Skip to content
Commits on Jul 20, 2012
  1. @supercurio

    Merge branch 'nexus-s-voodoo-sound' into nexus-s-voodoo

    supercurio committed
    Update to Jelly Bean kernel source code into Voodoo branch
  2. @supercurio

    Merge branch 'android-samsung-3.0-jb' into nexus-s-voodoo-sound

    supercurio committed
    Update to Jelly Bean kernel source code
Commits on Jun 20, 2012
  1. @ajs1984

    gpu: pvr: Flush the whole CPU cache when allocating uncached.

    ajs1984 committed with Android Partner Code Review
    This is a workaround for range-based CPU cache invalidation
    apparently not working reliably.
    
    Change-Id: I62fe3d10a465ee630729d7edf054f7bc82a422d8
Commits on Jun 18, 2012
  1. Revert "gpu: pvr: Add thread safety checks to SGX MMU code."

    jp abgrall committed with Android Partner Code Review
    Not needed anymore, was just to get a prebuild that QA could use.
    
    This reverts commit ca5c535
  2. Revert "gpu: pvr: Fix a bug that prevented MMU PTs from being dumped."

    jp abgrall committed with Android Partner Code Review
    Not needed anymore, was just to get a prebuild that QA could use.
    
    This reverts commit 5a7b953
  3. Revert "gpu: pvr: Harden MMU code's use-after-free checks."

    jp abgrall committed with Android Partner Code Review
    Not needed anymore, was just to get a prebuild that QA could use.
    
    
    This reverts commit d00c00e
  4. Revert "gpu: pvr: Intentionally leak SGX MMU PTs."

    jp abgrall committed with Android Partner Code Review
    Not needed anymore, was just to get a prebuild that QA could use.
    
    This reverts commit ce3815d
Commits on Jun 15, 2012
  1. @ajs1984

    gpu: pvr: Intentionally leak SGX MMU PTs.

    ajs1984 committed with JP Abgrall
    When page tables would normally be freed, leak them instead.
    
    This experiment is to try to prove a distinction between a use-after-free
    type bug and another driver corrupting our page tables. At the point the
    asserts go off, we don't expect the page to have been freed yet. So it
    should contain only valid PTEs. If however the PT is being used after
    free, it might contain junk from other kernel drivers. If we don't free
    the PTs, the latter should never happen.
    
    Change-Id: I69714cfd0ee81adb9a60a996d11e8373e209dfa9
  2. Revert "herring_defconfig: Enable SLUB / PAGEALLOC debug."

    jp abgrall committed with Android Partner Code Review
    It needs to be undone for experiment 3.
    
    This reverts commit 8a055ff
  3. @ajs1984

    herring_defconfig: Enable SLUB / PAGEALLOC debug.

    ajs1984 committed with Android Partner Code Review
    To help us diagnose an issue relating to SGX MMU page table corruption.
    
    Change-Id: Idede40ddcb4fe6216513de83be955df1897750fb
  4. Revert "gpu: pvr: Intentionally leak SGX MMU PTs."

    JP Abgrall committed
    Clicked submit in the wrong window. It needs to go after some other change.
    
    This reverts commit cc92b07
    Change-Id: Ie2cc8d1715af6b9c40145f5f3e857051916bea0a
    Signed-off-by: JP Abgrall <jpa@google.com>
  5. @ajs1984

    gpu: pvr: Harden MMU code's use-after-free checks.

    ajs1984 committed with Android Partner Code Review
    Add signature words to MMU_CONTEXT and MMU_PT_INFO. These words
    are initialized at alloc time and zeroed at free time. They are
    checked any time a page table is validated before being freed.
    
    Change-Id: I74aa57c1e8b48e89bfbb1f4f8a4120c54859122f
  6. @ajs1984

    gpu: pvr: Intentionally leak SGX MMU PTs.

    ajs1984 committed with Android Partner Code Review
    When page tables would normally be freed, leak them instead.
    
    This experiment is to try to prove a distinction between a use-after-free
    type bug and another driver corrupting our page tables. At the point the
    asserts go off, we don't expect the page to have been freed yet. So it
    should contain only valid PTEs. If however the PT is being used after
    free, it might contain junk from other kernel drivers. If we don't free
    the PTs, the latter should never happen.
    
