diff options
634 files changed, 34341 insertions, 13419 deletions
diff --git a/Documentation/DocBook/Makefile b/Documentation/DocBook/Makefile index 6fd1646d320..08687e45e19 100644 --- a/Documentation/DocBook/Makefile +++ b/Documentation/DocBook/Makefile @@ -15,11 +15,11 @@ DOCBOOKS := wanbook.xml z8530book.xml mcabook.xml videobook.xml \ ### # The build process is as follows (targets): -# (xmldocs) -# file.tmpl --> file.xml +--> file.ps (psdocs) -# +--> file.pdf (pdfdocs) -# +--> DIR=file (htmldocs) -# +--> man/ (mandocs) +# (xmldocs) [by docproc] +# file.tmpl --> file.xml +--> file.ps (psdocs) [by db2ps or xmlto] +# +--> file.pdf (pdfdocs) [by db2pdf or xmlto] +# +--> DIR=file (htmldocs) [by xmlto] +# +--> man/ (mandocs) [by xmlto] # for PDF and PS output you can choose between xmlto and docbook-utils tools diff --git a/Documentation/DocBook/kernel-api.tmpl b/Documentation/DocBook/kernel-api.tmpl index a0af560ed74..eb42bf9847c 100644 --- a/Documentation/DocBook/kernel-api.tmpl +++ b/Documentation/DocBook/kernel-api.tmpl @@ -159,7 +159,6 @@ X!Ilib/string.c !Earch/i386/lib/usercopy.c </sect1> <sect1><title>More Memory Management Functions</title> -!Iinclude/linux/rmap.h !Emm/readahead.c !Emm/filemap.c !Emm/memory.c diff --git a/Documentation/connector/cn_test.c b/Documentation/connector/cn_test.c index 3e73231695b..be7af146dd3 100644 --- a/Documentation/connector/cn_test.c +++ b/Documentation/connector/cn_test.c @@ -124,9 +124,8 @@ static void cn_test_timer_func(unsigned long __data) struct cn_msg *m; char data[32]; - m = kmalloc(sizeof(*m) + sizeof(data), GFP_ATOMIC); + m = kzalloc(sizeof(*m) + sizeof(data), GFP_ATOMIC); if (m) { - memset(m, 0, sizeof(*m) + sizeof(data)); memcpy(&m->id, &cn_test_id, sizeof(m->id)); m->seq = cn_test_timer_counter; diff --git a/Documentation/console/console.txt b/Documentation/console/console.txt index d3e17447321..877a1b26cc3 100644 --- a/Documentation/console/console.txt +++ b/Documentation/console/console.txt @@ -29,7 +29,7 @@ In newer kernels, the following are also available: If sysfs is enabled, the contents of /sys/class/vtconsole can be examined. This shows the console backends currently registered by the -system which are named vtcon<n> where <n> is an integer fro 0 to 15. Thus: +system which are named vtcon<n> where <n> is an integer from 0 to 15. Thus: ls /sys/class/vtconsole . .. vtcon0 vtcon1 diff --git a/Documentation/drivers/edac/edac.txt b/Documentation/drivers/edac/edac.txt index 3c5a9e4297b..a5c36842ece 100644 --- a/Documentation/drivers/edac/edac.txt +++ b/Documentation/drivers/edac/edac.txt @@ -2,22 +2,42 @@ EDAC - Error Detection And Correction -Written by Doug Thompson <norsk5@xmission.com> +Written by Doug Thompson <dougthompson@xmission.com> 7 Dec 2005 +17 Jul 2007 Updated -EDAC was written by: - Thayne Harbaugh, - modified by Dave Peterson, Doug Thompson, et al, - from the bluesmoke.sourceforge.net project. +EDAC is maintained and written by: + Doug Thompson, Dave Jiang, Dave Peterson et al, + original author: Thayne Harbaugh, + +Contact: + website: bluesmoke.sourceforge.net + mailing list: bluesmoke-devel@lists.sourceforge.net + +"bluesmoke" was the name for this device driver when it was "out-of-tree" +and maintained at sourceforge.net. When it was pushed into 2.6.16 for the +first time, it was renamed to 'EDAC'. + +The bluesmoke project at sourceforge.net is now utilized as a 'staging area' +for EDAC development, before it is sent upstream to kernel.org + +At the bluesmoke/EDAC project site, is a series of quilt patches against +recent kernels, stored in a SVN respository. For easier downloading, there +is also a tarball snapshot available. ============================================================================ EDAC PURPOSE The 'edac' kernel module goal is to detect and report errors that occur -within the computer system. In the initial release, memory Correctable Errors -(CE) and Uncorrectable Errors (UE) are the primary errors being harvested. +within the computer system running under linux. + +MEMORY + +In the initial release, memory Correctable Errors (CE) and Uncorrectable +Errors (UE) are the primary errors being harvested. These types of errors +are harvested by the 'edac_mc' class of device. Detecting CE events, then harvesting those events and reporting them, CAN be a predictor of future UE events. With CE events, the system can @@ -25,9 +45,27 @@ continue to operate, but with less safety. Preventive maintenance and proactive part replacement of memory DIMMs exhibiting CEs can reduce the likelihood of the dreaded UE events and system 'panics'. +NON-MEMORY + +A new feature for EDAC, the edac_device class of device, was added in +the 2.6.23 version of the kernel. + +This new device type allows for non-memory type of ECC hardware detectors +to have their states harvested and presented to userspace via the sysfs +interface. + +Some architectures have ECC detectors for L1, L2 and L3 caches, along with DMA +engines, fabric switches, main data path switches, interconnections, +and various other hardware data paths. If the hardware reports it, then +a edac_device device probably can be constructed to harvest and present +that to userspace. + + +PCI BUS SCANNING In addition, PCI Bus Parity and SERR Errors are scanned for on PCI devices in order to determine if errors are occurring on data transfers. + The presence of PCI Parity errors must be examined with a grain of salt. There are several add-in adapters that do NOT follow the PCI specification with regards to Parity generation and reporting. The specification says @@ -35,11 +73,17 @@ the vendor should tie the parity status bits to 0 if they do not intend to generate parity. Some vendors do not do this, and thus the parity bit can "float" giving false positives. -[There are patches in the kernel queue which will allow for storage of -quirks of PCI devices reporting false parity positives. The 2.6.18 -kernel should have those patches included. When that becomes available, -then EDAC will be patched to utilize that information to "skip" such -devices.] +In the kernel there is a pci device attribute located in sysfs that is +checked by the EDAC PCI scanning code. If that attribute is set, +PCI parity/error scannining is skipped for that device. The attribute +is: + + broken_parity_status + +as is located in /sys/devices/pci<XXX>/0000:XX:YY.Z directorys for +PCI devices. + +FUTURE HARDWARE SCANNING EDAC will have future error detectors that will be integrated with EDAC or added to it, in the following list: @@ -57,13 +101,14 @@ and the like. ============================================================================ EDAC VERSIONING -EDAC is composed of a "core" module (edac_mc.ko) and several Memory +EDAC is composed of a "core" module (edac_core.ko) and several Memory Controller (MC) driver modules. On a given system, the CORE is loaded and one MC driver will be loaded. Both the CORE and -the MC driver have individual versions that reflect current release -level of their respective modules. Thus, to "report" on what version -a system is running, one must report both the CORE's and the -MC driver's versions. +the MC driver (or edac_device driver) have individual versions that reflect +current release level of their respective modules. + +Thus, to "report" on what version a system is running, one must report both +the CORE's and the MC driver's versions. LOADING @@ -88,8 +133,9 @@ EDAC sysfs INTERFACE EDAC presents a 'sysfs' interface for control, reporting and attribute reporting purposes. -EDAC lives in the /sys/devices/system/edac directory. Within this directory -there currently reside 2 'edac' components: +EDAC lives in the /sys/devices/system/edac directory. + +Within this directory there currently reside 2 'edac' components: mc memory controller(s) system pci PCI control and status system @@ -188,7 +234,7 @@ In directory 'mc' are EDAC system overall control and attribute files: Panic on UE control file: - 'panic_on_ue' + 'edac_mc_panic_on_ue' An uncorrectable error will cause a machine panic. This is usually desirable. It is a bad idea to continue when an uncorrectable error @@ -199,12 +245,12 @@ Panic on UE control file: LOAD TIME: module/kernel parameter: panic_on_ue=[0|1] - RUN TIME: echo "1" >/sys/devices/system/edac/mc/panic_on_ue + RUN TIME: echo "1" >/sys/devices/system/edac/mc/edac_mc_panic_on_ue Log UE control file: - 'log_ue' + 'edac_mc_log_ue' Generate kernel messages describing uncorrectable errors. These errors are reported through the system message log system. UE statistics @@ -212,12 +258,12 @@ Log UE control file: LOAD TIME: module/kernel parameter: log_ue=[0|1] - RUN TIME: echo "1" >/sys/devices/system/edac/mc/log_ue + RUN TIME: echo "1" >/sys/devices/system/edac/mc/edac_mc_log_ue Log CE control file: - 'log_ce' + 'edac_mc_log_ce' Generate kernel messages describing correctable errors. These errors are reported through the system message log system. @@ -225,12 +271,12 @@ Log CE control file: LOAD TIME: module/kernel parameter: log_ce=[0|1] - RUN TIME: echo "1" >/sys/devices/system/edac/mc/log_ce + RUN TIME: echo "1" >/sys/devices/system/edac/mc/edac_mc_log_ce Polling period control file: - 'poll_msec' + 'edac_mc_poll_msec' The time period, in milliseconds, for polling for error information. Too small a value wastes resources. Too large a value might delay @@ -241,7 +287,7 @@ Polling period control file: LOAD TIME: module/kernel parameter: poll_msec=[0|1] - RUN TIME: echo "1000" >/sys/devices/system/edac/mc/poll_msec + RUN TIME: echo "1000" >/sys/devices/system/edac/mc/edac_mc_poll_msec ============================================================================ @@ -587,3 +633,95 @@ Parity Count: ======================================================================= + + +EDAC_DEVICE type of device + +In the header file, edac_core.h, there is a series of edac_device structures +and APIs for the EDAC_DEVICE. + +User space access to an edac_device is through the sysfs interface. + +At the location /sys/devices/system/edac (sysfs) new edac_device devices will +appear. + +There is a three level tree beneath the above 'edac' directory. For example, +the 'test_device_edac' device (found at the bluesmoke.sourceforget.net website) +installs itself as: + + /sys/devices/systm/edac/test-instance + +in this directory are various controls, a symlink and one or more 'instance' +directorys. + +The standard default controls are: + + log_ce boolean to log CE events + log_ue boolean to log UE events + panic_on_ue boolean to 'panic' the system if an UE is encountered + (default off, can be set true via startup script) + poll_msec time period between POLL cycles for events + +The test_device_edac device adds at least one of its own custom control: + + test_bits which in the current test driver does nothing but + show how it is installed. A ported driver can + add one or more such controls and/or attributes + for specific uses. + One out-of-tree driver uses controls here to allow + for ERROR INJECTION operations to hardware + injection registers + +The symlink points to the 'struct dev' that is registered for this edac_device. + +INSTANCES + +One or more instance directories are present. For the 'test_device_edac' case: + + test-instance0 + + +In this directory there are two default counter attributes, which are totals of +counter in deeper subdirectories. + + ce_count total of CE events of subdirectories + ue_count total of UE events of subdirectories + +BLOCKS + +At the lowest directory level is the 'block' directory. There can be 0, 1 +or more blocks specified in each instance. + + test-block0 + + +In this directory the default attributes are: + + ce_count which is counter of CE events for this 'block' + of hardware being monitored + ue_count which is counter of UE events for this 'block' + of hardware being monitored + + +The 'test_device_edac' device adds 4 attributes and 1 control: + + test-block-bits-0 for every POLL cycle this counter + is incremented + test-block-bits-1 every 10 cycles, this counter is bumped once, + and test-block-bits-0 is set to 0 + test-block-bits-2 every 100 cycles, this counter is bumped once, + and test-block-bits-1 is set to 0 + test-block-bits-3 every 1000 cycles, this counter is bumped once, + and test-block-bits-2 is set to 0 + + + reset-counters writing ANY thing to this control will + reset all the above counters. + + +Use of the 'test_device_edac' driver should any others to create their own +unique drivers for their hardware systems. + +The 'test_device_edac' sample driver is located at the +bluesmoke.sourceforge.net project site for EDAC. + diff --git a/Documentation/feature-removal-schedule.txt b/Documentation/feature-removal-schedule.txt index 66c8b4b165c..a5cb7839a67 100644 --- a/Documentation/feature-removal-schedule.txt +++ b/Documentation/feature-removal-schedule.txt @@ -51,6 +51,7 @@ Who: David Miller <davem@davemloft.net> What: Video4Linux API 1 ioctls and video_decoder.h from Video devices. When: December 2006 Files: include/linux/video_decoder.h +Check: include/linux/video_decoder.h Why: V4L1 AP1 was replaced by V4L2 API. during migration from 2.4 to 2.6 series. The old API have lots of drawbacks and don't provide enough means to work with all video and audio standards. The newer API is @@ -84,7 +85,7 @@ Who: Dominik Brodowski <linux@brodo.de> What: remove EXPORT_SYMBOL(kernel_thread) When: August 2006 Files: arch/*/kernel/*_ksyms.c -Funcs: kernel_thread +Check: kernel_thread Why: kernel_thread is a low-level implementation detail. Drivers should use the <linux/kthread.h> API instead which shields them from implementation details and provides a higherlevel interface that @@ -135,6 +136,15 @@ Who: Greg Kroah-Hartman <gregkh@suse.de> --------------------------- +What: vm_ops.nopage +When: Soon, provided in-kernel callers have been converted +Why: This interface is replaced by vm_ops.fault, but it has been around + forever, is used by a lot of drivers, and doesn't cost much to + maintain. +Who: Nick Piggin <npiggin@suse.de> + +--------------------------- + What: Interrupt only SA_* flags When: September 2007 Why: The interrupt related SA_* flags are replaced by IRQF_* to move them @@ -154,15 +164,6 @@ Who: Kay Sievers <kay.sievers@suse.de> --------------------------- -What: i2c-isa -When: December 2006 -Why: i2c-isa is a non-sense and doesn't fit in the device driver - model. Drivers relying on it are better implemented as platform - drivers. -Who: Jean Delvare <khali@linux-fr.org> - ---------------------------- - What: i2c_adapter.list When: July 2007 Why: Superfluous, this list duplicates the one maintained by the driver diff --git a/Documentation/filesystems/Locking b/Documentation/filesystems/Locking index d866551be03..f0f825808ca 100644 --- a/Documentation/filesystems/Locking +++ b/Documentation/filesystems/Locking @@ -510,13 +510,24 @@ More details about quota locking can be found in fs/dquot.c. prototypes: void (*open)(struct vm_area_struct*); void (*close)(struct vm_area_struct*); + int (*fault)(struct vm_area_struct*, struct vm_fault *); struct page *(*nopage)(struct vm_area_struct*, unsigned long, int *); + int (*page_mkwrite)(struct vm_area_struct *, struct page *); locking rules: - BKL mmap_sem + BKL mmap_sem PageLocked(page) open: no yes close: no yes +fault: no yes nopage: no yes +page_mkwrite: no yes no + + ->page_mkwrite() is called when a previously read-only page is +about to become writeable. The file system is responsible for +protecting against truncate races. Once appropriate action has been +taking to lock out truncate, the page range should be verified to be +within i_size. The page mapping should also be checked that it is not +NULL. ================================================================================ Dubious stuff diff --git a/Documentation/filesystems/configfs/configfs_example.c b/Documentation/filesystems/configfs/configfs_example.c index e56d49264b3..25151fd5c2c 100644 --- a/Documentation/filesystems/configfs/configfs_example.c +++ b/Documentation/filesystems/configfs/configfs_example.c @@ -277,11 +277,10 @@ static struct config_item *simple_children_make_item(struct config_group *group, { struct simple_child *simple_child; - simple_child = kmalloc(sizeof(struct simple_child), GFP_KERNEL); + simple_child = kzalloc(sizeof(struct simple_child), GFP_KERNEL); if (!simple_child) return NULL; - memset(simple_child, 0, sizeof(struct simple_child)); config_item_init_type_name(&simple_child->item, name, &simple_child_type); @@ -364,12 +363,11 @@ static struct config_group *group_children_make_group(struct config_group *group { struct simple_children *simple_children; - simple_children = kmalloc(sizeof(struct simple_children), + simple_children = kzalloc(sizeof(struct simple_children), GFP_KERNEL); if (!simple_children) return NULL; - memset(simple_children, 0, sizeof(struct simple_children)); config_group_init_type_name(&simple_children->group, name, &simple_children_type); diff --git a/Documentation/filesystems/proc.txt b/Documentation/filesystems/proc.txt index ebffdffb3d9..4a37e25e694 100644 --- a/Documentation/filesystems/proc.txt +++ b/Documentation/filesystems/proc.txt @@ -42,6 +42,7 @@ Table of Contents 2.12 /proc/<pid>/oom_adj - Adjust the oom-killer score 2.13 /proc/<pid>/oom_score - Display current oom-killer score 2.14 /proc/<pid>/io - Display the IO accounting fields + 2.15 /proc/<pid>/coredump_filter - Core dump filtering settings ------------------------------------------------------------------------------ Preface @@ -1065,6 +1066,13 @@ check the amount of free space (value is in seconds). Default settings are: 4, resume it if we have a value of 3 or more percent; consider information about the amount of free space valid for 30 seconds +audit_argv_kb +------------- + +The file contains a single value denoting the limit on the argv array size +for execve (in KiB). This limit is only applied when system call auditing for +execve is enabled, otherwise the value is ignored. + ctrl-alt-del ------------ @@ -2177,4 +2185,41 @@ those 64-bit counters, process A could see an intermediate result. More information about this can be found within the taskstats documentation in Documentation/accounting. +2.15 /proc/<pid>/coredump_filter - Core dump filtering settings +--------------------------------------------------------------- +When a process is dumped, all anonymous memory is written to a core file as +long as the size of the core file isn't limited. But sometimes we don't want +to dump some memory segments, for example, huge shared memory. Conversely, +sometimes we want to save file-backed memory segments into a core file, not +only the individual files. + +/proc/<pid>/coredump_filter allows you to customize which memory segments +will be dumped when the <pid> process is dumped. coredump_filter is a bitmask +of memory types. If a bit of the bitmask is set, memory segments of the +corresponding memory type are dumped, otherwise they are not dumped. + +The following 4 memory types are supported: + - (bit 0) anonymous private memory + - (bit 1) anonymous shared memory + - (bit 2) file-backed private memory + - (bit 3) file-backed shared memory + + Note that MMIO pages such as frame buffer are never dumped and vDSO pages + are always dumped regardless of the bitmask status. + +Default value of coredump_filter is 0x3; this means all anonymous memory +segments are dumped. + +If you don't want to dump all shared memory segments attached to pid 1234, +write 1 to the process's proc file. + + $ echo 0x1 > /proc/1234/coredump_filter + +When a new process is created, the process inherits the bitmask status from its +parent. It is useful to set up coredump_filter before the program runs. +For example: + + $ echo 0x7 > /proc/self/coredump_filter + $ ./some_program + ------------------------------------------------------------------------------ diff --git a/Documentation/gpio.txt b/Documentation/gpio.txt index 36af58eba13..218a8650f48 100644 --- a/Documentation/gpio.txt +++ b/Documentation/gpio.txt @@ -75,6 +75,9 @@ using the include file: If you stick to this convention then it'll be easier for other developers to see what your code is doing, and help maintain it. +Note that these operations include I/O barriers on platforms which need to +use them; drivers don't need to add them explicitly. + Identifying GPIOs ----------------- diff --git a/Documentation/hwmon/abituguru b/Documentation/hwmon/abituguru index b2c0d61b39a..87ffa0f5ec7 100644 --- a/Documentation/hwmon/abituguru +++ b/Documentation/hwmon/abituguru @@ -2,7 +2,7 @@ Kernel driver abituguru ======================= Supported chips: - * Abit uGuru revision 1-3 (Hardware Monitor part only) + * Abit uGuru revision 1 & 2 (Hardware Monitor part only) Prefix: 'abituguru' Addresses scanned: ISA 0x0E0 Datasheet: Not available, this driver is based on reverse engineering. @@ -20,8 +20,8 @@ Supported chips: uGuru 2.1.0.0 ~ 2.1.2.8 (AS8, AV8, AA8, AG8, AA8XE, AX8) uGuru 2.2.0.0 ~ 2.2.0.6 (AA8 Fatal1ty) uGuru 2.3.0.0 ~ 2.3.0.9 (AN8) - uGuru 3.0.0.0 ~ 3.0.1.2 (AW8, AL8, NI8) - uGuru 4.xxxxx? (AT8 32X) (2) + uGuru 3.0.0.0 ~ 3.0.x.x (AW8, AL8, AT8, NI8 SLI, AT8 32X, AN8 32X, + AW9D-MAX) (2) 1) For revisions 2 and 3 uGuru's the driver can autodetect the sensortype (Volt or Temp) for bank1 sensors, for revision 1 uGuru's this doesnot always work. For these uGuru's the autodection can @@ -30,8 +30,9 @@ Supported chips: bank1_types=1,1,0,0,0,0,0,2,0,0,0,0,2,0,0,1 You may also need to specify the fan_sensors option for these boards fan_sensors=5 - 2) The current version of the abituguru driver is known to NOT work - on these Motherboards + 2) There is a seperate abituguru3 driver for these motherboards, + the abituguru (without the 3 !) driver will not work on these + motherboards (and visa versa)! Authors: Hans de Goede <j.w.r.degoede@hhs.nl>, @@ -43,8 +44,10 @@ Module Parameters ----------------- * force: bool Force detection. Note this parameter only causes the - detection to be skipped, if the uGuru can't be read - the module initialization (insmod) will still fail. + detection to be skipped, and thus the insmod to + succeed. If the uGuru can't be read the actual hwmon + driver will not load and thus no hwmon device will get + registered. * bank1_types: int[] Bank1 sensortype autodetection override: -1 autodetect (default) 0 volt sensor @@ -69,13 +72,15 @@ dmesg | grep abituguru Description ----------- -This driver supports the hardware monitoring features of the Abit uGuru chip -found on Abit uGuru featuring motherboards (most modern Abit motherboards). +This driver supports the hardware monitoring features of the first and +second revision of the Abit uGuru chip found on Abit uGuru featuring +motherboards (most modern Abit motherboards). -The uGuru chip in reality is a Winbond W83L950D in disguise (despite Abit -claiming it is "a new microprocessor designed by the ABIT Engineers"). -Unfortunatly this doesn't help since the W83L950D is a generic -microcontroller with a custom Abit application running on it. +The first and second revision of the uGuru chip in reality is a Winbond +W83L950D in disguise (despite Abit claiming it is "a new microprocessor +designed by the ABIT Engineers"). Unfortunatly this doesn't help since the +W83L950D is a generic microcontroller with a custom Abit application running +on it. Despite Abit not releasing any information regarding the uGuru, Olle Sandberg <ollebull@gmail.com> has managed to reverse engineer the sensor part diff --git a/Documentation/hwmon/abituguru3 b/Documentation/hwmon/abituguru3 new file mode 100644 index 00000000000..fa598aac22f --- /dev/null +++ b/Documentation/hwmon/abituguru3 @@ -0,0 +1,65 @@ +Kernel driver abituguru3 +======================== + +Supported chips: + * Abit uGuru revision 3 (Hardware Monitor part, reading only) + Prefix: 'abituguru3' + Addresses scanned: ISA 0x0E0 + Datasheet: Not available, this driver is based on reverse engineering. + Note: + The uGuru is a microcontroller with onboard firmware which programs + it to behave as a hwmon IC. There are many different revisions of the + firmware and thus effectivly many different revisions of the uGuru. + Below is an incomplete list with which revisions are used for which + Motherboards: + uGuru 1.00 ~ 1.24 (AI7, KV8-MAX3, AN7) + uGuru 2.0.0.0 ~ 2.0.4.2 (KV8-PRO) + uGuru 2.1.0.0 ~ 2.1.2.8 (AS8, AV8, AA8, AG8, AA8XE, AX8) + uGuru 2.3.0.0 ~ 2.3.0.9 (AN8) + uGuru 3.0.0.0 ~ 3.0.x.x (AW8, AL8, AT8, NI8 SLI, AT8 32X, AN8 32X, + AW9D-MAX) + The abituguru3 driver is only for revison 3.0.x.x motherboards, + this driver will not work on older motherboards. For older + motherboards use the abituguru (without the 3 !) driver. + +Authors: + Hans de Goede <j.w.r.degoede@hhs.nl>, + (Initial reverse engineering done by Louis Kruger) + + +Module Parameters +----------------- + +* force: bool Force detection. Note this parameter only causes the + detection to be skipped, and thus the insmod to + succeed. If the uGuru can't be read the actual hwmon + driver will not load and thus no hwmon device will get + registered. +* verbose: bool Should the driver be verbose? + 0/off/false normal output + 1/on/true + verbose error reporting (default) + Default: 1 (the driver is still in the testing phase) + +Description +----------- + +This driver supports the hardware monitoring features of the third revision of +the Abit uGuru chip, found on recent Abit uGuru featuring motherboards. + +The 3rd revision of the uGuru chip in reality is a Winbond W83L951G. +Unfortunatly this doesn't help since the W83L951G is a generic microcontroller +with a custom Abit application running on it. + +Despite Abit not releasing any information regarding the uGuru revision 3, +Louis Kruger has managed to reverse engineer the sensor part of the uGuru. +Without his work this driver would not have been possible. + +Known Issues +------------ + +The voltage and frequency control parts of the Abit uGuru are not supported, +neither is writing any of the sensor settings and writing / reading the +fanspeed control registers (FanEQ) + +If you encounter any problems please mail me <j.w.r.degoede@hhs.nl> and +include the output of: "dmesg | grep abituguru" diff --git a/Documentation/hwmon/dme1737 b/Documentation/hwmon/dme1737 new file mode 100644 index 00000000000..1a0f3d64ab8 --- /dev/null +++ b/Documentation/hwmon/dme1737 @@ -0,0 +1,257 @@ +Kernel driver dme1737 +===================== + +Supported chips: + * SMSC DME1737 and compatibles (like Asus A8000) + Prefix: 'dme1737' + Addresses scanned: I2C 0x2c, 0x2d, 0x2e + Datasheet: Provided by SMSC upon request and under NDA + +Authors: + Juerg Haefliger <juergh@gmail.com> + + +Module Parameters +----------------- + +* force_start: bool Enables the monitoring of voltage, fan and temp inputs + and PWM output control functions. Using this parameter + shouldn't be required since the BIOS usually takes care + of this. + +Note that there is no need to use this parameter if the driver loads without +complaining. The driver will say so if it is necessary. + + +Description +----------- + +This driver implements support for the hardware monitoring capabilities of the +SMSC DME1737 and Asus A8000 (which are the same) Super-I/O chips. This chip +features monitoring of 3 temp sensors temp[1-3] (2 remote diodes and 1 +internal), 7 voltages in[0-6] (6 external and 1 internal) and 6 fan speeds +fan[1-6]. Additionally, the chip implements 5 PWM outputs pwm[1-3,5-6] for +controlling fan speeds both manually and automatically. + +Fan[3-6] and pwm[3,5-6] are optional features and their availability is +dependent on the configuration of the chip. The driver will detect which +features are present during initialization and create the sysfs attributes +accordingly. + + +Voltage Monitoring +------------------ + +The voltage inputs are sampled with 12-bit resolution and have internal +scaling resistors. The values returned by the driver therefore reflect true +millivolts and don't need scaling. The voltage inputs are mapped as follows +(the last column indicates the input ranges): + + in0: +5VTR (+5V standby) 0V - 6.64V + in1: Vccp (processor core) 0V - 3V + in2: VCC (internal +3.3V) 0V - 4.38V + in3: +5V 0V - 6.64V + in4: +12V 0V - 16V + in5: VTR (+3.3V standby) 0V - 4.38V + in6: Vbat (+3.0V) 0V - 4.38V + +Each voltage input has associated min and max limits which trigger an alarm +when crossed. + + +Temperature Monitoring +---------------------- + +Temperatures are measured with 12-bit resolution and reported in millidegree +Celsius. The chip also features offsets for all 3 temperature inputs which - +when programmed - get added to the input readings. The chip does all the +scaling by itself and the driver therefore reports true temperatures that don't +need any user-space adjustments. The temperature inputs are mapped as follows +(the last column indicates the input ranges): + + temp1: Remote diode 1 (3904 type) temperature -127C - +127C + temp2: DME1737 internal temperature -127C - +127C + temp3: Remote diode 2 (3904 type) temperature -127C - +127C + +Each temperature input has associated min and max limits which trigger an alarm +when crossed. Additionally, each temperature input has a fault attribute that +returns 1 when a faulty diode or an unconnected input is detected and 0 +otherwise. + + +Fan Monitoring +-------------- + +Fan RPMs are measured with 16-bit resolution. The chip provides inputs for 6 +fan tachometers. All 6 inputs have an associated min limit which triggers an +alarm when crossed. Fan inputs 1-4 provide type attributes that need to be set +to the number of pulses per fan revolution that the connected tachometer +generates. Supported values are 1, 2, and 4. Fan inputs 5-6 only support fans +that generate 2 pulses per revolution. Fan inputs 5-6 also provide a max +attribute that needs to be set to the maximum attainable RPM (fan at 100% duty- +cycle) of the input. The chip adjusts the sampling rate based on this value. + + +PWM Output Control +------------------ + +This chip features 5 PWM outputs. PWM outputs 1-3 are associated with fan +inputs 1-3 and PWM outputs 5-6 are associated with fan inputs 5-6. PWM outputs +1-3 can be configured to operate either in manual or automatic mode by setting +the appropriate enable attribute accordingly. PWM outputs 5-6 can only operate +in manual mode, their enable attributes are therefore read-only. When set to +manual mode, the fan speed is set by writing the duty-cycle value to the +appropriate PWM attribute. In automatic mode, the PWM attribute returns the +current duty-cycle as set by the fan controller in the chip. All PWM outputs +support the setting of the output frequency via the freq attribute. + +In automatic mode, the chip supports the setting of the PWM ramp rate which +defines how fast the PWM output is adjusting to changes of the associated +temperature input. Associating PWM outputs to temperature inputs is done via +temperature zones. The chip features 3 zones whose assignments to temperature +inputs is static and determined during initialization. These assignments can +be retrieved via the zone[1-3]_auto_channels_temp attributes. Each PWM output +is assigned to one (or hottest of multiple) temperature zone(s) through the +pwm[1-3]_auto_channels_zone attributes. Each PWM output has 3 distinct output +duty-cycles: full, low, and min. Full is internally hard-wired to 255 (100%) +and low and min can be programmed via pwm[1-3]_auto_point1_pwm and +pwm[1-3]_auto_pwm_min, respectively. The thermal thresholds of the zones are +programmed via zone[1-3]_auto_point[1-3]_temp and +zone[1-3]_auto_point1_temp_hyst: + + pwm[1-3]_auto_point2_pwm full-speed duty-cycle (255, i.e., 100%) + pwm[1-3]_auto_point1_pwm low-speed duty-cycle + pwm[1-3]_auto_pwm_min min-speed duty-cycle + + zone[1-3]_auto_point3_temp full-speed temp (all outputs) + zone[1-3]_auto_point2_temp full-speed temp + zone[1-3]_auto_point1_temp low-speed temp + zone[1-3]_auto_point1_temp_hyst min-speed temp + +The chip adjusts the output duty-cycle linearly in the range of auto_point1_pwm +to auto_point2_pwm if the temperature of the associated zone is between +auto_point1_temp and auto_point2_temp. If the temperature drops below the +auto_point1_temp_hyst value, the output duty-cycle is set to the auto_pwm_min +value which only supports two values: 0 or auto_point1_pwm. That means that the +fan either turns completely off or keeps spinning with the low-speed +duty-cycle. If any of the temperatures rise above the auto_point3_temp value, +all PWM outputs are set to 100% duty-cycle. + +Following is another representation of how the chip sets the output duty-cycle +based on the temperature of the associated thermal zone: + + Duty-Cycle Duty-Cycle + Temperature Rising Temp Falling Temp + ----------- ----------- ------------ + full-speed full-speed full-speed + + < linearly adjusted duty-cycle > + + low-speed low-speed low-speed + min-speed low-speed + min-speed min-speed min-speed + min-speed min-speed + + +Sysfs Attributes +---------------- + +Following is a list of all sysfs attributes that the driver provides, their +permissions and a short description: + +Name Perm Description +---- ---- ----------- +cpu0_vid RO CPU core reference voltage in + millivolts. +vrm RW Voltage regulator module version + number. + +in[0-6]_input RO Measured voltage in millivolts. +in[0-6]_min RW Low limit for voltage input. +in[0-6]_max RW High limit for voltage input. +in[0-6]_alarm RO Voltage input alarm. Returns 1 if + voltage input is or went outside the + associated min-max range, 0 otherwise. + +temp[1-3]_input RO Measured temperature in millidegree + Celsius. +temp[1-3]_min RW Low limit for temp input. +temp[1-3]_max RW High limit for temp input. +temp[1-3]_offset RW Offset for temp input. This value will + be added by the chip to the measured + temperature. +temp[1-3]_alarm RO Alarm for temp input. Returns 1 if temp + input is or went outside the associated + min-max range, 0 otherwise. +temp[1-3]_fault RO Temp input fault. Returns 1 if the chip + detects a faulty thermal diode or an + unconnected temp input, 0 otherwise. + +zone[1-3]_auto_channels_temp RO Temperature zone to temperature input + mapping. This attribute is a bitfield + and supports the following values: + 1: temp1 + 2: temp2 + 4: temp3 +zone[1-3]_auto_point1_temp_hyst RW Auto PWM temp point1 hysteresis. The + output of the corresponding PWM is set + to the pwm_auto_min value if the temp + falls below the auto_point1_temp_hyst + value. +zone[1-3]_auto_point[1-3]_temp RW Auto PWM temp points. Auto_point1 is + the low-speed temp, auto_point2 is the + full-speed temp, and auto_point3 is the + temp at which all PWM outputs are set + to full-speed (100% duty-cycle). + +fan[1-6]_input RO Measured fan speed in RPM. +fan[1-6]_min RW Low limit for fan input. +fan[1-6]_alarm RO Alarm for fan input. Returns 1 if fan + input is or went below the associated + min value, 0 otherwise. +fan[1-4]_type RW Type of attached fan. Expressed in + number of pulses per revolution that + the fan generates. Supported values are + 1, 2, and 4. +fan[5-6]_max RW Max attainable RPM at 100% duty-cycle. + Required for chip to adjust the + sampling rate accordingly. + +pmw[1-3,5-6] RO/RW Duty-cycle of PWM output. Supported + values are 0-255 (0%-100%). Only + writeable if the associated PWM is in + manual mode. +pwm[1-3]_enable RW Enable of PWM outputs 1-3. Supported + values are: + 0: turned off (output @ 100%) + 1: manual mode + 2: automatic mode +pwm[5-6]_enable RO Enable of PWM outputs 5-6. Always + returns 1 since these 2 outputs are + hard-wired to manual mode. +pmw[1-3,5-6]_freq RW Frequency of PWM output. Supported + values are in the range 11Hz-30000Hz + (default is 25000Hz). +pmw[1-3]_ramp_rate RW Ramp rate of PWM output. Determines how + fast the PWM duty-cycle will change + when the PWM is in automatic mode. + Expressed in ms per PWM step. Supported + values are in the range 0ms-206ms + (default is 0, which means the duty- + cycle changes instantly). +pwm[1-3]_auto_channels_zone RW PWM output to temperature zone mapping. + This attribute is a bitfield and + supports the following values: + 1: zone1 + 2: zone2 + 4: zone3 + 6: highest of zone[2-3] + 7: highest of zone[1-3] +pwm[1-3]_auto_pwm_min RW Auto PWM min pwm. Minimum PWM duty- + cycle. Supported values are 0 or + auto_point1_pwm. +pwm[1-3]_auto_point1_pwm RW Auto PWM pwm point. Auto_point1 is the + low-speed duty-cycle. +pwm[1-3]_auto_point2_pwm RO Auto PWM pwm point. Auto_point2 is the + full-speed duty-cycle which is hard- + wired to 255 (100% duty-cycle). diff --git a/Documentation/hwmon/f71805f b/Documentation/hwmon/f71805f index bfd0f154959..94e0d2cbd3d 100644 --- a/Documentation/hwmon/f71805f +++ b/Documentation/hwmon/f71805f @@ -5,11 +5,11 @@ Supported chips: * Fintek F71805F/FG Prefix: 'f71805f' Addresses scanned: none, address read from Super I/O config space - Datasheet: Provided by Fintek on request + Datasheet: Available from the Fintek website * Fintek F71872F/FG Prefix: 'f71872f' Addresses scanned: none, address read from Super I/O config space - Datasheet: Provided by Fintek on request + Datasheet: Available from the Fintek website Author: Jean Delvare <khali@linux-fr.org> @@ -128,7 +128,9 @@ it. When the PWM method is used, you can select the operating frequency, from 187.5 kHz (default) to 31 Hz. The best frequency depends on the fan model. As a rule of thumb, lower frequencies seem to give better -control, but may generate annoying high-pitch noise. Fintek recommends +control, but may generate annoying high-pitch noise. So a frequency just +above the audible range, such as 25 kHz, may be a good choice; if this +doesn't give you good linear control, try reducing it. Fintek recommends not going below 1 kHz, as the fan tachometers get confused by lower frequencies as well. @@ -136,16 +138,23 @@ When the DC method is used, Fintek recommends not going below 5 V, which corresponds to a pwm value of 106 for the driver. The driver doesn't enforce this limit though. -Three different fan control modes are supported: +Three different fan control modes are supported; the mode number is written +to the pwm<n>_enable file. -* Manual mode - You ask for a specific PWM duty cycle or DC voltage. +* 1: Manual mode + You ask for a specific PWM duty cycle or DC voltage by writing to the + pwm<n> file. -* Fan speed mode - You ask for a specific fan speed. This mode assumes that pwm1 - corresponds to fan1, pwm2 to fan2 and pwm3 to fan3. +* 2: Temperature mode + You define 3 temperature/fan speed trip points using the + pwm<n>_auto_point<m>_temp and _fan files. These define a staircase + relationship between temperature and fan speed with two additional points + interpolated between the values that you define. When the temperature + is below auto_point1_temp the fan is switched off. -* Temperature mode - You define 3 temperature/fan speed trip points, and the fan speed is - adjusted depending on the measured temperature, using interpolation. - This mode is not yet supported by the driver. +* 3: Fan speed mode + You ask for a specific fan speed by writing to the fan<n>_target file. + +Both of the automatic modes require that pwm1 corresponds to fan1, pwm2 to +fan2 and pwm3 to fan3. Temperature mode also requires that temp1 corresponds +to pwm1 and fan1, etc. diff --git a/Documentation/hwmon/it87 b/Documentation/hwmon/it87 index c0528d6f9ac..81ecc7e41c5 100644 --- a/Documentation/hwmon/it87 +++ b/Documentation/hwmon/it87 @@ -12,11 +12,12 @@ Supported chips: Addresses scanned: from Super I/O config space (8 I/O ports) Datasheet: Publicly available at the ITE website http://www.ite.com.tw/ - * IT8716F + * IT8716F/IT8726F Prefix: 'it8716' Addresses scanned: from Super I/O config space (8 I/O ports) Datasheet: Publicly available at the ITE website http://www.ite.com.tw/product_info/file/pc/IT8716F_V0.3.ZIP + http://www.ite.com.tw/product_info/file/pc/IT8726F_V0.3.pdf * IT8718F Prefix: 'it8718' Addresses scanned: from Super I/O config space (8 I/O ports) @@ -68,7 +69,7 @@ Description ----------- This driver implements support for the IT8705F, IT8712F, IT8716F, -IT8718F and SiS950 chips. +IT8718F, IT8726F and SiS950 chips. These chips are 'Super I/O chips', supporting floppy disks, infrared ports, joysticks and other miscellaneous stuff. For hardware monitoring, they @@ -97,6 +98,10 @@ clock divider mess) but not compatible with the older chips and revisions. For now, the driver only uses the 16-bit mode on the IT8716F and IT8718F. +The IT8726F is just bit enhanced IT8716F with additional hardware +for AMD power sequencing. Therefore the chip will appear as IT8716F +to userspace applications. + Temperatures are measured in degrees Celsius. An alarm is triggered once when the Overtemperature Shutdown limit is crossed. diff --git a/Documentation/hwmon/lm90 b/Documentation/hwmon/lm90 index 438cb24cee5..aa4a0ec2008 100644 --- a/Documentation/hwmon/lm90 +++ b/Documentation/hwmon/lm90 @@ -48,6 +48,18 @@ Supported chips: Addresses scanned: I2C 0x4c, 0x4d (unsupported 0x4e) Datasheet: Publicly available at the Maxim website http://www.maxim-ic.com/quick_view2.cfm/qv_pk/2578 + * Maxim MAX6680 + Prefix: 'max6680' + Addresses scanned: I2C 0x18, 0x19, 0x1a, 0x29, 0x2a, 0x2b, + 0x4c, 0x4d and 0x4e + Datasheet: Publicly available at the Maxim website + http://www.maxim-ic.com/quick_view2.cfm/qv_pk/3370 + * Maxim MAX6681 + Prefix: 'max6680' + Addresses scanned: I2C 0x18, 0x19, 0x1a, 0x29, 0x2a, 0x2b, + 0x4c, 0x4d and 0x4e + Datasheet: Publicly available at the Maxim website + http://www.maxim-ic.com/quick_view2.cfm/qv_pk/3370 Author: Jean Delvare <khali@linux-fr.org> @@ -59,11 +71,15 @@ Description The LM90 is a digital temperature sensor. It senses its own temperature as well as the temperature of up to one external diode. It is compatible with many other devices such as the LM86, the LM89, the LM99, the ADM1032, -the MAX6657, MAX6658 and the MAX6659 all of which are supported by this driver. -Note that there is no easy way to differentiate between the last three -variants. The extra address and features of the MAX6659 are not supported by -this driver. Additionally, the ADT7461 is supported if found in ADM1032 -compatibility mode. +the MAX6657, MAX6658, MAX6659, MAX6680 and the MAX6681 all of which are +supported by this driver. + +Note that there is no easy way to differentiate between the MAX6657, +MAX6658 and MAX6659 variants. The extra address and features of the +MAX6659 are not supported by this driver. The MAX6680 and MAX6681 only +differ in their pinout, therefore they obviously can't (and don't need to) +be distinguished. Additionally, the ADT7461 is supported if found in +ADM1032 compatibility mode. The specificity of this family of chipsets over the ADM1021/LM84 family is that it features critical limits with hysteresis, and an @@ -93,18 +109,22 @@ ADM1032: * ALERT is triggered by open remote sensor. * SMBus PEC support for Write Byte and Receive Byte transactions. -ADT7461 +ADT7461: * Extended temperature range (breaks compatibility) * Lower resolution for remote temperature MAX6657 and MAX6658: * Remote sensor type selection -MAX6659 +MAX6659: * Selectable address * Second critical temperature limit * Remote sensor type selection +MAX6680 and MAX6681: + * Selectable address + * Remote sensor type selection + All temperature values are given in degrees Celsius. Resolution is 1.0 degree for the local temperature, 0.125 degree for the remote temperature. @@ -141,7 +161,7 @@ SMBus Read Byte, and PEC will work properly. Additionally, the ADM1032 doesn't support SMBus Send Byte with PEC. Instead, it will try to write the PEC value to the register (because the SMBus Send Byte transaction with PEC is similar to a Write Byte transaction -without PEC), which is not what we want. Thus, PEC is explicitely disabled +without PEC), which is not what we want. Thus, PEC is explicitly disabled on SMBus Send Byte transactions in the lm90 driver. PEC on byte data transactions represents a significant increase in bandwidth diff --git a/Documentation/hwmon/lm93 b/Documentation/hwmon/lm93 new file mode 100644 index 00000000000..4e4a1dc1d2d --- /dev/null +++ b/Documentation/hwmon/lm93 @@ -0,0 +1,412 @@ +Kernel driver lm93 +================== + +Supported chips: + * National Semiconductor LM93 + Prefix 'lm93' + Addresses scanned: I2C 0x2c-0x2e + Datasheet: http://www.national.com/ds.cgi/LM/LM93.pdf + +Author: + Mark M. Hoffman <mhoffman@lightlink.com> + Ported to 2.6 by Eric J. Bowersox <ericb@aspsys.com> + Adapted to 2.6.20 by Carsten Emde <ce@osadl.org> + Modified for mainline integration by Hans J. Koch <hjk@linutronix.de> + +Module Parameters +----------------- + +(specific to LM93) +* init: integer + Set to non-zero to force some initializations (default is 0). +* disable_block: integer + A "0" allows SMBus block data transactions if the host supports them. A "1" + disables SMBus block data transactions. The default is 0. +* vccp_limit_type: integer array (2) + Configures in7 and in8 limit type, where 0 means absolute and non-zero + means relative. "Relative" here refers to "Dynamic Vccp Monitoring using + VID" from the datasheet. It greatly simplifies the interface to allow + only one set of limits (absolute or relative) to be in operation at a + time (even though the hardware is capable of enabling both). There's + not a compelling use case for enabling both at once, anyway. The default + is "0,0". +* vid_agtl: integer + A "0" configures the VID pins for V(ih) = 2.1V min, V(il) = 0.8V max. + A "1" configures the VID pins for V(ih) = 0.8V min, V(il) = 0.4V max. + (The latter setting is referred to as AGTL+ Compatible in the datasheet.) + I.e. this parameter controls the VID pin input thresholds; if your VID + inputs are not working, try changing this. The default value is "0". + +(common among sensor drivers) +* force: short array (min = 1, max = 48) + List of adapter,address pairs to assume to be present. Autodetection + of the target device will still be attempted. Use one of the more + specific force directives below if this doesn't detect the device. +* force_lm93: short array (min = 1, max = 48) + List of adapter,address pairs which are unquestionably assumed to contain + a 'lm93' chip +* ignore: short array (min = 1, max = 48) + List of adapter,address pairs not to scan +* ignore_range: short array (min = 1, max = 48) + List of adapter,start-addr,end-addr triples not to scan +* probe: short array (min = 1, max = 48) + List of adapter,address pairs to scan additionally +* probe_range: short array (min = 1, max = 48) + List of adapter,start-addr,end-addr triples to scan additionally + + +Hardware Description +-------------------- + +(from the datasheet) + +The LM93, hardware monitor, has a two wire digital interface compatible with +SMBus 2.0. Using an 8-bit ADC, the LM93 measures the temperature of two remote +diode connected transistors as well as its own die and 16 power supply +voltages. To set fan speed, the LM93 has two PWM outputs that are each +controlled by up to four temperature zones. The fancontrol algorithm is lookup +table based. The LM93 includes a digital filter that can be invoked to smooth +temperature readings for better control of fan speed. The LM93 has four +tachometer inputs to measure fan speed. Limit and status registers for all +measured values are included. The LM93 builds upon the functionality of +previous motherboard management ASICs and uses some of the LM85 s features +(i.e. smart tachometer mode). It also adds measurement and control support +for dynamic Vccp monitoring and PROCHOT. It is designed to monitor a dual +processor Xeon class motherboard with a minimum of external components. + + +Driver Description +------------------ + +This driver implements support for the National Semiconductor LM93. + + +User Interface +-------------- + +#PROCHOT: + +The LM93 can monitor two #PROCHOT signals. The results are found in the +sysfs files prochot1, prochot2, prochot1_avg, prochot2_avg, prochot1_max, +and prochot2_max. prochot1_max and prochot2_max contain the user limits +for #PROCHOT1 and #PROCHOT2, respectively. prochot1 and prochot2 contain +the current readings for the most recent complete time interval. The +value of prochot1_avg and prochot2_avg is something like a 2 period +exponential moving average (but not quite - check the datasheet). Note +that this third value is calculated by the chip itself. All values range +from 0-255 where 0 indicates no throttling, and 255 indicates > 99.6%. + +The monitoring intervals for the two #PROCHOT signals is also configurable. +These intervals can be found in the sysfs files prochot1_interval and +prochot2_interval. The values in these files specify the intervals for +#P1_PROCHOT and #P2_PROCHOT, respectively. Selecting a value not in this +list will cause the driver to use the next largest interval. The available +intervals are: + +#PROCHOT intervals: 0.73, 1.46, 2.9, 5.8, 11.7, 23.3, 46.6, 93.2, 186, 372 + +It is possible to configure the LM93 to logically short the two #PROCHOT +signals. I.e. when #P1_PROCHOT is asserted, the LM93 will automatically +assert #P2_PROCHOT, and vice-versa. This mode is enabled by writing a +non-zero integer to the sysfs file prochot_short. + +The LM93 can also override the #PROCHOT pins by driving a PWM signal onto +one or both of them. When overridden, the signal has a period of 3.56 mS, +a minimum pulse width of 5 clocks (at 22.5kHz => 6.25% duty cycle), and +a maximum pulse width of 80 clocks (at 22.5kHz => 99.88% duty cycle). + +The sysfs files prochot1_override and prochot2_override contain boolean +intgers which enable or disable the override function for #P1_PROCHOT and +#P2_PROCHOT, respectively. The sysfs file prochot_override_duty_cycle +contains a value controlling the duty cycle for the PWM signal used when +the override function is enabled. This value ranges from 0 to 15, with 0 +indicating minimum duty cycle and 15 indicating maximum. + +#VRD_HOT: + +The LM93 can monitor two #VRD_HOT signals. The results are found in the +sysfs files vrdhot1 and vrdhot2. There is one value per file: a boolean for +which 1 indicates #VRD_HOT is asserted and 0 indicates it is negated. These +files are read-only. + +Smart Tach Mode: + +(from the datasheet) + + If a fan is driven using a low-side drive PWM, the tachometer + output of the fan is corrupted. The LM93 includes smart tachometer + circuitry that allows an accurate tachometer reading to be + achieved despite the signal corruption. In smart tach mode all + four signals are measured within 4 seconds. + +Smart tach mode is enabled by the driver by writing 1 or 2 (associating the +the fan tachometer with a pwm) to the sysfs file fan<n>_smart_tach. A zero +will disable the function for that fan. Note that Smart tach mode cannot be +enabled if the PWM output frequency is 22500 Hz (see below). + +Manual PWM: + +The LM93 has a fixed or override mode for the two PWM outputs (although, there +are still some conditions that will override even this mode - see section +15.10.6 of the datasheet for details.) The sysfs files pwm1_override +and pwm2_override are used to enable this mode; each is a boolean integer +where 0 disables and 1 enables the manual control mode. The sysfs files pwm1 +and pwm2 are used to set the manual duty cycle; each is an integer (0-255) +where 0 is 0% duty cycle, and 255 is 100%. Note that the duty cycle values +are constrained by the hardware. Selecting a value which is not available +will cause the driver to use the next largest value. Also note: when manual +PWM mode is disabled, the value of pwm1 and pwm2 indicates the current duty +cycle chosen by the h/w. + +PWM Output Frequency: + +The LM93 supports several different frequencies for the PWM output channels. +The sysfs files pwm1_freq and pwm2_freq are used to select the frequency. The +frequency values are constrained by the hardware. Selecting a value which is +not available will cause the driver to use the next largest value. Also note +that this parameter has implications for the Smart Tach Mode (see above). + +PWM Output Frequencies: 12, 36, 48, 60, 72, 84, 96, 22500 (h/w default) + +Automatic PWM: + +The LM93 is capable of complex automatic fan control, with many different +points of configuration. To start, each PWM output can be bound to any +combination of eight control sources. The final PWM is the largest of all +individual control sources to which the PWM output is bound. + +The eight control sources are: temp1-temp4 (aka "zones" in the datasheet), +#PROCHOT 1 & 2, and #VRDHOT 1 & 2. The bindings are expressed as a bitmask +in the sysfs files pwm<n>_auto_channels, where a "1" enables the binding, and + a "0" disables it. The h/w default is 0x0f (all temperatures bound). + + 0x01 - Temp 1 + 0x02 - Temp 2 + 0x04 - Temp 3 + 0x08 - Temp 4 + 0x10 - #PROCHOT 1 + 0x20 - #PROCHOT 2 + 0x40 - #VRDHOT 1 + 0x80 - #VRDHOT 2 + +The function y = f(x) takes a source temperature x to a PWM output y. This +function of the LM93 is derived from a base temperature and a table of 12 +temperature offsets. The base temperature is expressed in degrees C in the +sysfs files temp<n>_auto_base. The offsets are expressed in cumulative +degrees C, with the value of offset <i> for temperature value <n> being +contained in the file temp<n>_auto_offset<i>. E.g. if the base temperature +is 40C: + + offset # temp<n>_auto_offset<i> range pwm + 1 0 - 25.00% + 2 0 - 28.57% + 3 1 40C - 41C 32.14% + 4 1 41C - 42C 35.71% + 5 2 42C - 44C 39.29% + 6 2 44C - 46C 42.86% + 7 2 48C - 50C 46.43% + 8 2 50C - 52C 50.00% + 9 2 52C - 54C 53.57% + 10 2 54C - 56C 57.14% + 11 2 56C - 58C 71.43% + 12 2 58C - 60C 85.71% + > 60C 100.00% + +Valid offsets are in the range 0C <= x <= 7.5C in 0.5C increments. + +There is an independent base temperature for each temperature channel. Note, +however, there are only two tables of offsets: one each for temp[12] and +temp[34]. Therefore, any change to e.g. temp1_auto_offset<i> will also +affect temp2_auto_offset<i>. + +The LM93 can also apply hysteresis to the offset table, to prevent unwanted +oscillation between two steps in the offsets table. These values are found in +the sysfs files temp<n>_auto_offset_hyst. The value in this file has the +same representation as in temp<n>_auto_offset<i>. + +If a temperature reading falls below the base value for that channel, the LM93 +will use the minimum PWM value. These values are found in the sysfs files +temp<n>_auto_pwm_min. Note, there are only two minimums: one each for temp[12] +and temp[34]. Therefore, any change to e.g. temp1_auto_pwm_min will also +affect temp2_auto_pwm_min. + +PWM Spin-Up Cycle: + +A spin-up cycle occurs when a PWM output is commanded from 0% duty cycle to +some value > 0%. The LM93 supports a minimum duty cycle during spin-up. These +values are found in the sysfs files pwm<n>_auto_spinup_min. The value in this +file has the same representation as other PWM duty cycle values. The +duration of the spin-up cycle is also configurable. These values are found in +the sysfs files pwm<n>_auto_spinup_time. The value in this file is +the spin-up time in seconds. The available spin-up times are constrained by +the hardware. Selecting a value which is not available will cause the driver +to use the next largest value. + +Spin-up Durations: 0 (disabled, h/w default), 0.1, 0.25, 0.4, 0.7, 1.0, + 2.0, 4.0 + +#PROCHOT and #VRDHOT PWM Ramping: + +If the #PROCHOT or #VRDHOT signals are asserted while bound to a PWM output +channel, the LM93 will ramp the PWM output up to 100% duty cycle in discrete +steps. The duration of each step is configurable. There are two files, with +one value each in seconds: pwm_auto_prochot_ramp and pwm_auto_vrdhot_ramp. +The available ramp times are constrained by the hardware. Selecting a value +which is not available will cause the driver to use the next largest value. + +Ramp Times: 0 (disabled, h/w default) to 0.75 in 0.05 second intervals + +Fan Boost: + +For each temperature channel, there is a boost temperature: if the channel +exceeds this limit, the LM93 will immediately drive both PWM outputs to 100%. +This limit is expressed in degrees C in the sysfs files temp<n>_auto_boost. +There is also a hysteresis temperature for this function: after the boost +limit is reached, the temperature channel must drop below this value before +the boost function is disabled. This temperature is also expressed in degrees +C in the sysfs files temp<n>_auto_boost_hyst. + +GPIO Pins: + +The LM93 can monitor the logic level of four dedicated GPIO pins as well as the +four tach input pins. GPIO0-GPIO3 correspond to (fan) tach 1-4, respectively. +All eight GPIOs are read by reading the bitmask in the sysfs file gpio. The +LSB is GPIO0, and the MSB is GPIO7. + + +LM93 Unique sysfs Files +----------------------- + + file description + ------------------------------------------------------------- + + prochot<n> current #PROCHOT % + + prochot<n>_avg moving average #PROCHOT % + + prochot<n>_max limit #PROCHOT % + + prochot_short enable or disable logical #PROCHOT pin short + + prochot<n>_override force #PROCHOT assertion as PWM + + prochot_override_duty_cycle + duty cycle for the PWM signal used when + #PROCHOT is overridden + + prochot<n>_interval #PROCHOT PWM sampling interval + + vrdhot<n> 0 means negated, 1 means asserted + + fan<n>_smart_tach enable or disable smart tach mode + + pwm<n>_auto_channels select control sources for PWM outputs + + pwm<n>_auto_spinup_min minimum duty cycle during spin-up + + pwm<n>_auto_spinup_time duration of spin-up + + pwm_auto_prochot_ramp ramp time per step when #PROCHOT asserted + + pwm_auto_vrdhot_ramp ramp time per step when #VRDHOT asserted + + temp<n>_auto_base temperature channel base + + temp<n>_auto_offset[1-12] + temperature channel offsets + + temp<n>_auto_offset_hyst + temperature channel offset hysteresis + + temp<n>_auto_boost temperature channel boost (PWMs to 100%) limit + + temp<n>_auto_boost_hyst temperature channel boost hysteresis + + gpio input state of 8 GPIO pins; read-only + + +Sample Configuration File +------------------------- + +Here is a sample LM93 chip config for sensors.conf: + +---------- cut here ---------- +chip "lm93-*" + +# VOLTAGE INPUTS + + # labels and scaling based on datasheet recommendations + label in1 "+12V1" + compute in1 @ * 12.945, @ / 12.945 + set in1_min 12 * 0.90 + set in1_max 12 * 1.10 + + label in2 "+12V2" + compute in2 @ * 12.945, @ / 12.945 + set in2_min 12 * 0.90 + set in2_max 12 * 1.10 + + label in3 "+12V3" + compute in3 @ * 12.945, @ / 12.945 + set in3_min 12 * 0.90 + set in3_max 12 * 1.10 + + label in4 "FSB_Vtt" + + label in5 "3GIO" + + label in6 "ICH_Core" + + label in7 "Vccp1" + + label in8 "Vccp2" + + label in9 "+3.3V" + set in9_min 3.3 * 0.90 + set in9_max 3.3 * 1.10 + + label in10 "+5V" + set in10_min 5.0 * 0.90 + set in10_max 5.0 * 1.10 + + label in11 "SCSI_Core" + + label in12 "Mem_Core" + + label in13 "Mem_Vtt" + + label in14 "Gbit_Core" + + # Assuming R1/R2 = 4.1143, and 3.3V reference + # -12V = (4.1143 + 1) * (@ - 3.3) + 3.3 + label in15 "-12V" + compute in15 @ * 5.1143 - 13.57719, (@ + 13.57719) / 5.1143 + set in15_min -12 * 0.90 + set in15_max -12 * 1.10 + + label in16 "+3.3VSB" + set in16_min 3.3 * 0.90 + set in16_max 3.3 * 1.10 + +# TEMPERATURE INPUTS + + label temp1 "CPU1" + label temp2 "CPU2" + label temp3 "LM93" + +# TACHOMETER INPUTS + + label fan1 "Fan1" + set fan1_min 3000 + label fan2 "Fan2" + set fan2_min 3000 + label fan3 "Fan3" + set fan3_min 3000 + label fan4 "Fan4" + set fan4_min 3000 + +# PWM OUTPUTS + + label pwm1 "CPU1" + label pwm2 "CPU2" + diff --git a/Documentation/hwmon/smsc47b397 b/Documentation/hwmon/smsc47b397 index 20682f15ae4..3a43b694892 100644 --- a/Documentation/hwmon/smsc47b397 +++ b/Documentation/hwmon/smsc47b397 @@ -4,6 +4,7 @@ Kernel driver smsc47b397 Supported chips: * SMSC LPC47B397-NC * SMSC SCH5307-NS + * SMSC SCH5317 Prefix: 'smsc47b397' Addresses scanned: none, address read from Super I/O config space Datasheet: In this file @@ -18,8 +19,8 @@ The following specification describes the SMSC LPC47B397-NC[1] sensor chip provided by Craig Kelly (In-Store Broadcast Network) and edited/corrected by Mark M. Hoffman <mhoffman@lightlink.com>. -[1] And SMSC SCH5307-NS, which has a different device ID but is otherwise -compatible. +[1] And SMSC SCH5307-NS and SCH5317, which have different device IDs but are +otherwise compatible. * * * * * @@ -131,7 +132,7 @@ OUT DX,AL The registers of interest for identifying the SIO on the dc7100 are Device ID (0x20) and Device Rev (0x21). -The Device ID will read 0x6F (for SCH5307-NS, 0x81) +The Device ID will read 0x6F (0x81 for SCH5307-NS, and 0x85 for SCH5317) The Device Rev currently reads 0x01 Obtaining the HWM Base Address. diff --git a/Documentation/hwmon/sysfs-interface b/Documentation/hwmon/sysfs-interface index a9a18ad0d17..b3a9e1b9dbd 100644 --- a/Documentation/hwmon/sysfs-interface +++ b/Documentation/hwmon/sysfs-interface @@ -172,11 +172,10 @@ pwm[1-*] Pulse width modulation fan control. 255 is max or 100%. pwm[1-*]_enable - Switch PWM on and off. - Not always present even if pwmN is. - 0: turn off - 1: turn on in manual mode - 2+: turn on in automatic mode + Fan speed control method: + 0: no fan speed control (i.e. fan at full speed) + 1: manual fan speed control enabled (using pwm[1-*]) + 2+: automatic fan speed control enabled Check individual chip documentation files for automatic mode details. RW @@ -343,9 +342,9 @@ to notify open diodes, unconnected fans etc. where the hardware supports it. When this boolean has value 1, the measurement for that channel should not be trusted. -in[0-*]_input_fault -fan[1-*]_input_fault -temp[1-*]_input_fault +in[0-*]_fault +fan[1-*]_fault +temp[1-*]_fault Input fault condition 0: no fault occured 1: fault condition diff --git a/Documentation/hwmon/w83627ehf b/Documentation/hwmon/w83627ehf index 030fac6cec7..ccc2bcb6106 100644 --- a/Documentation/hwmon/w83627ehf +++ b/Documentation/hwmon/w83627ehf @@ -22,9 +22,9 @@ This driver implements support for the Winbond W83627EHF, W83627EHG, and W83627DHG super I/O chips. We will refer to them collectively as Winbond chips. The chips implement three temperature sensors, five fan rotation -speed sensors, ten analog voltage sensors (only nine for the 627DHG), alarms -with beep warnings (control unimplemented), and some automatic fan regulation -strategies (plus manual fan control mode). +speed sensors, ten analog voltage sensors (only nine for the 627DHG), one +VID (6 pins), alarms with beep warnings (control unimplemented), and +some automatic fan regulation strategies (plus manual fan control mode). Temperatures are measured in degrees Celsius and measurement resolution is 1 degC for temp1 and 0.5 degC for temp2 and temp3. An alarm is triggered when diff --git a/Documentation/kbuild/makefiles.txt b/Documentation/kbuild/makefiles.txt index bb5306e9a5c..e08ef8759a0 100644 --- a/Documentation/kbuild/makefiles.txt +++ b/Documentation/kbuild/makefiles.txt @@ -501,6 +501,20 @@ more details, with real examples. The third parameter may be a text as in this example, but it may also be an expanded variable or a macro. + cc-fullversion + cc-fullversion is useful when the exact version of gcc is needed. + One typical use-case is when a specific GCC version is broken. + cc-fullversion points out a more specific version than cc-version does. + + Example: + #arch/powerpc/Makefile + $(Q)if test "$(call cc-fullversion)" = "040200" ; then \ + echo -n '*** GCC-4.2.0 cannot compile the 64-bit powerpc ' ; \ + false ; \ + fi + + In this example for a specific GCC version the build will error out explaining + to the user why it stops. === 4 Host Program support diff --git a/Documentation/kprobes.txt b/Documentation/kprobes.txt index da5404ab756..cb12ae175aa 100644 --- a/Documentation/kprobes.txt +++ b/Documentation/kprobes.txt @@ -247,12 +247,6 @@ control to Kprobes.) If the probed function is declared asmlinkage, fastcall, or anything else that affects how args are passed, the handler's declaration must match. -NOTE: A macro JPROBE_ENTRY is provided to handle architecture-specific -aliasing of jp->entry. In the interest of portability, it is advised -to use: - - jp->entry = JPROBE_ENTRY(handler); - register_jprobe() returns 0 on success, or a negative errno otherwise. 4.3 register_kretprobe @@ -518,7 +512,7 @@ long jdo_fork(unsigned long clone_flags, unsigned long stack_start, } static struct jprobe my_jprobe = { - .entry = JPROBE_ENTRY(jdo_fork) + .entry = jdo_fork }; static int __init jprobe_init(void) diff --git a/Documentation/lguest/Makefile b/Documentation/lguest/Makefile new file mode 100644 index 00000000000..b9b9427376e --- /dev/null +++ b/Documentation/lguest/Makefile @@ -0,0 +1,27 @@ +# This creates the demonstration utility "lguest" which runs a Linux guest. + +# For those people that have a separate object dir, look there for .config +KBUILD_OUTPUT := ../.. +ifdef O + ifeq ("$(origin O)", "command line") + KBUILD_OUTPUT := $(O) + endif +endif +# We rely on CONFIG_PAGE_OFFSET to know where to put lguest binary. +include $(KBUILD_OUTPUT)/.config +LGUEST_GUEST_TOP := ($(CONFIG_PAGE_OFFSET) - 0x08000000) + +CFLAGS:=-Wall -Wmissing-declarations -Wmissing-prototypes -O3 \ + -static -DLGUEST_GUEST_TOP="$(LGUEST_GUEST_TOP)" -Wl,-T,lguest.lds +LDLIBS:=-lz + +all: lguest.lds lguest + +# The linker script on x86 is so complex the only way of creating one +# which will link our binary in the right place is to mangle the +# default one. +lguest.lds: + $(LD) --verbose | awk '/^==========/ { PRINT=1; next; } /SIZEOF_HEADERS/ { gsub(/0x[0-9A-F]*/, "$(LGUEST_GUEST_TOP)") } { if (PRINT) print $$0; }' > $@ + +clean: + rm -f lguest.lds lguest diff --git a/Documentation/lguest/lguest.c b/Documentation/lguest/lguest.c new file mode 100644 index 00000000000..1432b502a2d --- /dev/null +++ b/Documentation/lguest/lguest.c @@ -0,0 +1,1012 @@ +/* Simple program to layout "physical" memory for new lguest guest. + * Linked high to avoid likely physical memory. */ +#define _LARGEFILE64_SOURCE +#define _GNU_SOURCE +#include <stdio.h> +#include <string.h> +#include <unistd.h> +#include <err.h> +#include <stdint.h> +#include <stdlib.h> +#include <elf.h> +#include <sys/mman.h> +#include <sys/types.h> +#include <sys/stat.h> +#include <sys/wait.h> +#include <fcntl.h> +#include <stdbool.h> +#include <errno.h> +#include <ctype.h> +#include <sys/socket.h> +#include <sys/ioctl.h> +#include <sys/time.h> +#include <time.h> +#include <netinet/in.h> +#include <net/if.h> +#include <linux/sockios.h> +#include <linux/if_tun.h> +#include <sys/uio.h> +#include <termios.h> +#include <getopt.h> +#include <zlib.h> +typedef unsigned long long u64; +typedef uint32_t u32; +typedef uint16_t u16; +typedef uint8_t u8; +#include "../../include/linux/lguest_launcher.h" +#include "../../include/asm-i386/e820.h" + +#define PAGE_PRESENT 0x7 /* Present, RW, Execute */ +#define NET_PEERNUM 1 +#define BRIDGE_PFX "bridge:" +#ifndef SIOCBRADDIF +#define SIOCBRADDIF 0x89a2 /* add interface to bridge */ +#endif + +static bool verbose; +#define verbose(args...) \ + do { if (verbose) printf(args); } while(0) +static int waker_fd; + +struct device_list +{ + fd_set infds; + int max_infd; + + struct device *dev; + struct device **lastdev; +}; + +struct device +{ + struct device *next; + struct lguest_device_desc *desc; + void *mem; + + /* Watch this fd if handle_input non-NULL. */ + int fd; + bool (*handle_input)(int fd, struct device *me); + + /* Watch DMA to this key if handle_input non-NULL. */ + unsigned long watch_key; + u32 (*handle_output)(int fd, const struct iovec *iov, + unsigned int num, struct device *me); + + /* Device-specific data. */ + void *priv; +}; + +static int open_or_die(const char *name, int flags) +{ + int fd = open(name, flags); + if (fd < 0) + err(1, "Failed to open %s", name); + return fd; +} + +static void *map_zeroed_pages(unsigned long addr, unsigned int num) +{ + static int fd = -1; + + if (fd == -1) + fd = open_or_die("/dev/zero", O_RDONLY); + + if (mmap((void *)addr, getpagesize() * num, + PROT_READ|PROT_WRITE|PROT_EXEC, MAP_FIXED|MAP_PRIVATE, fd, 0) + != (void *)addr) + err(1, "Mmaping %u pages of /dev/zero @%p", num, (void *)addr); + return (void *)addr; +} + +/* Find magic string marking entry point, return entry point. */ +static unsigned long entry_point(void *start, void *end, + unsigned long page_offset) +{ + void *p; + + for (p = start; p < end; p++) + if (memcmp(p, "GenuineLguest", strlen("GenuineLguest")) == 0) + return (long)p + strlen("GenuineLguest") + page_offset; + + err(1, "Is this image a genuine lguest?"); +} + +/* Returns the entry point */ +static unsigned long map_elf(int elf_fd, const Elf32_Ehdr *ehdr, + unsigned long *page_offset) +{ + void *addr; + Elf32_Phdr phdr[ehdr->e_phnum]; + unsigned int i; + unsigned long start = -1UL, end = 0; + + /* Sanity checks. */ + if (ehdr->e_type != ET_EXEC + || ehdr->e_machine != EM_386 + || ehdr->e_phentsize != sizeof(Elf32_Phdr) + || ehdr->e_phnum < 1 || ehdr->e_phnum > 65536U/sizeof(Elf32_Phdr)) + errx(1, "Malformed elf header"); + + if (lseek(elf_fd, ehdr->e_phoff, SEEK_SET) < 0) + err(1, "Seeking to program headers"); + if (read(elf_fd, phdr, sizeof(phdr)) != sizeof(phdr)) + err(1, "Reading program headers"); + + *page_offset = 0; + /* We map the loadable segments at virtual addresses corresponding + * to their physical addresses (our virtual == guest physical). */ + for (i = 0; i < ehdr->e_phnum; i++) { + if (phdr[i].p_type != PT_LOAD) + continue; + + verbose("Section %i: size %i addr %p\n", + i, phdr[i].p_memsz, (void *)phdr[i].p_paddr); + + /* We expect linear address space. */ + if (!*page_offset) + *page_offset = phdr[i].p_vaddr - phdr[i].p_paddr; + else if (*page_offset != phdr[i].p_vaddr - phdr[i].p_paddr) + errx(1, "Page offset of section %i different", i); + + if (phdr[i].p_paddr < start) + start = phdr[i].p_paddr; + if (phdr[i].p_paddr + phdr[i].p_filesz > end) + end = phdr[i].p_paddr + phdr[i].p_filesz; + + /* We map everything private, writable. */ + addr = mmap((void *)phdr[i].p_paddr, + phdr[i].p_filesz, + PROT_READ|PROT_WRITE|PROT_EXEC, + MAP_FIXED|MAP_PRIVATE, + elf_fd, phdr[i].p_offset); + if (addr != (void *)phdr[i].p_paddr) + err(1, "Mmaping vmlinux seg %i gave %p not %p", + i, addr, (void *)phdr[i].p_paddr); + } + + return entry_point((void *)start, (void *)end, *page_offset); +} + +/* This is amazingly reliable. */ +static unsigned long intuit_page_offset(unsigned char *img, unsigned long len) +{ + unsigned int i, possibilities[256] = { 0 }; + + for (i = 0; i + 4 < len; i++) { + /* mov 0xXXXXXXXX,%eax */ + if (img[i] == 0xA1 && ++possibilities[img[i+4]] > 3) + return (unsigned long)img[i+4] << 24; + } + errx(1, "could not determine page offset"); +} + +static unsigned long unpack_bzimage(int fd, unsigned long *page_offset) +{ + gzFile f; + int ret, len = 0; + void *img = (void *)0x100000; + + f = gzdopen(fd, "rb"); + while ((ret = gzread(f, img + len, 65536)) > 0) + len += ret; + if (ret < 0) + err(1, "reading image from bzImage"); + + verbose("Unpacked size %i addr %p\n", len, img); + *page_offset = intuit_page_offset(img, len); + + return entry_point(img, img + len, *page_offset); +} + +static unsigned long load_bzimage(int fd, unsigned long *page_offset) +{ + unsigned char c; + int state = 0; + + /* Ugly brute force search for gzip header. */ + while (read(fd, &c, 1) == 1) { + switch (state) { + case 0: + if (c == 0x1F) + state++; + break; + case 1: + if (c == 0x8B) + state++; + else + state = 0; + break; + case 2 ... 8: + state++; + break; + case 9: + lseek(fd, -10, SEEK_CUR); + if (c != 0x03) /* Compressed under UNIX. */ + state = -1; + else + return unpack_bzimage(fd, page_offset); + } + } + errx(1, "Could not find kernel in bzImage"); +} + +static unsigned long load_kernel(int fd, unsigned long *page_offset) +{ + Elf32_Ehdr hdr; + + if (read(fd, &hdr, sizeof(hdr)) != sizeof(hdr)) + err(1, "Reading kernel"); + + if (memcmp(hdr.e_ident, ELFMAG, SELFMAG) == 0) + return map_elf(fd, &hdr, page_offset); + + return load_bzimage(fd, page_offset); +} + +static inline unsigned long page_align(unsigned long addr) +{ + return ((addr + getpagesize()-1) & ~(getpagesize()-1)); +} + +/* initrd gets loaded at top of memory: return length. */ +static unsigned long load_initrd(const char *name, unsigned long mem) +{ + int ifd; + struct stat st; + unsigned long len; + void *iaddr; + + ifd = open_or_die(name, O_RDONLY); + if (fstat(ifd, &st) < 0) + err(1, "fstat() on initrd '%s'", name); + + len = page_align(st.st_size); + iaddr = mmap((void *)mem - len, st.st_size, + PROT_READ|PROT_EXEC|PROT_WRITE, + MAP_FIXED|MAP_PRIVATE, ifd, 0); + if (iaddr != (void *)mem - len) + err(1, "Mmaping initrd '%s' returned %p not %p", + name, iaddr, (void *)mem - len); + close(ifd); + verbose("mapped initrd %s size=%lu @ %p\n", name, st.st_size, iaddr); + return len; +} + +static unsigned long setup_pagetables(unsigned long mem, + unsigned long initrd_size, + unsigned long page_offset) +{ + u32 *pgdir, *linear; + unsigned int mapped_pages, i, linear_pages; + unsigned int ptes_per_page = getpagesize()/sizeof(u32); + + /* If we can map all of memory above page_offset, we do so. */ + if (mem <= -page_offset) + mapped_pages = mem/getpagesize(); + else + mapped_pages = -page_offset/getpagesize(); + + /* Each linear PTE page can map ptes_per_page pages. */ + linear_pages = (mapped_pages + ptes_per_page-1)/ptes_per_page; + + /* We lay out top-level then linear mapping immediately below initrd */ + pgdir = (void *)mem - initrd_size - getpagesize(); + linear = (void *)pgdir - linear_pages*getpagesize(); + + for (i = 0; i < mapped_pages; i++) + linear[i] = ((i * getpagesize()) | PAGE_PRESENT); + + /* Now set up pgd so that this memory is at page_offset */ + for (i = 0; i < mapped_pages; i += ptes_per_page) { + pgdir[(i + page_offset/getpagesize())/ptes_per_page] + = (((u32)linear + i*sizeof(u32)) | PAGE_PRESENT); + } + + verbose("Linear mapping of %u pages in %u pte pages at %p\n", + mapped_pages, linear_pages, linear); + + return (unsigned long)pgdir; +} + +static void concat(char *dst, char *args[]) +{ + unsigned int i, len = 0; + + for (i = 0; args[i]; i++) { + strcpy(dst+len, args[i]); + strcat(dst+len, " "); + len += strlen(args[i]) + 1; + } + /* In case it's empty. */ + dst[len] = '\0'; +} + +static int tell_kernel(u32 pgdir, u32 start, u32 page_offset) +{ + u32 args[] = { LHREQ_INITIALIZE, + LGUEST_GUEST_TOP/getpagesize(), /* Just below us */ + pgdir, start, page_offset }; + int fd; + + fd = open_or_die("/dev/lguest", O_RDWR); + if (write(fd, args, sizeof(args)) < 0) + err(1, "Writing to /dev/lguest"); + return fd; +} + +static void set_fd(int fd, struct device_list *devices) +{ + FD_SET(fd, &devices->infds); + if (fd > devices->max_infd) + devices->max_infd = fd; +} + +/* When input arrives, we tell the kernel to kick lguest out with -EAGAIN. */ +static void wake_parent(int pipefd, int lguest_fd, struct device_list *devices) +{ + set_fd(pipefd, devices); + + for (;;) { + fd_set rfds = devices->infds; + u32 args[] = { LHREQ_BREAK, 1 }; + + select(devices->max_infd+1, &rfds, NULL, NULL, NULL); + if (FD_ISSET(pipefd, &rfds)) { + int ignorefd; + if (read(pipefd, &ignorefd, sizeof(ignorefd)) == 0) + exit(0); + FD_CLR(ignorefd, &devices->infds); + } else + write(lguest_fd, args, sizeof(args)); + } +} + +static int setup_waker(int lguest_fd, struct device_list *device_list) +{ + int pipefd[2], child; + + pipe(pipefd); + child = fork(); + if (child == -1) + err(1, "forking"); + + if (child == 0) { + close(pipefd[1]); + wake_parent(pipefd[0], lguest_fd, device_list); + } + close(pipefd[0]); + + return pipefd[1]; +} + +static void *_check_pointer(unsigned long addr, unsigned int size, + unsigned int line) +{ + if (addr >= LGUEST_GUEST_TOP || addr + size >= LGUEST_GUEST_TOP) + errx(1, "%s:%i: Invalid address %li", __FILE__, line, addr); + return (void *)addr; +} +#define check_pointer(addr,size) _check_pointer(addr, size, __LINE__) + +/* Returns pointer to dma->used_len */ +static u32 *dma2iov(unsigned long dma, struct iovec iov[], unsigned *num) +{ + unsigned int i; + struct lguest_dma *udma; + + udma = check_pointer(dma, sizeof(*udma)); + for (i = 0; i < LGUEST_MAX_DMA_SECTIONS; i++) { + if (!udma->len[i]) + break; + + iov[i].iov_base = check_pointer(udma->addr[i], udma->len[i]); + iov[i].iov_len = udma->len[i]; + } + *num = i; + return &udma->used_len; +} + +static u32 *get_dma_buffer(int fd, void *key, + struct iovec iov[], unsigned int *num, u32 *irq) +{ + u32 buf[] = { LHREQ_GETDMA, (u32)key }; + unsigned long udma; + u32 *res; + + udma = write(fd, buf, sizeof(buf)); + if (udma == (unsigned long)-1) + return NULL; + + /* Kernel stashes irq in ->used_len. */ + res = dma2iov(udma, iov, num); + *irq = *res; + return res; +} + +static void trigger_irq(int fd, u32 irq) +{ + u32 buf[] = { LHREQ_IRQ, irq }; + if (write(fd, buf, sizeof(buf)) != 0) + err(1, "Triggering irq %i", irq); +} + +static void discard_iovec(struct iovec *iov, unsigned int *num) +{ + static char discard_buf[1024]; + *num = 1; + iov->iov_base = discard_buf; + iov->iov_len = sizeof(discard_buf); +} + +static struct termios orig_term; +static void restore_term(void) +{ + tcsetattr(STDIN_FILENO, TCSANOW, &orig_term); +} + +struct console_abort +{ + int count; + struct timeval start; +}; + +/* We DMA input to buffer bound at start of console page. */ +static bool handle_console_input(int fd, struct device *dev) +{ + u32 irq = 0, *lenp; + int len; + unsigned int num; + struct iovec iov[LGUEST_MAX_DMA_SECTIONS]; + struct console_abort *abort = dev->priv; + + lenp = get_dma_buffer(fd, dev->mem, iov, &num, &irq); + if (!lenp) { + warn("console: no dma buffer!"); + discard_iovec(iov, &num); + } + + len = readv(dev->fd, iov, num); + if (len <= 0) { + warnx("Failed to get console input, ignoring console."); + len = 0; + } + + if (lenp) { + *lenp = len; + trigger_irq(fd, irq); + } + + /* Three ^C within one second? Exit. */ + if (len == 1 && ((char *)iov[0].iov_base)[0] == 3) { + if (!abort->count++) + gettimeofday(&abort->start, NULL); + else if (abort->count == 3) { + struct timeval now; + gettimeofday(&now, NULL); + if (now.tv_sec <= abort->start.tv_sec+1) { + /* Make sure waker is not blocked in BREAK */ + u32 args[] = { LHREQ_BREAK, 0 }; + close(waker_fd); + write(fd, args, sizeof(args)); + exit(2); + } + abort->count = 0; + } + } else + abort->count = 0; + + if (!len) { + restore_term(); + return false; + } + return true; +} + +static u32 handle_console_output(int fd, const struct iovec *iov, + unsigned num, struct device*dev) +{ + return writev(STDOUT_FILENO, iov, num); +} + +static u32 handle_tun_output(int fd, const struct iovec *iov, + unsigned num, struct device *dev) +{ + /* Now we've seen output, we should warn if we can't get buffers. */ + *(bool *)dev->priv = true; + return writev(dev->fd, iov, num); +} + +static unsigned long peer_offset(unsigned int peernum) +{ + return 4 * peernum; +} + +static bool handle_tun_input(int fd, struct device *dev) +{ + u32 irq = 0, *lenp; + int len; + unsigned num; + struct iovec iov[LGUEST_MAX_DMA_SECTIONS]; + + lenp = get_dma_buffer(fd, dev->mem+peer_offset(NET_PEERNUM), iov, &num, + &irq); + if (!lenp) { + if (*(bool *)dev->priv) + warn("network: no dma buffer!"); + discard_iovec(iov, &num); + } + + len = readv(dev->fd, iov, num); + if (len <= 0) + err(1, "reading network"); + if (lenp) { + *lenp = len; + trigger_irq(fd, irq); + } + verbose("tun input packet len %i [%02x %02x] (%s)\n", len, + ((u8 *)iov[0].iov_base)[0], ((u8 *)iov[0].iov_base)[1], + lenp ? "sent" : "discarded"); + return true; +} + +static u32 handle_block_output(int fd, const struct iovec *iov, + unsigned num, struct device *dev) +{ + struct lguest_block_page *p = dev->mem; + u32 irq, *lenp; + unsigned int len, reply_num; + struct iovec reply[LGUEST_MAX_DMA_SECTIONS]; + off64_t device_len, off = (off64_t)p->sector * 512; + + device_len = *(off64_t *)dev->priv; + + if (off >= device_len) + err(1, "Bad offset %llu vs %llu", off, device_len); + if (lseek64(dev->fd, off, SEEK_SET) != off) + err(1, "Bad seek to sector %i", p->sector); + + verbose("Block: %s at offset %llu\n", p->type ? "WRITE" : "READ", off); + + lenp = get_dma_buffer(fd, dev->mem, reply, &reply_num, &irq); + if (!lenp) + err(1, "Block request didn't give us a dma buffer"); + + if (p->type) { + len = writev(dev->fd, iov, num); + if (off + len > device_len) { + ftruncate(dev->fd, device_len); + errx(1, "Write past end %llu+%u", off, len); + } + *lenp = 0; + } else { + len = readv(dev->fd, reply, reply_num); + *lenp = len; + } + + p->result = 1 + (p->bytes != len); + trigger_irq(fd, irq); + return 0; +} + +static void handle_output(int fd, unsigned long dma, unsigned long key, + struct device_list *devices) +{ + struct device *i; + u32 *lenp; + struct iovec iov[LGUEST_MAX_DMA_SECTIONS]; + unsigned num = 0; + + lenp = dma2iov(dma, iov, &num); + for (i = devices->dev; i; i = i->next) { + if (i->handle_output && key == i->watch_key) { + *lenp = i->handle_output(fd, iov, num, i); + return; + } + } + warnx("Pending dma %p, key %p", (void *)dma, (void *)key); +} + +static void handle_input(int fd, struct device_list *devices) +{ + struct timeval poll = { .tv_sec = 0, .tv_usec = 0 }; + + for (;;) { + struct device *i; + fd_set fds = devices->infds; + + if (select(devices->max_infd+1, &fds, NULL, NULL, &poll) == 0) + break; + + for (i = devices->dev; i; i = i->next) { + if (i->handle_input && FD_ISSET(i->fd, &fds)) { + if (!i->handle_input(fd, i)) { + FD_CLR(i->fd, &devices->infds); + /* Tell waker to ignore it too... */ + write(waker_fd, &i->fd, sizeof(i->fd)); + } + } + } + } +} + +static struct lguest_device_desc *new_dev_desc(u16 type, u16 features, + u16 num_pages) +{ + static unsigned long top = LGUEST_GUEST_TOP; + struct lguest_device_desc *desc; + + desc = malloc(sizeof(*desc)); + desc->type = type; + desc->num_pages = num_pages; + desc->features = features; + desc->status = 0; + if (num_pages) { + top -= num_pages*getpagesize(); + map_zeroed_pages(top, num_pages); + desc->pfn = top / getpagesize(); + } else + desc->pfn = 0; + return desc; +} + +static struct device *new_device(struct device_list *devices, + u16 type, u16 num_pages, u16 features, + int fd, + bool (*handle_input)(int, struct device *), + unsigned long watch_off, + u32 (*handle_output)(int, + const struct iovec *, + unsigned, + struct device *)) +{ + struct device *dev = malloc(sizeof(*dev)); + + /* Append to device list. */ + *devices->lastdev = dev; + dev->next = NULL; + devices->lastdev = &dev->next; + + dev->fd = fd; + if (handle_input) + set_fd(dev->fd, devices); + dev->desc = new_dev_desc(type, features, num_pages); + dev->mem = (void *)(dev->desc->pfn * getpagesize()); + dev->handle_input = handle_input; + dev->watch_key = (unsigned long)dev->mem + watch_off; + dev->handle_output = handle_output; + return dev; +} + +static void setup_console(struct device_list *devices) +{ + struct device *dev; + + if (tcgetattr(STDIN_FILENO, &orig_term) == 0) { + struct termios term = orig_term; + term.c_lflag &= ~(ISIG|ICANON|ECHO); + tcsetattr(STDIN_FILENO, TCSANOW, &term); + atexit(restore_term); + } + + /* We don't currently require a page for the console. */ + dev = new_device(devices, LGUEST_DEVICE_T_CONSOLE, 0, 0, + STDIN_FILENO, handle_console_input, + LGUEST_CONSOLE_DMA_KEY, handle_console_output); + dev->priv = malloc(sizeof(struct console_abort)); + ((struct console_abort *)dev->priv)->count = 0; + verbose("device %p: console\n", + (void *)(dev->desc->pfn * getpagesize())); +} + +static void setup_block_file(const char *filename, struct device_list *devices) +{ + int fd; + struct device *dev; + off64_t *device_len; + struct lguest_block_page *p; + + fd = open_or_die(filename, O_RDWR|O_LARGEFILE|O_DIRECT); + dev = new_device(devices, LGUEST_DEVICE_T_BLOCK, 1, + LGUEST_DEVICE_F_RANDOMNESS, + fd, NULL, 0, handle_block_output); + device_len = dev->priv = malloc(sizeof(*device_len)); + *device_len = lseek64(fd, 0, SEEK_END); + p = dev->mem; + + p->num_sectors = *device_len/512; + verbose("device %p: block %i sectors\n", + (void *)(dev->desc->pfn * getpagesize()), p->num_sectors); +} + +/* We use fnctl locks to reserve network slots (autocleanup!) */ +static unsigned int find_slot(int netfd, const char *filename) +{ + struct flock fl; + + fl.l_type = F_WRLCK; + fl.l_whence = SEEK_SET; + fl.l_len = 1; + for (fl.l_start = 0; + fl.l_start < getpagesize()/sizeof(struct lguest_net); + fl.l_start++) { + if (fcntl(netfd, F_SETLK, &fl) == 0) + return fl.l_start; + } + errx(1, "No free slots in network file %s", filename); +} + +static void setup_net_file(const char *filename, + struct device_list *devices) +{ + int netfd; + struct device *dev; + + netfd = open(filename, O_RDWR, 0); + if (netfd < 0) { + if (errno == ENOENT) { + netfd = open(filename, O_RDWR|O_CREAT, 0600); + if (netfd >= 0) { + char page[getpagesize()]; + memset(page, 0, sizeof(page)); + write(netfd, page, sizeof(page)); + } + } + if (netfd < 0) + err(1, "cannot open net file '%s'", filename); + } + + dev = new_device(devices, LGUEST_DEVICE_T_NET, 1, + find_slot(netfd, filename)|LGUEST_NET_F_NOCSUM, + -1, NULL, 0, NULL); + + /* We overwrite the /dev/zero mapping with the actual file. */ + if (mmap(dev->mem, getpagesize(), PROT_READ|PROT_WRITE, + MAP_FIXED|MAP_SHARED, netfd, 0) != dev->mem) + err(1, "could not mmap '%s'", filename); + verbose("device %p: shared net %s, peer %i\n", + (void *)(dev->desc->pfn * getpagesize()), filename, + dev->desc->features & ~LGUEST_NET_F_NOCSUM); +} + +static u32 str2ip(const char *ipaddr) +{ + unsigned int byte[4]; + + sscanf(ipaddr, "%u.%u.%u.%u", &byte[0], &byte[1], &byte[2], &byte[3]); + return (byte[0] << 24) | (byte[1] << 16) | (byte[2] << 8) | byte[3]; +} + +/* adapted from libbridge */ +static void add_to_bridge(int fd, const char *if_name, const char *br_name) +{ + int ifidx; + struct ifreq ifr; + + if (!*br_name) + errx(1, "must specify bridge name"); + + ifidx = if_nametoindex(if_name); + if (!ifidx) + errx(1, "interface %s does not exist!", if_name); + + strncpy(ifr.ifr_name, br_name, IFNAMSIZ); + ifr.ifr_ifindex = ifidx; + if (ioctl(fd, SIOCBRADDIF, &ifr) < 0) + err(1, "can't add %s to bridge %s", if_name, br_name); +} + +static void configure_device(int fd, const char *devname, u32 ipaddr, + unsigned char hwaddr[6]) +{ + struct ifreq ifr; + struct sockaddr_in *sin = (struct sockaddr_in *)&ifr.ifr_addr; + + memset(&ifr, 0, sizeof(ifr)); + strcpy(ifr.ifr_name, devname); + sin->sin_family = AF_INET; + sin->sin_addr.s_addr = htonl(ipaddr); + if (ioctl(fd, SIOCSIFADDR, &ifr) != 0) + err(1, "Setting %s interface address", devname); + ifr.ifr_flags = IFF_UP; + if (ioctl(fd, SIOCSIFFLAGS, &ifr) != 0) + err(1, "Bringing interface %s up", devname); + + if (ioctl(fd, SIOCGIFHWADDR, &ifr) != 0) + err(1, "getting hw address for %s", devname); + + memcpy(hwaddr, ifr.ifr_hwaddr.sa_data, 6); +} + +static void setup_tun_net(const char *arg, struct device_list *devices) +{ + struct device *dev; + struct ifreq ifr; + int netfd, ipfd; + u32 ip; + const char *br_name = NULL; + + netfd = open_or_die("/dev/net/tun", O_RDWR); + memset(&ifr, 0, sizeof(ifr)); + ifr.ifr_flags = IFF_TAP | IFF_NO_PI; + strcpy(ifr.ifr_name, "tap%d"); + if (ioctl(netfd, TUNSETIFF, &ifr) != 0) + err(1, "configuring /dev/net/tun"); + ioctl(netfd, TUNSETNOCSUM, 1); + + /* You will be peer 1: we should create enough jitter to randomize */ + dev = new_device(devices, LGUEST_DEVICE_T_NET, 1, + NET_PEERNUM|LGUEST_DEVICE_F_RANDOMNESS, netfd, + handle_tun_input, peer_offset(0), handle_tun_output); + dev->priv = malloc(sizeof(bool)); + *(bool *)dev->priv = false; + + ipfd = socket(PF_INET, SOCK_DGRAM, IPPROTO_IP); + if (ipfd < 0) + err(1, "opening IP socket"); + + if (!strncmp(BRIDGE_PFX, arg, strlen(BRIDGE_PFX))) { + ip = INADDR_ANY; + br_name = arg + strlen(BRIDGE_PFX); + add_to_bridge(ipfd, ifr.ifr_name, br_name); + } else + ip = str2ip(arg); + + /* We are peer 0, ie. first slot. */ + configure_device(ipfd, ifr.ifr_name, ip, dev->mem); + + /* Set "promisc" bit: we want every single packet. */ + *((u8 *)dev->mem) |= 0x1; + + close(ipfd); + + verbose("device %p: tun net %u.%u.%u.%u\n", + (void *)(dev->desc->pfn * getpagesize()), + (u8)(ip>>24), (u8)(ip>>16), (u8)(ip>>8), (u8)ip); + if (br_name) + verbose("attached to bridge: %s\n", br_name); +} + +/* Now we know how much memory we have, we copy in device descriptors */ +static void map_device_descriptors(struct device_list *devs, unsigned long mem) +{ + struct device *i; + unsigned int num; + struct lguest_device_desc *descs; + + /* Device descriptor array sits just above top of normal memory */ + descs = map_zeroed_pages(mem, 1); + + for (i = devs->dev, num = 0; i; i = i->next, num++) { + if (num == LGUEST_MAX_DEVICES) + errx(1, "too many devices"); + verbose("Device %i: %s\n", num, + i->desc->type == LGUEST_DEVICE_T_NET ? "net" + : i->desc->type == LGUEST_DEVICE_T_CONSOLE ? "console" + : i->desc->type == LGUEST_DEVICE_T_BLOCK ? "block" + : "unknown"); + descs[num] = *i->desc; + free(i->desc); + i->desc = &descs[num]; + } +} + +static void __attribute__((noreturn)) +run_guest(int lguest_fd, struct device_list *device_list) +{ + for (;;) { + u32 args[] = { LHREQ_BREAK, 0 }; + unsigned long arr[2]; + int readval; + + /* We read from the /dev/lguest device to run the Guest. */ + readval = read(lguest_fd, arr, sizeof(arr)); + + if (readval == sizeof(arr)) { + handle_output(lguest_fd, arr[0], arr[1], device_list); + continue; + } else if (errno == ENOENT) { + char reason[1024] = { 0 }; + read(lguest_fd, reason, sizeof(reason)-1); + errx(1, "%s", reason); + } else if (errno != EAGAIN) + err(1, "Running guest failed"); + handle_input(lguest_fd, device_list); + if (write(lguest_fd, args, sizeof(args)) < 0) + err(1, "Resetting break"); + } +} + +static struct option opts[] = { + { "verbose", 0, NULL, 'v' }, + { "sharenet", 1, NULL, 's' }, + { "tunnet", 1, NULL, 't' }, + { "block", 1, NULL, 'b' }, + { "initrd", 1, NULL, 'i' }, + { NULL }, +}; +static void usage(void) +{ + errx(1, "Usage: lguest [--verbose] " + "[--sharenet=<filename>|--tunnet=(<ipaddr>|bridge:<bridgename>)\n" + "|--block=<filename>|--initrd=<filename>]...\n" + "<mem-in-mb> vmlinux [args...]"); +} + +int main(int argc, char *argv[]) +{ + unsigned long mem, pgdir, start, page_offset, initrd_size = 0; + int c, lguest_fd; + struct device_list device_list; + void *boot = (void *)0; + const char *initrd_name = NULL; + + device_list.max_infd = -1; + device_list.dev = NULL; + device_list.lastdev = &device_list.dev; + FD_ZERO(&device_list.infds); + + while ((c = getopt_long(argc, argv, "v", opts, NULL)) != EOF) { + switch (c) { + case 'v': + verbose = true; + break; + case 's': + setup_net_file(optarg, &device_list); + break; + case 't': + setup_tun_net(optarg, &device_list); + break; + case 'b': + setup_block_file(optarg, &device_list); + break; + case 'i': + initrd_name = optarg; + break; + default: + warnx("Unknown argument %s", argv[optind]); + usage(); + } + } + if (optind + 2 > argc) + usage(); + + /* We need a console device */ + setup_console(&device_list); + + /* First we map /dev/zero over all of guest-physical memory. */ + mem = atoi(argv[optind]) * 1024 * 1024; + map_zeroed_pages(0, mem / getpagesize()); + + /* Now we load the kernel */ + start = load_kernel(open_or_die(argv[optind+1], O_RDONLY), + &page_offset); + + /* Write the device descriptors into memory. */ + map_device_descriptors(&device_list, mem); + + /* Map the initrd image if requested */ + if (initrd_name) { + initrd_size = load_initrd(initrd_name, mem); + *(unsigned long *)(boot+0x218) = mem - initrd_size; + *(unsigned long *)(boot+0x21c) = initrd_size; + *(unsigned char *)(boot+0x210) = 0xFF; + } + + /* Set up the initial linar pagetables. */ + pgdir = setup_pagetables(mem, initrd_size, page_offset); + + /* E820 memory map: ours is a simple, single region. */ + *(char*)(boot+E820NR) = 1; + *((struct e820entry *)(boot+E820MAP)) + = ((struct e820entry) { 0, mem, E820_RAM }); + /* Command line pointer and command line (at 4096) */ + *(void **)(boot + 0x228) = boot + 4096; + concat(boot + 4096, argv+optind+2); + /* Paravirt type: 1 == lguest */ + *(int *)(boot + 0x23c) = 1; + + lguest_fd = tell_kernel(pgdir, start, page_offset); + waker_fd = setup_waker(lguest_fd, &device_list); + + run_guest(lguest_fd, &device_list); +} diff --git a/Documentation/lguest/lguest.txt b/Documentation/lguest/lguest.txt new file mode 100644 index 00000000000..821617bd6c0 --- /dev/null +++ b/Documentation/lguest/lguest.txt @@ -0,0 +1,129 @@ +Rusty's Remarkably Unreliable Guide to Lguest + - or, A Young Coder's Illustrated Hypervisor +http://lguest.ozlabs.org + +Lguest is designed to be a minimal hypervisor for the Linux kernel, for +Linux developers and users to experiment with virtualization with the +minimum of complexity. Nonetheless, it should have sufficient +features to make it useful for specific tasks, and, of course, you are +encouraged to fork and enhance it. + +Features: + +- Kernel module which runs in a normal kernel. +- Simple I/O model for communication. +- Simple program to create new guests. +- Logo contains cute puppies: http://lguest.ozlabs.org + +Developer features: + +- Fun to hack on. +- No ABI: being tied to a specific kernel anyway, you can change anything. +- Many opportunities for improvement or feature implementation. + +Running Lguest: + +- Lguest runs the same kernel as guest and host. You can configure + them differently, but usually it's easiest not to. + + You will need to configure your kernel with the following options: + + CONFIG_HIGHMEM64G=n ("High Memory Support" "64GB")[1] + CONFIG_TUN=y/m ("Universal TUN/TAP device driver support") + CONFIG_EXPERIMENTAL=y ("Prompt for development and/or incomplete code/drivers") + CONFIG_PARAVIRT=y ("Paravirtualization support (EXPERIMENTAL)") + CONFIG_LGUEST=y/m ("Linux hypervisor example code") + + and I recommend: + CONFIG_HZ=100 ("Timer frequency")[2] + +- A tool called "lguest" is available in this directory: type "make" + to build it. If you didn't build your kernel in-tree, use "make + O=<builddir>". + +- Create or find a root disk image. There are several useful ones + around, such as the xm-test tiny root image at + http://xm-test.xensource.com/ramdisks/initrd-1.1-i386.img + + For more serious work, I usually use a distribution ISO image and + install it under qemu, then make multiple copies: + + dd if=/dev/zero of=rootfile bs=1M count=2048 + qemu -cdrom image.iso -hda rootfile -net user -net nic -boot d + +- "modprobe lg" if you built it as a module. + +- Run an lguest as root: + + Documentation/lguest/lguest 64m vmlinux --tunnet=192.168.19.1 --block=rootfile root=/dev/lgba + + Explanation: + 64m: the amount of memory to use. + + vmlinux: the kernel image found in the top of your build directory. You + can also use a standard bzImage. + + --tunnet=192.168.19.1: configures a "tap" device for networking with this + IP address. + + --block=rootfile: a file or block device which becomes /dev/lgba + inside the guest. + + root=/dev/lgba: this (and anything else on the command line) are + kernel boot parameters. + +- Configuring networking. I usually have the host masquerade, using + "iptables -t nat -A POSTROUTING -o eth0 -j MASQUERADE" and "echo 1 > + /proc/sys/net/ipv4/ip_forward". In this example, I would configure + eth0 inside the guest at 192.168.19.2. + + Another method is to bridge the tap device to an external interface + using --tunnet=bridge:<bridgename>, and perhaps run dhcp on the guest + to obtain an IP address. The bridge needs to be configured first: + this option simply adds the tap interface to it. + + A simple example on my system: + + ifconfig eth0 0.0.0.0 + brctl addbr lg0 + ifconfig lg0 up + brctl addif lg0 eth0 + dhclient lg0 + + Then use --tunnet=bridge:lg0 when launching the guest. + + See http://linux-net.osdl.org/index.php/Bridge for general information + on how to get bridging working. + +- You can also create an inter-guest network using + "--sharenet=<filename>": any two guests using the same file are on + the same network. This file is created if it does not exist. + +Lguest I/O model: + +Lguest uses a simplified DMA model plus shared memory for I/O. Guests +can communicate with each other if they share underlying memory +(usually by the lguest program mmaping the same file), but they can +use any non-shared memory to communicate with the lguest process. + +Guests can register DMA buffers at any key (must be a valid physical +address) using the LHCALL_BIND_DMA(key, dmabufs, num<<8|irq) +hypercall. "dmabufs" is the physical address of an array of "num" +"struct lguest_dma": each contains a used_len, and an array of +physical addresses and lengths. When a transfer occurs, the +"used_len" field of one of the buffers which has used_len 0 will be +set to the length transferred and the irq will fire. + +Using an irq value of 0 unbinds the dma buffers. + +To send DMA, the LHCALL_SEND_DMA(key, dma_physaddr) hypercall is used, +and the bytes used is written to the used_len field. This can be 0 if +noone else has bound a DMA buffer to that key or some other error. +DMA buffers bound by the same guest are ignored. + +Cheers! +Rusty Russell rusty@rustcorp.com.au. + +[1] These are on various places on the TODO list, waiting for you to + get annoyed enough at the limitation to fix it. +[2] Lguest is not yet tickless when idle. See [1]. diff --git a/Documentation/power/notifiers.txt b/Documentation/power/notifiers.txt new file mode 100644 index 00000000000..9293e4bc857 --- /dev/null +++ b/Documentation/power/notifiers.txt @@ -0,0 +1,50 @@ +Suspend notifiers + (C) 2007 Rafael J. Wysocki <rjw@sisk.pl>, GPL + +There are some operations that device drivers may want to carry out in their +.suspend() routines, but shouldn't, because they can cause the hibernation or +suspend to fail. For example, a driver may want to allocate a substantial amount +of memory (like 50 MB) in .suspend(), but that shouldn't be done after the +swsusp's memory shrinker has run. + +Also, there may be some operations, that subsystems want to carry out before a +hibernation/suspend or after a restore/resume, requiring the system to be fully +functional, so the drivers' .suspend() and .resume() routines are not suitable +for this purpose. For example, device drivers may want to upload firmware to +their devices after a restore from a hibernation image, but they cannot do it by +calling request_firmware() from their .resume() routines (user land processes +are frozen at this point). The solution may be to load the firmware into +memory before processes are frozen and upload it from there in the .resume() +routine. Of course, a hibernation notifier may be used for this purpose. + +The subsystems that have such needs can register suspend notifiers that will be +called upon the following events by the suspend core: + +PM_HIBERNATION_PREPARE The system is going to hibernate or suspend, tasks will + be frozen immediately. + +PM_POST_HIBERNATION The system memory state has been restored from a + hibernation image or an error occured during the + hibernation. Device drivers' .resume() callbacks have + been executed and tasks have been thawed. + +PM_SUSPEND_PREPARE The system is preparing for a suspend. + +PM_POST_SUSPEND The system has just resumed or an error occured during + the suspend. Device drivers' .resume() callbacks have + been executed and tasks have been thawed. + +It is generally assumed that whatever the notifiers do for +PM_HIBERNATION_PREPARE, should be undone for PM_POST_HIBERNATION. Analogously, +operations performed for PM_SUSPEND_PREPARE should be reversed for +PM_POST_SUSPEND. Additionally, all of the notifiers are called for +PM_POST_HIBERNATION if one of them fails for PM_HIBERNATION_PREPARE, and +all of the notifiers are called for PM_POST_SUSPEND if one of them fails for +PM_SUSPEND_PREPARE. + +The hibernation and suspend notifiers are called with pm_mutex held. They are +defined in the usual way, but their last argument is meaningless (it is always +NULL). To register and/or unregister a suspend notifier use the functions +register_pm_notifier() and unregister_pm_notifier(), respectively, defined in +include/linux/suspend.h . If you don't need to unregister the notifier, you can +also use the pm_notifier() macro defined in include/linux/suspend.h . @@ -13,6 +13,7 @@ offsets-file := include/asm-$(ARCH)/asm-offsets.h always := $(offsets-file) targets := $(offsets-file) targets += arch/$(ARCH)/kernel/asm-offsets.s +clean-files := $(addprefix $(objtree)/,$(targets)) # Default sed regexp - multiline due to syntax constraints define sed-y diff --git a/MAINTAINERS b/MAINTAINERS index f6b2665ccb2..fbe0dca1c0e 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -607,6 +607,12 @@ W: http://sourceforge.net/projects/acpi4asus W: http://xf.iksaif.net/acpi4asus S: Maintained +ASUS ASB100 HARDWARE MONITOR DRIVER +P: Mark M. Hoffman +M: mhoffman@lightlink.com +L: lm-sensors@lm-sensors.org +S: Maintained + ASUS LAPTOP EXTRAS DRIVER P: Corentin Chary M: corentincj@iksaif.net @@ -1273,6 +1279,12 @@ M: shannon.nelson@intel.com L: linux-kernel@vger.kernel.org S: Maintained +DME1737 HARDWARE MONITOR DRIVER +P: Juerg Haefliger +M: juergh@gmail.com +L: lm-sensors@lm-sensors.org +S: Maintained + DOCBOOK FOR DOCUMENTATION P: Randy Dunlap M: rdunlap@xenotime.net @@ -1352,21 +1364,60 @@ S: Supported EDAC-CORE P: Doug Thompson -M: norsk5@xmission.com +M: dougthompson@xmission.com L: bluesmoke-devel@lists.sourceforge.net W: bluesmoke.sourceforge.net S: Supported EDAC-E752X P: Mark Gross +P: Doug Thompson M: mark.gross@intel.com +M: dougthompson@xmission.com L: bluesmoke-devel@lists.sourceforge.net W: bluesmoke.sourceforge.net S: Maintained EDAC-E7XXX P: Doug Thompson -M: norsk5@xmission.com +M: dougthompson@xmission.com +L: bluesmoke-devel@lists.sourceforge.net +W: bluesmoke.sourceforge.net +S: Maintained + +EDAC-I82443BXGX +P: Tim Small +M: tim@buttersideup.com +L: bluesmoke-devel@lists.sourceforge.net +W: bluesmoke.sourceforge.net +S: Maintained + +EDAC-I3000 +P: Jason Uhlenkott +M: juhlenko@akamai.com +L: bluesmoke-devel@lists.sourceforge.net +W: bluesmoke.sourceforge.net +S: Maintained + +EDAC-I5000 +P: Doug Thompson +M: dougthompson@xmission.com +L: bluesmoke-devel@lists.sourceforge.net +W: bluesmoke.sourceforge.net +S: Maintained + +EDAC-I82975X +P: Ranganathan Desikan +P: Arvind R. +M: rdesikan@jetzbroadband.com +M: arvind@acarlab.com +L: bluesmoke-devel@lists.sourceforge.net +W: bluesmoke.sourceforge.net +S: Maintained + +EDAC-PASEMI +P: Egor Martovetsky +M: egor@pasemi.com L: bluesmoke-devel@lists.sourceforge.net W: bluesmoke.sourceforge.net S: Maintained @@ -1584,11 +1635,11 @@ W: http://gigaset307x.sourceforge.net/ S: Maintained HARDWARE MONITORING -P: Jean Delvare -M: khali@linux-fr.org +P: Mark M. Hoffman +M: mhoffman@lightlink.com L: lm-sensors@lm-sensors.org W: http://www.lm-sensors.org/ -T: quilt http://khali.linux-fr.org/devel/linux-2.6/jdelvare-hwmon/ +T: git lm-sensors.org:/kernel/mhoffman/hwmon-2.6.git S: Maintained HARDWARE RANDOM NUMBER GENERATOR CORE @@ -1724,6 +1775,12 @@ P: William Irwin M: wli@holomorphy.com S: Maintained +I2C/SMBUS STUB DRIVER +P: Mark M. Hoffman +M: mhoffman@lightlink.com +L: lm-sensors@lm-sensors.org +S: Maintained + I2C SUBSYSTEM P: Jean Delvare M: khali@linux-fr.org @@ -2857,8 +2914,8 @@ L: linux-kernel@vger.kernel.org S: Maintained POSIX CLOCKS and TIMERS -P: George Anzinger -M: george@mvista.com +P: Thomas Gleixner +M: tglx@linutronix.de L: linux-kernel@vger.kernel.org S: Supported @@ -3253,6 +3310,12 @@ W: http://www.brownhat.org/sis900.html L: netdev@vger.kernel.org S: Maintained +SIS 96X I2C/SMBUS DRIVER +P: Mark M. Hoffman +M: mhoffman@lightlink.com +L: lm-sensors@lm-sensors.org +S: Maintained + SIS FRAMEBUFFER DRIVER P: Thomas Winischhofer M: thomas@winischhofer.net @@ -3270,6 +3333,12 @@ P: Nicolas Pitre M: nico@cam.org S: Maintained +SMSC47B397 HARDWARE MONITOR DRIVER +P: Mark M. Hoffman +M: mhoffman@lightlink.com +L: lm-sensors@lm-sensors.org +S: Maintained + SOFTMAC LAYER (IEEE 802.11) P: Johannes Berg M: johannes@sipsolutions.net @@ -3289,9 +3358,19 @@ M: neilb@suse.de L: linux-raid@vger.kernel.org S: Supported -SOFTWARE SUSPEND: +HIBERNATION (aka Software Suspend, aka swsusp): +P: Pavel Machek +M: pavel@suse.cz +P: Rafael J. Wysocki +M: rjw@sisk.pl +L: linux-pm@lists.linux-foundation.org +S: Supported + +SUSPEND TO RAM: P: Pavel Machek M: pavel@suse.cz +P: Rafael J. Wysocki +M: rjw@sisk.pl L: linux-pm@lists.linux-foundation.org S: Maintained @@ -492,7 +492,7 @@ endif include $(srctree)/arch/$(ARCH)/Makefile ifdef CONFIG_FRAME_POINTER -CFLAGS += -fno-omit-frame-pointer $(call cc-option,-fno-optimize-sibling-calls,) +CFLAGS += -fno-omit-frame-pointer -fno-optimize-sibling-calls else CFLAGS += -fomit-frame-pointer endif @@ -514,6 +514,12 @@ CFLAGS += $(call cc-option,-Wdeclaration-after-statement,) # disable pointer signed / unsigned warnings in gcc 4.0 CFLAGS += $(call cc-option,-Wno-pointer-sign,) +# Use --build-id when available. +LDFLAGS_BUILD_ID = $(patsubst -Wl$(comma)%,%,\ + $(call ld-option, -Wl$(comma)--build-id,)) +LDFLAGS_MODULE += $(LDFLAGS_BUILD_ID) +LDFLAGS_vmlinux += $(LDFLAGS_BUILD_ID) + # Default kernel image to build when no specific target is given. # KBUILD_IMAGE may be overruled on the command line or # set in the environment @@ -612,7 +618,7 @@ quiet_cmd_vmlinux__ ?= LD $@ cmd_vmlinux__ ?= $(LD) $(LDFLAGS) $(LDFLAGS_vmlinux) -o $@ \ -T $(vmlinux-lds) $(vmlinux-init) \ --start-group $(vmlinux-main) --end-group \ - $(filter-out $(vmlinux-lds) $(vmlinux-init) $(vmlinux-main) FORCE ,$^) + $(filter-out $(vmlinux-lds) $(vmlinux-init) $(vmlinux-main) vmlinux.o FORCE ,$^) # Generate new vmlinux version quiet_cmd_vmlinux_version = GEN .version @@ -736,15 +742,31 @@ debug_kallsyms: .tmp_map$(last_kallsyms) endif # ifdef CONFIG_KALLSYMS +# Do modpost on a prelinked vmlinux. The finally linked vmlinux has +# relevant sections renamed as per the linker script. +quiet_cmd_vmlinux-modpost = LD $@ + cmd_vmlinux-modpost = $(LD) $(LDFLAGS) -r -o $@ \ + $(vmlinux-init) --start-group $(vmlinux-main) --end-group \ + $(filter-out $(vmlinux-init) $(vmlinux-main) $(vmlinux-lds) FORCE ,$^) +define rule_vmlinux-modpost + : + +$(call cmd,vmlinux-modpost) + $(Q)$(MAKE) -f $(srctree)/scripts/Makefile.modpost $@ + $(Q)echo 'cmd_$@ := $(cmd_vmlinux-modpost)' > $(dot-target).cmd +endef + # vmlinux image - including updated kernel symbols -vmlinux: $(vmlinux-lds) $(vmlinux-init) $(vmlinux-main) $(kallsyms.o) FORCE +vmlinux: $(vmlinux-lds) $(vmlinux-init) $(vmlinux-main) $(kallsyms.o) vmlinux.o FORCE ifdef CONFIG_HEADERS_CHECK $(Q)$(MAKE) -f $(srctree)/Makefile headers_check endif + $(call vmlinux-modpost) $(call if_changed_rule,vmlinux__) - $(Q)$(MAKE) -f $(srctree)/scripts/Makefile.modpost $@ $(Q)rm -f .old_version +vmlinux.o: $(vmlinux-lds) $(vmlinux-init) $(vmlinux-main) $(kallsyms.o) FORCE + $(call if_changed_rule,vmlinux-modpost) + # The actual objects are generated when descending, # make sure no implicit rule kicks in $(sort $(vmlinux-init) $(vmlinux-main)) $(vmlinux-lds): $(vmlinux-dirs) ; @@ -1317,7 +1339,7 @@ define xtags -I __initdata,__exitdata,__acquires,__releases \ -I EXPORT_SYMBOL,EXPORT_SYMBOL_GPL \ --extra=+f --c-kinds=+px \ - --regex-asm='/ENTRY\(([^)]*)\).*/\1/'; \ + --regex-asm='/^ENTRY\(([^)]*)\).*/\1/'; \ $(all-kconfigs) | xargs $1 -a \ --langdef=kconfig \ --language-force=kconfig \ diff --git a/arch/alpha/kernel/module.c b/arch/alpha/kernel/module.c index bd03dc94c72..026ba9af6d6 100644 --- a/arch/alpha/kernel/module.c +++ b/arch/alpha/kernel/module.c @@ -119,8 +119,7 @@ module_frob_arch_sections(Elf64_Ehdr *hdr, Elf64_Shdr *sechdrs, } nsyms = symtab->sh_size / sizeof(Elf64_Sym); - chains = kmalloc(nsyms * sizeof(struct got_entry), GFP_KERNEL); - memset(chains, 0, nsyms * sizeof(struct got_entry)); + chains = kcalloc(nsyms, sizeof(struct got_entry), GFP_KERNEL); got->sh_size = 0; got->sh_addralign = 8; diff --git a/arch/alpha/kernel/vmlinux.lds.S b/arch/alpha/kernel/vmlinux.lds.S index 449e76f118d..fe13daa5cb2 100644 --- a/arch/alpha/kernel/vmlinux.lds.S +++ b/arch/alpha/kernel/vmlinux.lds.S @@ -3,7 +3,7 @@ OUTPUT_FORMAT("elf64-alpha") OUTPUT_ARCH(alpha) ENTRY(__start) -PHDRS { kernel PT_LOAD ; } +PHDRS { kernel PT_LOAD; note PT_NOTE; } jiffies = jiffies_64; SECTIONS { @@ -28,6 +28,9 @@ SECTIONS __ex_table : { *(__ex_table) } __stop___ex_table = .; + NOTES :kernel :note + .dummy : { *(.dummy) } :kernel + RODATA /* Will be freed after init */ @@ -69,10 +72,7 @@ SECTIONS . = ALIGN(8); SECURITY_INIT - . = ALIGN(8192); - __per_cpu_start = .; - .data.percpu : { *(.data.percpu) } - __per_cpu_end = .; + PERCPU(8192) . = ALIGN(2*8192); __init_end = .; diff --git a/arch/alpha/mm/fault.c b/arch/alpha/mm/fault.c index f5862792a16..a0e18da594d 100644 --- a/arch/alpha/mm/fault.c +++ b/arch/alpha/mm/fault.c @@ -148,21 +148,17 @@ do_page_fault(unsigned long address, unsigned long mmcsr, the fault. */ fault = handle_mm_fault(mm, vma, address, cause > 0); up_read(&mm->mmap_sem); - - switch (fault) { - case VM_FAULT_MINOR: - current->min_flt++; - break; - case VM_FAULT_MAJOR: - current->maj_flt++; - break; - case VM_FAULT_SIGBUS: - goto do_sigbus; - case VM_FAULT_OOM: - goto out_of_memory; - default: + if (unlikely(fault & VM_FAULT_ERROR)) { + if (fault & VM_FAULT_OOM) + goto out_of_memory; + else if (fault & VM_FAULT_SIGBUS) + goto do_sigbus; BUG(); } + if (fault & VM_FAULT_MAJOR) + current->maj_flt++; + else + current->min_flt++; return; /* Something tried to access memory that isn't in our memory map. diff --git a/arch/arm/configs/badge4_defconfig b/arch/arm/configs/badge4_defconfig index 821865f7560..b2bbf217c70 100644 --- a/arch/arm/configs/badge4_defconfig +++ b/arch/arm/configs/badge4_defconfig @@ -708,7 +708,6 @@ CONFIG_I2C_ALGOPCF=m # I2C Hardware Bus support # CONFIG_I2C_ELEKTOR=m -# CONFIG_I2C_ISA is not set # CONFIG_I2C_PARPORT is not set # CONFIG_I2C_PARPORT_LIGHT is not set # CONFIG_I2C_STUB is not set diff --git a/arch/arm/configs/clps7500_defconfig b/arch/arm/configs/clps7500_defconfig index af9ae538913..49e9f9d8b3d 100644 --- a/arch/arm/configs/clps7500_defconfig +++ b/arch/arm/configs/clps7500_defconfig @@ -536,7 +536,6 @@ CONFIG_I2C_ALGOBIT=y # I2C Hardware Bus support # # CONFIG_I2C_ELEKTOR is not set -# CONFIG_I2C_ISA is not set # CONFIG_I2C_PARPORT is not set # CONFIG_I2C_PARPORT_LIGHT is not set # CONFIG_I2C_PCA_ISA is not set diff --git a/arch/arm/configs/footbridge_defconfig b/arch/arm/configs/footbridge_defconfig index 2a612d23120..299dc22294a 100644 --- a/arch/arm/configs/footbridge_defconfig +++ b/arch/arm/configs/footbridge_defconfig @@ -748,7 +748,6 @@ CONFIG_I2C=m # CONFIG_I2C_ELEKTOR is not set # CONFIG_I2C_I801 is not set # CONFIG_I2C_I810 is not set -# CONFIG_I2C_ISA is not set # CONFIG_I2C_NFORCE2 is not set # CONFIG_I2C_PARPORT is not set # CONFIG_I2C_PARPORT_LIGHT is not set diff --git a/arch/arm/configs/neponset_defconfig b/arch/arm/configs/neponset_defconfig index e86794a10fc..92ccdc6492f 100644 --- a/arch/arm/configs/neponset_defconfig +++ b/arch/arm/configs/neponset_defconfig @@ -698,7 +698,6 @@ CONFIG_I2C_ALGOBIT=y # I2C Hardware Bus support # # CONFIG_I2C_ELEKTOR is not set -# CONFIG_I2C_ISA is not set # CONFIG_I2C_PARPORT_LIGHT is not set # CONFIG_I2C_STUB is not set # CONFIG_I2C_PCA_ISA is not set diff --git a/arch/arm/configs/picotux200_defconfig b/arch/arm/configs/picotux200_defconfig index 339c48953a6..3c0c4f192dc 100644 --- a/arch/arm/configs/picotux200_defconfig +++ b/arch/arm/configs/picotux200_defconfig @@ -735,7 +735,6 @@ CONFIG_I2C_CHARDEV=m # I2C Hardware Bus support # CONFIG_I2C_AT91=m -CONFIG_I2C_ISA=m # CONFIG_I2C_OCORES is not set # CONFIG_I2C_PARPORT_LIGHT is not set # CONFIG_I2C_STUB is not set diff --git a/arch/arm/configs/rpc_defconfig b/arch/arm/configs/rpc_defconfig index bc091264d35..8452dc8c7cc 100644 --- a/arch/arm/configs/rpc_defconfig +++ b/arch/arm/configs/rpc_defconfig @@ -558,7 +558,6 @@ CONFIG_I2C_ALGOBIT=y # # I2C Hardware Bus support # -# CONFIG_I2C_ISA is not set # CONFIG_I2C_PARPORT is not set # CONFIG_I2C_PARPORT_LIGHT is not set # CONFIG_I2C_STUB is not set diff --git a/arch/arm/configs/s3c2410_defconfig b/arch/arm/configs/s3c2410_defconfig index a850da377a2..1d5150e4d6b 100644 --- a/arch/arm/configs/s3c2410_defconfig +++ b/arch/arm/configs/s3c2410_defconfig @@ -826,7 +826,6 @@ CONFIG_I2C_ALGOBIT=m # I2C Hardware Bus support # # CONFIG_I2C_ELEKTOR is not set -CONFIG_I2C_ISA=m # CONFIG_I2C_OCORES is not set # CONFIG_I2C_PARPORT is not set # CONFIG_I2C_PARPORT_LIGHT is not set diff --git a/arch/arm/kernel/vmlinux.lds.S b/arch/arm/kernel/vmlinux.lds.S index 2b7a8f5d8cf..5ff5406666b 100644 --- a/arch/arm/kernel/vmlinux.lds.S +++ b/arch/arm/kernel/vmlinux.lds.S @@ -66,6 +66,7 @@ SECTIONS . = ALIGN(4096); __per_cpu_start = .; *(.data.percpu) + *(.data.percpu.shared_aligned) __per_cpu_end = .; #ifndef CONFIG_XIP_KERNEL __init_begin = _stext; diff --git a/arch/arm/mach-iop13xx/pci.c b/arch/arm/mach-iop13xx/pci.c index 9d63d7f260c..99d94cb1baf 100644 --- a/arch/arm/mach-iop13xx/pci.c +++ b/arch/arm/mach-iop13xx/pci.c @@ -1002,11 +1002,10 @@ int iop13xx_pci_setup(int nr, struct pci_sys_data *sys) if (nr > 1) return 0; - res = kmalloc(sizeof(struct resource) * 2, GFP_KERNEL); + res = kcalloc(2, sizeof(struct resource), GFP_KERNEL); if (!res) panic("PCI: unable to alloc resources"); - memset(res, 0, sizeof(struct resource) * 2); /* 'nr' assumptions: * ATUX is always 0 diff --git a/arch/arm/mm/fault.c b/arch/arm/mm/fault.c index 75d491448e4..c04124a095c 100644 --- a/arch/arm/mm/fault.c +++ b/arch/arm/mm/fault.c @@ -183,20 +183,20 @@ good_area: */ survive: fault = handle_mm_fault(mm, vma, addr & PAGE_MASK, fsr & (1 << 11)); - - /* - * Handle the "normal" cases first - successful and sigbus - */ - switch (fault) { - case VM_FAULT_MAJOR: + if (unlikely(fault & VM_FAULT_ERROR)) { + if (fault & VM_FAULT_OOM) + goto out_of_memory; + else if (fault & VM_FAULT_SIGBUS) + return fault; + BUG(); + } + if (fault & VM_FAULT_MAJOR) tsk->maj_flt++; - return fault; - case VM_FAULT_MINOR: + else tsk->min_flt++; - case VM_FAULT_SIGBUS: - return fault; - } + return fault; +out_of_memory: if (!is_init(tsk)) goto out; @@ -249,7 +249,7 @@ do_page_fault(unsigned long addr, unsigned int fsr, struct pt_regs *regs) /* * Handle the "normal" case first - VM_FAULT_MAJOR / VM_FAULT_MINOR */ - if (fault >= VM_FAULT_MINOR) + if (likely(!(fault & VM_FAULT_ERROR))) return 0; /* @@ -259,8 +259,7 @@ do_page_fault(unsigned long addr, unsigned int fsr, struct pt_regs *regs) if (!user_mode(regs)) goto no_context; - switch (fault) { - case VM_FAULT_OOM: + if (fault & VM_FAULT_OOM) { /* * We ran out of memory, or some other thing * happened to us that made us unable to handle @@ -269,17 +268,15 @@ do_page_fault(unsigned long addr, unsigned int fsr, struct pt_regs *regs) printk("VM: killing process %s\n", tsk->comm); do_exit(SIGKILL); return 0; - - case VM_FAULT_SIGBUS: + } + if (fault & VM_FAULT_SIGBUS) { /* * We had some memory, but were unable to * successfully fix up this page fault. */ sig = SIGBUS; code = BUS_ADRERR; - break; - - default: + } else { /* * Something tried to access memory that * isn't in our memory map.. @@ -287,7 +284,6 @@ do_page_fault(unsigned long addr, unsigned int fsr, struct pt_regs *regs) sig = SIGSEGV; code = fault == VM_FAULT_BADACCESS ? SEGV_ACCERR : SEGV_MAPERR; - break; } __do_user_fault(tsk, addr, fsr, sig, code, regs); diff --git a/arch/arm26/mm/fault.c b/arch/arm26/mm/fault.c index 93c0cee0fb5..dec638a0c8d 100644 --- a/arch/arm26/mm/fault.c +++ b/arch/arm26/mm/fault.c @@ -170,20 +170,20 @@ good_area: */ survive: fault = handle_mm_fault(mm, vma, addr & PAGE_MASK, DO_COW(fsr)); - - /* - * Handle the "normal" cases first - successful and sigbus - */ - switch (fault) { - case VM_FAULT_MAJOR: + if (unlikely(fault & VM_FAULT_ERROR)) { + if (fault & VM_FAULT_OOM) + goto out_of_memory; + else if (fault & VM_FAULT_SIGBUS) + return fault; + BUG(); + } + if (fault & VM_FAULT_MAJOR) tsk->maj_flt++; - return fault; - case VM_FAULT_MINOR: + else tsk->min_flt++; - case VM_FAULT_SIGBUS: - return fault; - } + return fault; +out_of_memory: fault = -3; /* out of memory */ if (!is_init(tsk)) goto out; @@ -225,13 +225,11 @@ int do_page_fault(unsigned long addr, unsigned int fsr, struct pt_regs *regs) /* * Handle the "normal" case first */ - switch (fault) { - case VM_FAULT_MINOR: - case VM_FAULT_MAJOR: + if (likely(!(fault & VM_FAULT_ERROR))) return 0; - case VM_FAULT_SIGBUS: + if (fault & VM_FAULT_SIGBUS) goto do_sigbus; - } + /* else VM_FAULT_OOM */ /* * If we are in kernel mode at this point, we diff --git a/arch/avr32/mm/fault.c b/arch/avr32/mm/fault.c index 4b2495285d9..ae2d2c593b2 100644 --- a/arch/avr32/mm/fault.c +++ b/arch/avr32/mm/fault.c @@ -64,6 +64,7 @@ asmlinkage void do_page_fault(unsigned long ecr, struct pt_regs *regs) int writeaccess; long signr; int code; + int fault; if (notify_page_fault(regs, ecr)) return; @@ -132,20 +133,18 @@ good_area: * fault. */ survive: - switch (handle_mm_fault(mm, vma, address, writeaccess)) { - case VM_FAULT_MINOR: - tsk->min_flt++; - break; - case VM_FAULT_MAJOR: - tsk->maj_flt++; - break; - case VM_FAULT_SIGBUS: - goto do_sigbus; - case VM_FAULT_OOM: - goto out_of_memory; - default: + fault = handle_mm_fault(mm, vma, address, writeaccess); + if (unlikely(fault & VM_FAULT_ERROR)) { + if (fault & VM_FAULT_OOM) + goto out_of_memory; + else if (fault & VM_FAULT_SIGBUS) + goto do_sigbus; BUG(); } + if (fault & VM_FAULT_MAJOR) + tsk->maj_flt++; + else + tsk->min_flt++; up_read(&mm->mmap_sem); return; diff --git a/arch/blackfin/mm/blackfin_sram.c b/arch/blackfin/mm/blackfin_sram.c index 16c6169ed01..b99ea883cd2 100644 --- a/arch/blackfin/mm/blackfin_sram.c +++ b/arch/blackfin/mm/blackfin_sram.c @@ -521,10 +521,9 @@ void *sram_alloc_with_lsl(size_t size, unsigned long flags) struct sram_list_struct *lsl = NULL; struct mm_struct *mm = current->mm; - lsl = kmalloc(sizeof(struct sram_list_struct), GFP_KERNEL); + lsl = kzalloc(sizeof(struct sram_list_struct), GFP_KERNEL); if (!lsl) return NULL; - memset(lsl, 0, sizeof(*lsl)); if (flags & L1_INST_SRAM) addr = l1_inst_sram_alloc(size); diff --git a/arch/cris/arch-v10/drivers/pcf8563.c b/arch/cris/arch-v10/drivers/pcf8563.c index d47cfbf98d6..1de0026bb94 100644 --- a/arch/cris/arch-v10/drivers/pcf8563.c +++ b/arch/cris/arch-v10/drivers/pcf8563.c @@ -180,9 +180,7 @@ err: void __exit pcf8563_exit(void) { - if (unregister_chrdev(PCF8563_MAJOR, DEVICE_NAME) < 0) { - printk(KERN_INFO "%s: Unable to unregister device.\n", PCF8563_NAME); - } + unregister_chrdev(PCF8563_MAJOR, DEVICE_NAME); } /* diff --git a/arch/cris/arch-v10/vmlinux.lds.S b/arch/cris/arch-v10/vmlinux.lds.S index 4b348b38cf3..9859d49d088 100644 --- a/arch/cris/arch-v10/vmlinux.lds.S +++ b/arch/cris/arch-v10/vmlinux.lds.S @@ -44,7 +44,7 @@ SECTIONS ___data_start = . ; __Sdata = . ; .data : { /* Data */ - *(.data) + DATA_DATA } __edata = . ; /* End of data section */ _edata = . ; diff --git a/arch/cris/arch-v32/drivers/pcf8563.c b/arch/cris/arch-v32/drivers/pcf8563.c index fa8d50007e4..da479a14f83 100644 --- a/arch/cris/arch-v32/drivers/pcf8563.c +++ b/arch/cris/arch-v32/drivers/pcf8563.c @@ -193,9 +193,7 @@ err: void __exit pcf8563_exit(void) { - if (unregister_chrdev(PCF8563_MAJOR, DEVICE_NAME) < 0) { - printk(KERN_INFO "%s: Unable to unregister device.\n", PCF8563_NAME); - } + unregister_chrdev(PCF8563_MAJOR, DEVICE_NAME); } /* diff --git a/arch/cris/arch-v32/drivers/pci/dma.c b/arch/cris/arch-v32/drivers/pci/dma.c index 832fc63504d..66f9500fbc0 100644 --- a/arch/cris/arch-v32/drivers/pci/dma.c +++ b/arch/cris/arch-v32/drivers/pci/dma.c @@ -91,14 +91,12 @@ int dma_declare_coherent_memory(struct device *dev, dma_addr_t bus_addr, if (!mem_base) goto out; - dev->dma_mem = kmalloc(sizeof(struct dma_coherent_mem), GFP_KERNEL); + dev->dma_mem = kzalloc(sizeof(struct dma_coherent_mem), GFP_KERNEL); if (!dev->dma_mem) goto out; - memset(dev->dma_mem, 0, sizeof(struct dma_coherent_mem)); - dev->dma_mem->bitmap = kmalloc(bitmap_size, GFP_KERNEL); + dev->dma_mem->bitmap = kzalloc(bitmap_size, GFP_KERNEL); if (!dev->dma_mem->bitmap) goto free1_out; - memset(dev->dma_mem->bitmap, 0, bitmap_size); dev->dma_mem->virt_base = mem_base; dev->dma_mem->device_base = device_addr; diff --git a/arch/cris/arch-v32/vmlinux.lds.S b/arch/cris/arch-v32/vmlinux.lds.S index dfa25e1542b..b076c134c0b 100644 --- a/arch/cris/arch-v32/vmlinux.lds.S +++ b/arch/cris/arch-v32/vmlinux.lds.S @@ -49,7 +49,7 @@ SECTIONS ___data_start = . ; __Sdata = . ; .data : { /* Data */ - *(.data) + DATA_DATA } __edata = . ; /* End of data section. */ _edata = . ; @@ -91,10 +91,7 @@ SECTIONS } SECURITY_INIT - . = ALIGN (8192); - __per_cpu_start = .; - .data.percpu : { *(.data.percpu) } - __per_cpu_end = .; + PERCPU(8192) #ifdef CONFIG_BLK_DEV_INITRD .init.ramfs : { diff --git a/arch/cris/mm/fault.c b/arch/cris/mm/fault.c index c73e91f1299..8672ab7d797 100644 --- a/arch/cris/mm/fault.c +++ b/arch/cris/mm/fault.c @@ -179,6 +179,7 @@ do_page_fault(unsigned long address, struct pt_regs *regs, struct mm_struct *mm; struct vm_area_struct * vma; siginfo_t info; + int fault; D(printk("Page fault for %lX on %X at %lX, prot %d write %d\n", address, smp_processor_id(), instruction_pointer(regs), @@ -283,18 +284,18 @@ do_page_fault(unsigned long address, struct pt_regs *regs, * the fault. */ - switch (handle_mm_fault(mm, vma, address, writeaccess & 1)) { - case VM_FAULT_MINOR: - tsk->min_flt++; - break; - case VM_FAULT_MAJOR: - tsk->maj_flt++; - break; - case VM_FAULT_SIGBUS: - goto do_sigbus; - default: - goto out_of_memory; + fault = handle_mm_fault(mm, vma, address, writeaccess & 1); + if (unlikely(fault & VM_FAULT_ERROR)) { + if (fault & VM_FAULT_OOM) + goto out_of_memory; + else if (fault & VM_FAULT_SIGBUS) + goto do_sigbus; + BUG(); } + if (fault & VM_FAULT_MAJOR) + tsk->maj_flt++; + else + tsk->min_flt++; up_read(&mm->mmap_sem); return; diff --git a/arch/frv/Makefile b/arch/frv/Makefile index 038e3a8457e..9bf7345c5cc 100644 --- a/arch/frv/Makefile +++ b/arch/frv/Makefile @@ -88,7 +88,7 @@ ASFLAGS += -mno-fdpic # make sure the .S files get compiled with debug info # and disable optimisations that are unhelpful whilst debugging ifdef CONFIG_DEBUG_INFO -CFLAGS += -O1 +#CFLAGS += -O1 AFLAGS += -Wa,--gdwarf2 ASFLAGS += -Wa,--gdwarf2 endif diff --git a/arch/frv/kernel/vmlinux.lds.S b/arch/frv/kernel/vmlinux.lds.S index 481dc137464..3b71e0c8639 100644 --- a/arch/frv/kernel/vmlinux.lds.S +++ b/arch/frv/kernel/vmlinux.lds.S @@ -57,10 +57,7 @@ SECTIONS __alt_instructions_end = .; .altinstr_replacement : { *(.altinstr_replacement) } - . = ALIGN(4096); - __per_cpu_start = .; - .data.percpu : { *(.data.percpu) } - __per_cpu_end = .; + PERCPU(4096) #ifdef CONFIG_BLK_DEV_INITRD . = ALIGN(4096); diff --git a/arch/frv/mm/fault.c b/arch/frv/mm/fault.c index 3f12296c368..6798fa0257b 100644 --- a/arch/frv/mm/fault.c +++ b/arch/frv/mm/fault.c @@ -40,6 +40,7 @@ asmlinkage void do_page_fault(int datammu, unsigned long esr0, unsigned long ear pud_t *pue; pte_t *pte; int write; + int fault; #if 0 const char *atxc[16] = { @@ -162,18 +163,18 @@ asmlinkage void do_page_fault(int datammu, unsigned long esr0, unsigned long ear * make sure we exit gracefully rather than endlessly redo * the fault. */ - switch (handle_mm_fault(mm, vma, ear0, write)) { - case VM_FAULT_MINOR: - current->min_flt++; - break; - case VM_FAULT_MAJOR: - current->maj_flt++; - break; - case VM_FAULT_SIGBUS: - goto do_sigbus; - default: - goto out_of_memory; + fault = handle_mm_fault(mm, vma, ear0, write); + if (unlikely(fault & VM_FAULT_ERROR)) { + if (fault & VM_FAULT_OOM) + goto out_of_memory; + else if (fault & VM_FAULT_SIGBUS) + goto do_sigbus; + BUG(); } + if (fault & VM_FAULT_MAJOR) + current->maj_flt++; + else + current->min_flt++; up_read(&mm->mmap_sem); return; diff --git a/arch/i386/Kconfig.cpu b/arch/i386/Kconfig.cpu index 9cbe76c3aa3..11a24d54f27 100644 --- a/arch/i386/Kconfig.cpu +++ b/arch/i386/Kconfig.cpu @@ -297,11 +297,6 @@ config X86_POPAD_OK depends on !M386 default y -config X86_CMPXCHG64 - bool - depends on X86_PAE - default y - config X86_ALIGNMENT_16 bool depends on MWINCHIP3D || MWINCHIP2 || MWINCHIPC6 || MCYRIXIII || X86_ELAN || MK6 || M586MMX || M586TSC || M586 || M486 || MVIAC3_2 || MGEODEGX1 diff --git a/arch/i386/boot/.gitignore b/arch/i386/boot/.gitignore index 495f20c085d..18465143cfa 100644 --- a/arch/i386/boot/.gitignore +++ b/arch/i386/boot/.gitignore @@ -1,3 +1,5 @@ bootsect bzImage setup +setup.bin +setup.elf diff --git a/arch/i386/defconfig b/arch/i386/defconfig index 1a3a2217b7c..0ac62cdcd3b 100644 --- a/arch/i386/defconfig +++ b/arch/i386/defconfig @@ -166,7 +166,6 @@ CONFIG_X86_WP_WORKS_OK=y CONFIG_X86_INVLPG=y CONFIG_X86_BSWAP=y CONFIG_X86_POPAD_OK=y -CONFIG_X86_CMPXCHG64=y CONFIG_X86_GOOD_APIC=y CONFIG_X86_INTEL_USERCOPY=y CONFIG_X86_USE_PPRO_CHECKSUM=y diff --git a/arch/i386/kernel/acpi/sleep.c b/arch/i386/kernel/acpi/sleep.c index 4ee83577bf6..c42b5ab49de 100644 --- a/arch/i386/kernel/acpi/sleep.c +++ b/arch/i386/kernel/acpi/sleep.c @@ -14,7 +14,7 @@ /* address in low memory of the wakeup routine. */ unsigned long acpi_wakeup_address = 0; -unsigned long acpi_video_flags; +unsigned long acpi_realmode_flags; extern char wakeup_start, wakeup_end; extern unsigned long FASTCALL(acpi_copy_wakeup_routine(unsigned long)); @@ -68,9 +68,11 @@ static int __init acpi_sleep_setup(char *str) { while ((str != NULL) && (*str != '\0')) { if (strncmp(str, "s3_bios", 7) == 0) - acpi_video_flags = 1; + acpi_realmode_flags |= 1; if (strncmp(str, "s3_mode", 7) == 0) - acpi_video_flags |= 2; + acpi_realmode_flags |= 2; + if (strncmp(str, "s3_beep", 7) == 0) + acpi_realmode_flags |= 4; str = strchr(str, ','); if (str != NULL) str += strspn(str, ", \t"); @@ -80,9 +82,11 @@ static int __init acpi_sleep_setup(char *str) __setup("acpi_sleep=", acpi_sleep_setup); +/* Ouch, we want to delete this. We already have better version in userspace, in + s2ram from suspend.sf.net project */ static __init int reset_videomode_after_s3(struct dmi_system_id *d) { - acpi_video_flags |= 2; + acpi_realmode_flags |= 2; return 0; } diff --git a/arch/i386/kernel/acpi/wakeup.S b/arch/i386/kernel/acpi/wakeup.S index a2295a34b2c..ed0a0f2c159 100644 --- a/arch/i386/kernel/acpi/wakeup.S +++ b/arch/i386/kernel/acpi/wakeup.S @@ -13,6 +13,21 @@ # cs = 0x1234, eip = 0x05 # +#define BEEP \ + inb $97, %al; \ + outb %al, $0x80; \ + movb $3, %al; \ + outb %al, $97; \ + outb %al, $0x80; \ + movb $-74, %al; \ + outb %al, $67; \ + outb %al, $0x80; \ + movb $-119, %al; \ + outb %al, $66; \ + outb %al, $0x80; \ + movb $15, %al; \ + outb %al, $66; + ALIGN .align 4096 ENTRY(wakeup_start) @@ -31,6 +46,11 @@ wakeup_code: movw %cs, %ax movw %ax, %ds # Make ds:0 point to wakeup_start movw %ax, %ss + + testl $4, realmode_flags - wakeup_code + jz 1f + BEEP +1: mov $(wakeup_stack - wakeup_code), %sp # Private stack is needed for ASUS board movw $0x0e00 + 'S', %fs:(0x12) @@ -41,7 +61,7 @@ wakeup_code: cmpl $0x12345678, %eax jne bogus_real_magic - testl $1, video_flags - wakeup_code + testl $1, realmode_flags - wakeup_code jz 1f lcall $0xc000,$3 movw %cs, %ax @@ -49,7 +69,7 @@ wakeup_code: movw %ax, %ss 1: - testl $2, video_flags - wakeup_code + testl $2, realmode_flags - wakeup_code jz 1f mov video_mode - wakeup_code, %ax call mode_set @@ -88,7 +108,11 @@ wakeup_code: cmpl $0x12345678, %eax jne bogus_real_magic - ljmpl $__KERNEL_CS,$wakeup_pmode_return + testl $8, realmode_flags - wakeup_code + jz 1f + BEEP +1: + ljmpl $__KERNEL_CS, $wakeup_pmode_return real_save_gdt: .word 0 .long 0 @@ -97,7 +121,8 @@ real_save_cr3: .long 0 real_save_cr4: .long 0 real_magic: .long 0 video_mode: .long 0 -video_flags: .long 0 +realmode_flags: .long 0 +beep_flags: .long 0 real_efer_save_restore: .long 0 real_save_efer_edx: .long 0 real_save_efer_eax: .long 0 @@ -260,8 +285,8 @@ ENTRY(acpi_copy_wakeup_routine) movl saved_videomode, %edx movl %edx, video_mode - wakeup_start (%eax) - movl acpi_video_flags, %edx - movl %edx, video_flags - wakeup_start (%eax) + movl acpi_realmode_flags, %edx + movl %edx, realmode_flags - wakeup_start (%eax) movl $0x12345678, real_magic - wakeup_start (%eax) movl $0x12345678, saved_magic popl %ebx diff --git a/arch/i386/kernel/asm-offsets.c b/arch/i386/kernel/asm-offsets.c index 25f7eb51392..7288ac88d74 100644 --- a/arch/i386/kernel/asm-offsets.c +++ b/arch/i386/kernel/asm-offsets.c @@ -19,6 +19,11 @@ #include <xen/interface/xen.h> +#ifdef CONFIG_LGUEST_GUEST +#include <linux/lguest.h> +#include "../../../drivers/lguest/lg.h" +#endif + #define DEFINE(sym, val) \ asm volatile("\n->" #sym " %0 " #val : : "i" (val)) @@ -124,4 +129,19 @@ void foo(void) OFFSET(XEN_vcpu_info_mask, vcpu_info, evtchn_upcall_mask); OFFSET(XEN_vcpu_info_pending, vcpu_info, evtchn_upcall_pending); #endif + +#ifdef CONFIG_LGUEST_GUEST + BLANK(); + OFFSET(LGUEST_DATA_irq_enabled, lguest_data, irq_enabled); + OFFSET(LGUEST_PAGES_host_gdt_desc, lguest_pages, state.host_gdt_desc); + OFFSET(LGUEST_PAGES_host_idt_desc, lguest_pages, state.host_idt_desc); + OFFSET(LGUEST_PAGES_host_cr3, lguest_pages, state.host_cr3); + OFFSET(LGUEST_PAGES_host_sp, lguest_pages, state.host_sp); + OFFSET(LGUEST_PAGES_guest_gdt_desc, lguest_pages,state.guest_gdt_desc); + OFFSET(LGUEST_PAGES_guest_idt_desc, lguest_pages,state.guest_idt_desc); + OFFSET(LGUEST_PAGES_guest_gdt, lguest_pages, state.guest_gdt); + OFFSET(LGUEST_PAGES_regs_trapnum, lguest_pages, regs.trapnum); + OFFSET(LGUEST_PAGES_regs_errcode, lguest_pages, regs.errcode); + OFFSET(LGUEST_PAGES_regs, lguest_pages, regs); +#endif } diff --git a/arch/i386/kernel/init_task.c b/arch/i386/kernel/init_task.c index cff95d10a4d..d26fc063a76 100644 --- a/arch/i386/kernel/init_task.c +++ b/arch/i386/kernel/init_task.c @@ -42,5 +42,5 @@ EXPORT_SYMBOL(init_task); * per-CPU TSS segments. Threads are completely 'soft' on Linux, * no more per-task TSS's. */ -DEFINE_PER_CPU(struct tss_struct, init_tss) ____cacheline_internodealigned_in_smp = INIT_TSS; +DEFINE_PER_CPU_SHARED_ALIGNED(struct tss_struct, init_tss) = INIT_TSS; diff --git a/arch/i386/kernel/irq.c b/arch/i386/kernel/irq.c index d2daf672f4a..ba44d40b066 100644 --- a/arch/i386/kernel/irq.c +++ b/arch/i386/kernel/irq.c @@ -21,7 +21,7 @@ #include <asm/apic.h> #include <asm/uaccess.h> -DEFINE_PER_CPU(irq_cpustat_t, irq_stat) ____cacheline_internodealigned_in_smp; +DEFINE_PER_CPU_SHARED_ALIGNED(irq_cpustat_t, irq_stat); EXPORT_PER_CPU_SYMBOL(irq_stat); DEFINE_PER_CPU(struct pt_regs *, irq_regs); diff --git a/arch/i386/kernel/traps.c b/arch/i386/kernel/traps.c index d32fd4b6f78..3e7753c78b9 100644 --- a/arch/i386/kernel/traps.c +++ b/arch/i386/kernel/traps.c @@ -41,6 +41,10 @@ #include <linux/mca.h> #endif +#if defined(CONFIG_EDAC) +#include <linux/edac.h> +#endif + #include <asm/processor.h> #include <asm/system.h> #include <asm/io.h> @@ -638,6 +642,14 @@ mem_parity_error(unsigned char reason, struct pt_regs * regs) printk(KERN_EMERG "Uhhuh. NMI received for unknown reason %02x on " "CPU %d.\n", reason, smp_processor_id()); printk(KERN_EMERG "You have some hardware problem, likely on the PCI bus.\n"); + +#if defined(CONFIG_EDAC) + if(edac_handler_set()) { + edac_atomic_assert_error(); + return; + } +#endif + if (panic_on_unrecovered_nmi) panic("NMI: Not continuing"); @@ -1056,6 +1068,7 @@ asmlinkage void math_state_restore(void) thread->status |= TS_USEDFPU; /* So we fnsave on switch_to() */ tsk->fpu_counter++; } +EXPORT_SYMBOL_GPL(math_state_restore); #ifndef CONFIG_MATH_EMULATION diff --git a/arch/i386/kernel/tsc.c b/arch/i386/kernel/tsc.c index 252f9010f28..debd7dbb415 100644 --- a/arch/i386/kernel/tsc.c +++ b/arch/i386/kernel/tsc.c @@ -27,6 +27,7 @@ static int tsc_enabled; * an extra value to store the TSC freq */ unsigned int tsc_khz; +EXPORT_SYMBOL_GPL(tsc_khz); int tsc_disable; @@ -58,10 +59,11 @@ __setup("notsc", tsc_setup); */ static int tsc_unstable; -static inline int check_tsc_unstable(void) +int check_tsc_unstable(void) { return tsc_unstable; } +EXPORT_SYMBOL_GPL(check_tsc_unstable); /* Accellerators for sched_clock() * convert from cycles(64bits) => nanoseconds (64bits) diff --git a/arch/i386/kernel/vmlinux.lds.S b/arch/i386/kernel/vmlinux.lds.S index 00f1bc47d3a..7d72cce0052 100644 --- a/arch/i386/kernel/vmlinux.lds.S +++ b/arch/i386/kernel/vmlinux.lds.S @@ -60,7 +60,9 @@ SECTIONS __stop___ex_table = .; } - BUG_TABLE + NOTES :text :note + + BUG_TABLE :text . = ALIGN(4); .tracedata : AT(ADDR(.tracedata) - LOAD_OFFSET) { @@ -181,6 +183,7 @@ SECTIONS .data.percpu : AT(ADDR(.data.percpu) - LOAD_OFFSET) { __per_cpu_start = .; *(.data.percpu) + *(.data.percpu.shared_aligned) __per_cpu_end = .; } . = ALIGN(4096); @@ -207,6 +210,4 @@ SECTIONS STABS_DEBUG DWARF_DEBUG - - NOTES } diff --git a/arch/i386/mach-generic/es7000.c b/arch/i386/mach-generic/es7000.c index b47f951c0ec..4742626f08c 100644 --- a/arch/i386/mach-generic/es7000.c +++ b/arch/i386/mach-generic/es7000.c @@ -66,4 +66,4 @@ static int __init acpi_madt_oem_check(char *oem_id, char *oem_table_id) } #endif -struct genapic apic_es7000 = APIC_INIT("es7000", probe_es7000); +struct genapic __initdata_refok apic_es7000 = APIC_INIT("es7000", probe_es7000); diff --git a/arch/i386/mm/fault.c b/arch/i386/mm/fault.c index 1ecb3e43b52..e92a1012493 100644 --- a/arch/i386/mm/fault.c +++ b/arch/i386/mm/fault.c @@ -303,6 +303,7 @@ fastcall void __kprobes do_page_fault(struct pt_regs *regs, struct vm_area_struct * vma; unsigned long address; int write, si_code; + int fault; /* get the address */ address = read_cr2(); @@ -422,20 +423,18 @@ good_area: * make sure we exit gracefully rather than endlessly redo * the fault. */ - switch (handle_mm_fault(mm, vma, address, write)) { - case VM_FAULT_MINOR: - tsk->min_flt++; - break; - case VM_FAULT_MAJOR: - tsk->maj_flt++; - break; - case VM_FAULT_SIGBUS: - goto do_sigbus; - case VM_FAULT_OOM: + fault = handle_mm_fault(mm, vma, address, write); + if (unlikely(fault & VM_FAULT_ERROR)) { + if (fault & VM_FAULT_OOM) goto out_of_memory; - default: - BUG(); + else if (fault & VM_FAULT_SIGBUS) + goto do_sigbus; + BUG(); } + if (fault & VM_FAULT_MAJOR) + tsk->maj_flt++; + else + tsk->min_flt++; /* * Did it hit the DOS screen memory VA from vm86 mode? diff --git a/arch/ia64/ia32/binfmt_elf32.c b/arch/ia64/ia32/binfmt_elf32.c index 6f4d3d06f0e..e1189ba1ca5 100644 --- a/arch/ia64/ia32/binfmt_elf32.c +++ b/arch/ia64/ia32/binfmt_elf32.c @@ -195,62 +195,27 @@ ia64_elf32_init (struct pt_regs *regs) ia32_load_state(current); } +/* + * Undo the override of setup_arg_pages() without this ia32_setup_arg_pages() + * will suffer infinite self recursion. + */ +#undef setup_arg_pages + int ia32_setup_arg_pages (struct linux_binprm *bprm, int executable_stack) { - unsigned long stack_base; - struct vm_area_struct *mpnt; - struct mm_struct *mm = current->mm; - int i, ret; - - stack_base = IA32_STACK_TOP - MAX_ARG_PAGES*PAGE_SIZE; - mm->arg_start = bprm->p + stack_base; - - bprm->p += stack_base; - if (bprm->loader) - bprm->loader += stack_base; - bprm->exec += stack_base; - - mpnt = kmem_cache_zalloc(vm_area_cachep, GFP_KERNEL); - if (!mpnt) - return -ENOMEM; - - down_write(¤t->mm->mmap_sem); - { - mpnt->vm_mm = current->mm; - mpnt->vm_start = PAGE_MASK & (unsigned long) bprm->p; - mpnt->vm_end = IA32_STACK_TOP; - if (executable_stack == EXSTACK_ENABLE_X) - mpnt->vm_flags = VM_STACK_FLAGS | VM_EXEC; - else if (executable_stack == EXSTACK_DISABLE_X) - mpnt->vm_flags = VM_STACK_FLAGS & ~VM_EXEC; - else - mpnt->vm_flags = VM_STACK_FLAGS; - mpnt->vm_page_prot = (mpnt->vm_flags & VM_EXEC)? - PAGE_COPY_EXEC: PAGE_COPY; - if ((ret = insert_vm_struct(current->mm, mpnt))) { - up_write(¤t->mm->mmap_sem); - kmem_cache_free(vm_area_cachep, mpnt); - return ret; - } - current->mm->stack_vm = current->mm->total_vm = vma_pages(mpnt); + int ret; + + ret = setup_arg_pages(bprm, IA32_STACK_TOP, executable_stack); + if (!ret) { + /* + * Can't do it in ia64_elf32_init(). Needs to be done before + * calls to elf32_map() + */ + current->thread.ppl = ia32_init_pp_list(); } - for (i = 0 ; i < MAX_ARG_PAGES ; i++) { - struct page *page = bprm->page[i]; - if (page) { - bprm->page[i] = NULL; - install_arg_page(mpnt, page, stack_base); - } - stack_base += PAGE_SIZE; - } - up_write(¤t->mm->mmap_sem); - - /* Can't do it in ia64_elf32_init(). Needs to be done before calls to - elf32_map() */ - current->thread.ppl = ia32_init_pp_list(); - - return 0; + return ret; } static void diff --git a/arch/ia64/kernel/kprobes.c b/arch/ia64/kernel/kprobes.c index 5bc46f15134..5dc98b5abcf 100644 --- a/arch/ia64/kernel/kprobes.c +++ b/arch/ia64/kernel/kprobes.c @@ -936,10 +936,15 @@ static void ia64_get_bsp_cfm(struct unw_frame_info *info, void *arg) return; } +unsigned long arch_deref_entry_point(void *entry) +{ + return ((struct fnptr *)entry)->ip; +} + int __kprobes setjmp_pre_handler(struct kprobe *p, struct pt_regs *regs) { struct jprobe *jp = container_of(p, struct jprobe, kp); - unsigned long addr = ((struct fnptr *)(jp->entry))->ip; + unsigned long addr = arch_deref_entry_point(jp->entry); struct kprobe_ctlblk *kcb = get_kprobe_ctlblk(); struct param_bsp_cfm pa; int bytes; diff --git a/arch/ia64/kernel/setup.c b/arch/ia64/kernel/setup.c index 4d9864cc92c..cf06fe79904 100644 --- a/arch/ia64/kernel/setup.c +++ b/arch/ia64/kernel/setup.c @@ -980,15 +980,6 @@ cpu_init (void) pm_idle = default_idle; } -/* - * On SMP systems, when the scheduler does migration-cost autodetection, - * it needs a way to flush as much of the CPU's caches as possible. - */ -void sched_cacheflush(void) -{ - ia64_sal_cache_flush(3); -} - void __init check_bugs (void) { diff --git a/arch/ia64/kernel/smp.c b/arch/ia64/kernel/smp.c index b3a47f986e1..9f72838db26 100644 --- a/arch/ia64/kernel/smp.c +++ b/arch/ia64/kernel/smp.c @@ -82,7 +82,7 @@ static volatile struct call_data_struct *call_data; #define IPI_KDUMP_CPU_STOP 3 /* This needs to be cacheline aligned because it is written to by *other* CPUs. */ -static DEFINE_PER_CPU(u64, ipi_operation) ____cacheline_aligned; +static DEFINE_PER_CPU_SHARED_ALIGNED(u64, ipi_operation); extern void cpu_halt (void); diff --git a/arch/ia64/kernel/vmlinux.lds.S b/arch/ia64/kernel/vmlinux.lds.S index 5a65965c8b5..860f251d2fc 100644 --- a/arch/ia64/kernel/vmlinux.lds.S +++ b/arch/ia64/kernel/vmlinux.lds.S @@ -206,6 +206,7 @@ SECTIONS { __per_cpu_start = .; *(.data.percpu) + *(.data.percpu.shared_aligned) __per_cpu_end = .; } . = __phys_per_cpu_start + PERCPU_PAGE_SIZE; /* ensure percpu data fits diff --git a/arch/ia64/mm/fault.c b/arch/ia64/mm/fault.c index b87f785c241..73ccb6010c0 100644 --- a/arch/ia64/mm/fault.c +++ b/arch/ia64/mm/fault.c @@ -80,6 +80,7 @@ ia64_do_page_fault (unsigned long address, unsigned long isr, struct pt_regs *re struct mm_struct *mm = current->mm; struct siginfo si; unsigned long mask; + int fault; /* mmap_sem is performance critical.... */ prefetchw(&mm->mmap_sem); @@ -147,26 +148,25 @@ ia64_do_page_fault (unsigned long address, unsigned long isr, struct pt_regs *re * sure we exit gracefully rather than endlessly redo the * fault. */ - switch (handle_mm_fault(mm, vma, address, (mask & VM_WRITE) != 0)) { - case VM_FAULT_MINOR: - ++current->min_flt; - break; - case VM_FAULT_MAJOR: - ++current->maj_flt; - break; - case VM_FAULT_SIGBUS: + fault = handle_mm_fault(mm, vma, address, (mask & VM_WRITE) != 0); + if (unlikely(fault & VM_FAULT_ERROR)) { /* * We ran out of memory, or some other thing happened * to us that made us unable to handle the page fault * gracefully. */ - signal = SIGBUS; - goto bad_area; - case VM_FAULT_OOM: - goto out_of_memory; - default: + if (fault & VM_FAULT_OOM) { + goto out_of_memory; + } else if (fault & VM_FAULT_SIGBUS) { + signal = SIGBUS; + goto bad_area; + } BUG(); } + if (fault & VM_FAULT_MAJOR) + current->maj_flt++; + else + current->min_flt++; up_read(&mm->mmap_sem); return; diff --git a/arch/m32r/kernel/vmlinux.lds.S b/arch/m32r/kernel/vmlinux.lds.S index 4e2d5b9f0a9..942a8c7a441 100644 --- a/arch/m32r/kernel/vmlinux.lds.S +++ b/arch/m32r/kernel/vmlinux.lds.S @@ -110,10 +110,7 @@ SECTIONS __initramfs_end = .; #endif - . = ALIGN(4096); - __per_cpu_start = .; - .data.percpu : { *(.data.percpu) } - __per_cpu_end = .; + PERCPU(4096) . = ALIGN(4096); __init_end = .; /* freed after init ends here */ diff --git a/arch/m32r/m32104ut/defconfig.m32104ut b/arch/m32r/m32104ut/defconfig.m32104ut index 7b68fe8d921..1f88f493a9e 100644 --- a/arch/m32r/m32104ut/defconfig.m32104ut +++ b/arch/m32r/m32104ut/defconfig.m32104ut @@ -699,7 +699,6 @@ CONFIG_I2C_ALGOPCF=m # I2C Hardware Bus support # CONFIG_I2C_ELEKTOR=m -CONFIG_I2C_ISA=m # CONFIG_I2C_OCORES is not set # CONFIG_I2C_PARPORT is not set # CONFIG_I2C_PARPORT_LIGHT is not set diff --git a/arch/m32r/mm/fault.c b/arch/m32r/mm/fault.c index f3935ba2494..676a1c443d2 100644 --- a/arch/m32r/mm/fault.c +++ b/arch/m32r/mm/fault.c @@ -80,6 +80,7 @@ asmlinkage void do_page_fault(struct pt_regs *regs, unsigned long error_code, struct vm_area_struct * vma; unsigned long page, addr; int write; + int fault; siginfo_t info; /* @@ -195,20 +196,18 @@ survive: */ addr = (address & PAGE_MASK); set_thread_fault_code(error_code); - switch (handle_mm_fault(mm, vma, addr, write)) { - case VM_FAULT_MINOR: - tsk->min_flt++; - break; - case VM_FAULT_MAJOR: - tsk->maj_flt++; - break; - case VM_FAULT_SIGBUS: - goto do_sigbus; - case VM_FAULT_OOM: + fault = handle_mm_fault(mm, vma, addr, write); + if (unlikely(fault & VM_FAULT_ERROR)) { + if (fault & VM_FAULT_OOM) goto out_of_memory; - default: - BUG(); + else if (fault & VM_FAULT_SIGBUS) + goto do_sigbus; + BUG(); } + if (fault & VM_FAULT_MAJOR) + tsk->maj_flt++; + else + tsk->min_flt++; set_thread_fault_code(0); up_read(&mm->mmap_sem); return; diff --git a/arch/m68k/mm/fault.c b/arch/m68k/mm/fault.c index 2adbeb16e1b..578b48f47b9 100644 --- a/arch/m68k/mm/fault.c +++ b/arch/m68k/mm/fault.c @@ -159,18 +159,17 @@ good_area: #ifdef DEBUG printk("handle_mm_fault returns %d\n",fault); #endif - switch (fault) { - case VM_FAULT_MINOR: - current->min_flt++; - break; - case VM_FAULT_MAJOR: - current->maj_flt++; - break; - case VM_FAULT_SIGBUS: - goto bus_err; - default: - goto out_of_memory; + if (unlikely(fault & VM_FAULT_ERROR)) { + if (fault & VM_FAULT_OOM) + goto out_of_memory; + else if (fault & VM_FAULT_SIGBUS) + goto bus_err; + BUG(); } + if (fault & VM_FAULT_MAJOR) + current->maj_flt++; + else + current->min_flt++; up_read(&mm->mmap_sem); return 0; diff --git a/arch/m68knommu/Kconfig b/arch/m68knommu/Kconfig index adc64a2bafb..1175ceff8b2 100644 --- a/arch/m68knommu/Kconfig +++ b/arch/m68knommu/Kconfig @@ -45,6 +45,10 @@ config GENERIC_HWEIGHT bool default y +config GENERIC_HARDIRQS + bool + default y + config GENERIC_CALIBRATE_DELAY bool default y diff --git a/arch/m68knommu/kernel/Makefile b/arch/m68knommu/kernel/Makefile index 1c6cd1ab571..1524b39ad63 100644 --- a/arch/m68knommu/kernel/Makefile +++ b/arch/m68knommu/kernel/Makefile @@ -4,8 +4,8 @@ extra-y := vmlinux.lds -obj-y += dma.o entry.o init_task.o m68k_ksyms.o process.o ptrace.o semaphore.o \ - setup.o signal.o syscalltable.o sys_m68k.o time.o traps.o +obj-y += dma.o entry.o init_task.o irq.o m68k_ksyms.o process.o ptrace.o \ + semaphore.o setup.o signal.o syscalltable.o sys_m68k.o time.o traps.o obj-$(CONFIG_MODULES) += module.o obj-$(CONFIG_COMEMPCI) += comempci.o diff --git a/arch/m68knommu/kernel/asm-offsets.c b/arch/m68knommu/kernel/asm-offsets.c index 7cd183d346e..d97b89bae53 100644 --- a/arch/m68knommu/kernel/asm-offsets.c +++ b/arch/m68knommu/kernel/asm-offsets.c @@ -15,7 +15,6 @@ #include <linux/hardirq.h> #include <asm/bootinfo.h> #include <asm/irq.h> -#include <asm/irqnode.h> #include <asm/thread_info.h> #define DEFINE(sym, val) \ @@ -72,10 +71,6 @@ int main(void) #else /* bitfields are a bit difficult */ DEFINE(PT_VECTOR, offsetof(struct pt_regs, pc) + 4); - /* offsets into the irq_handler struct */ - DEFINE(IRQ_HANDLER, offsetof(struct irq_node, handler)); - DEFINE(IRQ_DEVID, offsetof(struct irq_node, dev_id)); - DEFINE(IRQ_NEXT, offsetof(struct irq_node, next)); #endif /* offsets into the kernel_stat struct */ diff --git a/arch/m68knommu/kernel/irq.c b/arch/m68knommu/kernel/irq.c new file mode 100644 index 00000000000..bba1bb48a21 --- /dev/null +++ b/arch/m68knommu/kernel/irq.c @@ -0,0 +1,82 @@ +/* + * irq.c + * + * (C) Copyright 2007, Greg Ungerer <gerg@snapgear.com> + * + * This file is subject to the terms and conditions of the GNU General Public + * License. See the file COPYING in the main directory of this archive + * for more details. + */ + +#include <linux/types.h> +#include <linux/init.h> +#include <linux/kernel.h> +#include <linux/kernel_stat.h> +#include <linux/interrupt.h> +#include <linux/irq.h> +#include <linux/seq_file.h> +#include <asm/system.h> +#include <asm/traps.h> + +asmlinkage void do_IRQ(int irq, struct pt_regs *regs) +{ + struct pt_regs *oldregs = set_irq_regs(regs); + + irq_enter(); + __do_IRQ(irq); + irq_exit(); + + set_irq_regs(oldregs); +} + +void ack_bad_irq(unsigned int irq) +{ + printk(KERN_ERR "IRQ: unexpected irq=%d\n", irq); +} + +static struct irq_chip m_irq_chip = { + .name = "M68K-INTC", + .enable = enable_vector, + .disable = disable_vector, + .ack = ack_vector, +}; + +void __init init_IRQ(void) +{ + int irq; + + init_vectors(); + + for (irq = 0; (irq < NR_IRQS); irq++) { + irq_desc[irq].status = IRQ_DISABLED; + irq_desc[irq].action = NULL; + irq_desc[irq].depth = 1; + irq_desc[irq].chip = &m_irq_chip; + } +} + +int show_interrupts(struct seq_file *p, void *v) +{ + struct irqaction *ap; + int irq = *((loff_t *) v); + + if (irq == 0) + seq_puts(p, " CPU0\n"); + + if (irq < NR_IRQS) { + ap = irq_desc[irq].action; + if (ap) { + seq_printf(p, "%3d: ", irq); + seq_printf(p, "%10u ", kstat_irqs(irq)); + seq_printf(p, "%14s ", irq_desc[irq].chip->name); + + seq_printf(p, "%s", ap->name); + for (ap = ap->next; ap; ap = ap->next) + seq_printf(p, ", %s", ap->name); + seq_putc(p, '\n'); + } + } + + return 0; +} + diff --git a/arch/m68knommu/kernel/m68k_ksyms.c b/arch/m68knommu/kernel/m68k_ksyms.c index 25327c9eadd..f795062aba1 100644 --- a/arch/m68knommu/kernel/m68k_ksyms.c +++ b/arch/m68knommu/kernel/m68k_ksyms.c @@ -81,8 +81,6 @@ EXPORT_SYMBOL(__mulsi3); EXPORT_SYMBOL(__udivsi3); EXPORT_SYMBOL(__umodsi3); -EXPORT_SYMBOL(is_in_rom); - #ifdef CONFIG_COLDFIRE extern unsigned int *dma_device_address; extern unsigned long dma_base_addr, _ramend; diff --git a/arch/m68knommu/kernel/traps.c b/arch/m68knommu/kernel/traps.c index fde04e1757f..437a061d8b9 100644 --- a/arch/m68knommu/kernel/traps.c +++ b/arch/m68knommu/kernel/traps.c @@ -62,8 +62,6 @@ static char const * const vec_names[] = { void __init trap_init(void) { - if (mach_trap_init) - mach_trap_init(); } void die_if_kernel(char *str, struct pt_regs *fp, int nr) @@ -82,7 +80,7 @@ void die_if_kernel(char *str, struct pt_regs *fp, int nr) printk(KERN_EMERG "Process %s (pid: %d, stackpage=%08lx)\n", current->comm, current->pid, PAGE_SIZE+(unsigned long)current); - show_stack(NULL, (unsigned long *)fp); + show_stack(NULL, (unsigned long *)(fp + 1)); add_taint(TAINT_DIE); do_exit(SIGSEGV); } diff --git a/arch/m68knommu/mm/memory.c b/arch/m68knommu/mm/memory.c index 1a66b71035a..f93b88b51f9 100644 --- a/arch/m68knommu/mm/memory.c +++ b/arch/m68knommu/mm/memory.c @@ -33,23 +33,3 @@ unsigned long kernel_map(unsigned long paddr, unsigned long size, return paddr; } - -int is_in_rom(unsigned long addr) -{ - extern unsigned long _ramstart, _ramend; - - /* - * What we are really trying to do is determine if addr is - * in an allocated kernel memory region. If not then assume - * we cannot free it or otherwise de-allocate it. Ideally - * we could restrict this to really being in a ROM or flash, - * but that would need to be done on a board by board basis, - * not globally. - */ - if ((addr < _ramstart) || (addr >= _ramend)) - return(1); - - /* Default case, not in ROM */ - return(0); -} - diff --git a/arch/m68knommu/platform/5307/Makefile b/arch/m68knommu/platform/5307/Makefile index 2fd37dcc309..719a313494b 100644 --- a/arch/m68knommu/platform/5307/Makefile +++ b/arch/m68knommu/platform/5307/Makefile @@ -16,7 +16,7 @@ ifdef CONFIG_FULLDEBUG AFLAGS += -DDEBUGGER_COMPATIBLE_CACHE=1 endif -obj-$(CONFIG_COLDFIRE) += entry.o vectors.o ints.o +obj-$(CONFIG_COLDFIRE) += entry.o vectors.o obj-$(CONFIG_M5206) += timers.o obj-$(CONFIG_M5206e) += timers.o obj-$(CONFIG_M520x) += pit.o diff --git a/arch/m68knommu/platform/5307/entry.S b/arch/m68knommu/platform/5307/entry.S index f0dba84d910..c358aebe0af 100644 --- a/arch/m68knommu/platform/5307/entry.S +++ b/arch/m68knommu/platform/5307/entry.S @@ -1,7 +1,7 @@ /* * linux/arch/m68knommu/platform/5307/entry.S * - * Copyright (C) 1999-2002, Greg Ungerer (gerg@snapgear.com) + * Copyright (C) 1999-2007, Greg Ungerer (gerg@snapgear.com) * Copyright (C) 1998 D. Jeff Dionne <jeff@lineo.ca>, * Kenneth Albanowski <kjahds@kjahds.com>, * Copyright (C) 2000 Lineo Inc. (www.lineo.com) @@ -155,34 +155,21 @@ Lsignal_return: /* * This is the generic interrupt handler (for all hardware interrupt - * sources). It figures out the vector number and calls the appropriate - * interrupt service routine directly. + * sources). Calls upto high level code to do all the work. */ ENTRY(inthandler) SAVE_ALL moveq #-1,%d0 movel %d0,%sp@(PT_ORIG_D0) - addql #1,local_irq_count movew %sp@(PT_FORMATVEC),%d0 /* put exception # in d0 */ andl #0x03fc,%d0 /* mask out vector only */ - leal per_cpu__kstat+STAT_IRQ,%a0 - addql #1,%a0@(%d0) - + movel %sp,%sp@- /* push regs arg */ lsrl #2,%d0 /* calculate real vector # */ - movel %d0,%d1 /* calculate array offset */ - lsll #4,%d1 - lea irq_list,%a0 - addl %d1,%a0 /* pointer to array struct */ - - movel %sp,%sp@- /* push regs arg onto stack */ - movel %a0@(8),%sp@- /* push devid arg */ - movel %d0,%sp@- /* push vector # on stack */ - - movel %a0@,%a0 /* get function to call */ - jbsr %a0@ /* call vector handler */ - lea %sp@(12),%sp /* pop parameters off stack */ + movel %d0,%sp@- /* push vector number */ + jbsr do_IRQ /* call high level irq handler */ + lea %sp@(8),%sp /* pop args off stack */ bra ret_from_interrupt /* this was fallthrough */ @@ -198,24 +185,15 @@ ENTRY(fasthandler) movew %sp@(PT_FORMATVEC),%d0 andl #0x03fc,%d0 /* mask out vector only */ - leal per_cpu__kstat+STAT_IRQ,%a0 - addql #1,%a0@(%d0) - - movel %sp,%sp@- /* push regs arg onto stack */ - clrl %sp@- /* push devid arg */ + movel %sp,%sp@- /* push regs arg */ lsrl #2,%d0 /* calculate real vector # */ - movel %d0,%sp@- /* push vector # on stack */ - - lsll #4,%d0 /* adjust for array offset */ - lea irq_list,%a0 - movel %a0@(%d0),%a0 /* get function to call */ - jbsr %a0@ /* call vector handler */ - lea %sp@(12),%sp /* pop parameters off stack */ + movel %d0,%sp@- /* push vector number */ + jbsr do_IRQ /* call high level irq handler */ + lea %sp@(8),%sp /* pop args off stack */ RESTORE_LOCAL ENTRY(ret_from_interrupt) - subql #1,local_irq_count jeq 2f 1: RESTORE_ALL diff --git a/arch/m68knommu/platform/5307/ints.c b/arch/m68knommu/platform/5307/ints.c deleted file mode 100644 index 751633038c4..00000000000 --- a/arch/m68knommu/platform/5307/ints.c +++ /dev/null @@ -1,279 +0,0 @@ -/* - * linux/arch/m68knommu/kernel/ints.c -- General interrupt handling code - * - * Copyright (C) 1999-2002 Greg Ungerer (gerg@snapgear.com) - * Copyright (C) 1998 D. Jeff Dionne <jeff@lineo.ca>, - * Kenneth Albanowski <kjahds@kjahds.com>, - * Copyright (C) 2000 Lineo Inc. (www.lineo.com) - * - * Based on: - * - * linux/arch/m68k/kernel/ints.c -- Linux/m68k general interrupt handling code - * - * This file is subject to the terms and conditions of the GNU General Public - * License. See the file COPYING in the main directory of this archive - * for more details. - */ - -#include <linux/module.h> -#include <linux/types.h> -#include <linux/init.h> -#include <linux/sched.h> -#include <linux/interrupt.h> -#include <linux/kernel_stat.h> -#include <linux/errno.h> -#include <linux/seq_file.h> - -#include <asm/system.h> -#include <asm/irq.h> -#include <asm/irqnode.h> -#include <asm/traps.h> -#include <asm/page.h> -#include <asm/machdep.h> - -/* - * This table stores the address info for each vector handler. - */ -struct irq_entry irq_list[SYS_IRQS]; - -#define NUM_IRQ_NODES 16 -static irq_node_t nodes[NUM_IRQ_NODES]; - -/* The number of spurious interrupts */ -volatile unsigned int num_spurious; - -unsigned int local_irq_count[NR_CPUS]; - -static irqreturn_t default_irq_handler(int irq, void *ptr) -{ -#if 1 - printk(KERN_INFO "%s(%d): default irq handler vec=%d [0x%x]\n", - __FILE__, __LINE__, irq, irq); -#endif - return(IRQ_HANDLED); -} - -/* - * void init_IRQ(void) - * - * Parameters: None - * - * Returns: Nothing - * - * This function should be called during kernel startup to initialize - * the IRQ handling routines. - */ - -void __init init_IRQ(void) -{ - int i; - - for (i = 0; i < SYS_IRQS; i++) { - if (mach_default_handler) - irq_list[i].handler = mach_default_handler; - else - irq_list[i].handler = default_irq_handler; - irq_list[i].flags = IRQ_FLG_STD; - irq_list[i].dev_id = NULL; - irq_list[i].devname = NULL; - } - - for (i = 0; i < NUM_IRQ_NODES; i++) - nodes[i].handler = NULL; - - if (mach_init_IRQ) - mach_init_IRQ(); -} - -irq_node_t *new_irq_node(void) -{ - irq_node_t *node; - short i; - - for (node = nodes, i = NUM_IRQ_NODES-1; i >= 0; node++, i--) - if (!node->handler) - return node; - - printk(KERN_INFO "new_irq_node: out of nodes\n"); - return NULL; -} - -int request_irq( - unsigned int irq, - irq_handler_t handler, - unsigned long flags, - const char *devname, - void *dev_id) -{ - if (irq < 0 || irq >= NR_IRQS) { - printk(KERN_WARNING "%s: Incorrect IRQ %d from %s\n", __FUNCTION__, - irq, devname); - return -ENXIO; - } - - if (!(irq_list[irq].flags & IRQ_FLG_STD)) { - if (irq_list[irq].flags & IRQ_FLG_LOCK) { - printk(KERN_WARNING "%s: IRQ %d from %s is not replaceable\n", - __FUNCTION__, irq, irq_list[irq].devname); - return -EBUSY; - } - if (flags & IRQ_FLG_REPLACE) { - printk(KERN_WARNING "%s: %s can't replace IRQ %d from %s\n", - __FUNCTION__, devname, irq, irq_list[irq].devname); - return -EBUSY; - } - } - - if (flags & IRQ_FLG_FAST) { - extern asmlinkage void fasthandler(void); - extern void set_evector(int vecnum, void (*handler)(void)); - set_evector(irq, fasthandler); - } - - irq_list[irq].handler = handler; - irq_list[irq].flags = flags; - irq_list[irq].dev_id = dev_id; - irq_list[irq].devname = devname; - return 0; -} - -EXPORT_SYMBOL(request_irq); - -void free_irq(unsigned int irq, void *dev_id) -{ - if (irq >= NR_IRQS) { - printk(KERN_WARNING "%s: Incorrect IRQ %d\n", __FUNCTION__, irq); - return; - } - - if (irq_list[irq].dev_id != dev_id) - printk(KERN_WARNING "%s: Removing probably wrong IRQ %d from %s\n", - __FUNCTION__, irq, irq_list[irq].devname); - - if (irq_list[irq].flags & IRQ_FLG_FAST) { - extern asmlinkage void inthandler(void); - extern void set_evector(int vecnum, void (*handler)(void)); - set_evector(irq, inthandler); - } - - if (mach_default_handler) - irq_list[irq].handler = mach_default_handler; - else - irq_list[irq].handler = default_irq_handler; - irq_list[irq].flags = IRQ_FLG_STD; - irq_list[irq].dev_id = NULL; - irq_list[irq].devname = NULL; -} - -EXPORT_SYMBOL(free_irq); - - -int sys_request_irq(unsigned int irq, irq_handler_t handler, - unsigned long flags, const char *devname, void *dev_id) -{ - if (irq > IRQ7) { - printk(KERN_WARNING "%s: Incorrect IRQ %d from %s\n", - __FUNCTION__, irq, devname); - return -ENXIO; - } - -#if 0 - if (!(irq_list[irq].flags & IRQ_FLG_STD)) { - if (irq_list[irq].flags & IRQ_FLG_LOCK) { - printk(KERN_WARNING "%s: IRQ %d from %s is not replaceable\n", - __FUNCTION__, irq, irq_list[irq].devname); - return -EBUSY; - } - if (!(flags & IRQ_FLG_REPLACE)) { - printk(KERN_WARNING "%s: %s can't replace IRQ %d from %s\n", - __FUNCTION__, devname, irq, irq_list[irq].devname); - return -EBUSY; - } - } -#endif - - irq_list[irq].handler = handler; - irq_list[irq].flags = flags; - irq_list[irq].dev_id = dev_id; - irq_list[irq].devname = devname; - return 0; -} - -void sys_free_irq(unsigned int irq, void *dev_id) -{ - if (irq > IRQ7) { - printk(KERN_WARNING "%s: Incorrect IRQ %d\n", __FUNCTION__, irq); - return; - } - - if (irq_list[irq].dev_id != dev_id) - printk(KERN_WARNING "%s: Removing probably wrong IRQ %d from %s\n", - __FUNCTION__, irq, irq_list[irq].devname); - - irq_list[irq].handler = mach_default_handler; - irq_list[irq].flags = 0; - irq_list[irq].dev_id = NULL; - irq_list[irq].devname = NULL; -} - -/* - * Do we need these probe functions on the m68k? - * - * ... may be useful with ISA devices - */ -unsigned long probe_irq_on (void) -{ - return 0; -} - -EXPORT_SYMBOL(probe_irq_on); - -int probe_irq_off (unsigned long irqs) -{ - return 0; -} - -EXPORT_SYMBOL(probe_irq_off); - -asmlinkage void process_int(unsigned long vec, struct pt_regs *fp) -{ - if (vec >= VEC_INT1 && vec <= VEC_INT7) { - vec -= VEC_SPUR; - kstat_cpu(0).irqs[vec]++; - irq_list[vec].handler(vec, irq_list[vec].dev_id); - } else { - if (mach_process_int) - mach_process_int(vec, fp); - else - panic("Can't process interrupt vector %ld\n", vec); - return; - } -} - - -int show_interrupts(struct seq_file *p, void *v) -{ - int i = *(loff_t *) v; - - if (i < NR_IRQS) { - if (! (irq_list[i].flags & IRQ_FLG_STD)) { - seq_printf(p, "%3d: %10u ", i, - (i ? kstat_cpu(0).irqs[i] : num_spurious)); - if (irq_list[i].flags & IRQ_FLG_LOCK) - seq_printf(p, "L "); - else - seq_printf(p, " "); - seq_printf(p, "%s\n", irq_list[i].devname); - } - } - - if (i == NR_IRQS && mach_get_irq_list) - mach_get_irq_list(p, v); - return 0; -} - -void init_irq_proc(void) -{ - /* Insert /proc/irq driver here */ -} - diff --git a/arch/m68knommu/platform/5307/vectors.c b/arch/m68knommu/platform/5307/vectors.c index 2a8b0d044ce..6cf89462023 100644 --- a/arch/m68knommu/platform/5307/vectors.c +++ b/arch/m68knommu/platform/5307/vectors.c @@ -3,23 +3,17 @@ /* * linux/arch/m68knommu/platform/5307/vectors.c * - * Copyright (C) 1999-2003, Greg Ungerer <gerg@snapgear.com> + * Copyright (C) 1999-2007, Greg Ungerer <gerg@snapgear.com> */ /***************************************************************************/ #include <linux/kernel.h> -#include <linux/sched.h> -#include <linux/param.h> #include <linux/init.h> -#include <linux/unistd.h> -#include <linux/delay.h> -#include <asm/irq.h> -#include <asm/dma.h> +#include <linux/irq.h> #include <asm/traps.h> #include <asm/machdep.h> #include <asm/coldfire.h> -#include <asm/mcftimer.h> #include <asm/mcfsim.h> #include <asm/mcfdma.h> #include <asm/mcfwdebug.h> @@ -56,7 +50,7 @@ asmlinkage void trap(void); asmlinkage void system_call(void); asmlinkage void inthandler(void); -void __init coldfire_trap_init(void) +void __init init_vectors(void) { int i; @@ -86,6 +80,23 @@ void __init coldfire_trap_init(void) /***************************************************************************/ +void enable_vector(unsigned int irq) +{ + /* Currently no action on ColdFire */ +} + +void disable_vector(unsigned int irq) +{ + /* Currently no action on ColdFire */ +} + +void ack_vector(unsigned int irq) +{ + /* Currently no action on ColdFire */ +} + +/***************************************************************************/ + void coldfire_reset(void) { HARD_RESET_NOW(); diff --git a/arch/m68knommu/platform/68328/entry.S b/arch/m68knommu/platform/68328/entry.S index f9786271545..b1aef72f3ba 100644 --- a/arch/m68knommu/platform/68328/entry.S +++ b/arch/m68knommu/platform/68328/entry.S @@ -133,7 +133,6 @@ Lreturn: */ inthandler1: SAVE_ALL - addql #1,local_irq_count /* put exception # in d0*/ movew %sp@(PT_VECTOR), %d0 and #0x3ff, %d0 @@ -145,7 +144,6 @@ inthandler1: inthandler2: SAVE_ALL - addql #1,local_irq_count /* put exception # in d0*/ movew %sp@(PT_VECTOR), %d0 and #0x3ff, %d0 @@ -157,7 +155,6 @@ inthandler2: inthandler3: SAVE_ALL - addql #1,local_irq_count /* put exception # in d0*/ movew %sp@(PT_VECTOR), %d0 and #0x3ff, %d0 @@ -169,7 +166,6 @@ inthandler3: inthandler4: SAVE_ALL - addql #1,local_irq_count /* put exception # in d0*/ movew %sp@(PT_VECTOR), %d0 and #0x3ff, %d0 @@ -181,7 +177,6 @@ inthandler4: inthandler5: SAVE_ALL - addql #1,local_irq_count /* put exception # in d0*/ movew %sp@(PT_VECTOR), %d0 and #0x3ff, %d0 @@ -193,7 +188,6 @@ inthandler5: inthandler6: SAVE_ALL - addql #1,local_irq_count /* put exception # in d0*/ movew %sp@(PT_VECTOR), %d0 and #0x3ff, %d0 @@ -205,7 +199,6 @@ inthandler6: inthandler7: SAVE_ALL - addql #1,local_irq_count /* put exception # in d0*/ movew %sp@(PT_VECTOR), %d0 and #0x3ff, %d0 @@ -217,7 +210,6 @@ inthandler7: inthandler: SAVE_ALL - addql #1,local_irq_count /* put exception # in d0*/ movew %sp@(PT_VECTOR), %d0 and #0x3ff, %d0 @@ -228,7 +220,6 @@ inthandler: bra ret_from_interrupt ret_from_interrupt: - subql #1,local_irq_count jeq 1f 2: RESTORE_ALL @@ -238,7 +229,6 @@ ret_from_interrupt: jhi 2b /* check if we need to do software interrupts */ - movel local_irq_count,%d0 jeq ret_from_exception pea ret_from_exception diff --git a/arch/m68knommu/platform/68328/ints.c b/arch/m68knommu/platform/68328/ints.c index 3de6e337554..72e56d554f4 100644 --- a/arch/m68knommu/platform/68328/ints.c +++ b/arch/m68knommu/platform/68328/ints.c @@ -9,21 +9,14 @@ * Copyright 1999 D. Jeff Dionne <jeff@rt-control.com> */ -#include <linux/module.h> #include <linux/types.h> #include <linux/kernel.h> -#include <linux/sched.h> -#include <linux/kernel_stat.h> -#include <linux/errno.h> +#include <linux/init.h> #include <linux/interrupt.h> - -#include <asm/system.h> -#include <asm/irq.h> -#include <asm/irqnode.h> +#include <linux/irq.h> #include <asm/traps.h> #include <asm/io.h> #include <asm/machdep.h> -#include <asm/setup.h> #if defined(CONFIG_M68328) #include <asm/MC68328.h> @@ -79,16 +72,12 @@ extern e_vector *_ramvec; /* The number of spurious interrupts */ volatile unsigned int num_spurious; -unsigned int local_irq_count[NR_CPUS]; - -/* irq node variables for the 32 (potential) on chip sources */ -static irq_node_t int_irq_list[NR_IRQS]; /* * This function should be called during kernel startup to initialize - * the IRQ handling routines. + * the machine vector table. */ -void init_IRQ(void) +void __init init_vectors(void) { int i; @@ -108,96 +97,10 @@ void init_IRQ(void) IVR = 0x40; /* Set DragonBall IVR (interrupt base) to 64 */ - /* initialize handlers */ - for (i = 0; i < NR_IRQS; i++) { - int_irq_list[i].handler = bad_interrupt; - int_irq_list[i].flags = IRQ_FLG_STD; - int_irq_list[i].dev_id = NULL; - int_irq_list[i].devname = NULL; - } - /* turn off all interrupts */ IMR = ~0; } -int request_irq( - unsigned int irq, - irq_handler_t handler, - unsigned long flags, - const char *devname, - void *dev_id) -{ - if (irq >= NR_IRQS) { - printk (KERN_ERR "%s: Unknown IRQ %d from %s\n", __FUNCTION__, irq, devname); - return -ENXIO; - } - - if (!(int_irq_list[irq].flags & IRQ_FLG_STD)) { - if (int_irq_list[irq].flags & IRQ_FLG_LOCK) { - printk(KERN_ERR "%s: IRQ %d from %s is not replaceable\n", - __FUNCTION__, irq, int_irq_list[irq].devname); - return -EBUSY; - } - if (flags & IRQ_FLG_REPLACE) { - printk(KERN_ERR "%s: %s can't replace IRQ %d from %s\n", - __FUNCTION__, devname, irq, int_irq_list[irq].devname); - return -EBUSY; - } - } - - int_irq_list[irq].handler = handler; - int_irq_list[irq].flags = flags; - int_irq_list[irq].dev_id = dev_id; - int_irq_list[irq].devname = devname; - - IMR &= ~(1<<irq); - - return 0; -} - -EXPORT_SYMBOL(request_irq); - -void free_irq(unsigned int irq, void *dev_id) -{ - if (irq >= NR_IRQS) { - printk (KERN_ERR "%s: Unknown IRQ %d\n", __FUNCTION__, irq); - return; - } - - if (int_irq_list[irq].dev_id != dev_id) - printk(KERN_INFO "%s: removing probably wrong IRQ %d from %s\n", - __FUNCTION__, irq, int_irq_list[irq].devname); - - int_irq_list[irq].handler = bad_interrupt; - int_irq_list[irq].flags = IRQ_FLG_STD; - int_irq_list[irq].dev_id = NULL; - int_irq_list[irq].devname = NULL; - - IMR |= 1<<irq; -} - -EXPORT_SYMBOL(free_irq); - -int show_interrupts(struct seq_file *p, void *v) -{ - int i = *(loff_t *) v; - - if (i < NR_IRQS) { - if (int_irq_list[i].devname) { - seq_printf(p, "%3d: %10u ", i, kstat_cpu(0).irqs[i]); - if (int_irq_list[i].flags & IRQ_FLG_LOCK) - seq_printf(p, "L "); - else - seq_printf(p, " "); - seq_printf(p, "%s\n", int_irq_list[i].devname); - } - } - if (i == NR_IRQS) - seq_printf(p, " : %10u spurious\n", num_spurious); - - return 0; -} - /* The 68k family did not have a good way to determine the source * of interrupts until later in the family. The EC000 core does * not provide the vector number on the stack, we vector everything @@ -255,14 +158,23 @@ void process_int(int vec, struct pt_regs *fp) irq++; } - kstat_cpu(0).irqs[irq]++; - - if (int_irq_list[irq].handler) { - int_irq_list[irq].handler(irq, int_irq_list[irq].dev_id, fp); - } else { - printk(KERN_ERR "unregistered interrupt %d!\nTurning it off in the IMR...\n", irq); - IMR |= mask; - } + do_IRQ(irq, fp); pend &= ~mask; } } + +void enable_vector(unsigned int irq) +{ + IMR &= ~(1<<irq); +} + +void disable_vector(unsigned int irq) +{ + IMR |= (1<<irq); +} + +void ack_vector(unsigned int irq) +{ + /* Nothing needed */ +} + diff --git a/arch/m68knommu/platform/68360/entry.S b/arch/m68knommu/platform/68360/entry.S index f1af8977f29..55dfefe3864 100644 --- a/arch/m68knommu/platform/68360/entry.S +++ b/arch/m68knommu/platform/68360/entry.S @@ -120,23 +120,21 @@ Lreturn: RESTORE_ALL /* - * This is the main interrupt handler, responsible for calling process_int() + * This is the main interrupt handler, responsible for calling do_IRQ() */ inthandler: SAVE_ALL - addql #1,local_irq_count /* put exception # in d0*/ movew %sp@(PT_VECTOR), %d0 and.l #0x3ff, %d0 lsr.l #0x02, %d0 movel %sp,%sp@- movel %d0,%sp@- /* put vector # on stack*/ - jbsr process_int /* process the IRQ*/ + jbsr do_IRQ /* process the IRQ*/ 3: addql #8,%sp /* pop parameters off stack*/ bra ret_from_interrupt ret_from_interrupt: - subql #1,local_irq_count jeq 1f 2: RESTORE_ALL diff --git a/arch/m68knommu/platform/68360/ints.c b/arch/m68knommu/platform/68360/ints.c index 4df3c146eb7..c36781157e0 100644 --- a/arch/m68knommu/platform/68360/ints.c +++ b/arch/m68knommu/platform/68360/ints.c @@ -10,20 +10,13 @@ * Copyright (c) 1999 D. Jeff Dionne <jeff@uclinux.org> */ -#include <linux/module.h> #include <linux/types.h> #include <linux/kernel.h> -#include <linux/sched.h> -#include <linux/kernel_stat.h> -#include <linux/errno.h> - -#include <asm/system.h> -#include <asm/irq.h> -#include <asm/irqnode.h> +#include <linux/init.h> +#include <linux/interrupt.h> +#include <linux/irq.h> #include <asm/traps.h> -#include <asm/io.h> #include <asm/machdep.h> -#include <asm/setup.h> #include <asm/m68360.h> /* from quicc/commproc.c: */ @@ -36,26 +29,19 @@ extern void cpm_interrupt_init(void); asmlinkage void system_call(void); asmlinkage void buserr(void); asmlinkage void trap(void); -asmlinkage irqreturn_t bad_interrupt(void); -asmlinkage irqreturn_t inthandler(void); +asmlinkage void bad_interrupt(void); +asmlinkage void inthandler(void); extern void *_ramvec[]; /* The number of spurious interrupts */ volatile unsigned int num_spurious; -unsigned int local_irq_count[NR_CPUS]; - -/* irq node variables for the 32 (potential) on chip sources */ -static irq_node_t int_irq_list[INTERNAL_IRQS]; - -static short int_irq_ablecount[INTERNAL_IRQS]; /* * This function should be called during kernel startup to initialize - * IRQ handling routines. + * the vector table. */ - -void init_IRQ(void) +void init_vectors(void) { int i; int vba = (CPM_VECTOR_BASE<<4); @@ -79,7 +65,6 @@ void init_IRQ(void) _ramvec[32] = system_call; _ramvec[33] = trap; - cpm_interrupt_init(); /* set up CICR for vector base address and irq level */ @@ -124,212 +109,20 @@ void init_IRQ(void) /* turn off all CPM interrupts */ pquicc->intr_cimr = 0x00000000; - - /* initialize handlers */ - for (i = 0; i < INTERNAL_IRQS; i++) { - int_irq_list[i].handler = NULL; - int_irq_list[i].flags = IRQ_FLG_STD; - int_irq_list[i].dev_id = NULL; - int_irq_list[i].devname = NULL; - } -} - -#if 0 -void M68360_insert_irq(irq_node_t **list, irq_node_t *node) -{ - unsigned long flags; - irq_node_t *cur; - - if (!node->dev_id) - printk(KERN_INFO "%s: Warning: dev_id of %s is zero\n", - __FUNCTION__, node->devname); - - local_irq_save(flags); - - cur = *list; - - while (cur) { - list = &cur->next; - cur = cur->next; - } - - node->next = cur; - *list = node; - - local_irq_restore(flags); } -void M68360_delete_irq(irq_node_t **list, void *dev_id) +void enable_vector(unsigned int irq) { - unsigned long flags; - irq_node_t *node; - - local_irq_save(flags); - - for (node = *list; node; list = &node->next, node = *list) { - if (node->dev_id == dev_id) { - *list = node->next; - /* Mark it as free. */ - node->handler = NULL; - local_irq_restore(flags); - return; - } - } - local_irq_restore(flags); - printk (KERN_INFO "%s: tried to remove invalid irq\n", __FUNCTION__); + pquicc->intr_cimr |= (1 << irq); } -#endif -int request_irq( - unsigned int irq, - irqreturn_t (*handler)(int, void *, struct pt_regs *), - unsigned long flags, - const char *devname, - void *dev_id) +void disable_vector(unsigned int irq) { - int mask = (1<<irq); - - irq += (CPM_VECTOR_BASE<<4); - - if (irq >= INTERNAL_IRQS) { - printk (KERN_ERR "%s: Unknown IRQ %d from %s\n", __FUNCTION__, irq, devname); - return -ENXIO; - } - - if (!(int_irq_list[irq].flags & IRQ_FLG_STD)) { - if (int_irq_list[irq].flags & IRQ_FLG_LOCK) { - printk(KERN_ERR "%s: IRQ %d from %s is not replaceable\n", - __FUNCTION__, irq, int_irq_list[irq].devname); - return -EBUSY; - } - if (flags & IRQ_FLG_REPLACE) { - printk(KERN_ERR "%s: %s can't replace IRQ %d from %s\n", - __FUNCTION__, devname, irq, int_irq_list[irq].devname); - return -EBUSY; - } - } - int_irq_list[irq].handler = handler; - int_irq_list[irq].flags = flags; - int_irq_list[irq].dev_id = dev_id; - int_irq_list[irq].devname = devname; - - /* enable in the CIMR */ - if (!int_irq_ablecount[irq]) - pquicc->intr_cimr |= mask; - /* *(volatile unsigned long *)0xfffff304 &= ~(1<<irq); */ - - return 0; + pquicc->intr_cimr &= ~(1 << irq); } -EXPORT_SYMBOL(request_irq); - -void free_irq(unsigned int irq, void *dev_id) +void ack_vector(unsigned int irq) { - if (irq >= INTERNAL_IRQS) { - printk (KERN_ERR "%s: Unknown IRQ %d\n", __FUNCTION__, irq); - return; - } - - if (int_irq_list[irq].dev_id != dev_id) - printk(KERN_INFO "%s: removing probably wrong IRQ %d from %s\n", - __FUNCTION__, irq, int_irq_list[irq].devname); - int_irq_list[irq].handler = NULL; - int_irq_list[irq].flags = IRQ_FLG_STD; - int_irq_list[irq].dev_id = NULL; - int_irq_list[irq].devname = NULL; - - *(volatile unsigned long *)0xfffff304 |= 1<<irq; + pquicc->intr_cisr = (1 << irq); } -EXPORT_SYMBOL(free_irq); - -#if 0 -/* - * Enable/disable a particular machine specific interrupt source. - * Note that this may affect other interrupts in case of a shared interrupt. - * This function should only be called for a _very_ short time to change some - * internal data, that may not be changed by the interrupt at the same time. - * int_(enable|disable)_irq calls may also be nested. - */ -void M68360_enable_irq(unsigned int irq) -{ - if (irq >= INTERNAL_IRQS) { - printk(KERN_ERR "%s: Unknown IRQ %d\n", __FUNCTION__, irq); - return; - } - - if (--int_irq_ablecount[irq]) - return; - - /* enable the interrupt */ - *(volatile unsigned long *)0xfffff304 &= ~(1<<irq); -} - -void M68360_disable_irq(unsigned int irq) -{ - if (irq >= INTERNAL_IRQS) { - printk(KERN_ERR "%s: Unknown IRQ %d\n", __FUNCTION__, irq); - return; - } - - if (int_irq_ablecount[irq]++) - return; - - /* disable the interrupt */ - *(volatile unsigned long *)0xfffff304 |= 1<<irq; -} -#endif - -int show_interrupts(struct seq_file *p, void *v) -{ - int i = *(loff_t *) v; - - if (i < NR_IRQS) { - if (int_irq_list[i].devname) { - seq_printf(p, "%3d: %10u ", i, kstat_cpu(0).irqs[i]); - if (int_irq_list[i].flags & IRQ_FLG_LOCK) - seq_printf(p, "L "); - else - seq_printf(p, " "); - seq_printf(p, "%s\n", int_irq_list[i].devname); - } - } - if (i == NR_IRQS) - seq_printf(p, " : %10u spurious\n", num_spurious); - - return 0; -} - -/* The 68k family did not have a good way to determine the source - * of interrupts until later in the family. The EC000 core does - * not provide the vector number on the stack, we vector everything - * into one vector and look in the blasted mask register... - * This code is designed to be fast, almost constant time, not clean! - */ -void process_int(int vec, struct pt_regs *fp) -{ - int irq; - int mask; - - /* unsigned long pend = *(volatile unsigned long *)0xfffff30c; */ - - /* irq = vec + (CPM_VECTOR_BASE<<4); */ - irq = vec; - - /* unsigned long pend = *(volatile unsigned long *)pquicc->intr_cipr; */ - - /* Bugger all that weirdness. For the moment, I seem to know where I came from; - * vec is passed from a specific ISR, so I'll use it. */ - - if (int_irq_list[irq].handler) { - int_irq_list[irq].handler(irq , int_irq_list[irq].dev_id, fp); - kstat_cpu(0).irqs[irq]++; - pquicc->intr_cisr = (1 << vec); /* indicate that irq has been serviced */ - } else { - printk(KERN_ERR "unregistered interrupt %d!\nTurning it off in the CIMR...\n", irq); - /* *(volatile unsigned long *)0xfffff304 |= mask; */ - pquicc->intr_cimr &= ~(1 << vec); - num_spurious += 1; - } - return(IRQ_HANDLED); -} diff --git a/arch/mips/kernel/vmlinux.lds.S b/arch/mips/kernel/vmlinux.lds.S index 9b9992cd562..bc9bae2a73f 100644 --- a/arch/mips/kernel/vmlinux.lds.S +++ b/arch/mips/kernel/vmlinux.lds.S @@ -119,10 +119,7 @@ SECTIONS .init.ramfs : { *(.init.ramfs) } __initramfs_end = .; #endif - . = ALIGN(_PAGE_SIZE); - __per_cpu_start = .; - .data.percpu : { *(.data.percpu) } - __per_cpu_end = .; + PERCPU(_PAGE_SIZE) . = ALIGN(_PAGE_SIZE); __init_end = .; /* freed after init ends here */ diff --git a/arch/mips/mm/fault.c b/arch/mips/mm/fault.c index 7ebea331edb..521771b373d 100644 --- a/arch/mips/mm/fault.c +++ b/arch/mips/mm/fault.c @@ -39,6 +39,7 @@ asmlinkage void do_page_fault(struct pt_regs *regs, unsigned long write, struct mm_struct *mm = tsk->mm; const int field = sizeof(unsigned long) * 2; siginfo_t info; + int fault; #if 0 printk("Cpu%d[%s:%d:%0*lx:%ld:%0*lx]\n", raw_smp_processor_id(), @@ -102,20 +103,18 @@ survive: * make sure we exit gracefully rather than endlessly redo * the fault. */ - switch (handle_mm_fault(mm, vma, address, write)) { - case VM_FAULT_MINOR: - tsk->min_flt++; - break; - case VM_FAULT_MAJOR: - tsk->maj_flt++; - break; - case VM_FAULT_SIGBUS: - goto do_sigbus; - case VM_FAULT_OOM: - goto out_of_memory; - default: + fault = handle_mm_fault(mm, vma, address, write); + if (unlikely(fault & VM_FAULT_ERROR)) { + if (fault & VM_FAULT_OOM) + goto out_of_memory; + else if (fault & VM_FAULT_SIGBUS) + goto do_sigbus; BUG(); } + if (fault & VM_FAULT_MAJOR) + tsk->maj_flt++; + else + tsk->min_flt++; up_read(&mm->mmap_sem); return; diff --git a/arch/parisc/kernel/vmlinux.lds.S b/arch/parisc/kernel/vmlinux.lds.S index 4d96ba4b984..d4e6a93c8d9 100644 --- a/arch/parisc/kernel/vmlinux.lds.S +++ b/arch/parisc/kernel/vmlinux.lds.S @@ -181,10 +181,9 @@ SECTIONS .init.ramfs : { *(.init.ramfs) } __initramfs_end = .; #endif - . = ALIGN(ASM_PAGE_SIZE); - __per_cpu_start = .; - .data.percpu : { *(.data.percpu) } - __per_cpu_end = .; + + PERCPU(ASM_PAGE_SIZE) + . = ALIGN(ASM_PAGE_SIZE); __init_end = .; /* freed after init ends here */ diff --git a/arch/parisc/mm/fault.c b/arch/parisc/mm/fault.c index f6f67554c62..7899ab87785 100644 --- a/arch/parisc/mm/fault.c +++ b/arch/parisc/mm/fault.c @@ -147,6 +147,7 @@ void do_page_fault(struct pt_regs *regs, unsigned long code, struct mm_struct *mm = tsk->mm; const struct exception_table_entry *fix; unsigned long acc_type; + int fault; if (in_atomic() || !mm) goto no_context; @@ -173,23 +174,23 @@ good_area: * fault. */ - switch (handle_mm_fault(mm, vma, address, (acc_type & VM_WRITE) != 0)) { - case VM_FAULT_MINOR: - ++current->min_flt; - break; - case VM_FAULT_MAJOR: - ++current->maj_flt; - break; - case VM_FAULT_SIGBUS: + fault = handle_mm_fault(mm, vma, address, (acc_type & VM_WRITE) != 0); + if (unlikely(fault & VM_FAULT_ERROR)) { /* * We hit a shared mapping outside of the file, or some * other thing happened to us that made us unable to * handle the page fault gracefully. */ - goto bad_area; - default: - goto out_of_memory; + if (fault & VM_FAULT_OOM) + goto out_of_memory; + else if (fault & VM_FAULT_SIGBUS) + goto bad_area; + BUG(); } + if (fault & VM_FAULT_MAJOR) + current->maj_flt++; + else + current->min_flt++; up_read(&mm->mmap_sem); return; diff --git a/arch/powerpc/Makefile b/arch/powerpc/Makefile index 187a39af3e1..6c1e36c33fa 100644 --- a/arch/powerpc/Makefile +++ b/arch/powerpc/Makefile @@ -201,6 +201,14 @@ checkbin: false; \ fi ; \ fi + @if test "$(call cc-fullversion)" = "040200" \ + && test "x${CONFIG_MODULES}${CONFIG_PPC64}" = "xyy" ; then \ + echo -n '*** GCC-4.2.0 cannot compile the 64-bit powerpc ' ; \ + echo 'kernel with modules enabled.' ; \ + echo -n '*** Please use a different GCC version or ' ; \ + echo 'disable kernel modules' ; \ + false ; \ + fi @if ! /bin/echo dssall | $(AS) -many -o $(TOUT) >/dev/null 2>&1 ; then \ echo -n '*** ${VERSION}.${PATCHLEVEL} kernels no longer build ' ; \ echo 'correctly with old versions of binutils.' ; \ diff --git a/arch/powerpc/kernel/kprobes.c b/arch/powerpc/kernel/kprobes.c index 0c96611f02f..440f5a87271 100644 --- a/arch/powerpc/kernel/kprobes.c +++ b/arch/powerpc/kernel/kprobes.c @@ -492,6 +492,13 @@ int __kprobes kprobe_exceptions_notify(struct notifier_block *self, return ret; } +#ifdef CONFIG_PPC64 +unsigned long arch_deref_entry_point(void *entry) +{ + return (unsigned long)(((func_descr_t *)entry)->entry); +} +#endif + int __kprobes setjmp_pre_handler(struct kprobe *p, struct pt_regs *regs) { struct jprobe *jp = container_of(p, struct jprobe, kp); @@ -500,11 +507,9 @@ int __kprobes setjmp_pre_handler(struct kprobe *p, struct pt_regs *regs) memcpy(&kcb->jprobe_saved_regs, regs, sizeof(struct pt_regs)); /* setup return addr to the jprobe handler routine */ + regs->nip = arch_deref_entry_point(jp->entry); #ifdef CONFIG_PPC64 - regs->nip = (unsigned long)(((func_descr_t *)jp->entry)->entry); regs->gpr[2] = (unsigned long)(((func_descr_t *)jp->entry)->toc); -#else - regs->nip = (unsigned long)jp->entry; #endif return 1; diff --git a/arch/powerpc/kernel/lparcfg.c b/arch/powerpc/kernel/lparcfg.c index c492cee90e0..6444eaa30a2 100644 --- a/arch/powerpc/kernel/lparcfg.c +++ b/arch/powerpc/kernel/lparcfg.c @@ -248,7 +248,7 @@ static void parse_system_parameter_string(struct seq_file *m) } else { int splpar_strlen; int idx, w_idx; - char *workbuffer = kmalloc(SPLPAR_MAXLENGTH, GFP_KERNEL); + char *workbuffer = kzalloc(SPLPAR_MAXLENGTH, GFP_KERNEL); if (!workbuffer) { printk(KERN_ERR "%s %s kmalloc failure at line %d \n", __FILE__, __FUNCTION__, __LINE__); @@ -261,7 +261,6 @@ static void parse_system_parameter_string(struct seq_file *m) splpar_strlen = local_buffer[0] * 256 + local_buffer[1]; local_buffer += 2; /* step over strlen value */ - memset(workbuffer, 0, SPLPAR_MAXLENGTH); w_idx = 0; idx = 0; while ((*local_buffer) && (idx < splpar_strlen)) { diff --git a/arch/powerpc/kernel/of_platform.c b/arch/powerpc/kernel/of_platform.c index 9536ed7f247..8ded4e7dc87 100644 --- a/arch/powerpc/kernel/of_platform.c +++ b/arch/powerpc/kernel/of_platform.c @@ -222,10 +222,9 @@ struct of_device* of_platform_device_create(struct device_node *np, { struct of_device *dev; - dev = kmalloc(sizeof(*dev), GFP_KERNEL); + dev = kzalloc(sizeof(*dev), GFP_KERNEL); if (!dev) return NULL; - memset(dev, 0, sizeof(*dev)); dev->node = of_node_get(np); dev->dma_mask = 0xffffffffUL; diff --git a/arch/powerpc/kernel/vmlinux.lds.S b/arch/powerpc/kernel/vmlinux.lds.S index ae4acd84143..0c458556399 100644 --- a/arch/powerpc/kernel/vmlinux.lds.S +++ b/arch/powerpc/kernel/vmlinux.lds.S @@ -63,6 +63,8 @@ SECTIONS __stop___ex_table = .; } + NOTES + BUG_TABLE /* @@ -144,6 +146,7 @@ SECTIONS .data.percpu : { __per_cpu_start = .; *(.data.percpu) + *(.data.percpu.shared_aligned) __per_cpu_end = .; } @@ -174,7 +177,9 @@ SECTIONS } #else .data : { - *(.data .data.rel* .toc1) + DATA_DATA + *(.data.rel*) + *(.toc1) *(.branch_lt) } diff --git a/arch/powerpc/mm/fault.c b/arch/powerpc/mm/fault.c index 0ece51310bf..3767211b3d0 100644 --- a/arch/powerpc/mm/fault.c +++ b/arch/powerpc/mm/fault.c @@ -145,7 +145,7 @@ int __kprobes do_page_fault(struct pt_regs *regs, unsigned long address, struct mm_struct *mm = current->mm; siginfo_t info; int code = SEGV_MAPERR; - int is_write = 0; + int is_write = 0, ret; int trap = TRAP(regs); int is_exec = trap == 0x400; @@ -330,22 +330,18 @@ good_area: * the fault. */ survive: - switch (handle_mm_fault(mm, vma, address, is_write)) { - - case VM_FAULT_MINOR: - current->min_flt++; - break; - case VM_FAULT_MAJOR: - current->maj_flt++; - break; - case VM_FAULT_SIGBUS: - goto do_sigbus; - case VM_FAULT_OOM: - goto out_of_memory; - default: + ret = handle_mm_fault(mm, vma, address, is_write); + if (unlikely(ret & VM_FAULT_ERROR)) { + if (ret & VM_FAULT_OOM) + goto out_of_memory; + else if (ret & VM_FAULT_SIGBUS) + goto do_sigbus; BUG(); } - + if (ret & VM_FAULT_MAJOR) + current->maj_flt++; + else + current->min_flt++; up_read(&mm->mmap_sem); return 0; diff --git a/arch/powerpc/platforms/cell/spufs/fault.c b/arch/powerpc/platforms/cell/spufs/fault.c index e064d0c0d80..f53a0743747 100644 --- a/arch/powerpc/platforms/cell/spufs/fault.c +++ b/arch/powerpc/platforms/cell/spufs/fault.c @@ -75,22 +75,20 @@ good_area: } ret = 0; *flt = handle_mm_fault(mm, vma, ea, is_write); - switch (*flt) { - case VM_FAULT_MINOR: - current->min_flt++; - break; - case VM_FAULT_MAJOR: - current->maj_flt++; - break; - case VM_FAULT_SIGBUS: - ret = -EFAULT; - goto bad_area; - case VM_FAULT_OOM: - ret = -ENOMEM; - goto bad_area; - default: + if (unlikely(*flt & VM_FAULT_ERROR)) { + if (*flt & VM_FAULT_OOM) { + ret = -ENOMEM; + goto bad_area; + } else if (*flt & VM_FAULT_SIGBUS) { + ret = -EFAULT; + goto bad_area; + } BUG(); } + if (*flt & VM_FAULT_MAJOR) + current->maj_flt++; + else + current->min_flt++; up_read(&mm->mmap_sem); return ret; @@ -212,15 +210,15 @@ int spufs_handle_class1(struct spu_context *ctx) * In case of unhandled error report the problem to user space. */ if (!ret) { - if (flt == VM_FAULT_MINOR) - ctx->stats.min_flt++; - else + if (flt & VM_FAULT_MAJOR) ctx->stats.maj_flt++; + else + ctx->stats.min_flt++; if (ctx->state == SPU_STATE_RUNNABLE) { - if (flt == VM_FAULT_MINOR) - ctx->spu->stats.min_flt++; - else + if (flt & VM_FAULT_MAJOR) ctx->spu->stats.maj_flt++; + else + ctx->spu->stats.min_flt++; } if (ctx->spu) diff --git a/arch/ppc/configs/ev64260_defconfig b/arch/ppc/configs/ev64260_defconfig index 84cc142a67b..587e9a3b949 100644 --- a/arch/ppc/configs/ev64260_defconfig +++ b/arch/ppc/configs/ev64260_defconfig @@ -531,7 +531,6 @@ CONFIG_I2C_CHARDEV=m # CONFIG_I2C_AMD8111 is not set # CONFIG_I2C_I801 is not set # CONFIG_I2C_I810 is not set -# CONFIG_I2C_ISA is not set # CONFIG_I2C_NFORCE2 is not set # CONFIG_I2C_PARPORT_LIGHT is not set # CONFIG_I2C_PIIX4 is not set diff --git a/arch/ppc/configs/mpc8540_ads_defconfig b/arch/ppc/configs/mpc8540_ads_defconfig index c5c86025e26..bf676ebd99a 100644 --- a/arch/ppc/configs/mpc8540_ads_defconfig +++ b/arch/ppc/configs/mpc8540_ads_defconfig @@ -452,7 +452,6 @@ CONFIG_I2C_CHARDEV=y # CONFIG_I2C_AMD8111 is not set # CONFIG_I2C_I801 is not set # CONFIG_I2C_I810 is not set -# CONFIG_I2C_ISA is not set CONFIG_I2C_MPC=y # CONFIG_I2C_NFORCE2 is not set # CONFIG_I2C_PARPORT_LIGHT is not set diff --git a/arch/ppc/configs/mpc8548_cds_defconfig b/arch/ppc/configs/mpc8548_cds_defconfig index abe034f24b8..f36fc5db540 100644 --- a/arch/ppc/configs/mpc8548_cds_defconfig +++ b/arch/ppc/configs/mpc8548_cds_defconfig @@ -413,7 +413,6 @@ CONFIG_I2C_CHARDEV=y # # I2C Hardware Bus support # -# CONFIG_I2C_ISA is not set CONFIG_I2C_MPC=y # CONFIG_I2C_PARPORT_LIGHT is not set # CONFIG_I2C_PCA_ISA is not set diff --git a/arch/ppc/configs/mpc8555_cds_defconfig b/arch/ppc/configs/mpc8555_cds_defconfig index 15abebf46b9..4f1e320acfb 100644 --- a/arch/ppc/configs/mpc8555_cds_defconfig +++ b/arch/ppc/configs/mpc8555_cds_defconfig @@ -518,7 +518,6 @@ CONFIG_I2C_CHARDEV=y # CONFIG_I2C_I801 is not set # CONFIG_I2C_I810 is not set # CONFIG_I2C_PIIX4 is not set -# CONFIG_I2C_ISA is not set CONFIG_I2C_MPC=y # CONFIG_I2C_NFORCE2 is not set # CONFIG_I2C_PARPORT_LIGHT is not set diff --git a/arch/ppc/configs/mpc8560_ads_defconfig b/arch/ppc/configs/mpc8560_ads_defconfig index f834fb541ad..f12d48fcbba 100644 --- a/arch/ppc/configs/mpc8560_ads_defconfig +++ b/arch/ppc/configs/mpc8560_ads_defconfig @@ -489,7 +489,6 @@ CONFIG_I2C_CHARDEV=y # CONFIG_I2C_I801 is not set # CONFIG_I2C_I810 is not set # CONFIG_I2C_PIIX4 is not set -# CONFIG_I2C_ISA is not set CONFIG_I2C_MPC=y # CONFIG_I2C_NFORCE2 is not set # CONFIG_I2C_PARPORT_LIGHT is not set diff --git a/arch/ppc/configs/radstone_ppc7d_defconfig b/arch/ppc/configs/radstone_ppc7d_defconfig index ca4d1fd0ca0..9f64532f2a8 100644 --- a/arch/ppc/configs/radstone_ppc7d_defconfig +++ b/arch/ppc/configs/radstone_ppc7d_defconfig @@ -710,7 +710,6 @@ CONFIG_I2C_CHARDEV=y # CONFIG_I2C_I801 is not set # CONFIG_I2C_I810 is not set # CONFIG_I2C_PIIX4 is not set -# CONFIG_I2C_ISA is not set # CONFIG_I2C_MPC is not set # CONFIG_I2C_NFORCE2 is not set # CONFIG_I2C_PARPORT_LIGHT is not set diff --git a/arch/ppc/configs/stx_gp3_defconfig b/arch/ppc/configs/stx_gp3_defconfig index 3fedc43e44a..70d6f842aa9 100644 --- a/arch/ppc/configs/stx_gp3_defconfig +++ b/arch/ppc/configs/stx_gp3_defconfig @@ -661,7 +661,6 @@ CONFIG_I2C_ALGOBIT=m # CONFIG_I2C_I801 is not set # CONFIG_I2C_I810 is not set # CONFIG_I2C_PIIX4 is not set -# CONFIG_I2C_ISA is not set # CONFIG_I2C_MPC is not set # CONFIG_I2C_NFORCE2 is not set # CONFIG_I2C_PARPORT is not set diff --git a/arch/ppc/configs/sycamore_defconfig b/arch/ppc/configs/sycamore_defconfig index 758114cfea5..6996cca18f3 100644 --- a/arch/ppc/configs/sycamore_defconfig +++ b/arch/ppc/configs/sycamore_defconfig @@ -461,7 +461,6 @@ CONFIG_I2C_CHARDEV=y # CONFIG_I2C_I801 is not set # CONFIG_I2C_I810 is not set # CONFIG_I2C_IBM_IIC is not set -# CONFIG_I2C_ISA is not set # CONFIG_I2C_NFORCE2 is not set # CONFIG_I2C_PARPORT_LIGHT is not set # CONFIG_I2C_PIIX4 is not set diff --git a/arch/ppc/kernel/vmlinux.lds.S b/arch/ppc/kernel/vmlinux.lds.S index 19db8746ff1..c0aac3ff9e9 100644 --- a/arch/ppc/kernel/vmlinux.lds.S +++ b/arch/ppc/kernel/vmlinux.lds.S @@ -130,10 +130,7 @@ SECTIONS __ftr_fixup : { *(__ftr_fixup) } __stop___ftr_fixup = .; - . = ALIGN(4096); - __per_cpu_start = .; - .data.percpu : { *(.data.percpu) } - __per_cpu_end = .; + PERCPU(4096) #ifdef CONFIG_BLK_DEV_INITRD . = ALIGN(4096); diff --git a/arch/ppc/mm/fault.c b/arch/ppc/mm/fault.c index 465f451f3bc..b98244e277f 100644 --- a/arch/ppc/mm/fault.c +++ b/arch/ppc/mm/fault.c @@ -96,6 +96,7 @@ int do_page_fault(struct pt_regs *regs, unsigned long address, struct mm_struct *mm = current->mm; siginfo_t info; int code = SEGV_MAPERR; + int fault; #if defined(CONFIG_4xx) || defined (CONFIG_BOOKE) int is_write = error_code & ESR_DST; #else @@ -249,20 +250,18 @@ good_area: * the fault. */ survive: - switch (handle_mm_fault(mm, vma, address, is_write)) { - case VM_FAULT_MINOR: - current->min_flt++; - break; - case VM_FAULT_MAJOR: - current->maj_flt++; - break; - case VM_FAULT_SIGBUS: - goto do_sigbus; - case VM_FAULT_OOM: - goto out_of_memory; - default: + fault = handle_mm_fault(mm, vma, address, is_write); + if (unlikely(fault & VM_FAULT_ERROR)) { + if (fault & VM_FAULT_OOM) + goto out_of_memory; + else if (fault & VM_FAULT_SIGBUS) + goto do_sigbus; BUG(); } + if (fault & VM_FAULT_MAJOR) + current->maj_flt++; + else + current->min_flt++; up_read(&mm->mmap_sem); /* diff --git a/arch/s390/kernel/vmlinux.lds.S b/arch/s390/kernel/vmlinux.lds.S index 7158a804a5e..6ab7d4ee13a 100644 --- a/arch/s390/kernel/vmlinux.lds.S +++ b/arch/s390/kernel/vmlinux.lds.S @@ -45,6 +45,8 @@ SECTIONS __ex_table : { *(__ex_table) } __stop___ex_table = .; + NOTES + BUG_TABLE .data : { /* Data */ @@ -107,10 +109,7 @@ SECTIONS . = ALIGN(2); __initramfs_end = .; #endif - . = ALIGN(4096); - __per_cpu_start = .; - .data.percpu : { *(.data.percpu) } - __per_cpu_end = .; + PERCPU(4096) . = ALIGN(4096); __init_end = .; /* freed after init ends here */ diff --git a/arch/s390/lib/uaccess_pt.c b/arch/s390/lib/uaccess_pt.c index 63181671e3e..60604b2819b 100644 --- a/arch/s390/lib/uaccess_pt.c +++ b/arch/s390/lib/uaccess_pt.c @@ -20,6 +20,7 @@ static int __handle_fault(struct mm_struct *mm, unsigned long address, { struct vm_area_struct *vma; int ret = -EFAULT; + int fault; if (in_atomic()) return ret; @@ -44,20 +45,18 @@ static int __handle_fault(struct mm_struct *mm, unsigned long address, } survive: - switch (handle_mm_fault(mm, vma, address, write_access)) { - case VM_FAULT_MINOR: - current->min_flt++; - break; - case VM_FAULT_MAJOR: - current->maj_flt++; - break; - case VM_FAULT_SIGBUS: - goto out_sigbus; - case VM_FAULT_OOM: - goto out_of_memory; - default: + fault = handle_mm_fault(mm, vma, address, write_access); + if (unlikely(fault & VM_FAULT_ERROR)) { + if (fault & VM_FAULT_OOM) + goto out_of_memory; + else if (fault & VM_FAULT_SIGBUS) + goto out_sigbus; BUG(); } + if (fault & VM_FAULT_MAJOR) + current->maj_flt++; + else + current->min_flt++; ret = 0; out: up_read(&mm->mmap_sem); diff --git a/arch/s390/mm/fault.c b/arch/s390/mm/fault.c index d855cdbf8fb..54055194e9a 100644 --- a/arch/s390/mm/fault.c +++ b/arch/s390/mm/fault.c @@ -307,6 +307,7 @@ do_exception(struct pt_regs *regs, unsigned long error_code, int write) unsigned long address; int space; int si_code; + int fault; if (notify_page_fault(regs, error_code)) return; @@ -377,23 +378,22 @@ survive: * make sure we exit gracefully rather than endlessly redo * the fault. */ - switch (handle_mm_fault(mm, vma, address, write)) { - case VM_FAULT_MINOR: - tsk->min_flt++; - break; - case VM_FAULT_MAJOR: - tsk->maj_flt++; - break; - case VM_FAULT_SIGBUS: - do_sigbus(regs, error_code, address); - return; - case VM_FAULT_OOM: - if (do_out_of_memory(regs, error_code, address)) - goto survive; - return; - default: + fault = handle_mm_fault(mm, vma, address, write); + if (unlikely(fault & VM_FAULT_ERROR)) { + if (fault & VM_FAULT_OOM) { + if (do_out_of_memory(regs, error_code, address)) + goto survive; + return; + } else if (fault & VM_FAULT_SIGBUS) { + do_sigbus(regs, error_code, address); + return; + } BUG(); } + if (fault & VM_FAULT_MAJOR) + tsk->maj_flt++; + else + tsk->min_flt++; up_read(&mm->mmap_sem); /* diff --git a/arch/sh/kernel/vmlinux.lds.S b/arch/sh/kernel/vmlinux.lds.S index 0696402f446..5ba216180b3 100644 --- a/arch/sh/kernel/vmlinux.lds.S +++ b/arch/sh/kernel/vmlinux.lds.S @@ -60,10 +60,7 @@ SECTIONS . = ALIGN(PAGE_SIZE); __nosave_end = .; - . = ALIGN(PAGE_SIZE); - __per_cpu_start = .; - .data.percpu : { *(.data.percpu) } - __per_cpu_end = .; + PERCPU(PAGE_SIZE) .data.cacheline_aligned : { *(.data.cacheline_aligned) } _edata = .; /* End of data section */ diff --git a/arch/sh/mm/fault.c b/arch/sh/mm/fault.c index 0b3eaf6fbb2..964c6767dc7 100644 --- a/arch/sh/mm/fault.c +++ b/arch/sh/mm/fault.c @@ -33,6 +33,7 @@ asmlinkage void __kprobes do_page_fault(struct pt_regs *regs, struct mm_struct *mm; struct vm_area_struct * vma; int si_code; + int fault; siginfo_t info; trace_hardirqs_on(); @@ -124,20 +125,18 @@ good_area: * the fault. */ survive: - switch (handle_mm_fault(mm, vma, address, writeaccess)) { - case VM_FAULT_MINOR: - tsk->min_flt++; - break; - case VM_FAULT_MAJOR: - tsk->maj_flt++; - break; - case VM_FAULT_SIGBUS: - goto do_sigbus; - case VM_FAULT_OOM: + fault = handle_mm_fault(mm, vma, address, writeaccess); + if (unlikely(fault & VM_FAULT_ERROR)) { + if (fault & VM_FAULT_OOM) goto out_of_memory; - default: - BUG(); + else if (fault & VM_FAULT_SIGBUS) + goto do_sigbus; + BUG(); } + if (fault & VM_FAULT_MAJOR) + tsk->maj_flt++; + else + tsk->min_flt++; up_read(&mm->mmap_sem); return; diff --git a/arch/sh64/kernel/vmlinux.lds.S b/arch/sh64/kernel/vmlinux.lds.S index 02aea86c590..8ac9c7c5f84 100644 --- a/arch/sh64/kernel/vmlinux.lds.S +++ b/arch/sh64/kernel/vmlinux.lds.S @@ -87,7 +87,10 @@ SECTIONS . = ALIGN(PAGE_SIZE); __per_cpu_start = .; - .data.percpu : C_PHYS(.data.percpu) { *(.data.percpu) } + .data.percpu : C_PHYS(.data.percpu) { + *(.data.percpu) + *(.data.percpu.shared_aligned) + } __per_cpu_end = . ; .data.cacheline_aligned : C_PHYS(.data.cacheline_aligned) { *(.data.cacheline_aligned) } diff --git a/arch/sh64/mm/fault.c b/arch/sh64/mm/fault.c index 3cd93ba5d82..0d069d82141 100644 --- a/arch/sh64/mm/fault.c +++ b/arch/sh64/mm/fault.c @@ -127,6 +127,7 @@ asmlinkage void do_page_fault(struct pt_regs *regs, unsigned long writeaccess, struct vm_area_struct * vma; const struct exception_table_entry *fixup; pte_t *pte; + int fault; #if defined(CONFIG_SH64_PROC_TLB) ++calls_to_do_slow_page_fault; @@ -221,18 +222,19 @@ good_area: * the fault. */ survive: - switch (handle_mm_fault(mm, vma, address, writeaccess)) { - case VM_FAULT_MINOR: - tsk->min_flt++; - break; - case VM_FAULT_MAJOR: - tsk->maj_flt++; - break; - case VM_FAULT_SIGBUS: - goto do_sigbus; - default: - goto out_of_memory; + fault = handle_mm_fault(mm, vma, address, writeaccess); + if (unlikely(fault & VM_FAULT_ERROR)) { + if (fault & VM_FAULT_OOM) + goto out_of_memory; + else if (fault & VM_FAULT_SIGBUS) + goto do_sigbus; + BUG(); } + if (fault & VM_FAULT_MAJOR) + tsk->maj_flt++; + else + tsk->min_flt++; + /* If we get here, the page fault has been handled. Do the TLB refill now from the newly-setup PTE, to avoid having to fault again right away on the same instruction. */ diff --git a/arch/sparc/kernel/vmlinux.lds.S b/arch/sparc/kernel/vmlinux.lds.S index f75a1b82278..47583887abc 100644 --- a/arch/sparc/kernel/vmlinux.lds.S +++ b/arch/sparc/kernel/vmlinux.lds.S @@ -65,10 +65,7 @@ SECTIONS __initramfs_end = .; #endif - . = ALIGN(4096); - __per_cpu_start = .; - .data.percpu : { *(.data.percpu) } - __per_cpu_end = .; + PERCPU(4096) . = ALIGN(4096); __init_end = .; . = ALIGN(32); diff --git a/arch/sparc/mm/fault.c b/arch/sparc/mm/fault.c index c3483365db4..50747fe4435 100644 --- a/arch/sparc/mm/fault.c +++ b/arch/sparc/mm/fault.c @@ -226,6 +226,7 @@ asmlinkage void do_sparc_fault(struct pt_regs *regs, int text_fault, int write, unsigned long g2; siginfo_t info; int from_user = !(regs->psr & PSR_PS); + int fault; if(text_fault) address = regs->pc; @@ -289,19 +290,18 @@ good_area: * make sure we exit gracefully rather than endlessly redo * the fault. */ - switch (handle_mm_fault(mm, vma, address, write)) { - case VM_FAULT_SIGBUS: - goto do_sigbus; - case VM_FAULT_OOM: - goto out_of_memory; - case VM_FAULT_MAJOR: + fault = handle_mm_fault(mm, vma, address, write); + if (unlikely(fault & VM_FAULT_ERROR)) { + if (fault & VM_FAULT_OOM) + goto out_of_memory; + else if (fault & VM_FAULT_SIGBUS) + goto do_sigbus; + BUG(); + } + if (fault & VM_FAULT_MAJOR) current->maj_flt++; - break; - case VM_FAULT_MINOR: - default: + else current->min_flt++; - break; - } up_read(&mm->mmap_sem); return; diff --git a/arch/sparc64/kernel/vmlinux.lds.S b/arch/sparc64/kernel/vmlinux.lds.S index 3ad10f3027e..481861764de 100644 --- a/arch/sparc64/kernel/vmlinux.lds.S +++ b/arch/sparc64/kernel/vmlinux.lds.S @@ -90,10 +90,8 @@ SECTIONS __initramfs_end = .; #endif - . = ALIGN(PAGE_SIZE); - __per_cpu_start = .; - .data.percpu : { *(.data.percpu) } - __per_cpu_end = .; + PERCPU(PAGE_SIZE) + . = ALIGN(PAGE_SIZE); __init_end = .; __bss_start = .; diff --git a/arch/sparc64/mm/fault.c b/arch/sparc64/mm/fault.c index b582024d219..17123e9ecf7 100644 --- a/arch/sparc64/mm/fault.c +++ b/arch/sparc64/mm/fault.c @@ -278,7 +278,7 @@ asmlinkage void __kprobes do_sparc64_fault(struct pt_regs *regs) struct mm_struct *mm = current->mm; struct vm_area_struct *vma; unsigned int insn = 0; - int si_code, fault_code; + int si_code, fault_code, fault; unsigned long address, mm_rss; fault_code = get_thread_fault_code(); @@ -415,20 +415,18 @@ good_area: goto bad_area; } - switch (handle_mm_fault(mm, vma, address, (fault_code & FAULT_CODE_WRITE))) { - case VM_FAULT_MINOR: - current->min_flt++; - break; - case VM_FAULT_MAJOR: - current->maj_flt++; - break; - case VM_FAULT_SIGBUS: - goto do_sigbus; - case VM_FAULT_OOM: - goto out_of_memory; - default: + fault = handle_mm_fault(mm, vma, address, (fault_code & FAULT_CODE_WRITE)); + if (unlikely(fault & VM_FAULT_ERROR)) { + if (fault & VM_FAULT_OOM) + goto out_of_memory; + else if (fault & VM_FAULT_SIGBUS) + goto do_sigbus; BUG(); } + if (fault & VM_FAULT_MAJOR) + current->maj_flt++; + else + current->min_flt++; up_read(&mm->mmap_sem); diff --git a/arch/sparc64/solaris/socksys.c b/arch/sparc64/solaris/socksys.c index e94f6e5d945..7736411f244 100644 --- a/arch/sparc64/solaris/socksys.c +++ b/arch/sparc64/solaris/socksys.c @@ -199,6 +199,5 @@ int __init init_socksys(void) void __exit cleanup_socksys(void) { - if (unregister_chrdev(30, "socksys")) - printk ("Couldn't unregister socksys character device\n"); + unregister_chrdev(30, "socksys"); } diff --git a/arch/um/defconfig b/arch/um/defconfig index a25cd25d55d..1e0f677c2f4 100644 --- a/arch/um/defconfig +++ b/arch/um/defconfig @@ -52,7 +52,6 @@ CONFIG_X86_WP_WORKS_OK=y CONFIG_X86_INVLPG=y CONFIG_X86_BSWAP=y CONFIG_X86_POPAD_OK=y -CONFIG_X86_CMPXCHG64=y CONFIG_X86_GOOD_APIC=y CONFIG_X86_USE_PPRO_CHECKSUM=y CONFIG_X86_TSC=y diff --git a/arch/um/kernel/trap.c b/arch/um/kernel/trap.c index abab90c3803..3850d53f79f 100644 --- a/arch/um/kernel/trap.c +++ b/arch/um/kernel/trap.c @@ -76,23 +76,24 @@ good_area: goto out; do { + int fault; survive: - switch (handle_mm_fault(mm, vma, address, is_write)){ - case VM_FAULT_MINOR: - current->min_flt++; - break; - case VM_FAULT_MAJOR: - current->maj_flt++; - break; - case VM_FAULT_SIGBUS: - err = -EACCES; - goto out; - case VM_FAULT_OOM: - err = -ENOMEM; - goto out_of_memory; - default: + fault = handle_mm_fault(mm, vma, address, is_write); + if (unlikely(fault & VM_FAULT_ERROR)) { + if (fault & VM_FAULT_OOM) { + err = -ENOMEM; + goto out_of_memory; + } else if (fault & VM_FAULT_SIGBUS) { + err = -EACCES; + goto out; + } BUG(); } + if (fault & VM_FAULT_MAJOR) + current->maj_flt++; + else + current->min_flt++; + pgd = pgd_offset(mm, address); pud = pud_offset(pgd, address); pmd = pmd_offset(pud, address); diff --git a/arch/x86_64/boot/.gitignore b/arch/x86_64/boot/.gitignore index 495f20c085d..18465143cfa 100644 --- a/arch/x86_64/boot/.gitignore +++ b/arch/x86_64/boot/.gitignore @@ -1,3 +1,5 @@ bootsect bzImage setup +setup.bin +setup.elf diff --git a/arch/x86_64/ia32/ia32_aout.c b/arch/x86_64/ia32/ia32_aout.c index fe83edb93c1..08781370256 100644 --- a/arch/x86_64/ia32/ia32_aout.c +++ b/arch/x86_64/ia32/ia32_aout.c @@ -404,7 +404,7 @@ beyond_if: set_brk(current->mm->start_brk, current->mm->brk); - retval = ia32_setup_arg_pages(bprm, IA32_STACK_TOP, EXSTACK_DEFAULT); + retval = setup_arg_pages(bprm, IA32_STACK_TOP, EXSTACK_DEFAULT); if (retval < 0) { /* Someone check-me: is this error path enough? */ send_sig(SIGKILL, current, 0); diff --git a/arch/x86_64/ia32/ia32_binfmt.c b/arch/x86_64/ia32/ia32_binfmt.c index 185399baaf6..ed56a8806ea 100644 --- a/arch/x86_64/ia32/ia32_binfmt.c +++ b/arch/x86_64/ia32/ia32_binfmt.c @@ -232,9 +232,6 @@ do { \ #define load_elf_binary load_elf32_binary #define ELF_PLAT_INIT(r, load_addr) elf32_init(r) -#define setup_arg_pages(bprm, stack_top, exec_stack) \ - ia32_setup_arg_pages(bprm, stack_top, exec_stack) -int ia32_setup_arg_pages(struct linux_binprm *bprm, unsigned long stack_top, int executable_stack); #undef start_thread #define start_thread(regs,new_rip,new_rsp) do { \ @@ -286,61 +283,6 @@ static void elf32_init(struct pt_regs *regs) me->thread.es = __USER_DS; } -int ia32_setup_arg_pages(struct linux_binprm *bprm, unsigned long stack_top, - int executable_stack) -{ - unsigned long stack_base; - struct vm_area_struct *mpnt; - struct mm_struct *mm = current->mm; - int i, ret; - - stack_base = stack_top - MAX_ARG_PAGES * PAGE_SIZE; - mm->arg_start = bprm->p + stack_base; - - bprm->p += stack_base; - if (bprm->loader) - bprm->loader += stack_base; - bprm->exec += stack_base; - - mpnt = kmem_cache_zalloc(vm_area_cachep, GFP_KERNEL); - if (!mpnt) - return -ENOMEM; - - down_write(&mm->mmap_sem); - { - mpnt->vm_mm = mm; - mpnt->vm_start = PAGE_MASK & (unsigned long) bprm->p; - mpnt->vm_end = stack_top; - if (executable_stack == EXSTACK_ENABLE_X) - mpnt->vm_flags = VM_STACK_FLAGS | VM_EXEC; - else if (executable_stack == EXSTACK_DISABLE_X) - mpnt->vm_flags = VM_STACK_FLAGS & ~VM_EXEC; - else - mpnt->vm_flags = VM_STACK_FLAGS; - mpnt->vm_page_prot = (mpnt->vm_flags & VM_EXEC) ? - PAGE_COPY_EXEC : PAGE_COPY; - if ((ret = insert_vm_struct(mm, mpnt))) { - up_write(&mm->mmap_sem); - kmem_cache_free(vm_area_cachep, mpnt); - return ret; - } - mm->stack_vm = mm->total_vm = vma_pages(mpnt); - } - - for (i = 0 ; i < MAX_ARG_PAGES ; i++) { - struct page *page = bprm->page[i]; - if (page) { - bprm->page[i] = NULL; - install_arg_page(mpnt, page, stack_base); - } - stack_base += PAGE_SIZE; - } - up_write(&mm->mmap_sem); - - return 0; -} -EXPORT_SYMBOL(ia32_setup_arg_pages); - #ifdef CONFIG_SYSCTL /* Register vsyscall32 into the ABI table */ #include <linux/sysctl.h> diff --git a/arch/x86_64/kernel/acpi/sleep.c b/arch/x86_64/kernel/acpi/sleep.c index 195b7034a14..4277f2b27e6 100644 --- a/arch/x86_64/kernel/acpi/sleep.c +++ b/arch/x86_64/kernel/acpi/sleep.c @@ -55,7 +55,7 @@ /* address in low memory of the wakeup routine. */ unsigned long acpi_wakeup_address = 0; -unsigned long acpi_video_flags; +unsigned long acpi_realmode_flags; extern char wakeup_start, wakeup_end; extern unsigned long acpi_copy_wakeup_routine(unsigned long); @@ -103,9 +103,11 @@ static int __init acpi_sleep_setup(char *str) { while ((str != NULL) && (*str != '\0')) { if (strncmp(str, "s3_bios", 7) == 0) - acpi_video_flags = 1; + acpi_realmode_flags |= 1; if (strncmp(str, "s3_mode", 7) == 0) - acpi_video_flags |= 2; + acpi_realmode_flags |= 2; + if (strncmp(str, "s3_beep", 7) == 0) + acpi_realmode_flags |= 4; str = strchr(str, ','); if (str != NULL) str += strspn(str, ", \t"); diff --git a/arch/x86_64/kernel/acpi/wakeup.S b/arch/x86_64/kernel/acpi/wakeup.S index 8550a6ffa27..13f1480cbec 100644 --- a/arch/x86_64/kernel/acpi/wakeup.S +++ b/arch/x86_64/kernel/acpi/wakeup.S @@ -16,6 +16,21 @@ # cs = 0x1234, eip = 0x05 # +#define BEEP \ + inb $97, %al; \ + outb %al, $0x80; \ + movb $3, %al; \ + outb %al, $97; \ + outb %al, $0x80; \ + movb $-74, %al; \ + outb %al, $67; \ + outb %al, $0x80; \ + movb $-119, %al; \ + outb %al, $66; \ + outb %al, $0x80; \ + movb $15, %al; \ + outb %al, $66; + ALIGN .align 16 @@ -33,6 +48,13 @@ wakeup_code: movw %cs, %ax movw %ax, %ds # Make ds:0 point to wakeup_start movw %ax, %ss + + # Data segment must be set up before we can see whether to beep. + testl $4, realmode_flags - wakeup_code + jz 1f + BEEP +1: + # Private stack is needed for ASUS board mov $(wakeup_stack - wakeup_code), %sp @@ -48,7 +70,7 @@ wakeup_code: testl %eax, %eax jnz no_longmode - testl $1, video_flags - wakeup_code + testl $1, realmode_flags - wakeup_code jz 1f lcall $0xc000,$3 movw %cs, %ax @@ -56,7 +78,7 @@ wakeup_code: movw %ax, %ss 1: - testl $2, video_flags - wakeup_code + testl $2, realmode_flags - wakeup_code jz 1f mov video_mode - wakeup_code, %ax call mode_seta @@ -230,7 +252,7 @@ gdt_48a: real_magic: .quad 0 video_mode: .quad 0 -video_flags: .quad 0 +realmode_flags: .quad 0 .code16 bogus_real_magic: @@ -346,8 +368,8 @@ ENTRY(acpi_copy_wakeup_routine) movl saved_video_mode, %edx movl %edx, video_mode - wakeup_start (,%rdi) - movl acpi_video_flags, %edx - movl %edx, video_flags - wakeup_start (,%rdi) + movl acpi_realmode_flags, %edx + movl %edx, realmode_flags - wakeup_start (,%rdi) movq $0x12345678, real_magic - wakeup_start (,%rdi) movq $0x123456789abcdef0, %rdx movq %rdx, saved_magic diff --git a/arch/x86_64/kernel/init_task.c b/arch/x86_64/kernel/init_task.c index 3dc5854ba21..4ff33d4f855 100644 --- a/arch/x86_64/kernel/init_task.c +++ b/arch/x86_64/kernel/init_task.c @@ -44,7 +44,7 @@ EXPORT_SYMBOL(init_task); * section. Since TSS's are completely CPU-local, we want them * on exact cacheline boundaries, to eliminate cacheline ping-pong. */ -DEFINE_PER_CPU(struct tss_struct, init_tss) ____cacheline_internodealigned_in_smp = INIT_TSS; +DEFINE_PER_CPU_SHARED_ALIGNED(struct tss_struct, init_tss) = INIT_TSS; /* Copies of the original ist values from the tss are only accessed during * debugging, no special alignment required. diff --git a/arch/x86_64/kernel/traps.c b/arch/x86_64/kernel/traps.c index 74cbeb2e99a..8713ad4a4db 100644 --- a/arch/x86_64/kernel/traps.c +++ b/arch/x86_64/kernel/traps.c @@ -34,6 +34,10 @@ #include <linux/bug.h> #include <linux/kdebug.h> +#if defined(CONFIG_EDAC) +#include <linux/edac.h> +#endif + #include <asm/system.h> #include <asm/io.h> #include <asm/atomic.h> @@ -719,6 +723,13 @@ mem_parity_error(unsigned char reason, struct pt_regs * regs) reason); printk(KERN_EMERG "You have some hardware problem, likely on the PCI bus.\n"); +#if defined(CONFIG_EDAC) + if(edac_handler_set()) { + edac_atomic_assert_error(); + return; + } +#endif + if (panic_on_unrecovered_nmi) panic("NMI: Not continuing"); diff --git a/arch/x86_64/kernel/tsc.c b/arch/x86_64/kernel/tsc.c index 48f9a8e6aa9..e850aa01e1b 100644 --- a/arch/x86_64/kernel/tsc.c +++ b/arch/x86_64/kernel/tsc.c @@ -44,7 +44,7 @@ unsigned long long sched_clock(void) static int tsc_unstable; -static inline int check_tsc_unstable(void) +inline int check_tsc_unstable(void) { return tsc_unstable; } diff --git a/arch/x86_64/kernel/vmlinux.lds.S b/arch/x86_64/kernel/vmlinux.lds.S index dbccfda8364..5c57ea4591c 100644 --- a/arch/x86_64/kernel/vmlinux.lds.S +++ b/arch/x86_64/kernel/vmlinux.lds.S @@ -48,7 +48,9 @@ SECTIONS __ex_table : AT(ADDR(__ex_table) - LOAD_OFFSET) { *(__ex_table) } __stop___ex_table = .; - BUG_TABLE + NOTES :text :note + + BUG_TABLE :text RODATA @@ -194,10 +196,8 @@ SECTIONS __initramfs_end = .; #endif - . = ALIGN(4096); - __per_cpu_start = .; - .data.percpu : AT(ADDR(.data.percpu) - LOAD_OFFSET) { *(.data.percpu) } - __per_cpu_end = .; + PERCPU(4096) + . = ALIGN(4096); __init_end = .; diff --git a/arch/x86_64/mm/fault.c b/arch/x86_64/mm/fault.c index 635e58d443d..84f11728fc7 100644 --- a/arch/x86_64/mm/fault.c +++ b/arch/x86_64/mm/fault.c @@ -317,7 +317,7 @@ asmlinkage void __kprobes do_page_fault(struct pt_regs *regs, struct vm_area_struct * vma; unsigned long address; const struct exception_table_entry *fixup; - int write; + int write, fault; unsigned long flags; siginfo_t info; @@ -450,19 +450,18 @@ good_area: * make sure we exit gracefully rather than endlessly redo * the fault. */ - switch (handle_mm_fault(mm, vma, address, write)) { - case VM_FAULT_MINOR: - tsk->min_flt++; - break; - case VM_FAULT_MAJOR: - tsk->maj_flt++; - break; - case VM_FAULT_SIGBUS: - goto do_sigbus; - default: - goto out_of_memory; + fault = handle_mm_fault(mm, vma, address, write); + if (unlikely(fault & VM_FAULT_ERROR)) { + if (fault & VM_FAULT_OOM) + goto out_of_memory; + else if (fault & VM_FAULT_SIGBUS) + goto do_sigbus; + BUG(); } - + if (fault & VM_FAULT_MAJOR) + tsk->maj_flt++; + else + tsk->min_flt++; up_read(&mm->mmap_sem); return; diff --git a/arch/xtensa/kernel/vmlinux.lds.S b/arch/xtensa/kernel/vmlinux.lds.S index b0582c3c5f8..ac4ed52034d 100644 --- a/arch/xtensa/kernel/vmlinux.lds.S +++ b/arch/xtensa/kernel/vmlinux.lds.S @@ -118,7 +118,8 @@ SECTIONS _fdata = .; .data : { - *(.data) CONSTRUCTORS + DATA_DATA + CONSTRUCTORS . = ALIGN(XCHAL_ICACHE_LINESIZE); *(.data.cacheline_aligned) } @@ -190,10 +191,7 @@ SECTIONS __initramfs_end = .; #endif - . = ALIGN(4096); - __per_cpu_start = .; - .data.percpu : { *(.data.percpu) } - __per_cpu_end = .; + PERCPU(4096) /* We need this dummy segment here */ diff --git a/arch/xtensa/mm/fault.c b/arch/xtensa/mm/fault.c index 3dc6f2f07bb..16004067add 100644 --- a/arch/xtensa/mm/fault.c +++ b/arch/xtensa/mm/fault.c @@ -41,6 +41,7 @@ void do_page_fault(struct pt_regs *regs) siginfo_t info; int is_write, is_exec; + int fault; info.si_code = SEGV_MAPERR; @@ -102,20 +103,18 @@ good_area: * the fault. */ survive: - switch (handle_mm_fault(mm, vma, address, is_write)) { - case VM_FAULT_MINOR: - current->min_flt++; - break; - case VM_FAULT_MAJOR: - current->maj_flt++; - break; - case VM_FAULT_SIGBUS: - goto do_sigbus; - case VM_FAULT_OOM: - goto out_of_memory; - default: + fault = handle_mm_fault(mm, vma, address, is_write); + if (unlikely(fault & VM_FAULT_ERROR)) { + if (fault & VM_FAULT_OOM) + goto out_of_memory; + else if (fault & VM_FAULT_SIGBUS) + goto do_sigbus; BUG(); } + if (fault & VM_FAULT_MAJOR) + current->maj_flt++; + else + current->min_flt++; up_read(&mm->mmap_sem); return; diff --git a/block/scsi_ioctl.c b/block/scsi_ioctl.c index a26ba07955f..71bdf88884b 100644 --- a/block/scsi_ioctl.c +++ b/block/scsi_ioctl.c @@ -436,11 +436,10 @@ int sg_scsi_ioctl(struct file *file, struct request_queue *q, bytes = max(in_len, out_len); if (bytes) { - buffer = kmalloc(bytes, q->bounce_gfp | GFP_USER| __GFP_NOWARN); + buffer = kzalloc(bytes, q->bounce_gfp | GFP_USER| __GFP_NOWARN); if (!buffer) return -ENOMEM; - memset(buffer, 0, bytes); } rq = blk_get_request(q, in_len ? WRITE : READ, __GFP_WAIT); diff --git a/drivers/Kconfig b/drivers/Kconfig index ae01d86070b..707650ab77a 100644 --- a/drivers/Kconfig +++ b/drivers/Kconfig @@ -85,4 +85,6 @@ source "drivers/auxdisplay/Kconfig" source "drivers/kvm/Kconfig" source "drivers/uio/Kconfig" + +source "drivers/lguest/Kconfig" endmenu diff --git a/drivers/Makefile b/drivers/Makefile index c34c8efff60..0ea8e3237c0 100644 --- a/drivers/Makefile +++ b/drivers/Makefile @@ -73,6 +73,7 @@ obj-$(CONFIG_ISDN) += isdn/ obj-$(CONFIG_EDAC) += edac/ obj-$(CONFIG_MCA) += mca/ obj-$(CONFIG_EISA) += eisa/ +obj-$(CONFIG_LGUEST_GUEST) += lguest/ obj-$(CONFIG_CPU_FREQ) += cpufreq/ obj-$(CONFIG_MMC) += mmc/ obj-$(CONFIG_NEW_LEDS) += leds/ diff --git a/drivers/acpi/sleep/main.c b/drivers/acpi/sleep/main.c index bc7e16ec839..42127c0d612 100644 --- a/drivers/acpi/sleep/main.c +++ b/drivers/acpi/sleep/main.c @@ -217,10 +217,26 @@ static void acpi_hibernation_finish(void) } } +static int acpi_hibernation_pre_restore(void) +{ + acpi_status status; + + status = acpi_hw_disable_all_gpes(); + + return ACPI_SUCCESS(status) ? 0 : -EFAULT; +} + +static void acpi_hibernation_restore_cleanup(void) +{ + acpi_hw_enable_all_runtime_gpes(); +} + static struct hibernation_ops acpi_hibernation_ops = { .prepare = acpi_hibernation_prepare, .enter = acpi_hibernation_enter, .finish = acpi_hibernation_finish, + .pre_restore = acpi_hibernation_pre_restore, + .restore_cleanup = acpi_hibernation_restore_cleanup, }; #endif /* CONFIG_SOFTWARE_SUSPEND */ diff --git a/drivers/acpi/sleep/poweroff.c b/drivers/acpi/sleep/poweroff.c index d9801eff648..39e40d56b03 100644 --- a/drivers/acpi/sleep/poweroff.c +++ b/drivers/acpi/sleep/poweroff.c @@ -39,7 +39,13 @@ int acpi_sleep_prepare(u32 acpi_state) #ifdef CONFIG_PM -void acpi_power_off(void) +static void acpi_power_off_prepare(void) +{ + /* Prepare to power off the system */ + acpi_sleep_prepare(ACPI_STATE_S5); +} + +static void acpi_power_off(void) { /* acpi_sleep_prepare(ACPI_STATE_S5) should have already been called */ printk("%s called\n", __FUNCTION__); @@ -48,30 +54,6 @@ void acpi_power_off(void) acpi_enter_sleep_state(ACPI_STATE_S5); } -static int acpi_shutdown(struct sys_device *x) -{ - switch (system_state) { - case SYSTEM_POWER_OFF: - /* Prepare to power off the system */ - return acpi_sleep_prepare(ACPI_STATE_S5); - case SYSTEM_SUSPEND_DISK: - /* Prepare to suspend the system to disk */ - return acpi_sleep_prepare(ACPI_STATE_S4); - default: - return 0; - } -} - -static struct sysdev_class acpi_sysclass = { - set_kset_name("acpi"), - .shutdown = acpi_shutdown -}; - -static struct sys_device device_acpi = { - .id = 0, - .cls = &acpi_sysclass, -}; - static int acpi_poweroff_init(void) { if (!acpi_disabled) { @@ -81,13 +63,8 @@ static int acpi_poweroff_init(void) status = acpi_get_sleep_type_data(ACPI_STATE_S5, &type_a, &type_b); if (ACPI_SUCCESS(status)) { - int error; - error = sysdev_class_register(&acpi_sysclass); - if (!error) - error = sysdev_register(&device_acpi); - if (!error) - pm_power_off = acpi_power_off; - return error; + pm_power_off_prepare = acpi_power_off_prepare; + pm_power_off = acpi_power_off; } } return 0; diff --git a/drivers/base/power/Makefile b/drivers/base/power/Makefile index fff17800720..966a5e28741 100644 --- a/drivers/base/power/Makefile +++ b/drivers/base/power/Makefile @@ -5,6 +5,6 @@ obj-$(CONFIG_PM_TRACE) += trace.o ifeq ($(CONFIG_DEBUG_DRIVER),y) EXTRA_CFLAGS += -DDEBUG endif -ifeq ($(CONFIG_PM_DEBUG),y) +ifeq ($(CONFIG_PM_VERBOSE),y) EXTRA_CFLAGS += -DDEBUG endif diff --git a/drivers/block/Makefile b/drivers/block/Makefile index 3e31532df0e..819c829125f 100644 --- a/drivers/block/Makefile +++ b/drivers/block/Makefile @@ -30,3 +30,4 @@ obj-$(CONFIG_BLK_DEV_SX8) += sx8.o obj-$(CONFIG_BLK_DEV_UB) += ub.o obj-$(CONFIG_XEN_BLKDEV_FRONTEND) += xen-blkfront.o +obj-$(CONFIG_LGUEST_GUEST) += lguest_blk.o diff --git a/drivers/block/lguest_blk.c b/drivers/block/lguest_blk.c new file mode 100644 index 00000000000..1634c2dd25e --- /dev/null +++ b/drivers/block/lguest_blk.c @@ -0,0 +1,275 @@ +/* A simple block driver for lguest. + * + * Copyright 2006 Rusty Russell <rusty@rustcorp.com.au> IBM Corporation + * + * 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. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA + */ +//#define DEBUG +#include <linux/init.h> +#include <linux/types.h> +#include <linux/blkdev.h> +#include <linux/interrupt.h> +#include <linux/lguest_bus.h> + +static char next_block_index = 'a'; + +struct blockdev +{ + spinlock_t lock; + + /* The disk structure for the kernel. */ + struct gendisk *disk; + + /* The major number for this disk. */ + int major; + int irq; + + unsigned long phys_addr; + /* The mapped block page. */ + struct lguest_block_page *lb_page; + + /* We only have a single request outstanding at a time. */ + struct lguest_dma dma; + struct request *req; +}; + +/* Jens gave me this nice helper to end all chunks of a request. */ +static void end_entire_request(struct request *req, int uptodate) +{ + if (end_that_request_first(req, uptodate, req->hard_nr_sectors)) + BUG(); + add_disk_randomness(req->rq_disk); + blkdev_dequeue_request(req); + end_that_request_last(req, uptodate); +} + +static irqreturn_t lgb_irq(int irq, void *_bd) +{ + struct blockdev *bd = _bd; + unsigned long flags; + + if (!bd->req) { + pr_debug("No work!\n"); + return IRQ_NONE; + } + + if (!bd->lb_page->result) { + pr_debug("No result!\n"); + return IRQ_NONE; + } + + spin_lock_irqsave(&bd->lock, flags); + end_entire_request(bd->req, bd->lb_page->result == 1); + bd->req = NULL; + bd->dma.used_len = 0; + blk_start_queue(bd->disk->queue); + spin_unlock_irqrestore(&bd->lock, flags); + return IRQ_HANDLED; +} + +static unsigned int req_to_dma(struct request *req, struct lguest_dma *dma) +{ + unsigned int i = 0, idx, len = 0; + struct bio *bio; + + rq_for_each_bio(bio, req) { + struct bio_vec *bvec; + bio_for_each_segment(bvec, bio, idx) { + BUG_ON(i == LGUEST_MAX_DMA_SECTIONS); + BUG_ON(!bvec->bv_len); + dma->addr[i] = page_to_phys(bvec->bv_page) + + bvec->bv_offset; + dma->len[i] = bvec->bv_len; + len += bvec->bv_len; + i++; + } + } + if (i < LGUEST_MAX_DMA_SECTIONS) + dma->len[i] = 0; + return len; +} + +static void empty_dma(struct lguest_dma *dma) +{ + dma->len[0] = 0; +} + +static void setup_req(struct blockdev *bd, + int type, struct request *req, struct lguest_dma *dma) +{ + bd->lb_page->type = type; + bd->lb_page->sector = req->sector; + bd->lb_page->result = 0; + bd->req = req; + bd->lb_page->bytes = req_to_dma(req, dma); +} + +static void do_write(struct blockdev *bd, struct request *req) +{ + struct lguest_dma send; + + pr_debug("lgb: WRITE sector %li\n", (long)req->sector); + setup_req(bd, 1, req, &send); + + lguest_send_dma(bd->phys_addr, &send); +} + +static void do_read(struct blockdev *bd, struct request *req) +{ + struct lguest_dma ping; + + pr_debug("lgb: READ sector %li\n", (long)req->sector); + setup_req(bd, 0, req, &bd->dma); + + empty_dma(&ping); + lguest_send_dma(bd->phys_addr, &ping); +} + +static void do_lgb_request(request_queue_t *q) +{ + struct blockdev *bd; + struct request *req; + +again: + req = elv_next_request(q); + if (!req) + return; + + bd = req->rq_disk->private_data; + /* Sometimes we get repeated requests after blk_stop_queue. */ + if (bd->req) + return; + + if (!blk_fs_request(req)) { + pr_debug("Got non-command 0x%08x\n", req->cmd_type); + req->errors++; + end_entire_request(req, 0); + goto again; + } + + if (rq_data_dir(req) == WRITE) + do_write(bd, req); + else + do_read(bd, req); + + /* Wait for interrupt to tell us it's done. */ + blk_stop_queue(q); +} + +static struct block_device_operations lguestblk_fops = { + .owner = THIS_MODULE, +}; + +static int lguestblk_probe(struct lguest_device *lgdev) +{ + struct blockdev *bd; + int err; + int irqflags = IRQF_SHARED; + + bd = kmalloc(sizeof(*bd), GFP_KERNEL); + if (!bd) + return -ENOMEM; + + spin_lock_init(&bd->lock); + bd->irq = lgdev_irq(lgdev); + bd->req = NULL; + bd->dma.used_len = 0; + bd->dma.len[0] = 0; + bd->phys_addr = (lguest_devices[lgdev->index].pfn << PAGE_SHIFT); + + bd->lb_page = lguest_map(bd->phys_addr, 1); + if (!bd->lb_page) { + err = -ENOMEM; + goto out_free_bd; + } + + bd->major = register_blkdev(0, "lguestblk"); + if (bd->major < 0) { + err = bd->major; + goto out_unmap; + } + + bd->disk = alloc_disk(1); + if (!bd->disk) { + err = -ENOMEM; + goto out_unregister_blkdev; + } + + bd->disk->queue = blk_init_queue(do_lgb_request, &bd->lock); + if (!bd->disk->queue) { + err = -ENOMEM; + goto out_put_disk; + } + + /* We can only handle a certain number of sg entries */ + blk_queue_max_hw_segments(bd->disk->queue, LGUEST_MAX_DMA_SECTIONS); + /* Buffers must not cross page boundaries */ + blk_queue_segment_boundary(bd->disk->queue, PAGE_SIZE-1); + + sprintf(bd->disk->disk_name, "lgb%c", next_block_index++); + if (lguest_devices[lgdev->index].features & LGUEST_DEVICE_F_RANDOMNESS) + irqflags |= IRQF_SAMPLE_RANDOM; + err = request_irq(bd->irq, lgb_irq, irqflags, bd->disk->disk_name, bd); + if (err) + goto out_cleanup_queue; + + err = lguest_bind_dma(bd->phys_addr, &bd->dma, 1, bd->irq); + if (err) + goto out_free_irq; + + bd->disk->major = bd->major; + bd->disk->first_minor = 0; + bd->disk->private_data = bd; + bd->disk->fops = &lguestblk_fops; + /* This is initialized to the disk size by the other end. */ + set_capacity(bd->disk, bd->lb_page->num_sectors); + add_disk(bd->disk); + + printk(KERN_INFO "%s: device %i at major %d\n", + bd->disk->disk_name, lgdev->index, bd->major); + + lgdev->private = bd; + return 0; + +out_free_irq: + free_irq(bd->irq, bd); +out_cleanup_queue: + blk_cleanup_queue(bd->disk->queue); +out_put_disk: + put_disk(bd->disk); +out_unregister_blkdev: + unregister_blkdev(bd->major, "lguestblk"); +out_unmap: + lguest_unmap(bd->lb_page); +out_free_bd: + kfree(bd); + return err; +} + +static struct lguest_driver lguestblk_drv = { + .name = "lguestblk", + .owner = THIS_MODULE, + .device_type = LGUEST_DEVICE_T_BLOCK, + .probe = lguestblk_probe, +}; + +static __init int lguestblk_init(void) +{ + return register_lguest_driver(&lguestblk_drv); +} +module_init(lguestblk_init); + +MODULE_DESCRIPTION("Lguest block driver"); +MODULE_LICENSE("GPL"); diff --git a/drivers/block/sx8.c b/drivers/block/sx8.c index 54509eb3391..949ae93499e 100644 --- a/drivers/block/sx8.c +++ b/drivers/block/sx8.c @@ -1608,7 +1608,7 @@ static int carm_init_one (struct pci_dev *pdev, const struct pci_device_id *ent) } #endif - host = kmalloc(sizeof(*host), GFP_KERNEL); + host = kzalloc(sizeof(*host), GFP_KERNEL); if (!host) { printk(KERN_ERR DRV_NAME "(%s): memory alloc failure\n", pci_name(pdev)); @@ -1616,7 +1616,6 @@ static int carm_init_one (struct pci_dev *pdev, const struct pci_device_id *ent) goto err_out_regions; } - memset(host, 0, sizeof(*host)); host->pdev = pdev; host->flags = pci_dac ? FL_DAC : 0; spin_lock_init(&host->lock); diff --git a/drivers/char/Makefile b/drivers/char/Makefile index 8852b8d643c..4e6f387fd18 100644 --- a/drivers/char/Makefile +++ b/drivers/char/Makefile @@ -42,6 +42,7 @@ obj-$(CONFIG_SYNCLINK_GT) += synclink_gt.o obj-$(CONFIG_N_HDLC) += n_hdlc.o obj-$(CONFIG_AMIGA_BUILTIN_SERIAL) += amiserial.o obj-$(CONFIG_SX) += sx.o generic_serial.o +obj-$(CONFIG_LGUEST_GUEST) += hvc_lguest.o obj-$(CONFIG_RIO) += rio/ generic_serial.o obj-$(CONFIG_HVC_CONSOLE) += hvc_vio.o hvsi.o obj-$(CONFIG_HVC_ISERIES) += hvc_iseries.o diff --git a/drivers/char/amiserial.c b/drivers/char/amiserial.c index 7b02bf1289a..3d468f502d2 100644 --- a/drivers/char/amiserial.c +++ b/drivers/char/amiserial.c @@ -1721,12 +1721,11 @@ static int get_async_struct(int line, struct async_struct **ret_info) *ret_info = sstate->info; return 0; } - info = kmalloc(sizeof(struct async_struct), GFP_KERNEL); + info = kzalloc(sizeof(struct async_struct), GFP_KERNEL); if (!info) { sstate->count--; return -ENOMEM; } - memset(info, 0, sizeof(struct async_struct)); #ifdef DECLARE_WAITQUEUE init_waitqueue_head(&info->open_wait); init_waitqueue_head(&info->close_wait); diff --git a/drivers/char/drm/via_dmablit.c b/drivers/char/drm/via_dmablit.c index fdb8609dd76..832de1d9ba7 100644 --- a/drivers/char/drm/via_dmablit.c +++ b/drivers/char/drm/via_dmablit.c @@ -273,10 +273,9 @@ via_alloc_desc_pages(drm_via_sg_info_t *vsg) vsg->num_desc_pages = (vsg->num_desc + vsg->descriptors_per_page - 1) / vsg->descriptors_per_page; - if (NULL == (vsg->desc_pages = kmalloc(sizeof(void *) * vsg->num_desc_pages, GFP_KERNEL))) + if (NULL == (vsg->desc_pages = kcalloc(vsg->num_desc_pages, sizeof(void *), GFP_KERNEL))) return DRM_ERR(ENOMEM); - memset(vsg->desc_pages, 0, sizeof(void *) * vsg->num_desc_pages); vsg->state = dr_via_desc_pages_alloc; for (i=0; i<vsg->num_desc_pages; ++i) { if (NULL == (vsg->desc_pages[i] = diff --git a/drivers/char/esp.c b/drivers/char/esp.c index 74cd5118af5..2e7ae42a550 100644 --- a/drivers/char/esp.c +++ b/drivers/char/esp.c @@ -2459,7 +2459,7 @@ static int __init espserial_init(void) return 1; } - info = kmalloc(sizeof(struct esp_struct), GFP_KERNEL); + info = kzalloc(sizeof(struct esp_struct), GFP_KERNEL); if (!info) { @@ -2469,7 +2469,6 @@ static int __init espserial_init(void) return 1; } - memset((void *)info, 0, sizeof(struct esp_struct)); spin_lock_init(&info->lock); /* rx_trigger, tx_trigger are needed by autoconfig */ info->config.rx_trigger = rx_trigger; @@ -2527,7 +2526,7 @@ static int __init espserial_init(void) if (!dma) info->stat_flags |= ESP_STAT_NEVER_DMA; - info = kmalloc(sizeof(struct esp_struct), GFP_KERNEL); + info = kzalloc(sizeof(struct esp_struct), GFP_KERNEL); if (!info) { printk(KERN_ERR "Couldn't allocate memory for esp serial device information\n"); @@ -2536,7 +2535,6 @@ static int __init espserial_init(void) return 0; } - memset((void *)info, 0, sizeof(struct esp_struct)); /* rx_trigger, tx_trigger are needed by autoconfig */ info->config.rx_trigger = rx_trigger; info->config.tx_trigger = tx_trigger; diff --git a/drivers/char/hvc_lguest.c b/drivers/char/hvc_lguest.c new file mode 100644 index 00000000000..e7b889e404a --- /dev/null +++ b/drivers/char/hvc_lguest.c @@ -0,0 +1,102 @@ +/* Simple console for lguest. + * + * Copyright (C) 2006 Rusty Russell, IBM Corporation + * + * 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. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA + */ +#include <linux/err.h> +#include <linux/init.h> +#include <linux/lguest_bus.h> +#include "hvc_console.h" + +static char inbuf[256]; +static struct lguest_dma cons_input = { .used_len = 0, + .addr[0] = __pa(inbuf), + .len[0] = sizeof(inbuf), + .len[1] = 0 }; + +static int put_chars(u32 vtermno, const char *buf, int count) +{ + struct lguest_dma dma; + + /* FIXME: what if it's over a page boundary? */ + dma.len[0] = count; + dma.len[1] = 0; + dma.addr[0] = __pa(buf); + + lguest_send_dma(LGUEST_CONSOLE_DMA_KEY, &dma); + return count; +} + +static int get_chars(u32 vtermno, char *buf, int count) +{ + static int cons_offset; + + if (!cons_input.used_len) + return 0; + + if (cons_input.used_len - cons_offset < count) + count = cons_input.used_len - cons_offset; + + memcpy(buf, inbuf + cons_offset, count); + cons_offset += count; + if (cons_offset == cons_input.used_len) { + cons_offset = 0; + cons_input.used_len = 0; + } + return count; +} + +static struct hv_ops lguest_cons = { + .get_chars = get_chars, + .put_chars = put_chars, +}; + +static int __init cons_init(void) +{ + if (strcmp(paravirt_ops.name, "lguest") != 0) + return 0; + + return hvc_instantiate(0, 0, &lguest_cons); +} +console_initcall(cons_init); + +static int lguestcons_probe(struct lguest_device *lgdev) +{ + int err; + + lgdev->private = hvc_alloc(0, lgdev_irq(lgdev), &lguest_cons, 256); + if (IS_ERR(lgdev->private)) + return PTR_ERR(lgdev->private); + + err = lguest_bind_dma(LGUEST_CONSOLE_DMA_KEY, &cons_input, 1, + lgdev_irq(lgdev)); + if (err) + printk("lguest console: failed to bind buffer.\n"); + return err; +} + +static struct lguest_driver lguestcons_drv = { + .name = "lguestcons", + .owner = THIS_MODULE, + .device_type = LGUEST_DEVICE_T_CONSOLE, + .probe = lguestcons_probe, +}; + +static int __init hvc_lguest_init(void) +{ + return register_lguest_driver(&lguestcons_drv); +} +module_init(hvc_lguest_init); diff --git a/drivers/char/hvcs.c b/drivers/char/hvcs.c index 207f7343ba6..17f96e04266 100644 --- a/drivers/char/hvcs.c +++ b/drivers/char/hvcs.c @@ -784,12 +784,10 @@ static int __devinit hvcs_probe( return -EFAULT; } - hvcsd = kmalloc(sizeof(*hvcsd), GFP_KERNEL); + hvcsd = kzalloc(sizeof(*hvcsd), GFP_KERNEL); if (!hvcsd) return -ENODEV; - /* hvcsd->tty is zeroed out with the memset */ - memset(hvcsd, 0x00, sizeof(*hvcsd)); spin_lock_init(&hvcsd->lock); /* Automatically incs the refcount the first time */ diff --git a/drivers/char/ip2/ip2main.c b/drivers/char/ip2/ip2main.c index 83c7258d358..6005b522577 100644 --- a/drivers/char/ip2/ip2main.c +++ b/drivers/char/ip2/ip2main.c @@ -425,9 +425,7 @@ cleanup_module(void) printk(KERN_ERR "IP2: failed to unregister tty driver (%d)\n", err); } put_tty_driver(ip2_tty_driver); - if ( ( err = unregister_chrdev ( IP2_IPL_MAJOR, pcIpl ) ) ) { - printk(KERN_ERR "IP2: failed to unregister IPL driver (%d)\n", err); - } + unregister_chrdev(IP2_IPL_MAJOR, pcIpl); remove_proc_entry("ip2mem", &proc_root); // free memory diff --git a/drivers/char/ipmi/ipmi_msghandler.c b/drivers/char/ipmi/ipmi_msghandler.c index b5df7e61aeb..6a01dd9e43f 100644 --- a/drivers/char/ipmi/ipmi_msghandler.c +++ b/drivers/char/ipmi/ipmi_msghandler.c @@ -2639,10 +2639,9 @@ int ipmi_register_smi(struct ipmi_smi_handlers *handlers, return -ENODEV; } - intf = kmalloc(sizeof(*intf), GFP_KERNEL); + intf = kzalloc(sizeof(*intf), GFP_KERNEL); if (!intf) return -ENOMEM; - memset(intf, 0, sizeof(*intf)); intf->ipmi_version_major = ipmi_version_major(device_id); intf->ipmi_version_minor = ipmi_version_minor(device_id); diff --git a/drivers/char/mbcs.c b/drivers/char/mbcs.c index 57f9115a456..7ee5d944492 100644 --- a/drivers/char/mbcs.c +++ b/drivers/char/mbcs.c @@ -39,14 +39,14 @@ #else #define DBG(fmt...) #endif -int mbcs_major; +static int mbcs_major; -LIST_HEAD(soft_list); +static LIST_HEAD(soft_list); /* * file operations */ -const struct file_operations mbcs_ops = { +static const struct file_operations mbcs_ops = { .open = mbcs_open, .llseek = mbcs_sram_llseek, .read = mbcs_sram_read, @@ -377,7 +377,7 @@ dmaread_exit: return rv; } -int mbcs_open(struct inode *ip, struct file *fp) +static int mbcs_open(struct inode *ip, struct file *fp) { struct mbcs_soft *soft; int minor; @@ -394,7 +394,7 @@ int mbcs_open(struct inode *ip, struct file *fp) return -ENODEV; } -ssize_t mbcs_sram_read(struct file * fp, char __user *buf, size_t len, loff_t * off) +static ssize_t mbcs_sram_read(struct file * fp, char __user *buf, size_t len, loff_t * off) { struct cx_dev *cx_dev = fp->private_data; struct mbcs_soft *soft = cx_dev->soft; @@ -418,7 +418,7 @@ ssize_t mbcs_sram_read(struct file * fp, char __user *buf, size_t len, loff_t * return rv; } -ssize_t +static ssize_t mbcs_sram_write(struct file * fp, const char __user *buf, size_t len, loff_t * off) { struct cx_dev *cx_dev = fp->private_data; @@ -443,7 +443,7 @@ mbcs_sram_write(struct file * fp, const char __user *buf, size_t len, loff_t * o return rv; } -loff_t mbcs_sram_llseek(struct file * filp, loff_t off, int whence) +static loff_t mbcs_sram_llseek(struct file * filp, loff_t off, int whence) { loff_t newpos; @@ -491,7 +491,7 @@ static void mbcs_gscr_pioaddr_set(struct mbcs_soft *soft) soft->gscr_addr = mbcs_pioaddr(soft, MBCS_GSCR_START); } -int mbcs_gscr_mmap(struct file *fp, struct vm_area_struct *vma) +static int mbcs_gscr_mmap(struct file *fp, struct vm_area_struct *vma) { struct cx_dev *cx_dev = fp->private_data; struct mbcs_soft *soft = cx_dev->soft; @@ -793,7 +793,7 @@ static int mbcs_remove(struct cx_dev *dev) return 0; } -const struct cx_device_id __devinitdata mbcs_id_table[] = { +static const struct cx_device_id __devinitdata mbcs_id_table[] = { { .part_num = MBCS_PART_NUM, .mfg_num = MBCS_MFG_NUM, @@ -807,7 +807,7 @@ const struct cx_device_id __devinitdata mbcs_id_table[] = { MODULE_DEVICE_TABLE(cx, mbcs_id_table); -struct cx_drv mbcs_driver = { +static struct cx_drv mbcs_driver = { .name = DEVICE_NAME, .id_table = mbcs_id_table, .probe = mbcs_probe, @@ -816,12 +816,7 @@ struct cx_drv mbcs_driver = { static void __exit mbcs_exit(void) { - int rv; - - rv = unregister_chrdev(mbcs_major, DEVICE_NAME); - if (rv < 0) - DBG(KERN_ALERT "Error in unregister_chrdev: %d\n", rv); - + unregister_chrdev(mbcs_major, DEVICE_NAME); cx_driver_unregister(&mbcs_driver); } diff --git a/drivers/char/mbcs.h b/drivers/char/mbcs.h index e7fd47e4325..c9905a3c335 100644 --- a/drivers/char/mbcs.h +++ b/drivers/char/mbcs.h @@ -542,12 +542,12 @@ struct mbcs_soft { struct semaphore algolock; }; -extern int mbcs_open(struct inode *ip, struct file *fp); -extern ssize_t mbcs_sram_read(struct file *fp, char __user *buf, size_t len, +static int mbcs_open(struct inode *ip, struct file *fp); +static ssize_t mbcs_sram_read(struct file *fp, char __user *buf, size_t len, loff_t * off); -extern ssize_t mbcs_sram_write(struct file *fp, const char __user *buf, size_t len, +static ssize_t mbcs_sram_write(struct file *fp, const char __user *buf, size_t len, loff_t * off); -extern loff_t mbcs_sram_llseek(struct file *filp, loff_t off, int whence); -extern int mbcs_gscr_mmap(struct file *fp, struct vm_area_struct *vma); +static loff_t mbcs_sram_llseek(struct file *filp, loff_t off, int whence); +static int mbcs_gscr_mmap(struct file *fp, struct vm_area_struct *vma); #endif // __MBCS_H__ diff --git a/drivers/char/pcmcia/synclink_cs.c b/drivers/char/pcmcia/synclink_cs.c index 13808f6083a..2b889317461 100644 --- a/drivers/char/pcmcia/synclink_cs.c +++ b/drivers/char/pcmcia/synclink_cs.c @@ -540,13 +540,12 @@ static int mgslpc_probe(struct pcmcia_device *link) if (debug_level >= DEBUG_LEVEL_INFO) printk("mgslpc_attach\n"); - info = kmalloc(sizeof(MGSLPC_INFO), GFP_KERNEL); + info = kzalloc(sizeof(MGSLPC_INFO), GFP_KERNEL); if (!info) { printk("Error can't allocate device instance data\n"); return -ENOMEM; } - memset(info, 0, sizeof(MGSLPC_INFO)); info->magic = MGSLPC_MAGIC; INIT_WORK(&info->task, bh_handler); info->max_frame_size = 4096; diff --git a/drivers/char/random.c b/drivers/char/random.c index 7f5271272f9..397c714cf2b 100644 --- a/drivers/char/random.c +++ b/drivers/char/random.c @@ -693,9 +693,14 @@ static void xfer_secondary_pool(struct entropy_store *r, size_t nbytes) if (r->pull && r->entropy_count < nbytes * 8 && r->entropy_count < r->poolinfo->POOLBITS) { - int bytes = max_t(int, random_read_wakeup_thresh / 8, - min_t(int, nbytes, sizeof(tmp))); + /* If we're limited, always leave two wakeup worth's BITS */ int rsvd = r->limit ? 0 : random_read_wakeup_thresh/4; + int bytes = nbytes; + + /* pull at least as many as BYTES as wakeup BITS */ + bytes = max_t(int, bytes, random_read_wakeup_thresh / 8); + /* but never more than the buffer size */ + bytes = min_t(int, bytes, sizeof(tmp)); DEBUG_ENT("going to reseed %s with %d bits " "(%d of %d requested)\n", diff --git a/drivers/char/rio/rio_linux.c b/drivers/char/rio/rio_linux.c index 294e9cb0c44..0ce96670f97 100644 --- a/drivers/char/rio/rio_linux.c +++ b/drivers/char/rio/rio_linux.c @@ -803,9 +803,7 @@ static void *ckmalloc(int size) { void *p; - p = kmalloc(size, GFP_KERNEL); - if (p) - memset(p, 0, size); + p = kzalloc(size, GFP_KERNEL); return p; } diff --git a/drivers/char/rio/riocmd.c b/drivers/char/rio/riocmd.c index 8cc60b69346..7321d002c34 100644 --- a/drivers/char/rio/riocmd.c +++ b/drivers/char/rio/riocmd.c @@ -556,9 +556,7 @@ struct CmdBlk *RIOGetCmdBlk(void) { struct CmdBlk *CmdBlkP; - CmdBlkP = kmalloc(sizeof(struct CmdBlk), GFP_ATOMIC); - if (CmdBlkP) - memset(CmdBlkP, 0, sizeof(struct CmdBlk)); + CmdBlkP = kzalloc(sizeof(struct CmdBlk), GFP_ATOMIC); return CmdBlkP; } diff --git a/drivers/char/rio/riotable.c b/drivers/char/rio/riotable.c index 7e988357326..991119c9f47 100644 --- a/drivers/char/rio/riotable.c +++ b/drivers/char/rio/riotable.c @@ -863,8 +863,7 @@ int RIOReMapPorts(struct rio_info *p, struct Host *HostP, struct Map *HostMapP) if (PortP->TxRingBuffer) memset(PortP->TxRingBuffer, 0, p->RIOBufferSize); else if (p->RIOBufferSize) { - PortP->TxRingBuffer = kmalloc(p->RIOBufferSize, GFP_KERNEL); - memset(PortP->TxRingBuffer, 0, p->RIOBufferSize); + PortP->TxRingBuffer = kzalloc(p->RIOBufferSize, GFP_KERNEL); } PortP->TxBufferOut = 0; PortP->TxBufferIn = 0; diff --git a/drivers/char/rocket.c b/drivers/char/rocket.c index 0270080ff0c..56cbba7b6ec 100644 --- a/drivers/char/rocket.c +++ b/drivers/char/rocket.c @@ -635,12 +635,11 @@ static void init_r_port(int board, int aiop, int chan, struct pci_dev *pci_dev) ctlp = sCtlNumToCtlPtr(board); /* Get a r_port struct for the port, fill it in and save it globally, indexed by line number */ - info = kmalloc(sizeof (struct r_port), GFP_KERNEL); + info = kzalloc(sizeof (struct r_port), GFP_KERNEL); if (!info) { printk(KERN_INFO "Couldn't allocate info struct for line #%d\n", line); return; } - memset(info, 0, sizeof (struct r_port)); info->magic = RPORT_MAGIC; info->line = line; diff --git a/drivers/char/stallion.c b/drivers/char/stallion.c index 93d0bb8b4c0..4a80b2f864e 100644 --- a/drivers/char/stallion.c +++ b/drivers/char/stallion.c @@ -4795,7 +4795,6 @@ static void __exit stallion_module_exit(void) { struct stlbrd *brdp; unsigned int i, j; - int retval; pr_debug("cleanup_module()\n"); @@ -4818,9 +4817,7 @@ static void __exit stallion_module_exit(void) for (i = 0; i < 4; i++) class_device_destroy(stallion_class, MKDEV(STL_SIOMEMMAJOR, i)); - if ((retval = unregister_chrdev(STL_SIOMEMMAJOR, "staliomem"))) - printk("STALLION: failed to un-register serial memory device, " - "errno=%d\n", -retval); + unregister_chrdev(STL_SIOMEMMAJOR, "staliomem"); class_destroy(stallion_class); pci_unregister_driver(&stl_pcidriver); diff --git a/drivers/char/synclink.c b/drivers/char/synclink.c index f53e51ddb9d..fdc256b380b 100644 --- a/drivers/char/synclink.c +++ b/drivers/char/synclink.c @@ -4324,13 +4324,12 @@ static struct mgsl_struct* mgsl_allocate_device(void) { struct mgsl_struct *info; - info = kmalloc(sizeof(struct mgsl_struct), + info = kzalloc(sizeof(struct mgsl_struct), GFP_KERNEL); if (!info) { printk("Error can't allocate device instance data\n"); } else { - memset(info, 0, sizeof(struct mgsl_struct)); info->magic = MGSL_MAGIC; INIT_WORK(&info->task, mgsl_bh_handler); info->max_frame_size = 4096; diff --git a/drivers/char/synclink_gt.c b/drivers/char/synclink_gt.c index 428b514201f..372a37e2562 100644 --- a/drivers/char/synclink_gt.c +++ b/drivers/char/synclink_gt.c @@ -3414,13 +3414,12 @@ static struct slgt_info *alloc_dev(int adapter_num, int port_num, struct pci_dev { struct slgt_info *info; - info = kmalloc(sizeof(struct slgt_info), GFP_KERNEL); + info = kzalloc(sizeof(struct slgt_info), GFP_KERNEL); if (!info) { DBGERR(("%s device alloc failed adapter=%d port=%d\n", driver_name, adapter_num, port_num)); } else { - memset(info, 0, sizeof(struct slgt_info)); info->magic = MGSL_MAGIC; INIT_WORK(&info->task, bh_handler); info->max_frame_size = 4096; diff --git a/drivers/char/synclinkmp.c b/drivers/char/synclinkmp.c index a65407b3207..c63013b2fc3 100644 --- a/drivers/char/synclinkmp.c +++ b/drivers/char/synclinkmp.c @@ -3786,14 +3786,13 @@ static SLMP_INFO *alloc_dev(int adapter_num, int port_num, struct pci_dev *pdev) { SLMP_INFO *info; - info = kmalloc(sizeof(SLMP_INFO), + info = kzalloc(sizeof(SLMP_INFO), GFP_KERNEL); if (!info) { printk("%s(%d) Error can't allocate device instance data for adapter %d, port %d\n", __FILE__,__LINE__, adapter_num, port_num); } else { - memset(info, 0, sizeof(SLMP_INFO)); info->magic = MGSL_MAGIC; INIT_WORK(&info->task, bh_handler); info->max_frame_size = 4096; diff --git a/drivers/char/viotape.c b/drivers/char/viotape.c index db57277117b..e12275df6ea 100644 --- a/drivers/char/viotape.c +++ b/drivers/char/viotape.c @@ -1098,15 +1098,10 @@ static int chg_state(int index, unsigned char new_state, struct file *file) /* Cleanup */ static void __exit viotap_exit(void) { - int ret; - remove_proc_entry("iSeries/viotape", NULL); vio_unregister_driver(&viotape_driver); class_destroy(tape_class); - ret = unregister_chrdev(VIOTAPE_MAJOR, "viotape"); - if (ret < 0) - printk(VIOTAPE_KERN_WARN "Error unregistering device: %d\n", - ret); + unregister_chrdev(VIOTAPE_MAJOR, "viotape"); if (viotape_unitinfo) dma_free_coherent(iSeries_vio_dev, sizeof(viotape_unitinfo[0]) * VIOTAPE_MAX_TAPE, diff --git a/drivers/char/watchdog/mpcore_wdt.c b/drivers/char/watchdog/mpcore_wdt.c index e88947f8fe5..0d2b2773541 100644 --- a/drivers/char/watchdog/mpcore_wdt.c +++ b/drivers/char/watchdog/mpcore_wdt.c @@ -328,12 +328,11 @@ static int __devinit mpcore_wdt_probe(struct platform_device *dev) goto err_out; } - wdt = kmalloc(sizeof(struct mpcore_wdt), GFP_KERNEL); + wdt = kzalloc(sizeof(struct mpcore_wdt), GFP_KERNEL); if (!wdt) { ret = -ENOMEM; goto err_out; } - memset(wdt, 0, sizeof(struct mpcore_wdt)); wdt->dev = &dev->dev; wdt->irq = platform_get_irq(dev, 0); diff --git a/drivers/char/watchdog/pcwd_usb.c b/drivers/char/watchdog/pcwd_usb.c index 1e7a6719d5b..0f3fd6c9c35 100644 --- a/drivers/char/watchdog/pcwd_usb.c +++ b/drivers/char/watchdog/pcwd_usb.c @@ -626,12 +626,11 @@ static int usb_pcwd_probe(struct usb_interface *interface, const struct usb_devi maxp = usb_maxpacket(udev, pipe, usb_pipeout(pipe)); /* allocate memory for our device and initialize it */ - usb_pcwd = kmalloc (sizeof(struct usb_pcwd_private), GFP_KERNEL); + usb_pcwd = kzalloc (sizeof(struct usb_pcwd_private), GFP_KERNEL); if (usb_pcwd == NULL) { printk(KERN_ERR PFX "Out of memory\n"); goto error; } - memset (usb_pcwd, 0x00, sizeof (*usb_pcwd)); usb_pcwd_device = usb_pcwd; diff --git a/drivers/edac/Kconfig b/drivers/edac/Kconfig index debf1d8e8b4..1724c41d241 100644 --- a/drivers/edac/Kconfig +++ b/drivers/edac/Kconfig @@ -3,18 +3,18 @@ # Copyright (c) 2003 Linux Networx # Licensed and distributed under the GPL # -# $Id: Kconfig,v 1.4.2.7 2005/07/08 22:05:38 dsp_llnl Exp $ -# menuconfig EDAC - tristate "EDAC - error detection and reporting (EXPERIMENTAL)" + bool "EDAC - error detection and reporting (EXPERIMENTAL)" depends on HAS_IOMEM - depends on X86 && EXPERIMENTAL + depends on EXPERIMENTAL + depends on X86 || MIPS || PPC help EDAC is designed to report errors in the core system. These are low-level errors that are reported in the CPU or - supporting chipset: memory errors, cache errors, PCI errors, - thermal throttling, etc.. If unsure, select 'Y'. + supporting chipset or other subsystems: + memory errors, cache errors, PCI errors, thermal throttling, etc.. + If unsure, select 'Y'. If this code is reporting problems on your system, please see the EDAC project web pages for more information at: @@ -73,6 +73,14 @@ config EDAC_E752X Support for error detection and correction on the Intel E7520, E7525, E7320 server chipsets. +config EDAC_I82443BXGX + tristate "Intel 82443BX/GX (440BX/GX)" + depends on EDAC_MM_EDAC && PCI && X86_32 + depends on BROKEN + help + Support for error detection and correction on the Intel + 82443BX/GX memory controllers (440BX/GX chipsets). + config EDAC_I82875P tristate "Intel 82875p (D82875P, E7210)" depends on EDAC_MM_EDAC && PCI && X86_32 @@ -80,6 +88,20 @@ config EDAC_I82875P Support for error detection and correction on the Intel DP82785P and E7210 server chipsets. +config EDAC_I82975X + tristate "Intel 82975x (D82975x)" + depends on EDAC_MM_EDAC && PCI && X86 + help + Support for error detection and correction on the Intel + DP82975x server chipsets. + +config EDAC_I3000 + tristate "Intel 3000/3010" + depends on EDAC_MM_EDAC && PCI && X86_32 + help + Support for error detection and correction on the Intel + 3000 and 3010 server chipsets. + config EDAC_I82860 tristate "Intel 82860" depends on EDAC_MM_EDAC && PCI && X86_32 @@ -94,15 +116,20 @@ config EDAC_R82600 Support for error detection and correction on the Radisys 82600 embedded chipset. -choice - prompt "Error detecting method" - default EDAC_POLL +config EDAC_I5000 + tristate "Intel Greencreek/Blackford chipset" + depends on EDAC_MM_EDAC && X86 && PCI + help + Support for error detection and correction the Intel + Greekcreek/Blackford chipsets. -config EDAC_POLL - bool "Poll for errors" +config EDAC_PASEMI + tristate "PA Semi PWRficient" + depends on EDAC_MM_EDAC && PCI + depends on PPC help - Poll the chipset periodically to detect errors. + Support for error detection and correction on PA Semi + PWRficient. -endchoice endif # EDAC diff --git a/drivers/edac/Makefile b/drivers/edac/Makefile index 93137fdab4b..02c09f0ff15 100644 --- a/drivers/edac/Makefile +++ b/drivers/edac/Makefile @@ -5,14 +5,27 @@ # This file may be distributed under the terms of the # GNU General Public License. # -# $Id: Makefile,v 1.4.2.3 2005/07/08 22:05:38 dsp_llnl Exp $ -obj-$(CONFIG_EDAC_MM_EDAC) += edac_mc.o +obj-$(CONFIG_EDAC) := edac_stub.o +obj-$(CONFIG_EDAC_MM_EDAC) += edac_core.o + +edac_core-objs := edac_mc.o edac_device.o edac_mc_sysfs.o edac_pci_sysfs.o +edac_core-objs += edac_module.o edac_device_sysfs.o + +ifdef CONFIG_PCI +edac_core-objs += edac_pci.o edac_pci_sysfs.o +endif + obj-$(CONFIG_EDAC_AMD76X) += amd76x_edac.o +obj-$(CONFIG_EDAC_I5000) += i5000_edac.o obj-$(CONFIG_EDAC_E7XXX) += e7xxx_edac.o obj-$(CONFIG_EDAC_E752X) += e752x_edac.o +obj-$(CONFIG_EDAC_I82443BXGX) += i82443bxgx_edac.o obj-$(CONFIG_EDAC_I82875P) += i82875p_edac.o +obj-$(CONFIG_EDAC_I82975X) += i82975x_edac.o +obj-$(CONFIG_EDAC_I3000) += i3000_edac.o obj-$(CONFIG_EDAC_I82860) += i82860_edac.o obj-$(CONFIG_EDAC_R82600) += r82600_edac.o +obj-$(CONFIG_EDAC_PASEMI) += pasemi_edac.o diff --git a/drivers/edac/amd76x_edac.c b/drivers/edac/amd76x_edac.c index f79f6b587bf..f2207541059 100644 --- a/drivers/edac/amd76x_edac.c +++ b/drivers/edac/amd76x_edac.c @@ -17,9 +17,9 @@ #include <linux/pci.h> #include <linux/pci_ids.h> #include <linux/slab.h> -#include "edac_mc.h" +#include "edac_core.h" -#define AMD76X_REVISION " Ver: 2.0.1 " __DATE__ +#define AMD76X_REVISION " Ver: 2.0.2 " __DATE__ #define EDAC_MOD_STR "amd76x_edac" #define amd76x_printk(level, fmt, arg...) \ @@ -86,13 +86,13 @@ struct amd76x_dev_info { static const struct amd76x_dev_info amd76x_devs[] = { [AMD761] = { - .ctl_name = "AMD761" - }, + .ctl_name = "AMD761"}, [AMD762] = { - .ctl_name = "AMD762" - }, + .ctl_name = "AMD762"}, }; +static struct edac_pci_ctl_info *amd76x_pci; + /** * amd76x_get_error_info - fetch error information * @mci: Memory controller @@ -102,21 +102,21 @@ static const struct amd76x_dev_info amd76x_devs[] = { * on the chip so that further errors will be reported */ static void amd76x_get_error_info(struct mem_ctl_info *mci, - struct amd76x_error_info *info) + struct amd76x_error_info *info) { struct pci_dev *pdev; pdev = to_pci_dev(mci->dev); pci_read_config_dword(pdev, AMD76X_ECC_MODE_STATUS, - &info->ecc_mode_status); + &info->ecc_mode_status); if (info->ecc_mode_status & BIT(8)) pci_write_bits32(pdev, AMD76X_ECC_MODE_STATUS, - (u32) BIT(8), (u32) BIT(8)); + (u32) BIT(8), (u32) BIT(8)); if (info->ecc_mode_status & BIT(9)) pci_write_bits32(pdev, AMD76X_ECC_MODE_STATUS, - (u32) BIT(9), (u32) BIT(9)); + (u32) BIT(9), (u32) BIT(9)); } /** @@ -130,7 +130,8 @@ static void amd76x_get_error_info(struct mem_ctl_info *mci, * then attempt to handle and clean up after the error */ static int amd76x_process_error_info(struct mem_ctl_info *mci, - struct amd76x_error_info *info, int handle_errors) + struct amd76x_error_info *info, + int handle_errors) { int error_found; u32 row; @@ -138,7 +139,7 @@ static int amd76x_process_error_info(struct mem_ctl_info *mci, error_found = 0; /* - * Check for an uncorrectable error + * Check for an uncorrectable error */ if (info->ecc_mode_status & BIT(8)) { error_found = 1; @@ -146,12 +147,12 @@ static int amd76x_process_error_info(struct mem_ctl_info *mci, if (handle_errors) { row = (info->ecc_mode_status >> 4) & 0xf; edac_mc_handle_ue(mci, mci->csrows[row].first_page, 0, - row, mci->ctl_name); + row, mci->ctl_name); } } /* - * Check for a correctable error + * Check for a correctable error */ if (info->ecc_mode_status & BIT(9)) { error_found = 1; @@ -159,7 +160,7 @@ static int amd76x_process_error_info(struct mem_ctl_info *mci, if (handle_errors) { row = info->ecc_mode_status & 0xf; edac_mc_handle_ce(mci, mci->csrows[row].first_page, 0, - 0, row, 0, mci->ctl_name); + 0, row, 0, mci->ctl_name); } } @@ -182,7 +183,7 @@ static void amd76x_check(struct mem_ctl_info *mci) } static void amd76x_init_csrows(struct mem_ctl_info *mci, struct pci_dev *pdev, - enum edac_type edac_mode) + enum edac_type edac_mode) { struct csrow_info *csrow; u32 mba, mba_base, mba_mask, dms; @@ -193,8 +194,7 @@ static void amd76x_init_csrows(struct mem_ctl_info *mci, struct pci_dev *pdev, /* find the DRAM Chip Select Base address and mask */ pci_read_config_dword(pdev, - AMD76X_MEM_BASE_ADDR + (index * 4), - &mba); + AMD76X_MEM_BASE_ADDR + (index * 4), &mba); if (!(mba & BIT(0))) continue; @@ -238,7 +238,7 @@ static int amd76x_probe1(struct pci_dev *pdev, int dev_idx) debugf0("%s()\n", __func__); pci_read_config_dword(pdev, AMD76X_ECC_MODE_STATUS, &ems); ems_mode = (ems >> 10) & 0x3; - mci = edac_mc_alloc(0, AMD76X_NR_CSROWS, AMD76X_NR_CHANS); + mci = edac_mc_alloc(0, AMD76X_NR_CSROWS, AMD76X_NR_CHANS, 0); if (mci == NULL) { return -ENOMEM; @@ -249,24 +249,36 @@ static int amd76x_probe1(struct pci_dev *pdev, int dev_idx) mci->mtype_cap = MEM_FLAG_RDDR; mci->edac_ctl_cap = EDAC_FLAG_NONE | EDAC_FLAG_EC | EDAC_FLAG_SECDED; mci->edac_cap = ems_mode ? - (EDAC_FLAG_EC | EDAC_FLAG_SECDED) : EDAC_FLAG_NONE; + (EDAC_FLAG_EC | EDAC_FLAG_SECDED) : EDAC_FLAG_NONE; mci->mod_name = EDAC_MOD_STR; mci->mod_ver = AMD76X_REVISION; mci->ctl_name = amd76x_devs[dev_idx].ctl_name; + mci->dev_name = pci_name(pdev); mci->edac_check = amd76x_check; mci->ctl_page_to_phys = NULL; amd76x_init_csrows(mci, pdev, ems_modes[ems_mode]); - amd76x_get_error_info(mci, &discard); /* clear counters */ + amd76x_get_error_info(mci, &discard); /* clear counters */ /* Here we assume that we will never see multiple instances of this * type of memory controller. The ID is therefore hardcoded to 0. */ - if (edac_mc_add_mc(mci,0)) { + if (edac_mc_add_mc(mci)) { debugf3("%s(): failed edac_mc_add_mc()\n", __func__); goto fail; } + /* allocating generic PCI control info */ + amd76x_pci = edac_pci_create_generic_ctl(&pdev->dev, EDAC_MOD_STR); + if (!amd76x_pci) { + printk(KERN_WARNING + "%s(): Unable to create PCI control\n", + __func__); + printk(KERN_WARNING + "%s(): PCI error report via EDAC not setup\n", + __func__); + } + /* get this far and it's successful */ debugf3("%s(): success\n", __func__); return 0; @@ -278,7 +290,7 @@ fail: /* returns count (>= 0), or negative on error */ static int __devinit amd76x_init_one(struct pci_dev *pdev, - const struct pci_device_id *ent) + const struct pci_device_id *ent) { debugf0("%s()\n", __func__); @@ -300,6 +312,9 @@ static void __devexit amd76x_remove_one(struct pci_dev *pdev) debugf0("%s()\n", __func__); + if (amd76x_pci) + edac_pci_release_generic_ctl(amd76x_pci); + if ((mci = edac_mc_del_mc(&pdev->dev)) == NULL) return; @@ -308,16 +323,14 @@ static void __devexit amd76x_remove_one(struct pci_dev *pdev) static const struct pci_device_id amd76x_pci_tbl[] __devinitdata = { { - PCI_VEND_DEV(AMD, FE_GATE_700C), PCI_ANY_ID, PCI_ANY_ID, 0, 0, - AMD762 - }, + PCI_VEND_DEV(AMD, FE_GATE_700C), PCI_ANY_ID, PCI_ANY_ID, 0, 0, + AMD762}, { - PCI_VEND_DEV(AMD, FE_GATE_700E), PCI_ANY_ID, PCI_ANY_ID, 0, 0, - AMD761 - }, + PCI_VEND_DEV(AMD, FE_GATE_700E), PCI_ANY_ID, PCI_ANY_ID, 0, 0, + AMD761}, { - 0, - } /* 0 terminated list. */ + 0, + } /* 0 terminated list. */ }; MODULE_DEVICE_TABLE(pci, amd76x_pci_tbl); diff --git a/drivers/edac/e752x_edac.c b/drivers/edac/e752x_edac.c index 8bcc887692a..3bba224cb55 100644 --- a/drivers/edac/e752x_edac.c +++ b/drivers/edac/e752x_edac.c @@ -22,13 +22,16 @@ #include <linux/pci.h> #include <linux/pci_ids.h> #include <linux/slab.h> -#include "edac_mc.h" +#include <linux/edac.h> +#include "edac_core.h" -#define E752X_REVISION " Ver: 2.0.1 " __DATE__ +#define E752X_REVISION " Ver: 2.0.2 " __DATE__ #define EDAC_MOD_STR "e752x_edac" static int force_function_unhide; +static struct edac_pci_ctl_info *e752x_pci; + #define e752x_printk(level, fmt, arg...) \ edac_printk(level, "e752x", fmt, ##arg) @@ -203,25 +206,22 @@ static const struct e752x_dev_info e752x_devs[] = { [E7520] = { .err_dev = PCI_DEVICE_ID_INTEL_7520_1_ERR, .ctl_dev = PCI_DEVICE_ID_INTEL_7520_0, - .ctl_name = "E7520" - }, + .ctl_name = "E7520"}, [E7525] = { .err_dev = PCI_DEVICE_ID_INTEL_7525_1_ERR, .ctl_dev = PCI_DEVICE_ID_INTEL_7525_0, - .ctl_name = "E7525" - }, + .ctl_name = "E7525"}, [E7320] = { .err_dev = PCI_DEVICE_ID_INTEL_7320_1_ERR, .ctl_dev = PCI_DEVICE_ID_INTEL_7320_0, - .ctl_name = "E7320" - }, + .ctl_name = "E7320"}, }; static unsigned long ctl_page_to_phys(struct mem_ctl_info *mci, - unsigned long page) + unsigned long page) { u32 remap; - struct e752x_pvt *pvt = (struct e752x_pvt *) mci->pvt_info; + struct e752x_pvt *pvt = (struct e752x_pvt *)mci->pvt_info; debugf3("%s()\n", __func__); @@ -241,13 +241,13 @@ static unsigned long ctl_page_to_phys(struct mem_ctl_info *mci, } static void do_process_ce(struct mem_ctl_info *mci, u16 error_one, - u32 sec1_add, u16 sec1_syndrome) + u32 sec1_add, u16 sec1_syndrome) { u32 page; int row; int channel; int i; - struct e752x_pvt *pvt = (struct e752x_pvt *) mci->pvt_info; + struct e752x_pvt *pvt = (struct e752x_pvt *)mci->pvt_info; debugf3("%s()\n", __func__); @@ -261,7 +261,8 @@ static void do_process_ce(struct mem_ctl_info *mci, u16 error_one, e752x_printk(KERN_WARNING, "Test row %d Table %d %d %d %d %d %d %d %d\n", row, pvt->map[0], pvt->map[1], pvt->map[2], pvt->map[3], - pvt->map[4], pvt->map[5], pvt->map[6], pvt->map[7]); + pvt->map[4], pvt->map[5], pvt->map[6], + pvt->map[7]); /* test for channel remapping */ for (i = 0; i < 8; i++) { @@ -275,24 +276,22 @@ static void do_process_ce(struct mem_ctl_info *mci, u16 error_one, row = i; else e752x_mc_printk(mci, KERN_WARNING, - "row %d not found in remap table\n", row); + "row %d not found in remap table\n", + row); } else row = edac_mc_find_csrow_by_page(mci, page); /* 0 = channel A, 1 = channel B */ channel = !