Commits on Sep 9, 2009
  1. Linux 2.6.31

    committed Sep 9, 2009
  2. aoe: allocate unused request_queue for sysfs

    Andy Whitcroft reported an oops in aoe triggered by use of an
    incorrectly initialised request_queue object:
      [ 2645.959090] kobject '<NULL>' (ffff880059ca22c0): tried to add
    		an uninitialized object, something is seriously wrong.
      [ 2645.959104] Pid: 6, comm: events/0 Not tainted 2.6.31-5-generic #24-Ubuntu
      [ 2645.959107] Call Trace:
      [ 2645.959139] [<ffffffff8126ca2f>] kobject_add+0x5f/0x70
      [ 2645.959151] [<ffffffff8125b4ab>] blk_register_queue+0x8b/0xf0
      [ 2645.959155] [<ffffffff8126043f>] add_disk+0x8f/0x160
      [ 2645.959161] [<ffffffffa01673c4>] aoeblk_gdalloc+0x164/0x1c0 [aoe]
    The request queue of an aoe device is not used but can be allocated in
    code that does not sleep.
    Bruno bisected this regression down to
      block: Expose stacked device queues in sysfs
    "This seems to generate /sys/block/$device/queue and its contents for
     everyone who is using queues, not just for those queues that have a
     non-NULL queue->request_fn."
    Note that embedding a queue inside another object has always been
    an illegal construct, since the queues are reference counted and
    must persist until the last reference is dropped. So aoe was
    always buggy in this respect (Jens).
    Signed-off-by: Ed Cashin <>
    Cc: Andy Whitcroft <>
    Cc: "Rafael J. Wysocki" <>
    Cc: Bruno Premont <>
    Cc: Martin K. Petersen <>
    Cc: Andrew Morton <>
    Signed-off-by: Jens Axboe <>
    ecashin committed with Jens Axboe Sep 9, 2009
  3. i915: disable interrupts before tearing down GEM state

    Reinette Chatre reports a frozen system (with blinking keyboard LEDs)
    when switching from graphics mode to the text console, or when
    suspending (which does the same thing). With netconsole, the oops
    turned out to be
    	BUG: unable to handle kernel NULL pointer dereference at 0000000000000084
    	IP: [<ffffffffa03ecaab>] i915_driver_irq_handler+0x26b/0xd20 [i915]
    and it's due to the i915_gem.c code doing drm_irq_uninstall() after
    having done i915_gem_idle(). And the i915_gem_idle() path will do
      i915_gem_idle() ->
        i915_gem_cleanup_ringbuffer() ->
          i915_gem_cleanup_hws() ->
            dev_priv->hw_status_page = NULL;
    but if an i915 interrupt comes in after this stage, it may want to
    access that hw_status_page, and gets the above NULL pointer dereference.
    And since the NULL pointer dereference happens from within an interrupt,
    and with the screen still in graphics mode, the common end result is
    simply a silently hung machine.
    Fix it by simply uninstalling the irq handler before idling rather than
    after. Fixes
    Reported-and-tested-by: Reinette Chatre <>
    Acked-by: Jesse Barnes <>
    Signed-off-by: Linus Torvalds <>
    committed Sep 9, 2009
Commits on Sep 8, 2009
  1. drm/i915: fix mask bits setting

    eDP is exclusive connector too, and add missing crtc_mask
    setting for TV.
    This fixes

    Signed-off-by: Zhenyu Wang <>
    Reported-and-tested-by: Carlos R. Mafra <>
    Signed-off-by: Linus Torvalds <>
    Zhenyu Wang committed with Sep 8, 2009
Commits on Sep 7, 2009
  1. Merge branch 'drm-fixes' of git://…

    * 'drm-fixes' of git://
      drm/radeon/kms: add LTE/GTE discard + rv515 two sided stencil register.
    committed Sep 7, 2009
  2. Merge branch 'for-linus' of git://…

    * 'for-linus' of git://
      IMA: update ima_counts_put
    committed Sep 7, 2009
  3. Merge git://

    * git://
      gianfar: Fix build.
    committed Sep 7, 2009
  4. Merge git://

    * git://
      pcmcia: add CNF-CDROM-ID for ide
    committed Sep 7, 2009
  5. Merge branch 'for-linus' of git://…

    * 'for-linus' of git://
      agp/intel: support for new chip variant of IGDNG mobile
      drm/i915: Unref old_obj on get_fence_reg() error path
      drm/i915: increase default latency constant (v2 w/comment)
    committed Sep 7, 2009
  6. drm/radeon/kms: add LTE/GTE discard + rv515 two sided stencil register.

