[arch-general] imagemagick 126.96.36.199-1 is slower than 188.8.131.52-1
archlinux at cryptocrack.de
Sat Oct 18 06:22:33 UTC 2014
On Fri, 17 Oct 2014 at 20:58:22, Lukas Fleischer wrote:
> Anatol and I were able to reproduce this issue. It seems to be related
> to __memcpy_avx_unaligned() in glibc which means that it only occurs on
> architectures with the AVX extension.
> If you have a look at the memcpy-avx-unaligned.S source code , you
> will notice that there are several branches that copy blocks of
> different sizes. Now, for some reason, 184.108.40.206-1 always (or almost
> always) picks the L(less_32bytes) branch which means that only small
> blocks are copied, while 220.127.116.11-1 copies larger blocks. I do not have
> the time to debug this in detail but maybe you can add this information
> to the upstream report?
After a more thorough analysis by Jan, we found out that the issue is
caused by a bug in the new OpenCL benchmark code. On modern CPUs, the
benchmark is executed on every start of ImageMagick which leads to huge
number of __memcpy_avx_unaligned() calls with a small block size. Jan
prepared a patch and he will submit it upstream.
>  https://sourceware.org/git/?p=glibc.git;a=blob_plain;f=sysdeps/x86_64/multiarch/memcpy-avx-unaligned.S;hb=HEAD
More information about the arch-general