(error_one & 1); - if (!pvt->map_type) - row = 7 - row; - /* e752x mc reads 34:6 of the DRAM linear address */ edac_mc_handle_ce(mci, page, offset_in_page(sec1_add << 4), sec1_syndrome, row, channel, "e752x CE"); } static inline void process_ce(struct mem_ctl_info *mci, u16 error_one, - u32 sec1_add, u16 sec1_syndrome, int *error_found, - int handle_error) + u32 sec1_add, u16 sec1_syndrome, int *error_found, + int handle_error) { *error_found = 1; @@ -301,11 +300,11 @@ static inline void process_ce(struct mem_ctl_info *mci, u16 error_one, } static void do_process_ue(struct mem_ctl_info *mci, u16 error_one, - u32 ded_add, u32 scrb_add) + u32 ded_add, u32 scrb_add) { u32 error_2b, block_page; int row; - struct e752x_pvt *pvt = (struct e752x_pvt *) mci->pvt_info; + struct e752x_pvt *pvt = (struct e752x_pvt *)mci->pvt_info; debugf3("%s()\n", __func__); @@ -316,14 +315,14 @@ static void do_process_ue(struct mem_ctl_info *mci, u16 error_one, block_page = error_2b >> (PAGE_SHIFT - 4); row = pvt->mc_symmetric ? - /* chip select are bits 14 & 13 */ + /* chip select are bits 14 & 13 */ ((block_page >> 1) & 3) : edac_mc_find_csrow_by_page(mci, block_page); /* e752x mc reads 34:6 of the DRAM linear address */ edac_mc_handle_ue(mci, block_page, - offset_in_page(error_2b << 4), - row, "e752x UE from Read"); + offset_in_page(error_2b << 4), + row, "e752x UE from Read"); } if (error_one & 0x0404) { error_2b = scrb_add; @@ -332,19 +331,20 @@ static void do_process_ue(struct mem_ctl_info *mci, u16 error_one, block_page = error_2b >> (PAGE_SHIFT - 4); row = pvt->mc_symmetric ? - /* chip select are bits 14 & 13 */ + /* chip select are bits 14 & 13 */ ((block_page >> 1) & 3) : edac_mc_find_csrow_by_page(mci, block_page); /* e752x mc reads 34:6 of the DRAM linear address */ edac_mc_handle_ue(mci, block_page, - offset_in_page(error_2b << 4), - row, "e752x UE from Scruber"); + offset_in_page(error_2b << 4), + row, "e752x UE from Scruber"); } } static inline void process_ue(struct mem_ctl_info *mci, u16 error_one, - u32 ded_add, u32 scrb_add, int *error_found, int handle_error) + u32 ded_add, u32 scrb_add, int *error_found, + int handle_error) { *error_found = 1; @@ -353,7 +353,7 @@ static inline void process_ue(struct mem_ctl_info *mci, u16 error_one, } static inline void process_ue_no_info_wr(struct mem_ctl_info *mci, - int *error_found, int handle_error) + int *error_found, int handle_error) { *error_found = 1; @@ -365,24 +365,24 @@ static inline void process_ue_no_info_wr(struct mem_ctl_info *mci, } static void do_process_ded_retry(struct mem_ctl_info *mci, u16 error, - u32 retry_add) + u32 retry_add) { u32 error_1b, page; int row; - struct e752x_pvt *pvt = (struct e752x_pvt *) mci->pvt_info; + struct e752x_pvt *pvt = (struct e752x_pvt *)mci->pvt_info; error_1b = retry_add; - page = error_1b >> (PAGE_SHIFT - 4); /* convert the addr to 4k page */ - row = pvt->mc_symmetric ? - ((page >> 1) & 3) : /* chip select are bits 14 & 13 */ + page = error_1b >> (PAGE_SHIFT - 4); /* convert the addr to 4k page */ + row = pvt->mc_symmetric ? ((page >> 1) & 3) : /* chip select are bits 14 & 13 */ edac_mc_find_csrow_by_page(mci, page); e752x_mc_printk(mci, KERN_WARNING, - "CE page 0x%lx, row %d : Memory read retry\n", - (long unsigned int) page, row); + "CE page 0x%lx, row %d : Memory read retry\n", + (long unsigned int)page, row); } static inline void process_ded_retry(struct mem_ctl_info *mci, u16 error, - u32 retry_add, int *error_found, int handle_error) + u32 retry_add, int *error_found, + int handle_error) { *error_found = 1; @@ -391,7 +391,7 @@ static inline void process_ded_retry(struct mem_ctl_info *mci, u16 error, } static inline void process_threshold_ce(struct mem_ctl_info *mci, u16 error, - int *error_found, int handle_error) + int *error_found, int handle_error) { *error_found = 1; @@ -420,7 +420,7 @@ static void do_global_error(int fatal, u32 errors) } static inline void global_error(int fatal, u32 errors, int *error_found, - int handle_error) + int handle_error) { *error_found = 1; @@ -447,7 +447,7 @@ static void do_hub_error(int fatal, u8 errors) } static inline void hub_error(int fatal, u8 errors, int *error_found, - int handle_error) + int handle_error) { *error_found = 1; @@ -505,7 +505,7 @@ static void do_sysbus_error(int fatal, u32 errors) } static inline void sysbus_error(int fatal, u32 errors, int *error_found, - int handle_error) + int handle_error) { *error_found = 1; @@ -514,7 +514,7 @@ static inline void sysbus_error(int fatal, u32 errors, int *error_found, } static void e752x_check_hub_interface(struct e752x_error_info *info, - int *error_found, int handle_error) + int *error_found, int handle_error) { u8 stat8; @@ -522,33 +522,32 @@ static void e752x_check_hub_interface(struct e752x_error_info *info, stat8 = info->hi_ferr; - if(stat8 & 0x7f) { /* Error, so process */ + if (stat8 & 0x7f) { /* Error, so process */ stat8 &= 0x7f; - if(stat8 & 0x2b) + if (stat8 & 0x2b) hub_error(1, stat8 & 0x2b, error_found, handle_error); - if(stat8 & 0x54) + if (stat8 & 0x54) hub_error(0, stat8 & 0x54, error_found, handle_error); } - //pci_read_config_byte(dev,E752X_HI_NERR,&stat8); stat8 = info->hi_nerr; - if(stat8 & 0x7f) { /* Error, so process */ + if (stat8 & 0x7f) { /* Error, so process */ stat8 &= 0x7f; if (stat8 & 0x2b) hub_error(1, stat8 & 0x2b, error_found, handle_error); - if(stat8 & 0x54) + if (stat8 & 0x54) hub_error(0, stat8 & 0x54, error_found, handle_error); } } static void e752x_check_sysbus(struct e752x_error_info *info, - int *error_found, int handle_error) + int *error_found, int handle_error) { u32 stat32, error32; @@ -556,47 +555,47 @@ static void e752x_check_sysbus(struct e752x_error_info *info, stat32 = info->sysbus_ferr + (info->sysbus_nerr << 16); if (stat32 == 0) - return; /* no errors */ + return; /* no errors */ error32 = (stat32 >> 16) & 0x3ff; stat32 = stat32 & 0x3ff; - if(stat32 & 0x087) + if (stat32 & 0x087) sysbus_error(1, stat32 & 0x087, error_found, handle_error); - if(stat32 & 0x378) + if (stat32 & 0x378) sysbus_error(0, stat32 & 0x378, error_found, handle_error); - if(error32 & 0x087) + if (error32 & 0x087) sysbus_error(1, error32 & 0x087, error_found, handle_error); - if(error32 & 0x378) + if (error32 & 0x378) sysbus_error(0, error32 & 0x378, error_found, handle_error); } -static void e752x_check_membuf (struct e752x_error_info *info, - int *error_found, int handle_error) +static void e752x_check_membuf(struct e752x_error_info *info, + int *error_found, int handle_error) { u8 stat8; stat8 = info->buf_ferr; - if (stat8 & 0x0f) { /* Error, so process */ + if (stat8 & 0x0f) { /* Error, so process */ stat8 &= 0x0f; membuf_error(stat8, error_found, handle_error); } stat8 = info->buf_nerr; - if (stat8 & 0x0f) { /* Error, so process */ + if (stat8 & 0x0f) { /* Error, so process */ stat8 &= 0x0f; membuf_error(stat8, error_found, handle_error); } } -static void e752x_check_dram (struct mem_ctl_info *mci, - struct e752x_error_info *info, int *error_found, - int handle_error) +static void e752x_check_dram(struct mem_ctl_info *mci, + struct e752x_error_info *info, int *error_found, + int handle_error) { u16 error_one, error_next; @@ -604,55 +603,52 @@ static void e752x_check_dram (struct mem_ctl_info *mci, error_next = info->dram_nerr; /* decode and report errors */ - if(error_one & 0x0101) /* check first error correctable */ + if (error_one & 0x0101) /* check first error correctable */ process_ce(mci, error_one, info->dram_sec1_add, - info->dram_sec1_syndrome, error_found, - handle_error); + info->dram_sec1_syndrome, error_found, handle_error); - if(error_next & 0x0101) /* check next error correctable */ + if (error_next & 0x0101) /* check next error correctable */ process_ce(mci, error_next, info->dram_sec2_add, - info->dram_sec2_syndrome, error_found, - handle_error); + info->dram_sec2_syndrome, error_found, handle_error); - if(error_one & 0x4040) + if (error_one & 0x4040) process_ue_no_info_wr(mci, error_found, handle_error); - if(error_next & 0x4040) + if (error_next & 0x4040) process_ue_no_info_wr(mci, error_found, handle_error); - if(error_one & 0x2020) + if (error_one & 0x2020) process_ded_retry(mci, error_one, info->dram_retr_add, - error_found, handle_error); + error_found, handle_error); - if(error_next & 0x2020) + if (error_next & 0x2020) process_ded_retry(mci, error_next, info->dram_retr_add, - error_found, handle_error); + error_found, handle_error); - if(error_one & 0x0808) - process_threshold_ce(mci, error_one, error_found, - handle_error); + if (error_one & 0x0808) + process_threshold_ce(mci, error_one, error_found, handle_error); - if(error_next & 0x0808) + if (error_next & 0x0808) process_threshold_ce(mci, error_next, error_found, - handle_error); + handle_error); - if(error_one & 0x0606) + if (error_one & 0x0606) process_ue(mci, error_one, info->dram_ded_add, - info->dram_scrb_add, error_found, handle_error); + info->dram_scrb_add, error_found, handle_error); - if(error_next & 0x0606) + if (error_next & 0x0606) process_ue(mci, error_next, info->dram_ded_add, - info->dram_scrb_add, error_found, handle_error); + info->dram_scrb_add, error_found, handle_error); } -static void e752x_get_error_info (struct mem_ctl_info *mci, - struct e752x_error_info *info) +static void e752x_get_error_info(struct mem_ctl_info *mci, + struct e752x_error_info *info) { struct pci_dev *dev; struct e752x_pvt *pvt; memset(info, 0, sizeof(*info)); - pvt = (struct e752x_pvt *) mci->pvt_info; + pvt = (struct e752x_pvt *)mci->pvt_info; dev = pvt->dev_d0f1; pci_read_config_dword(dev, E752X_FERR_GLOBAL, &info->ferr_global); @@ -661,8 +657,7 @@ static void e752x_get_error_info (struct mem_ctl_info *mci, pci_read_config_word(dev, E752X_SYSBUS_FERR, &info->sysbus_ferr); pci_read_config_byte(dev, E752X_BUF_FERR, &info->buf_ferr); - pci_read_config_word(dev, E752X_DRAM_FERR, - &info->dram_ferr); + pci_read_config_word(dev, E752X_DRAM_FERR, &info->dram_ferr); pci_read_config_dword(dev, E752X_DRAM_SEC1_ADD, &info->dram_sec1_add); pci_read_config_word(dev, E752X_DRAM_SEC1_SYNDROME, @@ -688,7 +683,7 @@ static void e752x_get_error_info (struct mem_ctl_info *mci, if (info->dram_ferr) pci_write_bits16(pvt->bridge_ck, E752X_DRAM_FERR, - info->dram_ferr, info->dram_ferr); + info->dram_ferr, info->dram_ferr); pci_write_config_dword(dev, E752X_FERR_GLOBAL, info->ferr_global); @@ -701,8 +696,7 @@ static void e752x_get_error_info (struct mem_ctl_info *mci, pci_read_config_word(dev, E752X_SYSBUS_NERR, &info->sysbus_nerr); pci_read_config_byte(dev, E752X_BUF_NERR, &info->buf_nerr); - pci_read_config_word(dev, E752X_DRAM_NERR, - &info->dram_nerr); + pci_read_config_word(dev, E752X_DRAM_NERR, &info->dram_nerr); pci_read_config_dword(dev, E752X_DRAM_SEC2_ADD, &info->dram_sec2_add); pci_read_config_word(dev, E752X_DRAM_SEC2_SYNDROME, @@ -722,15 +716,16 @@ static void e752x_get_error_info (struct mem_ctl_info *mci, if (info->dram_nerr) pci_write_bits16(pvt->bridge_ck, E752X_DRAM_NERR, - info->dram_nerr, info->dram_nerr); + info->dram_nerr, info->dram_nerr); pci_write_config_dword(dev, E752X_NERR_GLOBAL, info->nerr_global); } } -static int e752x_process_error_info (struct mem_ctl_info *mci, - struct e752x_error_info *info, int handle_errors) +static int e752x_process_error_info(struct mem_ctl_info *mci, + struct e752x_error_info *info, + int handle_errors) { u32 error32, stat32; int error_found; @@ -776,26 +771,38 @@ static inline int dual_channel_active(u16 ddrcsr) return (((ddrcsr >> 12) & 3) == 3); } +/* Remap csrow index numbers if map_type is "reverse" + */ +static inline int remap_csrow_index(struct mem_ctl_info *mci, int index) +{ + struct e752x_pvt *pvt = mci->pvt_info; + + if (!pvt->map_type) + return (7 - index); + + return (index); +} + static void e752x_init_csrows(struct mem_ctl_info *mci, struct pci_dev *pdev, - u16 ddrcsr) + u16 ddrcsr) { struct csrow_info *csrow; unsigned long last_cumul_size; int index, mem_dev, drc_chan; - int drc_drbg; /* DRB granularity 0=64mb, 1=128mb */ - int drc_ddim; /* DRAM Data Integrity Mode 0=none, 2=edac */ + int drc_drbg; /* DRB granularity 0=64mb, 1=128mb */ + int drc_ddim; /* DRAM Data Integrity Mode 0=none, 2=edac */ u8 value; u32 dra, drc, cumul_size; dra = 0; - for (index=0; index < 4; index++) { + for (index = 0; index < 4; index++) { u8 dra_reg; - pci_read_config_byte(pdev, E752X_DRA+index, &dra_reg); + pci_read_config_byte(pdev, E752X_DRA + index, &dra_reg); dra |= dra_reg << (index * 8); } pci_read_config_dword(pdev, E752X_DRC, &drc); drc_chan = dual_channel_active(ddrcsr); - drc_drbg = drc_chan + 1; /* 128 in dual mode, 64 in single */ + drc_drbg = drc_chan + 1; /* 128 in dual mode, 64 in single */ drc_ddim = (drc >> 20) & 0x3; /* The dram row boundary (DRB) reg values are boundary address for @@ -806,7 +813,7 @@ static void e752x_init_csrows(struct mem_ctl_info *mci, struct pci_dev *pdev, for (last_cumul_size = index = 0; index < mci->nr_csrows; index++) { /* mem_dev 0=x8, 1=x4 */ mem_dev = (dra >> (index * 4 + 2)) & 0x3; - csrow = &mci->csrows[index]; + csrow = &mci->csrows[remap_csrow_index(mci, index)]; mem_dev = (mem_dev == 2); pci_read_config_byte(pdev, E752X_DRB + index, &value); @@ -843,10 +850,10 @@ static void e752x_init_csrows(struct mem_ctl_info *mci, struct pci_dev *pdev, } static void e752x_init_mem_map_table(struct pci_dev *pdev, - struct e752x_pvt *pvt) + struct e752x_pvt *pvt) { int index; - u8 value, last, row, stat8; + u8 value, last, row; last = 0; row = 0; @@ -858,7 +865,7 @@ static void e752x_init_mem_map_table(struct pci_dev *pdev, /* no dimm in the slot, so flag it as empty */ pvt->map[index] = 0xff; pvt->map[index + 1] = 0xff; - } else { /* there is a dimm in the slot */ + } else { /* there is a dimm in the slot */ pvt->map[index] = row; row++; last = value; @@ -866,31 +873,25 @@ static void e752x_init_mem_map_table(struct pci_dev *pdev, * sided */ pci_read_config_byte(pdev, E752X_DRB + index + 1, - &value); - pvt->map[index + 1] = (value == last) ? - 0xff : /* the dimm is single sided, - so flag as empty */ - row; /* this is a double sided dimm - to save the next row # */ + &value); + + /* the dimm is single sided, so flag as empty */ + /* this is a double sided dimm to save the next row #*/ + pvt->map[index + 1] = (value == last) ? 0xff : row; row++; last = value; } } - - /* set the map type. 1 = normal, 0 = reversed */ - pci_read_config_byte(pdev, E752X_DRM, &stat8); - pvt->map_type = ((stat8 & 0x0f) > ((stat8 >> 4) & 0x0f)); } /* Return 0 on success or 1 on failure. */ static int e752x_get_devs(struct pci_dev *pdev, int dev_idx, - struct e752x_pvt *pvt) + struct e752x_pvt *pvt) { struct pci_dev *dev; pvt->bridge_ck = pci_get_device(PCI_VENDOR_ID_INTEL, - pvt->dev_info->err_dev, - pvt->bridge_ck); + pvt->dev_info->err_dev, pvt->bridge_ck); if (pvt->bridge_ck == NULL) pvt->bridge_ck = pci_scan_single_device(pdev->bus, @@ -898,13 +899,13 @@ static int e752x_get_devs(struct pci_dev *pdev, int dev_idx, if (pvt->bridge_ck == NULL) { e752x_printk(KERN_ERR, "error reporting device not found:" - "vendor %x device 0x%x (broken BIOS?)\n", - PCI_VENDOR_ID_INTEL, e752x_devs[dev_idx].err_dev); + "vendor %x device 0x%x (broken BIOS?)\n", + PCI_VENDOR_ID_INTEL, e752x_devs[dev_idx].err_dev); return 1; } dev = pci_get_device(PCI_VENDOR_ID_INTEL, e752x_devs[dev_idx].ctl_dev, - NULL); + NULL); if (dev == NULL) goto fail; @@ -942,12 +943,22 @@ static int e752x_probe1(struct pci_dev *pdev, int dev_idx) struct mem_ctl_info *mci; struct e752x_pvt *pvt; u16 ddrcsr; - int drc_chan; /* Number of channels 0=1chan,1=2chan */ + int drc_chan; /* Number of channels 0=1chan,1=2chan */ struct e752x_error_info discard; debugf0("%s(): mci\n", __func__); debugf0("Starting Probe1\n"); + /* make sure error reporting method is sane */ + switch (edac_op_state) { + case EDAC_OPSTATE_POLL: + case EDAC_OPSTATE_NMI: + break; + default: + edac_op_state = EDAC_OPSTATE_POLL; + break; + } + /* check to see if device 0 function 1 is enabled; if it isn't, we * assume the BIOS has reserved it for a reason and is expecting * exclusive access, we take care not to violate that assumption and @@ -966,7 +977,7 @@ static int e752x_probe1(struct pci_dev *pdev, int dev_idx) /* Dual channel = 1, Single channel = 0 */ drc_chan = dual_channel_active(ddrcsr); - mci = edac_mc_alloc(sizeof(*pvt), E752X_NR_CSROWS, drc_chan + 1); + mci = edac_mc_alloc(sizeof(*pvt), E752X_NR_CSROWS, drc_chan + 1, 0); if (mci == NULL) { return -ENOMEM; @@ -975,14 +986,14 @@ static int e752x_probe1(struct pci_dev *pdev, int dev_idx) debugf3("%s(): init mci\n", __func__); mci->mtype_cap = MEM_FLAG_RDDR; mci->edac_ctl_cap = EDAC_FLAG_NONE | EDAC_FLAG_SECDED | - EDAC_FLAG_S4ECD4ED; + EDAC_FLAG_S4ECD4ED; /* FIXME - what if different memory types are in different csrows? */ mci->mod_name = EDAC_MOD_STR; mci->mod_ver = E752X_REVISION; mci->dev = &pdev->dev; debugf3("%s(): init pvt\n", __func__); - pvt = (struct e752x_pvt *) mci->pvt_info; + pvt = (struct e752x_pvt *)mci->pvt_info; pvt->dev_info = &e752x_devs[dev_idx]; pvt->mc_symmetric = ((ddrcsr & 0x10) != 0); @@ -993,16 +1004,20 @@ static int e752x_probe1(struct pci_dev *pdev, int dev_idx) debugf3("%s(): more mci init\n", __func__); mci->ctl_name = pvt->dev_info->ctl_name; + mci->dev_name = pci_name(pdev); mci->edac_check = e752x_check; mci->ctl_page_to_phys = ctl_page_to_phys; - e752x_init_csrows(mci, pdev, ddrcsr); - e752x_init_mem_map_table(pdev, pvt); - - /* set the map type. 1 = normal, 0 = reversed */ + /* set the map type. 1 = normal, 0 = reversed + * Must be set before e752x_init_csrows in case csrow mapping + * is reversed. + */ pci_read_config_byte(pdev, E752X_DRM, &stat8); pvt->map_type = ((stat8 & 0x0f) > ((stat8 >> 4) & 0x0f)); + e752x_init_csrows(mci, pdev, ddrcsr); + e752x_init_mem_map_table(pdev, pvt); + mci->edac_cap |= EDAC_FLAG_NONE; debugf3("%s(): tolm, remapbase, remaplimit\n", __func__); @@ -1014,19 +1029,29 @@ static int e752x_probe1(struct pci_dev *pdev, int dev_idx) pci_read_config_word(pdev, E752X_REMAPLIMIT, &pci_data); pvt->remaplimit = ((u32) pci_data) << 14; e752x_printk(KERN_INFO, - "tolm = %x, remapbase = %x, remaplimit = %x\n", pvt->tolm, - pvt->remapbase, pvt->remaplimit); + "tolm = %x, remapbase = %x, remaplimit = %x\n", + pvt->tolm, pvt->remapbase, pvt->remaplimit); /* Here we assume that we will never see multiple instances of this * type of memory controller. The ID is therefore hardcoded to 0. */ - if (edac_mc_add_mc(mci,0)) { + if (edac_mc_add_mc(mci)) { debugf3("%s(): failed edac_mc_add_mc()\n", __func__); goto fail; } e752x_init_error_reporting_regs(pvt); - e752x_get_error_info(mci, &discard); /* clear other MCH errors */ + e752x_get_error_info(mci, &discard); /* clear other MCH errors */ + + /* allocating generic PCI control info */ + e752x_pci = edac_pci_create_generic_ctl(&pdev->dev, EDAC_MOD_STR); + if (!e752x_pci) { + printk(KERN_WARNING + "%s(): Unable to create PCI control\n", __func__); + printk(KERN_WARNING + "%s(): PCI error report via EDAC not setup\n", + __func__); + } /* get this far and it's successful */ debugf3("%s(): success\n", __func__); @@ -1043,12 +1068,12 @@ fail: /* returns count (>= 0), or negative on error */ static int __devinit e752x_init_one(struct pci_dev *pdev, - const struct pci_device_id *ent) + const struct pci_device_id *ent) { debugf0("%s()\n", __func__); /* wake up and enable device */ - if(pci_enable_device(pdev) < 0) + if (pci_enable_device(pdev) < 0) return -EIO; return e752x_probe1(pdev, ent->driver_data); @@ -1061,10 +1086,13 @@ static void __devexit e752x_remove_one(struct pci_dev *pdev) debugf0("%s()\n", __func__); + if (e752x_pci) + edac_pci_release_generic_ctl(e752x_pci); + if ((mci = edac_mc_del_mc(&pdev->dev)) == NULL) return; - pvt = (struct e752x_pvt *) mci->pvt_info; + pvt = (struct e752x_pvt *)mci->pvt_info; pci_dev_put(pvt->dev_d0f0); pci_dev_put(pvt->dev_d0f1); pci_dev_put(pvt->bridge_ck); @@ -1073,20 +1101,17 @@ static void __devexit e752x_remove_one(struct pci_dev *pdev) static const struct pci_device_id e752x_pci_tbl[] __devinitdata = { { - PCI_VEND_DEV(INTEL, 7520_0), PCI_ANY_ID, PCI_ANY_ID, 0, 0, - E7520 - }, + PCI_VEND_DEV(INTEL, 7520_0), PCI_ANY_ID, PCI_ANY_ID, 0, 0, + E7520}, { - PCI_VEND_DEV(INTEL, 7525_0), PCI_ANY_ID, PCI_ANY_ID, 0, 0, - E7525 - }, + PCI_VEND_DEV(INTEL, 7525_0), PCI_ANY_ID, PCI_ANY_ID, 0, 0, + E7525}, { - PCI_VEND_DEV(INTEL, 7320_0), PCI_ANY_ID, PCI_ANY_ID, 0, 0, - E7320 - }, + PCI_VEND_DEV(INTEL, 7320_0), PCI_ANY_ID, PCI_ANY_ID, 0, 0, + E7320}, { - 0, - } /* 0 terminated list. */ + 0, + } /* 0 terminated list. */ }; MODULE_DEVICE_TABLE(pci, e752x_pci_tbl); @@ -1122,5 +1147,6 @@ MODULE_DESCRIPTION("MC support for Intel e752x memory controllers"); module_param(force_function_unhide, int, 0444); MODULE_PARM_DESC(force_function_unhide, "if BIOS sets Dev0:Fun1 up as hidden:" -" 1=force unhide and hope BIOS doesn't fight driver for Dev0:Fun1 access"); - + " 1=force unhide and hope BIOS doesn't fight driver for Dev0:Fun1 access"); +module_param(edac_op_state, int, 0444); +MODULE_PARM_DESC(edac_op_state, "EDAC Error Reporting state: 0=Poll,1=NMI"); diff --git a/drivers/edac/e7xxx_edac.c b/drivers/edac/e7xxx_edac.c index 310d91b41c9..96ecc492664 100644 --- a/drivers/edac/e7xxx_edac.c +++ b/drivers/edac/e7xxx_edac.c @@ -27,9 +27,10 @@ #include <linux/pci.h> #include <linux/pci_ids.h> #include <linux/slab.h> -#include "edac_mc.h" +#include <linux/edac.h> +#include "edac_core.h" -#define E7XXX_REVISION " Ver: 2.0.1 " __DATE__ +#define E7XXX_REVISION " Ver: 2.0.2 " __DATE__ #define EDAC_MOD_STR "e7xxx_edac" #define e7xxx_printk(level, fmt, arg...) \ @@ -143,23 +144,21 @@ struct e7xxx_error_info { u32 dram_uelog_add; }; +static struct edac_pci_ctl_info *e7xxx_pci; + static const struct e7xxx_dev_info e7xxx_devs[] = { [E7500] = { .err_dev = PCI_DEVICE_ID_INTEL_7500_1_ERR, - .ctl_name = "E7500" - }, + .ctl_name = "E7500"}, [E7501] = { .err_dev = PCI_DEVICE_ID_INTEL_7501_1_ERR, - .ctl_name = "E7501" - }, + .ctl_name = "E7501"}, [E7505] = { .err_dev = PCI_DEVICE_ID_INTEL_7505_1_ERR, - .ctl_name = "E7505" - }, + .ctl_name = "E7505"}, [E7205] = { .err_dev = PCI_DEVICE_ID_INTEL_7205_1_ERR, - .ctl_name = "E7205" - }, + .ctl_name = "E7205"}, }; /* FIXME - is this valid for both SECDED and S4ECD4ED? */ @@ -180,15 +179,15 @@ static inline int e7xxx_find_channel(u16 syndrome) } static unsigned long ctl_page_to_phys(struct mem_ctl_info *mci, - unsigned long page) + unsigned long page) { u32 remap; - struct e7xxx_pvt *pvt = (struct e7xxx_pvt *) mci->pvt_info; + struct e7xxx_pvt *pvt = (struct e7xxx_pvt *)mci->pvt_info; debugf3("%s()\n", __func__); if ((page < pvt->tolm) || - ((page >= 0x100000) && (page < pvt->remapbase))) + ((page >= 0x100000) && (page < pvt->remapbase))) return page; remap = (page - pvt->tolm) + pvt->remapbase; @@ -200,8 +199,7 @@ static unsigned long ctl_page_to_phys(struct mem_ctl_info *mci, return pvt->tolm - 1; } -static void process_ce(struct mem_ctl_info *mci, - struct e7xxx_error_info *info) +static void process_ce(struct mem_ctl_info *mci, struct e7xxx_error_info *info) { u32 error_1b, page; u16 syndrome; @@ -212,7 +210,7 @@ static void process_ce(struct mem_ctl_info *mci, /* read the error address */ error_1b = info->dram_celog_add; /* FIXME - should use PAGE_SHIFT */ - page = error_1b >> 6; /* convert the address to 4k page */ + page = error_1b >> 6; /* convert the address to 4k page */ /* read the syndrome */ syndrome = info->dram_celog_syndrome; /* FIXME - check for -1 */ @@ -228,8 +226,7 @@ static void process_ce_no_info(struct mem_ctl_info *mci) edac_mc_handle_ce_no_info(mci, "e7xxx CE log register overflow"); } -static void process_ue(struct mem_ctl_info *mci, - struct e7xxx_error_info *info) +static void process_ue(struct mem_ctl_info *mci, struct e7xxx_error_info *info) { u32 error_2b, block_page; int row; @@ -238,7 +235,7 @@ static void process_ue(struct mem_ctl_info *mci, /* read the error address */ error_2b = info->dram_uelog_add; /* FIXME - should use PAGE_SHIFT */ - block_page = error_2b >> 6; /* convert to 4k address */ + block_page = error_2b >> 6; /* convert to 4k address */ row = edac_mc_find_csrow_by_page(mci, block_page); edac_mc_handle_ue(mci, block_page, 0, row, "e7xxx UE"); } @@ -249,16 +246,14 @@ static void process_ue_no_info(struct mem_ctl_info *mci) edac_mc_handle_ue_no_info(mci, "e7xxx UE log register overflow"); } -static void e7xxx_get_error_info (struct mem_ctl_info *mci, - struct e7xxx_error_info *info) +static void e7xxx_get_error_info(struct mem_ctl_info *mci, + struct e7xxx_error_info *info) { struct e7xxx_pvt *pvt; - pvt = (struct e7xxx_pvt *) mci->pvt_info; - pci_read_config_byte(pvt->bridge_ck, E7XXX_DRAM_FERR, - &info->dram_ferr); - pci_read_config_byte(pvt->bridge_ck, E7XXX_DRAM_NERR, - &info->dram_nerr); + pvt = (struct e7xxx_pvt *)mci->pvt_info; + pci_read_config_byte(pvt->bridge_ck, E7XXX_DRAM_FERR, &info->dram_ferr); + pci_read_config_byte(pvt->bridge_ck, E7XXX_DRAM_NERR, &info->dram_nerr); if ((info->dram_ferr & 1) || (info->dram_nerr & 1)) { pci_read_config_dword(pvt->bridge_ck, E7XXX_DRAM_CELOG_ADD, @@ -279,8 +274,9 @@ static void e7xxx_get_error_info (struct mem_ctl_info *mci, pci_write_bits8(pvt->bridge_ck, E7XXX_DRAM_NERR, 0x03, 0x03); } -static int e7xxx_process_error_info (struct mem_ctl_info *mci, - struct e7xxx_error_info *info, int handle_errors) +static int e7xxx_process_error_info(struct mem_ctl_info *mci, + struct e7xxx_error_info *info, + int handle_errors) { int error_found; @@ -341,7 +337,6 @@ static inline int dual_channel_active(u32 drc, int dev_idx) return (dev_idx == E7501) ? ((drc >> 22) & 0x1) : 1; } - /* Return DRB granularity (0=32mb, 1=64mb). */ static inline int drb_granularity(u32 drc, int dev_idx) { @@ -349,9 +344,8 @@ static inline int drb_granularity(u32 drc, int dev_idx) return (dev_idx == E7501) ? ((drc >> 18) & 0x3) : 1; } - static void e7xxx_init_csrows(struct mem_ctl_info *mci, struct pci_dev *pdev, - int dev_idx, u32 drc) + int dev_idx, u32 drc) { unsigned long last_cumul_size; int index; @@ -419,10 +413,21 @@ static int e7xxx_probe1(struct pci_dev *pdev, int dev_idx) struct e7xxx_error_info discard; debugf0("%s(): mci\n", __func__); + + /* make sure error reporting method is sane */ + switch (edac_op_state) { + case EDAC_OPSTATE_POLL: + case EDAC_OPSTATE_NMI: + break; + default: + edac_op_state = EDAC_OPSTATE_POLL; + break; + } + pci_read_config_dword(pdev, E7XXX_DRC, &drc); drc_chan = dual_channel_active(drc, dev_idx); - mci = edac_mc_alloc(sizeof(*pvt), E7XXX_NR_CSROWS, drc_chan + 1); + mci = edac_mc_alloc(sizeof(*pvt), E7XXX_NR_CSROWS, drc_chan + 1, 0); if (mci == NULL) return -ENOMEM; @@ -430,17 +435,16 @@ static int e7xxx_probe1(struct pci_dev *pdev, int dev_idx) debugf3("%s(): init mci\n", __func__); mci->mtype_cap = MEM_FLAG_RDDR; mci->edac_ctl_cap = EDAC_FLAG_NONE | EDAC_FLAG_SECDED | - EDAC_FLAG_S4ECD4ED; + EDAC_FLAG_S4ECD4ED; /* FIXME - what if different memory types are in different csrows? */ mci->mod_name = EDAC_MOD_STR; mci->mod_ver = E7XXX_REVISION; mci->dev = &pdev->dev; debugf3("%s(): init pvt\n", __func__); - pvt = (struct e7xxx_pvt *) mci->pvt_info; + pvt = (struct e7xxx_pvt *)mci->pvt_info; pvt->dev_info = &e7xxx_devs[dev_idx]; pvt->bridge_ck = pci_get_device(PCI_VENDOR_ID_INTEL, - pvt->dev_info->err_dev, - pvt->bridge_ck); + pvt->dev_info->err_dev, pvt->bridge_ck); if (!pvt->bridge_ck) { e7xxx_printk(KERN_ERR, "error reporting device not found:" @@ -451,6 +455,7 @@ static int e7xxx_probe1(struct pci_dev *pdev, int dev_idx) debugf3("%s(): more mci init\n", __func__); mci->ctl_name = pvt->dev_info->ctl_name; + mci->dev_name = pci_name(pdev); mci->edac_check = e7xxx_check; mci->ctl_page_to_phys = ctl_page_to_phys; e7xxx_init_csrows(mci, pdev, dev_idx, drc); @@ -473,11 +478,22 @@ static int e7xxx_probe1(struct pci_dev *pdev, int dev_idx) /* Here we assume that we will never see multiple instances of this * type of memory controller. The ID is therefore hardcoded to 0. */ - if (edac_mc_add_mc(mci,0)) { + if (edac_mc_add_mc(mci)) { debugf3("%s(): failed edac_mc_add_mc()\n", __func__); goto fail1; } + /* allocating generic PCI control info */ + e7xxx_pci = edac_pci_create_generic_ctl(&pdev->dev, EDAC_MOD_STR); + if (!e7xxx_pci) { + printk(KERN_WARNING + "%s(): Unable to create PCI control\n", + __func__); + printk(KERN_WARNING + "%s(): PCI error report via EDAC not setup\n", + __func__); + } + /* get this far and it's successful */ debugf3("%s(): success\n", __func__); return 0; @@ -493,7 +509,7 @@ fail0: /* returns count (>= 0), or negative on error */ static int __devinit e7xxx_init_one(struct pci_dev *pdev, - const struct pci_device_id *ent) + const struct pci_device_id *ent) { debugf0("%s()\n", __func__); @@ -509,34 +525,33 @@ static void __devexit e7xxx_remove_one(struct pci_dev *pdev) debugf0("%s()\n", __func__); + if (e7xxx_pci) + edac_pci_release_generic_ctl(e7xxx_pci); + if ((mci = edac_mc_del_mc(&pdev->dev)) == NULL) return; - pvt = (struct e7xxx_pvt *) mci->pvt_info; + pvt = (struct e7xxx_pvt *)mci->pvt_info; pci_dev_put(pvt->bridge_ck); edac_mc_free(mci); } static const struct pci_device_id e7xxx_pci_tbl[] __devinitdata = { { - PCI_VEND_DEV(INTEL, 7205_0), PCI_ANY_ID, PCI_ANY_ID, 0, 0, - E7205 - }, + PCI_VEND_DEV(INTEL, 7205_0), PCI_ANY_ID, PCI_ANY_ID, 0, 0, + E7205}, { - PCI_VEND_DEV(INTEL, 7500_0), PCI_ANY_ID, PCI_ANY_ID, 0, 0, - E7500 - }, + PCI_VEND_DEV(INTEL, 7500_0), PCI_ANY_ID, PCI_ANY_ID, 0, 0, + E7500}, { - PCI_VEND_DEV(INTEL, 7501_0), PCI_ANY_ID, PCI_ANY_ID, 0, 0, - E7501 - }, + PCI_VEND_DEV(INTEL, 7501_0), PCI_ANY_ID, PCI_ANY_ID, 0, 0, + E7501}, { - PCI_VEND_DEV(INTEL, 7505_0), PCI_ANY_ID, PCI_ANY_ID, 0, 0, - E7505 - }, + PCI_VEND_DEV(INTEL, 7505_0), PCI_ANY_ID, PCI_ANY_ID, 0, 0, + E7505}, { - 0, - } /* 0 terminated list. */ + 0, + } /* 0 terminated list. */ }; MODULE_DEVICE_TABLE(pci, e7xxx_pci_tbl); @@ -563,5 +578,7 @@ module_exit(e7xxx_exit); MODULE_LICENSE("GPL"); MODULE_AUTHOR("Linux Networx (http://lnxi.com) Thayne Harbaugh et al\n" - "Based on.work by Dan Hollis et al"); + "Based on.work by Dan Hollis et al"); MODULE_DESCRIPTION("MC support for Intel e7xxx memory controllers"); +module_param(edac_op_state, int, 0444); +MODULE_PARM_DESC(edac_op_state, "EDAC Error Reporting state: 0=Poll,1=NMI"); diff --git a/drivers/edac/edac_mc.h b/drivers/edac/edac_core.h index 713444cc410..4e6bad15c4b 100644 --- a/drivers/edac/edac_mc.h +++ b/drivers/edac/edac_core.h @@ -1,6 +1,7 @@ /* - * MC kernel module - * (C) 2003 Linux Networx (http://lnxi.com) + * Defines, structures, APIs for edac_core module + * + * (C) 2007 Linux Networx (http://lnxi.com) * This file may be distributed under the terms of the * GNU General Public License. * @@ -11,12 +12,13 @@ * NMI handling support added by * Dave Peterson <dsp@llnl.gov> <dave_peterson@pobox.com> * - * $Id: edac_mc.h,v 1.4.2.10 2005/10/05 00:43:44 dsp_llnl Exp $ + * Refactored for multi-source files: + * Doug Thompson <norsk5@xmission.com> * */ -#ifndef _EDAC_MC_H_ -#define _EDAC_MC_H_ +#ifndef _EDAC_CORE_H_ +#define _EDAC_CORE_H_ #include <linux/kernel.h> #include <linux/types.h> @@ -30,9 +32,14 @@ #include <linux/completion.h> #include <linux/kobject.h> #include <linux/platform_device.h> +#include <linux/sysdev.h> +#include <linux/workqueue.h> +#include <linux/version.h> #define EDAC_MC_LABEL_LEN 31 -#define MC_PROC_NAME_MAX_LEN 7 +#define EDAC_DEVICE_NAME_LEN 31 +#define EDAC_ATTRIB_VALUE_LEN 15 +#define MC_PROC_NAME_MAX_LEN 7 #if PAGE_SHIFT < 20 #define PAGES_TO_MiB( pages ) ( ( pages ) >> ( 20 - PAGE_SHIFT ) ) @@ -49,6 +56,14 @@ #define edac_mc_chipset_printk(mci, level, prefix, fmt, arg...) \ printk(level "EDAC " prefix " MC%d: " fmt, mci->mc_idx, ##arg) +/* edac_device printk */ +#define edac_device_printk(ctl, level, fmt, arg...) \ + printk(level "EDAC DEVICE%d: " fmt, ctl->dev_idx, ##arg) + +/* edac_pci printk */ +#define edac_pci_printk(ctl, level, fmt, arg...) \ + printk(level "EDAC PCI%d: " fmt, ctl->pci_idx, ##arg) + /* prefixes for edac_printk() and edac_mc_printk() */ #define EDAC_MC "MC" #define EDAC_PCI "PCI" @@ -60,7 +75,7 @@ extern int edac_debug_level; #define edac_debug_printk(level, fmt, arg...) \ do { \ if (level <= edac_debug_level) \ - edac_printk(KERN_DEBUG, EDAC_DEBUG, fmt, ##arg); \ + edac_printk(KERN_EMERG, EDAC_DEBUG, fmt, ##arg); \ } while(0) #define debugf0( ... ) edac_debug_printk(0, __VA_ARGS__ ) @@ -69,7 +84,7 @@ extern int edac_debug_level; #define debugf3( ... ) edac_debug_printk(3, __VA_ARGS__ ) #define debugf4( ... ) edac_debug_printk(4, __VA_ARGS__ ) -#else /* !CONFIG_EDAC_DEBUG */ +#else /* !CONFIG_EDAC_DEBUG */ #define debugf0( ... ) #define debugf1( ... ) @@ -77,18 +92,14 @@ extern int edac_debug_level; #define debugf3( ... ) #define debugf4( ... ) -#endif /* !CONFIG_EDAC_DEBUG */ +#endif /* !CONFIG_EDAC_DEBUG */ #define BIT(x) (1 << (x)) #define PCI_VEND_DEV(vend, dev) PCI_VENDOR_ID_ ## vend, \ PCI_DEVICE_ID_ ## vend ## _ ## dev -#if defined(CONFIG_X86) && defined(CONFIG_PCI) -#define dev_name(dev) pci_name(to_pci_dev(dev)) -#else -#define dev_name(dev) to_platform_device(dev)->name -#endif +#define dev_name(dev) (dev)->dev_name /* memory devices */ enum dev_type { @@ -124,8 +135,9 @@ enum mem_type { MEM_DDR, /* Double data rate SDRAM */ MEM_RDDR, /* Registered Double data rate SDRAM */ MEM_RMBS, /* Rambus DRAM */ - MEM_DDR2, /* DDR2 RAM */ - MEM_FB_DDR2, /* fully buffered DDR2 */ + MEM_DDR2, /* DDR2 RAM */ + MEM_FB_DDR2, /* fully buffered DDR2 */ + MEM_RDDR2, /* Registered DDR2 RAM */ }; #define MEM_FLAG_EMPTY BIT(MEM_EMPTY) @@ -141,6 +153,7 @@ enum mem_type { #define MEM_FLAG_RMBS BIT(MEM_RMBS) #define MEM_FLAG_DDR2 BIT(MEM_DDR2) #define MEM_FLAG_FB_DDR2 BIT(MEM_FB_DDR2) +#define MEM_FLAG_RDDR2 BIT(MEM_RDDR2) /* chipset Error Detection and Correction capabilities and mode */ enum edac_type { @@ -181,16 +194,23 @@ enum scrub_type { }; #define SCRUB_FLAG_SW_PROG BIT(SCRUB_SW_PROG) -#define SCRUB_FLAG_SW_SRC BIT(SCRUB_SW_SRC_CORR) -#define SCRUB_FLAG_SW_PROG_SRC BIT(SCRUB_SW_PROG_SRC_CORR) +#define SCRUB_FLAG_SW_SRC BIT(SCRUB_SW_SRC) +#define SCRUB_FLAG_SW_PROG_SRC BIT(SCRUB_SW_PROG_SRC) #define SCRUB_FLAG_SW_TUN BIT(SCRUB_SW_SCRUB_TUNABLE) #define SCRUB_FLAG_HW_PROG BIT(SCRUB_HW_PROG) -#define SCRUB_FLAG_HW_SRC BIT(SCRUB_HW_SRC_CORR) -#define SCRUB_FLAG_HW_PROG_SRC BIT(SCRUB_HW_PROG_SRC_CORR) +#define SCRUB_FLAG_HW_SRC BIT(SCRUB_HW_SRC) +#define SCRUB_FLAG_HW_PROG_SRC BIT(SCRUB_HW_PROG_SRC) #define SCRUB_FLAG_HW_TUN BIT(SCRUB_HW_TUNABLE) /* FIXME - should have notify capabilities: NMI, LOG, PROC, etc */ +/* EDAC internal operation states */ +#define OP_ALLOC 0x100 +#define OP_RUNNING_POLL 0x201 +#define OP_RUNNING_INTERRUPT 0x202 +#define OP_RUNNING_POLL_INTR 0x203 +#define OP_OFFLINE 0x300 + /* * There are several things to be aware of that aren't at all obvious: * @@ -276,7 +296,7 @@ enum scrub_type { struct channel_info { int chan_idx; /* channel index */ u32 ce_count; /* Correctable Errors for this CHANNEL */ - char label[EDAC_MC_LABEL_LEN + 1]; /* DIMM label on motherboard */ + char label[EDAC_MC_LABEL_LEN + 1]; /* DIMM label on motherboard */ struct csrow_info *csrow; /* the parent */ }; @@ -297,15 +317,29 @@ struct csrow_info { struct mem_ctl_info *mci; /* the parent */ struct kobject kobj; /* sysfs kobject for this csrow */ - struct completion kobj_complete; - /* FIXME the number of CHANNELs might need to become dynamic */ + /* channel information for this csrow */ u32 nr_channels; struct channel_info *channels; }; +/* mcidev_sysfs_attribute structure + * used for driver sysfs attributes and in mem_ctl_info + * sysfs top level entries + */ +struct mcidev_sysfs_attribute { + struct attribute attr; + ssize_t (*show)(struct mem_ctl_info *,char *); + ssize_t (*store)(struct mem_ctl_info *, const char *,size_t); +}; + +/* MEMORY controller information structure + */ struct mem_ctl_info { - struct list_head link; /* for global list of mem_ctl_info structs */ + struct list_head link; /* for global list of mem_ctl_info structs */ + + struct module *owner; /* Module owner of this control struct */ + unsigned long mtype_cap; /* memory types supported by mc */ unsigned long edac_ctl_cap; /* Mem controller EDAC capabilities */ unsigned long edac_cap; /* configuration capabilities - this is @@ -322,14 +356,15 @@ struct mem_ctl_info { /* Translates sdram memory scrub rate given in bytes/sec to the internal representation and configures whatever else needs to be configured. - */ - int (*set_sdram_scrub_rate) (struct mem_ctl_info *mci, u32 *bw); + */ + int (*set_sdram_scrub_rate) (struct mem_ctl_info * mci, u32 * bw); /* Get the current sdram memory scrub rate from the internal representation and converts it to the closest matching bandwith in bytes/sec. - */ - int (*get_sdram_scrub_rate) (struct mem_ctl_info *mci, u32 *bw); + */ + int (*get_sdram_scrub_rate) (struct mem_ctl_info * mci, u32 * bw); + /* pointer to edac checking routine */ void (*edac_check) (struct mem_ctl_info * mci); @@ -340,7 +375,7 @@ struct mem_ctl_info { */ /* FIXME - why not send the phys page to begin with? */ unsigned long (*ctl_page_to_phys) (struct mem_ctl_info * mci, - unsigned long page); + unsigned long page); int mc_idx; int nr_csrows; struct csrow_info *csrows; @@ -353,6 +388,7 @@ struct mem_ctl_info { const char *mod_name; const char *mod_ver; const char *ctl_name; + const char *dev_name; char proc_name[MC_PROC_NAME_MAX_LEN + 1]; void *pvt_info; u32 ue_noinfo_count; /* Uncorrectable Errors w/o info */ @@ -369,14 +405,327 @@ struct mem_ctl_info { /* edac sysfs device control */ struct kobject edac_mci_kobj; - struct completion kobj_complete; + + /* Additional top controller level attributes, but specified + * by the low level driver. + * + * Set by the low level driver to provide attributes at the + * controller level, same level as 'ue_count' and 'ce_count' above. + * An array of structures, NULL terminated + * + * If attributes are desired, then set to array of attributes + * If no attributes are desired, leave NULL + */ + struct mcidev_sysfs_attribute *mc_driver_sysfs_attributes; + + /* work struct for this MC */ + struct delayed_work work; + + /* the internal state of this controller instance */ + int op_state; +}; + +/* + * The following are the structures to provide for a generic + * or abstract 'edac_device'. This set of structures and the + * code that implements the APIs for the same, provide for + * registering EDAC type devices which are NOT standard memory. + * + * CPU caches (L1 and L2) + * DMA engines + * Core CPU swithces + * Fabric switch units + * PCIe interface controllers + * other EDAC/ECC type devices that can be monitored for + * errors, etc. + * + * It allows for a 2 level set of hiearchry. For example: + * + * cache could be composed of L1, L2 and L3 levels of cache. + * Each CPU core would have its own L1 cache, while sharing + * L2 and maybe L3 caches. + * + * View them arranged, via the sysfs presentation: + * /sys/devices/system/edac/.. + * + * mc/ <existing memory device directory> + * cpu/cpu0/.. <L1 and L2 block directory> + * /L1-cache/ce_count + * /ue_count + * /L2-cache/ce_count + * /ue_count + * cpu/cpu1/.. <L1 and L2 block directory> + * /L1-cache/ce_count + * /ue_count + * /L2-cache/ce_count + * /ue_count + * ... + * + * the L1 and L2 directories would be "edac_device_block's" + */ + +struct edac_device_counter { + u32 ue_count; + u32 ce_count; +}; + +/* forward reference */ +struct edac_device_ctl_info; +struct edac_device_block; + +/* edac_dev_sysfs_attribute structure + * used for driver sysfs attributes in mem_ctl_info + * for extra controls and attributes: + * like high level error Injection controls + */ +struct edac_dev_sysfs_attribute { + struct attribute attr; + ssize_t (*show)(struct edac_device_ctl_info *, char *); + ssize_t (*store)(struct edac_device_ctl_info *, const char *, size_t); +}; + +/* edac_dev_sysfs_block_attribute structure + * + * used in leaf 'block' nodes for adding controls/attributes + * + * each block in each instance of the containing control structure + * can have an array of the following. The show and store functions + * will be filled in with the show/store function in the + * low level driver. + * + * The 'value' field will be the actual value field used for + * counting + */ +struct edac_dev_sysfs_block_attribute { + struct attribute attr; + ssize_t (*show)(struct kobject *, struct attribute *, char *); + ssize_t (*store)(struct kobject *, struct attribute *, + const char *, size_t); + struct edac_device_block *block; + + unsigned int value; +}; + +/* device block control structure */ +struct edac_device_block { + struct edac_device_instance *instance; /* Up Pointer */ + char name[EDAC_DEVICE_NAME_LEN + 1]; + + struct edac_device_counter counters; /* basic UE and CE counters */ + + int nr_attribs; /* how many attributes */ + + /* this block's attributes, could be NULL */ + struct edac_dev_sysfs_block_attribute *block_attributes; + + /* edac sysfs device control */ + struct kobject kobj; +}; + +/* device instance control structure */ +struct edac_device_instance { + struct edac_device_ctl_info *ctl; /* Up pointer */ + char name[EDAC_DEVICE_NAME_LEN + 4]; + + struct edac_device_counter counters; /* instance counters */ + + u32 nr_blocks; /* how many blocks */ + struct edac_device_block *blocks; /* block array */ + + /* edac sysfs device control */ + struct kobject kobj; +}; + + +/* + * Abstract edac_device control info structure + * + */ +struct edac_device_ctl_info { + /* for global list of edac_device_ctl_info structs */ + struct list_head link; + + struct module *owner; /* Module owner of this control struct */ + + int dev_idx; + + /* Per instance controls for this edac_device */ + int log_ue; /* boolean for logging UEs */ + int log_ce; /* boolean for logging CEs */ + int panic_on_ue; /* boolean for panic'ing on an UE */ + unsigned poll_msec; /* number of milliseconds to poll interval */ + unsigned long delay; /* number of jiffies for poll_msec */ + + /* Additional top controller level attributes, but specified + * by the low level driver. + * + * Set by the low level driver to provide attributes at the + * controller level, same level as 'ue_count' and 'ce_count' above. + * An array of structures, NULL terminated + * + * If attributes are desired, then set to array of attributes + * If no attributes are desired, leave NULL + */ + struct edac_dev_sysfs_attribute *sysfs_attributes; + + /* pointer to main 'edac' class in sysfs */ + struct sysdev_class *edac_class; + + /* the internal state of this controller instance */ + int op_state; + /* work struct for this instance */ + struct delayed_work work; + + /* pointer to edac polling checking routine: + * If NOT NULL: points to polling check routine + * If NULL: Then assumes INTERRUPT operation, where + * MC driver will receive events + */ + void (*edac_check) (struct edac_device_ctl_info * edac_dev); + + struct device *dev; /* pointer to device structure */ + + const char *mod_name; /* module name */ + const char *ctl_name; /* edac controller name */ + const char *dev_name; /* pci/platform/etc... name */ + + void *pvt_info; /* pointer to 'private driver' info */ + + unsigned long start_time; /* edac_device load start time (jiffies) */ + + /* these are for safe removal of mc devices from global list while + * NMI handlers may be traversing list + */ + struct rcu_head rcu; + struct completion removal_complete; + + /* sysfs top name under 'edac' directory + * and instance name: + * cpu/cpu0/... + * cpu/cpu1/... + * cpu/cpu2/... + * ... + */ + char name[EDAC_DEVICE_NAME_LEN + 1]; + + /* Number of instances supported on this control structure + * and the array of those instances + */ + u32 nr_instances; + struct edac_device_instance *instances; + + /* Event counters for the this whole EDAC Device */ + struct edac_device_counter counters; + + /* edac sysfs device control for the 'name' + * device this structure controls + */ + struct kobject kobj; }; +/* To get from the instance's wq to the beginning of the ctl structure */ +#define to_edac_mem_ctl_work(w) \ + container_of(w, struct mem_ctl_info, work) + +#define to_edac_device_ctl_work(w) \ + container_of(w,struct edac_device_ctl_info,work) + +/* + * The alloc() and free() functions for the 'edac_device' control info + * structure. A MC driver will allocate one of these for each edac_device + * it is going to control/register with the EDAC CORE. + */ +extern struct edac_device_ctl_info *edac_device_alloc_ctl_info( + unsigned sizeof_private, + char *edac_device_name, unsigned nr_instances, + char *edac_block_name, unsigned nr_blocks, + unsigned offset_value, + struct edac_dev_sysfs_block_attribute *block_attributes, + unsigned nr_attribs, + int device_index); + +/* The offset value can be: + * -1 indicating no offset value + * 0 for zero-based block numbers + * 1 for 1-based block number + * other for other-based block number + */ +#define BLOCK_OFFSET_VALUE_OFF ((unsigned) -1) + +extern void edac_device_free_ctl_info(struct edac_device_ctl_info *ctl_info); + #ifdef CONFIG_PCI +struct edac_pci_counter { + atomic_t pe_count; + atomic_t npe_count; +}; + +/* + * Abstract edac_pci control info structure + * + */ +struct edac_pci_ctl_info { + /* for global list of edac_pci_ctl_info structs */ + struct list_head link; + + int pci_idx; + + struct sysdev_class *edac_class; /* pointer to class */ + + /* the internal state of this controller instance */ + int op_state; + /* work struct for this instance */ + struct delayed_work work; + + /* pointer to edac polling checking routine: + * If NOT NULL: points to polling check routine + * If NULL: Then assumes INTERRUPT operation, where + * MC driver will receive events + */ + void (*edac_check) (struct edac_pci_ctl_info * edac_dev); + + struct device *dev; /* pointer to device structure */ + + const char *mod_name; /* module name */ + const char *ctl_name; /* edac controller name */ + const char *dev_name; /* pci/platform/etc... name */ + + void *pvt_info; /* pointer to 'private driver' info */ + + unsigned long start_time; /* edac_pci load start time (jiffies) */ + + /* these are for safe removal of devices from global list while + * NMI handlers may be traversing list + */ + struct rcu_head rcu; + struct completion complete; + + /* sysfs top name under 'edac' directory + * and instance name: + * cpu/cpu0/... + * cpu/cpu1/... + * cpu/cpu2/... + * ... + */ + char name[EDAC_DEVICE_NAME_LEN + 1]; + + /* Event counters for the this whole EDAC Device */ + struct edac_pci_counter counters; + + /* edac sysfs device control for the 'name' + * device this structure controls + */ + struct kobject kobj; + struct completion kobj_complete; +}; + +#define to_edac_pci_ctl_work(w) \ + container_of(w, struct edac_pci_ctl_info,work) + /* write all or some bits in a byte-register*/ static inline void pci_write_bits8(struct pci_dev *pdev, int offset, u8 value, - u8 mask) + u8 mask) { if (mask != 0xff) { u8 buf; @@ -392,7 +741,7 @@ static inline void pci_write_bits8(struct pci_dev *pdev, int offset, u8 value, /* write all or some bits in a word-register*/ static inline void pci_write_bits16(struct pci_dev *pdev, int offset, - u16 value, u16 mask) + u16 value, u16 mask) { if (mask != 0xffff) { u16 buf; @@ -408,7 +757,7 @@ static inline void pci_write_bits16(struct pci_dev *pdev, int offset, /* write all or some bits in a dword-register*/ static inline void pci_write_bits32(struct pci_dev *pdev, int offset, - u32 value, u32 mask) + u32 value, u32 mask) { if (mask != 0xffff) { u32 buf; @@ -422,20 +771,16 @@ static inline void pci_write_bits32(struct pci_dev *pdev, int offset, pci_write_config_dword(pdev, offset, value); } -#endif /* CONFIG_PCI */ +#endif /* CONFIG_PCI */ -#ifdef CONFIG_EDAC_DEBUG -void edac_mc_dump_channel(struct channel_info *chan); -void edac_mc_dump_mci(struct mem_ctl_info *mci); -void edac_mc_dump_csrow(struct csrow_info *csrow); -#endif /* CONFIG_EDAC_DEBUG */ - -extern int edac_mc_add_mc(struct mem_ctl_info *mci,int mc_idx); -extern struct mem_ctl_info * edac_mc_del_mc(struct device *dev); +extern struct mem_ctl_info *edac_mc_alloc(unsigned sz_pvt, unsigned nr_csrows, + unsigned nr_chans, int edac_index); +extern int edac_mc_add_mc(struct mem_ctl_info *mci); +extern void edac_mc_free(struct mem_ctl_info *mci); +extern struct mem_ctl_info *edac_mc_find(int idx); +extern struct mem_ctl_info *edac_mc_del_mc(struct device *dev); extern int edac_mc_find_csrow_by_page(struct mem_ctl_info *mci, - unsigned long page); -extern void edac_mc_scrub_block(unsigned long page, unsigned long offset, - u32 size); + unsigned long page); /* * The no info errors are used when error overflows are reported. @@ -448,34 +793,59 @@ extern void edac_mc_scrub_block(unsigned long page, unsigned long offset, * statement clutter and extra function arguments. */ extern void edac_mc_handle_ce(struct mem_ctl_info *mci, - unsigned long page_frame_number, unsigned long offset_in_page, - unsigned long syndrome, int row, int channel, - const char *msg); + unsigned long page_frame_number, + unsigned long offset_in_page, + unsigned long syndrome, int row, int channel, + const char *msg); extern void edac_mc_handle_ce_no_info(struct mem_ctl_info *mci, - const char *msg); + const char *msg); extern void edac_mc_handle_ue(struct mem_ctl_info *mci, - unsigned long page_frame_number, unsigned long offset_in_page, - int row, const char *msg); + unsigned long page_frame_number, + unsigned long offset_in_page, int row, + const char *msg); extern void edac_mc_handle_ue_no_info(struct mem_ctl_info *mci, - const char *msg); -extern void edac_mc_handle_fbd_ue(struct mem_ctl_info *mci, - unsigned int csrow, - unsigned int channel0, - unsigned int channel1, - char *msg); -extern void edac_mc_handle_fbd_ce(struct mem_ctl_info *mci, - unsigned int csrow, - unsigned int channel, - char *msg); + const char *msg); +extern void edac_mc_handle_fbd_ue(struct mem_ctl_info *mci, unsigned int csrow, + unsigned int channel0, unsigned int channel1, + char *msg); +extern void edac_mc_handle_fbd_ce(struct mem_ctl_info *mci, unsigned int csrow, + unsigned int channel, char *msg); /* - * This kmalloc's and initializes all the structures. - * Can't be used if all structures don't have the same lifetime. + * edac_device APIs */ -extern struct mem_ctl_info *edac_mc_alloc(unsigned sz_pvt, unsigned nr_csrows, - unsigned nr_chans); +extern int edac_device_add_device(struct edac_device_ctl_info *edac_dev); +extern struct edac_device_ctl_info *edac_device_del_device(struct device *dev); +extern void edac_device_handle_ue(struct edac_device_ctl_info *edac_dev, + int inst_nr, int block_nr, const char *msg); +extern void edac_device_handle_ce(struct edac_device_ctl_info *edac_dev, + int inst_nr, int block_nr, const char *msg); -/* Free an mc previously allocated by edac_mc_alloc() */ -extern void edac_mc_free(struct mem_ctl_info *mci); +/* + * edac_pci APIs + */ +extern struct edac_pci_ctl_info *edac_pci_alloc_ctl_info(unsigned int sz_pvt, + const char *edac_pci_name); + +extern void edac_pci_free_ctl_info(struct edac_pci_ctl_info *pci); + +extern void edac_pci_reset_delay_period(struct edac_pci_ctl_info *pci, + unsigned long value); + +extern int edac_pci_add_device(struct edac_pci_ctl_info *pci, int edac_idx); +extern struct edac_pci_ctl_info *edac_pci_del_device(struct device *dev); + +extern struct edac_pci_ctl_info *edac_pci_create_generic_ctl( + struct device *dev, + const char *mod_name); + +extern void edac_pci_release_generic_ctl(struct edac_pci_ctl_info *pci); +extern int edac_pci_create_sysfs(struct edac_pci_ctl_info *pci); +extern void edac_pci_remove_sysfs(struct edac_pci_ctl_info *pci); + +/* + * edac misc APIs + */ +extern char *edac_op_state_to_string(int op_state); -#endif /* _EDAC_MC_H_ */ +#endif /* _EDAC_CORE_H_ */ diff --git a/drivers/edac/edac_device.c b/drivers/edac/edac_device.c new file mode 100644 index 00000000000..f3690a697cf --- /dev/null +++ b/drivers/edac/edac_device.c @@ -0,0 +1,746 @@ + +/* + * edac_device.c + * (C) 2007 www.douglaskthompson.com + * + * This file may be distributed under the terms of the + * GNU General Public License. + * + * Written by Doug Thompson <norsk5@xmission.com> + * + * edac_device API implementation + * 19 Jan 2007 + */ + +#include <linux/module.h> +#include <linux/types.h> +#include <linux/smp.h> +#include <linux/init.h> +#include <linux/sysctl.h> +#include <linux/highmem.h> +#include <linux/timer.h> +#include <linux/slab.h> +#include <linux/jiffies.h> +#include <linux/spinlock.h> +#include <linux/list.h> +#include <linux/sysdev.h> +#include <linux/ctype.h> +#include <linux/workqueue.h> +#include <asm/uaccess.h> +#include <asm/page.h> + +#include "edac_core.h" +#include "edac_module.h" + +/* lock for the list: 'edac_device_list', manipulation of this list + * is protected by the 'device_ctls_mutex' lock + */ +static DEFINE_MUTEX(device_ctls_mutex); +static struct list_head edac_device_list = LIST_HEAD_INIT(edac_device_list); + +#ifdef CONFIG_EDAC_DEBUG +static void edac_device_dump_device(struct edac_device_ctl_info *edac_dev) +{ + debugf3("\tedac_dev = %p dev_idx=%d \n", edac_dev, edac_dev->dev_idx); + debugf4("\tedac_dev->edac_check = %p\n", edac_dev->edac_check); + debugf3("\tdev = %p\n", edac_dev->dev); + debugf3("\tmod_name:ctl_name = %s:%s\n", + edac_dev->mod_name, edac_dev->ctl_name); + debugf3("\tpvt_info = %p\n\n", edac_dev->pvt_info); +} +#endif /* CONFIG_EDAC_DEBUG */ + + +/* + * edac_device_alloc_ctl_info() + * Allocate a new edac device control info structure + * + * The control structure is allocated in complete chunk + * from the OS. It is in turn sub allocated to the + * various objects that compose the struture + * + * The structure has a 'nr_instance' array within itself. + * Each instance represents a major component + * Example: L1 cache and L2 cache are 2 instance components + * + * Within each instance is an array of 'nr_blocks' blockoffsets + */ +struct edac_device_ctl_info *edac_device_alloc_ctl_info( + unsigned sz_private, + char *edac_device_name, unsigned nr_instances, + char *edac_block_name, unsigned nr_blocks, + unsigned offset_value, /* zero, 1, or other based offset */ + struct edac_dev_sysfs_block_attribute *attrib_spec, unsigned nr_attrib, + int device_index) +{ + struct edac_device_ctl_info *dev_ctl; + struct edac_device_instance *dev_inst, *inst; + struct edac_device_block *dev_blk, *blk_p, *blk; + struct edac_dev_sysfs_block_attribute *dev_attrib, *attrib_p, *attrib; + unsigned total_size; + unsigned count; + unsigned instance, block, attr; + void *pvt; + int err; + + debugf4("%s() instances=%d blocks=%d\n", + __func__, nr_instances, nr_blocks); + + /* Calculate the size of memory we need to allocate AND + * determine the offsets of the various item arrays + * (instance,block,attrib) from the start of an allocated structure. + * We want the alignment of each item (instance,block,attrib) + * to be at least as stringent as what the compiler would + * provide if we could simply hardcode everything into a single struct. + */ + dev_ctl = (struct edac_device_ctl_info *)NULL; + + /* Calc the 'end' offset past end of ONE ctl_info structure + * which will become the start of the 'instance' array + */ + dev_inst = edac_align_ptr(&dev_ctl[1], sizeof(*dev_inst)); + + /* Calc the 'end' offset past the instance array within the ctl_info + * which will become the start of the block array + */ + dev_blk = edac_align_ptr(&dev_inst[nr_instances], sizeof(*dev_blk)); + + /* Calc the 'end' offset past the dev_blk array + * which will become the start of the attrib array, if any. + */ + count = nr_instances * nr_blocks; + dev_attrib = edac_align_ptr(&dev_blk[count], sizeof(*dev_attrib)); + + /* Check for case of when an attribute array is specified */ + if (nr_attrib > 0) { + /* calc how many nr_attrib we need */ + count *= nr_attrib; + + /* Calc the 'end' offset past the attributes array */ + pvt = edac_align_ptr(&dev_attrib[count], sz_private); + } else { + /* no attribute array specificed */ + pvt = edac_align_ptr(dev_attrib, sz_private); + } + + /* 'pvt' now points to where the private data area is. + * At this point 'pvt' (like dev_inst,dev_blk and dev_attrib) + * is baselined at ZERO + */ + total_size = ((unsigned long)pvt) + sz_private; + + /* Allocate the amount of memory for the set of control structures */ + dev_ctl = kzalloc(total_size, GFP_KERNEL); + if (dev_ctl == NULL) + return NULL; + + /* Adjust pointers so they point within the actual memory we + * just allocated rather than an imaginary chunk of memory + * located at address 0. + * 'dev_ctl' points to REAL memory, while the others are + * ZERO based and thus need to be adjusted to point within + * the allocated memory. + */ + dev_inst = (struct edac_device_instance *) + (((char *)dev_ctl) + ((unsigned long)dev_inst)); + dev_blk = (struct edac_device_block *) + (((char *)dev_ctl) + ((unsigned long)dev_blk)); + dev_attrib = (struct edac_dev_sysfs_block_attribute *) + (((char *)dev_ctl) + ((unsigned long)dev_attrib)); + pvt = sz_private ? (((char *)dev_ctl) + ((unsigned long)pvt)) : NULL; + + /* Begin storing the information into the control info structure */ + dev_ctl->dev_idx = device_index; + dev_ctl->nr_instances = nr_instances; + dev_ctl->instances = dev_inst; + dev_ctl->pvt_info = pvt; + + /* Name of this edac device */ + snprintf(dev_ctl->name,sizeof(dev_ctl->name),"%s",edac_device_name); + + debugf4("%s() edac_dev=%p next after end=%p\n", + __func__, dev_ctl, pvt + sz_private ); + + /* Initialize every Instance */ + for (instance = 0; instance < nr_instances; instance++) { + inst = &dev_inst[instance]; + inst->ctl = dev_ctl; + inst->nr_blocks = nr_blocks; + blk_p = &dev_blk[instance * nr_blocks]; + inst->blocks = blk_p; + + /* name of this instance */ + snprintf(inst->name, sizeof(inst->name), + "%s%u", edac_device_name, instance); + + /* Initialize every block in each instance */ + for (block = 0; block < nr_blocks; block++) { + blk = &blk_p[block]; + blk->instance = inst; + snprintf(blk->name, sizeof(blk->name), + "%s%d", edac_block_name, block+offset_value); + + debugf4("%s() instance=%d inst_p=%p block=#%d " + "block_p=%p name='%s'\n", + __func__, instance, inst, block, + blk, blk->name); + + /* if there are NO attributes OR no attribute pointer + * then continue on to next block iteration + */ + if ((nr_attrib == 0) || (attrib_spec == NULL)) + continue; + + /* setup the attribute array for this block */ + blk->nr_attribs = nr_attrib; + attrib_p = &dev_attrib[block*nr_instances*nr_attrib]; + blk->block_attributes = attrib_p; + + debugf4("%s() THIS BLOCK_ATTRIB=%p\n", + __func__, blk->block_attributes); + + /* Initialize every user specified attribute in this + * block with the data the caller passed in + * Each block gets its own copy of pointers, + * and its unique 'value' + */ + for (attr = 0; attr < nr_attrib; attr++) { + attrib = &attrib_p[attr]; + + /* populate the unique per attrib + * with the code pointers and info + */ + attrib->attr = attrib_spec[attr].attr; + attrib->show = attrib_spec[attr].show; + attrib->store = attrib_spec[attr].store; + + attrib->block = blk; /* up link */ + + debugf4("%s() alloc-attrib=%p attrib_name='%s' " + "attrib-spec=%p spec-name=%s\n", + __func__, attrib, attrib->attr.name, + &attrib_spec[attr], + attrib_spec[attr].attr.name + ); + } + } + } + + /* Mark this instance as merely ALLOCATED */ + dev_ctl->op_state = OP_ALLOC; + + /* + * Initialize the 'root' kobj for the edac_device controller + */ + err = edac_device_register_sysfs_main_kobj(dev_ctl); + if (err) { + kfree(dev_ctl); + return NULL; + } + + /* at this point, the root kobj is valid, and in order to + * 'free' the object, then the function: + * edac_device_unregister_sysfs_main_kobj() must be called + * which will perform kobj unregistration and the actual free + * will occur during the kobject callback operation + */ + + return dev_ctl; +} +EXPORT_SYMBOL_GPL(edac_device_alloc_ctl_info); + +/* + * edac_device_free_ctl_info() + * frees the memory allocated by the edac_device_alloc_ctl_info() + * function + */ +void edac_device_free_ctl_info(struct edac_device_ctl_info *ctl_info) +{ + edac_device_unregister_sysfs_main_kobj(ctl_info); +} +EXPORT_SYMBOL_GPL(edac_device_free_ctl_info); + +/* + * find_edac_device_by_dev + * scans the edac_device list for a specific 'struct device *' + * + * lock to be held prior to call: device_ctls_mutex + * + * Return: + * pointer to control structure managing 'dev' + * NULL if not found on list + */ +static struct edac_device_ctl_info *find_edac_device_by_dev(struct device *dev) +{ + struct edac_device_ctl_info *edac_dev; + struct list_head *item; + + debugf0("%s()\n", __func__); + + list_for_each(item, &edac_device_list) { + edac_dev = list_entry(item, struct edac_device_ctl_info, link); + + if (edac_dev->dev == dev) + return edac_dev; + } + + return NULL; +} + +/* + * add_edac_dev_to_global_list + * Before calling this function, caller must + * assign a unique value to edac_dev->dev_idx. + * + * lock to be held prior to call: device_ctls_mutex + * + * Return: + * 0 on success + * 1 on failure. + */ +static int add_edac_dev_to_global_list(struct edac_device_ctl_info *edac_dev) +{ + struct list_head *item, *insert_before; + struct edac_device_ctl_info *rover; + + insert_before = &edac_device_list; + + /* Determine if already on the list */ + rover = find_edac_device_by_dev(edac_dev->dev); + if (unlikely(rover != NULL)) + goto fail0; + + /* Insert in ascending order by 'dev_idx', so find position */ + list_for_each(item, &edac_device_list) { + rover = list_entry(item, struct edac_device_ctl_info, link); + + if (rover->dev_idx >= edac_dev->dev_idx) { + if (unlikely(rover->dev_idx == edac_dev->dev_idx)) + goto fail1; + + insert_before = item; + break; + } + } + + list_add_tail_rcu(&edac_dev->link, insert_before); + return 0; + +fail0: + edac_printk(KERN_WARNING, EDAC_MC, + "%s (%s) %s %s already assigned %d\n", + rover->dev->bus_id, dev_name(rover), + rover->mod_name, rover->ctl_name, rover->dev_idx); + return 1; + +fail1: + edac_printk(KERN_WARNING, EDAC_MC, + "bug in low-level driver: attempt to assign\n" + " duplicate dev_idx %d in %s()\n", rover->dev_idx, + __func__); + return 1; +} + +/* + * complete_edac_device_list_del + * + * callback function when reference count is zero + */ +static void complete_edac_device_list_del(struct rcu_head *head) +{ + struct edac_device_ctl_info *edac_dev; + + edac_dev = container_of(head, struct edac_device_ctl_info, rcu); + INIT_LIST_HEAD(&edac_dev->link); + complete(&edac_dev->removal_complete); +} + +/* + * del_edac_device_from_global_list + * + * remove the RCU, setup for a callback call, + * then wait for the callback to occur + */ +static void del_edac_device_from_global_list(struct edac_device_ctl_info + *edac_device) +{ + list_del_rcu(&edac_device->link); + + init_completion(&edac_device->removal_complete); + call_rcu(&edac_device->rcu, complete_edac_device_list_del); + wait_for_completion(&edac_device->removal_complete); +} + +/** + * edac_device_find + * Search for a edac_device_ctl_info structure whose index is 'idx'. + * + * If found, return a pointer to the structure. + * Else return NULL. + * + * Caller must hold device_ctls_mutex. + */ +struct edac_device_ctl_info *edac_device_find(int idx) +{ + struct list_head *item; + struct edac_device_ctl_info *edac_dev; + + /* Iterate over list, looking for exact match of ID */ + list_for_each(item, &edac_device_list) { + edac_dev = list_entry(item, struct edac_device_ctl_info, link); + + if (edac_dev->dev_idx >= idx) { + if (edac_dev->dev_idx == idx) + return edac_dev; + + /* not on list, so terminate early */ + break; + } + } + + return NULL; +} +EXPORT_SYMBOL_GPL(edac_device_find); + +/* + * edac_device_workq_function + * performs the operation scheduled by a workq request + * + * this workq is embedded within an edac_device_ctl_info + * structure, that needs to be polled for possible error events. + * + * This operation is to acquire the list mutex lock + * (thus preventing insertation or deletion) + * and then call the device's poll function IFF this device is + * running polled and there is a poll function defined. + */ +static void edac_device_workq_function(struct work_struct *work_req) +{ + struct delayed_work *d_work = (struct delayed_work *)work_req; + struct edac_device_ctl_info *edac_dev = to_edac_device_ctl_work(d_work); + + mutex_lock(&device_ctls_mutex); + + /* Only poll controllers that are running polled and have a check */ + if ((edac_dev->op_state == OP_RUNNING_POLL) && + (edac_dev->edac_check != NULL)) { + edac_dev->edac_check(edac_dev); + } + + mutex_unlock(&device_ctls_mutex); + + /* Reschedule the workq for the next time period to start again + * if the number of msec is for 1 sec, then adjust to the next + * whole one second to save timers fireing all over the period + * between integral seconds + */ + if (edac_dev->poll_msec == 1000) + queue_delayed_work(edac_workqueue, &edac_dev->work, + round_jiffies(edac_dev->delay)); + else + queue_delayed_work(edac_workqueue, &edac_dev->work, + edac_dev->delay); +} + +/* + * edac_device_workq_setup + * initialize a workq item for this edac_device instance + * passing in the new delay period in msec + */ +void edac_device_workq_setup(struct edac_device_ctl_info *edac_dev, + unsigned msec) +{ + debugf0("%s()\n", __func__); + + /* take the arg 'msec' and set it into the control structure + * to used in the time period calculation + * then calc the number of jiffies that represents + */ + edac_dev->poll_msec = msec; + edac_dev->delay = msecs_to_jiffies(msec); + + INIT_DELAYED_WORK(&edac_dev->work, edac_device_workq_function); + + /* optimize here for the 1 second case, which will be normal value, to + * fire ON the 1 second time event. This helps reduce all sorts of + * timers firing on sub-second basis, while they are happy + * to fire together on the 1 second exactly + */ + if (edac_dev->poll_msec == 1000) + queue_delayed_work(edac_workqueue, &edac_dev->work, + round_jiffies(edac_dev->delay)); + else + queue_delayed_work(edac_workqueue, &edac_dev->work, + edac_dev->delay); +} + +/* + * edac_device_workq_teardown + * stop the workq processing on this edac_dev + */ +void edac_device_workq_teardown(struct edac_device_ctl_info *edac_dev) +{ + int status; + + status = cancel_delayed_work(&edac_dev->work); + if (status == 0) { + /* workq instance might be running, wait for it */ + flush_workqueue(edac_workqueue); + } +} + +/* + * edac_device_reset_delay_period + * + * need to stop any outstanding workq queued up at this time + * because we will be resetting the sleep time. + * Then restart the workq on the new delay + */ +void edac_device_reset_delay_period(struct edac_device_ctl_info *edac_dev, + unsigned long value) +{ + /* cancel the current workq request, without the mutex lock */ + edac_device_workq_teardown(edac_dev); + + /* acquire the mutex before doing the workq setup */ + mutex_lock(&device_ctls_mutex); + + /* restart the workq request, with new delay value */ + edac_device_workq_setup(edac_dev, value); + + mutex_unlock(&device_ctls_mutex); +} + +/** + * edac_device_add_device: Insert the 'edac_dev' structure into the + * edac_device global list and create sysfs entries associated with + * edac_device structure. + * @edac_device: pointer to the edac_device structure to be added to the list + * 'edac_device' structure. + * + * Return: + * 0 Success + * !0 Failure + */ +int edac_device_add_device(struct edac_device_ctl_info *edac_dev) +{ + debugf0("%s()\n", __func__); + +#ifdef CONFIG_EDAC_DEBUG + if (edac_debug_level >= 3) + edac_device_dump_device(edac_dev); +#endif + mutex_lock(&device_ctls_mutex); + + if (add_edac_dev_to_global_list(edac_dev)) + goto fail0; + + /* set load time so that error rate can be tracked */ + edac_dev->start_time = jiffies; + + /* create this instance's sysfs entries */ + if (edac_device_create_sysfs(edac_dev)) { + edac_device_printk(edac_dev, KERN_WARNING, + "failed to create sysfs device\n"); + goto fail1; + } + + /* If there IS a check routine, then we are running POLLED */ + if (edac_dev->edac_check != NULL) { + /* This instance is NOW RUNNING */ + edac_dev->op_state = OP_RUNNING_POLL; + + /* + * enable workq processing on this instance, + * default = 1000 msec + */ + edac_device_workq_setup(edac_dev, 1000); + } else { + edac_dev->op_state = OP_RUNNING_INTERRUPT; + } + + /* Report action taken */ + edac_device_printk(edac_dev, KERN_INFO, + "Giving out device to module '%s' controller " + "'%s': DEV '%s' (%s)\n", + edac_dev->mod_name, + edac_dev->ctl_name, + dev_name(edac_dev), + edac_op_state_to_string(edac_dev->op_state)); + + mutex_unlock(&device_ctls_mutex); + return 0; + +fail1: + /* Some error, so remove the entry from the lsit */ + del_edac_device_from_global_list(edac_dev); + +fail0: + mutex_unlock(&device_ctls_mutex); + return 1; +} +EXPORT_SYMBOL_GPL(edac_device_add_device); + +/** + * edac_device_del_device: + * Remove sysfs entries for specified edac_device structure and + * then remove edac_device structure from global list + * + * @pdev: + * Pointer to 'struct device' representing edac_device + * structure to remove. + * + * Return: + * Pointer to removed edac_device structure, + * OR NULL if device not found. + */ +struct edac_device_ctl_info *edac_device_del_device(struct device *dev) +{ + struct edac_device_ctl_info *edac_dev; + + debugf0("%s()\n", __func__); + + mutex_lock(&device_ctls_mutex); + + /* Find the structure on the list, if not there, then leave */ + edac_dev = find_edac_device_by_dev(dev); + if (edac_dev == NULL) { + mutex_unlock(&device_ctls_mutex); + return NULL; + } + + /* mark this instance as OFFLINE */ + edac_dev->op_state = OP_OFFLINE; + + /* clear workq processing on this instance */ + edac_device_workq_teardown(edac_dev); + + /* deregister from global list */ + del_edac_device_from_global_list(edac_dev); + + mutex_unlock(&device_ctls_mutex); + + /* Tear down the sysfs entries for this instance */ + edac_device_remove_sysfs(edac_dev); + + edac_printk(KERN_INFO, EDAC_MC, + "Removed device %d for %s %s: DEV %s\n", + edac_dev->dev_idx, + edac_dev->mod_name, edac_dev->ctl_name, dev_name(edac_dev)); + + return edac_dev; +} +EXPORT_SYMBOL_GPL(edac_device_del_device); + +static inline int edac_device_get_log_ce(struct edac_device_ctl_info *edac_dev) +{ + return edac_dev->log_ce; +} + +static inline int edac_device_get_log_ue(struct edac_device_ctl_info *edac_dev) +{ + return edac_dev->log_ue; +} + +static inline int edac_device_get_panic_on_ue(struct edac_device_ctl_info + *edac_dev) +{ + return edac_dev->panic_on_ue; +} + +/* + * edac_device_handle_ce + * perform a common output and handling of an 'edac_dev' CE event + */ +void edac_device_handle_ce(struct edac_device_ctl_info *edac_dev, + int inst_nr, int block_nr, const char *msg) +{ + struct edac_device_instance *instance; + struct edac_device_block *block = NULL; + + if ((inst_nr >= edac_dev->nr_instances) || (inst_nr < 0)) { + edac_device_printk(edac_dev, KERN_ERR, + "INTERNAL ERROR: 'instance' out of range " + "(%d >= %d)\n", inst_nr, + edac_dev->nr_instances); + return; + } + + instance = edac_dev->instances + inst_nr; + + if ((block_nr >= instance->nr_blocks) || (block_nr < 0)) { + edac_device_printk(edac_dev, KERN_ERR, + "INTERNAL ERROR: instance %d 'block' " + "out of range (%d >= %d)\n", + inst_nr, block_nr, + instance->nr_blocks); + return; + } + + if (instance->nr_blocks > 0) { + block = instance->blocks + block_nr; + block->counters.ce_count++; + } + + /* Propogate the count up the 'totals' tree */ + instance->counters.ce_count++; + edac_dev->counters.ce_count++; + + if (edac_device_get_log_ce(edac_dev)) + edac_device_printk(edac_dev, KERN_WARNING, + "CE: %s instance: %s block: %s '%s'\n", + edac_dev->ctl_name, instance->name, + block ? block->name : "N/A", msg); +} +EXPORT_SYMBOL_GPL(edac_device_handle_ce); + +/* + * edac_device_handle_ue + * perform a common output and handling of an 'edac_dev' UE event + */ +void edac_device_handle_ue(struct edac_device_ctl_info *edac_dev, + int inst_nr, int block_nr, const char *msg) +{ + struct edac_device_instance *instance; + struct edac_device_block *block = NULL; + + if ((inst_nr >= edac_dev->nr_instances) || (inst_nr < 0)) { + edac_device_printk(edac_dev, KERN_ERR, + "INTERNAL ERROR: 'instance' out of range " + "(%d >= %d)\n", inst_nr, + edac_dev->nr_instances); + return; + } + + instance = edac_dev->instances + inst_nr; + + if ((block_nr >= instance->nr_blocks) || (block_nr < 0)) { + edac_device_printk(edac_dev, KERN_ERR, + "INTERNAL ERROR: instance %d 'block' " + "out of range (%d >= %d)\n", + inst_nr, block_nr, + instance->nr_blocks); + return; + } + + if (instance->nr_blocks > 0) { + block = instance->blocks + block_nr; + block->counters.ue_count++; + } + + /* Propogate the count up the 'totals' tree */ + instance->counters.ue_count++; + edac_dev->counters.ue_count++; + + if (edac_device_get_log_ue(edac_dev)) + edac_device_printk(edac_dev, KERN_EMERG, + "UE: %s instance: %s block: %s '%s'\n", + edac_dev->ctl_name, instance->name, + block ? block->name : "N/A", msg); + + if (edac_device_get_panic_on_ue(edac_dev)) + panic("EDAC %s: UE instance: %s block %s '%s'\n", + edac_dev->ctl_name, instance->name, + block ? block->name : "N/A", msg); +} +EXPORT_SYMBOL_GPL(edac_device_handle_ue); diff --git a/drivers/edac/edac_device_sysfs.c b/drivers/edac/edac_device_sysfs.c new file mode 100644 index 00000000000..70b837f23c4 --- /dev/null +++ b/drivers/edac/edac_device_sysfs.c @@ -0,0 +1,896 @@ +/* + * file for managing the edac_device class of devices for EDAC + * + * (C) 2007 SoftwareBitMaker (http://www.softwarebitmaker.com) + * + * This file may be distributed under the terms of the + * GNU General Public License. + * + * Written Doug Thompson <norsk5@xmission.com> + * + */ + +#include <linux/ctype.h> +#include <linux/module.h> + +#include "edac_core.h" +#include "edac_module.h" + +#define EDAC_DEVICE_SYMLINK "device" + +#define to_edacdev(k) container_of(k, struct edac_device_ctl_info, kobj) +#define to_edacdev_attr(a) container_of(a, struct edacdev_attribute, attr) + + +/* + * Set of edac_device_ctl_info attribute store/show functions + */ + +/* 'log_ue' */ +static ssize_t edac_device_ctl_log_ue_show(struct edac_device_ctl_info + *ctl_info, char *data) +{ + return sprintf(data, "%u\n", ctl_info->log_ue); +} + +static ssize_t edac_device_ctl_log_ue_store(struct edac_device_ctl_info + *ctl_info, const char *data, + size_t count) +{ + /* if parameter is zero, turn off flag, if non-zero turn on flag */ + ctl_info->log_ue = (simple_strtoul(data, NULL, 0) != 0); + + return count; +} + +/* 'log_ce' */ +static ssize_t edac_device_ctl_log_ce_show(struct edac_device_ctl_info + *ctl_info, char *data) +{ + return sprintf(data, "%u\n", ctl_info->log_ce); +} + +static ssize_t edac_device_ctl_log_ce_store(struct edac_device_ctl_info + *ctl_info, const char *data, + size_t count) +{ + /* if parameter is zero, turn off flag, if non-zero turn on flag */ + ctl_info->log_ce = (simple_strtoul(data, NULL, 0) != 0); + + return count; +} + +/* 'panic_on_ue' */ +static ssize_t edac_device_ctl_panic_on_ue_show(struct edac_device_ctl_info + *ctl_info, char *data) +{ + return sprintf(data, "%u\n", ctl_info->panic_on_ue); +} + +static ssize_t edac_device_ctl_panic_on_ue_store(struct edac_device_ctl_info + *ctl_info, const char *data, + size_t count) +{ + /* if parameter is zero, turn off flag, if non-zero turn on flag */ + ctl_info->panic_on_ue = (simple_strtoul(data, NULL, 0) != 0); + + return count; +} + +/* 'poll_msec' show and store functions*/ +static ssize_t edac_device_ctl_poll_msec_show(struct edac_device_ctl_info + *ctl_info, char *data) +{ + return sprintf(data, "%u\n", ctl_info->poll_msec); +} + +static ssize_t edac_device_ctl_poll_msec_store(struct edac_device_ctl_info + *ctl_info, const char *data, + size_t count) +{ + unsigned long value; + + /* get the value and enforce that it is non-zero, must be at least + * one millisecond for the delay period, between scans + * Then cancel last outstanding delay for the work request + * and set a new one. + */ + value = simple_strtoul(data, NULL, 0); + edac_device_reset_delay_period(ctl_info, value); + + return count; +} + +/* edac_device_ctl_info specific attribute structure */ +struct ctl_info_attribute { + struct attribute attr; + ssize_t(*show) (struct edac_device_ctl_info *, char *); + ssize_t(*store) (struct edac_device_ctl_info *, const char *, size_t); +}; + +#define to_ctl_info(k) container_of(k, struct edac_device_ctl_info, kobj) +#define to_ctl_info_attr(a) container_of(a,struct ctl_info_attribute,attr) + +/* Function to 'show' fields from the edac_dev 'ctl_info' structure */ +static ssize_t edac_dev_ctl_info_show(struct kobject *kobj, + struct attribute *attr, char *buffer) +{ + struct edac_device_ctl_info *edac_dev = to_ctl_info(kobj); + struct ctl_info_attribute *ctl_info_attr = to_ctl_info_attr(attr); + + if (ctl_info_attr->show) + return ctl_info_attr->show(edac_dev, buffer); + return -EIO; +} + +/* Function to 'store' fields into the edac_dev 'ctl_info' structure */ +static ssize_t edac_dev_ctl_info_store(struct kobject *kobj, + struct attribute *attr, + const char *buffer, size_t count) +{ + struct edac_device_ctl_info *edac_dev = to_ctl_info(kobj); + struct ctl_info_attribute *ctl_info_attr = to_ctl_info_attr(attr); + + if (ctl_info_attr->store) + return ctl_info_attr->store(edac_dev, buffer, count); + return -EIO; +} + +/* edac_dev file operations for an 'ctl_info' */ +static struct sysfs_ops device_ctl_info_ops = { + .show = edac_dev_ctl_info_show, + .store = edac_dev_ctl_info_store +}; + +#define CTL_INFO_ATTR(_name,_mode,_show,_store) \ +static struct ctl_info_attribute attr_ctl_info_##_name = { \ + .attr = {.name = __stringify(_name), .mode = _mode }, \ + .show = _show, \ + .store = _store, \ +}; + +/* Declare the various ctl_info attributes here and their respective ops */ +CTL_INFO_ATTR(log_ue, S_IRUGO | S_IWUSR, + edac_device_ctl_log_ue_show, edac_device_ctl_log_ue_store); +CTL_INFO_ATTR(log_ce, S_IRUGO | S_IWUSR, + edac_device_ctl_log_ce_show, edac_device_ctl_log_ce_store); +CTL_INFO_ATTR(panic_on_ue, S_IRUGO | S_IWUSR, + edac_device_ctl_panic_on_ue_show, + edac_device_ctl_panic_on_ue_store); +CTL_INFO_ATTR(poll_msec, S_IRUGO | S_IWUSR, + edac_device_ctl_poll_msec_show, edac_device_ctl_poll_msec_store); + +/* Base Attributes of the EDAC_DEVICE ECC object */ +static struct ctl_info_attribute *device_ctrl_attr[] = { + &attr_ctl_info_panic_on_ue, + &attr_ctl_info_log_ue, + &attr_ctl_info_log_ce, + &attr_ctl_info_poll_msec, + NULL, +}; + +/* + * edac_device_ctrl_master_release + * + * called when the reference count for the 'main' kobj + * for a edac_device control struct reaches zero + * + * Reference count model: + * One 'main' kobject for each control structure allocated. + * That main kobj is initially set to one AND + * the reference count for the EDAC 'core' module is + * bumped by one, thus added 'keep in memory' dependency. + * + * Each new internal kobj (in instances and blocks) then + * bumps the 'main' kobject. + * + * When they are released their release functions decrement + * the 'main' kobj. + * + * When the main kobj reaches zero (0) then THIS function + * is called which then decrements the EDAC 'core' module. + * When the module reference count reaches zero then the + * module no longer has dependency on keeping the release + * function code in memory and module can be unloaded. + * + * This will support several control objects as well, each + * with its own 'main' kobj. + */ +static void edac_device_ctrl_master_release(struct kobject *kobj) +{ + struct edac_device_ctl_info *edac_dev = to_edacdev(kobj); + + debugf4("%s() control index=%d\n", __func__, edac_dev->dev_idx); + + /* decrement the EDAC CORE module ref count */ + module_put(edac_dev->owner); + + /* free the control struct containing the 'main' kobj + * passed in to this routine + */ + kfree(edac_dev); +} + +/* ktype for the main (master) kobject */ +static struct kobj_type ktype_device_ctrl = { + .release = edac_device_ctrl_master_release, + .sysfs_ops = &device_ctl_info_ops, + .default_attrs = (struct attribute **)device_ctrl_attr, +}; + +/* + * edac_device_register_sysfs_main_kobj + * + * perform the high level setup for the new edac_device instance + * + * Return: 0 SUCCESS + * !0 FAILURE + */ +int edac_device_register_sysfs_main_kobj(struct edac_device_ctl_info *edac_dev) +{ + struct sysdev_class *edac_class; + int err; + + debugf1("%s()\n", __func__); + + /* get the /sys/devices/system/edac reference */ + edac_class = edac_get_edac_class(); + if (edac_class == NULL) { + debugf1("%s() no edac_class error\n", __func__); + err = -ENODEV; + goto err_out; + } + + /* Point to the 'edac_class' this instance 'reports' to */ + edac_dev->edac_class = edac_class; + + /* Init the devices's kobject */ + memset(&edac_dev->kobj, 0, sizeof(struct kobject)); + edac_dev->kobj.ktype = &ktype_device_ctrl; + + /* set this new device under the edac_class kobject */ + edac_dev->kobj.parent = &edac_class->kset.kobj; + + /* generate sysfs "..../edac/<name>" */ + debugf4("%s() set name of kobject to: %s\n", __func__, edac_dev->name); + err = kobject_set_name(&edac_dev->kobj, "%s", edac_dev->name); + if (err) + goto err_out; + + /* Record which module 'owns' this control structure + * and bump the ref count of the module + */ + edac_dev->owner = THIS_MODULE; + + if (!try_module_get(edac_dev->owner)) { + err = -ENODEV; + goto err_out; + } + + /* register */ + err = kobject_register(&edac_dev->kobj); + if (err) { + debugf1("%s()Failed to register '.../edac/%s'\n", + __func__, edac_dev->name); + goto err_kobj_reg; + } + + /* At this point, to 'free' the control struct, + * edac_device_unregister_sysfs_main_kobj() must be used + */ + + debugf4("%s() Registered '.../edac/%s' kobject\n", + __func__, edac_dev->name); + + return 0; + + /* Error exit stack */ +err_kobj_reg: + module_put(edac_dev->owner); + +err_out: + return err; +} + +/* + * edac_device_unregister_sysfs_main_kobj: + * the '..../edac/<name>' kobject + */ +void edac_device_unregister_sysfs_main_kobj( + struct edac_device_ctl_info *edac_dev) +{ + debugf0("%s()\n", __func__); + debugf4("%s() name of kobject is: %s\n", + __func__, kobject_name(&edac_dev->kobj)); + + /* + * Unregister the edac device's kobject and + * allow for reference count to reach 0 at which point + * the callback will be called to: + * a) module_put() this module + * b) 'kfree' the memory + */ + kobject_unregister(&edac_dev->kobj); +} + +/* edac_dev -> instance information */ + +/* + * Set of low-level instance attribute show functions + */ +static ssize_t instance_ue_count_show(struct edac_device_instance *instance, + char *data) +{ + return sprintf(data, "%u\n", instance->counters.ue_count); +} + +static ssize_t instance_ce_count_show(struct edac_device_instance *instance, + char *data) +{ + return sprintf(data, "%u\n", instance->counters.ce_count); +} + +#define to_instance(k) container_of(k, struct edac_device_instance, kobj) +#define to_instance_attr(a) container_of(a,struct instance_attribute,attr) + +/* DEVICE instance kobject release() function */ +static void edac_device_ctrl_instance_release(struct kobject *kobj) +{ + struct edac_device_instance *instance; + + debugf1("%s()\n", __func__); + + /* map from this kobj to the main control struct + * and then dec the main kobj count + */ + instance = to_instance(kobj); + kobject_put(&instance->ctl->kobj); +} + +/* instance specific attribute structure */ +struct instance_attribute { + struct attribute attr; + ssize_t(*show) (struct edac_device_instance *, char *); + ssize_t(*store) (struct edac_device_instance *, const char *, size_t); +}; + +/* Function to 'show' fields from the edac_dev 'instance' structure */ +static ssize_t edac_dev_instance_show(struct kobject *kobj, + struct attribute *attr, char *buffer) +{ + struct edac_device_instance *instance = to_instance(kobj); + struct instance_attribute *instance_attr = to_instance_attr(attr); + + if (instance_attr->show) + return instance_attr->show(instance, buffer); + return -EIO; +} + +/* Function to 'store' fields into the edac_dev 'instance' structure */ +static ssize_t edac_dev_instance_store(struct kobject *kobj, + struct attribute *attr, + const char *buffer, size_t count) +{ + struct edac_device_instance *instance = to_instance(kobj); + struct instance_attribute *instance_attr = to_instance_attr(attr); + + if (instance_attr->store) + return instance_attr->store(instance, buffer, count); + return -EIO; +} + +/* edac_dev file operations for an 'instance' */ +static struct sysfs_ops device_instance_ops = { + .show = edac_dev_instance_show, + .store = edac_dev_instance_store +}; + +#define INSTANCE_ATTR(_name,_mode,_show,_store) \ +static struct instance_attribute attr_instance_##_name = { \ + .attr = {.name = __stringify(_name), .mode = _mode }, \ + .show = _show, \ + .store = _store, \ +}; + +/* + * Define attributes visible for the edac_device instance object + * Each contains a pointer to a show and an optional set + * function pointer that does the low level output/input + */ +INSTANCE_ATTR(ce_count, S_IRUGO, instance_ce_count_show, NULL); +INSTANCE_ATTR(ue_count, S_IRUGO, instance_ue_count_show, NULL); + +/* list of edac_dev 'instance' attributes */ +static struct instance_attribute *device_instance_attr[] = { + &attr_instance_ce_count, + &attr_instance_ue_count, + NULL, +}; + +/* The 'ktype' for each edac_dev 'instance' */ +static struct kobj_type ktype_instance_ctrl = { + .release = edac_device_ctrl_instance_release, + .sysfs_ops = &device_instance_ops, + .default_attrs = (struct attribute **)device_instance_attr, +}; + +/* edac_dev -> instance -> block information */ + +#define to_block(k) container_of(k, struct edac_device_block, kobj) +#define to_block_attr(a) \ + container_of(a, struct edac_dev_sysfs_block_attribute, attr) + +/* + * Set of low-level block attribute show functions + */ +static ssize_t block_ue_count_show(struct kobject *kobj, + struct attribute *attr, char *data) +{ + struct edac_device_block *block = to_block(kobj); + + return sprintf(data, "%u\n", block->counters.ue_count); +} + +static ssize_t block_ce_count_show(struct kobject *kobj, + struct attribute *attr, char *data) +{ + struct edac_device_block *block = to_block(kobj); + + return sprintf(data, "%u\n", block->counters.ce_count); +} + +/* DEVICE block kobject release() function */ +static void edac_device_ctrl_block_release(struct kobject *kobj) +{ + struct edac_device_block *block; + + debugf1("%s()\n", __func__); + + /* get the container of the kobj */ + block = to_block(kobj); + + /* map from 'block kobj' to 'block->instance->controller->main_kobj' + * now 'release' the block kobject + */ + kobject_put(&block->instance->ctl->kobj); +} + + +/* Function to 'show' fields from the edac_dev 'block' structure */ +static ssize_t edac_dev_block_show(struct kobject *kobj, + struct attribute *attr, char *buffer) +{ + struct edac_dev_sysfs_block_attribute *block_attr = + to_block_attr(attr); + + if (block_attr->show) + return block_attr->show(kobj, attr, buffer); + return -EIO; +} + +/* Function to 'store' fields into the edac_dev 'block' structure */ +static ssize_t edac_dev_block_store(struct kobject *kobj, + struct attribute *attr, + const char *buffer, size_t count) +{ + struct edac_dev_sysfs_block_attribute *block_attr; + + block_attr = to_block_attr(attr); + + if (block_attr->store) + return block_attr->store(kobj, attr, buffer, count); + return -EIO; +} + +/* edac_dev file operations for a 'block' */ +static struct sysfs_ops device_block_ops = { + .show = edac_dev_block_show, + .store = edac_dev_block_store +}; + +#define BLOCK_ATTR(_name,_mode,_show,_store) \ +static struct edac_dev_sysfs_block_attribute attr_block_##_name = { \ + .attr = {.name = __stringify(_name), .mode = _mode }, \ + .show = _show, \ + .store = _store, \ +}; + +BLOCK_ATTR(ce_count, S_IRUGO, block_ce_count_show, NULL); +BLOCK_ATTR(ue_count, S_IRUGO, block_ue_count_show, NULL); + +/* list of edac_dev 'block' attributes */ +static struct edac_dev_sysfs_block_attribute *device_block_attr[] = { + &attr_block_ce_count, + &attr_block_ue_count, + NULL, +}; + +/* The 'ktype' for each edac_dev 'block' */ +static struct kobj_type ktype_block_ctrl = { + .release = edac_device_ctrl_block_release, + .sysfs_ops = &device_block_ops, + .default_attrs = (struct attribute **)device_block_attr, +}; + +/* block ctor/dtor code */ + +/* + * edac_device_create_block + */ +static int edac_device_create_block(struct edac_device_ctl_info *edac_dev, + struct edac_device_instance *instance, + struct edac_device_block *block) +{ + int i; + int err; + struct edac_dev_sysfs_block_attribute *sysfs_attrib; + struct kobject *main_kobj; + + debugf4("%s() Instance '%s' inst_p=%p block '%s' block_p=%p\n", + __func__, instance->name, instance, block->name, block); + debugf4("%s() block kobj=%p block kobj->parent=%p\n", + __func__, &block->kobj, &block->kobj.parent); + + /* init this block's kobject */ + memset(&block->kobj, 0, sizeof(struct kobject)); + block->kobj.parent = &instance->kobj; + block->kobj.ktype = &ktype_block_ctrl; + + err = kobject_set_name(&block->kobj, "%s", block->name); + if (err) + return err; + + /* bump the main kobject's reference count for this controller + * and this instance is dependant on the main + */ + main_kobj = kobject_get(&edac_dev->kobj); + if (!main_kobj) { + err = -ENODEV; + goto err_out; + } + + /* Add this block's kobject */ + err = kobject_register(&block->kobj); + if (err) { + debugf1("%s() Failed to register instance '%s'\n", + __func__, block->name); + kobject_put(main_kobj); + err = -ENODEV; + goto err_out; + } + + /* If there are driver level block attributes, then added them + * to the block kobject + */ + sysfs_attrib = block->block_attributes; + if (sysfs_attrib && block->nr_attribs) { + for (i = 0; i < block->nr_attribs; i++, sysfs_attrib++) { + + debugf4("%s() creating block attrib='%s' " + "attrib->%p to kobj=%p\n", + __func__, + sysfs_attrib->attr.name, + sysfs_attrib, &block->kobj); + + /* Create each block_attribute file */ + err = sysfs_create_file(&block->kobj, + &sysfs_attrib->attr); + if (err) + goto err_on_attrib; + } + } + + return 0; + + /* Error unwind stack */ +err_on_attrib: + kobject_unregister(&block->kobj); + +err_out: + return err; +} + +/* + * edac_device_delete_block(edac_dev,block); + */ +static void edac_device_delete_block(struct edac_device_ctl_info *edac_dev, + struct edac_device_block *block) +{ + struct edac_dev_sysfs_block_attribute *sysfs_attrib; + int i; + + /* if this block has 'attributes' then we need to iterate over the list + * and 'remove' the attributes on this block + */ + sysfs_attrib = block->block_attributes; + if (sysfs_attrib && block->nr_attribs) { + for (i = 0; i < block->nr_attribs; i++, sysfs_attrib++) { + + /* remove each block_attrib file */ + sysfs_remove_file(&block->kobj, + (struct attribute *) sysfs_attrib); + } + } + + /* unregister this block's kobject, SEE: + * edac_device_ctrl_block_release() callback operation + */ + kobject_unregister(&block->kobj); +} + +/* instance ctor/dtor code */ + +/* + * edac_device_create_instance + * create just one instance of an edac_device 'instance' + */ +static int edac_device_create_instance(struct edac_device_ctl_info *edac_dev, + int idx) +{ + int i, j; + int err; + struct edac_device_instance *instance; + struct kobject *main_kobj; + + instance = &edac_dev->instances[idx]; + + /* Init the instance's kobject */ + memset(&instance->kobj, 0, sizeof(struct kobject)); + + /* set this new device under the edac_device main kobject */ + instance->kobj.parent = &edac_dev->kobj; + instance->kobj.ktype = &ktype_instance_ctrl; + instance->ctl = edac_dev; + + err = kobject_set_name(&instance->kobj, "%s", instance->name); + if (err) + goto err_out; + + /* bump the main kobject's reference count for this controller + * and this instance is dependant on the main + */ + main_kobj = kobject_get(&edac_dev->kobj); + if (!main_kobj) { + err = -ENODEV; + goto err_out; + } + + /* Formally register this instance's kobject */ + err = kobject_register(&instance->kobj); + if (err != 0) { + debugf2("%s() Failed to register instance '%s'\n", + __func__, instance->name); + kobject_put(main_kobj); + goto err_out; + } + + debugf4("%s() now register '%d' blocks for instance %d\n", + __func__, instance->nr_blocks, idx); + + /* register all blocks of this instance */ + for (i = 0; i < instance->nr_blocks; i++) { + err = edac_device_create_block(edac_dev, instance, + &instance->blocks[i]); + if (err) { + /* If any fail, remove all previous ones */ + for (j = 0; j < i; j++) + edac_device_delete_block(edac_dev, + &instance->blocks[j]); + goto err_release_instance_kobj; + } + } + + debugf4("%s() Registered instance %d '%s' kobject\n", + __func__, idx, instance->name); + + return 0; + + /* error unwind stack */ +err_release_instance_kobj: + kobject_unregister(&instance->kobj); + +err_out: + return err; +} + +/* + * edac_device_remove_instance + * remove an edac_device instance + */ +static void edac_device_delete_instance(struct edac_device_ctl_info *edac_dev, + int idx) +{ + struct edac_device_instance *instance; + int i; + + instance = &edac_dev->instances[idx]; + + /* unregister all blocks in this instance */ + for (i = 0; i < instance->nr_blocks; i++) + edac_device_delete_block(edac_dev, &instance->blocks[i]); + + /* unregister this instance's kobject, SEE: + * edac_device_ctrl_instance_release() for callback operation + */ + kobject_unregister(&instance->kobj); +} + +/* + * edac_device_create_instances + * create the first level of 'instances' for this device + * (ie 'cache' might have 'cache0', 'cache1', 'cache2', etc + */ +static int edac_device_create_instances(struct edac_device_ctl_info *edac_dev) +{ + int i, j; + int err; + + debugf0("%s()\n", __func__); + + /* iterate over creation of the instances */ + for (i = 0; i < edac_dev->nr_instances; i++) { + err = edac_device_create_instance(edac_dev, i); + if (err) { + /* unwind previous instances on error */ + for (j = 0; j < i; j++) + edac_device_delete_instance(edac_dev, j); + return err; + } + } + + return 0; +} + +/* + * edac_device_delete_instances(edac_dev); + * unregister all the kobjects of the instances + */ +static void edac_device_delete_instances(struct edac_device_ctl_info *edac_dev) +{ + int i; + + /* iterate over creation of the instances */ + for (i = 0; i < edac_dev->nr_instances; i++) + edac_device_delete_instance(edac_dev, i); +} + +/* edac_dev sysfs ctor/dtor code */ + +/* + * edac_device_add_main_sysfs_attributes + * add some attributes to this instance's main kobject + */ +static int edac_device_add_main_sysfs_attributes( + struct edac_device_ctl_info *edac_dev) +{ + struct edac_dev_sysfs_attribute *sysfs_attrib; + int err = 0; + + sysfs_attrib = edac_dev->sysfs_attributes; + if (sysfs_attrib) { + /* iterate over the array and create an attribute for each + * entry in the list + */ + while (sysfs_attrib->attr.name != NULL) { + err = sysfs_create_file(&edac_dev->kobj, + (struct attribute*) sysfs_attrib); + if (err) + goto err_out; + + sysfs_attrib++; + } + } + +err_out: + return err; +} + +/* + * edac_device_remove_main_sysfs_attributes + * remove any attributes to this instance's main kobject + */ +static void edac_device_remove_main_sysfs_attributes( + struct edac_device_ctl_info *edac_dev) +{ + struct edac_dev_sysfs_attribute *sysfs_attrib; + + /* if there are main attributes, defined, remove them. First, + * point to the start of the array and iterate over it + * removing each attribute listed from this device's instance's kobject + */ + sysfs_attrib = edac_dev->sysfs_attributes; + if (sysfs_attrib) { + while (sysfs_attrib->attr.name != NULL) { + sysfs_remove_file(&edac_dev->kobj, + (struct attribute *) sysfs_attrib); + sysfs_attrib++; + } + } +} + +/* + * edac_device_create_sysfs() Constructor + * + * accept a created edac_device control structure + * and 'export' it to sysfs. The 'main' kobj should already have been + * created. 'instance' and 'block' kobjects should be registered + * along with any 'block' attributes from the low driver. In addition, + * the main attributes (if any) are connected to the main kobject of + * the control structure. + * + * Return: + * 0 Success + * !0 Failure + */ +int edac_device_create_sysfs(struct edac_device_ctl_info *edac_dev) +{ + int err; + struct kobject *edac_kobj = &edac_dev->kobj; + + debugf0("%s() idx=%d\n", __func__, edac_dev->dev_idx); + + /* go create any main attributes callers wants */ + err = edac_device_add_main_sysfs_attributes(edac_dev); + if (err) { + debugf0("%s() failed to add sysfs attribs\n", __func__); + goto err_out; + } + + /* create a symlink from the edac device + * to the platform 'device' being used for this + */ + err = sysfs_create_link(edac_kobj, + &edac_dev->dev->kobj, EDAC_DEVICE_SYMLINK); + if (err) { + debugf0("%s() sysfs_create_link() returned err= %d\n", + __func__, err); + goto err_remove_main_attribs; + } + + /* Create the first level instance directories + * In turn, the nested blocks beneath the instances will + * be registered as well + */ + err = edac_device_create_instances(edac_dev); + if (err) { + debugf0("%s() edac_device_create_instances() " + "returned err= %d\n", __func__, err); + goto err_remove_link; + } + + + debugf4("%s() create-instances done, idx=%d\n", + __func__, edac_dev->dev_idx); + + return 0; + + /* Error unwind stack */ +err_remove_link: + /* remove the sym link */ + sysfs_remove_link(&edac_dev->kobj, EDAC_DEVICE_SYMLINK); + +err_remove_main_attribs: + edac_device_remove_main_sysfs_attributes(edac_dev); + +err_out: + return err; +} + +/* + * edac_device_remove_sysfs() destructor + * + * given an edac_device struct, tear down the kobject resources + */ +void edac_device_remove_sysfs(struct edac_device_ctl_info *edac_dev) +{ + debugf0("%s()\n", __func__); + + /* remove any main attributes for this device */ + edac_device_remove_main_sysfs_attributes(edac_dev); + + /* remove the device sym link */ + sysfs_remove_link(&edac_dev->kobj, EDAC_DEVICE_SYMLINK); + + /* walk the instance/block kobject tree, deconstructing it */ + edac_device_delete_instances(edac_dev); +} diff --git a/drivers/edac/edac_mc.c b/drivers/edac/edac_mc.c index 804875de580..4471be36259 100644 --- a/drivers/edac/edac_mc.c +++ b/drivers/edac/edac_mc.c @@ -27,1200 +27,20 @@ #include <linux/list.h> #include <linux/sysdev.h> #include <linux/ctype.h> -#include <linux/kthread.h> -#include <linux/freezer.h> +#include <linux/edac.h> #include <asm/uaccess.h> #include <asm/page.h> #include <asm/edac.h> -#include "edac_mc.h" - -#define EDAC_MC_VERSION "Ver: 2.0.1 " __DATE__ - - -#ifdef CONFIG_EDAC_DEBUG -/* Values of 0 to 4 will generate output */ -int edac_debug_level = 1; -EXPORT_SYMBOL_GPL(edac_debug_level); -#endif - -/* EDAC Controls, setable by module parameter, and sysfs */ -static int log_ue = 1; -static int log_ce = 1; -static int panic_on_ue; -static int poll_msec = 1000; +#include "edac_core.h" +#include "edac_module.h" /* lock to memory controller's control array */ -static DECLARE_MUTEX(mem_ctls_mutex); +static DEFINE_MUTEX(mem_ctls_mutex); static struct list_head mc_devices = LIST_HEAD_INIT(mc_devices); -static struct task_struct *edac_thread; - -#ifdef CONFIG_PCI -static int check_pci_parity = 0; /* default YES check PCI parity */ -static int panic_on_pci_parity; /* default no panic on PCI Parity */ -static atomic_t pci_parity_count = ATOMIC_INIT(0); - -static struct kobject edac_pci_kobj; /* /sys/devices/system/edac/pci */ -static struct completion edac_pci_kobj_complete; -#endif /* CONFIG_PCI */ - -/* START sysfs data and methods */ - - -static const char *mem_types[] = { - [MEM_EMPTY] = "Empty", - [MEM_RESERVED] = "Reserved", - [MEM_UNKNOWN] = "Unknown", - [MEM_FPM] = "FPM", - [MEM_EDO] = "EDO", - [MEM_BEDO] = "BEDO", - [MEM_SDR] = "Unbuffered-SDR", - [MEM_RDR] = "Registered-SDR", - [MEM_DDR] = "Unbuffered-DDR", - [MEM_RDDR] = "Registered-DDR", - [MEM_RMBS] = "RMBS" -}; - -static const char *dev_types[] = { - [DEV_UNKNOWN] = "Unknown", - [DEV_X1] = "x1", - [DEV_X2] = "x2", - [DEV_X4] = "x4", - [DEV_X8] = "x8", - [DEV_X16] = "x16", - [DEV_X32] = "x32", - [DEV_X64] = "x64" -}; - -static const char *edac_caps[] = { - [EDAC_UNKNOWN] = "Unknown", - [EDAC_NONE] = "None", - [EDAC_RESERVED] = "Reserved", - [EDAC_PARITY] = "PARITY", - [EDAC_EC] = "EC", - [EDAC_SECDED] = "SECDED", - [EDAC_S2ECD2ED] = "S2ECD2ED", - [EDAC_S4ECD4ED] = "S4ECD4ED", - [EDAC_S8ECD8ED] = "S8ECD8ED", - [EDAC_S16ECD16ED] = "S16ECD16ED" -}; - -/* sysfs object: /sys/devices/system/edac */ -static struct sysdev_class edac_class = { - set_kset_name("edac"), -}; - -/* sysfs object: - * /sys/devices/system/edac/mc - */ -static struct kobject edac_memctrl_kobj; - -/* We use these to wait for the reference counts on edac_memctrl_kobj and - * edac_pci_kobj to reach 0. - */ -static struct completion edac_memctrl_kobj_complete; - -/* - * /sys/devices/system/edac/mc; - * data structures and methods - */ -static ssize_t memctrl_int_show(void *ptr, char *buffer) -{ - int *value = (int*) ptr; - return sprintf(buffer, "%u\n", *value); -} - -static ssize_t memctrl_int_store(void *ptr, const char *buffer, size_t count) -{ - int *value = (int*) ptr; - - if (isdigit(*buffer)) - *value = simple_strtoul(buffer, NULL, 0); - - return count; -} - -struct memctrl_dev_attribute { - struct attribute attr; - void *value; - ssize_t (*show)(void *,char *); - ssize_t (*store)(void *, const char *, size_t); -}; - -/* Set of show/store abstract level functions for memory control object */ -static ssize_t memctrl_dev_show(struct kobject *kobj, - struct attribute *attr, char *buffer) -{ - struct memctrl_dev_attribute *memctrl_dev; - memctrl_dev = (struct memctrl_dev_attribute*)attr; - - if (memctrl_dev->show) - return memctrl_dev->show(memctrl_dev->value, buffer); - - return -EIO; -} - -static ssize_t memctrl_dev_store(struct kobject *kobj, struct attribute *attr, - const char *buffer, size_t count) -{ - struct memctrl_dev_attribute *memctrl_dev; - memctrl_dev = (struct memctrl_dev_attribute*)attr; - - if (memctrl_dev->store) - return memctrl_dev->store(memctrl_dev->value, buffer, count); - - return -EIO; -} - -static struct sysfs_ops memctrlfs_ops = { - .show = memctrl_dev_show, - .store = memctrl_dev_store -}; - -#define MEMCTRL_ATTR(_name,_mode,_show,_store) \ -struct memctrl_dev_attribute attr_##_name = { \ - .attr = {.name = __stringify(_name), .mode = _mode }, \ - .value = &_name, \ - .show = _show, \ - .store = _store, \ -}; - -#define MEMCTRL_STRING_ATTR(_name,_data,_mode,_show,_store) \ -struct memctrl_dev_attribute attr_##_name = { \ - .attr = {.name = __stringify(_name), .mode = _mode }, \ - .value = _data, \ - .show = _show, \ - .store = _store, \ -}; - -/* csrow<id> control files */ -MEMCTRL_ATTR(panic_on_ue,S_IRUGO|S_IWUSR,memctrl_int_show,memctrl_int_store); -MEMCTRL_ATTR(log_ue,S_IRUGO|S_IWUSR,memctrl_int_show,memctrl_int_store); -MEMCTRL_ATTR(log_ce,S_IRUGO|S_IWUSR,memctrl_int_show,memctrl_int_store); -MEMCTRL_ATTR(poll_msec,S_IRUGO|S_IWUSR,memctrl_int_show,memctrl_int_store); - -/* Base Attributes of the memory ECC object */ -static struct memctrl_dev_attribute *memctrl_attr[] = { - &attr_panic_on_ue, - &attr_log_ue, - &attr_log_ce, - &attr_poll_msec, - NULL, -}; - -/* Main MC kobject release() function */ -static void edac_memctrl_master_release(struct kobject *kobj) -{ - debugf1("%s()\n", __func__); - complete(&edac_memctrl_kobj_complete); -} - -static struct kobj_type ktype_memctrl = { - .release = edac_memctrl_master_release, - .sysfs_ops = &memctrlfs_ops, - .default_attrs = (struct attribute **) memctrl_attr, -}; - -/* Initialize the main sysfs entries for edac: - * /sys/devices/system/edac - * - * and children - * - * Return: 0 SUCCESS - * !0 FAILURE - */ -static int edac_sysfs_memctrl_setup(void) -{ - int err = 0; - - debugf1("%s()\n", __func__); - - /* create the /sys/devices/system/edac directory */ - err = sysdev_class_register(&edac_class); - - if (err) { - debugf1("%s() error=%d\n", __func__, err); - return err; - } - - /* Init the MC's kobject */ - memset(&edac_memctrl_kobj, 0, sizeof (edac_memctrl_kobj)); - edac_memctrl_kobj.parent = &edac_class.kset.kobj; - edac_memctrl_kobj.ktype = &ktype_memctrl; - - /* generate sysfs "..../edac/mc" */ - err = kobject_set_name(&edac_memctrl_kobj,"mc"); - - if (err) - goto fail; - - /* FIXME: maybe new sysdev_create_subdir() */ - err = kobject_register(&edac_memctrl_kobj); - - if (err) { - debugf1("Failed to register '.../edac/mc'\n"); - goto fail; - } - - debugf1("Registered '.../edac/mc' kobject\n"); - - return 0; - -fail: - sysdev_class_unregister(&edac_class); - return err; -} - -/* - * MC teardown: - * the '..../edac/mc' kobject followed by '..../edac' itself - */ -static void edac_sysfs_memctrl_teardown(void) -{ - debugf0("MC: " __FILE__ ": %s()\n", __func__); - - /* Unregister the MC's kobject and wait for reference count to reach - * 0. - */ - init_completion(&edac_memctrl_kobj_complete); - kobject_unregister(&edac_memctrl_kobj); - wait_for_completion(&edac_memctrl_kobj_complete); - - /* Unregister the 'edac' object */ - sysdev_class_unregister(&edac_class); -} - -#ifdef CONFIG_PCI -static ssize_t edac_pci_int_show(void *ptr, char *buffer) -{ - int *value = ptr; - return sprintf(buffer,"%d\n",*value); -} - -static ssize_t edac_pci_int_store(void *ptr, const char *buffer, size_t count) -{ - int *value = ptr; - - if (isdigit(*buffer)) - *value = simple_strtoul(buffer,NULL,0); - - return count; -} - -struct edac_pci_dev_attribute { - struct attribute attr; - void *value; - ssize_t (*show)(void *,char *); - ssize_t (*store)(void *, const char *,size_t); -}; - -/* Set of show/store abstract level functions for PCI Parity object */ -static ssize_t edac_pci_dev_show(struct kobject *kobj, struct attribute *attr, - char *buffer) -{ - struct edac_pci_dev_attribute *edac_pci_dev; - edac_pci_dev= (struct edac_pci_dev_attribute*)attr; - - if (edac_pci_dev->show) - return edac_pci_dev->show(edac_pci_dev->value, buffer); - return -EIO; -} - -static ssize_t edac_pci_dev_store(struct kobject *kobj, - struct attribute *attr, const char *buffer, size_t count) -{ - struct edac_pci_dev_attribute *edac_pci_dev; - edac_pci_dev= (struct edac_pci_dev_attribute*)attr; - - if (edac_pci_dev->show) - return edac_pci_dev->store(edac_pci_dev->value, buffer, count); - return -EIO; -} - -static struct sysfs_ops edac_pci_sysfs_ops = { - .show = edac_pci_dev_show, - .store = edac_pci_dev_store -}; - -#define EDAC_PCI_ATTR(_name,_mode,_show,_store) \ -struct edac_pci_dev_attribute edac_pci_attr_##_name = { \ - .attr = {.name = __stringify(_name), .mode = _mode }, \ - .value = &_name, \ - .show = _show, \ - .store = _store, \ -}; - -#define EDAC_PCI_STRING_ATTR(_name,_data,_mode,_show,_store) \ -struct edac_pci_dev_attribute edac_pci_attr_##_name = { \ - .attr = {.name = __stringify(_name), .mode = _mode }, \ - .value = _data, \ - .show = _show, \ - .store = _store, \ -}; - -/* PCI Parity control files */ -EDAC_PCI_ATTR(check_pci_parity, S_IRUGO|S_IWUSR, edac_pci_int_show, - edac_pci_int_store); -EDAC_PCI_ATTR(panic_on_pci_parity, S_IRUGO|S_IWUSR, edac_pci_int_show, - edac_pci_int_store); -EDAC_PCI_ATTR(pci_parity_count, S_IRUGO, edac_pci_int_show, NULL); - -/* Base Attributes of the memory ECC object */ -static struct edac_pci_dev_attribute *edac_pci_attr[] = { - &edac_pci_attr_check_pci_parity, - &edac_pci_attr_panic_on_pci_parity, - &edac_pci_attr_pci_parity_count, - NULL, -}; - -/* No memory to release */ -static void edac_pci_release(struct kobject *kobj) -{ - debugf1("%s()\n", __func__); - complete(&edac_pci_kobj_complete); -} - -static struct kobj_type ktype_edac_pci = { - .release = edac_pci_release, - .sysfs_ops = &edac_pci_sysfs_ops, - .default_attrs = (struct attribute **) edac_pci_attr, -}; - -/** - * edac_sysfs_pci_setup() - * - */ -static int edac_sysfs_pci_setup(void) -{ - int err; - - debugf1("%s()\n", __func__); - - memset(&edac_pci_kobj, 0, sizeof(edac_pci_kobj)); - edac_pci_kobj.parent = &edac_class.kset.kobj; - edac_pci_kobj.ktype = &ktype_edac_pci; - err = kobject_set_name(&edac_pci_kobj, "pci"); - - if (!err) { - /* Instanstiate the csrow object */ - /* FIXME: maybe new sysdev_create_subdir() */ - err = kobject_register(&edac_pci_kobj); - - if (err) - debugf1("Failed to register '.../edac/pci'\n"); - else - debugf1("Registered '.../edac/pci' kobject\n"); - } - - return err; -} - -static void edac_sysfs_pci_teardown(void) -{ - debugf0("%s()\n", __func__); - init_completion(&edac_pci_kobj_complete); - kobject_unregister(&edac_pci_kobj); - wait_for_completion(&edac_pci_kobj_complete); -} - - -static u16 get_pci_parity_status(struct pci_dev *dev, int secondary) -{ - int where; - u16 status; - - where = secondary ? PCI_SEC_STATUS : PCI_STATUS; - pci_read_config_word(dev, where, &status); - - /* If we get back 0xFFFF then we must suspect that the card has been - * pulled but the Linux PCI layer has not yet finished cleaning up. - * We don't want to report on such devices - */ - - if (status == 0xFFFF) { - u32 sanity; - - pci_read_config_dword(dev, 0, &sanity); - - if (sanity == 0xFFFFFFFF) - return 0; - } - - status &= PCI_STATUS_DETECTED_PARITY | PCI_STATUS_SIG_SYSTEM_ERROR | - PCI_STATUS_PARITY; - - if (status) - /* reset only the bits we are interested in */ - pci_write_config_word(dev, where, status); - - return status; -} - -typedef void (*pci_parity_check_fn_t) (struct pci_dev *dev); - -/* Clear any PCI parity errors logged by this device. */ -static void edac_pci_dev_parity_clear(struct pci_dev *dev) -{ - u8 header_type; - - get_pci_parity_status(dev, 0); - - /* read the device TYPE, looking for bridges */ - pci_read_config_byte(dev, PCI_HEADER_TYPE, &header_type); - - if ((header_type & 0x7F) == PCI_HEADER_TYPE_BRIDGE) - get_pci_parity_status(dev, 1); -} - -/* - * PCI Parity polling - * - */ -static void edac_pci_dev_parity_test(struct pci_dev *dev) -{ - u16 status; - u8 header_type; - - /* read the STATUS register on this device - */ - status = get_pci_parity_status(dev, 0); - - debugf2("PCI STATUS= 0x%04x %s\n", status, dev->dev.bus_id ); - - /* check the status reg for errors */ - if (status) { - if (status & (PCI_STATUS_SIG_SYSTEM_ERROR)) - edac_printk(KERN_CRIT, EDAC_PCI, - "Signaled System Error on %s\n", - pci_name(dev)); - - if (status & (PCI_STATUS_PARITY)) { - edac_printk(KERN_CRIT, EDAC_PCI, - "Master Data Parity Error on %s\n", - pci_name(dev)); - - atomic_inc(&pci_parity_count); - } - - if (status & (PCI_STATUS_DETECTED_PARITY)) { - edac_printk(KERN_CRIT, EDAC_PCI, - "Detected Parity Error on %s\n", - pci_name(dev)); - - atomic_inc(&pci_parity_count); - } - } - - /* read the device TYPE, looking for bridges */ - pci_read_config_byte(dev, PCI_HEADER_TYPE, &header_type); - - debugf2("PCI HEADER TYPE= 0x%02x %s\n", header_type, dev->dev.bus_id ); - - if ((header_type & 0x7F) == PCI_HEADER_TYPE_BRIDGE) { - /* On bridges, need to examine secondary status register */ - status = get_pci_parity_status(dev, 1); - - debugf2("PCI SEC_STATUS= 0x%04x %s\n", - status, dev->dev.bus_id ); - - /* check the secondary status reg for errors */ - if (status) { - if (status & (PCI_STATUS_SIG_SYSTEM_ERROR)) - edac_printk(KERN_CRIT, EDAC_PCI, "Bridge " - "Signaled System Error on %s\n", - pci_name(dev)); - - if (status & (PCI_STATUS_PARITY)) { - edac_printk(KERN_CRIT, EDAC_PCI, "Bridge " - "Master Data Parity Error on " - "%s\n", pci_name(dev)); - - atomic_inc(&pci_parity_count); - } - - if (status & (PCI_STATUS_DETECTED_PARITY)) { - edac_printk(KERN_CRIT, EDAC_PCI, "Bridge " - "Detected Parity Error on %s\n", - pci_name(dev)); - - atomic_inc(&pci_parity_count); - } - } - } -} - -/* - * pci_dev parity list iterator - * Scan the PCI device list for one iteration, looking for SERRORs - * Master Parity ERRORS or Parity ERRORs on primary or secondary devices - */ -static inline void edac_pci_dev_parity_iterator(pci_parity_check_fn_t fn) -{ - struct pci_dev *dev = NULL; - - /* request for kernel access to the next PCI device, if any, - * and while we are looking at it have its reference count - * bumped until we are done with it - */ - while((dev = pci_get_device(PCI_ANY_ID, PCI_ANY_ID, dev)) != NULL) { - fn(dev); - } -} - -static void do_pci_parity_check(void) -{ - unsigned long flags; - int before_count; - - debugf3("%s()\n", __func__); - - if (!check_pci_parity) - return; - - before_count = atomic_read(&pci_parity_count); - - /* scan all PCI devices looking for a Parity Error on devices and - * bridges - */ - local_irq_save(flags); - edac_pci_dev_parity_iterator(edac_pci_dev_parity_test); - local_irq_restore(flags); - - /* Only if operator has selected panic on PCI Error */ - if (panic_on_pci_parity) { - /* If the count is different 'after' from 'before' */ - if (before_count != atomic_read(&pci_parity_count)) - panic("EDAC: PCI Parity Error"); - } -} - -static inline void clear_pci_parity_errors(void) -{ - /* Clear any PCI bus parity errors that devices initially have logged - * in their registers. - */ - edac_pci_dev_parity_iterator(edac_pci_dev_parity_clear); -} - -#else /* CONFIG_PCI */ - -/* pre-process these away */ -#define do_pci_parity_check() -#define clear_pci_parity_errors() -#define edac_sysfs_pci_teardown() -#define edac_sysfs_pci_setup() (0) - -#endif /* CONFIG_PCI */ - -/* EDAC sysfs CSROW data structures and methods - */ - -/* Set of more default csrow<id> attribute show/store functions */ -static ssize_t csrow_ue_count_show(struct csrow_info *csrow, char *data, int private) -{ - return sprintf(data,"%u\n", csrow->ue_count); -} - -static ssize_t csrow_ce_count_show(struct csrow_info *csrow, char *data, int private) -{ - return sprintf(data,"%u\n", csrow->ce_count); -} - -static ssize_t csrow_size_show(struct csrow_info *csrow, char *data, int private) -{ - return sprintf(data,"%u\n", PAGES_TO_MiB(csrow->nr_pages)); -} - -static ssize_t csrow_mem_type_show(struct csrow_info *csrow, char *data, int private) -{ - return sprintf(data,"%s\n", mem_types[csrow->mtype]); -} - -static ssize_t csrow_dev_type_show(struct csrow_info *csrow, char *data, int private) -{ - return sprintf(data,"%s\n", dev_types[csrow->dtype]); -} - -static ssize_t csrow_edac_mode_show(struct csrow_info *csrow, char *data, int private) -{ - return sprintf(data,"%s\n", edac_caps[csrow->edac_mode]); -} - -/* show/store functions for DIMM Label attributes */ -static ssize_t channel_dimm_label_show(struct csrow_info *csrow, - char *data, int channel) -{ - return snprintf(data, EDAC_MC_LABEL_LEN,"%s", - csrow->channels[channel].label); -} - -static ssize_t channel_dimm_label_store(struct csrow_info *csrow, - const char *data, - size_t count, - int channel) -{ - ssize_t max_size = 0; - - max_size = min((ssize_t)count,(ssize_t)EDAC_MC_LABEL_LEN-1); - strncpy(csrow->channels[channel].label, data, max_size); - csrow->channels[channel].label[max_size] = '\0'; - - return max_size; -} - -/* show function for dynamic chX_ce_count attribute */ -static ssize_t channel_ce_count_show(struct csrow_info *csrow, - char *data, - int channel) -{ - return sprintf(data, "%u\n", csrow->channels[channel].ce_count); -} - -/* csrow specific attribute structure */ -struct csrowdev_attribute { - struct attribute attr; - ssize_t (*show)(struct csrow_info *,char *,int); - ssize_t (*store)(struct csrow_info *, const char *,size_t,int); - int private; -}; - -#define to_csrow(k) container_of(k, struct csrow_info, kobj) -#define to_csrowdev_attr(a) container_of(a, struct csrowdev_attribute, attr) - -/* Set of show/store higher level functions for default csrow attributes */ -static ssize_t csrowdev_show(struct kobject *kobj, - struct attribute *attr, - char *buffer) -{ - struct csrow_info *csrow = to_csrow(kobj); - struct csrowdev_attribute *csrowdev_attr = to_csrowdev_attr(attr); - - if (csrowdev_attr->show) - return csrowdev_attr->show(csrow, - buffer, - csrowdev_attr->private); - return -EIO; -} - -static ssize_t csrowdev_store(struct kobject *kobj, struct attribute *attr, - const char *buffer, size_t count) -{ - struct csrow_info *csrow = to_csrow(kobj); - struct csrowdev_attribute * csrowdev_attr = to_csrowdev_attr(attr); - - if (csrowdev_attr->store) - return csrowdev_attr->store(csrow, - buffer, - count, - csrowdev_attr->private); - return -EIO; -} - -static struct sysfs_ops csrowfs_ops = { - .show = csrowdev_show, - .store = csrowdev_store -}; - -#define CSROWDEV_ATTR(_name,_mode,_show,_store,_private) \ -struct csrowdev_attribute attr_##_name = { \ - .attr = {.name = __stringify(_name), .mode = _mode }, \ - .show = _show, \ - .store = _store, \ - .private = _private, \ -}; - -/* default cwrow<id>/attribute files */ -CSROWDEV_ATTR(size_mb,S_IRUGO,csrow_size_show,NULL,0); -CSROWDEV_ATTR(dev_type,S_IRUGO,csrow_dev_type_show,NULL,0); -CSROWDEV_ATTR(mem_type,S_IRUGO,csrow_mem_type_show,NULL,0); -CSROWDEV_ATTR(edac_mode,S_IRUGO,csrow_edac_mode_show,NULL,0); -CSROWDEV_ATTR(ue_count,S_IRUGO,csrow_ue_count_show,NULL,0); -CSROWDEV_ATTR(ce_count,S_IRUGO,csrow_ce_count_show,NULL,0); - -/* default attributes of the CSROW<id> object */ -static struct csrowdev_attribute *default_csrow_attr[] = { - &attr_dev_type, - &attr_mem_type, - &attr_edac_mode, - &attr_size_mb, - &attr_ue_count, - &attr_ce_count, - NULL, -}; - - -/* possible dynamic channel DIMM Label attribute files */ -CSROWDEV_ATTR(ch0_dimm_label,S_IRUGO|S_IWUSR, - channel_dimm_label_show, - channel_dimm_label_store, - 0 ); -CSROWDEV_ATTR(ch1_dimm_label,S_IRUGO|S_IWUSR, - channel_dimm_label_show, - channel_dimm_label_store, - 1 ); -CSROWDEV_ATTR(ch2_dimm_label,S_IRUGO|S_IWUSR, - channel_dimm_label_show, - channel_dimm_label_store, - 2 ); -CSROWDEV_ATTR(ch3_dimm_label,S_IRUGO|S_IWUSR, - channel_dimm_label_show, - channel_dimm_label_store, - 3 ); -CSROWDEV_ATTR(ch4_dimm_label,S_IRUGO|S_IWUSR, - channel_dimm_label_show, - channel_dimm_label_store, - 4 ); -CSROWDEV_ATTR(ch5_dimm_label,S_IRUGO|S_IWUSR, - channel_dimm_label_show, - channel_dimm_label_store, - 5 ); - -/* Total possible dynamic DIMM Label attribute file table */ -static struct csrowdev_attribute *dynamic_csrow_dimm_attr[] = { - &attr_ch0_dimm_label, - &attr_ch1_dimm_label, - &attr_ch2_dimm_label, - &attr_ch3_dimm_label, - &attr_ch4_dimm_label, - &attr_ch5_dimm_label -}; - -/* possible dynamic channel ce_count attribute files */ -CSROWDEV_ATTR(ch0_ce_count,S_IRUGO|S_IWUSR, - channel_ce_count_show, - NULL, - 0 ); -CSROWDEV_ATTR(ch1_ce_count,S_IRUGO|S_IWUSR, - channel_ce_count_show, - NULL, - 1 ); -CSROWDEV_ATTR(ch2_ce_count,S_IRUGO|S_IWUSR, - channel_ce_count_show, - NULL, - 2 ); -CSROWDEV_ATTR(ch3_ce_count,S_IRUGO|S_IWUSR, - channel_ce_count_show, - NULL, - 3 ); -CSROWDEV_ATTR(ch4_ce_count,S_IRUGO|S_IWUSR, - channel_ce_count_show, - NULL, - 4 ); -CSROWDEV_ATTR(ch5_ce_count,S_IRUGO|S_IWUSR, - channel_ce_count_show, - NULL, - 5 ); - -/* Total possible dynamic ce_count attribute file table */ -static struct csrowdev_attribute *dynamic_csrow_ce_count_attr[] = { - &attr_ch0_ce_count, - &attr_ch1_ce_count, - &attr_ch2_ce_count, - &attr_ch3_ce_count, - &attr_ch4_ce_count, - &attr_ch5_ce_count -}; - - -#define EDAC_NR_CHANNELS 6 - -/* Create dynamic CHANNEL files, indexed by 'chan', under specifed CSROW */ -static int edac_create_channel_files(struct kobject *kobj, int chan) -{ - int err=-ENODEV; - - if (chan >= EDAC_NR_CHANNELS) - return err; - - /* create the DIMM label attribute file */ - err = sysfs_create_file(kobj, - (struct attribute *) dynamic_csrow_dimm_attr[chan]); - - if (!err) { - /* create the CE Count attribute file */ - err = sysfs_create_file(kobj, - (struct attribute *) dynamic_csrow_ce_count_attr[chan]); - } else { - debugf1("%s() dimm labels and ce_count files created", __func__); - } - - return err; -} - -/* No memory to release for this kobj */ -static void edac_csrow_instance_release(struct kobject *kobj) -{ - struct csrow_info *cs; - - cs = container_of(kobj, struct csrow_info, kobj); - complete(&cs->kobj_complete); -} - -/* the kobj_type instance for a CSROW */ -static struct kobj_type ktype_csrow = { - .release = edac_csrow_instance_release, - .sysfs_ops = &csrowfs_ops, - .default_attrs = (struct attribute **) default_csrow_attr, -}; - -/* Create a CSROW object under specifed edac_mc_device */ -static int edac_create_csrow_object( - struct kobject *edac_mci_kobj, - struct csrow_info *csrow, - int index) -{ - int err = 0; - int chan; - - memset(&csrow->kobj, 0, sizeof(csrow->kobj)); - - /* generate ..../edac/mc/mc<id>/csrow<index> */ - - csrow->kobj.parent = edac_mci_kobj; - csrow->kobj.ktype = &ktype_csrow; - - /* name this instance of csrow<id> */ - err = kobject_set_name(&csrow->kobj,"csrow%d",index); - if (err) - goto error_exit; - - /* Instanstiate the csrow object */ - err = kobject_register(&csrow->kobj); - if (!err) { - /* Create the dyanmic attribute files on this csrow, - * namely, the DIMM labels and the channel ce_count - */ - for (chan = 0; chan < csrow->nr_channels; chan++) { - err = edac_create_channel_files(&csrow->kobj,chan); - if (err) - break; - } - } - -error_exit: - return err; -} - -/* default sysfs methods and data structures for the main MCI kobject */ - -static ssize_t mci_reset_counters_store(struct mem_ctl_info *mci, - const char *data, size_t count) -{ - int row, chan; - - mci->ue_noinfo_count = 0; - mci->ce_noinfo_count = 0; - mci->ue_count = 0; - mci->ce_count = 0; - - for (row = 0; row < mci->nr_csrows; row++) { - struct csrow_info *ri = &mci->csrows[row]; - - ri->ue_count = 0; - ri->ce_count = 0; - - for (chan = 0; chan < ri->nr_channels; chan++) - ri->channels[chan].ce_count = 0; - } - - mci->start_time = jiffies; - return count; -} - -/* memory scrubbing */ -static ssize_t mci_sdram_scrub_rate_store(struct mem_ctl_info *mci, - const char *data, size_t count) -{ - u32 bandwidth = -1; - - if (mci->set_sdram_scrub_rate) { - - memctrl_int_store(&bandwidth, data, count); - - if (!(*mci->set_sdram_scrub_rate)(mci, &bandwidth)) { - edac_printk(KERN_DEBUG, EDAC_MC, - "Scrub rate set successfully, applied: %d\n", - bandwidth); - } else { - /* FIXME: error codes maybe? */ - edac_printk(KERN_DEBUG, EDAC_MC, - "Scrub rate set FAILED, could not apply: %d\n", - bandwidth); - } - } else { - /* FIXME: produce "not implemented" ERROR for user-side. */ - edac_printk(KERN_WARNING, EDAC_MC, - "Memory scrubbing 'set'control is not implemented!\n"); - } - return count; -} - -static ssize_t mci_sdram_scrub_rate_show(struct mem_ctl_info *mci, char *data) -{ - u32 bandwidth = -1; - - if (mci->get_sdram_scrub_rate) { - if (!(*mci->get_sdram_scrub_rate)(mci, &bandwidth)) { - edac_printk(KERN_DEBUG, EDAC_MC, - "Scrub rate successfully, fetched: %d\n", - bandwidth); - } else { - /* FIXME: error codes maybe? */ - edac_printk(KERN_DEBUG, EDAC_MC, - "Scrub rate fetch FAILED, got: %d\n", - bandwidth); - } - } else { - /* FIXME: produce "not implemented" ERROR for user-side. */ - edac_printk(KERN_WARNING, EDAC_MC, - "Memory scrubbing 'get' control is not implemented!\n"); - } - return sprintf(data, "%d\n", bandwidth); -} - -/* default attribute files for the MCI object */ -static ssize_t mci_ue_count_show(struct mem_ctl_info *mci, char *data) -{ - return sprintf(data,"%d\n", mci->ue_count); -} - -static ssize_t mci_ce_count_show(struct mem_ctl_info *mci, char *data) -{ - return sprintf(data,"%d\n", mci->ce_count); -} - -static ssize_t mci_ce_noinfo_show(struct mem_ctl_info *mci, char *data) -{ - return sprintf(data,"%d\n", mci->ce_noinfo_count); -} - -static ssize_t mci_ue_noinfo_show(struct mem_ctl_info *mci, char *data) -{ - return sprintf(data,"%d\n", mci->ue_noinfo_count); -} - -static ssize_t mci_seconds_show(struct mem_ctl_info *mci, char *data) -{ - return sprintf(data,"%ld\n", (jiffies - mci->start_time) / HZ); -} - -static ssize_t mci_ctl_name_show(struct mem_ctl_info *mci, char *data) -{ - return sprintf(data,"%s\n", mci->ctl_name); -} - -static ssize_t mci_size_mb_show(struct mem_ctl_info *mci, char *data) -{ - int total_pages, csrow_idx; - - for (total_pages = csrow_idx = 0; csrow_idx < mci->nr_csrows; - csrow_idx++) { - struct csrow_info *csrow = &mci->csrows[csrow_idx]; - - if (!csrow->nr_pages) - continue; - - total_pages += csrow->nr_pages; - } - - return sprintf(data,"%u\n", PAGES_TO_MiB(total_pages)); -} - -struct mcidev_attribute { - struct attribute attr; - ssize_t (*show)(struct mem_ctl_info *,char *); - ssize_t (*store)(struct mem_ctl_info *, const char *,size_t); -}; - -#define to_mci(k) container_of(k, struct mem_ctl_info, edac_mci_kobj) -#define to_mcidev_attr(a) container_of(a, struct mcidev_attribute, attr) - -/* MCI show/store functions for top most object */ -static ssize_t mcidev_show(struct kobject *kobj, struct attribute *attr, - char *buffer) -{ - struct mem_ctl_info *mem_ctl_info = to_mci(kobj); - struct mcidev_attribute * mcidev_attr = to_mcidev_attr(attr); - - if (mcidev_attr->show) - return mcidev_attr->show(mem_ctl_info, buffer); - - return -EIO; -} - -static ssize_t mcidev_store(struct kobject *kobj, struct attribute *attr, - const char *buffer, size_t count) -{ - struct mem_ctl_info *mem_ctl_info = to_mci(kobj); - struct mcidev_attribute * mcidev_attr = to_mcidev_attr(attr); - - if (mcidev_attr->store) - return mcidev_attr->store(mem_ctl_info, buffer, count); - - return -EIO; -} - -static struct sysfs_ops mci_ops = { - .show = mcidev_show, - .store = mcidev_store -}; - -#define MCIDEV_ATTR(_name,_mode,_show,_store) \ -struct mcidev_attribute mci_attr_##_name = { \ - .attr = {.name = __stringify(_name), .mode = _mode }, \ - .show = _show, \ - .store = _store, \ -}; - -/* default Control file */ -MCIDEV_ATTR(reset_counters,S_IWUSR,NULL,mci_reset_counters_store); - -/* default Attribute files */ -MCIDEV_ATTR(mc_name,S_IRUGO,mci_ctl_name_show,NULL); -MCIDEV_ATTR(size_mb,S_IRUGO,mci_size_mb_show,NULL); -MCIDEV_ATTR(seconds_since_reset,S_IRUGO,mci_seconds_show,NULL); -MCIDEV_ATTR(ue_noinfo_count,S_IRUGO,mci_ue_noinfo_show,NULL); -MCIDEV_ATTR(ce_noinfo_count,S_IRUGO,mci_ce_noinfo_show,NULL); -MCIDEV_ATTR(ue_count,S_IRUGO,mci_ue_count_show,NULL); -MCIDEV_ATTR(ce_count,S_IRUGO,mci_ce_count_show,NULL); - -/* memory scrubber attribute file */ -MCIDEV_ATTR(sdram_scrub_rate,S_IRUGO|S_IWUSR,mci_sdram_scrub_rate_show,mci_sdram_scrub_rate_store); - -static struct mcidev_attribute *mci_attr[] = { - &mci_attr_reset_counters, - &mci_attr_mc_name, - &mci_attr_size_mb, - &mci_attr_seconds_since_reset, - &mci_attr_ue_noinfo_count, - &mci_attr_ce_noinfo_count, - &mci_attr_ue_count, - &mci_attr_ce_count, - &mci_attr_sdram_scrub_rate, - NULL -}; - -/* - * Release of a MC controlling instance - */ -static void edac_mci_instance_release(struct kobject *kobj) -{ - struct mem_ctl_info *mci; - - mci = to_mci(kobj); - debugf0("%s() idx=%d\n", __func__, mci->mc_idx); - complete(&mci->kobj_complete); -} - -static struct kobj_type ktype_mci = { - .release = edac_mci_instance_release, - .sysfs_ops = &mci_ops, - .default_attrs = (struct attribute **) mci_attr, -}; - - -#define EDAC_DEVICE_SYMLINK "device" - -/* - * Create a new Memory Controller kobject instance, - * mc<id> under the 'mc' directory - * - * Return: - * 0 Success - * !0 Failure - */ -static int edac_create_sysfs_mci_device(struct mem_ctl_info *mci) -{ - int i; - int err; - struct csrow_info *csrow; - struct kobject *edac_mci_kobj=&mci->edac_mci_kobj; - - debugf0("%s() idx=%d\n", __func__, mci->mc_idx); - memset(edac_mci_kobj, 0, sizeof(*edac_mci_kobj)); - - /* set the name of the mc<id> object */ - err = kobject_set_name(edac_mci_kobj,"mc%d",mci->mc_idx); - if (err) - return err; - - /* link to our parent the '..../edac/mc' object */ - edac_mci_kobj->parent = &edac_memctrl_kobj; - edac_mci_kobj->ktype = &ktype_mci; - - /* register the mc<id> kobject */ - err = kobject_register(edac_mci_kobj); - if (err) - return err; - - /* create a symlink for the device */ - err = sysfs_create_link(edac_mci_kobj, &mci->dev->kobj, - EDAC_DEVICE_SYMLINK); - if (err) - goto fail0; - - /* Make directories for each CSROW object - * under the mc<id> kobject - */ - for (i = 0; i < mci->nr_csrows; i++) { - csrow = &mci->csrows[i]; - - /* Only expose populated CSROWs */ - if (csrow->nr_pages > 0) { - err = edac_create_csrow_object(edac_mci_kobj,csrow,i); - if (err) - goto fail1; - } - } - - return 0; - - /* CSROW error: backout what has already been registered, */ -fail1: - for ( i--; i >= 0; i--) { - if (csrow->nr_pages > 0) { - init_completion(&csrow->kobj_complete); - kobject_unregister(&mci->csrows[i].kobj); - wait_for_completion(&csrow->kobj_complete); - } - } - -fail0: - init_completion(&mci->kobj_complete); - kobject_unregister(edac_mci_kobj); - wait_for_completion(&mci->kobj_complete); - return err; -} - -/* - * remove a Memory Controller instance - */ -static void edac_remove_sysfs_mci_device(struct mem_ctl_info *mci) -{ - int i; - - debugf0("%s()\n", __func__); - - /* remove all csrow kobjects */ - for (i = 0; i < mci->nr_csrows; i++) { - if (mci->csrows[i].nr_pages > 0) { - init_completion(&mci->csrows[i].kobj_complete); - kobject_unregister(&mci->csrows[i].kobj); - wait_for_completion(&mci->csrows[i].kobj_complete); - } - } - - sysfs_remove_link(&mci->edac_mci_kobj, EDAC_DEVICE_SYMLINK); - init_completion(&mci->kobj_complete); - kobject_unregister(&mci->edac_mci_kobj); - wait_for_completion(&mci->kobj_complete); -} - -/* END OF sysfs data and methods */ - #ifdef CONFIG_EDAC_DEBUG -void edac_mc_dump_channel(struct channel_info *chan) +static void edac_mc_dump_channel(struct channel_info *chan) { debugf4("\tchannel = %p\n", chan); debugf4("\tchannel->chan_idx = %d\n", chan->chan_idx); @@ -1228,25 +48,21 @@ void edac_mc_dump_channel(struct channel_info *chan) debugf4("\tchannel->label = '%s'\n", chan->label); debugf4("\tchannel->csrow = %p\n\n", chan->csrow); } -EXPORT_SYMBOL_GPL(edac_mc_dump_channel); -void edac_mc_dump_csrow(struct csrow_info *csrow) +static void edac_mc_dump_csrow(struct csrow_info *csrow) { debugf4("\tcsrow = %p\n", csrow); debugf4("\tcsrow->csrow_idx = %d\n", csrow->csrow_idx); - debugf4("\tcsrow->first_page = 0x%lx\n", - csrow->first_page); + debugf4("\tcsrow->first_page = 0x%lx\n", csrow->first_page); debugf4("\tcsrow->last_page = 0x%lx\n", csrow->last_page); debugf4("\tcsrow->page_mask = 0x%lx\n", csrow->page_mask); debugf4("\tcsrow->nr_pages = 0x%x\n", csrow->nr_pages); - debugf4("\tcsrow->nr_channels = %d\n", - csrow->nr_channels); + debugf4("\tcsrow->nr_channels = %d\n", csrow->nr_channels); debugf4("\tcsrow->channels = %p\n", csrow->channels); debugf4("\tcsrow->mci = %p\n\n", csrow->mci); } -EXPORT_SYMBOL_GPL(edac_mc_dump_csrow); -void edac_mc_dump_mci(struct mem_ctl_info *mci) +static void edac_mc_dump_mci(struct mem_ctl_info *mci) { debugf3("\tmci = %p\n", mci); debugf3("\tmci->mtype_cap = %lx\n", mci->mtype_cap); @@ -1256,13 +72,11 @@ void edac_mc_dump_mci(struct mem_ctl_info *mci) debugf3("\tmci->nr_csrows = %d, csrows = %p\n", mci->nr_csrows, mci->csrows); debugf3("\tdev = %p\n", mci->dev); - debugf3("\tmod_name:ctl_name = %s:%s\n", - mci->mod_name, mci->ctl_name); + debugf3("\tmod_name:ctl_name = %s:%s\n", mci->mod_name, mci->ctl_name); debugf3("\tpvt_info = %p\n\n", mci->pvt_info); } -EXPORT_SYMBOL_GPL(edac_mc_dump_mci); -#endif /* CONFIG_EDAC_DEBUG */ +#endif /* CONFIG_EDAC_DEBUG */ /* 'ptr' points to a possibly unaligned item X such that sizeof(X) is 'size'. * Adjust 'ptr' so that its alignment is at least as stringent as what the @@ -1271,7 +85,7 @@ EXPORT_SYMBOL_GPL(edac_mc_dump_mci); * If 'size' is a constant, the compiler will optimize this whole function * down to either a no-op or the addition of a constant to the value of 'ptr'. */ -static inline char * align_ptr(void *ptr, unsigned size) +void *edac_align_ptr(void *ptr, unsigned size) { unsigned align, r; @@ -1288,14 +102,14 @@ static inline char * align_ptr(void *ptr, unsigned size) else if (size > sizeof(char)) align = sizeof(short); else - return (char *) ptr; + return (char *)ptr; r = size % align; if (r == 0) - return (char *) ptr; + return (char *)ptr; - return (char *) (((unsigned long) ptr) + align - r); + return (void *)(((unsigned long)ptr) + align - r); } /** @@ -1315,7 +129,7 @@ static inline char * align_ptr(void *ptr, unsigned size) * struct mem_ctl_info pointer */ struct mem_ctl_info *edac_mc_alloc(unsigned sz_pvt, unsigned nr_csrows, - unsigned nr_chans) + unsigned nr_chans, int edac_index) { struct mem_ctl_info *mci; struct csrow_info *csi, *csrow; @@ -1323,30 +137,32 @@ struct mem_ctl_info *edac_mc_alloc(unsigned sz_pvt, unsigned nr_csrows, void *pvt; unsigned size; int row, chn; + int err; /* Figure out the offsets of the various items from the start of an mc * structure. We want the alignment of each item to be at least as * stringent as what the compiler would provide if we could simply * hardcode everything into a single struct. */ - mci = (struct mem_ctl_info *) 0; - csi = (struct csrow_info *)align_ptr(&mci[1], sizeof(*csi)); - chi = (struct channel_info *) - align_ptr(&csi[nr_csrows], sizeof(*chi)); - pvt = align_ptr(&chi[nr_chans * nr_csrows], sz_pvt); - size = ((unsigned long) pvt) + sz_pvt; - - if ((mci = kmalloc(size, GFP_KERNEL)) == NULL) + mci = (struct mem_ctl_info *)0; + csi = edac_align_ptr(&mci[1], sizeof(*csi)); + chi = edac_align_ptr(&csi[nr_csrows], sizeof(*chi)); + pvt = edac_align_ptr(&chi[nr_chans * nr_csrows], sz_pvt); + size = ((unsigned long)pvt) + sz_pvt; + + mci = kzalloc(size, GFP_KERNEL); + if (mci == NULL) return NULL; /* Adjust pointers so they point within the memory we just allocated * rather than an imaginary chunk of memory located at address 0. */ - csi = (struct csrow_info *) (((char *) mci) + ((unsigned long) csi)); - chi = (struct channel_info *) (((char *) mci) + ((unsigned long) chi)); - pvt = sz_pvt ? (((char *) mci) + ((unsigned long) pvt)) : NULL; + csi = (struct csrow_info *)(((char *)mci) + ((unsigned long)csi)); + chi = (struct channel_info *)(((char *)mci) + ((unsigned long)chi)); + pvt = sz_pvt ? (((char *)mci) + ((unsigned long)pvt)) : NULL; - memset(mci, 0, size); /* clear all fields */ + /* setup index and various internal pointers */ + mci->mc_idx = edac_index; mci->csrows = csi; mci->pvt_info = pvt; mci->nr_csrows = nr_csrows; @@ -1366,17 +182,35 @@ struct mem_ctl_info *edac_mc_alloc(unsigned sz_pvt, unsigned nr_csrows, } } + mci->op_state = OP_ALLOC; + + /* + * Initialize the 'root' kobj for the edac_mc controller + */ + err = edac_mc_register_sysfs_main_kobj(mci); + if (err) { + kfree(mci); + return NULL; + } + + /* at this point, the root kobj is valid, and in order to + * 'free' the object, then the function: + * edac_mc_unregister_sysfs_main_kobj() must be called + * which will perform kobj unregistration and the actual free + * will occur during the kobject callback operation + */ return mci; } EXPORT_SYMBOL_GPL(edac_mc_alloc); /** - * edac_mc_free: Free a previously allocated 'mci' structure + * edac_mc_free + * 'Free' a previously allocated 'mci' structure * @mci: pointer to a struct mem_ctl_info structure */ void edac_mc_free(struct mem_ctl_info *mci) { - kfree(mci); + edac_mc_unregister_sysfs_main_kobj(mci); } EXPORT_SYMBOL_GPL(edac_mc_free); @@ -1397,18 +231,136 @@ static struct mem_ctl_info *find_mci_by_dev(struct device *dev) return NULL; } +/* + * handler for EDAC to check if NMI type handler has asserted interrupt + */ +static int edac_mc_assert_error_check_and_clear(void) +{ + int old_state; + + if (edac_op_state == EDAC_OPSTATE_POLL) + return 1; + + old_state = edac_err_assert; + edac_err_assert = 0; + + return old_state; +} + +/* + * edac_mc_workq_function + * performs the operation scheduled by a workq request + */ +static void edac_mc_workq_function(struct work_struct *work_req) +{ + struct delayed_work *d_work = (struct delayed_work *)work_req; + struct mem_ctl_info *mci = to_edac_mem_ctl_work(d_work); + + mutex_lock(&mem_ctls_mutex); + + /* if this control struct has movd to offline state, we are done */ + if (mci->op_state == OP_OFFLINE) { + mutex_unlock(&mem_ctls_mutex); + return; + } + + /* Only poll controllers that are running polled and have a check */ + if (edac_mc_assert_error_check_and_clear() && (mci->edac_check != NULL)) + mci->edac_check(mci); + + /* + * FIXME: temp place holder for PCI checks, + * goes away when we break out PCI + */ + edac_pci_do_parity_check(); + + mutex_unlock(&mem_ctls_mutex); + + /* Reschedule */ + queue_delayed_work(edac_workqueue, &mci->work, + msecs_to_jiffies(edac_mc_get_poll_msec())); +} + +/* + * edac_mc_workq_setup + * initialize a workq item for this mci + * passing in the new delay period in msec + * + * locking model: + * + * called with the mem_ctls_mutex held + */ +static void edac_mc_workq_setup(struct mem_ctl_info *mci, unsigned msec) +{ + debugf0("%s()\n", __func__); + + /* if this instance is not in the POLL state, then simply return */ + if (mci->op_state != OP_RUNNING_POLL) + return; + + INIT_DELAYED_WORK(&mci->work, edac_mc_workq_function); + queue_delayed_work(edac_workqueue, &mci->work, msecs_to_jiffies(msec)); +} + +/* + * edac_mc_workq_teardown + * stop the workq processing on this mci + * + * locking model: + * + * called WITHOUT lock held + */ +static void edac_mc_workq_teardown(struct mem_ctl_info *mci) +{ + int status; + + /* if not running POLL, leave now */ + if (mci->op_state == OP_RUNNING_POLL) { + status = cancel_delayed_work(&mci->work); + if (status == 0) { + debugf0("%s() not canceled, flush the queue\n", + __func__); + + /* workq instance might be running, wait for it */ + flush_workqueue(edac_workqueue); + } + } +} + +/* + * edac_reset_delay_period + */ +static void edac_reset_delay_period(struct mem_ctl_info *mci, unsigned long value) +{ + /* cancel the current workq request */ + edac_mc_workq_teardown(mci); + + /* lock the list of devices for the new setup */ + mutex_lock(&mem_ctls_mutex); + + /* restart the workq request, with new delay value */ + edac_mc_workq_setup(mci, value); + + mutex_unlock(&mem_ctls_mutex); +} + /* Return 0 on success, 1 on failure. * Before calling this function, caller must * assign a unique value to mci->mc_idx. + * + * locking model: + * + * called with the mem_ctls_mutex lock held */ -static int add_mc_to_global_list (struct mem_ctl_info *mci) +static int add_mc_to_global_list(struct mem_ctl_info *mci) { struct list_head *item, *insert_before; struct mem_ctl_info *p; insert_before = &mc_devices; - if (unlikely((p = find_mci_by_dev(mci->dev)) != NULL)) + p = find_mci_by_dev(mci->dev); + if (unlikely(p != NULL)) goto fail0; list_for_each(item, &mc_devices) { @@ -1424,18 +376,19 @@ static int add_mc_to_global_list (struct mem_ctl_info *mci) } list_add_tail_rcu(&mci->link, insert_before); + atomic_inc(&edac_handlers); return 0; fail0: edac_printk(KERN_WARNING, EDAC_MC, - "%s (%s) %s %s already assigned %d\n", p->dev->bus_id, - dev_name(p->dev), p->mod_name, p->ctl_name, p->mc_idx); + "%s (%s) %s %s already assigned %d\n", p->dev->bus_id, + dev_name(mci), p->mod_name, p->ctl_name, p->mc_idx); return 1; fail1: edac_printk(KERN_WARNING, EDAC_MC, - "bug in low-level driver: attempt to assign\n" - " duplicate mc_idx %d in %s()\n", p->mc_idx, __func__); + "bug in low-level driver: attempt to assign\n" + " duplicate mc_idx %d in %s()\n", p->mc_idx, __func__); return 1; } @@ -1450,6 +403,7 @@ static void complete_mc_list_del(struct rcu_head *head) static void del_mc_from_global_list(struct mem_ctl_info *mci) { + atomic_dec(&edac_handlers); list_del_rcu(&mci->link); init_completion(&mci->complete); call_rcu(&mci->rcu, complete_mc_list_del); @@ -1457,6 +411,34 @@ static void del_mc_from_global_list(struct mem_ctl_info *mci) } /** + * edac_mc_find: Search for a mem_ctl_info structure whose index is 'idx'. + * + * If found, return a pointer to the structure. + * Else return NULL. + * + * Caller must hold mem_ctls_mutex. + */ +struct mem_ctl_info *edac_mc_find(int idx) +{ + struct list_head *item; + struct mem_ctl_info *mci; + + list_for_each(item, &mc_devices) { + mci = list_entry(item, struct mem_ctl_info, link); + + if (mci->mc_idx >= idx) { + if (mci->mc_idx == idx) + return mci; + + break; + } + } + + return NULL; +} +EXPORT_SYMBOL(edac_mc_find); + +/** * edac_mc_add_mc: Insert the 'mci' structure into the mci global list and * create sysfs entries associated with mci structure * @mci: pointer to the mci structure to be added to the list @@ -1468,10 +450,10 @@ static void del_mc_from_global_list(struct mem_ctl_info *mci) */ /* FIXME - should a warning be printed if no error detection? correction? */ -int edac_mc_add_mc(struct mem_ctl_info *mci, int mc_idx) +int edac_mc_add_mc(struct mem_ctl_info *mci) { debugf0("%s()\n", __func__); - mci->mc_idx = mc_idx; + #ifdef CONFIG_EDAC_DEBUG if (edac_debug_level >= 3) edac_mc_dump_mci(mci); @@ -1484,12 +466,12 @@ int edac_mc_add_mc(struct mem_ctl_info *mci, int mc_idx) edac_mc_dump_csrow(&mci->csrows[i]); for (j = 0; j < mci->csrows[i].nr_channels; j++) - edac_mc_dump_channel( - &mci->csrows[i].channels[j]); + edac_mc_dump_channel(&mci->csrows[i]. + channels[j]); } } #endif - down(&mem_ctls_mutex); + mutex_lock(&mem_ctls_mutex); if (add_mc_to_global_list(mci)) goto fail0; @@ -1503,18 +485,28 @@ int edac_mc_add_mc(struct mem_ctl_info *mci, int mc_idx) goto fail1; } + /* If there IS a check routine, then we are running POLLED */ + if (mci->edac_check != NULL) { + /* This instance is NOW RUNNING */ + mci->op_state = OP_RUNNING_POLL; + + edac_mc_workq_setup(mci, edac_mc_get_poll_msec()); + } else { + mci->op_state = OP_RUNNING_INTERRUPT; + } + /* Report action taken */ - edac_mc_printk(mci, KERN_INFO, "Giving out device to %s %s: DEV %s\n", - mci->mod_name, mci->ctl_name, dev_name(mci->dev)); + edac_mc_printk(mci, KERN_INFO, "Giving out device to '%s' '%s':" + " DEV %s\n", mci->mod_name, mci->ctl_name, dev_name(mci)); - up(&mem_ctls_mutex); + mutex_unlock(&mem_ctls_mutex); return 0; fail1: del_mc_from_global_list(mci); fail0: - up(&mem_ctls_mutex); + mutex_unlock(&mem_ctls_mutex); return 1; } EXPORT_SYMBOL_GPL(edac_mc_add_mc); @@ -1526,29 +518,41 @@ EXPORT_SYMBOL_GPL(edac_mc_add_mc); * * Return pointer to removed mci structure, or NULL if device not found. */ -struct mem_ctl_info * edac_mc_del_mc(struct device *dev) +struct mem_ctl_info *edac_mc_del_mc(struct device *dev) { struct mem_ctl_info *mci; - debugf0("MC: %s()\n", __func__); - down(&mem_ctls_mutex); + debugf0("%s()\n", __func__); + + mutex_lock(&mem_ctls_mutex); - if ((mci = find_mci_by_dev(dev)) == NULL) { - up(&mem_ctls_mutex); + /* find the requested mci struct in the global list */ + mci = find_mci_by_dev(dev); + if (mci == NULL) { + mutex_unlock(&mem_ctls_mutex); return NULL; } - edac_remove_sysfs_mci_device(mci); + /* marking MCI offline */ + mci->op_state = OP_OFFLINE; + del_mc_from_global_list(mci); - up(&mem_ctls_mutex); + mutex_unlock(&mem_ctls_mutex); + + /* flush workq processes and remove sysfs */ + edac_mc_workq_teardown(mci); + edac_remove_sysfs_mci_device(mci); + edac_printk(KERN_INFO, EDAC_MC, "Removed device %d for %s %s: DEV %s\n", mci->mc_idx, - mci->mod_name, mci->ctl_name, dev_name(mci->dev)); + mci->mod_name, mci->ctl_name, dev_name(mci)); + return mci; } EXPORT_SYMBOL_GPL(edac_mc_del_mc); -void edac_mc_scrub_block(unsigned long page, unsigned long offset, u32 size) +static void edac_mc_scrub_block(unsigned long page, unsigned long offset, + u32 size) { struct page *pg; void *virt_addr; @@ -1557,7 +561,7 @@ void edac_mc_scrub_block(unsigned long page, unsigned long offset, u32 size) debugf3("%s()\n", __func__); /* ECC error page was not in our memory. Ignore it. */ - if(!pfn_valid(page)) + if (!pfn_valid(page)) return; /* Find the actual page structure then map it and fix */ @@ -1577,7 +581,6 @@ void edac_mc_scrub_block(unsigned long page, unsigned long offset, u32 size) if (PageHighMem(pg)) local_irq_restore(flags); } -EXPORT_SYMBOL_GPL(edac_mc_scrub_block); /* FIXME - should return -1 */ int edac_mc_find_csrow_by_page(struct mem_ctl_info *mci, unsigned long page) @@ -1611,7 +614,7 @@ int edac_mc_find_csrow_by_page(struct mem_ctl_info *mci, unsigned long page) if (row == -1) edac_mc_printk(mci, KERN_ERR, "could not look up page error address %lx\n", - (unsigned long) page); + (unsigned long)page); return row; } @@ -1620,8 +623,9 @@ EXPORT_SYMBOL_GPL(edac_mc_find_csrow_by_page); /* FIXME - setable log (warning/emerg) levels */ /* FIXME - integrate with evlog: http://evlog.sourceforge.net/ */ void edac_mc_handle_ce(struct mem_ctl_info *mci, - unsigned long page_frame_number, unsigned long offset_in_page, - unsigned long syndrome, int row, int channel, const char *msg) + unsigned long page_frame_number, + unsigned long offset_in_page, unsigned long syndrome, + int row, int channel, const char *msg) { unsigned long remapped_page; @@ -1647,7 +651,7 @@ void edac_mc_handle_ce(struct mem_ctl_info *mci, return; } - if (log_ce) + if (edac_mc_get_log_ce()) /* FIXME - put in DIMM location */ edac_mc_printk(mci, KERN_WARNING, "CE page 0x%lx, offset 0x%lx, grain %d, syndrome " @@ -1671,18 +675,18 @@ void edac_mc_handle_ce(struct mem_ctl_info *mci, * page - which can then be scrubbed. */ remapped_page = mci->ctl_page_to_phys ? - mci->ctl_page_to_phys(mci, page_frame_number) : - page_frame_number; + mci->ctl_page_to_phys(mci, page_frame_number) : + page_frame_number; edac_mc_scrub_block(remapped_page, offset_in_page, - mci->csrows[row].grain); + mci->csrows[row].grain); } } EXPORT_SYMBOL_GPL(edac_mc_handle_ce); void edac_mc_handle_ce_no_info(struct mem_ctl_info *mci, const char *msg) { - if (log_ce) + if (edac_mc_get_log_ce()) edac_mc_printk(mci, KERN_WARNING, "CE - no information available: %s\n", msg); @@ -1692,8 +696,8 @@ void edac_mc_handle_ce_no_info(struct mem_ctl_info *mci, const char *msg) EXPORT_SYMBOL_GPL(edac_mc_handle_ce_no_info); void edac_mc_handle_ue(struct mem_ctl_info *mci, - unsigned long page_frame_number, unsigned long offset_in_page, - int row, const char *msg) + unsigned long page_frame_number, + unsigned long offset_in_page, int row, const char *msg) { int len = EDAC_MC_LABEL_LEN * 4; char labels[len + 1]; @@ -1714,26 +718,26 @@ void edac_mc_handle_ue(struct mem_ctl_info *mci, } chars = snprintf(pos, len + 1, "%s", - mci->csrows[row].channels[0].label); + mci->csrows[row].channels[0].label); len -= chars; pos += chars; for (chan = 1; (chan < mci->csrows[row].nr_channels) && (len > 0); - chan++) { + chan++) { chars = snprintf(pos, len + 1, ":%s", - mci->csrows[row].channels[chan].label); + mci->csrows[row].channels[chan].label); len -= chars; pos += chars; } - if (log_ue) + if (edac_mc_get_log_ue()) edac_mc_printk(mci, KERN_EMERG, "UE page 0x%lx, offset 0x%lx, grain %d, row %d, " "labels \"%s\": %s\n", page_frame_number, - offset_in_page, mci->csrows[row].grain, row, labels, - msg); + offset_in_page, mci->csrows[row].grain, row, + labels, msg); - if (panic_on_ue) + if (edac_mc_get_panic_on_ue()) panic("EDAC MC%d: UE page 0x%lx, offset 0x%lx, grain %d, " "row %d, labels \"%s\": %s\n", mci->mc_idx, page_frame_number, offset_in_page, @@ -1746,10 +750,10 @@ EXPORT_SYMBOL_GPL(edac_mc_handle_ue); void edac_mc_handle_ue_no_info(struct mem_ctl_info *mci, const char *msg) { - if (panic_on_ue) + if (edac_mc_get_panic_on_ue()) panic("EDAC MC%d: Uncorrected Error", mci->mc_idx); - if (log_ue) + if (edac_mc_get_log_ue()) edac_mc_printk(mci, KERN_WARNING, "UE - no information available: %s\n", msg); mci->ue_noinfo_count++; @@ -1757,16 +761,14 @@ void edac_mc_handle_ue_no_info(struct mem_ctl_info *mci, const char *msg) } EXPORT_SYMBOL_GPL(edac_mc_handle_ue_no_info); - /************************************************************* * On Fully Buffered DIMM modules, this help function is * called to process UE events */ void edac_mc_handle_fbd_ue(struct mem_ctl_info *mci, - unsigned int csrow, - unsigned int channela, - unsigned int channelb, - char *msg) + unsigned int csrow, + unsigned int channela, + unsigned int channelb, char *msg) { int len = EDAC_MC_LABEL_LEN * 4; char labels[len + 1]; @@ -1808,20 +810,21 @@ void edac_mc_handle_fbd_ue(struct mem_ctl_info *mci, /* Generate the DIMM labels from the specified channels */ chars = snprintf(pos, len + 1, "%s", mci->csrows[csrow].channels[channela].label); - len -= chars; pos += chars; + len -= chars; + pos += chars; chars = snprintf(pos, len + 1, "-%s", mci->csrows[csrow].channels[channelb].label); - if (log_ue) + if (edac_mc_get_log_ue()) edac_mc_printk(mci, KERN_EMERG, "UE row %d, channel-a= %d channel-b= %d " "labels \"%s\": %s\n", csrow, channela, channelb, labels, msg); - if (panic_on_ue) + if (edac_mc_get_panic_on_ue()) panic("UE row %d, channel-a= %d channel-b= %d " - "labels \"%s\": %s\n", csrow, channela, - channelb, labels, msg); + "labels \"%s\": %s\n", csrow, channela, + channelb, labels, msg); } EXPORT_SYMBOL(edac_mc_handle_fbd_ue); @@ -1830,9 +833,7 @@ EXPORT_SYMBOL(edac_mc_handle_fbd_ue); * called to process CE events */ void edac_mc_handle_fbd_ce(struct mem_ctl_info *mci, - unsigned int csrow, - unsigned int channel, - char *msg) + unsigned int csrow, unsigned int channel, char *msg) { /* Ensure boundary values */ @@ -1853,13 +854,12 @@ void edac_mc_handle_fbd_ce(struct mem_ctl_info *mci, return; } - if (log_ce) + if (edac_mc_get_log_ce()) /* FIXME - put in DIMM location */ edac_mc_printk(mci, KERN_WARNING, "CE row %d, channel %d, label \"%s\": %s\n", csrow, channel, - mci->csrows[csrow].channels[channel].label, - msg); + mci->csrows[csrow].channels[channel].label, msg); mci->ce_count++; mci->csrows[csrow].ce_count++; @@ -1867,17 +867,16 @@ void edac_mc_handle_fbd_ce(struct mem_ctl_info *mci, } EXPORT_SYMBOL(edac_mc_handle_fbd_ce); - /* * Iterate over all MC instances and check for ECC, et al, errors */ -static inline void check_mc_devices(void) +void edac_check_mc_devices(void) { struct list_head *item; struct mem_ctl_info *mci; debugf3("%s()\n", __func__); - down(&mem_ctls_mutex); + mutex_lock(&mem_ctls_mutex); list_for_each(item, &mc_devices) { mci = list_entry(item, struct mem_ctl_info, link); @@ -1886,120 +885,5 @@ static inline void check_mc_devices(void) mci->edac_check(mci); } - up(&mem_ctls_mutex); -} - -/* - * Check MC status every poll_msec. - * Check PCI status every poll_msec as well. - * - * This where the work gets done for edac. - * - * SMP safe, doesn't use NMI, and auto-rate-limits. - */ -static void do_edac_check(void) -{ - debugf3("%s()\n", __func__); - check_mc_devices(); - do_pci_parity_check(); -} - -static int edac_kernel_thread(void *arg) -{ - set_freezable(); - while (!kthread_should_stop()) { - do_edac_check(); - - /* goto sleep for the interval */ - schedule_timeout_interruptible((HZ * poll_msec) / 1000); - try_to_freeze(); - } - - return 0; + mutex_unlock(&mem_ctls_mutex); } - -/* - * edac_mc_init - * module initialization entry point - */ -static int __init edac_mc_init(void) -{ - edac_printk(KERN_INFO, EDAC_MC, EDAC_MC_VERSION "\n"); - - /* - * Harvest and clear any boot/initialization PCI parity errors - * - * FIXME: This only clears errors logged by devices present at time of - * module initialization. We should also do an initial clear - * of each newly hotplugged device. - */ - clear_pci_parity_errors(); - - /* Create the MC sysfs entries */ - if (edac_sysfs_memctrl_setup()) { - edac_printk(KERN_ERR, EDAC_MC, - "Error initializing sysfs code\n"); - return -ENODEV; - } - - /* Create the PCI parity sysfs entries */ - if (edac_sysfs_pci_setup()) { - edac_sysfs_memctrl_teardown(); - edac_printk(KERN_ERR, EDAC_MC, - "EDAC PCI: Error initializing sysfs code\n"); - return -ENODEV; - } - - /* create our kernel thread */ - edac_thread = kthread_run(edac_kernel_thread, NULL, "kedac"); - - if (IS_ERR(edac_thread)) { - /* remove the sysfs entries */ - edac_sysfs_memctrl_teardown(); - edac_sysfs_pci_teardown(); - return PTR_ERR(edac_thread); - } - - return 0; -} - -/* - * edac_mc_exit() - * module exit/termination functioni - */ -static void __exit edac_mc_exit(void) -{ - debugf0("%s()\n", __func__); - kthread_stop(edac_thread); - - /* tear down the sysfs device */ - edac_sysfs_memctrl_teardown(); - edac_sysfs_pci_teardown(); -} - -module_init(edac_mc_init); -module_exit(edac_mc_exit); - -MODULE_LICENSE("GPL"); -MODULE_AUTHOR("Linux Networx (http://lnxi.com) Thayne Harbaugh et al\n" - "Based on work by Dan Hollis et al"); -MODULE_DESCRIPTION("Core library routines for MC reporting"); - -module_param(panic_on_ue, int, 0644); -MODULE_PARM_DESC(panic_on_ue, "Panic on uncorrected error: 0=off 1=on"); -#ifdef CONFIG_PCI -module_param(check_pci_parity, int, 0644); -MODULE_PARM_DESC(check_pci_parity, "Check for PCI bus parity errors: 0=off 1=on"); -module_param(panic_on_pci_parity, int, 0644); -MODULE_PARM_DESC(panic_on_pci_parity, "Panic on PCI Bus Parity error: 0=off 1=on"); -#endif -module_param(log_ue, int, 0644); -MODULE_PARM_DESC(log_ue, "Log uncorrectable error to console: 0=off 1=on"); -module_param(log_ce, int, 0644); -MODULE_PARM_DESC(log_ce, "Log correctable error to console: 0=off 1=on"); -module_param(poll_msec, int, 0644); -MODULE_PARM_DESC(poll_msec, "Polling period in milliseconds"); -#ifdef CONFIG_EDAC_DEBUG -module_param(edac_debug_level, int, 0644); -MODULE_PARM_DESC(edac_debug_level, "Debug level"); -#endif diff --git a/drivers/edac/edac_mc_sysfs.c b/drivers/edac/edac_mc_sysfs.c new file mode 100644 index 00000000000..cd090b0677a --- /dev/null +++ b/drivers/edac/edac_mc_sysfs.c @@ -0,0 +1,1024 @@ +/* + * edac_mc kernel module + * (C) 2005-2007 Linux Networx (http://lnxi.com) + * + * This file may be distributed under the terms of the + * GNU General Public License. + * + * Written Doug Thompson <norsk5@xmission.com> www.softwarebitmaker.com + * + */ + +#include <linux/ctype.h> +#include <linux/bug.h> + +#include "edac_core.h" +#include "edac_module.h" + + +/* MC EDAC Controls, setable by module parameter, and sysfs */ +static int edac_mc_log_ue = 1; +static int edac_mc_log_ce = 1; +static int edac_mc_panic_on_ue; +static int edac_mc_poll_msec = 1000; + +/* Getter functions for above */ +int edac_mc_get_log_ue(void) +{ + return edac_mc_log_ue; +} + +int edac_mc_get_log_ce(void) +{ + return edac_mc_log_ce; +} + +int edac_mc_get_panic_on_ue(void) +{ + return edac_mc_panic_on_ue; +} + +/* this is temporary */ +int edac_mc_get_poll_msec(void) +{ + return edac_mc_poll_msec; +} + +/* Parameter declarations for above */ +module_param(edac_mc_panic_on_ue, int, 0644); +MODULE_PARM_DESC(edac_mc_panic_on_ue, "Panic on uncorrected error: 0=off 1=on"); +module_param(edac_mc_log_ue, int, 0644); +MODULE_PARM_DESC(edac_mc_log_ue, + "Log uncorrectable error to console: 0=off 1=on"); +module_param(edac_mc_log_ce, int, 0644); +MODULE_PARM_DESC(edac_mc_log_ce, + "Log correctable error to console: 0=off 1=on"); +module_param(edac_mc_poll_msec, int, 0644); +MODULE_PARM_DESC(edac_mc_poll_msec, "Polling period in milliseconds"); + +/* + * various constants for Memory Controllers + */ +static const char *mem_types[] = { + [MEM_EMPTY] = "Empty", + [MEM_RESERVED] = "Reserved", + [MEM_UNKNOWN] = "Unknown", + [MEM_FPM] = "FPM", + [MEM_EDO] = "EDO", + [MEM_BEDO] = "BEDO", + [MEM_SDR] = "Unbuffered-SDR", + [MEM_RDR] = "Registered-SDR", + [MEM_DDR] = "Unbuffered-DDR", + [MEM_RDDR] = "Registered-DDR", + [MEM_RMBS] = "RMBS", + [MEM_DDR2] = "Unbuffered-DDR2", + [MEM_FB_DDR2] = "FullyBuffered-DDR2", + [MEM_RDDR2] = "Registered-DDR2" +}; + +static const char *dev_types[] = { + [DEV_UNKNOWN] = "Unknown", + [DEV_X1] = "x1", + [DEV_X2] = "x2", + [DEV_X4] = "x4", + [DEV_X8] = "x8", + [DEV_X16] = "x16", + [DEV_X32] = "x32", + [DEV_X64] = "x64" +}; + +static const char *edac_caps[] = { + [EDAC_UNKNOWN] = "Unknown", + [EDAC_NONE] = "None", + [EDAC_RESERVED] = "Reserved", + [EDAC_PARITY] = "PARITY", + [EDAC_EC] = "EC", + [EDAC_SECDED] = "SECDED", + [EDAC_S2ECD2ED] = "S2ECD2ED", + [EDAC_S4ECD4ED] = "S4ECD4ED", + [EDAC_S8ECD8ED] = "S8ECD8ED", + [EDAC_S16ECD16ED] = "S16ECD16ED" +}; + + + +/* + * /sys/devices/system/edac/mc; + * data structures and methods + */ +static ssize_t memctrl_int_show(void *ptr, char *buffer) +{ + int *value = (int *)ptr; + return sprintf(buffer, "%u\n", *value); +} + +static ssize_t memctrl_int_store(void *ptr, const char *buffer, size_t count) +{ + int *value = (int *)ptr; + + if (isdigit(*buffer)) + *value = simple_strtoul(buffer, NULL, 0); + + return count; +} + + +/* EDAC sysfs CSROW data structures and methods + */ + +/* Set of more default csrow<id> attribute show/store functions */ +static ssize_t csrow_ue_count_show(struct csrow_info *csrow, char *data, + int private) +{ + return sprintf(data, "%u\n", csrow->ue_count); +} + +static ssize_t csrow_ce_count_show(struct csrow_info *csrow, char *data, + int private) +{ + return sprintf(data, "%u\n", csrow->ce_count); +} + +static ssize_t csrow_size_show(struct csrow_info *csrow, char *data, + int private) +{ + return sprintf(data, "%u\n", PAGES_TO_MiB(csrow->nr_pages)); +} + +static ssize_t csrow_mem_type_show(struct csrow_info *csrow, char *data, + int private) +{ + return sprintf(data, "%s\n", mem_types[csrow->mtype]); +} + +static ssize_t csrow_dev_type_show(struct csrow_info *csrow, char *data, + int private) +{ + return sprintf(data, "%s\n", dev_types[csrow->dtype]); +} + +static ssize_t csrow_edac_mode_show(struct csrow_info *csrow, char *data, + int private) +{ + return sprintf(data, "%s\n", edac_caps[csrow->edac_mode]); +} + +/* show/store functions for DIMM Label attributes */ +static ssize_t channel_dimm_label_show(struct csrow_info *csrow, + char *data, int channel) +{ + return snprintf(data, EDAC_MC_LABEL_LEN, "%s", + csrow->channels[channel].label); +} + +static ssize_t channel_dimm_label_store(struct csrow_info *csrow, + const char *data, + size_t count, int channel) +{ + ssize_t max_size = 0; + + max_size = min((ssize_t) count, (ssize_t) EDAC_MC_LABEL_LEN - 1); + strncpy(csrow->channels[channel].label, data, max_size); + csrow->channels[channel].label[max_size] = '\0'; + + return max_size; +} + +/* show function for dynamic chX_ce_count attribute */ +static ssize_t channel_ce_count_show(struct csrow_info *csrow, + char *data, int channel) +{ + return sprintf(data, "%u\n", csrow->channels[channel].ce_count); +} + +/* csrow specific attribute structure */ +struct csrowdev_attribute { + struct attribute attr; + ssize_t(*show) (struct csrow_info *, char *, int); + ssize_t(*store) (struct csrow_info *, const char *, size_t, int); + int private; +}; + +#define to_csrow(k) container_of(k, struct csrow_info, kobj) +#define to_csrowdev_attr(a) container_of(a, struct csrowdev_attribute, attr) + +/* Set of show/store higher level functions for default csrow attributes */ +static ssize_t csrowdev_show(struct kobject *kobj, + struct attribute *attr, char *buffer) +{ + struct csrow_info *csrow = to_csrow(kobj); + struct csrowdev_attribute *csrowdev_attr = to_csrowdev_attr(attr); + + if (csrowdev_attr->show) + return csrowdev_attr->show(csrow, + buffer, csrowdev_attr->private); + return -EIO; +} + +static ssize_t csrowdev_store(struct kobject *kobj, struct attribute *attr, + const char *buffer, size_t count) +{ + struct csrow_info *csrow = to_csrow(kobj); + struct csrowdev_attribute *csrowdev_attr = to_csrowdev_attr(attr); + + if (csrowdev_attr->store) + return csrowdev_attr->store(csrow, + buffer, + count, csrowdev_attr->private); + return -EIO; +} + +static struct sysfs_ops csrowfs_ops = { + .show = csrowdev_show, + .store = csrowdev_store +}; + +#define CSROWDEV_ATTR(_name,_mode,_show,_store,_private) \ +static struct csrowdev_attribute attr_##_name = { \ + .attr = {.name = __stringify(_name), .mode = _mode }, \ + .show = _show, \ + .store = _store, \ + .private = _private, \ +}; + +/* default cwrow<id>/attribute files */ +CSROWDEV_ATTR(size_mb, S_IRUGO, csrow_size_show, NULL, 0); +CSROWDEV_ATTR(dev_type, S_IRUGO, csrow_dev_type_show, NULL, 0); +CSROWDEV_ATTR(mem_type, S_IRUGO, csrow_mem_type_show, NULL, 0); +CSROWDEV_ATTR(edac_mode, S_IRUGO, csrow_edac_mode_show, NULL, 0); +CSROWDEV_ATTR(ue_count, S_IRUGO, csrow_ue_count_show, NULL, 0); +CSROWDEV_ATTR(ce_count, S_IRUGO, csrow_ce_count_show, NULL, 0); + +/* default attributes of the CSROW<id> object */ +static struct csrowdev_attribute *default_csrow_attr[] = { + &attr_dev_type, + &attr_mem_type, + &attr_edac_mode, + &attr_size_mb, + &attr_ue_count, + &attr_ce_count, + NULL, +}; + +/* possible dynamic channel DIMM Label attribute files */ +CSROWDEV_ATTR(ch0_dimm_label, S_IRUGO | S_IWUSR, + channel_dimm_label_show, channel_dimm_label_store, 0); +CSROWDEV_ATTR(ch1_dimm_label, S_IRUGO | S_IWUSR, + channel_dimm_label_show, channel_dimm_label_store, 1); +CSROWDEV_ATTR(ch2_dimm_label, S_IRUGO | S_IWUSR, + channel_dimm_label_show, channel_dimm_label_store, 2); +CSROWDEV_ATTR(ch3_dimm_label, S_IRUGO | S_IWUSR, + channel_dimm_label_show, channel_dimm_label_store, 3); +CSROWDEV_ATTR(ch4_dimm_label, S_IRUGO | S_IWUSR, + channel_dimm_label_show, channel_dimm_label_store, 4); +CSROWDEV_ATTR(ch5_dimm_label, S_IRUGO | S_IWUSR, + channel_dimm_label_show, channel_dimm_label_store, 5); + +/* Total possible dynamic DIMM Label attribute file table */ +static struct csrowdev_attribute *dynamic_csrow_dimm_attr[] = { + &attr_ch0_dimm_label, + &attr_ch1_dimm_label, + &attr_ch2_dimm_label, + &attr_ch3_dimm_label, + &attr_ch4_dimm_label, + &attr_ch5_dimm_label +}; + +/* possible dynamic channel ce_count attribute files */ +CSROWDEV_ATTR(ch0_ce_count, S_IRUGO | S_IWUSR, channel_ce_count_show, NULL, 0); +CSROWDEV_ATTR(ch1_ce_count, S_IRUGO | S_IWUSR, channel_ce_count_show, NULL, 1); +CSROWDEV_ATTR(ch2_ce_count, S_IRUGO | S_IWUSR, channel_ce_count_show, NULL, 2); +CSROWDEV_ATTR(ch3_ce_count, S_IRUGO | S_IWUSR, channel_ce_count_show, NULL, 3); +CSROWDEV_ATTR(ch4_ce_count, S_IRUGO | S_IWUSR, channel_ce_count_show, NULL, 4); +CSROWDEV_ATTR(ch5_ce_count, S_IRUGO | S_IWUSR, channel_ce_count_show, NULL, 5); + +/* Total possible dynamic ce_count attribute file table */ +static struct csrowdev_attribute *dynamic_csrow_ce_count_attr[] = { + &attr_ch0_ce_count, + &attr_ch1_ce_count, + &attr_ch2_ce_count, + &attr_ch3_ce_count, + &attr_ch4_ce_count, + &attr_ch5_ce_count +}; + +#define EDAC_NR_CHANNELS 6 + +/* Create dynamic CHANNEL files, indexed by 'chan', under specifed CSROW */ +static int edac_create_channel_files(struct kobject *kobj, int chan) +{ + int err = -ENODEV; + + if (chan >= EDAC_NR_CHANNELS) + return err; + + /* create the DIMM label attribute file */ + err = sysfs_create_file(kobj, + (struct attribute *) + dynamic_csrow_dimm_attr[chan]); + + if (!err) { + /* create the CE Count attribute file */ + err = sysfs_create_file(kobj, + (struct attribute *) + dynamic_csrow_ce_count_attr[chan]); + } else { + debugf1("%s() dimm labels and ce_count files created", + __func__); + } + + return err; +} + +/* No memory to release for this kobj */ +static void edac_csrow_instance_release(struct kobject *kobj) +{ + struct mem_ctl_info *mci; + struct csrow_info *cs; + + debugf1("%s()\n", __func__); + + cs = container_of(kobj, struct csrow_info, kobj); + mci = cs->mci; + + kobject_put(&mci->edac_mci_kobj); +} + +/* the kobj_type instance for a CSROW */ +static struct kobj_type ktype_csrow = { + .release = edac_csrow_instance_release, + .sysfs_ops = &csrowfs_ops, + .default_attrs = (struct attribute **)default_csrow_attr, +}; + +/* Create a CSROW object under specifed edac_mc_device */ +static int edac_create_csrow_object(struct mem_ctl_info *mci, + struct csrow_info *csrow, int index) +{ + struct kobject *kobj_mci = &mci->edac_mci_kobj; + struct kobject *kobj; + int chan; + int err; + + /* generate ..../edac/mc/mc<id>/csrow<index> */ + memset(&csrow->kobj, 0, sizeof(csrow->kobj)); + csrow->mci = mci; /* include container up link */ + csrow->kobj.parent = kobj_mci; + csrow->kobj.ktype = &ktype_csrow; + + /* name this instance of csrow<id> */ + err = kobject_set_name(&csrow->kobj, "csrow%d", index); + if (err) + goto err_out; + + /* bump the mci instance's kobject's ref count */ + kobj = kobject_get(&mci->edac_mci_kobj); + if (!kobj) { + err = -ENODEV; + goto err_out; + } + + /* Instanstiate the csrow object */ + err = kobject_register(&csrow->kobj); + if (err) + goto err_release_top_kobj; + + /* At this point, to release a csrow kobj, one must + * call the kobject_unregister and allow that tear down + * to work the releasing + */ + + /* Create the dyanmic attribute files on this csrow, + * namely, the DIMM labels and the channel ce_count + */ + for (chan = 0; chan < csrow->nr_channels; chan++) { + err = edac_create_channel_files(&csrow->kobj, chan); + if (err) { + /* special case the unregister here */ + kobject_unregister(&csrow->kobj); + goto err_out; + } + } + + return 0; + + /* error unwind stack */ +err_release_top_kobj: + kobject_put(&mci->edac_mci_kobj); + +err_out: + return err; +} + +/* default sysfs methods and data structures for the main MCI kobject */ + +static ssize_t mci_reset_counters_store(struct mem_ctl_info *mci, + const char *data, size_t count) +{ + int row, chan; + + mci->ue_noinfo_count = 0; + mci->ce_noinfo_count = 0; + mci->ue_count = 0; + mci->ce_count = 0; + + for (row = 0; row < mci->nr_csrows; row++) { + struct csrow_info *ri = &mci->csrows[row]; + + ri->ue_count = 0; + ri->ce_count = 0; + + for (chan = 0; chan < ri->nr_channels; chan++) + ri->channels[chan].ce_count = 0; + } + + mci->start_time = jiffies; + return count; +} + +/* memory scrubbing */ +static ssize_t mci_sdram_scrub_rate_store(struct mem_ctl_info *mci, + const char *data, size_t count) +{ + u32 bandwidth = -1; + + if (mci->set_sdram_scrub_rate) { + + memctrl_int_store(&bandwidth, data, count); + + if (!(*mci->set_sdram_scrub_rate) (mci, &bandwidth)) { + edac_printk(KERN_DEBUG, EDAC_MC, + "Scrub rate set successfully, applied: %d\n", + bandwidth); + } else { + /* FIXME: error codes maybe? */ + edac_printk(KERN_DEBUG, EDAC_MC, + "Scrub rate set FAILED, could not apply: %d\n", + bandwidth); + } + } else { + /* FIXME: produce "not implemented" ERROR for user-side. */ + edac_printk(KERN_WARNING, EDAC_MC, + "Memory scrubbing 'set'control is not implemented!\n"); + } + return count; +} + +static ssize_t mci_sdram_scrub_rate_show(struct mem_ctl_info *mci, char *data) +{ + u32 bandwidth = -1; + + if (mci->get_sdram_scrub_rate) { + if (!(*mci->get_sdram_scrub_rate) (mci, &bandwidth)) { + edac_printk(KERN_DEBUG, EDAC_MC, + "Scrub rate successfully, fetched: %d\n", + bandwidth); + } else { + /* FIXME: error codes maybe? */ + edac_printk(KERN_DEBUG, EDAC_MC, + "Scrub rate fetch FAILED, got: %d\n", + bandwidth); + } + } else { + /* FIXME: produce "not implemented" ERROR for user-side. */ + edac_printk(KERN_WARNING, EDAC_MC, + "Memory scrubbing 'get' control is not implemented\n"); + } + return sprintf(data, "%d\n", bandwidth); +} + +/* default attribute files for the MCI object */ +static ssize_t mci_ue_count_show(struct mem_ctl_info *mci, char *data) +{ + return sprintf(data, "%d\n", mci->ue_count); +} + +static ssize_t mci_ce_count_show(struct mem_ctl_info *mci, char *data) +{ + return sprintf(data, "%d\n", mci->ce_count); +} + +static ssize_t mci_ce_noinfo_show(struct mem_ctl_info *mci, char *data) +{ + return sprintf(data, "%d\n", mci->ce_noinfo_count); +} + +static ssize_t mci_ue_noinfo_show(struct mem_ctl_info *mci, char *data) +{ + return sprintf(data, "%d\n", mci->ue_noinfo_count); +} + +static ssize_t mci_seconds_show(struct mem_ctl_info *mci, char *data) +{ + return sprintf(data, "%ld\n", (jiffies - mci->start_time) / HZ); +} + +static ssize_t mci_ctl_name_show(struct mem_ctl_info *mci, char *data) +{ + return sprintf(data, "%s\n", mci->ctl_name); +} + +static ssize_t mci_size_mb_show(struct mem_ctl_info *mci, char *data) +{ + int total_pages, csrow_idx; + + for (total_pages = csrow_idx = 0; csrow_idx < mci->nr_csrows; + csrow_idx++) { + struct csrow_info *csrow = &mci->csrows[csrow_idx]; + + if (!csrow->nr_pages) + continue; + + total_pages += csrow->nr_pages; + } + + return sprintf(data, "%u\n", PAGES_TO_MiB(total_pages)); +} + +#define to_mci(k) container_of(k, struct mem_ctl_info, edac_mci_kobj) +#define to_mcidev_attr(a) container_of(a,struct mcidev_sysfs_attribute,attr) + +/* MCI show/store functions for top most object */ +static ssize_t mcidev_show(struct kobject *kobj, struct attribute *attr, + char *buffer) +{ + struct mem_ctl_info *mem_ctl_info = to_mci(kobj); + struct mcidev_sysfs_attribute *mcidev_attr = to_mcidev_attr(attr); + + if (mcidev_attr->show) + return mcidev_attr->show(mem_ctl_info, buffer); + + return -EIO; +} + +static ssize_t mcidev_store(struct kobject *kobj, struct attribute *attr, + const char *buffer, size_t count) +{ + struct mem_ctl_info *mem_ctl_info = to_mci(kobj); + struct mcidev_sysfs_attribute *mcidev_attr = to_mcidev_attr(attr); + + if (mcidev_attr->store) + return mcidev_attr->store(mem_ctl_info, buffer, count); + + return -EIO; +} + +/* Intermediate show/store table */ +static struct sysfs_ops mci_ops = { + .show = mcidev_show, + .store = mcidev_store +}; + +#define MCIDEV_ATTR(_name,_mode,_show,_store) \ +static struct mcidev_sysfs_attribute mci_attr_##_name = { \ + .attr = {.name = __stringify(_name), .mode = _mode }, \ + .show = _show, \ + .store = _store, \ +}; + +/* default Control file */ +MCIDEV_ATTR(reset_counters, S_IWUSR, NULL, mci_reset_counters_store); + +/* default Attribute files */ +MCIDEV_ATTR(mc_name, S_IRUGO, mci_ctl_name_show, NULL); +MCIDEV_ATTR(size_mb, S_IRUGO, mci_size_mb_show, NULL); +MCIDEV_ATTR(seconds_since_reset, S_IRUGO, mci_seconds_show, NULL); +MCIDEV_ATTR(ue_noinfo_count, S_IRUGO, mci_ue_noinfo_show, NULL); +MCIDEV_ATTR(ce_noinfo_count, S_IRUGO, mci_ce_noinfo_show, NULL); +MCIDEV_ATTR(ue_count, S_IRUGO, mci_ue_count_show, NULL); +MCIDEV_ATTR(ce_count, S_IRUGO, mci_ce_count_show, NULL); + +/* memory scrubber attribute file */ +MCIDEV_ATTR(sdram_scrub_rate, S_IRUGO | S_IWUSR, mci_sdram_scrub_rate_show, + mci_sdram_scrub_rate_store); + +static struct mcidev_sysfs_attribute *mci_attr[] = { + &mci_attr_reset_counters, + &mci_attr_mc_name, + &mci_attr_size_mb, + &mci_attr_seconds_since_reset, + &mci_attr_ue_noinfo_count, + &mci_attr_ce_noinfo_count, + &mci_attr_ue_count, + &mci_attr_ce_count, + &mci_attr_sdram_scrub_rate, + NULL +}; + + +/* + * Release of a MC controlling instance + * + * each MC control instance has the following resources upon entry: + * a) a ref count on the top memctl kobj + * b) a ref count on this module + * + * this function must decrement those ref counts and then + * issue a free on the instance's memory + */ +static void edac_mci_control_release(struct kobject *kobj) +{ + struct mem_ctl_info *mci; + + mci = to_mci(kobj); + + debugf0("%s() mci instance idx=%d releasing\n", __func__, mci->mc_idx); + + /* decrement the module ref count */ + module_put(mci->owner); + + /* free the mci instance memory here */ + kfree(mci); +} + +static struct kobj_type ktype_mci = { + .release = edac_mci_control_release, + .sysfs_ops = &mci_ops, + .default_attrs = (struct attribute **)mci_attr, +}; + +/* show/store, tables, etc for the MC kset */ + + +struct memctrl_dev_attribute { + struct attribute attr; + void *value; + ssize_t(*show) (void *, char *); + ssize_t(*store) (void *, const char *, size_t); +}; + +/* Set of show/store abstract level functions for memory control object */ +static ssize_t memctrl_dev_show(struct kobject *kobj, + struct attribute *attr, char *buffer) +{ + struct memctrl_dev_attribute *memctrl_dev; + memctrl_dev = (struct memctrl_dev_attribute *)attr; + + if (memctrl_dev->show) + return memctrl_dev->show(memctrl_dev->value, buffer); + + return -EIO; +} + +static ssize_t memctrl_dev_store(struct kobject *kobj, struct attribute *attr, + const char *buffer, size_t count) +{ + struct memctrl_dev_attribute *memctrl_dev; + memctrl_dev = (struct memctrl_dev_attribute *)attr; + + if (memctrl_dev->store) + return memctrl_dev->store(memctrl_dev->value, buffer, count); + + return -EIO; +} + +static struct sysfs_ops memctrlfs_ops = { + .show = memctrl_dev_show, + .store = memctrl_dev_store +}; + +#define MEMCTRL_ATTR(_name, _mode, _show, _store) \ +static struct memctrl_dev_attribute attr_##_name = { \ + .attr = {.name = __stringify(_name), .mode = _mode }, \ + .value = &_name, \ + .show = _show, \ + .store = _store, \ +}; + +#define MEMCTRL_STRING_ATTR(_name, _data, _mode, _show, _store) \ +static struct memctrl_dev_attribute attr_##_name = { \ + .attr = {.name = __stringify(_name), .mode = _mode }, \ + .value = _data, \ + .show = _show, \ + .store = _store, \ +}; + +/* csrow<id> control files */ +MEMCTRL_ATTR(edac_mc_panic_on_ue, + S_IRUGO | S_IWUSR, memctrl_int_show, memctrl_int_store); + +MEMCTRL_ATTR(edac_mc_log_ue, + S_IRUGO | S_IWUSR, memctrl_int_show, memctrl_int_store); + +MEMCTRL_ATTR(edac_mc_log_ce, + S_IRUGO | S_IWUSR, memctrl_int_show, memctrl_int_store); + +MEMCTRL_ATTR(edac_mc_poll_msec, + S_IRUGO | S_IWUSR, memctrl_int_show, memctrl_int_store); + +/* Base Attributes of the memory ECC object */ +static struct memctrl_dev_attribute *memctrl_attr[] = { + &attr_edac_mc_panic_on_ue, + &attr_edac_mc_log_ue, + &attr_edac_mc_log_ce, + &attr_edac_mc_poll_msec, + NULL, +}; + + +/* the ktype for the mc_kset internal kobj */ +static struct kobj_type ktype_mc_set_attribs = { + .sysfs_ops = &memctrlfs_ops, + .default_attrs = (struct attribute **)memctrl_attr, +}; + +/* EDAC memory controller sysfs kset: + * /sys/devices/system/edac/mc + */ +static struct kset mc_kset = { + .kobj = {.name = "mc", .ktype = &ktype_mc_set_attribs }, + .ktype = &ktype_mci, +}; + + +/* + * edac_mc_register_sysfs_main_kobj + * + * setups and registers the main kobject for each mci + */ +int edac_mc_register_sysfs_main_kobj(struct mem_ctl_info *mci) +{ + struct kobject *kobj_mci; + int err; + + debugf1("%s()\n", __func__); + + kobj_mci = &mci->edac_mci_kobj; + + /* Init the mci's kobject */ + memset(kobj_mci, 0, sizeof(*kobj_mci)); + + /* this instance become part of the mc_kset */ + kobj_mci->kset = &mc_kset; + + /* set the name of the mc<id> object */ + err = kobject_set_name(kobj_mci, "mc%d", mci->mc_idx); + if (err) + goto fail_out; + + /* Record which module 'owns' this control structure + * and bump the ref count of the module + */ + mci->owner = THIS_MODULE; + + /* bump ref count on this module */ + if (!try_module_get(mci->owner)) { + err = -ENODEV; + goto fail_out; + } + + /* register the mc<id> kobject to the mc_kset */ + err = kobject_register(kobj_mci); + if (err) { + debugf1("%s()Failed to register '.../edac/mc%d'\n", + __func__, mci->mc_idx); + goto kobj_reg_fail; + } + + /* At this point, to 'free' the control struct, + * edac_mc_unregister_sysfs_main_kobj() must be used + */ + + debugf1("%s() Registered '.../edac/mc%d' kobject\n", + __func__, mci->mc_idx); + + return 0; + + /* Error exit stack */ + +kobj_reg_fail: + module_put(mci->owner); + +fail_out: + return err; +} + +/* + * edac_mc_register_sysfs_main_kobj + * + * tears down and the main mci kobject from the mc_kset + */ +void edac_mc_unregister_sysfs_main_kobj(struct mem_ctl_info *mci) +{ + /* delete the kobj from the mc_kset */ + kobject_unregister(&mci->edac_mci_kobj); +} + +#define EDAC_DEVICE_SYMLINK "device" + +/* + * edac_create_mci_instance_attributes + * create MC driver specific attributes at the topmost level + * directory of this mci instance. + */ +static int edac_create_mci_instance_attributes(struct mem_ctl_info *mci) +{ + int err; + struct mcidev_sysfs_attribute *sysfs_attrib; + + /* point to the start of the array and iterate over it + * adding each attribute listed to this mci instance's kobject + */ + sysfs_attrib = mci->mc_driver_sysfs_attributes; + + while (sysfs_attrib && sysfs_attrib->attr.name) { + err = sysfs_create_file(&mci->edac_mci_kobj, + (struct attribute*) sysfs_attrib); + if (err) { + return err; + } + + sysfs_attrib++; + } + + return 0; +} + +/* + * edac_remove_mci_instance_attributes + * remove MC driver specific attributes at the topmost level + * directory of this mci instance. + */ +static void edac_remove_mci_instance_attributes(struct mem_ctl_info *mci) +{ + struct mcidev_sysfs_attribute *sysfs_attrib; + + /* point to the start of the array and iterate over it + * adding each attribute listed to this mci instance's kobject + */ + sysfs_attrib = mci->mc_driver_sysfs_attributes; + + /* loop if there are attributes and until we hit a NULL entry */ + while (sysfs_attrib && sysfs_attrib->attr.name) { + sysfs_remove_file(&mci->edac_mci_kobj, + (struct attribute *) sysfs_attrib); + sysfs_attrib++; + } +} + + +/* + * Create a new Memory Controller kobject instance, + * mc<id> under the 'mc' directory + * + * Return: + * 0 Success + * !0 Failure + */ +int edac_create_sysfs_mci_device(struct mem_ctl_info *mci) +{ + int i; + int err; + struct csrow_info *csrow; + struct kobject *kobj_mci = &mci->edac_mci_kobj; + + debugf0("%s() idx=%d\n", __func__, mci->mc_idx); + + /* create a symlink for the device */ + err = sysfs_create_link(kobj_mci, &mci->dev->kobj, + EDAC_DEVICE_SYMLINK); + if (err) { + debugf1("%s() failure to create symlink\n", __func__); + goto fail0; + } + + /* If the low level driver desires some attributes, + * then create them now for the driver. + */ + if (mci->mc_driver_sysfs_attributes) { + err = edac_create_mci_instance_attributes(mci); + if (err) { + debugf1("%s() failure to create mci attributes\n", + __func__); + goto fail0; + } + } + + /* Make directories for each CSROW object under the mc<id> kobject + */ + for (i = 0; i < mci->nr_csrows; i++) { + csrow = &mci->csrows[i]; + + /* Only expose populated CSROWs */ + if (csrow->nr_pages > 0) { + err = edac_create_csrow_object(mci, csrow, i); + if (err) { + debugf1("%s() failure: create csrow %d obj\n", + __func__, i); + goto fail1; + } + } + } + + return 0; + + /* CSROW error: backout what has already been registered, */ +fail1: + for (i--; i >= 0; i--) { + if (csrow->nr_pages > 0) { + kobject_unregister(&mci->csrows[i].kobj); + } + } + + /* remove the mci instance's attributes, if any */ + edac_remove_mci_instance_attributes(mci); + + /* remove the symlink */ + sysfs_remove_link(kobj_mci, EDAC_DEVICE_SYMLINK); + +fail0: + return err; +} + +/* + * remove a Memory Controller instance + */ +void edac_remove_sysfs_mci_device(struct mem_ctl_info *mci) +{ + int i; + + debugf0("%s()\n", __func__); + + /* remove all csrow kobjects */ + for (i = 0; i < mci->nr_csrows; i++) { + if (mci->csrows[i].nr_pages > 0) { + debugf0("%s() unreg csrow-%d\n", __func__, i); + kobject_unregister(&mci->csrows[i].kobj); + } + } + + debugf0("%s() remove_link\n", __func__); + + /* remove the symlink */ + sysfs_remove_link(&mci->edac_mci_kobj, EDAC_DEVICE_SYMLINK); + + debugf0("%s() remove_mci_instance\n", __func__); + + /* remove this mci instance's attribtes */ + edac_remove_mci_instance_attributes(mci); + + debugf0("%s() unregister this mci kobj\n", __func__); + + /* unregister this instance's kobject */ + kobject_unregister(&mci->edac_mci_kobj); +} + + + + +/* + * edac_setup_sysfs_mc_kset(void) + * + * Initialize the mc_kset for the 'mc' entry + * This requires creating the top 'mc' directory with a kset + * and its controls/attributes. + * + * To this 'mc' kset, instance 'mci' will be grouped as children. + * + * Return: 0 SUCCESS + * !0 FAILURE error code + */ +int edac_sysfs_setup_mc_kset(void) +{ + int err = 0; + struct sysdev_class *edac_class; + + debugf1("%s()\n", __func__); + + /* get the /sys/devices/system/edac class reference */ + edac_class = edac_get_edac_class(); + if (edac_class == NULL) { + debugf1("%s() no edac_class error=%d\n", __func__, err); + goto fail_out; + } + + /* Init the MC's kobject */ + mc_kset.kobj.parent = &edac_class->kset.kobj; + + /* register the mc_kset */ + err = kset_register(&mc_kset); + if (err) { + debugf1("%s() Failed to register '.../edac/mc'\n", __func__); + goto fail_out; + } + + debugf1("%s() Registered '.../edac/mc' kobject\n", __func__); + + return 0; + + + /* error unwind stack */ +fail_out: + return err; +} + +/* + * edac_sysfs_teardown_mc_kset + * + * deconstruct the mc_ket for memory controllers + */ +void edac_sysfs_teardown_mc_kset(void) +{ + kset_unregister(&mc_kset); +} + diff --git a/drivers/edac/edac_module.c b/drivers/edac/edac_module.c new file mode 100644 index 00000000000..e0c4a408605 --- /dev/null +++ b/drivers/edac/edac_module.c @@ -0,0 +1,222 @@ +/* + * edac_module.c + * + * (C) 2007 www.softwarebitmaker.com + * + * This file is licensed under the terms of the GNU General Public + * License version 2. This program is licensed "as is" without any + * warranty of any kind, whether express or implied. + * + * Author: Doug Thompson <dougthompson@xmission.com> + * + */ +#include <linux/edac.h> + +#include "edac_core.h" +#include "edac_module.h" + +#define EDAC_VERSION "Ver: 2.1.0 " __DATE__ + +#ifdef CONFIG_EDAC_DEBUG +/* Values of 0 to 4 will generate output */ +int edac_debug_level = 2; +EXPORT_SYMBOL_GPL(edac_debug_level); +#endif + +/* scope is to module level only */ +struct workqueue_struct *edac_workqueue; + +/* + * sysfs object: /sys/devices/system/edac + * need to export to other files in this modules + */ +static struct sysdev_class edac_class = { + set_kset_name("edac"), +}; +static int edac_class_valid; + +/* + * edac_op_state_to_string() + */ +char *edac_op_state_to_string(int opstate) +{ + if (opstate == OP_RUNNING_POLL) + return "POLLED"; + else if (opstate == OP_RUNNING_INTERRUPT) + return "INTERRUPT"; + else if (opstate == OP_RUNNING_POLL_INTR) + return "POLL-INTR"; + else if (opstate == OP_ALLOC) + return "ALLOC"; + else if (opstate == OP_OFFLINE) + return "OFFLINE"; + + return "UNKNOWN"; +} + +/* + * edac_get_edac_class() + * + * return pointer to the edac class of 'edac' + */ +struct sysdev_class *edac_get_edac_class(void) +{ + struct sysdev_class *classptr = NULL; + + if (edac_class_valid) + classptr = &edac_class; + + return classptr; +} + +/* + * edac_register_sysfs_edac_name() + * + * register the 'edac' into /sys/devices/system + * + * return: + * 0 success + * !0 error + */ +static int edac_register_sysfs_edac_name(void) +{ + int err; + + /* create the /sys/devices/system/edac directory */ + err = sysdev_class_register(&edac_class); + + if (err) { + debugf1("%s() error=%d\n", __func__, err); + return err; + } + + edac_class_valid = 1; + return 0; +} + +/* + * sysdev_class_unregister() + * + * unregister the 'edac' from /sys/devices/system + */ +static void edac_unregister_sysfs_edac_name(void) +{ + /* only if currently registered, then unregister it */ + if (edac_class_valid) + sysdev_class_unregister(&edac_class); + + edac_class_valid = 0; +} + +/* + * edac_workqueue_setup + * initialize the edac work queue for polling operations + */ +static int edac_workqueue_setup(void) +{ + edac_workqueue = create_singlethread_workqueue("edac-poller"); + if (edac_workqueue == NULL) + return -ENODEV; + else + return 0; +} + +/* + * edac_workqueue_teardown + * teardown the edac workqueue + */ +static void edac_workqueue_teardown(void) +{ + if (edac_workqueue) { + flush_workqueue(edac_workqueue); + destroy_workqueue(edac_workqueue); + edac_workqueue = NULL; + } +} + +/* + * edac_init + * module initialization entry point + */ +static int __init edac_init(void) +{ + int err = 0; + + edac_printk(KERN_INFO, EDAC_MC, EDAC_VERSION "\n"); + + /* + * Harvest and clear any boot/initialization PCI parity errors + * + * FIXME: This only clears errors logged by devices present at time of + * module initialization. We should also do an initial clear + * of each newly hotplugged device. + */ + edac_pci_clear_parity_errors(); + + /* + * perform the registration of the /sys/devices/system/edac class object + */ + if (edac_register_sysfs_edac_name()) { + edac_printk(KERN_ERR, EDAC_MC, + "Error initializing 'edac' kobject\n"); + err = -ENODEV; + goto error; + } + + /* + * now set up the mc_kset under the edac class object + */ + err = edac_sysfs_setup_mc_kset(); + if (err) + goto sysfs_setup_fail; + + /* Setup/Initialize the workq for this core */ + err = edac_workqueue_setup(); + if (err) { + edac_printk(KERN_ERR, EDAC_MC, "init WorkQueue failure\n"); + goto workq_fail; + } + + return 0; + + /* Error teardown stack */ +workq_fail: + edac_sysfs_teardown_mc_kset(); + +sysfs_setup_fail: + edac_unregister_sysfs_edac_name(); + +error: + return err; +} + +/* + * edac_exit() + * module exit/termination function + */ +static void __exit edac_exit(void) +{ + debugf0("%s()\n", __func__); + + /* tear down the various subsystems */ + edac_workqueue_teardown(); + edac_sysfs_teardown_mc_kset(); + edac_unregister_sysfs_edac_name(); +} + +/* + * Inform the kernel of our entry and exit points + */ +module_init(edac_init); +module_exit(edac_exit); + +MODULE_LICENSE("GPL"); +MODULE_AUTHOR("Doug Thompson www.softwarebitmaker.com, et al"); +MODULE_DESCRIPTION("Core library routines for EDAC reporting"); + +/* refer to *_sysfs.c files for parameters that are exported via sysfs */ + +#ifdef CONFIG_EDAC_DEBUG +module_param(edac_debug_level, int, 0644); +MODULE_PARM_DESC(edac_debug_level, "Debug level"); +#endif diff --git a/drivers/edac/edac_module.h b/drivers/edac/edac_module.h new file mode 100644 index 00000000000..a2134dfc3cc --- /dev/null +++ b/drivers/edac/edac_module.h @@ -0,0 +1,77 @@ + +/* + * edac_module.h + * + * For defining functions/data for within the EDAC_CORE module only + * + * written by doug thompson <norsk5@xmission.h> + */ + +#ifndef __EDAC_MODULE_H__ +#define __EDAC_MODULE_H__ + +#include <linux/sysdev.h> + +#include "edac_core.h" + +/* + * INTERNAL EDAC MODULE: + * EDAC memory controller sysfs create/remove functions + * and setup/teardown functions + * + * edac_mc objects + */ +extern int edac_sysfs_setup_mc_kset(void); +extern void edac_sysfs_teardown_mc_kset(void); +extern int edac_mc_register_sysfs_main_kobj(struct mem_ctl_info *mci); +extern void edac_mc_unregister_sysfs_main_kobj(struct mem_ctl_info *mci); +extern int edac_create_sysfs_mci_device(struct mem_ctl_info *mci); +extern void edac_remove_sysfs_mci_device(struct mem_ctl_info *mci); +extern void edac_check_mc_devices(void); +extern int edac_get_log_ue(void); +extern int edac_get_log_ce(void); +extern int edac_get_panic_on_ue(void); +extern int edac_mc_get_log_ue(void); +extern int edac_mc_get_log_ce(void); +extern int edac_mc_get_panic_on_ue(void); +extern int edac_get_poll_msec(void); +extern int edac_mc_get_poll_msec(void); + +extern int edac_device_register_sysfs_main_kobj( + struct edac_device_ctl_info *edac_dev); +extern void edac_device_unregister_sysfs_main_kobj( + struct edac_device_ctl_info *edac_dev); +extern int edac_device_create_sysfs(struct edac_device_ctl_info *edac_dev); +extern void edac_device_remove_sysfs(struct edac_device_ctl_info *edac_dev); +extern struct sysdev_class *edac_get_edac_class(void); + +/* edac core workqueue: single CPU mode */ +extern struct workqueue_struct *edac_workqueue; +extern void edac_device_workq_setup(struct edac_device_ctl_info *edac_dev, + unsigned msec); +extern void edac_device_workq_teardown(struct edac_device_ctl_info *edac_dev); +extern void edac_device_reset_delay_period(struct edac_device_ctl_info + *edac_dev, unsigned long value); +extern void *edac_align_ptr(void *ptr, unsigned size); + +/* + * EDAC PCI functions + */ +#ifdef CONFIG_PCI +extern void edac_pci_do_parity_check(void); +extern void edac_pci_clear_parity_errors(void); +extern int edac_sysfs_pci_setup(void); +extern void edac_sysfs_pci_teardown(void); +extern int edac_pci_get_check_errors(void); +extern int edac_pci_get_poll_msec(void); +#else /* CONFIG_PCI */ +/* pre-process these away */ +#define edac_pci_do_parity_check() +#define edac_pci_clear_parity_errors() +#define edac_sysfs_pci_setup() (0) +#define edac_sysfs_pci_teardown() +#define edac_pci_get_check_errors() +#define edac_pci_get_poll_msec() +#endif /* CONFIG_PCI */ + +#endif /* __EDAC_MODULE_H__ */ diff --git a/drivers/edac/edac_pci.c b/drivers/edac/edac_pci.c new file mode 100644 index 00000000000..d9cd5e048ce --- /dev/null +++ b/drivers/edac/edac_pci.c @@ -0,0 +1,433 @@ +/* + * EDAC PCI component + * + * Author: Dave Jiang <djiang@mvista.com> + * + * 2007 (c) MontaVista Software, Inc. This file is licensed under + * the terms of the GNU General Public License version 2. This program + * is licensed "as is" without any warranty of any kind, whether express + * or implied. + * + */ +#include <linux/module.h> +#include <linux/types.h> +#include <linux/smp.h> +#include <linux/init.h> +#include <linux/sysctl.h> +#include <linux/highmem.h> +#include <linux/timer.h> +#include <linux/slab.h> +#include <linux/spinlock.h> +#include <linux/list.h> +#include <linux/sysdev.h> +#include <linux/ctype.h> +#include <linux/workqueue.h> +#include <asm/uaccess.h> +#include <asm/page.h> + +#include "edac_core.h" +#include "edac_module.h" + +static DEFINE_MUTEX(edac_pci_ctls_mutex); +static struct list_head edac_pci_list = LIST_HEAD_INIT(edac_pci_list); + +static inline void edac_lock_pci_list(void) +{ + mutex_lock(&edac_pci_ctls_mutex); +} + +static inline void edac_unlock_pci_list(void) +{ + mutex_unlock(&edac_pci_ctls_mutex); +} + +/* + * The alloc() and free() functions for the 'edac_pci' control info + * structure. The chip driver will allocate one of these for each + * edac_pci it is going to control/register with the EDAC CORE. + */ +struct edac_pci_ctl_info *edac_pci_alloc_ctl_info(unsigned int sz_pvt, + const char *edac_pci_name) +{ + struct edac_pci_ctl_info *pci; + void *pvt; + unsigned int size; + + pci = (struct edac_pci_ctl_info *)0; + pvt = edac_align_ptr(&pci[1], sz_pvt); + size = ((unsigned long)pvt) + sz_pvt; + + if ((pci = kzalloc(size, GFP_KERNEL)) == NULL) + return NULL; + + pvt = sz_pvt ? ((char *)pci) + ((unsigned long)pvt) : NULL; + + pci->pvt_info = pvt; + + pci->op_state = OP_ALLOC; + + snprintf(pci->name, strlen(edac_pci_name) + 1, "%s", edac_pci_name); + + return pci; +} + +EXPORT_SYMBOL_GPL(edac_pci_alloc_ctl_info); + +/* + * edac_pci_free_ctl_info() + * frees the memory allocated by edac_pci_alloc_ctl_info() function + */ +void edac_pci_free_ctl_info(struct edac_pci_ctl_info *pci) +{ + kfree(pci); +} + +EXPORT_SYMBOL_GPL(edac_pci_free_ctl_info); + +/* + * find_edac_pci_by_dev() + * scans the edac_pci list for a specific 'struct device *' + */ +static struct edac_pci_ctl_info *find_edac_pci_by_dev(struct device *dev) +{ + struct edac_pci_ctl_info *pci; + struct list_head *item; + + debugf3("%s()\n", __func__); + + list_for_each(item, &edac_pci_list) { + pci = list_entry(item, struct edac_pci_ctl_info, link); + + if (pci->dev == dev) + return pci; + } + + return NULL; +} + +/* + * add_edac_pci_to_global_list + * Before calling this function, caller must assign a unique value to + * edac_dev->pci_idx. + * Return: + * 0 on success + * 1 on failure + */ +static int add_edac_pci_to_global_list(struct edac_pci_ctl_info *pci) +{ + struct list_head *item, *insert_before; + struct edac_pci_ctl_info *rover; + + insert_before = &edac_pci_list; + + /* Determine if already on the list */ + if (unlikely((rover = find_edac_pci_by_dev(pci->dev)) != NULL)) + goto fail0; + + /* Insert in ascending order by 'pci_idx', so find position */ + list_for_each(item, &edac_pci_list) { + rover = list_entry(item, struct edac_pci_ctl_info, link); + + if (rover->pci_idx >= pci->pci_idx) { + if (unlikely(rover->pci_idx == pci->pci_idx)) + goto fail1; + + insert_before = item; + break; + } + } + + list_add_tail_rcu(&pci->link, insert_before); + return 0; + +fail0: + edac_printk(KERN_WARNING, EDAC_PCI, + "%s (%s) %s %s already assigned %d\n", + rover->dev->bus_id, dev_name(rover), + rover->mod_name, rover->ctl_name, rover->pci_idx); + return 1; + +fail1: + edac_printk(KERN_WARNING, EDAC_PCI, + "but in low-level driver: attempt to assign\n" + "\tduplicate pci_idx %d in %s()\n", rover->pci_idx, + __func__); + return 1; +} + +/* + * complete_edac_pci_list_del + */ +static void complete_edac_pci_list_del(struct rcu_head *head) +{ + struct edac_pci_ctl_info *pci; + + pci = container_of(head, struct edac_pci_ctl_info, rcu); + INIT_LIST_HEAD(&pci->link); + complete(&pci->complete); +} + +/* + * del_edac_pci_from_global_list + */ +static void del_edac_pci_from_global_list(struct edac_pci_ctl_info *pci) +{ + list_del_rcu(&pci->link); + init_completion(&pci->complete); + call_rcu(&pci->rcu, complete_edac_pci_list_del); + wait_for_completion(&pci->complete); +} + +/* + * edac_pci_find() + * Search for an edac_pci_ctl_info structure whose index is 'idx' + * + * If found, return a pointer to the structure + * Else return NULL. + * + * Caller must hold pci_ctls_mutex. + */ +struct edac_pci_ctl_info *edac_pci_find(int idx) +{ + struct list_head *item; + struct edac_pci_ctl_info *pci; + + /* Iterage over list, looking for exact match of ID */ + list_for_each(item, &edac_pci_list) { + pci = list_entry(item, struct edac_pci_ctl_info, link); + + if (pci->pci_idx >= idx) { + if (pci->pci_idx == idx) + return pci; + + /* not on list, so terminate early */ + break; + } + } + + return NULL; +} + +EXPORT_SYMBOL_GPL(edac_pci_find); + +/* + * edac_pci_workq_function() + * performs the operation scheduled by a workq request + */ +static void edac_pci_workq_function(struct work_struct *work_req) +{ + struct delayed_work *d_work = (struct delayed_work *)work_req; + struct edac_pci_ctl_info *pci = to_edac_pci_ctl_work(d_work); + + edac_lock_pci_list(); + + if ((pci->op_state == OP_RUNNING_POLL) && + (pci->edac_check != NULL) && (edac_pci_get_check_errors())) + pci->edac_check(pci); + + edac_unlock_pci_list(); + + /* Reschedule */ + queue_delayed_work(edac_workqueue, &pci->work, + msecs_to_jiffies(edac_pci_get_poll_msec())); +} + +/* + * edac_pci_workq_setup() + * initialize a workq item for this edac_pci instance + * passing in the new delay period in msec + */ +static void edac_pci_workq_setup(struct edac_pci_ctl_info *pci, + unsigned int msec) +{ + debugf0("%s()\n", __func__); + + INIT_DELAYED_WORK(&pci->work, edac_pci_workq_function); + queue_delayed_work(edac_workqueue, &pci->work, + msecs_to_jiffies(edac_pci_get_poll_msec())); +} + +/* + * edac_pci_workq_teardown() + * stop the workq processing on this edac_pci instance + */ +static void edac_pci_workq_teardown(struct edac_pci_ctl_info *pci) +{ + int status; + + status = cancel_delayed_work(&pci->work); + if (status == 0) + flush_workqueue(edac_workqueue); +} + +/* + * edac_pci_reset_delay_period + */ +void edac_pci_reset_delay_period(struct edac_pci_ctl_info *pci, + unsigned long value) +{ + edac_lock_pci_list(); + + edac_pci_workq_teardown(pci); + + edac_pci_workq_setup(pci, value); + + edac_unlock_pci_list(); +} + +EXPORT_SYMBOL_GPL(edac_pci_reset_delay_period); + +/* + * edac_pci_add_device: Insert the 'edac_dev' structure into the + * edac_pci global list and create sysfs entries associated with + * edac_pci structure. + * @pci: pointer to the edac_device structure to be added to the list + * @edac_idx: A unique numeric identifier to be assigned to the + * 'edac_pci' structure. + * + * Return: + * 0 Success + * !0 Failure + */ +int edac_pci_add_device(struct edac_pci_ctl_info *pci, int edac_idx) +{ + debugf0("%s()\n", __func__); + + pci->pci_idx = edac_idx; + + edac_lock_pci_list(); + + if (add_edac_pci_to_global_list(pci)) + goto fail0; + + pci->start_time = jiffies; + + if (edac_pci_create_sysfs(pci)) { + edac_pci_printk(pci, KERN_WARNING, + "failed to create sysfs pci\n"); + goto fail1; + } + + if (pci->edac_check != NULL) { + pci->op_state = OP_RUNNING_POLL; + + edac_pci_workq_setup(pci, 1000); + } else { + pci->op_state = OP_RUNNING_INTERRUPT; + } + + edac_pci_printk(pci, KERN_INFO, + "Giving out device to module '%s' controller '%s':" + " DEV '%s' (%s)\n", + pci->mod_name, + pci->ctl_name, + dev_name(pci), edac_op_state_to_string(pci->op_state)); + + edac_unlock_pci_list(); + return 0; + +fail1: + del_edac_pci_from_global_list(pci); +fail0: + edac_unlock_pci_list(); + return 1; +} + +EXPORT_SYMBOL_GPL(edac_pci_add_device); + +/* + * edac_pci_del_device() + * Remove sysfs entries for specified edac_pci structure and + * then remove edac_pci structure from global list + * + * @dev: + * Pointer to 'struct device' representing edac_pci structure + * to remove + * + * Return: + * Pointer to removed edac_pci structure, + * or NULL if device not found + */ +struct edac_pci_ctl_info *edac_pci_del_device(struct device *dev) +{ + struct edac_pci_ctl_info *pci; + + debugf0("%s()\n", __func__); + + edac_lock_pci_list(); + + if ((pci = find_edac_pci_by_dev(dev)) == NULL) { + edac_unlock_pci_list(); + return NULL; + } + + pci->op_state = OP_OFFLINE; + + edac_pci_workq_teardown(pci); + + edac_pci_remove_sysfs(pci); + + del_edac_pci_from_global_list(pci); + + edac_unlock_pci_list(); + + edac_printk(KERN_INFO, EDAC_PCI, + "Removed device %d for %s %s: DEV %s\n", + pci->pci_idx, pci->mod_name, pci->ctl_name, dev_name(pci)); + + return pci; +} + +EXPORT_SYMBOL_GPL(edac_pci_del_device); + +void edac_pci_generic_check(struct edac_pci_ctl_info *pci) +{ + edac_pci_do_parity_check(); +} + +static int edac_pci_idx; +#define EDAC_PCI_GENCTL_NAME "EDAC PCI controller" + +struct edac_pci_gen_data { + int edac_idx; +}; + +struct edac_pci_ctl_info *edac_pci_create_generic_ctl(struct device *dev, + const char *mod_name) +{ + struct edac_pci_ctl_info *pci; + struct edac_pci_gen_data *pdata; + + pci = edac_pci_alloc_ctl_info(sizeof(*pdata), EDAC_PCI_GENCTL_NAME); + if (!pci) + return NULL; + + pdata = pci->pvt_info; + pci->dev = dev; + dev_set_drvdata(pci->dev, pci); + pci->dev_name = pci_name(to_pci_dev(dev)); + + pci->mod_name = mod_name; + pci->ctl_name = EDAC_PCI_GENCTL_NAME; + pci->edac_check = edac_pci_generic_check; + + pdata->edac_idx = edac_pci_idx++; + + if (edac_pci_add_device(pci, pdata->edac_idx) > 0) { + debugf3("%s(): failed edac_pci_add_device()\n", __func__); + edac_pci_free_ctl_info(pci); + return NULL; + } + + return pci; +} + +EXPORT_SYMBOL_GPL(edac_pci_create_generic_ctl); + +void edac_pci_release_generic_ctl(struct edac_pci_ctl_info *pci) +{ + edac_pci_del_device(pci->dev); + edac_pci_free_ctl_info(pci); +} + +EXPORT_SYMBOL_GPL(edac_pci_release_generic_ctl); diff --git a/drivers/edac/edac_pci_sysfs.c b/drivers/edac/edac_pci_sysfs.c new file mode 100644 index 00000000000..fac94cae2c3 --- /dev/null +++ b/drivers/edac/edac_pci_sysfs.c @@ -0,0 +1,620 @@ +/* + * (C) 2005, 2006 Linux Networx (http://lnxi.com) + * This file may be distributed under the terms of the + * GNU General Public License. + * + * Written Doug Thompson <norsk5@xmission.com> + * + */ +#include <linux/module.h> +#include <linux/sysdev.h> +#include <linux/ctype.h> + +#include "edac_core.h" +#include "edac_module.h" + +#ifdef CONFIG_PCI + +#define EDAC_PCI_SYMLINK "device" + +static int check_pci_errors; /* default YES check PCI parity */ +static int edac_pci_panic_on_pe; /* default no panic on PCI Parity */ +static int edac_pci_log_pe = 1; /* log PCI parity errors */ +static int edac_pci_log_npe = 1; /* log PCI non-parity error errors */ +static atomic_t pci_parity_count = ATOMIC_INIT(0); +static atomic_t pci_nonparity_count = ATOMIC_INIT(0); +static int edac_pci_poll_msec = 1000; + +static struct kobject edac_pci_kobj; /* /sys/devices/system/edac/pci */ +static struct completion edac_pci_kobj_complete; +static atomic_t edac_pci_sysfs_refcount = ATOMIC_INIT(0); + +int edac_pci_get_check_errors(void) +{ + return check_pci_errors; +} + +int edac_pci_get_log_pe(void) +{ + return edac_pci_log_pe; +} + +int edac_pci_get_log_npe(void) +{ + return edac_pci_log_npe; +} + +int edac_pci_get_panic_on_pe(void) +{ + return edac_pci_panic_on_pe; +} + +int edac_pci_get_poll_msec(void) +{ + return edac_pci_poll_msec; +} + +/**************************** EDAC PCI sysfs instance *******************/ +static ssize_t instance_pe_count_show(struct edac_pci_ctl_info *pci, char *data) +{ + return sprintf(data, "%u\n", atomic_read(&pci->counters.pe_count)); +} + +static ssize_t instance_npe_count_show(struct edac_pci_ctl_info *pci, + char *data) +{ + return sprintf(data, "%u\n", atomic_read(&pci->counters.npe_count)); +} + +#define to_instance(k) container_of(k, struct edac_pci_ctl_info, kobj) +#define to_instance_attr(a) container_of(a, struct instance_attribute, attr) + +/* DEVICE instance kobject release() function */ +static void edac_pci_instance_release(struct kobject *kobj) +{ + struct edac_pci_ctl_info *pci; + + debugf1("%s()\n", __func__); + + pci = to_instance(kobj); + complete(&pci->kobj_complete); +} + +/* instance specific attribute structure */ +struct instance_attribute { + struct attribute attr; + ssize_t(*show) (struct edac_pci_ctl_info *, char *); + ssize_t(*store) (struct edac_pci_ctl_info *, const char *, size_t); +}; + +/* Function to 'show' fields from the edac_pci 'instance' structure */ +static ssize_t edac_pci_instance_show(struct kobject *kobj, + struct attribute *attr, char *buffer) +{ + struct edac_pci_ctl_info *pci = to_instance(kobj); + struct instance_attribute *instance_attr = to_instance_attr(attr); + + if (instance_attr->show) + return instance_attr->show(pci, buffer); + return -EIO; +} + +/* Function to 'store' fields into the edac_pci 'instance' structure */ +static ssize_t edac_pci_instance_store(struct kobject *kobj, + struct attribute *attr, + const char *buffer, size_t count) +{ + struct edac_pci_ctl_info *pci = to_instance(kobj); + struct instance_attribute *instance_attr = to_instance_attr(attr); + + if (instance_attr->store) + return instance_attr->store(pci, buffer, count); + return -EIO; +} + +static struct sysfs_ops pci_instance_ops = { + .show = edac_pci_instance_show, + .store = edac_pci_instance_store +}; + +#define INSTANCE_ATTR(_name, _mode, _show, _store) \ +static struct instance_attribute attr_instance_##_name = { \ + .attr = {.name = __stringify(_name), .mode = _mode }, \ + .show = _show, \ + .store = _store, \ +}; + +INSTANCE_ATTR(pe_count, S_IRUGO, instance_pe_count_show, NULL); +INSTANCE_ATTR(npe_count, S_IRUGO, instance_npe_count_show, NULL); + +/* pci instance attributes */ +static struct instance_attribute *pci_instance_attr[] = { + &attr_instance_pe_count, + &attr_instance_npe_count, + NULL +}; + +/* the ktype for pci instance */ +static struct kobj_type ktype_pci_instance = { + .release = edac_pci_instance_release, + .sysfs_ops = &pci_instance_ops, + .default_attrs = (struct attribute **)pci_instance_attr, +}; + +static int edac_pci_create_instance_kobj(struct edac_pci_ctl_info *pci, int idx) +{ + int err; + + pci->kobj.parent = &edac_pci_kobj; + pci->kobj.ktype = &ktype_pci_instance; + + err = kobject_set_name(&pci->kobj, "pci%d", idx); + if (err) + return err; + + err = kobject_register(&pci->kobj); + if (err != 0) { + debugf2("%s() failed to register instance pci%d\n", + __func__, idx); + return err; + } + + debugf1("%s() Register instance 'pci%d' kobject\n", __func__, idx); + + return 0; +} + +static void +edac_pci_delete_instance_kobj(struct edac_pci_ctl_info *pci, int idx) +{ + init_completion(&pci->kobj_complete); + kobject_unregister(&pci->kobj); + wait_for_completion(&pci->kobj_complete); +} + +/***************************** EDAC PCI sysfs root **********************/ +#define to_edacpci(k) container_of(k, struct edac_pci_ctl_info, kobj) +#define to_edacpci_attr(a) container_of(a, struct edac_pci_attr, attr) + +static ssize_t edac_pci_int_show(void *ptr, char *buffer) +{ + int *value = ptr; + return sprintf(buffer, "%d\n", *value); +} + +static ssize_t edac_pci_int_store(void *ptr, const char *buffer, size_t count) +{ + int *value = ptr; + + if (isdigit(*buffer)) + *value = simple_strtoul(buffer, NULL, 0); + + return count; +} + +struct edac_pci_dev_attribute { + struct attribute attr; + void *value; + ssize_t(*show) (void *, char *); + ssize_t(*store) (void *, const char *, size_t); +}; + +/* Set of show/store abstract level functions for PCI Parity object */ +static ssize_t edac_pci_dev_show(struct kobject *kobj, struct attribute *attr, + char *buffer) +{ + struct edac_pci_dev_attribute *edac_pci_dev; + edac_pci_dev = (struct edac_pci_dev_attribute *)attr; + + if (edac_pci_dev->show) + return edac_pci_dev->show(edac_pci_dev->value, buffer); + return -EIO; +} + +static ssize_t edac_pci_dev_store(struct kobject *kobj, + struct attribute *attr, const char *buffer, + size_t count) +{ + struct edac_pci_dev_attribute *edac_pci_dev; + edac_pci_dev = (struct edac_pci_dev_attribute *)attr; + + if (edac_pci_dev->show) + return edac_pci_dev->store(edac_pci_dev->value, buffer, count); + return -EIO; +} + +static struct sysfs_ops edac_pci_sysfs_ops = { + .show = edac_pci_dev_show, + .store = edac_pci_dev_store +}; + +#define EDAC_PCI_ATTR(_name,_mode,_show,_store) \ +static struct edac_pci_dev_attribute edac_pci_attr_##_name = { \ + .attr = {.name = __stringify(_name), .mode = _mode }, \ + .value = &_name, \ + .show = _show, \ + .store = _store, \ +}; + +#define EDAC_PCI_STRING_ATTR(_name,_data,_mode,_show,_store) \ +static struct edac_pci_dev_attribute edac_pci_attr_##_name = { \ + .attr = {.name = __stringify(_name), .mode = _mode }, \ + .value = _data, \ + .show = _show, \ + .store = _store, \ +}; + +/* PCI Parity control files */ +EDAC_PCI_ATTR(check_pci_errors, S_IRUGO | S_IWUSR, edac_pci_int_show, + edac_pci_int_store); +EDAC_PCI_ATTR(edac_pci_log_pe, S_IRUGO | S_IWUSR, edac_pci_int_show, + edac_pci_int_store); +EDAC_PCI_ATTR(edac_pci_log_npe, S_IRUGO | S_IWUSR, edac_pci_int_show, + edac_pci_int_store); +EDAC_PCI_ATTR(edac_pci_panic_on_pe, S_IRUGO | S_IWUSR, edac_pci_int_show, + edac_pci_int_store); +EDAC_PCI_ATTR(pci_parity_count, S_IRUGO, edac_pci_int_show, NULL); +EDAC_PCI_ATTR(pci_nonparity_count, S_IRUGO, edac_pci_int_show, NULL); + +/* Base Attributes of the memory ECC object */ +static struct edac_pci_dev_attribute *edac_pci_attr[] = { + &edac_pci_attr_check_pci_errors, + &edac_pci_attr_edac_pci_log_pe, + &edac_pci_attr_edac_pci_log_npe, + &edac_pci_attr_edac_pci_panic_on_pe, + &edac_pci_attr_pci_parity_count, + &edac_pci_attr_pci_nonparity_count, + NULL, +}; + +/* No memory to release */ +static void edac_pci_release(struct kobject *kobj) +{ + struct edac_pci_ctl_info *pci; + + pci = to_edacpci(kobj); + + debugf1("%s()\n", __func__); + complete(&pci->kobj_complete); +} + +static struct kobj_type ktype_edac_pci = { + .release = edac_pci_release, + .sysfs_ops = &edac_pci_sysfs_ops, + .default_attrs = (struct attribute **)edac_pci_attr, +}; + +/** + * edac_sysfs_pci_setup() + * + * setup the sysfs for EDAC PCI attributes + * assumes edac_class has already been initialized + */ +int edac_pci_register_main_kobj(void) +{ + int err; + struct sysdev_class *edac_class; + + debugf1("%s()\n", __func__); + + edac_class = edac_get_edac_class(); + if (edac_class == NULL) { + debugf1("%s() no edac_class\n", __func__); + return -ENODEV; + } + + edac_pci_kobj.ktype = &ktype_edac_pci; + + edac_pci_kobj.parent = &edac_class->kset.kobj; + + err = kobject_set_name(&edac_pci_kobj, "pci"); + if (err) + return err; + + /* Instanstiate the pci object */ + /* FIXME: maybe new sysdev_create_subdir() */ + err = kobject_register(&edac_pci_kobj); + + if (err) { + debugf1("Failed to register '.../edac/pci'\n"); + return err; + } + + debugf1("Registered '.../edac/pci' kobject\n"); + + return 0; +} + +/* + * edac_pci_unregister_main_kobj() + * + * perform the sysfs teardown for the PCI attributes + */ +void edac_pci_unregister_main_kobj(void) +{ + debugf0("%s()\n", __func__); + init_completion(&edac_pci_kobj_complete); + kobject_unregister(&edac_pci_kobj); + wait_for_completion(&edac_pci_kobj_complete); +} + +int edac_pci_create_sysfs(struct edac_pci_ctl_info *pci) +{ + int err; + struct kobject *edac_kobj = &pci->kobj; + + if (atomic_inc_return(&edac_pci_sysfs_refcount) == 1) { + err = edac_pci_register_main_kobj(); + if (err) { + atomic_dec(&edac_pci_sysfs_refcount); + return err; + } + } + + err = edac_pci_create_instance_kobj(pci, pci->pci_idx); + if (err) { + if (atomic_dec_return(&edac_pci_sysfs_refcount) == 0) + edac_pci_unregister_main_kobj(); + } + + debugf0("%s() idx=%d\n", __func__, pci->pci_idx); + + err = sysfs_create_link(edac_kobj, &pci->dev->kobj, EDAC_PCI_SYMLINK); + if (err) { + debugf0("%s() sysfs_create_link() returned err= %d\n", + __func__, err); + return err; + } + + return 0; +} + +void edac_pci_remove_sysfs(struct edac_pci_ctl_info *pci) +{ + debugf0("%s()\n", __func__); + + edac_pci_delete_instance_kobj(pci, pci->pci_idx); + + sysfs_remove_link(&pci->kobj, EDAC_PCI_SYMLINK); + + if (atomic_dec_return(&edac_pci_sysfs_refcount) == 0) + edac_pci_unregister_main_kobj(); +} + +/************************ PCI error handling *************************/ +static u16 get_pci_parity_status(struct pci_dev *dev, int secondary) +{ + int where; + u16 status; + + where = secondary ? PCI_SEC_STATUS : PCI_STATUS; + pci_read_config_word(dev, where, &status); + + /* If we get back 0xFFFF then we must suspect that the card has been + * pulled but the Linux PCI layer has not yet finished cleaning up. + * We don't want to report on such devices + */ + + if (status == 0xFFFF) { + u32 sanity; + + pci_read_config_dword(dev, 0, &sanity); + + if (sanity == 0xFFFFFFFF) + return 0; + } + + status &= PCI_STATUS_DETECTED_PARITY | PCI_STATUS_SIG_SYSTEM_ERROR | + PCI_STATUS_PARITY; + + if (status) + /* reset only the bits we are interested in */ + pci_write_config_word(dev, where, status); + + return status; +} + +typedef void (*pci_parity_check_fn_t) (struct pci_dev * dev); + +/* Clear any PCI parity errors logged by this device. */ +static void edac_pci_dev_parity_clear(struct pci_dev *dev) +{ + u8 header_type; + + get_pci_parity_status(dev, 0); + + /* read the device TYPE, looking for bridges */ + pci_read_config_byte(dev, PCI_HEADER_TYPE, &header_type); + + if ((header_type & 0x7F) == PCI_HEADER_TYPE_BRIDGE) + get_pci_parity_status(dev, 1); +} + +/* + * PCI Parity polling + * + */ +static void edac_pci_dev_parity_test(struct pci_dev *dev) +{ + u16 status; + u8 header_type; + + /* read the STATUS register on this device + */ + status = get_pci_parity_status(dev, 0); + + debugf2("PCI STATUS= 0x%04x %s\n", status, dev->dev.bus_id); + + /* check the status reg for errors */ + if (status) { + if (status & (PCI_STATUS_SIG_SYSTEM_ERROR)) { + edac_printk(KERN_CRIT, EDAC_PCI, + "Signaled System Error on %s\n", + pci_name(dev)); + atomic_inc(&pci_nonparity_count); + } + + if (status & (PCI_STATUS_PARITY)) { + edac_printk(KERN_CRIT, EDAC_PCI, + "Master Data Parity Error on %s\n", + pci_name(dev)); + + atomic_inc(&pci_parity_count); + } + + if (status & (PCI_STATUS_DETECTED_PARITY)) { + edac_printk(KERN_CRIT, EDAC_PCI, + "Detected Parity Error on %s\n", + pci_name(dev)); + + atomic_inc(&pci_parity_count); + } + } + + /* read the device TYPE, looking for bridges */ + pci_read_config_byte(dev, PCI_HEADER_TYPE, &header_type); + + debugf2("PCI HEADER TYPE= 0x%02x %s\n", header_type, dev->dev.bus_id); + + if ((header_type & 0x7F) == PCI_HEADER_TYPE_BRIDGE) { + /* On bridges, need to examine secondary status register */ + status = get_pci_parity_status(dev, 1); + + debugf2("PCI SEC_STATUS= 0x%04x %s\n", status, dev->dev.bus_id); + + /* check the secondary status reg for errors */ + if (status) { + if (status & (PCI_STATUS_SIG_SYSTEM_ERROR)) { + edac_printk(KERN_CRIT, EDAC_PCI, "Bridge " + "Signaled System Error on %s\n", + pci_name(dev)); + atomic_inc(&pci_nonparity_count); + } + + if (status & (PCI_STATUS_PARITY)) { + edac_printk(KERN_CRIT, EDAC_PCI, "Bridge " + "Master Data Parity Error on " + "%s\n", pci_name(dev)); + + atomic_inc(&pci_parity_count); + } + + if (status & (PCI_STATUS_DETECTED_PARITY)) { + edac_printk(KERN_CRIT, EDAC_PCI, "Bridge " + "Detected Parity Error on %s\n", + pci_name(dev)); + + atomic_inc(&pci_parity_count); + } + } + } +} + +/* + * pci_dev parity list iterator + * Scan the PCI device list for one iteration, looking for SERRORs + * Master Parity ERRORS or Parity ERRORs on primary or secondary devices + */ +static inline void edac_pci_dev_parity_iterator(pci_parity_check_fn_t fn) +{ + struct pci_dev *dev = NULL; + + /* request for kernel access to the next PCI device, if any, + * and while we are looking at it have its reference count + * bumped until we are done with it + */ + while ((dev = pci_get_device(PCI_ANY_ID, PCI_ANY_ID, dev)) != NULL) { + fn(dev); + } +} + +/* + * edac_pci_do_parity_check + * + * performs the actual PCI parity check operation + */ +void edac_pci_do_parity_check(void) +{ + unsigned long flags; + int before_count; + + debugf3("%s()\n", __func__); + + if (!check_pci_errors) + return; + + before_count = atomic_read(&pci_parity_count); + + /* scan all PCI devices looking for a Parity Error on devices and + * bridges + */ + local_irq_save(flags); + edac_pci_dev_parity_iterator(edac_pci_dev_parity_test); + local_irq_restore(flags); + + /* Only if operator has selected panic on PCI Error */ + if (edac_pci_get_panic_on_pe()) { + /* If the count is different 'after' from 'before' */ + if (before_count != atomic_read(&pci_parity_count)) + panic("EDAC: PCI Parity Error"); + } +} + +void edac_pci_clear_parity_errors(void) +{ + /* Clear any PCI bus parity errors that devices initially have logged + * in their registers. + */ + edac_pci_dev_parity_iterator(edac_pci_dev_parity_clear); +} +void edac_pci_handle_pe(struct edac_pci_ctl_info *pci, const char *msg) +{ + + /* global PE counter incremented by edac_pci_do_parity_check() */ + atomic_inc(&pci->counters.pe_count); + + if (edac_pci_get_log_pe()) + edac_pci_printk(pci, KERN_WARNING, + "Parity Error ctl: %s %d: %s\n", + pci->ctl_name, pci->pci_idx, msg); + + /* + * poke all PCI devices and see which one is the troublemaker + * panic() is called if set + */ + edac_pci_do_parity_check(); +} + +EXPORT_SYMBOL_GPL(edac_pci_handle_pe); + +void edac_pci_handle_npe(struct edac_pci_ctl_info *pci, const char *msg) +{ + + /* global NPE counter incremented by edac_pci_do_parity_check() */ + atomic_inc(&pci->counters.npe_count); + + if (edac_pci_get_log_npe()) + edac_pci_printk(pci, KERN_WARNING, + "Non-Parity Error ctl: %s %d: %s\n", + pci->ctl_name, pci->pci_idx, msg); + + /* + * poke all PCI devices and see which one is the troublemaker + * panic() is called if set + */ + edac_pci_do_parity_check(); +} + +EXPORT_SYMBOL_GPL(edac_pci_handle_npe); + +/* + * Define the PCI parameter to the module + */ +module_param(check_pci_errors, int, 0644); +MODULE_PARM_DESC(check_pci_errors, + "Check for PCI bus parity errors: 0=off 1=on"); +module_param(edac_pci_panic_on_pe, int, 0644); +MODULE_PARM_DESC(edac_pci_panic_on_pe, + "Panic on PCI Bus Parity error: 0=off 1=on"); + +#endif /* CONFIG_PCI */ diff --git a/drivers/edac/edac_stub.c b/drivers/edac/edac_stub.c new file mode 100644 index 00000000000..20b428aa155 --- /dev/null +++ b/drivers/edac/edac_stub.c @@ -0,0 +1,46 @@ +/* + * common EDAC components that must be in kernel + * + * Author: Dave Jiang <djiang@mvista.com> + * + * 2007 (c) MontaVista Software, Inc. This file is licensed under + * the terms of the GNU General Public License version 2. This program + * is licensed "as is" without any warranty of any kind, whether express + * or implied. + * + */ +#include <linux/module.h> +#include <linux/edac.h> +#include <asm/atomic.h> +#include <asm/edac.h> + +int edac_op_state = EDAC_OPSTATE_INVAL; +EXPORT_SYMBOL_GPL(edac_op_state); + +atomic_t edac_handlers = ATOMIC_INIT(0); +EXPORT_SYMBOL_GPL(edac_handlers); + +int edac_err_assert = 0; +EXPORT_SYMBOL_GPL(edac_err_assert); + +/* + * called to determine if there is an EDAC driver interested in + * knowing an event (such as NMI) occurred + */ +int edac_handler_set(void) +{ + if (edac_op_state == EDAC_OPSTATE_POLL) + return 0; + + return atomic_read(&edac_handlers); +} +EXPORT_SYMBOL_GPL(edac_handler_set); + +/* + * handler for NMI type of interrupts to assert error + */ +void edac_atomic_assert_error(void) +{ + edac_err_assert++; +} +EXPORT_SYMBOL_GPL(edac_atomic_assert_error); diff --git a/drivers/edac/i3000_edac.c b/drivers/edac/i3000_edac.c new file mode 100644 index 00000000000..0ecfdc432f8 --- /dev/null +++ b/drivers/edac/i3000_edac.c @@ -0,0 +1,506 @@ +/* + * Intel 3000/3010 Memory Controller kernel module + * Copyright (C) 2007 Akamai Technologies, Inc. + * Shamelessly copied from: + * Intel D82875P Memory Controller kernel module + * (C) 2003 Linux Networx (http://lnxi.com) + * + * This file may be distributed under the terms of the + * GNU General Public License. + */ + +#include <linux/module.h> +#include <linux/init.h> +#include <linux/pci.h> +#include <linux/pci_ids.h> +#include <linux/slab.h> +#include "edac_core.h" + +#define I3000_REVISION "1.1" + +#define EDAC_MOD_STR "i3000_edac" + +#define I3000_RANKS 8 +#define I3000_RANKS_PER_CHANNEL 4 +#define I3000_CHANNELS 2 + +/* Intel 3000 register addresses - device 0 function 0 - DRAM Controller */ + +#define I3000_MCHBAR 0x44 /* MCH Memory Mapped Register BAR */ +#define I3000_MCHBAR_MASK 0xffffc000 +#define I3000_MMR_WINDOW_SIZE 16384 + +#define I3000_EDEAP 0x70 /* Extended DRAM Error Address Pointer (8b) + * + * 7:1 reserved + * 0 bit 32 of address + */ +#define I3000_DEAP 0x58 /* DRAM Error Address Pointer (32b) + * + * 31:7 address + * 6:1 reserved + * 0 Error channel 0/1 + */ +#define I3000_DEAP_GRAIN (1 << 7) +#define I3000_DEAP_PFN(edeap, deap) ((((edeap) & 1) << (32 - PAGE_SHIFT)) | \ + ((deap) >> PAGE_SHIFT)) +#define I3000_DEAP_OFFSET(deap) ((deap) & ~(I3000_DEAP_GRAIN-1) & ~PAGE_MASK) +#define I3000_DEAP_CHANNEL(deap) ((deap) & 1) + +#define I3000_DERRSYN 0x5c /* DRAM Error Syndrome (8b) + * + * 7:0 DRAM ECC Syndrome + */ + +#define I3000_ERRSTS 0xc8 /* Error Status Register (16b) + * + * 15:12 reserved + * 11 MCH Thermal Sensor Event for SMI/SCI/SERR + * 10 reserved + * 9 LOCK to non-DRAM Memory Flag (LCKF) + * 8 Received Refresh Timeout Flag (RRTOF) + * 7:2 reserved + * 1 Multiple-bit DRAM ECC Error Flag (DMERR) + * 0 Single-bit DRAM ECC Error Flag (DSERR) + */ +#define I3000_ERRSTS_BITS 0x0b03 /* bits which indicate errors */ +#define I3000_ERRSTS_UE 0x0002 +#define I3000_ERRSTS_CE 0x0001 + +#define I3000_ERRCMD 0xca /* Error Command (16b) + * + * 15:12 reserved + * 11 SERR on MCH Thermal Sensor Event (TSESERR) + * 10 reserved + * 9 SERR on LOCK to non-DRAM Memory (LCKERR) + * 8 SERR on DRAM Refresh Timeout (DRTOERR) + * 7:2 reserved + * 1 SERR Multiple-Bit DRAM ECC Error (DMERR) + * 0 SERR on Single-Bit ECC Error (DSERR) + */ + +/* Intel MMIO register space - device 0 function 0 - MMR space */ + +#define I3000_DRB_SHIFT 25 /* 32MiB grain */ + +#define I3000_C0DRB 0x100 /* Channel 0 DRAM Rank Boundary (8b x 4) + * + * 7:0 Channel 0 DRAM Rank Boundary Address + */ +#define I3000_C1DRB 0x180 /* Channel 1 DRAM Rank Boundary (8b x 4) + * + * 7:0 Channel 1 DRAM Rank Boundary Address + */ + +#define I3000_C0DRA 0x108 /* Channel 0 DRAM Rank Attribute (8b x 2) + * + * 7 reserved + * 6:4 DRAM odd Rank Attribute + * 3 reserved + * 2:0 DRAM even Rank Attribute + * + * Each attribute defines the page + * size of the corresponding rank: + * 000: unpopulated + * 001: reserved + * 010: 4 KB + * 011: 8 KB + * 100: 16 KB + * Others: reserved + */ +#define I3000_C1DRA 0x188 /* Channel 1 DRAM Rank Attribute (8b x 2) */ +#define ODD_RANK_ATTRIB(dra) (((dra) & 0x70) >> 4) +#define EVEN_RANK_ATTRIB(dra) ((dra) & 0x07) + +#define I3000_C0DRC0 0x120 /* DRAM Controller Mode 0 (32b) + * + * 31:30 reserved + * 29 Initialization Complete (IC) + * 28:11 reserved + * 10:8 Refresh Mode Select (RMS) + * 7 reserved + * 6:4 Mode Select (SMS) + * 3:2 reserved + * 1:0 DRAM Type (DT) + */ + +#define I3000_C0DRC1 0x124 /* DRAM Controller Mode 1 (32b) + * + * 31 Enhanced Addressing Enable (ENHADE) + * 30:0 reserved + */ + +enum i3000p_chips { + I3000 = 0, +}; + +struct i3000_dev_info { + const char *ctl_name; +}; + +struct i3000_error_info { + u16 errsts; + u8 derrsyn; + u8 edeap; + u32 deap; + u16 errsts2; +}; + +static const struct i3000_dev_info i3000_devs[] = { + [I3000] = { + .ctl_name = "i3000"}, +}; + +static struct pci_dev *mci_pdev; +static int i3000_registered = 1; +static struct edac_pci_ctl_info *i3000_pci; + +static void i3000_get_error_info(struct mem_ctl_info *mci, + struct i3000_error_info *info) +{ + struct pci_dev *pdev; + + pdev = to_pci_dev(mci->dev); + + /* + * This is a mess because there is no atomic way to read all the + * registers at once and the registers can transition from CE being + * overwritten by UE. + */ + pci_read_config_word(pdev, I3000_ERRSTS, &info->errsts); + if (!(info->errsts & I3000_ERRSTS_BITS)) + return; + pci_read_config_byte(pdev, I3000_EDEAP, &info->edeap); + pci_read_config_dword(pdev, I3000_DEAP, &info->deap); + pci_read_config_byte(pdev, I3000_DERRSYN, &info->derrsyn); + pci_read_config_word(pdev, I3000_ERRSTS, &info->errsts2); + + /* + * If the error is the same for both reads then the first set + * of reads is valid. If there is a change then there is a CE + * with no info and the second set of reads is valid and + * should be UE info. + */ + if ((info->errsts ^ info->errsts2) & I3000_ERRSTS_BITS) { + pci_read_config_byte(pdev, I3000_EDEAP, &info->edeap); + pci_read_config_dword(pdev, I3000_DEAP, &info->deap); + pci_read_config_byte(pdev, I3000_DERRSYN, &info->derrsyn); + } + + /* Clear any error bits. + * (Yes, we really clear bits by writing 1 to them.) + */ + pci_write_bits16(pdev, I3000_ERRSTS, I3000_ERRSTS_BITS, + I3000_ERRSTS_BITS); +} + +static int i3000_process_error_info(struct mem_ctl_info *mci, + struct i3000_error_info *info, + int handle_errors) +{ + int row, multi_chan; + int pfn, offset, channel; + + multi_chan = mci->csrows[0].nr_channels - 1; + + if (!(info->errsts & I3000_ERRSTS_BITS)) + return 0; + + if (!handle_errors) + return 1; + + if ((info->errsts ^ info->errsts2) & I3000_ERRSTS_BITS) { + edac_mc_handle_ce_no_info(mci, "UE overwrote CE"); + info->errsts = info->errsts2; + } + + pfn = I3000_DEAP_PFN(info->edeap, info->deap); + offset = I3000_DEAP_OFFSET(info->deap); + channel = I3000_DEAP_CHANNEL(info->deap); + + row = edac_mc_find_csrow_by_page(mci, pfn); + + if (info->errsts & I3000_ERRSTS_UE) + edac_mc_handle_ue(mci, pfn, offset, row, "i3000 UE"); + else + edac_mc_handle_ce(mci, pfn, offset, info->derrsyn, row, + multi_chan ? channel : 0, "i3000 CE"); + + return 1; +} + +static void i3000_check(struct mem_ctl_info *mci) +{ + struct i3000_error_info info; + + debugf1("MC%d: %s()\n", mci->mc_idx, __func__); + i3000_get_error_info(mci, &info); + i3000_process_error_info(mci, &info, 1); +} + +static int i3000_is_interleaved(const unsigned char *c0dra, + const unsigned char *c1dra, + const unsigned char *c0drb, + const unsigned char *c1drb) +{ + int i; + + /* If the channels aren't populated identically then + * we're not interleaved. + */ + for (i = 0; i < I3000_RANKS_PER_CHANNEL / 2; i++) + if (ODD_RANK_ATTRIB(c0dra[i]) != ODD_RANK_ATTRIB(c1dra[i]) || + EVEN_RANK_ATTRIB(c0dra[i]) != + EVEN_RANK_ATTRIB(c1dra[i])) + return 0; + + /* If the rank boundaries for the two channels are different + * then we're not interleaved. + */ + for (i = 0; i < I3000_RANKS_PER_CHANNEL; i++) + if (c0drb[i] != c1drb[i]) + return 0; + + return 1; +} + +static int i3000_probe1(struct pci_dev *pdev, int dev_idx) +{ + int rc; + int i; + struct mem_ctl_info *mci = NULL; + unsigned long last_cumul_size; + int interleaved, nr_channels; + unsigned char dra[I3000_RANKS / 2], drb[I3000_RANKS]; + unsigned char *c0dra = dra, *c1dra = &dra[I3000_RANKS_PER_CHANNEL / 2]; + unsigned char *c0drb = drb, *c1drb = &drb[I3000_RANKS_PER_CHANNEL]; + unsigned long mchbar; + void *window; + + debugf0("MC: %s()\n", __func__); + + pci_read_config_dword(pdev, I3000_MCHBAR, (u32 *) & mchbar); + mchbar &= I3000_MCHBAR_MASK; + window = ioremap_nocache(mchbar, I3000_MMR_WINDOW_SIZE); + if (!window) { + printk(KERN_ERR "i3000: cannot map mmio space at 0x%lx\n", + mchbar); + return -ENODEV; + } + + c0dra[0] = readb(window + I3000_C0DRA + 0); /* ranks 0,1 */ + c0dra[1] = readb(window + I3000_C0DRA + 1); /* ranks 2,3 */ + c1dra[0] = readb(window + I3000_C1DRA + 0); /* ranks 0,1 */ + c1dra[1] = readb(window + I3000_C1DRA + 1); /* ranks 2,3 */ + + for (i = 0; i < I3000_RANKS_PER_CHANNEL; i++) { + c0drb[i] = readb(window + I3000_C0DRB + i); + c1drb[i] = readb(window + I3000_C1DRB + i); + } + + iounmap(window); + + /* Figure out how many channels we have. + * + * If we have what the datasheet calls "asymmetric channels" + * (essentially the same as what was called "virtual single + * channel mode" in the i82875) then it's a single channel as + * far as EDAC is concerned. + */ + interleaved = i3000_is_interleaved(c0dra, c1dra, c0drb, c1drb); + nr_channels = interleaved ? 2 : 1; + mci = edac_mc_alloc(0, I3000_RANKS / nr_channels, nr_channels, 0); + if (!mci) + return -ENOMEM; + + debugf3("MC: %s(): init mci\n", __func__); + + mci->dev = &pdev->dev; + mci->mtype_cap = MEM_FLAG_DDR2; + + mci->edac_ctl_cap = EDAC_FLAG_SECDED; + mci->edac_cap = EDAC_FLAG_SECDED; + + mci->mod_name = EDAC_MOD_STR; + mci->mod_ver = I3000_REVISION; + mci->ctl_name = i3000_devs[dev_idx].ctl_name; + mci->dev_name = pci_name(pdev); + mci->edac_check = i3000_check; + mci->ctl_page_to_phys = NULL; + + /* + * The dram rank boundary (DRB) reg values are boundary addresses + * for each DRAM rank with a granularity of 32MB. DRB regs are + * cumulative; the last one will contain the total memory + * contained in all ranks. + * + * If we're in interleaved mode then we're only walking through + * the ranks of controller 0, so we double all the values we see. + */ + for (last_cumul_size = i = 0; i < mci->nr_csrows; i++) { + u8 value; + u32 cumul_size; + struct csrow_info *csrow = &mci->csrows[i]; + + value = drb[i]; + cumul_size = value << (I3000_DRB_SHIFT - PAGE_SHIFT); + if (interleaved) + cumul_size <<= 1; + debugf3("MC: %s(): (%d) cumul_size 0x%x\n", + __func__, i, cumul_size); + if (cumul_size == last_cumul_size) { + csrow->mtype = MEM_EMPTY; + continue; + } + + csrow->first_page = last_cumul_size; + csrow->last_page = cumul_size - 1; + csrow->nr_pages = cumul_size - last_cumul_size; + last_cumul_size = cumul_size; + csrow->grain = I3000_DEAP_GRAIN; + csrow->mtype = MEM_DDR2; + csrow->dtype = DEV_UNKNOWN; + csrow->edac_mode = EDAC_UNKNOWN; + } + + /* Clear any error bits. + * (Yes, we really clear bits by writing 1 to them.) + */ + pci_write_bits16(pdev, I3000_ERRSTS, I3000_ERRSTS_BITS, + I3000_ERRSTS_BITS); + + rc = -ENODEV; + if (edac_mc_add_mc(mci)) { + debugf3("MC: %s(): failed edac_mc_add_mc()\n", __func__); + goto fail; + } + + /* allocating generic PCI control info */ + i3000_pci = edac_pci_create_generic_ctl(&pdev->dev, EDAC_MOD_STR); + if (!i3000_pci) { + printk(KERN_WARNING + "%s(): Unable to create PCI control\n", + __func__); + printk(KERN_WARNING + "%s(): PCI error report via EDAC not setup\n", + __func__); + } + + /* get this far and it's successful */ + debugf3("MC: %s(): success\n", __func__); + return 0; + + fail: + if (mci) + edac_mc_free(mci); + + return rc; +} + +/* returns count (>= 0), or negative on error */ +static int __devinit i3000_init_one(struct pci_dev *pdev, + const struct pci_device_id *ent) +{ + int rc; + + debugf0("MC: %s()\n", __func__); + + if (pci_enable_device(pdev) < 0) + return -EIO; + + rc = i3000_probe1(pdev, ent->driver_data); + if (mci_pdev == NULL) + mci_pdev = pci_dev_get(pdev); + + return rc; +} + +static void __devexit i3000_remove_one(struct pci_dev *pdev) +{ + struct mem_ctl_info *mci; + + debugf0("%s()\n", __func__); + + if (i3000_pci) + edac_pci_release_generic_ctl(i3000_pci); + + if ((mci = edac_mc_del_mc(&pdev->dev)) == NULL) + return; + + edac_mc_free(mci); +} + +static const struct pci_device_id i3000_pci_tbl[] __devinitdata = { + { + PCI_VEND_DEV(INTEL, 3000_HB), PCI_ANY_ID, PCI_ANY_ID, 0, 0, + I3000}, + { + 0, + } /* 0 terminated list. */ +}; + +MODULE_DEVICE_TABLE(pci, i3000_pci_tbl); + +static struct pci_driver i3000_driver = { + .name = EDAC_MOD_STR, + .probe = i3000_init_one, + .remove = __devexit_p(i3000_remove_one), + .id_table = i3000_pci_tbl, +}; + +static int __init i3000_init(void) +{ + int pci_rc; + + debugf3("MC: %s()\n", __func__); + pci_rc = pci_register_driver(&i3000_driver); + if (pci_rc < 0) + goto fail0; + + if (mci_pdev == NULL) { + i3000_registered = 0; + mci_pdev = pci_get_device(PCI_VENDOR_ID_INTEL, + PCI_DEVICE_ID_INTEL_3000_HB, NULL); + if (!mci_pdev) { + debugf0("i3000 pci_get_device fail\n"); + pci_rc = -ENODEV; + goto fail1; + } + + pci_rc = i3000_init_one(mci_pdev, i3000_pci_tbl); + if (pci_rc < 0) { + debugf0("i3000 init fail\n"); + pci_rc = -ENODEV; + goto fail1; + } + } + + return 0; + +fail1: + pci_unregister_driver(&i3000_driver); + +fail0: + if (mci_pdev) + pci_dev_put(mci_pdev); + + return pci_rc; +} + +static void __exit i3000_exit(void) +{ + debugf3("MC: %s()\n", __func__); + + pci_unregister_driver(&i3000_driver); + if (!i3000_registered) { + i3000_remove_one(mci_pdev); + pci_dev_put(mci_pdev); + } +} + +module_init(i3000_init); +module_exit(i3000_exit); + +MODULE_LICENSE("GPL"); +MODULE_AUTHOR("Akamai Technologies Arthur Ulfeldt/Jason Uhlenkott"); +MODULE_DESCRIPTION("MC support for Intel 3000 memory hub controllers"); diff --git a/drivers/edac/i5000_edac.c b/drivers/edac/i5000_edac.c new file mode 100644 index 00000000000..96f7e63e399 --- /dev/null +++ b/drivers/edac/i5000_edac.c @@ -0,0 +1,1505 @@ +/* + * Intel 5000(P/V/X) class Memory Controllers kernel module + * + * This file may be distributed under the terms of the + * GNU General Public License. + * + * Written by Douglas Thompson Linux Networx (http://lnxi.com) + * norsk5@xmission.com + * + * This module is based on the following document: + * + * Intel 5000X Chipset Memory Controller Hub (MCH) - Datasheet + * http://developer.intel.com/design/chipsets/datashts/313070.htm + * + */ + +#include <linux/module.h> +#include <linux/init.h> +#include <linux/pci.h> +#include <linux/pci_ids.h> +#include <linux/slab.h> +#include <linux/edac.h> +#include <asm/mmzone.h> + +#include "edac_core.h" + +/* + * Alter this version for the I5000 module when modifications are made + */ +#define I5000_REVISION " Ver: 2.0.12 " __DATE__ +#define EDAC_MOD_STR "i5000_edac" + +#define i5000_printk(level, fmt, arg...) \ + edac_printk(level, "i5000", fmt, ##arg) + +#define i5000_mc_printk(mci, level, fmt, arg...) \ + edac_mc_chipset_printk(mci, level, "i5000", fmt, ##arg) + +#ifndef PCI_DEVICE_ID_INTEL_FBD_0 +#define PCI_DEVICE_ID_INTEL_FBD_0 0x25F5 +#endif +#ifndef PCI_DEVICE_ID_INTEL_FBD_1 +#define PCI_DEVICE_ID_INTEL_FBD_1 0x25F6 +#endif + +/* Device 16, + * Function 0: System Address + * Function 1: Memory Branch Map, Control, Errors Register + * Function 2: FSB Error Registers + * + * All 3 functions of Device 16 (0,1,2) share the SAME DID + */ +#define PCI_DEVICE_ID_INTEL_I5000_DEV16 0x25F0 + +/* OFFSETS for Function 0 */ + +/* OFFSETS for Function 1 */ +#define AMBASE 0x48 +#define MAXCH 0x56 +#define MAXDIMMPERCH 0x57 +#define TOLM 0x6C +#define REDMEMB 0x7C +#define RED_ECC_LOCATOR(x) ((x) & 0x3FFFF) +#define REC_ECC_LOCATOR_EVEN(x) ((x) & 0x001FF) +#define REC_ECC_LOCATOR_ODD(x) ((x) & 0x3FE00) +#define MIR0 0x80 +#define MIR1 0x84 +#define MIR2 0x88 +#define AMIR0 0x8C +#define AMIR1 0x90 +#define AMIR2 0x94 + +#define FERR_FAT_FBD 0x98 +#define NERR_FAT_FBD 0x9C +#define EXTRACT_FBDCHAN_INDX(x) (((x)>>28) & 0x3) +#define FERR_FAT_FBDCHAN 0x30000000 +#define FERR_FAT_M3ERR 0x00000004 +#define FERR_FAT_M2ERR 0x00000002 +#define FERR_FAT_M1ERR 0x00000001 +#define FERR_FAT_MASK (FERR_FAT_M1ERR | \ + FERR_FAT_M2ERR | \ + FERR_FAT_M3ERR) + +#define FERR_NF_FBD 0xA0 + +/* Thermal and SPD or BFD errors */ +#define FERR_NF_M28ERR 0x01000000 +#define FERR_NF_M27ERR 0x00800000 +#define FERR_NF_M26ERR 0x00400000 +#define FERR_NF_M25ERR 0x00200000 +#define FERR_NF_M24ERR 0x00100000 +#define FERR_NF_M23ERR 0x00080000 +#define FERR_NF_M22ERR 0x00040000 +#define FERR_NF_M21ERR 0x00020000 + +/* Correctable errors */ +#define FERR_NF_M20ERR 0x00010000 +#define FERR_NF_M19ERR 0x00008000 +#define FERR_NF_M18ERR 0x00004000 +#define FERR_NF_M17ERR 0x00002000 + +/* Non-Retry or redundant Retry errors */ +#define FERR_NF_M16ERR 0x00001000 +#define FERR_NF_M15ERR 0x00000800 +#define FERR_NF_M14ERR 0x00000400 +#define FERR_NF_M13ERR 0x00000200 + +/* Uncorrectable errors */ +#define FERR_NF_M12ERR 0x00000100 +#define FERR_NF_M11ERR 0x00000080 +#define FERR_NF_M10ERR 0x00000040 +#define FERR_NF_M9ERR 0x00000020 +#define FERR_NF_M8ERR 0x00000010 +#define FERR_NF_M7ERR 0x00000008 +#define FERR_NF_M6ERR 0x00000004 +#define FERR_NF_M5ERR 0x00000002 +#define FERR_NF_M4ERR 0x00000001 + +#define FERR_NF_UNCORRECTABLE (FERR_NF_M12ERR | \ + FERR_NF_M11ERR | \ + FERR_NF_M10ERR | \ + FERR_NF_M8ERR | \ + FERR_NF_M7ERR | \ + FERR_NF_M6ERR | \ + FERR_NF_M5ERR | \ + FERR_NF_M4ERR) +#define FERR_NF_CORRECTABLE (FERR_NF_M20ERR | \ + FERR_NF_M19ERR | \ + FERR_NF_M18ERR | \ + FERR_NF_M17ERR) +#define FERR_NF_DIMM_SPARE (FERR_NF_M27ERR | \ + FERR_NF_M28ERR) +#define FERR_NF_THERMAL (FERR_NF_M26ERR | \ + FERR_NF_M25ERR | \ + FERR_NF_M24ERR | \ + FERR_NF_M23ERR) +#define FERR_NF_SPD_PROTOCOL (FERR_NF_M22ERR) +#define FERR_NF_NORTH_CRC (FERR_NF_M21ERR) +#define FERR_NF_NON_RETRY (FERR_NF_M13ERR | \ + FERR_NF_M14ERR | \ + FERR_NF_M15ERR) + +#define NERR_NF_FBD 0xA4 +#define FERR_NF_MASK (FERR_NF_UNCORRECTABLE | \ + FERR_NF_CORRECTABLE | \ + FERR_NF_DIMM_SPARE | \ + FERR_NF_THERMAL | \ + FERR_NF_SPD_PROTOCOL | \ + FERR_NF_NORTH_CRC | \ + FERR_NF_NON_RETRY) + +#define EMASK_FBD 0xA8 +#define EMASK_FBD_M28ERR 0x08000000 +#define EMASK_FBD_M27ERR 0x04000000 +#define EMASK_FBD_M26ERR 0x02000000 +#define EMASK_FBD_M25ERR 0x01000000 +#define EMASK_FBD_M24ERR 0x00800000 +#define EMASK_FBD_M23ERR 0x00400000 +#define EMASK_FBD_M22ERR 0x00200000 +#define EMASK_FBD_M21ERR 0x00100000 +#define EMASK_FBD_M20ERR 0x00080000 +#define EMASK_FBD_M19ERR 0x00040000 +#define EMASK_FBD_M18ERR 0x00020000 +#define EMASK_FBD_M17ERR 0x00010000 + +#define EMASK_FBD_M15ERR 0x00004000 +#define EMASK_FBD_M14ERR 0x00002000 +#define EMASK_FBD_M13ERR 0x00001000 +#define EMASK_FBD_M12ERR 0x00000800 +#define EMASK_FBD_M11ERR 0x00000400 +#define EMASK_FBD_M10ERR 0x00000200 +#define EMASK_FBD_M9ERR 0x00000100 +#define EMASK_FBD_M8ERR 0x00000080 +#define EMASK_FBD_M7ERR 0x00000040 +#define EMASK_FBD_M6ERR 0x00000020 +#define EMASK_FBD_M5ERR 0x00000010 +#define EMASK_FBD_M4ERR 0x00000008 +#define EMASK_FBD_M3ERR 0x00000004 +#define EMASK_FBD_M2ERR 0x00000002 +#define EMASK_FBD_M1ERR 0x00000001 + +#define ENABLE_EMASK_FBD_FATAL_ERRORS (EMASK_FBD_M1ERR | \ + EMASK_FBD_M2ERR | \ + EMASK_FBD_M3ERR) + +#define ENABLE_EMASK_FBD_UNCORRECTABLE (EMASK_FBD_M4ERR | \ + EMASK_FBD_M5ERR | \ + EMASK_FBD_M6ERR | \ + EMASK_FBD_M7ERR | \ + EMASK_FBD_M8ERR | \ + EMASK_FBD_M9ERR | \ + EMASK_FBD_M10ERR | \ + EMASK_FBD_M11ERR | \ + EMASK_FBD_M12ERR) +#define ENABLE_EMASK_FBD_CORRECTABLE (EMASK_FBD_M17ERR | \ + EMASK_FBD_M18ERR | \ + EMASK_FBD_M19ERR | \ + EMASK_FBD_M20ERR) +#define ENABLE_EMASK_FBD_DIMM_SPARE (EMASK_FBD_M27ERR | \ + EMASK_FBD_M28ERR) +#define ENABLE_EMASK_FBD_THERMALS (EMASK_FBD_M26ERR | \ + EMASK_FBD_M25ERR | \ + EMASK_FBD_M24ERR | \ + EMASK_FBD_M23ERR) +#define ENABLE_EMASK_FBD_SPD_PROTOCOL (EMASK_FBD_M22ERR) +#define ENABLE_EMASK_FBD_NORTH_CRC (EMASK_FBD_M21ERR) +#define ENABLE_EMASK_FBD_NON_RETRY (EMASK_FBD_M15ERR | \ + EMASK_FBD_M14ERR | \ + EMASK_FBD_M13ERR) + +#define ENABLE_EMASK_ALL (ENABLE_EMASK_FBD_NON_RETRY | \ + ENABLE_EMASK_FBD_NORTH_CRC | \ + ENABLE_EMASK_FBD_SPD_PROTOCOL | \ + ENABLE_EMASK_FBD_THERMALS | \ + ENABLE_EMASK_FBD_DIMM_SPARE | \ + ENABLE_EMASK_FBD_FATAL_ERRORS | \ + ENABLE_EMASK_FBD_CORRECTABLE | \ + ENABLE_EMASK_FBD_UNCORRECTABLE) + +#define ERR0_FBD 0xAC +#define ERR1_FBD 0xB0 +#define ERR2_FBD 0xB4 +#define MCERR_FBD 0xB8 +#define NRECMEMA 0xBE +#define NREC_BANK(x) (((x)>>12) & 0x7) +#define NREC_RDWR(x) (((x)>>11) & 1) +#define NREC_RANK(x) (((x)>>8) & 0x7) +#define NRECMEMB 0xC0 +#define NREC_CAS(x) (((x)>>16) & 0xFFFFFF) +#define NREC_RAS(x) ((x) & 0x7FFF) +#define NRECFGLOG 0xC4 +#define NREEECFBDA 0xC8 +#define NREEECFBDB 0xCC +#define NREEECFBDC 0xD0 +#define NREEECFBDD 0xD4 +#define NREEECFBDE 0xD8 +#define REDMEMA 0xDC +#define RECMEMA 0xE2 +#define REC_BANK(x) (((x)>>12) & 0x7) +#define REC_RDWR(x) (((x)>>11) & 1) +#define REC_RANK(x) (((x)>>8) & 0x7) +#define RECMEMB 0xE4 +#define REC_CAS(x) (((x)>>16) & 0xFFFFFF) +#define REC_RAS(x) ((x) & 0x7FFF) +#define RECFGLOG 0xE8 +#define RECFBDA 0xEC +#define RECFBDB 0xF0 +#define RECFBDC 0xF4 +#define RECFBDD 0xF8 +#define RECFBDE 0xFC + +/* OFFSETS for Function 2 */ + +/* + * Device 21, + * Function 0: Memory Map Branch 0 + * + * Device 22, + * Function 0: Memory Map Branch 1 + */ +#define PCI_DEVICE_ID_I5000_BRANCH_0 0x25F5 +#define PCI_DEVICE_ID_I5000_BRANCH_1 0x25F6 + +#define AMB_PRESENT_0 0x64 +#define AMB_PRESENT_1 0x66 +#define MTR0 0x80 +#define MTR1 0x84 +#define MTR2 0x88 +#define MTR3 0x8C + +#define NUM_MTRS 4 +#define CHANNELS_PER_BRANCH (2) + +/* Defines to extract the vaious fields from the + * MTRx - Memory Technology Registers + */ +#define MTR_DIMMS_PRESENT(mtr) ((mtr) & (0x1 << 8)) +#define MTR_DRAM_WIDTH(mtr) ((((mtr) >> 6) & 0x1) ? 8 : 4) +#define MTR_DRAM_BANKS(mtr) ((((mtr) >> 5) & 0x1) ? 8 : 4) +#define MTR_DRAM_BANKS_ADDR_BITS(mtr) ((MTR_DRAM_BANKS(mtr) == 8) ? 3 : 2) +#define MTR_DIMM_RANK(mtr) (((mtr) >> 4) & 0x1) +#define MTR_DIMM_RANK_ADDR_BITS(mtr) (MTR_DIMM_RANK(mtr) ? 2 : 1) +#define MTR_DIMM_ROWS(mtr) (((mtr) >> 2) & 0x3) +#define MTR_DIMM_ROWS_ADDR_BITS(mtr) (MTR_DIMM_ROWS(mtr) + 13) +#define MTR_DIMM_COLS(mtr) ((mtr) & 0x3) +#define MTR_DIMM_COLS_ADDR_BITS(mtr) (MTR_DIMM_COLS(mtr) + 10) + +#ifdef CONFIG_EDAC_DEBUG +static char *numrow_toString[] = { + "8,192 - 13 rows", + "16,384 - 14 rows", + "32,768 - 15 rows", + "reserved" +}; + +static char *numcol_toString[] = { + "1,024 - 10 columns", + "2,048 - 11 columns", + "4,096 - 12 columns", + "reserved" +}; +#endif + +/* Enumeration of supported devices */ +enum i5000_chips { + I5000P = 0, + I5000V = 1, /* future */ + I5000X = 2 /* future */ +}; + +/* Device name and register DID (Device ID) */ +struct i5000_dev_info { + const char *ctl_name; /* name for this device */ + u16 fsb_mapping_errors; /* DID for the branchmap,control */ +}; + +/* Table of devices attributes supported by this driver */ +static const struct i5000_dev_info i5000_devs[] = { + [I5000P] = { + .ctl_name = "I5000", + .fsb_mapping_errors = PCI_DEVICE_ID_INTEL_I5000_DEV16, + }, +}; + +struct i5000_dimm_info { + int megabytes; /* size, 0 means not present */ + int dual_rank; +}; + +#define MAX_CHANNELS 6 /* max possible channels */ +#define MAX_CSROWS (8*2) /* max possible csrows per channel */ + +/* driver private data structure */ +struct i5000_pvt { + struct pci_dev *system_address; /* 16.0 */ + struct pci_dev *branchmap_werrors; /* 16.1 */ + struct pci_dev *fsb_error_regs; /* 16.2 */ + struct pci_dev *branch_0; /* 21.0 */ + struct pci_dev *branch_1; /* 22.0 */ + + u16 tolm; /* top of low memory */ + u64 ambase; /* AMB BAR */ + + u16 mir0, mir1, mir2; + + u16 b0_mtr[NUM_MTRS]; /* Memory Technlogy Reg */ + u16 b0_ambpresent0; /* Branch 0, Channel 0 */ + u16 b0_ambpresent1; /* Brnach 0, Channel 1 */ + + u16 b1_mtr[NUM_MTRS]; /* Memory Technlogy Reg */ + u16 b1_ambpresent0; /* Branch 1, Channel 8 */ + u16 b1_ambpresent1; /* Branch 1, Channel 1 */ + + /* DIMM infomation matrix, allocating architecture maximums */ + struct i5000_dimm_info dimm_info[MAX_CSROWS][MAX_CHANNELS]; + + /* Actual values for this controller */ + int maxch; /* Max channels */ + int maxdimmperch; /* Max DIMMs per channel */ +}; + +/* I5000 MCH error information retrieved from Hardware */ +struct i5000_error_info { + + /* These registers are always read from the MC */ + u32 ferr_fat_fbd; /* First Errors Fatal */ + u32 nerr_fat_fbd; /* Next Errors Fatal */ + u32 ferr_nf_fbd; /* First Errors Non-Fatal */ + u32 nerr_nf_fbd; /* Next Errors Non-Fatal */ + + /* These registers are input ONLY if there was a Recoverable Error */ + u32 redmemb; /* Recoverable Mem Data Error log B */ + u16 recmema; /* Recoverable Mem Error log A */ + u32 recmemb; /* Recoverable Mem Error log B */ + + /* These registers are input ONLY if there was a + * Non-Recoverable Error */ + u16 nrecmema; /* Non-Recoverable Mem log A */ + u16 nrecmemb; /* Non-Recoverable Mem log B */ + +}; + +static struct edac_pci_ctl_info *i5000_pci; + +/* + * i5000_get_error_info Retrieve the hardware error information from + * the hardware and cache it in the 'info' + * structure + */ +static void i5000_get_error_info(struct mem_ctl_info *mci, + struct i5000_error_info *info) +{ + struct i5000_pvt *pvt; + u32 value; + + pvt = mci->pvt_info; + + /* read in the 1st FATAL error register */ + pci_read_config_dword(pvt->branchmap_werrors, FERR_FAT_FBD, &value); + + /* Mask only the bits that the doc says are valid + */ + value &= (FERR_FAT_FBDCHAN | FERR_FAT_MASK); + + /* If there is an error, then read in the */ + /* NEXT FATAL error register and the Memory Error Log Register A */ + if (value & FERR_FAT_MASK) { + info->ferr_fat_fbd = value; + + /* harvest the various error data we need */ + pci_read_config_dword(pvt->branchmap_werrors, + NERR_FAT_FBD, &info->nerr_fat_fbd); + pci_read_config_word(pvt->branchmap_werrors, + NRECMEMA, &info->nrecmema); + pci_read_config_word(pvt->branchmap_werrors, + NRECMEMB, &info->nrecmemb); + + /* Clear the error bits, by writing them back */ + pci_write_config_dword(pvt->branchmap_werrors, + FERR_FAT_FBD, value); + } else { + info->ferr_fat_fbd = 0; + info->nerr_fat_fbd = 0; + info->nrecmema = 0; + info->nrecmemb = 0; + } + + /* read in the 1st NON-FATAL error register */ + pci_read_config_dword(pvt->branchmap_werrors, FERR_NF_FBD, &value); + + /* If there is an error, then read in the 1st NON-FATAL error + * register as well */ + if (value & FERR_NF_MASK) { + info->ferr_nf_fbd = value; + + /* harvest the various error data we need */ + pci_read_config_dword(pvt->branchmap_werrors, + NERR_NF_FBD, &info->nerr_nf_fbd); + pci_read_config_word(pvt->branchmap_werrors, + RECMEMA, &info->recmema); + pci_read_config_dword(pvt->branchmap_werrors, + RECMEMB, &info->recmemb); + pci_read_config_dword(pvt->branchmap_werrors, + REDMEMB, &info->redmemb); + + /* Clear the error bits, by writing them back */ + pci_write_config_dword(pvt->branchmap_werrors, + FERR_NF_FBD, value); + } else { + info->ferr_nf_fbd = 0; + info->nerr_nf_fbd = 0; + info->recmema = 0; + info->recmemb = 0; + info->redmemb = 0; + } +} + +/* + * i5000_process_fatal_error_info(struct mem_ctl_info *mci, + * struct i5000_error_info *info, + * int handle_errors); + * + * handle the Intel FATAL errors, if any + */ +static void i5000_process_fatal_error_info(struct mem_ctl_info *mci, + struct i5000_error_info *info, + int handle_errors) +{ + char msg[EDAC_MC_LABEL_LEN + 1 + 90]; + u32 allErrors; + int branch; + int channel; + int bank; + int rank; + int rdwr; + int ras, cas; + + /* mask off the Error bits that are possible */ + allErrors = (info->ferr_fat_fbd & FERR_FAT_MASK); + if (!allErrors) + return; /* if no error, return now */ + + /* ONLY ONE of the possible error bits will be set, as per the docs */ + i5000_mc_printk(mci, KERN_ERR, + "FATAL ERRORS Found!!! 1st FATAL Err Reg= 0x%x\n", + allErrors); + + branch = EXTRACT_FBDCHAN_INDX(info->ferr_fat_fbd); + channel = branch; + + /* Use the NON-Recoverable macros to extract data */ + bank = NREC_BANK(info->nrecmema); + rank = NREC_RANK(info->nrecmema); + rdwr = NREC_RDWR(info->nrecmema); + ras = NREC_RAS(info->nrecmemb); + cas = NREC_CAS(info->nrecmemb); + + debugf0("\t\tCSROW= %d Channels= %d,%d (Branch= %d " + "DRAM Bank= %d rdwr= %s ras= %d cas= %d)\n", + rank, channel, channel + 1, branch >> 1, bank, + rdwr ? "Write" : "Read", ras, cas); + + /* Only 1 bit will be on */ + if (allErrors & FERR_FAT_M1ERR) { + i5000_mc_printk(mci, KERN_ERR, + "Alert on non-redundant retry or fast " + "reset timeout\n"); + + } else if (allErrors & FERR_FAT_M2ERR) { + i5000_mc_printk(mci, KERN_ERR, + "Northbound CRC error on non-redundant " + "retry\n"); + + } else if (allErrors & FERR_FAT_M3ERR) { + i5000_mc_printk(mci, KERN_ERR, + ">Tmid Thermal event with intelligent " + "throttling disabled\n"); + } + + /* Form out message */ + snprintf(msg, sizeof(msg), + "(Branch=%d DRAM-Bank=%d RDWR=%s RAS=%d CAS=%d " + "FATAL Err=0x%x)", + branch >> 1, bank, rdwr ? "Write" : "Read", ras, cas, + allErrors); + + /* Call the helper to output message */ + edac_mc_handle_fbd_ue(mci, rank, channel, channel + 1, msg); +} + +/* + * i5000_process_fatal_error_info(struct mem_ctl_info *mci, + * struct i5000_error_info *info, + * int handle_errors); + * + * handle the Intel NON-FATAL errors, if any + */ +static void i5000_process_nonfatal_error_info(struct mem_ctl_info *mci, + struct i5000_error_info *info, + int handle_errors) +{ + char msg[EDAC_MC_LABEL_LEN + 1 + 90]; + u32 allErrors; + u32 ue_errors; + u32 ce_errors; + u32 misc_errors; + int branch; + int channel; + int bank; + int rank; + int rdwr; + int ras, cas; + + /* mask off the Error bits that are possible */ + allErrors = (info->ferr_nf_fbd & FERR_NF_MASK); + if (!allErrors) + return; /* if no error, return now */ + + /* ONLY ONE of the possible error bits will be set, as per the docs */ + i5000_mc_printk(mci, KERN_WARNING, + "NON-FATAL ERRORS Found!!! 1st NON-FATAL Err " + "Reg= 0x%x\n", allErrors); + + ue_errors = allErrors & FERR_NF_UNCORRECTABLE; + if (ue_errors) { + debugf0("\tUncorrected bits= 0x%x\n", ue_errors); + + branch = EXTRACT_FBDCHAN_INDX(info->ferr_nf_fbd); + channel = branch; + bank = NREC_BANK(info->nrecmema); + rank = NREC_RANK(info->nrecmema); + rdwr = NREC_RDWR(info->nrecmema); + ras = NREC_RAS(info->nrecmemb); + cas = NREC_CAS(info->nrecmemb); + + debugf0 + ("\t\tCSROW= %d Channels= %d,%d (Branch= %d " + "DRAM Bank= %d rdwr= %s ras= %d cas= %d)\n", + rank, channel, channel + 1, branch >> 1, bank, + rdwr ? "Write" : "Read", ras, cas); + + /* Form out message */ + snprintf(msg, sizeof(msg), + "(Branch=%d DRAM-Bank=%d RDWR=%s RAS=%d " + "CAS=%d, UE Err=0x%x)", + branch >> 1, bank, rdwr ? "Write" : "Read", ras, cas, + ue_errors); + + /* Call the helper to output message */ + edac_mc_handle_fbd_ue(mci, rank, channel, channel + 1, msg); + } + + /* Check correctable errors */ + ce_errors = allErrors & FERR_NF_CORRECTABLE; + if (ce_errors) { + debugf0("\tCorrected bits= 0x%x\n", ce_errors); + + branch = EXTRACT_FBDCHAN_INDX(info->ferr_nf_fbd); + + channel = 0; + if (REC_ECC_LOCATOR_ODD(info->redmemb)) + channel = 1; + + /* Convert channel to be based from zero, instead of + * from branch base of 0 */ + channel += branch; + + bank = REC_BANK(info->recmema); + rank = REC_RANK(info->recmema); + rdwr = REC_RDWR(info->recmema); + ras = REC_RAS(info->recmemb); + cas = REC_CAS(info->recmemb); + + debugf0("\t\tCSROW= %d Channel= %d (Branch %d " + "DRAM Bank= %d rdwr= %s ras= %d cas= %d)\n", + rank, channel, branch >> 1, bank, + rdwr ? "Write" : "Read", ras, cas); + + /* Form out message */ + snprintf(msg, sizeof(msg), + "(Branch=%d DRAM-Bank=%d RDWR=%s RAS=%d " + "CAS=%d, CE Err=0x%x)", branch >> 1, bank, + rdwr ? "Write" : "Read", ras, cas, ce_errors); + + /* Call the helper to output message */ + edac_mc_handle_fbd_ce(mci, rank, channel, msg); + } + + /* See if any of the thermal errors have fired */ + misc_errors = allErrors & FERR_NF_THERMAL; + if (misc_errors) { + i5000_printk(KERN_WARNING, "\tTHERMAL Error, bits= 0x%x\n", + misc_errors); + } + + /* See if any of the thermal errors have fired */ + misc_errors = allErrors & FERR_NF_NON_RETRY; + if (misc_errors) { + i5000_printk(KERN_WARNING, "\tNON-Retry Errors, bits= 0x%x\n", + misc_errors); + } + + /* See if any of the thermal errors have fired */ + misc_errors = allErrors & FERR_NF_NORTH_CRC; + if (misc_errors) { + i5000_printk(KERN_WARNING, + "\tNORTHBOUND CRC Error, bits= 0x%x\n", + misc_errors); + } + + /* See if any of the thermal errors have fired */ + misc_errors = allErrors & FERR_NF_SPD_PROTOCOL; + if (misc_errors) { + i5000_printk(KERN_WARNING, + "\tSPD Protocol Error, bits= 0x%x\n", + misc_errors); + } + + /* See if any of the thermal errors have fired */ + misc_errors = allErrors & FERR_NF_DIMM_SPARE; + if (misc_errors) { + i5000_printk(KERN_WARNING, "\tDIMM-Spare Error, bits= 0x%x\n", + misc_errors); + } +} + +/* + * i5000_process_error_info Process the error info that is + * in the 'info' structure, previously retrieved from hardware + */ +static void i5000_process_error_info(struct mem_ctl_info *mci, + struct i5000_error_info *info, + int handle_errors) +{ + /* First handle any fatal errors that occurred */ + i5000_process_fatal_error_info(mci, info, handle_errors); + + /* now handle any non-fatal errors that occurred */ + i5000_process_nonfatal_error_info(mci, info, handle_errors); +} + +/* + * i5000_clear_error Retrieve any error from the hardware + * but do NOT process that error. + * Used for 'clearing' out of previous errors + * Called by the Core module. + */ +static void i5000_clear_error(struct mem_ctl_info *mci) +{ + struct i5000_error_info info; + + i5000_get_error_info(mci, &info); +} + +/* + * i5000_check_error Retrieve and process errors reported by the + * hardware. Called by the Core module. + */ +static void i5000_check_error(struct mem_ctl_info *mci) +{ + struct i5000_error_info info; + debugf4("MC%d: " __FILE__ ": %s()\n", mci->mc_idx, __func__); + i5000_get_error_info(mci, &info); + i5000_process_error_info(mci, &info, 1); +} + +/* + * i5000_get_devices Find and perform 'get' operation on the MCH's + * device/functions we want to reference for this driver + * + * Need to 'get' device 16 func 1 and func 2 + */ +static int i5000_get_devices(struct mem_ctl_info *mci, int dev_idx) +{ + //const struct i5000_dev_info *i5000_dev = &i5000_devs[dev_idx]; + struct i5000_pvt *pvt; + struct pci_dev *pdev; + + pvt = mci->pvt_info; + + /* Attempt to 'get' the MCH register we want */ + pdev = NULL; + while (1) { + pdev = pci_get_device(PCI_VENDOR_ID_INTEL, + PCI_DEVICE_ID_INTEL_I5000_DEV16, pdev); + + /* End of list, leave */ + if (pdev == NULL) { + i5000_printk(KERN_ERR, + "'system address,Process Bus' " + "device not found:" + "vendor 0x%x device 0x%x FUNC 1 " + "(broken BIOS?)\n", + PCI_VENDOR_ID_INTEL, + PCI_DEVICE_ID_INTEL_I5000_DEV16); + + return 1; + } + + /* Scan for device 16 func 1 */ + if (PCI_FUNC(pdev->devfn) == 1) + break; + } + + pvt->branchmap_werrors = pdev; + + /* Attempt to 'get' the MCH register we want */ + pdev = NULL; + while (1) { + pdev = pci_get_device(PCI_VENDOR_ID_INTEL, + PCI_DEVICE_ID_INTEL_I5000_DEV16, pdev); + + if (pdev == NULL) { + i5000_printk(KERN_ERR, + "MC: 'branchmap,control,errors' " + "device not found:" + "vendor 0x%x device 0x%x Func 2 " + "(broken BIOS?)\n", + PCI_VENDOR_ID_INTEL, + PCI_DEVICE_ID_INTEL_I5000_DEV16); + + pci_dev_put(pvt->branchmap_werrors); + return 1; + } + + /* Scan for device 16 func 1 */ + if (PCI_FUNC(pdev->devfn) == 2) + break; + } + + pvt->fsb_error_regs = pdev; + + debugf1("System Address, processor bus- PCI Bus ID: %s %x:%x\n", + pci_name(pvt->system_address), + pvt->system_address->vendor, pvt->system_address->device); + debugf1("Branchmap, control and errors - PCI Bus ID: %s %x:%x\n", + pci_name(pvt->branchmap_werrors), + pvt->branchmap_werrors->vendor, pvt->branchmap_werrors->device); + debugf1("FSB Error Regs - PCI Bus ID: %s %x:%x\n", + pci_name(pvt->fsb_error_regs), + pvt->fsb_error_regs->vendor, pvt->fsb_error_regs->device); + + pdev = NULL; + pdev = pci_get_device(PCI_VENDOR_ID_INTEL, + PCI_DEVICE_ID_I5000_BRANCH_0, pdev); + + if (pdev == NULL) { + i5000_printk(KERN_ERR, + "MC: 'BRANCH 0' device not found:" + "vendor 0x%x device 0x%x Func 0 (broken BIOS?)\n", + PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_I5000_BRANCH_0); + + pci_dev_put(pvt->branchmap_werrors); + pci_dev_put(pvt->fsb_error_regs); + return 1; + } + + pvt->branch_0 = pdev; + + /* If this device claims to have more than 2 channels then + * fetch Branch 1's information + */ + if (pvt->maxch >= CHANNELS_PER_BRANCH) { + pdev = NULL; + pdev = pci_get_device(PCI_VENDOR_ID_INTEL, + PCI_DEVICE_ID_I5000_BRANCH_1, pdev); + + if (pdev == NULL) { + i5000_printk(KERN_ERR, + "MC: 'BRANCH 1' device not found:" + "vendor 0x%x device 0x%x Func 0 " + "(broken BIOS?)\n", + PCI_VENDOR_ID_INTEL, + PCI_DEVICE_ID_I5000_BRANCH_1); + + pci_dev_put(pvt->branchmap_werrors); + pci_dev_put(pvt->fsb_error_regs); + pci_dev_put(pvt->branch_0); + return 1; + } + + pvt->branch_1 = pdev; + } + + return 0; +} + +/* + * i5000_put_devices 'put' all the devices that we have + * reserved via 'get' + */ +static void i5000_put_devices(struct mem_ctl_info *mci) +{ + struct i5000_pvt *pvt; + + pvt = mci->pvt_info; + + pci_dev_put(pvt->branchmap_werrors); /* FUNC 1 */ + pci_dev_put(pvt->fsb_error_regs); /* FUNC 2 */ + pci_dev_put(pvt->branch_0); /* DEV 21 */ + + /* Only if more than 2 channels do we release the second branch */ + if (pvt->maxch >= CHANNELS_PER_BRANCH) + pci_dev_put(pvt->branch_1); /* DEV 22 */ +} + +/* + * determine_amb_resent + * + * the information is contained in NUM_MTRS different registers + * determineing which of the NUM_MTRS requires knowing + * which channel is in question + * + * 2 branches, each with 2 channels + * b0_ambpresent0 for channel '0' + * b0_ambpresent1 for channel '1' + * b1_ambpresent0 for channel '2' + * b1_ambpresent1 for channel '3' + */ +static int determine_amb_present_reg(struct i5000_pvt *pvt, int channel) +{ + int amb_present; + + if (channel < CHANNELS_PER_BRANCH) { + if (channel & 0x1) + amb_present = pvt->b0_ambpresent1; + else + amb_present = pvt->b0_ambpresent0; + } else { + if (channel & 0x1) + amb_present = pvt->b1_ambpresent1; + else + amb_present = pvt->b1_ambpresent0; + } + + return amb_present; +} + +/* + * determine_mtr(pvt, csrow, channel) + * + * return the proper MTR register as determine by the csrow and channel desired + */ +static int determine_mtr(struct i5000_pvt *pvt, int csrow, int channel) +{ + int mtr; + + if (channel < CHANNELS_PER_BRANCH) + mtr = pvt->b0_mtr[csrow >> 1]; + else + mtr = pvt->b1_mtr[csrow >> 1]; + + return mtr; +} + +/* + */ +static void decode_mtr(int slot_row, u16 mtr) +{ + int ans; + + ans = MTR_DIMMS_PRESENT(mtr); + + debugf2("\tMTR%d=0x%x: DIMMs are %s\n", slot_row, mtr, + ans ? "Present" : "NOT Present"); + if (!ans) + return; + + debugf2("\t\tWIDTH: x%d\n", MTR_DRAM_WIDTH(mtr)); + debugf2("\t\tNUMBANK: %d bank(s)\n", MTR_DRAM_BANKS(mtr)); + debugf2("\t\tNUMRANK: %s\n", MTR_DIMM_RANK(mtr) ? "double" : "single"); + debugf2("\t\tNUMROW: %s\n", numrow_toString[MTR_DIMM_ROWS(mtr)]); + debugf2("\t\tNUMCOL: %s\n", numcol_toString[MTR_DIMM_COLS(mtr)]); +} + +static void handle_channel(struct i5000_pvt *pvt, int csrow, int channel, + struct i5000_dimm_info *dinfo) +{ + int mtr; + int amb_present_reg; + int addrBits; + + mtr = determine_mtr(pvt, csrow, channel); + if (MTR_DIMMS_PRESENT(mtr)) { + amb_present_reg = determine_amb_present_reg(pvt, channel); + + /* Determine if there is a DIMM present in this DIMM slot */ + if (amb_present_reg & (1 << (csrow >> 1))) { + dinfo->dual_rank = MTR_DIMM_RANK(mtr); + + if (!