firefly-linux-kernel-4.4.55.git
11 years agoipc,sem: fix semctl(..., GETNCNT)
Rik van Riel [Thu, 9 May 2013 20:59:59 +0000 (16:59 -0400)]
ipc,sem: fix semctl(..., GETNCNT)

The semctl GETNCNT returns the number of semops waiting for the
specified semaphore to become nonzero.  After commit 9f1bc2c9022c
("ipc,sem: have only one list in struct sem_queue"), the semops waiting
on just one semaphore are waiting on that semaphore's list.

In order to return the correct count, we have to walk that list too, in
addition to the sem_array's list for complex operations.

Signed-off-by: Rik van Riel <riel@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
11 years agoipc,sem: fix semctl(..., GETZCNT)
Rik van Riel [Thu, 9 May 2013 20:53:28 +0000 (16:53 -0400)]
ipc,sem: fix semctl(..., GETZCNT)

The semctl GETZCNT returns the number of semops waiting for the
specified semaphore to become zero.  After commit 9f1bc2c9022c
("ipc,sem: have only one list in struct sem_queue"), the semops waiting
on just one semaphore are waiting on that semaphore's list.

In order to return the correct count, we have to walk that list too, in
addition to the sem_array's list for complex operations.

This bug broke dbench; it works again with this patch applied.

Signed-off-by: Rik van Riel <riel@redhat.com>
Reported-by: Kent Overstreet <koverstreet@google.com>
Tested-by: Kent Overstreet <koverstreet@google.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
11 years agoMerge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs
Linus Torvalds [Thu, 9 May 2013 20:44:35 +0000 (13:44 -0700)]
Merge branch 'for-linus' of git://git./linux/kernel/git/viro/vfs

Pull more vfs fixes from Al Viro:
 "Regression fix from Geert + yet another open-coded kernel_read()"

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs:
  ecryptfs: don't open-code kernel_read()
  xtensa simdisk: Fix proc_create_data() conversion fallout

11 years agoMerge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mason/linux...
Linus Torvalds [Thu, 9 May 2013 20:07:40 +0000 (13:07 -0700)]
Merge branch 'for-linus' of git://git./linux/kernel/git/mason/linux-btrfs

Pull btrfs update from Chris Mason:
 "These are mostly fixes.  The biggest exceptions are Josef's skinny
  extents and Jan Schmidt's code to rebuild our quota indexes if they
  get out of sync (or you enable quotas on an existing filesystem).

  The skinny extents are off by default because they are a new variation
  on the extent allocation tree format.  btrfstune -x enables them, and
  the new format makes the extent allocation tree about 30% smaller.

  I rebased this a few days ago to rework Dave Sterba's crc checks on
  the super block, but almost all of these go back to rc6, since I
  though 3.9 was due any minute.

  The biggest missing fix is the tracepoint bug that was hit late in
  3.9.  I ran into problems with that in overnight testing and I'm still
  tracking it down.  I'll definitely have that fixed for rc2."

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mason/linux-btrfs: (101 commits)
  Btrfs: allow superblock mismatch from older mkfs
  btrfs: enhance superblock checks
  btrfs: fix misleading variable name for flags
  btrfs: use unsigned long type for extent state bits
  Btrfs: improve the loop of scrub_stripe
  btrfs: read entire device info under lock
  btrfs: remove unused gfp mask parameter from release_extent_buffer callchain
  btrfs: handle errors returned from get_tree_block_key
  btrfs: make static code static & remove dead code
  Btrfs: deal with errors in write_dev_supers
  Btrfs: remove almost all of the BUG()'s from tree-log.c
  Btrfs: deal with free space cache errors while replaying log
  Btrfs: automatic rescan after "quota enable" command
  Btrfs: rescan for qgroups
  Btrfs: split btrfs_qgroup_account_ref into four functions
  Btrfs: allocate new chunks if the space is not enough for global rsv
  Btrfs: separate sequence numbers for delayed ref tracking and tree mod log
  btrfs: move leak debug code to functions
  Btrfs: return free space in cow error path
  Btrfs: set UUID in root_item for created trees
  ...

11 years agoMerge tag 'for-linus-v3.10-rc1-2' of git://oss.sgi.com/xfs/xfs
Linus Torvalds [Thu, 9 May 2013 20:06:20 +0000 (13:06 -0700)]
Merge tag 'for-linus-v3.10-rc1-2' of git://oss.sgi.com/xfs/xfs

Pull xfs update (#2) from Ben Myers:

 - add CONFIG_XFS_WARN, a step between zero debugging and
   CONFIG_XFS_DEBUG.

 - fix attrmulti and attrlist to fall back to vmalloc when kmalloc
   fails.

* tag 'for-linus-v3.10-rc1-2' of git://oss.sgi.com/xfs/xfs:
  xfs: fallback to vmalloc for large buffers in xfs_compat_attrlist_by_handle
  xfs: fallback to vmalloc for large buffers in xfs_attrlist_by_handle
  xfs: introduce CONFIG_XFS_WARN

11 years agoecryptfs: don't open-code kernel_read()
Al Viro [Wed, 8 May 2013 02:28:48 +0000 (22:28 -0400)]
ecryptfs: don't open-code kernel_read()

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
11 years agoMerge tag 'nfs-for-3.10-2' of git://git.linux-nfs.org/projects/trondmy/linux-nfs
Linus Torvalds [Thu, 9 May 2013 17:24:54 +0000 (10:24 -0700)]
Merge tag 'nfs-for-3.10-2' of git://git.linux-nfs.org/projects/trondmy/linux-nfs

Pull more NFS client bugfixes from Trond Myklebust:

 - Ensure that we match the 'sec=' mount flavour against the server list

 - Fix the NFSv4 byte range locking in the presence of delegations

 - Ensure that we conform to the NFSv4.1 spec w.r.t.  freeing lock
   stateids

 - Fix a pNFS data server connection race

* tag 'nfs-for-3.10-2' of git://git.linux-nfs.org/projects/trondmy/linux-nfs:
  NFS4.1 Fix data server connection race
  NFSv3: match sec= flavor against server list
  NFSv4.1: Ensure that we free the lock stateid on the server
  NFSv4: Convert nfs41_free_stateid to use an asynchronous RPC call
  SUNRPC: Don't spam syslog with "Pseudoflavor not found" messages
  NFSv4.x: Fix handling of partially delegated locks

11 years agoMerge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net
Linus Torvalds [Thu, 9 May 2013 17:23:43 +0000 (10:23 -0700)]
Merge git://git./linux/kernel/git/davem/net

Pull networking update from David Miller:

 1) Propagate return error values properly in irda, spider_net, sfc, and
    bfin_mac.  From Wei Yongjun.

 2) Fix fec driver OOPS on rapid link up/down, from Frank Li.

 3) FIX VF resource allocation and chip message payload length errors in
    be2net driver, from Sathya Perla.

 4) Fix inner protocol inspection during GSO from Pravin B Shelar.

* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net:
  if_cablemodem.h: Add parenthesis around ioctl macros
  gso: Handle Trans-Ether-Bridging protocol in skb_network_protocol()
  net: fec: fix kernel oops when plug/unplug cable many times
  bfin_mac: fix error return code in bfin_mac_probe()
  sfc: fix return value check in efx_ptp_probe_channel()
  net/spider_net: fix error return code in spider_net_open()
  net/irda: fix error return code in bfin_sir_open()
  net: of_mdio: fix behavior on missing phy device
  sierra_net: keep status interrupt URB active
  usbnet: allow status interrupt URB to always be active
  qmi_wwan/cdc_ether: add device IDs for Dell 5804 (Novatel E371) WWAN card
  be2net: disable TX in be_close()
  be2net: fix EQ from getting full while cleaning RX CQ
  be2net: fix payload_len value for GET_MAC_LIST cmd req
  be2net: provision VF resources before enabling SR-IOV

11 years agoMerge tag 'pci-v3.10-fixes-1' of git://git.kernel.org/pub/scm/linux/kernel/git/helgaa...
Linus Torvalds [Thu, 9 May 2013 17:21:44 +0000 (10:21 -0700)]
Merge tag 'pci-v3.10-fixes-1' of git://git./linux/kernel/git/helgaas/pci

Pull PCI updates from Bjorn Helgaas:
 "MSI:
      PCI: Set ->mask_pos correctly
  Hotplug:
      PCI: Delay final fixups until resources are assigned
  Moorestown:
      x86/pci/mrst: Use configuration mechanism 1 for 00:00.0, 00:02.0, 00:03.0"

* tag 'pci-v3.10-fixes-1' of git://git.kernel.org/pub/scm/linux/kernel/git/helgaas/pci:
  PCI: Delay final fixups until resources are assigned
  x86/pci/mrst: Use configuration mechanism 1 for 00:00.0, 00:02.0, 00:03.0
  PCI: Set ->mask_pos correctly

11 years agoxtensa simdisk: Fix proc_create_data() conversion fallout
Geert Uytterhoeven [Thu, 9 May 2013 12:08:31 +0000 (14:08 +0200)]
xtensa simdisk: Fix proc_create_data() conversion fallout

arch/xtensa/platforms/iss/simdisk.c: In function 'proc_read_simdisk':
arch/xtensa/platforms/iss/simdisk.c:220:12: warning: initialization discards 'const' qualifier from pointer target type [enabled by default]
arch/xtensa/platforms/iss/simdisk.c: In function 'proc_write_simdisk':
arch/xtensa/platforms/iss/simdisk.c:241:38: error: 'buffer' undeclared (first use in this function)
arch/xtensa/platforms/iss/simdisk.c:241:38: note: each undeclared identifier is reported only once for each function it appears in

Introduced by commit a69755b187749e7cc020e17127a54f395aea4eaa ("xtensa
simdisk: switch to proc_create_data()")

Signed-off-by: Geert Uytterhoeven <geert@linux-m68k.org>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
11 years agoMerge tag 'for-linus-20130509' of git://git.infradead.org/linux-mtd
Linus Torvalds [Thu, 9 May 2013 17:15:46 +0000 (10:15 -0700)]
Merge tag 'for-linus-20130509' of git://git.infradead.org/linux-mtd

Pull MTD update from David Woodhouse:

 - Lots of cleanups from Artem, including deletion of some obsolete
   drivers

 - Support partitions larger than 4GiB in device tree

 - Support for new SPI chips

* tag 'for-linus-20130509' of git://git.infradead.org/linux-mtd: (83 commits)
  mtd: omap2: Use module_platform_driver()
  mtd: bf5xx_nand: Use module_platform_driver()
  mtd: denali_dt: Remove redundant use of of_match_ptr
  mtd: denali_dt: Change return value to fix smatch warning
  mtd: denali_dt: Use module_platform_driver()
  mtd: denali_dt: Fix incorrect error check
  mtd: nand: subpage write support for hardware based ECC schemes
  mtd: omap2: use msecs_to_jiffies()
  mtd: nand_ids: use size macros
  mtd: nand_ids: improve LEGACY_ID_NAND macro a bit
  mtd: add 4 Toshiba nand chips for the full-id case
  mtd: add the support to parse out the full-id nand type
  mtd: add new fields to nand_flash_dev{}
  mtd: sh_flctl: Use of_match_ptr() macro
  mtd: gpio: Use of_match_ptr() macro
  mtd: gpio: Use devm_kzalloc()
  mtd: davinci_nand: Use of_match_ptr()
  mtd: dataflash: Use of_match_ptr() macro
  mtd: remove h720x flash support
  mtd: onenand: remove OneNAND simulator
  ...

11 years agoMerge tag 'firewire-updates' of git://git.kernel.org/pub/scm/linux/kernel/git/ieee139...
Linus Torvalds [Thu, 9 May 2013 17:11:48 +0000 (10:11 -0700)]
Merge tag 'firewire-updates' of git://git./linux/kernel/git/ieee1394/linux1394

Pull firewure updates from Stefan Richter:
  - fix controller removal when controller is in suspended state
  - fix video reception on VIA VT6306 with gstreamer, MythTV, and maybe dv4l
  - fix a startup issue with Agere/LSI FW643-e2
  - error logging improvements and other small updates

* tag 'firewire-updates' of git://git.kernel.org/pub/scm/linux/kernel/git/ieee1394/linux1394:
  firewire: ohci: dump_stack() for PHY regs read/write failures
  firewire: ohci: Improve bus reset error messages
  firewire: ohci: Alias dev_* log functions
  firewire: ohci: Fix 'failed to read phy reg' on FW643 rev8
  firewire: ohci: fix VIA VT6306 video reception
  firewire: ohci: Check LPS before register access on pci removal
  firewire: ohci: Fix double free_irq()
  firewire: remove unnecessary alloc/OOM messages
  firewire: sbp2: replace BUG_ON by WARN_ON
  firewire: core: remove an always false test
  firewire: Remove two unneeded checks for macros

11 years agoMerge tag 'edac_fixes_for_3.10' of git://git.kernel.org/pub/scm/linux/kernel/git...
Linus Torvalds [Thu, 9 May 2013 17:11:08 +0000 (10:11 -0700)]
Merge tag 'edac_fixes_for_3.10' of git://git./linux/kernel/git/bp/bp

Pull two small EDAC fixes from Borislav Petkov.

