Openwall GNU/*/Linux - a small security-enhanced Linux distro for servers
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Date: Wed, 10 Apr 2013 04:25:47 +0400
From: Solar Designer <solar@...nwall.com>
To: john-dev@...ts.openwall.com
Subject: Re: [patch] sse/xop implementation of raw-sha256

On Tue, Apr 09, 2013 at 08:29:12PM +0200, magnum wrote:
> On 9 Apr, 2013, at 19:00 , Jeremi Gosney <epixoip@...dshell.nl> wrote:
> > I was surprised to learn there was no SIMD implementation of SHA2 in
> > JTR, so I wrote one. I've completed SHA256, and have attached a patch
> > for it. I will begin work on SHA512 shortly.
> 
> Cool, thanks! Committed now.

First of all, I really appreciate Jeremi's work on this.  Thanks!

Then, I think we should convert this format to non-pluggable, because it
only makes sense on SSE2+ (as currently implemented), so the struct in
john.c should only be expected and registered in those builds.  Also,
the check for __GNUC__ is probably not needed - or maybe it should be
around the #pragma optimize.  Non-gcc compilers have these intrinsics too.

So I think both the new source file's body and the added lines in john.c
should be wrapped in #ifdef __SSE2__ ... #endif.  Don't we do it this
way for Tavis' raw-sha1-ng?  I think it's a similar situation.

Alexander

Powered by blists - more mailing lists

Your e-mail address:

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