((dinfo->dual_rank == 0) && + ((csrow & 0x1) == 0x1))) { + /* Start with the number of bits for a Bank + * on the DRAM */ + addrBits = MTR_DRAM_BANKS_ADDR_BITS(mtr); + /* Add thenumber of ROW bits */ + addrBits += MTR_DIMM_ROWS_ADDR_BITS(mtr); + /* add the number of COLUMN bits */ + addrBits += MTR_DIMM_COLS_ADDR_BITS(mtr); + + addrBits += 6; /* add 64 bits per DIMM */ + addrBits -= 20; /* divide by 2^^20 */ + addrBits -= 3; /* 8 bits per bytes */ + + dinfo->megabytes = 1 << addrBits; + } + } + } +} + +/* + * calculate_dimm_size + * + * also will output a DIMM matrix map, if debug is enabled, for viewing + * how the DIMMs are populated + */ +static void calculate_dimm_size(struct i5000_pvt *pvt) +{ + struct i5000_dimm_info *dinfo; + int csrow, max_csrows; + char *p, *mem_buffer; + int space, n; + int channel; + + /* ================= Generate some debug output ================= */ + space = PAGE_SIZE; + mem_buffer = p = kmalloc(space, GFP_KERNEL); + if (p == NULL) { + i5000_printk(KERN_ERR, "MC: %s:%s() kmalloc() failed\n", + __FILE__, __func__); + return; + } + + n = snprintf(p, space, "\n"); + p += n; + space -= n; + + /* Scan all the actual CSROWS (which is # of DIMMS * 2) + * and calculate the information for each DIMM + * Start with the highest csrow first, to display it first + * and work toward the 0th csrow + */ + max_csrows = pvt->maxdimmperch * 2; + for (csrow = max_csrows - 1; csrow >= 0; csrow--) { + + /* on an odd csrow, first output a 'boundary' marker, + * then reset the message buffer */ + if (csrow & 0x1) { + n = snprintf(p, space, "---------------------------" + "--------------------------------"); + p += n; + space -= n; + debugf2("%s\n", mem_buffer); + p = mem_buffer; + space = PAGE_SIZE; + } + n = snprintf(p, space, "csrow %2d ", csrow); + p += n; + space -= n; + + for (channel = 0; channel < pvt->maxch; channel++) { + dinfo = &pvt->dimm_info[csrow][channel]; + handle_channel(pvt, csrow, channel, dinfo); + n = snprintf(p, space, "%4d MB | ", dinfo->megabytes); + p += n; + space -= n; + } + n = snprintf(p, space, "\n"); + p += n; + space -= n; + } + + /* Output the last bottom 'boundary' marker */ + n = snprintf(p, space, "---------------------------" + "--------------------------------\n"); + p += n; + space -= n; + + /* now output the 'channel' labels */ + n = snprintf(p, space, " "); + p += n; + space -= n; + for (channel = 0; channel < pvt->maxch; channel++) { + n = snprintf(p, space, "channel %d | ", channel); + p += n; + space -= n; + } + n = snprintf(p, space, "\n"); + p += n; + space -= n; + + /* output the last message and free buffer */ + debugf2("%s\n", mem_buffer); + kfree(mem_buffer); +} + +/* + * i5000_get_mc_regs read in the necessary registers and + * cache locally + * + * Fills in the private data members + */ +static void i5000_get_mc_regs(struct mem_ctl_info *mci) +{ + struct i5000_pvt *pvt; + u32 actual_tolm; + u16 limit; + int slot_row; + int maxch; + int maxdimmperch; + int way0, way1; + + pvt = mci->pvt_info; + + pci_read_config_dword(pvt->system_address, AMBASE, + (u32 *) & pvt->ambase); + pci_read_config_dword(pvt->system_address, AMBASE + sizeof(u32), + ((u32 *) & pvt->ambase) + sizeof(u32)); + + maxdimmperch = pvt->maxdimmperch; + maxch = pvt->maxch; + + debugf2("AMBASE= 0x%lx MAXCH= %d MAX-DIMM-Per-CH= %d\n", + (long unsigned int)pvt->ambase, pvt->maxch, pvt->maxdimmperch); + + /* Get the Branch Map regs */ + pci_read_config_word(pvt->branchmap_werrors, TOLM, &pvt->tolm); + pvt->tolm >>= 12; + debugf2("\nTOLM (number of 256M regions) =%u (0x%x)\n", pvt->tolm, + pvt->tolm); + + actual_tolm = pvt->tolm << 28; + debugf2("Actual TOLM byte addr=%u (0x%x)\n", actual_tolm, actual_tolm); + + pci_read_config_word(pvt->branchmap_werrors, MIR0, &pvt->mir0); + pci_read_config_word(pvt->branchmap_werrors, MIR1, &pvt->mir1); + pci_read_config_word(pvt->branchmap_werrors, MIR2, &pvt->mir2); + + /* Get the MIR[0-2] regs */ + limit = (pvt->mir0 >> 4) & 0x0FFF; + way0 = pvt->mir0 & 0x1; + way1 = pvt->mir0 & 0x2; + debugf2("MIR0: limit= 0x%x WAY1= %u WAY0= %x\n", limit, way1, way0); + limit = (pvt->mir1 >> 4) & 0x0FFF; + way0 = pvt->mir1 & 0x1; + way1 = pvt->mir1 & 0x2; + debugf2("MIR1: limit= 0x%x WAY1= %u WAY0= %x\n", limit, way1, way0); + limit = (pvt->mir2 >> 4) & 0x0FFF; + way0 = pvt->mir2 & 0x1; + way1 = pvt->mir2 & 0x2; + debugf2("MIR2: limit= 0x%x WAY1= %u WAY0= %x\n", limit, way1, way0); + + /* Get the MTR[0-3] regs */ + for (slot_row = 0; slot_row < NUM_MTRS; slot_row++) { + int where = MTR0 + (slot_row * sizeof(u32)); + + pci_read_config_word(pvt->branch_0, where, + &pvt->b0_mtr[slot_row]); + + debugf2("MTR%d where=0x%x B0 value=0x%x\n", slot_row, where, + pvt->b0_mtr[slot_row]); + + if (pvt->maxch >= CHANNELS_PER_BRANCH) { + pci_read_config_word(pvt->branch_1, where, + &pvt->b1_mtr[slot_row]); + debugf2("MTR%d where=0x%x B1 value=0x%x\n", slot_row, + where, pvt->b0_mtr[slot_row]); + } else { + pvt->b1_mtr[slot_row] = 0; + } + } + + /* Read and dump branch 0's MTRs */ + debugf2("\nMemory Technology Registers:\n"); + debugf2(" Branch 0:\n"); + for (slot_row = 0; slot_row < NUM_MTRS; slot_row++) { + decode_mtr(slot_row, pvt->b0_mtr[slot_row]); + } + pci_read_config_word(pvt->branch_0, AMB_PRESENT_0, + &pvt->b0_ambpresent0); + debugf2("\t\tAMB-Branch 0-present0 0x%x:\n", pvt->b0_ambpresent0); + pci_read_config_word(pvt->branch_0, AMB_PRESENT_1, + &pvt->b0_ambpresent1); + debugf2("\t\tAMB-Branch 0-present1 0x%x:\n", pvt->b0_ambpresent1); + + /* Only if we have 2 branchs (4 channels) */ + if (pvt->maxch < CHANNELS_PER_BRANCH) { + pvt->b1_ambpresent0 = 0; + pvt->b1_ambpresent1 = 0; + } else { + /* Read and dump branch 1's MTRs */ + debugf2(" Branch 1:\n"); + for (slot_row = 0; slot_row < NUM_MTRS; slot_row++) { + decode_mtr(slot_row, pvt->b1_mtr[slot_row]); + } + pci_read_config_word(pvt->branch_1, AMB_PRESENT_0, + &pvt->b1_ambpresent0); + debugf2("\t\tAMB-Branch 1-present0 0x%x:\n", + pvt->b1_ambpresent0); + pci_read_config_word(pvt->branch_1, AMB_PRESENT_1, + &pvt->b1_ambpresent1); + debugf2("\t\tAMB-Branch 1-present1 0x%x:\n", + pvt->b1_ambpresent1); + } + + /* Go and determine the size of each DIMM and place in an + * orderly matrix */ + calculate_dimm_size(pvt); +} + +/* + * i5000_init_csrows Initialize the 'csrows' table within + * the mci control structure with the + * addressing of memory. + * + * return: + * 0 success + * 1 no actual memory found on this MC + */ +static int i5000_init_csrows(struct mem_ctl_info *mci) +{ + struct i5000_pvt *pvt; + struct csrow_info *p_csrow; + int empty, channel_count; + int max_csrows; + int mtr; + int csrow_megs; + int channel; + int csrow; + + pvt = mci->pvt_info; + + channel_count = pvt->maxch; + max_csrows = pvt->maxdimmperch * 2; + + empty = 1; /* Assume NO memory */ + + for (csrow = 0; csrow < max_csrows; csrow++) { + p_csrow = &mci->csrows[csrow]; + + p_csrow->csrow_idx = csrow; + + /* use branch 0 for the basis */ + mtr = pvt->b0_mtr[csrow >> 1]; + + /* if no DIMMS on this row, continue */ + if (!MTR_DIMMS_PRESENT(mtr)) + continue; + + /* FAKE OUT VALUES, FIXME */ + p_csrow->first_page = 0 + csrow * 20; + p_csrow->last_page = 9 + csrow * 20; + p_csrow->page_mask = 0xFFF; + + p_csrow->grain = 8; + + csrow_megs = 0; + for (channel = 0; channel < pvt->maxch; channel++) { + csrow_megs += pvt->dimm_info[csrow][channel].megabytes; + } + + p_csrow->nr_pages = csrow_megs << 8; + + /* Assume DDR2 for now */ + p_csrow->mtype = MEM_FB_DDR2; + + /* ask what device type on this row */ + if (MTR_DRAM_WIDTH(mtr)) + p_csrow->dtype = DEV_X8; + else + p_csrow->dtype = DEV_X4; + + p_csrow->edac_mode = EDAC_S8ECD8ED; + + empty = 0; + } + + return empty; +} + +/* + * i5000_enable_error_reporting + * Turn on the memory reporting features of the hardware + */ +static void i5000_enable_error_reporting(struct mem_ctl_info *mci) +{ + struct i5000_pvt *pvt; + u32 fbd_error_mask; + + pvt = mci->pvt_info; + + /* Read the FBD Error Mask Register */ + pci_read_config_dword(pvt->branchmap_werrors, EMASK_FBD, + &fbd_error_mask); + + /* Enable with a '0' */ + fbd_error_mask &= ~(ENABLE_EMASK_ALL); + + pci_write_config_dword(pvt->branchmap_werrors, EMASK_FBD, + fbd_error_mask); +} + +/* + * i5000_get_dimm_and_channel_counts(pdev, &num_csrows, &num_channels) + * + * ask the device how many channels are present and how many CSROWS + * as well + */ +static void i5000_get_dimm_and_channel_counts(struct pci_dev *pdev, + int *num_dimms_per_channel, + int *num_channels) +{ + u8 value; + + /* Need to retrieve just how many channels and dimms per channel are + * supported on this memory controller + */ + pci_read_config_byte(pdev, MAXDIMMPERCH, &value); + *num_dimms_per_channel = (int)value *2; + + pci_read_config_byte(pdev, MAXCH, &value); + *num_channels = (int)value; +} + +/* + * i5000_probe1 Probe for ONE instance of device to see if it is + * present. + * return: + * 0 for FOUND a device + * < 0 for error code + */ +static int i5000_probe1(struct pci_dev *pdev, int dev_idx) +{ + struct mem_ctl_info *mci; + struct i5000_pvt *pvt; + int num_channels; + int num_dimms_per_channel; + int num_csrows; + + debugf0("MC: " __FILE__ ": %s(), pdev bus %u dev=0x%x fn=0x%x\n", + __func__, + pdev->bus->number, + PCI_SLOT(pdev->devfn), PCI_FUNC(pdev->devfn)); + + /* We only are looking for func 0 of the set */ + if (PCI_FUNC(pdev->devfn) != 0) + return -ENODEV; + + /* make sure error reporting method is sane */ + switch (edac_op_state) { + case EDAC_OPSTATE_POLL: + case EDAC_OPSTATE_NMI: + break; + default: + edac_op_state = EDAC_OPSTATE_POLL; + break; + } + + /* Ask the devices for the number of CSROWS and CHANNELS so + * that we can calculate the memory resources, etc + * + * The Chipset will report what it can handle which will be greater + * or equal to what the motherboard manufacturer will implement. + * + * As we don't have a motherboard identification routine to determine + * actual number of slots/dimms per channel, we thus utilize the + * resource as specified by the chipset. Thus, we might have + * have more DIMMs per channel than actually on the mobo, but this + * allows the driver to support upto the chipset max, without + * some fancy mobo determination. + */ + i5000_get_dimm_and_channel_counts(pdev, &num_dimms_per_channel, + &num_channels); + num_csrows = num_dimms_per_channel * 2; + + debugf0("MC: %s(): Number of - Channels= %d DIMMS= %d CSROWS= %d\n", + __func__, num_channels, num_dimms_per_channel, num_csrows); + + /* allocate a new MC control structure */ + mci = edac_mc_alloc(sizeof(*pvt), num_csrows, num_channels, 0); + + if (mci == NULL) + return -ENOMEM; + + debugf0("MC: " __FILE__ ": %s(): mci = %p\n", __func__, mci); + + mci->dev = &pdev->dev; /* record ptr to the generic device */ + + pvt = mci->pvt_info; + pvt->system_address = pdev; /* Record this device in our private */ + pvt->maxch = num_channels; + pvt->maxdimmperch = num_dimms_per_channel; + + /* 'get' the pci devices we want to reserve for our use */ + if (i5000_get_devices(mci, dev_idx)) + goto fail0; + + /* Time to get serious */ + i5000_get_mc_regs(mci); /* retrieve the hardware registers */ + + mci->mc_idx = 0; + mci->mtype_cap = MEM_FLAG_FB_DDR2; + mci->edac_ctl_cap = EDAC_FLAG_NONE; + mci->edac_cap = EDAC_FLAG_NONE; + mci->mod_name = "i5000_edac.c"; + mci->mod_ver = I5000_REVISION; + mci->ctl_name = i5000_devs[dev_idx].ctl_name; + mci->dev_name = pci_name(pdev); + mci->ctl_page_to_phys = NULL; + + /* Set the function pointer to an actual operation function */ + mci->edac_check = i5000_check_error; + + /* initialize the MC control structure 'csrows' table + * with the mapping and control information */ + if (i5000_init_csrows(mci)) { + debugf0("MC: Setting mci->edac_cap to EDAC_FLAG_NONE\n" + " because i5000_init_csrows() returned nonzero " + "value\n"); + mci->edac_cap = EDAC_FLAG_NONE; /* no csrows found */ + } else { + debugf1("MC: Enable error reporting now\n"); + i5000_enable_error_reporting(mci); + } + + /* add this new MC control structure to EDAC's list of MCs */ + if (edac_mc_add_mc(mci)) { + debugf0("MC: " __FILE__ + ": %s(): failed edac_mc_add_mc()\n", __func__); + /* FIXME: perhaps some code should go here that disables error + * reporting if we just enabled it + */ + goto fail1; + } + + i5000_clear_error(mci); + + /* allocating generic PCI control info */ + i5000_pci = edac_pci_create_generic_ctl(&pdev->dev, EDAC_MOD_STR); + if (!i5000_pci) { + printk(KERN_WARNING + "%s(): Unable to create PCI control\n", + __func__); + printk(KERN_WARNING + "%s(): PCI error report via EDAC not setup\n", + __func__); + } + + return 0; + + /* Error exit unwinding stack */ +fail1: + + i5000_put_devices(mci); + +fail0: + edac_mc_free(mci); + return -ENODEV; +} + +/* + * i5000_init_one constructor for one instance of device + * + * returns: + * negative on error + * count (>= 0) + */ +static int __devinit i5000_init_one(struct pci_dev *pdev, + const struct pci_device_id *id) +{ + int rc; + + debugf0("MC: " __FILE__ ": %s()\n", __func__); + + /* wake up device */ + rc = pci_enable_device(pdev); + if (rc == -EIO) + return rc; + + /* now probe and enable the device */ + return i5000_probe1(pdev, id->driver_data); +} + +/* + * i5000_remove_one destructor for one instance of device + * + */ +static void __devexit i5000_remove_one(struct pci_dev *pdev) +{ + struct mem_ctl_info *mci; + + debugf0(__FILE__ ": %s()\n", __func__); + + if (i5000_pci) + edac_pci_release_generic_ctl(i5000_pci); + + if ((mci = edac_mc_del_mc(&pdev->dev)) == NULL) + return; + + /* retrieve references to resources, and free those resources */ + i5000_put_devices(mci); + + edac_mc_free(mci); +} + +/* + * pci_device_id table for which devices we are looking for + * + * The "E500P" device is the first device supported. + */ +static const struct pci_device_id i5000_pci_tbl[] __devinitdata = { + {PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_I5000_DEV16), + .driver_data = I5000P}, + + {0,} /* 0 terminated list. */ +}; + +MODULE_DEVICE_TABLE(pci, i5000_pci_tbl); + +/* + * i5000_driver pci_driver structure for this module + * + */ +static struct pci_driver i5000_driver = { + .name = __stringify(KBUILD_BASENAME), + .probe = i5000_init_one, + .remove = __devexit_p(i5000_remove_one), + .id_table = i5000_pci_tbl, +}; + +/* + * i5000_init Module entry function + * Try to initialize this module for its devices + */ +static int __init i5000_init(void) +{ + int pci_rc; + + debugf2("MC: " __FILE__ ": %s()\n", __func__); + + pci_rc = pci_register_driver(&i5000_driver); + + return (pci_rc < 0) ? pci_rc : 0; +} + +/* + * i5000_exit() Module exit function + * Unregister the driver + */ +static void __exit i5000_exit(void) +{ + debugf2("MC: " __FILE__ ": %s()\n", __func__); + pci_unregister_driver(&i5000_driver); +} + +module_init(i5000_init); +module_exit(i5000_exit); + +MODULE_LICENSE("GPL"); +MODULE_AUTHOR + ("Linux Networx (http://lnxi.com) Doug Thompson <norsk5@xmission.com>"); +MODULE_DESCRIPTION("MC Driver for Intel I5000 memory controllers - " + I5000_REVISION); +module_param(edac_op_state, int, 0444); +MODULE_PARM_DESC(edac_op_state, "EDAC Error Reporting state: 0=Poll,1=NMI"); diff --git a/drivers/edac/i82443bxgx_edac.c b/drivers/edac/i82443bxgx_edac.c new file mode 100644 index 00000000000..83bfe37c4bb --- /dev/null +++ b/drivers/edac/i82443bxgx_edac.c @@ -0,0 +1,402 @@ +/* + * Intel 82443BX/GX (440BX/GX chipset) Memory Controller EDAC kernel + * module (C) 2006 Tim Small + * + * This file may be distributed under the terms of the GNU General + * Public License. + * + * Written by Tim Small <tim@buttersideup.com>, based on work by Linux + * Networx, Thayne Harbaugh, Dan Hollis <goemon at anime dot net> and + * others. + * + * 440GX fix by Jason Uhlenkott <juhlenko@akamai.com>. + * + * Written with reference to 82443BX Host Bridge Datasheet: + * http://www.intel.com/design/chipsets/440/documentation.htm + * references to this document given in []. + * + * This module doesn't support the 440LX, but it may be possible to + * make it do so (the 440LX's register definitions are different, but + * not completely so - I haven't studied them in enough detail to know + * how easy this would be). + */ + +#include <linux/module.h> +#include <linux/init.h> + +#include <linux/pci.h> +#include <linux/pci_ids.h> + +#include <linux/slab.h> + +#include "edac_core.h" + +#define I82443_REVISION "0.1" + +#define EDAC_MOD_STR "i82443bxgx_edac" + +/* The 82443BX supports SDRAM, or EDO (EDO for mobile only), "Memory + * Size: 8 MB to 512 MB (1GB with Registered DIMMs) with eight memory + * rows" "The 82443BX supports multiple-bit error detection and + * single-bit error correction when ECC mode is enabled and + * single/multi-bit error detection when correction is disabled. + * During writes to the DRAM, the 82443BX generates ECC for the data + * on a QWord basis. Partial QWord writes require a read-modify-write + * cycle when ECC is enabled." +*/ + +/* "Additionally, the 82443BX ensures that the data is corrected in + * main memory so that accumulation of errors is prevented. Another + * error within the same QWord would result in a double-bit error + * which is unrecoverable. This is known as hardware scrubbing since + * it requires no software intervention to correct the data in memory." + */ + +/* [Also see page 100 (section 4.3), "DRAM Interface"] + * [Also see page 112 (section 4.6.1.4), ECC] + */ + +#define I82443BXGX_NR_CSROWS 8 +#define I82443BXGX_NR_CHANS 1 +#define I82443BXGX_NR_DIMMS 4 + +/* 82443 PCI Device 0 */ +#define I82443BXGX_NBXCFG 0x50 /* 32bit register starting at this PCI + * config space offset */ +#define I82443BXGX_NBXCFG_OFFSET_NON_ECCROW 24 /* Array of bits, zero if + * row is non-ECC */ +#define I82443BXGX_NBXCFG_OFFSET_DRAM_FREQ 12 /* 2 bits,00=100MHz,10=66 MHz */ + +#define I82443BXGX_NBXCFG_OFFSET_DRAM_INTEGRITY 7 /* 2 bits: */ +#define I82443BXGX_NBXCFG_INTEGRITY_NONE 0x0 /* 00 = Non-ECC */ +#define I82443BXGX_NBXCFG_INTEGRITY_EC 0x1 /* 01 = EC (only) */ +#define I82443BXGX_NBXCFG_INTEGRITY_ECC 0x2 /* 10 = ECC */ +#define I82443BXGX_NBXCFG_INTEGRITY_SCRUB 0x3 /* 11 = ECC + HW Scrub */ + +#define I82443BXGX_NBXCFG_OFFSET_ECC_DIAG_ENABLE 6 + +/* 82443 PCI Device 0 */ +#define I82443BXGX_EAP 0x80 /* 32bit register starting at this PCI + * config space offset, Error Address + * Pointer Register */ +#define I82443BXGX_EAP_OFFSET_EAP 12 /* High 20 bits of error address */ +#define I82443BXGX_EAP_OFFSET_MBE BIT(1) /* Err at EAP was multi-bit (W1TC) */ +#define I82443BXGX_EAP_OFFSET_SBE BIT(0) /* Err at EAP was single-bit (W1TC) */ + +#define I82443BXGX_ERRCMD 0x90 /* 8bit register starting at this PCI + * config space offset. */ +#define I82443BXGX_ERRCMD_OFFSET_SERR_ON_MBE BIT(1) /* 1 = enable */ +#define I82443BXGX_ERRCMD_OFFSET_SERR_ON_SBE BIT(0) /* 1 = enable */ + +#define I82443BXGX_ERRSTS 0x91 /* 16bit register starting at this PCI + * config space offset. */ +#define I82443BXGX_ERRSTS_OFFSET_MBFRE 5 /* 3 bits - first err row multibit */ +#define I82443BXGX_ERRSTS_OFFSET_MEF BIT(4) /* 1 = MBE occurred */ +#define I82443BXGX_ERRSTS_OFFSET_SBFRE 1 /* 3 bits - first err row singlebit */ +#define I82443BXGX_ERRSTS_OFFSET_SEF BIT(0) /* 1 = SBE occurred */ + +#define I82443BXGX_DRAMC 0x57 /* 8bit register starting at this PCI + * config space offset. */ +#define I82443BXGX_DRAMC_OFFSET_DT 3 /* 2 bits, DRAM Type */ +#define I82443BXGX_DRAMC_DRAM_IS_EDO 0 /* 00 = EDO */ +#define I82443BXGX_DRAMC_DRAM_IS_SDRAM 1 /* 01 = SDRAM */ +#define I82443BXGX_DRAMC_DRAM_IS_RSDRAM 2 /* 10 = Registered SDRAM */ + +#define I82443BXGX_DRB 0x60 /* 8x 8bit registers starting at this PCI + * config space offset. */ + +/* FIXME - don't poll when ECC disabled? */ + +struct i82443bxgx_edacmc_error_info { + u32 eap; +}; + +static struct edac_pci_ctl_info *i82443bxgx_pci; + +static void i82443bxgx_edacmc_get_error_info(struct mem_ctl_info *mci, + struct i82443bxgx_edacmc_error_info + *info) +{ + struct pci_dev *pdev; + pdev = to_pci_dev(mci->dev); + pci_read_config_dword(pdev, I82443BXGX_EAP, &info->eap); + if (info->eap & I82443BXGX_EAP_OFFSET_SBE) + /* Clear error to allow next error to be reported [p.61] */ + pci_write_bits32(pdev, I82443BXGX_EAP, + I82443BXGX_EAP_OFFSET_SBE, + I82443BXGX_EAP_OFFSET_SBE); + + if (info->eap & I82443BXGX_EAP_OFFSET_MBE) + /* Clear error to allow next error to be reported [p.61] */ + pci_write_bits32(pdev, I82443BXGX_EAP, + I82443BXGX_EAP_OFFSET_MBE, + I82443BXGX_EAP_OFFSET_MBE); +} + +static int i82443bxgx_edacmc_process_error_info(struct mem_ctl_info *mci, + struct + i82443bxgx_edacmc_error_info + *info, int handle_errors) +{ + int error_found = 0; + u32 eapaddr, page, pageoffset; + + /* bits 30:12 hold the 4kb block in which the error occurred + * [p.61] */ + eapaddr = (info->eap & 0xfffff000); + page = eapaddr >> PAGE_SHIFT; + pageoffset = eapaddr - (page << PAGE_SHIFT); + + if (info->eap & I82443BXGX_EAP_OFFSET_SBE) { + error_found = 1; + if (handle_errors) + edac_mc_handle_ce(mci, page, pageoffset, + /* 440BX/GX don't make syndrome information + * available */ + 0, edac_mc_find_csrow_by_page(mci, page), 0, + mci->ctl_name); + } + + if (info->eap & I82443BXGX_EAP_OFFSET_MBE) { + error_found = 1; + if (handle_errors) + edac_mc_handle_ue(mci, page, pageoffset, + edac_mc_find_csrow_by_page(mci, page), + mci->ctl_name); + } + + return error_found; +} + +static void i82443bxgx_edacmc_check(struct mem_ctl_info *mci) +{ + struct i82443bxgx_edacmc_error_info info; + + debugf1("MC%d: " __FILE__ ": %s()\n", mci->mc_idx, __func__); + i82443bxgx_edacmc_get_error_info(mci, &info); + i82443bxgx_edacmc_process_error_info(mci, &info, 1); +} + +static void i82443bxgx_init_csrows(struct mem_ctl_info *mci, + struct pci_dev *pdev, + enum edac_type edac_mode, + enum mem_type mtype) +{ + struct csrow_info *csrow; + int index; + u8 drbar, dramc; + u32 row_base, row_high_limit, row_high_limit_last; + + pci_read_config_byte(pdev, I82443BXGX_DRAMC, &dramc); + row_high_limit_last = 0; + for (index = 0; index < mci->nr_csrows; index++) { + csrow = &mci->csrows[index]; + pci_read_config_byte(pdev, I82443BXGX_DRB + index, &drbar); + debugf1("MC%d: " __FILE__ ": %s() Row=%d DRB = %#0x\n", + mci->mc_idx, __func__, index, drbar); + row_high_limit = ((u32) drbar << 23); + /* find the DRAM Chip Select Base address and mask */ + debugf1("MC%d: " __FILE__ ": %s() Row=%d, " + "Boundry Address=%#0x, Last = %#0x \n", + mci->mc_idx, __func__, index, row_high_limit, + row_high_limit_last); + + /* 440GX goes to 2GB, represented with a DRB of 0. */ + if (row_high_limit_last && !row_high_limit) + row_high_limit = 1UL << 31; + + /* This row is empty [p.49] */ + if (row_high_limit == row_high_limit_last) + continue; + row_base = row_high_limit_last; + csrow->first_page = row_base >> PAGE_SHIFT; + csrow->last_page = (row_high_limit >> PAGE_SHIFT) - 1; + csrow->nr_pages = csrow->last_page - csrow->first_page + 1; + /* EAP reports in 4kilobyte granularity [61] */ + csrow->grain = 1 << 12; + csrow->mtype = mtype; + /* I don't think 440BX can tell you device type? FIXME? */ + csrow->dtype = DEV_UNKNOWN; + /* Mode is global to all rows on 440BX */ + csrow->edac_mode = edac_mode; + row_high_limit_last = row_high_limit; + } +} + +static int i82443bxgx_edacmc_probe1(struct pci_dev *pdev, int dev_idx) +{ + struct mem_ctl_info *mci; + u8 dramc; + u32 nbxcfg, ecc_mode; + enum mem_type mtype; + enum edac_type edac_mode; + + debugf0("MC: " __FILE__ ": %s()\n", __func__); + + /* Something is really hosed if PCI config space reads from + * the MC aren't working. + */ + if (pci_read_config_dword(pdev, I82443BXGX_NBXCFG, &nbxcfg)) + return -EIO; + + mci = edac_mc_alloc(0, I82443BXGX_NR_CSROWS, I82443BXGX_NR_CHANS, 0); + + if (mci == NULL) + return -ENOMEM; + + debugf0("MC: " __FILE__ ": %s(): mci = %p\n", __func__, mci); + mci->dev = &pdev->dev; + mci->mtype_cap = MEM_FLAG_EDO | MEM_FLAG_SDR | MEM_FLAG_RDR; + mci->edac_ctl_cap = EDAC_FLAG_NONE | EDAC_FLAG_EC | EDAC_FLAG_SECDED; + pci_read_config_byte(pdev, I82443BXGX_DRAMC, &dramc); + switch ((dramc >> I82443BXGX_DRAMC_OFFSET_DT) & (BIT(0) | BIT(1))) { + case I82443BXGX_DRAMC_DRAM_IS_EDO: + mtype = MEM_EDO; + break; + case I82443BXGX_DRAMC_DRAM_IS_SDRAM: + mtype = MEM_SDR; + break; + case I82443BXGX_DRAMC_DRAM_IS_RSDRAM: + mtype = MEM_RDR; + break; + default: + debugf0("Unknown/reserved DRAM type value " + "in DRAMC register!\n"); + mtype = -MEM_UNKNOWN; + } + + if ((mtype == MEM_SDR) || (mtype == MEM_RDR)) + mci->edac_cap = mci->edac_ctl_cap; + else + mci->edac_cap = EDAC_FLAG_NONE; + + mci->scrub_cap = SCRUB_FLAG_HW_SRC; + pci_read_config_dword(pdev, I82443BXGX_NBXCFG, &nbxcfg); + ecc_mode = ((nbxcfg >> I82443BXGX_NBXCFG_OFFSET_DRAM_INTEGRITY) & + (BIT(0) | BIT(1))); + + mci->scrub_mode = (ecc_mode == I82443BXGX_NBXCFG_INTEGRITY_SCRUB) + ? SCRUB_HW_SRC : SCRUB_NONE; + + switch (ecc_mode) { + case I82443BXGX_NBXCFG_INTEGRITY_NONE: + edac_mode = EDAC_NONE; + break; + case I82443BXGX_NBXCFG_INTEGRITY_EC: + edac_mode = EDAC_EC; + break; + case I82443BXGX_NBXCFG_INTEGRITY_ECC: + case I82443BXGX_NBXCFG_INTEGRITY_SCRUB: + edac_mode = EDAC_SECDED; + break; + default: + debugf0("%s(): Unknown/reserved ECC state " + "in NBXCFG register!\n", __func__); + edac_mode = EDAC_UNKNOWN; + break; + } + + i82443bxgx_init_csrows(mci, pdev, edac_mode, mtype); + + /* Many BIOSes don't clear error flags on boot, so do this + * here, or we get "phantom" errors occuring at module-load + * time. */ + pci_write_bits32(pdev, I82443BXGX_EAP, + (I82443BXGX_EAP_OFFSET_SBE | + I82443BXGX_EAP_OFFSET_MBE), + (I82443BXGX_EAP_OFFSET_SBE | + I82443BXGX_EAP_OFFSET_MBE)); + + mci->mod_name = EDAC_MOD_STR; + mci->mod_ver = I82443_REVISION; + mci->ctl_name = "I82443BXGX"; + mci->dev_name = pci_name(pdev); + mci->edac_check = i82443bxgx_edacmc_check; + mci->ctl_page_to_phys = NULL; + + if (edac_mc_add_mc(mci)) { + debugf3("%s(): failed edac_mc_add_mc()\n", __func__); + goto fail; + } + + /* allocating generic PCI control info */ + i82443bxgx_pci = edac_pci_create_generic_ctl(&pdev->dev, EDAC_MOD_STR); + if (!i82443bxgx_pci) { + printk(KERN_WARNING + "%s(): Unable to create PCI control\n", + __func__); + printk(KERN_WARNING + "%s(): PCI error report via EDAC not setup\n", + __func__); + } + + debugf3("MC: " __FILE__ ": %s(): success\n", __func__); + return 0; + +fail: + edac_mc_free(mci); + return -ENODEV; +} + +EXPORT_SYMBOL_GPL(i82443bxgx_edacmc_probe1); + +/* returns count (>= 0), or negative on error */ +static int __devinit i82443bxgx_edacmc_init_one(struct pci_dev *pdev, + const struct pci_device_id *ent) +{ + debugf0("MC: " __FILE__ ": %s()\n", __func__); + + /* don't need to call pci_device_enable() */ + return i82443bxgx_edacmc_probe1(pdev, ent->driver_data); +} + +static void __devexit i82443bxgx_edacmc_remove_one(struct pci_dev *pdev) +{ + struct mem_ctl_info *mci; + + debugf0(__FILE__ ": %s()\n", __func__); + + if (i82443bxgx_pci) + edac_pci_release_generic_ctl(i82443bxgx_pci); + + if ((mci = edac_mc_del_mc(&pdev->dev)) == NULL) + return; + + edac_mc_free(mci); +} + +EXPORT_SYMBOL_GPL(i82443bxgx_edacmc_remove_one); + +static const struct pci_device_id i82443bxgx_pci_tbl[] __devinitdata = { + {PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_82443BX_0)}, + {PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_82443BX_2)}, + {PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_82443GX_0)}, + {PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_82443GX_2)}, + {0,} /* 0 terminated list. */ +}; + +MODULE_DEVICE_TABLE(pci, i82443bxgx_pci_tbl); + +static struct pci_driver i82443bxgx_edacmc_driver = { + .name = EDAC_MOD_STR, + .probe = i82443bxgx_edacmc_init_one, + .remove = __devexit_p(i82443bxgx_edacmc_remove_one), + .id_table = i82443bxgx_pci_tbl, +}; + +static int __init i82443bxgx_edacmc_init(void) +{ + return pci_register_driver(&i82443bxgx_edacmc_driver); +} + +static void __exit i82443bxgx_edacmc_exit(void) +{ + pci_unregister_driver(&i82443bxgx_edacmc_driver); +} + +module_init(i82443bxgx_edacmc_init); +module_exit(i82443bxgx_edacmc_exit); + +MODULE_LICENSE("GPL"); +MODULE_AUTHOR("Tim Small <tim@buttersideup.com> - WPAD"); +MODULE_DESCRIPTION("EDAC MC support for Intel 82443BX/GX memory controllers"); diff --git a/drivers/edac/i82860_edac.c b/drivers/edac/i82860_edac.c index e4bb298e613..f5ecd2c4d81 100644 --- a/drivers/edac/i82860_edac.c +++ b/drivers/edac/i82860_edac.c @@ -14,9 +14,9 @@ #include <linux/pci.h> #include <linux/pci_ids.h> #include <linux/slab.h> -#include "edac_mc.h" +#include "edac_core.h" -#define I82860_REVISION " Ver: 2.0.1 " __DATE__ +#define I82860_REVISION " Ver: 2.0.2 " __DATE__ #define EDAC_MOD_STR "i82860_edac" #define i82860_printk(level, fmt, arg...) \ @@ -54,16 +54,16 @@ struct i82860_error_info { static const struct i82860_dev_info i82860_devs[] = { [I82860] = { - .ctl_name = "i82860" - }, + .ctl_name = "i82860"}, }; -static struct pci_dev *mci_pdev = NULL; /* init dev: in case that AGP code +static struct pci_dev *mci_pdev; /* init dev: in case that AGP code * has already registered driver */ +static struct edac_pci_ctl_info *i82860_pci; static void i82860_get_error_info(struct mem_ctl_info *mci, - struct i82860_error_info *info) + struct i82860_error_info *info) { struct pci_dev *pdev; @@ -91,13 +91,13 @@ static void i82860_get_error_info(struct mem_ctl_info *mci, if ((info->errsts ^ info->errsts2) & 0x0003) { pci_read_config_dword(pdev, I82860_EAP, &info->eap); - pci_read_config_word(pdev, I82860_DERRCTL_STS, - &info->derrsyn); + pci_read_config_word(pdev, I82860_DERRCTL_STS, &info->derrsyn); } } static int i82860_process_error_info(struct mem_ctl_info *mci, - struct i82860_error_info *info, int handle_errors) + struct i82860_error_info *info, + int handle_errors) { int row; @@ -136,7 +136,7 @@ static void i82860_check(struct mem_ctl_info *mci) static void i82860_init_csrows(struct mem_ctl_info *mci, struct pci_dev *pdev) { unsigned long last_cumul_size; - u16 mchcfg_ddim; /* DRAM Data Integrity Mode 0=none, 2=edac */ + u16 mchcfg_ddim; /* DRAM Data Integrity Mode 0=none, 2=edac */ u16 value; u32 cumul_size; struct csrow_info *csrow; @@ -155,7 +155,7 @@ static void i82860_init_csrows(struct mem_ctl_info *mci, struct pci_dev *pdev) csrow = &mci->csrows[index]; pci_read_config_word(pdev, I82860_GBA + index * 2, &value); cumul_size = (value & I82860_GBA_MASK) << - (I82860_GBA_SHIFT - PAGE_SHIFT); + (I82860_GBA_SHIFT - PAGE_SHIFT); debugf3("%s(): (%d) cumul_size 0x%x\n", __func__, index, cumul_size); @@ -186,7 +186,7 @@ static int i82860_probe1(struct pci_dev *pdev, int dev_idx) the channel and the GRA registers map to physical devices so we are going to make 1 channel for group. */ - mci = edac_mc_alloc(0, 16, 1); + mci = edac_mc_alloc(0, 16, 1, 0); if (!mci) return -ENOMEM; @@ -200,19 +200,31 @@ static int i82860_probe1(struct pci_dev *pdev, int dev_idx) mci->mod_name = EDAC_MOD_STR; mci->mod_ver = I82860_REVISION; mci->ctl_name = i82860_devs[dev_idx].ctl_name; + mci->dev_name = pci_name(pdev); mci->edac_check = i82860_check; mci->ctl_page_to_phys = NULL; i82860_init_csrows(mci, pdev); - i82860_get_error_info(mci, &discard); /* clear counters */ + i82860_get_error_info(mci, &discard); /* clear counters */ /* Here we assume that we will never see multiple instances of this * type of memory controller. The ID is therefore hardcoded to 0. */ - if (edac_mc_add_mc(mci,0)) { + if (edac_mc_add_mc(mci)) { debugf3("%s(): failed edac_mc_add_mc()\n", __func__); goto fail; } + /* allocating generic PCI control info */ + i82860_pci = edac_pci_create_generic_ctl(&pdev->dev, EDAC_MOD_STR); + if (!i82860_pci) { + printk(KERN_WARNING + "%s(): Unable to create PCI control\n", + __func__); + printk(KERN_WARNING + "%s(): PCI error report via EDAC not setup\n", + __func__); + } + /* get this far and it's successful */ debugf3("%s(): success\n", __func__); @@ -225,7 +237,7 @@ fail: /* returns count (>= 0), or negative on error */ static int __devinit i82860_init_one(struct pci_dev *pdev, - const struct pci_device_id *ent) + const struct pci_device_id *ent) { int rc; @@ -249,6 +261,9 @@ static void __devexit i82860_remove_one(struct pci_dev *pdev) debugf0("%s()\n", __func__); + if (i82860_pci) + edac_pci_release_generic_ctl(i82860_pci); + if ((mci = edac_mc_del_mc(&pdev->dev)) == NULL) return; @@ -257,12 +272,11 @@ static void __devexit i82860_remove_one(struct pci_dev *pdev) static const struct pci_device_id i82860_pci_tbl[] __devinitdata = { { - PCI_VEND_DEV(INTEL, 82860_0), PCI_ANY_ID, PCI_ANY_ID, 0, 0, - I82860 - }, + PCI_VEND_DEV(INTEL, 82860_0), PCI_ANY_ID, PCI_ANY_ID, 0, 0, + I82860}, { - 0, - } /* 0 terminated list. */ + 0, + } /* 0 terminated list. */ }; MODULE_DEVICE_TABLE(pci, i82860_pci_tbl); @@ -329,5 +343,5 @@ module_exit(i82860_exit); MODULE_LICENSE("GPL"); MODULE_AUTHOR("Red Hat Inc. (http://www.redhat.com) " - "Ben Woodard <woodard@redhat.com>"); + "Ben Woodard <woodard@redhat.com>"); MODULE_DESCRIPTION("ECC support for Intel 82860 memory hub controllers"); diff --git a/drivers/edac/i82875p_edac.c b/drivers/edac/i82875p_edac.c index 2800b3e614a..031abadc439 100644 --- a/drivers/edac/i82875p_edac.c +++ b/drivers/edac/i82875p_edac.c @@ -18,9 +18,9 @@ #include <linux/pci.h> #include <linux/pci_ids.h> #include <linux/slab.h> -#include "edac_mc.h" +#include "edac_core.h" -#define I82875P_REVISION " Ver: 2.0.1 " __DATE__ +#define I82875P_REVISION " Ver: 2.0.2 " __DATE__ #define EDAC_MOD_STR "i82875p_edac" #define i82875p_printk(level, fmt, arg...) \ @@ -174,18 +174,19 @@ struct i82875p_error_info { static const struct i82875p_dev_info i82875p_devs[] = { [I82875P] = { - .ctl_name = "i82875p" - }, + .ctl_name = "i82875p"}, }; -static struct pci_dev *mci_pdev = NULL; /* init dev: in case that AGP code has +static struct pci_dev *mci_pdev; /* init dev: in case that AGP code has * already registered driver */ static int i82875p_registered = 1; +static struct edac_pci_ctl_info *i82875p_pci; + static void i82875p_get_error_info(struct mem_ctl_info *mci, - struct i82875p_error_info *info) + struct i82875p_error_info *info) { struct pci_dev *pdev; @@ -197,38 +198,39 @@ static void i82875p_get_error_info(struct mem_ctl_info *mci, * overwritten by UE. */ pci_read_config_word(pdev, I82875P_ERRSTS, &info->errsts); + + if (!(info->errsts & 0x0081)) + return; + pci_read_config_dword(pdev, I82875P_EAP, &info->eap); pci_read_config_byte(pdev, I82875P_DES, &info->des); pci_read_config_byte(pdev, I82875P_DERRSYN, &info->derrsyn); pci_read_config_word(pdev, I82875P_ERRSTS, &info->errsts2); - pci_write_bits16(pdev, I82875P_ERRSTS, 0x0081, 0x0081); - /* * If the error is the same then we can for both reads then * the first set of reads is valid. If there is a change then * there is a CE no info and the second set of reads is valid * and should be UE info. */ - if (!(info->errsts2 & 0x0081)) - return; - if ((info->errsts ^ info->errsts2) & 0x0081) { pci_read_config_dword(pdev, I82875P_EAP, &info->eap); pci_read_config_byte(pdev, I82875P_DES, &info->des); - pci_read_config_byte(pdev, I82875P_DERRSYN, - &info->derrsyn); + pci_read_config_byte(pdev, I82875P_DERRSYN, &info->derrsyn); } + + pci_write_bits16(pdev, I82875P_ERRSTS, 0x0081, 0x0081); } static int i82875p_process_error_info(struct mem_ctl_info *mci, - struct i82875p_error_info *info, int handle_errors) + struct i82875p_error_info *info, + int handle_errors) { int row, multi_chan; multi_chan = mci->csrows[0].nr_channels - 1; - if (!(info->errsts2 & 0x0081)) + if (!(info->errsts & 0x0081)) return 0; if (!handle_errors) @@ -263,10 +265,12 @@ static void i82875p_check(struct mem_ctl_info *mci) /* Return 0 on success or 1 on failure. */ static int i82875p_setup_overfl_dev(struct pci_dev *pdev, - struct pci_dev **ovrfl_pdev, void __iomem **ovrfl_window) + struct pci_dev **ovrfl_pdev, + void __iomem **ovrfl_window) { struct pci_dev *dev; void __iomem *window; + int err; *ovrfl_pdev = NULL; *ovrfl_window = NULL; @@ -284,14 +288,19 @@ static int i82875p_setup_overfl_dev(struct pci_dev *pdev, if (dev == NULL) return 1; - pci_bus_add_device(dev); + err = pci_bus_add_device(dev); + if (err) { + i82875p_printk(KERN_ERR, + "%s(): pci_bus_add_device() Failed\n", + __func__); + } } *ovrfl_pdev = dev; if (pci_enable_device(dev)) { i82875p_printk(KERN_ERR, "%s(): Failed to enable overflow " - "device\n", __func__); + "device\n", __func__); return 1; } @@ -307,7 +316,7 @@ static int i82875p_setup_overfl_dev(struct pci_dev *pdev, if (window == NULL) { i82875p_printk(KERN_ERR, "%s(): Failed to ioremap bar6\n", - __func__); + __func__); goto fail1; } @@ -325,21 +334,20 @@ fail0: return 1; } - /* Return 1 if dual channel mode is active. Else return 0. */ static inline int dual_channel_active(u32 drc) { return (drc >> 21) & 0x1; } - static void i82875p_init_csrows(struct mem_ctl_info *mci, - struct pci_dev *pdev, void __iomem *ovrfl_window, u32 drc) + struct pci_dev *pdev, + void __iomem * ovrfl_window, u32 drc) { struct csrow_info *csrow; unsigned long last_cumul_size; u8 value; - u32 drc_ddim; /* DRAM Data Integrity Mode 0=none,2=edac */ + u32 drc_ddim; /* DRAM Data Integrity Mode 0=none,2=edac */ u32 cumul_size; int index; @@ -392,7 +400,7 @@ static int i82875p_probe1(struct pci_dev *pdev, int dev_idx) drc = readl(ovrfl_window + I82875P_DRC); nr_chans = dual_channel_active(drc) + 1; mci = edac_mc_alloc(sizeof(*pvt), I82875P_NR_CSROWS(nr_chans), - nr_chans); + nr_chans, 0); if (!mci) { rc = -ENOMEM; @@ -407,23 +415,35 @@ static int i82875p_probe1(struct pci_dev *pdev, int dev_idx) mci->mod_name = EDAC_MOD_STR; mci->mod_ver = I82875P_REVISION; mci->ctl_name = i82875p_devs[dev_idx].ctl_name; + mci->dev_name = pci_name(pdev); mci->edac_check = i82875p_check; mci->ctl_page_to_phys = NULL; debugf3("%s(): init pvt\n", __func__); - pvt = (struct i82875p_pvt *) mci->pvt_info; + pvt = (struct i82875p_pvt *)mci->pvt_info; pvt->ovrfl_pdev = ovrfl_pdev; pvt->ovrfl_window = ovrfl_window; i82875p_init_csrows(mci, pdev, ovrfl_window, drc); - i82875p_get_error_info(mci, &discard); /* clear counters */ + i82875p_get_error_info(mci, &discard); /* clear counters */ /* Here we assume that we will never see multiple instances of this * type of memory controller. The ID is therefore hardcoded to 0. */ - if (edac_mc_add_mc(mci,0)) { + if (edac_mc_add_mc(mci)) { debugf3("%s(): failed edac_mc_add_mc()\n", __func__); goto fail1; } + /* allocating generic PCI control info */ + i82875p_pci = edac_pci_create_generic_ctl(&pdev->dev, EDAC_MOD_STR); + if (!i82875p_pci) { + printk(KERN_WARNING + "%s(): Unable to create PCI control\n", + __func__); + printk(KERN_WARNING + "%s(): PCI error report via EDAC not setup\n", + __func__); + } + /* get this far and it's successful */ debugf3("%s(): success\n", __func__); return 0; @@ -442,7 +462,7 @@ fail0: /* returns count (>= 0), or negative on error */ static int __devinit i82875p_init_one(struct pci_dev *pdev, - const struct pci_device_id *ent) + const struct pci_device_id *ent) { int rc; @@ -467,10 +487,13 @@ static void __devexit i82875p_remove_one(struct pci_dev *pdev) debugf0("%s()\n", __func__); + if (i82875p_pci) + edac_pci_release_generic_ctl(i82875p_pci); + if ((mci = edac_mc_del_mc(&pdev->dev)) == NULL) return; - pvt = (struct i82875p_pvt *) mci->pvt_info; + pvt = (struct i82875p_pvt *)mci->pvt_info; if (pvt->ovrfl_window) iounmap(pvt->ovrfl_window); @@ -488,12 +511,11 @@ static void __devexit i82875p_remove_one(struct pci_dev *pdev) static const struct pci_device_id i82875p_pci_tbl[] __devinitdata = { { - PCI_VEND_DEV(INTEL, 82875_0), PCI_ANY_ID, PCI_ANY_ID, 0, 0, - I82875P - }, + PCI_VEND_DEV(INTEL, 82875_0), PCI_ANY_ID, PCI_ANY_ID, 0, 0, + I82875P}, { - 0, - } /* 0 terminated list. */ + 0, + } /* 0 terminated list. */ }; MODULE_DEVICE_TABLE(pci, i82875p_pci_tbl); @@ -517,7 +539,7 @@ static int __init i82875p_init(void) if (mci_pdev == NULL) { mci_pdev = pci_get_device(PCI_VENDOR_ID_INTEL, - PCI_DEVICE_ID_INTEL_82875_0, NULL); + PCI_DEVICE_ID_INTEL_82875_0, NULL); if (!mci_pdev) { debugf0("875p pci_get_device fail\n"); diff --git a/drivers/edac/i82975x_edac.c b/drivers/edac/i82975x_edac.c new file mode 100644 index 00000000000..0ee88845693 --- /dev/null +++ b/drivers/edac/i82975x_edac.c @@ -0,0 +1,666 @@ +/* + * Intel 82975X Memory Controller kernel module + * (C) 2007 aCarLab (India) Pvt. Ltd. (http://acarlab.com) + * (C) 2007 jetzbroadband (http://jetzbroadband.com) + * This file may be distributed under the terms of the + * GNU General Public License. + * + * Written by Arvind R. + * Copied from i82875p_edac.c source: + */ + +#include <linux/module.h> +#include <linux/init.h> +#include <linux/pci.h> +#include <linux/pci_ids.h> +#include <linux/slab.h> + +#include "edac_core.h" + +#define I82975X_REVISION " Ver: 1.0.0 " __DATE__ +#define EDAC_MOD_STR "i82975x_edac" + +#define i82975x_printk(level, fmt, arg...) \ + edac_printk(level, "i82975x", fmt, ##arg) + +#define i82975x_mc_printk(mci, level, fmt, arg...) \ + edac_mc_chipset_printk(mci, level, "i82975x", fmt, ##arg) + +#ifndef PCI_DEVICE_ID_INTEL_82975_0 +#define PCI_DEVICE_ID_INTEL_82975_0 0x277c +#endif /* PCI_DEVICE_ID_INTEL_82975_0 */ + +#define I82975X_NR_CSROWS(nr_chans) (8/(nr_chans)) + +/* Intel 82975X register addresses - device 0 function 0 - DRAM Controller */ +#define I82975X_EAP 0x58 /* Dram Error Address Pointer (32b) + * + * 31:7 128 byte cache-line address + * 6:1 reserved + * 0 0: CH0; 1: CH1 + */ + +#define I82975X_DERRSYN 0x5c /* Dram Error SYNdrome (8b) + * + * 7:0 DRAM ECC Syndrome + */ + +#define I82975X_DES 0x5d /* Dram ERRor DeSTination (8b) + * 0h: Processor Memory Reads + * 1h:7h reserved + * More - See Page 65 of Intel DocSheet. + */ + +#define I82975X_ERRSTS 0xc8 /* Error Status Register (16b) + * + * 15:12 reserved + * 11 Thermal Sensor Event + * 10 reserved + * 9 non-DRAM lock error (ndlock) + * 8 Refresh Timeout + * 7:2 reserved + * 1 ECC UE (multibit DRAM error) + * 0 ECC CE (singlebit DRAM error) + */ + +/* Error Reporting is supported by 3 mechanisms: + 1. DMI SERR generation ( ERRCMD ) + 2. SMI DMI generation ( SMICMD ) + 3. SCI DMI generation ( SCICMD ) +NOTE: Only ONE of the three must be enabled +*/ +#define I82975X_ERRCMD 0xca /* Error Command (16b) + * + * 15:12 reserved + * 11 Thermal Sensor Event + * 10 reserved + * 9 non-DRAM lock error (ndlock) + * 8 Refresh Timeout + * 7:2 reserved + * 1 ECC UE (multibit DRAM error) + * 0 ECC CE (singlebit DRAM error) + */ + +#define I82975X_SMICMD 0xcc /* Error Command (16b) + * + * 15:2 reserved + * 1 ECC UE (multibit DRAM error) + * 0 ECC CE (singlebit DRAM error) + */ + +#define I82975X_SCICMD 0xce /* Error Command (16b) + * + * 15:2 reserved + * 1 ECC UE (multibit DRAM error) + * 0 ECC CE (singlebit DRAM error) + */ + +#define I82975X_XEAP 0xfc /* Extended Dram Error Address Pointer (8b) + * + * 7:1 reserved + * 0 Bit32 of the Dram Error Address + */ + +#define I82975X_MCHBAR 0x44 /* + * + * 31:14 Base Addr of 16K memory-mapped + * configuration space + * 13:1 reserverd + * 0 mem-mapped config space enable + */ + +/* NOTE: Following addresses have to indexed using MCHBAR offset (44h, 32b) */ +/* Intel 82975x memory mapped register space */ + +#define I82975X_DRB_SHIFT 25 /* fixed 32MiB grain */ + +#define I82975X_DRB 0x100 /* DRAM Row Boundary (8b x 8) + * + * 7 set to 1 in highest DRB of + * channel if 4GB in ch. + * 6:2 upper boundary of rank in + * 32MB grains + * 1:0 set to 0 + */ +#define I82975X_DRB_CH0R0 0x100 +#define I82975X_DRB_CH0R1 0x101 +#define I82975X_DRB_CH0R2 0x102 +#define I82975X_DRB_CH0R3 0x103 +#define I82975X_DRB_CH1R0 0x180 +#define I82975X_DRB_CH1R1 0x181 +#define I82975X_DRB_CH1R2 0x182 +#define I82975X_DRB_CH1R3 0x183 + + +#define I82975X_DRA 0x108 /* DRAM Row Attribute (4b x 8) + * defines the PAGE SIZE to be used + * for the rank + * 7 reserved + * 6:4 row attr of odd rank, i.e. 1 + * 3 reserved + * 2:0 row attr of even rank, i.e. 0 + * + * 000 = unpopulated + * 001 = reserved + * 010 = 4KiB + * 011 = 8KiB + * 100 = 16KiB + * others = reserved + */ +#define I82975X_DRA_CH0R01 0x108 +#define I82975X_DRA_CH0R23 0x109 +#define I82975X_DRA_CH1R01 0x188 +#define I82975X_DRA_CH1R23 0x189 + + +#define I82975X_BNKARC 0x10e /* Type of device in each rank - Bank Arch (16b) + * + * 15:8 reserved + * 7:6 Rank 3 architecture + * 5:4 Rank 2 architecture + * 3:2 Rank 1 architecture + * 1:0 Rank 0 architecture + * + * 00 => x16 devices; i.e 4 banks + * 01 => x8 devices; i.e 8 banks + */ +#define I82975X_C0BNKARC 0x10e +#define I82975X_C1BNKARC 0x18e + + + +#define I82975X_DRC 0x120 /* DRAM Controller Mode0 (32b) + * + * 31:30 reserved + * 29 init complete + * 28:11 reserved, according to Intel + * 22:21 number of channels + * 00=1 01=2 in 82875 + * seems to be ECC mode + * bits in 82975 in Asus + * P5W + * 19:18 Data Integ Mode + * 00=none 01=ECC in 82875 + * 10:8 refresh mode + * 7 reserved + * 6:4 mode select + * 3:2 reserved + * 1:0 DRAM type 10=Second Revision + * DDR2 SDRAM + * 00, 01, 11 reserved + */ +#define I82975X_DRC_CH0M0 0x120 +#define I82975X_DRC_CH1M0 0x1A0 + + +#define I82975X_DRC_M1 0x124 /* DRAM Controller Mode1 (32b) + * 31 0=Standard Address Map + * 1=Enhanced Address Map + * 30:0 reserved + */ + +#define I82975X_DRC_CH0M1 0x124 +#define I82975X_DRC_CH1M1 0x1A4 + +enum i82975x_chips { + I82975X = 0, +}; + +struct i82975x_pvt { + void __iomem *mch_window; +}; + +struct i82975x_dev_info { + const char *ctl_name; +}; + +struct i82975x_error_info { + u16 errsts; + u32 eap; + u8 des; + u8 derrsyn; + u16 errsts2; + u8 chan; /* the channel is bit 0 of EAP */ + u8 xeap; /* extended eap bit */ +}; + +static const struct i82975x_dev_info i82975x_devs[] = { + [I82975X] = { + .ctl_name = "i82975x" + }, +}; + +static struct pci_dev *mci_pdev; /* init dev: in case that AGP code has + * already registered driver + */ + +static int i82975x_registered = 1; + +static void i82975x_get_error_info(struct mem_ctl_info *mci, + struct i82975x_error_info *info) +{ + struct pci_dev *pdev; + + pdev = to_pci_dev(mci->dev); + + /* + * This is a mess because there is no atomic way to read all the + * registers at once and the registers can transition from CE being + * overwritten by UE. + */ + pci_read_config_word(pdev, I82975X_ERRSTS, &info->errsts); + pci_read_config_dword(pdev, I82975X_EAP, &info->eap); + pci_read_config_byte(pdev, I82975X_XEAP, &info->xeap); + pci_read_config_byte(pdev, I82975X_DES, &info->des); + pci_read_config_byte(pdev, I82975X_DERRSYN, &info->derrsyn); + pci_read_config_word(pdev, I82975X_ERRSTS, &info->errsts2); + + pci_write_bits16(pdev, I82975X_ERRSTS, 0x0003, 0x0003); + + /* + * If the error is the same then we can for both reads then + * the first set of reads is valid. If there is a change then + * there is a CE no info and the second set of reads is valid + * and should be UE info. + */ + if (!(info->errsts2 & 0x0003)) + return; + + if ((info->errsts ^ info->errsts2) & 0x0003) { + pci_read_config_dword(pdev, I82975X_EAP, &info->eap); + pci_read_config_byte(pdev, I82975X_XEAP, &info->xeap); + pci_read_config_byte(pdev, I82975X_DES, &info->des); + pci_read_config_byte(pdev, I82975X_DERRSYN, + &info->derrsyn); + } +} + +static int i82975x_process_error_info(struct mem_ctl_info *mci, + struct i82975x_error_info *info, int handle_errors) +{ + int row, multi_chan, chan; + + multi_chan = mci->csrows[0].nr_channels - 1; + + if (!(info->errsts2 & 0x0003)) + return 0; + + if (!handle_errors) + return 1; + + if ((info->errsts ^ info->errsts2) & 0x0003) { + edac_mc_handle_ce_no_info(mci, "UE overwrote CE"); + info->errsts = info->errsts2; + } + + chan = info->eap & 1; + info->eap >>= 1; + if (info->xeap ) + info->eap |= 0x80000000; + info->eap >>= PAGE_SHIFT; + row = edac_mc_find_csrow_by_page(mci, info->eap); + + if (info->errsts & 0x0002) + edac_mc_handle_ue(mci, info->eap, 0, row, "i82975x UE"); + else + edac_mc_handle_ce(mci, info->eap, 0, info->derrsyn, row, + multi_chan ? chan : 0, + "i82975x CE"); + + return 1; +} + +static void i82975x_check(struct mem_ctl_info *mci) +{ + struct i82975x_error_info info; + + debugf1("MC%d: %s()\n", mci->mc_idx, __func__); + i82975x_get_error_info(mci, &info); + i82975x_process_error_info(mci, &info, 1); +} + +/* Return 1 if dual channel mode is active. Else return 0. */ +static int dual_channel_active(void __iomem *mch_window) +{ + /* + * We treat interleaved-symmetric configuration as dual-channel - EAP's + * bit-0 giving the channel of the error location. + * + * All other configurations are treated as single channel - the EAP's + * bit-0 will resolve ok in symmetric area of mixed + * (symmetric/asymmetric) configurations + */ + u8 drb[4][2]; + int row; + int dualch; + + for (dualch = 1, row = 0; dualch && (row < 4); row++) { + drb[row][0] = readb(mch_window + I82975X_DRB + row); + drb[row][1] = readb(mch_window + I82975X_DRB + row + 0x80); + dualch = dualch && (drb[row][0] == drb[row][1]); + } + return dualch; +} + +static enum dev_type i82975x_dram_type(void __iomem *mch_window, int rank) +{ + /* + * ASUS P5W DH either does not program this register or programs + * it wrong! + * ECC is possible on i92975x ONLY with DEV_X8 which should mean 'val' + * for each rank should be 01b - the LSB of the word should be 0x55; + * but it reads 0! + */ + return DEV_X8; +} + +static void i82975x_init_csrows(struct mem_ctl_info *mci, + struct pci_dev *pdev, void __iomem *mch_window) +{ + struct csrow_info *csrow; + unsigned long last_cumul_size; + u8 value; + u32 cumul_size; + int index; + + last_cumul_size = 0; + + /* + * 82875 comment: + * The dram row boundary (DRB) reg values are boundary address + * for each DRAM row with a granularity of 32 or 64MB (single/dual + * channel operation). DRB regs are cumulative; therefore DRB7 will + * contain the total memory contained in all eight rows. + * + * FIXME: + * EDAC currently works for Dual-channel Interleaved configuration. + * Other configurations, which the chip supports, need fixing/testing. + * + */ + + for (index = 0; index < mci->nr_csrows; index++) { + csrow = &mci->csrows[index]; + + value = readb(mch_window + I82975X_DRB + index + + ((index >= 4) ? 0x80 : 0)); + cumul_size = value; + cumul_size <<= (I82975X_DRB_SHIFT - PAGE_SHIFT); + debugf3("%s(): (%d) cumul_size 0x%x\n", __func__, index, + cumul_size); + if (cumul_size == last_cumul_size) + continue; /* not populated */ + + csrow->first_page = last_cumul_size; + csrow->last_page = cumul_size - 1; + csrow->nr_pages = cumul_size - last_cumul_size; + last_cumul_size = cumul_size; + csrow->grain = 1 << 7; /* I82975X_EAP has 128B resolution */ + csrow->mtype = MEM_DDR; /* i82975x supports only DDR2 */ + csrow->dtype = i82975x_dram_type(mch_window, index); + csrow->edac_mode = EDAC_SECDED; /* only supported */ + } +} + +/* #define i82975x_DEBUG_IOMEM */ + +#ifdef i82975x_DEBUG_IOMEM +static void i82975x_print_dram_timings(void __iomem *mch_window) +{ + /* + * The register meanings are from Intel specs; + * (shows 13-5-5-5 for 800-DDR2) + * Asus P5W Bios reports 15-5-4-4 + * What's your religion? + */ + static const int caslats[4] = { 5, 4, 3, 6 }; + u32 dtreg[2]; + + dtreg[0] = readl(mch_window + 0x114); + dtreg[1] = readl(mch_window + 0x194); + i82975x_printk(KERN_INFO, "DRAM Timings : Ch0 Ch1\n" + " RAS Active Min = %d %d\n" + " CAS latency = %d %d\n" + " RAS to CAS = %d %d\n" + " RAS precharge = %d %d\n", + (dtreg[0] >> 19 ) & 0x0f, + (dtreg[1] >> 19) & 0x0f, + caslats[(dtreg[0] >> 8) & 0x03], + caslats[(dtreg[1] >> 8) & 0x03], + ((dtreg[0] >> 4) & 0x07) + 2, + ((dtreg[1] >> 4) & 0x07) + 2, + (dtreg[0] & 0x07) + 2, + (dtreg[1] & 0x07) + 2 + ); + +} +#endif + +static int i82975x_probe1(struct pci_dev *pdev, int dev_idx) +{ + int rc = -ENODEV; + struct mem_ctl_info *mci; + struct i82975x_pvt *pvt; + void __iomem *mch_window; + u32 mchbar; + u32 drc[2]; + struct i82975x_error_info discard; + int chans; +#ifdef i82975x_DEBUG_IOMEM + u8 c0drb[4]; + u8 c1drb[4]; +#endif + + debugf0("%s()\n", __func__); + + pci_read_config_dword(pdev, I82975X_MCHBAR, &mchbar); + if (!(mchbar & 1)) { + debugf3("%s(): failed, MCHBAR disabled!\n", __func__); + goto fail0; + } + mchbar &= 0xffffc000; /* bits 31:14 used for 16K window */ + mch_window = ioremap_nocache(mchbar, 0x1000); + +#ifdef i82975x_DEBUG_IOMEM + i82975x_printk(KERN_INFO, "MCHBAR real = %0x, remapped = %p\n", + mchbar, mch_window); + + c0drb[0] = readb(mch_window + I82975X_DRB_CH0R0); + c0drb[1] = readb(mch_window + I82975X_DRB_CH0R1); + c0drb[2] = readb(mch_window + I82975X_DRB_CH0R2); + c0drb[3] = readb(mch_window + I82975X_DRB_CH0R3); + c1drb[0] = readb(mch_window + I82975X_DRB_CH1R0); + c1drb[1] = readb(mch_window + I82975X_DRB_CH1R1); + c1drb[2] = readb(mch_window + I82975X_DRB_CH1R2); + c1drb[3] = readb(mch_window + I82975X_DRB_CH1R3); + i82975x_printk(KERN_INFO, "DRBCH0R0 = 0x%02x\n", c0drb[0]); + i82975x_printk(KERN_INFO, "DRBCH0R1 = 0x%02x\n", c0drb[1]); + i82975x_printk(KERN_INFO, "DRBCH0R2 = 0x%02x\n", c0drb[2]); + i82975x_printk(KERN_INFO, "DRBCH0R3 = 0x%02x\n", c0drb[3]); + i82975x_printk(KERN_INFO, "DRBCH1R0 = 0x%02x\n", c1drb[0]); + i82975x_printk(KERN_INFO, "DRBCH1R1 = 0x%02x\n", c1drb[1]); + i82975x_printk(KERN_INFO, "DRBCH1R2 = 0x%02x\n", c1drb[2]); + i82975x_printk(KERN_INFO, "DRBCH1R3 = 0x%02x\n", c1drb[3]); +#endif + + drc[0] = readl(mch_window + I82975X_DRC_CH0M0); + drc[1] = readl(mch_window + I82975X_DRC_CH1M0); +#ifdef i82975x_DEBUG_IOMEM + i82975x_printk(KERN_INFO, "DRC_CH0 = %0x, %s\n", drc[0], + ((drc[0] >> 21) & 3) == 1 ? + "ECC enabled" : "ECC disabled"); + i82975x_printk(KERN_INFO, "DRC_CH1 = %0x, %s\n", drc[1], + ((drc[1] >> 21) & 3) == 1 ? + "ECC enabled" : "ECC disabled"); + + i82975x_printk(KERN_INFO, "C0 BNKARC = %0x\n", + readw(mch_window + I82975X_C0BNKARC)); + i82975x_printk(KERN_INFO, "C1 BNKARC = %0x\n", + readw(mch_window + I82975X_C1BNKARC)); + i82975x_print_dram_timings(mch_window); + goto fail1; +#endif + if (!(((drc[0] >> 21) & 3) == 1 || ((drc[1] >> 21) & 3) == 1)) { + i82975x_printk(KERN_INFO, "ECC disabled on both channels.\n"); + goto fail1; + } + + chans = dual_channel_active(mch_window) + 1; + + /* assuming only one controller, index thus is 0 */ + mci = edac_mc_alloc(sizeof(*pvt), I82975X_NR_CSROWS(chans), + chans, 0); + if (!mci) { + rc = -ENOMEM; + goto fail1; + } + + debugf3("%s(): init mci\n", __func__); + mci->dev = &pdev->dev; + mci->mtype_cap = MEM_FLAG_DDR; + mci->edac_ctl_cap = EDAC_FLAG_NONE | EDAC_FLAG_SECDED; + mci->edac_cap = EDAC_FLAG_NONE | EDAC_FLAG_SECDED; + mci->mod_name = EDAC_MOD_STR; + mci->mod_ver = I82975X_REVISION; + mci->ctl_name = i82975x_devs[dev_idx].ctl_name; + mci->edac_check = i82975x_check; + mci->ctl_page_to_phys = NULL; + debugf3("%s(): init pvt\n", __func__); + pvt = (struct i82975x_pvt *) mci->pvt_info; + pvt->mch_window = mch_window; + i82975x_init_csrows(mci, pdev, mch_window); + i82975x_get_error_info(mci, &discard); /* clear counters */ + + /* finalize this instance of memory controller with edac core */ + if (edac_mc_add_mc(mci)) { + debugf3("%s(): failed edac_mc_add_mc()\n", __func__); + goto fail2; + } + + /* get this far and it's successful */ + debugf3("%s(): success\n", __func__); + return 0; + +fail2: + edac_mc_free(mci); + +fail1: + iounmap(mch_window); +fail0: + return rc; +} + +/* returns count (>= 0), or negative on error */ +static int __devinit i82975x_init_one(struct pci_dev *pdev, + const struct pci_device_id *ent) +{ + int rc; + + debugf0("%s()\n", __func__); + + if (pci_enable_device(pdev) < 0) + return -EIO; + + rc = i82975x_probe1(pdev, ent->driver_data); + + if (mci_pdev == NULL) + mci_pdev = pci_dev_get(pdev); + + return rc; +} + +static void __devexit i82975x_remove_one(struct pci_dev *pdev) +{ + struct mem_ctl_info *mci; + struct i82975x_pvt *pvt; + + debugf0("%s()\n", __func__); + + mci = edac_mc_del_mc(&pdev->dev); + if (mci == NULL) + return; + + pvt = mci->pvt_info; + if (pvt->mch_window) + iounmap( pvt->mch_window ); + + edac_mc_free(mci); +} + +static const struct pci_device_id i82975x_pci_tbl[] __devinitdata = { + { + PCI_VEND_DEV(INTEL, 82975_0), PCI_ANY_ID, PCI_ANY_ID, 0, 0, + I82975X + }, + { + 0, + } /* 0 terminated list. */ +}; + +MODULE_DEVICE_TABLE(pci, i82975x_pci_tbl); + +static struct pci_driver i82975x_driver = { + .name = EDAC_MOD_STR, + .probe = i82975x_init_one, + .remove = __devexit_p(i82975x_remove_one), + .id_table = i82975x_pci_tbl, +}; + +static int __init i82975x_init(void) +{ + int pci_rc; + + debugf3("%s()\n", __func__); + + pci_rc = pci_register_driver(&i82975x_driver); + if (pci_rc < 0) + goto fail0; + + if (mci_pdev == NULL) { + mci_pdev = pci_get_device(PCI_VENDOR_ID_INTEL, + PCI_DEVICE_ID_INTEL_82975_0, NULL); + + if (!mci_pdev) { + debugf0("i82975x pci_get_device fail\n"); + pci_rc = -ENODEV; + goto fail1; + } + + pci_rc = i82975x_init_one(mci_pdev, i82975x_pci_tbl); + + if (pci_rc < 0) { + debugf0("i82975x init fail\n"); + pci_rc = -ENODEV; + goto fail1; + } + } + + return 0; + +fail1: + pci_unregister_driver(&i82975x_driver); + +fail0: + if (mci_pdev != NULL) + pci_dev_put(mci_pdev); + + return pci_rc; +} + +static void __exit i82975x_exit(void) +{ + debugf3("%s()\n", __func__); + + pci_unregister_driver(&i82975x_driver); + + if (!i82975x_registered) { + i82975x_remove_one(mci_pdev); + pci_dev_put(mci_pdev); + } +} + +module_init(i82975x_init); +module_exit(i82975x_exit); + +MODULE_LICENSE("GPL"); +MODULE_AUTHOR("Arvind R. <arvind@acarlab.com>"); +MODULE_DESCRIPTION("MC support for Intel 82975 memory hub controllers"); diff --git a/drivers/edac/pasemi_edac.c b/drivers/edac/pasemi_edac.c new file mode 100644 index 00000000000..e66cdd42a39 --- /dev/null +++ b/drivers/edac/pasemi_edac.c @@ -0,0 +1,299 @@ +/* + * Copyright (C) 2006-2007 PA Semi, Inc + * + * Author: Egor Martovetsky <egor@pasemi.com> + * Maintained by: Olof Johansson <olof@lixom.net> + * + * Driver for the PWRficient onchip memory controllers + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA + */ + + +#include <linux/module.h> +#include <linux/init.h> +#include <linux/pci.h> +#include <linux/pci_ids.h> +#include <linux/slab.h> +#include "edac_core.h" + +#define MODULE_NAME "pasemi_edac" + +#define MCCFG_MCEN 0x300 +#define MCCFG_MCEN_MMC_EN 0x00000001 +#define MCCFG_ERRCOR 0x388 +#define MCCFG_ERRCOR_RNK_FAIL_DET_EN 0x00000100 +#define MCCFG_ERRCOR_ECC_GEN_EN 0x00000010 +#define MCCFG_ERRCOR_ECC_CRR_EN 0x00000001 +#define MCCFG_SCRUB 0x384 +#define MCCFG_SCRUB_RGLR_SCRB_EN 0x00000001 +#define MCDEBUG_ERRCTL1 0x728 +#define MCDEBUG_ERRCTL1_RFL_LOG_EN 0x00080000 +#define MCDEBUG_ERRCTL1_MBE_LOG_EN 0x00040000 +#define MCDEBUG_ERRCTL1_SBE_LOG_EN 0x00020000 +#define MCDEBUG_ERRSTA 0x730 +#define MCDEBUG_ERRSTA_RFL_STATUS 0x00000004 +#define MCDEBUG_ERRSTA_MBE_STATUS 0x00000002 +#define MCDEBUG_ERRSTA_SBE_STATUS 0x00000001 +#define MCDEBUG_ERRCNT1 0x734 +#define MCDEBUG_ERRCNT1_SBE_CNT_OVRFLO 0x00000080 +#define MCDEBUG_ERRLOG1A 0x738 +#define MCDEBUG_ERRLOG1A_MERR_TYPE_M 0x30000000 +#define MCDEBUG_ERRLOG1A_MERR_TYPE_NONE 0x00000000 +#define MCDEBUG_ERRLOG1A_MERR_TYPE_SBE 0x10000000 +#define MCDEBUG_ERRLOG1A_MERR_TYPE_MBE 0x20000000 +#define MCDEBUG_ERRLOG1A_MERR_TYPE_RFL 0x30000000 +#define MCDEBUG_ERRLOG1A_MERR_BA_M 0x00700000 +#define MCDEBUG_ERRLOG1A_MERR_BA_S 20 +#define MCDEBUG_ERRLOG1A_MERR_CS_M 0x00070000 +#define MCDEBUG_ERRLOG1A_MERR_CS_S 16 +#define MCDEBUG_ERRLOG1A_SYNDROME_M 0x0000ffff +#define MCDRAM_RANKCFG 0x114 +#define MCDRAM_RANKCFG_EN 0x00000001 +#define MCDRAM_RANKCFG_TYPE_SIZE_M 0x000001c0 +#define MCDRAM_RANKCFG_TYPE_SIZE_S 6 + +#define PASEMI_EDAC_NR_CSROWS 8 +#define PASEMI_EDAC_NR_CHANS 1 +#define PASEMI_EDAC_ERROR_GRAIN 64 + +static int last_page_in_mmc; +static int system_mmc_id; + + +static u32 pasemi_edac_get_error_info(struct mem_ctl_info *mci) +{ + struct pci_dev *pdev = to_pci_dev(mci->dev); + u32 tmp; + + pci_read_config_dword(pdev, MCDEBUG_ERRSTA, + &tmp); + + tmp &= (MCDEBUG_ERRSTA_RFL_STATUS | MCDEBUG_ERRSTA_MBE_STATUS + | MCDEBUG_ERRSTA_SBE_STATUS); + + if (tmp) { + if (tmp & MCDEBUG_ERRSTA_SBE_STATUS) + pci_write_config_dword(pdev, MCDEBUG_ERRCNT1, + MCDEBUG_ERRCNT1_SBE_CNT_OVRFLO); + pci_write_config_dword(pdev, MCDEBUG_ERRSTA, tmp); + } + + return tmp; +} + +static void pasemi_edac_process_error_info(struct mem_ctl_info *mci, u32 errsta) +{ + struct pci_dev *pdev = to_pci_dev(mci->dev); + u32 errlog1a; + u32 cs; + + if (!errsta) + return; + + pci_read_config_dword(pdev, MCDEBUG_ERRLOG1A, &errlog1a); + + cs = (errlog1a & MCDEBUG_ERRLOG1A_MERR_CS_M) >> + MCDEBUG_ERRLOG1A_MERR_CS_S; + + /* uncorrectable/multi-bit errors */ + if (errsta & (MCDEBUG_ERRSTA_MBE_STATUS | + MCDEBUG_ERRSTA_RFL_STATUS)) { + edac_mc_handle_ue(mci, mci->csrows[cs].first_page, 0, + cs, mci->ctl_name); + } + + /* correctable/single-bit errors */ + if (errsta & MCDEBUG_ERRSTA_SBE_STATUS) { + edac_mc_handle_ce(mci, mci->csrows[cs].first_page, 0, + 0, cs, 0, mci->ctl_name); + } +} + +static void pasemi_edac_check(struct mem_ctl_info *mci) +{ + u32 errsta; + + errsta = pasemi_edac_get_error_info(mci); + if (errsta) + pasemi_edac_process_error_info(mci, errsta); +} + +static int pasemi_edac_init_csrows(struct mem_ctl_info *mci, + struct pci_dev *pdev, + enum edac_type edac_mode) +{ + struct csrow_info *csrow; + u32 rankcfg; + int index; + + for (index = 0; index < mci->nr_csrows; index++) { + csrow = &mci->csrows[index]; + + pci_read_config_dword(pdev, + MCDRAM_RANKCFG + (index * 12), + &rankcfg); + + if (!(rankcfg & MCDRAM_RANKCFG_EN)) + continue; + + switch ((rankcfg & MCDRAM_RANKCFG_TYPE_SIZE_M) >> + MCDRAM_RANKCFG_TYPE_SIZE_S) { + case 0: + csrow->nr_pages = 128 << (20 - PAGE_SHIFT); + break; + case 1: + csrow->nr_pages = 256 << (20 - PAGE_SHIFT); + break; + case 2: + case 3: + csrow->nr_pages = 512 << (20 - PAGE_SHIFT); + break; + case 4: + csrow->nr_pages = 1024 << (20 - PAGE_SHIFT); + break; + case 5: + csrow->nr_pages = 2048 << (20 - PAGE_SHIFT); + break; + default: + edac_mc_printk(mci, KERN_ERR, + "Unrecognized Rank Config. rankcfg=%u\n", + rankcfg); + return -EINVAL; + } + + csrow->first_page = last_page_in_mmc; + csrow->last_page = csrow->first_page + csrow->nr_pages - 1; + last_page_in_mmc += csrow->nr_pages; + csrow->page_mask = 0; + csrow->grain = PASEMI_EDAC_ERROR_GRAIN; + csrow->mtype = MEM_DDR; + csrow->dtype = DEV_UNKNOWN; + csrow->edac_mode = edac_mode; + } + return 0; +} + +static int __devinit pasemi_edac_probe(struct pci_dev *pdev, + const struct pci_device_id *ent) +{ + struct mem_ctl_info *mci = NULL; + u32 errctl1, errcor, scrub, mcen; + + pci_read_config_dword(pdev, MCCFG_MCEN, &mcen); + if (!(mcen & MCCFG_MCEN_MMC_EN)) + return -ENODEV; + + /* + * We should think about enabling other error detection later on + */ + + pci_read_config_dword(pdev, MCDEBUG_ERRCTL1, &errctl1); + errctl1 |= MCDEBUG_ERRCTL1_SBE_LOG_EN | + MCDEBUG_ERRCTL1_MBE_LOG_EN | + MCDEBUG_ERRCTL1_RFL_LOG_EN; + pci_write_config_dword(pdev, MCDEBUG_ERRCTL1, errctl1); + + mci = edac_mc_alloc(0, PASEMI_EDAC_NR_CSROWS, PASEMI_EDAC_NR_CHANS, + system_mmc_id++); + + if (mci == NULL) + return -ENOMEM; + + pci_read_config_dword(pdev, MCCFG_ERRCOR, &errcor); + errcor |= MCCFG_ERRCOR_RNK_FAIL_DET_EN | + MCCFG_ERRCOR_ECC_GEN_EN | + MCCFG_ERRCOR_ECC_CRR_EN; + + mci->dev = &pdev->dev; + mci->mtype_cap = MEM_FLAG_DDR | MEM_FLAG_RDDR; + mci->edac_ctl_cap = EDAC_FLAG_NONE | EDAC_FLAG_EC | EDAC_FLAG_SECDED; + mci->edac_cap = (errcor & MCCFG_ERRCOR_ECC_GEN_EN) ? + ((errcor & MCCFG_ERRCOR_ECC_CRR_EN) ? + (EDAC_FLAG_EC | EDAC_FLAG_SECDED) : EDAC_FLAG_EC) : + EDAC_FLAG_NONE; + mci->mod_name = MODULE_NAME; + mci->dev_name = pci_name(pdev); + mci->ctl_name = "pasemi,1682m-mc"; + mci->edac_check = pasemi_edac_check; + mci->ctl_page_to_phys = NULL; + pci_read_config_dword(pdev, MCCFG_SCRUB, &scrub); + mci->scrub_cap = SCRUB_FLAG_HW_PROG | SCRUB_FLAG_HW_SRC; + mci->scrub_mode = + ((errcor & MCCFG_ERRCOR_ECC_CRR_EN) ? SCRUB_FLAG_HW_SRC : 0) | + ((scrub & MCCFG_SCRUB_RGLR_SCRB_EN) ? SCRUB_FLAG_HW_PROG : 0); + + if (pasemi_edac_init_csrows(mci, pdev, + (mci->edac_cap & EDAC_FLAG_SECDED) ? + EDAC_SECDED : + ((mci->edac_cap & EDAC_FLAG_EC) ? + EDAC_EC : EDAC_NONE))) + goto fail; + + /* + * Clear status + */ + pasemi_edac_get_error_info(mci); + + if (edac_mc_add_mc(mci)) + goto fail; + + /* get this far and it's successful */ + return 0; + +fail: + edac_mc_free(mci); + return -ENODEV; +} + +static void __devexit pasemi_edac_remove(struct pci_dev *pdev) +{ + struct mem_ctl_info *mci = edac_mc_del_mc(&pdev->dev); + + if (!mci) + return; + + edac_mc_free(mci); +} + + +static const struct pci_device_id pasemi_edac_pci_tbl[] = { + { PCI_DEVICE(PCI_VENDOR_ID_PASEMI, 0xa00a) }, +}; + +MODULE_DEVICE_TABLE(pci, pasemi_edac_pci_tbl); + +static struct pci_driver pasemi_edac_driver = { + .name = MODULE_NAME, + .probe = pasemi_edac_probe, + .remove = __devexit_p(pasemi_edac_remove), + .id_table = pasemi_edac_pci_tbl, +}; + +static int __init pasemi_edac_init(void) +{ + return pci_register_driver(&pasemi_edac_driver); +} + +static void __exit pasemi_edac_exit(void) +{ + pci_unregister_driver(&pasemi_edac_driver); +} + +module_init(pasemi_edac_init); +module_exit(pasemi_edac_exit); + +MODULE_LICENSE("GPL"); +MODULE_AUTHOR("Egor Martovetsky <egor@pasemi.com>"); +MODULE_DESCRIPTION("MC support for PA Semi PA6T-1682M memory controller"); diff --git a/drivers/edac/r82600_edac.c b/drivers/edac/r82600_edac.c index a49cf0a3939..e25f712f2dc 100644 --- a/drivers/edac/r82600_edac.c +++ b/drivers/edac/r82600_edac.c @@ -11,7 +11,7 @@ * * Written with reference to 82600 High Integration Dual PCI System * Controller Data Book: - * http://www.radisys.com/files/support_downloads/007-01277-0002.82600DataBook.pdf + * www.radisys.com/files/support_downloads/007-01277-0002.82600DataBook.pdf * references to this document given in [] */ @@ -20,9 +20,9 @@ #include <linux/pci.h> #include <linux/pci_ids.h> #include <linux/slab.h> -#include "edac_mc.h" +#include "edac_core.h" -#define R82600_REVISION " Ver: 2.0.1 " __DATE__ +#define R82600_REVISION " Ver: 2.0.2 " __DATE__ #define EDAC_MOD_STR "r82600_edac" #define r82600_printk(level, fmt, arg...) \ @@ -131,10 +131,12 @@ struct r82600_error_info { u32 eapr; }; -static unsigned int disable_hardware_scrub = 0; +static unsigned int disable_hardware_scrub; -static void r82600_get_error_info (struct mem_ctl_info *mci, - struct r82600_error_info *info) +static struct edac_pci_ctl_info *r82600_pci; + +static void r82600_get_error_info(struct mem_ctl_info *mci, + struct r82600_error_info *info) { struct pci_dev *pdev; @@ -144,18 +146,19 @@ static void r82600_get_error_info (struct mem_ctl_info *mci, if (info->eapr & BIT(0)) /* Clear error to allow next error to be reported [p.62] */ pci_write_bits32(pdev, R82600_EAP, - ((u32) BIT(0) & (u32) BIT(1)), - ((u32) BIT(0) & (u32) BIT(1))); + ((u32) BIT(0) & (u32) BIT(1)), + ((u32) BIT(0) & (u32) BIT(1))); if (info->eapr & BIT(1)) /* Clear error to allow next error to be reported [p.62] */ pci_write_bits32(pdev, R82600_EAP, - ((u32) BIT(0) & (u32) BIT(1)), - ((u32) BIT(0) & (u32) BIT(1))); + ((u32) BIT(0) & (u32) BIT(1)), + ((u32) BIT(0) & (u32) BIT(1))); } -static int r82600_process_error_info (struct mem_ctl_info *mci, - struct r82600_error_info *info, int handle_errors) +static int r82600_process_error_info(struct mem_ctl_info *mci, + struct r82600_error_info *info, + int handle_errors) { int error_found; u32 eapaddr, page; @@ -172,25 +175,24 @@ static int r82600_process_error_info (struct mem_ctl_info *mci, * granularity (upper 19 bits only) */ page = eapaddr >> PAGE_SHIFT; - if (info->eapr & BIT(0)) { /* CE? */ + if (info->eapr & BIT(0)) { /* CE? */ error_found = 1; if (handle_errors) - edac_mc_handle_ce(mci, page, 0, /* not avail */ + edac_mc_handle_ce(mci, page, 0, /* not avail */ syndrome, edac_mc_find_csrow_by_page(mci, page), - 0, /* channel */ - mci->ctl_name); + 0, mci->ctl_name); } - if (info->eapr & BIT(1)) { /* UE? */ + if (info->eapr & BIT(1)) { /* UE? */ error_found = 1; if (handle_errors) /* 82600 doesn't give enough info */ edac_mc_handle_ue(mci, page, 0, - edac_mc_find_csrow_by_page(mci, page), - mci->ctl_name); + edac_mc_find_csrow_by_page(mci, page), + mci->ctl_name); } return error_found; @@ -211,11 +213,11 @@ static inline int ecc_enabled(u8 dramcr) } static void r82600_init_csrows(struct mem_ctl_info *mci, struct pci_dev *pdev, - u8 dramcr) + u8 dramcr) { struct csrow_info *csrow; int index; - u8 drbar; /* SDRAM Row Boundry Address Register */ + u8 drbar; /* SDRAM Row Boundry Address Register */ u32 row_high_limit, row_high_limit_last; u32 reg_sdram, ecc_on, row_base; @@ -276,7 +278,7 @@ static int r82600_probe1(struct pci_dev *pdev, int dev_idx) debugf2("%s(): sdram refresh rate = %#0x\n", __func__, sdram_refresh_rate); debugf2("%s(): DRAMC register = %#0x\n", __func__, dramcr); - mci = edac_mc_alloc(0, R82600_NR_CSROWS, R82600_NR_CHANS); + mci = edac_mc_alloc(0, R82600_NR_CSROWS, R82600_NR_CHANS, 0); if (mci == NULL) return -ENOMEM; @@ -305,15 +307,16 @@ static int r82600_probe1(struct pci_dev *pdev, int dev_idx) mci->mod_name = EDAC_MOD_STR; mci->mod_ver = R82600_REVISION; mci->ctl_name = "R82600"; + mci->dev_name = pci_name(pdev); mci->edac_check = r82600_check; mci->ctl_page_to_phys = NULL; r82600_init_csrows(mci, pdev, dramcr); - r82600_get_error_info(mci, &discard); /* clear counters */ + r82600_get_error_info(mci, &discard); /* clear counters */ /* Here we assume that we will never see multiple instances of this * type of memory controller. The ID is therefore hardcoded to 0. */ - if (edac_mc_add_mc(mci,0)) { + if (edac_mc_add_mc(mci)) { debugf3("%s(): failed edac_mc_add_mc()\n", __func__); goto fail; } @@ -326,6 +329,17 @@ static int r82600_probe1(struct pci_dev *pdev, int dev_idx) pci_write_bits32(pdev, R82600_EAP, BIT(31), BIT(31)); } + /* allocating generic PCI control info */ + r82600_pci = edac_pci_create_generic_ctl(&pdev->dev, EDAC_MOD_STR); + if (!r82600_pci) { + printk(KERN_WARNING + "%s(): Unable to create PCI control\n", + __func__); + printk(KERN_WARNING + "%s(): PCI error report via EDAC not setup\n", + __func__); + } + debugf3("%s(): success\n", __func__); return 0; @@ -336,7 +350,7 @@ fail: /* returns count (>= 0), or negative on error */ static int __devinit r82600_init_one(struct pci_dev *pdev, - const struct pci_device_id *ent) + const struct pci_device_id *ent) { debugf0("%s()\n", __func__); @@ -350,6 +364,9 @@ static void __devexit r82600_remove_one(struct pci_dev *pdev) debugf0("%s()\n", __func__); + if (r82600_pci) + edac_pci_release_generic_ctl(r82600_pci); + if ((mci = edac_mc_del_mc(&pdev->dev)) == NULL) return; @@ -358,11 +375,11 @@ static void __devexit r82600_remove_one(struct pci_dev *pdev) static const struct pci_device_id r82600_pci_tbl[] __devinitdata = { { - PCI_DEVICE(PCI_VENDOR_ID_RADISYS, R82600_BRIDGE_ID) - }, + PCI_DEVICE(PCI_VENDOR_ID_RADISYS, R82600_BRIDGE_ID) + }, { - 0, - } /* 0 terminated list. */ + 0, + } /* 0 terminated list. */ }; MODULE_DEVICE_TABLE(pci, r82600_pci_tbl); @@ -389,7 +406,7 @@ module_exit(r82600_exit); MODULE_LICENSE("GPL"); MODULE_AUTHOR("Tim Small <tim@buttersideup.com> - WPAD Ltd. " - "on behalf of EADS Astrium"); + "on behalf of EADS Astrium"); MODULE_DESCRIPTION("MC support for Radisys 82600 memory controllers"); module_param(disable_hardware_scrub, bool, 0644); diff --git a/drivers/hwmon/Kconfig b/drivers/hwmon/Kconfig index 13eea47dceb..dbdca6f10e4 100644 --- a/drivers/hwmon/Kconfig +++ b/drivers/hwmon/Kconfig @@ -29,17 +29,34 @@ config HWMON_VID default n config SENSORS_ABITUGURU - tristate "Abit uGuru" + tristate "Abit uGuru (rev 1 & 2)" depends on EXPERIMENTAL help - If you say yes here you get support for the Abit uGuru chips - sensor part. The voltage and frequency control parts of the Abit - uGuru are not supported. The Abit uGuru chip can be found on Abit - uGuru featuring motherboards (most modern Abit motherboards). + If you say yes here you get support for the sensor part of the first + and second revision of the Abit uGuru chip. The voltage and frequency + control parts of the Abit uGuru are not supported. The Abit uGuru + chip can be found on Abit uGuru featuring motherboards (most modern + Abit motherboards from before end 2005). For more info and a list + of which motherboards have which revision see + Documentation/hwmon/abituguru This driver can also be built as a module. If so, the module will be called abituguru. +config SENSORS_ABITUGURU3 + tristate "Abit uGuru (rev 3)" + depends on HWMON && EXPERIMENTAL + help + If you say yes here you get support for the sensor part of the + third revision of the Abit uGuru chip. Only reading the sensors + and their settings is supported. The third revision of the Abit + uGuru chip can be found on recent Abit motherboards (since end + 2005). For more info and a list of which motherboards have which + revision see Documentation/hwmon/abituguru3 + + This driver can also be built as a module. If so, the module + will be called abituguru3. + config SENSORS_AD7418 tristate "Analog Devices AD7416, AD7417 and AD7418" depends on I2C && EXPERIMENTAL @@ -250,12 +267,10 @@ config SENSORS_CORETEMP config SENSORS_IT87 tristate "ITE IT87xx and compatibles" - depends on I2C - select I2C_ISA select HWMON_VID help If you say yes here you get support for ITE IT8705F, IT8712F, - IT8716F and IT8718F sensor chips, and the SiS960 clone. + IT8716F, IT8718F and IT8726F sensor chips, and the SiS960 clone. This driver can also be built as a module. If so, the module will be called it87. @@ -365,8 +380,8 @@ config SENSORS_LM90 depends on I2C help If you say yes here you get support for National Semiconductor LM90, - LM86, LM89 and LM99, Analog Devices ADM1032 and Maxim MAX6657 and - MAX6658 sensor chips. + LM86, LM89 and LM99, Analog Devices ADM1032 and Maxim MAX6657, + MAX6658, MAX6659, MAX6680 and MAX6681 sensor chips. The Analog Devices ADT7461 sensor chip is also supported, but only if found in ADM1032 compatibility mode. @@ -384,6 +399,17 @@ config SENSORS_LM92 This driver can also be built as a module. If so, the module will be called lm92. +config SENSORS_LM93 + tristate "National Semiconductor LM93 and compatibles" + depends on HWMON && I2C + select HWMON_VID + help + If you say yes here you get support for National Semiconductor LM93 + sensor chips. + + This driver can also be built as a module. If so, the module + will be called lm93. + config SENSORS_MAX1619 tristate "Maxim MAX1619 sensor chip" depends on I2C @@ -405,8 +431,6 @@ config SENSORS_MAX6650 config SENSORS_PC87360 tristate "National Semiconductor PC87360 family" - depends on I2C && EXPERIMENTAL - select I2C_ISA select HWMON_VID help If you say yes here you get access to the hardware monitoring @@ -433,8 +457,7 @@ config SENSORS_PC87427 config SENSORS_SIS5595 tristate "Silicon Integrated Systems Corp. SiS5595" - depends on I2C && PCI && EXPERIMENTAL - select I2C_ISA + depends on PCI help If you say yes here you get support for the integrated sensors in SiS5595 South Bridges. @@ -442,6 +465,18 @@ config SENSORS_SIS5595 This driver can also be built as a module. If so, the module will be called sis5595. +config SENSORS_DME1737 + tristate "SMSC DME1737 and compatibles" + depends on I2C && EXPERIMENTAL + select HWMON_VID + help + If you say yes here you get support for the hardware monitoring + and fan control features of the SMSC DME1737 (and compatibles + like the Asus A8000) Super-I/O chip. + + This driver can also be built as a module. If so, the module + will be called dme1737. + config SENSORS_SMSC47M1 tristate "SMSC LPC47M10x and compatibles" help @@ -487,8 +522,7 @@ config SENSORS_SMSC47B397 config SENSORS_VIA686A tristate "VIA686A" - depends on I2C && PCI - select I2C_ISA + depends on PCI help If you say yes here you get support for the integrated sensors in Via 686A/B South Bridges. @@ -509,9 +543,8 @@ config SENSORS_VT1211 config SENSORS_VT8231 tristate "VIA VT8231" - depends on I2C && PCI && EXPERIMENTAL + depends on PCI select HWMON_VID - select I2C_ISA help If you say yes here then you get support for the integrated sensors in the VIA VT8231 device. @@ -584,17 +617,16 @@ config SENSORS_W83627HF will be called w83627hf. config SENSORS_W83627EHF - tristate "Winbond W83627EHF" - depends on I2C && EXPERIMENTAL - select I2C_ISA + tristate "Winbond W83627EHF/DHG" + select HWMON_VID help - If you say yes here you get preliminary support for the hardware + If you say yes here you get support for the hardware monitoring functionality of the Winbond W83627EHF Super-I/O chip. - Only fan and temperature inputs are supported at the moment, while - the chip does much more than that. This driver also supports the W83627EHG, which is the lead-free - version of the W83627EHF. + version of the W83627EHF, and the W83627DHG, which is a similar + chip suited for specific Intel processors that use PECI such as + the Core 2 Duo. This driver can also be built as a module. If so, the module will be called w83627ehf. diff --git a/drivers/hwmon/Makefile b/drivers/hwmon/Makefile index cfaf338919d..59f81fae40a 100644 --- a/drivers/hwmon/Makefile +++ b/drivers/hwmon/Makefile @@ -14,6 +14,7 @@ obj-$(CONFIG_SENSORS_W83781D) += w83781d.o obj-$(CONFIG_SENSORS_W83791D) += w83791d.o obj-$(CONFIG_SENSORS_ABITUGURU) += abituguru.o +obj-$(CONFIG_SENSORS_ABITUGURU3)+= abituguru3.o obj-$(CONFIG_SENSORS_AD7418) += ad7418.o obj-$(CONFIG_SENSORS_ADM1021) += adm1021.o obj-$(CONFIG_SENSORS_ADM1025) += adm1025.o @@ -25,6 +26,7 @@ obj-$(CONFIG_SENSORS_APPLESMC) += applesmc.o obj-$(CONFIG_SENSORS_AMS) += ams/ obj-$(CONFIG_SENSORS_ATXP1) += atxp1.o obj-$(CONFIG_SENSORS_CORETEMP) += coretemp.o +obj-$(CONFIG_SENSORS_DME1737) += dme1737.o obj-$(CONFIG_SENSORS_DS1621) += ds1621.o obj-$(CONFIG_SENSORS_F71805F) += f71805f.o obj-$(CONFIG_SENSORS_FSCHER) += fscher.o @@ -45,6 +47,7 @@ obj-$(CONFIG_SENSORS_LM85) += lm85.o obj-$(CONFIG_SENSORS_LM87) += lm87.o obj-$(CONFIG_SENSORS_LM90) += lm90.o obj-$(CONFIG_SENSORS_LM92) += lm92.o +obj-$(CONFIG_SENSORS_LM93) += lm93.o obj-$(CONFIG_SENSORS_MAX1619) += max1619.o obj-$(CONFIG_SENSORS_MAX6650) += max6650.o obj-$(CONFIG_SENSORS_PC87360) += pc87360.o diff --git a/drivers/hwmon/abituguru.c b/drivers/hwmon/abituguru.c index bede4d990ea..d575ee958de 100644 --- a/drivers/hwmon/abituguru.c +++ b/drivers/hwmon/abituguru.c @@ -16,9 +16,9 @@ Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA. */ /* - This driver supports the sensor part of the custom Abit uGuru chip found - on Abit uGuru motherboards. Note: because of lack of specs the CPU / RAM / - etc voltage & frequency control is not supported! + This driver supports the sensor part of the first and second revision of + the custom Abit uGuru chip found on Abit uGuru motherboards. Note: because + of lack of specs the CPU/RAM voltage & frequency control is not supported! */ #include <linux/module.h> #include <linux/sched.h> @@ -31,6 +31,7 @@ #include <linux/platform_device.h> #include <linux/hwmon.h> #include <linux/hwmon-sysfs.h> +#include <linux/dmi.h> #include <asm/io.h> /* Banks */ @@ -418,7 +419,7 @@ static int __devinit abituguru_detect_bank1_sensor_type(struct abituguru_data *data, u8 sensor_addr) { - u8 val, buf[3]; + u8 val, test_flag, buf[3]; int i, ret = -ENODEV; /* error is the most common used retval :| */ /* If overriden by the user return the user selected type */ @@ -436,7 +437,7 @@ abituguru_detect_bank1_sensor_type(struct abituguru_data *data, return -ENODEV; /* Test val is sane / usable for sensor type detection. */ - if ((val < 10u) || (val > 240u)) { + if ((val < 10u) || (val > 250u)) { printk(KERN_WARNING ABIT_UGURU_NAME ": bank1-sensor: %d reading (%d) too close to limits, " "unable to determine sensor type, skipping sensor\n", @@ -449,10 +450,20 @@ abituguru_detect_bank1_sensor_type(struct abituguru_data *data, ABIT_UGURU_DEBUG(2, "testing bank1 sensor %d\n", (int)sensor_addr); /* Volt sensor test, enable volt low alarm, set min value ridicously - high. If its a volt sensor this should always give us an alarm. */ - buf[0] = ABIT_UGURU_VOLT_LOW_ALARM_ENABLE; - buf[1] = 245; - buf[2] = 250; + high, or vica versa if the reading is very high. If its a volt + sensor this should always give us an alarm. */ + if (val <= 240u) { + buf[0] = ABIT_UGURU_VOLT_LOW_ALARM_ENABLE; + buf[1] = 245; + buf[2] = 250; + test_flag = ABIT_UGURU_VOLT_LOW_ALARM_FLAG; + } else { + buf[0] = ABIT_UGURU_VOLT_HIGH_ALARM_ENABLE; + buf[1] = 5; + buf[2] = 10; + test_flag = ABIT_UGURU_VOLT_HIGH_ALARM_FLAG; + } + if (abituguru_write(data, ABIT_UGURU_SENSOR_BANK1 + 2, sensor_addr, buf, 3) != 3) goto abituguru_detect_bank1_sensor_type_exit; @@ -469,13 +480,13 @@ abituguru_detect_bank1_sensor_type(struct abituguru_data *data, sensor_addr, buf, 3, ABIT_UGURU_MAX_RETRIES) != 3) goto abituguru_detect_bank1_sensor_type_exit; - if (buf[0] & ABIT_UGURU_VOLT_LOW_ALARM_FLAG) { + if (buf[0] & test_flag) { ABIT_UGURU_DEBUG(2, " found volt sensor\n"); ret = ABIT_UGURU_IN_SENSOR; goto abituguru_detect_bank1_sensor_type_exit; } else ABIT_UGURU_DEBUG(2, " alarm raised during volt " - "sensor test, but volt low flag not set\n"); + "sensor test, but volt range flag not set\n"); } else ABIT_UGURU_DEBUG(2, " alarm not raised during volt sensor " "test\n"); @@ -1287,6 +1298,7 @@ abituguru_probe_error: for (i = 0; i < ARRAY_SIZE(abituguru_sysfs_attr); i++) device_remove_file(&pdev->dev, &abituguru_sysfs_attr[i].dev_attr); + platform_set_drvdata(pdev, NULL); kfree(data); return res; } @@ -1296,13 +1308,13 @@ static int __devexit abituguru_remove(struct platform_device *pdev) int i; struct abituguru_data *data = platform_get_drvdata(pdev); - platform_set_drvdata(pdev, NULL); hwmon_device_unregister(data->class_dev); for (i = 0; data->sysfs_attr[i].dev_attr.attr.name; i++) device_remove_file(&pdev->dev, &data->sysfs_attr[i].dev_attr); for (i = 0; i < ARRAY_SIZE(abituguru_sysfs_attr); i++) device_remove_file(&pdev->dev, &abituguru_sysfs_attr[i].dev_attr); + platform_set_drvdata(pdev, NULL); kfree(data); return 0; @@ -1436,6 +1448,15 @@ static int __init abituguru_init(void) int address, err; struct resource res = { .flags = IORESOURCE_IO }; +#ifdef CONFIG_DMI + char *board_vendor = dmi_get_system_info(DMI_BOARD_VENDOR); + + /* safety check, refuse to load on non Abit motherboards */ + if (!force && (!board_vendor || + strcmp(board_vendor, "http://www.abit.com.tw/"))) + return -ENODEV; +#endif + address = abituguru_detect(); if (address < 0) return address; diff --git a/drivers/hwmon/abituguru3.c b/drivers/hwmon/abituguru3.c new file mode 100644 index 00000000000..a003d104ca4 --- /dev/null +++ b/drivers/hwmon/abituguru3.c @@ -0,0 +1,1140 @@ +/* + abituguru3.c Copyright (c) 2006 Hans de Goede <j.w.r.degoede@hhs.nl> + + 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. + + You should have received a copy of the GNU General Public License + along with this program; if not, write to the Free Software + Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA. +*/ +/* + This driver supports the sensor part of revision 3 of the custom Abit uGuru + chip found on newer Abit uGuru motherboards. Note: because of lack of specs + only reading the sensors and their settings is supported. +*/ +#include <linux/module.h> +#include <linux/init.h> +#include <linux/slab.h> +#include <linux/jiffies.h> +#include <linux/mutex.h> +#include <linux/err.h> +#include <linux/delay.h> +#include <linux/platform_device.h> +#include <linux/hwmon.h> +#include <linux/hwmon-sysfs.h> +#include <asm/io.h> + +/* uGuru3 bank addresses */ +#define ABIT_UGURU3_SETTINGS_BANK 0x01 +#define ABIT_UGURU3_SENSORS_BANK 0x08 +#define ABIT_UGURU3_MISC_BANK 0x09 +#define ABIT_UGURU3_ALARMS_START 0x1E +#define ABIT_UGURU3_SETTINGS_START 0x24 +#define ABIT_UGURU3_VALUES_START 0x80 +#define ABIT_UGURU3_BOARD_ID 0x0A +/* uGuru3 sensor bank flags */ /* Alarm if: */ +#define ABIT_UGURU3_TEMP_HIGH_ALARM_ENABLE 0x01 /* temp over warn */ +#define ABIT_UGURU3_VOLT_HIGH_ALARM_ENABLE 0x02 /* volt over max */ +#define ABIT_UGURU3_VOLT_LOW_ALARM_ENABLE 0x04 /* volt under min */ +#define ABIT_UGURU3_TEMP_HIGH_ALARM_FLAG 0x10 /* temp is over warn */ +#define ABIT_UGURU3_VOLT_HIGH_ALARM_FLAG 0x20 /* volt is over max */ +#define ABIT_UGURU3_VOLT_LOW_ALARM_FLAG 0x40 /* volt is under min */ +#define ABIT_UGURU3_FAN_LOW_ALARM_ENABLE 0x01 /* fan under min */ +#define ABIT_UGURU3_BEEP_ENABLE 0x08 /* beep if alarm */ +#define ABIT_UGURU3_SHUTDOWN_ENABLE 0x80 /* shutdown if alarm */ +/* sensor types */ +#define ABIT_UGURU3_IN_SENSOR 0 +#define ABIT_UGURU3_TEMP_SENSOR 1 +#define ABIT_UGURU3_FAN_SENSOR 2 + +/* Timeouts / Retries, if these turn out to need a lot of fiddling we could + convert them to params. Determined by trial and error. I assume this is + cpu-speed independent, since the ISA-bus and not the CPU should be the + bottleneck. */ +#define ABIT_UGURU3_WAIT_TIMEOUT 250 +/* Normally the 0xAC at the end of synchronize() is reported after the + first read, but sometimes not and we need to poll */ +#define ABIT_UGURU3_SYNCHRONIZE_TIMEOUT 5 +/* utility macros */ +#define ABIT_UGURU3_NAME "abituguru3" +#define ABIT_UGURU3_DEBUG(format, arg...) \ + if (verbose) \ + printk(KERN_DEBUG ABIT_UGURU3_NAME ": " format , ## arg) + +/* Macros to help calculate the sysfs_names array length */ +#define ABIT_UGURU3_MAX_NO_SENSORS 26 +/* sum of strlen +1 of: in??_input\0, in??_{min,max}\0, in??_{min,max}_alarm\0, + in??_{min,max}_alarm_enable\0, in??_beep\0, in??_shutdown\0, in??_label\0 */ +#define ABIT_UGURU3_IN_NAMES_LENGTH (11 + 2 * 9 + 2 * 15 + 2 * 22 + 10 + 14 + 11) +/* sum of strlen +1 of: temp??_input\0, temp??_max\0, temp??_crit\0, + temp??_alarm\0, temp??_alarm_enable\0, temp??_beep\0, temp??_shutdown\0, + temp??_label\0 */ +#define ABIT_UGURU3_TEMP_NAMES_LENGTH (13 + 11 + 12 + 13 + 20 + 12 + 16 + 13) +/* sum of strlen +1 of: fan??_input\0, fan??_min\0, fan??_alarm\0, + fan??_alarm_enable\0, fan??_beep\0, fan??_shutdown\0, fan??_label\0 */ +#define ABIT_UGURU3_FAN_NAMES_LENGTH (12 + 10 + 12 + 19 + 11 + 15 + 12) +/* Worst case scenario 16 in sensors (longest names_length) and the rest + temp sensors (second longest names_length). */ +#define ABIT_UGURU3_SYSFS_NAMES_LENGTH (16 * ABIT_UGURU3_IN_NAMES_LENGTH + \ + (ABIT_UGURU3_MAX_NO_SENSORS - 16) * ABIT_UGURU3_TEMP_NAMES_LENGTH) + +/* All the macros below are named identical to the openguru2 program + reverse engineered by Louis Kruger, hence the names might not be 100% + logical. I could come up with better names, but I prefer keeping the names + identical so that this driver can be compared with his work more easily. */ +/* Two i/o-ports are used by uGuru */ +#define ABIT_UGURU3_BASE 0x00E0 +#define ABIT_UGURU3_CMD 0x00 +#define ABIT_UGURU3_DATA 0x04 +#define ABIT_UGURU3_REGION_LENGTH 5 +/* The wait_xxx functions return this on success and the last contents + of the DATA register (0-255) on failure. */ +#define ABIT_UGURU3_SUCCESS -1 +/* uGuru status flags */ +#define ABIT_UGURU3_STATUS_READY_FOR_READ 0x01 +#define ABIT_UGURU3_STATUS_BUSY 0x02 + + +/* Structures */ +struct abituguru3_sensor_info { + const char* name; + int port; + int type; + int multiplier; + int divisor; + int offset; +}; + +struct abituguru3_motherboard_info { + u16 id; + const char *name; + /* + 1 -> end of sensors indicated by a sensor with name == NULL */ + struct abituguru3_sensor_info sensors[ABIT_UGURU3_MAX_NO_SENSORS + 1]; +}; + +/* For the Abit uGuru, we need to keep some data in memory. + The structure is dynamically allocated, at the same time when a new + abituguru3 device is allocated. */ +struct abituguru3_data { + struct class_device *class_dev; /* hwmon registered device */ + struct mutex update_lock; /* protect access to data and uGuru */ + unsigned short addr; /* uguru base address */ + char valid; /* !=0 if following fields are valid */ + unsigned long last_updated; /* In jiffies */ + + /* For convenience the sysfs attr and their names are generated + automatically. We have max 10 entries per sensor (for in sensors) */ + struct sensor_device_attribute_2 sysfs_attr[ABIT_UGURU3_MAX_NO_SENSORS + * 10]; + + /* Buffer to store the dynamically generated sysfs names */ + char sysfs_names[ABIT_UGURU3_SYSFS_NAMES_LENGTH]; + + /* Pointer to the sensors info for the detected motherboard */ + const struct abituguru3_sensor_info *sensors; + + /* The abituguru3 supports upto 48 sensors, and thus has registers + sets for 48 sensors, for convienence reasons / simplicity of the + code we always read and store all registers for all 48 sensors */ + + /* Alarms for all 48 sensors (1 bit per sensor) */ + u8 alarms[48/8]; + + /* Value of all 48 sensors */ + u8 value[48]; + + /* Settings of all 48 sensors, note in and temp sensors (the first 32 + sensors) have 3 bytes of settings, while fans only have 2 bytes, + for convenience we use 3 bytes for all sensors */ + u8 settings[48][3]; +}; + + +/* Constants */ +static const struct abituguru3_motherboard_info abituguru3_motherboards[] = { + { 0x000C, "unknown", { + { "CPU Core", 0, 0, 10, 1, 0 }, + { "DDR", 1, 0, 10, 1, 0 }, + { "DDR VTT", 2, 0, 10, 1, 0 }, + { "CPU VTT 1.2V", 3, 0, 10, 1, 0 }, + { "MCH & PCIE 1.5V", 4, 0, 10, 1, 0 }, + { "MCH 2.5V", 5, 0, 20, 1, 0 }, + { "ICH 1.05V", 6, 0, 10, 1, 0 }, + { "ATX +12V (24-Pin)", 7, 0, 60, 1, 0 }, + { "ATX +12V (4-pin)", 8, 0, 60, 1, 0 }, + { "ATX +5V", 9, 0, 30, 1, 0 }, + { "+3.3V", 10, 0, 20, 1, 0 }, + { "5VSB", 11, 0, 30, 1, 0 }, + { "CPU", 24, 1, 1, 1, 0 }, + { "System ", 25, 1, 1, 1, 0 }, + { "PWM", 26, 1, 1, 1, 0 }, + { "CPU Fan", 32, 2, 60, 1, 0 }, + { "NB Fan", 33, 2, 60, 1, 0 }, + { "SYS FAN", 34, 2, 60, 1, 0 }, + { "AUX1 Fan", 35, 2, 60, 1, 0 }, + { NULL, 0, 0, 0, 0, 0 } } + }, + { 0x000D, "Abit AW8", { + { "CPU Core", 0, 0, 10, 1, 0 }, + { "DDR", 1, 0, 10, 1, 0 }, + { "DDR VTT", 2, 0, 10, 1, 0 }, + { "CPU VTT 1.2V", 3, 0, 10, 1, 0 }, + { "MCH & PCIE 1.5V", 4, 0, 10, 1, 0 }, + { "MCH 2.5V", 5, 0, 20, 1, 0 }, + { "ICH 1.05V", 6, 0, 10, 1, 0 }, + { "ATX +12V (24-Pin)", 7, 0, 60, 1, 0 }, + { "ATX +12V (4-pin)", 8, 0, 60, 1, 0 }, + { "ATX +5V", 9, 0, 30, 1, 0 }, + { "+3.3V", 10, 0, 20, 1, 0 }, + { "5VSB", 11, 0, 30, 1, 0 }, + { "CPU", 24, 1, 1, 1, 0 }, + { "System ", 25, 1, 1, 1, 0 }, + { "PWM1", 26, 1, 1, 1, 0 }, + { "PWM2", 27, 1, 1, 1, 0 }, + { "PWM3", 28, 1, 1, 1, 0 }, + { "PWM4", 29, 1, 1, 1, 0 }, + { "CPU Fan", 32, 2, 60, 1, 0 }, + { "NB Fan", 33, 2, 60, 1, 0 }, + { "SYS Fan", 34, 2, 60, 1, 0 }, + { "AUX1 Fan", 35, 2, 60, 1, 0 }, + { "AUX2 Fan", 36, 2, 60, 1, 0 }, + { "AUX3 Fan", 37, 2, 60, 1, 0 }, + { "AUX4 Fan", 38, 2, 60, 1, 0 }, + { "AUX5 Fan", 39, 2, 60, 1, 0 }, + { NULL, 0, 0, 0, 0, 0 } } + }, + { 0x000E, "AL-8", { + { "CPU Core", 0, 0, 10, 1, 0 }, + { "DDR", 1, 0, 10, 1, 0 }, + { "DDR VTT", 2, 0, 10, 1, 0 }, + { "CPU VTT 1.2V", 3, 0, 10, 1, 0 }, + { "MCH & PCIE 1.5V", 4, 0, 10, 1, 0 }, + { "MCH 2.5V", 5, 0, 20, 1, 0 }, + { "ICH 1.05V", 6, 0, 10, 1, 0 }, + { "ATX +12V (24-Pin)", 7, 0, 60, 1, 0 }, + { "ATX +12V (4-pin)", 8, 0, 60, 1, 0 }, + { "ATX +5V", 9, 0, 30, 1, 0 }, + { "+3.3V", 10, 0, 20, 1, 0 }, + { "5VSB", 11, 0, 30, 1, 0 }, + { "CPU", 24, 1, 1, 1, 0 }, + { "System ", 25, 1, 1, 1, 0 }, + { "PWM", 26, 1, 1, 1, 0 }, + { "CPU Fan", 32, 2, 60, 1, 0 }, + { "NB Fan", 33, 2, 60, 1, 0 }, + { "SYS Fan", 34, 2, 60, 1, 0 }, + { NULL, 0, 0, 0, 0, 0 } } + }, + { 0x000F, "unknown", { + { "CPU Core", 0, 0, 10, 1, 0 }, + { "DDR", 1, 0, 10, 1, 0 }, + { "DDR VTT", 2, 0, 10, 1, 0 }, + { "CPU VTT 1.2V", 3, 0, 10, 1, 0 }, + { "MCH & PCIE 1.5V", 4, 0, 10, 1, 0 }, + { "MCH 2.5V", 5, 0, 20, 1, 0 }, + { "ICH 1.05V", 6, 0, 10, 1, 0 }, + { "ATX +12V (24-Pin)", 7, 0, 60, 1, 0 }, + { "ATX +12V (4-pin)", 8, 0, 60, 1, 0 }, + { "ATX +5V", 9, 0, 30, 1, 0 }, + { "+3.3V", 10, 0, 20, 1, 0 }, + { "5VSB", 11, 0, 30, 1, 0 }, + { "CPU", 24, 1, 1, 1, 0 }, + { "System ", 25, 1, 1, 1, 0 }, + { "PWM", 26, 1, 1, 1, 0 }, + { "CPU Fan", 32, 2, 60, 1, 0 }, + { "NB Fan", 33, 2, 60, 1, 0 }, + { "SYS Fan", 34, 2, 60, 1, 0 }, + { NULL, 0, 0, 0, 0, 0 } } + }, + { 0x0010, "Abit NI8 SLI GR", { + { "CPU Core", 0, 0, 10, 1, 0 }, + { "DDR", 1, 0, 10, 1, 0 }, + { "DDR VTT", 2, 0, 10, 1, 0 }, + { "CPU VTT 1.2V", 3, 0, 10, 1, 0 }, + { "NB 1.4V", 4, 0, 10, 1, 0 }, + { "SB 1.5V", 6, 0, 10, 1, 0 }, + { "ATX +12V (24-Pin)", 7, 0, 60, 1, 0 }, + { "ATX +12V (4-pin)", 8, 0, 60, 1, 0 }, + { "ATX +5V", 9, 0, 30, 1, 0 }, + { "+3.3V", 10, 0, 20, 1, 0 }, + { "5VSB", 11, 0, 30, 1, 0 }, + { "CPU", 24, 1, 1, 1, 0 }, + { "SYS", 25, 1, 1, 1, 0 }, + { "PWM", 26, 1, 1, 1, 0 }, + { "CPU Fan", 32, 2, 60, 1, 0 }, + { "NB Fan", 33, 2, 60, 1, 0 }, + { "SYS Fan", 34, 2, 60, 1, 0 }, + { "AUX1 Fan", 35, 2, 60, 1, 0 }, + { "OTES1 Fan", 36, 2, 60, 1, 0 }, + { NULL, 0, 0, 0, 0, 0 } } + }, + { 0x0011, "Abit AT8 32X", { + { "CPU Core", 0, 0, 10, 1, 0 }, + { "DDR", 1, 0, 20, 1, 0 }, + { "DDR VTT", 2, 0, 10, 1, 0 }, + { "CPU VDDA 2.5V", 6, 0, 20, 1, 0 }, + { "NB 1.8V", 4, 0, 10, 1, 0 }, + { "NB 1.8V Dual", 5, 0, 10, 1, 0 }, + { "HTV 1.2", 3, 0, 10, 1, 0 }, + { "PCIE 1.2V", 12, 0, 10, 1, 0 }, + { "NB 1.2V", 13, 0, 10, 1, 0 }, + { "ATX +12V (24-Pin)", 7, 0, 60, 1, 0 }, + { "ATX +12V (4-pin)", 8, 0, 60, 1, 0 }, + { "ATX +5V", 9, 0, 30, 1, 0 }, + { "+3.3V", 10, 0, 20, 1, 0 }, + { "5VSB", 11, 0, 30, 1, 0 }, + { "CPU", 24, 1, 1, 1, 0 }, + { "NB", 25, 1, 1, 1, 0 }, + { "System", 26, 1, 1, 1, 0 }, + { "PWM", 27, 1, 1, 1, 0 }, + { "CPU Fan", 32, 2, 60, 1, 0 }, + { "NB Fan", 33, 2, 60, 1, 0 }, + { "SYS Fan", 34, 2, 60, 1, 0 }, + { "AUX1 Fan", 35, 2, 60, 1, 0 }, + { "AUX2 Fan", 36, 2, 60, 1, 0 }, + { NULL, 0, 0, 0, 0, 0 } } + }, + { 0x0012, "Abit AN8 32X", { + { "CPU Core", 0, 0, 10, 1, 0 }, + { "DDR", 1, 0, 20, 1, 0 }, + { "DDR VTT", 2, 0, 10, 1, 0 }, + { "HyperTransport", 3, 0, 10, 1, 0 }, + { "CPU VDDA 2.5V", 5, 0, 20, 1, 0 }, + { "NB", 4, 0, 10, 1, 0 }, + { "SB", 6, 0, 10, 1, 0 }, + { "ATX +12V (24-Pin)", 7, 0, 60, 1, 0 }, + { "ATX +12V (4-pin)", 8, 0, 60, 1, 0 }, + { "ATX +5V", 9, 0, 30, 1, 0 }, + { "+3.3V", 10, 0, 20, 1, 0 }, + { "5VSB", 11, 0, 30, 1, 0 }, + { "CPU", 24, 1, 1, 1, 0 }, + { "SYS", 25, 1, 1, 1, 0 }, + { "PWM", 26, 1, 1, 1, 0 }, + { "CPU Fan", 32, 2, 60, 1, 0 }, + { "NB Fan", 33, 2, 60, 1, 0 }, + { "SYS Fan", 34, 2, 60, 1, 0 }, + { "AUX1 Fan", 36, 2, 60, 1, 0 }, + { NULL, 0, 0, 0, 0, 0 } } + }, + { 0x0013, "unknown", { + { "CPU Core", 0, 0, 10, 1, 0 }, + { "DDR", 1, 0, 10, 1, 0 }, + { "DDR VTT", 2, 0, 10, 1, 0 }, + { "CPU VTT 1.2V", 3, 0, 10, 1, 0 }, + { "MCH & PCIE 1.5V", 4, 0, 10, 1, 0 }, + { "MCH 2.5V", 5, 0, 20, 1, 0 }, + { "ICH 1.05V", 6, 0, 10, 1, 0 }, + { "ATX +12V (24-Pin)", 7, 0, 60, 1, 0 }, + { "ATX +12V (4-pin)", 8, 0, 60, 1, 0 }, + { "ATX +5V", 9, 0, 30, 1, 0 }, + { "+3.3V", 10, 0, 20, 1, 0 }, + { "5VSB", 11, 0, 30, 1, 0 }, + { "CPU", 24, 1, 1, 1, 0 }, + { "System ", 25, 1, 1, 1, 0 }, + { "PWM1", 26, 1, 1, 1, 0 }, + { "PWM2", 27, 1, 1, 1, 0 }, + { "PWM3", 28, 1, 1, 1, 0 }, + { "PWM4", 29, 1, 1, 1, 0 }, + { "CPU Fan", 32, 2, 60, 1, 0 }, + { "NB Fan", 33, 2, 60, 1, 0 }, + { "SYS Fan", 34, 2, 60, 1, 0 }, + { "AUX1 Fan", 35, 2, 60, 1, 0 }, + { "AUX2 Fan", 36, 2, 60, 1, 0 }, + { "AUX3 Fan", 37, 2, 60, 1, 0 }, + { "AUX4 Fan", 38, 2, 60, 1, 0 }, + { NULL, 0, 0, 0, 0, 0 } } + }, + { 0x0014, "Abit AB9 Pro", { + { "CPU Core", 0, 0, 10, 1, 0 }, + { "DDR", 1, 0, 10, 1, 0 }, + { "DDR VTT", 2, 0, 10, 1, 0 }, + { "CPU VTT 1.2V", 3, 0, 10, 1, 0 }, + { "MCH & PCIE 1.5V", 4, 0, 10, 1, 0 }, + { "MCH 2.5V", 5, 0, 20, 1, 0 }, + { "ICH 1.05V", 6, 0, 10, 1, 0 }, + { "ATX +12V (24-Pin)", 7, 0, 60, 1, 0 }, + { "ATX +12V (4-pin)", 8, 0, 60, 1, 0 }, + { "ATX +5V", 9, 0, 30, 1, 0 }, + { "+3.3V", 10, 0, 20, 1, 0 }, + { "5VSB", 11, 0, 30, 1, 0 }, + { "CPU", 24, 1, 1, 1, 0 }, + { "System ", 25, 1, 1, 1, 0 }, + { "PWM", 26, 1, 1, 1, 0 }, + { "CPU Fan", 32, 2, 60, 1, 0 }, + { "NB Fan", 33, 2, 60, 1, 0 }, + { "SYS Fan", 34, 2, 60, 1, 0 }, + { NULL, 0, 0, 0, 0, 0 } } + }, + { 0x0015, "unknown", { + { "CPU Core", 0, 0, 10, 1, 0 }, + { "DDR", 1, 0, 20, 1, 0 }, + { "DDR VTT", 2, 0, 10, 1, 0 }, + { "HyperTransport", 3, 0, 10, 1, 0 }, + { "CPU VDDA 2.5V", 5, 0, 20, 1, 0 }, + { "NB", 4, 0, 10, 1, 0 }, + { "SB", 6, 0, 10, 1, 0 }, + { "ATX +12V (24-Pin)", 7, 0, 60, 1, 0 }, + { "ATX +12V (4-pin)", 8, 0, 60, 1, 0 }, + { "ATX +5V", 9, 0, 30, 1, 0 }, + { "+3.3V", 10, 0, 20, 1, 0 }, + { "5VSB", 11, 0, 30, 1, 0 }, + { "CPU", 24, 1, 1, 1, 0 }, + { "SYS", 25, 1, 1, 1, 0 }, + { "PWM", 26, 1, 1, 1, 0 }, + { "CPU Fan", 32, 2, 60, 1, 0 }, + { "NB Fan", 33, 2, 60, 1, 0 }, + { "SYS Fan", 34, 2, 60, 1, 0 }, + { "AUX1 Fan", 33, 2, 60, 1, 0 }, + { "AUX2 Fan", 35, 2, 60, 1, 0 }, + { "AUX3 Fan", 36, 2, 60, 1, 0 }, + { NULL, 0, 0, 0, 0, 0 } } + }, + { 0x0016, "AW9D-MAX", { + { "CPU Core", 0, 0, 10, 1, 0 }, + { "DDR2", 1, 0, 20, 1, 0 }, + { "DDR2 VTT", 2, 0, 10, 1, 0 }, + { "CPU VTT 1.2V", 3, 0, 10, 1, 0 }, + { "MCH & PCIE 1.5V", 4, 0, 10, 1, 0 }, + { "MCH 2.5V", 5, 0, 20, 1, 0 }, + { "ICH 1.05V", 6, 0, 10, 1, 0 }, + { "ATX +12V (24-Pin)", 7, 0, 60, 1, 0 }, + { "ATX +12V (4-pin)", 8, 0, 60, 1, 0 }, + { "ATX +5V", 9, 0, 30, 1, 0 }, + { "+3.3V", 10, 0, 20, 1, 0 }, + { "5VSB", 11, 0, 30, 1, 0 }, + { "CPU", 24, 1, 1, 1, 0 }, + { "System ", 25, 1, 1, 1, 0 }, + { "PWM1", 26, 1, 1, 1, 0 }, + { "PWM2", 27, 1, 1, 1, 0 }, + { "PWM3", 28, 1, 1, 1, 0 }, + { "PWM4", 29, 1, 1, 1, 0 }, + { "CPU Fan", 32, 2, 60, 1, 0 }, + { "NB Fan", 33, 2, 60, 1, 0 }, + { "SYS Fan", 34, 2, 60, 1, 0 }, + { "AUX1 Fan", 35, 2, 60, 1, 0 }, + { "AUX2 Fan", 36, 2, 60, 1, 0 }, + { "AUX3 Fan", 37, 2, 60, 1, 0 }, + { "OTES1 Fan", 38, 2, 60, 1, 0 }, + { NULL, 0, 0, 0, 0, 0 } } + }, + { 0x0017, "unknown", { + { "CPU Core", 0, 0, 10, 1, 0 }, + { "DDR2", 1, 0, 20, 1, 0 }, + { "DDR2 VTT", 2, 0, 10, 1, 0 }, + { "HyperTransport", 3, 0, 10, 1, 0 }, + { "CPU VDDA 2.5V", 6, 0, 20, 1, 0 }, + { "NB 1.8V", 4, 0, 10, 1, 0 }, + { "NB 1.2V ", 13, 0, 10, 1, 0 }, + { "SB 1.2V", 5, 0, 10, 1, 0 }, + { "PCIE 1.2V", 12, 0, 10, 1, 0 }, + { "ATX +12V (24-Pin)", 7, 0, 60, 1, 0 }, + { "ATX +12V (4-pin)", 8, 0, 60, 1, 0 }, + { "ATX +5V", 9, 0, 30, 1, 0 }, + { "ATX +3.3V", 10, 0, 20, 1, 0 }, + { "ATX 5VSB", 11, 0, 30, 1, 0 }, + { "CPU", 24, 1, 1, 1, 0 }, + { "System ", 26, 1, 1, 1, 0 }, + { "PWM", 27, 1, 1, 1, 0 }, + { "CPU FAN", 32, 2, 60, 1, 0 }, + { "SYS FAN", 34, 2, 60, 1, 0 }, + { "AUX1 FAN", 35, 2, 60, 1, 0 }, + { "AUX2 FAN", 36, 2, 60, 1, 0 }, + { "AUX3 FAN", 37, 2, 60, 1, 0 }, + { NULL, 0, 0, 0, 0, 0 } } + }, + { 0x0018, "unknown", { + { "CPU Core", 0, 0, 10, 1, 0 }, + { "DDR2", 1, 0, 20, 1, 0 }, + { "DDR2 VTT", 2, 0, 10, 1, 0 }, + { "CPU VTT", 3, 0, 10, 1, 0 }, + { "MCH 1.25V", 4, 0, 10, 1, 0 }, + { "ICHIO 1.5V", 5, 0, 10, 1, 0 }, + { "ICH 1.05V", 6, 0, 10, 1, 0 }, + { "ATX +12V (24-Pin)", 7, 0, 60, 1, 0 }, + { "ATX +12V (4-pin)", 8, 0, 60, 1, 0 }, + { "ATX +5V", 9, 0, 30, 1, 0 }, + { "+3.3V", 10, 0, 20, 1, 0 }, + { "5VSB", 11, 0, 30, 1, 0 }, + { "CPU", 24, 1, 1, 1, 0 }, + { "System ", 25, 1, 1, 1, 0 }, + { "PWM Phase1", 26, 1, 1, 1, 0 }, + { "PWM Phase2", 27, 1, 1, 1, 0 }, + { "PWM Phase3", 28, 1, 1, 1, 0 }, + { "PWM Phase4", 29, 1, 1, 1, 0 }, + { "PWM Phase5", 30, 1, 1, 1, 0 }, + { "CPU Fan", 32, 2, 60, 1, 0 }, + { "SYS Fan", 34, 2, 60, 1, 0 }, + { "AUX1 Fan", 33, 2, 60, 1, 0 }, + { "AUX2 Fan", 35, 2, 60, 1, 0 }, + { "AUX3 Fan", 36, 2, 60, 1, 0 }, + { NULL, 0, 0, 0, 0, 0 } } + }, + { 0x0019, "unknown", { + { "CPU Core", 7, 0, 10, 1, 0 }, + { "DDR2", 13, 0, 20, 1, 0 }, + { "DDR2 VTT", 14, 0, 10, 1, 0 }, + { "CPU VTT", 3, 0, 20, 1, 0 }, + { "NB 1.2V ", 4, 0, 10, 1, 0 }, + { "SB 1.5V", 6, 0, 10, 1, 0 }, + { "HyperTransport", 5, 0, 10, 1, 0 }, + { "ATX +12V (24-Pin)", 12, 0, 60, 1, 0 }, + { "ATX +12V (4-pin)", 8, 0, 60, 1, 0 }, + { "ATX +5V", 9, 0, 30, 1, 0 }, + { "ATX +3.3V", 10, 0, 20, 1, 0 }, + { "ATX 5VSB", 11, 0, 30, 1, 0 }, + { "CPU", 24, 1, 1, 1, 0 }, + { "System ", 25, 1, 1, 1, 0 }, + { "PWM Phase1", 26, 1, 1, 1, 0 }, + { "PWM Phase2", 27, 1, 1, 1, 0 }, + { "PWM Phase3", 28, 1, 1, 1, 0 }, + { "PWM Phase4", 29, 1, 1, 1, 0 }, + { "PWM Phase5", 30, 1, 1, 1, 0 }, + { "CPU FAN", 32, 2, 60, 1, 0 }, + { "SYS FAN", 34, 2, 60, 1, 0 }, + { "AUX1 FAN", 33, 2, 60, 1, 0 }, + { "AUX2 FAN", 35, 2, 60, 1, 0 }, + { "AUX3 FAN", 36, 2, 60, 1, 0 }, + { NULL, 0, 0, 0, 0, 0 } } + }, + { 0x001A, "unknown", { + { "CPU Core", 0, 0, 10, 1, 0 }, + { "DDR2", 1, 0, 20, 1, 0 }, + { "DDR2 VTT", 2, 0, 10, 1, 0 }, + { "CPU VTT 1.2V", 3, 0, 10, 1, 0 }, + { "MCH 1.25V", 4, 0, 10, 1, 0 }, + { "ICHIO 1.5V", 5, 0, 10, 1, 0 }, + { "ICH 1.05V", 6, 0, 10, 1, 0 }, + { "ATX +12V (24-Pin)", 7, 0, 60, 1, 0 }, + { "ATX +12V (8-pin)", 8, 0, 60, 1, 0 }, + { "ATX +5V", 9, 0, 30, 1, 0 }, + { "+3.3V", 10, 0, 20, 1, 0 }, + { "5VSB", 11, 0, 30, 1, 0 }, + { "CPU", 24, 1, 1, 1, 0 }, + { "System ", 25, 1, 1, 1, 0 }, + { "PWM ", 26, 1, 1, 1, 0 }, + { "PWM Phase2", 27, 1, 1, 1, 0 }, + { "PWM Phase3", 28, 1, 1, 1, 0 }, + { "PWM Phase4", 29, 1, 1, 1, 0 }, + { "PWM Phase5", 30, 1, 1, 1, 0 }, + { "CPU Fan", 32, 2, 60, 1, 0 }, + { "SYS Fan", 34, 2, 60, 1, 0 }, + { "AUX1 Fan", 33, 2, 60, 1, 0 }, + { "AUX2 Fan", 35, 2, 60, 1, 0 }, + { "AUX3 Fan", 36, 2, 60, 1, 0 }, + { NULL, 0, 0, 0, 0, 0 } } + }, + { 0x0000, NULL, { { NULL, 0, 0, 0, 0, 0 } } } +}; + + +/* Insmod parameters */ +static int force; +module_param(force, bool, 0); +MODULE_PARM_DESC(force, "Set to one to force detection."); +/* Default verbose is 1, since this driver is still in the testing phase */ +static int verbose = 1; +module_param(verbose, bool, 0644); +MODULE_PARM_DESC(verbose, "Enable/disable verbose error reporting"); + + +/* wait while the uguru is busy (usually after a write) */ +static int abituguru3_wait_while_busy(struct abituguru3_data *data) +{ + u8 x; + int timeout = ABIT_UGURU3_WAIT_TIMEOUT; + + while ((x = inb_p(data->addr + ABIT_UGURU3_DATA)) & + ABIT_UGURU3_STATUS_BUSY) { + timeout--; + if (timeout == 0) + return x; + /* sleep a bit before our last try, to give the uGuru3 one + last chance to respond. */ + if (timeout == 1) + msleep(1); + } + return ABIT_UGURU3_SUCCESS; +} + +/* wait till uguru is ready to be read */ +static int abituguru3_wait_for_read(struct abituguru3_data *data) +{ + u8 x; + int timeout = ABIT_UGURU3_WAIT_TIMEOUT; + + while (!((x = inb_p(data->addr + ABIT_UGURU3_DATA)) & + ABIT_UGURU3_STATUS_READY_FOR_READ)) { + timeout--; + if (timeout == 0) + return x; + /* sleep a bit before our last try, to give the uGuru3 one + last chance to respond. */ + if (timeout == 1) + msleep(1); + } + return ABIT_UGURU3_SUCCESS; +} + +/* This synchronizes us with the uGuru3's protocol state machine, this + must be done before each command. */ +static int abituguru3_synchronize(struct abituguru3_data *data) +{ + int x, timeout = ABIT_UGURU3_SYNCHRONIZE_TIMEOUT; + + if ((x = abituguru3_wait_while_busy(data)) != ABIT_UGURU3_SUCCESS) { + ABIT_UGURU3_DEBUG("synchronize timeout during initial busy " + "wait, status: 0x%02x\n", x); + return -EIO; + } + + outb(0x20, data->addr + ABIT_UGURU3_DATA); + if ((x = abituguru3_wait_while_busy(data)) != ABIT_UGURU3_SUCCESS) { + ABIT_UGURU3_DEBUG("synchronize timeout after sending 0x20, " + "status: 0x%02x\n", x); + return -EIO; + } + + outb(0x10, data->addr + ABIT_UGURU3_CMD); + if ((x = abituguru3_wait_while_busy(data)) != ABIT_UGURU3_SUCCESS) { + ABIT_UGURU3_DEBUG("synchronize timeout after sending 0x10, " + "status: 0x%02x\n", x); + return -EIO; + } + + outb(0x00, data->addr + ABIT_UGURU3_CMD); + if ((x = abituguru3_wait_while_busy(data)) != ABIT_UGURU3_SUCCESS) { + ABIT_UGURU3_DEBUG("synchronize timeout after sending 0x00, " + "status: 0x%02x\n", x); + return -EIO; + } + + if ((x = abituguru3_wait_for_read(data)) != ABIT_UGURU3_SUCCESS) { + ABIT_UGURU3_DEBUG("synchronize timeout waiting for read, " + "status: 0x%02x\n", x); + return -EIO; + } + + while ((x = inb(data->addr + ABIT_UGURU3_CMD)) != 0xAC) { + timeout--; + if (timeout == 0) { + ABIT_UGURU3_DEBUG("synchronize timeout cmd does not " + "hold 0xAC after synchronize, cmd: 0x%02x\n", + x); + return -EIO; + } + msleep(1); + } + return 0; +} + +/* Read count bytes from sensor sensor_addr in bank bank_addr and store the + result in buf */ +static int abituguru3_read(struct abituguru3_data *data, u8 bank, u8 offset, + u8 count, u8 *buf) +{ + int i, x; + + if ((x = abituguru3_synchronize(data))) + return x; + + outb(0x1A, data->addr + ABIT_UGURU3_DATA); + if ((x = abituguru3_wait_while_busy(data)) != ABIT_UGURU3_SUCCESS) { + ABIT_UGURU3_DEBUG("read from 0x%02x:0x%02x timed out after " + "sending 0x1A, status: 0x%02x\n", (unsigned int)bank, + (unsigned int)offset, x); + return -EIO; + } + + outb(bank, data->addr + ABIT_UGURU3_CMD); + if ((x = abituguru3_wait_while_busy(data)) != ABIT_UGURU3_SUCCESS) { + ABIT_UGURU3_DEBUG("read from 0x%02x:0x%02x timed out after " + "sending the bank, status: 0x%02x\n", + (unsigned int)bank, (unsigned int)offset, x); + return -EIO; + } + + outb(offset, data->addr + ABIT_UGURU3_CMD); + if ((x = abituguru3_wait_while_busy(data)) != ABIT_UGURU3_SUCCESS) { + ABIT_UGURU3_DEBUG("read from 0x%02x:0x%02x timed out after " + "sending the offset, status: 0x%02x\n", + (unsigned int)bank, (unsigned int)offset, x); + return -EIO; + } + + outb(count, data->addr + ABIT_UGURU3_CMD); + if ((x = abituguru3_wait_while_busy(data)) != ABIT_UGURU3_SUCCESS) { + ABIT_UGURU3_DEBUG("read from 0x%02x:0x%02x timed out after " + "sending the count, status: 0x%02x\n", + (unsigned int)bank, (unsigned int)offset, x); + return -EIO; + } + + for (i = 0; i < count; i++) { + if ((x = abituguru3_wait_for_read(data)) != + ABIT_UGURU3_SUCCESS) { + ABIT_UGURU3_DEBUG("timeout reading byte %d from " + "0x%02x:0x%02x, status: 0x%02x\n", i, + (unsigned int)bank, (unsigned int)offset, x); + break; + } + buf[i] = inb(data->addr + ABIT_UGURU3_CMD); + } + return i; +} + +/* Sensor settings are stored 1 byte per offset with the bytes + placed add consecutive offsets. */ +int abituguru3_read_increment_offset(struct abituguru3_data *data, u8 bank, + u8 offset, u8 count, u8 *buf, int offset_count) +{ + int i, x; + + for (i = 0; i < offset_count; i++) + if ((x = abituguru3_read(data, bank, offset + i, count, + buf + i * count)) != count) + return i * count + (i && (x < 0)) ? 0 : x; + + return i * count; +} + +/* Following are the sysfs callback functions. These functions expect: + sensor_device_attribute_2->index: index into the data->sensors array + sensor_device_attribute_2->nr: register offset, bitmask or NA. */ +static struct abituguru3_data *abituguru3_update_device(struct device *dev); + +static ssize_t show_value(struct device *dev, + struct device_attribute *devattr, char *buf) +{ + int value; + struct sensor_device_attribute_2 *attr = to_sensor_dev_attr_2(devattr); + struct abituguru3_data *data = abituguru3_update_device(dev); + const struct abituguru3_sensor_info *sensor; + + if (!data) + return -EIO; + + sensor = &data->sensors[attr->index]; + + /* are we reading a setting, or is this a normal read? */ + if (attr->nr) + value = data->settings[sensor->port][attr->nr]; + else + value = data->value[sensor->port]; + + /* convert the value */ + value = (value * sensor->multiplier) / sensor->divisor + + sensor->offset; + + /* alternatively we could update the sensors settings struct for this, + but then its contents would differ from the windows sw ini files */ + if (sensor->type == ABIT_UGURU3_TEMP_SENSOR) + value *= 1000; + + return sprintf(buf, "%d\n", value); +} + +static ssize_t show_alarm(struct device *dev, + struct device_attribute *devattr, char *buf) +{ + int port; + struct sensor_device_attribute_2 *attr = to_sensor_dev_attr_2(devattr); + struct abituguru3_data *data = abituguru3_update_device(dev); + + if (!data) + return -EIO; + + port = data->sensors[attr->index].port; + + /* See if the alarm bit for this sensor is set and if a bitmask is + given in attr->nr also check if the alarm matches the type of alarm + we're looking for (for volt it can be either low or high). The type + is stored in a few readonly bits in the settings of the sensor. */ + if ((data->alarms[port / 8] & (0x01 << (port % 8))) && + (!attr->nr || (data->settings[port][0] & attr->nr))) + return sprintf(buf, "1\n"); + else + return sprintf(buf, "0\n"); +} + +static ssize_t show_mask(struct device *dev, + struct device_attribute *devattr, char *buf) +{ + struct sensor_device_attribute_2 *attr = to_sensor_dev_attr_2(devattr); + struct abituguru3_data *data = dev_get_drvdata(dev); + + if (data->settings[data->sensors[attr->index].port][0] & attr->nr) + return sprintf(buf, "1\n"); + else + return sprintf(buf, "0\n"); +} + +static ssize_t show_label(struct device *dev, + struct device_attribute *devattr, char *buf) +{ + struct sensor_device_attribute_2 *attr = to_sensor_dev_attr_2(devattr); + struct abituguru3_data *data = dev_get_drvdata(dev); + + return sprintf(buf, "%s\n", data->sensors[attr->index].name); +} + +static ssize_t show_name(struct device *dev, + struct device_attribute *devattr, char *buf) +{ + return sprintf(buf, "%s\n", ABIT_UGURU3_NAME); +} + +/* Sysfs attr templates, the real entries are generated automatically. */ +static const +struct sensor_device_attribute_2 abituguru3_sysfs_templ[3][10] = { { + SENSOR_ATTR_2(in%d_input, 0444, show_value, NULL, 0, 0), + SENSOR_ATTR_2(in%d_min, 0444, show_value, NULL, 1, 0), + SENSOR_ATTR_2(in%d_max, 0444, show_value, NULL, 2, 0), + SENSOR_ATTR_2(in%d_min_alarm, 0444, show_alarm, NULL, + ABIT_UGURU3_VOLT_LOW_ALARM_FLAG, 0), + SENSOR_ATTR_2(in%d_max_alarm, 0444, show_alarm, NULL, + ABIT_UGURU3_VOLT_HIGH_ALARM_FLAG, 0), + SENSOR_ATTR_2(in%d_beep, 0444, show_mask, NULL, + ABIT_UGURU3_BEEP_ENABLE, 0), + SENSOR_ATTR_2(in%d_shutdown, 0444, show_mask, NULL, + ABIT_UGURU3_SHUTDOWN_ENABLE, 0), + SENSOR_ATTR_2(in%d_min_alarm_enable, 0444, show_mask, NULL, + ABIT_UGURU3_VOLT_LOW_ALARM_ENABLE, 0), + SENSOR_ATTR_2(in%d_max_alarm_enable, 0444, show_mask, NULL, + ABIT_UGURU3_VOLT_HIGH_ALARM_ENABLE, 0), + SENSOR_ATTR_2(in%d_label, 0444, show_label, NULL, 0, 0) + }, { + SENSOR_ATTR_2(temp%d_input, 0444, show_value, NULL, 0, 0), + SENSOR_ATTR_2(temp%d_max, 0444, show_value, NULL, 1, 0), + SENSOR_ATTR_2(temp%d_crit, 0444, show_value, NULL, 2, 0), + SENSOR_ATTR_2(temp%d_alarm, 0444, show_alarm, NULL, 0, 0), + SENSOR_ATTR_2(temp%d_beep, 0444, show_mask, NULL, + ABIT_UGURU3_BEEP_ENABLE, 0), + SENSOR_ATTR_2(temp%d_shutdown, 0444, show_mask, NULL, + ABIT_UGURU3_SHUTDOWN_ENABLE, 0), + SENSOR_ATTR_2(temp%d_alarm_enable, 0444, show_mask, NULL, + ABIT_UGURU3_TEMP_HIGH_ALARM_ENABLE, 0), + SENSOR_ATTR_2(temp%d_label, 0444, show_label, NULL, 0, 0) + }, { + SENSOR_ATTR_2(fan%d_input, 0444, show_value, NULL, 0, 0), + SENSOR_ATTR_2(fan%d_min, 0444, show_value, NULL, 1, 0), + SENSOR_ATTR_2(fan%d_alarm, 0444, show_alarm, NULL, 0, 0), + SENSOR_ATTR_2(fan%d_beep, 0444, show_mask, NULL, + ABIT_UGURU3_BEEP_ENABLE, 0), + SENSOR_ATTR_2(fan%d_shutdown, 0444, show_mask, NULL, + ABIT_UGURU3_SHUTDOWN_ENABLE, 0), + SENSOR_ATTR_2(fan%d_alarm_enable, 0444, show_mask, NULL, + ABIT_UGURU3_FAN_LOW_ALARM_ENABLE, 0), + SENSOR_ATTR_2(fan%d_label, 0444, show_label, NULL, 0, 0) +} }; + +static struct sensor_device_attribute_2 abituguru3_sysfs_attr[] = { + SENSOR_ATTR_2(name, 0444, show_name, NULL, 0, 0), +}; + +static int __devinit abituguru3_probe(struct platform_device *pdev) +{ + const int no_sysfs_attr[3] = { 10, 8, 7 }; + int sensor_index[3] = { 0, 1, 1 }; + struct abituguru3_data *data; + int i, j, type, used, sysfs_names_free, sysfs_attr_i, res = -ENODEV; + char *sysfs_filename; + u8 buf[2]; + u16 id; + + if (!(data = kzalloc(sizeof(struct abituguru3_data), GFP_KERNEL))) + return -ENOMEM; + + data->addr = platform_get_resource(pdev, IORESOURCE_IO, 0)->start; + mutex_init(&data->update_lock); + platform_set_drvdata(pdev, data); + + /* Read the motherboard ID */ + if ((i = abituguru3_read(data, ABIT_UGURU3_MISC_BANK, + ABIT_UGURU3_BOARD_ID, 2, buf)) != 2) { + goto abituguru3_probe_error; + } + + /* Completely read the uGuru to see if one really is there */ + if (!abituguru3_update_device(&pdev->dev)) + goto abituguru3_probe_error; + + /* lookup the ID in our motherboard table */ + id = ((u16)buf[0] << 8) | (u16)buf[1]; + for (i = 0; abituguru3_motherboards[i].id; i++) + if (abituguru3_motherboards[i].id == id) + break; + if (!abituguru3_motherboards[i].id) { + printk(KERN_ERR ABIT_UGURU3_NAME ": error unknown motherboard " + "ID: %04X. Please report this to the abituguru3 " + "maintainer (see MAINTAINERS)\n", (unsigned int)id); + goto abituguru3_probe_error; + } + data->sensors = abituguru3_motherboards[i].sensors; + printk(KERN_INFO ABIT_UGURU3_NAME ": found Abit uGuru3, motherboard " + "ID: %04X (%s)\n", (unsigned int)id, + abituguru3_motherboards[i].name); + + /* Fill the sysfs attr array */ + sysfs_attr_i = 0; + sysfs_filename = data->sysfs_names; + sysfs_names_free = ABIT_UGURU3_SYSFS_NAMES_LENGTH; + for (i = 0; data->sensors[i].name; i++) { + /* Fail safe check, this should never happen! */ + if (i >= ABIT_UGURU3_MAX_NO_SENSORS) { + printk(KERN_ERR ABIT_UGURU3_NAME + ": Fatal error motherboard has more sensors " + "then ABIT_UGURU3_MAX_NO_SENSORS. This should " + "never happen please report to the abituguru3 " + "maintainer (see MAINTAINERS)\n"); + res = -ENAMETOOLONG; + goto abituguru3_probe_error; + } + type = data->sensors[i].type; + for (j = 0; j < no_sysfs_attr[type]; j++) { + used = snprintf(sysfs_filename, sysfs_names_free, + abituguru3_sysfs_templ[type][j].dev_attr.attr. + name, sensor_index[type]) + 1; + data->sysfs_attr[sysfs_attr_i] = + abituguru3_sysfs_templ[type][j]; + data->sysfs_attr[sysfs_attr_i].dev_attr.attr.name = + sysfs_filename; + data->sysfs_attr[sysfs_attr_i].index = i; + sysfs_filename += used; + sysfs_names_free -= used; + sysfs_attr_i++; + } + sensor_index[type]++; + } + /* Fail safe check, this should never happen! */ + if (sysfs_names_free < 0) { + printk(KERN_ERR ABIT_UGURU3_NAME + ": Fatal error ran out of space for sysfs attr names. " + "This should never happen please report to the " + "abituguru3 maintainer (see MAINTAINERS)\n"); + res = -ENAMETOOLONG; + goto abituguru3_probe_error; + } + + /* Register sysfs hooks */ + for (i = 0; i < sysfs_attr_i; i++) + if (device_create_file(&pdev->dev, + &data->sysfs_attr[i].dev_attr)) + goto abituguru3_probe_error; + for (i = 0; i < ARRAY_SIZE(abituguru3_sysfs_attr); i++) + if (device_create_file(&pdev->dev, + &abituguru3_sysfs_attr[i].dev_attr)) + goto abituguru3_probe_error; + + data->class_dev = hwmon_device_register(&pdev->dev); + if (IS_ERR(data->class_dev)) { + res = PTR_ERR(data->class_dev); + goto abituguru3_probe_error; + } + + return 0; /* success */ + +abituguru3_probe_error: + for (i = 0; data->sysfs_attr[i].dev_attr.attr.name; i++) + device_remove_file(&pdev->dev, &data->sysfs_attr[i].dev_attr); + for (i = 0; i < ARRAY_SIZE(abituguru3_sysfs_attr); i++) + device_remove_file(&pdev->dev, + &abituguru3_sysfs_attr[i].dev_attr); + kfree(data); + return res; +} + +static int __devexit abituguru3_remove(struct platform_device *pdev) +{ + int i; + struct abituguru3_data *data = platform_get_drvdata(pdev); + + platform_set_drvdata(pdev, NULL); + hwmon_device_unregister(data->class_dev); + for (i = 0; data->sysfs_attr[i].dev_attr.attr.name; i++) + device_remove_file(&pdev->dev, &data->sysfs_attr[i].dev_attr); + for (i = 0; i < ARRAY_SIZE(abituguru3_sysfs_attr); i++) + device_remove_file(&pdev->dev, + &abituguru3_sysfs_attr[i].dev_attr); + kfree(data); + + return 0; +} + +static struct abituguru3_data *abituguru3_update_device(struct device *dev) +{ + int i; + struct abituguru3_data *data = dev_get_drvdata(dev); + + mutex_lock(&data->update_lock); + if (!data->valid || time_after(jiffies, data->last_updated + HZ)) { + /* Clear data->valid while updating */ + data->valid = 0; + /* Read alarms */ + if (abituguru3_read_increment_offset(data, + ABIT_UGURU3_SETTINGS_BANK, + ABIT_UGURU3_ALARMS_START, + 1, data->alarms, 48/8) != (48/8)) + goto LEAVE_UPDATE; + /* Read in and temp sensors (3 byte settings / sensor) */ + for (i = 0; i < 32; i++) { + if (abituguru3_read(data, ABIT_UGURU3_SENSORS_BANK, + ABIT_UGURU3_VALUES_START + i, + 1, &data->value[i]) != 1) + goto LEAVE_UPDATE; + if (abituguru3_read_increment_offset(data, + ABIT_UGURU3_SETTINGS_BANK, + ABIT_UGURU3_SETTINGS_START + i * 3, + 1, + data->settings[i], 3) != 3) + goto LEAVE_UPDATE; + } + /* Read temp sensors (2 byte settings / sensor) */ + for (i = 0; i < 16; i++) { + if (abituguru3_read(data, ABIT_UGURU3_SENSORS_BANK, + ABIT_UGURU3_VALUES_START + 32 + i, + 1, &data->value[32 + i]) != 1) + goto LEAVE_UPDATE; + if (abituguru3_read_increment_offset(data, + ABIT_UGURU3_SETTINGS_BANK, + ABIT_UGURU3_SETTINGS_START + 32 * 3 + + i * 2, 1, + data->settings[32 + i], 2) != 2) + goto LEAVE_UPDATE; + } + data->last_updated = jiffies; + data->valid = 1; + } +LEAVE_UPDATE: + mutex_unlock(&data->update_lock); + if (data->valid) + return data; + else + return NULL; +} + +#ifdef CONFIG_PM +static int abituguru3_suspend(struct platform_device *pdev, pm_message_t state) +{ + struct abituguru3_data *data = platform_get_drvdata(pdev); + /* make sure all communications with the uguru3 are done and no new + ones are started */ + mutex_lock(&data->update_lock); + return 0; +} + +static int abituguru3_resume(struct platform_device *pdev) +{ + struct abituguru3_data *data = platform_get_drvdata(pdev); + mutex_unlock(&data->update_lock); + return 0; +} +#else +#define abituguru3_suspend NULL +#define abituguru3_resume NULL +#endif /* CONFIG_PM */ + +static struct platform_driver abituguru3_driver = { + .driver = { + .owner = THIS_MODULE, + .name = ABIT_UGURU3_NAME, + }, + .probe = abituguru3_probe, + .remove = __devexit_p(abituguru3_remove), + .suspend = abituguru3_suspend, + .resume = abituguru3_resume +}; + +static int __init abituguru3_detect(void) +{ + /* See if there is an uguru3 there. An idle uGuru3 will hold 0x00 or + 0x08 at DATA and 0xAC at CMD. Sometimes the uGuru3 will hold 0x05 + at CMD instead, why is unknown. So we test for 0x05 too. */ + u8 data_val = inb_p(ABIT_UGURU3_BASE + ABIT_UGURU3_DATA); + u8 cmd_val = inb_p(ABIT_UGURU3_BASE + ABIT_UGURU3_CMD); + if (((data_val == 0x00) || (data_val == 0x08)) && + ((cmd_val == 0xAC) || (cmd_val == 0x05))) + return ABIT_UGURU3_BASE; + + ABIT_UGURU3_DEBUG("no Abit uGuru3 found, data = 0x%02X, cmd = " + "0x%02X\n", (unsigned int)data_val, (unsigned int)cmd_val); + + if (force) { + printk(KERN_INFO ABIT_UGURU3_NAME ": Assuming Abit uGuru3 is " + "present because of \"force\" parameter\n"); + return ABIT_UGURU3_BASE; + } + + /* No uGuru3 found */ + return -ENODEV; +} + +static struct platform_device *abituguru3_pdev; + +static int __init abituguru3_init(void) +{ + int address, err; + struct resource res = { .flags = IORESOURCE_IO }; + + address = abituguru3_detect(); + if (address < 0) + return address; + + err = platform_driver_register(&abituguru3_driver); + if (err) + goto exit; + + abituguru3_pdev = platform_device_alloc(ABIT_UGURU3_NAME, address); + if (!abituguru3_pdev) { + printk(KERN_ERR ABIT_UGURU3_NAME + ": Device allocation failed\n"); + err = -ENOMEM; + goto exit_driver_unregister; + } + + res.start = address; + res.end = address + ABIT_UGURU3_REGION_LENGTH - 1; + res.name = ABIT_UGURU3_NAME; + + err = platform_device_add_resources(abituguru3_pdev, &res, 1); + if (err) { + printk(KERN_ERR ABIT_UGURU3_NAME + ": Device resource addition failed (%d)\n", err); + goto exit_device_put; + } + + err = platform_device_add(abituguru3_pdev); + if (err) { + printk(KERN_ERR ABIT_UGURU3_NAME + ": Device addition failed (%d)\n", err); + goto exit_device_put; + } + + return 0; + +exit_device_put: + platform_device_put(abituguru3_pdev); +exit_driver_unregister: + platform_driver_unregister(&abituguru3_driver); +exit: + return err; +} + +static void __exit abituguru3_exit(void) +{ + platform_device_unregister(abituguru3_pdev); + platform_driver_unregister(&abituguru3_driver); +} + +MODULE_AUTHOR("Hans de Goede <j.w.r.degoede@hhs.nl>"); +MODULE_DESCRIPTION("Abit uGuru3 Sensor device"); +MODULE_LICENSE("GPL"); + +module_init(abituguru3_init); +module_exit(abituguru3_exit); diff --git a/drivers/hwmon/coretemp.c b/drivers/hwmon/coretemp.c index 6d54c8caed7..7c1795225b0 100644 --- a/drivers/hwmon/coretemp.c +++ b/drivers/hwmon/coretemp.c @@ -318,7 +318,7 @@ exit: } #ifdef CONFIG_HOTPLUG_CPU -void coretemp_device_remove(unsigned int cpu) +static void coretemp_device_remove(unsigned int cpu) { struct pdev_entry *p, *n; mutex_lock(&pdev_list_mutex); diff --git a/drivers/hwmon/dme1737.c b/drivers/hwmon/dme1737.c new file mode 100644 index 00000000000..be3aaa5d0b9 --- /dev/null +++ b/drivers/hwmon/dme1737.c @@ -0,0 +1,2080 @@ +/* + * dme1737.c - driver for the SMSC DME1737 and Asus A8000 Super-I/O chips + * integrated hardware monitoring features. + * Copyright (c) 2007 Juerg Haefliger <juergh@gmail.com> + * + * This driver is based on the LM85 driver. The hardware monitoring + * capabilities of the DME1737 are very similar to the LM85 with some + * additional features. Even though the DME1737 is a Super-I/O chip, the + * hardware monitoring registers are only accessible via SMBus. + * + * 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. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA. + */ + +#include <linux/module.h> +#include <linux/init.h> +#include <linux/slab.h> +#include <linux/jiffies.h> +#include <linux/i2c.h> +#include <linux/hwmon.h> +#include <linux/hwmon-sysfs.h> +#include <linux/hwmon-vid.h> +#include <linux/err.h> +#include <linux/mutex.h> +#include <asm/io.h> + +/* Module load parameters */ +static int force_start; +module_param(force_start, bool, 0); +MODULE_PARM_DESC(force_start, "Force the chip to start monitoring inputs"); + +/* Addresses to scan */ +static unsigned short normal_i2c[] = {0x2c, 0x2d, 0x2e, I2C_CLIENT_END}; + +/* Insmod parameters */ +I2C_CLIENT_INSMOD_1(dme1737); + +/* --------------------------------------------------------------------- + * Registers + * + * The sensors are defined as follows: + * + * Voltages Temperatures + * -------- ------------ + * in0 +5VTR (+5V stdby) temp1 Remote diode 1 + * in1 Vccp (proc core) temp2 Internal temp + * in2 VCC (internal +3.3V) temp3 Remote diode 2 + * in3 +5V + * in4 +12V + * in5 VTR (+3.3V stby) + * in6 Vbat + * + * --------------------------------------------------------------------- */ + +/* Voltages (in) numbered 0-6 (ix) */ +#define DME1737_REG_IN(ix) ((ix) < 5 ? 0x20 + (ix) \ + : 0x94 + (ix)) +#define DME1737_REG_IN_MIN(ix) ((ix) < 5 ? 0x44 + (ix) * 2 \ + : 0x91 + (ix) * 2) +#define DME1737_REG_IN_MAX(ix) ((ix) < 5 ? 0x45 + (ix) * 2 \ + : 0x92 + (ix) * 2) + +/* Temperatures (temp) numbered 0-2 (ix) */ +#define DME1737_REG_TEMP(ix) (0x25 + (ix)) +#define DME1737_REG_TEMP_MIN(ix) (0x4e + (ix) * 2) +#define DME1737_REG_TEMP_MAX(ix) (0x4f + (ix) * 2) +#define DME1737_REG_TEMP_OFFSET(ix) ((ix) == 0 ? 0x1f \ + : 0x1c + (ix)) + +/* Voltage and temperature LSBs + * The LSBs (4 bits each) are stored in 5 registers with the following layouts: + * IN_TEMP_LSB(0) = [in5, in6] + * IN_TEMP_LSB(1) = [temp3, temp1] + * IN_TEMP_LSB(2) = [in4, temp2] + * IN_TEMP_LSB(3) = [in3, in0] + * IN_TEMP_LSB(4) = [in2, in1] */ +#define DME1737_REG_IN_TEMP_LSB(ix) (0x84 + (ix)) +static const u8 DME1737_REG_IN_LSB[] = {3, 4, 4, 3, 2, 0, 0}; +static const u8 DME1737_REG_IN_LSB_SHL[] = {4, 4, 0, 0, 0, 0, 4}; +static const u8 DME1737_REG_TEMP_LSB[] = {1, 2, 1}; +static const u8 DME1737_REG_TEMP_LSB_SHL[] = {4, 4, 0}; + +/* Fans numbered 0-5 (ix) */ +#define DME1737_REG_FAN(ix) ((ix) < 4 ? 0x28 + (ix) * 2 \ + : 0xa1 + (ix) * 2) +#define DME1737_REG_FAN_MIN(ix) ((ix) < 4 ? 0x54 + (ix) * 2 \ + : 0xa5 + (ix) * 2) +#define DME1737_REG_FAN_OPT(ix) ((ix) < 4 ? 0x90 + (ix) \ + : 0xb2 + (ix)) +#define DME1737_REG_FAN_MAX(ix) (0xb4 + (ix)) /* only for fan[4-5] */ + +/* PWMs numbered 0-2, 4-5 (ix) */ +#define DME1737_REG_PWM(ix) ((ix) < 3 ? 0x30 + (ix) \ + : 0xa1 + (ix)) +#define DME1737_REG_PWM_CONFIG(ix) (0x5c + (ix)) /* only for pwm[0-2] */ +#define DME1737_REG_PWM_MIN(ix) (0x64 + (ix)) /* only for pwm[0-2] */ +#define DME1737_REG_PWM_FREQ(ix) ((ix) < 3 ? 0x5f + (ix) \ + : 0xa3 + (ix)) +/* The layout of the ramp rate registers is different from the other pwm + * registers. The bits for the 3 PWMs are stored in 2 registers: + * PWM_RR(0) = [OFF3, OFF2, OFF1, RES, RR1E, RR1-2, RR1-1, RR1-0] + * PWM_RR(1) = [RR2E, RR2-2, RR2-1, RR2-0, RR3E, RR3-2, RR3-1, RR3-0] */ +#define DME1737_REG_PWM_RR(ix) (0x62 + (ix)) /* only for pwm[0-2] */ + +/* Thermal zones 0-2 */ +#define DME1737_REG_ZONE_LOW(ix) (0x67 + (ix)) +#define DME1737_REG_ZONE_ABS(ix) (0x6a + (ix)) +/* The layout of the hysteresis registers is different from the other zone + * registers. The bits for the 3 zones are stored in 2 registers: + * ZONE_HYST(0) = [H1-3, H1-2, H1-1, H1-0, H2-3, H2-2, H2-1, H2-0] + * ZONE_HYST(1) = [H3-3, H3-2, H3-1, H3-0, RES, RES, RES, RES] */ +#define DME1737_REG_ZONE_HYST(ix) (0x6d + (ix)) + +/* Alarm registers and bit mapping + * The 3 8-bit alarm registers will be concatenated to a single 32-bit + * alarm value [0, ALARM3, ALARM2, ALARM1]. */ +#define DME1737_REG_ALARM1 0x41 +#define DME1737_REG_ALARM2 0x42 +#define DME1737_REG_ALARM3 0x83 +static const u8 DME1737_BIT_ALARM_IN[] = {0, 1, 2, 3, 8, 16, 17}; +static const u8 DME1737_BIT_ALARM_TEMP[] = {4, 5, 6}; +static const u8 DME1737_BIT_ALARM_FAN[] = {10, 11, 12, 13, 22, 23}; + +/* Miscellaneous registers */ +#define DME1737_REG_COMPANY 0x3e +#define DME1737_REG_VERSTEP 0x3f +#define DME1737_REG_CONFIG 0x40 +#define DME1737_REG_CONFIG2 0x7f +#define DME1737_REG_VID 0x43 +#define DME1737_REG_TACH_PWM 0x81 + +/* --------------------------------------------------------------------- + * Misc defines + * --------------------------------------------------------------------- */ + +/* Chip identification */ +#define DME1737_COMPANY_SMSC 0x5c +#define DME1737_VERSTEP 0x88 +#define DME1737_VERSTEP_MASK 0xf8 + +/* --------------------------------------------------------------------- + * Data structures and manipulation thereof + * --------------------------------------------------------------------- */ + +struct dme1737_data { + struct i2c_client client; + struct class_device *class_dev; + + struct mutex update_lock; + int valid; /* !=0 if following fields are valid */ + unsigned long last_update; /* in jiffies */ + unsigned long last_vbat; /* in jiffies */ + + u8 vid; + u8 pwm_rr_en; + u8 has_pwm; + u8 has_fan; + + /* Register values */ + u16 in[7]; + u8 in_min[7]; + u8 in_max[7]; + s16 temp[3]; + s8 temp_min[3]; + s8 temp_max[3]; + s8 temp_offset[3]; + u8 config; + u8 config2; + u8 vrm; + u16 fan[6]; + u16 fan_min[6]; + u8 fan_max[2]; + u8 fan_opt[6]; + u8 pwm[6]; + u8 pwm_min[3]; + u8 pwm_config[3]; + u8 pwm_acz[3]; + u8 pwm_freq[6]; + u8 pwm_rr[2]; + u8 zone_low[3]; + u8 zone_abs[3]; + u8 zone_hyst[2]; + u32 alarms; +}; + +/* Nominal voltage values */ +static const int IN_NOMINAL[] = {5000, 2250, 3300, 5000, 12000, 3300, 3300}; + +/* Voltage input + * Voltage inputs have 16 bits resolution, limit values have 8 bits + * resolution. */ +static inline int IN_FROM_REG(int reg, int ix, int res) +{ + return (reg * IN_NOMINAL[ix] + (3 << (res - 3))) / (3 << (res - 2)); +} + +static inline int IN_TO_REG(int val, int ix) +{ + return SENSORS_LIMIT((val * 192 + IN_NOMINAL[ix] / 2) / + IN_NOMINAL[ix], 0, 255); +} + +/* Temperature input + * The register values represent temperatures in 2's complement notation from + * -127 degrees C to +127 degrees C. Temp inputs have 16 bits resolution, limit + * values have 8 bits resolution. */ +static inline int TEMP_FROM_REG(int reg, int res) +{ + return (reg * 1000) >> (res - 8); +} + +static inline int TEMP_TO_REG(int val) +{ + return SENSORS_LIMIT((val < 0 ? val - 500 : val + 500) / 1000, + -128, 127); +} + +/* Temperature range */ +static const int TEMP_RANGE[] = {2000, 2500, 3333, 4000, 5000, 6666, 8000, + 10000, 13333, 16000, 20000, 26666, 32000, + 40000, 53333, 80000}; + +static inline int TEMP_RANGE_FROM_REG(int reg) +{ + return TEMP_RANGE[(reg >> 4) & 0x0f]; +} + +static int TEMP_RANGE_TO_REG(int val, int reg) +{ + int i; + + for (i = 15; i > 0; i--) { + if (val > (TEMP_RANGE[i] + TEMP_RANGE[i - 1] + 1) / 2) { + break; + } + } + + return (reg & 0x0f) | (i << 4); +} + +/* Temperature hysteresis + * Register layout: + * reg[0] = [H1-3, H1-2, H1-1, H1-0, H2-3, H2-2, H2-1, H2-0] + * reg[1] = [H3-3, H3-2, H3-1, H3-0, xxxx, xxxx, xxxx, xxxx] */ +static inline int TEMP_HYST_FROM_REG(int reg, int ix) +{ + return (((ix == 1) ? reg : reg >> 4) & 0x0f) * 1000; +} + +static inline int TEMP_HYST_TO_REG(int val, int ix, int reg) +{ + int hyst = SENSORS_LIMIT((val + 500) / 1000, 0, 15); + + return (ix == 1) ? (reg & 0xf0) | hyst : (reg & 0x0f) | (hyst << 4); +} + +/* Fan input RPM */ +static inline int FAN_FROM_REG(int reg, int tpc) +{ + return (reg == 0 || reg == 0xffff) ? 0 : + (tpc == 0) ? 90000 * 60 / reg : tpc * reg; +} + +static inline int FAN_TO_REG(int val, int tpc) +{ + return SENSORS_LIMIT((tpc == 0) ? 90000 * 60 / val : val / tpc, + 0, 0xffff); +} + +/* Fan TPC (tach pulse count) + * Converts a register value to a TPC multiplier or returns 0 if the tachometer + * is configured in legacy (non-tpc) mode */ +static inline int FAN_TPC_FROM_REG(int reg) +{ + return (reg & 0x20) ? 0 : 60 >> (reg & 0x03); +} + +/* Fan type + * The type of a fan is expressed in number of pulses-per-revolution that it + * emits */ +static inline int FAN_TYPE_FROM_REG(int reg) +{ + int edge = (reg >> 1) & 0x03; + + return (edge > 0) ? 1 << (edge - 1) : 0; +} + +static inline int FAN_TYPE_TO_REG(int val, int reg) +{ + int edge = (val == 4) ? 3 : val; + + return (reg & 0xf9) | (edge << 1); +} + +/* Fan max RPM */ +static const int FAN_MAX[] = {0x54, 0x38, 0x2a, 0x21, 0x1c, 0x18, 0x15, 0x12, + 0x11, 0x0f, 0x0e}; + +static int FAN_MAX_FROM_REG(int reg) +{ + int i; + + for (i = 10; i > 0; i--) { + if (reg == FAN_MAX[i]) { + break; + } + } + + return 1000 + i * 500; +} + +static int FAN_MAX_TO_REG(int val) +{ + int i; + + for (i = 10; i > 0; i--) { + if (val > (1000 + (i - 1) * 500)) { + break; + } + } + + return FAN_MAX[i]; +} + +/* PWM enable + * Register to enable mapping: + * 000: 2 fan on zone 1 auto + * 001: 2 fan on zone 2 auto + * 010: 2 fan on zone 3 auto + * 011: 0 fan full on + * 100: -1 fan disabled + * 101: 2 fan on hottest of zones 2,3 auto + * 110: 2 fan on hottest of zones 1,2,3 auto + * 111: 1 fan in manual mode */ +static inline int PWM_EN_FROM_REG(int reg) +{ + static const int en[] = {2, 2, 2, 0, -1, 2, 2, 1}; + + return en[(reg >> 5) & 0x07]; +} + +static inline int PWM_EN_TO_REG(int val, int reg) +{ + int en = (val == 1) ? 7 : 3; + + return (reg & 0x1f) | ((en & 0x07) << 5); +} + +/* PWM auto channels zone + * Register to auto channels zone mapping (ACZ is a bitfield with bit x + * corresponding to zone x+1): + * 000: 001 fan on zone 1 auto + * 001: 010 fan on zone 2 auto + * 010: 100 fan on zone 3 auto + * 011: 000 fan full on + * 100: 000 fan disabled + * 101: 110 fan on hottest of zones 2,3 auto + * 110: 111 fan on hottest of zones 1,2,3 auto + * 111: 000 fan in manual mode */ +static inline int PWM_ACZ_FROM_REG(int reg) +{ + static const int acz[] = {1, 2, 4, 0, 0, 6, 7, 0}; + + return acz[(reg >> 5) & 0x07]; +} + +static inline int PWM_ACZ_TO_REG(int val, int reg) +{ + int acz = (val == 4) ? 2 : val - 1; + + return (reg & 0x1f) | ((acz & 0x07) << 5); +} + +/* PWM frequency */ +static const int PWM_FREQ[] = {11, 15, 22, 29, 35, 44, 59, 88, + 15000, 20000, 30000, 25000, 0, 0, 0, 0}; + +static inline int PWM_FREQ_FROM_REG(int reg) +{ + return PWM_FREQ[reg & 0x0f]; +} + +static int PWM_FREQ_TO_REG(int val, int reg) +{ + int i; + + /* the first two cases are special - stupid chip design! */ + if (val > 27500) { + i = 10; + } else if (val > 22500) { + i = 11; + } else { + for (i = 9; i > 0; i--) { + if (val > (PWM_FREQ[i] + PWM_FREQ[i - 1] + 1) / 2) { + break; + } + } + } + + return (reg & 0xf0) | i; +} + +/* PWM ramp rate + * Register layout: + * reg[0] = [OFF3, OFF2, OFF1, RES, RR1-E, RR1-2, RR1-1, RR1-0] + * reg[1] = [RR2-E, RR2-2, RR2-1, RR2-0, RR3-E, RR3-2, RR3-1, RR3-0] */ +static const u8 PWM_RR[] = {206, 104, 69, 41, 26, 18, 10, 5}; + +static inline int PWM_RR_FROM_REG(int reg, int ix) +{ + int rr = (ix == 1) ? reg >> 4 : reg; + + return (rr & 0x08) ? PWM_RR[rr & 0x07] : 0; +} + +static int PWM_RR_TO_REG(int val, int ix, int reg) +{ + int i; + + for (i = 0; i < 7; i++) { + if (val > (PWM_RR[i] + PWM_RR[i + 1] + 1) / 2) { + break; + } + } + + return (ix == 1) ? (reg & 0x8f) | (i << 4) : (reg & 0xf8) | i; +} + +/* PWM ramp rate enable */ +static inline int PWM_RR_EN_FROM_REG(int reg, int ix) +{ + return PWM_RR_FROM_REG(reg, ix) ? 1 : 0; +} + +static inline int PWM_RR_EN_TO_REG(int val, int ix, int reg) +{ + int en = (ix == 1) ? 0x80 : 0x08; + + return val ? reg | en : reg & ~en; +} + +/* PWM min/off + * The PWM min/off bits are part of the PMW ramp rate register 0 (see above for + * the register layout). */ +static inline int PWM_OFF_FROM_REG(int reg, int ix) +{ + return (reg >> (ix + 5)) & 0x01; +} + +static inline int PWM_OFF_TO_REG(int val, int ix, int reg) +{ + return (reg & ~(1 << (ix + 5))) | ((val & 0x01) << (ix + 5)); +} + +/* --------------------------------------------------------------------- + * Device I/O access + * --------------------------------------------------------------------- */ + +static u8 dme1737_read(struct i2c_client *client, u8 reg) +{ + s32 val = i2c_smbus_read_byte_data(client, reg); + + if (val < 0) { + dev_warn(&client->dev, "Read from register 0x%02x failed! " + "Please report to the driver maintainer.\n", reg); + } + + return val; +} + +static s32 dme1737_write(struct i2c_client *client, u8 reg, u8 value) +{ + s32 res = i2c_smbus_write_byte_data(client, reg, value); + + if (res < 0) { + dev_warn(&client->dev, "Write to register 0x%02x failed! " + "Please report to the driver maintainer.\n", reg); + } + + return res; +} + +static struct dme1737_data *dme1737_update_device(struct device *dev) +{ + struct i2c_client *client = to_i2c_client(dev); + struct dme1737_data *data = i2c_get_clientdata(client); + int ix; + u8 lsb[5]; + + mutex_lock(&data->update_lock); + + /* Enable a Vbat monitoring cycle every 10 mins */ + if (time_after(jiffies, data->last_vbat + 600 * HZ) || !data->valid) { + dme1737_write(client, DME1737_REG_CONFIG, dme1737_read(client, + DME1737_REG_CONFIG) | 0x10); + data->last_vbat = jiffies; + } + + /* Sample register contents every 1 sec */ + if (time_after(jiffies, data->last_update + HZ) || !data->valid) { + data->vid = dme1737_read(client, DME1737_REG_VID) & 0x3f; + + /* In (voltage) registers */ + for (ix = 0; ix < ARRAY_SIZE(data->in); ix++) { + /* Voltage inputs are stored as 16 bit values even + * though they have only 12 bits resolution. This is + * to make it consistent with the temp inputs. */ + data->in[ix] = dme1737_read(client, + DME1737_REG_IN(ix)) << 8; + data->in_min[ix] = dme1737_read(client, + DME1737_REG_IN_MIN(ix)); + data->in_max[ix] = dme1737_read(client, + DME1737_REG_IN_MAX(ix)); + } + + /* Temp registers */ + for (ix = 0; ix < ARRAY_SIZE(data->temp); ix++) { + /* Temp inputs are stored as 16 bit values even + * though they have only 12 bits resolution. This is + * to take advantage of implicit conversions between + * register values (2's complement) and temp values + * (signed decimal). */ + data->temp[ix] = dme1737_read(client, + DME1737_REG_TEMP(ix)) << 8; + data->temp_min[ix] = dme1737_read(client, + DME1737_REG_TEMP_MIN(ix)); + data->temp_max[ix] = dme1737_read(client, + DME1737_REG_TEMP_MAX(ix)); + data->temp_offset[ix] = dme1737_read(client, + DME1737_REG_TEMP_OFFSET(ix)); + } + + /* In and temp LSB registers + * The LSBs are latched when the MSBs are read, so the order in + * which the registers are read (MSB first, then LSB) is + * important! */ + for (ix = 0; ix < ARRAY_SIZE(lsb); ix++) { + lsb[ix] = dme1737_read(client, + DME1737_REG_IN_TEMP_LSB(ix)); + } + for (ix = 0; ix < ARRAY_SIZE(data->in); ix++) { + data->in[ix] |= (lsb[DME1737_REG_IN_LSB[ix]] << + DME1737_REG_IN_LSB_SHL[ix]) & 0xf0; + } + for (ix = 0; ix < ARRAY_SIZE(data->temp); ix++) { + data->temp[ix] |= (lsb[DME1737_REG_TEMP_LSB[ix]] << + DME1737_REG_TEMP_LSB_SHL[ix]) & 0xf0; + } + + /* Fan registers */ + for (ix = 0; ix < ARRAY_SIZE(data->fan); ix++) { + /* Skip reading registers if optional fans are not + * present */ + if (!(data->has_fan & (1 << ix))) { + continue; + } + data->fan[ix] = dme1737_read(client, + DME1737_REG_FAN(ix)); + data->fan[ix] |= dme1737_read(client, + DME1737_REG_FAN(ix) + 1) << 8; + data->fan_min[ix] = dme1737_read(client, + DME1737_REG_FAN_MIN(ix)); + data->fan_min[ix] |= dme1737_read(client, + DME1737_REG_FAN_MIN(ix) + 1) << 8; + data->fan_opt[ix] = dme1737_read(client, + DME1737_REG_FAN_OPT(ix)); + /* fan_max exists only for fan[5-6] */ + if (ix > 3) { + data->fan_max[ix - 4] = dme1737_read(client, + DME1737_REG_FAN_MAX(ix)); + } + } + + /* PWM registers */ + for (ix = 0; ix < ARRAY_SIZE(data->pwm); ix++) { + /* Skip reading registers if optional PWMs are not + * present */ + if (!(data->has_pwm & (1 << ix))) { + continue; + } + data->pwm[ix] = dme1737_read(client, + DME1737_REG_PWM(ix)); + data->pwm_freq[ix] = dme1737_read(client, + DME1737_REG_PWM_FREQ(ix)); + /* pwm_config and pwm_min exist only for pwm[1-3] */ + if (ix < 3) { + data->pwm_config[ix] = dme1737_read(client, + DME1737_REG_PWM_CONFIG(ix)); + data->pwm_min[ix] = dme1737_read(client, + DME1737_REG_PWM_MIN(ix)); + } + } + for (ix = 0; ix < ARRAY_SIZE(data->pwm_rr); ix++) { + data->pwm_rr[ix] = dme1737_read(client, + DME1737_REG_PWM_RR(ix)); + } + + /* Thermal zone registers */ + for (ix = 0; ix < ARRAY_SIZE(data->zone_low); ix++) { + data->zone_low[ix] = dme1737_read(client, + DME1737_REG_ZONE_LOW(ix)); + data->zone_abs[ix] = dme1737_read(client, + DME1737_REG_ZONE_ABS(ix)); + } + for (ix = 0; ix < ARRAY_SIZE(data->zone_hyst); ix++) { + data->zone_hyst[ix] = dme1737_read(client, + DME1737_REG_ZONE_HYST(ix)); + } + + /* Alarm registers */ + data->alarms = dme1737_read(client, + DME1737_REG_ALARM1); + /* Bit 7 tells us if the other alarm registers are non-zero and + * therefore also need to be read */ + if (data->alarms & 0x80) { + data->alarms |= dme1737_read(client, + DME1737_REG_ALARM2) << 8; + data->alarms |= dme1737_read(client, + DME1737_REG_ALARM3) << 16; + } + + data->last_update = jiffies; + data->valid = 1; + } + + mutex_unlock(&data->update_lock); + + return data; +} + +/* --------------------------------------------------------------------- + * Voltage sysfs attributes + * ix = [0-5] + * --------------------------------------------------------------------- */ + +#define SYS_IN_INPUT 0 +#define SYS_IN_MIN 1 +#define SYS_IN_MAX 2 +#define SYS_IN_ALARM 3 + +static ssize_t show_in(struct device *dev, struct device_attribute *attr, + char *buf) +{ + struct dme1737_data *data = dme1737_update_device(dev); + struct sensor_device_attribute_2 + *sensor_attr_2 = to_sensor_dev_attr_2(attr); + int ix = sensor_attr_2->index; + int fn = sensor_attr_2->nr; + int res; + + switch (fn) { + case SYS_IN_INPUT: + res = IN_FROM_REG(data->in[ix], ix, 16); + break; + case SYS_IN_MIN: + res = IN_FROM_REG(data->in_min[ix], ix, 8); + break; + case SYS_IN_MAX: + res = IN_FROM_REG(data->in_max[ix], ix, 8); + break; + case SYS_IN_ALARM: + res = (data->alarms >> DME1737_BIT_ALARM_IN[ix]) & 0x01; + break; + default: + res = 0; + dev_dbg(dev, "Unknown attr fetch (%d)\n", fn); + } + + return sprintf(buf, "%d\n", res); +} + +static ssize_t set_in(struct device *dev, struct device_attribute *attr, + const char *buf, size_t count) +{ + struct i2c_client *client = to_i2c_client(dev); + struct dme1737_data *data = i2c_get_clientdata(client); + struct sensor_device_attribute_2 + *sensor_attr_2 = to_sensor_dev_attr_2(attr); + int ix = sensor_attr_2->index; + int fn = sensor_attr_2->nr; + long val = simple_strtol(buf, NULL, 10); + + mutex_lock(&data->update_lock); + switch (fn) { + case SYS_IN_MIN: + data->in_min[ix] = IN_TO_REG(val, ix); + dme1737_write(client, DME1737_REG_IN_MIN(ix), + data->in_min[ix]); + break; + case SYS_IN_MAX: + data->in_max[ix] = IN_TO_REG(val, ix); + dme1737_write(client, DME1737_REG_IN_MAX(ix), + data->in_max[ix]); + break; + default: + dev_dbg(dev, "Unknown attr fetch (%d)\n", fn); + } + mutex_unlock(&data->update_lock); + + return count; +} + +/* --------------------------------------------------------------------- + * Temperature sysfs attributes + * ix = [0-2] + * --------------------------------------------------------------------- */ + +#define SYS_TEMP_INPUT 0 +#define SYS_TEMP_MIN 1 +#define SYS_TEMP_MAX 2 +#define SYS_TEMP_OFFSET 3 +#define SYS_TEMP_ALARM 4 +#define SYS_TEMP_FAULT 5 + +static ssize_t show_temp(struct device *dev, struct device_attribute *attr, + char *buf) +{ + struct dme1737_data *data = dme1737_update_device(dev); + struct sensor_device_attribute_2 + *sensor_attr_2 = to_sensor_dev_attr_2(attr); + int ix = sensor_attr_2->index; + int fn = sensor_attr_2->nr; + int res; + + switch (fn) { + case SYS_TEMP_INPUT: + res = TEMP_FROM_REG(data->temp[ix], 16); + break; + case SYS_TEMP_MIN: + res = TEMP_FROM_REG(data->temp_min[ix], 8); + break; + case SYS_TEMP_MAX: + res = TEMP_FROM_REG(data->temp_max[ix], 8); + break; + case SYS_TEMP_OFFSET: + res = TEMP_FROM_REG(data->temp_offset[ix], 8); + break; + case SYS_TEMP_ALARM: + res = (data->alarms >> DME1737_BIT_ALARM_TEMP[ix]) & 0x01; + break; + case SYS_TEMP_FAULT: + res = (data->temp[ix] == 0x0800); + break; + default: + res = 0; + dev_dbg(dev, "Unknown attr fetch (%d)\n", fn); + } + + return sprintf(buf, "%d\n", res); +} + +static ssize_t set_temp(struct device *dev, struct device_attribute *attr, + const char *buf, size_t count) +{ + struct i2c_client *client = to_i2c_client(dev); + struct dme1737_data *data = i2c_get_clientdata(client); + struct sensor_device_attribute_2 + *sensor_attr_2 = to_sensor_dev_attr_2(attr); + int ix = sensor_attr_2->index; + int fn = sensor_attr_2->nr; + long val = simple_strtol(buf, NULL, 10); + + mutex_lock(&data->update_lock); + switch (fn) { + case SYS_TEMP_MIN: + data->temp_min[ix] = TEMP_TO_REG(val); + dme1737_write(client, DME1737_REG_TEMP_MIN(ix), + data->temp_min[ix]); + break; + case SYS_TEMP_MAX: + data->temp_max[ix] = TEMP_TO_REG(val); + dme1737_write(client, DME1737_REG_TEMP_MAX(ix), + data->temp_max[ix]); + break; + case SYS_TEMP_OFFSET: + data->temp_offset[ix] = TEMP_TO_REG(val); + dme1737_write(client, DME1737_REG_TEMP_OFFSET(ix), + data->temp_offset[ix]); + break; + default: + dev_dbg(dev, "Unknown attr fetch (%d)\n", fn); + } + mutex_unlock(&data->update_lock); + + return count; +} + +/* --------------------------------------------------------------------- + * Zone sysfs attributes + * ix = [0-2] + * --------------------------------------------------------------------- */ + +#define SYS_ZONE_AUTO_CHANNELS_TEMP 0 +#define SYS_ZONE_AUTO_POINT1_TEMP_HYST 1 +#define SYS_ZONE_AUTO_POINT1_TEMP 2 +#define SYS_ZONE_AUTO_POINT2_TEMP 3 +#define SYS_ZONE_AUTO_POINT3_TEMP 4 + +static ssize_t show_zone(struct device *dev, struct device_attribute *attr, + char *buf) +{ + struct dme1737_data *data = dme1737_update_device(dev); + struct sensor_device_attribute_2 + *sensor_attr_2 = to_sensor_dev_attr_2(attr); + int ix = sensor_attr_2->index; + int fn = sensor_attr_2->nr; + int res; + + switch (fn) { + case SYS_ZONE_AUTO_CHANNELS_TEMP: + /* check config2 for non-standard temp-to-zone mapping */ + if ((ix == 1) && (data->config2 & 0x02)) { + res = 4; + } else { + res = 1 << ix; + } + break; + case SYS_ZONE_AUTO_POINT1_TEMP_HYST: + res = TEMP_FROM_REG(data->zone_low[ix], 8) - + TEMP_HYST_FROM_REG(data->zone_hyst[ix == 2], ix); + break; + case SYS_ZONE_AUTO_POINT1_TEMP: + res = TEMP_FROM_REG(data->zone_low[ix], 8); + break; + case SYS_ZONE_AUTO_POINT2_TEMP: + /* pwm_freq holds the temp range bits in the upper nibble */ + res = TEMP_FROM_REG(data->zone_low[ix], 8) + + TEMP_RANGE_FROM_REG(data->pwm_freq[ix]); + break; + case SYS_ZONE_AUTO_POINT3_TEMP: + res = TEMP_FROM_REG(data->zone_abs[ix], 8); + break; + default: + res = 0; + dev_dbg(dev, "Unknown attr fetch (%d)\n", fn); + } + + return sprintf(buf, "%d\n", res); +} + +static ssize_t set_zone(struct device *dev, struct device_attribute *attr, + const char *buf, size_t count) +{ + struct i2c_client *client = to_i2c_client(dev); + struct dme1737_data *data = i2c_get_clientdata(client); + struct sensor_device_attribute_2 + *sensor_attr_2 = to_sensor_dev_attr_2(attr); + int ix = sensor_attr_2->index; + int fn = sensor_attr_2->nr; + long val = simple_strtol(buf, NULL, 10); + + mutex_lock(&data->update_lock); + switch (fn) { + case SYS_ZONE_AUTO_POINT1_TEMP_HYST: + /* Refresh the cache */ + data->zone_low[ix] = dme1737_read(client, + DME1737_REG_ZONE_LOW(ix)); + /* Modify the temp hyst value */ + data->zone_hyst[ix == 2] = TEMP_HYST_TO_REG( + TEMP_FROM_REG(data->zone_low[ix], 8) - + val, ix, dme1737_read(client, + DME1737_REG_ZONE_HYST(ix == 2))); + dme1737_write(client, DME1737_REG_ZONE_HYST(ix == 2), + data->zone_hyst[ix == 2]); + break; + case SYS_ZONE_AUTO_POINT1_TEMP: + data->zone_low[ix] = TEMP_TO_REG(val); + dme1737_write(client, DME1737_REG_ZONE_LOW(ix), + data->zone_low[ix]); + break; + case SYS_ZONE_AUTO_POINT2_TEMP: + /* Refresh the cache */ + data->zone_low[ix] = dme1737_read(client, + DME1737_REG_ZONE_LOW(ix)); + /* Modify the temp range value (which is stored in the upper + * nibble of the pwm_freq register) */ + data->pwm_freq[ix] = TEMP_RANGE_TO_REG(val - + TEMP_FROM_REG(data->zone_low[ix], 8), + dme1737_read(client, + DME1737_REG_PWM_FREQ(ix))); + dme1737_write(client, DME1737_REG_PWM_FREQ(ix), + data->pwm_freq[ix]); + break; + case SYS_ZONE_AUTO_POINT3_TEMP: + data->zone_abs[ix] = TEMP_TO_REG(val); + dme1737_write(client, DME1737_REG_ZONE_ABS(ix), + data->zone_abs[ix]); + break; + default: + dev_dbg(dev, "Unknown attr fetch (%d)\n", fn); + } + mutex_unlock(&data->update_lock); + + return count; +} + +/* --------------------------------------------------------------------- + * Fan sysfs attributes + * ix = [0-5] + * --------------------------------------------------------------------- */ + +#define SYS_FAN_INPUT 0 +#define SYS_FAN_MIN 1 +#define SYS_FAN_MAX 2 +#define SYS_FAN_ALARM 3 +#define SYS_FAN_TYPE 4 + +static ssize_t show_fan(struct device *dev, struct device_attribute *attr, + char *buf) +{ + struct dme1737_data *data = dme1737_update_device(dev); + struct sensor_device_attribute_2 + *sensor_attr_2 = to_sensor_dev_attr_2(attr); + int ix = sensor_attr_2->index; + int fn = sensor_attr_2->nr; + int res; + + switch (fn) { + case SYS_FAN_INPUT: + res = FAN_FROM_REG(data->fan[ix], + ix < 4 ? 0 : + FAN_TPC_FROM_REG(data->fan_opt[ix])); + break; + case SYS_FAN_MIN: + res = FAN_FROM_REG(data->fan_min[ix], + ix < 4 ? 0 : + FAN_TPC_FROM_REG(data->fan_opt[ix])); + break; + case SYS_FAN_MAX: + /* only valid for fan[5-6] */ + res = FAN_MAX_FROM_REG(data->fan_max[ix - 4]); + break; + case SYS_FAN_ALARM: + res = (data->alarms >> DME1737_BIT_ALARM_FAN[ix]) & 0x01; + break; + case SYS_FAN_TYPE: + /* only valid for fan[1-4] */ + res = FAN_TYPE_FROM_REG(data->fan_opt[ix]); + break; + default: + res = 0; + dev_dbg(dev, "Unknown attr fetch (%d)\n", fn); + } + + return sprintf(buf, "%d\n", res); +} + +static ssize_t set_fan(struct device *dev, struct device_attribute *attr, + const char *buf, size_t count) +{ + struct i2c_client *client = to_i2c_client(dev); + struct dme1737_data *data = i2c_get_clientdata(client); + struct sensor_device_attribute_2 + *sensor_attr_2 = to_sensor_dev_attr_2(attr); + int ix = sensor_attr_2->index; + int fn = sensor_attr_2->nr; + long val = simple_strtol(buf, NULL, 10); + + mutex_lock(&data->update_lock); + switch (fn) { + case SYS_FAN_MIN: + if (ix < 4) { + data->fan_min[ix] = FAN_TO_REG(val, 0); + } else { + /* Refresh the cache */ + data->fan_opt[ix] = dme1737_read(client, + DME1737_REG_FAN_OPT(ix)); + /* Modify the fan min value */ + data->fan_min[ix] = FAN_TO_REG(val, + FAN_TPC_FROM_REG(data->fan_opt[ix])); + } + dme1737_write(client, DME1737_REG_FAN_MIN(ix), + data->fan_min[ix] & 0xff); + dme1737_write(client, DME1737_REG_FAN_MIN(ix) + 1, + data->fan_min[ix] >> 8); + break; + case SYS_FAN_MAX: + /* Only valid for fan[5-6] */ + data->fan_max[ix - 4] = FAN_MAX_TO_REG(val); + dme1737_write(client, DME1737_REG_FAN_MAX(ix), + data->fan_max[ix - 4]); + break; + case SYS_FAN_TYPE: + /* Only valid for fan[1-4] */ + if (!(val == 1 || val == 2 || val == 4)) { + count = -EINVAL; + dev_warn(&client->dev, "Fan type value %ld not " + "supported. Choose one of 1, 2, or 4.\n", + val); + goto exit; + } + data->fan_opt[ix] = FAN_TYPE_TO_REG(val, dme1737_read(client, + DME1737_REG_FAN_OPT(ix))); + dme1737_write(client, DME1737_REG_FAN_OPT(ix), + data->fan_opt[ix]); + break; + default: + dev_dbg(dev, "Unknown attr fetch (%d)\n", fn); + } +exit: + mutex_unlock(&data->update_lock); + + return count; +} + +/* --------------------------------------------------------------------- + * PWM sysfs attributes + * ix = [0-4] + * --------------------------------------------------------------------- */ + +#define SYS_PWM 0 +#define SYS_PWM_FREQ 1 +#define SYS_PWM_ENABLE 2 +#define SYS_PWM_RAMP_RATE 3 +#define SYS_PWM_AUTO_CHANNELS_ZONE 4 +#define SYS_PWM_AUTO_PWM_MIN 5 +#define SYS_PWM_AUTO_POINT1_PWM 6 +#define SYS_PWM_AUTO_POINT2_PWM 7 + +static ssize_t show_pwm(struct device *dev, struct device_attribute *attr, + char *buf) +{ + struct dme1737_data *data = dme1737_update_device(dev); + struct sensor_device_attribute_2 + *sensor_attr_2 = to_sensor_dev_attr_2(attr); + int ix = sensor_attr_2->index; + int fn = sensor_attr_2->nr; + int res; + + switch (fn) { + case SYS_PWM: + if (PWM_EN_FROM_REG(data->pwm_config[ix]) == 0) { + res = 255; + } else { + res = data->pwm[ix]; + } + break; + case SYS_PWM_FREQ: + res = PWM_FREQ_FROM_REG(data->pwm_freq[ix]); + break; + case SYS_PWM_ENABLE: + if (ix > 3) { + res = 1; /* pwm[5-6] hard-wired to manual mode */ + } else { + res = PWM_EN_FROM_REG(data->pwm_config[ix]); + } + break; + case SYS_PWM_RAMP_RATE: + /* Only valid for pwm[1-3] */ + res = PWM_RR_FROM_REG(data->pwm_rr[ix > 0], ix); + break; + case SYS_PWM_AUTO_CHANNELS_ZONE: + /* Only valid for pwm[1-3] */ + if (PWM_EN_FROM_REG(data->pwm_config[ix]) == 2) { + res = PWM_ACZ_FROM_REG(data->pwm_config[ix]); + } else { + res = data->pwm_acz[ix]; + } + break; + case SYS_PWM_AUTO_PWM_MIN: + /* Only valid for pwm[1-3] */ + if (PWM_OFF_FROM_REG(data->pwm_rr[0], ix)) { + res = data->pwm_min[ix]; + } else { + res = 0; + } + break; + case SYS_PWM_AUTO_POINT1_PWM: + /* Only valid for pwm[1-3] */ + res = data->pwm_min[ix]; + break; + case SYS_PWM_AUTO_POINT2_PWM: + /* Only valid for pwm[1-3] */ + res = 255; /* hard-wired */ + break; + default: + res = 0; + dev_dbg(dev, "Unknown attr fetch (%d)\n", fn); + } + + return sprintf(buf, "%d\n", res); +} + +static struct attribute *dme1737_attr_pwm[]; +static void dme1737_chmod_file(struct i2c_client*, struct attribute*, mode_t); + +static ssize_t set_pwm(struct device *dev, struct device_attribute *attr, + const char *buf, size_t count) +{ + struct i2c_client *client = to_i2c_client(dev); + struct dme1737_data *data = i2c_get_clientdata(client); + struct sensor_device_attribute_2 + *sensor_attr_2 = to_sensor_dev_attr_2(attr); + int ix = sensor_attr_2->index; + int fn = sensor_attr_2->nr; + long val = simple_strtol(buf, NULL, 10); + + mutex_lock(&data->update_lock); + switch (fn) { + case SYS_PWM: + data->pwm[ix] = SENSORS_LIMIT(val, 0, 255); + dme1737_write(client, DME1737_REG_PWM(ix), data->pwm[ix]); + break; + case SYS_PWM_FREQ: + data->pwm_freq[ix] = PWM_FREQ_TO_REG(val, dme1737_read(client, + DME1737_REG_PWM_FREQ(ix))); + dme1737_write(client, DME1737_REG_PWM_FREQ(ix), + data->pwm_freq[ix]); + break; + case SYS_PWM_ENABLE: + /* Only valid for pwm[1-3] */ + if (val < 0 || val > 2) { + count = -EINVAL; + dev_warn(&client->dev, "PWM enable %ld not " + "supported. Choose one of 0, 1, or 2.\n", + val); + goto exit; + } + /* Refresh the cache */ + data->pwm_config[ix] = dme1737_read(client, + DME1737_REG_PWM_CONFIG(ix)); + if (val == PWM_EN_FROM_REG(data->pwm_config[ix])) { + /* Bail out if no change */ + goto exit; + } + /* Do some housekeeping if we are currently in auto mode */ + if (PWM_EN_FROM_REG(data->pwm_config[ix]) == 2) { + /* Save the current zone channel assignment */ + data->pwm_acz[ix] = PWM_ACZ_FROM_REG( + data->pwm_config[ix]); + /* Save the current ramp rate state and disable it */ + data->pwm_rr[ix > 0] = dme1737_read(client, + DME1737_REG_PWM_RR(ix > 0)); + data->pwm_rr_en &= ~(1 << ix); + if (PWM_RR_EN_FROM_REG(data->pwm_rr[ix > 0], ix)) { + data->pwm_rr_en |= (1 << ix); + data->pwm_rr[ix > 0] = PWM_RR_EN_TO_REG(0, ix, + data->pwm_rr[ix > 0]); + dme1737_write(client, + DME1737_REG_PWM_RR(ix > 0), + data->pwm_rr[ix > 0]); + } + } + /* Set the new PWM mode */ + switch (val) { + case 0: + /* Change permissions of pwm[ix] to read-only */ + dme1737_chmod_file(client, dme1737_attr_pwm[ix], + S_IRUGO); + /* Turn fan fully on */ + data->pwm_config[ix] = PWM_EN_TO_REG(0, + data->pwm_config[ix]); + dme1737_write(client, DME1737_REG_PWM_CONFIG(ix), + data->pwm_config[ix]); + break; + case 1: + /* Turn on manual mode */ + data->pwm_config[ix] = PWM_EN_TO_REG(1, + data->pwm_config[ix]); + dme1737_write(client, DME1737_REG_PWM_CONFIG(ix), + data->pwm_config[ix]); + /* Change permissions of pwm[ix] to read-writeable */ + dme1737_chmod_file(client, dme1737_attr_pwm[ix], + S_IRUGO | S_IWUSR); + break; + case 2: + /* Change permissions of pwm[ix] to read-only */ + dme1737_chmod_file(client, dme1737_attr_pwm[ix], + S_IRUGO); + /* Turn on auto mode using the saved zone channel + * assignment */ + data->pwm_config[ix] = PWM_ACZ_TO_REG( + data->pwm_acz[ix], + data->pwm_config[ix]); + dme1737_write(client, DME1737_REG_PWM_CONFIG(ix), + data->pwm_config[ix]); + /* Enable PWM ramp rate if previously enabled */ + if (data->pwm_rr_en & (1 << ix)) { + data->pwm_rr[ix > 0] = PWM_RR_EN_TO_REG(1, ix, + dme1737_read(client, + DME1737_REG_PWM_RR(ix > 0))); + dme1737_write(client, + DME1737_REG_PWM_RR(ix > 0), + data->pwm_rr[ix > 0]); + } + break; + } + break; + case SYS_PWM_RAMP_RATE: + /* Only valid for pwm[1-3] */ + /* Refresh the cache */ + data->pwm_config[ix] = dme1737_read(client, + DME1737_REG_PWM_CONFIG(ix)); + data->pwm_rr[ix > 0] = dme1737_read(client, + DME1737_REG_PWM_RR(ix > 0)); + /* Set the ramp rate value */ + if (val > 0) { + data->pwm_rr[ix > 0] = PWM_RR_TO_REG(val, ix, + data->pwm_rr[ix > 0]); + } + /* Enable/disable the feature only if the associated PWM + * output is in automatic mode. */ + if (PWM_EN_FROM_REG(data->pwm_config[ix]) == 2) { + data->pwm_rr[ix > 0] = PWM_RR_EN_TO_REG(val > 0, ix, + data->pwm_rr[ix > 0]); + } + dme1737_write(client, DME1737_REG_PWM_RR(ix > 0), + data->pwm_rr[ix > 0]); + break; + case SYS_PWM_AUTO_CHANNELS_ZONE: + /* Only valid for pwm[1-3] */ + if (!(val == 1 || val == 2 || val == 4 || + val == 6 || val == 7)) { + count = -EINVAL; + dev_warn(&client->dev, "PWM auto channels zone %ld " + "not supported. Choose one of 1, 2, 4, 6, " + "or 7.\n", val); + goto exit; + } + /* Refresh the cache */ + data->pwm_config[ix] = dme1737_read(client, + DME1737_REG_PWM_CONFIG(ix)); + if (PWM_EN_FROM_REG(data->pwm_config[ix]) == 2) { + /* PWM is already in auto mode so update the temp + * channel assignment */ + data->pwm_config[ix] = PWM_ACZ_TO_REG(val, + data->pwm_config[ix]); + dme1737_write(client, DME1737_REG_PWM_CONFIG(ix), + data->pwm_config[ix]); + } else { + /* PWM is not in auto mode so we save the temp + * channel assignment for later use */ + data->pwm_acz[ix] = val; + } + break; + case SYS_PWM_AUTO_PWM_MIN: + /* Only valid for pwm[1-3] */ + /* Refresh the cache */ + data->pwm_min[ix] = dme1737_read(client, + DME1737_REG_PWM_MIN(ix)); + /* There are only 2 values supported for the auto_pwm_min + * value: 0 or auto_point1_pwm. So if the temperature drops + * below the auto_point1_temp_hyst value, the fan either turns + * off or runs at auto_point1_pwm duty-cycle. */ + if (val > ((data->pwm_min[ix] + 1) / 2)) { + data->pwm_rr[0] = PWM_OFF_TO_REG(1, ix, + dme1737_read(client, + DME1737_REG_PWM_RR(0))); + + } else { + data->pwm_rr[0] = PWM_OFF_TO_REG(0, ix, + dme1737_read(client, + DME1737_REG_PWM_RR(0))); + + } + dme1737_write(client, DME1737_REG_PWM_RR(0), + data->pwm_rr[0]); + break; + case SYS_PWM_AUTO_POINT1_PWM: + /* Only valid for pwm[1-3] */ + data->pwm_min[ix] = SENSORS_LIMIT(val, 0, 255); + dme1737_write(client, DME1737_REG_PWM_MIN(ix), + data->pwm_min[ix]); + break; + default: + dev_dbg(dev, "Unknown attr fetch (%d)\n", fn); + } +exit: + mutex_unlock(&data->update_lock); + + return count; +} + +/* --------------------------------------------------------------------- + * Miscellaneous sysfs attributes + * --------------------------------------------------------------------- */ + +static ssize_t show_vrm(struct device *dev, struct device_attribute *attr, + char *buf) +{ + struct i2c_client *client = to_i2c_client(dev); + struct dme1737_data *data = i2c_get_clientdata(client); + + return sprintf(buf, "%d\n", data->vrm); +} + +static ssize_t set_vrm(struct device *dev, struct device_attribute *attr, + const char *buf, size_t count) +{ + struct i2c_client *client = to_i2c_client(dev); + struct dme1737_data *data = i2c_get_clientdata(client); + long val = simple_strtol(buf, NULL, 10); + + data->vrm = val; + return count; +} + +static ssize_t show_vid(struct device *dev, struct device_attribute *attr, + char *buf) +{ + struct dme1737_data *data = dme1737_update_device(dev); + + return sprintf(buf, "%d\n", vid_from_reg(data->vid, data->vrm)); +} + +/* --------------------------------------------------------------------- + * Sysfs device attribute defines and structs + * --------------------------------------------------------------------- */ + +/* Voltages 0-6 */ + +#define SENSOR_DEVICE_ATTR_IN(ix) \ +static SENSOR_DEVICE_ATTR_2(in##ix##_input, S_IRUGO, \ + show_in, NULL, SYS_IN_INPUT, ix); \ +static SENSOR_DEVICE_ATTR_2(in##ix##_min, S_IRUGO | S_IWUSR, \ + show_in, set_in, SYS_IN_MIN, ix); \ +static SENSOR_DEVICE_ATTR_2(in##ix##_max, S_IRUGO | S_IWUSR, \ + show_in, set_in, SYS_IN_MAX, ix); \ +static SENSOR_DEVICE_ATTR_2(in##ix##_alarm, S_IRUGO, \ + show_in, NULL, SYS_IN_ALARM, ix) + +SENSOR_DEVICE_ATTR_IN(0); +SENSOR_DEVICE_ATTR_IN(1); +SENSOR_DEVICE_ATTR_IN(2); +SENSOR_DEVICE_ATTR_IN(3); +SENSOR_DEVICE_ATTR_IN(4); +SENSOR_DEVICE_ATTR_IN(5); +SENSOR_DEVICE_ATTR_IN(6); + +/* Temperatures 1-3 */ + +#define SENSOR_DEVICE_ATTR_TEMP(ix) \ +static SENSOR_DEVICE_ATTR_2(temp##ix##_input, S_IRUGO, \ + show_temp, NULL, SYS_TEMP_INPUT, ix-1); \ +static SENSOR_DEVICE_ATTR_2(temp##ix##_min, S_IRUGO | S_IWUSR, \ + show_temp, set_temp, SYS_TEMP_MIN, ix-1); \ +static SENSOR_DEVICE_ATTR_2(temp##ix##_max, S_IRUGO | S_IWUSR, \ + show_temp, set_temp, SYS_TEMP_MAX, ix-1); \ +static SENSOR_DEVICE_ATTR_2(temp##ix##_offset, S_IRUGO, \ + show_temp, set_temp, SYS_TEMP_OFFSET, ix-1); \ +static SENSOR_DEVICE_ATTR_2(temp##ix##_alarm, S_IRUGO, \ + show_temp, NULL, SYS_TEMP_ALARM, ix-1); \ +static SENSOR_DEVICE_ATTR_2(temp##ix##_fault, S_IRUGO, \ + show_temp, NULL, SYS_TEMP_FAULT, ix-1) + +SENSOR_DEVICE_ATTR_TEMP(1); +SENSOR_DEVICE_ATTR_TEMP(2); +SENSOR_DEVICE_ATTR_TEMP(3); + +/* Zones 1-3 */ + +#define SENSOR_DEVICE_ATTR_ZONE(ix) \ +static SENSOR_DEVICE_ATTR_2(zone##ix##_auto_channels_temp, S_IRUGO, \ + show_zone, NULL, SYS_ZONE_AUTO_CHANNELS_TEMP, ix-1); \ +static SENSOR_DEVICE_ATTR_2(zone##ix##_auto_point1_temp_hyst, S_IRUGO, \ + show_zone, set_zone, SYS_ZONE_AUTO_POINT1_TEMP_HYST, ix-1); \ +static SENSOR_DEVICE_ATTR_2(zone##ix##_auto_point1_temp, S_IRUGO, \ + show_zone, set_zone, SYS_ZONE_AUTO_POINT1_TEMP, ix-1); \ +static SENSOR_DEVICE_ATTR_2(zone##ix##_auto_point2_temp, S_IRUGO, \ + show_zone, set_zone, SYS_ZONE_AUTO_POINT2_TEMP, ix-1); \ +static SENSOR_DEVICE_ATTR_2(zone##ix##_auto_point3_temp, S_IRUGO, \ + show_zone, set_zone, SYS_ZONE_AUTO_POINT3_TEMP, ix-1) + +SENSOR_DEVICE_ATTR_ZONE(1); +SENSOR_DEVICE_ATTR_ZONE(2); +SENSOR_DEVICE_ATTR_ZONE(3); + +/* Fans 1-4 */ + +#define SENSOR_DEVICE_ATTR_FAN_1TO4(ix) \ +static SENSOR_DEVICE_ATTR_2(fan##ix##_input, S_IRUGO, \ + show_fan, NULL, SYS_FAN_INPUT, ix-1); \ +static SENSOR_DEVICE_ATTR_2(fan##ix##_min, S_IRUGO | S_IWUSR, \ + show_fan, set_fan, SYS_FAN_MIN, ix-1); \ +static SENSOR_DEVICE_ATTR_2(fan##ix##_alarm, S_IRUGO, \ + show_fan, NULL, SYS_FAN_ALARM, ix-1); \ +static SENSOR_DEVICE_ATTR_2(fan##ix##_type, S_IRUGO | S_IWUSR, \ + show_fan, set_fan, SYS_FAN_TYPE, ix-1) + +SENSOR_DEVICE_ATTR_FAN_1TO4(1); +SENSOR_DEVICE_ATTR_FAN_1TO4(2); +SENSOR_DEVICE_ATTR_FAN_1TO4(3); +SENSOR_DEVICE_ATTR_FAN_1TO4(4); + +/* Fans 5-6 */ + +#define SENSOR_DEVICE_ATTR_FAN_5TO6(ix) \ +static SENSOR_DEVICE_ATTR_2(fan##ix##_input, S_IRUGO, \ + show_fan, NULL, SYS_FAN_INPUT, ix-1); \ +static SENSOR_DEVICE_ATTR_2(fan##ix##_min, S_IRUGO | S_IWUSR, \ + show_fan, set_fan, SYS_FAN_MIN, ix-1); \ +static SENSOR_DEVICE_ATTR_2(fan##ix##_alarm, S_IRUGO, \ + show_fan, NULL, SYS_FAN_ALARM, ix-1); \ +static SENSOR_DEVICE_ATTR_2(fan##ix##_max, S_IRUGO | S_IWUSR, \ + show_fan, set_fan, SYS_FAN_MAX, ix-1) + +SENSOR_DEVICE_ATTR_FAN_5TO6(5); +SENSOR_DEVICE_ATTR_FAN_5TO6(6); + +/* PWMs 1-3 */ + +#define SENSOR_DEVICE_ATTR_PWM_1TO3(ix) \ +static SENSOR_DEVICE_ATTR_2(pwm##ix, S_IRUGO, \ + show_pwm, set_pwm, SYS_PWM, ix-1); \ +static SENSOR_DEVICE_ATTR_2(pwm##ix##_freq, S_IRUGO, \ + show_pwm, set_pwm, SYS_PWM_FREQ, ix-1); \ +static SENSOR_DEVICE_ATTR_2(pwm##ix##_enable, S_IRUGO, \ + show_pwm, set_pwm, SYS_PWM_ENABLE, ix-1); \ +static SENSOR_DEVICE_ATTR_2(pwm##ix##_ramp_rate, S_IRUGO, \ + show_pwm, set_pwm, SYS_PWM_RAMP_RATE, ix-1); \ +static SENSOR_DEVICE_ATTR_2(pwm##ix##_auto_channels_zone, S_IRUGO, \ + show_pwm, set_pwm, SYS_PWM_AUTO_CHANNELS_ZONE, ix-1); \ +static SENSOR_DEVICE_ATTR_2(pwm##ix##_auto_pwm_min, S_IRUGO, \ + show_pwm, set_pwm, SYS_PWM_AUTO_PWM_MIN, ix-1); \ +static SENSOR_DEVICE_ATTR_2(pwm##ix##_auto_point1_pwm, S_IRUGO, \ + show_pwm, set_pwm, SYS_PWM_AUTO_POINT1_PWM, ix-1); \ +static SENSOR_DEVICE_ATTR_2(pwm##ix##_auto_point2_pwm, S_IRUGO, \ + show_pwm, NULL, SYS_PWM_AUTO_POINT2_PWM, ix-1) + +SENSOR_DEVICE_ATTR_PWM_1TO3(1); +SENSOR_DEVICE_ATTR_PWM_1TO3(2); +SENSOR_DEVICE_ATTR_PWM_1TO3(3); + +/* PWMs 5-6 */ + +#define SENSOR_DEVICE_ATTR_PWM_5TO6(ix) \ +static SENSOR_DEVICE_ATTR_2(pwm##ix, S_IRUGO | S_IWUSR, \ + show_pwm, set_pwm, SYS_PWM, ix-1); \ +static SENSOR_DEVICE_ATTR_2(pwm##ix##_freq, S_IRUGO | S_IWUSR, \ + show_pwm, set_pwm, SYS_PWM_FREQ, ix-1); \ +static SENSOR_DEVICE_ATTR_2(pwm##ix##_enable, S_IRUGO, \ + show_pwm, NULL, SYS_PWM_ENABLE, ix-1) + +SENSOR_DEVICE_ATTR_PWM_5TO6(5); +SENSOR_DEVICE_ATTR_PWM_5TO6(6); + +/* Misc */ + +static DEVICE_ATTR(vrm, S_IRUGO | S_IWUSR, show_vrm, set_vrm); +static DEVICE_ATTR(cpu0_vid, S_IRUGO, show_vid, NULL); + +#define SENSOR_DEV_ATTR_IN(ix) \ +&sensor_dev_attr_in##ix##_input.dev_attr.attr, \ +&sensor_dev_attr_in##ix##_min.dev_attr.attr, \ +&sensor_dev_attr_in##ix##_max.dev_attr.attr, \ +&sensor_dev_attr_in##ix##_alarm.dev_attr.attr + +/* These attributes are read-writeable only if the chip is *not* locked */ +#define SENSOR_DEV_ATTR_TEMP_LOCK(ix) \ +&sensor_dev_attr_temp##ix##_offset.dev_attr.attr + +#define SENSOR_DEV_ATTR_TEMP(ix) \ +SENSOR_DEV_ATTR_TEMP_LOCK(ix), \ +&sensor_dev_attr_temp##ix##_input.dev_attr.attr, \ +&sensor_dev_attr_temp##ix##_min.dev_attr.attr, \ +&sensor_dev_attr_temp##ix##_max.dev_attr.attr, \ +&sensor_dev_attr_temp##ix##_alarm.dev_attr.attr, \ +&sensor_dev_attr_temp##ix##_fault.dev_attr.attr + +/* These attributes are read-writeable only if the chip is *not* locked */ +#define SENSOR_DEV_ATTR_ZONE_LOCK(ix) \ +&sensor_dev_attr_zone##ix##_auto_point1_temp_hyst.dev_attr.attr, \ +&sensor_dev_attr_zone##ix##_auto_point1_temp.dev_attr.attr, \ +&sensor_dev_attr_zone##ix##_auto_point2_temp.dev_attr.attr, \ +&sensor_dev_attr_zone##ix##_auto_point3_temp.dev_attr.attr + +#define SENSOR_DEV_ATTR_ZONE(ix) \ +SENSOR_DEV_ATTR_ZONE_LOCK(ix), \ +&sensor_dev_attr_zone##ix##_auto_channels_temp.dev_attr.attr + +#define SENSOR_DEV_ATTR_FAN_1TO4(ix) \ +&sensor_dev_attr_fan##ix##_input.dev_attr.attr, \ +&sensor_dev_attr_fan##ix##_min.dev_attr.attr, \ +&sensor_dev_attr_fan##ix##_alarm.dev_attr.attr, \ +&sensor_dev_attr_fan##ix##_type.dev_attr.attr + +#define SENSOR_DEV_ATTR_FAN_5TO6(ix) \ +&sensor_dev_attr_fan##ix##_input.dev_attr.attr, \ +&sensor_dev_attr_fan##ix##_min.dev_attr.attr, \ +&sensor_dev_attr_fan##ix##_alarm.dev_attr.attr, \ +&sensor_dev_attr_fan##ix##_max.dev_attr.attr + +/* These attributes are read-writeable only if the chip is *not* locked */ +#define SENSOR_DEV_ATTR_PWM_1TO3_LOCK(ix) \ +&sensor_dev_attr_pwm##ix##_freq.dev_attr.attr, \ +&sensor_dev_attr_pwm##ix##_enable.dev_attr.attr, \ +&sensor_dev_attr_pwm##ix##_ramp_rate.dev_attr.attr, \ +&sensor_dev_attr_pwm##ix##_auto_channels_zone.dev_attr.attr, \ +&sensor_dev_attr_pwm##ix##_auto_pwm_min.dev_attr.attr, \ +&sensor_dev_attr_pwm##ix##_auto_point1_pwm.dev_attr.attr + +#define SENSOR_DEV_ATTR_PWM_1TO3(ix) \ +SENSOR_DEV_ATTR_PWM_1TO3_LOCK(ix), \ +&sensor_dev_attr_pwm##ix.dev_attr.attr, \ +&sensor_dev_attr_pwm##ix##_auto_point2_pwm.dev_attr.attr + +/* These attributes are read-writeable only if the chip is *not* locked */ +#define SENSOR_DEV_ATTR_PWM_5TO6_LOCK(ix) \ +&sensor_dev_attr_pwm##ix.dev_attr.attr, \ +&sensor_dev_attr_pwm##ix##_freq.dev_attr.attr + +#define SENSOR_DEV_ATTR_PWM_5TO6(ix) \ +SENSOR_DEV_ATTR_PWM_5TO6_LOCK(ix), \ +&sensor_dev_attr_pwm##ix##_enable.dev_attr.attr + +/* This struct holds all the attributes that are always present and need to be + * created unconditionally. The attributes that need modification of their + * permissions are created read-only and write permissions are added or removed + * on the fly when required */ +static struct attribute *dme1737_attr[] ={ + /* Voltages */ + SENSOR_DEV_ATTR_IN(0), + SENSOR_DEV_ATTR_IN(1), + SENSOR_DEV_ATTR_IN(2), + SENSOR_DEV_ATTR_IN(3), + SENSOR_DEV_ATTR_IN(4), + SENSOR_DEV_ATTR_IN(5), + SENSOR_DEV_ATTR_IN(6), + /* Temperatures */ + SENSOR_DEV_ATTR_TEMP(1), + SENSOR_DEV_ATTR_TEMP(2), + SENSOR_DEV_ATTR_TEMP(3), + /* Zones */ + SENSOR_DEV_ATTR_ZONE(1), + SENSOR_DEV_ATTR_ZONE(2), + SENSOR_DEV_ATTR_ZONE(3), + /* Misc */ + &dev_attr_vrm.attr, + &dev_attr_cpu0_vid.attr, + NULL +}; + +static const struct attribute_group dme1737_group = { + .attrs = dme1737_attr, +}; + +/* The following structs hold the PWM attributes, some of which are optional. + * Their creation depends on the chip configuration which is determined during + * module load. */ +static struct attribute *dme1737_attr_pwm1[] = { + SENSOR_DEV_ATTR_PWM_1TO3(1), + NULL +}; +static struct attribute *dme1737_attr_pwm2[] = { + SENSOR_DEV_ATTR_PWM_1TO3(2), + NULL +}; +static struct attribute *dme1737_attr_pwm3[] = { + SENSOR_DEV_ATTR_PWM_1TO3(3), + NULL +}; +static struct attribute *dme1737_attr_pwm5[] = { + SENSOR_DEV_ATTR_PWM_5TO6(5), + NULL +}; +static struct attribute *dme1737_attr_pwm6[] = { + SENSOR_DEV_ATTR_PWM_5TO6(6), + NULL +}; + +static const struct attribute_group dme1737_pwm_group[] = { + { .attrs = dme1737_attr_pwm1 }, + { .attrs = dme1737_attr_pwm2 }, + { .attrs = dme1737_attr_pwm3 }, + { .attrs = NULL }, + { .attrs = dme1737_attr_pwm5 }, + { .attrs = dme1737_attr_pwm6 }, +}; + +/* The following structs hold the fan attributes, some of which are optional. + * Their creation depends on the chip configuration which is determined during + * module load. */ +static struct attribute *dme1737_attr_fan1[] = { + SENSOR_DEV_ATTR_FAN_1TO4(1), + NULL +}; +static struct attribute *dme1737_attr_fan2[] = { + SENSOR_DEV_ATTR_FAN_1TO4(2), + NULL +}; +static struct attribute *dme1737_attr_fan3[] = { + SENSOR_DEV_ATTR_FAN_1TO4(3), + NULL +}; +static struct attribute *dme1737_attr_fan4[] = { + SENSOR_DEV_ATTR_FAN_1TO4(4), + NULL +}; +static struct attribute *dme1737_attr_fan5[] = { + SENSOR_DEV_ATTR_FAN_5TO6(5), + NULL +}; +static struct attribute *dme1737_attr_fan6[] = { + SENSOR_DEV_ATTR_FAN_5TO6(6), + NULL +}; + +static const struct attribute_group dme1737_fan_group[] = { + { .attrs = dme1737_attr_fan1 }, + { .attrs = dme1737_attr_fan2 }, + { .attrs = dme1737_attr_fan3 }, + { .attrs = dme1737_attr_fan4 }, + { .attrs = dme1737_attr_fan5 }, + { .attrs = dme1737_attr_fan6 }, +}; + +/* The permissions of all of the following attributes are changed to read- + * writeable if the chip is *not* locked. Otherwise they stay read-only. */ +static struct attribute *dme1737_attr_lock[] = { + /* Temperatures */ + SENSOR_DEV_ATTR_TEMP_LOCK(1), + SENSOR_DEV_ATTR_TEMP_LOCK(2), + SENSOR_DEV_ATTR_TEMP_LOCK(3), + /* Zones */ + SENSOR_DEV_ATTR_ZONE_LOCK(1), + SENSOR_DEV_ATTR_ZONE_LOCK(2), + SENSOR_DEV_ATTR_ZONE_LOCK(3), + NULL +}; + +static const struct attribute_group dme1737_lock_group = { + .attrs = dme1737_attr_lock, +}; + +/* The permissions of the following PWM attributes are changed to read- + * writeable if the chip is *not* locked and the respective PWM is available. + * Otherwise they stay read-only. */ +static struct attribute *dme1737_attr_pwm1_lock[] = { + SENSOR_DEV_ATTR_PWM_1TO3_LOCK(1), + NULL +}; +static struct attribute *dme1737_attr_pwm2_lock[] = { + SENSOR_DEV_ATTR_PWM_1TO3_LOCK(2), + NULL +}; +static struct attribute *dme1737_attr_pwm3_lock[] = { + SENSOR_DEV_ATTR_PWM_1TO3_LOCK(3), + NULL +}; +static struct attribute *dme1737_attr_pwm5_lock[] = { + SENSOR_DEV_ATTR_PWM_5TO6_LOCK(5), + NULL +}; +static struct attribute *dme1737_attr_pwm6_lock[] = { + SENSOR_DEV_ATTR_PWM_5TO6_LOCK(6), + NULL +}; + +static const struct attribute_group dme1737_pwm_lock_group[] = { + { .attrs = dme1737_attr_pwm1_lock }, + { .attrs = dme1737_attr_pwm2_lock }, + { .attrs = dme1737_attr_pwm3_lock }, + { .attrs = NULL }, + { .attrs = dme1737_attr_pwm5_lock }, + { .attrs = dme1737_attr_pwm6_lock }, +}; + +/* Pwm[1-3] are read-writeable if the associated pwm is in manual mode and the + * chip is not locked. Otherwise they are read-only. */ +static struct attribute *dme1737_attr_pwm[] = { + &sensor_dev_attr_pwm1.dev_attr.attr, + &sensor_dev_attr_pwm2.dev_attr.attr, + &sensor_dev_attr_pwm3.dev_attr.attr, +}; + +/* --------------------------------------------------------------------- + * Super-IO functions + * --------------------------------------------------------------------- */ + +static inline int dme1737_sio_inb(int sio_cip, int reg) +{ + outb(reg, sio_cip); + return inb(sio_cip + 1); +} + +static inline void dme1737_sio_outb(int sio_cip, int reg, int val) +{ + outb(reg, sio_cip); + outb(val, sio_cip + 1); +} + +static int dme1737_sio_get_features(int sio_cip, struct i2c_client *client) +{ + struct dme1737_data *data = i2c_get_clientdata(client); + int err = 0, reg; + u16 addr; + + /* Enter configuration mode */ + outb(0x55, sio_cip); + + /* Check device ID + * The DME1737 can return either 0x78 or 0x77 as its device ID. */ + reg = dme1737_sio_inb(sio_cip, 0x20); + if (!(reg == 0x77 || reg == 0x78)) { + err = -ENODEV; + goto exit; + } + + /* Select logical device A (runtime registers) */ + dme1737_sio_outb(sio_cip, 0x07, 0x0a); + + /* Get the base address of the runtime registers */ + if (!(addr = (dme1737_sio_inb(sio_cip, 0x60) << 8) | + dme1737_sio_inb(sio_cip, 0x61))) { + err = -ENODEV; + goto exit; + } + + /* Read the runtime registers to determine which optional features + * are enabled and available. Bits [3:2] of registers 0x43-0x46 are set + * to '10' if the respective feature is enabled. */ + if ((inb(addr + 0x43) & 0x0c) == 0x08) { /* fan6 */ + data->has_fan |= (1 << 5); + } + if ((inb(addr + 0x44) & 0x0c) == 0x08) { /* pwm6 */ + data->has_pwm |= (1 << 5); + } + if ((inb(addr + 0x45) & 0x0c) == 0x08) { /* fan5 */ + data->has_fan |= (1 << 4); + } + if ((inb(addr + 0x46) & 0x0c) == 0x08) { /* pwm5 */ + data->has_pwm |= (1 << 4); + } + +exit: + /* Exit configuration mode */ + outb(0xaa, sio_cip); + + return err; +} + +/* --------------------------------------------------------------------- + * Device detection, registration and initialization + * --------------------------------------------------------------------- */ + +static struct i2c_driver dme1737_driver; + +static void dme1737_chmod_file(struct i2c_client *client, + struct attribute *attr, mode_t mode) +{ + if (sysfs_chmod_file(&client->dev.kobj, attr, mode)) { + dev_warn(&client->dev, "Failed to change permissions of %s.\n", + attr->name); + } +} + +static void dme1737_chmod_group(struct i2c_client *client, + const struct attribute_group *group, + mode_t mode) +{ + struct attribute **attr; + + for (attr = group->attrs; *attr; attr++) { + dme1737_chmod_file(client, *attr, mode); + } +} + +static int dme1737_init_client(struct i2c_client *client) +{ + struct dme1737_data *data = i2c_get_clientdata(client); + int ix; + u8 reg; + + data->config = dme1737_read(client, DME1737_REG_CONFIG); + /* Inform if part is not monitoring/started */ + if (!(data->config & 0x01)) { + if (!force_start) { + dev_err(&client->dev, "Device is not monitoring. " + "Use the force_start load parameter to " + "override.\n"); + return -EFAULT; + } + + /* Force monitoring */ + data->config |= 0x01; + dme1737_write(client, DME1737_REG_CONFIG, data->config); + } + /* Inform if part is not ready */ + if (!(data->config & 0x04)) { + dev_err(&client->dev, "Device is not ready.\n"); + return -EFAULT; + } + + data->config2 = dme1737_read(client, DME1737_REG_CONFIG2); + /* Check if optional fan3 input is enabled */ + if (data->config2 & 0x04) { + data->has_fan |= (1 << 2); + } + + /* Fan4 and pwm3 are only available if the client's I2C address + * is the default 0x2e. Otherwise the I/Os associated with these + * functions are used for addr enable/select. */ + if (client->addr == 0x2e) { + data->has_fan |= (1 << 3); + data->has_pwm |= (1 << 2); + } + + /* Determine if the optional fan[5-6] and/or pwm[5-6] are enabled. + * For this, we need to query the runtime registers through the + * Super-IO LPC interface. Try both config ports 0x2e and 0x4e. */ + if (dme1737_sio_get_features(0x2e, client) && + dme1737_sio_get_features(0x4e, client)) { + dev_warn(&client->dev, "Failed to query Super-IO for optional " + "features.\n"); + } + + /* Fan1, fan2, pwm1, and pwm2 are always present */ + data->has_fan |= 0x03; + data->has_pwm |= 0x03; + + dev_info(&client->dev, "Optional features: pwm3=%s, pwm5=%s, pwm6=%s, " + "fan3=%s, fan4=%s, fan5=%s, fan6=%s.\n", + (data->has_pwm & (1 << 2)) ? "yes" : "no", + (data->has_pwm & (1 << 4)) ? "yes" : "no", + (data->has_pwm & (1 << 5)) ? "yes" : "no", + (data->has_fan & (1 << 2)) ? "yes" : "no", + (data->has_fan & (1 << 3)) ? "yes" : "no", + (data->has_fan & (1 << 4)) ? "yes" : "no", + (data->has_fan & (1 << 5)) ? "yes" : "no"); + + reg = dme1737_read(client, DME1737_REG_TACH_PWM); + /* Inform if fan-to-pwm mapping differs from the default */ + if (reg != 0xa4) { + dev_warn(&client->dev, "Non-standard fan to pwm mapping: " + "fan1->pwm%d, fan2->pwm%d, fan3->pwm%d, " + "fan4->pwm%d. Please report to the driver " + "maintainer.\n", + (reg & 0x03) + 1, ((reg >> 2) & 0x03) + 1, + ((reg >> 4) & 0x03) + 1, ((reg >> 6) & 0x03) + 1); + } + + /* Switch pwm[1-3] to manual mode if they are currently disabled and + * set the duty-cycles to 0% (which is identical to the PWMs being + * disabled). */ + if (!(data->config & 0x02)) { + for (ix = 0; ix < 3; ix++) { + data->pwm_config[ix] = dme1737_read(client, + DME1737_REG_PWM_CONFIG(ix)); + if ((data->has_pwm & (1 << ix)) && + (PWM_EN_FROM_REG(data->pwm_config[ix]) == -1)) { + dev_info(&client->dev, "Switching pwm%d to " + "manual mode.\n", ix + 1); + data->pwm_config[ix] = PWM_EN_TO_REG(1, + data->pwm_config[ix]); + dme1737_write(client, DME1737_REG_PWM(ix), 0); + dme1737_write(client, + DME1737_REG_PWM_CONFIG(ix), + data->pwm_config[ix]); + } + } + } + + /* Initialize the default PWM auto channels zone (acz) assignments */ + data->pwm_acz[0] = 1; /* pwm1 -> zone1 */ + data->pwm_acz[1] = 2; /* pwm2 -> zone2 */ + data->pwm_acz[2] = 4; /* pwm3 -> zone3 */ + + /* Set VRM */ + data->vrm = vid_which_vrm(); + + return 0; +} + +static int dme1737_detect(struct i2c_adapter *adapter, int address, + int kind) +{ + u8 company, verstep = 0; + struct i2c_client *client; + struct dme1737_data *data; + int ix, err = 0; + const char *name; + + if (!i2c_check_functionality(adapter, I2C_FUNC_SMBUS_BYTE_DATA)) { + goto exit; + } + + if (!(data = kzalloc(sizeof(struct dme1737_data), GFP_KERNEL))) { + err = -ENOMEM; + goto exit; + } + + client = &data->client; + i2c_set_clientdata(client, data); + client->addr = address; + client->adapter = adapter; + client->driver = &dme1737_driver; + + /* A negative kind means that the driver was loaded with no force + * parameter (default), so we must identify the chip. */ + if (kind < 0) { + company = dme1737_read(client, DME1737_REG_COMPANY); + verstep = dme1737_read(client, DME1737_REG_VERSTEP); + + if (!((company == DME1737_COMPANY_SMSC) && + ((verstep & DME1737_VERSTEP_MASK) == DME1737_VERSTEP))) { + err = -ENODEV; + goto exit_kfree; + } + } + + kind = dme1737; + name = "dme1737"; + + /* Fill in the remaining client fields and put it into the global + * list */ + strlcpy(client->name, name, I2C_NAME_SIZE); + mutex_init(&data->update_lock); + + /* Tell the I2C layer a new client has arrived */ + if ((err = i2c_attach_client(client))) { + goto exit_kfree; + } + + /* Initialize the DME1737 chip */ + if ((err = dme1737_init_client(client))) { + goto exit_detach; + } + + /* Create standard sysfs attributes */ + if ((err = sysfs_create_group(&client->dev.kobj, &dme1737_group))) { + goto exit_detach; + } + + /* Create fan sysfs attributes */ + for (ix = 0; ix < ARRAY_SIZE(dme1737_fan_group); ix++) { + if (data->has_fan & (1 << ix)) { + if ((err = sysfs_create_group(&client->dev.kobj, + &dme1737_fan_group[ix]))) { + goto exit_remove; + } + } + } + + /* Create PWM sysfs attributes */ + for (ix = 0; ix < ARRAY_SIZE(dme1737_pwm_group); ix++) { + if (data->has_pwm & (1 << ix)) { + if ((err = sysfs_create_group(&client->dev.kobj, + &dme1737_pwm_group[ix]))) { + goto exit_remove; + } + } + } + + /* Inform if the device is locked. Otherwise change the permissions of + * selected attributes from read-only to read-writeable. */ + if (data->config & 0x02) { + dev_info(&client->dev, "Device is locked. Some attributes " + "will be read-only.\n"); + } else { + /* Change permissions of standard attributes */ + dme1737_chmod_group(client, &dme1737_lock_group, + S_IRUGO | S_IWUSR); + + /* Change permissions of PWM attributes */ + for (ix = 0; ix < ARRAY_SIZE(dme1737_pwm_lock_group); ix++) { + if (data->has_pwm & (1 << ix)) { + dme1737_chmod_group(client, + &dme1737_pwm_lock_group[ix], + S_IRUGO | S_IWUSR); + } + } + + /* Change permissions of pwm[1-3] if in manual mode */ + for (ix = 0; ix < 3; ix++) { + if ((data->has_pwm & (1 << ix)) && + (PWM_EN_FROM_REG(data->pwm_config[ix]) == 1)) { + dme1737_chmod_file(client, + dme1737_attr_pwm[ix], + S_IRUGO | S_IWUSR); + } + } + } + + /* Register device */ + data->class_dev = hwmon_device_register(&client->dev); + if (IS_ERR(data->class_dev)) { + err = PTR_ERR(data->class_dev); + goto exit_remove; + } + + dev_info(&adapter->dev, "Found a DME1737 chip at 0x%02x " + "(rev 0x%02x)\n", client->addr, verstep); + + return 0; + +exit_remove: + for (ix = 0; ix < ARRAY_SIZE(dme1737_fan_group); ix++) { + if (data->has_fan & (1 << ix)) { + sysfs_remove_group(&client->dev.kobj, + &dme1737_fan_group[ix]); + } + } + for (ix = 0; ix < ARRAY_SIZE(dme1737_pwm_group); ix++) { + if (data->has_pwm & (1 << ix)) { + sysfs_remove_group(&client->dev.kobj, + &dme1737_pwm_group[ix]); + } + } + sysfs_remove_group(&client->dev.kobj, &dme1737_group); +exit_detach: + i2c_detach_client(client); +exit_kfree: + kfree(data); +exit: + return err; +} + +static int dme1737_attach_adapter(struct i2c_adapter *adapter) +{ + if (!(adapter->class & I2C_CLASS_HWMON)) { + return 0; + } + + return i2c_probe(adapter, &addr_data, dme1737_detect); +} + +static int dme1737_detach_client(struct i2c_client *client) +{ + struct dme1737_data *data = i2c_get_clientdata(client); + int ix, err; + + hwmon_device_unregister(data->class_dev); + + for (ix = 0; ix < ARRAY_SIZE(dme1737_fan_group); ix++) { + if (data->has_fan & (1 << ix)) { + sysfs_remove_group(&client->dev.kobj, + &dme1737_fan_group[ix]); + } + } + for (ix = 0; ix < ARRAY_SIZE(dme1737_pwm_group); ix++) { + if (data->has_pwm & (1 << ix)) { + sysfs_remove_group(&client->dev.kobj, + &dme1737_pwm_group[ix]); + } + } + sysfs_remove_group(&client->dev.kobj, &dme1737_group); + + if ((err = i2c_detach_client(client))) { + return err; + } + + kfree(data); + return 0; +} + +static struct i2c_driver dme1737_driver = { + .driver = { + .name = "dme1737", + }, + .attach_adapter = dme1737_attach_adapter, + .detach_client = dme1737_detach_client, +}; + +static int __init dme1737_init(void) +{ + return i2c_add_driver(&dme1737_driver); +} + +static void __exit dme1737_exit(void) +{ + i2c_del_driver(&dme1737_driver); +} + +MODULE_AUTHOR("Juerg Haefliger <juergh@gmail.com>"); +MODULE_DESCRIPTION("DME1737 sensors"); +MODULE_LICENSE("GPL"); + +module_init(dme1737_init); +module_exit(dme1737_exit); diff --git a/drivers/hwmon/ds1621.c b/drivers/hwmon/ds1621.c index d5ac422d73b..1212d6b7f31 100644 --- a/drivers/hwmon/ds1621.c +++ b/drivers/hwmon/ds1621.c @@ -27,6 +27,7 @@ #include <linux/jiffies.h> #include <linux/i2c.h> #include <linux/hwmon.h> +#include <linux/hwmon-sysfs.h> #include <linux/err.h> #include <linux/mutex.h> #include <linux/sysfs.h> @@ -52,9 +53,11 @@ MODULE_PARM_DESC(polarity, "Output's polarity: 0 = active high, 1 = active low") #define DS1621_REG_CONFIG_DONE 0x80 /* The DS1621 registers */ -#define DS1621_REG_TEMP 0xAA /* word, RO */ -#define DS1621_REG_TEMP_MIN 0xA2 /* word, RW */ -#define DS1621_REG_TEMP_MAX 0xA1 /* word, RW */ +static const u8 DS1621_REG_TEMP[3] = { + 0xAA, /* input, word, RO */ + 0xA2, /* min, word, RW */ + 0xA1, /* max, word, RW */ +}; #define DS1621_REG_CONF 0xAC /* byte, RW */ #define DS1621_COM_START 0xEE /* no data */ #define DS1621_COM_STOP 0x22 /* no data */ @@ -63,10 +66,7 @@ MODULE_PARM_DESC(polarity, "Output's polarity: 0 = active high, 1 = active low") #define DS1621_ALARM_TEMP_HIGH 0x40 #define DS1621_ALARM_TEMP_LOW 0x20 -/* Conversions. Rounding and limit checking is only done on the TO_REG - variants. Note that you should be a bit careful with which arguments - these macros are called: arguments may be evaluated more than once. - Fixing this is just not worth it. */ +/* Conversions */ #define ALARMS_FROM_REG(val) ((val) & \ (DS1621_ALARM_TEMP_HIGH | DS1621_ALARM_TEMP_LOW)) @@ -78,7 +78,7 @@ struct ds1621_data { char valid; /* !=0 if following fields are valid */ unsigned long last_updated; /* In jiffies */ - u16 temp, temp_min, temp_max; /* Register values, word */ + u16 temp[3]; /* Register values, word */ u8 conf; /* Register encoding, combined */ }; @@ -101,7 +101,7 @@ static struct i2c_driver ds1621_driver = { /* All registers are word-sized, except for the configuration register. DS1621 uses a high-byte first convention, which is exactly opposite to - the usual practice. */ + the SMBus standard. */ static int ds1621_read_value(struct i2c_client *client, u8 reg) { if (reg == DS1621_REG_CONF) @@ -110,9 +110,6 @@ static int ds1621_read_value(struct i2c_client *client, u8 reg) return swab16(i2c_smbus_read_word_data(client, reg)); } -/* All registers are word-sized, except for the configuration register. - DS1621 uses a high-byte first convention, which is exactly opposite to - the usual practice. */ static int ds1621_write_value(struct i2c_client *client, u8 reg, u16 value) { if (reg == DS1621_REG_CONF) @@ -139,50 +136,61 @@ static void ds1621_init_client(struct i2c_client *client) i2c_smbus_write_byte(client, DS1621_COM_START); } -#define show(value) \ -static ssize_t show_##value(struct device *dev, struct device_attribute *attr, char *buf) \ -{ \ - struct ds1621_data *data = ds1621_update_client(dev); \ - return sprintf(buf, "%d\n", LM75_TEMP_FROM_REG(data->value)); \ +static ssize_t show_temp(struct device *dev, struct device_attribute *da, + char *buf) +{ + struct sensor_device_attribute *attr = to_sensor_dev_attr(da); + struct ds1621_data *data = ds1621_update_client(dev); + return sprintf(buf, "%d\n", + LM75_TEMP_FROM_REG(data->temp[attr->index])); } -show(temp); -show(temp_min); -show(temp_max); - -#define set_temp(suffix, value, reg) \ -static ssize_t set_temp_##suffix(struct device *dev, struct device_attribute *attr, const char *buf, \ - size_t count) \ -{ \ - struct i2c_client *client = to_i2c_client(dev); \ - struct ds1621_data *data = ds1621_update_client(dev); \ - u16 val = LM75_TEMP_TO_REG(simple_strtoul(buf, NULL, 10)); \ - \ - mutex_lock(&data->update_lock); \ - data->value = val; \ - ds1621_write_value(client, reg, data->value); \ - mutex_unlock(&data->update_lock); \ - return count; \ -} +static ssize_t set_temp(struct device *dev, struct device_attribute *da, + const char *buf, size_t count) +{ + struct sensor_device_attribute *attr = to_sensor_dev_attr(da); + struct i2c_client *client = to_i2c_client(dev); + struct ds1621_data *data = ds1621_update_client(dev); + u16 val = LM75_TEMP_TO_REG(simple_strtoul(buf, NULL, 10)); -set_temp(min, temp_min, DS1621_REG_TEMP_MIN); -set_temp(max, temp_max, DS1621_REG_TEMP_MAX); + mutex_lock(&data->update_lock); + data->temp[attr->index] = val; + ds1621_write_value(client, DS1621_REG_TEMP[attr->index], + data->temp[attr->index]); + mutex_unlock(&data->update_lock); + return count; +} -static ssize_t show_alarms(struct device *dev, struct device_attribute *attr, char *buf) +static ssize_t show_alarms(struct device *dev, struct device_attribute *da, + char *buf) { struct ds1621_data *data = ds1621_update_client(dev); return sprintf(buf, "%d\n", ALARMS_FROM_REG(data->conf)); } +static ssize_t show_alarm(struct device *dev, struct device_attribute *da, + char *buf) +{ + struct sensor_device_attribute *attr = to_sensor_dev_attr(da); + struct ds1621_data *data = ds1621_update_client(dev); + return sprintf(buf, "%d\n", !!(data->conf & attr->index)); +} + static DEVICE_ATTR(alarms, S_IRUGO, show_alarms, NULL); -static DEVICE_ATTR(temp1_input, S_IRUGO , show_temp, NULL); -static DEVICE_ATTR(temp1_min, S_IWUSR | S_IRUGO , show_temp_min, set_temp_min); -static DEVICE_ATTR(temp1_max, S_IWUSR | S_IRUGO, show_temp_max, set_temp_max); +static SENSOR_DEVICE_ATTR(temp1_input, S_IRUGO, show_temp, NULL, 0); +static SENSOR_DEVICE_ATTR(temp1_min, S_IWUSR | S_IRUGO, show_temp, set_temp, 1); +static SENSOR_DEVICE_ATTR(temp1_max, S_IWUSR | S_IRUGO, show_temp, set_temp, 2); +static SENSOR_DEVICE_ATTR(temp1_min_alarm, S_IRUGO, show_alarm, NULL, + DS1621_ALARM_TEMP_LOW); +static SENSOR_DEVICE_ATTR(temp1_max_alarm, S_IRUGO, show_alarm, NULL, + DS1621_ALARM_TEMP_HIGH); static struct attribute *ds1621_attributes[] = { - &dev_attr_temp1_input.attr, - &dev_attr_temp1_min.attr, - &dev_attr_temp1_max.attr, + &sensor_dev_attr_temp1_input.dev_attr.attr, + &sensor_dev_attr_temp1_min.dev_attr.attr, + &sensor_dev_attr_temp1_max.dev_attr.attr, + &sensor_dev_attr_temp1_min_alarm.dev_attr.attr, + &sensor_dev_attr_temp1_max_alarm.dev_attr.attr, &dev_attr_alarms.attr, NULL }; @@ -204,9 +212,9 @@ static int ds1621_detect(struct i2c_adapter *adapter, int address, int kind) { int conf, temp; - struct i2c_client *new_client; + struct i2c_client *client; struct ds1621_data *data; - int err = 0; + int i, err = 0; if (!i2c_check_functionality(adapter, I2C_FUNC_SMBUS_BYTE_DATA | I2C_FUNC_SMBUS_WORD_DATA @@ -221,55 +229,44 @@ static int ds1621_detect(struct i2c_adapter *adapter, int address, goto exit; } - new_client = &data->client; - i2c_set_clientdata(new_client, data); - new_client->addr = address; - new_client->adapter = adapter; - new_client->driver = &ds1621_driver; - new_client->flags = 0; - + client = &data->client; + i2c_set_clientdata(client, data); + client->addr = address; + client->adapter = adapter; + client->driver = &ds1621_driver; /* Now, we do the remaining detection. It is lousy. */ if (kind < 0) { /* The NVB bit should be low if no EEPROM write has been requested during the latest 10ms, which is highly improbable in our case. */ - conf = ds1621_read_value(new_client, DS1621_REG_CONF); + conf = ds1621_read_value(client, DS1621_REG_CONF); if (conf & DS1621_REG_CONFIG_NVB) goto exit_free; /* The 7 lowest bits of a temperature should always be 0. */ - temp = ds1621_read_value(new_client, DS1621_REG_TEMP); - if (temp & 0x007f) - goto exit_free; - temp = ds1621_read_value(new_client, DS1621_REG_TEMP_MIN); - if (temp & 0x007f) - goto exit_free; - temp = ds1621_read_value(new_client, DS1621_REG_TEMP_MAX); - if (temp & 0x007f) - goto exit_free; + for (i = 0; i < ARRAY_SIZE(data->temp); i++) { + temp = ds1621_read_value(client, DS1621_REG_TEMP[i]); + if (temp & 0x007f) + goto exit_free; + } } - /* Determine the chip type - only one kind supported! */ - if (kind <= 0) - kind = ds1621; - /* Fill in remaining client fields and put it into the global list */ - strlcpy(new_client->name, "ds1621", I2C_NAME_SIZE); - data->valid = 0; + strlcpy(client->name, "ds1621", I2C_NAME_SIZE); mutex_init(&data->update_lock); /* Tell the I2C layer a new client has arrived */ - if ((err = i2c_attach_client(new_client))) + if ((err = i2c_attach_client(client))) goto exit_free; /* Initialize the DS1621 chip */ - ds1621_init_client(new_client); + ds1621_init_client(client); /* Register sysfs hooks */ - if ((err = sysfs_create_group(&new_client->dev.kobj, &ds1621_group))) + if ((err = sysfs_create_group(&client->dev.kobj, &ds1621_group))) goto exit_detach; - data->class_dev = hwmon_device_register(&new_client->dev); + data->class_dev = hwmon_device_register(&client->dev); if (IS_ERR(data->class_dev)) { err = PTR_ERR(data->class_dev); goto exit_remove_files; @@ -278,9 +275,9 @@ static int ds1621_detect(struct i2c_adapter *adapter, int address, return 0; exit_remove_files: - sysfs_remove_group(&new_client->dev.kobj, &ds1621_group); + sysfs_remove_group(&client->dev.kobj, &ds1621_group); exit_detach: - i2c_detach_client(new_client); + i2c_detach_client(client); exit_free: kfree(data); exit: @@ -314,23 +311,21 @@ static struct ds1621_data *ds1621_update_client(struct device *dev) if (time_after(jiffies, data->last_updated + HZ + HZ / 2) || !data->valid) { + int i; dev_dbg(&client->dev, "Starting ds1621 update\n"); data->conf = ds1621_read_value(client, DS1621_REG_CONF); - data->temp = ds1621_read_value(client, DS1621_REG_TEMP); - - data->temp_min = ds1621_read_value(client, - DS1621_REG_TEMP_MIN); - data->temp_max = ds1621_read_value(client, - DS1621_REG_TEMP_MAX); + for (i = 0; i < ARRAY_SIZE(data->temp); i++) + data->temp[i] = ds1621_read_value(client, + DS1621_REG_TEMP[i]); /* reset alarms if necessary */ new_conf = data->conf; - if (data->temp > data->temp_min) + if (data->temp[0] > data->temp[1]) /* input > min */ new_conf &= ~DS1621_ALARM_TEMP_LOW; - if (data->temp < data->temp_max) + if (data->temp[0] < data->temp[2]) /* input < max */ new_conf &= ~DS1621_ALARM_TEMP_HIGH; if (data->conf != new_conf) ds1621_write_value(client, DS1621_REG_CONF, diff --git a/drivers/hwmon/f71805f.c b/drivers/hwmon/f71805f.c index cdbe309b8fc..6f60715f34f 100644 --- a/drivers/hwmon/f71805f.c +++ b/drivers/hwmon/f71805f.c @@ -127,6 +127,13 @@ superio_exit(int base) #define F71805F_REG_TEMP_HIGH(nr) (0x54 + 2 * (nr)) #define F71805F_REG_TEMP_HYST(nr) (0x55 + 2 * (nr)) #define F71805F_REG_TEMP_MODE 0x01 +/* pwm/fan pwmnr from 0 to 2, auto point apnr from 0 to 2 */ +/* map Fintek numbers to our numbers as follows: 9->0, 5->1, 1->2 */ +#define F71805F_REG_PWM_AUTO_POINT_TEMP(pwmnr, apnr) \ + (0xA0 + 0x10 * (pwmnr) + (2 - (apnr))) +#define F71805F_REG_PWM_AUTO_POINT_FAN(pwmnr, apnr) \ + (0xA4 + 0x10 * (pwmnr) + \ + 2 * (2 - (apnr))) #define F71805F_REG_START 0x00 /* status nr from 0 to 2 */ @@ -144,6 +151,11 @@ superio_exit(int base) * Data structures and manipulation thereof */ +struct f71805f_auto_point { + u8 temp[3]; + u16 fan[3]; +}; + struct f71805f_data { unsigned short addr; const char *name; @@ -170,6 +182,7 @@ struct f71805f_data { u8 temp_hyst[3]; u8 temp_mode; unsigned long alarms; + struct f71805f_auto_point auto_points[3]; }; struct f71805f_sio_data { @@ -312,7 +325,7 @@ static void f71805f_write16(struct f71805f_data *data, u8 reg, u16 val) static struct f71805f_data *f71805f_update_device(struct device *dev) { struct f71805f_data *data = dev_get_drvdata(dev); - int nr; + int nr, apnr; mutex_lock(&data->update_lock); @@ -342,6 +355,18 @@ static struct f71805f_data *f71805f_update_device(struct device *dev) F71805F_REG_TEMP_HYST(nr)); } data->temp_mode = f71805f_read8(data, F71805F_REG_TEMP_MODE); + for (nr = 0; nr < 3; nr++) { + for (apnr = 0; apnr < 3; apnr++) { + data->auto_points[nr].temp[apnr] = + f71805f_read8(data, + F71805F_REG_PWM_AUTO_POINT_TEMP(nr, + apnr)); + data->auto_points[nr].fan[apnr] = + f71805f_read16(data, + F71805F_REG_PWM_AUTO_POINT_FAN(nr, + apnr)); + } + } data->last_limits = jiffies; } @@ -705,6 +730,70 @@ static ssize_t set_pwm_freq(struct device *dev, struct device_attribute return count; } +static ssize_t show_pwm_auto_point_temp(struct device *dev, + struct device_attribute *devattr, + char* buf) +{ + struct f71805f_data *data = dev_get_drvdata(dev); + struct sensor_device_attribute_2 *attr = to_sensor_dev_attr_2(devattr); + int pwmnr = attr->nr; + int apnr = attr->index; + + return sprintf(buf, "%ld\n", + temp_from_reg(data->auto_points[pwmnr].temp[apnr])); +} + +static ssize_t set_pwm_auto_point_temp(struct device *dev, + struct device_attribute *devattr, + const char* buf, size_t count) +{ + struct f71805f_data *data = dev_get_drvdata(dev); + struct sensor_device_attribute_2 *attr = to_sensor_dev_attr_2(devattr); + int pwmnr = attr->nr; + int apnr = attr->index; + unsigned long val = simple_strtol(buf, NULL, 10); + + mutex_lock(&data->update_lock); + data->auto_points[pwmnr].temp[apnr] = temp_to_reg(val); + f71805f_write8(data, F71805F_REG_PWM_AUTO_POINT_TEMP(pwmnr, apnr), + data->auto_points[pwmnr].temp[apnr]); + mutex_unlock(&data->update_lock); + + return count; +} + +static ssize_t show_pwm_auto_point_fan(struct device *dev, + struct device_attribute *devattr, + char* buf) +{ + struct f71805f_data *data = dev_get_drvdata(dev); + struct sensor_device_attribute_2 *attr = to_sensor_dev_attr_2(devattr); + int pwmnr = attr->nr; + int apnr = attr->index; + + return sprintf(buf, "%ld\n", + fan_from_reg(data->auto_points[pwmnr].fan[apnr])); +} + +static ssize_t set_pwm_auto_point_fan(struct device *dev, + struct device_attribute *devattr, + const char* buf, size_t count) +{ + struct f71805f_data *data = dev_get_drvdata(dev); + struct sensor_device_attribute_2 *attr = to_sensor_dev_attr_2(devattr); + int pwmnr = attr->nr; + int apnr = attr->index; + unsigned long val = simple_strtoul(buf, NULL, 10); + + mutex_lock(&data->update_lock); + data->auto_points[pwmnr].fan[apnr] = fan_to_reg(val); + f71805f_write16(data, F71805F_REG_PWM_AUTO_POINT_FAN(pwmnr, apnr), + data->auto_points[pwmnr].fan[apnr]); + mutex_unlock(&data->update_lock); + + return count; +} + static ssize_t show_temp(struct device *dev, struct device_attribute *devattr, char *buf) { @@ -932,6 +1021,63 @@ static SENSOR_DEVICE_ATTR(pwm3_freq, S_IRUGO | S_IWUSR, show_pwm_freq, set_pwm_freq, 2); static SENSOR_DEVICE_ATTR(pwm3_mode, S_IRUGO, show_pwm_mode, NULL, 2); +static SENSOR_DEVICE_ATTR_2(pwm1_auto_point1_temp, S_IRUGO | S_IWUSR, + show_pwm_auto_point_temp, set_pwm_auto_point_temp, + 0, 0); +static SENSOR_DEVICE_ATTR_2(pwm1_auto_point1_fan, S_IRUGO | S_IWUSR, + show_pwm_auto_point_fan, set_pwm_auto_point_fan, + 0, 0); +static SENSOR_DEVICE_ATTR_2(pwm1_auto_point2_temp, S_IRUGO | S_IWUSR, + show_pwm_auto_point_temp, set_pwm_auto_point_temp, + 0, 1); +static SENSOR_DEVICE_ATTR_2(pwm1_auto_point2_fan, S_IRUGO | S_IWUSR, + show_pwm_auto_point_fan, set_pwm_auto_point_fan, + 0, 1); +static SENSOR_DEVICE_ATTR_2(pwm1_auto_point3_temp, S_IRUGO | S_IWUSR, + show_pwm_auto_point_temp, set_pwm_auto_point_temp, + 0, 2); +static SENSOR_DEVICE_ATTR_2(pwm1_auto_point3_fan, S_IRUGO | S_IWUSR, + show_pwm_auto_point_fan, set_pwm_auto_point_fan, + 0, 2); + +static SENSOR_DEVICE_ATTR_2(pwm2_auto_point1_temp, S_IRUGO | S_IWUSR, + show_pwm_auto_point_temp, set_pwm_auto_point_temp, + 1, 0); +static SENSOR_DEVICE_ATTR_2(pwm2_auto_point1_fan, S_IRUGO | S_IWUSR, + show_pwm_auto_point_fan, set_pwm_auto_point_fan, + 1, 0); +static SENSOR_DEVICE_ATTR_2(pwm2_auto_point2_temp, S_IRUGO | S_IWUSR, + show_pwm_auto_point_temp, set_pwm_auto_point_temp, + 1, 1); +static SENSOR_DEVICE_ATTR_2(pwm2_auto_point2_fan, S_IRUGO | S_IWUSR, + show_pwm_auto_point_fan, set_pwm_auto_point_fan, + 1, 1); +static SENSOR_DEVICE_ATTR_2(pwm2_auto_point3_temp, S_IRUGO | S_IWUSR, + show_pwm_auto_point_temp, set_pwm_auto_point_temp, + 1, 2); +static SENSOR_DEVICE_ATTR_2(pwm2_auto_point3_fan, S_IRUGO | S_IWUSR, + show_pwm_auto_point_fan, set_pwm_auto_point_fan, + 1, 2); + +static SENSOR_DEVICE_ATTR_2(pwm3_auto_point1_temp, S_IRUGO | S_IWUSR, + show_pwm_auto_point_temp, set_pwm_auto_point_temp, + 2, 0); +static SENSOR_DEVICE_ATTR_2(pwm3_auto_point1_fan, S_IRUGO | S_IWUSR, + show_pwm_auto_point_fan, set_pwm_auto_point_fan, + 2, 0); +static SENSOR_DEVICE_ATTR_2(pwm3_auto_point2_temp, S_IRUGO | S_IWUSR, + show_pwm_auto_point_temp, set_pwm_auto_point_temp, + 2, 1); +static SENSOR_DEVICE_ATTR_2(pwm3_auto_point2_fan, S_IRUGO | S_IWUSR, + show_pwm_auto_point_fan, set_pwm_auto_point_fan, + 2, 1); +static SENSOR_DEVICE_ATTR_2(pwm3_auto_point3_temp, S_IRUGO | S_IWUSR, + show_pwm_auto_point_temp, set_pwm_auto_point_temp, + 2, 2); +static SENSOR_DEVICE_ATTR_2(pwm3_auto_point3_fan, S_IRUGO | S_IWUSR, + show_pwm_auto_point_fan, set_pwm_auto_point_fan, + 2, 2); + static SENSOR_DEVICE_ATTR(in0_alarm, S_IRUGO, show_alarm, NULL, 0); static SENSOR_DEVICE_ATTR(in1_alarm, S_IRUGO, show_alarm, NULL, 1); static SENSOR_DEVICE_ATTR(in2_alarm, S_IRUGO, show_alarm, NULL, 2); @@ -1014,6 +1160,25 @@ static struct attribute *f71805f_attributes[] = { &sensor_dev_attr_temp3_max_hyst.dev_attr.attr, &sensor_dev_attr_temp3_type.dev_attr.attr, + &sensor_dev_attr_pwm1_auto_point1_temp.dev_attr.attr, + &sensor_dev_attr_pwm1_auto_point1_fan.dev_attr.attr, + &sensor_dev_attr_pwm1_auto_point2_temp.dev_attr.attr, + &sensor_dev_attr_pwm1_auto_point2_fan.dev_attr.attr, + &sensor_dev_attr_pwm1_auto_point3_temp.dev_attr.attr, + &sensor_dev_attr_pwm1_auto_point3_fan.dev_attr.attr, + &sensor_dev_attr_pwm2_auto_point1_temp.dev_attr.attr, + &sensor_dev_attr_pwm2_auto_point1_fan.dev_attr.attr, + &sensor_dev_attr_pwm2_auto_point2_temp.dev_attr.attr, + &sensor_dev_attr_pwm2_auto_point2_fan.dev_attr.attr, + &sensor_dev_attr_pwm2_auto_point3_temp.dev_attr.attr, + &sensor_dev_attr_pwm2_auto_point3_fan.dev_attr.attr, + &sensor_dev_attr_pwm3_auto_point1_temp.dev_attr.attr, + &sensor_dev_attr_pwm3_auto_point1_fan.dev_attr.attr, + &sensor_dev_attr_pwm3_auto_point2_temp.dev_attr.attr, + &sensor_dev_attr_pwm3_auto_point2_fan.dev_attr.attr, + &sensor_dev_attr_pwm3_auto_point3_temp.dev_attr.attr, + &sensor_dev_attr_pwm3_auto_point3_fan.dev_attr.attr, + &sensor_dev_attr_in0_alarm.dev_attr.attr, &sensor_dev_attr_in1_alarm.dev_attr.attr, &sensor_dev_attr_in2_alarm.dev_attr.attr, @@ -1242,12 +1407,12 @@ static int __devexit f71805f_remove(struct platform_device *pdev) struct resource *res; int i; - platform_set_drvdata(pdev, NULL); hwmon_device_unregister(data->class_dev); sysfs_remove_group(&pdev->dev.kobj, &f71805f_group); for (i = 0; i < 4; i++) sysfs_remove_group(&pdev->dev.kobj, &f71805f_group_optin[i]); sysfs_remove_group(&pdev->dev.kobj, &f71805f_group_pwm_freq); + platform_set_drvdata(pdev, NULL); kfree(data); res = platform_get_resource(pdev, IORESOURCE_IO, 0); @@ -1290,15 +1455,12 @@ static int __init f71805f_device_add(unsigned short address, goto exit_device_put; } - pdev->dev.platform_data = kmalloc(sizeof(struct f71805f_sio_data), - GFP_KERNEL); - if (!pdev->dev.platform_data) { - err = -ENOMEM; + err = platform_device_add_data(pdev, sio_data, + sizeof(struct f71805f_sio_data)); + if (err) { printk(KERN_ERR DRVNAME ": Platform data allocation failed\n"); goto exit_device_put; } - memcpy(pdev->dev.platform_data, sio_data, - sizeof(struct f71805f_sio_data)); err = platform_device_add(pdev); if (err) { diff --git a/drivers/hwmon/it87.c b/drivers/hwmon/it87.c index 62afc63708a..eff6036e15c 100644 --- a/drivers/hwmon/it87.c +++ b/drivers/hwmon/it87.c @@ -6,6 +6,7 @@ IT8712F Super I/O chip w/LPC interface IT8716F Super I/O chip w/LPC interface IT8718F Super I/O chip w/LPC interface + IT8726F Super I/O chip w/LPC interface Sis950 A clone of the IT8705F Copyright (C) 2001 Chris Gauthron <chrisg@0-in.com> @@ -30,8 +31,7 @@ #include <linux/init.h> #include <linux/slab.h> #include <linux/jiffies.h> -#include <linux/i2c.h> -#include <linux/i2c-isa.h> +#include <linux/platform_device.h> #include <linux/hwmon.h> #include <linux/hwmon-sysfs.h> #include <linux/hwmon-vid.h> @@ -40,10 +40,12 @@ #include <linux/sysfs.h> #include <asm/io.h> +#define DRVNAME "it87" -static unsigned short isa_address; enum chips { it87, it8712, it8716, it8718 }; +static struct platform_device *pdev; + #define REG 0x2e /* The register to read/write */ #define DEV 0x07 /* Register: Logical device select */ #define VAL 0x2f /* The value to read/write */ @@ -97,6 +99,7 @@ superio_exit(void) #define IT8705F_DEVID 0x8705 #define IT8716F_DEVID 0x8716 #define IT8718F_DEVID 0x8718 +#define IT8726F_DEVID 0x8726 #define IT87_ACT_REG 0x30 #define IT87_BASE_REG 0x60 @@ -110,10 +113,6 @@ static int update_vbat; /* Not all BIOSes properly configure the PWM registers */ static int fix_pwm_polarity; -/* Values read from Super-I/O config space */ -static u16 chip_type; -static u8 vid_value; - /* Many IT87 constants specified below */ /* Length of ISA address segment */ @@ -214,13 +213,20 @@ static const unsigned int pwm_freq[8] = { }; +struct it87_sio_data { + enum chips type; + /* Values read from Super-I/O config space */ + u8 vid_value; +}; + /* For each registered chip, we need to keep some data in memory. The structure is dynamically allocated. */ struct it87_data { - struct i2c_client client; struct class_device *class_dev; enum chips type; + unsigned short addr; + const char *name; struct mutex update_lock; char valid; /* !=0 if following fields are valid */ unsigned long last_updated; /* In jiffies */ @@ -245,26 +251,25 @@ struct it87_data { }; -static int it87_detect(struct i2c_adapter *adapter); -static int it87_detach_client(struct i2c_client *client); +static int it87_probe(struct platform_device *pdev); +static int it87_remove(struct platform_device *pdev); -static int it87_read_value(struct i2c_client *client, u8 reg); -static void it87_write_value(struct i2c_client *client, u8 reg, u8 value); +static int it87_read_value(struct it87_data *data, u8 reg); +static void it87_write_value(struct it87_data *data, u8 reg, u8 value); static struct it87_data *it87_update_device(struct device *dev); -static int it87_check_pwm(struct i2c_client *client); -static void it87_init_client(struct i2c_client *client, struct it87_data *data); +static int it87_check_pwm(struct device *dev); +static void it87_init_device(struct platform_device *pdev); -static struct i2c_driver it87_isa_driver = { +static struct platform_driver it87_driver = { .driver = { .owner = THIS_MODULE, - .name = "it87-isa", + .name = DRVNAME, }, - .attach_adapter = it87_detect, - .detach_client = it87_detach_client, + .probe = it87_probe, + .remove = __devexit_p(it87_remove), }; - static ssize_t show_in(struct device *dev, struct device_attribute *attr, char *buf) { @@ -301,13 +306,12 @@ static ssize_t set_in_min(struct device *dev, struct device_attribute *attr, struct sensor_device_attribute *sensor_attr = to_sensor_dev_attr(attr); int nr = sensor_attr->index; - struct i2c_client *client = to_i2c_client(dev); - struct it87_data *data = i2c_get_clientdata(client); + struct it87_data *data = dev_get_drvdata(dev); unsigned long val = simple_strtoul(buf, NULL, 10); mutex_lock(&data->update_lock); data->in_min[nr] = IN_TO_REG(val); - it87_write_value(client, IT87_REG_VIN_MIN(nr), + it87_write_value(data, IT87_REG_VIN_MIN(nr), data->in_min[nr]); mutex_unlock(&data->update_lock); return count; @@ -318,13 +322,12 @@ static ssize_t set_in_max(struct device *dev, struct device_attribute *attr, struct sensor_device_attribute *sensor_attr = to_sensor_dev_attr(attr); int nr = sensor_attr->index; - struct i2c_client *client = to_i2c_client(dev); - struct it87_data *data = i2c_get_clientdata(client); + struct it87_data *data = dev_get_drvdata(dev); unsigned long val = simple_strtoul(buf, NULL, 10); mutex_lock(&data->update_lock); data->in_max[nr] = IN_TO_REG(val); - it87_write_value(client, IT87_REG_VIN_MAX(nr), + it87_write_value(data, IT87_REG_VIN_MAX(nr), data->in_max[nr]); mutex_unlock(&data->update_lock); return count; @@ -392,13 +395,12 @@ static ssize_t set_temp_max(struct device *dev, struct device_attribute *attr, struct sensor_device_attribute *sensor_attr = to_sensor_dev_attr(attr); int nr = sensor_attr->index; - struct i2c_client *client = to_i2c_client(dev); - struct it87_data *data = i2c_get_clientdata(client); + struct it87_data *data = dev_get_drvdata(dev); int val = simple_strtol(buf, NULL, 10); mutex_lock(&data->update_lock); data->temp_high[nr] = TEMP_TO_REG(val); - it87_write_value(client, IT87_REG_TEMP_HIGH(nr), data->temp_high[nr]); + it87_write_value(data, IT87_REG_TEMP_HIGH(nr), data->temp_high[nr]); mutex_unlock(&data->update_lock); return count; } @@ -408,13 +410,12 @@ static ssize_t set_temp_min(struct device *dev, struct device_attribute *attr, struct sensor_device_attribute *sensor_attr = to_sensor_dev_attr(attr); int nr = sensor_attr->index; - struct i2c_client *client = to_i2c_client(dev); - struct it87_data *data = i2c_get_clientdata(client); + struct it87_data *data = dev_get_drvdata(dev); int val = simple_strtol(buf, NULL, 10); mutex_lock(&data->update_lock); data->temp_low[nr] = TEMP_TO_REG(val); - it87_write_value(client, IT87_REG_TEMP_LOW(nr), data->temp_low[nr]); + it87_write_value(data, IT87_REG_TEMP_LOW(nr), data->temp_low[nr]); mutex_unlock(&data->update_lock); return count; } @@ -451,8 +452,7 @@ static ssize_t set_sensor(struct device *dev, struct device_attribute *attr, struct sensor_device_attribute *sensor_attr = to_sensor_dev_attr(attr); int nr = sensor_attr->index; - struct i2c_client *client = to_i2c_client(dev); - struct it87_data *data = i2c_get_clientdata(client); + struct it87_data *data = dev_get_drvdata(dev); int val = simple_strtol(buf, NULL, 10); mutex_lock(&data->update_lock); @@ -468,7 +468,7 @@ static ssize_t set_sensor(struct device *dev, struct device_attribute *attr, mutex_unlock(&data->update_lock); return -EINVAL; } - it87_write_value(client, IT87_REG_TEMP_ENABLE, data->sensor); + it87_write_value(data, IT87_REG_TEMP_ENABLE, data->sensor); mutex_unlock(&data->update_lock); return count; } @@ -542,13 +542,12 @@ static ssize_t set_fan_min(struct device *dev, struct device_attribute *attr, struct sensor_device_attribute *sensor_attr = to_sensor_dev_attr(attr); int nr = sensor_attr->index; - struct i2c_client *client = to_i2c_client(dev); - struct it87_data *data = i2c_get_clientdata(client); + struct it87_data *data = dev_get_drvdata(dev); int val = simple_strtol(buf, NULL, 10); u8 reg; mutex_lock(&data->update_lock); - reg = it87_read_value(client, IT87_REG_FAN_DIV); + reg = it87_read_value(data, IT87_REG_FAN_DIV); switch (nr) { case 0: data->fan_div[nr] = reg & 0x07; break; case 1: data->fan_div[nr] = (reg >> 3) & 0x07; break; @@ -556,7 +555,7 @@ static ssize_t set_fan_min(struct device *dev, struct device_attribute *attr, } data->fan_min[nr] = FAN_TO_REG(val, DIV_FROM_REG(data->fan_div[nr])); - it87_write_value(client, IT87_REG_FAN_MIN(nr), data->fan_min[nr]); + it87_write_value(data, IT87_REG_FAN_MIN(nr), data->fan_min[nr]); mutex_unlock(&data->update_lock); return count; } @@ -566,14 +565,13 @@ static ssize_t set_fan_div(struct device *dev, struct device_attribute *attr, struct sensor_device_attribute *sensor_attr = to_sensor_dev_attr(attr); int nr = sensor_attr->index; - struct i2c_client *client = to_i2c_client(dev); - struct it87_data *data = i2c_get_clientdata(client); + struct it87_data *data = dev_get_drvdata(dev); unsigned long val = simple_strtoul(buf, NULL, 10); int min; u8 old; mutex_lock(&data->update_lock); - old = it87_read_value(client, IT87_REG_FAN_DIV); + old = it87_read_value(data, IT87_REG_FAN_DIV); /* Save fan min limit */ min = FAN_FROM_REG(data->fan_min[nr], DIV_FROM_REG(data->fan_div[nr])); @@ -594,11 +592,11 @@ static ssize_t set_fan_div(struct device *dev, struct device_attribute *attr, val |= (data->fan_div[1] & 0x07) << 3; if (data->fan_div[2] == 3) val |= 0x1 << 6; - it87_write_value(client, IT87_REG_FAN_DIV, val); + it87_write_value(data, IT87_REG_FAN_DIV, val); /* Restore fan min limit */ data->fan_min[nr] = FAN_TO_REG(min, DIV_FROM_REG(data->fan_div[nr])); - it87_write_value(client, IT87_REG_FAN_MIN(nr), data->fan_min[nr]); + it87_write_value(data, IT87_REG_FAN_MIN(nr), data->fan_min[nr]); mutex_unlock(&data->update_lock); return count; @@ -609,8 +607,7 @@ static ssize_t set_pwm_enable(struct device *dev, struct sensor_device_attribute *sensor_attr = to_sensor_dev_attr(attr); int nr = sensor_attr->index; - struct i2c_client *client = to_i2c_client(dev); - struct it87_data *data = i2c_get_clientdata(client); + struct it87_data *data = dev_get_drvdata(dev); int val = simple_strtol(buf, NULL, 10); mutex_lock(&data->update_lock); @@ -618,17 +615,17 @@ static ssize_t set_pwm_enable(struct device *dev, if (val == 0) { int tmp; /* make sure the fan is on when in on/off mode */ - tmp = it87_read_value(client, IT87_REG_FAN_CTL); - it87_write_value(client, IT87_REG_FAN_CTL, tmp | (1 << nr)); + tmp = it87_read_value(data, IT87_REG_FAN_CTL); + it87_write_value(data, IT87_REG_FAN_CTL, tmp | (1 << nr)); /* set on/off mode */ data->fan_main_ctrl &= ~(1 << nr); - it87_write_value(client, IT87_REG_FAN_MAIN_CTRL, data->fan_main_ctrl); + it87_write_value(data, IT87_REG_FAN_MAIN_CTRL, data->fan_main_ctrl); } else if (val == 1) { /* set SmartGuardian mode */ data->fan_main_ctrl |= (1 << nr); - it87_write_value(client, IT87_REG_FAN_MAIN_CTRL, data->fan_main_ctrl); + it87_write_value(data, IT87_REG_FAN_MAIN_CTRL, data->fan_main_ctrl); /* set saved pwm value, clear FAN_CTLX PWM mode bit */ - it87_write_value(client, IT87_REG_PWM(nr), PWM_TO_REG(data->manual_pwm_ctl[nr])); + it87_write_value(data, IT87_REG_PWM(nr), PWM_TO_REG(data->manual_pwm_ctl[nr])); } else { mutex_unlock(&data->update_lock); return -EINVAL; @@ -643,8 +640,7 @@ static ssize_t set_pwm(struct device *dev, struct device_attribute *attr, struct sensor_device_attribute *sensor_attr = to_sensor_dev_attr(attr); int nr = sensor_attr->index; - struct i2c_client *client = to_i2c_client(dev); - struct it87_data *data = i2c_get_clientdata(client); + struct it87_data *data = dev_get_drvdata(dev); int val = simple_strtol(buf, NULL, 10); if (val < 0 || val > 255) @@ -653,15 +649,14 @@ static ssize_t set_pwm(struct device *dev, struct device_attribute *attr, mutex_lock(&data->update_lock); data->manual_pwm_ctl[nr] = val; if (data->fan_main_ctrl & (1 << nr)) - it87_write_value(client, IT87_REG_PWM(nr), PWM_TO_REG(data->manual_pwm_ctl[nr])); + it87_write_value(data, IT87_REG_PWM(nr), PWM_TO_REG(data->manual_pwm_ctl[nr])); mutex_unlock(&data->update_lock); return count; } static ssize_t set_pwm_freq(struct device *dev, struct device_attribute *attr, const char *buf, size_t count) { - struct i2c_client *client = to_i2c_client(dev); - struct it87_data *data = i2c_get_clientdata(client); + struct it87_data *data = dev_get_drvdata(dev); unsigned long val = simple_strtoul(buf, NULL, 10); int i; @@ -672,9 +667,9 @@ static ssize_t set_pwm_freq(struct device *dev, } mutex_lock(&data->update_lock); - data->fan_ctl = it87_read_value(client, IT87_REG_FAN_CTL) & 0x8f; + data->fan_ctl = it87_read_value(data, IT87_REG_FAN_CTL) & 0x8f; data->fan_ctl |= i << 4; - it87_write_value(client, IT87_REG_FAN_CTL, data->fan_ctl); + it87_write_value(data, IT87_REG_FAN_CTL, data->fan_ctl); mutex_unlock(&data->update_lock); return count; @@ -729,15 +724,14 @@ static ssize_t set_fan16_min(struct device *dev, struct device_attribute *attr, { struct sensor_device_attribute *sensor_attr = to_sensor_dev_attr(attr); int nr = sensor_attr->index; - struct i2c_client *client = to_i2c_client(dev); - struct it87_data *data = i2c_get_clientdata(client); + struct it87_data *data = dev_get_drvdata(dev); int val = simple_strtol(buf, NULL, 10); mutex_lock(&data->update_lock); data->fan_min[nr] = FAN16_TO_REG(val); - it87_write_value(client, IT87_REG_FAN_MIN(nr), + it87_write_value(data, IT87_REG_FAN_MIN(nr), data->fan_min[nr] & 0xff); - it87_write_value(client, IT87_REG_FANX_MIN(nr), + it87_write_value(data, IT87_REG_FANX_MIN(nr), data->fan_min[nr] >> 8); mutex_unlock(&data->update_lock); return count; @@ -775,8 +769,7 @@ show_vrm_reg(struct device *dev, struct device_attribute *attr, char *buf) static ssize_t store_vrm_reg(struct device *dev, struct device_attribute *attr, const char *buf, size_t count) { - struct i2c_client *client = to_i2c_client(dev); - struct it87_data *data = i2c_get_clientdata(client); + struct it87_data *data = dev_get_drvdata(dev); u32 val; val = simple_strtoul(buf, NULL, 10); @@ -794,6 +787,14 @@ show_vid_reg(struct device *dev, struct device_attribute *attr, char *buf) } static DEVICE_ATTR(cpu0_vid, S_IRUGO, show_vid_reg, NULL); +static ssize_t show_name(struct device *dev, struct device_attribute + *devattr, char *buf) +{ + struct it87_data *data = dev_get_drvdata(dev); + return sprintf(buf, "%s\n", data->name); +} +static DEVICE_ATTR(name, S_IRUGO, show_name, NULL); + static struct attribute *it87_attributes[] = { &sensor_dev_attr_in0_input.dev_attr.attr, &sensor_dev_attr_in1_input.dev_attr.attr, @@ -835,6 +836,7 @@ static struct attribute *it87_attributes[] = { &sensor_dev_attr_temp3_type.dev_attr.attr, &dev_attr_alarms.attr, + &dev_attr_name.attr, NULL }; @@ -877,17 +879,36 @@ static const struct attribute_group it87_group_opt = { }; /* SuperIO detection - will change isa_address if a chip is found */ -static int __init it87_find(unsigned short *address) +static int __init it87_find(unsigned short *address, + struct it87_sio_data *sio_data) { int err = -ENODEV; + u16 chip_type; superio_enter(); chip_type = superio_inw(DEVID); - if (chip_type != IT8712F_DEVID - && chip_type != IT8716F_DEVID - && chip_type != IT8718F_DEVID - && chip_type != IT8705F_DEVID) - goto exit; + + switch (chip_type) { + case IT8705F_DEVID: + sio_data->type = it87; + break; + case IT8712F_DEVID: + sio_data->type = it8712; + break; + case IT8716F_DEVID: + case IT8726F_DEVID: + sio_data->type = it8716; + break; + case IT8718F_DEVID: + sio_data->type = it8718; + break; + case 0xffff: /* No device at all */ + goto exit; + default: + pr_debug(DRVNAME ": Unsupported chip (DEVID=0x%x)\n", + chip_type); + goto exit; + } superio_select(PME); if (!(superio_inb(IT87_ACT_REG) & 0x01)) { @@ -911,7 +932,7 @@ static int __init it87_find(unsigned short *address) superio_select(GPIO); if (chip_type == it8718) - vid_value = superio_inb(IT87_SIO_VID_REG); + sio_data->vid_value = superio_inb(IT87_SIO_VID_REG); reg = superio_inb(IT87_SIO_PINX2_REG); if (reg & (1 << 0)) @@ -925,18 +946,26 @@ exit: return err; } -/* This function is called by i2c_probe */ -static int it87_detect(struct i2c_adapter *adapter) +static int __devinit it87_probe(struct platform_device *pdev) { - struct i2c_client *new_client; struct it87_data *data; + struct resource *res; + struct device *dev = &pdev->dev; + struct it87_sio_data *sio_data = dev->platform_data; int err = 0; - const char *name; int enable_pwm_interface; - - /* Reserve the ISA region */ - if (!request_region(isa_address, IT87_EXTENT, - it87_isa_driver.driver.name)){ + static const char *names[] = { + "it87", + "it8712", + "it8716", + "it8718", + }; + + res = platform_get_resource(pdev, IORESOURCE_IO, 0); + if (!request_region(res->start, IT87_EXTENT, DRVNAME)) { + dev_err(dev, "Failed to request region 0x%lx-0x%lx\n", + (unsigned long)res->start, + (unsigned long)(res->start + IT87_EXTENT - 1)); err = -EBUSY; goto ERROR0; } @@ -946,129 +975,104 @@ static int it87_detect(struct i2c_adapter *adapter) goto ERROR1; } - new_client = &data->client; - i2c_set_clientdata(new_client, data); - new_client->addr = isa_address; - new_client->adapter = adapter; - new_client->driver = &it87_isa_driver; + data->addr = res->start; + data->type = sio_data->type; + data->name = names[sio_data->type]; /* Now, we do the remaining detection. */ - if ((it87_read_value(new_client, IT87_REG_CONFIG) & 0x80) - || it87_read_value(new_client, IT87_REG_CHIPID) != 0x90) { + if ((it87_read_value(data, IT87_REG_CONFIG) & 0x80) + || it87_read_value(data, IT87_REG_CHIPID) != 0x90) { err = -ENODEV; goto ERROR2; } - /* Determine the chip type. */ - switch (chip_type) { - case IT8712F_DEVID: - data->type = it8712; - name = "it8712"; - break; - case IT8716F_DEVID: - data->type = it8716; - name = "it8716"; - break; - case IT8718F_DEVID: - data->type = it8718; - name = "it8718"; - break; - default: - data->type = it87; - name = "it87"; - } + platform_set_drvdata(pdev, data); - /* Fill in the remaining client fields and put it into the global list */ - strlcpy(new_client->name, name, I2C_NAME_SIZE); mutex_init(&data->update_lock); - /* Tell the I2C layer a new client has arrived */ - if ((err = i2c_attach_client(new_client))) - goto ERROR2; - /* Check PWM configuration */ - enable_pwm_interface = it87_check_pwm(new_client); + enable_pwm_interface = it87_check_pwm(dev); /* Initialize the IT87 chip */ - it87_init_client(new_client, data); + it87_init_device(pdev); /* Register sysfs hooks */ - if ((err = sysfs_create_group(&new_client->dev.kobj, &it87_group))) - goto ERROR3; + if ((err = sysfs_create_group(&dev->kobj, &it87_group))) + goto ERROR2; /* Do not create fan files for disabled fans */ if (data->type == it8716 || data->type == it8718) { /* 16-bit tachometers */ if (data->has_fan & (1 << 0)) { - if ((err = device_create_file(&new_client->dev, + if ((err = device_create_file(dev, &sensor_dev_attr_fan1_input16.dev_attr)) - || (err = device_create_file(&new_client->dev, + || (err = device_create_file(dev, &sensor_dev_attr_fan1_min16.dev_attr))) goto ERROR4; } if (data->has_fan & (1 << 1)) { - if ((err = device_create_file(&new_client->dev, + if ((err = device_create_file(dev, &sensor_dev_attr_fan2_input16.dev_attr)) - || (err = device_create_file(&new_client->dev, + || (err = device_create_file(dev, &sensor_dev_attr_fan2_min16.dev_attr))) goto ERROR4; } if (data->has_fan & (1 << 2)) { - if ((err = device_create_file(&new_client->dev, + if ((err = device_create_file(dev, &sensor_dev_attr_fan3_input16.dev_attr)) - || (err = device_create_file(&new_client->dev, + || (err = device_create_file(dev, &sensor_dev_attr_fan3_min16.dev_attr))) goto ERROR4; } } else { /* 8-bit tachometers with clock divider */ if (data->has_fan & (1 << 0)) { - if ((err = device_create_file(&new_client->dev, + if ((err = device_create_file(dev, &sensor_dev_attr_fan1_input.dev_attr)) - || (err = device_create_file(&new_client->dev, + || (err = device_create_file(dev, &sensor_dev_attr_fan1_min.dev_attr)) - || (err = device_create_file(&new_client->dev, + || (err = device_create_file(dev, &sensor_dev_attr_fan1_div.dev_attr))) goto ERROR4; } if (data->has_fan & (1 << 1)) { - if ((err = device_create_file(&new_client->dev, + if ((err = device_create_file(dev, &sensor_dev_attr_fan2_input.dev_attr)) - || (err = device_create_file(&new_client->dev, + || (err = device_create_file(dev, &sensor_dev_attr_fan2_min.dev_attr)) - || (err = device_create_file(&new_client->dev, + || (err = device_create_file(dev, &sensor_dev_attr_fan2_div.dev_attr))) goto ERROR4; } if (data->has_fan & (1 << 2)) { - if ((err = device_create_file(&new_client->dev, + if ((err = device_create_file(dev, &sensor_dev_attr_fan3_input.dev_attr)) - || (err = device_create_file(&new_client->dev, + || (err = device_create_file(dev, &sensor_dev_attr_fan3_min.dev_attr)) - || (err = device_create_file(&new_client->dev, + || (err = device_create_file(dev, &sensor_dev_attr_fan3_div.dev_attr))) goto ERROR4; } } if (enable_pwm_interface) { - if ((err = device_create_file(&new_client->dev, + if ((err = device_create_file(dev, &sensor_dev_attr_pwm1_enable.dev_attr)) - || (err = device_create_file(&new_client->dev, + || (err = device_create_file(dev, &sensor_dev_attr_pwm2_enable.dev_attr)) - || (err = device_create_file(&new_client->dev, + || (err = device_create_file(dev, &sensor_dev_attr_pwm3_enable.dev_attr)) - || (err = device_create_file(&new_client->dev, + || (err = device_create_file(dev, &sensor_dev_attr_pwm1.dev_attr)) - || (err = device_create_file(&new_client->dev, + || (err = device_create_file(dev, &sensor_dev_attr_pwm2.dev_attr)) - || (err = device_create_file(&new_client->dev, + || (err = device_create_file(dev, &sensor_dev_attr_pwm3.dev_attr)) - || (err = device_create_file(&new_client->dev, + || (err = device_create_file(dev, &dev_attr_pwm1_freq)) - || (err = device_create_file(&new_client->dev, + || (err = device_create_file(dev, &dev_attr_pwm2_freq)) - || (err = device_create_file(&new_client->dev, + || (err = device_create_file(dev, &dev_attr_pwm3_freq))) goto ERROR4; } @@ -1077,15 +1081,15 @@ static int it87_detect(struct i2c_adapter *adapter) || data->type == it8718) { data->vrm = vid_which_vrm(); /* VID reading from Super-I/O config space if available */ - data->vid = vid_value; - if ((err = device_create_file(&new_client->dev, + data->vid = sio_data->vid_value; + if ((err = device_create_file(dev, &dev_attr_vrm)) - || (err = device_create_file(&new_client->dev, + || (err = device_create_file(dev, &dev_attr_cpu0_vid))) goto ERROR4; } - data->class_dev = hwmon_device_register(&new_client->dev); + data->class_dev = hwmon_device_register(dev); if (IS_ERR(data->class_dev)) { err = PTR_ERR(data->class_dev); goto ERROR4; @@ -1094,31 +1098,27 @@ static int it87_detect(struct i2c_adapter *adapter) return 0; ERROR4: - sysfs_remove_group(&new_client->dev.kobj, &it87_group); - sysfs_remove_group(&new_client->dev.kobj, &it87_group_opt); -ERROR3: - i2c_detach_client(new_client); + sysfs_remove_group(&dev->kobj, &it87_group); + sysfs_remove_group(&dev->kobj, &it87_group_opt); ERROR2: + platform_set_drvdata(pdev, NULL); kfree(data); ERROR1: - release_region(isa_address, IT87_EXTENT); + release_region(res->start, IT87_EXTENT); ERROR0: return err; } -static int it87_detach_client(struct i2c_client *client) +static int __devexit it87_remove(struct platform_device *pdev) { - struct it87_data *data = i2c_get_clientdata(client); - int err; + struct it87_data *data = platform_get_drvdata(pdev); hwmon_device_unregister(data->class_dev); - sysfs_remove_group(&client->dev.kobj, &it87_group); - sysfs_remove_group(&client->dev.kobj, &it87_group_opt); + sysfs_remove_group(&pdev->dev.kobj, &it87_group); + sysfs_remove_group(&pdev->dev.kobj, &it87_group_opt); - if ((err = i2c_detach_client(client))) - return err; - - release_region(client->addr, IT87_EXTENT); + release_region(data->addr, IT87_EXTENT); + platform_set_drvdata(pdev, NULL); kfree(data); return 0; @@ -1127,28 +1127,29 @@ static int it87_detach_client(struct i2c_client *client) /* Must be called with data->update_lock held, except during initialization. We ignore the IT87 BUSY flag at this moment - it could lead to deadlocks, would slow down the IT87 access and should not be necessary. */ -static int it87_read_value(struct i2c_client *client, u8 reg) +static int it87_read_value(struct it87_data *data, u8 reg) { - outb_p(reg, client->addr + IT87_ADDR_REG_OFFSET); - return inb_p(client->addr + IT87_DATA_REG_OFFSET); + outb_p(reg, data->addr + IT87_ADDR_REG_OFFSET); + return inb_p(data->addr + IT87_DATA_REG_OFFSET); } /* Must be called with data->update_lock held, except during initialization. We ignore the IT87 BUSY flag at this moment - it could lead to deadlocks, would slow down the IT87 access and should not be necessary. */ -static void it87_write_value(struct i2c_client *client, u8 reg, u8 value) +static void it87_write_value(struct it87_data *data, u8 reg, u8 value) { - outb_p(reg, client->addr + IT87_ADDR_REG_OFFSET); - outb_p(value, client->addr + IT87_DATA_REG_OFFSET); + outb_p(reg, data->addr + IT87_ADDR_REG_OFFSET); + outb_p(value, data->addr + IT87_DATA_REG_OFFSET); } /* Return 1 if and only if the PWM interface is safe to use */ -static int it87_check_pwm(struct i2c_client *client) +static int __devinit it87_check_pwm(struct device *dev) { + struct it87_data *data = dev_get_drvdata(dev); /* Some BIOSes fail to correctly configure the IT87 fans. All fans off * and polarity set to active low is sign that this is the case so we * disable pwm control to protect the user. */ - int tmp = it87_read_value(client, IT87_REG_FAN_CTL); + int tmp = it87_read_value(data, IT87_REG_FAN_CTL); if ((tmp & 0x87) == 0) { if (fix_pwm_polarity) { /* The user asks us to attempt a chip reconfiguration. @@ -1158,7 +1159,7 @@ static int it87_check_pwm(struct i2c_client *client) u8 pwm[3]; for (i = 0; i < 3; i++) - pwm[i] = it87_read_value(client, + pwm[i] = it87_read_value(data, IT87_REG_PWM(i)); /* If any fan is in automatic pwm mode, the polarity @@ -1166,26 +1167,26 @@ static int it87_check_pwm(struct i2c_client *client) * better don't change anything (but still disable the * PWM interface). */ if (!((pwm[0] | pwm[1] | pwm[2]) & 0x80)) { - dev_info(&client->dev, "Reconfiguring PWM to " + dev_info(dev, "Reconfiguring PWM to " "active high polarity\n"); - it87_write_value(client, IT87_REG_FAN_CTL, + it87_write_value(data, IT87_REG_FAN_CTL, tmp | 0x87); for (i = 0; i < 3; i++) - it87_write_value(client, + it87_write_value(data, IT87_REG_PWM(i), 0x7f & ~pwm[i]); return 1; } - dev_info(&client->dev, "PWM configuration is " + dev_info(dev, "PWM configuration is " "too broken to be fixed\n"); } - dev_info(&client->dev, "Detected broken BIOS " + dev_info(dev, "Detected broken BIOS " "defaults, disabling PWM interface\n"); return 0; } else if (fix_pwm_polarity) { - dev_info(&client->dev, "PWM configuration looks " + dev_info(dev, "PWM configuration looks " "sane, won't touch\n"); } @@ -1193,8 +1194,9 @@ static int it87_check_pwm(struct i2c_client *client) } /* Called when we have found a new IT87. */ -static void it87_init_client(struct i2c_client *client, struct it87_data *data) +static void __devinit it87_init_device(struct platform_device *pdev) { + struct it87_data *data = platform_get_drvdata(pdev); int tmp, i; /* initialize to sane defaults: @@ -1214,48 +1216,48 @@ static void it87_init_client(struct i2c_client *client, struct it87_data *data) * means -1 degree C, which surprisingly doesn't trigger an alarm, * but is still confusing, so change to 127 degrees C. */ for (i = 0; i < 8; i++) { - tmp = it87_read_value(client, IT87_REG_VIN_MIN(i)); + tmp = it87_read_value(data, IT87_REG_VIN_MIN(i)); if (tmp == 0xff) - it87_write_value(client, IT87_REG_VIN_MIN(i), 0); + it87_write_value(data, IT87_REG_VIN_MIN(i), 0); } for (i = 0; i < 3; i++) { - tmp = it87_read_value(client, IT87_REG_TEMP_HIGH(i)); + tmp = it87_read_value(data, IT87_REG_TEMP_HIGH(i)); if (tmp == 0xff) - it87_write_value(client, IT87_REG_TEMP_HIGH(i), 127); + it87_write_value(data, IT87_REG_TEMP_HIGH(i), 127); } /* Check if temperature channnels are reset manually or by some reason */ - tmp = it87_read_value(client, IT87_REG_TEMP_ENABLE); + tmp = it87_read_value(data, IT87_REG_TEMP_ENABLE); if ((tmp & 0x3f) == 0) { /* Temp1,Temp3=thermistor; Temp2=thermal diode */ tmp = (tmp & 0xc0) | 0x2a; - it87_write_value(client, IT87_REG_TEMP_ENABLE, tmp); + it87_write_value(data, IT87_REG_TEMP_ENABLE, tmp); } data->sensor = tmp; /* Check if voltage monitors are reset manually or by some reason */ - tmp = it87_read_value(client, IT87_REG_VIN_ENABLE); + tmp = it87_read_value(data, IT87_REG_VIN_ENABLE); if ((tmp & 0xff) == 0) { /* Enable all voltage monitors */ - it87_write_value(client, IT87_REG_VIN_ENABLE, 0xff); + it87_write_value(data, IT87_REG_VIN_ENABLE, 0xff); } /* Check if tachometers are reset manually or by some reason */ - data->fan_main_ctrl = it87_read_value(client, IT87_REG_FAN_MAIN_CTRL); + data->fan_main_ctrl = it87_read_value(data, IT87_REG_FAN_MAIN_CTRL); if ((data->fan_main_ctrl & 0x70) == 0) { /* Enable all fan tachometers */ data->fan_main_ctrl |= 0x70; - it87_write_value(client, IT87_REG_FAN_MAIN_CTRL, data->fan_main_ctrl); + it87_write_value(data, IT87_REG_FAN_MAIN_CTRL, data->fan_main_ctrl); } data->has_fan = (data->fan_main_ctrl >> 4) & 0x07; /* Set tachometers to 16-bit mode if needed */ if (data->type == it8716 || data->type == it8718) { - tmp = it87_read_value(client, IT87_REG_FAN_16BIT); + tmp = it87_read_value(data, IT87_REG_FAN_16BIT); if (~tmp & 0x07 & data->has_fan) { - dev_dbg(&client->dev, + dev_dbg(&pdev->dev, "Setting fan1-3 to 16-bit mode\n"); - it87_write_value(client, IT87_REG_FAN_16BIT, + it87_write_value(data, IT87_REG_FAN_16BIT, tmp | 0x07); } } @@ -1265,7 +1267,7 @@ static void it87_init_client(struct i2c_client *client, struct it87_data *data) for (i = 0; i < 3; i++) { if (data->fan_main_ctrl & (1 << i)) { /* pwm mode */ - tmp = it87_read_value(client, IT87_REG_PWM(i)); + tmp = it87_read_value(data, IT87_REG_PWM(i)); if (tmp & 0x80) { /* automatic pwm - not yet implemented, but * leave the settings made by the BIOS alone @@ -1279,15 +1281,14 @@ static void it87_init_client(struct i2c_client *client, struct it87_data *data) } /* Start monitoring */ - it87_write_value(client, IT87_REG_CONFIG, - (it87_read_value(client, IT87_REG_CONFIG) & 0x36) + it87_write_value(data, IT87_REG_CONFIG, + (it87_read_value(data, IT87_REG_CONFIG) & 0x36) | (update_vbat ? 0x41 : 0x01)); } static struct it87_data *it87_update_device(struct device *dev) { - struct i2c_client *client = to_i2c_client(dev); - struct it87_data *data = i2c_get_clientdata(client); + struct it87_data *data = dev_get_drvdata(dev); int i; mutex_lock(&data->update_lock); @@ -1298,20 +1299,20 @@ static struct it87_data *it87_update_device(struct device *dev) if (update_vbat) { /* Cleared after each update, so reenable. Value returned by this read will be previous value */ - it87_write_value(client, IT87_REG_CONFIG, - it87_read_value(client, IT87_REG_CONFIG) | 0x40); + it87_write_value(data, IT87_REG_CONFIG, + it87_read_value(data, IT87_REG_CONFIG) | 0x40); } for (i = 0; i <= 7; i++) { data->in[i] = - it87_read_value(client, IT87_REG_VIN(i)); + it87_read_value(data, IT87_REG_VIN(i)); data->in_min[i] = - it87_read_value(client, IT87_REG_VIN_MIN(i)); + it87_read_value(data, IT87_REG_VIN_MIN(i)); data->in_max[i] = - it87_read_value(client, IT87_REG_VIN_MAX(i)); + it87_read_value(data, IT87_REG_VIN_MAX(i)); } /* in8 (battery) has no limit registers */ data->in[8] = - it87_read_value(client, IT87_REG_VIN(8)); + it87_read_value(data, IT87_REG_VIN(8)); for (i = 0; i < 3; i++) { /* Skip disabled fans */ @@ -1319,46 +1320,47 @@ static struct it87_data *it87_update_device(struct device *dev) continue; data->fan_min[i] = - it87_read_value(client, IT87_REG_FAN_MIN(i)); - data->fan[i] = it87_read_value(client, + it87_read_value(data, IT87_REG_FAN_MIN(i)); + data->fan[i] = it87_read_value(data, IT87_REG_FAN(i)); /* Add high byte if in 16-bit mode */ if (data->type == it8716 || data->type == it8718) { - data->fan[i] |= it87_read_value(client, + data->fan[i] |= it87_read_value(data, IT87_REG_FANX(i)) << 8; - data->fan_min[i] |= it87_read_value(client, + data->fan_min[i] |= it87_read_value(data, IT87_REG_FANX_MIN(i)) << 8; } } for (i = 0; i < 3; i++) { data->temp[i] = - it87_read_value(client, IT87_REG_TEMP(i)); + it87_read_value(data, IT87_REG_TEMP(i)); data->temp_high[i] = - it87_read_value(client, IT87_REG_TEMP_HIGH(i)); + it87_read_value(data, IT87_REG_TEMP_HIGH(i)); data->temp_low[i] = - it87_read_value(client, IT87_REG_TEMP_LOW(i)); + it87_read_value(data, IT87_REG_TEMP_LOW(i)); } /* Newer chips don't have clock dividers */ if ((data->has_fan & 0x07) && data->type != it8716 && data->type != it8718) { - i = it87_read_value(client, IT87_REG_FAN_DIV); + i = it87_read_value(data, IT87_REG_FAN_DIV); data->fan_div[0] = i & 0x07; data->fan_div[1] = (i >> 3) & 0x07; data->fan_div[2] = (i & 0x40) ? 3 : 1; } data->alarms = - it87_read_value(client, IT87_REG_ALARM1) | - (it87_read_value(client, IT87_REG_ALARM2) << 8) | - (it87_read_value(client, IT87_REG_ALARM3) << 16); - data->fan_main_ctrl = it87_read_value(client, IT87_REG_FAN_MAIN_CTRL); - data->fan_ctl = it87_read_value(client, IT87_REG_FAN_CTL); - - data->sensor = it87_read_value(client, IT87_REG_TEMP_ENABLE); + it87_read_value(data, IT87_REG_ALARM1) | + (it87_read_value(data, IT87_REG_ALARM2) << 8) | + (it87_read_value(data, IT87_REG_ALARM3) << 16); + data->fan_main_ctrl = it87_read_value(data, + IT87_REG_FAN_MAIN_CTRL); + data->fan_ctl = it87_read_value(data, IT87_REG_FAN_CTL); + + data->sensor = it87_read_value(data, IT87_REG_TEMP_ENABLE); /* The 8705 does not have VID capability */ if (data->type == it8712 || data->type == it8716) { - data->vid = it87_read_value(client, IT87_REG_VID); + data->vid = it87_read_value(data, IT87_REG_VID); /* The older IT8712F revisions had only 5 VID pins, but we assume it is always safe to read 6 bits. */ data->vid &= 0x3f; @@ -1372,24 +1374,85 @@ static struct it87_data *it87_update_device(struct device *dev) return data; } +static int __init it87_device_add(unsigned short address, + const struct it87_sio_data *sio_data) +{ + struct resource res = { + .start = address , + .end = address + IT87_EXTENT - 1, + .name = DRVNAME, + .flags = IORESOURCE_IO, + }; + int err; + + pdev = platform_device_alloc(DRVNAME, address); + if (!pdev) { + err = -ENOMEM; + printk(KERN_ERR DRVNAME ": Device allocation failed\n"); + goto exit; + } + + err = platform_device_add_resources(pdev, &res, 1); + if (err) { + printk(KERN_ERR DRVNAME ": Device resource addition failed " + "(%d)\n", err); + goto exit_device_put; + } + + err = platform_device_add_data(pdev, sio_data, + sizeof(struct it87_sio_data)); + if (err) { + printk(KERN_ERR DRVNAME ": Platform data allocation failed\n"); + goto exit_device_put; + } + + err = platform_device_add(pdev); + if (err) { + printk(KERN_ERR DRVNAME ": Device addition failed (%d)\n", + err); + goto exit_device_put; + } + + return 0; + +exit_device_put: + platform_device_put(pdev); +exit: + return err; +} + static int __init sm_it87_init(void) { - int res; + int err; + unsigned short isa_address=0; + struct it87_sio_data sio_data; + + err = it87_find(&isa_address, &sio_data); + if (err) + return err; + err = platform_driver_register(&it87_driver); + if (err) + return err; - if ((res = it87_find(&isa_address))) - return res; - return i2c_isa_add_driver(&it87_isa_driver); + err = it87_device_add(isa_address, &sio_data); + if (err){ + platform_driver_unregister(&it87_driver); + return err; + } + + return 0; } static void __exit sm_it87_exit(void) { - i2c_isa_del_driver(&it87_isa_driver); + platform_device_unregister(pdev); + platform_driver_unregister(&it87_driver); } MODULE_AUTHOR("Chris Gauthron <chrisg@0-in.com>, " "Jean Delvare <khali@linux-fr.org>"); -MODULE_DESCRIPTION("IT8705F/8712F/8716F/8718F, SiS950 driver"); +MODULE_DESCRIPTION("IT8705F/8712F/8716F/8718F/8726F, SiS950 driver"); module_param(update_vbat, bool, 0); MODULE_PARM_DESC(update_vbat, "Update vbat if set else return powerup value"); module_param(fix_pwm_polarity, bool, 0); diff --git a/drivers/hwmon/lm63.c b/drivers/hwmon/lm63.c index d69f3cf0712..2162d69a8c0 100644 --- a/drivers/hwmon/lm63.c +++ b/drivers/hwmon/lm63.c @@ -364,7 +364,7 @@ static DEVICE_ATTR(temp2_crit_hyst, S_IWUSR | S_IRUGO, show_temp2_crit_hyst, /* Individual alarm files */ static SENSOR_DEVICE_ATTR(fan1_min_alarm, S_IRUGO, show_alarm, NULL, 0); static SENSOR_DEVICE_ATTR(temp2_crit_alarm, S_IRUGO, show_alarm, NULL, 1); -static SENSOR_DEVICE_ATTR(temp2_input_fault, S_IRUGO, show_alarm, NULL, 2); +static SENSOR_DEVICE_ATTR(temp2_fault, S_IRUGO, show_alarm, NULL, 2); static SENSOR_DEVICE_ATTR(temp2_min_alarm, S_IRUGO, show_alarm, NULL, 3); static SENSOR_DEVICE_ATTR(temp2_max_alarm, S_IRUGO, show_alarm, NULL, 4); static SENSOR_DEVICE_ATTR(temp1_max_alarm, S_IRUGO, show_alarm, NULL, 6); @@ -383,7 +383,7 @@ static struct attribute *lm63_attributes[] = { &dev_attr_temp2_crit_hyst.attr, &sensor_dev_attr_temp2_crit_alarm.dev_attr.attr, - &sensor_dev_attr_temp2_input_fault.dev_attr.attr, + &sensor_dev_attr_temp2_fault.dev_attr.attr, &sensor_dev_attr_temp2_min_alarm.dev_attr.attr, &sensor_dev_attr_temp2_max_alarm.dev_attr.attr, &sensor_dev_attr_temp1_max_alarm.dev_attr.attr, diff --git a/drivers/hwmon/lm83.c b/drivers/hwmon/lm83.c index feb87b41e98..654c0f73464 100644 --- a/drivers/hwmon/lm83.c +++ b/drivers/hwmon/lm83.c @@ -223,14 +223,14 @@ static SENSOR_DEVICE_ATTR(temp4_crit, S_IRUGO, show_temp, NULL, 8); /* Individual alarm files */ static SENSOR_DEVICE_ATTR(temp1_crit_alarm, S_IRUGO, show_alarm, NULL, 0); static SENSOR_DEVICE_ATTR(temp3_crit_alarm, S_IRUGO, show_alarm, NULL, 1); -static SENSOR_DEVICE_ATTR(temp3_input_fault, S_IRUGO, show_alarm, NULL, 2); +static SENSOR_DEVICE_ATTR(temp3_fault, S_IRUGO, show_alarm, NULL, 2); static SENSOR_DEVICE_ATTR(temp3_max_alarm, S_IRUGO, show_alarm, NULL, 4); static SENSOR_DEVICE_ATTR(temp1_max_alarm, S_IRUGO, show_alarm, NULL, 6); static SENSOR_DEVICE_ATTR(temp2_crit_alarm, S_IRUGO, show_alarm, NULL, 8); static SENSOR_DEVICE_ATTR(temp4_crit_alarm, S_IRUGO, show_alarm, NULL, 9); -static SENSOR_DEVICE_ATTR(temp4_input_fault, S_IRUGO, show_alarm, NULL, 10); +static SENSOR_DEVICE_ATTR(temp4_fault, S_IRUGO, show_alarm, NULL, 10); static SENSOR_DEVICE_ATTR(temp4_max_alarm, S_IRUGO, show_alarm, NULL, 12); -static SENSOR_DEVICE_ATTR(temp2_input_fault, S_IRUGO, show_alarm, NULL, 13); +static SENSOR_DEVICE_ATTR(temp2_fault, S_IRUGO, show_alarm, NULL, 13); static SENSOR_DEVICE_ATTR(temp2_max_alarm, S_IRUGO, show_alarm, NULL, 15); /* Raw alarm file for compatibility */ static DEVICE_ATTR(alarms, S_IRUGO, show_alarms, NULL); @@ -245,7 +245,7 @@ static struct attribute *lm83_attributes[] = { &sensor_dev_attr_temp1_crit_alarm.dev_attr.attr, &sensor_dev_attr_temp3_crit_alarm.dev_attr.attr, - &sensor_dev_attr_temp3_input_fault.dev_attr.attr, + &sensor_dev_attr_temp3_fault.dev_attr.attr, &sensor_dev_attr_temp3_max_alarm.dev_attr.attr, &sensor_dev_attr_temp1_max_alarm.dev_attr.attr, &dev_attr_alarms.attr, @@ -266,9 +266,9 @@ static struct attribute *lm83_attributes_opt[] = { &sensor_dev_attr_temp2_crit_alarm.dev_attr.attr, &sensor_dev_attr_temp4_crit_alarm.dev_attr.attr, - &sensor_dev_attr_temp4_input_fault.dev_attr.attr, + &sensor_dev_attr_temp4_fault.dev_attr.attr, &sensor_dev_attr_temp4_max_alarm.dev_attr.attr, - &sensor_dev_attr_temp2_input_fault.dev_attr.attr, + &sensor_dev_attr_temp2_fault.dev_attr.attr, &sensor_dev_attr_temp2_max_alarm.dev_attr.attr, NULL }; diff --git a/drivers/hwmon/lm90.c b/drivers/hwmon/lm90.c index 6882ce75fee..48833fff492 100644 --- a/drivers/hwmon/lm90.c +++ b/drivers/hwmon/lm90.c @@ -43,6 +43,13 @@ * variants. The extra address and features of the MAX6659 are not * supported by this driver. * + * This driver also supports the MAX6680 and MAX6681, two other sensor + * chips made by Maxim. These are quite similar to the other Maxim + * chips. Complete datasheet can be obtained at: + * http://www.maxim-ic.com/quick_view2.cfm/qv_pk/3370 + * The MAX6680 and MAX6681 only differ in the pinout so they can be + * treated identically. + * * This driver also supports the ADT7461 chip from Analog Devices but * only in its "compatability mode". If an ADT7461 chip is found but * is configured in non-compatible mode (where its temperature @@ -84,20 +91,25 @@ /* * Addresses to scan * Address is fully defined internally and cannot be changed except for - * MAX6659. + * MAX6659, MAX6680 and MAX6681. * LM86, LM89, LM90, LM99, ADM1032, ADM1032-1, ADT7461, MAX6657 and MAX6658 * have address 0x4c. * ADM1032-2, ADT7461-2, LM89-1, and LM99-1 have address 0x4d. * MAX6659 can have address 0x4c, 0x4d or 0x4e (unsupported). + * MAX6680 and MAX6681 can have address 0x18, 0x19, 0x1a, 0x29, 0x2a, 0x2b, + * 0x4c, 0x4d or 0x4e. */ -static unsigned short normal_i2c[] = { 0x4c, 0x4d, I2C_CLIENT_END }; +static unsigned short normal_i2c[] = { 0x18, 0x19, 0x1a, + 0x29, 0x2a, 0x2b, + 0x4c, 0x4d, 0x4e, + I2C_CLIENT_END }; /* * Insmod parameters */ -I2C_CLIENT_INSMOD_6(lm90, adm1032, lm99, lm86, max6657, adt7461); +I2C_CLIENT_INSMOD_7(lm90, adm1032, lm99, lm86, max6657, adt7461, max6680); /* * The LM90 registers @@ -359,7 +371,7 @@ static SENSOR_DEVICE_ATTR(temp2_crit_hyst, S_IRUGO, show_temphyst, NULL, 4); /* Individual alarm files */ static SENSOR_DEVICE_ATTR(temp1_crit_alarm, S_IRUGO, show_alarm, NULL, 0); static SENSOR_DEVICE_ATTR(temp2_crit_alarm, S_IRUGO, show_alarm, NULL, 1); -static SENSOR_DEVICE_ATTR(temp2_input_fault, S_IRUGO, show_alarm, NULL, 2); +static SENSOR_DEVICE_ATTR(temp2_fault, S_IRUGO, show_alarm, NULL, 2); static SENSOR_DEVICE_ATTR(temp2_min_alarm, S_IRUGO, show_alarm, NULL, 3); static SENSOR_DEVICE_ATTR(temp2_max_alarm, S_IRUGO, show_alarm, NULL, 4); static SENSOR_DEVICE_ATTR(temp1_min_alarm, S_IRUGO, show_alarm, NULL, 5); @@ -381,7 +393,7 @@ static struct attribute *lm90_attributes[] = { &sensor_dev_attr_temp1_crit_alarm.dev_attr.attr, &sensor_dev_attr_temp2_crit_alarm.dev_attr.attr, - &sensor_dev_attr_temp2_input_fault.dev_attr.attr, + &sensor_dev_attr_temp2_fault.dev_attr.attr, &sensor_dev_attr_temp2_min_alarm.dev_attr.attr, &sensor_dev_attr_temp2_max_alarm.dev_attr.attr, &sensor_dev_attr_temp1_min_alarm.dev_attr.attr, @@ -429,7 +441,7 @@ static DEVICE_ATTR(pec, S_IWUSR | S_IRUGO, show_pec, set_pec); */ /* The ADM1032 supports PEC but not on write byte transactions, so we need - to explicitely ask for a transaction without PEC. */ + to explicitly ask for a transaction without PEC. */ static inline s32 adm1032_write_byte(struct i2c_client *client, u8 value) { return i2c_smbus_xfer(client->adapter, client->addr, @@ -525,7 +537,8 @@ static int lm90_detect(struct i2c_adapter *adapter, int address, int kind) ®_convrate) < 0) goto exit_free; - if (man_id == 0x01) { /* National Semiconductor */ + if ((address == 0x4C || address == 0x4D) + && man_id == 0x01) { /* National Semiconductor */ u8 reg_config2; if (lm90_read_reg(new_client, LM90_REG_R_CONFIG2, @@ -548,7 +561,8 @@ static int lm90_detect(struct i2c_adapter *adapter, int address, int kind) } } } else - if (man_id == 0x41) { /* Analog Devices */ + if ((address == 0x4C || address == 0x4D) + && man_id == 0x41) { /* Analog Devices */ if ((chip_id & 0xF0) == 0x40 /* ADM1032 */ && (reg_config1 & 0x3F) == 0x00 && reg_convrate <= 0x0A) { @@ -562,18 +576,30 @@ static int lm90_detect(struct i2c_adapter *adapter, int address, int kind) } else if (man_id == 0x4D) { /* Maxim */ /* - * The Maxim variants do NOT have a chip_id register. - * Reading from that address will return the last read - * value, which in our case is those of the man_id - * register. Likewise, the config1 register seems to - * lack a low nibble, so the value will be those of the - * previous read, so in our case those of the man_id - * register. + * The MAX6657, MAX6658 and MAX6659 do NOT have a + * chip_id register. Reading from that address will + * return the last read value, which in our case is + * those of the man_id register. Likewise, the config1 + * register seems to lack a low nibble, so the value + * will be those of the previous read, so in our case + * those of the man_id register. */ if (chip_id == man_id + && (address == 0x4F || address == 0x4D) && (reg_config1 & 0x1F) == (man_id & 0x0F) && reg_convrate <= 0x09) { kind = max6657; + } else + /* The chip_id register of the MAX6680 and MAX6681 + * holds the revision of the chip. + * the lowest bit of the config1 register is unused + * and should return zero when read, so should the + * second to last bit of config1 (software reset) + */ + if (chip_id == 0x01 + && (reg_config1 & 0x03) == 0x00 + && reg_convrate <= 0x07) { + kind = max6680; } } @@ -599,6 +625,8 @@ static int lm90_detect(struct i2c_adapter *adapter, int address, int kind) name = "lm86"; } else if (kind == max6657) { name = "max6657"; + } else if (kind == max6680) { + name = "max6680"; } else if (kind == adt7461) { name = "adt7461"; } @@ -646,7 +674,8 @@ exit: static void lm90_init_client(struct i2c_client *client) { - u8 config; + u8 config, config_orig; + struct lm90_data *data = i2c_get_clientdata(client); /* * Start the conversions. @@ -657,9 +686,20 @@ static void lm90_init_client(struct i2c_client *client) dev_warn(&client->dev, "Initialization failed!\n"); return; } - if (config & 0x40) - i2c_smbus_write_byte_data(client, LM90_REG_W_CONFIG1, - config & 0xBF); /* run */ + config_orig = config; + + /* + * Put MAX6680/MAX8881 into extended resolution (bit 0x10, + * 0.125 degree resolution) and range (0x08, extend range + * to -64 degree) mode for the remote temperature sensor. + */ + if (data->kind == max6680) { + config |= 0x18; + } + + config &= 0xBF; /* run */ + if (config != config_orig) /* Only write if changed */ + i2c_smbus_write_byte_data(client, LM90_REG_W_CONFIG1, config); } static int lm90_detach_client(struct i2c_client *client) diff --git a/drivers/hwmon/lm93.c b/drivers/hwmon/lm93.c new file mode 100644 index 00000000000..23edf4fe422 --- /dev/null +++ b/drivers/hwmon/lm93.c @@ -0,0 +1,2655 @@ +/* + lm93.c - Part of lm_sensors, Linux kernel modules for hardware monitoring + + Author/Maintainer: Mark M. Hoffman <mhoffman@lightlink.com> + Copyright (c) 2004 Utilitek Systems, Inc. + + derived in part from lm78.c: + Copyright (c) 1998, 1999 Frodo Looijaard <frodol@dds.nl> + + derived in part from lm85.c: + Copyright (c) 2002, 2003 Philip Pokorny <ppokorny@penguincomputing.com> + Copyright (c) 2003 Margit Schubert-While <margitsw@t-online.de> + + derived in part from w83l785ts.c: + Copyright (c) 2003-2004 Jean Delvare <khali@linux-fr.org> + + Ported to Linux 2.6 by Eric J. Bowersox <ericb@aspsys.com> + Copyright (c) 2005 Aspen Systems, Inc. + + Adapted to 2.6.20 by Carsten Emde <cbe@osadl.org> + Copyright (c) 2006 Carsten Emde, Open Source Automation Development Lab + + Modified for mainline integration by Hans J. Koch <hjk@linutronix.de> + Copyright (c) 2007 Hans J. Koch, Linutronix GmbH + + 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. + + You should have received a copy of the GNU General Public License + along with this program; if not, write to the Free Software + Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA. +*/ + +#include <linux/module.h> +#include <linux/init.h> +#include <linux/slab.h> +#include <linux/i2c.h> +#include <linux/hwmon.h> +#include <linux/hwmon-sysfs.h> +#include <linux/hwmon-vid.h> +#include <linux/err.h> +#include <linux/delay.h> + +/* LM93 REGISTER ADDRESSES */ + +/* miscellaneous */ +#define LM93_REG_MFR_ID 0x3e +#define LM93_REG_VER 0x3f +#define LM93_REG_STATUS_CONTROL 0xe2 +#define LM93_REG_CONFIG 0xe3 +#define LM93_REG_SLEEP_CONTROL 0xe4 + +/* alarm values start here */ +#define LM93_REG_HOST_ERROR_1 0x48 + +/* voltage inputs: in1-in16 (nr => 0-15) */ +#define LM93_REG_IN(nr) (0x56 + (nr)) +#define LM93_REG_IN_MIN(nr) (0x90 + (nr) * 2) +#define LM93_REG_IN_MAX(nr) (0x91 + (nr) * 2) + +/* temperature inputs: temp1-temp4 (nr => 0-3) */ +#define LM93_REG_TEMP(nr) (0x50 + (nr)) +#define LM93_REG_TEMP_MIN(nr) (0x78 + (nr) * 2) +#define LM93_REG_TEMP_MAX(nr) (0x79 + (nr) * 2) + +/* temp[1-4]_auto_boost (nr => 0-3) */ +#define LM93_REG_BOOST(nr) (0x80 + (nr)) + +/* #PROCHOT inputs: prochot1-prochot2 (nr => 0-1) */ +#define LM93_REG_PROCHOT_CUR(nr) (0x67 + (nr) * 2) +#define LM93_REG_PROCHOT_AVG(nr) (0x68 + (nr) * 2) +#define LM93_REG_PROCHOT_MAX(nr) (0xb0 + (nr)) + +/* fan tach inputs: fan1-fan4 (nr => 0-3) */ +#define LM93_REG_FAN(nr) (0x6e + (nr) * 2) +#define LM93_REG_FAN_MIN(nr) (0xb4 + (nr) * 2) + +/* pwm outputs: pwm1-pwm2 (nr => 0-1, reg => 0-3) */ +#define LM93_REG_PWM_CTL(nr,reg) (0xc8 + (reg) + (nr) * 4) +#define LM93_PWM_CTL1 0x0 +#define LM93_PWM_CTL2 0x1 +#define LM93_PWM_CTL3 0x2 +#define LM93_PWM_CTL4 0x3 + +/* GPIO input state */ +#define LM93_REG_GPI 0x6b + +/* vid inputs: vid1-vid2 (nr => 0-1) */ +#define LM93_REG_VID(nr) (0x6c + (nr)) + +/* vccp1 & vccp2: VID relative inputs (nr => 0-1) */ +#define LM93_REG_VCCP_LIMIT_OFF(nr) (0xb2 + (nr)) + +/* temp[1-4]_auto_boost_hyst */ +#define LM93_REG_BOOST_HYST_12 0xc0 +#define LM93_REG_BOOST_HYST_34 0xc1 +#define LM93_REG_BOOST_HYST(nr) (0xc0 + (nr)/2) + +/* temp[1-4]_auto_pwm_[min|hyst] */ +#define LM93_REG_PWM_MIN_HYST_12 0xc3 +#define LM93_REG_PWM_MIN_HYST_34 0xc4 +#define LM93_REG_PWM_MIN_HYST(nr) (0xc3 + (nr)/2) + +/* prochot_override & prochot_interval */ +#define LM93_REG_PROCHOT_OVERRIDE 0xc6 +#define LM93_REG_PROCHOT_INTERVAL 0xc7 + +/* temp[1-4]_auto_base (nr => 0-3) */ +#define LM93_REG_TEMP_BASE(nr) (0xd0 + (nr)) + +/* temp[1-4]_auto_offsets (step => 0-11) */ +#define LM93_REG_TEMP_OFFSET(step) (0xd4 + (step)) + +/* #PROCHOT & #VRDHOT PWM ramp control */ +#define LM93_REG_PWM_RAMP_CTL 0xbf + +/* miscellaneous */ +#define LM93_REG_SFC1 0xbc +#define LM93_REG_SFC2 0xbd +#define LM93_REG_GPI_VID_CTL 0xbe +#define LM93_REG_SF_TACH_TO_PWM 0xe0 + +/* error masks */ +#define LM93_REG_GPI_ERR_MASK 0xec +#define LM93_REG_MISC_ERR_MASK 0xed + +/* LM93 REGISTER VALUES */ +#define LM93_MFR_ID 0x73 +#define LM93_MFR_ID_PROTOTYPE 0x72 + +/* SMBus capabilities */ +#define LM93_SMBUS_FUNC_FULL (I2C_FUNC_SMBUS_BYTE_DATA | \ + I2C_FUNC_SMBUS_WORD_DATA | I2C_FUNC_SMBUS_BLOCK_DATA) +#define LM93_SMBUS_FUNC_MIN (I2C_FUNC_SMBUS_BYTE_DATA | \ + I2C_FUNC_SMBUS_WORD_DATA) + +/* Addresses to scan */ +static unsigned short normal_i2c[] = { 0x2c, 0x2d, 0x2e, I2C_CLIENT_END }; + +/* Insmod parameters */ +I2C_CLIENT_INSMOD_1(lm93); + +static int disable_block; +module_param(disable_block, bool, 0); +MODULE_PARM_DESC(disable_block, + "Set to non-zero to disable SMBus block data transactions."); + +static int init; +module_param(init, bool, 0); +MODULE_PARM_DESC(init, "Set to non-zero to force chip initialization."); + +static int vccp_limit_type[2] = {0,0}; +module_param_array(vccp_limit_type, int, NULL, 0); +MODULE_PARM_DESC(vccp_limit_type, "Configures in7 and in8 limit modes."); + +static int vid_agtl; +module_param(vid_agtl, int, 0); +MODULE_PARM_DESC(vid_agtl, "Configures VID pin input thresholds."); + +/* Driver data */ +static struct i2c_driver lm93_driver; + +/* LM93 BLOCK READ COMMANDS */ +static const struct { u8 cmd; u8 len; } lm93_block_read_cmds[12] = { + { 0xf2, 8 }, + { 0xf3, 8 }, + { 0xf4, 6 }, + { 0xf5, 16 }, + { 0xf6, 4 }, + { 0xf7, 8 }, + { 0xf8, 12 }, + { 0xf9, 32 }, + { 0xfa, 8 }, + { 0xfb, 8 }, + { 0xfc, 16 }, + { 0xfd, 9 }, +}; + +/* ALARMS: SYSCTL format described further below + REG: 64 bits in 8 registers, as immediately below */ +struct block1_t { + u8 host_status_1; + u8 host_status_2; + u8 host_status_3; + u8 host_status_4; + u8 p1_prochot_status; + u8 p2_prochot_status; + u8 gpi_status; + u8 fan_status; +}; + +/* + * Client-specific data + */ +struct lm93_data { + struct i2c_client client; + struct class_device *class_dev; + + struct mutex update_lock; + unsigned long last_updated; /* In jiffies */ + + /* client update function */ + void (*update)(struct lm93_data *, struct i2c_client *); + + char valid; /* !=0 if following fields are valid */ + + /* register values, arranged by block read groups */ + struct block1_t block1; + + /* temp1 - temp4: unfiltered readings + temp1 - temp2: filtered readings */ + u8 block2[6]; + + /* vin1 - vin16: readings */ + u8 block3[16]; + + /* prochot1 - prochot2: readings */ + struct { + u8 cur; + u8 avg; + } block4[2]; + + /* fan counts 1-4 => 14-bits, LE, *left* justified */ + u16 block5[4]; + + /* block6 has a lot of data we don't need */ + struct { + u8 min; + u8 max; + } temp_lim[3]; + + /* vin1 - vin16: low and high limits */ + struct { + u8 min; + u8 max; + } block7[16]; + + /* fan count limits 1-4 => same format as block5 */ + u16 block8[4]; + + /* pwm control registers (2 pwms, 4 regs) */ + u8 block9[2][4]; + + /* auto/pwm base temp and offset temp registers */ + struct { + u8 base[4]; + u8 offset[12]; + } block10; + + /* master config register */ + u8 config; + + /* VID1 & VID2 => register format, 6-bits, right justified */ + u8 vid[2]; + + /* prochot1 - prochot2: limits */ + u8 prochot_max[2]; + + /* vccp1 & vccp2 (in7 & in8): VID relative limits (register format) */ + u8 vccp_limits[2]; + + /* GPIO input state (register format, i.e. inverted) */ + u8 gpi; + + /* #PROCHOT override (register format) */ + u8 prochot_override; + + /* #PROCHOT intervals (register format) */ + u8 prochot_interval; + + /* Fan Boost Temperatures (register format) */ + u8 boost[4]; + + /* Fan Boost Hysteresis (register format) */ + u8 boost_hyst[2]; + + /* Temperature Zone Min. PWM & Hysteresis (register format) */ + u8 auto_pwm_min_hyst[2]; + + /* #PROCHOT & #VRDHOT PWM Ramp Control */ + u8 pwm_ramp_ctl; + + /* miscellaneous setup regs */ + u8 sfc1; + u8 sfc2; + u8 sf_tach_to_pwm; + + /* The two PWM CTL2 registers can read something other than what was + last written for the OVR_DC field (duty cycle override). So, we + save the user-commanded value here. */ + u8 pwm_override[2]; +}; + +/* VID: mV + REG: 6-bits, right justified, *always* using Intel VRM/VRD 10 */ +static int LM93_VID_FROM_REG(u8 reg) +{ + return vid_from_reg((reg & 0x3f), 100); +} + +/* min, max, and nominal register values, per channel (u8) */ +static const u8 lm93_vin_reg_min[16] = { + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0xae, +}; +static const u8 lm93_vin_reg_max[16] = { + 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, + 0xff, 0xfa, 0xff, 0xff, 0xff, 0xff, 0xff, 0xd1, +}; +/* Values from the datasheet. They're here for documentation only. +static const u8 lm93_vin_reg_nom[16] = { + 0xc0, 0xc0, 0xc0, 0xc0, 0xc0, 0xc0, 0xc0, 0xc0, + 0xc0, 0xc0, 0xc0, 0xc0, 0xc0, 0xc0, 0x40, 0xc0, +}; +*/ + +/* min, max, and nominal voltage readings, per channel (mV)*/ +static const unsigned long lm93_vin_val_min[16] = { + 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 3000, +}; + +static const unsigned long lm93_vin_val_max[16] = { + 1236, 1236, 1236, 1600, 2000, 2000, 1600, 1600, + 4400, 6500, 3333, 2625, 1312, 1312, 1236, 3600, +}; +/* Values from the datasheet. They're here for documentation only. +static const unsigned long lm93_vin_val_nom[16] = { + 927, 927, 927, 1200, 1500, 1500, 1200, 1200, + 3300, 5000, 2500, 1969, 984, 984, 309, 3300, +}; +*/ + +static unsigned LM93_IN_FROM_REG(int nr, u8 reg) +{ + const long uV_max = lm93_vin_val_max[nr] * 1000; + const long uV_min = lm93_vin_val_min[nr] * 1000; + + const long slope = (uV_max - uV_min) / + (lm93_vin_reg_max[nr] - lm93_vin_reg_min[nr]); + const long intercept = uV_min - slope * lm93_vin_reg_min[nr]; + + return (slope * reg + intercept + 500) / 1000; +} + +/* IN: mV, limits determined by channel nr + REG: scaling determined by channel nr */ +static u8 LM93_IN_TO_REG(int nr, unsigned val) +{ + /* range limit */ + const long mV = SENSORS_LIMIT(val, + lm93_vin_val_min[nr], lm93_vin_val_max[nr]); + + /* try not to lose too much precision here */ + const long uV = mV * 1000; + const long uV_max = lm93_vin_val_max[nr] * 1000; + const long uV_min = lm93_vin_val_min[nr] * 1000; + + /* convert */ + const long slope = (uV_max - uV_min) / + (lm93_vin_reg_max[nr] - lm93_vin_reg_min[nr]); + const long intercept = uV_min - slope * lm93_vin_reg_min[nr]; + + u8 result = ((uV - intercept + (slope/2)) / slope); + result = SENSORS_LIMIT(result, + lm93_vin_reg_min[nr], lm93_vin_reg_max[nr]); + return result; +} + +/* vid in mV, upper == 0 indicates low limit, otherwise upper limit */ +static unsigned LM93_IN_REL_FROM_REG(u8 reg, int upper, int vid) +{ + const long uV_offset = upper ? (((reg >> 4 & 0x0f) + 1) * 12500) : + (((reg >> 0 & 0x0f) + 1) * -25000); + const long uV_vid = vid * 1000; + return (uV_vid + uV_offset + 5000) / 10000; +} + +#define LM93_IN_MIN_FROM_REG(reg,vid) LM93_IN_REL_FROM_REG(reg,0,vid) +#define LM93_IN_MAX_FROM_REG(reg,vid) LM93_IN_REL_FROM_REG(reg,1,vid) + +/* vid in mV , upper == 0 indicates low limit, otherwise upper limit + upper also determines which nibble of the register is returned + (the other nibble will be 0x0) */ +static u8 LM93_IN_REL_TO_REG(unsigned val, int upper, int vid) +{ + long uV_offset = vid * 1000 - val * 10000; + if (upper) { + uV_offset = SENSORS_LIMIT(uV_offset, 12500, 200000); + return (u8)((uV_offset / 12500 - 1) << 4); + } else { + uV_offset = SENSORS_LIMIT(uV_offset, -400000, -25000); + return (u8)((uV_offset / -25000 - 1) << 0); + } +} + +/* TEMP: 1/1000 degrees C (-128C to +127C) + REG: 1C/bit, two's complement */ +static int LM93_TEMP_FROM_REG(u8 reg) +{ + return (s8)reg * 1000; +} + +#define LM93_TEMP_MIN (-128000) +#define LM93_TEMP_MAX ( 127000) + +/* TEMP: 1/1000 degrees C (-128C to +127C) + REG: 1C/bit, two's complement */ +static u8 LM93_TEMP_TO_REG(int temp) +{ + int ntemp = SENSORS_LIMIT(temp, LM93_TEMP_MIN, LM93_TEMP_MAX); + ntemp += (ntemp<0 ? -500 : 500); + return (u8)(ntemp / 1000); +} + +/* Determine 4-bit temperature offset resolution */ +static int LM93_TEMP_OFFSET_MODE_FROM_REG(u8 sfc2, int nr) +{ + /* mode: 0 => 1C/bit, nonzero => 0.5C/bit */ + return sfc2 & (nr < 2 ? 0x10 : 0x20); +} + +/* This function is common to all 4-bit temperature offsets + reg is 4 bits right justified + mode 0 => 1C/bit, mode !0 => 0.5C/bit */ +static int LM93_TEMP_OFFSET_FROM_REG(u8 reg, int mode) +{ + return (reg & 0x0f) * (mode ? 5 : 10); +} + +#define LM93_TEMP_OFFSET_MIN ( 0) +#define LM93_TEMP_OFFSET_MAX0 (150) +#define LM93_TEMP_OFFSET_MAX1 ( 75) + +/* This function is common to all 4-bit temperature offsets + returns 4 bits right justified + mode 0 => 1C/bit, mode !0 => 0.5C/bit */ +static u8 LM93_TEMP_OFFSET_TO_REG(int off, int mode) +{ + int factor = mode ? 5 : 10; + + off = SENSORS_LIMIT(off, LM93_TEMP_OFFSET_MIN, + mode ? LM93_TEMP_OFFSET_MAX1 : LM93_TEMP_OFFSET_MAX0); + return (u8)((off + factor/2) / factor); +} + +/* 0 <= nr <= 3 */ +static int LM93_TEMP_AUTO_OFFSET_FROM_REG(u8 reg, int nr, int mode) +{ + /* temp1-temp2 (nr=0,1) use lower nibble */ + if (nr < 2) + return LM93_TEMP_OFFSET_FROM_REG(reg & 0x0f, mode); + + /* temp3-temp4 (nr=2,3) use upper nibble */ + else + return LM93_TEMP_OFFSET_FROM_REG(reg >> 4 & 0x0f, mode); +} + +/* TEMP: 1/10 degrees C (0C to +15C (mode 0) or +7.5C (mode non-zero)) + REG: 1.0C/bit (mode 0) or 0.5C/bit (mode non-zero) + 0 <= nr <= 3 */ +static u8 LM93_TEMP_AUTO_OFFSET_TO_REG(u8 old, int off, int nr, int mode) +{ + u8 new = LM93_TEMP_OFFSET_TO_REG(off, mode); + + /* temp1-temp2 (nr=0,1) use lower nibble */ + if (nr < 2) + return (old & 0xf0) | (new & 0x0f); + + /* temp3-temp4 (nr=2,3) use upper nibble */ + else + return (new << 4 & 0xf0) | (old & 0x0f); +} + +static int LM93_AUTO_BOOST_HYST_FROM_REGS(struct lm93_data *data, int nr, + int mode) +{ + u8 reg; + + switch (nr) { + case 0: + reg = data->boost_hyst[0] & 0x0f; + break; + case 1: + reg = data->boost_hyst[0] >> 4 & 0x0f; + break; + case 2: + reg = data->boost_hyst[1] & 0x0f; + break; + case 3: + default: + reg = data->boost_hyst[1] >> 4 & 0x0f; + break; + } + + return LM93_TEMP_FROM_REG(data->boost[nr]) - + LM93_TEMP_OFFSET_FROM_REG(reg, mode); +} + +static u8 LM93_AUTO_BOOST_HYST_TO_REG(struct lm93_data *data, long hyst, + int nr, int mode) +{ + u8 reg = LM93_TEMP_OFFSET_TO_REG( + (LM93_TEMP_FROM_REG(data->boost[nr]) - hyst), mode); + + switch (nr) { + case 0: + reg = (data->boost_hyst[0] & 0xf0) | (reg & 0x0f); + break; + case 1: + reg = (reg << 4 & 0xf0) | (data->boost_hyst[0] & 0x0f); + break; + case 2: + reg = (data->boost_hyst[1] & 0xf0) | (reg & 0x0f); + break; + case 3: + default: + reg = (reg << 4 & 0xf0) | (data->boost_hyst[1] & 0x0f); + break; + } + + return reg; +} + +/* PWM: 0-255 per sensors documentation + REG: 0-13 as mapped below... right justified */ +typedef enum { LM93_PWM_MAP_HI_FREQ, LM93_PWM_MAP_LO_FREQ } pwm_freq_t; +static int lm93_pwm_map[2][16] = { + { + 0x00, /* 0.00% */ 0x40, /* 25.00% */ + 0x50, /* 31.25% */ 0x60, /* 37.50% */ + 0x70, /* 43.75% */ 0x80, /* 50.00% */ + 0x90, /* 56.25% */ 0xa0, /* 62.50% */ + 0xb0, /* 68.75% */ 0xc0, /* 75.00% */ + 0xd0, /* 81.25% */ 0xe0, /* 87.50% */ + 0xf0, /* 93.75% */ 0xff, /* 100.00% */ + 0xff, 0xff, /* 14, 15 are reserved and should never occur */ + }, + { + 0x00, /* 0.00% */ 0x40, /* 25.00% */ + 0x49, /* 28.57% */ 0x52, /* 32.14% */ + 0x5b, /* 35.71% */ 0x64, /* 39.29% */ + 0x6d, /* 42.86% */ 0x76, /* 46.43% */ + 0x80, /* 50.00% */ 0x89, /* 53.57% */ + 0x92, /* 57.14% */ 0xb6, /* 71.43% */ + 0xdb, /* 85.71% */ 0xff, /* 100.00% */ + 0xff, 0xff, /* 14, 15 are reserved and should never occur */ + }, +}; + +static int LM93_PWM_FROM_REG(u8 reg, pwm_freq_t freq) +{ + return lm93_pwm_map[freq][reg & 0x0f]; +} + +/* round up to nearest match */ +static u8 LM93_PWM_TO_REG(int pwm, pwm_freq_t freq) +{ + int i; + for (i = 0; i < 13; i++) + if (pwm <= lm93_pwm_map[freq][i]) + break; + + /* can fall through with i==13 */ + return (u8)i; +} + +static int LM93_FAN_FROM_REG(u16 regs) +{ + const u16 count = le16_to_cpu(regs) >> 2; + return count==0 ? -1 : count==0x3fff ? 0: 1350000 / count; +} + +/* + * RPM: (82.5 to 1350000) + * REG: 14-bits, LE, *left* justified + */ +static u16 LM93_FAN_TO_REG(long rpm) +{ + u16 count, regs; + + if (rpm == 0) { + count = 0x3fff; + } else { + rpm = SENSORS_LIMIT(rpm, 1, 1000000); + count = SENSORS_LIMIT((1350000 + rpm) / rpm, 1, 0x3ffe); + } + + regs = count << 2; + return cpu_to_le16(regs); +} + +/* PWM FREQ: HZ + REG: 0-7 as mapped below */ +static int lm93_pwm_freq_map[8] = { + 22500, 96, 84, 72, 60, 48, 36, 12 +}; + +static int LM93_PWM_FREQ_FROM_REG(u8 reg) +{ + return lm93_pwm_freq_map[reg & 0x07]; +} + +/* round up to nearest match */ +static u8 LM93_PWM_FREQ_TO_REG(int freq) +{ + int i; + for (i = 7; i > 0; i--) + if (freq <= lm93_pwm_freq_map[i]) + break; + + /* can fall through with i==0 */ + return (u8)i; +} + +/* TIME: 1/100 seconds + * REG: 0-7 as mapped below */ +static int lm93_spinup_time_map[8] = { + 0, 10, 25, 40, 70, 100, 200, 400, +}; + +static int LM93_SPINUP_TIME_FROM_REG(u8 reg) +{ + return lm93_spinup_time_map[reg >> 5 & 0x07]; +} + +/* round up to nearest match */ +static u8 LM93_SPINUP_TIME_TO_REG(int time) +{ + int i; + for (i = 0; i < 7; i++) + if (time <= lm93_spinup_time_map[i]) + break; + + /* can fall through with i==8 */ + return (u8)i; +} + +#define LM93_RAMP_MIN 0 +#define LM93_RAMP_MAX 75 + +static int LM93_RAMP_FROM_REG(u8 reg) +{ + return (reg & 0x0f) * 5; +} + +/* RAMP: 1/100 seconds + REG: 50mS/bit 4-bits right justified */ +static u8 LM93_RAMP_TO_REG(int ramp) +{ + ramp = SENSORS_LIMIT(ramp, LM93_RAMP_MIN, LM93_RAMP_MAX); + return (u8)((ramp + 2) / 5); +} + +/* PROCHOT: 0-255, 0 => 0%, 255 => > 96.6% + * REG: (same) */ +static u8 LM93_PROCHOT_TO_REG(long prochot) +{ + prochot = SENSORS_LIMIT(prochot, 0, 255); + return (u8)prochot; +} + +/* PROCHOT-INTERVAL: 73 - 37200 (1/100 seconds) + * REG: 0-9 as mapped below */ +static int lm93_interval_map[10] = { + 73, 146, 290, 580, 1170, 2330, 4660, 9320, 18600, 37200, +}; + +static int LM93_INTERVAL_FROM_REG(u8 reg) +{ + return lm93_interval_map[reg & 0x0f]; +} + +/* round up to nearest match */ +static u8 LM93_INTERVAL_TO_REG(long interval) +{ + int i; + for (i = 0; i < 9; i++) + if (interval <= lm93_interval_map[i]) + break; + + /* can fall through with i==9 */ + return (u8)i; +} + +/* GPIO: 0-255, GPIO0 is LSB + * REG: inverted */ +static unsigned LM93_GPI_FROM_REG(u8 reg) +{ + return ~reg & 0xff; +} + +/* alarm bitmask definitions + The LM93 has nearly 64 bits of error status... I've pared that down to + what I think is a useful subset in order to fit it into 32 bits. + + Especially note that the #VRD_HOT alarms are missing because we provide + that information as values in another sysfs file. + + If libsensors is extended to support 64 bit values, this could be revisited. +*/ +#define LM93_ALARM_IN1 0x00000001 +#define LM93_ALARM_IN2 0x00000002 +#define LM93_ALARM_IN3 0x00000004 +#define LM93_ALARM_IN4 0x00000008 +#define LM93_ALARM_IN5 0x00000010 +#define LM93_ALARM_IN6 0x00000020 +#define LM93_ALARM_IN7 0x00000040 +#define LM93_ALARM_IN8 0x00000080 +#define LM93_ALARM_IN9 0x00000100 +#define LM93_ALARM_IN10 0x00000200 +#define LM93_ALARM_IN11 0x00000400 +#define LM93_ALARM_IN12 0x00000800 +#define LM93_ALARM_IN13 0x00001000 +#define LM93_ALARM_IN14 0x00002000 +#define LM93_ALARM_IN15 0x00004000 +#define LM93_ALARM_IN16 0x00008000 +#define LM93_ALARM_FAN1 0x00010000 +#define LM93_ALARM_FAN2 0x00020000 +#define LM93_ALARM_FAN3 0x00040000 +#define LM93_ALARM_FAN4 0x00080000 +#define LM93_ALARM_PH1_ERR 0x00100000 +#define LM93_ALARM_PH2_ERR 0x00200000 +#define LM93_ALARM_SCSI1_ERR 0x00400000 +#define LM93_ALARM_SCSI2_ERR 0x00800000 +#define LM93_ALARM_DVDDP1_ERR 0x01000000 +#define LM93_ALARM_DVDDP2_ERR 0x02000000 +#define LM93_ALARM_D1_ERR 0x04000000 +#define LM93_ALARM_D2_ERR 0x08000000 +#define LM93_ALARM_TEMP1 0x10000000 +#define LM93_ALARM_TEMP2 0x20000000 +#define LM93_ALARM_TEMP3 0x40000000 + +static unsigned LM93_ALARMS_FROM_REG(struct block1_t b1) +{ + unsigned result; + result = b1.host_status_2 & 0x3f; + + if (vccp_limit_type[0]) + result |= (b1.host_status_4 & 0x10) << 2; + else + result |= b1.host_status_2 & 0x40; + + if (vccp_limit_type[1]) + result |= (b1.host_status_4 & 0x20) << 2; + else + result |= b1.host_status_2 & 0x80; + + result |= b1.host_status_3 << 8; + result |= (b1.fan_status & 0x0f) << 16; + result |= (b1.p1_prochot_status & 0x80) << 13; + result |= (b1.p2_prochot_status & 0x80) << 14; + result |= (b1.host_status_4 & 0xfc) << 20; + result |= (b1.host_status_1 & 0x07) << 28; + return result; +} + +#define MAX_RETRIES 5 + +static u8 lm93_read_byte(struct i2c_client *client, u8 reg) +{ + int value, i; + + /* retry in case of read errors */ + for (i=1; i<=MAX_RETRIES; i++) { + if ((value = i2c_smbus_read_byte_data(client, reg)) >= 0) { + return value; + } else { + dev_warn(&client->dev,"lm93: read byte data failed, " + "address 0x%02x.\n", reg); + mdelay(i + 3); + } + + } + + /* <TODO> what to return in case of error? */ + dev_err(&client->dev,"lm93: All read byte retries failed!!\n"); + return 0; +} + +static int lm93_write_byte(struct i2c_client *client, u8 reg, u8 value) +{ + int result; + + /* <TODO> how to handle write errors? */ + result = i2c_smbus_write_byte_data(client, reg, value); + + if (result < 0) + dev_warn(&client->dev,"lm93: write byte data failed, " + "0x%02x at address 0x%02x.\n", value, reg); + + return result; +} + +static u16 lm93_read_word(struct i2c_client *client, u8 reg) +{ + int value, i; + + /* retry in case of read errors */ + for (i=1; i<=MAX_RETRIES; i++) { + if ((value = i2c_smbus_read_word_data(client, reg)) >= 0) { + return value; + } else { + dev_warn(&client->dev,"lm93: read word data failed, " + "address 0x%02x.\n", reg); + mdelay(i + 3); + } + + } + + /* <TODO> what to return in case of error? */ + dev_err(&client->dev,"lm93: All read word retries failed!!\n"); + return 0; +} + +static int lm93_write_word(struct i2c_client *client, u8 reg, u16 value) +{ + int result; + + /* <TODO> how to handle write errors? */ + result = i2c_smbus_write_word_data(client, reg, value); + + if (result < 0) + dev_warn(&client->dev,"lm93: write word data failed, " + "0x%04x at address 0x%02x.\n", value, reg); + + return result; +} + +static u8 lm93_block_buffer[I2C_SMBUS_BLOCK_MAX]; + +/* + read block data into values, retry if not expected length + fbn => index to lm93_block_read_cmds table + (Fixed Block Number - section 14.5.2 of LM93 datasheet) +*/ +static void lm93_read_block(struct i2c_client *client, u8 fbn, u8 *values) +{ + int i, result=0; + + for (i = 1; i <= MAX_RETRIES; i++) { + result = i2c_smbus_read_block_data(client, + lm93_block_read_cmds[fbn].cmd, lm93_block_buffer); + + if (result == lm93_block_read_cmds[fbn].len) { + break; + } else { + dev_warn(&client->dev,"lm93: block read data failed, " + "command 0x%02x.