Openwall GNU/*/Linux - a small security-enhanced Linux distro for servers
[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Date: Fri, 28 Aug 2015 04:55:37 +0200
From: magnum <john.magnum@...hmail.com>
To: john-dev@...ts.openwall.com
Subject: Re: Long duration formats' many-salts figure incorrect
 in short benchmarks

On 2015-08-18 23:41, Solar Designer wrote:
> On Tue, Aug 18, 2015 at 11:31:52PM +0200, magnum wrote:
>> [...] the benchmark timer ends before we even get to try the second
>> salt even once!
> [...]
>> Longer runs show better figures (within limits). This problem is with
>> "your code", but it never surfaces in non-Jumbo. I'm not sure whether to
>> fix this, or how. Should we simply ensure the while loop runs long
>> enough for both salts to be tested at least once? Maybe that would end
>> up showing a mere 2x boost in this case? Should we ensure the loop does
>> all BENCHMARK_MANY virtual salts? But that would likely end up in some
>> GPU formats benchmarking for hours...
>
> Yes, there's this problem with no obviously correct solution.  Testing
> every salt at least once makes sense to me because we've already done so
> during the self-test.  This might increase the test+benchmark time by
> more than a factor of 2, though, since the self-tests run up to a lower
> crypt_all() count argument for most of its invocations.  But it
> shouldn't be worse than 4x overall.  What's worse is that, as you
> correctly point out, this might not achieve the full BENCHMARK_MANY
> speed, requiring the user to manually increase the benchmark duration to
> see that.

Days ago, I simply added a warning so at least you'd know.

FYI I now added an undocumented hack (I have no intentions to document 
it): Now if you give the benchmark time as a negative, as in "-test=-1", 
it will be that time minimum but it will also wait until all salts are 
tested.

$ ../run/john -form:7z -test
Will run 8 OpenMP threads
Benchmarking: 7z, 7-Zip (512K iterations) [SHA256 128/128 AVX 4x AES]... 
(8xOMP) DONE
Speed for cost 1 (iteration count) of 524288
Warning: "Many salts" test limited: 1/256
Many salts:	94.1 c/s real, 14.6 c/s virtual
Only one salt:	91.4 c/s real, 14.6 c/s virtual

$ ../run/john -form:7z -test=-1
Will run 8 OpenMP threads
Benchmarking: 7z, 7-Zip (512K iterations) [SHA256 128/128 AVX 4x AES]... 
(8xOMP) DONE
Speed for cost 1 (iteration count) of 524288
Many salts:	11538 c/s real, 1853 c/s virtual
Only one salt:	94.1 c/s real, 14.6 c/s virtual

$ ../run/john -form:oldoffice -test
Will run 8 OpenMP threads
Benchmarking: oldoffice, MS Office <= 2003 [MD5/SHA1 RC4 32/64]... 
(8xOMP) DONE
Speed for cost 1 (hash type) of 1 and 0
Warning: "Many salts" test limited: 1/256
Many salts:	204800 c/s real, 204800 c/s virtual
Only one salt:	204800 c/s real, 204800 c/s virtual

$ ../run/john -form:oldoffice -test=-1
Will run 8 OpenMP threads
Benchmarking: oldoffice, MS Office <= 2003 [MD5/SHA1 RC4 32/64]... 
(8xOMP) DONE
Speed for cost 1 (hash type) of 1 and 0
Many salts:	2383K c/s real, 371835 c/s virtual
Only one salt:	204800 c/s real, 204800 c/s virtual

magnum

Powered by blists - more mailing lists

Your e-mail address:

Powered by Openwall GNU/*/Linux - Powered by OpenVZ