    This adds some rv350+ register for LTE/GTE discard,
    and enables the rv515 two sided stencil register.
    It also disables the DEPTHXY_OFFSET register which
    can be used to workaround the CS checker.
    Moves rs690 to proper place in rs600 and uses correct
    table on rs600.
    Signed-off-by: Dave Airlie <>
    Dave Airlie committed Sep 7, 2009
  7. IMA: update ima_counts_put

    - As ima_counts_put() may be called after the inode has been freed,
    verify that the inode is not NULL, before dereferencing it.
    - Maintain the IMA file counters in may_open() properly, decrementing
    any counter increments on subsequent errors.
    Reported-by: Ciprian Docan <>
    Reported-by: J.R. Okajima <>
    Signed-off-by: Mimi Zohar <>
    Acked-by: Eric Paris <
    Signed-off-by: James Morris <>
    Mimi Zohar committed with James Morris Sep 4, 2009
Commits on Sep 6, 2009
  1. gianfar: Fix build.

    Reported by Michael Guntsche <>
    38bddf0 gianfar: gfar_remove needs to call unregister_netdev()
    breaks the build of the gianfar driver because "dev" is undefined in
    this function. To quickly test rc9 I changed this to priv->ndev but I do
    not know if this is the correct one.
    Signed-off-by: David S. Miller <>
    davem330 committed Sep 6, 2009
Commits on Sep 5, 2009
  1. Linux 2.6.31-rc9

    committed Sep 5, 2009
  2. Merge branch 'for-linus' of git://…

    * 'for-linus' of git://
      firewire: sbp2: fix freeing of unallocated memory
      firewire: ohci: fix Ricoh R5C832, video reception
      firewire: ohci: fix Agere FW643 and multiple cameras
      firewire: core: fix crash in iso resource management
    committed Sep 5, 2009
  3. powerpc: Fix i8259 interrupt driver kernel crash on ML510

    This patch fixes a null pointer exception caused by removal of
    'ack()' for level interrupts in the Xilinx interrupt driver.  A recent
    change to the xilinx interrupt controller removed the ack hook for
    level irqs.
    Signed-off-by: Roderick Colenbrander <>
    Signed-off-by: Grant Likely <>
    Acked-by: Benjamin Herrenschmidt <>
    Signed-off-by: Linus Torvalds <>
    Roderick Colenbrander committed with Sep 3, 2009
  4. Merge git://

    * git://
      JFFS2: add missing verify buffer allocation/deallocation
      mtd: nftl: fix offset alignments
      mtd: nftl: write support is broken
      mtd: m25p80: fix null pointer dereference bug
    committed Sep 5, 2009
  5. Merge branch 'for-linus' of git://

    * 'for-linus' of git://
      block: Allow changing max_sectors_kb above the default 512
    committed Sep 5, 2009
  6. Merge branch 'fix/oxygen' of git://…

    * 'fix/oxygen' of git://
      sound: oxygen: handle cards with missing EEPROM
      sound: oxygen: fix MCLK rate for 192 kHz playback
    committed Sep 5, 2009
  7. Merge git://

    * git://
      tc: Fix unitialized kernel memory leak
      pkt_sched: Revert tasklet_hrtimer changes.
      net: sk_free() should be allowed right after sk_alloc()
      gianfar: gfar_remove needs to call unregister_netdev()
      ipw2200: firmware DMA loading rework
    committed Sep 5, 2009
  8. Merge git://

    * git://
      crypto: skcipher - Fix skcipher_dequeue_givcrypt NULL test
    committed Sep 5, 2009
  9. Merge branch 'fixes' of git://…

    * 'fixes' of git://
      [CPUFREQ] Re-enable cpufreq suspend and resume code
    committed Sep 5, 2009
  10. Merge branch 'release' of git://…

    * 'release' of git://
      [IA64] fix csum_ipv6_magic()
      [IA64] Fix warning in dma-mapping.c
    committed Sep 5, 2009
  11. Merge branch 'for-linus' of git://

    * 'for-linus' of git://
      xfs: actually enable the swapext compat handler
    committed Sep 5, 2009
  12. Merge branch 'for-linus' of git://…

    * 'for-linus' of git://
      nilfs2: fix preempt count underflow in nilfs_btnode_prepare_change_key
    committed Sep 5, 2009
  13. Merge branch 'for-linus' of git://…

    * 'for-linus' of git://
      percpu: don't assume existence of cpu0
    committed Sep 5, 2009
  14. Merge branch 'slab/urgent' of git://…

