[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Date: Mon, 18 Feb 2013 22:23:20 +0100
From: magnum <john.magnum@...hmail.com>
To: john-dev@...ts.openwall.com
Subject: Re: remaining GPU issues
On 18 Feb, 2013, at 22:09 , Claudio André <claudioandre.br@...il.com> wrote:
> Em 18-02-2013 18:06, Solar Designer escreveu:
>> On Sun, Feb 10, 2013 at 08:37:48PM +0100, magnum wrote:
>>> I always thought the main problem was the driver, and you upgraded it. Do we get ASIC hangs from any format now?
>> We didn't have any ASIC hangs on bull for 37 days, but Claudio has just
>> managed to trigger one.
>>
>> Alexander
>
>
> Two (i guess).
>
> The 1st (I tried to use LWS=1 GWS=1). Not sure it matters. The second, (if confirmed) was a test=0.
> -----
>
>
> Benchmarking: ZIP-AES PBKDF2-HMAC-SHA-1 [OpenCL]... DONE
> Raw: 25.0 c/s real, 100 c/s virtual
>
> Device 1: Tahiti (AMD Radeon HD 7900 Series)
> Benchmarking: OpenBSD Blowfish (x32) [OpenCL]... -
> ../../../thread/semaphore.cpp:87: sem_wait() failed
> ^C^C
> Broadcast message from root@...l
> (unknown) at 0:39 ...
>
> The system is going down for reboot in 1 minute!
>
That format kills most cards I have tried it on. It does not care about LWS or GWS environment variables (from what I know) and it does not even honor device's reported limits. I'm tempted to just disable it in the repo.
DES is also ignoring device limits but it usually doesn't kill the GPU.
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.
