This is the mail archive of the
libc-alpha@sourceware.org
mailing list for the glibc project.
Re: [PATCH] Improve unaligned memcpy and memmove.
- From: OndÅej BÃlka <neleai at seznam dot cz>
- To: Liubov Dmitrieva <liubov dot dmitrieva at gmail dot com>
- Cc: GNU C Library <libc-alpha at sourceware dot org>
- Date: Fri, 23 Aug 2013 18:05:59 +0200
- Subject: Re: [PATCH] Improve unaligned memcpy and memmove.
- References: <20130819085220 dot GB19541 at domone> <CAHjhQ925SRFa4ztrwZPJi20PLZwT5pG08T+CMLLdGgMN59=h-Q at mail dot gmail dot com>
On Fri, Aug 23, 2013 at 04:04:01PM +0400, Liubov Dmitrieva wrote:
> We need to retest on Haswell and Silvermont where the same
> implementation is currently used due to bit_Fast_Unaligned_Load is set
> on.
> Can you provide your profile to retest for final check.
>
Sure, A benchmarks that I used are here.
http://kam.mff.cuni.cz/~ondra/benchmark_string/memcpy_profile_loop230813.tar.bz2
http://kam.mff.cuni.cz/~ondra/benchmark_string/memmove_profile230813.tar.bz2
I now recheck with them to also get data from amd processors that were off.
http://kam.mff.cuni.cz/~ondra/benchmark_string/memcpy_profile_loop.html
http://kam.mff.cuni.cz/~ondra/benchmark_string/memmove_profile.html