Openwall GNU/*/Linux - a small security-enhanced Linux distro for servers
[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Date: Thu, 10 Jan 2013 19:43:29 +0100
From: magnum <john.magnum@...hmail.com>
To: john-dev@...ts.openwall.com
Subject: Re: OpenCL formats failing test

On 10 Jan, 2013, at 8:37 , Solar Designer <solar@...nwall.com> wrote:
> After upgrading Catalyst to 12.8 and APP SDK to 2.7, these two formats
> are failing test on 7970:
> 
> $ ./john -pla=1 -te -fo=zip-opencl
> OpenCL platform 1: AMD Accelerated Parallel Processing, 2 device(s).
> Using device 0: Tahiti
> Local work size (LWS) 192, Global work size (GWS) 9216
> Benchmarking: ZIP-AES PBKDF2-HMAC-SHA-1 [OpenCL]... FAILED (cmp_all(1))

Fixed, in shared pbkdf2 kernel. Due to endianness, derived keys with non-word-boundary lengths did not get the last bytes(s) filled in.

> $ ./john -pla=1 -te -fo=office2013-opencl
> OpenCL platform 1: AMD Accelerated Parallel Processing, 2 device(s).
> Using device 0: Tahiti
> Local worksize (LWS) 64, Global worksize (GWS) 16384
> Benchmarking: Office 2013 SHA-512 AES (100,000 iterations) [OpenCL]... Segmentation fault

Still debugging this. It works fine on some devices. It does fail on some nvidias too, and OTOH it does work on some AMD CPU drivers. Weird.

magnum

Powered by blists - more mailing lists

Your e-mail address:

Confused about mailing lists and their use? Read about mailing lists on Wikipedia and check out these guidelines on proper formatting of your messages.