    Change-Id: I3046bb81896ed6ae4ea1f2de19a62a0e5e89e063
  7. @ajs1984

    gpu: pvr: Fix a bug that prevented MMU PTs from being dumped.

    ajs1984 committed with Android Partner Code Review
    DumpPT already had a static inline if PT_DEBUG or PT_DUMP
    weren't defined, so it shouldn't have been #ifdef guarded.
    
    Furthermore, PT_DEBUG isn't enabled so DumpPT would not have been
    called to dump the page tables in one case. This is incorrect as
    it should have been called if PT_DUMP was enabled.
    
    Change-Id: I4a0d2a3665b7586a5dfa4390098df4edb6edfe74
  8. @ajs1984

    gpu: pvr: Add thread safety checks to SGX MMU code.

    ajs1984 committed with Android Partner Code Review
    The SGX MMU code was not designed to be used from multiple driver
    threads. One theory is that somehow the driver does sometimes use
    the MMU code from multiple threads, which could cause corruption
    of meta data.
    
    We must allow PID 1 through because the kernel uses this ID before
    userspace is up, and the module_init code interacts with the MMU
    subsystem.
    
    Otherwise, check that the outer bridge lock is held. If it is held,
    check the PID of the caller matches the PID holding the bridge lock.
    
    Change-Id: I35f42f251316b158576e5d853f8099e24942b4f2
Commits on Jun 8, 2012
  1. Merge remote-tracking branch 'common/android-3.0' into android-samsun…

    JP Abgrall committed
    …g-30-wip-mergedown
    
    * common/android-3.0:
      ext4: add missing save_error_info() to ext4_error()
      net: wireless: bcmdhd: Update to version 5.90.195.75
      net: wireless: bcmdhd: Fix P2P driver crash for MFG firmware
      net: wireless: bcmdhd: Make responce waiting uninterruptible
  2. @tytso

    ext4: add missing save_error_info() to ext4_error()

    tytso committed with Ken Sumrall
    The ext4_error() function is missing a call to save_error_info().
    Since this is the function which marks the file system as containing
    an error, this oversight (which was introduced in 2.6.36) is quite
    significant, and should be backported to older stable kernels with
    high urgency.
    
    Change-Id: Ia1eb8d91f37ceb67faf3b79d6bc79b899f1d6bfc
    Reported-by: Ken Sumrall <ksumrall@google.com>
    Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
    Cc: ksumrall@google.com
    Cc: stable@kernel.org
    Signed-off-by: Ken Sumrall <ksumrall@android.com>
  3. ARM: s5pv210: herring: FIMC1 buffers are not used, so reduce them.

    JP Abgrall committed
    jung-min.oh@samsung.com suggested ...
    """
     Some memory can be reduced in crespo.
    
     For normal video playback, liboverlay in GB uses the reserved memory of
     FIMC1
    
     but libhwcomposer in ICS uses the reserved memory of FIMD.
    """
    
    Change-Id: Id3f40d78795fb96900678fb97ee8e0189daa79d3
    Signed-off-by: JP Abgrall <jpa@google.com>
Commits on Jun 6, 2012
  1. @ajs1984

    gpu: pvr: Update to DDK 1.8@945322

    ajs1984 committed
    This is a version number change only, for compatibility with the
    user-mode driver.
    
    Change-Id: I272428a490fbbd18ce9016e92e82829c5dc6b898
Commits on Jun 5, 2012
  1. Revert "gpu: pvr: Update to DDK 1.8@943950"

    JP Abgrall committed
    This reverts commit a4f4aa1.
    
    The DDK change is not yet qualified as a JB release.
    But there is a USB VBUS fix needed.
  2. Revert "power: enable USB VBUS of cp and WiMAX"

    JP Abgrall committed with Android Partner Code Review
    This reverts commit b029380.
    
    Breaks UART handling in the rild.
      06-01 19:23:04.156    83    83 E RIL(s)  : UART_Read : select - timeout occur!
    
