Openwall GNU/*/Linux - a small security-enhanced Linux distro for servers
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Date: Sat, 30 Jun 2012 16:48:33 -0400
From:  <jfoug@....net>
To: john-dev@...ts.openwall.com
Subject: Re: SHA2 added to bleeding

---- jfoug <jfoug@....net> wrote: 
> I have added sha2.c and sha2.h and modified a lot of formats, to remove the
> oSSL version dependency for SHA224/256/384/512 from JtR.
> 

Here are some quick and dirty 'speed' tests:


Speed tests for new JtR sha2.c code (generic vs oSSL)
All tests on: AMD Atholon64-3400 dual core, non OMP build.

linux-x86-64 build
          raw-sha256    384    512
OpenSSL        2517K  1999k  2086k
Generic        2303k  1751k  1808k
Percent        91.5%  87.6%  86.7%

win32-cygwin-x86-sse2i build  (old 3.4.4 gcc)
          raw-sha256    384    512
OpenSSL        1910K   920k   949k
Generic        1807k   205k   209k
Percent        94.6%    22%    22%

VC 2005 build  (32 bit release)
          raw-sha256    384    512
OpenSSL        1804K   937k   973k
Generic        1807k   427k   437k
Percent         100%  45.5%  44.8%

Hopefully the above formats somewhat OK.  I did this from a smart phone, VPN'd in to my desktop.

I was glad to see that the 64 bit SHA2 formats did acceptable on a 64 bit build.  They suck on 32 bit, but I am pretty sure that the ossl code is all in asm on the 32 bit stuff (or at least they must have asm for ror64(x) )

All tests were done on the same older Athlon 64, fully idle.

Jim.

Powered by blists - more mailing lists

Your e-mail address:

Confused about mailing lists and their use? Read about mailing lists on Wikipedia and check out these guidelines on proper formatting of your messages.