Follow @Openwall on Twitter for new release announcements and other news
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20120617163154.GA9294@openwall.com>
Date: Sun, 17 Jun 2012 20:31:54 +0400
From: Solar Designer <solar@...nwall.com>
To: john-dev@...ts.openwall.com
Subject: Re: [patch] optional new raw sha1 implemetation

On Sun, Jun 17, 2012 at 08:02:43PM +0400, Solar Designer wrote:
> (different invocations of the same code).  We should be able to improve
> this further to 30M+ c/s.  I am not making use of _mm_cmov_si128() yet.

Well, I did not get it to 30M, but it is close.  Best invocation:

Benchmarking: Raw SHA-1 [taviso sse4]... DONE
Raw:    29696K c/s real, 29696K c/s virtual

Worst:

Benchmarking: Raw SHA-1 [taviso sse4]... DONE
Raw:    28199K c/s real, 28199K c/s virtual

Patch attached.  While at it, I also added a fix to sse-intrinsics.c for
32-bit builds with SSSE3.

Alexander

View attachment "taviso-sha1-xop.diff" of type "text/plain" (5837 bytes)

Powered by blists - more mailing lists

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