Follow @Openwall on Twitter for new release announcements and other news
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Date: Sun, 24 Nov 2013 17:35:19 +0100
From: magnum <>
Subject: Re: md5crypt-opencl

On 2013-11-24 12:47, Lukas Odzioba wrote:
> Solar stated this on private list:
> 2013/11/24 Solar Designer <>:
>> BTW, I guess one of the reasons hashcat/md5crypt/AMD is so good is that
>> hashcat groups candidate passwords by length.  When there are different
>> length passwords being tested on a GPU at the same time, you're probably
>> wasting some local memory on supporting excess lengths for the shorter
>> ones of the passwords.  You could try implementing something similar -
>> e.g. per-length OpenCL kernel invocations out of one crypt_all(), or
>> simply detection of and optimization for the special case when all
>> passwords in a crypt_all() are same length (maybe you do this already?)

While that is probably a good idea, our currently active test vectors 
are only of length 8 so IRL we're even worse than the benchmark :-(

BTW isn't some of what we do in SSE-intrinsics.c appliable to a GPU? I 
think it is but I never really groked that SSE code.

> So what's my current status, I have code that pass self-test
> (saltlen=8,pass=8). c/s results are the following:
> 570: 1.1k c/s
> 7970 1.6k c/s
> Titan 2.2k c/s

You mean x1000 in all of the above, no? I get 194k using a laptop GPU 
with latest bleeding.

The HC+ figures: 3445k c/s on 7970 and 1044 c/s on GTX 580.


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.