Follow @Openwall on Twitter for new release announcements and other news
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Date: Sun, 12 Aug 2012 04:58:11 +0200
From: magnum <john.magnum@...hmail.com>
To: john-dev@...ts.openwall.com
Subject: Re: Jumbo 7

On 2012-08-12 03:43, magnum wrote:
> This is a good test and we need to look into the problems, but most of
> the offending formats do not exist in the Jumbo-7 branch (actually named
> 1.7.9-jumbo-6-fixes).

OK, some of the failed ones are actually in -fixes. Most of these are
Samuele's formats and he does not seem to maintain them anymore. All of
these pass on bull's GTX 570.

> On 2012-08-12 01:27, Lukas Odzioba wrote:
>> OpenCL platform 0: NVIDIA CUDA, 1 device(s).
>> Using device 0: GeForce GTX 460
>> Local work size (LWS) 256, Global work size (GWS) 2097152
>> Benchmarking: MySQL 4.1 double-SHA-1 [OpenCL]... FAILED
>> (get_hash[0](0))

>> OpenCL platform 0: NVIDIA CUDA, 1 device(s).
>> Using device 0: GeForce GTX 460
>> Local work size (LWS) 256, Global work size (GWS) 2097152
>> Benchmarking: Raw SHA-1 OpenCL [OpenCL]... FAILED (get_hash[1](0))

>> OpenCL platform 0: NVIDIA CUDA, 1 device(s).
>> Using device 0: GeForce GTX 460
>> Global work size = 524288
>> Benchmarking: Raw SHA-512 [OpenCL]... FAILED (get_hash[0](0))

>> OpenCL platform 0: NVIDIA CUDA, 1 device(s).
>> Using device 0: GeForce GTX 460
>> Building the kernel, this could take a while
>> ./jtr_bencher.sh: line 6: 27532 Segmentation fault      $dir/john -test -fo=$fmt

I believe the above was sha512crypt-opencl. Segfault while building?

>> OpenCL platform 0: NVIDIA CUDA, 1 device(s).
>> Using device 0: GeForce GTX 460
>> Local work size (LWS) 256, Global work size (GWS) 4194304
>> Benchmarking: Netscape LDAP salted SHA-1 [OpenCL]... FAILED (get_hash[0](0))

>> OpenCL platform 0: NVIDIA CUDA, 1 device(s).
>> Using device 0: GeForce GTX 460
>> Local work size (LWS) 128, Global work size (GWS) 524288
>> Benchmarking: Mac OS X 10.7+ salted SHA-512 [OpenCL]... FAILED (get_hash[0](0))


BTW, I tried pegging EncFS LWS at 32, and it works - but it takes a long
time:

magnum@...l:src [magnum-jumbo]$ time ../run/john -t -fo:encfs-opencl
OpenCL platform 0: NVIDIA CUDA, 1 device(s).
Using device 0: GeForce GTX 570
Benchmarking: EncFS PBKDF2 AES / Blowfish [OpenCL]... DONE
Raw:    348 c/s real, 348 c/s virtual

real    11m25.636s
user    5m32.137s
sys     5m53.206s

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.