Re: missing madvise functionality

Eric Dumazet wrote:
A page fault is not that expensive. But clearing N*PAGE_SIZE bytes is,
because it potentially evicts a large part of CPU cache.

*A* page fault is not that expensive. The problem is that you get a
page fault for every single page. For 200k allocated you get 50 page
faults. It quickly adds up.

--
➧ Ulrich Drepper ➧ Red Hat, Inc. ➧ 444 Castro St ➧ Mountain View, CA ❖

Attachment: signature.asc
Description: OpenPGP digital signature

Relevant Pages

  • Re: missing madvise functionality
    ... because it potentially evicts a large part of CPU cache. ... *A* page fault is not that expensive. ... More majordomo info at http://vger.kernel.org/majordomo-info.html ...
    (Linux-Kernel)
  • Re: [PATCH] garbage values in file /proc/net/sockstat
    ... On Monday 23 January 2006 14:28, Eric Dumazet wrote: ... > Shouldnt we force a page fault for not possible cpus in cpu_data ...
    (Linux-Kernel)
  • Re: missing madvise functionality
    ... Eric Dumazet wrote: ... madvisecan relax this requirement (if the pages were ... then a page fault could bring a new page with random content) ... Please read the FAQ at http://www.tux.org/lkml/ ...
    (Linux-Kernel)