* tag 'edac_fixes_for_3.10' of git://git.kernel.org/pub/scm/linux/kernel/git/bp/bp:
  EDAC: Don't give write permission to read-only files
  EDAC, mc_sysfs.c: Fix string array pointer types

11 years agoMerge git://www.linux-watchdog.org/linux-watchdog
Linus Torvalds [Thu, 9 May 2013 17:09:14 +0000 (10:09 -0700)]
Merge git://www.linux-watchdog.org/linux-watchdog

Pull watchdog update from Wim Van Sebroeck:
 "Fix a kdump issue in hpwdt and a possible NULL dereference"

* git://www.linux-watchdog.org/linux-watchdog:
  watchdog: Fix race condition in registration code
  watchdog: Convert to devm_ioremap_resource()

11 years agoMerge tag 'gpio-for-linus' of git://git.secretlab.ca/git/linux
Linus Torvalds [Thu, 9 May 2013 16:59:16 +0000 (09:59 -0700)]
Merge tag 'gpio-for-linus' of git://git.secretlab.ca/git/linux

Pull removal of GENERIC_GPIO from Grant Likely:
 "GENERIC_GPIO now synonymous with GPIOLIB.  There are no longer any
  valid cases for enableing GENERIC_GPIO without GPIOLIB, even though it
  is possible to do so which has been causing confusion and breakage.
  This branch does the work to completely eliminate GENERIC_GPIO."

* tag 'gpio-for-linus' of git://git.secretlab.ca/git/linux:
  gpio: update gpio Chinese documentation
  Remove GENERIC_GPIO config option
  Convert selectors of GENERIC_GPIO to GPIOLIB
  blackfin: force use of gpiolib
  m68k: coldfire: use gpiolib
  mips: pnx833x: remove requirement for GENERIC_GPIO
  openrisc: default GENERIC_GPIO to false
  avr32: default GENERIC_GPIO to false
  xtensa: remove explicit selection of GENERIC_GPIO
  sh: replace CONFIG_GENERIC_GPIO by CONFIG_GPIOLIB
  powerpc: remove redundant GENERIC_GPIO selection
  unicore32: default GENERIC_GPIO to false
  unicore32: remove unneeded select GENERIC_GPIO
  arm: plat-orion: use GPIO driver on CONFIG_GPIOLIB
  arm: remove redundant GENERIC_GPIO selection
  mips: alchemy: require gpiolib
  mips: txx9: change GENERIC_GPIO to GPIOLIB
  mips: loongson: use GPIO driver on CONFIG_GPIOLIB
  mips: remove redundant GENERIC_GPIO select

11 years agoMerge branch 'for-linus' of git://git.infradead.org/users/vkoul/slave-dma
Linus Torvalds [Thu, 9 May 2013 16:46:45 +0000 (09:46 -0700)]
Merge branch 'for-linus' of git://git.infradead.org/users/vkoul/slave-dma

Pull slave-dmaengine updates from Vinod Koul:
 "This time we have dmatest improvements from Andy along with dw_dmac
  fixes.  He has also done support for acpi for dmanegine.

  Also we have bunch of fixes going in DT support for dmanegine for
  various folks.  Then Haswell and other ioat changes from Dave and
  SUDMAC support from Shimoda."

* 'for-linus' of git://git.infradead.org/users/vkoul/slave-dma: (53 commits)
  dma: tegra: implement suspend/resume callbacks
  dma:of: Use a mutex to protect the of_dma_list
  dma: of: Fix of_node reference leak
  dmaengine: sirf: move driver init from module_init to subsys_initcall
  sudmac: add support for SUDMAC
  dma: sh: add Kconfig
  at_hdmac: move to generic DMA binding
  ioatdma: ioat3_alloc_sed can be static
  ioatdma: Adding write back descriptor error status support for ioatdma 3.3
  ioatdma: S1200 platforms ioatdma channel 2 and 3 falsely advertise RAID cap
  ioatdma: Adding support for 16 src PQ ops and super extended descriptors
  ioatdma: Removing hw bug workaround for CB3.x .2 and earlier
  dw_dmac: add ACPI support
  dmaengine: call acpi_dma_request_slave_channel as well
  dma: acpi-dma: introduce ACPI DMA helpers
  dma: of: Remove unnecessary list_empty check
  DMA: OF: Check properties value before running be32_to_cpup() on it
  DMA: of: Constant names
  ioatdma: skip silicon bug workaround for pq_align for cb3.3
  ioatdma: Removing PQ val disable for cb3.3
  ...

11 years agoMerge branch 'next' of git://git.kernel.org/pub/scm/linux/kernel/git/rzhang/linux
Linus Torvalds [Thu, 9 May 2013 16:40:49 +0000 (09:40 -0700)]
Merge branch 'next' of git://git./linux/kernel/git/rzhang/linux

Pull thermal management update from Zhang Rui:
 "The most important one is to build thermal core and governor and cpu
  cooling code into one module.  This fixes a regression that thermal
  core does not work if it is built as module, since 3.7.  I'll backport
  them to stable kernel once those changes are in upstream.

  The largest batch is the thermal kernel-doc & coding style
  updates/cleanups from Eduardo.

  Highlights:

   - build all thermal framework code into one module to fix a
     regression that thermal does not work if it is built as module.

   - Marvell Armada 370/XP thermal sensor driver

   - thermal core/cpu cooling kernel-doc & coding style updates and
     cleanups.

   - Add Eduardo Valentin as thermal sub-maintainer, both in mailing
     list and patchwork.  He will help me on arm thermal drivers."

* 'next' of git://git.kernel.org/pub/scm/linux/kernel/git/rzhang/linux: (68 commits)
  thermal: db8500_cpufreq_cooling: remove usage of IS_ERR_OR_NULL()
  thermal: thermal_core: remove usage of IS_ERR_OR_NULL
  thermal: cpu_cooling: improve line breaking
  thermal: cpu_cooling: alignment improvements
  thermal: cpu_cooling: remove checkpatch.pl warning
  thermal: cpu_cooling: remove trailing blank line
  thermal: cpu_cooling: align on open parenthesis
  thermal: cpu_cooling: standardize comment style
  thermal: cpu_cooling: standardize end of function
  thermal: cpu_cooling: remove trailing white spaces
  Thermal: update documentation for thermal_zone_device_register
  thermal: update kernel-doc for thermal_zone_device_register
  thermal: update kernel-doc for create_trip_attrs
  thermal: update kernel-doc for thermal_cooling_device_register
  thermal: update kernel-doc for thermal_zone_unbind_cooling_device
  thermal: update kernel-doc for thermal_zone_bind_cooling_device
  thermal: use EXPORT_SYMBOL_GPL
  thermal: rename notify_thermal_framework to thermal_notify_framework
  thermal: update driver license
  thermal: use strlcpy instead of strcpy
  ...

11 years agoEDAC: Don't give write permission to read-only files
Srivatsa S. Bhat [Tue, 30 Apr 2013 09:47:16 +0000 (15:17 +0530)]
EDAC: Don't give write permission to read-only files

I get the following warning on boot:

------------[ cut here ]------------
WARNING: at drivers/base/core.c:575 device_create_file+0x9a/0xa0()
Hardware name:  -[8737R2A]-
Write permission without 'store'
...
</snip>

Drilling down, this is related to dynamic channel ce_count attribute
files sporting a S_IWUSR mode without a ->store() function. Looking
around, it appears that they aren't supposed to have a ->store()
function. So remove the bogus write permission to get rid of the
warning.

Signed-off-by: Srivatsa S. Bhat <srivatsa.bhat@linux.vnet.ibm.com>
Cc: Mauro Carvalho Chehab <mchehab@redhat.com>
Cc: <stable@vger.kernel.org> # 3.[89]
[ shorten commit message ]
Signed-off-by: Borislav Petkov <bp@suse.de>
11 years agowatchdog: Fix race condition in registration code
Guenter Roeck [Sat, 6 Apr 2013 04:22:43 +0000 (21:22 -0700)]
watchdog: Fix race condition in registration code

A race condition exists when registering the first watchdog device.
Sequence of events:

- watchdog_register_device calls watchdog_dev_register
- watchdog_dev_register creates the watchdog misc device by calling
  misc_register.
  At that time, the matching character device (/dev/watchdog0) does not yet
  exist, and old_wdd is not set either.
- Userspace gets an event and opens /dev/watchdog
- watchdog_open is called and sets wdd = old_wdd, which is still NULL,
  and tries to dereference it. This causes the kernel to panic.

Seen with systemd trying to open /dev/watchdog immediately after
it was created.

Reported-by: Arkadiusz Miskiewicz <arekm@maven.pl>
Signed-off-by: Guenter Roeck <linux@roeck-us.net>
Tested-by: Arkadiusz Miskiewicz <arekm@maven.pl>
Signed-off-by: Wim Van Sebroeck <wim@iguana.be>
11 years agowatchdog: Convert to devm_ioremap_resource()
Sachin Kamat [Mon, 4 Mar 2013 05:06:41 +0000 (10:36 +0530)]
watchdog: Convert to devm_ioremap_resource()

Use the newly introduced devm_ioremap_resource() instead of
devm_request_and_ioremap() which provides more consistent error handling.

devm_ioremap_resource() provides its own error messages; so all explicit
error messages can be removed from the failure code paths.

Signed-off-by: Sachin Kamat <sachin.kamat@linaro.org>
Reviewed-by: Thierry Reding <thierry.reding@avionic-design.de>
Cc: Gabor Juhos <juhosg@openwrt.org>
Cc: Paul Mundt <lethal@linux-sh.org>
Signed-off-by: Wim Van Sebroeck <wim@iguana.be>
11 years agoMerge tag 'rdma-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/roland...
Linus Torvalds [Wed, 8 May 2013 22:29:48 +0000 (15:29 -0700)]
Merge tag 'rdma-for-linus' of git://git./linux/kernel/git/roland/infiniband

Pull InfiniBand/RDMA changes from Roland Dreier:
 - XRC transport fixes
 - Fix DHCP on IPoIB
 - mlx4 preparations for flow steering
 - iSER fixes
 - miscellaneous other fixes

* tag 'rdma-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/roland/infiniband: (23 commits)
  IB/iser: Add support for iser CM REQ additional info
  IB/iser: Return error to upper layers on EAGAIN registration failures
  IB/iser: Move informational messages from error to info level
  IB/iser: Add module version
  mlx4_core: Expose a few helpers to fill DMFS HW strucutures
  mlx4_core: Directly expose fields of DMFS HW rule control segment
  mlx4_core: Change a few DMFS fields names to match firmare spec
  mlx4: Match DMFS promiscuous field names to firmware spec
  mlx4_core: Move DMFS HW structs to common header file
  IB/mlx4: Set link type for RAW PACKET QPs in the QP context
  IB/mlx4: Disable VLAN stripping for RAW PACKET QPs
  mlx4_core: Reduce warning message for SRQ_LIMIT event to debug level
  RDMA/iwcm: Don't touch cmid after dropping reference
  IB/qib: Correct qib_verbs_register_sysfs() error handling
  IB/ipath: Correct ipath_verbs_register_sysfs() error handling
  RDMA/cxgb4: Fix SQ allocation when on-chip SQ is disabled
  SRPT: Fix odd use of WARN_ON()
  IPoIB: Fix ipoib_hard_header() return value
  RDMA: Rename random32() to prandom_u32()
  RDMA/cxgb3: Fix uninitialized variable
  ...

11 years agoMerge tag 'arm64-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/cmarinas...
Linus Torvalds [Wed, 8 May 2013 22:15:27 +0000 (15:15 -0700)]
Merge tag 'arm64-for-linus' of git://git./linux/kernel/git/cmarinas/linux-aarch64

Pull arm64 update from Catalin Marinas:

 - Since drivers/irqchip/irq-gic.c no longer has dependencies on arm32
   specifics (the 'gic' branch merged), it can be enabled on arm64.

 - Enable arm64 support for poweroff/restart (for code under
   drivers/power/reset/).

