Follow @Openwall on Twitter for new release announcements and other news
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Date: Wed, 04 Apr 2012 01:32:13 +0200
From: magnum <>
Subject: Re: RAR OpenCL committed to git

On 04/04/2012 01:11 AM, Solar Designer wrote:
> On Wed, Apr 04, 2012 at 12:23:17AM +0200, magnum wrote:
>> On a GTX580, it currently does around 4000 c/s in -hp mode (cRARk OpenCL
>> does ~5500 c/s using same card). I got it to 4500 c/s by moving GPU
>> arrays to __local memory but then the self-test fails (although no
>> warnings from building etc - I hate that) and I have no idea what is
>> wrong. On an old 9600GT, I get 290 c/s.
> user@...l:~/john/magnum-jumbo/src$ ../run/john -te -fo=rar
> No protocol specified
> No protocol specified

What's that "No protocol" stuff?

> OpenCL platform 0: NVIDIA CUDA, 1 device(s).
> Using device 0: GeForce GTX 570
> Max local work size 576, best multiple 32
> Local work size (LWS) 512, Keys per crypt (KPC) 8192
> AES-NI engine not available
> Benchmarking: RAR3 (6 characters) [OpenCL]... (8xOMP) DONE
> Raw:    2269 c/s real, 2079 c/s virtual
> I wonder if something needs to be re-tuned for GTX 570.  It shouldn't be
> that much slower than 580.  Oh, maybe your 4000 c/s is for a system with

Strange. No, AES-NI won't make any difference for short blocks like -hp
mode or the tiny inlined -p mode files, nor will OMP. I would expect an
overclocked 570 to be similar to a 580. Note that the usual LWS/KPC
environment variables and/or john.conf options can be used (but there's
currently no find-best enumeration built in if setting them to zero).

I got best result (but with long runtime) from LWS=576 KPC=9216


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.