[lkml]   [2011]   [Jun]   [28]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
Messages in this thread
    SubjectRe: [PATCH v3] futex: Fix regression with read only mappings

    On 06/28/2011 03:54 AM, Peter Zijlstra wrote:
    > On Mon, 2011-06-27 at 17:22 -0500, Shawn Bohrer wrote:
    >> commit 7485d0d3758e8e6491a5c9468114e74dc050785d (futexes: Remove rw
    >> parameter from get_futex_key()) in 2.6.33 introduced a user-mode
    >> regression in that it additionally prevented futex operations on a
    >> region within a read only memory mapped file. For example this breaks
    >> workloads that have one or more reader processes doing a FUTEX_WAIT on a
    >> futex within a read only shared mapping, and a writer processes that has
    >> a writable mapping issuing the FUTEX_WAKE.
    >> This fixes the regression for futex operations that should be valid on
    >> RO mappings by trying a RO get_user_pages_fast() when the RW
    >> get_user_pages_fast() fails so as not to slow down the common path of
    >> writable anonymous maps and bailing when we used the RO path on
    >> anonymous memory.
    >> While fixing the regression this patch opens up two possible bad
    >> scenarios as identified by KOSAKI Motohiro:
    >> 1) This patch also allows FUTEX_WAIT on RO private mappings which have
    >> the following corner case.
    >> Thread-A: call futex(FUTEX_WAIT, memory-region-A).
    >> get_futex_key() return inode based key.
    >> sleep on the key
    >> Thread-B: call mprotect(PROT_READ|PROT_WRITE, memory-region-A)
    >> Thread-B: write memory-region-A.
    >> COW happen. This process's memory-region-A become related
    >> to new COWed private (ie PageAnon=1) page.
    >> Thread-B: call futex(FUETX_WAKE, memory-region-A).
    >> get_futex_key() return mm based key.
    >> IOW, we fail to wake up Thread-A.
    >> 2) Current futex code doesn't handle zero page properly.
    >> Read mode get_user_pages() can return zero page, but current futex
    >> code doesn't handle it at all. Then, zero page makes infinite loop
    >> internally.
    >> This Patch is based on Peter Zijlstra's initial patch with modifications to
    >> only allow RO mappings for futex operations that need VERIFY_READ access.
    >> Reported-by: David Oliver <>
    >> Signed-off-by: Shawn Bohrer <>
    > Acked-by: Peter Zijlstra <>

    I think we need to address #2 above first.

    1) Is the loop killable?
    2) If not, and we try to catch zero_page scenario, we need to ensure RO
    sparse file mapping work. Peter notes that these are probably fine,
    filemap.c doesn't seem to use the zero page - but we need to test.

    It's agreed, I believe, between myself, Peter, and Shawn that we will
    document Private RW mappings as unsupported.

    Darren Hart
    Intel Open Source Technology Center
    Yocto Project - Linux Kernel

     \ /
      Last update: 2011-06-28 16:59    [W:0.027 / U:7.236 seconds]
    ©2003-2016 Jasper Spaans. hosted at Digital OceanAdvertise on this site