Openwall GNU/*/Linux - a small security-enhanced Linux distro for servers
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Date: Wed, 29 Feb 2012 11:07:38 -0300
From: Claudio André <claudioandre.br@...il.com>
To: john-dev@...ts.openwall.com
Subject: Re: Re: OpenCL errors

Hi

On the notebook I first tried the AMD Catalyst drivers i get at the AMD 
site. Everything was ok, except, OpenCL. So i tried the driver provided 
by Ubuntu and it simply worked. I know it shouldn't, but it happened. 
I'm using what worked and was selectec/provided by Canonical.

I noticed there are some OpenCL files provided by the video driver, and 
they are not overriden by 2.6 SDK. Maybe one of these 2.5 SDK files is 
helping in the sem_wait area, even after i installed the 2.6 SDK.

I tried and It works fine on 2.5.

My opinion, AMD should test everything better. Meanwhile, forget about 
SDK 2.6.

Claudio


Em 28-02-2012 20:19, Samuele Giovanni Tonon escreveu:
> On 02/28/12 23:48, Claudio André wrote:
>> Hi, below (and attached) more information.
>>
>> The error only happens while testing with ./john --test or ./john --test=0.
>>
>> It works fine if i test only one format.
>> - ./john -test -fo:raw-sha1-opencl or ./john -test=0 -fo:raw-sha1-opencl
> this is good: seems a problem related to the fact that when doing
> multiple test there's no "destructor" of the previous format.
>
>> Catalyst: 11.11, driver: 8.911, more detais on the image.
> oh this is strange: i thought amd sdk 2.6 only worked with 12.1, maybe
> this is why you are not getting the sem_wait problem.
>
> does this happens when using sdk 2.5 ? supposing you are using
> the one from amd and just untarring it should not be a problem, if you
> could try it would be great however i don't want to mess your notebook
> so feel free to not do that if you are not confident you can go back to
> 2.6.
>
>
> Cheers
> Samuele
>
>

Powered by blists - more mailing lists

Your e-mail address:

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