\n", + lm93_block_read_cmds[fbn].cmd); + mdelay(i + 3); + } + } + + if (result == lm93_block_read_cmds[fbn].len) { + memcpy(values,lm93_block_buffer,lm93_block_read_cmds[fbn].len); + } else { + /* <TODO> what to do in case of error? */ + } +} + +static struct lm93_data *lm93_update_device(struct device *dev) +{ + struct i2c_client *client = to_i2c_client(dev); + struct lm93_data *data = i2c_get_clientdata(client); + const unsigned long interval = HZ + (HZ / 2); + + mutex_lock(&data->update_lock); + + if (time_after(jiffies, data->last_updated + interval) || + !data->valid) { + + data->update(data, client); + data->last_updated = jiffies; + data->valid = 1; + } + + mutex_unlock(&data->update_lock); + return data; +} + +/* update routine for data that has no corresponding SMBus block command */ +static void lm93_update_client_common(struct lm93_data *data, + struct i2c_client *client) +{ + int i; + u8 *ptr; + + /* temp1 - temp4: limits */ + for (i = 0; i < 4; i++) { + data->temp_lim[i].min = + lm93_read_byte(client, LM93_REG_TEMP_MIN(i)); + data->temp_lim[i].max = + lm93_read_byte(client, LM93_REG_TEMP_MAX(i)); + } + + /* config register */ + data->config = lm93_read_byte(client, LM93_REG_CONFIG); + + /* vid1 - vid2: values */ + for (i = 0; i < 2; i++) + data->vid[i] = lm93_read_byte(client, LM93_REG_VID(i)); + + /* prochot1 - prochot2: limits */ + for (i = 0; i < 2; i++) + data->prochot_max[i] = lm93_read_byte(client, + LM93_REG_PROCHOT_MAX(i)); + + /* vccp1 - vccp2: VID relative limits */ + for (i = 0; i < 2; i++) + data->vccp_limits[i] = lm93_read_byte(client, + LM93_REG_VCCP_LIMIT_OFF(i)); + + /* GPIO input state */ + data->gpi = lm93_read_byte(client, LM93_REG_GPI); + + /* #PROCHOT override state */ + data->prochot_override = lm93_read_byte(client, + LM93_REG_PROCHOT_OVERRIDE); + + /* #PROCHOT intervals */ + data->prochot_interval = lm93_read_byte(client, + LM93_REG_PROCHOT_INTERVAL); + + /* Fan Boost Termperature registers */ + for (i = 0; i < 4; i++) + data->boost[i] = lm93_read_byte(client, LM93_REG_BOOST(i)); + + /* Fan Boost Temperature Hyst. registers */ + data->boost_hyst[0] = lm93_read_byte(client, LM93_REG_BOOST_HYST_12); + data->boost_hyst[1] = lm93_read_byte(client, LM93_REG_BOOST_HYST_34); + + /* Temperature Zone Min. PWM & Hysteresis registers */ + data->auto_pwm_min_hyst[0] = + lm93_read_byte(client, LM93_REG_PWM_MIN_HYST_12); + data->auto_pwm_min_hyst[1] = + lm93_read_byte(client, LM93_REG_PWM_MIN_HYST_34); + + /* #PROCHOT & #VRDHOT PWM Ramp Control register */ + data->pwm_ramp_ctl = lm93_read_byte(client, LM93_REG_PWM_RAMP_CTL); + + /* misc setup registers */ + data->sfc1 = lm93_read_byte(client, LM93_REG_SFC1); + data->sfc2 = lm93_read_byte(client, LM93_REG_SFC2); + data->sf_tach_to_pwm = lm93_read_byte(client, + LM93_REG_SF_TACH_TO_PWM); + + /* write back alarm values to clear */ + for (i = 0, ptr = (u8 *)(&data->block1); i < 8; i++) + lm93_write_byte(client, LM93_REG_HOST_ERROR_1 + i, *(ptr + i)); +} + +/* update routine which uses SMBus block data commands */ +static void lm93_update_client_full(struct lm93_data *data, + struct i2c_client *client) +{ + dev_dbg(&client->dev,"starting device update (block data enabled)\n"); + + /* in1 - in16: values & limits */ + lm93_read_block(client, 3, (u8 *)(data->block3)); + lm93_read_block(client, 7, (u8 *)(data->block7)); + + /* temp1 - temp4: values */ + lm93_read_block(client, 2, (u8 *)(data->block2)); + + /* prochot1 - prochot2: values */ + lm93_read_block(client, 4, (u8 *)(data->block4)); + + /* fan1 - fan4: values & limits */ + lm93_read_block(client, 5, (u8 *)(data->block5)); + lm93_read_block(client, 8, (u8 *)(data->block8)); + + /* pmw control registers */ + lm93_read_block(client, 9, (u8 *)(data->block9)); + + /* alarm values */ + lm93_read_block(client, 1, (u8 *)(&data->block1)); + + /* auto/pwm registers */ + lm93_read_block(client, 10, (u8 *)(&data->block10)); + + lm93_update_client_common(data, client); +} + +/* update routine which uses SMBus byte/word data commands only */ +static void lm93_update_client_min(struct lm93_data *data, + struct i2c_client *client) +{ + int i,j; + u8 *ptr; + + dev_dbg(&client->dev,"starting device update (block data disabled)\n"); + + /* in1 - in16: values & limits */ + for (i = 0; i < 16; i++) { + data->block3[i] = + lm93_read_byte(client, LM93_REG_IN(i)); + data->block7[i].min = + lm93_read_byte(client, LM93_REG_IN_MIN(i)); + data->block7[i].max = + lm93_read_byte(client, LM93_REG_IN_MAX(i)); + } + + /* temp1 - temp4: values */ + for (i = 0; i < 4; i++) { + data->block2[i] = + lm93_read_byte(client, LM93_REG_TEMP(i)); + } + + /* prochot1 - prochot2: values */ + for (i = 0; i < 2; i++) { + data->block4[i].cur = + lm93_read_byte(client, LM93_REG_PROCHOT_CUR(i)); + data->block4[i].avg = + lm93_read_byte(client, LM93_REG_PROCHOT_AVG(i)); + } + + /* fan1 - fan4: values & limits */ + for (i = 0; i < 4; i++) { + data->block5[i] = + lm93_read_word(client, LM93_REG_FAN(i)); + data->block8[i] = + lm93_read_word(client, LM93_REG_FAN_MIN(i)); + } + + /* pwm control registers */ + for (i = 0; i < 2; i++) { + for (j = 0; j < 4; j++) { + data->block9[i][j] = + lm93_read_byte(client, LM93_REG_PWM_CTL(i,j)); + } + } + + /* alarm values */ + for (i = 0, ptr = (u8 *)(&data->block1); i < 8; i++) { + *(ptr + i) = + lm93_read_byte(client, LM93_REG_HOST_ERROR_1 + i); + } + + /* auto/pwm (base temp) registers */ + for (i = 0; i < 4; i++) { + data->block10.base[i] = + lm93_read_byte(client, LM93_REG_TEMP_BASE(i)); + } + + /* auto/pwm (offset temp) registers */ + for (i = 0; i < 12; i++) { + data->block10.offset[i] = + lm93_read_byte(client, LM93_REG_TEMP_OFFSET(i)); + } + + lm93_update_client_common(data, client); +} + +/* following are the sysfs callback functions */ +static ssize_t show_in(struct device *dev, struct device_attribute *attr, + char *buf) +{ + int nr = (to_sensor_dev_attr(attr))->index; + + struct lm93_data *data = lm93_update_device(dev); + return sprintf(buf, "%d\n", LM93_IN_FROM_REG(nr, data->block3[nr])); +} + +static SENSOR_DEVICE_ATTR(in1_input, S_IRUGO, show_in, NULL, 0); +static SENSOR_DEVICE_ATTR(in2_input, S_IRUGO, show_in, NULL, 1); +static SENSOR_DEVICE_ATTR(in3_input, S_IRUGO, show_in, NULL, 2); +static SENSOR_DEVICE_ATTR(in4_input, S_IRUGO, show_in, NULL, 3); +static SENSOR_DEVICE_ATTR(in5_input, S_IRUGO, show_in, NULL, 4); +static SENSOR_DEVICE_ATTR(in6_input, S_IRUGO, show_in, NULL, 5); +static SENSOR_DEVICE_ATTR(in7_input, S_IRUGO, show_in, NULL, 6); +static SENSOR_DEVICE_ATTR(in8_input, S_IRUGO, show_in, NULL, 7); +static SENSOR_DEVICE_ATTR(in9_input, S_IRUGO, show_in, NULL, 8); +static SENSOR_DEVICE_ATTR(in10_input, S_IRUGO, show_in, NULL, 9); +static SENSOR_DEVICE_ATTR(in11_input, S_IRUGO, show_in, NULL, 10); +static SENSOR_DEVICE_ATTR(in12_input, S_IRUGO, show_in, NULL, 11); +static SENSOR_DEVICE_ATTR(in13_input, S_IRUGO, show_in, NULL, 12); +static SENSOR_DEVICE_ATTR(in14_input, S_IRUGO, show_in, NULL, 13); +static SENSOR_DEVICE_ATTR(in15_input, S_IRUGO, show_in, NULL, 14); +static SENSOR_DEVICE_ATTR(in16_input, S_IRUGO, show_in, NULL, 15); + +static ssize_t show_in_min(struct device *dev, + struct device_attribute *attr, char *buf) +{ + int nr = (to_sensor_dev_attr(attr))->index; + struct lm93_data *data = lm93_update_device(dev); + int vccp = nr - 6; + long rc, vid; + + if ((nr==6 || nr==7) && (vccp_limit_type[vccp])) { + vid = LM93_VID_FROM_REG(data->vid[vccp]); + rc = LM93_IN_MIN_FROM_REG(data->vccp_limits[vccp], vid); + } + else { + rc = LM93_IN_FROM_REG(nr, data->block7[nr].min); \ + } + return sprintf(buf, "%ld\n", rc); \ +} + +static ssize_t store_in_min(struct device *dev, struct device_attribute *attr, + const char *buf, size_t count) +{ + int nr = (to_sensor_dev_attr(attr))->index; + struct i2c_client *client = to_i2c_client(dev); + struct lm93_data *data = i2c_get_clientdata(client); + u32 val = simple_strtoul(buf, NULL, 10); + int vccp = nr - 6; + long vid; + + mutex_lock(&data->update_lock); + if ((nr==6 || nr==7) && (vccp_limit_type[vccp])) { + vid = LM93_VID_FROM_REG(data->vid[vccp]); + data->vccp_limits[vccp] = (data->vccp_limits[vccp] & 0xf0) | + LM93_IN_REL_TO_REG(val, 0, vid); + lm93_write_byte(client, LM93_REG_VCCP_LIMIT_OFF(vccp), + data->vccp_limits[vccp]); + } + else { + data->block7[nr].min = LM93_IN_TO_REG(nr,val); + lm93_write_byte(client, LM93_REG_IN_MIN(nr), + data->block7[nr].min); + } + mutex_unlock(&data->update_lock); + return count; +} + +static SENSOR_DEVICE_ATTR(in1_min, S_IWUSR | S_IRUGO, + show_in_min, store_in_min, 0); +static SENSOR_DEVICE_ATTR(in2_min, S_IWUSR | S_IRUGO, + show_in_min, store_in_min, 1); +static SENSOR_DEVICE_ATTR(in3_min, S_IWUSR | S_IRUGO, + show_in_min, store_in_min, 2); +static SENSOR_DEVICE_ATTR(in4_min, S_IWUSR | S_IRUGO, + show_in_min, store_in_min, 3); +static SENSOR_DEVICE_ATTR(in5_min, S_IWUSR | S_IRUGO, + show_in_min, store_in_min, 4); +static SENSOR_DEVICE_ATTR(in6_min, S_IWUSR | S_IRUGO, + show_in_min, store_in_min, 5); +static SENSOR_DEVICE_ATTR(in7_min, S_IWUSR | S_IRUGO, + show_in_min, store_in_min, 6); +static SENSOR_DEVICE_ATTR(in8_min, S_IWUSR | S_IRUGO, + show_in_min, store_in_min, 7); +static SENSOR_DEVICE_ATTR(in9_min, S_IWUSR | S_IRUGO, + show_in_min, store_in_min, 8); +static SENSOR_DEVICE_ATTR(in10_min, S_IWUSR | S_IRUGO, + show_in_min, store_in_min, 9); +static SENSOR_DEVICE_ATTR(in11_min, S_IWUSR | S_IRUGO, + show_in_min, store_in_min, 10); +static SENSOR_DEVICE_ATTR(in12_min, S_IWUSR | S_IRUGO, + show_in_min, store_in_min, 11); +static SENSOR_DEVICE_ATTR(in13_min, S_IWUSR | S_IRUGO, + show_in_min, store_in_min, 12); +static SENSOR_DEVICE_ATTR(in14_min, S_IWUSR | S_IRUGO, + show_in_min, store_in_min, 13); +static SENSOR_DEVICE_ATTR(in15_min, S_IWUSR | S_IRUGO, + show_in_min, store_in_min, 14); +static SENSOR_DEVICE_ATTR(in16_min, S_IWUSR | S_IRUGO, + show_in_min, store_in_min, 15); + +static ssize_t show_in_max(struct device *dev, + struct device_attribute *attr, char *buf) +{ + int nr = (to_sensor_dev_attr(attr))->index; + struct lm93_data *data = lm93_update_device(dev); + int vccp = nr - 6; + long rc, vid; + + if ((nr==6 || nr==7) && (vccp_limit_type[vccp])) { + vid = LM93_VID_FROM_REG(data->vid[vccp]); + rc = LM93_IN_MAX_FROM_REG(data->vccp_limits[vccp],vid); + } + else { + rc = LM93_IN_FROM_REG(nr,data->block7[nr].max); \ + } + return sprintf(buf,"%ld\n",rc); \ +} + +static ssize_t store_in_max(struct device *dev, struct device_attribute *attr, + const char *buf, size_t count) +{ + int nr = (to_sensor_dev_attr(attr))->index; + struct i2c_client *client = to_i2c_client(dev); + struct lm93_data *data = i2c_get_clientdata(client); + u32 val = simple_strtoul(buf, NULL, 10); + int vccp = nr - 6; + long vid; + + mutex_lock(&data->update_lock); + if ((nr==6 || nr==7) && (vccp_limit_type[vccp])) { + vid = LM93_VID_FROM_REG(data->vid[vccp]); + data->vccp_limits[vccp] = (data->vccp_limits[vccp] & 0x0f) | + LM93_IN_REL_TO_REG(val, 1, vid); + lm93_write_byte(client, LM93_REG_VCCP_LIMIT_OFF(vccp), + data->vccp_limits[vccp]); + } + else { + data->block7[nr].max = LM93_IN_TO_REG(nr,val); + lm93_write_byte(client, LM93_REG_IN_MAX(nr), + data->block7[nr].max); + } + mutex_unlock(&data->update_lock); + return count; +} + +static SENSOR_DEVICE_ATTR(in1_max, S_IWUSR | S_IRUGO, + show_in_max, store_in_max, 0); +static SENSOR_DEVICE_ATTR(in2_max, S_IWUSR | S_IRUGO, + show_in_max, store_in_max, 1); +static SENSOR_DEVICE_ATTR(in3_max, S_IWUSR | S_IRUGO, + show_in_max, store_in_max, 2); +static SENSOR_DEVICE_ATTR(in4_max, S_IWUSR | S_IRUGO, + show_in_max, store_in_max, 3); +static SENSOR_DEVICE_ATTR(in5_max, S_IWUSR | S_IRUGO, + show_in_max, store_in_max, 4); +static SENSOR_DEVICE_ATTR(in6_max, S_IWUSR | S_IRUGO, + show_in_max, store_in_max, 5); +static SENSOR_DEVICE_ATTR(in7_max, S_IWUSR | S_IRUGO, + show_in_max, store_in_max, 6); +static SENSOR_DEVICE_ATTR(in8_max, S_IWUSR | S_IRUGO, + show_in_max, store_in_max, 7); +static SENSOR_DEVICE_ATTR(in9_max, S_IWUSR | S_IRUGO, + show_in_max, store_in_max, 8); +static SENSOR_DEVICE_ATTR(in10_max, S_IWUSR | S_IRUGO, + show_in_max, store_in_max, 9); +static SENSOR_DEVICE_ATTR(in11_max, S_IWUSR | S_IRUGO, + show_in_max, store_in_max, 10); +static SENSOR_DEVICE_ATTR(in12_max, S_IWUSR | S_IRUGO, + show_in_max, store_in_max, 11); +static SENSOR_DEVICE_ATTR(in13_max, S_IWUSR | S_IRUGO, + show_in_max, store_in_max, 12); +static SENSOR_DEVICE_ATTR(in14_max, S_IWUSR | S_IRUGO, + show_in_max, store_in_max, 13); +static SENSOR_DEVICE_ATTR(in15_max, S_IWUSR | S_IRUGO, + show_in_max, store_in_max, 14); +static SENSOR_DEVICE_ATTR(in16_max, S_IWUSR | S_IRUGO, + show_in_max, store_in_max, 15); + +static ssize_t show_temp(struct device *dev, + struct device_attribute *attr, char *buf) +{ + int nr = (to_sensor_dev_attr(attr))->index; + struct lm93_data *data = lm93_update_device(dev); + return sprintf(buf,"%d\n",LM93_TEMP_FROM_REG(data->block2[nr])); +} + +static SENSOR_DEVICE_ATTR(temp1_input, S_IRUGO, show_temp, NULL, 0); +static SENSOR_DEVICE_ATTR(temp2_input, S_IRUGO, show_temp, NULL, 1); +static SENSOR_DEVICE_ATTR(temp3_input, S_IRUGO, show_temp, NULL, 2); + +static ssize_t show_temp_min(struct device *dev, + struct device_attribute *attr, char *buf) +{ + int nr = (to_sensor_dev_attr(attr))->index; + struct lm93_data *data = lm93_update_device(dev); + return sprintf(buf,"%d\n",LM93_TEMP_FROM_REG(data->temp_lim[nr].min)); +} + +static ssize_t store_temp_min(struct device *dev, struct device_attribute *attr, + const char *buf, size_t count) +{ + int nr = (to_sensor_dev_attr(attr))->index; + struct i2c_client *client = to_i2c_client(dev); + struct lm93_data *data = i2c_get_clientdata(client); + u32 val = simple_strtoul(buf, NULL, 10); + + mutex_lock(&data->update_lock); + data->temp_lim[nr].min = LM93_TEMP_TO_REG(val); + lm93_write_byte(client, LM93_REG_TEMP_MIN(nr), data->temp_lim[nr].min); + mutex_unlock(&data->update_lock); + return count; +} + +static SENSOR_DEVICE_ATTR(temp1_min, S_IWUSR | S_IRUGO, + show_temp_min, store_temp_min, 0); +static SENSOR_DEVICE_ATTR(temp2_min, S_IWUSR | S_IRUGO, + show_temp_min, store_temp_min, 1); +static SENSOR_DEVICE_ATTR(temp3_min, S_IWUSR | S_IRUGO, + show_temp_min, store_temp_min, 2); + +static ssize_t show_temp_max(struct device *dev, + struct device_attribute *attr, char *buf) +{ + int nr = (to_sensor_dev_attr(attr))->index; + struct lm93_data *data = lm93_update_device(dev); + return sprintf(buf,"%d\n",LM93_TEMP_FROM_REG(data->temp_lim[nr].max)); +} + +static ssize_t store_temp_max(struct device *dev, struct device_attribute *attr, + const char *buf, size_t count) +{ + int nr = (to_sensor_dev_attr(attr))->index; + struct i2c_client *client = to_i2c_client(dev); + struct lm93_data *data = i2c_get_clientdata(client); + u32 val = simple_strtoul(buf, NULL, 10); + + mutex_lock(&data->update_lock); + data->temp_lim[nr].max = LM93_TEMP_TO_REG(val); + lm93_write_byte(client, LM93_REG_TEMP_MAX(nr), data->temp_lim[nr].max); + mutex_unlock(&data->update_lock); + return count; +} + +static SENSOR_DEVICE_ATTR(temp1_max, S_IWUSR | S_IRUGO, + show_temp_max, store_temp_max, 0); +static SENSOR_DEVICE_ATTR(temp2_max, S_IWUSR | S_IRUGO, + show_temp_max, store_temp_max, 1); +static SENSOR_DEVICE_ATTR(temp3_max, S_IWUSR | S_IRUGO, + show_temp_max, store_temp_max, 2); + +static ssize_t show_temp_auto_base(struct device *dev, + struct device_attribute *attr, char *buf) +{ + int nr = (to_sensor_dev_attr(attr))->index; + struct lm93_data *data = lm93_update_device(dev); + return sprintf(buf,"%d\n",LM93_TEMP_FROM_REG(data->block10.base[nr])); +} + +static ssize_t store_temp_auto_base(struct device *dev, + struct device_attribute *attr, + const char *buf, size_t count) +{ + int nr = (to_sensor_dev_attr(attr))->index; + struct i2c_client *client = to_i2c_client(dev); + struct lm93_data *data = i2c_get_clientdata(client); + u32 val = simple_strtoul(buf, NULL, 10); + + mutex_lock(&data->update_lock); + data->block10.base[nr] = LM93_TEMP_TO_REG(val); + lm93_write_byte(client, LM93_REG_TEMP_BASE(nr), data->block10.base[nr]); + mutex_unlock(&data->update_lock); + return count; +} + +static SENSOR_DEVICE_ATTR(temp1_auto_base, S_IWUSR | S_IRUGO, + show_temp_auto_base, store_temp_auto_base, 0); +static SENSOR_DEVICE_ATTR(temp2_auto_base, S_IWUSR | S_IRUGO, + show_temp_auto_base, store_temp_auto_base, 1); +static SENSOR_DEVICE_ATTR(temp3_auto_base, S_IWUSR | S_IRUGO, + show_temp_auto_base, store_temp_auto_base, 2); + +static ssize_t show_temp_auto_boost(struct device *dev, + struct device_attribute *attr,char *buf) +{ + int nr = (to_sensor_dev_attr(attr))->index; + struct lm93_data *data = lm93_update_device(dev); + return sprintf(buf,"%d\n",LM93_TEMP_FROM_REG(data->boost[nr])); +} + +static ssize_t store_temp_auto_boost(struct device *dev, + struct device_attribute *attr, + const char *buf, size_t count) +{ + int nr = (to_sensor_dev_attr(attr))->index; + struct i2c_client *client = to_i2c_client(dev); + struct lm93_data *data = i2c_get_clientdata(client); + u32 val = simple_strtoul(buf, NULL, 10); + + mutex_lock(&data->update_lock); + data->boost[nr] = LM93_TEMP_TO_REG(val); + lm93_write_byte(client, LM93_REG_BOOST(nr), data->boost[nr]); + mutex_unlock(&data->update_lock); + return count; +} + +static SENSOR_DEVICE_ATTR(temp1_auto_boost, S_IWUSR | S_IRUGO, + show_temp_auto_boost, store_temp_auto_boost, 0); +static SENSOR_DEVICE_ATTR(temp2_auto_boost, S_IWUSR | S_IRUGO, + show_temp_auto_boost, store_temp_auto_boost, 1); +static SENSOR_DEVICE_ATTR(temp3_auto_boost, S_IWUSR | S_IRUGO, + show_temp_auto_boost, store_temp_auto_boost, 2); + +static ssize_t show_temp_auto_boost_hyst(struct device *dev, + struct device_attribute *attr, + char *buf) +{ + int nr = (to_sensor_dev_attr(attr))->index; + struct lm93_data *data = lm93_update_device(dev); + int mode = LM93_TEMP_OFFSET_MODE_FROM_REG(data->sfc2, nr); + return sprintf(buf,"%d\n", + LM93_AUTO_BOOST_HYST_FROM_REGS(data, nr, mode)); +} + +static ssize_t store_temp_auto_boost_hyst(struct device *dev, + struct device_attribute *attr, + const char *buf, size_t count) +{ + int nr = (to_sensor_dev_attr(attr))->index; + struct i2c_client *client = to_i2c_client(dev); + struct lm93_data *data = i2c_get_clientdata(client); + u32 val = simple_strtoul(buf, NULL, 10); + + mutex_lock(&data->update_lock); + /* force 0.5C/bit mode */ + data->sfc2 = lm93_read_byte(client, LM93_REG_SFC2); + data->sfc2 |= ((nr < 2) ? 0x10 : 0x20); + lm93_write_byte(client, LM93_REG_SFC2, data->sfc2); + data->boost_hyst[nr/2] = LM93_AUTO_BOOST_HYST_TO_REG(data, val, nr, 1); + lm93_write_byte(client, LM93_REG_BOOST_HYST(nr), + data->boost_hyst[nr/2]); + mutex_unlock(&data->update_lock); + return count; +} + +static SENSOR_DEVICE_ATTR(temp1_auto_boost_hyst, S_IWUSR | S_IRUGO, + show_temp_auto_boost_hyst, + store_temp_auto_boost_hyst, 0); +static SENSOR_DEVICE_ATTR(temp2_auto_boost_hyst, S_IWUSR | S_IRUGO, + show_temp_auto_boost_hyst, + store_temp_auto_boost_hyst, 1); +static SENSOR_DEVICE_ATTR(temp3_auto_boost_hyst, S_IWUSR | S_IRUGO, + show_temp_auto_boost_hyst, + store_temp_auto_boost_hyst, 2); + +static ssize_t show_temp_auto_offset(struct device *dev, + struct device_attribute *attr, char *buf) +{ + struct sensor_device_attribute_2 *s_attr = to_sensor_dev_attr_2(attr); + int nr = s_attr->index; + int ofs = s_attr->nr; + struct lm93_data *data = lm93_update_device(dev); + int mode = LM93_TEMP_OFFSET_MODE_FROM_REG(data->sfc2, nr); + return sprintf(buf,"%d\n", + LM93_TEMP_AUTO_OFFSET_FROM_REG(data->block10.offset[ofs], + nr,mode)); +} + +static ssize_t store_temp_auto_offset(struct device *dev, + struct device_attribute *attr, + const char *buf, size_t count) +{ + struct sensor_device_attribute_2 *s_attr = to_sensor_dev_attr_2(attr); + int nr = s_attr->index; + int ofs = s_attr->nr; + struct i2c_client *client = to_i2c_client(dev); + struct lm93_data *data = i2c_get_clientdata(client); + u32 val = simple_strtoul(buf, NULL, 10); + + mutex_lock(&data->update_lock); + /* force 0.5C/bit mode */ + data->sfc2 = lm93_read_byte(client, LM93_REG_SFC2); + data->sfc2 |= ((nr < 2) ? 0x10 : 0x20); + lm93_write_byte(client, LM93_REG_SFC2, data->sfc2); + data->block10.offset[ofs] = LM93_TEMP_AUTO_OFFSET_TO_REG( + data->block10.offset[ofs], val, nr, 1); + lm93_write_byte(client, LM93_REG_TEMP_OFFSET(ofs), + data->block10.offset[ofs]); + mutex_unlock(&data->update_lock); + return count; +} + +static SENSOR_DEVICE_ATTR_2(temp1_auto_offset1, S_IWUSR | S_IRUGO, + show_temp_auto_offset, store_temp_auto_offset, 0, 0); +static SENSOR_DEVICE_ATTR_2(temp1_auto_offset2, S_IWUSR | S_IRUGO, + show_temp_auto_offset, store_temp_auto_offset, 1, 0); +static SENSOR_DEVICE_ATTR_2(temp1_auto_offset3, S_IWUSR | S_IRUGO, + show_temp_auto_offset, store_temp_auto_offset, 2, 0); +static SENSOR_DEVICE_ATTR_2(temp1_auto_offset4, S_IWUSR | S_IRUGO, + show_temp_auto_offset, store_temp_auto_offset, 3, 0); +static SENSOR_DEVICE_ATTR_2(temp1_auto_offset5, S_IWUSR | S_IRUGO, + show_temp_auto_offset, store_temp_auto_offset, 4, 0); +static SENSOR_DEVICE_ATTR_2(temp1_auto_offset6, S_IWUSR | S_IRUGO, + show_temp_auto_offset, store_temp_auto_offset, 5, 0); +static SENSOR_DEVICE_ATTR_2(temp1_auto_offset7, S_IWUSR | S_IRUGO, + show_temp_auto_offset, store_temp_auto_offset, 6, 0); +static SENSOR_DEVICE_ATTR_2(temp1_auto_offset8, S_IWUSR | S_IRUGO, + show_temp_auto_offset, store_temp_auto_offset, 7, 0); +static SENSOR_DEVICE_ATTR_2(temp1_auto_offset9, S_IWUSR | S_IRUGO, + show_temp_auto_offset, store_temp_auto_offset, 8, 0); +static SENSOR_DEVICE_ATTR_2(temp1_auto_offset10, S_IWUSR | S_IRUGO, + show_temp_auto_offset, store_temp_auto_offset, 9, 0); +static SENSOR_DEVICE_ATTR_2(temp1_auto_offset11, S_IWUSR | S_IRUGO, + show_temp_auto_offset, store_temp_auto_offset, 10, 0); +static SENSOR_DEVICE_ATTR_2(temp1_auto_offset12, S_IWUSR | S_IRUGO, + show_temp_auto_offset, store_temp_auto_offset, 11, 0); +static SENSOR_DEVICE_ATTR_2(temp2_auto_offset1, S_IWUSR | S_IRUGO, + show_temp_auto_offset, store_temp_auto_offset, 0, 1); +static SENSOR_DEVICE_ATTR_2(temp2_auto_offset2, S_IWUSR | S_IRUGO, + show_temp_auto_offset, store_temp_auto_offset, 1, 1); +static SENSOR_DEVICE_ATTR_2(temp2_auto_offset3, S_IWUSR | S_IRUGO, + show_temp_auto_offset, store_temp_auto_offset, 2, 1); +static SENSOR_DEVICE_ATTR_2(temp2_auto_offset4, S_IWUSR | S_IRUGO, + show_temp_auto_offset, store_temp_auto_offset, 3, 1); +static SENSOR_DEVICE_ATTR_2(temp2_auto_offset5, S_IWUSR | S_IRUGO, + show_temp_auto_offset, store_temp_auto_offset, 4, 1); +static SENSOR_DEVICE_ATTR_2(temp2_auto_offset6, S_IWUSR | S_IRUGO, + show_temp_auto_offset, store_temp_auto_offset, 5, 1); +static SENSOR_DEVICE_ATTR_2(temp2_auto_offset7, S_IWUSR | S_IRUGO, + show_temp_auto_offset, store_temp_auto_offset, 6, 1); +static SENSOR_DEVICE_ATTR_2(temp2_auto_offset8, S_IWUSR | S_IRUGO, + show_temp_auto_offset, store_temp_auto_offset, 7, 1); +static SENSOR_DEVICE_ATTR_2(temp2_auto_offset9, S_IWUSR | S_IRUGO, + show_temp_auto_offset, store_temp_auto_offset, 8, 1); +static SENSOR_DEVICE_ATTR_2(temp2_auto_offset10, S_IWUSR | S_IRUGO, + show_temp_auto_offset, store_temp_auto_offset, 9, 1); +static SENSOR_DEVICE_ATTR_2(temp2_auto_offset11, S_IWUSR | S_IRUGO, + show_temp_auto_offset, store_temp_auto_offset, 10, 1); +static SENSOR_DEVICE_ATTR_2(temp2_auto_offset12, S_IWUSR | S_IRUGO, + show_temp_auto_offset, store_temp_auto_offset, 11, 1); +static SENSOR_DEVICE_ATTR_2(temp3_auto_offset1, S_IWUSR | S_IRUGO, + show_temp_auto_offset, store_temp_auto_offset, 0, 2); +static SENSOR_DEVICE_ATTR_2(temp3_auto_offset2, S_IWUSR | S_IRUGO, + show_temp_auto_offset, store_temp_auto_offset, 1, 2); +static SENSOR_DEVICE_ATTR_2(temp3_auto_offset3, S_IWUSR | S_IRUGO, + show_temp_auto_offset, store_temp_auto_offset, 2, 2); +static SENSOR_DEVICE_ATTR_2(temp3_auto_offset4, S_IWUSR | S_IRUGO, + show_temp_auto_offset, store_temp_auto_offset, 3, 2); +static SENSOR_DEVICE_ATTR_2(temp3_auto_offset5, S_IWUSR | S_IRUGO, + show_temp_auto_offset, store_temp_auto_offset, 4, 2); +static SENSOR_DEVICE_ATTR_2(temp3_auto_offset6, S_IWUSR | S_IRUGO, + show_temp_auto_offset, store_temp_auto_offset, 5, 2); +static SENSOR_DEVICE_ATTR_2(temp3_auto_offset7, S_IWUSR | S_IRUGO, + show_temp_auto_offset, store_temp_auto_offset, 6, 2); +static SENSOR_DEVICE_ATTR_2(temp3_auto_offset8, S_IWUSR | S_IRUGO, + show_temp_auto_offset, store_temp_auto_offset, 7, 2); +static SENSOR_DEVICE_ATTR_2(temp3_auto_offset9, S_IWUSR | S_IRUGO, + show_temp_auto_offset, store_temp_auto_offset, 8, 2); +static SENSOR_DEVICE_ATTR_2(temp3_auto_offset10, S_IWUSR | S_IRUGO, + show_temp_auto_offset, store_temp_auto_offset, 9, 2); +static SENSOR_DEVICE_ATTR_2(temp3_auto_offset11, S_IWUSR | S_IRUGO, + show_temp_auto_offset, store_temp_auto_offset, 10, 2); +static SENSOR_DEVICE_ATTR_2(temp3_auto_offset12, S_IWUSR | S_IRUGO, + show_temp_auto_offset, store_temp_auto_offset, 11, 2); + +static ssize_t show_temp_auto_pwm_min(struct device *dev, + struct device_attribute *attr, char *buf) +{ + int nr = (to_sensor_dev_attr(attr))->index; + u8 reg, ctl4; + struct lm93_data *data = lm93_update_device(dev); + reg = data->auto_pwm_min_hyst[nr/2] >> 4 & 0x0f; + ctl4 = data->block9[nr][LM93_PWM_CTL4]; + return sprintf(buf,"%d\n",LM93_PWM_FROM_REG(reg, (ctl4 & 0x07) ? + LM93_PWM_MAP_LO_FREQ : LM93_PWM_MAP_HI_FREQ)); +} + +static ssize_t store_temp_auto_pwm_min(struct device *dev, + struct device_attribute *attr, + const char *buf, size_t count) +{ + int nr = (to_sensor_dev_attr(attr))->index; + struct i2c_client *client = to_i2c_client(dev); + struct lm93_data *data = i2c_get_clientdata(client); + u32 val = simple_strtoul(buf, NULL, 10); + u8 reg, ctl4; + + mutex_lock(&data->update_lock); + reg = lm93_read_byte(client, LM93_REG_PWM_MIN_HYST(nr)); + ctl4 = lm93_read_byte(client, LM93_REG_PWM_CTL(nr,LM93_PWM_CTL4)); + reg = (reg & 0x0f) | + LM93_PWM_TO_REG(val, (ctl4 & 0x07) ? + LM93_PWM_MAP_LO_FREQ : + LM93_PWM_MAP_HI_FREQ) << 4; + data->auto_pwm_min_hyst[nr/2] = reg; + lm93_write_byte(client, LM93_REG_PWM_MIN_HYST(nr), reg); + mutex_unlock(&data->update_lock); + return count; +} + +static SENSOR_DEVICE_ATTR(temp1_auto_pwm_min, S_IWUSR | S_IRUGO, + show_temp_auto_pwm_min, + store_temp_auto_pwm_min, 0); +static SENSOR_DEVICE_ATTR(temp2_auto_pwm_min, S_IWUSR | S_IRUGO, + show_temp_auto_pwm_min, + store_temp_auto_pwm_min, 1); +static SENSOR_DEVICE_ATTR(temp3_auto_pwm_min, S_IWUSR | S_IRUGO, + show_temp_auto_pwm_min, + store_temp_auto_pwm_min, 2); + +static ssize_t show_temp_auto_offset_hyst(struct device *dev, + struct device_attribute *attr, char *buf) +{ + int nr = (to_sensor_dev_attr(attr))->index; + struct lm93_data *data = lm93_update_device(dev); + int mode = LM93_TEMP_OFFSET_MODE_FROM_REG(data->sfc2, nr); + return sprintf(buf,"%d\n",LM93_TEMP_OFFSET_FROM_REG( + data->auto_pwm_min_hyst[nr/2], mode)); +} + +static ssize_t store_temp_auto_offset_hyst(struct device *dev, + struct device_attribute *attr, + const char *buf, size_t count) +{ + int nr = (to_sensor_dev_attr(attr))->index; + struct i2c_client *client = to_i2c_client(dev); + struct lm93_data *data = i2c_get_clientdata(client); + u32 val = simple_strtoul(buf, NULL, 10); + u8 reg; + + mutex_lock(&data->update_lock); + /* force 0.5C/bit mode */ + data->sfc2 = lm93_read_byte(client, LM93_REG_SFC2); + data->sfc2 |= ((nr < 2) ? 0x10 : 0x20); + lm93_write_byte(client, LM93_REG_SFC2, data->sfc2); + reg = data->auto_pwm_min_hyst[nr/2]; + reg = (reg & 0xf0) | (LM93_TEMP_OFFSET_TO_REG(val, 1) & 0x0f); + data->auto_pwm_min_hyst[nr/2] = reg; + lm93_write_byte(client, LM93_REG_PWM_MIN_HYST(nr), reg); + mutex_unlock(&data->update_lock); + return count; +} + +static SENSOR_DEVICE_ATTR(temp1_auto_offset_hyst, S_IWUSR | S_IRUGO, + show_temp_auto_offset_hyst, + store_temp_auto_offset_hyst, 0); +static SENSOR_DEVICE_ATTR(temp2_auto_offset_hyst, S_IWUSR | S_IRUGO, + show_temp_auto_offset_hyst, + store_temp_auto_offset_hyst, 1); +static SENSOR_DEVICE_ATTR(temp3_auto_offset_hyst, S_IWUSR | S_IRUGO, + show_temp_auto_offset_hyst, + store_temp_auto_offset_hyst, 2); + +static ssize_t show_fan_input(struct device *dev, + struct device_attribute *attr, char *buf) +{ + struct sensor_device_attribute *s_attr = to_sensor_dev_attr(attr); + int nr = s_attr->index; + struct lm93_data *data = lm93_update_device(dev); + + return sprintf(buf,"%d\n",LM93_FAN_FROM_REG(data->block5[nr])); +} + +static SENSOR_DEVICE_ATTR(fan1_input, S_IRUGO, show_fan_input, NULL, 0); +static SENSOR_DEVICE_ATTR(fan2_input, S_IRUGO, show_fan_input, NULL, 1); +static SENSOR_DEVICE_ATTR(fan3_input, S_IRUGO, show_fan_input, NULL, 2); +static SENSOR_DEVICE_ATTR(fan4_input, S_IRUGO, show_fan_input, NULL, 3); + +static ssize_t show_fan_min(struct device *dev, + struct device_attribute *attr, char *buf) +{ + int nr = (to_sensor_dev_attr(attr))->index; + struct lm93_data *data = lm93_update_device(dev); + + return sprintf(buf,"%d\n",LM93_FAN_FROM_REG(data->block8[nr])); +} + +static ssize_t store_fan_min(struct device *dev, struct device_attribute *attr, + const char *buf, size_t count) +{ + int nr = (to_sensor_dev_attr(attr))->index; + struct i2c_client *client = to_i2c_client(dev); + struct lm93_data *data = i2c_get_clientdata(client); + u32 val = simple_strtoul(buf, NULL, 10); + + mutex_lock(&data->update_lock); + data->block8[nr] = LM93_FAN_TO_REG(val); + lm93_write_word(client,LM93_REG_FAN_MIN(nr),data->block8[nr]); + mutex_unlock(&data->update_lock); + return count; +} + +static SENSOR_DEVICE_ATTR(fan1_min, S_IWUSR | S_IRUGO, + show_fan_min, store_fan_min, 0); +static SENSOR_DEVICE_ATTR(fan2_min, S_IWUSR | S_IRUGO, + show_fan_min, store_fan_min, 1); +static SENSOR_DEVICE_ATTR(fan3_min, S_IWUSR | S_IRUGO, + show_fan_min, store_fan_min, 2); +static SENSOR_DEVICE_ATTR(fan4_min, S_IWUSR | S_IRUGO, + show_fan_min, store_fan_min, 3); + +/* some tedious bit-twiddling here to deal with the register format: + + data->sf_tach_to_pwm: (tach to pwm mapping bits) + + bit | 7 | 6 | 5 | 4 | 3 | 2 | 1 | 0 + T4:P2 T4:P1 T3:P2 T3:P1 T2:P2 T2:P1 T1:P2 T1:P1 + + data->sfc2: (enable bits) + + bit | 3 | 2 | 1 | 0 + T4 T3 T2 T1 +*/ + +static ssize_t show_fan_smart_tach(struct device *dev, + struct device_attribute *attr, char *buf) +{ + int nr = (to_sensor_dev_attr(attr))->index; + struct lm93_data *data = lm93_update_device(dev); + long rc = 0; + int mapping; + + /* extract the relevant mapping */ + mapping = (data->sf_tach_to_pwm >> (nr * 2)) & 0x03; + + /* if there's a mapping and it's enabled */ + if (mapping && ((data->sfc2 >> nr) & 0x01)) + rc = mapping; + return sprintf(buf,"%ld\n",rc); +} + +/* helper function - must grab data->update_lock before calling + fan is 0-3, indicating fan1-fan4 */ +static void lm93_write_fan_smart_tach(struct i2c_client *client, + struct lm93_data *data, int fan, long value) +{ + /* insert the new mapping and write it out */ + data->sf_tach_to_pwm = lm93_read_byte(client, LM93_REG_SF_TACH_TO_PWM); + data->sf_tach_to_pwm &= ~(0x3 << fan * 2); + data->sf_tach_to_pwm |= value << fan * 2; + lm93_write_byte(client, LM93_REG_SF_TACH_TO_PWM, data->sf_tach_to_pwm); + + /* insert the enable bit and write it out */ + data->sfc2 = lm93_read_byte(client, LM93_REG_SFC2); + if (value) + data->sfc2 |= 1 << fan; + else + data->sfc2 &= ~(1 << fan); + lm93_write_byte(client, LM93_REG_SFC2, data->sfc2); +} + +static ssize_t store_fan_smart_tach(struct device *dev, + struct device_attribute *attr, + const char *buf, size_t count) +{ + int nr = (to_sensor_dev_attr(attr))->index; + struct i2c_client *client = to_i2c_client(dev); + struct lm93_data *data = i2c_get_clientdata(client); + u32 val = simple_strtoul(buf, NULL, 10); + + mutex_lock(&data->update_lock); + /* sanity test, ignore the write otherwise */ + if (0 <= val && val <= 2) { + /* can't enable if pwm freq is 22.5KHz */ + if (val) { + u8 ctl4 = lm93_read_byte(client, + LM93_REG_PWM_CTL(val-1,LM93_PWM_CTL4)); + if ((ctl4 & 0x07) == 0) + val = 0; + } + lm93_write_fan_smart_tach(client, data, nr, val); + } + mutex_unlock(&data->update_lock); + return count; +} + +static SENSOR_DEVICE_ATTR(fan1_smart_tach, S_IWUSR | S_IRUGO, + show_fan_smart_tach, store_fan_smart_tach, 0); +static SENSOR_DEVICE_ATTR(fan2_smart_tach, S_IWUSR | S_IRUGO, + show_fan_smart_tach, store_fan_smart_tach, 1); +static SENSOR_DEVICE_ATTR(fan3_smart_tach, S_IWUSR | S_IRUGO, + show_fan_smart_tach, store_fan_smart_tach, 2); +static SENSOR_DEVICE_ATTR(fan4_smart_tach, S_IWUSR | S_IRUGO, + show_fan_smart_tach, store_fan_smart_tach, 3); + +static ssize_t show_pwm(struct device *dev, struct device_attribute *attr, + char *buf) +{ + int nr = (to_sensor_dev_attr(attr))->index; + struct lm93_data *data = lm93_update_device(dev); + u8 ctl2, ctl4; + long rc; + + ctl2 = data->block9[nr][LM93_PWM_CTL2]; + ctl4 = data->block9[nr][LM93_PWM_CTL4]; + if (ctl2 & 0x01) /* show user commanded value if enabled */ + rc = data->pwm_override[nr]; + else /* show present h/w value if manual pwm disabled */ + rc = LM93_PWM_FROM_REG(ctl2 >> 4, (ctl4 & 0x07) ? + LM93_PWM_MAP_LO_FREQ : LM93_PWM_MAP_HI_FREQ); + return sprintf(buf,"%ld\n",rc); +} + +static ssize_t store_pwm(struct device *dev, struct device_attribute *attr, + const char *buf, size_t count) +{ + int nr = (to_sensor_dev_attr(attr))->index; + struct i2c_client *client = to_i2c_client(dev); + struct lm93_data *data = i2c_get_clientdata(client); + u32 val = simple_strtoul(buf, NULL, 10); + u8 ctl2, ctl4; + + mutex_lock(&data->update_lock); + ctl2 = lm93_read_byte(client,LM93_REG_PWM_CTL(nr,LM93_PWM_CTL2)); + ctl4 = lm93_read_byte(client, LM93_REG_PWM_CTL(nr,LM93_PWM_CTL4)); + ctl2 = (ctl2 & 0x0f) | LM93_PWM_TO_REG(val,(ctl4 & 0x07) ? + LM93_PWM_MAP_LO_FREQ : LM93_PWM_MAP_HI_FREQ) << 4; + /* save user commanded value */ + data->pwm_override[nr] = LM93_PWM_FROM_REG(ctl2 >> 4, + (ctl4 & 0x07) ? LM93_PWM_MAP_LO_FREQ : + LM93_PWM_MAP_HI_FREQ); + lm93_write_byte(client,LM93_REG_PWM_CTL(nr,LM93_PWM_CTL2),ctl2); + mutex_unlock(&data->update_lock); + return count; +} + +static SENSOR_DEVICE_ATTR(pwm1, S_IWUSR | S_IRUGO, show_pwm, store_pwm, 0); +static SENSOR_DEVICE_ATTR(pwm2, S_IWUSR | S_IRUGO, show_pwm, store_pwm, 1); + +static ssize_t show_pwm_enable(struct device *dev, + struct device_attribute *attr, char *buf) +{ + int nr = (to_sensor_dev_attr(attr))->index; + struct lm93_data *data = lm93_update_device(dev); + u8 ctl2; + long rc; + + ctl2 = data->block9[nr][LM93_PWM_CTL2]; + if (ctl2 & 0x01) /* manual override enabled ? */ + rc = ((ctl2 & 0xF0) == 0xF0) ? 0 : 1; + else + rc = 2; + return sprintf(buf,"%ld\n",rc); +} + +static ssize_t store_pwm_enable(struct device *dev, + struct device_attribute *attr, + const char *buf, size_t count) +{ + int nr = (to_sensor_dev_attr(attr))->index; + struct i2c_client *client = to_i2c_client(dev); + struct lm93_data *data = i2c_get_clientdata(client); + u32 val = simple_strtoul(buf, NULL, 10); + u8 ctl2; + + mutex_lock(&data->update_lock); + ctl2 = lm93_read_byte(client,LM93_REG_PWM_CTL(nr,LM93_PWM_CTL2)); + + switch (val) { + case 0: + ctl2 |= 0xF1; /* enable manual override, set PWM to max */ + break; + case 1: ctl2 |= 0x01; /* enable manual override */ + break; + case 2: ctl2 &= ~0x01; /* disable manual override */ + break; + default: + mutex_unlock(&data->update_lock); + return -EINVAL; + } + + lm93_write_byte(client,LM93_REG_PWM_CTL(nr,LM93_PWM_CTL2),ctl2); + mutex_unlock(&data->update_lock); + return count; +} + +static SENSOR_DEVICE_ATTR(pwm1_enable, S_IWUSR | S_IRUGO, + show_pwm_enable, store_pwm_enable, 0); +static SENSOR_DEVICE_ATTR(pwm2_enable, S_IWUSR | S_IRUGO, + show_pwm_enable, store_pwm_enable, 1); + +static ssize_t show_pwm_freq(struct device *dev, struct device_attribute *attr, + char *buf) +{ + int nr = (to_sensor_dev_attr(attr))->index; + struct lm93_data *data = lm93_update_device(dev); + u8 ctl4; + + ctl4 = data->block9[nr][LM93_PWM_CTL4]; + return sprintf(buf,"%d\n",LM93_PWM_FREQ_FROM_REG(ctl4)); +} + +/* helper function - must grab data->update_lock before calling + pwm is 0-1, indicating pwm1-pwm2 + this disables smart tach for all tach channels bound to the given pwm */ +static void lm93_disable_fan_smart_tach(struct i2c_client *client, + struct lm93_data *data, int pwm) +{ + int mapping = lm93_read_byte(client, LM93_REG_SF_TACH_TO_PWM); + int mask; + + /* collapse the mapping into a mask of enable bits */ + mapping = (mapping >> pwm) & 0x55; + mask = mapping & 0x01; + mask |= (mapping & 0x04) >> 1; + mask |= (mapping & 0x10) >> 2; + mask |= (mapping & 0x40) >> 3; + + /* disable smart tach according to the mask */ + data->sfc2 = lm93_read_byte(client, LM93_REG_SFC2); + data->sfc2 &= ~mask; + lm93_write_byte(client, LM93_REG_SFC2, data->sfc2); +} + +static ssize_t store_pwm_freq(struct device *dev, + struct device_attribute *attr, + const char *buf, size_t count) +{ + int nr = (to_sensor_dev_attr(attr))->index; + struct i2c_client *client = to_i2c_client(dev); + struct lm93_data *data = i2c_get_clientdata(client); + u32 val = simple_strtoul(buf, NULL, 10); + u8 ctl4; + + mutex_lock(&data->update_lock); + ctl4 = lm93_read_byte(client,LM93_REG_PWM_CTL(nr,LM93_PWM_CTL4)); + ctl4 = (ctl4 & 0xf8) | LM93_PWM_FREQ_TO_REG(val); + data->block9[nr][LM93_PWM_CTL4] = ctl4; + /* ctl4 == 0 -> 22.5KHz -> disable smart tach */ + if (!ctl4) + lm93_disable_fan_smart_tach(client, data, nr); + lm93_write_byte(client, LM93_REG_PWM_CTL(nr,LM93_PWM_CTL4), ctl4); + mutex_unlock(&data->update_lock); + return count; +} + +static SENSOR_DEVICE_ATTR(pwm1_freq, S_IWUSR | S_IRUGO, + show_pwm_freq, store_pwm_freq, 0); +static SENSOR_DEVICE_ATTR(pwm2_freq, S_IWUSR | S_IRUGO, + show_pwm_freq, store_pwm_freq, 1); + +static ssize_t show_pwm_auto_channels(struct device *dev, + struct device_attribute *attr, char *buf) +{ + int nr = (to_sensor_dev_attr(attr))->index; + struct lm93_data *data = lm93_update_device(dev); + return sprintf(buf,"%d\n",data->block9[nr][LM93_PWM_CTL1]); +} + +static ssize_t store_pwm_auto_channels(struct device *dev, + struct device_attribute *attr, + const char *buf, size_t count) +{ + int nr = (to_sensor_dev_attr(attr))->index; + struct i2c_client *client = to_i2c_client(dev); + struct lm93_data *data = i2c_get_clientdata(client); + u32 val = simple_strtoul(buf, NULL, 10); + + mutex_lock(&data->update_lock); + data->block9[nr][LM93_PWM_CTL1] = SENSORS_LIMIT(val, 0, 255); + lm93_write_byte(client, LM93_REG_PWM_CTL(nr,LM93_PWM_CTL1), + data->block9[nr][LM93_PWM_CTL1]); + mutex_unlock(&data->update_lock); + return count; +} + +static SENSOR_DEVICE_ATTR(pwm1_auto_channels, S_IWUSR | S_IRUGO, + show_pwm_auto_channels, store_pwm_auto_channels, 0); +static SENSOR_DEVICE_ATTR(pwm2_auto_channels, S_IWUSR | S_IRUGO, + show_pwm_auto_channels, store_pwm_auto_channels, 1); + +static ssize_t show_pwm_auto_spinup_min(struct device *dev, + struct device_attribute *attr,char *buf) +{ + int nr = (to_sensor_dev_attr(attr))->index; + struct lm93_data *data = lm93_update_device(dev); + u8 ctl3, ctl4; + + ctl3 = data->block9[nr][LM93_PWM_CTL3]; + ctl4 = data->block9[nr][LM93_PWM_CTL4]; + return sprintf(buf,"%d\n", + LM93_PWM_FROM_REG(ctl3 & 0x0f, (ctl4 & 0x07) ? + LM93_PWM_MAP_LO_FREQ : LM93_PWM_MAP_HI_FREQ)); +} + +static ssize_t store_pwm_auto_spinup_min(struct device *dev, + struct device_attribute *attr, + const char *buf, size_t count) +{ + int nr = (to_sensor_dev_attr(attr))->index; + struct i2c_client *client = to_i2c_client(dev); + struct lm93_data *data = i2c_get_clientdata(client); + u32 val = simple_strtoul(buf, NULL, 10); + u8 ctl3, ctl4; + + mutex_lock(&data->update_lock); + ctl3 = lm93_read_byte(client,LM93_REG_PWM_CTL(nr, LM93_PWM_CTL3)); + ctl4 = lm93_read_byte(client,LM93_REG_PWM_CTL(nr, LM93_PWM_CTL4)); + ctl3 = (ctl3 & 0xf0) | LM93_PWM_TO_REG(val, (ctl4 & 0x07) ? + LM93_PWM_MAP_LO_FREQ : + LM93_PWM_MAP_HI_FREQ); + data->block9[nr][LM93_PWM_CTL3] = ctl3; + lm93_write_byte(client,LM93_REG_PWM_CTL(nr, LM93_PWM_CTL3), ctl3); + mutex_unlock(&data->update_lock); + return count; +} + +static SENSOR_DEVICE_ATTR(pwm1_auto_spinup_min, S_IWUSR | S_IRUGO, + show_pwm_auto_spinup_min, + store_pwm_auto_spinup_min, 0); +static SENSOR_DEVICE_ATTR(pwm2_auto_spinup_min, S_IWUSR | S_IRUGO, + show_pwm_auto_spinup_min, + store_pwm_auto_spinup_min, 1); + +static ssize_t show_pwm_auto_spinup_time(struct device *dev, + struct device_attribute *attr, char *buf) +{ + int nr = (to_sensor_dev_attr(attr))->index; + struct lm93_data *data = lm93_update_device(dev); + return sprintf(buf,"%d\n",LM93_SPINUP_TIME_FROM_REG( + data->block9[nr][LM93_PWM_CTL3])); +} + +static ssize_t store_pwm_auto_spinup_time(struct device *dev, + struct device_attribute *attr, + const char *buf, size_t count) +{ + int nr = (to_sensor_dev_attr(attr))->index; + struct i2c_client *client = to_i2c_client(dev); + struct lm93_data *data = i2c_get_clientdata(client); + u32 val = simple_strtoul(buf, NULL, 10); + u8 ctl3; + + mutex_lock(&data->update_lock); + ctl3 = lm93_read_byte(client,LM93_REG_PWM_CTL(nr, LM93_PWM_CTL3)); + ctl3 = (ctl3 & 0x1f) | (LM93_SPINUP_TIME_TO_REG(val) << 5 & 0xe0); + data->block9[nr][LM93_PWM_CTL3] = ctl3; + lm93_write_byte(client,LM93_REG_PWM_CTL(nr, LM93_PWM_CTL3), ctl3); + mutex_unlock(&data->update_lock); + return count; +} + +static SENSOR_DEVICE_ATTR(pwm1_auto_spinup_time, S_IWUSR | S_IRUGO, + show_pwm_auto_spinup_time, + store_pwm_auto_spinup_time, 0); +static SENSOR_DEVICE_ATTR(pwm2_auto_spinup_time, S_IWUSR | S_IRUGO, + show_pwm_auto_spinup_time, + store_pwm_auto_spinup_time, 1); + +static ssize_t show_pwm_auto_prochot_ramp(struct device *dev, + struct device_attribute *attr, char *buf) +{ + struct lm93_data *data = lm93_update_device(dev); + return sprintf(buf,"%d\n", + LM93_RAMP_FROM_REG(data->pwm_ramp_ctl >> 4 & 0x0f)); +} + +static ssize_t store_pwm_auto_prochot_ramp(struct device *dev, + struct device_attribute *attr, + const char *buf, size_t count) +{ + struct i2c_client *client = to_i2c_client(dev); + struct lm93_data *data = i2c_get_clientdata(client); + u32 val = simple_strtoul(buf, NULL, 10); + u8 ramp; + + mutex_lock(&data->update_lock); + ramp = lm93_read_byte(client, LM93_REG_PWM_RAMP_CTL); + ramp = (ramp & 0x0f) | (LM93_RAMP_TO_REG(val) << 4 & 0xf0); + lm93_write_byte(client, LM93_REG_PWM_RAMP_CTL, ramp); + mutex_unlock(&data->update_lock); + return count; +} + +static DEVICE_ATTR(pwm_auto_prochot_ramp, S_IRUGO | S_IWUSR, + show_pwm_auto_prochot_ramp, + store_pwm_auto_prochot_ramp); + +static ssize_t show_pwm_auto_vrdhot_ramp(struct device *dev, + struct device_attribute *attr, char *buf) +{ + struct lm93_data *data = lm93_update_device(dev); + return sprintf(buf,"%d\n", + LM93_RAMP_FROM_REG(data->pwm_ramp_ctl & 0x0f)); +} + +static ssize_t store_pwm_auto_vrdhot_ramp(struct device *dev, + struct device_attribute *attr, + const char *buf, size_t count) +{ + struct i2c_client *client = to_i2c_client(dev); + struct lm93_data *data = i2c_get_clientdata(client); + u32 val = simple_strtoul(buf, NULL, 10); + u8 ramp; + + mutex_lock(&data->update_lock); + ramp = lm93_read_byte(client, LM93_REG_PWM_RAMP_CTL); + ramp = (ramp & 0xf0) | (LM93_RAMP_TO_REG(val) & 0x0f); + lm93_write_byte(client, LM93_REG_PWM_RAMP_CTL, ramp); + mutex_unlock(&data->update_lock); + return 0; +} + +static DEVICE_ATTR(pwm_auto_vrdhot_ramp, S_IRUGO | S_IWUSR, + show_pwm_auto_vrdhot_ramp, + store_pwm_auto_vrdhot_ramp); + +static ssize_t show_vid(struct device *dev, struct device_attribute *attr, + char *buf) +{ + int nr = (to_sensor_dev_attr(attr))->index; + struct lm93_data *data = lm93_update_device(dev); + return sprintf(buf,"%d\n",LM93_VID_FROM_REG(data->vid[nr])); +} + +static SENSOR_DEVICE_ATTR(vid1, S_IRUGO, show_vid, NULL, 0); +static SENSOR_DEVICE_ATTR(vid2, S_IRUGO, show_vid, NULL, 1); + +static ssize_t show_prochot(struct device *dev, struct device_attribute *attr, + char *buf) +{ + int nr = (to_sensor_dev_attr(attr))->index; + struct lm93_data *data = lm93_update_device(dev); + return sprintf(buf,"%d\n",data->block4[nr].cur); +} + +static SENSOR_DEVICE_ATTR(prochot1, S_IRUGO, show_prochot, NULL, 0); +static SENSOR_DEVICE_ATTR(prochot2, S_IRUGO, show_prochot, NULL, 1); + +static ssize_t show_prochot_avg(struct device *dev, + struct device_attribute *attr, char *buf) +{ + int nr = (to_sensor_dev_attr(attr))->index; + struct lm93_data *data = lm93_update_device(dev); + return sprintf(buf,"%d\n",data->block4[nr].avg); +} + +static SENSOR_DEVICE_ATTR(prochot1_avg, S_IRUGO, show_prochot_avg, NULL, 0); +static SENSOR_DEVICE_ATTR(prochot2_avg, S_IRUGO, show_prochot_avg, NULL, 1); + +static ssize_t show_prochot_max(struct device *dev, + struct device_attribute *attr, char *buf) +{ + int nr = (to_sensor_dev_attr(attr))->index; + struct lm93_data *data = lm93_update_device(dev); + return sprintf(buf,"%d\n",data->prochot_max[nr]); +} + +static ssize_t store_prochot_max(struct device *dev, + struct device_attribute *attr, + const char *buf, size_t count) +{ + int nr = (to_sensor_dev_attr(attr))->index; + struct i2c_client *client = to_i2c_client(dev); + struct lm93_data *data = i2c_get_clientdata(client); + u32 val = simple_strtoul(buf, NULL, 10); + + mutex_lock(&data->update_lock); + data->prochot_max[nr] = LM93_PROCHOT_TO_REG(val); + lm93_write_byte(client, LM93_REG_PROCHOT_MAX(nr), + data->prochot_max[nr]); + mutex_unlock(&data->update_lock); + return count; +} + +static SENSOR_DEVICE_ATTR(prochot1_max, S_IWUSR | S_IRUGO, + show_prochot_max, store_prochot_max, 0); +static SENSOR_DEVICE_ATTR(prochot2_max, S_IWUSR | S_IRUGO, + show_prochot_max, store_prochot_max, 1); + +static const u8 prochot_override_mask[] = { 0x80, 0x40 }; + +static ssize_t show_prochot_override(struct device *dev, + struct device_attribute *attr, char *buf) +{ + int nr = (to_sensor_dev_attr(attr))->index; + struct lm93_data *data = lm93_update_device(dev); + return sprintf(buf,"%d\n", + (data->prochot_override & prochot_override_mask[nr]) ? 1 : 0); +} + +static ssize_t store_prochot_override(struct device *dev, + struct device_attribute *attr, + const char *buf, size_t count) +{ + int nr = (to_sensor_dev_attr(attr))->index; + struct i2c_client *client = to_i2c_client(dev); + struct lm93_data *data = i2c_get_clientdata(client); + u32 val = simple_strtoul(buf, NULL, 10); + + mutex_lock(&data->update_lock); + if (val) + data->prochot_override |= prochot_override_mask[nr]; + else + data->prochot_override &= (~prochot_override_mask[nr]); + lm93_write_byte(client, LM93_REG_PROCHOT_OVERRIDE, + data->prochot_override); + mutex_unlock(&data->update_lock); + return count; +} + +static SENSOR_DEVICE_ATTR(prochot1_override, S_IWUSR | S_IRUGO, + show_prochot_override, store_prochot_override, 0); +static SENSOR_DEVICE_ATTR(prochot2_override, S_IWUSR | S_IRUGO, + show_prochot_override, store_prochot_override, 1); + +static ssize_t show_prochot_interval(struct device *dev, + struct device_attribute *attr, char *buf) +{ + int nr = (to_sensor_dev_attr(attr))->index; + struct lm93_data *data = lm93_update_device(dev); + u8 tmp; + if (nr==1) + tmp = (data->prochot_interval & 0xf0) >> 4; + else + tmp = data->prochot_interval & 0x0f; + return sprintf(buf,"%d\n",LM93_INTERVAL_FROM_REG(tmp)); +} + +static ssize_t store_prochot_interval(struct device *dev, + struct device_attribute *attr, + const char *buf, size_t count) +{ + int nr = (to_sensor_dev_attr(attr))->index; + struct i2c_client *client = to_i2c_client(dev); + struct lm93_data *data = i2c_get_clientdata(client); + u32 val = simple_strtoul(buf, NULL, 10); + u8 tmp; + + mutex_lock(&data->update_lock); + tmp = lm93_read_byte(client, LM93_REG_PROCHOT_INTERVAL); + if (nr==1) + tmp = (tmp & 0x0f) | (LM93_INTERVAL_TO_REG(val) << 4); + else + tmp = (tmp & 0xf0) | LM93_INTERVAL_TO_REG(val); + data->prochot_interval = tmp; + lm93_write_byte(client, LM93_REG_PROCHOT_INTERVAL, tmp); + mutex_unlock(&data->update_lock); + return count; +} + +static SENSOR_DEVICE_ATTR(prochot1_interval, S_IWUSR | S_IRUGO, + show_prochot_interval, store_prochot_interval, 0); +static SENSOR_DEVICE_ATTR(prochot2_interval, S_IWUSR | S_IRUGO, + show_prochot_interval, store_prochot_interval, 1); + +static ssize_t show_prochot_override_duty_cycle(struct device *dev, + struct device_attribute *attr, + char *buf) +{ + struct lm93_data *data = lm93_update_device(dev); + return sprintf(buf,"%d\n",data->prochot_override & 0x0f); +} + +static ssize_t store_prochot_override_duty_cycle(struct device *dev, + struct device_attribute *attr, + const char *buf, size_t count) +{ + struct i2c_client *client = to_i2c_client(dev); + struct lm93_data *data = i2c_get_clientdata(client); + u32 val = simple_strtoul(buf, NULL, 10); + + mutex_lock(&data->update_lock); + data->prochot_override = (data->prochot_override & 0xf0) | + SENSORS_LIMIT(val, 0, 15); + lm93_write_byte(client, LM93_REG_PROCHOT_OVERRIDE, + data->prochot_override); + mutex_unlock(&data->update_lock); + return count; +} + +static DEVICE_ATTR(prochot_override_duty_cycle, S_IRUGO | S_IWUSR, + show_prochot_override_duty_cycle, + store_prochot_override_duty_cycle); + +static ssize_t show_prochot_short(struct device *dev, + struct device_attribute *attr, char *buf) +{ + struct lm93_data *data = lm93_update_device(dev); + return sprintf(buf,"%d\n",(data->config & 0x10) ? 1 : 0); +} + +static ssize_t store_prochot_short(struct device *dev, + struct device_attribute *attr, + const char *buf, size_t count) +{ + struct i2c_client *client = to_i2c_client(dev); + struct lm93_data *data = i2c_get_clientdata(client); + u32 val = simple_strtoul(buf, NULL, 10); + + mutex_lock(&data->update_lock); + if (val) + data->config |= 0x10; + else + data->config &= ~0x10; + lm93_write_byte(client, LM93_REG_CONFIG, data->config); + mutex_unlock(&data->update_lock); + return count; +} + +static DEVICE_ATTR(prochot_short, S_IRUGO | S_IWUSR, + show_prochot_short, store_prochot_short); + +static ssize_t show_vrdhot(struct device *dev, struct device_attribute *attr, + char *buf) +{ + int nr = (to_sensor_dev_attr(attr))->index; + struct lm93_data *data = lm93_update_device(dev); + return sprintf(buf,"%d\n", + data->block1.host_status_1 & (1 << (nr+4)) ? 1 : 0); +} + +static SENSOR_DEVICE_ATTR(vrdhot1, S_IRUGO, show_vrdhot, NULL, 0); +static SENSOR_DEVICE_ATTR(vrdhot2, S_IRUGO, show_vrdhot, NULL, 1); + +static ssize_t show_gpio(struct device *dev, struct device_attribute *attr, + char *buf) +{ + struct lm93_data *data = lm93_update_device(dev); + return sprintf(buf,"%d\n",LM93_GPI_FROM_REG(data->gpi)); +} + +static DEVICE_ATTR(gpio, S_IRUGO, show_gpio, NULL); + +static ssize_t show_alarms(struct device *dev, struct device_attribute *attr, + char *buf) +{ + struct lm93_data *data = lm93_update_device(dev); + return sprintf(buf,"%d\n",LM93_ALARMS_FROM_REG(data->block1)); +} + +static DEVICE_ATTR(alarms, S_IRUGO, show_alarms, NULL); + +static struct attribute *lm93_attrs[] = { + &sensor_dev_attr_in1_input.dev_attr.attr, + &sensor_dev_attr_in2_input.dev_attr.attr, + &sensor_dev_attr_in3_input.dev_attr.attr, + &sensor_dev_attr_in4_input.dev_attr.attr, + &sensor_dev_attr_in5_input.dev_attr.attr, + &sensor_dev_attr_in6_input.dev_attr.attr, + &sensor_dev_attr_in7_input.dev_attr.attr, + &sensor_dev_attr_in8_input.dev_attr.attr, + &sensor_dev_attr_in9_input.dev_attr.attr, + &sensor_dev_attr_in10_input.dev_attr.attr, + &sensor_dev_attr_in11_input.dev_attr.attr, + &sensor_dev_attr_in12_input.dev_attr.attr, + &sensor_dev_attr_in13_input.dev_attr.attr, + &sensor_dev_attr_in14_input.dev_attr.attr, + &sensor_dev_attr_in15_input.dev_attr.attr, + &sensor_dev_attr_in16_input.dev_attr.attr, + &sensor_dev_attr_in1_min.dev_attr.attr, + &sensor_dev_attr_in2_min.dev_attr.attr, + &sensor_dev_attr_in3_min.dev_attr.attr, + &sensor_dev_attr_in4_min.dev_attr.attr, + &sensor_dev_attr_in5_min.dev_attr.attr, + &sensor_dev_attr_in6_min.dev_attr.attr, + &sensor_dev_attr_in7_min.dev_attr.attr, + &sensor_dev_attr_in8_min.dev_attr.attr, + &sensor_dev_attr_in9_min.dev_attr.attr, + &sensor_dev_attr_in10_min.dev_attr.attr, + &sensor_dev_attr_in11_min.dev_attr.attr, + &sensor_dev_attr_in12_min.dev_attr.attr, + &sensor_dev_attr_in13_min.dev_attr.attr, + &sensor_dev_attr_in14_min.dev_attr.attr, + &sensor_dev_attr_in15_min.dev_attr.attr, + &sensor_dev_attr_in16_min.dev_attr.attr, + &sensor_dev_attr_in1_max.dev_attr.attr, + &sensor_dev_attr_in2_max.dev_attr.attr, + &sensor_dev_attr_in3_max.dev_attr.attr, + &sensor_dev_attr_in4_max.dev_attr.attr, + &sensor_dev_attr_in5_max.dev_attr.attr, + &sensor_dev_attr_in6_max.dev_attr.attr, + &sensor_dev_attr_in7_max.dev_attr.attr, + &sensor_dev_attr_in8_max.dev_attr.attr, + &sensor_dev_attr_in9_max.dev_attr.attr, + &sensor_dev_attr_in10_max.dev_attr.attr, + &sensor_dev_attr_in11_max.dev_attr.attr, + &sensor_dev_attr_in12_max.dev_attr.attr, + &sensor_dev_attr_in13_max.dev_attr.attr, + &sensor_dev_attr_in14_max.dev_attr.attr, + &sensor_dev_attr_in15_max.dev_attr.attr, + &sensor_dev_attr_in16_max.dev_attr.attr, + &sensor_dev_attr_temp1_input.dev_attr.attr, + &sensor_dev_attr_temp2_input.dev_attr.attr, + &sensor_dev_attr_temp3_input.dev_attr.attr, + &sensor_dev_attr_temp1_min.dev_attr.attr, + &sensor_dev_attr_temp2_min.dev_attr.attr, + &sensor_dev_attr_temp3_min.dev_attr.attr, + &sensor_dev_attr_temp1_max.dev_attr.attr, + &sensor_dev_attr_temp2_max.dev_attr.attr, + &sensor_dev_attr_temp3_max.dev_attr.attr, + &sensor_dev_attr_temp1_auto_base.dev_attr.attr, + &sensor_dev_attr_temp2_auto_base.dev_attr.attr, + &sensor_dev_attr_temp3_auto_base.dev_attr.attr, + &sensor_dev_attr_temp1_auto_boost.dev_attr.attr, + &sensor_dev_attr_temp2_auto_boost.dev_attr.attr, + &sensor_dev_attr_temp3_auto_boost.dev_attr.attr, + &sensor_dev_attr_temp1_auto_boost_hyst.dev_attr.attr, + &sensor_dev_attr_temp2_auto_boost_hyst.dev_attr.attr, + &sensor_dev_attr_temp3_auto_boost_hyst.dev_attr.attr, + &sensor_dev_attr_temp1_auto_offset1.dev_attr.attr, + &sensor_dev_attr_temp1_auto_offset2.dev_attr.attr, + &sensor_dev_attr_temp1_auto_offset3.dev_attr.attr, + &sensor_dev_attr_temp1_auto_offset4.dev_attr.attr, + &sensor_dev_attr_temp1_auto_offset5.dev_attr.attr, + &sensor_dev_attr_temp1_auto_offset6.dev_attr.attr, + &sensor_dev_attr_temp1_auto_offset7.dev_attr.attr, + &sensor_dev_attr_temp1_auto_offset8.dev_attr.attr, + &sensor_dev_attr_temp1_auto_offset9.dev_attr.attr, + &sensor_dev_attr_temp1_auto_offset10.dev_attr.attr, + &sensor_dev_attr_temp1_auto_offset11.dev_attr.attr, + &sensor_dev_attr_temp1_auto_offset12.dev_attr.attr, + &sensor_dev_attr_temp2_auto_offset1.dev_attr.attr, + &sensor_dev_attr_temp2_auto_offset2.dev_attr.attr, + &sensor_dev_attr_temp2_auto_offset3.dev_attr.attr, + &sensor_dev_attr_temp2_auto_offset4.dev_attr.attr, + &sensor_dev_attr_temp2_auto_offset5.dev_attr.attr, + &sensor_dev_attr_temp2_auto_offset6.dev_attr.attr, + &sensor_dev_attr_temp2_auto_offset7.dev_attr.attr, + &sensor_dev_attr_temp2_auto_offset8.dev_attr.attr, + &sensor_dev_attr_temp2_auto_offset9.dev_attr.attr, + &sensor_dev_attr_temp2_auto_offset10.dev_attr.attr, + &sensor_dev_attr_temp2_auto_offset11.dev_attr.attr, + &sensor_dev_attr_temp2_auto_offset12.dev_attr.attr, + &sensor_dev_attr_temp3_auto_offset1.dev_attr.attr, + &sensor_dev_attr_temp3_auto_offset2.dev_attr.attr, + &sensor_dev_attr_temp3_auto_offset3.dev_attr.attr, + &sensor_dev_attr_temp3_auto_offset4.dev_attr.attr, + &sensor_dev_attr_temp3_auto_offset5.dev_attr.attr, + &sensor_dev_attr_temp3_auto_offset6.dev_attr.attr, + &sensor_dev_attr_temp3_auto_offset7.dev_attr.attr, + &sensor_dev_attr_temp3_auto_offset8.dev_attr.attr, + &sensor_dev_attr_temp3_auto_offset9.dev_attr.attr, + &sensor_dev_attr_temp3_auto_offset10.dev_attr.attr, + &sensor_dev_attr_temp3_auto_offset11.dev_attr.attr, + &sensor_dev_attr_temp3_auto_offset12.dev_attr.attr, + &sensor_dev_attr_temp1_auto_pwm_min.dev_attr.attr, + &sensor_dev_attr_temp2_auto_pwm_min.dev_attr.attr, + &sensor_dev_attr_temp3_auto_pwm_min.dev_attr.attr, + &sensor_dev_attr_temp1_auto_offset_hyst.dev_attr.attr, + &sensor_dev_attr_temp2_auto_offset_hyst.dev_attr.attr, + &sensor_dev_attr_temp3_auto_offset_hyst.dev_attr.attr, + &sensor_dev_attr_fan1_input.dev_attr.attr, + &sensor_dev_attr_fan2_input.dev_attr.attr, + &sensor_dev_attr_fan3_input.dev_attr.attr, + &sensor_dev_attr_fan4_input.dev_attr.attr, + &sensor_dev_attr_fan1_min.dev_attr.attr, + &sensor_dev_attr_fan2_min.dev_attr.attr, + &sensor_dev_attr_fan3_min.dev_attr.attr, + &sensor_dev_attr_fan4_min.dev_attr.attr, + &sensor_dev_attr_fan1_smart_tach.dev_attr.attr, + &sensor_dev_attr_fan2_smart_tach.dev_attr.attr, + &sensor_dev_attr_fan3_smart_tach.dev_attr.attr, + &sensor_dev_attr_fan4_smart_tach.dev_attr.attr, + &sensor_dev_attr_pwm1.dev_attr.attr, + &sensor_dev_attr_pwm2.dev_attr.attr, + &sensor_dev_attr_pwm1_enable.dev_attr.attr, + &sensor_dev_attr_pwm2_enable.dev_attr.attr, + &sensor_dev_attr_pwm1_freq.dev_attr.attr, + &sensor_dev_attr_pwm2_freq.dev_attr.attr, + &sensor_dev_attr_pwm1_auto_channels.dev_attr.attr, + &sensor_dev_attr_pwm2_auto_channels.dev_attr.attr, + &sensor_dev_attr_pwm1_auto_spinup_min.dev_attr.attr, + &sensor_dev_attr_pwm2_auto_spinup_min.dev_attr.attr, + &sensor_dev_attr_pwm1_auto_spinup_time.dev_attr.attr, + &sensor_dev_attr_pwm2_auto_spinup_time.dev_attr.attr, + &dev_attr_pwm_auto_prochot_ramp.attr, + &dev_attr_pwm_auto_vrdhot_ramp.attr, + &sensor_dev_attr_vid1.dev_attr.attr, + &sensor_dev_attr_vid2.dev_attr.attr, + &sensor_dev_attr_prochot1.dev_attr.attr, + &sensor_dev_attr_prochot2.dev_attr.attr, + &sensor_dev_attr_prochot1_avg.dev_attr.attr, + &sensor_dev_attr_prochot2_avg.dev_attr.attr, + &sensor_dev_attr_prochot1_max.dev_attr.attr, + &sensor_dev_attr_prochot2_max.dev_attr.attr, + &sensor_dev_attr_prochot1_override.dev_attr.attr, + &sensor_dev_attr_prochot2_override.dev_attr.attr, + &sensor_dev_attr_prochot1_interval.dev_attr.attr, + &sensor_dev_attr_prochot2_interval.dev_attr.attr, + &dev_attr_prochot_override_duty_cycle.attr, + &dev_attr_prochot_short.attr, + &sensor_dev_attr_vrdhot1.dev_attr.attr, + &sensor_dev_attr_vrdhot2.dev_attr.attr, + &dev_attr_gpio.attr, + &dev_attr_alarms.attr, + NULL +}; + +static struct attribute_group lm93_attr_grp = { + .attrs = lm93_attrs, +}; + +static void lm93_init_client(struct i2c_client *client) +{ + int i; + u8 reg; + + /* configure VID pin input thresholds */ + reg = lm93_read_byte(client, LM93_REG_GPI_VID_CTL); + lm93_write_byte(client, LM93_REG_GPI_VID_CTL, + reg | (vid_agtl ? 0x03 : 0x00)); + + if (init) { + /* enable #ALERT pin */ + reg = lm93_read_byte(client, LM93_REG_CONFIG); + lm93_write_byte(client, LM93_REG_CONFIG, reg | 0x08); + + /* enable ASF mode for BMC status registers */ + reg = lm93_read_byte(client, LM93_REG_STATUS_CONTROL); + lm93_write_byte(client, LM93_REG_STATUS_CONTROL, reg | 0x02); + + /* set sleep state to S0 */ + lm93_write_byte(client, LM93_REG_SLEEP_CONTROL, 0); + + /* unmask #VRDHOT and dynamic VCCP (if nec) error events */ + reg = lm93_read_byte(client, LM93_REG_MISC_ERR_MASK); + reg &= ~0x03; + reg &= ~(vccp_limit_type[0] ? 0x10 : 0); + reg &= ~(vccp_limit_type[1] ? 0x20 : 0); + lm93_write_byte(client, LM93_REG_MISC_ERR_MASK, reg); + } + + /* start monitoring */ + reg = lm93_read_byte(client, LM93_REG_CONFIG); + lm93_write_byte(client, LM93_REG_CONFIG, reg | 0x01); + + /* spin until ready */ + for (i=0; i<20; i++) { + msleep(10); + if ((lm93_read_byte(client, LM93_REG_CONFIG) & 0x80) == 0x80) + return; + } + + dev_warn(&client->dev,"timed out waiting for sensor " + "chip to signal ready!\n"); +} + +static int lm93_detect(struct i2c_adapter *adapter, int address, int kind) +{ + struct lm93_data *data; + struct i2c_client *client; + + int err = -ENODEV, func; + void (*update)(struct lm93_data *, struct i2c_client *); + + /* choose update routine based on bus capabilities */ + func = i2c_get_functionality(adapter); + if ( ((LM93_SMBUS_FUNC_FULL & func) == LM93_SMBUS_FUNC_FULL) && + (!disable_block) ) { + dev_dbg(&adapter->dev,"using SMBus block data transactions\n"); + update = lm93_update_client_full; + } else if ((LM93_SMBUS_FUNC_MIN & func) == LM93_SMBUS_FUNC_MIN) { + dev_dbg(&adapter->dev,"disabled SMBus block data " + "transactions\n"); + update = lm93_update_client_min; + } else { + dev_dbg(&adapter->dev,"detect failed, " + "smbus byte and/or word data not supported!\n"); + goto err_out; + } + + /* OK. For now, we presume we have a valid client. We now create the + client structure, even though we cannot fill it completely yet. + But it allows us to access lm78_{read,write}_value. */ + + if ( !(data = kzalloc(sizeof(struct lm93_data), GFP_KERNEL))) { + dev_dbg(&adapter->dev,"out of memory!\n"); + err = -ENOMEM; + goto err_out; + } + + client = &data->client; + i2c_set_clientdata(client, data); + client->addr = address; + client->adapter = adapter; + client->driver = &lm93_driver; + + /* detection */ + if (kind < 0) { + int mfr = lm93_read_byte(client, LM93_REG_MFR_ID); + + if (mfr != 0x01) { + dev_dbg(&adapter->dev,"detect failed, " + "bad manufacturer id 0x%02x!\n", mfr); + goto err_free; + } + } + + if (kind <= 0) { + int ver = lm93_read_byte(client, LM93_REG_VER); + + if ((ver == LM93_MFR_ID) || (ver == LM93_MFR_ID_PROTOTYPE)) { + kind = lm93; + } else { + dev_dbg(&adapter->dev,"detect failed, " + "bad version id 0x%02x!\n", ver); + if (kind == 0) + dev_dbg(&adapter->dev, + "(ignored 'force' parameter)\n"); + goto err_free; + } + } + + /* fill in remaining client fields */ + strlcpy(client->name, "lm93", I2C_NAME_SIZE); + dev_dbg(&adapter->dev,"loading %s at %d,0x%02x\n", + client->name, i2c_adapter_id(client->adapter), + client->addr); + + /* housekeeping */ + data->valid = 0; + data->update = update; + mutex_init(&data->update_lock); + + /* tell the I2C layer a new client has arrived */ + if ((err = i2c_attach_client(client))) + goto err_free; + + /* initialize the chip */ + lm93_init_client(client); + + err = sysfs_create_group(&client->dev.kobj, &lm93_attr_grp); + if (err) + goto err_detach; + + /* Register hwmon driver class */ + data->class_dev = hwmon_device_register(&client->dev); + if ( !IS_ERR(data->class_dev)) + return 0; + + err = PTR_ERR(data->class_dev); + dev_err(&client->dev, "error registering hwmon device.\n"); + sysfs_remove_group(&client->dev.kobj, &lm93_attr_grp); +err_detach: + i2c_detach_client(client); +err_free: + kfree(data); +err_out: + return err; +} + +/* This function is called when: + * lm93_driver is inserted (when this module is loaded), for each + available adapter + * when a new adapter is inserted (and lm93_driver is still present) */ +static int lm93_attach_adapter(struct i2c_adapter *adapter) +{ + return i2c_probe(adapter, &addr_data, lm93_detect); +} + +static int lm93_detach_client(struct i2c_client *client) +{ + struct lm93_data *data = i2c_get_clientdata(client); + int err = 0; + + hwmon_device_unregister(data->class_dev); + sysfs_remove_group(&client->dev.kobj, &lm93_attr_grp); + + err = i2c_detach_client(client); + if (!err) + kfree(data); + return err; +} + +static struct i2c_driver lm93_driver = { + .driver = { + .name = "lm93", + }, + .attach_adapter = lm93_attach_adapter, + .detach_client = lm93_detach_client, +}; + +static int __init lm93_init(void) +{ + return i2c_add_driver(&lm93_driver); +} + +static void __exit lm93_exit(void) +{ + i2c_del_driver(&lm93_driver); +} + +MODULE_AUTHOR("Mark M. Hoffman <mhoffman@lightlink.com>, " + "Hans J. Koch <hjk@linutronix.de"); +MODULE_DESCRIPTION("LM93 driver"); +MODULE_LICENSE("GPL"); + +module_init(lm93_init); +module_exit(lm93_exit); diff --git a/drivers/hwmon/pc87360.c b/drivers/hwmon/pc87360.c index c8a21be09d8..cb72526c346 100644 --- a/drivers/hwmon/pc87360.c +++ b/drivers/hwmon/pc87360.c @@ -1,7 +1,7 @@ /* * pc87360.c - Part of lm_sensors, Linux kernel modules * for hardware monitoring - * Copyright (C) 2004 Jean Delvare <khali@linux-fr.org> + * Copyright (C) 2004, 2007 Jean Delvare <khali@linux-fr.org> * * Copied from smsc47m1.c: * Copyright (C) 2002 Mark D. Studebaker <mdsxyz123@yahoo.com> @@ -37,8 +37,7 @@ #include <linux/init.h> #include <linux/slab.h> #include <linux/jiffies.h> -#include <linux/i2c.h> -#include <linux/i2c-isa.h> +#include <linux/platform_device.h> #include <linux/hwmon.h> #include <linux/hwmon-sysfs.h> #include <linux/hwmon-vid.h> @@ -47,12 +46,10 @@ #include <asm/io.h> static u8 devid; -static unsigned short address; +static struct platform_device *pdev; static unsigned short extra_isa[3]; static u8 confreg[4]; -enum chips { any_chip, pc87360, pc87363, pc87364, pc87365, pc87366 }; - static int init = 1; module_param(init, int, 0); MODULE_PARM_DESC(init, @@ -178,11 +175,11 @@ static inline u8 PWM_TO_REG(int val, int inv) ((val) + 500) / 1000) /* - * Client data (each client gets its own) + * Device data */ struct pc87360_data { - struct i2c_client client; + const char *name; struct class_device *class_dev; struct mutex lock; struct mutex update_lock; @@ -222,27 +219,28 @@ struct pc87360_data { * Functions declaration */ -static int pc87360_detect(struct i2c_adapter *adapter); -static int pc87360_detach_client(struct i2c_client *client); +static int pc87360_probe(struct platform_device *pdev); +static int pc87360_remove(struct platform_device *pdev); static int pc87360_read_value(struct pc87360_data *data, u8 ldi, u8 bank, u8 reg); static void pc87360_write_value(struct pc87360_data *data, u8 ldi, u8 bank, u8 reg, u8 value); -static void pc87360_init_client(struct i2c_client *client, int use_thermistors); +static void pc87360_init_device(struct platform_device *pdev, + int use_thermistors); static struct pc87360_data *pc87360_update_device(struct device *dev); /* - * Driver data (common to all clients) + * Driver data */ -static struct i2c_driver pc87360_driver = { +static struct platform_driver pc87360_driver = { .driver = { .owner = THIS_MODULE, .name = "pc87360", }, - .attach_adapter = pc87360_detect, - .detach_client = pc87360_detach_client, + .probe = pc87360_probe, + .remove = __devexit_p(pc87360_remove), }; /* @@ -281,8 +279,7 @@ static ssize_t set_fan_min(struct device *dev, struct device_attribute *devattr, size_t count) { struct sensor_device_attribute *attr = to_sensor_dev_attr(devattr); - struct i2c_client *client = to_i2c_client(dev); - struct pc87360_data *data = i2c_get_clientdata(client); + struct pc87360_data *data = dev_get_drvdata(dev); long fan_min = simple_strtol(buf, NULL, 10); mutex_lock(&data->update_lock); @@ -347,8 +344,7 @@ static ssize_t set_pwm(struct device *dev, struct device_attribute *devattr, con size_t count) { struct sensor_device_attribute *attr = to_sensor_dev_attr(devattr); - struct i2c_client *client = to_i2c_client(dev); - struct pc87360_data *data = i2c_get_clientdata(client); + struct pc87360_data *data = dev_get_drvdata(dev); long val = simple_strtol(buf, NULL, 10); mutex_lock(&data->update_lock); @@ -410,8 +406,7 @@ static ssize_t set_in_min(struct device *dev, struct device_attribute *devattr, size_t count) { struct sensor_device_attribute *attr = to_sensor_dev_attr(devattr); - struct i2c_client *client = to_i2c_client(dev); - struct pc87360_data *data = i2c_get_clientdata(client); + struct pc87360_data *data = dev_get_drvdata(dev); long val = simple_strtol(buf, NULL, 10); mutex_lock(&data->update_lock); @@ -425,8 +420,7 @@ static ssize_t set_in_max(struct device *dev, struct device_attribute *devattr, size_t count) { struct sensor_device_attribute *attr = to_sensor_dev_attr(devattr); - struct i2c_client *client = to_i2c_client(dev); - struct pc87360_data *data = i2c_get_clientdata(client); + struct pc87360_data *data = dev_get_drvdata(dev); long val = simple_strtol(buf, NULL, 10); mutex_lock(&data->update_lock); @@ -511,8 +505,7 @@ static ssize_t show_vrm(struct device *dev, struct device_attribute *attr, char } static ssize_t set_vrm(struct device *dev, struct device_attribute *attr, const char *buf, size_t count) { - struct i2c_client *client = to_i2c_client(dev); - struct pc87360_data *data = i2c_get_clientdata(client); + struct pc87360_data *data = dev_get_drvdata(dev); data->vrm = simple_strtoul(buf, NULL, 10); return count; } @@ -584,8 +577,7 @@ static ssize_t set_therm_min(struct device *dev, struct device_attribute *devatt size_t count) { struct sensor_device_attribute *attr = to_sensor_dev_attr(devattr); - struct i2c_client *client = to_i2c_client(dev); - struct pc87360_data *data = i2c_get_clientdata(client); + struct pc87360_data *data = dev_get_drvdata(dev); long val = simple_strtol(buf, NULL, 10); mutex_lock(&data->update_lock); @@ -599,8 +591,7 @@ static ssize_t set_therm_max(struct device *dev, struct device_attribute *devatt size_t count) { struct sensor_device_attribute *attr = to_sensor_dev_attr(devattr); - struct i2c_client *client = to_i2c_client(dev); - struct pc87360_data *data = i2c_get_clientdata(client); + struct pc87360_data *data = dev_get_drvdata(dev); long val = simple_strtol(buf, NULL, 10); mutex_lock(&data->update_lock); @@ -614,8 +605,7 @@ static ssize_t set_therm_crit(struct device *dev, struct device_attribute *devat size_t count) { struct sensor_device_attribute *attr = to_sensor_dev_attr(devattr); - struct i2c_client *client = to_i2c_client(dev); - struct pc87360_data *data = i2c_get_clientdata(client); + struct pc87360_data *data = dev_get_drvdata(dev); long val = simple_strtol(buf, NULL, 10); mutex_lock(&data->update_lock); @@ -715,8 +705,7 @@ static ssize_t set_temp_min(struct device *dev, struct device_attribute *devattr size_t count) { struct sensor_device_attribute *attr = to_sensor_dev_attr(devattr); - struct i2c_client *client = to_i2c_client(dev); - struct pc87360_data *data = i2c_get_clientdata(client); + struct pc87360_data *data = dev_get_drvdata(dev); long val = simple_strtol(buf, NULL, 10); mutex_lock(&data->update_lock); @@ -730,8 +719,7 @@ static ssize_t set_temp_max(struct device *dev, struct device_attribute *devattr size_t count) { struct sensor_device_attribute *attr = to_sensor_dev_attr(devattr); - struct i2c_client *client = to_i2c_client(dev); - struct pc87360_data *data = i2c_get_clientdata(client); + struct pc87360_data *data = dev_get_drvdata(dev); long val = simple_strtol(buf, NULL, 10); mutex_lock(&data->update_lock); @@ -745,8 +733,7 @@ static ssize_t set_temp_crit(struct device *dev, struct device_attribute *devatt size_t count) { struct sensor_device_attribute *attr = to_sensor_dev_attr(devattr); - struct i2c_client *client = to_i2c_client(dev); - struct pc87360_data *data = i2c_get_clientdata(client); + struct pc87360_data *data = dev_get_drvdata(dev); long val = simple_strtol(buf, NULL, 10); mutex_lock(&data->update_lock); @@ -818,6 +805,14 @@ static const struct attribute_group pc8736x_temp_group = { .attrs = pc8736x_temp_attr_array, }; +static ssize_t show_name(struct device *dev, struct device_attribute + *devattr, char *buf) +{ + struct pc87360_data *data = dev_get_drvdata(dev); + return sprintf(buf, "%s\n", data->name); +} +static DEVICE_ATTR(name, S_IRUGO, show_name, NULL); + /* * Device detection, registration and update */ @@ -912,28 +907,18 @@ static int __init pc87360_find(int sioaddr, u8 *devid, unsigned short *addresses return 0; } -static int pc87360_detect(struct i2c_adapter *adapter) +static int __devinit pc87360_probe(struct platform_device *pdev) { int i; - struct i2c_client *client; struct pc87360_data *data; int err = 0; const char *name = "pc87360"; int use_thermistors = 0; - struct device *dev; + struct device *dev = &pdev->dev; if (!(data = kzalloc(sizeof(struct pc87360_data), GFP_KERNEL))) return -ENOMEM; - client = &data->client; - dev = &client->dev; - i2c_set_clientdata(client, data); - client->addr = address; - mutex_init(&data->lock); - client->adapter = adapter; - client->driver = &pc87360_driver; - client->flags = 0; - data->fannr = 2; data->innr = 0; data->tempnr = 0; @@ -960,15 +945,17 @@ static int pc87360_detect(struct i2c_adapter *adapter) break; } - strlcpy(client->name, name, sizeof(client->name)); + data->name = name; data->valid = 0; + mutex_init(&data->lock); mutex_init(&data->update_lock); + platform_set_drvdata(pdev, data); for (i = 0; i < 3; i++) { if (((data->address[i] = extra_isa[i])) && !request_region(extra_isa[i], PC87360_EXTENT, pc87360_driver.driver.name)) { - dev_err(&client->dev, "Region 0x%x-0x%x already " + dev_err(dev, "Region 0x%x-0x%x already " "in use!\n", extra_isa[i], extra_isa[i]+PC87360_EXTENT-1); for (i--; i >= 0; i--) @@ -982,9 +969,6 @@ static int pc87360_detect(struct i2c_adapter *adapter) if (data->fannr) data->fan_conf = confreg[0] | (confreg[1] << 8); - if ((err = i2c_attach_client(client))) - goto ERROR2; - /* Use the correct reference voltage Unless both the VLM and the TMS logical devices agree to use an external Vref, the internal one is used. */ @@ -996,7 +980,7 @@ static int pc87360_detect(struct i2c_adapter *adapter) PC87365_REG_TEMP_CONFIG); } data->in_vref = (i&0x02) ? 3025 : 2966; - dev_dbg(&client->dev, "Using %s reference voltage\n", + dev_dbg(dev, "Using %s reference voltage\n", (i&0x02) ? "external" : "internal"); data->vid_conf = confreg[3]; @@ -1015,18 +999,18 @@ static int pc87360_detect(struct i2c_adapter *adapter) if (devid == 0xe9 && data->address[1]) /* PC87366 */ use_thermistors = confreg[2] & 0x40; - pc87360_init_client(client, use_thermistors); + pc87360_init_device(pdev, use_thermistors); } /* Register all-or-nothing sysfs groups */ if (data->innr && - (err = sysfs_create_group(&client->dev.kobj, + (err = sysfs_create_group(&dev->kobj, &pc8736x_vin_group))) goto ERROR3; if (data->innr == 14 && - (err = sysfs_create_group(&client->dev.kobj, + (err = sysfs_create_group(&dev->kobj, &pc8736x_therm_group))) goto ERROR3; @@ -1067,7 +1051,10 @@ static int pc87360_detect(struct i2c_adapter *adapter) goto ERROR3; } - data->class_dev = hwmon_device_register(&client->dev); + if ((err = device_create_file(dev, &dev_attr_name))) + goto ERROR3; + + data->class_dev = hwmon_device_register(dev); if (IS_ERR(data->class_dev)) { err = PTR_ERR(data->class_dev); goto ERROR3; @@ -1075,14 +1062,12 @@ static int pc87360_detect(struct i2c_adapter *adapter) return 0; ERROR3: + device_remove_file(dev, &dev_attr_name); /* can still remove groups whose members were added individually */ - sysfs_remove_group(&client->dev.kobj, &pc8736x_temp_group); - sysfs_remove_group(&client->dev.kobj, &pc8736x_fan_group); - sysfs_remove_group(&client->dev.kobj, &pc8736x_therm_group); - sysfs_remove_group(&client->dev.kobj, &pc8736x_vin_group); - - i2c_detach_client(client); -ERROR2: + sysfs_remove_group(&dev->kobj, &pc8736x_temp_group); + sysfs_remove_group(&dev->kobj, &pc8736x_fan_group); + sysfs_remove_group(&dev->kobj, &pc8736x_therm_group); + sysfs_remove_group(&dev->kobj, &pc8736x_vin_group); for (i = 0; i < 3; i++) { if (data->address[i]) { release_region(data->address[i], PC87360_EXTENT); @@ -1093,20 +1078,18 @@ ERROR1: return err; } -static int pc87360_detach_client(struct i2c_client *client) +static int __devexit pc87360_remove(struct platform_device *pdev) { - struct pc87360_data *data = i2c_get_clientdata(client); + struct pc87360_data *data = platform_get_drvdata(pdev); int i; hwmon_device_unregister(data->class_dev); - sysfs_remove_group(&client->dev.kobj, &pc8736x_temp_group); - sysfs_remove_group(&client->dev.kobj, &pc8736x_fan_group); - sysfs_remove_group(&client->dev.kobj, &pc8736x_therm_group); - sysfs_remove_group(&client->dev.kobj, &pc8736x_vin_group); - - if ((i = i2c_detach_client(client))) - return i; + device_remove_file(&pdev->dev, &dev_attr_name); + sysfs_remove_group(&pdev->dev.kobj, &pc8736x_temp_group); + sysfs_remove_group(&pdev->dev.kobj, &pc8736x_fan_group); + sysfs_remove_group(&pdev->dev.kobj, &pc8736x_therm_group); + sysfs_remove_group(&pdev->dev.kobj, &pc8736x_vin_group); for (i = 0; i < 3; i++) { if (data->address[i]) { @@ -1144,9 +1127,10 @@ static void pc87360_write_value(struct pc87360_data *data, u8 ldi, u8 bank, mutex_unlock(&(data->lock)); } -static void pc87360_init_client(struct i2c_client *client, int use_thermistors) +static void pc87360_init_device(struct platform_device *pdev, + int use_thermistors) { - struct pc87360_data *data = i2c_get_clientdata(client); + struct pc87360_data *data = platform_get_drvdata(pdev); int i, nr; const u8 init_in[14] = { 2, 2, 2, 2, 2, 2, 2, 1, 1, 3, 1, 2, 2, 2 }; const u8 init_temp[3] = { 2, 2, 1 }; @@ -1155,7 +1139,7 @@ static void pc87360_init_client(struct i2c_client *client, int use_thermistors) if (init >= 2 && data->innr) { reg = pc87360_read_value(data, LD_IN, NO_BANK, PC87365_REG_IN_CONVRATE); - dev_info(&client->dev, "VLM conversion set to " + dev_info(&pdev->dev, "VLM conversion set to " "1s period, 160us delay\n"); pc87360_write_value(data, LD_IN, NO_BANK, PC87365_REG_IN_CONVRATE, @@ -1169,7 +1153,7 @@ static void pc87360_init_client(struct i2c_client *client, int use_thermistors) reg = pc87360_read_value(data, LD_IN, i, PC87365_REG_IN_STATUS); if (!(reg & 0x01)) { - dev_dbg(&client->dev, "Forcibly " + dev_dbg(&pdev->dev, "Forcibly " "enabling in%d\n", i); pc87360_write_value(data, LD_IN, i, PC87365_REG_IN_STATUS, @@ -1193,7 +1177,7 @@ static void pc87360_init_client(struct i2c_client *client, int use_thermistors) reg = pc87360_read_value(data, LD_TEMP, i, PC87365_REG_TEMP_STATUS); if (!(reg & 0x01)) { - dev_dbg(&client->dev, "Forcibly " + dev_dbg(&pdev->dev, "Forcibly " "enabling temp%d\n", i+1); pc87360_write_value(data, LD_TEMP, i, PC87365_REG_TEMP_STATUS, @@ -1210,7 +1194,7 @@ static void pc87360_init_client(struct i2c_client *client, int use_thermistors) reg = pc87360_read_value(data, LD_TEMP, (i-11)/2, PC87365_REG_TEMP_STATUS); if (reg & 0x01) { - dev_dbg(&client->dev, "Skipping " + dev_dbg(&pdev->dev, "Skipping " "temp%d, pin already in use " "by temp%d\n", i-7, (i-11)/2); continue; @@ -1220,7 +1204,7 @@ static void pc87360_init_client(struct i2c_client *client, int use_thermistors) reg = pc87360_read_value(data, LD_IN, i, PC87365_REG_IN_STATUS); if (!(reg & 0x01)) { - dev_dbg(&client->dev, "Forcibly " + dev_dbg(&pdev->dev, "Forcibly " "enabling temp%d\n", i-7); pc87360_write_value(data, LD_IN, i, PC87365_REG_TEMP_STATUS, @@ -1234,7 +1218,7 @@ static void pc87360_init_client(struct i2c_client *client, int use_thermistors) reg = pc87360_read_value(data, LD_IN, NO_BANK, PC87365_REG_IN_CONFIG); if (reg & 0x01) { - dev_dbg(&client->dev, "Forcibly " + dev_dbg(&pdev->dev, "Forcibly " "enabling monitoring (VLM)\n"); pc87360_write_value(data, LD_IN, NO_BANK, PC87365_REG_IN_CONFIG, @@ -1246,7 +1230,7 @@ static void pc87360_init_client(struct i2c_client *client, int use_thermistors) reg = pc87360_read_value(data, LD_TEMP, NO_BANK, PC87365_REG_TEMP_CONFIG); if (reg & 0x01) { - dev_dbg(&client->dev, "Forcibly enabling " + dev_dbg(&pdev->dev, "Forcibly enabling " "monitoring (TMS)\n"); pc87360_write_value(data, LD_TEMP, NO_BANK, PC87365_REG_TEMP_CONFIG, @@ -1268,9 +1252,9 @@ static void pc87360_init_client(struct i2c_client *client, int use_thermistors) } } -static void pc87360_autodiv(struct i2c_client *client, int nr) +static void pc87360_autodiv(struct device *dev, int nr) { - struct pc87360_data *data = i2c_get_clientdata(client); + struct pc87360_data *data = dev_get_drvdata(dev); u8 old_min = data->fan_min[nr]; /* Increase clock divider if needed and possible */ @@ -1280,7 +1264,7 @@ static void pc87360_autodiv(struct i2c_client *client, int nr) data->fan_status[nr] += 0x20; data->fan_min[nr] >>= 1; data->fan[nr] >>= 1; - dev_dbg(&client->dev, "Increasing " + dev_dbg(dev, "Increasing " "clock divider to %d for fan %d\n", FAN_DIV_FROM_REG(data->fan_status[nr]), nr+1); } @@ -1292,7 +1276,7 @@ static void pc87360_autodiv(struct i2c_client *client, int nr) data->fan_status[nr] -= 0x20; data->fan_min[nr] <<= 1; data->fan[nr] <<= 1; - dev_dbg(&client->dev, "Decreasing " + dev_dbg(dev, "Decreasing " "clock divider to %d for fan %d\n", FAN_DIV_FROM_REG(data->fan_status[nr]), nr+1); @@ -1309,14 +1293,13 @@ static void pc87360_autodiv(struct i2c_client *client, int nr) static struct pc87360_data *pc87360_update_device(struct device *dev) { - struct i2c_client *client = to_i2c_client(dev); - struct pc87360_data *data = i2c_get_clientdata(client); + struct pc87360_data *data = dev_get_drvdata(dev); u8 i; mutex_lock(&data->update_lock); if (time_after(jiffies, data->last_updated + HZ * 2) || !data->valid) { - dev_dbg(&client->dev, "Data update\n"); + dev_dbg(dev, "Data update\n"); /* Fans */ for (i = 0; i < data->fannr; i++) { @@ -1330,7 +1313,7 @@ static struct pc87360_data *pc87360_update_device(struct device *dev) LD_FAN, NO_BANK, PC87360_REG_FAN_MIN(i)); /* Change clock divider if needed */ - pc87360_autodiv(client, i); + pc87360_autodiv(dev, i); /* Clear bits and write new divider */ pc87360_write_value(data, LD_FAN, NO_BANK, PC87360_REG_FAN_STATUS(i), @@ -1418,9 +1401,53 @@ static struct pc87360_data *pc87360_update_device(struct device *dev) return data; } +static int __init pc87360_device_add(unsigned short address) +{ + struct resource res = { + .name = "pc87360", + .flags = IORESOURCE_IO, + }; + int err, i; + + pdev = platform_device_alloc("pc87360", address); + if (!pdev) { + err = -ENOMEM; + printk(KERN_ERR "pc87360: Device allocation failed\n"); + goto exit; + } + + for (i = 0; i < 3; i++) { + if (!extra_isa[i]) + continue; + res.start = extra_isa[i]; + res.end = extra_isa[i] + PC87360_EXTENT - 1; + err = platform_device_add_resources(pdev, &res, 1); + if (err) { + printk(KERN_ERR "pc87360: Device resource[%d] " + "addition failed (%d)\n", i, err); + goto exit_device_put; + } + } + + err = platform_device_add(pdev); + if (err) { + printk(KERN_ERR "pc87360: Device addition failed (%d)\n", + err); + goto exit_device_put; + } + + return 0; + +exit_device_put: + platform_device_put(pdev); +exit: + return err; +} + static int __init pc87360_init(void) { - int i; + int err, i; + unsigned short address = 0; if (pc87360_find(0x2e, &devid, extra_isa) && pc87360_find(0x4e, &devid, extra_isa)) { @@ -1443,12 +1470,27 @@ static int __init pc87360_init(void) return -ENODEV; } - return i2c_isa_add_driver(&pc87360_driver); + err = platform_driver_register(&pc87360_driver); + if (err) + goto exit; + + /* Sets global pdev as a side effect */ + err = pc87360_device_add(address); + if (err) + goto exit_driver; + + return 0; + + exit_driver: + platform_driver_unregister(&pc87360_driver); + exit: + return err; } static void __exit pc87360_exit(void) { - i2c_isa_del_driver(&pc87360_driver); + platform_device_unregister(pdev); + platform_driver_unregister(&pc87360_driver); } diff --git a/drivers/hwmon/pc87427.c b/drivers/hwmon/pc87427.c index 29354fa26f8..2915bc4ad0d 100644 --- a/drivers/hwmon/pc87427.c +++ b/drivers/hwmon/pc87427.c @@ -484,7 +484,6 @@ static int __devexit pc87427_remove(struct platform_device *pdev) struct resource *res; int i; - platform_set_drvdata(pdev, NULL); hwmon_device_unregister(data->class_dev); device_remove_file(&pdev->dev, &dev_attr_name); for (i = 0; i < 8; i++) { @@ -492,6 +491,7 @@ static int __devexit pc87427_remove(struct platform_device *pdev) continue; sysfs_remove_group(&pdev->dev.kobj, &pc87427_group_fan[i]); } + platform_set_drvdata(pdev, NULL); kfree(data); res = platform_get_resource(pdev, IORESOURCE_IO, 0); diff --git a/drivers/hwmon/sis5595.c b/drivers/hwmon/sis5595.c index 3f400263fc0..83321b28cf0 100644 --- a/drivers/hwmon/sis5595.c +++ b/drivers/hwmon/sis5595.c @@ -54,9 +54,9 @@ #include <linux/slab.h> #include <linux/ioport.h> #include <linux/pci.h> -#include <linux/i2c.h> -#include <linux/i2c-isa.h> +#include <linux/platform_device.h> #include <linux/hwmon.h> +#include <linux/hwmon-sysfs.h> #include <linux/err.h> #include <linux/init.h> #include <linux/jiffies.h> @@ -72,17 +72,13 @@ module_param(force_addr, ushort, 0); MODULE_PARM_DESC(force_addr, "Initialize the base address of the sensors"); -/* Device address - Note that we can't determine the ISA address until we have initialized - our module */ -static unsigned short address; +static struct platform_device *pdev; /* Many SIS5595 constants specified below */ /* Length of ISA address segment */ #define SIS5595_EXTENT 8 /* PCI Config Registers */ -#define SIS5595_REVISION_REG 0x08 #define SIS5595_BASE_REG 0x68 #define SIS5595_PIN_REG 0x7A #define SIS5595_ENABLE_REG 0x7B @@ -165,7 +161,8 @@ static inline u8 DIV_TO_REG(int val) /* For each registered chip, we need to keep some data in memory. The structure is dynamically allocated. */ struct sis5595_data { - struct i2c_client client; + unsigned short addr; + const char *name; struct class_device *class_dev; struct mutex lock; @@ -189,102 +186,88 @@ struct sis5595_data { static struct pci_dev *s_bridge; /* pointer to the (only) sis5595 */ -static int sis5595_detect(struct i2c_adapter *adapter); -static int sis5595_detach_client(struct i2c_client *client); +static int sis5595_probe(struct platform_device *pdev); +static int sis5595_remove(struct platform_device *pdev); -static int sis5595_read_value(struct i2c_client *client, u8 reg); -static int sis5595_write_value(struct i2c_client *client, u8 reg, u8 value); +static int sis5595_read_value(struct sis5595_data *data, u8 reg); +static void sis5595_write_value(struct sis5595_data *data, u8 reg, u8 value); static struct sis5595_data *sis5595_update_device(struct device *dev); -static void sis5595_init_client(struct i2c_client *client); +static void sis5595_init_device(struct sis5595_data *data); -static struct i2c_driver sis5595_driver = { +static struct platform_driver sis5595_driver = { .driver = { .owner = THIS_MODULE, .name = "sis5595", }, - .attach_adapter = sis5595_detect, - .detach_client = sis5595_detach_client, + .probe = sis5595_probe, + .remove = __devexit_p(sis5595_remove), }; /* 4 Voltages */ -static ssize_t show_in(struct device *dev, char *buf, int nr) +static ssize_t show_in(struct device *dev, struct device_attribute *da, + char *buf) { struct sis5595_data *data = sis5595_update_device(dev); + struct sensor_device_attribute *attr = to_sensor_dev_attr(da); + int nr = attr->index; return sprintf(buf, "%d\n", IN_FROM_REG(data->in[nr])); } -static ssize_t show_in_min(struct device *dev, char *buf, int nr) +static ssize_t show_in_min(struct device *dev, struct device_attribute *da, + char *buf) { struct sis5595_data *data = sis5595_update_device(dev); + struct sensor_device_attribute *attr = to_sensor_dev_attr(da); + int nr = attr->index; return sprintf(buf, "%d\n", IN_FROM_REG(data->in_min[nr])); } -static ssize_t show_in_max(struct device *dev, char *buf, int nr) +static ssize_t show_in_max(struct device *dev, struct device_attribute *da, + char *buf) { struct sis5595_data *data = sis5595_update_device(dev); + struct sensor_device_attribute *attr = to_sensor_dev_attr(da); + int nr = attr->index; return sprintf(buf, "%d\n", IN_FROM_REG(data->in_max[nr])); } -static ssize_t set_in_min(struct device *dev, const char *buf, - size_t count, int nr) +static ssize_t set_in_min(struct device *dev, struct device_attribute *da, + const char *buf, size_t count) { - struct i2c_client *client = to_i2c_client(dev); - struct sis5595_data *data = i2c_get_clientdata(client); + struct sis5595_data *data = dev_get_drvdata(dev); + struct sensor_device_attribute *attr = to_sensor_dev_attr(da); + int nr = attr->index; unsigned long val = simple_strtoul(buf, NULL, 10); mutex_lock(&data->update_lock); data->in_min[nr] = IN_TO_REG(val); - sis5595_write_value(client, SIS5595_REG_IN_MIN(nr), data->in_min[nr]); + sis5595_write_value(data, SIS5595_REG_IN_MIN(nr), data->in_min[nr]); mutex_unlock(&data->update_lock); return count; } -static ssize_t set_in_max(struct device *dev, const char *buf, - size_t count, int nr) +static ssize_t set_in_max(struct device *dev, struct device_attribute *da, + const char *buf, size_t count) { - struct i2c_client *client = to_i2c_client(dev); - struct sis5595_data *data = i2c_get_clientdata(client); + struct sis5595_data *data = dev_get_drvdata(dev); + struct sensor_device_attribute *attr = to_sensor_dev_attr(da); + int nr = attr->index; unsigned long val = simple_strtoul(buf, NULL, 10); mutex_lock(&data->update_lock); data->in_max[nr] = IN_TO_REG(val); - sis5595_write_value(client, SIS5595_REG_IN_MAX(nr), data->in_max[nr]); + sis5595_write_value(data, SIS5595_REG_IN_MAX(nr), data->in_max[nr]); mutex_unlock(&data->update_lock); return count; } #define show_in_offset(offset) \ -static ssize_t \ - show_in##offset (struct device *dev, struct device_attribute *attr, char *buf) \ -{ \ - return show_in(dev, buf, offset); \ -} \ -static DEVICE_ATTR(in##offset##_input, S_IRUGO, \ - show_in##offset, NULL); \ -static ssize_t \ - show_in##offset##_min (struct device *dev, struct device_attribute *attr, char *buf) \ -{ \ - return show_in_min(dev, buf, offset); \ -} \ -static ssize_t \ - show_in##offset##_max (struct device *dev, struct device_attribute *attr, char *buf) \ -{ \ - return show_in_max(dev, buf, offset); \ -} \ -static ssize_t set_in##offset##_min (struct device *dev, struct device_attribute *attr, \ - const char *buf, size_t count) \ -{ \ - return set_in_min(dev, buf, count, offset); \ -} \ -static ssize_t set_in##offset##_max (struct device *dev, struct device_attribute *attr, \ - const char *buf, size_t count) \ -{ \ - return set_in_max(dev, buf, count, offset); \ -} \ -static DEVICE_ATTR(in##offset##_min, S_IRUGO | S_IWUSR, \ - show_in##offset##_min, set_in##offset##_min); \ -static DEVICE_ATTR(in##offset##_max, S_IRUGO | S_IWUSR, \ - show_in##offset##_max, set_in##offset##_max); +static SENSOR_DEVICE_ATTR(in##offset##_input, S_IRUGO, \ + show_in, NULL, offset); \ +static SENSOR_DEVICE_ATTR(in##offset##_min, S_IRUGO | S_IWUSR, \ + show_in_min, set_in_min, offset); \ +static SENSOR_DEVICE_ATTR(in##offset##_max, S_IRUGO | S_IWUSR, \ + show_in_max, set_in_max, offset); show_in_offset(0); show_in_offset(1); @@ -307,13 +290,12 @@ static ssize_t show_temp_over(struct device *dev, struct device_attribute *attr, static ssize_t set_temp_over(struct device *dev, struct device_attribute *attr, const char *buf, size_t count) { - struct i2c_client *client = to_i2c_client(dev); - struct sis5595_data *data = i2c_get_clientdata(client); + struct sis5595_data *data = dev_get_drvdata(dev); long val = simple_strtol(buf, NULL, 10); mutex_lock(&data->update_lock); data->temp_over = TEMP_TO_REG(val); - sis5595_write_value(client, SIS5595_REG_TEMP_OVER, data->temp_over); + sis5595_write_value(data, SIS5595_REG_TEMP_OVER, data->temp_over); mutex_unlock(&data->update_lock); return count; } @@ -326,13 +308,12 @@ static ssize_t show_temp_hyst(struct device *dev, struct device_attribute *attr, static ssize_t set_temp_hyst(struct device *dev, struct device_attribute *attr, const char *buf, size_t count) { - struct i2c_client *client = to_i2c_client(dev); - struct sis5595_data *data = i2c_get_clientdata(client); + struct sis5595_data *data = dev_get_drvdata(dev); long val = simple_strtol(buf, NULL, 10); mutex_lock(&data->update_lock); data->temp_hyst = TEMP_TO_REG(val); - sis5595_write_value(client, SIS5595_REG_TEMP_HYST, data->temp_hyst); + sis5595_write_value(data, SIS5595_REG_TEMP_HYST, data->temp_hyst); mutex_unlock(&data->update_lock); return count; } @@ -344,37 +325,47 @@ static DEVICE_ATTR(temp1_max_hyst, S_IRUGO | S_IWUSR, show_temp_hyst, set_temp_hyst); /* 2 Fans */ -static ssize_t show_fan(struct device *dev, char *buf, int nr) +static ssize_t show_fan(struct device *dev, struct device_attribute *da, + char *buf) { struct sis5595_data *data = sis5595_update_device(dev); + struct sensor_device_attribute *attr = to_sensor_dev_attr(da); + int nr = attr->index; return sprintf(buf, "%d\n", FAN_FROM_REG(data->fan[nr], DIV_FROM_REG(data->fan_div[nr])) ); } -static ssize_t show_fan_min(struct device *dev, char *buf, int nr) +static ssize_t show_fan_min(struct device *dev, struct device_attribute *da, + char *buf) { struct sis5595_data *data = sis5595_update_device(dev); + struct sensor_device_attribute *attr = to_sensor_dev_attr(da); + int nr = attr->index; return sprintf(buf,"%d\n", FAN_FROM_REG(data->fan_min[nr], DIV_FROM_REG(data->fan_div[nr])) ); } -static ssize_t set_fan_min(struct device *dev, const char *buf, - size_t count, int nr) +static ssize_t set_fan_min(struct device *dev, struct device_attribute *da, + const char *buf, size_t count) { - struct i2c_client *client = to_i2c_client(dev); - struct sis5595_data *data = i2c_get_clientdata(client); + struct sis5595_data *data = dev_get_drvdata(dev); + struct sensor_device_attribute *attr = to_sensor_dev_attr(da); + int nr = attr->index; unsigned long val = simple_strtoul(buf, NULL, 10); mutex_lock(&data->update_lock); data->fan_min[nr] = FAN_TO_REG(val, DIV_FROM_REG(data->fan_div[nr])); - sis5595_write_value(client, SIS5595_REG_FAN_MIN(nr), data->fan_min[nr]); + sis5595_write_value(data, SIS5595_REG_FAN_MIN(nr), data->fan_min[nr]); mutex_unlock(&data->update_lock); return count; } -static ssize_t show_fan_div(struct device *dev, char *buf, int nr) +static ssize_t show_fan_div(struct device *dev, struct device_attribute *da, + char *buf) { struct sis5595_data *data = sis5595_update_device(dev); + struct sensor_device_attribute *attr = to_sensor_dev_attr(da); + int nr = attr->index; return sprintf(buf, "%d\n", DIV_FROM_REG(data->fan_div[nr]) ); } @@ -382,11 +373,12 @@ static ssize_t show_fan_div(struct device *dev, char *buf, int nr) determined in part by the fan divisor. This follows the principle of least surprise; the user doesn't expect the fan minimum to change just because the divisor changed. */ -static ssize_t set_fan_div(struct device *dev, const char *buf, - size_t count, int nr) +static ssize_t set_fan_div(struct device *dev, struct device_attribute *da, + const char *buf, size_t count) { - struct i2c_client *client = to_i2c_client(dev); - struct sis5595_data *data = i2c_get_clientdata(client); + struct sis5595_data *data = dev_get_drvdata(dev); + struct sensor_device_attribute *attr = to_sensor_dev_attr(da); + int nr = attr->index; unsigned long min; unsigned long val = simple_strtoul(buf, NULL, 10); int reg; @@ -394,7 +386,7 @@ static ssize_t set_fan_div(struct device *dev, const char *buf, mutex_lock(&data->update_lock); min = FAN_FROM_REG(data->fan_min[nr], DIV_FROM_REG(data->fan_div[nr])); - reg = sis5595_read_value(client, SIS5595_REG_FANDIV); + reg = sis5595_read_value(data, SIS5595_REG_FANDIV); switch (val) { case 1: data->fan_div[nr] = 0; break; @@ -402,7 +394,7 @@ static ssize_t set_fan_div(struct device *dev, const char *buf, case 4: data->fan_div[nr] = 2; break; case 8: data->fan_div[nr] = 3; break; default: - dev_err(&client->dev, "fan_div value %ld not " + dev_err(dev, "fan_div value %ld not " "supported. Choose one of 1, 2, 4 or 8!\n", val); mutex_unlock(&data->update_lock); return -EINVAL; @@ -416,55 +408,25 @@ static ssize_t set_fan_div(struct device *dev, const char *buf, reg = (reg & 0x3f) | (data->fan_div[nr] << 6); break; } - sis5595_write_value(client, SIS5595_REG_FANDIV, reg); + sis5595_write_value(data, SIS5595_REG_FANDIV, reg); data->fan_min[nr] = FAN_TO_REG(min, DIV_FROM_REG(data->fan_div[nr])); - sis5595_write_value(client, SIS5595_REG_FAN_MIN(nr), data->fan_min[nr]); + sis5595_write_value(data, SIS5595_REG_FAN_MIN(nr), data->fan_min[nr]); mutex_unlock(&data->update_lock); return count; } #define show_fan_offset(offset) \ -static ssize_t show_fan_##offset (struct device *dev, struct device_attribute *attr, char *buf) \ -{ \ - return show_fan(dev, buf, offset - 1); \ -} \ -static ssize_t show_fan_##offset##_min (struct device *dev, struct device_attribute *attr, char *buf) \ -{ \ - return show_fan_min(dev, buf, offset - 1); \ -} \ -static ssize_t show_fan_##offset##_div (struct device *dev, struct device_attribute *attr, char *buf) \ -{ \ - return show_fan_div(dev, buf, offset - 1); \ -} \ -static ssize_t set_fan_##offset##_min (struct device *dev, struct device_attribute *attr, \ - const char *buf, size_t count) \ -{ \ - return set_fan_min(dev, buf, count, offset - 1); \ -} \ -static DEVICE_ATTR(fan##offset##_input, S_IRUGO, show_fan_##offset, NULL);\ -static DEVICE_ATTR(fan##offset##_min, S_IRUGO | S_IWUSR, \ - show_fan_##offset##_min, set_fan_##offset##_min); +static SENSOR_DEVICE_ATTR(fan##offset##_input, S_IRUGO, \ + show_fan, NULL, offset - 1); \ +static SENSOR_DEVICE_ATTR(fan##offset##_min, S_IRUGO | S_IWUSR, \ + show_fan_min, set_fan_min, offset - 1); \ +static SENSOR_DEVICE_ATTR(fan##offset##_div, S_IRUGO | S_IWUSR, \ + show_fan_div, set_fan_div, offset - 1); show_fan_offset(1); show_fan_offset(2); -static ssize_t set_fan_1_div(struct device *dev, struct device_attribute *attr, const char *buf, - size_t count) -{ - return set_fan_div(dev, buf, count, 0) ; -} - -static ssize_t set_fan_2_div(struct device *dev, struct device_attribute *attr, const char *buf, - size_t count) -{ - return set_fan_div(dev, buf, count, 1) ; -} -static DEVICE_ATTR(fan1_div, S_IRUGO | S_IWUSR, - show_fan_1_div, set_fan_1_div); -static DEVICE_ATTR(fan2_div, S_IRUGO | S_IWUSR, - show_fan_2_div, set_fan_2_div); - /* Alarms */ static ssize_t show_alarms(struct device *dev, struct device_attribute *attr, char *buf) { @@ -473,28 +435,37 @@ static ssize_t show_alarms(struct device *dev, struct device_attribute *attr, ch } static DEVICE_ATTR(alarms, S_IRUGO, show_alarms, NULL); +static ssize_t show_name(struct device *dev, struct device_attribute *attr, + char *buf) +{ + struct sis5595_data *data = dev_get_drvdata(dev); + return sprintf(buf, "%s\n", data->name); +} +static DEVICE_ATTR(name, S_IRUGO, show_name, NULL); + static struct attribute *sis5595_attributes[] = { - &dev_attr_in0_input.attr, - &dev_attr_in0_min.attr, - &dev_attr_in0_max.attr, - &dev_attr_in1_input.attr, - &dev_attr_in1_min.attr, - &dev_attr_in1_max.attr, - &dev_attr_in2_input.attr, - &dev_attr_in2_min.attr, - &dev_attr_in2_max.attr, - &dev_attr_in3_input.attr, - &dev_attr_in3_min.attr, - &dev_attr_in3_max.attr, - - &dev_attr_fan1_input.attr, - &dev_attr_fan1_min.attr, - &dev_attr_fan1_div.attr, - &dev_attr_fan2_input.attr, - &dev_attr_fan2_min.attr, - &dev_attr_fan2_div.attr, + &sensor_dev_attr_in0_input.dev_attr.attr, + &sensor_dev_attr_in0_min.dev_attr.attr, + &sensor_dev_attr_in0_max.dev_attr.attr, + &sensor_dev_attr_in1_input.dev_attr.attr, + &sensor_dev_attr_in1_min.dev_attr.attr, + &sensor_dev_attr_in1_max.dev_attr.attr, + &sensor_dev_attr_in2_input.dev_attr.attr, + &sensor_dev_attr_in2_min.dev_attr.attr, + &sensor_dev_attr_in2_max.dev_attr.attr, + &sensor_dev_attr_in3_input.dev_attr.attr, + &sensor_dev_attr_in3_min.dev_attr.attr, + &sensor_dev_attr_in3_max.dev_attr.attr, + + &sensor_dev_attr_fan1_input.dev_attr.attr, + &sensor_dev_attr_fan1_min.dev_attr.attr, + &sensor_dev_attr_fan1_div.dev_attr.attr, + &sensor_dev_attr_fan2_input.dev_attr.attr, + &sensor_dev_attr_fan2_min.dev_attr.attr, + &sensor_dev_attr_fan2_div.dev_attr.attr, &dev_attr_alarms.attr, + &dev_attr_name.attr, NULL }; @@ -503,9 +474,9 @@ static const struct attribute_group sis5595_group = { }; static struct attribute *sis5595_attributes_opt[] = { - &dev_attr_in4_input.attr, - &dev_attr_in4_min.attr, - &dev_attr_in4_max.attr, + &sensor_dev_attr_in4_input.dev_attr.attr, + &sensor_dev_attr_in4_min.dev_attr.attr, + &sensor_dev_attr_in4_max.dev_attr.attr, &dev_attr_temp1_input.attr, &dev_attr_temp1_max.attr, @@ -518,68 +489,35 @@ static const struct attribute_group sis5595_group_opt = { }; /* This is called when the module is loaded */ -static int sis5595_detect(struct i2c_adapter *adapter) +static int __devinit sis5595_probe(struct platform_device *pdev) { int err = 0; int i; - struct i2c_client *new_client; struct sis5595_data *data; + struct resource *res; char val; - u16 a; - if (force_addr) - address = force_addr & ~(SIS5595_EXTENT - 1); /* Reserve the ISA region */ - if (!request_region(address, SIS5595_EXTENT, + res = platform_get_resource(pdev, IORESOURCE_IO, 0); + if (!request_region(res->start, SIS5595_EXTENT, sis5595_driver.driver.name)) { err = -EBUSY; goto exit; } - if (force_addr) { - dev_warn(&adapter->dev, "forcing ISA address 0x%04X\n", address); - if (PCIBIOS_SUCCESSFUL != - pci_write_config_word(s_bridge, SIS5595_BASE_REG, address)) - goto exit_release; - if (PCIBIOS_SUCCESSFUL != - pci_read_config_word(s_bridge, SIS5595_BASE_REG, &a)) - goto exit_release; - if ((a & ~(SIS5595_EXTENT - 1)) != address) - /* doesn't work for some chips? */ - goto exit_release; - } - - if (PCIBIOS_SUCCESSFUL != - pci_read_config_byte(s_bridge, SIS5595_ENABLE_REG, &val)) { - goto exit_release; - } - if ((val & 0x80) == 0) { - if (PCIBIOS_SUCCESSFUL != - pci_write_config_byte(s_bridge, SIS5595_ENABLE_REG, - val | 0x80)) - goto exit_release; - if (PCIBIOS_SUCCESSFUL != - pci_read_config_byte(s_bridge, SIS5595_ENABLE_REG, &val)) - goto exit_release; - if ((val & 0x80) == 0) - /* doesn't work for some chips! */ - goto exit_release; - } if (!(data = kzalloc(sizeof(struct sis5595_data), GFP_KERNEL))) { err = -ENOMEM; goto exit_release; } - new_client = &data->client; - new_client->addr = address; mutex_init(&data->lock); - i2c_set_clientdata(new_client, data); - new_client->adapter = adapter; - new_client->driver = &sis5595_driver; - new_client->flags = 0; + mutex_init(&data->update_lock); + data->addr = res->start; + data->name = "sis5595"; + platform_set_drvdata(pdev, data); /* Check revision and pin registers to determine whether 4 or 5 voltages */ - pci_read_config_byte(s_bridge, SIS5595_REVISION_REG, &(data->revision)); + pci_read_config_byte(s_bridge, PCI_REVISION_ID, &data->revision); /* 4 voltages, 1 temp */ data->maxins = 3; if (data->revision >= REV2MIN) { @@ -589,47 +527,37 @@ static int sis5595_detect(struct i2c_adapter *adapter) data->maxins = 4; } - /* Fill in the remaining client fields and put it into the global list */ - strlcpy(new_client->name, "sis5595", I2C_NAME_SIZE); - - data->valid = 0; - mutex_init(&data->update_lock); - - /* Tell the I2C layer a new client has arrived */ - if ((err = i2c_attach_client(new_client))) - goto exit_free; - /* Initialize the SIS5595 chip */ - sis5595_init_client(new_client); + sis5595_init_device(data); /* A few vars need to be filled upon startup */ for (i = 0; i < 2; i++) { - data->fan_min[i] = sis5595_read_value(new_client, + data->fan_min[i] = sis5595_read_value(data, SIS5595_REG_FAN_MIN(i)); } /* Register sysfs hooks */ - if ((err = sysfs_create_group(&new_client->dev.kobj, &sis5595_group))) - goto exit_detach; + if ((err = sysfs_create_group(&pdev->dev.kobj, &sis5595_group))) + goto exit_free; if (data->maxins == 4) { - if ((err = device_create_file(&new_client->dev, - &dev_attr_in4_input)) - || (err = device_create_file(&new_client->dev, - &dev_attr_in4_min)) - || (err = device_create_file(&new_client->dev, - &dev_attr_in4_max))) + if ((err = device_create_file(&pdev->dev, + &sensor_dev_attr_in4_input.dev_attr)) + || (err = device_create_file(&pdev->dev, + &sensor_dev_attr_in4_min.dev_attr)) + || (err = device_create_file(&pdev->dev, + &sensor_dev_attr_in4_max.dev_attr))) goto exit_remove_files; } else { - if ((err = device_create_file(&new_client->dev, + if ((err = device_create_file(&pdev->dev, &dev_attr_temp1_input)) - || (err = device_create_file(&new_client->dev, + || (err = device_create_file(&pdev->dev, &dev_attr_temp1_max)) - || (err = device_create_file(&new_client->dev, + || (err = device_create_file(&pdev->dev, &dev_attr_temp1_max_hyst))) goto exit_remove_files; } - data->class_dev = hwmon_device_register(&new_client->dev); + data->class_dev = hwmon_device_register(&pdev->dev); if (IS_ERR(data->class_dev)) { err = PTR_ERR(data->class_dev); goto exit_remove_files; @@ -638,32 +566,26 @@ static int sis5595_detect(struct i2c_adapter *adapter) return 0; exit_remove_files: - sysfs_remove_group(&new_client->dev.kobj, &sis5595_group); - sysfs_remove_group(&new_client->dev.kobj, &sis5595_group_opt); -exit_detach: - i2c_detach_client(new_client); + sysfs_remove_group(&pdev->dev.kobj, &sis5595_group); + sysfs_remove_group(&pdev->dev.kobj, &sis5595_group_opt); exit_free: kfree(data); exit_release: - release_region(address, SIS5595_EXTENT); + release_region(res->start, SIS5595_EXTENT); exit: return err; } -static int sis5595_detach_client(struct i2c_client *client) +static int __devexit sis5595_remove(struct platform_device *pdev) { - struct sis5595_data *data = i2c_get_clientdata(client); - int err; + struct sis5595_data *data = platform_get_drvdata(pdev); hwmon_device_unregister(data->class_dev); - sysfs_remove_group(&client->dev.kobj, &sis5595_group); - sysfs_remove_group(&client->dev.kobj, &sis5595_group_opt); - - if ((err = i2c_detach_client(client))) - return err; - - release_region(client->addr, SIS5595_EXTENT); + sysfs_remove_group(&pdev->dev.kobj, &sis5595_group); + sysfs_remove_group(&pdev->dev.kobj, &sis5595_group_opt); + release_region(data->addr, SIS5595_EXTENT); + platform_set_drvdata(pdev, NULL); kfree(data); return 0; @@ -671,41 +593,37 @@ static int sis5595_detach_client(struct i2c_client *client) /* ISA access must be locked explicitly. */ -static int sis5595_read_value(struct i2c_client *client, u8 reg) +static int sis5595_read_value(struct sis5595_data *data, u8 reg) { int res; - struct sis5595_data *data = i2c_get_clientdata(client); mutex_lock(&data->lock); - outb_p(reg, client->addr + SIS5595_ADDR_REG_OFFSET); - res = inb_p(client->addr + SIS5595_DATA_REG_OFFSET); + outb_p(reg, data->addr + SIS5595_ADDR_REG_OFFSET); + res = inb_p(data->addr + SIS5595_DATA_REG_OFFSET); mutex_unlock(&data->lock); return res; } -static int sis5595_write_value(struct i2c_client *client, u8 reg, u8 value) +static void sis5595_write_value(struct sis5595_data *data, u8 reg, u8 value) { - struct sis5595_data *data = i2c_get_clientdata(client); mutex_lock(&data->lock); - outb_p(reg, client->addr + SIS5595_ADDR_REG_OFFSET); - outb_p(value, client->addr + SIS5595_DATA_REG_OFFSET); + outb_p(reg, data->addr + SIS5595_ADDR_REG_OFFSET); + outb_p(value, data->addr + SIS5595_DATA_REG_OFFSET); mutex_unlock(&data->lock); - return 0; } /* Called when we have found a new SIS5595. */ -static void sis5595_init_client(struct i2c_client *client) +static void __devinit sis5595_init_device(struct sis5595_data *data) { - u8 config = sis5595_read_value(client, SIS5595_REG_CONFIG); + u8 config = sis5595_read_value(data, SIS5595_REG_CONFIG); if (!(config & 0x01)) - sis5595_write_value(client, SIS5595_REG_CONFIG, + sis5595_write_value(data, SIS5595_REG_CONFIG, (config & 0xf7) | 0x01); } static struct sis5595_data *sis5595_update_device(struct device *dev) { - struct i2c_client *client = to_i2c_client(dev); - struct sis5595_data *data = i2c_get_clientdata(client); + struct sis5595_data *data = dev_get_drvdata(dev); int i; mutex_lock(&data->update_lock); @@ -715,35 +633,35 @@ static struct sis5595_data *sis5595_update_device(struct device *dev) for (i = 0; i <= data->maxins; i++) { data->in[i] = - sis5595_read_value(client, SIS5595_REG_IN(i)); + sis5595_read_value(data, SIS5595_REG_IN(i)); data->in_min[i] = - sis5595_read_value(client, + sis5595_read_value(data, SIS5595_REG_IN_MIN(i)); data->in_max[i] = - sis5595_read_value(client, + sis5595_read_value(data, SIS5595_REG_IN_MAX(i)); } for (i = 0; i < 2; i++) { data->fan[i] = - sis5595_read_value(client, SIS5595_REG_FAN(i)); + sis5595_read_value(data, SIS5595_REG_FAN(i)); data->fan_min[i] = - sis5595_read_value(client, + sis5595_read_value(data, SIS5595_REG_FAN_MIN(i)); } if (data->maxins == 3) { data->temp = - sis5595_read_value(client, SIS5595_REG_TEMP); + sis5595_read_value(data, SIS5595_REG_TEMP); data->temp_over = - sis5595_read_value(client, SIS5595_REG_TEMP_OVER); + sis5595_read_value(data, SIS5595_REG_TEMP_OVER); data->temp_hyst = - sis5595_read_value(client, SIS5595_REG_TEMP_HYST); + sis5595_read_value(data, SIS5595_REG_TEMP_HYST); } - i = sis5595_read_value(client, SIS5595_REG_FANDIV); + i = sis5595_read_value(data, SIS5595_REG_FANDIV); data->fan_div[0] = (i >> 4) & 0x03; data->fan_div[1] = i >> 6; data->alarms = - sis5595_read_value(client, SIS5595_REG_ALARM1) | - (sis5595_read_value(client, SIS5595_REG_ALARM2) << 8); + sis5595_read_value(data, SIS5595_REG_ALARM1) | + (sis5595_read_value(data, SIS5595_REG_ALARM2) << 8); data->last_updated = jiffies; data->valid = 1; } @@ -774,10 +692,50 @@ static int blacklist[] __devinitdata = { PCI_DEVICE_ID_SI_5598, 0 }; +static int __devinit sis5595_device_add(unsigned short address) +{ + struct resource res = { + .start = address, + .end = address + SIS5595_EXTENT - 1, + .name = "sis5595", + .flags = IORESOURCE_IO, + }; + int err; + + pdev = platform_device_alloc("sis5595", address); + if (!pdev) { + err = -ENOMEM; + printk(KERN_ERR "sis5595: Device allocation failed\n"); + goto exit; + } + + err = platform_device_add_resources(pdev, &res, 1); + if (err) { + printk(KERN_ERR "sis5595: Device resource addition failed " + "(%d)\n", err); + goto exit_device_put; + } + + err = platform_device_add(pdev); + if (err) { + printk(KERN_ERR "sis5595: Device addition failed (%d)\n", + err); + goto exit_device_put; + } + + return 0; + +exit_device_put: + platform_device_put(pdev); +exit: + return err; +} + static int __devinit sis5595_pci_probe(struct pci_dev *dev, const struct pci_device_id *id) { - u16 val; + u16 address; + u8 enable; int *i; for (i = blacklist; *i != 0; i++) { @@ -790,27 +748,68 @@ static int __devinit sis5595_pci_probe(struct pci_dev *dev, } } + force_addr &= ~(SIS5595_EXTENT - 1); + if (force_addr) { + dev_warn(&dev->dev, "Forcing ISA address 0x%x\n", force_addr); + pci_write_config_word(dev, SIS5595_BASE_REG, force_addr); + } + if (PCIBIOS_SUCCESSFUL != - pci_read_config_word(dev, SIS5595_BASE_REG, &val)) + pci_read_config_word(dev, SIS5595_BASE_REG, &address)) { + dev_err(&dev->dev, "Failed to read ISA address\n"); return -ENODEV; + } - address = val & ~(SIS5595_EXTENT - 1); - if (address == 0 && force_addr == 0) { + address &= ~(SIS5595_EXTENT - 1); + if (!address) { dev_err(&dev->dev, "Base address not set - upgrade BIOS or use force_addr=0xaddr\n"); return -ENODEV; } + if (force_addr && address != force_addr) { + /* doesn't work for some chips? */ + dev_err(&dev->dev, "Failed to force ISA address\n"); + return -ENODEV; + } - s_bridge = pci_dev_get(dev); - if (i2c_isa_add_driver(&sis5595_driver)) { - pci_dev_put(s_bridge); - s_bridge = NULL; + if (PCIBIOS_SUCCESSFUL != + pci_read_config_byte(dev, SIS5595_ENABLE_REG, &enable)) { + dev_err(&dev->dev, "Failed to read enable register\n"); + return -ENODEV; + } + if (!(enable & 0x80)) { + if ((PCIBIOS_SUCCESSFUL != + pci_write_config_byte(dev, SIS5595_ENABLE_REG, + enable | 0x80)) + || (PCIBIOS_SUCCESSFUL != + pci_read_config_byte(dev, SIS5595_ENABLE_REG, &enable)) + || (!(enable & 0x80))) { + /* doesn't work for some chips! */ + dev_err(&dev->dev, "Failed to enable HWM device\n"); + return -ENODEV; + } } + if (platform_driver_register(&sis5595_driver)) { + dev_dbg(&dev->dev, "Failed to register sis5595 driver\n"); + goto exit; + } + + s_bridge = pci_dev_get(dev); + /* Sets global pdev as a side effect */ + if (sis5595_device_add(address)) + goto exit_unregister; + /* Always return failure here. This is to allow other drivers to bind * to this pci device. We don't really want to have control over the * pci device, we only wanted to read as few register values from it. */ return -ENODEV; + +exit_unregister: + pci_dev_put(dev); + platform_driver_unregister(&sis5595_driver); +exit: + return -ENODEV; } static struct pci_driver sis5595_pci_driver = { @@ -828,7 +827,8 @@ static void __exit sm_sis5595_exit(void) { pci_unregister_driver(&sis5595_pci_driver); if (s_bridge != NULL) { - i2c_isa_del_driver(&sis5595_driver); + platform_device_unregister(pdev); + platform_driver_unregister(&sis5595_driver); pci_dev_put(s_bridge); s_bridge = NULL; } diff --git a/drivers/hwmon/smsc47b397.c b/drivers/hwmon/smsc47b397.c index 943abbd95ab..45266b30ce1 100644 --- a/drivers/hwmon/smsc47b397.c +++ b/drivers/hwmon/smsc47b397.c @@ -174,6 +174,8 @@ static SENSOR_DEVICE_ATTR(temp4_input, S_IRUGO, show_temp, NULL, 3); REG: count of 90kHz pulses / revolution */ static int fan_from_reg(u16 reg) { + if (reg == 0 || reg == 0xffff) + return 0; return 90000 * 60 / reg; } @@ -333,7 +335,7 @@ static int __init smsc47b397_find(unsigned short *addr) superio_enter(); id = superio_inb(SUPERIO_REG_DEVID); - if ((id != 0x6f) && (id != 0x81)) { + if ((id != 0x6f) && (id != 0x81) && (id != 0x85)) { superio_exit(); return -ENODEV; } @@ -346,7 +348,8 @@ static int __init smsc47b397_find(unsigned short *addr) printk(KERN_INFO DRVNAME ": found SMSC %s " "(base address 0x%04x, revision %u)\n", - id == 0x81 ? "SCH5307-NS" : "LPC47B397-NC", *addr, rev); + id == 0x81 ? "SCH5307-NS" : id == 0x85 ? "SCH5317" : + "LPC47B397-NC", *addr, rev); superio_exit(); return 0; diff --git a/drivers/hwmon/smsc47m1.c b/drivers/hwmon/smsc47m1.c index 1e21c8cc948..1de2f2be870 100644 --- a/drivers/hwmon/smsc47m1.c +++ b/drivers/hwmon/smsc47m1.c @@ -597,6 +597,7 @@ static int __devinit smsc47m1_probe(struct platform_device *pdev) error_remove_files: sysfs_remove_group(&dev->kobj, &smsc47m1_group); error_free: + platform_set_drvdata(pdev, NULL); kfree(data); error_release: release_region(res->start, SMSC_EXTENT); @@ -608,12 +609,12 @@ static int __devexit smsc47m1_remove(struct platform_device *pdev) struct smsc47m1_data *data = platform_get_drvdata(pdev); struct resource *res; - platform_set_drvdata(pdev, NULL); hwmon_device_unregister(data->class_dev); sysfs_remove_group(&pdev->dev.kobj, &smsc47m1_group); res = platform_get_resource(pdev, IORESOURCE_IO, 0); release_region(res->start, SMSC_EXTENT); + platform_set_drvdata(pdev, NULL); kfree(data); return 0; @@ -693,15 +694,12 @@ static int __init smsc47m1_device_add(unsigned short address, goto exit_device_put; } - pdev->dev.platform_data = kmalloc(sizeof(struct smsc47m1_sio_data), - GFP_KERNEL); - if (!pdev->dev.platform_data) { - err = -ENOMEM; + err = platform_device_add_data(pdev, sio_data, + sizeof(struct smsc47m1_sio_data)); + if (err) { printk(KERN_ERR DRVNAME ": Platform data allocation failed\n"); goto exit_device_put; } - memcpy(pdev->dev.platform_data, sio_data, - sizeof(struct smsc47m1_sio_data)); err = platform_device_add(pdev); if (err) { diff --git a/drivers/hwmon/smsc47m192.c b/drivers/hwmon/smsc47m192.c index a012f396f35..d3a3ba04cb0 100644 --- a/drivers/hwmon/smsc47m192.c +++ b/drivers/hwmon/smsc47m192.c @@ -31,6 +31,7 @@ #include <linux/hwmon-vid.h> #include <linux/err.h> #include <linux/sysfs.h> +#include <linux/mutex.h> /* Addresses to scan */ static unsigned short normal_i2c[] = { 0x2c, 0x2d, I2C_CLIENT_END }; @@ -97,7 +98,7 @@ static inline int TEMP_FROM_REG(s8 val) struct smsc47m192_data { struct i2c_client client; struct class_device *class_dev; - struct semaphore update_lock; + struct mutex update_lock; char valid; /* !=0 if following fields are valid */ unsigned long last_updated; /* In jiffies */ @@ -164,11 +165,11 @@ static ssize_t set_in_min(struct device *dev, struct device_attribute *attr, struct smsc47m192_data *data = i2c_get_clientdata(client); unsigned long val = simple_strtoul(buf, NULL, 10); - down(&data->update_lock); + mutex_lock(&data->update_lock); data->in_min[nr] = IN_TO_REG(val, nr); i2c_smbus_write_byte_data(client, SMSC47M192_REG_IN_MIN(nr), data->in_min[nr]); - up(&data->update_lock); + mutex_unlock(&data->update_lock); return count; } @@ -181,11 +182,11 @@ static ssize_t set_in_max(struct device *dev, struct device_attribute *attr, struct smsc47m192_data *data = i2c_get_clientdata(client); unsigned long val = simple_strtoul(buf, NULL, 10); - down(&data->update_lock); + mutex_lock(&data->update_lock); data->in_max[nr] = IN_TO_REG(val, nr); i2c_smbus_write_byte_data(client, SMSC47M192_REG_IN_MAX(nr), data->in_max[nr]); - up(&data->update_lock); + mutex_unlock(&data->update_lock); return count; } @@ -243,11 +244,11 @@ static ssize_t set_temp_min(struct device *dev, struct device_attribute *attr, struct smsc47m192_data *data = i2c_get_clientdata(client); long val = simple_strtol(buf, NULL, 10); - down(&data->update_lock); + mutex_lock(&data->update_lock); data->temp_min[nr] = TEMP_TO_REG(val); i2c_smbus_write_byte_data(client, SMSC47M192_REG_TEMP_MIN[nr], data->temp_min[nr]); - up(&data->update_lock); + mutex_unlock(&data->update_lock); return count; } @@ -260,11 +261,11 @@ static ssize_t set_temp_max(struct device *dev, struct device_attribute *attr, struct smsc47m192_data *data = i2c_get_clientdata(client); long val = simple_strtol(buf, NULL, 10); - down(&data->update_lock); + mutex_lock(&data->update_lock); data->temp_max[nr] = TEMP_TO_REG(val); i2c_smbus_write_byte_data(client, SMSC47M192_REG_TEMP_MAX[nr], data->temp_max[nr]); - up(&data->update_lock); + mutex_unlock(&data->update_lock); return count; } @@ -287,7 +288,7 @@ static ssize_t set_temp_offset(struct device *dev, struct device_attribute u8 sfr = i2c_smbus_read_byte_data(client, SMSC47M192_REG_SFR); long val = simple_strtol(buf, NULL, 10); - down(&data->update_lock); + mutex_lock(&data->update_lock); data->temp_offset[nr] = TEMP_TO_REG(val); if (nr>1) i2c_smbus_write_byte_data(client, @@ -303,7 +304,7 @@ static ssize_t set_temp_offset(struct device *dev, struct device_attribute } else if ((sfr & 0x10) == (nr==0 ? 0x10 : 0)) i2c_smbus_write_byte_data(client, SMSC47M192_REG_TEMP_OFFSET(nr), 0); - up(&data->update_lock); + mutex_unlock(&data->update_lock); return count; } @@ -360,8 +361,8 @@ static ssize_t show_alarm(struct device *dev, struct device_attribute *attr, static SENSOR_DEVICE_ATTR(temp1_alarm, S_IRUGO, show_alarm, NULL, 0x0010); static SENSOR_DEVICE_ATTR(temp2_alarm, S_IRUGO, show_alarm, NULL, 0x0020); static SENSOR_DEVICE_ATTR(temp3_alarm, S_IRUGO, show_alarm, NULL, 0x0040); -static SENSOR_DEVICE_ATTR(temp2_input_fault, S_IRUGO, show_alarm, NULL, 0x4000); -static SENSOR_DEVICE_ATTR(temp3_input_fault, S_IRUGO, show_alarm, NULL, 0x8000); +static SENSOR_DEVICE_ATTR(temp2_fault, S_IRUGO, show_alarm, NULL, 0x4000); +static SENSOR_DEVICE_ATTR(temp3_fault, S_IRUGO, show_alarm, NULL, 0x8000); static SENSOR_DEVICE_ATTR(in0_alarm, S_IRUGO, show_alarm, NULL, 0x0001); static SENSOR_DEVICE_ATTR(in1_alarm, S_IRUGO, show_alarm, NULL, 0x0002); static SENSOR_DEVICE_ATTR(in2_alarm, S_IRUGO, show_alarm, NULL, 0x0004); @@ -411,13 +412,13 @@ static struct attribute *smsc47m192_attributes[] = { &sensor_dev_attr_temp2_min.dev_attr.attr, &sensor_dev_attr_temp2_offset.dev_attr.attr, &sensor_dev_attr_temp2_alarm.dev_attr.attr, - &sensor_dev_attr_temp2_input_fault.dev_attr.attr, + &sensor_dev_attr_temp2_fault.dev_attr.attr, &sensor_dev_attr_temp3_input.dev_attr.attr, &sensor_dev_attr_temp3_max.dev_attr.attr, &sensor_dev_attr_temp3_min.dev_attr.attr, &sensor_dev_attr_temp3_offset.dev_attr.attr, &sensor_dev_attr_temp3_alarm.dev_attr.attr, - &sensor_dev_attr_temp3_input_fault.dev_attr.attr, + &sensor_dev_attr_temp3_fault.dev_attr.attr, &dev_attr_cpu0_vid.attr, &dev_attr_vrm.attr, @@ -531,7 +532,7 @@ static int smsc47m192_detect(struct i2c_adapter *adapter, int address, /* Fill in the remaining client fields and put into the global list */ strlcpy(client->name, "smsc47m192", I2C_NAME_SIZE); data->vrm = vid_which_vrm(); - init_MUTEX(&data->update_lock); + mutex_init(&data->update_lock); /* Tell the I2C layer a new client has arrived */ if ((err = i2c_attach_client(client))) @@ -594,7 +595,7 @@ static struct smsc47m192_data *smsc47m192_update_device(struct device *dev) struct smsc47m192_data *data = i2c_get_clientdata(client); int i, config; - down(&data->update_lock); + mutex_lock(&data->update_lock); if (time_after(jiffies, data->last_updated + HZ + HZ / 2) || !data->valid) { @@ -645,7 +646,7 @@ static struct smsc47m192_data *smsc47m192_update_device(struct device *dev) data->valid = 1; } - up(&data->update_lock); + mutex_unlock(&data->update_lock); return data; } diff --git a/drivers/hwmon/via686a.c b/drivers/hwmon/via686a.c index 9a440c8cc52..24a6851491d 100644 --- a/drivers/hwmon/via686a.c +++ b/drivers/hwmon/via686a.c @@ -34,9 +34,9 @@ #include <linux/slab.h> #include <linux/pci.h> #include <linux/jiffies.h> -#include <linux/i2c.h> -#include <linux/i2c-isa.h> +#include <linux/platform_device.h> #include <linux/hwmon.h> +#include <linux/hwmon-sysfs.h> #include <linux/err.h> #include <linux/init.h> #include <linux/mutex.h> @@ -51,10 +51,7 @@ module_param(force_addr, ushort, 0); MODULE_PARM_DESC(force_addr, "Initialize the base address of the sensors"); -/* Device address - Note that we can't determine the ISA address until we have initialized - our module */ -static unsigned short address; +static struct platform_device *pdev; /* The Via 686a southbridge has a LM78-like chip integrated on the same IC. @@ -295,7 +292,8 @@ static inline long TEMP_FROM_REG10(u16 val) /* For each registered chip, we need to keep some data in memory. The structure is dynamically allocated. */ struct via686a_data { - struct i2c_client client; + unsigned short addr; + const char *name; struct class_device *class_dev; struct mutex update_lock; char valid; /* !=0 if following fields are valid */ @@ -315,98 +313,85 @@ struct via686a_data { static struct pci_dev *s_bridge; /* pointer to the (only) via686a */ -static int via686a_detect(struct i2c_adapter *adapter); -static int via686a_detach_client(struct i2c_client *client); +static int via686a_probe(struct platform_device *pdev); +static int via686a_remove(struct platform_device *pdev); -static inline int via686a_read_value(struct i2c_client *client, u8 reg) +static inline int via686a_read_value(struct via686a_data *data, u8 reg) { - return (inb_p(client->addr + reg)); + return inb_p(data->addr + reg); } -static inline void via686a_write_value(struct i2c_client *client, u8 reg, +static inline void via686a_write_value(struct via686a_data *data, u8 reg, u8 value) { - outb_p(value, client->addr + reg); + outb_p(value, data->addr + reg); } static struct via686a_data *via686a_update_device(struct device *dev); -static void via686a_init_client(struct i2c_client *client); +static void via686a_init_device(struct via686a_data *data); /* following are the sysfs callback functions */ /* 7 voltage sensors */ -static ssize_t show_in(struct device *dev, char *buf, int nr) { +static ssize_t show_in(struct device *dev, struct device_attribute *da, + char *buf) { struct via686a_data *data = via686a_update_device(dev); + struct sensor_device_attribute *attr = to_sensor_dev_attr(da); + int nr = attr->index; return sprintf(buf, "%ld\n", IN_FROM_REG(data->in[nr], nr)); } -static ssize_t show_in_min(struct device *dev, char *buf, int nr) { +static ssize_t show_in_min(struct device *dev, struct device_attribute *da, + char *buf) { struct via686a_data *data = via686a_update_device(dev); + struct sensor_device_attribute *attr = to_sensor_dev_attr(da); + int nr = attr->index; return sprintf(buf, "%ld\n", IN_FROM_REG(data->in_min[nr], nr)); } -static ssize_t show_in_max(struct device *dev, char *buf, int nr) { +static ssize_t show_in_max(struct device *dev, struct device_attribute *da, + char *buf) { struct via686a_data *data = via686a_update_device(dev); + struct sensor_device_attribute *attr = to_sensor_dev_attr(da); + int nr = attr->index; return sprintf(buf, "%ld\n", IN_FROM_REG(data->in_max[nr], nr)); } -static ssize_t set_in_min(struct device *dev, const char *buf, - size_t count, int nr) { - struct i2c_client *client = to_i2c_client(dev); - struct via686a_data *data = i2c_get_clientdata(client); +static ssize_t set_in_min(struct device *dev, struct device_attribute *da, + const char *buf, size_t count) { + struct via686a_data *data = dev_get_drvdata(dev); + struct sensor_device_attribute *attr = to_sensor_dev_attr(da); + int nr = attr->index; unsigned long val = simple_strtoul(buf, NULL, 10); mutex_lock(&data->update_lock); data->in_min[nr] = IN_TO_REG(val, nr); - via686a_write_value(client, VIA686A_REG_IN_MIN(nr), + via686a_write_value(data, VIA686A_REG_IN_MIN(nr), data->in_min[nr]); mutex_unlock(&data->update_lock); return count; } -static ssize_t set_in_max(struct device *dev, const char *buf, - size_t count, int nr) { - struct i2c_client *client = to_i2c_client(dev); - struct via686a_data *data = i2c_get_clientdata(client); +static ssize_t set_in_max(struct device *dev, struct device_attribute *da, + const char *buf, size_t count) { + struct via686a_data *data = dev_get_drvdata(dev); + struct sensor_device_attribute *attr = to_sensor_dev_attr(da); + int nr = attr->index; unsigned long val = simple_strtoul(buf, NULL, 10); mutex_lock(&data->update_lock); data->in_max[nr] = IN_TO_REG(val, nr); - via686a_write_value(client, VIA686A_REG_IN_MAX(nr), + via686a_write_value(data, VIA686A_REG_IN_MAX(nr), data->in_max[nr]); mutex_unlock(&data->update_lock); return count; } #define show_in_offset(offset) \ -static ssize_t \ - show_in##offset (struct device *dev, struct device_attribute *attr, char *buf) \ -{ \ - return show_in(dev, buf, offset); \ -} \ -static ssize_t \ - show_in##offset##_min (struct device *dev, struct device_attribute *attr, char *buf) \ -{ \ - return show_in_min(dev, buf, offset); \ -} \ -static ssize_t \ - show_in##offset##_max (struct device *dev, struct device_attribute *attr, char *buf) \ -{ \ - return show_in_max(dev, buf, offset); \ -} \ -static ssize_t set_in##offset##_min (struct device *dev, struct device_attribute *attr, \ - const char *buf, size_t count) \ -{ \ - return set_in_min(dev, buf, count, offset); \ -} \ -static ssize_t set_in##offset##_max (struct device *dev, struct device_attribute *attr, \ - const char *buf, size_t count) \ -{ \ - return set_in_max(dev, buf, count, offset); \ -} \ -static DEVICE_ATTR(in##offset##_input, S_IRUGO, show_in##offset, NULL);\ -static DEVICE_ATTR(in##offset##_min, S_IRUGO | S_IWUSR, \ - show_in##offset##_min, set_in##offset##_min); \ -static DEVICE_ATTR(in##offset##_max, S_IRUGO | S_IWUSR, \ - show_in##offset##_max, set_in##offset##_max); +static SENSOR_DEVICE_ATTR(in##offset##_input, S_IRUGO, \ + show_in, NULL, offset); \ +static SENSOR_DEVICE_ATTR(in##offset##_min, S_IRUGO | S_IWUSR, \ + show_in_min, set_in_min, offset); \ +static SENSOR_DEVICE_ATTR(in##offset##_max, S_IRUGO | S_IWUSR, \ + show_in_max, set_in_max, offset); show_in_offset(0); show_in_offset(1); @@ -415,150 +400,128 @@ show_in_offset(3); show_in_offset(4); /* 3 temperatures */ -static ssize_t show_temp(struct device *dev, char *buf, int nr) { +static ssize_t show_temp(struct device *dev, struct device_attribute *da, + char *buf) { struct via686a_data *data = via686a_update_device(dev); + struct sensor_device_attribute *attr = to_sensor_dev_attr(da); + int nr = attr->index; return sprintf(buf, "%ld\n", TEMP_FROM_REG10(data->temp[nr])); } -static ssize_t show_temp_over(struct device *dev, char *buf, int nr) { +static ssize_t show_temp_over(struct device *dev, struct device_attribute *da, + char *buf) { struct via686a_data *data = via686a_update_device(dev); + struct sensor_device_attribute *attr = to_sensor_dev_attr(da); + int nr = attr->index; return sprintf(buf, "%ld\n", TEMP_FROM_REG(data->temp_over[nr])); } -static ssize_t show_temp_hyst(struct device *dev, char *buf, int nr) { +static ssize_t show_temp_hyst(struct device *dev, struct device_attribute *da, + char *buf) { struct via686a_data *data = via686a_update_device(dev); + struct sensor_device_attribute *attr = to_sensor_dev_attr(da); + int nr = attr->index; return sprintf(buf, "%ld\n", TEMP_FROM_REG(data->temp_hyst[nr])); } -static ssize_t set_temp_over(struct device *dev, const char *buf, - size_t count, int nr) { - struct i2c_client *client = to_i2c_client(dev); - struct via686a_data *data = i2c_get_clientdata(client); +static ssize_t set_temp_over(struct device *dev, struct device_attribute *da, + const char *buf, size_t count) { + struct via686a_data *data = dev_get_drvdata(dev); + struct sensor_device_attribute *attr = to_sensor_dev_attr(da); + int nr = attr->index; int val = simple_strtol(buf, NULL, 10); mutex_lock(&data->update_lock); data->temp_over[nr] = TEMP_TO_REG(val); - via686a_write_value(client, VIA686A_REG_TEMP_OVER[nr], + via686a_write_value(data, VIA686A_REG_TEMP_OVER[nr], data->temp_over[nr]); mutex_unlock(&data->update_lock); return count; } -static ssize_t set_temp_hyst(struct device *dev, const char *buf, - size_t count, int nr) { - struct i2c_client *client = to_i2c_client(dev); - struct via686a_data *data = i2c_get_clientdata(client); +static ssize_t set_temp_hyst(struct device *dev, struct device_attribute *da, + const char *buf, size_t count) { + struct via686a_data *data = dev_get_drvdata(dev); + struct sensor_device_attribute *attr = to_sensor_dev_attr(da); + int nr = attr->index; int val = simple_strtol(buf, NULL, 10); mutex_lock(&data->update_lock); data->temp_hyst[nr] = TEMP_TO_REG(val); - via686a_write_value(client, VIA686A_REG_TEMP_HYST[nr], + via686a_write_value(data, VIA686A_REG_TEMP_HYST[nr], data->temp_hyst[nr]); mutex_unlock(&data->update_lock); return count; } #define show_temp_offset(offset) \ -static ssize_t show_temp_##offset (struct device *dev, struct device_attribute *attr, char *buf) \ -{ \ - return show_temp(dev, buf, offset - 1); \ -} \ -static ssize_t \ -show_temp_##offset##_over (struct device *dev, struct device_attribute *attr, char *buf) \ -{ \ - return show_temp_over(dev, buf, offset - 1); \ -} \ -static ssize_t \ -show_temp_##offset##_hyst (struct device *dev, struct device_attribute *attr, char *buf) \ -{ \ - return show_temp_hyst(dev, buf, offset - 1); \ -} \ -static ssize_t set_temp_##offset##_over (struct device *dev, struct device_attribute *attr, \ - const char *buf, size_t count) \ -{ \ - return set_temp_over(dev, buf, count, offset - 1); \ -} \ -static ssize_t set_temp_##offset##_hyst (struct device *dev, struct device_attribute *attr, \ - const char *buf, size_t count) \ -{ \ - return set_temp_hyst(dev, buf, count, offset - 1); \ -} \ -static DEVICE_ATTR(temp##offset##_input, S_IRUGO, show_temp_##offset, NULL);\ -static DEVICE_ATTR(temp##offset##_max, S_IRUGO | S_IWUSR, \ - show_temp_##offset##_over, set_temp_##offset##_over); \ -static DEVICE_ATTR(temp##offset##_max_hyst, S_IRUGO | S_IWUSR, \ - show_temp_##offset##_hyst, set_temp_##offset##_hyst); +static SENSOR_DEVICE_ATTR(temp##offset##_input, S_IRUGO, \ + show_temp, NULL, offset - 1); \ +static SENSOR_DEVICE_ATTR(temp##offset##_max, S_IRUGO | S_IWUSR, \ + show_temp_over, set_temp_over, offset - 1); \ +static SENSOR_DEVICE_ATTR(temp##offset##_max_hyst, S_IRUGO | S_IWUSR, \ + show_temp_hyst, set_temp_hyst, offset - 1); show_temp_offset(1); show_temp_offset(2); show_temp_offset(3); /* 2 Fans */ -static ssize_t show_fan(struct device *dev, char *buf, int nr) { +static ssize_t show_fan(struct device *dev, struct device_attribute *da, + char *buf) { struct via686a_data *data = via686a_update_device(dev); + struct sensor_device_attribute *attr = to_sensor_dev_attr(da); + int nr = attr->index; return sprintf(buf, "%d\n", FAN_FROM_REG(data->fan[nr], DIV_FROM_REG(data->fan_div[nr])) ); } -static ssize_t show_fan_min(struct device *dev, char *buf, int nr) { +static ssize_t show_fan_min(struct device *dev, struct device_attribute *da, + char *buf) { struct via686a_data *data = via686a_update_device(dev); + struct sensor_device_attribute *attr = to_sensor_dev_attr(da); + int nr = attr->index; return sprintf(buf, "%d\n", FAN_FROM_REG(data->fan_min[nr], DIV_FROM_REG(data->fan_div[nr])) ); } -static ssize_t show_fan_div(struct device *dev, char *buf, int nr) { +static ssize_t show_fan_div(struct device *dev, struct device_attribute *da, + char *buf) { struct via686a_data *data = via686a_update_device(dev); + struct sensor_device_attribute *attr = to_sensor_dev_attr(da); + int nr = attr->index; return sprintf(buf, "%d\n", DIV_FROM_REG(data->fan_div[nr]) ); } -static ssize_t set_fan_min(struct device *dev, const char *buf, - size_t count, int nr) { - struct i2c_client *client = to_i2c_client(dev); - struct via686a_data *data = i2c_get_clientdata(client); +static ssize_t set_fan_min(struct device *dev, struct device_attribute *da, + const char *buf, size_t count) { + struct via686a_data *data = dev_get_drvdata(dev); + struct sensor_device_attribute *attr = to_sensor_dev_attr(da); + int nr = attr->index; int val = simple_strtol(buf, NULL, 10); mutex_lock(&data->update_lock); data->fan_min[nr] = FAN_TO_REG(val, DIV_FROM_REG(data->fan_div[nr])); - via686a_write_value(client, VIA686A_REG_FAN_MIN(nr+1), data->fan_min[nr]); + via686a_write_value(data, VIA686A_REG_FAN_MIN(nr+1), data->fan_min[nr]); mutex_unlock(&data->update_lock); return count; } -static ssize_t set_fan_div(struct device *dev, const char *buf, - size_t count, int nr) { - struct i2c_client *client = to_i2c_client(dev); - struct via686a_data *data = i2c_get_clientdata(client); +static ssize_t set_fan_div(struct device *dev, struct device_attribute *da, + const char *buf, size_t count) { + struct via686a_data *data = dev_get_drvdata(dev); + struct sensor_device_attribute *attr = to_sensor_dev_attr(da); + int nr = attr->index; int val = simple_strtol(buf, NULL, 10); int old; mutex_lock(&data->update_lock); - old = via686a_read_value(client, VIA686A_REG_FANDIV); + old = via686a_read_value(data, VIA686A_REG_FANDIV); data->fan_div[nr] = DIV_TO_REG(val); old = (old & 0x0f) | (data->fan_div[1] << 6) | (data->fan_div[0] << 4); - via686a_write_value(client, VIA686A_REG_FANDIV, old); + via686a_write_value(data, VIA686A_REG_FANDIV, old); mutex_unlock(&data->update_lock); return count; } #define show_fan_offset(offset) \ -static ssize_t show_fan_##offset (struct device *dev, struct device_attribute *attr, char *buf) \ -{ \ - return show_fan(dev, buf, offset - 1); \ -} \ -static ssize_t show_fan_##offset##_min (struct device *dev, struct device_attribute *attr, char *buf) \ -{ \ - return show_fan_min(dev, buf, offset - 1); \ -} \ -static ssize_t show_fan_##offset##_div (struct device *dev, struct device_attribute *attr, char *buf) \ -{ \ - return show_fan_div(dev, buf, offset - 1); \ -} \ -static ssize_t set_fan_##offset##_min (struct device *dev, struct device_attribute *attr, \ - const char *buf, size_t count) \ -{ \ - return set_fan_min(dev, buf, count, offset - 1); \ -} \ -static ssize_t set_fan_##offset##_div (struct device *dev, struct device_attribute *attr, \ - const char *buf, size_t count) \ -{ \ - return set_fan_div(dev, buf, count, offset - 1); \ -} \ -static DEVICE_ATTR(fan##offset##_input, S_IRUGO, show_fan_##offset, NULL);\ -static DEVICE_ATTR(fan##offset##_min, S_IRUGO | S_IWUSR, \ - show_fan_##offset##_min, set_fan_##offset##_min); \ -static DEVICE_ATTR(fan##offset##_div, S_IRUGO | S_IWUSR, \ - show_fan_##offset##_div, set_fan_##offset##_div); +static SENSOR_DEVICE_ATTR(fan##offset##_input, S_IRUGO, \ + show_fan, NULL, offset - 1); \ +static SENSOR_DEVICE_ATTR(fan##offset##_min, S_IRUGO | S_IWUSR, \ + show_fan_min, set_fan_min, offset - 1); \ +static SENSOR_DEVICE_ATTR(fan##offset##_div, S_IRUGO | S_IWUSR, \ + show_fan_div, set_fan_div, offset - 1); show_fan_offset(1); show_fan_offset(2); @@ -570,41 +533,50 @@ static ssize_t show_alarms(struct device *dev, struct device_attribute *attr, ch } static DEVICE_ATTR(alarms, S_IRUGO, show_alarms, NULL); +static ssize_t show_name(struct device *dev, struct device_attribute + *devattr, char *buf) +{ + struct via686a_data *data = dev_get_drvdata(dev); + return sprintf(buf, "%s\n", data->name); +} +static DEVICE_ATTR(name, S_IRUGO, show_name, NULL); + static struct attribute *via686a_attributes[] = { - &dev_attr_in0_input.attr, - &dev_attr_in1_input.attr, - &dev_attr_in2_input.attr, - &dev_attr_in3_input.attr, - &dev_attr_in4_input.attr, - &dev_attr_in0_min.attr, - &dev_attr_in1_min.attr, - &dev_attr_in2_min.attr, - &dev_attr_in3_min.attr, - &dev_attr_in4_min.attr, - &dev_attr_in0_max.attr, - &dev_attr_in1_max.attr, - &dev_attr_in2_max.attr, - &dev_attr_in3_max.attr, - &dev_attr_in4_max.attr, - - &dev_attr_temp1_input.attr, - &dev_attr_temp2_input.attr, - &dev_attr_temp3_input.attr, - &dev_attr_temp1_max.attr, - &dev_attr_temp2_max.attr, - &dev_attr_temp3_max.attr, - &dev_attr_temp1_max_hyst.attr, - &dev_attr_temp2_max_hyst.attr, - &dev_attr_temp3_max_hyst.attr, - - &dev_attr_fan1_input.attr, - &dev_attr_fan2_input.attr, - &dev_attr_fan1_min.attr, - &dev_attr_fan2_min.attr, - &dev_attr_fan1_div.attr, - &dev_attr_fan2_div.attr, + &sensor_dev_attr_in0_input.dev_attr.attr, + &sensor_dev_attr_in1_input.dev_attr.attr, + &sensor_dev_attr_in2_input.dev_attr.attr, + &sensor_dev_attr_in3_input.dev_attr.attr, + &sensor_dev_attr_in4_input.dev_attr.attr, + &sensor_dev_attr_in0_min.dev_attr.attr, + &sensor_dev_attr_in1_min.dev_attr.attr, + &sensor_dev_attr_in2_min.dev_attr.attr, + &sensor_dev_attr_in3_min.dev_attr.attr, + &sensor_dev_attr_in4_min.dev_attr.attr, + &sensor_dev_attr_in0_max.dev_attr.attr, + &sensor_dev_attr_in1_max.dev_attr.attr, + &sensor_dev_attr_in2_max.dev_attr.attr, + &sensor_dev_attr_in3_max.dev_attr.attr, + &sensor_dev_attr_in4_max.dev_attr.attr, + + &sensor_dev_attr_temp1_input.dev_attr.attr, + &sensor_dev_attr_temp2_input.dev_attr.attr, + &sensor_dev_attr_temp3_input.dev_attr.attr, + &sensor_dev_attr_temp1_max.dev_attr.attr, + &sensor_dev_attr_temp2_max.dev_attr.attr, + &sensor_dev_attr_temp3_max.dev_attr.attr, + &sensor_dev_attr_temp1_max_hyst.dev_attr.attr, + &sensor_dev_attr_temp2_max_hyst.dev_attr.attr, + &sensor_dev_attr_temp3_max_hyst.dev_attr.attr, + + &sensor_dev_attr_fan1_input.dev_attr.attr, + &sensor_dev_attr_fan2_input.dev_attr.attr, + &sensor_dev_attr_fan1_min.dev_attr.attr, + &sensor_dev_attr_fan2_min.dev_attr.attr, + &sensor_dev_attr_fan1_div.dev_attr.attr, + &sensor_dev_attr_fan2_div.dev_attr.attr, &dev_attr_alarms.attr, + &dev_attr_name.attr, NULL }; @@ -612,58 +584,29 @@ static const struct attribute_group via686a_group = { .attrs = via686a_attributes, }; -/* The driver. I choose to use type i2c_driver, as at is identical to both - smbus_driver and isa_driver, and clients could be of either kind */ -static struct i2c_driver via686a_driver = { +static struct platform_driver via686a_driver = { .driver = { .owner = THIS_MODULE, .name = "via686a", }, - .attach_adapter = via686a_detect, - .detach_client = via686a_detach_client, + .probe = via686a_probe, + .remove = __devexit_p(via686a_remove), }; /* This is called when the module is loaded */ -static int via686a_detect(struct i2c_adapter *adapter) +static int __devinit via686a_probe(struct platform_device *pdev) { - struct i2c_client *new_client; struct via686a_data *data; - int err = 0; - const char client_name[] = "via686a"; - u16 val; - - /* 8231 requires multiple of 256, we enforce that on 686 as well */ - if (force_addr) { - address = force_addr & 0xFF00; - dev_warn(&adapter->dev, "forcing ISA address 0x%04X\n", - address); - if (PCIBIOS_SUCCESSFUL != - pci_write_config_word(s_bridge, VIA686A_BASE_REG, address)) - return -ENODEV; - } - if (PCIBIOS_SUCCESSFUL != - pci_read_config_word(s_bridge, VIA686A_ENABLE_REG, &val)) - return -ENODEV; - if (!(val & 0x0001)) { - if (force_addr) { - dev_info(&adapter->dev, "enabling sensors\n"); - if (PCIBIOS_SUCCESSFUL != - pci_write_config_word(s_bridge, VIA686A_ENABLE_REG, - val | 0x0001)) - return -ENODEV; - } else { - dev_warn(&adapter->dev, "sensors disabled - enable " - "with force_addr=0x%x\n", address); - return -ENODEV; - } - } + struct resource *res; + int err; /* Reserve the ISA region */ - if (!request_region(address, VIA686A_EXTENT, + res = platform_get_resource(pdev, IORESOURCE_IO, 0); + if (!request_region(res->start, VIA686A_EXTENT, via686a_driver.driver.name)) { - dev_err(&adapter->dev, "region 0x%x already in use!\n", - address); + dev_err(&pdev->dev, "Region 0x%lx-0x%lx already in use!\n", + (unsigned long)res->start, (unsigned long)res->end); return -ENODEV; } @@ -672,30 +615,19 @@ static int via686a_detect(struct i2c_adapter *adapter) goto exit_release; } - new_client = &data->client; - i2c_set_clientdata(new_client, data); - new_client->addr = address; - new_client->adapter = adapter; - new_client->driver = &via686a_driver; - new_client->flags = 0; - - /* Fill in the remaining client fields and put into the global list */ - strlcpy(new_client->name, client_name, I2C_NAME_SIZE); - - data->valid = 0; + platform_set_drvdata(pdev, data); + data->addr = res->start; + data->name = "via686a"; mutex_init(&data->update_lock); - /* Tell the I2C layer a new client has arrived */ - if ((err = i2c_attach_client(new_client))) - goto exit_free; /* Initialize the VIA686A chip */ - via686a_init_client(new_client); + via686a_init_device(data); /* Register sysfs hooks */ - if ((err = sysfs_create_group(&new_client->dev.kobj, &via686a_group))) - goto exit_detach; + if ((err = sysfs_create_group(&pdev->dev.kobj, &via686a_group))) + goto exit_free; - data->class_dev = hwmon_device_register(&new_client->dev); + data->class_dev = hwmon_device_register(&pdev->dev); if (IS_ERR(data->class_dev)) { err = PTR_ERR(data->class_dev); goto exit_remove_files; @@ -704,51 +636,46 @@ static int via686a_detect(struct i2c_adapter *adapter) return 0; exit_remove_files: - sysfs_remove_group(&new_client->dev.kobj, &via686a_group); -exit_detach: - i2c_detach_client(new_client); + sysfs_remove_group(&pdev->dev.kobj, &via686a_group); exit_free: kfree(data); exit_release: - release_region(address, VIA686A_EXTENT); + release_region(res->start, VIA686A_EXTENT); return err; } -static int via686a_detach_client(struct i2c_client *client) +static int __devexit via686a_remove(struct platform_device *pdev) { - struct via686a_data *data = i2c_get_clientdata(client); - int err; + struct via686a_data *data = platform_get_drvdata(pdev); hwmon_device_unregister(data->class_dev); - sysfs_remove_group(&client->dev.kobj, &via686a_group); + sysfs_remove_group(&pdev->dev.kobj, &via686a_group); - if ((err = i2c_detach_client(client))) - return err; - - release_region(client->addr, VIA686A_EXTENT); + release_region(data->addr, VIA686A_EXTENT); + platform_set_drvdata(pdev, NULL); kfree(data); return 0; } -static void via686a_init_client(struct i2c_client *client) +static void __devinit via686a_init_device(struct via686a_data *data) { u8 reg; /* Start monitoring */ - reg = via686a_read_value(client, VIA686A_REG_CONFIG); - via686a_write_value(client, VIA686A_REG_CONFIG, (reg|0x01)&0x7F); + reg = via686a_read_value(data, VIA686A_REG_CONFIG); + via686a_write_value(data, VIA686A_REG_CONFIG, (reg | 0x01) & 0x7F); /* Configure temp interrupt mode for continuous-interrupt operation */ - via686a_write_value(client, VIA686A_REG_TEMP_MODE, - via686a_read_value(client, VIA686A_REG_TEMP_MODE) & - !(VIA686A_TEMP_MODE_MASK | VIA686A_TEMP_MODE_CONTINUOUS)); + reg = via686a_read_value(data, VIA686A_REG_TEMP_MODE); + via686a_write_value(data, VIA686A_REG_TEMP_MODE, + (reg & ~VIA686A_TEMP_MODE_MASK) + | VIA686A_TEMP_MODE_CONTINUOUS); } static struct via686a_data *via686a_update_device(struct device *dev) { - struct i2c_client *client = to_i2c_client(dev); - struct via686a_data *data = i2c_get_clientdata(client); + struct via686a_data *data = dev_get_drvdata(dev); int i; mutex_lock(&data->update_lock); @@ -757,27 +684,27 @@ static struct via686a_data *via686a_update_device(struct device *dev) || !data->valid) { for (i = 0; i <= 4; i++) { data->in[i] = - via686a_read_value(client, VIA686A_REG_IN(i)); - data->in_min[i] = via686a_read_value(client, + via686a_read_value(data, VIA686A_REG_IN(i)); + data->in_min[i] = via686a_read_value(data, VIA686A_REG_IN_MIN (i)); data->in_max[i] = - via686a_read_value(client, VIA686A_REG_IN_MAX(i)); + via686a_read_value(data, VIA686A_REG_IN_MAX(i)); } for (i = 1; i <= 2; i++) { data->fan[i - 1] = - via686a_read_value(client, VIA686A_REG_FAN(i)); - data->fan_min[i - 1] = via686a_read_value(client, + via686a_read_value(data, VIA686A_REG_FAN(i)); + data->fan_min[i - 1] = via686a_read_value(data, VIA686A_REG_FAN_MIN(i)); } for (i = 0; i <= 2; i++) { - data->temp[i] = via686a_read_value(client, + data->temp[i] = via686a_read_value(data, VIA686A_REG_TEMP[i]) << 2; data->temp_over[i] = - via686a_read_value(client, + via686a_read_value(data, VIA686A_REG_TEMP_OVER[i]); data->temp_hyst[i] = - via686a_read_value(client, + via686a_read_value(data, VIA686A_REG_TEMP_HYST[i]); } /* add in lower 2 bits @@ -785,23 +712,23 @@ static struct via686a_data *via686a_update_device(struct device *dev) temp2 uses bits 5-4 of VIA686A_REG_TEMP_LOW23 temp3 uses bits 7-6 of VIA686A_REG_TEMP_LOW23 */ - data->temp[0] |= (via686a_read_value(client, + data->temp[0] |= (via686a_read_value(data, VIA686A_REG_TEMP_LOW1) & 0xc0) >> 6; data->temp[1] |= - (via686a_read_value(client, VIA686A_REG_TEMP_LOW23) & + (via686a_read_value(data, VIA686A_REG_TEMP_LOW23) & 0x30) >> 4; data->temp[2] |= - (via686a_read_value(client, VIA686A_REG_TEMP_LOW23) & + (via686a_read_value(data, VIA686A_REG_TEMP_LOW23) & 0xc0) >> 6; - i = via686a_read_value(client, VIA686A_REG_FANDIV); + i = via686a_read_value(data, VIA686A_REG_FANDIV); data->fan_div[0] = (i >> 4) & 0x03; data->fan_div[1] = i >> 6; data->alarms = - via686a_read_value(client, + via686a_read_value(data, VIA686A_REG_ALARM1) | - (via686a_read_value(client, VIA686A_REG_ALARM2) << 8); + (via686a_read_value(data, VIA686A_REG_ALARM2) << 8); data->last_updated = jiffies; data->valid = 1; } @@ -818,32 +745,102 @@ static struct pci_device_id via686a_pci_ids[] = { MODULE_DEVICE_TABLE(pci, via686a_pci_ids); +static int __devinit via686a_device_add(unsigned short address) +{ + struct resource res = { + .start = address, + .end = address + VIA686A_EXTENT - 1, + .name = "via686a", + .flags = IORESOURCE_IO, + }; + int err; + + pdev = platform_device_alloc("via686a", address); + if (!pdev) { + err = -ENOMEM; + printk(KERN_ERR "via686a: Device allocation failed\n"); + goto exit; + } + + err = platform_device_add_resources(pdev, &res, 1); + if (err) { + printk(KERN_ERR "via686a: Device resource addition failed " + "(%d)\n", err); + goto exit_device_put; + } + + err = platform_device_add(pdev); + if (err) { + printk(KERN_ERR "via686a: Device addition failed (%d)\n", + err); + goto exit_device_put; + } + + return 0; + +exit_device_put: + platform_device_put(pdev); +exit: + return err; +} + static int __devinit via686a_pci_probe(struct pci_dev *dev, const struct pci_device_id *id) { - u16 val; + u16 address, val; + if (force_addr) { + address = force_addr & ~(VIA686A_EXTENT - 1); + dev_warn(&dev->dev, "Forcing ISA address 0x%x\n", address); + if (PCIBIOS_SUCCESSFUL != + pci_write_config_word(dev, VIA686A_BASE_REG, address | 1)) + return -ENODEV; + } if (PCIBIOS_SUCCESSFUL != pci_read_config_word(dev, VIA686A_BASE_REG, &val)) return -ENODEV; address = val & ~(VIA686A_EXTENT - 1); - if (address == 0 && force_addr == 0) { + if (address == 0) { dev_err(&dev->dev, "base address not set - upgrade BIOS " "or use force_addr=0xaddr\n"); return -ENODEV; } - s_bridge = pci_dev_get(dev); - if (i2c_isa_add_driver(&via686a_driver)) { - pci_dev_put(s_bridge); - s_bridge = NULL; + if (PCIBIOS_SUCCESSFUL != + pci_read_config_word(dev, VIA686A_ENABLE_REG, &val)) + return -ENODEV; + if (!(val & 0x0001)) { + if (!force_addr) { + dev_warn(&dev->dev, "Sensors disabled, enable " + "with force_addr=0x%x\n", address); + return -ENODEV; + } + + dev_warn(&dev->dev, "Enabling sensors\n"); + if (PCIBIOS_SUCCESSFUL != + pci_write_config_word(dev, VIA686A_ENABLE_REG, + val | 0x0001)) + return -ENODEV; } + if (platform_driver_register(&via686a_driver)) + goto exit; + + /* Sets global pdev as a side effect */ + if (via686a_device_add(address)) + goto exit_unregister; + /* Always return failure here. This is to allow other drivers to bind * to this pci device. We don't really want to have control over the * pci device, we only wanted to read as few register values from it. */ + s_bridge = pci_dev_get(dev); + return -ENODEV; + +exit_unregister: + platform_driver_unregister(&via686a_driver); +exit: return -ENODEV; } @@ -862,7 +859,8 @@ static void __exit sm_via686a_exit(void) { pci_unregister_driver(&via686a_pci_driver); if (s_bridge != NULL) { - i2c_isa_del_driver(&via686a_driver); + platform_device_unregister(pdev); + platform_driver_unregister(&via686a_driver); pci_dev_put(s_bridge); s_bridge = NULL; } diff --git a/drivers/hwmon/vt8231.c b/drivers/hwmon/vt8231.c index a6a4aa0eee1..c604972f018 100644 --- a/drivers/hwmon/vt8231.c +++ b/drivers/hwmon/vt8231.c @@ -29,8 +29,7 @@ #include <linux/slab.h> #include <linux/pci.h> #include <linux/jiffies.h> -#include <linux/i2c.h> -#include <linux/i2c-isa.h> +#include <linux/platform_device.h> #include <linux/hwmon.h> #include <linux/hwmon-sysfs.h> #include <linux/hwmon-vid.h> @@ -42,10 +41,7 @@ static int force_addr; module_param(force_addr, int, 0); MODULE_PARM_DESC(force_addr, "Initialize the base address of the sensors"); -/* Device address - Note that we can't determine the ISA address until we have initialized - our module */ -static unsigned short isa_address; +static struct platform_device *pdev; #define VT8231_EXTENT 0x80 #define VT8231_BASE_REG 0x70 @@ -148,7 +144,9 @@ static inline u8 FAN_TO_REG(long rpm, int div) #define FAN_FROM_REG(val, div) ((val) == 0 ? 0 : 1310720 / ((val) * (div))) struct vt8231_data { - struct i2c_client client; + unsigned short addr; + const char *name; + struct mutex update_lock; struct class_device *class_dev; char valid; /* !=0 if following fields are valid */ @@ -168,20 +166,20 @@ struct vt8231_data { }; static struct pci_dev *s_bridge; -static int vt8231_detect(struct i2c_adapter *adapter); -static int vt8231_detach_client(struct i2c_client *client); +static int vt8231_probe(struct platform_device *pdev); +static int vt8231_remove(struct platform_device *pdev); static struct vt8231_data *vt8231_update_device(struct device *dev); -static void vt8231_init_client(struct i2c_client *client); +static void vt8231_init_device(struct vt8231_data *data); -static inline int vt8231_read_value(struct i2c_client *client, u8 reg) +static inline int vt8231_read_value(struct vt8231_data *data, u8 reg) { - return inb_p(client->addr + reg); + return inb_p(data->addr + reg); } -static inline void vt8231_write_value(struct i2c_client *client, u8 reg, +static inline void vt8231_write_value(struct vt8231_data *data, u8 reg, u8 value) { - outb_p(value, client->addr + reg); + outb_p(value, data->addr + reg); } /* following are the sysfs callback functions */ @@ -220,13 +218,12 @@ static ssize_t set_in_min(struct device *dev, struct device_attribute *attr, { struct sensor_device_attribute *sensor_attr = to_sensor_dev_attr(attr); int nr = sensor_attr->index; - struct i2c_client *client = to_i2c_client(dev); - struct vt8231_data *data = i2c_get_clientdata(client); + struct vt8231_data *data = dev_get_drvdata(dev); unsigned long val = simple_strtoul(buf, NULL, 10); mutex_lock(&data->update_lock); data->in_min[nr] = SENSORS_LIMIT(((val * 958) / 10000) + 3, 0, 255); - vt8231_write_value(client, regvoltmin[nr], data->in_min[nr]); + vt8231_write_value(data, regvoltmin[nr], data->in_min[nr]); mutex_unlock(&data->update_lock); return count; } @@ -236,13 +233,12 @@ static ssize_t set_in_max(struct device *dev, struct device_attribute *attr, { struct sensor_device_attribute *sensor_attr = to_sensor_dev_attr(attr); int nr = sensor_attr->index; - struct i2c_client *client = to_i2c_client(dev); - struct vt8231_data *data = i2c_get_clientdata(client); + struct vt8231_data *data = dev_get_drvdata(dev); unsigned long val = simple_strtoul(buf, NULL, 10); mutex_lock(&data->update_lock); data->in_max[nr] = SENSORS_LIMIT(((val * 958) / 10000) + 3, 0, 255); - vt8231_write_value(client, regvoltmax[nr], data->in_max[nr]); + vt8231_write_value(data, regvoltmax[nr], data->in_max[nr]); mutex_unlock(&data->update_lock); return count; } @@ -278,14 +274,13 @@ static ssize_t show_in5_max(struct device *dev, struct device_attribute *attr, static ssize_t set_in5_min(struct device *dev, struct device_attribute *attr, const char *buf, size_t count) { - struct i2c_client *client = to_i2c_client(dev); - struct vt8231_data *data = i2c_get_clientdata(client); + struct vt8231_data *data = dev_get_drvdata(dev); unsigned long val = simple_strtoul(buf, NULL, 10); mutex_lock(&data->update_lock); data->in_min[5] = SENSORS_LIMIT(((val * 958 * 34) / (10000 * 54)) + 3, 0, 255); - vt8231_write_value(client, regvoltmin[5], data->in_min[5]); + vt8231_write_value(data, regvoltmin[5], data->in_min[5]); mutex_unlock(&data->update_lock); return count; } @@ -293,14 +288,13 @@ static ssize_t set_in5_min(struct device *dev, struct device_attribute *attr, static ssize_t set_in5_max(struct device *dev, struct device_attribute *attr, const char *buf, size_t count) { - struct i2c_client *client = to_i2c_client(dev); - struct vt8231_data *data = i2c_get_clientdata(client); + struct vt8231_data *data = dev_get_drvdata(dev); unsigned long val = simple_strtoul(buf, NULL, 10); mutex_lock(&data->update_lock); data->in_max[5] = SENSORS_LIMIT(((val * 958 * 34) / (10000 * 54)) + 3, 0, 255); - vt8231_write_value(client, regvoltmax[5], data->in_max[5]); + vt8231_write_value(data, regvoltmax[5], data->in_max[5]); mutex_unlock(&data->update_lock); return count; } @@ -348,26 +342,24 @@ static ssize_t show_temp0_min(struct device *dev, struct device_attribute *attr, static ssize_t set_temp0_max(struct device *dev, struct device_attribute *attr, const char *buf, size_t count) { - struct i2c_client *client = to_i2c_client(dev); - struct vt8231_data *data = i2c_get_clientdata(client); + struct vt8231_data *data = dev_get_drvdata(dev); int val = simple_strtol(buf, NULL, 10); mutex_lock(&data->update_lock); data->temp_max[0] = SENSORS_LIMIT((val + 500) / 1000, 0, 255); - vt8231_write_value(client, regtempmax[0], data->temp_max[0]); + vt8231_write_value(data, regtempmax[0], data->temp_max[0]); mutex_unlock(&data->update_lock); return count; } static ssize_t set_temp0_min(struct device *dev, struct device_attribute *attr, const char *buf, size_t count) { - struct i2c_client *client = to_i2c_client(dev); - struct vt8231_data *data = i2c_get_clientdata(client); + struct vt8231_data *data = dev_get_drvdata(dev); int val = simple_strtol(buf, NULL, 10); mutex_lock(&data->update_lock); data->temp_min[0] = SENSORS_LIMIT((val + 500) / 1000, 0, 255); - vt8231_write_value(client, regtempmin[0], data->temp_min[0]); + vt8231_write_value(data, regtempmin[0], data->temp_min[0]); mutex_unlock(&data->update_lock); return count; } @@ -404,13 +396,12 @@ static ssize_t set_temp_max(struct device *dev, struct device_attribute *attr, { struct sensor_device_attribute *sensor_attr = to_sensor_dev_attr(attr); int nr = sensor_attr->index; - struct i2c_client *client = to_i2c_client(dev); - struct vt8231_data *data = i2c_get_clientdata(client); + struct vt8231_data *data = dev_get_drvdata(dev); int val = simple_strtol(buf, NULL, 10); mutex_lock(&data->update_lock); data->temp_max[nr] = SENSORS_LIMIT(TEMP_MAXMIN_TO_REG(val), 0, 255); - vt8231_write_value(client, regtempmax[nr], data->temp_max[nr]); + vt8231_write_value(data, regtempmax[nr], data->temp_max[nr]); mutex_unlock(&data->update_lock); return count; } @@ -419,13 +410,12 @@ static ssize_t set_temp_min(struct device *dev, struct device_attribute *attr, { struct sensor_device_attribute *sensor_attr = to_sensor_dev_attr(attr); int nr = sensor_attr->index; - struct i2c_client *client = to_i2c_client(dev); - struct vt8231_data *data = i2c_get_clientdata(client); + struct vt8231_data *data = dev_get_drvdata(dev); int val = simple_strtol(buf, NULL, 10); mutex_lock(&data->update_lock); data->temp_min[nr] = SENSORS_LIMIT(TEMP_MAXMIN_TO_REG(val), 0, 255); - vt8231_write_value(client, regtempmin[nr], data->temp_min[nr]); + vt8231_write_value(data, regtempmin[nr], data->temp_min[nr]); mutex_unlock(&data->update_lock); return count; } @@ -486,13 +476,12 @@ static ssize_t set_fan_min(struct device *dev, struct device_attribute *attr, { struct sensor_device_attribute *sensor_attr = to_sensor_dev_attr(attr); int nr = sensor_attr->index; - struct i2c_client *client = to_i2c_client(dev); - struct vt8231_data *data = i2c_get_clientdata(client); + struct vt8231_data *data = dev_get_drvdata(dev); int val = simple_strtoul(buf, NULL, 10); mutex_lock(&data->update_lock); data->fan_min[nr] = FAN_TO_REG(val, DIV_FROM_REG(data->fan_div[nr])); - vt8231_write_value(client, VT8231_REG_FAN_MIN(nr), data->fan_min[nr]); + vt8231_write_value(data, VT8231_REG_FAN_MIN(nr), data->fan_min[nr]); mutex_unlock(&data->update_lock); return count; } @@ -500,12 +489,11 @@ static ssize_t set_fan_min(struct device *dev, struct device_attribute *attr, static ssize_t set_fan_div(struct device *dev, struct device_attribute *attr, const char *buf, size_t count) { - struct i2c_client *client = to_i2c_client(dev); - struct vt8231_data *data = i2c_get_clientdata(client); + struct vt8231_data *data = dev_get_drvdata(dev); struct sensor_device_attribute *sensor_attr = to_sensor_dev_attr(attr); unsigned long val = simple_strtoul(buf, NULL, 10); int nr = sensor_attr->index; - int old = vt8231_read_value(client, VT8231_REG_FANDIV); + int old = vt8231_read_value(data, VT8231_REG_FANDIV); long min = FAN_FROM_REG(data->fan_min[nr], DIV_FROM_REG(data->fan_div[nr])); @@ -516,7 +504,7 @@ static ssize_t set_fan_div(struct device *dev, struct device_attribute *attr, case 4: data->fan_div[nr] = 2; break; case 8: data->fan_div[nr] = 3; break; default: - dev_err(&client->dev, "fan_div value %ld not supported." + dev_err(dev, "fan_div value %ld not supported." "Choose one of 1, 2, 4 or 8!\n", val); mutex_unlock(&data->update_lock); return -EINVAL; @@ -524,10 +512,10 @@ static ssize_t set_fan_div(struct device *dev, struct device_attribute *attr, /* Correct the fan minimum speed */ data->fan_min[nr] = FAN_TO_REG(min, DIV_FROM_REG(data->fan_div[nr])); - vt8231_write_value(client, VT8231_REG_FAN_MIN(nr), data->fan_min[nr]); + vt8231_write_value(data, VT8231_REG_FAN_MIN(nr), data->fan_min[nr]); old = (old & 0x0f) | (data->fan_div[1] << 6) | (data->fan_div[0] << 4); - vt8231_write_value(client, VT8231_REG_FANDIV, old); + vt8231_write_value(data, VT8231_REG_FANDIV, old); mutex_unlock(&data->update_lock); return count; } @@ -551,9 +539,16 @@ static ssize_t show_alarms(struct device *dev, struct device_attribute *attr, struct vt8231_data *data = vt8231_update_device(dev); return sprintf(buf, "%d\n", data->alarms); } - static DEVICE_ATTR(alarms, S_IRUGO, show_alarms, NULL); +static ssize_t show_name(struct device *dev, struct device_attribute + *devattr, char *buf) +{ + struct vt8231_data *data = dev_get_drvdata(dev); + return sprintf(buf, "%s\n", data->name); +} +static DEVICE_ATTR(name, S_IRUGO, show_name, NULL); + static struct attribute *vt8231_attributes_temps[6][4] = { { &dev_attr_temp1_input.attr, @@ -648,6 +643,7 @@ static struct attribute *vt8231_attributes[] = { &sensor_dev_attr_fan1_div.dev_attr.attr, &sensor_dev_attr_fan2_div.dev_attr.attr, &dev_attr_alarms.attr, + &dev_attr_name.attr, NULL }; @@ -655,13 +651,13 @@ static const struct attribute_group vt8231_group = { .attrs = vt8231_attributes, }; -static struct i2c_driver vt8231_driver = { +static struct platform_driver vt8231_driver = { .driver = { .owner = THIS_MODULE, .name = "vt8231", }, - .attach_adapter = vt8231_detect, - .detach_client = vt8231_detach_client, + .probe = vt8231_probe, + .remove = __devexit_p(vt8231_remove), }; static struct pci_device_id vt8231_pci_ids[] = { @@ -680,40 +676,18 @@ static struct pci_driver vt8231_pci_driver = { .probe = vt8231_pci_probe, }; -int vt8231_detect(struct i2c_adapter *adapter) +int vt8231_probe(struct platform_device *pdev) { - struct i2c_client *client; + struct resource *res; struct vt8231_data *data; int err = 0, i; - u16 val; - - /* 8231 requires multiple of 256 */ - if (force_addr) { - isa_address = force_addr & 0xFF00; - dev_warn(&adapter->dev, "forcing ISA address 0x%04X\n", - isa_address); - if (PCIBIOS_SUCCESSFUL != pci_write_config_word(s_bridge, - VT8231_BASE_REG, isa_address)) - return -ENODEV; - } - - if (PCIBIOS_SUCCESSFUL != - pci_read_config_word(s_bridge, VT8231_ENABLE_REG, &val)) - return -ENODEV; - - if (!(val & 0x0001)) { - dev_warn(&adapter->dev, "enabling sensors\n"); - if (PCIBIOS_SUCCESSFUL != - pci_write_config_word(s_bridge, VT8231_ENABLE_REG, - val | 0x0001)) - return -ENODEV; - } /* Reserve the ISA region */ - if (!request_region(isa_address, VT8231_EXTENT, - vt8231_pci_driver.name)) { - dev_err(&adapter->dev, "region 0x%x already in use!\n", - isa_address); + res = platform_get_resource(pdev, IORESOURCE_IO, 0); + if (!request_region(res->start, VT8231_EXTENT, + vt8231_driver.driver.name)) { + dev_err(&pdev->dev, "Region 0x%lx-0x%lx already in use!\n", + (unsigned long)res->start, (unsigned long)res->end); return -ENODEV; } @@ -722,33 +696,23 @@ int vt8231_detect(struct i2c_adapter *adapter) goto exit_release; } - client = &data->client; - i2c_set_clientdata(client, data); - client->addr = isa_address; - client->adapter = adapter; - client->driver = &vt8231_driver; - - /* Fill in the remaining client fields and put into the global list */ - strlcpy(client->name, "vt8231", I2C_NAME_SIZE); + platform_set_drvdata(pdev, data); + data->addr = res->start; + data->name = "vt8231"; mutex_init(&data->update_lock); - - /* Tell the I2C layer a new client has arrived */ - if ((err = i2c_attach_client(client))) - goto exit_free; - - vt8231_init_client(client); + vt8231_init_device(data); /* Register sysfs hooks */ - if ((err = sysfs_create_group(&client->dev.kobj, &vt8231_group))) - goto exit_detach; + if ((err = sysfs_create_group(&pdev->dev.kobj, &vt8231_group))) + goto exit_free; /* Must update device information to find out the config field */ - data->uch_config = vt8231_read_value(client, VT8231_REG_UCH_CONFIG); + data->uch_config = vt8231_read_value(data, VT8231_REG_UCH_CONFIG); for (i = 0; i < ARRAY_SIZE(vt8231_group_temps); i++) { if (ISTEMP(i, data->uch_config)) { - if ((err = sysfs_create_group(&client->dev.kobj, + if ((err = sysfs_create_group(&pdev->dev.kobj, &vt8231_group_temps[i]))) goto exit_remove_files; } @@ -756,13 +720,13 @@ int vt8231_detect(struct i2c_adapter *adapter) for (i = 0; i < ARRAY_SIZE(vt8231_group_volts); i++) { if (ISVOLT(i, data->uch_config)) { - if ((err = sysfs_create_group(&client->dev.kobj, + if ((err = sysfs_create_group(&pdev->dev.kobj, &vt8231_group_volts[i]))) goto exit_remove_files; } } - data->class_dev = hwmon_device_register(&client->dev); + data->class_dev = hwmon_device_register(&pdev->dev); if (IS_ERR(data->class_dev)) { err = PTR_ERR(data->class_dev); goto exit_remove_files; @@ -771,56 +735,52 @@ int vt8231_detect(struct i2c_adapter *adapter) exit_remove_files: for (i = 0; i < ARRAY_SIZE(vt8231_group_volts); i++) - sysfs_remove_group(&client->dev.kobj, &vt8231_group_volts[i]); + sysfs_remove_group(&pdev->dev.kobj, &vt8231_group_volts[i]); for (i = 0; i < ARRAY_SIZE(vt8231_group_temps); i++) - sysfs_remove_group(&client->dev.kobj, &vt8231_group_temps[i]); + sysfs_remove_group(&pdev->dev.kobj, &vt8231_group_temps[i]); + + sysfs_remove_group(&pdev->dev.kobj, &vt8231_group); - sysfs_remove_group(&client->dev.kobj, &vt8231_group); -exit_detach: - i2c_detach_client(client); exit_free: + platform_set_drvdata(pdev, NULL); kfree(data); + exit_release: - release_region(isa_address, VT8231_EXTENT); + release_region(res->start, VT8231_EXTENT); return err; } -static int vt8231_detach_client(struct i2c_client *client) +static int vt8231_remove(struct platform_device *pdev) { - struct vt8231_data *data = i2c_get_clientdata(client); - int err, i; + struct vt8231_data *data = platform_get_drvdata(pdev); + int i; hwmon_device_unregister(data->class_dev); for (i = 0; i < ARRAY_SIZE(vt8231_group_volts); i++) - sysfs_remove_group(&client->dev.kobj, &vt8231_group_volts[i]); + sysfs_remove_group(&pdev->dev.kobj, &vt8231_group_volts[i]); for (i = 0; i < ARRAY_SIZE(vt8231_group_temps); i++) - sysfs_remove_group(&client->dev.kobj, &vt8231_group_temps[i]); + sysfs_remove_group(&pdev->dev.kobj, &vt8231_group_temps[i]); - sysfs_remove_group(&client->dev.kobj, &vt8231_group); + sysfs_remove_group(&pdev->dev.kobj, &vt8231_group); - if ((err = i2c_detach_client(client))) { - return err; - } - - release_region(client->addr, VT8231_EXTENT); + release_region(data->addr, VT8231_EXTENT); + platform_set_drvdata(pdev, NULL); kfree(data); - return 0; } -static void vt8231_init_client(struct i2c_client *client) +static void vt8231_init_device(struct vt8231_data *data) { - vt8231_write_value(client, VT8231_REG_TEMP1_CONFIG, 0); - vt8231_write_value(client, VT8231_REG_TEMP2_CONFIG, 0); + vt8231_write_value(data, VT8231_REG_TEMP1_CONFIG, 0); + vt8231_write_value(data, VT8231_REG_TEMP2_CONFIG, 0); } static struct vt8231_data *vt8231_update_device(struct device *dev) { - struct i2c_client *client = to_i2c_client(dev); - struct vt8231_data *data = i2c_get_clientdata(client); + struct vt8231_data *data = dev_get_drvdata(dev); int i; u16 low; @@ -830,41 +790,41 @@ static struct vt8231_data *vt8231_update_device(struct device *dev) || !data->valid) { for (i = 0; i < 6; i++) { if (ISVOLT(i, data->uch_config)) { - data->in[i] = vt8231_read_value(client, + data->in[i] = vt8231_read_value(data, regvolt[i]); - data->in_min[i] = vt8231_read_value(client, + data->in_min[i] = vt8231_read_value(data, regvoltmin[i]); - data->in_max[i] = vt8231_read_value(client, + data->in_max[i] = vt8231_read_value(data, regvoltmax[i]); } } for (i = 0; i < 2; i++) { - data->fan[i] = vt8231_read_value(client, + data->fan[i] = vt8231_read_value(data, VT8231_REG_FAN(i)); - data->fan_min[i] = vt8231_read_value(client, + data->fan_min[i] = vt8231_read_value(data, VT8231_REG_FAN_MIN(i)); } - low = vt8231_read_value(client, VT8231_REG_TEMP_LOW01); + low = vt8231_read_value(data, VT8231_REG_TEMP_LOW01); low = (low >> 6) | ((low & 0x30) >> 2) - | (vt8231_read_value(client, VT8231_REG_TEMP_LOW25) << 4); + | (vt8231_read_value(data, VT8231_REG_TEMP_LOW25) << 4); for (i = 0; i < 6; i++) { if (ISTEMP(i, data->uch_config)) { - data->temp[i] = (vt8231_read_value(client, + data->temp[i] = (vt8231_read_value(data, regtemp[i]) << 2) | ((low >> (2 * i)) & 0x03); - data->temp_max[i] = vt8231_read_value(client, + data->temp_max[i] = vt8231_read_value(data, regtempmax[i]); - data->temp_min[i] = vt8231_read_value(client, + data->temp_min[i] = vt8231_read_value(data, regtempmin[i]); } } - i = vt8231_read_value(client, VT8231_REG_FANDIV); + i = vt8231_read_value(data, VT8231_REG_FANDIV); data->fan_div[0] = (i >> 4) & 0x03; data->fan_div[1] = i >> 6; - data->alarms = vt8231_read_value(client, VT8231_REG_ALARM1) | - (vt8231_read_value(client, VT8231_REG_ALARM2) << 8); + data->alarms = vt8231_read_value(data, VT8231_REG_ALARM1) | + (vt8231_read_value(data, VT8231_REG_ALARM2) << 8); /* Set alarm flags correctly */ if (!data->fan[0] && data->fan_min[0]) { @@ -888,33 +848,102 @@ static struct vt8231_data *vt8231_update_device(struct device *dev) return data; } +static int __devinit vt8231_device_add(unsigned short address) +{ + struct resource res = { + .start = address, + .end = address + VT8231_EXTENT - 1, + .name = "vt8231", + .flags = IORESOURCE_IO, + }; + int err; + + pdev = platform_device_alloc("vt8231", address); + if (!pdev) { + err = -ENOMEM; + printk(KERN_ERR "vt8231: Device allocation failed\n"); + goto exit; + } + + err = platform_device_add_resources(pdev, &res, 1); + if (err) { + printk(KERN_ERR "vt8231: Device resource addition failed " + "(%d)\n", err); + goto exit_device_put; + } + + err = platform_device_add(pdev); + if (err) { + printk(KERN_ERR "vt8231: Device addition failed (%d)\n", + err); + goto exit_device_put; + } + + return 0; + +exit_device_put: + platform_device_put(pdev); +exit: + return err; +} + static int __devinit vt8231_pci_probe(struct pci_dev *dev, const struct pci_device_id *id) { - u16 val; + u16 address, val; + if (force_addr) { + address = force_addr & 0xff00; + dev_warn(&dev->dev, "Forcing ISA address 0x%x\n", + address); + + if (PCIBIOS_SUCCESSFUL != + pci_write_config_word(dev, VT8231_BASE_REG, address | 1)) + return -ENODEV; + } if (PCIBIOS_SUCCESSFUL != pci_read_config_word(dev, VT8231_BASE_REG, &val)) return -ENODEV; - isa_address = val & ~(VT8231_EXTENT - 1); - if (isa_address == 0 && force_addr == 0) { + address = val & ~(VT8231_EXTENT - 1); + if (address == 0) { dev_err(&dev->dev, "base address not set -\ upgrade BIOS or use force_addr=0xaddr\n"); return -ENODEV; } - s_bridge = pci_dev_get(dev); + if (PCIBIOS_SUCCESSFUL != pci_read_config_word(dev, VT8231_ENABLE_REG, + &val)) + return -ENODEV; - if (i2c_isa_add_driver(&vt8231_driver)) { - pci_dev_put(s_bridge); - s_bridge = NULL; + if (!(val & 0x0001)) { + dev_warn(&dev->dev, "enabling sensors\n"); + if (PCIBIOS_SUCCESSFUL != + pci_write_config_word(dev, VT8231_ENABLE_REG, + val | 0x0001)) + return -ENODEV; } + if (platform_driver_register(&vt8231_driver)) + goto exit; + + /* Sets global pdev as a side effect */ + if (vt8231_device_add(address)) + goto exit_unregister; + /* Always return failure here. This is to allow other drivers to bind * to this pci device. We don't really want to have control over the * pci device, we only wanted to read as few register values from it. */ + + /* We do, however, mark ourselves as using the PCI device to stop it + getting unloaded. */ + s_bridge = pci_dev_get(dev); + return -ENODEV; + +exit_unregister: + platform_driver_unregister(&vt8231_driver); +exit: return -ENODEV; } @@ -927,7 +956,8 @@ static void __exit sm_vt8231_exit(void) { pci_unregister_driver(&vt8231_pci_driver); if (s_bridge != NULL) { - i2c_isa_del_driver(&vt8231_driver); + platform_device_unregister(pdev); + platform_driver_unregister(&vt8231_driver); pci_dev_put(s_bridge); s_bridge = NULL; } diff --git a/drivers/hwmon/w83627ehf.c b/drivers/hwmon/w83627ehf.c index 30a76404f0a..c51ae2e1775 100644 --- a/drivers/hwmon/w83627ehf.c +++ b/drivers/hwmon/w83627ehf.c @@ -41,41 +41,39 @@ #include <linux/module.h> #include <linux/init.h> #include <linux/slab.h> -#include <linux/i2c.h> -#include <linux/i2c-isa.h> +#include <linux/jiffies.h> +#include <linux/platform_device.h> #include <linux/hwmon.h> #include <linux/hwmon-sysfs.h> +#include <linux/hwmon-vid.h> #include <linux/err.h> #include <linux/mutex.h> #include <asm/io.h> #include "lm75.h" -/* The actual ISA address is read from Super-I/O configuration space */ -static unsigned short address; +enum kinds { w83627ehf, w83627dhg }; -/* - * Super-I/O constants and functions - */ +/* used to set data->name = w83627ehf_device_names[data->sio_kind] */ +static const char * w83627ehf_device_names[] = { + "w83627ehf", + "w83627dhg", +}; + +#define DRVNAME "w83627ehf" /* - * The three following globals are initialized in w83627ehf_find(), before - * the i2c-isa device is created. Otherwise, they could be stored in - * w83627ehf_data. This is ugly, but necessary, and when the driver is next - * updated to become a platform driver, the globals will disappear. + * Super-I/O constants and functions */ -static int REG; /* The register to read/write */ -static int VAL; /* The value to read/write */ -/* The w83627ehf/ehg have 10 voltage inputs, but the w83627dhg has 9. This - * value is also used in w83627ehf_detect() to export a device name in sysfs - * (e.g. w83627ehf or w83627dhg) */ -static int w83627ehf_num_in; #define W83627EHF_LD_HWM 0x0b #define SIO_REG_LDSEL 0x07 /* Logical device select */ #define SIO_REG_DEVID 0x20 /* Device ID (2 bytes) */ +#define SIO_REG_EN_VRM10 0x2C /* GPIO3, GPIO4 selection */ #define SIO_REG_ENABLE 0x30 /* Logical device enable */ #define SIO_REG_ADDR 0x60 /* Logical device address (2 bytes) */ +#define SIO_REG_VID_CTRL 0xF0 /* VID control */ +#define SIO_REG_VID_DATA 0xF1 /* VID data */ #define SIO_W83627EHF_ID 0x8850 #define SIO_W83627EHG_ID 0x8860 @@ -83,38 +81,38 @@ static int w83627ehf_num_in; #define SIO_ID_MASK 0xFFF0 static inline void -superio_outb(int reg, int val) +superio_outb(int ioreg, int reg, int val) { - outb(reg, REG); - outb(val, VAL); + outb(reg, ioreg); + outb(val, ioreg + 1); } static inline int -superio_inb(int reg) +superio_inb(int ioreg, int reg) { - outb(reg, REG); - return inb(VAL); + outb(reg, ioreg); + return inb(ioreg + 1); } static inline void -superio_select(int ld) +superio_select(int ioreg, int ld) { - outb(SIO_REG_LDSEL, REG); - outb(ld, VAL); + outb(SIO_REG_LDSEL, ioreg); + outb(ld, ioreg + 1); } static inline void -superio_enter(void) +superio_enter(int ioreg) { - outb(0x87, REG); - outb(0x87, REG); + outb(0x87, ioreg); + outb(0x87, ioreg); } static inline void -superio_exit(void) +superio_exit(int ioreg) { - outb(0x02, REG); - outb(0x02, VAL); + outb(0x02, ioreg); + outb(0x02, ioreg + 1); } /* @@ -124,8 +122,8 @@ superio_exit(void) #define IOREGION_ALIGNMENT ~7 #define IOREGION_OFFSET 5 #define IOREGION_LENGTH 2 -#define ADDR_REG_OFFSET 5 -#define DATA_REG_OFFSET 6 +#define ADDR_REG_OFFSET 0 +#define DATA_REG_OFFSET 1 #define W83627EHF_REG_BANK 0x4E #define W83627EHF_REG_CONFIG 0x40 @@ -255,7 +253,9 @@ static inline u8 in_to_reg(u32 val, u8 nr) */ struct w83627ehf_data { - struct i2c_client client; + int addr; /* IO base of hw monitor block */ + const char *name; + struct class_device *class_dev; struct mutex lock; @@ -264,6 +264,7 @@ struct w83627ehf_data { unsigned long last_updated; /* In jiffies */ /* Register values */ + u8 in_num; /* number of in inputs we have */ u8 in[10]; /* Register value */ u8 in_max[10]; /* Register value */ u8 in_min[10]; /* Register value */ @@ -271,6 +272,7 @@ struct w83627ehf_data { u8 fan_min[5]; u8 fan_div[5]; u8 has_fan; /* some fan inputs can be disabled */ + u8 temp_type[3]; s8 temp1; s8 temp1_max; s8 temp1_max_hyst; @@ -288,6 +290,14 @@ struct w83627ehf_data { u8 fan_min_output[4]; /* minimum fan speed */ u8 fan_stop_time[4]; + + u8 vid; + u8 vrm; +}; + +struct w83627ehf_sio_data { + int sioreg; + enum kinds kind; }; static inline int is_word_sized(u16 reg) @@ -303,156 +313,152 @@ static inline int is_word_sized(u16 reg) nothing for registers which live in bank 0. For others, they respectively set the bank register to the correct value (before the register is accessed), and back to 0 (afterwards). */ -static inline void w83627ehf_set_bank(struct i2c_client *client, u16 reg) +static inline void w83627ehf_set_bank(struct w83627ehf_data *data, u16 reg) { if (reg & 0xff00) { - outb_p(W83627EHF_REG_BANK, client->addr + ADDR_REG_OFFSET); - outb_p(reg >> 8, client->addr + DATA_REG_OFFSET); + outb_p(W83627EHF_REG_BANK, data->addr + ADDR_REG_OFFSET); + outb_p(reg >> 8, data->addr + DATA_REG_OFFSET); } } -static inline void w83627ehf_reset_bank(struct i2c_client *client, u16 reg) +static inline void w83627ehf_reset_bank(struct w83627ehf_data *data, u16 reg) { if (reg & 0xff00) { - outb_p(W83627EHF_REG_BANK, client->addr + ADDR_REG_OFFSET); - outb_p(0, client->addr + DATA_REG_OFFSET); + outb_p(W83627EHF_REG_BANK, data->addr + ADDR_REG_OFFSET); + outb_p(0, data->addr + DATA_REG_OFFSET); } } -static u16 w83627ehf_read_value(struct i2c_client *client, u16 reg) +static u16 w83627ehf_read_value(struct w83627ehf_data *data, u16 reg) { - struct w83627ehf_data *data = i2c_get_clientdata(client); int res, word_sized = is_word_sized(reg); mutex_lock(&data->lock); - w83627ehf_set_bank(client, reg); - outb_p(reg & 0xff, client->addr + ADDR_REG_OFFSET); - res = inb_p(client->addr + DATA_REG_OFFSET); + w83627ehf_set_bank(data, reg); + outb_p(reg & 0xff, data->addr + ADDR_REG_OFFSET); + res = inb_p(data->addr + DATA_REG_OFFSET); if (word_sized) { outb_p((reg & 0xff) + 1, - client->addr + ADDR_REG_OFFSET); - res = (res << 8) + inb_p(client->addr + DATA_REG_OFFSET); + data->addr + ADDR_REG_OFFSET); + res = (res << 8) + inb_p(data->addr + DATA_REG_OFFSET); } - w83627ehf_reset_bank(client, reg); + w83627ehf_reset_bank(data, reg); mutex_unlock(&data->lock); return res; } -static int w83627ehf_write_value(struct i2c_client *client, u16 reg, u16 value) +static int w83627ehf_write_value(struct w83627ehf_data *data, u16 reg, u16 value) { - struct w83627ehf_data *data = i2c_get_clientdata(client); int word_sized = is_word_sized(reg); mutex_lock(&data->lock); - w83627ehf_set_bank(client, reg); - outb_p(reg & 0xff, client->addr + ADDR_REG_OFFSET); + w83627ehf_set_bank(data, reg); + outb_p(reg & 0xff, data->addr + ADDR_REG_OFFSET); if (word_sized) { - outb_p(value >> 8, client->addr + DATA_REG_OFFSET); + outb_p(value >> 8, data->addr + DATA_REG_OFFSET); outb_p((reg & 0xff) + 1, - client->addr + ADDR_REG_OFFSET); + data->addr + ADDR_REG_OFFSET); } - outb_p(value & 0xff, client->addr + DATA_REG_OFFSET); - w83627ehf_reset_bank(client, reg); + outb_p(value & 0xff, data->addr + DATA_REG_OFFSET); + w83627ehf_reset_bank(data, reg); mutex_unlock(&data->lock); return 0; } /* This function assumes that the caller holds data->update_lock */ -static void w83627ehf_write_fan_div(struct i2c_client *client, int nr) +static void w83627ehf_write_fan_div(struct w83627ehf_data *data, int nr) { - struct w83627ehf_data *data = i2c_get_clientdata(client); u8 reg; switch (nr) { case 0: - reg = (w83627ehf_read_value(client, W83627EHF_REG_FANDIV1) & 0xcf) + reg = (w83627ehf_read_value(data, W83627EHF_REG_FANDIV1) & 0xcf) | ((data->fan_div[0] & 0x03) << 4); /* fan5 input control bit is write only, compute the value */ reg |= (data->has_fan & (1 << 4)) ? 1 : 0; - w83627ehf_write_value(client, W83627EHF_REG_FANDIV1, reg); - reg = (w83627ehf_read_value(client, W83627EHF_REG_VBAT) & 0xdf) + w83627ehf_write_value(data, W83627EHF_REG_FANDIV1, reg); + reg = (w83627ehf_read_value(data, W83627EHF_REG_VBAT) & 0xdf) | ((data->fan_div[0] & 0x04) << 3); - w83627ehf_write_value(client, W83627EHF_REG_VBAT, reg); + w83627ehf_write_value(data, W83627EHF_REG_VBAT, reg); break; case 1: - reg = (w83627ehf_read_value(client, W83627EHF_REG_FANDIV1) & 0x3f) + reg = (w83627ehf_read_value(data, W83627EHF_REG_FANDIV1) & 0x3f) | ((data->fan_div[1] & 0x03) << 6); /* fan5 input control bit is write only, compute the value */ reg |= (data->has_fan & (1 << 4)) ? 1 : 0; - w83627ehf_write_value(client, W83627EHF_REG_FANDIV1, reg); - reg = (w83627ehf_read_value(client, W83627EHF_REG_VBAT) & 0xbf) + w83627ehf_write_value(data, W83627EHF_REG_FANDIV1, reg); + reg = (w83627ehf_read_value(data, W83627EHF_REG_VBAT) & 0xbf) | ((data->fan_div[1] & 0x04) << 4); - w83627ehf_write_value(client, W83627EHF_REG_VBAT, reg); + w83627ehf_write_value(data, W83627EHF_REG_VBAT, reg); break; case 2: - reg = (w83627ehf_read_value(client, W83627EHF_REG_FANDIV2) & 0x3f) + reg = (w83627ehf_read_value(data, W83627EHF_REG_FANDIV2) & 0x3f) | ((data->fan_div[2] & 0x03) << 6); - w83627ehf_write_value(client, W83627EHF_REG_FANDIV2, reg); - reg = (w83627ehf_read_value(client, W83627EHF_REG_VBAT) & 0x7f) + w83627ehf_write_value(data, W83627EHF_REG_FANDIV2, reg); + reg = (w83627ehf_read_value(data, W83627EHF_REG_VBAT) & 0x7f) | ((data->fan_div[2] & 0x04) << 5); - w83627ehf_write_value(client, W83627EHF_REG_VBAT, reg); + w83627ehf_write_value(data, W83627EHF_REG_VBAT, reg); break; case 3: - reg = (w83627ehf_read_value(client, W83627EHF_REG_DIODE) & 0xfc) + reg = (w83627ehf_read_value(data, W83627EHF_REG_DIODE) & 0xfc) | (data->fan_div[3] & 0x03); - w83627ehf_write_value(client, W83627EHF_REG_DIODE, reg); - reg = (w83627ehf_read_value(client, W83627EHF_REG_SMI_OVT) & 0x7f) + w83627ehf_write_value(data, W83627EHF_REG_DIODE, reg); + reg = (w83627ehf_read_value(data, W83627EHF_REG_SMI_OVT) & 0x7f) | ((data->fan_div[3] & 0x04) << 5); - w83627ehf_write_value(client, W83627EHF_REG_SMI_OVT, reg); + w83627ehf_write_value(data, W83627EHF_REG_SMI_OVT, reg); break; case 4: - reg = (w83627ehf_read_value(client, W83627EHF_REG_DIODE) & 0x73) + reg = (w83627ehf_read_value(data, W83627EHF_REG_DIODE) & 0x73) | ((data->fan_div[4] & 0x03) << 2) | ((data->fan_div[4] & 0x04) << 5); - w83627ehf_write_value(client, W83627EHF_REG_DIODE, reg); + w83627ehf_write_value(data, W83627EHF_REG_DIODE, reg); break; } } static struct w83627ehf_data *w83627ehf_update_device(struct device *dev) { - struct i2c_client *client = to_i2c_client(dev); - struct w83627ehf_data *data = i2c_get_clientdata(client); + struct w83627ehf_data *data = dev_get_drvdata(dev); int pwmcfg = 0, tolerance = 0; /* shut up the compiler */ int i; mutex_lock(&data->update_lock); - if (time_after(jiffies, data->last_updated + HZ) + if (time_after(jiffies, data->last_updated + HZ + HZ/2) || !data->valid) { /* Fan clock dividers */ - i = w83627ehf_read_value(client, W83627EHF_REG_FANDIV1); + i = w83627ehf_read_value(data, W83627EHF_REG_FANDIV1); data->fan_div[0] = (i >> 4) & 0x03; data->fan_div[1] = (i >> 6) & 0x03; - i = w83627ehf_read_value(client, W83627EHF_REG_FANDIV2); + i = w83627ehf_read_value(data, W83627EHF_REG_FANDIV2); data->fan_div[2] = (i >> 6) & 0x03; - i = w83627ehf_read_value(client, W83627EHF_REG_VBAT); + i = w83627ehf_read_value(data, W83627EHF_REG_VBAT); data->fan_div[0] |= (i >> 3) & 0x04; data->fan_div[1] |= (i >> 4) & 0x04; data->fan_div[2] |= (i >> 5) & 0x04; if (data->has_fan & ((1 << 3) | (1 << 4))) { - i = w83627ehf_read_value(client, W83627EHF_REG_DIODE); + i = w83627ehf_read_value(data, W83627EHF_REG_DIODE); data->fan_div[3] = i & 0x03; data->fan_div[4] = ((i >> 2) & 0x03) | ((i >> 5) & 0x04); } if (data->has_fan & (1 << 3)) { - i = w83627ehf_read_value(client, W83627EHF_REG_SMI_OVT); + i = w83627ehf_read_value(data, W83627EHF_REG_SMI_OVT); data->fan_div[3] |= (i >> 5) & 0x04; } /* Measured voltages and limits */ - for (i = 0; i < w83627ehf_num_in; i++) { - data->in[i] = w83627ehf_read_value(client, + for (i = 0; i < data->in_num; i++) { + data->in[i] = w83627ehf_read_value(data, W83627EHF_REG_IN(i)); - data->in_min[i] = w83627ehf_read_value(client, + data->in_min[i] = w83627ehf_read_value(data, W83627EHF_REG_IN_MIN(i)); - data->in_max[i] = w83627ehf_read_value(client, + data->in_max[i] = w83627ehf_read_value(data, W83627EHF_REG_IN_MAX(i)); } @@ -461,9 +467,9 @@ static struct w83627ehf_data *w83627ehf_update_device(struct device *dev) if (!(data->has_fan & (1 << i))) continue; - data->fan[i] = w83627ehf_read_value(client, + data->fan[i] = w83627ehf_read_value(data, W83627EHF_REG_FAN[i]); - data->fan_min[i] = w83627ehf_read_value(client, + data->fan_min[i] = w83627ehf_read_value(data, W83627EHF_REG_FAN_MIN[i]); /* If we failed to measure the fan speed and clock @@ -471,16 +477,16 @@ static struct w83627ehf_data *w83627ehf_update_device(struct device *dev) time */ if (data->fan[i] == 0xff && data->fan_div[i] < 0x07) { - dev_dbg(&client->dev, "Increasing fan%d " + dev_dbg(dev, "Increasing fan%d " "clock divider from %u to %u\n", i + 1, div_from_reg(data->fan_div[i]), div_from_reg(data->fan_div[i] + 1)); data->fan_div[i]++; - w83627ehf_write_fan_div(client, i); + w83627ehf_write_fan_div(data, i); /* Preserve min limit if possible */ if (data->fan_min[i] >= 2 && data->fan_min[i] != 255) - w83627ehf_write_value(client, + w83627ehf_write_value(data, W83627EHF_REG_FAN_MIN[i], (data->fan_min[i] /= 2)); } @@ -489,9 +495,9 @@ static struct w83627ehf_data *w83627ehf_update_device(struct device *dev) for (i = 0; i < 4; i++) { /* pwmcfg, tolarance mapped for i=0, i=1 to same reg */ if (i != 1) { - pwmcfg = w83627ehf_read_value(client, + pwmcfg = w83627ehf_read_value(data, W83627EHF_REG_PWM_ENABLE[i]); - tolerance = w83627ehf_read_value(client, + tolerance = w83627ehf_read_value(data, W83627EHF_REG_TOLERANCE[i]); } data->pwm_mode[i] = @@ -500,14 +506,14 @@ static struct w83627ehf_data *w83627ehf_update_device(struct device *dev) data->pwm_enable[i] = ((pwmcfg >> W83627EHF_PWM_ENABLE_SHIFT[i]) & 3) + 1; - data->pwm[i] = w83627ehf_read_value(client, + data->pwm[i] = w83627ehf_read_value(data, W83627EHF_REG_PWM[i]); - data->fan_min_output[i] = w83627ehf_read_value(client, + data->fan_min_output[i] = w83627ehf_read_value(data, W83627EHF_REG_FAN_MIN_OUTPUT[i]); - data->fan_stop_time[i] = w83627ehf_read_value(client, + data->fan_stop_time[i] = w83627ehf_read_value(data, W83627EHF_REG_FAN_STOP_TIME[i]); data->target_temp[i] = - w83627ehf_read_value(client, + w83627ehf_read_value(data, W83627EHF_REG_TARGET[i]) & (data->pwm_mode[i] == 1 ? 0x7f : 0xff); data->tolerance[i] = (tolerance >> (i == 1 ? 4 : 0)) @@ -515,26 +521,26 @@ static struct w83627ehf_data *w83627ehf_update_device(struct device *dev) } /* Measured temperatures and limits */ - data->temp1 = w83627ehf_read_value(client, + data->temp1 = w83627ehf_read_value(data, W83627EHF_REG_TEMP1); - data->temp1_max = w83627ehf_read_value(client, + data->temp1_max = w83627ehf_read_value(data, W83627EHF_REG_TEMP1_OVER); - data->temp1_max_hyst = w83627ehf_read_value(client, + data->temp1_max_hyst = w83627ehf_read_value(data, W83627EHF_REG_TEMP1_HYST); for (i = 0; i < 2; i++) { - data->temp[i] = w83627ehf_read_value(client, + data->temp[i] = w83627ehf_read_value(data, W83627EHF_REG_TEMP[i]); - data->temp_max[i] = w83627ehf_read_value(client, + data->temp_max[i] = w83627ehf_read_value(data, W83627EHF_REG_TEMP_OVER[i]); - data->temp_max_hyst[i] = w83627ehf_read_value(client, + data->temp_max_hyst[i] = w83627ehf_read_value(data, W83627EHF_REG_TEMP_HYST[i]); } - data->alarms = w83627ehf_read_value(client, + data->alarms = w83627ehf_read_value(data, W83627EHF_REG_ALARM1) | - (w83627ehf_read_value(client, + (w83627ehf_read_value(data, W83627EHF_REG_ALARM2) << 8) | - (w83627ehf_read_value(client, + (w83627ehf_read_value(data, W83627EHF_REG_ALARM3) << 16); data->last_updated = jiffies; @@ -567,15 +573,14 @@ static ssize_t \ store_in_##reg (struct device *dev, struct device_attribute *attr, \ const char *buf, size_t count) \ { \ - struct i2c_client *client = to_i2c_client(dev); \ - struct w83627ehf_data *data = i2c_get_clientdata(client); \ + struct w83627ehf_data *data = dev_get_drvdata(dev); \ struct sensor_device_attribute *sensor_attr = to_sensor_dev_attr(attr); \ int nr = sensor_attr->index; \ u32 val = simple_strtoul(buf, NULL, 10); \ \ mutex_lock(&data->update_lock); \ data->in_##reg[nr] = in_to_reg(val, nr); \ - w83627ehf_write_value(client, W83627EHF_REG_IN_##REG(nr), \ + w83627ehf_write_value(data, W83627EHF_REG_IN_##REG(nr), \ data->in_##reg[nr]); \ mutex_unlock(&data->update_lock); \ return count; \ @@ -673,8 +678,7 @@ static ssize_t store_fan_min(struct device *dev, struct device_attribute *attr, const char *buf, size_t count) { - struct i2c_client *client = to_i2c_client(dev); - struct w83627ehf_data *data = i2c_get_clientdata(client); + struct w83627ehf_data *data = dev_get_drvdata(dev); struct sensor_device_attribute *sensor_attr = to_sensor_dev_attr(attr); int nr = sensor_attr->index; unsigned int val = simple_strtoul(buf, NULL, 10); @@ -716,18 +720,25 @@ store_fan_min(struct device *dev, struct device_attribute *attr, /* Write both the fan clock divider (if it changed) and the new fan min (unconditionally) */ if (new_div != data->fan_div[nr]) { - if (new_div > data->fan_div[nr]) - data->fan[nr] >>= (data->fan_div[nr] - new_div); - else - data->fan[nr] <<= (new_div - data->fan_div[nr]); + /* Preserve the fan speed reading */ + if (data->fan[nr] != 0xff) { + if (new_div > data->fan_div[nr]) + data->fan[nr] >>= new_div - data->fan_div[nr]; + else if (data->fan[nr] & 0x80) + data->fan[nr] = 0xff; + else + data->fan[nr] <<= data->fan_div[nr] - new_div; + } dev_dbg(dev, "fan%u clock divider changed from %u to %u\n", nr + 1, div_from_reg(data->fan_div[nr]), div_from_reg(new_div)); data->fan_div[nr] = new_div; - w83627ehf_write_fan_div(client, nr); + w83627ehf_write_fan_div(data, nr); + /* Give the chip time to sample a new speed value */ + data->last_updated = jiffies; } - w83627ehf_write_value(client, W83627EHF_REG_FAN_MIN[nr], + w83627ehf_write_value(data, W83627EHF_REG_FAN_MIN[nr], data->fan_min[nr]); mutex_unlock(&data->update_lock); @@ -788,13 +799,12 @@ static ssize_t \ store_temp1_##reg(struct device *dev, struct device_attribute *attr, \ const char *buf, size_t count) \ { \ - struct i2c_client *client = to_i2c_client(dev); \ - struct w83627ehf_data *data = i2c_get_clientdata(client); \ + struct w83627ehf_data *data = dev_get_drvdata(dev); \ u32 val = simple_strtoul(buf, NULL, 10); \ \ mutex_lock(&data->update_lock); \ data->temp1_##reg = temp1_to_reg(val, -128000, 127000); \ - w83627ehf_write_value(client, W83627EHF_REG_TEMP1_##REG, \ + w83627ehf_write_value(data, W83627EHF_REG_TEMP1_##REG, \ data->temp1_##reg); \ mutex_unlock(&data->update_lock); \ return count; \ @@ -822,15 +832,14 @@ static ssize_t \ store_##reg(struct device *dev, struct device_attribute *attr, \ const char *buf, size_t count) \ { \ - struct i2c_client *client = to_i2c_client(dev); \ - struct w83627ehf_data *data = i2c_get_clientdata(client); \ + struct w83627ehf_data *data = dev_get_drvdata(dev); \ struct sensor_device_attribute *sensor_attr = to_sensor_dev_attr(attr); \ int nr = sensor_attr->index; \ u32 val = simple_strtoul(buf, NULL, 10); \ \ mutex_lock(&data->update_lock); \ data->reg[nr] = LM75_TEMP_TO_REG(val); \ - w83627ehf_write_value(client, W83627EHF_REG_TEMP_##REG[nr], \ + w83627ehf_write_value(data, W83627EHF_REG_TEMP_##REG[nr], \ data->reg[nr]); \ mutex_unlock(&data->update_lock); \ return count; \ @@ -838,6 +847,15 @@ store_##reg(struct device *dev, struct device_attribute *attr, \ store_temp_reg(OVER, temp_max); store_temp_reg(HYST, temp_max_hyst); +static ssize_t +show_temp_type(struct device *dev, struct device_attribute *attr, char *buf) +{ + struct w83627ehf_data *data = w83627ehf_update_device(dev); + struct sensor_device_attribute *sensor_attr = to_sensor_dev_attr(attr); + int nr = sensor_attr->index; + return sprintf(buf, "%d\n", (int)data->temp_type[nr]); +} + static struct sensor_device_attribute sda_temp[] = { SENSOR_ATTR(temp1_input, S_IRUGO, show_temp1, NULL, 0), SENSOR_ATTR(temp2_input, S_IRUGO, show_temp, NULL, 0), @@ -857,6 +875,9 @@ static struct sensor_device_attribute sda_temp[] = { SENSOR_ATTR(temp1_alarm, S_IRUGO, show_alarm, NULL, 4), SENSOR_ATTR(temp2_alarm, S_IRUGO, show_alarm, NULL, 5), SENSOR_ATTR(temp3_alarm, S_IRUGO, show_alarm, NULL, 13), + SENSOR_ATTR(temp1_type, S_IRUGO, show_temp_type, NULL, 0), + SENSOR_ATTR(temp2_type, S_IRUGO, show_temp_type, NULL, 1), + SENSOR_ATTR(temp3_type, S_IRUGO, show_temp_type, NULL, 2), }; #define show_pwm_reg(reg) \ @@ -877,8 +898,7 @@ static ssize_t store_pwm_mode(struct device *dev, struct device_attribute *attr, const char *buf, size_t count) { - struct i2c_client *client = to_i2c_client(dev); - struct w83627ehf_data *data = i2c_get_clientdata(client); + struct w83627ehf_data *data = dev_get_drvdata(dev); struct sensor_device_attribute *sensor_attr = to_sensor_dev_attr(attr); int nr = sensor_attr->index; u32 val = simple_strtoul(buf, NULL, 10); @@ -887,12 +907,12 @@ store_pwm_mode(struct device *dev, struct device_attribute *attr, if (val > 1) return -EINVAL; mutex_lock(&data->update_lock); - reg = w83627ehf_read_value(client, W83627EHF_REG_PWM_ENABLE[nr]); + reg = w83627ehf_read_value(data, W83627EHF_REG_PWM_ENABLE[nr]); data->pwm_mode[nr] = val; reg &= ~(1 << W83627EHF_PWM_MODE_SHIFT[nr]); if (!val) reg |= 1 << W83627EHF_PWM_MODE_SHIFT[nr]; - w83627ehf_write_value(client, W83627EHF_REG_PWM_ENABLE[nr], reg); + w83627ehf_write_value(data, W83627EHF_REG_PWM_ENABLE[nr], reg); mutex_unlock(&data->update_lock); return count; } @@ -901,15 +921,14 @@ static ssize_t store_pwm(struct device *dev, struct device_attribute *attr, const char *buf, size_t count) { - struct i2c_client *client = to_i2c_client(dev); - struct w83627ehf_data *data = i2c_get_clientdata(client); + struct w83627ehf_data *data = dev_get_drvdata(dev); struct sensor_device_attribute *sensor_attr = to_sensor_dev_attr(attr); int nr = sensor_attr->index; u32 val = SENSORS_LIMIT(simple_strtoul(buf, NULL, 10), 0, 255); mutex_lock(&data->update_lock); data->pwm[nr] = val; - w83627ehf_write_value(client, W83627EHF_REG_PWM[nr], val); + w83627ehf_write_value(data, W83627EHF_REG_PWM[nr], val); mutex_unlock(&data->update_lock); return count; } @@ -918,8 +937,7 @@ static ssize_t store_pwm_enable(struct device *dev, struct device_attribute *attr, const char *buf, size_t count) { - struct i2c_client *client = to_i2c_client(dev); - struct w83627ehf_data *data = i2c_get_clientdata(client); + struct w83627ehf_data *data = dev_get_drvdata(dev); struct sensor_device_attribute *sensor_attr = to_sensor_dev_attr(attr); int nr = sensor_attr->index; u32 val = simple_strtoul(buf, NULL, 10); @@ -928,11 +946,11 @@ store_pwm_enable(struct device *dev, struct device_attribute *attr, if (!val || (val > 2)) /* only modes 1 and 2 are supported */ return -EINVAL; mutex_lock(&data->update_lock); - reg = w83627ehf_read_value(client, W83627EHF_REG_PWM_ENABLE[nr]); + reg = w83627ehf_read_value(data, W83627EHF_REG_PWM_ENABLE[nr]); data->pwm_enable[nr] = val; reg &= ~(0x03 << W83627EHF_PWM_ENABLE_SHIFT[nr]); reg |= (val - 1) << W83627EHF_PWM_ENABLE_SHIFT[nr]; - w83627ehf_write_value(client, W83627EHF_REG_PWM_ENABLE[nr], reg); + w83627ehf_write_value(data, W83627EHF_REG_PWM_ENABLE[nr], reg); mutex_unlock(&data->update_lock); return count; } @@ -955,15 +973,14 @@ static ssize_t store_target_temp(struct device *dev, struct device_attribute *attr, const char *buf, size_t count) { - struct i2c_client *client = to_i2c_client(dev); - struct w83627ehf_data *data = i2c_get_clientdata(client); + struct w83627ehf_data *data = dev_get_drvdata(dev); struct sensor_device_attribute *sensor_attr = to_sensor_dev_attr(attr); int nr = sensor_attr->index; u8 val = temp1_to_reg(simple_strtoul(buf, NULL, 10), 0, 127000); mutex_lock(&data->update_lock); data->target_temp[nr] = val; - w83627ehf_write_value(client, W83627EHF_REG_TARGET[nr], val); + w83627ehf_write_value(data, W83627EHF_REG_TARGET[nr], val); mutex_unlock(&data->update_lock); return count; } @@ -972,8 +989,7 @@ static ssize_t store_tolerance(struct device *dev, struct device_attribute *attr, const char *buf, size_t count) { - struct i2c_client *client = to_i2c_client(dev); - struct w83627ehf_data *data = i2c_get_clientdata(client); + struct w83627ehf_data *data = dev_get_drvdata(dev); struct sensor_device_attribute *sensor_attr = to_sensor_dev_attr(attr); int nr = sensor_attr->index; u16 reg; @@ -981,13 +997,13 @@ store_tolerance(struct device *dev, struct device_attribute *attr, u8 val = temp1_to_reg(simple_strtoul(buf, NULL, 10), 0, 15000); mutex_lock(&data->update_lock); - reg = w83627ehf_read_value(client, W83627EHF_REG_TOLERANCE[nr]); + reg = w83627ehf_read_value(data, W83627EHF_REG_TOLERANCE[nr]); data->tolerance[nr] = val; if (nr == 1) reg = (reg & 0x0f) | (val << 4); else reg = (reg & 0xf0) | val; - w83627ehf_write_value(client, W83627EHF_REG_TOLERANCE[nr], reg); + w83627ehf_write_value(data, W83627EHF_REG_TOLERANCE[nr], reg); mutex_unlock(&data->update_lock); return count; } @@ -1058,14 +1074,13 @@ static ssize_t \ store_##reg(struct device *dev, struct device_attribute *attr, \ const char *buf, size_t count) \ {\ - struct i2c_client *client = to_i2c_client(dev); \ - struct w83627ehf_data *data = i2c_get_clientdata(client); \ + struct w83627ehf_data *data = dev_get_drvdata(dev); \ struct sensor_device_attribute *sensor_attr = to_sensor_dev_attr(attr); \ int nr = sensor_attr->index; \ u32 val = SENSORS_LIMIT(simple_strtoul(buf, NULL, 10), 1, 255); \ mutex_lock(&data->update_lock); \ data->reg[nr] = val; \ - w83627ehf_write_value(client, W83627EHF_REG_##REG[nr], val); \ + w83627ehf_write_value(data, W83627EHF_REG_##REG[nr], val); \ mutex_unlock(&data->update_lock); \ return count; \ } @@ -1087,21 +1102,28 @@ static ssize_t \ store_##reg(struct device *dev, struct device_attribute *attr, \ const char *buf, size_t count) \ { \ - struct i2c_client *client = to_i2c_client(dev); \ - struct w83627ehf_data *data = i2c_get_clientdata(client); \ + struct w83627ehf_data *data = dev_get_drvdata(dev); \ struct sensor_device_attribute *sensor_attr = to_sensor_dev_attr(attr); \ int nr = sensor_attr->index; \ u8 val = step_time_to_reg(simple_strtoul(buf, NULL, 10), \ data->pwm_mode[nr]); \ mutex_lock(&data->update_lock); \ data->reg[nr] = val; \ - w83627ehf_write_value(client, W83627EHF_REG_##REG[nr], val); \ + w83627ehf_write_value(data, W83627EHF_REG_##REG[nr], val); \ mutex_unlock(&data->update_lock); \ return count; \ } \ fan_time_functions(fan_stop_time, FAN_STOP_TIME) +static ssize_t show_name(struct device *dev, struct device_attribute *attr, + char *buf) +{ + struct w83627ehf_data *data = dev_get_drvdata(dev); + + return sprintf(buf, "%s\n", data->name); +} +static DEVICE_ATTR(name, S_IRUGO, show_name, NULL); static struct sensor_device_attribute sda_sf3_arrays_fan4[] = { SENSOR_ATTR(pwm4_stop_time, S_IWUSR | S_IRUGO, show_fan_stop_time, @@ -1125,8 +1147,16 @@ static struct sensor_device_attribute sda_sf3_arrays[] = { store_fan_min_output, 2), }; +static ssize_t +show_vid(struct device *dev, struct device_attribute *attr, char *buf) +{ + struct w83627ehf_data *data = dev_get_drvdata(dev); + return sprintf(buf, "%d\n", vid_from_reg(data->vid, data->vrm)); +} +static DEVICE_ATTR(cpu0_vid, S_IRUGO, show_vid, NULL); + /* - * Driver and client management + * Driver and device management */ static void w83627ehf_device_remove_files(struct device *dev) @@ -1134,12 +1164,13 @@ static void w83627ehf_device_remove_files(struct device *dev) /* some entries in the following arrays may not have been used in * device_create_file(), but device_remove_file() will ignore them */ int i; + struct w83627ehf_data *data = dev_get_drvdata(dev); for (i = 0; i < ARRAY_SIZE(sda_sf3_arrays); i++) device_remove_file(dev, &sda_sf3_arrays[i].dev_attr); for (i = 0; i < ARRAY_SIZE(sda_sf3_arrays_fan4); i++) device_remove_file(dev, &sda_sf3_arrays_fan4[i].dev_attr); - for (i = 0; i < w83627ehf_num_in; i++) { + for (i = 0; i < data->in_num; i++) { device_remove_file(dev, &sda_in_input[i].dev_attr); device_remove_file(dev, &sda_in_alarm[i].dev_attr); device_remove_file(dev, &sda_in_min[i].dev_attr); @@ -1160,43 +1191,64 @@ static void w83627ehf_device_remove_files(struct device *dev) } for (i = 0; i < ARRAY_SIZE(sda_temp); i++) device_remove_file(dev, &sda_temp[i].dev_attr); -} -static struct i2c_driver w83627ehf_driver; + device_remove_file(dev, &dev_attr_name); + if (data->vid != 0x3f) + device_remove_file(dev, &dev_attr_cpu0_vid); +} -static void w83627ehf_init_client(struct i2c_client *client) +/* Get the monitoring functions started */ +static inline void __devinit w83627ehf_init_device(struct w83627ehf_data *data) { int i; - u8 tmp; + u8 tmp, diode; /* Start monitoring is needed */ - tmp = w83627ehf_read_value(client, W83627EHF_REG_CONFIG); + tmp = w83627ehf_read_value(data, W83627EHF_REG_CONFIG); if (!(tmp & 0x01)) - w83627ehf_write_value(client, W83627EHF_REG_CONFIG, + w83627ehf_write_value(data, W83627EHF_REG_CONFIG, tmp | 0x01); /* Enable temp2 and temp3 if needed */ for (i = 0; i < 2; i++) { - tmp = w83627ehf_read_value(client, + tmp = w83627ehf_read_value(data, W83627EHF_REG_TEMP_CONFIG[i]); if (tmp & 0x01) - w83627ehf_write_value(client, + w83627ehf_write_value(data, W83627EHF_REG_TEMP_CONFIG[i], tmp & 0xfe); } + + /* Enable VBAT monitoring if needed */ + tmp = w83627ehf_read_value(data, W83627EHF_REG_VBAT); + if (!(tmp & 0x01)) + w83627ehf_write_value(data, W83627EHF_REG_VBAT, tmp | 0x01); + + /* Get thermal sensor types */ + diode = w83627ehf_read_value(data, W83627EHF_REG_DIODE); + for (i = 0; i < 3; i++) { + if ((tmp & (0x02 << i))) + data->temp_type[i] = (diode & (0x10 << i)) ? 1 : 2; + else + data->temp_type[i] = 4; /* thermistor */ + } } -static int w83627ehf_detect(struct i2c_adapter *adapter) +static int __devinit w83627ehf_probe(struct platform_device *pdev) { - struct i2c_client *client; + struct device *dev = &pdev->dev; + struct w83627ehf_sio_data *sio_data = dev->platform_data; struct w83627ehf_data *data; - struct device *dev; - u8 fan4pin, fan5pin; + struct resource *res; + u8 fan4pin, fan5pin, en_vrm10; int i, err = 0; - if (!request_region(address + IOREGION_OFFSET, IOREGION_LENGTH, - w83627ehf_driver.driver.name)) { + res = platform_get_resource(pdev, IORESOURCE_IO, 0); + if (!request_region(res->start, IOREGION_LENGTH, DRVNAME)) { err = -EBUSY; + dev_err(dev, "Failed to request region 0x%lx-0x%lx\n", + (unsigned long)res->start, + (unsigned long)res->start + IOREGION_LENGTH - 1); goto exit; } @@ -1205,41 +1257,47 @@ static int w83627ehf_detect(struct i2c_adapter *adapter) goto exit_release; } - client = &data->client; - i2c_set_clientdata(client, data); - client->addr = address; + data->addr = res->start; mutex_init(&data->lock); - client->adapter = adapter; - client->driver = &w83627ehf_driver; - client->flags = 0; - dev = &client->dev; - - if (w83627ehf_num_in == 9) - strlcpy(client->name, "w83627dhg", I2C_NAME_SIZE); - else /* just say ehf. 627EHG is 627EHF in lead-free packaging. */ - strlcpy(client->name, "w83627ehf", I2C_NAME_SIZE); - - data->valid = 0; mutex_init(&data->update_lock); + data->name = w83627ehf_device_names[sio_data->kind]; + platform_set_drvdata(pdev, data); - /* Tell the i2c layer a new client has arrived */ - if ((err = i2c_attach_client(client))) - goto exit_free; + /* 627EHG and 627EHF have 10 voltage inputs; DHG has 9 */ + data->in_num = (sio_data->kind == w83627dhg) ? 9 : 10; /* Initialize the chip */ - w83627ehf_init_client(client); - - /* A few vars need to be filled upon startup */ - for (i = 0; i < 5; i++) - data->fan_min[i] = w83627ehf_read_value(client, - W83627EHF_REG_FAN_MIN[i]); + w83627ehf_init_device(data); + + data->vrm = vid_which_vrm(); + superio_enter(sio_data->sioreg); + /* Set VID input sensibility if needed. In theory the BIOS should + have set it, but in practice it's not always the case. */ + en_vrm10 = superio_inb(sio_data->sioreg, SIO_REG_EN_VRM10); + if ((en_vrm10 & 0x08) && data->vrm != 100) { + dev_warn(dev, "Setting VID input voltage to TTL\n"); + superio_outb(sio_data->sioreg, SIO_REG_EN_VRM10, + en_vrm10 & ~0x08); + } else if (!(en_vrm10 & 0x08) && data->vrm == 100) { + dev_warn(dev, "Setting VID input voltage to VRM10\n"); + superio_outb(sio_data->sioreg, SIO_REG_EN_VRM10, + en_vrm10 | 0x08); + } + /* Read VID value */ + superio_select(sio_data->sioreg, W83627EHF_LD_HWM); + if (superio_inb(sio_data->sioreg, SIO_REG_VID_CTRL) & 0x80) + data->vid = superio_inb(sio_data->sioreg, SIO_REG_VID_DATA) & 0x3f; + else { + dev_info(dev, "VID pins in output mode, CPU VID not " + "available\n"); + data->vid = 0x3f; + } /* fan4 and fan5 share some pins with the GPIO and serial flash */ - superio_enter(); - fan5pin = superio_inb(0x24) & 0x2; - fan4pin = superio_inb(0x29) & 0x6; - superio_exit(); + fan5pin = superio_inb(sio_data->sioreg, 0x24) & 0x2; + fan4pin = superio_inb(sio_data->sioreg, 0x29) & 0x6; + superio_exit(sio_data->sioreg); /* It looks like fan4 and fan5 pins can be alternatively used as fan on/off switches, but fan5 control is write only :/ @@ -1248,7 +1306,7 @@ static int w83627ehf_detect(struct i2c_adapter *adapter) is not the default. */ data->has_fan = 0x07; /* fan1, fan2 and fan3 */ - i = w83627ehf_read_value(client, W83627EHF_REG_FANDIV1); + i = w83627ehf_read_value(data, W83627EHF_REG_FANDIV1); if ((i & (1 << 2)) && (!fan4pin)) data->has_fan |= (1 << 3); if (!(i & (1 << 1)) && (!fan5pin)) @@ -1268,7 +1326,7 @@ static int w83627ehf_detect(struct i2c_adapter *adapter) goto exit_remove; } - for (i = 0; i < w83627ehf_num_in; i++) + for (i = 0; i < data->in_num; i++) if ((err = device_create_file(dev, &sda_in_input[i].dev_attr)) || (err = device_create_file(dev, &sda_in_alarm[i].dev_attr)) @@ -1308,6 +1366,16 @@ static int w83627ehf_detect(struct i2c_adapter *adapter) if ((err = device_create_file(dev, &sda_temp[i].dev_attr))) goto exit_remove; + err = device_create_file(dev, &dev_attr_name); + if (err) + goto exit_remove; + + if (data->vid != 0x3f) { + err = device_create_file(dev, &dev_attr_cpu0_vid); + if (err) + goto exit_remove; + } + data->class_dev = hwmon_device_register(dev); if (IS_ERR(data->class_dev)) { err = PTR_ERR(data->class_dev); @@ -1318,95 +1386,172 @@ static int w83627ehf_detect(struct i2c_adapter *adapter) exit_remove: w83627ehf_device_remove_files(dev); - i2c_detach_client(client); -exit_free: kfree(data); + platform_set_drvdata(pdev, NULL); exit_release: - release_region(address + IOREGION_OFFSET, IOREGION_LENGTH); + release_region(res->start, IOREGION_LENGTH); exit: return err; } -static int w83627ehf_detach_client(struct i2c_client *client) +static int __devexit w83627ehf_remove(struct platform_device *pdev) { - struct w83627ehf_data *data = i2c_get_clientdata(client); - int err; + struct w83627ehf_data *data = platform_get_drvdata(pdev); hwmon_device_unregister(data->class_dev); - w83627ehf_device_remove_files(&client->dev); - - if ((err = i2c_detach_client(client))) - return err; - release_region(client->addr + IOREGION_OFFSET, IOREGION_LENGTH); + w83627ehf_device_remove_files(&pdev->dev); + release_region(data->addr, IOREGION_LENGTH); + platform_set_drvdata(pdev, NULL); kfree(data); return 0; } -static struct i2c_driver w83627ehf_driver = { +static struct platform_driver w83627ehf_driver = { .driver = { .owner = THIS_MODULE, - .name = "w83627ehf", + .name = DRVNAME, }, - .attach_adapter = w83627ehf_detect, - .detach_client = w83627ehf_detach_client, + .probe = w83627ehf_probe, + .remove = __devexit_p(w83627ehf_remove), }; -static int __init w83627ehf_find(int sioaddr, unsigned short *addr) +/* w83627ehf_find() looks for a '627 in the Super-I/O config space */ +static int __init w83627ehf_find(int sioaddr, unsigned short *addr, + struct w83627ehf_sio_data *sio_data) { + static const char __initdata sio_name_W83627EHF[] = "W83627EHF"; + static const char __initdata sio_name_W83627EHG[] = "W83627EHG"; + static const char __initdata sio_name_W83627DHG[] = "W83627DHG"; + u16 val; + const char *sio_name; - REG = sioaddr; - VAL = sioaddr + 1; - superio_enter(); + superio_enter(sioaddr); - val = (superio_inb(SIO_REG_DEVID) << 8) - | superio_inb(SIO_REG_DEVID + 1); + val = (superio_inb(sioaddr, SIO_REG_DEVID) << 8) + | superio_inb(sioaddr, SIO_REG_DEVID + 1); switch (val & SIO_ID_MASK) { - case SIO_W83627DHG_ID: - w83627ehf_num_in = 9; - break; case SIO_W83627EHF_ID: + sio_data->kind = w83627ehf; + sio_name = sio_name_W83627EHF; + break; case SIO_W83627EHG_ID: - w83627ehf_num_in = 10; + sio_data->kind = w83627ehf; + sio_name = sio_name_W83627EHG; + break; + case SIO_W83627DHG_ID: + sio_data->kind = w83627dhg; + sio_name = sio_name_W83627DHG; break; default: - printk(KERN_WARNING "w83627ehf: unsupported chip ID: 0x%04x\n", - val); - superio_exit(); + if (val != 0xffff) + pr_debug(DRVNAME ": unsupported chip ID: 0x%04x\n", + val); + superio_exit(sioaddr); return -ENODEV; } - superio_select(W83627EHF_LD_HWM); - val = (superio_inb(SIO_REG_ADDR) << 8) - | superio_inb(SIO_REG_ADDR + 1); + /* We have a known chip, find the HWM I/O address */ + superio_select(sioaddr, W83627EHF_LD_HWM); + val = (superio_inb(sioaddr, SIO_REG_ADDR) << 8) + | superio_inb(sioaddr, SIO_REG_ADDR + 1); *addr = val & IOREGION_ALIGNMENT; if (*addr == 0) { - superio_exit(); + printk(KERN_ERR DRVNAME ": Refusing to enable a Super-I/O " + "device with a base I/O port 0.\n"); + superio_exit(sioaddr); return -ENODEV; } /* Activate logical device if needed */ - val = superio_inb(SIO_REG_ENABLE); - if (!(val & 0x01)) - superio_outb(SIO_REG_ENABLE, val | 0x01); + val = superio_inb(sioaddr, SIO_REG_ENABLE); + if (!(val & 0x01)) { + printk(KERN_WARNING DRVNAME ": Forcibly enabling Super-I/O. " + "Sensor is probably unusable.\n"); + superio_outb(sioaddr, SIO_REG_ENABLE, val | 0x01); + } + + superio_exit(sioaddr); + pr_info(DRVNAME ": Found %s chip at %#x\n", sio_name, *addr); + sio_data->sioreg = sioaddr; - superio_exit(); return 0; } +/* when Super-I/O functions move to a separate file, the Super-I/O + * bus will manage the lifetime of the device and this module will only keep + * track of the w83627ehf driver. But since we platform_device_alloc(), we + * must keep track of the device */ +static struct platform_device *pdev; + static int __init sensors_w83627ehf_init(void) { - if (w83627ehf_find(0x2e, &address) - && w83627ehf_find(0x4e, &address)) + int err; + unsigned short address; + struct resource res; + struct w83627ehf_sio_data sio_data; + + /* initialize sio_data->kind and sio_data->sioreg. + * + * when Super-I/O functions move to a separate file, the Super-I/O + * driver will probe 0x2e and 0x4e and auto-detect the presence of a + * w83627ehf hardware monitor, and call probe() */ + if (w83627ehf_find(0x2e, &address, &sio_data) && + w83627ehf_find(0x4e, &address, &sio_data)) return -ENODEV; - return i2c_isa_add_driver(&w83627ehf_driver); + err = platform_driver_register(&w83627ehf_driver); + if (err) + goto exit; + + if (!(pdev = platform_device_alloc(DRVNAME, address))) { + err = -ENOMEM; + printk(KERN_ERR DRVNAME ": Device allocation failed\n"); + goto exit_unregister; + } + + err = platform_device_add_data(pdev, &sio_data, + sizeof(struct w83627ehf_sio_data)); + if (err) { + printk(KERN_ERR DRVNAME ": Platform data allocation failed\n"); + goto exit_device_put; + } + + memset(&res, 0, sizeof(res)); + res.name = DRVNAME; + res.start = address + IOREGION_OFFSET; + res.end = address + IOREGION_OFFSET + IOREGION_LENGTH - 1; + res.flags = IORESOURCE_IO; + err = platform_device_add_resources(pdev, &res, 1); + if (err) { + printk(KERN_ERR DRVNAME ": Device resource addition failed " + "(%d)\n", err); + goto exit_device_put; + } + + /* platform_device_add calls probe() */ + err = platform_device_add(pdev); + if (err) { + printk(KERN_ERR DRVNAME ": Device addition failed (%d)\n", + err); + goto exit_device_put; + } + + return 0; + +exit_device_put: + platform_device_put(pdev); +exit_unregister: + platform_driver_unregister(&w83627ehf_driver); +exit: + return err; } static void __exit sensors_w83627ehf_exit(void) { - i2c_isa_del_driver(&w83627ehf_driver); + platform_device_unregister(pdev); + platform_driver_unregister(&w83627ehf_driver); } MODULE_AUTHOR("Jean Delvare <khali@linux-fr.org>"); diff --git a/drivers/hwmon/w83627hf.c b/drivers/hwmon/w83627hf.c index 12cb40a975d..1ce78179b00 100644 --- a/drivers/hwmon/w83627hf.c +++ b/drivers/hwmon/w83627hf.c @@ -220,6 +220,18 @@ static const u8 regpwm[] = { W83627THF_REG_PWM1, W83627THF_REG_PWM2, #define W836X7HF_REG_PWM(type, nr) (((type) == w83627hf) ? \ regpwm_627hf[(nr) - 1] : regpwm[(nr) - 1]) +#define W83627HF_REG_PWM_FREQ 0x5C /* Only for the 627HF */ + +#define W83637HF_REG_PWM_FREQ1 0x00 /* 697HF/687THF too */ +#define W83637HF_REG_PWM_FREQ2 0x02 /* 697HF/687THF too */ +#define W83637HF_REG_PWM_FREQ3 0x10 /* 687THF too */ + +static const u8 W83637HF_REG_PWM_FREQ[] = { W83637HF_REG_PWM_FREQ1, + W83637HF_REG_PWM_FREQ2, + W83637HF_REG_PWM_FREQ3 }; + +#define W83627HF_BASE_PWM_FREQ 46870 + #define W83781D_REG_I2C_ADDR 0x48 #define W83781D_REG_I2C_SUBADDR 0x4A @@ -267,6 +279,49 @@ static int TEMP_FROM_REG(u8 reg) #define PWM_TO_REG(val) (SENSORS_LIMIT((val),0,255)) +static inline unsigned long pwm_freq_from_reg_627hf(u8 reg) +{ + unsigned long freq; + freq = W83627HF_BASE_PWM_FREQ >> reg; + return freq; +} +static inline u8 pwm_freq_to_reg_627hf(unsigned long val) +{ + u8 i; + /* Only 5 dividers (1 2 4 8 16) + Search for the nearest available frequency */ + for (i = 0; i < 4; i++) { + if (val > (((W83627HF_BASE_PWM_FREQ >> i) + + (W83627HF_BASE_PWM_FREQ >> (i+1))) / 2)) + break; + } + return i; +} + +static inline unsigned long pwm_freq_from_reg(u8 reg) +{ + /* Clock bit 8 -> 180 kHz or 24 MHz */ + unsigned long clock = (reg & 0x80) ? 180000UL : 24000000UL; + + reg &= 0x7f; + /* This should not happen but anyway... */ + if (reg == 0) + reg++; + return (clock / (reg << 8)); +} +static inline u8 pwm_freq_to_reg(unsigned long val) +{ + /* Minimum divider value is 0x01 and maximum is 0x7F */ + if (val >= 93750) /* The highest we can do */ + return 0x01; + if (val >= 720) /* Use 24 MHz clock */ + return (24000000UL / (val << 8)); + if (val < 6) /* The lowest we can do */ + return 0xFF; + else /* Use 180 kHz clock */ + return (0x80 | (180000UL / (val << 8))); +} + #define BEEP_MASK_FROM_REG(val) (val) #define BEEP_MASK_TO_REG(val) ((val) & 0xffffff) #define BEEP_ENABLE_TO_REG(val) ((val)?1:0) @@ -316,6 +371,7 @@ struct w83627hf_data { u32 beep_mask; /* Register encoding, combined */ u8 beep_enable; /* Boolean */ u8 pwm[3]; /* Register value */ + u8 pwm_freq[3]; /* Register value */ u16 sens[3]; /* 782D/783S only. 1 = pentium diode; 2 = 3904 diode; 3000-5000 = thermistor beta. @@ -852,6 +908,64 @@ sysfs_pwm(2); sysfs_pwm(3); static ssize_t +show_pwm_freq_reg(struct device *dev, char *buf, int nr) +{ + struct w83627hf_data *data = w83627hf_update_device(dev); + if (data->type == w83627hf) + return sprintf(buf, "%ld\n", + pwm_freq_from_reg_627hf(data->pwm_freq[nr - 1])); + else + return sprintf(buf, "%ld\n", + pwm_freq_from_reg(data->pwm_freq[nr - 1])); +} + +static ssize_t +store_pwm_freq_reg(struct device *dev, const char *buf, size_t count, int nr) +{ + struct w83627hf_data *data = dev_get_drvdata(dev); + static const u8 mask[]={0xF8, 0x8F}; + u32 val; + + val = simple_strtoul(buf, NULL, 10); + + mutex_lock(&data->update_lock); + + if (data->type == w83627hf) { + data->pwm_freq[nr - 1] = pwm_freq_to_reg_627hf(val); + w83627hf_write_value(data, W83627HF_REG_PWM_FREQ, + (data->pwm_freq[nr - 1] << ((nr - 1)*4)) | + (w83627hf_read_value(data, + W83627HF_REG_PWM_FREQ) & mask[nr - 1])); + } else { + data->pwm_freq[nr - 1] = pwm_freq_to_reg(val); + w83627hf_write_value(data, W83637HF_REG_PWM_FREQ[nr - 1], + data->pwm_freq[nr - 1]); + } + + mutex_unlock(&data->update_lock); + return count; +} + +#define sysfs_pwm_freq(offset) \ +static ssize_t show_regs_pwm_freq_##offset(struct device *dev, \ + struct device_attribute *attr, char *buf) \ +{ \ + return show_pwm_freq_reg(dev, buf, offset); \ +} \ +static ssize_t \ +store_regs_pwm_freq_##offset(struct device *dev, \ + struct device_attribute *attr, const char *buf, size_t count) \ +{ \ + return store_pwm_freq_reg(dev, buf, count, offset); \ +} \ +static DEVICE_ATTR(pwm##offset##_freq, S_IRUGO | S_IWUSR, \ + show_regs_pwm_freq_##offset, store_regs_pwm_freq_##offset); + +sysfs_pwm_freq(1); +sysfs_pwm_freq(2); +sysfs_pwm_freq(3); + +static ssize_t show_sensor_reg(struct device *dev, char *buf, int nr) { struct w83627hf_data *data = w83627hf_update_device(dev); @@ -1077,6 +1191,9 @@ static struct attribute *w83627hf_attributes_opt[] = { &dev_attr_pwm3.attr, + &dev_attr_pwm1_freq.attr, + &dev_attr_pwm2_freq.attr, + &dev_attr_pwm3_freq.attr, NULL }; @@ -1139,7 +1256,9 @@ static int __devinit w83627hf_probe(struct platform_device *pdev) || (err = device_create_file(dev, &dev_attr_in5_max)) || (err = device_create_file(dev, &dev_attr_in6_input)) || (err = device_create_file(dev, &dev_attr_in6_min)) - || (err = device_create_file(dev, &dev_attr_in6_max))) + || (err = device_create_file(dev, &dev_attr_in6_max)) + || (err = device_create_file(dev, &dev_attr_pwm1_freq)) + || (err = device_create_file(dev, &dev_attr_pwm2_freq))) goto ERROR4; if (data->type != w83697hf) @@ -1169,6 +1288,12 @@ static int __devinit w83627hf_probe(struct platform_device *pdev) if ((err = device_create_file(dev, &dev_attr_pwm3))) goto ERROR4; + if (data->type == w83637hf || data->type == w83687thf) + if ((err = device_create_file(dev, &dev_attr_pwm1_freq)) + || (err = device_create_file(dev, &dev_attr_pwm2_freq)) + || (err = device_create_file(dev, &dev_attr_pwm3_freq))) + goto ERROR4; + data->class_dev = hwmon_device_register(dev); if (IS_ERR(data->class_dev)) { err = PTR_ERR(data->class_dev); @@ -1181,6 +1306,7 @@ static int __devinit w83627hf_probe(struct platform_device *pdev) sysfs_remove_group(&dev->kobj, &w83627hf_group); sysfs_remove_group(&dev->kobj, &w83627hf_group_opt); ERROR3: + platform_set_drvdata(pdev, NULL); kfree(data); ERROR1: release_region(res->start, WINB_REGION_SIZE); @@ -1193,11 +1319,11 @@ static int __devexit w83627hf_remove(struct platform_device *pdev) struct w83627hf_data *data = platform_get_drvdata(pdev); struct resource *res; - platform_set_drvdata(pdev, NULL); hwmon_device_unregister(data->class_dev); sysfs_remove_group(&pdev->dev.kobj, &w83627hf_group); sysfs_remove_group(&pdev->dev.kobj, &w83627hf_group_opt); + platform_set_drvdata(pdev, NULL); kfree(data); res = platform_get_resource(pdev, IORESOURCE_IO, 0); @@ -1472,6 +1598,20 @@ static struct w83627hf_data *w83627hf_update_device(struct device *dev) (data->type == w83627hf || data->type == w83697hf)) break; } + if (data->type == w83627hf) { + u8 tmp = w83627hf_read_value(data, + W83627HF_REG_PWM_FREQ); + data->pwm_freq[0] = tmp & 0x07; + data->pwm_freq[1] = (tmp >> 4) & 0x07; + } else if (data->type != w83627thf) { + for (i = 1; i <= 3; i++) { + data->pwm_freq[i - 1] = + w83627hf_read_value(data, + W83637HF_REG_PWM_FREQ[i - 1]); + if (i == 2 && (data->type == w83697hf)) + break; + } + } data->temp = w83627hf_read_value(data, W83781D_REG_TEMP(1)); data->temp_max = @@ -1548,15 +1688,12 @@ static int __init w83627hf_device_add(unsigned short address, goto exit_device_put; } - pdev->dev.platform_data = kmalloc(sizeof(struct w83627hf_sio_data), - GFP_KERNEL); - if (!pdev->dev.platform_data) { - err = -ENOMEM; + err = platform_device_add_data(pdev, sio_data, + sizeof(struct w83627hf_sio_data)); + if (err) { printk(KERN_ERR DRVNAME ": Platform data allocation failed\n"); goto exit_device_put; } - memcpy(pdev->dev.platform_data, sio_data, - sizeof(struct w83627hf_sio_data)); err = platform_device_add(pdev); if (err) { diff --git a/drivers/i2c/busses/Kconfig b/drivers/i2c/busses/Kconfig index 1c77e14480d..da1647869f9 100644 --- a/drivers/i2c/busses/Kconfig +++ b/drivers/i2c/busses/Kconfig @@ -237,9 +237,6 @@ config I2C_IOP3XX This driver can also be built as a module. If so, the module will be called i2c-iop3xx. -config I2C_ISA - tristate - config I2C_IXP4XX tristate "IXP4xx GPIO-Based I2C Interface (DEPRECATED)" depends on ARCH_IXP4XX diff --git a/drivers/i2c/busses/Makefile b/drivers/i2c/busses/Makefile index a6db4e38bda..5b752e4e191 100644 --- a/drivers/i2c/busses/Makefile +++ b/drivers/i2c/busses/Makefile @@ -18,7 +18,6 @@ obj-$(CONFIG_I2C_I801) += i2c-i801.o obj-$(CONFIG_I2C_I810) += i2c-i810.o obj-$(CONFIG_I2C_IBM_IIC) += i2c-ibm_iic.o obj-$(CONFIG_I2C_IOP3XX) += i2c-iop3xx.o -obj-$(CONFIG_I2C_ISA) += i2c-isa.o obj-$(CONFIG_I2C_IXP2000) += i2c-ixp2000.o obj-$(CONFIG_I2C_IXP4XX) += i2c-ixp4xx.o obj-$(CONFIG_I2C_POWERMAC) += i2c-powermac.o diff --git a/drivers/i2c/busses/i2c-isa.c b/drivers/i2c/busses/i2c-isa.c deleted file mode 100644 index b0e1370075d..00000000000 --- a/drivers/i2c/busses/i2c-isa.c +++ /dev/null @@ -1,192 +0,0 @@ -/* - i2c-isa.c - an i2c-core-like thing for ISA hardware monitoring chips - Copyright (C) 2005 Jean Delvare <khali@linux-fr.org> - - Based on the i2c-isa pseudo-adapter from the lm_sensors project - Copyright (c) 1998, 1999 Frodo Looijaard <frodol@dds.nl> - - 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. - - You should have received a copy of the GNU General Public License - along with this program; if not, write to the Free Software - Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA. -*/ - -/* This implements an i2c-core-like thing for ISA hardware monitoring - chips. Such chips are linked to the i2c subsystem for historical - reasons (because the early ISA hardware monitoring chips such as the - LM78 had both an I2C and an ISA interface). They used to be - registered with the main i2c-core, but as a first step in the - direction of a clean separation between I2C and ISA chip drivers, - we now have this separate core for ISA ones. It is significantly - more simple than the real one, of course, because we don't have to - handle multiple busses: there is only one (fake) ISA adapter. - It is worth noting that we still rely on i2c-core for some things - at the moment - but hopefully this won't last. */ - -#include <linux/init.h> -#include <linux/module.h> -#include <linux/kernel.h> -#include <linux/errno.h> -#include <linux/i2c.h> -#include <linux/i2c-isa.h> -#include <linux/platform_device.h> -#include <linux/completion.h> - -/* Exported by i2c-core for i2c-isa only */ -extern void i2c_adapter_dev_release(struct device *dev); -extern struct class i2c_adapter_class; - -static u32 isa_func(struct i2c_adapter *adapter); - -/* This is the actual algorithm we define */ -static const struct i2c_algorithm isa_algorithm = { - .functionality = isa_func, -}; - -/* There can only be one... */ -static struct i2c_adapter isa_adapter = { - .owner = THIS_MODULE, - .id = I2C_HW_ISA, - .class = I2C_CLASS_HWMON, - .algo = &isa_algorithm, - .name = "ISA main adapter", -}; - -/* We can't do a thing... */ -static u32 isa_func(struct i2c_adapter *adapter) -{ - return 0; -} - - -/* We implement an interface which resembles i2c_{add,del}_driver, - but for i2c-isa drivers. We don't have to remember and handle lists - of drivers and adapters so this is much more simple, of course. */ - -int i2c_isa_add_driver(struct i2c_driver *driver) -{ - int res; - - /* Add the driver to the list of i2c drivers in the driver core */ - driver->driver.bus = &i2c_bus_type; - res = driver_register(&driver->driver); - if (res) - return res; - dev_dbg(&isa_adapter.dev, "Driver %s registered\n", driver->driver.name); - - /* Now look for clients */ - res = driver->attach_adapter(&isa_adapter); - if (res) { - dev_dbg(&isa_adapter.dev, - "Driver %s failed to attach adapter, unregistering\n", - driver->driver.name); - driver_unregister(&driver->driver); - } - return res; -} - -int i2c_isa_del_driver(struct i2c_driver *driver) -{ - struct list_head *item, *_n; - struct i2c_client *client; - int res; - - /* Detach all clients belonging to this one driver */ - list_for_each_safe(item, _n, &isa_adapter.clients) { - client = list_entry(item, struct i2c_client, list); - if (client->driver != driver) - continue; - dev_dbg(&isa_adapter.dev, "Detaching client %s at 0x%x\n", - client->name, client->addr); - if ((res = driver->detach_client(client))) { - dev_err(&isa_adapter.dev, "Failed, driver " - "%s not unregistered!\n", - driver->driver.name); - return res; - } - } - - /* Get the driver off the core list */ - driver_unregister(&driver->driver); - dev_dbg(&isa_adapter.dev, "Driver %s unregistered\n", driver->driver.name); - - return 0; -} - - -static int __init i2c_isa_init(void) -{ - int err; - - mutex_init(&isa_adapter.clist_lock); - INIT_LIST_HEAD(&isa_adapter.clients); - - isa_adapter.nr = ANY_I2C_ISA_BUS; - isa_adapter.dev.parent = &platform_bus; - sprintf(isa_adapter.dev.bus_id, "i2c-%d", isa_adapter.nr); - isa_adapter.dev.release = &i2c_adapter_dev_release; - isa_adapter.dev.class = &i2c_adapter_class; - err = device_register(&isa_adapter.dev); - if (err) { - printk(KERN_ERR "i2c-isa: Failed to register device\n"); - goto exit; - } - - dev_dbg(&isa_adapter.dev, "%s registered\n", isa_adapter.name); - - return 0; - -exit: - return err; -} - -static void __exit i2c_isa_exit(void) -{ -#ifdef DEBUG - struct list_head *item, *_n; - struct i2c_client *client = NULL; -#endif - - /* There should be no more active client */ -#ifdef DEBUG - dev_dbg(&isa_adapter.dev, "Looking for clients\n"); - list_for_each_safe(item, _n, &isa_adapter.clients) { - client = list_entry(item, struct i2c_client, list); - dev_err(&isa_adapter.dev, "Driver %s still has an active " - "ISA client at 0x%x\n", client->driver->driver.name, - client->addr); - } - if (client != NULL) - return; -#endif - - /* Clean up the sysfs representation */ - dev_dbg(&isa_adapter.dev, "Unregistering from sysfs\n"); - init_completion(&isa_adapter.dev_released); - device_unregister(&isa_adapter.dev); - - /* Wait for sysfs to drop all references */ - dev_dbg(&isa_adapter.dev, "Waiting for sysfs completion\n"); - wait_for_completion(&isa_adapter.dev_released); - - dev_dbg(&isa_adapter.dev, "%s unregistered\n", isa_adapter.name); -} - -EXPORT_SYMBOL(i2c_isa_add_driver); -EXPORT_SYMBOL(i2c_isa_del_driver); - -MODULE_AUTHOR("Jean Delvare <khali@linux-fr.org>"); -MODULE_DESCRIPTION("ISA bus access through i2c"); -MODULE_LICENSE("GPL"); - -module_init(i2c_isa_init); -module_exit(i2c_isa_exit); diff --git a/drivers/i2c/i2c-core.c b/drivers/i2c/i2c-core.c index 6971a62397d..d663e6960d9 100644 --- a/drivers/i2c/i2c-core.c +++ b/drivers/i2c/i2c-core.c @@ -288,7 +288,6 @@ void i2c_adapter_dev_release(struct device *dev) struct i2c_adapter *adap = to_i2c_adapter(dev); complete(&adap->dev_released); } -EXPORT_SYMBOL_GPL(i2c_adapter_dev_release); /* exported to i2c-isa */ static ssize_t show_adapter_name(struct device *dev, struct device_attribute *attr, char *buf) @@ -307,7 +306,6 @@ struct class i2c_adapter_class = { .name = "i2c-adapter", .dev_attrs = i2c_adapter_attrs, }; -EXPORT_SYMBOL_GPL(i2c_adapter_class); /* exported to i2c-isa */ static void i2c_scan_static_board_info(struct i2c_adapter *adapter) { diff --git a/drivers/ide/ide-io.c b/drivers/ide/ide-io.c index c5b5011da56..f9de7984441 100644 --- a/drivers/ide/ide-io.c +++ b/drivers/ide/ide-io.c @@ -55,7 +55,7 @@ #include <asm/bitops.h> static int __ide_end_request(ide_drive_t *drive, struct request *rq, - int uptodate, int nr_sectors) + int uptodate, unsigned int nr_bytes) { int ret = 1; @@ -64,7 +64,7 @@ static int __ide_end_request(ide_drive_t *drive, struct request *rq, * complete the whole request right now */ if (blk_noretry_request(rq) && end_io_error(uptodate)) - nr_sectors = rq->hard_nr_sectors; + nr_bytes = rq->hard_nr_sectors << 9; if (!blk_fs_request(rq) && end_io_error(uptodate) && !rq->errors) rq->errors = -EIO; @@ -78,7 +78,7 @@ static int __ide_end_request(ide_drive_t *drive, struct request *rq, HWGROUP(drive)->hwif->ide_dma_on(drive); } - if (!end_that_request_first(rq, uptodate, nr_sectors)) { + if (!end_that_request_chunk(rq, uptodate, nr_bytes)) { add_disk_randomness(rq->rq_disk); if (!list_empty(&rq->queuelist)) blkdev_dequeue_request(rq); @@ -103,6 +103,7 @@ static int __ide_end_request(ide_drive_t *drive, struct request *rq, int ide_end_request (ide_drive_t *drive, int uptodate, int nr_sectors) { + unsigned int nr_bytes = nr_sectors << 9; struct request *rq; unsigned long flags; int ret = 1; @@ -114,10 +115,14 @@ int ide_end_request (ide_drive_t *drive, int uptodate, int nr_sectors) spin_lock_irqsave(&ide_lock, flags); rq = HWGROUP(drive)->rq; - if (!nr_sectors) - nr_sectors = rq->hard_cur_sectors; + if (!nr_bytes) { + if (blk_pc_request(rq)) + nr_bytes = rq->data_len; + else + nr_bytes = rq->hard_cur_sectors << 9; + } - ret = __ide_end_request(drive, rq, uptodate, nr_sectors); + ret = __ide_end_request(drive, rq, uptodate, nr_bytes); spin_unlock_irqrestore(&ide_lock, flags); return ret; diff --git a/drivers/ide/mips/swarm.c b/drivers/ide/mips/swarm.c index 6e935d7c63f..c2e29571b00 100644 --- a/drivers/ide/mips/swarm.c +++ b/drivers/ide/mips/swarm.c @@ -165,12 +165,11 @@ static int __devinit swarm_ide_init_module(void) goto out; } - if (!(pldev = kmalloc(sizeof (*pldev), GFP_KERNEL))) { + if (!(pldev = kzalloc(sizeof (*pldev), GFP_KERNEL))) { err = -ENOMEM; goto out_unregister_driver; } - memset (pldev, 0, sizeof (*pldev)); pldev->name = swarm_ide_string; pldev->id = 0; pldev->dev.release = swarm_ide_platform_release; diff --git a/drivers/infiniband/core/addr.c b/drivers/infiniband/core/addr.c index a91001c59b6..c5c33d35f87 100644 --- a/drivers/infiniband/core/addr.c +++ b/drivers/infiniband/core/addr.c @@ -295,10 +295,9 @@ int rdma_resolve_ip(struct rdma_addr_client *client, struct addr_req *req; int ret = 0; - req = kmalloc(sizeof *req, GFP_KERNEL); + req = kzalloc(sizeof *req, GFP_KERNEL); if (!req) return -ENOMEM; - memset(req, 0, sizeof *req); if (src_addr) memcpy(&req->src_addr, src_addr, ip_addr_size(src_addr)); diff --git a/drivers/infiniband/hw/cxgb3/iwch_cm.c b/drivers/infiniband/hw/cxgb3/iwch_cm.c index 5dc68cd5621..9574088f0d4 100644 --- a/drivers/infiniband/hw/cxgb3/iwch_cm.c +++ b/drivers/infiniband/hw/cxgb3/iwch_cm.c @@ -229,9 +229,8 @@ static void *alloc_ep(int size, gfp_t gfp) { struct iwch_ep_common *epc; - epc = kmalloc(size, gfp); + epc = kzalloc(size, gfp); if (epc) { - memset(epc, 0, size); kref_init(&epc->kref); spin_lock_init(&epc->lock); init_waitqueue_head(&epc->waitq); diff --git a/drivers/input/serio/ambakmi.c b/drivers/input/serio/ambakmi.c index 5a7b49c3553..b10ffae7c39 100644 --- a/drivers/input/serio/ambakmi.c +++ b/drivers/input/serio/ambakmi.c @@ -117,15 +117,13 @@ static int amba_kmi_probe(struct amba_device *dev, void *id) if (ret) return ret; - kmi = kmalloc(sizeof(struct amba_kmi_port), GFP_KERNEL); - io = kmalloc(sizeof(struct serio), GFP_KERNEL); + kmi = kzalloc(sizeof(struct amba_kmi_port), GFP_KERNEL); + io = kzalloc(sizeof(struct serio), GFP_KERNEL); if (!kmi || !io) { ret = -ENOMEM; goto out; } - memset(kmi, 0, sizeof(struct amba_kmi_port)); - memset(io, 0, sizeof(struct serio)); io->id.type = SERIO_8042; io->write = amba_kmi_write; diff --git a/drivers/input/serio/pcips2.c b/drivers/input/serio/pcips2.c index ea5e3c6ddb6..1b404f9e3bf 100644 --- a/drivers/input/serio/pcips2.c +++ b/drivers/input/serio/pcips2.c @@ -140,15 +140,13 @@ static int __devinit pcips2_probe(struct pci_dev *dev, const struct pci_device_i if (ret) goto disable; - ps2if = kmalloc(sizeof(struct pcips2_data), GFP_KERNEL); - serio = kmalloc(sizeof(struct serio), GFP_KERNEL); + ps2if = kzalloc(sizeof(struct pcips2_data), GFP_KERNEL); + serio = kzalloc(sizeof(struct serio), GFP_KERNEL); if (!ps2if || !serio) { ret = -ENOMEM; goto release; } - memset(ps2if, 0, sizeof(struct pcips2_data)); - memset(serio, 0, sizeof(struct serio)); serio->id.type = SERIO_8042; serio->write = pcips2_write; diff --git a/drivers/input/serio/sa1111ps2.c b/drivers/input/serio/sa1111ps2.c index d31ece8f68e..2ad88780a17 100644 --- a/drivers/input/serio/sa1111ps2.c +++ b/drivers/input/serio/sa1111ps2.c @@ -234,15 +234,13 @@ static int __devinit ps2_probe(struct sa1111_dev *dev) struct serio *serio; int ret; - ps2if = kmalloc(sizeof(struct ps2if), GFP_KERNEL); - serio = kmalloc(sizeof(struct serio), GFP_KERNEL); + ps2if = kzalloc(sizeof(struct ps2if), GFP_KERNEL); + serio = kzalloc(sizeof(struct serio), GFP_KERNEL); if (!ps2if || !serio) { ret = -ENOMEM; goto free; } - memset(ps2if, 0, sizeof(struct ps2if)); - memset(serio, 0, sizeof(struct serio)); serio->id.type = SERIO_8042; serio->write = ps2_write; diff --git a/drivers/isdn/sc/card.h b/drivers/isdn/sc/card.h index 4fbfa825c3a..5992f63c383 100644 --- a/drivers/isdn/sc/card.h +++ b/drivers/isdn/sc/card.h @@ -125,7 +125,7 @@ int sendmessage(int card, unsigned int procid, unsigned int type, int receivemessage(int card, RspMessage *rspmsg); int sc_ioctl(int card, scs_ioctl *data); int setup_buffers(int card, int c); -void check_reset(unsigned long data); +void sc_check_reset(unsigned long data); void check_phystat(unsigned long data); #endif /* CARD_H */ diff --git a/drivers/isdn/sc/command.c b/drivers/isdn/sc/command.c index b7bb7cbcf50..0e4969c2ef9 100644 --- a/drivers/isdn/sc/command.c +++ b/drivers/isdn/sc/command.c @@ -344,7 +344,7 @@ int reset(int card) spin_lock_irqsave(&sc_adapter[card]->lock, flags); init_timer(&sc_adapter[card]->reset_timer); - sc_adapter[card]->reset_timer.function = check_reset; + sc_adapter[card]->reset_timer.function = sc_check_reset; sc_adapter[card]->reset_timer.data = card; sc_adapter[card]->reset_timer.expires = jiffies + CHECKRESET_TIME; add_timer(&sc_adapter[card]->reset_timer); diff --git a/drivers/isdn/sc/timer.c b/drivers/isdn/sc/timer.c index cc1b8861be2..91fbe0dc28e 100644 --- a/drivers/isdn/sc/timer.c +++ b/drivers/isdn/sc/timer.c @@ -43,7 +43,7 @@ static void setup_ports(int card) * Then, check to see if the signate has been set. Next, set the * signature to a known value and issue a startproc if needed. */ -void check_reset(unsigned long data) +void sc_check_reset(unsigned long data) { unsigned long flags; unsigned long sig; diff --git a/drivers/kvm/Kconfig b/drivers/kvm/Kconfig index 33fa28a8c19..2f661e5f0da 100644 --- a/drivers/kvm/Kconfig +++ b/drivers/kvm/Kconfig @@ -11,7 +11,6 @@ if VIRTUALIZATION config KVM tristate "Kernel-based Virtual Machine (KVM) support" depends on X86 && EXPERIMENTAL - depends on X86_CMPXCHG64 || 64BIT ---help--- Support hosting fully virtualized guest machines using hardware virtualization extensions. You will need a fairly recent diff --git a/drivers/lguest/Kconfig b/drivers/lguest/Kconfig new file mode 100644 index 00000000000..43d901fdc77 --- /dev/null +++ b/drivers/lguest/Kconfig @@ -0,0 +1,20 @@ +config LGUEST + tristate "Linux hypervisor example code" + depends on X86 && PARAVIRT && NET && EXPERIMENTAL && !X86_PAE + select LGUEST_GUEST + select HVC_DRIVER + ---help--- + This is a very simple module which allows you to run + multiple instances of the same Linux kernel, using the + "lguest" command found in the Documentation/lguest directory. + Note that "lguest" is pronounced to rhyme with "fell quest", + not "rustyvisor". See Documentation/lguest/lguest.txt. + + If unsure, say N. If curious, say M. If masochistic, say Y. + +config LGUEST_GUEST + bool + help + The guest needs code built-in, even if the host has lguest + support as a module. The drivers are tiny, so we build them + in too. diff --git a/drivers/lguest/Makefile b/drivers/lguest/Makefile new file mode 100644 index 00000000000..55382c7d799 --- /dev/null +++ b/drivers/lguest/Makefile @@ -0,0 +1,7 @@ +# Guest requires the paravirt_ops replacement and the bus driver. +obj-$(CONFIG_LGUEST_GUEST) += lguest.o lguest_asm.o lguest_bus.o + +# Host requires the other files, which can be a module. +obj-$(CONFIG_LGUEST) += lg.o +lg-y := core.o hypercalls.o page_tables.o interrupts_and_traps.o \ + segments.o io.o lguest_user.o switcher.o diff --git a/drivers/lguest/core.c b/drivers/lguest/core.c new file mode 100644 index 00000000000..ce909ec5749 --- /dev/null +++ b/drivers/lguest/core.c @@ -0,0 +1,462 @@ +/* World's simplest hypervisor, to test paravirt_ops and show + * unbelievers that virtualization is the future. Plus, it's fun! */ +#include <linux/module.h> +#include <linux/stringify.h> +#include <linux/stddef.h> +#include <linux/io.h> +#include <linux/mm.h> +#include <linux/vmalloc.h> +#include <linux/cpu.h> +#include <linux/freezer.h> +#include <asm/paravirt.h> +#include <asm/desc.h> +#include <asm/pgtable.h> +#include <asm/uaccess.h> +#include <asm/poll.h> +#include <asm/highmem.h> +#include <asm/asm-offsets.h> +#include <asm/i387.h> +#include "lg.h" + +/* Found in switcher.S */ +extern char start_switcher_text[], end_switcher_text[], switch_to_guest[]; +extern unsigned long default_idt_entries[]; + +/* Every guest maps the core switcher code. */ +#define SHARED_SWITCHER_PAGES \ + DIV_ROUND_UP(end_switcher_text - start_switcher_text, PAGE_SIZE) +/* Pages for switcher itself, then two pages per cpu */ +#define TOTAL_SWITCHER_PAGES (SHARED_SWITCHER_PAGES + 2 * NR_CPUS) + +/* We map at -4M for ease of mapping into the guest (one PTE page). */ +#define SWITCHER_ADDR 0xFFC00000 + +static struct vm_struct *switcher_vma; +static struct page **switcher_page; + +static int cpu_had_pge; +static struct { + unsigned long offset; + unsigned short segment; +} lguest_entry; + +/* This One Big lock protects all inter-guest data structures. */ +DEFINE_MUTEX(lguest_lock); +static DEFINE_PER_CPU(struct lguest *, last_guest); + +/* FIXME: Make dynamic. */ +#define MAX_LGUEST_GUESTS 16 +struct lguest lguests[MAX_LGUEST_GUESTS]; + +/* Offset from where switcher.S was compiled to where we've copied it */ +static unsigned long switcher_offset(void) +{ + return SWITCHER_ADDR - (unsigned long)start_switcher_text; +} + +/* This cpu's struct lguest_pages. */ +static struct lguest_pages *lguest_pages(unsigned int cpu) +{ + return &(((struct lguest_pages *) + (SWITCHER_ADDR + SHARED_SWITCHER_PAGES*PAGE_SIZE))[cpu]); +} + +static __init int map_switcher(void) +{ + int i, err; + struct page **pagep; + + switcher_page = kmalloc(sizeof(switcher_page[0])*TOTAL_SWITCHER_PAGES, + GFP_KERNEL); + if (!switcher_page) { + err = -ENOMEM; + goto out; + } + + for (i = 0; i < TOTAL_SWITCHER_PAGES; i++) { + unsigned long addr = get_zeroed_page(GFP_KERNEL); + if (!addr) { + err = -ENOMEM; + goto free_some_pages; + } + switcher_page[i] = virt_to_page(addr); + } + + switcher_vma = __get_vm_area(TOTAL_SWITCHER_PAGES * PAGE_SIZE, + VM_ALLOC, SWITCHER_ADDR, VMALLOC_END); + if (!switcher_vma) { + err = -ENOMEM; + printk("lguest: could not map switcher pages high\n"); + goto free_pages; + } + + pagep = switcher_page; + err = map_vm_area(switcher_vma, PAGE_KERNEL, &pagep); + if (err) { + printk("lguest: map_vm_area failed: %i\n", err); + goto free_vma; + } + memcpy(switcher_vma->addr, start_switcher_text, + end_switcher_text - start_switcher_text); + + /* Fix up IDT entries to point into copied text. */ + for (i = 0; i < IDT_ENTRIES; i++) + default_idt_entries[i] += switcher_offset(); + + for_each_possible_cpu(i) { + struct lguest_pages *pages = lguest_pages(i); + struct lguest_ro_state *state = &pages->state; + + /* These fields are static: rest done in copy_in_guest_info */ + state->host_gdt_desc.size = GDT_SIZE-1; + state->host_gdt_desc.address = (long)get_cpu_gdt_table(i); + store_idt(&state->host_idt_desc); + state->guest_idt_desc.size = sizeof(state->guest_idt)-1; + state->guest_idt_desc.address = (long)&state->guest_idt; + state->guest_gdt_desc.size = sizeof(state->guest_gdt)-1; + state->guest_gdt_desc.address = (long)&state->guest_gdt; + state->guest_tss.esp0 = (long)(&pages->regs + 1); + state->guest_tss.ss0 = LGUEST_DS; + /* No I/O for you! */ + state->guest_tss.io_bitmap_base = sizeof(state->guest_tss); + setup_default_gdt_entries(state); + setup_default_idt_entries(state, default_idt_entries); + + /* Setup LGUEST segments on all cpus */ + get_cpu_gdt_table(i)[GDT_ENTRY_LGUEST_CS] = FULL_EXEC_SEGMENT; + get_cpu_gdt_table(i)[GDT_ENTRY_LGUEST_DS] = FULL_SEGMENT; + } + + /* Initialize entry point into switcher. */ + lguest_entry.offset = (long)switch_to_guest + switcher_offset(); + lguest_entry.segment = LGUEST_CS; + + printk(KERN_INFO "lguest: mapped switcher at %p\n", + switcher_vma->addr); + return 0; + +free_vma: + vunmap(switcher_vma->addr); +free_pages: + i = TOTAL_SWITCHER_PAGES; +free_some_pages: + for (--i; i >= 0; i--) + __free_pages(switcher_page[i], 0); + kfree(switcher_page); +out: + return err; +} + +static void unmap_switcher(void) +{ + unsigned int i; + + vunmap(switcher_vma->addr); + for (i = 0; i < TOTAL_SWITCHER_PAGES; i++) + __free_pages(switcher_page[i], 0); +} + +/* IN/OUT insns: enough to get us past boot-time probing. */ +static int emulate_insn(struct lguest *lg) +{ + u8 insn; + unsigned int insnlen = 0, in = 0, shift = 0; + unsigned long physaddr = guest_pa(lg, lg->regs->eip); + + /* This only works for addresses in linear mapping... */ + if (lg->regs->eip < lg->page_offset) + return 0; + lgread(lg, &insn, physaddr, 1); + + /* Operand size prefix means it's actually for ax. */ + if (insn == 0x66) { + shift = 16; + insnlen = 1; + lgread(lg, &insn, physaddr + insnlen, 1); + } + + switch (insn & 0xFE) { + case 0xE4: /* in <next byte>,%al */ + insnlen += 2; + in = 1; + break; + case 0xEC: /* in (%dx),%al */ + insnlen += 1; + in = 1; + break; + case 0xE6: /* out %al,<next byte> */ + insnlen += 2; + break; + case 0xEE: /* out %al,(%dx) */ + insnlen += 1; + break; + default: + return 0; + } + + if (in) { + /* Lower bit tells is whether it's a 16 or 32 bit access */ + if (insn & 0x1) + lg->regs->eax = 0xFFFFFFFF; + else + lg->regs->eax |= (0xFFFF << shift); + } + lg->regs->eip += insnlen; + return 1; +} + +int lguest_address_ok(const struct lguest *lg, + unsigned long addr, unsigned long len) +{ + return (addr+len) / PAGE_SIZE < lg->pfn_limit && (addr+len >= addr); +} + +/* Just like get_user, but don't let guest access lguest binary. */ +u32 lgread_u32(struct lguest *lg, unsigned long addr) +{ + u32 val = 0; + + /* Don't let them access lguest binary */ + if (!lguest_address_ok(lg, addr, sizeof(val)) + || get_user(val, (u32 __user *)addr) != 0) + kill_guest(lg, "bad read address %#lx", addr); + return val; +} + +void lgwrite_u32(struct lguest *lg, unsigned long addr, u32 val) +{ + if (!lguest_address_ok(lg, addr, sizeof(val)) + || put_user(val, (u32 __user *)addr) != 0) + kill_guest(lg, "bad write address %#lx", addr); +} + +void lgread(struct lguest *lg, void *b, unsigned long addr, unsigned bytes) +{ + if (!lguest_address_ok(lg, addr, bytes) + || copy_from_user(b, (void __user *)addr, bytes) != 0) { + /* copy_from_user should do this, but as we rely on it... */ + memset(b, 0, bytes); + kill_guest(lg, "bad read address %#lx len %u", addr, bytes); + } +} + +void lgwrite(struct lguest *lg, unsigned long addr, const void *b, + unsigned bytes) +{ + if (!lguest_address_ok(lg, addr, bytes) + || copy_to_user((void __user *)addr, b, bytes) != 0) + kill_guest(lg, "bad write address %#lx len %u", addr, bytes); +} + +static void set_ts(void) +{ + u32 cr0; + + cr0 = read_cr0(); + if (!(cr0 & 8)) + write_cr0(cr0|8); +} + +static void copy_in_guest_info(struct lguest *lg, struct lguest_pages *pages) +{ + if (__get_cpu_var(last_guest) != lg || lg->last_pages != pages) { + __get_cpu_var(last_guest) = lg; + lg->last_pages = pages; + lg->changed = CHANGED_ALL; + } + + /* These are pretty cheap, so we do them unconditionally. */ + pages->state.host_cr3 = __pa(current->mm->pgd); + map_switcher_in_guest(lg, pages); + pages->state.guest_tss.esp1 = lg->esp1; + pages->state.guest_tss.ss1 = lg->ss1; + + /* Copy direct trap entries. */ + if (lg->changed & CHANGED_IDT) + copy_traps(lg, pages->state.guest_idt, default_idt_entries); + + /* Copy all GDT entries but the TSS. */ + if (lg->changed & CHANGED_GDT) + copy_gdt(lg, pages->state.guest_gdt); + /* If only the TLS entries have changed, copy them. */ + else if (lg->changed & CHANGED_GDT_TLS) + copy_gdt_tls(lg, pages->state.guest_gdt); + + lg->changed = 0; +} + +static void run_guest_once(struct lguest *lg, struct lguest_pages *pages) +{ + unsigned int clobber; + + copy_in_guest_info(lg, pages); + + /* Put eflags on stack, lcall does rest: suitable for iret return. */ + asm volatile("pushf; lcall *lguest_entry" + : "=a"(clobber), "=b"(clobber) + : "0"(pages), "1"(__pa(lg->pgdirs[lg->pgdidx].pgdir)) + : "memory", "%edx", "%ecx", "%edi", "%esi"); +} + +int run_guest(struct lguest *lg, unsigned long __user *user) +{ + while (!lg->dead) { + unsigned int cr2 = 0; /* Damn gcc */ + + /* Hypercalls first: we might have been out to userspace */ + do_hypercalls(lg); + if (lg->dma_is_pending) { + if (put_user(lg->pending_dma, user) || + put_user(lg->pending_key, user+1)) + return -EFAULT; + return sizeof(unsigned long)*2; + } + + if (signal_pending(current)) + return -ERESTARTSYS; + + /* If Waker set break_out, return to Launcher. */ + if (lg->break_out) + return -EAGAIN; + + maybe_do_interrupt(lg); + + try_to_freeze(); + + if (lg->dead) + break; + + if (lg->halted) { + set_current_state(TASK_INTERRUPTIBLE); + schedule(); + continue; + } + + local_irq_disable(); + + /* Even if *we* don't want FPU trap, guest might... */ + if (lg->ts) + set_ts(); + + /* Don't let Guest do SYSENTER: we can't handle it. */ + if (boot_cpu_has(X86_FEATURE_SEP)) + wrmsr(MSR_IA32_SYSENTER_CS, 0, 0); + + run_guest_once(lg, lguest_pages(raw_smp_processor_id())); + + /* Save cr2 now if we page-faulted. */ + if (lg->regs->trapnum == 14) + cr2 = read_cr2(); + else if (lg->regs->trapnum == 7) + math_state_restore(); + + if (boot_cpu_has(X86_FEATURE_SEP)) + wrmsr(MSR_IA32_SYSENTER_CS, __KERNEL_CS, 0); + local_irq_enable(); + + switch (lg->regs->trapnum) { + case 13: /* We've intercepted a GPF. */ + if (lg->regs->errcode == 0) { + if (emulate_insn(lg)) + continue; + } + break; + case 14: /* We've intercepted a page fault. */ + if (demand_page(lg, cr2, lg->regs->errcode)) + continue; + + /* If lguest_data is NULL, this won't hurt. */ + if (put_user(cr2, &lg->lguest_data->cr2)) + kill_guest(lg, "Writing cr2"); + break; + case 7: /* We've intercepted a Device Not Available fault. */ + /* If they don't want to know, just absorb it. */ + if (!lg->ts) + continue; + break; + case 32 ... 255: /* Real interrupt, fall thru */ + cond_resched(); + case LGUEST_TRAP_ENTRY: /* Handled at top of loop */ + continue; + } + + if (deliver_trap(lg, lg->regs->trapnum)) + continue; + + kill_guest(lg, "unhandled trap %li at %#lx (%#lx)", + lg->regs->trapnum, lg->regs->eip, + lg->regs->trapnum == 14 ? cr2 : lg->regs->errcode); + } + return -ENOENT; +} + +int find_free_guest(void) +{ + unsigned int i; + for (i = 0; i < MAX_LGUEST_GUESTS; i++) + if (!lguests[i].tsk) + return i; + return -1; +} + +static void adjust_pge(void *on) +{ + if (on) + write_cr4(read_cr4() | X86_CR4_PGE); + else + write_cr4(read_cr4() & ~X86_CR4_PGE); +} + +static int __init init(void) +{ + int err; + + if (paravirt_enabled()) { + printk("lguest is afraid of %s\n", paravirt_ops.name); + return -EPERM; + } + + err = map_switcher(); + if (err) + return err; + + err = init_pagetables(switcher_page, SHARED_SWITCHER_PAGES); + if (err) { + unmap_switcher(); + return err; + } + lguest_io_init(); + + err = lguest_device_init(); + if (err) { + free_pagetables(); + unmap_switcher(); + return err; + } + lock_cpu_hotplug(); + if (cpu_has_pge) { /* We have a broader idea of "global". */ + cpu_had_pge = 1; + on_each_cpu(adjust_pge, (void *)0, 0, 1); + clear_bit(X86_FEATURE_PGE, boot_cpu_data.x86_capability); + } + unlock_cpu_hotplug(); + return 0; +} + +static void __exit fini(void) +{ + lguest_device_remove(); + free_pagetables(); + unmap_switcher(); + lock_cpu_hotplug(); + if (cpu_had_pge) { + set_bit(X86_FEATURE_PGE, boot_cpu_data.x86_capability); + on_each_cpu(adjust_pge, (void *)1, 0, 1); + } + unlock_cpu_hotplug(); +} + +module_init(init); +module_exit(fini); +MODULE_LICENSE("GPL"); +MODULE_AUTHOR("Rusty Russell <rusty@rustcorp.com.au>"); diff --git a/drivers/lguest/hypercalls.c b/drivers/lguest/hypercalls.c new file mode 100644 index 00000000000..ea52ca451f7 --- /dev/null +++ b/drivers/lguest/hypercalls.c @@ -0,0 +1,192 @@ +/* Actual hypercalls, which allow guests to actually do something. + Copyright (C) 2006 Rusty Russell IBM Corporation + + 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. + + You should have received a copy of the GNU General Public License + along with this program; if not, write to the Free Software + Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA +*/ +#include <linux/uaccess.h> +#include <linux/syscalls.h> +#include <linux/mm.h> +#include <asm/page.h> +#include <asm/pgtable.h> +#include <irq_vectors.h> +#include "lg.h" + +static void do_hcall(struct lguest *lg, struct lguest_regs *regs) +{ + switch (regs->eax) { + case LHCALL_FLUSH_ASYNC: + break; + case LHCALL_LGUEST_INIT: + kill_guest(lg, "already have lguest_data"); + break; + case LHCALL_CRASH: { + char msg[128]; + lgread(lg, msg, regs->edx, sizeof(msg)); + msg[sizeof(msg)-1] = '\0'; + kill_guest(lg, "CRASH: %s", msg); + break; + } + case LHCALL_FLUSH_TLB: + if (regs->edx) + guest_pagetable_clear_all(lg); + else + guest_pagetable_flush_user(lg); + break; + case LHCALL_GET_WALLCLOCK: { + struct timespec ts; + ktime_get_real_ts(&ts); + regs->eax = ts.tv_sec; + break; + } + case LHCALL_BIND_DMA: + regs->eax = bind_dma(lg, regs->edx, regs->ebx, + regs->ecx >> 8, regs->ecx & 0xFF); + break; + case LHCALL_SEND_DMA: + send_dma(lg, regs->edx, regs->ebx); + break; + case LHCALL_LOAD_GDT: + load_guest_gdt(lg, regs->edx, regs->ebx); + break; + case LHCALL_LOAD_IDT_ENTRY: + load_guest_idt_entry(lg, regs->edx, regs->ebx, regs->ecx); + break; + case LHCALL_NEW_PGTABLE: + guest_new_pagetable(lg, regs->edx); + break; + case LHCALL_SET_STACK: + guest_set_stack(lg, regs->edx, regs->ebx, regs->ecx); + break; + case LHCALL_SET_PTE: + guest_set_pte(lg, regs->edx, regs->ebx, mkgpte(regs->ecx)); + break; + case LHCALL_SET_PMD: + guest_set_pmd(lg, regs->edx, regs->ebx); + break; + case LHCALL_LOAD_TLS: + guest_load_tls(lg, regs->edx); + break; + case LHCALL_SET_CLOCKEVENT: + guest_set_clockevent(lg, regs->edx); + break; + case LHCALL_TS: + lg->ts = regs->edx; + break; + case LHCALL_HALT: + lg->halted = 1; + break; + default: + kill_guest(lg, "Bad hypercall %li\n", regs->eax); + } +} + +/* We always do queued calls before actual hypercall. */ +static void do_async_hcalls(struct lguest *lg) +{ + unsigned int i; + u8 st[LHCALL_RING_SIZE]; + + if (copy_from_user(&st, &lg->lguest_data->hcall_status, sizeof(st))) + return; + + for (i = 0; i < ARRAY_SIZE(st); i++) { + struct lguest_regs regs; + unsigned int n = lg->next_hcall; + + if (st[n] == 0xFF) + break; + + if (++lg->next_hcall == LHCALL_RING_SIZE) + lg->next_hcall = 0; + + if (get_user(regs.eax, &lg->lguest_data->hcalls[n].eax) + || get_user(regs.edx, &lg->lguest_data->hcalls[n].edx) + || get_user(regs.ecx, &lg->lguest_data->hcalls[n].ecx) + || get_user(regs.ebx, &lg->lguest_data->hcalls[n].ebx)) { + kill_guest(lg, "Fetching async hypercalls"); + break; + } + + do_hcall(lg, ®s); + if (put_user(0xFF, &lg->lguest_data->hcall_status[n])) { + kill_guest(lg, "Writing result for async hypercall"); + break; + } + + if (lg->dma_is_pending) + break; + } +} + +static void initialize(struct lguest *lg) +{ + u32 tsc_speed; + + if (lg->regs->eax != LHCALL_LGUEST_INIT) { + kill_guest(lg, "hypercall %li before LGUEST_INIT", + lg->regs->eax); + return; + } + + /* We only tell the guest to use the TSC if it's reliable. */ + if (boot_cpu_has(X86_FEATURE_CONSTANT_TSC) && !check_tsc_unstable()) + tsc_speed = tsc_khz; + else + tsc_speed = 0; + + lg->lguest_data = (struct lguest_data __user *)lg->regs->edx; + /* We check here so we can simply copy_to_user/from_user */ + if (!lguest_address_ok(lg, lg->regs->edx, sizeof(*lg->lguest_data))) { + kill_guest(lg, "bad guest page %p", lg->lguest_data); + return; + } + if (get_user(lg->noirq_start, &lg->lguest_data->noirq_start) + || get_user(lg->noirq_end, &lg->lguest_data->noirq_end) + /* We reserve the top pgd entry. */ + || put_user(4U*1024*1024, &lg->lguest_data->reserve_mem) + || put_user(tsc_speed, &lg->lguest_data->tsc_khz) + || put_user(lg->guestid, &lg->lguest_data->guestid)) + kill_guest(lg, "bad guest page %p", lg->lguest_data); + + /* This is the one case where the above accesses might have + * been the first write to a Guest page. This may have caused + * a copy-on-write fault, but the Guest might be referring to + * the old (read-only) page. */ + guest_pagetable_clear_all(lg); +} + +/* Even if we go out to userspace and come back, we don't want to do + * the hypercall again. */ +static void clear_hcall(struct lguest *lg) +{ + lg->regs->trapnum = 255; +} + +void do_hypercalls(struct lguest *lg) +{ + if (unlikely(!lg->lguest_data)) { + if (lg->regs->trapnum == LGUEST_TRAP_ENTRY) { + initialize(lg); + clear_hcall(lg); + } + return; + } + + do_async_hcalls(lg); + if (!lg->dma_is_pending && lg->regs->trapnum == LGUEST_TRAP_ENTRY) { + do_hcall(lg, lg->regs); + clear_hcall(lg); + } +} diff --git a/drivers/lguest/interrupts_and_traps.c b/drivers/lguest/interrupts_and_traps.c new file mode 100644 index 00000000000..d9de5bbc613 --- /dev/null +++ b/drivers/lguest/interrupts_and_traps.c @@ -0,0 +1,268 @@ +#include <linux/uaccess.h> +#include "lg.h" + +static unsigned long idt_address(u32 lo, u32 hi) +{ + return (lo & 0x0000FFFF) | (hi & 0xFFFF0000); +} + +static int idt_type(u32 lo, u32 hi) +{ + return (hi >> 8) & 0xF; +} + +static int idt_present(u32 lo, u32 hi) +{ + return (hi & 0x8000); +} + +static void push_guest_stack(struct lguest *lg, unsigned long *gstack, u32 val) +{ + *gstack -= 4; + lgwrite_u32(lg, *gstack, val); +} + +static void set_guest_interrupt(struct lguest *lg, u32 lo, u32 hi, int has_err) +{ + unsigned long gstack; + u32 eflags, ss, irq_enable; + + /* If they want a ring change, we use new stack and push old ss/esp */ + if ((lg->regs->ss&0x3) != GUEST_PL) { + gstack = guest_pa(lg, lg->esp1); + ss = lg->ss1; + push_guest_stack(lg, &gstack, lg->regs->ss); + push_guest_stack(lg, &gstack, lg->regs->esp); + } else { + gstack = guest_pa(lg, lg->regs->esp); + ss = lg->regs->ss; + } + + /* We use IF bit in eflags to indicate whether irqs were disabled + (it's always 0, since irqs are enabled when guest is running). */ + eflags = lg->regs->eflags; + if (get_user(irq_enable, &lg->lguest_data->irq_enabled)) + irq_enable = 0; + eflags |= (irq_enable & X86_EFLAGS_IF); + + push_guest_stack(lg, &gstack, eflags); + push_guest_stack(lg, &gstack, lg->regs->cs); + push_guest_stack(lg, &gstack, lg->regs->eip); + + if (has_err) + push_guest_stack(lg, &gstack, lg->regs->errcode); + + /* Change the real stack so switcher returns to trap handler */ + lg->regs->ss = ss; + lg->regs->esp = gstack + lg->page_offset; + lg->regs->cs = (__KERNEL_CS|GUEST_PL); + lg->regs->eip = idt_address(lo, hi); + + /* Disable interrupts for an interrupt gate. */ + if (idt_type(lo, hi) == 0xE) + if (put_user(0, &lg->lguest_data->irq_enabled)) + kill_guest(lg, "Disabling interrupts"); +} + +void maybe_do_interrupt(struct lguest *lg) +{ + unsigned int irq; + DECLARE_BITMAP(blk, LGUEST_IRQS); + struct desc_struct *idt; + + if (!lg->lguest_data) + return; + + /* Mask out any interrupts they have blocked. */ + if (copy_from_user(&blk, lg->lguest_data->blocked_interrupts, + sizeof(blk))) + return; + + bitmap_andnot(blk, lg->irqs_pending, blk, LGUEST_IRQS); + + irq = find_first_bit(blk, LGUEST_IRQS); + if (irq >= LGUEST_IRQS) + return; + + if (lg->regs->eip >= lg->noirq_start && lg->regs->eip < lg->noirq_end) + return; + + /* If they're halted, we re-enable interrupts. */ + if (lg->halted) { + /* Re-enable interrupts. */ + if (put_user(X86_EFLAGS_IF, &lg->lguest_data->irq_enabled)) + kill_guest(lg, "Re-enabling interrupts"); + lg->halted = 0; + } else { + /* Maybe they have interrupts disabled? */ + u32 irq_enabled; + if (get_user(irq_enabled, &lg->lguest_data->irq_enabled)) + irq_enabled = 0; + if (!irq_enabled) + return; + } + + idt = &lg->idt[FIRST_EXTERNAL_VECTOR+irq]; + if (idt_present(idt->a, idt->b)) { + clear_bit(irq, lg->irqs_pending); + set_guest_interrupt(lg, idt->a, idt->b, 0); + } +} + +static int has_err(unsigned int trap) +{ + return (trap == 8 || (trap >= 10 && trap <= 14) || trap == 17); +} + +int deliver_trap(struct lguest *lg, unsigned int num) +{ + u32 lo = lg->idt[num].a, hi = lg->idt[num].b; + + if (!idt_present(lo, hi)) + return 0; + set_guest_interrupt(lg, lo, hi, has_err(num)); + return 1; +} + +static int direct_trap(const struct lguest *lg, + const struct desc_struct *trap, + unsigned int num) +{ + /* Hardware interrupts don't go to guest (except syscall). */ + if (num >= FIRST_EXTERNAL_VECTOR && num != SYSCALL_VECTOR) + return 0; + + /* We intercept page fault (demand shadow paging & cr2 saving) + protection fault (in/out emulation) and device not + available (TS handling), and hypercall */ + if (num == 14 || num == 13 || num == 7 || num == LGUEST_TRAP_ENTRY) + return 0; + + /* Interrupt gates (0xE) or not present (0x0) can't go direct. */ + return idt_type(trap->a, trap->b) == 0xF; +} + +void pin_stack_pages(struct lguest *lg) +{ + unsigned int i; + + for (i = 0; i < lg->stack_pages; i++) + pin_page(lg, lg->esp1 - i * PAGE_SIZE); +} + +void guest_set_stack(struct lguest *lg, u32 seg, u32 esp, unsigned int pages) +{ + /* You cannot have a stack segment with priv level 0. */ + if ((seg & 0x3) != GUEST_PL) + kill_guest(lg, "bad stack segment %i", seg); + if (pages > 2) + kill_guest(lg, "bad stack pages %u", pages); + lg->ss1 = seg; + lg->esp1 = esp; + lg->stack_pages = pages; + pin_stack_pages(lg); +} + +/* Set up trap in IDT. */ +static void set_trap(struct lguest *lg, struct desc_struct *trap, + unsigned int num, u32 lo, u32 hi) +{ + u8 type = idt_type(lo, hi); + + if (!idt_present(lo, hi)) { + trap->a = trap->b = 0; + return; + } + + if (type != 0xE && type != 0xF) + kill_guest(lg, "bad IDT type %i", type); + + trap->a = ((__KERNEL_CS|GUEST_PL)<<16) | (lo&0x0000FFFF); + trap->b = (hi&0xFFFFEF00); +} + +void load_guest_idt_entry(struct lguest *lg, unsigned int num, u32 lo, u32 hi) +{ + /* Guest never handles: NMI, doublefault, hypercall, spurious irq. */ + if (num == 2 || num == 8 || num == 15 || num == LGUEST_TRAP_ENTRY) + return; + + lg->changed |= CHANGED_IDT; + if (num < ARRAY_SIZE(lg->idt)) + set_trap(lg, &lg->idt[num], num, lo, hi); + else if (num == SYSCALL_VECTOR) + set_trap(lg, &lg->syscall_idt, num, lo, hi); +} + +static void default_idt_entry(struct desc_struct *idt, + int trap, + const unsigned long handler) +{ + u32 flags = 0x8e00; + + /* They can't "int" into any of them except hypercall. */ + if (trap == LGUEST_TRAP_ENTRY) + flags |= (GUEST_PL << 13); + + idt->a = (LGUEST_CS<<16) | (handler&0x0000FFFF); + idt->b = (handler&0xFFFF0000) | flags; +} + +void setup_default_idt_entries(struct lguest_ro_state *state, + const unsigned long *def) +{ + unsigned int i; + + for (i = 0; i < ARRAY_SIZE(state->guest_idt); i++) + default_idt_entry(&state->guest_idt[i], i, def[i]); +} + +void copy_traps(const struct lguest *lg, struct desc_struct *idt, + const unsigned long *def) +{ + unsigned int i; + + /* All hardware interrupts are same whatever the guest: only the + * traps might be different. */ + for (i = 0; i < FIRST_EXTERNAL_VECTOR; i++) { + if (direct_trap(lg, &lg->idt[i], i)) + idt[i] = lg->idt[i]; + else + default_idt_entry(&idt[i], i, def[i]); + } + i = SYSCALL_VECTOR; + if (direct_trap(lg, &lg->syscall_idt, i)) + idt[i] = lg->syscall_idt; + else + default_idt_entry(&idt[i], i, def[i]); +} + +void guest_set_clockevent(struct lguest *lg, unsigned long delta) +{ + ktime_t expires; + + if (unlikely(delta == 0)) { + /* Clock event device is shutting down. */ + hrtimer_cancel(&lg->hrt); + return; + } + + expires = ktime_add_ns(ktime_get_real(), delta); + hrtimer_start(&lg->hrt, expires, HRTIMER_MODE_ABS); +} + +static enum hrtimer_restart clockdev_fn(struct hrtimer *timer) +{ + struct lguest *lg = container_of(timer, struct lguest, hrt); + + set_bit(0, lg->irqs_pending); + if (lg->halted) + wake_up_process(lg->tsk); + return HRTIMER_NORESTART; +} + +void init_clockdev(struct lguest *lg) +{ + hrtimer_init(&lg->hrt, CLOCK_REALTIME, HRTIMER_MODE_ABS); + lg->hrt.function = clockdev_fn; +} diff --git a/drivers/lguest/io.c b/drivers/lguest/io.c new file mode 100644 index 00000000000..06bdba2337e --- /dev/null +++ b/drivers/lguest/io.c @@ -0,0 +1,399 @@ +/* Simple I/O model for guests, based on shared memory. + * Copyright (C) 2006 Rusty Russell IBM Corporation + * + * 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. