Openwall GNU/*/Linux - a small security-enhanced Linux distro for servers
[<prev] [next>] [thread-next>] [day] [month] [year] [list]
Date: Fri, 23 Dec 2011 19:49:08 +0400
From: Solar Designer <solar@...nwall.com>
To: john-dev@...ts.openwall.com
Subject: MD5 intrinsics compile-time condition

magnum, Jim -

I don't know if you noticed, but the benchmark results that W.A. posted
to john-users indicate a problem:

http://www.openwall.com/lists/john-users/2011/12/19/3

Benchmarking: FreeBSD MD5 [32/64 X2]... (4xMPI) DONE
Raw:	34034 c/s real, 34034 c/s virtual

Benchmarking: dynamic_27: FreeBSD MD5 [SSE2i 4x3]... (4xMPI) DONE
Raw:	67824 c/s real, 67152 c/s virtual

Benchmarking: dynamic_28: Apache MD5 [SSE2i 4x3]... (4xMPI) DONE
Raw:	68388 c/s real, 68388 c/s virtual

Apparently, the condition that enables the use of intrinsics is not the
same for md5 vs. dynamic_27 and 28, and apparently it is non-optimal for
md5 for certain gcc version(s) (I guess Apple's gcc 4.2).

Would you look into this, please?

Thanks,

Alexander

Powered by blists - more mailing lists

Your e-mail address:

Powered by Openwall GNU/*/Linux - Powered by OpenVZ