Follow @Openwall on Twitter for new release announcements and other news
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Date: Sat, 10 Nov 2012 02:28:30 +0100
From: magnum <>
Subject: Re: Split kernel for OpenCL WPA-PSK

On 9 Nov, 2012, at 20:00 , magnum <> wrote:
> On 11/08/2012 08:09 PM, magnum wrote:
>> On 8 Nov, 2012, at 19:12 , magnum <> wrote:
>>> Using device 0: Tahiti
>>> Local worksize (LWS) 192, Global worksize (GWS) 196608
>>> Benchmarking: WPA-PSK PBKDF2-HMAC-SHA-1 [OpenCL]... DONE
>>> Raw:    66197 c/s real, 137970 c/s virtual
>>> This code too does over 2.1 billion SHA1/second, but CPU post-processing nearly halves the speed (without OMP). So I'm in the process of moving all of that post-processing to GPU. It's just a couple HMACs more, so I hope to exceed 120K c/s with that in place.
> Done, but not committed yet. It now does NO post processing on CPU:
> Using device 0: Tahiti
> Local worksize (LWS) 128, Global worksize (GWS) 262144
> Benchmarking: WPA-PSK PBKDF2-HMAC-SHA-1 [OpenCL]... DONE
> Raw:    129453 c/s real, 52428K c/s virtual
> This is a tad faster than HashCat unless Atom has tweaked it since the figure I found (otherwise he will now... OTOH I'm not done yet =). Jumbo-7 does 42723 c/s out of the box. 

This is now committed. Please test. NOTE that the version 1 hashes is completely untested (I believe that goes for the CUDA and CPU versions too). We lack test vectors!

Plaintext length was bumped from 15 to 32 (OpenCL only) without performance regression. IMHO 15 is too short unless there are performance reasons.


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.