[lockup] Re: objrmap-core-1 (rmap removal for file mappings to avoid 4:4 in <=16G machines)

From: Ingo Molnar (mingo_at_elte.hu)
Date: 03/09/04

  • Next message: Stefan Smietanowski: "Re: disable partitioning!" 
    Date:	Tue, 9 Mar 2004 11:52:26 +0100
To: Andrea Arcangeli <andrea@suse.de>

    * Andrea Arcangeli <andrea@suse.de> wrote:

    > This patch avoids the allocation of rmap for shared memory and it uses
    > the objrmap framework to do find the mapping-ptes starting from a
    > page_t which is zero memory cost, (and zero cpu cost for the fast
    > paths)

    this patch locks up the VM.

    To reproduce, run the attached, very simple test-mmap.c code (as
    unprivileged user) which maps 80MB worth of shared memory in a
    finegrained way, creating ~19K vmas, and sleeps. Keep this process
    around.

    Then try to create any sort of VM swap pressure. (start a few desktop
    apps or generate pagecache pressure.) [the 500 MHz P3 system i tried
    this on has 256 MB of RAM and 300 MB of swap.]

    stock 2.6.4-rc2-mm1 handles it just fine - it starts swapping and
    recovers. The system is responsive and behaves just fine.

    with 2.6.4-rc2-mm1 + your objrmap patch the box in essence locks up and
    it's not possible to do anything. The VM is looping within the objrmap
    functions. (a sample trace attached.)

    Note that the test-mmap.c app does nothing that a normal user cannot do.
    In fact it's not even hostile - it only has lots of vmas but is
    otherwise not actively pushing the VM, it's just sleeping. (Also, the
    test is a very far cry from Oracle's workload of gigabytes of shm mapped
    in a finegrained way to hundreds of processes.) All in one, currently i
    believe the patch is pretty unacceptable in its present form.

            Ingo

    Pid: 7, comm: kswapd0
    EIP: 0060:[<c013ee6d>] CPU: 0
    EIP is at page_referenced_obj+0xdd/0x120
     EFLAGS: 00000246 Not tainted
    EAX: cb311808 EBX: cb311820 ECX: 40a2d000 EDX: cb311848
    ESI: cfe202fc EDI: cfe2033c EBP: cfdf9dc4 DS: 007b ES: 007b
    CR0: 8005003b CR2: 40507000 CR3: 0b11e000 CR4: 00000290
    Call Trace:
     [<c013ef71>] page_referenced+0xc1/0xd0
     [<c0137bad>] refill_inactive_zone+0x3fd/0x4c0
     [<c01376bc>] shrink_cache+0x26c/0x360
     [<c0137d11>] shrink_zone+0xa1/0xb0
     [<c01380d7>] balance_pgdat+0x1a7/0x200
     [<c013820b>] kswapd+0xdb/0xe0
     [<c01180b0>] autoremove_wake_function+0x0/0x50
     [<c01180b0>] autoremove_wake_function+0x0/0x50
     [<c0138130>] kswapd+0x0/0xe0
     [<c01050f9>] kernel_thread_helper+0x5/0xc

    -
    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: Stefan Smietanowski: "Re: disable partitioning!"

    Relevant Pages

    • Re: [PATCH] anobjrmap 1/6 objrmap
      ... >> as soon as it hits swap). ... >> whole patch and compare it with your other below results. ... but you must use my objrmap patch too ... send the line "unsubscribe linux-kernel" in ...
      (Linux-Kernel)
    • Re: [PATCH] anobjrmap 1/6 objrmap
      ... >> Mmmm, if you have a broken out patch, it'd be preferable. ... > this one is against mainline, but you must use my objrmap patch too ... Your whole aa3 tree runs nicely, ... send the line "unsubscribe linux-kernel" in ...
      (Linux-Kernel)
    • Re: [PATCH] anobjrmap 1/6 objrmap
      ... >> this one is against mainline, but you must use my objrmap patch too ... Your whole aa3 tree runs nicely, ... > to have the broken out patch before publishing comparisons, ... find_trylock_page cost is a per-page pagecache thing absolutely ...
      (Linux-Kernel)
    • Re: [lockup] Re: objrmap-core-1 (rmap removal for file mappings to avoid 4:4 in <=16G machines)
      ... Ugh. ... I kind of like objrmap and things may be fixable... ... lists, that could well fix the problem you're running ... send the line "unsubscribe linux-kernel" in ...
      (Linux-Kernel)
    • Re: anon_vma RFC2
      ... Dave's objrmap, the second is your anobjrmap-1, the third is my anon_vma ... my patch is not stable yet, it crashes during swapping and the debugging ... send the line "unsubscribe linux-kernel" in ...
      (Linux-Kernel)