    * 'slab/urgent' of git://
      slub: Fix kmem_cache_destroy() with SLAB_DESTROY_BY_RCU
    committed Sep 5, 2009
  15. Merge git://

    * git://
      dm snapshot: fix on disk chunk size validation
      dm exception store: split set_chunk_size
      dm snapshot: fix header corruption race on invalidation
      dm snapshot: refactor zero_disk_area to use chunk_io
      dm log: userspace add luid to distinguish between concurrent log instances
      dm raid1: do not allow log_failure variable to unset after being set
      dm log: remove incorrect field from userspace table output
      dm log: fix userspace status output
      dm stripe: expose correct io hints
      dm table: add more context to terse warning messages
      dm table: fix queue_limit checking device iterator
      dm snapshot: implement iterate devices
      dm multipath: fix oops when request based io fails when no paths
    committed Sep 5, 2009
  16. Merge branch 'for-linus' of git://…

    * 'for-linus' of git://
      PCI SR-IOV: correct broken resource alignment calculations
    committed Sep 5, 2009
  17. Merge git://

    * git://
      sparc64: Fix bootup with mcount in some configs.
      sparc64: Kill spurious NMI watchdog triggers by increasing limit to 30 seconds.
    committed Sep 5, 2009
  18. Merge branch 'perfcounters-fixes-for-linus' of git://…

    * 'perfcounters-fixes-for-linus' of git://
      perf_counter/powerpc: Fix cache event codes for POWER7
      perf_counter: Fix /0 bug in swcounters
      perf_counters: Increase paranoia level
    committed Sep 5, 2009
  19. Merge branch 'for-linus' of git://…

    * 'for-linus' of git://
      Input: atkbd - add Compaq Presario R4000-series repeat quirk
      Input: i8042 - add Acer Aspire 5536 to the nomux list
    committed Sep 5, 2009
  20. ext2: fix unbalanced kmap()/kunmap()

    In ext2_rename(), dir_page is acquired through ext2_dotdot().  It is
    then released through ext2_set_link() but only if old_dir != new_dir.
    Failing that, the pkmap reference count is never decremented and the
    page remains pinned forever.  Repeat that a couple times with highmem
    pages and all pkmap slots get exhausted, and every further kmap() calls
    end up stalling on the pkmap_map_wait queue at which point the whole
    system comes to a halt.
    Signed-off-by: Nicolas Pitre <>
    Acked-by: Theodore Ts'o <>
    Signed-off-by: Linus Torvalds <>
    Nicolas Pitre committed with Sep 5, 2009
  21. Merge branch 'upstream-linus' of git://…

    * 'upstream-linus' of git://
      ocfs2: ocfs2_write_begin_nolock() should handle len=0
      ocfs2: invalidate dentry if its dentry_lock isn't initialized.
    committed Sep 5, 2009
  22. pty: don't limit the writes to 'pty_space()' inside 'pty_write()'

    The whole write-room thing is something that is up to the _caller_ to
    worry about, not the pty layer itself.  The total buffer space will
    still be limited by the buffering routines themselves, so there is no
    advantage or need in having pty_write() artificially limit the size
    And what happened was that the caller (the n_tty line discipline, in
    this case) may have verified that there is room for 2 bytes to be
    written (for NL -> CRNL expansion), and it used to then do those writes
    as two single-byte writes.  And if the first byte written (CR) then
    caused a new tty buffer to be allocated, pty_space() may have returned
    zero when trying to write the second byte (LF), and then incorrectly
    failed the write - leading to a lost newline character.
    This should finally fix

    Reported-by: Mikael Pettersson <>
    Acked-by: Alan Cox <>
    Signed-off-by: Linus Torvalds <>
    committed Sep 5, 2009
  23. n_tty: do O_ONLCR translation as a single write

    When translating CR to CRNL in the n_tty line discipline, we did it as
    two tty_put_char() calls.  Which works, but is stupid, and has caused
    problems before too with bad interactions with the write_room() logic.
    The generic USB serial driver had that problem, for example.
    Now the pty layer had similar issues after being moved to the generic
    tty buffering code (in commit d945cb9:
    "pty: Rework the pty layer to use the normal buffering logic").
    So stop doing the silly separate two writes, and do it as a single write
    instead.  That's what the n_tty layer already does for the space
    expansion of tabs (XTABS), and it means that we'll now always have just
    a single write for the CRNL to match the single 'tty_write_room()' test,
    which hopefully means that the next time somebody screws up buffering,
    it won't cause weeks of debugging.
    Signed-off-by: Linus Torvalds <>
    committed Sep 5, 2009