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA + */ +#include <linux/types.h> +#include <linux/futex.h> +#include <linux/jhash.h> +#include <linux/mm.h> +#include <linux/highmem.h> +#include <linux/uaccess.h> +#include "lg.h" + +static struct list_head dma_hash[61]; + +void lguest_io_init(void) +{ + unsigned int i; + + for (i = 0; i < ARRAY_SIZE(dma_hash); i++) + INIT_LIST_HEAD(&dma_hash[i]); +} + +/* FIXME: allow multi-page lengths. */ +static int check_dma_list(struct lguest *lg, const struct lguest_dma *dma) +{ + unsigned int i; + + for (i = 0; i < LGUEST_MAX_DMA_SECTIONS; i++) { + if (!dma->len[i]) + return 1; + if (!lguest_address_ok(lg, dma->addr[i], dma->len[i])) + goto kill; + if (dma->len[i] > PAGE_SIZE) + goto kill; + /* We could do over a page, but is it worth it? */ + if ((dma->addr[i] % PAGE_SIZE) + dma->len[i] > PAGE_SIZE) + goto kill; + } + return 1; + +kill: + kill_guest(lg, "bad DMA entry: %u@%#lx", dma->len[i], dma->addr[i]); + return 0; +} + +static unsigned int hash(const union futex_key *key) +{ + return jhash2((u32*)&key->both.word, + (sizeof(key->both.word)+sizeof(key->both.ptr))/4, + key->both.offset) + % ARRAY_SIZE(dma_hash); +} + +static inline int key_eq(const union futex_key *a, const union futex_key *b) +{ + return (a->both.word == b->both.word + && a->both.ptr == b->both.ptr + && a->both.offset == b->both.offset); +} + +/* Must hold read lock on dmainfo owner's current->mm->mmap_sem */ +static void unlink_dma(struct lguest_dma_info *dmainfo) +{ + BUG_ON(!mutex_is_locked(&lguest_lock)); + dmainfo->interrupt = 0; + list_del(&dmainfo->list); + drop_futex_key_refs(&dmainfo->key); +} + +static int unbind_dma(struct lguest *lg, + const union futex_key *key, + unsigned long dmas) +{ + int i, ret = 0; + + for (i = 0; i < LGUEST_MAX_DMA; i++) { + if (key_eq(key, &lg->dma[i].key) && dmas == lg->dma[i].dmas) { + unlink_dma(&lg->dma[i]); + ret = 1; + break; + } + } + return ret; +} + +int bind_dma(struct lguest *lg, + unsigned long ukey, unsigned long dmas, u16 numdmas, u8 interrupt) +{ + unsigned int i; + int ret = 0; + union futex_key key; + struct rw_semaphore *fshared = ¤t->mm->mmap_sem; + + if (interrupt >= LGUEST_IRQS) + return 0; + + mutex_lock(&lguest_lock); + down_read(fshared); + if (get_futex_key((u32 __user *)ukey, fshared, &key) != 0) { + kill_guest(lg, "bad dma key %#lx", ukey); + goto unlock; + } + get_futex_key_refs(&key); + + if (interrupt == 0) + ret = unbind_dma(lg, &key, dmas); + else { + for (i = 0; i < LGUEST_MAX_DMA; i++) { + if (lg->dma[i].interrupt) + continue; + + lg->dma[i].dmas = dmas; + lg->dma[i].num_dmas = numdmas; + lg->dma[i].next_dma = 0; + lg->dma[i].key = key; + lg->dma[i].guestid = lg->guestid; + lg->dma[i].interrupt = interrupt; + list_add(&lg->dma[i].list, &dma_hash[hash(&key)]); + ret = 1; + goto unlock; + } + } + drop_futex_key_refs(&key); +unlock: + up_read(fshared); + mutex_unlock(&lguest_lock); + return ret; +} + +/* lgread from another guest */ +static int lgread_other(struct lguest *lg, + void *buf, u32 addr, unsigned bytes) +{ + if (!lguest_address_ok(lg, addr, bytes) + || access_process_vm(lg->tsk, addr, buf, bytes, 0) != bytes) { + memset(buf, 0, bytes); + kill_guest(lg, "bad address in registered DMA struct"); + return 0; + } + return 1; +} + +/* lgwrite to another guest */ +static int lgwrite_other(struct lguest *lg, u32 addr, + const void *buf, unsigned bytes) +{ + if (!lguest_address_ok(lg, addr, bytes) + || (access_process_vm(lg->tsk, addr, (void *)buf, bytes, 1) + != bytes)) { + kill_guest(lg, "bad address writing to registered DMA"); + return 0; + } + return 1; +} + +static u32 copy_data(struct lguest *srclg, + const struct lguest_dma *src, + const struct lguest_dma *dst, + struct page *pages[]) +{ + unsigned int totlen, si, di, srcoff, dstoff; + void *maddr = NULL; + + totlen = 0; + si = di = 0; + srcoff = dstoff = 0; + while (si < LGUEST_MAX_DMA_SECTIONS && src->len[si] + && di < LGUEST_MAX_DMA_SECTIONS && dst->len[di]) { + u32 len = min(src->len[si] - srcoff, dst->len[di] - dstoff); + + if (!maddr) + maddr = kmap(pages[di]); + + /* FIXME: This is not completely portable, since + archs do different things for copy_to_user_page. */ + if (copy_from_user(maddr + (dst->addr[di] + dstoff)%PAGE_SIZE, + (void *__user)src->addr[si], len) != 0) { + kill_guest(srclg, "bad address in sending DMA"); + totlen = 0; + break; + } + + totlen += len; + srcoff += len; + dstoff += len; + if (srcoff == src->len[si]) { + si++; + srcoff = 0; + } + if (dstoff == dst->len[di]) { + kunmap(pages[di]); + maddr = NULL; + di++; + dstoff = 0; + } + } + + if (maddr) + kunmap(pages[di]); + + return totlen; +} + +/* Src is us, ie. current. */ +static u32 do_dma(struct lguest *srclg, const struct lguest_dma *src, + struct lguest *dstlg, const struct lguest_dma *dst) +{ + int i; + u32 ret; + struct page *pages[LGUEST_MAX_DMA_SECTIONS]; + + if (!check_dma_list(dstlg, dst) || !check_dma_list(srclg, src)) + return 0; + + /* First get the destination pages */ + for (i = 0; i < LGUEST_MAX_DMA_SECTIONS; i++) { + if (dst->len[i] == 0) + break; + if (get_user_pages(dstlg->tsk, dstlg->mm, + dst->addr[i], 1, 1, 1, pages+i, NULL) + != 1) { + kill_guest(dstlg, "Error mapping DMA pages"); + ret = 0; + goto drop_pages; + } + } + + /* Now copy until we run out of src or dst. */ + ret = copy_data(srclg, src, dst, pages); + +drop_pages: + while (--i >= 0) + put_page(pages[i]); + return ret; +} + +static int dma_transfer(struct lguest *srclg, + unsigned long udma, + struct lguest_dma_info *dst) +{ + struct lguest_dma dst_dma, src_dma; + struct lguest *dstlg; + u32 i, dma = 0; + + dstlg = &lguests[dst->guestid]; + /* Get our dma list. */ + lgread(srclg, &src_dma, udma, sizeof(src_dma)); + + /* We can't deadlock against them dmaing to us, because this + * is all under the lguest_lock. */ + down_read(&dstlg->mm->mmap_sem); + + for (i = 0; i < dst->num_dmas; i++) { + dma = (dst->next_dma + i) % dst->num_dmas; + if (!lgread_other(dstlg, &dst_dma, + dst->dmas + dma * sizeof(struct lguest_dma), + sizeof(dst_dma))) { + goto fail; + } + if (!dst_dma.used_len) + break; + } + if (i != dst->num_dmas) { + unsigned long used_lenp; + unsigned int ret; + + ret = do_dma(srclg, &src_dma, dstlg, &dst_dma); + /* Put used length in src. */ + lgwrite_u32(srclg, + udma+offsetof(struct lguest_dma, used_len), ret); + if (ret == 0 && src_dma.len[0] != 0) + goto fail; + + /* Make sure destination sees contents before length. */ + wmb(); + used_lenp = dst->dmas + + dma * sizeof(struct lguest_dma) + + offsetof(struct lguest_dma, used_len); + lgwrite_other(dstlg, used_lenp, &ret, sizeof(ret)); + dst->next_dma++; + } + up_read(&dstlg->mm->mmap_sem); + + /* Do this last so dst doesn't simply sleep on lock. */ + set_bit(dst->interrupt, dstlg->irqs_pending); + wake_up_process(dstlg->tsk); + return i == dst->num_dmas; + +fail: + up_read(&dstlg->mm->mmap_sem); + return 0; +} + +void send_dma(struct lguest *lg, unsigned long ukey, unsigned long udma) +{ + union futex_key key; + int empty = 0; + struct rw_semaphore *fshared = ¤t->mm->mmap_sem; + +again: + mutex_lock(&lguest_lock); + down_read(fshared); + if (get_futex_key((u32 __user *)ukey, fshared, &key) != 0) { + kill_guest(lg, "bad sending DMA key"); + goto unlock; + } + /* Shared mapping? Look for other guests... */ + if (key.shared.offset & 1) { + struct lguest_dma_info *i; + list_for_each_entry(i, &dma_hash[hash(&key)], list) { + if (i->guestid == lg->guestid) + continue; + if (!key_eq(&key, &i->key)) + continue; + + empty += dma_transfer(lg, udma, i); + break; + } + if (empty == 1) { + /* Give any recipients one chance to restock. */ + up_read(¤t->mm->mmap_sem); + mutex_unlock(&lguest_lock); + empty++; + goto again; + } + } else { + /* Private mapping: tell our userspace. */ + lg->dma_is_pending = 1; + lg->pending_dma = udma; + lg->pending_key = ukey; + } +unlock: + up_read(fshared); + mutex_unlock(&lguest_lock); +} + +void release_all_dma(struct lguest *lg) +{ + unsigned int i; + + BUG_ON(!mutex_is_locked(&lguest_lock)); + + down_read(&lg->mm->mmap_sem); + for (i = 0; i < LGUEST_MAX_DMA; i++) { + if (lg->dma[i].interrupt) + unlink_dma(&lg->dma[i]); + } + up_read(&lg->mm->mmap_sem); +} + +/* Userspace wants a dma buffer from this guest. */ +unsigned long get_dma_buffer(struct lguest *lg, + unsigned long ukey, unsigned long *interrupt) +{ + unsigned long ret = 0; + union futex_key key; + struct lguest_dma_info *i; + struct rw_semaphore *fshared = ¤t->mm->mmap_sem; + + mutex_lock(&lguest_lock); + down_read(fshared); + if (get_futex_key((u32 __user *)ukey, fshared, &key) != 0) { + kill_guest(lg, "bad registered DMA buffer"); + goto unlock; + } + list_for_each_entry(i, &dma_hash[hash(&key)], list) { + if (key_eq(&key, &i->key) && i->guestid == lg->guestid) { + unsigned int j; + for (j = 0; j < i->num_dmas; j++) { + struct lguest_dma dma; + + ret = i->dmas + j * sizeof(struct lguest_dma); + lgread(lg, &dma, ret, sizeof(dma)); + if (dma.used_len == 0) + break; + } + *interrupt = i->interrupt; + break; + } + } +unlock: + up_read(fshared); + mutex_unlock(&lguest_lock); + return ret; +} + diff --git a/drivers/lguest/lg.h b/drivers/lguest/lg.h new file mode 100644 index 00000000000..3e2ddfbc816 --- /dev/null +++ b/drivers/lguest/lg.h @@ -0,0 +1,261 @@ +#ifndef _LGUEST_H +#define _LGUEST_H + +#include <asm/desc.h> + +#define GDT_ENTRY_LGUEST_CS 10 +#define GDT_ENTRY_LGUEST_DS 11 +#define LGUEST_CS (GDT_ENTRY_LGUEST_CS * 8) +#define LGUEST_DS (GDT_ENTRY_LGUEST_DS * 8) + +#ifndef __ASSEMBLY__ +#include <linux/types.h> +#include <linux/init.h> +#include <linux/stringify.h> +#include <linux/binfmts.h> +#include <linux/futex.h> +#include <linux/lguest.h> +#include <linux/lguest_launcher.h> +#include <linux/wait.h> +#include <linux/err.h> +#include <asm/semaphore.h> +#include "irq_vectors.h" + +#define GUEST_PL 1 + +struct lguest_regs +{ + /* Manually saved part. */ + unsigned long ebx, ecx, edx; + unsigned long esi, edi, ebp; + unsigned long gs; + unsigned long eax; + unsigned long fs, ds, es; + unsigned long trapnum, errcode; + /* Trap pushed part */ + unsigned long eip; + unsigned long cs; + unsigned long eflags; + unsigned long esp; + unsigned long ss; +}; + +void free_pagetables(void); +int init_pagetables(struct page **switcher_page, unsigned int pages); + +/* Full 4G segment descriptors, suitable for CS and DS. */ +#define FULL_EXEC_SEGMENT ((struct desc_struct){0x0000ffff, 0x00cf9b00}) +#define FULL_SEGMENT ((struct desc_struct){0x0000ffff, 0x00cf9300}) + +struct lguest_dma_info +{ + struct list_head list; + union futex_key key; + unsigned long dmas; + u16 next_dma; + u16 num_dmas; + u16 guestid; + u8 interrupt; /* 0 when not registered */ +}; + +/* We have separate types for the guest's ptes & pgds and the shadow ptes & + * pgds. Since this host might use three-level pagetables and the guest and + * shadow pagetables don't, we can't use the normal pte_t/pgd_t. */ +typedef union { + struct { unsigned flags:12, pfn:20; }; + struct { unsigned long val; } raw; +} spgd_t; +typedef union { + struct { unsigned flags:12, pfn:20; }; + struct { unsigned long val; } raw; +} spte_t; +typedef union { + struct { unsigned flags:12, pfn:20; }; + struct { unsigned long val; } raw; +} gpgd_t; +typedef union { + struct { unsigned flags:12, pfn:20; }; + struct { unsigned long val; } raw; +} gpte_t; +#define mkgpte(_val) ((gpte_t){.raw.val = _val}) +#define mkgpgd(_val) ((gpgd_t){.raw.val = _val}) + +struct pgdir +{ + unsigned long cr3; + spgd_t *pgdir; +}; + +/* This is a guest-specific page (mapped ro) into the guest. */ +struct lguest_ro_state +{ + /* Host information we need to restore when we switch back. */ + u32 host_cr3; + struct Xgt_desc_struct host_idt_desc; + struct Xgt_desc_struct host_gdt_desc; + u32 host_sp; + + /* Fields which are used when guest is running. */ + struct Xgt_desc_struct guest_idt_desc; + struct Xgt_desc_struct guest_gdt_desc; + struct i386_hw_tss guest_tss; + struct desc_struct guest_idt[IDT_ENTRIES]; + struct desc_struct guest_gdt[GDT_ENTRIES]; +}; + +/* We have two pages shared with guests, per cpu. */ +struct lguest_pages +{ + /* This is the stack page mapped rw in guest */ + char spare[PAGE_SIZE - sizeof(struct lguest_regs)]; + struct lguest_regs regs; + + /* This is the host state & guest descriptor page, ro in guest */ + struct lguest_ro_state state; +} __attribute__((aligned(PAGE_SIZE))); + +#define CHANGED_IDT 1 +#define CHANGED_GDT 2 +#define CHANGED_GDT_TLS 4 /* Actually a subset of CHANGED_GDT */ +#define CHANGED_ALL 3 + +/* The private info the thread maintains about the guest. */ +struct lguest +{ + /* At end of a page shared mapped over lguest_pages in guest. */ + unsigned long regs_page; + struct lguest_regs *regs; + struct lguest_data __user *lguest_data; + struct task_struct *tsk; + struct mm_struct *mm; /* == tsk->mm, but that becomes NULL on exit */ + u16 guestid; + u32 pfn_limit; + u32 page_offset; + u32 cr2; + int halted; + int ts; + u32 next_hcall; + u32 esp1; + u8 ss1; + + /* Do we need to stop what we're doing and return to userspace? */ + int break_out; + wait_queue_head_t break_wq; + + /* Bitmap of what has changed: see CHANGED_* above. */ + int changed; + struct lguest_pages *last_pages; + + /* We keep a small number of these. */ + u32 pgdidx; + struct pgdir pgdirs[4]; + + /* Cached wakeup: we hold a reference to this task. */ + struct task_struct *wake; + + unsigned long noirq_start, noirq_end; + int dma_is_pending; + unsigned long pending_dma; /* struct lguest_dma */ + unsigned long pending_key; /* address they're sending to */ + + unsigned int stack_pages; + u32 tsc_khz; + + struct lguest_dma_info dma[LGUEST_MAX_DMA]; + + /* Dead? */ + const char *dead; + + /* The GDT entries copied into lguest_ro_state when running. */ + struct desc_struct gdt[GDT_ENTRIES]; + + /* The IDT entries: some copied into lguest_ro_state when running. */ + struct desc_struct idt[FIRST_EXTERNAL_VECTOR+LGUEST_IRQS]; + struct desc_struct syscall_idt; + + /* Virtual clock device */ + struct hrtimer hrt; + + /* Pending virtual interrupts */ + DECLARE_BITMAP(irqs_pending, LGUEST_IRQS); +}; + +extern struct lguest lguests[]; +extern struct mutex lguest_lock; + +/* core.c: */ +u32 lgread_u32(struct lguest *lg, unsigned long addr); +void lgwrite_u32(struct lguest *lg, unsigned long addr, u32 val); +void lgread(struct lguest *lg, void *buf, unsigned long addr, unsigned len); +void lgwrite(struct lguest *lg, unsigned long, const void *buf, unsigned len); +int find_free_guest(void); +int lguest_address_ok(const struct lguest *lg, + unsigned long addr, unsigned long len); +int run_guest(struct lguest *lg, unsigned long __user *user); + + +/* interrupts_and_traps.c: */ +void maybe_do_interrupt(struct lguest *lg); +int deliver_trap(struct lguest *lg, unsigned int num); +void load_guest_idt_entry(struct lguest *lg, unsigned int i, u32 low, u32 hi); +void guest_set_stack(struct lguest *lg, u32 seg, u32 esp, unsigned int pages); +void pin_stack_pages(struct lguest *lg); +void setup_default_idt_entries(struct lguest_ro_state *state, + const unsigned long *def); +void copy_traps(const struct lguest *lg, struct desc_struct *idt, + const unsigned long *def); +void guest_set_clockevent(struct lguest *lg, unsigned long delta); +void init_clockdev(struct lguest *lg); + +/* segments.c: */ +void setup_default_gdt_entries(struct lguest_ro_state *state); +void setup_guest_gdt(struct lguest *lg); +void load_guest_gdt(struct lguest *lg, unsigned long table, u32 num); +void guest_load_tls(struct lguest *lg, unsigned long tls_array); +void copy_gdt(const struct lguest *lg, struct desc_struct *gdt); +void copy_gdt_tls(const struct lguest *lg, struct desc_struct *gdt); + +/* page_tables.c: */ +int init_guest_pagetable(struct lguest *lg, unsigned long pgtable); +void free_guest_pagetable(struct lguest *lg); +void guest_new_pagetable(struct lguest *lg, unsigned long pgtable); +void guest_set_pmd(struct lguest *lg, unsigned long cr3, u32 i); +void guest_pagetable_clear_all(struct lguest *lg); +void guest_pagetable_flush_user(struct lguest *lg); +void guest_set_pte(struct lguest *lg, unsigned long cr3, + unsigned long vaddr, gpte_t val); +void map_switcher_in_guest(struct lguest *lg, struct lguest_pages *pages); +int demand_page(struct lguest *info, unsigned long cr2, int errcode); +void pin_page(struct lguest *lg, unsigned long vaddr); + +/* lguest_user.c: */ +int lguest_device_init(void); +void lguest_device_remove(void); + +/* io.c: */ +void lguest_io_init(void); +int bind_dma(struct lguest *lg, + unsigned long key, unsigned long udma, u16 numdmas, u8 interrupt); +void send_dma(struct lguest *info, unsigned long key, unsigned long udma); +void release_all_dma(struct lguest *lg); +unsigned long get_dma_buffer(struct lguest *lg, unsigned long key, + unsigned long *interrupt); + +/* hypercalls.c: */ +void do_hypercalls(struct lguest *lg); + +#define kill_guest(lg, fmt...) \ +do { \ + if (!(lg)->dead) { \ + (lg)->dead = kasprintf(GFP_ATOMIC, fmt); \ + if (!(lg)->dead) \ + (lg)->dead = ERR_PTR(-ENOMEM); \ + } \ +} while(0) + +static inline unsigned long guest_pa(struct lguest *lg, unsigned long vaddr) +{ + return vaddr - lg->page_offset; +} +#endif /* __ASSEMBLY__ */ +#endif /* _LGUEST_H */ diff --git a/drivers/lguest/lguest.c b/drivers/lguest/lguest.c new file mode 100644 index 00000000000..b9a58b78c99 --- /dev/null +++ b/drivers/lguest/lguest.c @@ -0,0 +1,621 @@ +/* + * Lguest specific paravirt-ops implementation + * + * Copyright (C) 2006, Rusty Russell <rusty@rustcorp.com.au> IBM Corporation. + * + * 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, GOOD TITLE or + * NON INFRINGEMENT. See the GNU General Public License for more + * details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA. + */ +#include <linux/kernel.h> +#include <linux/start_kernel.h> +#include <linux/string.h> +#include <linux/console.h> +#include <linux/screen_info.h> +#include <linux/irq.h> +#include <linux/interrupt.h> +#include <linux/clocksource.h> +#include <linux/clockchips.h> +#include <linux/lguest.h> +#include <linux/lguest_launcher.h> +#include <linux/lguest_bus.h> +#include <asm/paravirt.h> +#include <asm/param.h> +#include <asm/page.h> +#include <asm/pgtable.h> +#include <asm/desc.h> +#include <asm/setup.h> +#include <asm/e820.h> +#include <asm/mce.h> +#include <asm/io.h> +//#include <asm/sched-clock.h> + +/* Declarations for definitions in lguest_guest.S */ +extern char lguest_noirq_start[], lguest_noirq_end[]; +extern const char lgstart_cli[], lgend_cli[]; +extern const char lgstart_sti[], lgend_sti[]; +extern const char lgstart_popf[], lgend_popf[]; +extern const char lgstart_pushf[], lgend_pushf[]; +extern const char lgstart_iret[], lgend_iret[]; +extern void lguest_iret(void); + +struct lguest_data lguest_data = { + .hcall_status = { [0 ... LHCALL_RING_SIZE-1] = 0xFF }, + .noirq_start = (u32)lguest_noirq_start, + .noirq_end = (u32)lguest_noirq_end, + .blocked_interrupts = { 1 }, /* Block timer interrupts */ +}; +struct lguest_device_desc *lguest_devices; + +static enum paravirt_lazy_mode lazy_mode; +static void lguest_lazy_mode(enum paravirt_lazy_mode mode) +{ + if (mode == PARAVIRT_LAZY_FLUSH) { + if (unlikely(lazy_mode != PARAVIRT_LAZY_NONE)) + hcall(LHCALL_FLUSH_ASYNC, 0, 0, 0); + } else { + lazy_mode = mode; + if (mode == PARAVIRT_LAZY_NONE) + hcall(LHCALL_FLUSH_ASYNC, 0, 0, 0); + } +} + +static void lazy_hcall(unsigned long call, + unsigned long arg1, + unsigned long arg2, + unsigned long arg3) +{ + if (lazy_mode == PARAVIRT_LAZY_NONE) + hcall(call, arg1, arg2, arg3); + else + async_hcall(call, arg1, arg2, arg3); +} + +void async_hcall(unsigned long call, + unsigned long arg1, unsigned long arg2, unsigned long arg3) +{ + /* Note: This code assumes we're uniprocessor. */ + static unsigned int next_call; + unsigned long flags; + + local_irq_save(flags); + if (lguest_data.hcall_status[next_call] != 0xFF) { + /* Table full, so do normal hcall which will flush table. */ + hcall(call, arg1, arg2, arg3); + } else { + lguest_data.hcalls[next_call].eax = call; + lguest_data.hcalls[next_call].edx = arg1; + lguest_data.hcalls[next_call].ebx = arg2; + lguest_data.hcalls[next_call].ecx = arg3; + /* Make sure host sees arguments before "valid" flag. */ + wmb(); + lguest_data.hcall_status[next_call] = 0; + if (++next_call == LHCALL_RING_SIZE) + next_call = 0; + } + local_irq_restore(flags); +} + +void lguest_send_dma(unsigned long key, struct lguest_dma *dma) +{ + dma->used_len = 0; + hcall(LHCALL_SEND_DMA, key, __pa(dma), 0); +} + +int lguest_bind_dma(unsigned long key, struct lguest_dma *dmas, + unsigned int num, u8 irq) +{ + if (!hcall(LHCALL_BIND_DMA, key, __pa(dmas), (num << 8) | irq)) + return -ENOMEM; + return 0; +} + +void lguest_unbind_dma(unsigned long key, struct lguest_dma *dmas) +{ + hcall(LHCALL_BIND_DMA, key, __pa(dmas), 0); +} + +/* For guests, device memory can be used as normal memory, so we cast away the + * __iomem to quieten sparse. */ +void *lguest_map(unsigned long phys_addr, unsigned long pages) +{ + return (__force void *)ioremap(phys_addr, PAGE_SIZE*pages); +} + +void lguest_unmap(void *addr) +{ + iounmap((__force void __iomem *)addr); +} + +static unsigned long save_fl(void) +{ + return lguest_data.irq_enabled; +} + +static void restore_fl(unsigned long flags) +{ + /* FIXME: Check if interrupt pending... */ + lguest_data.irq_enabled = flags; +} + +static void irq_disable(void) +{ + lguest_data.irq_enabled = 0; +} + +static void irq_enable(void) +{ + /* FIXME: Check if interrupt pending... */ + lguest_data.irq_enabled = X86_EFLAGS_IF; +} + +static void lguest_write_idt_entry(struct desc_struct *dt, + int entrynum, u32 low, u32 high) +{ + write_dt_entry(dt, entrynum, low, high); + hcall(LHCALL_LOAD_IDT_ENTRY, entrynum, low, high); +} + +static void lguest_load_idt(const struct Xgt_desc_struct *desc) +{ + unsigned int i; + struct desc_struct *idt = (void *)desc->address; + + for (i = 0; i < (desc->size+1)/8; i++) + hcall(LHCALL_LOAD_IDT_ENTRY, i, idt[i].a, idt[i].b); +} + +static void lguest_load_gdt(const struct Xgt_desc_struct *desc) +{ + BUG_ON((desc->size+1)/8 != GDT_ENTRIES); + hcall(LHCALL_LOAD_GDT, __pa(desc->address), GDT_ENTRIES, 0); +} + +static void lguest_write_gdt_entry(struct desc_struct *dt, + int entrynum, u32 low, u32 high) +{ + write_dt_entry(dt, entrynum, low, high); + hcall(LHCALL_LOAD_GDT, __pa(dt), GDT_ENTRIES, 0); +} + +static void lguest_load_tls(struct thread_struct *t, unsigned int cpu) +{ + lazy_hcall(LHCALL_LOAD_TLS, __pa(&t->tls_array), cpu, 0); +} + +static void lguest_set_ldt(const void *addr, unsigned entries) +{ +} + +static void lguest_load_tr_desc(void) +{ +} + +static void lguest_cpuid(unsigned int *eax, unsigned int *ebx, + unsigned int *ecx, unsigned int *edx) +{ + int function = *eax; + + native_cpuid(eax, ebx, ecx, edx); + switch (function) { + case 1: /* Basic feature request. */ + /* We only allow kernel to see SSE3, CMPXCHG16B and SSSE3 */ + *ecx &= 0x00002201; + /* SSE, SSE2, FXSR, MMX, CMOV, CMPXCHG8B, FPU. */ + *edx &= 0x07808101; + /* Host wants to know when we flush kernel pages: set PGE. */ + *edx |= 0x00002000; + break; + case 0x80000000: + /* Futureproof this a little: if they ask how much extended + * processor information, limit it to known fields. */ + if (*eax > 0x80000008) + *eax = 0x80000008; + break; + } +} + +static unsigned long current_cr0, current_cr3; +static void lguest_write_cr0(unsigned long val) +{ + lazy_hcall(LHCALL_TS, val & 8, 0, 0); + current_cr0 = val; +} + +static unsigned long lguest_read_cr0(void) +{ + return current_cr0; +} + +static void lguest_clts(void) +{ + lazy_hcall(LHCALL_TS, 0, 0, 0); + current_cr0 &= ~8U; +} + +static unsigned long lguest_read_cr2(void) +{ + return lguest_data.cr2; +} + +static void lguest_write_cr3(unsigned long cr3) +{ + lazy_hcall(LHCALL_NEW_PGTABLE, cr3, 0, 0); + current_cr3 = cr3; +} + +static unsigned long lguest_read_cr3(void) +{ + return current_cr3; +} + +/* Used to enable/disable PGE, but we don't care. */ +static unsigned long lguest_read_cr4(void) +{ + return 0; +} + +static void lguest_write_cr4(unsigned long val) +{ +} + +static void lguest_set_pte_at(struct mm_struct *mm, unsigned long addr, + pte_t *ptep, pte_t pteval) +{ + *ptep = pteval; + lazy_hcall(LHCALL_SET_PTE, __pa(mm->pgd), addr, pteval.pte_low); +} + +/* We only support two-level pagetables at the moment. */ +static void lguest_set_pmd(pmd_t *pmdp, pmd_t pmdval) +{ + *pmdp = pmdval; + lazy_hcall(LHCALL_SET_PMD, __pa(pmdp)&PAGE_MASK, + (__pa(pmdp)&(PAGE_SIZE-1))/4, 0); +} + +/* FIXME: Eliminate all callers of this. */ +static void lguest_set_pte(pte_t *ptep, pte_t pteval) +{ + *ptep = pteval; + /* Don't bother with hypercall before initial setup. */ + if (current_cr3) + lazy_hcall(LHCALL_FLUSH_TLB, 1, 0, 0); +} + +static void lguest_flush_tlb_single(unsigned long addr) +{ + /* Simply set it to zero, and it will fault back in. */ + lazy_hcall(LHCALL_SET_PTE, current_cr3, addr, 0); +} + +static void lguest_flush_tlb_user(void) +{ + lazy_hcall(LHCALL_FLUSH_TLB, 0, 0, 0); +} + +static void lguest_flush_tlb_kernel(void) +{ + lazy_hcall(LHCALL_FLUSH_TLB, 1, 0, 0); +} + +static void disable_lguest_irq(unsigned int irq) +{ + set_bit(irq, lguest_data.blocked_interrupts); +} + +static void enable_lguest_irq(unsigned int irq) +{ + clear_bit(irq, lguest_data.blocked_interrupts); + /* FIXME: If it's pending? */ +} + +static struct irq_chip lguest_irq_controller = { + .name = "lguest", + .mask = disable_lguest_irq, + .mask_ack = disable_lguest_irq, + .unmask = enable_lguest_irq, +}; + +static void __init lguest_init_IRQ(void) +{ + unsigned int i; + + for (i = 0; i < LGUEST_IRQS; i++) { + int vector = FIRST_EXTERNAL_VECTOR + i; + if (vector != SYSCALL_VECTOR) { + set_intr_gate(vector, interrupt[i]); + set_irq_chip_and_handler(i, &lguest_irq_controller, + handle_level_irq); + } + } + irq_ctx_init(smp_processor_id()); +} + +static unsigned long lguest_get_wallclock(void) +{ + return hcall(LHCALL_GET_WALLCLOCK, 0, 0, 0); +} + +static cycle_t lguest_clock_read(void) +{ + if (lguest_data.tsc_khz) + return native_read_tsc(); + else + return jiffies; +} + +/* This is what we tell the kernel is our clocksource. */ +static struct clocksource lguest_clock = { + .name = "lguest", + .rating = 400, + .read = lguest_clock_read, +}; + +/* We also need a "struct clock_event_device": Linux asks us to set it to go + * off some time in the future. Actually, James Morris figured all this out, I + * just applied the patch. */ +static int lguest_clockevent_set_next_event(unsigned long delta, + struct clock_event_device *evt) +{ + if (delta < LG_CLOCK_MIN_DELTA) { + if (printk_ratelimit()) + printk(KERN_DEBUG "%s: small delta %lu ns\n", + __FUNCTION__, delta); + return -ETIME; + } + hcall(LHCALL_SET_CLOCKEVENT, delta, 0, 0); + return 0; +} + +static void lguest_clockevent_set_mode(enum clock_event_mode mode, + struct clock_event_device *evt) +{ + switch (mode) { + case CLOCK_EVT_MODE_UNUSED: + case CLOCK_EVT_MODE_SHUTDOWN: + /* A 0 argument shuts the clock down. */ + hcall(LHCALL_SET_CLOCKEVENT, 0, 0, 0); + break; + case CLOCK_EVT_MODE_ONESHOT: + /* This is what we expect. */ + break; + case CLOCK_EVT_MODE_PERIODIC: + BUG(); + } +} + +/* This describes our primitive timer chip. */ +static struct clock_event_device lguest_clockevent = { + .name = "lguest", + .features = CLOCK_EVT_FEAT_ONESHOT, + .set_next_event = lguest_clockevent_set_next_event, + .set_mode = lguest_clockevent_set_mode, + .rating = INT_MAX, + .mult = 1, + .shift = 0, + .min_delta_ns = LG_CLOCK_MIN_DELTA, + .max_delta_ns = LG_CLOCK_MAX_DELTA, +}; + +/* This is the Guest timer interrupt handler (hardware interrupt 0). We just + * call the clockevent infrastructure and it does whatever needs doing. */ +static void lguest_time_irq(unsigned int irq, struct irq_desc *desc) +{ + unsigned long flags; + + /* Don't interrupt us while this is running. */ + local_irq_save(flags); + lguest_clockevent.event_handler(&lguest_clockevent); + local_irq_restore(flags); +} + +static void lguest_time_init(void) +{ + set_irq_handler(0, lguest_time_irq); + + /* We use the TSC if the Host tells us we can, otherwise a dumb + * jiffies-based clock. */ + if (lguest_data.tsc_khz) { + lguest_clock.shift = 22; + lguest_clock.mult = clocksource_khz2mult(lguest_data.tsc_khz, + lguest_clock.shift); + lguest_clock.mask = CLOCKSOURCE_MASK(64); + lguest_clock.flags = CLOCK_SOURCE_IS_CONTINUOUS; + } else { + /* To understand this, start at kernel/time/jiffies.c... */ + lguest_clock.shift = 8; + lguest_clock.mult = (((u64)NSEC_PER_SEC<<8)/ACTHZ) << 8; + lguest_clock.mask = CLOCKSOURCE_MASK(32); + } + clocksource_register(&lguest_clock); + + /* We can't set cpumask in the initializer: damn C limitations! */ + lguest_clockevent.cpumask = cpumask_of_cpu(0); + clockevents_register_device(&lguest_clockevent); + + enable_lguest_irq(0); +} + +static void lguest_load_esp0(struct tss_struct *tss, + struct thread_struct *thread) +{ + lazy_hcall(LHCALL_SET_STACK, __KERNEL_DS|0x1, thread->esp0, + THREAD_SIZE/PAGE_SIZE); +} + +static void lguest_set_debugreg(int regno, unsigned long value) +{ + /* FIXME: Implement */ +} + +static void lguest_wbinvd(void) +{ +} + +#ifdef CONFIG_X86_LOCAL_APIC +static void lguest_apic_write(unsigned long reg, unsigned long v) +{ +} + +static unsigned long lguest_apic_read(unsigned long reg) +{ + return 0; +} +#endif + +static void lguest_safe_halt(void) +{ + hcall(LHCALL_HALT, 0, 0, 0); +} + +static void lguest_power_off(void) +{ + hcall(LHCALL_CRASH, __pa("Power down"), 0, 0); +} + +static int lguest_panic(struct notifier_block *nb, unsigned long l, void *p) +{ + hcall(LHCALL_CRASH, __pa(p), 0, 0); + return NOTIFY_DONE; +} + +static struct notifier_block paniced = { + .notifier_call = lguest_panic +}; + +static __init char *lguest_memory_setup(void) +{ + /* We do this here because lockcheck barfs if before start_kernel */ + atomic_notifier_chain_register(&panic_notifier_list, &paniced); + + add_memory_region(E820_MAP->addr, E820_MAP->size, E820_MAP->type); + return "LGUEST"; +} + +static const struct lguest_insns +{ + const char *start, *end; +} lguest_insns[] = { + [PARAVIRT_PATCH(irq_disable)] = { lgstart_cli, lgend_cli }, + [PARAVIRT_PATCH(irq_enable)] = { lgstart_sti, lgend_sti }, + [PARAVIRT_PATCH(restore_fl)] = { lgstart_popf, lgend_popf }, + [PARAVIRT_PATCH(save_fl)] = { lgstart_pushf, lgend_pushf }, +}; +static unsigned lguest_patch(u8 type, u16 clobber, void *insns, unsigned len) +{ + unsigned int insn_len; + + /* Don't touch it if we don't have a replacement */ + if (type >= ARRAY_SIZE(lguest_insns) || !lguest_insns[type].start) + return paravirt_patch_default(type, clobber, insns, len); + + insn_len = lguest_insns[type].end - lguest_insns[type].start; + + /* Similarly if we can't fit replacement. */ + if (len < insn_len) + return paravirt_patch_default(type, clobber, insns, len); + + memcpy(insns, lguest_insns[type].start, insn_len); + return insn_len; +} + +__init void lguest_init(void *boot) +{ + /* Copy boot parameters first. */ + memcpy(&boot_params, boot, PARAM_SIZE); + memcpy(boot_command_line, __va(boot_params.hdr.cmd_line_ptr), + COMMAND_LINE_SIZE); + + paravirt_ops.name = "lguest"; + paravirt_ops.paravirt_enabled = 1; + paravirt_ops.kernel_rpl = 1; + + paravirt_ops.save_fl = save_fl; + paravirt_ops.restore_fl = restore_fl; + paravirt_ops.irq_disable = irq_disable; + paravirt_ops.irq_enable = irq_enable; + paravirt_ops.load_gdt = lguest_load_gdt; + paravirt_ops.memory_setup = lguest_memory_setup; + paravirt_ops.cpuid = lguest_cpuid; + paravirt_ops.write_cr3 = lguest_write_cr3; + paravirt_ops.flush_tlb_user = lguest_flush_tlb_user; + paravirt_ops.flush_tlb_single = lguest_flush_tlb_single; + paravirt_ops.flush_tlb_kernel = lguest_flush_tlb_kernel; + paravirt_ops.set_pte = lguest_set_pte; + paravirt_ops.set_pte_at = lguest_set_pte_at; + paravirt_ops.set_pmd = lguest_set_pmd; +#ifdef CONFIG_X86_LOCAL_APIC + paravirt_ops.apic_write = lguest_apic_write; + paravirt_ops.apic_write_atomic = lguest_apic_write; + paravirt_ops.apic_read = lguest_apic_read; +#endif + paravirt_ops.load_idt = lguest_load_idt; + paravirt_ops.iret = lguest_iret; + paravirt_ops.load_esp0 = lguest_load_esp0; + paravirt_ops.load_tr_desc = lguest_load_tr_desc; + paravirt_ops.set_ldt = lguest_set_ldt; + paravirt_ops.load_tls = lguest_load_tls; + paravirt_ops.set_debugreg = lguest_set_debugreg; + paravirt_ops.clts = lguest_clts; + paravirt_ops.read_cr0 = lguest_read_cr0; + paravirt_ops.write_cr0 = lguest_write_cr0; + paravirt_ops.init_IRQ = lguest_init_IRQ; + paravirt_ops.read_cr2 = lguest_read_cr2; + paravirt_ops.read_cr3 = lguest_read_cr3; + paravirt_ops.read_cr4 = lguest_read_cr4; + paravirt_ops.write_cr4 = lguest_write_cr4; + paravirt_ops.write_gdt_entry = lguest_write_gdt_entry; + paravirt_ops.write_idt_entry = lguest_write_idt_entry; + paravirt_ops.patch = lguest_patch; + paravirt_ops.safe_halt = lguest_safe_halt; + paravirt_ops.get_wallclock = lguest_get_wallclock; + paravirt_ops.time_init = lguest_time_init; + paravirt_ops.set_lazy_mode = lguest_lazy_mode; + paravirt_ops.wbinvd = lguest_wbinvd; + + hcall(LHCALL_LGUEST_INIT, __pa(&lguest_data), 0, 0); + + /* We use top of mem for initial pagetables. */ + init_pg_tables_end = __pa(pg0); + + asm volatile ("mov %0, %%fs" : : "r" (__KERNEL_DS) : "memory"); + + reserve_top_address(lguest_data.reserve_mem); + + lockdep_init(); + + paravirt_disable_iospace(); + + cpu_detect(&new_cpu_data); + /* head.S usually sets up the first capability word, so do it here. */ + new_cpu_data.x86_capability[0] = cpuid_edx(1); + + /* Math is always hard! */ + new_cpu_data.hard_math = 1; + +#ifdef CONFIG_X86_MCE + mce_disabled = 1; +#endif + +#ifdef CONFIG_ACPI + acpi_disabled = 1; + acpi_ht = 0; +#endif + + add_preferred_console("hvc", 0, NULL); + + pm_power_off = lguest_power_off; + start_kernel(); +} diff --git a/drivers/lguest/lguest_asm.S b/drivers/lguest/lguest_asm.S new file mode 100644 index 00000000000..00046c57b5b --- /dev/null +++ b/drivers/lguest/lguest_asm.S @@ -0,0 +1,56 @@ +#include <linux/linkage.h> +#include <linux/lguest.h> +#include <asm/asm-offsets.h> +#include <asm/thread_info.h> + +/* FIXME: Once asm/processor-flags.h goes in, include that */ +#define X86_EFLAGS_IF 0x00000200 + +/* + * This is where we begin: we have a magic signature which the launcher looks + * for. The plan is that the Linux boot protocol will be extended with a + * "platform type" field which will guide us here from the normal entry point, + * but for the moment this suffices. We pass the virtual address of the boot + * info to lguest_init(). + * + * We put it in .init.text will be discarded after boot. + */ +.section .init.text, "ax", @progbits +.ascii "GenuineLguest" + /* Set up initial stack. */ + movl $(init_thread_union+THREAD_SIZE),%esp + movl %esi, %eax + addl $__PAGE_OFFSET, %eax + jmp lguest_init + +/* The templates for inline patching. */ +#define LGUEST_PATCH(name, insns...) \ + lgstart_##name: insns; lgend_##name:; \ + .globl lgstart_##name; .globl lgend_##name + +LGUEST_PATCH(cli, movl $0, lguest_data+LGUEST_DATA_irq_enabled) +LGUEST_PATCH(sti, movl $X86_EFLAGS_IF, lguest_data+LGUEST_DATA_irq_enabled) +LGUEST_PATCH(popf, movl %eax, lguest_data+LGUEST_DATA_irq_enabled) +LGUEST_PATCH(pushf, movl lguest_data+LGUEST_DATA_irq_enabled, %eax) + +.text +/* These demark the EIP range where host should never deliver interrupts. */ +.global lguest_noirq_start +.global lguest_noirq_end + +/* + * We move eflags word to lguest_data.irq_enabled to restore interrupt state. + * For page faults, gpfs and virtual interrupts, the hypervisor has saved + * eflags manually, otherwise it was delivered directly and so eflags reflects + * the real machine IF state, ie. interrupts on. Since the kernel always dies + * if it takes such a trap with interrupts disabled anyway, turning interrupts + * back on unconditionally here is OK. + */ +ENTRY(lguest_iret) + pushl %eax + movl 12(%esp), %eax +lguest_noirq_start: + movl %eax,%ss:lguest_data+LGUEST_DATA_irq_enabled + popl %eax + iret +lguest_noirq_end: diff --git a/drivers/lguest/lguest_bus.c b/drivers/lguest/lguest_bus.c new file mode 100644 index 00000000000..18d6ab21a43 --- /dev/null +++ b/drivers/lguest/lguest_bus.c @@ -0,0 +1,148 @@ +#include <linux/init.h> +#include <linux/bootmem.h> +#include <linux/lguest_bus.h> +#include <asm/io.h> + +static ssize_t type_show(struct device *_dev, + struct device_attribute *attr, char *buf) +{ + struct lguest_device *dev = container_of(_dev,struct lguest_device,dev); + return sprintf(buf, "%hu", lguest_devices[dev->index].type); +} +static ssize_t features_show(struct device *_dev, + struct device_attribute *attr, char *buf) +{ + struct lguest_device *dev = container_of(_dev,struct lguest_device,dev); + return sprintf(buf, "%hx", lguest_devices[dev->index].features); +} +static ssize_t pfn_show(struct device *_dev, + struct device_attribute *attr, char *buf) +{ + struct lguest_device *dev = container_of(_dev,struct lguest_device,dev); + return sprintf(buf, "%u", lguest_devices[dev->index].pfn); +} +static ssize_t status_show(struct device *_dev, + struct device_attribute *attr, char *buf) +{ + struct lguest_device *dev = container_of(_dev,struct lguest_device,dev); + return sprintf(buf, "%hx", lguest_devices[dev->index].status); +} +static ssize_t status_store(struct device *_dev, struct device_attribute *attr, + const char *buf, size_t count) +{ + struct lguest_device *dev = container_of(_dev,struct lguest_device,dev); + if (sscanf(buf, "%hi", &lguest_devices[dev->index].status) != 1) + return -EINVAL; + return count; +} +static struct device_attribute lguest_dev_attrs[] = { + __ATTR_RO(type), + __ATTR_RO(features), + __ATTR_RO(pfn), + __ATTR(status, 0644, status_show, status_store), + __ATTR_NULL +}; + +static int lguest_dev_match(struct device *_dev, struct device_driver *_drv) +{ + struct lguest_device *dev = container_of(_dev,struct lguest_device,dev); + struct lguest_driver *drv = container_of(_drv,struct lguest_driver,drv); + + return (drv->device_type == lguest_devices[dev->index].type); +} + +struct lguest_bus { + struct bus_type bus; + struct device dev; +}; + +static struct lguest_bus lguest_bus = { + .bus = { + .name = "lguest", + .match = lguest_dev_match, + .dev_attrs = lguest_dev_attrs, + }, + .dev = { + .parent = NULL, + .bus_id = "lguest", + } +}; + +static int lguest_dev_probe(struct device *_dev) +{ + int ret; + struct lguest_device *dev = container_of(_dev,struct lguest_device,dev); + struct lguest_driver *drv = container_of(dev->dev.driver, + struct lguest_driver, drv); + + lguest_devices[dev->index].status |= LGUEST_DEVICE_S_DRIVER; + ret = drv->probe(dev); + if (ret == 0) + lguest_devices[dev->index].status |= LGUEST_DEVICE_S_DRIVER_OK; + return ret; +} + +int register_lguest_driver(struct lguest_driver *drv) +{ + if (!lguest_devices) + return 0; + + drv->drv.bus = &lguest_bus.bus; + drv->drv.name = drv->name; + drv->drv.owner = drv->owner; + drv->drv.probe = lguest_dev_probe; + + return driver_register(&drv->drv); +} +EXPORT_SYMBOL_GPL(register_lguest_driver); + +static void add_lguest_device(unsigned int index) +{ + struct lguest_device *new; + + lguest_devices[index].status |= LGUEST_DEVICE_S_ACKNOWLEDGE; + new = kmalloc(sizeof(struct lguest_device), GFP_KERNEL); + if (!new) { + printk(KERN_EMERG "Cannot allocate lguest device %u\n", index); + lguest_devices[index].status |= LGUEST_DEVICE_S_FAILED; + return; + } + + new->index = index; + new->private = NULL; + memset(&new->dev, 0, sizeof(new->dev)); + new->dev.parent = &lguest_bus.dev; + new->dev.bus = &lguest_bus.bus; + sprintf(new->dev.bus_id, "%u", index); + if (device_register(&new->dev) != 0) { + printk(KERN_EMERG "Cannot register lguest device %u\n", index); + lguest_devices[index].status |= LGUEST_DEVICE_S_FAILED; + kfree(new); + } +} + +static void scan_devices(void) +{ + unsigned int i; + + for (i = 0; i < LGUEST_MAX_DEVICES; i++) + if (lguest_devices[i].type) + add_lguest_device(i); +} + +static int __init lguest_bus_init(void) +{ + if (strcmp(paravirt_ops.name, "lguest") != 0) + return 0; + + /* Devices are in page above top of "normal" mem. */ + lguest_devices = lguest_map(max_pfn<<PAGE_SHIFT, 1); + + if (bus_register(&lguest_bus.bus) != 0 + || device_register(&lguest_bus.dev) != 0) + panic("lguest bus registration failed"); + + scan_devices(); + return 0; +} +postcore_initcall(lguest_bus_init); diff --git a/drivers/lguest/lguest_user.c b/drivers/lguest/lguest_user.c new file mode 100644 index 00000000000..e90d7a783da --- /dev/null +++ b/drivers/lguest/lguest_user.c @@ -0,0 +1,236 @@ +/* Userspace control of the guest, via /dev/lguest. */ +#include <linux/uaccess.h> +#include <linux/miscdevice.h> +#include <linux/fs.h> +#include "lg.h" + +static void setup_regs(struct lguest_regs *regs, unsigned long start) +{ + /* Write out stack in format lguest expects, so we can switch to it. */ + regs->ds = regs->es = regs->ss = __KERNEL_DS|GUEST_PL; + regs->cs = __KERNEL_CS|GUEST_PL; + regs->eflags = 0x202; /* Interrupts enabled. */ + regs->eip = start; + /* esi points to our boot information (physical address 0) */ +} + +/* + addr */ +static long user_get_dma(struct lguest *lg, const u32 __user *input) +{ + unsigned long key, udma, irq; + + if (get_user(key, input) != 0) + return -EFAULT; + udma = get_dma_buffer(lg, key, &irq); + if (!udma) + return -ENOENT; + + /* We put irq number in udma->used_len. */ + lgwrite_u32(lg, udma + offsetof(struct lguest_dma, used_len), irq); + return udma; +} + +/* To force the Guest to stop running and return to the Launcher, the + * Waker sets writes LHREQ_BREAK and the value "1" to /dev/lguest. The + * Launcher then writes LHREQ_BREAK and "0" to release the Waker. */ +static int break_guest_out(struct lguest *lg, const u32 __user *input) +{ + unsigned long on; + + /* Fetch whether they're turning break on or off.. */ + if (get_user(on, input) != 0) + return -EFAULT; + + if (on) { + lg->break_out = 1; + /* Pop it out (may be running on different CPU) */ + wake_up_process(lg->tsk); + /* Wait for them to reset it */ + return wait_event_interruptible(lg->break_wq, !lg->break_out); + } else { + lg->break_out = 0; + wake_up(&lg->break_wq); + return 0; + } +} + +/* + irq */ +static int user_send_irq(struct lguest *lg, const u32 __user *input) +{ + u32 irq; + + if (get_user(irq, input) != 0) + return -EFAULT; + if (irq >= LGUEST_IRQS) + return -EINVAL; + set_bit(irq, lg->irqs_pending); + return 0; +} + +static ssize_t read(struct file *file, char __user *user, size_t size,loff_t*o) +{ + struct lguest *lg = file->private_data; + + if (!lg) + return -EINVAL; + + /* If you're not the task which owns the guest, go away. */ + if (current != lg->tsk) + return -EPERM; + + if (lg->dead) { + size_t len; + + if (IS_ERR(lg->dead)) + return PTR_ERR(lg->dead); + + len = min(size, strlen(lg->dead)+1); + if (copy_to_user(user, lg->dead, len) != 0) + return -EFAULT; + return len; + } + + if (lg->dma_is_pending) + lg->dma_is_pending = 0; + + return run_guest(lg, (unsigned long __user *)user); +} + +/* Take: pfnlimit, pgdir, start, pageoffset. */ +static int initialize(struct file *file, const u32 __user *input) +{ + struct lguest *lg; + int err, i; + u32 args[4]; + + /* We grab the Big Lguest lock, which protects the global array + * "lguests" and multiple simultaneous initializations. */ + mutex_lock(&lguest_lock); + + if (file->private_data) { + err = -EBUSY; + goto unlock; + } + + if (copy_from_user(args, input, sizeof(args)) != 0) { + err = -EFAULT; + goto unlock; + } + + i = find_free_guest(); + if (i < 0) { + err = -ENOSPC; + goto unlock; + } + lg = &lguests[i]; + lg->guestid = i; + lg->pfn_limit = args[0]; + lg->page_offset = args[3]; + lg->regs_page = get_zeroed_page(GFP_KERNEL); + if (!lg->regs_page) { + err = -ENOMEM; + goto release_guest; + } + lg->regs = (void *)lg->regs_page + PAGE_SIZE - sizeof(*lg->regs); + + err = init_guest_pagetable(lg, args[1]); + if (err) + goto free_regs; + + setup_regs(lg->regs, args[2]); + setup_guest_gdt(lg); + init_clockdev(lg); + lg->tsk = current; + lg->mm = get_task_mm(lg->tsk); + init_waitqueue_head(&lg->break_wq); + lg->last_pages = NULL; + file->private_data = lg; + + mutex_unlock(&lguest_lock); + + return sizeof(args); + +free_regs: + free_page(lg->regs_page); +release_guest: + memset(lg, 0, sizeof(*lg)); +unlock: + mutex_unlock(&lguest_lock); + return err; +} + +static ssize_t write(struct file *file, const char __user *input, + size_t size, loff_t *off) +{ + struct lguest *lg = file->private_data; + u32 req; + + if (get_user(req, input) != 0) + return -EFAULT; + input += sizeof(req); + + if (req != LHREQ_INITIALIZE && !lg) + return -EINVAL; + if (lg && lg->dead) + return -ENOENT; + + /* If you're not the task which owns the Guest, you can only break */ + if (lg && current != lg->tsk && req != LHREQ_BREAK) + return -EPERM; + + switch (req) { + case LHREQ_INITIALIZE: + return initialize(file, (const u32 __user *)input); + case LHREQ_GETDMA: + return user_get_dma(lg, (const u32 __user *)input); + case LHREQ_IRQ: + return user_send_irq(lg, (const u32 __user *)input); + case LHREQ_BREAK: + return break_guest_out(lg, (const u32 __user *)input); + default: + return -EINVAL; + } +} + +static int close(struct inode *inode, struct file *file) +{ + struct lguest *lg = file->private_data; + + if (!lg) + return 0; + + mutex_lock(&lguest_lock); + /* Cancels the hrtimer set via LHCALL_SET_CLOCKEVENT. */ + hrtimer_cancel(&lg->hrt); + release_all_dma(lg); + free_guest_pagetable(lg); + mmput(lg->mm); + if (!IS_ERR(lg->dead)) + kfree(lg->dead); + free_page(lg->regs_page); + memset(lg, 0, sizeof(*lg)); + mutex_unlock(&lguest_lock); + return 0; +} + +static struct file_operations lguest_fops = { + .owner = THIS_MODULE, + .release = close, + .write = write, + .read = read, +}; +static struct miscdevice lguest_dev = { + .minor = MISC_DYNAMIC_MINOR, + .name = "lguest", + .fops = &lguest_fops, +}; + +int __init lguest_device_init(void) +{ + return misc_register(&lguest_dev); +} + +void __exit lguest_device_remove(void) +{ + misc_deregister(&lguest_dev); +} diff --git a/drivers/lguest/page_tables.c b/drivers/lguest/page_tables.c new file mode 100644 index 00000000000..1b0ba09b126 --- /dev/null +++ b/drivers/lguest/page_tables.c @@ -0,0 +1,411 @@ +/* Shadow page table operations. + * Copyright (C) Rusty Russell IBM Corporation 2006. + * GPL v2 and any later version */ +#include <linux/mm.h> +#include <linux/types.h> +#include <linux/spinlock.h> +#include <linux/random.h> +#include <linux/percpu.h> +#include <asm/tlbflush.h> +#include "lg.h" + +#define PTES_PER_PAGE_SHIFT 10 +#define PTES_PER_PAGE (1 << PTES_PER_PAGE_SHIFT) +#define SWITCHER_PGD_INDEX (PTES_PER_PAGE - 1) + +static DEFINE_PER_CPU(spte_t *, switcher_pte_pages); +#define switcher_pte_page(cpu) per_cpu(switcher_pte_pages, cpu) + +static unsigned vaddr_to_pgd_index(unsigned long vaddr) +{ + return vaddr >> (PAGE_SHIFT + PTES_PER_PAGE_SHIFT); +} + +/* These access the shadow versions (ie. the ones used by the CPU). */ +static spgd_t *spgd_addr(struct lguest *lg, u32 i, unsigned long vaddr) +{ + unsigned int index = vaddr_to_pgd_index(vaddr); + + if (index >= SWITCHER_PGD_INDEX) { + kill_guest(lg, "attempt to access switcher pages"); + index = 0; + } + return &lg->pgdirs[i].pgdir[index]; +} + +static spte_t *spte_addr(struct lguest *lg, spgd_t spgd, unsigned long vaddr) +{ + spte_t *page = __va(spgd.pfn << PAGE_SHIFT); + BUG_ON(!(spgd.flags & _PAGE_PRESENT)); + return &page[(vaddr >> PAGE_SHIFT) % PTES_PER_PAGE]; +} + +/* These access the guest versions. */ +static unsigned long gpgd_addr(struct lguest *lg, unsigned long vaddr) +{ + unsigned int index = vaddr >> (PAGE_SHIFT + PTES_PER_PAGE_SHIFT); + return lg->pgdirs[lg->pgdidx].cr3 + index * sizeof(gpgd_t); +} + +static unsigned long gpte_addr(struct lguest *lg, + gpgd_t gpgd, unsigned long vaddr) +{ + unsigned long gpage = gpgd.pfn << PAGE_SHIFT; + BUG_ON(!(gpgd.flags & _PAGE_PRESENT)); + return gpage + ((vaddr>>PAGE_SHIFT) % PTES_PER_PAGE) * sizeof(gpte_t); +} + +/* Do a virtual -> physical mapping on a user page. */ +static unsigned long get_pfn(unsigned long virtpfn, int write) +{ + struct page *page; + unsigned long ret = -1UL; + + down_read(¤t->mm->mmap_sem); + if (get_user_pages(current, current->mm, virtpfn << PAGE_SHIFT, + 1, write, 1, &page, NULL) == 1) + ret = page_to_pfn(page); + up_read(¤t->mm->mmap_sem); + return ret; +} + +static spte_t gpte_to_spte(struct lguest *lg, gpte_t gpte, int write) +{ + spte_t spte; + unsigned long pfn; + + /* We ignore the global flag. */ + spte.flags = (gpte.flags & ~_PAGE_GLOBAL); + pfn = get_pfn(gpte.pfn, write); + if (pfn == -1UL) { + kill_guest(lg, "failed to get page %u", gpte.pfn); + /* Must not put_page() bogus page on cleanup. */ + spte.flags = 0; + } + spte.pfn = pfn; + return spte; +} + +static void release_pte(spte_t pte) +{ + if (pte.flags & _PAGE_PRESENT) + put_page(pfn_to_page(pte.pfn)); +} + +static void check_gpte(struct lguest *lg, gpte_t gpte) +{ + if ((gpte.flags & (_PAGE_PWT|_PAGE_PSE)) || gpte.pfn >= lg->pfn_limit) + kill_guest(lg, "bad page table entry"); +} + +static void check_gpgd(struct lguest *lg, gpgd_t gpgd) +{ + if ((gpgd.flags & ~_PAGE_TABLE) || gpgd.pfn >= lg->pfn_limit) + kill_guest(lg, "bad page directory entry"); +} + +/* FIXME: We hold reference to pages, which prevents them from being + swapped. It'd be nice to have a callback when Linux wants to swap out. */ + +/* We fault pages in, which allows us to update accessed/dirty bits. + * Return true if we got page. */ +int demand_page(struct lguest *lg, unsigned long vaddr, int errcode) +{ + gpgd_t gpgd; + spgd_t *spgd; + unsigned long gpte_ptr; + gpte_t gpte; + spte_t *spte; + + gpgd = mkgpgd(lgread_u32(lg, gpgd_addr(lg, vaddr))); + if (!(gpgd.flags & _PAGE_PRESENT)) + return 0; + + spgd = spgd_addr(lg, lg->pgdidx, vaddr); + if (!(spgd->flags & _PAGE_PRESENT)) { + /* Get a page of PTEs for them. */ + unsigned long ptepage = get_zeroed_page(GFP_KERNEL); + /* FIXME: Steal from self in this case? */ + if (!ptepage) { + kill_guest(lg, "out of memory allocating pte page"); + return 0; + } + check_gpgd(lg, gpgd); + spgd->raw.val = (__pa(ptepage) | gpgd.flags); + } + + gpte_ptr = gpte_addr(lg, gpgd, vaddr); + gpte = mkgpte(lgread_u32(lg, gpte_ptr)); + + /* No page? */ + if (!(gpte.flags & _PAGE_PRESENT)) + return 0; + + /* Write to read-only page? */ + if ((errcode & 2) && !(gpte.flags & _PAGE_RW)) + return 0; + + /* User access to a non-user page? */ + if ((errcode & 4) && !(gpte.flags & _PAGE_USER)) + return 0; + + check_gpte(lg, gpte); + gpte.flags |= _PAGE_ACCESSED; + if (errcode & 2) + gpte.flags |= _PAGE_DIRTY; + + /* We're done with the old pte. */ + spte = spte_addr(lg, *spgd, vaddr); + release_pte(*spte); + + /* We don't make it writable if this isn't a write: later + * write will fault so we can set dirty bit in guest. */ + if (gpte.flags & _PAGE_DIRTY) + *spte = gpte_to_spte(lg, gpte, 1); + else { + gpte_t ro_gpte = gpte; + ro_gpte.flags &= ~_PAGE_RW; + *spte = gpte_to_spte(lg, ro_gpte, 0); + } + + /* Now we update dirty/accessed on guest. */ + lgwrite_u32(lg, gpte_ptr, gpte.raw.val); + return 1; +} + +/* This is much faster than the full demand_page logic. */ +static int page_writable(struct lguest *lg, unsigned long vaddr) +{ + spgd_t *spgd; + unsigned long flags; + + spgd = spgd_addr(lg, lg->pgdidx, vaddr); + if (!(spgd->flags & _PAGE_PRESENT)) + return 0; + + flags = spte_addr(lg, *spgd, vaddr)->flags; + return (flags & (_PAGE_PRESENT|_PAGE_RW)) == (_PAGE_PRESENT|_PAGE_RW); +} + +void pin_page(struct lguest *lg, unsigned long vaddr) +{ + if (!page_writable(lg, vaddr) && !demand_page(lg, vaddr, 2)) + kill_guest(lg, "bad stack page %#lx", vaddr); +} + +static void release_pgd(struct lguest *lg, spgd_t *spgd) +{ + if (spgd->flags & _PAGE_PRESENT) { + unsigned int i; + spte_t *ptepage = __va(spgd->pfn << PAGE_SHIFT); + for (i = 0; i < PTES_PER_PAGE; i++) + release_pte(ptepage[i]); + free_page((long)ptepage); + spgd->raw.val = 0; + } +} + +static void flush_user_mappings(struct lguest *lg, int idx) +{ + unsigned int i; + for (i = 0; i < vaddr_to_pgd_index(lg->page_offset); i++) + release_pgd(lg, lg->pgdirs[idx].pgdir + i); +} + +void guest_pagetable_flush_user(struct lguest *lg) +{ + flush_user_mappings(lg, lg->pgdidx); +} + +static unsigned int find_pgdir(struct lguest *lg, unsigned long pgtable) +{ + unsigned int i; + for (i = 0; i < ARRAY_SIZE(lg->pgdirs); i++) + if (lg->pgdirs[i].cr3 == pgtable) + break; + return i; +} + +static unsigned int new_pgdir(struct lguest *lg, + unsigned long cr3, + int *blank_pgdir) +{ + unsigned int next; + + next = random32() % ARRAY_SIZE(lg->pgdirs); + if (!lg->pgdirs[next].pgdir) { + lg->pgdirs[next].pgdir = (spgd_t *)get_zeroed_page(GFP_KERNEL); + if (!lg->pgdirs[next].pgdir) + next = lg->pgdidx; + else + /* There are no mappings: you'll need to re-pin */ + *blank_pgdir = 1; + } + lg->pgdirs[next].cr3 = cr3; + /* Release all the non-kernel mappings. */ + flush_user_mappings(lg, next); + + return next; +} + +void guest_new_pagetable(struct lguest *lg, unsigned long pgtable) +{ + int newpgdir, repin = 0; + + newpgdir = find_pgdir(lg, pgtable); + if (newpgdir == ARRAY_SIZE(lg->pgdirs)) + newpgdir = new_pgdir(lg, pgtable, &repin); + lg->pgdidx = newpgdir; + if (repin) + pin_stack_pages(lg); +} + +static void release_all_pagetables(struct lguest *lg) +{ + unsigned int i, j; + + for (i = 0; i < ARRAY_SIZE(lg->pgdirs); i++) + if (lg->pgdirs[i].pgdir) + for (j = 0; j < SWITCHER_PGD_INDEX; j++) + release_pgd(lg, lg->pgdirs[i].pgdir + j); +} + +void guest_pagetable_clear_all(struct lguest *lg) +{ + release_all_pagetables(lg); + pin_stack_pages(lg); +} + +static void do_set_pte(struct lguest *lg, int idx, + unsigned long vaddr, gpte_t gpte) +{ + spgd_t *spgd = spgd_addr(lg, idx, vaddr); + if (spgd->flags & _PAGE_PRESENT) { + spte_t *spte = spte_addr(lg, *spgd, vaddr); + release_pte(*spte); + if (gpte.flags & (_PAGE_DIRTY | _PAGE_ACCESSED)) { + check_gpte(lg, gpte); + *spte = gpte_to_spte(lg, gpte, gpte.flags&_PAGE_DIRTY); + } else + spte->raw.val = 0; + } +} + +void guest_set_pte(struct lguest *lg, + unsigned long cr3, unsigned long vaddr, gpte_t gpte) +{ + /* Kernel mappings must be changed on all top levels. */ + if (vaddr >= lg->page_offset) { + unsigned int i; + for (i = 0; i < ARRAY_SIZE(lg->pgdirs); i++) + if (lg->pgdirs[i].pgdir) + do_set_pte(lg, i, vaddr, gpte); + } else { + int pgdir = find_pgdir(lg, cr3); + if (pgdir != ARRAY_SIZE(lg->pgdirs)) + do_set_pte(lg, pgdir, vaddr, gpte); + } +} + +void guest_set_pmd(struct lguest *lg, unsigned long cr3, u32 idx) +{ + int pgdir; + + if (idx >= SWITCHER_PGD_INDEX) + return; + + pgdir = find_pgdir(lg, cr3); + if (pgdir < ARRAY_SIZE(lg->pgdirs)) + release_pgd(lg, lg->pgdirs[pgdir].pgdir + idx); +} + +int init_guest_pagetable(struct lguest *lg, unsigned long pgtable) +{ + /* We assume this in flush_user_mappings, so check now */ + if (vaddr_to_pgd_index(lg->page_offset) >= SWITCHER_PGD_INDEX) + return -EINVAL; + lg->pgdidx = 0; + lg->pgdirs[lg->pgdidx].cr3 = pgtable; + lg->pgdirs[lg->pgdidx].pgdir = (spgd_t*)get_zeroed_page(GFP_KERNEL); + if (!lg->pgdirs[lg->pgdidx].pgdir) + return -ENOMEM; + return 0; +} + +void free_guest_pagetable(struct lguest *lg) +{ + unsigned int i; + + release_all_pagetables(lg); + for (i = 0; i < ARRAY_SIZE(lg->pgdirs); i++) + free_page((long)lg->pgdirs[i].pgdir); +} + +/* Caller must be preempt-safe */ +void map_switcher_in_guest(struct lguest *lg, struct lguest_pages *pages) +{ + spte_t *switcher_pte_page = __get_cpu_var(switcher_pte_pages); + spgd_t switcher_pgd; + spte_t regs_pte; + + /* Since switcher less that 4MB, we simply mug top pte page. */ + switcher_pgd.pfn = __pa(switcher_pte_page) >> PAGE_SHIFT; + switcher_pgd.flags = _PAGE_KERNEL; + lg->pgdirs[lg->pgdidx].pgdir[SWITCHER_PGD_INDEX] = switcher_pgd; + + /* Map our regs page over stack page. */ + regs_pte.pfn = __pa(lg->regs_page) >> PAGE_SHIFT; + regs_pte.flags = _PAGE_KERNEL; + switcher_pte_page[(unsigned long)pages/PAGE_SIZE%PTES_PER_PAGE] + = regs_pte; +} + +static void free_switcher_pte_pages(void) +{ + unsigned int i; + + for_each_possible_cpu(i) + free_page((long)switcher_pte_page(i)); +} + +static __init void populate_switcher_pte_page(unsigned int cpu, + struct page *switcher_page[], + unsigned int pages) +{ + unsigned int i; + spte_t *pte = switcher_pte_page(cpu); + + for (i = 0; i < pages; i++) { + pte[i].pfn = page_to_pfn(switcher_page[i]); + pte[i].flags = _PAGE_PRESENT|_PAGE_ACCESSED; + } + + /* We only map this CPU's pages, so guest can't see others. */ + i = pages + cpu*2; + + /* First page (regs) is rw, second (state) is ro. */ + pte[i].pfn = page_to_pfn(switcher_page[i]); + pte[i].flags = _PAGE_PRESENT|_PAGE_ACCESSED|_PAGE_RW; + pte[i+1].pfn = page_to_pfn(switcher_page[i+1]); + pte[i+1].flags = _PAGE_PRESENT|_PAGE_ACCESSED; +} + +__init int init_pagetables(struct page **switcher_page, unsigned int pages) +{ + unsigned int i; + + for_each_possible_cpu(i) { + switcher_pte_page(i) = (spte_t *)get_zeroed_page(GFP_KERNEL); + if (!switcher_pte_page(i)) { + free_switcher_pte_pages(); + return -ENOMEM; + } + populate_switcher_pte_page(i, switcher_page, pages); + } + return 0; +} + +void free_pagetables(void) +{ + free_switcher_pte_pages(); +} diff --git a/drivers/lguest/segments.c b/drivers/lguest/segments.c new file mode 100644 index 00000000000..1b2cfe89dcd --- /dev/null +++ b/drivers/lguest/segments.c @@ -0,0 +1,125 @@ +#include "lg.h" + +static int desc_ok(const struct desc_struct *gdt) +{ + /* MBZ=0, P=1, DT=1 */ + return ((gdt->b & 0x00209000) == 0x00009000); +} + +static int segment_present(const struct desc_struct *gdt) +{ + return gdt->b & 0x8000; +} + +static int ignored_gdt(unsigned int num) +{ + return (num == GDT_ENTRY_TSS + || num == GDT_ENTRY_LGUEST_CS + || num == GDT_ENTRY_LGUEST_DS + || num == GDT_ENTRY_DOUBLEFAULT_TSS); +} + +/* We don't allow removal of CS, DS or SS; it doesn't make sense. */ +static void check_segment_use(struct lguest *lg, unsigned int desc) +{ + if (lg->regs->gs / 8 == desc) + lg->regs->gs = 0; + if (lg->regs->fs / 8 == desc) + lg->regs->fs = 0; + if (lg->regs->es / 8 == desc) + lg->regs->es = 0; + if (lg->regs->ds / 8 == desc + || lg->regs->cs / 8 == desc + || lg->regs->ss / 8 == desc) + kill_guest(lg, "Removed live GDT entry %u", desc); +} + +static void fixup_gdt_table(struct lguest *lg, unsigned start, unsigned end) +{ + unsigned int i; + + for (i = start; i < end; i++) { + /* We never copy these ones to real gdt */ + if (ignored_gdt(i)) + continue; + + /* We could fault in switch_to_guest if they are using + * a removed segment. */ + if (!segment_present(&lg->gdt[i])) { + check_segment_use(lg, i); + continue; + } + + if (!desc_ok(&lg->gdt[i])) + kill_guest(lg, "Bad GDT descriptor %i", i); + + /* DPL 0 presumably means "for use by guest". */ + if ((lg->gdt[i].b & 0x00006000) == 0) + lg->gdt[i].b |= (GUEST_PL << 13); + + /* Set accessed bit, since gdt isn't writable. */ + lg->gdt[i].b |= 0x00000100; + } +} + +void setup_default_gdt_entries(struct lguest_ro_state *state) +{ + struct desc_struct *gdt = state->guest_gdt; + unsigned long tss = (unsigned long)&state->guest_tss; + + /* Hypervisor segments. */ + gdt[GDT_ENTRY_LGUEST_CS] = FULL_EXEC_SEGMENT; + gdt[GDT_ENTRY_LGUEST_DS] = FULL_SEGMENT; + + /* This is the one which we *cannot* copy from guest, since tss + is depended on this lguest_ro_state, ie. this cpu. */ + gdt[GDT_ENTRY_TSS].a = 0x00000067 | (tss << 16); + gdt[GDT_ENTRY_TSS].b = 0x00008900 | (tss & 0xFF000000) + | ((tss >> 16) & 0x000000FF); +} + +void setup_guest_gdt(struct lguest *lg) +{ + lg->gdt[GDT_ENTRY_KERNEL_CS] = FULL_EXEC_SEGMENT; + lg->gdt[GDT_ENTRY_KERNEL_DS] = FULL_SEGMENT; + lg->gdt[GDT_ENTRY_KERNEL_CS].b |= (GUEST_PL << 13); + lg->gdt[GDT_ENTRY_KERNEL_DS].b |= (GUEST_PL << 13); +} + +/* This is a fast version for the common case where only the three TLS entries + * have changed. */ +void copy_gdt_tls(const struct lguest *lg, struct desc_struct *gdt) +{ + unsigned int i; + + for (i = GDT_ENTRY_TLS_MIN; i <= GDT_ENTRY_TLS_MAX; i++) + gdt[i] = lg->gdt[i]; +} + +void copy_gdt(const struct lguest *lg, struct desc_struct *gdt) +{ + unsigned int i; + + for (i = 0; i < GDT_ENTRIES; i++) + if (!ignored_gdt(i)) + gdt[i] = lg->gdt[i]; +} + +void load_guest_gdt(struct lguest *lg, unsigned long table, u32 num) +{ + if (num > ARRAY_SIZE(lg->gdt)) + kill_guest(lg, "too many gdt entries %i", num); + + lgread(lg, lg->gdt, table, num * sizeof(lg->gdt[0])); + fixup_gdt_table(lg, 0, ARRAY_SIZE(lg->gdt)); + lg->changed |= CHANGED_GDT; +} + +void guest_load_tls(struct lguest *lg, unsigned long gtls) +{ + struct desc_struct *tls = &lg->gdt[GDT_ENTRY_TLS_MIN]; + + lgread(lg, tls, gtls, sizeof(*tls)*GDT_ENTRY_TLS_ENTRIES); + fixup_gdt_table(lg, GDT_ENTRY_TLS_MIN, GDT_ENTRY_TLS_MAX+1); + lg->changed |= CHANGED_GDT_TLS; +} diff --git a/drivers/lguest/switcher.S b/drivers/lguest/switcher.S new file mode 100644 index 00000000000..eadd4cc299d --- /dev/null +++ b/drivers/lguest/switcher.S @@ -0,0 +1,159 @@ +/* This code sits at 0xFFC00000 to do the low-level guest<->host switch. + + There is are two pages above us for this CPU (struct lguest_pages). + The second page (struct lguest_ro_state) becomes read-only after the + context switch. The first page (the stack for traps) remains writable, + but while we're in here, the guest cannot be running. +*/ +#include <linux/linkage.h> +#include <asm/asm-offsets.h> +#include "lg.h" + +.text +ENTRY(start_switcher_text) + +/* %eax points to lguest pages for this CPU. %ebx contains cr3 value. + All normal registers can be clobbered! */ +ENTRY(switch_to_guest) + /* Save host segments on host stack. */ + pushl %es + pushl %ds + pushl %gs + pushl %fs + /* With CONFIG_FRAME_POINTER, gcc doesn't let us clobber this! */ + pushl %ebp + /* Save host stack. */ + movl %esp, LGUEST_PAGES_host_sp(%eax) + /* Switch to guest stack: if we get NMI we expect to be there. */ + movl %eax, %edx + addl $LGUEST_PAGES_regs, %edx + movl %edx, %esp + /* Switch to guest's GDT, IDT. */ + lgdt LGUEST_PAGES_guest_gdt_desc(%eax) + lidt LGUEST_PAGES_guest_idt_desc(%eax) + /* Switch to guest's TSS while GDT still writable. */ + movl $(GDT_ENTRY_TSS*8), %edx + ltr %dx + /* Set host's TSS GDT entry to available (clear byte 5 bit 2). */ + movl (LGUEST_PAGES_host_gdt_desc+2)(%eax), %edx + andb $0xFD, (GDT_ENTRY_TSS*8 + 5)(%edx) + /* Switch to guest page tables: lguest_pages->state now read-only. */ + movl %ebx, %cr3 + /* Restore guest regs */ + popl %ebx + popl %ecx + popl %edx + popl %esi + popl %edi + popl %ebp + popl %gs + popl %eax + popl %fs + popl %ds + popl %es + /* Skip error code and trap number */ + addl $8, %esp + iret + +#define SWITCH_TO_HOST \ + /* Save guest state */ \ + pushl %es; \ + pushl %ds; \ + pushl %fs; \ + pushl %eax; \ + pushl %gs; \ + pushl %ebp; \ + pushl %edi; \ + pushl %esi; \ + pushl %edx; \ + pushl %ecx; \ + pushl %ebx; \ + /* Load lguest ds segment for convenience. */ \ + movl $(LGUEST_DS), %eax; \ + movl %eax, %ds; \ + /* Figure out where we are, based on stack (at top of regs). */ \ + movl %esp, %eax; \ + subl $LGUEST_PAGES_regs, %eax; \ + /* Put trap number in %ebx before we switch cr3 and lose it. */ \ + movl LGUEST_PAGES_regs_trapnum(%eax), %ebx; \ + /* Switch to host page tables (host GDT, IDT and stack are in host \ + mem, so need this first) */ \ + movl LGUEST_PAGES_host_cr3(%eax), %edx; \ + movl %edx, %cr3; \ + /* Set guest's TSS to available (clear byte 5 bit 2). */ \ + andb $0xFD, (LGUEST_PAGES_guest_gdt+GDT_ENTRY_TSS*8+5)(%eax); \ + /* Switch to host's GDT & IDT. */ \ + lgdt LGUEST_PAGES_host_gdt_desc(%eax); \ + lidt LGUEST_PAGES_host_idt_desc(%eax); \ + /* Switch to host's stack. */ \ + movl LGUEST_PAGES_host_sp(%eax), %esp; \ + /* Switch to host's TSS */ \ + movl $(GDT_ENTRY_TSS*8), %edx; \ + ltr %dx; \ + popl %ebp; \ + popl %fs; \ + popl %gs; \ + popl %ds; \ + popl %es + +/* Return to run_guest_once. */ +return_to_host: + SWITCH_TO_HOST + iret + +deliver_to_host: + SWITCH_TO_HOST + /* Decode IDT and jump to hosts' irq handler. When that does iret, it + * will return to run_guest_once. This is a feature. */ + movl (LGUEST_PAGES_host_idt_desc+2)(%eax), %edx + leal (%edx,%ebx,8), %eax + movzwl (%eax),%edx + movl 4(%eax), %eax + xorw %ax, %ax + orl %eax, %edx + jmp *%edx + +/* Real hardware interrupts are delivered straight to the host. Others + cause us to return to run_guest_once so it can decide what to do. Note + that some of these are overridden by the guest to deliver directly, and + never enter here (see load_guest_idt_entry). */ +.macro IRQ_STUB N TARGET + .data; .long 1f; .text; 1: + /* Make an error number for most traps, which don't have one. */ + .if (\N <> 8) && (\N < 10 || \N > 14) && (\N <> 17) + pushl $0 + .endif + pushl $\N + jmp \TARGET + ALIGN +.endm + +.macro IRQ_STUBS FIRST LAST TARGET + irq=\FIRST + .rept \LAST-\FIRST+1 + IRQ_STUB irq \TARGET + irq=irq+1 + .endr +.endm + +/* We intercept every interrupt, because we may need to switch back to + * host. Unfortunately we can't tell them apart except by entry + * point, so we need 256 entry points. + */ +.data +.global default_idt_entries +default_idt_entries: +.text + IRQ_STUBS 0 1 return_to_host /* First two traps */ + IRQ_STUB 2 handle_nmi /* NMI */ + IRQ_STUBS 3 31 return_to_host /* Rest of traps */ + IRQ_STUBS 32 127 deliver_to_host /* Real interrupts */ + IRQ_STUB 128 return_to_host /* System call (overridden) */ + IRQ_STUBS 129 255 deliver_to_host /* Other real interrupts */ + +/* We ignore NMI and return. */ +handle_nmi: + addl $8, %esp + iret + +ENTRY(end_switcher_text) diff --git a/drivers/macintosh/macio_asic.c b/drivers/macintosh/macio_asic.c index c96b7fe882a..ec9e5f32f0a 100644 --- a/drivers/macintosh/macio_asic.c +++ b/drivers/macintosh/macio_asic.c @@ -365,10 +365,9 @@ static struct macio_dev * macio_add_one_device(struct macio_chip *chip, if (np == NULL) return NULL; - dev = kmalloc(sizeof(*dev), GFP_KERNEL); + dev = kzalloc(sizeof(*dev), GFP_KERNEL); if (!dev) return NULL; - memset(dev, 0, sizeof(*dev)); dev->bus = &chip->lbus; dev->media_bay = in_bay; diff --git a/drivers/macintosh/smu.c b/drivers/macintosh/smu.c index f8e1a135bf9..d409f675948 100644 --- a/drivers/macintosh/smu.c +++ b/drivers/macintosh/smu.c @@ -1053,10 +1053,9 @@ static int smu_open(struct inode *inode, struct file *file) struct smu_private *pp; unsigned long flags; - pp = kmalloc(sizeof(struct smu_private), GFP_KERNEL); + pp = kzalloc(sizeof(struct smu_private), GFP_KERNEL); if (pp == 0) return -ENOMEM; - memset(pp, 0, sizeof(struct smu_private)); spin_lock_init(&pp->lock); pp->mode = smu_file_commands; init_waitqueue_head(&pp->wait); diff --git a/drivers/macintosh/therm_pm72.c b/drivers/macintosh/therm_pm72.c index 3d90fc00209..e43554e754a 100644 --- a/drivers/macintosh/therm_pm72.c +++ b/drivers/macintosh/therm_pm72.c @@ -318,10 +318,9 @@ static struct i2c_client *attach_i2c_chip(int id, const char *name) if (adap == NULL) return NULL; - clt = kmalloc(sizeof(struct i2c_client), GFP_KERNEL); + clt = kzalloc(sizeof(struct i2c_client), GFP_KERNEL); if (clt == NULL) return NULL; - memset(clt, 0, sizeof(struct i2c_client)); clt->addr = (id >> 1) & 0x7f; clt->adapter = adap; diff --git a/drivers/macintosh/therm_windtunnel.c b/drivers/macintosh/therm_windtunnel.c index 3d0354e96a9..5452da1bb1a 100644 --- a/drivers/macintosh/therm_windtunnel.c +++ b/drivers/macintosh/therm_windtunnel.c @@ -431,9 +431,8 @@ do_probe( struct i2c_adapter *adapter, int addr, int kind ) | I2C_FUNC_SMBUS_WRITE_BYTE) ) return 0; - if( !(cl=kmalloc(sizeof(*cl), GFP_KERNEL)) ) + if( !(cl=kzalloc(sizeof(*cl), GFP_KERNEL)) ) return -ENOMEM; - memset( cl, 0, sizeof(struct i2c_client) ); cl->addr = addr; cl->adapter = adapter; diff --git a/drivers/macintosh/windfarm_lm75_sensor.c b/drivers/macintosh/windfarm_lm75_sensor.c index a0fabf3c200..7e10c3ab4d5 100644 --- a/drivers/macintosh/windfarm_lm75_sensor.c +++ b/drivers/macintosh/windfarm_lm75_sensor.c @@ -117,10 +117,9 @@ static struct wf_lm75_sensor *wf_lm75_create(struct i2c_adapter *adapter, DBG("wf_lm75: creating %s device at address 0x%02x\n", ds1775 ? "ds1775" : "lm75", addr); - lm = kmalloc(sizeof(struct wf_lm75_sensor), GFP_KERNEL); + lm = kzalloc(sizeof(struct wf_lm75_sensor), GFP_KERNEL); if (lm == NULL) return NULL; - memset(lm, 0, sizeof(struct wf_lm75_sensor)); /* Usual rant about sensor names not beeing very consistent in * the device-tree, oh well ... diff --git a/drivers/md/dm-raid1.c b/drivers/md/dm-raid1.c index 1a876f9965e..144071e70a9 100644 --- a/drivers/md/dm-raid1.c +++ b/drivers/md/dm-raid1.c @@ -951,13 +951,12 @@ static struct mirror_set *alloc_context(unsigned int nr_mirrors, len = sizeof(*ms) + (sizeof(ms->mirror[0]) * nr_mirrors); - ms = kmalloc(len, GFP_KERNEL); + ms = kzalloc(len, GFP_KERNEL); if (!ms) { ti->error = "Cannot allocate mirror context"; return NULL; } - memset(ms, 0, len); spin_lock_init(&ms->lock); ms->ti = ti; diff --git a/drivers/media/dvb/cinergyT2/cinergyT2.c b/drivers/media/dvb/cinergyT2/cinergyT2.c index 5a1449f485c..28929b618e2 100644 --- a/drivers/media/dvb/cinergyT2/cinergyT2.c +++ b/drivers/media/dvb/cinergyT2/cinergyT2.c @@ -905,12 +905,11 @@ static int cinergyt2_probe (struct usb_interface *intf, struct cinergyt2 *cinergyt2; int err; - if (!(cinergyt2 = kmalloc (sizeof(struct cinergyt2), GFP_KERNEL))) { + if (!(cinergyt2 = kzalloc (sizeof(struct cinergyt2), GFP_KERNEL))) { dprintk(1, "out of memory?!?\n"); return -ENOMEM; } - memset (cinergyt2, 0, sizeof (struct cinergyt2)); usb_set_intfdata (intf, (void *) cinergyt2); mutex_init(&cinergyt2->sem); diff --git a/drivers/media/video/cpia2/cpia2_core.c b/drivers/media/video/cpia2/cpia2_core.c index 55aab8d3888..a76bd786cf1 100644 --- a/drivers/media/video/cpia2/cpia2_core.c +++ b/drivers/media/video/cpia2/cpia2_core.c @@ -2224,15 +2224,13 @@ struct camera_data *cpia2_init_camera_struct(void) { struct camera_data *cam; - cam = kmalloc(sizeof(*cam), GFP_KERNEL); + cam = kzalloc(sizeof(*cam), GFP_KERNEL); if (!cam) { ERR("couldn't kmalloc cpia2 struct\n"); return NULL; } - /* Default everything to 0 */ - memset(cam, 0, sizeof(struct camera_data)); cam->present = 1; mutex_init(&cam->busy_lock); diff --git a/drivers/media/video/msp3400-driver.c b/drivers/media/video/msp3400-driver.c index 507b1d4260e..11cfcf18ec3 100644 --- a/drivers/media/video/msp3400-driver.c +++ b/drivers/media/video/msp3400-driver.c @@ -812,10 +812,9 @@ static int msp_attach(struct i2c_adapter *adapter, int address, int kind) int msp_product, msp_prod_hi, msp_prod_lo; int msp_rom; - client = kmalloc(sizeof(*client), GFP_KERNEL); + client = kzalloc(sizeof(*client), GFP_KERNEL); if (client == NULL) return -ENOMEM; - memset(client, 0, sizeof(*client)); client->addr = address; client->adapter = adapter; client->driver = &i2c_driver; diff --git a/drivers/media/video/planb.c b/drivers/media/video/planb.c index 1455a8f4e93..4ab1af74a97 100644 --- a/drivers/media/video/planb.c +++ b/drivers/media/video/planb.c @@ -353,9 +353,8 @@ static int planb_prepare_open(struct planb *pb) * PLANB_DUMMY)*sizeof(struct dbdma_cmd) +(PLANB_MAXLINES*((PLANB_MAXPIXELS+7)& ~7))/8 +MAX_GBUFFERS*sizeof(unsigned int); - if ((pb->priv_space = kmalloc (size, GFP_KERNEL)) == 0) + if ((pb->priv_space = kzalloc (size, GFP_KERNEL)) == 0) return -ENOMEM; - memset ((void *) pb->priv_space, 0, size); pb->overlay_last1 = pb->ch1_cmd = (volatile struct dbdma_cmd *) DBDMA_ALIGN (pb->priv_space); pb->overlay_last2 = pb->ch2_cmd = pb->ch1_cmd + pb->tab_size; diff --git a/drivers/media/video/usbvideo/vicam.c b/drivers/media/video/usbvideo/vicam.c index 2d9c0dd3b73..ff555129c82 100644 --- a/drivers/media/video/usbvideo/vicam.c +++ b/drivers/media/video/usbvideo/vicam.c @@ -1130,13 +1130,12 @@ vicam_probe( struct usb_interface *intf, const struct usb_device_id *id) } if ((cam = - kmalloc(sizeof (struct vicam_camera), GFP_KERNEL)) == NULL) { + kzalloc(sizeof (struct vicam_camera), GFP_KERNEL)) == NULL) { printk(KERN_WARNING "could not allocate kernel memory for vicam_camera struct\n"); return -ENOMEM; } - memset(cam, 0, sizeof (struct vicam_camera)); cam->shutter_speed = 15; diff --git a/drivers/mfd/mcp-core.c b/drivers/mfd/mcp-core.c index 75f401d52fd..b4ed57e0272 100644 --- a/drivers/mfd/mcp-core.c +++ b/drivers/mfd/mcp-core.c @@ -200,9 +200,8 @@ struct mcp *mcp_host_alloc(struct device *parent, size_t size) { struct mcp *mcp; - mcp = kmalloc(sizeof(struct mcp) + size, GFP_KERNEL); + mcp = kzalloc(sizeof(struct mcp) + size, GFP_KERNEL); if (mcp) { - memset(mcp, 0, sizeof(struct mcp) + size); spin_lock_init(&mcp->lock); mcp->attached_device.parent = parent; mcp->attached_device.bus = &mcp_bus_type; diff --git a/drivers/mfd/ucb1x00-core.c b/drivers/mfd/ucb1x00-core.c index 149810a084f..e03f1bcd4f9 100644 --- a/drivers/mfd/ucb1x00-core.c +++ b/drivers/mfd/ucb1x00-core.c @@ -484,12 +484,11 @@ static int ucb1x00_probe(struct mcp *mcp) goto err_disable; } - ucb = kmalloc(sizeof(struct ucb1x00), GFP_KERNEL); + ucb = kzalloc(sizeof(struct ucb1x00), GFP_KERNEL); ret = -ENOMEM; if (!ucb) goto err_disable; - memset(ucb, 0, sizeof(struct ucb1x00)); ucb->cdev.class = &ucb1x00_class; ucb->cdev.dev = &mcp->attached_device; diff --git a/drivers/misc/asus-laptop.c b/drivers/misc/asus-laptop.c index 7798f590e5a..f7530605997 100644 --- a/drivers/misc/asus-laptop.c +++ b/drivers/misc/asus-laptop.c @@ -979,10 +979,9 @@ static int asus_hotk_add(struct acpi_device *device) printk(ASUS_NOTICE "Asus Laptop Support version %s\n", ASUS_LAPTOP_VERSION); - hotk = kmalloc(sizeof(struct asus_hotk), GFP_KERNEL); + hotk = kzalloc(sizeof(struct asus_hotk), GFP_KERNEL); if (!hotk) return -ENOMEM; - memset(hotk, 0, sizeof(struct asus_hotk)); hotk->handle = device->handle; strcpy(acpi_device_name(device), ASUS_HOTK_DEVICE_NAME); diff --git a/drivers/misc/ibmasm/command.c b/drivers/misc/ibmasm/command.c index b5df347c81b..6497872df52 100644 --- a/drivers/misc/ibmasm/command.c +++ b/drivers/misc/ibmasm/command.c @@ -41,18 +41,16 @@ struct command *ibmasm_new_command(struct service_processor *sp, size_t buffer_s if (buffer_size > IBMASM_CMD_MAX_BUFFER_SIZE) return NULL; - cmd = kmalloc(sizeof(struct command), GFP_KERNEL); + cmd = kzalloc(sizeof(struct command), GFP_KERNEL); if (cmd == NULL) return NULL; - memset(cmd, 0, sizeof(*cmd)); - cmd->buffer = kmalloc(buffer_size, GFP_KERNEL); + cmd->buffer = kzalloc(buffer_size, GFP_KERNEL); if (cmd->buffer == NULL) { kfree(cmd); return NULL; } - memset(cmd->buffer, 0, buffer_size); cmd->buffer_size = buffer_size; kobject_init(&cmd->kobj); diff --git a/drivers/misc/ibmasm/ibmasmfs.c b/drivers/misc/ibmasm/ibmasmfs.c index eb7b073734b..22a7e8ba211 100644 --- a/drivers/misc/ibmasm/ibmasmfs.c +++ b/drivers/misc/ibmasm/ibmasmfs.c @@ -563,11 +563,10 @@ static ssize_t remote_settings_file_write(struct file *file, const char __user * if (*offset != 0) return 0; - buff = kmalloc (count + 1, GFP_KERNEL); + buff = kzalloc (count + 1, GFP_KERNEL); if (!buff) return -ENOMEM; - memset(buff, 0x0, count + 1); if (copy_from_user(buff, ubuff, count)) { kfree(buff); diff --git a/drivers/misc/ibmasm/module.c b/drivers/misc/ibmasm/module.c index fb03a853fac..4f9d4a9da98 100644 --- a/drivers/misc/ibmasm/module.c +++ b/drivers/misc/ibmasm/module.c @@ -77,13 +77,12 @@ static int __devinit ibmasm_init_one(struct pci_dev *pdev, const struct pci_devi /* vnc client won't work without bus-mastering */ pci_set_master(pdev); - sp = kmalloc(sizeof(struct service_processor), GFP_KERNEL); + sp = kzalloc(sizeof(struct service_processor), GFP_KERNEL); if (sp == NULL) { dev_err(&pdev->dev, "Failed to allocate memory\n"); result = -ENOMEM; goto error_kmalloc; } - memset(sp, 0, sizeof(struct service_processor)); spin_lock_init(&sp->lock); INIT_LIST_HEAD(&sp->command_queue); diff --git a/drivers/mmc/card/block.c b/drivers/mmc/card/block.c index cbd4b6e3e17..93fe2e5dd61 100644 --- a/drivers/mmc/card/block.c +++ b/drivers/mmc/card/block.c @@ -414,13 +414,12 @@ static struct mmc_blk_data *mmc_blk_alloc(struct mmc_card *card) return ERR_PTR(-ENOSPC); __set_bit(devidx, dev_use); - md = kmalloc(sizeof(struct mmc_blk_data), GFP_KERNEL); + md = kzalloc(sizeof(struct mmc_blk_data), GFP_KERNEL); if (!md) { ret = -ENOMEM; goto out; } - memset(md, 0, sizeof(struct mmc_blk_data)); /* * Set the read-only status based on the supported commands diff --git a/drivers/net/Makefile b/drivers/net/Makefile index aaaa0a04bb4..336af0635df 100644 --- a/drivers/net/Makefile +++ b/drivers/net/Makefile @@ -177,6 +177,7 @@ obj-$(CONFIG_ZORRO8390) += zorro8390.o obj-$(CONFIG_HPLANCE) += hplance.o 7990.o obj-$(CONFIG_MVME147_NET) += mvme147.o 7990.o obj-$(CONFIG_EQUALIZER) += eql.o +obj-$(CONFIG_LGUEST_GUEST) += lguest_net.o obj-$(CONFIG_MIPS_JAZZ_SONIC) += jazzsonic.o obj-$(CONFIG_MIPS_AU1X00_ENET) += au1000_eth.o obj-$(CONFIG_MIPS_SIM_NET) += mipsnet.o diff --git a/drivers/net/b44.c b/drivers/net/b44.c index 96fb0ec905a..37f1b6ff5c1 100644 --- a/drivers/net/b44.c +++ b/drivers/net/b44.c @@ -1519,14 +1519,13 @@ static void b44_setup_pseudo_magicp(struct b44 *bp) u8 *pwol_pattern; u8 pwol_mask[B44_PMASK_SIZE]; - pwol_pattern = kmalloc(B44_PATTERN_SIZE, GFP_KERNEL); + pwol_pattern = kzalloc(B44_PATTERN_SIZE, GFP_KERNEL); if (!pwol_pattern) { printk(KERN_ERR PFX "Memory not available for WOL\n"); return; } /* Ipv4 magic packet pattern - pattern 0.*/ - memset(pwol_pattern, 0, B44_PATTERN_SIZE); memset(pwol_mask, 0, B44_PMASK_SIZE); plen0 = b44_magic_pattern(bp->dev->dev_addr, pwol_pattern, pwol_mask, B44_ETHIPV4UDP_HLEN); diff --git a/drivers/net/bsd_comp.c b/drivers/net/bsd_comp.c index 7845eaf6f29..202d4a4ef75 100644 --- a/drivers/net/bsd_comp.c +++ b/drivers/net/bsd_comp.c @@ -395,14 +395,13 @@ static void *bsd_alloc (unsigned char *options, int opt_len, int decomp) * Allocate the main control structure for this instance. */ maxmaxcode = MAXCODE(bits); - db = kmalloc(sizeof (struct bsd_db), + db = kzalloc(sizeof (struct bsd_db), GFP_KERNEL); if (!db) { return NULL; } - memset (db, 0, sizeof(struct bsd_db)); /* * Allocate space for the dictionary. This may be more than one page in * length. diff --git a/drivers/net/forcedeth.c b/drivers/net/forcedeth.c index 136827f8dc2..6d1d50a1978 100644 --- a/drivers/net/forcedeth.c +++ b/drivers/net/forcedeth.c @@ -5137,12 +5137,10 @@ static int __devinit nv_probe(struct pci_dev *pci_dev, const struct pci_device_i goto out_unmap; np->tx_ring.ex = &np->rx_ring.ex[np->rx_ring_size]; } - np->rx_skb = kmalloc(sizeof(struct nv_skb_map) * np->rx_ring_size, GFP_KERNEL); - np->tx_skb = kmalloc(sizeof(struct nv_skb_map) * np->tx_ring_size, GFP_KERNEL); + np->rx_skb = kcalloc(np->rx_ring_size, sizeof(struct nv_skb_map), GFP_KERNEL); + np->tx_skb = kcalloc(np->tx_ring_size, sizeof(struct nv_skb_map), GFP_KERNEL); if (!np->rx_skb || !np->tx_skb) goto out_freering; - memset(np->rx_skb, 0, sizeof(struct nv_skb_map) * np->rx_ring_size); - memset(np->tx_skb, 0, sizeof(struct nv_skb_map) * np->tx_ring_size); dev->open = nv_open; dev->stop = nv_close; diff --git a/drivers/net/hamradio/dmascc.c b/drivers/net/hamradio/dmascc.c index 3be8c504759..205f0967249 100644 --- a/drivers/net/hamradio/dmascc.c +++ b/drivers/net/hamradio/dmascc.c @@ -453,8 +453,8 @@ static int __init setup_adapter(int card_base, int type, int n) int scc_base = card_base + hw[type].scc_offset; char *chipnames[] = CHIPNAMES; - /* Allocate memory */ - info = kmalloc(sizeof(struct scc_info), GFP_KERNEL | GFP_DMA); + /* Initialize what is necessary for write_scc and write_scc_data */ + info = kzalloc(sizeof(struct scc_info), GFP_KERNEL | GFP_DMA); if (!info) { printk(KERN_ERR "dmascc: " "could not allocate memory for %s at %#3x\n", @@ -462,8 +462,6 @@ static int __init setup_adapter(int card_base, int type, int n) goto out; } - /* Initialize what is necessary for write_scc and write_scc_data */ - memset(info, 0, sizeof(struct scc_info)); info->dev[0] = alloc_netdev(0, "", dev_setup); if (!info->dev[0]) { diff --git a/drivers/net/irda/irport.c b/drivers/net/irda/irport.c index 3078c419cb0..20732458f5a 100644 --- a/drivers/net/irda/irport.c +++ b/drivers/net/irda/irport.c @@ -164,14 +164,13 @@ irport_open(int i, unsigned int iobase, unsigned int irq) /* Allocate memory if needed */ if (self->tx_buff.truesize > 0) { - self->tx_buff.head = kmalloc(self->tx_buff.truesize, + self->tx_buff.head = kzalloc(self->tx_buff.truesize, GFP_KERNEL); if (self->tx_buff.head == NULL) { IRDA_ERROR("%s(), can't allocate memory for " "transmit buffer!\n", __FUNCTION__); goto err_out4; } - memset(self->tx_buff.head, 0, self->tx_buff.truesize); } self->tx_buff.data = self->tx_buff.head; diff --git a/drivers/net/irda/irtty-sir.c b/drivers/net/irda/irtty-sir.c index ad1857364d5..6f5f697ec9f 100644 --- a/drivers/net/irda/irtty-sir.c +++ b/drivers/net/irda/irtty-sir.c @@ -505,10 +505,9 @@ static int irtty_open(struct tty_struct *tty) } /* allocate private device info block */ - priv = kmalloc(sizeof(*priv), GFP_KERNEL); + priv = kzalloc(sizeof(*priv), GFP_KERNEL); if (!priv) goto out_put; - memset(priv, 0, sizeof(*priv)); priv->magic = IRTTY_MAGIC; priv->tty = tty; diff --git a/drivers/net/iseries_veth.c b/drivers/net/iseries_veth.c index 347d50cd77d..0433c41f902 100644 --- a/drivers/net/iseries_veth.c +++ b/drivers/net/iseries_veth.c @@ -822,10 +822,9 @@ static int veth_init_connection(u8 rlp) || ! HvLpConfig_doLpsCommunicateOnVirtualLan(this_lp, rlp) ) return 0; - cnx = kmalloc(sizeof(*cnx), GFP_KERNEL); + cnx = kzalloc(sizeof(*cnx), GFP_KERNEL); if (! cnx) return -ENOMEM; - memset(cnx, 0, sizeof(*cnx)); cnx->remote_lp = rlp; spin_lock_init(&cnx->lock); @@ -852,14 +851,13 @@ static int veth_init_connection(u8 rlp) if (rc != 0) return rc; - msgs = kmalloc(VETH_NUMBUFFERS * sizeof(struct veth_msg), GFP_KERNEL); + msgs = kcalloc(VETH_NUMBUFFERS, sizeof(struct veth_msg), GFP_KERNEL); if (! msgs) { veth_error("Can't allocate buffers for LPAR %d.\n", rlp); return -ENOMEM; } cnx->msgs = msgs; - memset(msgs, 0, VETH_NUMBUFFERS * sizeof(struct veth_msg)); for (i = 0; i < VETH_NUMBUFFERS; i++) { msgs[i].token = i; diff --git a/drivers/net/lance.c b/drivers/net/lance.c index a2f37e52b92..a4e5fab1262 100644 --- a/drivers/net/lance.c +++ b/drivers/net/lance.c @@ -533,11 +533,10 @@ static int __init lance_probe1(struct net_device *dev, int ioaddr, int irq, int dev->base_addr = ioaddr; /* Make certain the data structures used by the LANCE are aligned and DMAble. */ - lp = kmalloc(sizeof(*lp), GFP_DMA | GFP_KERNEL); + lp = kzalloc(sizeof(*lp), GFP_DMA | GFP_KERNEL); if(lp==NULL) return -ENODEV; if (lance_debug > 6) printk(" (#0x%05lx)", (unsigned long)lp); - memset(lp, 0, sizeof(*lp)); dev->priv = lp; lp->name = chipname; lp->rx_buffs = (unsigned long)kmalloc(PKT_BUF_SZ*RX_RING_SIZE, diff --git a/drivers/net/lguest_net.c b/drivers/net/lguest_net.c new file mode 100644 index 00000000000..112778652f7 --- /dev/null +++ b/drivers/net/lguest_net.c @@ -0,0 +1,354 @@ +/* A simple network driver for lguest. + * + * Copyright 2006 Rusty Russell <rusty@rustcorp.com.au> IBM Corporation + * + * 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. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA + */ +//#define DEBUG +#include <linux/netdevice.h> +#include <linux/etherdevice.h> +#include <linux/module.h> +#include <linux/mm_types.h> +#include <linux/io.h> +#include <linux/lguest_bus.h> + +#define SHARED_SIZE PAGE_SIZE +#define MAX_LANS 4 +#define NUM_SKBS 8 + +struct lguestnet_info +{ + /* The shared page(s). */ + struct lguest_net *peer; + unsigned long peer_phys; + unsigned long mapsize; + + /* The lguest_device I come from */ + struct lguest_device *lgdev; + + /* My peerid. */ + unsigned int me; + + /* Receive queue. */ + struct sk_buff *skb[NUM_SKBS]; + struct lguest_dma dma[NUM_SKBS]; +}; + +/* How many bytes left in this page. */ +static unsigned int rest_of_page(void *data) +{ + return PAGE_SIZE - ((unsigned long)data % PAGE_SIZE); +} + +/* Simple convention: offset 4 * peernum. */ +static unsigned long peer_key(struct lguestnet_info *info, unsigned peernum) +{ + return info->peer_phys + 4 * peernum; +} + +static void skb_to_dma(const struct sk_buff *skb, unsigned int headlen, + struct lguest_dma *dma) +{ + unsigned int i, seg; + + for (i = seg = 0; i < headlen; seg++, i += rest_of_page(skb->data+i)) { + dma->addr[seg] = virt_to_phys(skb->data + i); + dma->len[seg] = min((unsigned)(headlen - i), + rest_of_page(skb->data + i)); + } + for (i = 0; i < skb_shinfo(skb)->nr_frags; i++, seg++) { + const skb_frag_t *f = &skb_shinfo(skb)->frags[i]; + /* Should not happen with MTU less than 64k - 2 * PAGE_SIZE. */ + if (seg == LGUEST_MAX_DMA_SECTIONS) { + printk("Woah dude! Megapacket!\n"); + break; + } + dma->addr[seg] = page_to_phys(f->page) + f->page_offset; + dma->len[seg] = f->size; + } + if (seg < LGUEST_MAX_DMA_SECTIONS) + dma->len[seg] = 0; +} + +/* We overload multicast bit to show promiscuous mode. */ +#define PROMISC_BIT 0x01 + +static void lguestnet_set_multicast(struct net_device *dev) +{ + struct lguestnet_info *info = netdev_priv(dev); + + if ((dev->flags & (IFF_PROMISC|IFF_ALLMULTI)) || dev->mc_count) + info->peer[info->me].mac[0] |= PROMISC_BIT; + else + info->peer[info->me].mac[0] &= ~PROMISC_BIT; +} + +static int promisc(struct lguestnet_info *info, unsigned int peer) +{ + return info->peer[peer].mac[0] & PROMISC_BIT; +} + +static int mac_eq(const unsigned char mac[ETH_ALEN], + struct lguestnet_info *info, unsigned int peer) +{ + /* Ignore multicast bit, which peer turns on to mean promisc. */ + if ((info->peer[peer].mac[0] & (~PROMISC_BIT)) != mac[0]) + return 0; + return memcmp(mac+1, info->peer[peer].mac+1, ETH_ALEN-1) == 0; +} + +static void transfer_packet(struct net_device *dev, + struct sk_buff *skb, + unsigned int peernum) +{ + struct lguestnet_info *info = netdev_priv(dev); + struct lguest_dma dma; + + skb_to_dma(skb, skb_headlen(skb), &dma); + pr_debug("xfer length %04x (%u)\n", htons(skb->len), skb->len); + + lguest_send_dma(peer_key(info, peernum), &dma); + if (dma.used_len != skb->len) { + dev->stats.tx_carrier_errors++; + pr_debug("Bad xfer to peer %i: %i of %i (dma %p/%i)\n", + peernum, dma.used_len, skb->len, + (void *)dma.addr[0], dma.len[0]); + } else { + dev->stats.tx_bytes += skb->len; + dev->stats.tx_packets++; + } +} + +static int unused_peer(const struct lguest_net peer[], unsigned int num) +{ + return peer[num].mac[0] == 0; +} + +static int lguestnet_start_xmit(struct sk_buff *skb, struct net_device *dev) +{ + unsigned int i; + int broadcast; + struct lguestnet_info *info = netdev_priv(dev); + const unsigned char *dest = ((struct ethhdr *)skb->data)->h_dest; + + pr_debug("%s: xmit %02x:%02x:%02x:%02x:%02x:%02x\n", + dev->name, dest[0],dest[1],dest[2],dest[3],dest[4],dest[5]); + + broadcast = is_multicast_ether_addr(dest); + for (i = 0; i < info->mapsize/sizeof(struct lguest_net); i++) { + if (i == info->me || unused_peer(info->peer, i)) + continue; + + if (!broadcast && !promisc(info, i) && !mac_eq(dest, info, i)) + continue; + + pr_debug("lguestnet %s: sending from %i to %i\n", + dev->name, info->me, i); + transfer_packet(dev, skb, i); + } + dev_kfree_skb(skb); + return 0; +} + +/* Find a new skb to put in this slot in shared mem. */ +static int fill_slot(struct net_device *dev, unsigned int slot) +{ + struct lguestnet_info *info = netdev_priv(dev); + /* Try to create and register a new one. */ + info->skb[slot] = netdev_alloc_skb(dev, ETH_HLEN + ETH_DATA_LEN); + if (!info->skb[slot]) { + printk("%s: could not fill slot %i\n", dev->name, slot); + return -ENOMEM; + } + + skb_to_dma(info->skb[slot], ETH_HLEN + ETH_DATA_LEN, &info->dma[slot]); + wmb(); + /* Now we tell hypervisor it can use the slot. */ + info->dma[slot].used_len = 0; + return 0; +} + +static irqreturn_t lguestnet_rcv(int irq, void *dev_id) +{ + struct net_device *dev = dev_id; + struct lguestnet_info *info = netdev_priv(dev); + unsigned int i, done = 0; + + for (i = 0; i < ARRAY_SIZE(info->dma); i++) { + unsigned int length; + struct sk_buff *skb; + + length = info->dma[i].used_len; + if (length == 0) + continue; + + done++; + skb = info->skb[i]; + fill_slot(dev, i); + + if (length < ETH_HLEN || length > ETH_HLEN + ETH_DATA_LEN) { + pr_debug(KERN_WARNING "%s: unbelievable skb len: %i\n", + dev->name, length); + dev_kfree_skb(skb); + continue; + } + + skb_put(skb, length); + skb->protocol = eth_type_trans(skb, dev); + /* This is a reliable transport. */ + if (dev->features & NETIF_F_NO_CSUM) + skb->ip_summed = CHECKSUM_UNNECESSARY; + pr_debug("Receiving skb proto 0x%04x len %i type %i\n", + ntohs(skb->protocol), skb->len, skb->pkt_type); + + dev->stats.rx_bytes += skb->len; + dev->stats.rx_packets++; + netif_rx(skb); + } + return done ? IRQ_HANDLED : IRQ_NONE; +} + +static int lguestnet_open(struct net_device *dev) +{ + int i; + struct lguestnet_info *info = netdev_priv(dev); + + /* Set up our MAC address */ + memcpy(info->peer[info->me].mac, dev->dev_addr, ETH_ALEN); + + /* Turn on promisc mode if needed */ + lguestnet_set_multicast(dev); + + for (i = 0; i < ARRAY_SIZE(info->dma); i++) { + if (fill_slot(dev, i) != 0) + goto cleanup; + } + if (lguest_bind_dma(peer_key(info,info->me), info->dma, + NUM_SKBS, lgdev_irq(info->lgdev)) != 0) + goto cleanup; + return 0; + +cleanup: + while (--i >= 0) + dev_kfree_skb(info->skb[i]); + return -ENOMEM; +} + +static int lguestnet_close(struct net_device *dev) +{ + unsigned int i; + struct lguestnet_info *info = netdev_priv(dev); + + /* Clear all trace: others might deliver packets, we'll ignore it. */ + memset(&info->peer[info->me], 0, sizeof(info->peer[info->me])); + + /* Deregister sg lists. */ + lguest_unbind_dma(peer_key(info, info->me), info->dma); + for (i = 0; i < ARRAY_SIZE(info->dma); i++) + dev_kfree_skb(info->skb[i]); + return 0; +} + +static int lguestnet_probe(struct lguest_device *lgdev) +{ + int err, irqf = IRQF_SHARED; + struct net_device *dev; + struct lguestnet_info *info; + struct lguest_device_desc *desc = &lguest_devices[lgdev->index]; + + pr_debug("lguest_net: probing for device %i\n", lgdev->index); + + dev = alloc_etherdev(sizeof(struct lguestnet_info)); + if (!dev) + return -ENOMEM; + + SET_MODULE_OWNER(dev); + + /* Ethernet defaults with some changes */ + ether_setup(dev); + dev->set_mac_address = NULL; + + dev->dev_addr[0] = 0x02; /* set local assignment bit (IEEE802) */ + dev->dev_addr[1] = 0x00; + memcpy(&dev->dev_addr[2], &lguest_data.guestid, 2); + dev->dev_addr[4] = 0x00; + dev->dev_addr[5] = 0x00; + + dev->open = lguestnet_open; + dev->stop = lguestnet_close; + dev->hard_start_xmit = lguestnet_start_xmit; + + /* Turning on/off promisc will call dev->set_multicast_list. + * We don't actually support multicast yet */ + dev->set_multicast_list = lguestnet_set_multicast; + SET_NETDEV_DEV(dev, &lgdev->dev); + if (desc->features & LGUEST_NET_F_NOCSUM) + dev->features = NETIF_F_SG|NETIF_F_NO_CSUM; + + info = netdev_priv(dev); + info->mapsize = PAGE_SIZE * desc->num_pages; + info->peer_phys = ((unsigned long)desc->pfn << PAGE_SHIFT); + info->lgdev = lgdev; + info->peer = lguest_map(info->peer_phys, desc->num_pages); + if (!info->peer) { + err = -ENOMEM; + goto free; + } + + /* This stores our peerid (upper bits reserved for future). */ + info->me = (desc->features & (info->mapsize-1)); + + err = register_netdev(dev); + if (err) { + pr_debug("lguestnet: registering device failed\n"); + goto unmap; + } + + if (lguest_devices[lgdev->index].features & LGUEST_DEVICE_F_RANDOMNESS) + irqf |= IRQF_SAMPLE_RANDOM; + if (request_irq(lgdev_irq(lgdev), lguestnet_rcv, irqf, "lguestnet", + dev) != 0) { + pr_debug("lguestnet: cannot get irq %i\n", lgdev_irq(lgdev)); + goto unregister; + } + + pr_debug("lguestnet: registered device %s\n", dev->name); + lgdev->private = dev; + return 0; + +unregister: + unregister_netdev(dev); +unmap: + lguest_unmap(info->peer); +free: + free_netdev(dev); + return err; +} + +static struct lguest_driver lguestnet_drv = { + .name = "lguestnet", + .owner = THIS_MODULE, + .device_type = LGUEST_DEVICE_T_NET, + .probe = lguestnet_probe, +}; + +static __init int lguestnet_init(void) +{ + return register_lguest_driver(&lguestnet_drv); +} +module_init(lguestnet_init); + +MODULE_DESCRIPTION("Lguest network driver"); +MODULE_LICENSE("GPL"); diff --git a/drivers/net/pcmcia/com20020_cs.c b/drivers/net/pcmcia/com20020_cs.c index 0d1c7a41c9c..ea9414c4d90 100644 --- a/drivers/net/pcmcia/com20020_cs.c +++ b/drivers/net/pcmcia/com20020_cs.c @@ -147,7 +147,7 @@ static int com20020_probe(struct pcmcia_device *p_dev) DEBUG(0, "com20020_attach()\n"); /* Create new network device */ - info = kmalloc(sizeof(struct com20020_dev_t), GFP_KERNEL); + info = kzalloc(sizeof(struct com20020_dev_t), GFP_KERNEL); if (!info) goto fail_alloc_info; @@ -155,7 +155,6 @@ static int com20020_probe(struct pcmcia_device *p_dev) if (!dev) goto fail_alloc_dev; - memset(info, 0, sizeof(struct com20020_dev_t)); lp = dev->priv; lp->timeout = timeout; lp->backplane = backplane; diff --git a/drivers/net/pcmcia/ibmtr_cs.c b/drivers/net/pcmcia/ibmtr_cs.c index 4ecb8ca5a99..4eafa4f42cf 100644 --- a/drivers/net/pcmcia/ibmtr_cs.c +++ b/drivers/net/pcmcia/ibmtr_cs.c @@ -146,9 +146,8 @@ static int __devinit ibmtr_attach(struct pcmcia_device *link) DEBUG(0, "ibmtr_attach()\n"); /* Create new token-ring device */ - info = kmalloc(sizeof(*info), GFP_KERNEL); + info = kzalloc(sizeof(*info), GFP_KERNEL); if (!info) return -ENOMEM; - memset(info,0,sizeof(*info)); dev = alloc_trdev(sizeof(struct tok_info)); if (!dev) { kfree(info); diff --git a/drivers/net/ppp_async.c b/drivers/net/ppp_async.c index caabbc408c3..27f5b904f48 100644 --- a/drivers/net/ppp_async.c +++ b/drivers/net/ppp_async.c @@ -159,12 +159,11 @@ ppp_asynctty_open(struct tty_struct *tty) int err; err = -ENOMEM; - ap = kmalloc(sizeof(*ap), GFP_KERNEL); + ap = kzalloc(sizeof(*ap), GFP_KERNEL); if (ap == 0) goto out; /* initialize the asyncppp structure */ - memset(ap, 0, sizeof(*ap)); ap->tty = tty; ap->mru = PPP_MRU; spin_lock_init(&ap->xmit_lock); diff --git a/drivers/net/ppp_deflate.c b/drivers/net/ppp_deflate.c index 72c8d6628f5..eb98b661efb 100644 --- a/drivers/net/ppp_deflate.c +++ b/drivers/net/ppp_deflate.c @@ -121,12 +121,11 @@ static void *z_comp_alloc(unsigned char *options, int opt_len) if (w_size < DEFLATE_MIN_SIZE || w_size > DEFLATE_MAX_SIZE) return NULL; - state = kmalloc(sizeof(*state), + state = kzalloc(sizeof(*state), GFP_KERNEL); if (state == NULL) return NULL; - memset (state, 0, sizeof (struct ppp_deflate_state)); state->strm.next_in = NULL; state->w_size = w_size; state->strm.workspace = vmalloc(zlib_deflate_workspacesize()); @@ -341,11 +340,10 @@ static void *z_decomp_alloc(unsigned char *options, int opt_len) if (w_size < DEFLATE_MIN_SIZE || w_size > DEFLATE_MAX_SIZE) return NULL; - state = kmalloc(sizeof(*state), GFP_KERNEL); + state = kzalloc(sizeof(*state), GFP_KERNEL); if (state == NULL) return NULL; - memset (state, 0, sizeof (struct ppp_deflate_state)); state->w_size = w_size; state->strm.next_out = NULL; state->strm.workspace = kmalloc(zlib_inflate_workspacesize(), diff --git a/drivers/net/ppp_generic.c b/drivers/net/ppp_generic.c index 3ef0092dc09..ef3325b6923 100644 --- a/drivers/net/ppp_generic.c +++ b/drivers/net/ppp_generic.c @@ -2684,8 +2684,7 @@ static void __exit ppp_cleanup(void) if (atomic_read(&ppp_unit_count) || atomic_read(&channel_count)) printk(KERN_ERR "PPP: removing module but units remain!\n"); cardmap_destroy(&all_ppp_units); - if (unregister_chrdev(PPP_MAJOR, "ppp") != 0) - printk(KERN_ERR "PPP: failed to unregister PPP device\n"); + unregister_chrdev(PPP_MAJOR, "ppp"); device_destroy(ppp_class, MKDEV(PPP_MAJOR, 0)); class_destroy(ppp_class); } diff --git a/drivers/net/ppp_mppe.c b/drivers/net/ppp_mppe.c index d5bdd257465..f79cf87a2bf 100644 --- a/drivers/net/ppp_mppe.c +++ b/drivers/net/ppp_mppe.c @@ -200,11 +200,10 @@ static void *mppe_alloc(unsigned char *options, int optlen) || options[0] != CI_MPPE || options[1] != CILEN_MPPE) goto out; - state = kmalloc(sizeof(*state), GFP_KERNEL); + state = kzalloc(sizeof(*state), GFP_KERNEL); if (state == NULL) goto out; - memset(state, 0, sizeof(*state)); state->arc4 = crypto_alloc_blkcipher("ecb(arc4)", 0, CRYPTO_ALG_ASYNC); if (IS_ERR(state->arc4)) { diff --git a/drivers/net/ppp_synctty.c b/drivers/net/ppp_synctty.c index 5918fab3834..ce64032a465 100644 --- a/drivers/net/ppp_synctty.c +++ b/drivers/net/ppp_synctty.c @@ -207,13 +207,12 @@ ppp_sync_open(struct tty_struct *tty) struct syncppp *ap; int err; - ap = kmalloc(sizeof(*ap), GFP_KERNEL); + ap = kzalloc(sizeof(*ap), GFP_KERNEL); err = -ENOMEM; if (ap == 0) goto out; /* initialize the syncppp structure */ - memset(ap, 0, sizeof(*ap)); ap->tty = tty; ap->mru = PPP_MRU; spin_lock_init(&ap->xmit_lock); diff --git a/drivers/net/shaper.c b/drivers/net/shaper.c index e886e8d7cfd..4c3d98ff4cd 100644 --- a/drivers/net/shaper.c +++ b/drivers/net/shaper.c @@ -600,10 +600,9 @@ static int __init shaper_init(void) return -ENODEV; alloc_size = sizeof(*dev) * shapers; - devs = kmalloc(alloc_size, GFP_KERNEL); + devs = kzalloc(alloc_size, GFP_KERNEL); if (!devs) return -ENOMEM; - memset(devs, 0, alloc_size); for (i = 0; i < shapers; i++) { diff --git a/drivers/net/tg3.c b/drivers/net/tg3.c index 5ee14764fd7..887b9a5cfe4 100644 --- a/drivers/net/tg3.c +++ b/drivers/net/tg3.c @@ -64,8 +64,8 @@ #define DRV_MODULE_NAME "tg3" #define PFX DRV_MODULE_NAME ": " -#define DRV_MODULE_VERSION "3.78" -#define DRV_MODULE_RELDATE "July 11, 2007" +#define DRV_MODULE_VERSION "3.79" +#define DRV_MODULE_RELDATE "July 18, 2007" #define TG3_DEF_MAC_MODE 0 #define TG3_DEF_RX_MODE 0 @@ -4847,6 +4847,59 @@ static int tg3_poll_fw(struct tg3 *tp) return 0; } +/* Save PCI command register before chip reset */ +static void tg3_save_pci_state(struct tg3 *tp) +{ + u32 val; + + pci_read_config_dword(tp->pdev, TG3PCI_COMMAND, &val); + tp->pci_cmd = val; +} + +/* Restore PCI state after chip reset */ +static void tg3_restore_pci_state(struct tg3 *tp) +{ + u32 val; + + /* Re-enable indirect register accesses. */ + pci_write_config_dword(tp->pdev, TG3PCI_MISC_HOST_CTRL, + tp->misc_host_ctrl); + + /* Set MAX PCI retry to zero. */ + val = (PCISTATE_ROM_ENABLE | PCISTATE_ROM_RETRY_ENABLE); + if (tp->pci_chip_rev_id == CHIPREV_ID_5704_A0 && + (tp->tg3_flags & TG3_FLAG_PCIX_MODE)) + val |= PCISTATE_RETRY_SAME_DMA; + pci_write_config_dword(tp->pdev, TG3PCI_PCISTATE, val); + + pci_write_config_dword(tp->pdev, TG3PCI_COMMAND, tp->pci_cmd); + + /* Make sure PCI-X relaxed ordering bit is clear. */ + pci_read_config_dword(tp->pdev, TG3PCI_X_CAPS, &val); + val &= ~PCIX_CAPS_RELAXED_ORDERING; + pci_write_config_dword(tp->pdev, TG3PCI_X_CAPS, val); + + if (tp->tg3_flags2 & TG3_FLG2_5780_CLASS) { + u32 val; + + /* Chip reset on 5780 will reset MSI enable bit, + * so need to restore it. + */ + if (tp->tg3_flags2 & TG3_FLG2_USING_MSI) { + u16 ctrl; + + pci_read_config_word(tp->pdev, + tp->msi_cap + PCI_MSI_FLAGS, + &ctrl); + pci_write_config_word(tp->pdev, + tp->msi_cap + PCI_MSI_FLAGS, + ctrl | PCI_MSI_FLAGS_ENABLE); + val = tr32(MSGINT_MODE); + tw32(MSGINT_MODE, val | MSGINT_MODE_ENABLE); + } + } +} + static void tg3_stop_fw(struct tg3 *); /* tp->lock is held. */ @@ -4863,6 +4916,12 @@ static int tg3_chip_reset(struct tg3 *tp) */ tp->nvram_lock_cnt = 0; + /* GRC_MISC_CFG core clock reset will clear the memory + * enable bit in PCI register 4 and the MSI enable bit + * on some chips, so we save relevant registers here. + */ + tg3_save_pci_state(tp); + if (GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5752 || GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5755 || GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5787) @@ -4961,50 +5020,14 @@ static int tg3_chip_reset(struct tg3 *tp) pci_write_config_dword(tp->pdev, 0xd8, 0xf5000); } - /* Re-enable indirect register accesses. */ - pci_write_config_dword(tp->pdev, TG3PCI_MISC_HOST_CTRL, - tp->misc_host_ctrl); - - /* Set MAX PCI retry to zero. */ - val = (PCISTATE_ROM_ENABLE | PCISTATE_ROM_RETRY_ENABLE); - if (tp->pci_chip_rev_id == CHIPREV_ID_5704_A0 && - (tp->tg3_flags & TG3_FLAG_PCIX_MODE)) - val |= PCISTATE_RETRY_SAME_DMA; - pci_write_config_dword(tp->pdev, TG3PCI_PCISTATE, val); - - pci_restore_state(tp->pdev); + tg3_restore_pci_state(tp); tp->tg3_flags &= ~TG3_FLAG_CHIP_RESETTING; - /* Make sure PCI-X relaxed ordering bit is clear. */ - pci_read_config_dword(tp->pdev, TG3PCI_X_CAPS, &val); - val &= ~PCIX_CAPS_RELAXED_ORDERING; - pci_write_config_dword(tp->pdev, TG3PCI_X_CAPS, val); - - if (tp->tg3_flags2 & TG3_FLG2_5780_CLASS) { - u32 val; - - /* Chip reset on 5780 will reset MSI enable bit, - * so need to restore it. - */ - if (tp->tg3_flags2 & TG3_FLG2_USING_MSI) { - u16 ctrl; - - pci_read_config_word(tp->pdev, - tp->msi_cap + PCI_MSI_FLAGS, - &ctrl); - pci_write_config_word(tp->pdev, - tp->msi_cap + PCI_MSI_FLAGS, - ctrl | PCI_MSI_FLAGS_ENABLE); - val = tr32(MSGINT_MODE); - tw32(MSGINT_MODE, val | MSGINT_MODE_ENABLE); - } - + val = 0; + if (tp->tg3_flags2 & TG3_FLG2_5780_CLASS) val = tr32(MEMARB_MODE); - tw32(MEMARB_MODE, val | MEMARB_MODE_ENABLE); - - } else - tw32(MEMARB_MODE, MEMARB_MODE_ENABLE); + tw32(MEMARB_MODE, val | MEMARB_MODE_ENABLE); if (tp->pci_chip_rev_id == CHIPREV_ID_5750_A3) { tg3_stop_fw(tp); @@ -11978,7 +12001,6 @@ static int __devinit tg3_init_one(struct pci_dev *pdev, */ if ((tr32(HOSTCC_MODE) & HOSTCC_MODE_ENABLE) || (tr32(WDMAC_MODE) & WDMAC_MODE_ENABLE)) { - pci_save_state(tp->pdev); tw32(MEMARB_MODE, MEMARB_MODE_ENABLE); tg3_halt(tp, RESET_KIND_SHUTDOWN, 1); } @@ -12007,12 +12029,6 @@ static int __devinit tg3_init_one(struct pci_dev *pdev, tg3_init_coal(tp); - /* Now that we have fully setup the chip, save away a snapshot - * of the PCI config space. We need to restore this after - * GRC_MISC_CFG core clock resets and some resume events. - */ - pci_save_state(tp->pdev); - pci_set_drvdata(pdev, dev); err = register_netdev(dev); diff --git a/drivers/net/tg3.h b/drivers/net/tg3.h index d84e75e7365..5c21f49026c 100644 --- a/drivers/net/tg3.h +++ b/drivers/net/tg3.h @@ -2345,6 +2345,7 @@ struct tg3 { #define PHY_REV_BCM5411_X0 0x1 /* Found on Netgear GA302T */ u32 led_ctrl; + u32 pci_cmd; char board_part_number[24]; char fw_ver[16]; diff --git a/drivers/net/wan/c101.c b/drivers/net/wan/c101.c index 6b63b350cd5..8ead774d14c 100644 --- a/drivers/net/wan/c101.c +++ b/drivers/net/wan/c101.c @@ -315,12 +315,11 @@ static int __init c101_run(unsigned long irq, unsigned long winbase) return -ENODEV; } - card = kmalloc(sizeof(card_t), GFP_KERNEL); + card = kzalloc(sizeof(card_t), GFP_KERNEL); if (card == NULL) { printk(KERN_ERR "c101: unable to allocate memory\n"); return -ENOBUFS; } - memset(card, 0, sizeof(card_t)); card->dev = alloc_hdlcdev(card); if (!card->dev) { diff --git a/drivers/net/wan/cosa.c b/drivers/net/wan/cosa.c index 9ef49ce148b..26058b4f8f3 100644 --- a/drivers/net/wan/cosa.c +++ b/drivers/net/wan/cosa.c @@ -572,13 +572,11 @@ static int cosa_probe(int base, int irq, int dma) sprintf(cosa->name, "cosa%d", cosa->num); /* Initialize the per-channel data */ - cosa->chan = kmalloc(sizeof(struct channel_data)*cosa->nchannels, - GFP_KERNEL); + cosa->chan = kcalloc(cosa->nchannels, sizeof(struct channel_data), GFP_KERNEL); if (!cosa->chan) { err = -ENOMEM; goto err_out3; } - memset(cosa->chan, 0, sizeof(struct channel_data)*cosa->nchannels); for (i=0; i<cosa->nchannels; i++) { cosa->chan[i].cosa = cosa; cosa->chan[i].num = i; diff --git a/drivers/net/wan/cycx_main.c b/drivers/net/wan/cycx_main.c index 6e5f1c89851..a0e8611ad8e 100644 --- a/drivers/net/wan/cycx_main.c +++ b/drivers/net/wan/cycx_main.c @@ -113,12 +113,10 @@ static int __init cycx_init(void) /* Verify number of cards and allocate adapter data space */ cycx_ncards = min_t(int, cycx_ncards, CYCX_MAX_CARDS); cycx_ncards = max_t(int, cycx_ncards, 1); - cycx_card_array = kmalloc(sizeof(struct cycx_device) * cycx_ncards, - GFP_KERNEL); + cycx_card_array = kcalloc(cycx_ncards, sizeof(struct cycx_device), GFP_KERNEL); if (!cycx_card_array) goto out; - memset(cycx_card_array, 0, sizeof(struct cycx_device) * cycx_ncards); /* Register adapters with WAN router */ for (cnt = 0; cnt < cycx_ncards; ++cnt) { diff --git a/drivers/net/wan/cycx_x25.c b/drivers/net/wan/cycx_x25.c index 016b3ff3ea5..a8af28b273d 100644 --- a/drivers/net/wan/cycx_x25.c +++ b/drivers/net/wan/cycx_x25.c @@ -376,11 +376,10 @@ static int cycx_wan_new_if(struct wan_device *wandev, struct net_device *dev, } /* allocate and initialize private data */ - chan = kmalloc(sizeof(struct cycx_x25_channel), GFP_KERNEL); + chan = kzalloc(sizeof(struct cycx_x25_channel), GFP_KERNEL); if (!chan) return -ENOMEM; - memset(chan, 0, sizeof(*chan)); strcpy(chan->name, conf->name); chan->card = card; chan->link = conf->port; diff --git a/drivers/net/wan/dscc4.c b/drivers/net/wan/dscc4.c index dca02447145..50d2f9108dc 100644 --- a/drivers/net/wan/dscc4.c +++ b/drivers/net/wan/dscc4.c @@ -890,12 +890,11 @@ static int dscc4_found1(struct pci_dev *pdev, void __iomem *ioaddr) struct dscc4_dev_priv *root; int i, ret = -ENOMEM; - root = kmalloc(dev_per_card*sizeof(*root), GFP_KERNEL); + root = kcalloc(dev_per_card, sizeof(*root), GFP_KERNEL); if (!root) { printk(KERN_ERR "%s: can't allocate data\n", DRV_NAME); goto err_out; } - memset(root, 0, dev_per_card*sizeof(*root)); for (i = 0; i < dev_per_card; i++) { root[i].dev = alloc_hdlcdev(root + i); @@ -903,12 +902,11 @@ static int dscc4_found1(struct pci_dev *pdev, void __iomem *ioaddr) goto err_free_dev; } - ppriv = kmalloc(sizeof(*ppriv), GFP_KERNEL); + ppriv = kzalloc(sizeof(*ppriv), GFP_KERNEL); if (!ppriv) { printk(KERN_ERR "%s: can't allocate private data\n", DRV_NAME); goto err_free_dev; } - memset(ppriv, 0, sizeof(struct dscc4_pci_priv)); ppriv->root = root; spin_lock_init(&ppriv->lock); diff --git a/drivers/net/wan/farsync.c b/drivers/net/wan/farsync.c index 58a53b6d9b4..12dae8e2484 100644 --- a/drivers/net/wan/farsync.c +++ b/drivers/net/wan/farsync.c @@ -2476,13 +2476,12 @@ fst_add_one(struct pci_dev *pdev, const struct pci_device_id *ent) } /* Allocate driver private data */ - card = kmalloc(sizeof (struct fst_card_info), GFP_KERNEL); + card = kzalloc(sizeof (struct fst_card_info), GFP_KERNEL); if (card == NULL) { printk_err("FarSync card found but insufficient memory for" " driver storage\n"); return -ENOMEM; } - memset(card, 0, sizeof (struct fst_card_info)); /* Try to enable the device */ if ((err = pci_enable_device(pdev)) != 0) { diff --git a/drivers/net/wan/hostess_sv11.c b/drivers/net/wan/hostess_sv11.c index 9ba3e4ee6ec..bf5f8d9b5c8 100644 --- a/drivers/net/wan/hostess_sv11.c +++ b/drivers/net/wan/hostess_sv11.c @@ -231,11 +231,10 @@ static struct sv11_device *sv11_init(int iobase, int irq) return NULL; } - sv = kmalloc(sizeof(struct sv11_device), GFP_KERNEL); + sv = kzalloc(sizeof(struct sv11_device), GFP_KERNEL); if(!sv) goto fail3; - memset(sv, 0, sizeof(*sv)); sv->if_ptr=&sv->netdev; sv->netdev.dev = alloc_netdev(0, "hdlc%d", sv11_setup); diff --git a/drivers/net/wan/n2.c b/drivers/net/wan/n2.c index 5c322dfb79f..cbdf0b748bd 100644 --- a/drivers/net/wan/n2.c +++ b/drivers/net/wan/n2.c @@ -351,12 +351,11 @@ static int __init n2_run(unsigned long io, unsigned long irq, return -ENODEV; } - card = kmalloc(sizeof(card_t), GFP_KERNEL); + card = kzalloc(sizeof(card_t), GFP_KERNEL); if (card == NULL) { printk(KERN_ERR "n2: unable to allocate memory\n"); return -ENOBUFS; } - memset(card, 0, sizeof(card_t)); card->ports[0].dev = alloc_hdlcdev(&card->ports[0]); card->ports[1].dev = alloc_hdlcdev(&card->ports[1]); diff --git a/drivers/net/wan/pc300_drv.c b/drivers/net/wan/pc300_drv.c index 5d8c78ee2cd..99fee2f1d01 100644 --- a/drivers/net/wan/pc300_drv.c +++ b/drivers/net/wan/pc300_drv.c @@ -3456,7 +3456,7 @@ cpc_init_one(struct pci_dev *pdev, const struct pci_device_id *ent) if ((err = pci_enable_device(pdev)) < 0) return err; - card = kmalloc(sizeof(pc300_t), GFP_KERNEL); + card = kzalloc(sizeof(pc300_t), GFP_KERNEL); if (card == NULL) { printk("PC300 found at RAM 0x%016llx, " "but could not allocate card structure.\n", @@ -3464,7 +3464,6 @@ cpc_init_one(struct pci_dev *pdev, const struct pci_device_id *ent) err = -ENOMEM; goto err_disable_dev; } - memset(card, 0, sizeof(pc300_t)); err = -ENODEV; diff --git a/drivers/net/wan/pc300too.c b/drivers/net/wan/pc300too.c index dfbd3b00f03..6353cb5c658 100644 --- a/drivers/net/wan/pc300too.c +++ b/drivers/net/wan/pc300too.c @@ -334,14 +334,13 @@ static int __devinit pc300_pci_init_one(struct pci_dev *pdev, return i; } - card = kmalloc(sizeof(card_t), GFP_KERNEL); + card = kzalloc(sizeof(card_t), GFP_KERNEL); if (card == NULL) { printk(KERN_ERR "pc300: unable to allocate memory\n"); pci_release_regions(pdev); pci_disable_device(pdev); return -ENOBUFS; } - memset(card, 0, sizeof(card_t)); pci_set_drvdata(pdev, card); if (pdev->device == PCI_DEVICE_ID_PC300_TE_1 || diff --git a/drivers/net/wan/pci200syn.c b/drivers/net/wan/pci200syn.c index 7f720de2e9f..092e51d8903 100644 --- a/drivers/net/wan/pci200syn.c +++ b/drivers/net/wan/pci200syn.c @@ -312,14 +312,13 @@ static int __devinit pci200_pci_init_one(struct pci_dev *pdev, return i; } - card = kmalloc(sizeof(card_t), GFP_KERNEL); + card = kzalloc(sizeof(card_t), GFP_KERNEL); if (card == NULL) { printk(KERN_ERR "pci200syn: unable to allocate memory\n"); pci_release_regions(pdev); pci_disable_device(pdev); return -ENOBUFS; } - memset(card, 0, sizeof(card_t)); pci_set_drvdata(pdev, card); card->ports[0].dev = alloc_hdlcdev(&card->ports[0]); card->ports[1].dev = alloc_hdlcdev(&card->ports[1]); diff --git a/drivers/net/wan/sdla.c b/drivers/net/wan/sdla.c index 6a485f0556f..792e588d7d6 100644 --- a/drivers/net/wan/sdla.c +++ b/drivers/net/wan/sdla.c @@ -1196,10 +1196,9 @@ static int sdla_xfer(struct net_device *dev, struct sdla_mem __user *info, int r if (read) { - temp = kmalloc(mem.len, GFP_KERNEL); + temp = kzalloc(mem.len, GFP_KERNEL); if (!temp) return(-ENOMEM); - memset(temp, 0, mem.len); sdla_read(dev, mem.addr, temp, mem.len); if(copy_to_user(mem.data, temp, mem.len)) { diff --git a/drivers/net/wan/sealevel.c b/drivers/net/wan/sealevel.c index 131358108c5..11276bf3149 100644 --- a/drivers/net/wan/sealevel.c +++ b/drivers/net/wan/sealevel.c @@ -270,11 +270,10 @@ static __init struct slvl_board *slvl_init(int iobase, int irq, return NULL; } - b = kmalloc(sizeof(struct slvl_board), GFP_KERNEL); + b = kzalloc(sizeof(struct slvl_board), GFP_KERNEL); if(!b) goto fail3; - memset(b, 0, sizeof(*b)); if (!(b->dev[0]= slvl_alloc(iobase, irq))) goto fail2; diff --git a/drivers/net/wan/wanxl.c b/drivers/net/wan/wanxl.c index c7360157433..3c78f985638 100644 --- a/drivers/net/wan/wanxl.c +++ b/drivers/net/wan/wanxl.c @@ -599,7 +599,7 @@ static int __devinit wanxl_pci_init_one(struct pci_dev *pdev, } alloc_size = sizeof(card_t) + ports * sizeof(port_t); - card = kmalloc(alloc_size, GFP_KERNEL); + card = kzalloc(alloc_size, GFP_KERNEL); if (card == NULL) { printk(KERN_ERR "wanXL %s: unable to allocate memory\n", pci_name(pdev)); @@ -607,7 +607,6 @@ static int __devinit wanxl_pci_init_one(struct pci_dev *pdev, pci_disable_device(pdev); return -ENOBUFS; } - memset(card, 0, alloc_size); pci_set_drvdata(pdev, card); card->pdev = pdev; diff --git a/drivers/net/wan/x25_asy.c b/drivers/net/wan/x25_asy.c index 1c9edd97acc..c48b1cc63fd 100644 --- a/drivers/net/wan/x25_asy.c +++ b/drivers/net/wan/x25_asy.c @@ -786,14 +786,12 @@ static int __init init_x25_asy(void) printk(KERN_INFO "X.25 async: version 0.00 ALPHA " "(dynamic channels, max=%d).\n", x25_asy_maxdev ); - x25_asy_devs = kmalloc(sizeof(struct net_device *)*x25_asy_maxdev, - GFP_KERNEL); + x25_asy_devs = kcalloc(x25_asy_maxdev, sizeof(struct net_device*), GFP_KERNEL); if (!x25_asy_devs) { printk(KERN_WARNING "X25 async: Can't allocate x25_asy_ctrls[] " "array! Uaargh! (-> No X.25 available)\n"); return -ENOMEM; } - memset(x25_asy_devs, 0, sizeof(struct net_device *)*x25_asy_maxdev); return tty_register_ldisc(N_X25, &x25_ldisc); } diff --git a/drivers/nubus/nubus.c b/drivers/nubus/nubus.c index 3a0a3a73493..e503c9c9803 100644 --- a/drivers/nubus/nubus.c +++ b/drivers/nubus/nubus.c @@ -466,9 +466,8 @@ static struct nubus_dev* __init parent->base, dir.base); /* Actually we should probably panic if this fails */ - if ((dev = kmalloc(sizeof(*dev), GFP_ATOMIC)) == NULL) + if ((dev = kzalloc(sizeof(*dev), GFP_ATOMIC)) == NULL) return NULL; - memset(dev, 0, sizeof(*dev)); dev->resid = parent->type; dev->directory = dir.base; dev->board = board; @@ -800,9 +799,8 @@ static struct nubus_board* __init nubus_add_board(int slot, int bytelanes) nubus_rewind(&rp, FORMAT_BLOCK_SIZE, bytelanes); /* Actually we should probably panic if this fails */ - if ((board = kmalloc(sizeof(*board), GFP_ATOMIC)) == NULL) + if ((board = kzalloc(sizeof(*board), GFP_ATOMIC)) == NULL) return NULL; - memset(board, 0, sizeof(*board)); board->fblock = rp; /* Dump the format block for debugging purposes */ diff --git a/drivers/parport/parport_cs.c b/drivers/parport/parport_cs.c index 8b7d84eca05..802a81d4736 100644 --- a/drivers/parport/parport_cs.c +++ b/drivers/parport/parport_cs.c @@ -105,9 +105,8 @@ static int parport_probe(struct pcmcia_device *link) DEBUG(0, "parport_attach()\n"); /* Create new parport device */ - info = kmalloc(sizeof(*info), GFP_KERNEL); + info = kzalloc(sizeof(*info), GFP_KERNEL); if (!info) return -ENOMEM; - memset(info, 0, sizeof(*info)); link->priv = info; info->p_dev = link; diff --git a/drivers/parport/parport_serial.c b/drivers/parport/parport_serial.c index 90ea3b8b99b..bd6ad8b3816 100644 --- a/drivers/parport/parport_serial.c +++ b/drivers/parport/parport_serial.c @@ -324,10 +324,9 @@ static int __devinit parport_serial_pci_probe (struct pci_dev *dev, struct parport_serial_private *priv; int err; - priv = kmalloc (sizeof *priv, GFP_KERNEL); + priv = kzalloc (sizeof *priv, GFP_KERNEL); if (!priv) return -ENOMEM; - memset(priv, 0, sizeof(struct parport_serial_private)); pci_set_drvdata (dev, priv); err = pci_enable_device (dev); diff --git a/drivers/pci/pcie/aer/aerdrv.c b/drivers/pci/pcie/aer/aerdrv.c index 6846fb42b39..ad90a01b0df 100644 --- a/drivers/pci/pcie/aer/aerdrv.c +++ b/drivers/pci/pcie/aer/aerdrv.c @@ -148,11 +148,10 @@ static struct aer_rpc* aer_alloc_rpc(struct pcie_device *dev) { struct aer_rpc *rpc; - if (!(rpc = kmalloc(sizeof(struct aer_rpc), + if (!(rpc = kzalloc(sizeof(struct aer_rpc), GFP_KERNEL))) return NULL; - memset(rpc, 0, sizeof(struct aer_rpc)); /* * Initialize Root lock access, e_lock, to Root Error Status Reg, * Root Error ID Reg, and Root error producer/consumer index. diff --git a/drivers/pnp/core.c b/drivers/pnp/core.c index 3e20b1cc777..8e7b2dd3881 100644 --- a/drivers/pnp/core.c +++ b/drivers/pnp/core.c @@ -35,12 +35,11 @@ void *pnp_alloc(long size) { void *result; - result = kmalloc(size, GFP_KERNEL); + result = kzalloc(size, GFP_KERNEL); if (!result){ printk(KERN_ERR "pnp: Out of Memory\n"); return NULL; } - memset(result, 0, size); return result; } diff --git a/drivers/rapidio/rio-scan.c b/drivers/rapidio/rio-scan.c index f935c1f71a5..44420723a35 100644 --- a/drivers/rapidio/rio-scan.c +++ b/drivers/rapidio/rio-scan.c @@ -297,11 +297,10 @@ static struct rio_dev *rio_setup_device(struct rio_net *net, struct rio_switch *rswitch; int result, rdid; - rdev = kmalloc(sizeof(struct rio_dev), GFP_KERNEL); + rdev = kzalloc(sizeof(struct rio_dev), GFP_KERNEL); if (!rdev) goto out; - memset(rdev, 0, sizeof(struct rio_dev)); rdev->net = net; rio_mport_read_config_32(port, destid, hopcount, RIO_DEV_ID_CAR, &result); @@ -801,9 +800,8 @@ static struct rio_net __devinit *rio_alloc_net(struct rio_mport *port) { struct rio_net *net; - net = kmalloc(sizeof(struct rio_net), GFP_KERNEL); + net = kzalloc(sizeof(struct rio_net), GFP_KERNEL); if (net) { - memset(net, 0, sizeof(struct rio_net)); INIT_LIST_HEAD(&net->node); INIT_LIST_HEAD(&net->devices); INIT_LIST_HEAD(&net->mports); diff --git a/drivers/rtc/rtc-cmos.c b/drivers/rtc/rtc-cmos.c index e24ea82dc35..5d760bb6c2c 100644 --- a/drivers/rtc/rtc-cmos.c +++ b/drivers/rtc/rtc-cmos.c @@ -235,7 +235,7 @@ static int cmos_set_alarm(struct device *dev, struct rtc_wkalrm *t) return 0; } -static int cmos_set_freq(struct device *dev, int freq) +static int cmos_irq_set_freq(struct device *dev, int freq) { struct cmos_rtc *cmos = dev_get_drvdata(dev); int f; @@ -259,6 +259,34 @@ static int cmos_set_freq(struct device *dev, int freq) return 0; } +static int cmos_irq_set_state(struct device *dev, int enabled) +{ + struct cmos_rtc *cmos = dev_get_drvdata(dev); + unsigned char rtc_control, rtc_intr; + unsigned long flags; + + if (!is_valid_irq(cmos->irq)) + return -ENXIO; + + spin_lock_irqsave(&rtc_lock, flags); + rtc_control = CMOS_READ(RTC_CONTROL); + + if (enabled) + rtc_control |= RTC_PIE; + else + rtc_control &= ~RTC_PIE; + + CMOS_WRITE(rtc_control, RTC_CONTROL); + + rtc_intr = CMOS_READ(RTC_INTR_FLAGS); + rtc_intr &= (rtc_control & RTC_IRQMASK) | RTC_IRQF; + if (is_intr(rtc_intr)) + rtc_update_irq(cmos->rtc, 1, rtc_intr); + + spin_unlock_irqrestore(&rtc_lock, flags); + return 0; +} + #if defined(CONFIG_RTC_INTF_DEV) || defined(CONFIG_RTC_INTF_DEV_MODULE) static int @@ -360,7 +388,8 @@ static const struct rtc_class_ops cmos_rtc_ops = { .read_alarm = cmos_read_alarm, .set_alarm = cmos_set_alarm, .proc = cmos_procfs, - .irq_set_freq = cmos_set_freq, + .irq_set_freq = cmos_irq_set_freq, + .irq_set_state = cmos_irq_set_state, }; /*----------------------------------------------------------------*/ diff --git a/drivers/s390/char/tape_34xx.c b/drivers/s390/char/tape_34xx.c index e765875e8db..80e7a537e7d 100644 --- a/drivers/s390/char/tape_34xx.c +++ b/drivers/s390/char/tape_34xx.c @@ -131,10 +131,9 @@ tape_34xx_schedule_work(struct tape_device *device, enum tape_op op) { struct tape_34xx_work *p; - if ((p = kmalloc(sizeof(*p), GFP_ATOMIC)) == NULL) + if ((p = kzalloc(sizeof(*p), GFP_ATOMIC)) == NULL) return -ENOMEM; - memset(p, 0, sizeof(*p)); INIT_WORK(&p->work, tape_34xx_work_handler); p->device = tape_get_device_reference(device); diff --git a/drivers/s390/net/claw.c b/drivers/s390/net/claw.c index 348bb7b8277..023455a0b34 100644 --- a/drivers/s390/net/claw.c +++ b/drivers/s390/net/claw.c @@ -317,8 +317,8 @@ claw_probe(struct ccwgroup_device *cgdev) CLAW_DBF_TEXT_(2,setup,"probex%d",-ENOMEM); return -ENOMEM; } - privptr->p_mtc_envelope= kmalloc( MAX_ENVELOPE_SIZE, GFP_KERNEL); - privptr->p_env = kmalloc(sizeof(struct claw_env), GFP_KERNEL); + privptr->p_mtc_envelope= kzalloc( MAX_ENVELOPE_SIZE, GFP_KERNEL); + privptr->p_env = kzalloc(sizeof(struct claw_env), GFP_KERNEL); if ((privptr->p_mtc_envelope==NULL) || (privptr->p_env==NULL)) { probe_error(cgdev); put_device(&cgdev->dev); @@ -327,8 +327,6 @@ claw_probe(struct ccwgroup_device *cgdev) CLAW_DBF_TEXT_(2,setup,"probex%d",-ENOMEM); return -ENOMEM; } - memset(privptr->p_mtc_envelope, 0x00, MAX_ENVELOPE_SIZE); - memset(privptr->p_env, 0x00, sizeof(struct claw_env)); memcpy(privptr->p_env->adapter_name,WS_NAME_NOT_DEF,8); memcpy(privptr->p_env->host_name,WS_NAME_NOT_DEF,8); memcpy(privptr->p_env->api_type,WS_NAME_NOT_DEF,8); @@ -3924,7 +3922,7 @@ add_channel(struct ccw_device *cdev,int i,struct claw_privbk *privptr) snprintf(p_ch->id, CLAW_ID_SIZE, "cl-%s", cdev->dev.bus_id); ccw_device_get_id(cdev, &dev_id); p_ch->devno = dev_id.devno; - if ((p_ch->irb = kmalloc(sizeof (struct irb),GFP_KERNEL)) == NULL) { + if ((p_ch->irb = kzalloc(sizeof (struct irb),GFP_KERNEL)) == NULL) { printk(KERN_WARNING "%s Out of memory in %s for irb\n", p_ch->id,__FUNCTION__); #ifdef FUNCTRACE @@ -3933,7 +3931,6 @@ add_channel(struct ccw_device *cdev,int i,struct claw_privbk *privptr) #endif return -ENOMEM; } - memset(p_ch->irb, 0, sizeof (struct irb)); #ifdef FUNCTRACE printk(KERN_INFO "%s:%s Exit on line %d\n", cdev->dev.bus_id,__FUNCTION__,__LINE__); diff --git a/drivers/sbus/char/bbc_i2c.c b/drivers/sbus/char/bbc_i2c.c index 178155bf9db..fbadd4d761f 100644 --- a/drivers/sbus/char/bbc_i2c.c +++ b/drivers/sbus/char/bbc_i2c.c @@ -156,10 +156,9 @@ struct bbc_i2c_client *bbc_i2c_attach(struct linux_ebus_child *echild) if (!bp) return NULL; - client = kmalloc(sizeof(*client), GFP_KERNEL); + client = kzalloc(sizeof(*client), GFP_KERNEL); if (!client) return NULL; - memset(client, 0, sizeof(*client)); client->bp = bp; client->echild = echild; client->bus = echild->resource[0].start; diff --git a/drivers/sbus/char/vfc_dev.c b/drivers/sbus/char/vfc_dev.c index 6afc7e5df0d..26b1d2a17ed 100644 --- a/drivers/sbus/char/vfc_dev.c +++ b/drivers/sbus/char/vfc_dev.c @@ -656,12 +656,9 @@ static int vfc_probe(void) if (!cards) return -ENODEV; - vfc_dev_lst = kmalloc(sizeof(struct vfc_dev *) * - (cards+1), - GFP_KERNEL); + vfc_dev_lst = kcalloc(cards + 1, sizeof(struct vfc_dev*), GFP_KERNEL); if (vfc_dev_lst == NULL) return -ENOMEM; - memset(vfc_dev_lst, 0, sizeof(struct vfc_dev *) * (cards + 1)); vfc_dev_lst[cards] = NULL; ret = register_chrdev(VFC_MAJOR, vfcstr, &vfc_fops); diff --git a/drivers/scsi/3w-9xxx.c b/drivers/scsi/3w-9xxx.c index 76c09097175..6b49f6a2524 100644 --- a/drivers/scsi/3w-9xxx.c +++ b/drivers/scsi/3w-9xxx.c @@ -1160,13 +1160,12 @@ static int twa_initialize_device_extension(TW_Device_Extension *tw_dev) } /* Allocate event info space */ - tw_dev->event_queue[0] = kmalloc(sizeof(TW_Event) * TW_Q_LENGTH, GFP_KERNEL); + tw_dev->event_queue[0] = kcalloc(TW_Q_LENGTH, sizeof(TW_Event), GFP_KERNEL); if (!tw_dev->event_queue[0]) { TW_PRINTK(tw_dev->host, TW_DRIVER, 0x18, "Event info memory allocation failed"); goto out; } - memset(tw_dev->event_queue[0], 0, sizeof(TW_Event) * TW_Q_LENGTH); for (i = 0; i < TW_Q_LENGTH; i++) { tw_dev->event_queue[i] = (TW_Event *)((unsigned char *)tw_dev->event_queue[0] + (i * sizeof(TW_Event))); diff --git a/drivers/scsi/NCR53C9x.c b/drivers/scsi/NCR53C9x.c index 8b5334c56f0..773d11dd995 100644 --- a/drivers/scsi/NCR53C9x.c +++ b/drivers/scsi/NCR53C9x.c @@ -3606,11 +3606,10 @@ out: int esp_slave_alloc(struct scsi_device *SDptr) { struct esp_device *esp_dev = - kmalloc(sizeof(struct esp_device), GFP_ATOMIC); + kzalloc(sizeof(struct esp_device), GFP_ATOMIC); if (!esp_dev) return -ENOMEM; - memset(esp_dev, 0, sizeof(struct esp_device)); SDptr->hostdata = esp_dev; return 0; } diff --git a/drivers/scsi/NCR_D700.c b/drivers/scsi/NCR_D700.c index f12864abed2..3a8089705fe 100644 --- a/drivers/scsi/NCR_D700.c +++ b/drivers/scsi/NCR_D700.c @@ -181,13 +181,12 @@ NCR_D700_probe_one(struct NCR_D700_private *p, int siop, int irq, struct Scsi_Host *host; int ret; - hostdata = kmalloc(sizeof(*hostdata), GFP_KERNEL); + hostdata = kzalloc(sizeof(*hostdata), GFP_KERNEL); if (!hostdata) { printk(KERN_ERR "NCR D700: SIOP%d: Failed to allocate host" "data, detatching\n", siop); return -ENOMEM; } - memset(hostdata, 0, sizeof(*hostdata)); if (!request_region(region, 64, "NCR_D700")) { printk(KERN_ERR "NCR D700: Failed to reserve IO region 0x%x\n", diff --git a/drivers/scsi/NCR_Q720.c b/drivers/scsi/NCR_Q720.c index 778844c3544..a8bbdc2273b 100644 --- a/drivers/scsi/NCR_Q720.c +++ b/drivers/scsi/NCR_Q720.c @@ -148,11 +148,10 @@ NCR_Q720_probe(struct device *dev) __u32 base_addr, mem_size; void __iomem *mem_base; - p = kmalloc(sizeof(*p), GFP_KERNEL); + p = kzalloc(sizeof(*p), GFP_KERNEL); if (!p) return -ENOMEM; - memset(p, 0, sizeof(*p)); pos2 = mca_device_read_pos(mca_dev, 2); /* enable device */ pos2 |= NCR_Q720_POS2_BOARD_ENABLE | NCR_Q720_POS2_INTERRUPT_ENABLE; diff --git a/drivers/scsi/imm.c b/drivers/scsi/imm.c index 0464c182c57..005d2b05f32 100644 --- a/drivers/scsi/imm.c +++ b/drivers/scsi/imm.c @@ -1159,11 +1159,10 @@ static int __imm_attach(struct parport *pb) init_waitqueue_head(&waiting); - dev = kmalloc(sizeof(imm_struct), GFP_KERNEL); + dev = kzalloc(sizeof(imm_struct), GFP_KERNEL); if (!dev) return -ENOMEM; - memset(dev, 0, sizeof(imm_struct)); dev->base = -1; dev->mode = IMM_AUTODETECT; diff --git a/drivers/scsi/ips.c b/drivers/scsi/ips.c index 9f8ed6b8157..492a51bd6aa 100644 --- a/drivers/scsi/ips.c +++ b/drivers/scsi/ips.c @@ -7068,14 +7068,13 @@ ips_init_phase1(struct pci_dev *pci_dev, int *indexPtr) subdevice_id = pci_dev->subsystem_device; /* found a controller */ - ha = kmalloc(sizeof (ips_ha_t), GFP_KERNEL); + ha = kzalloc(sizeof (ips_ha_t), GFP_KERNEL); if (ha == NULL) { IPS_PRINTK(KERN_WARNING, pci_dev, "Unable to allocate temporary ha struct\n"); return -1; } - memset(ha, 0, sizeof (ips_ha_t)); ips_sh[index] = NULL; ips_ha[index] = ha; diff --git a/drivers/scsi/lasi700.c b/drivers/scsi/lasi700.c index 5c32a69e41b..3126824da36 100644 --- a/drivers/scsi/lasi700.c +++ b/drivers/scsi/lasi700.c @@ -101,13 +101,12 @@ lasi700_probe(struct parisc_device *dev) struct NCR_700_Host_Parameters *hostdata; struct Scsi_Host *host; - hostdata = kmalloc(sizeof(*hostdata), GFP_KERNEL); + hostdata = kzalloc(sizeof(*hostdata), GFP_KERNEL); if (!hostdata) { printk(KERN_ERR "%s: Failed to allocate host data\n", dev->dev.bus_id); return -ENOMEM; } - memset(hostdata, 0, sizeof(struct NCR_700_Host_Parameters)); hostdata->dev = &dev->dev; dma_set_mask(&dev->dev, DMA_32BIT_MASK); diff --git a/drivers/scsi/lpfc/lpfc_init.c b/drivers/scsi/lpfc/lpfc_init.c index f81f85ee190..07bd0dcdf0d 100644 --- a/drivers/scsi/lpfc/lpfc_init.c +++ b/drivers/scsi/lpfc/lpfc_init.c @@ -1830,7 +1830,7 @@ lpfc_pci_probe_one(struct pci_dev *pdev, const struct pci_device_id *pid) /* Initialize and populate the iocb list per host. */ INIT_LIST_HEAD(&phba->lpfc_iocb_list); for (i = 0; i < LPFC_IOCB_LIST_CNT; i++) { - iocbq_entry = kmalloc(sizeof(struct lpfc_iocbq), GFP_KERNEL); + iocbq_entry = kzalloc(sizeof(struct lpfc_iocbq), GFP_KERNEL); if (iocbq_entry == NULL) { printk(KERN_ERR "%s: only allocated %d iocbs of " "expected %d count. Unloading driver.\n", @@ -1839,7 +1839,6 @@ lpfc_pci_probe_one(struct pci_dev *pdev, const struct pci_device_id *pid) goto out_free_iocbq; } - memset(iocbq_entry, 0, sizeof(struct lpfc_iocbq)); iotag = lpfc_sli_next_iotag(phba, iocbq_entry); if (iotag == 0) { kfree (iocbq_entry); diff --git a/drivers/scsi/megaraid/megaraid_mbox.c b/drivers/scsi/megaraid/megaraid_mbox.c index c46685a03a9..c6a53dccc16 100644 --- a/drivers/scsi/megaraid/megaraid_mbox.c +++ b/drivers/scsi/megaraid/megaraid_mbox.c @@ -454,7 +454,7 @@ megaraid_probe_one(struct pci_dev *pdev, const struct pci_device_id *id) pci_set_master(pdev); // Allocate the per driver initialization structure - adapter = kmalloc(sizeof(adapter_t), GFP_KERNEL); + adapter = kzalloc(sizeof(adapter_t), GFP_KERNEL); if (adapter == NULL) { con_log(CL_ANN, (KERN_WARNING @@ -462,7 +462,6 @@ megaraid_probe_one(struct pci_dev *pdev, const struct pci_device_id *id) goto out_probe_one; } - memset(adapter, 0, sizeof(adapter_t)); // set up PCI related soft state and other pre-known parameters @@ -746,10 +745,9 @@ megaraid_init_mbox(adapter_t *adapter) * Allocate and initialize the init data structure for mailbox * controllers */ - raid_dev = kmalloc(sizeof(mraid_device_t), GFP_KERNEL); + raid_dev = kzalloc(sizeof(mraid_device_t), GFP_KERNEL); if (raid_dev == NULL) return -1; - memset(raid_dev, 0, sizeof(mraid_device_t)); /* * Attach the adapter soft state to raid device soft state @@ -1050,8 +1048,7 @@ megaraid_alloc_cmd_packets(adapter_t *adapter) * since the calling routine does not yet know the number of available * commands. */ - adapter->kscb_list = kmalloc(sizeof(scb_t) * MBOX_MAX_SCSI_CMDS, - GFP_KERNEL); + adapter->kscb_list = kcalloc(MBOX_MAX_SCSI_CMDS, sizeof(scb_t), GFP_KERNEL); if (adapter->kscb_list == NULL) { con_log(CL_ANN, (KERN_WARNING @@ -1059,7 +1056,6 @@ megaraid_alloc_cmd_packets(adapter_t *adapter) __LINE__)); goto out_free_ibuf; } - memset(adapter->kscb_list, 0, sizeof(scb_t) * MBOX_MAX_SCSI_CMDS); // memory allocation for our command packets if (megaraid_mbox_setup_dma_pools(adapter) != 0) { @@ -3495,8 +3491,7 @@ megaraid_cmm_register(adapter_t *adapter) int i; // Allocate memory for the base list of scb for management module. - adapter->uscb_list = kmalloc(sizeof(scb_t) * MBOX_MAX_USER_CMDS, - GFP_KERNEL); + adapter->uscb_list = kcalloc(MBOX_MAX_USER_CMDS, sizeof(scb_t), GFP_KERNEL); if (adapter->uscb_list == NULL) { con_log(CL_ANN, (KERN_WARNING @@ -3504,7 +3499,6 @@ megaraid_cmm_register(adapter_t *adapter) __LINE__)); return -1; } - memset(adapter->uscb_list, 0, sizeof(scb_t) * MBOX_MAX_USER_CMDS); // Initialize the synchronization parameters for resources for diff --git a/drivers/scsi/megaraid/megaraid_mm.c b/drivers/scsi/megaraid/megaraid_mm.c index 84d9c27133d..b6587a6d848 100644 --- a/drivers/scsi/megaraid/megaraid_mm.c +++ b/drivers/scsi/megaraid/megaraid_mm.c @@ -890,12 +890,11 @@ mraid_mm_register_adp(mraid_mmadp_t *lld_adp) if (lld_adp->drvr_type != DRVRTYPE_MBOX) return (-EINVAL); - adapter = kmalloc(sizeof(mraid_mmadp_t), GFP_KERNEL); + adapter = kzalloc(sizeof(mraid_mmadp_t), GFP_KERNEL); if (!adapter) return -ENOMEM; - memset(adapter, 0, sizeof(mraid_mmadp_t)); adapter->unique_id = lld_adp->unique_id; adapter->drvr_type = lld_adp->drvr_type; diff --git a/drivers/scsi/megaraid/megaraid_sas.c b/drivers/scsi/megaraid/megaraid_sas.c index b7f2e613c90..ebb948c016b 100644 --- a/drivers/scsi/megaraid/megaraid_sas.c +++ b/drivers/scsi/megaraid/megaraid_sas.c @@ -1636,15 +1636,13 @@ static int megasas_alloc_cmds(struct megasas_instance *instance) * Allocate the dynamic array first and then allocate individual * commands. */ - instance->cmd_list = kmalloc(sizeof(struct megasas_cmd *) * max_cmd, - GFP_KERNEL); + instance->cmd_list = kcalloc(max_cmd, sizeof(struct megasas_cmd*), GFP_KERNEL); if (!instance->cmd_list) { printk(KERN_DEBUG "megasas: out of memory\n"); return -ENOMEM; } - memset(instance->cmd_list, 0, sizeof(struct megasas_cmd *) * max_cmd); for (i = 0; i < max_cmd; i++) { instance->cmd_list[i] = kmalloc(sizeof(struct megasas_cmd), diff --git a/drivers/scsi/pcmcia/aha152x_stub.c b/drivers/scsi/pcmcia/aha152x_stub.c index 370802d24ac..2dd0dc9a9ae 100644 --- a/drivers/scsi/pcmcia/aha152x_stub.c +++ b/drivers/scsi/pcmcia/aha152x_stub.c @@ -106,9 +106,8 @@ static int aha152x_probe(struct pcmcia_device *link) DEBUG(0, "aha152x_attach()\n"); /* Create new SCSI device */ - info = kmalloc(sizeof(*info), GFP_KERNEL); + info = kzalloc(sizeof(*info), GFP_KERNEL); if (!info) return -ENOMEM; - memset(info, 0, sizeof(*info)); info->p_dev = link; link->priv = info; diff --git a/drivers/scsi/pcmcia/nsp_cs.c b/drivers/scsi/pcmcia/nsp_cs.c index c6f8c6e65e0..445cfbbca9b 100644 --- a/drivers/scsi/pcmcia/nsp_cs.c +++ b/drivers/scsi/pcmcia/nsp_cs.c @@ -1602,9 +1602,8 @@ static int nsp_cs_probe(struct pcmcia_device *link) nsp_dbg(NSP_DEBUG_INIT, "in"); /* Create new SCSI device */ - info = kmalloc(sizeof(*info), GFP_KERNEL); + info = kzalloc(sizeof(*info), GFP_KERNEL); if (info == NULL) { return -ENOMEM; } - memset(info, 0, sizeof(*info)); info->p_dev = link; link->priv = info; data->ScsiInfo = info; diff --git a/drivers/scsi/pcmcia/qlogic_stub.c b/drivers/scsi/pcmcia/qlogic_stub.c index 697cfb76c3a..67c5a58d17d 100644 --- a/drivers/scsi/pcmcia/qlogic_stub.c +++ b/drivers/scsi/pcmcia/qlogic_stub.c @@ -162,10 +162,9 @@ static int qlogic_probe(struct pcmcia_device *link) DEBUG(0, "qlogic_attach()\n"); /* Create new SCSI device */ - info = kmalloc(sizeof(*info), GFP_KERNEL); + info = kzalloc(sizeof(*info), GFP_KERNEL); if (!info) return -ENOMEM; - memset(info, 0, sizeof(*info)); info->p_dev = link; link->priv = info; link->io.NumPorts1 = 16; diff --git a/drivers/scsi/pcmcia/sym53c500_cs.c b/drivers/scsi/pcmcia/sym53c500_cs.c index 2695b7187b2..961839ecfe8 100644 --- a/drivers/scsi/pcmcia/sym53c500_cs.c +++ b/drivers/scsi/pcmcia/sym53c500_cs.c @@ -875,10 +875,9 @@ SYM53C500_probe(struct pcmcia_device *link) DEBUG(0, "SYM53C500_attach()\n"); /* Create new SCSI device */ - info = kmalloc(sizeof(*info), GFP_KERNEL); + info = kzalloc(sizeof(*info), GFP_KERNEL); if (!info) return -ENOMEM; - memset(info, 0, sizeof(*info)); info->p_dev = link; link->priv = info; link->io.NumPorts1 = 16; diff --git a/drivers/scsi/ppa.c b/drivers/scsi/ppa.c index 2f1fa1eb7e9..67b6d76a6c8 100644 --- a/drivers/scsi/ppa.c +++ b/drivers/scsi/ppa.c @@ -1014,10 +1014,9 @@ static int __ppa_attach(struct parport *pb) int modes, ppb, ppb_hi; int err = -ENOMEM; - dev = kmalloc(sizeof(ppa_struct), GFP_KERNEL); + dev = kzalloc(sizeof(ppa_struct), GFP_KERNEL); if (!dev) return -ENOMEM; - memset(dev, 0, sizeof(ppa_struct)); dev->base = -1; dev->mode = PPA_AUTODETECT; dev->recon_tmo = PPA_RECON_TMO; diff --git a/drivers/scsi/sim710.c b/drivers/scsi/sim710.c index 018c65f73ac..710f19de3d4 100644 --- a/drivers/scsi/sim710.c +++ b/drivers/scsi/sim710.c @@ -100,7 +100,7 @@ sim710_probe_common(struct device *dev, unsigned long base_addr, { struct Scsi_Host * host = NULL; struct NCR_700_Host_Parameters *hostdata = - kmalloc(sizeof(struct NCR_700_Host_Parameters), GFP_KERNEL); + kzalloc(sizeof(struct NCR_700_Host_Parameters), GFP_KERNEL); printk(KERN_NOTICE "sim710: %s\n", dev->bus_id); printk(KERN_NOTICE "sim710: irq = %d, clock = %d, base = 0x%lx, scsi_id = %d\n", @@ -110,7 +110,6 @@ sim710_probe_common(struct device *dev, unsigned long base_addr, printk(KERN_ERR "sim710: Failed to allocate host data\n"); goto out; } - memset(hostdata, 0, sizeof(struct NCR_700_Host_Parameters)); if(request_region(base_addr, 64, "sim710") == NULL) { printk(KERN_ERR "sim710: Failed to reserve IO region 0x%lx\n", diff --git a/drivers/scsi/tmscsim.c b/drivers/scsi/tmscsim.c index 14cba1ca38b..5db1520f8ba 100644 --- a/drivers/scsi/tmscsim.c +++ b/drivers/scsi/tmscsim.c @@ -2082,10 +2082,9 @@ static int dc390_slave_alloc(struct scsi_device *scsi_device) uint id = scsi_device->id; uint lun = scsi_device->lun; - pDCB = kmalloc(sizeof(struct dc390_dcb), GFP_KERNEL); + pDCB = kzalloc(sizeof(struct dc390_dcb), GFP_KERNEL); if (!pDCB) return -ENOMEM; - memset(pDCB, 0, sizeof(struct dc390_dcb)); if (!pACB->DCBCnt++) { pACB->pLinkDCB = pDCB; diff --git a/drivers/serial/amba-pl011.c b/drivers/serial/amba-pl011.c index 954073c6ce3..72229df9dc1 100644 --- a/drivers/serial/amba-pl011.c +++ b/drivers/serial/amba-pl011.c @@ -716,7 +716,7 @@ static int pl011_probe(struct amba_device *dev, void *id) goto out; } - uap = kmalloc(sizeof(struct uart_amba_port), GFP_KERNEL); + uap = kzalloc(sizeof(struct uart_amba_port), GFP_KERNEL); if (uap == NULL) { ret = -ENOMEM; goto out; @@ -728,7 +728,6 @@ static int pl011_probe(struct amba_device *dev, void *id) goto free; } - memset(uap, 0, sizeof(struct uart_amba_port)); uap->clk = clk_get(&dev->dev, "UARTCLK"); if (IS_ERR(uap->clk)) { ret = PTR_ERR(uap->clk); diff --git a/drivers/sh/superhyway/superhyway.c b/drivers/sh/superhyway/superhyway.c index 94b22903119..7d873b3b051 100644 --- a/drivers/sh/superhyway/superhyway.c +++ b/drivers/sh/superhyway/superhyway.c @@ -56,11 +56,10 @@ int superhyway_add_device(unsigned long base, struct superhyway_device *sdev, struct superhyway_device *dev = sdev; if (!dev) { - dev = kmalloc(sizeof(struct superhyway_device), GFP_KERNEL); + dev = kzalloc(sizeof(struct superhyway_device), GFP_KERNEL); if (!dev) return -ENOMEM; - memset(dev, 0, sizeof(struct superhyway_device)); } dev->bus = bus; diff --git a/drivers/sn/ioc3.c b/drivers/sn/ioc3.c index 2dd6eed50aa..29fcd6d0301 100644 --- a/drivers/sn/ioc3.c +++ b/drivers/sn/ioc3.c @@ -629,7 +629,7 @@ static int ioc3_probe(struct pci_dev *pdev, const struct pci_device_id *pci_id) #endif /* Set up per-IOC3 data */ - idd = kmalloc(sizeof(struct ioc3_driver_data), GFP_KERNEL); + idd = kzalloc(sizeof(struct ioc3_driver_data), GFP_KERNEL); if (!idd) { printk(KERN_WARNING "%s: Failed to allocate IOC3 data for pci_dev %s.\n", @@ -637,7 +637,6 @@ static int ioc3_probe(struct pci_dev *pdev, const struct pci_device_id *pci_id) ret = -ENODEV; goto out_idd; } - memset(idd, 0, sizeof(struct ioc3_driver_data)); spin_lock_init(&idd->ir_lock); spin_lock_init(&idd->gpio_lock); idd->pdev = pdev; diff --git a/drivers/telephony/ixj_pcmcia.c b/drivers/telephony/ixj_pcmcia.c index 3e658dc7c2d..ff9a29b7633 100644 --- a/drivers/telephony/ixj_pcmcia.c +++ b/drivers/telephony/ixj_pcmcia.c @@ -45,11 +45,10 @@ static int ixj_probe(struct pcmcia_device *p_dev) p_dev->io.Attributes2 = IO_DATA_PATH_WIDTH_8; p_dev->io.IOAddrLines = 3; p_dev->conf.IntType = INT_MEMORY_AND_IO; - p_dev->priv = kmalloc(sizeof(struct ixj_info_t), GFP_KERNEL); + p_dev->priv = kzalloc(sizeof(struct ixj_info_t), GFP_KERNEL); if (!p_dev->priv) { return -ENOMEM; } - memset(p_dev->priv, 0, sizeof(struct ixj_info_t)); return ixj_config(p_dev); } diff --git a/drivers/usb/gadget/goku_udc.c b/drivers/usb/gadget/goku_udc.c index d6c5f1150ae..349b8166f34 100644 --- a/drivers/usb/gadget/goku_udc.c +++ b/drivers/usb/gadget/goku_udc.c @@ -1777,14 +1777,13 @@ static int goku_probe(struct pci_dev *pdev, const struct pci_device_id *id) } /* alloc, and start init */ - dev = kmalloc (sizeof *dev, GFP_KERNEL); + dev = kzalloc (sizeof *dev, GFP_KERNEL); if (dev == NULL){ pr_debug("enomem %s\n", pci_name(pdev)); retval = -ENOMEM; goto done; } - memset(dev, 0, sizeof *dev); spin_lock_init(&dev->lock); dev->pdev = pdev; dev->gadget.ops = &goku_ops; diff --git a/drivers/usb/gadget/serial.c b/drivers/usb/gadget/serial.c index dd33ff0ae4c..38138bb9ddb 100644 --- a/drivers/usb/gadget/serial.c +++ b/drivers/usb/gadget/serial.c @@ -1427,7 +1427,7 @@ static int __init gs_bind(struct usb_gadget *gadget) gs_acm_config_desc.bmAttributes |= USB_CONFIG_ATT_WAKEUP; } - gs_device = dev = kmalloc(sizeof(struct gs_dev), GFP_KERNEL); + gs_device = dev = kzalloc(sizeof(struct gs_dev), GFP_KERNEL); if (dev == NULL) return -ENOMEM; @@ -1435,7 +1435,6 @@ static int __init gs_bind(struct usb_gadget *gadget) init_utsname()->sysname, init_utsname()->release, gadget->name); - memset(dev, 0, sizeof(struct gs_dev)); dev->dev_gadget = gadget; spin_lock_init(&dev->dev_lock); INIT_LIST_HEAD(&dev->dev_req_list); diff --git a/drivers/usb/host/ohci-hcd.c b/drivers/usb/host/ohci-hcd.c index 2038125b7f8..6edf4097d2d 100644 --- a/drivers/usb/host/ohci-hcd.c +++ b/drivers/usb/host/ohci-hcd.c @@ -171,11 +171,10 @@ static int ohci_urb_enqueue ( } /* allocate the private part of the URB */ - urb_priv = kmalloc (sizeof (urb_priv_t) + size * sizeof (struct td *), + urb_priv = kzalloc (sizeof (urb_priv_t) + size * sizeof (struct td *), mem_flags); if (!urb_priv) return -ENOMEM; - memset (urb_priv, 0, sizeof (urb_priv_t) + size * sizeof (struct td *)); INIT_LIST_HEAD (&urb_priv->pending); urb_priv->length = size; urb_priv->ed = ed; diff --git a/drivers/usb/host/sl811_cs.c b/drivers/usb/host/sl811_cs.c index 2d0e73b2009..5da63f53500 100644 --- a/drivers/usb/host/sl811_cs.c +++ b/drivers/usb/host/sl811_cs.c @@ -278,10 +278,9 @@ static int sl811_cs_probe(struct pcmcia_device *link) { local_info_t *local; - local = kmalloc(sizeof(local_info_t), GFP_KERNEL); + local = kzalloc(sizeof(local_info_t), GFP_KERNEL); if (!local) return -ENOMEM; - memset(local, 0, sizeof(local_info_t)); local->p_dev = link; link->priv = local; diff --git a/drivers/video/amba-clcd.c b/drivers/video/amba-clcd.c index 6c9dc2e69c8..a7a1c891bfa 100644 --- a/drivers/video/amba-clcd.c +++ b/drivers/video/amba-clcd.c @@ -447,13 +447,12 @@ static int clcdfb_probe(struct amba_device *dev, void *id) goto out; } - fb = kmalloc(sizeof(struct clcd_fb), GFP_KERNEL); + fb = kzalloc(sizeof(struct clcd_fb), GFP_KERNEL); if (!fb) { printk(KERN_INFO "CLCD: could not allocate new clcd_fb struct\n"); ret = -ENOMEM; goto free_region; } - memset(fb, 0, sizeof(struct clcd_fb)); fb->dev = dev; fb->board = board; diff --git a/drivers/video/aty/atyfb_base.c b/drivers/video/aty/atyfb_base.c index ef330e34d03..0c7bf75732e 100644 --- a/drivers/video/aty/atyfb_base.c +++ b/drivers/video/aty/atyfb_base.c @@ -2937,12 +2937,11 @@ static int __devinit atyfb_setup_sparc(struct pci_dev *pdev, /* nothing */ ; j = i + 4; - par->mmap_map = kmalloc(j * sizeof(*par->mmap_map), GFP_ATOMIC); + par->mmap_map = kcalloc(j, sizeof(*par->mmap_map), GFP_ATOMIC); if (!par->mmap_map) { PRINTKE("atyfb_setup_sparc() can't alloc mmap_map\n"); return -ENOMEM; } - memset(par->mmap_map, 0, j * sizeof(*par->mmap_map)); for (i = 0, j = 2; i < 6 && pdev->resource[i].start; i++) { struct resource *rp = &pdev->resource[i]; diff --git a/drivers/video/au1200fb.c b/drivers/video/au1200fb.c index dbf4ec3f6d5..03e57ef8837 100644 --- a/drivers/video/au1200fb.c +++ b/drivers/video/au1200fb.c @@ -1589,11 +1589,10 @@ static int au1200fb_init_fbinfo(struct au1200fb_device *fbdev) return -EFAULT; } - fbi->pseudo_palette = kmalloc(sizeof(u32) * 16, GFP_KERNEL); + fbi->pseudo_palette = kcalloc(16, sizeof(u32), GFP_KERNEL); if (!fbi->pseudo_palette) { return -ENOMEM; } - memset(fbi->pseudo_palette, 0, sizeof(u32) * 16); if (fb_alloc_cmap(&fbi->cmap, AU1200_LCD_NBR_PALETTE_ENTRIES, 0) < 0) { print_err("Fail to allocate colormap (%d entries)", diff --git a/drivers/video/clps711xfb.c b/drivers/video/clps711xfb.c index 50b78af0fa2..dea6579941b 100644 --- a/drivers/video/clps711xfb.c +++ b/drivers/video/clps711xfb.c @@ -366,11 +366,10 @@ int __init clps711xfb_init(void) if (fb_get_options("clps711xfb", NULL)) return -ENODEV; - cfb = kmalloc(sizeof(*cfb), GFP_KERNEL); + cfb = kzalloc(sizeof(*cfb), GFP_KERNEL); if (!cfb) goto out; - memset(cfb, 0, sizeof(*cfb)); strcpy(cfb->fix.id, "clps711x"); cfb->fbops = &clps7111fb_ops; diff --git a/drivers/video/cyber2000fb.c b/drivers/video/cyber2000fb.c index 7a6eeda5ae9..30ede6e8830 100644 --- a/drivers/video/cyber2000fb.c +++ b/drivers/video/cyber2000fb.c @@ -1221,11 +1221,10 @@ cyberpro_alloc_fb_info(unsigned int id, char *name) { struct cfb_info *cfb; - cfb = kmalloc(sizeof(struct cfb_info), GFP_KERNEL); + cfb = kzalloc(sizeof(struct cfb_info), GFP_KERNEL); if (!cfb) return NULL; - memset(cfb, 0, sizeof(struct cfb_info)); cfb->id = id; diff --git a/drivers/video/logo/logo.c b/drivers/video/logo/logo.c index 80c03618eb5..2b0f799aa8d 100644 --- a/drivers/video/logo/logo.c +++ b/drivers/video/logo/logo.c @@ -34,8 +34,11 @@ extern const struct linux_logo logo_superh_vga16; extern const struct linux_logo logo_superh_clut224; extern const struct linux_logo logo_m32r_clut224; - -const struct linux_logo *fb_find_logo(int depth) +/* logo's are marked __initdata. Use __init_refok to tell + * modpost that it is intended that this function uses data + * marked __initdata. + */ +const struct linux_logo * __init_refok fb_find_logo(int depth) { const struct linux_logo *logo = NULL; diff --git a/drivers/video/pvr2fb.c b/drivers/video/pvr2fb.c index 0f88c30f94f..f9300266044 100644 --- a/drivers/video/pvr2fb.c +++ b/drivers/video/pvr2fb.c @@ -1082,13 +1082,12 @@ static int __init pvr2fb_init(void) #endif size = sizeof(struct fb_info) + sizeof(struct pvr2fb_par) + 16 * sizeof(u32); - fb_info = kmalloc(size, GFP_KERNEL); + fb_info = kzalloc(size, GFP_KERNEL); if (!fb_info) { printk(KERN_ERR "Failed to allocate memory for fb_info\n"); return -ENOMEM; } - memset(fb_info, 0, size); currentpar = (struct pvr2fb_par *)(fb_info + 1); diff --git a/drivers/video/savage/savagefb_driver.c b/drivers/video/savage/savagefb_driver.c index 3d7507ad55f..b855f4a34af 100644 --- a/drivers/video/savage/savagefb_driver.c +++ b/drivers/video/savage/savagefb_driver.c @@ -2174,11 +2174,10 @@ static int __devinit savage_init_fb_info(struct fb_info *info, #if defined(CONFIG_FB_SAVAGE_ACCEL) /* FIFO size + padding for commands */ - info->pixmap.addr = kmalloc(8*1024, GFP_KERNEL); + info->pixmap.addr = kcalloc(8, 1024, GFP_KERNEL); err = -ENOMEM; if (info->pixmap.addr) { - memset(info->pixmap.addr, 0, 8*1024); info->pixmap.size = 8*1024; info->pixmap.scan_align = 4; info->pixmap.buf_align = 4; diff --git a/drivers/video/valkyriefb.c b/drivers/video/valkyriefb.c index ad66f070acb..7b0cef9ca8f 100644 --- a/drivers/video/valkyriefb.c +++ b/drivers/video/valkyriefb.c @@ -356,10 +356,9 @@ int __init valkyriefb_init(void) } #endif /* ppc (!CONFIG_MAC) */ - p = kmalloc(sizeof(*p), GFP_ATOMIC); + p = kzalloc(sizeof(*p), GFP_ATOMIC); if (p == 0) return -ENOMEM; - memset(p, 0, sizeof(*p)); /* Map in frame buffer and registers */ if (!request_mem_region(frame_buffer_phys, 0x100000, "valkyriefb")) { diff --git a/drivers/w1/masters/matrox_w1.c b/drivers/w1/masters/matrox_w1.c index 6f9d880ab2e..d356da5709f 100644 --- a/drivers/w1/masters/matrox_w1.c +++ b/drivers/w1/masters/matrox_w1.c @@ -164,7 +164,7 @@ static int __devinit matrox_w1_probe(struct pci_dev *pdev, const struct pci_devi if (pdev->vendor != PCI_VENDOR_ID_MATROX || pdev->device != PCI_DEVICE_ID_MATROX_G400) return -ENODEV; - dev = kmalloc(sizeof(struct matrox_device) + + dev = kzalloc(sizeof(struct matrox_device) + sizeof(struct w1_bus_master), GFP_KERNEL); if (!dev) { dev_err(&pdev->dev, @@ -173,7 +173,6 @@ static int __devinit matrox_w1_probe(struct pci_dev *pdev, const struct pci_devi return -ENOMEM; } - memset(dev, 0, sizeof(struct matrox_device) + sizeof(struct w1_bus_master)); dev->bus_master = (struct w1_bus_master *)(dev + 1); diff --git a/drivers/w1/slaves/w1_ds2433.c b/drivers/w1/slaves/w1_ds2433.c index cab56005dd4..858c16a544c 100644 --- a/drivers/w1/slaves/w1_ds2433.c +++ b/drivers/w1/slaves/w1_ds2433.c @@ -266,10 +266,9 @@ static int w1_f23_add_slave(struct w1_slave *sl) #ifdef CONFIG_W1_SLAVE_DS2433_CRC struct w1_f23_data *data; - data = kmalloc(sizeof(struct w1_f23_data), GFP_KERNEL); + data = kzalloc(sizeof(struct w1_f23_data), GFP_KERNEL); if (!data) return -ENOMEM; - memset(data, 0, sizeof(struct w1_f23_data)); sl->family_data = data; #endif /* CONFIG_W1_SLAVE_DS2433_CRC */ diff --git a/drivers/w1/w1.c b/drivers/w1/w1.c index c6332108f1c..8d7ab74170d 100644 --- a/drivers/w1/w1.c +++ b/drivers/w1/w1.c @@ -520,7 +520,7 @@ static int w1_attach_slave_device(struct w1_master *dev, struct w1_reg_num *rn) int err; struct w1_netlink_msg msg; - sl = kmalloc(sizeof(struct w1_slave), GFP_KERNEL); + sl = kzalloc(sizeof(struct w1_slave), GFP_KERNEL); if (!sl) { dev_err(&dev->dev, "%s: failed to allocate new slave device.\n", @@ -528,7 +528,6 @@ static int w1_attach_slave_device(struct w1_master *dev, struct w1_reg_num *rn) return -ENOMEM; } - memset(sl, 0, sizeof(*sl)); sl->owner = THIS_MODULE; sl->master = dev; diff --git a/drivers/w1/w1_int.c b/drivers/w1/w1_int.c index 258defdb2ef..2fbd8dd16df 100644 --- a/drivers/w1/w1_int.c +++ b/drivers/w1/w1_int.c @@ -41,7 +41,7 @@ static struct w1_master * w1_alloc_dev(u32 id, int slave_count, int slave_ttl, /* * We are in process context(kernel thread), so can sleep. */ - dev = kmalloc(sizeof(struct w1_master) + sizeof(struct w1_bus_master), GFP_KERNEL); + dev = kzalloc(sizeof(struct w1_master) + sizeof(struct w1_bus_master), GFP_KERNEL); if (!dev) { printk(KERN_ERR "Failed to allocate %zd bytes for new w1 device.\n", @@ -49,7 +49,6 @@ static struct w1_master * w1_alloc_dev(u32 id, int slave_count, int slave_ttl, return NULL; } - memset(dev, 0, sizeof(struct w1_master) + sizeof(struct w1_bus_master)); dev->bus_master = (struct w1_bus_master *)(dev + 1); diff --git a/fs/Kconfig b/fs/Kconfig index 6a649902c5a..58a0650293e 100644 --- a/fs/Kconfig +++ b/fs/Kconfig @@ -1674,7 +1674,7 @@ config NFSD_V3_ACL config NFSD_V4 bool "Provide NFSv4 server support (EXPERIMENTAL)" - depends on NFSD_V3 && EXPERIMENTAL + depends on NFSD && NFSD_V3 && EXPERIMENTAL select RPCSEC_GSS_KRB5 help If you would like to include the NFSv4 server as well as the NFSv2 diff --git a/fs/afs/flock.c b/fs/afs/flock.c index 8f07f8d1bfa..4f77f3caee9 100644 --- a/fs/afs/flock.c +++ b/fs/afs/flock.c @@ -456,7 +456,8 @@ static int afs_do_getlk(struct file *file, struct file_lock *fl) /* check local lock records first */ ret = 0; - if (posix_test_lock(file, fl) == 0) { + posix_test_lock(file, fl); + if (fl->fl_type == F_UNLCK) { /* no local locks; consult the server */ ret = afs_vnode_fetch_status(vnode, NULL, key); if (ret < 0) diff --git a/fs/binfmt_elf.c b/fs/binfmt_elf.c index a27e42bf340..ba24cb2ff6c 100644 --- a/fs/binfmt_elf.c +++ b/fs/binfmt_elf.c @@ -148,6 +148,7 @@ create_elf_tables(struct linux_binprm *bprm, struct elfhdr *exec, elf_addr_t *elf_info; int ei_index = 0; struct task_struct *tsk = current; + struct vm_area_struct *vma; /* * If this architecture has a platform capability string, copy it @@ -234,6 +235,15 @@ create_elf_tables(struct linux_binprm *bprm, struct elfhdr *exec, sp = (elf_addr_t __user *)bprm->p; #endif + + /* + * Grow the stack manually; some architectures have a limit on how + * far ahead a user-space access may be in order to grow the stack. + */ + vma = find_extend_vma(current->mm, bprm->p); + if (!vma) + return -EFAULT; + /* Now, let's put argc (and argv, envp if appropriate) on the stack */ if (__put_user(argc, sp++)) return -EFAULT; @@ -254,8 +264,8 @@ create_elf_tables(struct linux_binprm *bprm, struct elfhdr *exec, size_t len; if (__put_user((elf_addr_t)p, argv++)) return -EFAULT; - len = strnlen_user((void __user *)p, PAGE_SIZE*MAX_ARG_PAGES); - if (!len || len > PAGE_SIZE*MAX_ARG_PAGES) + len = strnlen_user((void __user *)p, MAX_ARG_STRLEN); + if (!len || len > MAX_ARG_STRLEN) return 0; p += len; } @@ -266,8 +276,8 @@ create_elf_tables(struct linux_binprm *bprm, struct elfhdr *exec, size_t len; if (__put_user((elf_addr_t)p, envp++)) return -EFAULT; - len = strnlen_user((void __user *)p, PAGE_SIZE*MAX_ARG_PAGES); - if (!len || len > PAGE_SIZE*MAX_ARG_PAGES) + len = strnlen_user((void __user *)p, MAX_ARG_STRLEN); + if (!len || len > MAX_ARG_STRLEN) return 0; p += len; } @@ -826,10 +836,6 @@ static int load_elf_binary(struct linux_binprm *bprm, struct pt_regs *regs) } /* OK, This is the point of no return */ - current->mm->start_data = 0; - current->mm->end_data = 0; - current->mm->end_code = 0; - current->mm->mmap = NULL; current->flags &= ~PF_FORKNOEXEC; current->mm->def_flags = def_flags; @@ -1051,9 +1057,13 @@ static int load_elf_binary(struct linux_binprm *bprm, struct pt_regs *regs) compute_creds(bprm); current->flags &= ~PF_FORKNOEXEC; - create_elf_tables(bprm, &loc->elf_ex, + retval = create_elf_tables(bprm, &loc->elf_ex, (interpreter_type == INTERPRETER_AOUT), load_addr, interp_load_addr); + if (retval < 0) { + send_sig(SIGKILL, current, 0); + goto out; + } /* N.B. passed_fileno might not be initialized? */ if (interpreter_type == INTERPRETER_AOUT) current->mm->arg_start += strlen(passed_fileno) + 1; @@ -1252,7 +1262,7 @@ static int dump_seek(struct file *file, loff_t off) * * I think we should skip something. But I am not sure how. H.J. */ -static int maydump(struct vm_area_struct *vma) +static int maydump(struct vm_area_struct *vma, unsigned long mm_flags) { /* The vma can be set up to tell us the answer directly. */ if (vma->vm_flags & VM_ALWAYSDUMP) @@ -1262,15 +1272,19 @@ static int maydump(struct vm_area_struct *vma) if (vma->vm_flags & (VM_IO | VM_RESERVED)) return 0; - /* Dump shared memory only if mapped from an anonymous file. */ - if (vma->vm_flags & VM_SHARED) - return vma->vm_file->f_path.dentry->d_inode->i_nlink == 0; + /* By default, dump shared memory if mapped from an anonymous file. */ + if (vma->vm_flags & VM_SHARED) { + if (vma->vm_file->f_path.dentry->d_inode->i_nlink == 0) + return test_bit(MMF_DUMP_ANON_SHARED, &mm_flags); + else + return test_bit(MMF_DUMP_MAPPED_SHARED, &mm_flags); + } - /* If it hasn't been written to, don't write it out */ + /* By default, if it hasn't been written to, don't write it out. */ if (!vma->anon_vma) - return 0; + return test_bit(MMF_DUMP_MAPPED_PRIVATE, &mm_flags); - return 1; + return test_bit(MMF_DUMP_ANON_PRIVATE, &mm_flags); } /* An ELF note in memory */ @@ -1562,6 +1576,7 @@ static int elf_core_dump(long signr, struct pt_regs *regs, struct file *file) #endif int thread_status_size = 0; elf_addr_t *auxv; + unsigned long mm_flags; #ifdef ELF_CORE_WRITE_EXTRA_NOTES int extra_notes_size; #endif @@ -1705,6 +1720,13 @@ static int elf_core_dump(long signr, struct pt_regs *regs, struct file *file) dataoff = offset = roundup(offset, ELF_EXEC_PAGESIZE); + /* + * We must use the same mm->flags while dumping core to avoid + * inconsistency between the program headers and bodies, otherwise an + * unusable core file can be generated. + */ + mm_flags = current->mm->flags; + /* Write program headers for segments dump */ for (vma = first_vma(current, gate_vma); vma != NULL; vma = next_vma(vma, gate_vma)) { @@ -1717,7 +1739,7 @@ static int elf_core_dump(long signr, struct pt_regs *regs, struct file *file) phdr.p_offset = offset; phdr.p_vaddr = vma->vm_start; phdr.p_paddr = 0; - phdr.p_filesz = maydump(vma) ? sz : 0; + phdr.p_filesz = maydump(vma, mm_flags) ? sz : 0; phdr.p_memsz = sz; offset += phdr.p_filesz; phdr.p_flags = vma->vm_flags & VM_READ ? PF_R : 0; @@ -1761,7 +1783,7 @@ static int elf_core_dump(long signr, struct pt_regs *regs, struct file *file) vma = next_vma(vma, gate_vma)) { unsigned long addr; - if (!maydump(vma)) + if (!maydump(vma, mm_flags)) continue; for (addr = vma->vm_start; diff --git a/fs/binfmt_elf_fdpic.c b/fs/binfmt_elf_fdpic.c index 9d62fbad3d4..2f5d8dbe676 100644 --- a/fs/binfmt_elf_fdpic.c +++ b/fs/binfmt_elf_fdpic.c @@ -621,8 +621,8 @@ static int create_elf_fdpic_tables(struct linux_binprm *bprm, p = (char __user *) current->mm->arg_start; for (loop = bprm->argc; loop > 0; loop--) { __put_user((elf_caddr_t) p, argv++); - len = strnlen_user(p, PAGE_SIZE * MAX_ARG_PAGES); - if (!len || len > PAGE_SIZE * MAX_ARG_PAGES) + len = strnlen_user(p, MAX_ARG_STRLEN); + if (!len || len > MAX_ARG_STRLEN) return -EINVAL; p += len; } @@ -633,8 +633,8 @@ static int create_elf_fdpic_tables(struct linux_binprm *bprm, current->mm->env_start = (unsigned long) p; for (loop = bprm->envc; loop > 0; loop--) { __put_user((elf_caddr_t)(unsigned long) p, envp++); - len = strnlen_user(p, PAGE_SIZE * MAX_ARG_PAGES); - if (!len || len > PAGE_SIZE * MAX_ARG_PAGES) + len = strnlen_user(p, MAX_ARG_STRLEN); + if (!len || len > MAX_ARG_STRLEN) return -EINVAL; p += len; } @@ -1181,8 +1181,10 @@ static int dump_seek(struct file *file, loff_t off) * * I think we should skip something. But I am not sure how. H.J. */ -static int maydump(struct vm_area_struct *vma) +static int maydump(struct vm_area_struct *vma, unsigned long mm_flags) { + int dump_ok; + /* Do not dump I/O mapped devices or special mappings */ if (vma->vm_flags & (VM_IO | VM_RESERVED)) { kdcore("%08lx: %08lx: no (IO)", vma->vm_start, vma->vm_flags); @@ -1197,27 +1199,35 @@ static int maydump(struct vm_area_struct *vma) return 0; } - /* Dump shared memory only if mapped from an anonymous file. */ + /* By default, dump shared memory if mapped from an anonymous file. */ if (vma->vm_flags & VM_SHARED) { if (vma->vm_file->f_path.dentry->d_inode->i_nlink == 0) { - kdcore("%08lx: %08lx: no (share)", vma->vm_start, vma->vm_flags); - return 1; + dump_ok = test_bit(MMF_DUMP_ANON_SHARED, &mm_flags); + kdcore("%08lx: %08lx: %s (share)", vma->vm_start, + vma->vm_flags, dump_ok ? "yes" : "no"); + return dump_ok; } - kdcore("%08lx: %08lx: no (share)", vma->vm_start, vma->vm_flags); - return 0; + dump_ok = test_bit(MMF_DUMP_MAPPED_SHARED, &mm_flags); + kdcore("%08lx: %08lx: %s (share)", vma->vm_start, + vma->vm_flags, dump_ok ? "yes" : "no"); + return dump_ok; } #ifdef CONFIG_MMU - /* If it hasn't been written to, don't write it out */ + /* By default, if it hasn't been written to, don't write it out */ if (!vma->anon_vma) { - kdcore("%08lx: %08lx: no (!anon)", vma->vm_start, vma->vm_flags); - return 0; + dump_ok = test_bit(MMF_DUMP_MAPPED_PRIVATE, &mm_flags); + kdcore("%08lx: %08lx: %s (!anon)", vma->vm_start, + vma->vm_flags, dump_ok ? "yes" : "no"); + return dump_ok; } #endif - kdcore("%08lx: %08lx: yes", vma->vm_start, vma->vm_flags); - return 1; + dump_ok = test_bit(MMF_DUMP_ANON_PRIVATE, &mm_flags); + kdcore("%08lx: %08lx: %s", vma->vm_start, vma->vm_flags, + dump_ok ? "yes" : "no"); + return dump_ok; } /* An ELF note in memory */ @@ -1456,15 +1466,15 @@ static int elf_dump_thread_status(long signr, struct elf_thread_status *t) * dump the segments for an MMU process */ #ifdef CONFIG_MMU -static int elf_fdpic_dump_segments(struct file *file, struct mm_struct *mm, - size_t *size, unsigned long *limit) +static int elf_fdpic_dump_segments(struct file *file, size_t *size, + unsigned long *limit, unsigned long mm_flags) { struct vm_area_struct *vma; for (vma = current->mm->mmap; vma; vma = vma->vm_next) { unsigned long addr; - if (!maydump(vma)) + if (!maydump(vma, mm_flags)) continue; for (addr = vma->vm_start; @@ -1511,15 +1521,15 @@ end_coredump: * dump the segments for a NOMMU process */ #ifndef CONFIG_MMU -static int elf_fdpic_dump_segments(struct file *file, struct mm_struct *mm, - size_t *size, unsigned long *limit) +static int elf_fdpic_dump_segments(struct file *file, size_t *size, + unsigned long *limit, unsigned long mm_flags) { struct vm_list_struct *vml; for (vml = current->mm->context.vmlist; vml; vml = vml->next) { struct vm_area_struct *vma = vml->vma; - if (!maydump(vma)) + if (!maydump(vma, mm_flags)) continue; if ((*size += PAGE_SIZE) > *limit) @@ -1570,6 +1580,7 @@ static int elf_fdpic_core_dump(long signr, struct pt_regs *regs, struct vm_list_struct *vml; #endif elf_addr_t *auxv; + unsigned long mm_flags; /* * We no longer stop all VM operations. @@ -1707,6 +1718,13 @@ static int elf_fdpic_core_dump(long signr, struct pt_regs *regs, /* Page-align dumped data */ dataoff = offset = roundup(offset, ELF_EXEC_PAGESIZE); + /* + * We must use the same mm->flags while dumping core to avoid + * inconsistency between the program headers and bodies, otherwise an + * unusable core file can be generated. + */ + mm_flags = current->mm->flags; + /* write program headers for segments dump */ for ( #ifdef CONFIG_MMU @@ -1728,7 +1746,7 @@ static int elf_fdpic_core_dump(long signr, struct pt_regs *regs, phdr.p_offset = offset; phdr.p_vaddr = vma->vm_start; phdr.p_paddr = 0; - phdr.p_filesz = maydump(vma) ? sz : 0; + phdr.p_filesz = maydump(vma, mm_flags) ? sz : 0; phdr.p_memsz = sz; offset += phdr.p_filesz; phdr.p_flags = vma->vm_flags & VM_READ ? PF_R : 0; @@ -1762,7 +1780,7 @@ static int elf_fdpic_core_dump(long signr, struct pt_regs *regs, DUMP_SEEK(dataoff); - if (elf_fdpic_dump_segments(file, current->mm, &size, &limit) < 0) + if (elf_fdpic_dump_segments(file, &size, &limit, mm_flags) < 0) goto end_coredump; #ifdef ELF_CORE_WRITE_EXTRA_DATA diff --git a/fs/binfmt_misc.c b/fs/binfmt_misc.c index 330fd3fe854..42e94b3ab7b 100644 --- a/fs/binfmt_misc.c +++ b/fs/binfmt_misc.c @@ -126,7 +126,9 @@ static int load_misc_binary(struct linux_binprm *bprm, struct pt_regs *regs) goto _ret; if (!(fmt->flags & MISC_FMT_PRESERVE_ARGV0)) { - remove_arg_zero(bprm); + retval = remove_arg_zero(bprm); + if (retval) + goto _ret; } if (fmt->flags & MISC_FMT_OPEN_BINARY) { diff --git a/fs/binfmt_script.c b/fs/binfmt_script.c index 304c88544d8..4d0e0f6d327 100644 --- a/fs/binfmt_script.c +++ b/fs/binfmt_script.c @@ -67,7 +67,9 @@ static int load_script(struct linux_binprm *bprm,struct pt_regs *regs) * This is done in reverse order, because of how the * user environment and arguments are stored. */ - remove_arg_zero(bprm); + retval = remove_arg_zero(bprm); + if (retval) + return retval; retval = copy_strings_kernel(1, &bprm->interp, bprm); if (retval < 0) return retval; bprm->argc++; diff --git a/fs/buffer.c b/fs/buffer.c index 0f900671423..02ebb1f1d3b 100644 --- a/fs/buffer.c +++ b/fs/buffer.c @@ -2194,6 +2194,52 @@ int generic_commit_write(struct file *file, struct page *page, return 0; } +/* + * block_page_mkwrite() is not allowed to change the file size as it gets + * called from a page fault handler when a page is first dirtied. Hence we must + * be careful to check for EOF conditions here. We set the page up correctly + * for a written page which means we get ENOSPC checking when writing into + * holes and correct delalloc and unwritten extent mapping on filesystems that + * support these features. + * + * We are not allowed to take the i_mutex here so we have to play games to + * protect against truncate races as the page could now be beyond EOF. Because + * vmtruncate() writes the inode size before removing pages, once we have the + * page lock we can determine safely if the page is beyond EOF. If it is not + * beyond EOF, then the page is guaranteed safe against truncation until we + * unlock the page. + */ +int +block_page_mkwrite(struct vm_area_struct *vma, struct page *page, + get_block_t get_block) +{ + struct inode *inode = vma->vm_file->f_path.dentry->d_inode; + unsigned long end; + loff_t size; + int ret = -EINVAL; + + lock_page(page); + size = i_size_read(inode); + if ((page->mapping != inode->i_mapping) || + ((page->index << PAGE_CACHE_SHIFT) > size)) { + /* page got truncated out from underneath us */ + goto out_unlock; + } + + /* page is wholly or partially inside EOF */ + if (((page->index + 1) << PAGE_CACHE_SHIFT) > size) + end = size & ~PAGE_CACHE_MASK; + else + end = PAGE_CACHE_SIZE; + + ret = block_prepare_write(page, 0, end, get_block); + if (!ret) + ret = block_commit_write(page, 0, end); + +out_unlock: + unlock_page(page); + return ret; +} /* * nobh_prepare_write()'s prereads are special: the buffer_heads are freed @@ -2977,6 +3023,7 @@ EXPORT_SYMBOL(__brelse); EXPORT_SYMBOL(__wait_on_buffer); EXPORT_SYMBOL(block_commit_write); EXPORT_SYMBOL(block_prepare_write); +EXPORT_SYMBOL(block_page_mkwrite); EXPORT_SYMBOL(block_read_full_page); EXPORT_SYMBOL(block_sync_page); EXPORT_SYMBOL(block_truncate_page); diff --git a/fs/char_dev.c b/fs/char_dev.c index 164a45cdaf5..bbbf07baa14 100644 --- a/fs/char_dev.c +++ b/fs/char_dev.c @@ -321,14 +321,13 @@ void unregister_chrdev_region(dev_t from, unsigned count) } } -int unregister_chrdev(unsigned int major, const char *name) +void unregister_chrdev(unsigned int major, const char *name) { struct char_device_struct *cd; cd = __unregister_chrdev_region(major, 0, 256); if (cd && cd->cdev) cdev_del(cd->cdev); kfree(cd); - return 0; } static DEFINE_SPINLOCK(cdev_lock); diff --git a/fs/coda/cache.c b/fs/coda/cache.c index fcb88fa8d2f..8a2370341c7 100644 --- a/fs/coda/cache.c +++ b/fs/coda/cache.c @@ -43,17 +43,12 @@ void coda_cache_enter(struct inode *inode, int mask) void coda_cache_clear_inode(struct inode *inode) { struct coda_inode_info *cii = ITOC(inode); - cii->c_cached_perm = 0; + cii->c_cached_epoch = atomic_read(&permission_epoch) - 1; } /* remove all acl caches */ void coda_cache_clear_all(struct super_block *sb) { - struct coda_sb_info *sbi; - - sbi = coda_sbp(sb); - BUG_ON(!sbi); - atomic_inc(&permission_epoch); } diff --git a/fs/coda/cnode.c b/fs/coda/cnode.c index 28c872747f8..a7a780929ee 100644 --- a/fs/coda/cnode.c +++ b/fs/coda/cnode.c @@ -55,11 +55,6 @@ static int coda_set_inode(struct inode *inode, void *data) return 0; } -static int coda_fail_inode(struct inode *inode, void *data) -{ - return -1; -} - struct inode * coda_iget(struct super_block * sb, struct CodaFid * fid, struct coda_vattr * attr) { @@ -141,7 +136,7 @@ struct inode *coda_fid_to_inode(struct CodaFid *fid, struct super_block *sb) return NULL; } - inode = iget5_locked(sb, hash, coda_test_inode, coda_fail_inode, fid); + inode = ilookup5(sb, hash, coda_test_inode, fid); if ( !inode ) return NULL; diff --git a/fs/coda/coda_int.h b/fs/coda/coda_int.h index 9e6338fea51..8ccd5ed81d9 100644 --- a/fs/coda/coda_int.h +++ b/fs/coda/coda_int.h @@ -1,12 +1,19 @@ #ifndef _CODA_INT_ #define _CODA_INT_ +struct dentry; + extern struct file_system_type coda_fs_type; +extern unsigned long coda_timeout; +extern int coda_hard; +extern int coda_fake_statfs; void coda_destroy_inodecache(void); int coda_init_inodecache(void); int coda_fsync(struct file *coda_file, struct dentry *coda_dentry, int datasync); +void coda_sysctl_init(void); +void coda_sysctl_clean(void); #endif /* _CODA_INT_ */ diff --git a/fs/coda/dir.c b/fs/coda/dir.c index 898a86dde8f..04a3dd84c99 100644 --- a/fs/coda/dir.c +++ b/fs/coda/dir.c @@ -25,7 +25,6 @@ #include <linux/coda_psdev.h> #include <linux/coda_fs_i.h> #include <linux/coda_cache.h> -#include <linux/coda_proc.h> #include "coda_int.h" @@ -43,15 +42,15 @@ static int coda_rename(struct inode *old_inode, struct dentry *old_dentry, struct inode *new_inode, struct dentry *new_dentry); /* dir file-ops */ -static int coda_readdir(struct file *file, void *dirent, filldir_t filldir); +static int coda_readdir(struct file *file, void *buf, filldir_t filldir); /* dentry ops */ static int coda_dentry_revalidate(struct dentry *de, struct nameidata *nd); static int coda_dentry_delete(struct dentry *); /* support routines */ -static int coda_venus_readdir(struct file *filp, filldir_t filldir, - void *dirent, struct dentry *dir); +static int coda_venus_readdir(struct file *coda_file, void *buf, + filldir_t filldir); /* same as fs/bad_inode.c */ static int coda_return_EIO(void) @@ -97,58 +96,45 @@ const struct file_operations coda_dir_operations = { /* access routines: lookup, readlink, permission */ static struct dentry *coda_lookup(struct inode *dir, struct dentry *entry, struct nameidata *nd) { - struct inode *res_inode = NULL; + struct inode *inode = NULL; struct CodaFid resfid = { { 0, } }; - int dropme = 0; /* to indicate entry should not be cached */ int type = 0; int error = 0; const char *name = entry->d_name.name; size_t length = entry->d_name.len; - - if ( length > CODA_MAXNAMLEN ) { - printk("name too long: lookup, %s (%*s)\n", + + if (length > CODA_MAXNAMLEN) { + printk(KERN_ERR "name too long: lookup, %s (%*s)\n", coda_i2s(dir), (int)length, name); return ERR_PTR(-ENAMETOOLONG); } + /* control object, create inode on the fly */ + if (coda_isroot(dir) && coda_iscontrol(name, length)) { + error = coda_cnode_makectl(&inode, dir->i_sb); + type = CODA_NOCACHE; + goto exit; + } + lock_kernel(); - /* control object, create inode on the fly */ - if (coda_isroot(dir) && coda_iscontrol(name, length)) { - error = coda_cnode_makectl(&res_inode, dir->i_sb); - dropme = 1; - goto exit; - } - error = venus_lookup(dir->i_sb, coda_i2f(dir), - (const char *)name, length, &type, &resfid); + error = venus_lookup(dir->i_sb, coda_i2f(dir), name, length, + &type, &resfid); + if (!error) + error = coda_cnode_make(&inode, &resfid, dir->i_sb); - res_inode = NULL; - if (!error) { - if (type & CODA_NOCACHE) { - type &= (~CODA_NOCACHE); - dropme = 1; - } + unlock_kernel(); - error = coda_cnode_make(&res_inode, &resfid, dir->i_sb); - if (error) { - unlock_kernel(); - return ERR_PTR(error); - } - } else if (error != -ENOENT) { - unlock_kernel(); + if (error && error != -ENOENT) return ERR_PTR(error); - } exit: - entry->d_time = 0; entry->d_op = &coda_dentry_operations; - d_add(entry, res_inode); - if ( dropme ) { - d_drop(entry); - coda_flag_inode(res_inode, C_VATTR); - } - unlock_kernel(); - return NULL; + + if (inode && (type & CODA_NOCACHE)) + coda_flag_inode(inode, C_VATTR | C_PURGE); + + return d_splice_alias(inode, entry); } @@ -161,8 +147,6 @@ int coda_permission(struct inode *inode, int mask, struct nameidata *nd) lock_kernel(); - coda_vfs_stat.permission++; - if (coda_cache_check(inode, mask)) goto out; @@ -173,12 +157,11 @@ int coda_permission(struct inode *inode, int mask, struct nameidata *nd) out: unlock_kernel(); - - return error; + return error; } -static inline void coda_dir_changed(struct inode *dir, int link) +static inline void coda_dir_update_mtime(struct inode *dir) { #ifdef REQUERY_VENUS_FOR_MTIME /* invalidate the directory cnode's attributes so we refetch the @@ -186,12 +169,27 @@ static inline void coda_dir_changed(struct inode *dir, int link) coda_flag_inode(dir, C_VATTR); #else /* optimistically we can also act as if our nose bleeds. The - * granularity of the mtime is coarse anyways so we might actually be - * right most of the time. Note: we only do this for directories. */ + * granularity of the mtime is coarse anyways so we might actually be + * right most of the time. Note: we only do this for directories. */ dir->i_mtime = dir->i_ctime = CURRENT_TIME_SEC; #endif - if (link) - dir->i_nlink += link; +} + +/* we have to wrap inc_nlink/drop_nlink because sometimes userspace uses a + * trick to fool GNU find's optimizations. If we can't be sure of the link + * (because of volume mount points) we set i_nlink to 1 which forces find + * to consider every child as a possible directory. We should also never + * see an increment or decrement for deleted directories where i_nlink == 0 */ +static inline void coda_dir_inc_nlink(struct inode *dir) +{ + if (dir->i_nlink >= 2) + inc_nlink(dir); +} + +static inline void coda_dir_drop_nlink(struct inode *dir) +{ + if (dir->i_nlink > 2) + drop_nlink(dir); } /* creation routines: create, mknod, mkdir, link, symlink */ @@ -205,7 +203,6 @@ static int coda_create(struct inode *dir, struct dentry *de, int mode, struct na struct coda_vattr attrs; lock_kernel(); - coda_vfs_stat.create++; if (coda_isroot(dir) && coda_iscontrol(name, length)) { unlock_kernel(); @@ -229,10 +226,10 @@ static int coda_create(struct inode *dir, struct dentry *de, int mode, struct na } /* invalidate the directory cnode's attributes */ - coda_dir_changed(dir, 0); + coda_dir_update_mtime(dir); unlock_kernel(); d_instantiate(de, inode); - return 0; + return 0; } static int coda_mkdir(struct inode *dir, struct dentry *de, int mode) @@ -245,7 +242,6 @@ static int coda_mkdir(struct inode *dir, struct dentry *de, int mode) struct CodaFid newfid; lock_kernel(); - coda_vfs_stat.mkdir++; if (coda_isroot(dir) && coda_iscontrol(name, len)) { unlock_kernel(); @@ -268,12 +264,13 @@ static int coda_mkdir(struct inode *dir, struct dentry *de, int mode) d_drop(de); return PTR_ERR(inode); } - + /* invalidate the directory cnode's attributes */ - coda_dir_changed(dir, 1); + coda_dir_inc_nlink(dir); + coda_dir_update_mtime(dir); unlock_kernel(); d_instantiate(de, inode); - return 0; + return 0; } /* try to make de an entry in dir_inodde linked to source_de */ @@ -286,7 +283,6 @@ static int coda_link(struct dentry *source_de, struct inode *dir_inode, int error; lock_kernel(); - coda_vfs_stat.link++; if (coda_isroot(dir_inode) && coda_iscontrol(name, len)) { unlock_kernel(); @@ -296,16 +292,16 @@ static int coda_link(struct dentry *source_de, struct inode *dir_inode, error = venus_link(dir_inode->i_sb, coda_i2f(inode), coda_i2f(dir_inode), (const char *)name, len); - if (error) { + if (error) { d_drop(de); goto out; } - coda_dir_changed(dir_inode, 0); + coda_dir_update_mtime(dir_inode); atomic_inc(&inode->i_count); d_instantiate(de, inode); inc_nlink(inode); - + out: unlock_kernel(); return(error); @@ -318,10 +314,9 @@ static int coda_symlink(struct inode *dir_inode, struct dentry *de, const char *name = de->d_name.name; int len = de->d_name.len; int symlen; - int error=0; - + int error = 0; + lock_kernel(); - coda_vfs_stat.symlink++; if (coda_isroot(dir_inode) && coda_iscontrol(name, len)) { unlock_kernel(); @@ -336,18 +331,18 @@ static int coda_symlink(struct inode *dir_inode, struct dentry *de, /* * This entry is now negative. Since we do not create - * an inode for the entry we have to drop it. + * an inode for the entry we have to drop it. */ d_drop(de); - error = venus_symlink(dir_inode->i_sb, coda_i2f(dir_inode), name, len, + error = venus_symlink(dir_inode->i_sb, coda_i2f(dir_inode), name, len, symname, symlen); /* mtime is no good anymore */ if ( !error ) - coda_dir_changed(dir_inode, 0); + coda_dir_update_mtime(dir_inode); unlock_kernel(); - return error; + return error; } /* destruction routines: unlink, rmdir */ @@ -358,79 +353,70 @@ int coda_unlink(struct inode *dir, struct dentry *de) int len = de->d_name.len; lock_kernel(); - coda_vfs_stat.unlink++; - error = venus_remove(dir->i_sb, coda_i2f(dir), name, len); - if ( error ) { + error = venus_remove(dir->i_sb, coda_i2f(dir), name, len); + if ( error ) { unlock_kernel(); - return error; - } + return error; + } - coda_dir_changed(dir, 0); + coda_dir_update_mtime(dir); drop_nlink(de->d_inode); unlock_kernel(); - - return 0; + return 0; } int coda_rmdir(struct inode *dir, struct dentry *de) { const char *name = de->d_name.name; int len = de->d_name.len; - int error; + int error; lock_kernel(); - coda_vfs_stat.rmdir++; - if (!d_unhashed(de)) { - unlock_kernel(); - return -EBUSY; - } error = venus_rmdir(dir->i_sb, coda_i2f(dir), name, len); + if (!error) { + /* VFS may delete the child */ + if (de->d_inode) + de->d_inode->i_nlink = 0; - if ( error ) { - unlock_kernel(); - return error; - } - - coda_dir_changed(dir, -1); - drop_nlink(de->d_inode); - d_delete(de); + /* fix the link count of the parent */ + coda_dir_drop_nlink(dir); + coda_dir_update_mtime(dir); + } unlock_kernel(); - - return 0; + return error; } /* rename */ -static int coda_rename(struct inode *old_dir, struct dentry *old_dentry, +static int coda_rename(struct inode *old_dir, struct dentry *old_dentry, struct inode *new_dir, struct dentry *new_dentry) { - const char *old_name = old_dentry->d_name.name; - const char *new_name = new_dentry->d_name.name; + const char *old_name = old_dentry->d_name.name; + const char *new_name = new_dentry->d_name.name; int old_length = old_dentry->d_name.len; int new_length = new_dentry->d_name.len; - int link_adjust = 0; - int error; + int error; lock_kernel(); - coda_vfs_stat.rename++; - error = venus_rename(old_dir->i_sb, coda_i2f(old_dir), - coda_i2f(new_dir), old_length, new_length, + error = venus_rename(old_dir->i_sb, coda_i2f(old_dir), + coda_i2f(new_dir), old_length, new_length, (const char *) old_name, (const char *)new_name); - if ( !error ) { + if ( !error ) { if ( new_dentry->d_inode ) { - if ( S_ISDIR(new_dentry->d_inode->i_mode) ) - link_adjust = 1; - - coda_dir_changed(old_dir, -link_adjust); - coda_dir_changed(new_dir, link_adjust); + if ( S_ISDIR(new_dentry->d_inode->i_mode) ) { + coda_dir_drop_nlink(old_dir); + coda_dir_inc_nlink(new_dir); + } + coda_dir_update_mtime(old_dir); + coda_dir_update_mtime(new_dir); coda_flag_inode(new_dentry->d_inode, C_VATTR); } else { coda_flag_inode(old_dir, C_VATTR); coda_flag_inode(new_dir, C_VATTR); - } + } } unlock_kernel(); @@ -439,44 +425,41 @@ static int coda_rename(struct inode *old_dir, struct dentry *old_dentry, /* file operations for directories */ -int coda_readdir(struct file *coda_file, void *dirent, filldir_t filldir) +int coda_readdir(struct file *coda_file, void *buf, filldir_t filldir) { - struct dentry *coda_dentry = coda_file->f_path.dentry; struct coda_file_info *cfi; struct file *host_file; - struct inode *host_inode; int ret; cfi = CODA_FTOC(coda_file); BUG_ON(!cfi || cfi->cfi_magic != CODA_MAGIC); host_file = cfi->cfi_container; - coda_vfs_stat.readdir++; + if (!host_file->f_op) + return -ENOTDIR; - host_inode = host_file->f_path.dentry->d_inode; - mutex_lock(&host_inode->i_mutex); - host_file->f_pos = coda_file->f_pos; + 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. + * and as such we need grab the inode mutex. */ + struct inode *host_inode = host_file->f_path.dentry->d_inode; - if (!host_file->f_op->readdir) { - /* Venus: we must read Venus dirents from the file */ - ret = coda_venus_readdir(host_file, filldir, dirent, coda_dentry); - } else { - /* potemkin case: we were handed a directory inode. */ - /* Yuk, we can't call vfs_readdir because we are already - * holding the inode semaphore. */ - ret = -ENOTDIR; - if (!host_file->f_op || !host_file->f_op->readdir) - goto out; + mutex_lock(&host_inode->i_mutex); + host_file->f_pos = coda_file->f_pos; ret = -ENOENT; if (!IS_DEADDIR(host_inode)) { - ret = host_file->f_op->readdir(host_file, dirent, filldir); + ret = host_file->f_op->readdir(host_file, buf, filldir); file_accessed(host_file); } + + coda_file->f_pos = host_file->f_pos; + mutex_unlock(&host_inode->i_mutex); } -out: - coda_file->f_pos = host_file->f_pos; - mutex_unlock(&host_inode->i_mutex); + else /* Venus: we must read Venus dirents from a file */ + ret = coda_venus_readdir(coda_file, buf, filldir); return ret; } @@ -501,57 +484,68 @@ static inline unsigned int CDT2DT(unsigned char cdt) } /* support routines */ -static int coda_venus_readdir(struct file *filp, filldir_t filldir, - void *dirent, struct dentry *dir) +static int coda_venus_readdir(struct file *coda_file, void *buf, + filldir_t filldir) { int result = 0; /* # of entries returned */ + struct coda_file_info *cfi; + struct coda_inode_info *cii; + struct file *host_file; + struct dentry *de; struct venus_dirent *vdir; unsigned long vdir_size = (unsigned long)(&((struct venus_dirent *)0)->d_name); unsigned int type; struct qstr name; ino_t ino; - int ret, i; + int ret; + + cfi = CODA_FTOC(coda_file); + BUG_ON(!cfi || cfi->cfi_magic != CODA_MAGIC); + host_file = cfi->cfi_container; + + de = coda_file->f_path.dentry; + cii = ITOC(de->d_inode); vdir = kmalloc(sizeof(*vdir), GFP_KERNEL); if (!vdir) return -ENOMEM; - i = filp->f_pos; - switch(i) { + switch (coda_file->f_pos) { case 0: - ret = filldir(dirent, ".", 1, 0, dir->d_inode->i_ino, DT_DIR); + ret = filldir(buf, ".", 1, 0, de->d_inode->i_ino, DT_DIR); if (ret < 0) break; result++; - filp->f_pos++; + coda_file->f_pos++; /* fallthrough */ case 1: - ret = filldir(dirent, "..", 2, 1, dir->d_parent->d_inode->i_ino, DT_DIR); + ret = filldir(buf, "..", 2, 1, de->d_parent->d_inode->i_ino, DT_DIR); if (ret < 0) break; result++; - filp->f_pos++; + coda_file->f_pos++; /* fallthrough */ default: while (1) { /* read entries from the directory file */ - ret = kernel_read(filp, filp->f_pos - 2, (char *)vdir, + ret = kernel_read(host_file, coda_file->f_pos - 2, (char *)vdir, sizeof(*vdir)); if (ret < 0) { - printk("coda_venus_readdir: read dir failed %d\n", ret); + printk(KERN_ERR "coda readdir: read dir %s failed %d\n", + coda_f2s(&cii->c_fid), ret); break; } if (ret == 0) break; /* end of directory file reached */ /* catch truncated reads */ if (ret < vdir_size || ret < vdir_size + vdir->d_namlen) { - printk("coda_venus_readdir: short read: %ld\n", - filp->f_path.dentry->d_inode->i_ino); + printk(KERN_ERR "coda readdir: short read on %s\n", + coda_f2s(&cii->c_fid)); ret = -EBADF; break; } /* validate whether the directory file actually makes sense */ if (vdir->d_reclen < vdir_size + vdir->d_namlen) { - printk("coda_venus_readdir: Invalid dir: %ld\n", - filp->f_path.dentry->d_inode->i_ino); + printk(KERN_ERR "coda readdir: invalid dir %s\n", + coda_f2s(&cii->c_fid)); ret = -EBADF; break; } @@ -570,21 +564,21 @@ static int coda_venus_readdir(struct file *filp, filldir_t filldir, * userspace doesn't have to worry about breaking * getcwd by having mismatched inode numbers for * internal volume mountpoints. */ - ino = find_inode_number(dir, &name); + ino = find_inode_number(de, &name); if (!ino) ino = vdir->d_fileno; type = CDT2DT(vdir->d_type); - ret = filldir(dirent, name.name, name.len, filp->f_pos, - ino, type); + ret = filldir(buf, name.name, name.len, + coda_file->f_pos, ino, type); /* failure means no space for filling in this round */ if (ret < 0) break; result++; } /* we'll always have progress because d_reclen is unsigned and * we've already established it is non-zero. */ - filp->f_pos += vdir->d_reclen; + coda_file->f_pos += vdir->d_reclen; + } } - } kfree(vdir); return result ? result : ret; } diff --git a/fs/coda/file.c b/fs/coda/file.c index 99dbe866816..7594962604c 100644 --- a/fs/coda/file.c +++ b/fs/coda/file.c @@ -22,7 +22,6 @@ #include <linux/coda_linux.h> #include <linux/coda_fs_i.h> #include <linux/coda_psdev.h> -#include <linux/coda_proc.h> #include "coda_int.h" @@ -134,8 +133,6 @@ int coda_open(struct inode *coda_inode, struct file *coda_file) unsigned short coda_flags = coda_flags_to_cflags(flags); struct coda_file_info *cfi; - coda_vfs_stat.open++; - cfi = kmalloc(sizeof(struct coda_file_info), GFP_KERNEL); if (!cfi) return -ENOMEM; @@ -143,8 +140,11 @@ int coda_open(struct inode *coda_inode, struct file *coda_file) lock_kernel(); error = venus_open(coda_inode->i_sb, coda_i2f(coda_inode), coda_flags, - &host_file); - if (error || !host_file) { + &host_file); + if (!host_file) + error = -EIO; + + if (error) { kfree(cfi); unlock_kernel(); return error; @@ -173,8 +173,6 @@ int coda_flush(struct file *coda_file, fl_owner_t id) lock_kernel(); - coda_vfs_stat.flush++; - /* last close semantics */ fcnt = file_count(coda_file); if (fcnt > 1) @@ -216,8 +214,7 @@ int coda_release(struct inode *coda_inode, struct file *coda_file) int err = 0; lock_kernel(); - coda_vfs_stat.release++; - + if (!use_coda_close) { err = venus_release(coda_inode->i_sb, coda_i2f(coda_inode), coda_flags); @@ -268,8 +265,6 @@ int coda_fsync(struct file *coda_file, struct dentry *coda_dentry, int datasync) BUG_ON(!cfi || cfi->cfi_magic != CODA_MAGIC); host_file = cfi->cfi_container; - coda_vfs_stat.fsync++; - if (host_file->f_op && host_file->f_op->fsync) { host_dentry = host_file->f_path.dentry; host_inode = host_dentry->d_inode; diff --git a/fs/coda/inode.c b/fs/coda/inode.c index dbff1bd4fb9..6771a4271e3 100644 --- a/fs/coda/inode.c +++ b/fs/coda/inode.c @@ -83,7 +83,7 @@ void coda_destroy_inodecache(void) static int coda_remount(struct super_block *sb, int *flags, char *data) { - *flags |= MS_NODIRATIME; + *flags |= MS_NOATIME; return 0; } @@ -141,11 +141,10 @@ static int get_device_index(struct coda_mount_data *data) static int coda_fill_super(struct super_block *sb, void *data, int silent) { - struct inode *root = NULL; - struct coda_sb_info *sbi = NULL; + struct inode *root = NULL; struct venus_comm *vc = NULL; struct CodaFid fid; - int error; + int error; int idx; idx = get_device_index((struct coda_mount_data *) data); @@ -167,21 +166,14 @@ static int coda_fill_super(struct super_block *sb, void *data, int silent) return -EBUSY; } - sbi = kmalloc(sizeof(struct coda_sb_info), GFP_KERNEL); - if(!sbi) { - return -ENOMEM; - } - vc->vc_sb = sb; - sbi->sbi_vcomm = vc; - - sb->s_fs_info = sbi; - sb->s_flags |= MS_NODIRATIME; /* probably even noatime */ - sb->s_blocksize = 1024; /* XXXXX what do we put here?? */ - sb->s_blocksize_bits = 10; - sb->s_magic = CODA_SUPER_MAGIC; - sb->s_op = &coda_super_operations; + sb->s_fs_info = vc; + sb->s_flags |= MS_NOATIME; + sb->s_blocksize = 4096; /* XXXXX what do we put here?? */ + sb->s_blocksize_bits = 12; + sb->s_magic = CODA_SUPER_MAGIC; + sb->s_op = &coda_super_operations; /* get root fid from Venus: this needs the root inode */ error = venus_rootfid(sb, &fid); @@ -207,26 +199,20 @@ static int coda_fill_super(struct super_block *sb, void *data, int silent) return 0; error: - if (sbi) { - kfree(sbi); - if(vc) - vc->vc_sb = NULL; - } if (root) - iput(root); + iput(root); + if (vc) + vc->vc_sb = NULL; - return -EINVAL; + return -EINVAL; } static void coda_put_super(struct super_block *sb) { - struct coda_sb_info *sbi; - - sbi = coda_sbp(sb); - sbi->sbi_vcomm->vc_sb = NULL; + coda_vcp(sb)->vc_sb = NULL; + sb->s_fs_info = NULL; printk("Coda: Bye bye.\n"); - kfree(sbi); } static void coda_clear_inode(struct inode *inode) @@ -296,7 +282,7 @@ static int coda_statfs(struct dentry *dentry, struct kstatfs *buf) /* and fill in the rest */ buf->f_type = CODA_SUPER_MAGIC; - buf->f_bsize = 1024; + buf->f_bsize = 4096; buf->f_namelen = CODA_MAXNAMLEN; return 0; diff --git a/fs/coda/psdev.c b/fs/coda/psdev.c index 803aacf0d49..dcc6aead70f 100644 --- a/fs/coda/psdev.c +++ b/fs/coda/psdev.c @@ -45,12 +45,9 @@ #include <linux/coda_linux.h> #include <linux/coda_fs_i.h> #include <linux/coda_psdev.h> -#include <linux/coda_proc.h> #include "coda_int.h" -#define upc_free(r) kfree(r) - /* statistics */ int coda_hard; /* allows signals during upcalls */ unsigned long coda_timeout = 30; /* .. secs, then signals will dequeue */ @@ -195,7 +192,8 @@ static ssize_t coda_psdev_write(struct file *file, const char __user *buf, if (req->uc_opcode == CODA_OPEN_BY_FD) { struct coda_open_by_fd_out *outp = (struct coda_open_by_fd_out *)req->uc_data; - outp->fh = fget(outp->fd); + if (!outp->oh.result) + outp->fh = fget(outp->fd); } wake_up(&req->uc_sleep); @@ -263,7 +261,7 @@ static ssize_t coda_psdev_read(struct file * file, char __user * buf, } CODA_FREE(req->uc_data, sizeof(struct coda_in_hdr)); - upc_free(req); + kfree(req); out: unlock_kernel(); return (count ? count : retval); @@ -271,71 +269,70 @@ out: static int coda_psdev_open(struct inode * inode, struct file * file) { - struct venus_comm *vcp; - int idx; + struct venus_comm *vcp; + int idx, err; - lock_kernel(); idx = iminor(inode); - if(idx >= MAX_CODADEVS) { - unlock_kernel(); + if (idx < 0 || idx >= MAX_CODADEVS) return -ENODEV; - } + lock_kernel(); + + err = -EBUSY; vcp = &coda_comms[idx]; - if(vcp->vc_inuse) { - unlock_kernel(); - return -EBUSY; - } - - if (!vcp->vc_inuse++) { + if (!vcp->vc_inuse) { + vcp->vc_inuse++; + INIT_LIST_HEAD(&vcp->vc_pending); INIT_LIST_HEAD(&vcp->vc_processing); init_waitqueue_head(&vcp->vc_waitq); vcp->vc_sb = NULL; vcp->vc_seq = 0; + + file->private_data = vcp; + err = 0; } - - file->private_data = vcp; unlock_kernel(); - return 0; + return err; } static int coda_psdev_release(struct inode * inode, struct file * file) { - struct venus_comm *vcp = (struct venus_comm *) file->private_data; - struct upc_req *req, *tmp; + struct venus_comm *vcp = (struct venus_comm *) file->private_data; + struct upc_req *req, *tmp; - lock_kernel(); - if ( !vcp->vc_inuse ) { - unlock_kernel(); + if (!vcp || !vcp->vc_inuse ) { printk("psdev_release: Not open.\n"); return -1; } - if (--vcp->vc_inuse) { - unlock_kernel(); - return 0; - } - - /* Wakeup clients so they can return. */ + lock_kernel(); + + /* Wakeup clients so they can return. */ list_for_each_entry_safe(req, tmp, &vcp->vc_pending, uc_chain) { + list_del(&req->uc_chain); + /* Async requests need to be freed here */ if (req->uc_flags & REQ_ASYNC) { CODA_FREE(req->uc_data, sizeof(struct coda_in_hdr)); - upc_free(req); + kfree(req); continue; } req->uc_flags |= REQ_ABORT; wake_up(&req->uc_sleep); - } - - list_for_each_entry(req, &vcp->vc_processing, uc_chain) { + } + + list_for_each_entry_safe(req, tmp, &vcp->vc_processing, uc_chain) { + list_del(&req->uc_chain); + req->uc_flags |= REQ_ABORT; - wake_up(&req->uc_sleep); - } + wake_up(&req->uc_sleep); + } + file->private_data = NULL; + vcp->vc_inuse--; unlock_kernel(); return 0; } @@ -376,21 +373,20 @@ out: return err; } - -MODULE_AUTHOR("Peter J. Braam <braam@cs.cmu.edu>"); +MODULE_AUTHOR("Jan Harkes, Peter J. Braam"); +MODULE_DESCRIPTION("Coda Distributed File System VFS interface"); +MODULE_ALIAS_CHARDEV_MAJOR(CODA_PSDEV_MAJOR); MODULE_LICENSE("GPL"); +#ifdef CONFIG_CODA_FS_OLD_API +MODULE_VERSION("5.3.21"); +#else +MODULE_VERSION("6.6"); +#endif static int __init init_coda(void) { int status; int i; - printk(KERN_INFO "Coda Kernel/Venus communications, " -#ifdef CONFIG_CODA_FS_OLD_API - "v5.3.20" -#else - "v6.0.0" -#endif - ", coda@cs.cmu.edu\n"); status = coda_init_inodecache(); if (status) diff --git a/fs/coda/symlink.c b/fs/coda/symlink.c index 76e00a65a75..4513b725845 100644 --- a/fs/coda/symlink.c +++ b/fs/coda/symlink.c @@ -20,7 +20,6 @@ #include <linux/coda_linux.h> #include <linux/coda_psdev.h> #include <linux/coda_fs_i.h> -#include <linux/coda_proc.h> static int coda_symlink_filler(struct file *file, struct page *page) { @@ -32,7 +31,6 @@ static int coda_symlink_filler(struct file *file, struct page *page) lock_kernel(); cii = ITOC(inode); - coda_vfs_stat.follow_link++; error = venus_readlink(inode->i_sb, &cii->c_fid, p, &len); unlock_kernel(); diff --git a/fs/coda/sysctl.c b/fs/coda/sysctl.c index c57a1fa7cf2..81b7771c646 100644 --- a/fs/coda/sysctl.c +++ b/fs/coda/sysctl.c @@ -5,181 +5,14 @@ * * Carnegie Mellon encourages users to contribute improvements to * the Coda project. Contact Peter Braam (coda@cs.cmu.edu). - * - * CODA operation statistics - * (c) March, 1998 Zhanyong Wan <zhanyong.wan@yale.edu> - * */ -#include <linux/time.h> -#include <linux/mm.h> #include <linux/sysctl.h> -#include <linux/proc_fs.h> -#include <linux/seq_file.h> -#include <linux/slab.h> -#include <linux/stat.h> -#include <linux/ctype.h> -#include <linux/bitops.h> -#include <asm/uaccess.h> -#include <linux/utsname.h> -#include <linux/module.h> -#include <linux/coda.h> -#include <linux/coda_linux.h> -#include <linux/coda_fs_i.h> -#include <linux/coda_psdev.h> -#include <linux/coda_cache.h> -#include <linux/coda_proc.h> +#include "coda_int.h" static struct ctl_table_header *fs_table_header; -#define CODA_TIMEOUT 3 /* timeout on upcalls to become intrble */ -#define CODA_HARD 5 /* mount type "hard" or "soft" */ -#define CODA_VFS 6 /* vfs statistics */ -#define CODA_CACHE_INV 9 /* cache invalidation statistics */ -#define CODA_FAKE_STATFS 10 /* don't query venus for actual cache usage */ - -struct coda_vfs_stats coda_vfs_stat; -static struct coda_cache_inv_stats coda_cache_inv_stat; - -static void reset_coda_vfs_stats( void ) -{ - memset( &coda_vfs_stat, 0, sizeof( coda_vfs_stat ) ); -} - -static void reset_coda_cache_inv_stats( void ) -{ - memset( &coda_cache_inv_stat, 0, sizeof( coda_cache_inv_stat ) ); -} - -static int do_reset_coda_vfs_stats( ctl_table * table, int write, - struct file * filp, void __user * buffer, - size_t * lenp, loff_t * ppos ) -{ - if ( write ) { - reset_coda_vfs_stats(); - - *ppos += *lenp; - } else { - *lenp = 0; - } - - return 0; -} - -static int do_reset_coda_cache_inv_stats( ctl_table * table, int write, - struct file * filp, - void __user * buffer, - size_t * lenp, loff_t * ppos ) -{ - if ( write ) { - reset_coda_cache_inv_stats(); - - *ppos += *lenp; - } else { - *lenp = 0; - } - - return 0; -} - -static int proc_vfs_stats_show(struct seq_file *m, void *v) -{ - struct coda_vfs_stats * ps = & coda_vfs_stat; - - seq_printf(m, - "Coda VFS statistics\n" - "===================\n\n" - "File Operations:\n" - "\topen\t\t%9d\n" - "\tflush\t\t%9d\n" - "\trelease\t\t%9d\n" - "\tfsync\t\t%9d\n\n" - "Dir Operations:\n" - "\treaddir\t\t%9d\n\n" - "Inode Operations\n" - "\tcreate\t\t%9d\n" - "\tlookup\t\t%9d\n" - "\tlink\t\t%9d\n" - "\tunlink\t\t%9d\n" - "\tsymlink\t\t%9d\n" - "\tmkdir\t\t%9d\n" - "\trmdir\t\t%9d\n" - "\trename\t\t%9d\n" - "\tpermission\t%9d\n", - - /* file operations */ - ps->open, - ps->flush, - ps->release, - ps->fsync, - - /* dir operations */ - ps->readdir, - - /* inode operations */ - ps->create, - ps->lookup, - ps->link, - ps->unlink, - ps->symlink, - ps->mkdir, - ps->rmdir, - ps->rename, - ps->permission); - return 0; -} - -static int proc_cache_inv_stats_show(struct seq_file *m, void *v) -{ - struct coda_cache_inv_stats * ps = & coda_cache_inv_stat; - - seq_printf(m, - "Coda cache invalidation statistics\n" - "==================================\n\n" - "flush\t\t%9d\n" - "purge user\t%9d\n" - "zap_dir\t\t%9d\n" - "zap_file\t%9d\n" - "zap_vnode\t%9d\n" - "purge_fid\t%9d\n" - "replace\t\t%9d\n", - ps->flush, - ps->purge_user, - ps->zap_dir, - ps->zap_file, - ps->zap_vnode, - ps->purge_fid, - ps->replace ); - return 0; -} - -static int proc_vfs_stats_open(struct inode *inode, struct file *file) -{ - return single_open(file, proc_vfs_stats_show, NULL); -} - -static int proc_cache_inv_stats_open(struct inode *inode, struct file *file) -{ - return single_open(file, proc_cache_inv_stats_show, NULL); -} - -static const struct file_operations proc_vfs_stats_fops = { - .owner = THIS_MODULE, - .open = proc_vfs_stats_open, - .read = seq_read, - .llseek = seq_lseek, - .release = single_release, -}; - -static const struct file_operations proc_cache_inv_stats_fops = { - .owner = THIS_MODULE, - .open = proc_cache_inv_stats_open, - .read = seq_read, - .llseek = seq_lseek, - .release = single_release, -}; - static ctl_table coda_table[] = { { .ctl_name = CTL_UNNUMBERED, @@ -199,22 +32,6 @@ static ctl_table coda_table[] = { }, { .ctl_name = CTL_UNNUMBERED, - .procname = "vfs_stats", - .data = NULL, - .maxlen = 0, - .mode = 0644, - .proc_handler = &do_reset_coda_vfs_stats - }, - { - .ctl_name = CTL_UNNUMBERED, - .procname = "cache_inv_stats", - .data = NULL, - .maxlen = 0, - .mode = 0644, - .proc_handler = &do_reset_coda_cache_inv_stats - }, - { - .ctl_name = CTL_UNNUMBERED, .procname = "fake_statfs", .data = &coda_fake_statfs, .maxlen = sizeof(int), @@ -235,59 +52,20 @@ static ctl_table fs_table[] = { }; -#ifdef CONFIG_PROC_FS - -/* - target directory structure: - /proc/fs (see linux/fs/proc/root.c) - /proc/fs/coda - /proc/fs/coda/{vfs_stats, - -*/ - -static struct proc_dir_entry* proc_fs_coda; - -#endif - void coda_sysctl_init(void) { - reset_coda_vfs_stats(); - reset_coda_cache_inv_stats(); - -#ifdef CONFIG_PROC_FS - proc_fs_coda = proc_mkdir("coda", proc_root_fs); - if (proc_fs_coda) { - struct proc_dir_entry *pde; - - proc_fs_coda->owner = THIS_MODULE; - pde = create_proc_entry("vfs_stats", 0, proc_fs_coda); - if (pde) - pde->proc_fops = &proc_vfs_stats_fops; - pde = create_proc_entry("cache_inv_stats", 0, proc_fs_coda); - if (pde) - pde->proc_fops = &proc_cache_inv_stats_fops; - } -#endif - #ifdef CONFIG_SYSCTL if ( !fs_table_header ) fs_table_header = register_sysctl_table(fs_table); -#endif +#endif } -void coda_sysctl_clean(void) +void coda_sysctl_clean(void) { - #ifdef CONFIG_SYSCTL if ( fs_table_header ) { unregister_sysctl_table(fs_table_header); fs_table_header = NULL; } #endif - -#ifdef CONFIG_PROC_FS - remove_proc_entry("cache_inv_stats", proc_fs_coda); - remove_proc_entry("vfs_stats", proc_fs_coda); - remove_proc_entry("coda", proc_root_fs); -#endif } diff --git a/fs/coda/upcall.c b/fs/coda/upcall.c index 5faacdb1a47..e4e766e5557 100644 --- a/fs/coda/upcall.c +++ b/fs/coda/upcall.c @@ -35,12 +35,10 @@ #include <linux/coda_psdev.h> #include <linux/coda_fs_i.h> #include <linux/coda_cache.h> -#include <linux/coda_proc.h> -#define upc_alloc() kmalloc(sizeof(struct upc_req), GFP_KERNEL) -#define upc_free(r) kfree(r) +#include "coda_int.h" -static int coda_upcall(struct coda_sb_info *mntinfo, int inSize, int *outSize, +static int coda_upcall(struct venus_comm *vc, int inSize, int *outSize, union inputArgs *buffer); static void *alloc_upcall(int opcode, int size) @@ -86,13 +84,9 @@ int venus_rootfid(struct super_block *sb, struct CodaFid *fidp) insize = SIZE(root); UPARG(CODA_ROOT); - error = coda_upcall(coda_sbp(sb), insize, &outsize, inp); - - if (error) { - printk("coda_get_rootfid: error %d\n", error); - } else { + error = coda_upcall(coda_vcp(sb), insize, &outsize, inp); + if (!error) *fidp = outp->coda_root.VFid; - } CODA_FREE(inp, insize); return error; @@ -109,9 +103,9 @@ int venus_getattr(struct super_block *sb, struct CodaFid *fid, UPARG(CODA_GETATTR); inp->coda_getattr.VFid = *fid; - error = coda_upcall(coda_sbp(sb), insize, &outsize, inp); - - *attr = outp->coda_getattr.attr; + error = coda_upcall(coda_vcp(sb), insize, &outsize, inp); + if (!error) + *attr = outp->coda_getattr.attr; CODA_FREE(inp, insize); return error; @@ -130,7 +124,7 @@ int venus_setattr(struct super_block *sb, struct CodaFid *fid, inp->coda_setattr.VFid = *fid; inp->coda_setattr.attr = *vattr; - error = coda_upcall(coda_sbp(sb), insize, &outsize, inp); + error = coda_upcall(coda_vcp(sb), insize, &outsize, inp); CODA_FREE(inp, insize); return error; @@ -156,10 +150,11 @@ int venus_lookup(struct super_block *sb, struct CodaFid *fid, memcpy((char *)(inp) + offset, name, length); *((char *)inp + offset + length) = '\0'; - error = coda_upcall(coda_sbp(sb), insize, &outsize, inp); - - *resfid = outp->coda_lookup.VFid; - *type = outp->coda_lookup.vtype; + error = coda_upcall(coda_vcp(sb), insize, &outsize, inp); + if (!error) { + *resfid = outp->coda_lookup.VFid; + *type = outp->coda_lookup.vtype; + } CODA_FREE(inp, insize); return error; @@ -188,7 +183,7 @@ int venus_store(struct super_block *sb, struct CodaFid *fid, int flags, inp->coda_store.VFid = *fid; inp->coda_store.flags = flags; - error = coda_upcall(coda_sbp(sb), insize, &outsize, inp); + error = coda_upcall(coda_vcp(sb), insize, &outsize, inp); CODA_FREE(inp, insize); return error; @@ -206,7 +201,7 @@ int venus_release(struct super_block *sb, struct CodaFid *fid, int flags) inp->coda_release.VFid = *fid; inp->coda_release.flags = flags; - error = coda_upcall(coda_sbp(sb), insize, &outsize, inp); + error = coda_upcall(coda_vcp(sb), insize, &outsize, inp); CODA_FREE(inp, insize); return error; @@ -235,7 +230,7 @@ int venus_close(struct super_block *sb, struct CodaFid *fid, int flags, inp->coda_close.VFid = *fid; inp->coda_close.flags = flags; - error = coda_upcall(coda_sbp(sb), insize, &outsize, inp); + error = coda_upcall(coda_vcp(sb), insize, &outsize, inp); CODA_FREE(inp, insize); return error; @@ -251,12 +246,12 @@ int venus_open(struct super_block *sb, struct CodaFid *fid, insize = SIZE(open_by_fd); UPARG(CODA_OPEN_BY_FD); - inp->coda_open.VFid = *fid; - inp->coda_open.flags = flags; + inp->coda_open_by_fd.VFid = *fid; + inp->coda_open_by_fd.flags = flags; - error = coda_upcall(coda_sbp(sb), insize, &outsize, inp); - - *fh = outp->coda_open_by_fd.fh; + error = coda_upcall(coda_vcp(sb), insize, &outsize, inp); + if (!error) + *fh = outp->coda_open_by_fd.fh; CODA_FREE(inp, insize); return error; @@ -281,11 +276,12 @@ int venus_mkdir(struct super_block *sb, struct CodaFid *dirfid, /* Venus must get null terminated string */ memcpy((char *)(inp) + offset, name, length); *((char *)inp + offset + length) = '\0'; - - error = coda_upcall(coda_sbp(sb), insize, &outsize, inp); - *attrs = outp->coda_mkdir.attr; - *newfid = outp->coda_mkdir.VFid; + error = coda_upcall(coda_vcp(sb), insize, &outsize, inp); + if (!error) { + *attrs = outp->coda_mkdir.attr; + *newfid = outp->coda_mkdir.VFid; + } CODA_FREE(inp, insize); return error; @@ -323,7 +319,7 @@ int venus_rename(struct super_block *sb, struct CodaFid *old_fid, memcpy((char *)(inp) + offset, new_name, new_length); *((char *)inp + offset + new_length) = '\0'; - error = coda_upcall(coda_sbp(sb), insize, &outsize, inp); + error = coda_upcall(coda_vcp(sb), insize, &outsize, inp); CODA_FREE(inp, insize); return error; @@ -351,11 +347,12 @@ int venus_create(struct super_block *sb, struct CodaFid *dirfid, /* Venus must get null terminated string */ memcpy((char *)(inp) + offset, name, length); *((char *)inp + offset + length) = '\0'; - - error = coda_upcall(coda_sbp(sb), insize, &outsize, inp); - *attrs = outp->coda_create.attr; - *newfid = outp->coda_create.VFid; + error = coda_upcall(coda_vcp(sb), insize, &outsize, inp); + if (!error) { + *attrs = outp->coda_create.attr; + *newfid = outp->coda_create.VFid; + } CODA_FREE(inp, insize); return error; @@ -377,8 +374,8 @@ int venus_rmdir(struct super_block *sb, struct CodaFid *dirfid, inp->coda_rmdir.name = offset; memcpy((char *)(inp) + offset, name, length); *((char *)inp + offset + length) = '\0'; - - error = coda_upcall(coda_sbp(sb), insize, &outsize, inp); + + error = coda_upcall(coda_vcp(sb), insize, &outsize, inp); CODA_FREE(inp, insize); return error; @@ -399,8 +396,8 @@ int venus_remove(struct super_block *sb, struct CodaFid *dirfid, inp->coda_remove.name = offset; memcpy((char *)(inp) + offset, name, length); *((char *)inp + offset + length) = '\0'; - - error = coda_upcall(coda_sbp(sb), insize, &outsize, inp); + + error = coda_upcall(coda_vcp(sb), insize, &outsize, inp); CODA_FREE(inp, insize); return error; @@ -420,19 +417,18 @@ int venus_readlink(struct super_block *sb, struct CodaFid *fid, UPARG(CODA_READLINK); inp->coda_readlink.VFid = *fid; - - error = coda_upcall(coda_sbp(sb), insize, &outsize, inp); - - if (! error) { - retlen = outp->coda_readlink.count; + + error = coda_upcall(coda_vcp(sb), insize, &outsize, inp); + if (!error) { + retlen = outp->coda_readlink.count; if ( retlen > *length ) - retlen = *length; + retlen = *length; *length = retlen; result = (char *)outp + (long)outp->coda_readlink.data; memcpy(buffer, result, retlen); *(buffer + retlen) = '\0'; } - + CODA_FREE(inp, insize); return error; } @@ -458,8 +454,8 @@ int venus_link(struct super_block *sb, struct CodaFid *fid, /* make sure strings are null terminated */ memcpy((char *)(inp) + offset, name, len); *((char *)inp + offset + len) = '\0'; - - error = coda_upcall(coda_sbp(sb), insize, &outsize, inp); + + error = coda_upcall(coda_vcp(sb), insize, &outsize, inp); CODA_FREE(inp, insize); return error; @@ -494,7 +490,7 @@ int venus_symlink(struct super_block *sb, struct CodaFid *fid, memcpy((char *)(inp) + offset, name, len); *((char *)inp + offset + len) = '\0'; - error = coda_upcall(coda_sbp(sb), insize, &outsize, inp); + error = coda_upcall(coda_vcp(sb), insize, &outsize, inp); CODA_FREE(inp, insize); return error; @@ -509,9 +505,9 @@ int venus_fsync(struct super_block *sb, struct CodaFid *fid) insize=SIZE(fsync); UPARG(CODA_FSYNC); - inp->coda_fsync.VFid = *fid; - error = coda_upcall(coda_sbp(sb), sizeof(union inputArgs), - &outsize, inp); + inp->coda_fsync.VFid = *fid; + error = coda_upcall(coda_vcp(sb), sizeof(union inputArgs), + &outsize, inp); CODA_FREE(inp, insize); return error; @@ -529,7 +525,7 @@ int venus_access(struct super_block *sb, struct CodaFid *fid, int mask) inp->coda_access.VFid = *fid; inp->coda_access.flags = mask; - error = coda_upcall(coda_sbp(sb), insize, &outsize, inp); + error = coda_upcall(coda_vcp(sb), insize, &outsize, inp); CODA_FREE(inp, insize); return error; @@ -578,9 +574,9 @@ int venus_pioctl(struct super_block *sb, struct CodaFid *fid, goto exit; } - error = coda_upcall(coda_sbp(sb), SIZE(ioctl) + data->vi.in_size, - &outsize, inp); - + error = coda_upcall(coda_vcp(sb), SIZE(ioctl) + data->vi.in_size, + &outsize, inp); + if (error) { printk("coda_pioctl: Venus returns: %d for %s\n", error, coda_f2s(fid)); @@ -620,16 +616,13 @@ int venus_statfs(struct dentry *dentry, struct kstatfs *sfs) insize = max_t(unsigned int, INSIZE(statfs), OUTSIZE(statfs)); UPARG(CODA_STATFS); - error = coda_upcall(coda_sbp(dentry->d_sb), insize, &outsize, inp); - - if (!error) { + error = coda_upcall(coda_vcp(dentry->d_sb), insize, &outsize, inp); + if (!error) { sfs->f_blocks = outp->coda_statfs.stat.f_blocks; sfs->f_bfree = outp->coda_statfs.stat.f_bfree; sfs->f_bavail = outp->coda_statfs.stat.f_bavail; sfs->f_files = outp->coda_statfs.stat.f_files; sfs->f_ffree = outp->coda_statfs.stat.f_ffree; - } else { - printk("coda_statfs: Venus returns: %d\n", error); } CODA_FREE(inp, insize); @@ -638,96 +631,129 @@ int venus_statfs(struct dentry *dentry, struct kstatfs *sfs) /* * coda_upcall and coda_downcall routines. - * */ +static void block_signals(sigset_t *old) +{ + spin_lock_irq(¤t->sighand->siglock); + *old = current->blocked; + + sigfillset(¤t->blocked); + sigdelset(¤t->blocked, SIGKILL); + sigdelset(¤t->blocked, SIGSTOP); + sigdelset(¤t->blocked, SIGINT); + + recalc_sigpending(); + spin_unlock_irq(¤t->sighand->siglock); +} + +static void unblock_signals(sigset_t *old) +{ + spin_lock_irq(¤t->sighand->siglock); + current->blocked = *old; + recalc_sigpending(); + spin_unlock_irq(¤t->sighand->siglock); +} + +/* Don't allow signals to interrupt the following upcalls before venus + * has seen them, + * - CODA_CLOSE or CODA_RELEASE upcall (to avoid reference count problems) + * - CODA_STORE (to avoid data loss) + */ +#define CODA_INTERRUPTIBLE(r) (!coda_hard && \ + (((r)->uc_opcode != CODA_CLOSE && \ + (r)->uc_opcode != CODA_STORE && \ + (r)->uc_opcode != CODA_RELEASE) || \ + (r)->uc_flags & REQ_READ)) -static inline void coda_waitfor_upcall(struct upc_req *vmp, - struct venus_comm *vcommp) +static inline void coda_waitfor_upcall(struct upc_req *req) { DECLARE_WAITQUEUE(wait, current); + unsigned long timeout = jiffies + coda_timeout * HZ; + sigset_t old; + int blocked; - vmp->uc_posttime = jiffies; + block_signals(&old); + blocked = 1; - add_wait_queue(&vmp->uc_sleep, &wait); + add_wait_queue(&req->uc_sleep, &wait); for (;;) { - if ( !coda_hard && vmp->uc_opcode != CODA_CLOSE ) + if (CODA_INTERRUPTIBLE(req)) set_current_state(TASK_INTERRUPTIBLE); else set_current_state(TASK_UNINTERRUPTIBLE); - /* venus died */ - if ( !vcommp->vc_inuse ) - break; - /* got a reply */ - if ( vmp->uc_flags & ( REQ_WRITE | REQ_ABORT ) ) + if (req->uc_flags & (REQ_WRITE | REQ_ABORT)) break; - if ( !coda_hard && vmp->uc_opcode != CODA_CLOSE && signal_pending(current) ) { - /* if this process really wants to die, let it go */ - if ( sigismember(&(current->pending.signal), SIGKILL) || - sigismember(&(current->pending.signal), SIGINT) ) - break; - /* signal is present: after timeout always return - really smart idea, probably useless ... */ - if ( jiffies - vmp->uc_posttime > coda_timeout * HZ ) - break; + if (blocked && time_after(jiffies, timeout) && + CODA_INTERRUPTIBLE(req)) + { + unblock_signals(&old); + blocked = 0; + } + + if (signal_pending(current)) { + list_del(&req->uc_chain); + break; } - schedule(); + + if (blocked) + schedule_timeout(HZ); + else + schedule(); } - remove_wait_queue(&vmp->uc_sleep, &wait); - set_current_state(TASK_RUNNING); + if (blocked) + unblock_signals(&old); - return; + remove_wait_queue(&req->uc_sleep, &wait); + set_current_state(TASK_RUNNING); } -/* - * coda_upcall will return an error in the case of +/* + * coda_upcall will return an error in the case of * failed communication with Venus _or_ will peek at Venus * reply and return Venus' error. * * As venus has 2 types of errors, normal errors (positive) and internal * errors (negative), normal errors are negated, while internal errors * are all mapped to -EINTR, while showing a nice warning message. (jh) - * */ -static int coda_upcall(struct coda_sb_info *sbi, - int inSize, int *outSize, - union inputArgs *buffer) +static int coda_upcall(struct venus_comm *vcp, + int inSize, int *outSize, + union inputArgs *buffer) { - struct venus_comm *vcommp; union outputArgs *out; - struct upc_req *req; + union inputArgs *sig_inputArgs; + struct upc_req *req, *sig_req; int error = 0; - vcommp = sbi->sbi_vcomm; - if ( !vcommp->vc_inuse ) { - printk("No pseudo device in upcall comms at %p\n", vcommp); - return -ENXIO; + if (!vcp->vc_inuse) { + printk(KERN_NOTICE "coda: Venus dead, not sending upcall\n"); + return -ENXIO; } /* Format the request message. */ - req = upc_alloc(); - if (!req) { - printk("Failed to allocate upc_req structure\n"); + req = kmalloc(sizeof(struct upc_req), GFP_KERNEL); + if (!req) return -ENOMEM; - } + req->uc_data = (void *)buffer; req->uc_flags = 0; req->uc_inSize = inSize; req->uc_outSize = *outSize ? *outSize : inSize; req->uc_opcode = ((union inputArgs *)buffer)->ih.opcode; - req->uc_unique = ++vcommp->vc_seq; + req->uc_unique = ++vcp->vc_seq; init_waitqueue_head(&req->uc_sleep); - + /* Fill in the common input args. */ ((union inputArgs *)buffer)->ih.unique = req->uc_unique; /* Append msg to pending queue and poke Venus. */ - list_add_tail(&(req->uc_chain), &vcommp->vc_pending); - - wake_up_interruptible(&vcommp->vc_waitq); + list_add_tail(&req->uc_chain, &vcp->vc_pending); + + wake_up_interruptible(&vcp->vc_waitq); /* We can be interrupted while we wait for Venus to process * our request. If the interrupt occurs before Venus has read * the request, we dequeue and return. If it occurs after the @@ -738,67 +764,60 @@ static int coda_upcall(struct coda_sb_info *sbi, * ENODEV. */ /* Go to sleep. Wake up on signals only after the timeout. */ - coda_waitfor_upcall(req, vcommp); + coda_waitfor_upcall(req); - if (vcommp->vc_inuse) { /* i.e. Venus is still alive */ - /* Op went through, interrupt or not... */ - if (req->uc_flags & REQ_WRITE) { + /* Op went through, interrupt or not... */ + if (req->uc_flags & REQ_WRITE) { out = (union outputArgs *)req->uc_data; /* here we map positive Venus errors to kernel errors */ error = -out->oh.result; *outSize = req->uc_outSize; goto exit; - } - if ( !(req->uc_flags & REQ_READ) && signal_pending(current)) { - /* Interrupted before venus read it. */ - list_del(&(req->uc_chain)); - /* perhaps the best way to convince the app to - give up? */ - error = -EINTR; + } + + error = -EINTR; + if ((req->uc_flags & REQ_ABORT) || !signal_pending(current)) { + printk(KERN_WARNING "coda: Unexpected interruption.\n"); goto exit; - } - if ( (req->uc_flags & REQ_READ) && signal_pending(current) ) { - /* interrupted after Venus did its read, send signal */ - union inputArgs *sig_inputArgs; - struct upc_req *sig_req; - - list_del(&(req->uc_chain)); - error = -ENOMEM; - sig_req = upc_alloc(); - if (!sig_req) goto exit; - - CODA_ALLOC((sig_req->uc_data), char *, sizeof(struct coda_in_hdr)); - if (!sig_req->uc_data) { - upc_free(sig_req); - goto exit; - } - - error = -EINTR; - sig_inputArgs = (union inputArgs *)sig_req->uc_data; - sig_inputArgs->ih.opcode = CODA_SIGNAL; - sig_inputArgs->ih.unique = req->uc_unique; - - sig_req->uc_flags = REQ_ASYNC; - sig_req->uc_opcode = sig_inputArgs->ih.opcode; - sig_req->uc_unique = sig_inputArgs->ih.unique; - sig_req->uc_inSize = sizeof(struct coda_in_hdr); - sig_req->uc_outSize = sizeof(struct coda_in_hdr); - - /* insert at head of queue! */ - list_add(&(sig_req->uc_chain), &vcommp->vc_pending); - wake_up_interruptible(&vcommp->vc_waitq); - } else { - printk("Coda: Strange interruption..\n"); - error = -EINTR; - } - } else { /* If venus died i.e. !VC_OPEN(vcommp) */ - printk("coda_upcall: Venus dead on (op,un) (%d.%d) flags %d\n", - req->uc_opcode, req->uc_unique, req->uc_flags); - error = -ENODEV; } - exit: - upc_free(req); + /* Interrupted before venus read it. */ + if (!(req->uc_flags & REQ_READ)) + goto exit; + + /* Venus saw the upcall, make sure we can send interrupt signal */ + if (!vcp->vc_inuse) { + printk(KERN_INFO "coda: Venus dead, not sending signal.\n"); + goto exit; + } + + error = -ENOMEM; + sig_req = kmalloc(sizeof(struct upc_req), GFP_KERNEL); + if (!sig_req) goto exit; + + CODA_ALLOC((sig_req->uc_data), char *, sizeof(struct coda_in_hdr)); + if (!sig_req->uc_data) { + kfree(sig_req); + goto exit; + } + + error = -EINTR; + sig_inputArgs = (union inputArgs *)sig_req->uc_data; + sig_inputArgs->ih.opcode = CODA_SIGNAL; + sig_inputArgs->ih.unique = req->uc_unique; + + sig_req->uc_flags = REQ_ASYNC; + sig_req->uc_opcode = sig_inputArgs->ih.opcode; + sig_req->uc_unique = sig_inputArgs->ih.unique; + sig_req->uc_inSize = sizeof(struct coda_in_hdr); + sig_req->uc_outSize = sizeof(struct coda_in_hdr); + + /* insert at head of queue! */ + list_add(&(sig_req->uc_chain), &vcp->vc_pending); + wake_up_interruptible(&vcp->vc_waitq); + +exit: + kfree(req); return error; } @@ -838,77 +857,66 @@ static int coda_upcall(struct coda_sb_info *sbi, int coda_downcall(int opcode, union outputArgs * out, struct super_block *sb) { + struct inode *inode = NULL; + struct CodaFid *fid, *newfid; + /* Handle invalidation requests. */ - if ( !sb || !sb->s_root || !sb->s_root->d_inode) - return 0; - - switch (opcode) { - - case CODA_FLUSH : { - coda_cache_clear_all(sb); - shrink_dcache_sb(sb); - coda_flag_inode(sb->s_root->d_inode, C_FLUSH); - return(0); - } - - case CODA_PURGEUSER : { - coda_cache_clear_all(sb); - return(0); - } - - case CODA_ZAPDIR : { - struct inode *inode; - struct CodaFid *fid = &out->coda_zapdir.CodaFid; - - inode = coda_fid_to_inode(fid, sb); - if (inode) { - coda_flag_inode_children(inode, C_PURGE); - coda_flag_inode(inode, C_VATTR); - iput(inode); - } - - return(0); - } - - case CODA_ZAPFILE : { - struct inode *inode; - struct CodaFid *fid = &out->coda_zapfile.CodaFid; - inode = coda_fid_to_inode(fid, sb); - if ( inode ) { - coda_flag_inode(inode, C_VATTR); - iput(inode); - } - return 0; - } - - case CODA_PURGEFID : { - struct inode *inode; - struct CodaFid *fid = &out->coda_purgefid.CodaFid; - inode = coda_fid_to_inode(fid, sb); - if ( inode ) { + if ( !sb || !sb->s_root) + return 0; + + switch (opcode) { + case CODA_FLUSH: + coda_cache_clear_all(sb); + shrink_dcache_sb(sb); + if (sb->s_root->d_inode) + coda_flag_inode(sb->s_root->d_inode, C_FLUSH); + break; + + case CODA_PURGEUSER: + coda_cache_clear_all(sb); + break; + + case CODA_ZAPDIR: + fid = &out->coda_zapdir.CodaFid; + inode = coda_fid_to_inode(fid, sb); + if (inode) { + coda_flag_inode_children(inode, C_PURGE); + coda_flag_inode(inode, C_VATTR); + } + break; + + case CODA_ZAPFILE: + fid = &out->coda_zapfile.CodaFid; + inode = coda_fid_to_inode(fid, sb); + if (inode) + coda_flag_inode(inode, C_VATTR); + break; + + case CODA_PURGEFID: + fid = &out->coda_purgefid.CodaFid; + inode = coda_fid_to_inode(fid, sb); + if (inode) { coda_flag_inode_children(inode, C_PURGE); /* catch the dentries later if some are still busy */ coda_flag_inode(inode, C_PURGE); d_prune_aliases(inode); - iput(inode); - } - return 0; - } - - case CODA_REPLACE : { - struct inode *inode; - struct CodaFid *oldfid = &out->coda_replace.OldFid; - struct CodaFid *newfid = &out->coda_replace.NewFid; - inode = coda_fid_to_inode(oldfid, sb); - if ( inode ) { - coda_replace_fid(inode, oldfid, newfid); - iput(inode); - } - return 0; - } - } - return 0; + } + break; + + case CODA_REPLACE: + fid = &out->coda_replace.OldFid; + newfid = &out->coda_replace.NewFid; + inode = coda_fid_to_inode(fid, sb); + if (inode) + coda_replace_fid(inode, fid, newfid); + break; + } + + if (inode) + iput(inode); + + return 0; } diff --git a/fs/compat.c b/fs/compat.c index 4db6216e526..15078ce4c04 100644 --- a/fs/compat.c +++ b/fs/compat.c @@ -1257,6 +1257,7 @@ static int compat_copy_strings(int argc, compat_uptr_t __user *argv, { struct page *kmapped_page = NULL; char *kaddr = NULL; + unsigned long kpos = 0; int ret; while (argc-- > 0) { @@ -1265,92 +1266,84 @@ static int compat_copy_strings(int argc, compat_uptr_t __user *argv, unsigned long pos; if (get_user(str, argv+argc) || - !(len = strnlen_user(compat_ptr(str), bprm->p))) { + !(len = strnlen_user(compat_ptr(str), MAX_ARG_STRLEN))) { ret = -EFAULT; goto out; } - if (bprm->p < len) { + if (len > MAX_ARG_STRLEN) { ret = -E2BIG; goto out; } - bprm->p -= len; - /* XXX: add architecture specific overflow check here. */ + /* We're going to work our way backwords. */ pos = bprm->p; + str += len; + bprm->p -= len; while (len > 0) { - int i, new, err; int offset, bytes_to_copy; - struct page *page; offset = pos % PAGE_SIZE; - i = pos/PAGE_SIZE; - page = bprm->page[i]; - new = 0; - if (!page) { - page = alloc_page(GFP_HIGHUSER); - bprm->page[i] = page; - if (!page) { - ret = -ENOMEM; + if (offset == 0) + offset = PAGE_SIZE; + + bytes_to_copy = offset; + if (bytes_to_copy > len) + bytes_to_copy = len; + + offset -= bytes_to_copy; + pos -= bytes_to_copy; + str -= bytes_to_copy; + len -= bytes_to_copy; + + if (!kmapped_page || kpos != (pos & PAGE_MASK)) { + struct page *page; + +#ifdef CONFIG_STACK_GROWSUP + ret = expand_stack_downwards(bprm->vma, pos); + if (ret < 0) { + /* We've exceed the stack rlimit. */ + ret = -E2BIG; + goto out; + } +#endif + ret = get_user_pages(current, bprm->mm, pos, + 1, 1, 1, &page, NULL); + if (ret <= 0) { + /* We've exceed the stack rlimit. */ + ret = -E2BIG; goto out; } - new = 1; - } - if (page != kmapped_page) { - if (kmapped_page) + if (kmapped_page) { + flush_kernel_dcache_page(kmapped_page); kunmap(kmapped_page); + put_page(kmapped_page); + } kmapped_page = page; kaddr = kmap(kmapped_page); + kpos = pos & PAGE_MASK; + flush_cache_page(bprm->vma, kpos, + page_to_pfn(kmapped_page)); } - if (new && offset) - memset(kaddr, 0, offset); - bytes_to_copy = PAGE_SIZE - offset; - if (bytes_to_copy > len) { - bytes_to_copy = len; - if (new) - memset(kaddr+offset+len, 0, - PAGE_SIZE-offset-len); - } - err = copy_from_user(kaddr+offset, compat_ptr(str), - bytes_to_copy); - if (err) { + if (copy_from_user(kaddr+offset, compat_ptr(str), + bytes_to_copy)) { ret = -EFAULT; goto out; } - - pos += bytes_to_copy; - str += bytes_to_copy; - len -= bytes_to_copy; } } ret = 0; out: - if (kmapped_page) + if (kmapped_page) { + flush_kernel_dcache_page(kmapped_page); kunmap(kmapped_page); - return ret; -} - -#ifdef CONFIG_MMU - -#define free_arg_pages(bprm) do { } while (0) - -#else - -static inline void free_arg_pages(struct linux_binprm *bprm) -{ - int i; - - for (i = 0; i < MAX_ARG_PAGES; i++) { - if (bprm->page[i]) - __free_page(bprm->page[i]); - bprm->page[i] = NULL; + put_page(kmapped_page); } + return ret; } -#endif /* CONFIG_MMU */ - /* * compat_do_execve() is mostly a copy of do_execve(), with the exception * that it processes 32 bit argv and envp pointers. @@ -1363,7 +1356,6 @@ int compat_do_execve(char * filename, struct linux_binprm *bprm; struct file *file; int retval; - int i; retval = -ENOMEM; bprm = kzalloc(sizeof(*bprm), GFP_KERNEL); @@ -1377,24 +1369,19 @@ int compat_do_execve(char * filename, sched_exec(); - bprm->p = PAGE_SIZE*MAX_ARG_PAGES-sizeof(void *); bprm->file = file; bprm->filename = filename; bprm->interp = filename; - bprm->mm = mm_alloc(); - retval = -ENOMEM; - if (!bprm->mm) - goto out_file; - retval = init_new_context(current, bprm->mm); - if (retval < 0) - goto out_mm; + retval = bprm_mm_init(bprm); + if (retval) + goto out_file; - bprm->argc = compat_count(argv, bprm->p / sizeof(compat_uptr_t)); + bprm->argc = compat_count(argv, MAX_ARG_STRINGS); if ((retval = bprm->argc) < 0) goto out_mm; - bprm->envc = compat_count(envp, bprm->p / sizeof(compat_uptr_t)); + bprm->envc = compat_count(envp, MAX_ARG_STRINGS); if ((retval = bprm->envc) < 0) goto out_mm; @@ -1421,8 +1408,6 @@ int compat_do_execve(char * filename, retval = search_binary_handler(bprm, regs); if (retval >= 0) { - free_arg_pages(bprm); - /* execve success */ security_bprm_free(bprm); acct_update_integrals(current); @@ -1431,19 +1416,12 @@ int compat_do_execve(char * filename, } out: - /* Something went wrong, return the inode and free the argument pages*/ - for (i = 0 ; i < MAX_ARG_PAGES ; i++) { - struct page * page = bprm->page[i]; - if (page) - __free_page(page); - } - if (bprm->security) security_bprm_free(bprm); out_mm: if (bprm->mm) - mmdrop(bprm->mm); + mmput(bprm->mm); out_file: if (bprm->file) { diff --git a/fs/dlm/memory.c b/fs/dlm/memory.c index f858fef6e41..fb9e2ee998a 100644 --- a/fs/dlm/memory.c +++ b/fs/dlm/memory.c @@ -39,9 +39,7 @@ char *allocate_lvb(struct dlm_ls *ls) { char *p; - p = kmalloc(ls->ls_lvblen, GFP_KERNEL); - if (p) - memset(p, 0, ls->ls_lvblen); + p = kzalloc(ls->ls_lvblen, GFP_KERNEL); return p; } @@ -59,9 +57,7 @@ struct dlm_rsb *allocate_rsb(struct dlm_ls *ls, int namelen) DLM_ASSERT(namelen <= DLM_RESNAME_MAXLEN,); - r = kmalloc(sizeof(*r) + namelen, GFP_KERNEL); - if (r) - memset(r, 0, sizeof(*r) + namelen); + r = kzalloc(sizeof(*r) + namelen, GFP_KERNEL); return r; } @@ -101,9 +97,7 @@ struct dlm_direntry *allocate_direntry(struct dlm_ls *ls, int namelen) DLM_ASSERT(namelen <= DLM_RESNAME_MAXLEN, printk("namelen = %d\n", namelen);); - de = kmalloc(sizeof(*de) + namelen, GFP_KERNEL); - if (de) - memset(de, 0, sizeof(*de) + namelen); + de = kzalloc(sizeof(*de) + namelen, GFP_KERNEL); return de; } diff --git a/fs/ecryptfs/inode.c b/fs/ecryptfs/inode.c index e77a2ec71aa..0a50942b437 100644 --- a/fs/ecryptfs/inode.c +++ b/fs/ecryptfs/inode.c @@ -902,8 +902,9 @@ static int ecryptfs_setattr(struct dentry *dentry, struct iattr *ia) mutex_lock(&crypt_stat->cs_mutex); if (S_ISDIR(dentry->d_inode->i_mode)) crypt_stat->flags &= ~(ECRYPTFS_ENCRYPTED); - else if (!(crypt_stat->flags & ECRYPTFS_POLICY_APPLIED) - || !(crypt_stat->flags & ECRYPTFS_KEY_VALID)) { + else if (S_ISREG(dentry->d_inode->i_mode) + && (!(crypt_stat->flags & ECRYPTFS_POLICY_APPLIED) + || !(crypt_stat->flags & ECRYPTFS_KEY_VALID))) { struct vfsmount *lower_mnt; struct file *lower_file = NULL; struct ecryptfs_mount_crypt_stat *mount_crypt_stat; diff --git a/fs/exec.c b/fs/exec.c index f20561ff452..7bdea7937ee 100644 --- a/fs/exec.c +++ b/fs/exec.c @@ -54,6 +54,7 @@ #include <asm/uaccess.h> #include <asm/mmu_context.h> +#include <asm/tlb.h> #ifdef CONFIG_KMOD #include <linux/kmod.h> @@ -178,6 +179,207 @@ exit: goto out; } +#ifdef CONFIG_MMU + +static struct page *get_arg_page(struct linux_binprm *bprm, unsigned long pos, + int write) +{ + struct page *page; + int ret; + +#ifdef CONFIG_STACK_GROWSUP + if (write) { + ret = expand_stack_downwards(bprm->vma, pos); + if (ret < 0) + return NULL; + } +#endif + ret = get_user_pages(current, bprm->mm, pos, + 1, write, 1, &page, NULL); + if (ret <= 0) + return NULL; + + if (write) { + struct rlimit *rlim = current->signal->rlim; + unsigned long size = bprm->vma->vm_end - bprm->vma->vm_start; + + /* + * Limit to 1/4-th the stack size for the argv+env strings. + * This ensures that: + * - the remaining binfmt code will not run out of stack space, + * - the program will have a reasonable amount of stack left + * to work from. + */ + if (size > rlim[RLIMIT_STACK].rlim_cur / 4) { + put_page(page); + return NULL; + } + } + + return page; +} + +static void put_arg_page(struct page *page) +{ + put_page(page); +} + +static void free_arg_page(struct linux_binprm *bprm, int i) +{ +} + +static void free_arg_pages(struct linux_binprm *bprm) +{ +} + +static void flush_arg_page(struct linux_binprm *bprm, unsigned long pos, + struct page *page) +{ + flush_cache_page(bprm->vma, pos, page_to_pfn(page)); +} + +static int __bprm_mm_init(struct linux_binprm *bprm) +{ + int err = -ENOMEM; + struct vm_area_struct *vma = NULL; + struct mm_struct *mm = bprm->mm; + + bprm->vma = vma = kmem_cache_zalloc(vm_area_cachep, GFP_KERNEL); + if (!vma) + goto err; + + down_write(&mm->mmap_sem); + vma->vm_mm = mm; + + /* + * Place the stack at the largest stack address the architecture + * supports. Later, we'll move this to an appropriate place. We don't + * use STACK_TOP because that can depend on attributes which aren't + * configured yet. + */ + vma->vm_end = STACK_TOP_MAX; + vma->vm_start = vma->vm_end - PAGE_SIZE; + + vma->vm_flags = VM_STACK_FLAGS; + vma->vm_page_prot = protection_map[vma->vm_flags & 0x7]; + err = insert_vm_struct(mm, vma); + if (err) { + up_write(&mm->mmap_sem); + goto err; + } + + mm->stack_vm = mm->total_vm = 1; + up_write(&mm->mmap_sem); + + bprm->p = vma->vm_end - sizeof(void *); + + return 0; + +err: + if (vma) { + bprm->vma = NULL; + kmem_cache_free(vm_area_cachep, vma); + } + + return err; +} + +static bool valid_arg_len(struct linux_binprm *bprm, long len) +{ + return len <= MAX_ARG_STRLEN; +} + +#else + +static struct page *get_arg_page(struct linux_binprm *bprm, unsigned long pos, + int write) +{ + struct page *page; + + page = bprm->page[pos / PAGE_SIZE]; + if (!page && write) { + page = alloc_page(GFP_HIGHUSER|__GFP_ZERO); + if (!page) + return NULL; + bprm->page[pos / PAGE_SIZE] = page; + } + + return page; +} + +static void put_arg_page(struct page *page) +{ +} + +static void free_arg_page(struct linux_binprm *bprm, int i) +{ + if (bprm->page[i]) { + __free_page(bprm->page[i]); + bprm->page[i] = NULL; + } +} + +static void free_arg_pages(struct linux_binprm *bprm) +{ + int i; + + for (i = 0; i < MAX_ARG_PAGES; i++) + free_arg_page(bprm, i); +} + +static void flush_arg_page(struct linux_binprm *bprm, unsigned long pos, + struct page *page) +{ +} + +static int __bprm_mm_init(struct linux_binprm *bprm) +{ + bprm->p = PAGE_SIZE * MAX_ARG_PAGES - sizeof(void *); + return 0; +} + +static bool valid_arg_len(struct linux_binprm *bprm, long len) +{ + return len <= bprm->p; +} + +#endif /* CONFIG_MMU */ + +/* + * Create a new mm_struct and populate it with a temporary stack + * vm_area_struct. We don't have enough context at this point to set the stack + * flags, permissions, and offset, so we use temporary values. We'll update + * them later in setup_arg_pages(). + */ +int bprm_mm_init(struct linux_binprm *bprm) +{ + int err; + struct mm_struct *mm = NULL; + + bprm->mm = mm = mm_alloc(); + err = -ENOMEM; + if (!mm) + goto err; + + err = init_new_context(current, mm); + if (err) + goto err; + + err = __bprm_mm_init(bprm); + if (err) + goto err; + + return 0; + +err: + if (mm) { + bprm->mm = NULL; + mmdrop(mm); + } + + return err; +} + /* * count() counts the number of strings in array ARGV. */ @@ -203,15 +405,16 @@ static int count(char __user * __user * argv, int max) } /* - * 'copy_strings()' copies argument/environment strings from user - * memory to free pages in kernel mem. These are in a format ready - * to be put directly into the top of new user memory. + * 'copy_strings()' copies argument/environment strings from the old + * processes's memory to the new process's stack. The call to get_user_pages() + * ensures the destination page is created and not swapped out. */ static int copy_strings(int argc, char __user * __user * argv, struct linux_binprm *bprm) { struct page *kmapped_page = NULL; char *kaddr = NULL; + unsigned long kpos = 0; int ret; while (argc-- > 0) { @@ -220,69 +423,69 @@ static int copy_strings(int argc, char __user * __user * argv, unsigned long pos; if (get_user(str, argv+argc) || - !(len = strnlen_user(str, bprm->p))) { + !(len = strnlen_user(str, MAX_ARG_STRLEN))) { ret = -EFAULT; goto out; } - if (bprm->p < len) { + if (!valid_arg_len(bprm, len)) { ret = -E2BIG; goto out; } - bprm->p -= len; - /* XXX: add architecture specific overflow check here. */ + /* We're going to work our way backwords. */ pos = bprm->p; + str += len; + bprm->p -= len; while (len > 0) { - int i, new, err; int offset, bytes_to_copy; - struct page *page; offset = pos % PAGE_SIZE; - i = pos/PAGE_SIZE; - page = bprm->page[i]; - new = 0; - if (!page) { - page = alloc_page(GFP_HIGHUSER); - bprm->page[i] = page; + if (offset == 0) + offset = PAGE_SIZE; + + bytes_to_copy = offset; + if (bytes_to_copy > len) + bytes_to_copy = len; + + offset -= bytes_to_copy; + pos -= bytes_to_copy; + str -= bytes_to_copy; + len -= bytes_to_copy; + + if (!kmapped_page || kpos != (pos & PAGE_MASK)) { + struct page *page; + + page = get_arg_page(bprm, pos, 1); if (!page) { - ret = -ENOMEM; + ret = -E2BIG; goto out; } - new = 1; - } - if (page != kmapped_page) { - if (kmapped_page) + if (kmapped_page) { + flush_kernel_dcache_page(kmapped_page); kunmap(kmapped_page); + put_arg_page(kmapped_page); + } kmapped_page = page; kaddr = kmap(kmapped_page); + kpos = pos & PAGE_MASK; + flush_arg_page(bprm, kpos, kmapped_page); } - if (new && offset) - memset(kaddr, 0, offset); - bytes_to_copy = PAGE_SIZE - offset; - if (bytes_to_copy > len) { - bytes_to_copy = len; - if (new) - memset(kaddr+offset+len, 0, - PAGE_SIZE-offset-len); - } - err = copy_from_user(kaddr+offset, str, bytes_to_copy); - if (err) { + if (copy_from_user(kaddr+offset, str, bytes_to_copy)) { ret = -EFAULT; goto out; } - - pos += bytes_to_copy; - str += bytes_to_copy; - len -= bytes_to_copy; } } ret = 0; out: - if (kmapped_page) + if (kmapped_page) { + flush_kernel_dcache_page(kmapped_page); kunmap(kmapped_page); + put_arg_page(kmapped_page); + } return ret; } @@ -298,181 +501,172 @@ int copy_strings_kernel(int argc,char ** argv, struct linux_binprm *bprm) set_fs(oldfs); return r; } - EXPORT_SYMBOL(copy_strings_kernel); #ifdef CONFIG_MMU + /* - * This routine is used to map in a page into an address space: needed by - * execve() for the initial stack and environment pages. + * During bprm_mm_init(), we create a temporary stack at STACK_TOP_MAX. Once + * the binfmt code determines where the new stack should reside, we shift it to + * its final location. The process proceeds as follows: * - * vma->vm_mm->mmap_sem is held for writing. + * 1) Use shift to calculate the new vma endpoints. + * 2) Extend vma to cover both the old and new ranges. This ensures the + * arguments passed to subsequent functions are consistent. + * 3) Move vma's page tables to the new range. + * 4) Free up any cleared pgd range. + * 5) Shrink the vma to cover only the new range. */ -void install_arg_page(struct vm_area_struct *vma, - struct page *page, unsigned long address) +static int shift_arg_pages(struct vm_area_struct *vma, unsigned long shift) { struct mm_struct *mm = vma->vm_mm; - pte_t * pte; - spinlock_t *ptl; + unsigned long old_start = vma->vm_start; + unsigned long old_end = vma->vm_end; + unsigned long length = old_end - old_start; + unsigned long new_start = old_start - shift; + unsigned long new_end = old_end - shift; + struct mmu_gather *tlb; - if (unlikely(anon_vma_prepare(vma))) - goto out; + BUG_ON(new_start > new_end); - flush_dcache_page(page); - pte = get_locked_pte(mm, address, &ptl); - if (!pte) - goto out; - if (!pte_none(*pte)) { - pte_unmap_unlock(pte, ptl); - goto out; + /* + * ensure there are no vmas between where we want to go + * and where we are + */ + if (vma != find_vma(mm, new_start)) + return -EFAULT; + + /* + * cover the whole range: [new_start, old_end) + */ + vma_adjust(vma, new_start, old_end, vma->vm_pgoff, NULL); + + /* + * move the page tables downwards, on failure we rely on + * process cleanup to remove whatever mess we made. + */ + if (length != move_page_tables(vma, old_start, + vma, new_start, length)) + return -ENOMEM; + + lru_add_drain(); + tlb = tlb_gather_mmu(mm, 0); + if (new_end > old_start) { + /* + * when the old and new regions overlap clear from new_end. + */ + free_pgd_range(&tlb, new_end, old_end, new_end, + vma->vm_next ? vma->vm_next->vm_start : 0); + } else { + /* + * otherwise, clean from old_start; this is done to not touch + * the address space in [new_end, old_start) some architectures + * have constraints on va-space that make this illegal (IA64) - + * for the others its just a little faster. + */ + free_pgd_range(&tlb, old_start, old_end, new_end, + vma->vm_next ? vma->vm_next->vm_start : 0); } - inc_mm_counter(mm, anon_rss); - lru_cache_add_active(page); - set_pte_at(mm, address, pte, pte_mkdirty(pte_mkwrite(mk_pte( - page, vma->vm_page_prot)))); - page_add_new_anon_rmap(page, vma, address); - pte_unmap_unlock(pte, ptl); - - /* no need for flush_tlb */ - return; -out: - __free_page(page); - force_sig(SIGKILL, current); + tlb_finish_mmu(tlb, new_end, old_end); + + /* + * shrink the vma to just the new range. + */ + vma_adjust(vma, new_start, new_end, vma->vm_pgoff, NULL); + + return 0; } #define EXTRA_STACK_VM_PAGES 20 /* random */ +/* + * Finalizes the stack vm_area_struct. The flags and permissions are updated, + * the stack is optionally relocated, and some extra space is added. + */ int setup_arg_pages(struct linux_binprm *bprm, unsigned long stack_top, int executable_stack) { - unsigned long stack_base; - struct vm_area_struct *mpnt; + unsigned long ret; + unsigned long stack_shift; struct mm_struct *mm = current->mm; - int i, ret; - long arg_size; + struct vm_area_struct *vma = bprm->vma; + struct vm_area_struct *prev = NULL; + unsigned long vm_flags; + unsigned long stack_base; #ifdef CONFIG_STACK_GROWSUP - /* Move the argument and environment strings to the bottom of the - * stack space. - */ - int offset, j; - char *to, *from; - - /* Start by shifting all the pages down */ - i = 0; - for (j = 0; j < MAX_ARG_PAGES; j++) { - struct page *page = bprm->page[j]; - if (!page) - continue; - bprm->page[i++] = page; - } - - /* Now move them within their pages */ - offset = bprm->p % PAGE_SIZE; - to = kmap(bprm->page[0]); - for (j = 1; j < i; j++) { - memmove(to, to + offset, PAGE_SIZE - offset); - from = kmap(bprm->page[j]); - memcpy(to + PAGE_SIZE - offset, from, offset); - kunmap(bprm->page[j - 1]); - to = from; - } - memmove(to, to + offset, PAGE_SIZE - offset); - kunmap(bprm->page[j - 1]); - /* Limit stack size to 1GB */ stack_base = current->signal->rlim[RLIMIT_STACK].rlim_max; if (stack_base > (1 << 30)) stack_base = 1 << 30; - stack_base = PAGE_ALIGN(stack_top - stack_base); - /* Adjust bprm->p to point to the end of the strings. */ - bprm->p = stack_base + PAGE_SIZE * i - offset; + /* Make sure we didn't let the argument array grow too large. */ + if (vma->vm_end - vma->vm_start > stack_base) + return -ENOMEM; - mm->arg_start = stack_base; - arg_size = i << PAGE_SHIFT; + stack_base = PAGE_ALIGN(stack_top - stack_base); - /* zero pages that were copied above */ - while (i < MAX_ARG_PAGES) - bprm->page[i++] = NULL; + stack_shift = vma->vm_start - stack_base; + mm->arg_start = bprm->p - stack_shift; + bprm->p = vma->vm_end - stack_shift; #else - stack_base = arch_align_stack(stack_top - MAX_ARG_PAGES*PAGE_SIZE); - stack_base = PAGE_ALIGN(stack_base); - bprm->p += stack_base; + stack_top = arch_align_stack(stack_top); + stack_top = PAGE_ALIGN(stack_top); + stack_shift = vma->vm_end - stack_top; + + bprm->p -= stack_shift; mm->arg_start = bprm->p; - arg_size = stack_top - (PAGE_MASK & (unsigned long) mm->arg_start); #endif - arg_size += EXTRA_STACK_VM_PAGES * PAGE_SIZE; - if (bprm->loader) - bprm->loader += stack_base; - bprm->exec += stack_base; - - mpnt = kmem_cache_zalloc(vm_area_cachep, GFP_KERNEL); - if (!mpnt) - return -ENOMEM; + bprm->loader -= stack_shift; + bprm->exec -= stack_shift; down_write(&mm->mmap_sem); - { - mpnt->vm_mm = mm; -#ifdef CONFIG_STACK_GROWSUP - mpnt->vm_start = stack_base; - mpnt->vm_end = stack_base + arg_size; -#else - mpnt->vm_end = stack_top; - mpnt->vm_start = mpnt->vm_end - arg_size; -#endif - /* Adjust stack execute permissions; explicitly enable - * for EXSTACK_ENABLE_X, disable for EXSTACK_DISABLE_X - * and leave alone (arch default) otherwise. */ - if (unlikely(executable_stack == EXSTACK_ENABLE_X)) - mpnt->vm_flags = VM_STACK_FLAGS | VM_EXEC; - else if (executable_stack == EXSTACK_DISABLE_X) - mpnt->vm_flags = VM_STACK_FLAGS & ~VM_EXEC; - else - mpnt->vm_flags = VM_STACK_FLAGS; - mpnt->vm_flags |= mm->def_flags; - mpnt->vm_page_prot = protection_map[mpnt->vm_flags & 0x7]; - if ((ret = insert_vm_struct(mm, mpnt))) { + vm_flags = vma->vm_flags; + + /* + * Adjust stack execute permissions; explicitly enable for + * EXSTACK_ENABLE_X, disable for EXSTACK_DISABLE_X and leave alone + * (arch default) otherwise. + */ + if (unlikely(executable_stack == EXSTACK_ENABLE_X)) + vm_flags |= VM_EXEC; + else if (executable_stack == EXSTACK_DISABLE_X) + vm_flags &= ~VM_EXEC; + vm_flags |= mm->def_flags; + + ret = mprotect_fixup(vma, &prev, vma->vm_start, vma->vm_end, + vm_flags); + if (ret) + goto out_unlock; + BUG_ON(prev != vma); + + /* Move stack pages down in memory. */ + if (stack_shift) { + ret = shift_arg_pages(vma, stack_shift); + if (ret) { up_write(&mm->mmap_sem); - kmem_cache_free(vm_area_cachep, mpnt); return ret; } - mm->stack_vm = mm->total_vm = vma_pages(mpnt); } - for (i = 0 ; i < MAX_ARG_PAGES ; i++) { - struct page *page = bprm->page[i]; - if (page) { - bprm->page[i] = NULL; - install_arg_page(mpnt, page, stack_base); - } - stack_base += PAGE_SIZE; - } +#ifdef CONFIG_STACK_GROWSUP + stack_base = vma->vm_end + EXTRA_STACK_VM_PAGES * PAGE_SIZE; +#else + stack_base = vma->vm_start - EXTRA_STACK_VM_PAGES * PAGE_SIZE; +#endif + ret = expand_stack(vma, stack_base); + if (ret) + ret = -EFAULT; + +out_unlock: up_write(&mm->mmap_sem); - return 0; } - EXPORT_SYMBOL(setup_arg_pages); -#define free_arg_pages(bprm) do { } while (0) - -#else - -static inline void free_arg_pages(struct linux_binprm *bprm) -{ - int i; - - for (i = 0; i < MAX_ARG_PAGES; i++) { - if (bprm->page[i]) - __free_page(bprm->page[i]); - bprm->page[i] = NULL; - } -} - #endif /* CONFIG_MMU */ struct file *open_exec(const char *name) @@ -864,9 +1058,9 @@ int flush_old_exec(struct linux_binprm * bprm) current->sas_ss_sp = current->sas_ss_size = 0; if (current->euid == current->uid && current->egid == current->gid) - current->mm->dumpable = 1; + set_dumpable(current->mm, 1); else - current->mm->dumpable = suid_dumpable; + set_dumpable(current->mm, suid_dumpable); name = bprm->filename; @@ -894,7 +1088,7 @@ int flush_old_exec(struct linux_binprm * bprm) file_permission(bprm->file, MAY_READ) || (bprm->interp_flags & BINPRM_FLAGS_ENFORCE_NONDUMP)) { suid_keys(current); - current->mm->dumpable = suid_dumpable; + set_dumpable(current->mm, suid_dumpable); } /* An exec changes our domain. We are no longer part of the thread @@ -1000,43 +1194,42 @@ EXPORT_SYMBOL(compute_creds); * points to; chop off the first by relocating brpm->p to right after * the first '\0' encountered. */ -void remove_arg_zero(struct linux_binprm *bprm) +int remove_arg_zero(struct linux_binprm *bprm) { - if (bprm->argc) { - char ch; + int ret = 0; + unsigned long offset; + char *kaddr; + struct page *page; - do { - unsigned long offset; - unsigned long index; - char *kaddr; - struct page *page; - - offset = bprm->p & ~PAGE_MASK; - index = bprm->p >> PAGE_SHIFT; + if (!bprm->argc) + return 0; - page = bprm->page[index]; - kaddr = kmap_atomic(page, KM_USER0); + do { + offset = bprm->p & ~PAGE_MASK; + page = get_arg_page(bprm, bprm->p, 0); + if (!page) { + ret = -EFAULT; + goto out; + } + kaddr = kmap_atomic(page, KM_USER0); - /* run through page until we reach end or find NUL */ - do { - ch = *(kaddr + offset); + for (; offset < PAGE_SIZE && kaddr[offset]; + offset++, bprm->p++) + ; - /* discard that character... */ - bprm->p++; - offset++; - } while (offset < PAGE_SIZE && ch != '\0'); + kunmap_atomic(kaddr, KM_USER0); + put_arg_page(page); - kunmap_atomic(kaddr, KM_USER0); + if (offset == PAGE_SIZE) + free_arg_page(bprm, (bprm->p >> PAGE_SHIFT) - 1); + } while (offset == PAGE_SIZE); - /* free the old page */ - if (offset == PAGE_SIZE) { - __free_page(page); - bprm->page[index] = NULL; - } - } while (ch != '\0'); + bprm->p++; + bprm->argc--; + ret = 0; - bprm->argc--; - } +out: + return ret; } EXPORT_SYMBOL(remove_arg_zero); @@ -1062,7 +1255,7 @@ int search_binary_handler(struct linux_binprm *bprm,struct pt_regs *regs) fput(bprm->file); bprm->file = NULL; - loader = PAGE_SIZE*MAX_ARG_PAGES-sizeof(void *); + loader = bprm->vma->vm_end - sizeof(void *); file = open_exec("/sbin/loader"); retval = PTR_ERR(file); @@ -1154,8 +1347,8 @@ int do_execve(char * filename, { struct linux_binprm *bprm; struct file *file; + unsigned long env_p; int retval; - int i; retval = -ENOMEM; bprm = kzalloc(sizeof(*bprm), GFP_KERNEL); @@ -1169,25 +1362,19 @@ int do_execve(char * filename, sched_exec(); - bprm->p = PAGE_SIZE*MAX_ARG_PAGES-sizeof(void *); - bprm->file = file; bprm->filename = filename; bprm->interp = filename; - bprm->mm = mm_alloc(); - retval = -ENOMEM; - if (!bprm->mm) - goto out_file; - retval = init_new_context(current, bprm->mm); - if (retval < 0) - goto out_mm; + retval = bprm_mm_init(bprm); + if (retval) + goto out_file; - bprm->argc = count(argv, bprm->p / sizeof(void *)); + bprm->argc = count(argv, MAX_ARG_STRINGS); if ((retval = bprm->argc) < 0) goto out_mm; - bprm->envc = count(envp, bprm->p / sizeof(void *)); + bprm->envc = count(envp, MAX_ARG_STRINGS); if ((retval = bprm->envc) < 0) goto out_mm; @@ -1208,15 +1395,16 @@ int do_execve(char * filename, if (retval < 0) goto out; + env_p = bprm->p; retval = copy_strings(bprm->argc, argv, bprm); if (retval < 0) goto out; + bprm->argv_len = env_p - bprm->p; retval = search_binary_handler(bprm,regs); if (retval >= 0) { - free_arg_pages(bprm); - /* execve success */ + free_arg_pages(bprm); security_bprm_free(bprm); acct_update_integrals(current); kfree(bprm); @@ -1224,26 +1412,19 @@ int do_execve(char * filename, } out: - /* Something went wrong, return the inode and free the argument pages*/ - for (i = 0 ; i < MAX_ARG_PAGES ; i++) { - struct page * page = bprm->page[i]; - if (page) - __free_page(page); - } - + free_arg_pages(bprm); if (bprm->security) security_bprm_free(bprm); out_mm: if (bprm->mm) - mmdrop(bprm->mm); + mmput (bprm->mm); out_file: if (bprm->file) { allow_write_access(bprm->file); fput(bprm->file); } - out_kfree: kfree(bprm); @@ -1484,6 +1665,56 @@ fail: return core_waiters; } +/* + * set_dumpable converts traditional three-value dumpable to two flags and + * stores them into mm->flags. It modifies lower two bits of mm->flags, but + * these bits are not changed atomically. So get_dumpable can observe the + * intermediate state. To avoid doing unexpected behavior, get get_dumpable + * return either old dumpable or new one by paying attention to the order of + * modifying the bits. + * + * dumpable | mm->flags (binary) + * old new | initial interim final + * ---------+----------------------- + * 0 1 | 00 01 01 + * 0 2 | 00 10(*) 11 + * 1 0 | 01 00 00 + * 1 2 | 01 11 11 + * 2 0 | 11 10(*) 00 + * 2 1 | 11 11 01 + * + * (*) get_dumpable regards interim value of 10 as 11. + */ +void set_dumpable(struct mm_struct *mm, int value) +{ + switch (value) { + case 0: + clear_bit(MMF_DUMPABLE, &mm->flags); + smp_wmb(); + clear_bit(MMF_DUMP_SECURELY, &mm->flags); + break; + case 1: + set_bit(MMF_DUMPABLE, &mm->flags); + smp_wmb(); + clear_bit(MMF_DUMP_SECURELY, &mm->flags); + break; + case 2: + set_bit(MMF_DUMP_SECURELY, &mm->flags); + smp_wmb(); + set_bit(MMF_DUMPABLE, &mm->flags); + break; + } +} +EXPORT_SYMBOL_GPL(set_dumpable); + +int get_dumpable(struct mm_struct *mm) +{ + int ret; + + ret = mm->flags & 0x3; + return (ret >= 2) ? 2 : ret; +} + int do_coredump(long signr, int exit_code, struct pt_regs * regs) { char corename[CORENAME_MAX_SIZE + 1]; @@ -1502,7 +1733,7 @@ int do_coredump(long signr, int exit_code, struct pt_regs * regs) if (!binfmt || !binfmt->core_dump) goto fail; down_write(&mm->mmap_sem); - if (!mm->dumpable) { + if (!get_dumpable(mm)) { up_write(&mm->mmap_sem); goto fail; } @@ -1512,11 +1743,11 @@ int do_coredump(long signr, int exit_code, struct pt_regs * regs) * process nor do we know its entire history. We only know it * was tainted so we dump it as root in mode 2. */ - if (mm->dumpable == 2) { /* Setuid core dump mode */ + if (get_dumpable(mm) == 2) { /* Setuid core dump mode */ flag = O_EXCL; /* Stop rewrite attacks */ current->fsuid = 0; /* Dump root private */ } - mm->dumpable = 0; + set_dumpable(mm, 0); retval = coredump_wait(exit_code); if (retval < 0) diff --git a/fs/ext2/super.c b/fs/ext2/super.c index 3eefa97fe20..a6b1072daea 100644 --- a/fs/ext2/super.c +++ b/fs/ext2/super.c @@ -883,13 +883,11 @@ static int ext2_fill_super(struct super_block *sb, void *data, int silent) goto failed_mount; } bgl_lock_init(&sbi->s_blockgroup_lock); - sbi->s_debts = kmalloc(sbi->s_groups_count * sizeof(*sbi->s_debts), - GFP_KERNEL); + sbi->s_debts = kcalloc(sbi->s_groups_count, sizeof(*sbi->s_debts), GFP_KERNEL); if (!sbi->s_debts) { printk ("EXT2-fs: not enough memory\n"); goto failed_mount_group_desc; } - memset(sbi->s_debts, 0, sbi->s_groups_count * sizeof(*sbi->s_debts)); for (i = 0; i < db_count; i++) { block = descriptor_loc(sb, logic_sb_block, i); sbi->s_group_desc[i] = sb_bread(sb, block); diff --git a/fs/ext3/dir.c b/fs/ext3/dir.c index 852869840f2..c00723a99f4 100644 --- a/fs/ext3/dir.c +++ b/fs/ext3/dir.c @@ -136,12 +136,14 @@ static int ext3_readdir(struct file * filp, err = ext3_get_blocks_handle(NULL, inode, blk, 1, &map_bh, 0, 0); if (err > 0) { - page_cache_readahead(sb->s_bdev->bd_inode->i_mapping, - &filp->f_ra, - filp, - map_bh.b_blocknr >> - (PAGE_CACHE_SHIFT - inode->i_blkbits), - 1); + pgoff_t index = map_bh.b_blocknr >> + (PAGE_CACHE_SHIFT - inode->i_blkbits); + if (!ra_has_index(&filp->f_ra, index)) + page_cache_sync_readahead( + sb->s_bdev->bd_inode->i_mapping, + &filp->f_ra, filp, + index, 1); + filp->f_ra.prev_index = index; bh = ext3_bread(NULL, inode, blk, 0, &err); } diff --git a/fs/ext4/dir.c b/fs/ext4/dir.c index e8ad06e2831..3ab01c04e00 100644 --- a/fs/ext4/dir.c +++ b/fs/ext4/dir.c @@ -135,12 +135,14 @@ static int ext4_readdir(struct file * filp, map_bh.b_state = 0; err = ext4_get_blocks_wrap(NULL, inode, blk, 1, &map_bh, 0, 0); if (err > 0) { - page_cache_readahead(sb->s_bdev->bd_inode->i_mapping, - &filp->f_ra, - filp, - map_bh.b_blocknr >> - (PAGE_CACHE_SHIFT - inode->i_blkbits), - 1); + pgoff_t index = map_bh.b_blocknr >> + (PAGE_CACHE_SHIFT - inode->i_blkbits); + if (!ra_has_index(&filp->f_ra, index)) + page_cache_sync_readahead( + sb->s_bdev->bd_inode->i_mapping, + &filp->f_ra, filp, + index, 1); + filp->f_ra.prev_index = index; bh = ext4_bread(NULL, inode, blk, 0, &err); } diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c index de26c25d6a1..a4848e04a5e 100644 --- a/fs/ext4/inode.c +++ b/fs/ext4/inode.c @@ -2903,7 +2903,7 @@ int ext4_write_inode(struct inode *inode, int wait) return 0; if (ext4_journal_current_handle()) { - jbd_debug(0, "called recursively, non-PF_MEMALLOC!\n"); + jbd_debug(1, "called recursively, non-PF_MEMALLOC!\n"); dump_stack(); return -EIO; } diff --git a/fs/gfs2/ops_address.c b/fs/gfs2/ops_address.c index 26c888890c2..ce90032c010 100644 --- a/fs/gfs2/ops_address.c +++ b/fs/gfs2/ops_address.c @@ -251,7 +251,7 @@ static int gfs2_readpage(struct file *file, struct page *page) if (file) { gf = file->private_data; if (test_bit(GFF_EXLOCK, &gf->f_flags)) - /* gfs2_sharewrite_nopage has grabbed the ip->i_gl already */ + /* gfs2_sharewrite_fault has grabbed the ip->i_gl already */ goto skip_lock; } gfs2_holder_init(ip->i_gl, LM_ST_SHARED, GL_ATIME|LM_FLAG_TRY_1CB, &gh); diff --git a/fs/gfs2/ops_vm.c b/fs/gfs2/ops_vm.c index 404b7cc9f8c..927d739d468 100644 --- a/fs/gfs2/ops_vm.c +++ b/fs/gfs2/ops_vm.c @@ -27,13 +27,12 @@ #include "trans.h" #include "util.h" -static struct page *gfs2_private_nopage(struct vm_area_struct *area, - unsigned long address, int *type) +static int gfs2_private_fault(struct vm_area_struct *vma, struct vm_fault *vmf) { - struct gfs2_inode *ip = GFS2_I(area->vm_file->f_mapping->host); + struct gfs2_inode *ip = GFS2_I(vma->vm_file->f_mapping->host); set_bit(GIF_PAGED, &ip->i_flags); - return filemap_nopage(area, address, type); + return filemap_fault(vma, vmf); } static int alloc_page_backing(struct gfs2_inode *ip, struct page *page) @@ -104,58 +103,67 @@ out: return error; } -static struct page *gfs2_sharewrite_nopage(struct vm_area_struct *area, - unsigned long address, int *type) +static int gfs2_sharewrite_fault(struct vm_area_struct *vma, + struct vm_fault *vmf) { - struct file *file = area->vm_file; + struct file *file = vma->vm_file; struct gfs2_file *gf = file->private_data; struct gfs2_inode *ip = GFS2_I(file->f_mapping->host); struct gfs2_holder i_gh; - struct page *result = NULL; - unsigned long index = ((address - area->vm_start) >> PAGE_CACHE_SHIFT) + - area->vm_pgoff; int alloc_required; int error; + int ret = 0; error = gfs2_glock_nq_init(ip->i_gl, LM_ST_EXCLUSIVE, 0, &i_gh); if (error) - return NULL; + goto out; set_bit(GIF_PAGED, &ip->i_flags); set_bit(GIF_SW_PAGED, &ip->i_flags); - error = gfs2_write_alloc_required(ip, (u64)index << PAGE_CACHE_SHIFT, - PAGE_CACHE_SIZE, &alloc_required); - if (error) - goto out; + error = gfs2_write_alloc_required(ip, + (u64)vmf->pgoff << PAGE_CACHE_SHIFT, + PAGE_CACHE_SIZE, &alloc_required); + if (error) { + ret = VM_FAULT_OOM; /* XXX: are these right? */ + goto out_unlock; + } set_bit(GFF_EXLOCK, &gf->f_flags); - result = filemap_nopage(area, address, type); + ret = filemap_fault(vma, vmf); clear_bit(GFF_EXLOCK, &gf->f_flags); - if (!result || result == NOPAGE_OOM) - goto out; + if (ret & VM_FAULT_ERROR) + goto out_unlock; if (alloc_required) { - error = alloc_page_backing(ip, result); + /* XXX: do we need to drop page lock around alloc_page_backing?*/ + error = alloc_page_backing(ip, vmf->page); if (error) { - page_cache_release(result); - result = NULL; - goto out; + /* + * VM_FAULT_LOCKED should always be the case for + * filemap_fault, but it may not be in a future + * implementation. + */ + if (ret & VM_FAULT_LOCKED) + unlock_page(vmf->page); + page_cache_release(vmf->page); + ret = VM_FAULT_OOM; + goto out_unlock; } - set_page_dirty(result); + set_page_dirty(vmf->page); } -out: +out_unlock: gfs2_glock_dq_uninit(&i_gh); - - return result; +out: + return ret; } struct vm_operations_struct gfs2_vm_ops_private = { - .nopage = gfs2_private_nopage, + .fault = gfs2_private_fault, }; struct vm_operations_struct gfs2_vm_ops_sharewrite = { - .nopage = gfs2_sharewrite_nopage, + .fault = gfs2_sharewrite_fault, }; diff --git a/fs/jbd2/recovery.c b/fs/jbd2/recovery.c index e7730a045b9..b50be8a044e 100644 --- a/fs/jbd2/recovery.c +++ b/fs/jbd2/recovery.c @@ -251,10 +251,10 @@ int jbd2_journal_recover(journal_t *journal) if (!err) err = do_one_pass(journal, &info, PASS_REPLAY); - jbd_debug(0, "JBD: recovery, exit status %d, " + jbd_debug(1, "JBD: recovery, exit status %d, " "recovered transactions %u to %u\n", err, info.start_transaction, info.end_transaction); - jbd_debug(0, "JBD: Replayed %d and revoked %d/%d blocks\n", + jbd_debug(1, "JBD: Replayed %d and revoked %d/%d blocks\n", info.nr_replays, info.nr_revoke_hits, info.nr_revokes); /* Restart the log at the next transaction ID, thus invalidating @@ -298,7 +298,7 @@ int jbd2_journal_skip_recovery(journal_t *journal) #ifdef CONFIG_JBD2_DEBUG int dropped = info.end_transaction - be32_to_cpu(sb->s_sequence); #endif - jbd_debug(0, + jbd_debug(1, "JBD: ignoring %d transaction%s from the journal.\n", dropped, (dropped == 1) ? "" : "s"); journal->j_transaction_sequence = ++info.end_transaction; diff --git a/fs/namei.c b/fs/namei.c index defaa47c11d..a83160acd74 100644 --- a/fs/namei.c +++ b/fs/namei.c @@ -107,6 +107,8 @@ * any extra contention... */ +static int fastcall link_path_walk(const char *name, struct nameidata *nd); + /* In order to reduce some races, while at the same time doing additional * checking and hopefully speeding things up, we copy filenames to the * kernel data space before using them.. @@ -998,7 +1000,7 @@ return_err: * Retry the whole path once, forcing real lookup requests * instead of relying on the dcache. */ -int fastcall link_path_walk(const char *name, struct nameidata *nd) +static int fastcall link_path_walk(const char *name, struct nameidata *nd) { struct nameidata save = *nd; int result; @@ -1022,7 +1024,7 @@ int fastcall link_path_walk(const char *name, struct nameidata *nd) return result; } -int fastcall path_walk(const char * name, struct nameidata *nd) +static int fastcall path_walk(const char * name, struct nameidata *nd) { current->total_link_count = 0; return link_path_walk(name, nd); @@ -1172,6 +1174,37 @@ int fastcall path_lookup(const char *name, unsigned int flags, return do_path_lookup(AT_FDCWD, name, flags, nd); } +/** + * vfs_path_lookup - lookup a file path relative to a dentry-vfsmount pair + * @dentry: pointer to dentry of the base directory + * @mnt: pointer to vfs mount of the base directory + * @name: pointer to file name + * @flags: lookup flags + * @nd: pointer to nameidata + */ +int vfs_path_lookup(struct dentry *dentry, struct vfsmount *mnt, + const char *name, unsigned int flags, + struct nameidata *nd) +{ + int retval; + + /* same as do_path_lookup */ + nd->last_type = LAST_ROOT; + nd->flags = flags; + nd->depth = 0; + + nd->mnt = mntget(mnt); + nd->dentry = dget(dentry); + + retval = path_walk(name, nd); + if (unlikely(!retval && !audit_dummy_context() && nd->dentry && + nd->dentry->d_inode)) + audit_inode(name, nd->dentry->d_inode); + + return retval; + +} + static int __path_lookup_intent_open(int dfd, const char *name, unsigned int lookup_flags, struct nameidata *nd, int open_flags, int create_mode) @@ -2774,8 +2807,8 @@ EXPORT_SYMBOL(__page_symlink); EXPORT_SYMBOL(page_symlink); EXPORT_SYMBOL(page_symlink_inode_operations); EXPORT_SYMBOL(path_lookup); +EXPORT_SYMBOL(vfs_path_lookup); EXPORT_SYMBOL(path_release); -EXPORT_SYMBOL(path_walk); EXPORT_SYMBOL(permission); EXPORT_SYMBOL(vfs_permission); EXPORT_SYMBOL(file_permission); diff --git a/fs/ncpfs/mmap.c b/fs/ncpfs/mmap.c index 70a69115500..a94473d3072 100644 --- a/fs/ncpfs/mmap.c +++ b/fs/ncpfs/mmap.c @@ -24,31 +24,35 @@ /* * Fill in the supplied page for mmap + * XXX: how are we excluding truncate/invalidate here? Maybe need to lock + * page? */ -static struct page* ncp_file_mmap_nopage(struct vm_area_struct *area, - unsigned long address, int *type) +static int ncp_file_mmap_fault(struct vm_area_struct *area, + struct vm_fault *vmf) { struct file *file = area->vm_file; struct dentry *dentry = file->f_path.dentry; struct inode *inode = dentry->d_inode; - struct page* page; char *pg_addr; unsigned int already_read; unsigned int count; int bufsize; - int pos; + int pos; /* XXX: loff_t ? */ - page = alloc_page(GFP_HIGHUSER); /* ncpfs has nothing against high pages - as long as recvmsg and memset works on it */ - if (!page) - return page; - pg_addr = kmap(page); - address &= PAGE_MASK; - pos = address - area->vm_start + (area->vm_pgoff << PAGE_SHIFT); + /* + * ncpfs has nothing against high pages as long + * as recvmsg and memset works on it + */ + vmf->page = alloc_page(GFP_HIGHUSER); + if (!vmf->page) + return VM_FAULT_OOM; + pg_addr = kmap(vmf->page); + pos = vmf->pgoff << PAGE_SHIFT; count = PAGE_SIZE; - if (address + PAGE_SIZE > area->vm_end) { - count = area->vm_end - address; + if ((unsigned long)vmf->virtual_address + PAGE_SIZE > area->vm_end) { + WARN_ON(1); /* shouldn't happen? */ + count = area->vm_end - (unsigned long)vmf->virtual_address; } /* what we can read in one go */ bufsize = NCP_SERVER(inode)->buffer_size; @@ -83,23 +87,21 @@ static struct page* ncp_file_mmap_nopage(struct vm_area_struct *area, if (already_read < PAGE_SIZE) memset(pg_addr + already_read, 0, PAGE_SIZE - already_read); - flush_dcache_page(page); - kunmap(page); + flush_dcache_page(vmf->page); + kunmap(vmf->page); /* * If I understand ncp_read_kernel() properly, the above always * fetches from the network, here the analogue of disk. * -- wli */ - if (type) - *type = VM_FAULT_MAJOR; count_vm_event(PGMAJFAULT); - return page; + return VM_FAULT_MAJOR; } static struct vm_operations_struct ncp_file_mmap = { - .nopage = ncp_file_mmap_nopage, + .fault = ncp_file_mmap_fault, }; diff --git a/fs/nfs/callback_xdr.c b/fs/nfs/callback_xdr.c index 849a2029975..058ade7efe7 100644 --- a/fs/nfs/callback_xdr.c +++ b/fs/nfs/callback_xdr.c @@ -179,7 +179,7 @@ static __be32 decode_getattr_args(struct svc_rqst *rqstp, struct xdr_stream *xdr args->addr = svc_addr_in(rqstp); status = decode_bitmap(xdr, args->bitmap); out: - dprintk("%s: exit with status = %d\n", __FUNCTION__, status); + dprintk("%s: exit with status = %d\n", __FUNCTION__, ntohl(status)); return status; } @@ -200,7 +200,7 @@ static __be32 decode_recall_args(struct svc_rqst *rqstp, struct xdr_stream *xdr, args->truncate = ntohl(*p); status = decode_fh(xdr, &args->fh); out: - dprintk("%s: exit with status = %d\n", __FUNCTION__, status); + dprintk("%s: exit with status = %d\n", __FUNCTION__, ntohl(status)); return status; } @@ -349,7 +349,7 @@ static __be32 encode_getattr_res(struct svc_rqst *rqstp, struct xdr_stream *xdr, status = encode_attr_mtime(xdr, res->bitmap, &res->mtime); *savep = htonl((unsigned int)((char *)xdr->p - (char *)(savep+1))); out: - dprintk("%s: exit with status = %d\n", __FUNCTION__, status); + dprintk("%s: exit with status = %d\n", __FUNCTION__, ntohl(status)); return status; } @@ -392,7 +392,7 @@ static __be32 process_op(struct svc_rqst *rqstp, status = res; if (op->encode_res != NULL && status == 0) status = op->encode_res(rqstp, xdr_out, resp); - dprintk("%s: done, status = %d\n", __FUNCTION__, status); + dprintk("%s: done, status = %d\n", __FUNCTION__, ntohl(status)); return status; } @@ -431,7 +431,7 @@ static __be32 nfs4_callback_compound(struct svc_rqst *rqstp, void *argp, void *r } *hdr_res.status = status; *hdr_res.nops = htonl(nops); - dprintk("%s: done, status = %u\n", __FUNCTION__, status); + dprintk("%s: done, status = %u\n", __FUNCTION__, ntohl(status)); return rpc_success; } diff --git a/fs/nfs/dir.c b/fs/nfs/dir.c index 322141f4ab4..ea97408e423 100644 --- a/fs/nfs/dir.c +++ b/fs/nfs/dir.c @@ -654,7 +654,7 @@ static int nfs_check_verifier(struct inode *dir, struct dentry *dentry) if (IS_ROOT(dentry)) return 1; - verf = (unsigned long)dentry->d_fsdata; + verf = dentry->d_time; if (nfs_caches_unstable(dir) || verf != NFS_I(dir)->cache_change_attribute) return 0; @@ -663,7 +663,7 @@ static int nfs_check_verifier(struct inode *dir, struct dentry *dentry) static inline void nfs_set_verifier(struct dentry * dentry, unsigned long verf) { - dentry->d_fsdata = (void *)verf; + dentry->d_time = verf; } static void nfs_refresh_verifier(struct dentry * dentry, unsigned long verf) @@ -869,7 +869,7 @@ static void nfs_dentry_iput(struct dentry *dentry, struct inode *inode) if (dentry->d_flags & DCACHE_NFSFS_RENAMED) { lock_kernel(); drop_nlink(inode); - nfs_complete_unlink(dentry); + nfs_complete_unlink(dentry, inode); unlock_kernel(); } /* When creating a negative dentry, we want to renew d_time */ @@ -1411,7 +1411,7 @@ static int nfs_sillyrename(struct inode *dir, struct dentry *dentry) nfs_renew_times(dentry); nfs_set_verifier(dentry, nfs_save_change_attribute(dir)); d_move(dentry, sdentry); - error = nfs_async_unlink(dentry); + error = nfs_async_unlink(dir, dentry); /* If we return 0 we don't unlink */ } dput(sdentry); diff --git a/fs/nfs/nfs2xdr.c b/fs/nfs/nfs2xdr.c index 7fcc78f2aa7..c5fce756720 100644 --- a/fs/nfs/nfs2xdr.c +++ b/fs/nfs/nfs2xdr.c @@ -43,6 +43,7 @@ #define NFS_entry_sz (NFS_filename_sz+3) #define NFS_diropargs_sz (NFS_fhandle_sz+NFS_filename_sz) +#define NFS_removeargs_sz (NFS_fhandle_sz+NFS_filename_sz) #define NFS_sattrargs_sz (NFS_fhandle_sz+NFS_sattr_sz) #define NFS_readlinkargs_sz (NFS_fhandle_sz) #define NFS_readargs_sz (NFS_fhandle_sz+3) @@ -66,7 +67,7 @@ * Common NFS XDR functions as inlines */ static inline __be32 * -xdr_encode_fhandle(__be32 *p, struct nfs_fh *fhandle) +xdr_encode_fhandle(__be32 *p, const struct nfs_fh *fhandle) { memcpy(p, fhandle->data, NFS2_FHSIZE); return p + XDR_QUADLEN(NFS2_FHSIZE); @@ -204,7 +205,7 @@ nfs_xdr_sattrargs(struct rpc_rqst *req, __be32 *p, struct nfs_sattrargs *args) /* * Encode directory ops argument - * LOOKUP, REMOVE, RMDIR + * LOOKUP, RMDIR */ static int nfs_xdr_diropargs(struct rpc_rqst *req, __be32 *p, struct nfs_diropargs *args) @@ -216,6 +217,18 @@ nfs_xdr_diropargs(struct rpc_rqst *req, __be32 *p, struct nfs_diropargs *args) } /* + * Encode REMOVE argument + */ +static int +nfs_xdr_removeargs(struct rpc_rqst *req, __be32 *p, const struct nfs_removeargs *args) +{ + p = xdr_encode_fhandle(p, args->fh); + p = xdr_encode_array(p, args->name.name, args->name.len); + req->rq_slen = xdr_adjust_iovec(req->rq_svec, p); + return 0; +} + +/* * Arguments to a READ call. Since we read data directly into the page * cache, we also set up the reply iovec here so that iov[1] points * exactly to the page we want to fetch. @@ -705,7 +718,7 @@ struct rpc_procinfo nfs_procedures[] = { PROC(READ, readargs, readres, 3), PROC(WRITE, writeargs, writeres, 4), PROC(CREATE, createargs, diropres, 0), - PROC(REMOVE, diropargs, stat, 0), + PROC(REMOVE, removeargs, stat, 0), PROC(RENAME, renameargs, stat, 0), PROC(LINK, linkargs, stat, 0), PROC(SYMLINK, symlinkargs, stat, 0), diff --git a/fs/nfs/nfs3proc.c b/fs/nfs/nfs3proc.c index 814d886b6aa..c7ca5d70870 100644 --- a/fs/nfs/nfs3proc.c +++ b/fs/nfs/nfs3proc.c @@ -349,62 +349,42 @@ out: static int nfs3_proc_remove(struct inode *dir, struct qstr *name) { - struct nfs_fattr dir_attr; - struct nfs3_diropargs arg = { - .fh = NFS_FH(dir), - .name = name->name, - .len = name->len + struct nfs_removeargs arg = { + .fh = NFS_FH(dir), + .name.len = name->len, + .name.name = name->name, }; - struct rpc_message msg = { - .rpc_proc = &nfs3_procedures[NFS3PROC_REMOVE], - .rpc_argp = &arg, - .rpc_resp = &dir_attr, + struct nfs_removeres res; + struct rpc_message msg = { + .rpc_proc = &nfs3_procedures[NFS3PROC_REMOVE], + .rpc_argp = &arg, + .rpc_resp = &res, }; int status; dprintk("NFS call remove %s\n", name->name); - nfs_fattr_init(&dir_attr); + nfs_fattr_init(&res.dir_attr); status = rpc_call_sync(NFS_CLIENT(dir), &msg, 0); - nfs_post_op_update_inode(dir, &dir_attr); + nfs_post_op_update_inode(dir, &res.dir_attr); dprintk("NFS reply remove: %d\n", status); return status; } -static int -nfs3_proc_unlink_setup(struct rpc_message *msg, struct dentry *dir, struct qstr *name) +static void +nfs3_proc_unlink_setup(struct rpc_message *msg, struct inode *dir) { - struct unlinkxdr { - struct nfs3_diropargs arg; - struct nfs_fattr res; - } *ptr; - - ptr = kmalloc(sizeof(*ptr), GFP_KERNEL); - if (!ptr) - return -ENOMEM; - ptr->arg.fh = NFS_FH(dir->d_inode); - ptr->arg.name = name->name; - ptr->arg.len = name->len; - nfs_fattr_init(&ptr->res); msg->rpc_proc = &nfs3_procedures[NFS3PROC_REMOVE]; - msg->rpc_argp = &ptr->arg; - msg->rpc_resp = &ptr->res; - return 0; } static int -nfs3_proc_unlink_done(struct dentry *dir, struct rpc_task *task) +nfs3_proc_unlink_done(struct rpc_task *task, struct inode *dir) { - struct rpc_message *msg = &task->tk_msg; - struct nfs_fattr *dir_attr; - - if (nfs3_async_handle_jukebox(task, dir->d_inode)) - return 1; - if (msg->rpc_argp) { - dir_attr = (struct nfs_fattr*)msg->rpc_resp; - nfs_post_op_update_inode(dir->d_inode, dir_attr); - kfree(msg->rpc_argp); - } - return 0; + struct nfs_removeres *res; + if (nfs3_async_handle_jukebox(task, dir)) + return 0; + res = task->tk_msg.rpc_resp; + nfs_post_op_update_inode(dir, &res->dir_attr); + return 1; } static int diff --git a/fs/nfs/nfs3xdr.c b/fs/nfs/nfs3xdr.c index b4647a22f34..d9e08f0cf2a 100644 --- a/fs/nfs/nfs3xdr.c +++ b/fs/nfs/nfs3xdr.c @@ -50,6 +50,7 @@ #define NFS3_sattrargs_sz (NFS3_fh_sz+NFS3_sattr_sz+3) #define NFS3_diropargs_sz (NFS3_fh_sz+NFS3_filename_sz) +#define NFS3_removeargs_sz (NFS3_fh_sz+NFS3_filename_sz) #define NFS3_accessargs_sz (NFS3_fh_sz+1) #define NFS3_readlinkargs_sz (NFS3_fh_sz) #define NFS3_readargs_sz (NFS3_fh_sz+3) @@ -65,6 +66,7 @@ #define NFS3_attrstat_sz (1+NFS3_fattr_sz) #define NFS3_wccstat_sz (1+NFS3_wcc_data_sz) +#define NFS3_removeres_sz (NFS3_wccstat_sz) #define NFS3_lookupres_sz (1+NFS3_fh_sz+(2 * NFS3_post_op_attr_sz)) #define NFS3_accessres_sz (1+NFS3_post_op_attr_sz+1) #define NFS3_readlinkres_sz (1+NFS3_post_op_attr_sz+1) @@ -106,7 +108,7 @@ static struct { * Common NFS XDR functions as inlines */ static inline __be32 * -xdr_encode_fhandle(__be32 *p, struct nfs_fh *fh) +xdr_encode_fhandle(__be32 *p, const struct nfs_fh *fh) { return xdr_encode_array(p, fh->data, fh->size); } @@ -300,6 +302,18 @@ nfs3_xdr_diropargs(struct rpc_rqst *req, __be32 *p, struct nfs3_diropargs *args) } /* + * Encode REMOVE argument + */ +static int +nfs3_xdr_removeargs(struct rpc_rqst *req, __be32 *p, const struct nfs_removeargs *args) +{ + p = xdr_encode_fhandle(p, args->fh); + p = xdr_encode_array(p, args->name.name, args->name.len); + req->rq_slen = xdr_adjust_iovec(req->rq_svec, p); + return 0; +} + +/* * Encode access() argument */ static int @@ -736,6 +750,12 @@ nfs3_xdr_wccstat(struct rpc_rqst *req, __be32 *p, struct nfs_fattr *fattr) return status; } +static int +nfs3_xdr_removeres(struct rpc_rqst *req, __be32 *p, struct nfs_removeres *res) +{ + return nfs3_xdr_wccstat(req, p, &res->dir_attr); +} + /* * Decode LOOKUP reply */ @@ -1126,7 +1146,7 @@ struct rpc_procinfo nfs3_procedures[] = { PROC(MKDIR, mkdirargs, createres, 0), PROC(SYMLINK, symlinkargs, createres, 0), PROC(MKNOD, mknodargs, createres, 0), - PROC(REMOVE, diropargs, wccstat, 0), + PROC(REMOVE, removeargs, removeres, 0), PROC(RMDIR, diropargs, wccstat, 0), PROC(RENAME, renameargs, renameres, 0), PROC(LINK, linkargs, linkres, 0), diff --git a/fs/nfs/nfs4_fs.h b/fs/nfs/nfs4_fs.h index 6c028e734fe..d2802b1ca3b 100644 --- a/fs/nfs/nfs4_fs.h +++ b/fs/nfs/nfs4_fs.h @@ -182,7 +182,7 @@ extern int nfs4_do_close(struct path *path, struct nfs4_state *state); extern struct dentry *nfs4_atomic_open(struct inode *, struct dentry *, struct nameidata *); extern int nfs4_open_revalidate(struct inode *, struct dentry *, int, struct nameidata *); extern int nfs4_server_capabilities(struct nfs_server *server, struct nfs_fh *fhandle); -extern int nfs4_proc_fs_locations(struct inode *dir, struct qstr *name, +extern int nfs4_proc_fs_locations(struct inode *dir, const struct qstr *name, struct nfs4_fs_locations *fs_locations, struct page *page); extern struct nfs4_state_recovery_ops nfs4_reboot_recovery_ops; diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c index fee2da856c9..6ca2795ccd9 100644 --- a/fs/nfs/nfs4proc.c +++ b/fs/nfs/nfs4proc.c @@ -66,6 +66,8 @@ static int _nfs4_proc_access(struct inode *inode, struct nfs_access_entry *entry static int nfs4_handle_exception(const struct nfs_server *server, int errorcode, struct nfs4_exception *exception); static int nfs4_wait_clnt_recover(struct rpc_clnt *clnt, struct nfs_client *clp); static int _nfs4_do_access(struct inode *inode, struct rpc_cred *cred, int openflags); +static int _nfs4_proc_lookup(struct inode *dir, const struct qstr *name, struct nfs_fh *fhandle, struct nfs_fattr *fattr); +static int _nfs4_proc_getattr(struct nfs_server *server, struct nfs_fh *fhandle, struct nfs_fattr *fattr); /* Prevent leaks of NFSv4 errors into userland */ int nfs4_map_errors(int err) @@ -552,6 +554,18 @@ static struct nfs_open_context *nfs4_state_find_open_context(struct nfs4_state * return ERR_PTR(-ENOENT); } +static struct nfs4_opendata *nfs4_open_recoverdata_alloc(struct nfs_open_context *ctx, struct nfs4_state *state) +{ + struct nfs4_opendata *opendata; + + opendata = nfs4_opendata_alloc(&ctx->path, state->owner, 0, NULL); + if (opendata == NULL) + return ERR_PTR(-ENOMEM); + opendata->state = state; + atomic_inc(&state->count); + return opendata; +} + static int nfs4_open_recover_helper(struct nfs4_opendata *opendata, mode_t openflags, struct nfs4_state **res) { struct nfs4_state *newstate; @@ -626,12 +640,11 @@ static int _nfs4_do_open_reclaim(struct nfs_open_context *ctx, struct nfs4_state int delegation_type = 0; int status; - opendata = nfs4_opendata_alloc(&ctx->path, state->owner, 0, NULL); - if (opendata == NULL) - return -ENOMEM; + opendata = nfs4_open_recoverdata_alloc(ctx, state); + if (IS_ERR(opendata)) + return PTR_ERR(opendata); opendata->o_arg.claim = NFS4_OPEN_CLAIM_PREVIOUS; opendata->o_arg.fh = NFS_FH(state->inode); - nfs_copy_fh(&opendata->o_res.fh, opendata->o_arg.fh); rcu_read_lock(); delegation = rcu_dereference(NFS_I(state->inode)->delegation); if (delegation != NULL && (delegation->flags & NFS_DELEGATION_NEED_RECLAIM) != 0) @@ -672,13 +685,12 @@ static int nfs4_open_reclaim(struct nfs4_state_owner *sp, struct nfs4_state *sta static int _nfs4_open_delegation_recall(struct nfs_open_context *ctx, struct nfs4_state *state, const nfs4_stateid *stateid) { - struct nfs4_state_owner *sp = state->owner; struct nfs4_opendata *opendata; int ret; - opendata = nfs4_opendata_alloc(&ctx->path, sp, 0, NULL); - if (opendata == NULL) - return -ENOMEM; + opendata = nfs4_open_recoverdata_alloc(ctx, state); + if (IS_ERR(opendata)) + return PTR_ERR(opendata); opendata->o_arg.claim = NFS4_OPEN_CLAIM_DELEGATE_CUR; memcpy(opendata->o_arg.u.delegation.data, stateid->data, sizeof(opendata->o_arg.u.delegation.data)); @@ -823,8 +835,10 @@ static void nfs4_open_prepare(struct rpc_task *task, void *calldata) /* Update sequence id. */ data->o_arg.id = sp->so_owner_id.id; data->o_arg.clientid = sp->so_client->cl_clientid; - if (data->o_arg.claim == NFS4_OPEN_CLAIM_PREVIOUS) + if (data->o_arg.claim == NFS4_OPEN_CLAIM_PREVIOUS) { msg.rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_OPEN_NOATTR]; + nfs_copy_fh(&data->o_res.fh, data->o_arg.fh); + } data->timestamp = jiffies; rpc_call_setup(task, &msg, 0); return; @@ -918,6 +932,9 @@ static int _nfs4_proc_open(struct nfs4_opendata *data) if (status != 0 || !data->rpc_done) return status; + if (o_res->fh.size == 0) + _nfs4_proc_lookup(dir, o_arg->name, &o_res->fh, o_res->f_attr); + if (o_arg->open_flags & O_CREAT) { update_changeattr(dir, &o_res->cinfo); nfs_post_op_update_inode(dir, o_res->dir_attr); @@ -929,7 +946,7 @@ static int _nfs4_proc_open(struct nfs4_opendata *data) return status; } if (!