Openwall GNU/*/Linux - a small security-enhanced Linux distro for servers
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Date: Tue, 21 Aug 2012 20:30:06 +0200
From: magnum <john.magnum@...hmail.com>
To: john-dev@...ts.openwall.com
Subject: Re: OpenCL optimization trick for md*

On 2012-08-21 18:51, Alain Espinosa wrote:
> Hi. Share some small OpenCL optimization i will use in next Hash Suite.

Thanks! These are already in place in some of the formats, and often
ifdef'ed like this (they harm nvidia performance, at least this one):

#ifdef cl_nv_pragma_unroll
#define NVIDIA
#endif
...
#ifdef NVIDIA
#define F(x,y,z)        (z ^ (x & (y ^ z)))
#else
#define F(x,y,z)        bitselect(z, y, x)
#endif


But someone should check all formats. I tried variants like the below
for a quick check:

$ grep -H -n "&" opencl/*cl|sed 's/ //g'|grep '.^(.&(.^.)'

$ grep -H -n "&" opencl/*cl|sed 's/ //g'|grep '((.&(.|.))|(.&.))'


magnum

Powered by blists - more mailing lists

Your e-mail address:

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