[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Date: Thu, 22 Nov 2012 20:23:31 +0100
From: magnum <john.magnum@...hmail.com>
To: john-users@...ts.openwall.com
Subject: Re: scalar mode
That's great. So maybe I should re-enable vectorizing by default for some AMD devices? I only know Cayman and Juniper benefits from it. Maybe I'll just single these two out.
magnum
On 22 Nov, 2012, at 16:29 , Marc Brinkmann <marc.brinkmann@...il.com> wrote:
> Hey Magnum,
>
> you are truly a genius and made it work again.
> Recent version:
>
> make[1]: Verlasse Verzeichnis '/home/marc/JohnTheRipper/src'
> cp opencl/*.cl ../run/
> cp opencl_*.h ../run/
> marc@...berd:~/JohnTheRipper/src$ cd ../run/
> marc@...berd:~/JohnTheRipper/run$ ./john --test --format=office2010-opencl
> OpenCL platform 0: AMD Accelerated Parallel Processing, 2 device(s).
> Using device 0: Juniper
> Local worksize (LWS) 64, Global worksize (GWS) 8192
> Benchmarking: Office 2010 SHA-1 AES (100,000 iterations) [OpenCL]... DONE
> Raw: 3103 c/s real, 204800 c/s virtual
>
> marc@...berd:~/JohnTheRipper/run$ ./john --test --request-vectorize
> --format=office2010-opencl
> OpenCL platform 0: AMD Accelerated Parallel Processing, 2 device(s).
> Using device 0: Juniper
> Local worksize (LWS) 64, Global worksize (GWS) 4096
> Benchmarking: Office 2010 SHA-1 AES (100,000 iterations) [OpenCL 4x]... DONE
> Raw: 5251 c/s real, 327680 c/s virtual
>
> marc@...berd:~/JohnTheRipper/run$
>
>
> -> Scalar and Vectorized are working
>
> Amazing :)
> BR Marc
>
>
> 2012/11/17 magnum <john.magnum@...hmail.com>
>
>> On 17 Nov, 2012, at 17:13 , Marc Brinkmann <marc.brinkmann@...il.com>
>> wrote:
>>> Hmm, i think i missed the < in my patch-command.
>>> i ran $ patch <bughunt.diff
>>> Then i recompiled and testet John
>>>
>>> I'm sorry but see for yourself, segfault each time:
>>
>> OK, thanks. I think I'll just give up for now. Sorry.
>>
>> magnum
>>
>>
Powered by blists - more mailing lists
Powered by Openwall GNU/*/Linux -
Powered by OpenVZ