(o_res->f_attr->valid & NFS_ATTR_FATTR)) - return server->nfs_client->rpc_ops->getattr(server, &o_res->fh, o_res->f_attr); + _nfs4_proc_getattr(server, &o_res->fh, o_res->f_attr); return 0; } @@ -989,9 +1006,9 @@ static int _nfs4_open_expired(struct nfs_open_context *ctx, struct nfs4_state *s struct nfs4_opendata *opendata; int ret; - opendata = nfs4_opendata_alloc(&ctx->path, state->owner, 0, NULL); - if (opendata == NULL) - return -ENOMEM; + opendata = nfs4_open_recoverdata_alloc(ctx, state); + if (IS_ERR(opendata)) + return PTR_ERR(opendata); ret = nfs4_open_recover(opendata, state); if (ret == -ESTALE) { /* Invalidate the state owner so we don't ever use it again */ @@ -1553,7 +1570,7 @@ static int nfs4_proc_get_root(struct nfs_server *server, struct nfs_fh *fhandle, * Note that we'll actually follow the referral later when * we detect fsid mismatch in inode revalidation */ -static int nfs4_get_referral(struct inode *dir, struct qstr *name, struct nfs_fattr *fattr, struct nfs_fh *fhandle) +static int nfs4_get_referral(struct inode *dir, const struct qstr *name, struct nfs_fattr *fattr, struct nfs_fh *fhandle) { int status = -ENOMEM; struct page *page = NULL; @@ -1668,8 +1685,8 @@ nfs4_proc_setattr(struct dentry *dentry, struct nfs_fattr *fattr, return status; } -static int _nfs4_proc_lookupfh(struct nfs_server *server, struct nfs_fh *dirfh, - struct qstr *name, struct nfs_fh *fhandle, +static int _nfs4_proc_lookupfh(struct nfs_server *server, const struct nfs_fh *dirfh, + const struct qstr *name, struct nfs_fh *fhandle, struct nfs_fattr *fattr) { int status; @@ -1715,7 +1732,7 @@ static int nfs4_proc_lookupfh(struct nfs_server *server, struct nfs_fh *dirfh, return err; } -static int _nfs4_proc_lookup(struct inode *dir, struct qstr *name, +static int _nfs4_proc_lookup(struct inode *dir, const struct qstr *name, struct nfs_fh *fhandle, struct nfs_fattr *fattr) { int status; @@ -1908,28 +1925,27 @@ out: static int _nfs4_proc_remove(struct inode *dir, struct qstr *name) { struct nfs_server *server = NFS_SERVER(dir); - struct nfs4_remove_arg args = { + struct nfs_removeargs args = { .fh = NFS_FH(dir), - .name = name, + .name.len = name->len, + .name.name = name->name, .bitmask = server->attr_bitmask, }; - struct nfs_fattr dir_attr; - struct nfs4_remove_res res = { + struct nfs_removeres res = { .server = server, - .dir_attr = &dir_attr, }; struct rpc_message msg = { - .rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_REMOVE], - .rpc_argp = &args, - .rpc_resp = &res, + .rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_REMOVE], + .rpc_argp = &args, + .rpc_resp = &res, }; int status; - nfs_fattr_init(res.dir_attr); + nfs_fattr_init(&res.dir_attr); status = rpc_call_sync(server->client, &msg, 0); if (status == 0) { update_changeattr(dir, &res.cinfo); - nfs_post_op_update_inode(dir, res.dir_attr); + nfs_post_op_update_inode(dir, &res.dir_attr); } return status; } @@ -1946,48 +1962,26 @@ static int nfs4_proc_remove(struct inode *dir, struct qstr *name) return err; } -struct unlink_desc { - struct nfs4_remove_arg args; - struct nfs4_remove_res res; - struct nfs_fattr dir_attr; -}; - -static int nfs4_proc_unlink_setup(struct rpc_message *msg, struct dentry *dir, - struct qstr *name) +static void nfs4_proc_unlink_setup(struct rpc_message *msg, struct inode *dir) { - struct nfs_server *server = NFS_SERVER(dir->d_inode); - struct unlink_desc *up; + struct nfs_server *server = NFS_SERVER(dir); + struct nfs_removeargs *args = msg->rpc_argp; + struct nfs_removeres *res = msg->rpc_resp; - up = kmalloc(sizeof(*up), GFP_KERNEL); - if (!up) - return -ENOMEM; - - up->args.fh = NFS_FH(dir->d_inode); - up->args.name = name; - up->args.bitmask = server->attr_bitmask; - up->res.server = server; - up->res.dir_attr = &up->dir_attr; - + args->bitmask = server->attr_bitmask; + res->server = server; msg->rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_REMOVE]; - msg->rpc_argp = &up->args; - msg->rpc_resp = &up->res; - return 0; } -static int nfs4_proc_unlink_done(struct dentry *dir, struct rpc_task *task) +static int nfs4_proc_unlink_done(struct rpc_task *task, struct inode *dir) { - struct rpc_message *msg = &task->tk_msg; - struct unlink_desc *up; - - if (msg->rpc_resp != NULL) { - up = container_of(msg->rpc_resp, struct unlink_desc, res); - update_changeattr(dir->d_inode, &up->res.cinfo); - nfs_post_op_update_inode(dir->d_inode, up->res.dir_attr); - kfree(up); - msg->rpc_resp = NULL; - msg->rpc_argp = NULL; - } - return 0; + struct nfs_removeres *res = task->tk_msg.rpc_resp; + + if (nfs4_async_handle_error(task, res->server) == -EAGAIN) + return 0; + update_changeattr(dir, &res->cinfo); + nfs_post_op_update_inode(dir, &res->dir_attr); + return 1; } static int _nfs4_proc_rename(struct inode *old_dir, struct qstr *old_name, @@ -3672,7 +3666,7 @@ ssize_t nfs4_listxattr(struct dentry *dentry, char *buf, size_t buflen) return len; } -int nfs4_proc_fs_locations(struct inode *dir, struct qstr *name, +int nfs4_proc_fs_locations(struct inode *dir, const struct qstr *name, struct nfs4_fs_locations *fs_locations, struct page *page) { struct nfs_server *server = NFS_SERVER(dir); diff --git a/fs/nfs/nfs4xdr.c b/fs/nfs/nfs4xdr.c index c08738441f7..badd73b7ca1 100644 --- a/fs/nfs/nfs4xdr.c +++ b/fs/nfs/nfs4xdr.c @@ -72,10 +72,15 @@ static int nfs4_stat_to_errno(int); */ #define open_owner_id_maxsz (1 + 4) #define lock_owner_id_maxsz (1 + 4) +#define decode_lockowner_maxsz (1 + XDR_QUADLEN(IDMAP_NAMESZ)) #define compound_encode_hdr_maxsz (3 + (NFS4_MAXTAGLEN >> 2)) #define compound_decode_hdr_maxsz (3 + (NFS4_MAXTAGLEN >> 2)) #define op_encode_hdr_maxsz (1) #define op_decode_hdr_maxsz (2) +#define encode_stateid_maxsz (XDR_QUADLEN(NFS4_STATEID_SIZE)) +#define decode_stateid_maxsz (XDR_QUADLEN(NFS4_STATEID_SIZE)) +#define encode_verifier_maxsz (XDR_QUADLEN(NFS4_VERIFIER_SIZE)) +#define decode_verifier_maxsz (XDR_QUADLEN(NFS4_VERIFIER_SIZE)) #define encode_putfh_maxsz (op_encode_hdr_maxsz + 1 + \ (NFS4_FHSIZE >> 2)) #define decode_putfh_maxsz (op_decode_hdr_maxsz) @@ -96,6 +101,11 @@ static int nfs4_stat_to_errno(int); #define nfs4_fattr_maxsz (nfs4_fattr_bitmap_maxsz + \ nfs4_fattr_value_maxsz) #define decode_getattr_maxsz (op_decode_hdr_maxsz + nfs4_fattr_maxsz) +#define encode_attrs_maxsz (nfs4_fattr_bitmap_maxsz + \ + 1 + 2 + 1 + \ + nfs4_owner_maxsz + \ + nfs4_group_maxsz + \ + 4 + 4) #define encode_savefh_maxsz (op_encode_hdr_maxsz) #define decode_savefh_maxsz (op_decode_hdr_maxsz) #define encode_restorefh_maxsz (op_encode_hdr_maxsz) @@ -123,7 +133,7 @@ static int nfs4_stat_to_errno(int); #define decode_lookup_maxsz (op_decode_hdr_maxsz) #define encode_share_access_maxsz \ (2) -#define encode_createmode_maxsz (1 + nfs4_fattr_maxsz) +#define encode_createmode_maxsz (1 + encode_attrs_maxsz) #define encode_opentype_maxsz (1 + encode_createmode_maxsz) #define encode_claim_null_maxsz (1 + nfs4_name_maxsz) #define encode_open_maxsz (op_encode_hdr_maxsz + \ @@ -132,14 +142,52 @@ static int nfs4_stat_to_errno(int); encode_opentype_maxsz + \ encode_claim_null_maxsz) #define decode_ace_maxsz (3 + nfs4_owner_maxsz) -#define decode_delegation_maxsz (1 + XDR_QUADLEN(NFS4_STATEID_SIZE) + 1 + \ +#define decode_delegation_maxsz (1 + decode_stateid_maxsz + 1 + \ decode_ace_maxsz) #define decode_change_info_maxsz (5) #define decode_open_maxsz (op_decode_hdr_maxsz + \ - XDR_QUADLEN(NFS4_STATEID_SIZE) + \ + decode_stateid_maxsz + \ decode_change_info_maxsz + 1 + \ nfs4_fattr_bitmap_maxsz + \ decode_delegation_maxsz) +#define encode_open_confirm_maxsz \ + (op_encode_hdr_maxsz + \ + encode_stateid_maxsz + 1) +#define decode_open_confirm_maxsz \ + (op_decode_hdr_maxsz + \ + decode_stateid_maxsz) +#define encode_open_downgrade_maxsz \ + (op_encode_hdr_maxsz + \ + encode_stateid_maxsz + 1 + \ + encode_share_access_maxsz) +#define decode_open_downgrade_maxsz \ + (op_decode_hdr_maxsz + \ + decode_stateid_maxsz) +#define encode_close_maxsz (op_encode_hdr_maxsz + \ + 1 + encode_stateid_maxsz) +#define decode_close_maxsz (op_decode_hdr_maxsz + \ + decode_stateid_maxsz) +#define encode_setattr_maxsz (op_encode_hdr_maxsz + \ + encode_stateid_maxsz + \ + encode_attrs_maxsz) +#define decode_setattr_maxsz (op_decode_hdr_maxsz + \ + nfs4_fattr_bitmap_maxsz) +#define encode_read_maxsz (op_encode_hdr_maxsz + \ + encode_stateid_maxsz + 3) +#define decode_read_maxsz (op_decode_hdr_maxsz + 2) +#define encode_readdir_maxsz (op_encode_hdr_maxsz + \ + 2 + encode_verifier_maxsz + 5) +#define decode_readdir_maxsz (op_decode_hdr_maxsz + \ + decode_verifier_maxsz) +#define encode_readlink_maxsz (op_encode_hdr_maxsz) +#define decode_readlink_maxsz (op_decode_hdr_maxsz + 1) +#define encode_write_maxsz (op_encode_hdr_maxsz + \ + encode_stateid_maxsz + 4) +#define decode_write_maxsz (op_decode_hdr_maxsz + \ + 2 + decode_verifier_maxsz) +#define encode_commit_maxsz (op_encode_hdr_maxsz + 3) +#define decode_commit_maxsz (op_decode_hdr_maxsz + \ + decode_verifier_maxsz) #define encode_remove_maxsz (op_encode_hdr_maxsz + \ nfs4_name_maxsz) #define encode_rename_maxsz (op_encode_hdr_maxsz + \ @@ -148,19 +196,44 @@ static int nfs4_stat_to_errno(int); #define encode_link_maxsz (op_encode_hdr_maxsz + \ nfs4_name_maxsz) #define decode_link_maxsz (op_decode_hdr_maxsz + 5) +#define encode_lock_maxsz (op_encode_hdr_maxsz + \ + 7 + \ + 1 + encode_stateid_maxsz + 8) +#define decode_lock_denied_maxsz \ + (8 + decode_lockowner_maxsz) +#define decode_lock_maxsz (op_decode_hdr_maxsz + \ + decode_lock_denied_maxsz) +#define encode_lockt_maxsz (op_encode_hdr_maxsz + 12) +#define decode_lockt_maxsz (op_decode_hdr_maxsz + \ + decode_lock_denied_maxsz) +#define encode_locku_maxsz (op_encode_hdr_maxsz + 3 + \ + encode_stateid_maxsz + \ + 4) +#define decode_locku_maxsz (op_decode_hdr_maxsz + \ + decode_stateid_maxsz) +#define encode_access_maxsz (op_encode_hdr_maxsz + 1) +#define decode_access_maxsz (op_decode_hdr_maxsz + 2) #define encode_symlink_maxsz (op_encode_hdr_maxsz + \ 1 + nfs4_name_maxsz + \ 1 + \ nfs4_fattr_maxsz) #define decode_symlink_maxsz (op_decode_hdr_maxsz + 8) #define encode_create_maxsz (op_encode_hdr_maxsz + \ - 2 + nfs4_name_maxsz + \ - nfs4_fattr_maxsz) + 1 + 2 + nfs4_name_maxsz + \ + encode_attrs_maxsz) #define decode_create_maxsz (op_decode_hdr_maxsz + \ decode_change_info_maxsz + \ nfs4_fattr_bitmap_maxsz) +#define encode_statfs_maxsz (encode_getattr_maxsz) +#define decode_statfs_maxsz (decode_getattr_maxsz) #define encode_delegreturn_maxsz (op_encode_hdr_maxsz + 4) #define decode_delegreturn_maxsz (op_decode_hdr_maxsz) +#define encode_getacl_maxsz (encode_getattr_maxsz) +#define decode_getacl_maxsz (op_decode_hdr_maxsz + \ + nfs4_fattr_bitmap_maxsz + 1) +#define encode_setacl_maxsz (op_encode_hdr_maxsz + \ + encode_stateid_maxsz + 3) +#define decode_setacl_maxsz (decode_setattr_maxsz) #define encode_fs_locations_maxsz \ (encode_getattr_maxsz) #define decode_fs_locations_maxsz \ @@ -169,37 +242,37 @@ static int nfs4_stat_to_errno(int); #define NFS4_dec_compound_sz (1024) /* XXX: large enough? */ #define NFS4_enc_read_sz (compound_encode_hdr_maxsz + \ encode_putfh_maxsz + \ - op_encode_hdr_maxsz + 7) + encode_read_maxsz) #define NFS4_dec_read_sz (compound_decode_hdr_maxsz + \ decode_putfh_maxsz + \ - op_decode_hdr_maxsz + 2) + decode_read_maxsz) #define NFS4_enc_readlink_sz (compound_encode_hdr_maxsz + \ encode_putfh_maxsz + \ - op_encode_hdr_maxsz) + encode_readlink_maxsz) #define NFS4_dec_readlink_sz (compound_decode_hdr_maxsz + \ decode_putfh_maxsz + \ - op_decode_hdr_maxsz) + decode_readlink_maxsz) #define NFS4_enc_readdir_sz (compound_encode_hdr_maxsz + \ encode_putfh_maxsz + \ - op_encode_hdr_maxsz + 9) + encode_readdir_maxsz) #define NFS4_dec_readdir_sz (compound_decode_hdr_maxsz + \ decode_putfh_maxsz + \ - op_decode_hdr_maxsz + 2) + decode_readdir_maxsz) #define NFS4_enc_write_sz (compound_encode_hdr_maxsz + \ encode_putfh_maxsz + \ - op_encode_hdr_maxsz + 8 + \ + encode_write_maxsz + \ encode_getattr_maxsz) #define NFS4_dec_write_sz (compound_decode_hdr_maxsz + \ decode_putfh_maxsz + \ - op_decode_hdr_maxsz + 4 + \ + decode_write_maxsz + \ decode_getattr_maxsz) #define NFS4_enc_commit_sz (compound_encode_hdr_maxsz + \ encode_putfh_maxsz + \ - op_encode_hdr_maxsz + 3 + \ + encode_commit_maxsz + \ encode_getattr_maxsz) #define NFS4_dec_commit_sz (compound_decode_hdr_maxsz + \ decode_putfh_maxsz + \ - op_decode_hdr_maxsz + 2 + \ + decode_commit_maxsz + \ decode_getattr_maxsz) #define NFS4_enc_open_sz (compound_encode_hdr_maxsz + \ encode_putfh_maxsz + \ @@ -217,13 +290,14 @@ static int nfs4_stat_to_errno(int); decode_getattr_maxsz + \ decode_restorefh_maxsz + \ decode_getattr_maxsz) -#define NFS4_enc_open_confirm_sz \ - (compound_encode_hdr_maxsz + \ - encode_putfh_maxsz + \ - op_encode_hdr_maxsz + 5) -#define NFS4_dec_open_confirm_sz (compound_decode_hdr_maxsz + \ - decode_putfh_maxsz + \ - op_decode_hdr_maxsz + 4) +#define NFS4_enc_open_confirm_sz \ + (compound_encode_hdr_maxsz + \ + encode_putfh_maxsz + \ + encode_open_confirm_maxsz) +#define NFS4_dec_open_confirm_sz \ + (compound_decode_hdr_maxsz + \ + decode_putfh_maxsz + \ + decode_open_confirm_maxsz) #define NFS4_enc_open_noattr_sz (compound_encode_hdr_maxsz + \ encode_putfh_maxsz + \ encode_open_maxsz + \ @@ -234,31 +308,30 @@ static int nfs4_stat_to_errno(int); decode_getattr_maxsz) #define NFS4_enc_open_downgrade_sz \ (compound_encode_hdr_maxsz + \ - encode_putfh_maxsz + \ - op_encode_hdr_maxsz + 7 + \ - encode_getattr_maxsz) + encode_putfh_maxsz + \ + encode_open_downgrade_maxsz + \ + encode_getattr_maxsz) #define NFS4_dec_open_downgrade_sz \ (compound_decode_hdr_maxsz + \ - decode_putfh_maxsz + \ - op_decode_hdr_maxsz + 4 + \ - decode_getattr_maxsz) -#define NFS4_enc_close_sz (compound_encode_hdr_maxsz + \ - encode_putfh_maxsz + \ - op_encode_hdr_maxsz + 5 + \ - encode_getattr_maxsz) -#define NFS4_dec_close_sz (compound_decode_hdr_maxsz + \ - decode_putfh_maxsz + \ - op_decode_hdr_maxsz + 4 + \ - decode_getattr_maxsz) -#define NFS4_enc_setattr_sz (compound_encode_hdr_maxsz + \ - encode_putfh_maxsz + \ - op_encode_hdr_maxsz + 4 + \ - nfs4_fattr_maxsz + \ - encode_getattr_maxsz) -#define NFS4_dec_setattr_sz (compound_decode_hdr_maxsz + \ - decode_putfh_maxsz + \ - op_decode_hdr_maxsz + 3 + \ - nfs4_fattr_maxsz) + decode_putfh_maxsz + \ + decode_open_downgrade_maxsz + \ + decode_getattr_maxsz) +#define NFS4_enc_close_sz (compound_encode_hdr_maxsz + \ + encode_putfh_maxsz + \ + encode_close_maxsz + \ + encode_getattr_maxsz) +#define NFS4_dec_close_sz (compound_decode_hdr_maxsz + \ + decode_putfh_maxsz + \ + decode_close_maxsz + \ + decode_getattr_maxsz) +#define NFS4_enc_setattr_sz (compound_encode_hdr_maxsz + \ + encode_putfh_maxsz + \ + encode_setattr_maxsz + \ + encode_getattr_maxsz) +#define NFS4_dec_setattr_sz (compound_decode_hdr_maxsz + \ + decode_putfh_maxsz + \ + decode_setattr_maxsz + \ + decode_getattr_maxsz) #define NFS4_enc_fsinfo_sz (compound_encode_hdr_maxsz + \ encode_putfh_maxsz + \ encode_fsinfo_maxsz) @@ -285,39 +358,28 @@ static int nfs4_stat_to_errno(int); decode_fsinfo_maxsz) #define NFS4_enc_lock_sz (compound_encode_hdr_maxsz + \ encode_putfh_maxsz + \ - encode_getattr_maxsz + \ - op_encode_hdr_maxsz + \ - 1 + 1 + 2 + 2 + \ - 1 + 4 + 1 + 2 + \ - lock_owner_id_maxsz) + encode_lock_maxsz) #define NFS4_dec_lock_sz (compound_decode_hdr_maxsz + \ decode_putfh_maxsz + \ - decode_getattr_maxsz + \ - op_decode_hdr_maxsz + \ - 2 + 2 + 1 + 2 + \ - lock_owner_id_maxsz) + decode_lock_maxsz) #define NFS4_enc_lockt_sz (compound_encode_hdr_maxsz + \ encode_putfh_maxsz + \ - encode_getattr_maxsz + \ - op_encode_hdr_maxsz + \ - 1 + 2 + 2 + 2 + \ - lock_owner_id_maxsz) -#define NFS4_dec_lockt_sz (NFS4_dec_lock_sz) + encode_lockt_maxsz) +#define NFS4_dec_lockt_sz (compound_decode_hdr_maxsz + \ + decode_putfh_maxsz + \ + decode_lockt_maxsz) #define NFS4_enc_locku_sz (compound_encode_hdr_maxsz + \ encode_putfh_maxsz + \ - encode_getattr_maxsz + \ - op_encode_hdr_maxsz + \ - 1 + 1 + 4 + 2 + 2) + encode_locku_maxsz) #define NFS4_dec_locku_sz (compound_decode_hdr_maxsz + \ decode_putfh_maxsz + \ - decode_getattr_maxsz + \ - op_decode_hdr_maxsz + 4) + decode_locku_maxsz) #define NFS4_enc_access_sz (compound_encode_hdr_maxsz + \ encode_putfh_maxsz + \ - op_encode_hdr_maxsz + 1) + encode_access_maxsz) #define NFS4_dec_access_sz (compound_decode_hdr_maxsz + \ decode_putfh_maxsz + \ - op_decode_hdr_maxsz + 2) + decode_access_maxsz) #define NFS4_enc_getattr_sz (compound_encode_hdr_maxsz + \ encode_putfh_maxsz + \ encode_getattr_maxsz) @@ -416,10 +478,10 @@ static int nfs4_stat_to_errno(int); decode_getattr_maxsz) #define NFS4_enc_statfs_sz (compound_encode_hdr_maxsz + \ encode_putfh_maxsz + \ - encode_getattr_maxsz) + encode_statfs_maxsz) #define NFS4_dec_statfs_sz (compound_decode_hdr_maxsz + \ decode_putfh_maxsz + \ - op_decode_hdr_maxsz + 12) + decode_statfs_maxsz) #define NFS4_enc_server_caps_sz (compound_encode_hdr_maxsz + \ encode_putfh_maxsz + \ encode_getattr_maxsz) @@ -435,18 +497,16 @@ static int nfs4_stat_to_errno(int); decode_getattr_maxsz) #define NFS4_enc_getacl_sz (compound_encode_hdr_maxsz + \ encode_putfh_maxsz + \ - encode_getattr_maxsz) + encode_getacl_maxsz) #define NFS4_dec_getacl_sz (compound_decode_hdr_maxsz + \ decode_putfh_maxsz + \ - op_decode_hdr_maxsz + \ - nfs4_fattr_bitmap_maxsz + 1) + decode_getacl_maxsz) #define NFS4_enc_setacl_sz (compound_encode_hdr_maxsz + \ encode_putfh_maxsz + \ - op_encode_hdr_maxsz + 4 + \ - nfs4_fattr_bitmap_maxsz + 1) + encode_setacl_maxsz) #define NFS4_dec_setacl_sz (compound_decode_hdr_maxsz + \ decode_putfh_maxsz + \ - op_decode_hdr_maxsz + nfs4_fattr_bitmap_maxsz) + decode_setacl_maxsz) #define NFS4_enc_fs_locations_sz \ (compound_encode_hdr_maxsz + \ encode_putfh_maxsz + \ @@ -1108,12 +1168,10 @@ static int encode_read(struct xdr_stream *xdr, const struct nfs_readargs *args) static int encode_readdir(struct xdr_stream *xdr, const struct nfs4_readdir_arg *readdir, struct rpc_rqst *req) { - struct rpc_auth *auth = req->rq_task->tk_msg.rpc_cred->cr_auth; uint32_t attrs[2] = { FATTR4_WORD0_RDATTR_ERROR|FATTR4_WORD0_FILEID, FATTR4_WORD1_MOUNTED_ON_FILEID, }; - int replen; __be32 *p; RESERVE_SPACE(12+NFS4_VERIFIER_SIZE+20); @@ -1138,37 +1196,16 @@ static int encode_readdir(struct xdr_stream *xdr, const struct nfs4_readdir_arg attrs[0] & readdir->bitmask[0], attrs[1] & readdir->bitmask[1]); - /* set up reply kvec - * toplevel_status + taglen + rescount + OP_PUTFH + status - * + OP_READDIR + status + verifer(2) = 9 - */ - replen = (RPC_REPHDRSIZE + auth->au_rslack + 9) << 2; - xdr_inline_pages(&req->rq_rcv_buf, replen, readdir->pages, - readdir->pgbase, readdir->count); - dprintk("%s: inlined page args = (%u, %p, %u, %u)\n", - __FUNCTION__, replen, readdir->pages, - readdir->pgbase, readdir->count); - return 0; } static int encode_readlink(struct xdr_stream *xdr, const struct nfs4_readlink *readlink, struct rpc_rqst *req) { - struct rpc_auth *auth = req->rq_task->tk_msg.rpc_cred->cr_auth; - unsigned int replen; __be32 *p; RESERVE_SPACE(4); WRITE32(OP_READLINK); - /* set up reply kvec - * toplevel_status + taglen + rescount + OP_PUTFH + status - * + OP_READLINK + status + string length = 8 - */ - replen = (RPC_REPHDRSIZE + auth->au_rslack + 8) << 2; - xdr_inline_pages(&req->rq_rcv_buf, replen, readlink->pages, - readlink->pgbase, readlink->pglen); - return 0; } @@ -1398,7 +1435,7 @@ out: /* * Encode REMOVE request */ -static int nfs4_xdr_enc_remove(struct rpc_rqst *req, __be32 *p, const struct nfs4_remove_arg *args) +static int nfs4_xdr_enc_remove(struct rpc_rqst *req, __be32 *p, const struct nfs_removeargs *args) { struct xdr_stream xdr; struct compound_hdr hdr = { @@ -1410,7 +1447,7 @@ static int nfs4_xdr_enc_remove(struct rpc_rqst *req, __be32 *p, const struct nfs encode_compound_hdr(&xdr, &hdr); if ((status = encode_putfh(&xdr, args->fh)) != 0) goto out; - if ((status = encode_remove(&xdr, args->name)) != 0) + if ((status = encode_remove(&xdr, &args->name)) != 0) goto out; status = encode_getfattr(&xdr, args->bitmask); out: @@ -1734,6 +1771,8 @@ static int nfs4_xdr_enc_readlink(struct rpc_rqst *req, __be32 *p, const struct n struct compound_hdr hdr = { .nops = 2, }; + struct rpc_auth *auth = req->rq_task->tk_msg.rpc_cred->cr_auth; + unsigned int replen; int status; xdr_init_encode(&xdr, &req->rq_snd_buf, p); @@ -1742,6 +1781,15 @@ static int nfs4_xdr_enc_readlink(struct rpc_rqst *req, __be32 *p, const struct n if(status) goto out; status = encode_readlink(&xdr, args, req); + + /* set up reply kvec + * toplevel_status + taglen + rescount + OP_PUTFH + status + * + OP_READLINK + status + string length = 8 + */ + replen = (RPC_REPHDRSIZE + auth->au_rslack + NFS4_dec_readlink_sz) << 2; + xdr_inline_pages(&req->rq_rcv_buf, replen, args->pages, + args->pgbase, args->pglen); + out: return status; } @@ -1755,6 +1803,8 @@ static int nfs4_xdr_enc_readdir(struct rpc_rqst *req, __be32 *p, const struct nf struct compound_hdr hdr = { .nops = 2, }; + struct rpc_auth *auth = req->rq_task->tk_msg.rpc_cred->cr_auth; + int replen; int status; xdr_init_encode(&xdr, &req->rq_snd_buf, p); @@ -1763,6 +1813,18 @@ static int nfs4_xdr_enc_readdir(struct rpc_rqst *req, __be32 *p, const struct nf if(status) goto out; status = encode_readdir(&xdr, args, req); + + /* set up reply kvec + * toplevel_status + taglen + rescount + OP_PUTFH + status + * + OP_READDIR + status + verifer(2) = 9 + */ + replen = (RPC_REPHDRSIZE + auth->au_rslack + NFS4_dec_readdir_sz) << 2; + xdr_inline_pages(&req->rq_rcv_buf, replen, args->pages, + args->pgbase, args->count); + dprintk("%s: inlined page args = (%u, %p, %u, %u)\n", + __FUNCTION__, replen, args->pages, + args->pgbase, args->count); + out: return status; } @@ -3161,11 +3223,12 @@ static int decode_getfh(struct xdr_stream *xdr, struct nfs_fh *fh) uint32_t len; int status; + /* Zero handle first to allow comparisons */ + memset(fh, 0, sizeof(*fh)); + status = decode_op_hdr(xdr, OP_GETFH); if (status) return status; - /* Zero handle first to allow comparisons */ - memset(fh, 0, sizeof(*fh)); READ_BUF(4); READ32(len); @@ -3772,7 +3835,7 @@ out: /* * Decode REMOVE response */ -static int nfs4_xdr_dec_remove(struct rpc_rqst *rqstp, __be32 *p, struct nfs4_remove_res *res) +static int nfs4_xdr_dec_remove(struct rpc_rqst *rqstp, __be32 *p, struct nfs_removeres *res) { struct xdr_stream xdr; struct compound_hdr hdr; @@ -3785,7 +3848,7 @@ static int nfs4_xdr_dec_remove(struct rpc_rqst *rqstp, __be32 *p, struct nfs4_re goto out; if ((status = decode_remove(&xdr, &res->cinfo)) != 0) goto out; - decode_getfattr(&xdr, res->dir_attr, res->server); + decode_getfattr(&xdr, &res->dir_attr, res->server); out: return status; } @@ -4030,12 +4093,11 @@ static int nfs4_xdr_dec_open(struct rpc_rqst *rqstp, __be32 *p, struct nfs_openr status = decode_open(&xdr, res); if (status) goto out; - status = decode_getfh(&xdr, &res->fh); - if (status) + if (decode_getfh(&xdr, &res->fh) != 0) goto out; if (decode_getfattr(&xdr, res->f_attr, res->server) != 0) goto out; - if ((status = decode_restorefh(&xdr)) != 0) + if (decode_restorefh(&xdr) != 0) goto out; decode_getfattr(&xdr, res->dir_attr, res->server); out: diff --git a/fs/nfs/proc.c b/fs/nfs/proc.c index 7be0ee2782c..845cdde1d8b 100644 --- a/fs/nfs/proc.c +++ b/fs/nfs/proc.c @@ -272,14 +272,14 @@ nfs_proc_mknod(struct inode *dir, struct dentry *dentry, struct iattr *sattr, static int nfs_proc_remove(struct inode *dir, struct qstr *name) { - struct nfs_diropargs arg = { - .fh = NFS_FH(dir), - .name = name->name, - .len = name->len + struct nfs_removeargs arg = { + .fh = NFS_FH(dir), + .name.len = name->len, + .name.name = name->name, }; - struct rpc_message msg = { - .rpc_proc = &nfs_procedures[NFSPROC_REMOVE], - .rpc_argp = &arg, + struct rpc_message msg = { + .rpc_proc = &nfs_procedures[NFSPROC_REMOVE], + .rpc_argp = &arg, }; int status; @@ -291,32 +291,16 @@ nfs_proc_remove(struct inode *dir, struct qstr *name) return status; } -static int -nfs_proc_unlink_setup(struct rpc_message *msg, struct dentry *dir, struct qstr *name) +static void +nfs_proc_unlink_setup(struct rpc_message *msg, struct inode *dir) { - struct nfs_diropargs *arg; - - arg = kmalloc(sizeof(*arg), GFP_KERNEL); - if (!arg) - return -ENOMEM; - arg->fh = NFS_FH(dir->d_inode); - arg->name = name->name; - arg->len = name->len; msg->rpc_proc = &nfs_procedures[NFSPROC_REMOVE]; - msg->rpc_argp = arg; - return 0; } -static int -nfs_proc_unlink_done(struct dentry *dir, struct rpc_task *task) +static int nfs_proc_unlink_done(struct rpc_task *task, struct inode *dir) { - struct rpc_message *msg = &task->tk_msg; - - if (msg->rpc_argp) { - nfs_mark_for_revalidate(dir->d_inode); - kfree(msg->rpc_argp); - } - return 0; + nfs_mark_for_revalidate(dir); + return 1; } static int diff --git a/fs/nfs/super.c b/fs/nfs/super.c index adffe1615c5..b34b7a711d5 100644 --- a/fs/nfs/super.c +++ b/fs/nfs/super.c @@ -1685,6 +1685,9 @@ static int nfs4_validate_mount_data(struct nfs4_mount_data **options, dprintk("MNTPATH: %s\n", *mntpath); + if (args.client_address == NULL) + goto out_no_client_address; + *ip_addr = args.client_address; break; @@ -1705,6 +1708,10 @@ out_inval_auth: out_no_address: dfprintk(MOUNT, "NFS4: mount program didn't pass remote address\n"); return -EINVAL; + +out_no_client_address: + dfprintk(MOUNT, "NFS4: mount program didn't pass callback address\n"); + return -EINVAL; } /* diff --git a/fs/nfs/unlink.c b/fs/nfs/unlink.c index 0e28189c215..045ab805c17 100644 --- a/fs/nfs/unlink.c +++ b/fs/nfs/unlink.c @@ -3,7 +3,6 @@ * * nfs sillydelete handling * - * NOTE: we rely on holding the BKL for list manipulation protection. */ #include <linux/slab.h> @@ -15,46 +14,23 @@ struct nfs_unlinkdata { - struct nfs_unlinkdata *next; - struct dentry *dir, *dentry; - struct qstr name; - struct rpc_task task; + struct nfs_removeargs args; + struct nfs_removeres res; + struct inode *dir; struct rpc_cred *cred; - unsigned int count; }; -static struct nfs_unlinkdata *nfs_deletes; -static RPC_WAITQ(nfs_delete_queue, "nfs_delete_queue"); - -/** - * nfs_detach_unlinkdata - Remove asynchronous unlink from global list - * @data: pointer to descriptor - */ -static inline void -nfs_detach_unlinkdata(struct nfs_unlinkdata *data) -{ - struct nfs_unlinkdata **q; - - for (q = &nfs_deletes; *q != NULL; q = &((*q)->next)) { - if (*q == data) { - *q = data->next; - break; - } - } -} - /** - * nfs_put_unlinkdata - release data from a sillydelete operation. + * nfs_free_unlinkdata - release data from a sillydelete operation. * @data: pointer to unlink structure. */ static void -nfs_put_unlinkdata(struct nfs_unlinkdata *data) +nfs_free_unlinkdata(struct nfs_unlinkdata *data) { - if (--data->count == 0) { - nfs_detach_unlinkdata(data); - kfree(data->name.name); - kfree(data); - } + iput(data->dir); + put_rpccred(data->cred); + kfree(data->args.name.name); + kfree(data); } #define NAME_ALLOC_LEN(len) ((len+16) & ~15) @@ -63,50 +39,36 @@ nfs_put_unlinkdata(struct nfs_unlinkdata *data) * @dentry: pointer to dentry * @data: nfs_unlinkdata */ -static inline void -nfs_copy_dname(struct dentry *dentry, struct nfs_unlinkdata *data) +static int nfs_copy_dname(struct dentry *dentry, struct nfs_unlinkdata *data) { char *str; int len = dentry->d_name.len; - str = kmalloc(NAME_ALLOC_LEN(len), GFP_KERNEL); + str = kmemdup(dentry->d_name.name, NAME_ALLOC_LEN(len), GFP_KERNEL); if (!str) - return; - memcpy(str, dentry->d_name.name, len); - if (!data->name.len) { - data->name.len = len; - data->name.name = str; - } else - kfree(str); + return -ENOMEM; + data->args.name.len = len; + data->args.name.name = str; + return 0; } /** * nfs_async_unlink_init - Initialize the RPC info - * @task: rpc_task of the sillydelete - * - * We delay initializing RPC info until after the call to dentry_iput() - * in order to minimize races against rename(). + * task: rpc_task of the sillydelete */ static void nfs_async_unlink_init(struct rpc_task *task, void *calldata) { - struct nfs_unlinkdata *data = calldata; - struct dentry *dir = data->dir; - struct rpc_message msg = { - .rpc_cred = data->cred, + struct nfs_unlinkdata *data = calldata; + struct inode *dir = data->dir; + struct rpc_message msg = { + .rpc_argp = &data->args, + .rpc_resp = &data->res, + .rpc_cred = data->cred, }; - int status = -ENOENT; - - if (!data->name.len) - goto out_err; - status = NFS_PROTO(dir->d_inode)->unlink_setup(&msg, dir, &data->name); - if (status < 0) - goto out_err; - nfs_begin_data_update(dir->d_inode); + nfs_begin_data_update(dir); + NFS_PROTO(dir)->unlink_setup(&msg, dir); rpc_call_setup(task, &msg, 0); - return; - out_err: - rpc_exit(task, status); } /** @@ -117,19 +79,13 @@ static void nfs_async_unlink_init(struct rpc_task *task, void *calldata) */ static void nfs_async_unlink_done(struct rpc_task *task, void *calldata) { - struct nfs_unlinkdata *data = calldata; - struct dentry *dir = data->dir; - struct inode *dir_i; - - if (!dir) - return; - dir_i = dir->d_inode; - nfs_end_data_update(dir_i); - if (NFS_PROTO(dir_i)->unlink_done(dir, task)) - return; - put_rpccred(data->cred); - data->cred = NULL; - dput(dir); + struct nfs_unlinkdata *data = calldata; + struct inode *dir = data->dir; + + if (!NFS_PROTO(dir)->unlink_done(task, dir)) + rpc_restart_call(task); + else + nfs_end_data_update(dir); } /** @@ -142,7 +98,7 @@ static void nfs_async_unlink_done(struct rpc_task *task, void *calldata) static void nfs_async_unlink_release(void *calldata) { struct nfs_unlinkdata *data = calldata; - nfs_put_unlinkdata(data); + nfs_free_unlinkdata(data); } static const struct rpc_call_ops nfs_unlink_ops = { @@ -151,73 +107,94 @@ static const struct rpc_call_ops nfs_unlink_ops = { .rpc_release = nfs_async_unlink_release, }; +static int nfs_call_unlink(struct dentry *dentry, struct nfs_unlinkdata *data) +{ + struct rpc_task *task; + struct dentry *parent; + struct inode *dir; + + if (nfs_copy_dname(dentry, data) < 0) + goto out_free; + + parent = dget_parent(dentry); + if (parent == NULL) + goto out_free; + dir = igrab(parent->d_inode); + dput(parent); + if (dir == NULL) + goto out_free; + + data->dir = dir; + data->args.fh = NFS_FH(dir); + nfs_fattr_init(&data->res.dir_attr); + + task = rpc_run_task(NFS_CLIENT(dir), RPC_TASK_ASYNC, &nfs_unlink_ops, data); + if (!IS_ERR(task)) + rpc_put_task(task); + return 1; +out_free: + return 0; +} + /** * nfs_async_unlink - asynchronous unlinking of a file + * @dir: parent directory of dentry * @dentry: dentry to unlink */ int -nfs_async_unlink(struct dentry *dentry) +nfs_async_unlink(struct inode *dir, struct dentry *dentry) { - struct dentry *dir = dentry->d_parent; - struct nfs_unlinkdata *data; - struct rpc_clnt *clnt = NFS_CLIENT(dir->d_inode); - int status = -ENOMEM; + struct nfs_unlinkdata *data; + int status = -ENOMEM; data = kzalloc(sizeof(*data), GFP_KERNEL); - if (!data) + if (data == NULL) goto out; - data->cred = rpcauth_lookupcred(clnt->cl_auth, 0); + data->cred = rpcauth_lookupcred(NFS_CLIENT(dir)->cl_auth, 0); if (IS_ERR(data->cred)) { status = PTR_ERR(data->cred); goto out_free; } - data->dir = dget(dir); - data->dentry = dentry; - - data->next = nfs_deletes; - nfs_deletes = data; - data->count = 1; - - rpc_init_task(&data->task, clnt, RPC_TASK_ASYNC, &nfs_unlink_ops, data); + status = -EBUSY; spin_lock(&dentry->d_lock); + if (dentry->d_flags & DCACHE_NFSFS_RENAMED) + goto out_unlock; dentry->d_flags |= DCACHE_NFSFS_RENAMED; + dentry->d_fsdata = data; spin_unlock(&dentry->d_lock); - - rpc_sleep_on(&nfs_delete_queue, &data->task, NULL, NULL); - status = 0; - out: - return status; + return 0; +out_unlock: + spin_unlock(&dentry->d_lock); + put_rpccred(data->cred); out_free: kfree(data); +out: return status; } /** * nfs_complete_unlink - Initialize completion of the sillydelete * @dentry: dentry to delete + * @inode: inode * * Since we're most likely to be called by dentry_iput(), we * only use the dentry to find the sillydelete. We then copy the name * into the qstr. */ void -nfs_complete_unlink(struct dentry *dentry) +nfs_complete_unlink(struct dentry *dentry, struct inode *inode) { - struct nfs_unlinkdata *data; + struct nfs_unlinkdata *data = NULL; - for(data = nfs_deletes; data != NULL; data = data->next) { - if (dentry == data->dentry) - break; - } - if (!data) - return; - data->count++; - nfs_copy_dname(dentry, data); spin_lock(&dentry->d_lock); - dentry->d_flags &= ~DCACHE_NFSFS_RENAMED; + if (dentry->d_flags & DCACHE_NFSFS_RENAMED) { + dentry->d_flags &= ~DCACHE_NFSFS_RENAMED; + data = dentry->d_fsdata; + } spin_unlock(&dentry->d_lock); - rpc_wake_up_task(&data->task); - nfs_put_unlinkdata(data); + + if (data != NULL && (NFS_STALE(inode) || !nfs_call_unlink(dentry, data))) + nfs_free_unlinkdata(data); } diff --git a/fs/nfsctl.c b/fs/nfsctl.c index c043136a82c..51f1b31acbf 100644 --- a/fs/nfsctl.c +++ b/fs/nfsctl.c @@ -23,19 +23,15 @@ static struct file *do_open(char *name, int flags) { struct nameidata nd; + struct vfsmount *mnt; int error; - nd.mnt = do_kern_mount("nfsd", 0, "nfsd", NULL); + mnt = do_kern_mount("nfsd", 0, "nfsd", NULL); + if (IS_ERR(mnt)) + return (struct file *)mnt; - if (IS_ERR(nd.mnt)) - return (struct file *)nd.mnt; - - nd.dentry = dget(nd.mnt->mnt_root); - nd.last_type = LAST_ROOT; - nd.flags = 0; - nd.depth = 0; - - error = path_walk(name, &nd); + error = vfs_path_lookup(mnt->mnt_root, mnt, name, 0, &nd); + mntput(mnt); /* drop do_kern_mount reference */ if (error) return ERR_PTR(error); diff --git a/fs/nfsd/auth.c b/fs/nfsd/auth.c index cf61dc8ae94..21928056e35 100644 --- a/fs/nfsd/auth.c +++ b/fs/nfsd/auth.c @@ -9,10 +9,11 @@ #include <linux/sunrpc/svc.h> #include <linux/sunrpc/svcauth.h> #include <linux/nfsd/nfsd.h> +#include <linux/nfsd/export.h> #define CAP_NFSD_MASK (CAP_FS_MASK|CAP_TO_MASK(CAP_SYS_RESOURCE)) -static int nfsexp_flags(struct svc_rqst *rqstp, struct svc_export *exp) +int nfsexp_flags(struct svc_rqst *rqstp, struct svc_export *exp) { struct exp_flavor_info *f; struct exp_flavor_info *end = exp->ex_flavors + exp->ex_nflavors; diff --git a/fs/nfsd/export.c b/fs/nfsd/export.c index c7bbf460b00..6ab8de40904 100644 --- a/fs/nfsd/export.c +++ b/fs/nfsd/export.c @@ -1265,7 +1265,7 @@ struct svc_export * rqst_exp_get_by_name(struct svc_rqst *rqstp, struct vfsmount *mnt, struct dentry *dentry) { - struct svc_export *gssexp, *exp = NULL; + struct svc_export *gssexp, *exp = ERR_PTR(-ENOENT); if (rqstp->rq_client == NULL) goto gss; @@ -1288,7 +1288,7 @@ gss: &rqstp->rq_chandle); if (PTR_ERR(gssexp) == -ENOENT) return exp; - if (exp && !IS_ERR(exp)) + if (!IS_ERR(exp)) exp_put(exp); return gssexp; } @@ -1296,7 +1296,7 @@ gss: struct svc_export * rqst_exp_find(struct svc_rqst *rqstp, int fsid_type, u32 *fsidv) { - struct svc_export *gssexp, *exp = NULL; + struct svc_export *gssexp, *exp = ERR_PTR(-ENOENT); if (rqstp->rq_client == NULL) goto gss; @@ -1318,7 +1318,7 @@ gss: &rqstp->rq_chandle); if (PTR_ERR(gssexp) == -ENOENT) return exp; - if (exp && !IS_ERR(exp)) + if (!IS_ERR(exp)) exp_put(exp); return gssexp; } diff --git a/fs/nfsd/vfs.c b/fs/nfsd/vfs.c index e90f4a8a1d0..ee96a897a29 100644 --- a/fs/nfsd/vfs.c +++ b/fs/nfsd/vfs.c @@ -120,14 +120,14 @@ nfsd_cross_mnt(struct svc_rqst *rqstp, struct dentry **dpp, mntput(mnt); goto out; } - if (exp2 && ((exp->ex_flags & NFSEXP_CROSSMOUNT) || EX_NOHIDE(exp2))) { + if ((exp->ex_flags & NFSEXP_CROSSMOUNT) || EX_NOHIDE(exp2)) { /* successfully crossed mount point */ exp_put(exp); *expp = exp2; dput(dentry); *dpp = mounts; } else { - if (exp2) exp_put(exp2); + exp_put(exp2); dput(mounts); } mntput(mnt); @@ -1797,6 +1797,11 @@ nfsd_statfs(struct svc_rqst *rqstp, struct svc_fh *fhp, struct kstatfs *stat) return err; } +static int exp_rdonly(struct svc_rqst *rqstp, struct svc_export *exp) +{ + return nfsexp_flags(rqstp, exp) & NFSEXP_READONLY; +} + /* * Check for a user's access permissions to this inode. */ @@ -1833,7 +1838,7 @@ nfsd_permission(struct svc_rqst *rqstp, struct svc_export *exp, */ if (!(acc & MAY_LOCAL_ACCESS)) if (acc & (MAY_WRITE | MAY_SATTR | MAY_TRUNC)) { - if (EX_RDONLY(exp, rqstp) || IS_RDONLY(inode)) + if (exp_rdonly(rqstp, exp) || IS_RDONLY(inode)) return nfserr_rofs; if (/* (acc & MAY_WRITE) && */ IS_IMMUTABLE(inode)) return nfserr_perm; @@ -1916,7 +1921,7 @@ nfsd_racache_init(int cache_size) raparm_hash[i].pb_head = NULL; spin_lock_init(&raparm_hash[i].pb_lock); } - nperbucket = cache_size >> RAPARM_HASH_BITS; + nperbucket = DIV_ROUND_UP(cache_size, RAPARM_HASH_SIZE); for (i = 0; i < cache_size - 1; i++) { if (i % nperbucket == 0) raparm_hash[j++].pb_head = raparml + i; diff --git a/fs/ocfs2/aops.c b/fs/ocfs2/aops.c index 84bf6e79de2..460d440310f 100644 --- a/fs/ocfs2/aops.c +++ b/fs/ocfs2/aops.c @@ -232,7 +232,7 @@ static int ocfs2_readpage(struct file *file, struct page *page) * might now be discovering a truncate that hit on another node. * block_read_full_page->get_block freaks out if it is asked to read * beyond the end of a file, so we check here. Callers - * (generic_file_read, fault->nopage) are clever enough to check i_size + * (generic_file_read, vm_ops->fault) are clever enough to check i_size * and notice that the page they just read isn't needed. * * XXX sys_readahead() seems to get that wrong? diff --git a/fs/ocfs2/file.c b/fs/ocfs2/file.c index 004c2abbc73..5727cd18302 100644 --- a/fs/ocfs2/file.c +++ b/fs/ocfs2/file.c @@ -34,6 +34,7 @@ #include <linux/splice.h> #include <linux/mount.h> #include <linux/writeback.h> +#include <linux/falloc.h> #define MLOG_MASK_PREFIX ML_INODE #include <cluster/masklog.h> @@ -1504,30 +1505,19 @@ out: /* * Parts of this function taken from xfs_change_file_space() */ -int ocfs2_change_file_space(struct file *file, unsigned int cmd, - struct ocfs2_space_resv *sr) +static int __ocfs2_change_file_space(struct file *file, struct inode *inode, + loff_t f_pos, unsigned int cmd, + struct ocfs2_space_resv *sr, + int change_size) { int ret; s64 llen; - struct inode *inode = file->f_path.dentry->d_inode; + loff_t size; struct ocfs2_super *osb = OCFS2_SB(inode->i_sb); struct buffer_head *di_bh = NULL; handle_t *handle; unsigned long long max_off = ocfs2_max_file_offset(inode->i_sb->s_blocksize_bits); - if ((cmd == OCFS2_IOC_RESVSP || cmd == OCFS2_IOC_RESVSP64) && - !ocfs2_writes_unwritten_extents(osb)) - return -ENOTTY; - else if ((cmd == OCFS2_IOC_UNRESVSP || cmd == OCFS2_IOC_UNRESVSP64) && - !ocfs2_sparse_alloc(osb)) - return -ENOTTY; - - if (!S_ISREG(inode->i_mode)) - return -EINVAL; - - if (!(file->f_mode & FMODE_WRITE)) - return -EBADF; - if (ocfs2_is_hard_readonly(osb) || ocfs2_is_soft_readonly(osb)) return -EROFS; @@ -1557,7 +1547,7 @@ int ocfs2_change_file_space(struct file *file, unsigned int cmd, case 0: /*SEEK_SET*/ break; case 1: /*SEEK_CUR*/ - sr->l_start += file->f_pos; + sr->l_start += f_pos; break; case 2: /*SEEK_END*/ sr->l_start += i_size_read(inode); @@ -1577,6 +1567,7 @@ int ocfs2_change_file_space(struct file *file, unsigned int cmd, ret = -EINVAL; goto out_meta_unlock; } + size = sr->l_start + sr->l_len; if (cmd == OCFS2_IOC_RESVSP || cmd == OCFS2_IOC_RESVSP64) { if (sr->l_len <= 0) { @@ -1585,7 +1576,7 @@ int ocfs2_change_file_space(struct file *file, unsigned int cmd, } } - if (should_remove_suid(file->f_path.dentry)) { + if (file && should_remove_suid(file->f_path.dentry)) { ret = __ocfs2_write_remove_suid(inode, di_bh); if (ret) { mlog_errno(ret); @@ -1628,6 +1619,9 @@ int ocfs2_change_file_space(struct file *file, unsigned int cmd, goto out_meta_unlock; } + if (change_size && i_size_read(inode) < size) + i_size_write(inode, size); + inode->i_ctime = inode->i_mtime = CURRENT_TIME; ret = ocfs2_mark_inode_dirty(handle, inode, di_bh); if (ret < 0) @@ -1646,6 +1640,52 @@ out: return ret; } +int ocfs2_change_file_space(struct file *file, unsigned int cmd, + struct ocfs2_space_resv *sr) +{ + struct inode *inode = file->f_path.dentry->d_inode; + struct ocfs2_super *osb = OCFS2_SB(inode->i_sb);; + + if ((cmd == OCFS2_IOC_RESVSP || cmd == OCFS2_IOC_RESVSP64) && + !ocfs2_writes_unwritten_extents(osb)) + return -ENOTTY; + else if ((cmd == OCFS2_IOC_UNRESVSP || cmd == OCFS2_IOC_UNRESVSP64) && + !ocfs2_sparse_alloc(osb)) + return -ENOTTY; + + if (!S_ISREG(inode->i_mode)) + return -EINVAL; + + if (!(file->f_mode & FMODE_WRITE)) + return -EBADF; + + return __ocfs2_change_file_space(file, inode, file->f_pos, cmd, sr, 0); +} + +static long ocfs2_fallocate(struct inode *inode, int mode, loff_t offset, + loff_t len) +{ + struct ocfs2_super *osb = OCFS2_SB(inode->i_sb); + struct ocfs2_space_resv sr; + int change_size = 1; + + if (!ocfs2_writes_unwritten_extents(osb)) + return -EOPNOTSUPP; + + if (S_ISDIR(inode->i_mode)) + return -ENODEV; + + if (mode & FALLOC_FL_KEEP_SIZE) + change_size = 0; + + sr.l_whence = 0; + sr.l_start = (s64)offset; + sr.l_len = (s64)len; + + return __ocfs2_change_file_space(NULL, inode, offset, + OCFS2_IOC_RESVSP64, &sr, change_size); +} + static int ocfs2_prepare_inode_for_write(struct dentry *dentry, loff_t *ppos, size_t count, @@ -2312,6 +2352,7 @@ const struct inode_operations ocfs2_file_iops = { .setattr = ocfs2_setattr, .getattr = ocfs2_getattr, .permission = ocfs2_permission, + .fallocate = ocfs2_fallocate, }; const struct inode_operations ocfs2_special_file_iops = { diff --git a/fs/ocfs2/mmap.c b/fs/ocfs2/mmap.c index d79aa12137d..ee64749e2ee 100644 --- a/fs/ocfs2/mmap.c +++ b/fs/ocfs2/mmap.c @@ -60,31 +60,28 @@ static inline int ocfs2_vm_op_unblock_sigs(sigset_t *oldset) return sigprocmask(SIG_SETMASK, oldset, NULL); } -static struct page *ocfs2_nopage(struct vm_area_struct * area, - unsigned long address, - int *type) +static int ocfs2_fault(struct vm_area_struct *area, struct vm_fault *vmf) { - struct page *page = NOPAGE_SIGBUS; sigset_t blocked, oldset; - int ret; + int error, ret; - mlog_entry("(area=%p, address=%lu, type=%p)\n", area, address, - type); + mlog_entry("(area=%p, page offset=%lu)\n", area, vmf->pgoff); - ret = ocfs2_vm_op_block_sigs(&blocked, &oldset); - if (ret < 0) { - mlog_errno(ret); + error = ocfs2_vm_op_block_sigs(&blocked, &oldset); + if (error < 0) { + mlog_errno(error); + ret = VM_FAULT_SIGBUS; goto out; } - page = filemap_nopage(area, address, type); + ret = filemap_fault(area, vmf); - ret = ocfs2_vm_op_unblock_sigs(&oldset); - if (ret < 0) - mlog_errno(ret); + error = ocfs2_vm_op_unblock_sigs(&oldset); + if (error < 0) + mlog_errno(error); out: - mlog_exit_ptr(page); - return page; + mlog_exit_ptr(vmf->page); + return ret; } static int __ocfs2_page_mkwrite(struct inode *inode, struct buffer_head *di_bh, @@ -209,7 +206,7 @@ out: } static struct vm_operations_struct ocfs2_file_vm_ops = { - .nopage = ocfs2_nopage, + .fault = ocfs2_fault, .page_mkwrite = ocfs2_page_mkwrite, }; @@ -226,6 +223,7 @@ int ocfs2_mmap(struct file *file, struct vm_area_struct *vma) ocfs2_meta_unlock(file->f_dentry->d_inode, lock_level); out: vma->vm_ops = &ocfs2_file_vm_ops; + vma->vm_flags |= VM_CAN_NONLINEAR; return 0; } diff --git a/fs/partitions/check.c b/fs/partitions/check.c index 98e0b85a9bb..783c57ec07d 100644 --- a/fs/partitions/check.c +++ b/fs/partitions/check.c @@ -372,11 +372,10 @@ void add_partition(struct gendisk *disk, int part, sector_t start, sector_t len, { struct hd_struct *p; - p = kmalloc(sizeof(*p), GFP_KERNEL); + p = kzalloc(sizeof(*p), GFP_KERNEL); if (!p) return; - memset(p, 0, sizeof(*p)); p->start_sect = start; p->nr_sects = len; p->partno = part; diff --git a/fs/proc/base.c b/fs/proc/base.c index 42cb4f5613b..3c77d5a64e7 100644 --- a/fs/proc/base.c +++ b/fs/proc/base.c @@ -72,6 +72,7 @@ #include <linux/poll.h> #include <linux/nsproxy.h> #include <linux/oom.h> +#include <linux/elf.h> #include "internal.h" /* NOTE: @@ -1014,7 +1015,7 @@ static int task_dumpable(struct task_struct *task) task_lock(task); mm = task->mm; if (mm) - dumpable = mm->dumpable; + dumpable = get_dumpable(mm); task_unlock(task); if(dumpable == 1) return 1; @@ -1785,6 +1786,91 @@ static const struct inode_operations proc_attr_dir_inode_operations = { #endif +#if defined(USE_ELF_CORE_DUMP) && defined(CONFIG_ELF_CORE) +static ssize_t proc_coredump_filter_read(struct file *file, char __user *buf, + size_t count, loff_t *ppos) +{ + struct task_struct *task = get_proc_task(file->f_dentry->d_inode); + struct mm_struct *mm; + char buffer[PROC_NUMBUF]; + size_t len; + int ret; + + if (!task) + return -ESRCH; + + ret = 0; + mm = get_task_mm(task); + if (mm) { + len = snprintf(buffer, sizeof(buffer), "%08lx\n", + ((mm->flags & MMF_DUMP_FILTER_MASK) >> + MMF_DUMP_FILTER_SHIFT)); + mmput(mm); + ret = simple_read_from_buffer(buf, count, ppos, buffer, len); + } + + put_task_struct(task); + + return ret; +} + +static ssize_t proc_coredump_filter_write(struct file *file, + const char __user *buf, + size_t count, + loff_t *ppos) +{ + struct task_struct *task; + struct mm_struct *mm; + char buffer[PROC_NUMBUF], *end; + unsigned int val; + int ret; + int i; + unsigned long mask; + + ret = -EFAULT; + memset(buffer, 0, sizeof(buffer)); + if (count > sizeof(buffer) - 1) + count = sizeof(buffer) - 1; + if (copy_from_user(buffer, buf, count)) + goto out_no_task; + + ret = -EINVAL; + val = (unsigned int)simple_strtoul(buffer, &end, 0); + if (*end == '\n') + end++; + if (end - buffer == 0) + goto out_no_task; + + ret = -ESRCH; + task = get_proc_task(file->f_dentry->d_inode); + if (!task) + goto out_no_task; + + ret = end - buffer; + mm = get_task_mm(task); + if (!mm) + goto out_no_mm; + + for (i = 0, mask = 1; i < MMF_DUMP_FILTER_BITS; i++, mask <<= 1) { + if (val & mask) + set_bit(i + MMF_DUMP_FILTER_SHIFT, &mm->flags); + else + clear_bit(i + MMF_DUMP_FILTER_SHIFT, &mm->flags); + } + + mmput(mm); + out_no_mm: + put_task_struct(task); + out_no_task: + return ret; +} + +static const struct file_operations proc_coredump_filter_operations = { + .read = proc_coredump_filter_read, + .write = proc_coredump_filter_write, +}; +#endif + /* * /proc/self: */ @@ -2005,6 +2091,9 @@ static const struct pid_entry tgid_base_stuff[] = { #ifdef CONFIG_FAULT_INJECTION REG("make-it-fail", S_IRUGO|S_IWUSR, fault_inject), #endif +#if defined(USE_ELF_CORE_DUMP) && defined(CONFIG_ELF_CORE) + REG("coredump_filter", S_IRUGO|S_IWUSR, coredump_filter), +#endif #ifdef CONFIG_TASK_IO_ACCOUNTING INF("io", S_IRUGO, pid_io_accounting), #endif diff --git a/fs/proc/proc_misc.c b/fs/proc/proc_misc.c index d24b8d46059..f133afebed7 100644 --- a/fs/proc/proc_misc.c +++ b/fs/proc/proc_misc.c @@ -445,6 +445,11 @@ static int show_stat(struct seq_file *p, void *v) cputime64_t user, nice, system, idle, iowait, irq, softirq, steal; u64 sum = 0; struct timespec boottime; + unsigned int *per_irq_sum; + + per_irq_sum = kzalloc(sizeof(unsigned int)*NR_IRQS, GFP_KERNEL); + if (!per_irq_sum) + return -ENOMEM; user = nice = system = idle = iowait = irq = softirq = steal = cputime64_zero; @@ -462,8 +467,11 @@ static int show_stat(struct seq_file *p, void *v) irq = cputime64_add(irq, kstat_cpu(i).cpustat.irq); softirq = cputime64_add(softirq, kstat_cpu(i).cpustat.softirq); steal = cputime64_add(steal, kstat_cpu(i).cpustat.steal); - for (j = 0 ; j < NR_IRQS ; j++) - sum += kstat_cpu(i).irqs[j]; + for (j = 0; j < NR_IRQS; j++) { + unsigned int temp = kstat_cpu(i).irqs[j]; + sum += temp; + per_irq_sum[j] += temp; + } } seq_printf(p, "cpu %llu %llu %llu %llu %llu %llu %llu %llu\n", @@ -501,7 +509,7 @@ static int show_stat(struct seq_file *p, void *v) #if !defined(CONFIG_PPC64) && !defined(CONFIG_ALPHA) && !defined(CONFIG_IA64) for (i = 0; i < NR_IRQS; i++) - seq_printf(p, " %u", kstat_irqs(i)); + seq_printf(p, " %u", per_irq_sum[i]); #endif seq_printf(p, @@ -516,6 +524,7 @@ static int show_stat(struct seq_file *p, void *v) nr_running(), nr_iowait()); + kfree(per_irq_sum); return 0; } diff --git a/fs/splice.c b/fs/splice.c index 53fc2082a46..22496d2a73f 100644 --- a/fs/splice.c +++ b/fs/splice.c @@ -265,7 +265,7 @@ __generic_file_splice_read(struct file *in, loff_t *ppos, unsigned int flags) { struct address_space *mapping = in->f_mapping; - unsigned int loff, nr_pages; + unsigned int loff, nr_pages, req_pages; struct page *pages[PIPE_BUFFERS]; struct partial_page partial[PIPE_BUFFERS]; struct page *page; @@ -281,28 +281,24 @@ __generic_file_splice_read(struct file *in, loff_t *ppos, index = *ppos >> PAGE_CACHE_SHIFT; loff = *ppos & ~PAGE_CACHE_MASK; - nr_pages = (len + loff + PAGE_CACHE_SIZE - 1) >> PAGE_CACHE_SHIFT; - - if (nr_pages > PIPE_BUFFERS) - nr_pages = PIPE_BUFFERS; - - /* - * Don't try to 2nd guess the read-ahead logic, call into - * page_cache_readahead() like the page cache reads would do. - */ - page_cache_readahead(mapping, &in->f_ra, in, index, nr_pages); + req_pages = (len + loff + PAGE_CACHE_SIZE - 1) >> PAGE_CACHE_SHIFT; + nr_pages = min(req_pages, (unsigned)PIPE_BUFFERS); /* * Lookup the (hopefully) full range of pages we need. */ spd.nr_pages = find_get_pages_contig(mapping, index, nr_pages, pages); + index += spd.nr_pages; /* * If find_get_pages_contig() returned fewer pages than we needed, - * allocate the rest and fill in the holes. + * readahead/allocate the rest and fill in the holes. */ + if (spd.nr_pages < nr_pages) + page_cache_sync_readahead(mapping, &in->f_ra, in, + index, req_pages - spd.nr_pages); + error = 0; - index += spd.nr_pages; while (spd.nr_pages < nr_pages) { /* * Page could be there, find_get_pages_contig() breaks on @@ -311,12 +307,6 @@ __generic_file_splice_read(struct file *in, loff_t *ppos, page = find_get_page(mapping, index); if (!page) { /* - * Make sure the read-ahead engine is notified - * about this failure. - */ - handle_ra_miss(mapping, &in->f_ra, index); - - /* * page didn't exist, allocate one. */ page = page_cache_alloc_cold(mapping); @@ -361,6 +351,10 @@ __generic_file_splice_read(struct file *in, loff_t *ppos, this_len = min_t(unsigned long, len, PAGE_CACHE_SIZE - loff); page = pages[page_nr]; + if (PageReadahead(page)) + page_cache_async_readahead(mapping, &in->f_ra, in, + page, index, req_pages - page_nr); + /* * If the page isn't uptodate, we may need to start io on it */ @@ -453,6 +447,7 @@ fill_it: */ while (page_nr < nr_pages) page_cache_release(pages[page_nr++]); + in->f_ra.prev_index = index; if (spd.nr_pages) return splice_to_pipe(pipe, &spd); diff --git a/fs/udf/balloc.c b/fs/udf/balloc.c index 4cec9101568..ef48d094dd2 100644 --- a/fs/udf/balloc.c +++ b/fs/udf/balloc.c @@ -41,18 +41,17 @@ #define uint(x) xuint(x) #define xuint(x) __le ## x -static inline int find_next_one_bit (void * addr, int size, int offset) +static inline int find_next_one_bit(void *addr, int size, int offset) { - uintBPL_t * p = ((uintBPL_t *) addr) + (offset / BITS_PER_LONG); - int result = offset & ~(BITS_PER_LONG-1); + uintBPL_t *p = ((uintBPL_t *) addr) + (offset / BITS_PER_LONG); + int result = offset & ~(BITS_PER_LONG - 1); unsigned long tmp; if (offset >= size) return size; size -= result; - offset &= (BITS_PER_LONG-1); - if (offset) - { + offset &= (BITS_PER_LONG - 1); + if (offset) { tmp = leBPL_to_cpup(p++); tmp &= ~0UL << offset; if (size < BITS_PER_LONG) @@ -62,8 +61,7 @@ static inline int find_next_one_bit (void * addr, int size, int offset) size -= BITS_PER_LONG; result += BITS_PER_LONG; } - while (size & ~(BITS_PER_LONG-1)) - { + while (size & ~(BITS_PER_LONG - 1)) { if ((tmp = leBPL_to_cpup(p++))) goto found_middle; result += BITS_PER_LONG; @@ -72,17 +70,18 @@ static inline int find_next_one_bit (void * addr, int size, int offset) if (!size) return result; tmp = leBPL_to_cpup(p); -found_first: - tmp &= ~0UL >> (BITS_PER_LONG-size); -found_middle: + found_first: + tmp &= ~0UL >> (BITS_PER_LONG - size); + found_middle: return result + ffz(~tmp); } #define find_first_one_bit(addr, size)\ find_next_one_bit((addr), (size), 0) -static int read_block_bitmap(struct super_block * sb, - struct udf_bitmap *bitmap, unsigned int block, unsigned long bitmap_nr) +static int read_block_bitmap(struct super_block *sb, + struct udf_bitmap *bitmap, unsigned int block, + unsigned long bitmap_nr) { struct buffer_head *bh = NULL; int retval = 0; @@ -92,38 +91,39 @@ static int read_block_bitmap(struct super_block * sb, loc.partitionReferenceNum = UDF_SB_PARTITION(sb); bh = udf_tread(sb, udf_get_lb_pblock(sb, loc, block)); - if (!bh) - { + if (!bh) { retval = -EIO; } bitmap->s_block_bitmap[bitmap_nr] = bh; return retval; } -static int __load_block_bitmap(struct super_block * sb, - struct udf_bitmap *bitmap, unsigned int block_group) +static int __load_block_bitmap(struct super_block *sb, + struct udf_bitmap *bitmap, + unsigned int block_group) { int retval = 0; int nr_groups = bitmap->s_nr_groups; - if (block_group >= nr_groups) - { - udf_debug("block_group (%d) > nr_groups (%d)\n", block_group, nr_groups); + if (block_group >= nr_groups) { + udf_debug("block_group (%d) > nr_groups (%d)\n", block_group, + nr_groups); } if (bitmap->s_block_bitmap[block_group]) return block_group; - else - { - retval = read_block_bitmap(sb, bitmap, block_group, block_group); + else { + retval = + read_block_bitmap(sb, bitmap, block_group, block_group); if (retval < 0) return retval; return block_group; } } -static inline int load_block_bitmap(struct super_block * sb, - struct udf_bitmap *bitmap, unsigned int block_group) +static inline int load_block_bitmap(struct super_block *sb, + struct udf_bitmap *bitmap, + unsigned int block_group) { int slot; @@ -138,13 +138,14 @@ static inline int load_block_bitmap(struct super_block * sb, return slot; } -static void udf_bitmap_free_blocks(struct super_block * sb, - struct inode * inode, - struct udf_bitmap *bitmap, - kernel_lb_addr bloc, uint32_t offset, uint32_t count) +static void udf_bitmap_free_blocks(struct super_block *sb, + struct inode *inode, + struct udf_bitmap *bitmap, + kernel_lb_addr bloc, uint32_t offset, + uint32_t count) { struct udf_sb_info *sbi = UDF_SB(sb); - struct buffer_head * bh = NULL; + struct buffer_head *bh = NULL; unsigned long block; unsigned long block_group; unsigned long bit; @@ -154,17 +155,22 @@ static void udf_bitmap_free_blocks(struct super_block * sb, mutex_lock(&sbi->s_alloc_mutex); if (bloc.logicalBlockNum < 0 || - (bloc.logicalBlockNum + count) > UDF_SB_PARTLEN(sb, bloc.partitionReferenceNum)) - { - udf_debug("%d < %d || %d + %d > %d\n", - bloc.logicalBlockNum, 0, bloc.logicalBlockNum, count, - UDF_SB_PARTLEN(sb, bloc.partitionReferenceNum)); + (bloc.logicalBlockNum + count) > UDF_SB_PARTLEN(sb, + bloc. + partitionReferenceNum)) + { + udf_debug("%d < %d || %d + %d > %d\n", bloc.logicalBlockNum, 0, + bloc.logicalBlockNum, count, UDF_SB_PARTLEN(sb, + bloc. + partitionReferenceNum)); goto error_return; } - block = bloc.logicalBlockNum + offset + (sizeof(struct spaceBitmapDesc) << 3); + block = + bloc.logicalBlockNum + offset + + (sizeof(struct spaceBitmapDesc) << 3); -do_more: + do_more: overflow = 0; block_group = block >> (sb->s_blocksize_bits + 3); bit = block % (sb->s_blocksize << 3); @@ -172,8 +178,7 @@ do_more: /* * Check to see if we are freeing blocks across a group boundary. */ - if (bit + count > (sb->s_blocksize << 3)) - { + if (bit + count > (sb->s_blocksize << 3)) { overflow = bit + count - (sb->s_blocksize << 3); count -= overflow; } @@ -182,32 +187,31 @@ do_more: goto error_return; bh = bitmap->s_block_bitmap[bitmap_nr]; - for (i=0; i < count; i++) - { - if (udf_set_bit(bit + i, bh->b_data)) - { + for (i = 0; i < count; i++) { + if (udf_set_bit(bit + i, bh->b_data)) { udf_debug("bit %ld already set\n", bit + i); - udf_debug("byte=%2x\n", ((char *)bh->b_data)[(bit + i) >> 3]); - } - else - { + udf_debug("byte=%2x\n", + ((char *)bh->b_data)[(bit + i) >> 3]); + } else { if (inode) DQUOT_FREE_BLOCK(inode, 1); - if (UDF_SB_LVIDBH(sb)) - { - UDF_SB_LVID(sb)->freeSpaceTable[UDF_SB_PARTITION(sb)] = - cpu_to_le32(le32_to_cpu(UDF_SB_LVID(sb)->freeSpaceTable[UDF_SB_PARTITION(sb)])+1); + if (UDF_SB_LVIDBH(sb)) { + UDF_SB_LVID(sb)-> + freeSpaceTable[UDF_SB_PARTITION(sb)] = + cpu_to_le32(le32_to_cpu + (UDF_SB_LVID(sb)-> + freeSpaceTable[UDF_SB_PARTITION + (sb)]) + 1); } } } mark_buffer_dirty(bh); - if (overflow) - { + if (overflow) { block += count; count = overflow; goto do_more; } -error_return: + error_return: sb->s_dirt = 1; if (UDF_SB_LVIDBH(sb)) mark_buffer_dirty(UDF_SB_LVIDBH(sb)); @@ -215,10 +219,11 @@ error_return: return; } -static int udf_bitmap_prealloc_blocks(struct super_block * sb, - struct inode * inode, - struct udf_bitmap *bitmap, uint16_t partition, uint32_t first_block, - uint32_t block_count) +static int udf_bitmap_prealloc_blocks(struct super_block *sb, + struct inode *inode, + struct udf_bitmap *bitmap, + uint16_t partition, uint32_t first_block, + uint32_t block_count) { struct udf_sb_info *sbi = UDF_SB(sb); int alloc_count = 0; @@ -233,9 +238,10 @@ static int udf_bitmap_prealloc_blocks(struct super_block * sb, if (first_block + block_count > UDF_SB_PARTLEN(sb, partition)) block_count = UDF_SB_PARTLEN(sb, partition) - first_block; -repeat: + repeat: nr_groups = (UDF_SB_PARTLEN(sb, partition) + - (sizeof(struct spaceBitmapDesc) << 3) + (sb->s_blocksize * 8) - 1) / (sb->s_blocksize * 8); + (sizeof(struct spaceBitmapDesc) << 3) + + (sb->s_blocksize * 8) - 1) / (sb->s_blocksize * 8); block = first_block + (sizeof(struct spaceBitmapDesc) << 3); block_group = block >> (sb->s_blocksize_bits + 3); group_start = block_group ? 0 : sizeof(struct spaceBitmapDesc); @@ -247,31 +253,30 @@ repeat: bit = block % (sb->s_blocksize << 3); - while (bit < (sb->s_blocksize << 3) && block_count > 0) - { + while (bit < (sb->s_blocksize << 3) && block_count > 0) { if (!udf_test_bit(bit, bh->b_data)) goto out; else if (DQUOT_PREALLOC_BLOCK(inode, 1)) goto out; - else if (!udf_clear_bit(bit, bh->b_data)) - { + else if (!udf_clear_bit(bit, bh->b_data)) { udf_debug("bit already cleared for block %d\n", bit); DQUOT_FREE_BLOCK(inode, 1); goto out; } - block_count --; - alloc_count ++; - bit ++; - block ++; + block_count--; + alloc_count++; + bit++; + block++; } mark_buffer_dirty(bh); if (block_count > 0) goto repeat; -out: - if (UDF_SB_LVIDBH(sb)) - { + out: + if (UDF_SB_LVIDBH(sb)) { UDF_SB_LVID(sb)->freeSpaceTable[partition] = - cpu_to_le32(le32_to_cpu(UDF_SB_LVID(sb)->freeSpaceTable[partition])-alloc_count); + cpu_to_le32(le32_to_cpu + (UDF_SB_LVID(sb)->freeSpaceTable[partition]) - + alloc_count); mark_buffer_dirty(UDF_SB_LVIDBH(sb)); } sb->s_dirt = 1; @@ -279,12 +284,13 @@ out: return alloc_count; } -static int udf_bitmap_new_block(struct super_block * sb, - struct inode * inode, - struct udf_bitmap *bitmap, uint16_t partition, uint32_t goal, int *err) +static int udf_bitmap_new_block(struct super_block *sb, + struct inode *inode, + struct udf_bitmap *bitmap, uint16_t partition, + uint32_t goal, int *err) { struct udf_sb_info *sbi = UDF_SB(sb); - int newbit, bit=0, block, block_group, group_start; + int newbit, bit = 0, block, block_group, group_start; int end_goal, nr_groups, bitmap_nr, i; struct buffer_head *bh = NULL; char *ptr; @@ -293,7 +299,7 @@ static int udf_bitmap_new_block(struct super_block * sb, *err = -ENOSPC; mutex_lock(&sbi->s_alloc_mutex); -repeat: + repeat: if (goal < 0 || goal >= UDF_SB_PARTLEN(sb, partition)) goal = 0; @@ -306,38 +312,39 @@ repeat: if (bitmap_nr < 0) goto error_return; bh = bitmap->s_block_bitmap[bitmap_nr]; - ptr = memscan((char *)bh->b_data + group_start, 0xFF, sb->s_blocksize - group_start); + ptr = + memscan((char *)bh->b_data + group_start, 0xFF, + sb->s_blocksize - group_start); - if ((ptr - ((char *)bh->b_data)) < sb->s_blocksize) - { + if ((ptr - ((char *)bh->b_data)) < sb->s_blocksize) { bit = block % (sb->s_blocksize << 3); - if (udf_test_bit(bit, bh->b_data)) - { + if (udf_test_bit(bit, bh->b_data)) { goto got_block; } end_goal = (bit + 63) & ~63; bit = udf_find_next_one_bit(bh->b_data, end_goal, bit); if (bit < end_goal) goto got_block; - ptr = memscan((char *)bh->b_data + (bit >> 3), 0xFF, sb->s_blocksize - ((bit + 7) >> 3)); + ptr = + memscan((char *)bh->b_data + (bit >> 3), 0xFF, + sb->s_blocksize - ((bit + 7) >> 3)); newbit = (ptr - ((char *)bh->b_data)) << 3; - if (newbit < sb->s_blocksize << 3) - { + if (newbit < sb->s_blocksize << 3) { bit = newbit; goto search_back; } - newbit = udf_find_next_one_bit(bh->b_data, sb->s_blocksize << 3, bit); - if (newbit < sb->s_blocksize << 3) - { + newbit = + udf_find_next_one_bit(bh->b_data, sb->s_blocksize << 3, + bit); + if (newbit < sb->s_blocksize << 3) { bit = newbit; goto got_block; } } - for (i=0; i<(nr_groups*2); i++) - { - block_group ++; + for (i = 0; i < (nr_groups * 2); i++) { + block_group++; if (block_group >= nr_groups) block_group = 0; group_start = block_group ? 0 : sizeof(struct spaceBitmapDesc); @@ -346,67 +353,69 @@ repeat: if (bitmap_nr < 0) goto error_return; bh = bitmap->s_block_bitmap[bitmap_nr]; - if (i < nr_groups) - { - ptr = memscan((char *)bh->b_data + group_start, 0xFF, sb->s_blocksize - group_start); - if ((ptr - ((char *)bh->b_data)) < sb->s_blocksize) - { + if (i < nr_groups) { + ptr = + memscan((char *)bh->b_data + group_start, 0xFF, + sb->s_blocksize - group_start); + if ((ptr - ((char *)bh->b_data)) < sb->s_blocksize) { bit = (ptr - ((char *)bh->b_data)) << 3; break; } - } - else - { - bit = udf_find_next_one_bit((char *)bh->b_data, sb->s_blocksize << 3, group_start << 3); + } else { + bit = + udf_find_next_one_bit((char *)bh->b_data, + sb->s_blocksize << 3, + group_start << 3); if (bit < sb->s_blocksize << 3) break; } } - if (i >= (nr_groups*2)) - { + if (i >= (nr_groups * 2)) { mutex_unlock(&sbi->s_alloc_mutex); return newblock; } if (bit < sb->s_blocksize << 3) goto search_back; else - bit = udf_find_next_one_bit(bh->b_data, sb->s_blocksize << 3, group_start << 3); - if (bit >= sb->s_blocksize << 3) - { + bit = + udf_find_next_one_bit(bh->b_data, sb->s_blocksize << 3, + group_start << 3); + if (bit >= sb->s_blocksize << 3) { mutex_unlock(&sbi->s_alloc_mutex); return 0; } -search_back: - for (i=0; i<7 && bit > (group_start << 3) && udf_test_bit(bit - 1, bh->b_data); i++, bit--); + search_back: + for (i = 0; + i < 7 && bit > (group_start << 3) + && udf_test_bit(bit - 1, bh->b_data); i++, bit--) ; -got_block: + got_block: /* * Check quota for allocation of this block. */ - if (inode && DQUOT_ALLOC_BLOCK(inode, 1)) - { + if (inode && DQUOT_ALLOC_BLOCK(inode, 1)) { mutex_unlock(&sbi->s_alloc_mutex); *err = -EDQUOT; return 0; } newblock = bit + (block_group << (sb->s_blocksize_bits + 3)) - - (sizeof(struct spaceBitmapDesc) << 3); + (sizeof(struct spaceBitmapDesc) << 3); - if (!udf_clear_bit(bit, bh->b_data)) - { + if (!udf_clear_bit(bit, bh->b_data)) { udf_debug("bit already cleared for block %d\n", bit); goto repeat; } mark_buffer_dirty(bh); - if (UDF_SB_LVIDBH(sb)) - { + if (UDF_SB_LVIDBH(sb)) { UDF_SB_LVID(sb)->freeSpaceTable[partition] = - cpu_to_le32(le32_to_cpu(UDF_SB_LVID(sb)->freeSpaceTable[partition])-1); + cpu_to_le32(le32_to_cpu + (UDF_SB_LVID(sb)->freeSpaceTable[partition]) - + 1); mark_buffer_dirty(UDF_SB_LVIDBH(sb)); } sb->s_dirt = 1; @@ -414,16 +423,17 @@ got_block: *err = 0; return newblock; -error_return: + error_return: *err = -EIO; mutex_unlock(&sbi->s_alloc_mutex); return 0; } -static void udf_table_free_blocks(struct super_block * sb, - struct inode * inode, - struct inode * table, - kernel_lb_addr bloc, uint32_t offset, uint32_t count) +static void udf_table_free_blocks(struct super_block *sb, + struct inode *inode, + struct inode *table, + kernel_lb_addr bloc, uint32_t offset, + uint32_t count) { struct udf_sb_info *sbi = UDF_SB(sb); uint32_t start, end; @@ -435,11 +445,14 @@ static void udf_table_free_blocks(struct super_block * sb, mutex_lock(&sbi->s_alloc_mutex); if (bloc.logicalBlockNum < 0 || - (bloc.logicalBlockNum + count) > UDF_SB_PARTLEN(sb, bloc.partitionReferenceNum)) - { - udf_debug("%d < %d || %d + %d > %d\n", - bloc.logicalBlockNum, 0, bloc.logicalBlockNum, count, - UDF_SB_PARTLEN(sb, bloc.partitionReferenceNum)); + (bloc.logicalBlockNum + count) > UDF_SB_PARTLEN(sb, + bloc. + partitionReferenceNum)) + { + udf_debug("%d < %d || %d + %d > %d\n", bloc.logicalBlockNum, 0, + bloc.logicalBlockNum, count, UDF_SB_PARTLEN(sb, + bloc. + partitionReferenceNum)); goto error_return; } @@ -447,10 +460,11 @@ static void udf_table_free_blocks(struct super_block * sb, but.. oh well */ if (inode) DQUOT_FREE_BLOCK(inode, count); - if (UDF_SB_LVIDBH(sb)) - { + if (UDF_SB_LVIDBH(sb)) { UDF_SB_LVID(sb)->freeSpaceTable[UDF_SB_PARTITION(sb)] = - cpu_to_le32(le32_to_cpu(UDF_SB_LVID(sb)->freeSpaceTable[UDF_SB_PARTITION(sb)])+count); + cpu_to_le32(le32_to_cpu + (UDF_SB_LVID(sb)-> + freeSpaceTable[UDF_SB_PARTITION(sb)]) + count); mark_buffer_dirty(UDF_SB_LVIDBH(sb)); } @@ -463,73 +477,75 @@ static void udf_table_free_blocks(struct super_block * sb, epos.bh = oepos.bh = NULL; while (count && (etype = - udf_next_aext(table, &epos, &eloc, &elen, 1)) != -1) - { + udf_next_aext(table, &epos, &eloc, &elen, 1)) != -1) { if (((eloc.logicalBlockNum + (elen >> sb->s_blocksize_bits)) == - start)) - { - if ((0x3FFFFFFF - elen) < (count << sb->s_blocksize_bits)) - { - count -= ((0x3FFFFFFF - elen) >> sb->s_blocksize_bits); - start += ((0x3FFFFFFF - elen) >> sb->s_blocksize_bits); - elen = (etype << 30) | (0x40000000 - sb->s_blocksize); - } - else - { + start)) { + if ((0x3FFFFFFF - elen) < + (count << sb->s_blocksize_bits)) { + count -= + ((0x3FFFFFFF - + elen) >> sb->s_blocksize_bits); + start += + ((0x3FFFFFFF - + elen) >> sb->s_blocksize_bits); + elen = + (etype << 30) | (0x40000000 - + sb->s_blocksize); + } else { elen = (etype << 30) | - (elen + (count << sb->s_blocksize_bits)); + (elen + (count << sb->s_blocksize_bits)); start += count; count = 0; } udf_write_aext(table, &oepos, eloc, elen, 1); - } - else if (eloc.logicalBlockNum == (end + 1)) - { - if ((0x3FFFFFFF - elen) < (count << sb->s_blocksize_bits)) - { - count -= ((0x3FFFFFFF - elen) >> sb->s_blocksize_bits); - end -= ((0x3FFFFFFF - elen) >> sb->s_blocksize_bits); + } else if (eloc.logicalBlockNum == (end + 1)) { + if ((0x3FFFFFFF - elen) < + (count << sb->s_blocksize_bits)) { + count -= + ((0x3FFFFFFF - + elen) >> sb->s_blocksize_bits); + end -= + ((0x3FFFFFFF - + elen) >> sb->s_blocksize_bits); eloc.logicalBlockNum -= - ((0x3FFFFFFF - elen) >> sb->s_blocksize_bits); - elen = (etype << 30) | (0x40000000 - sb->s_blocksize); - } - else - { + ((0x3FFFFFFF - + elen) >> sb->s_blocksize_bits); + elen = + (etype << 30) | (0x40000000 - + sb->s_blocksize); + } else { eloc.logicalBlockNum = start; elen = (etype << 30) | - (elen + (count << sb->s_blocksize_bits)); + (elen + (count << sb->s_blocksize_bits)); end -= count; count = 0; } udf_write_aext(table, &oepos, eloc, elen, 1); } - if (epos.bh != oepos.bh) - { + if (epos.bh != oepos.bh) { i = -1; oepos.block = epos.block; brelse(oepos.bh); get_bh(epos.bh); oepos.bh = epos.bh; oepos.offset = 0; - } - else + } else oepos.offset = epos.offset; } - if (count) - { + if (count) { /* NOTE: we CANNOT use udf_add_aext here, as it can try to allocate - a new block, and since we hold the super block lock already - very bad things would happen :) + a new block, and since we hold the super block lock already + very bad things would happen :) - We copy the behavior of udf_add_aext, but instead of - trying to allocate a new block close to the existing one, - we just steal a block from the extent we are trying to add. + We copy the behavior of udf_add_aext, but instead of + trying to allocate a new block close to the existing one, + we just steal a block from the extent we are trying to add. - It would be nice if the blocks were close together, but it - isn't required. - */ + It would be nice if the blocks were close together, but it + isn't required. + */ int adsize; short_ad *sad = NULL; @@ -537,121 +553,124 @@ static void udf_table_free_blocks(struct super_block * sb, struct allocExtDesc *aed; eloc.logicalBlockNum = start; - elen = EXT_RECORDED_ALLOCATED | - (count << sb->s_blocksize_bits); + elen = EXT_RECORDED_ALLOCATED | (count << sb->s_blocksize_bits); if (UDF_I_ALLOCTYPE(table) == ICBTAG_FLAG_AD_SHORT) adsize = sizeof(short_ad); else if (UDF_I_ALLOCTYPE(table) == ICBTAG_FLAG_AD_LONG) adsize = sizeof(long_ad); - else - { + else { brelse(oepos.bh); brelse(epos.bh); goto error_return; } - if (epos.offset + (2 * adsize) > sb->s_blocksize) - { + if (epos.offset + (2 * adsize) > sb->s_blocksize) { char *sptr, *dptr; int loffset; - + brelse(oepos.bh); oepos = epos; /* Steal a block from the extent being free'd */ epos.block.logicalBlockNum = eloc.logicalBlockNum; - eloc.logicalBlockNum ++; + eloc.logicalBlockNum++; elen -= sb->s_blocksize; if (!(epos.bh = udf_tread(sb, - udf_get_lb_pblock(sb, epos.block, 0)))) - { + udf_get_lb_pblock(sb, + epos.block, + 0)))) { brelse(oepos.bh); goto error_return; } aed = (struct allocExtDesc *)(epos.bh->b_data); - aed->previousAllocExtLocation = cpu_to_le32(oepos.block.logicalBlockNum); - if (epos.offset + adsize > sb->s_blocksize) - { + aed->previousAllocExtLocation = + cpu_to_le32(oepos.block.logicalBlockNum); + if (epos.offset + adsize > sb->s_blocksize) { loffset = epos.offset; aed->lengthAllocDescs = cpu_to_le32(adsize); sptr = UDF_I_DATA(inode) + epos.offset - - udf_file_entry_alloc_offset(inode) + - UDF_I_LENEATTR(inode) - adsize; - dptr = epos.bh->b_data + sizeof(struct allocExtDesc); + udf_file_entry_alloc_offset(inode) + + UDF_I_LENEATTR(inode) - adsize; + dptr = + epos.bh->b_data + + sizeof(struct allocExtDesc); memcpy(dptr, sptr, adsize); - epos.offset = sizeof(struct allocExtDesc) + adsize; - } - else - { + epos.offset = + sizeof(struct allocExtDesc) + adsize; + } else { loffset = epos.offset + adsize; aed->lengthAllocDescs = cpu_to_le32(0); sptr = oepos.bh->b_data + epos.offset; epos.offset = sizeof(struct allocExtDesc); - if (oepos.bh) - { - aed = (struct allocExtDesc *)oepos.bh->b_data; + if (oepos.bh) { + aed = + (struct allocExtDesc *)oepos.bh-> + b_data; aed->lengthAllocDescs = - cpu_to_le32(le32_to_cpu(aed->lengthAllocDescs) + adsize); - } - else - { + cpu_to_le32(le32_to_cpu + (aed-> + lengthAllocDescs) + + adsize); + } else { UDF_I_LENALLOC(table) += adsize; mark_inode_dirty(table); } } if (UDF_SB_UDFREV(sb) >= 0x0200) - udf_new_tag(epos.bh->b_data, TAG_IDENT_AED, 3, 1, - epos.block.logicalBlockNum, sizeof(tag)); + udf_new_tag(epos.bh->b_data, TAG_IDENT_AED, 3, + 1, epos.block.logicalBlockNum, + sizeof(tag)); else - udf_new_tag(epos.bh->b_data, TAG_IDENT_AED, 2, 1, - epos.block.logicalBlockNum, sizeof(tag)); - switch (UDF_I_ALLOCTYPE(table)) - { - case ICBTAG_FLAG_AD_SHORT: + udf_new_tag(epos.bh->b_data, TAG_IDENT_AED, 2, + 1, epos.block.logicalBlockNum, + sizeof(tag)); + switch (UDF_I_ALLOCTYPE(table)) { + case ICBTAG_FLAG_AD_SHORT: { - sad = (short_ad *)sptr; - sad->extLength = cpu_to_le32( - EXT_NEXT_EXTENT_ALLOCDECS | - sb->s_blocksize); - sad->extPosition = cpu_to_le32(epos.block.logicalBlockNum); + sad = (short_ad *) sptr; + sad->extLength = + cpu_to_le32 + (EXT_NEXT_EXTENT_ALLOCDECS | sb-> + s_blocksize); + sad->extPosition = + cpu_to_le32(epos.block. + logicalBlockNum); break; } - case ICBTAG_FLAG_AD_LONG: + case ICBTAG_FLAG_AD_LONG: { - lad = (long_ad *)sptr; - lad->extLength = cpu_to_le32( - EXT_NEXT_EXTENT_ALLOCDECS | - sb->s_blocksize); - lad->extLocation = cpu_to_lelb(epos.block); + lad = (long_ad *) sptr; + lad->extLength = + cpu_to_le32 + (EXT_NEXT_EXTENT_ALLOCDECS | sb-> + s_blocksize); + lad->extLocation = + cpu_to_lelb(epos.block); break; } } - if (oepos.bh) - { + if (oepos.bh) { udf_update_tag(oepos.bh->b_data, loffset); mark_buffer_dirty(oepos.bh); - } - else + } else mark_inode_dirty(table); } - if (elen) /* It's possible that stealing the block emptied the extent */ - { + if (elen) { /* It's possible that stealing the block emptied the extent */ udf_write_aext(table, &epos, eloc, elen, 1); - if (!epos.bh) - { + if (!epos.bh) { UDF_I_LENALLOC(table) += adsize; mark_inode_dirty(table); - } - else - { + } else { aed = (struct allocExtDesc *)epos.bh->b_data; aed->lengthAllocDescs = - cpu_to_le32(le32_to_cpu(aed->lengthAllocDescs) + adsize); + cpu_to_le32(le32_to_cpu + (aed->lengthAllocDescs) + + adsize); udf_update_tag(epos.bh->b_data, epos.offset); mark_buffer_dirty(epos.bh); } @@ -661,16 +680,16 @@ static void udf_table_free_blocks(struct super_block * sb, brelse(epos.bh); brelse(oepos.bh); -error_return: + error_return: sb->s_dirt = 1; mutex_unlock(&sbi->s_alloc_mutex); return; } -static int udf_table_prealloc_blocks(struct super_block * sb, - struct inode * inode, - struct inode *table, uint16_t partition, uint32_t first_block, - uint32_t block_count) +static int udf_table_prealloc_blocks(struct super_block *sb, + struct inode *inode, + struct inode *table, uint16_t partition, + uint32_t first_block, uint32_t block_count) { struct udf_sb_info *sbi = UDF_SB(sb); int alloc_count = 0; @@ -696,39 +715,46 @@ static int udf_table_prealloc_blocks(struct super_block * sb, eloc.logicalBlockNum = 0xFFFFFFFF; while (first_block != eloc.logicalBlockNum && (etype = - udf_next_aext(table, &epos, &eloc, &elen, 1)) != -1) - { + udf_next_aext(table, + &epos, + &eloc, + &elen, + 1)) != + -1) { udf_debug("eloc=%d, elen=%d, first_block=%d\n", - eloc.logicalBlockNum, elen, first_block); - ; /* empty loop body */ + eloc.logicalBlockNum, elen, first_block); + ; /* empty loop body */ } - if (first_block == eloc.logicalBlockNum) - { + if (first_block == eloc.logicalBlockNum) { epos.offset -= adsize; alloc_count = (elen >> sb->s_blocksize_bits); - if (inode && DQUOT_PREALLOC_BLOCK(inode, alloc_count > block_count ? block_count : alloc_count)) + if (inode + && DQUOT_PREALLOC_BLOCK(inode, + alloc_count > + block_count ? block_count : + alloc_count)) alloc_count = 0; - else if (alloc_count > block_count) - { + else if (alloc_count > block_count) { alloc_count = block_count; eloc.logicalBlockNum += alloc_count; elen -= (alloc_count << sb->s_blocksize_bits); - udf_write_aext(table, &epos, eloc, (etype << 30) | elen, 1); - } - else - udf_delete_aext(table, epos, eloc, (etype << 30) | elen); - } - else + udf_write_aext(table, &epos, eloc, (etype << 30) | elen, + 1); + } else + udf_delete_aext(table, epos, eloc, + (etype << 30) | elen); + } else alloc_count = 0; brelse(epos.bh); - if (alloc_count && UDF_SB_LVIDBH(sb)) - { + if (alloc_count && UDF_SB_LVIDBH(sb)) { UDF_SB_LVID(sb)->freeSpaceTable[partition] = - cpu_to_le32(le32_to_cpu(UDF_SB_LVID(sb)->freeSpaceTable[partition])-alloc_count); + cpu_to_le32(le32_to_cpu + (UDF_SB_LVID(sb)->freeSpaceTable[partition]) - + alloc_count); mark_buffer_dirty(UDF_SB_LVIDBH(sb)); sb->s_dirt = 1; } @@ -736,9 +762,10 @@ static int udf_table_prealloc_blocks(struct super_block * sb, return alloc_count; } -static int udf_table_new_block(struct super_block * sb, - struct inode * inode, - struct inode *table, uint16_t partition, uint32_t goal, int *err) +static int udf_table_new_block(struct super_block *sb, + struct inode *inode, + struct inode *table, uint16_t partition, + uint32_t goal, int *err) { struct udf_sb_info *sbi = UDF_SB(sb); uint32_t spread = 0xFFFFFFFF, nspread = 0xFFFFFFFF; @@ -765,30 +792,27 @@ static int udf_table_new_block(struct super_block * sb, we stop. Otherwise we keep going till we run out of extents. We store the buffer_head, bloc, and extoffset of the current closest match and use that when we are done. - */ + */ epos.offset = sizeof(struct unallocSpaceEntry); epos.block = UDF_I_LOCATION(table); epos.bh = goal_epos.bh = NULL; while (spread && (etype = - udf_next_aext(table, &epos, &eloc, &elen, 1)) != -1) - { - if (goal >= eloc.logicalBlockNum) - { - if (goal < eloc.logicalBlockNum + (elen >> sb->s_blocksize_bits)) + udf_next_aext(table, &epos, &eloc, &elen, 1)) != -1) { + if (goal >= eloc.logicalBlockNum) { + if (goal < + eloc.logicalBlockNum + + (elen >> sb->s_blocksize_bits)) nspread = 0; else nspread = goal - eloc.logicalBlockNum - - (elen >> sb->s_blocksize_bits); - } - else + (elen >> sb->s_blocksize_bits); + } else nspread = eloc.logicalBlockNum - goal; - if (nspread < spread) - { + if (nspread < spread) { spread = nspread; - if (goal_epos.bh != epos.bh) - { + if (goal_epos.bh != epos.bh) { brelse(goal_epos.bh); goal_epos.bh = epos.bh; get_bh(goal_epos.bh); @@ -802,8 +826,7 @@ static int udf_table_new_block(struct super_block * sb, brelse(epos.bh); - if (spread == 0xFFFFFFFF) - { + if (spread == 0xFFFFFFFF) { brelse(goal_epos.bh); mutex_unlock(&sbi->s_alloc_mutex); return 0; @@ -815,11 +838,10 @@ static int udf_table_new_block(struct super_block * sb, /* This works, but very poorly.... */ newblock = goal_eloc.logicalBlockNum; - goal_eloc.logicalBlockNum ++; + goal_eloc.logicalBlockNum++; goal_elen -= sb->s_blocksize; - if (inode && DQUOT_ALLOC_BLOCK(inode, 1)) - { + if (inode && DQUOT_ALLOC_BLOCK(inode, 1)) { brelse(goal_epos.bh); mutex_unlock(&sbi->s_alloc_mutex); *err = -EDQUOT; @@ -832,10 +854,11 @@ static int udf_table_new_block(struct super_block * sb, udf_delete_aext(table, goal_epos, goal_eloc, goal_elen); brelse(goal_epos.bh); - if (UDF_SB_LVIDBH(sb)) - { + if (UDF_SB_LVIDBH(sb)) { UDF_SB_LVID(sb)->freeSpaceTable[partition] = - cpu_to_le32(le32_to_cpu(UDF_SB_LVID(sb)->freeSpaceTable[partition])-1); + cpu_to_le32(le32_to_cpu + (UDF_SB_LVID(sb)->freeSpaceTable[partition]) - + 1); mark_buffer_dirty(UDF_SB_LVIDBH(sb)); } @@ -845,105 +868,99 @@ static int udf_table_new_block(struct super_block * sb, return newblock; } -inline void udf_free_blocks(struct super_block * sb, - struct inode * inode, - kernel_lb_addr bloc, uint32_t offset, uint32_t count) +inline void udf_free_blocks(struct super_block *sb, + struct inode *inode, + kernel_lb_addr bloc, uint32_t offset, + uint32_t count) { uint16_t partition = bloc.partitionReferenceNum; - if (UDF_SB_PARTFLAGS(sb, partition) & UDF_PART_FLAG_UNALLOC_BITMAP) - { + if (UDF_SB_PARTFLAGS(sb, partition) & UDF_PART_FLAG_UNALLOC_BITMAP) { return udf_bitmap_free_blocks(sb, inode, - UDF_SB_PARTMAPS(sb)[partition].s_uspace.s_bitmap, - bloc, offset, count); - } - else if (UDF_SB_PARTFLAGS(sb, partition) & UDF_PART_FLAG_UNALLOC_TABLE) - { + UDF_SB_PARTMAPS(sb)[partition]. + s_uspace.s_bitmap, bloc, offset, + count); + } else if (UDF_SB_PARTFLAGS(sb, partition) & + UDF_PART_FLAG_UNALLOC_TABLE) { return udf_table_free_blocks(sb, inode, - UDF_SB_PARTMAPS(sb)[partition].s_uspace.s_table, - bloc, offset, count); - } - else if (UDF_SB_PARTFLAGS(sb, partition) & UDF_PART_FLAG_FREED_BITMAP) - { + UDF_SB_PARTMAPS(sb)[partition]. + s_uspace.s_table, bloc, offset, + count); + } else if (UDF_SB_PARTFLAGS(sb, partition) & UDF_PART_FLAG_FREED_BITMAP) { return udf_bitmap_free_blocks(sb, inode, - UDF_SB_PARTMAPS(sb)[partition].s_fspace.s_bitmap, - bloc, offset, count); - } - else if (UDF_SB_PARTFLAGS(sb, partition) & UDF_PART_FLAG_FREED_TABLE) - { + UDF_SB_PARTMAPS(sb)[partition]. + s_fspace.s_bitmap, bloc, offset, + count); + } else if (UDF_SB_PARTFLAGS(sb, partition) & UDF_PART_FLAG_FREED_TABLE) { return udf_table_free_blocks(sb, inode, - UDF_SB_PARTMAPS(sb)[partition].s_fspace.s_table, - bloc, offset, count); - } - else + UDF_SB_PARTMAPS(sb)[partition]. + s_fspace.s_table, bloc, offset, + count); + } else return; } -inline int udf_prealloc_blocks(struct super_block * sb, - struct inode * inode, - uint16_t partition, uint32_t first_block, uint32_t block_count) +inline int udf_prealloc_blocks(struct super_block *sb, + struct inode *inode, + uint16_t partition, uint32_t first_block, + uint32_t block_count) { - if (UDF_SB_PARTFLAGS(sb, partition) & UDF_PART_FLAG_UNALLOC_BITMAP) - { + if (UDF_SB_PARTFLAGS(sb, partition) & UDF_PART_FLAG_UNALLOC_BITMAP) { return udf_bitmap_prealloc_blocks(sb, inode, - UDF_SB_PARTMAPS(sb)[partition].s_uspace.s_bitmap, - partition, first_block, block_count); - } - else if (UDF_SB_PARTFLAGS(sb, partition) & UDF_PART_FLAG_UNALLOC_TABLE) - { + UDF_SB_PARTMAPS(sb) + [partition].s_uspace.s_bitmap, + partition, first_block, + block_count); + } else if (UDF_SB_PARTFLAGS(sb, partition) & + UDF_PART_FLAG_UNALLOC_TABLE) { return udf_table_prealloc_blocks(sb, inode, - UDF_SB_PARTMAPS(sb)[partition].s_uspace.s_table, - partition, first_block, block_count); - } - else if (UDF_SB_PARTFLAGS(sb, partition) & UDF_PART_FLAG_FREED_BITMAP) - { + UDF_SB_PARTMAPS(sb)[partition]. + s_uspace.s_table, partition, + first_block, block_count); + } else if (UDF_SB_PARTFLAGS(sb, partition) & UDF_PART_FLAG_FREED_BITMAP) { return udf_bitmap_prealloc_blocks(sb, inode, - UDF_SB_PARTMAPS(sb)[partition].s_fspace.s_bitmap, - partition, first_block, block_count); - } - else if (UDF_SB_PARTFLAGS(sb, partition) & UDF_PART_FLAG_FREED_TABLE) - { + UDF_SB_PARTMAPS(sb) + [partition].s_fspace.s_bitmap, + partition, first_block, + block_count); + } else if (UDF_SB_PARTFLAGS(sb, partition) & UDF_PART_FLAG_FREED_TABLE) { return udf_table_prealloc_blocks(sb, inode, - UDF_SB_PARTMAPS(sb)[partition].s_fspace.s_table, - partition, first_block, block_count); - } - else + UDF_SB_PARTMAPS(sb)[partition]. + s_fspace.s_table, partition, + first_block, block_count); + } else return 0; } -inline int udf_new_block(struct super_block * sb, - struct inode * inode, - uint16_t partition, uint32_t goal, int *err) +inline int udf_new_block(struct super_block *sb, + struct inode *inode, + uint16_t partition, uint32_t goal, int *err) { int ret; - if (UDF_SB_PARTFLAGS(sb, partition) & UDF_PART_FLAG_UNALLOC_BITMAP) - { + if (UDF_SB_PARTFLAGS(sb, partition) & UDF_PART_FLAG_UNALLOC_BITMAP) { ret = udf_bitmap_new_block(sb, inode, - UDF_SB_PARTMAPS(sb)[partition].s_uspace.s_bitmap, - partition, goal, err); + UDF_SB_PARTMAPS(sb)[partition]. + s_uspace.s_bitmap, partition, goal, + err); return ret; - } - else if (UDF_SB_PARTFLAGS(sb, partition) & UDF_PART_FLAG_UNALLOC_TABLE) - { + } else if (UDF_SB_PARTFLAGS(sb, partition) & + UDF_PART_FLAG_UNALLOC_TABLE) { return udf_table_new_block(sb, inode, - UDF_SB_PARTMAPS(sb)[partition].s_uspace.s_table, - partition, goal, err); - } - else if (UDF_SB_PARTFLAGS(sb, partition) & UDF_PART_FLAG_FREED_BITMAP) - { + UDF_SB_PARTMAPS(sb)[partition]. + s_uspace.s_table, partition, goal, + err); + } else if (UDF_SB_PARTFLAGS(sb, partition) & UDF_PART_FLAG_FREED_BITMAP) { return udf_bitmap_new_block(sb, inode, - UDF_SB_PARTMAPS(sb)[partition].s_fspace.s_bitmap, - partition, goal, err); - } - else if (UDF_SB_PARTFLAGS(sb, partition) & UDF_PART_FLAG_FREED_TABLE) - { + UDF_SB_PARTMAPS(sb)[partition]. + s_fspace.s_bitmap, partition, goal, + err); + } else if (UDF_SB_PARTFLAGS(sb, partition) & UDF_PART_FLAG_FREED_TABLE) { return udf_table_new_block(sb, inode, - UDF_SB_PARTMAPS(sb)[partition].s_fspace.s_table, - partition, goal, err); - } - else - { + UDF_SB_PARTMAPS(sb)[partition]. + s_fspace.s_table, partition, goal, + err); + } else { *err = -EIO; return 0; } diff --git a/fs/udf/crc.c b/fs/udf/crc.c index ef2bfaa19d7..ae3d4979094 100644 --- a/fs/udf/crc.c +++ b/fs/udf/crc.c @@ -79,8 +79,7 @@ static uint16_t crc_table[256] = { * July 21, 1997 - Andrew E. Mileski * Adapted from OSTA-UDF(tm) 1.50 standard. */ -uint16_t -udf_crc(uint8_t *data, uint32_t size, uint16_t crc) +uint16_t udf_crc(uint8_t * data, uint32_t size, uint16_t crc) { while (size--) crc = crc_table[(crc >> 8 ^ *(data++)) & 0xffU] ^ (crc << 8); @@ -112,7 +111,7 @@ int main(void) return 0; } -#endif /* defined(TEST) */ +#endif /* defined(TEST) */ /****************************************************************************/ #if defined(GENERATE) @@ -138,7 +137,7 @@ int main(int argc, char **argv) /* Get the polynomial */ sscanf(argv[1], "%lo", &poly); - if (poly & 0xffff0000U){ + if (poly & 0xffff0000U) { fprintf(stderr, "polynomial is too large\en"); exit(1); } @@ -147,22 +146,22 @@ int main(int argc, char **argv) /* Create a table */ printf("static unsigned short crc_table[256] = {\n"); - for (n = 0; n < 256; n++){ + for (n = 0; n < 256; n++) { if (n % 8 == 0) printf("\t"); crc = n << 8; - for (i = 0; i < 8; i++){ - if(crc & 0x8000U) + for (i = 0; i < 8; i++) { + if (crc & 0x8000U) crc = (crc << 1) ^ poly; else crc <<= 1; - crc &= 0xFFFFU; + crc &= 0xFFFFU; } if (n == 255) printf("0x%04xU ", crc); else printf("0x%04xU, ", crc); - if(n % 8 == 7) + if (n % 8 == 7) printf("\n"); } printf("};\n"); @@ -170,4 +169,4 @@ int main(int argc, char **argv) return 0; } -#endif /* defined(GENERATE) */ +#endif /* defined(GENERATE) */ diff --git a/fs/udf/dir.c b/fs/udf/dir.c index e45f86b5e7b..79bab9fe120 100644 --- a/fs/udf/dir.c +++ b/fs/udf/dir.c @@ -43,10 +43,10 @@ static int do_udf_readdir(struct inode *, struct file *, filldir_t, void *); /* readdir and lookup functions */ const struct file_operations udf_dir_operations = { - .read = generic_read_dir, - .readdir = udf_readdir, - .ioctl = udf_ioctl, - .fsync = udf_fsync_file, + .read = generic_read_dir, + .readdir = udf_readdir, + .ioctl = udf_ioctl, + .fsync = udf_fsync_file, }; /* @@ -82,26 +82,26 @@ int udf_readdir(struct file *filp, void *dirent, filldir_t filldir) lock_kernel(); - if ( filp->f_pos == 0 ) - { - if (filldir(dirent, ".", 1, filp->f_pos, dir->i_ino, DT_DIR) < 0) - { + if (filp->f_pos == 0) { + if (filldir(dirent, ".", 1, filp->f_pos, dir->i_ino, DT_DIR) < + 0) { unlock_kernel(); return 0; } - filp->f_pos ++; + filp->f_pos++; } result = do_udf_readdir(dir, filp, filldir, dirent); unlock_kernel(); - return result; + return result; } -static int -do_udf_readdir(struct inode * dir, struct file *filp, filldir_t filldir, void *dirent) +static int +do_udf_readdir(struct inode *dir, struct file *filp, filldir_t filldir, + void *dirent) { struct udf_fileident_bh fibh; - struct fileIdentDesc *fi=NULL; + struct fileIdentDesc *fi = NULL; struct fileIdentDesc cfi; int block, iblock; loff_t nf_pos = filp->f_pos - 1; @@ -117,7 +117,7 @@ do_udf_readdir(struct inode * dir, struct file *filp, filldir_t filldir, void *d sector_t offset; int i, num; unsigned int dt_type; - struct extent_position epos = { NULL, 0, {0, 0}}; + struct extent_position epos = { NULL, 0, {0, 0} }; if (nf_pos >= size) return 0; @@ -125,65 +125,61 @@ do_udf_readdir(struct inode * dir, struct file *filp, filldir_t filldir, void *d if (nf_pos == 0) nf_pos = (udf_ext0_offset(dir) >> 2); - fibh.soffset = fibh.eoffset = (nf_pos & ((dir->i_sb->s_blocksize - 1) >> 2)) << 2; + fibh.soffset = fibh.eoffset = + (nf_pos & ((dir->i_sb->s_blocksize - 1) >> 2)) << 2; if (UDF_I_ALLOCTYPE(dir) == ICBTAG_FLAG_AD_IN_ICB) fibh.sbh = fibh.ebh = NULL; else if (inode_bmap(dir, nf_pos >> (dir->i_sb->s_blocksize_bits - 2), - &epos, &eloc, &elen, &offset) == (EXT_RECORDED_ALLOCATED >> 30)) - { + &epos, &eloc, &elen, + &offset) == (EXT_RECORDED_ALLOCATED >> 30)) { block = udf_get_lb_pblock(dir->i_sb, eloc, offset); - if ((++offset << dir->i_sb->s_blocksize_bits) < elen) - { + if ((++offset << dir->i_sb->s_blocksize_bits) < elen) { if (UDF_I_ALLOCTYPE(dir) == ICBTAG_FLAG_AD_SHORT) epos.offset -= sizeof(short_ad); else if (UDF_I_ALLOCTYPE(dir) == ICBTAG_FLAG_AD_LONG) epos.offset -= sizeof(long_ad); - } - else + } else offset = 0; - if (!(fibh.sbh = fibh.ebh = udf_tread(dir->i_sb, block))) - { + if (!(fibh.sbh = fibh.ebh = udf_tread(dir->i_sb, block))) { brelse(epos.bh); return -EIO; } - - if (!(offset & ((16 >> (dir->i_sb->s_blocksize_bits - 9))-1))) - { + + if (!(offset & ((16 >> (dir->i_sb->s_blocksize_bits - 9)) - 1))) { i = 16 >> (dir->i_sb->s_blocksize_bits - 9); - if (i+offset > (elen >> dir->i_sb->s_blocksize_bits)) - i = (elen >> dir->i_sb->s_blocksize_bits)-offset; - for (num=0; i>0; i--) - { - block = udf_get_lb_pblock(dir->i_sb, eloc, offset+i); + if (i + offset > (elen >> dir->i_sb->s_blocksize_bits)) + i = (elen >> dir->i_sb->s_blocksize_bits) - + offset; + for (num = 0; i > 0; i--) { + block = + udf_get_lb_pblock(dir->i_sb, eloc, + offset + i); tmp = udf_tgetblk(dir->i_sb, block); - if (tmp && !buffer_uptodate(tmp) && !buffer_locked(tmp)) + if (tmp && !buffer_uptodate(tmp) + && !buffer_locked(tmp)) bha[num++] = tmp; else brelse(tmp); } - if (num) - { + if (num) { ll_rw_block(READA, num, bha); - for (i=0; i<num; i++) + for (i = 0; i < num; i++) brelse(bha[i]); } } - } - else - { + } else { brelse(epos.bh); return -ENOENT; } - while ( nf_pos < size ) - { + while (nf_pos < size) { filp->f_pos = nf_pos + 1; - fi = udf_fileident_read(dir, &nf_pos, &fibh, &cfi, &epos, &eloc, &elen, &offset); + fi = udf_fileident_read(dir, &nf_pos, &fibh, &cfi, &epos, &eloc, + &elen, &offset); - if (!fi) - { + if (!fi) { if (fibh.sbh != fibh.ebh) brelse(fibh.ebh); brelse(fibh.sbh); @@ -196,43 +192,41 @@ do_udf_readdir(struct inode * dir, struct file *filp, filldir_t filldir, void *d if (fibh.sbh == fibh.ebh) nameptr = fi->fileIdent + liu; - else - { + else { int poffset; /* Unpaded ending offset */ - poffset = fibh.soffset + sizeof(struct fileIdentDesc) + liu + lfi; + poffset = + fibh.soffset + sizeof(struct fileIdentDesc) + liu + + lfi; if (poffset >= lfi) - nameptr = (char *)(fibh.ebh->b_data + poffset - lfi); - else - { + nameptr = + (char *)(fibh.ebh->b_data + poffset - lfi); + else { nameptr = fname; - memcpy(nameptr, fi->fileIdent + liu, lfi - poffset); - memcpy(nameptr + lfi - poffset, fibh.ebh->b_data, poffset); + memcpy(nameptr, fi->fileIdent + liu, + lfi - poffset); + memcpy(nameptr + lfi - poffset, + fibh.ebh->b_data, poffset); } } - if ( (cfi.fileCharacteristics & FID_FILE_CHAR_DELETED) != 0 ) - { - if ( !UDF_QUERY_FLAG(dir->i_sb, UDF_FLAG_UNDELETE) ) + if ((cfi.fileCharacteristics & FID_FILE_CHAR_DELETED) != 0) { + if (!UDF_QUERY_FLAG(dir->i_sb, UDF_FLAG_UNDELETE)) continue; } - - if ( (cfi.fileCharacteristics & FID_FILE_CHAR_HIDDEN) != 0 ) - { - if ( !UDF_QUERY_FLAG(dir->i_sb, UDF_FLAG_UNHIDE) ) + + if ((cfi.fileCharacteristics & FID_FILE_CHAR_HIDDEN) != 0) { + if (!UDF_QUERY_FLAG(dir->i_sb, UDF_FLAG_UNHIDE)) continue; } - if ( cfi.fileCharacteristics & FID_FILE_CHAR_PARENT ) - { + if (cfi.fileCharacteristics & FID_FILE_CHAR_PARENT) { iblock = parent_ino(filp->f_path.dentry); flen = 2; memcpy(fname, "..", flen); dt_type = DT_DIR; - } - else - { + } else { kernel_lb_addr tloc = lelb_to_cpu(cfi.icb.extLocation); iblock = udf_get_lb_pblock(dir->i_sb, tloc, 0); @@ -240,18 +234,18 @@ do_udf_readdir(struct inode * dir, struct file *filp, filldir_t filldir, void *d dt_type = DT_UNKNOWN; } - if (flen) - { - if (filldir(dirent, fname, flen, filp->f_pos, iblock, dt_type) < 0) - { + if (flen) { + if (filldir + (dirent, fname, flen, filp->f_pos, iblock, + dt_type) < 0) { if (fibh.sbh != fibh.ebh) brelse(fibh.ebh); brelse(fibh.sbh); brelse(epos.bh); - return 0; + return 0; } } - } /* end while */ + } /* end while */ filp->f_pos = nf_pos + 1; diff --git a/fs/udf/directory.c b/fs/udf/directory.c index 198caa33027..8adc77c1d57 100644 --- a/fs/udf/directory.c +++ b/fs/udf/directory.c @@ -19,10 +19,10 @@ #include <linux/buffer_head.h> #if 0 -static uint8_t * -udf_filead_read(struct inode *dir, uint8_t *tmpad, uint8_t ad_size, - kernel_lb_addr fe_loc, int *pos, int *offset, - struct buffer_head **bh, int *error) +static uint8_t *udf_filead_read(struct inode *dir, uint8_t * tmpad, + uint8_t ad_size, kernel_lb_addr fe_loc, + int *pos, int *offset, struct buffer_head **bh, + int *error) { int loffset = *offset; int block; @@ -31,31 +31,27 @@ udf_filead_read(struct inode *dir, uint8_t *tmpad, uint8_t ad_size, *error = 0; - ad = (uint8_t *)(*bh)->b_data + *offset; + ad = (uint8_t *) (*bh)->b_data + *offset; *offset += ad_size; - if (!ad) - { + if (!ad) { brelse(*bh); *error = 1; return NULL; } - if (*offset == dir->i_sb->s_blocksize) - { + if (*offset == dir->i_sb->s_blocksize) { brelse(*bh); block = udf_get_lb_pblock(dir->i_sb, fe_loc, ++*pos); if (!block) return NULL; if (!(*bh = udf_tread(dir->i_sb, block))) return NULL; - } - else if (*offset > dir->i_sb->s_blocksize) - { + } else if (*offset > dir->i_sb->s_blocksize) { ad = tmpad; remainder = dir->i_sb->s_blocksize - loffset; - memcpy((uint8_t *)ad, (*bh)->b_data + loffset, remainder); + memcpy((uint8_t *) ad, (*bh)->b_data + loffset, remainder); brelse(*bh); block = udf_get_lb_pblock(dir->i_sb, fe_loc, ++*pos); @@ -64,56 +60,56 @@ udf_filead_read(struct inode *dir, uint8_t *tmpad, uint8_t ad_size, if (!((*bh) = udf_tread(dir->i_sb, block))) return NULL; - memcpy((uint8_t *)ad + remainder, (*bh)->b_data, ad_size - remainder); + memcpy((uint8_t *) ad + remainder, (*bh)->b_data, + ad_size - remainder); *offset = ad_size - remainder; } return ad; } #endif -struct fileIdentDesc * -udf_fileident_read(struct inode *dir, loff_t *nf_pos, - struct udf_fileident_bh *fibh, - struct fileIdentDesc *cfi, - struct extent_position *epos, - kernel_lb_addr *eloc, uint32_t *elen, - sector_t *offset) +struct fileIdentDesc *udf_fileident_read(struct inode *dir, loff_t * nf_pos, + struct udf_fileident_bh *fibh, + struct fileIdentDesc *cfi, + struct extent_position *epos, + kernel_lb_addr * eloc, uint32_t * elen, + sector_t * offset) { struct fileIdentDesc *fi; int i, num, block; - struct buffer_head * tmp, * bha[16]; + struct buffer_head *tmp, *bha[16]; fibh->soffset = fibh->eoffset; - if (UDF_I_ALLOCTYPE(dir) == ICBTAG_FLAG_AD_IN_ICB) - { + if (UDF_I_ALLOCTYPE(dir) == ICBTAG_FLAG_AD_IN_ICB) { fi = udf_get_fileident(UDF_I_DATA(dir) - - (UDF_I_EFE(dir) ? - sizeof(struct extendedFileEntry) : - sizeof(struct fileEntry)), - dir->i_sb->s_blocksize, &(fibh->eoffset)); + (UDF_I_EFE(dir) ? + sizeof(struct extendedFileEntry) : + sizeof(struct fileEntry)), + dir->i_sb->s_blocksize, + &(fibh->eoffset)); if (!fi) return NULL; *nf_pos += ((fibh->eoffset - fibh->soffset) >> 2); - memcpy((uint8_t *)cfi, (uint8_t *)fi, sizeof(struct fileIdentDesc)); + memcpy((uint8_t *) cfi, (uint8_t *) fi, + sizeof(struct fileIdentDesc)); return fi; } - if (fibh->eoffset == dir->i_sb->s_blocksize) - { + if (fibh->eoffset == dir->i_sb->s_blocksize) { int lextoffset = epos->offset; if (udf_next_aext(dir, epos, eloc, elen, 1) != - (EXT_RECORDED_ALLOCATED >> 30)) + (EXT_RECORDED_ALLOCATED >> 30)) return NULL; block = udf_get_lb_pblock(dir->i_sb, *eloc, *offset); - (*offset) ++; + (*offset)++; if ((*offset << dir->i_sb->s_blocksize_bits) >= *elen) *offset = 0; @@ -125,57 +121,57 @@ udf_fileident_read(struct inode *dir, loff_t *nf_pos, return NULL; fibh->soffset = fibh->eoffset = 0; - if (!(*offset & ((16 >> (dir->i_sb->s_blocksize_bits - 9))-1))) + if (! + (*offset & ((16 >> (dir->i_sb->s_blocksize_bits - 9)) - 1))) { i = 16 >> (dir->i_sb->s_blocksize_bits - 9); - if (i+*offset > (*elen >> dir->i_sb->s_blocksize_bits)) - i = (*elen >> dir->i_sb->s_blocksize_bits)-*offset; - for (num=0; i>0; i--) - { - block = udf_get_lb_pblock(dir->i_sb, *eloc, *offset+i); + if (i + *offset > + (*elen >> dir->i_sb->s_blocksize_bits)) + i = (*elen >> dir->i_sb->s_blocksize_bits) - + *offset; + for (num = 0; i > 0; i--) { + block = + udf_get_lb_pblock(dir->i_sb, *eloc, + *offset + i); tmp = udf_tgetblk(dir->i_sb, block); - if (tmp && !buffer_uptodate(tmp) && !buffer_locked(tmp)) + if (tmp && !buffer_uptodate(tmp) + && !buffer_locked(tmp)) bha[num++] = tmp; else brelse(tmp); } - if (num) - { + if (num) { ll_rw_block(READA, num, bha); - for (i=0; i<num; i++) + for (i = 0; i < num; i++) brelse(bha[i]); } } - } - else if (fibh->sbh != fibh->ebh) - { + } else if (fibh->sbh != fibh->ebh) { brelse(fibh->sbh); fibh->sbh = fibh->ebh; } fi = udf_get_fileident(fibh->sbh->b_data, dir->i_sb->s_blocksize, - &(fibh->eoffset)); + &(fibh->eoffset)); if (!fi) return NULL; *nf_pos += ((fibh->eoffset - fibh->soffset) >> 2); - if (fibh->eoffset <= dir->i_sb->s_blocksize) - { - memcpy((uint8_t *)cfi, (uint8_t *)fi, sizeof(struct fileIdentDesc)); - } - else if (fibh->eoffset > dir->i_sb->s_blocksize) - { + if (fibh->eoffset <= dir->i_sb->s_blocksize) { + memcpy((uint8_t *) cfi, (uint8_t *) fi, + sizeof(struct fileIdentDesc)); + } else if (fibh->eoffset > dir->i_sb->s_blocksize) { int lextoffset = epos->offset; if (udf_next_aext(dir, epos, eloc, elen, 1) != - (EXT_RECORDED_ALLOCATED >> 30)) + (EXT_RECORDED_ALLOCATED >> 30)) return NULL; block = udf_get_lb_pblock(dir->i_sb, *eloc, *offset); - (*offset) ++; + (*offset)++; if ((*offset << dir->i_sb->s_blocksize_bits) >= *elen) *offset = 0; @@ -188,62 +184,62 @@ udf_fileident_read(struct inode *dir, loff_t *nf_pos, if (!(fibh->ebh = udf_tread(dir->i_sb, block))) return NULL; - if (sizeof(struct fileIdentDesc) > - fibh->soffset) - { + if (sizeof(struct fileIdentDesc) > -fibh->soffset) { int fi_len; - memcpy((uint8_t *)cfi, (uint8_t *)fi, - fibh->soffset); - memcpy((uint8_t *)cfi - fibh->soffset, fibh->ebh->b_data, - sizeof(struct fileIdentDesc) + fibh->soffset); + memcpy((uint8_t *) cfi, (uint8_t *) fi, -fibh->soffset); + memcpy((uint8_t *) cfi - fibh->soffset, + fibh->ebh->b_data, + sizeof(struct fileIdentDesc) + fibh->soffset); - fi_len = (sizeof(struct fileIdentDesc) + cfi->lengthFileIdent + - le16_to_cpu(cfi->lengthOfImpUse) + 3) & ~3; + fi_len = + (sizeof(struct fileIdentDesc) + + cfi->lengthFileIdent + + le16_to_cpu(cfi->lengthOfImpUse) + 3) & ~3; - *nf_pos += ((fi_len - (fibh->eoffset - fibh->soffset)) >> 2); + *nf_pos += + ((fi_len - (fibh->eoffset - fibh->soffset)) >> 2); fibh->eoffset = fibh->soffset + fi_len; - } - else - { - memcpy((uint8_t *)cfi, (uint8_t *)fi, sizeof(struct fileIdentDesc)); + } else { + memcpy((uint8_t *) cfi, (uint8_t *) fi, + sizeof(struct fileIdentDesc)); } } return fi; } -struct fileIdentDesc * -udf_get_fileident(void * buffer, int bufsize, int * offset) +struct fileIdentDesc *udf_get_fileident(void *buffer, int bufsize, int *offset) { struct fileIdentDesc *fi; int lengthThisIdent; - uint8_t * ptr; + uint8_t *ptr; int padlen; - if ( (!buffer) || (!offset) ) { - udf_debug("invalidparms\n, buffer=%p, offset=%p\n", buffer, offset); + if ((!buffer) || (!offset)) { + udf_debug("invalidparms\n, buffer=%p, offset=%p\n", buffer, + offset); return NULL; } ptr = buffer; - if ( (*offset > 0) && (*offset < bufsize) ) { + if ((*offset > 0) && (*offset < bufsize)) { ptr += *offset; } - fi=(struct fileIdentDesc *)ptr; - if (le16_to_cpu(fi->descTag.tagIdent) != TAG_IDENT_FID) - { + fi = (struct fileIdentDesc *)ptr; + if (le16_to_cpu(fi->descTag.tagIdent) != TAG_IDENT_FID) { udf_debug("0x%x != TAG_IDENT_FID\n", - le16_to_cpu(fi->descTag.tagIdent)); + le16_to_cpu(fi->descTag.tagIdent)); udf_debug("offset: %u sizeof: %lu bufsize: %u\n", - *offset, (unsigned long)sizeof(struct fileIdentDesc), bufsize); + *offset, (unsigned long)sizeof(struct fileIdentDesc), + bufsize); return NULL; } - if ( (*offset + sizeof(struct fileIdentDesc)) > bufsize ) - { + if ((*offset + sizeof(struct fileIdentDesc)) > bufsize) { lengthThisIdent = sizeof(struct fileIdentDesc); - } - else + } else lengthThisIdent = sizeof(struct fileIdentDesc) + - fi->lengthFileIdent + le16_to_cpu(fi->lengthOfImpUse); + fi->lengthFileIdent + le16_to_cpu(fi->lengthOfImpUse); /* we need to figure padding, too! */ padlen = lengthThisIdent % UDF_NAME_PAD; @@ -255,56 +251,53 @@ udf_get_fileident(void * buffer, int bufsize, int * offset) } #if 0 -static extent_ad * -udf_get_fileextent(void * buffer, int bufsize, int * offset) +static extent_ad *udf_get_fileextent(void *buffer, int bufsize, int *offset) { - extent_ad * ext; + extent_ad *ext; struct fileEntry *fe; - uint8_t * ptr; + uint8_t *ptr; - if ( (!buffer) || (!offset) ) - { + if ((!buffer) || (!offset)) { printk(KERN_ERR "udf: udf_get_fileextent() invalidparms\n"); return NULL; } fe = (struct fileEntry *)buffer; - if ( le16_to_cpu(fe->descTag.tagIdent) != TAG_IDENT_FE ) - { + if (le16_to_cpu(fe->descTag.tagIdent) != TAG_IDENT_FE) { udf_debug("0x%x != TAG_IDENT_FE\n", - le16_to_cpu(fe->descTag.tagIdent)); + le16_to_cpu(fe->descTag.tagIdent)); return NULL; } - ptr=(uint8_t *)(fe->extendedAttr) + le32_to_cpu(fe->lengthExtendedAttr); + ptr = + (uint8_t *) (fe->extendedAttr) + + le32_to_cpu(fe->lengthExtendedAttr); - if ( (*offset > 0) && (*offset < le32_to_cpu(fe->lengthAllocDescs)) ) - { + if ((*offset > 0) && (*offset < le32_to_cpu(fe->lengthAllocDescs))) { ptr += *offset; } - ext = (extent_ad *)ptr; + ext = (extent_ad *) ptr; *offset = *offset + sizeof(extent_ad); return ext; } #endif -short_ad * -udf_get_fileshortad(uint8_t *ptr, int maxoffset, int *offset, int inc) +short_ad *udf_get_fileshortad(uint8_t * ptr, int maxoffset, int *offset, + int inc) { short_ad *sa; - if ( (!ptr) || (!offset) ) - { + if ((!ptr) || (!offset)) { printk(KERN_ERR "udf: udf_get_fileshortad() invalidparms\n"); return NULL; } - if ( (*offset < 0) || ((*offset + sizeof(short_ad)) > maxoffset) ) + if ((*offset < 0) || ((*offset + sizeof(short_ad)) > maxoffset)) return NULL; - else if ((sa = (short_ad *)ptr)->extLength == 0) + else if ((sa = (short_ad *) ptr)->extLength == 0) return NULL; if (inc) @@ -312,20 +305,18 @@ udf_get_fileshortad(uint8_t *ptr, int maxoffset, int *offset, int inc) return sa; } -long_ad * -udf_get_filelongad(uint8_t *ptr, int maxoffset, int * offset, int inc) +long_ad *udf_get_filelongad(uint8_t * ptr, int maxoffset, int *offset, int inc) { long_ad *la; - if ( (!ptr) || (!offset) ) - { + if ((!ptr) || (!offset)) { printk(KERN_ERR "udf: udf_get_filelongad() invalidparms\n"); return NULL; } - if ( (*offset < 0) || ((*offset + sizeof(long_ad)) > maxoffset) ) + if ((*offset < 0) || ((*offset + sizeof(long_ad)) > maxoffset)) return NULL; - else if ((la = (long_ad *)ptr)->extLength == 0) + else if ((la = (long_ad *) ptr)->extLength == 0) return NULL; if (inc) diff --git a/fs/udf/ecma_167.h b/fs/udf/ecma_167.h index f81f2ebbf50..294ce2daa03 100644 --- a/fs/udf/ecma_167.h +++ b/fs/udf/ecma_167.h @@ -38,10 +38,9 @@ #define _ECMA_167_H 1 /* Character set specification (ECMA 167r3 1/7.2.1) */ -typedef struct -{ - uint8_t charSetType; - uint8_t charSetInfo[63]; +typedef struct { + uint8_t charSetType; + uint8_t charSetInfo[63]; } __attribute__ ((packed)) charspec; /* Character Set Type (ECMA 167r3 1/7.2.1.1) */ @@ -55,35 +54,33 @@ typedef struct #define CHARSPEC_TYPE_CS7 0x07 /* (1/7.2.9) */ #define CHARSPEC_TYPE_CS8 0x08 /* (1/7.2.10) */ -typedef uint8_t dstring; +typedef uint8_t dstring; /* Timestamp (ECMA 167r3 1/7.3) */ -typedef struct -{ - __le16 typeAndTimezone; - __le16 year; - uint8_t month; - uint8_t day; - uint8_t hour; - uint8_t minute; - uint8_t second; - uint8_t centiseconds; - uint8_t hundredsOfMicroseconds; - uint8_t microseconds; +typedef struct { + __le16 typeAndTimezone; + __le16 year; + uint8_t month; + uint8_t day; + uint8_t hour; + uint8_t minute; + uint8_t second; + uint8_t centiseconds; + uint8_t hundredsOfMicroseconds; + uint8_t microseconds; } __attribute__ ((packed)) timestamp; -typedef struct -{ - uint16_t typeAndTimezone; - int16_t year; - uint8_t month; - uint8_t day; - uint8_t hour; - uint8_t minute; - uint8_t second; - uint8_t centiseconds; - uint8_t hundredsOfMicroseconds; - uint8_t microseconds; +typedef struct { + uint16_t typeAndTimezone; + int16_t year; + uint8_t month; + uint8_t day; + uint8_t hour; + uint8_t minute; + uint8_t second; + uint8_t centiseconds; + uint8_t hundredsOfMicroseconds; + uint8_t microseconds; } __attribute__ ((packed)) kernel_timestamp; /* Type and Time Zone (ECMA 167r3 1/7.3.1) */ @@ -94,11 +91,10 @@ typedef struct #define TIMESTAMP_TIMEZONE_MASK 0x0FFF /* Entity identifier (ECMA 167r3 1/7.4) */ -typedef struct -{ - uint8_t flags; - uint8_t ident[23]; - uint8_t identSuffix[8]; +typedef struct { + uint8_t flags; + uint8_t ident[23]; + uint8_t identSuffix[8]; } __attribute__ ((packed)) regid; /* Flags (ECMA 167r3 1/7.4.1) */ @@ -107,12 +103,11 @@ typedef struct /* Volume Structure Descriptor (ECMA 167r3 2/9.1) */ #define VSD_STD_ID_LEN 5 -struct volStructDesc -{ - uint8_t structType; - uint8_t stdIdent[VSD_STD_ID_LEN]; - uint8_t structVersion; - uint8_t structData[2041]; +struct volStructDesc { + uint8_t structType; + uint8_t stdIdent[VSD_STD_ID_LEN]; + uint8_t structVersion; + uint8_t structData[2041]; } __attribute__ ((packed)); /* Standard Identifier (EMCA 167r2 2/9.1.2) */ @@ -127,69 +122,63 @@ struct volStructDesc #define VSD_STD_ID_TEA01 "TEA01" /* (2/9.3) */ /* Beginning Extended Area Descriptor (ECMA 167r3 2/9.2) */ -struct beginningExtendedAreaDesc -{ - uint8_t structType; - uint8_t stdIdent[VSD_STD_ID_LEN]; - uint8_t structVersion; - uint8_t structData[2041]; +struct beginningExtendedAreaDesc { + uint8_t structType; + uint8_t stdIdent[VSD_STD_ID_LEN]; + uint8_t structVersion; + uint8_t structData[2041]; } __attribute__ ((packed)); /* Terminating Extended Area Descriptor (ECMA 167r3 2/9.3) */ -struct terminatingExtendedAreaDesc -{ - uint8_t structType; - uint8_t stdIdent[VSD_STD_ID_LEN]; - uint8_t structVersion; - uint8_t structData[2041]; +struct terminatingExtendedAreaDesc { + uint8_t structType; + uint8_t stdIdent[VSD_STD_ID_LEN]; + uint8_t structVersion; + uint8_t structData[2041]; } __attribute__ ((packed)); /* Boot Descriptor (ECMA 167r3 2/9.4) */ -struct bootDesc -{ - uint8_t structType; - uint8_t stdIdent[VSD_STD_ID_LEN]; - uint8_t structVersion; - uint8_t reserved1; - regid archType; - regid bootIdent; - __le32 bootExtLocation; - __le32 bootExtLength; - __le64 loadAddress; - __le64 startAddress; - timestamp descCreationDateAndTime; - __le16 flags; - uint8_t reserved2[32]; - uint8_t bootUse[1906]; +struct bootDesc { + uint8_t structType; + uint8_t stdIdent[VSD_STD_ID_LEN]; + uint8_t structVersion; + uint8_t reserved1; + regid archType; + regid bootIdent; + __le32 bootExtLocation; + __le32 bootExtLength; + __le64 loadAddress; + __le64 startAddress; + timestamp descCreationDateAndTime; + __le16 flags; + uint8_t reserved2[32]; + uint8_t bootUse[1906]; } __attribute__ ((packed)); /* Flags (ECMA 167r3 2/9.4.12) */ #define BOOT_FLAGS_ERASE 0x01 /* Extent Descriptor (ECMA 167r3 3/7.1) */ -typedef struct -{ - __le32 extLength; - __le32 extLocation; +typedef struct { + __le32 extLength; + __le32 extLocation; } __attribute__ ((packed)) extent_ad; -typedef struct -{ - uint32_t extLength; - uint32_t extLocation; +typedef struct { + uint32_t extLength; + uint32_t extLocation; } kernel_extent_ad; /* Descriptor Tag (ECMA 167r3 3/7.2) */ -typedef struct -{ - __le16 tagIdent; - __le16 descVersion; - uint8_t tagChecksum; - uint8_t reserved; - __le16 tagSerialNum; - __le16 descCRC; - __le16 descCRCLength; - __le32 tagLocation; +typedef struct { + __le16 tagIdent; + __le16 descVersion; + uint8_t tagChecksum; + uint8_t reserved; + __le16 tagSerialNum; + __le16 descCRC; + __le16 descCRCLength; + __le32 tagLocation; } __attribute__ ((packed)) tag; /* Tag Identifier (ECMA 167r3 3/7.2.1) */ @@ -204,87 +193,81 @@ typedef struct #define TAG_IDENT_LVID 0x0009 /* NSR Descriptor (ECMA 167r3 3/9.1) */ -struct NSRDesc -{ - uint8_t structType; - uint8_t stdIdent[VSD_STD_ID_LEN]; - uint8_t structVersion; - uint8_t reserved; - uint8_t structData[2040]; -} __attribute__ ((packed)); - +struct NSRDesc { + uint8_t structType; + uint8_t stdIdent[VSD_STD_ID_LEN]; + uint8_t structVersion; + uint8_t reserved; + uint8_t structData[2040]; +} __attribute__ ((packed)); + /* Primary Volume Descriptor (ECMA 167r3 3/10.1) */ -struct primaryVolDesc -{ - tag descTag; - __le32 volDescSeqNum; - __le32 primaryVolDescNum; - dstring volIdent[32]; - __le16 volSeqNum; - __le16 maxVolSeqNum; - __le16 interchangeLvl; - __le16 maxInterchangeLvl; - __le32 charSetList; - __le32 maxCharSetList; - dstring volSetIdent[128]; - charspec descCharSet; - charspec explanatoryCharSet; - extent_ad volAbstract; - extent_ad volCopyright; - regid appIdent; - timestamp recordingDateAndTime; - regid impIdent; - uint8_t impUse[64]; - __le32 predecessorVolDescSeqLocation; - __le16 flags; - uint8_t reserved[22]; +struct primaryVolDesc { + tag descTag; + __le32 volDescSeqNum; + __le32 primaryVolDescNum; + dstring volIdent[32]; + __le16 volSeqNum; + __le16 maxVolSeqNum; + __le16 interchangeLvl; + __le16 maxInterchangeLvl; + __le32 charSetList; + __le32 maxCharSetList; + dstring volSetIdent[128]; + charspec descCharSet; + charspec explanatoryCharSet; + extent_ad volAbstract; + extent_ad volCopyright; + regid appIdent; + timestamp recordingDateAndTime; + regid impIdent; + uint8_t impUse[64]; + __le32 predecessorVolDescSeqLocation; + __le16 flags; + uint8_t reserved[22]; } __attribute__ ((packed)); /* Flags (ECMA 167r3 3/10.1.21) */ #define PVD_FLAGS_VSID_COMMON 0x0001 /* Anchor Volume Descriptor Pointer (ECMA 167r3 3/10.2) */ -struct anchorVolDescPtr -{ - tag descTag; - extent_ad mainVolDescSeqExt; - extent_ad reserveVolDescSeqExt; - uint8_t reserved[480]; +struct anchorVolDescPtr { + tag descTag; + extent_ad mainVolDescSeqExt; + extent_ad reserveVolDescSeqExt; + uint8_t reserved[480]; } __attribute__ ((packed)); /* Volume Descriptor Pointer (ECMA 167r3 3/10.3) */ -struct volDescPtr -{ - tag descTag; - __le32 volDescSeqNum; - extent_ad nextVolDescSeqExt; - uint8_t reserved[484]; +struct volDescPtr { + tag descTag; + __le32 volDescSeqNum; + extent_ad nextVolDescSeqExt; + uint8_t reserved[484]; } __attribute__ ((packed)); /* Implementation Use Volume Descriptor (ECMA 167r3 3/10.4) */ -struct impUseVolDesc -{ - tag descTag; - __le32 volDescSeqNum; - regid impIdent; - uint8_t impUse[460]; +struct impUseVolDesc { + tag descTag; + __le32 volDescSeqNum; + regid impIdent; + uint8_t impUse[460]; } __attribute__ ((packed)); /* Partition Descriptor (ECMA 167r3 3/10.5) */ -struct partitionDesc -{ - tag descTag; - __le32 volDescSeqNum; - __le16 partitionFlags; - __le16 partitionNumber; - regid partitionContents; - uint8_t partitionContentsUse[128]; - __le32 accessType; - __le32 partitionStartingLocation; - __le32 partitionLength; - regid impIdent; - uint8_t impUse[128]; - uint8_t reserved[156]; +struct partitionDesc { + tag descTag; + __le32 volDescSeqNum; + __le16 partitionFlags; + __le16 partitionNumber; + regid partitionContents; + uint8_t partitionContentsUse[128]; + __le32 accessType; + __le32 partitionStartingLocation; + __le32 partitionLength; + regid impIdent; + uint8_t impUse[128]; + uint8_t reserved[156]; } __attribute__ ((packed)); /* Partition Flags (ECMA 167r3 3/10.5.3) */ @@ -307,29 +290,27 @@ struct partitionDesc #define PD_ACCESS_TYPE_OVERWRITABLE 0x00000004 /* Logical Volume Descriptor (ECMA 167r3 3/10.6) */ -struct logicalVolDesc -{ - tag descTag; - __le32 volDescSeqNum; - charspec descCharSet; - dstring logicalVolIdent[128]; - __le32 logicalBlockSize; - regid domainIdent; - uint8_t logicalVolContentsUse[16]; - __le32 mapTableLength; - __le32 numPartitionMaps; - regid impIdent; - uint8_t impUse[128]; - extent_ad integritySeqExt; - uint8_t partitionMaps[0]; +struct logicalVolDesc { + tag descTag; + __le32 volDescSeqNum; + charspec descCharSet; + dstring logicalVolIdent[128]; + __le32 logicalBlockSize; + regid domainIdent; + uint8_t logicalVolContentsUse[16]; + __le32 mapTableLength; + __le32 numPartitionMaps; + regid impIdent; + uint8_t impUse[128]; + extent_ad integritySeqExt; + uint8_t partitionMaps[0]; } __attribute__ ((packed)); /* Generic Partition Map (ECMA 167r3 3/10.7.1) */ -struct genericPartitionMap -{ - uint8_t partitionMapType; - uint8_t partitionMapLength; - uint8_t partitionMapping[0]; +struct genericPartitionMap { + uint8_t partitionMapType; + uint8_t partitionMapLength; + uint8_t partitionMapping[0]; } __attribute__ ((packed)); /* Partition Map Type (ECMA 167r3 3/10.7.1.1) */ @@ -338,51 +319,46 @@ struct genericPartitionMap #define GP_PARTITION_MAP_TYPE_2 0x02 /* Type 1 Partition Map (ECMA 167r3 3/10.7.2) */ -struct genericPartitionMap1 -{ - uint8_t partitionMapType; - uint8_t partitionMapLength; - __le16 volSeqNum; - __le16 partitionNum; +struct genericPartitionMap1 { + uint8_t partitionMapType; + uint8_t partitionMapLength; + __le16 volSeqNum; + __le16 partitionNum; } __attribute__ ((packed)); /* Type 2 Partition Map (ECMA 167r3 3/10.7.3) */ -struct genericPartitionMap2 -{ - uint8_t partitionMapType; - uint8_t partitionMapLength; - uint8_t partitionIdent[62]; +struct genericPartitionMap2 { + uint8_t partitionMapType; + uint8_t partitionMapLength; + uint8_t partitionIdent[62]; } __attribute__ ((packed)); /* Unallocated Space Descriptor (ECMA 167r3 3/10.8) */ -struct unallocSpaceDesc -{ - tag descTag; - __le32 volDescSeqNum; - __le32 numAllocDescs; - extent_ad allocDescs[0]; +struct unallocSpaceDesc { + tag descTag; + __le32 volDescSeqNum; + __le32 numAllocDescs; + extent_ad allocDescs[0]; } __attribute__ ((packed)); /* Terminating Descriptor (ECMA 167r3 3/10.9) */ -struct terminatingDesc -{ - tag descTag; - uint8_t reserved[496]; +struct terminatingDesc { + tag descTag; + uint8_t reserved[496]; } __attribute__ ((packed)); /* Logical Volume Integrity Descriptor (ECMA 167r3 3/10.10) */ -struct logicalVolIntegrityDesc -{ - tag descTag; - timestamp recordingDateAndTime; - __le32 integrityType; - extent_ad nextIntegrityExt; - uint8_t logicalVolContentsUse[32]; - __le32 numOfPartitions; - __le32 lengthOfImpUse; - __le32 freeSpaceTable[0]; - __le32 sizeTable[0]; - uint8_t impUse[0]; +struct logicalVolIntegrityDesc { + tag descTag; + timestamp recordingDateAndTime; + __le32 integrityType; + extent_ad nextIntegrityExt; + uint8_t logicalVolContentsUse[32]; + __le32 numOfPartitions; + __le32 lengthOfImpUse; + __le32 freeSpaceTable[0]; + __le32 sizeTable[0]; + uint8_t impUse[0]; } __attribute__ ((packed)); /* Integrity Type (ECMA 167r3 3/10.10.3) */ @@ -390,56 +366,49 @@ struct logicalVolIntegrityDesc #define LVID_INTEGRITY_TYPE_CLOSE 0x00000001 /* Recorded Address (ECMA 167r3 4/7.1) */ -typedef struct -{ - __le32 logicalBlockNum; - __le16 partitionReferenceNum; +typedef struct { + __le32 logicalBlockNum; + __le16 partitionReferenceNum; } __attribute__ ((packed)) lb_addr; /* ... and its in-core analog */ -typedef struct -{ - uint32_t logicalBlockNum; - uint16_t partitionReferenceNum; +typedef struct { + uint32_t logicalBlockNum; + uint16_t partitionReferenceNum; } kernel_lb_addr; /* Short Allocation Descriptor (ECMA 167r3 4/14.14.1) */ -typedef struct -{ - __le32 extLength; - __le32 extPosition; +typedef struct { + __le32 extLength; + __le32 extPosition; } __attribute__ ((packed)) short_ad; /* Long Allocation Descriptor (ECMA 167r3 4/14.14.2) */ -typedef struct -{ - __le32 extLength; - lb_addr extLocation; - uint8_t impUse[6]; +typedef struct { + __le32 extLength; + lb_addr extLocation; + uint8_t impUse[6]; } __attribute__ ((packed)) long_ad; -typedef struct -{ - uint32_t extLength; - kernel_lb_addr extLocation; - uint8_t impUse[6]; +typedef struct { + uint32_t extLength; + kernel_lb_addr extLocation; + uint8_t impUse[6]; } kernel_long_ad; /* Extended Allocation Descriptor (ECMA 167r3 4/14.14.3) */ -typedef struct -{ - __le32 extLength; - __le32 recordedLength; - __le32 informationLength; - lb_addr extLocation; +typedef struct { + __le32 extLength; + __le32 recordedLength; + __le32 informationLength; + lb_addr extLocation; } __attribute__ ((packed)) ext_ad; -typedef struct -{ - uint32_t extLength; - uint32_t recordedLength; - uint32_t informationLength; - kernel_lb_addr extLocation; +typedef struct { + uint32_t extLength; + uint32_t recordedLength; + uint32_t informationLength; + kernel_lb_addr extLocation; } kernel_ext_ad; /* Descriptor Tag (ECMA 167r3 4/7.2 - See 3/7.2) */ @@ -458,52 +427,49 @@ typedef struct #define TAG_IDENT_EFE 0x010A /* File Set Descriptor (ECMA 167r3 4/14.1) */ -struct fileSetDesc -{ - tag descTag; - timestamp recordingDateAndTime; - __le16 interchangeLvl; - __le16 maxInterchangeLvl; - __le32 charSetList; - __le32 maxCharSetList; - __le32 fileSetNum; - __le32 fileSetDescNum; - charspec logicalVolIdentCharSet; - dstring logicalVolIdent[128]; - charspec fileSetCharSet; - dstring fileSetIdent[32]; - dstring copyrightFileIdent[32]; - dstring abstractFileIdent[32]; - long_ad rootDirectoryICB; - regid domainIdent; - long_ad nextExt; - long_ad streamDirectoryICB; - uint8_t reserved[32]; +struct fileSetDesc { + tag descTag; + timestamp recordingDateAndTime; + __le16 interchangeLvl; + __le16 maxInterchangeLvl; + __le32 charSetList; + __le32 maxCharSetList; + __le32 fileSetNum; + __le32 fileSetDescNum; + charspec logicalVolIdentCharSet; + dstring logicalVolIdent[128]; + charspec fileSetCharSet; + dstring fileSetIdent[32]; + dstring copyrightFileIdent[32]; + dstring abstractFileIdent[32]; + long_ad rootDirectoryICB; + regid domainIdent; + long_ad nextExt; + long_ad streamDirectoryICB; + uint8_t reserved[32]; } __attribute__ ((packed)); /* Partition Header Descriptor (ECMA 167r3 4/14.3) */ -struct partitionHeaderDesc -{ - short_ad unallocSpaceTable; - short_ad unallocSpaceBitmap; - short_ad partitionIntegrityTable; - short_ad freedSpaceTable; - short_ad freedSpaceBitmap; - uint8_t reserved[88]; +struct partitionHeaderDesc { + short_ad unallocSpaceTable; + short_ad unallocSpaceBitmap; + short_ad partitionIntegrityTable; + short_ad freedSpaceTable; + short_ad freedSpaceBitmap; + uint8_t reserved[88]; } __attribute__ ((packed)); /* File Identifier Descriptor (ECMA 167r3 4/14.4) */ -struct fileIdentDesc -{ - tag descTag; - __le16 fileVersionNum; - uint8_t fileCharacteristics; - uint8_t lengthFileIdent; - long_ad icb; - __le16 lengthOfImpUse; - uint8_t impUse[0]; - uint8_t fileIdent[0]; - uint8_t padding[0]; +struct fileIdentDesc { + tag descTag; + __le16 fileVersionNum; + uint8_t fileCharacteristics; + uint8_t lengthFileIdent; + long_ad icb; + __le16 lengthOfImpUse; + uint8_t impUse[0]; + uint8_t fileIdent[0]; + uint8_t padding[0]; } __attribute__ ((packed)); /* File Characteristics (ECMA 167r3 4/14.4.3) */ @@ -514,24 +480,22 @@ struct fileIdentDesc #define FID_FILE_CHAR_METADATA 0x10 /* Allocation Ext Descriptor (ECMA 167r3 4/14.5) */ -struct allocExtDesc -{ - tag descTag; - __le32 previousAllocExtLocation; - __le32 lengthAllocDescs; +struct allocExtDesc { + tag descTag; + __le32 previousAllocExtLocation; + __le32 lengthAllocDescs; } __attribute__ ((packed)); /* ICB Tag (ECMA 167r3 4/14.6) */ -typedef struct -{ - __le32 priorRecordedNumDirectEntries; - __le16 strategyType; - __le16 strategyParameter; - __le16 numEntries; - uint8_t reserved; - uint8_t fileType; - lb_addr parentICBLocation; - __le16 flags; +typedef struct { + __le32 priorRecordedNumDirectEntries; + __le16 strategyType; + __le16 strategyParameter; + __le16 numEntries; + uint8_t reserved; + uint8_t fileType; + lb_addr parentICBLocation; + __le16 flags; } __attribute__ ((packed)) icbtag; /* Strategy Type (ECMA 167r3 4/14.6.2) */ @@ -576,45 +540,42 @@ typedef struct #define ICBTAG_FLAG_STREAM 0x2000 /* Indirect Entry (ECMA 167r3 4/14.7) */ -struct indirectEntry -{ - tag descTag; - icbtag icbTag; - long_ad indirectICB; +struct indirectEntry { + tag descTag; + icbtag icbTag; + long_ad indirectICB; } __attribute__ ((packed)); /* Terminal Entry (ECMA 167r3 4/14.8) */ -struct terminalEntry -{ - tag descTag; - icbtag icbTag; +struct terminalEntry { + tag descTag; + icbtag icbTag; } __attribute__ ((packed)); /* File Entry (ECMA 167r3 4/14.9) */ -struct fileEntry -{ - tag descTag; - icbtag icbTag; - __le32 uid; - __le32 gid; - __le32 permissions; - __le16 fileLinkCount; - uint8_t recordFormat; - uint8_t recordDisplayAttr; - __le32 recordLength; - __le64 informationLength; - __le64 logicalBlocksRecorded; - timestamp accessTime; - timestamp modificationTime; - timestamp attrTime; - __le32 checkpoint; - long_ad extendedAttrICB; - regid impIdent; - __le64 uniqueID; - __le32 lengthExtendedAttr; - __le32 lengthAllocDescs; - uint8_t extendedAttr[0]; - uint8_t allocDescs[0]; +struct fileEntry { + tag descTag; + icbtag icbTag; + __le32 uid; + __le32 gid; + __le32 permissions; + __le16 fileLinkCount; + uint8_t recordFormat; + uint8_t recordDisplayAttr; + __le32 recordLength; + __le64 informationLength; + __le64 logicalBlocksRecorded; + timestamp accessTime; + timestamp modificationTime; + timestamp attrTime; + __le32 checkpoint; + long_ad extendedAttrICB; + regid impIdent; + __le64 uniqueID; + __le32 lengthExtendedAttr; + __le32 lengthAllocDescs; + uint8_t extendedAttr[0]; + uint8_t allocDescs[0]; } __attribute__ ((packed)); /* Permissions (ECMA 167r3 4/14.9.5) */ @@ -655,57 +616,52 @@ struct fileEntry #define FE_RECORD_DISPLAY_ATTR_3 0x03 /* Extended Attribute Header Descriptor (ECMA 167r3 4/14.10.1) */ -struct extendedAttrHeaderDesc -{ - tag descTag; - __le32 impAttrLocation; - __le32 appAttrLocation; +struct extendedAttrHeaderDesc { + tag descTag; + __le32 impAttrLocation; + __le32 appAttrLocation; } __attribute__ ((packed)); /* Generic Format (ECMA 167r3 4/14.10.2) */ -struct genericFormat -{ - __le32 attrType; - uint8_t attrSubtype; - uint8_t reserved[3]; - __le32 attrLength; - uint8_t attrData[0]; +struct genericFormat { + __le32 attrType; + uint8_t attrSubtype; + uint8_t reserved[3]; + __le32 attrLength; + uint8_t attrData[0]; } __attribute__ ((packed)); /* Character Set Information (ECMA 167r3 4/14.10.3) */ -struct charSetInfo -{ - __le32 attrType; - uint8_t attrSubtype; - uint8_t reserved[3]; - __le32 attrLength; - __le32 escapeSeqLength; - uint8_t charSetType; - uint8_t escapeSeq[0]; +struct charSetInfo { + __le32 attrType; + uint8_t attrSubtype; + uint8_t reserved[3]; + __le32 attrLength; + __le32 escapeSeqLength; + uint8_t charSetType; + uint8_t escapeSeq[0]; } __attribute__ ((packed)); /* Alternate Permissions (ECMA 167r3 4/14.10.4) */ -struct altPerms -{ - __le32 attrType; - uint8_t attrSubtype; - uint8_t reserved[3]; - __le32 attrLength; - __le16 ownerIdent; - __le16 groupIdent; - __le16 permission; +struct altPerms { + __le32 attrType; + uint8_t attrSubtype; + uint8_t reserved[3]; + __le32 attrLength; + __le16 ownerIdent; + __le16 groupIdent; + __le16 permission; } __attribute__ ((packed)); /* File Times Extended Attribute (ECMA 167r3 4/14.10.5) */ -struct fileTimesExtAttr -{ - __le32 attrType; - uint8_t attrSubtype; - uint8_t reserved[3]; - __le32 attrLength; - __le32 dataLength; - __le32 fileTimeExistence; - uint8_t fileTimes; +struct fileTimesExtAttr { + __le32 attrType; + uint8_t attrSubtype; + uint8_t reserved[3]; + __le32 attrLength; + __le32 dataLength; + __le32 fileTimeExistence; + uint8_t fileTimes; } __attribute__ ((packed)); /* FileTimeExistence (ECMA 167r3 4/14.10.5.6) */ @@ -715,52 +671,48 @@ struct fileTimesExtAttr #define FTE_BACKUP 0x00000002 /* Information Times Extended Attribute (ECMA 167r3 4/14.10.6) */ -struct infoTimesExtAttr -{ - __le32 attrType; - uint8_t attrSubtype; - uint8_t reserved[3]; - __le32 attrLength; - __le32 dataLength; - __le32 infoTimeExistence; - uint8_t infoTimes[0]; +struct infoTimesExtAttr { + __le32 attrType; + uint8_t attrSubtype; + uint8_t reserved[3]; + __le32 attrLength; + __le32 dataLength; + __le32 infoTimeExistence; + uint8_t infoTimes[0]; } __attribute__ ((packed)); /* Device Specification (ECMA 167r3 4/14.10.7) */ -struct deviceSpec -{ - __le32 attrType; - uint8_t attrSubtype; - uint8_t reserved[3]; - __le32 attrLength; - __le32 impUseLength; - __le32 majorDeviceIdent; - __le32 minorDeviceIdent; - uint8_t impUse[0]; +struct deviceSpec { + __le32 attrType; + uint8_t attrSubtype; + uint8_t reserved[3]; + __le32 attrLength; + __le32 impUseLength; + __le32 majorDeviceIdent; + __le32 minorDeviceIdent; + uint8_t impUse[0]; } __attribute__ ((packed)); /* Implementation Use Extended Attr (ECMA 167r3 4/14.10.8) */ -struct impUseExtAttr -{ - __le32 attrType; - uint8_t attrSubtype; - uint8_t reserved[3]; - __le32 attrLength; - __le32 impUseLength; - regid impIdent; - uint8_t impUse[0]; +struct impUseExtAttr { + __le32 attrType; + uint8_t attrSubtype; + uint8_t reserved[3]; + __le32 attrLength; + __le32 impUseLength; + regid impIdent; + uint8_t impUse[0]; } __attribute__ ((packed)); /* Application Use Extended Attribute (ECMA 167r3 4/14.10.9) */ -struct appUseExtAttr -{ - __le32 attrType; - uint8_t attrSubtype; - uint8_t reserved[3]; - __le32 attrLength; - __le32 appUseLength; - regid appIdent; - uint8_t appUse[0]; +struct appUseExtAttr { + __le32 attrType; + uint8_t attrSubtype; + uint8_t reserved[3]; + __le32 attrLength; + __le32 appUseLength; + regid appIdent; + uint8_t appUse[0]; } __attribute__ ((packed)); #define EXTATTR_CHAR_SET 1 @@ -771,35 +723,31 @@ struct appUseExtAttr #define EXTATTR_IMP_USE 2048 #define EXTATTR_APP_USE 65536 - /* Unallocated Space Entry (ECMA 167r3 4/14.11) */ -struct unallocSpaceEntry -{ - tag descTag; - icbtag icbTag; - __le32 lengthAllocDescs; - uint8_t allocDescs[0]; +struct unallocSpaceEntry { + tag descTag; + icbtag icbTag; + __le32 lengthAllocDescs; + uint8_t allocDescs[0]; } __attribute__ ((packed)); /* Space Bitmap Descriptor (ECMA 167r3 4/14.12) */ -struct spaceBitmapDesc -{ - tag descTag; - __le32 numOfBits; - __le32 numOfBytes; - uint8_t bitmap[0]; +struct spaceBitmapDesc { + tag descTag; + __le32 numOfBits; + __le32 numOfBytes; + uint8_t bitmap[0]; } __attribute__ ((packed)); /* Partition Integrity Entry (ECMA 167r3 4/14.13) */ -struct partitionIntegrityEntry -{ - tag descTag; - icbtag icbTag; - timestamp recordingDateAndTime; - uint8_t integrityType; - uint8_t reserved[175]; - regid impIdent; - uint8_t impUse[256]; +struct partitionIntegrityEntry { + tag descTag; + icbtag icbTag; + timestamp recordingDateAndTime; + uint8_t integrityType; + uint8_t reserved[175]; + regid impIdent; + uint8_t impUse[256]; } __attribute__ ((packed)); /* Short Allocation Descriptor (ECMA 167r3 4/14.14.1) */ @@ -815,50 +763,47 @@ struct partitionIntegrityEntry /* Extended Allocation Descriptor (ECMA 167r3 4/14.14.3) */ /* Logical Volume Header Descriptor (ECMA 167r3 4/14.15) */ -struct logicalVolHeaderDesc -{ - __le64 uniqueID; - uint8_t reserved[24]; +struct logicalVolHeaderDesc { + __le64 uniqueID; + uint8_t reserved[24]; } __attribute__ ((packed)); /* Path Component (ECMA 167r3 4/14.16.1) */ -struct pathComponent -{ - uint8_t componentType; - uint8_t lengthComponentIdent; - __le16 componentFileVersionNum; - dstring componentIdent[0]; +struct pathComponent { + uint8_t componentType; + uint8_t lengthComponentIdent; + __le16 componentFileVersionNum; + dstring componentIdent[0]; } __attribute__ ((packed)); /* File Entry (ECMA 167r3 4/14.17) */ -struct extendedFileEntry -{ - tag descTag; - icbtag icbTag; - __le32 uid; - __le32 gid; - __le32 permissions; - __le16 fileLinkCount; - uint8_t recordFormat; - uint8_t recordDisplayAttr; - __le32 recordLength; - __le64 informationLength; - __le64 objectSize; - __le64 logicalBlocksRecorded; - timestamp accessTime; - timestamp modificationTime; - timestamp createTime; - timestamp attrTime; - __le32 checkpoint; - __le32 reserved; - long_ad extendedAttrICB; - long_ad streamDirectoryICB; - regid impIdent; - __le64 uniqueID; - __le32 lengthExtendedAttr; - __le32 lengthAllocDescs; - uint8_t extendedAttr[0]; - uint8_t allocDescs[0]; -} __attribute__ ((packed)); - -#endif /* _ECMA_167_H */ +struct extendedFileEntry { + tag descTag; + icbtag icbTag; + __le32 uid; + __le32 gid; + __le32 permissions; + __le16 fileLinkCount; + uint8_t recordFormat; + uint8_t recordDisplayAttr; + __le32 recordLength; + __le64 informationLength; + __le64 objectSize; + __le64 logicalBlocksRecorded; + timestamp accessTime; + timestamp modificationTime; + timestamp createTime; + timestamp attrTime; + __le32 checkpoint; + __le32 reserved; + long_ad extendedAttrICB; + long_ad streamDirectoryICB; + regid impIdent; + __le64 uniqueID; + __le32 lengthExtendedAttr; + __le32 lengthAllocDescs; + uint8_t extendedAttr[0]; + uint8_t allocDescs[0]; +} __attribute__ ((packed)); + +#endif /* _ECMA_167_H */ diff --git a/fs/udf/file.c b/fs/udf/file.c index df070bee8d4..67bf36bd3e6 100644 --- a/fs/udf/file.c +++ b/fs/udf/file.c @@ -30,7 +30,7 @@ #include <linux/udf_fs.h> #include <asm/uaccess.h> #include <linux/kernel.h> -#include <linux/string.h> /* memset */ +#include <linux/string.h> /* memset */ #include <linux/capability.h> #include <linux/errno.h> #include <linux/smp_lock.h> @@ -41,7 +41,7 @@ #include "udf_i.h" #include "udf_sb.h" -static int udf_adinicb_readpage(struct file *file, struct page * page) +static int udf_adinicb_readpage(struct file *file, struct page *page) { struct inode *inode = page->mapping->host; char *kaddr; @@ -58,7 +58,8 @@ static int udf_adinicb_readpage(struct file *file, struct page * page) return 0; } -static int udf_adinicb_writepage(struct page *page, struct writeback_control *wbc) +static int udf_adinicb_writepage(struct page *page, + struct writeback_control *wbc) { struct inode *inode = page->mapping->host; char *kaddr; @@ -74,19 +75,21 @@ static int udf_adinicb_writepage(struct page *page, struct writeback_control *wb return 0; } -static int udf_adinicb_prepare_write(struct file *file, struct page *page, unsigned offset, unsigned to) +static int udf_adinicb_prepare_write(struct file *file, struct page *page, + unsigned offset, unsigned to) { kmap(page); return 0; } -static int udf_adinicb_commit_write(struct file *file, struct page *page, unsigned offset, unsigned to) +static int udf_adinicb_commit_write(struct file *file, struct page *page, + unsigned offset, unsigned to) { struct inode *inode = page->mapping->host; char *kaddr = page_address(page); memcpy(UDF_I_DATA(inode) + UDF_I_LENEATTR(inode) + offset, - kaddr + offset, to - offset); + kaddr + offset, to - offset); mark_inode_dirty(inode); SetPageUptodate(page); kunmap(page); @@ -97,15 +100,15 @@ static int udf_adinicb_commit_write(struct file *file, struct page *page, unsign } const struct address_space_operations udf_adinicb_aops = { - .readpage = udf_adinicb_readpage, - .writepage = udf_adinicb_writepage, - .sync_page = block_sync_page, - .prepare_write = udf_adinicb_prepare_write, - .commit_write = udf_adinicb_commit_write, + .readpage = udf_adinicb_readpage, + .writepage = udf_adinicb_writepage, + .sync_page = block_sync_page, + .prepare_write = udf_adinicb_prepare_write, + .commit_write = udf_adinicb_commit_write, }; static ssize_t udf_file_aio_write(struct kiocb *iocb, const struct iovec *iov, - unsigned long nr_segs, loff_t ppos) + unsigned long nr_segs, loff_t ppos) { ssize_t retval; struct file *file = iocb->ki_filp; @@ -113,25 +116,20 @@ static ssize_t udf_file_aio_write(struct kiocb *iocb, const struct iovec *iov, int err, pos; size_t count = iocb->ki_left; - if (UDF_I_ALLOCTYPE(inode) == ICBTAG_FLAG_AD_IN_ICB) - { + if (UDF_I_ALLOCTYPE(inode) == ICBTAG_FLAG_AD_IN_ICB) { if (file->f_flags & O_APPEND) pos = inode->i_size; else pos = ppos; - if (inode->i_sb->s_blocksize < (udf_file_entry_alloc_offset(inode) + - pos + count)) - { + if (inode->i_sb->s_blocksize < + (udf_file_entry_alloc_offset(inode) + pos + count)) { udf_expand_file_adinicb(inode, pos + count, &err); - if (UDF_I_ALLOCTYPE(inode) == ICBTAG_FLAG_AD_IN_ICB) - { + if (UDF_I_ALLOCTYPE(inode) == ICBTAG_FLAG_AD_IN_ICB) { udf_debug("udf_expand_adinicb: err=%d\n", err); return err; } - } - else - { + } else { if (pos + count > inode->i_size) UDF_I_LENALLOC(inode) = pos + count; else @@ -181,48 +179,47 @@ static ssize_t udf_file_aio_write(struct kiocb *iocb, const struct iovec *iov, * Written, tested, and released. */ int udf_ioctl(struct inode *inode, struct file *filp, unsigned int cmd, - unsigned long arg) + unsigned long arg) { int result = -EINVAL; - if ( file_permission(filp, MAY_READ) != 0 ) - { - udf_debug("no permission to access inode %lu\n", - inode->i_ino); + if (file_permission(filp, MAY_READ) != 0) { + udf_debug("no permission to access inode %lu\n", inode->i_ino); return -EPERM; } - if ( !arg ) - { + if (!arg) { udf_debug("invalid argument to udf_ioctl\n"); return -EINVAL; } - switch (cmd) - { - case UDF_GETVOLIDENT: - return copy_to_user((char __user *)arg, - UDF_SB_VOLIDENT(inode->i_sb), 32) ? -EFAULT : 0; - case UDF_RELOCATE_BLOCKS: + switch (cmd) { + case UDF_GETVOLIDENT: + return copy_to_user((char __user *)arg, + UDF_SB_VOLIDENT(inode->i_sb), + 32) ? -EFAULT : 0; + case UDF_RELOCATE_BLOCKS: { long old, new; - if (!capable(CAP_SYS_ADMIN)) return -EACCES; - if (get_user(old, (long __user *)arg)) return -EFAULT; + if (!capable(CAP_SYS_ADMIN)) + return -EACCES; + if (get_user(old, (long __user *)arg)) + return -EFAULT; if ((result = udf_relocate_blocks(inode->i_sb, - old, &new)) == 0) + old, &new)) == 0) result = put_user(new, (long __user *)arg); return result; } - case UDF_GETEASIZE: - result = put_user(UDF_I_LENEATTR(inode), (int __user *)arg); - break; + case UDF_GETEASIZE: + result = put_user(UDF_I_LENEATTR(inode), (int __user *)arg); + break; - case UDF_GETEABLOCK: - result = copy_to_user((char __user *)arg, UDF_I_DATA(inode), - UDF_I_LENEATTR(inode)) ? -EFAULT : 0; - break; + case UDF_GETEABLOCK: + result = copy_to_user((char __user *)arg, UDF_I_DATA(inode), + UDF_I_LENEATTR(inode)) ? -EFAULT : 0; + break; } return result; @@ -240,10 +237,9 @@ int udf_ioctl(struct inode *inode, struct file *filp, unsigned int cmd, * HISTORY * */ -static int udf_release_file(struct inode * inode, struct file * filp) +static int udf_release_file(struct inode *inode, struct file *filp) { - if (filp->f_mode & FMODE_WRITE) - { + if (filp->f_mode & FMODE_WRITE) { lock_kernel(); udf_discard_prealloc(inode); unlock_kernel(); @@ -252,18 +248,18 @@ static int udf_release_file(struct inode * inode, struct file * filp) } const struct file_operations udf_file_operations = { - .read = do_sync_read, - .aio_read = generic_file_aio_read, - .ioctl = udf_ioctl, - .open = generic_file_open, - .mmap = generic_file_mmap, - .write = do_sync_write, - .aio_write = udf_file_aio_write, - .release = udf_release_file, - .fsync = udf_fsync_file, - .splice_read = generic_file_splice_read, + .read = do_sync_read, + .aio_read = generic_file_aio_read, + .ioctl = udf_ioctl, + .open = generic_file_open, + .mmap = generic_file_mmap, + .write = do_sync_write, + .aio_write = udf_file_aio_write, + .release = udf_release_file, + .fsync = udf_fsync_file, + .splice_read = generic_file_splice_read, }; const struct inode_operations udf_file_inode_operations = { - .truncate = udf_truncate, + .truncate = udf_truncate, }; diff --git a/fs/udf/fsync.c b/fs/udf/fsync.c index 6ded93e7c44..7f0901c4f1f 100644 --- a/fs/udf/fsync.c +++ b/fs/udf/fsync.c @@ -29,7 +29,7 @@ static int udf_fsync_inode(struct inode *, int); * even pass file to fsync ? */ -int udf_fsync_file(struct file * file, struct dentry *dentry, int datasync) +int udf_fsync_file(struct file *file, struct dentry *dentry, int datasync) { struct inode *inode = dentry->d_inode; return udf_fsync_inode(inode, datasync); @@ -45,6 +45,6 @@ static int udf_fsync_inode(struct inode *inode, int datasync) if (datasync && !(inode->i_state & I_DIRTY_DATASYNC)) return err; - err |= udf_sync_inode (inode); + err |= udf_sync_inode(inode); return err ? -EIO : 0; } diff --git a/fs/udf/ialloc.c b/fs/udf/ialloc.c index 10f3188738a..2eb503806bc 100644 --- a/fs/udf/ialloc.c +++ b/fs/udf/ialloc.c @@ -28,7 +28,7 @@ #include "udf_i.h" #include "udf_sb.h" -void udf_free_inode(struct inode * inode) +void udf_free_inode(struct inode *inode) { struct super_block *sb = inode->i_sb; struct udf_sb_info *sbi = UDF_SB(sb); @@ -46,10 +46,12 @@ void udf_free_inode(struct inode * inode) if (sbi->s_lvidbh) { if (S_ISDIR(inode->i_mode)) UDF_SB_LVIDIU(sb)->numDirs = - cpu_to_le32(le32_to_cpu(UDF_SB_LVIDIU(sb)->numDirs) - 1); + cpu_to_le32(le32_to_cpu(UDF_SB_LVIDIU(sb)->numDirs) + - 1); else UDF_SB_LVIDIU(sb)->numFiles = - cpu_to_le32(le32_to_cpu(UDF_SB_LVIDIU(sb)->numFiles) - 1); + cpu_to_le32(le32_to_cpu(UDF_SB_LVIDIU(sb)->numFiles) + - 1); mark_buffer_dirty(sbi->s_lvidbh); } @@ -58,18 +60,17 @@ void udf_free_inode(struct inode * inode) udf_free_blocks(sb, NULL, UDF_I_LOCATION(inode), 0, 1); } -struct inode * udf_new_inode (struct inode *dir, int mode, int * err) +struct inode *udf_new_inode(struct inode *dir, int mode, int *err) { struct super_block *sb = dir->i_sb; struct udf_sb_info *sbi = UDF_SB(sb); - struct inode * inode; + struct inode *inode; int block; uint32_t start = UDF_I_LOCATION(dir).logicalBlockNum; inode = new_inode(sb); - if (!inode) - { + if (!inode) { *err = -ENOMEM; return NULL; } @@ -81,26 +82,30 @@ struct inode * udf_new_inode (struct inode *dir, int mode, int * err) UDF_I_NEXT_ALLOC_GOAL(inode) = 0; UDF_I_STRAT4096(inode) = 0; - block = udf_new_block(dir->i_sb, NULL, UDF_I_LOCATION(dir).partitionReferenceNum, - start, err); - if (*err) - { + block = + udf_new_block(dir->i_sb, NULL, + UDF_I_LOCATION(dir).partitionReferenceNum, start, + err); + if (*err) { iput(inode); return NULL; } mutex_lock(&sbi->s_alloc_mutex); - if (UDF_SB_LVIDBH(sb)) - { + if (UDF_SB_LVIDBH(sb)) { struct logicalVolHeaderDesc *lvhd; uint64_t uniqueID; - lvhd = (struct logicalVolHeaderDesc *)(UDF_SB_LVID(sb)->logicalVolContentsUse); + lvhd = + (struct logicalVolHeaderDesc *)(UDF_SB_LVID(sb)-> + logicalVolContentsUse); if (S_ISDIR(mode)) UDF_SB_LVIDIU(sb)->numDirs = - cpu_to_le32(le32_to_cpu(UDF_SB_LVIDIU(sb)->numDirs) + 1); + cpu_to_le32(le32_to_cpu(UDF_SB_LVIDIU(sb)->numDirs) + + 1); else UDF_SB_LVIDIU(sb)->numFiles = - cpu_to_le32(le32_to_cpu(UDF_SB_LVIDIU(sb)->numFiles) + 1); + cpu_to_le32(le32_to_cpu(UDF_SB_LVIDIU(sb)->numFiles) + + 1); UDF_I_UNIQUE(inode) = uniqueID = le64_to_cpu(lvhd->uniqueID); if (!(++uniqueID & 0x00000000FFFFFFFFUL)) uniqueID += 16; @@ -109,35 +114,34 @@ struct inode * udf_new_inode (struct inode *dir, int mode, int * err) } inode->i_mode = mode; inode->i_uid = current->fsuid; - if (dir->i_mode & S_ISGID) - { + if (dir->i_mode & S_ISGID) { inode->i_gid = dir->i_gid; if (S_ISDIR(mode)) mode |= S_ISGID; - } - else + } else inode->i_gid = current->fsgid; UDF_I_LOCATION(inode).logicalBlockNum = block; - UDF_I_LOCATION(inode).partitionReferenceNum = UDF_I_LOCATION(dir).partitionReferenceNum; + UDF_I_LOCATION(inode).partitionReferenceNum = + UDF_I_LOCATION(dir).partitionReferenceNum; inode->i_ino = udf_get_lb_pblock(sb, UDF_I_LOCATION(inode), 0); inode->i_blocks = 0; UDF_I_LENEATTR(inode) = 0; UDF_I_LENALLOC(inode) = 0; UDF_I_USE(inode) = 0; - if (UDF_QUERY_FLAG(inode->i_sb, UDF_FLAG_USE_EXTENDED_FE)) - { + if (UDF_QUERY_FLAG(inode->i_sb, UDF_FLAG_USE_EXTENDED_FE)) { UDF_I_EFE(inode) = 1; UDF_UPDATE_UDFREV(inode->i_sb, UDF_VERS_USE_EXTENDED_FE); - UDF_I_DATA(inode) = kzalloc(inode->i_sb->s_blocksize - sizeof(struct extendedFileEntry), GFP_KERNEL); - } - else - { + UDF_I_DATA(inode) = + kzalloc(inode->i_sb->s_blocksize - + sizeof(struct extendedFileEntry), GFP_KERNEL); + } else { UDF_I_EFE(inode) = 0; - UDF_I_DATA(inode) = kzalloc(inode->i_sb->s_blocksize - sizeof(struct fileEntry), GFP_KERNEL); + UDF_I_DATA(inode) = + kzalloc(inode->i_sb->s_blocksize - sizeof(struct fileEntry), + GFP_KERNEL); } - if (!UDF_I_DATA(inode)) - { + if (!UDF_I_DATA(inode)) { iput(inode); *err = -ENOMEM; mutex_unlock(&sbi->s_alloc_mutex); @@ -150,13 +154,12 @@ struct inode * udf_new_inode (struct inode *dir, int mode, int * err) else UDF_I_ALLOCTYPE(inode) = ICBTAG_FLAG_AD_LONG; inode->i_mtime = inode->i_atime = inode->i_ctime = - UDF_I_CRTIME(inode) = current_fs_time(inode->i_sb); + UDF_I_CRTIME(inode) = current_fs_time(inode->i_sb); insert_inode_hash(inode); mark_inode_dirty(inode); mutex_unlock(&sbi->s_alloc_mutex); - if (DQUOT_ALLOC_INODE(inode)) - { + if (DQUOT_ALLOC_INODE(inode)) { DQUOT_DROP(inode); inode->i_flags |= S_NOQUOTA; inode->i_nlink = 0; diff --git a/fs/udf/inode.c b/fs/udf/inode.c index 5b82e489af7..be6326f449a 100644 --- a/fs/udf/inode.c +++ b/fs/udf/inode.c @@ -51,18 +51,18 @@ static int udf_update_inode(struct inode *, int); static void udf_fill_inode(struct inode *, struct buffer_head *); static int udf_alloc_i_data(struct inode *inode, size_t size); static struct buffer_head *inode_getblk(struct inode *, sector_t, int *, - long *, int *); + long *, int *); static int8_t udf_insert_aext(struct inode *, struct extent_position, - kernel_lb_addr, uint32_t); + kernel_lb_addr, uint32_t); static void udf_split_extents(struct inode *, int *, int, int, - kernel_long_ad [EXTENT_MERGE_SIZE], int *); + kernel_long_ad[EXTENT_MERGE_SIZE], int *); static void udf_prealloc_extents(struct inode *, int, int, - kernel_long_ad [EXTENT_MERGE_SIZE], int *); + kernel_long_ad[EXTENT_MERGE_SIZE], int *); static void udf_merge_extents(struct inode *, - kernel_long_ad [EXTENT_MERGE_SIZE], int *); + kernel_long_ad[EXTENT_MERGE_SIZE], int *); static void udf_update_extents(struct inode *, - kernel_long_ad [EXTENT_MERGE_SIZE], int, int, - struct extent_position *); + kernel_long_ad[EXTENT_MERGE_SIZE], int, int, + struct extent_position *); static int udf_get_block(struct inode *, sector_t, struct buffer_head *, int); /* @@ -81,7 +81,7 @@ static int udf_get_block(struct inode *, sector_t, struct buffer_head *, int); * * Called at the last iput() if i_nlink is zero. */ -void udf_delete_inode(struct inode * inode) +void udf_delete_inode(struct inode *inode) { truncate_inode_pages(&inode->i_data, 0); @@ -97,7 +97,7 @@ void udf_delete_inode(struct inode * inode) unlock_kernel(); return; -no_delete: + no_delete: clear_inode(inode); } @@ -132,26 +132,27 @@ static int udf_readpage(struct file *file, struct page *page) return block_read_full_page(page, udf_get_block); } -static int udf_prepare_write(struct file *file, struct page *page, unsigned from, unsigned to) +static int udf_prepare_write(struct file *file, struct page *page, + unsigned from, unsigned to) { return block_prepare_write(page, from, to, udf_get_block); } static sector_t udf_bmap(struct address_space *mapping, sector_t block) { - return generic_block_bmap(mapping,block,udf_get_block); + return generic_block_bmap(mapping, block, udf_get_block); } const struct address_space_operations udf_aops = { - .readpage = udf_readpage, - .writepage = udf_writepage, - .sync_page = block_sync_page, - .prepare_write = udf_prepare_write, - .commit_write = generic_commit_write, - .bmap = udf_bmap, + .readpage = udf_readpage, + .writepage = udf_writepage, + .sync_page = block_sync_page, + .prepare_write = udf_prepare_write, + .commit_write = generic_commit_write, + .bmap = udf_bmap, }; -void udf_expand_file_adinicb(struct inode * inode, int newsize, int * err) +void udf_expand_file_adinicb(struct inode *inode, int newsize, int *err) { struct page *page; char *kaddr; @@ -163,8 +164,7 @@ void udf_expand_file_adinicb(struct inode * inode, int newsize, int * err) /* from now on we have normal address_space methods */ inode->i_data.a_ops = &udf_aops; - if (!UDF_I_LENALLOC(inode)) - { + if (!UDF_I_LENALLOC(inode)) { if (UDF_QUERY_FLAG(inode->i_sb, UDF_FLAG_USE_SHORT_AD)) UDF_I_ALLOCTYPE(inode) = ICBTAG_FLAG_AD_SHORT; else @@ -176,19 +176,18 @@ void udf_expand_file_adinicb(struct inode * inode, int newsize, int * err) page = grab_cache_page(inode->i_mapping, 0); BUG_ON(!PageLocked(page)); - if (!PageUptodate(page)) - { + if (!PageUptodate(page)) { kaddr = kmap(page); memset(kaddr + UDF_I_LENALLOC(inode), 0x00, - PAGE_CACHE_SIZE - UDF_I_LENALLOC(inode)); + PAGE_CACHE_SIZE - UDF_I_LENALLOC(inode)); memcpy(kaddr, UDF_I_DATA(inode) + UDF_I_LENEATTR(inode), - UDF_I_LENALLOC(inode)); + UDF_I_LENALLOC(inode)); flush_dcache_page(page); SetPageUptodate(page); kunmap(page); } memset(UDF_I_DATA(inode) + UDF_I_LENEATTR(inode), 0x00, - UDF_I_LENALLOC(inode)); + UDF_I_LENALLOC(inode)); UDF_I_LENALLOC(inode) = 0; if (UDF_QUERY_FLAG(inode->i_sb, UDF_FLAG_USE_SHORT_AD)) UDF_I_ALLOCTYPE(inode) = ICBTAG_FLAG_AD_SHORT; @@ -201,7 +200,8 @@ void udf_expand_file_adinicb(struct inode * inode, int newsize, int * err) mark_inode_dirty(inode); } -struct buffer_head * udf_expand_dir_adinicb(struct inode *inode, int *block, int *err) +struct buffer_head *udf_expand_dir_adinicb(struct inode *inode, int *block, + int *err) { int newblock; struct buffer_head *dbh = NULL; @@ -220,8 +220,7 @@ struct buffer_head * udf_expand_dir_adinicb(struct inode *inode, int *block, int else alloctype = ICBTAG_FLAG_AD_LONG; - if (!inode->i_size) - { + if (!inode->i_size) { UDF_I_ALLOCTYPE(inode) = alloctype; mark_inode_dirty(inode); return NULL; @@ -229,13 +228,14 @@ struct buffer_head * udf_expand_dir_adinicb(struct inode *inode, int *block, int /* alloc block, and copy data to it */ *block = udf_new_block(inode->i_sb, inode, - UDF_I_LOCATION(inode).partitionReferenceNum, - UDF_I_LOCATION(inode).logicalBlockNum, err); + UDF_I_LOCATION(inode).partitionReferenceNum, + UDF_I_LOCATION(inode).logicalBlockNum, err); if (!(*block)) return NULL; newblock = udf_get_pblock(inode->i_sb, *block, - UDF_I_LOCATION(inode).partitionReferenceNum, 0); + UDF_I_LOCATION(inode).partitionReferenceNum, + 0); if (!newblock) return NULL; dbh = udf_tgetblk(inode->i_sb, newblock); @@ -247,16 +247,17 @@ struct buffer_head * udf_expand_dir_adinicb(struct inode *inode, int *block, int unlock_buffer(dbh); mark_buffer_dirty_inode(dbh, inode); - sfibh.soffset = sfibh.eoffset = (f_pos & ((inode->i_sb->s_blocksize - 1) >> 2)) << 2; + sfibh.soffset = sfibh.eoffset = + (f_pos & ((inode->i_sb->s_blocksize - 1) >> 2)) << 2; sfibh.sbh = sfibh.ebh = NULL; dfibh.soffset = dfibh.eoffset = 0; dfibh.sbh = dfibh.ebh = dbh; - while ( (f_pos < size) ) - { + while ((f_pos < size)) { UDF_I_ALLOCTYPE(inode) = ICBTAG_FLAG_AD_IN_ICB; - sfi = udf_fileident_read(inode, &f_pos, &sfibh, &cfi, NULL, NULL, NULL, NULL); - if (!sfi) - { + sfi = + udf_fileident_read(inode, &f_pos, &sfibh, &cfi, NULL, NULL, + NULL, NULL); + if (!sfi) { brelse(dbh); return NULL; } @@ -266,8 +267,8 @@ struct buffer_head * udf_expand_dir_adinicb(struct inode *inode, int *block, int dfibh.eoffset += (sfibh.eoffset - sfibh.soffset); dfi = (struct fileIdentDesc *)(dbh->b_data + dfibh.soffset); if (udf_write_fi(inode, sfi, dfi, &dfibh, sfi->impUse, - sfi->fileIdent + le16_to_cpu(sfi->lengthOfImpUse))) - { + sfi->fileIdent + + le16_to_cpu(sfi->lengthOfImpUse))) { UDF_I_ALLOCTYPE(inode) = ICBTAG_FLAG_AD_IN_ICB; brelse(dbh); return NULL; @@ -275,10 +276,12 @@ struct buffer_head * udf_expand_dir_adinicb(struct inode *inode, int *block, int } mark_buffer_dirty_inode(dbh, inode); - memset(UDF_I_DATA(inode) + UDF_I_LENEATTR(inode), 0, UDF_I_LENALLOC(inode)); + memset(UDF_I_DATA(inode) + UDF_I_LENEATTR(inode), 0, + UDF_I_LENALLOC(inode)); UDF_I_LENALLOC(inode) = 0; eloc.logicalBlockNum = *block; - eloc.partitionReferenceNum = UDF_I_LOCATION(inode).partitionReferenceNum; + eloc.partitionReferenceNum = + UDF_I_LOCATION(inode).partitionReferenceNum; elen = inode->i_size; UDF_I_LENEXTENTS(inode) = elen; epos.bh = NULL; @@ -292,14 +295,14 @@ struct buffer_head * udf_expand_dir_adinicb(struct inode *inode, int *block, int return dbh; } -static int udf_get_block(struct inode *inode, sector_t block, struct buffer_head *bh_result, int create) +static int udf_get_block(struct inode *inode, sector_t block, + struct buffer_head *bh_result, int create) { int err, new; struct buffer_head *bh; unsigned long phys; - if (!create) - { + if (!create) { phys = udf_block_map(inode, block); if (phys) map_bh(bh_result, inode->i_sb, phys); @@ -315,10 +318,9 @@ static int udf_get_block(struct inode *inode, sector_t block, struct buffer_head if (block < 0) goto abort_negative; - if (block == UDF_I_NEXT_ALLOC_BLOCK(inode) + 1) - { - UDF_I_NEXT_ALLOC_BLOCK(inode) ++; - UDF_I_NEXT_ALLOC_GOAL(inode) ++; + if (block == UDF_I_NEXT_ALLOC_BLOCK(inode) + 1) { + UDF_I_NEXT_ALLOC_BLOCK(inode)++; + UDF_I_NEXT_ALLOC_GOAL(inode)++; } err = 0; @@ -332,29 +334,27 @@ static int udf_get_block(struct inode *inode, sector_t block, struct buffer_head if (new) set_buffer_new(bh_result); map_bh(bh_result, inode->i_sb, phys); -abort: + abort: unlock_kernel(); return err; -abort_negative: + abort_negative: udf_warning(inode->i_sb, "udf_get_block", "block < 0"); goto abort; } -static struct buffer_head * -udf_getblk(struct inode *inode, long block, int create, int *err) +static struct buffer_head *udf_getblk(struct inode *inode, long block, + int create, int *err) { struct buffer_head dummy; dummy.b_state = 0; dummy.b_blocknr = -1000; *err = udf_get_block(inode, block, &dummy, create); - if (!*err && buffer_mapped(&dummy)) - { + if (!*err && buffer_mapped(&dummy)) { struct buffer_head *bh; bh = sb_getblk(inode->i_sb, dummy.b_blocknr); - if (buffer_new(&dummy)) - { + if (buffer_new(&dummy)) { lock_buffer(bh); memset(bh->b_data, 0x00, inode->i_sb->s_blocksize); set_buffer_uptodate(bh); @@ -368,12 +368,12 @@ udf_getblk(struct inode *inode, long block, int create, int *err) /* Extend the file by 'blocks' blocks, return the number of extents added */ int udf_extend_file(struct inode *inode, struct extent_position *last_pos, - kernel_long_ad *last_ext, sector_t blocks) + kernel_long_ad * last_ext, sector_t blocks) { sector_t add; int count = 0, fake = !(last_ext->extLength & UDF_EXTENT_LENGTH_MASK); struct super_block *sb = inode->i_sb; - kernel_lb_addr prealloc_loc = {0, 0}; + kernel_lb_addr prealloc_loc = { 0, 0 }; int prealloc_len = 0; /* The previous extent is fake and we should not extend by anything @@ -383,28 +383,32 @@ int udf_extend_file(struct inode *inode, struct extent_position *last_pos, /* Round the last extent up to a multiple of block size */ if (last_ext->extLength & (sb->s_blocksize - 1)) { last_ext->extLength = - (last_ext->extLength & UDF_EXTENT_FLAG_MASK) | - (((last_ext->extLength & UDF_EXTENT_LENGTH_MASK) + - sb->s_blocksize - 1) & ~(sb->s_blocksize - 1)); + (last_ext->extLength & UDF_EXTENT_FLAG_MASK) | + (((last_ext->extLength & UDF_EXTENT_LENGTH_MASK) + + sb->s_blocksize - 1) & ~(sb->s_blocksize - 1)); UDF_I_LENEXTENTS(inode) = - (UDF_I_LENEXTENTS(inode) + sb->s_blocksize - 1) & - ~(sb->s_blocksize - 1); + (UDF_I_LENEXTENTS(inode) + sb->s_blocksize - 1) & + ~(sb->s_blocksize - 1); } /* Last extent are just preallocated blocks? */ - if ((last_ext->extLength & UDF_EXTENT_FLAG_MASK) == EXT_NOT_RECORDED_ALLOCATED) { + if ((last_ext->extLength & UDF_EXTENT_FLAG_MASK) == + EXT_NOT_RECORDED_ALLOCATED) { /* Save the extent so that we can reattach it to the end */ prealloc_loc = last_ext->extLocation; prealloc_len = last_ext->extLength; /* Mark the extent as a hole */ last_ext->extLength = EXT_NOT_RECORDED_NOT_ALLOCATED | - (last_ext->extLength & UDF_EXTENT_LENGTH_MASK); + (last_ext->extLength & UDF_EXTENT_LENGTH_MASK); last_ext->extLocation.logicalBlockNum = 0; - last_ext->extLocation.partitionReferenceNum = 0; + last_ext->extLocation.partitionReferenceNum = 0; } /* Can we merge with the previous extent? */ - if ((last_ext->extLength & UDF_EXTENT_FLAG_MASK) == EXT_NOT_RECORDED_NOT_ALLOCATED) { - add = ((1<<30) - sb->s_blocksize - (last_ext->extLength & - UDF_EXTENT_LENGTH_MASK)) >> sb->s_blocksize_bits; + if ((last_ext->extLength & UDF_EXTENT_FLAG_MASK) == + EXT_NOT_RECORDED_NOT_ALLOCATED) { + add = + ((1 << 30) - sb->s_blocksize - + (last_ext->extLength & UDF_EXTENT_LENGTH_MASK)) >> sb-> + s_blocksize_bits; if (add > blocks) add = blocks; blocks -= add; @@ -413,40 +417,42 @@ int udf_extend_file(struct inode *inode, struct extent_position *last_pos, if (fake) { udf_add_aext(inode, last_pos, last_ext->extLocation, - last_ext->extLength, 1); + last_ext->extLength, 1); count++; - } - else - udf_write_aext(inode, last_pos, last_ext->extLocation, last_ext->extLength, 1); + } else + udf_write_aext(inode, last_pos, last_ext->extLocation, + last_ext->extLength, 1); /* Managed to do everything necessary? */ if (!blocks) goto out; /* All further extents will be NOT_RECORDED_NOT_ALLOCATED */ last_ext->extLocation.logicalBlockNum = 0; - last_ext->extLocation.partitionReferenceNum = 0; - add = (1 << (30-sb->s_blocksize_bits)) - 1; - last_ext->extLength = EXT_NOT_RECORDED_NOT_ALLOCATED | (add << sb->s_blocksize_bits); + last_ext->extLocation.partitionReferenceNum = 0; + add = (1 << (30 - sb->s_blocksize_bits)) - 1; + last_ext->extLength = + EXT_NOT_RECORDED_NOT_ALLOCATED | (add << sb->s_blocksize_bits); /* Create enough extents to cover the whole hole */ while (blocks > add) { blocks -= add; if (udf_add_aext(inode, last_pos, last_ext->extLocation, - last_ext->extLength, 1) == -1) + last_ext->extLength, 1) == -1) return -1; count++; } if (blocks) { last_ext->extLength = EXT_NOT_RECORDED_NOT_ALLOCATED | - (blocks << sb->s_blocksize_bits); + (blocks << sb->s_blocksize_bits); if (udf_add_aext(inode, last_pos, last_ext->extLocation, - last_ext->extLength, 1) == -1) + last_ext->extLength, 1) == -1) return -1; count++; } -out: + out: /* Do we have some preallocated blocks saved? */ if (prealloc_len) { - if (udf_add_aext(inode, last_pos, prealloc_loc, prealloc_len, 1) == -1) + if (udf_add_aext(inode, last_pos, prealloc_loc, prealloc_len, 1) + == -1) return -1; last_ext->extLocation = prealloc_loc; last_ext->extLength = prealloc_len; @@ -462,8 +468,8 @@ out: return count; } -static struct buffer_head * inode_getblk(struct inode * inode, sector_t block, - int *err, long *phys, int *new) +static struct buffer_head *inode_getblk(struct inode *inode, sector_t block, + int *err, long *phys, int *new) { static sector_t last_block; struct buffer_head *result = NULL; @@ -484,21 +490,18 @@ static struct buffer_head * inode_getblk(struct inode * inode, sector_t block, prev_epos.block = UDF_I_LOCATION(inode); prev_epos.bh = NULL; cur_epos = next_epos = prev_epos; - b_off = (loff_t)block << inode->i_sb->s_blocksize_bits; + b_off = (loff_t) block << inode->i_sb->s_blocksize_bits; /* find the extent which contains the block we are looking for. - alternate between laarr[0] and laarr[1] for locations of the - current extent, and the previous extent */ - do - { - if (prev_epos.bh != cur_epos.bh) - { + alternate between laarr[0] and laarr[1] for locations of the + current extent, and the previous extent */ + do { + if (prev_epos.bh != cur_epos.bh) { brelse(prev_epos.bh); get_bh(cur_epos.bh); prev_epos.bh = cur_epos.bh; } - if (cur_epos.bh != next_epos.bh) - { + if (cur_epos.bh != next_epos.bh) { brelse(cur_epos.bh); get_bh(next_epos.bh); cur_epos.bh = next_epos.bh; @@ -512,7 +515,8 @@ static struct buffer_head * inode_getblk(struct inode * inode, sector_t block, prev_epos.offset = cur_epos.offset; cur_epos.offset = next_epos.offset; - if ((etype = udf_next_aext(inode, &next_epos, &eloc, &elen, 1)) == -1) + if ((etype = + udf_next_aext(inode, &next_epos, &eloc, &elen, 1)) == -1) break; c = !c; @@ -522,10 +526,10 @@ static struct buffer_head * inode_getblk(struct inode * inode, sector_t block, if (etype != (EXT_NOT_RECORDED_NOT_ALLOCATED >> 30)) pgoal = eloc.logicalBlockNum + - ((elen + inode->i_sb->s_blocksize - 1) >> - inode->i_sb->s_blocksize_bits); + ((elen + inode->i_sb->s_blocksize - 1) >> + inode->i_sb->s_blocksize_bits); - count ++; + count++; } while (lbcount + elen <= b_off); b_off -= lbcount; @@ -538,15 +542,13 @@ static struct buffer_head * inode_getblk(struct inode * inode, sector_t block, udf_next_aext(inode, &cur_epos, &tmpeloc, &tmpelen, 0); /* if the extent is allocated and recorded, return the block - if the extent is not a multiple of the blocksize, round up */ + if the extent is not a multiple of the blocksize, round up */ - if (etype == (EXT_RECORDED_ALLOCATED >> 30)) - { - if (elen & (inode->i_sb->s_blocksize - 1)) - { + if (etype == (EXT_RECORDED_ALLOCATED >> 30)) { + if (elen & (inode->i_sb->s_blocksize - 1)) { elen = EXT_RECORDED_ALLOCATED | - ((elen + inode->i_sb->s_blocksize - 1) & - ~(inode->i_sb->s_blocksize - 1)); + ((elen + inode->i_sb->s_blocksize - 1) & + ~(inode->i_sb->s_blocksize - 1)); etype = udf_write_aext(inode, &cur_epos, eloc, elen, 1); } brelse(prev_epos.bh); @@ -559,18 +561,17 @@ static struct buffer_head * inode_getblk(struct inode * inode, sector_t block, last_block = block; /* Are we beyond EOF? */ - if (etype == -1) - { + if (etype == -1) { int ret; if (count) { if (c) laarr[0] = laarr[1]; startnum = 1; - } - else { + } else { /* Create a fake extent when there's not one */ - memset(&laarr[0].extLocation, 0x00, sizeof(kernel_lb_addr)); + memset(&laarr[0].extLocation, 0x00, + sizeof(kernel_lb_addr)); laarr[0].extLength = EXT_NOT_RECORDED_NOT_ALLOCATED; /* Will udf_extend_file() create real extent from a fake one? */ startnum = (offset > 0); @@ -590,26 +591,26 @@ static struct buffer_head * inode_getblk(struct inode * inode, sector_t block, offset = 0; count += ret; /* We are not covered by a preallocated extent? */ - if ((laarr[0].extLength & UDF_EXTENT_FLAG_MASK) != EXT_NOT_RECORDED_ALLOCATED) { + if ((laarr[0].extLength & UDF_EXTENT_FLAG_MASK) != + EXT_NOT_RECORDED_ALLOCATED) { /* Is there any real extent? - otherwise we overwrite * the fake one... */ if (count) c = !c; laarr[c].extLength = EXT_NOT_RECORDED_NOT_ALLOCATED | - inode->i_sb->s_blocksize; - memset(&laarr[c].extLocation, 0x00, sizeof(kernel_lb_addr)); - count ++; - endnum ++; + inode->i_sb->s_blocksize; + memset(&laarr[c].extLocation, 0x00, + sizeof(kernel_lb_addr)); + count++; + endnum++; } - endnum = c+1; + endnum = c + 1; lastblock = 1; - } - else { + } else { endnum = startnum = ((count > 2) ? 2 : count); /* if the current extent is in position 0, swap it with the previous */ - if (!c && count != 1) - { + if (!c && count != 1) { laarr[2] = laarr[0]; laarr[0] = laarr[1]; laarr[1] = laarr[2]; @@ -617,37 +618,37 @@ static struct buffer_head * inode_getblk(struct inode * inode, sector_t block, } /* if the current block is located in an extent, read the next extent */ - if ((etype = udf_next_aext(inode, &next_epos, &eloc, &elen, 0)) != -1) - { - laarr[c+1].extLength = (etype << 30) | elen; - laarr[c+1].extLocation = eloc; - count ++; - startnum ++; - endnum ++; - } - else { + if ((etype = + udf_next_aext(inode, &next_epos, &eloc, &elen, 0)) != -1) { + laarr[c + 1].extLength = (etype << 30) | elen; + laarr[c + 1].extLocation = eloc; + count++; + startnum++; + endnum++; + } else { lastblock = 1; } } /* if the current extent is not recorded but allocated, get the - block in the extent corresponding to the requested block */ + block in the extent corresponding to the requested block */ if ((laarr[c].extLength >> 30) == (EXT_NOT_RECORDED_ALLOCATED >> 30)) newblocknum = laarr[c].extLocation.logicalBlockNum + offset; - else /* otherwise, allocate a new block */ - { + else { /* otherwise, allocate a new block */ + if (UDF_I_NEXT_ALLOC_BLOCK(inode) == block) goal = UDF_I_NEXT_ALLOC_GOAL(inode); - if (!goal) - { + if (!goal) { if (!(goal = pgoal)) - goal = UDF_I_LOCATION(inode).logicalBlockNum + 1; + goal = + UDF_I_LOCATION(inode).logicalBlockNum + 1; } if (!(newblocknum = udf_new_block(inode->i_sb, inode, - UDF_I_LOCATION(inode).partitionReferenceNum, goal, err))) - { + UDF_I_LOCATION(inode). + partitionReferenceNum, goal, + err))) { brelse(prev_epos.bh); *err = -ENOSPC; return NULL; @@ -656,8 +657,8 @@ static struct buffer_head * inode_getblk(struct inode * inode, sector_t block, } /* if the extent the requsted block is located in contains multiple blocks, - split the extent into at most three extents. blocks prior to requested - block, requested block, and blocks after requested block */ + split the extent into at most three extents. blocks prior to requested + block, requested block, and blocks after requested block */ udf_split_extents(inode, &c, offset, newblocknum, laarr, &endnum); #ifdef UDF_PREALLOCATE @@ -669,15 +670,15 @@ static struct buffer_head * inode_getblk(struct inode * inode, sector_t block, udf_merge_extents(inode, laarr, &endnum); /* write back the new extents, inserting new extents if the new number - of extents is greater than the old number, and deleting extents if - the new number of extents is less than the old number */ + of extents is greater than the old number, and deleting extents if + the new number of extents is less than the old number */ udf_update_extents(inode, laarr, startnum, endnum, &prev_epos); brelse(prev_epos.bh); if (!(newblock = udf_get_pblock(inode->i_sb, newblocknum, - UDF_I_LOCATION(inode).partitionReferenceNum, 0))) - { + UDF_I_LOCATION(inode). + partitionReferenceNum, 0))) { return NULL; } *phys = newblock; @@ -694,283 +695,329 @@ static struct buffer_head * inode_getblk(struct inode * inode, sector_t block, return result; } -static void udf_split_extents(struct inode *inode, int *c, int offset, int newblocknum, - kernel_long_ad laarr[EXTENT_MERGE_SIZE], int *endnum) +static void udf_split_extents(struct inode *inode, int *c, int offset, + int newblocknum, + kernel_long_ad laarr[EXTENT_MERGE_SIZE], + int *endnum) { if ((laarr[*c].extLength >> 30) == (EXT_NOT_RECORDED_ALLOCATED >> 30) || - (laarr[*c].extLength >> 30) == (EXT_NOT_RECORDED_NOT_ALLOCATED >> 30)) - { + (laarr[*c].extLength >> 30) == + (EXT_NOT_RECORDED_NOT_ALLOCATED >> 30)) { int curr = *c; int blen = ((laarr[curr].extLength & UDF_EXTENT_LENGTH_MASK) + - inode->i_sb->s_blocksize - 1) >> inode->i_sb->s_blocksize_bits; + inode->i_sb->s_blocksize - + 1) >> inode->i_sb->s_blocksize_bits; int8_t etype = (laarr[curr].extLength >> 30); - if (blen == 1) - ; - else if (!offset || blen == offset + 1) - { - laarr[curr+2] = laarr[curr+1]; - laarr[curr+1] = laarr[curr]; - } - else - { - laarr[curr+3] = laarr[curr+1]; - laarr[curr+2] = laarr[curr+1] = laarr[curr]; - } - - if (offset) - { - if (etype == (EXT_NOT_RECORDED_ALLOCATED >> 30)) - { - udf_free_blocks(inode->i_sb, inode, laarr[curr].extLocation, 0, offset); - laarr[curr].extLength = EXT_NOT_RECORDED_NOT_ALLOCATED | - (offset << inode->i_sb->s_blocksize_bits); + if (blen == 1) ; + else if (!offset || blen == offset + 1) { + laarr[curr + 2] = laarr[curr + 1]; + laarr[curr + 1] = laarr[curr]; + } else { + laarr[curr + 3] = laarr[curr + 1]; + laarr[curr + 2] = laarr[curr + 1] = laarr[curr]; + } + + if (offset) { + if (etype == (EXT_NOT_RECORDED_ALLOCATED >> 30)) { + udf_free_blocks(inode->i_sb, inode, + laarr[curr].extLocation, 0, + offset); + laarr[curr].extLength = + EXT_NOT_RECORDED_NOT_ALLOCATED | (offset << + inode-> + i_sb-> + s_blocksize_bits); laarr[curr].extLocation.logicalBlockNum = 0; - laarr[curr].extLocation.partitionReferenceNum = 0; - } - else + laarr[curr].extLocation.partitionReferenceNum = + 0; + } else laarr[curr].extLength = (etype << 30) | - (offset << inode->i_sb->s_blocksize_bits); - curr ++; - (*c) ++; - (*endnum) ++; + (offset << inode->i_sb->s_blocksize_bits); + curr++; + (*c)++; + (*endnum)++; } laarr[curr].extLocation.logicalBlockNum = newblocknum; if (etype == (EXT_NOT_RECORDED_NOT_ALLOCATED >> 30)) laarr[curr].extLocation.partitionReferenceNum = - UDF_I_LOCATION(inode).partitionReferenceNum; + UDF_I_LOCATION(inode).partitionReferenceNum; laarr[curr].extLength = EXT_RECORDED_ALLOCATED | - inode->i_sb->s_blocksize; - curr ++; + inode->i_sb->s_blocksize; + curr++; - if (blen != offset + 1) - { + if (blen != offset + 1) { if (etype == (EXT_NOT_RECORDED_ALLOCATED >> 30)) - laarr[curr].extLocation.logicalBlockNum += (offset + 1); - laarr[curr].extLength = (etype << 30) | - ((blen - (offset + 1)) << inode->i_sb->s_blocksize_bits); - curr ++; - (*endnum) ++; + laarr[curr].extLocation.logicalBlockNum += + (offset + 1); + laarr[curr].extLength = + (etype << 30) | ((blen - (offset + 1)) << inode-> + i_sb->s_blocksize_bits); + curr++; + (*endnum)++; } } } static void udf_prealloc_extents(struct inode *inode, int c, int lastblock, - kernel_long_ad laarr[EXTENT_MERGE_SIZE], int *endnum) + kernel_long_ad laarr[EXTENT_MERGE_SIZE], + int *endnum) { int start, length = 0, currlength = 0, i; - if (*endnum >= (c+1)) - { + if (*endnum >= (c + 1)) { if (!lastblock) return; else start = c; - } - else - { - if ((laarr[c+1].extLength >> 30) == (EXT_NOT_RECORDED_ALLOCATED >> 30)) - { - start = c+1; - length = currlength = (((laarr[c+1].extLength & UDF_EXTENT_LENGTH_MASK) + - inode->i_sb->s_blocksize - 1) >> inode->i_sb->s_blocksize_bits); - } - else + } else { + if ((laarr[c + 1].extLength >> 30) == + (EXT_NOT_RECORDED_ALLOCATED >> 30)) { + start = c + 1; + length = currlength = + (((laarr[c + 1]. + extLength & UDF_EXTENT_LENGTH_MASK) + + inode->i_sb->s_blocksize - + 1) >> inode->i_sb->s_blocksize_bits); + } else start = c; } - for (i=start+1; i<=*endnum; i++) - { - if (i == *endnum) - { + for (i = start + 1; i <= *endnum; i++) { + if (i == *endnum) { if (lastblock) length += UDF_DEFAULT_PREALLOC_BLOCKS; - } - else if ((laarr[i].extLength >> 30) == (EXT_NOT_RECORDED_NOT_ALLOCATED >> 30)) - length += (((laarr[i].extLength & UDF_EXTENT_LENGTH_MASK) + - inode->i_sb->s_blocksize - 1) >> inode->i_sb->s_blocksize_bits); + } else if ((laarr[i].extLength >> 30) == + (EXT_NOT_RECORDED_NOT_ALLOCATED >> 30)) + length += + (((laarr[i].extLength & UDF_EXTENT_LENGTH_MASK) + + inode->i_sb->s_blocksize - + 1) >> inode->i_sb->s_blocksize_bits); else break; } - if (length) - { + if (length) { int next = laarr[start].extLocation.logicalBlockNum + - (((laarr[start].extLength & UDF_EXTENT_LENGTH_MASK) + - inode->i_sb->s_blocksize - 1) >> inode->i_sb->s_blocksize_bits); + (((laarr[start].extLength & UDF_EXTENT_LENGTH_MASK) + + inode->i_sb->s_blocksize - + 1) >> inode->i_sb->s_blocksize_bits); int numalloc = udf_prealloc_blocks(inode->i_sb, inode, - laarr[start].extLocation.partitionReferenceNum, - next, (UDF_DEFAULT_PREALLOC_BLOCKS > length ? length : - UDF_DEFAULT_PREALLOC_BLOCKS) - currlength); - - if (numalloc) - { - if (start == (c+1)) + laarr[start].extLocation. + partitionReferenceNum, + next, + (UDF_DEFAULT_PREALLOC_BLOCKS + > + length ? length : + UDF_DEFAULT_PREALLOC_BLOCKS) + - currlength); + + if (numalloc) { + if (start == (c + 1)) laarr[start].extLength += - (numalloc << inode->i_sb->s_blocksize_bits); - else - { - memmove(&laarr[c+2], &laarr[c+1], - sizeof(long_ad) * (*endnum - (c+1))); - (*endnum) ++; - laarr[c+1].extLocation.logicalBlockNum = next; - laarr[c+1].extLocation.partitionReferenceNum = - laarr[c].extLocation.partitionReferenceNum; - laarr[c+1].extLength = EXT_NOT_RECORDED_ALLOCATED | - (numalloc << inode->i_sb->s_blocksize_bits); - start = c+1; + (numalloc << inode->i_sb->s_blocksize_bits); + else { + memmove(&laarr[c + 2], &laarr[c + 1], + sizeof(long_ad) * (*endnum - (c + 1))); + (*endnum)++; + laarr[c + 1].extLocation.logicalBlockNum = next; + laarr[c + 1].extLocation.partitionReferenceNum = + laarr[c].extLocation.partitionReferenceNum; + laarr[c + 1].extLength = + EXT_NOT_RECORDED_ALLOCATED | (numalloc << + inode->i_sb-> + s_blocksize_bits); + start = c + 1; } - for (i=start+1; numalloc && i<*endnum; i++) - { - int elen = ((laarr[i].extLength & UDF_EXTENT_LENGTH_MASK) + - inode->i_sb->s_blocksize - 1) >> inode->i_sb->s_blocksize_bits; + for (i = start + 1; numalloc && i < *endnum; i++) { + int elen = + ((laarr[i]. + extLength & UDF_EXTENT_LENGTH_MASK) + + inode->i_sb->s_blocksize - + 1) >> inode->i_sb->s_blocksize_bits; - if (elen > numalloc) - { + if (elen > numalloc) { laarr[i].extLength -= - (numalloc << inode->i_sb->s_blocksize_bits); + (numalloc << inode->i_sb-> + s_blocksize_bits); numalloc = 0; - } - else - { + } else { numalloc -= elen; - if (*endnum > (i+1)) - memmove(&laarr[i], &laarr[i+1], - sizeof(long_ad) * (*endnum - (i+1))); - i --; - (*endnum) --; + if (*endnum > (i + 1)) + memmove(&laarr[i], + &laarr[i + 1], + sizeof(long_ad) * + (*endnum - (i + 1))); + i--; + (*endnum)--; } } - UDF_I_LENEXTENTS(inode) += numalloc << inode->i_sb->s_blocksize_bits; + UDF_I_LENEXTENTS(inode) += + numalloc << inode->i_sb->s_blocksize_bits; } } } static void udf_merge_extents(struct inode *inode, - kernel_long_ad laarr[EXTENT_MERGE_SIZE], int *endnum) + kernel_long_ad laarr[EXTENT_MERGE_SIZE], + int *endnum) { int i; - for (i=0; i<(*endnum-1); i++) - { - if ((laarr[i].extLength >> 30) == (laarr[i+1].extLength >> 30)) - { - if (((laarr[i].extLength >> 30) == (EXT_NOT_RECORDED_NOT_ALLOCATED >> 30)) || - ((laarr[i+1].extLocation.logicalBlockNum - laarr[i].extLocation.logicalBlockNum) == - (((laarr[i].extLength & UDF_EXTENT_LENGTH_MASK) + - inode->i_sb->s_blocksize - 1) >> inode->i_sb->s_blocksize_bits))) - { - if (((laarr[i].extLength & UDF_EXTENT_LENGTH_MASK) + - (laarr[i+1].extLength & UDF_EXTENT_LENGTH_MASK) + - inode->i_sb->s_blocksize - 1) & ~UDF_EXTENT_LENGTH_MASK) - { - laarr[i+1].extLength = (laarr[i+1].extLength - - (laarr[i].extLength & UDF_EXTENT_LENGTH_MASK) + - UDF_EXTENT_LENGTH_MASK) & ~(inode->i_sb->s_blocksize-1); - laarr[i].extLength = (laarr[i].extLength & UDF_EXTENT_FLAG_MASK) + - (UDF_EXTENT_LENGTH_MASK + 1) - inode->i_sb->s_blocksize; - laarr[i+1].extLocation.logicalBlockNum = - laarr[i].extLocation.logicalBlockNum + - ((laarr[i].extLength & UDF_EXTENT_LENGTH_MASK) >> - inode->i_sb->s_blocksize_bits); - } - else - { - laarr[i].extLength = laarr[i+1].extLength + - (((laarr[i].extLength & UDF_EXTENT_LENGTH_MASK) + - inode->i_sb->s_blocksize - 1) & ~(inode->i_sb->s_blocksize-1)); - if (*endnum > (i+2)) - memmove(&laarr[i+1], &laarr[i+2], - sizeof(long_ad) * (*endnum - (i+2))); - i --; - (*endnum) --; + for (i = 0; i < (*endnum - 1); i++) { + if ((laarr[i].extLength >> 30) == + (laarr[i + 1].extLength >> 30)) { + if (((laarr[i].extLength >> 30) == + (EXT_NOT_RECORDED_NOT_ALLOCATED >> 30)) + || + ((laarr[i + 1].extLocation.logicalBlockNum - + laarr[i].extLocation.logicalBlockNum) == + (((laarr[i].extLength & UDF_EXTENT_LENGTH_MASK) + + inode->i_sb->s_blocksize - + 1) >> inode->i_sb->s_blocksize_bits))) { + if (((laarr[i]. + extLength & UDF_EXTENT_LENGTH_MASK) + + (laarr[i + 1]. + extLength & UDF_EXTENT_LENGTH_MASK) + + inode->i_sb->s_blocksize - + 1) & ~UDF_EXTENT_LENGTH_MASK) { + laarr[i + 1].extLength = + (laarr[i + 1].extLength - + (laarr[i]. + extLength & + UDF_EXTENT_LENGTH_MASK) + + UDF_EXTENT_LENGTH_MASK) & ~(inode-> + i_sb-> + s_blocksize + - 1); + laarr[i].extLength = + (laarr[i]. + extLength & UDF_EXTENT_FLAG_MASK) + + (UDF_EXTENT_LENGTH_MASK + 1) - + inode->i_sb->s_blocksize; + laarr[i + + 1].extLocation.logicalBlockNum = + laarr[i].extLocation. + logicalBlockNum + + ((laarr[i]. + extLength & + UDF_EXTENT_LENGTH_MASK) >> inode-> + i_sb->s_blocksize_bits); + } else { + laarr[i].extLength = + laarr[i + 1].extLength + + (((laarr[i]. + extLength & + UDF_EXTENT_LENGTH_MASK) + + inode->i_sb->s_blocksize - + 1) & ~(inode->i_sb->s_blocksize - + 1)); + if (*endnum > (i + 2)) + memmove(&laarr[i + 1], + &laarr[i + 2], + sizeof(long_ad) * + (*endnum - (i + 2))); + i--; + (*endnum)--; } } - } - else if (((laarr[i].extLength >> 30) == (EXT_NOT_RECORDED_ALLOCATED >> 30)) && - ((laarr[i+1].extLength >> 30) == (EXT_NOT_RECORDED_NOT_ALLOCATED >> 30))) - { - udf_free_blocks(inode->i_sb, inode, laarr[i].extLocation, 0, - ((laarr[i].extLength & UDF_EXTENT_LENGTH_MASK) + - inode->i_sb->s_blocksize - 1) >> inode->i_sb->s_blocksize_bits); + } else + if (((laarr[i].extLength >> 30) == + (EXT_NOT_RECORDED_ALLOCATED >> 30)) + && ((laarr[i + 1].extLength >> 30) == + (EXT_NOT_RECORDED_NOT_ALLOCATED >> 30))) { + udf_free_blocks(inode->i_sb, inode, + laarr[i].extLocation, 0, + ((laarr[i]. + extLength & UDF_EXTENT_LENGTH_MASK) + + inode->i_sb->s_blocksize - + 1) >> inode->i_sb->s_blocksize_bits); laarr[i].extLocation.logicalBlockNum = 0; laarr[i].extLocation.partitionReferenceNum = 0; if (((laarr[i].extLength & UDF_EXTENT_LENGTH_MASK) + - (laarr[i+1].extLength & UDF_EXTENT_LENGTH_MASK) + - inode->i_sb->s_blocksize - 1) & ~UDF_EXTENT_LENGTH_MASK) - { - laarr[i+1].extLength = (laarr[i+1].extLength - - (laarr[i].extLength & UDF_EXTENT_LENGTH_MASK) + - UDF_EXTENT_LENGTH_MASK) & ~(inode->i_sb->s_blocksize-1); - laarr[i].extLength = (laarr[i].extLength & UDF_EXTENT_FLAG_MASK) + - (UDF_EXTENT_LENGTH_MASK + 1) - inode->i_sb->s_blocksize; + (laarr[i + 1].extLength & UDF_EXTENT_LENGTH_MASK) + + inode->i_sb->s_blocksize - + 1) & ~UDF_EXTENT_LENGTH_MASK) { + laarr[i + 1].extLength = + (laarr[i + 1].extLength - + (laarr[i]. + extLength & UDF_EXTENT_LENGTH_MASK) + + UDF_EXTENT_LENGTH_MASK) & ~(inode->i_sb-> + s_blocksize - + 1); + laarr[i].extLength = + (laarr[i]. + extLength & UDF_EXTENT_FLAG_MASK) + + (UDF_EXTENT_LENGTH_MASK + 1) - + inode->i_sb->s_blocksize; + } else { + laarr[i].extLength = laarr[i + 1].extLength + + (((laarr[i]. + extLength & UDF_EXTENT_LENGTH_MASK) + + inode->i_sb->s_blocksize - + 1) & ~(inode->i_sb->s_blocksize - 1)); + if (*endnum > (i + 2)) + memmove(&laarr[i + 1], &laarr[i + 2], + sizeof(long_ad) * (*endnum - + (i + 2))); + i--; + (*endnum)--; } - else - { - laarr[i].extLength = laarr[i+1].extLength + - (((laarr[i].extLength & UDF_EXTENT_LENGTH_MASK) + - inode->i_sb->s_blocksize - 1) & ~(inode->i_sb->s_blocksize-1)); - if (*endnum > (i+2)) - memmove(&laarr[i+1], &laarr[i+2], - sizeof(long_ad) * (*endnum - (i+2))); - i --; - (*endnum) --; - } - } - else if ((laarr[i].extLength >> 30) == (EXT_NOT_RECORDED_ALLOCATED >> 30)) - { - udf_free_blocks(inode->i_sb, inode, laarr[i].extLocation, 0, - ((laarr[i].extLength & UDF_EXTENT_LENGTH_MASK) + - inode->i_sb->s_blocksize - 1) >> inode->i_sb->s_blocksize_bits); + } else if ((laarr[i].extLength >> 30) == + (EXT_NOT_RECORDED_ALLOCATED >> 30)) { + udf_free_blocks(inode->i_sb, inode, + laarr[i].extLocation, 0, + ((laarr[i]. + extLength & UDF_EXTENT_LENGTH_MASK) + + inode->i_sb->s_blocksize - + 1) >> inode->i_sb->s_blocksize_bits); laarr[i].extLocation.logicalBlockNum = 0; laarr[i].extLocation.partitionReferenceNum = 0; - laarr[i].extLength = (laarr[i].extLength & UDF_EXTENT_LENGTH_MASK) | - EXT_NOT_RECORDED_NOT_ALLOCATED; + laarr[i].extLength = + (laarr[i]. + extLength & UDF_EXTENT_LENGTH_MASK) | + EXT_NOT_RECORDED_NOT_ALLOCATED; } } } static void udf_update_extents(struct inode *inode, - kernel_long_ad laarr[EXTENT_MERGE_SIZE], int startnum, int endnum, - struct extent_position *epos) + kernel_long_ad laarr[EXTENT_MERGE_SIZE], + int startnum, int endnum, + struct extent_position *epos) { int start = 0, i; kernel_lb_addr tmploc; uint32_t tmplen; - if (startnum > endnum) - { - for (i=0; i<(startnum-endnum); i++) + if (startnum > endnum) { + for (i = 0; i < (startnum - endnum); i++) udf_delete_aext(inode, *epos, laarr[i].extLocation, - laarr[i].extLength); - } - else if (startnum < endnum) - { - for (i=0; i<(endnum-startnum); i++) - { + laarr[i].extLength); + } else if (startnum < endnum) { + for (i = 0; i < (endnum - startnum); i++) { udf_insert_aext(inode, *epos, laarr[i].extLocation, - laarr[i].extLength); + laarr[i].extLength); udf_next_aext(inode, epos, &laarr[i].extLocation, - &laarr[i].extLength, 1); - start ++; + &laarr[i].extLength, 1); + start++; } } - for (i=start; i<endnum; i++) - { + for (i = start; i < endnum; i++) { udf_next_aext(inode, epos, &tmploc, &tmplen, 0); udf_write_aext(inode, epos, laarr[i].extLocation, - laarr[i].extLength, 1); + laarr[i].extLength, 1); } } -struct buffer_head * udf_bread(struct inode * inode, int block, - int create, int * err) +struct buffer_head *udf_bread(struct inode *inode, int block, + int create, int *err) { - struct buffer_head * bh = NULL; + struct buffer_head *bh = NULL; bh = udf_getblk(inode, block, create, err); if (!bh) @@ -987,56 +1034,51 @@ struct buffer_head * udf_bread(struct inode * inode, int block, return NULL; } -void udf_truncate(struct inode * inode) +void udf_truncate(struct inode *inode) { int offset; int err; if (!(S_ISREG(inode->i_mode) || S_ISDIR(inode->i_mode) || - S_ISLNK(inode->i_mode))) + S_ISLNK(inode->i_mode))) return; if (IS_APPEND(inode) || IS_IMMUTABLE(inode)) return; lock_kernel(); - if (UDF_I_ALLOCTYPE(inode) == ICBTAG_FLAG_AD_IN_ICB) - { - if (inode->i_sb->s_blocksize < (udf_file_entry_alloc_offset(inode) + - inode->i_size)) - { + if (UDF_I_ALLOCTYPE(inode) == ICBTAG_FLAG_AD_IN_ICB) { + if (inode->i_sb->s_blocksize < + (udf_file_entry_alloc_offset(inode) + inode->i_size)) { udf_expand_file_adinicb(inode, inode->i_size, &err); - if (UDF_I_ALLOCTYPE(inode) == ICBTAG_FLAG_AD_IN_ICB) - { + if (UDF_I_ALLOCTYPE(inode) == ICBTAG_FLAG_AD_IN_ICB) { inode->i_size = UDF_I_LENALLOC(inode); unlock_kernel(); return; - } - else + } else udf_truncate_extents(inode); - } - else - { + } else { offset = inode->i_size & (inode->i_sb->s_blocksize - 1); - memset(UDF_I_DATA(inode) + UDF_I_LENEATTR(inode) + offset, 0x00, inode->i_sb->s_blocksize - offset - udf_file_entry_alloc_offset(inode)); + memset(UDF_I_DATA(inode) + UDF_I_LENEATTR(inode) + + offset, 0x00, + inode->i_sb->s_blocksize - offset - + udf_file_entry_alloc_offset(inode)); UDF_I_LENALLOC(inode) = inode->i_size; } - } - else - { - block_truncate_page(inode->i_mapping, inode->i_size, udf_get_block); + } else { + block_truncate_page(inode->i_mapping, inode->i_size, + udf_get_block); udf_truncate_extents(inode); } inode->i_mtime = inode->i_ctime = current_fs_time(inode->i_sb); if (IS_SYNC(inode)) - udf_sync_inode (inode); + udf_sync_inode(inode); else mark_inode_dirty(inode); unlock_kernel(); } -static void -__udf_read_inode(struct inode *inode) +static void __udf_read_inode(struct inode *inode) { struct buffer_head *bh = NULL; struct fileEntry *fe; @@ -1056,19 +1098,18 @@ __udf_read_inode(struct inode *inode) */ bh = udf_read_ptagged(inode->i_sb, UDF_I_LOCATION(inode), 0, &ident); - if (!bh) - { + if (!bh) { printk(KERN_ERR "udf: udf_read_inode(ino %ld) failed !bh\n", - inode->i_ino); + inode->i_ino); make_bad_inode(inode); return; } if (ident != TAG_IDENT_FE && ident != TAG_IDENT_EFE && - ident != TAG_IDENT_USE) - { - printk(KERN_ERR "udf: udf_read_inode(ino %ld) failed ident=%d\n", - inode->i_ino, ident); + ident != TAG_IDENT_USE) { + printk(KERN_ERR + "udf: udf_read_inode(ino %ld) failed ident=%d\n", + inode->i_ino, ident); brelse(bh); make_bad_inode(inode); return; @@ -1076,51 +1117,46 @@ __udf_read_inode(struct inode *inode) fe = (struct fileEntry *)bh->b_data; - if (le16_to_cpu(fe->icbTag.strategyType) == 4096) - { + if (le16_to_cpu(fe->icbTag.strategyType) == 4096) { struct buffer_head *ibh = NULL, *nbh = NULL; struct indirectEntry *ie; - ibh = udf_read_ptagged(inode->i_sb, UDF_I_LOCATION(inode), 1, &ident); - if (ident == TAG_IDENT_IE) - { - if (ibh) - { + ibh = + udf_read_ptagged(inode->i_sb, UDF_I_LOCATION(inode), 1, + &ident); + if (ident == TAG_IDENT_IE) { + if (ibh) { kernel_lb_addr loc; ie = (struct indirectEntry *)ibh->b_data; loc = lelb_to_cpu(ie->indirectICB.extLocation); if (ie->indirectICB.extLength && - (nbh = udf_read_ptagged(inode->i_sb, loc, 0, &ident))) - { - if (ident == TAG_IDENT_FE || - ident == TAG_IDENT_EFE) - { - memcpy(&UDF_I_LOCATION(inode), &loc, sizeof(kernel_lb_addr)); + (nbh = + udf_read_ptagged(inode->i_sb, loc, 0, + &ident))) { + if (ident == TAG_IDENT_FE + || ident == TAG_IDENT_EFE) { + memcpy(&UDF_I_LOCATION(inode), + &loc, + sizeof(kernel_lb_addr)); brelse(bh); brelse(ibh); brelse(nbh); __udf_read_inode(inode); return; - } - else - { + } else { brelse(nbh); brelse(ibh); } - } - else + } else brelse(ibh); } - } - else + } else brelse(ibh); - } - else if (le16_to_cpu(fe->icbTag.strategyType) != 4) - { + } else if (le16_to_cpu(fe->icbTag.strategyType) != 4) { printk(KERN_ERR "udf: unsupported strategy type: %d\n", - le16_to_cpu(fe->icbTag.strategyType)); + le16_to_cpu(fe->icbTag.strategyType)); brelse(bh); make_bad_inode(inode); return; @@ -1143,62 +1179,70 @@ static void udf_fill_inode(struct inode *inode, struct buffer_head *bh) if (le16_to_cpu(fe->icbTag.strategyType) == 4) UDF_I_STRAT4096(inode) = 0; - else /* if (le16_to_cpu(fe->icbTag.strategyType) == 4096) */ + else /* if (le16_to_cpu(fe->icbTag.strategyType) == 4096) */ UDF_I_STRAT4096(inode) = 1; - UDF_I_ALLOCTYPE(inode) = le16_to_cpu(fe->icbTag.flags) & ICBTAG_FLAG_AD_MASK; + UDF_I_ALLOCTYPE(inode) = + le16_to_cpu(fe->icbTag.flags) & ICBTAG_FLAG_AD_MASK; UDF_I_UNIQUE(inode) = 0; UDF_I_LENEATTR(inode) = 0; UDF_I_LENEXTENTS(inode) = 0; UDF_I_LENALLOC(inode) = 0; UDF_I_NEXT_ALLOC_BLOCK(inode) = 0; UDF_I_NEXT_ALLOC_GOAL(inode) = 0; - if (le16_to_cpu(fe->descTag.tagIdent) == TAG_IDENT_EFE) - { + if (le16_to_cpu(fe->descTag.tagIdent) == TAG_IDENT_EFE) { UDF_I_EFE(inode) = 1; UDF_I_USE(inode) = 0; - if (udf_alloc_i_data(inode, inode->i_sb->s_blocksize - sizeof(struct extendedFileEntry))) - { + if (udf_alloc_i_data + (inode, + inode->i_sb->s_blocksize - + sizeof(struct extendedFileEntry))) { make_bad_inode(inode); return; } - memcpy(UDF_I_DATA(inode), bh->b_data + sizeof(struct extendedFileEntry), inode->i_sb->s_blocksize - sizeof(struct extendedFileEntry)); - } - else if (le16_to_cpu(fe->descTag.tagIdent) == TAG_IDENT_FE) - { + memcpy(UDF_I_DATA(inode), + bh->b_data + sizeof(struct extendedFileEntry), + inode->i_sb->s_blocksize - + sizeof(struct extendedFileEntry)); + } else if (le16_to_cpu(fe->descTag.tagIdent) == TAG_IDENT_FE) { UDF_I_EFE(inode) = 0; UDF_I_USE(inode) = 0; - if (udf_alloc_i_data(inode, inode->i_sb->s_blocksize - sizeof(struct fileEntry))) - { + if (udf_alloc_i_data + (inode, + inode->i_sb->s_blocksize - sizeof(struct fileEntry))) { make_bad_inode(inode); return; } - memcpy(UDF_I_DATA(inode), bh->b_data + sizeof(struct fileEntry), inode->i_sb->s_blocksize - sizeof(struct fileEntry)); - } - else if (le16_to_cpu(fe->descTag.tagIdent) == TAG_IDENT_USE) - { + memcpy(UDF_I_DATA(inode), bh->b_data + sizeof(struct fileEntry), + inode->i_sb->s_blocksize - sizeof(struct fileEntry)); + } else if (le16_to_cpu(fe->descTag.tagIdent) == TAG_IDENT_USE) { UDF_I_EFE(inode) = 0; UDF_I_USE(inode) = 1; UDF_I_LENALLOC(inode) = - le32_to_cpu( - ((struct unallocSpaceEntry *)bh->b_data)->lengthAllocDescs); - if (udf_alloc_i_data(inode, inode->i_sb->s_blocksize - sizeof(struct unallocSpaceEntry))) - { + le32_to_cpu(((struct unallocSpaceEntry *)bh->b_data)-> + lengthAllocDescs); + if (udf_alloc_i_data + (inode, + inode->i_sb->s_blocksize - + sizeof(struct unallocSpaceEntry))) { make_bad_inode(inode); return; } - memcpy(UDF_I_DATA(inode), bh->b_data + sizeof(struct unallocSpaceEntry), inode->i_sb->s_blocksize - sizeof(struct unallocSpaceEntry)); + memcpy(UDF_I_DATA(inode), + bh->b_data + sizeof(struct unallocSpaceEntry), + inode->i_sb->s_blocksize - + sizeof(struct unallocSpaceEntry)); return; } inode->i_uid = le32_to_cpu(fe->uid); if (inode->i_uid == -1 || UDF_QUERY_FLAG(inode->i_sb, - UDF_FLAG_UID_IGNORE)) + UDF_FLAG_UID_IGNORE)) inode->i_uid = UDF_SB(inode->i_sb)->s_uid; inode->i_gid = le32_to_cpu(fe->gid); if (inode->i_gid == -1 || UDF_QUERY_FLAG(inode->i_sb, - UDF_FLAG_GID_IGNORE)) + UDF_FLAG_GID_IGNORE)) inode->i_gid = UDF_SB(inode->i_sb)->s_gid; inode->i_nlink = le16_to_cpu(fe->fileLinkCount); @@ -1211,41 +1255,31 @@ static void udf_fill_inode(struct inode *inode, struct buffer_head *bh) inode->i_mode = udf_convert_permissions(fe); inode->i_mode &= ~UDF_SB(inode->i_sb)->s_umask; - if (UDF_I_EFE(inode) == 0) - { + if (UDF_I_EFE(inode) == 0) { inode->i_blocks = le64_to_cpu(fe->logicalBlocksRecorded) << - (inode->i_sb->s_blocksize_bits - 9); + (inode->i_sb->s_blocksize_bits - 9); - if ( udf_stamp_to_time(&convtime, &convtime_usec, - lets_to_cpu(fe->accessTime)) ) - { + if (udf_stamp_to_time(&convtime, &convtime_usec, + lets_to_cpu(fe->accessTime))) { inode->i_atime.tv_sec = convtime; inode->i_atime.tv_nsec = convtime_usec * 1000; - } - else - { + } else { inode->i_atime = UDF_SB_RECORDTIME(inode->i_sb); } - if ( udf_stamp_to_time(&convtime, &convtime_usec, - lets_to_cpu(fe->modificationTime)) ) - { + if (udf_stamp_to_time(&convtime, &convtime_usec, + lets_to_cpu(fe->modificationTime))) { inode->i_mtime.tv_sec = convtime; inode->i_mtime.tv_nsec = convtime_usec * 1000; - } - else - { + } else { inode->i_mtime = UDF_SB_RECORDTIME(inode->i_sb); } - if ( udf_stamp_to_time(&convtime, &convtime_usec, - lets_to_cpu(fe->attrTime)) ) - { + if (udf_stamp_to_time(&convtime, &convtime_usec, + lets_to_cpu(fe->attrTime))) { inode->i_ctime.tv_sec = convtime; inode->i_ctime.tv_nsec = convtime_usec * 1000; - } - else - { + } else { inode->i_ctime = UDF_SB_RECORDTIME(inode->i_sb); } @@ -1253,65 +1287,51 @@ static void udf_fill_inode(struct inode *inode, struct buffer_head *bh) UDF_I_LENEATTR(inode) = le32_to_cpu(fe->lengthExtendedAttr); UDF_I_LENALLOC(inode) = le32_to_cpu(fe->lengthAllocDescs); offset = sizeof(struct fileEntry) + UDF_I_LENEATTR(inode); - } - else - { + } else { inode->i_blocks = le64_to_cpu(efe->logicalBlocksRecorded) << - (inode->i_sb->s_blocksize_bits - 9); + (inode->i_sb->s_blocksize_bits - 9); - if ( udf_stamp_to_time(&convtime, &convtime_usec, - lets_to_cpu(efe->accessTime)) ) - { + if (udf_stamp_to_time(&convtime, &convtime_usec, + lets_to_cpu(efe->accessTime))) { inode->i_atime.tv_sec = convtime; inode->i_atime.tv_nsec = convtime_usec * 1000; - } - else - { + } else { inode->i_atime = UDF_SB_RECORDTIME(inode->i_sb); } - if ( udf_stamp_to_time(&convtime, &convtime_usec, - lets_to_cpu(efe->modificationTime)) ) - { + if (udf_stamp_to_time(&convtime, &convtime_usec, + lets_to_cpu(efe->modificationTime))) { inode->i_mtime.tv_sec = convtime; inode->i_mtime.tv_nsec = convtime_usec * 1000; - } - else - { + } else { inode->i_mtime = UDF_SB_RECORDTIME(inode->i_sb); } - if ( udf_stamp_to_time(&convtime, &convtime_usec, - lets_to_cpu(efe->createTime)) ) - { + if (udf_stamp_to_time(&convtime, &convtime_usec, + lets_to_cpu(efe->createTime))) { UDF_I_CRTIME(inode).tv_sec = convtime; UDF_I_CRTIME(inode).tv_nsec = convtime_usec * 1000; - } - else - { + } else { UDF_I_CRTIME(inode) = UDF_SB_RECORDTIME(inode->i_sb); } - if ( udf_stamp_to_time(&convtime, &convtime_usec, - lets_to_cpu(efe->attrTime)) ) - { + if (udf_stamp_to_time(&convtime, &convtime_usec, + lets_to_cpu(efe->attrTime))) { inode->i_ctime.tv_sec = convtime; inode->i_ctime.tv_nsec = convtime_usec * 1000; - } - else - { + } else { inode->i_ctime = UDF_SB_RECORDTIME(inode->i_sb); } UDF_I_UNIQUE(inode) = le64_to_cpu(efe->uniqueID); UDF_I_LENEATTR(inode) = le32_to_cpu(efe->lengthExtendedAttr); UDF_I_LENALLOC(inode) = le32_to_cpu(efe->lengthAllocDescs); - offset = sizeof(struct extendedFileEntry) + UDF_I_LENEATTR(inode); + offset = + sizeof(struct extendedFileEntry) + UDF_I_LENEATTR(inode); } - switch (fe->icbTag.fileType) - { - case ICBTAG_FILE_TYPE_DIRECTORY: + switch (fe->icbTag.fileType) { + case ICBTAG_FILE_TYPE_DIRECTORY: { inode->i_op = &udf_dir_inode_operations; inode->i_fop = &udf_dir_operations; @@ -1319,9 +1339,9 @@ static void udf_fill_inode(struct inode *inode, struct buffer_head *bh) inc_nlink(inode); break; } - case ICBTAG_FILE_TYPE_REALTIME: - case ICBTAG_FILE_TYPE_REGULAR: - case ICBTAG_FILE_TYPE_UNDEF: + case ICBTAG_FILE_TYPE_REALTIME: + case ICBTAG_FILE_TYPE_REGULAR: + case ICBTAG_FILE_TYPE_UNDEF: { if (UDF_I_ALLOCTYPE(inode) == ICBTAG_FLAG_AD_IN_ICB) inode->i_data.a_ops = &udf_adinicb_aops; @@ -1332,56 +1352,54 @@ static void udf_fill_inode(struct inode *inode, struct buffer_head *bh) inode->i_mode |= S_IFREG; break; } - case ICBTAG_FILE_TYPE_BLOCK: + case ICBTAG_FILE_TYPE_BLOCK: { inode->i_mode |= S_IFBLK; break; } - case ICBTAG_FILE_TYPE_CHAR: + case ICBTAG_FILE_TYPE_CHAR: { inode->i_mode |= S_IFCHR; break; } - case ICBTAG_FILE_TYPE_FIFO: + case ICBTAG_FILE_TYPE_FIFO: { init_special_inode(inode, inode->i_mode | S_IFIFO, 0); break; } - case ICBTAG_FILE_TYPE_SOCKET: + case ICBTAG_FILE_TYPE_SOCKET: { init_special_inode(inode, inode->i_mode | S_IFSOCK, 0); break; } - case ICBTAG_FILE_TYPE_SYMLINK: + case ICBTAG_FILE_TYPE_SYMLINK: { inode->i_data.a_ops = &udf_symlink_aops; inode->i_op = &page_symlink_inode_operations; - inode->i_mode = S_IFLNK|S_IRWXUGO; + inode->i_mode = S_IFLNK | S_IRWXUGO; break; } - default: + default: { - printk(KERN_ERR "udf: udf_fill_inode(ino %ld) failed unknown file type=%d\n", - inode->i_ino, fe->icbTag.fileType); + printk(KERN_ERR + "udf: udf_fill_inode(ino %ld) failed unknown file type=%d\n", + inode->i_ino, fe->icbTag.fileType); make_bad_inode(inode); return; } } - if (S_ISCHR(inode->i_mode) || S_ISBLK(inode->i_mode)) - { - struct deviceSpec *dsea = - (struct deviceSpec *) - udf_get_extendedattr(inode, 12, 1); + if (S_ISCHR(inode->i_mode) || S_ISBLK(inode->i_mode)) { + struct deviceSpec *dsea = (struct deviceSpec *) + udf_get_extendedattr(inode, 12, 1); - if (dsea) - { - init_special_inode(inode, inode->i_mode, MKDEV( - le32_to_cpu(dsea->majorDeviceIdent), - le32_to_cpu(dsea->minorDeviceIdent))); + if (dsea) { + init_special_inode(inode, inode->i_mode, + MKDEV(le32_to_cpu + (dsea->majorDeviceIdent), + le32_to_cpu(dsea-> + minorDeviceIdent))); /* Developer ID ??? */ - } - else - { + } else { make_bad_inode(inode); } } @@ -1391,9 +1409,9 @@ static int udf_alloc_i_data(struct inode *inode, size_t size) { UDF_I_DATA(inode) = kmalloc(size, GFP_KERNEL); - if (!UDF_I_DATA(inode)) - { - printk(KERN_ERR "udf:udf_alloc_i_data (ino %ld) no free memory\n", + if (!UDF_I_DATA(inode)) { + printk(KERN_ERR + "udf:udf_alloc_i_data (ino %ld) no free memory\n", inode->i_ino); return -ENOMEM; } @@ -1401,8 +1419,7 @@ static int udf_alloc_i_data(struct inode *inode, size_t size) return 0; } -static mode_t -udf_convert_permissions(struct fileEntry *fe) +static mode_t udf_convert_permissions(struct fileEntry *fe) { mode_t mode; uint32_t permissions; @@ -1411,12 +1428,12 @@ udf_convert_permissions(struct fileEntry *fe) permissions = le32_to_cpu(fe->permissions); flags = le16_to_cpu(fe->icbTag.flags); - mode = (( permissions ) & S_IRWXO) | - (( permissions >> 2 ) & S_IRWXG) | - (( permissions >> 4 ) & S_IRWXU) | - (( flags & ICBTAG_FLAG_SETUID) ? S_ISUID : 0) | - (( flags & ICBTAG_FLAG_SETGID) ? S_ISGID : 0) | - (( flags & ICBTAG_FLAG_STICKY) ? S_ISVTX : 0); + mode = ((permissions) & S_IRWXO) | + ((permissions >> 2) & S_IRWXG) | + ((permissions >> 4) & S_IRWXU) | + ((flags & ICBTAG_FLAG_SETUID) ? S_ISUID : 0) | + ((flags & ICBTAG_FLAG_SETGID) ? S_ISGID : 0) | + ((flags & ICBTAG_FLAG_STICKY) ? S_ISVTX : 0); return mode; } @@ -1436,7 +1453,7 @@ udf_convert_permissions(struct fileEntry *fe) * Written, tested, and released. */ -int udf_write_inode(struct inode * inode, int sync) +int udf_write_inode(struct inode *inode, int sync) { int ret; lock_kernel(); @@ -1445,13 +1462,12 @@ int udf_write_inode(struct inode * inode, int sync) return ret; } -int udf_sync_inode(struct inode * inode) +int udf_sync_inode(struct inode *inode) { return udf_update_inode(inode, 1); } -static int -udf_update_inode(struct inode *inode, int do_sync) +static int udf_update_inode(struct inode *inode, int do_sync) { struct buffer_head *bh = NULL; struct fileEntry *fe; @@ -1464,10 +1480,10 @@ udf_update_inode(struct inode *inode, int do_sync) int err = 0; bh = udf_tread(inode->i_sb, - udf_get_lb_pblock(inode->i_sb, UDF_I_LOCATION(inode), 0)); + udf_get_lb_pblock(inode->i_sb, UDF_I_LOCATION(inode), + 0)); - if (!bh) - { + if (!bh) { udf_debug("bread failure\n"); return -EIO; } @@ -1477,23 +1493,29 @@ udf_update_inode(struct inode *inode, int do_sync) fe = (struct fileEntry *)bh->b_data; efe = (struct extendedFileEntry *)bh->b_data; - if (le16_to_cpu(fe->descTag.tagIdent) == TAG_IDENT_USE) - { + if (le16_to_cpu(fe->descTag.tagIdent) == TAG_IDENT_USE) { struct unallocSpaceEntry *use = - (struct unallocSpaceEntry *)bh->b_data; + (struct unallocSpaceEntry *)bh->b_data; use->lengthAllocDescs = cpu_to_le32(UDF_I_LENALLOC(inode)); - memcpy(bh->b_data + sizeof(struct unallocSpaceEntry), UDF_I_DATA(inode), inode->i_sb->s_blocksize - sizeof(struct unallocSpaceEntry)); - crclen = sizeof(struct unallocSpaceEntry) + UDF_I_LENALLOC(inode) - - sizeof(tag); - use->descTag.tagLocation = cpu_to_le32(UDF_I_LOCATION(inode).logicalBlockNum); + memcpy(bh->b_data + sizeof(struct unallocSpaceEntry), + UDF_I_DATA(inode), + inode->i_sb->s_blocksize - + sizeof(struct unallocSpaceEntry)); + crclen = + sizeof(struct unallocSpaceEntry) + UDF_I_LENALLOC(inode) - + sizeof(tag); + use->descTag.tagLocation = + cpu_to_le32(UDF_I_LOCATION(inode).logicalBlockNum); use->descTag.descCRCLength = cpu_to_le16(crclen); - use->descTag.descCRC = cpu_to_le16(udf_crc((char *)use + sizeof(tag), crclen, 0)); + use->descTag.descCRC = + cpu_to_le16(udf_crc((char *)use + sizeof(tag), crclen, 0)); use->descTag.tagChecksum = 0; - for (i=0; i<16; i++) + for (i = 0; i < 16; i++) if (i != 4) - use->descTag.tagChecksum += ((uint8_t *)&(use->descTag))[i]; + use->descTag.tagChecksum += + ((uint8_t *) & (use->descTag))[i]; mark_buffer_dirty(bh); brelse(bh); @@ -1502,20 +1524,21 @@ udf_update_inode(struct inode *inode, int do_sync) if (UDF_QUERY_FLAG(inode->i_sb, UDF_FLAG_UID_FORGET)) fe->uid = cpu_to_le32(-1); - else fe->uid = cpu_to_le32(inode->i_uid); + else + fe->uid = cpu_to_le32(inode->i_uid); if (UDF_QUERY_FLAG(inode->i_sb, UDF_FLAG_GID_FORGET)) fe->gid = cpu_to_le32(-1); - else fe->gid = cpu_to_le32(inode->i_gid); + else + fe->gid = cpu_to_le32(inode->i_gid); - udfperms = ((inode->i_mode & S_IRWXO) ) | - ((inode->i_mode & S_IRWXG) << 2) | - ((inode->i_mode & S_IRWXU) << 4); + udfperms = ((inode->i_mode & S_IRWXO)) | + ((inode->i_mode & S_IRWXG) << 2) | ((inode->i_mode & S_IRWXU) << 4); - udfperms |= (le32_to_cpu(fe->permissions) & - (FE_PERM_O_DELETE | FE_PERM_O_CHATTR | - FE_PERM_G_DELETE | FE_PERM_G_CHATTR | - FE_PERM_U_DELETE | FE_PERM_U_CHATTR)); + udfperms |= (le32_to_cpu(fe->permissions) & + (FE_PERM_O_DELETE | FE_PERM_O_CHATTR | + FE_PERM_G_DELETE | FE_PERM_G_CHATTR | + FE_PERM_U_DELETE | FE_PERM_U_CHATTR)); fe->permissions = cpu_to_le32(udfperms); if (S_ISDIR(inode->i_mode)) @@ -1525,26 +1548,24 @@ udf_update_inode(struct inode *inode, int do_sync) fe->informationLength = cpu_to_le64(inode->i_size); - if (S_ISCHR(inode->i_mode) || S_ISBLK(inode->i_mode)) - { + if (S_ISCHR(inode->i_mode) || S_ISBLK(inode->i_mode)) { regid *eid; - struct deviceSpec *dsea = - (struct deviceSpec *) - udf_get_extendedattr(inode, 12, 1); + struct deviceSpec *dsea = (struct deviceSpec *) + udf_get_extendedattr(inode, 12, 1); - if (!dsea) - { + if (!dsea) { dsea = (struct deviceSpec *) - udf_add_extendedattr(inode, - sizeof(struct deviceSpec) + - sizeof(regid), 12, 0x3); + udf_add_extendedattr(inode, + sizeof(struct deviceSpec) + + sizeof(regid), 12, 0x3); dsea->attrType = cpu_to_le32(12); dsea->attrSubtype = 1; - dsea->attrLength = cpu_to_le32(sizeof(struct deviceSpec) + - sizeof(regid)); + dsea->attrLength = + cpu_to_le32(sizeof(struct deviceSpec) + + sizeof(regid)); dsea->impUseLength = cpu_to_le32(sizeof(regid)); } - eid = (regid *)dsea->impUse; + eid = (regid *) dsea->impUse; memset(eid, 0, sizeof(regid)); strcpy(eid->ident, UDF_ID_DEVELOPER); eid->identSuffix[0] = UDF_OS_CLASS_UNIX; @@ -1553,12 +1574,13 @@ udf_update_inode(struct inode *inode, int do_sync) dsea->minorDeviceIdent = cpu_to_le32(iminor(inode)); } - if (UDF_I_EFE(inode) == 0) - { - memcpy(bh->b_data + sizeof(struct fileEntry), UDF_I_DATA(inode), inode->i_sb->s_blocksize - sizeof(struct fileEntry)); - fe->logicalBlocksRecorded = cpu_to_le64( - (inode->i_blocks + (1 << (inode->i_sb->s_blocksize_bits - 9)) - 1) >> - (inode->i_sb->s_blocksize_bits - 9)); + if (UDF_I_EFE(inode) == 0) { + memcpy(bh->b_data + sizeof(struct fileEntry), UDF_I_DATA(inode), + inode->i_sb->s_blocksize - sizeof(struct fileEntry)); + fe->logicalBlocksRecorded = + cpu_to_le64((inode->i_blocks + + (1 << (inode->i_sb->s_blocksize_bits - 9)) - + 1) >> (inode->i_sb->s_blocksize_bits - 9)); if (udf_time_to_stamp(&cpu_time, inode->i_atime)) fe->accessTime = cpu_to_lets(cpu_time); @@ -1575,31 +1597,34 @@ udf_update_inode(struct inode *inode, int do_sync) fe->lengthAllocDescs = cpu_to_le32(UDF_I_LENALLOC(inode)); fe->descTag.tagIdent = cpu_to_le16(TAG_IDENT_FE); crclen = sizeof(struct fileEntry); - } - else - { - memcpy(bh->b_data + sizeof(struct extendedFileEntry), UDF_I_DATA(inode), inode->i_sb->s_blocksize - sizeof(struct extendedFileEntry)); + } else { + memcpy(bh->b_data + sizeof(struct extendedFileEntry), + UDF_I_DATA(inode), + inode->i_sb->s_blocksize - + sizeof(struct extendedFileEntry)); efe->objectSize = cpu_to_le64(inode->i_size); - efe->logicalBlocksRecorded = cpu_to_le64( - (inode->i_blocks + (1 << (inode->i_sb->s_blocksize_bits - 9)) - 1) >> - (inode->i_sb->s_blocksize_bits - 9)); + efe->logicalBlocksRecorded = cpu_to_le64((inode->i_blocks + + (1 << + (inode->i_sb-> + s_blocksize_bits - + 9)) - + 1) >> (inode->i_sb-> + s_blocksize_bits + - 9)); if (UDF_I_CRTIME(inode).tv_sec > inode->i_atime.tv_sec || - (UDF_I_CRTIME(inode).tv_sec == inode->i_atime.tv_sec && - UDF_I_CRTIME(inode).tv_nsec > inode->i_atime.tv_nsec)) - { + (UDF_I_CRTIME(inode).tv_sec == inode->i_atime.tv_sec && + UDF_I_CRTIME(inode).tv_nsec > inode->i_atime.tv_nsec)) { UDF_I_CRTIME(inode) = inode->i_atime; } if (UDF_I_CRTIME(inode).tv_sec > inode->i_mtime.tv_sec || - (UDF_I_CRTIME(inode).tv_sec == inode->i_mtime.tv_sec && - UDF_I_CRTIME(inode).tv_nsec > inode->i_mtime.tv_nsec)) - { + (UDF_I_CRTIME(inode).tv_sec == inode->i_mtime.tv_sec && + UDF_I_CRTIME(inode).tv_nsec > inode->i_mtime.tv_nsec)) { UDF_I_CRTIME(inode) = inode->i_mtime; } if (UDF_I_CRTIME(inode).tv_sec > inode->i_ctime.tv_sec || - (UDF_I_CRTIME(inode).tv_sec == inode->i_ctime.tv_sec && - UDF_I_CRTIME(inode).tv_nsec > inode->i_ctime.tv_nsec)) - { + (UDF_I_CRTIME(inode).tv_sec == inode->i_ctime.tv_sec && + UDF_I_CRTIME(inode).tv_nsec > inode->i_ctime.tv_nsec)) { UDF_I_CRTIME(inode) = inode->i_ctime; } @@ -1622,14 +1647,11 @@ udf_update_inode(struct inode *inode, int do_sync) efe->descTag.tagIdent = cpu_to_le16(TAG_IDENT_EFE); crclen = sizeof(struct extendedFileEntry); } - if (UDF_I_STRAT4096(inode)) - { + if (UDF_I_STRAT4096(inode)) { fe->icbTag.strategyType = cpu_to_le16(4096); fe->icbTag.strategyParameter = cpu_to_le16(1); fe->icbTag.numEntries = cpu_to_le16(2); - } - else - { + } else { fe->icbTag.strategyType = cpu_to_le16(4); fe->icbTag.numEntries = cpu_to_le16(1); } @@ -1649,13 +1671,13 @@ udf_update_inode(struct inode *inode, int do_sync) else if (S_ISSOCK(inode->i_mode)) fe->icbTag.fileType = ICBTAG_FILE_TYPE_SOCKET; - icbflags = UDF_I_ALLOCTYPE(inode) | - ((inode->i_mode & S_ISUID) ? ICBTAG_FLAG_SETUID : 0) | - ((inode->i_mode & S_ISGID) ? ICBTAG_FLAG_SETGID : 0) | - ((inode->i_mode & S_ISVTX) ? ICBTAG_FLAG_STICKY : 0) | - (le16_to_cpu(fe->icbTag.flags) & - ~(ICBTAG_FLAG_AD_MASK | ICBTAG_FLAG_SETUID | - ICBTAG_FLAG_SETGID | ICBTAG_FLAG_STICKY)); + icbflags = UDF_I_ALLOCTYPE(inode) | + ((inode->i_mode & S_ISUID) ? ICBTAG_FLAG_SETUID : 0) | + ((inode->i_mode & S_ISGID) ? ICBTAG_FLAG_SETGID : 0) | + ((inode->i_mode & S_ISVTX) ? ICBTAG_FLAG_STICKY : 0) | + (le16_to_cpu(fe->icbTag.flags) & + ~(ICBTAG_FLAG_AD_MASK | ICBTAG_FLAG_SETUID | + ICBTAG_FLAG_SETGID | ICBTAG_FLAG_STICKY)); fe->icbTag.flags = cpu_to_le16(icbflags); if (UDF_SB_UDFREV(inode->i_sb) >= 0x0200) @@ -1663,25 +1685,26 @@ udf_update_inode(struct inode *inode, int do_sync) else fe->descTag.descVersion = cpu_to_le16(2); fe->descTag.tagSerialNum = cpu_to_le16(UDF_SB_SERIALNUM(inode->i_sb)); - fe->descTag.tagLocation = cpu_to_le32(UDF_I_LOCATION(inode).logicalBlockNum); + fe->descTag.tagLocation = + cpu_to_le32(UDF_I_LOCATION(inode).logicalBlockNum); crclen += UDF_I_LENEATTR(inode) + UDF_I_LENALLOC(inode) - sizeof(tag); fe->descTag.descCRCLength = cpu_to_le16(crclen); - fe->descTag.descCRC = cpu_to_le16(udf_crc((char *)fe + sizeof(tag), crclen, 0)); + fe->descTag.descCRC = + cpu_to_le16(udf_crc((char *)fe + sizeof(tag), crclen, 0)); fe->descTag.tagChecksum = 0; - for (i=0; i<16; i++) + for (i = 0; i < 16; i++) if (i != 4) - fe->descTag.tagChecksum += ((uint8_t *)&(fe->descTag))[i]; + fe->descTag.tagChecksum += + ((uint8_t *) & (fe->descTag))[i]; /* write the data blocks */ mark_buffer_dirty(bh); - if (do_sync) - { + if (do_sync) { sync_dirty_buffer(bh); - if (buffer_req(bh) && !buffer_uptodate(bh)) - { + if (buffer_req(bh) && !buffer_uptodate(bh)) { printk("IO error syncing udf inode [%s:%08lx]\n", - inode->i_sb->s_id, inode->i_ino); + inode->i_sb->s_id, inode->i_ino); err = -EIO; } } @@ -1689,8 +1712,7 @@ udf_update_inode(struct inode *inode, int do_sync) return err; } -struct inode * -udf_iget(struct super_block *sb, kernel_lb_addr ino) +struct inode *udf_iget(struct super_block *sb, kernel_lb_addr ino) { unsigned long block = udf_get_lb_pblock(sb, ino, 0); struct inode *inode = iget_locked(sb, block); @@ -1707,22 +1729,23 @@ udf_iget(struct super_block *sb, kernel_lb_addr ino) if (is_bad_inode(inode)) goto out_iput; - if (ino.logicalBlockNum >= UDF_SB_PARTLEN(sb, ino.partitionReferenceNum)) { + if (ino.logicalBlockNum >= + UDF_SB_PARTLEN(sb, ino.partitionReferenceNum)) { udf_debug("block=%d, partition=%d out of range\n", - ino.logicalBlockNum, ino.partitionReferenceNum); + ino.logicalBlockNum, ino.partitionReferenceNum); make_bad_inode(inode); goto out_iput; } return inode; - out_iput: + out_iput: iput(inode); return NULL; } -int8_t udf_add_aext(struct inode *inode, struct extent_position *epos, - kernel_lb_addr eloc, uint32_t elen, int inc) +int8_t udf_add_aext(struct inode * inode, struct extent_position * epos, + kernel_lb_addr eloc, uint32_t elen, int inc) { int adsize; short_ad *sad = NULL; @@ -1732,7 +1755,9 @@ int8_t udf_add_aext(struct inode *inode, struct extent_position *epos, uint8_t *ptr; if (!epos->bh) - ptr = UDF_I_DATA(inode) + epos->offset - udf_file_entry_alloc_offset(inode) + UDF_I_LENEATTR(inode); + ptr = + UDF_I_DATA(inode) + epos->offset - + udf_file_entry_alloc_offset(inode) + UDF_I_LENEATTR(inode); else ptr = epos->bh->b_data + epos->offset; @@ -1743,21 +1768,24 @@ int8_t udf_add_aext(struct inode *inode, struct extent_position *epos, else return -1; - if (epos->offset + (2 * adsize) > inode->i_sb->s_blocksize) - { + if (epos->offset + (2 * adsize) > inode->i_sb->s_blocksize) { char *sptr, *dptr; struct buffer_head *nbh; int err, loffset; kernel_lb_addr obloc = epos->block; - if (!(epos->block.logicalBlockNum = udf_new_block(inode->i_sb, NULL, - obloc.partitionReferenceNum, obloc.logicalBlockNum, &err))) - { + if (! + (epos->block.logicalBlockNum = + udf_new_block(inode->i_sb, NULL, + obloc.partitionReferenceNum, + obloc.logicalBlockNum, &err))) { return -1; } - if (!(nbh = udf_tgetblk(inode->i_sb, udf_get_lb_pblock(inode->i_sb, - epos->block, 0)))) - { + if (! + (nbh = + udf_tgetblk(inode->i_sb, + udf_get_lb_pblock(inode->i_sb, epos->block, + 0)))) { return -1; } lock_buffer(nbh); @@ -1768,144 +1796,142 @@ int8_t udf_add_aext(struct inode *inode, struct extent_position *epos, aed = (struct allocExtDesc *)(nbh->b_data); if (!UDF_QUERY_FLAG(inode->i_sb, UDF_FLAG_STRICT)) - aed->previousAllocExtLocation = cpu_to_le32(obloc.logicalBlockNum); - if (epos->offset + adsize > inode->i_sb->s_blocksize) - { + aed->previousAllocExtLocation = + cpu_to_le32(obloc.logicalBlockNum); + if (epos->offset + adsize > inode->i_sb->s_blocksize) { loffset = epos->offset; aed->lengthAllocDescs = cpu_to_le32(adsize); sptr = ptr - adsize; dptr = nbh->b_data + sizeof(struct allocExtDesc); memcpy(dptr, sptr, adsize); epos->offset = sizeof(struct allocExtDesc) + adsize; - } - else - { + } else { loffset = epos->offset + adsize; aed->lengthAllocDescs = cpu_to_le32(0); sptr = ptr; epos->offset = sizeof(struct allocExtDesc); - if (epos->bh) - { + if (epos->bh) { aed = (struct allocExtDesc *)epos->bh->b_data; aed->lengthAllocDescs = - cpu_to_le32(le32_to_cpu(aed->lengthAllocDescs) + adsize); - } - else - { + cpu_to_le32(le32_to_cpu + (aed->lengthAllocDescs) + + adsize); + } else { UDF_I_LENALLOC(inode) += adsize; mark_inode_dirty(inode); } } if (UDF_SB_UDFREV(inode->i_sb) >= 0x0200) udf_new_tag(nbh->b_data, TAG_IDENT_AED, 3, 1, - epos->block.logicalBlockNum, sizeof(tag)); + epos->block.logicalBlockNum, sizeof(tag)); else udf_new_tag(nbh->b_data, TAG_IDENT_AED, 2, 1, - epos->block.logicalBlockNum, sizeof(tag)); - switch (UDF_I_ALLOCTYPE(inode)) - { - case ICBTAG_FLAG_AD_SHORT: + epos->block.logicalBlockNum, sizeof(tag)); + switch (UDF_I_ALLOCTYPE(inode)) { + case ICBTAG_FLAG_AD_SHORT: { - sad = (short_ad *)sptr; - sad->extLength = cpu_to_le32( - EXT_NEXT_EXTENT_ALLOCDECS | - inode->i_sb->s_blocksize); - sad->extPosition = cpu_to_le32(epos->block.logicalBlockNum); + sad = (short_ad *) sptr; + sad->extLength = + cpu_to_le32(EXT_NEXT_EXTENT_ALLOCDECS | + inode->i_sb->s_blocksize); + sad->extPosition = + cpu_to_le32(epos->block.logicalBlockNum); break; } - case ICBTAG_FLAG_AD_LONG: + case ICBTAG_FLAG_AD_LONG: { - lad = (long_ad *)sptr; - lad->extLength = cpu_to_le32( - EXT_NEXT_EXTENT_ALLOCDECS | - inode->i_sb->s_blocksize); + lad = (long_ad *) sptr; + lad->extLength = + cpu_to_le32(EXT_NEXT_EXTENT_ALLOCDECS | + inode->i_sb->s_blocksize); lad->extLocation = cpu_to_lelb(epos->block); memset(lad->impUse, 0x00, sizeof(lad->impUse)); break; } } - if (epos->bh) - { - if (!UDF_QUERY_FLAG(inode->i_sb, UDF_FLAG_STRICT) || UDF_SB_UDFREV(inode->i_sb) >= 0x0201) + if (epos->bh) { + if (!UDF_QUERY_FLAG(inode->i_sb, UDF_FLAG_STRICT) + || UDF_SB_UDFREV(inode->i_sb) >= 0x0201) udf_update_tag(epos->bh->b_data, loffset); else - udf_update_tag(epos->bh->b_data, sizeof(struct allocExtDesc)); + udf_update_tag(epos->bh->b_data, + sizeof(struct allocExtDesc)); mark_buffer_dirty_inode(epos->bh, inode); brelse(epos->bh); - } - else + } else mark_inode_dirty(inode); epos->bh = nbh; } etype = udf_write_aext(inode, epos, eloc, elen, inc); - if (!epos->bh) - { + if (!epos->bh) { UDF_I_LENALLOC(inode) += adsize; mark_inode_dirty(inode); - } - else - { + } else { aed = (struct allocExtDesc *)epos->bh->b_data; aed->lengthAllocDescs = - cpu_to_le32(le32_to_cpu(aed->lengthAllocDescs) + adsize); - if (!UDF_QUERY_FLAG(inode->i_sb, UDF_FLAG_STRICT) || UDF_SB_UDFREV(inode->i_sb) >= 0x0201) - udf_update_tag(epos->bh->b_data, epos->offset + (inc ? 0 : adsize)); + cpu_to_le32(le32_to_cpu(aed->lengthAllocDescs) + adsize); + if (!UDF_QUERY_FLAG(inode->i_sb, UDF_FLAG_STRICT) + || UDF_SB_UDFREV(inode->i_sb) >= 0x0201) + udf_update_tag(epos->bh->b_data, + epos->offset + (inc ? 0 : adsize)); else - udf_update_tag(epos->bh->b_data, sizeof(struct allocExtDesc)); + udf_update_tag(epos->bh->b_data, + sizeof(struct allocExtDesc)); mark_buffer_dirty_inode(epos->bh, inode); } return etype; } -int8_t udf_write_aext(struct inode *inode, struct extent_position *epos, - kernel_lb_addr eloc, uint32_t elen, int inc) +int8_t udf_write_aext(struct inode * inode, struct extent_position * epos, + kernel_lb_addr eloc, uint32_t elen, int inc) { int adsize; uint8_t *ptr; if (!epos->bh) - ptr = UDF_I_DATA(inode) + epos->offset - udf_file_entry_alloc_offset(inode) + UDF_I_LENEATTR(inode); + ptr = + UDF_I_DATA(inode) + epos->offset - + udf_file_entry_alloc_offset(inode) + UDF_I_LENEATTR(inode); else ptr = epos->bh->b_data + epos->offset; - switch (UDF_I_ALLOCTYPE(inode)) - { - case ICBTAG_FLAG_AD_SHORT: + switch (UDF_I_ALLOCTYPE(inode)) { + case ICBTAG_FLAG_AD_SHORT: { - short_ad *sad = (short_ad *)ptr; + short_ad *sad = (short_ad *) ptr; sad->extLength = cpu_to_le32(elen); sad->extPosition = cpu_to_le32(eloc.logicalBlockNum); adsize = sizeof(short_ad); break; } - case ICBTAG_FLAG_AD_LONG: + case ICBTAG_FLAG_AD_LONG: { - long_ad *lad = (long_ad *)ptr; + long_ad *lad = (long_ad *) ptr; lad->extLength = cpu_to_le32(elen); lad->extLocation = cpu_to_lelb(eloc); memset(lad->impUse, 0x00, sizeof(lad->impUse)); adsize = sizeof(long_ad); break; } - default: - return -1; + default: + return -1; } - if (epos->bh) - { - if (!UDF_QUERY_FLAG(inode->i_sb, UDF_FLAG_STRICT) || UDF_SB_UDFREV(inode->i_sb) >= 0x0201) - { - struct allocExtDesc *aed = (struct allocExtDesc *)epos->bh->b_data; + if (epos->bh) { + if (!UDF_QUERY_FLAG(inode->i_sb, UDF_FLAG_STRICT) + || UDF_SB_UDFREV(inode->i_sb) >= 0x0201) { + struct allocExtDesc *aed = + (struct allocExtDesc *)epos->bh->b_data; udf_update_tag(epos->bh->b_data, - le32_to_cpu(aed->lengthAllocDescs) + sizeof(struct allocExtDesc)); + le32_to_cpu(aed->lengthAllocDescs) + + sizeof(struct allocExtDesc)); } mark_buffer_dirty_inode(epos->bh, inode); - } - else + } else mark_inode_dirty(inode); if (inc) @@ -1913,21 +1939,24 @@ int8_t udf_write_aext(struct inode *inode, struct extent_position *epos, return (elen >> 30); } -int8_t udf_next_aext(struct inode *inode, struct extent_position *epos, - kernel_lb_addr *eloc, uint32_t *elen, int inc) +int8_t udf_next_aext(struct inode * inode, struct extent_position * epos, + kernel_lb_addr * eloc, uint32_t * elen, int inc) { int8_t etype; while ((etype = udf_current_aext(inode, epos, eloc, elen, inc)) == - (EXT_NEXT_EXTENT_ALLOCDECS >> 30)) - { + (EXT_NEXT_EXTENT_ALLOCDECS >> 30)) { epos->block = *eloc; epos->offset = sizeof(struct allocExtDesc); brelse(epos->bh); - if (!(epos->bh = udf_tread(inode->i_sb, udf_get_lb_pblock(inode->i_sb, epos->block, 0)))) - { + if (! + (epos->bh = + udf_tread(inode->i_sb, + udf_get_lb_pblock(inode->i_sb, epos->block, + 0)))) { udf_debug("reading block %d failed!\n", - udf_get_lb_pblock(inode->i_sb, epos->block, 0)); + udf_get_lb_pblock(inode->i_sb, epos->block, + 0)); return -1; } } @@ -1935,58 +1964,71 @@ int8_t udf_next_aext(struct inode *inode, struct extent_position *epos, return etype; } -int8_t udf_current_aext(struct inode *inode, struct extent_position *epos, - kernel_lb_addr *eloc, uint32_t *elen, int inc) +int8_t udf_current_aext(struct inode * inode, struct extent_position * epos, + kernel_lb_addr * eloc, uint32_t * elen, int inc) { int alen; int8_t etype; uint8_t *ptr; - if (!epos->bh) - { + if (!epos->bh) { if (!epos->offset) epos->offset = udf_file_entry_alloc_offset(inode); - ptr = UDF_I_DATA(inode) + epos->offset - udf_file_entry_alloc_offset(inode) + UDF_I_LENEATTR(inode); - alen = udf_file_entry_alloc_offset(inode) + UDF_I_LENALLOC(inode); - } - else - { + ptr = + UDF_I_DATA(inode) + epos->offset - + udf_file_entry_alloc_offset(inode) + UDF_I_LENEATTR(inode); + alen = + udf_file_entry_alloc_offset(inode) + UDF_I_LENALLOC(inode); + } else { if (!epos->offset) epos->offset = sizeof(struct allocExtDesc); ptr = epos->bh->b_data + epos->offset; - alen = sizeof(struct allocExtDesc) + le32_to_cpu(((struct allocExtDesc *)epos->bh->b_data)->lengthAllocDescs); + alen = + sizeof(struct allocExtDesc) + + le32_to_cpu(((struct allocExtDesc *)epos->bh->b_data)-> + lengthAllocDescs); } - switch (UDF_I_ALLOCTYPE(inode)) - { - case ICBTAG_FLAG_AD_SHORT: + switch (UDF_I_ALLOCTYPE(inode)) { + case ICBTAG_FLAG_AD_SHORT: { short_ad *sad; - if (!(sad = udf_get_fileshortad(ptr, alen, &epos->offset, inc))) + if (! + (sad = + udf_get_fileshortad(ptr, alen, &epos->offset, + inc))) return -1; etype = le32_to_cpu(sad->extLength) >> 30; eloc->logicalBlockNum = le32_to_cpu(sad->extPosition); - eloc->partitionReferenceNum = UDF_I_LOCATION(inode).partitionReferenceNum; - *elen = le32_to_cpu(sad->extLength) & UDF_EXTENT_LENGTH_MASK; + eloc->partitionReferenceNum = + UDF_I_LOCATION(inode).partitionReferenceNum; + *elen = + le32_to_cpu(sad-> + extLength) & UDF_EXTENT_LENGTH_MASK; break; } - case ICBTAG_FLAG_AD_LONG: + case ICBTAG_FLAG_AD_LONG: { long_ad *lad; - if (!(lad = udf_get_filelongad(ptr, alen, &epos->offset, inc))) + if (! + (lad = + udf_get_filelongad(ptr, alen, &epos->offset, inc))) return -1; etype = le32_to_cpu(lad->extLength) >> 30; *eloc = lelb_to_cpu(lad->extLocation); - *elen = le32_to_cpu(lad->extLength) & UDF_EXTENT_LENGTH_MASK; + *elen = + le32_to_cpu(lad-> + extLength) & UDF_EXTENT_LENGTH_MASK; break; } - default: + default: { - udf_debug("alloc_type = %d unsupported\n", UDF_I_ALLOCTYPE(inode)); + udf_debug("alloc_type = %d unsupported\n", + UDF_I_ALLOCTYPE(inode)); return -1; } } @@ -2005,8 +2047,7 @@ udf_insert_aext(struct inode *inode, struct extent_position epos, if (epos.bh) get_bh(epos.bh); - while ((etype = udf_next_aext(inode, &epos, &oeloc, &oelen, 0)) != -1) - { + while ((etype = udf_next_aext(inode, &epos, &oeloc, &oelen, 0)) != -1) { udf_write_aext(inode, &epos, neloc, nelen, 1); neloc = oeloc; @@ -2017,16 +2058,15 @@ udf_insert_aext(struct inode *inode, struct extent_position epos, return (nelen >> 30); } -int8_t udf_delete_aext(struct inode *inode, struct extent_position epos, - kernel_lb_addr eloc, uint32_t elen) +int8_t udf_delete_aext(struct inode * inode, struct extent_position epos, + kernel_lb_addr eloc, uint32_t elen) { struct extent_position oepos; int adsize; int8_t etype; struct allocExtDesc *aed; - if (epos.bh) - { + if (epos.bh) { get_bh(epos.bh); get_bh(epos.bh); } @@ -2042,11 +2082,9 @@ int8_t udf_delete_aext(struct inode *inode, struct extent_position epos, if (udf_next_aext(inode, &epos, &eloc, &elen, 1) == -1) return -1; - while ((etype = udf_next_aext(inode, &epos, &eloc, &elen, 1)) != -1) - { + while ((etype = udf_next_aext(inode, &epos, &eloc, &elen, 1)) != -1) { udf_write_aext(inode, &oepos, eloc, (etype << 30) | elen, 1); - if (oepos.bh != epos.bh) - { + if (oepos.bh != epos.bh) { oepos.block = epos.block; brelse(oepos.bh); get_bh(epos.bh); @@ -2057,45 +2095,44 @@ int8_t udf_delete_aext(struct inode *inode, struct extent_position epos, memset(&eloc, 0x00, sizeof(kernel_lb_addr)); elen = 0; - if (epos.bh != oepos.bh) - { + if (epos.bh != oepos.bh) { udf_free_blocks(inode->i_sb, inode, epos.block, 0, 1); udf_write_aext(inode, &oepos, eloc, elen, 1); udf_write_aext(inode, &oepos, eloc, elen, 1); - if (!oepos.bh) - { + if (!oepos.bh) { UDF_I_LENALLOC(inode) -= (adsize * 2); mark_inode_dirty(inode); - } - else - { + } else { aed = (struct allocExtDesc *)oepos.bh->b_data; aed->lengthAllocDescs = - cpu_to_le32(le32_to_cpu(aed->lengthAllocDescs) - (2*adsize)); - if (!UDF_QUERY_FLAG(inode->i_sb, UDF_FLAG_STRICT) || UDF_SB_UDFREV(inode->i_sb) >= 0x0201) - udf_update_tag(oepos.bh->b_data, oepos.offset - (2*adsize)); + cpu_to_le32(le32_to_cpu(aed->lengthAllocDescs) - + (2 * adsize)); + if (!UDF_QUERY_FLAG(inode->i_sb, UDF_FLAG_STRICT) + || UDF_SB_UDFREV(inode->i_sb) >= 0x0201) + udf_update_tag(oepos.bh->b_data, + oepos.offset - (2 * adsize)); else - udf_update_tag(oepos.bh->b_data, sizeof(struct allocExtDesc)); + udf_update_tag(oepos.bh->b_data, + sizeof(struct allocExtDesc)); mark_buffer_dirty_inode(oepos.bh, inode); } - } - else - { + } else { udf_write_aext(inode, &oepos, eloc, elen, 1); - if (!oepos.bh) - { + if (!oepos.bh) { UDF_I_LENALLOC(inode) -= adsize; mark_inode_dirty(inode); - } - else - { + } else { aed = (struct allocExtDesc *)oepos.bh->b_data; aed->lengthAllocDescs = - cpu_to_le32(le32_to_cpu(aed->lengthAllocDescs) - adsize); - if (!UDF_QUERY_FLAG(inode->i_sb, UDF_FLAG_STRICT) || UDF_SB_UDFREV(inode->i_sb) >= 0x0201) - udf_update_tag(oepos.bh->b_data, epos.offset - adsize); + cpu_to_le32(le32_to_cpu(aed->lengthAllocDescs) - + adsize); + if (!UDF_QUERY_FLAG(inode->i_sb, UDF_FLAG_STRICT) + || UDF_SB_UDFREV(inode->i_sb) >= 0x0201) + udf_update_tag(oepos.bh->b_data, + epos.offset - adsize); else - udf_update_tag(oepos.bh->b_data, sizeof(struct allocExtDesc)); + udf_update_tag(oepos.bh->b_data, + sizeof(struct allocExtDesc)); mark_buffer_dirty_inode(oepos.bh, inode); } } @@ -2105,14 +2142,15 @@ int8_t udf_delete_aext(struct inode *inode, struct extent_position epos, return (elen >> 30); } -int8_t inode_bmap(struct inode *inode, sector_t block, struct extent_position *pos, - kernel_lb_addr *eloc, uint32_t *elen, sector_t *offset) +int8_t inode_bmap(struct inode * inode, sector_t block, + struct extent_position * pos, kernel_lb_addr * eloc, + uint32_t * elen, sector_t * offset) { - loff_t lbcount = 0, bcount = (loff_t)block << inode->i_sb->s_blocksize_bits; + loff_t lbcount = 0, bcount = + (loff_t) block << inode->i_sb->s_blocksize_bits; int8_t etype; - if (block < 0) - { + if (block < 0) { printk(KERN_ERR "udf: inode_bmap: block < 0\n"); return -1; } @@ -2122,11 +2160,10 @@ int8_t inode_bmap(struct inode *inode, sector_t block, struct extent_position *p pos->bh = NULL; *elen = 0; - do - { - if ((etype = udf_next_aext(inode, pos, eloc, elen, 1)) == -1) - { - *offset = (bcount - lbcount) >> inode->i_sb->s_blocksize_bits; + do { + if ((etype = udf_next_aext(inode, pos, eloc, elen, 1)) == -1) { + *offset = + (bcount - lbcount) >> inode->i_sb->s_blocksize_bits; UDF_I_LENEXTENTS(inode) = lbcount; return -1; } @@ -2143,12 +2180,13 @@ long udf_block_map(struct inode *inode, sector_t block) kernel_lb_addr eloc; uint32_t elen; sector_t offset; - struct extent_position epos = { NULL, 0, { 0, 0}}; + struct extent_position epos = { NULL, 0, {0, 0} }; int ret; lock_kernel(); - if (inode_bmap(inode, block, &epos, &eloc, &elen, &offset) == (EXT_RECORDED_ALLOCATED >> 30)) + if (inode_bmap(inode, block, &epos, &eloc, &elen, &offset) == + (EXT_RECORDED_ALLOCATED >> 30)) ret = udf_get_lb_pblock(inode->i_sb, eloc, offset); else ret = 0; diff --git a/fs/udf/lowlevel.c b/fs/udf/lowlevel.c index 08421610766..4826c3616ee 100644 --- a/fs/udf/lowlevel.c +++ b/fs/udf/lowlevel.c @@ -26,43 +26,38 @@ #include <linux/udf_fs.h> #include "udf_sb.h" -unsigned int -udf_get_last_session(struct super_block *sb) +unsigned int udf_get_last_session(struct super_block *sb) { struct cdrom_multisession ms_info; unsigned int vol_desc_start; struct block_device *bdev = sb->s_bdev; int i; - vol_desc_start=0; - ms_info.addr_format=CDROM_LBA; - i = ioctl_by_bdev(bdev, CDROMMULTISESSION, (unsigned long) &ms_info); + vol_desc_start = 0; + ms_info.addr_format = CDROM_LBA; + i = ioctl_by_bdev(bdev, CDROMMULTISESSION, (unsigned long)&ms_info); #define WE_OBEY_THE_WRITTEN_STANDARDS 1 - if (i == 0) - { + if (i == 0) { udf_debug("XA disk: %s, vol_desc_start=%d\n", - (ms_info.xa_flag ? "yes" : "no"), ms_info.addr.lba); + (ms_info.xa_flag ? "yes" : "no"), ms_info.addr.lba); #if WE_OBEY_THE_WRITTEN_STANDARDS - if (ms_info.xa_flag) /* necessary for a valid ms_info.addr */ + if (ms_info.xa_flag) /* necessary for a valid ms_info.addr */ #endif vol_desc_start = ms_info.addr.lba; - } - else - { + } else { udf_debug("CDROMMULTISESSION not supported: rc=%d\n", i); } return vol_desc_start; } -unsigned long -udf_get_last_block(struct super_block *sb) +unsigned long udf_get_last_block(struct super_block *sb) { struct block_device *bdev = sb->s_bdev; unsigned long lblock = 0; - if (ioctl_by_bdev(bdev, CDROM_LAST_WRITTEN, (unsigned long) &lblock)) + if (ioctl_by_bdev(bdev, CDROM_LAST_WRITTEN, (unsigned long)&lblock)) lblock = bdev->bd_inode->i_size >> sb->s_blocksize_bits; if (lblock) diff --git a/fs/udf/misc.c b/fs/udf/misc.c index a2b2a98ce78..a7f57277a96 100644 --- a/fs/udf/misc.c +++ b/fs/udf/misc.c @@ -29,8 +29,7 @@ #include "udf_i.h" #include "udf_sb.h" -struct buffer_head * -udf_tgetblk(struct super_block *sb, int block) +struct buffer_head *udf_tgetblk(struct super_block *sb, int block) { if (UDF_QUERY_FLAG(sb, UDF_FLAG_VARCONV)) return sb_getblk(sb, udf_fixed_to_variable(block)); @@ -38,8 +37,7 @@ udf_tgetblk(struct super_block *sb, int block) return sb_getblk(sb, block); } -struct buffer_head * -udf_tread(struct super_block *sb, int block) +struct buffer_head *udf_tread(struct super_block *sb, int block) { if (UDF_QUERY_FLAG(sb, UDF_FLAG_VARCONV)) return sb_bread(sb, udf_fixed_to_variable(block)); @@ -47,9 +45,8 @@ udf_tread(struct super_block *sb, int block) return sb_bread(sb, block); } -struct genericFormat * -udf_add_extendedattr(struct inode * inode, uint32_t size, uint32_t type, - uint8_t loc) +struct genericFormat *udf_add_extendedattr(struct inode *inode, uint32_t size, + uint32_t type, uint8_t loc) { uint8_t *ea = NULL, *ad = NULL; int offset; @@ -59,78 +56,76 @@ udf_add_extendedattr(struct inode * inode, uint32_t size, uint32_t type, ea = UDF_I_DATA(inode); if (UDF_I_LENEATTR(inode)) ad = UDF_I_DATA(inode) + UDF_I_LENEATTR(inode); - else - { + else { ad = ea; size += sizeof(struct extendedAttrHeaderDesc); } offset = inode->i_sb->s_blocksize - udf_file_entry_alloc_offset(inode) - - UDF_I_LENALLOC(inode); + UDF_I_LENALLOC(inode); /* TODO - Check for FreeEASpace */ - if (loc & 0x01 && offset >= size) - { + if (loc & 0x01 && offset >= size) { struct extendedAttrHeaderDesc *eahd; eahd = (struct extendedAttrHeaderDesc *)ea; - if (UDF_I_LENALLOC(inode)) - { + if (UDF_I_LENALLOC(inode)) { memmove(&ad[size], ad, UDF_I_LENALLOC(inode)); } - if (UDF_I_LENEATTR(inode)) - { + if (UDF_I_LENEATTR(inode)) { /* check checksum/crc */ - if (le16_to_cpu(eahd->descTag.tagIdent) != TAG_IDENT_EAHD || - le32_to_cpu(eahd->descTag.tagLocation) != UDF_I_LOCATION(inode).logicalBlockNum) - { + if (le16_to_cpu(eahd->descTag.tagIdent) != + TAG_IDENT_EAHD + || le32_to_cpu(eahd->descTag.tagLocation) != + UDF_I_LOCATION(inode).logicalBlockNum) { return NULL; } - } - else - { + } else { size -= sizeof(struct extendedAttrHeaderDesc); - UDF_I_LENEATTR(inode) += sizeof(struct extendedAttrHeaderDesc); + UDF_I_LENEATTR(inode) += + sizeof(struct extendedAttrHeaderDesc); eahd->descTag.tagIdent = cpu_to_le16(TAG_IDENT_EAHD); if (UDF_SB_UDFREV(inode->i_sb) >= 0x0200) eahd->descTag.descVersion = cpu_to_le16(3); else eahd->descTag.descVersion = cpu_to_le16(2); - eahd->descTag.tagSerialNum = cpu_to_le16(UDF_SB_SERIALNUM(inode->i_sb)); - eahd->descTag.tagLocation = cpu_to_le32(UDF_I_LOCATION(inode).logicalBlockNum); + eahd->descTag.tagSerialNum = + cpu_to_le16(UDF_SB_SERIALNUM(inode->i_sb)); + eahd->descTag.tagLocation = + cpu_to_le32(UDF_I_LOCATION(inode).logicalBlockNum); eahd->impAttrLocation = cpu_to_le32(0xFFFFFFFF); eahd->appAttrLocation = cpu_to_le32(0xFFFFFFFF); } offset = UDF_I_LENEATTR(inode); - if (type < 2048) - { - if (le32_to_cpu(eahd->appAttrLocation) < UDF_I_LENEATTR(inode)) - { - uint32_t aal = le32_to_cpu(eahd->appAttrLocation); - memmove(&ea[offset - aal + size], - &ea[aal], offset - aal); + if (type < 2048) { + if (le32_to_cpu(eahd->appAttrLocation) < + UDF_I_LENEATTR(inode)) { + uint32_t aal = + le32_to_cpu(eahd->appAttrLocation); + memmove(&ea[offset - aal + size], &ea[aal], + offset - aal); offset -= aal; eahd->appAttrLocation = cpu_to_le32(aal + size); } - if (le32_to_cpu(eahd->impAttrLocation) < UDF_I_LENEATTR(inode)) - { - uint32_t ial = le32_to_cpu(eahd->impAttrLocation); - memmove(&ea[offset - ial + size], - &ea[ial], offset - ial); + if (le32_to_cpu(eahd->impAttrLocation) < + UDF_I_LENEATTR(inode)) { + uint32_t ial = + le32_to_cpu(eahd->impAttrLocation); + memmove(&ea[offset - ial + size], &ea[ial], + offset - ial); offset -= ial; eahd->impAttrLocation = cpu_to_le32(ial + size); } - } - else if (type < 65536) - { - if (le32_to_cpu(eahd->appAttrLocation) < UDF_I_LENEATTR(inode)) - { - uint32_t aal = le32_to_cpu(eahd->appAttrLocation); - memmove(&ea[offset - aal + size], - &ea[aal], offset - aal); + } else if (type < 65536) { + if (le32_to_cpu(eahd->appAttrLocation) < + UDF_I_LENEATTR(inode)) { + uint32_t aal = + le32_to_cpu(eahd->appAttrLocation); + memmove(&ea[offset - aal + size], &ea[aal], + offset - aal); offset -= aal; eahd->appAttrLocation = cpu_to_le32(aal + size); } @@ -138,22 +133,23 @@ udf_add_extendedattr(struct inode * inode, uint32_t size, uint32_t type, /* rewrite CRC + checksum of eahd */ crclen = sizeof(struct extendedAttrHeaderDesc) - sizeof(tag); eahd->descTag.descCRCLength = cpu_to_le16(crclen); - eahd->descTag.descCRC = cpu_to_le16(udf_crc((char *)eahd + sizeof(tag), crclen, 0)); + eahd->descTag.descCRC = + cpu_to_le16(udf_crc((char *)eahd + sizeof(tag), crclen, 0)); eahd->descTag.tagChecksum = 0; - for (i=0; i<16; i++) + for (i = 0; i < 16; i++) if (i != 4) - eahd->descTag.tagChecksum += ((uint8_t *)&(eahd->descTag))[i]; + eahd->descTag.tagChecksum += + ((uint8_t *) & (eahd->descTag))[i]; UDF_I_LENEATTR(inode) += size; return (struct genericFormat *)&ea[offset]; } - if (loc & 0x02) - { + if (loc & 0x02) { } return NULL; } -struct genericFormat * -udf_get_extendedattr(struct inode *inode, uint32_t type, uint8_t subtype) +struct genericFormat *udf_get_extendedattr(struct inode *inode, uint32_t type, + uint8_t subtype) { struct genericFormat *gaf; uint8_t *ea = NULL; @@ -161,18 +157,17 @@ udf_get_extendedattr(struct inode *inode, uint32_t type, uint8_t subtype) ea = UDF_I_DATA(inode); - if (UDF_I_LENEATTR(inode)) - { + if (UDF_I_LENEATTR(inode)) { struct extendedAttrHeaderDesc *eahd; eahd = (struct extendedAttrHeaderDesc *)ea; /* check checksum/crc */ if (le16_to_cpu(eahd->descTag.tagIdent) != TAG_IDENT_EAHD || - le32_to_cpu(eahd->descTag.tagLocation) != UDF_I_LOCATION(inode).logicalBlockNum) - { + le32_to_cpu(eahd->descTag.tagLocation) != + UDF_I_LOCATION(inode).logicalBlockNum) { return NULL; } - + if (type < 2048) offset = sizeof(struct extendedAttrHeaderDesc); else if (type < 65536) @@ -180,10 +175,10 @@ udf_get_extendedattr(struct inode *inode, uint32_t type, uint8_t subtype) else offset = le32_to_cpu(eahd->appAttrLocation); - while (offset < UDF_I_LENEATTR(inode)) - { + while (offset < UDF_I_LENEATTR(inode)) { gaf = (struct genericFormat *)&ea[offset]; - if (le32_to_cpu(gaf->attrType) == type && gaf->attrSubtype == subtype) + if (le32_to_cpu(gaf->attrType) == type + && gaf->attrSubtype == subtype) return gaf; else offset += le32_to_cpu(gaf->attrLength); @@ -202,8 +197,8 @@ udf_get_extendedattr(struct inode *inode, uint32_t type, uint8_t subtype) * July 1, 1997 - Andrew E. Mileski * Written, tested, and released. */ -struct buffer_head * -udf_read_tagged(struct super_block *sb, uint32_t block, uint32_t location, uint16_t *ident) +struct buffer_head *udf_read_tagged(struct super_block *sb, uint32_t block, + uint32_t location, uint16_t * ident) { tag *tag_p; struct buffer_head *bh = NULL; @@ -215,29 +210,29 @@ udf_read_tagged(struct super_block *sb, uint32_t block, uint32_t location, uint1 return NULL; bh = udf_tread(sb, block + UDF_SB_SESSION(sb)); - if (!bh) - { - udf_debug("block=%d, location=%d: read failed\n", block + UDF_SB_SESSION(sb), location); + if (!bh) { + udf_debug("block=%d, location=%d: read failed\n", + block + UDF_SB_SESSION(sb), location); return NULL; } - tag_p = (tag *)(bh->b_data); + tag_p = (tag *) (bh->b_data); *ident = le16_to_cpu(tag_p->tagIdent); - if ( location != le32_to_cpu(tag_p->tagLocation) ) - { + if (location != le32_to_cpu(tag_p->tagLocation)) { udf_debug("location mismatch block %u, tag %u != %u\n", - block + UDF_SB_SESSION(sb), le32_to_cpu(tag_p->tagLocation), location); + block + UDF_SB_SESSION(sb), + le32_to_cpu(tag_p->tagLocation), location); goto error_out; } - + /* Verify the tag checksum */ checksum = 0U; for (i = 0; i < 4; i++) - checksum += (uint8_t)(bh->b_data[i]); + checksum += (uint8_t) (bh->b_data[i]); for (i = 5; i < 16; i++) - checksum += (uint8_t)(bh->b_data[i]); + checksum += (uint8_t) (bh->b_data[i]); if (checksum != tag_p->tagChecksum) { printk(KERN_ERR "udf: tag checksum failed block %d\n", block); goto error_out; @@ -245,38 +240,39 @@ udf_read_tagged(struct super_block *sb, uint32_t block, uint32_t location, uint1 /* Verify the tag version */ if (le16_to_cpu(tag_p->descVersion) != 0x0002U && - le16_to_cpu(tag_p->descVersion) != 0x0003U) - { + le16_to_cpu(tag_p->descVersion) != 0x0003U) { udf_debug("tag version 0x%04x != 0x0002 || 0x0003 block %d\n", - le16_to_cpu(tag_p->descVersion), block); + le16_to_cpu(tag_p->descVersion), block); goto error_out; } /* Verify the descriptor CRC */ if (le16_to_cpu(tag_p->descCRCLength) + sizeof(tag) > sb->s_blocksize || - le16_to_cpu(tag_p->descCRC) == udf_crc(bh->b_data + sizeof(tag), - le16_to_cpu(tag_p->descCRCLength), 0)) - { + le16_to_cpu(tag_p->descCRC) == udf_crc(bh->b_data + sizeof(tag), + le16_to_cpu(tag_p-> + descCRCLength), + 0)) { return bh; } udf_debug("Crc failure block %d: crc = %d, crclen = %d\n", - block + UDF_SB_SESSION(sb), le16_to_cpu(tag_p->descCRC), le16_to_cpu(tag_p->descCRCLength)); + block + UDF_SB_SESSION(sb), le16_to_cpu(tag_p->descCRC), + le16_to_cpu(tag_p->descCRCLength)); -error_out: + error_out: brelse(bh); return NULL; } -struct buffer_head * -udf_read_ptagged(struct super_block *sb, kernel_lb_addr loc, uint32_t offset, uint16_t *ident) +struct buffer_head *udf_read_ptagged(struct super_block *sb, kernel_lb_addr loc, + uint32_t offset, uint16_t * ident) { return udf_read_tagged(sb, udf_get_lb_pblock(sb, loc, offset), - loc.logicalBlockNum + offset, ident); + loc.logicalBlockNum + offset, ident); } void udf_update_tag(char *data, int length) { - tag *tptr = (tag *)data; + tag *tptr = (tag *) data; int i; length -= sizeof(tag); @@ -285,15 +281,15 @@ void udf_update_tag(char *data, int length) tptr->descCRCLength = cpu_to_le16(length); tptr->descCRC = cpu_to_le16(udf_crc(data + sizeof(tag), length, 0)); - for (i=0; i<16; i++) + for (i = 0; i < 16; i++) if (i != 4) - tptr->tagChecksum += (uint8_t)(data[i]); + tptr->tagChecksum += (uint8_t) (data[i]); } void udf_new_tag(char *data, uint16_t ident, uint16_t version, uint16_t snum, - uint32_t loc, int length) + uint32_t loc, int length) { - tag *tptr = (tag *)data; + tag *tptr = (tag *) data; tptr->tagIdent = cpu_to_le16(ident); tptr->descVersion = cpu_to_le16(version); tptr->tagSerialNum = cpu_to_le16(snum); diff --git a/fs/udf/namei.c b/fs/udf/namei.c index 51fe307dc0e..334d363a090 100644 --- a/fs/udf/namei.c +++ b/fs/udf/namei.c @@ -32,7 +32,8 @@ #include <linux/buffer_head.h> #include <linux/sched.h> -static inline int udf_match(int len1, const char *name1, int len2, const char *name2) +static inline int udf_match(int len1, const char *name1, int len2, + const char *name2) { if (len1 != len2) return 0; @@ -40,8 +41,8 @@ static inline int udf_match(int len1, const char *name1, int len2, const char *n } int udf_write_fi(struct inode *inode, struct fileIdentDesc *cfi, - struct fileIdentDesc *sfi, struct udf_fileident_bh *fibh, - uint8_t *impuse, uint8_t *fileident) + struct fileIdentDesc *sfi, struct udf_fileident_bh *fibh, + uint8_t * impuse, uint8_t * fileident) { uint16_t crclen = fibh->eoffset - fibh->soffset - sizeof(tag); uint16_t crc; @@ -51,7 +52,7 @@ int udf_write_fi(struct inode *inode, struct fileIdentDesc *cfi, uint16_t liu = le16_to_cpu(cfi->lengthOfImpUse); uint8_t lfi = cfi->lengthFileIdent; int padlen = fibh->eoffset - fibh->soffset - liu - lfi - - sizeof(struct fileIdentDesc); + sizeof(struct fileIdentDesc); int adinicb = 0; if (UDF_I_ALLOCTYPE(inode) == ICBTAG_FLAG_AD_IN_ICB) @@ -59,83 +60,86 @@ int udf_write_fi(struct inode *inode, struct fileIdentDesc *cfi, offset = fibh->soffset + sizeof(struct fileIdentDesc); - if (impuse) - { + if (impuse) { if (adinicb || (offset + liu < 0)) - memcpy((uint8_t *)sfi->impUse, impuse, liu); + memcpy((uint8_t *) sfi->impUse, impuse, liu); else if (offset >= 0) memcpy(fibh->ebh->b_data + offset, impuse, liu); - else - { - memcpy((uint8_t *)sfi->impUse, impuse, -offset); - memcpy(fibh->ebh->b_data, impuse - offset, liu + offset); + else { + memcpy((uint8_t *) sfi->impUse, impuse, -offset); + memcpy(fibh->ebh->b_data, impuse - offset, + liu + offset); } } offset += liu; - if (fileident) - { + if (fileident) { if (adinicb || (offset + lfi < 0)) - memcpy((uint8_t *)sfi->fileIdent + liu, fileident, lfi); + memcpy((uint8_t *) sfi->fileIdent + liu, fileident, + lfi); else if (offset >= 0) memcpy(fibh->ebh->b_data + offset, fileident, lfi); - else - { - memcpy((uint8_t *)sfi->fileIdent + liu, fileident, -offset); - memcpy(fibh->ebh->b_data, fileident - offset, lfi + offset); + else { + memcpy((uint8_t *) sfi->fileIdent + liu, fileident, + -offset); + memcpy(fibh->ebh->b_data, fileident - offset, + lfi + offset); } } offset += lfi; if (adinicb || (offset + padlen < 0)) - memset((uint8_t *)sfi->padding + liu + lfi, 0x00, padlen); + memset((uint8_t *) sfi->padding + liu + lfi, 0x00, padlen); else if (offset >= 0) memset(fibh->ebh->b_data + offset, 0x00, padlen); - else - { - memset((uint8_t *)sfi->padding + liu + lfi, 0x00, -offset); + else { + memset((uint8_t *) sfi->padding + liu + lfi, 0x00, -offset); memset(fibh->ebh->b_data, 0x00, padlen + offset); } - crc = udf_crc((uint8_t *)cfi + sizeof(tag), sizeof(struct fileIdentDesc) - - sizeof(tag), 0); + crc = + udf_crc((uint8_t *) cfi + sizeof(tag), + sizeof(struct fileIdentDesc) - sizeof(tag), 0); if (fibh->sbh == fibh->ebh) - crc = udf_crc((uint8_t *)sfi->impUse, - crclen + sizeof(tag) - sizeof(struct fileIdentDesc), crc); + crc = udf_crc((uint8_t *) sfi->impUse, + crclen + sizeof(tag) - + sizeof(struct fileIdentDesc), crc); else if (sizeof(struct fileIdentDesc) >= -fibh->soffset) - crc = udf_crc(fibh->ebh->b_data + sizeof(struct fileIdentDesc) + fibh->soffset, - crclen + sizeof(tag) - sizeof(struct fileIdentDesc), crc); - else - { - crc = udf_crc((uint8_t *)sfi->impUse, - -fibh->soffset - sizeof(struct fileIdentDesc), crc); + crc = + udf_crc(fibh->ebh->b_data + sizeof(struct fileIdentDesc) + + fibh->soffset, + crclen + sizeof(tag) - sizeof(struct fileIdentDesc), + crc); + else { + crc = udf_crc((uint8_t *) sfi->impUse, + -fibh->soffset - sizeof(struct fileIdentDesc), + crc); crc = udf_crc(fibh->ebh->b_data, fibh->eoffset, crc); } cfi->descTag.descCRC = cpu_to_le16(crc); cfi->descTag.descCRCLength = cpu_to_le16(crclen); - for (i=0; i<16; i++) + for (i = 0; i < 16; i++) if (i != 4) - checksum += ((uint8_t *)&cfi->descTag)[i]; + checksum += ((uint8_t *) & cfi->descTag)[i]; cfi->descTag.tagChecksum = checksum; if (adinicb || (sizeof(struct fileIdentDesc) <= -fibh->soffset)) - memcpy((uint8_t *)sfi, (uint8_t *)cfi, sizeof(struct fileIdentDesc)); - else - { - memcpy((uint8_t *)sfi, (uint8_t *)cfi, -fibh->soffset); - memcpy(fibh->ebh->b_data, (uint8_t *)cfi - fibh->soffset, - sizeof(struct fileIdentDesc) + fibh->soffset); + memcpy((uint8_t *) sfi, (uint8_t *) cfi, + sizeof(struct fileIdentDesc)); + else { + memcpy((uint8_t *) sfi, (uint8_t *) cfi, -fibh->soffset); + memcpy(fibh->ebh->b_data, (uint8_t *) cfi - fibh->soffset, + sizeof(struct fileIdentDesc) + fibh->soffset); } if (adinicb) mark_inode_dirty(inode); - else - { + else { if (fibh->sbh != fibh->ebh) mark_buffer_dirty_inode(fibh->ebh, inode); mark_buffer_dirty_inode(fibh->sbh, inode); @@ -143,12 +147,12 @@ int udf_write_fi(struct inode *inode, struct fileIdentDesc *cfi, return 0; } -static struct fileIdentDesc * -udf_find_entry(struct inode *dir, struct dentry *dentry, - struct udf_fileident_bh *fibh, - struct fileIdentDesc *cfi) +static struct fileIdentDesc *udf_find_entry(struct inode *dir, + struct dentry *dentry, + struct udf_fileident_bh *fibh, + struct fileIdentDesc *cfi) { - struct fileIdentDesc *fi=NULL; + struct fileIdentDesc *fi = NULL; loff_t f_pos; int block, flen; char fname[UDF_NAME_LEN]; @@ -159,46 +163,41 @@ udf_find_entry(struct inode *dir, struct dentry *dentry, kernel_lb_addr eloc; uint32_t elen; sector_t offset; - struct extent_position epos = { NULL, 0, { 0, 0}}; + struct extent_position epos = { NULL, 0, {0, 0} }; size = (udf_ext0_offset(dir) + dir->i_size) >> 2; f_pos = (udf_ext0_offset(dir) >> 2); - fibh->soffset = fibh->eoffset = (f_pos & ((dir->i_sb->s_blocksize - 1) >> 2)) << 2; + fibh->soffset = fibh->eoffset = + (f_pos & ((dir->i_sb->s_blocksize - 1) >> 2)) << 2; if (UDF_I_ALLOCTYPE(dir) == ICBTAG_FLAG_AD_IN_ICB) fibh->sbh = fibh->ebh = NULL; else if (inode_bmap(dir, f_pos >> (dir->i_sb->s_blocksize_bits - 2), - &epos, &eloc, &elen, &offset) == (EXT_RECORDED_ALLOCATED >> 30)) - { + &epos, &eloc, &elen, + &offset) == (EXT_RECORDED_ALLOCATED >> 30)) { block = udf_get_lb_pblock(dir->i_sb, eloc, offset); - if ((++offset << dir->i_sb->s_blocksize_bits) < elen) - { + if ((++offset << dir->i_sb->s_blocksize_bits) < elen) { if (UDF_I_ALLOCTYPE(dir) == ICBTAG_FLAG_AD_SHORT) epos.offset -= sizeof(short_ad); else if (UDF_I_ALLOCTYPE(dir) == ICBTAG_FLAG_AD_LONG) epos.offset -= sizeof(long_ad); - } - else + } else offset = 0; - if (!(fibh->sbh = fibh->ebh = udf_tread(dir->i_sb, block))) - { + if (!(fibh->sbh = fibh->ebh = udf_tread(dir->i_sb, block))) { brelse(epos.bh); return NULL; } - } - else - { + } else { brelse(epos.bh); return NULL; } - while ( (f_pos < size) ) - { - fi = udf_fileident_read(dir, &f_pos, fibh, cfi, &epos, &eloc, &elen, &offset); + while ((f_pos < size)) { + fi = udf_fileident_read(dir, &f_pos, fibh, cfi, &epos, &eloc, + &elen, &offset); - if (!fi) - { + if (!fi) { if (fibh->sbh != fibh->ebh) brelse(fibh->ebh); brelse(fibh->sbh); @@ -209,45 +208,45 @@ udf_find_entry(struct inode *dir, struct dentry *dentry, liu = le16_to_cpu(cfi->lengthOfImpUse); lfi = cfi->lengthFileIdent; - if (fibh->sbh == fibh->ebh) - { + if (fibh->sbh == fibh->ebh) { nameptr = fi->fileIdent + liu; - } - else - { + } else { int poffset; /* Unpaded ending offset */ - poffset = fibh->soffset + sizeof(struct fileIdentDesc) + liu + lfi; + poffset = + fibh->soffset + sizeof(struct fileIdentDesc) + liu + + lfi; if (poffset >= lfi) - nameptr = (uint8_t *)(fibh->ebh->b_data + poffset - lfi); - else - { + nameptr = + (uint8_t *) (fibh->ebh->b_data + poffset - + lfi); + else { nameptr = fname; - memcpy(nameptr, fi->fileIdent + liu, lfi - poffset); - memcpy(nameptr + lfi - poffset, fibh->ebh->b_data, poffset); + memcpy(nameptr, fi->fileIdent + liu, + lfi - poffset); + memcpy(nameptr + lfi - poffset, + fibh->ebh->b_data, poffset); } } - if ( (cfi->fileCharacteristics & FID_FILE_CHAR_DELETED) != 0 ) - { - if ( !UDF_QUERY_FLAG(dir->i_sb, UDF_FLAG_UNDELETE) ) + if ((cfi->fileCharacteristics & FID_FILE_CHAR_DELETED) != 0) { + if (!UDF_QUERY_FLAG(dir->i_sb, UDF_FLAG_UNDELETE)) continue; } - - if ( (cfi->fileCharacteristics & FID_FILE_CHAR_HIDDEN) != 0 ) - { - if ( !UDF_QUERY_FLAG(dir->i_sb, UDF_FLAG_UNHIDE) ) + + if ((cfi->fileCharacteristics & FID_FILE_CHAR_HIDDEN) != 0) { + if (!UDF_QUERY_FLAG(dir->i_sb, UDF_FLAG_UNHIDE)) continue; } if (!lfi) continue; - if ((flen = udf_get_filename(dir->i_sb, nameptr, fname, lfi))) - { - if (udf_match(flen, fname, dentry->d_name.len, dentry->d_name.name)) - { + if ((flen = udf_get_filename(dir->i_sb, nameptr, fname, lfi))) { + if (udf_match + (flen, fname, dentry->d_name.len, + dentry->d_name.name)) { brelse(epos.bh); return fi; } @@ -293,41 +292,37 @@ udf_find_entry(struct inode *dir, struct dentry *dentry, * Written, tested, and released. */ -static struct dentry * -udf_lookup(struct inode *dir, struct dentry *dentry, struct nameidata *nd) +static struct dentry *udf_lookup(struct inode *dir, struct dentry *dentry, + struct nameidata *nd) { struct inode *inode = NULL; struct fileIdentDesc cfi; struct udf_fileident_bh fibh; - if (dentry->d_name.len > UDF_NAME_LEN-2) + if (dentry->d_name.len > UDF_NAME_LEN - 2) return ERR_PTR(-ENAMETOOLONG); lock_kernel(); #ifdef UDF_RECOVERY /* temporary shorthand for specifying files by inode number */ - if (!strncmp(dentry->d_name.name, ".B=", 3) ) - { - kernel_lb_addr lb = { 0, simple_strtoul(dentry->d_name.name+3, NULL, 0) }; + if (!strncmp(dentry->d_name.name, ".B=", 3)) { + kernel_lb_addr lb = + { 0, simple_strtoul(dentry->d_name.name + 3, NULL, 0) }; inode = udf_iget(dir->i_sb, lb); - if (!inode) - { + if (!inode) { unlock_kernel(); return ERR_PTR(-EACCES); } - } - else -#endif /* UDF_RECOVERY */ + } else +#endif /* UDF_RECOVERY */ - if (udf_find_entry(dir, dentry, &fibh, &cfi)) - { + if (udf_find_entry(dir, dentry, &fibh, &cfi)) { if (fibh.sbh != fibh.ebh) brelse(fibh.ebh); brelse(fibh.sbh); inode = udf_iget(dir->i_sb, lelb_to_cpu(cfi.icb.extLocation)); - if ( !inode ) - { + if (!inode) { unlock_kernel(); return ERR_PTR(-EACCES); } @@ -337,13 +332,13 @@ udf_lookup(struct inode *dir, struct dentry *dentry, struct nameidata *nd) return NULL; } -static struct fileIdentDesc * -udf_add_entry(struct inode *dir, struct dentry *dentry, - struct udf_fileident_bh *fibh, - struct fileIdentDesc *cfi, int *err) +static struct fileIdentDesc *udf_add_entry(struct inode *dir, + struct dentry *dentry, + struct udf_fileident_bh *fibh, + struct fileIdentDesc *cfi, int *err) { struct super_block *sb; - struct fileIdentDesc *fi=NULL; + struct fileIdentDesc *fi = NULL; char name[UDF_NAME_LEN], fname[UDF_NAME_LEN]; int namelen; loff_t f_pos; @@ -357,50 +352,47 @@ udf_add_entry(struct inode *dir, struct dentry *dentry, kernel_lb_addr eloc; uint32_t elen; sector_t offset; - struct extent_position epos = { NULL, 0, { 0, 0 }}; + struct extent_position epos = { NULL, 0, {0, 0} }; sb = dir->i_sb; - if (dentry) - { - if (!dentry->d_name.len) - { + if (dentry) { + if (!dentry->d_name.len) { *err = -EINVAL; return NULL; } - if ( !(namelen = udf_put_filename(sb, dentry->d_name.name, name, dentry->d_name.len))) - { + if (! + (namelen = + udf_put_filename(sb, dentry->d_name.name, name, + dentry->d_name.len))) { *err = -ENAMETOOLONG; return NULL; } - } - else + } else namelen = 0; nfidlen = (sizeof(struct fileIdentDesc) + namelen + 3) & ~3; f_pos = (udf_ext0_offset(dir) >> 2); - fibh->soffset = fibh->eoffset = (f_pos & ((dir->i_sb->s_blocksize - 1) >> 2)) << 2; + fibh->soffset = fibh->eoffset = + (f_pos & ((dir->i_sb->s_blocksize - 1) >> 2)) << 2; if (UDF_I_ALLOCTYPE(dir) == ICBTAG_FLAG_AD_IN_ICB) fibh->sbh = fibh->ebh = NULL; else if (inode_bmap(dir, f_pos >> (dir->i_sb->s_blocksize_bits - 2), - &epos, &eloc, &elen, &offset) == (EXT_RECORDED_ALLOCATED >> 30)) - { + &epos, &eloc, &elen, + &offset) == (EXT_RECORDED_ALLOCATED >> 30)) { block = udf_get_lb_pblock(dir->i_sb, eloc, offset); - if ((++offset << dir->i_sb->s_blocksize_bits) < elen) - { + if ((++offset << dir->i_sb->s_blocksize_bits) < elen) { if (UDF_I_ALLOCTYPE(dir) == ICBTAG_FLAG_AD_SHORT) epos.offset -= sizeof(short_ad); else if (UDF_I_ALLOCTYPE(dir) == ICBTAG_FLAG_AD_LONG) epos.offset -= sizeof(long_ad); - } - else + } else offset = 0; - if (!(fibh->sbh = fibh->ebh = udf_tread(dir->i_sb, block))) - { + if (!(fibh->sbh = fibh->ebh = udf_tread(dir->i_sb, block))) { brelse(epos.bh); *err = -EIO; return NULL; @@ -408,21 +400,18 @@ udf_add_entry(struct inode *dir, struct dentry *dentry, block = UDF_I_LOCATION(dir).logicalBlockNum; - } - else - { + } else { block = udf_get_lb_pblock(dir->i_sb, UDF_I_LOCATION(dir), 0); fibh->sbh = fibh->ebh = NULL; fibh->soffset = fibh->eoffset = sb->s_blocksize; goto add; } - while ( (f_pos < size) ) - { - fi = udf_fileident_read(dir, &f_pos, fibh, cfi, &epos, &eloc, &elen, &offset); + while ((f_pos < size)) { + fi = udf_fileident_read(dir, &f_pos, fibh, cfi, &epos, &eloc, + &elen, &offset); - if (!fi) - { + if (!fi) { if (fibh->sbh != fibh->ebh) brelse(fibh->ebh); brelse(fibh->sbh); @@ -436,36 +425,38 @@ udf_add_entry(struct inode *dir, struct dentry *dentry, if (fibh->sbh == fibh->ebh) nameptr = fi->fileIdent + liu; - else - { + else { int poffset; /* Unpaded ending offset */ - poffset = fibh->soffset + sizeof(struct fileIdentDesc) + liu + lfi; + poffset = + fibh->soffset + sizeof(struct fileIdentDesc) + liu + + lfi; if (poffset >= lfi) - nameptr = (char *)(fibh->ebh->b_data + poffset - lfi); - else - { + nameptr = + (char *)(fibh->ebh->b_data + poffset - lfi); + else { nameptr = fname; - memcpy(nameptr, fi->fileIdent + liu, lfi - poffset); - memcpy(nameptr + lfi - poffset, fibh->ebh->b_data, poffset); + memcpy(nameptr, fi->fileIdent + liu, + lfi - poffset); + memcpy(nameptr + lfi - poffset, + fibh->ebh->b_data, poffset); } } - if ( (cfi->fileCharacteristics & FID_FILE_CHAR_DELETED) != 0 ) - { - if (((sizeof(struct fileIdentDesc) + liu + lfi + 3) & ~3) == nfidlen) - { + if ((cfi->fileCharacteristics & FID_FILE_CHAR_DELETED) != 0) { + if (((sizeof(struct fileIdentDesc) + liu + lfi + + 3) & ~3) == nfidlen) { brelse(epos.bh); cfi->descTag.tagSerialNum = cpu_to_le16(1); cfi->fileVersionNum = cpu_to_le16(1); cfi->fileCharacteristics = 0; cfi->lengthFileIdent = namelen; cfi->lengthOfImpUse = cpu_to_le16(0); - if (!udf_write_fi(dir, cfi, fi, fibh, NULL, name)) + if (!udf_write_fi + (dir, cfi, fi, fibh, NULL, name)) return fi; - else - { + else { *err = -EIO; return NULL; } @@ -476,8 +467,8 @@ udf_add_entry(struct inode *dir, struct dentry *dentry, continue; if ((flen = udf_get_filename(dir->i_sb, nameptr, fname, lfi)) && - udf_match(flen, fname, dentry->d_name.len, dentry->d_name.name)) - { + udf_match(flen, fname, dentry->d_name.len, + dentry->d_name.name)) { if (fibh->sbh != fibh->ebh) brelse(fibh->ebh); brelse(fibh->sbh); @@ -487,12 +478,11 @@ udf_add_entry(struct inode *dir, struct dentry *dentry, } } -add: + add: f_pos += nfidlen; if (UDF_I_ALLOCTYPE(dir) == ICBTAG_FLAG_AD_IN_ICB && - sb->s_blocksize - fibh->eoffset < nfidlen) - { + sb->s_blocksize - fibh->eoffset < nfidlen) { brelse(epos.bh); epos.bh = NULL; fibh->soffset -= udf_ext0_offset(dir); @@ -501,11 +491,14 @@ add: if (fibh->sbh != fibh->ebh) brelse(fibh->ebh); brelse(fibh->sbh); - if (!(fibh->sbh = fibh->ebh = udf_expand_dir_adinicb(dir, &block, err))) + if (! + (fibh->sbh = fibh->ebh = + udf_expand_dir_adinicb(dir, &block, err))) return NULL; epos.block = UDF_I_LOCATION(dir); eloc.logicalBlockNum = block; - eloc.partitionReferenceNum = UDF_I_LOCATION(dir).partitionReferenceNum; + eloc.partitionReferenceNum = + UDF_I_LOCATION(dir).partitionReferenceNum; elen = dir->i_sb->s_blocksize; epos.offset = udf_file_entry_alloc_offset(dir); if (UDF_I_ALLOCTYPE(dir) == ICBTAG_FLAG_AD_SHORT) @@ -514,89 +507,84 @@ add: epos.offset += sizeof(long_ad); } - if (sb->s_blocksize - fibh->eoffset >= nfidlen) - { + if (sb->s_blocksize - fibh->eoffset >= nfidlen) { fibh->soffset = fibh->eoffset; fibh->eoffset += nfidlen; - if (fibh->sbh != fibh->ebh) - { + if (fibh->sbh != fibh->ebh) { brelse(fibh->sbh); fibh->sbh = fibh->ebh; } - if (UDF_I_ALLOCTYPE(dir) == ICBTAG_FLAG_AD_IN_ICB) - { + if (UDF_I_ALLOCTYPE(dir) == ICBTAG_FLAG_AD_IN_ICB) { block = UDF_I_LOCATION(dir).logicalBlockNum; - fi = (struct fileIdentDesc *)(UDF_I_DATA(dir) + fibh->soffset - udf_ext0_offset(dir) + UDF_I_LENEATTR(dir)); - } - else - { + fi = (struct fileIdentDesc *)(UDF_I_DATA(dir) + + fibh->soffset - + udf_ext0_offset(dir) + + UDF_I_LENEATTR(dir)); + } else { block = eloc.logicalBlockNum + ((elen - 1) >> - dir->i_sb->s_blocksize_bits); - fi = (struct fileIdentDesc *)(fibh->sbh->b_data + fibh->soffset); + dir->i_sb-> + s_blocksize_bits); + fi = (struct fileIdentDesc *)(fibh->sbh->b_data + + fibh->soffset); } - } - else - { + } else { fibh->soffset = fibh->eoffset - sb->s_blocksize; fibh->eoffset += nfidlen - sb->s_blocksize; - if (fibh->sbh != fibh->ebh) - { + if (fibh->sbh != fibh->ebh) { brelse(fibh->sbh); fibh->sbh = fibh->ebh; } block = eloc.logicalBlockNum + ((elen - 1) >> - dir->i_sb->s_blocksize_bits); + dir->i_sb->s_blocksize_bits); - if (!(fibh->ebh = udf_bread(dir, f_pos >> (dir->i_sb->s_blocksize_bits - 2), 1, err))) - { + if (! + (fibh->ebh = + udf_bread(dir, f_pos >> (dir->i_sb->s_blocksize_bits - 2), + 1, err))) { brelse(epos.bh); brelse(fibh->sbh); return NULL; } - if (!(fibh->soffset)) - { + if (!(fibh->soffset)) { if (udf_next_aext(dir, &epos, &eloc, &elen, 1) == - (EXT_RECORDED_ALLOCATED >> 30)) - { + (EXT_RECORDED_ALLOCATED >> 30)) { block = eloc.logicalBlockNum + ((elen - 1) >> - dir->i_sb->s_blocksize_bits); - } - else - block ++; + dir->i_sb-> + s_blocksize_bits); + } else + block++; brelse(fibh->sbh); fibh->sbh = fibh->ebh; fi = (struct fileIdentDesc *)(fibh->sbh->b_data); - } - else - { + } else { fi = (struct fileIdentDesc *) - (fibh->sbh->b_data + sb->s_blocksize + fibh->soffset); + (fibh->sbh->b_data + sb->s_blocksize + + fibh->soffset); } } memset(cfi, 0, sizeof(struct fileIdentDesc)); if (UDF_SB_UDFREV(sb) >= 0x0200) - udf_new_tag((char *)cfi, TAG_IDENT_FID, 3, 1, block, sizeof(tag)); + udf_new_tag((char *)cfi, TAG_IDENT_FID, 3, 1, block, + sizeof(tag)); else - udf_new_tag((char *)cfi, TAG_IDENT_FID, 2, 1, block, sizeof(tag)); + udf_new_tag((char *)cfi, TAG_IDENT_FID, 2, 1, block, + sizeof(tag)); cfi->fileVersionNum = cpu_to_le16(1); cfi->lengthFileIdent = namelen; cfi->lengthOfImpUse = cpu_to_le16(0); - if (!udf_write_fi(dir, cfi, fi, fibh, NULL, name)) - { + if (!udf_write_fi(dir, cfi, fi, fibh, NULL, name)) { brelse(epos.bh); dir->i_size += nfidlen; if (UDF_I_ALLOCTYPE(dir) == ICBTAG_FLAG_AD_IN_ICB) UDF_I_LENALLOC(dir) += nfidlen; mark_inode_dirty(dir); return fi; - } - else - { + } else { brelse(epos.bh); if (fibh->sbh != fibh->ebh) brelse(fibh->ebh); @@ -607,7 +595,8 @@ add: } static int udf_delete_entry(struct inode *inode, struct fileIdentDesc *fi, - struct udf_fileident_bh *fibh, struct fileIdentDesc *cfi) + struct udf_fileident_bh *fibh, + struct fileIdentDesc *cfi) { cfi->fileCharacteristics |= FID_FILE_CHAR_DELETED; if (UDF_QUERY_FLAG(inode->i_sb, UDF_FLAG_STRICT)) @@ -615,7 +604,8 @@ static int udf_delete_entry(struct inode *inode, struct fileIdentDesc *fi, return udf_write_fi(inode, cfi, fi, fibh, NULL, NULL); } -static int udf_create(struct inode *dir, struct dentry *dentry, int mode, struct nameidata *nd) +static int udf_create(struct inode *dir, struct dentry *dentry, int mode, + struct nameidata *nd) { struct udf_fileident_bh fibh; struct inode *inode; @@ -624,8 +614,7 @@ static int udf_create(struct inode *dir, struct dentry *dentry, int mode, struct lock_kernel(); inode = udf_new_inode(dir, mode, &err); - if (!inode) - { + if (!inode) { unlock_kernel(); return err; } @@ -639,9 +628,8 @@ static int udf_create(struct inode *dir, struct dentry *dentry, int mode, struct inode->i_mode = mode; mark_inode_dirty(inode); - if (!(fi = udf_add_entry(dir, dentry, &fibh, &cfi, &err))) - { - inode->i_nlink --; + if (!(fi = udf_add_entry(dir, dentry, &fibh, &cfi, &err))) { + inode->i_nlink--; mark_inode_dirty(inode); iput(inode); unlock_kernel(); @@ -649,11 +637,10 @@ static int udf_create(struct inode *dir, struct dentry *dentry, int mode, struct } cfi.icb.extLength = cpu_to_le32(inode->i_sb->s_blocksize); cfi.icb.extLocation = cpu_to_lelb(UDF_I_LOCATION(inode)); - *(__le32 *)((struct allocDescImpUse *)cfi.icb.impUse)->impUse = - cpu_to_le32(UDF_I_UNIQUE(inode) & 0x00000000FFFFFFFFUL); + *(__le32 *) ((struct allocDescImpUse *)cfi.icb.impUse)->impUse = + cpu_to_le32(UDF_I_UNIQUE(inode) & 0x00000000FFFFFFFFUL); udf_write_fi(dir, &cfi, fi, &fibh, NULL, NULL); - if (UDF_I_ALLOCTYPE(dir) == ICBTAG_FLAG_AD_IN_ICB) - { + if (UDF_I_ALLOCTYPE(dir) == ICBTAG_FLAG_AD_IN_ICB) { mark_inode_dirty(dir); } if (fibh.sbh != fibh.ebh) @@ -664,9 +651,10 @@ static int udf_create(struct inode *dir, struct dentry *dentry, int mode, struct return 0; } -static int udf_mknod(struct inode * dir, struct dentry * dentry, int mode, dev_t rdev) +static int udf_mknod(struct inode *dir, struct dentry *dentry, int mode, + dev_t rdev) { - struct inode * inode; + struct inode *inode; struct udf_fileident_bh fibh; struct fileIdentDesc cfi, *fi; int err; @@ -682,9 +670,8 @@ static int udf_mknod(struct inode * dir, struct dentry * dentry, int mode, dev_t inode->i_uid = current->fsuid; init_special_inode(inode, mode, rdev); - if (!(fi = udf_add_entry(dir, dentry, &fibh, &cfi, &err))) - { - inode->i_nlink --; + if (!(fi = udf_add_entry(dir, dentry, &fibh, &cfi, &err))) { + inode->i_nlink--; mark_inode_dirty(inode); iput(inode); unlock_kernel(); @@ -692,11 +679,10 @@ static int udf_mknod(struct inode * dir, struct dentry * dentry, int mode, dev_t } cfi.icb.extLength = cpu_to_le32(inode->i_sb->s_blocksize); cfi.icb.extLocation = cpu_to_lelb(UDF_I_LOCATION(inode)); - *(__le32 *)((struct allocDescImpUse *)cfi.icb.impUse)->impUse = - cpu_to_le32(UDF_I_UNIQUE(inode) & 0x00000000FFFFFFFFUL); + *(__le32 *) ((struct allocDescImpUse *)cfi.icb.impUse)->impUse = + cpu_to_le32(UDF_I_UNIQUE(inode) & 0x00000000FFFFFFFFUL); udf_write_fi(dir, &cfi, fi, &fibh, NULL, NULL); - if (UDF_I_ALLOCTYPE(dir) == ICBTAG_FLAG_AD_IN_ICB) - { + if (UDF_I_ALLOCTYPE(dir) == ICBTAG_FLAG_AD_IN_ICB) { mark_inode_dirty(dir); } mark_inode_dirty(inode); @@ -706,21 +692,21 @@ static int udf_mknod(struct inode * dir, struct dentry * dentry, int mode, dev_t brelse(fibh.sbh); d_instantiate(dentry, inode); err = 0; -out: + out: unlock_kernel(); return err; } -static int udf_mkdir(struct inode * dir, struct dentry * dentry, int mode) +static int udf_mkdir(struct inode *dir, struct dentry *dentry, int mode) { - struct inode * inode; + struct inode *inode; struct udf_fileident_bh fibh; struct fileIdentDesc cfi, *fi; int err; lock_kernel(); err = -EMLINK; - if (dir->i_nlink >= (256<<sizeof(dir->i_nlink))-1) + if (dir->i_nlink >= (256 << sizeof(dir->i_nlink)) - 1) goto out; err = -EIO; @@ -730,8 +716,7 @@ static int udf_mkdir(struct inode * dir, struct dentry * dentry, int mode) inode->i_op = &udf_dir_inode_operations; inode->i_fop = &udf_dir_operations; - if (!(fi = udf_add_entry(inode, NULL, &fibh, &cfi, &err))) - { + if (!(fi = udf_add_entry(inode, NULL, &fibh, &cfi, &err))) { inode->i_nlink--; mark_inode_dirty(inode); iput(inode); @@ -740,9 +725,10 @@ static int udf_mkdir(struct inode * dir, struct dentry * dentry, int mode) inode->i_nlink = 2; cfi.icb.extLength = cpu_to_le32(inode->i_sb->s_blocksize); cfi.icb.extLocation = cpu_to_lelb(UDF_I_LOCATION(dir)); - *(__le32 *)((struct allocDescImpUse *)cfi.icb.impUse)->impUse = - cpu_to_le32(UDF_I_UNIQUE(dir) & 0x00000000FFFFFFFFUL); - cfi.fileCharacteristics = FID_FILE_CHAR_DIRECTORY | FID_FILE_CHAR_PARENT; + *(__le32 *) ((struct allocDescImpUse *)cfi.icb.impUse)->impUse = + cpu_to_le32(UDF_I_UNIQUE(dir) & 0x00000000FFFFFFFFUL); + cfi.fileCharacteristics = + FID_FILE_CHAR_DIRECTORY | FID_FILE_CHAR_PARENT; udf_write_fi(inode, &cfi, fi, &fibh, NULL, NULL); brelse(fibh.sbh); inode->i_mode = S_IFDIR | mode; @@ -750,8 +736,7 @@ static int udf_mkdir(struct inode * dir, struct dentry * dentry, int mode) inode->i_mode |= S_ISGID; mark_inode_dirty(inode); - if (!(fi = udf_add_entry(dir, dentry, &fibh, &cfi, &err))) - { + if (!(fi = udf_add_entry(dir, dentry, &fibh, &cfi, &err))) { inode->i_nlink = 0; mark_inode_dirty(inode); iput(inode); @@ -759,8 +744,8 @@ static int udf_mkdir(struct inode * dir, struct dentry * dentry, int mode) } cfi.icb.extLength = cpu_to_le32(inode->i_sb->s_blocksize); cfi.icb.extLocation = cpu_to_lelb(UDF_I_LOCATION(inode)); - *(__le32 *)((struct allocDescImpUse *)cfi.icb.impUse)->impUse = - cpu_to_le32(UDF_I_UNIQUE(inode) & 0x00000000FFFFFFFFUL); + *(__le32 *) ((struct allocDescImpUse *)cfi.icb.impUse)->impUse = + cpu_to_le32(UDF_I_UNIQUE(inode) & 0x00000000FFFFFFFFUL); cfi.fileCharacteristics |= FID_FILE_CHAR_DIRECTORY; udf_write_fi(dir, &cfi, fi, &fibh, NULL, NULL); inc_nlink(dir); @@ -770,7 +755,7 @@ static int udf_mkdir(struct inode * dir, struct dentry * dentry, int mode) brelse(fibh.ebh); brelse(fibh.sbh); err = 0; -out: + out: unlock_kernel(); return err; } @@ -785,47 +770,41 @@ static int empty_dir(struct inode *dir) kernel_lb_addr eloc; uint32_t elen; sector_t offset; - struct extent_position epos = { NULL, 0, { 0, 0}}; + struct extent_position epos = { NULL, 0, {0, 0} }; f_pos = (udf_ext0_offset(dir) >> 2); - fibh.soffset = fibh.eoffset = (f_pos & ((dir->i_sb->s_blocksize - 1) >> 2)) << 2; + fibh.soffset = fibh.eoffset = + (f_pos & ((dir->i_sb->s_blocksize - 1) >> 2)) << 2; if (UDF_I_ALLOCTYPE(dir) == ICBTAG_FLAG_AD_IN_ICB) fibh.sbh = fibh.ebh = NULL; else if (inode_bmap(dir, f_pos >> (dir->i_sb->s_blocksize_bits - 2), - &epos, &eloc, &elen, &offset) == (EXT_RECORDED_ALLOCATED >> 30)) - { + &epos, &eloc, &elen, + &offset) == (EXT_RECORDED_ALLOCATED >> 30)) { block = udf_get_lb_pblock(dir->i_sb, eloc, offset); - if ((++offset << dir->i_sb->s_blocksize_bits) < elen) - { + if ((++offset << dir->i_sb->s_blocksize_bits) < elen) { if (UDF_I_ALLOCTYPE(dir) == ICBTAG_FLAG_AD_SHORT) epos.offset -= sizeof(short_ad); else if (UDF_I_ALLOCTYPE(dir) == ICBTAG_FLAG_AD_LONG) epos.offset -= sizeof(long_ad); - } - else + } else offset = 0; - if (!(fibh.sbh = fibh.ebh = udf_tread(dir->i_sb, block))) - { + if (!