Improve memcpy for Atom

H.J. Lu hjl.tools@gmail.com
Wed Oct 19 16:21:00 GMT 2011


On Wed, Oct 19, 2011 at 6:11 AM, Michael Zolotukhin
<michael.v.zolotukhin@gmail.com> wrote:
> Hi,
>
> This patch contains one function:
> __memcpy_ssse3
>
> It improves memcpy on small sizes and on sizes between half of shared
> cache size and shared cache size for Atom (up to 40% performance
> gain).
>
> The patch was tested on Atom.
>
> Change Log:
> 2011-10-11  Michael Zolotukhin  <michael.v.zolotukhin@gmail.com>
>
>        * sysdeps/i386/i686/multiarch/memcpy-ssse3.S: Update.
>        XMM-moves are used for copying on small sizes. Use
> SHARED_CACHE_SIZE instead of
>        SHARED_CACHE_SIZE_HALF.
>

We use SHARED_CACHE_SIZE_HALF on purpose.  memcpy in one
process may be slower in some cases, but so it won't starve other
processes for cache.


-- 
H.J.



More information about the Libc-alpha mailing list