[lkml]   [2011]   [Oct]   [18]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
Messages in this thread
    SubjectRe: [PATCH] mm: munlock use mapcount to avoid terrible overhead
    On Tue, 18 Oct 2011, Andi Kleen wrote:
    > Hugh Dickins <> writes:
    > > A process spent 30 minutes exiting, just munlocking the pages of a large
    > > anonymous area that had been alternately mprotected into page-sized vmas:
    > > for every single page there's an anon_vma walk through all the other
    > > little vmas to find the right one.
    > We had the same problem recently after a mmap+touch workload: in this
    > case it was hugepaged walking all these anon_vmas and the list was over
    > 100k long.
    > Had some data on this at plumbers:
    > > A general fix to that would be a lot more complicated (use prio_tree on
    > > anon_vma?), but there's one very simple thing we can do to speed up the
    > > common case: if a page to be munlocked is mapped only once, then it is
    > > our vma that it is mapped into, and there's no need whatever to walk
    > > through all the others.
    > I think we need a generic fix, this problem does not only happen
    > in munmap.

    Thanks for the pointer, Andi, I'll have to look into it when I've a
    moment; but I don't look forward to making this area more complicated.


     \ /
      Last update: 2011-10-19 03:03    [W:0.022 / U:42.852 seconds]
    ©2003-2016 Jasper Spaans. hosted at Digital OceanAdvertise on this site