    Bug: 6601359
    Change-Id: Iffe12ae24e4738c0944d960fad49d9fe94d9b59d
    Signed-off-by: JP Abgrall <jpa@google.com>
Commits on Jun 1, 2012
  1. gpu: pvr: Update to DDK 1.8@943950

    Ben Jones committed
    This is a version number change only, for compatibility with the
    user-mode driver.
    
    Change-Id: I0cfc6d73c1a386597f96efa97a3bd2e2d366e760
Commits on May 31, 2012
  1. net: wireless: bcmdhd: Update to version 5.90.195.75

    Dmitry Shmidt committed
    - Fix false PCB-OVERLAP issue
    - Fix simultanious connect request on two P2P devices
    
    Signed-off-by: Dmitry Shmidt <dimitrysh@google.com>
Commits on May 30, 2012
  1. net: wireless: bcmdhd: Fix P2P driver crash for MFG firmware

    Leslie Yu committed with Dmitry Shmidt
    Signed-off-by: Dmitry Shmidt <dimitrysh@google.com>
  2. net: wireless: bcmdhd: Make responce waiting uninterruptible

    Dmitry Shmidt committed
    Signed-off-by: Dmitry Shmidt <dimitrysh@google.com>
  3. Merge remote-tracking branch 'common/android-3.0' into android-samsun…

    JP Abgrall committed
    …g-30-wip-mergedown
    
    * common/android-3.0:
      USB: gadget: f_audio_source: Adjust packet timing to reduce glitches
      usb: gadget: android: Fix product name
      net: wireless: bcmdhd: Fix WPS PBC overlap failure
      usb: gadget: composite: Fix corruption when changing configuration
Commits on May 29, 2012
  1. USB: gadget: f_audio_source: Adjust packet timing to reduce glitches

    Mike Lockwood committed
    Increase max packet size and clean up timing logic so we can better
    recover from not getting an interrupt in time for a SOF.
    
    Signed-off-by: Mike Lockwood <lockwood@google.com>
  2. @benoitgoby

    usb: gadget: android: Fix product name

    benoitgoby committed
    Product names may contain spaces and scanf %s only matches the 1st word.
    Use strlcpy instead.
    
    Change-Id: Ie8703fea9775f7fc17fe615a42597ca3816d36b0
    Signed-off-by: Benoit Goby <benoit@android.com>
Commits on May 24, 2012
  1. net: wireless: bcmdhd: Fix WPS PBC overlap failure

    Neeraj Kumar Garg committed with Dmitry Shmidt
    Signed-off-by: Dmitry Shmidt <dimitrysh@google.com>
Commits on May 22, 2012
  1. @benoitgoby

    usb: gadget: composite: Fix corruption when changing configuration

    benoitgoby committed
    Remove the config from the configs list before releasing the spinlock.
    Otherwise the other cpu might be processing a SET_CONFIGURATION that
    will switch to the configuration that is being released.
    
    Bug: 6521576
    Change-Id: Id4da0d0e18ead63e20cb236cd1d3e8e6d116acce
    Signed-off-by: Benoit Goby <benoit@android.com>
Commits on May 17, 2012
  1. Merge branch 'android-3.0' into android-samsung-3.0-wip

    Dmitry Shmidt committed
Commits on May 16, 2012
  1. net: wireless: bcmdhd: Ignore signal_pending() while waiting in IOCTL

    Dmitry Shmidt committed
    Signed-off-by: Dmitry Shmidt <dimitrysh@google.com>
  2. net: wireless: bcmdhd: Check return value from dhd_dev_init_ioctl()

    Dmitry Shmidt committed
    Signed-off-by: Dmitry Shmidt <dimitrysh@google.com>
  3. net: wireless: bcmdhd: Fix WARN_ON(!res->pub.channel)

    Dmitry Shmidt committed
    Signed-off-by: Dmitry Shmidt <dimitrysh@google.com>
  4. net: wireless: bcmdhd: Change singal pending return value from -110 t…

    Dmitry Shmidt committed
    …o -4
    
    - ETIMEDOUT is interpreted as FW is not responding,
      so return EINTR instead
    
    Signed-off-by: Dmitry Shmidt <dimitrysh@google.com>
Something went wrong with that request. Please try again.