lkml.org 
[lkml]   [1997]   [Feb]   [6]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
    /
    From
    SubjectRe: MMX performance....
    Date
    In article <Pine.LNX.3.95.970206204244.3277A-100000@ferret.lmh.ox.ac.uk> Chris Evans <chris@ferret.lmh.ox.ac.uk> writes:

    On 6 Feb 1997, Robert Krawitz wrote:

    > The Pentium memcpy() patch, BTW, has a lot of overhead of its own; it
    > dumps and restores the FPU state (when it's in use it dumps the
    > registers; when not, it dumps just the rest of the state). That's why
    > it's configured to operate only when the amount of data to be copied
    > is large. The overhead is well worth it, though, since memory
    > bandwidth on write is used so much more efficiently.

    Is the Pentium memcpy patch in 2.1.x yet? If not, why not? Such a low
    level change should go into the development kernels as early as possible,
    surely...

    I've been sending it off to Linus periodically. He hasn't responded.
    I would like to see it in the kernel.

    Actually it's about time I started using development kernels on my
    personal box. I like to always run with the memcpy patch though. Is there
    at least a 2.1.x patch for it?

    Please see my web page http://www.tiac.net/users/rlk/linux.html.
    --
    Robert Krawitz <rlk@tiac.net> http://www.tiac.net/users/rlk/

    Member of the League for Programming Freedom -- mail lpf@uunet.uu.net
    Tall Clubs International -- http://www.tall.org/ or 1-800-521-2512

    \
     
     \ /
      Last update: 2005-03-22 13:38    [W:7.714 / U:0.116 seconds]
    ©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site