 - Fixes (dts file, exception handling, bitops)

* tag 'arm64-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/cmarinas/linux-aarch64:
  arm64: Treat the bitops index argument as an 'int'
  arm64: Ignore the 'write' ESR flag on cache maintenance faults
  arm64: dts: fix #address-cells for foundation-v8
  arm64: vexpress: Add support for poweroff/restart
  arm64: Enable support for the ARM GIC interrupt controller

11 years agoMerge tag 'f2fs-for-v3.10' of git://git.kernel.org/pub/scm/linux/kernel/git/jaegeuk...
Linus Torvalds [Wed, 8 May 2013 22:11:48 +0000 (15:11 -0700)]
Merge tag 'f2fs-for-v3.10' of git://git./linux/kernel/git/jaegeuk/f2fs

Pull f2fs updates from Jaegeuk Kim:
 "This patch-set includes the following major enhancement patches.
   - introduce a new gloabl lock scheme
   - add tracepoints on several major functions
   - fix the overall cleaning process focused on victim selection
   - apply the block plugging to merge IOs as much as possible
   - enhance management of free nids and its list
   - enhance the readahead mode for node pages
   - address several cretical deadlock conditions
   - reduce lock_page calls

  The other minor bug fixes and enhancements are as follows.
   - calculation mistakes: overflow
   - bio types: READ, READA, and READ_SYNC
   - fix the recovery flow, data races, and null pointer errors"

* tag 'f2fs-for-v3.10' of git://git.kernel.org/pub/scm/linux/kernel/git/jaegeuk/f2fs: (68 commits)
  f2fs: cover free_nid management with spin_lock
  f2fs: optimize scan_nat_page()
  f2fs: code cleanup for scan_nat_page() and build_free_nids()
  f2fs: bugfix for alloc_nid_failed()
  f2fs: recover when journal contains deleted files
  f2fs: continue to mount after failing recovery
  f2fs: avoid deadlock during evict after f2fs_gc
  f2fs: modify the number of issued pages to merge IOs
  f2fs: remove useless #include <linux/proc_fs.h> as we're now using sysfs as debug entry.
  f2fs: fix inconsistent using of NM_WOUT_THRESHOLD
  f2fs: check truncation of mapping after lock_page
  f2fs: enhance alloc_nid and build_free_nids flows
  f2fs: add a tracepoint on f2fs_new_inode
  f2fs: check nid == 0 in add_free_nid
  f2fs: add REQ_META about metadata requests for submit
  f2fs: give a chance to merge IOs by IO scheduler
  f2fs: avoid frequent background GC
  f2fs: add tracepoints to debug checkpoint request
  f2fs: add tracepoints for write page operations
  f2fs: add tracepoints to debug the block allocation
  ...

11 years agoMerge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/rkuo/linux...
Linus Torvalds [Wed, 8 May 2013 22:08:59 +0000 (15:08 -0700)]
Merge branch 'for-linus' of git://git./linux/kernel/git/rkuo/linux-hexagon-kernel

Pull Hexagon fixes from Richard Kuo:
 "A bug fix and a Kconfig cleanup"

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/rkuo/linux-hexagon-kernel:
  HEXAGON: Remove non existent reference to GENERIC_KERNEL_EXECVE & GENERIC_KERNEL_THREAD
  Hexagon: fix register used to call do_work_pending

11 years agomm/slab: Fix crash during slab init
Chris Mason [Wed, 8 May 2013 19:56:28 +0000 (15:56 -0400)]
mm/slab: Fix crash during slab init

Commit 8a965b3baa89 ("mm, slab_common: Fix bootstrap creation of kmalloc
caches") introduced a regression that caused us to crash early during
boot.  The commit was introducing ordering of slab creation, making sure
two odd-sized slabs were created after specific powers of two sizes.

But, if any of the power of two slabs were created earlier during boot,
slabs at index 1 or 2 might not get created at all.  This patch makes
sure none of the slabs get skipped.

Tony Lindgren bisected this down to the offending commit, which really
helped because bisect kept bringing me to almost but not quite this one.

Signed-off-by: Chris Mason <chris.mason@fusionio.com>
Acked-by: Christoph Lameter <cl@linux.com>
Acked-by: Tony Lindgren <tony@atomide.com>
Acked-by: Soren Brinkmann <soren.brinkmann@xilinx.com>
Tested-by: Tetsuo Handa <penguin-kernel@i-love.sakura.ne.jp>
Tested-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
11 years agoNFS4.1 Fix data server connection race
Andy Adamson [Wed, 8 May 2013 20:21:18 +0000 (16:21 -0400)]
NFS4.1 Fix data server connection race

Unlike meta data server mounts which support multiple mount points to
the same server via struct nfs_server, data servers support a single connection.

Concurrent calls to setup the data server connection can race where the first
call allocates the nfs_client struct, and before the cache struct nfs_client
pointer can be set, a second call also tries to setup the connection, finds the
already allocated nfs_client, bumps the reference count, re-initializes the
session,etc. This results in a hanging data server session after umount.

Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
11 years agoMerge branches 'cxgb4', 'ipoib', 'iser', 'misc', 'mlx4', 'qib' and 'srp' into for...
Roland Dreier [Wed, 8 May 2013 21:12:37 +0000 (14:12 -0700)]
Merge branches 'cxgb4', 'ipoib', 'iser', 'misc', 'mlx4', 'qib' and 'srp' into for-next

11 years agoif_cablemodem.h: Add parenthesis around ioctl macros
Josh Boyer [Wed, 8 May 2013 09:45:47 +0000 (09:45 +0000)]
if_cablemodem.h: Add parenthesis around ioctl macros

Protect the SIOCGCM* ioctl macros with parenthesis.

Reported-by: Paul Wouters <pwouters@redhat.com>
Signed-off-by: Josh Boyer <jwboyer@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
11 years agogso: Handle Trans-Ether-Bridging protocol in skb_network_protocol()
Pravin B Shelar [Tue, 7 May 2013 20:41:07 +0000 (20:41 +0000)]
gso: Handle Trans-Ether-Bridging protocol in skb_network_protocol()

Rather than having logic to calculate inner protocol in every
tunnel gso handler move it to gso code. This simplifies code.

Cc: Eric Dumazet <eric.dumazet@gmail.com>
Cc: Cong Wang <amwang@redhat.com>
Cc: David S. Miller <davem@davemloft.net>
Signed-off-by: Pravin B Shelar <pshelar@nicira.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
11 years agonet: fec: fix kernel oops when plug/unplug cable many times
Frank Li [Tue, 7 May 2013 14:08:44 +0000 (14:08 +0000)]
net: fec: fix kernel oops when plug/unplug cable many times

reproduce steps
 1. flood ping from other machine
  ping -f -s 41000 IP
 2. run below script
    while [ 1 ]; do ethtool -s eth0 autoneg off;
    sleep 3;ethtool -s eth0 autoneg on; sleep 4; done;

You can see oops in one hour.

The reason is fec_restart clear BD but NAPI may use it.
The solution is disable NAPI and stop xmit when reset BD.
disable NAPI may sleep, so fec_restart can't be call in
atomic context.

Signed-off-by: Frank Li <Frank.Li@freescale.com>
Reviewed-by: Lucas Stach <l.stach@pengutronix.de>
Tested-by: Lucas Stach <l.stach@pengutronix.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
11 years agobfin_mac: fix error return code in bfin_mac_probe()
Wei Yongjun [Tue, 7 May 2013 02:23:38 +0000 (02:23 +0000)]
bfin_mac: fix error return code in bfin_mac_probe()

Fix to return a negative error code from the error handling
case instead of 0, as done elsewhere in this function.

Signed-off-by: Wei Yongjun <yongjun_wei@trendmicro.com.cn>
Signed-off-by: David S. Miller <davem@davemloft.net>
11 years agosfc: fix return value check in efx_ptp_probe_channel()
Wei Yongjun [Tue, 7 May 2013 02:19:25 +0000 (02:19 +0000)]
sfc: fix return value check in efx_ptp_probe_channel()

In case of error, the function ptp_clock_register() returns ERR_PTR()
and never returns NULL. The NULL test in the return value check should
be replaced with IS_ERR().

Signed-off-by: Wei Yongjun <yongjun_wei@trendmicro.com.cn>
Reviewed-by: Ben Hutchings <bhutchings@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
11 years agonet/spider_net: fix error return code in spider_net_open()
Wei Yongjun [Tue, 7 May 2013 02:16:00 +0000 (02:16 +0000)]
net/spider_net: fix error return code in spider_net_open()

Fix to return a negative error code from the error handling
case instead of 0, as done elsewhere in this function.

Signed-off-by: Wei Yongjun <yongjun_wei@trendmicro.com.cn>
Signed-off-by: David S. Miller <davem@davemloft.net>
11 years agonet/irda: fix error return code in bfin_sir_open()
Wei Yongjun [Tue, 7 May 2013 02:14:33 +0000 (02:14 +0000)]
net/irda: fix error return code in bfin_sir_open()

Fix to return -ENOMEM in the irlap_open() error handling case instead
of 0(overwrite to 0 by bfin_sir_startup()), as done elsewhere in this
function.

Signed-off-by: Wei Yongjun <yongjun_wei@trendmicro.com.cn>
Signed-off-by: David S. Miller <davem@davemloft.net>
11 years agonet: of_mdio: fix behavior on missing phy device
Sebastian Hesselbarth [Mon, 6 May 2013 23:49:31 +0000 (23:49 +0000)]
net: of_mdio: fix behavior on missing phy device

of_mdiobus_register creates a phy_device even if get_phy_device failed
to create it previously. This causes indefinite polling on non-existent
PHYs. This fix makes of_mdio_register rely on get_phy_device to
properly create the device or fail otherwise.

Signed-off-by: Sebastian Hesselbarth <sebastian.hesselbarth@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
11 years agosierra_net: keep status interrupt URB active
Dan Williams [Mon, 6 May 2013 11:34:56 +0000 (11:34 +0000)]
sierra_net: keep status interrupt URB active

The driver and firmware sync up through SYNC messages, and the
firmware's affirmative reply to these SYNC messages appears to be the
"Reset" indication received via the status interrupt endpoint.  Thus the
driver needs the status interrupt endpoint always active so that the
Reset indication can be received even if the netdev is closed, which is
the case right after device insertion.

If the Reset indication is not received by the driver, it continues
sending SYNC messages to the firmware, which crashes about 10 seconds
later and the device stops responding.

Signed-off-by: Dan Williams <dcbw@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
11 years agousbnet: allow status interrupt URB to always be active
Dan Williams [Mon, 6 May 2013 11:29:23 +0000 (11:29 +0000)]
usbnet: allow status interrupt URB to always be active

Some drivers (sierra_net) need the status interrupt URB
active even when the device is closed, because they receive
custom indications from firmware.  Add functions to refcount
the status interrupt URB submit/kill operation so that
sub-drivers and the generic driver don't fight over whether
the status interrupt URB is active or not.

A sub-driver can call usbnet_status_start() at any time, but
the URB is only submitted the first time the function is
called.  Likewise, when the sub-driver is done with the URB,
it calls usbnet_status_stop() but the URB is only killed when
all users have stopped it.  The URB is still killed and
re-submitted for suspend/resume, as before, with the same
refcount it had at suspend.

Signed-off-by: Dan Williams <dcbw@redhat.com>
Acked-by: Oliver Neukum <oliver@neukum.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
11 years agoqmi_wwan/cdc_ether: add device IDs for Dell 5804 (Novatel E371) WWAN card
Dan Williams [Mon, 6 May 2013 11:17:37 +0000 (11:17 +0000)]
qmi_wwan/cdc_ether: add device IDs for Dell 5804 (Novatel E371) WWAN card

A rebranded Novatel E371 for AT&T's LTE bands.  qmi_wwan should drive this
device, while cdc_ether should ignore it.  Even though the USB descriptors
are plain CDC-ETHER that USB interface is a QMI interface.

Cc: <stable@vger.kernel.org>
Signed-off-by: Dan Williams <dcbw@redhat.com>
Acked-by: Bjørn Mork <bjorn@mork.no>
Signed-off-by: David S. Miller <davem@davemloft.net>
11 years agobe2net: disable TX in be_close()
Sathya Perla [Wed, 8 May 2013 02:05:50 +0000 (02:05 +0000)]
be2net: disable TX in be_close()

be_close() followed by be_clear() is called as a part of cleanup in the
EEH/AER flow.  This patch stops TX in be_close() before cleaning/freeing
up the TX queues in be_clear(). This prevents be_xmit() from being called
while TX queues no longer exist.

Signed-off-by: Sathya Perla <sathya.perla@emulex.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
11 years agobe2net: fix EQ from getting full while cleaning RX CQ
Sathya Perla [Wed, 8 May 2013 02:05:49 +0000 (02:05 +0000)]
be2net: fix EQ from getting full while cleaning RX CQ

While cleaning RX queues, the CQ DB may be rung several times (with rearm)
while waiting for the flush compl. Each CQ-notify with rearm can result in
an event. The EQ may get full resulting in a HW error.

