Openwall GNU/*/Linux - a small security-enhanced Linux distro for servers
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Date: Tue, 30 Oct 2012 21:54:07 +0100
From: magnum <john.magnum@...hmail.com>
To: john-dev@...ts.openwall.com
Subject: Re: Serious bug in md5crypt-cuda - or in driver?

On 10/30/12 21:45, Lukas Odzioba wrote:
> 2012/10/30 magnum <john.magnum@...hmail.com>:
>> The current md5crypt-cuda code works fine with Fermi and Kepler cards but on a 9600GT I noticed today I got unreasonably good speed:
>>
>> Benchmarking: md5crypt [CUDA]... DONE
>> Raw:    63351K c/s real, 63351K c/s virtual
>>
>> At first I did not spot anything unusual but hey wait a minute, is that a K tucked on the figures?!
>>
>> So I ran it through the Test Suite which very quickly ends up claiming everything passed. But some manual testing revels that it will simply accept ANY password as valid for ANY hash. We should enhance the Test Suite (and possibly the self tests?) so it can detect false positives like these.
>
> It accepts any pass for any hash only on 9600GT, for rest what have
> you tested it it worked as it should?

I only see problems on this 9600GT. The heftier cards has no problems 
that I can see, no false positives (I explicitly verified that). And 
speed figures are sane. It seems the kernel call on 9600GT returns 
immediately (some kind of silent crash?) and for some reason the 
"cracked" array is set. Is there any more error checking that can be 
made in CUDA than already is the case?

magnum



Powered by blists - more mailing lists

Your e-mail address:

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