(fibh.sbh = fibh.ebh = udf_tread(dir->i_sb, block))) { brelse(epos.bh); return 0; } - } - else - { + } else { brelse(epos.bh); return 0; } + while ((f_pos < size)) { + fi = udf_fileident_read(dir, &f_pos, &fibh, &cfi, &epos, &eloc, + &elen, &offset); - while ( (f_pos < size) ) - { - fi = udf_fileident_read(dir, &f_pos, &fibh, &cfi, &epos, &eloc, &elen, &offset); - - if (!fi) - { + if (!fi) { if (fibh.sbh != fibh.ebh) brelse(fibh.ebh); brelse(fibh.sbh); @@ -833,8 +812,8 @@ static int empty_dir(struct inode *dir) return 0; } - if (cfi.lengthFileIdent && (cfi.fileCharacteristics & FID_FILE_CHAR_DELETED) == 0) - { + if (cfi.lengthFileIdent + && (cfi.fileCharacteristics & FID_FILE_CHAR_DELETED) == 0) { if (fibh.sbh != fibh.ebh) brelse(fibh.ebh); brelse(fibh.sbh); @@ -849,10 +828,10 @@ static int empty_dir(struct inode *dir) return 1; } -static int udf_rmdir(struct inode * dir, struct dentry * dentry) +static int udf_rmdir(struct inode *dir, struct dentry *dentry) { int retval; - struct inode * inode = dentry->d_inode; + struct inode *inode = dentry->d_inode; struct udf_fileident_bh fibh; struct fileIdentDesc *fi, cfi; kernel_lb_addr tloc; @@ -875,27 +854,28 @@ static int udf_rmdir(struct inode * dir, struct dentry * dentry) goto end_rmdir; if (inode->i_nlink != 2) udf_warning(inode->i_sb, "udf_rmdir", - "empty directory has nlink != 2 (%d)", - inode->i_nlink); + "empty directory has nlink != 2 (%d)", + inode->i_nlink); clear_nlink(inode); inode->i_size = 0; inode_dec_link_count(dir); - inode->i_ctime = dir->i_ctime = dir->i_mtime = current_fs_time(dir->i_sb); + inode->i_ctime = dir->i_ctime = dir->i_mtime = + current_fs_time(dir->i_sb); mark_inode_dirty(dir); -end_rmdir: + end_rmdir: if (fibh.sbh != fibh.ebh) brelse(fibh.ebh); brelse(fibh.sbh); -out: + out: unlock_kernel(); return retval; } -static int udf_unlink(struct inode * dir, struct dentry * dentry) +static int udf_unlink(struct inode *dir, struct dentry *dentry) { int retval; - struct inode * inode = dentry->d_inode; + struct inode *inode = dentry->d_inode; struct udf_fileident_bh fibh; struct fileIdentDesc *fi; struct fileIdentDesc cfi; @@ -912,10 +892,9 @@ static int udf_unlink(struct inode * dir, struct dentry * dentry) if (udf_get_lb_pblock(dir->i_sb, tloc, 0) != inode->i_ino) goto end_unlink; - if (!inode->i_nlink) - { + if (!inode->i_nlink) { udf_debug("Deleting nonexistent file (%lu), %d\n", - inode->i_ino, inode->i_nlink); + inode->i_ino, inode->i_nlink); inode->i_nlink = 1; } retval = udf_delete_entry(dir, fi, &fibh, &cfi); @@ -927,22 +906,23 @@ static int udf_unlink(struct inode * dir, struct dentry * dentry) inode->i_ctime = dir->i_ctime; retval = 0; -end_unlink: + end_unlink: if (fibh.sbh != fibh.ebh) brelse(fibh.ebh); brelse(fibh.sbh); -out: + out: unlock_kernel(); return retval; } -static int udf_symlink(struct inode * dir, struct dentry * dentry, const char * symname) +static int udf_symlink(struct inode *dir, struct dentry *dentry, + const char *symname) { - struct inode * inode; + struct inode *inode; struct pathComponent *pc; char *compstart; struct udf_fileident_bh fibh; - struct extent_position epos = { NULL, 0, {0, 0}}; + struct extent_position epos = { NULL, 0, {0, 0} }; int eoffset, elen = 0; struct fileIdentDesc *fi; struct fileIdentDesc cfi; @@ -960,28 +940,31 @@ static int udf_symlink(struct inode * dir, struct dentry * dentry, const char * inode->i_data.a_ops = &udf_symlink_aops; inode->i_op = &page_symlink_inode_operations; - if (UDF_I_ALLOCTYPE(inode) != ICBTAG_FLAG_AD_IN_ICB) - { + if (UDF_I_ALLOCTYPE(inode) != ICBTAG_FLAG_AD_IN_ICB) { kernel_lb_addr eloc; uint32_t elen; block = udf_new_block(inode->i_sb, inode, - UDF_I_LOCATION(inode).partitionReferenceNum, - UDF_I_LOCATION(inode).logicalBlockNum, &err); + UDF_I_LOCATION(inode). + partitionReferenceNum, + UDF_I_LOCATION(inode).logicalBlockNum, + &err); if (!block) goto out_no_entry; epos.block = UDF_I_LOCATION(inode); epos.offset = udf_file_entry_alloc_offset(inode); epos.bh = NULL; eloc.logicalBlockNum = block; - eloc.partitionReferenceNum = UDF_I_LOCATION(inode).partitionReferenceNum; + eloc.partitionReferenceNum = + UDF_I_LOCATION(inode).partitionReferenceNum; elen = inode->i_sb->s_blocksize; UDF_I_LENEXTENTS(inode) = elen; udf_add_aext(inode, &epos, eloc, elen, 0); brelse(epos.bh); block = udf_get_pblock(inode->i_sb, block, - UDF_I_LOCATION(inode).partitionReferenceNum, 0); + UDF_I_LOCATION(inode). + partitionReferenceNum, 0); epos.bh = udf_tread(inode->i_sb, block); lock_buffer(epos.bh); memset(epos.bh->b_data, 0x00, inode->i_sb->s_blocksize); @@ -989,17 +972,14 @@ static int udf_symlink(struct inode * dir, struct dentry * dentry, const char * unlock_buffer(epos.bh); mark_buffer_dirty_inode(epos.bh, inode); ea = epos.bh->b_data + udf_ext0_offset(inode); - } - else + } else ea = UDF_I_DATA(inode) + UDF_I_LENEATTR(inode); eoffset = inode->i_sb->s_blocksize - udf_ext0_offset(inode); pc = (struct pathComponent *)ea; - if (*symname == '/') - { - do - { + if (*symname == '/') { + do { symname++; } while (*symname == '/'); @@ -1012,8 +992,7 @@ static int udf_symlink(struct inode * dir, struct dentry * dentry, const char * err = -ENAMETOOLONG; - while (*symname) - { + while (*symname) { if (elen + sizeof(struct pathComponent) > eoffset) goto out_no_entry; @@ -1021,28 +1000,30 @@ static int udf_symlink(struct inode * dir, struct dentry * dentry, const char * compstart = (char *)symname; - do - { + do { symname++; } while (*symname && *symname != '/'); pc->componentType = 5; pc->lengthComponentIdent = 0; pc->componentFileVersionNum = 0; - if (compstart[0] == '.') - { - if ((symname-compstart) == 1) + if (compstart[0] == '.') { + if ((symname - compstart) == 1) pc->componentType = 4; - else if ((symname-compstart) == 2 && compstart[1] == '.') + else if ((symname - compstart) == 2 + && compstart[1] == '.') pc->componentType = 3; } - if (pc->componentType == 5) - { - if ( !(namelen = udf_put_filename(inode->i_sb, compstart, name, symname-compstart))) + if (pc->componentType == 5) { + if (! + (namelen = + udf_put_filename(inode->i_sb, compstart, name, + symname - compstart))) goto out_no_entry; - if (elen + sizeof(struct pathComponent) + namelen > eoffset) + if (elen + sizeof(struct pathComponent) + namelen > + eoffset) goto out_no_entry; else pc->lengthComponentIdent = namelen; @@ -1052,10 +1033,8 @@ static int udf_symlink(struct inode * dir, struct dentry * dentry, const char * elen += sizeof(struct pathComponent) + pc->lengthComponentIdent; - if (*symname) - { - do - { + if (*symname) { + do { symname++; } while (*symname == '/'); } @@ -1071,22 +1050,22 @@ static int udf_symlink(struct inode * dir, struct dentry * dentry, const char * goto out_no_entry; cfi.icb.extLength = cpu_to_le32(inode->i_sb->s_blocksize); cfi.icb.extLocation = cpu_to_lelb(UDF_I_LOCATION(inode)); - if (UDF_SB_LVIDBH(inode->i_sb)) - { + if (UDF_SB_LVIDBH(inode->i_sb)) { struct logicalVolHeaderDesc *lvhd; uint64_t uniqueID; - lvhd = (struct logicalVolHeaderDesc *)(UDF_SB_LVID(inode->i_sb)->logicalVolContentsUse); + lvhd = + (struct logicalVolHeaderDesc *)(UDF_SB_LVID(inode->i_sb)-> + logicalVolContentsUse); uniqueID = le64_to_cpu(lvhd->uniqueID); - *(__le32 *)((struct allocDescImpUse *)cfi.icb.impUse)->impUse = - cpu_to_le32(uniqueID & 0x00000000FFFFFFFFUL); + *(__le32 *) ((struct allocDescImpUse *)cfi.icb.impUse)->impUse = + cpu_to_le32(uniqueID & 0x00000000FFFFFFFFUL); if (!(++uniqueID & 0x00000000FFFFFFFFUL)) uniqueID += 16; lvhd->uniqueID = cpu_to_le64(uniqueID); mark_buffer_dirty(UDF_SB_LVIDBH(inode->i_sb)); } udf_write_fi(dir, &cfi, fi, &fibh, NULL, NULL); - if (UDF_I_ALLOCTYPE(dir) == ICBTAG_FLAG_AD_IN_ICB) - { + if (UDF_I_ALLOCTYPE(dir) == ICBTAG_FLAG_AD_IN_ICB) { mark_inode_dirty(dir); } if (fibh.sbh != fibh.ebh) @@ -1095,18 +1074,18 @@ static int udf_symlink(struct inode * dir, struct dentry * dentry, const char * d_instantiate(dentry, inode); err = 0; -out: + out: unlock_kernel(); return err; -out_no_entry: + out_no_entry: inode_dec_link_count(inode); iput(inode); goto out; } -static int udf_link(struct dentry * old_dentry, struct inode * dir, - struct dentry *dentry) +static int udf_link(struct dentry *old_dentry, struct inode *dir, + struct dentry *dentry) { struct inode *inode = old_dentry->d_inode; struct udf_fileident_bh fibh; @@ -1114,35 +1093,33 @@ static int udf_link(struct dentry * old_dentry, struct inode * dir, int err; lock_kernel(); - if (inode->i_nlink >= (256<<sizeof(inode->i_nlink))-1) - { + if (inode->i_nlink >= (256 << sizeof(inode->i_nlink)) - 1) { unlock_kernel(); return -EMLINK; } - if (!(fi = udf_add_entry(dir, dentry, &fibh, &cfi, &err))) - { + if (!(fi = udf_add_entry(dir, dentry, &fibh, &cfi, &err))) { unlock_kernel(); return err; } cfi.icb.extLength = cpu_to_le32(inode->i_sb->s_blocksize); cfi.icb.extLocation = cpu_to_lelb(UDF_I_LOCATION(inode)); - if (UDF_SB_LVIDBH(inode->i_sb)) - { + if (UDF_SB_LVIDBH(inode->i_sb)) { struct logicalVolHeaderDesc *lvhd; uint64_t uniqueID; - lvhd = (struct logicalVolHeaderDesc *)(UDF_SB_LVID(inode->i_sb)->logicalVolContentsUse); + lvhd = + (struct logicalVolHeaderDesc *)(UDF_SB_LVID(inode->i_sb)-> + logicalVolContentsUse); uniqueID = le64_to_cpu(lvhd->uniqueID); - *(__le32 *)((struct allocDescImpUse *)cfi.icb.impUse)->impUse = - cpu_to_le32(uniqueID & 0x00000000FFFFFFFFUL); + *(__le32 *) ((struct allocDescImpUse *)cfi.icb.impUse)->impUse = + cpu_to_le32(uniqueID & 0x00000000FFFFFFFFUL); if (!(++uniqueID & 0x00000000FFFFFFFFUL)) uniqueID += 16; lvhd->uniqueID = cpu_to_le64(uniqueID); mark_buffer_dirty(UDF_SB_LVIDBH(inode->i_sb)); } udf_write_fi(dir, &cfi, fi, &fibh, NULL, NULL); - if (UDF_I_ALLOCTYPE(dir) == ICBTAG_FLAG_AD_IN_ICB) - { + if (UDF_I_ALLOCTYPE(dir) == ICBTAG_FLAG_AD_IN_ICB) { mark_inode_dirty(dir); } if (fibh.sbh != fibh.ebh) @@ -1160,80 +1137,80 @@ static int udf_link(struct dentry * old_dentry, struct inode * dir, /* Anybody can rename anything with this: the permission checks are left to the * higher-level routines. */ -static int udf_rename (struct inode * old_dir, struct dentry * old_dentry, - struct inode * new_dir, struct dentry * new_dentry) +static int udf_rename(struct inode *old_dir, struct dentry *old_dentry, + struct inode *new_dir, struct dentry *new_dentry) { - struct inode * old_inode = old_dentry->d_inode; - struct inode * new_inode = new_dentry->d_inode; + struct inode *old_inode = old_dentry->d_inode; + struct inode *new_inode = new_dentry->d_inode; struct udf_fileident_bh ofibh, nfibh; - struct fileIdentDesc *ofi = NULL, *nfi = NULL, *dir_fi = NULL, ocfi, ncfi; + struct fileIdentDesc *ofi = NULL, *nfi = NULL, *dir_fi = + NULL, ocfi, ncfi; struct buffer_head *dir_bh = NULL; int retval = -ENOENT; kernel_lb_addr tloc; lock_kernel(); - if ((ofi = udf_find_entry(old_dir, old_dentry, &ofibh, &ocfi))) - { + if ((ofi = udf_find_entry(old_dir, old_dentry, &ofibh, &ocfi))) { if (ofibh.sbh != ofibh.ebh) brelse(ofibh.ebh); brelse(ofibh.sbh); } tloc = lelb_to_cpu(ocfi.icb.extLocation); if (!ofi || udf_get_lb_pblock(old_dir->i_sb, tloc, 0) - != old_inode->i_ino) + != old_inode->i_ino) goto end_rename; nfi = udf_find_entry(new_dir, new_dentry, &nfibh, &ncfi); - if (nfi) - { - if (!new_inode) - { + if (nfi) { + if (!new_inode) { if (nfibh.sbh != nfibh.ebh) brelse(nfibh.ebh); brelse(nfibh.sbh); nfi = NULL; } } - if (S_ISDIR(old_inode->i_mode)) - { + if (S_ISDIR(old_inode->i_mode)) { uint32_t offset = udf_ext0_offset(old_inode); - if (new_inode) - { + if (new_inode) { retval = -ENOTEMPTY; if (!empty_dir(new_inode)) goto end_rename; } retval = -EIO; - if (UDF_I_ALLOCTYPE(old_inode) == ICBTAG_FLAG_AD_IN_ICB) - { + if (UDF_I_ALLOCTYPE(old_inode) == ICBTAG_FLAG_AD_IN_ICB) { dir_fi = udf_get_fileident(UDF_I_DATA(old_inode) - - (UDF_I_EFE(old_inode) ? - sizeof(struct extendedFileEntry) : - sizeof(struct fileEntry)), - old_inode->i_sb->s_blocksize, &offset); - } - else - { + (UDF_I_EFE(old_inode) ? + sizeof(struct + extendedFileEntry) : + sizeof(struct fileEntry)), + old_inode->i_sb->s_blocksize, + &offset); + } else { dir_bh = udf_bread(old_inode, 0, 0, &retval); if (!dir_bh) goto end_rename; - dir_fi = udf_get_fileident(dir_bh->b_data, old_inode->i_sb->s_blocksize, &offset); + dir_fi = + udf_get_fileident(dir_bh->b_data, + old_inode->i_sb->s_blocksize, + &offset); } if (!dir_fi) goto end_rename; tloc = lelb_to_cpu(dir_fi->icb.extLocation); if (udf_get_lb_pblock(old_inode->i_sb, tloc, 0) - != old_dir->i_ino) + != old_dir->i_ino) goto end_rename; retval = -EMLINK; - if (!new_inode && new_dir->i_nlink >= (256<<sizeof(new_dir->i_nlink))-1) + if (!new_inode + && new_dir->i_nlink >= + (256 << sizeof(new_dir->i_nlink)) - 1) goto end_rename; } - if (!nfi) - { - nfi = udf_add_entry(new_dir, new_dentry, &nfibh, &ncfi, &retval); + if (!nfi) { + nfi = + udf_add_entry(new_dir, new_dentry, &nfibh, &ncfi, &retval); if (!nfi) goto end_rename; } @@ -1257,39 +1234,33 @@ static int udf_rename (struct inode * old_dir, struct dentry * old_dentry, ofi = udf_find_entry(old_dir, old_dentry, &ofibh, &ocfi); udf_delete_entry(old_dir, ofi, &ofibh, &ocfi); - if (new_inode) - { + if (new_inode) { new_inode->i_ctime = current_fs_time(new_inode->i_sb); inode_dec_link_count(new_inode); } old_dir->i_ctime = old_dir->i_mtime = current_fs_time(old_dir->i_sb); mark_inode_dirty(old_dir); - if (dir_fi) - { + if (dir_fi) { dir_fi->icb.extLocation = cpu_to_lelb(UDF_I_LOCATION(new_dir)); udf_update_tag((char *)dir_fi, (sizeof(struct fileIdentDesc) + - le16_to_cpu(dir_fi->lengthOfImpUse) + 3) & ~3); - if (UDF_I_ALLOCTYPE(old_inode) == ICBTAG_FLAG_AD_IN_ICB) - { + le16_to_cpu(dir_fi-> + lengthOfImpUse) + + 3) & ~3); + if (UDF_I_ALLOCTYPE(old_inode) == ICBTAG_FLAG_AD_IN_ICB) { mark_inode_dirty(old_inode); - } - else + } else mark_buffer_dirty_inode(dir_bh, old_inode); inode_dec_link_count(old_dir); - if (new_inode) - { + if (new_inode) { inode_dec_link_count(new_inode); - } - else - { + } else { inc_nlink(new_dir); mark_inode_dirty(new_dir); } } - if (ofi) - { + if (ofi) { if (ofibh.sbh != ofibh.ebh) brelse(ofibh.ebh); brelse(ofibh.sbh); @@ -1297,10 +1268,9 @@ static int udf_rename (struct inode * old_dir, struct dentry * old_dentry, retval = 0; -end_rename: + end_rename: brelse(dir_bh); - if (nfi) - { + if (nfi) { if (nfibh.sbh != nfibh.ebh) brelse(nfibh.ebh); brelse(nfibh.sbh); @@ -1310,13 +1280,13 @@ end_rename: } const struct inode_operations udf_dir_inode_operations = { - .lookup = udf_lookup, - .create = udf_create, - .link = udf_link, - .unlink = udf_unlink, - .symlink = udf_symlink, - .mkdir = udf_mkdir, - .rmdir = udf_rmdir, - .mknod = udf_mknod, - .rename = udf_rename, + .lookup = udf_lookup, + .create = udf_create, + .link = udf_link, + .unlink = udf_unlink, + .symlink = udf_symlink, + .mkdir = udf_mkdir, + .rmdir = udf_rmdir, + .mknod = udf_mknod, + .rename = udf_rename, }; diff --git a/fs/udf/osta_udf.h b/fs/udf/osta_udf.h index e82aae65269..bec5d340d8c 100644 --- a/fs/udf/osta_udf.h +++ b/fs/udf/osta_udf.h @@ -65,153 +65,140 @@ #define IS_DF_HARD_WRITE_PROTECT 0x01 #define IS_DF_SOFT_WRITE_PROTECT 0x02 -struct UDFIdentSuffix -{ - __le16 UDFRevision; - uint8_t OSClass; - uint8_t OSIdentifier; - uint8_t reserved[4]; +struct UDFIdentSuffix { + __le16 UDFRevision; + uint8_t OSClass; + uint8_t OSIdentifier; + uint8_t reserved[4]; } __attribute__ ((packed)); -struct impIdentSuffix -{ - uint8_t OSClass; - uint8_t OSIdentifier; - uint8_t reserved[6]; +struct impIdentSuffix { + uint8_t OSClass; + uint8_t OSIdentifier; + uint8_t reserved[6]; } __attribute__ ((packed)); -struct appIdentSuffix -{ - uint8_t impUse[8]; +struct appIdentSuffix { + uint8_t impUse[8]; } __attribute__ ((packed)); /* Logical Volume Integrity Descriptor (UDF 2.50 2.2.6) */ /* Implementation Use (UDF 2.50 2.2.6.4) */ -struct logicalVolIntegrityDescImpUse -{ - regid impIdent; - __le32 numFiles; - __le32 numDirs; - __le16 minUDFReadRev; - __le16 minUDFWriteRev; - __le16 maxUDFWriteRev; - uint8_t impUse[0]; +struct logicalVolIntegrityDescImpUse { + regid impIdent; + __le32 numFiles; + __le32 numDirs; + __le16 minUDFReadRev; + __le16 minUDFWriteRev; + __le16 maxUDFWriteRev; + uint8_t impUse[0]; } __attribute__ ((packed)); /* Implementation Use Volume Descriptor (UDF 2.50 2.2.7) */ /* Implementation Use (UDF 2.50 2.2.7.2) */ -struct impUseVolDescImpUse -{ - charspec LVICharset; - dstring logicalVolIdent[128]; - dstring LVInfo1[36]; - dstring LVInfo2[36]; - dstring LVInfo3[36]; - regid impIdent; - uint8_t impUse[128]; +struct impUseVolDescImpUse { + charspec LVICharset; + dstring logicalVolIdent[128]; + dstring LVInfo1[36]; + dstring LVInfo2[36]; + dstring LVInfo3[36]; + regid impIdent; + uint8_t impUse[128]; } __attribute__ ((packed)); -struct udfPartitionMap2 -{ - uint8_t partitionMapType; - uint8_t partitionMapLength; - uint8_t reserved1[2]; - regid partIdent; - __le16 volSeqNum; - __le16 partitionNum; +struct udfPartitionMap2 { + uint8_t partitionMapType; + uint8_t partitionMapLength; + uint8_t reserved1[2]; + regid partIdent; + __le16 volSeqNum; + __le16 partitionNum; } __attribute__ ((packed)); /* Virtual Partition Map (UDF 2.50 2.2.8) */ -struct virtualPartitionMap -{ - uint8_t partitionMapType; - uint8_t partitionMapLength; - uint8_t reserved1[2]; - regid partIdent; - __le16 volSeqNum; - __le16 partitionNum; - uint8_t reserved2[24]; +struct virtualPartitionMap { + uint8_t partitionMapType; + uint8_t partitionMapLength; + uint8_t reserved1[2]; + regid partIdent; + __le16 volSeqNum; + __le16 partitionNum; + uint8_t reserved2[24]; } __attribute__ ((packed)); /* Sparable Partition Map (UDF 2.50 2.2.9) */ -struct sparablePartitionMap -{ - uint8_t partitionMapType; - uint8_t partitionMapLength; - uint8_t reserved1[2]; - regid partIdent; - __le16 volSeqNum; - __le16 partitionNum; - __le16 packetLength; - uint8_t numSparingTables; - uint8_t reserved2[1]; - __le32 sizeSparingTable; - __le32 locSparingTable[4]; +struct sparablePartitionMap { + uint8_t partitionMapType; + uint8_t partitionMapLength; + uint8_t reserved1[2]; + regid partIdent; + __le16 volSeqNum; + __le16 partitionNum; + __le16 packetLength; + uint8_t numSparingTables; + uint8_t reserved2[1]; + __le32 sizeSparingTable; + __le32 locSparingTable[4]; } __attribute__ ((packed)); /* Metadata Partition Map (UDF 2.4.0 2.2.10) */ -struct metadataPartitionMap -{ - uint8_t partitionMapType; - uint8_t partitionMapLength; - uint8_t reserved1[2]; - regid partIdent; - __le16 volSeqNum; - __le16 partitionNum; - __le32 metadataFileLoc; - __le32 metadataMirrorFileLoc; - __le32 metadataBitmapFileLoc; - __le32 allocUnitSize; - __le16 alignUnitSize; - uint8_t flags; - uint8_t reserved2[5]; +struct metadataPartitionMap { + uint8_t partitionMapType; + uint8_t partitionMapLength; + uint8_t reserved1[2]; + regid partIdent; + __le16 volSeqNum; + __le16 partitionNum; + __le32 metadataFileLoc; + __le32 metadataMirrorFileLoc; + __le32 metadataBitmapFileLoc; + __le32 allocUnitSize; + __le16 alignUnitSize; + uint8_t flags; + uint8_t reserved2[5]; } __attribute__ ((packed)); /* Virtual Allocation Table (UDF 1.5 2.2.10) */ -struct virtualAllocationTable15 -{ - __le32 VirtualSector[0]; - regid vatIdent; - __le32 previousVATICBLoc; -} __attribute__ ((packed)); +struct virtualAllocationTable15 { + __le32 VirtualSector[0]; + regid vatIdent; + __le32 previousVATICBLoc; +} __attribute__ ((packed)); #define ICBTAG_FILE_TYPE_VAT15 0x00U /* Virtual Allocation Table (UDF 2.50 2.2.11) */ -struct virtualAllocationTable20 -{ - __le16 lengthHeader; - __le16 lengthImpUse; - dstring logicalVolIdent[128]; - __le32 previousVATICBLoc; - __le32 numFiles; - __le32 numDirs; - __le16 minReadRevision; - __le16 minWriteRevision; - __le16 maxWriteRevision; - __le16 reserved; - uint8_t impUse[0]; - __le32 vatEntry[0]; +struct virtualAllocationTable20 { + __le16 lengthHeader; + __le16 lengthImpUse; + dstring logicalVolIdent[128]; + __le32 previousVATICBLoc; + __le32 numFiles; + __le32 numDirs; + __le16 minReadRevision; + __le16 minWriteRevision; + __le16 maxWriteRevision; + __le16 reserved; + uint8_t impUse[0]; + __le32 vatEntry[0]; } __attribute__ ((packed)); #define ICBTAG_FILE_TYPE_VAT20 0xF8U /* Sparing Table (UDF 2.50 2.2.12) */ -struct sparingEntry -{ - __le32 origLocation; - __le32 mappedLocation; +struct sparingEntry { + __le32 origLocation; + __le32 mappedLocation; } __attribute__ ((packed)); -struct sparingTable -{ - tag descTag; - regid sparingIdent; - __le16 reallocationTableLen; - __le16 reserved; - __le32 sequenceNum; +struct sparingTable { + tag descTag; + regid sparingIdent; + __le16 reallocationTableLen; + __le16 reserved; + __le32 sequenceNum; struct sparingEntry - mapEntry[0]; + mapEntry[0]; } __attribute__ ((packed)); /* Metadata File (and Metadata Mirror File) (UDF 2.50 2.2.13.1) */ @@ -220,10 +207,9 @@ struct sparingTable #define ICBTAG_FILE_TYPE_BITMAP 0xFC /* struct long_ad ICB - ADImpUse (UDF 2.50 2.2.4.3) */ -struct allocDescImpUse -{ - __le16 flags; - uint8_t impUse[4]; +struct allocDescImpUse { + __le16 flags; + uint8_t impUse[4]; } __attribute__ ((packed)); #define AD_IU_EXT_ERASED 0x0001 @@ -233,27 +219,24 @@ struct allocDescImpUse /* Implementation Use Extended Attribute (UDF 2.50 3.3.4.5) */ /* FreeEASpace (UDF 2.50 3.3.4.5.1.1) */ -struct freeEaSpace -{ - __le16 headerChecksum; - uint8_t freeEASpace[0]; +struct freeEaSpace { + __le16 headerChecksum; + uint8_t freeEASpace[0]; } __attribute__ ((packed)); /* DVD Copyright Management Information (UDF 2.50 3.3.4.5.1.2) */ -struct DVDCopyrightImpUse -{ - __le16 headerChecksum; - uint8_t CGMSInfo; - uint8_t dataType; - uint8_t protectionSystemInfo[4]; +struct DVDCopyrightImpUse { + __le16 headerChecksum; + uint8_t CGMSInfo; + uint8_t dataType; + uint8_t protectionSystemInfo[4]; } __attribute__ ((packed)); /* Application Use Extended Attribute (UDF 2.50 3.3.4.6) */ /* FreeAppEASpace (UDF 2.50 3.3.4.6.1) */ -struct freeAppEASpace -{ - __le16 headerChecksum; - uint8_t freeEASpace[0]; +struct freeAppEASpace { + __le16 headerChecksum; + uint8_t freeEASpace[0]; } __attribute__ ((packed)); /* UDF Defined System Stream (UDF 2.50 3.3.7) */ @@ -293,4 +276,4 @@ struct freeAppEASpace #define UDF_OS_ID_BEOS 0x00U #define UDF_OS_ID_WINCE 0x00U -#endif /* _OSTA_UDF_H */ +#endif /* _OSTA_UDF_H */ diff --git a/fs/udf/partition.c b/fs/udf/partition.c index 467a26171cd..a95d830a674 100644 --- a/fs/udf/partition.c +++ b/fs/udf/partition.c @@ -28,106 +28,120 @@ #include <linux/slab.h> #include <linux/buffer_head.h> -inline uint32_t udf_get_pblock(struct super_block *sb, uint32_t block, uint16_t partition, uint32_t offset) +inline uint32_t udf_get_pblock(struct super_block *sb, uint32_t block, + uint16_t partition, uint32_t offset) { - if (partition >= UDF_SB_NUMPARTS(sb)) - { - udf_debug("block=%d, partition=%d, offset=%d: invalid partition\n", - block, partition, offset); + if (partition >= UDF_SB_NUMPARTS(sb)) { + udf_debug + ("block=%d, partition=%d, offset=%d: invalid partition\n", + block, partition, offset); return 0xFFFFFFFF; } if (UDF_SB_PARTFUNC(sb, partition)) - return UDF_SB_PARTFUNC(sb, partition)(sb, block, partition, offset); + return UDF_SB_PARTFUNC(sb, partition) (sb, block, partition, + offset); else return UDF_SB_PARTROOT(sb, partition) + block + offset; } -uint32_t udf_get_pblock_virt15(struct super_block *sb, uint32_t block, uint16_t partition, uint32_t offset) +uint32_t udf_get_pblock_virt15(struct super_block * sb, uint32_t block, + uint16_t partition, uint32_t offset) { struct buffer_head *bh = NULL; uint32_t newblock; uint32_t index; uint32_t loc; - index = (sb->s_blocksize - UDF_SB_TYPEVIRT(sb,partition).s_start_offset) / sizeof(uint32_t); + index = + (sb->s_blocksize - + UDF_SB_TYPEVIRT(sb, partition).s_start_offset) / sizeof(uint32_t); - if (block > UDF_SB_TYPEVIRT(sb,partition).s_num_entries) - { - udf_debug("Trying to access block beyond end of VAT (%d max %d)\n", - block, UDF_SB_TYPEVIRT(sb,partition).s_num_entries); + if (block > UDF_SB_TYPEVIRT(sb, partition).s_num_entries) { + udf_debug + ("Trying to access block beyond end of VAT (%d max %d)\n", + block, UDF_SB_TYPEVIRT(sb, partition).s_num_entries); return 0xFFFFFFFF; } - if (block >= index) - { + if (block >= index) { block -= index; newblock = 1 + (block / (sb->s_blocksize / sizeof(uint32_t))); index = block % (sb->s_blocksize / sizeof(uint32_t)); - } - else - { + } else { newblock = 0; - index = UDF_SB_TYPEVIRT(sb,partition).s_start_offset / sizeof(uint32_t) + block; + index = + UDF_SB_TYPEVIRT(sb, + partition).s_start_offset / + sizeof(uint32_t) + block; } loc = udf_block_map(UDF_SB_VAT(sb), newblock); - if (!(bh = sb_bread(sb, loc))) - { + if (!(bh = sb_bread(sb, loc))) { udf_debug("get_pblock(UDF_VIRTUAL_MAP:%p,%d,%d) VAT: %d[%d]\n", - sb, block, partition, loc, index); + sb, block, partition, loc, index); return 0xFFFFFFFF; } - loc = le32_to_cpu(((__le32 *)bh->b_data)[index]); + loc = le32_to_cpu(((__le32 *) bh->b_data)[index]); brelse(bh); - if (UDF_I_LOCATION(UDF_SB_VAT(sb)).partitionReferenceNum == partition) - { + if (UDF_I_LOCATION(UDF_SB_VAT(sb)).partitionReferenceNum == partition) { udf_debug("recursive call to udf_get_pblock!\n"); return 0xFFFFFFFF; } - return udf_get_pblock(sb, loc, UDF_I_LOCATION(UDF_SB_VAT(sb)).partitionReferenceNum, offset); + return udf_get_pblock(sb, loc, + UDF_I_LOCATION(UDF_SB_VAT(sb)). + partitionReferenceNum, offset); } -inline uint32_t udf_get_pblock_virt20(struct super_block *sb, uint32_t block, uint16_t partition, uint32_t offset) +inline uint32_t udf_get_pblock_virt20(struct super_block * sb, uint32_t block, + uint16_t partition, uint32_t offset) { return udf_get_pblock_virt15(sb, block, partition, offset); } -uint32_t udf_get_pblock_spar15(struct super_block *sb, uint32_t block, uint16_t partition, uint32_t offset) +uint32_t udf_get_pblock_spar15(struct super_block * sb, uint32_t block, + uint16_t partition, uint32_t offset) { int i; struct sparingTable *st = NULL; - uint32_t packet = (block + offset) & ~(UDF_SB_TYPESPAR(sb,partition).s_packet_len - 1); + uint32_t packet = + (block + offset) & ~(UDF_SB_TYPESPAR(sb, partition).s_packet_len - + 1); - for (i=0; i<4; i++) - { - if (UDF_SB_TYPESPAR(sb,partition).s_spar_map[i] != NULL) - { - st = (struct sparingTable *)UDF_SB_TYPESPAR(sb,partition).s_spar_map[i]->b_data; + for (i = 0; i < 4; i++) { + if (UDF_SB_TYPESPAR(sb, partition).s_spar_map[i] != NULL) { + st = (struct sparingTable *)UDF_SB_TYPESPAR(sb, + partition). + s_spar_map[i]->b_data; break; } } - if (st) - { - for (i=0; i<le16_to_cpu(st->reallocationTableLen); i++) - { - if (le32_to_cpu(st->mapEntry[i].origLocation) >= 0xFFFFFFF0) + if (st) { + for (i = 0; i < le16_to_cpu(st->reallocationTableLen); i++) { + if (le32_to_cpu(st->mapEntry[i].origLocation) >= + 0xFFFFFFF0) break; - else if (le32_to_cpu(st->mapEntry[i].origLocation) == packet) - { - return le32_to_cpu(st->mapEntry[i].mappedLocation) + - ((block + offset) & (UDF_SB_TYPESPAR(sb,partition).s_packet_len - 1)); - } - else if (le32_to_cpu(st->mapEntry[i].origLocation) > packet) + else if (le32_to_cpu(st->mapEntry[i].origLocation) == + packet) { + return le32_to_cpu(st->mapEntry[i]. + mappedLocation) + ((block + + offset) & + (UDF_SB_TYPESPAR + (sb, + partition). + s_packet_len + - 1)); + } else if (le32_to_cpu(st->mapEntry[i].origLocation) > + packet) break; } } - return UDF_SB_PARTROOT(sb,partition) + block + offset; + return UDF_SB_PARTROOT(sb, partition) + block + offset; } int udf_relocate_blocks(struct super_block *sb, long old_block, long *new_block) @@ -138,19 +152,21 @@ int udf_relocate_blocks(struct super_block *sb, long old_block, long *new_block) uint32_t packet; int i, j, k, l; - for (i=0; i<UDF_SB_NUMPARTS(sb); i++) - { - if (old_block > UDF_SB_PARTROOT(sb,i) && - old_block < UDF_SB_PARTROOT(sb,i) + UDF_SB_PARTLEN(sb,i)) + for (i = 0; i < UDF_SB_NUMPARTS(sb); i++) { + if (old_block > UDF_SB_PARTROOT(sb, i) && + old_block < UDF_SB_PARTROOT(sb, i) + UDF_SB_PARTLEN(sb, i)) { - sdata = &UDF_SB_TYPESPAR(sb,i); - packet = (old_block - UDF_SB_PARTROOT(sb,i)) & ~(sdata->s_packet_len - 1); - - for (j=0; j<4; j++) - { - if (UDF_SB_TYPESPAR(sb,i).s_spar_map[j] != NULL) - { - st = (struct sparingTable *)sdata->s_spar_map[j]->b_data; + sdata = &UDF_SB_TYPESPAR(sb, i); + packet = + (old_block - + UDF_SB_PARTROOT(sb, + i)) & ~(sdata->s_packet_len - 1); + + for (j = 0; j < 4; j++) { + if (UDF_SB_TYPESPAR(sb, i).s_spar_map[j] != + NULL) { + st = (struct sparingTable *)sdata-> + s_spar_map[j]->b_data; break; } } @@ -158,60 +174,123 @@ int udf_relocate_blocks(struct super_block *sb, long old_block, long *new_block) if (!st) return 1; - for (k=0; k<le16_to_cpu(st->reallocationTableLen); k++) - { - if (le32_to_cpu(st->mapEntry[k].origLocation) == 0xFFFFFFFF) - { - for (; j<4; j++) - { - if (sdata->s_spar_map[j]) - { - st = (struct sparingTable *)sdata->s_spar_map[j]->b_data; - st->mapEntry[k].origLocation = cpu_to_le32(packet); - udf_update_tag((char *)st, sizeof(struct sparingTable) + le16_to_cpu(st->reallocationTableLen) * sizeof(struct sparingEntry)); - mark_buffer_dirty(sdata->s_spar_map[j]); + for (k = 0; k < le16_to_cpu(st->reallocationTableLen); + k++) { + if (le32_to_cpu(st->mapEntry[k].origLocation) == + 0xFFFFFFFF) { + for (; j < 4; j++) { + if (sdata->s_spar_map[j]) { + st = (struct + sparingTable *) + sdata-> + s_spar_map[j]-> + b_data; + st->mapEntry[k]. + origLocation = + cpu_to_le32(packet); + udf_update_tag((char *) + st, + sizeof + (struct + sparingTable) + + + le16_to_cpu + (st-> + reallocationTableLen) + * + sizeof + (struct + sparingEntry)); + mark_buffer_dirty + (sdata-> + s_spar_map[j]); } } - *new_block = le32_to_cpu(st->mapEntry[k].mappedLocation) + - ((old_block - UDF_SB_PARTROOT(sb,i)) & (sdata->s_packet_len - 1)); + *new_block = + le32_to_cpu(st->mapEntry[k]. + mappedLocation) + + ((old_block - + UDF_SB_PARTROOT(sb, + i)) & (sdata-> + s_packet_len + - 1)); return 0; - } - else if (le32_to_cpu(st->mapEntry[k].origLocation) == packet) - { - *new_block = le32_to_cpu(st->mapEntry[k].mappedLocation) + - ((old_block - UDF_SB_PARTROOT(sb,i)) & (sdata->s_packet_len - 1)); + } else + if (le32_to_cpu + (st->mapEntry[k].origLocation) == + packet) { + *new_block = + le32_to_cpu(st->mapEntry[k]. + mappedLocation) + + ((old_block - + UDF_SB_PARTROOT(sb, + i)) & (sdata-> + s_packet_len + - 1)); return 0; - } - else if (le32_to_cpu(st->mapEntry[k].origLocation) > packet) + } else + if (le32_to_cpu + (st->mapEntry[k].origLocation) > packet) break; } - for (l=k; l<le16_to_cpu(st->reallocationTableLen); l++) - { - if (le32_to_cpu(st->mapEntry[l].origLocation) == 0xFFFFFFFF) - { - for (; j<4; j++) - { - if (sdata->s_spar_map[j]) - { - st = (struct sparingTable *)sdata->s_spar_map[j]->b_data; - mapEntry = st->mapEntry[l]; - mapEntry.origLocation = cpu_to_le32(packet); - memmove(&st->mapEntry[k+1], &st->mapEntry[k], (l-k)*sizeof(struct sparingEntry)); - st->mapEntry[k] = mapEntry; - udf_update_tag((char *)st, sizeof(struct sparingTable) + le16_to_cpu(st->reallocationTableLen) * sizeof(struct sparingEntry)); - mark_buffer_dirty(sdata->s_spar_map[j]); + for (l = k; l < le16_to_cpu(st->reallocationTableLen); + l++) { + if (le32_to_cpu(st->mapEntry[l].origLocation) == + 0xFFFFFFFF) { + for (; j < 4; j++) { + if (sdata->s_spar_map[j]) { + st = (struct + sparingTable *) + sdata-> + s_spar_map[j]-> + b_data; + mapEntry = + st->mapEntry[l]; + mapEntry.origLocation = + cpu_to_le32(packet); + memmove(&st-> + mapEntry[k + 1], + &st-> + mapEntry[k], + (l - + k) * + sizeof(struct + sparingEntry)); + st->mapEntry[k] = + mapEntry; + udf_update_tag((char *) + st, + sizeof + (struct + sparingTable) + + + le16_to_cpu + (st-> + reallocationTableLen) + * + sizeof + (struct + sparingEntry)); + mark_buffer_dirty + (sdata-> + s_spar_map[j]); } } - *new_block = le32_to_cpu(st->mapEntry[k].mappedLocation) + - ((old_block - UDF_SB_PARTROOT(sb,i)) & (sdata->s_packet_len - 1)); + *new_block = + le32_to_cpu(st->mapEntry[k]. + mappedLocation) + + ((old_block - + UDF_SB_PARTROOT(sb, + i)) & (sdata-> + s_packet_len + - 1)); return 0; } } return 1; } } - if (i == UDF_SB_NUMPARTS(sb)) - { + if (i == UDF_SB_NUMPARTS(sb)) { /* outside of partitions */ /* for now, fail =) */ return 1; diff --git a/fs/udf/super.c b/fs/udf/super.c index d6a504f5d75..911387aa181 100644 --- a/fs/udf/super.c +++ b/fs/udf/super.c @@ -38,7 +38,7 @@ * 12/20/98 find the free space bitmap (if it exists) */ -#include "udfdecl.h" +#include "udfdecl.h" #include <linux/blkdev.h> #include <linux/slab.h> @@ -80,12 +80,15 @@ static int udf_remount_fs(struct super_block *, int *, char *); static int udf_check_valid(struct super_block *, int, int); static int udf_vrs(struct super_block *sb, int silent); static int udf_load_partition(struct super_block *, kernel_lb_addr *); -static int udf_load_logicalvol(struct super_block *, struct buffer_head *, kernel_lb_addr *); +static int udf_load_logicalvol(struct super_block *, struct buffer_head *, + kernel_lb_addr *); static void udf_load_logicalvolint(struct super_block *, kernel_extent_ad); static void udf_find_anchor(struct super_block *); -static int udf_find_fileset(struct super_block *, kernel_lb_addr *, kernel_lb_addr *); +static int udf_find_fileset(struct super_block *, kernel_lb_addr *, + kernel_lb_addr *); static void udf_load_pvoldesc(struct super_block *, struct buffer_head *); -static void udf_load_fileset(struct super_block *, struct buffer_head *, kernel_lb_addr *); +static void udf_load_fileset(struct super_block *, struct buffer_head *, + kernel_lb_addr *); static void udf_load_partdesc(struct super_block *, struct buffer_head *); static void udf_open_lvid(struct super_block *); static void udf_close_lvid(struct super_block *); @@ -94,25 +97,27 @@ static int udf_statfs(struct dentry *, struct kstatfs *); /* UDF filesystem type */ static int udf_get_sb(struct file_system_type *fs_type, - int flags, const char *dev_name, void *data, struct vfsmount *mnt) + int flags, const char *dev_name, void *data, + struct vfsmount *mnt) { return get_sb_bdev(fs_type, flags, dev_name, data, udf_fill_super, mnt); } static struct file_system_type udf_fstype = { - .owner = THIS_MODULE, - .name = "udf", - .get_sb = udf_get_sb, - .kill_sb = kill_block_super, - .fs_flags = FS_REQUIRES_DEV, + .owner = THIS_MODULE, + .name = "udf", + .get_sb = udf_get_sb, + .kill_sb = kill_block_super, + .fs_flags = FS_REQUIRES_DEV, }; -static struct kmem_cache * udf_inode_cachep; +static struct kmem_cache *udf_inode_cachep; static struct inode *udf_alloc_inode(struct super_block *sb) { struct udf_inode_info *ei; - ei = (struct udf_inode_info *)kmem_cache_alloc(udf_inode_cachep, GFP_KERNEL); + ei = (struct udf_inode_info *)kmem_cache_alloc(udf_inode_cachep, + GFP_KERNEL); if (!ei) return NULL; @@ -130,9 +135,9 @@ static void udf_destroy_inode(struct inode *inode) kmem_cache_free(udf_inode_cachep, UDF_I(inode)); } -static void init_once(void * foo, struct kmem_cache * cachep, unsigned long flags) +static void init_once(void *foo, struct kmem_cache *cachep, unsigned long flags) { - struct udf_inode_info *ei = (struct udf_inode_info *) foo; + struct udf_inode_info *ei = (struct udf_inode_info *)foo; ei->i_ext.i_data = NULL; inode_init_once(&ei->vfs_inode); @@ -142,8 +147,8 @@ static int init_inodecache(void) { udf_inode_cachep = kmem_cache_create("udf_inode_cache", sizeof(struct udf_inode_info), - 0, (SLAB_RECLAIM_ACCOUNT| - SLAB_MEM_SPREAD), + 0, (SLAB_RECLAIM_ACCOUNT | + SLAB_MEM_SPREAD), init_once, NULL); if (udf_inode_cachep == NULL) return -ENOMEM; @@ -157,19 +162,18 @@ static void destroy_inodecache(void) /* Superblock operations */ static const struct super_operations udf_sb_ops = { - .alloc_inode = udf_alloc_inode, - .destroy_inode = udf_destroy_inode, - .write_inode = udf_write_inode, - .delete_inode = udf_delete_inode, - .clear_inode = udf_clear_inode, - .put_super = udf_put_super, - .write_super = udf_write_super, - .statfs = udf_statfs, - .remount_fs = udf_remount_fs, + .alloc_inode = udf_alloc_inode, + .destroy_inode = udf_destroy_inode, + .write_inode = udf_write_inode, + .delete_inode = udf_delete_inode, + .clear_inode = udf_clear_inode, + .put_super = udf_put_super, + .write_super = udf_write_super, + .statfs = udf_statfs, + .remount_fs = udf_remount_fs, }; -struct udf_options -{ +struct udf_options { unsigned char novrs; unsigned int blocksize; unsigned int session; @@ -196,9 +200,9 @@ static int __init init_udf_fs(void) if (err) goto out; return 0; -out: + out: destroy_inodecache(); -out1: + out1: return err; } @@ -209,7 +213,7 @@ static void __exit exit_udf_fs(void) } module_init(init_udf_fs) -module_exit(exit_udf_fs) + module_exit(exit_udf_fs) /* * udf_parse_options @@ -264,7 +268,6 @@ module_exit(exit_udf_fs) * July 1, 1997 - Andrew E. Mileski * Written, tested, and released. */ - enum { Opt_novrs, Opt_nostrict, Opt_bs, Opt_unhide, Opt_undelete, Opt_noadinicb, Opt_adinicb, Opt_shortad, Opt_longad, @@ -303,8 +306,7 @@ static match_table_t tokens = { {Opt_err, NULL} }; -static int -udf_parse_options(char *options, struct udf_options *uopt) +static int udf_parse_options(char *options, struct udf_options *uopt) { char *p; int option; @@ -323,126 +325,123 @@ udf_parse_options(char *options, struct udf_options *uopt) if (!options) return 1; - while ((p = strsep(&options, ",")) != NULL) - { + while ((p = strsep(&options, ",")) != NULL) { substring_t args[MAX_OPT_ARGS]; int token; if (!*p) continue; token = match_token(p, tokens, args); - switch (token) - { - case Opt_novrs: - uopt->novrs = 1; - case Opt_bs: - if (match_int(&args[0], &option)) - return 0; - uopt->blocksize = option; - break; - case Opt_unhide: - uopt->flags |= (1 << UDF_FLAG_UNHIDE); - break; - case Opt_undelete: - uopt->flags |= (1 << UDF_FLAG_UNDELETE); - break; - case Opt_noadinicb: - uopt->flags &= ~(1 << UDF_FLAG_USE_AD_IN_ICB); - break; - case Opt_adinicb: - uopt->flags |= (1 << UDF_FLAG_USE_AD_IN_ICB); - break; - case Opt_shortad: - uopt->flags |= (1 << UDF_FLAG_USE_SHORT_AD); - break; - case Opt_longad: - uopt->flags &= ~(1 << UDF_FLAG_USE_SHORT_AD); - break; - case Opt_gid: - if (match_int(args, &option)) - return 0; - uopt->gid = option; - break; - case Opt_uid: - if (match_int(args, &option)) - return 0; - uopt->uid = option; - break; - case Opt_umask: - if (match_octal(args, &option)) - return 0; - uopt->umask = option; - break; - case Opt_nostrict: - uopt->flags &= ~(1 << UDF_FLAG_STRICT); - break; - case Opt_session: - if (match_int(args, &option)) - return 0; - uopt->session = option; - break; - case Opt_lastblock: - if (match_int(args, &option)) - return 0; - uopt->lastblock = option; - break; - case Opt_anchor: - if (match_int(args, &option)) - return 0; - uopt->anchor = option; - break; - case Opt_volume: - if (match_int(args, &option)) - return 0; - uopt->volume = option; - break; - case Opt_partition: - if (match_int(args, &option)) - return 0; - uopt->partition = option; - break; - case Opt_fileset: - if (match_int(args, &option)) - return 0; - uopt->fileset = option; - break; - case Opt_rootdir: - if (match_int(args, &option)) - return 0; - uopt->rootdir = option; - break; - case Opt_utf8: - uopt->flags |= (1 << UDF_FLAG_UTF8); - break; + switch (token) { + case Opt_novrs: + uopt->novrs = 1; + case Opt_bs: + if (match_int(&args[0], &option)) + return 0; + uopt->blocksize = option; + break; + case Opt_unhide: + uopt->flags |= (1 << UDF_FLAG_UNHIDE); + break; + case Opt_undelete: + uopt->flags |= (1 << UDF_FLAG_UNDELETE); + break; + case Opt_noadinicb: + uopt->flags &= ~(1 << UDF_FLAG_USE_AD_IN_ICB); + break; + case Opt_adinicb: + uopt->flags |= (1 << UDF_FLAG_USE_AD_IN_ICB); + break; + case Opt_shortad: + uopt->flags |= (1 << UDF_FLAG_USE_SHORT_AD); + break; + case Opt_longad: + uopt->flags &= ~(1 << UDF_FLAG_USE_SHORT_AD); + break; + case Opt_gid: + if (match_int(args, &option)) + return 0; + uopt->gid = option; + break; + case Opt_uid: + if (match_int(args, &option)) + return 0; + uopt->uid = option; + break; + case Opt_umask: + if (match_octal(args, &option)) + return 0; + uopt->umask = option; + break; + case Opt_nostrict: + uopt->flags &= ~(1 << UDF_FLAG_STRICT); + break; + case Opt_session: + if (match_int(args, &option)) + return 0; + uopt->session = option; + break; + case Opt_lastblock: + if (match_int(args, &option)) + return 0; + uopt->lastblock = option; + break; + case Opt_anchor: + if (match_int(args, &option)) + return 0; + uopt->anchor = option; + break; + case Opt_volume: + if (match_int(args, &option)) + return 0; + uopt->volume = option; + break; + case Opt_partition: + if (match_int(args, &option)) + return 0; + uopt->partition = option; + break; + case Opt_fileset: + if (match_int(args, &option)) + return 0; + uopt->fileset = option; + break; + case Opt_rootdir: + if (match_int(args, &option)) + return 0; + uopt->rootdir = option; + break; + case Opt_utf8: + uopt->flags |= (1 << UDF_FLAG_UTF8); + break; #ifdef CONFIG_UDF_NLS - case Opt_iocharset: - uopt->nls_map = load_nls(args[0].from); - uopt->flags |= (1 << UDF_FLAG_NLS_MAP); - break; + case Opt_iocharset: + uopt->nls_map = load_nls(args[0].from); + uopt->flags |= (1 << UDF_FLAG_NLS_MAP); + break; #endif - case Opt_uignore: - uopt->flags |= (1 << UDF_FLAG_UID_IGNORE); - break; - case Opt_uforget: - uopt->flags |= (1 << UDF_FLAG_UID_FORGET); - break; - case Opt_gignore: - uopt->flags |= (1 << UDF_FLAG_GID_IGNORE); - break; - case Opt_gforget: - uopt->flags |= (1 << UDF_FLAG_GID_FORGET); - break; - default: - printk(KERN_ERR "udf: bad mount option \"%s\" " - "or missing value\n", p); + case Opt_uignore: + uopt->flags |= (1 << UDF_FLAG_UID_IGNORE); + break; + case Opt_uforget: + uopt->flags |= (1 << UDF_FLAG_UID_FORGET); + break; + case Opt_gignore: + uopt->flags |= (1 << UDF_FLAG_GID_IGNORE); + break; + case Opt_gforget: + uopt->flags |= (1 << UDF_FLAG_GID_FORGET); + break; + default: + printk(KERN_ERR "udf: bad mount option \"%s\" " + "or missing value\n", p); return 0; } } return 1; } -void -udf_write_super(struct super_block *sb) +void udf_write_super(struct super_block *sb) { lock_kernel(); if (!(sb->s_flags & MS_RDONLY)) @@ -451,22 +450,21 @@ udf_write_super(struct super_block *sb) unlock_kernel(); } -static int -udf_remount_fs(struct super_block *sb, int *flags, char *options) +static int udf_remount_fs(struct super_block *sb, int *flags, char *options) { struct udf_options uopt; - uopt.flags = UDF_SB(sb)->s_flags ; - uopt.uid = UDF_SB(sb)->s_uid ; - uopt.gid = UDF_SB(sb)->s_gid ; - uopt.umask = UDF_SB(sb)->s_umask ; + uopt.flags = UDF_SB(sb)->s_flags; + uopt.uid = UDF_SB(sb)->s_uid; + uopt.gid = UDF_SB(sb)->s_gid; + uopt.umask = UDF_SB(sb)->s_umask; - if ( !udf_parse_options(options, &uopt) ) + if (!udf_parse_options(options, &uopt)) return -EINVAL; UDF_SB(sb)->s_flags = uopt.flags; - UDF_SB(sb)->s_uid = uopt.uid; - UDF_SB(sb)->s_gid = uopt.gid; + UDF_SB(sb)->s_uid = uopt.uid; + UDF_SB(sb)->s_gid = uopt.gid; UDF_SB(sb)->s_umask = uopt.umask; if (UDF_SB_LVIDBH(sb)) { @@ -512,8 +510,7 @@ udf_remount_fs(struct super_block *sb, int *flags, char *options) * July 1, 1997 - Andrew E. Mileski * Written, tested, and released. */ -static int -udf_set_blocksize(struct super_block *sb, int bsize) +static int udf_set_blocksize(struct super_block *sb, int bsize) { if (!sb_min_blocksize(sb, bsize)) { udf_debug("Bad block size (%d)\n", bsize); @@ -523,16 +520,15 @@ udf_set_blocksize(struct super_block *sb, int bsize) return sb->s_blocksize; } -static int -udf_vrs(struct super_block *sb, int silent) +static int udf_vrs(struct super_block *sb, int silent) { struct volStructDesc *vsd = NULL; int sector = 32768; int sectorsize; struct buffer_head *bh = NULL; - int iso9660=0; - int nsr02=0; - int nsr03=0; + int iso9660 = 0; + int nsr02 = 0; + int nsr03 = 0; /* Block size must be a multiple of 512 */ if (sb->s_blocksize & 511) @@ -546,10 +542,9 @@ udf_vrs(struct super_block *sb, int silent) sector += (UDF_SB_SESSION(sb) << sb->s_blocksize_bits); udf_debug("Starting at sector %u (%ld byte sectors)\n", - (sector >> sb->s_blocksize_bits), sb->s_blocksize); + (sector >> sb->s_blocksize_bits), sb->s_blocksize); /* Process the sequence (if applicable) */ - for (;!nsr02 && !nsr03; sector += sectorsize) - { + for (; !nsr02 && !nsr03; sector += sectorsize) { /* Read a block */ bh = udf_tread(sb, sector >> sb->s_blocksize_bits); if (!bh) @@ -557,52 +552,56 @@ udf_vrs(struct super_block *sb, int silent) /* Look for ISO descriptors */ vsd = (struct volStructDesc *)(bh->b_data + - (sector & (sb->s_blocksize - 1))); + (sector & + (sb->s_blocksize - 1))); - if (vsd->stdIdent[0] == 0) - { + if (vsd->stdIdent[0] == 0) { brelse(bh); break; - } - else if (!strncmp(vsd->stdIdent, VSD_STD_ID_CD001, VSD_STD_ID_LEN)) - { + } else + if (!strncmp + (vsd->stdIdent, VSD_STD_ID_CD001, VSD_STD_ID_LEN)) { iso9660 = sector; - switch (vsd->structType) - { - case 0: - udf_debug("ISO9660 Boot Record found\n"); - break; - case 1: - udf_debug("ISO9660 Primary Volume Descriptor found\n"); - break; - case 2: - udf_debug("ISO9660 Supplementary Volume Descriptor found\n"); - break; - case 3: - udf_debug("ISO9660 Volume Partition Descriptor found\n"); - break; - case 255: - udf_debug("ISO9660 Volume Descriptor Set Terminator found\n"); - break; - default: - udf_debug("ISO9660 VRS (%u) found\n", vsd->structType); - break; + switch (vsd->structType) { + case 0: + udf_debug("ISO9660 Boot Record found\n"); + break; + case 1: + udf_debug + ("ISO9660 Primary Volume Descriptor found\n"); + break; + case 2: + udf_debug + ("ISO9660 Supplementary Volume Descriptor found\n"); + break; + case 3: + udf_debug + ("ISO9660 Volume Partition Descriptor found\n"); + break; + case 255: + udf_debug + ("ISO9660 Volume Descriptor Set Terminator found\n"); + break; + default: + udf_debug("ISO9660 VRS (%u) found\n", + vsd->structType); + break; } - } - else if (!strncmp(vsd->stdIdent, VSD_STD_ID_BEA01, VSD_STD_ID_LEN)) - { - } - else if (!strncmp(vsd->stdIdent, VSD_STD_ID_TEA01, VSD_STD_ID_LEN)) - { + } else + if (!strncmp + (vsd->stdIdent, VSD_STD_ID_BEA01, VSD_STD_ID_LEN)) { + } else + if (!strncmp + (vsd->stdIdent, VSD_STD_ID_TEA01, VSD_STD_ID_LEN)) { brelse(bh); break; - } - else if (!strncmp(vsd->stdIdent, VSD_STD_ID_NSR02, VSD_STD_ID_LEN)) - { + } else + if (!strncmp + (vsd->stdIdent, VSD_STD_ID_NSR02, VSD_STD_ID_LEN)) { nsr02 = sector; - } - else if (!strncmp(vsd->stdIdent, VSD_STD_ID_NSR03, VSD_STD_ID_LEN)) - { + } else + if (!strncmp + (vsd->stdIdent, VSD_STD_ID_NSR03, VSD_STD_ID_LEN)) { nsr03 = sector; } brelse(bh); @@ -635,8 +634,7 @@ udf_vrs(struct super_block *sb, int silent) * July 1, 1997 - Andrew E. Mileski * Written, tested, and released. */ -static void -udf_find_anchor(struct super_block *sb) +static void udf_find_anchor(struct super_block *sb) { int lastblock = UDF_SB_LASTBLOCK(sb); struct buffer_head *bh = NULL; @@ -644,13 +642,13 @@ udf_find_anchor(struct super_block *sb) uint32_t location; int i; - if (lastblock) - { + if (lastblock) { int varlastblock = udf_variable_to_fixed(lastblock); - int last[] = { lastblock, lastblock - 2, - lastblock - 150, lastblock - 152, - varlastblock, varlastblock - 2, - varlastblock - 150, varlastblock - 152 }; + int last[] = { lastblock, lastblock - 2, + lastblock - 150, lastblock - 152, + varlastblock, varlastblock - 2, + varlastblock - 150, varlastblock - 152 + }; lastblock = 0; @@ -663,90 +661,103 @@ udf_find_anchor(struct super_block *sb) * however, if the disc isn't closed, it could be 512 */ for (i = 0; !lastblock && i < ARRAY_SIZE(last); i++) { - if (last[i] < 0 || !(bh = sb_bread(sb, last[i]))) - { + if (last[i] < 0 || !(bh = sb_bread(sb, last[i]))) { ident = location = 0; - } - else - { - ident = le16_to_cpu(((tag *)bh->b_data)->tagIdent); - location = le32_to_cpu(((tag *)bh->b_data)->tagLocation); + } else { + ident = + le16_to_cpu(((tag *) bh->b_data)->tagIdent); + location = + le32_to_cpu(((tag *) bh->b_data)-> + tagLocation); brelse(bh); } - if (ident == TAG_IDENT_AVDP) - { - if (location == last[i] - UDF_SB_SESSION(sb)) - { - lastblock = UDF_SB_ANCHOR(sb)[0] = last[i] - UDF_SB_SESSION(sb); - UDF_SB_ANCHOR(sb)[1] = last[i] - 256 - UDF_SB_SESSION(sb); - } - else if (location == udf_variable_to_fixed(last[i]) - UDF_SB_SESSION(sb)) - { + if (ident == TAG_IDENT_AVDP) { + if (location == last[i] - UDF_SB_SESSION(sb)) { + lastblock = UDF_SB_ANCHOR(sb)[0] = + last[i] - UDF_SB_SESSION(sb); + UDF_SB_ANCHOR(sb)[1] = + last[i] - 256 - UDF_SB_SESSION(sb); + } else if (location == + udf_variable_to_fixed(last[i]) - + UDF_SB_SESSION(sb)) { UDF_SET_FLAG(sb, UDF_FLAG_VARCONV); - lastblock = UDF_SB_ANCHOR(sb)[0] = udf_variable_to_fixed(last[i]) - UDF_SB_SESSION(sb); - UDF_SB_ANCHOR(sb)[1] = lastblock - 256 - UDF_SB_SESSION(sb); - } - else - udf_debug("Anchor found at block %d, location mismatch %d.\n", - last[i], location); - } - else if (ident == TAG_IDENT_FE || ident == TAG_IDENT_EFE) - { + lastblock = UDF_SB_ANCHOR(sb)[0] = + udf_variable_to_fixed(last[i]) - + UDF_SB_SESSION(sb); + UDF_SB_ANCHOR(sb)[1] = + lastblock - 256 - + UDF_SB_SESSION(sb); + } else + udf_debug + ("Anchor found at block %d, location mismatch %d.\n", + last[i], location); + } else if (ident == TAG_IDENT_FE + || ident == TAG_IDENT_EFE) { lastblock = last[i]; UDF_SB_ANCHOR(sb)[3] = 512; - } - else - { - if (last[i] < 256 || !(bh = sb_bread(sb, last[i] - 256))) - { + } else { + if (last[i] < 256 + || !(bh = sb_bread(sb, last[i] - 256))) { ident = location = 0; - } - else - { - ident = le16_to_cpu(((tag *)bh->b_data)->tagIdent); - location = le32_to_cpu(((tag *)bh->b_data)->tagLocation); + } else { + ident = + le16_to_cpu(((tag *) bh->b_data)-> + tagIdent); + location = + le32_to_cpu(((tag *) bh->b_data)-> + tagLocation); brelse(bh); } - + if (ident == TAG_IDENT_AVDP && - location == last[i] - 256 - UDF_SB_SESSION(sb)) - { + location == + last[i] - 256 - UDF_SB_SESSION(sb)) { lastblock = last[i]; UDF_SB_ANCHOR(sb)[1] = last[i] - 256; - } - else - { - if (last[i] < 312 + UDF_SB_SESSION(sb) || !(bh = sb_bread(sb, last[i] - 312 - UDF_SB_SESSION(sb)))) + } else { + if (last[i] < 312 + UDF_SB_SESSION(sb) + || !(bh = + sb_bread(sb, + last[i] - 312 - + UDF_SB_SESSION(sb)))) { ident = location = 0; - } - else - { - ident = le16_to_cpu(((tag *)bh->b_data)->tagIdent); - location = le32_to_cpu(((tag *)bh->b_data)->tagLocation); + } else { + ident = + le16_to_cpu(((tag *) bh-> + b_data)-> + tagIdent); + location = + le32_to_cpu(((tag *) bh-> + b_data)-> + tagLocation); brelse(bh); } - + if (ident == TAG_IDENT_AVDP && - location == udf_variable_to_fixed(last[i]) - 256) - { - UDF_SET_FLAG(sb, UDF_FLAG_VARCONV); - lastblock = udf_variable_to_fixed(last[i]); - UDF_SB_ANCHOR(sb)[1] = lastblock - 256; + location == + udf_variable_to_fixed(last[i]) - + 256) { + UDF_SET_FLAG(sb, + UDF_FLAG_VARCONV); + lastblock = + udf_variable_to_fixed(last + [i]); + UDF_SB_ANCHOR(sb)[1] = + lastblock - 256; } } } } } - if (!lastblock) - { + if (!lastblock) { /* We havn't found the lastblock. check 312 */ - if ((bh = sb_bread(sb, 312 + UDF_SB_SESSION(sb)))) - { - ident = le16_to_cpu(((tag *)bh->b_data)->tagIdent); - location = le32_to_cpu(((tag *)bh->b_data)->tagLocation); + if ((bh = sb_bread(sb, 312 + UDF_SB_SESSION(sb)))) { + ident = le16_to_cpu(((tag *) bh->b_data)->tagIdent); + location = + le32_to_cpu(((tag *) bh->b_data)->tagLocation); brelse(bh); if (ident == TAG_IDENT_AVDP && location == 256) @@ -755,18 +766,19 @@ udf_find_anchor(struct super_block *sb) } for (i = 0; i < ARRAY_SIZE(UDF_SB_ANCHOR(sb)); i++) { - if (UDF_SB_ANCHOR(sb)[i]) - { + if (UDF_SB_ANCHOR(sb)[i]) { if (!(bh = udf_read_tagged(sb, - UDF_SB_ANCHOR(sb)[i], UDF_SB_ANCHOR(sb)[i], &ident))) - { + UDF_SB_ANCHOR(sb)[i], + UDF_SB_ANCHOR(sb)[i], + &ident))) { UDF_SB_ANCHOR(sb)[i] = 0; - } - else - { + } else { brelse(bh); if ((ident != TAG_IDENT_AVDP) && (i || - (ident != TAG_IDENT_FE && ident != TAG_IDENT_EFE))) + (ident != + TAG_IDENT_FE + && ident != + TAG_IDENT_EFE))) { UDF_SB_ANCHOR(sb)[i] = 0; } @@ -777,72 +789,75 @@ udf_find_anchor(struct super_block *sb) UDF_SB_LASTBLOCK(sb) = lastblock; } -static int -udf_find_fileset(struct super_block *sb, kernel_lb_addr *fileset, kernel_lb_addr *root) +static int +udf_find_fileset(struct super_block *sb, kernel_lb_addr * fileset, + kernel_lb_addr * root) { struct buffer_head *bh = NULL; long lastblock; uint16_t ident; if (fileset->logicalBlockNum != 0xFFFFFFFF || - fileset->partitionReferenceNum != 0xFFFF) - { + fileset->partitionReferenceNum != 0xFFFF) { bh = udf_read_ptagged(sb, *fileset, 0, &ident); if (!bh) return 1; - else if (ident != TAG_IDENT_FSD) - { + else if (ident != TAG_IDENT_FSD) { brelse(bh); return 1; } - + } - if (!bh) /* Search backwards through the partitions */ - { + if (!bh) { /* Search backwards through the partitions */ kernel_lb_addr newfileset; return 1; - - for (newfileset.partitionReferenceNum=UDF_SB_NUMPARTS(sb)-1; - (newfileset.partitionReferenceNum != 0xFFFF && - fileset->logicalBlockNum == 0xFFFFFFFF && - fileset->partitionReferenceNum == 0xFFFF); - newfileset.partitionReferenceNum--) - { - lastblock = UDF_SB_PARTLEN(sb, newfileset.partitionReferenceNum); + + for (newfileset.partitionReferenceNum = UDF_SB_NUMPARTS(sb) - 1; + (newfileset.partitionReferenceNum != 0xFFFF && + fileset->logicalBlockNum == 0xFFFFFFFF && + fileset->partitionReferenceNum == 0xFFFF); + newfileset.partitionReferenceNum--) { + lastblock = + UDF_SB_PARTLEN(sb, + newfileset.partitionReferenceNum); newfileset.logicalBlockNum = 0; - do - { - bh = udf_read_ptagged(sb, newfileset, 0, &ident); - if (!bh) - { - newfileset.logicalBlockNum ++; + do { + bh = udf_read_ptagged(sb, newfileset, 0, + &ident); + if (!bh) { + newfileset.logicalBlockNum++; continue; } - switch (ident) - { - case TAG_IDENT_SBD: + switch (ident) { + case TAG_IDENT_SBD: { struct spaceBitmapDesc *sp; - sp = (struct spaceBitmapDesc *)bh->b_data; - newfileset.logicalBlockNum += 1 + - ((le32_to_cpu(sp->numOfBytes) + sizeof(struct spaceBitmapDesc) - 1) - >> sb->s_blocksize_bits); + sp = (struct spaceBitmapDesc *) + bh->b_data; + newfileset.logicalBlockNum += + 1 + + ((le32_to_cpu + (sp->numOfBytes) + + sizeof(struct + spaceBitmapDesc) - + 1) + >> sb->s_blocksize_bits); brelse(bh); break; } - case TAG_IDENT_FSD: + case TAG_IDENT_FSD: { *fileset = newfileset; break; } - default: + default: { - newfileset.logicalBlockNum ++; + newfileset.logicalBlockNum++; brelse(bh); bh = NULL; break; @@ -850,16 +865,16 @@ udf_find_fileset(struct super_block *sb, kernel_lb_addr *fileset, kernel_lb_addr } } while (newfileset.logicalBlockNum < lastblock && - fileset->logicalBlockNum == 0xFFFFFFFF && - fileset->partitionReferenceNum == 0xFFFF); + fileset->logicalBlockNum == 0xFFFFFFFF && + fileset->partitionReferenceNum == 0xFFFF); } } if ((fileset->logicalBlockNum != 0xFFFFFFFF || - fileset->partitionReferenceNum != 0xFFFF) && bh) - { + fileset->partitionReferenceNum != 0xFFFF) && bh) { udf_debug("Fileset at block=%d, partition=%d\n", - fileset->logicalBlockNum, fileset->partitionReferenceNum); + fileset->logicalBlockNum, + fileset->partitionReferenceNum); UDF_SB_PARTITION(sb) = fileset->partitionReferenceNum; udf_load_fileset(sb, bh, root); @@ -869,8 +884,7 @@ udf_find_fileset(struct super_block *sb, kernel_lb_addr *fileset, kernel_lb_addr return 1; } -static void -udf_load_pvoldesc(struct super_block *sb, struct buffer_head *bh) +static void udf_load_pvoldesc(struct super_block *sb, struct buffer_head *bh) { struct primaryVolDesc *pvoldesc; time_t recording; @@ -880,37 +894,35 @@ udf_load_pvoldesc(struct super_block *sb, struct buffer_head *bh) pvoldesc = (struct primaryVolDesc *)bh->b_data; - if ( udf_stamp_to_time(&recording, &recording_usec, - lets_to_cpu(pvoldesc->recordingDateAndTime)) ) - { + if (udf_stamp_to_time(&recording, &recording_usec, + lets_to_cpu(pvoldesc->recordingDateAndTime))) { kernel_timestamp ts; ts = lets_to_cpu(pvoldesc->recordingDateAndTime); - udf_debug("recording time %ld/%ld, %04u/%02u/%02u %02u:%02u (%x)\n", - recording, recording_usec, - ts.year, ts.month, ts.day, ts.hour, ts.minute, ts.typeAndTimezone); + udf_debug + ("recording time %ld/%ld, %04u/%02u/%02u %02u:%02u (%x)\n", + recording, recording_usec, ts.year, ts.month, ts.day, + ts.hour, ts.minute, ts.typeAndTimezone); UDF_SB_RECORDTIME(sb).tv_sec = recording; UDF_SB_RECORDTIME(sb).tv_nsec = recording_usec * 1000; } - if ( !udf_build_ustr(&instr, pvoldesc->volIdent, 32) ) - { - if (udf_CS0toUTF8(&outstr, &instr)) - { - strncpy( UDF_SB_VOLIDENT(sb), outstr.u_name, + if (!udf_build_ustr(&instr, pvoldesc->volIdent, 32)) { + if (udf_CS0toUTF8(&outstr, &instr)) { + strncpy(UDF_SB_VOLIDENT(sb), outstr.u_name, outstr.u_len > 31 ? 31 : outstr.u_len); udf_debug("volIdent[] = '%s'\n", UDF_SB_VOLIDENT(sb)); } } - if ( !udf_build_ustr(&instr, pvoldesc->volSetIdent, 128) ) - { + if (!udf_build_ustr(&instr, pvoldesc->volSetIdent, 128)) { if (udf_CS0toUTF8(&outstr, &instr)) udf_debug("volSetIdent[] = '%s'\n", outstr.u_name); } } -static void -udf_load_fileset(struct super_block *sb, struct buffer_head *bh, kernel_lb_addr *root) +static void +udf_load_fileset(struct super_block *sb, struct buffer_head *bh, + kernel_lb_addr * root) { struct fileSetDesc *fset; @@ -920,109 +932,154 @@ udf_load_fileset(struct super_block *sb, struct buffer_head *bh, kernel_lb_addr UDF_SB_SERIALNUM(sb) = le16_to_cpu(fset->descTag.tagSerialNum); - udf_debug("Rootdir at block=%d, partition=%d\n", - root->logicalBlockNum, root->partitionReferenceNum); + udf_debug("Rootdir at block=%d, partition=%d\n", + root->logicalBlockNum, root->partitionReferenceNum); } -static void -udf_load_partdesc(struct super_block *sb, struct buffer_head *bh) +static void udf_load_partdesc(struct super_block *sb, struct buffer_head *bh) { struct partitionDesc *p; int i; p = (struct partitionDesc *)bh->b_data; - for (i=0; i<UDF_SB_NUMPARTS(sb); i++) - { - udf_debug("Searching map: (%d == %d)\n", - UDF_SB_PARTMAPS(sb)[i].s_partition_num, le16_to_cpu(p->partitionNumber)); - if (UDF_SB_PARTMAPS(sb)[i].s_partition_num == le16_to_cpu(p->partitionNumber)) - { - UDF_SB_PARTLEN(sb,i) = le32_to_cpu(p->partitionLength); /* blocks */ - UDF_SB_PARTROOT(sb,i) = le32_to_cpu(p->partitionStartingLocation); - if (le32_to_cpu(p->accessType) == PD_ACCESS_TYPE_READ_ONLY) - UDF_SB_PARTFLAGS(sb,i) |= UDF_PART_FLAG_READ_ONLY; - if (le32_to_cpu(p->accessType) == PD_ACCESS_TYPE_WRITE_ONCE) - UDF_SB_PARTFLAGS(sb,i) |= UDF_PART_FLAG_WRITE_ONCE; - if (le32_to_cpu(p->accessType) == PD_ACCESS_TYPE_REWRITABLE) - UDF_SB_PARTFLAGS(sb,i) |= UDF_PART_FLAG_REWRITABLE; - if (le32_to_cpu(p->accessType) == PD_ACCESS_TYPE_OVERWRITABLE) - UDF_SB_PARTFLAGS(sb,i) |= UDF_PART_FLAG_OVERWRITABLE; - - if (!strcmp(p->partitionContents.ident, PD_PARTITION_CONTENTS_NSR02) || - !strcmp(p->partitionContents.ident, PD_PARTITION_CONTENTS_NSR03)) - { + for (i = 0; i < UDF_SB_NUMPARTS(sb); i++) { + udf_debug("Searching map: (%d == %d)\n", + UDF_SB_PARTMAPS(sb)[i].s_partition_num, + le16_to_cpu(p->partitionNumber)); + if (UDF_SB_PARTMAPS(sb)[i].s_partition_num == + le16_to_cpu(p->partitionNumber)) { + UDF_SB_PARTLEN(sb, i) = le32_to_cpu(p->partitionLength); /* blocks */ + UDF_SB_PARTROOT(sb, i) = + le32_to_cpu(p->partitionStartingLocation); + if (le32_to_cpu(p->accessType) == + PD_ACCESS_TYPE_READ_ONLY) + UDF_SB_PARTFLAGS(sb, i) |= + UDF_PART_FLAG_READ_ONLY; + if (le32_to_cpu(p->accessType) == + PD_ACCESS_TYPE_WRITE_ONCE) + UDF_SB_PARTFLAGS(sb, i) |= + UDF_PART_FLAG_WRITE_ONCE; + if (le32_to_cpu(p->accessType) == + PD_ACCESS_TYPE_REWRITABLE) + UDF_SB_PARTFLAGS(sb, i) |= + UDF_PART_FLAG_REWRITABLE; + if (le32_to_cpu(p->accessType) == + PD_ACCESS_TYPE_OVERWRITABLE) + UDF_SB_PARTFLAGS(sb, i) |= + UDF_PART_FLAG_OVERWRITABLE; + + if (!strcmp + (p->partitionContents.ident, + PD_PARTITION_CONTENTS_NSR02) + || !strcmp(p->partitionContents.ident, + PD_PARTITION_CONTENTS_NSR03)) { struct partitionHeaderDesc *phd; - phd = (struct partitionHeaderDesc *)(p->partitionContentsUse); - if (phd->unallocSpaceTable.extLength) - { - kernel_lb_addr loc = { le32_to_cpu(phd->unallocSpaceTable.extPosition), i }; - - UDF_SB_PARTMAPS(sb)[i].s_uspace.s_table = - udf_iget(sb, loc); - UDF_SB_PARTFLAGS(sb,i) |= UDF_PART_FLAG_UNALLOC_TABLE; - udf_debug("unallocSpaceTable (part %d) @ %ld\n", - i, UDF_SB_PARTMAPS(sb)[i].s_uspace.s_table->i_ino); + phd = + (struct partitionHeaderDesc *)(p-> + partitionContentsUse); + if (phd->unallocSpaceTable.extLength) { + kernel_lb_addr loc = + { le32_to_cpu(phd-> + unallocSpaceTable. + extPosition), i }; + + UDF_SB_PARTMAPS(sb)[i].s_uspace. + s_table = udf_iget(sb, loc); + UDF_SB_PARTFLAGS(sb, i) |= + UDF_PART_FLAG_UNALLOC_TABLE; + udf_debug + ("unallocSpaceTable (part %d) @ %ld\n", + i, + UDF_SB_PARTMAPS(sb)[i].s_uspace. + s_table->i_ino); } - if (phd->unallocSpaceBitmap.extLength) - { + if (phd->unallocSpaceBitmap.extLength) { UDF_SB_ALLOC_BITMAP(sb, i, s_uspace); - if (UDF_SB_PARTMAPS(sb)[i].s_uspace.s_bitmap != NULL) - { - UDF_SB_PARTMAPS(sb)[i].s_uspace.s_bitmap->s_extLength = - le32_to_cpu(phd->unallocSpaceBitmap.extLength); - UDF_SB_PARTMAPS(sb)[i].s_uspace.s_bitmap->s_extPosition = - le32_to_cpu(phd->unallocSpaceBitmap.extPosition); - UDF_SB_PARTFLAGS(sb,i) |= UDF_PART_FLAG_UNALLOC_BITMAP; - udf_debug("unallocSpaceBitmap (part %d) @ %d\n", - i, UDF_SB_PARTMAPS(sb)[i].s_uspace.s_bitmap->s_extPosition); + if (UDF_SB_PARTMAPS(sb)[i].s_uspace. + s_bitmap != NULL) { + UDF_SB_PARTMAPS(sb)[i].s_uspace. + s_bitmap->s_extLength = + le32_to_cpu(phd-> + unallocSpaceBitmap. + extLength); + UDF_SB_PARTMAPS(sb)[i].s_uspace. + s_bitmap->s_extPosition = + le32_to_cpu(phd-> + unallocSpaceBitmap. + extPosition); + UDF_SB_PARTFLAGS(sb, i) |= + UDF_PART_FLAG_UNALLOC_BITMAP; + udf_debug + ("unallocSpaceBitmap (part %d) @ %d\n", + i, + UDF_SB_PARTMAPS(sb)[i]. + s_uspace.s_bitmap-> + s_extPosition); } } if (phd->partitionIntegrityTable.extLength) - udf_debug("partitionIntegrityTable (part %d)\n", i); - if (phd->freedSpaceTable.extLength) - { - kernel_lb_addr loc = { le32_to_cpu(phd->freedSpaceTable.extPosition), i }; - - UDF_SB_PARTMAPS(sb)[i].s_fspace.s_table = - udf_iget(sb, loc); - UDF_SB_PARTFLAGS(sb,i) |= UDF_PART_FLAG_FREED_TABLE; - udf_debug("freedSpaceTable (part %d) @ %ld\n", - i, UDF_SB_PARTMAPS(sb)[i].s_fspace.s_table->i_ino); + udf_debug + ("partitionIntegrityTable (part %d)\n", + i); + if (phd->freedSpaceTable.extLength) { + kernel_lb_addr loc = + { le32_to_cpu(phd->freedSpaceTable. + extPosition), i }; + + UDF_SB_PARTMAPS(sb)[i].s_fspace. + s_table = udf_iget(sb, loc); + UDF_SB_PARTFLAGS(sb, i) |= + UDF_PART_FLAG_FREED_TABLE; + udf_debug + ("freedSpaceTable (part %d) @ %ld\n", + i, + UDF_SB_PARTMAPS(sb)[i].s_fspace. + s_table->i_ino); } - if (phd->freedSpaceBitmap.extLength) - { + if (phd->freedSpaceBitmap.extLength) { UDF_SB_ALLOC_BITMAP(sb, i, s_fspace); - if (UDF_SB_PARTMAPS(sb)[i].s_fspace.s_bitmap != NULL) - { - UDF_SB_PARTMAPS(sb)[i].s_fspace.s_bitmap->s_extLength = - le32_to_cpu(phd->freedSpaceBitmap.extLength); - UDF_SB_PARTMAPS(sb)[i].s_fspace.s_bitmap->s_extPosition = - le32_to_cpu(phd->freedSpaceBitmap.extPosition); - UDF_SB_PARTFLAGS(sb,i) |= UDF_PART_FLAG_FREED_BITMAP; - udf_debug("freedSpaceBitmap (part %d) @ %d\n", - i, UDF_SB_PARTMAPS(sb)[i].s_fspace.s_bitmap->s_extPosition); + if (UDF_SB_PARTMAPS(sb)[i].s_fspace. + s_bitmap != NULL) { + UDF_SB_PARTMAPS(sb)[i].s_fspace. + s_bitmap->s_extLength = + le32_to_cpu(phd-> + freedSpaceBitmap. + extLength); + UDF_SB_PARTMAPS(sb)[i].s_fspace. + s_bitmap->s_extPosition = + le32_to_cpu(phd-> + freedSpaceBitmap. + extPosition); + UDF_SB_PARTFLAGS(sb, i) |= + UDF_PART_FLAG_FREED_BITMAP; + udf_debug + ("freedSpaceBitmap (part %d) @ %d\n", + i, + UDF_SB_PARTMAPS(sb)[i]. + s_fspace.s_bitmap-> + s_extPosition); } } } break; } } - if (i == UDF_SB_NUMPARTS(sb)) - { - udf_debug("Partition (%d) not found in partition map\n", le16_to_cpu(p->partitionNumber)); - } - else - { - udf_debug("Partition (%d:%d type %x) starts at physical %d, block length %d\n", - le16_to_cpu(p->partitionNumber), i, UDF_SB_PARTTYPE(sb,i), - UDF_SB_PARTROOT(sb,i), UDF_SB_PARTLEN(sb,i)); + if (i == UDF_SB_NUMPARTS(sb)) { + udf_debug("Partition (%d) not found in partition map\n", + le16_to_cpu(p->partitionNumber)); + } else { + udf_debug + ("Partition (%d:%d type %x) starts at physical %d, block length %d\n", + le16_to_cpu(p->partitionNumber), i, UDF_SB_PARTTYPE(sb, i), + UDF_SB_PARTROOT(sb, i), UDF_SB_PARTLEN(sb, i)); } } -static int -udf_load_logicalvol(struct super_block *sb, struct buffer_head * bh, kernel_lb_addr *fileset) +static int +udf_load_logicalvol(struct super_block *sb, struct buffer_head *bh, + kernel_lb_addr * fileset) { struct logicalVolDesc *lvd; int i, j, offset; @@ -1032,82 +1089,114 @@ udf_load_logicalvol(struct super_block *sb, struct buffer_head * bh, kernel_lb_a UDF_SB_ALLOC_PARTMAPS(sb, le32_to_cpu(lvd->numPartitionMaps)); - for (i=0,offset=0; - i<UDF_SB_NUMPARTS(sb) && offset<le32_to_cpu(lvd->mapTableLength); - i++,offset+=((struct genericPartitionMap *)&(lvd->partitionMaps[offset]))->partitionMapLength) - { - type = ((struct genericPartitionMap *)&(lvd->partitionMaps[offset]))->partitionMapType; - if (type == 1) - { - struct genericPartitionMap1 *gpm1 = (struct genericPartitionMap1 *)&(lvd->partitionMaps[offset]); - UDF_SB_PARTTYPE(sb,i) = UDF_TYPE1_MAP15; - UDF_SB_PARTVSN(sb,i) = le16_to_cpu(gpm1->volSeqNum); - UDF_SB_PARTNUM(sb,i) = le16_to_cpu(gpm1->partitionNum); - UDF_SB_PARTFUNC(sb,i) = NULL; - } - else if (type == 2) - { - struct udfPartitionMap2 *upm2 = (struct udfPartitionMap2 *)&(lvd->partitionMaps[offset]); - if (!strncmp(upm2->partIdent.ident, UDF_ID_VIRTUAL, strlen(UDF_ID_VIRTUAL))) - { - if (le16_to_cpu(((__le16 *)upm2->partIdent.identSuffix)[0]) == 0x0150) - { - UDF_SB_PARTTYPE(sb,i) = UDF_VIRTUAL_MAP15; - UDF_SB_PARTFUNC(sb,i) = udf_get_pblock_virt15; - } - else if (le16_to_cpu(((__le16 *)upm2->partIdent.identSuffix)[0]) == 0x0200) - { - UDF_SB_PARTTYPE(sb,i) = UDF_VIRTUAL_MAP20; - UDF_SB_PARTFUNC(sb,i) = udf_get_pblock_virt20; + for (i = 0, offset = 0; + i < UDF_SB_NUMPARTS(sb) + && offset < le32_to_cpu(lvd->mapTableLength); + i++, offset += + ((struct genericPartitionMap *)&(lvd->partitionMaps[offset]))-> + partitionMapLength) { + type = + ((struct genericPartitionMap *) + &(lvd->partitionMaps[offset]))->partitionMapType; + if (type == 1) { + struct genericPartitionMap1 *gpm1 = + (struct genericPartitionMap1 *)&(lvd-> + partitionMaps + [offset]); + UDF_SB_PARTTYPE(sb, i) = UDF_TYPE1_MAP15; + UDF_SB_PARTVSN(sb, i) = le16_to_cpu(gpm1->volSeqNum); + UDF_SB_PARTNUM(sb, i) = le16_to_cpu(gpm1->partitionNum); + UDF_SB_PARTFUNC(sb, i) = NULL; + } else if (type == 2) { + struct udfPartitionMap2 *upm2 = + (struct udfPartitionMap2 *)&(lvd-> + partitionMaps[offset]); + if (!strncmp + (upm2->partIdent.ident, UDF_ID_VIRTUAL, + strlen(UDF_ID_VIRTUAL))) { + if (le16_to_cpu + (((__le16 *) upm2->partIdent. + identSuffix)[0]) == 0x0150) { + UDF_SB_PARTTYPE(sb, i) = + UDF_VIRTUAL_MAP15; + UDF_SB_PARTFUNC(sb, i) = + udf_get_pblock_virt15; + } else + if (le16_to_cpu + (((__le16 *) upm2->partIdent. + identSuffix)[0]) == 0x0200) { + UDF_SB_PARTTYPE(sb, i) = + UDF_VIRTUAL_MAP20; + UDF_SB_PARTFUNC(sb, i) = + udf_get_pblock_virt20; } - } - else if (!strncmp(upm2->partIdent.ident, UDF_ID_SPARABLE, strlen(UDF_ID_SPARABLE))) - { + } else + if (!strncmp + (upm2->partIdent.ident, UDF_ID_SPARABLE, + strlen(UDF_ID_SPARABLE))) { uint32_t loc; uint16_t ident; struct sparingTable *st; - struct sparablePartitionMap *spm = (struct sparablePartitionMap *)&(lvd->partitionMaps[offset]); - - UDF_SB_PARTTYPE(sb,i) = UDF_SPARABLE_MAP15; - UDF_SB_TYPESPAR(sb,i).s_packet_len = le16_to_cpu(spm->packetLength); - for (j=0; j<spm->numSparingTables; j++) - { - loc = le32_to_cpu(spm->locSparingTable[j]); - UDF_SB_TYPESPAR(sb,i).s_spar_map[j] = - udf_read_tagged(sb, loc, loc, &ident); - if (UDF_SB_TYPESPAR(sb,i).s_spar_map[j] != NULL) - { - st = (struct sparingTable *)UDF_SB_TYPESPAR(sb,i).s_spar_map[j]->b_data; - if (ident != 0 || - strncmp(st->sparingIdent.ident, UDF_ID_SPARING, strlen(UDF_ID_SPARING))) + struct sparablePartitionMap *spm = + (struct sparablePartitionMap *)&(lvd-> + partitionMaps + [offset]); + + UDF_SB_PARTTYPE(sb, i) = UDF_SPARABLE_MAP15; + UDF_SB_TYPESPAR(sb, i).s_packet_len = + le16_to_cpu(spm->packetLength); + for (j = 0; j < spm->numSparingTables; j++) { + loc = + le32_to_cpu(spm-> + locSparingTable[j]); + UDF_SB_TYPESPAR(sb, i).s_spar_map[j] = + udf_read_tagged(sb, loc, loc, + &ident); + if (UDF_SB_TYPESPAR(sb, i). + s_spar_map[j] != NULL) { + st = (struct sparingTable *) + UDF_SB_TYPESPAR(sb, + i). + s_spar_map[j]->b_data; + if (ident != 0 + || strncmp(st->sparingIdent. + ident, + UDF_ID_SPARING, + strlen + (UDF_ID_SPARING))) { - brelse(UDF_SB_TYPESPAR(sb,i).s_spar_map[j]); - UDF_SB_TYPESPAR(sb,i).s_spar_map[j] = NULL; + brelse(UDF_SB_TYPESPAR + (sb, + i). + s_spar_map[j]); + UDF_SB_TYPESPAR(sb, + i). + s_spar_map[j] = + NULL; } } } - UDF_SB_PARTFUNC(sb,i) = udf_get_pblock_spar15; - } - else - { - udf_debug("Unknown ident: %s\n", upm2->partIdent.ident); + UDF_SB_PARTFUNC(sb, i) = udf_get_pblock_spar15; + } else { + udf_debug("Unknown ident: %s\n", + upm2->partIdent.ident); continue; } - UDF_SB_PARTVSN(sb,i) = le16_to_cpu(upm2->volSeqNum); - UDF_SB_PARTNUM(sb,i) = le16_to_cpu(upm2->partitionNum); + UDF_SB_PARTVSN(sb, i) = le16_to_cpu(upm2->volSeqNum); + UDF_SB_PARTNUM(sb, i) = le16_to_cpu(upm2->partitionNum); } udf_debug("Partition (%d:%d) type %d on volume %d\n", - i, UDF_SB_PARTNUM(sb,i), type, UDF_SB_PARTVSN(sb,i)); + i, UDF_SB_PARTNUM(sb, i), type, UDF_SB_PARTVSN(sb, + i)); } - if (fileset) - { - long_ad *la = (long_ad *)&(lvd->logicalVolContentsUse[0]); + if (fileset) { + long_ad *la = (long_ad *) & (lvd->logicalVolContentsUse[0]); *fileset = lelb_to_cpu(la->extLocation); - udf_debug("FileSet found in LogicalVolDesc at block=%d, partition=%d\n", - fileset->logicalBlockNum, - fileset->partitionReferenceNum); + udf_debug + ("FileSet found in LogicalVolDesc at block=%d, partition=%d\n", + fileset->logicalBlockNum, fileset->partitionReferenceNum); } if (lvd->integritySeqExt.extLength) udf_load_logicalvolint(sb, leea_to_cpu(lvd->integritySeqExt)); @@ -1118,26 +1207,26 @@ udf_load_logicalvol(struct super_block *sb, struct buffer_head * bh, kernel_lb_a * udf_load_logicalvolint * */ -static void -udf_load_logicalvolint(struct super_block *sb, kernel_extent_ad loc) +static void udf_load_logicalvolint(struct super_block *sb, kernel_extent_ad loc) { struct buffer_head *bh = NULL; uint16_t ident; while (loc.extLength > 0 && - (bh = udf_read_tagged(sb, loc.extLocation, - loc.extLocation, &ident)) && - ident == TAG_IDENT_LVID) - { + (bh = udf_read_tagged(sb, loc.extLocation, + loc.extLocation, &ident)) && + ident == TAG_IDENT_LVID) { UDF_SB_LVIDBH(sb) = bh; - + if (UDF_SB_LVID(sb)->nextIntegrityExt.extLength) - udf_load_logicalvolint(sb, leea_to_cpu(UDF_SB_LVID(sb)->nextIntegrityExt)); - + udf_load_logicalvolint(sb, + leea_to_cpu(UDF_SB_LVID(sb)-> + nextIntegrityExt)); + if (UDF_SB_LVIDBH(sb) != bh) brelse(bh); loc.extLength -= sb->s_blocksize; - loc.extLocation ++; + loc.extLocation++; } if (UDF_SB_LVIDBH(sb) != bh) brelse(bh); @@ -1158,15 +1247,16 @@ udf_load_logicalvolint(struct super_block *sb, kernel_extent_ad loc) * July 1, 1997 - Andrew E. Mileski * Written, tested, and released. */ -static int -udf_process_sequence(struct super_block *sb, long block, long lastblock, kernel_lb_addr *fileset) +static int +udf_process_sequence(struct super_block *sb, long block, long lastblock, + kernel_lb_addr * fileset) { struct buffer_head *bh = NULL; struct udf_vds_record vds[VDS_POS_LENGTH]; struct generic_desc *gd; struct volDescPtr *vdp; - int done=0; - int i,j; + int done = 0; + int i, j; uint32_t vdsn; uint16_t ident; long next_s = 0, next_e = 0; @@ -1174,93 +1264,92 @@ udf_process_sequence(struct super_block *sb, long block, long lastblock, kernel_ memset(vds, 0, sizeof(struct udf_vds_record) * VDS_POS_LENGTH); /* Read the main descriptor sequence */ - for (;(!done && block <= lastblock); block++) - { + for (; (!done && block <= lastblock); block++) { bh = udf_read_tagged(sb, block, block, &ident); - if (!bh) + if (!bh) break; /* Process each descriptor (ISO 13346 3/8.3-8.4) */ gd = (struct generic_desc *)bh->b_data; vdsn = le32_to_cpu(gd->volDescSeqNum); - switch (ident) - { - case TAG_IDENT_PVD: /* ISO 13346 3/10.1 */ - if (vdsn >= vds[VDS_POS_PRIMARY_VOL_DESC].volDescSeqNum) - { - vds[VDS_POS_PRIMARY_VOL_DESC].volDescSeqNum = vdsn; - vds[VDS_POS_PRIMARY_VOL_DESC].block = block; - } - break; - case TAG_IDENT_VDP: /* ISO 13346 3/10.3 */ - if (vdsn >= vds[VDS_POS_VOL_DESC_PTR].volDescSeqNum) - { - vds[VDS_POS_VOL_DESC_PTR].volDescSeqNum = vdsn; - vds[VDS_POS_VOL_DESC_PTR].block = block; - - vdp = (struct volDescPtr *)bh->b_data; - next_s = le32_to_cpu(vdp->nextVolDescSeqExt.extLocation); - next_e = le32_to_cpu(vdp->nextVolDescSeqExt.extLength); - next_e = next_e >> sb->s_blocksize_bits; - next_e += next_s; - } - break; - case TAG_IDENT_IUVD: /* ISO 13346 3/10.4 */ - if (vdsn >= vds[VDS_POS_IMP_USE_VOL_DESC].volDescSeqNum) - { - vds[VDS_POS_IMP_USE_VOL_DESC].volDescSeqNum = vdsn; - vds[VDS_POS_IMP_USE_VOL_DESC].block = block; - } - break; - case TAG_IDENT_PD: /* ISO 13346 3/10.5 */ - if (!vds[VDS_POS_PARTITION_DESC].block) - vds[VDS_POS_PARTITION_DESC].block = block; - break; - case TAG_IDENT_LVD: /* ISO 13346 3/10.6 */ - if (vdsn >= vds[VDS_POS_LOGICAL_VOL_DESC].volDescSeqNum) - { - vds[VDS_POS_LOGICAL_VOL_DESC].volDescSeqNum = vdsn; - vds[VDS_POS_LOGICAL_VOL_DESC].block = block; - } - break; - case TAG_IDENT_USD: /* ISO 13346 3/10.8 */ - if (vdsn >= vds[VDS_POS_UNALLOC_SPACE_DESC].volDescSeqNum) - { - vds[VDS_POS_UNALLOC_SPACE_DESC].volDescSeqNum = vdsn; - vds[VDS_POS_UNALLOC_SPACE_DESC].block = block; - } - break; - case TAG_IDENT_TD: /* ISO 13346 3/10.9 */ - vds[VDS_POS_TERMINATING_DESC].block = block; - if (next_e) - { - block = next_s; - lastblock = next_e; - next_s = next_e = 0; - } - else - done = 1; - break; + switch (ident) { + case TAG_IDENT_PVD: /* ISO 13346 3/10.1 */ + if (vdsn >= vds[VDS_POS_PRIMARY_VOL_DESC].volDescSeqNum) { + vds[VDS_POS_PRIMARY_VOL_DESC].volDescSeqNum = + vdsn; + vds[VDS_POS_PRIMARY_VOL_DESC].block = block; + } + break; + case TAG_IDENT_VDP: /* ISO 13346 3/10.3 */ + if (vdsn >= vds[VDS_POS_VOL_DESC_PTR].volDescSeqNum) { + vds[VDS_POS_VOL_DESC_PTR].volDescSeqNum = vdsn; + vds[VDS_POS_VOL_DESC_PTR].block = block; + + vdp = (struct volDescPtr *)bh->b_data; + next_s = + le32_to_cpu(vdp->nextVolDescSeqExt. + extLocation); + next_e = + le32_to_cpu(vdp->nextVolDescSeqExt. + extLength); + next_e = next_e >> sb->s_blocksize_bits; + next_e += next_s; + } + break; + case TAG_IDENT_IUVD: /* ISO 13346 3/10.4 */ + if (vdsn >= vds[VDS_POS_IMP_USE_VOL_DESC].volDescSeqNum) { + vds[VDS_POS_IMP_USE_VOL_DESC].volDescSeqNum = + vdsn; + vds[VDS_POS_IMP_USE_VOL_DESC].block = block; + } + break; + case TAG_IDENT_PD: /* ISO 13346 3/10.5 */ + if (!vds[VDS_POS_PARTITION_DESC].block) + vds[VDS_POS_PARTITION_DESC].block = block; + break; + case TAG_IDENT_LVD: /* ISO 13346 3/10.6 */ + if (vdsn >= vds[VDS_POS_LOGICAL_VOL_DESC].volDescSeqNum) { + vds[VDS_POS_LOGICAL_VOL_DESC].volDescSeqNum = + vdsn; + vds[VDS_POS_LOGICAL_VOL_DESC].block = block; + } + break; + case TAG_IDENT_USD: /* ISO 13346 3/10.8 */ + if (vdsn >= + vds[VDS_POS_UNALLOC_SPACE_DESC].volDescSeqNum) { + vds[VDS_POS_UNALLOC_SPACE_DESC].volDescSeqNum = + vdsn; + vds[VDS_POS_UNALLOC_SPACE_DESC].block = block; + } + break; + case TAG_IDENT_TD: /* ISO 13346 3/10.9 */ + vds[VDS_POS_TERMINATING_DESC].block = block; + if (next_e) { + block = next_s; + lastblock = next_e; + next_s = next_e = 0; + } else + done = 1; + break; } brelse(bh); } - for (i=0; i<VDS_POS_LENGTH; i++) - { - if (vds[i].block) - { - bh = udf_read_tagged(sb, vds[i].block, vds[i].block, &ident); + for (i = 0; i < VDS_POS_LENGTH; i++) { + if (vds[i].block) { + bh = udf_read_tagged(sb, vds[i].block, vds[i].block, + &ident); if (i == VDS_POS_PRIMARY_VOL_DESC) udf_load_pvoldesc(sb, bh); else if (i == VDS_POS_LOGICAL_VOL_DESC) udf_load_logicalvol(sb, bh, fileset); - else if (i == VDS_POS_PARTITION_DESC) - { + else if (i == VDS_POS_PARTITION_DESC) { struct buffer_head *bh2 = NULL; udf_load_partdesc(sb, bh); - for (j=vds[i].block+1; j<vds[VDS_POS_TERMINATING_DESC].block; j++) - { + for (j = vds[i].block + 1; + j < vds[VDS_POS_TERMINATING_DESC].block; + j++) { bh2 = udf_read_tagged(sb, j, j, &ident); gd = (struct generic_desc *)bh2->b_data; if (ident == TAG_IDENT_PD) @@ -1278,31 +1367,27 @@ udf_process_sequence(struct super_block *sb, long block, long lastblock, kernel_ /* * udf_check_valid() */ -static int -udf_check_valid(struct super_block *sb, int novrs, int silent) +static int udf_check_valid(struct super_block *sb, int novrs, int silent) { long block; - if (novrs) - { + if (novrs) { udf_debug("Validity check skipped because of novrs option\n"); return 0; } /* Check that it is NSR02 compliant */ /* Process any "CD-ROM Volume Descriptor Set" (ECMA 167 2/8.3.1) */ - else if ((block = udf_vrs(sb, silent)) == -1) - { - udf_debug("Failed to read byte 32768. Assuming open disc. Skipping validity check\n"); + else if ((block = udf_vrs(sb, silent)) == -1) { + udf_debug + ("Failed to read byte 32768. Assuming open disc. Skipping validity check\n"); if (!UDF_SB_LASTBLOCK(sb)) UDF_SB_LASTBLOCK(sb) = udf_get_last_block(sb); return 0; - } - else + } else return !block; } -static int -udf_load_partition(struct super_block *sb, kernel_lb_addr *fileset) +static int udf_load_partition(struct super_block *sb, kernel_lb_addr * fileset) { struct anchorVolDescPtr *anchor; uint16_t ident; @@ -1315,19 +1400,27 @@ udf_load_partition(struct super_block *sb, kernel_lb_addr *fileset) for (i = 0; i < ARRAY_SIZE(UDF_SB_ANCHOR(sb)); i++) { if (UDF_SB_ANCHOR(sb)[i] && (bh = udf_read_tagged(sb, - UDF_SB_ANCHOR(sb)[i], UDF_SB_ANCHOR(sb)[i], &ident))) - { + UDF_SB_ANCHOR + (sb)[i], + UDF_SB_ANCHOR + (sb)[i], + &ident))) { anchor = (struct anchorVolDescPtr *)bh->b_data; /* Locate the main sequence */ - main_s = le32_to_cpu( anchor->mainVolDescSeqExt.extLocation ); - main_e = le32_to_cpu( anchor->mainVolDescSeqExt.extLength ); + main_s = + le32_to_cpu(anchor->mainVolDescSeqExt.extLocation); + main_e = + le32_to_cpu(anchor->mainVolDescSeqExt.extLength); main_e = main_e >> sb->s_blocksize_bits; main_e += main_s; /* Locate the reserve sequence */ - reserve_s = le32_to_cpu(anchor->reserveVolDescSeqExt.extLocation); - reserve_e = le32_to_cpu(anchor->reserveVolDescSeqExt.extLength); + reserve_s = + le32_to_cpu(anchor->reserveVolDescSeqExt. + extLocation); + reserve_e = + le32_to_cpu(anchor->reserveVolDescSeqExt.extLength); reserve_e = reserve_e >> sb->s_blocksize_bits; reserve_e += reserve_s; @@ -1335,9 +1428,10 @@ udf_load_partition(struct super_block *sb, kernel_lb_addr *fileset) /* Process the main & reserve sequences */ /* responsible for finding the PartitionDesc(s) */ - if (!(udf_process_sequence(sb, main_s, main_e, fileset) && - udf_process_sequence(sb, reserve_s, reserve_e, fileset))) - { + if (! + (udf_process_sequence(sb, main_s, main_e, fileset) + && udf_process_sequence(sb, reserve_s, reserve_e, + fileset))) { break; } } @@ -1349,36 +1443,37 @@ udf_load_partition(struct super_block *sb, kernel_lb_addr *fileset) } else udf_debug("Using anchor in block %d\n", UDF_SB_ANCHOR(sb)[i]); - for (i=0; i<UDF_SB_NUMPARTS(sb); i++) - { - switch (UDF_SB_PARTTYPE(sb, i)) - { - case UDF_VIRTUAL_MAP15: - case UDF_VIRTUAL_MAP20: + for (i = 0; i < UDF_SB_NUMPARTS(sb); i++) { + switch (UDF_SB_PARTTYPE(sb, i)) { + case UDF_VIRTUAL_MAP15: + case UDF_VIRTUAL_MAP20: { kernel_lb_addr uninitialized_var(ino); - if (!UDF_SB_LASTBLOCK(sb)) - { - UDF_SB_LASTBLOCK(sb) = udf_get_last_block(sb); + if (!UDF_SB_LASTBLOCK(sb)) { + UDF_SB_LASTBLOCK(sb) = + udf_get_last_block(sb); udf_find_anchor(sb); } - if (!UDF_SB_LASTBLOCK(sb)) - { - udf_debug("Unable to determine Lastblock (For Virtual Partition)\n"); + if (!UDF_SB_LASTBLOCK(sb)) { + udf_debug + ("Unable to determine Lastblock (For Virtual Partition)\n"); return 1; } - for (j=0; j<UDF_SB_NUMPARTS(sb); j++) - { + for (j = 0; j < UDF_SB_NUMPARTS(sb); j++) { if (j != i && - UDF_SB_PARTVSN(sb,i) == UDF_SB_PARTVSN(sb,j) && - UDF_SB_PARTNUM(sb,i) == UDF_SB_PARTNUM(sb,j)) - { + UDF_SB_PARTVSN(sb, + i) == + UDF_SB_PARTVSN(sb, j) + && UDF_SB_PARTNUM(sb, + i) == + UDF_SB_PARTNUM(sb, j)) { ino.partitionReferenceNum = j; - ino.logicalBlockNum = UDF_SB_LASTBLOCK(sb) - - UDF_SB_PARTROOT(sb,j); + ino.logicalBlockNum = + UDF_SB_LASTBLOCK(sb) - + UDF_SB_PARTROOT(sb, j); break; } } @@ -1389,13 +1484,13 @@ udf_load_partition(struct super_block *sb, kernel_lb_addr *fileset) if (!(UDF_SB_VAT(sb) = udf_iget(sb, ino))) return 1; - if (UDF_SB_PARTTYPE(sb,i) == UDF_VIRTUAL_MAP15) - { - UDF_SB_TYPEVIRT(sb,i).s_start_offset = udf_ext0_offset(UDF_SB_VAT(sb)); - UDF_SB_TYPEVIRT(sb,i).s_num_entries = (UDF_SB_VAT(sb)->i_size - 36) >> 2; - } - else if (UDF_SB_PARTTYPE(sb,i) == UDF_VIRTUAL_MAP20) - { + if (UDF_SB_PARTTYPE(sb, i) == UDF_VIRTUAL_MAP15) { + UDF_SB_TYPEVIRT(sb, i).s_start_offset = + udf_ext0_offset(UDF_SB_VAT(sb)); + UDF_SB_TYPEVIRT(sb, i).s_num_entries = + (UDF_SB_VAT(sb)->i_size - 36) >> 2; + } else if (UDF_SB_PARTTYPE(sb, i) == + UDF_VIRTUAL_MAP20) { struct buffer_head *bh = NULL; uint32_t pos; @@ -1403,15 +1498,26 @@ udf_load_partition(struct super_block *sb, kernel_lb_addr *fileset) bh = sb_bread(sb, pos); if (!bh) return 1; - UDF_SB_TYPEVIRT(sb,i).s_start_offset = - le16_to_cpu(((struct virtualAllocationTable20 *)bh->b_data + udf_ext0_offset(UDF_SB_VAT(sb)))->lengthHeader) + - udf_ext0_offset(UDF_SB_VAT(sb)); - UDF_SB_TYPEVIRT(sb,i).s_num_entries = (UDF_SB_VAT(sb)->i_size - - UDF_SB_TYPEVIRT(sb,i).s_start_offset) >> 2; + UDF_SB_TYPEVIRT(sb, i).s_start_offset = + le16_to_cpu(((struct + virtualAllocationTable20 + *)bh->b_data + + udf_ext0_offset + (UDF_SB_VAT(sb)))-> + lengthHeader) + + udf_ext0_offset(UDF_SB_VAT(sb)); + UDF_SB_TYPEVIRT(sb, i).s_num_entries = + (UDF_SB_VAT(sb)->i_size - + UDF_SB_TYPEVIRT(sb, + i). + s_start_offset) >> 2; brelse(bh); } - UDF_SB_PARTROOT(sb,i) = udf_get_pblock(sb, 0, i, 0); - UDF_SB_PARTLEN(sb,i) = UDF_SB_PARTLEN(sb,ino.partitionReferenceNum); + UDF_SB_PARTROOT(sb, i) = + udf_get_pblock(sb, 0, i, 0); + UDF_SB_PARTLEN(sb, i) = + UDF_SB_PARTLEN(sb, + ino.partitionReferenceNum); } } } @@ -1420,26 +1526,28 @@ udf_load_partition(struct super_block *sb, kernel_lb_addr *fileset) static void udf_open_lvid(struct super_block *sb) { - if (UDF_SB_LVIDBH(sb)) - { + if (UDF_SB_LVIDBH(sb)) { int i; kernel_timestamp cpu_time; UDF_SB_LVIDIU(sb)->impIdent.identSuffix[0] = UDF_OS_CLASS_UNIX; UDF_SB_LVIDIU(sb)->impIdent.identSuffix[1] = UDF_OS_ID_LINUX; if (udf_time_to_stamp(&cpu_time, CURRENT_TIME)) - UDF_SB_LVID(sb)->recordingDateAndTime = cpu_to_lets(cpu_time); + UDF_SB_LVID(sb)->recordingDateAndTime = + cpu_to_lets(cpu_time); UDF_SB_LVID(sb)->integrityType = LVID_INTEGRITY_TYPE_OPEN; UDF_SB_LVID(sb)->descTag.descCRC = - cpu_to_le16(udf_crc((char *)UDF_SB_LVID(sb) + sizeof(tag), - le16_to_cpu(UDF_SB_LVID(sb)->descTag.descCRCLength), 0)); + cpu_to_le16(udf_crc((char *)UDF_SB_LVID(sb) + sizeof(tag), + le16_to_cpu(UDF_SB_LVID(sb)->descTag. + descCRCLength), 0)); UDF_SB_LVID(sb)->descTag.tagChecksum = 0; - for (i=0; i<16; i++) + for (i = 0; i < 16; i++) if (i != 4) UDF_SB_LVID(sb)->descTag.tagChecksum += - ((uint8_t *)&(UDF_SB_LVID(sb)->descTag))[i]; + ((uint8_t *) & + (UDF_SB_LVID(sb)->descTag))[i]; mark_buffer_dirty(UDF_SB_LVIDBH(sb)); } @@ -1448,32 +1556,41 @@ static void udf_open_lvid(struct super_block *sb) static void udf_close_lvid(struct super_block *sb) { if (UDF_SB_LVIDBH(sb) && - UDF_SB_LVID(sb)->integrityType == LVID_INTEGRITY_TYPE_OPEN) - { + UDF_SB_LVID(sb)->integrityType == LVID_INTEGRITY_TYPE_OPEN) { int i; kernel_timestamp cpu_time; UDF_SB_LVIDIU(sb)->impIdent.identSuffix[0] = UDF_OS_CLASS_UNIX; UDF_SB_LVIDIU(sb)->impIdent.identSuffix[1] = UDF_OS_ID_LINUX; if (udf_time_to_stamp(&cpu_time, CURRENT_TIME)) - UDF_SB_LVID(sb)->recordingDateAndTime = cpu_to_lets(cpu_time); - if (UDF_MAX_WRITE_VERSION > le16_to_cpu(UDF_SB_LVIDIU(sb)->maxUDFWriteRev)) - UDF_SB_LVIDIU(sb)->maxUDFWriteRev = cpu_to_le16(UDF_MAX_WRITE_VERSION); - if (UDF_SB_UDFREV(sb) > le16_to_cpu(UDF_SB_LVIDIU(sb)->minUDFReadRev)) - UDF_SB_LVIDIU(sb)->minUDFReadRev = cpu_to_le16(UDF_SB_UDFREV(sb)); - if (UDF_SB_UDFREV(sb) > le16_to_cpu(UDF_SB_LVIDIU(sb)->minUDFWriteRev)) - UDF_SB_LVIDIU(sb)->minUDFWriteRev = cpu_to_le16(UDF_SB_UDFREV(sb)); - UDF_SB_LVID(sb)->integrityType = cpu_to_le32(LVID_INTEGRITY_TYPE_CLOSE); + UDF_SB_LVID(sb)->recordingDateAndTime = + cpu_to_lets(cpu_time); + if (UDF_MAX_WRITE_VERSION > + le16_to_cpu(UDF_SB_LVIDIU(sb)->maxUDFWriteRev)) + UDF_SB_LVIDIU(sb)->maxUDFWriteRev = + cpu_to_le16(UDF_MAX_WRITE_VERSION); + if (UDF_SB_UDFREV(sb) > + le16_to_cpu(UDF_SB_LVIDIU(sb)->minUDFReadRev)) + UDF_SB_LVIDIU(sb)->minUDFReadRev = + cpu_to_le16(UDF_SB_UDFREV(sb)); + if (UDF_SB_UDFREV(sb) > + le16_to_cpu(UDF_SB_LVIDIU(sb)->minUDFWriteRev)) + UDF_SB_LVIDIU(sb)->minUDFWriteRev = + cpu_to_le16(UDF_SB_UDFREV(sb)); + UDF_SB_LVID(sb)->integrityType = + cpu_to_le32(LVID_INTEGRITY_TYPE_CLOSE); UDF_SB_LVID(sb)->descTag.descCRC = - cpu_to_le16(udf_crc((char *)UDF_SB_LVID(sb) + sizeof(tag), - le16_to_cpu(UDF_SB_LVID(sb)->descTag.descCRCLength), 0)); + cpu_to_le16(udf_crc((char *)UDF_SB_LVID(sb) + sizeof(tag), + le16_to_cpu(UDF_SB_LVID(sb)->descTag. + descCRCLength), 0)); UDF_SB_LVID(sb)->descTag.tagChecksum = 0; - for (i=0; i<16; i++) + for (i = 0; i < 16; i++) if (i != 4) UDF_SB_LVID(sb)->descTag.tagChecksum += - ((uint8_t *)&(UDF_SB_LVID(sb)->descTag))[i]; + ((uint8_t *) & + (UDF_SB_LVID(sb)->descTag))[i]; mark_buffer_dirty(UDF_SB_LVIDBH(sb)); } @@ -1498,7 +1615,7 @@ static void udf_close_lvid(struct super_block *sb) static int udf_fill_super(struct super_block *sb, void *options, int silent) { int i; - struct inode *inode=NULL; + struct inode *inode = NULL; struct udf_options uopt; kernel_lb_addr rootdir, fileset; struct udf_sb_info *sbi; @@ -1520,15 +1637,13 @@ static int udf_fill_super(struct super_block *sb, void *options, int silent) goto error_out; if (uopt.flags & (1 << UDF_FLAG_UTF8) && - uopt.flags & (1 << UDF_FLAG_NLS_MAP)) - { + uopt.flags & (1 << UDF_FLAG_NLS_MAP)) { udf_error(sb, "udf_read_super", - "utf8 cannot be combined with iocharset\n"); + "utf8 cannot be combined with iocharset\n"); goto error_out; } #ifdef CONFIG_UDF_NLS - if ((uopt.flags & (1 << UDF_FLAG_NLS_MAP)) && !uopt.nls_map) - { + if ((uopt.flags & (1 << UDF_FLAG_NLS_MAP)) && !uopt.nls_map) { uopt.nls_map = load_nls_default(); if (!uopt.nls_map) uopt.flags &= ~(1 << UDF_FLAG_NLS_MAP); @@ -1552,7 +1667,7 @@ static int udf_fill_super(struct super_block *sb, void *options, int silent) if (!udf_set_blocksize(sb, uopt.blocksize)) goto error_out; - if ( uopt.session == 0xFFFFFFFF ) + if (uopt.session == 0xFFFFFFFF) UDF_SB_SESSION(sb) = udf_get_last_session(sb); else UDF_SB_SESSION(sb) = uopt.session; @@ -1564,10 +1679,9 @@ static int udf_fill_super(struct super_block *sb, void *options, int silent) UDF_SB_ANCHOR(sb)[2] = uopt.anchor; UDF_SB_ANCHOR(sb)[3] = 256; - if (udf_check_valid(sb, uopt.novrs, silent)) /* read volume recognition sequences */ - { + if (udf_check_valid(sb, uopt.novrs, silent)) { /* read volume recognition sequences */ printk("UDF-fs: No VRS found\n"); - goto error_out; + goto error_out; } udf_find_anchor(sb); @@ -1579,29 +1693,26 @@ static int udf_fill_super(struct super_block *sb, void *options, int silent) sb->s_magic = UDF_SUPER_MAGIC; sb->s_time_gran = 1000; - if (udf_load_partition(sb, &fileset)) - { + if (udf_load_partition(sb, &fileset)) { printk("UDF-fs: No partition found (1)\n"); goto error_out; } udf_debug("Lastblock=%d\n", UDF_SB_LASTBLOCK(sb)); - if ( UDF_SB_LVIDBH(sb) ) - { - uint16_t minUDFReadRev = le16_to_cpu(UDF_SB_LVIDIU(sb)->minUDFReadRev); - uint16_t minUDFWriteRev = le16_to_cpu(UDF_SB_LVIDIU(sb)->minUDFWriteRev); + if (UDF_SB_LVIDBH(sb)) { + uint16_t minUDFReadRev = + le16_to_cpu(UDF_SB_LVIDIU(sb)->minUDFReadRev); + uint16_t minUDFWriteRev = + le16_to_cpu(UDF_SB_LVIDIU(sb)->minUDFWriteRev); /* uint16_t maxUDFWriteRev = le16_to_cpu(UDF_SB_LVIDIU(sb)->maxUDFWriteRev); */ - if (minUDFReadRev > UDF_MAX_READ_VERSION) - { + if (minUDFReadRev > UDF_MAX_READ_VERSION) { printk("UDF-fs: minUDFReadRev=%x (max is %x)\n", - le16_to_cpu(UDF_SB_LVIDIU(sb)->minUDFReadRev), - UDF_MAX_READ_VERSION); + le16_to_cpu(UDF_SB_LVIDIU(sb)->minUDFReadRev), + UDF_MAX_READ_VERSION); goto error_out; - } - else if (minUDFWriteRev > UDF_MAX_WRITE_VERSION) - { + } else if (minUDFWriteRev > UDF_MAX_WRITE_VERSION) { sb->s_flags |= MS_RDONLY; } @@ -1613,31 +1724,30 @@ static int udf_fill_super(struct super_block *sb, void *options, int silent) UDF_SET_FLAG(sb, UDF_FLAG_USE_STREAMS); } - if ( !UDF_SB_NUMPARTS(sb) ) - { + if (!UDF_SB_NUMPARTS(sb)) { printk("UDF-fs: No partition found (2)\n"); goto error_out; } - if (UDF_SB_PARTFLAGS(sb, UDF_SB_PARTITION(sb)) & UDF_PART_FLAG_READ_ONLY) { - printk("UDF-fs: Partition marked readonly; forcing readonly mount\n"); + if (UDF_SB_PARTFLAGS(sb, UDF_SB_PARTITION(sb)) & + UDF_PART_FLAG_READ_ONLY) { + printk + ("UDF-fs: Partition marked readonly; forcing readonly mount\n"); sb->s_flags |= MS_RDONLY; } - if ( udf_find_fileset(sb, &fileset, &rootdir) ) - { + if (udf_find_fileset(sb, &fileset, &rootdir)) { printk("UDF-fs: No fileset found\n"); goto error_out; } - if (!silent) - { + if (!silent) { kernel_timestamp ts; udf_time_to_stamp(&ts, UDF_SB_RECORDTIME(sb)); - udf_info("UDF %s (%s) Mounting volume '%s', timestamp %04u/%02u/%02u %02u:%02u (%x)\n", - UDFFS_VERSION, UDFFS_DATE, - UDF_SB_VOLIDENT(sb), ts.year, ts.month, ts.day, ts.hour, ts.minute, - ts.typeAndTimezone); + udf_info + ("UDF %s (%s) Mounting volume '%s', timestamp %04u/%02u/%02u %02u:%02u (%x)\n", + UDFFS_VERSION, UDFFS_DATE, UDF_SB_VOLIDENT(sb), ts.year, + ts.month, ts.day, ts.hour, ts.minute, ts.typeAndTimezone); } if (!(sb->s_flags & MS_RDONLY)) udf_open_lvid(sb); @@ -1645,18 +1755,16 @@ static int udf_fill_super(struct super_block *sb, void *options, int silent) /* Assign the root inode */ /* assign inodes by physical block number */ /* perhaps it's not extensible enough, but for now ... */ - inode = udf_iget(sb, rootdir); - if (!inode) - { + inode = udf_iget(sb, rootdir); + if (!inode) { printk("UDF-fs: Error in udf_iget, block=%d, partition=%d\n", - rootdir.logicalBlockNum, rootdir.partitionReferenceNum); + rootdir.logicalBlockNum, rootdir.partitionReferenceNum); goto error_out; } /* Allocate a dentry for the root inode */ sb->s_root = d_alloc_root(inode); - if (!sb->s_root) - { + if (!sb->s_root) { printk("UDF-fs: Couldn't allocate root dentry\n"); iput(inode); goto error_out; @@ -1664,23 +1772,30 @@ static int udf_fill_super(struct super_block *sb, void *options, int silent) sb->s_maxbytes = MAX_LFS_FILESIZE; return 0; -error_out: + error_out: if (UDF_SB_VAT(sb)) iput(UDF_SB_VAT(sb)); - if (UDF_SB_NUMPARTS(sb)) - { - if (UDF_SB_PARTFLAGS(sb, UDF_SB_PARTITION(sb)) & UDF_PART_FLAG_UNALLOC_TABLE) - iput(UDF_SB_PARTMAPS(sb)[UDF_SB_PARTITION(sb)].s_uspace.s_table); - if (UDF_SB_PARTFLAGS(sb, UDF_SB_PARTITION(sb)) & UDF_PART_FLAG_FREED_TABLE) - iput(UDF_SB_PARTMAPS(sb)[UDF_SB_PARTITION(sb)].s_fspace.s_table); - if (UDF_SB_PARTFLAGS(sb, UDF_SB_PARTITION(sb)) & UDF_PART_FLAG_UNALLOC_BITMAP) - UDF_SB_FREE_BITMAP(sb,UDF_SB_PARTITION(sb),s_uspace); - if (UDF_SB_PARTFLAGS(sb, UDF_SB_PARTITION(sb)) & UDF_PART_FLAG_FREED_BITMAP) - UDF_SB_FREE_BITMAP(sb,UDF_SB_PARTITION(sb),s_fspace); - if (UDF_SB_PARTTYPE(sb, UDF_SB_PARTITION(sb)) == UDF_SPARABLE_MAP15) - { - for (i=0; i<4; i++) - brelse(UDF_SB_TYPESPAR(sb, UDF_SB_PARTITION(sb)).s_spar_map[i]); + if (UDF_SB_NUMPARTS(sb)) { + if (UDF_SB_PARTFLAGS(sb, UDF_SB_PARTITION(sb)) & + UDF_PART_FLAG_UNALLOC_TABLE) + iput(UDF_SB_PARTMAPS(sb)[UDF_SB_PARTITION(sb)].s_uspace. + s_table); + if (UDF_SB_PARTFLAGS(sb, UDF_SB_PARTITION(sb)) & + UDF_PART_FLAG_FREED_TABLE) + iput(UDF_SB_PARTMAPS(sb)[UDF_SB_PARTITION(sb)].s_fspace. + s_table); + if (UDF_SB_PARTFLAGS(sb, UDF_SB_PARTITION(sb)) & + UDF_PART_FLAG_UNALLOC_BITMAP) + UDF_SB_FREE_BITMAP(sb, UDF_SB_PARTITION(sb), s_uspace); + if (UDF_SB_PARTFLAGS(sb, UDF_SB_PARTITION(sb)) & + UDF_PART_FLAG_FREED_BITMAP) + UDF_SB_FREE_BITMAP(sb, UDF_SB_PARTITION(sb), s_fspace); + if (UDF_SB_PARTTYPE(sb, UDF_SB_PARTITION(sb)) == + UDF_SPARABLE_MAP15) { + for (i = 0; i < 4; i++) + brelse(UDF_SB_TYPESPAR + (sb, + UDF_SB_PARTITION(sb)).s_spar_map[i]); } } #ifdef CONFIG_UDF_NLS @@ -1697,32 +1812,31 @@ error_out: } void udf_error(struct super_block *sb, const char *function, - const char *fmt, ...) + const char *fmt, ...) { va_list args; - if (!(sb->s_flags & MS_RDONLY)) - { + if (!(sb->s_flags & MS_RDONLY)) { /* mark sb error */ sb->s_dirt = 1; } va_start(args, fmt); vsnprintf(error_buf, sizeof(error_buf), fmt, args); va_end(args); - printk (KERN_CRIT "UDF-fs error (device %s): %s: %s\n", - sb->s_id, function, error_buf); + printk(KERN_CRIT "UDF-fs error (device %s): %s: %s\n", + sb->s_id, function, error_buf); } void udf_warning(struct super_block *sb, const char *function, - const char *fmt, ...) + const char *fmt, ...) { va_list args; - va_start (args, fmt); + va_start(args, fmt); vsnprintf(error_buf, sizeof(error_buf), fmt, args); va_end(args); printk(KERN_WARNING "UDF-fs warning (device %s): %s: %s\n", - sb->s_id, function, error_buf); + sb->s_id, function, error_buf); } /* @@ -1738,27 +1852,33 @@ void udf_warning(struct super_block *sb, const char *function, * July 1, 1997 - Andrew E. Mileski * Written, tested, and released. */ -static void -udf_put_super(struct super_block *sb) +static void udf_put_super(struct super_block *sb) { int i; if (UDF_SB_VAT(sb)) iput(UDF_SB_VAT(sb)); - if (UDF_SB_NUMPARTS(sb)) - { - if (UDF_SB_PARTFLAGS(sb, UDF_SB_PARTITION(sb)) & UDF_PART_FLAG_UNALLOC_TABLE) - iput(UDF_SB_PARTMAPS(sb)[UDF_SB_PARTITION(sb)].s_uspace.s_table); - if (UDF_SB_PARTFLAGS(sb, UDF_SB_PARTITION(sb)) & UDF_PART_FLAG_FREED_TABLE) - iput(UDF_SB_PARTMAPS(sb)[UDF_SB_PARTITION(sb)].s_fspace.s_table); - if (UDF_SB_PARTFLAGS(sb, UDF_SB_PARTITION(sb)) & UDF_PART_FLAG_UNALLOC_BITMAP) - UDF_SB_FREE_BITMAP(sb,UDF_SB_PARTITION(sb),s_uspace); - if (UDF_SB_PARTFLAGS(sb, UDF_SB_PARTITION(sb)) & UDF_PART_FLAG_FREED_BITMAP) - UDF_SB_FREE_BITMAP(sb,UDF_SB_PARTITION(sb),s_fspace); - if (UDF_SB_PARTTYPE(sb, UDF_SB_PARTITION(sb)) == UDF_SPARABLE_MAP15) - { - for (i=0; i<4; i++) - brelse(UDF_SB_TYPESPAR(sb, UDF_SB_PARTITION(sb)).s_spar_map[i]); + if (UDF_SB_NUMPARTS(sb)) { + if (UDF_SB_PARTFLAGS(sb, UDF_SB_PARTITION(sb)) & + UDF_PART_FLAG_UNALLOC_TABLE) + iput(UDF_SB_PARTMAPS(sb)[UDF_SB_PARTITION(sb)].s_uspace. + s_table); + if (UDF_SB_PARTFLAGS(sb, UDF_SB_PARTITION(sb)) & + UDF_PART_FLAG_FREED_TABLE) + iput(UDF_SB_PARTMAPS(sb)[UDF_SB_PARTITION(sb)].s_fspace. + s_table); + if (UDF_SB_PARTFLAGS(sb, UDF_SB_PARTITION(sb)) & + UDF_PART_FLAG_UNALLOC_BITMAP) + UDF_SB_FREE_BITMAP(sb, UDF_SB_PARTITION(sb), s_uspace); + if (UDF_SB_PARTFLAGS(sb, UDF_SB_PARTITION(sb)) & + UDF_PART_FLAG_FREED_BITMAP) + UDF_SB_FREE_BITMAP(sb, UDF_SB_PARTITION(sb), s_fspace); + if (UDF_SB_PARTTYPE(sb, UDF_SB_PARTITION(sb)) == + UDF_SPARABLE_MAP15) { + for (i = 0; i < 4; i++) + brelse(UDF_SB_TYPESPAR + (sb, + UDF_SB_PARTITION(sb)).s_spar_map[i]); } } #ifdef CONFIG_UDF_NLS @@ -1786,8 +1906,7 @@ udf_put_super(struct super_block *sb) * July 1, 1997 - Andrew E. Mileski * Written, tested, and released. */ -static int -udf_statfs(struct dentry *dentry, struct kstatfs *buf) +static int udf_statfs(struct dentry *dentry, struct kstatfs *buf) { struct super_block *sb = dentry->d_sb; @@ -1797,11 +1916,12 @@ udf_statfs(struct dentry *dentry, struct kstatfs *buf) buf->f_bfree = udf_count_free(sb); buf->f_bavail = buf->f_bfree; buf->f_files = (UDF_SB_LVIDBH(sb) ? - (le32_to_cpu(UDF_SB_LVIDIU(sb)->numFiles) + - le32_to_cpu(UDF_SB_LVIDIU(sb)->numDirs)) : 0) + buf->f_bfree; + (le32_to_cpu(UDF_SB_LVIDIU(sb)->numFiles) + + le32_to_cpu(UDF_SB_LVIDIU(sb)->numDirs)) : 0) + + buf->f_bfree; buf->f_ffree = buf->f_bfree; /* __kernel_fsid_t f_fsid */ - buf->f_namelen = UDF_NAME_LEN-2; + buf->f_namelen = UDF_NAME_LEN - 2; return 0; } @@ -1830,13 +1950,10 @@ udf_count_free_bitmap(struct super_block *sb, struct udf_bitmap *bitmap) loc.partitionReferenceNum = UDF_SB_PARTITION(sb); bh = udf_read_ptagged(sb, loc, 0, &ident); - if (!bh) - { + if (!bh) { printk(KERN_ERR "udf: udf_count_free failed\n"); goto out; - } - else if (ident != TAG_IDENT_SBD) - { + } else if (ident != TAG_IDENT_SBD) { brelse(bh); printk(KERN_ERR "udf: udf_count_free failed\n"); goto out; @@ -1844,43 +1961,39 @@ udf_count_free_bitmap(struct super_block *sb, struct udf_bitmap *bitmap) bm = (struct spaceBitmapDesc *)bh->b_data; bytes = le32_to_cpu(bm->numOfBytes); - index = sizeof(struct spaceBitmapDesc); /* offset in first block only */ - ptr = (uint8_t *)bh->b_data; + index = sizeof(struct spaceBitmapDesc); /* offset in first block only */ + ptr = (uint8_t *) bh->b_data; - while ( bytes > 0 ) - { - while ((bytes > 0) && (index < sb->s_blocksize)) - { + while (bytes > 0) { + while ((bytes > 0) && (index < sb->s_blocksize)) { value = ptr[index]; - accum += udf_bitmap_lookup[ value & 0x0f ]; - accum += udf_bitmap_lookup[ value >> 4 ]; + accum += udf_bitmap_lookup[value & 0x0f]; + accum += udf_bitmap_lookup[value >> 4]; index++; bytes--; } - if ( bytes ) - { + if (bytes) { brelse(bh); newblock = udf_get_lb_pblock(sb, loc, ++block); bh = udf_tread(sb, newblock); - if (!bh) - { + if (!bh) { udf_debug("read failed\n"); goto out; } index = 0; - ptr = (uint8_t *)bh->b_data; + ptr = (uint8_t *) bh->b_data; } } brelse(bh); -out: + out: unlock_kernel(); return accum; } static unsigned int -udf_count_free_table(struct super_block *sb, struct inode * table) +udf_count_free_table(struct super_block *sb, struct inode *table) { unsigned int accum = 0; uint32_t elen; @@ -1902,17 +2015,17 @@ udf_count_free_table(struct super_block *sb, struct inode * table) return accum; } - -static unsigned int -udf_count_free(struct super_block *sb) + +static unsigned int udf_count_free(struct super_block *sb) { unsigned int accum = 0; - if (UDF_SB_LVIDBH(sb)) - { - if (le32_to_cpu(UDF_SB_LVID(sb)->numOfPartitions) > UDF_SB_PARTITION(sb)) - { - accum = le32_to_cpu(UDF_SB_LVID(sb)->freeSpaceTable[UDF_SB_PARTITION(sb)]); + if (UDF_SB_LVIDBH(sb)) { + if (le32_to_cpu(UDF_SB_LVID(sb)->numOfPartitions) > + UDF_SB_PARTITION(sb)) { + accum = + le32_to_cpu(UDF_SB_LVID(sb)-> + freeSpaceTable[UDF_SB_PARTITION(sb)]); if (accum == 0xFFFFFFFF) accum = 0; @@ -1922,28 +2035,40 @@ udf_count_free(struct super_block *sb) if (accum) return accum; - if (UDF_SB_PARTFLAGS(sb,UDF_SB_PARTITION(sb)) & UDF_PART_FLAG_UNALLOC_BITMAP) - { - accum += udf_count_free_bitmap(sb, - UDF_SB_PARTMAPS(sb)[UDF_SB_PARTITION(sb)].s_uspace.s_bitmap); + if (UDF_SB_PARTFLAGS(sb, UDF_SB_PARTITION(sb)) & + UDF_PART_FLAG_UNALLOC_BITMAP) { + accum += + udf_count_free_bitmap(sb, + UDF_SB_PARTMAPS(sb)[UDF_SB_PARTITION + (sb)].s_uspace. + s_bitmap); } - if (UDF_SB_PARTFLAGS(sb,UDF_SB_PARTITION(sb)) & UDF_PART_FLAG_FREED_BITMAP) - { - accum += udf_count_free_bitmap(sb, - UDF_SB_PARTMAPS(sb)[UDF_SB_PARTITION(sb)].s_fspace.s_bitmap); + if (UDF_SB_PARTFLAGS(sb, UDF_SB_PARTITION(sb)) & + UDF_PART_FLAG_FREED_BITMAP) { + accum += + udf_count_free_bitmap(sb, + UDF_SB_PARTMAPS(sb)[UDF_SB_PARTITION + (sb)].s_fspace. + s_bitmap); } if (accum) return accum; - if (UDF_SB_PARTFLAGS(sb,UDF_SB_PARTITION(sb)) & UDF_PART_FLAG_UNALLOC_TABLE) - { - accum += udf_count_free_table(sb, - UDF_SB_PARTMAPS(sb)[UDF_SB_PARTITION(sb)].s_uspace.s_table); + if (UDF_SB_PARTFLAGS(sb, UDF_SB_PARTITION(sb)) & + UDF_PART_FLAG_UNALLOC_TABLE) { + accum += + udf_count_free_table(sb, + UDF_SB_PARTMAPS(sb)[UDF_SB_PARTITION + (sb)].s_uspace. + s_table); } - if (UDF_SB_PARTFLAGS(sb,UDF_SB_PARTITION(sb)) & UDF_PART_FLAG_FREED_TABLE) - { - accum += udf_count_free_table(sb, - UDF_SB_PARTMAPS(sb)[UDF_SB_PARTITION(sb)].s_fspace.s_table); + if (UDF_SB_PARTFLAGS(sb, UDF_SB_PARTITION(sb)) & + UDF_PART_FLAG_FREED_TABLE) { + accum += + udf_count_free_table(sb, + UDF_SB_PARTMAPS(sb)[UDF_SB_PARTITION + (sb)].s_fspace. + s_table); } return accum; diff --git a/fs/udf/symlink.c b/fs/udf/symlink.c index 12613b680cc..c4b82a92008 100644 --- a/fs/udf/symlink.c +++ b/fs/udf/symlink.c @@ -33,41 +33,40 @@ #include <linux/buffer_head.h> #include "udf_i.h" -static void udf_pc_to_char(struct super_block *sb, char *from, int fromlen, char *to) +static void udf_pc_to_char(struct super_block *sb, char *from, int fromlen, + char *to) { struct pathComponent *pc; int elen = 0; char *p = to; - while (elen < fromlen) - { + while (elen < fromlen) { pc = (struct pathComponent *)(from + elen); - switch (pc->componentType) - { - case 1: - if (pc->lengthComponentIdent == 0) - { - p = to; - *p++ = '/'; - } - break; - case 3: - memcpy(p, "../", 3); - p += 3; - break; - case 4: - memcpy(p, "./", 2); - p += 2; - /* that would be . - just ignore */ - break; - case 5: - p += udf_get_filename(sb, pc->componentIdent, p, pc->lengthComponentIdent); + switch (pc->componentType) { + case 1: + if (pc->lengthComponentIdent == 0) { + p = to; *p++ = '/'; - break; + } + break; + case 3: + memcpy(p, "../", 3); + p += 3; + break; + case 4: + memcpy(p, "./", 2); + p += 2; + /* that would be . - just ignore */ + break; + case 5: + p += udf_get_filename(sb, pc->componentIdent, p, + pc->lengthComponentIdent); + *p++ = '/'; + break; } elen += sizeof(struct pathComponent) + pc->lengthComponentIdent; } - if (p > to+1) + if (p > to + 1) p[-1] = '\0'; else p[0] = '\0'; @@ -84,8 +83,7 @@ static int udf_symlink_filler(struct file *file, struct page *page) lock_kernel(); if (UDF_I_ALLOCTYPE(inode) == ICBTAG_FLAG_AD_IN_ICB) symlink = UDF_I_DATA(inode) + UDF_I_LENEATTR(inode); - else - { + else { bh = sb_bread(inode->i_sb, udf_block_map(inode, 0)); if (!bh) @@ -102,7 +100,7 @@ static int udf_symlink_filler(struct file *file, struct page *page) kunmap(page); unlock_page(page); return 0; -out: + out: unlock_kernel(); SetPageError(page); kunmap(page); @@ -114,5 +112,5 @@ out: * symlinks can't do much... */ const struct address_space_operations udf_symlink_aops = { - .readpage = udf_symlink_filler, + .readpage = udf_symlink_filler, }; diff --git a/fs/udf/truncate.c b/fs/udf/truncate.c index 60d27764424..b2002da0a5c 100644 --- a/fs/udf/truncate.c +++ b/fs/udf/truncate.c @@ -28,35 +28,38 @@ #include "udf_i.h" #include "udf_sb.h" -static void extent_trunc(struct inode * inode, struct extent_position *epos, - kernel_lb_addr eloc, int8_t etype, uint32_t elen, uint32_t nelen) +static void extent_trunc(struct inode *inode, struct extent_position *epos, + kernel_lb_addr eloc, int8_t etype, uint32_t elen, + uint32_t nelen) { kernel_lb_addr neloc = { 0, 0 }; - int last_block = (elen + inode->i_sb->s_blocksize - 1) >> inode->i_sb->s_blocksize_bits; - int first_block = (nelen + inode->i_sb->s_blocksize - 1) >> inode->i_sb->s_blocksize_bits; + int last_block = + (elen + inode->i_sb->s_blocksize - + 1) >> inode->i_sb->s_blocksize_bits; + int first_block = + (nelen + inode->i_sb->s_blocksize - + 1) >> inode->i_sb->s_blocksize_bits; - if (nelen) - { - if (etype == (EXT_NOT_RECORDED_ALLOCATED >> 30)) - { - udf_free_blocks(inode->i_sb, inode, eloc, 0, last_block); + if (nelen) { + if (etype == (EXT_NOT_RECORDED_ALLOCATED >> 30)) { + udf_free_blocks(inode->i_sb, inode, eloc, 0, + last_block); etype = (EXT_NOT_RECORDED_NOT_ALLOCATED >> 30); - } - else + } else neloc = eloc; nelen = (etype << 30) | nelen; } - if (elen != nelen) - { + if (elen != nelen) { udf_write_aext(inode, epos, neloc, nelen, 0); - if (last_block - first_block > 0) - { + if (last_block - first_block > 0) { if (etype == (EXT_RECORDED_ALLOCATED >> 30)) mark_inode_dirty(inode); if (etype != (EXT_NOT_RECORDED_NOT_ALLOCATED >> 30)) - udf_free_blocks(inode->i_sb, inode, eloc, first_block, last_block - first_block); + udf_free_blocks(inode->i_sb, inode, eloc, + first_block, + last_block - first_block); } } } @@ -67,7 +70,7 @@ static void extent_trunc(struct inode * inode, struct extent_position *epos, */ void udf_truncate_tail_extent(struct inode *inode) { - struct extent_position epos = { NULL, 0, {0, 0}}; + struct extent_position epos = { NULL, 0, {0, 0} }; kernel_lb_addr eloc; uint32_t elen, nelen; uint64_t lbcount = 0; @@ -89,8 +92,7 @@ void udf_truncate_tail_extent(struct inode *inode) BUG(); /* Find the last extent in the file */ - while ((netype = udf_next_aext(inode, &epos, &eloc, &elen, 1)) != -1) - { + while ((netype = udf_next_aext(inode, &epos, &eloc, &elen, 1)) != -1) { etype = netype; lbcount += elen; if (lbcount > inode->i_size) { @@ -123,7 +125,7 @@ void udf_truncate_tail_extent(struct inode *inode) void udf_discard_prealloc(struct inode *inode) { - struct extent_position epos = { NULL, 0, {0, 0}}; + struct extent_position epos = { NULL, 0, {0, 0} }; kernel_lb_addr eloc; uint32_t elen; uint64_t lbcount = 0; @@ -131,7 +133,7 @@ void udf_discard_prealloc(struct inode *inode) int adsize; if (UDF_I_ALLOCTYPE(inode) == ICBTAG_FLAG_AD_IN_ICB || - inode->i_size == UDF_I_LENEXTENTS(inode)) + inode->i_size == UDF_I_LENEXTENTS(inode)) return; if (UDF_I_ALLOCTYPE(inode) == ICBTAG_FLAG_AD_SHORT) @@ -153,15 +155,21 @@ void udf_discard_prealloc(struct inode *inode) lbcount -= elen; extent_trunc(inode, &epos, eloc, etype, elen, 0); if (!epos.bh) { - UDF_I_LENALLOC(inode) = epos.offset - udf_file_entry_alloc_offset(inode); + UDF_I_LENALLOC(inode) = + epos.offset - udf_file_entry_alloc_offset(inode); mark_inode_dirty(inode); } else { - struct allocExtDesc *aed = (struct allocExtDesc *)(epos.bh->b_data); - aed->lengthAllocDescs = cpu_to_le32(epos.offset - sizeof(struct allocExtDesc)); - if (!UDF_QUERY_FLAG(inode->i_sb, UDF_FLAG_STRICT) || UDF_SB_UDFREV(inode->i_sb) >= 0x0201) + struct allocExtDesc *aed = + (struct allocExtDesc *)(epos.bh->b_data); + aed->lengthAllocDescs = + cpu_to_le32(epos.offset - + sizeof(struct allocExtDesc)); + if (!UDF_QUERY_FLAG(inode->i_sb, UDF_FLAG_STRICT) + || UDF_SB_UDFREV(inode->i_sb) >= 0x0201) udf_update_tag(epos.bh->b_data, epos.offset); else - udf_update_tag(epos.bh->b_data, sizeof(struct allocExtDesc)); + udf_update_tag(epos.bh->b_data, + sizeof(struct allocExtDesc)); mark_buffer_dirty_inode(epos.bh, inode); } } @@ -171,7 +179,7 @@ void udf_discard_prealloc(struct inode *inode) brelse(epos.bh); } -void udf_truncate_extents(struct inode * inode) +void udf_truncate_extents(struct inode *inode) { struct extent_position epos; kernel_lb_addr eloc, neloc = { 0, 0 }; @@ -190,9 +198,10 @@ void udf_truncate_extents(struct inode * inode) BUG(); etype = inode_bmap(inode, first_block, &epos, &eloc, &elen, &offset); - byte_offset = (offset << sb->s_blocksize_bits) + (inode->i_size & (sb->s_blocksize-1)); - if (etype != -1) - { + byte_offset = + (offset << sb->s_blocksize_bits) + + (inode->i_size & (sb->s_blocksize - 1)); + if (etype != -1) { epos.offset -= adsize; extent_trunc(inode, &epos, eloc, etype, elen, byte_offset); epos.offset += adsize; @@ -206,86 +215,98 @@ void udf_truncate_extents(struct inode * inode) else lenalloc -= sizeof(struct allocExtDesc); - while ((etype = udf_current_aext(inode, &epos, &eloc, &elen, 0)) != -1) - { - if (etype == (EXT_NEXT_EXTENT_ALLOCDECS >> 30)) - { + while ((etype = + udf_current_aext(inode, &epos, &eloc, &elen, + 0)) != -1) { + if (etype == (EXT_NEXT_EXTENT_ALLOCDECS >> 30)) { udf_write_aext(inode, &epos, neloc, nelen, 0); - if (indirect_ext_len) - { + if (indirect_ext_len) { /* We managed to free all extents in the * indirect extent - free it too */ if (!epos.bh) BUG(); - udf_free_blocks(sb, inode, epos.block, 0, indirect_ext_len); - } - else - { - if (!epos.bh) - { - UDF_I_LENALLOC(inode) = lenalloc; + udf_free_blocks(sb, inode, epos.block, + 0, indirect_ext_len); + } else { + if (!epos.bh) { + UDF_I_LENALLOC(inode) = + lenalloc; mark_inode_dirty(inode); - } - else - { - struct allocExtDesc *aed = (struct allocExtDesc *)(epos.bh->b_data); - aed->lengthAllocDescs = cpu_to_le32(lenalloc); - if (!UDF_QUERY_FLAG(sb, UDF_FLAG_STRICT) || UDF_SB_UDFREV(sb) >= 0x0201) - udf_update_tag(epos.bh->b_data, lenalloc + - sizeof(struct allocExtDesc)); + } else { + struct allocExtDesc *aed = + (struct allocExtDesc + *)(epos.bh->b_data); + aed->lengthAllocDescs = + cpu_to_le32(lenalloc); + if (!UDF_QUERY_FLAG + (sb, UDF_FLAG_STRICT) + || UDF_SB_UDFREV(sb) >= + 0x0201) + udf_update_tag(epos.bh-> + b_data, + lenalloc + + + sizeof + (struct + allocExtDesc)); else - udf_update_tag(epos.bh->b_data, sizeof(struct allocExtDesc)); - mark_buffer_dirty_inode(epos.bh, inode); + udf_update_tag(epos.bh-> + b_data, + sizeof + (struct + allocExtDesc)); + mark_buffer_dirty_inode(epos.bh, + inode); } } brelse(epos.bh); epos.offset = sizeof(struct allocExtDesc); epos.block = eloc; - epos.bh = udf_tread(sb, udf_get_lb_pblock(sb, eloc, 0)); + epos.bh = + udf_tread(sb, + udf_get_lb_pblock(sb, eloc, 0)); if (elen) indirect_ext_len = (elen + - sb->s_blocksize - 1) >> - sb->s_blocksize_bits; + sb->s_blocksize - + 1) >> sb-> + s_blocksize_bits; else indirect_ext_len = 1; - } - else - { - extent_trunc(inode, &epos, eloc, etype, elen, 0); + } else { + extent_trunc(inode, &epos, eloc, etype, elen, + 0); epos.offset += adsize; } } - if (indirect_ext_len) - { + if (indirect_ext_len) { if (!epos.bh) BUG(); - udf_free_blocks(sb, inode, epos.block, 0, indirect_ext_len); - } - else - { - if (!epos.bh) - { + udf_free_blocks(sb, inode, epos.block, 0, + indirect_ext_len); + } else { + if (!epos.bh) { UDF_I_LENALLOC(inode) = lenalloc; mark_inode_dirty(inode); - } - else - { - struct allocExtDesc *aed = (struct allocExtDesc *)(epos.bh->b_data); + } else { + struct allocExtDesc *aed = + (struct allocExtDesc *)(epos.bh->b_data); aed->lengthAllocDescs = cpu_to_le32(lenalloc); - if (!UDF_QUERY_FLAG(sb, UDF_FLAG_STRICT) || UDF_SB_UDFREV(sb) >= 0x0201) - udf_update_tag(epos.bh->b_data, lenalloc + - sizeof(struct allocExtDesc)); + if (!UDF_QUERY_FLAG(sb, UDF_FLAG_STRICT) + || UDF_SB_UDFREV(sb) >= 0x0201) + udf_update_tag(epos.bh->b_data, + lenalloc + + sizeof(struct + allocExtDesc)); else - udf_update_tag(epos.bh->b_data, sizeof(struct allocExtDesc)); + udf_update_tag(epos.bh->b_data, + sizeof(struct + allocExtDesc)); mark_buffer_dirty_inode(epos.bh, inode); } } - } - else if (inode->i_size) - { - if (byte_offset) - { + } else if (inode->i_size) { + if (byte_offset) { kernel_long_ad extent; /* @@ -293,21 +314,33 @@ void udf_truncate_extents(struct inode * inode) * no extent above inode->i_size => truncate is * extending the file by 'offset' blocks. */ - if ((!epos.bh && epos.offset == udf_file_entry_alloc_offset(inode)) || - (epos.bh && epos.offset == sizeof(struct allocExtDesc))) { + if ((!epos.bh + && epos.offset == + udf_file_entry_alloc_offset(inode)) || (epos.bh + && epos. + offset == + sizeof + (struct + allocExtDesc))) + { /* File has no extents at all or has empty last * indirect extent! Create a fake extent... */ extent.extLocation.logicalBlockNum = 0; extent.extLocation.partitionReferenceNum = 0; - extent.extLength = EXT_NOT_RECORDED_NOT_ALLOCATED; - } - else { + extent.extLength = + EXT_NOT_RECORDED_NOT_ALLOCATED; + } else { epos.offset -= adsize; etype = udf_next_aext(inode, &epos, - &extent.extLocation, &extent.extLength, 0); + &extent.extLocation, + &extent.extLength, 0); extent.extLength |= etype << 30; } - udf_extend_file(inode, &epos, &extent, offset+((inode->i_size & (sb->s_blocksize-1)) != 0)); + udf_extend_file(inode, &epos, &extent, + offset + + ((inode-> + i_size & (sb->s_blocksize - 1)) != + 0)); } } UDF_I_LENEXTENTS(inode) = inode->i_size; diff --git a/fs/udf/udf_i.h b/fs/udf/udf_i.h index d7dbe6f3ba0..bee4308a811 100644 --- a/fs/udf/udf_i.h +++ b/fs/udf/udf_i.h @@ -23,4 +23,4 @@ static inline struct udf_inode_info *UDF_I(struct inode *inode) #define UDF_I_LAD(X) ( UDF_I(X)->i_ext.i_lad ) #define UDF_I_DATA(X) ( UDF_I(X)->i_ext.i_data ) -#endif /* !defined(_LINUX_UDF_I_H) */ +#endif /* !defined(_LINUX_UDF_I_H) */ diff --git a/fs/udf/udf_sb.h b/fs/udf/udf_sb.h index 3b2e6c8cb15..60f31d8cebe 100644 --- a/fs/udf/udf_sb.h +++ b/fs/udf/udf_sb.h @@ -20,8 +20,8 @@ #define UDF_FLAG_VARCONV 8 #define UDF_FLAG_NLS_MAP 9 #define UDF_FLAG_UTF8 10 -#define UDF_FLAG_UID_FORGET 11 /* save -1 for uid to disk */ -#define UDF_FLAG_UID_IGNORE 12 /* use sb uid instead of on disk uid */ +#define UDF_FLAG_UID_FORGET 11 /* save -1 for uid to disk */ +#define UDF_FLAG_UID_IGNORE 12 /* use sb uid instead of on disk uid */ #define UDF_FLAG_GID_FORGET 13 #define UDF_FLAG_GID_IGNORE 14 @@ -139,4 +139,4 @@ static inline struct udf_sb_info *UDF_SB(struct super_block *sb) #define UDF_SB_FLAGS(X) ( UDF_SB(X)->s_flags ) #define UDF_SB_VAT(X) ( UDF_SB(X)->s_vat ) -#endif /* __LINUX_UDF_SB_H */ +#endif /* __LINUX_UDF_SB_H */ diff --git a/fs/udf/udfdecl.h b/fs/udf/udfdecl.h index f581f2f69c0..76f2b82a39d 100644 --- a/fs/udf/udfdecl.h +++ b/fs/udf/udfdecl.h @@ -50,30 +50,26 @@ extern const struct address_space_operations udf_aops; extern const struct address_space_operations udf_adinicb_aops; extern const struct address_space_operations udf_symlink_aops; -struct udf_fileident_bh -{ +struct udf_fileident_bh { struct buffer_head *sbh; struct buffer_head *ebh; int soffset; int eoffset; }; -struct udf_vds_record -{ +struct udf_vds_record { uint32_t block; uint32_t volDescSeqNum; }; -struct generic_desc -{ - tag descTag; - __le32 volDescSeqNum; +struct generic_desc { + tag descTag; + __le32 volDescSeqNum; }; -struct ustr -{ +struct ustr { uint8_t u_cmpID; - uint8_t u_name[UDF_NAME_LEN-2]; + uint8_t u_name[UDF_NAME_LEN - 2]; uint8_t u_len; }; @@ -83,44 +79,58 @@ struct extent_position { kernel_lb_addr block; }; - /* super.c */ extern void udf_error(struct super_block *, const char *, const char *, ...); extern void udf_warning(struct super_block *, const char *, const char *, ...); /* namei.c */ -extern int udf_write_fi(struct inode *inode, struct fileIdentDesc *, struct fileIdentDesc *, struct udf_fileident_bh *, uint8_t *, uint8_t *); +extern int udf_write_fi(struct inode *inode, struct fileIdentDesc *, + struct fileIdentDesc *, struct udf_fileident_bh *, + uint8_t *, uint8_t *); /* file.c */ -extern int udf_ioctl(struct inode *, struct file *, unsigned int, unsigned long); +extern int udf_ioctl(struct inode *, struct file *, unsigned int, + unsigned long); /* inode.c */ extern struct inode *udf_iget(struct super_block *, kernel_lb_addr); extern int udf_sync_inode(struct inode *); extern void udf_expand_file_adinicb(struct inode *, int, int *); -extern struct buffer_head * udf_expand_dir_adinicb(struct inode *, int *, int *); -extern struct buffer_head * udf_bread(struct inode *, int, int, int *); +extern struct buffer_head *udf_expand_dir_adinicb(struct inode *, int *, int *); +extern struct buffer_head *udf_bread(struct inode *, int, int, int *); extern void udf_truncate(struct inode *); extern void udf_read_inode(struct inode *); extern void udf_delete_inode(struct inode *); extern void udf_clear_inode(struct inode *); extern int udf_write_inode(struct inode *, int); extern long udf_block_map(struct inode *, sector_t); -extern int udf_extend_file(struct inode *, struct extent_position *, kernel_long_ad *, sector_t); -extern int8_t inode_bmap(struct inode *, sector_t, struct extent_position *, kernel_lb_addr *, uint32_t *, sector_t *); -extern int8_t udf_add_aext(struct inode *, struct extent_position *, kernel_lb_addr, uint32_t, int); -extern int8_t udf_write_aext(struct inode *, struct extent_position *, kernel_lb_addr, uint32_t, int); -extern int8_t udf_delete_aext(struct inode *, struct extent_position, kernel_lb_addr, uint32_t); -extern int8_t udf_next_aext(struct inode *, struct extent_position *, kernel_lb_addr *, uint32_t *, int); -extern int8_t udf_current_aext(struct inode *, struct extent_position *, kernel_lb_addr *, uint32_t *, int); +extern int udf_extend_file(struct inode *, struct extent_position *, + kernel_long_ad *, sector_t); +extern int8_t inode_bmap(struct inode *, sector_t, struct extent_position *, + kernel_lb_addr *, uint32_t *, sector_t *); +extern int8_t udf_add_aext(struct inode *, struct extent_position *, + kernel_lb_addr, uint32_t, int); +extern int8_t udf_write_aext(struct inode *, struct extent_position *, + kernel_lb_addr, uint32_t, int); +extern int8_t udf_delete_aext(struct inode *, struct extent_position, + kernel_lb_addr, uint32_t); +extern int8_t udf_next_aext(struct inode *, struct extent_position *, + kernel_lb_addr *, uint32_t *, int); +extern int8_t udf_current_aext(struct inode *, struct extent_position *, + kernel_lb_addr *, uint32_t *, int); /* misc.c */ extern struct buffer_head *udf_tgetblk(struct super_block *, int); extern struct buffer_head *udf_tread(struct super_block *, int); -extern struct genericFormat *udf_add_extendedattr(struct inode *, uint32_t, uint32_t, uint8_t); -extern struct genericFormat *udf_get_extendedattr(struct inode *, uint32_t, uint8_t); -extern struct buffer_head *udf_read_tagged(struct super_block *, uint32_t, uint32_t, uint16_t *); -extern struct buffer_head *udf_read_ptagged(struct super_block *, kernel_lb_addr, uint32_t, uint16_t *); +extern struct genericFormat *udf_add_extendedattr(struct inode *, uint32_t, + uint32_t, uint8_t); +extern struct genericFormat *udf_get_extendedattr(struct inode *, uint32_t, + uint8_t); +extern struct buffer_head *udf_read_tagged(struct super_block *, uint32_t, + uint32_t, uint16_t *); +extern struct buffer_head *udf_read_ptagged(struct super_block *, + kernel_lb_addr, uint32_t, + uint16_t *); extern void udf_update_tag(char *, int); extern void udf_new_tag(char *, uint16_t, uint16_t, uint16_t, uint32_t, int); @@ -129,21 +139,26 @@ extern unsigned int udf_get_last_session(struct super_block *); extern unsigned long udf_get_last_block(struct super_block *); /* partition.c */ -extern uint32_t udf_get_pblock(struct super_block *, uint32_t, uint16_t, uint32_t); -extern uint32_t udf_get_pblock_virt15(struct super_block *, uint32_t, uint16_t, uint32_t); -extern uint32_t udf_get_pblock_virt20(struct super_block *, uint32_t, uint16_t, uint32_t); -extern uint32_t udf_get_pblock_spar15(struct super_block *, uint32_t, uint16_t, uint32_t); +extern uint32_t udf_get_pblock(struct super_block *, uint32_t, uint16_t, + uint32_t); +extern uint32_t udf_get_pblock_virt15(struct super_block *, uint32_t, uint16_t, + uint32_t); +extern uint32_t udf_get_pblock_virt20(struct super_block *, uint32_t, uint16_t, + uint32_t); +extern uint32_t udf_get_pblock_spar15(struct super_block *, uint32_t, uint16_t, + uint32_t); extern int udf_relocate_blocks(struct super_block *, long, long *); /* unicode.c */ extern int udf_get_filename(struct super_block *, uint8_t *, uint8_t *, int); -extern int udf_put_filename(struct super_block *, const uint8_t *, uint8_t *, int); +extern int udf_put_filename(struct super_block *, const uint8_t *, uint8_t *, + int); extern int udf_build_ustr(struct ustr *, dstring *, int); extern int udf_CS0toUTF8(struct ustr *, struct ustr *); /* ialloc.c */ extern void udf_free_inode(struct inode *); -extern struct inode * udf_new_inode (struct inode *, int, int *); +extern struct inode *udf_new_inode(struct inode *, int, int *); /* truncate.c */ extern void udf_truncate_tail_extent(struct inode *); @@ -151,18 +166,27 @@ extern void udf_discard_prealloc(struct inode *); extern void udf_truncate_extents(struct inode *); /* balloc.c */ -extern void udf_free_blocks(struct super_block *, struct inode *, kernel_lb_addr, uint32_t, uint32_t); -extern int udf_prealloc_blocks(struct super_block *, struct inode *, uint16_t, uint32_t, uint32_t); -extern int udf_new_block(struct super_block *, struct inode *, uint16_t, uint32_t, int *); +extern void udf_free_blocks(struct super_block *, struct inode *, + kernel_lb_addr, uint32_t, uint32_t); +extern int udf_prealloc_blocks(struct super_block *, struct inode *, uint16_t, + uint32_t, uint32_t); +extern int udf_new_block(struct super_block *, struct inode *, uint16_t, + uint32_t, int *); /* fsync.c */ extern int udf_fsync_file(struct file *, struct dentry *, int); /* directory.c */ -extern struct fileIdentDesc * udf_fileident_read(struct inode *, loff_t *, struct udf_fileident_bh *, struct fileIdentDesc *, struct extent_position *, kernel_lb_addr *, uint32_t *, sector_t *); -extern struct fileIdentDesc * udf_get_fileident(void * buffer, int bufsize, int * offset); -extern long_ad * udf_get_filelongad(uint8_t *, int, int *, int); -extern short_ad * udf_get_fileshortad(uint8_t *, int, int *, int); +extern struct fileIdentDesc *udf_fileident_read(struct inode *, loff_t *, + struct udf_fileident_bh *, + struct fileIdentDesc *, + struct extent_position *, + kernel_lb_addr *, uint32_t *, + sector_t *); +extern struct fileIdentDesc *udf_get_fileident(void *buffer, int bufsize, + int *offset); +extern long_ad *udf_get_filelongad(uint8_t *, int, int *, int); +extern short_ad *udf_get_fileshortad(uint8_t *, int, int *, int); /* crc.c */ extern uint16_t udf_crc(uint8_t *, uint32_t, uint16_t); @@ -171,4 +195,4 @@ extern uint16_t udf_crc(uint8_t *, uint32_t, uint16_t); extern time_t *udf_stamp_to_time(time_t *, long *, kernel_timestamp); extern kernel_timestamp *udf_time_to_stamp(kernel_timestamp *, struct timespec); -#endif /* __UDF_DECL_H */ +#endif /* __UDF_DECL_H */ diff --git a/fs/udf/udfend.h b/fs/udf/udfend.h index 17d37887956..450daab35a1 100644 --- a/fs/udf/udfend.h +++ b/fs/udf/udfend.h @@ -78,4 +78,4 @@ static inline timestamp cpu_to_lets(kernel_timestamp in) return out; } -#endif /* __UDF_ENDIAN_H */ +#endif /* __UDF_ENDIAN_H */ diff --git a/fs/udf/udftime.c b/fs/udf/udftime.c index 85d8dbe843f..b9f3198080e 100644 --- a/fs/udf/udftime.c +++ b/fs/udf/udftime.c @@ -46,37 +46,36 @@ #endif /* How many days come before each month (0-12). */ -static const unsigned short int __mon_yday[2][13] = -{ +static const unsigned short int __mon_yday[2][13] = { /* Normal years. */ - { 0, 31, 59, 90, 120, 151, 181, 212, 243, 273, 304, 334, 365 }, + {0, 31, 59, 90, 120, 151, 181, 212, 243, 273, 304, 334, 365}, /* Leap years. */ - { 0, 31, 60, 91, 121, 152, 182, 213, 244, 274, 305, 335, 366 } + {0, 31, 60, 91, 121, 152, 182, 213, 244, 274, 305, 335, 366} }; #define MAX_YEAR_SECONDS 69 -#define SPD 0x15180 /*3600*24*/ +#define SPD 0x15180 /*3600*24 */ #define SPY(y,l,s) (SPD * (365*y+l)+s) -static time_t year_seconds[MAX_YEAR_SECONDS]= { -/*1970*/ SPY( 0, 0,0), SPY( 1, 0,0), SPY( 2, 0,0), SPY( 3, 1,0), -/*1974*/ SPY( 4, 1,0), SPY( 5, 1,0), SPY( 6, 1,0), SPY( 7, 2,0), -/*1978*/ SPY( 8, 2,0), SPY( 9, 2,0), SPY(10, 2,0), SPY(11, 3,0), -/*1982*/ SPY(12, 3,0), SPY(13, 3,0), SPY(14, 3,0), SPY(15, 4,0), -/*1986*/ SPY(16, 4,0), SPY(17, 4,0), SPY(18, 4,0), SPY(19, 5,0), -/*1990*/ SPY(20, 5,0), SPY(21, 5,0), SPY(22, 5,0), SPY(23, 6,0), -/*1994*/ SPY(24, 6,0), SPY(25, 6,0), SPY(26, 6,0), SPY(27, 7,0), -/*1998*/ SPY(28, 7,0), SPY(29, 7,0), SPY(30, 7,0), SPY(31, 8,0), -/*2002*/ SPY(32, 8,0), SPY(33, 8,0), SPY(34, 8,0), SPY(35, 9,0), -/*2006*/ SPY(36, 9,0), SPY(37, 9,0), SPY(38, 9,0), SPY(39,10,0), -/*2010*/ SPY(40,10,0), SPY(41,10,0), SPY(42,10,0), SPY(43,11,0), -/*2014*/ SPY(44,11,0), SPY(45,11,0), SPY(46,11,0), SPY(47,12,0), -/*2018*/ SPY(48,12,0), SPY(49,12,0), SPY(50,12,0), SPY(51,13,0), -/*2022*/ SPY(52,13,0), SPY(53,13,0), SPY(54,13,0), SPY(55,14,0), -/*2026*/ SPY(56,14,0), SPY(57,14,0), SPY(58,14,0), SPY(59,15,0), -/*2030*/ SPY(60,15,0), SPY(61,15,0), SPY(62,15,0), SPY(63,16,0), -/*2034*/ SPY(64,16,0), SPY(65,16,0), SPY(66,16,0), SPY(67,17,0), -/*2038*/ SPY(68,17,0) +static time_t year_seconds[MAX_YEAR_SECONDS] = { +/*1970*/ SPY(0, 0, 0), SPY(1, 0, 0), SPY(2, 0, 0), SPY(3, 1, 0), +/*1974*/ SPY(4, 1, 0), SPY(5, 1, 0), SPY(6, 1, 0), SPY(7, 2, 0), +/*1978*/ SPY(8, 2, 0), SPY(9, 2, 0), SPY(10, 2, 0), SPY(11, 3, 0), +/*1982*/ SPY(12, 3, 0), SPY(13, 3, 0), SPY(14, 3, 0), SPY(15, 4, 0), +/*1986*/ SPY(16, 4, 0), SPY(17, 4, 0), SPY(18, 4, 0), SPY(19, 5, 0), +/*1990*/ SPY(20, 5, 0), SPY(21, 5, 0), SPY(22, 5, 0), SPY(23, 6, 0), +/*1994*/ SPY(24, 6, 0), SPY(25, 6, 0), SPY(26, 6, 0), SPY(27, 7, 0), +/*1998*/ SPY(28, 7, 0), SPY(29, 7, 0), SPY(30, 7, 0), SPY(31, 8, 0), +/*2002*/ SPY(32, 8, 0), SPY(33, 8, 0), SPY(34, 8, 0), SPY(35, 9, 0), +/*2006*/ SPY(36, 9, 0), SPY(37, 9, 0), SPY(38, 9, 0), SPY(39, 10, 0), +/*2010*/ SPY(40, 10, 0), SPY(41, 10, 0), SPY(42, 10, 0), SPY(43, 11, 0), +/*2014*/ SPY(44, 11, 0), SPY(45, 11, 0), SPY(46, 11, 0), SPY(47, 12, 0), +/*2018*/ SPY(48, 12, 0), SPY(49, 12, 0), SPY(50, 12, 0), SPY(51, 13, 0), +/*2022*/ SPY(52, 13, 0), SPY(53, 13, 0), SPY(54, 13, 0), SPY(55, 14, 0), +/*2026*/ SPY(56, 14, 0), SPY(57, 14, 0), SPY(58, 14, 0), SPY(59, 15, 0), +/*2030*/ SPY(60, 15, 0), SPY(61, 15, 0), SPY(62, 15, 0), SPY(63, 16, 0), +/*2034*/ SPY(64, 16, 0), SPY(65, 16, 0), SPY(66, 16, 0), SPY(67, 17, 0), +/*2038*/ SPY(68, 17, 0) }; extern struct timezone sys_tz; @@ -84,27 +83,23 @@ extern struct timezone sys_tz; #define SECS_PER_HOUR (60 * 60) #define SECS_PER_DAY (SECS_PER_HOUR * 24) -time_t * -udf_stamp_to_time(time_t *dest, long *dest_usec, kernel_timestamp src) +time_t *udf_stamp_to_time(time_t * dest, long *dest_usec, kernel_timestamp src) { int yday; uint8_t type = src.typeAndTimezone >> 12; int16_t offset; - if (type == 1) - { + if (type == 1) { offset = src.typeAndTimezone << 4; /* sign extent offset */ offset = (offset >> 4); - if (offset == -2047) /* unspecified offset */ + if (offset == -2047) /* unspecified offset */ offset = 0; - } - else + } else offset = 0; if ((src.year < EPOCH_YEAR) || - (src.year >= EPOCH_YEAR+MAX_YEAR_SECONDS)) - { + (src.year >= EPOCH_YEAR + MAX_YEAR_SECONDS)) { *dest = -1; *dest_usec = -1; return NULL; @@ -112,16 +107,16 @@ udf_stamp_to_time(time_t *dest, long *dest_usec, kernel_timestamp src) *dest = year_seconds[src.year - EPOCH_YEAR]; *dest -= offset * 60; - yday = ((__mon_yday[__isleap (src.year)] - [src.month-1]) + (src.day-1)); - *dest += ( ( (yday* 24) + src.hour ) * 60 + src.minute ) * 60 + src.second; - *dest_usec = src.centiseconds * 10000 + src.hundredsOfMicroseconds * 100 + src.microseconds; + yday = ((__mon_yday[__isleap(src.year)] + [src.month - 1]) + (src.day - 1)); + *dest += (((yday * 24) + src.hour) * 60 + src.minute) * 60 + src.second; + *dest_usec = + src.centiseconds * 10000 + src.hundredsOfMicroseconds * 100 + + src.microseconds; return dest; } - -kernel_timestamp * -udf_time_to_stamp(kernel_timestamp *dest, struct timespec ts) +kernel_timestamp *udf_time_to_stamp(kernel_timestamp * dest, struct timespec ts) { long int days, rem, y; const unsigned short int *ip; @@ -146,28 +141,28 @@ udf_time_to_stamp(kernel_timestamp *dest, struct timespec ts) #define DIV(a,b) ((a) / (b) - ((a) % (b) < 0)) #define LEAPS_THRU_END_OF(y) (DIV (y, 4) - DIV (y, 100) + DIV (y, 400)) - while (days < 0 || days >= (__isleap(y) ? 366 : 365)) - { + while (days < 0 || days >= (__isleap(y) ? 366 : 365)) { long int yg = y + days / 365 - (days % 365 < 0); /* Adjust DAYS and Y to match the guessed year. */ - days -= ((yg - y) * 365 - + LEAPS_THRU_END_OF (yg - 1) - - LEAPS_THRU_END_OF (y - 1)); + days -= ((yg - y) * 365 + LEAPS_THRU_END_OF(yg - 1) + - LEAPS_THRU_END_OF(y - 1)); y = yg; } dest->year = y; ip = __mon_yday[__isleap(y)]; - for (y = 11; days < (long int) ip[y]; --y) + for (y = 11; days < (long int)ip[y]; --y) continue; days -= ip[y]; dest->month = y + 1; dest->day = days + 1; dest->centiseconds = ts.tv_nsec / 10000000; - dest->hundredsOfMicroseconds = (ts.tv_nsec / 1000 - dest->centiseconds * 10000) / 100; - dest->microseconds = (ts.tv_nsec / 1000 - dest->centiseconds * 10000 - - dest->hundredsOfMicroseconds * 100); + dest->hundredsOfMicroseconds = + (ts.tv_nsec / 1000 - dest->centiseconds * 10000) / 100; + dest->microseconds = + (ts.tv_nsec / 1000 - dest->centiseconds * 10000 - + dest->hundredsOfMicroseconds * 100); return dest; } diff --git a/fs/udf/unicode.c b/fs/udf/unicode.c index 706c92e1dcc..46835240275 100644 --- a/fs/udf/unicode.c +++ b/fs/udf/unicode.c @@ -29,9 +29,9 @@ static int udf_translate_to_linux(uint8_t *, uint8_t *, int, uint8_t *, int); -static int udf_char_to_ustr(struct ustr *dest, const uint8_t *src, int strlen) +static int udf_char_to_ustr(struct ustr *dest, const uint8_t * src, int strlen) { - if ( (!dest) || (!src) || (!strlen) || (strlen > UDF_NAME_LEN-2) ) + if ((!dest) || (!src) || (!strlen) || (strlen > UDF_NAME_LEN - 2)) return 0; memset(dest, 0, sizeof(struct ustr)); memcpy(dest->u_name, src, strlen); @@ -43,33 +43,33 @@ static int udf_char_to_ustr(struct ustr *dest, const uint8_t *src, int strlen) /* * udf_build_ustr */ -int udf_build_ustr(struct ustr *dest, dstring *ptr, int size) +int udf_build_ustr(struct ustr *dest, dstring * ptr, int size) { int usesize; - if ( (!dest) || (!ptr) || (!size) ) + if ((!dest) || (!ptr) || (!size)) return -1; memset(dest, 0, sizeof(struct ustr)); - usesize= (size > UDF_NAME_LEN) ? UDF_NAME_LEN : size; - dest->u_cmpID=ptr[0]; - dest->u_len=ptr[size-1]; - memcpy(dest->u_name, ptr+1, usesize-1); + usesize = (size > UDF_NAME_LEN) ? UDF_NAME_LEN : size; + dest->u_cmpID = ptr[0]; + dest->u_len = ptr[size - 1]; + memcpy(dest->u_name, ptr + 1, usesize - 1); return 0; } /* * udf_build_ustr_exact */ -static int udf_build_ustr_exact(struct ustr *dest, dstring *ptr, int exactsize) +static int udf_build_ustr_exact(struct ustr *dest, dstring * ptr, int exactsize) { - if ( (!dest) || (!ptr) || (!exactsize) ) + if ((!dest) || (!ptr) || (!exactsize)) return -1; memset(dest, 0, sizeof(struct ustr)); - dest->u_cmpID=ptr[0]; - dest->u_len=exactsize-1; - memcpy(dest->u_name, ptr+1, exactsize-1); + dest->u_cmpID = ptr[0]; + dest->u_len = exactsize - 1; + memcpy(dest->u_name, ptr + 1, exactsize - 1); return 0; } @@ -108,22 +108,20 @@ int udf_CS0toUTF8(struct ustr *utf_o, struct ustr *ocu_i) cmp_id = ocu_i->u_cmpID; utf_o->u_len = 0; - if (ocu_len == 0) - { + if (ocu_len == 0) { memset(utf_o, 0, sizeof(struct ustr)); utf_o->u_cmpID = 0; utf_o->u_len = 0; return 0; } - if ((cmp_id != 8) && (cmp_id != 16)) - { - printk(KERN_ERR "udf: unknown compression code (%d) stri=%s\n", cmp_id, ocu_i->u_name); + if ((cmp_id != 8) && (cmp_id != 16)) { + printk(KERN_ERR "udf: unknown compression code (%d) stri=%s\n", + cmp_id, ocu_i->u_name); return 0; } - for (i = 0; (i < ocu_len) && (utf_o->u_len <= (UDF_NAME_LEN-3)) ;) - { + for (i = 0; (i < ocu_len) && (utf_o->u_len <= (UDF_NAME_LEN - 3));) { /* Expand OSTA compressed Unicode to Unicode */ c = ocu[i++]; @@ -132,20 +130,22 @@ int udf_CS0toUTF8(struct ustr *utf_o, struct ustr *ocu_i) /* Compress Unicode to UTF-8 */ if (c < 0x80U) - utf_o->u_name[utf_o->u_len++] = (uint8_t)c; - else if (c < 0x800U) - { - utf_o->u_name[utf_o->u_len++] = (uint8_t)(0xc0 | (c >> 6)); - utf_o->u_name[utf_o->u_len++] = (uint8_t)(0x80 | (c & 0x3f)); - } - else - { - utf_o->u_name[utf_o->u_len++] = (uint8_t)(0xe0 | (c >> 12)); - utf_o->u_name[utf_o->u_len++] = (uint8_t)(0x80 | ((c >> 6) & 0x3f)); - utf_o->u_name[utf_o->u_len++] = (uint8_t)(0x80 | (c & 0x3f)); + utf_o->u_name[utf_o->u_len++] = (uint8_t) c; + else if (c < 0x800U) { + utf_o->u_name[utf_o->u_len++] = + (uint8_t) (0xc0 | (c >> 6)); + utf_o->u_name[utf_o->u_len++] = + (uint8_t) (0x80 | (c & 0x3f)); + } else { + utf_o->u_name[utf_o->u_len++] = + (uint8_t) (0xe0 | (c >> 12)); + utf_o->u_name[utf_o->u_len++] = + (uint8_t) (0x80 | ((c >> 6) & 0x3f)); + utf_o->u_name[utf_o->u_len++] = + (uint8_t) (0x80 | (c & 0x3f)); } } - utf_o->u_cmpID=8; + utf_o->u_cmpID = 8; return utf_o->u_len; } @@ -173,7 +173,7 @@ int udf_CS0toUTF8(struct ustr *utf_o, struct ustr *ocu_i) * November 12, 1997 - Andrew E. Mileski * Written, tested, and released. */ -static int udf_UTF8toCS0(dstring *ocu, struct ustr *utf, int length) +static int udf_UTF8toCS0(dstring * ocu, struct ustr *utf, int length) { unsigned c, i, max_val, utf_char; int utf_cnt, u_len; @@ -182,53 +182,38 @@ static int udf_UTF8toCS0(dstring *ocu, struct ustr *utf, int length) ocu[0] = 8; max_val = 0xffU; -try_again: + try_again: u_len = 0U; utf_char = 0U; utf_cnt = 0U; - for (i = 0U; i < utf->u_len; i++) - { - c = (uint8_t)utf->u_name[i]; + for (i = 0U; i < utf->u_len; i++) { + c = (uint8_t) utf->u_name[i]; /* Complete a multi-byte UTF-8 character */ - if (utf_cnt) - { + if (utf_cnt) { utf_char = (utf_char << 6) | (c & 0x3fU); if (--utf_cnt) continue; - } - else - { + } else { /* Check for a multi-byte UTF-8 character */ - if (c & 0x80U) - { + if (c & 0x80U) { /* Start a multi-byte UTF-8 character */ - if ((c & 0xe0U) == 0xc0U) - { + if ((c & 0xe0U) == 0xc0U) { utf_char = c & 0x1fU; utf_cnt = 1; - } - else if ((c & 0xf0U) == 0xe0U) - { + } else if ((c & 0xf0U) == 0xe0U) { utf_char = c & 0x0fU; utf_cnt = 2; - } - else if ((c & 0xf8U) == 0xf0U) - { + } else if ((c & 0xf8U) == 0xf0U) { utf_char = c & 0x07U; utf_cnt = 3; - } - else if ((c & 0xfcU) == 0xf8U) - { + } else if ((c & 0xfcU) == 0xf8U) { utf_char = c & 0x03U; utf_cnt = 4; - } - else if ((c & 0xfeU) == 0xfcU) - { + } else if ((c & 0xfeU) == 0xfcU) { utf_char = c & 0x01U; utf_cnt = 5; - } - else + } else goto error_out; continue; } else @@ -237,37 +222,33 @@ try_again: } /* Choose no compression if necessary */ - if (utf_char > max_val) - { - if ( 0xffU == max_val ) - { + if (utf_char > max_val) { + if (0xffU == max_val) { max_val = 0xffffU; - ocu[0] = (uint8_t)0x10U; + ocu[0] = (uint8_t) 0x10U; goto try_again; } goto error_out; } - if (max_val == 0xffffU) - { - ocu[++u_len] = (uint8_t)(utf_char >> 8); + if (max_val == 0xffffU) { + ocu[++u_len] = (uint8_t) (utf_char >> 8); } - ocu[++u_len] = (uint8_t)(utf_char & 0xffU); + ocu[++u_len] = (uint8_t) (utf_char & 0xffU); } - - if (utf_cnt) - { -error_out: + if (utf_cnt) { + error_out: ocu[++u_len] = '?'; printk(KERN_DEBUG "udf: bad UTF-8 character\n"); } - ocu[length - 1] = (uint8_t)u_len + 1; + ocu[length - 1] = (uint8_t) u_len + 1; return u_len + 1; } -static int udf_CS0toNLS(struct nls_table *nls, struct ustr *utf_o, struct ustr *ocu_i) +static int udf_CS0toNLS(struct nls_table *nls, struct ustr *utf_o, + struct ustr *ocu_i) { uint8_t *ocu; uint32_t c; @@ -280,36 +261,35 @@ static int udf_CS0toNLS(struct nls_table *nls, struct ustr *utf_o, struct ustr * cmp_id = ocu_i->u_cmpID; utf_o->u_len = 0; - if (ocu_len == 0) - { + if (ocu_len == 0) { memset(utf_o, 0, sizeof(struct ustr)); utf_o->u_cmpID = 0; utf_o->u_len = 0; return 0; } - if ((cmp_id != 8) && (cmp_id != 16)) - { - printk(KERN_ERR "udf: unknown compression code (%d) stri=%s\n", cmp_id, ocu_i->u_name); + if ((cmp_id != 8) && (cmp_id != 16)) { + printk(KERN_ERR "udf: unknown compression code (%d) stri=%s\n", + cmp_id, ocu_i->u_name); return 0; } - for (i = 0; (i < ocu_len) && (utf_o->u_len <= (UDF_NAME_LEN-3)) ;) - { + for (i = 0; (i < ocu_len) && (utf_o->u_len <= (UDF_NAME_LEN - 3));) { /* Expand OSTA compressed Unicode to Unicode */ c = ocu[i++]; if (cmp_id == 16) c = (c << 8) | ocu[i++]; - utf_o->u_len += nls->uni2char(c, &utf_o->u_name[utf_o->u_len], - UDF_NAME_LEN - utf_o->u_len); + utf_o->u_len += nls->uni2char(c, &utf_o->u_name[utf_o->u_len], + UDF_NAME_LEN - utf_o->u_len); } - utf_o->u_cmpID=8; + utf_o->u_cmpID = 8; return utf_o->u_len; } -static int udf_NLStoCS0(struct nls_table *nls, dstring *ocu, struct ustr *uni, int length) +static int udf_NLStoCS0(struct nls_table *nls, dstring * ocu, struct ustr *uni, + int length) { unsigned len, i, max_val; uint16_t uni_char; @@ -319,93 +299,87 @@ static int udf_NLStoCS0(struct nls_table *nls, dstring *ocu, struct ustr *uni, i ocu[0] = 8; max_val = 0xffU; -try_again: + try_again: u_len = 0U; - for (i = 0U; i < uni->u_len; i++) - { - len = nls->char2uni(&uni->u_name[i], uni->u_len-i, &uni_char); + for (i = 0U; i < uni->u_len; i++) { + len = nls->char2uni(&uni->u_name[i], uni->u_len - i, &uni_char); if (len <= 0) continue; - if (uni_char > max_val) - { + if (uni_char > max_val) { max_val = 0xffffU; - ocu[0] = (uint8_t)0x10U; + ocu[0] = (uint8_t) 0x10U; goto try_again; } - + if (max_val == 0xffffU) - ocu[++u_len] = (uint8_t)(uni_char >> 8); - ocu[++u_len] = (uint8_t)(uni_char & 0xffU); + ocu[++u_len] = (uint8_t) (uni_char >> 8); + ocu[++u_len] = (uint8_t) (uni_char & 0xffU); i += len - 1; } - ocu[length - 1] = (uint8_t)u_len + 1; + ocu[length - 1] = (uint8_t) u_len + 1; return u_len + 1; } -int udf_get_filename(struct super_block *sb, uint8_t *sname, uint8_t *dname, int flen) +int udf_get_filename(struct super_block *sb, uint8_t * sname, uint8_t * dname, + int flen) { struct ustr filename, unifilename; int len; - if (udf_build_ustr_exact(&unifilename, sname, flen)) - { + if (udf_build_ustr_exact(&unifilename, sname, flen)) { return 0; } - if (UDF_QUERY_FLAG(sb, UDF_FLAG_UTF8)) - { - if (!udf_CS0toUTF8(&filename, &unifilename) ) - { - udf_debug("Failed in udf_get_filename: sname = %s\n", sname); + if (UDF_QUERY_FLAG(sb, UDF_FLAG_UTF8)) { + if (!udf_CS0toUTF8(&filename, &unifilename)) { + udf_debug("Failed in udf_get_filename: sname = %s\n", + sname); return 0; } - } - else if (UDF_QUERY_FLAG(sb, UDF_FLAG_NLS_MAP)) - { - if (!udf_CS0toNLS(UDF_SB(sb)->s_nls_map, &filename, &unifilename) ) - { - udf_debug("Failed in udf_get_filename: sname = %s\n", sname); + } else if (UDF_QUERY_FLAG(sb, UDF_FLAG_NLS_MAP)) { + if (!udf_CS0toNLS + (UDF_SB(sb)->s_nls_map, &filename, &unifilename)) { + udf_debug("Failed in udf_get_filename: sname = %s\n", + sname); return 0; } - } - else + } else return 0; - if ((len = udf_translate_to_linux(dname, filename.u_name, filename.u_len, - unifilename.u_name, unifilename.u_len))) - { + if ((len = + udf_translate_to_linux(dname, filename.u_name, filename.u_len, + unifilename.u_name, unifilename.u_len))) { return len; } return 0; } -int udf_put_filename(struct super_block *sb, const uint8_t *sname, uint8_t *dname, int flen) +int udf_put_filename(struct super_block *sb, const uint8_t * sname, + uint8_t * dname, int flen) { struct ustr unifilename; int namelen; - if ( !(udf_char_to_ustr(&unifilename, sname, flen)) ) - { + if (!(udf_char_to_ustr(&unifilename, sname, flen))) { return 0; } - if (UDF_QUERY_FLAG(sb, UDF_FLAG_UTF8)) - { - if ( !(namelen = udf_UTF8toCS0(dname, &unifilename, UDF_NAME_LEN)) ) - { + if (UDF_QUERY_FLAG(sb, UDF_FLAG_UTF8)) { + if (! + (namelen = + udf_UTF8toCS0(dname, &unifilename, UDF_NAME_LEN))) { return 0; } - } - else if (UDF_QUERY_FLAG(sb, UDF_FLAG_NLS_MAP)) - { - if ( !(namelen = udf_NLStoCS0(UDF_SB(sb)->s_nls_map, dname, &unifilename, UDF_NAME_LEN)) ) - { + } else if (UDF_QUERY_FLAG(sb, UDF_FLAG_NLS_MAP)) { + if (! + (namelen = + udf_NLStoCS0(UDF_SB(sb)->s_nls_map, dname, &unifilename, + UDF_NAME_LEN))) { return 0; } - } - else + } else return 0; return namelen; @@ -416,40 +390,36 @@ int udf_put_filename(struct super_block *sb, const uint8_t *sname, uint8_t *dnam #define CRC_MARK '#' #define EXT_SIZE 5 -static int udf_translate_to_linux(uint8_t *newName, uint8_t *udfName, int udfLen, uint8_t *fidName, int fidNameLen) +static int udf_translate_to_linux(uint8_t * newName, uint8_t * udfName, + int udfLen, uint8_t * fidName, int fidNameLen) { - int index, newIndex = 0, needsCRC = 0; + int index, newIndex = 0, needsCRC = 0; int extIndex = 0, newExtIndex = 0, hasExt = 0; unsigned short valueCRC; uint8_t curr; const uint8_t hexChar[] = "0123456789ABCDEF"; if (udfName[0] == '.' && (udfLen == 1 || - (udfLen == 2 && udfName[1] == '.'))) - { + (udfLen == 2 && udfName[1] == '.'))) { needsCRC = 1; newIndex = udfLen; memcpy(newName, udfName, udfLen); - } - else - { - for (index = 0; index < udfLen; index++) - { + } else { + for (index = 0; index < udfLen; index++) { curr = udfName[index]; - if (curr == '/' || curr == 0) - { + if (curr == '/' || curr == 0) { needsCRC = 1; curr = ILLEGAL_CHAR_MARK; - while (index+1 < udfLen && (udfName[index+1] == '/' || - udfName[index+1] == 0)) + while (index + 1 < udfLen + && (udfName[index + 1] == '/' + || udfName[index + 1] == 0)) index++; } - if (curr == EXT_MARK && (udfLen - index - 1) <= EXT_SIZE) - { + if (curr == EXT_MARK + && (udfLen - index - 1) <= EXT_SIZE) { if (udfLen == index + 1) hasExt = 0; - else - { + else { hasExt = 1; extIndex = index; newExtIndex = newIndex; @@ -461,26 +431,29 @@ static int udf_translate_to_linux(uint8_t *newName, uint8_t *udfName, int udfLen needsCRC = 1; } } - if (needsCRC) - { + if (needsCRC) { uint8_t ext[EXT_SIZE]; int localExtIndex = 0; - if (hasExt) - { + if (hasExt) { int maxFilenameLen; - for(index = 0; index<EXT_SIZE && extIndex + index +1 < udfLen; - index++ ) - { + for (index = 0; + index < EXT_SIZE && extIndex + index + 1 < udfLen; + index++) { curr = udfName[extIndex + index + 1]; - if (curr == '/' || curr == 0) - { + if (curr == '/' || curr == 0) { needsCRC = 1; curr = ILLEGAL_CHAR_MARK; - while(extIndex + index + 2 < udfLen && (index + 1 < EXT_SIZE - && (udfName[extIndex + index + 2] == '/' || - udfName[extIndex + index + 2] == 0))) + while (extIndex + index + 2 < udfLen + && (index + 1 < EXT_SIZE + && + (udfName + [extIndex + index + 2] == + '/' + || udfName[extIndex + + index + 2] == + 0))) index++; } ext[localExtIndex++] = curr; @@ -490,8 +463,7 @@ static int udf_translate_to_linux(uint8_t *newName, uint8_t *udfName, int udfLen newIndex = maxFilenameLen; else newIndex = newExtIndex; - } - else if (newIndex > 250) + } else if (newIndex > 250) newIndex = 250; newName[newIndex++] = CRC_MARK; valueCRC = udf_crc(fidName, fidNameLen, 0); @@ -500,10 +472,9 @@ static int udf_translate_to_linux(uint8_t *newName, uint8_t *udfName, int udfLen newName[newIndex++] = hexChar[(valueCRC & 0x00f0) >> 4]; newName[newIndex++] = hexChar[(valueCRC & 0x000f)]; - if (hasExt) - { + if (hasExt) { newName[newIndex++] = EXT_MARK; - for (index = 0;index < localExtIndex ;index++ ) + for (index = 0; index < localExtIndex; index++) newName[newIndex++] = ext[index]; } } diff --git a/fs/xfs/linux-2.6/xfs_file.c b/fs/xfs/linux-2.6/xfs_file.c index cbcd40c8c2a..0d4001eafd1 100644 --- a/fs/xfs/linux-2.6/xfs_file.c +++ b/fs/xfs/linux-2.6/xfs_file.c @@ -212,19 +212,18 @@ xfs_file_fsync( } #ifdef CONFIG_XFS_DMAPI -STATIC struct page * -xfs_vm_nopage( - struct vm_area_struct *area, - unsigned long address, - int *type) +STATIC int +xfs_vm_fault( + struct vm_area_struct *vma, + struct vm_fault *vmf) { - struct inode *inode = area->vm_file->f_path.dentry->d_inode; + struct inode *inode = vma->vm_file->f_path.dentry->d_inode; bhv_vnode_t *vp = vn_from_inode(inode); ASSERT_ALWAYS(vp->v_vfsp->vfs_flag & VFS_DMI); - if (XFS_SEND_MMAP(XFS_VFSTOM(vp->v_vfsp), area, 0)) - return NULL; - return filemap_nopage(area, address, type); + if (XFS_SEND_MMAP(XFS_VFSTOM(vp->v_vfsp), vma, 0)) + return VM_FAULT_SIGBUS; + return filemap_fault(vma, vmf); } #endif /* CONFIG_XFS_DMAPI */ @@ -310,6 +309,7 @@ xfs_file_mmap( struct vm_area_struct *vma) { vma->vm_ops = &xfs_file_vm_ops; + vma->vm_flags |= VM_CAN_NONLINEAR; #ifdef CONFIG_XFS_DMAPI if (vn_from_inode(filp->f_path.dentry->d_inode)->v_vfsp->vfs_flag & VFS_DMI) @@ -413,6 +413,20 @@ xfs_file_open_exec( } #endif /* HAVE_FOP_OPEN_EXEC */ +/* + * mmap()d file has taken write protection fault and is being made + * writable. We can set the page state up correctly for a writable + * page, which means we can do correct delalloc accounting (ENOSPC + * checking!) and unwritten extent mapping. + */ +STATIC int +xfs_vm_page_mkwrite( + struct vm_area_struct *vma, + struct page *page) +{ + return block_page_mkwrite(vma, page, xfs_get_blocks); +} + const struct file_operations xfs_file_operations = { .llseek = generic_file_llseek, .read = do_sync_read, @@ -464,14 +478,14 @@ const struct file_operations xfs_dir_file_operations = { }; static struct vm_operations_struct xfs_file_vm_ops = { - .nopage = filemap_nopage, - .populate = filemap_populate, + .fault = filemap_fault, + .page_mkwrite = xfs_vm_page_mkwrite, }; #ifdef CONFIG_XFS_DMAPI static struct vm_operations_struct xfs_dmapi_file_vm_ops = { - .nopage = xfs_vm_nopage, - .populate = filemap_populate, + .fault = xfs_vm_fault, + .page_mkwrite = xfs_vm_page_mkwrite, #ifdef HAVE_VMOP_MPROTECT .mprotect = xfs_vm_mprotect, #endif diff --git a/fs/xfs/xfs_vnodeops.c b/fs/xfs/xfs_vnodeops.c index 79b522779aa..1a5ad8cd97b 100644 --- a/fs/xfs/xfs_vnodeops.c +++ b/fs/xfs/xfs_vnodeops.c @@ -589,7 +589,30 @@ xfs_setattr( code = xfs_igrow_start(ip, vap->va_size, credp); } xfs_iunlock(ip, XFS_ILOCK_EXCL); - vn_iowait(vp); /* wait for the completion of any pending DIOs */ + + /* + * We are going to log the inode size change in this + * transaction so any previous writes that are beyond the on + * disk EOF and the new EOF that have not been written out need + * to be written here. If we do not write the data out, we + * expose ourselves to the null files problem. + * + * Only flush from the on disk size to the smaller of the in + * memory file size or the new size as that's the range we + * really care about here and prevents waiting for other data + * not within the range we care about here. + */ + if (!code && + (ip->i_size != ip->i_d.di_size) && + (vap->va_size > ip->i_d.di_size)) { + code = bhv_vop_flush_pages(XFS_ITOV(ip), + ip->i_d.di_size, vap->va_size, + XFS_B_ASYNC, FI_NONE); + } + + /* wait for all I/O to complete */ + vn_iowait(vp); + if (!code) code = xfs_itruncate_data(ip, vap->va_size); if (code) { @@ -4434,9 +4457,12 @@ xfs_free_file_space( while (!error && !done) { /* - * allocate and setup the transaction + * allocate and setup the transaction. Allow this + * transaction to dip into the reserve blocks to ensure + * the freeing of the space succeeds at ENOSPC. */ tp = xfs_trans_alloc(mp, XFS_TRANS_DIOSTRAT); + tp->t_flags |= XFS_TRANS_RESERVE; error = xfs_trans_reserve(tp, resblks, XFS_WRITE_LOG_RES(mp), diff --git a/include/asm-alpha/a.out.h b/include/asm-alpha/a.out.h index d97daf42753..e43cf61649a 100644 --- a/include/asm-alpha/a.out.h +++ b/include/asm-alpha/a.out.h @@ -101,6 +101,8 @@ struct exec #define STACK_TOP \ (current->personality & ADDR_LIMIT_32BIT ? 0x80000000 : 0x00120000000UL) +#define STACK_TOP_MAX 0x00120000000UL + #endif #endif /* __A_OUT_GNU_H__ */ diff --git a/include/asm-alpha/system.h b/include/asm-alpha/system.h index cf1021a97b2..620c4d86cbf 100644 --- a/include/asm-alpha/system.h +++ b/include/asm-alpha/system.h @@ -139,16 +139,6 @@ extern void halt(void) __attribute__((noreturn)); struct task_struct; extern struct task_struct *alpha_switch_to(unsigned long, struct task_struct*); -/* - * On SMP systems, when the scheduler does migration-cost autodetection, - * it needs a way to flush as much of the CPU's caches as possible. - * - * TODO: fill this in! - */ -static inline void sched_cacheflush(void) -{ -} - #define imb() \ __asm__ __volatile__ ("call_pal %0 #imb" : : "i" (PAL_imb) : "memory") diff --git a/include/asm-arm/a.out.h b/include/asm-arm/a.out.h index 3e5fe64c439..d7165e86df2 100644 --- a/include/asm-arm/a.out.h +++ b/include/asm-arm/a.out.h @@ -30,6 +30,7 @@ struct exec #ifdef __KERNEL__ #define STACK_TOP ((current->personality == PER_LINUX_32BIT) ? \ TASK_SIZE : TASK_SIZE_26) +#define STACK_TOP_MAX TASK_SIZE #endif #ifndef LIBRARY_START_TEXT diff --git a/include/asm-arm/system.h b/include/asm-arm/system.h index 6f8e6a69dc5..94ea8c6dc1a 100644 --- a/include/asm-arm/system.h +++ b/include/asm-arm/system.h @@ -254,16 +254,6 @@ do { \ last = __switch_to(prev,task_thread_info(prev), task_thread_info(next)); \ } while (0) -/* - * On SMP systems, when the scheduler does migration-cost autodetection, - * it needs a way to flush as much of the CPU's caches as possible. - * - * TODO: fill this in! - */ -static inline void sched_cacheflush(void) -{ -} - #if defined(CONFIG_CPU_SA1100) || defined(CONFIG_CPU_SA110) /* * On the StrongARM, "swp" is terminally broken since it bypasses the diff --git a/include/asm-arm26/a.out.h b/include/asm-arm26/a.out.h index 9b2702c42c8..7167f54ae3f 100644 --- a/include/asm-arm26/a.out.h +++ b/include/asm-arm26/a.out.h @@ -29,6 +29,7 @@ struct exec #ifdef __KERNEL__ #define STACK_TOP TASK_SIZE +#define STACK_TOP_MAX STACK_TOP #endif #ifndef LIBRARY_START_TEXT diff --git a/include/asm-arm26/system.h b/include/asm-arm26/system.h index 4703593b3bb..e09da5ff1f5 100644 --- a/include/asm-arm26/system.h +++ b/include/asm-arm26/system.h @@ -110,16 +110,6 @@ do { \ } while (0) /* - * On SMP systems, when the scheduler does migration-cost autodetection, - * it needs a way to flush as much of the CPU's caches as possible. - * - * TODO: fill this in! - */ -static inline void sched_cacheflush(void) -{ -} - -/* * Save the current interrupt enable state & disable IRQs */ #define local_irq_save(x) \ diff --git a/include/asm-avr32/a.out.h b/include/asm-avr32/a.out.h index 50bf6e31a14..9f398ab28ed 100644 --- a/include/asm-avr32/a.out.h +++ b/include/asm-avr32/a.out.h @@ -20,6 +20,7 @@ struct exec #ifdef __KERNEL__ #define STACK_TOP TASK_SIZE +#define STACK_TOP_MAX STACK_TOP #endif diff --git a/include/asm-cris/a.out.h b/include/asm-cris/a.out.h index 770734ce54a..919b34a084f 100644 --- a/include/asm-cris/a.out.h +++ b/include/asm-cris/a.out.h @@ -8,6 +8,7 @@ /* grabbed from the intel stuff */ #define STACK_TOP TASK_SIZE +#define STACK_TOP_MAX STACK_TOP struct exec diff --git a/include/asm-frv/mem-layout.h b/include/asm-frv/mem-layout.h index a025dd4514e..aaf2a773d9d 100644 --- a/include/asm-frv/mem-layout.h +++ b/include/asm-frv/mem-layout.h @@ -60,6 +60,7 @@ */ #define BRK_BASE __UL(2 * 1024 * 1024 + PAGE_SIZE) #define STACK_TOP __UL(2 * 1024 * 1024) +#define STACK_TOP_MAX STACK_TOP /* userspace process size */ #ifdef CONFIG_MMU diff --git a/include/asm-generic/percpu.h b/include/asm-generic/percpu.h index d984a904143..d85172e9ed4 100644 --- a/include/asm-generic/percpu.h +++ b/include/asm-generic/percpu.h @@ -14,6 +14,11 @@ extern unsigned long __per_cpu_offset[NR_CPUS]; #define DEFINE_PER_CPU(type, name) \ __attribute__((__section__(".data.percpu"))) __typeof__(type) per_cpu__##name +#define DEFINE_PER_CPU_SHARED_ALIGNED(type, name) \ + __attribute__((__section__(".data.percpu.shared_aligned"))) \ + __typeof__(type) per_cpu__##name \ + ____cacheline_aligned_in_smp + /* var is in discarded region: offset to particular copy we want */ #define per_cpu(var, cpu) (*({ \ extern int simple_identifier_##var(void); \ @@ -34,6 +39,9 @@ do { \ #define DEFINE_PER_CPU(type, name) \ __typeof__(type) per_cpu__##name +#define DEFINE_PER_CPU_SHARED_ALIGNED(type, name) \ + DEFINE_PER_CPU(type, name) + #define per_cpu(var, cpu) (*((void)(cpu), &per_cpu__##var)) #define __get_cpu_var(var) per_cpu__##var #define __raw_get_cpu_var(var) per_cpu__##var diff --git a/include/asm-generic/vmlinux.lds.h b/include/asm-generic/vmlinux.lds.h index 84155eb67f1..0240e0506a0 100644 --- a/include/asm-generic/vmlinux.lds.h +++ b/include/asm-generic/vmlinux.lds.h @@ -224,7 +224,11 @@ } #define NOTES \ - .notes : { *(.note.*) } :note + .notes : AT(ADDR(.notes) - LOAD_OFFSET) { \ + VMLINUX_SYMBOL(__start_notes) = .; \ + *(.note.*) \ + VMLINUX_SYMBOL(__stop_notes) = .; \ + } #define INITCALLS \ *(.initcall0.init) \ @@ -245,3 +249,11 @@ *(.initcall7.init) \ *(.initcall7s.init) +#define PERCPU(align) \ + . = ALIGN(align); \ + __per_cpu_start = .; \ + .data.percpu : AT(ADDR(.data.percpu) - LOAD_OFFSET) { \ + *(.data.percpu) \ + *(.data.percpu.shared_aligned) \ + } \ + __per_cpu_end = .; diff --git a/include/asm-h8300/a.out.h b/include/asm-h8300/a.out.h index 3c70939f9f0..aa5d2277823 100644 --- a/include/asm-h8300/a.out.h +++ b/include/asm-h8300/a.out.h @@ -20,6 +20,7 @@ struct exec #ifdef __KERNEL__ #define STACK_TOP TASK_SIZE +#define STACK_TOP_MAX STACK_TOP #endif diff --git a/include/asm-i386/a.out.h b/include/asm-i386/a.out.h index ab17bb8e546..851a60f8258 100644 --- a/include/asm-i386/a.out.h +++ b/include/asm-i386/a.out.h @@ -20,6 +20,7 @@ struct exec #ifdef __KERNEL__ #define STACK_TOP TASK_SIZE +#define STACK_TOP_MAX STACK_TOP #endif diff --git a/include/asm-i386/cmpxchg.h b/include/asm-i386/cmpxchg.h index 7adcef0cd53..64dcdf46117 100644 --- a/include/asm-i386/cmpxchg.h +++ b/include/asm-i386/cmpxchg.h @@ -3,14 +3,16 @@ #include <linux/bitops.h> /* for LOCK_PREFIX */ +/* + * Note: if you use set64_bit(), __cmpxchg64(), or their variants, you + * you need to test for the feature in boot_cpu_data. + */ + #define xchg(ptr,v) ((__typeof__(*(ptr)))__xchg((unsigned long)(v),(ptr),sizeof(*(ptr)))) struct __xchg_dummy { unsigned long a[100]; }; #define __xg(x) ((struct __xchg_dummy *)(x)) - -#ifdef CONFIG_X86_CMPXCHG64 - /* * The semantics of XCHGCMP8B are a bit strange, this is why * there is a loop and the loading of %%eax and %%edx has to @@ -65,8 +67,6 @@ static inline void __set_64bit_var (unsigned long long *ptr, __set_64bit(ptr, (unsigned int)(value), (unsigned int)((value)>>32ULL) ) : \ __set_64bit(ptr, ll_low(value), ll_high(value)) ) -#endif - /* * Note: no "lock" prefix even on SMP: xchg always implies lock anyway * Note 2: xchg has side effect, so that attribute volatile is necessary, @@ -252,8 +252,6 @@ static inline unsigned long cmpxchg_386(volatile void *ptr, unsigned long old, }) #endif -#ifdef CONFIG_X86_CMPXCHG64 - static inline unsigned long long __cmpxchg64(volatile void *ptr, unsigned long long old, unsigned long long new) { @@ -289,5 +287,3 @@ static inline unsigned long long __cmpxchg64_local(volatile void *ptr, ((__typeof__(*(ptr)))__cmpxchg64_local((ptr),(unsigned long long)(o),\ (unsigned long long)(n))) #endif - -#endif diff --git a/include/asm-i386/kprobes.h b/include/asm-i386/kprobes.h index 8774d06689d..06f7303c30c 100644 --- a/include/asm-i386/kprobes.h +++ b/include/asm-i386/kprobes.h @@ -42,7 +42,6 @@ typedef u8 kprobe_opcode_t; ? (MAX_STACK_SIZE) \ : (((unsigned long)current_thread_info()) + THREAD_SIZE - (ADDR))) -#define JPROBE_ENTRY(pentry) (kprobe_opcode_t *)pentry #define ARCH_SUPPORTS_KRETPROBES #define ARCH_INACTIVE_KPROBE_COUNT 0 #define flush_insn_slot(p) do { } while (0) diff --git a/include/asm-i386/percpu.h b/include/asm-i386/percpu.h index f54830b5d5a..a7ebd436f3c 100644 --- a/include/asm-i386/percpu.h +++ b/include/asm-i386/percpu.h @@ -54,6 +54,11 @@ extern unsigned long __per_cpu_offset[]; #define DEFINE_PER_CPU(type, name) \ __attribute__((__section__(".data.percpu"))) __typeof__(type) per_cpu__##name +#define DEFINE_PER_CPU_SHARED_ALIGNED(type, name) \ + __attribute__((__section__(".data.percpu.shared_aligned"))) \ + __typeof__(type) per_cpu__##name \ + ____cacheline_aligned_in_smp + /* We can use this directly for local CPU (faster). */ DECLARE_PER_CPU(unsigned long, this_cpu_off); diff --git a/include/asm-i386/required-features.h b/include/asm-i386/required-features.h index 65848a00705..618feb98f9f 100644 --- a/include/asm-i386/required-features.h +++ b/include/asm-i386/required-features.h @@ -29,7 +29,7 @@ # define NEED_CMOV 0 #endif -#ifdef CONFIG_X86_CMPXCHG64 +#ifdef CONFIG_X86_PAE # define NEED_CX8 (1<<(X86_FEATURE_CX8 & 31)) #else # define NEED_CX8 0 diff --git a/include/asm-i386/system.h b/include/asm-i386/system.h index 94ed3686a5f..609756c6167 100644 --- a/include/asm-i386/system.h +++ b/include/asm-i386/system.h @@ -310,15 +310,6 @@ void enable_hlt(void); extern int es7000_plat; void cpu_idle_wait(void); -/* - * On SMP systems, when the scheduler does migration-cost autodetection, - * it needs a way to flush as much of the CPU's caches as possible: - */ -static inline void sched_cacheflush(void) -{ - wbinvd(); -} - extern unsigned long arch_align_stack(unsigned long sp); extern void free_init_pages(char *what, unsigned long begin, unsigned long end); diff --git a/include/asm-i386/tsc.h b/include/asm-i386/tsc.h index 62c091ffccc..a4d806610b7 100644 --- a/include/asm-i386/tsc.h +++ b/include/asm-i386/tsc.h @@ -63,6 +63,7 @@ extern void tsc_init(void); extern void mark_tsc_unstable(char *reason); extern int unsynchronized_tsc(void); extern void init_tsc_clocksource(void); +int check_tsc_unstable(void); /* * Boot-time check whether the TSCs are synchronized across diff --git a/include/asm-ia64/kprobes.h b/include/asm-ia64/kprobes.h index 6382e52ec22..067d9dea68f 100644 --- a/include/asm-ia64/kprobes.h +++ b/include/asm-ia64/kprobes.h @@ -82,8 +82,6 @@ struct kprobe_ctlblk { struct prev_kprobe prev_kprobe[ARCH_PREV_KPROBE_SZ]; }; -#define JPROBE_ENTRY(pentry) (kprobe_opcode_t *)pentry - #define ARCH_SUPPORTS_KRETPROBES #define ARCH_INACTIVE_KPROBE_COUNT 1 diff --git a/include/asm-ia64/percpu.h b/include/asm-ia64/percpu.h index fbe5cf3ab8d..43a7aac414e 100644 --- a/include/asm-ia64/percpu.h +++ b/include/asm-ia64/percpu.h @@ -29,6 +29,16 @@ __attribute__((__section__(".data.percpu"))) \ __SMALL_ADDR_AREA __typeof__(type) per_cpu__##name +#ifdef CONFIG_SMP +#define DEFINE_PER_CPU_SHARED_ALIGNED(type, name) \ + __attribute__((__section__(".data.percpu.shared_aligned"))) \ + __SMALL_ADDR_AREA __typeof__(type) per_cpu__##name \ + ____cacheline_aligned_in_smp +#else +#define DEFINE_PER_CPU_SHARED_ALIGNED(type, name) \ + DEFINE_PER_CPU(type, name) +#endif + /* * Pretty much a literal copy of asm-generic/percpu.h, except that percpu_modcopy() is an * external routine, to avoid include-hell. diff --git a/include/asm-ia64/processor.h b/include/asm-ia64/processor.h index db81ba406ce..6251c76437d 100644 --- a/include/asm-ia64/processor.h +++ b/include/asm-ia64/processor.h @@ -295,9 +295,9 @@ struct thread_struct { regs->ar_bspstore = current->thread.rbs_bot; \ regs->ar_fpsr = FPSR_DEFAULT; \ regs->loadrs = 0; \ - regs->r8 = current->mm->dumpable; /* set "don't zap registers" flag */ \ + regs->r8 = get_dumpable(current->mm); /* set "don't zap registers" flag */ \ regs->r12 = new_sp - 16; /* allocate 16 byte scratch area */ \ - if (unlikely(!current->mm->dumpable)) { \ + if (unlikely(!get_dumpable(current->mm))) { \ /* \ * Zap scratch regs to avoid leaking bits between processes with different \ * uid/privileges. \ diff --git a/include/asm-ia64/system.h b/include/asm-ia64/system.h index 384fbf7f2a0..91bb8e00066 100644 --- a/include/asm-ia64/system.h +++ b/include/asm-ia64/system.h @@ -259,7 +259,6 @@ extern void ia64_load_extra (struct task_struct *task); #define ia64_platform_is(x) (strcmp(x, platform_name) == 0) void cpu_idle_wait(void); -void sched_cacheflush(void); #define arch_align_stack(x) (x) diff --git a/include/asm-ia64/ustack.h b/include/asm-ia64/ustack.h index a349467913e..504167c35b8 100644 --- a/include/asm-ia64/ustack.h +++ b/include/asm-ia64/ustack.h @@ -11,6 +11,7 @@ /* The absolute hard limit for stack size is 1/2 of the mappable space in the region */ #define MAX_USER_STACK_SIZE (RGN_MAP_LIMIT/2) #define STACK_TOP (0x6000000000000000UL + RGN_MAP_LIMIT) +#define STACK_TOP_MAX STACK_TOP #endif /* Make a default stack size of 2GiB */ diff --git a/include/asm-m32r/a.out.h b/include/asm-m32r/a.out.h index 9a4a5d20160..6a1b5d42f32 100644 --- a/include/asm-m32r/a.out.h +++ b/include/asm-m32r/a.out.h @@ -20,6 +20,7 @@ struct exec #ifdef __KERNEL__ #define STACK_TOP TASK_SIZE +#define STACK_TOP_MAX STACK_TOP #endif diff --git a/include/asm-m32r/system.h b/include/asm-m32r/system.h index 8ee73d3f316..2365de5c295 100644 --- a/include/asm-m32r/system.h +++ b/include/asm-m32r/system.h @@ -54,16 +54,6 @@ ); \ } while(0) -/* - * On SMP systems, when the scheduler does migration-cost autodetection, - * it needs a way to flush as much of the CPU's caches as possible. - * - * TODO: fill this in! - */ -static inline void sched_cacheflush(void) -{ -} - /* Interrupt Control */ #if !defined(CONFIG_CHIP_M32102) && !defined(CONFIG_CHIP_M32104) #define local_irq_enable() \ diff --git a/include/asm-m68k/a.out.h b/include/asm-m68k/a.out.h index eda1662773b..6fc86a221a9 100644 --- a/include/asm-m68k/a.out.h +++ b/include/asm-m68k/a.out.h @@ -20,6 +20,7 @@ struct exec #ifdef __KERNEL__ #define STACK_TOP TASK_SIZE +#define STACK_TOP_MAX STACK_TOP #endif diff --git a/include/asm-m68knommu/irq.h b/include/asm-m68knommu/irq.h index 7b8f874f842..9373c31ac87 100644 --- a/include/asm-m68knommu/irq.h +++ b/include/asm-m68knommu/irq.h @@ -1,7 +1,5 @@ -#ifndef _M68K_IRQ_H_ -#define _M68K_IRQ_H_ - -#include <asm/ptrace.h> +#ifndef _M68KNOMMU_IRQ_H_ +#define _M68KNOMMU_IRQ_H_ #ifdef CONFIG_COLDFIRE /* @@ -17,75 +15,12 @@ /* * # of m68k interrupts */ -#define SYS_IRQS 8 -#define NR_IRQS (24+SYS_IRQS) +#define SYS_IRQS 8 +#define NR_IRQS (24 + SYS_IRQS) #endif /* CONFIG_COLDFIRE */ -/* - * Interrupt source definitions - * General interrupt sources are the level 1-7. - * Adding an interrupt service routine for one of these sources - * results in the addition of that routine to a chain of routines. - * Each one is called in succession. Each individual interrupt - * service routine should determine if the device associated with - * that routine requires service. - */ -#define IRQ1 (1) /* level 1 interrupt */ -#define IRQ2 (2) /* level 2 interrupt */ -#define IRQ3 (3) /* level 3 interrupt */ -#define IRQ4 (4) /* level 4 interrupt */ -#define IRQ5 (5) /* level 5 interrupt */ -#define IRQ6 (6) /* level 6 interrupt */ -#define IRQ7 (7) /* level 7 interrupt (non-maskable) */ - -/* - * Machine specific interrupt sources. - * - * Adding an interrupt service routine for a source with this bit - * set indicates a special machine specific interrupt source. - * The machine specific files define these sources. - * - * The IRQ_MACHSPEC bit is now gone - the only thing it did was to - * introduce unnecessary overhead. - * - * All interrupt handling is actually machine specific so it is better - * to use function pointers, as used by the Sparc port, and select the - * interrupt handling functions when initializing the kernel. This way - * we save some unnecessary overhead at run-time. - * 01/11/97 - Jes - */ - -extern void (*mach_enable_irq)(unsigned int); -extern void (*mach_disable_irq)(unsigned int); - -/* - * various flags for request_irq() - the Amiga now uses the standard - * mechanism like all other architectures - IRQF_DISABLED and - * IRQF_SHARED are your friends. - */ -#define IRQ_FLG_LOCK (0x0001) /* handler is not replaceable */ -#define IRQ_FLG_REPLACE (0x0002) /* replace existing handler */ -#define IRQ_FLG_FAST (0x0004) -#define IRQ_FLG_SLOW (0x0008) -#define IRQ_FLG_STD (0x8000) /* internally used */ - -#ifdef CONFIG_M68360 - -#define CPM_INTERRUPT IRQ4 - -/* see MC68360 User's Manual, p. 7-377 */ -#define CPM_VECTOR_BASE 0x04 /* 3 MSbits of CPM vector */ - -#endif /* CONFIG_M68360 */ - -/* - * Some drivers want these entry points - */ -#define enable_irq(x) do { } while (0) -#define disable_irq(x) do { } while (0) -#define disable_irq_nosync(x) disable_irq(x) #define irq_canonicalize(irq) (irq) -#endif /* _M68K_IRQ_H_ */ +#endif /* _M68KNOMMU_IRQ_H_ */ diff --git a/include/asm-m68knommu/irqnode.h b/include/asm-m68knommu/irqnode.h deleted file mode 100644 index 6132a9858b5..00000000000 --- a/include/asm-m68knommu/irqnode.h +++ /dev/null @@ -1,36 +0,0 @@ -#ifndef _M68K_IRQNODE_H_ -#define _M68K_IRQNODE_H_ - -#include <linux/interrupt.h> - -/* - * This structure is used to chain together the ISRs for a particular - * interrupt source (if it supports chaining). - */ -typedef struct irq_node { - irq_handler_t handler; - unsigned long flags; - void *dev_id; - const char *devname; - struct irq_node *next; -} irq_node_t; - -/* - * This structure has only 4 elements for speed reasons - */ -struct irq_entry { - irq_handler_t handler; - unsigned long flags; - void *dev_id; - const char *devname; -}; - -/* count of spurious interrupts */ -extern volatile unsigned int num_spurious; - -/* - * This function returns a new irq_node_t - */ -extern irq_node_t *new_irq_node(void); - -#endif /* _M68K_IRQNODE_H_ */ diff --git a/include/asm-m68knommu/m68360.h b/include/asm-m68knommu/m68360.h index dd11b070884..eb7d39ef285 100644 --- a/include/asm-m68knommu/m68360.h +++ b/include/asm-m68knommu/m68360.h @@ -3,3 +3,11 @@ #include "m68360_quicc.h" #include "m68360_enet.h" +#ifdef CONFIG_M68360 + +#define CPM_INTERRUPT 4 + +/* see MC68360 User's Manual, p. 7-377 */ +#define CPM_VECTOR_BASE 0x04 /* 3 MSbits of CPM vector */ + +#endif /* CONFIG_M68360 */ diff --git a/include/asm-m68knommu/pgtable.h b/include/asm-m68knommu/pgtable.h index 9dfbbc24aa7..e1e6a1d2333 100644 --- a/include/asm-m68knommu/pgtable.h +++ b/include/asm-m68knommu/pgtable.h @@ -49,7 +49,6 @@ static inline int pte_file(pte_t pte) { return 0; } * These would be in other places but having them here reduces the diffs. */ extern unsigned int kobjsize(const void *objp); -extern int is_in_rom(unsigned long); /* * No page table caches to initialise. diff --git a/include/asm-m68knommu/traps.h b/include/asm-m68knommu/traps.h index f2a81317cc1..d0671e5f8e2 100644 --- a/include/asm-m68knommu/traps.h +++ b/include/asm-m68knommu/traps.h @@ -16,6 +16,10 @@ typedef void (*e_vector)(void); extern e_vector vectors[]; +extern void init_vectors(void); +extern void enable_vector(unsigned int irq); +extern void disable_vector(unsigned int irq); +extern void ack_vector(unsigned int irq); #endif diff --git a/include/asm-m68knommu/uaccess.h b/include/asm-m68knommu/uaccess.h index 62b29b10bc6..9ed9169a884 100644 --- a/include/asm-m68knommu/uaccess.h +++ b/include/asm-m68knommu/uaccess.h @@ -15,12 +15,15 @@ #define access_ok(type,addr,size) _access_ok((unsigned long)(addr),(size)) +/* + * It is not enough to just have access_ok check for a real RAM address. + * This would disallow the case of code/ro-data running XIP in flash/rom. + * Ideally we would check the possible flash ranges too, but that is + * currently not so easy. + */ static inline int _access_ok(unsigned long addr, unsigned long size) { - extern unsigned long memory_start, memory_end; - - return (((addr >= memory_start) && (addr+size < memory_end)) || - (is_in_rom(addr) && is_in_rom(addr+size))); + return 1; } /* diff --git a/include/asm-mips/a.out.h b/include/asm-mips/a.out.h index ef33c3f1348..1ad60ba186d 100644 --- a/include/asm-mips/a.out.h +++ b/include/asm-mips/a.out.h @@ -40,6 +40,7 @@ struct exec #ifdef CONFIG_64BIT #define STACK_TOP (current->thread.mflags & MF_32BIT_ADDR ? TASK_SIZE32 : TASK_SIZE) #endif +#define STACK_TOP_MAX TASK_SIZE #endif diff --git a/include/asm-mips/system.h b/include/asm-mips/system.h index 46bdb3f566f..76339165bc2 100644 --- a/include/asm-mips/system.h +++ b/include/asm-mips/system.h @@ -71,16 +71,6 @@ do { \ write_c0_userlocal(task_thread_info(current)->tp_value);\ } while(0) -/* - * On SMP systems, when the scheduler does migration-cost autodetection, - * it needs a way to flush as much of the CPU's caches as possible. - * - * TODO: fill this in! - */ -static inline void sched_cacheflush(void) -{ -} - static inline unsigned long __xchg_u32(volatile int * m, unsigned int val) { __u32 retval; diff --git a/include/asm-parisc/a.out.h b/include/asm-parisc/a.out.h index 2a490cc9ec9..23e2c90943e 100644 --- a/include/asm-parisc/a.out.h +++ b/include/asm-parisc/a.out.h @@ -23,6 +23,7 @@ struct exec * prumpf */ #define STACK_TOP TASK_SIZE +#define STACK_TOP_MAX DEFAULT_TASK_SIZE #endif diff --git a/include/asm-parisc/system.h b/include/asm-parisc/system.h index 21fbfc5afd0..ee80c920b46 100644 --- a/include/asm-parisc/system.h +++ b/include/asm-parisc/system.h @@ -48,17 +48,6 @@ extern struct task_struct *_switch_to(struct task_struct *, struct task_struct * (last) = _switch_to(prev, next); \ } while(0) -/* - * On SMP systems, when the scheduler does migration-cost autodetection, - * it needs a way to flush as much of the CPU's caches as possible. - * - * TODO: fill this in! - */ -static inline void sched_cacheflush(void) -{ -} - - /* interrupt control */ #define local_save_flags(x) __asm__ __volatile__("ssm 0, %0" : "=r" (x) : : "memory") #define local_irq_disable() __asm__ __volatile__("rsm %0,%%r0\n" : : "i" (PSW_I) : "memory" ) diff --git a/include/asm-powerpc/a.out.h b/include/asm-powerpc/a.out.h index c7393a97736..5c5ea83f934 100644 --- a/include/asm-powerpc/a.out.h +++ b/include/asm-powerpc/a.out.h @@ -26,9 +26,12 @@ struct exec #define STACK_TOP (test_thread_flag(TIF_32BIT) ? \ STACK_TOP_USER32 : STACK_TOP_USER64) +#define STACK_TOP_MAX STACK_TOP_USER64 + #else /* __powerpc64__ */ #define STACK_TOP TASK_SIZE +#define STACK_TOP_MAX STACK_TOP #endif /* __powerpc64__ */ #endif /* __KERNEL__ */ diff --git a/include/asm-powerpc/kprobes.h b/include/asm-powerpc/kprobes.h index 9537fda238b..8b08b447d6f 100644 --- a/include/asm-powerpc/kprobes.h +++ b/include/asm-powerpc/kprobes.h @@ -73,12 +73,10 @@ typedef unsigned int kprobe_opcode_t; } \ } -#define JPROBE_ENTRY(pentry) (kprobe_opcode_t *)((func_descr_t *)pentry) #define is_trap(instr) (IS_TW(instr) || IS_TD(instr) || \ IS_TWI(instr) || IS_TDI(instr)) #else /* Use stock kprobe_lookup_name since ppc32 doesn't use function descriptors */ -#define JPROBE_ENTRY(pentry) (kprobe_opcode_t *)(pentry) #define is_trap(instr) (IS_TW(instr) || IS_TWI(instr)) #endif diff --git a/include/asm-powerpc/percpu.h b/include/asm-powerpc/percpu.h index 2f2e3024fa6..73dc8ba4010 100644 --- a/include/asm-powerpc/percpu.h +++ b/include/asm-powerpc/percpu.h @@ -20,6 +20,11 @@ #define DEFINE_PER_CPU(type, name) \ __attribute__((__section__(".data.percpu"))) __typeof__(type) per_cpu__##name +#define DEFINE_PER_CPU_SHARED_ALIGNED(type, name) \ + __attribute__((__section__(".data.percpu.shared_aligned"))) \ + __typeof__(type) per_cpu__##name \ + ____cacheline_aligned_in_smp + /* var is in discarded region: offset to particular copy we want */ #define per_cpu(var, cpu) (*RELOC_HIDE(&per_cpu__##var, __per_cpu_offset(cpu))) #define __get_cpu_var(var) (*RELOC_HIDE(&per_cpu__##var, __my_cpu_offset())) @@ -40,6 +45,8 @@ extern void setup_per_cpu_areas(void); #define DEFINE_PER_CPU(type, name) \ __typeof__(type) per_cpu__##name +#define DEFINE_PER_CPU_SHARED_ALIGNED(type, name) \ + DEFINE_PER_CPU(type, name) #define per_cpu(var, cpu) (*((void)(cpu), &per_cpu__##var)) #define __get_cpu_var(var) per_cpu__##var diff --git a/include/asm-powerpc/system.h b/include/asm-powerpc/system.h index 32aa42b748b..41520b7a7b7 100644 --- a/include/asm-powerpc/system.h +++ b/include/asm-powerpc/system.h @@ -184,16 +184,6 @@ struct thread_struct; extern struct task_struct *_switch(struct thread_struct *prev, struct thread_struct *next); -/* - * On SMP systems, when the scheduler does migration-cost autodetection, - * it needs a way to flush as much of the CPU's caches as possible. - * - * TODO: fill this in! - */ -static inline void sched_cacheflush(void) -{ -} - extern unsigned int rtas_data; extern int mem_init_done; /* set on boot once kmalloc can be called */ extern unsigned long memory_limit; diff --git a/include/asm-ppc/system.h b/include/asm-ppc/system.h index d84a3cf4d03..f1311a8f310 100644 --- a/include/asm-ppc/system.h +++ b/include/asm-ppc/system.h @@ -129,16 +129,6 @@ extern struct task_struct *__switch_to(struct task_struct *, struct task_struct *); #define switch_to(prev, next, last) ((last) = __switch_to((prev), (next))) -/* - * On SMP systems, when the scheduler does migration-cost autodetection, - * it needs a way to flush as much of the CPU's caches as possible. - * - * TODO: fill this in! - */ -static inline void sched_cacheflush(void) -{ -} - struct thread_struct; extern struct task_struct *_switch(struct thread_struct *prev, struct thread_struct *next); diff --git a/include/asm-s390/a.out.h b/include/asm-s390/a.out.h index 72adee6ef33..46158dcaf51 100644 --- a/include/asm-s390/a.out.h +++ b/include/asm-s390/a.out.h @@ -32,6 +32,7 @@ struct exec #ifdef __KERNEL__ #define STACK_TOP TASK_SIZE +#define STACK_TOP_MAX DEFAULT_TASK_SIZE #endif diff --git a/include/asm-s390/kprobes.h b/include/asm-s390/kprobes.h index 830fe4c4eea..340ba10446e 100644 --- a/include/asm-s390/kprobes.h +++ b/include/asm-s390/kprobes.h @@ -46,8 +46,6 @@ typedef u16 kprobe_opcode_t; ? (MAX_STACK_SIZE) \ : (((unsigned long)current_thread_info()) + THREAD_SIZE - (ADDR))) -#define JPROBE_ENTRY(pentry) (kprobe_opcode_t *)(pentry) - #define ARCH_SUPPORTS_KRETPROBES #define ARCH_INACTIVE_KPROBE_COUNT 0 diff --git a/include/asm-s390/percpu.h b/include/asm-s390/percpu.h index 9ea7f1023e5..545857e6444 100644 --- a/include/asm-s390/percpu.h +++ b/include/asm-s390/percpu.h @@ -41,6 +41,11 @@ extern unsigned long __per_cpu_offset[NR_CPUS]; __attribute__((__section__(".data.percpu"))) \ __typeof__(type) per_cpu__##name +#define DEFINE_PER_CPU_SHARED_ALIGNED(type, name) \ + __attribute__((__section__(".data.percpu.shared_aligned"))) \ + __typeof__(type) per_cpu__##name \ + ____cacheline_aligned_in_smp + #define __get_cpu_var(var) __reloc_hide(var,S390_lowcore.percpu_offset) #define __raw_get_cpu_var(var) __reloc_hide(var,S390_lowcore.percpu_offset) #define per_cpu(var,cpu) __reloc_hide(var,__per_cpu_offset[cpu]) @@ -59,6 +64,8 @@ do { \ #define DEFINE_PER_CPU(type, name) \ __typeof__(type) per_cpu__##name +#define DEFINE_PER_CPU_SHARED_ALIGNED(type, name) \ + DEFINE_PER_CPU(type, name) #define __get_cpu_var(var) __reloc_hide(var,0) #define __raw_get_cpu_var(var) __reloc_hide(var,0) diff --git a/include/asm-s390/system.h b/include/asm-s390/system.h index bbe137c3ed6..64a3cd05cae 100644 --- a/include/asm-s390/system.h +++ b/include/asm-s390/system.h @@ -97,16 +97,6 @@ static inline void restore_access_regs(unsigned int *acrs) prev = __switch_to(prev,next); \ } while (0) -/* - * On SMP systems, when the scheduler does migration-cost autodetection, - * it needs a way to flush as much of the CPU's caches as possible. - * - * TODO: fill this in! - */ -static inline void sched_cacheflush(void) -{ -} - #ifdef CONFIG_VIRT_CPU_ACCOUNTING extern void account_vtime(struct task_struct *); extern void account_tick_vtime(struct task_struct *); diff --git a/include/asm-sh/a.out.h b/include/asm-sh/a.out.h index 6e9fca9ee33..685d0f6125f 100644 --- a/include/asm-sh/a.out.h +++ b/include/asm-sh/a.out.h @@ -20,6 +20,7 @@ struct exec #ifdef __KERNEL__ #define STACK_TOP TASK_SIZE +#define STACK_TOP_MAX STACK_TOP #endif diff --git a/include/asm-sh/system.h b/include/asm-sh/system.h index 7c75045ae22..24504253720 100644 --- a/include/asm-sh/system.h +++ b/include/asm-sh/system.h @@ -64,16 +64,6 @@ struct task_struct *__switch_to(struct task_struct *prev, last = __last; \ } while (0) -/* - * On SMP systems, when the scheduler does migration-cost autodetection, - * it needs a way to flush as much of the CPU's caches as possible. - * - * TODO: fill this in! - */ -static inline void sched_cacheflush(void) -{ -} - #ifdef CONFIG_CPU_SH4A #define __icbi() \ { \ diff --git a/include/asm-sh64/a.out.h b/include/asm-sh64/a.out.h index e1995e86b66..237ee4e5b72 100644 --- a/include/asm-sh64/a.out.h +++ b/include/asm-sh64/a.out.h @@ -31,6 +31,7 @@ struct exec #ifdef __KERNEL__ #define STACK_TOP TASK_SIZE +#define STACK_TOP_MAX STACK_TOP #endif diff --git a/include/asm-sparc/a.out.h b/include/asm-sparc/a.out.h index 9090060a23e..917e0425069 100644 --- a/include/asm-sparc/a.out.h +++ b/include/asm-sparc/a.out.h @@ -92,6 +92,7 @@ struct relocation_info /* used when header.a_machtype == M_SPARC */ #include <asm/page.h> #define STACK_TOP (PAGE_OFFSET - PAGE_SIZE) +#define STACK_TOP_MAX STACK_TOP #endif /* __KERNEL__ */ diff --git a/include/asm-sparc/system.h b/include/asm-sparc/system.h index 8b4e23b3bb3..d1a2572e3f5 100644 --- a/include/asm-sparc/system.h +++ b/include/asm-sparc/system.h @@ -165,16 +165,6 @@ extern void fpsave(unsigned long *fpregs, unsigned long *fsr, } while(0) /* - * On SMP systems, when the scheduler does migration-cost autodetection, - * it needs a way to flush as much of the CPU's caches as possible. - * - * TODO: fill this in! - */ -static inline void sched_cacheflush(void) -{ -} - -/* * Changing the IRQ level on the Sparc. */ extern void local_irq_restore(unsigned long); diff --git a/include/asm-sparc64/a.out.h b/include/asm-sparc64/a.out.h index eb3b8e90b27..902e07f89a4 100644 --- a/include/asm-sparc64/a.out.h +++ b/include/asm-sparc64/a.out.h @@ -101,6 +101,8 @@ struct relocation_info /* used when header.a_machtype == M_SPARC */ #define STACK_TOP (test_thread_flag(TIF_32BIT) ? \ STACK_TOP32 : STACK_TOP64) +#define STACK_TOP_MAX STACK_TOP64 + #endif #endif /* !(__ASSEMBLY__) */ diff --git a/include/asm-sparc64/kprobes.h b/include/asm-sparc64/kprobes.h index a331b7b0dff..7f6774dca5f 100644 --- a/include/asm-sparc64/kprobes.h +++ b/include/asm-sparc64/kprobes.h @@ -10,7 +10,6 @@ typedef u32 kprobe_opcode_t; #define BREAKPOINT_INSTRUCTION_2 0x91d02071 /* ta 0x71 */ #define MAX_INSN_SIZE 2 -#define JPROBE_ENTRY(pentry) (kprobe_opcode_t *)pentry #define arch_remove_kprobe(p) do {} while (0) #define ARCH_INACTIVE_KPROBE_COUNT 0 diff --git a/include/asm-sparc64/percpu.h b/include/asm-sparc64/percpu.h index 88db872ce2f..caf8750792f 100644 --- a/include/asm-sparc64/percpu.h +++ b/include/asm-sparc64/percpu.h @@ -18,6 +18,11 @@ extern unsigned long __per_cpu_shift; #define DEFINE_PER_CPU(type, name) \ __attribute__((__section__(".data.percpu"))) __typeof__(type) per_cpu__##name +#define DEFINE_PER_CPU_SHARED_ALIGNED(type, name) \ + __attribute__((__section__(".data.percpu.shared_aligned"))) \ + __typeof__(type) per_cpu__##name \ + ____cacheline_aligned_in_smp + register unsigned long __local_per_cpu_offset asm("g5"); /* var is in discarded region: offset to particular copy we want */ @@ -38,6 +43,8 @@ do { \ #define real_setup_per_cpu_areas() do { } while (0) #define DEFINE_PER_CPU(type, name) \ __typeof__(type) per_cpu__##name +#define DEFINE_PER_CPU_SHARED_ALIGNED(type, name) \ + DEFINE_PER_CPU(type, name) #define per_cpu(var, cpu) (*((void)cpu, &per_cpu__##var)) #define __get_cpu_var(var) per_cpu__##var diff --git a/include/asm-sparc64/system.h b/include/asm-sparc64/system.h index 8ba380ec6da..409067408ee 100644 --- a/include/asm-sparc64/system.h +++ b/include/asm-sparc64/system.h @@ -204,16 +204,6 @@ do { if (test_thread_flag(TIF_PERFCTR)) { \ } \ } while(0) -/* - * On SMP systems, when the scheduler does migration-cost autodetection, - * it needs a way to flush as much of the CPU's caches as possible. - * - * TODO: fill this in! - */ -static inline void sched_cacheflush(void) -{ -} - static inline unsigned long xchg32(__volatile__ unsigned int *m, unsigned int val) { unsigned long tmp1, tmp2; diff --git a/include/asm-um/a.out.h b/include/asm-um/a.out.h index 7016b893ac9..78bc9eed26b 100644 --- a/include/asm-um/a.out.h +++ b/include/asm-um/a.out.h @@ -17,4 +17,6 @@ extern int honeypot; #define STACK_TOP \ CHOOSE_MODE((honeypot ? host_task_size : task_size), task_size) +#define STACK_TOP_MAX STACK_TOP + #endif diff --git a/include/asm-x86_64/a.out.h b/include/asm-x86_64/a.out.h index 7255cde0653..e789300e41a 100644 --- a/include/asm-x86_64/a.out.h +++ b/include/asm-x86_64/a.out.h @@ -21,7 +21,8 @@ struct exec #ifdef __KERNEL__ #include <linux/thread_info.h> -#define STACK_TOP TASK_SIZE +#define STACK_TOP TASK_SIZE +#define STACK_TOP_MAX TASK_SIZE64 #endif #endif /* __A_OUT_GNU_H__ */ diff --git a/include/asm-x86_64/kprobes.h b/include/asm-x86_64/kprobes.h index cf5317898fb..7db825403e0 100644 --- a/include/asm-x86_64/kprobes.h +++ b/include/asm-x86_64/kprobes.h @@ -41,7 +41,6 @@ typedef u8 kprobe_opcode_t; ? (MAX_STACK_SIZE) \ : (((unsigned long)current_thread_info()) + THREAD_SIZE - (ADDR))) -#define JPROBE_ENTRY(pentry) (kprobe_opcode_t *)pentry #define ARCH_SUPPORTS_KRETPROBES #define ARCH_INACTIVE_KPROBE_COUNT 1 diff --git a/include/asm-x86_64/percpu.h b/include/asm-x86_64/percpu.h index c6fbb67eac9..5abd4827010 100644 --- a/include/asm-x86_64/percpu.h +++ b/include/asm-x86_64/percpu.h @@ -20,6 +20,11 @@ #define DEFINE_PER_CPU(type, name) \ __attribute__((__section__(".data.percpu"))) __typeof__(type) per_cpu__##name +#define DEFINE_PER_CPU_SHARED_ALIGNED(type, name) \ + __attribute__((__section__(".data.percpu.shared_aligned"))) \ + __typeof__(type) per_cpu__##name \ + ____cacheline_internodealigned_in_smp + /* var is in discarded region: offset to particular copy we want */ #define per_cpu(var, cpu) (*({ \ extern int simple_identifier_##var(void); \ @@ -46,6 +51,8 @@ extern void setup_per_cpu_areas(void); #define DEFINE_PER_CPU(type, name) \ __typeof__(type) per_cpu__##name +#define DEFINE_PER_CPU_SHARED_ALIGNED(type, name) \ + DEFINE_PER_CPU(type, name) #define per_cpu(var, cpu) (*((void)(cpu), &per_cpu__##var)) #define __get_cpu_var(var) per_cpu__##var diff --git a/include/asm-x86_64/system.h b/include/asm-x86_64/system.h index ead9f9a5623..e4f246d62c4 100644 --- a/include/asm-x86_64/system.h +++ b/include/asm-x86_64/system.h @@ -111,15 +111,6 @@ static inline void write_cr4(unsigned long val) #define wbinvd() \ __asm__ __volatile__ ("wbinvd": : :"memory"); -/* - * On SMP systems, when the scheduler does migration-cost autodetection, - * it needs a way to flush as much of the CPU's caches as possible. - */ -static inline void sched_cacheflush(void) -{ - wbinvd(); -} - #endif /* __KERNEL__ */ #define nop() __asm__ __volatile__ ("nop") diff --git a/include/asm-xtensa/a.out.h b/include/asm-xtensa/a.out.h index ffc4dcfd6ac..05a2f67c676 100644 --- a/include/asm-xtensa/a.out.h +++ b/include/asm-xtensa/a.out.h @@ -17,6 +17,7 @@ /* Note: the kernel needs the a.out definitions, even if only ELF is used. */ #define STACK_TOP TASK_SIZE +#define STACK_TOP_MAX STACK_TOP struct exec { diff --git a/include/linux/acpi.h b/include/linux/acpi.h index fccd8b548d9..dc234c508a6 100644 --- a/include/linux/acpi.h +++ b/include/linux/acpi.h @@ -122,7 +122,7 @@ extern struct acpi_mcfg_allocation *pci_mmcfg_config; extern int pci_mmcfg_config_num; extern int sbf_port; -extern unsigned long acpi_video_flags; +extern unsigned long acpi_realmode_flags; #else /* !CONFIG_ACPI */ diff --git a/include/linux/aio.h b/include/linux/aio.h index b903fc02bdb..d10e608f232 100644 --- a/include/linux/aio.h +++ b/include/linux/aio.h @@ -86,7 +86,7 @@ struct kioctx; */ struct kiocb { struct list_head ki_run_list; - long ki_flags; + unsigned long ki_flags; int ki_users; unsigned ki_key; /* id of this request */ diff --git a/include/linux/binfmts.h b/include/linux/binfmts.h index e1a708337be..91c8c07fe8b 100644 --- a/include/linux/binfmts.h +++ b/include/linux/binfmts.h @@ -6,11 +6,13 @@ struct pt_regs; /* - * MAX_ARG_PAGES defines the number of pages allocated for arguments - * and envelope for the new program. 32 should suffice, this gives - * a maximum env+arg of 128kB w/4KB pages! + * These are the maximum length and maximum number of strings passed to the + * execve() system call. MAX_ARG_STRLEN is essentially random but serves to + * prevent the kernel from being unduly impacted by misaddressed pointers. + * MAX_ARG_STRINGS is chosen to fit in a signed 32-bit integer. */ -#define MAX_ARG_PAGES 32 +#define MAX_ARG_STRLEN (PAGE_SIZE * 32) +#define MAX_ARG_STRINGS 0x7FFFFFFF /* sizeof(linux_binprm->buf) */ #define BINPRM_BUF_SIZE 128 @@ -24,7 +26,12 @@ struct pt_regs; */ struct linux_binprm{ char buf[BINPRM_BUF_SIZE]; +#ifdef CONFIG_MMU + struct vm_area_struct *vma; +#else +# define MAX_ARG_PAGES 32 struct page *page[MAX_ARG_PAGES]; +#endif struct mm_struct *mm; unsigned long p; /* current top of mem */ int sh_bang; @@ -40,6 +47,7 @@ struct linux_binprm{ unsigned interp_flags; unsigned interp_data; unsigned long loader, exec; + unsigned long argv_len; }; #define BINPRM_FLAGS_ENFORCE_NONDUMP_BIT 0 @@ -68,7 +76,7 @@ extern int register_binfmt(struct linux_binfmt *); extern int unregister_binfmt(struct linux_binfmt *); extern int prepare_binprm(struct linux_binprm *); -extern void remove_arg_zero(struct linux_binprm *); +extern int __must_check remove_arg_zero(struct linux_binprm *); extern int search_binary_handler(struct linux_binprm *,struct pt_regs *); extern int flush_old_exec(struct linux_binprm * bprm); @@ -85,6 +93,7 @@ extern int suid_dumpable; extern int setup_arg_pages(struct linux_binprm * bprm, unsigned long stack_top, int executable_stack); +extern int bprm_mm_init(struct linux_binprm *bprm); extern int copy_strings_kernel(int argc,char ** argv,struct linux_binprm *bprm); extern void compute_creds(struct linux_binprm *binprm); extern int do_coredump(long signr, int exit_code, struct pt_regs * regs); diff --git a/include/linux/buffer_head.h b/include/linux/buffer_head.h index 5c6e12853a9..35cadad84b1 100644 --- a/include/linux/buffer_head.h +++ b/include/linux/buffer_head.h @@ -209,6 +209,8 @@ int cont_prepare_write(struct page*, unsigned, unsigned, get_block_t*, int generic_cont_expand(struct inode *inode, loff_t size); int generic_cont_expand_simple(struct inode *inode, loff_t size); int block_commit_write(struct page *page, unsigned from, unsigned to); +int block_page_mkwrite(struct vm_area_struct *vma, struct page *page, + get_block_t get_block); void block_sync_page(struct page *); sector_t generic_block_bmap(struct address_space *, sector_t, get_block_t *); int generic_commit_write(struct file *, struct page *, unsigned, unsigned); diff --git a/include/linux/coda_linux.h b/include/linux/coda_linux.h index e4ac016ad27..c4079b403e9 100644 --- a/include/linux/coda_linux.h +++ b/include/linux/coda_linux.h @@ -43,9 +43,6 @@ int coda_revalidate_inode(struct dentry *); int coda_getattr(struct vfsmount *, struct dentry *, struct kstat *); int coda_setattr(struct dentry *, struct iattr *); -/* global variables */ -extern int coda_fake_statfs; - /* this file: heloers */ static __inline__ struct CodaFid *coda_i2f(struct inode *); static __inline__ char *coda_i2s(struct inode *); diff --git a/include/linux/coda_proc.h b/include/linux/coda_proc.h deleted file mode 100644 index 0dc1b0458e7..00000000000 --- a/include/linux/coda_proc.h +++ /dev/null @@ -1,76 +0,0 @@ -/* - * coda_statis.h - * - * CODA operation statistics - * - * (c) March, 1998 - * by Michihiro Kuramochi, Zhenyu Xia and Zhanyong Wan - * zhanyong.wan@yale.edu - * - */ - -#ifndef _CODA_PROC_H -#define _CODA_PROC_H - -void coda_sysctl_init(void); -void coda_sysctl_clean(void); - -#include <linux/sysctl.h> -#include <linux/coda_fs_i.h> -#include <linux/coda.h> - -/* these four files are presented to show the result of the statistics: - * - * /proc/fs/coda/vfs_stats - * cache_inv_stats - * - * these four files are presented to reset the statistics to 0: - * - * /proc/sys/coda/vfs_stats - * cache_inv_stats - */ - -/* VFS operation statistics */ -struct coda_vfs_stats -{ - /* file operations */ - int open; - int flush; - int release; - int fsync; - - /* dir operations */ - int readdir; - - /* inode operations */ - int create; - int lookup; - int link; - int unlink; - int symlink; - int mkdir; - int rmdir; - int rename; - int permission; - - /* symlink operatoins*/ - int follow_link; - int readlink; -}; - -/* cache invalidation statistics */ -struct coda_cache_inv_stats -{ - int flush; - int purge_user; - int zap_dir; - int zap_file; - int zap_vnode; - int purge_fid; - int replace; -}; - -/* these global variables hold the actual statistics data */ -extern struct coda_vfs_stats coda_vfs_stat; - -#endif /* _CODA_PROC_H */ diff --git a/include/linux/coda_psdev.h b/include/linux/coda_psdev.h index b541bb3d1f4..aa8f454b3b7 100644 --- a/include/linux/coda_psdev.h +++ b/include/linux/coda_psdev.h @@ -8,11 +8,6 @@ struct kstatfs; -struct coda_sb_info -{ - struct venus_comm *sbi_vcomm; -}; - /* communication pending/processing queues */ struct venus_comm { u_long vc_seq; @@ -24,9 +19,9 @@ struct venus_comm { }; -static inline struct coda_sb_info *coda_sbp(struct super_block *sb) +static inline struct venus_comm *coda_vcp(struct super_block *sb) { - return ((struct coda_sb_info *)((sb)->s_fs_info)); + return (struct venus_comm *)((sb)->s_fs_info); } @@ -74,8 +69,6 @@ int venus_statfs(struct dentry *dentry, struct kstatfs *sfs); /* messages between coda filesystem in kernel and Venus */ -extern int coda_hard; -extern unsigned long coda_timeout; struct upc_req { struct list_head uc_chain; caddr_t uc_data; @@ -85,7 +78,6 @@ struct upc_req { u_short uc_opcode; /* copied from data to save lookup */ int uc_unique; wait_queue_head_t uc_sleep; /* process' wait queue */ - unsigned long uc_posttime; }; #define REQ_ASYNC 0x1 diff --git a/include/linux/edac.h b/include/linux/edac.h new file mode 100644 index 00000000000..eab451e69a9 --- /dev/null +++ b/include/linux/edac.h @@ -0,0 +1,29 @@ +/* + * Generic EDAC defs + * + * Author: Dave Jiang <djiang@mvista.com> + * + * 2006-2007 (c) MontaVista Software, Inc. This file is licensed under + * the terms of the GNU General Public License version 2. This program + * is licensed "as is" without any warranty of any kind, whether express + * or implied. + * + */ +#ifndef _LINUX_EDAC_H_ +#define _LINUX_EDAC_H_ + +#include <asm/atomic.h> + +#define EDAC_OPSTATE_INVAL -1 +#define EDAC_OPSTATE_POLL 0 +#define EDAC_OPSTATE_NMI 1 +#define EDAC_OPSTATE_INT 2 + +extern int edac_op_state; +extern int edac_err_assert; +extern atomic_t edac_handlers; + +extern int edac_handler_set(void); +extern void edac_atomic_assert_error(void); + +#endif diff --git a/include/linux/freezer.h b/include/linux/freezer.h index 2d38b1a7466..c8e02de737f 100644 --- a/include/linux/freezer.h +++ b/include/linux/freezer.h @@ -25,7 +25,7 @@ static inline int freezing(struct task_struct *p) /* * Request that a process be frozen */ -static inline void freeze(struct task_struct *p) +static inline void set_freeze_flag(struct task_struct *p) { set_tsk_thread_flag(p, TIF_FREEZE); } @@ -33,7 +33,7 @@ static inline void freeze(struct task_struct *p) /* * Sometimes we may need to cancel the previous 'freeze' request */ -static inline void do_not_freeze(struct task_struct *p) +static inline void clear_freeze_flag(struct task_struct *p) { clear_tsk_thread_flag(p, TIF_FREEZE); } @@ -56,7 +56,7 @@ static inline int thaw_process(struct task_struct *p) wake_up_process(p); return 1; } - clear_tsk_thread_flag(p, TIF_FREEZE); + clear_freeze_flag(p); task_unlock(p); return 0; } @@ -129,7 +129,8 @@ static inline void set_freezable(void) #else static inline int frozen(struct task_struct *p) { return 0; } static inline int freezing(struct task_struct *p) { return 0; } -static inline void freeze(struct task_struct *p) { BUG(); } +static inline void set_freeze_flag(struct task_struct *p) {} +static inline void clear_freeze_flag(struct task_struct *p) {} static inline int thaw_process(struct task_struct *p) { return 1; } static inline void refrigerator(void) {} diff --git a/include/linux/fs.h b/include/linux/fs.h index 9562a59b370..d33beadd9a4 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -697,20 +697,26 @@ struct fown_struct { * Track a single file's readahead state */ struct file_ra_state { - unsigned long start; /* Current window */ - unsigned long size; - unsigned long flags; /* ra flags RA_FLAG_xxx*/ - unsigned long cache_hit; /* cache hit count*/ - unsigned long prev_index; /* Cache last read() position */ - unsigned long ahead_start; /* Ahead window */ - unsigned long ahead_size; + pgoff_t start; /* where readahead started */ + unsigned long size; /* # of readahead pages */ + unsigned long async_size; /* do asynchronous readahead when + there are only # of pages ahead */ + unsigned long ra_pages; /* Maximum readahead window */ unsigned long mmap_hit; /* Cache hit stat for mmap accesses */ unsigned long mmap_miss; /* Cache miss stat for mmap accesses */ + unsigned long prev_index; /* Cache last read() position */ unsigned int prev_offset; /* Offset where last read() ended in a page */ }; -#define RA_FLAG_MISS 0x01 /* a cache miss occured against this file */ -#define RA_FLAG_INCACHE 0x02 /* file is already in cache */ + +/* + * Check if @index falls in the readahead windows. + */ +static inline int ra_has_index(struct file_ra_state *ra, pgoff_t index) +{ + return (index >= ra->start && + index < ra->start + ra->size); +} struct file { /* @@ -1463,7 +1469,7 @@ extern int alloc_chrdev_region(dev_t *, unsigned, unsigned, const char *); extern int register_chrdev_region(dev_t, unsigned, const char *); extern int register_chrdev(unsigned int, const char *, const struct file_operations *); -extern int unregister_chrdev(unsigned int, const char *); +extern void unregister_chrdev(unsigned int, const char *); extern void unregister_chrdev_region(dev_t, unsigned); extern int chrdev_open(struct inode *, struct file *); extern void chrdev_show(struct seq_file *,off_t); diff --git a/include/linux/genetlink.h b/include/linux/genetlink.h index f7a93770e1b..7da02c93002 100644 --- a/include/linux/genetlink.h +++ b/include/linux/genetlink.h @@ -39,6 +39,9 @@ enum { CTRL_CMD_NEWOPS, CTRL_CMD_DELOPS, CTRL_CMD_GETOPS, + CTRL_CMD_NEWMCAST_GRP, + CTRL_CMD_DELMCAST_GRP, + CTRL_CMD_GETMCAST_GRP, /* unused */ __CTRL_CMD_MAX, }; @@ -52,6 +55,7 @@ enum { CTRL_ATTR_HDRSIZE, CTRL_ATTR_MAXATTR, CTRL_ATTR_OPS, + CTRL_ATTR_MCAST_GROUPS, __CTRL_ATTR_MAX, }; @@ -66,4 +70,13 @@ enum { #define CTRL_ATTR_OP_MAX (__CTRL_ATTR_OP_MAX - 1) +enum { + CTRL_ATTR_MCAST_GRP_UNSPEC, + CTRL_ATTR_MCAST_GRP_NAME, + CTRL_ATTR_MCAST_GRP_ID, + __CTRL_ATTR_MCAST_GRP_MAX, +}; + +#define CTRL_ATTR_MCAST_GRP_MAX (__CTRL_ATTR_MCAST_GRP_MAX - 1) + #endif /* __LINUX_GENERIC_NETLINK_H */ diff --git a/include/linux/highmem.h b/include/linux/highmem.h index 12c5e4e3135..1fcb0033179 100644 --- a/include/linux/highmem.h +++ b/include/linux/highmem.h @@ -103,21 +103,6 @@ __alloc_zeroed_user_highpage(gfp_t movableflags, #endif /** - * alloc_zeroed_user_highpage - Allocate a zeroed HIGHMEM page for a VMA - * @vma: The VMA the page is to be allocated for - * @vaddr: The virtual address the page will be inserted into - * - * This function will allocate a page for a VMA that the caller knows will - * not be able to move in the future using move_pages() or reclaim. If it - * is known that the page can move, use alloc_zeroed_user_highpage_movable - */ -static inline struct page * -alloc_zeroed_user_highpage(struct vm_area_struct *vma, unsigned long vaddr) -{ - return __alloc_zeroed_user_highpage(0, vma, vaddr); -} - -/** * alloc_zeroed_user_highpage_movable - Allocate a zeroed HIGHMEM page for a VMA that the caller knows can move * @vma: The VMA the page is to be allocated for * @vaddr: The virtual address the page will be inserted into diff --git a/include/linux/i2c-isa.h b/include/linux/i2c-isa.h deleted file mode 100644 index 67e3598c4ce..00000000000 --- a/include/linux/i2c-isa.h +++ /dev/null @@ -1,36 +0,0 @@ -/* - * i2c-isa.h - definitions for the i2c-isa pseudo-i2c-adapter interface - * - * Copyright (C) 2005 Jean Delvare <khali@linux-fr.org> - * - * 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. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, write to the Free Software - * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA. - */ - -#ifndef _LINUX_I2C_ISA_H -#define _LINUX_I2C_ISA_H - -#include <linux/i2c.h> - -extern int i2c_isa_add_driver(struct i2c_driver *driver); -extern int i2c_isa_del_driver(struct i2c_driver *driver); - -/* Detect whether we are on the isa bus. This is only useful to hybrid - (i2c+isa) drivers. */ -#define i2c_is_isa_adapter(adapptr) \ - ((adapptr)->id == I2C_HW_ISA) -#define i2c_is_isa_client(clientptr) \ - i2c_is_isa_adapter((clientptr)->adapter) - -#endif /* _LINUX_I2C_ISA_H */ diff --git a/include/linux/i2c.h b/include/linux/i2c.h index 2eaba21b9b1..0c37a737a2b 100644 --- a/include/linux/i2c.h +++ b/include/linux/i2c.h @@ -368,7 +368,6 @@ struct i2c_client_address_data { /* The numbers to use to set I2C bus address */ #define ANY_I2C_BUS 0xffff -#define ANY_I2C_ISA_BUS 9191 /* ----- functions exported by i2c.o */ diff --git a/include/linux/kprobes.h b/include/linux/kprobes.h index 23adf6075ae..51464d12a4e 100644 --- a/include/linux/kprobes.h +++ b/include/linux/kprobes.h @@ -116,9 +116,12 @@ struct kprobe { */ struct jprobe { struct kprobe kp; - kprobe_opcode_t *entry; /* probe handling code to jump to */ + void *entry; /* probe handling code to jump to */ }; +/* For backward compatibility with old code using JPROBE_ENTRY() */ +#define JPROBE_ENTRY(handler) (handler) + DECLARE_PER_CPU(struct kprobe *, current_kprobe); DECLARE_PER_CPU(struct kprobe_ctlblk, kprobe_ctlblk); @@ -211,6 +214,7 @@ int longjmp_break_handler(struct kprobe *, struct pt_regs *); int register_jprobe(struct jprobe *p); void unregister_jprobe(struct jprobe *p); void jprobe_return(void); +unsigned long arch_deref_entry_point(void *); int register_kretprobe(struct kretprobe *rp); void unregister_kretprobe(struct kretprobe *rp); diff --git a/include/linux/lguest.h b/include/linux/lguest.h new file mode 100644 index 00000000000..500aace21ca --- /dev/null +++ b/include/linux/lguest.h @@ -0,0 +1,85 @@ +/* Things the lguest guest needs to know. Note: like all lguest interfaces, + * this is subject to wild and random change between versions. */ +#ifndef _ASM_LGUEST_H +#define _ASM_LGUEST_H + +#ifndef __ASSEMBLY__ +#include <asm/irq.h> + +#define LHCALL_FLUSH_ASYNC 0 +#define LHCALL_LGUEST_INIT 1 +#define LHCALL_CRASH 2 +#define LHCALL_LOAD_GDT 3 +#define LHCALL_NEW_PGTABLE 4 +#define LHCALL_FLUSH_TLB 5 +#define LHCALL_LOAD_IDT_ENTRY 6 +#define LHCALL_SET_STACK 7 +#define LHCALL_TS 8 +#define LHCALL_SET_CLOCKEVENT 9 +#define LHCALL_HALT 10 +#define LHCALL_GET_WALLCLOCK 11 +#define LHCALL_BIND_DMA 12 +#define LHCALL_SEND_DMA 13 +#define LHCALL_SET_PTE 14 +#define LHCALL_SET_PMD 15 +#define LHCALL_LOAD_TLS 16 + +#define LG_CLOCK_MIN_DELTA 100UL +#define LG_CLOCK_MAX_DELTA ULONG_MAX + +#define LGUEST_TRAP_ENTRY 0x1F + +static inline unsigned long +hcall(unsigned long call, + unsigned long arg1, unsigned long arg2, unsigned long arg3) +{ + asm volatile("int $" __stringify(LGUEST_TRAP_ENTRY) + : "=a"(call) + : "a"(call), "d"(arg1), "b"(arg2), "c"(arg3) + : "memory"); + return call; +} + +void async_hcall(unsigned long call, + unsigned long arg1, unsigned long arg2, unsigned long arg3); + +/* Can't use our min() macro here: needs to be a constant */ +#define LGUEST_IRQS (NR_IRQS < 32 ? NR_IRQS: 32) + +#define LHCALL_RING_SIZE 64 +struct hcall_ring +{ + u32 eax, edx, ebx, ecx; +}; + +/* All the good stuff happens here: guest registers it with LGUEST_INIT */ +struct lguest_data +{ +/* Fields which change during running: */ + /* 512 == enabled (same as eflags) */ + unsigned int irq_enabled; + /* Interrupts blocked by guest. */ + DECLARE_BITMAP(blocked_interrupts, LGUEST_IRQS); + + /* Virtual address of page fault. */ + unsigned long cr2; + + /* Async hypercall ring. 0xFF == done, 0 == pending. */ + u8 hcall_status[LHCALL_RING_SIZE]; + struct hcall_ring hcalls[LHCALL_RING_SIZE]; + +/* Fields initialized by the hypervisor at boot: */ + /* Memory not to try to access */ + unsigned long reserve_mem; + /* ID of this guest (used by network driver to set ethernet address) */ + u16 guestid; + /* KHz for the TSC clock. */ + u32 tsc_khz; + +/* Fields initialized by the guest at boot: */ + /* Instruction range to suppress interrupts even if enabled */ + unsigned long noirq_start, noirq_end; +}; +extern struct lguest_data lguest_data; +#endif /* __ASSEMBLY__ */ +#endif /* _ASM_LGUEST_H */ diff --git a/include/linux/lguest_bus.h b/include/linux/lguest_bus.h new file mode 100644 index 00000000000..c9b4e05fee4 --- /dev/null +++ b/include/linux/lguest_bus.h @@ -0,0 +1,48 @@ +#ifndef _ASM_LGUEST_DEVICE_H +#define _ASM_LGUEST_DEVICE_H +/* Everything you need to know about lguest devices. */ +#include <linux/device.h> +#include <linux/lguest.h> +#include <linux/lguest_launcher.h> + +struct lguest_device { + /* Unique busid, and index into lguest_page->devices[] */ + unsigned int index; + + struct device dev; + + /* Driver can hang data off here. */ + void *private; +}; + +/* By convention, each device can use irq index+1 if it wants to. */ +static inline int lgdev_irq(const struct lguest_device *dev) +{ + return dev->index + 1; +} + +/* dma args must not be vmalloced! */ +void lguest_send_dma(unsigned long key, struct lguest_dma *dma); +int lguest_bind_dma(unsigned long key, struct lguest_dma *dmas, + unsigned int num, u8 irq); +void lguest_unbind_dma(unsigned long key, struct lguest_dma *dmas); + +/* Map the virtual device space */ +void *lguest_map(unsigned long phys_addr, unsigned long pages); +void lguest_unmap(void *); + +struct lguest_driver { + const char *name; + struct module *owner; + u16 device_type; + int (*probe)(struct lguest_device *dev); + void (*remove)(struct lguest_device *dev); + + struct device_driver drv; +}; + +extern int register_lguest_driver(struct lguest_driver *drv); +extern void unregister_lguest_driver(struct lguest_driver *drv); + +extern struct lguest_device_desc *lguest_devices; /* Just past max_pfn */ +#endif /* _ASM_LGUEST_DEVICE_H */ diff --git a/include/linux/lguest_launcher.h b/include/linux/lguest_launcher.h new file mode 100644 index 00000000000..0ba414a40c8 --- /dev/null +++ b/include/linux/lguest_launcher.h @@ -0,0 +1,73 @@ +#ifndef _ASM_LGUEST_USER +#define _ASM_LGUEST_USER +/* Everything the "lguest" userspace program needs to know. */ +/* They can register up to 32 arrays of lguest_dma. */ +#define LGUEST_MAX_DMA 32 +/* At most we can dma 16 lguest_dma in one op. */ +#define LGUEST_MAX_DMA_SECTIONS 16 + +/* How many devices? Assume each one wants up to two dma arrays per device. */ +#define LGUEST_MAX_DEVICES (LGUEST_MAX_DMA/2) + +struct lguest_dma +{ + /* 0 if free to be used, filled by hypervisor. */ + u32 used_len; + unsigned long addr[LGUEST_MAX_DMA_SECTIONS]; + u16 len[LGUEST_MAX_DMA_SECTIONS]; +}; + +struct lguest_block_page +{ + /* 0 is a read, 1 is a write. */ + int type; + u32 sector; /* Offset in device = sector * 512. */ + u32 bytes; /* Length expected to be read/written in bytes */ + /* 0 = pending, 1 = done, 2 = done, error */ + int result; + u32 num_sectors; /* Disk length = num_sectors * 512 */ +}; + +/* There is a shared page of these. */ +struct lguest_net +{ + /* Simply the mac address (with multicast bit meaning promisc). */ + unsigned char mac[6]; +}; + +/* Where the Host expects the Guest to SEND_DMA console output to. */ +#define LGUEST_CONSOLE_DMA_KEY 0 + +/* We have a page of these descriptors in the lguest_device page. */ +struct lguest_device_desc { + u16 type; +#define LGUEST_DEVICE_T_CONSOLE 1 +#define LGUEST_DEVICE_T_NET 2 +#define LGUEST_DEVICE_T_BLOCK 3 + + u16 features; +#define LGUEST_NET_F_NOCSUM 0x4000 /* Don't bother checksumming */ +#define LGUEST_DEVICE_F_RANDOMNESS 0x8000 /* IRQ is fairly random */ + + u16 status; +/* 256 and above are device specific. */ +#define LGUEST_DEVICE_S_ACKNOWLEDGE 1 /* We have seen device. */ +#define LGUEST_DEVICE_S_DRIVER 2 /* We have found a driver */ +#define LGUEST_DEVICE_S_DRIVER_OK 4 /* Driver says OK! */ +#define LGUEST_DEVICE_S_REMOVED 8 /* Device has gone away. */ +#define LGUEST_DEVICE_S_REMOVED_ACK 16 /* Driver has been told. */ +#define LGUEST_DEVICE_S_FAILED 128 /* Something actually failed */ + + u16 num_pages; + u32 pfn; +}; + +/* Write command first word is a request. */ +enum lguest_req +{ + LHREQ_INITIALIZE, /* + pfnlimit, pgdir, start, pageoffset */ + LHREQ_GETDMA, /* + addr (returns &lguest_dma, irq in ->used_len) */ + LHREQ_IRQ, /* + irq */ + LHREQ_BREAK, /* + on/off flag (on blocks until someone does off) */ +}; +#endif /* _ASM_LGUEST_USER */ diff --git a/include/linux/lockdep.h b/include/linux/lockdep.h index 14c937d345c..0e843bf6587 100644 --- a/include/linux/lockdep.h +++ b/include/linux/lockdep.h @@ -1,7 +1,8 @@ /* * Runtime locking correctness validator * - * Copyright (C) 2006 Red Hat, Inc., Ingo Molnar <mingo@redhat.com> + * Copyright (C) 2006,2007 Red Hat, Inc., Ingo Molnar <mingo@redhat.com> + * Copyright (C) 2007 Red Hat, Inc., Peter Zijlstra <pzijlstr@redhat.com> * * see Documentation/lockdep-design.txt for more details. */ @@ -9,6 +10,7 @@ #define __LINUX_LOCKDEP_H struct task_struct; +struct lockdep_map; #ifdef CONFIG_LOCKDEP @@ -114,8 +116,44 @@ struct lock_class { const char *name; int name_version; + +#ifdef CONFIG_LOCK_STAT + unsigned long contention_point[4]; +#endif +}; + +#ifdef CONFIG_LOCK_STAT +struct lock_time { + s64 min; + s64 max; + s64 total; + unsigned long nr; +}; + +enum bounce_type { + bounce_acquired_write, + bounce_acquired_read, + bounce_contended_write, + bounce_contended_read, + nr_bounce_types, + + bounce_acquired = bounce_acquired_write, + bounce_contended = bounce_contended_write, }; +struct lock_class_stats { + unsigned long contention_point[4]; + struct lock_time read_waittime; + struct lock_time write_waittime; + struct lock_time read_holdtime; + struct lock_time write_holdtime; + unsigned long bounces[nr_bounce_types]; +}; + +struct lock_class_stats lock_stats(struct lock_class *class); +void clear_lock_stats(struct lock_class *class); +#endif + /* * Map the lock object (the lock instance) to the lock-class object. * This is embedded into specific lock instances: @@ -124,6 +162,9 @@ struct lockdep_map { struct lock_class_key *key; struct lock_class *class_cache; const char *name; +#ifdef CONFIG_LOCK_STAT + int cpu; +#endif }; /* @@ -165,6 +206,10 @@ struct held_lock { unsigned long acquire_ip; struct lockdep_map *instance; +#ifdef CONFIG_LOCK_STAT + u64 waittime_stamp; + u64 holdtime_stamp; +#endif /* * The lock-stack is unified in that the lock chains of interrupt * contexts nest ontop of process context chains, but we 'separate' @@ -281,6 +326,30 @@ struct lock_class_key { }; #endif /* !LOCKDEP */ +#ifdef CONFIG_LOCK_STAT + +extern void lock_contended(struct lockdep_map *lock, unsigned long ip); +extern void lock_acquired(struct lockdep_map *lock); + +#define LOCK_CONTENDED(_lock, try, lock) \ +do { \ + if (!try(_lock)) { \ + lock_contended(&(_lock)->dep_map, _RET_IP_); \ + lock(_lock); \ + } \ + lock_acquired(&(_lock)->dep_map); \ +} while (0) + +#else /* CONFIG_LOCK_STAT */ + +#define lock_contended(lockdep_map, ip) do {} while (0) +#define lock_acquired(lockdep_map) do {} while (0) + +#define LOCK_CONTENDED(_lock, try, lock) \ + lock(_lock) + +#endif /* CONFIG_LOCK_STAT */ + #if defined(CONFIG_TRACE_IRQFLAGS) && defined(CONFIG_GENERIC_HARDIRQS) extern void early_init_irq_lock_class(void); #else diff --git a/include/linux/mm.h b/include/linux/mm.h index a5c451816fd..c456c3a1c28 100644 --- a/include/linux/mm.h +++ b/include/linux/mm.h @@ -168,6 +168,8 @@ extern unsigned int kobjsize(const void *objp); #define VM_INSERTPAGE 0x02000000 /* The vma has had "vm_insert_page()" done on it */ #define VM_ALWAYSDUMP 0x04000000 /* Always include in core dumps */ +#define VM_CAN_NONLINEAR 0x08000000 /* Has ->fault & does nonlinear pages */ + #ifndef VM_STACK_DEFAULT_FLAGS /* arch can override this */ #define VM_STACK_DEFAULT_FLAGS VM_DATA_DEFAULT_FLAGS #endif @@ -190,6 +192,30 @@ extern unsigned int kobjsize(const void *objp); */ extern pgprot_t protection_map[16]; +#define FAULT_FLAG_WRITE 0x01 /* Fault was a write access */ +#define FAULT_FLAG_NONLINEAR 0x02 /* Fault was via a nonlinear mapping */ + + +/* + * vm_fault is filled by the the pagefault handler and passed to the vma's + * ->fault function. The vma's ->fault is responsible for returning a bitmask + * of VM_FAULT_xxx flags that give details about how the fault was handled. + * + * pgoff should be used in favour of virtual_address, if possible. If pgoff + * is used, one may set VM_CAN_NONLINEAR in the vma->vm_flags to get nonlinear + * mapping support. + */ +struct vm_fault { + unsigned int flags; /* FAULT_FLAG_xxx flags */ + pgoff_t pgoff; /* Logical page offset based on vma */ + void __user *virtual_address; /* Faulting virtual address */ + + struct page *page; /* ->fault handlers should return a + * page here, unless VM_FAULT_NOPAGE + * is set (which is also implied by + * VM_FAULT_ERROR). + */ +}; /* * These are the virtual MM functions - opening of an area, closing and @@ -199,9 +225,11 @@ extern pgprot_t protection_map[16]; struct vm_operations_struct { void (*open)(struct vm_area_struct * area); void (*close)(struct vm_area_struct * area); - struct page * (*nopage)(struct vm_area_struct * area, unsigned long address, int *type); - unsigned long (*nopfn)(struct vm_area_struct * area, unsigned long address); - int (*populate)(struct vm_area_struct * area, unsigned long address, unsigned long len, pgprot_t prot, unsigned long pgoff, int nonblock); + int (*fault)(struct vm_area_struct *vma, struct vm_fault *vmf); + struct page *(*nopage)(struct vm_area_struct *area, + unsigned long address, int *type); + unsigned long (*nopfn)(struct vm_area_struct *area, + unsigned long address); /* notification that a previously read-only page is about to become * writable, if an error is returned it will cause a SIGBUS */ @@ -655,7 +683,6 @@ static inline int page_mapped(struct page *page) */ #define NOPAGE_SIGBUS (NULL) #define NOPAGE_OOM ((struct page *) (-1)) -#define NOPAGE_REFAULT ((struct page *) (-2)) /* Return to userspace, rerun */ /* * Error return values for the *_nopfn functions @@ -669,16 +696,18 @@ static inline int page_mapped(struct page *page) * Used to decide whether a process gets delivered SIGBUS or * just gets major/minor fault counters bumped up. */ -#define VM_FAULT_OOM 0x00 -#define VM_FAULT_SIGBUS 0x01 -#define VM_FAULT_MINOR 0x02 -#define VM_FAULT_MAJOR 0x03 - -/* - * Special case for get_user_pages. - * Must be in a distinct bit from the above VM_FAULT_ flags. - */ -#define VM_FAULT_WRITE 0x10 + +#define VM_FAULT_MINOR 0 /* For backwards compat. Remove me quickly. */ + +#define VM_FAULT_OOM 0x0001 +#define VM_FAULT_SIGBUS 0x0002 +#define VM_FAULT_MAJOR 0x0004 +#define VM_FAULT_WRITE 0x0008 /* Special case for get_user_pages */ + +#define VM_FAULT_NOPAGE 0x0100 /* ->fault installed the pte, not return page */ +#define VM_FAULT_LOCKED 0x0200 /* ->fault locked the returned page */ + +#define VM_FAULT_ERROR (VM_FAULT_OOM | VM_FAULT_SIGBUS) #define offset_in_page(p) ((unsigned long)(p) & ~PAGE_MASK) @@ -762,20 +791,10 @@ static inline void unmap_shared_mapping_range(struct address_space *mapping, extern int vmtruncate(struct inode * inode, loff_t offset); extern int vmtruncate_range(struct inode * inode, loff_t offset, loff_t end); -extern int install_page(struct mm_struct *mm, struct vm_area_struct *vma, unsigned long addr, struct page *page, pgprot_t prot); -extern int install_file_pte(struct mm_struct *mm, struct vm_area_struct *vma, unsigned long addr, unsigned long pgoff, pgprot_t prot); #ifdef CONFIG_MMU -extern int __handle_mm_fault(struct mm_struct *mm,struct vm_area_struct *vma, +extern int handle_mm_fault(struct mm_struct *mm, struct vm_area_struct *vma, unsigned long address, int write_access); - -static inline int handle_mm_fault(struct mm_struct *mm, - struct vm_area_struct *vma, unsigned long address, - int write_access) -{ - return __handle_mm_fault(mm, vma, address, write_access) & - (~VM_FAULT_WRITE); -} #else static inline int handle_mm_fault(struct mm_struct *mm, struct vm_area_struct *vma, unsigned long address, @@ -789,7 +808,6 @@ static inline int handle_mm_fault(struct mm_struct *mm, extern int make_pages_present(unsigned long addr, unsigned long end); extern int access_process_vm(struct task_struct *tsk, unsigned long addr, void *buf, int len, int write); -void install_arg_page(struct vm_area_struct *, struct page *, unsigned long); int get_user_pages(struct task_struct *tsk, struct mm_struct *mm, unsigned long start, int len, int write, int force, struct page **pages, struct vm_area_struct **vmas); @@ -806,9 +824,15 @@ int FASTCALL(set_page_dirty(struct page *page)); int set_page_dirty_lock(struct page *page); int clear_page_dirty_for_io(struct page *page); +extern unsigned long move_page_tables(struct vm_area_struct *vma, + unsigned long old_addr, struct vm_area_struct *new_vma, + unsigned long new_addr, unsigned long len); extern unsigned long do_mremap(unsigned long addr, unsigned long old_len, unsigned long new_len, unsigned long flags, unsigned long new_addr); +extern int mprotect_fixup(struct vm_area_struct *vma, + struct vm_area_struct **pprev, unsigned long start, + unsigned long end, unsigned long newflags); /* * A callback you can register to apply pressure to ageable caches. @@ -1104,9 +1128,7 @@ extern void truncate_inode_pages_range(struct address_space *, loff_t lstart, loff_t lend); /* generic vm_area_ops exported for stackable file systems */ -extern struct page *filemap_nopage(struct vm_area_struct *, unsigned long, int *); -extern int filemap_populate(struct vm_area_struct *, unsigned long, - unsigned long, pgprot_t, unsigned long, int); +extern int filemap_fault(struct vm_area_struct *, struct vm_fault *); /* mm/page-writeback.c */ int write_one_page(struct page *page, int wait); @@ -1121,13 +1143,20 @@ int do_page_cache_readahead(struct address_space *mapping, struct file *filp, pgoff_t offset, unsigned long nr_to_read); int force_page_cache_readahead(struct address_space *mapping, struct file *filp, pgoff_t offset, unsigned long nr_to_read); -unsigned long page_cache_readahead(struct address_space *mapping, - struct file_ra_state *ra, - struct file *filp, - pgoff_t offset, - unsigned long size); -void handle_ra_miss(struct address_space *mapping, - struct file_ra_state *ra, pgoff_t offset); + +void page_cache_sync_readahead(struct address_space *mapping, + struct file_ra_state *ra, + struct file *filp, + pgoff_t offset, + unsigned long size); + +void page_cache_async_readahead(struct address_space *mapping, + struct file_ra_state *ra, + struct file *filp, + struct page *pg, + pgoff_t offset, + unsigned long size); + unsigned long max_sane_readahead(unsigned long nr); /* Do stack extension */ @@ -1135,6 +1164,8 @@ extern int expand_stack(struct vm_area_struct *vma, unsigned long address); #ifdef CONFIG_IA64 extern int expand_upwards(struct vm_area_struct *vma, unsigned long address); #endif +extern int expand_stack_downwards(struct vm_area_struct *vma, + unsigned long address); /* Look up the first VMA which satisfies addr < vm_end, NULL if none. */ extern struct vm_area_struct * find_vma(struct mm_struct * mm, unsigned long addr); diff --git a/include/linux/namei.h b/include/linux/namei.h index b7dd24917f0..6c38efbd810 100644 --- a/include/linux/namei.h +++ b/include/linux/namei.h @@ -69,8 +69,8 @@ extern int FASTCALL(__user_walk_fd(int dfd, const char __user *, unsigned, struc #define user_path_walk_link(name,nd) \ __user_walk_fd(AT_FDCWD, name, 0, nd) extern int FASTCALL(path_lookup(const char *, unsigned, struct nameidata *)); -extern int FASTCALL(path_walk(const char *, struct nameidata *)); -extern int FASTCALL(link_path_walk(const char *, struct nameidata *)); +extern int vfs_path_lookup(struct dentry *, struct vfsmount *, + const char *, unsigned int, struct nameidata *); extern void path_release(struct nameidata *); extern void path_release_on_umount(struct nameidata *); diff --git a/include/linux/netlink.h b/include/linux/netlink.h index 2e23353c28a..83d8239f0cc 100644 --- a/include/linux/netlink.h +++ b/include/linux/netlink.h @@ -161,6 +161,8 @@ extern struct sock *netlink_kernel_create(int unit, unsigned int groups, void (*input)(struct sock *sk, int len), struct mutex *cb_mutex, struct module *module); +extern int netlink_change_ngroups(struct sock *sk, unsigned int groups); +extern void netlink_clear_multicast_users(struct sock *sk, unsigned int group); extern void netlink_ack(struct sk_buff *in_skb, struct nlmsghdr *nlh, int err); extern int netlink_has_listeners(struct sock *sk, unsigned int group); extern int netlink_unicast(struct sock *ssk, struct sk_buff *skb, __u32 pid, int nonblock); diff --git a/include/linux/nfs_fs.h b/include/linux/nfs_fs.h index c098ae194f7..9ba4aec37c5 100644 --- a/include/linux/nfs_fs.h +++ b/include/linux/nfs_fs.h @@ -407,8 +407,8 @@ extern void nfs_release_automount_timer(void); /* * linux/fs/nfs/unlink.c */ -extern int nfs_async_unlink(struct dentry *); -extern void nfs_complete_unlink(struct dentry *); +extern int nfs_async_unlink(struct inode *dir, struct dentry *dentry); +extern void nfs_complete_unlink(struct dentry *dentry, struct inode *); /* * linux/fs/nfs/write.c diff --git a/include/linux/nfs_xdr.h b/include/linux/nfs_xdr.h index 38d77681cf2..cf74a4db84a 100644 --- a/include/linux/nfs_xdr.h +++ b/include/linux/nfs_xdr.h @@ -278,6 +278,21 @@ struct nfs_writeres { }; /* + * Common arguments to the unlink call + */ +struct nfs_removeargs { + const struct nfs_fh *fh; + struct qstr name; + const u32 * bitmask; +}; + +struct nfs_removeres { + const struct nfs_server *server; + struct nfs4_change_info cinfo; + struct nfs_fattr dir_attr; +}; + +/* * Argument struct for decode_entry function */ struct nfs_entry { @@ -631,18 +646,6 @@ struct nfs4_readlink { struct page ** pages; /* zero-copy data */ }; -struct nfs4_remove_arg { - const struct nfs_fh * fh; - const struct qstr * name; - const u32 * bitmask; -}; - -struct nfs4_remove_res { - const struct nfs_server * server; - struct nfs4_change_info cinfo; - struct nfs_fattr * dir_attr; -}; - struct nfs4_rename_arg { const struct nfs_fh * old_dir; const struct nfs_fh * new_dir; @@ -788,9 +791,8 @@ struct nfs_rpc_ops { int (*create) (struct inode *, struct dentry *, struct iattr *, int, struct nameidata *); int (*remove) (struct inode *, struct qstr *); - int (*unlink_setup) (struct rpc_message *, - struct dentry *, struct qstr *); - int (*unlink_done) (struct dentry *, struct rpc_task *); + void (*unlink_setup) (struct rpc_message *, struct inode *dir); + int (*unlink_done) (struct rpc_task *, struct inode *); int (*rename) (struct inode *, struct qstr *, struct inode *, struct qstr *); int (*link) (struct inode *, struct inode *, struct qstr *); diff --git a/include/linux/nfsd/export.h b/include/linux/nfsd/export.h index 78feb7beff7..5cd19246909 100644 --- a/include/linux/nfsd/export.h +++ b/include/linux/nfsd/export.h @@ -116,18 +116,7 @@ struct svc_expkey { #define EX_NOHIDE(exp) ((exp)->ex_flags & NFSEXP_NOHIDE) #define EX_WGATHER(exp) ((exp)->ex_flags & NFSEXP_GATHERED_WRITES) -static inline int EX_RDONLY(struct svc_export *exp, struct svc_rqst *rqstp) -{ - struct exp_flavor_info *f; - struct exp_flavor_info *end = exp->ex_flavors + exp->ex_nflavors; - - for (f = exp->ex_flavors; f < end; f++) { - if (f->pseudoflavor == rqstp->rq_flavor) - return f->flags & NFSEXP_READONLY; - } - return exp->ex_flags & NFSEXP_READONLY; -} - +int nfsexp_flags(struct svc_rqst *rqstp, struct svc_export *exp); __be32 check_nfsd_access(struct svc_export *exp, struct svc_rqst *rqstp); /* diff --git a/include/linux/notifier.h b/include/linux/notifier.h index 576f2bb34cc..be3f2bb6fcf 100644 --- a/include/linux/notifier.h +++ b/include/linux/notifier.h @@ -212,5 +212,11 @@ extern int __srcu_notifier_call_chain(struct srcu_notifier_head *nh, #define CPU_DEAD_FROZEN (CPU_DEAD | CPU_TASKS_FROZEN) #define CPU_DYING_FROZEN (CPU_DYING | CPU_TASKS_FROZEN) +/* Hibernation and suspend events */ +#define PM_HIBERNATION_PREPARE 0x0001 /* Going to hibernate */ +#define PM_POST_HIBERNATION 0x0002 /* Hibernation finished */ +#define PM_SUSPEND_PREPARE 0x0003 /* Going to suspend the system */ +#define PM_POST_SUSPEND 0x0004 /* Suspend finished */ + #endif /* __KERNEL__ */ #endif /* _LINUX_NOTIFIER_H */ diff --git a/include/linux/page-flags.h b/include/linux/page-flags.h index 731cd2ac322..209d3a47f50 100644 --- a/include/linux/page-flags.h +++ b/include/linux/page-flags.h @@ -90,6 +90,9 @@ #define PG_reclaim 17 /* To be reclaimed asap */ #define PG_buddy 19 /* Page is free, on buddy lists */ +/* PG_readahead is only used for file reads; PG_reclaim is only for writes */ +#define PG_readahead PG_reclaim /* Reminder to do async read-ahead */ + /* PG_owner_priv_1 users should have descriptive aliases */ #define PG_checked PG_owner_priv_1 /* Used by some filesystems */ #define PG_pinned PG_owner_priv_1 /* Xen pinned pagetable */ @@ -186,37 +189,15 @@ static inline void SetPageUptodate(struct page *page) #define __SetPagePrivate(page) __set_bit(PG_private, &(page)->flags) #define __ClearPagePrivate(page) __clear_bit(PG_private, &(page)->flags) +/* + * Only test-and-set exist for PG_writeback. The unconditional operators are + * risky: they bypass page accounting. + */ #define PageWriteback(page) test_bit(PG_writeback, &(page)->flags) -#define SetPageWriteback(page) \ - do { \ - if (!test_and_set_bit(PG_writeback, \ - &(page)->flags)) \ - inc_zone_page_state(page, NR_WRITEBACK); \ - } while (0) -#define TestSetPageWriteback(page) \ - ({ \ - int ret; \ - ret = test_and_set_bit(PG_writeback, \ - &(page)->flags); \ - if (!ret) \ - inc_zone_page_state(page, NR_WRITEBACK); \ - ret; \ - }) -#define ClearPageWriteback(page) \ - do { \ - if (test_and_clear_bit(PG_writeback, \ - &(page)->flags)) \ - dec_zone_page_state(page, NR_WRITEBACK); \ - } while (0) -#define TestClearPageWriteback(page) \ - ({ \ - int ret; \ - ret = test_and_clear_bit(PG_writeback, \ - &(page)->flags); \ - if (ret) \ - dec_zone_page_state(page, NR_WRITEBACK); \ - ret; \ - }) +#define TestSetPageWriteback(page) test_and_set_bit(PG_writeback, \ + &(page)->flags) +#define TestClearPageWriteback(page) test_and_clear_bit(PG_writeback, \ + &(page)->flags) #define PageBuddy(page) test_bit(PG_buddy, &(page)->flags) #define __SetPageBuddy(page) __set_bit(PG_buddy, &(page)->flags) @@ -226,6 +207,10 @@ static inline void SetPageUptodate(struct page *page) #define SetPageMappedToDisk(page) set_bit(PG_mappedtodisk, &(page)->flags) #define ClearPageMappedToDisk(page) clear_bit(PG_mappedtodisk, &(page)->flags) +#define PageReadahead(page) test_bit(PG_readahead, &(page)->flags) +#define SetPageReadahead(page) set_bit(PG_readahead, &(page)->flags) +#define ClearPageReadahead(page) clear_bit(PG_readahead, &(page)->flags) + #define PageReclaim(page) test_bit(PG_reclaim, &(page)->flags) #define SetPageReclaim(page) set_bit(PG_reclaim, &(page)->flags) #define ClearPageReclaim(page) clear_bit(PG_reclaim, &(page)->flags) diff --git a/include/linux/pci_ids.h b/include/linux/pci_ids.h index 2c7add16953..b15c6498fe6 100644 --- a/include/linux/pci_ids.h +++ b/include/linux/pci_ids.h @@ -495,6 +495,8 @@ #define PCI_VENDOR_ID_AMD 0x1022 #define PCI_DEVICE_ID_AMD_K8_NB 0x1100 +#define PCI_DEVICE_ID_AMD_K8_NB_ADDRMAP 0x1101 +#define PCI_DEVICE_ID_AMD_K8_NB_MEMCTL 0x1102 #define PCI_DEVICE_ID_AMD_K8_NB_MISC 0x1103 #define PCI_DEVICE_ID_AMD_LANCE 0x2000 #define PCI_DEVICE_ID_AMD_LANCE_HOME 0x2001 @@ -2209,6 +2211,7 @@ #define PCI_DEVICE_ID_INTEL_82915GM_IG 0x2592 #define PCI_DEVICE_ID_INTEL_82945G_HB 0x2770 #define PCI_DEVICE_ID_INTEL_82945G_IG 0x2772 +#define PCI_DEVICE_ID_INTEL_3000_HB 0x2778 #define PCI_DEVICE_ID_INTEL_82945GM_HB 0x27A0 #define PCI_DEVICE_ID_INTEL_82945GM_IG 0x27A2 #define PCI_DEVICE_ID_INTEL_ICH6_0 0x2640 diff --git a/include/linux/pm.h b/include/linux/pm.h index 2735b7cadd2..ad3cc2eb0d3 100644 --- a/include/linux/pm.h +++ b/include/linux/pm.h @@ -101,6 +101,7 @@ struct pm_dev */ extern void (*pm_idle)(void); extern void (*pm_power_off)(void); +extern void (*pm_power_off_prepare)(void); typedef int __bitwise suspend_state_t; diff --git a/include/linux/sched.h b/include/linux/sched.h index 731edaca8ff..33b9b4841ee 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -345,6 +345,27 @@ typedef unsigned long mm_counter_t; (mm)->hiwater_vm = (mm)->total_vm; \ } while (0) +extern void set_dumpable(struct mm_struct *mm, int value); +extern int get_dumpable(struct mm_struct *mm); + +/* mm flags */ +/* dumpable bits */ +#define MMF_DUMPABLE 0 /* core dump is permitted */ +#define MMF_DUMP_SECURELY 1 /* core file is readable only by root */ +#define MMF_DUMPABLE_BITS 2 + +/* coredump filter bits */ +#define MMF_DUMP_ANON_PRIVATE 2 +#define MMF_DUMP_ANON_SHARED 3 +#define MMF_DUMP_MAPPED_PRIVATE 4 +#define MMF_DUMP_MAPPED_SHARED 5 +#define MMF_DUMP_FILTER_SHIFT MMF_DUMPABLE_BITS +#define MMF_DUMP_FILTER_BITS 4 +#define MMF_DUMP_FILTER_MASK \ + (((1 << MMF_DUMP_FILTER_BITS) - 1) << MMF_DUMP_FILTER_SHIFT) +#define MMF_DUMP_FILTER_DEFAULT \ + ((1 << MMF_DUMP_ANON_PRIVATE) | (1 << MMF_DUMP_ANON_SHARED)) + struct mm_struct { struct vm_area_struct * mmap; /* list of VMAs */ struct rb_root mm_rb; @@ -402,7 +423,7 @@ struct mm_struct { unsigned int token_priority; unsigned int last_interval; - unsigned char dumpable:2; + unsigned long flags; /* Must use atomic bitops to access the bits */ /* coredumping support */ int core_waiters; @@ -1327,6 +1348,13 @@ static inline int set_cpus_allowed(struct task_struct *p, cpumask_t new_mask) #endif extern unsigned long long sched_clock(void); + +/* + * For kernel-internal use: high-speed (but slightly incorrect) per-cpu + * clock constructed from sched_clock(): + */ +extern unsigned long long cpu_clock(int cpu); + extern unsigned long long task_sched_runtime(struct task_struct *task); diff --git a/include/linux/spinlock_types.h b/include/linux/spinlock_types.h index 210549ba4ef..f6a3a951b79 100644 --- a/include/linux/spinlock_types.h +++ b/include/linux/spinlock_types.h @@ -9,14 +9,14 @@ * Released under the General Public License (GPL). */ -#include <linux/lockdep.h> - #if defined(CONFIG_SMP) # include <asm/spinlock_types.h> #else # include <linux/spinlock_types_up.h> #endif +#include <linux/lockdep.h> + typedef struct { raw_spinlock_t raw_lock; #if defined(CONFIG_PREEMPT) && defined(CONFIG_SMP) diff --git a/include/linux/spinlock_types_up.h b/include/linux/spinlock_types_up.h index 27644af20b7..04135b0e198 100644 --- a/include/linux/spinlock_types_up.h +++ b/include/linux/spinlock_types_up.h @@ -12,14 +12,10 @@ * Released under the General Public License (GPL). */ -#if defined(CONFIG_DEBUG_SPINLOCK) || \ - defined(CONFIG_DEBUG_LOCK_ALLOC) +#ifdef CONFIG_DEBUG_SPINLOCK typedef struct { volatile unsigned int slock; -#ifdef CONFIG_DEBUG_LOCK_ALLOC - struct lockdep_map dep_map; -#endif } raw_spinlock_t; #define __RAW_SPIN_LOCK_UNLOCKED { 1 } @@ -34,9 +30,6 @@ typedef struct { } raw_spinlock_t; typedef struct { /* no debug version on UP */ -#ifdef CONFIG_DEBUG_LOCK_ALLOC - struct lockdep_map dep_map; -#endif } raw_rwlock_t; #define __RAW_RW_LOCK_UNLOCKED { } diff --git a/include/linux/stacktrace.h b/include/linux/stacktrace.h index 1d2b084c018..e7fa657d0c4 100644 --- a/include/linux/stacktrace.h +++ b/include/linux/stacktrace.h @@ -13,7 +13,7 @@ extern void save_stack_trace(struct stack_trace *trace); extern void print_stack_trace(struct stack_trace *trace, int spaces); #else # define save_stack_trace(trace) do { } while (0) -# define print_stack_trace(trace) do { } while (0) +# define print_stack_trace(trace, spaces) do { } while (0) #endif #endif diff --git a/include/linux/sunrpc/xdr.h b/include/linux/sunrpc/xdr.h index 9e340fa23c0..c6b53d181bf 100644 --- a/include/linux/sunrpc/xdr.h +++ b/include/linux/sunrpc/xdr.h @@ -12,6 +12,7 @@ #include <linux/uio.h> #include <asm/byteorder.h> #include <linux/scatterlist.h> +#include <linux/smp_lock.h> /* * Buffer adjustment @@ -36,6 +37,21 @@ struct xdr_netobj { typedef int (*kxdrproc_t)(void *rqstp, __be32 *data, void *obj); /* + * We're still requiring the BKL in the xdr code until it's been + * more carefully audited, at which point this wrapper will become + * unnecessary. + */ +static inline int rpc_call_xdrproc(kxdrproc_t xdrproc, void *rqstp, __be32 *data, void *obj) +{ + int ret; + + lock_kernel(); + ret = xdrproc(rqstp, data, obj); + unlock_kernel(); + return ret; +} + +/* * Basic structure for transmission/reception of a client XDR message. * Features a header (for a linear buffer containing RPC headers * and the data payload for short messages), and then an array of diff --git a/include/linux/suspend.h b/include/linux/suspend.h index 9c7cb643066..e8e6da394c9 100644 --- a/include/linux/suspend.h +++ b/include/linux/suspend.h @@ -43,14 +43,19 @@ static inline void pm_restore_console(void) {} * @prepare: prepare system for hibernation * @enter: shut down system after state has been saved to disk * @finish: finish/clean up after state has been reloaded + * @pre_restore: prepare system for the restoration from a hibernation image + * @restore_cleanup: clean up after a failing image restoration */ struct hibernation_ops { int (*prepare)(void); int (*enter)(void); void (*finish)(void); + int (*pre_restore)(void); + void (*restore_cleanup)(void); }; -#if defined(CONFIG_PM) && defined(CONFIG_SOFTWARE_SUSPEND) +#ifdef CONFIG_PM +#ifdef CONFIG_SOFTWARE_SUSPEND /* kernel/power/snapshot.c */ extern void __register_nosave_region(unsigned long b, unsigned long e, int km); static inline void register_nosave_region(unsigned long b, unsigned long e) @@ -68,16 +73,14 @@ extern unsigned long get_safe_page(gfp_t gfp_mask); extern void hibernation_set_ops(struct hibernation_ops *ops); extern int hibernate(void); -#else -static inline void register_nosave_region(unsigned long b, unsigned long e) {} -static inline void register_nosave_region_late(unsigned long b, unsigned long e) {} +#else /* CONFIG_SOFTWARE_SUSPEND */ static inline int swsusp_page_is_forbidden(struct page *p) { return 0; } static inline void swsusp_set_page_free(struct page *p) {} static inline void swsusp_unset_page_free(struct page *p) {} static inline void hibernation_set_ops(struct hibernation_ops *ops) {} static inline int hibernate(void) { return -ENOSYS; } -#endif /* defined(CONFIG_PM) && defined(CONFIG_SOFTWARE_SUSPEND) */ +#endif /* CONFIG_SOFTWARE_SUSPEND */ void save_processor_state(void); void restore_processor_state(void); @@ -85,4 +88,43 @@ struct saved_context; void __save_processor_state(struct saved_context *ctxt); void __restore_processor_state(struct saved_context *ctxt); +/* kernel/power/main.c */ +extern struct blocking_notifier_head pm_chain_head; + +static inline int register_pm_notifier(struct notifier_block *nb) +{ + return blocking_notifier_chain_register(&pm_chain_head, nb); +} + +static inline int unregister_pm_notifier(struct notifier_block *nb) +{ + return blocking_notifier_chain_unregister(&pm_chain_head, nb); +} + +#define pm_notifier(fn, pri) { \ + static struct notifier_block fn##_nb = \ + { .notifier_call = fn, .priority = pri }; \ + register_pm_notifier(&fn##_nb); \ +} +#else /* CONFIG_PM */ + +static inline int register_pm_notifier(struct notifier_block *nb) +{ + return 0; +} + +static inline int unregister_pm_notifier(struct notifier_block *nb) +{ + return 0; +} + +#define pm_notifier(fn, pri) do { (void)(fn); } while (0) +#endif /* CONFIG_PM */ + +#if !defined CONFIG_SOFTWARE_SUSPEND || !defined(CONFIG_PM) +static inline void register_nosave_region(unsigned long b, unsigned long e) +{ +} +#endif + #endif /* _LINUX_SWSUSP_H */ diff --git a/include/linux/user_namespace.h b/include/linux/user_namespace.h index bb320573bb9..1101b0ce878 100644 --- a/include/linux/user_namespace.h +++ b/include/linux/user_namespace.h @@ -49,7 +49,7 @@ static inline struct user_namespace *copy_user_ns(int flags, if (flags & CLONE_NEWUSER) return ERR_PTR(-EINVAL); - return NULL; + return old_ns; } static inline void put_user_ns(struct user_namespace *ns) diff --git a/include/net/genetlink.h b/include/net/genetlink.h index b6eaca122db..decdda54682 100644 --- a/include/net/genetlink.h +++ b/include/net/genetlink.h @@ -5,6 +5,22 @@ #include <net/netlink.h> /** + * struct genl_multicast_group - generic netlink multicast group + * @name: name of the multicast group, names are per-family + * @id: multicast group ID, assigned by the core, to use with + * genlmsg_multicast(). + * @list: list entry for linking + * @family: pointer to family, need not be set before registering + */ +struct genl_multicast_group +{ + struct genl_family *family; /* private */ + struct list_head list; /* private */ + char name[GENL_NAMSIZ]; + u32 id; +}; + +/** * struct genl_family - generic netlink family * @id: protocol family idenfitier * @hdrsize: length of user specific header in bytes @@ -14,6 +30,7 @@ * @attrbuf: buffer to store parsed attributes * @ops_list: list of all assigned operations * @family_list: family list + * @mcast_groups: multicast groups list */ struct genl_family { @@ -25,6 +42,7 @@ struct genl_family struct nlattr ** attrbuf; /* private */ struct list_head ops_list; /* private */ struct list_head family_list; /* private */ + struct list_head mcast_groups; /* private */ }; /** @@ -73,6 +91,10 @@ extern int genl_register_family(struct genl_family *family); extern int genl_unregister_family(struct genl_family *family); extern int genl_register_ops(struct genl_family *, struct genl_ops *ops); extern int genl_unregister_ops(struct genl_family *, struct genl_ops *ops); +extern int genl_register_mc_group(struct genl_family *family, + struct genl_multicast_group *grp); +extern void genl_unregister_mc_group(struct genl_family *family, + struct genl_multicast_group *grp); extern struct sock *genl_sock; diff --git a/ipc/shm.c b/ipc/shm.c index 242c3f66493..d0259e3ad1c 100644 --- a/ipc/shm.c +++ b/ipc/shm.c @@ -224,13 +224,12 @@ static void shm_close(struct vm_area_struct *vma) mutex_unlock(&shm_ids(ns).mutex); } -static struct page *shm_nopage(struct vm_area_struct *vma, - unsigned long address, int *type) +static int shm_fault(struct vm_area_struct *vma, struct vm_fault *vmf) { struct file *file = vma->vm_file; struct shm_file_data *sfd = shm_file_data(file); - return sfd->vm_ops->nopage(vma, address, type); + return sfd->vm_ops->fault(vma, vmf); } #ifdef CONFIG_NUMA @@ -269,6 +268,7 @@ static int shm_mmap(struct file * file, struct vm_area_struct * vma) if (ret != 0) return ret; sfd->vm_ops = vma->vm_ops; + BUG_ON(!sfd->vm_ops->fault); vma->vm_ops = &shm_vm_ops; shm_open(vma); @@ -327,7 +327,7 @@ static const struct file_operations shm_file_operations = { static struct vm_operations_struct shm_vm_ops = { .open = shm_open, /* callback for a new vm-area open */ .close = shm_close, /* callback for when the vm-area is released */ - .nopage = shm_nopage, + .fault = shm_fault, #if defined(CONFIG_NUMA) .set_policy = shm_set_policy, .get_policy = shm_get_policy, diff --git a/kernel/auditsc.c b/kernel/auditsc.c index b7640a5f382..145cbb79c4b 100644 --- a/kernel/auditsc.c +++ b/kernel/auditsc.c @@ -153,7 +153,7 @@ struct audit_aux_data_execve { struct audit_aux_data d; int argc; int envc; - char mem[0]; + struct mm_struct *mm; }; struct audit_aux_data_socketcall { @@ -831,6 +831,55 @@ static int audit_log_pid_context(struct audit_context *context, pid_t pid, return rc; } +static void audit_log_execve_info(struct audit_buffer *ab, + struct audit_aux_data_execve *axi) +{ + int i; + long len, ret; + const char __user *p = (const char __user *)axi->mm->arg_start; + char *buf; + + if (axi->mm != current->mm) + return; /* execve failed, no additional info */ + + for (i = 0; i < axi->argc; i++, p += len) { + len = strnlen_user(p, MAX_ARG_STRLEN); + /* + * We just created this mm, if we can't find the strings + * we just copied into it something is _very_ wrong. Similar + * for strings that are too long, we should not have created + * any. + */ + if (!len || len > MAX_ARG_STRLEN) { + WARN_ON(1); + send_sig(SIGKILL, current, 0); + } + + buf = kmalloc(len, GFP_KERNEL); + if (!buf) { + audit_panic("out of memory for argv string\n"); + break; + } + + ret = copy_from_user(buf, p, len); + /* + * There is no reason for this copy to be short. We just + * copied them here, and the mm hasn't been exposed to user- + * space yet. + */ + if (!ret) { + WARN_ON(1); + send_sig(SIGKILL, current, 0); + } + + audit_log_format(ab, "a%d=", i); + audit_log_untrustedstring(ab, buf); + audit_log_format(ab, "\n"); + + kfree(buf); + } +} + static void audit_log_exit(struct audit_context *context, struct task_struct *tsk) { int i, call_panic = 0; @@ -971,13 +1020,7 @@ static void audit_log_exit(struct audit_context *context, struct task_struct *ts case AUDIT_EXECVE: { struct audit_aux_data_execve *axi = (void *)aux; - int i; - const char *p; - for (i = 0, p = axi->mem; i < axi->argc; i++) { - audit_log_format(ab, "a%d=", i); - p = audit_log_untrustedstring(ab, p); - audit_log_format(ab, "\n"); - } + audit_log_execve_info(ab, axi); break; } case AUDIT_SOCKETCALL: { @@ -1821,32 +1864,31 @@ int __audit_ipc_set_perm(unsigned long qbytes, uid_t uid, gid_t gid, mode_t mode return 0; } +int audit_argv_kb = 32; + int audit_bprm(struct linux_binprm *bprm) { struct audit_aux_data_execve *ax; struct audit_context *context = current->audit_context; - unsigned long p, next; - void *to; if (likely(!audit_enabled || !context || context->dummy)) return 0; - ax = kmalloc(sizeof(*ax) + PAGE_SIZE * MAX_ARG_PAGES - bprm->p, - GFP_KERNEL); + /* + * Even though the stack code doesn't limit the arg+env size any more, + * the audit code requires that _all_ arguments be logged in a single + * netlink skb. Hence cap it :-( + */ + if (bprm->argv_len > (audit_argv_kb << 10)) + return -E2BIG; + + ax = kmalloc(sizeof(*ax), GFP_KERNEL); if (!ax) return -ENOMEM; ax->argc = bprm->argc; ax->envc = bprm->envc; - for (p = bprm->p, to = ax->mem; p < MAX_ARG_PAGES*PAGE_SIZE; p = next) { - struct page *page = bprm->page[p / PAGE_SIZE]; - void *kaddr = kmap(page); - next = (p + PAGE_SIZE) & ~(PAGE_SIZE - 1); - memcpy(to, kaddr + (p & (PAGE_SIZE - 1)), next - p); - to += next - p; - kunmap(page); - } - + ax->mm = bprm->mm; ax->d.type = AUDIT_EXECVE; ax->d.next = context->aux; context->aux = (void *)ax; diff --git a/kernel/exit.c b/kernel/exit.c index e8af8d0c248..464c2b172f0 100644 --- a/kernel/exit.c +++ b/kernel/exit.c @@ -45,6 +45,7 @@ #include <linux/resource.h> #include <linux/blkdev.h> #include <linux/task_io_accounting_ops.h> +#include <linux/freezer.h> #include <asm/uaccess.h> #include <asm/unistd.h> @@ -594,6 +595,8 @@ static void exit_mm(struct task_struct * tsk) tsk->mm = NULL; up_read(&mm->mmap_sem); enter_lazy_tlb(mm, current); + /* We don't want this task to be frozen prematurely */ + clear_freeze_flag(tsk); task_unlock(tsk); mmput(mm); } diff --git a/kernel/fork.c b/kernel/fork.c index ba39bdb2a7b..46983899822 100644 --- a/kernel/fork.c +++ b/kernel/fork.c @@ -334,6 +334,8 @@ static struct mm_struct * mm_init(struct mm_struct * mm) atomic_set(&mm->mm_count, 1); init_rwsem(&mm->mmap_sem); INIT_LIST_HEAD(&mm->mmlist); + mm->flags = (current->mm) ? current->mm->flags + : MMF_DUMP_FILTER_DEFAULT; mm->core_waiters = 0; mm->nr_ptes = 0; set_mm_counter(mm, file_rss, 0); diff --git a/kernel/futex.c b/kernel/futex.c index 5c3f45d07c5..a12425051ee 100644 --- a/kernel/futex.c +++ b/kernel/futex.c @@ -346,15 +346,20 @@ static int futex_handle_fault(unsigned long address, vma = find_vma(mm, address); if (vma && address >= vma->vm_start && (vma->vm_flags & VM_WRITE)) { - switch (handle_mm_fault(mm, vma, address, 1)) { - case VM_FAULT_MINOR: - ret = 0; - current->min_flt++; - break; - case VM_FAULT_MAJOR: + int fault; + fault = handle_mm_fault(mm, vma, address, 1); + if (unlikely((fault & VM_FAULT_ERROR))) { +#if 0 + /* XXX: let's do this when we verify it is OK */ + if (ret & VM_FAULT_OOM) + ret = -ENOMEM; +#endif + } else { ret = 0; - current->maj_flt++; - break; + if (fault & VM_FAULT_MAJOR) + current->maj_flt++; + else + current->min_flt++; } } if (!fshared) diff --git a/kernel/kmod.c b/kernel/kmod.c index 78d365c524e..beedbdc6460 100644 --- a/kernel/kmod.c +++ b/kernel/kmod.c @@ -33,6 +33,8 @@ #include <linux/kernel.h> #include <linux/init.h> #include <linux/resource.h> +#include <linux/notifier.h> +#include <linux/suspend.h> #include <asm/uaccess.h> extern int max_threads; @@ -265,6 +267,88 @@ static void __call_usermodehelper(struct work_struct *work) } } +#ifdef CONFIG_PM +/* + * If set, call_usermodehelper_exec() will exit immediately returning -EBUSY + * (used for preventing user land processes from being created after the user + * land has been frozen during a system-wide hibernation or suspend operation). + */ +static int usermodehelper_disabled; + +/* Number of helpers running */ +static atomic_t running_helpers = ATOMIC_INIT(0); + +/* + * Wait queue head used by usermodehelper_pm_callback() to wait for all running + * helpers to finish. + */ +static DECLARE_WAIT_QUEUE_HEAD(running_helpers_waitq); + +/* + * Time to wait for running_helpers to become zero before the setting of + * usermodehelper_disabled in usermodehelper_pm_callback() fails + */ +#define RUNNING_HELPERS_TIMEOUT (5 * HZ) + +static int usermodehelper_pm_callback(struct notifier_block *nfb, + unsigned long action, + void *ignored) +{ + long retval; + + switch (action) { + case PM_HIBERNATION_PREPARE: + case PM_SUSPEND_PREPARE: + usermodehelper_disabled = 1; + smp_mb(); + /* + * From now on call_usermodehelper_exec() won't start any new + * helpers, so it is sufficient if running_helpers turns out to + * be zero at one point (it may be increased later, but that + * doesn't matter). + */ + retval = wait_event_timeout(running_helpers_waitq, + atomic_read(&running_helpers) == 0, + RUNNING_HELPERS_TIMEOUT); + if (retval) { + return NOTIFY_OK; + } else { + usermodehelper_disabled = 0; + return NOTIFY_BAD; + } + case PM_POST_HIBERNATION: + case PM_POST_SUSPEND: + usermodehelper_disabled = 0; + return NOTIFY_OK; + } + + return NOTIFY_DONE; +} + +static void helper_lock(void) +{ + atomic_inc(&running_helpers); + smp_mb__after_atomic_inc(); +} + +static void helper_unlock(void) +{ + if (atomic_dec_and_test(&running_helpers)) + wake_up(&running_helpers_waitq); +} + +static void register_pm_notifier_callback(void) +{ + pm_notifier(usermodehelper_pm_callback, 0); +} +#else /* CONFIG_PM */ +#define usermodehelper_disabled 0 + +static inline void helper_lock(void) {} +static inline void helper_unlock(void) {} +static inline void register_pm_notifier_callback(void) {} +#endif /* CONFIG_PM */ + /** * call_usermodehelper_setup - prepare to call a usermode helper * @path - path to usermode executable @@ -369,12 +453,13 @@ int call_usermodehelper_exec(struct subprocess_info *sub_info, DECLARE_COMPLETION_ONSTACK(done); int retval; + helper_lock(); if (sub_info->path[0] == '\0') { retval = 0; goto out; } - if (!khelper_wq) { + if (!khelper_wq || usermodehelper_disabled) { retval = -EBUSY; goto out; } @@ -390,6 +475,7 @@ int call_usermodehelper_exec(struct subprocess_info *sub_info, out: call_usermodehelper_freeinfo(sub_info); + helper_unlock(); return retval; } EXPORT_SYMBOL(call_usermodehelper_exec); @@ -431,4 +517,5 @@ void __init usermodehelper_init(void) { khelper_wq = create_singlethread_workqueue("khelper"); BUG_ON(!khelper_wq); + register_pm_notifier_callback(); } diff --git a/kernel/kprobes.c b/kernel/kprobes.c index 9e47d8c493f..3e9f513a728 100644 --- a/kernel/kprobes.c +++ b/kernel/kprobes.c @@ -675,9 +675,18 @@ static struct notifier_block kprobe_exceptions_nb = { .priority = 0x7fffffff /* we need to be notified first */ }; +unsigned long __weak arch_deref_entry_point(void *entry) +{ + return (unsigned long)entry; +} int __kprobes register_jprobe(struct jprobe *jp) { + unsigned long addr = arch_deref_entry_point(jp->entry); + + if (!kernel_text_address(addr)) + return -EINVAL; + /* Todo: Verify probepoint is a function entry point */ jp->kp.pre_handler = setjmp_pre_handler; jp->kp.break_handler = longjmp_break_handler; diff --git a/kernel/ksysfs.c b/kernel/ksysfs.c index 559deca5ed1..2565e1b6dd7 100644 --- a/kernel/ksysfs.c +++ b/kernel/ksysfs.c @@ -62,6 +62,28 @@ static ssize_t kexec_crash_loaded_show(struct kset *kset, char *page) KERNEL_ATTR_RO(kexec_crash_loaded); #endif /* CONFIG_KEXEC */ +/* + * Make /sys/kernel/notes give the raw contents of our kernel .notes section. + */ +extern const char __start_notes __attribute__((weak)); +extern const char __stop_notes __attribute__((weak)); +#define notes_size (&__stop_notes - &__start_notes) + +static ssize_t notes_read(struct kobject *kobj, struct bin_attribute *bin_attr, + char *buf, loff_t off, size_t count) +{ + memcpy(buf, &__start_notes + off, count); + return count; +} + +static struct bin_attribute notes_attr = { + .attr = { + .name = "notes", + .mode = S_IRUGO, + }, + .read = ¬es_read, +}; + decl_subsys(kernel, NULL, NULL); EXPORT_SYMBOL_GPL(kernel_subsys); @@ -88,6 +110,12 @@ static int __init ksysfs_init(void) error = sysfs_create_group(&kernel_subsys.kobj, &kernel_attr_group); + if (!error && notes_size > 0) { + notes_attr.size = notes_size; + error = sysfs_create_bin_file(&kernel_subsys.kobj, + ¬es_attr); + } + return error; } diff --git a/kernel/lockdep.c b/kernel/lockdep.c index edba2ffb43d..734da579ad1 100644 --- a/kernel/lockdep.c +++ b/kernel/lockdep.c @@ -5,7 +5,8 @@ * * Started by Ingo Molnar: * - * Copyright (C) 2006 Red Hat, Inc., Ingo Molnar <mingo@redhat.com> + * Copyright (C) 2006,2007 Red Hat, Inc., Ingo Molnar <mingo@redhat.com> + * Copyright (C) 2007 Red Hat, Inc., Peter Zijlstra <pzijlstr@redhat.com> * * this code maps all the lock dependencies as they occur in a live kernel * and will warn about the following classes of locking bugs: @@ -37,11 +38,26 @@ #include <linux/debug_locks.h> #include <linux/irqflags.h> #include <linux/utsname.h> +#include <linux/hash.h> #include <asm/sections.h> #include "lockdep_internals.h" +#ifdef CONFIG_PROVE_LOCKING +int prove_locking = 1; +module_param(prove_locking, int, 0644); +#else +#define prove_locking 0 +#endif + +#ifdef CONFIG_LOCK_STAT +int lock_stat = 1; +module_param(lock_stat, int, 0644); +#else +#define lock_stat 0 +#endif + /* * lockdep_lock: protects the lockdep graph, the hashes and the * class/list/hash allocators. @@ -96,23 +112,6 @@ unsigned long nr_list_entries; static struct lock_list list_entries[MAX_LOCKDEP_ENTRIES]; /* - * Allocate a lockdep entry. (assumes the graph_lock held, returns - * with NULL on failure) - */ -static struct lock_list *alloc_list_entry(void) -{ - if (nr_list_entries >= MAX_LOCKDEP_ENTRIES) { - if (!debug_locks_off_graph_unlock()) - return NULL; - - printk("BUG: MAX_LOCKDEP_ENTRIES too low!\n"); - printk("turning off the locking correctness validator.\n"); - return NULL; - } - return list_entries + nr_list_entries++; -} - -/* * All data structures here are protected by the global debug_lock. * * Mutex key structs only get allocated, once during bootup, and never @@ -121,6 +120,117 @@ static struct lock_list *alloc_list_entry(void) unsigned long nr_lock_classes; static struct lock_class lock_classes[MAX_LOCKDEP_KEYS]; +#ifdef CONFIG_LOCK_STAT +static DEFINE_PER_CPU(struct lock_class_stats[MAX_LOCKDEP_KEYS], lock_stats); + +static int lock_contention_point(struct lock_class *class, unsigned long ip) +{ + int i; + + for (i = 0; i < ARRAY_SIZE(class->contention_point); i++) { + if (class->contention_point[i] == 0) { + class->contention_point[i] = ip; + break; + } + if (class->contention_point[i] == ip) + break; + } + + return i; +} + +static void lock_time_inc(struct lock_time *lt, s64 time) +{ + if (time > lt->max) + lt->max = time; + + if (time < lt->min || !lt->min) + lt->min = time; + + lt->total += time; + lt->nr++; +} + +static inline void lock_time_add(struct lock_time *src, struct lock_time *dst) +{ + dst->min += src->min; + dst->max += src->max; + dst->total += src->total; + dst->nr += src->nr; +} + +struct lock_class_stats lock_stats(struct lock_class *class) +{ + struct lock_class_stats stats; + int cpu, i; + + memset(&stats, 0, sizeof(struct lock_class_stats)); + for_each_possible_cpu(cpu) { + struct lock_class_stats *pcs = + &per_cpu(lock_stats, cpu)[class - lock_classes]; + + for (i = 0; i < ARRAY_SIZE(stats.contention_point); i++) + stats.contention_point[i] += pcs->contention_point[i]; + + lock_time_add(&pcs->read_waittime, &stats.read_waittime); + lock_time_add(&pcs->write_waittime, &stats.write_waittime); + + lock_time_add(&pcs->read_holdtime, &stats.read_holdtime); + lock_time_add(&pcs->write_holdtime, &stats.write_holdtime); + + for (i = 0; i < ARRAY_SIZE(stats.bounces); i++) + stats.bounces[i] += pcs->bounces[i]; + } + + return stats; +} + +void clear_lock_stats(struct lock_class *class) +{ + int cpu; + + for_each_possible_cpu(cpu) { + struct lock_class_stats *cpu_stats = + &per_cpu(lock_stats, cpu)[class - lock_classes]; + + memset(cpu_stats, 0, sizeof(struct lock_class_stats)); + } + memset(class->contention_point, 0, sizeof(class->contention_point)); +} + +static struct lock_class_stats *get_lock_stats(struct lock_class *class) +{ + return &get_cpu_var(lock_stats)[class - lock_classes]; +} + +static void put_lock_stats(struct lock_class_stats *stats) +{ + put_cpu_var(lock_stats); +} + +static void lock_release_holdtime(struct held_lock *hlock) +{ + struct lock_class_stats *stats; + s64 holdtime; + + if (!lock_stat) + return; + + holdtime = sched_clock() - hlock->holdtime_stamp; + + stats = get_lock_stats(hlock->class); + if (hlock->read) + lock_time_inc(&stats->read_holdtime, holdtime); + else + lock_time_inc(&stats->write_holdtime, holdtime); + put_lock_stats(stats); +} +#else +static inline void lock_release_holdtime(struct held_lock *hlock) +{ +} +#endif + /* * We keep a global list of all lock classes. The list only grows, * never shrinks. The list is only accessed with the lockdep @@ -133,24 +243,18 @@ LIST_HEAD(all_lock_classes); */ #define CLASSHASH_BITS (MAX_LOCKDEP_KEYS_BITS - 1) #define CLASSHASH_SIZE (1UL << CLASSHASH_BITS) -#define CLASSHASH_MASK (CLASSHASH_SIZE - 1) -#define __classhashfn(key) ((((unsigned long)key >> CLASSHASH_BITS) + (unsigned long)key) & CLASSHASH_MASK) +#define __classhashfn(key) hash_long((unsigned long)key, CLASSHASH_BITS) #define classhashentry(key) (classhash_table + __classhashfn((key))) static struct list_head classhash_table[CLASSHASH_SIZE]; -unsigned long nr_lock_chains; -static struct lock_chain lock_chains[MAX_LOCKDEP_CHAINS]; - /* * We put the lock dependency chains into a hash-table as well, to cache * their existence: */ #define CHAINHASH_BITS (MAX_LOCKDEP_CHAINS_BITS-1) #define CHAINHASH_SIZE (1UL << CHAINHASH_BITS) -#define CHAINHASH_MASK (CHAINHASH_SIZE - 1) -#define __chainhashfn(chain) \ - (((chain >> CHAINHASH_BITS) + chain) & CHAINHASH_MASK) +#define __chainhashfn(chain) hash_long(chain, CHAINHASH_BITS) #define chainhashentry(chain) (chainhash_table + __chainhashfn((chain))) static struct list_head chainhash_table[CHAINHASH_SIZE]; @@ -223,26 +327,6 @@ static int verbose(struct lock_class *class) return 0; } -#ifdef CONFIG_TRACE_IRQFLAGS - -static int hardirq_verbose(struct lock_class *class) -{ -#if HARDIRQ_VERBOSE - return class_filter(class); -#endif - return 0; -} - -static int softirq_verbose(struct lock_class *class) -{ -#if SOFTIRQ_VERBOSE - return class_filter(class); -#endif - return 0; -} - -#endif - /* * Stack-trace: tightly packed array of stack backtrace * addresses. Protected by the graph_lock. @@ -291,6 +375,11 @@ unsigned int max_recursion_depth; * about it later on, in lockdep_info(). */ static int lockdep_init_error; +static unsigned long lockdep_init_trace_data[20]; +static struct stack_trace lockdep_init_trace = { + .max_entries = ARRAY_SIZE(lockdep_init_trace_data), + .entries = lockdep_init_trace_data, +}; /* * Various lockdep statistics: @@ -482,6 +571,262 @@ static void print_lock_dependencies(struct lock_class *class, int depth) } } +static void print_kernel_version(void) +{ + printk("%s %.*s\n", init_utsname()->release, + (int)strcspn(init_utsname()->version, " "), + init_utsname()->version); +} + +static int very_verbose(struct lock_class *class) +{ +#if VERY_VERBOSE + return class_filter(class); +#endif + return 0; +} + +/* + * Is this the address of a static object: + */ +static int static_obj(void *obj) +{ + unsigned long start = (unsigned long) &_stext, + end = (unsigned long) &_end, + addr = (unsigned long) obj; +#ifdef CONFIG_SMP + int i; +#endif + + /* + * static variable? + */ + if ((addr >= start) && (addr < end)) + return 1; + +#ifdef CONFIG_SMP + /* + * percpu var? + */ + for_each_possible_cpu(i) { + start = (unsigned long) &__per_cpu_start + per_cpu_offset(i); + end = (unsigned long) &__per_cpu_start + PERCPU_ENOUGH_ROOM + + per_cpu_offset(i); + + if ((addr >= start) && (addr < end)) + return 1; + } +#endif + + /* + * module var? + */ + return is_module_address(addr); +} + +/* + * To make lock name printouts unique, we calculate a unique + * class->name_version generation counter: + */ +static int count_matching_names(struct lock_class *new_class) +{ + struct lock_class *class; + int count = 0; + + if (!new_class->name) + return 0; + + list_for_each_entry(class, &all_lock_classes, lock_entry) { + if (new_class->key - new_class->subclass == class->key) + return class->name_version; + if (class->name && !strcmp(class->name, new_class->name)) + count = max(count, class->name_version); + } + + return count + 1; +} + +/* + * Register a lock's class in the hash-table, if the class is not present + * yet. Otherwise we look it up. We cache the result in the lock object + * itself, so actual lookup of the hash should be once per lock object. + */ +static inline struct lock_class * +look_up_lock_class(struct lockdep_map *lock, unsigned int subclass) +{ + struct lockdep_subclass_key *key; + struct list_head *hash_head; + struct lock_class *class; + +#ifdef CONFIG_DEBUG_LOCKDEP + /* + * If the architecture calls into lockdep before initializing + * the hashes then we'll warn about it later. (we cannot printk + * right now) + */ + if (unlikely(!lockdep_initialized)) { + lockdep_init(); + lockdep_init_error = 1; + save_stack_trace(&lockdep_init_trace); + } +#endif + + /* + * Static locks do not have their class-keys yet - for them the key + * is the lock object itself: + */ + if (unlikely(!lock->key)) + lock->key = (void *)lock; + + /* + * NOTE: the class-key must be unique. For dynamic locks, a static + * lock_class_key variable is passed in through the mutex_init() + * (or spin_lock_init()) call - which acts as the key. For static + * locks we use the lock object itself as the key. + */ + BUILD_BUG_ON(sizeof(struct lock_class_key) > + sizeof(struct lockdep_map)); + + key = lock->key->subkeys + subclass; + + hash_head = classhashentry(key); + + /* + * We can walk the hash lockfree, because the hash only + * grows, and we are careful when adding entries to the end: + */ + list_for_each_entry(class, hash_head, hash_entry) { + if (class->key == key) { + WARN_ON_ONCE(class->name != lock->name); + return class; + } + } + + return NULL; +} + +/* + * Register a lock's class in the hash-table, if the class is not present + * yet. Otherwise we look it up. We cache the result in the lock object + * itself, so actual lookup of the hash should be once per lock object. + */ +static inline struct lock_class * +register_lock_class(struct lockdep_map *lock, unsigned int subclass, int force) +{ + struct lockdep_subclass_key *key; + struct list_head *hash_head; + struct lock_class *class; + unsigned long flags; + + class = look_up_lock_class(lock, subclass); + if (likely(class)) + return class; + + /* + * Debug-check: all keys must be persistent! + */ + if (!static_obj(lock->key)) { + debug_locks_off(); + printk("INFO: trying to register non-static key.\n"); + printk("the code is fine but needs lockdep annotation.\n"); + printk("turning off the locking correctness validator.\n"); + dump_stack(); + + return NULL; + } + + key = lock->key->subkeys + subclass; + hash_head = classhashentry(key); + + raw_local_irq_save(flags); + if (!graph_lock()) { + raw_local_irq_restore(flags); + return NULL; + } + /* + * We have to do the hash-walk again, to avoid races + * with another CPU: + */ + list_for_each_entry(class, hash_head, hash_entry) + if (class->key == key) + goto out_unlock_set; + /* + * Allocate a new key from the static array, and add it to + * the hash: + */ + if (nr_lock_classes >= MAX_LOCKDEP_KEYS) { + if (!debug_locks_off_graph_unlock()) { + raw_local_irq_restore(flags); + return NULL; + } + raw_local_irq_restore(flags); + + printk("BUG: MAX_LOCKDEP_KEYS too low!\n"); + printk("turning off the locking correctness validator.\n"); + return NULL; + } + class = lock_classes + nr_lock_classes++; + debug_atomic_inc(&nr_unused_locks); + class->key = key; + class->name = lock->name; + class->subclass = subclass; + INIT_LIST_HEAD(&class->lock_entry); + INIT_LIST_HEAD(&class->locks_before); + INIT_LIST_HEAD(&class->locks_after); + class->name_version = count_matching_names(class); + /* + * We use RCU's safe list-add method to make + * parallel walking of the hash-list safe: + */ + list_add_tail_rcu(&class->hash_entry, hash_head); + + if (verbose(class)) { + graph_unlock(); + raw_local_irq_restore(flags); + + printk("\nnew class %p: %s", class->key, class->name); + if (class->name_version > 1) + printk("#%d", class->name_version); + printk("\n"); + dump_stack(); + + raw_local_irq_save(flags); + if (!graph_lock()) { + raw_local_irq_restore(flags); + return NULL; + } + } +out_unlock_set: + graph_unlock(); + raw_local_irq_restore(flags); + + if (!subclass || force) + lock->class_cache = class; + + if (DEBUG_LOCKS_WARN_ON(class->subclass != subclass)) + return NULL; + + return class; +} + +#ifdef CONFIG_PROVE_LOCKING +/* + * Allocate a lockdep entry. (assumes the graph_lock held, returns + * with NULL on failure) + */ +static struct lock_list *alloc_list_entry(void) +{ + if (nr_list_entries >= MAX_LOCKDEP_ENTRIES) { + if (!debug_locks_off_graph_unlock()) + return NULL; + + printk("BUG: MAX_LOCKDEP_ENTRIES too low!\n"); + printk("turning off the locking correctness validator.\n"); + return NULL; + } + return list_entries + nr_list_entries++; +} + /* * Add a new dependency to the head of the list: */ @@ -542,13 +887,6 @@ print_circular_bug_entry(struct lock_list *target, unsigned int depth) return 0; } -static void print_kernel_version(void) -{ - printk("%s %.*s\n", init_utsname()->release, - (int)strcspn(init_utsname()->version, " "), - init_utsname()->version); -} - /* * When a circular dependency is detected, print the * header first: @@ -640,15 +978,7 @@ check_noncircular(struct lock_class *source, unsigned int depth) return 1; } -static int very_verbose(struct lock_class *class) -{ -#if VERY_VERBOSE - return class_filter(class); -#endif - return 0; -} #ifdef CONFIG_TRACE_IRQFLAGS - /* * Forwards and backwards subgraph searching, for the purposes of * proving that two subgraphs can be connected by a new dependency @@ -821,6 +1151,78 @@ check_usage(struct task_struct *curr, struct held_lock *prev, bit_backwards, bit_forwards, irqclass); } +static int +check_prev_add_irq(struct task_struct *curr, struct held_lock *prev, + struct held_lock *next) +{ + /* + * Prove that the new dependency does not connect a hardirq-safe + * lock with a hardirq-unsafe lock - to achieve this we search + * the backwards-subgraph starting at <prev>, and the + * forwards-subgraph starting at <next>: + */ + if (!check_usage(curr, prev, next, LOCK_USED_IN_HARDIRQ, + LOCK_ENABLED_HARDIRQS, "hard")) + return 0; + + /* + * Prove that the new dependency does not connect a hardirq-safe-read + * lock with a hardirq-unsafe lock - to achieve this we search + * the backwards-subgraph starting at <prev>, and the + * forwards-subgraph starting at <next>: + */ + if (!check_usage(curr, prev, next, LOCK_USED_IN_HARDIRQ_READ, + LOCK_ENABLED_HARDIRQS, "hard-read")) + return 0; + + /* + * Prove that the new dependency does not connect a softirq-safe + * lock with a softirq-unsafe lock - to achieve this we search + * the backwards-subgraph starting at <prev>, and the + * forwards-subgraph starting at <next>: + */ + if (!check_usage(curr, prev, next, LOCK_USED_IN_SOFTIRQ, + LOCK_ENABLED_SOFTIRQS, "soft")) + return 0; + /* + * Prove that the new dependency does not connect a softirq-safe-read + * lock with a softirq-unsafe lock - to achieve this we search + * the backwards-subgraph starting at <prev>, and the + * forwards-subgraph starting at <next>: + */ + if (!check_usage(curr, prev, next, LOCK_USED_IN_SOFTIRQ_READ, + LOCK_ENABLED_SOFTIRQS, "soft")) + return 0; + + return 1; +} + +static void inc_chains(void) +{ + if (current->hardirq_context) + nr_hardirq_chains++; + else { + if (current->softirq_context) + nr_softirq_chains++; + else + nr_process_chains++; + } +} + +#else + +static inline int +check_prev_add_irq(struct task_struct *curr, struct held_lock *prev, + struct held_lock *next) +{ + return 1; +} + +static inline void inc_chains(void) +{ + nr_process_chains++; +} + #endif static int @@ -922,47 +1324,10 @@ check_prev_add(struct task_struct *curr, struct held_lock *prev, if (!