Fix this by not re-arming the CQ while notifying a valid completion.
Also, there's no need to wait for 1ms after destroying RXQ, as the code in
be_rx_cq_clean() waits for the flush compl to arrive.

Signed-off-by: Sathya Perla <sathya.perla@emulex.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
11 years agobe2net: fix payload_len value for GET_MAC_LIST cmd req
Sathya Perla [Wed, 8 May 2013 02:05:48 +0000 (02:05 +0000)]
be2net: fix payload_len value for GET_MAC_LIST cmd req

The buffer size for a FW cmd request must be big enough to fit the response,
else the cmd fails. For GET_MAC_LIST cmd, though the memory allocated for
the cmd is big enough to fit the response, the payload_len value in the
WRB hdr is being set to the request length only.
Fix this for GET_MAC_LIST cmd.

Signed-off-by: Sathya Perla <sathya.perla@emulex.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
11 years agobe2net: provision VF resources before enabling SR-IOV
Sathya Perla [Wed, 8 May 2013 02:05:47 +0000 (02:05 +0000)]
be2net: provision VF resources before enabling SR-IOV

When the PF driver calls pci_enable_sriov(), the VFs may be probed
inline before the call returns. So, the resources required for all VFs
must be provisioned by the PF driver *before* calling pci_enable_sriov();
else, VF probe will fail.

Signed-off-by: Sathya Perla <sathya.perla@emulex.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
11 years agoMerge branch 'for-3.10/drivers' of git://git.kernel.dk/linux-block
Linus Torvalds [Wed, 8 May 2013 18:51:05 +0000 (11:51 -0700)]
Merge branch 'for-3.10/drivers' of git://git.kernel.dk/linux-block

Pull block driver updates from Jens Axboe:
 "It might look big in volume, but when categorized, not a lot of
  drivers are touched.  The pull request contains:

   - mtip32xx fixes from Micron.

   - A slew of drbd updates, this time in a nicer series.

   - bcache, a flash/ssd caching framework from Kent.

   - Fixes for cciss"

* 'for-3.10/drivers' of git://git.kernel.dk/linux-block: (66 commits)
  bcache: Use bd_link_disk_holder()
  bcache: Allocator cleanup/fixes
  cciss: bug fix to prevent cciss from loading in kdump crash kernel
  cciss: add cciss_allow_hpsa module parameter
  drivers/block/mg_disk.c: add CONFIG_PM_SLEEP to suspend/resume functions
  mtip32xx: Workaround for unaligned writes
  bcache: Make sure blocksize isn't smaller than device blocksize
  bcache: Fix merge_bvec_fn usage for when it modifies the bvm
  bcache: Correctly check against BIO_MAX_PAGES
  bcache: Hack around stuff that clones up to bi_max_vecs
  bcache: Set ra_pages based on backing device's ra_pages
  bcache: Take data offset from the bdev superblock.
  mtip32xx: mtip32xx: Disable TRIM support
  mtip32xx: fix a smatch warning
  bcache: Disable broken btree fuzz tester
  bcache: Fix a format string overflow
  bcache: Fix a minor memory leak on device teardown
  bcache: Documentation updates
  bcache: Use WARN_ONCE() instead of __WARN()
  bcache: Add missing #include <linux/prefetch.h>
  ...

11 years agoMerge branch 'for-3.10/core' of git://git.kernel.dk/linux-block
Linus Torvalds [Wed, 8 May 2013 17:13:35 +0000 (10:13 -0700)]
Merge branch 'for-3.10/core' of git://git.kernel.dk/linux-block

Pull block core updates from Jens Axboe:

 - Major bit is Kents prep work for immutable bio vecs.

 - Stable candidate fix for a scheduling-while-atomic in the queue
   bypass operation.

 - Fix for the hang on exceeded rq->datalen 32-bit unsigned when merging
   discard bios.

 - Tejuns changes to convert the writeback thread pool to the generic
   workqueue mechanism.

 - Runtime PM framework, SCSI patches exists on top of these in James'
   tree.

 - A few random fixes.

* 'for-3.10/core' of git://git.kernel.dk/linux-block: (40 commits)
  relay: move remove_buf_file inside relay_close_buf
  partitions/efi.c: replace useless kzalloc's by kmalloc's
  fs/block_dev.c: fix iov_shorten() criteria in blkdev_aio_read()
  block: fix max discard sectors limit
  blkcg: fix "scheduling while atomic" in blk_queue_bypass_start
  Documentation: cfq-iosched: update documentation help for cfq tunables
  writeback: expose the bdi_wq workqueue
  writeback: replace custom worker pool implementation with unbound workqueue
  writeback: remove unused bdi_pending_list
  aoe: Fix unitialized var usage
  bio-integrity: Add explicit field for owner of bip_buf
  block: Add an explicit bio flag for bios that own their bvec
  block: Add bio_alloc_pages()
  block: Convert some code to bio_for_each_segment_all()
  block: Add bio_for_each_segment_all()
  bounce: Refactor __blk_queue_bounce to not use bi_io_vec
  raid1: use bio_copy_data()
  pktcdvd: Use bio_reset() in disabled code to kill bi_idx usage
  pktcdvd: use bio_copy_data()
  block: Add bio_copy_data()
  ...

11 years agof2fs: cover free_nid management with spin_lock
Jaegeuk Kim [Tue, 7 May 2013 11:47:40 +0000 (20:47 +0900)]
f2fs: cover free_nid management with spin_lock

After build_free_nids() searches free nid candidates from nat pages and
current journal blocks, it checks all the candidates if they are allocated
so that the nat cache has its nid with an allocated block address.

In this procedure, previously we used
    list_for_each_entry_safe(fnid, next_fnid, &nm_i->free_nid_list, list).
But, this is not covered by free_nid_list_lock, resulting in null pointer bug.

This patch moves this checking routine inside add_free_nid() in order not to use
the spin_lock.

Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
11 years agof2fs: optimize scan_nat_page()
Haicheng Li [Mon, 6 May 2013 15:15:43 +0000 (23:15 +0800)]
f2fs: optimize scan_nat_page()

When nm_i->fcnt > 2 * MAX_FREE_NIDS, stop scanning other NAT entries.

Signed-off-by: Haicheng Li <haicheng.li@linux.intel.com>
[Jaegeuk Kim: fix handling the return value of add_free_nid()]
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
11 years agof2fs: code cleanup for scan_nat_page() and build_free_nids()
Haicheng Li [Mon, 6 May 2013 15:15:42 +0000 (23:15 +0800)]
f2fs: code cleanup for scan_nat_page() and build_free_nids()

This patch does two cleanups:
1. remove unused variable "fcnt" in build_free_nids().
2. make scan_nat_page() as void type and remove useless variable "fcnt".

Signed-off-by: Haicheng Li <haicheng.li@linux.intel.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
11 years agof2fs: bugfix for alloc_nid_failed()
Haicheng Li [Mon, 6 May 2013 15:15:41 +0000 (23:15 +0800)]
f2fs: bugfix for alloc_nid_failed()

Directly drop the free_nid cache when nm_i->fcnt > 2 * MAX_FREE_NIDS

Since there is NOT nmi->free_nid_list_lock spinlock protection between
a sequential calling of alloc_nid() and alloc_nid_failed(), some other
threads may already add new free_nid to the free_nid_list during this
period.

We need to make sure nmi->fcnt is never > 2 * MAX_FREE_NIDS.

Signed-off-by: Haicheng Li <haicheng.li@linux.intel.com>
[Jaegeuk Kim: fit the coding style]
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
11 years agof2fs: recover when journal contains deleted files
Chris Fries [Thu, 2 May 2013 21:09:05 +0000 (16:09 -0500)]
f2fs: recover when journal contains deleted files

When recovering a journal file with fsync data for files that have
been deleted, don't bail out on recovery.

Signed-off-by: Chris Fries <C.Fries@motorola.com>
Reviewed-by: Russell Knize <rknize2@motorola.com>
Reviewed-by: Jason Hrycay <jason.hrycay@motorola.com>
[Jaegeuk Kim: fit the coding style]
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
11 years agof2fs: continue to mount after failing recovery
Chris Fries [Thu, 2 May 2013 21:07:34 +0000 (16:07 -0500)]
f2fs: continue to mount after failing recovery

When unable to roll forward the journal, we shouldn't bail out and
not mount, we should continue to attempt the mount.  Bad recovery data
is likely unrecoverable at this point, and requiring the user to try
to mount again doesn't solve any issues.

Signed-off-by: Chris Fries <C.Fries@motorola.com>
Reviewed-by: Russell Knize <rknize2@motorola.com>
Reviewed-by: Jason Hrycay <jason.hrycay@motorola.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
11 years agof2fs: avoid deadlock during evict after f2fs_gc
Jaegeuk Kim [Tue, 30 Apr 2013 02:33:27 +0000 (11:33 +0900)]
f2fs: avoid deadlock during evict after f2fs_gc

o Deadlock case #1

Thread 1:
- writeback_sb_inodes
 - do_writepages
  - f2fs_write_data_pages
   - write_cache_pages
    - f2fs_write_data_page
     - f2fs_balance_fs
      - wait mutex_lock(gc_mutex)

Thread 2:
- f2fs_balance_fs
 - mutex_lock(gc_mutex)
 - f2fs_gc
  - f2fs_iget
   - wait iget_locked(inode->i_lock)

Thread 3:
- do_unlinkat
 - iput
  - lock(inode->i_lock)
   - evict
    - inode_wait_for_writeback

o Deadlock case #2

Thread 1:
- __writeback_single_inode
 : set I_SYNC
  - do_writepages
   - f2fs_write_data_page
    - f2fs_balance_fs
     - f2fs_gc
      - iput
       - evict
        - inode_wait_for_writeback(I_SYNC)

In order to avoid this, even though iput is called with the zero-reference
count, we need to stop the eviction procedure if the inode is on writeback.
So this patch links f2fs_drop_inode which checks the I_SYNC flag.

Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
11 years agoarm64: Treat the bitops index argument as an 'int'
Catalin Marinas [Tue, 7 May 2013 17:02:58 +0000 (18:02 +0100)]
arm64: Treat the bitops index argument as an 'int'

The bitops prototype use an 'int' as the bit index type but the asm
implementation assume it to be a 'long'. Since the compiler does not
guarantee zeroing the upper 32-bits in a register when used as 'int',
change the bitops implementation accordingly.

Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
11 years agoarm64: Ignore the 'write' ESR flag on cache maintenance faults
Catalin Marinas [Tue, 7 May 2013 15:57:06 +0000 (16:57 +0100)]
arm64: Ignore the 'write' ESR flag on cache maintenance faults

ESR.WnR bit is always set on data cache maintenance faults even though
the page is not required to have write permission. If a translation
fault (page not yet mapped) happens for read-only user address range,
Linux incorrectly assumes a permission fault. This patch adds the check
of the ESR.CM bit during the page fault handling to ignore the 'write'
flag.

Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
Reported-by: Tim Northover <Tim.Northover@arm.com>
Cc: stable@vger.kernel.org
11 years agoarm64: dts: fix #address-cells for foundation-v8
Mark Rutland [Tue, 7 May 2013 13:04:03 +0000 (14:04 +0100)]
arm64: dts: fix #address-cells for foundation-v8

Commit 90556ca1 ("arm64: vexpress: Add dts files for the ARMv8 RTSM
models") added foundation-v8.dts, but erroneously set
/cpus/#address-cells = <1> while providing two cells in each cpus/cpu@N
node's reg property.

As of commit ea393a2e ("arm64: smp: honour #address-size when parsing
CPU reg property") we read in as many address cells as specified rather
than always reading two. This means that for foundation-v8.dts, we only
read the first reg cell (zero) for each cpu node, and receive a lot of
warnings at boot of the form "/cpus/cpu@1: duplicate cpu reg properties
in the DT".

This patch corrects foundation-v8.dts to have the correct value for
/cpus/#address-cells.

Signed-off-by: Mark Rutland <mark.rutland@arm.com>
Cc: Catalin Marinas <catalin.marinas@arm.com>
Cc: Pawel Moll <pawel.moll@arm.com>
Cc: Will Deacon <will.deacon@arm.com>
Tested-by: Marc Zyngier <marc.zyngier@arm.com>
Acked-by: Marc Zyngier <marc.zyngier@arm.com>
Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
11 years agoarm64: vexpress: Add support for poweroff/restart
Catalin Marinas [Thu, 28 Feb 2013 18:14:37 +0000 (18:14 +0000)]
arm64: vexpress: Add support for poweroff/restart

This patch adds the arm_pm_poweroff definition expected by the
vexpress-poweroff.c driver and enables the latter for arm64.

Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
Acked-by: Pawel Moll <pawel.moll@arm.com>
11 years agoarm64: Enable support for the ARM GIC interrupt controller
Catalin Marinas [Mon, 14 Jan 2013 12:39:31 +0000 (12:39 +0000)]
arm64: Enable support for the ARM GIC interrupt controller

This patch enables ARM_GIC on the arm64 kernel.

Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
11 years agoMerge branch 'gic' into HEAD
Catalin Marinas [Wed, 8 May 2013 09:22:07 +0000 (10:22 +0100)]
Merge branch 'gic' into HEAD

* arm64-prep-gic:
  irqchip: gic: Perform the gic_secondary_init() call via CPU notifier
  irqchip: gic: Call handle_bad_irq() directly
  arm: Move chained_irq_(enter|exit) to a generic file
  arm: Move the set_handle_irq and handle_arch_irq declarations to asm/irq.h

11 years agoMerge branch 'akpm' (incoming from Andrew)
Linus Torvalds [Wed, 8 May 2013 03:49:51 +0000 (20:49 -0700)]
Merge branch 'akpm' (incoming from Andrew)

Merge more incoming from Andrew Morton:

 - Various fixes which were stalled or which I picked up recently

 - A large rotorooting of the AIO code.  Allegedly to improve
   performance but I don't really have good performance numbers (I might
   have lost the email) and I can't raise Kent today.  I held this out
   of 3.9 and we could give it another cycle if it's all too late/scary.

I ended up taking only the first two thirds of the AIO rotorooting.  I
left the percpu parts and the batch completion for later.  - Linus

* emailed patches from Andrew Morton <akpm@linux-foundation.org>: (33 commits)
  aio: don't include aio.h in sched.h
  aio: kill ki_retry
  aio: kill ki_key
  aio: give shared kioctx fields their own cachelines
  aio: kill struct aio_ring_info
  aio: kill batch allocation
  aio: change reqs_active to include unreaped completions
  aio: use cancellation list lazily
  aio: use flush_dcache_page()
  aio: make aio_read_evt() more efficient, convert to hrtimers
  wait: add wait_event_hrtimeout()
  aio: refcounting cleanup
  aio: make aio_put_req() lockless
  aio: do fget() after aio_get_req()
  aio: dprintk() -> pr_debug()
  aio: move private stuff out of aio.h
  aio: add kiocb_cancel()
  aio: kill return value of aio_complete()
  char: add aio_{read,write} to /dev/{null,zero}
  aio: remove retry-based AIO
  ...

11 years agoaio: don't include aio.h in sched.h
Kent Overstreet [Tue, 7 May 2013 23:19:08 +0000 (16:19 -0700)]
aio: don't include aio.h in sched.h

Faster kernel compiles by way of fewer unnecessary includes.

[akpm@linux-foundation.org: fix fallout]
[akpm@linux-foundation.org: fix build]
Signed-off-by: Kent Overstreet <koverstreet@google.com>
Cc: Zach Brown <zab@redhat.com>
Cc: Felipe Balbi <balbi@ti.com>
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: Mark Fasheh <mfasheh@suse.com>
Cc: Joel Becker <jlbec@evilplan.org>
Cc: Rusty Russell <rusty@rustcorp.com.au>
Cc: Jens Axboe <axboe@kernel.dk>
Cc: Asai Thambi S P <asamymuthupa@micron.com>
Cc: Selvan Mani <smani@micron.com>
Cc: Sam Bradshaw <sbradshaw@micron.com>
Cc: Jeff Moyer <jmoyer@redhat.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Benjamin LaHaise <bcrl@kvack.org>
Reviewed-by: "Theodore Ts'o" <tytso@mit.edu>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
11 years agoaio: kill ki_retry
Kent Overstreet [Tue, 7 May 2013 23:19:11 +0000 (16:19 -0700)]
aio: kill ki_retry

Thanks to Zach Brown's work to rip out the retry infrastructure, we don't
need this anymore - ki_retry was only called right after the kiocb was
initialized.

This also refactors and trims some duplicated code, as well as cleaning up
the refcounting/error handling a bit.

[akpm@linux-foundation.org: use fmode_t in aio_run_iocb()]
[akpm@linux-foundation.org: fix file_start_write/file_end_write tests]
[akpm@linux-foundation.org: coding-style fixes]
Signed-off-by: Kent Overstreet <koverstreet@google.com>
Cc: Zach Brown <zab@redhat.com>
Cc: Felipe Balbi <balbi@ti.com>
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: Mark Fasheh <mfasheh@suse.com>
Cc: Joel Becker <jlbec@evilplan.org>
Cc: Rusty Russell <rusty@rustcorp.com.au>
Cc: Jens Axboe <axboe@kernel.dk>
Cc: Asai Thambi S P <asamymuthupa@micron.com>
Cc: Selvan Mani <smani@micron.com>
Cc: Sam Bradshaw <sbradshaw@micron.com>
Cc: Jeff Moyer <jmoyer@redhat.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Benjamin LaHaise <bcrl@kvack.org>
Reviewed-by: "Theodore Ts'o" <tytso@mit.edu>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
11 years agoaio: kill ki_key
Kent Overstreet [Tue, 7 May 2013 23:19:10 +0000 (16:19 -0700)]
aio: kill ki_key

ki_key wasn't actually used for anything previously - it was always 0.
Drop it to trim struct kiocb a bit.

Signed-off-by: Kent Overstreet <koverstreet@google.com>
Cc: Zach Brown <zab@redhat.com>
Cc: Felipe Balbi <balbi@ti.com>
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: Mark Fasheh <mfasheh@suse.com>
Cc: Joel Becker <jlbec@evilplan.org>
Cc: Rusty Russell <rusty@rustcorp.com.au>
Cc: Jens Axboe <axboe@kernel.dk>
Cc: Asai Thambi S P <asamymuthupa@micron.com>
Cc: Selvan Mani <smani@micron.com>
Cc: Sam Bradshaw <sbradshaw@micron.com>
Cc: Jeff Moyer <jmoyer@redhat.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Benjamin LaHaise <bcrl@kvack.org>
Reviewed-by: "Theodore Ts'o" <tytso@mit.edu>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
11 years agoaio: give shared kioctx fields their own cachelines
Kent Overstreet [Tue, 7 May 2013 23:18:56 +0000 (16:18 -0700)]
aio: give shared kioctx fields their own cachelines

[akpm@linux-foundation.org: make reqs_active __cacheline_aligned_in_smp]
Signed-off-by: Kent Overstreet <koverstreet@google.com>
Cc: Zach Brown <zab@redhat.com>
Cc: Felipe Balbi <balbi@ti.com>
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: Mark Fasheh <mfasheh@suse.com>
Cc: Joel Becker <jlbec@evilplan.org>
Cc: Rusty Russell <rusty@rustcorp.com.au>
Cc: Jens Axboe <axboe@kernel.dk>
Cc: Asai Thambi S P <asamymuthupa@micron.com>
Cc: Selvan Mani <smani@micron.com>
Cc: Sam Bradshaw <sbradshaw@micron.com>
Cc: Jeff Moyer <jmoyer@redhat.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Benjamin LaHaise <bcrl@kvack.org>
Reviewed-by: "Theodore Ts'o" <tytso@mit.edu>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
11 years agoaio: kill struct aio_ring_info
Kent Overstreet [Tue, 7 May 2013 23:18:55 +0000 (16:18 -0700)]
aio: kill struct aio_ring_info

struct aio_ring_info was kind of odd, the only place it's used is where
it's embedded in struct kioctx - there's no real need for it.

The next patch rearranges struct kioctx and puts various things on their
own cachelines - getting rid of struct aio_ring_info now makes that
reordering a bit clearer.

Signed-off-by: Kent Overstreet <koverstreet@google.com>
Cc: Zach Brown <zab@redhat.com>
Cc: Felipe Balbi <balbi@ti.com>
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: Mark Fasheh <mfasheh@suse.com>
Cc: Joel Becker <jlbec@evilplan.org>
Cc: Rusty Russell <rusty@rustcorp.com.au>
Cc: Jens Axboe <axboe@kernel.dk>
Cc: Asai Thambi S P <asamymuthupa@micron.com>
Cc: Selvan Mani <smani@micron.com>
Cc: Sam Bradshaw <sbradshaw@micron.com>
Cc: Jeff Moyer <jmoyer@redhat.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Benjamin LaHaise <bcrl@kvack.org>
Reviewed-by: "Theodore Ts'o" <tytso@mit.edu>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
11 years agoaio: kill batch allocation
Kent Overstreet [Tue, 7 May 2013 23:18:53 +0000 (16:18 -0700)]
aio: kill batch allocation

Previously, allocating a kiocb required touching quite a few global
(well, per kioctx) cachelines...  so batching up allocation to amortize
those was worthwhile.  But we've gotten rid of some of those, and in
another couple of patches kiocb allocation won't require writing to any
shared cachelines, so that means we can just rip this code out.

Signed-off-by: Kent Overstreet <koverstreet@google.com>
Cc: Zach Brown <zab@redhat.com>
Cc: Felipe Balbi <balbi@ti.com>
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: Mark Fasheh <mfasheh@suse.com>
Cc: Joel Becker <jlbec@evilplan.org>
Cc: Rusty Russell <rusty@rustcorp.com.au>
Cc: Jens Axboe <axboe@kernel.dk>
Cc: Asai Thambi S P <asamymuthupa@micron.com>
Cc: Selvan Mani <smani@micron.com>
Cc: Sam Bradshaw <sbradshaw@micron.com>
Cc: Jeff Moyer <jmoyer@redhat.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Benjamin LaHaise <bcrl@kvack.org>
Reviewed-by: "Theodore Ts'o" <tytso@mit.edu>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
11 years agoaio: change reqs_active to include unreaped completions
Kent Overstreet [Tue, 7 May 2013 23:18:51 +0000 (16:18 -0700)]
aio: change reqs_active to include unreaped completions

The aio code tries really hard to avoid having to deal with the
completion ringbuffer overflowing.  To do that, it has to keep track of
the number of outstanding kiocbs, and the number of completions
currently in the ringbuffer - and it's got to check that every time we
allocate a kiocb.  Ouch.

But - we can improve this quite a bit if we just change reqs_active to
mean "number of outstanding requests and unreaped completions" - that
means kiocb allocation doesn't have to look at the ringbuffer, which is
a fairly significant win.

Signed-off-by: Kent Overstreet <koverstreet@google.com>
Cc: Zach Brown <zab@redhat.com>
Cc: Felipe Balbi <balbi@ti.com>
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: Mark Fasheh <mfasheh@suse.com>
Cc: Joel Becker <jlbec@evilplan.org>
Cc: Rusty Russell <rusty@rustcorp.com.au>
Cc: Jens Axboe <axboe@kernel.dk>
Cc: Asai Thambi S P <asamymuthupa@micron.com>
Cc: Selvan Mani <smani@micron.com>
Cc: Sam Bradshaw <sbradshaw@micron.com>
Cc: Jeff Moyer <jmoyer@redhat.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Benjamin LaHaise <bcrl@kvack.org>
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
11 years agoaio: use cancellation list lazily
Kent Overstreet [Tue, 7 May 2013 23:18:49 +0000 (16:18 -0700)]
aio: use cancellation list lazily

Cancelling kiocbs requires adding them to a per kioctx linked list,
which is one of the few things we need to take the kioctx lock for in
the fast path.  But most kiocbs can't be cancelled - so if we just do
this lazily, we can avoid quite a bit of locking overhead.

While we're at it, instead of using a flag bit switch to using ki_cancel
itself to indicate that a kiocb has been cancelled/completed.  This lets
us get rid of ki_flags entirely.

[akpm@linux-foundation.org: remove buggy BUG()]
Signed-off-by: Kent Overstreet <koverstreet@google.com>
Cc: Zach Brown <zab@redhat.com>
Cc: Felipe Balbi <balbi@ti.com>
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: Mark Fasheh <mfasheh@suse.com>
Cc: Joel Becker <jlbec@evilplan.org>
Cc: Rusty Russell <rusty@rustcorp.com.au>
Cc: Jens Axboe <axboe@kernel.dk>
Cc: Asai Thambi S P <asamymuthupa@micron.com>
Cc: Selvan Mani <smani@micron.com>
Cc: Sam Bradshaw <sbradshaw@micron.com>
Cc: Jeff Moyer <jmoyer@redhat.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Benjamin LaHaise <bcrl@kvack.org>
Reviewed-by: "Theodore Ts'o" <tytso@mit.edu>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
11 years agoaio: use flush_dcache_page()
Kent Overstreet [Tue, 7 May 2013 23:18:47 +0000 (16:18 -0700)]
aio: use flush_dcache_page()

This wasn't causing problems before because it's not needed on x86, but
it is needed on other architectures.

