Follow @Openwall on Twitter for new release announcements and other news
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Date: Sun, 2 Jan 2011 16:52:13 -0800
From: Dhiru Kholia <dhiru.kholia@...il.com>
To: john-users@...ts.openwall.com
Subject: Re: NTLM and OpenCL 1.0

On Thu, Dec 30, 2010 at 11:23 PM, Dhiru Kholia <dhiru.kholia@...il.com> wrote:
> I have added preliminary OpenCL support for raw-MD4 hashes (combined
> OpenCL patch attached). Performance sucks right now but I am working
> on it :-).

I have added OpenCL support for raw-MD4 and raw-MD5 hashes.

$ ../run/john -test -format=raw-md5
Benchmarking: Raw MD5 [raw-md5 64x1]... DONE
Raw:	5874K c/s real, 5933K c/s virtual

$ ../run/john -test -format=raw-md5-opencl
Benchmarking: Raw MD5 [raw-md5-opencl]...
OpenCL Platform: <<<NVIDIA CUDA>>> and device: <<<GeForce GT 240>>>
DONE
Raw:	37374K c/s real, 111025K c/s virtual

$ ../run/john -test -format=raw-md4
Benchmarking: Raw MD4 [32/64]...
OpenCL Platform: <<<NVIDIA CUDA>>> and device: <<<GeForce GT 240>>>
DONE
Raw:	27698K c/s real, 59918K c/s virtual

Areas which needs more work are marked as TODO in the code. Thanks
Erik for all the testing and benchmarking reports!
...

GT 240 is strangely giving better performance (>2X) than AMD 6970
which I got recently. Drivers? Crappy compiler? I have started looking
into AMD CAL to get more performance.

-- 
Cheers,
Dhiru

Download attachment "john-1.7.6-jumbo-9-opencl-9.diff.gz" of type "application/x-gzip" (28514 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.