(check_noncircular(next->class, 0))) return print_circular_bug_tail(); -#ifdef CONFIG_TRACE_IRQFLAGS - /* - * Prove that the new dependency does not connect a hardirq-safe - * lock with a hardirq-unsafe lock - to achieve this we search - * the backwards-subgraph starting at <prev>, and the - * forwards-subgraph starting at <next>: - */ - if (!check_usage(curr, prev, next, LOCK_USED_IN_HARDIRQ, - LOCK_ENABLED_HARDIRQS, "hard")) + if (!check_prev_add_irq(curr, prev, next)) return 0; /* - * Prove that the new dependency does not connect a hardirq-safe-read - * lock with a hardirq-unsafe lock - to achieve this we search - * the backwards-subgraph starting at <prev>, and the - * forwards-subgraph starting at <next>: - */ - if (!check_usage(curr, prev, next, LOCK_USED_IN_HARDIRQ_READ, - LOCK_ENABLED_HARDIRQS, "hard-read")) - return 0; - - /* - * Prove that the new dependency does not connect a softirq-safe - * lock with a softirq-unsafe lock - to achieve this we search - * the backwards-subgraph starting at <prev>, and the - * forwards-subgraph starting at <next>: - */ - if (!check_usage(curr, prev, next, LOCK_USED_IN_SOFTIRQ, - LOCK_ENABLED_SOFTIRQS, "soft")) - return 0; - /* - * Prove that the new dependency does not connect a softirq-safe-read - * lock with a softirq-unsafe lock - to achieve this we search - * the backwards-subgraph starting at <prev>, and the - * forwards-subgraph starting at <next>: - */ - if (!check_usage(curr, prev, next, LOCK_USED_IN_SOFTIRQ_READ, - LOCK_ENABLED_SOFTIRQS, "soft")) - return 0; -#endif - /* * For recursive read-locks we do all the dependency checks, * but we dont store read-triggered dependencies (only * write-triggered dependencies). This ensures that only the @@ -1088,224 +1453,8 @@ out_bug: return 0; } - -/* - * Is this the address of a static object: - */ -static int static_obj(void *obj) -{ - unsigned long start = (unsigned long) &_stext, - end = (unsigned long) &_end, - addr = (unsigned long) obj; -#ifdef CONFIG_SMP - int i; -#endif - - /* - * static variable? - */ - if ((addr >= start) && (addr < end)) - return 1; - -#ifdef CONFIG_SMP - /* - * percpu var? - */ - for_each_possible_cpu(i) { - start = (unsigned long) &__per_cpu_start + per_cpu_offset(i); - end = (unsigned long) &__per_cpu_start + PERCPU_ENOUGH_ROOM - + per_cpu_offset(i); - - if ((addr >= start) && (addr < end)) - return 1; - } -#endif - - /* - * module var? - */ - return is_module_address(addr); -} - -/* - * To make lock name printouts unique, we calculate a unique - * class->name_version generation counter: - */ -static int count_matching_names(struct lock_class *new_class) -{ - struct lock_class *class; - int count = 0; - - if (!new_class->name) - return 0; - - list_for_each_entry(class, &all_lock_classes, lock_entry) { - if (new_class->key - new_class->subclass == class->key) - return class->name_version; - if (class->name && !strcmp(class->name, new_class->name)) - count = max(count, class->name_version); - } - - return count + 1; -} - -/* - * Register a lock's class in the hash-table, if the class is not present - * yet. Otherwise we look it up. We cache the result in the lock object - * itself, so actual lookup of the hash should be once per lock object. - */ -static inline struct lock_class * -look_up_lock_class(struct lockdep_map *lock, unsigned int subclass) -{ - struct lockdep_subclass_key *key; - struct list_head *hash_head; - struct lock_class *class; - -#ifdef CONFIG_DEBUG_LOCKDEP - /* - * If the architecture calls into lockdep before initializing - * the hashes then we'll warn about it later. (we cannot printk - * right now) - */ - if (unlikely(!lockdep_initialized)) { - lockdep_init(); - lockdep_init_error = 1; - } -#endif - - /* - * Static locks do not have their class-keys yet - for them the key - * is the lock object itself: - */ - if (unlikely(!lock->key)) - lock->key = (void *)lock; - - /* - * NOTE: the class-key must be unique. For dynamic locks, a static - * lock_class_key variable is passed in through the mutex_init() - * (or spin_lock_init()) call - which acts as the key. For static - * locks we use the lock object itself as the key. - */ - BUILD_BUG_ON(sizeof(struct lock_class_key) > sizeof(struct lock_class)); - - key = lock->key->subkeys + subclass; - - hash_head = classhashentry(key); - - /* - * We can walk the hash lockfree, because the hash only - * grows, and we are careful when adding entries to the end: - */ - list_for_each_entry(class, hash_head, hash_entry) - if (class->key == key) - return class; - - return NULL; -} - -/* - * Register a lock's class in the hash-table, if the class is not present - * yet. Otherwise we look it up. We cache the result in the lock object - * itself, so actual lookup of the hash should be once per lock object. - */ -static inline struct lock_class * -register_lock_class(struct lockdep_map *lock, unsigned int subclass, int force) -{ - struct lockdep_subclass_key *key; - struct list_head *hash_head; - struct lock_class *class; - unsigned long flags; - - class = look_up_lock_class(lock, subclass); - if (likely(class)) - return class; - - /* - * Debug-check: all keys must be persistent! - */ - if (!static_obj(lock->key)) { - debug_locks_off(); - printk("INFO: trying to register non-static key.\n"); - printk("the code is fine but needs lockdep annotation.\n"); - printk("turning off the locking correctness validator.\n"); - dump_stack(); - - return NULL; - } - - key = lock->key->subkeys + subclass; - hash_head = classhashentry(key); - - raw_local_irq_save(flags); - if (!graph_lock()) { - raw_local_irq_restore(flags); - return NULL; - } - /* - * We have to do the hash-walk again, to avoid races - * with another CPU: - */ - list_for_each_entry(class, hash_head, hash_entry) - if (class->key == key) - goto out_unlock_set; - /* - * Allocate a new key from the static array, and add it to - * the hash: - */ - if (nr_lock_classes >= MAX_LOCKDEP_KEYS) { - if (!debug_locks_off_graph_unlock()) { - raw_local_irq_restore(flags); - return NULL; - } - raw_local_irq_restore(flags); - - printk("BUG: MAX_LOCKDEP_KEYS too low!\n"); - printk("turning off the locking correctness validator.\n"); - return NULL; - } - class = lock_classes + nr_lock_classes++; - debug_atomic_inc(&nr_unused_locks); - class->key = key; - class->name = lock->name; - class->subclass = subclass; - INIT_LIST_HEAD(&class->lock_entry); - INIT_LIST_HEAD(&class->locks_before); - INIT_LIST_HEAD(&class->locks_after); - class->name_version = count_matching_names(class); - /* - * We use RCU's safe list-add method to make - * parallel walking of the hash-list safe: - */ - list_add_tail_rcu(&class->hash_entry, hash_head); - - if (verbose(class)) { - graph_unlock(); - raw_local_irq_restore(flags); - - printk("\nnew class %p: %s", class->key, class->name); - if (class->name_version > 1) - printk("#%d", class->name_version); - printk("\n"); - dump_stack(); - - raw_local_irq_save(flags); - if (!graph_lock()) { - raw_local_irq_restore(flags); - return NULL; - } - } -out_unlock_set: - graph_unlock(); - raw_local_irq_restore(flags); - - if (!subclass || force) - lock->class_cache = class; - - if (DEBUG_LOCKS_WARN_ON(class->subclass != subclass)) - return NULL; - - return class; -} +unsigned long nr_lock_chains; +static struct lock_chain lock_chains[MAX_LOCKDEP_CHAINS]; /* * Look up a dependency chain. If the key is not present yet then @@ -1366,21 +1515,72 @@ cache_hit: chain->chain_key = chain_key; list_add_tail_rcu(&chain->entry, hash_head); debug_atomic_inc(&chain_lookup_misses); -#ifdef CONFIG_TRACE_IRQFLAGS - if (current->hardirq_context) - nr_hardirq_chains++; - else { - if (current->softirq_context) - nr_softirq_chains++; - else - nr_process_chains++; - } -#else - nr_process_chains++; -#endif + inc_chains(); + + return 1; +} + +static int validate_chain(struct task_struct *curr, struct lockdep_map *lock, + struct held_lock *hlock, int chain_head) +{ + /* + * Trylock needs to maintain the stack of held locks, but it + * does not add new dependencies, because trylock can be done + * in any order. + * + * We look up the chain_key and do the O(N^2) check and update of + * the dependencies only if this is a new dependency chain. + * (If lookup_chain_cache() returns with 1 it acquires + * graph_lock for us) + */ + if (!hlock->trylock && (hlock->check == 2) && + lookup_chain_cache(curr->curr_chain_key, hlock->class)) { + /* + * Check whether last held lock: + * + * - is irq-safe, if this lock is irq-unsafe + * - is softirq-safe, if this lock is hardirq-unsafe + * + * And check whether the new lock's dependency graph + * could lead back to the previous lock. + * + * any of these scenarios could lead to a deadlock. If + * All validations + */ + int ret = check_deadlock(curr, hlock, lock, hlock->read); + + if (!ret) + return 0; + /* + * Mark recursive read, as we jump over it when + * building dependencies (just like we jump over + * trylock entries): + */ + if (ret == 2) + hlock->read = 2; + /* + * Add dependency only if this lock is not the head + * of the chain, and if it's not a secondary read-lock: + */ + if (!chain_head && ret != 2) + if (!check_prevs_add(curr, hlock)) + return 0; + graph_unlock(); + } else + /* after lookup_chain_cache(): */ + if (unlikely(!debug_locks)) + return 0; return 1; } +#else +static inline int validate_chain(struct task_struct *curr, + struct lockdep_map *lock, struct held_lock *hlock, + int chain_head) +{ + return 1; +} +#endif /* * We are building curr_chain_key incrementally, so double-check @@ -1425,6 +1625,57 @@ static void check_chain_key(struct task_struct *curr) #endif } +static int +print_usage_bug(struct task_struct *curr, struct held_lock *this, + enum lock_usage_bit prev_bit, enum lock_usage_bit new_bit) +{ + if (!debug_locks_off_graph_unlock() || debug_locks_silent) + return 0; + + printk("\n=================================\n"); + printk( "[ INFO: inconsistent lock state ]\n"); + print_kernel_version(); + printk( "---------------------------------\n"); + + printk("inconsistent {%s} -> {%s} usage.\n", + usage_str[prev_bit], usage_str[new_bit]); + + printk("%s/%d [HC%u[%lu]:SC%u[%lu]:HE%u:SE%u] takes:\n", + curr->comm, curr->pid, + trace_hardirq_context(curr), hardirq_count() >> HARDIRQ_SHIFT, + trace_softirq_context(curr), softirq_count() >> SOFTIRQ_SHIFT, + trace_hardirqs_enabled(curr), + trace_softirqs_enabled(curr)); + print_lock(this); + + printk("{%s} state was registered at:\n", usage_str[prev_bit]); + print_stack_trace(this->class->usage_traces + prev_bit, 1); + + print_irqtrace_events(curr); + printk("\nother info that might help us debug this:\n"); + lockdep_print_held_locks(curr); + + printk("\nstack backtrace:\n"); + dump_stack(); + + return 0; +} + +/* + * Print out an error if an invalid bit is set: + */ +static inline int +valid_state(struct task_struct *curr, struct held_lock *this, + enum lock_usage_bit new_bit, enum lock_usage_bit bad_bit) +{ + if (unlikely(this->class->usage_mask & (1 << bad_bit))) + return print_usage_bug(curr, this, bad_bit, new_bit); + return 1; +} + +static int mark_lock(struct task_struct *curr, struct held_lock *this, + enum lock_usage_bit new_bit); + #ifdef CONFIG_TRACE_IRQFLAGS /* @@ -1518,90 +1769,30 @@ void print_irqtrace_events(struct task_struct *curr) print_ip_sym(curr->softirq_disable_ip); } -#endif - -static int -print_usage_bug(struct task_struct *curr, struct held_lock *this, - enum lock_usage_bit prev_bit, enum lock_usage_bit new_bit) +static int hardirq_verbose(struct lock_class *class) { - if (!debug_locks_off_graph_unlock() || debug_locks_silent) - return 0; - - printk("\n=================================\n"); - printk( "[ INFO: inconsistent lock state ]\n"); - print_kernel_version(); - printk( "---------------------------------\n"); - - printk("inconsistent {%s} -> {%s} usage.\n", - usage_str[prev_bit], usage_str[new_bit]); - - printk("%s/%d [HC%u[%lu]:SC%u[%lu]:HE%u:SE%u] takes:\n", - curr->comm, curr->pid, - trace_hardirq_context(curr), hardirq_count() >> HARDIRQ_SHIFT, - trace_softirq_context(curr), softirq_count() >> SOFTIRQ_SHIFT, - trace_hardirqs_enabled(curr), - trace_softirqs_enabled(curr)); - print_lock(this); - - printk("{%s} state was registered at:\n", usage_str[prev_bit]); - print_stack_trace(this->class->usage_traces + prev_bit, 1); - - print_irqtrace_events(curr); - printk("\nother info that might help us debug this:\n"); - lockdep_print_held_locks(curr); - - printk("\nstack backtrace:\n"); - dump_stack(); - +#if HARDIRQ_VERBOSE + return class_filter(class); +#endif return 0; } -/* - * Print out an error if an invalid bit is set: - */ -static inline int -valid_state(struct task_struct *curr, struct held_lock *this, - enum lock_usage_bit new_bit, enum lock_usage_bit bad_bit) +static int softirq_verbose(struct lock_class *class) { - if (unlikely(this->class->usage_mask & (1 << bad_bit))) - return print_usage_bug(curr, this, bad_bit, new_bit); - return 1; +#if SOFTIRQ_VERBOSE + return class_filter(class); +#endif + return 0; } #define STRICT_READ_CHECKS 1 -/* - * Mark a lock with a usage bit, and validate the state transition: - */ -static int mark_lock(struct task_struct *curr, struct held_lock *this, - enum lock_usage_bit new_bit) +static int mark_lock_irq(struct task_struct *curr, struct held_lock *this, + enum lock_usage_bit new_bit) { - unsigned int new_mask = 1 << new_bit, ret = 1; - - /* - * If already set then do not dirty the cacheline, - * nor do any checks: - */ - if (likely(this->class->usage_mask & new_mask)) - return 1; - - if (!graph_lock()) - return 0; - /* - * Make sure we didnt race: - */ - if (unlikely(this->class->usage_mask & new_mask)) { - graph_unlock(); - return 1; - } - - this->class->usage_mask |= new_mask; + int ret = 1; - if (!save_trace(this->class->usage_traces + new_bit)) - return 0; - - switch (new_bit) { -#ifdef CONFIG_TRACE_IRQFLAGS + switch(new_bit) { case LOCK_USED_IN_HARDIRQ: if (!valid_state(curr, this, new_bit, LOCK_ENABLED_HARDIRQS)) return 0; @@ -1760,37 +1951,14 @@ static int mark_lock(struct task_struct *curr, struct held_lock *this, if (softirq_verbose(this->class)) ret = 2; break; -#endif - case LOCK_USED: - /* - * Add it to the global list of classes: - */ - list_add_tail_rcu(&this->class->lock_entry, &all_lock_classes); - debug_atomic_dec(&nr_unused_locks); - break; default: - if (!debug_locks_off_graph_unlock()) - return 0; WARN_ON(1); - return 0; - } - - graph_unlock(); - - /* - * We must printk outside of the graph_lock: - */ - if (ret == 2) { - printk("\nmarked lock as {%s}:\n", usage_str[new_bit]); - print_lock(this); - print_irqtrace_events(curr); - dump_stack(); + break; } return ret; } -#ifdef CONFIG_TRACE_IRQFLAGS /* * Mark all held locks with a usage bit: */ @@ -1973,9 +2141,176 @@ void trace_softirqs_off(unsigned long ip) debug_atomic_inc(&redundant_softirqs_off); } +static int mark_irqflags(struct task_struct *curr, struct held_lock *hlock) +{ + /* + * If non-trylock use in a hardirq or softirq context, then + * mark the lock as used in these contexts: + */ + if (!hlock->trylock) { + if (hlock->read) { + if (curr->hardirq_context) + if (!mark_lock(curr, hlock, + LOCK_USED_IN_HARDIRQ_READ)) + return 0; + if (curr->softirq_context) + if (!mark_lock(curr, hlock, + LOCK_USED_IN_SOFTIRQ_READ)) + return 0; + } else { + if (curr->hardirq_context) + if (!mark_lock(curr, hlock, LOCK_USED_IN_HARDIRQ)) + return 0; + if (curr->softirq_context) + if (!mark_lock(curr, hlock, LOCK_USED_IN_SOFTIRQ)) + return 0; + } + } + if (!hlock->hardirqs_off) { + if (hlock->read) { + if (!mark_lock(curr, hlock, + LOCK_ENABLED_HARDIRQS_READ)) + return 0; + if (curr->softirqs_enabled) + if (!mark_lock(curr, hlock, + LOCK_ENABLED_SOFTIRQS_READ)) + return 0; + } else { + if (!mark_lock(curr, hlock, + LOCK_ENABLED_HARDIRQS)) + return 0; + if (curr->softirqs_enabled) + if (!mark_lock(curr, hlock, + LOCK_ENABLED_SOFTIRQS)) + return 0; + } + } + + return 1; +} + +static int separate_irq_context(struct task_struct *curr, + struct held_lock *hlock) +{ + unsigned int depth = curr->lockdep_depth; + + /* + * Keep track of points where we cross into an interrupt context: + */ + hlock->irq_context = 2*(curr->hardirq_context ? 1 : 0) + + curr->softirq_context; + if (depth) { + struct held_lock *prev_hlock; + + prev_hlock = curr->held_locks + depth-1; + /* + * If we cross into another context, reset the + * hash key (this also prevents the checking and the + * adding of the dependency to 'prev'): + */ + if (prev_hlock->irq_context != hlock->irq_context) + return 1; + } + return 0; +} + +#else + +static inline +int mark_lock_irq(struct task_struct *curr, struct held_lock *this, + enum lock_usage_bit new_bit) +{ + WARN_ON(1); + return 1; +} + +static inline int mark_irqflags(struct task_struct *curr, + struct held_lock *hlock) +{ + return 1; +} + +static inline int separate_irq_context(struct task_struct *curr, + struct held_lock *hlock) +{ + return 0; +} + #endif /* + * Mark a lock with a usage bit, and validate the state transition: + */ +static int mark_lock(struct task_struct *curr, struct held_lock *this, + enum lock_usage_bit new_bit) +{ + unsigned int new_mask = 1 << new_bit, ret = 1; + + /* + * If already set then do not dirty the cacheline, + * nor do any checks: + */ + if (likely(this->class->usage_mask & new_mask)) + return 1; + + if (!graph_lock()) + return 0; + /* + * Make sure we didnt race: + */ + if (unlikely(this->class->usage_mask & new_mask)) { + graph_unlock(); + return 1; + } + + this->class->usage_mask |= new_mask; + + if (!save_trace(this->class->usage_traces + new_bit)) + return 0; + + switch (new_bit) { + case LOCK_USED_IN_HARDIRQ: + case LOCK_USED_IN_SOFTIRQ: + case LOCK_USED_IN_HARDIRQ_READ: + case LOCK_USED_IN_SOFTIRQ_READ: + case LOCK_ENABLED_HARDIRQS: + case LOCK_ENABLED_SOFTIRQS: + case LOCK_ENABLED_HARDIRQS_READ: + case LOCK_ENABLED_SOFTIRQS_READ: + ret = mark_lock_irq(curr, this, new_bit); + if (!ret) + return 0; + break; + case LOCK_USED: + /* + * Add it to the global list of classes: + */ + list_add_tail_rcu(&this->class->lock_entry, &all_lock_classes); + debug_atomic_dec(&nr_unused_locks); + break; + default: + if (!debug_locks_off_graph_unlock()) + return 0; + WARN_ON(1); + return 0; + } + + graph_unlock(); + + /* + * We must printk outside of the graph_lock: + */ + if (ret == 2) { + printk("\nmarked lock as {%s}:\n", usage_str[new_bit]); + print_lock(this); + print_irqtrace_events(curr); + dump_stack(); + } + + return ret; +} + +/* * Initialize a lock instance's lock-class mapping info: */ void lockdep_init_map(struct lockdep_map *lock, const char *name, @@ -1999,6 +2334,9 @@ void lockdep_init_map(struct lockdep_map *lock, const char *name, lock->name = name; lock->key = key; lock->class_cache = NULL; +#ifdef CONFIG_LOCK_STAT + lock->cpu = raw_smp_processor_id(); +#endif if (subclass) register_lock_class(lock, subclass, 1); } @@ -2020,6 +2358,9 @@ static int __lock_acquire(struct lockdep_map *lock, unsigned int subclass, int chain_head = 0; u64 chain_key; + if (!prove_locking) + check = 1; + if (unlikely(!debug_locks)) return 0; @@ -2070,57 +2411,18 @@ static int __lock_acquire(struct lockdep_map *lock, unsigned int subclass, hlock->read = read; hlock->check = check; hlock->hardirqs_off = hardirqs_off; - - if (check != 2) - goto out_calc_hash; -#ifdef CONFIG_TRACE_IRQFLAGS - /* - * If non-trylock use in a hardirq or softirq context, then - * mark the lock as used in these contexts: - */ - if (!trylock) { - if (read) { - if (curr->hardirq_context) - if (!mark_lock(curr, hlock, - LOCK_USED_IN_HARDIRQ_READ)) - return 0; - if (curr->softirq_context) - if (!mark_lock(curr, hlock, - LOCK_USED_IN_SOFTIRQ_READ)) - return 0; - } else { - if (curr->hardirq_context) - if (!mark_lock(curr, hlock, LOCK_USED_IN_HARDIRQ)) - return 0; - if (curr->softirq_context) - if (!mark_lock(curr, hlock, LOCK_USED_IN_SOFTIRQ)) - return 0; - } - } - if (!hardirqs_off) { - if (read) { - if (!mark_lock(curr, hlock, - LOCK_ENABLED_HARDIRQS_READ)) - return 0; - if (curr->softirqs_enabled) - if (!mark_lock(curr, hlock, - LOCK_ENABLED_SOFTIRQS_READ)) - return 0; - } else { - if (!mark_lock(curr, hlock, - LOCK_ENABLED_HARDIRQS)) - return 0; - if (curr->softirqs_enabled) - if (!mark_lock(curr, hlock, - LOCK_ENABLED_SOFTIRQS)) - return 0; - } - } +#ifdef CONFIG_LOCK_STAT + hlock->waittime_stamp = 0; + hlock->holdtime_stamp = sched_clock(); #endif + + if (check == 2 && !mark_irqflags(curr, hlock)) + return 0; + /* mark it as used: */ if (!mark_lock(curr, hlock, LOCK_USED)) return 0; -out_calc_hash: + /* * Calculate the chain hash: it's the combined has of all the * lock keys along the dependency chain. We save the hash value @@ -2143,77 +2445,15 @@ out_calc_hash: } hlock->prev_chain_key = chain_key; - -#ifdef CONFIG_TRACE_IRQFLAGS - /* - * Keep track of points where we cross into an interrupt context: - */ - hlock->irq_context = 2*(curr->hardirq_context ? 1 : 0) + - curr->softirq_context; - if (depth) { - struct held_lock *prev_hlock; - - prev_hlock = curr->held_locks + depth-1; - /* - * If we cross into another context, reset the - * hash key (this also prevents the checking and the - * adding of the dependency to 'prev'): - */ - if (prev_hlock->irq_context != hlock->irq_context) { - chain_key = 0; - chain_head = 1; - } + if (separate_irq_context(curr, hlock)) { + chain_key = 0; + chain_head = 1; } -#endif chain_key = iterate_chain_key(chain_key, id); curr->curr_chain_key = chain_key; - /* - * Trylock needs to maintain the stack of held locks, but it - * does not add new dependencies, because trylock can be done - * in any order. - * - * We look up the chain_key and do the O(N^2) check and update of - * the dependencies only if this is a new dependency chain. - * (If lookup_chain_cache() returns with 1 it acquires - * graph_lock for us) - */ - if (!trylock && (check == 2) && lookup_chain_cache(chain_key, class)) { - /* - * Check whether last held lock: - * - * - is irq-safe, if this lock is irq-unsafe - * - is softirq-safe, if this lock is hardirq-unsafe - * - * And check whether the new lock's dependency graph - * could lead back to the previous lock. - * - * any of these scenarios could lead to a deadlock. If - * All validations - */ - int ret = check_deadlock(curr, hlock, lock, read); - - if (!ret) - return 0; - /* - * Mark recursive read, as we jump over it when - * building dependencies (just like we jump over - * trylock entries): - */ - if (ret == 2) - hlock->read = 2; - /* - * Add dependency only if this lock is not the head - * of the chain, and if it's not a secondary read-lock: - */ - if (!chain_head && ret != 2) - if (!check_prevs_add(curr, hlock)) - return 0; - graph_unlock(); - } else - /* after lookup_chain_cache(): */ - if (unlikely(!debug_locks)) - return 0; + if (!validate_chain(curr, lock, hlock, chain_head)) + return 0; curr->lockdep_depth++; check_chain_key(curr); @@ -2315,6 +2555,8 @@ lock_release_non_nested(struct task_struct *curr, return print_unlock_inbalance_bug(curr, lock, ip); found_it: + lock_release_holdtime(hlock); + /* * We have the right lock to unlock, 'hlock' points to it. * Now we remove it from the stack, and add back the other @@ -2367,6 +2609,8 @@ static int lock_release_nested(struct task_struct *curr, curr->curr_chain_key = hlock->prev_chain_key; + lock_release_holdtime(hlock); + #ifdef CONFIG_DEBUG_LOCKDEP hlock->prev_chain_key = 0; hlock->class = NULL; @@ -2441,6 +2685,9 @@ void lock_acquire(struct lockdep_map *lock, unsigned int subclass, { unsigned long flags; + if (unlikely(!lock_stat && !prove_locking)) + return; + if (unlikely(current->lockdep_recursion)) return; @@ -2460,6 +2707,9 @@ void lock_release(struct lockdep_map *lock, int nested, unsigned long ip) { unsigned long flags; + if (unlikely(!lock_stat && !prove_locking)) + return; + if (unlikely(current->lockdep_recursion)) return; @@ -2473,6 +2723,166 @@ void lock_release(struct lockdep_map *lock, int nested, unsigned long ip) EXPORT_SYMBOL_GPL(lock_release); +#ifdef CONFIG_LOCK_STAT +static int +print_lock_contention_bug(struct task_struct *curr, struct lockdep_map *lock, + unsigned long ip) +{ + if (!debug_locks_off()) + return 0; + if (debug_locks_silent) + return 0; + + printk("\n=================================\n"); + printk( "[ BUG: bad contention detected! ]\n"); + printk( "---------------------------------\n"); + printk("%s/%d is trying to contend lock (", + curr->comm, curr->pid); + print_lockdep_cache(lock); + printk(") at:\n"); + print_ip_sym(ip); + printk("but there are no locks held!\n"); + printk("\nother info that might help us debug this:\n"); + lockdep_print_held_locks(curr); + + printk("\nstack backtrace:\n"); + dump_stack(); + + return 0; +} + +static void +__lock_contended(struct lockdep_map *lock, unsigned long ip) +{ + struct task_struct *curr = current; + struct held_lock *hlock, *prev_hlock; + struct lock_class_stats *stats; + unsigned int depth; + int i, point; + + depth = curr->lockdep_depth; + if (DEBUG_LOCKS_WARN_ON(!depth)) + return; + + prev_hlock = NULL; + for (i = depth-1; i >= 0; i--) { + hlock = curr->held_locks + i; + /* + * We must not cross into another context: + */ + if (prev_hlock && prev_hlock->irq_context != hlock->irq_context) + break; + if (hlock->instance == lock) + goto found_it; + prev_hlock = hlock; + } + print_lock_contention_bug(curr, lock, ip); + return; + +found_it: + hlock->waittime_stamp = sched_clock(); + + point = lock_contention_point(hlock->class, ip); + + stats = get_lock_stats(hlock->class); + if (point < ARRAY_SIZE(stats->contention_point)) + stats->contention_point[i]++; + if (lock->cpu != smp_processor_id()) + stats->bounces[bounce_contended + !!hlock->read]++; + put_lock_stats(stats); +} + +static void +__lock_acquired(struct lockdep_map *lock) +{ + struct task_struct *curr = current; + struct held_lock *hlock, *prev_hlock; + struct lock_class_stats *stats; + unsigned int depth; + u64 now; + s64 waittime = 0; + int i, cpu; + + depth = curr->lockdep_depth; + if (DEBUG_LOCKS_WARN_ON(!depth)) + return; + + prev_hlock = NULL; + for (i = depth-1; i >= 0; i--) { + hlock = curr->held_locks + i; + /* + * We must not cross into another context: + */ + if (prev_hlock && prev_hlock->irq_context != hlock->irq_context) + break; + if (hlock->instance == lock) + goto found_it; + prev_hlock = hlock; + } + print_lock_contention_bug(curr, lock, _RET_IP_); + return; + +found_it: + cpu = smp_processor_id(); + if (hlock->waittime_stamp) { + now = sched_clock(); + waittime = now - hlock->waittime_stamp; + hlock->holdtime_stamp = now; + } + + stats = get_lock_stats(hlock->class); + if (waittime) { + if (hlock->read) + lock_time_inc(&stats->read_waittime, waittime); + else + lock_time_inc(&stats->write_waittime, waittime); + } + if (lock->cpu != cpu) + stats->bounces[bounce_acquired + !!hlock->read]++; + put_lock_stats(stats); + + lock->cpu = cpu; +} + +void lock_contended(struct lockdep_map *lock, unsigned long ip) +{ + unsigned long flags; + + if (unlikely(!lock_stat)) + return; + + if (unlikely(current->lockdep_recursion)) + return; + + raw_local_irq_save(flags); + check_flags(flags); + current->lockdep_recursion = 1; + __lock_contended(lock, ip); + current->lockdep_recursion = 0; + raw_local_irq_restore(flags); +} +EXPORT_SYMBOL_GPL(lock_contended); + +void lock_acquired(struct lockdep_map *lock) +{ + unsigned long flags; + + if (unlikely(!lock_stat)) + return; + + if (unlikely(current->lockdep_recursion)) + return; + + raw_local_irq_save(flags); + check_flags(flags); + current->lockdep_recursion = 1; + __lock_acquired(lock); + current->lockdep_recursion = 0; + raw_local_irq_restore(flags); +} +EXPORT_SYMBOL_GPL(lock_acquired); +#endif + /* * Used by the testsuite, sanitize the validator state * after a simulated failure: @@ -2636,8 +3046,11 @@ void __init lockdep_info(void) sizeof(struct held_lock) * MAX_LOCK_DEPTH); #ifdef CONFIG_DEBUG_LOCKDEP - if (lockdep_init_error) - printk("WARNING: lockdep init error! Arch code didnt call lockdep_init() early enough?\n"); + if (lockdep_init_error) { + printk("WARNING: lockdep init error! Arch code didn't call lockdep_init() early enough?\n"); + printk("Call stack leading to lockdep invocation was:\n"); + print_stack_trace(&lockdep_init_trace, 0); + } #endif } diff --git a/kernel/lockdep_proc.c b/kernel/lockdep_proc.c index 58f35e586ee..9f17af4a249 100644 --- a/kernel/lockdep_proc.c +++ b/kernel/lockdep_proc.c @@ -5,7 +5,8 @@ * * Started by Ingo Molnar: * - * Copyright (C) 2006 Red Hat, Inc., Ingo Molnar <mingo@redhat.com> + * Copyright (C) 2006,2007 Red Hat, Inc., Ingo Molnar <mingo@redhat.com> + * Copyright (C) 2007 Red Hat, Inc., Peter Zijlstra <pzijlstr@redhat.com> * * Code for /proc/lockdep and /proc/lockdep_stats: * @@ -15,6 +16,10 @@ #include <linux/seq_file.h> #include <linux/kallsyms.h> #include <linux/debug_locks.h> +#include <linux/vmalloc.h> +#include <linux/sort.h> +#include <asm/uaccess.h> +#include <asm/div64.h> #include "lockdep_internals.h" @@ -271,8 +276,10 @@ static int lockdep_stats_show(struct seq_file *m, void *v) if (nr_list_entries) factor = sum_forward_deps / nr_list_entries; +#ifdef CONFIG_PROVE_LOCKING seq_printf(m, " dependency chains: %11lu [max: %lu]\n", nr_lock_chains, MAX_LOCKDEP_CHAINS); +#endif #ifdef CONFIG_TRACE_IRQFLAGS seq_printf(m, " in-hardirq chains: %11u\n", @@ -342,6 +349,292 @@ static const struct file_operations proc_lockdep_stats_operations = { .release = seq_release, }; +#ifdef CONFIG_LOCK_STAT + +struct lock_stat_data { + struct lock_class *class; + struct lock_class_stats stats; +}; + +struct lock_stat_seq { + struct lock_stat_data *iter; + struct lock_stat_data *iter_end; + struct lock_stat_data stats[MAX_LOCKDEP_KEYS]; +}; + +/* + * sort on absolute number of contentions + */ +static int lock_stat_cmp(const void *l, const void *r) +{ + const struct lock_stat_data *dl = l, *dr = r; + unsigned long nl, nr; + + nl = dl->stats.read_waittime.nr + dl->stats.write_waittime.nr; + nr = dr->stats.read_waittime.nr + dr->stats.write_waittime.nr; + + return nr - nl; +} + +static void seq_line(struct seq_file *m, char c, int offset, int length) +{ + int i; + + for (i = 0; i < offset; i++) + seq_puts(m, " "); + for (i = 0; i < length; i++) + seq_printf(m, "%c", c); + seq_puts(m, "\n"); +} + +static void snprint_time(char *buf, size_t bufsiz, s64 nr) +{ + unsigned long rem; + + rem = do_div(nr, 1000); /* XXX: do_div_signed */ + snprintf(buf, bufsiz, "%lld.%02d", (long long)nr, ((int)rem+5)/10); +} + +static void seq_time(struct seq_file *m, s64 time) +{ + char num[15]; + + snprint_time(num, sizeof(num), time); + seq_printf(m, " %14s", num); +} + +static void seq_lock_time(struct seq_file *m, struct lock_time *lt) +{ + seq_printf(m, "%14lu", lt->nr); + seq_time(m, lt->min); + seq_time(m, lt->max); + seq_time(m, lt->total); +} + +static void seq_stats(struct seq_file *m, struct lock_stat_data *data) +{ + char name[39]; + struct lock_class *class; + struct lock_class_stats *stats; + int i, namelen; + + class = data->class; + stats = &data->stats; + + namelen = 38; + if (class->name_version > 1) + namelen -= 2; /* XXX truncates versions > 9 */ + if (class->subclass) + namelen -= 2; + + if (!class->name) { + char str[KSYM_NAME_LEN]; + const char *key_name; + + key_name = __get_key_name(class->key, str); + snprintf(name, namelen, "%s", key_name); + } else { + snprintf(name, namelen, "%s", class->name); + } + namelen = strlen(name); + if (class->name_version > 1) { + snprintf(name+namelen, 3, "#%d", class->name_version); + namelen += 2; + } + if (class->subclass) { + snprintf(name+namelen, 3, "/%d", class->subclass); + namelen += 2; + } + + if (stats->write_holdtime.nr) { + if (stats->read_holdtime.nr) + seq_printf(m, "%38s-W:", name); + else + seq_printf(m, "%40s:", name); + + seq_printf(m, "%14lu ", stats->bounces[bounce_contended_write]); + seq_lock_time(m, &stats->write_waittime); + seq_printf(m, " %14lu ", stats->bounces[bounce_acquired_write]); + seq_lock_time(m, &stats->write_holdtime); + seq_puts(m, "\n"); + } + + if (stats->read_holdtime.nr) { + seq_printf(m, "%38s-R:", name); + seq_printf(m, "%14lu ", stats->bounces[bounce_contended_read]); + seq_lock_time(m, &stats->read_waittime); + seq_printf(m, " %14lu ", stats->bounces[bounce_acquired_read]); + seq_lock_time(m, &stats->read_holdtime); + seq_puts(m, "\n"); + } + + if (stats->read_waittime.nr + stats->write_waittime.nr == 0) + return; + + if (stats->read_holdtime.nr) + namelen += 2; + + for (i = 0; i < ARRAY_SIZE(class->contention_point); i++) { + char sym[KSYM_SYMBOL_LEN]; + char ip[32]; + + if (class->contention_point[i] == 0) + break; + + if (!i) + seq_line(m, '-', 40-namelen, namelen); + + sprint_symbol(sym, class->contention_point[i]); + snprintf(ip, sizeof(ip), "[<%p>]", + (void *)class->contention_point[i]); + seq_printf(m, "%40s %14lu %29s %s\n", name, + stats->contention_point[i], + ip, sym); + } + if (i) { + seq_puts(m, "\n"); + seq_line(m, '.', 0, 40 + 1 + 10 * (14 + 1)); + seq_puts(m, "\n"); + } +} + +static void seq_header(struct seq_file *m) +{ + seq_printf(m, "lock_stat version 0.2\n"); + seq_line(m, '-', 0, 40 + 1 + 10 * (14 + 1)); + seq_printf(m, "%40s %14s %14s %14s %14s %14s %14s %14s %14s " + "%14s %14s\n", + "class name", + "con-bounces", + "contentions", + "waittime-min", + "waittime-max", + "waittime-total", + "acq-bounces", + "acquisitions", + "holdtime-min", + "holdtime-max", + "holdtime-total"); + seq_line(m, '-', 0, 40 + 1 + 10 * (14 + 1)); + seq_printf(m, "\n"); +} + +static void *ls_start(struct seq_file *m, loff_t *pos) +{ + struct lock_stat_seq *data = m->private; + + if (data->iter == data->stats) + seq_header(m); + + if (data->iter == data->iter_end) + data->iter = NULL; + + return data->iter; +} + +static void *ls_next(struct seq_file *m, void *v, loff_t *pos) +{ + struct lock_stat_seq *data = m->private; + + (*pos)++; + + data->iter = v; + data->iter++; + if (data->iter == data->iter_end) + data->iter = NULL; + + return data->iter; +} + +static void ls_stop(struct seq_file *m, void *v) +{ +} + +static int ls_show(struct seq_file *m, void *v) +{ + struct lock_stat_seq *data = m->private; + + seq_stats(m, data->iter); + return 0; +} + +static struct seq_operations lockstat_ops = { + .start = ls_start, + .next = ls_next, + .stop = ls_stop, + .show = ls_show, +}; + +static int lock_stat_open(struct inode *inode, struct file *file) +{ + int res; + struct lock_class *class; + struct lock_stat_seq *data = vmalloc(sizeof(struct lock_stat_seq)); + + if (!data) + return -ENOMEM; + + res = seq_open(file, &lockstat_ops); + if (!res) { + struct lock_stat_data *iter = data->stats; + struct seq_file *m = file->private_data; + + data->iter = iter; + list_for_each_entry(class, &all_lock_classes, lock_entry) { + iter->class = class; + iter->stats = lock_stats(class); + iter++; + } + data->iter_end = iter; + + sort(data->stats, data->iter_end - data->iter, + sizeof(struct lock_stat_data), + lock_stat_cmp, NULL); + + m->private = data; + } else + vfree(data); + + return res; +} + +static ssize_t lock_stat_write(struct file *file, const char __user *buf, + size_t count, loff_t *ppos) +{ + struct lock_class *class; + char c; + + if (count) { + if (get_user(c, buf)) + return -EFAULT; + + if (c != '0') + return count; + + list_for_each_entry(class, &all_lock_classes, lock_entry) + clear_lock_stats(class); + } + return count; +} + +static int lock_stat_release(struct inode *inode, struct file *file) +{ + struct seq_file *seq = file->private_data; + + vfree(seq->private); + seq->private = NULL; + return seq_release(inode, file); +} + +static const struct file_operations proc_lock_stat_operations = { + .open = lock_stat_open, + .write = lock_stat_write, + .read = seq_read, + .llseek = seq_lseek, + .release = lock_stat_release, +}; +#endif /* CONFIG_LOCK_STAT */ + static int __init lockdep_proc_init(void) { struct proc_dir_entry *entry; @@ -354,6 +647,12 @@ static int __init lockdep_proc_init(void) if (entry) entry->proc_fops = &proc_lockdep_stats_operations; +#ifdef CONFIG_LOCK_STAT + entry = create_proc_entry("lock_stat", S_IRUSR, NULL); + if (entry) + entry->proc_fops = &proc_lock_stat_operations; +#endif + return 0; } diff --git a/kernel/mutex.c b/kernel/mutex.c index 303eab18484..691b86564dd 100644 --- a/kernel/mutex.c +++ b/kernel/mutex.c @@ -139,6 +139,12 @@ __mutex_lock_common(struct mutex *lock, long state, unsigned int subclass) list_add_tail(&waiter.list, &lock->wait_list); waiter.task = task; + old_val = atomic_xchg(&lock->count, -1); + if (old_val == 1) + goto done; + + lock_contended(&lock->dep_map, _RET_IP_); + for (;;) { /* * Lets try to take the lock again - this is needed even if @@ -174,6 +180,8 @@ __mutex_lock_common(struct mutex *lock, long state, unsigned int subclass) spin_lock_mutex(&lock->wait_lock, flags); } +done: + lock_acquired(&lock->dep_map); /* got the lock - rejoice! */ mutex_remove_waiter(lock, &waiter, task_thread_info(task)); debug_mutex_set_owner(lock, task_thread_info(task)); diff --git a/kernel/power/Kconfig b/kernel/power/Kconfig index 73328476761..7358609e473 100644 --- a/kernel/power/Kconfig +++ b/kernel/power/Kconfig @@ -33,13 +33,20 @@ config PM_DEBUG bool "Power Management Debug Support" depends on PM ---help--- - This option enables verbose debugging support in the Power Management - code. This is helpful when debugging and reporting various PM bugs, - like suspend support. + This option enables various debugging support in the Power Management + code. This is helpful when debugging and reporting PM bugs, like + suspend support. + +config PM_VERBOSE + bool "Verbose Power Management debugging" + depends on PM_DEBUG + default n + ---help--- + This option enables verbose messages from the Power Management code. config DISABLE_CONSOLE_SUSPEND bool "Keep console(s) enabled during suspend/resume (DANGEROUS)" - depends on PM && PM_DEBUG + depends on PM_DEBUG default n ---help--- This option turns off the console suspend mechanism that prevents @@ -50,7 +57,7 @@ config DISABLE_CONSOLE_SUSPEND config PM_TRACE bool "Suspend/resume event tracing" - depends on PM && PM_DEBUG && X86_32 && EXPERIMENTAL + depends on PM_DEBUG && X86_32 && EXPERIMENTAL default n ---help--- This enables some cheesy code to save the last PM event point in the diff --git a/kernel/power/disk.c b/kernel/power/disk.c index f445b9cd60f..324ac0188ce 100644 --- a/kernel/power/disk.c +++ b/kernel/power/disk.c @@ -45,7 +45,7 @@ enum { static int hibernation_mode = HIBERNATION_SHUTDOWN; -struct hibernation_ops *hibernation_ops; +static struct hibernation_ops *hibernation_ops; /** * hibernation_set_ops - set the global hibernate operations @@ -54,7 +54,8 @@ struct hibernation_ops *hibernation_ops; void hibernation_set_ops(struct hibernation_ops *ops) { - if (ops && !(ops->prepare && ops->enter && ops->finish)) { + if (ops && !(ops->prepare && ops->enter && ops->finish + && ops->pre_restore && ops->restore_cleanup)) { WARN_ON(1); return; } @@ -74,9 +75,9 @@ void hibernation_set_ops(struct hibernation_ops *ops) * platform driver if so configured and return an error code if it fails */ -static int platform_prepare(void) +static int platform_prepare(int platform_mode) { - return (hibernation_mode == HIBERNATION_PLATFORM && hibernation_ops) ? + return (platform_mode && hibernation_ops) ? hibernation_ops->prepare() : 0; } @@ -85,13 +86,145 @@ static int platform_prepare(void) * using the platform driver (must be called after platform_prepare()) */ -static void platform_finish(void) +static void platform_finish(int platform_mode) { - if (hibernation_mode == HIBERNATION_PLATFORM && hibernation_ops) + if (platform_mode && hibernation_ops) hibernation_ops->finish(); } /** + * platform_pre_restore - prepare the platform for the restoration from a + * hibernation image. If the restore fails after this function has been + * called, platform_restore_cleanup() must be called. + */ + +static int platform_pre_restore(int platform_mode) +{ + return (platform_mode && hibernation_ops) ? + hibernation_ops->pre_restore() : 0; +} + +/** + * platform_restore_cleanup - switch the platform to the normal mode of + * operation after a failing restore. If platform_pre_restore() has been + * called before the failing restore, this function must be called too, + * regardless of the result of platform_pre_restore(). + */ + +static void platform_restore_cleanup(int platform_mode) +{ + if (platform_mode && hibernation_ops) + hibernation_ops->restore_cleanup(); +} + +/** + * hibernation_snapshot - quiesce devices and create the hibernation + * snapshot image. + * @platform_mode - if set, use the platform driver, if available, to + * prepare the platform frimware for the power transition. + * + * Must be called with pm_mutex held + */ + +int hibernation_snapshot(int platform_mode) +{ + int error; + + /* Free memory before shutting down devices. */ + error = swsusp_shrink_memory(); + if (error) + return error; + + suspend_console(); + error = device_suspend(PMSG_FREEZE); + if (error) + goto Resume_console; + + error = platform_prepare(platform_mode); + if (error) + goto Resume_devices; + + error = disable_nonboot_cpus(); + if (!error) { + if (hibernation_mode != HIBERNATION_TEST) { + in_suspend = 1; + error = swsusp_suspend(); + /* Control returns here after successful restore */ + } else { + printk("swsusp debug: Waiting for 5 seconds.\n"); + mdelay(5000); + } + } + enable_nonboot_cpus(); + Resume_devices: + platform_finish(platform_mode); + device_resume(); + Resume_console: + resume_console(); + return error; +} + +/** + * hibernation_restore - quiesce devices and restore the hibernation + * snapshot image. If successful, control returns in hibernation_snaphot() + * @platform_mode - if set, use the platform driver, if available, to + * prepare the platform frimware for the transition. + * + * Must be called with pm_mutex held + */ + +int hibernation_restore(int platform_mode) +{ + int error; + + pm_prepare_console(); + suspend_console(); + error = device_suspend(PMSG_PRETHAW); + if (error) + goto Finish; + + error = platform_pre_restore(platform_mode); + if (!error) { + error = disable_nonboot_cpus(); + if (!error) + error = swsusp_resume(); + enable_nonboot_cpus(); + } + platform_restore_cleanup(platform_mode); + device_resume(); + Finish: + resume_console(); + pm_restore_console(); + return error; +} + +/** + * hibernation_platform_enter - enter the hibernation state using the + * platform driver (if available) + */ + +int hibernation_platform_enter(void) +{ + int error; + + if (hibernation_ops) { + kernel_shutdown_prepare(SYSTEM_SUSPEND_DISK); + /* + * We have cancelled the power transition by running + * hibernation_ops->finish() before saving the image, so we + * should let the firmware know that we're going to enter the + * sleep state after all + */ + error = hibernation_ops->prepare(); + if (!error) + error = hibernation_ops->enter(); + } else { + error = -ENOSYS; + } + return error; +} + +/** * power_down - Shut the machine down for hibernation. * * Use the platform driver, if configured so; otherwise try @@ -111,11 +244,7 @@ static void power_down(void) kernel_restart(NULL); break; case HIBERNATION_PLATFORM: - if (hibernation_ops) { - kernel_shutdown_prepare(SYSTEM_SUSPEND_DISK); - hibernation_ops->enter(); - break; - } + hibernation_platform_enter(); } kernel_halt(); /* @@ -152,9 +281,16 @@ int hibernate(void) { int error; + mutex_lock(&pm_mutex); /* The snapshot device should not be opened while we're running */ - if (!atomic_add_unless(&snapshot_device_available, -1, 0)) - return -EBUSY; + if (!atomic_add_unless(&snapshot_device_available, -1, 0)) { + error = -EBUSY; + goto Unlock; + } + + error = pm_notifier_call_chain(PM_HIBERNATION_PREPARE); + if (error) + goto Exit; /* Allocate memory management structures */ error = create_basic_memory_bitmaps(); @@ -165,75 +301,35 @@ int hibernate(void) if (error) goto Finish; - mutex_lock(&pm_mutex); if (hibernation_mode == HIBERNATION_TESTPROC) { printk("swsusp debug: Waiting for 5 seconds.\n"); mdelay(5000); goto Thaw; } + error = hibernation_snapshot(hibernation_mode == HIBERNATION_PLATFORM); + if (in_suspend && !error) { + unsigned int flags = 0; - /* Free memory before shutting down devices. */ - error = swsusp_shrink_memory(); - if (error) - goto Thaw; - - error = platform_prepare(); - if (error) - goto Thaw; - - suspend_console(); - error = device_suspend(PMSG_FREEZE); - if (error) { - printk(KERN_ERR "PM: Some devices failed to suspend\n"); - goto Resume_devices; - } - error = disable_nonboot_cpus(); - if (error) - goto Enable_cpus; - - if (hibernation_mode == HIBERNATION_TEST) { - printk("swsusp debug: Waiting for 5 seconds.\n"); - mdelay(5000); - goto Enable_cpus; - } - - pr_debug("PM: snapshotting memory.\n"); - in_suspend = 1; - error = swsusp_suspend(); - if (error) - goto Enable_cpus; - - if (in_suspend) { - enable_nonboot_cpus(); - platform_finish(); - device_resume(); - resume_console(); + if (hibernation_mode == HIBERNATION_PLATFORM) + flags |= SF_PLATFORM_MODE; pr_debug("PM: writing image.\n"); - error = swsusp_write(); + error = swsusp_write(flags); + swsusp_free(); if (!error) power_down(); - else { - swsusp_free(); - goto Thaw; - } } else { pr_debug("PM: Image restored successfully.\n"); + swsusp_free(); } - - swsusp_free(); - Enable_cpus: - enable_nonboot_cpus(); - Resume_devices: - platform_finish(); - device_resume(); - resume_console(); Thaw: - mutex_unlock(&pm_mutex); unprepare_processes(); Finish: free_basic_memory_bitmaps(); Exit: + pm_notifier_call_chain(PM_POST_HIBERNATION); atomic_inc(&snapshot_device_available); + Unlock: + mutex_unlock(&pm_mutex); return error; } @@ -253,6 +349,7 @@ int hibernate(void) static int software_resume(void) { int error; + unsigned int flags; mutex_lock(&pm_mutex); if (!swsusp_resume_device) { @@ -300,30 +397,12 @@ static int software_resume(void) pr_debug("PM: Reading swsusp image.\n"); - error = swsusp_read(); - if (error) { - swsusp_free(); - goto Thaw; - } - - pr_debug("PM: Preparing devices for restore.\n"); - - suspend_console(); - error = device_suspend(PMSG_PRETHAW); - if (error) - goto Free; - - error = disable_nonboot_cpus(); + error = swsusp_read(&flags); if (!error) - swsusp_resume(); + hibernation_restore(flags & SF_PLATFORM_MODE); - enable_nonboot_cpus(); - Free: - swsusp_free(); - device_resume(); - resume_console(); - Thaw: printk(KERN_ERR "PM: Restore failed, recovering.\n"); + swsusp_free(); unprepare_processes(); Done: free_basic_memory_bitmaps(); @@ -333,7 +412,7 @@ static int software_resume(void) Unlock: mutex_unlock(&pm_mutex); pr_debug("PM: Resume from disk failed.\n"); - return 0; + return error; } late_initcall(software_resume); diff --git a/kernel/power/main.c b/kernel/power/main.c index fc45ed22620..32147b57c3b 100644 --- a/kernel/power/main.c +++ b/kernel/power/main.c @@ -23,6 +23,8 @@ #include "power.h" +BLOCKING_NOTIFIER_HEAD(pm_chain_head); + /*This is just an arbitrary number */ #define FREE_PAGE_NUMBER (100) @@ -63,14 +65,11 @@ static inline void pm_finish(suspend_state_t state) /** * suspend_prepare - Do prep work before entering low-power state. - * @state: State we're entering. * - * This is common code that is called for each state that we're - * entering. Allocate a console, stop all processes, then make sure - * the platform can enter the requested state. + * This is common code that is called for each state that we're entering. + * Run suspend notifiers, allocate a console and stop all processes. */ - -static int suspend_prepare(suspend_state_t state) +static int suspend_prepare(void) { int error; unsigned int free_pages; @@ -78,6 +77,10 @@ static int suspend_prepare(suspend_state_t state) if (!pm_ops || !pm_ops->enter) return -EPERM; + error = pm_notifier_call_chain(PM_SUSPEND_PREPARE); + if (error) + goto Finish; + pm_prepare_console(); if (freeze_processes()) { @@ -85,46 +88,23 @@ static int suspend_prepare(suspend_state_t state) goto Thaw; } - if ((free_pages = global_page_state(NR_FREE_PAGES)) - < FREE_PAGE_NUMBER) { + free_pages = global_page_state(NR_FREE_PAGES); + if (free_pages < FREE_PAGE_NUMBER) { pr_debug("PM: free some memory\n"); shrink_all_memory(FREE_PAGE_NUMBER - free_pages); if (nr_free_pages() < FREE_PAGE_NUMBER) { error = -ENOMEM; printk(KERN_ERR "PM: No enough memory\n"); - goto Thaw; } } - - if (pm_ops->set_target) { - error = pm_ops->set_target(state); - if (error) - goto Thaw; - } - suspend_console(); - error = device_suspend(PMSG_SUSPEND); - if (error) { - printk(KERN_ERR "Some devices failed to suspend\n"); - goto Resume_console; - } - if (pm_ops->prepare) { - if ((error = pm_ops->prepare(state))) - goto Resume_devices; - } - - error = disable_nonboot_cpus(); if (!error) return 0; - enable_nonboot_cpus(); - pm_finish(state); - Resume_devices: - device_resume(); - Resume_console: - resume_console(); Thaw: thaw_processes(); pm_restore_console(); + Finish: + pm_notifier_call_chain(PM_POST_SUSPEND); return error; } @@ -140,6 +120,12 @@ void __attribute__ ((weak)) arch_suspend_enable_irqs(void) local_irq_enable(); } +/** + * suspend_enter - enter the desired system sleep state. + * @state: state to enter + * + * This function should be called after devices have been suspended. + */ int suspend_enter(suspend_state_t state) { int error = 0; @@ -159,23 +145,58 @@ int suspend_enter(suspend_state_t state) return error; } +/** + * suspend_devices_and_enter - suspend devices and enter the desired system sleep + * state. + * @state: state to enter + */ +int suspend_devices_and_enter(suspend_state_t state) +{ + int error; + + if (!pm_ops) + return -ENOSYS; + + if (pm_ops->set_target) { + error = pm_ops->set_target(state); + if (error) + return error; + } + suspend_console(); + error = device_suspend(PMSG_SUSPEND); + if (error) { + printk(KERN_ERR "Some devices failed to suspend\n"); + goto Resume_console; + } + if (pm_ops->prepare) { + error = pm_ops->prepare(state); + if (error) + goto Resume_devices; + } + error = disable_nonboot_cpus(); + if (!error) + suspend_enter(state); + + enable_nonboot_cpus(); + pm_finish(state); + Resume_devices: + device_resume(); + Resume_console: + resume_console(); + return error; +} /** * suspend_finish - Do final work before exiting suspend sequence. - * @state: State we're coming out of. * * Call platform code to clean up, restart processes, and free the * console that we've allocated. This is not called for suspend-to-disk. */ - -static void suspend_finish(suspend_state_t state) +static void suspend_finish(void) { - enable_nonboot_cpus(); - pm_finish(state); - device_resume(); - resume_console(); thaw_processes(); pm_restore_console(); + pm_notifier_call_chain(PM_POST_SUSPEND); } @@ -207,7 +228,6 @@ static inline int valid_state(suspend_state_t state) * Then, do the setup for suspend, enter the state, and cleaup (after * we've woken up). */ - static int enter_state(suspend_state_t state) { int error; @@ -218,14 +238,14 @@ static int enter_state(suspend_state_t state) return -EBUSY; pr_debug("PM: Preparing system for %s sleep\n", pm_states[state]); - if ((error = suspend_prepare(state))) + if ((error = suspend_prepare())) goto Unlock; pr_debug("PM: Entering %s sleep\n", pm_states[state]); - error = suspend_enter(state); + error = suspend_devices_and_enter(state); pr_debug("PM: Finishing wakeup.\n"); - suspend_finish(state); + suspend_finish(); Unlock: mutex_unlock(&pm_mutex); return error; diff --git a/kernel/power/power.h b/kernel/power/power.h index 51381487103..5f24c786f8e 100644 --- a/kernel/power/power.h +++ b/kernel/power/power.h @@ -25,7 +25,10 @@ struct swsusp_info { */ #define SPARE_PAGES ((1024 * 1024) >> PAGE_SHIFT) -extern struct hibernation_ops *hibernation_ops; +/* kernel/power/disk.c */ +extern int hibernation_snapshot(int platform_mode); +extern int hibernation_restore(int platform_mode); +extern int hibernation_platform_enter(void); #endif extern int pfn_is_nosave(unsigned long); @@ -152,16 +155,34 @@ extern sector_t alloc_swapdev_block(int swap); extern void free_all_swap_pages(int swap); extern int swsusp_swap_in_use(void); +/* + * Flags that can be passed from the hibernatig hernel to the "boot" kernel in + * the image header. + */ +#define SF_PLATFORM_MODE 1 + +/* kernel/power/disk.c */ extern int swsusp_check(void); extern int swsusp_shrink_memory(void); extern void swsusp_free(void); extern int swsusp_suspend(void); extern int swsusp_resume(void); -extern int swsusp_read(void); -extern int swsusp_write(void); +extern int swsusp_read(unsigned int *flags_p); +extern int swsusp_write(unsigned int flags); extern void swsusp_close(void); -extern int suspend_enter(suspend_state_t state); struct timeval; +/* kernel/power/swsusp.c */ extern void swsusp_show_speed(struct timeval *, struct timeval *, unsigned int, char *); + +/* kernel/power/main.c */ +extern int suspend_enter(suspend_state_t state); +extern int suspend_devices_and_enter(suspend_state_t state); +extern struct blocking_notifier_head pm_chain_head; + +static inline int pm_notifier_call_chain(unsigned long val) +{ + return (blocking_notifier_call_chain(&pm_chain_head, val, NULL) + == NOTIFY_BAD) ? -EINVAL : 0; +} diff --git a/kernel/power/process.c b/kernel/power/process.c index e0233d8422b..3434940a3df 100644 --- a/kernel/power/process.c +++ b/kernel/power/process.c @@ -40,7 +40,7 @@ static inline void frozen_process(void) current->flags |= PF_FROZEN; wmb(); } - clear_tsk_thread_flag(current, TIF_FREEZE); + clear_freeze_flag(current); } /* Refrigerator is place where frozen processes are stored :-). */ @@ -72,20 +72,19 @@ void refrigerator(void) schedule(); } pr_debug("%s left refrigerator\n", current->comm); - current->state = save; + __set_current_state(save); } -static inline void freeze_process(struct task_struct *p) +static void freeze_task(struct task_struct *p) { unsigned long flags; if (!freezing(p)) { rmb(); if (!frozen(p)) { + set_freeze_flag(p); if (p->state == TASK_STOPPED) force_sig_specific(SIGSTOP, p); - - freeze(p); spin_lock_irqsave(&p->sighand->siglock, flags); signal_wake_up(p, p->state == TASK_STOPPED); spin_unlock_irqrestore(&p->sighand->siglock, flags); @@ -99,19 +98,14 @@ static void cancel_freezing(struct task_struct *p) if (freezing(p)) { pr_debug(" clean up: %s\n", p->comm); - do_not_freeze(p); + clear_freeze_flag(p); spin_lock_irqsave(&p->sighand->siglock, flags); recalc_sigpending_and_wake(p); spin_unlock_irqrestore(&p->sighand->siglock, flags); } } -static inline int is_user_space(struct task_struct *p) -{ - return p->mm && !(p->flags & PF_BORROWED_MM); -} - -static unsigned int try_to_freeze_tasks(int freeze_user_space) +static int try_to_freeze_tasks(int freeze_user_space) { struct task_struct *g, *p; unsigned long end_time; @@ -122,26 +116,40 @@ static unsigned int try_to_freeze_tasks(int freeze_user_space) todo = 0; read_lock(&tasklist_lock); do_each_thread(g, p) { - if (!freezeable(p)) + if (frozen(p) || !freezeable(p)) continue; - if (frozen(p)) - continue; - - if (p->state == TASK_TRACED && frozen(p->parent)) { - cancel_freezing(p); - continue; + if (freeze_user_space) { + if (p->state == TASK_TRACED && + frozen(p->parent)) { + cancel_freezing(p); + continue; + } + /* + * Kernel threads should not have TIF_FREEZE set + * at this point, so we must ensure that either + * p->mm is not NULL *and* PF_BORROWED_MM is + * unset, or TIF_FRREZE is left unset. + * The task_lock() is necessary to prevent races + * with exit_mm() or use_mm()/unuse_mm() from + * occuring. + */ + task_lock(p); + if (!p->mm || (p->flags & PF_BORROWED_MM)) { + task_unlock(p); + continue; + } + freeze_task(p); + task_unlock(p); + } else { + freeze_task(p); } - if (freeze_user_space && !is_user_space(p)) - continue; - - freeze_process(p); if (!freezer_should_skip(p)) todo++; } while_each_thread(g, p); read_unlock(&tasklist_lock); yield(); /* Yield is okay here */ - if (todo && time_after(jiffies, end_time)) + if (time_after(jiffies, end_time)) break; } while (todo); @@ -152,49 +160,41 @@ static unsigned int try_to_freeze_tasks(int freeze_user_space) * but it cleans up leftover PF_FREEZE requests. */ printk("\n"); - printk(KERN_ERR "Stopping %s timed out after %d seconds " + printk(KERN_ERR "Freezing of %s timed out after %d seconds " "(%d tasks refusing to freeze):\n", - freeze_user_space ? "user space processes" : - "kernel threads", + freeze_user_space ? "user space " : "tasks ", TIMEOUT / HZ, todo); + show_state(); read_lock(&tasklist_lock); do_each_thread(g, p) { - if (freeze_user_space && !is_user_space(p)) - continue; - task_lock(p); - if (freezeable(p) && !frozen(p) && - !freezer_should_skip(p)) + if (freezing(p) && !freezer_should_skip(p)) printk(KERN_ERR " %s\n", p->comm); - cancel_freezing(p); task_unlock(p); } while_each_thread(g, p); read_unlock(&tasklist_lock); } - return todo; + return todo ? -EBUSY : 0; } /** * freeze_processes - tell processes to enter the refrigerator - * - * Returns 0 on success, or the number of processes that didn't freeze, - * although they were told to. */ int freeze_processes(void) { - unsigned int nr_unfrozen; + int error; printk("Stopping tasks ... "); - nr_unfrozen = try_to_freeze_tasks(FREEZER_USER_SPACE); - if (nr_unfrozen) - return nr_unfrozen; + error = try_to_freeze_tasks(FREEZER_USER_SPACE); + if (error) + return error; sys_sync(); - nr_unfrozen = try_to_freeze_tasks(FREEZER_KERNEL_THREADS); - if (nr_unfrozen) - return nr_unfrozen; + error = try_to_freeze_tasks(FREEZER_KERNEL_THREADS); + if (error) + return error; printk("done.\n"); BUG_ON(in_atomic()); @@ -210,7 +210,7 @@ static void thaw_tasks(int thaw_user_space) if (!freezeable(p)) continue; - if (is_user_space(p) == !thaw_user_space) + if (!p->mm == thaw_user_space) continue; thaw_process(p); diff --git a/kernel/power/swap.c b/kernel/power/swap.c index 8b1a1b83714..917aba10057 100644 --- a/kernel/power/swap.c +++ b/kernel/power/swap.c @@ -33,8 +33,9 @@ extern char resume_file[]; #define SWSUSP_SIG "S1SUSPEND" struct swsusp_header { - char reserved[PAGE_SIZE - 20 - sizeof(sector_t)]; + char reserved[PAGE_SIZE - 20 - sizeof(sector_t) - sizeof(int)]; sector_t image; + unsigned int flags; /* Flags to pass to the "boot" kernel */ char orig_sig[10]; char sig[10]; } __attribute__((packed)); @@ -138,7 +139,7 @@ static int wait_on_bio_chain(struct bio **bio_chain) * Saving part */ -static int mark_swapfiles(sector_t start) +static int mark_swapfiles(sector_t start, unsigned int flags) { int error; @@ -148,6 +149,7 @@ static int mark_swapfiles(sector_t start) memcpy(swsusp_header->orig_sig,swsusp_header->sig, 10); memcpy(swsusp_header->sig,SWSUSP_SIG, 10); swsusp_header->image = start; + swsusp_header->flags = flags; error = bio_write_page(swsusp_resume_block, swsusp_header, NULL); } else { @@ -369,6 +371,7 @@ static int enough_swap(unsigned int nr_pages) /** * swsusp_write - Write entire image and metadata. + * @flags: flags to pass to the "boot" kernel in the image header * * It is important _NOT_ to umount filesystems at this point. We want * them synced (in case something goes wrong) but we DO not want to mark @@ -376,7 +379,7 @@ static int enough_swap(unsigned int nr_pages) * correctly, we'll mark system clean, anyway.) */ -int swsusp_write(void) +int swsusp_write(unsigned int flags) { struct swap_map_handle handle; struct snapshot_handle snapshot; @@ -415,7 +418,7 @@ int swsusp_write(void) if (!error) { flush_swap_writer(&handle); printk("S"); - error = mark_swapfiles(start); + error = mark_swapfiles(start, flags); printk("|\n"); } } @@ -540,13 +543,20 @@ static int load_image(struct swap_map_handle *handle, return error; } -int swsusp_read(void) +/** + * swsusp_read - read the hibernation image. + * @flags_p: flags passed by the "frozen" kernel in the image header should + * be written into this memeory location + */ + +int swsusp_read(unsigned int *flags_p) { int error; struct swap_map_handle handle; struct snapshot_handle snapshot; struct swsusp_info *header; + *flags_p = swsusp_header->flags; if (IS_ERR(resume_bdev)) { pr_debug("swsusp: block device not initialised\n"); return PTR_ERR(resume_bdev); diff --git a/kernel/power/user.c b/kernel/power/user.c index d65305b515b..bd0723a7df3 100644 --- a/kernel/power/user.c +++ b/kernel/power/user.c @@ -128,92 +128,6 @@ static ssize_t snapshot_write(struct file *filp, const char __user *buf, return res; } -static inline int platform_prepare(void) -{ - int error = 0; - - if (hibernation_ops) - error = hibernation_ops->prepare(); - - return error; -} - -static inline void platform_finish(void) -{ - if (hibernation_ops) - hibernation_ops->finish(); -} - -static inline int snapshot_suspend(int platform_suspend) -{ - int error; - - mutex_lock(&pm_mutex); - /* Free memory before shutting down devices. */ - error = swsusp_shrink_memory(); - if (error) - goto Finish; - - if (platform_suspend) { - error = platform_prepare(); - if (error) - goto Finish; - } - suspend_console(); - error = device_suspend(PMSG_FREEZE); - if (error) - goto Resume_devices; - - error = disable_nonboot_cpus(); - if (!error) { - in_suspend = 1; - error = swsusp_suspend(); - } - enable_nonboot_cpus(); - Resume_devices: - if (platform_suspend) - platform_finish(); - - device_resume(); - resume_console(); - Finish: - mutex_unlock(&pm_mutex); - return error; -} - -static inline int snapshot_restore(int platform_suspend) -{ - int error; - - mutex_lock(&pm_mutex); - pm_prepare_console(); - if (platform_suspend) { - error = platform_prepare(); - if (error) - goto Finish; - } - suspend_console(); - error = device_suspend(PMSG_PRETHAW); - if (error) - goto Resume_devices; - - error = disable_nonboot_cpus(); - if (!error) - error = swsusp_resume(); - - enable_nonboot_cpus(); - Resume_devices: - if (platform_suspend) - platform_finish(); - - device_resume(); - resume_console(); - Finish: - pm_restore_console(); - mutex_unlock(&pm_mutex); - return error; -} - static int snapshot_ioctl(struct inode *inode, struct file *filp, unsigned int cmd, unsigned long arg) { @@ -237,10 +151,14 @@ static int snapshot_ioctl(struct inode *inode, struct file *filp, if (data->frozen) break; mutex_lock(&pm_mutex); - if (freeze_processes()) { - thaw_processes(); - error = -EBUSY; + error = pm_notifier_call_chain(PM_HIBERNATION_PREPARE); + if (!error) { + error = freeze_processes(); + if (error) + thaw_processes(); } + if (error) + pm_notifier_call_chain(PM_POST_HIBERNATION); mutex_unlock(&pm_mutex); if (!error) data->frozen = 1; @@ -251,6 +169,7 @@ static int snapshot_ioctl(struct inode *inode, struct file *filp, break; mutex_lock(&pm_mutex); thaw_processes(); + pm_notifier_call_chain(PM_POST_HIBERNATION); mutex_unlock(&pm_mutex); data->frozen = 0; break; @@ -260,7 +179,7 @@ static int snapshot_ioctl(struct inode *inode, struct file *filp, error = -EPERM; break; } - error = snapshot_suspend(data->platform_suspend); + error = hibernation_snapshot(data->platform_suspend); if (!error) error = put_user(in_suspend, (unsigned int __user *)arg); if (!error) @@ -274,7 +193,7 @@ static int snapshot_ioctl(struct inode *inode, struct file *filp, error = -EPERM; break; } - error = snapshot_restore(data->platform_suspend); + error = hibernation_restore(data->platform_suspend); break; case SNAPSHOT_FREE: @@ -336,47 +255,19 @@ static int snapshot_ioctl(struct inode *inode, struct file *filp, break; case SNAPSHOT_S2RAM: - if (!pm_ops) { - error = -ENOSYS; - break; - } - if (!data->frozen) { error = -EPERM; break; } - if (!mutex_trylock(&pm_mutex)) { error = -EBUSY; break; } - - if (pm_ops->prepare) { - error = pm_ops->prepare(PM_SUSPEND_MEM); - if (error) - goto OutS3; - } - - /* Put devices to sleep */ - suspend_console(); - error = device_suspend(PMSG_SUSPEND); - if (error) { - printk(KERN_ERR "Failed to suspend some devices.\n"); - } else { - error = disable_nonboot_cpus(); - if (!error) { - /* Enter S3, system is already frozen */ - suspend_enter(PM_SUSPEND_MEM); - enable_nonboot_cpus(); - } - /* Wake up devices */ - device_resume(); - } - resume_console(); - if (pm_ops->finish) - pm_ops->finish(PM_SUSPEND_MEM); - - OutS3: + /* + * Tasks are frozen and the notifiers have been called with + * PM_HIBERNATION_PREPARE + */ + error = suspend_devices_and_enter(PM_SUSPEND_MEM); mutex_unlock(&pm_mutex); break; @@ -386,19 +277,14 @@ static int snapshot_ioctl(struct inode *inode, struct file *filp, switch (arg) { case PMOPS_PREPARE: - if (hibernation_ops) { - data->platform_suspend = 1; - error = 0; - } else { - error = -ENOSYS; - } + data->platform_suspend = 1; + error = 0; break; case PMOPS_ENTER: - if (data->platform_suspend) { - kernel_shutdown_prepare(SYSTEM_SUSPEND_DISK); - error = hibernation_ops->enter(); - } + if (data->platform_suspend) + error = hibernation_platform_enter(); + break; case PMOPS_FINISH: diff --git a/kernel/ptrace.c b/kernel/ptrace.c index 4a1745f1dad..82a558b655d 100644 --- a/kernel/ptrace.c +++ b/kernel/ptrace.c @@ -142,7 +142,7 @@ static int may_attach(struct task_struct *task) return -EPERM; smp_rmb(); if (task->mm) - dumpable = task->mm->dumpable; + dumpable = get_dumpable(task->mm); if (!dumpable && !capable(CAP_SYS_PTRACE)) return -EPERM; diff --git a/kernel/relay.c b/kernel/relay.c index a615a8f513f..510fbbd7b50 100644 --- a/kernel/relay.c +++ b/kernel/relay.c @@ -80,7 +80,7 @@ static struct vm_operations_struct relay_file_mmap_ops = { * * Caller should already have grabbed mmap_sem. */ -int relay_mmap_buf(struct rchan_buf *buf, struct vm_area_struct *vma) +static int relay_mmap_buf(struct rchan_buf *buf, struct vm_area_struct *vma) { unsigned long length = vma->vm_end - vma->vm_start; struct file *filp = vma->vm_file; @@ -145,7 +145,7 @@ depopulate: * * Returns channel buffer if successful, %NULL otherwise. */ -struct rchan_buf *relay_create_buf(struct rchan *chan) +static struct rchan_buf *relay_create_buf(struct rchan *chan) { struct rchan_buf *buf = kzalloc(sizeof(struct rchan_buf), GFP_KERNEL); if (!buf) @@ -175,7 +175,7 @@ free_buf: * * Should only be called from kref_put(). */ -void relay_destroy_channel(struct kref *kref) +static void relay_destroy_channel(struct kref *kref) { struct rchan *chan = container_of(kref, struct rchan, kref); kfree(chan); @@ -185,7 +185,7 @@ void relay_destroy_channel(struct kref *kref) * relay_destroy_buf - destroy an rchan_buf struct and associated buffer * @buf: the buffer struct */ -void relay_destroy_buf(struct rchan_buf *buf) +static void relay_destroy_buf(struct rchan_buf *buf) { struct rchan *chan = buf->chan; unsigned int i; @@ -210,7 +210,7 @@ void relay_destroy_buf(struct rchan_buf *buf) * rchan_buf_struct and the channel buffer. Should only be called from * kref_put(). */ -void relay_remove_buf(struct kref *kref) +static void relay_remove_buf(struct kref *kref) { struct rchan_buf *buf = container_of(kref, struct rchan_buf, kref); buf->chan->cb->remove_buf_file(buf->dentry); @@ -223,11 +223,10 @@ void relay_remove_buf(struct kref *kref) * * Returns 1 if the buffer is empty, 0 otherwise. */ -int relay_buf_empty(struct rchan_buf *buf) +static int relay_buf_empty(struct rchan_buf *buf) { return (buf->subbufs_produced - buf->subbufs_consumed) ? 0 : 1; } -EXPORT_SYMBOL_GPL(relay_buf_empty); /** * relay_buf_full - boolean, is the channel buffer full? diff --git a/kernel/rwsem.c b/kernel/rwsem.c index 9a87886b022..1ec620c0306 100644 --- a/kernel/rwsem.c +++ b/kernel/rwsem.c @@ -20,7 +20,7 @@ void down_read(struct rw_semaphore *sem) might_sleep(); rwsem_acquire_read(&sem->dep_map, 0, 0, _RET_IP_); - __down_read(sem); + LOCK_CONTENDED(sem, __down_read_trylock, __down_read); } EXPORT_SYMBOL(down_read); @@ -47,7 +47,7 @@ void down_write(struct rw_semaphore *sem) might_sleep(); rwsem_acquire(&sem->dep_map, 0, 0, _RET_IP_); - __down_write(sem); + LOCK_CONTENDED(sem, __down_write_trylock, __down_write); } EXPORT_SYMBOL(down_write); @@ -111,7 +111,7 @@ void down_read_nested(struct rw_semaphore *sem, int subclass) might_sleep(); rwsem_acquire_read(&sem->dep_map, subclass, 0, _RET_IP_); - __down_read(sem); + LOCK_CONTENDED(sem, __down_read_trylock, __down_read); } EXPORT_SYMBOL(down_read_nested); @@ -130,7 +130,7 @@ void down_write_nested(struct rw_semaphore *sem, int subclass) might_sleep(); rwsem_acquire(&sem->dep_map, subclass, 0, _RET_IP_); - __down_write_nested(sem, subclass); + LOCK_CONTENDED(sem, __down_write_trylock, __down_write); } EXPORT_SYMBOL(down_write_nested); diff --git a/kernel/sched.c b/kernel/sched.c index cb31fb4a137..93cf241cfbe 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -301,7 +301,7 @@ struct rq { struct lock_class_key rq_lock_key; }; -static DEFINE_PER_CPU(struct rq, runqueues) ____cacheline_aligned_in_smp; +static DEFINE_PER_CPU_SHARED_ALIGNED(struct rq, runqueues); static DEFINE_MUTEX(sched_hotcpu_mutex); static inline void check_preempt_curr(struct rq *rq, struct task_struct *p) @@ -379,6 +379,23 @@ static inline unsigned long long rq_clock(struct rq *rq) #define task_rq(p) cpu_rq(task_cpu(p)) #define cpu_curr(cpu) (cpu_rq(cpu)->curr) +/* + * For kernel-internal use: high-speed (but slightly incorrect) per-cpu + * clock constructed from sched_clock(): + */ +unsigned long long cpu_clock(int cpu) +{ + struct rq *rq = cpu_rq(cpu); + unsigned long long now; + unsigned long flags; + + spin_lock_irqsave(&rq->lock, flags); + now = rq_clock(rq); + spin_unlock_irqrestore(&rq->lock, flags); + + return now; +} + #ifdef CONFIG_FAIR_GROUP_SCHED /* Change a task's ->cfs_rq if it moves across CPUs */ static inline void set_task_cfs_rq(struct task_struct *p) @@ -2235,7 +2252,7 @@ find_busiest_group(struct sched_domain *sd, int this_cpu, rq = cpu_rq(i); - if (*sd_idle && !idle_cpu(i)) + if (*sd_idle && rq->nr_running) *sd_idle = 0; /* Bias balancing toward cpus of our domain */ @@ -2257,9 +2274,11 @@ find_busiest_group(struct sched_domain *sd, int this_cpu, /* * First idle cpu or the first cpu(busiest) in this sched group * is eligible for doing load balancing at this and above - * domains. + * domains. In the newly idle case, we will allow all the cpu's + * to do the newly idle load balance. */ - if (local_group && balance_cpu != this_cpu && balance) { + if (idle != CPU_NEWLY_IDLE && local_group && + balance_cpu != this_cpu && balance) { *balance = 0; goto ret; } @@ -2677,6 +2696,7 @@ load_balance_newidle(int this_cpu, struct rq *this_rq, struct sched_domain *sd) unsigned long imbalance; int nr_moved = 0; int sd_idle = 0; + int all_pinned = 0; cpumask_t cpus = CPU_MASK_ALL; /* @@ -2715,10 +2735,11 @@ redo: double_lock_balance(this_rq, busiest); nr_moved = move_tasks(this_rq, this_cpu, busiest, minus_1_or_zero(busiest->nr_running), - imbalance, sd, CPU_NEWLY_IDLE, NULL); + imbalance, sd, CPU_NEWLY_IDLE, + &all_pinned); spin_unlock(&busiest->lock); - if (!nr_moved) { + if (unlikely(all_pinned)) { cpu_clear(cpu_of(busiest), cpus); if (!cpus_empty(cpus)) goto redo; diff --git a/kernel/spinlock.c b/kernel/spinlock.c index 2c6c2bf8551..cd72424c266 100644 --- a/kernel/spinlock.c +++ b/kernel/spinlock.c @@ -72,7 +72,7 @@ void __lockfunc _read_lock(rwlock_t *lock) { preempt_disable(); rwlock_acquire_read(&lock->dep_map, 0, 0, _RET_IP_); - _raw_read_lock(lock); + LOCK_CONTENDED(lock, _raw_read_trylock, _raw_read_lock); } EXPORT_SYMBOL(_read_lock); @@ -88,8 +88,8 @@ unsigned long __lockfunc _spin_lock_irqsave(spinlock_t *lock) * _raw_spin_lock_flags() code, because lockdep assumes * that interrupts are not re-enabled during lock-acquire: */ -#ifdef CONFIG_PROVE_LOCKING - _raw_spin_lock(lock); +#ifdef CONFIG_LOCKDEP + LOCK_CONTENDED(lock, _raw_spin_trylock, _raw_spin_lock); #else _raw_spin_lock_flags(lock, &flags); #endif @@ -102,7 +102,7 @@ void __lockfunc _spin_lock_irq(spinlock_t *lock) local_irq_disable(); preempt_disable(); spin_acquire(&lock->dep_map, 0, 0, _RET_IP_); - _raw_spin_lock(lock); + LOCK_CONTENDED(lock, _raw_spin_trylock, _raw_spin_lock); } EXPORT_SYMBOL(_spin_lock_irq); @@ -111,7 +111,7 @@ void __lockfunc _spin_lock_bh(spinlock_t *lock) local_bh_disable(); preempt_disable(); spin_acquire(&lock->dep_map, 0, 0, _RET_IP_); - _raw_spin_lock(lock); + LOCK_CONTENDED(lock, _raw_spin_trylock, _raw_spin_lock); } EXPORT_SYMBOL(_spin_lock_bh); @@ -122,7 +122,7 @@ unsigned long __lockfunc _read_lock_irqsave(rwlock_t *lock) local_irq_save(flags); preempt_disable(); rwlock_acquire_read(&lock->dep_map, 0, 0, _RET_IP_); - _raw_read_lock(lock); + LOCK_CONTENDED(lock, _raw_read_trylock, _raw_read_lock); return flags; } EXPORT_SYMBOL(_read_lock_irqsave); @@ -132,7 +132,7 @@ void __lockfunc _read_lock_irq(rwlock_t *lock) local_irq_disable(); preempt_disable(); rwlock_acquire_read(&lock->dep_map, 0, 0, _RET_IP_); - _raw_read_lock(lock); + LOCK_CONTENDED(lock, _raw_read_trylock, _raw_read_lock); } EXPORT_SYMBOL(_read_lock_irq); @@ -141,7 +141,7 @@ void __lockfunc _read_lock_bh(rwlock_t *lock) local_bh_disable(); preempt_disable(); rwlock_acquire_read(&lock->dep_map, 0, 0, _RET_IP_); - _raw_read_lock(lock); + LOCK_CONTENDED(lock, _raw_read_trylock, _raw_read_lock); } EXPORT_SYMBOL(_read_lock_bh); @@ -152,7 +152,7 @@ unsigned long __lockfunc _write_lock_irqsave(rwlock_t *lock) local_irq_save(flags); preempt_disable(); rwlock_acquire(&lock->dep_map, 0, 0, _RET_IP_); - _raw_write_lock(lock); + LOCK_CONTENDED(lock, _raw_write_trylock, _raw_write_lock); return flags; } EXPORT_SYMBOL(_write_lock_irqsave); @@ -162,7 +162,7 @@ void __lockfunc _write_lock_irq(rwlock_t *lock) local_irq_disable(); preempt_disable(); rwlock_acquire(&lock->dep_map, 0, 0, _RET_IP_); - _raw_write_lock(lock); + LOCK_CONTENDED(lock, _raw_write_trylock, _raw_write_lock); } EXPORT_SYMBOL(_write_lock_irq); @@ -171,7 +171,7 @@ void __lockfunc _write_lock_bh(rwlock_t *lock) local_bh_disable(); preempt_disable(); rwlock_acquire(&lock->dep_map, 0, 0, _RET_IP_); - _raw_write_lock(lock); + LOCK_CONTENDED(lock, _raw_write_trylock, _raw_write_lock); } EXPORT_SYMBOL(_write_lock_bh); @@ -179,7 +179,7 @@ void __lockfunc _spin_lock(spinlock_t *lock) { preempt_disable(); spin_acquire(&lock->dep_map, 0, 0, _RET_IP_); - _raw_spin_lock(lock); + LOCK_CONTENDED(lock, _raw_spin_trylock, _raw_spin_lock); } EXPORT_SYMBOL(_spin_lock); @@ -188,7 +188,7 @@ void __lockfunc _write_lock(rwlock_t *lock) { preempt_disable(); rwlock_acquire(&lock->dep_map, 0, 0, _RET_IP_); - _raw_write_lock(lock); + LOCK_CONTENDED(lock, _raw_write_trylock, _raw_write_lock); } EXPORT_SYMBOL(_write_lock); @@ -289,7 +289,7 @@ void __lockfunc _spin_lock_nested(spinlock_t *lock, int subclass) { preempt_disable(); spin_acquire(&lock->dep_map, subclass, 0, _RET_IP_); - _raw_spin_lock(lock); + LOCK_CONTENDED(lock, _raw_spin_trylock, _raw_spin_lock); } EXPORT_SYMBOL(_spin_lock_nested); @@ -305,8 +305,8 @@ unsigned long __lockfunc _spin_lock_irqsave_nested(spinlock_t *lock, int subclas * _raw_spin_lock_flags() code, because lockdep assumes * that interrupts are not re-enabled during lock-acquire: */ -#ifdef CONFIG_PROVE_SPIN_LOCKING - _raw_spin_lock(lock); +#ifdef CONFIG_LOCKDEP + LOCK_CONTENDED(lock, _raw_spin_trylock, _raw_spin_lock); #else _raw_spin_lock_flags(lock, &flags); #endif diff --git a/kernel/sys.c b/kernel/sys.c index 18987c7f6ad..08562f41976 100644 --- a/kernel/sys.c +++ b/kernel/sys.c @@ -100,6 +100,13 @@ struct pid *cad_pid; EXPORT_SYMBOL(cad_pid); /* + * If set, this is used for preparing the system to power off. + */ + +void (*pm_power_off_prepare)(void); +EXPORT_SYMBOL(pm_power_off_prepare); + +/* * Notifier list for kernel code which wants to be called * at shutdown. This is used to stop any idling DMA operations * and the like. @@ -867,6 +874,8 @@ EXPORT_SYMBOL_GPL(kernel_halt); void kernel_power_off(void) { kernel_shutdown_prepare(SYSTEM_POWER_OFF); + if (pm_power_off_prepare) + pm_power_off_prepare(); printk(KERN_EMERG "Power down.\n"); machine_power_off(); } @@ -1027,7 +1036,7 @@ asmlinkage long sys_setregid(gid_t rgid, gid_t egid) return -EPERM; } if (new_egid != old_egid) { - current->mm->dumpable = suid_dumpable; + set_dumpable(current->mm, suid_dumpable); smp_wmb(); } if (rgid != (gid_t) -1 || @@ -1057,13 +1066,13 @@ asmlinkage long sys_setgid(gid_t gid) if (capable(CAP_SETGID)) { if (old_egid != gid) { - current->mm->dumpable = suid_dumpable; + set_dumpable(current->mm, suid_dumpable); smp_wmb(); } current->gid = current->egid = current->sgid = current->fsgid = gid; } else if ((gid == current->gid) || (gid == current->sgid)) { if (old_egid != gid) { - current->mm->dumpable = suid_dumpable; + set_dumpable(current->mm, suid_dumpable); smp_wmb(); } current->egid = current->fsgid = gid; @@ -1094,7 +1103,7 @@ static int set_user(uid_t new_ruid, int dumpclear) switch_uid(new_user); if (dumpclear) { - current->mm->dumpable = suid_dumpable; + set_dumpable(current->mm, suid_dumpable); smp_wmb(); } current->uid = new_ruid; @@ -1150,7 +1159,7 @@ asmlinkage long sys_setreuid(uid_t ruid, uid_t euid) return -EAGAIN; if (new_euid != old_euid) { - current->mm->dumpable = suid_dumpable; + set_dumpable(current->mm, suid_dumpable); smp_wmb(); } current->fsuid = current->euid = new_euid; @@ -1200,7 +1209,7 @@ asmlinkage long sys_setuid(uid_t uid) return -EPERM; if (old_euid != uid) { - current->mm->dumpable = suid_dumpable; + set_dumpable(current->mm, suid_dumpable); smp_wmb(); } current->fsuid = current->euid = uid; @@ -1245,7 +1254,7 @@ asmlinkage long sys_setresuid(uid_t ruid, uid_t euid, uid_t suid) } if (euid != (uid_t) -1) { if (euid != current->euid) { - current->mm->dumpable = suid_dumpable; + set_dumpable(current->mm, suid_dumpable); smp_wmb(); } current->euid = euid; @@ -1295,7 +1304,7 @@ asmlinkage long sys_setresgid(gid_t rgid, gid_t egid, gid_t sgid) } if (egid != (gid_t) -1) { if (egid != current->egid) { - current->mm->dumpable = suid_dumpable; + set_dumpable(current->mm, suid_dumpable); smp_wmb(); } current->egid = egid; @@ -1341,7 +1350,7 @@ asmlinkage long sys_setfsuid(uid_t uid) uid == current->suid || uid == current->fsuid || capable(CAP_SETUID)) { if (uid != old_fsuid) { - current->mm->dumpable = suid_dumpable; + set_dumpable(current->mm, suid_dumpable); smp_wmb(); } current->fsuid = uid; @@ -1370,7 +1379,7 @@ asmlinkage long sys_setfsgid(gid_t gid) gid == current->sgid || gid == current->fsgid || capable(CAP_SETGID)) { if (gid != old_fsgid) { - current->mm->dumpable = suid_dumpable; + set_dumpable(current->mm, suid_dumpable); smp_wmb(); } current->fsgid = gid; @@ -2167,14 +2176,14 @@ asmlinkage long sys_prctl(int option, unsigned long arg2, unsigned long arg3, error = put_user(current->pdeath_signal, (int __user *)arg2); break; case PR_GET_DUMPABLE: - error = current->mm->dumpable; + error = get_dumpable(current->mm); break; case PR_SET_DUMPABLE: if (arg2 < 0 || arg2 > 1) { error = -EINVAL; break; } - current->mm->dumpable = arg2; + set_dumpable(current->mm, arg2); break; case PR_SET_UNALIGN: diff --git a/kernel/sysctl.c b/kernel/sysctl.c index 44a1d699aad..222299844ad 100644 --- a/kernel/sysctl.c +++ b/kernel/sysctl.c @@ -78,6 +78,7 @@ extern int percpu_pagelist_fraction; extern int compat_log; extern int maps_protect; extern int sysctl_stat_interval; +extern int audit_argv_kb; /* this is needed for the proc_dointvec_minmax for [fs_]overflow UID and GID */ static int maxolduid = 65535; @@ -160,6 +161,8 @@ extern ctl_table inotify_table[]; int sysctl_legacy_va_layout; #endif +extern int prove_locking; +extern int lock_stat; /* The default sysctl tables: */ @@ -281,6 +284,26 @@ static ctl_table kern_table[] = { .mode = 0644, .proc_handler = &proc_dointvec, }, +#ifdef CONFIG_PROVE_LOCKING + { + .ctl_name = CTL_UNNUMBERED, + .procname = "prove_locking", + .data = &prove_locking, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = &proc_dointvec, + }, +#endif +#ifdef CONFIG_LOCK_STAT + { + .ctl_name = CTL_UNNUMBERED, + .procname = "lock_stat", + .data = &lock_stat, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = &proc_dointvec, + }, +#endif { .ctl_name = CTL_UNNUMBERED, .procname = "sched_features", @@ -306,6 +329,16 @@ static ctl_table kern_table[] = { .mode = 0644, .proc_handler = &proc_dointvec, }, +#ifdef CONFIG_AUDITSYSCALL + { + .ctl_name = CTL_UNNUMBERED, + .procname = "audit_argv_kb", + .data = &audit_argv_kb, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = &proc_dointvec, + }, +#endif { .ctl_name = KERN_CORE_PATTERN, .procname = "core_pattern", @@ -660,7 +693,7 @@ static ctl_table kern_table[] = { { .ctl_name = KERN_ACPI_VIDEO_FLAGS, .procname = "acpi_video_flags", - .data = &acpi_video_flags, + .data = &acpi_realmode_flags, .maxlen = sizeof (unsigned long), .mode = 0644, .proc_handler = &proc_doulongvec_minmax, @@ -715,13 +748,17 @@ static ctl_table kern_table[] = { .proc_handler = &proc_dostring, .strategy = &sysctl_string, }, - +/* + * NOTE: do not add new entries to this table unless you have read + * Documentation/sysctl/ctl_unnumbered.txt + */ { .ctl_name = 0 } }; /* Constants for minimum and maximum testing in vm_table. We use these as one-element integer vectors. */ static int zero; +static int two = 2; static int one_hundred = 100; @@ -1112,7 +1149,10 @@ static ctl_table fs_table[] = { .data = &lease_break_time, .maxlen = sizeof(int), .mode = 0644, - .proc_handler = &proc_dointvec, + .proc_handler = &proc_dointvec_minmax, + .strategy = &sysctl_intvec, + .extra1 = &zero, + .extra2 = &two, }, { .ctl_name = FS_AIO_NR, diff --git a/kernel/time/timekeeping.c b/kernel/time/timekeeping.c index 728cedfd3cb..89698776613 100644 --- a/kernel/time/timekeeping.c +++ b/kernel/time/timekeeping.c @@ -401,7 +401,7 @@ static __always_inline int clocksource_bigadjust(s64 error, s64 *interval, * this is optimized for the most common adjustments of -1,0,1, * for other values we can do a bit more work. */ -static void clocksource_adjust(struct clocksource *clock, s64 offset) +static void clocksource_adjust(s64 offset) { s64 error, interval = clock->cycle_interval; int adj; @@ -476,7 +476,7 @@ void update_wall_time(void) } /* correct the clock when NTP error is too big */ - clocksource_adjust(clock, offset); + clocksource_adjust(offset); /* store full nanoseconds into xtime */ xtime.tv_nsec = (s64)clock->xtime_nsec >> clock->shift; diff --git a/kernel/timer.c b/kernel/timer.c index b7792fb0338..d1e8b975c7a 100644 --- a/kernel/timer.c +++ b/kernel/timer.c @@ -103,14 +103,14 @@ static inline tvec_base_t *tbase_get_base(tvec_base_t *base) static inline void timer_set_deferrable(struct timer_list *timer) { timer->base = ((tvec_base_t *)((unsigned long)(timer->base) | - TBASE_DEFERRABLE_FLAG)); + TBASE_DEFERRABLE_FLAG)); } static inline void timer_set_base(struct timer_list *timer, tvec_base_t *new_base) { timer->base = (tvec_base_t *)((unsigned long)(new_base) | - tbase_get_deferrable(timer->base)); + tbase_get_deferrable(timer->base)); } /** @@ -445,10 +445,10 @@ EXPORT_SYMBOL(__mod_timer); void add_timer_on(struct timer_list *timer, int cpu) { tvec_base_t *base = per_cpu(tvec_bases, cpu); - unsigned long flags; + unsigned long flags; timer_stats_timer_set_start_info(timer); - BUG_ON(timer_pending(timer) || !timer->function); + BUG_ON(timer_pending(timer) || !timer->function); spin_lock_irqsave(&base->lock, flags); timer_set_base(timer, base); internal_add_timer(base, timer); @@ -627,7 +627,7 @@ static inline void __run_timers(tvec_base_t *base) while (time_after_eq(jiffies, base->timer_jiffies)) { struct list_head work_list; struct list_head *head = &work_list; - int index = base->timer_jiffies & TVR_MASK; + int index = base->timer_jiffies & TVR_MASK; /* * Cascade timers: @@ -644,8 +644,8 @@ static inline void __run_timers(tvec_base_t *base) unsigned long data; timer = list_first_entry(head, struct timer_list,entry); - fn = timer->function; - data = timer->data; + fn = timer->function; + data = timer->data; timer_stats_account_timer(timer); @@ -689,8 +689,8 @@ static unsigned long __next_timer_interrupt(tvec_base_t *base) index = slot = timer_jiffies & TVR_MASK; do { list_for_each_entry(nte, base->tv1.vec + slot, entry) { - if (tbase_get_deferrable(nte->base)) - continue; + if (tbase_get_deferrable(nte->base)) + continue; found = 1; expires = nte->expires; @@ -834,7 +834,7 @@ void update_process_times(int user_tick) if (rcu_pending(cpu)) rcu_check_callbacks(cpu, user_tick); scheduler_tick(); - run_posix_cpu_timers(p); + run_posix_cpu_timers(p); } /* @@ -909,7 +909,7 @@ static inline void update_times(unsigned long ticks) update_wall_time(); calc_load(ticks); } - + /* * The 64-bit jiffies value is not atomic - you MUST NOT read it * without sampling the sequence number in xtime_lock. @@ -1105,7 +1105,7 @@ asmlinkage long sys_gettid(void) /** * do_sysinfo - fill in sysinfo struct * @info: pointer to buffer to fill - */ + */ int do_sysinfo(struct sysinfo *info) { unsigned long mem_total, sav_total; diff --git a/lib/Kconfig.debug b/lib/Kconfig.debug index 640844024ff..f3e0c2abcbd 100644 --- a/lib/Kconfig.debug +++ b/lib/Kconfig.debug @@ -283,6 +283,17 @@ config LOCKDEP select KALLSYMS select KALLSYMS_ALL +config LOCK_STAT + bool "Lock usage statisitics" + depends on DEBUG_KERNEL && TRACE_IRQFLAGS_SUPPORT && STACKTRACE_SUPPORT && LOCKDEP_SUPPORT + select LOCKDEP + select DEBUG_SPINLOCK + select DEBUG_MUTEXES + select DEBUG_LOCK_ALLOC + default n + help + This feature enables tracking lock contention points + config DEBUG_LOCKDEP bool "Lock dependency engine debugging" depends on DEBUG_KERNEL && LOCKDEP diff --git a/mm/filemap.c b/mm/filemap.c index 5d5449f3d41..49a6fe375d0 100644 --- a/mm/filemap.c +++ b/mm/filemap.c @@ -891,15 +891,20 @@ void do_generic_mapping_read(struct address_space *mapping, unsigned long nr, ret; cond_resched(); - if (index == next_index) - next_index = page_cache_readahead(mapping, &ra, filp, - index, last_index - index); - find_page: page = find_get_page(mapping, index); - if (unlikely(page == NULL)) { - handle_ra_miss(mapping, &ra, index); - goto no_cached_page; + if (!page) { + page_cache_sync_readahead(mapping, + &ra, filp, + index, last_index - index); + page = find_get_page(mapping, index); + if (unlikely(page == NULL)) + goto no_cached_page; + } + if (PageReadahead(page)) { + page_cache_async_readahead(mapping, + &ra, filp, page, + index, last_index - index); } if (!PageUptodate(page)) goto page_not_up_to_date; @@ -1051,6 +1056,7 @@ no_cached_page: out: *_ra = ra; + _ra->prev_index = prev_index; *ppos = ((loff_t) index << PAGE_CACHE_SHIFT) + offset; if (cached_page) @@ -1301,62 +1307,62 @@ static int fastcall page_cache_read(struct file * file, unsigned long offset) #define MMAP_LOTSAMISS (100) /** - * filemap_nopage - read in file data for page fault handling - * @area: the applicable vm_area - * @address: target address to read in - * @type: returned with VM_FAULT_{MINOR,MAJOR} if not %NULL + * filemap_fault - read in file data for page fault handling + * @vma: vma in which the fault was taken + * @vmf: struct vm_fault containing details of the fault * - * filemap_nopage() is invoked via the vma operations vector for a + * filemap_fault() is invoked via the vma operations vector for a * mapped memory region to read in file data during a page fault. * * The goto's are kind of ugly, but this streamlines the normal case of having * it in the page cache, and handles the special cases reasonably without * having a lot of duplicated code. */ -struct page *filemap_nopage(struct vm_area_struct *area, - unsigned long address, int *type) +int filemap_fault(struct vm_area_struct *vma, struct vm_fault *vmf) { int error; - struct file *file = area->vm_file; + struct file *file = vma->vm_file; struct address_space *mapping = file->f_mapping; struct file_ra_state *ra = &file->f_ra; struct inode *inode = mapping->host; struct page *page; - unsigned long size, pgoff; - int did_readaround = 0, majmin = VM_FAULT_MINOR; - - pgoff = ((address-area->vm_start) >> PAGE_CACHE_SHIFT) + area->vm_pgoff; + unsigned long size; + int did_readaround = 0; + int ret = 0; -retry_all: size = (i_size_read(inode) + PAGE_CACHE_SIZE - 1) >> PAGE_CACHE_SHIFT; - if (pgoff >= size) + if (vmf->pgoff >= size) goto outside_data_content; /* If we don't want any read-ahead, don't bother */ - if (VM_RandomReadHint(area)) + if (VM_RandomReadHint(vma)) goto no_cached_page; /* - * The readahead code wants to be told about each and every page - * so it can build and shrink its windows appropriately - * - * For sequential accesses, we use the generic readahead logic. - */ - if (VM_SequentialReadHint(area)) - page_cache_readahead(mapping, ra, file, pgoff, 1); - - /* * Do we have something in the page cache already? */ retry_find: - page = find_get_page(mapping, pgoff); + page = find_lock_page(mapping, vmf->pgoff); + /* + * For sequential accesses, we use the generic readahead logic. + */ + if (VM_SequentialReadHint(vma)) { + if (!page) { + page_cache_sync_readahead(mapping, ra, file, + vmf->pgoff, 1); + page = find_lock_page(mapping, vmf->pgoff); + if (!page) + goto no_cached_page; + } + if (PageReadahead(page)) { + page_cache_async_readahead(mapping, ra, file, page, + vmf->pgoff, 1); + } + } + if (!page) { unsigned long ra_pages; - if (VM_SequentialReadHint(area)) { - handle_ra_miss(mapping, ra, pgoff); - goto no_cached_page; - } ra->mmap_miss++; /* @@ -1371,7 +1377,7 @@ retry_find: * check did_readaround, as this is an inner loop. */ if (!did_readaround) { - majmin = VM_FAULT_MAJOR; + ret = VM_FAULT_MAJOR; count_vm_event(PGMAJFAULT); } did_readaround = 1; @@ -1379,11 +1385,11 @@ retry_find: if (ra_pages) { pgoff_t start = 0; - if (pgoff > ra_pages / 2) - start = pgoff - ra_pages / 2; + if (vmf->pgoff > ra_pages / 2) + start = vmf->pgoff - ra_pages / 2; do_page_cache_readahead(mapping, file, start, ra_pages); } - page = find_get_page(mapping, pgoff); + page = find_lock_page(mapping, vmf->pgoff); if (!page) goto no_cached_page; } @@ -1392,35 +1398,42 @@ retry_find: ra->mmap_hit++; /* - * Ok, found a page in the page cache, now we need to check - * that it's up-to-date. + * We have a locked page in the page cache, now we need to check + * that it's up-to-date. If not, it is going to be due to an error. */ - if (!PageUptodate(page)) + if (unlikely(!PageUptodate(page))) goto page_not_uptodate; -success: + /* Must recheck i_size under page lock */ + size = (i_size_read(inode) + PAGE_CACHE_SIZE - 1) >> PAGE_CACHE_SHIFT; + if (unlikely(vmf->pgoff >= size)) { + unlock_page(page); + goto outside_data_content; + } + /* * Found the page and have a reference on it. */ mark_page_accessed(page); - if (type) - *type = majmin; - return page; + ra->prev_index = page->index; + vmf->page = page; + return ret | VM_FAULT_LOCKED; outside_data_content: /* * An external ptracer can access pages that normally aren't * accessible.. */ - if (area->vm_mm == current->mm) - return NOPAGE_SIGBUS; + if (vma->vm_mm == current->mm) + return VM_FAULT_SIGBUS; + /* Fall through to the non-read-ahead case */ no_cached_page: /* * We're only likely to ever get here if MADV_RANDOM is in * effect. */ - error = page_cache_read(file, pgoff); + error = page_cache_read(file, vmf->pgoff); /* * The page we want has now been added to the page cache. @@ -1436,12 +1449,13 @@ no_cached_page: * to schedule I/O. */ if (error == -ENOMEM) - return NOPAGE_OOM; - return NOPAGE_SIGBUS; + return VM_FAULT_OOM; + return VM_FAULT_SIGBUS; page_not_uptodate: + /* IO error path */ if (!did_readaround) { - majmin = VM_FAULT_MAJOR; + ret = VM_FAULT_MAJOR; count_vm_event(PGMAJFAULT); } @@ -1451,217 +1465,21 @@ page_not_uptodate: * because there really aren't any performance issues here * and we need to check for errors. */ - lock_page(page); - - /* Somebody truncated the page on us? */ - if (!page->mapping) { - unlock_page(page); - page_cache_release(page); - goto retry_all; - } - - /* Somebody else successfully read it in? */ - if (PageUptodate(page)) { - unlock_page(page); - goto success; - } ClearPageError(page); error = mapping->a_ops->readpage(file, page); - if (!error) { - wait_on_page_locked(page); - if (PageUptodate(page)) - goto success; - } else if (error == AOP_TRUNCATED_PAGE) { - page_cache_release(page); - goto retry_find; - } - - /* - * Things didn't work out. Return zero to tell the - * mm layer so, possibly freeing the page cache page first. - */ - shrink_readahead_size_eio(file, ra); page_cache_release(page); - return NOPAGE_SIGBUS; -} -EXPORT_SYMBOL(filemap_nopage); - -static struct page * filemap_getpage(struct file *file, unsigned long pgoff, - int nonblock) -{ - struct address_space *mapping = file->f_mapping; - struct page *page; - int error; - /* - * Do we have something in the page cache already? - */ -retry_find: - page = find_get_page(mapping, pgoff); - if (!page) { - if (nonblock) - return NULL; - goto no_cached_page; - } - - /* - * Ok, found a page in the page cache, now we need to check - * that it's up-to-date. - */ - if (!PageUptodate(page)) { - if (nonblock) { - page_cache_release(page); - return NULL; - } - goto page_not_uptodate; - } - -success: - /* - * Found the page and have a reference on it. - */ - mark_page_accessed(page); - return page; - -no_cached_page: - error = page_cache_read(file, pgoff); - - /* - * The page we want has now been added to the page cache. - * In the unlikely event that someone removed it in the - * meantime, we'll just come back here and read it again. - */ - if (error >= 0) + if (!error || error == AOP_TRUNCATED_PAGE) goto retry_find; - /* - * An error return from page_cache_read can result if the - * system is low on memory, or a problem occurs while trying - * to schedule I/O. - */ - return NULL; - -page_not_uptodate: - lock_page(page); - - /* Did it get truncated while we waited for it? */ - if (!page->mapping) { - unlock_page(page); - goto err; - } - - /* Did somebody else get it up-to-date? */ - if (PageUptodate(page)) { - unlock_page(page); - goto success; - } - - error = mapping->a_ops->readpage(file, page); - if (!error) { - wait_on_page_locked(page); - if (PageUptodate(page)) - goto success; - } else if (error == AOP_TRUNCATED_PAGE) { - page_cache_release(page); - goto retry_find; - } - - /* - * Umm, take care of errors if the page isn't up-to-date. - * Try to re-read it _once_. We do this synchronously, - * because there really aren't any performance issues here - * and we need to check for errors. - */ - lock_page(page); - - /* Somebody truncated the page on us? */ - if (!page->mapping) { - unlock_page(page); - goto err; - } - /* Somebody else successfully read it in? */ - if (PageUptodate(page)) { - unlock_page(page); - goto success; - } - - ClearPageError(page); - error = mapping->a_ops->readpage(file, page); - if (!error) { - wait_on_page_locked(page); - if (PageUptodate(page)) - goto success; - } else if (error == AOP_TRUNCATED_PAGE) { - page_cache_release(page); - goto retry_find; - } - - /* - * Things didn't work out. Return zero to tell the - * mm layer so, possibly freeing the page cache page first. - */ -err: - page_cache_release(page); - - return NULL; -} - -int filemap_populate(struct vm_area_struct *vma, unsigned long addr, - unsigned long len, pgprot_t prot, unsigned long pgoff, - int nonblock) -{ - struct file *file = vma->vm_file; - struct address_space *mapping = file->f_mapping; - struct inode *inode = mapping->host; - unsigned long size; - struct mm_struct *mm = vma->vm_mm; - struct page *page; - int err; - - if (!nonblock) - force_page_cache_readahead(mapping, vma->vm_file, - pgoff, len >> PAGE_CACHE_SHIFT); - -repeat: - size = (i_size_read(inode) + PAGE_CACHE_SIZE - 1) >> PAGE_CACHE_SHIFT; - if (pgoff + (len >> PAGE_CACHE_SHIFT) > size) - return -EINVAL; - - page = filemap_getpage(file, pgoff, nonblock); - - /* XXX: This is wrong, a filesystem I/O error may have happened. Fix that as - * done in shmem_populate calling shmem_getpage */ - if (!page && !nonblock) - return -ENOMEM; - - if (page) { - err = install_page(mm, vma, addr, page, prot); - if (err) { - page_cache_release(page); - return err; - } - } else if (vma->vm_flags & VM_NONLINEAR) { - /* No page was found just because we can't read it in now (being - * here implies nonblock != 0), but the page may exist, so set - * the PTE to fault it in later. */ - err = install_file_pte(mm, vma, addr, pgoff, prot); - if (err) - return err; - } - - len -= PAGE_SIZE; - addr += PAGE_SIZE; - pgoff++; - if (len) - goto repeat; - - return 0; + /* Things didn't work out. Return zero to tell the mm layer so. */ + shrink_readahead_size_eio(file, ra); + return VM_FAULT_SIGBUS; } -EXPORT_SYMBOL(filemap_populate); +EXPORT_SYMBOL(filemap_fault); struct vm_operations_struct generic_file_vm_ops = { - .nopage = filemap_nopage, - .populate = filemap_populate, + .fault = filemap_fault, }; /* This is used for a general mmap of a disk file */ @@ -1674,6 +1492,7 @@ int generic_file_mmap(struct file * file, struct vm_area_struct * vma) return -ENOEXEC; file_accessed(file); vma->vm_ops = &generic_file_vm_ops; + vma->vm_flags |= VM_CAN_NONLINEAR; return 0; } diff --git a/mm/filemap_xip.c b/mm/filemap_xip.c index 65ffc321f0c..53ee6a29963 100644 --- a/mm/filemap_xip.c +++ b/mm/filemap_xip.c @@ -205,62 +205,58 @@ __xip_unmap (struct address_space * mapping, } /* - * xip_nopage() is invoked via the vma operations vector for a + * xip_fault() is invoked via the vma operations vector for a * mapped memory region to read in file data during a page fault. * - * This function is derived from filemap_nopage, but used for execute in place + * This function is derived from filemap_fault, but used for execute in place */ -static struct page * -xip_file_nopage(struct vm_area_struct * area, - unsigned long address, - int *type) +static int xip_file_fault(struct vm_area_struct *area, struct vm_fault *vmf) { struct file *file = area->vm_file; struct address_space *mapping = file->f_mapping; struct inode *inode = mapping->host; struct page *page; - unsigned long size, pgoff, endoff; + pgoff_t size; - pgoff = ((address - area->vm_start) >> PAGE_CACHE_SHIFT) - + area->vm_pgoff; - endoff = ((area->vm_end - area->vm_start) >> PAGE_CACHE_SHIFT) - + area->vm_pgoff; + /* XXX: are VM_FAULT_ codes OK? */ size = (i_size_read(inode) + PAGE_CACHE_SIZE - 1) >> PAGE_CACHE_SHIFT; - if (pgoff >= size) - return NOPAGE_SIGBUS; + if (vmf->pgoff >= size) + return VM_FAULT_SIGBUS; - page = mapping->a_ops->get_xip_page(mapping, pgoff*(PAGE_SIZE/512), 0); + page = mapping->a_ops->get_xip_page(mapping, + vmf->pgoff*(PAGE_SIZE/512), 0); if (!IS_ERR(page)) goto out; if (PTR_ERR(page) != -ENODATA) - return NOPAGE_SIGBUS; + return VM_FAULT_OOM; /* sparse block */ if ((area->vm_flags & (VM_WRITE | VM_MAYWRITE)) && (area->vm_flags & (VM_SHARED| VM_MAYSHARE)) && (!(mapping->host->i_sb->s_flags & MS_RDONLY))) { /* maybe shared writable, allocate new block */ - page = mapping->a_ops->get_xip_page (mapping, - pgoff*(PAGE_SIZE/512), 1); + page = mapping->a_ops->get_xip_page(mapping, + vmf->pgoff*(PAGE_SIZE/512), 1); if (IS_ERR(page)) - return NOPAGE_SIGBUS; + return VM_FAULT_SIGBUS; /* unmap page at pgoff from all other vmas */ - __xip_unmap(mapping, pgoff); + __xip_unmap(mapping, vmf->pgoff); } else { /* not shared and writable, use xip_sparse_page() */ page = xip_sparse_page(); if (!page) - return NOPAGE_OOM; + return VM_FAULT_OOM; } out: page_cache_get(page); - return page; + vmf->page = page; + return 0; } static struct vm_operations_struct xip_file_vm_ops = { - .nopage = xip_file_nopage, + .fault = xip_file_fault, }; int xip_file_mmap(struct file * file, struct vm_area_struct * vma) @@ -269,6 +265,7 @@ int xip_file_mmap(struct file * file, struct vm_area_struct * vma) file_accessed(file); vma->vm_ops = &xip_file_vm_ops; + vma->vm_flags |= VM_CAN_NONLINEAR; return 0; } EXPORT_SYMBOL_GPL(xip_file_mmap); diff --git a/mm/fremap.c b/mm/fremap.c index 4e3f53dd5fd..c395b1abf08 100644 --- a/mm/fremap.c +++ b/mm/fremap.c @@ -20,13 +20,14 @@ #include <asm/cacheflush.h> #include <asm/tlbflush.h> -static int zap_pte(struct mm_struct *mm, struct vm_area_struct *vma, +static void zap_pte(struct mm_struct *mm, struct vm_area_struct *vma, unsigned long addr, pte_t *ptep) { pte_t pte = *ptep; - struct page *page = NULL; if (pte_present(pte)) { + struct page *page; + flush_cache_page(vma, addr, pte_pfn(pte)); pte = ptep_clear_flush(vma, addr, ptep); page = vm_normal_page(vma, addr, pte); @@ -35,68 +36,21 @@ static int zap_pte(struct mm_struct *mm, struct vm_area_struct *vma, set_page_dirty(page); page_remove_rmap(page, vma); page_cache_release(page); + update_hiwater_rss(mm); + dec_mm_counter(mm, file_rss); } } else { if (!pte_file(pte)) free_swap_and_cache(pte_to_swp_entry(pte)); pte_clear_not_present_full(mm, addr, ptep, 0); } - return !!page; } /* - * Install a file page to a given virtual memory address, release any - * previously existing mapping. - */ -int install_page(struct mm_struct *mm, struct vm_area_struct *vma, - unsigned long addr, struct page *page, pgprot_t prot) -{ - struct inode *inode; - pgoff_t size; - int err = -ENOMEM; - pte_t *pte; - pte_t pte_val; - spinlock_t *ptl; - - pte = get_locked_pte(mm, addr, &ptl); - if (!pte) - goto out; - - /* - * This page may have been truncated. Tell the - * caller about it. - */ - err = -EINVAL; - inode = vma->vm_file->f_mapping->host; - size = (i_size_read(inode) + PAGE_CACHE_SIZE - 1) >> PAGE_CACHE_SHIFT; - if (!page->mapping || page->index >= size) - goto unlock; - err = -ENOMEM; - if (page_mapcount(page) > INT_MAX/2) - goto unlock; - - if (pte_none(*pte) || !zap_pte(mm, vma, addr, pte)) - inc_mm_counter(mm, file_rss); - - flush_icache_page(vma, page); - pte_val = mk_pte(page, prot); - set_pte_at(mm, addr, pte, pte_val); - page_add_file_rmap(page); - update_mmu_cache(vma, addr, pte_val); - lazy_mmu_prot_update(pte_val); - err = 0; -unlock: - pte_unmap_unlock(pte, ptl); -out: - return err; -} -EXPORT_SYMBOL(install_page); - -/* * Install a file pte to a given virtual memory address, release any * previously existing mapping. */ -int install_file_pte(struct mm_struct *mm, struct vm_area_struct *vma, +static int install_file_pte(struct mm_struct *mm, struct vm_area_struct *vma, unsigned long addr, unsigned long pgoff, pgprot_t prot) { int err = -ENOMEM; @@ -107,10 +61,8 @@ int install_file_pte(struct mm_struct *mm, struct vm_area_struct *vma, if (!pte) goto out; - if (!pte_none(*pte) && zap_pte(mm, vma, addr, pte)) { - update_hiwater_rss(mm); - dec_mm_counter(mm, file_rss); - } + if (!pte_none(*pte)) + zap_pte(mm, vma, addr, pte); set_pte_at(mm, addr, pte, pgoff_to_pte(pgoff)); /* @@ -126,6 +78,25 @@ out: return err; } +static int populate_range(struct mm_struct *mm, struct vm_area_struct *vma, + unsigned long addr, unsigned long size, pgoff_t pgoff) +{ + int err; + + do { + err = install_file_pte(mm, vma, addr, pgoff, vma->vm_page_prot); + if (err) + return err; + + size -= PAGE_SIZE; + addr += PAGE_SIZE; + pgoff++; + } while (size); + + return 0; + +} + /*** * sys_remap_file_pages - remap arbitrary pages of a shared backing store * file within an existing vma. @@ -183,41 +154,77 @@ asmlinkage long sys_remap_file_pages(unsigned long start, unsigned long size, * the single existing vma. vm_private_data is used as a * swapout cursor in a VM_NONLINEAR vma. */ - if (vma && (vma->vm_flags & VM_SHARED) && - (!vma->vm_private_data || (vma->vm_flags & VM_NONLINEAR)) && - vma->vm_ops && vma->vm_ops->populate && - end > start && start >= vma->vm_start && - end <= vma->vm_end) { - - /* Must set VM_NONLINEAR before any pages are populated. */ - if (pgoff != linear_page_index(vma, start) && - !(vma->vm_flags & VM_NONLINEAR)) { - if (!has_write_lock) { - up_read(&mm->mmap_sem); - down_write(&mm->mmap_sem); - has_write_lock = 1; - goto retry; - } - mapping = vma->vm_file->f_mapping; - spin_lock(&mapping->i_mmap_lock); - flush_dcache_mmap_lock(mapping); - vma->vm_flags |= VM_NONLINEAR; - vma_prio_tree_remove(vma, &mapping->i_mmap); - vma_nonlinear_insert(vma, &mapping->i_mmap_nonlinear); - flush_dcache_mmap_unlock(mapping); - spin_unlock(&mapping->i_mmap_lock); - } + if (!vma || !(vma->vm_flags & VM_SHARED)) + goto out; - err = vma->vm_ops->populate(vma, start, size, - vma->vm_page_prot, - pgoff, flags & MAP_NONBLOCK); + if (vma->vm_private_data && !(vma->vm_flags & VM_NONLINEAR)) + goto out; + + if (!vma->vm_flags & VM_CAN_NONLINEAR) + goto out; + if (end <= start || start < vma->vm_start || end > vma->vm_end) + goto out; + + /* Must set VM_NONLINEAR before any pages are populated. */ + if (!(vma->vm_flags & VM_NONLINEAR)) { + /* Don't need a nonlinear mapping, exit success */ + if (pgoff == linear_page_index(vma, start)) { + err = 0; + goto out; + } + + if (!has_write_lock) { + up_read(&mm->mmap_sem); + down_write(&mm->mmap_sem); + has_write_lock = 1; + goto retry; + } + mapping = vma->vm_file->f_mapping; /* - * We can't clear VM_NONLINEAR because we'd have to do - * it after ->populate completes, and that would prevent - * downgrading the lock. (Locks can't be upgraded). + * page_mkclean doesn't work on nonlinear vmas, so if + * dirty pages need to be accounted, emulate with linear + * vmas. */ + if (mapping_cap_account_dirty(mapping)) { + unsigned long addr; + + flags &= MAP_NONBLOCK; + addr = mmap_region(vma->vm_file, start, size, + flags, vma->vm_flags, pgoff, 1); + if (IS_ERR_VALUE(addr)) { + err = addr; + } else { + BUG_ON(addr != start); + err = 0; + } + goto out; + } + spin_lock(&mapping->i_mmap_lock); + flush_dcache_mmap_lock(mapping); + vma->vm_flags |= VM_NONLINEAR; + vma_prio_tree_remove(vma, &mapping->i_mmap); + vma_nonlinear_insert(vma, &mapping->i_mmap_nonlinear); + flush_dcache_mmap_unlock(mapping); + spin_unlock(&mapping->i_mmap_lock); + } + + err = populate_range(mm, vma, start, size, pgoff); + if (!err && !(flags & MAP_NONBLOCK)) { + if (unlikely(has_write_lock)) { + downgrade_write(&mm->mmap_sem); + has_write_lock = 0; + } + make_pages_present(start, start+size); } + + /* + * We can't clear VM_NONLINEAR because we'd have to do + * it after ->populate completes, and that would prevent + * downgrading the lock. (Locks can't be upgraded). + */ + +out: if (likely(!has_write_lock)) up_read(&mm->mmap_sem); else diff --git a/mm/hugetlb.c b/mm/hugetlb.c index 6912bbf33fa..f127940ec24 100644 --- a/mm/hugetlb.c +++ b/mm/hugetlb.c @@ -78,16 +78,13 @@ static struct page *dequeue_huge_page(struct vm_area_struct *vma, for (z = zonelist->zones; *z; z++) { nid = zone_to_nid(*z); if (cpuset_zone_allowed_softwall(*z, htlb_alloc_mask) && - !list_empty(&hugepage_freelists[nid])) - break; - } - - if (*z) { - page = list_entry(hugepage_freelists[nid].next, - struct page, lru); - list_del(&page->lru); - free_huge_pages--; - free_huge_pages_node[nid]--; + !list_empty(&hugepage_freelists[nid])) { + page = list_entry(hugepage_freelists[nid].next, + struct page, lru); + list_del(&page->lru); + free_huge_pages--; + free_huge_pages_node[nid]--; + } } return page; } @@ -107,15 +104,19 @@ static int alloc_fresh_huge_page(void) { static int prev_nid; struct page *page; - static DEFINE_SPINLOCK(nid_lock); int nid; - spin_lock(&nid_lock); + /* + * Copy static prev_nid to local nid, work on that, then copy it + * back to prev_nid afterwards: otherwise there's a window in which + * a racer might pass invalid nid MAX_NUMNODES to alloc_pages_node. + * But we don't need to use a spin_lock here: it really doesn't + * matter if occasionally a racer chooses the same nid as we do. + */ nid = next_node(prev_nid, node_online_map); if (nid == MAX_NUMNODES) nid = first_node(node_online_map); prev_nid = nid; - spin_unlock(&nid_lock); page = alloc_pages_node(nid, htlb_alloc_mask|__GFP_COMP|__GFP_NOWARN, HUGETLB_PAGE_ORDER); @@ -207,7 +208,7 @@ static void update_and_free_page(struct page *page) 1 << PG_dirty | 1 << PG_active | 1 << PG_reserved | 1 << PG_private | 1<< PG_writeback); } - page[1].lru.next = NULL; + set_compound_page_dtor(page, NULL); set_page_refcounted(page); __free_pages(page, HUGETLB_PAGE_ORDER); } @@ -316,15 +317,14 @@ unsigned long hugetlb_total_pages(void) * hugegpage VMA. do_page_fault() is supposed to trap this, so BUG is we get * this far. */ -static struct page *hugetlb_nopage(struct vm_area_struct *vma, - unsigned long address, int *unused) +static int hugetlb_vm_op_fault(struct vm_area_struct *vma, struct vm_fault *vmf) { BUG(); - return NULL; + return 0; } struct vm_operations_struct hugetlb_vm_ops = { - .nopage = hugetlb_nopage, + .fault = hugetlb_vm_op_fault, }; static pte_t make_huge_pte(struct vm_area_struct *vma, struct page *page, @@ -470,7 +470,7 @@ static int hugetlb_cow(struct mm_struct *mm, struct vm_area_struct *vma, avoidcopy = (page_count(old_page) == 1); if (avoidcopy) { set_huge_ptep_writable(vma, address, ptep); - return VM_FAULT_MINOR; + return 0; } page_cache_get(old_page); @@ -495,7 +495,7 @@ static int hugetlb_cow(struct mm_struct *mm, struct vm_area_struct *vma, } page_cache_release(new_page); page_cache_release(old_page); - return VM_FAULT_MINOR; + return 0; } static int hugetlb_no_page(struct mm_struct *mm, struct vm_area_struct *vma, @@ -552,7 +552,7 @@ retry: if (idx >= size) goto backout; - ret = VM_FAULT_MINOR; + ret = 0; if (!pte_none(*ptep)) goto backout; @@ -603,7 +603,7 @@ int hugetlb_fault(struct mm_struct *mm, struct vm_area_struct *vma, return ret; } - ret = VM_FAULT_MINOR; + ret = 0; spin_lock(&mm->page_table_lock); /* Check for a racing update before calling hugetlb_cow */ @@ -642,7 +642,7 @@ int follow_hugetlb_page(struct mm_struct *mm, struct vm_area_struct *vma, spin_unlock(&mm->page_table_lock); ret = hugetlb_fault(mm, vma, vaddr, 0); spin_lock(&mm->page_table_lock); - if (ret == VM_FAULT_MINOR) + if (!(ret & VM_FAULT_MAJOR)) continue; remainder = 0; diff --git a/mm/memory.c b/mm/memory.c index 9c6ff7fffdc..8aace3db3a5 100644 --- a/mm/memory.c +++ b/mm/memory.c @@ -1047,7 +1047,8 @@ int get_user_pages(struct task_struct *tsk, struct mm_struct *mm, if (pages) foll_flags |= FOLL_GET; if (!write && !(vma->vm_flags & VM_LOCKED) && - (!vma->vm_ops || !vma->vm_ops->nopage)) + (!vma->vm_ops || (!vma->vm_ops->nopage && + !vma->vm_ops->fault))) foll_flags |= FOLL_ANON; do { @@ -1067,31 +1068,30 @@ int get_user_pages(struct task_struct *tsk, struct mm_struct *mm, cond_resched(); while (!(page = follow_page(vma, start, foll_flags))) { int ret; - ret = __handle_mm_fault(mm, vma, start, + ret = handle_mm_fault(mm, vma, start, foll_flags & FOLL_WRITE); + if (ret & VM_FAULT_ERROR) { + if (ret & VM_FAULT_OOM) + return i ? i : -ENOMEM; + else if (ret & VM_FAULT_SIGBUS) + return i ? i : -EFAULT; + BUG(); + } + if (ret & VM_FAULT_MAJOR) + tsk->maj_flt++; + else + tsk->min_flt++; + /* - * The VM_FAULT_WRITE bit tells us that do_wp_page has - * broken COW when necessary, even if maybe_mkwrite - * decided not to set pte_write. We can thus safely do - * subsequent page lookups as if they were reads. + * The VM_FAULT_WRITE bit tells us that + * do_wp_page has broken COW when necessary, + * even if maybe_mkwrite decided not to set + * pte_write. We can thus safely do subsequent + * page lookups as if they were reads. */ if (ret & VM_FAULT_WRITE) foll_flags &= ~FOLL_WRITE; - - switch (ret & ~VM_FAULT_WRITE) { - case VM_FAULT_MINOR: - tsk->min_flt++; - break; - case VM_FAULT_MAJOR: - tsk->maj_flt++; - break; - case VM_FAULT_SIGBUS: - return i ? i : -EFAULT; - case VM_FAULT_OOM: - return i ? i : -ENOMEM; - default: - BUG(); - } + cond_resched(); } if (pages) { @@ -1638,7 +1638,7 @@ static int do_wp_page(struct mm_struct *mm, struct vm_area_struct *vma, { struct page *old_page, *new_page; pte_t entry; - int reuse = 0, ret = VM_FAULT_MINOR; + int reuse = 0, ret = 0; struct page *dirty_page = NULL; old_page = vm_normal_page(vma, address, orig_pte); @@ -1765,6 +1765,15 @@ gotten: unlock: pte_unmap_unlock(page_table, ptl); if (dirty_page) { + /* + * Yes, Virginia, this is actually required to prevent a race + * with clear_page_dirty_for_io() from clearing the page dirty + * bit after it clear all dirty ptes, but before a racing + * do_wp_page installs a dirty pte. + * + * do_no_page is protected similarly. + */ + wait_on_page_locked(dirty_page); set_page_dirty_balance(dirty_page); put_page(dirty_page); } @@ -1831,6 +1840,13 @@ static int unmap_mapping_range_vma(struct vm_area_struct *vma, unsigned long restart_addr; int need_break; + /* + * files that support invalidating or truncating portions of the + * file from under mmaped areas must have their ->fault function + * return a locked page (and set VM_FAULT_LOCKED in the return). + * This provides synchronisation against concurrent unmapping here. + */ + again: restart_addr = vma->vm_truncate_count; if (is_restart_addr(restart_addr) && start_addr < restart_addr) { @@ -1959,17 +1975,8 @@ void unmap_mapping_range(struct address_space *mapping, spin_lock(&mapping->i_mmap_lock); - /* serialize i_size write against truncate_count write */ - smp_wmb(); - /* Protect against page faults, and endless unmapping loops */ + /* Protect against endless unmapping loops */ mapping->truncate_count++; - /* - * For archs where spin_lock has inclusive semantics like ia64 - * this smp_mb() will prevent to read pagetable contents - * before the truncate_count increment is visible to - * other cpus. - */ - smp_mb(); if (unlikely(is_restart_addr(mapping->truncate_count))) { if (mapping->truncate_count == 0) reset_vma_truncate_counts(mapping); @@ -2008,8 +2015,18 @@ int vmtruncate(struct inode * inode, loff_t offset) if (IS_SWAPFILE(inode)) goto out_busy; i_size_write(inode, offset); + + /* + * unmap_mapping_range is called twice, first simply for efficiency + * so that truncate_inode_pages does fewer single-page unmaps. However + * after this first call, and before truncate_inode_pages finishes, + * it is possible for private pages to be COWed, which remain after + * truncate_inode_pages finishes, hence the second unmap_mapping_range + * call must be made for correctness. + */ unmap_mapping_range(mapping, offset + PAGE_SIZE - 1, 0, 1); truncate_inode_pages(mapping, offset); + unmap_mapping_range(mapping, offset + PAGE_SIZE - 1, 0, 1); goto out_truncate; do_expand: @@ -2049,6 +2066,7 @@ int vmtruncate_range(struct inode *inode, loff_t offset, loff_t end) down_write(&inode->i_alloc_sem); unmap_mapping_range(mapping, offset, (end - offset), 1); truncate_inode_pages_range(mapping, offset, end); + unmap_mapping_range(mapping, offset, (end - offset), 1); inode->i_op->truncate_range(inode, offset, end); up_write(&inode->i_alloc_sem); mutex_unlock(&inode->i_mutex); @@ -2130,7 +2148,7 @@ static int do_swap_page(struct mm_struct *mm, struct vm_area_struct *vma, struct page *page; swp_entry_t entry; pte_t pte; - int ret = VM_FAULT_MINOR; + int ret = 0; if (!pte_unmap_same(mm, pmd, page_table, orig_pte)) goto out; @@ -2198,15 +2216,15 @@ static int do_swap_page(struct mm_struct *mm, struct vm_area_struct *vma, unlock_page(page); if (write_access) { + /* XXX: We could OR the do_wp_page code with this one? */ if (do_wp_page(mm, vma, address, - page_table, pmd, ptl, pte) == VM_FAULT_OOM) + page_table, pmd, ptl, pte) & VM_FAULT_OOM) ret = VM_FAULT_OOM; goto out; } /* No need to invalidate - it was non-present before */ update_mmu_cache(vma, address, pte); - lazy_mmu_prot_update(pte); unlock: pte_unmap_unlock(page_table, ptl); out: @@ -2271,7 +2289,7 @@ static int do_anonymous_page(struct mm_struct *mm, struct vm_area_struct *vma, lazy_mmu_prot_update(entry); unlock: pte_unmap_unlock(page_table, ptl); - return VM_FAULT_MINOR; + return 0; release: page_cache_release(page); goto unlock; @@ -2280,10 +2298,10 @@ oom: } /* - * do_no_page() tries to create a new page mapping. It aggressively + * __do_fault() tries to create a new page mapping. It aggressively * tries to share with existing pages, but makes a separate copy if - * the "write_access" parameter is true in order to avoid the next - * page fault. + * the FAULT_FLAG_WRITE is set in the flags parameter in order to avoid + * the next page fault. * * As this is called only for pages that do not currently exist, we * do not need to flush old virtual caches or the TLB. @@ -2292,90 +2310,100 @@ oom: * but allow concurrent faults), and pte mapped but not yet locked. * We return with mmap_sem still held, but pte unmapped and unlocked. */ -static int do_no_page(struct mm_struct *mm, struct vm_area_struct *vma, +static int __do_fault(struct mm_struct *mm, struct vm_area_struct *vma, unsigned long address, pte_t *page_table, pmd_t *pmd, - int write_access) + pgoff_t pgoff, unsigned int flags, pte_t orig_pte) { spinlock_t *ptl; - struct page *new_page; - struct address_space *mapping = NULL; + struct page *page; pte_t entry; - unsigned int sequence = 0; - int ret = VM_FAULT_MINOR; int anon = 0; struct page *dirty_page = NULL; + struct vm_fault vmf; + int ret; + + vmf.virtual_address = (void __user *)(address & PAGE_MASK); + vmf.pgoff = pgoff; + vmf.flags = flags; + vmf.page = NULL; pte_unmap(page_table); BUG_ON(vma->vm_flags & VM_PFNMAP); - if (vma->vm_file) { - mapping = vma->vm_file->f_mapping; - sequence = mapping->truncate_count; - smp_rmb(); /* serializes i_size against truncate_count */ + if (likely(vma->vm_ops->fault)) { + ret = vma->vm_ops->fault(vma, &vmf); + if (unlikely(ret & (VM_FAULT_ERROR | VM_FAULT_NOPAGE))) + return ret; + } else { + /* Legacy ->nopage path */ + ret = 0; + vmf.page = vma->vm_ops->nopage(vma, address & PAGE_MASK, &ret); + /* no page was available -- either SIGBUS or OOM */ + if (unlikely(vmf.page == NOPAGE_SIGBUS)) + return VM_FAULT_SIGBUS; + else if (unlikely(vmf.page == NOPAGE_OOM)) + return VM_FAULT_OOM; } -retry: - new_page = vma->vm_ops->nopage(vma, address & PAGE_MASK, &ret); + /* - * No smp_rmb is needed here as long as there's a full - * spin_lock/unlock sequence inside the ->nopage callback - * (for the pagecache lookup) that acts as an implicit - * smp_mb() and prevents the i_size read to happen - * after the next truncate_count read. + * For consistency in subsequent calls, make the faulted page always + * locked. */ - - /* no page was available -- either SIGBUS, OOM or REFAULT */ - if (unlikely(new_page == NOPAGE_SIGBUS)) - return VM_FAULT_SIGBUS; - else if (unlikely(new_page == NOPAGE_OOM)) - return VM_FAULT_OOM; - else if (unlikely(new_page == NOPAGE_REFAULT)) - return VM_FAULT_MINOR; + if (unlikely(!(ret & VM_FAULT_LOCKED))) + lock_page(vmf.page); + else + VM_BUG_ON(!PageLocked(vmf.page)); /* * Should we do an early C-O-W break? */ - if (write_access) { + page = vmf.page; + if (flags & FAULT_FLAG_WRITE) { if (!(vma->vm_flags & VM_SHARED)) { - struct page *page; - - if (unlikely(anon_vma_prepare(vma))) - goto oom; + anon = 1; + if (unlikely(anon_vma_prepare(vma))) { + ret = VM_FAULT_OOM; + goto out; + } page = alloc_page_vma(GFP_HIGHUSER_MOVABLE, vma, address); - if (!page) - goto oom; - copy_user_highpage(page, new_page, address, vma); - page_cache_release(new_page); - new_page = page; - anon = 1; - + if (!page) { + ret = VM_FAULT_OOM; + goto out; + } + copy_user_highpage(page, vmf.page, address, vma); } else { - /* if the page will be shareable, see if the backing + /* + * If the page will be shareable, see if the backing * address space wants to know that the page is about - * to become writable */ - if (vma->vm_ops->page_mkwrite && - vma->vm_ops->page_mkwrite(vma, new_page) < 0 - ) { - page_cache_release(new_page); - return VM_FAULT_SIGBUS; + * to become writable + */ + if (vma->vm_ops->page_mkwrite) { + unlock_page(page); + if (vma->vm_ops->page_mkwrite(vma, page) < 0) { + ret = VM_FAULT_SIGBUS; + anon = 1; /* no anon but release vmf.page */ + goto out_unlocked; + } + lock_page(page); + /* + * XXX: this is not quite right (racy vs + * invalidate) to unlock and relock the page + * like this, however a better fix requires + * reworking page_mkwrite locking API, which + * is better done later. + */ + if (!page->mapping) { + ret = 0; + anon = 1; /* no anon but release vmf.page */ + goto out; + } } } + } page_table = pte_offset_map_lock(mm, pmd, address, &ptl); - /* - * For a file-backed vma, someone could have truncated or otherwise - * invalidated this page. If unmap_mapping_range got called, - * retry getting the page. - */ - if (mapping && unlikely(sequence != mapping->truncate_count)) { - pte_unmap_unlock(page_table, ptl); - page_cache_release(new_page); - cond_resched(); - sequence = mapping->truncate_count; - smp_rmb(); - goto retry; - } /* * This silly early PAGE_DIRTY setting removes a race @@ -2388,45 +2416,63 @@ retry: * handle that later. */ /* Only go through if we didn't race with anybody else... */ - if (pte_none(*page_table)) { - flush_icache_page(vma, new_page); - entry = mk_pte(new_page, vma->vm_page_prot); - if (write_access) + if (likely(pte_same(*page_table, orig_pte))) { + flush_icache_page(vma, page); + entry = mk_pte(page, vma->vm_page_prot); + if (flags & FAULT_FLAG_WRITE) entry = maybe_mkwrite(pte_mkdirty(entry), vma); set_pte_at(mm, address, page_table, entry); if (anon) { - inc_mm_counter(mm, anon_rss); - lru_cache_add_active(new_page); - page_add_new_anon_rmap(new_page, vma, address); + inc_mm_counter(mm, anon_rss); + lru_cache_add_active(page); + page_add_new_anon_rmap(page, vma, address); } else { inc_mm_counter(mm, file_rss); - page_add_file_rmap(new_page); - if (write_access) { - dirty_page = new_page; + page_add_file_rmap(page); + if (flags & FAULT_FLAG_WRITE) { + dirty_page = page; get_page(dirty_page); } } + + /* no need to invalidate: a not-present page won't be cached */ + update_mmu_cache(vma, address, entry); + lazy_mmu_prot_update(entry); } else { - /* One of our sibling threads was faster, back out. */ - page_cache_release(new_page); - goto unlock; + if (anon) + page_cache_release(page); + else + anon = 1; /* no anon but release faulted_page */ } - /* no need to invalidate: a not-present page shouldn't be cached */ - update_mmu_cache(vma, address, entry); - lazy_mmu_prot_update(entry); -unlock: pte_unmap_unlock(page_table, ptl); - if (dirty_page) { + +out: + unlock_page(vmf.page); +out_unlocked: + if (anon) + page_cache_release(vmf.page); + else if (dirty_page) { set_page_dirty_balance(dirty_page); put_page(dirty_page); } + return ret; -oom: - page_cache_release(new_page); - return VM_FAULT_OOM; } +static int do_linear_fault(struct mm_struct *mm, struct vm_area_struct *vma, + unsigned long address, pte_t *page_table, pmd_t *pmd, + int write_access, pte_t orig_pte) +{ + pgoff_t pgoff = (((address & PAGE_MASK) + - vma->vm_start) >> PAGE_CACHE_SHIFT) + vma->vm_pgoff; + unsigned int flags = (write_access ? FAULT_FLAG_WRITE : 0); + + return __do_fault(mm, vma, address, page_table, pmd, pgoff, + flags, orig_pte); +} + + /* * do_no_pfn() tries to create a new page mapping for a page without * a struct_page backing it @@ -2450,7 +2496,6 @@ static noinline int do_no_pfn(struct mm_struct *mm, struct vm_area_struct *vma, spinlock_t *ptl; pte_t entry; unsigned long pfn; - int ret = VM_FAULT_MINOR; pte_unmap(page_table); BUG_ON(!(vma->vm_flags & VM_PFNMAP)); @@ -2462,7 +2507,7 @@ static noinline int do_no_pfn(struct mm_struct *mm, struct vm_area_struct *vma, else if (unlikely(pfn == NOPFN_SIGBUS)) return VM_FAULT_SIGBUS; else if (unlikely(pfn == NOPFN_REFAULT)) - return VM_FAULT_MINOR; + return 0; page_table = pte_offset_map_lock(mm, pmd, address, &ptl); @@ -2474,7 +2519,7 @@ static noinline int do_no_pfn(struct mm_struct *mm, struct vm_area_struct *vma, set_pte_at(mm, address, page_table, entry); } pte_unmap_unlock(page_table, ptl); - return ret; + return 0; } /* @@ -2486,33 +2531,30 @@ static noinline int do_no_pfn(struct mm_struct *mm, struct vm_area_struct *vma, * but allow concurrent faults), and pte mapped but not yet locked. * We return with mmap_sem still held, but pte unmapped and unlocked. */ -static int do_file_page(struct mm_struct *mm, struct vm_area_struct *vma, +static int do_nonlinear_fault(struct mm_struct *mm, struct vm_area_struct *vma, unsigned long address, pte_t *page_table, pmd_t *pmd, int write_access, pte_t orig_pte) { + unsigned int flags = FAULT_FLAG_NONLINEAR | + (write_access ? FAULT_FLAG_WRITE : 0); pgoff_t pgoff; - int err; if (!pte_unmap_same(mm, pmd, page_table, orig_pte)) - return VM_FAULT_MINOR; + return 0; - if (unlikely(!(vma->vm_flags & VM_NONLINEAR))) { + if (unlikely(!(vma->vm_flags & VM_NONLINEAR) || + !(vma->vm_flags & VM_CAN_NONLINEAR))) { /* * Page table corrupted: show pte and kill process. */ print_bad_pte(vma, orig_pte, address); return VM_FAULT_OOM; } - /* We can then assume vm->vm_ops && vma->vm_ops->populate */ pgoff = pte_to_pgoff(orig_pte); - err = vma->vm_ops->populate(vma, address & PAGE_MASK, PAGE_SIZE, - vma->vm_page_prot, pgoff, 0); - if (err == -ENOMEM) - return VM_FAULT_OOM; - if (err) - return VM_FAULT_SIGBUS; - return VM_FAULT_MAJOR; + + return __do_fault(mm, vma, address, page_table, pmd, pgoff, + flags, orig_pte); } /* @@ -2539,10 +2581,9 @@ static inline int handle_pte_fault(struct mm_struct *mm, if (!pte_present(entry)) { if (pte_none(entry)) { if (vma->vm_ops) { - if (vma->vm_ops->nopage) - return do_no_page(mm, vma, address, - pte, pmd, - write_access); + if (vma->vm_ops->fault || vma->vm_ops->nopage) + return do_linear_fault(mm, vma, address, + pte, pmd, write_access, entry); if (unlikely(vma->vm_ops->nopfn)) return do_no_pfn(mm, vma, address, pte, pmd, write_access); @@ -2551,7 +2592,7 @@ static inline int handle_pte_fault(struct mm_struct *mm, pte, pmd, write_access); } if (pte_file(entry)) - return do_file_page(mm, vma, address, + return do_nonlinear_fault(mm, vma, address, pte, pmd, write_access, entry); return do_swap_page(mm, vma, address, pte, pmd, write_access, entry); @@ -2583,13 +2624,13 @@ static inline int handle_pte_fault(struct mm_struct *mm, } unlock: pte_unmap_unlock(pte, ptl); - return VM_FAULT_MINOR; + return 0; } /* * By the time we get here, we already hold the mm semaphore */ -int __handle_mm_fault(struct mm_struct *mm, struct vm_area_struct *vma, +int handle_mm_fault(struct mm_struct *mm, struct vm_area_struct *vma, unsigned long address, int write_access) { pgd_t *pgd; @@ -2618,7 +2659,7 @@ int __handle_mm_fault(struct mm_struct *mm, struct vm_area_struct *vma, return handle_pte_fault(mm, vma, address, pte, pmd, write_access); } -EXPORT_SYMBOL_GPL(__handle_mm_fault); +EXPORT_SYMBOL_GPL(handle_mm_fault); #ifndef __PAGETABLE_PUD_FOLDED /* @@ -2824,3 +2865,4 @@ int access_process_vm(struct task_struct *tsk, unsigned long addr, void *buf, in return buf - old_buf; } +EXPORT_SYMBOL_GPL(access_process_vm); diff --git a/mm/mmap.c b/mm/mmap.c index 144b4a290f2..7afc7a7cec6 100644 --- a/mm/mmap.c +++ b/mm/mmap.c @@ -1165,12 +1165,8 @@ out: mm->locked_vm += len >> PAGE_SHIFT; make_pages_present(addr, addr + len); } - if (flags & MAP_POPULATE) { - up_write(&mm->mmap_sem); - sys_remap_file_pages(addr, len, 0, - pgoff, flags & MAP_NONBLOCK); - down_write(&mm->mmap_sem); - } + if ((flags & MAP_POPULATE) && !(flags & MAP_NONBLOCK)) + make_pages_present(addr, addr + len); return addr; unmap_and_free_vma: @@ -1575,33 +1571,11 @@ int expand_upwards(struct vm_area_struct *vma, unsigned long address) } #endif /* CONFIG_STACK_GROWSUP || CONFIG_IA64 */ -#ifdef CONFIG_STACK_GROWSUP -int expand_stack(struct vm_area_struct *vma, unsigned long address) -{ - return expand_upwards(vma, address); -} - -struct vm_area_struct * -find_extend_vma(struct mm_struct *mm, unsigned long addr) -{ - struct vm_area_struct *vma, *prev; - - addr &= PAGE_MASK; - vma = find_vma_prev(mm, addr, &prev); - if (vma && (vma->vm_start <= addr)) - return vma; - if (!prev || expand_stack(prev, addr)) - return NULL; - if (prev->vm_flags & VM_LOCKED) { - make_pages_present(addr, prev->vm_end); - } - return prev; -} -#else /* * vma is the first one with address < vma->vm_start. Have to extend vma. */ -int expand_stack(struct vm_area_struct *vma, unsigned long address) +static inline int expand_downwards(struct vm_area_struct *vma, + unsigned long address) { int error; @@ -1638,6 +1612,38 @@ int expand_stack(struct vm_area_struct *vma, unsigned long address) return error; } +int expand_stack_downwards(struct vm_area_struct *vma, unsigned long address) +{ + return expand_downwards(vma, address); +} + +#ifdef CONFIG_STACK_GROWSUP +int expand_stack(struct vm_area_struct *vma, unsigned long address) +{ + return expand_upwards(vma, address); +} + +struct vm_area_struct * +find_extend_vma(struct mm_struct *mm, unsigned long addr) +{ + struct vm_area_struct *vma, *prev; + + addr &= PAGE_MASK; + vma = find_vma_prev(mm, addr, &prev); + if (vma && (vma->vm_start <= addr)) + return vma; + if (!prev || expand_stack(prev, addr)) + return NULL; + if (prev->vm_flags & VM_LOCKED) + make_pages_present(addr, prev->vm_end); + return prev; +} +#else +int expand_stack(struct vm_area_struct *vma, unsigned long address) +{ + return expand_downwards(vma, address); +} + struct vm_area_struct * find_extend_vma(struct mm_struct * mm, unsigned long addr) { @@ -1655,9 +1661,8 @@ find_extend_vma(struct mm_struct * mm, unsigned long addr) start = vma->vm_start; if (expand_stack(vma, addr)) return NULL; - if (vma->vm_flags & VM_LOCKED) { + if (vma->vm_flags & VM_LOCKED) make_pages_present(addr, start); - } return vma; } #endif diff --git a/mm/mprotect.c b/mm/mprotect.c index 3b8f3c0c63f..e8346c30abe 100644 --- a/mm/mprotect.c +++ b/mm/mprotect.c @@ -128,7 +128,7 @@ static void change_protection(struct vm_area_struct *vma, flush_tlb_range(vma, start, end); } -static int +int mprotect_fixup(struct vm_area_struct *vma, struct vm_area_struct **pprev, unsigned long start, unsigned long end, unsigned long newflags) { diff --git a/mm/mremap.c b/mm/mremap.c index bc7c52efc71..8ea5c2412c6 100644 --- a/mm/mremap.c +++ b/mm/mremap.c @@ -120,7 +120,7 @@ static void move_ptes(struct vm_area_struct *vma, pmd_t *old_pmd, #define LATENCY_LIMIT (64 * PAGE_SIZE) -static unsigned long move_page_tables(struct vm_area_struct *vma, +unsigned long move_page_tables(struct vm_area_struct *vma, unsigned long old_addr, struct vm_area_struct *new_vma, unsigned long new_addr, unsigned long len) { diff --git a/mm/nommu.c b/mm/nommu.c index 8bbbf147a79..1b105d28949 100644 --- a/mm/nommu.c +++ b/mm/nommu.c @@ -1341,11 +1341,10 @@ int in_gate_area_no_task(unsigned long addr) return 0; } -struct page *filemap_nopage(struct vm_area_struct *area, - unsigned long address, int *type) +int filemap_fault(struct vm_area_struct *vma, struct vm_fault *vmf) { BUG(); - return NULL; + return 0; } /* diff --git a/mm/page-writeback.c b/mm/page-writeback.c index 886ea0d5a13..63512a9ed57 100644 --- a/mm/page-writeback.c +++ b/mm/page-writeback.c @@ -918,6 +918,9 @@ int clear_page_dirty_for_io(struct page *page) { struct address_space *mapping = page_mapping(page); + BUG_ON(!PageLocked(page)); + + ClearPageReclaim(page); if (mapping && mapping_cap_account_dirty(mapping)) { /* * Yes, Virginia, this is indeed insane. @@ -943,14 +946,19 @@ int clear_page_dirty_for_io(struct page *page) * We basically use the page "master dirty bit" * as a serialization point for all the different * threads doing their things. - * - * FIXME! We still have a race here: if somebody - * adds the page back to the page tables in - * between the "page_mkclean()" and the "TestClearPageDirty()", - * we might have it mapped without the dirty bit set. */ if (page_mkclean(page)) set_page_dirty(page); + /* + * We carefully synchronise fault handlers against + * installing a dirty pte and marking the page dirty + * at this point. We do this by having them hold the + * page lock at some point after installing their + * pte, but before marking the page dirty. + * Pages are always locked coming in here, so we get + * the desired exclusion. See mm/memory.c:do_wp_page() + * for more comments. + */ if (TestClearPageDirty(page)) { dec_zone_page_state(page, NR_FILE_DIRTY); return 1; @@ -979,6 +987,8 @@ int test_clear_page_writeback(struct page *page) } else { ret = TestClearPageWriteback(page); } + if (ret) + dec_zone_page_state(page, NR_WRITEBACK); return ret; } @@ -1004,6 +1014,8 @@ int test_set_page_writeback(struct page *page) } else { ret = TestSetPageWriteback(page); } + if (!ret) + inc_zone_page_state(page, NR_WRITEBACK); return ret; } diff --git a/mm/page_alloc.c b/mm/page_alloc.c index e2a10b957f2..43cb3b3e167 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -453,12 +453,6 @@ static inline int free_pages_check(struct page *page) 1 << PG_reserved | 1 << PG_buddy )))) bad_page(page); - /* - * PageReclaim == PageTail. It is only an error - * for PageReclaim to be set if PageCompound is clear. - */ - if (unlikely(!PageCompound(page) && PageReclaim(page))) - bad_page(page); if (PageDirty(page)) __ClearPageDirty(page); /* @@ -602,7 +596,6 @@ static int prep_new_page(struct page *page, int order, gfp_t gfp_flags) 1 << PG_locked | 1 << PG_active | 1 << PG_dirty | - 1 << PG_reclaim | 1 << PG_slab | 1 << PG_swapcache | 1 << PG_writeback | @@ -617,7 +610,7 @@ static int prep_new_page(struct page *page, int order, gfp_t gfp_flags) if (PageReserved(page)) return 1; - page->flags &= ~(1 << PG_uptodate | 1 << PG_error | + page->flags &= ~(1 << PG_uptodate | 1 << PG_error | 1 << PG_readahead | 1 << PG_referenced | 1 << PG_arch_1 | 1 << PG_owner_priv_1 | 1 << PG_mappedtodisk); set_page_private(page, 0); diff --git a/mm/readahead.c b/mm/readahead.c index 9861e883fe5..39bf45d4332 100644 --- a/mm/readahead.c +++ b/mm/readahead.c @@ -21,8 +21,16 @@ void default_unplug_io_fn(struct backing_dev_info *bdi, struct page *page) } EXPORT_SYMBOL(default_unplug_io_fn); +/* + * Convienent macros for min/max read-ahead pages. + * Note that MAX_RA_PAGES is rounded down, while MIN_RA_PAGES is rounded up. + * The latter is necessary for systems with large page size(i.e. 64k). + */ +#define MAX_RA_PAGES (VM_MAX_READAHEAD*1024 / PAGE_CACHE_SIZE) +#define MIN_RA_PAGES DIV_ROUND_UP(VM_MIN_READAHEAD*1024, PAGE_CACHE_SIZE) + struct backing_dev_info default_backing_dev_info = { - .ra_pages = (VM_MAX_READAHEAD * 1024) / PAGE_CACHE_SIZE, + .ra_pages = MAX_RA_PAGES, .state = 0, .capabilities = BDI_CAP_MAP_COPY, .unplug_io_fn = default_unplug_io_fn, @@ -41,82 +49,6 @@ file_ra_state_init(struct file_ra_state *ra, struct address_space *mapping) } EXPORT_SYMBOL_GPL(file_ra_state_init); -/* - * Return max readahead size for this inode in number-of-pages. - */ -static inline unsigned long get_max_readahead(struct file_ra_state *ra) -{ - return ra->ra_pages; -} - -static inline unsigned long get_min_readahead(struct file_ra_state *ra) -{ - return (VM_MIN_READAHEAD * 1024) / PAGE_CACHE_SIZE; -} - -static inline void reset_ahead_window(struct file_ra_state *ra) -{ - /* - * ... but preserve ahead_start + ahead_size value, - * see 'recheck:' label in page_cache_readahead(). - * Note: We never use ->ahead_size as rvalue without - * checking ->ahead_start != 0 first. - */ - ra->ahead_size += ra->ahead_start; - ra->ahead_start = 0; -} - -static inline void ra_off(struct file_ra_state *ra) -{ - ra->start = 0; - ra->flags = 0; - ra->size = 0; - reset_ahead_window(ra); - return; -} - -/* - * Set the initial window size, round to next power of 2 and square - * for small size, x 4 for medium, and x 2 for large - * for 128k (32 page) max ra - * 1-8 page = 32k initial, > 8 page = 128k initial - */ -static unsigned long get_init_ra_size(unsigned long size, unsigned long max) -{ - unsigned long newsize = roundup_pow_of_two(size); - - if (newsize <= max / 32) - newsize = newsize * 4; - else if (newsize <= max / 4) - newsize = newsize * 2; - else - newsize = max; - return newsize; -} - -/* - * Set the new window size, this is called only when I/O is to be submitted, - * not for each call to readahead. If a cache miss occured, reduce next I/O - * size, else increase depending on how close to max we are. - */ -static inline unsigned long get_next_ra_size(struct file_ra_state *ra) -{ - unsigned long max = get_max_readahead(ra); - unsigned long min = get_min_readahead(ra); - unsigned long cur = ra->size; - unsigned long newsize; - - if (ra->flags & RA_FLAG_MISS) { - ra->flags &= ~RA_FLAG_MISS; - newsize = max((cur - 2), min); - } else if (cur < max / 16) { - newsize = 4 * cur; - } else { - newsize = 2 * cur; - } - return min(newsize, max); -} - #define list_to_page(head) (list_entry((head)->prev, struct page, lru)) /** @@ -193,66 +125,6 @@ out: } /* - * Readahead design. - * - * The fields in struct file_ra_state represent the most-recently-executed - * readahead attempt: - * - * start: Page index at which we started the readahead - * size: Number of pages in that read - * Together, these form the "current window". - * Together, start and size represent the `readahead window'. - * prev_index: The page which the readahead algorithm most-recently inspected. - * It is mainly used to detect sequential file reading. - * If page_cache_readahead sees that it is again being called for - * a page which it just looked at, it can return immediately without - * making any state changes. - * offset: Offset in the prev_index where the last read ended - used for - * detection of sequential file reading. - * ahead_start, - * ahead_size: Together, these form the "ahead window". - * ra_pages: The externally controlled max readahead for this fd. - * - * When readahead is in the off state (size == 0), readahead is disabled. - * In this state, prev_index is used to detect the resumption of sequential I/O. - * - * The readahead code manages two windows - the "current" and the "ahead" - * windows. The intent is that while the application is walking the pages - * in the current window, I/O is underway on the ahead window. When the - * current window is fully traversed, it is replaced by the ahead window - * and the ahead window is invalidated. When this copying happens, the - * new current window's pages are probably still locked. So - * we submit a new batch of I/O immediately, creating a new ahead window. - * - * So: - * - * ----|----------------|----------------|----- - * ^start ^start+size - * ^ahead_start ^ahead_start+ahead_size - * - * ^ When this page is read, we submit I/O for the - * ahead window. - * - * A `readahead hit' occurs when a read request is made against a page which is - * the next sequential page. Ahead window calculations are done only when it - * is time to submit a new IO. The code ramps up the size agressively at first, - * but slow down as it approaches max_readhead. - * - * Any seek/ramdom IO will result in readahead being turned off. It will resume - * at the first sequential access. - * - * There is a special-case: if the first page which the application tries to - * read happens to be the first page of the file, it is assumed that a linear - * read is about to happen and the window is immediately set to the initial size - * based on I/O request size and the max_readahead. - * - * This function is to be called for every read request, rather than when - * it is time to perform readahead. It is called only once for the entire I/O - * regardless of size unless readahead is unable to start enough I/O to satisfy - * the request (I/O request > max_readahead). - */ - -/* * do_page_cache_readahead actually reads a chunk of disk. It allocates all * the pages first, then submits them all for I/O. This avoids the very bad * behaviour which would occur if page allocations are causing VM writeback. @@ -265,7 +137,8 @@ out: */ static int __do_page_cache_readahead(struct address_space *mapping, struct file *filp, - pgoff_t offset, unsigned long nr_to_read) + pgoff_t offset, unsigned long nr_to_read, + unsigned long lookahead_size) { struct inode *inode = mapping->host; struct page *page; @@ -278,7 +151,7 @@ __do_page_cache_readahead(struct address_space *mapping, struct file *filp, if (isize == 0) goto out; - end_index = ((isize - 1) >> PAGE_CACHE_SHIFT); + end_index = ((isize - 1) >> PAGE_CACHE_SHIFT); /* * Preallocate as many pages as we will need. @@ -286,7 +159,7 @@ __do_page_cache_readahead(struct address_space *mapping, struct file *filp, read_lock_irq(&mapping->tree_lock); for (page_idx = 0; page_idx < nr_to_read; page_idx++) { pgoff_t page_offset = offset + page_idx; - + if (page_offset > end_index) break; @@ -301,6 +174,8 @@ __do_page_cache_readahead(struct address_space *mapping, struct file *filp, break; page->index = page_offset; list_add(&page->lru, &page_pool); + if (page_idx == nr_to_read - lookahead_size) + SetPageReadahead(page); ret++; } read_unlock_irq(&mapping->tree_lock); @@ -337,7 +212,7 @@ int force_page_cache_readahead(struct address_space *mapping, struct file *filp, if (this_chunk > nr_to_read) this_chunk = nr_to_read; err = __do_page_cache_readahead(mapping, filp, - offset, this_chunk); + offset, this_chunk, 0); if (err < 0) { ret = err; break; @@ -350,28 +225,6 @@ int force_page_cache_readahead(struct address_space *mapping, struct file *filp, } /* - * Check how effective readahead is being. If the amount of started IO is - * less than expected then the file is partly or fully in pagecache and - * readahead isn't helping. - * - */ -static inline int check_ra_success(struct file_ra_state *ra, - unsigned long nr_to_read, unsigned long actual) -{ - if (actual == 0) { - ra->cache_hit += nr_to_read; - if (ra->cache_hit >= VM_MAX_CACHE_HIT) { - ra_off(ra); - ra->flags |= RA_FLAG_INCACHE; - return 0; - } - } else { - ra->cache_hit=0; - } - return 1; -} - -/* * This version skips the IO if the queue is read-congested, and will tell the * block layer to abandon the readahead if request allocation would block. * @@ -384,200 +237,237 @@ int do_page_cache_readahead(struct address_space *mapping, struct file *filp, if (bdi_read_congested(mapping->backing_dev_info)) return -1; - return __do_page_cache_readahead(mapping, filp, offset, nr_to_read); + return __do_page_cache_readahead(mapping, filp, offset, nr_to_read, 0); } /* - * Read 'nr_to_read' pages starting at page 'offset'. If the flag 'block' - * is set wait till the read completes. Otherwise attempt to read without - * blocking. - * Returns 1 meaning 'success' if read is successful without switching off - * readahead mode. Otherwise return failure. + * Given a desired number of PAGE_CACHE_SIZE readahead pages, return a + * sensible upper limit. */ -static int -blockable_page_cache_readahead(struct address_space *mapping, struct file *filp, - pgoff_t offset, unsigned long nr_to_read, - struct file_ra_state *ra, int block) +unsigned long max_sane_readahead(unsigned long nr) +{ + return min(nr, (node_page_state(numa_node_id(), NR_INACTIVE) + + node_page_state(numa_node_id(), NR_FREE_PAGES)) / 2); +} + +/* + * Submit IO for the read-ahead request in file_ra_state. + */ +static unsigned long ra_submit(struct file_ra_state *ra, + struct address_space *mapping, struct file *filp) { int actual; - if (!block && bdi_read_congested(mapping->backing_dev_info)) - return 0; + actual = __do_page_cache_readahead(mapping, filp, + ra->start, ra->size, ra->async_size); + + return actual; +} - actual = __do_page_cache_readahead(mapping, filp, offset, nr_to_read); +/* + * Set the initial window size, round to next power of 2 and square + * for small size, x 4 for medium, and x 2 for large + * for 128k (32 page) max ra + * 1-8 page = 32k initial, > 8 page = 128k initial + */ +static unsigned long get_init_ra_size(unsigned long size, unsigned long max) +{ + unsigned long newsize = roundup_pow_of_two(size); - return check_ra_success(ra, nr_to_read, actual); + if (newsize <= max / 32) + newsize = newsize * 4; + else if (newsize <= max / 4) + newsize = newsize * 2; + else + newsize = max; + + return newsize; } -static int make_ahead_window(struct address_space *mapping, struct file *filp, - struct file_ra_state *ra, int force) +/* + * Get the previous window size, ramp it up, and + * return it as the new window size. + */ +static unsigned long get_next_ra_size(struct file_ra_state *ra, + unsigned long max) { - int block, ret; - - ra->ahead_size = get_next_ra_size(ra); - ra->ahead_start = ra->start + ra->size; - - block = force || (ra->prev_index >= ra->ahead_start); - ret = blockable_page_cache_readahead(mapping, filp, - ra->ahead_start, ra->ahead_size, ra, block); - - if (!ret && !force) { - /* A read failure in blocking mode, implies pages are - * all cached. So we can safely assume we have taken - * care of all the pages requested in this call. - * A read failure in non-blocking mode, implies we are - * reading more pages than requested in this call. So - * we safely assume we have taken care of all the pages - * requested in this call. - * - * Just reset the ahead window in case we failed due to - * congestion. The ahead window will any way be closed - * in case we failed due to excessive page cache hits. - */ - reset_ahead_window(ra); - } + unsigned long cur = ra->size; + unsigned long newsize; - return ret; + if (cur < max / 16) + newsize = 4 * cur; + else + newsize = 2 * cur; + + return min(newsize, max); } -/** - * page_cache_readahead - generic adaptive readahead - * @mapping: address_space which holds the pagecache and I/O vectors - * @ra: file_ra_state which holds the readahead state - * @filp: passed on to ->readpage() and ->readpages() - * @offset: start offset into @mapping, in PAGE_CACHE_SIZE units - * @req_size: hint: total size of the read which the caller is performing in - * PAGE_CACHE_SIZE units +/* + * On-demand readahead design. + * + * The fields in struct file_ra_state represent the most-recently-executed + * readahead attempt: + * + * |<----- async_size ---------| + * |------------------- size -------------------->| + * |==================#===========================| + * ^start ^page marked with PG_readahead * - * page_cache_readahead() is the main function. If performs the adaptive - * readahead window size management and submits the readahead I/O. + * To overlap application thinking time and disk I/O time, we do + * `readahead pipelining': Do not wait until the application consumed all + * readahead pages and stalled on the missing page at readahead_index; + * Instead, submit an asynchronous readahead I/O as soon as there are + * only async_size pages left in the readahead window. Normally async_size + * will be equal to size, for maximum pipelining. * - * Note that @filp is purely used for passing on to the ->readpage[s]() - * handler: it may refer to a different file from @mapping (so we may not use - * @filp->f_mapping or @filp->f_path.dentry->d_inode here). - * Also, @ra may not be equal to &@filp->f_ra. + * In interleaved sequential reads, concurrent streams on the same fd can + * be invalidating each other's readahead state. So we flag the new readahead + * page at (start+size-async_size) with PG_readahead, and use it as readahead + * indicator. The flag won't be set on already cached pages, to avoid the + * readahead-for-nothing fuss, saving pointless page cache lookups. + * + * prev_index tracks the last visited page in the _previous_ read request. + * It should be maintained by the caller, and will be used for detecting + * small random reads. Note that the readahead algorithm checks loosely + * for sequential patterns. Hence interleaved reads might be served as + * sequential ones. + * + * There is a special-case: if the first page which the application tries to + * read happens to be the first page of the file, it is assumed that a linear + * read is about to happen and the window is immediately set to the initial size + * based on I/O request size and the max_readahead. * + * The code ramps up the readahead size aggressively at first, but slow down as + * it approaches max_readhead. + */ + +/* + * A minimal readahead algorithm for trivial sequential/random reads. */ -unsigned long -page_cache_readahead(struct address_space *mapping, struct file_ra_state *ra, - struct file *filp, pgoff_t offset, unsigned long req_size) +static unsigned long +ondemand_readahead(struct address_space *mapping, + struct file_ra_state *ra, struct file *filp, + bool hit_readahead_marker, pgoff_t offset, + unsigned long req_size) { - unsigned long max, newsize; + unsigned long max; /* max readahead pages */ int sequential; - /* - * We avoid doing extra work and bogusly perturbing the readahead - * window expansion logic. - */ - if (offset == ra->prev_index && --req_size) - ++offset; - - /* Note that prev_index == -1 if it is a first read */ - sequential = (offset == ra->prev_index + 1); - ra->prev_index = offset; - ra->prev_offset = 0; - - max = get_max_readahead(ra); - newsize = min(req_size, max); - - /* No readahead or sub-page sized read or file already in cache */ - if (newsize == 0 || (ra->flags & RA_FLAG_INCACHE)) - goto out; - - ra->prev_index += newsize - 1; + max = ra->ra_pages; + sequential = (offset - ra->prev_index <= 1UL) || (req_size > max); /* - * Special case - first read at start of file. We'll assume it's - * a whole-file read and grow the window fast. Or detect first - * sequential access + * It's the expected callback offset, assume sequential access. + * Ramp up sizes, and push forward the readahead window. */ - if (sequential && ra->size == 0) { - ra->size = get_init_ra_size(newsize, max); - ra->start = offset; - if (!blockable_page_cache_readahead(mapping, filp, offset, - ra->size, ra, 1)) - goto out; - - /* - * If the request size is larger than our max readahead, we - * at least want to be sure that we get 2 IOs in flight and - * we know that we will definitly need the new I/O. - * once we do this, subsequent calls should be able to overlap - * IOs,* thus preventing stalls. so issue the ahead window - * immediately. - */ - if (req_size >= max) - make_ahead_window(mapping, filp, ra, 1); - - goto out; + if (offset && (offset == (ra->start + ra->size - ra->async_size) || + offset == (ra->start + ra->size))) { + ra->start += ra->size; + ra->size = get_next_ra_size(ra, max); + ra->async_size = ra->size; + goto readit; } /* - * Now handle the random case: - * partial page reads and first access were handled above, - * so this must be the next page otherwise it is random + * Standalone, small read. + * Read as is, and do not pollute the readahead state. */ - if (!sequential) { - ra_off(ra); - blockable_page_cache_readahead(mapping, filp, offset, - newsize, ra, 1); - goto out; + if (!hit_readahead_marker && !sequential) { + return __do_page_cache_readahead(mapping, filp, + offset, req_size, 0); } /* - * If we get here we are doing sequential IO and this was not the first - * occurence (ie we have an existing window) + * It may be one of + * - first read on start of file + * - sequential cache miss + * - oversize random read + * Start readahead for it. */ - if (ra->ahead_start == 0) { /* no ahead window yet */ - if (!make_ahead_window(mapping, filp, ra, 0)) - goto recheck; - } + ra->start = offset; + ra->size = get_init_ra_size(req_size, max); + ra->async_size = ra->size > req_size ? ra->size - req_size : ra->size; /* - * Already have an ahead window, check if we crossed into it. - * If so, shift windows and issue a new ahead window. - * Only return the #pages that are in the current window, so that - * we get called back on the first page of the ahead window which - * will allow us to submit more IO. + * Hit on a marked page without valid readahead state. + * E.g. interleaved reads. + * Not knowing its readahead pos/size, bet on the minimal possible one. */ - if (ra->prev_index >= ra->ahead_start) { - ra->start = ra->ahead_start; - ra->size = ra->ahead_size; - make_ahead_window(mapping, filp, ra, 0); -recheck: - /* prev_index shouldn't overrun the ahead window */ - ra->prev_index = min(ra->prev_index, - ra->ahead_start + ra->ahead_size - 1); + if (hit_readahead_marker) { + ra->start++; + ra->size = get_next_ra_size(ra, max); } -out: - return ra->prev_index + 1; +readit: + return ra_submit(ra, mapping, filp); } -EXPORT_SYMBOL_GPL(page_cache_readahead); -/* - * handle_ra_miss() is called when it is known that a page which should have - * been present in the pagecache (we just did some readahead there) was in fact - * not found. This will happen if it was evicted by the VM (readahead - * thrashing) +/** + * page_cache_sync_readahead - generic file readahead + * @mapping: address_space which holds the pagecache and I/O vectors + * @ra: file_ra_state which holds the readahead state + * @filp: passed on to ->readpage() and ->readpages() + * @offset: start offset into @mapping, in pagecache page-sized units + * @req_size: hint: total size of the read which the caller is performing in + * pagecache pages * - * Turn on the cache miss flag in the RA struct, this will cause the RA code - * to reduce the RA size on the next read. + * page_cache_sync_readahead() should be called when a cache miss happened: + * it will submit the read. The readahead logic may decide to piggyback more + * pages onto the read request if access patterns suggest it will improve + * performance. */ -void handle_ra_miss(struct address_space *mapping, - struct file_ra_state *ra, pgoff_t offset) +void page_cache_sync_readahead(struct address_space *mapping, + struct file_ra_state *ra, struct file *filp, + pgoff_t offset, unsigned long req_size) { - ra->flags |= RA_FLAG_MISS; - ra->flags &= ~RA_FLAG_INCACHE; - ra->cache_hit = 0; + /* no read-ahead */ + if (!ra->ra_pages) + return; + + /* do read-ahead */ + ondemand_readahead(mapping, ra, filp, false, offset, req_size); } +EXPORT_SYMBOL_GPL(page_cache_sync_readahead); -/* - * Given a desired number of PAGE_CACHE_SIZE readahead pages, return a - * sensible upper limit. - */ -unsigned long max_sane_readahead(unsigned long nr) +/** + * page_cache_async_readahead - file readahead for marked pages + * @mapping: address_space which holds the pagecache and I/O vectors + * @ra: file_ra_state which holds the readahead state + * @filp: passed on to ->readpage() and ->readpages() + * @page: the page at @offset which has the PG_readahead flag set + * @offset: start offset into @mapping, in pagecache page-sized units + * @req_size: hint: total size of the read which the caller is performing in + * pagecache pages + * + * page_cache_async_ondemand() should be called when a page is used which + * has the PG_readahead flag: this is a marker to suggest that the application + * has used up enough of the readahead window that we should start pulling in + * more pages. */ +void +page_cache_async_readahead(struct address_space *mapping, + struct file_ra_state *ra, struct file *filp, + struct page *page, pgoff_t offset, + unsigned long req_size) { - return min(nr, (node_page_state(numa_node_id(), NR_INACTIVE) - + node_page_state(numa_node_id(), NR_FREE_PAGES)) / 2); + /* no read-ahead */ + if (!ra->ra_pages) + return; + + /* + * Same bit is used for PG_readahead and PG_reclaim. + */ + if (PageWriteback(page)) + return; + + ClearPageReadahead(page); + + /* + * Defer asynchronous read-ahead on IO congestion. + */ + if (bdi_read_congested(mapping->backing_dev_info)) + return; + + /* do read-ahead */ + ondemand_readahead(mapping, ra, filp, true, offset, req_size); } +EXPORT_SYMBOL_GPL(page_cache_async_readahead); diff --git a/mm/rmap.c b/mm/rmap.c index 61e492597a0..fede5c7910b 100644 --- a/mm/rmap.c +++ b/mm/rmap.c @@ -621,8 +621,10 @@ void page_remove_rmap(struct page *page, struct vm_area_struct *vma) printk (KERN_EMERG " page->count = %x\n", page_count(page)); printk (KERN_EMERG " page->mapping = %p\n", page->mapping); print_symbol (KERN_EMERG " vma->vm_ops = %s\n", (unsigned long)vma->vm_ops); - if (vma->vm_ops) + if (vma->vm_ops) { print_symbol (KERN_EMERG " vma->vm_ops->nopage = %s\n", (unsigned long)vma->vm_ops->nopage); + print_symbol (KERN_EMERG " vma->vm_ops->fault = %s\n", (unsigned long)vma->vm_ops->fault); + } if (vma->vm_file && vma->vm_file->f_op) print_symbol (KERN_EMERG " vma->vm_file->f_op->mmap = %s\n", (unsigned long)vma->vm_file->f_op->mmap); BUG(); diff --git a/mm/shmem.c b/mm/shmem.c index 96fa79fb6ad..ad155c7745d 100644 --- a/mm/shmem.c +++ b/mm/shmem.c @@ -83,6 +83,7 @@ enum sgp_type { SGP_READ, /* don't exceed i_size, don't allocate page */ SGP_CACHE, /* don't exceed i_size, may allocate page */ SGP_WRITE, /* may exceed i_size, may allocate page */ + SGP_FAULT, /* same as SGP_CACHE, return with page locked */ }; static int shmem_getpage(struct inode *inode, unsigned long idx, @@ -1100,6 +1101,10 @@ static int shmem_getpage(struct inode *inode, unsigned long idx, if (idx >= SHMEM_MAX_INDEX) return -EFBIG; + + if (type) + *type = 0; + /* * Normally, filepage is NULL on entry, and either found * uptodate immediately, or allocated and zeroed, or read @@ -1133,9 +1138,9 @@ repeat: if (!swappage) { shmem_swp_unmap(entry); /* here we actually do the io */ - if (type && *type == VM_FAULT_MINOR) { + if (type && !(*type & VM_FAULT_MAJOR)) { __count_vm_event(PGMAJFAULT); - *type = VM_FAULT_MAJOR; + *type |= VM_FAULT_MAJOR; } spin_unlock(&info->lock); swappage = shmem_swapin(info, swap, idx); @@ -1289,8 +1294,10 @@ repeat: } done: if (*pagep != filepage) { - unlock_page(filepage); *pagep = filepage; + if (sgp != SGP_FAULT) + unlock_page(filepage); + } return 0; @@ -1302,72 +1309,21 @@ failed: return error; } -static struct page *shmem_nopage(struct vm_area_struct *vma, - unsigned long address, int *type) +static int shmem_fault(struct vm_area_struct *vma, struct vm_fault *vmf) { struct inode *inode = vma->vm_file->f_path.dentry->d_inode; - struct page *page = NULL; - unsigned long idx; int error; + int ret; - idx = (address - vma->vm_start) >> PAGE_SHIFT; - idx += vma->vm_pgoff; - idx >>= PAGE_CACHE_SHIFT - PAGE_SHIFT; - if (((loff_t) idx << PAGE_CACHE_SHIFT) >= i_size_read(inode)) - return NOPAGE_SIGBUS; + if (((loff_t)vmf->pgoff << PAGE_CACHE_SHIFT) >= i_size_read(inode)) + return VM_FAULT_SIGBUS; - error = shmem_getpage(inode, idx, &page, SGP_CACHE, type); + error = shmem_getpage(inode, vmf->pgoff, &vmf->page, SGP_FAULT, &ret); if (error) - return (error == -ENOMEM)? NOPAGE_OOM: NOPAGE_SIGBUS; - - mark_page_accessed(page); - return page; -} - -static int shmem_populate(struct vm_area_struct *vma, - unsigned long addr, unsigned long len, - pgprot_t prot, unsigned long pgoff, int nonblock) -{ - struct inode *inode = vma->vm_file->f_path.dentry->d_inode; - struct mm_struct *mm = vma->vm_mm; - enum sgp_type sgp = nonblock? SGP_QUICK: SGP_CACHE; - unsigned long size; + return ((error == -ENOMEM) ? VM_FAULT_OOM : VM_FAULT_SIGBUS); - size = (i_size_read(inode) + PAGE_SIZE - 1) >> PAGE_SHIFT; - if (pgoff >= size || pgoff + (len >> PAGE_SHIFT) > size) - return -EINVAL; - - while ((long) len > 0) { - struct page *page = NULL; - int err; - /* - * Will need changing if PAGE_CACHE_SIZE != PAGE_SIZE - */ - err = shmem_getpage(inode, pgoff, &page, sgp, NULL); - if (err) - return err; - /* Page may still be null, but only if nonblock was set. */ - if (page) { - mark_page_accessed(page); - err = install_page(mm, vma, addr, page, prot); - if (err) { - page_cache_release(page); - return err; - } - } else if (vma->vm_flags & VM_NONLINEAR) { - /* No page was found just because we can't read it in - * now (being here implies nonblock != 0), but the page - * may exist, so set the PTE to fault it in later. */ - err = install_file_pte(mm, vma, addr, pgoff, prot); - if (err) - return err; - } - - len -= PAGE_SIZE; - addr += PAGE_SIZE; - pgoff++; - } - return 0; + mark_page_accessed(vmf->page); + return ret | VM_FAULT_LOCKED; } #ifdef CONFIG_NUMA @@ -1414,6 +1370,7 @@ static int shmem_mmap(struct file *file, struct vm_area_struct *vma) { file_accessed(file); vma->vm_ops = &shmem_vm_ops; + vma->vm_flags |= VM_CAN_NONLINEAR; return 0; } @@ -2459,8 +2416,7 @@ static const struct super_operations shmem_ops = { }; static struct vm_operations_struct shmem_vm_ops = { - .nopage = shmem_nopage, - .populate = shmem_populate, + .fault = shmem_fault, #ifdef CONFIG_NUMA .set_policy = shmem_set_policy, .get_policy = shmem_get_policy, diff --git a/mm/slab.c b/mm/slab.c index 96d30ee256e..c3feeaab387 100644 --- a/mm/slab.c +++ b/mm/slab.c @@ -1163,7 +1163,7 @@ static int __cpuinit cpuup_callback(struct notifier_block *nfb, struct kmem_cache *cachep; struct kmem_list3 *l3 = NULL; int node = cpu_to_node(cpu); - int memsize = sizeof(struct kmem_list3); + const int memsize = sizeof(struct kmem_list3); switch (action) { case CPU_LOCK_ACQUIRE: @@ -3690,8 +3690,8 @@ static __always_inline void *__do_kmalloc(size_t size, gfp_t flags, * functions. */ cachep = __find_general_cachep(size, flags); - if (unlikely(cachep == NULL)) - return NULL; + if (unlikely(ZERO_OR_NULL_PTR(cachep))) + return cachep; return __cache_alloc(cachep, flags, caller); } diff --git a/mm/slub.c b/mm/slub.c index 52a4f44be39..322f3a5d72c 100644 --- a/mm/slub.c +++ b/mm/slub.c @@ -2394,7 +2394,7 @@ size_t ksize(const void *object) struct page *page; struct kmem_cache *s; - if (object == ZERO_SIZE_PTR) + if (ZERO_OR_NULL_PTR(object)) return 0; page = get_object_page(object); diff --git a/mm/truncate.c b/mm/truncate.c index f47e46d1be3..5cdfbc1a59f 100644 --- a/mm/truncate.c +++ b/mm/truncate.c @@ -82,7 +82,7 @@ EXPORT_SYMBOL(cancel_dirty_page); /* * If truncate cannot remove the fs-private metadata from the page, the page * becomes anonymous. It will be left on the LRU and may even be mapped into - * user pagetables if we're racing with filemap_nopage(). + * user pagetables if we're racing with filemap_fault(). * * We need to bale out if page->mapping is no longer equal to the original * mapping. This happens a) when the VM reclaimed the page while we waited on @@ -192,6 +192,11 @@ void truncate_inode_pages_range(struct address_space *mapping, unlock_page(page); continue; } + if (page_mapped(page)) { + unmap_mapping_range(mapping, + (loff_t)page_index<<PAGE_CACHE_SHIFT, + PAGE_CACHE_SIZE, 0); + } truncate_complete_page(mapping, page); unlock_page(page); } @@ -229,6 +234,11 @@ void truncate_inode_pages_range(struct address_space *mapping, break; lock_page(page); wait_on_page_writeback(page); + if (page_mapped(page)) { + unmap_mapping_range(mapping, + (loff_t)page->index<<PAGE_CACHE_SHIFT, + PAGE_CACHE_SIZE, 0); + } if (page->index > next) next = page->index; next++; @@ -405,7 +415,7 @@ int invalidate_inode_pages2_range(struct address_space *mapping, break; } wait_on_page_writeback(page); - while (page_mapped(page)) { + if (page_mapped(page)) { if (!did_range_unmap) { /* * Zap the rest of the file in one hit. @@ -425,6 +435,7 @@ int invalidate_inode_pages2_range(struct address_space *mapping, PAGE_CACHE_SIZE, 0); } } + BUG_ON(page_mapped(page)); ret = do_launder_page(mapping, page); if (ret == 0 && !invalidate_complete_page2(mapping, page)) ret = -EIO; diff --git a/mm/vmalloc.c b/mm/vmalloc.c index 3130c343088..3cee76a8c9f 100644 --- a/mm/vmalloc.c +++ b/mm/vmalloc.c @@ -164,6 +164,7 @@ int map_vm_area(struct vm_struct *area, pgprot_t prot, struct page ***pages) flush_cache_vmap((unsigned long) area->addr, end); return err; } +EXPORT_SYMBOL_GPL(map_vm_area); static struct vm_struct *__get_vm_area_node(unsigned long size, unsigned long flags, unsigned long start, unsigned long end, @@ -242,6 +243,7 @@ struct vm_struct *__get_vm_area(unsigned long size, unsigned long flags, { return __get_vm_area_node(size, flags, start, end, -1, GFP_KERNEL); } +EXPORT_SYMBOL_GPL(__get_vm_area); /** * get_vm_area - reserve a contingous kernel virtual area @@ -583,9 +585,9 @@ void *vmalloc_exec(unsigned long size) } #if defined(CONFIG_64BIT) && defined(CONFIG_ZONE_DMA32) -#define GFP_VMALLOC32 GFP_DMA32 +#define GFP_VMALLOC32 GFP_DMA32 | GFP_KERNEL #elif defined(CONFIG_64BIT) && defined(CONFIG_ZONE_DMA) -#define GFP_VMALLOC32 GFP_DMA +#define GFP_VMALLOC32 GFP_DMA | GFP_KERNEL #else #define GFP_VMALLOC32 GFP_KERNEL #endif diff --git a/net/ax25/af_ax25.c b/net/ax25/af_ax25.c index c83cf843297..dae2a42d3d8 100644 --- a/net/ax25/af_ax25.c +++ b/net/ax25/af_ax25.c @@ -1262,7 +1262,7 @@ static int __must_check ax25_connect(struct socket *sock, for (;;) { prepare_to_wait(sk->sk_sleep, &wait, - TASK_INTERRUPTIBLE); + TASK_INTERRUPTIBLE); if (sk->sk_state != TCP_SYN_SENT) break; if (!signal_pending(current)) { diff --git a/net/bluetooth/hci_core.c b/net/bluetooth/hci_core.c index f6d867e0179..63caa414945 100644 --- a/net/bluetooth/hci_core.c +++ b/net/bluetooth/hci_core.c @@ -982,7 +982,7 @@ int hci_recv_fragment(struct hci_dev *hdev, int type, void *data, int count) skb->dev = (void *) hdev; bt_cb(skb)->pkt_type = type; - + __reassembly(hdev, type) = skb; scb = (void *) skb->cb; diff --git a/net/core/dev.c b/net/core/dev.c index 6357f54c8ff..38212c3f997 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -2629,7 +2629,7 @@ void __dev_set_rx_mode(struct net_device *dev) return; if (!netif_device_present(dev)) - return; + return; if (dev->set_rx_mode) dev->set_rx_mode(dev); diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c index 864cbdf31ed..06eccca8cb5 100644 --- a/net/core/rtnetlink.c +++ b/net/core/rtnetlink.c @@ -98,7 +98,7 @@ int rtattr_parse(struct rtattr *tb[], int maxattr, struct rtattr *rta, int len) } int __rtattr_parse_nested_compat(struct rtattr *tb[], int maxattr, - struct rtattr *rta, int len) + struct rtattr *rta, int len) { if (RTA_PAYLOAD(rta) < len) return -1; diff --git a/net/core/sock.c b/net/core/sock.c index 091032a250c..239a08a6ff2 100644 --- a/net/core/sock.c +++ b/net/core/sock.c @@ -171,6 +171,19 @@ static const char *af_family_slock_key_strings[AF_MAX+1] = { "slock-AF_TIPC" , "slock-AF_BLUETOOTH", "slock-AF_IUCV" , "slock-AF_RXRPC" , "slock-AF_MAX" }; +static const char *af_family_clock_key_strings[AF_MAX+1] = { + "clock-AF_UNSPEC", "clock-AF_UNIX" , "clock-AF_INET" , + "clock-AF_AX25" , "clock-AF_IPX" , "clock-AF_APPLETALK", + "clock-AF_NETROM", "clock-AF_BRIDGE" , "clock-AF_ATMPVC" , + "clock-AF_X25" , "clock-AF_INET6" , "clock-AF_ROSE" , + "clock-AF_DECnet", "clock-AF_NETBEUI" , "clock-AF_SECURITY" , + "clock-AF_KEY" , "clock-AF_NETLINK" , "clock-AF_PACKET" , + "clock-AF_ASH" , "clock-AF_ECONET" , "clock-AF_ATMSVC" , + "clock-21" , "clock-AF_SNA" , "clock-AF_IRDA" , + "clock-AF_PPPOX" , "clock-AF_WANPIPE" , "clock-AF_LLC" , + "clock-27" , "clock-28" , "clock-29" , + "clock-AF_TIPC" , "clock-AF_BLUETOOTH", "clock-AF_MAX" +}; #endif /* @@ -217,7 +230,7 @@ static int sock_set_timeout(long *timeo_p, char __user *optval, int optlen) warned++; printk(KERN_INFO "sock_set_timeout: `%s' (pid %d) " "tries to set negative timeout\n", - current->comm, current->pid); + current->comm, current->pid); return 0; } *timeo_p = MAX_SCHEDULE_TIMEOUT; @@ -941,8 +954,9 @@ struct sock *sk_clone(const struct sock *sk, const gfp_t priority) rwlock_init(&newsk->sk_dst_lock); rwlock_init(&newsk->sk_callback_lock); - lockdep_set_class(&newsk->sk_callback_lock, - af_callback_keys + newsk->sk_family); + lockdep_set_class_and_name(&newsk->sk_callback_lock, + af_callback_keys + newsk->sk_family, + af_family_clock_key_strings[newsk->sk_family]); newsk->sk_dst_cache = NULL; newsk->sk_wmem_queued = 0; @@ -1530,8 +1544,9 @@ void sock_init_data(struct socket *sock, struct sock *sk) rwlock_init(&sk->sk_dst_lock); rwlock_init(&sk->sk_callback_lock); - lockdep_set_class(&sk->sk_callback_lock, - af_callback_keys + sk->sk_family); + lockdep_set_class_and_name(&sk->sk_callback_lock, + af_callback_keys + sk->sk_family, + af_family_clock_key_strings[sk->sk_family]); sk->sk_state_change = sock_def_wakeup; sk->sk_data_ready = sock_def_readable; diff --git a/net/dccp/ccids/lib/loss_interval.c b/net/dccp/ccids/lib/loss_interval.c index 515225f3a46..dd0fc992b04 100644 --- a/net/dccp/ccids/lib/loss_interval.c +++ b/net/dccp/ccids/lib/loss_interval.c @@ -227,7 +227,7 @@ void dccp_li_update_li(struct sock *sk, struct list_head *li_hist_list, struct list_head *hist_list, struct timeval *last_feedback, u16 s, u32 bytes_recv, - u32 previous_x_recv, u64 seq_loss, u8 win_loss) + u32 previous_x_recv, u64 seq_loss, u8 win_loss) { struct dccp_li_hist_entry *head; u64 seq_temp; diff --git a/net/dccp/probe.c b/net/dccp/probe.c index 43a3adb027e..bae10b0f2fc 100644 --- a/net/dccp/probe.c +++ b/net/dccp/probe.c @@ -112,7 +112,7 @@ static struct jprobe dccp_send_probe = { .kp = { .symbol_name = "dccp_sendmsg", }, - .entry = JPROBE_ENTRY(jdccp_sendmsg), + .entry = jdccp_sendmsg, }; static int dccpprobe_open(struct inode *inode, struct file *file) diff --git a/net/ipv4/fib_frontend.c b/net/ipv4/fib_frontend.c index 2eb909be804..eff6bce453e 100644 --- a/net/ipv4/fib_frontend.c +++ b/net/ipv4/fib_frontend.c @@ -817,7 +817,7 @@ static void nl_fib_input(struct sock *sk, int len) static void nl_fib_lookup_init(void) { netlink_kernel_create(NETLINK_FIB_LOOKUP, 0, nl_fib_input, NULL, - THIS_MODULE); + THIS_MODULE); } static void fib_disable_ip(struct net_device *dev, int force) diff --git a/net/ipv4/ip_forward.c b/net/ipv4/ip_forward.c index 9cb04df0054..8c95cf09f87 100644 --- a/net/ipv4/ip_forward.c +++ b/net/ipv4/ip_forward.c @@ -86,7 +86,7 @@ int ip_forward(struct sk_buff *skb) goto sr_failed; if (unlikely(skb->len > dst_mtu(&rt->u.dst) && - (ip_hdr(skb)->frag_off & htons(IP_DF))) && !skb->local_df) { + (ip_hdr(skb)->frag_off & htons(IP_DF))) && !skb->local_df) { IP_INC_STATS(IPSTATS_MIB_FRAGFAILS); icmp_send(skb, ICMP_DEST_UNREACH, ICMP_FRAG_NEEDED, htonl(dst_mtu(&rt->u.dst))); diff --git a/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4_compat.c b/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4_compat.c index 3da9d73d1b5..27c7918e442 100644 --- a/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4_compat.c +++ b/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4_compat.c @@ -177,7 +177,7 @@ static int ct_open(struct inode *inode, struct file *file) struct ct_iter_state *st; int ret; - st = kmalloc(sizeof(struct ct_iter_state), GFP_KERNEL); + st = kzalloc(sizeof(struct ct_iter_state), GFP_KERNEL); if (st == NULL) return -ENOMEM; ret = seq_open(file, &ct_seq_ops); @@ -185,7 +185,6 @@ static int ct_open(struct inode *inode, struct file *file) goto out_free; seq = file->private_data; seq->private = st; - memset(st, 0, sizeof(struct ct_iter_state)); return ret; out_free: kfree(st); diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c index 20aea1595c4..666d8a58d14 100644 --- a/net/ipv4/tcp_output.c +++ b/net/ipv4/tcp_output.c @@ -1615,7 +1615,7 @@ u32 __tcp_select_window(struct sock *sk) if (window <= free_space - mss || window > free_space) window = (free_space/mss)*mss; else if (mss == full_space && - free_space > window + full_space/2) + free_space > window + full_space/2) window = free_space; } diff --git a/net/ipv4/tcp_probe.c b/net/ipv4/tcp_probe.c index f37d5928921..b76398d1b45 100644 --- a/net/ipv4/tcp_probe.c +++ b/net/ipv4/tcp_probe.c @@ -130,7 +130,7 @@ static struct jprobe tcp_jprobe = { .kp = { .symbol_name = "tcp_rcv_established", }, - .entry = JPROBE_ENTRY(jtcp_rcv_established), + .entry = jtcp_rcv_established, }; static int tcpprobe_open(struct inode * inode, struct file * file) diff --git a/net/mac80211/Makefile b/net/mac80211/Makefile index e9738dad2d7..a9c2d0787d4 100644 --- a/net/mac80211/Makefile +++ b/net/mac80211/Makefile @@ -13,6 +13,7 @@ mac80211-objs := \ ieee80211_iface.o \ ieee80211_rate.o \ michael.o \ + regdomain.o \ tkip.o \ aes_ccm.o \ wme.o \ diff --git a/net/mac80211/debugfs_netdev.c b/net/mac80211/debugfs_netdev.c index a3e01d76d50..799a9208c4b 100644 --- a/net/mac80211/debugfs_netdev.c +++ b/net/mac80211/debugfs_netdev.c @@ -397,6 +397,8 @@ static int netdev_notify(struct notifier_block * nb, void *ndev) { struct net_device *dev = ndev; + struct dentry *dir; + struct ieee80211_sub_if_data *sdata = IEEE80211_DEV_TO_SUB_IF(dev); char buf[10+IFNAMSIZ]; if (state != NETDEV_CHANGENAME) @@ -408,10 +410,11 @@ static int netdev_notify(struct notifier_block * nb, if (dev->ieee80211_ptr->wiphy->privid != mac80211_wiphy_privid) return 0; - /* TODO sprintf(buf, "netdev:%s", dev->name); - debugfs_rename(IEEE80211_DEV_TO_SUB_IF(dev)->debugfsdir, buf); - */ + dir = sdata->debugfsdir; + if (!debugfs_rename(dir->d_parent, dir, dir->d_parent, buf)) + printk(KERN_ERR "mac80211: debugfs: failed to rename debugfs " + "dir to %s\n", buf); return 0; } diff --git a/net/mac80211/ieee80211.c b/net/mac80211/ieee80211.c index 2ddf4ef4065..c944b17d0fc 100644 --- a/net/mac80211/ieee80211.c +++ b/net/mac80211/ieee80211.c @@ -4986,8 +4986,8 @@ int ieee80211_register_hw(struct ieee80211_hw *hw) * and we need some headroom for passing the frame to monitor * interfaces, but never both at the same time. */ - local->tx_headroom = max(local->hw.extra_tx_headroom, - sizeof(struct ieee80211_tx_status_rtap_hdr)); + local->tx_headroom = max_t(unsigned int , local->hw.extra_tx_headroom, + sizeof(struct ieee80211_tx_status_rtap_hdr)); debugfs_hw_add(local); @@ -5095,7 +5095,7 @@ int ieee80211_register_hwmode(struct ieee80211_hw *hw, } if (!(hw->flags & IEEE80211_HW_DEFAULT_REG_DOMAIN_CONFIGURED)) - ieee80211_init_client(local->mdev); + ieee80211_set_default_regdomain(mode); return 0; } @@ -5246,6 +5246,7 @@ static int __init ieee80211_init(void) } ieee80211_debugfs_netdev_init(); + ieee80211_regdomain_init(); return 0; } diff --git a/net/mac80211/ieee80211_i.h b/net/mac80211/ieee80211_i.h index 055a2a91218..6f7bae7ef9c 100644 --- a/net/mac80211/ieee80211_i.h +++ b/net/mac80211/ieee80211_i.h @@ -759,7 +759,6 @@ void ieee80211_update_default_wep_only(struct ieee80211_local *local); /* ieee80211_ioctl.c */ int ieee80211_set_compression(struct ieee80211_local *local, struct net_device *dev, struct sta_info *sta); -int ieee80211_init_client(struct net_device *dev); int ieee80211_set_channel(struct ieee80211_local *local, int channel, int freq); /* ieee80211_sta.c */ void ieee80211_sta_timer(unsigned long data); @@ -798,6 +797,10 @@ void ieee80211_if_sdata_init(struct ieee80211_sub_if_data *sdata); int ieee80211_if_add_mgmt(struct ieee80211_local *local); void ieee80211_if_del_mgmt(struct ieee80211_local *local); +/* regdomain.c */ +void ieee80211_regdomain_init(void); +void ieee80211_set_default_regdomain(struct ieee80211_hw_mode *mode); + /* for wiphy privid */ extern void *mac80211_wiphy_privid; diff --git a/net/mac80211/ieee80211_ioctl.c b/net/mac80211/ieee80211_ioctl.c index 5918dd079e1..d0e1ab5589d 100644 --- a/net/mac80211/ieee80211_ioctl.c +++ b/net/mac80211/ieee80211_ioctl.c @@ -27,20 +27,6 @@ #include "aes_ccm.h" #include "debugfs_key.h" -static int ieee80211_regdom = 0x10; /* FCC */ -module_param(ieee80211_regdom, int, 0444); -MODULE_PARM_DESC(ieee80211_regdom, "IEEE 802.11 regulatory domain; 64=MKK"); - -/* - * If firmware is upgraded by the vendor, additional channels can be used based - * on the new Japanese regulatory rules. This is indicated by setting - * ieee80211_japan_5ghz module parameter to one when loading the 80211 kernel - * module. - */ -static int ieee80211_japan_5ghz /* = 0 */; -module_param(ieee80211_japan_5ghz, int, 0444); -MODULE_PARM_DESC(ieee80211_japan_5ghz, "Vendor-updated firmware for 5 GHz"); - static void ieee80211_set_hw_encryption(struct net_device *dev, struct sta_info *sta, u8 addr[ETH_ALEN], struct ieee80211_key *key) @@ -412,125 +398,6 @@ static int ieee80211_ioctl_giwrange(struct net_device *dev, } -struct ieee80211_channel_range { - short start_freq; - short end_freq; - unsigned char power_level; - unsigned char antenna_max; -}; - -static const struct ieee80211_channel_range ieee80211_fcc_channels[] = { - { 2412, 2462, 27, 6 } /* IEEE 802.11b/g, channels 1..11 */, - { 5180, 5240, 17, 6 } /* IEEE 802.11a, channels 36..48 */, - { 5260, 5320, 23, 6 } /* IEEE 802.11a, channels 52..64 */, - { 5745, 5825, 30, 6 } /* IEEE 802.11a, channels 149..165, outdoor */, - { 0 } -}; - -static const struct ieee80211_channel_range ieee80211_mkk_channels[] = { - { 2412, 2472, 20, 6 } /* IEEE 802.11b/g, channels 1..13 */, - { 5170, 5240, 20, 6 } /* IEEE 802.11a, channels 34..48 */, - { 5260, 5320, 20, 6 } /* IEEE 802.11a, channels 52..64 */, - { 0 } -}; - - -static const struct ieee80211_channel_range *channel_range = - ieee80211_fcc_channels; - - -static void ieee80211_unmask_channel(struct net_device *dev, int mode, - struct ieee80211_channel *chan) -{ - int i; - - chan->flag = 0; - - if (ieee80211_regdom == 64 && - (mode == MODE_ATHEROS_TURBO || mode == MODE_ATHEROS_TURBOG)) { - /* Do not allow Turbo modes in Japan. */ - return; - } - - for (i = 0; channel_range[i].start_freq; i++) { - const struct ieee80211_channel_range *r = &channel_range[i]; - if (r->start_freq <= chan->freq && r->end_freq >= chan->freq) { - if (ieee80211_regdom == 64 && !ieee80211_japan_5ghz && - chan->freq >= 5260 && chan->freq <= 5320) { - /* - * Skip new channels in Japan since the - * firmware was not marked having been upgraded - * by the vendor. - */ - continue; - } - - if (ieee80211_regdom == 0x10 && - (chan->freq == 5190 || chan->freq == 5210 || - chan->freq == 5230)) { - /* Skip MKK channels when in FCC domain. */ - continue; - } - - chan->flag |= IEEE80211_CHAN_W_SCAN | - IEEE80211_CHAN_W_ACTIVE_SCAN | - IEEE80211_CHAN_W_IBSS; - chan->power_level = r->power_level; - chan->antenna_max = r->antenna_max; - - if (ieee80211_regdom == 64 && - (chan->freq == 5170 || chan->freq == 5190 || - chan->freq == 5210 || chan->freq == 5230)) { - /* - * New regulatory rules in Japan have backwards - * compatibility with old channels in 5.15-5.25 - * GHz band, but the station is not allowed to - * use active scan on these old channels. - */ - chan->flag &= ~IEEE80211_CHAN_W_ACTIVE_SCAN; - } - - if (ieee80211_regdom == 64 && - (chan->freq == 5260 || chan->freq == 5280 || - chan->freq == 5300 || chan->freq == 5320)) { - /* - * IBSS is not allowed on 5.25-5.35 GHz band - * due to radar detection requirements. - */ - chan->flag &= ~IEEE80211_CHAN_W_IBSS; - } - - break; - } - } -} - - -static int ieee80211_unmask_channels(struct net_device *dev) -{ - struct ieee80211_local *local = wdev_priv(dev->ieee80211_ptr); - struct ieee80211_hw_mode *mode; - int c; - - list_for_each_entry(mode, &local->modes_list, list) { - for (c = 0; c < mode->num_channels; c++) { - ieee80211_unmask_channel(dev, mode->mode, - &mode->channels[c]); - } - } - return 0; -} - - -int ieee80211_init_client(struct net_device *dev) -{ - if (ieee80211_regdom == 0x40) - channel_range = ieee80211_mkk_channels; - ieee80211_unmask_channels(dev); - return 0; -} - - static int ieee80211_ioctl_siwmode(struct net_device *dev, struct iw_request_info *info, __u32 *mode, char *extra) diff --git a/net/mac80211/ieee80211_rate.c b/net/mac80211/ieee80211_rate.c index 16e850864b8..2118de04fc3 100644 --- a/net/mac80211/ieee80211_rate.c +++ b/net/mac80211/ieee80211_rate.c @@ -24,11 +24,10 @@ int ieee80211_rate_control_register(struct rate_control_ops *ops) { struct rate_control_alg *alg; - alg = kmalloc(sizeof(*alg), GFP_KERNEL); + alg = kzalloc(sizeof(*alg), GFP_KERNEL); if (alg == NULL) { return -ENOMEM; } - memset(alg, 0, sizeof(*alg)); alg->ops = ops; mutex_lock(&rate_ctrl_mutex); diff --git a/net/mac80211/ieee80211_sta.c b/net/mac80211/ieee80211_sta.c index ba2bf8f0a34..7ba352e3ffe 100644 --- a/net/mac80211/ieee80211_sta.c +++ b/net/mac80211/ieee80211_sta.c @@ -25,7 +25,6 @@ #include <linux/wireless.h> #include <linux/random.h> #include <linux/etherdevice.h> -#include <linux/rtnetlink.h> #include <net/iw_handler.h> #include <asm/types.h> @@ -1327,10 +1326,9 @@ ieee80211_rx_bss_add(struct net_device *dev, u8 *bssid) struct ieee80211_local *local = wdev_priv(dev->ieee80211_ptr); struct ieee80211_sta_bss *bss; - bss = kmalloc(sizeof(*bss), GFP_ATOMIC); + bss = kzalloc(sizeof(*bss), GFP_ATOMIC); if (!bss) return NULL; - memset(bss, 0, sizeof(*bss)); atomic_inc(&bss->users); atomic_inc(&bss->users); memcpy(bss->bssid, bssid, ETH_ALEN); @@ -2107,12 +2105,9 @@ static int ieee80211_sta_config_auth(struct net_device *dev, struct ieee80211_sta_bss *bss, *selected = NULL; int top_rssi = 0, freq; - rtnl_lock(); - if (!ifsta->auto_channel_sel && !ifsta->auto_bssid_sel && !ifsta->auto_ssid_sel) { ifsta->state = IEEE80211_AUTHENTICATE; - rtnl_unlock(); ieee80211_sta_reset_auth(dev, ifsta); return 0; } @@ -2155,7 +2150,6 @@ static int ieee80211_sta_config_auth(struct net_device *dev, ieee80211_sta_set_bssid(dev, selected->bssid); ieee80211_rx_bss_put(dev, selected); ifsta->state = IEEE80211_AUTHENTICATE; - rtnl_unlock(); ieee80211_sta_reset_auth(dev, ifsta); return 0; } else { @@ -2166,7 +2160,6 @@ static int ieee80211_sta_config_auth(struct net_device *dev, } else ifsta->state = IEEE80211_DISABLED; } - rtnl_unlock(); return -1; } diff --git a/net/mac80211/regdomain.c b/net/mac80211/regdomain.c new file mode 100644 index 00000000000..b697a2afbb4 --- /dev/null +++ b/net/mac80211/regdomain.c @@ -0,0 +1,158 @@ +/* + * Copyright 2002-2005, Instant802 Networks, Inc. + * Copyright 2005-2006, Devicescape Software, 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. + */ + +/* + * This regulatory domain control implementation is known to be incomplete + * and confusing. mac80211 regulatory domain control will be significantly + * reworked in the not-too-distant future. + * + * For now, drivers wishing to control which channels are and aren't available + * are advised as follows: + * - set the IEEE80211_HW_DEFAULT_REG_DOMAIN_CONFIGURED flag + * - continue to include *ALL* possible channels in the modes registered + * through ieee80211_register_hwmode() + * - for each allowable ieee80211_channel structure registered in the above + * call, set the flag member to some meaningful value such as + * IEEE80211_CHAN_W_SCAN | IEEE80211_CHAN_W_ACTIVE_SCAN | + * IEEE80211_CHAN_W_IBSS. + * - leave flag as 0 for non-allowable channels + * + * The usual implementation is for a driver to read a device EEPROM to + * determine which regulatory domain it should be operating under, then + * looking up the allowable channels in a driver-local table, then performing + * the above. + */ + +#include <linux/module.h> +#include <linux/netdevice.h> +#include <net/mac80211.h> +#include "ieee80211_i.h" + +static int ieee80211_regdom = 0x10; /* FCC */ +module_param(ieee80211_regdom, int, 0444); +MODULE_PARM_DESC(ieee80211_regdom, "IEEE 802.11 regulatory domain; 64=MKK"); + +/* + * If firmware is upgraded by the vendor, additional channels can be used based + * on the new Japanese regulatory rules. This is indicated by setting + * ieee80211_japan_5ghz module parameter to one when loading the 80211 kernel + * module. + */ +static int ieee80211_japan_5ghz /* = 0 */; +module_param(ieee80211_japan_5ghz, int, 0444); +MODULE_PARM_DESC(ieee80211_japan_5ghz, "Vendor-updated firmware for 5 GHz"); + + +struct ieee80211_channel_range { + short start_freq; + short end_freq; + unsigned char power_level; + unsigned char antenna_max; +}; + +static const struct ieee80211_channel_range ieee80211_fcc_channels[] = { + { 2412, 2462, 27, 6 } /* IEEE 802.11b/g, channels 1..11 */, + { 5180, 5240, 17, 6 } /* IEEE 802.11a, channels 36..48 */, + { 5260, 5320, 23, 6 } /* IEEE 802.11a, channels 52..64 */, + { 5745, 5825, 30, 6 } /* IEEE 802.11a, channels 149..165, outdoor */, + { 0 } +}; + +static const struct ieee80211_channel_range ieee80211_mkk_channels[] = { + { 2412, 2472, 20, 6 } /* IEEE 802.11b/g, channels 1..13 */, + { 5170, 5240, 20, 6 } /* IEEE 802.11a, channels 34..48 */, + { 5260, 5320, 20, 6 } /* IEEE 802.11a, channels 52..64 */, + { 0 } +}; + + +static const struct ieee80211_channel_range *channel_range = + ieee80211_fcc_channels; + + +static void ieee80211_unmask_channel(int mode, struct ieee80211_channel *chan) +{ + int i; + + chan->flag = 0; + + if (ieee80211_regdom == 64 && + (mode == MODE_ATHEROS_TURBO || mode == MODE_ATHEROS_TURBOG)) { + /* Do not allow Turbo modes in Japan. */ + return; + } + + for (i = 0; channel_range[i].start_freq; i++) { + const struct ieee80211_channel_range *r = &channel_range[i]; + if (r->start_freq <= chan->freq && r->end_freq >= chan->freq) { + if (ieee80211_regdom == 64 && !ieee80211_japan_5ghz && + chan->freq >= 5260 && chan->freq <= 5320) { + /* + * Skip new channels in Japan since the + * firmware was not marked having been upgraded + * by the vendor. + */ + continue; + } + + if (ieee80211_regdom == 0x10 && + (chan->freq == 5190 || chan->freq == 5210 || + chan->freq == 5230)) { + /* Skip MKK channels when in FCC domain. */ + continue; + } + + chan->flag |= IEEE80211_CHAN_W_SCAN | + IEEE80211_CHAN_W_ACTIVE_SCAN | + IEEE80211_CHAN_W_IBSS; + chan->power_level = r->power_level; + chan->antenna_max = r->antenna_max; + + if (ieee80211_regdom == 64 && + (chan->freq == 5170 || chan->freq == 5190 || + chan->freq == 5210 || chan->freq == 5230)) { + /* + * New regulatory rules in Japan have backwards + * compatibility with old channels in 5.15-5.25 + * GHz band, but the station is not allowed to + * use active scan on these old channels. + */ + chan->flag &= ~IEEE80211_CHAN_W_ACTIVE_SCAN; + } + + if (ieee80211_regdom == 64 && + (chan->freq == 5260 || chan->freq == 5280 || + chan->freq == 5300 || chan->freq == 5320)) { + /* + * IBSS is not allowed on 5.25-5.35 GHz band + * due to radar detection requirements. + */ + chan->flag &= ~IEEE80211_CHAN_W_IBSS; + } + + break; + } + } +} + + +void ieee80211_set_default_regdomain(struct ieee80211_hw_mode *mode) +{ + int c; + for (c = 0; c < mode->num_channels; c++) + ieee80211_unmask_channel(mode->mode, &mode->channels[c]); +} + + +void ieee80211_regdomain_init(void) +{ + if (ieee80211_regdom == 0x40) + channel_range = ieee80211_mkk_channels; +} + diff --git a/net/netfilter/nf_conntrack_standalone.c b/net/netfilter/nf_conntrack_standalone.c index ffb6ff8c352..a4ce5e88799 100644 --- a/net/netfilter/nf_conntrack_standalone.c +++ b/net/netfilter/nf_conntrack_standalone.c @@ -181,7 +181,7 @@ static int ct_seq_show(struct seq_file *s, void *v) if (seq_printf(s, "use=%u\n", atomic_read(&conntrack->ct_general.use))) return -ENOSPC; - + return 0; } @@ -198,7 +198,7 @@ static int ct_open(struct inode *inode, struct file *file) struct ct_iter_state *st; int ret; - st = kmalloc(sizeof(struct ct_iter_state), GFP_KERNEL); + st = kzalloc(sizeof(struct ct_iter_state), GFP_KERNEL); if (st == NULL) return -ENOMEM; ret = seq_open(file, &ct_seq_ops); @@ -206,7 +206,6 @@ static int ct_open(struct inode *inode, struct file *file) goto out_free; seq = file->private_data; seq->private = st; - memset(st, 0, sizeof(struct ct_iter_state)); return ret; out_free: kfree(st); diff --git a/net/netfilter/nf_log.c b/net/netfilter/nf_log.c index 94985792b79..d67c4fbf603 100644 --- a/net/netfilter/nf_log.c +++ b/net/netfilter/nf_log.c @@ -9,7 +9,7 @@ #include "nf_internals.h" -/* Internal logging interface, which relies on the real +/* Internal logging interface, which relies on the real LOG target modules */ #define NF_LOG_PREFIXLEN 128 diff --git a/net/netlink/af_netlink.c b/net/netlink/af_netlink.c index 641cfbc278d..5681ce3aebc 100644 --- a/net/netlink/af_netlink.c +++ b/net/netlink/af_netlink.c @@ -62,6 +62,7 @@ #include <net/netlink.h> #define NLGRPSZ(x) (ALIGN(x, sizeof(unsigned long) * 8) / 8) +#define NLGRPLONGS(x) (NLGRPSZ(x)/sizeof(unsigned long)) struct netlink_sock { /* struct sock has to be the first member of netlink_sock */ @@ -314,10 +315,12 @@ netlink_update_listeners(struct sock *sk) unsigned long mask; unsigned int i; - for (i = 0; i < NLGRPSZ(tbl->groups)/sizeof(unsigned long); i++) { + for (i = 0; i < NLGRPLONGS(tbl->groups); i++) { mask = 0; - sk_for_each_bound(sk, node, &tbl->mc_list) - mask |= nlk_sk(sk)->groups[i]; + sk_for_each_bound(sk, node, &tbl->mc_list) { + if (i < NLGRPLONGS(nlk_sk(sk)->ngroups)) + mask |= nlk_sk(sk)->groups[i]; + } tbl->listeners[i] = mask; } /* this function is only called with the netlink table "grabbed", which @@ -555,26 +558,37 @@ netlink_update_subscriptions(struct sock *sk, unsigned int subscriptions) nlk->subscriptions = subscriptions; } -static int netlink_alloc_groups(struct sock *sk) +static int netlink_realloc_groups(struct sock *sk) { struct netlink_sock *nlk = nlk_sk(sk); unsigned int groups; + unsigned long *new_groups; int err = 0; - netlink_lock_table(); + netlink_table_grab(); + groups = nl_table[sk->sk_protocol].groups; - if (!nl_table[sk->sk_protocol].registered) + if (!nl_table[sk->sk_protocol].registered) { err = -ENOENT; - netlink_unlock_table(); + goto out_unlock; + } - if (err) - return err; + if (nlk->ngroups >= groups) + goto out_unlock; - nlk->groups = kzalloc(NLGRPSZ(groups), GFP_KERNEL); - if (nlk->groups == NULL) - return -ENOMEM; + new_groups = krealloc(nlk->groups, NLGRPSZ(groups), GFP_ATOMIC); + if (new_groups == NULL) { + err = -ENOMEM; + goto out_unlock; + } + memset((char*)new_groups + NLGRPSZ(nlk->ngroups), 0, + NLGRPSZ(groups) - NLGRPSZ(nlk->ngroups)); + + nlk->groups = new_groups; nlk->ngroups = groups; - return 0; + out_unlock: + netlink_table_ungrab(); + return err; } static int netlink_bind(struct socket *sock, struct sockaddr *addr, int addr_len) @@ -591,11 +605,9 @@ static int netlink_bind(struct socket *sock, struct sockaddr *addr, int addr_len if (nladdr->nl_groups) { if (!netlink_capable(sock, NL_NONROOT_RECV)) return -EPERM; - if (nlk->groups == NULL) { - err = netlink_alloc_groups(sk); - if (err) - return err; - } + err = netlink_realloc_groups(sk); + if (err) + return err; } if (nlk->pid) { @@ -839,10 +851,18 @@ retry: int netlink_has_listeners(struct sock *sk, unsigned int group) { int res = 0; + unsigned long *listeners; BUG_ON(!(nlk_sk(sk)->flags & NETLINK_KERNEL_SOCKET)); + + rcu_read_lock(); + listeners = rcu_dereference(nl_table[sk->sk_protocol].listeners); + if (group - 1 < nl_table[sk->sk_protocol].groups) - res = test_bit(group - 1, nl_table[sk->sk_protocol].listeners); + res = test_bit(group - 1, listeners); + + rcu_read_unlock(); + return res; } EXPORT_SYMBOL_GPL(netlink_has_listeners); @@ -1007,6 +1027,23 @@ void netlink_set_err(struct sock *ssk, u32 pid, u32 group, int code) read_unlock(&nl_table_lock); } +/* must be called with netlink table grabbed */ +static void netlink_update_socket_mc(struct netlink_sock *nlk, + unsigned int group, + int is_new) +{ + int old, new = !!is_new, subscriptions; + + old = test_bit(group - 1, nlk->groups); + subscriptions = nlk->subscriptions - old + new; + if (new) + __set_bit(group - 1, nlk->groups); + else + __clear_bit(group - 1, nlk->groups); + netlink_update_subscriptions(&nlk->sk, subscriptions); + netlink_update_listeners(&nlk->sk); +} + static int netlink_setsockopt(struct socket *sock, int level, int optname, char __user *optval, int optlen) { @@ -1032,27 +1069,16 @@ static int netlink_setsockopt(struct socket *sock, int level, int optname, break; case NETLINK_ADD_MEMBERSHIP: case NETLINK_DROP_MEMBERSHIP: { - unsigned int subscriptions; - int old, new = optname == NETLINK_ADD_MEMBERSHIP ? 1 : 0; - if (!netlink_capable(sock, NL_NONROOT_RECV)) return -EPERM; - if (nlk->groups == NULL) { - err = netlink_alloc_groups(sk); - if (err) - return err; - } + err = netlink_realloc_groups(sk); + if (err) + return err; if (!val || val - 1 >= nlk->ngroups) return -EINVAL; netlink_table_grab(); - old = test_bit(val - 1, nlk->groups); - subscriptions = nlk->subscriptions - old + new; - if (new) - __set_bit(val - 1, nlk->groups); - else - __clear_bit(val - 1, nlk->groups); - netlink_update_subscriptions(sk, subscriptions); - netlink_update_listeners(sk); + netlink_update_socket_mc(nlk, val, + optname == NETLINK_ADD_MEMBERSHIP); netlink_table_ungrab(); err = 0; break; @@ -1328,6 +1354,71 @@ out_sock_release: return NULL; } +/** + * netlink_change_ngroups - change number of multicast groups + * + * This changes the number of multicast groups that are available + * on a certain netlink family. Note that it is not possible to + * change the number of groups to below 32. Also note that it does + * not implicitly call netlink_clear_multicast_users() when the + * number of groups is reduced. + * + * @sk: The kernel netlink socket, as returned by netlink_kernel_create(). + * @groups: The new number of groups. + */ +int netlink_change_ngroups(struct sock *sk, unsigned int groups) +{ + unsigned long *listeners, *old = NULL; + struct netlink_table *tbl = &nl_table[sk->sk_protocol]; + int err = 0; + + if (groups < 32) + groups = 32; + + netlink_table_grab(); + if (NLGRPSZ(tbl->groups) < NLGRPSZ(groups)) { + listeners = kzalloc(NLGRPSZ(groups), GFP_ATOMIC); + if (!listeners) { + err = -ENOMEM; + goto out_ungrab; + } + old = tbl->listeners; + memcpy(listeners, old, NLGRPSZ(tbl->groups)); + rcu_assign_pointer(tbl->listeners, listeners); + } + tbl->groups = groups; + + out_ungrab: + netlink_table_ungrab(); + synchronize_rcu(); + kfree(old); + return err; +} +EXPORT_SYMBOL(netlink_change_ngroups); + +/** + * netlink_clear_multicast_users - kick off multicast listeners + * + * This function removes all listeners from the given group. + * @ksk: The kernel netlink socket, as returned by + * netlink_kernel_create(). + * @group: The multicast group to clear. + */ +void netlink_clear_multicast_users(struct sock *ksk, unsigned int group) +{ + struct sock *sk; + struct hlist_node *node; + struct netlink_table *tbl = &nl_table[ksk->sk_protocol]; + + netlink_table_grab(); + + sk_for_each_bound(sk, node, &tbl->mc_list) + netlink_update_socket_mc(nlk_sk(sk), group, 0); + + netlink_table_ungrab(); +} +EXPORT_SYMBOL(netlink_clear_multicast_users); + void netlink_set_nonroot(int protocol, unsigned int flags) { if ((unsigned int)protocol < MAX_LINKS) diff --git a/net/netlink/genetlink.c b/net/netlink/genetlink.c index b9ab62f938d..e146531faf1 100644 --- a/net/netlink/genetlink.c +++ b/net/netlink/genetlink.c @@ -3,6 +3,7 @@ * * Authors: Jamal Hadi Salim * Thomas Graf <tgraf@suug.ch> + * Johannes Berg <johannes@sipsolutions.net> */ #include <linux/module.h> @@ -13,6 +14,7 @@ #include <linux/string.h> #include <linux/skbuff.h> #include <linux/mutex.h> +#include <linux/bitmap.h> #include <net/sock.h> #include <net/genetlink.h> @@ -42,6 +44,16 @@ static void genl_unlock(void) #define GENL_FAM_TAB_MASK (GENL_FAM_TAB_SIZE - 1) static struct list_head family_ht[GENL_FAM_TAB_SIZE]; +/* + * Bitmap of multicast groups that are currently in use. + * + * To avoid an allocation at boot of just one unsigned long, + * declare it global instead. + * Bit 0 is marked as already used since group 0 is invalid. + */ +static unsigned long mc_group_start = 0x1; +static unsigned long *mc_groups = &mc_group_start; +static unsigned long mc_groups_longs = 1; static int genl_ctrl_event(int event, void *data); @@ -116,6 +128,114 @@ static inline u16 genl_generate_id(void) return id_gen_idx; } +static struct genl_multicast_group notify_grp; + +/** + * genl_register_mc_group - register a multicast group + * + * Registers the specified multicast group and notifies userspace + * about the new group. + * + * Returns 0 on success or a negative error code. + * + * @family: The generic netlink family the group shall be registered for. + * @grp: The group to register, must have a name. + */ +int genl_register_mc_group(struct genl_family *family, + struct genl_multicast_group *grp) +{ + int id; + unsigned long *new_groups; + int err; + + BUG_ON(grp->name[0] == '\0'); + + genl_lock(); + + /* special-case our own group */ + if (grp == ¬ify_grp) + id = GENL_ID_CTRL; + else + id = find_first_zero_bit(mc_groups, + mc_groups_longs * BITS_PER_LONG); + + + if (id >= mc_groups_longs * BITS_PER_LONG) { + size_t nlen = (mc_groups_longs + 1) * sizeof(unsigned long); + + if (mc_groups == &mc_group_start) { + new_groups = kzalloc(nlen, GFP_KERNEL); + if (!new_groups) { + err = -ENOMEM; + goto out; + } + mc_groups = new_groups; + *mc_groups = mc_group_start; + } else { + new_groups = krealloc(mc_groups, nlen, GFP_KERNEL); + if (!new_groups) { + err = -ENOMEM; + goto out; + } + mc_groups = new_groups; + mc_groups[mc_groups_longs] = 0; + } + mc_groups_longs++; + } + + err = netlink_change_ngroups(genl_sock, + sizeof(unsigned long) * NETLINK_GENERIC); + if (err) + goto out; + + grp->id = id; + set_bit(id, mc_groups); + list_add_tail(&grp->list, &family->mcast_groups); + grp->family = family; + + genl_ctrl_event(CTRL_CMD_NEWMCAST_GRP, grp); + out: + genl_unlock(); + return 0; +} +EXPORT_SYMBOL(genl_register_mc_group); + +/** + * genl_unregister_mc_group - unregister a multicast group + * + * Unregisters the specified multicast group and notifies userspace + * about it. All current listeners on the group are removed. + * + * Note: It is not necessary to unregister all multicast groups before + * unregistering the family, unregistering the family will cause + * all assigned multicast groups to be unregistered automatically. + * + * @family: Generic netlink family the group belongs to. + * @grp: The group to unregister, must have been registered successfully + * previously. + */ +void genl_unregister_mc_group(struct genl_family *family, + struct genl_multicast_group *grp) +{ + BUG_ON(grp->family != family); + genl_lock(); + netlink_clear_multicast_users(genl_sock, grp->id); + clear_bit(grp->id, mc_groups); + list_del(&grp->list); + genl_ctrl_event(CTRL_CMD_DELMCAST_GRP, grp); + grp->id = 0; + grp->family = NULL; + genl_unlock(); +} + +static void genl_unregister_mc_groups(struct genl_family *family) +{ + struct genl_multicast_group *grp, *tmp; + + list_for_each_entry_safe(grp, tmp, &family->mcast_groups, list) + genl_unregister_mc_group(family, grp); +} + /** * genl_register_ops - register generic netlink operations * @family: generic netlink family @@ -216,6 +336,7 @@ int genl_register_family(struct genl_family *family) goto errout; INIT_LIST_HEAD(&family->ops_list); + INIT_LIST_HEAD(&family->mcast_groups); genl_lock(); @@ -275,6 +396,8 @@ int genl_unregister_family(struct genl_family *family) { struct genl_family *rc; + genl_unregister_mc_groups(family); + genl_lock(); list_for_each_entry(rc, genl_family_chain(family->id), family_list) { @@ -410,6 +533,67 @@ static int ctrl_fill_info(struct genl_family *family, u32 pid, u32 seq, nla_nest_end(skb, nla_ops); } + if (!list_empty(&family->mcast_groups)) { + struct genl_multicast_group *grp; + struct nlattr *nla_grps; + int idx = 1; + + nla_grps = nla_nest_start(skb, CTRL_ATTR_MCAST_GROUPS); + if (nla_grps == NULL) + goto nla_put_failure; + + list_for_each_entry(grp, &family->mcast_groups, list) { + struct nlattr *nest; + + nest = nla_nest_start(skb, idx++); + if (nest == NULL) + goto nla_put_failure; + + NLA_PUT_U32(skb, CTRL_ATTR_MCAST_GRP_ID, grp->id); + NLA_PUT_STRING(skb, CTRL_ATTR_MCAST_GRP_NAME, + grp->name); + + nla_nest_end(skb, nest); + } + nla_nest_end(skb, nla_grps); + } + + return genlmsg_end(skb, hdr); + +nla_put_failure: + return genlmsg_cancel(skb, hdr); +} + +static int ctrl_fill_mcgrp_info(struct genl_multicast_group *grp, u32 pid, + u32 seq, u32 flags, struct sk_buff *skb, + u8 cmd) +{ + void *hdr; + struct nlattr *nla_grps; + struct nlattr *nest; + + hdr = genlmsg_put(skb, pid, seq, &genl_ctrl, flags, cmd); + if (hdr == NULL) + return -1; + + NLA_PUT_STRING(skb, CTRL_ATTR_FAMILY_NAME, grp->family->name); + NLA_PUT_U16(skb, CTRL_ATTR_FAMILY_ID, grp->family->id); + + nla_grps = nla_nest_start(skb, CTRL_ATTR_MCAST_GROUPS); + if (nla_grps == NULL) + goto nla_put_failure; + + nest = nla_nest_start(skb, 1); + if (nest == NULL) + goto nla_put_failure; + + NLA_PUT_U32(skb, CTRL_ATTR_MCAST_GRP_ID, grp->id); + NLA_PUT_STRING(skb, CTRL_ATTR_MCAST_GRP_NAME, + grp->name); + + nla_nest_end(skb, nest); + nla_nest_end(skb, nla_grps); + return genlmsg_end(skb, hdr); nla_put_failure: @@ -453,8 +637,8 @@ errout: return skb->len; } -static struct sk_buff *ctrl_build_msg(struct genl_family *family, u32 pid, - int seq, u8 cmd) +static struct sk_buff *ctrl_build_family_msg(struct genl_family *family, + u32 pid, int seq, u8 cmd) { struct sk_buff *skb; int err; @@ -472,6 +656,25 @@ static struct sk_buff *ctrl_build_msg(struct genl_family *family, u32 pid, return skb; } +static struct sk_buff *ctrl_build_mcgrp_msg(struct genl_multicast_group *grp, + u32 pid, int seq, u8 cmd) +{ + struct sk_buff *skb; + int err; + + skb = nlmsg_new(NLMSG_DEFAULT_SIZE, GFP_KERNEL); + if (skb == NULL) + return ERR_PTR(-ENOBUFS); + + err = ctrl_fill_mcgrp_info(grp, pid, seq, 0, skb, cmd); + if (err < 0) { + nlmsg_free(skb); + return ERR_PTR(err); + } + + return skb; +} + static const struct nla_policy ctrl_policy[CTRL_ATTR_MAX+1] = { [CTRL_ATTR_FAMILY_ID] = { .type = NLA_U16 }, [CTRL_ATTR_FAMILY_NAME] = { .type = NLA_NUL_STRING, @@ -501,8 +704,8 @@ static int ctrl_getfamily(struct sk_buff *skb, struct genl_info *info) goto errout; } - msg = ctrl_build_msg(res, info->snd_pid, info->snd_seq, - CTRL_CMD_NEWFAMILY); + msg = ctrl_build_family_msg(res, info->snd_pid, info->snd_seq, + CTRL_CMD_NEWFAMILY); if (IS_ERR(msg)) { err = PTR_ERR(msg); goto errout; @@ -523,7 +726,15 @@ static int genl_ctrl_event(int event, void *data) switch (event) { case CTRL_CMD_NEWFAMILY: case CTRL_CMD_DELFAMILY: - msg = ctrl_build_msg(data, 0, 0, event); + msg = ctrl_build_family_msg(data, 0, 0, event); + if (IS_ERR(msg)) + return PTR_ERR(msg); + + genlmsg_multicast(msg, 0, GENL_ID_CTRL, GFP_KERNEL); + break; + case CTRL_CMD_NEWMCAST_GRP: + case CTRL_CMD_DELMCAST_GRP: + msg = ctrl_build_mcgrp_msg(data, 0, 0, event); if (IS_ERR(msg)) return PTR_ERR(msg); @@ -541,6 +752,10 @@ static struct genl_ops genl_ctrl_ops = { .policy = ctrl_policy, }; +static struct genl_multicast_group notify_grp = { + .name = "notify", +}; + static int __init genl_init(void) { int i, err; @@ -557,11 +772,17 @@ static int __init genl_init(void) goto errout_register; netlink_set_nonroot(NETLINK_GENERIC, NL_NONROOT_RECV); - genl_sock = netlink_kernel_create(NETLINK_GENERIC, GENL_MAX_ID, - genl_rcv, NULL, THIS_MODULE); + + /* we'll bump the group number right afterwards */ + genl_sock = netlink_kernel_create(NETLINK_GENERIC, 0, genl_rcv, + NULL, THIS_MODULE); if (genl_sock == NULL) panic("GENL: Cannot initialize generic netlink\n"); + err = genl_register_mc_group(&genl_ctrl, ¬ify_grp); + if (err < 0) + goto errout_register; + return 0; errout_register: diff --git a/net/netrom/af_netrom.c b/net/netrom/af_netrom.c index 5d66490dd29..dc9273295a3 100644 --- a/net/netrom/af_netrom.c +++ b/net/netrom/af_netrom.c @@ -720,7 +720,7 @@ static int nr_connect(struct socket *sock, struct sockaddr *uaddr, for (;;) { prepare_to_wait(sk->sk_sleep, &wait, - TASK_INTERRUPTIBLE); + TASK_INTERRUPTIBLE); if (sk->sk_state != TCP_SYN_SENT) break; if (!signal_pending(current)) { diff --git a/net/packet/af_packet.c b/net/packet/af_packet.c index 7c27bd389b7..1322d62b5d9 100644 --- a/net/packet/af_packet.c +++ b/net/packet/af_packet.c @@ -108,7 +108,7 @@ Outgoing, dev->hard_header!=NULL Incoming, dev->hard_header==NULL mac_header -> UNKNOWN position. It is very likely, that it points to ll header. PPP makes it, that is wrong, because introduce - assymetry between rx and tx paths. + assymetry between rx and tx paths. data -> data Outgoing, dev->hard_header==NULL diff --git a/net/rfkill/rfkill-input.c b/net/rfkill/rfkill-input.c index 230e35c5978..9f746be5885 100644 --- a/net/rfkill/rfkill-input.c +++ b/net/rfkill/rfkill-input.c @@ -83,7 +83,7 @@ static DEFINE_RFKILL_TASK(rfkill_wlan, RFKILL_TYPE_WLAN); static DEFINE_RFKILL_TASK(rfkill_bt, RFKILL_TYPE_BLUETOOTH); static void rfkill_event(struct input_handle *handle, unsigned int type, - unsigned int code, int down) + unsigned int code, int down) { if (type == EV_KEY && down == 1) { switch (code) { diff --git a/net/rfkill/rfkill.c b/net/rfkill/rfkill.c index f3986d498b4..db3395bfbcf 100644 --- a/net/rfkill/rfkill.c +++ b/net/rfkill/rfkill.c @@ -187,7 +187,7 @@ static ssize_t rfkill_claim_store(struct device *dev, static struct device_attribute rfkill_dev_attrs[] = { __ATTR(name, S_IRUGO, rfkill_name_show, NULL), __ATTR(type, S_IRUGO, rfkill_type_show, NULL), - __ATTR(state, S_IRUGO, rfkill_state_show, rfkill_state_store), + __ATTR(state, S_IRUGO|S_IWUSR, rfkill_state_show, rfkill_state_store), __ATTR(claim, S_IRUGO|S_IWUSR, rfkill_claim_show, rfkill_claim_store), __ATTR_NULL }; diff --git a/net/rose/af_rose.c b/net/rose/af_rose.c index f4d3aba0080..976c3cc86a2 100644 --- a/net/rose/af_rose.c +++ b/net/rose/af_rose.c @@ -816,7 +816,7 @@ rose_try_next_neigh: for (;;) { prepare_to_wait(sk->sk_sleep, &wait, - TASK_INTERRUPTIBLE); + TASK_INTERRUPTIBLE); if (sk->sk_state != TCP_SYN_SENT) break; if (!signal_pending(current)) { diff --git a/net/rxrpc/af_rxrpc.c b/net/rxrpc/af_rxrpc.c index 2c57df9c131..46f6d572ad2 100644 --- a/net/rxrpc/af_rxrpc.c +++ b/net/rxrpc/af_rxrpc.c @@ -805,26 +805,26 @@ static int __init af_rxrpc_init(void) } ret = proto_register(&rxrpc_proto, 1); - if (ret < 0) { - printk(KERN_CRIT "RxRPC: Cannot register protocol\n"); + if (ret < 0) { + printk(KERN_CRIT "RxRPC: Cannot register protocol\n"); goto error_proto; } ret = sock_register(&rxrpc_family_ops); if (ret < 0) { - printk(KERN_CRIT "RxRPC: Cannot register socket family\n"); + printk(KERN_CRIT "RxRPC: Cannot register socket family\n"); goto error_sock; } ret = register_key_type(&key_type_rxrpc); if (ret < 0) { - printk(KERN_CRIT "RxRPC: Cannot register client key type\n"); + printk(KERN_CRIT "RxRPC: Cannot register client key type\n"); goto error_key_type; } ret = register_key_type(&key_type_rxrpc_s); if (ret < 0) { - printk(KERN_CRIT "RxRPC: Cannot register server key type\n"); + printk(KERN_CRIT "RxRPC: Cannot register server key type\n"); goto error_key_type_s; } diff --git a/net/sctp/sm_statefuns.c b/net/sctp/sm_statefuns.c index f02ce3dddb7..fd2dfdd7d7f 100644 --- a/net/sctp/sm_statefuns.c +++ b/net/sctp/sm_statefuns.c @@ -1779,7 +1779,7 @@ static sctp_disposition_t sctp_sf_do_dupcook_d(const struct sctp_endpoint *ep, SCTP_COMM_UP, 0, asoc->c.sinit_num_ostreams, asoc->c.sinit_max_instreams, - NULL, GFP_ATOMIC); + NULL, GFP_ATOMIC); if (!ev) goto nomem; diff --git a/net/sctp/socket.c b/net/sctp/socket.c index b1917f68723..ee88f2ea510 100644 --- a/net/sctp/socket.c +++ b/net/sctp/socket.c @@ -4803,7 +4803,7 @@ static int sctp_getsockopt_partial_delivery_point(struct sock *sk, int len, char __user *optval, int __user *optlen) { - u32 val; + u32 val; if (len < sizeof(u32)) return -EINVAL; @@ -4827,7 +4827,7 @@ static int sctp_getsockopt_maxburst(struct sock *sk, int len, char __user *optval, int __user *optlen) { - int val; + int val; if (len < sizeof(int)) return -EINVAL; diff --git a/net/sunrpc/auth.c b/net/sunrpc/auth.c index 29a8ecc6092..1ea27559b1d 100644 --- a/net/sunrpc/auth.c +++ b/net/sunrpc/auth.c @@ -13,7 +13,6 @@ #include <linux/errno.h> #include <linux/sunrpc/clnt.h> #include <linux/spinlock.h> -#include <linux/smp_lock.h> #ifdef RPC_DEBUG # define RPCDBG_FACILITY RPCDBG_AUTH @@ -476,17 +475,13 @@ rpcauth_wrap_req(struct rpc_task *task, kxdrproc_t encode, void *rqstp, __be32 *data, void *obj) { struct rpc_cred *cred = task->tk_msg.rpc_cred; - int ret; dprintk("RPC: %5u using %s cred %p to wrap rpc data\n", task->tk_pid, cred->cr_ops->cr_name, cred); if (cred->cr_ops->crwrap_req) return cred->cr_ops->crwrap_req(task, encode, rqstp, data, obj); /* By default, we encode the arguments normally. */ - lock_kernel(); - ret = encode(rqstp, data, obj); - unlock_kernel(); - return ret; + return rpc_call_xdrproc(encode, rqstp, data, obj); } int @@ -494,7 +489,6 @@ rpcauth_unwrap_resp(struct rpc_task *task, kxdrproc_t decode, void *rqstp, __be32 *data, void *obj) { struct rpc_cred *cred = task->tk_msg.rpc_cred; - int ret; dprintk("RPC: %5u using %s cred %p to unwrap rpc data\n", task->tk_pid, cred->cr_ops->cr_name, cred); @@ -502,10 +496,7 @@ rpcauth_unwrap_resp(struct rpc_task *task, kxdrproc_t decode, void *rqstp, return cred->cr_ops->crunwrap_resp(task, decode, rqstp, data, obj); /* By default, we decode the arguments normally. */ - lock_kernel(); - ret = decode(rqstp, data, obj); - unlock_kernel(); - return ret; + return rpc_call_xdrproc(decode, rqstp, data, obj); } int diff --git a/net/sunrpc/auth_gss/auth_gss.c b/net/sunrpc/auth_gss/auth_gss.c index abfda33bac6..4bbc59cc237 100644 --- a/net/sunrpc/auth_gss/auth_gss.c +++ b/net/sunrpc/auth_gss/auth_gss.c @@ -43,7 +43,6 @@ #include <linux/types.h> #include <linux/slab.h> #include <linux/sched.h> -#include <linux/smp_lock.h> #include <linux/pagemap.h> #include <linux/sunrpc/clnt.h> #include <linux/sunrpc/auth.h> @@ -1000,9 +999,7 @@ gss_wrap_req_integ(struct rpc_cred *cred, struct gss_cl_ctx *ctx, offset = (u8 *)p - (u8 *)snd_buf->head[0].iov_base; *p++ = htonl(rqstp->rq_seqno); - lock_kernel(); - status = encode(rqstp, p, obj); - unlock_kernel(); + status = rpc_call_xdrproc(encode, rqstp, p, obj); if (status) return status; @@ -1096,9 +1093,7 @@ gss_wrap_req_priv(struct rpc_cred *cred, struct gss_cl_ctx *ctx, offset = (u8 *)p - (u8 *)snd_buf->head[0].iov_base; *p++ = htonl(rqstp->rq_seqno); - lock_kernel(); - status = encode(rqstp, p, obj); - unlock_kernel(); + status = rpc_call_xdrproc(encode, rqstp, p, obj); if (status) return status; @@ -1157,16 +1152,12 @@ gss_wrap_req(struct rpc_task *task, /* The spec seems a little ambiguous here, but I think that not * wrapping context destruction requests makes the most sense. */ - lock_kernel(); - status = encode(rqstp, p, obj); - unlock_kernel(); + status = rpc_call_xdrproc(encode, rqstp, p, obj); goto out; } switch (gss_cred->gc_service) { case RPC_GSS_SVC_NONE: - lock_kernel(); - status = encode(rqstp, p, obj); - unlock_kernel(); + status = rpc_call_xdrproc(encode, rqstp, p, obj); break; case RPC_GSS_SVC_INTEGRITY: status = gss_wrap_req_integ(cred, ctx, encode, @@ -1282,9 +1273,7 @@ gss_unwrap_resp(struct rpc_task *task, cred->cr_auth->au_rslack = cred->cr_auth->au_verfsize + (p - savedp) + (savedlen - head->iov_len); out_decode: - lock_kernel(); - status = decode(rqstp, p, obj); - unlock_kernel(); + status = rpc_call_xdrproc(decode, rqstp, p, obj); out: gss_put_ctx(ctx); dprintk("RPC: %5u gss_unwrap_resp returning %d\n", task->tk_pid, diff --git a/net/sunrpc/auth_gss/gss_krb5_crypto.c b/net/sunrpc/auth_gss/gss_krb5_crypto.c index f441aa0b26d..bfb6a29633d 100644 --- a/net/sunrpc/auth_gss/gss_krb5_crypto.c +++ b/net/sunrpc/auth_gss/gss_krb5_crypto.c @@ -67,7 +67,7 @@ krb5_encrypt( if (crypto_blkcipher_ivsize(tfm) > 16) { dprintk("RPC: gss_k5encrypt: tfm iv size to large %d\n", - crypto_blkcipher_ivsize(tfm)); + crypto_blkcipher_ivsize(tfm)); goto out; } diff --git a/net/sunrpc/rpc_pipe.c b/net/sunrpc/rpc_pipe.c index e787b6a43ee..5b2b6fb244f 100644 --- a/net/sunrpc/rpc_pipe.c +++ b/net/sunrpc/rpc_pipe.c @@ -460,21 +460,19 @@ static struct dentry_operations rpc_dentry_operations = { static int rpc_lookup_parent(char *path, struct nameidata *nd) { + struct vfsmount *mnt; + if (path[0] == '\0') return -ENOENT; - nd->mnt = rpc_get_mount(); - if (IS_ERR(nd->mnt)) { + + mnt = rpc_get_mount(); + if (IS_ERR(mnt)) { printk(KERN_WARNING "%s: %s failed to mount " "pseudofilesystem \n", __FILE__, __FUNCTION__); - return PTR_ERR(nd->mnt); + return PTR_ERR(mnt); } - mntget(nd->mnt); - nd->dentry = dget(rpc_mount->mnt_root); - nd->last_type = LAST_ROOT; - nd->flags = LOOKUP_PARENT; - nd->depth = 0; - if (path_walk(path, nd)) { + if (vfs_path_lookup(mnt->mnt_root, mnt, path, LOOKUP_PARENT, nd)) { printk(KERN_WARNING "%s: %s failed to find path %s\n", __FILE__, __FUNCTION__, path); rpc_put_mount(); diff --git a/net/tipc/name_table.c b/net/tipc/name_table.c index 9dfc9127acd..d8473eefcd2 100644 --- a/net/tipc/name_table.c +++ b/net/tipc/name_table.c @@ -1052,12 +1052,11 @@ int tipc_nametbl_init(void) { int array_size = sizeof(struct hlist_head) * tipc_nametbl_size; - table.types = kmalloc(array_size, GFP_ATOMIC); + table.types = kzalloc(array_size, GFP_ATOMIC); if (!table.types) return -ENOMEM; write_lock_bh(&tipc_nametbl_lock); - memset(table.types, 0, array_size); table.local_publ_count = 0; write_unlock_bh(&tipc_nametbl_lock); return 0; diff --git a/net/tipc/socket.c b/net/tipc/socket.c index 4a8f37f4876..84110172031 100644 --- a/net/tipc/socket.c +++ b/net/tipc/socket.c @@ -1629,8 +1629,8 @@ static struct proto_ops msg_ops = { .getsockopt = getsockopt, .sendmsg = send_msg, .recvmsg = recv_msg, - .mmap = sock_no_mmap, - .sendpage = sock_no_sendpage + .mmap = sock_no_mmap, + .sendpage = sock_no_sendpage }; static struct proto_ops packet_ops = { @@ -1650,8 +1650,8 @@ static struct proto_ops packet_ops = { .getsockopt = getsockopt, .sendmsg = send_packet, .recvmsg = recv_msg, - .mmap = sock_no_mmap, - .sendpage = sock_no_sendpage + .mmap = sock_no_mmap, + .sendpage = sock_no_sendpage }; static struct proto_ops stream_ops = { @@ -1671,8 +1671,8 @@ static struct proto_ops stream_ops = { .getsockopt = getsockopt, .sendmsg = send_stream, .recvmsg = recv_stream, - .mmap = sock_no_mmap, - .sendpage = sock_no_sendpage + .mmap = sock_no_mmap, + .sendpage = sock_no_sendpage }; static struct net_proto_family tipc_family_ops = { diff --git a/net/xfrm/xfrm_policy.c b/net/xfrm/xfrm_policy.c index b48f06fc9fd..cfaf17c8851 100644 --- a/net/xfrm/xfrm_policy.c +++ b/net/xfrm/xfrm_policy.c @@ -857,7 +857,7 @@ xfrm_policy_flush_secctx_check(u8 type, struct xfrm_audit *audit_info) pol, NULL); return err; } - } + } for (i = xfrm_policy_bydst[dir].hmask; i >= 0; i--) { hlist_for_each_entry(pol, entry, xfrm_policy_bydst[dir].table + i, diff --git a/net/xfrm/xfrm_state.c b/net/xfrm/xfrm_state.c index e070c3f938f..38f90ca75b1 100644 --- a/net/xfrm/xfrm_state.c +++ b/net/xfrm/xfrm_state.c @@ -407,7 +407,7 @@ xfrm_state_flush_secctx_check(u8 proto, struct xfrm_audit *audit_info) xfrm_audit_log(audit_info->loginuid, audit_info->secid, AUDIT_MAC_IPSEC_DELSA, - 0, NULL, x); + 0, NULL, x); return err; } diff --git a/scripts/Kbuild.include b/scripts/Kbuild.include index 06c1a377c4c..677bc6c175c 100644 --- a/scripts/Kbuild.include +++ b/scripts/Kbuild.include @@ -100,9 +100,14 @@ cc-option-align = $(subst -functions=0,,\ $(call cc-option,-falign-functions=0,-malign-functions=0)) # cc-version -# Usage gcc-ver := $(call cc-version,$(CC)) +# Usage gcc-ver := $(call cc-version) cc-version = $(shell $(CONFIG_SHELL) $(srctree)/scripts/gcc-version.sh $(CC)) +# cc-fullversion +# Usage gcc-ver := $(call cc-fullversion) +cc-fullversion = $(shell $(CONFIG_SHELL) \ + $(srctree)/scripts/gcc-version.sh -p $(CC)) + # cc-ifversion # Usage: EXTRA_CFLAGS += $(call cc-ifversion, -lt, 0402, -O1) cc-ifversion = $(shell [ $(call cc-version, $(CC)) $(1) $(2) ] && echo $(3)) diff --git a/scripts/Makefile.build b/scripts/Makefile.build index a525112847f..3f7b451f395 100644 --- a/scripts/Makefile.build +++ b/scripts/Makefile.build @@ -7,6 +7,22 @@ src := $(obj) PHONY := __build __build: +# Init all relevant variables used in kbuild files so +# 1) they have correct type +# 2) they do not inherit any value from the environment +obj-y := +obj-m := +lib-y := +lib-m := +always := +targets := +subdir-y := +subdir-m := +EXTRA_AFLAGS := +EXTRA_CFLAGS := +EXTRA_CPPFLAGS := +EXTRA_LDFLAGS := + # Read .config if it exist, otherwise ignore -include include/config/auto.conf diff --git a/scripts/Makefile.headersinst b/scripts/Makefile.headersinst index f98d772aac8..53dae3eb3d1 100644 --- a/scripts/Makefile.headersinst +++ b/scripts/Makefile.headersinst @@ -11,13 +11,13 @@ UNIFDEF := scripts/unifdef -U__KERNEL__ # Eliminate the contents of (and inclusions of) compiler.h HDRSED := sed -e "s/ inline / __inline__ /g" \ - -e "s/[[:space:]]__user[[:space:]]\+/ /g" \ - -e "s/(__user[[:space:]]\+/ (/g" \ - -e "s/[[:space:]]__force[[:space:]]\+/ /g" \ - -e "s/(__force[[:space:]]\+/ (/g" \ - -e "s/[[:space:]]__iomem[[:space:]]\+/ /g" \ - -e "s/(__iomem[[:space:]]\+/ (/g" \ - -e "s/[[:space:]]__attribute_const__[[:space:]]\+/\ /g" \ + -e "s/[[:space:]]__user[[:space:]]\{1,\}/ /g" \ + -e "s/(__user[[:space:]]\{1,\}/ (/g" \ + -e "s/[[:space:]]__force[[:space:]]\{1,\}/ /g" \ + -e "s/(__force[[:space:]]\{1,\}/ (/g" \ + -e "s/[[:space:]]__iomem[[:space:]]\{1,\}/ /g" \ + -e "s/(__iomem[[:space:]]\{1,\}/ (/g" \ + -e "s/[[:space:]]__attribute_const__[[:space:]]\{1,\}/\ /g" \ -e "s/[[:space:]]__attribute_const__$$//" \ -e "/^\#include <linux\/compiler.h>/d" diff --git a/scripts/Makefile.modpost b/scripts/Makefile.modpost index d5bbbcce31e..c6fcc597b3b 100644 --- a/scripts/Makefile.modpost +++ b/scripts/Makefile.modpost @@ -70,10 +70,10 @@ __modpost: $(modules:.ko=.o) FORCE $(call cmd,modpost) $(wildcard vmlinux) $(filter-out FORCE,$^) quiet_cmd_kernel-mod = MODPOST $@ - cmd_kernel-mod = $(cmd_modpost) $(KBUILD_VMLINUX_OBJS) + cmd_kernel-mod = $(cmd_modpost) $@ PHONY += vmlinux -vmlinux: FORCE +vmlinux.o: FORCE $(call cmd,kernel-mod) # Declare generated files as targets for modpost diff --git a/scripts/checkpatch.pl b/scripts/checkpatch.pl index 25e20a27fc5..73751ab6ec0 100755 --- a/scripts/checkpatch.pl +++ b/scripts/checkpatch.pl @@ -9,7 +9,7 @@ use strict; my $P = $0; $P =~ s@.*/@@g; -my $V = '0.07'; +my $V = '0.08'; use Getopt::Long qw(:config no_auto_abbrev); @@ -47,16 +47,14 @@ my $removal = 'Documentation/feature-removal-schedule.txt'; if ($tree && -f $removal) { open(REMOVE, "<$removal") || die "$P: $removal: open failed - $!\n"; while (<REMOVE>) { - if (/^Files:\s+(.*\S)/) { - for my $file (split(/[, ]+/, $1)) { - if ($file =~ m@include/(.*)@) { + if (/^Check:\s+(.*\S)/) { + for my $entry (split(/[, ]+/, $1)) { + if ($entry =~ m@include/(.*)@) { push(@dep_includes, $1); - } - } - } elsif (/^Funcs:\s+(.*\S)/) { - for my $func (split(/[, ]+/, $1)) { - push(@dep_functions, $func); + } elsif ($entry !~ m@/@) { + push(@dep_functions, $entry); + } } } } @@ -153,7 +151,7 @@ sub sanitise_line { } sub ctx_block_get { - my ($linenr, $remain, $outer, $open, $close) = @_; + my ($linenr, $remain, $outer, $open, $close, $off) = @_; my $line; my $start = $linenr - 1; my $blk = ''; @@ -161,38 +159,58 @@ sub ctx_block_get { my @c; my @res = (); + my $level = 0; for ($line = $start; $remain > 0; $line++) { next if ($rawlines[$line] =~ /^-/); $remain--; $blk .= $rawlines[$line]; + foreach my $c (split(//, $rawlines[$line])) { + ##print "C<$c>L<$level><$open$close>O<$off>\n"; + if ($off > 0) { + $off--; + next; + } - @o = ($blk =~ /$open/g); - @c = ($blk =~ /$close/g); + if ($c eq $close && $level > 0) { + $level--; + last if ($level == 0); + } elsif ($c eq $open) { + $level++; + } + } - if (!$outer || (scalar(@o) - scalar(@c)) == 1) { + if (!$outer || $level <= 1) { push(@res, $rawlines[$line]); } - last if (scalar(@o) == scalar(@c)); + last if ($level == 0); } - return @res; + return ($level, @res); } sub ctx_block_outer { my ($linenr, $remain) = @_; - return ctx_block_get($linenr, $remain, 1, '\{', '\}'); + my ($level, @r) = ctx_block_get($linenr, $remain, 1, '{', '}', 0); + return @r; } sub ctx_block { my ($linenr, $remain) = @_; - return ctx_block_get($linenr, $remain, 0, '\{', '\}'); + my ($level, @r) = ctx_block_get($linenr, $remain, 0, '{', '}', 0); + return @r; } sub ctx_statement { + my ($linenr, $remain, $off) = @_; + + my ($level, @r) = ctx_block_get($linenr, $remain, 0, '(', ')', $off); + return @r; +} +sub ctx_block_level { my ($linenr, $remain) = @_; - return ctx_block_get($linenr, $remain, 0, '\(', '\)'); + return ctx_block_get($linenr, $remain, 0, '{', '}', 0); } sub ctx_locate_comment { @@ -246,16 +264,23 @@ sub cat_vet { return $vet; } +my @report = (); +sub report { + push(@report, $_[0]); +} +sub report_dump { + @report; +} sub ERROR { - print "ERROR: $_[0]\n"; + report("ERROR: $_[0]\n"); our $clean = 0; } sub WARN { - print "WARNING: $_[0]\n"; + report("WARNING: $_[0]\n"); our $clean = 0; } sub CHK { - print "CHECK: $_[0]\n"; + report("CHECK: $_[0]\n"); our $clean = 0; } @@ -318,7 +343,10 @@ sub process { (?:\s*\*+\s*const|\s*\*+)? }x; my $Declare = qr{(?:$Storage\s+)?$Type}; - my $Attribute = qr{__read_mostly|__init|__initdata}; + my $Attribute = qr{const|__read_mostly|__init|__initdata|__meminit}; + + my $Member = qr{->$Ident|\.$Ident|\[[^]]*\]}; + my $Lval = qr{$Ident(?:$Member)*}; # Pre-scan the patch looking for any __setup documentation. my @setup_docs = (); @@ -509,7 +537,7 @@ sub process { # if/while/etc brace do not go on next line, unless defining a do while loop, # or if that brace on the next line is for something else if ($line =~ /\b(?:(if|while|for|switch)\s*\(|do\b|else\b)/ && $line !~ /^.#/) { - my @ctx = ctx_statement($linenr, $realcnt); + my @ctx = ctx_statement($linenr, $realcnt, 0); my $ctx_ln = $linenr + $#ctx + 1; my $ctx_cnt = $realcnt - $#ctx - 1; my $ctx = join("\n", @ctx); @@ -521,7 +549,7 @@ sub process { ##warn "line<$line>\nctx<$ctx>\nnext<$lines[$ctx_ln - 1]>"; if ($ctx !~ /{\s*/ && $ctx_cnt > 0 && $lines[$ctx_ln - 1] =~ /^\+\s*{/) { - ERROR("That { should be on the previous line\n" . + ERROR("That open brace { should be on the previous line\n" . "$here\n$ctx\n$lines[$ctx_ln - 1]"); } } @@ -535,6 +563,12 @@ sub process { next; } +# check for initialisation to aggregates open brace on the next line + if ($prevline =~ /$Declare\s*$Ident\s*=\s*$/ && + $line =~ /^.\s*{/) { + ERROR("That open brace { should be on the previous line\n" . $hereprev); + } + # # Checks which are anchored on the added line. # @@ -570,8 +604,13 @@ sub process { } } +# check for external initialisers. + if ($line =~ /^.$Type\s*$Ident\s*=\s*(0|NULL);/) { + ERROR("do not initialise externals to 0 or NULL\n" . + $herecurr); + } # check for static initialisers. - if ($line=~/\s*static\s.*=\s+(0|NULL);/) { + if ($line =~ /\s*static\s.*=\s*(0|NULL);/) { ERROR("do not initialise statics to 0 or NULL\n" . $herecurr); } @@ -593,11 +632,11 @@ sub process { ERROR("\"(foo $1 )\" should be \"(foo $1)\"\n" . $herecurr); - } elsif ($line =~ m{$NonptrType(\*+)(?:\s+const)?\s+[A-Za-z\d_]+}) { + } elsif ($line =~ m{$NonptrType(\*+)(?:\s+$Attribute)?\s+[A-Za-z\d_]+}) { ERROR("\"foo$1 bar\" should be \"foo $1bar\"\n" . $herecurr); - } elsif ($line =~ m{$NonptrType\s+(\*+)(?!\s+const)\s+[A-Za-z\d_]+}) { + } elsif ($line =~ m{$NonptrType\s+(\*+)(?!\s+$Attribute)\s+[A-Za-z\d_]+}) { ERROR("\"foo $1 bar\" should be \"foo $1bar\"\n" . $herecurr); } @@ -614,7 +653,7 @@ sub process { # to try and find and validate the current printk. In summary the current # printk includes all preceeding printk's which have no newline on the end. # we assume the first bad printk is the one to report. - if ($line =~ /\bprintk\((?!KERN_)/) { + if ($line =~ /\bprintk\((?!KERN_)\s*"/) { my $ok = 0; for (my $ln = $linenr - 1; $ln >= $first_line; $ln--) { #print "CHECK<$lines[$ln - 1]\n"; @@ -639,6 +678,12 @@ sub process { ERROR("open brace '{' following function declarations go on the next line\n" . $herecurr); } +# check for spaces between functions and their parentheses. + if ($line =~ /($Ident)\s+\(/ && + $1 !~ /^(?:if|for|while|switch|return|volatile)$/ && + $line !~ /$Type\s+\(/ && $line !~ /^.\#\s*define\b/) { + ERROR("no space between function name and open parenthesis '('\n" . $herecurr); + } # Check operator spacing. # Note we expand the line with the leading + as the real # line will be displayed with the leading + and the tabs @@ -647,7 +692,7 @@ sub process { $opline = expand_tabs($opline); $opline =~ s/^./ /; if (!($line=~/\#\s*include/)) { - my @elements = split(/(<<=|>>=|<=|>=|==|!=|\+=|-=|\*=|\/=|%=|\^=|\|=|&=|->|<<|>>|<|>|=|!|~|&&|\|\||,|\^|\+\+|--|;|&|\||\+|-|\*|\/\/|\/)/, $opline); + my @elements = split(/(<<=|>>=|<=|>=|==|!=|\+=|-=|\*=|\/=|%=|\^=|\|=|&=|=>|->|<<|>>|<|>|=|!|~|&&|\|\||,|\^|\+\+|--|;|&|\||\+|-|\*|\/\/|\/)/, $opline); my $off = 0; for (my $n = 0; $n < $#elements; $n += 2) { $off += length($elements[$n]); @@ -773,6 +818,18 @@ sub process { } } +# check for multiple assignments + if ($line =~ /^.\s*$Lval\s*=\s*$Lval\s*=(?!=)/) { + WARN("multiple assignments should be avoided\n" . $herecurr); + } + +# check for multiple declarations, allowing for a function declaration +# continuation. + if ($line =~ /^.\s*$Type\s+$Ident(?:\s*=[^,{]*)?\s*,\s*$Ident.*/ && + $line !~ /^.\s*$Type\s+$Ident(?:\s*=[^,{]*)?\s*,\s*$Type\s*$Ident.*/) { + WARN("declaring multiple variables together should be avoided\n" . $herecurr); + } + #need space before brace following if, while, etc if ($line =~ /\(.*\){/ || $line =~ /do{/) { ERROR("need a space before the open brace '{'\n" . $herecurr); @@ -847,13 +904,18 @@ sub process { # or the one below. my $ln = $linenr; my $cnt = $realcnt; + my $off = 0; - # If the macro starts on the define line start there. - if ($prevline !~ m{^.#\s*define\s*$Ident(?:\([^\)]*\))?\s*\\\s*$}) { + # If the macro starts on the define line start + # grabbing the statement after the identifier + $prevline =~ m{^(.#\s*define\s*$Ident(?:\([^\)]*\))?\s*)(.*)\\\s*$}; + ##print "1<$1> 2<$2>\n"; + if ($2 ne '') { + $off = length($1); $ln--; $cnt++; } - my @ctx = ctx_statement($ln, $cnt); + my @ctx = ctx_statement($ln, $cnt, $off); my $ctx_ln = $ln + $#ctx + 1; my $ctx = join("\n", @ctx); @@ -873,6 +935,44 @@ sub process { } } +# check for redundant bracing round if etc + if ($line =~ /\b(if|while|for|else)\b/) { + # Locate the end of the opening statement. + my @control = ctx_statement($linenr, $realcnt, 0); + my $nr = $linenr + (scalar(@control) - 1); + my $cnt = $realcnt - (scalar(@control) - 1); + + my $off = $realcnt - $cnt; + #print "$off: line<$line>end<" . $lines[$nr - 1] . ">\n"; + + # If this is is a braced statement group check it + if ($lines[$nr - 1] =~ /{\s*$/) { + my ($lvl, @block) = ctx_block_level($nr, $cnt); + + my $stmt = join(' ', @block); + $stmt =~ s/^[^{]*{//; + $stmt =~ s/}[^}]*$//; + + #print "block<" . join(' ', @block) . "><" . scalar(@block) . ">\n"; + #print "stmt<$stmt>\n\n"; + + # Count the ;'s if there is fewer than two + # then there can only be one statement, + # if there is a brace inside we cannot + # trivially detect if its one statement. + # Also nested if's often require braces to + # disambiguate the else binding so shhh there. + my @semi = ($stmt =~ /;/g); + ##print "semi<" . scalar(@semi) . ">\n"; + if ($lvl == 0 && scalar(@semi) < 2 && + $stmt !~ /{/ && $stmt !~ /\bif\b/) { + my $herectx = "$here\n" . join("\n", @control, @block[1 .. $#block]) . "\n"; + shift(@block); + ERROR("braces {} are not necessary for single statement blocks\n" . $herectx); + } + } + } + # don't include deprecated include files (uses RAW line) for my $inc (@dep_includes) { if ($rawline =~ m@\#\s*include\s*\<$inc>@) { @@ -898,6 +998,14 @@ sub process { $herecurr); } +# check for needless kfree() checks + if ($prevline =~ /\bif\s*\(([^\)]*)\)/) { + my $expr = $1; + if ($line =~ /\bkfree\(\Q$expr\E\);/) { + WARN("kfree(NULL) is safe this check is probabally not required\n" . $hereprev); + } + } + # warn about #ifdefs in C files # if ($line =~ /^.#\s*if(|n)def/ && ($realfile =~ /\.c$/)) { # print "#ifdef in C files should be avoided\n"; @@ -952,6 +1060,9 @@ sub process { ERROR("Missing Signed-off-by: line(s)\n"); } + if ($clean == 0 && ($chk_patch || $is_patch)) { + print report_dump(); + } if ($clean == 1 && $quiet == 0) { print "Your patch has no obvious style problems and is ready for submission.\n" } diff --git a/scripts/cleanfile b/scripts/cleanfile index f1ba8aa58a4..cefd29e5229 100755 --- a/scripts/cleanfile +++ b/scripts/cleanfile @@ -7,7 +7,9 @@ use bytes; use File::Basename; -# +# Default options +$max_width = 79; + # Clean up space-tab sequences, either by removing spaces or # replacing them with tabs. sub clean_space_tabs($) @@ -48,9 +50,49 @@ sub clean_space_tabs($) return $lo; } +# Compute the visual width of a string +sub strwidth($) { + no bytes; # Tab alignment depends on characters + + my($li) = @_; + my($c, $i); + my $pos = 0; + my $mlen = 0; + + for ($i = 0; $i < length($li); $i++) { + $c = substr($li,$i,1); + if ($c eq "\t") { + $pos = ($pos+8) & ~7; + } elsif ($c eq "\n") { + $mlen = $pos if ($pos > $mlen); + $pos = 0; + } else { + $pos++; + } + } + + $mlen = $pos if ($pos > $mlen); + return $mlen; +} + $name = basename($0); -foreach $f ( @ARGV ) { +@files = (); + +while (defined($a = shift(@ARGV))) { + if ($a =~ /^-/) { + if ($a eq '-width' || $a eq '-w') { + $max_width = shift(@ARGV)+0; + } else { + print STDERR "Usage: $name [-width #] files...\n"; + exit 1; + } + } else { + push(@files, $a); + } +} + +foreach $f ( @files ) { print STDERR "$name: $f\n"; if (! -f $f) { @@ -90,8 +132,10 @@ foreach $f ( @ARGV ) { @blanks = (); @lines = (); + $lineno = 0; while ( defined($line = <FILE>) ) { + $lineno++; $in_bytes += length($line); $line =~ s/[ \t\r]*$//; # Remove trailing spaces $line = clean_space_tabs($line); @@ -107,6 +151,12 @@ foreach $f ( @ARGV ) { @blanks = (); $blank_bytes = 0; } + + $l_width = strwidth($line); + if ($max_width && $l_width > $max_width) { + print STDERR + "$f:$lineno: line exceeds $max_width characters ($l_width)\n"; + } } # Any blanks at the end of the file are discarded diff --git a/scripts/cleanpatch b/scripts/cleanpatch index a53f987708f..9680d03ad2b 100755 --- a/scripts/cleanpatch +++ b/scripts/cleanpatch @@ -7,7 +7,9 @@ use bytes; use File::Basename; -# +# Default options +$max_width = 79; + # Clean up space-tab sequences, either by removing spaces or # replacing them with tabs. sub clean_space_tabs($) @@ -48,9 +50,49 @@ sub clean_space_tabs($) return $lo; } +# Compute the visual width of a string +sub strwidth($) { + no bytes; # Tab alignment depends on characters + + my($li) = @_; + my($c, $i); + my $pos = 0; + my $mlen = 0; + + for ($i = 0; $i < length($li); $i++) { + $c = substr($li,$i,1); + if ($c eq "\t") { + $pos = ($pos+8) & ~7; + } elsif ($c eq "\n") { + $mlen = $pos if ($pos > $mlen); + $pos = 0; + } else { + $pos++; + } + } + + $mlen = $pos if ($pos > $mlen); + return $mlen; +} + $name = basename($0); -foreach $f ( @ARGV ) { +@files = (); + +while (defined($a = shift(@ARGV))) { + if ($a =~ /^-/) { + if ($a eq '-width' || $a eq '-w') { + $max_width = shift(@ARGV)+0; + } else { + print STDERR "Usage: $name [-width #] files...\n"; + exit 1; + } + } else { + push(@files, $a); + } +} + +foreach $f ( @files ) { print STDERR "$name: $f\n"; if (! -f $f) { @@ -86,6 +128,7 @@ foreach $f ( @ARGV ) { $in_bytes = 0; $out_bytes = 0; + $lineno = 0; @lines = (); @@ -93,10 +136,12 @@ foreach $f ( @ARGV ) { $err = 0; while ( defined($line = <FILE>) ) { + $lineno++; $in_bytes += length($line); if (!$in_hunk) { - if ($line =~ /^\@\@\s+\-([0-9]+),([0-9]+)\s+\+([0-9]+),([0-9]+)\s\@\@/) { + if ($line =~ + /^\@\@\s+\-([0-9]+),([0-9]+)\s+\+([0-9]+),([0-9]+)\s\@\@/) { $minus_lines = $2; $plus_lines = $4; if ($minus_lines || $plus_lines) { @@ -117,6 +162,13 @@ foreach $f ( @ARGV ) { $text =~ s/[ \t\r]*$//; # Remove trailing spaces $text = clean_space_tabs($text); + $l_width = strwidth($text); + if ($max_width && $l_width > $max_width) { + print STDERR + "$f:$lineno: adds line exceeds $max_width ", + "characters ($l_width)\n"; + } + push(@hunk_lines, '+'.$text); } elsif ($line =~ /^\-/) { $minus_lines--; diff --git a/scripts/gcc-version.sh b/scripts/gcc-version.sh index bb4fbeab832..8a1d1879c7a 100644 --- a/scripts/gcc-version.sh +++ b/scripts/gcc-version.sh @@ -1,14 +1,23 @@ #!/bin/sh # -# gcc-version gcc-command +# gcc-version [-p] gcc-command # # Prints the gcc version of `gcc-command' in a canonical 4-digit form # such as `0295' for gcc-2.95, `0303' for gcc-3.3, etc. # +# With the -p option, prints the patchlevel as well, for example `029503' for +# gcc-2.95.3, `030301' for gcc-3.3.1, etc. +# + +if [ $1 = "-p" ] ; then with_patchlevel=1; shift; fi compiler="$*" MAJOR=$(echo __GNUC__ | $compiler -E -xc - | tail -n 1) MINOR=$(echo __GNUC_MINOR__ | $compiler -E -xc - | tail -n 1) -printf "%02d%02d\\n" $MAJOR $MINOR - +if [ "x$with_patchlevel" != "x" ] ; then + PATCHLEVEL=$(echo __GNUC_PATCHLEVEL__ | $compiler -E -xc - | tail -n 1) + printf "%02d%02d%02d\\n" $MAJOR $MINOR $PATCHLEVEL +else + printf "%02d%02d\\n" $MAJOR $MINOR +fi diff --git a/scripts/gen_initramfs_list.sh b/scripts/gen_initramfs_list.sh index 683eb12babb..684fb9cdc05 100644 --- a/scripts/gen_initramfs_list.sh +++ b/scripts/gen_initramfs_list.sh @@ -19,11 +19,11 @@ $0 [-o <file>] [-u <uid>] [-g <gid>] {-d | <cpio_source>} ... -o <file> Create gzipped initramfs file named <file> using gen_init_cpio and gzip -u <uid> User ID to map to user ID 0 (root). - <uid> is only meaningful if <cpio_source> - is a directory. + <uid> is only meaningful if <cpio_source> is a + directory. "squash" forces all files to uid 0. -g <gid> Group ID to map to group ID 0 (root). - <gid> is only meaningful if <cpio_source> - is a directory. + <gid> is only meaningful if <cpio_source> is a + directory. "squash" forces all files to gid 0. <cpio_source> File list or directory for cpio archive. If <cpio_source> is a .cpio file it will be used as direct input to initramfs. @@ -113,8 +113,8 @@ parse() { local gid="$4" local ftype=$(filetype "${location}") # remap uid/gid to 0 if necessary - [ "$uid" -eq "$root_uid" ] && uid=0 - [ "$gid" -eq "$root_gid" ] && gid=0 + [ "$root_uid" = "squash" ] && uid=0 || [ "$uid" -eq "$root_uid" ] && uid=0 + [ "$root_gid" = "squash" ] && gid=0 || [ "$gid" -eq "$root_gid" ] && gid=0 local str="${mode} ${uid} ${gid}" [ "${ftype}" == "invalid" ] && return 0 diff --git a/scripts/kallsyms.c b/scripts/kallsyms.c index 10b006694e5..1f11d848532 100644 --- a/scripts/kallsyms.c +++ b/scripts/kallsyms.c @@ -24,8 +24,6 @@ * */ -#define _GNU_SOURCE - #include <stdio.h> #include <stdlib.h> #include <string.h> @@ -378,6 +376,17 @@ static void build_initial_tok_table(void) table_cnt = pos; } +static void *find_token(unsigned char *str, int len, unsigned char *token) +{ + int i; + + for (i = 0; i < len - 1; i++) { + if (str[i] == token[0] && str[i+1] == token[1]) + return &str[i]; + } + return NULL; +} + /* replace a given token in all the valid symbols. Use the sampled symbols * to update the counts */ static void compress_symbols(unsigned char *str, int idx) @@ -391,7 +400,7 @@ static void compress_symbols(unsigned char *str, int idx) p1 = table[i].sym; /* find the token on the symbol */ - p2 = memmem(p1, len, str, 2); + p2 = find_token(p1, len, str); if (!p2) continue; /* decrease the counts for this symbol's tokens */ @@ -410,7 +419,7 @@ static void compress_symbols(unsigned char *str, int idx) if (size < 2) break; /* find the token on the symbol */ - p2 = memmem(p1, size, str, 2); + p2 = find_token(p1, size, str); } while (p2); diff --git a/scripts/kconfig/Makefile b/scripts/kconfig/Makefile index fb2bb3099dd..8986a48c8c4 100644 --- a/scripts/kconfig/Makefile +++ b/scripts/kconfig/Makefile @@ -22,24 +22,25 @@ oldconfig: $(obj)/conf silentoldconfig: $(obj)/conf $< -s arch/$(ARCH)/Kconfig +# Create new linux.po file +# Adjust charset to UTF-8 in .po file to accept UTF-8 in Kconfig files +# The symlink is used to repair a deficiency in arch/um update-po-config: $(obj)/kxgettext - xgettext --default-domain=linux \ - --add-comments --keyword=_ --keyword=N_ \ - --files-from=scripts/kconfig/POTFILES.in \ - --output scripts/kconfig/config.pot - $(Q)ln -fs Kconfig_i386 arch/um/Kconfig_arch - $(Q)for i in `ls arch/`; \ - do \ - scripts/kconfig/kxgettext arch/$$i/Kconfig \ - | msguniq -o scripts/kconfig/linux_$${i}.pot; \ - done - $(Q)msgcat scripts/kconfig/config.pot \ - `find scripts/kconfig/ -type f -name linux_*.pot` \ - --output scripts/kconfig/linux_raw.pot - $(Q)msguniq --sort-by-file scripts/kconfig/linux_raw.pot \ - --output scripts/kconfig/linux.pot - $(Q)rm -f arch/um/Kconfig_arch - $(Q)rm -f scripts/kconfig/linux_*.pot scripts/kconfig/config.pot + xgettext --default-domain=linux \ + --add-comments --keyword=_ --keyword=N_ \ + --from-code=UTF-8 \ + --files-from=scripts/kconfig/POTFILES.in \ + --output $(obj)/config.pot + $(Q)sed -i s/CHARSET/UTF-8/ $(obj)/config.pot + $(Q)ln -fs Kconfig.i386 arch/um/Kconfig.arch + (for i in `ls arch/`; \ + do \ + $(obj)/kxgettext arch/$$i/Kconfig; \ + done ) >> $(obj)/config.pot + msguniq --sort-by-file --to-code=UTF-8 $(obj)/config.pot \ + --output $(obj)/linux.pot + $(Q)rm -f arch/um/Kconfig.arch + $(Q)rm -f $(obj)/config.pot PHONY += randconfig allyesconfig allnoconfig allmodconfig defconfig diff --git a/scripts/kconfig/confdata.c b/scripts/kconfig/confdata.c index 664fe29dace..b2913e9da49 100644 --- a/scripts/kconfig/confdata.c +++ b/scripts/kconfig/confdata.c @@ -341,27 +341,42 @@ int conf_read(const char *name) conf_unsaved++; /* maybe print value in verbose mode... */ sym_ok: + if (!sym_is_choice(sym)) + continue; + /* The choice symbol only has a set value (and thus is not new) + * if all its visible childs have values. + */ + prop = sym_get_choice_prop(sym); + flags = sym->flags; + for (e = prop->expr; e; e = e->left.expr) + if (e->right.sym->visible != no) + flags &= e->right.sym->flags; + sym->flags &= flags | ~SYMBOL_DEF_USER; + } + + for_all_symbols(i, sym) { if (sym_has_value(sym) && !sym_is_choice_value(sym)) { - if (sym->visible == no) + /* Reset values of generates values, so they'll appear + * as new, if they should become visible, but that + * doesn't quite work if the Kconfig and the saved + * configuration disagree. + */ + if (sym->visible == no && !conf_unsaved) sym->flags &= ~SYMBOL_DEF_USER; switch (sym->type) { case S_STRING: case S_INT: case S_HEX: - if (!sym_string_within_range(sym, sym->def[S_DEF_USER].val)) - sym->flags &= ~(SYMBOL_VALID|SYMBOL_DEF_USER); + /* Reset a string value if it's out of range */ + if (sym_string_within_range(sym, sym->def[S_DEF_USER].val)) + break; + sym->flags &= ~(SYMBOL_VALID|SYMBOL_DEF_USER); + conf_unsaved++; + break; default: break; } } - if (!sym_is_choice(sym)) - continue; - prop = sym_get_choice_prop(sym); - flags = sym->flags; - for (e = prop->expr; e; e = e->left.expr) - if (e->right.sym->visible != no) - flags &= e->right.sym->flags; - sym->flags &= flags | ~SYMBOL_DEF_USER; } sym_add_change_count(conf_warnings || conf_unsaved); diff --git a/scripts/kconfig/kxgettext.c b/scripts/kconfig/kxgettext.c index abee55ca617..11f7dab9471 100644 --- a/scripts/kconfig/kxgettext.c +++ b/scripts/kconfig/kxgettext.c @@ -212,7 +212,9 @@ void menu__xgettext(void) struct message *m = message__list; while (m != NULL) { - message__print_gettext_msgid_msgstr(m); + /* skip empty lines ("") */ + if (strlen(m->msg) > sizeof("\"\"")) + message__print_gettext_msgid_msgstr(m); m = m->next; } } diff --git a/scripts/kconfig/lxdialog/check-lxdialog.sh b/scripts/kconfig/lxdialog/check-lxdialog.sh index cdca7388e0f..9681476b96e 100644 --- a/scripts/kconfig/lxdialog/check-lxdialog.sh +++ b/scripts/kconfig/lxdialog/check-lxdialog.sh @@ -51,7 +51,7 @@ usage() { printf "Usage: $0 [-check compiler options|-header|-library]\n" } -if [ $# == 0 ]; then +if [ $# -eq 0 ]; then usage exit 1 fi diff --git a/scripts/kconfig/mconf.c b/scripts/kconfig/mconf.c index d0e4fa594fc..d2c2a429887 100644 --- a/scripts/kconfig/mconf.c +++ b/scripts/kconfig/mconf.c @@ -419,11 +419,13 @@ static void search_conf(void) { struct symbol **sym_arr; struct gstr res; + char *dialog_input; int dres; again: dialog_clear(); dres = dialog_inputbox(_("Search Configuration Parameter"), - _("Enter CONFIG_ (sub)string to search for (omit CONFIG_)"), + _("Enter CONFIG_ (sub)string to search for " + "(with or without \"CONFIG\")"), 10, 75, ""); switch (dres) { case 0: @@ -435,7 +437,12 @@ again: return; } - sym_arr = sym_re_search(dialog_input_result); + /* strip CONFIG_ if necessary */ + dialog_input = dialog_input_result; + if (strncasecmp(dialog_input_result, "CONFIG_", 7) == 0) + dialog_input += 7; + + sym_arr = sym_re_search(dialog_input); res = get_relations_str(sym_arr); free(sym_arr); show_textbox(_("Search Results"), str_get(&res), 0, 0); diff --git a/scripts/kernel-doc b/scripts/kernel-doc index e5bf649e516..1f5835115ca 100755 --- a/scripts/kernel-doc +++ b/scripts/kernel-doc @@ -154,6 +154,7 @@ use strict; my $errors = 0; my $warnings = 0; +my $anon_struct_union = 0; # match expressions used to find embedded type information my $type_constant = '\%([-_\w]+)'; @@ -403,7 +404,11 @@ sub output_highlight { print $lineprefix, $blankline; } else { $line =~ s/\\\\\\/\&/g; - print $lineprefix, $line; + if ($output_mode eq "man" && substr($line, 0, 1) eq ".") { + print "\\&$line"; + } else { + print $lineprefix, $line; + } } print "\n"; } @@ -718,6 +723,7 @@ sub output_struct_xml(%) { # pointer-to-function print " $1 $parameter) ($2);\n"; } elsif ($type =~ m/^(.*?)\s*(:.*)/) { + # bitfield print " $1 $parameter$2;\n"; } else { print " ".$type." ".$parameter.";\n"; @@ -1260,6 +1266,7 @@ sub output_struct_text(%) { # pointer-to-function print "\t$1 $parameter) ($2);\n"; } elsif ($type =~ m/^(.*?)\s*(:.*)/) { + # bitfield print "\t$1 $parameter$2;\n"; } else { print "\t".$type." ".$parameter.";\n"; @@ -1510,8 +1517,13 @@ sub push_parameter($$$) { my $param = shift; my $type = shift; my $file = shift; - my $anon = 0; + if (($anon_struct_union == 1) && ($type eq "") && + ($param eq "}")) { + return; # ignore the ending }; from anon. struct/union + } + + $anon_struct_union = 0; my $param_name = $param; $param_name =~ s/\[.*//; @@ -1530,16 +1542,16 @@ sub push_parameter($$$) { # handle unnamed (anonymous) union or struct: { $type = $param; - $param = "{unnamed_" . $param. "}"; + $param = "{unnamed_" . $param . "}"; $parameterdescs{$param} = "anonymous\n"; - $anon = 1; + $anon_struct_union = 1; } # warn if parameter has no description # (but ignore ones starting with # as these are not parameters # but inline preprocessor statements); # also ignore unnamed structs/unions; - if (!$anon) { + if (!$anon_struct_union) { if (!defined $parameterdescs{$param_name} && $param_name !~ /^#/) { $parameterdescs{$param_name} = $undescribed; @@ -1691,6 +1703,8 @@ sub process_state3_function($$) { my $x = shift; my $file = shift; + $x =~ s@\/\/.*$@@gos; # strip C99-style comments to end of line + if ($x =~ m#\s*/\*\s+MACDOC\s*#io || ($x =~ /^#/ && $x !~ /^#define/)) { # do nothing } @@ -1713,6 +1727,8 @@ sub process_state3_type($$) { $x =~ s@[\r\n]+@ @gos; # strip newlines/cr's. $x =~ s@^\s+@@gos; # strip leading spaces $x =~ s@\s+$@@gos; # strip trailing spaces + $x =~ s@\/\/.*$@@gos; # strip C99-style comments to end of line + if ($x =~ /^#/) { # To distinguish preprocessor directive from regular declaration later. $x .= ";"; @@ -1796,7 +1812,7 @@ sub process_file($) { $state = 2; if (/-(.*)/) { - # strip leading/trailing/multiple spaces #RDD:T: + # strip leading/trailing/multiple spaces $descr= $1; $descr =~ s/^\s*//; $descr =~ s/\s*$//; diff --git a/scripts/mod/modpost.c b/scripts/mod/modpost.c index 3645e980da7..04579a51790 100644 --- a/scripts/mod/modpost.c +++ b/scripts/mod/modpost.c @@ -75,7 +75,8 @@ static int is_vmlinux(const char *modname) else myname = modname; - return strcmp(myname, "vmlinux") == 0; + return (strcmp(myname, "vmlinux") == 0) || + (strcmp(myname, "vmlinux.o") == 0); } void *do_nofail(void *ptr, const char *expr) @@ -374,6 +375,7 @@ static int parse_elf(struct elf_info *info, const char *filename) hdr->e_shstrndx = TO_NATIVE(hdr->e_shstrndx); hdr->e_shnum = TO_NATIVE(hdr->e_shnum); hdr->e_machine = TO_NATIVE(hdr->e_machine); + hdr->e_type = TO_NATIVE(hdr->e_type); sechdrs = (void *)hdr + hdr->e_shoff; info->sechdrs = sechdrs; @@ -384,6 +386,8 @@ static int parse_elf(struct elf_info *info, const char *filename) sechdrs[i].sh_size = TO_NATIVE(sechdrs[i].sh_size); sechdrs[i].sh_link = TO_NATIVE(sechdrs[i].sh_link); sechdrs[i].sh_name = TO_NATIVE(sechdrs[i].sh_name); + sechdrs[i].sh_info = TO_NATIVE(sechdrs[i].sh_info); + sechdrs[i].sh_addr = TO_NATIVE(sechdrs[i].sh_addr); } /* Find symbol table. */ for (i = 1; i < hdr->e_shnum; i++) { @@ -605,18 +609,14 @@ static int strrcmp(const char *s, const char *sub) * warn here. * the pattern is identified by: * tosec = .init.text | .exit.text | .init.data - * fromsec = .data - * atsym = *driver, *_template, *_sht, *_ops, *_probe, *probe_one, *_console + * fromsec = .data | .data.rel | .data.rel.* + * atsym = *driver, *_template, *_sht, *_ops, *_probe, *probe_one, *_console, *_timer * * Pattern 3: - * Whitelist all references from .pci_fixup* section to .init.text - * This is part of the PCI init when built-in - * - * Pattern 4: * Whitelist all refereces from .text.head to .init.data * Whitelist all refereces from .text.head to .init.text * - * Pattern 5: + * Pattern 4: * Some symbols belong to init section but still it is ok to reference * these from non-init sections as these symbols don't have any memory * allocated for them and symbol address and value are same. So even @@ -625,26 +625,6 @@ static int strrcmp(const char *s, const char *sub) * This pattern is identified by * refsymname = __init_begin, _sinittext, _einittext * - * Pattern 7: - * Logos used in drivers/video/logo reside in __initdata but the - * funtion that references them are EXPORT_SYMBOL() so cannot be - * marker __init. So we whitelist them here. - * The pattern is: - * tosec = .init.data - * fromsec = .text* - * refsymname = logo_ - * - * Pattern 8: - * Symbols contained in .paravirtprobe may safely reference .init.text. - * The pattern is: - * tosec = .init.text - * fromsec = .paravirtprobe - * - * Pattern 10: - * ia64 has machvec table for each platform and - * powerpc has a machine desc table for each platform. - * It is mixture of function pointers of .init.text and .text. - * fromsec = .machvec | .machine.desc **/ static int secref_whitelist(const char *modname, const char *tosec, const char *fromsec, const char *atsym, @@ -655,12 +635,12 @@ static int secref_whitelist(const char *modname, const char *tosec, const char *pat2sym[] = { "driver", "_template", /* scsi uses *_template a lot */ + "_timer", /* arm uses ops structures named _timer a lot */ "_sht", /* scsi also used *_sht to some extent */ "_ops", "_probe", "_probe_one", "_console", - "apic_es7000", NULL }; @@ -692,7 +672,9 @@ static int secref_whitelist(const char *modname, const char *tosec, (strcmp(tosec, ".exit.text") != 0) && (strcmp(tosec, ".init.data") != 0)) f2 = 0; - if (strcmp(fromsec, ".data") != 0) + if ((strcmp(fromsec, ".data") != 0) && + (strcmp(fromsec, ".data.rel") != 0) && + (strncmp(fromsec, ".data.rel.", strlen(".data.rel.")) != 0)) f2 = 0; for (s = pat2sym; *s; s++) @@ -702,37 +684,16 @@ static int secref_whitelist(const char *modname, const char *tosec, return 1; /* Check for pattern 3 */ - if ((strncmp(fromsec, ".pci_fixup", strlen(".pci_fixup")) == 0) && - (strcmp(tosec, ".init.text") == 0)) - return 1; - - /* Check for pattern 4 */ if ((strcmp(fromsec, ".text.head") == 0) && ((strcmp(tosec, ".init.data") == 0) || (strcmp(tosec, ".init.text") == 0))) return 1; - /* Check for pattern 5 */ + /* Check for pattern 4 */ for (s = pat3refsym; *s; s++) if (strcmp(refsymname, *s) == 0) return 1; - /* Check for pattern 7 */ - if ((strcmp(tosec, ".init.data") == 0) && - (strncmp(fromsec, ".text", strlen(".text")) == 0) && - (strncmp(refsymname, "logo_", strlen("logo_")) == 0)) - return 1; - - /* Check for pattern 8 */ - if ((strcmp(tosec, ".init.text") == 0) && - (strcmp(fromsec, ".paravirtprobe") == 0)) - return 1; - - /* Check for pattern 10 */ - if ((strcmp(fromsec, ".machvec") == 0) || - (strcmp(fromsec, ".machine.desc") == 0)) - return 1; - return 0; } @@ -753,6 +714,8 @@ static Elf_Sym *find_elf_symbol(struct elf_info *elf, Elf_Addr addr, for (sym = elf->symtab_start; sym < elf->symtab_stop; sym++) { if (sym->st_shndx != relsym->st_shndx) continue; + if (ELF_ST_TYPE(sym->st_info) == STT_SECTION) + continue; if (sym->st_value == addr) return sym; } @@ -864,11 +827,6 @@ static void warn_sec_mismatch(const char *modname, const char *fromsec, elf->strtab + before->st_name, refsymname)) return; - /* fromsec whitelist - without a valid 'before' - * powerpc has a GOT table in .got2 section */ - if (strcmp(fromsec, ".got2") == 0) - return; - if (before && after) { warn("%s(%s+0x%llx): Section mismatch: reference to %s:%s " "(between '%s' and '%s')\n", @@ -895,6 +853,78 @@ static void warn_sec_mismatch(const char *modname, const char *fromsec, } } +static unsigned int *reloc_location(struct elf_info *elf, + int rsection, Elf_Rela *r) +{ + Elf_Shdr *sechdrs = elf->sechdrs; + int section = sechdrs[rsection].sh_info; + + return (void *)elf->hdr + sechdrs[section].sh_offset + + (r->r_offset - sechdrs[section].sh_addr); +} + +static int addend_386_rel(struct elf_info *elf, int rsection, Elf_Rela *r) +{ + unsigned int r_typ = ELF_R_TYPE(r->r_info); + unsigned int *location = reloc_location(elf, rsection, r); + + switch (r_typ) { + case R_386_32: + r->r_addend = TO_NATIVE(*location); + break; + case R_386_PC32: + r->r_addend = TO_NATIVE(*location) + 4; + /* For CONFIG_RELOCATABLE=y */ + if (elf->hdr->e_type == ET_EXEC) + r->r_addend += r->r_offset; + break; + } + return 0; +} + +static int addend_arm_rel(struct elf_info *elf, int rsection, Elf_Rela *r) +{ + unsigned int r_typ = ELF_R_TYPE(r->r_info); + + switch (r_typ) { + case R_ARM_ABS32: + /* From ARM ABI: (S + A) | T */ + r->r_addend = (int)(long)(elf->symtab_start + ELF_R_SYM(r->r_info)); + break; + case R_ARM_PC24: + /* From ARM ABI: ((S + A) | T) - P */ + r->r_addend = (int)(long)(elf->hdr + elf->sechdrs[rsection].sh_offset + + (r->r_offset - elf->sechdrs[rsection].sh_addr)); + break; + default: + return 1; + } + return 0; +} + +static int addend_mips_rel(struct elf_info *elf, int rsection, Elf_Rela *r) +{ + unsigned int r_typ = ELF_R_TYPE(r->r_info); + unsigned int *location = reloc_location(elf, rsection, r); + unsigned int inst; + + if (r_typ == R_MIPS_HI16) + return 1; /* skip this */ + inst = TO_NATIVE(*location); + switch (r_typ) { + case R_MIPS_LO16: + r->r_addend = inst & 0xffff; + break; + case R_MIPS_26: + r->r_addend = (inst & 0x03ffffff) << 2; + break; + case R_MIPS_32: + r->r_addend = inst; + break; + } + return 0; +} + /** * A module includes a number of sections that are discarded * either when loaded or when used as built-in. @@ -938,8 +968,11 @@ static void check_sec_ref(struct module *mod, const char *modname, r.r_offset = TO_NATIVE(rela->r_offset); #if KERNEL_ELFCLASS == ELFCLASS64 if (hdr->e_machine == EM_MIPS) { + unsigned int r_typ; r_sym = ELF64_MIPS_R_SYM(rela->r_info); r_sym = TO_NATIVE(r_sym); + r_typ = ELF64_MIPS_R_TYPE(rela->r_info); + r.r_info = ELF64_R_INFO(r_sym, r_typ); } else { r.r_info = TO_NATIVE(rela->r_info); r_sym = ELF_R_SYM(r.r_info); @@ -972,8 +1005,11 @@ static void check_sec_ref(struct module *mod, const char *modname, r.r_offset = TO_NATIVE(rel->r_offset); #if KERNEL_ELFCLASS == ELFCLASS64 if (hdr->e_machine == EM_MIPS) { + unsigned int r_typ; r_sym = ELF64_MIPS_R_SYM(rel->r_info); r_sym = TO_NATIVE(r_sym); + r_typ = ELF64_MIPS_R_TYPE(rel->r_info); + r.r_info = ELF64_R_INFO(r_sym, r_typ); } else { r.r_info = TO_NATIVE(rel->r_info); r_sym = ELF_R_SYM(r.r_info); @@ -983,6 +1019,20 @@ static void check_sec_ref(struct module *mod, const char *modname, r_sym = ELF_R_SYM(r.r_info); #endif r.r_addend = 0; + switch (hdr->e_machine) { + case EM_386: + if (addend_386_rel(elf, i, &r)) + continue; + break; + case EM_ARM: + if(addend_arm_rel(elf, i, &r)) + continue; + break; + case EM_MIPS: + if (addend_mips_rel(elf, i, &r)) + continue; + break; + } sym = elf->symtab_start + r_sym; /* Skip special sections */ if (sym->st_shndx >= SHN_LORESERVE) @@ -998,6 +1048,63 @@ static void check_sec_ref(struct module *mod, const char *modname, } } +/* + * Identify sections from which references to either a + * .init or a .exit section is OK. + * + * [OPD] Keith Ownes <kaos@sgi.com> commented: + * For our future {in}sanity, add a comment that this is the ppc .opd + * section, not the ia64 .opd section. + * ia64 .opd should not point to discarded sections. + * [.rodata] like for .init.text we ignore .rodata references -same reason + */ +static int initexit_section_ref_ok(const char *name) +{ + const char **s; + /* Absolute section names */ + const char *namelist1[] = { + "__bug_table", /* used by powerpc for BUG() */ + "__ex_table", + ".altinstructions", + ".cranges", /* used by sh64 */ + ".fixup", + ".machvec", /* ia64 + powerpc uses these */ + ".machine.desc", + ".opd", /* See comment [OPD] */ + ".parainstructions", + ".pdr", + ".plt", /* seen on ARCH=um build on x86_64. Harmless */ + ".smp_locks", + ".stab", + NULL + }; + /* Start of section names */ + const char *namelist2[] = { + ".debug", + ".eh_frame", + ".note", /* ignore ELF notes - may contain anything */ + ".got", /* powerpc - global offset table */ + ".toc", /* powerpc - table of contents */ + NULL + }; + /* part of section name */ + const char *namelist3 [] = { + ".unwind", /* Sample: IA_64.unwind.exit.text */ + NULL + }; + + for (s = namelist1; *s; s++) + if (strcmp(*s, name) == 0) + return 1; + for (s = namelist2; *s; s++) + if (strncmp(*s, name, strlen(*s)) == 0) + return 1; + for (s = namelist3; *s; s++) + if (strstr(name, *s) != NULL) + return 1; + return 0; +} + /** * Functions used only during module init is marked __init and is stored in * a .init.text section. Likewise data is marked __initdata and stored in @@ -1014,7 +1121,7 @@ static int init_section(const char *name) return 0; } -/** +/* * Identify sections from which references to a .init section is OK. * * Unfortunately references to read only data that referenced .init @@ -1028,48 +1135,31 @@ static int init_section(const char *name) * * where vgacon_startup is __init. If you want to wade through the false * positives, take out the check for rodata. - **/ + */ static int init_section_ref_ok(const char *name) { const char **s; /* Absolute section names */ const char *namelist1[] = { - ".init", - ".opd", /* see comment [OPD] at exit_section_ref_ok() */ - ".toc1", /* used by ppc64 */ - ".stab", - ".data.rel.ro", /* used by parisc64 */ - ".parainstructions", - ".text.lock", - "__bug_table", /* used by powerpc for BUG() */ - ".pci_fixup_header", - ".pci_fixup_final", - ".pdr", - "__param", - "__ex_table", - ".fixup", - ".smp_locks", - ".plt", /* seen on ARCH=um build on x86_64. Harmless */ + "__dbe_table", /* MIPS generate these */ "__ftr_fixup", /* powerpc cpu feature fixup */ "__fw_ftr_fixup", /* powerpc firmware feature fixup */ - ".cranges", /* used by sh64 */ + "__param", + ".data.rel.ro", /* used by parisc64 */ + ".init", + ".text.lock", NULL }; /* Start of section names */ const char *namelist2[] = { ".init.", - ".altinstructions", - ".eh_frame", - ".debug", - ".parainstructions", + ".pci_fixup", ".rodata", NULL }; - /* part of section name */ - const char *namelist3 [] = { - ".unwind", /* sample: IA_64.unwind.init.text */ - NULL - }; + + if (initexit_section_ref_ok(name)) + return 1; for (s = namelist1; *s; s++) if (strcmp(*s, name) == 0) @@ -1077,9 +1167,10 @@ static int init_section_ref_ok(const char *name) for (s = namelist2; *s; s++) if (strncmp(*s, name, strlen(*s)) == 0) return 1; - for (s = namelist3; *s; s++) - if (strstr(name, *s) != NULL) - return 1; + + /* If section name ends with ".init" we allow references + * as is the case with .initcallN.init, .early_param.init, .taglist.init etc + */ if (strrcmp(name, ".init") == 0) return 1; return 0; @@ -1104,58 +1195,25 @@ static int exit_section(const char *name) /* * Identify sections from which references to a .exit section is OK. - * - * [OPD] Keith Ownes <kaos@sgi.com> commented: - * For our future {in}sanity, add a comment that this is the ppc .opd - * section, not the ia64 .opd section. - * ia64 .opd should not point to discarded sections. - * [.rodata] like for .init.text we ignore .rodata references -same reason - **/ + */ static int exit_section_ref_ok(const char *name) { const char **s; /* Absolute section names */ const char *namelist1[] = { - ".exit.text", ".exit.data", - ".init.text", - ".rodata", - ".opd", /* See comment [OPD] */ - ".toc1", /* used by ppc64 */ - ".altinstructions", - ".pdr", - "__bug_table", /* used by powerpc for BUG() */ + ".exit.text", ".exitcall.exit", - ".eh_frame", - ".parainstructions", - ".stab", - "__ex_table", - ".fixup", - ".smp_locks", - ".plt", /* seen on ARCH=um build on x86_64. Harmless */ - ".cranges", /* used by sh64 */ - NULL - }; - /* Start of section names */ - const char *namelist2[] = { - ".debug", - NULL - }; - /* part of section name */ - const char *namelist3 [] = { - ".unwind", /* Sample: IA_64.unwind.exit.text */ + ".rodata", NULL }; + if (initexit_section_ref_ok(name)) + return 1; + for (s = namelist1; *s; s++) if (strcmp(*s, name) == 0) return 1; - for (s = namelist2; *s; s++) - if (strncmp(*s, name, strlen(*s)) == 0) - return 1; - for (s = namelist3; *s; s++) - if (strstr(name, *s) != NULL) - return 1; return 0; } diff --git a/scripts/mod/modpost.h b/scripts/mod/modpost.h index 0858caa9c03..4156dd34c5d 100644 --- a/scripts/mod/modpost.h +++ b/scripts/mod/modpost.h @@ -60,6 +60,9 @@ typedef union #define ELF64_MIPS_R_SYM(i) \ ((__extension__ (_Elf64_Mips_R_Info_union)(i)).r_info_fields.r_sym) +#define ELF64_MIPS_R_TYPE(i) \ + ((__extension__ (_Elf64_Mips_R_Info_union)(i)).r_info_fields.r_type1) + #if KERNEL_ELFDATA != HOST_ELFDATA static inline void __endian(const void *src, void *dest, unsigned int size) diff --git a/security/commoncap.c b/security/commoncap.c index 384379ede4f..338606eb723 100644 --- a/security/commoncap.c +++ b/security/commoncap.c @@ -148,7 +148,7 @@ void cap_bprm_apply_creds (struct linux_binprm *bprm, int unsafe) if (bprm->e_uid != current->uid || bprm->e_gid != current->gid || !cap_issubset (new_permitted, current->cap_permitted)) { - current->mm->dumpable = suid_dumpable; + set_dumpable(current->mm, suid_dumpable); if (unsafe & ~LSM_UNSAFE_PTRACE_CAP) { if (!capable(CAP_SETUID)) { diff --git a/security/dummy.c b/security/dummy.c index d6a112ce297..19d813d5e08 100644 --- a/security/dummy.c +++ b/security/dummy.c @@ -130,7 +130,7 @@ static void dummy_bprm_free_security (struct linux_binprm *bprm) static void dummy_bprm_apply_creds (struct linux_binprm *bprm, int unsafe) { if (bprm->e_uid != current->uid || bprm->e_gid != current->gid) { - current->mm->dumpable = suid_dumpable; + set_dumpable(current->mm, suid_dumpable); if ((unsafe & ~LSM_UNSAFE_PTRACE_CAP) && !capable(CAP_SETUID)) { bprm->e_uid = current->uid; diff --git a/sound/core/seq/seq_virmidi.c b/sound/core/seq/seq_virmidi.c index 061a7c61402..e11790f6deb 100644 --- a/sound/core/seq/seq_virmidi.c +++ b/sound/core/seq/seq_virmidi.c @@ -363,7 +363,7 @@ static int snd_virmidi_dev_attach_seq(struct snd_virmidi_dev *rdev) if (rdev->client >= 0) return 0; - pinfo = kmalloc(sizeof(*pinfo), GFP_KERNEL); + pinfo = kzalloc(sizeof(*pinfo), GFP_KERNEL); if (!pinfo) { err = -ENOMEM; goto __error; @@ -380,7 +380,6 @@ static int snd_virmidi_dev_attach_seq(struct snd_virmidi_dev *rdev) rdev->client = client; /* create a port */ - memset(pinfo, 0, sizeof(*pinfo)); pinfo->addr.client = client; sprintf(pinfo->name, "VirMIDI %d-%d", rdev->card->number, rdev->device); /* set all capabilities */ diff --git a/sound/core/sound.c b/sound/core/sound.c index 70600df94d6..8dc7a3b32b9 100644 --- a/sound/core/sound.c +++ b/sound/core/sound.c @@ -446,8 +446,7 @@ static void __exit alsa_sound_exit(void) { snd_info_minor_unregister(); snd_info_done(); - if (unregister_chrdev(major, "alsa") != 0) - snd_printk(KERN_ERR "unable to unregister major device number %d\n", major); + unregister_chrdev(major, "alsa"); } module_init(alsa_sound_init) diff --git a/usr/gen_init_cpio.c b/usr/gen_init_cpio.c index 8365db6cfe0..7abc07f0fcd 100644 --- a/usr/gen_init_cpio.c +++ b/usr/gen_init_cpio.c @@ -498,7 +498,9 @@ int main (int argc, char *argv[]) exit(1); } - if (! (cpio_list = fopen(argv[1], "r"))) { + if (!strcmp(argv[1], "-")) + cpio_list = stdin; + else if (! (cpio_list = fopen(argv[1], "r"))) { fprintf(stderr, "ERROR: unable to open '%s': %s\n\n", argv[1], strerror(errno)); usage(argv[0]); |