Signed-off-by: Kent Overstreet <koverstreet@google.com>
Cc: Zach Brown <zab@redhat.com>
Cc: Felipe Balbi <balbi@ti.com>
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: Mark Fasheh <mfasheh@suse.com>
Cc: Joel Becker <jlbec@evilplan.org>
Cc: Rusty Russell <rusty@rustcorp.com.au>
Cc: Jens Axboe <axboe@kernel.dk>
Cc: Asai Thambi S P <asamymuthupa@micron.com>
Cc: Selvan Mani <smani@micron.com>
Cc: Sam Bradshaw <sbradshaw@micron.com>
Cc: Jeff Moyer <jmoyer@redhat.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Benjamin LaHaise <bcrl@kvack.org>
Cc: Theodore Ts'o <tytso@mit.edu>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
11 years agoaio: make aio_read_evt() more efficient, convert to hrtimers
Kent Overstreet [Tue, 7 May 2013 23:18:45 +0000 (16:18 -0700)]
aio: make aio_read_evt() more efficient, convert to hrtimers

Previously, aio_read_event() pulled a single completion off the
ringbuffer at a time, locking and unlocking each time.  Change it to
pull off as many events as it can at a time, and copy them directly to
userspace.

This also fixes a bug where if copying the event to userspace failed,
we'd lose the event.

Also convert it to wait_event_interruptible_hrtimeout(), which
simplifies it quite a bit.

[akpm@linux-foundation.org: coding-style fixes]
Signed-off-by: Kent Overstreet <koverstreet@google.com>
Cc: Zach Brown <zab@redhat.com>
Cc: Felipe Balbi <balbi@ti.com>
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: Mark Fasheh <mfasheh@suse.com>
Cc: Joel Becker <jlbec@evilplan.org>
Cc: Rusty Russell <rusty@rustcorp.com.au>
Cc: Jens Axboe <axboe@kernel.dk>
Cc: Asai Thambi S P <asamymuthupa@micron.com>
Cc: Selvan Mani <smani@micron.com>
Cc: Sam Bradshaw <sbradshaw@micron.com>
Cc: Jeff Moyer <jmoyer@redhat.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Benjamin LaHaise <bcrl@kvack.org>
Reviewed-by: "Theodore Ts'o" <tytso@mit.edu>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
11 years agowait: add wait_event_hrtimeout()
Kent Overstreet [Tue, 7 May 2013 23:18:43 +0000 (16:18 -0700)]
wait: add wait_event_hrtimeout()

Analagous to wait_event_timeout() and friends, this adds
wait_event_hrtimeout() and wait_event_interruptible_hrtimeout().

Note that unlike the versions that use regular timers, these don't
return the amount of time remaining when they return - instead, they
return 0 or -ETIME if they timed out.  because I was uncomfortable with
the semantics of doing it the other way (that I could get it right,
anyways).

If the timer expires, there's no real guarantee that expire_time -
current_time would be <= 0 - due to timer slack certainly, and I'm not
sure I want to know the implications of the different clock bases in
hrtimers.

If the timer does expire and the code calculates that the time remaining
is nonnegative, that could be even worse if the calling code then reuses
that timeout.  Probably safer to just return 0 then, but I could imagine
weird bugs or at least unintended behaviour arising from that too.

I came to the conclusion that if other users end up actually needing the
amount of time remaining, the sanest thing to do would be to create a
version that uses absolute timeouts instead of relative.

[akpm@linux-foundation.org: fix description of `timeout' arg]
Signed-off-by: Kent Overstreet <koverstreet@google.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Zach Brown <zab@redhat.com>
Cc: Felipe Balbi <balbi@ti.com>
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: Mark Fasheh <mfasheh@suse.com>
Cc: Joel Becker <jlbec@evilplan.org>
Cc: Rusty Russell <rusty@rustcorp.com.au>
Cc: Jens Axboe <axboe@kernel.dk>
Cc: Asai Thambi S P <asamymuthupa@micron.com>
Cc: Selvan Mani <smani@micron.com>
Cc: Sam Bradshaw <sbradshaw@micron.com>
Cc: Jeff Moyer <jmoyer@redhat.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Benjamin LaHaise <bcrl@kvack.org>
Reviewed-by: "Theodore Ts'o" <tytso@mit.edu>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
11 years agoaio: refcounting cleanup
Kent Overstreet [Tue, 7 May 2013 23:18:41 +0000 (16:18 -0700)]
aio: refcounting cleanup

The usage of ctx->dead was fubar - it makes no sense to explicitly check
it all over the place, especially when we're already using RCU.

Now, ctx->dead only indicates whether we've dropped the initial
refcount. The new teardown sequence is:

  set ctx->dead
  hlist_del_rcu();
  synchronize_rcu();

Now we know no system calls can take a new ref, and it's safe to drop
the initial ref:

  put_ioctx();

We also need to ensure there are no more outstanding kiocbs.  This was
done incorrectly - it was being done in kill_ctx(), and before dropping
the initial refcount.  At this point, other syscalls may still be
submitting kiocbs!

Now, we cancel and wait for outstanding kiocbs in free_ioctx(), after
kioctx->users has dropped to 0 and we know no more iocbs could be
submitted.

[akpm@linux-foundation.org: coding-style fixes]
Signed-off-by: Kent Overstreet <koverstreet@google.com>
Cc: Zach Brown <zab@redhat.com>
Cc: Felipe Balbi <balbi@ti.com>
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: Mark Fasheh <mfasheh@suse.com>
Cc: Joel Becker <jlbec@evilplan.org>
Cc: Rusty Russell <rusty@rustcorp.com.au>
Cc: Jens Axboe <axboe@kernel.dk>
Cc: Asai Thambi S P <asamymuthupa@micron.com>
Cc: Selvan Mani <smani@micron.com>
Cc: Sam Bradshaw <sbradshaw@micron.com>
Cc: Jeff Moyer <jmoyer@redhat.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Benjamin LaHaise <bcrl@kvack.org>
Reviewed-by: "Theodore Ts'o" <tytso@mit.edu>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
11 years agoaio: make aio_put_req() lockless
Kent Overstreet [Tue, 7 May 2013 23:18:39 +0000 (16:18 -0700)]
aio: make aio_put_req() lockless

Freeing a kiocb needed to touch the kioctx for three things:

 * Pull it off the reqs_active list
 * Decrementing reqs_active
 * Issuing a wakeup, if the kioctx was in the process of being freed.

This patch moves these to aio_complete(), for a couple reasons:

 * aio_complete() already has to issue the wakeup, so if we drop the
   kioctx refcount before aio_complete does its wakeup we don't have to
   do it twice.
 * aio_complete currently has to take the kioctx lock, so it makes sense
   for it to pull the kiocb off the reqs_active list too.
 * A later patch is going to change reqs_active to include unreaped
   completions - this will mean allocating a kiocb doesn't have to look
   at the ringbuffer. So taking the decrement of reqs_active out of
   kiocb_free() is useful prep work for that patch.

This doesn't really affect cancellation, since existing (usb) code that
implements a cancel function still calls aio_complete() - we just have
to make sure that aio_complete does the necessary teardown for cancelled
kiocbs.

It does affect code paths where we free kiocbs that were never
submitted; they need to decrement reqs_active and pull the kiocb off the
reqs_active list.  This occurs in two places: kiocb_batch_free(), which
is going away in a later patch, and the error path in io_submit_one.

[akpm@linux-foundation.org: coding-style fixes]
Signed-off-by: Kent Overstreet <koverstreet@google.com>
Cc: Zach Brown <zab@redhat.com>
Cc: Felipe Balbi <balbi@ti.com>
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: Mark Fasheh <mfasheh@suse.com>
Cc: Joel Becker <jlbec@evilplan.org>
Cc: Rusty Russell <rusty@rustcorp.com.au>
Cc: Jens Axboe <axboe@kernel.dk>
Cc: Asai Thambi S P <asamymuthupa@micron.com>
Cc: Selvan Mani <smani@micron.com>
Cc: Sam Bradshaw <sbradshaw@micron.com>
Acked-by: Jeff Moyer <jmoyer@redhat.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Benjamin LaHaise <bcrl@kvack.org>
Reviewed-by: "Theodore Ts'o" <tytso@mit.edu>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
11 years agoaio: do fget() after aio_get_req()
Kent Overstreet [Tue, 7 May 2013 23:18:37 +0000 (16:18 -0700)]
aio: do fget() after aio_get_req()

aio_get_req() will fail if we have the maximum number of requests
outstanding, which depending on the application may not be uncommon.  So
avoid doing an unnecessary fget().

Signed-off-by: Kent Overstreet <koverstreet@google.com>
Cc: Zach Brown <zab@redhat.com>
Cc: Felipe Balbi <balbi@ti.com>
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: Mark Fasheh <mfasheh@suse.com>
Cc: Joel Becker <jlbec@evilplan.org>
Cc: Rusty Russell <rusty@rustcorp.com.au>
Cc: Jens Axboe <axboe@kernel.dk>
Cc: Asai Thambi S P <asamymuthupa@micron.com>
Cc: Selvan Mani <smani@micron.com>
Cc: Sam Bradshaw <sbradshaw@micron.com>
Acked-by: Jeff Moyer <jmoyer@redhat.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Benjamin LaHaise <bcrl@kvack.org>
Reviewed-by: "Theodore Ts'o" <tytso@mit.edu>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
11 years agoaio: dprintk() -> pr_debug()
Kent Overstreet [Tue, 7 May 2013 23:18:35 +0000 (16:18 -0700)]
aio: dprintk() -> pr_debug()

Signed-off-by: Kent Overstreet <koverstreet@google.com>
Cc: Zach Brown <zab@redhat.com>
Cc: Felipe Balbi <balbi@ti.com>
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: Mark Fasheh <mfasheh@suse.com>
Cc: Joel Becker <jlbec@evilplan.org>
Cc: Rusty Russell <rusty@rustcorp.com.au>
Cc: Jens Axboe <axboe@kernel.dk>
Cc: Asai Thambi S P <asamymuthupa@micron.com>
Cc: Selvan Mani <smani@micron.com>
Cc: Sam Bradshaw <sbradshaw@micron.com>
Acked-by: Jeff Moyer <jmoyer@redhat.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Benjamin LaHaise <bcrl@kvack.org>
Cc: Theodore Ts'o <tytso@mit.edu>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
11 years agoaio: move private stuff out of aio.h
Kent Overstreet [Tue, 7 May 2013 23:18:33 +0000 (16:18 -0700)]
aio: move private stuff out of aio.h

Signed-off-by: Kent Overstreet <koverstreet@google.com>
Cc: Zach Brown <zab@redhat.com>
Cc: Felipe Balbi <balbi@ti.com>
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: Mark Fasheh <mfasheh@suse.com>
Cc: Joel Becker <jlbec@evilplan.org>
Cc: Rusty Russell <rusty@rustcorp.com.au>
Cc: Jens Axboe <axboe@kernel.dk>
Cc: Asai Thambi S P <asamymuthupa@micron.com>
Cc: Selvan Mani <smani@micron.com>
Cc: Sam Bradshaw <sbradshaw@micron.com>
Acked-by: Jeff Moyer <jmoyer@redhat.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Benjamin LaHaise <bcrl@kvack.org>
Cc: Theodore Ts'o <tytso@mit.edu>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
11 years agoaio: add kiocb_cancel()
Kent Overstreet [Tue, 7 May 2013 23:18:31 +0000 (16:18 -0700)]
aio: add kiocb_cancel()

Minor refactoring, to get rid of some duplicated code

[akpm@linux-foundation.org: fix warning]
Signed-off-by: Kent Overstreet <koverstreet@google.com>
Cc: Zach Brown <zab@redhat.com>
Cc: Felipe Balbi <balbi@ti.com>
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: Mark Fasheh <mfasheh@suse.com>
Cc: Joel Becker <jlbec@evilplan.org>
Cc: Rusty Russell <rusty@rustcorp.com.au>
Cc: Jens Axboe <axboe@kernel.dk>
Cc: Asai Thambi S P <asamymuthupa@micron.com>
Cc: Selvan Mani <smani@micron.com>
Cc: Sam Bradshaw <sbradshaw@micron.com>
Acked-by: Jeff Moyer <jmoyer@redhat.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Benjamin LaHaise <bcrl@kvack.org>
Reviewed-by: "Theodore Ts'o" <tytso@mit.edu>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
11 years agoaio: kill return value of aio_complete()
Kent Overstreet [Tue, 7 May 2013 23:18:29 +0000 (16:18 -0700)]
aio: kill return value of aio_complete()

Nothing used the return value, and it probably wasn't possible to use it
safely for the locked versions (aio_complete(), aio_put_req()).  Just
kill it.

