[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <f13214b7ada0264fb2c550d3eaf40886@smtp.hushmail.com>
Date: Wed, 20 Mar 2013 22:45:00 +0100
From: magnum <john.magnum@...hmail.com>
To: john-dev@...ts.openwall.com
Subject: Re: bf-opencl broken in bleeding
On 20 Mar, 2013, at 22:33 , Claudio André <claudioandre.br@...il.com> wrote:
> Em 20-03-2013 17:17, Solar Designer escreveu:
>> magnum, Sayantan -
>>
>> Somehow trying to run bf-opencl on -pla=1 on bull results in the CPU
>> kernel being built even when targeting the GPU, and then it segfaults.
>>
>> user@...l:~/john/bleeding-jumbo/run$ ./john -te -fo=bf-opencl -pla=1
>> Device 1: Tahiti (AMD Radeon HD 7900 Series)
>> Segmentation fault
>> user@...l:~/john/bleeding-jumbo/run$ ls -l kernels/bf*bin
>> -rw------- 1 user user 1124600 Mar 21 00:09 kernels/bf_cpu_kernel_Tahiti.bin
>>
>> Alexander
>
> Magnum, this patch seems (i tested on bull) to fix the problem.
Thanks, I was just going to say you could fix it much quicker than me :-)
I screwed up a merge conflict with latest code from unstable (only tested it on a one platform machine). I'm not used to the new device/platform stuff in bleeding.
Committed now.
magnum
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.
