Openwall GNU/*/Linux - a small security-enhanced Linux distro for servers
[<prev] [next>] [thread-next>] [day] [month] [year] [list]
Date: Mon, 6 Feb 2006 01:25:35 +0000 (UTC)
From:  Radim Horak <yesbody@...nam.cz>
To: john-users@...ts.openwall.com
Subject:  GPU acceleration?

Hi everybody,
maybe this is totally crazy idea, but I'm quite sober, so let me start with this 
citation that summarizes my inspiration:

"The SM3.0 GPUs have become general and powerful enough to perform a broader 
range of computations – beyond the role of pure graphical shading for which they 
were initially used."
http://www.havok.com/%20index.php?option=com_content&task=view&id=187&Itemid=77

So the 2 big questions for programmers are:
1] Is it feasible/plausible to use SM3 GPU for password cracking?
2] If yes, would S.D. or anybody consider implementing it in john?


Supplemental SM3 informations:
- Shader Model v.3 graphics accelerators include these widespread models:
nVidia 6x00 and 7x00 line, ATI Xx00 line
- They incorporate good parallelism
- the length of the shader programs is not limited
- technology development cycles of GPUs are faster than CPUs
- The upcoming Unified shader model should be even more universal
- high level languages: HLSL and Cg
- They are connected with fast 3-4 GBs PCI-express bus

My naive idea is, that the generated password candidates could be stored in 
texture formats (texture pixel can be represented as 4 bytes (RGBA), so psw. 
candidate/hash could "use" 2-4 pixels.), than encrypted/processed with pixel 
and/or vertex shader programs, then find the matches...


Relevant links, programming documentation:
http://msdn.microsoft.com/library/default.asp?url=/library/en-us/directx9_c/
HLSL_Shaders.asp
ftp://download.nvidia.com/developer/cg/Cg_1.4/Docs/CG_UserManual_1-4.pdf

-Radim


Powered by blists - more mailing lists

Your e-mail address:

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