Re: [PATCH linux-2.6.1-rc1-mm1] aiodio_fallback_bio_count.patch

From: Daniel McNeil (daniel_at_osdl.org)
Date: 12/31/03

  • Next message: Russell King: "[PATCH] Fix "echo -n 3 > /sys/.../power/state"" 
    To: Andrew Morton <akpm@osdl.org>
Date:	31 Dec 2003 14:41:15 -0800

    This is an update of AIO fallback patch and with the bio_count race
    fixed by changing bio_list_lock into bio_lock and using that for all
    the bio fields. I changed bio_count and bios_in_flight from atomics
    into int. They are now proctected by the bio_lock. I fixed the race,
    by in finished_one_bio() by leaving the bio_count at 1 until after the
    dio_complete() and then do the bio_count decrement and wakeup holding
    the bio_lock.

    I re-diffed to 2.6.1-rc1-mm1.

    I've been stressing this code for several weeks.

    Daniel

    -
    To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
    the body of a message to majordomo@vger.kernel.org
    More majordomo info at http://vger.kernel.org/majordomo-info.html
    Please read the FAQ at http://www.tux.org/lkml/

  • Next message: Russell King: "[PATCH] Fix "echo -n 3 > /sys/.../power/state""

    Relevant Pages

    • Re: [PATCH linux-2.6.1-rc1-mm1] aiodio_fallback_bio_count.patch
      ... Daniel McNeil wrote: ... > This is an update of AIO fallback patch and with the bio_count race ... I fixed the race, ... send the line "unsubscribe linux-kernel" in ...
      (Linux-Kernel)
    • Re: question about /proc/<PID>/mem in 2.4 (fwd)
      ... Kind regards ... > the whole design of the dump/suid race fixing work rather than me. ... I followed the advice of Alan Cox and looked up "Solar Designer" on ... send the line "unsubscribe linux-kernel" in ...
      (Linux-Kernel)
    • Re: [PATCH linux-2.6.1-rc1-mm1] filemap_fdatawait.patch
      ... and we should simply fix generic_file_direct_IO to avoid doing so. ... The race that we need to worry about is between background writeouts ... This is the same race that we have been discussing (background writer ... send the line "unsubscribe linux-kernel" in ...
      (Linux-Kernel)
    • Re: 2.6.8.1-mm2
      ... Having reiserfs dig into the radix tree looks like a layering ... This has the same race issue as any of the "hole punch" ... depends on the nopage functions not searching past i_size, ... send the line "unsubscribe linux-kernel" in ...
      (Linux-Kernel)
    • Re: [patch 2.6.13-rc4] fix get_user_pages bug
      ... > Since we will have dropped the page table lock when calling ... I don't think we can fix that race (which is probably pretty much ... send the line "unsubscribe linux-kernel" in ...
      (Linux-Kernel)