Signed-off-by: Kent Overstreet <koverstreet@google.com>
Acked-by: Zach Brown <zab@redhat.com>
Cc: Felipe Balbi <balbi@ti.com>
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: Mark Fasheh <mfasheh@suse.com>
Cc: Joel Becker <jlbec@evilplan.org>
Cc: Rusty Russell <rusty@rustcorp.com.au>
Cc: Jens Axboe <axboe@kernel.dk>
Cc: Asai Thambi S P <asamymuthupa@micron.com>
Cc: Selvan Mani <smani@micron.com>
Cc: Sam Bradshaw <sbradshaw@micron.com>
Acked-by: Jeff Moyer <jmoyer@redhat.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Benjamin LaHaise <bcrl@kvack.org>
Reviewed-by: "Theodore Ts'o" <tytso@mit.edu>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
11 years agochar: add aio_{read,write} to /dev/{null,zero}
Zach Brown [Tue, 7 May 2013 23:18:27 +0000 (16:18 -0700)]
char: add aio_{read,write} to /dev/{null,zero}

These are handy for measuring the cost of the aio infrastructure with
operations that do very little and complete immediately.

Signed-off-by: Zach Brown <zab@redhat.com>
Signed-off-by: Kent Overstreet <koverstreet@google.com>
Cc: Felipe Balbi <balbi@ti.com>
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: Mark Fasheh <mfasheh@suse.com>
Cc: Joel Becker <jlbec@evilplan.org>
Cc: Rusty Russell <rusty@rustcorp.com.au>
Cc: Jens Axboe <axboe@kernel.dk>
Cc: Asai Thambi S P <asamymuthupa@micron.com>
Cc: Selvan Mani <smani@micron.com>
Cc: Sam Bradshaw <sbradshaw@micron.com>
Acked-by: Jeff Moyer <jmoyer@redhat.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Benjamin LaHaise <bcrl@kvack.org>
Reviewed-by: "Theodore Ts'o" <tytso@mit.edu>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
11 years agoaio: remove retry-based AIO
Zach Brown [Tue, 7 May 2013 23:18:25 +0000 (16:18 -0700)]
aio: remove retry-based AIO

This removes the retry-based AIO infrastructure now that nothing in tree
is using it.

We want to remove retry-based AIO because it is fundemantally unsafe.
It retries IO submission from a kernel thread that has only assumed the
mm of the submitting task.  All other task_struct references in the IO
submission path will see the kernel thread, not the submitting task.
This design flaw means that nothing of any meaningful complexity can use
retry-based AIO.

This removes all the code and data associated with the retry machinery.
The most significant benefit of this is the removal of the locking
around the unused run list in the submission path.

[akpm@linux-foundation.org: coding-style fixes]
Signed-off-by: Kent Overstreet <koverstreet@google.com>
Signed-off-by: Zach Brown <zab@redhat.com>
Cc: Zach Brown <zab@redhat.com>
Cc: Felipe Balbi <balbi@ti.com>
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: Mark Fasheh <mfasheh@suse.com>
Cc: Joel Becker <jlbec@evilplan.org>
Cc: Rusty Russell <rusty@rustcorp.com.au>
Cc: Jens Axboe <axboe@kernel.dk>
Cc: Asai Thambi S P <asamymuthupa@micron.com>
Cc: Selvan Mani <smani@micron.com>
Cc: Sam Bradshaw <sbradshaw@micron.com>
Acked-by: Jeff Moyer <jmoyer@redhat.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Benjamin LaHaise <bcrl@kvack.org>
Reviewed-by: "Theodore Ts'o" <tytso@mit.edu>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
11 years agogadget: remove only user of aio retry
Zach Brown [Tue, 7 May 2013 23:18:23 +0000 (16:18 -0700)]
gadget: remove only user of aio retry

This removes the only in-tree user of aio retry.  This will let us
remove the retry code from the aio core.

Removing retry is relatively easy as the USB gadget wasn't using it to
retry IOs at all.  It always fully submitted the IO in the context of
the initial io_submit() call.  It only used the AIO retry facility to
get the submitter's mm context for copying the result of a read back to
user space.  This is easy to implement with use_mm() and a work struct,
much like kvm does with async_pf_execute() for get_user_pages().

[akpm@linux-foundation.org: coding-style fixes]
Signed-off-by: Zach Brown <zab@redhat.com>
Signed-off-by: Kent Overstreet <koverstreet@google.com>
Cc: Felipe Balbi <balbi@ti.com>
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: Mark Fasheh <mfasheh@suse.com>
Cc: Joel Becker <jlbec@evilplan.org>
Cc: Rusty Russell <rusty@rustcorp.com.au>
Cc: Jens Axboe <axboe@kernel.dk>
Cc: Asai Thambi S P <asamymuthupa@micron.com>
Cc: Selvan Mani <smani@micron.com>
Cc: Sam Bradshaw <sbradshaw@micron.com>
Acked-by: Jeff Moyer <jmoyer@redhat.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Benjamin LaHaise <bcrl@kvack.org>
Cc: Theodore Ts'o <tytso@mit.edu>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
11 years agoaio: remove dead code from aio.h
Zach Brown [Tue, 7 May 2013 23:18:21 +0000 (16:18 -0700)]
aio: remove dead code from aio.h

Signed-off-by: Zach Brown <zab@redhat.com>
Signed-off-by: Kent Overstreet <koverstreet@google.com>
Cc: Felipe Balbi <balbi@ti.com>
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: Mark Fasheh <mfasheh@suse.com>
Cc: Joel Becker <jlbec@evilplan.org>
Cc: Rusty Russell <rusty@rustcorp.com.au>
Cc: Jens Axboe <axboe@kernel.dk>
Cc: Asai Thambi S P <asamymuthupa@micron.com>
Cc: Selvan Mani <smani@micron.com>
Cc: Sam Bradshaw <sbradshaw@micron.com>
Acked-by: Jeff Moyer <jmoyer@redhat.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Benjamin LaHaise <bcrl@kvack.org>
Reviewed-by: "Theodore Ts'o" <tytso@mit.edu>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
11 years agomm: remove old aio use_mm() comment
Zach Brown [Tue, 7 May 2013 23:18:19 +0000 (16:18 -0700)]
mm: remove old aio use_mm() comment

Bunch of performance improvements and cleanups Zach Brown and I have
been working on.  The code should be pretty solid at this point, though
it could of course use more review and testing.

The results in my testing are pretty impressive, particularly when an
ioctx is being shared between multiple threads.  In my crappy synthetic
benchmark, with 4 threads submitting and one thread reaping completions,
I saw overhead in the aio code go from ~50% (mostly ioctx lock
contention) to low single digits.  Performance with ioctx per thread
improved too, but I'd have to rerun those benchmarks.

The reason I've been focused on performance when the ioctx is shared is
that for a fair number of real world completions, userspace needs the
completions aggregated somehow - in practice people just end up
implementing this aggregation in userspace today, but if it's done right
we can do it much more efficiently in the kernel.

Performance wise, the end result of this patch series is that submitting
a kiocb writes to _no_ shared cachelines - the penalty for sharing an
ioctx is gone there.  There's still going to be some cacheline
contention when we deliver the completions to the aio ringbuffer (at
least if you have interrupts being delivered on multiple cores, which
for high end stuff you do) but I have a couple more patches not in this
series that implement coalescing for that (by taking advantage of
interrupt coalescing).  With that, there's basically no bottlenecks or
performance issues to speak of in the aio code.

This patch:

use_mm() is used in more places than just aio.  There's no need to mention
callers when describing the function.

Signed-off-by: Zach Brown <zab@redhat.com>
Signed-off-by: Kent Overstreet <koverstreet@google.com>
Cc: Felipe Balbi <balbi@ti.com>
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: Mark Fasheh <mfasheh@suse.com>
Cc: Joel Becker <jlbec@evilplan.org>
Cc: Rusty Russell <rusty@rustcorp.com.au>
Cc: Jens Axboe <axboe@kernel.dk>
Cc: Asai Thambi S P <asamymuthupa@micron.com>
Cc: Selvan Mani <smani@micron.com>
Cc: Sam Bradshaw <sbradshaw@micron.com>
Acked-by: Jeff Moyer <jmoyer@redhat.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Benjamin LaHaise <bcrl@kvack.org>
Reviewed-by: "Theodore Ts'o" <tytso@mit.edu>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
11 years agomm/vmalloc.c: add vfree comment
Andrew Morton [Tue, 7 May 2013 23:18:18 +0000 (16:18 -0700)]
mm/vmalloc.c: add vfree comment

Cc: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
11 years agoremove unused random32() and srandom32()
Akinobu Mita [Tue, 7 May 2013 23:18:17 +0000 (16:18 -0700)]
remove unused random32() and srandom32()

After finishing a naming transition, remove unused backward
compatibility wrapper macros

Signed-off-by: Akinobu Mita <akinobu.mita@gmail.com>
Cc: "Theodore Ts'o" <tytso@mit.edu>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
11 years agodrivers/infiniband/hw: rename random32() to prandom_u32()
Andrew Morton [Tue, 7 May 2013 23:18:16 +0000 (16:18 -0700)]
drivers/infiniband/hw: rename random32() to prandom_u32()

Use preferable function name which implies using a pseudo-random number
generator.

Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
11 years agodrivers/net: rename random32() to prandom_u32()
Akinobu Mita [Tue, 7 May 2013 23:18:15 +0000 (16:18 -0700)]
drivers/net: rename random32() to prandom_u32()

Use preferable function name which implies using a pseudo-random number
generator.

[akpm@linux-foundation.org: convert team_mode_random.c]
Signed-off-by: Akinobu Mita <akinobu.mita@gmail.com>
Acked-by: Thomas Sailer <t.sailer@alumni.ethz.ch>
Acked-by: Bing Zhao <bzhao@marvell.com> [mwifiex]
Cc: "David S. Miller" <davem@davemloft.net>
Cc: Michael Chan <mchan@broadcom.com>
Cc: Thomas Sailer <t.sailer@alumni.ethz.ch>
Cc: Jean-Paul Roubelat <jpr@f6fbb.org>
Cc: Bing Zhao <bzhao@marvell.com>
Cc: Brett Rudley <brudley@broadcom.com>
Cc: Arend van Spriel <arend@broadcom.com>
Cc: "Franky (Zhenhui) Lin" <frankyl@broadcom.com>
Cc: Hante Meuleman <meuleman@broadcom.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
11 years agohugetlbfs: fix mmap failure in unaligned size request
Naoya Horiguchi [Tue, 7 May 2013 23:18:13 +0000 (16:18 -0700)]
hugetlbfs: fix mmap failure in unaligned size request

The current kernel returns -EINVAL unless a given mmap length is
"almost" hugepage aligned.  This is because in sys_mmap_pgoff() the
given length is passed to vm_mmap_pgoff() as it is without being aligned
with hugepage boundary.

This is a regression introduced in commit 40716e29243d ("hugetlbfs: fix
alignment of huge page requests"), where alignment code is pushed into
hugetlb_file_setup() and the variable len in caller side is not changed.

To fix this, this patch partially reverts that commit, and adds
alignment code in caller side.  And it also introduces hstate_sizelog()
in order to get proper hstate to specified hugepage size.

Addresses https://bugzilla.kernel.org/show_bug.cgi?id=56881

[akpm@linux-foundation.org: fix warning when CONFIG_HUGETLB_PAGE=n]
Signed-off-by: Naoya Horiguchi <n-horiguchi@ah.jp.nec.com>
Signed-off-by: Johannes Weiner <hannes@cmpxchg.org>
Reported-by: <iceman_dvd@yahoo.com>
Cc: Steven Truelove <steven.truelove@utoronto.ca>
Cc: Jianguo Wu <wujianguo@huawei.com>
Cc: Hugh Dickins <hughd@google.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
11 years agoparisc: remove the second argument of kmap_atomic()
Zhao Hongjiang [Tue, 7 May 2013 23:18:12 +0000 (16:18 -0700)]
parisc: remove the second argument of kmap_atomic()

kmap_atomic() requires only one argument now.

Signed-off-by: Zhao Hongjiang <zhaohongjiang@huawei.com>
Cc: "James E.J. Bottomley" <jejb@parisc-linux.org>
Cc: Helge Deller <deller@gmx.de>
Cc: Rolf Eike Beer <eike-kernel@sf-tec.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
11 years agodrivers/rtc/rtc-rs5c372.c: add R2221T/L variant to the driver
Lucas Stach [Tue, 7 May 2013 23:18:11 +0000 (16:18 -0700)]
drivers/rtc/rtc-rs5c372.c: add R2221T/L variant to the driver

Register layout is the same, so just add the variant to the appropriate
places.

Signed-off-by: Lucas Stach <l.stach@pengutronix.de>
Signed-off-by: Jan Luebbe <jlu@pengutronix.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
11 years agoinclude/linux/mm.h: complete the mm_walk definition
Andrew Morton [Tue, 7 May 2013 23:18:10 +0000 (16:18 -0700)]
include/linux/mm.h: complete the mm_walk definition

That nameless-function-arguments thing drives me batty.  Fix.

