[PATCH v1.2] Improve unaligned memcpy and memmove.

[Liubov Dmitrieva]

I am surprised that rep is faster on Atom because Atom is known for slow reps...
We should recheck it.

You probably should join that memcpy patches into one to simplify
review and to make clear what version for which processor will be
finally used.

--
Liubov

On Fri, Oct 4, 2013 at 5:29 PM, Ondřej Bílka <neleai@seznam.cz> wrote:
> On Fri, Oct 04, 2013 at 05:07:35PM +0400, Liubov Dmitrieva wrote:
>> Can we make "**back" versions clean up in this patch?
>> Are there any processors still use it?
>> Atom and core2 uses "***ssse3" version not the "***back" ones.
>> Do we need to handle these "***back" versions now?
>>
> Yes but not in this patch, they are used at memmmove.
>
> I do this in improved ssse3 version that I wrote here,
> http://www.sourceware.org/ml/libc-alpha/2013-08/msg00279.html
> I will prepare actual patch today.
>
> Ondra
>
>> --
>> Liubov
>>
>