Follow @Openwall on Twitter for new release announcements and other news
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Date: Fri, 14 Dec 2012 23:55:37 -0200
From: Claudio André <claudioandre.br@...il.com>
To: john-dev@...ts.openwall.com
Subject: Re: Benchmark versus incremental mode (speed)

Em 14-12-2012 21:17, magnum escreveu:
> On 14 Dec, 2012, at 23:19 , Claudio André <claudioandre.br@...il.com> wrote:
>> Magnum, all
>>
>> Why benchmark shows a speed so different from real cracking (even now)? Is there anything i missed? Should i do something in a different way?
>> If possible, could you please give me any hint or advice?
> It should be a lot better since I merged that self-test timing patch a week ago or so. Maybe we see problems now that were hidden before.
>
>> First point, which one is the real number? Please take a look in [1] and [2].
>> My test vector is not that bad comparing to --incremental mode. Or is it?
> Per definition, the real crack gives the real figure. If they differ it may be due to things like test vector length not matching the IRL run (in case the format suffers significantly from length) or if the IRL hash does not use 5000 iterations. If that's not it, I'm not sure.

If i change the test vector to this, it works as "expected" [3].

{"$5$LKO/Ute40T3FNF95$U0prpBQd4PloSGU0pnpM4z9wKn4vZ1.jsrzQfPqxph9", 
"U*U*U*U*"},
//{"$5$LKO/Ute40T3FNF95$fdgfoJEBoMajNxCv3Ru9LyQ0xZgv0OBMQoq80LQ/Qd.", 
"U*U***U"},
//{"$5$LKO/Ute40T3FNF95$8Ry82xGnnPI/6HtFYnvPBTYgOL23sdMXn8C29aO.x/A", 
"U*U***U*"},
//{"$5$9mx1HkCz7G1xho50$O7V7YgleJKLUhcfk9pgzdh3RapEaWqMtEp9UUBAKIPA", 
"*U*U*U*U"},
//{"$5$kc7lRD1fpYg0g.IP$d7CMTcEqJyTXyeq8hTdu/jB/I6DGkoo62NXbHIR7S43", ""},
     {NULL}

So, could you please think about it and see if you have any new ideas? I 
copied my test vector from CPU code.

> claudio@...l:~/unstable/src$ ../run/john -fo:sha256crypt-opencl ~/sha256 -inc --max-run-time=600
> OpenCL platform 0: NVIDIA CUDA, 1 device(s).
> Using device 0: GeForce GTX 570
> Building the kernel, this could take a while
> Local work size (LWS) 128, global work size (GWS) 23040
> guesses: 0  time: 0:00:11:40 0.00%  c/s: 19753  trying: tui1194 - sidgsb
> Session stopped (max run-time reached)
> Why do I not see any "Loaded x hashes with y different salts" in the above?

Sorry, the real command was:
  ../run/john -fo:sha256crypt-opencl ~/sha256 -inc --max-run-time=600 > 
~/teste1

You can see new complete results in [1].


> What happens if you change BENCHMARK_LENGH to 0 and re-run this benchmark?
Nothing new [2].

Thanks


[1]
claudio@...l:~/claudio/src$ ../run/john -fo:sha256crypt-opencl ~/sha256 
-inc --max-run-time=6
OpenCL platform 0: NVIDIA CUDA, 1 device(s).
Using device 0: GeForce GTX 570
Building the kernel, this could take a while
Local work size (LWS) 128, global work size (GWS) 23040
Loaded 1 password hash (sha256crypt [OpenCL])
guesses: 0  time: 0:00:00:08 0.00%  c/s: 18162  trying: tig4 - stures!
Session stopped (max run-time reached)

claudio@...l:~/claudio/src$ ../run/john -fo:sha256crypt-opencl ~/sha256 
-inc --max-run-time=600
OpenCL platform 0: NVIDIA CUDA, 1 device(s).
Using device 0: GeForce GTX 570
Building the kernel, this could take a while
Local work size (LWS) 128, global work size (GWS) 23040
Loaded 1 password hash (sha256crypt [OpenCL])
guesses: 0  time: 0:00:11:41 0.00%  c/s: 19739  trying: tui1194 - sidgsb
Session stopped (max run-time reached)

[2]
claudio@...l:~/claudio/src$ ../run/john -fo:sha256crypt-opencl -t
OpenCL platform 0: NVIDIA CUDA, 1 device(s).
Using device 0: GeForce GTX 570
Building the kernel, this could take a while
Local work size (LWS) 128, global work size (GWS) 23040
Benchmarking: sha256crypt (rounds=5000) [OpenCL]... DONE
Many salts:    13552 c/s real, 13592 c/s virtual
Only one salt:    13633 c/s real, 13552 c/s virtual

[3]
claudio@...l:~/claudio/src$ ../run/john -fo:sha256crypt-opencl -t
OpenCL platform 0: NVIDIA CUDA, 1 device(s).
Using device 0: GeForce GTX 570
Building the kernel, this could take a while
Local work size (LWS) 128, global work size (GWS) 23040
Benchmarking: sha256crypt (rounds=5000) [OpenCL]... DONE
Raw:    18885 c/s real, 18731 c/s virtual

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.