Cc: Dave Hansen <dave.hansen@intel.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
11 years agomm, memcg: add rss_huge stat to memory.stat
David Rientjes [Tue, 7 May 2013 23:18:09 +0000 (16:18 -0700)]
mm, memcg: add rss_huge stat to memory.stat

This exports the amount of anonymous transparent hugepages for each
memcg via the new "rss_huge" stat in memory.stat.  The units are in
bytes.

This is helpful to determine the hugepage utilization for individual
jobs on the system in comparison to rss and opportunities where
MADV_HUGEPAGE may be helpful.

The amount of anonymous transparent hugepages is also included in "rss"
for backwards compatibility.

Signed-off-by: David Rientjes <rientjes@google.com>
Acked-by: Michal Hocko <mhocko@suse.cz>
Acked-by: Johannes Weiner <hannes@cmpxchg.org>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
11 years agomm/SPARC: use common help functions to free reserved pages
Jiang Liu [Tue, 7 May 2013 23:18:08 +0000 (16:18 -0700)]
mm/SPARC: use common help functions to free reserved pages

Use common help functions to free reserved pages.

Signed-off-by: Jiang Liu <jiang.liu@huawei.com>
Acked-by: David S. Miller <davem@davemloft.net>
Acked-by: Sam Ravnborg <sam@ravnborg.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
11 years agoarm: fix mismerge of arch/arm/mach-omap2/timer.c
Linus Torvalds [Wed, 8 May 2013 00:59:53 +0000 (17:59 -0700)]
arm: fix mismerge of arch/arm/mach-omap2/timer.c

I badly screwed up the merge in commit 6fa52ed33bea ("Merge tag
'drivers-for-linus' of git://git.kernel.org/pub/.../arm-soc") by
incorrectly taking the arch/arm/mach-omap2/* data fully from the merge
target because the 'drivers-for-linus' branch seemed to be a proper
superset of the duplicate ARM commits.

That was bogus: commit ff931c821bab ("ARM: OMAP: clocks: Delay clk inits
atleast until slab is initialized") only existed in head, and the
changes to arch/arm/mach-omap2/timer.c from that commit got list.

Re-doing the merge more carefully, I do think this part was the only
thing I screwed up.  Knock wood.

Reported-by: Tony Lindgren <tony@atomide.com>
Cc: Arnd Bergmann <arnd@arndb.de>
Cc: Olof Johansson <olof@lixom.net>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
11 years agoxfs: fallback to vmalloc for large buffers in xfs_compat_attrlist_by_handle
Eric Sandeen [Thu, 25 Apr 2013 16:13:06 +0000 (11:13 -0500)]
xfs: fallback to vmalloc for large buffers in xfs_compat_attrlist_by_handle

Shamelessly copied from dchinner's:
ad650f5b xfs: fallback to vmalloc for large buffers in xfs_attrmulti_attr_get

xfsdump uses a large buffer for extended attributes, which has a
kmalloc'd shadow buffer in the kernel. This can fail after the
system has been running for some time as it is a high order
allocation. Add a fallback to vmalloc so that it doesn't require
contiguous memory and so won't randomly fail while xfsdump is
running.

This was done for xfs_attrlist_by_handle but
xfs_compat_attrlist_by_handle (the 32-bit version) needs the same
attention.

Signed-off-by: Eric Sandeen <sandeen@redhat.com>
Reviewed-by: Mark Tinguely <tinguely@sgi.com>
Signed-off-by: Ben Myers <bpm@sgi.com>
11 years agoxfs: fallback to vmalloc for large buffers in xfs_attrlist_by_handle
Eric Sandeen [Thu, 4 Apr 2013 05:05:13 +0000 (00:05 -0500)]
xfs: fallback to vmalloc for large buffers in xfs_attrlist_by_handle

Shamelessly copied from dchinner's:
ad650f5b xfs: fallback to vmalloc for large buffers in xfs_attrmulti_attr_get

xfsdump uses for a large buffer for extended attributes, which has a
kmalloc'd shadow buffer in the kernel. This can fail after the
system has been running for some time as it is a high order
allocation. Add a fallback to vmalloc so that it doesn't require
contiguous memory and so won't randomly fail while xfsdump is
running.

Signed-off-by: Eric Sandeen <sandeen@redhat.com>
Reviewed-by: Mark Tinguely <tinguely@sgi.com>
Signed-off-by: Ben Myers <bpm@sgi.com>
11 years agoxfs: introduce CONFIG_XFS_WARN
Dave Chinner [Tue, 30 Apr 2013 11:39:34 +0000 (21:39 +1000)]
xfs: introduce CONFIG_XFS_WARN

Running a CONFIG_XFS_DEBUG kernel in production environments is not
the best idea as it introduces significant overhead, can change
the behaviour of algorithms (such as allocation) to improve test
coverage, and (most importantly) panic the machine on non-fatal
errors.

There are many cases where all we want to do is run a
kernel with more bounds checking enabled, such as is provided by the
ASSERT() statements throughout the code, but without all the
potential overhead and drawbacks.

This patch converts all the ASSERT statements to evaluate as
WARN_ON(1) statements and hence if they fail dump a warning and a
stack trace to the log. This has minimal overhead and does not
change any algorithms, and will allow us to find strange "out of
bounds" problems more easily on production machines.

There are a few places where assert statements contain debug only
code. These are converted to be debug-or-warn only code so that we
still get all the assert checks in the code.

Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Brian Foster <bfoster@redhat.com>
Signed-off-by: Ben Myers <bpm@sgi.com>
11 years agorwsem: check counter to avoid cmpxchg calls
Davidlohr Bueso [Tue, 7 May 2013 22:39:03 +0000 (15:39 -0700)]
rwsem: check counter to avoid cmpxchg calls

This patch tries to reduce the amount of cmpxchg calls in the writer
failed path by checking the counter value first before issuing the
instruction.  If ->count is not set to RWSEM_WAITING_BIAS then there is
no point wasting a cmpxchg call.

Furthermore, Michel states "I suppose it helps due to the case where
someone else steals the lock while we're trying to acquire
sem->wait_lock."

Two very different workloads and machines were used to see how this
patch improves throughput: pgbench on a quad-core laptop and aim7 on a
large 8 socket box with 80 cores.

Some results comparing Michel's fast-path write lock stealing
(tps-rwsem) on a quad-core laptop running pgbench:

  | db_size | clients  |  tps-rwsem     |   tps-patch  |
  +---------+----------+----------------+--------------+
  | 160 MB   |       1 |           6906 |         9153 | + 32.5
  | 160 MB   |       2 |          15931 |        22487 | + 41.1%
  | 160 MB   |       4 |          33021 |        32503 |
  | 160 MB   |       8 |          34626 |        34695 |
  | 160 MB   |      16 |          33098 |        34003 |
  | 160 MB   |      20 |          31343 |        31440 |
  | 160 MB   |      30 |          28961 |        28987 |
  | 160 MB   |      40 |          26902 |        26970 |
  | 160 MB   |      50 |          25760 |        25810 |
  ------------------------------------------------------
  | 1.6 GB   |       1 |           7729 |         7537 |
  | 1.6 GB   |       2 |          19009 |        23508 | + 23.7%
  | 1.6 GB   |       4 |          33185 |        32666 |
  | 1.6 GB   |       8 |          34550 |        34318 |
  | 1.6 GB   |      16 |          33079 |        32689 |
  | 1.6 GB   |      20 |          31494 |        31702 |
  | 1.6 GB   |      30 |          28535 |        28755 |
  | 1.6 GB   |      40 |          27054 |        27017 |
  | 1.6 GB   |      50 |          25591 |        25560 |
  ------------------------------------------------------
  | 7.6 GB   |       1 |           6224 |         7469 | + 20.0%
  | 7.6 GB   |       2 |          13611 |        12778 |
  | 7.6 GB   |       4 |          33108 |        32927 |
  | 7.6 GB   |       8 |          34712 |        34878 |
  | 7.6 GB   |      16 |          32895 |        33003 |
  | 7.6 GB   |      20 |          31689 |        31974 |
  | 7.6 GB   |      30 |          29003 |        28806 |
  | 7.6 GB   |      40 |          26683 |        26976 |
  | 7.6 GB   |      50 |          25925 |        25652 |
  ------------------------------------------------------

For the aim7 worloads, they overall improved on top of Michel's
patchset.  For full graphs on how the rwsem series plus this patch
behaves on a large 8 socket machine against a vanilla kernel:

  http://stgolabs.net/rwsem-aim7-results.tar.gz

Signed-off-by: Davidlohr Bueso <davidlohr.bueso@hp.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
11 years agokref: minor cleanup
Anatol Pomozov [Tue, 7 May 2013 22:37:48 +0000 (15:37 -0700)]
kref: minor cleanup

 - make warning smp-safe
 - result of atomic _unless_zero functions should be checked by caller
   to avoid use-after-free error
 - trivial whitespace fix.

Link: https://lkml.org/lkml/2013/4/12/391
Tested: compile x86, boot machine and run xfstests
Signed-off-by: Anatol Pomozov <anatol.pomozov@gmail.com>
[ Removed line-break, changed to use WARN_ON_ONCE()  - Linus ]
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
11 years agoMerge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs
Linus Torvalds [Tue, 7 May 2013 22:14:53 +0000 (15:14 -0700)]
Merge branch 'for-linus' of git://git./linux/kernel/git/viro/vfs

Pull more vfs updates from Al Viro:
 "A couple of fixes + getting rid of __blkdev_put() return value"

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs:
  proc: Use PDE attribute setting accessor functions
  make blkdev_put() return void
  block_device_operations->release() should return void
  mtd_blktrans_ops->release() should return void
  hfs: SMP race on directory close()

11 years agoMerge branch 'parisc-for-3.10' of git://git.kernel.org/pub/scm/linux/kernel/git/delle...
Linus Torvalds [Tue, 7 May 2013 22:13:48 +0000 (15:13 -0700)]
Merge branch 'parisc-for-3.10' of git://git./linux/kernel/git/deller/parisc-linux

Pull parisc updates from Helge Deller:
 "Main fixes and updates in this patch series are:
   - we faced kernel stack corruptions because of multiple delivery of
     interrupts
   - added kernel stack overflow checks
   - added possibility to use dedicated stacks for irq processing
   - initial support for page sizes > 4k
   - more information in /proc/interrupts (e.g.  TLB flushes and number
     of IPI calls)
   - documented how the parisc gateway page works
   - and of course quite some other smaller cleanups and fixes."

* 'parisc-for-3.10' of git://git.kernel.org/pub/scm/linux/kernel/git/deller/parisc-linux:
  parisc: tlb flush counting fix for SMP and UP
  parisc: more irq statistics in /proc/interrupts
  parisc: implement irq stacks
  parisc: add kernel stack overflow check
  parisc: only re-enable interrupts if we need to schedule or deliver signals when returning to userspace
  parisc: implement atomic64_dec_if_positive()
  parisc: use long branch in fork_like macro
  parisc: fix NATIVE set up in build
  parisc: document the parisc gateway page
  parisc: fix partly 16/64k PAGE_SIZE boot
  parisc: Provide default implementation for dma_{alloc, free}_attrs
  parisc: fix whitespace errors in arch/parisc/kernel/traps.c
  parisc: remove the second argument of kmap_atomic

11 years agoMerge tag '3.9-rc3-smp-6-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/sstabe...
Linus Torvalds [Tue, 7 May 2013 22:11:43 +0000 (15:11 -0700)]
Merge tag '3.9-rc3-smp-6-tag' of git://git./linux/kernel/git/sstabellini/xen

Pull ARM Xen SMP updates from Stefano Stabellini:
 "This contains a bunch of Xen/ARM specific changes, including some
  fixes, SMP support for Xen on ARM, and moving the xenvm machine from
  mach-vexpress to mach-virt.

  The non-Xen files that are touched are arch/arm/Kconfig, to select
  ARM_PSCI on XEN, and arch/arm/boot/dts/Makefile, to build the xenvm
  DTB if CONFIG_ARCH_VIRT.

  Highlights:

   - Move xenvm to mach-virt.

   - Implement SMP support in Xen on ARM.

   - Add support for machine reboot and power off via Xen hypercalls"

* tag '3.9-rc3-smp-6-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/sstabellini/xen:
  xen/arm: remove duplicated include from enlighten.c
  xen/arm: use sched_op hypercalls for machine reboot and power off
  xenvm: add a simple PSCI node and a second cpu
  xen/arm: XEN selects ARM_PSCI
  xen: move the xenvm machine to mach-virt
  xen/arm: SMP support
  xen/arm: implement HYPERVISOR_vcpu_op
  xen/arm: actually pass a non-NULL percpu pointer to request_percpu_irq