Openwall GNU/*/Linux - a small security-enhanced Linux distro for servers
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Date: Tue, 29 Nov 2011 18:17:07 -0500 (EST)
From: "Brad Tilley" <brad@...ystems.com>
To: john-users@...ts.openwall.com
Subject: Re: AMD X6 multicore issue

I think I did this right. Feel free to add it to the wiki on my behalf.
Some over-zealous "security admins" disallow the entire TLD of .info on my
current network.

Brad

--------------------------

model name	: AMD Phenom(tm) II X6 1090T Processor
stepping	: 0
cpu MHz		: 3210.559
cache size	: 512 KB

--------------------------

1.7.9 Without OpenMP

cd src && make clean && make clean linux-x86-64

./john --test
Benchmarking: Traditional DES [128/128 BS SSE2-16]... DONE
Many salts:	3376K c/s real, 3382K c/s virtual
Only one salt:	3218K c/s real, 3218K c/s virtual

Benchmarking: BSDI DES (x725) [128/128 BS SSE2-16]... DONE
Many salts:	110848 c/s real, 110848 c/s virtual
Only one salt:	108646 c/s real, 108646 c/s virtual

Benchmarking: FreeBSD MD5 [32/64 X2]... DONE
Raw:	16089 c/s real, 16089 c/s virtual

Benchmarking: OpenBSD Blowfish (x32) [32/64 X2]... DONE
Raw:	1070 c/s real, 1070 c/s virtual

Benchmarking: Kerberos AFS DES [48/64 4K]... DONE
Short:	520345 c/s real, 520345 c/s virtual
Long:	1477K c/s real, 1477K c/s virtual

Benchmarking: LM DES [128/128 BS SSE2-16]... DONE
Raw:	43999K c/s real, 43999K c/s virtual

Benchmarking: generic crypt(3) [?/64]... DONE
Many salts:	306681 c/s real, 306681 c/s virtual
Only one salt:	305203 c/s real, 305203 c/s virtual

Benchmarking: Tripcode DES [48/64 4K]... DONE
Raw:	413900 c/s real, 413900 c/s virtual

Benchmarking: dummy [N/A]... DONE
Raw:	108892K c/s real, 108675K c/s virtual


--------------------

1.7.9 With OpenMP (uncommented the two gcc lines in Makefile)

cd src && make clean && make clean linux-x86-64

./john --test
Benchmarking: Traditional DES [128/128 BS SSE2-16]... DONE
Many salts:	17758K c/s real, 2959K c/s virtual
Only one salt:	14091K c/s real, 2348K c/s virtual

Benchmarking: BSDI DES (x725) [128/128 BS SSE2-16]... DONE
Many salts:	585523 c/s real, 97587 c/s virtual
Only one salt:	511795 c/s real, 86189 c/s virtual

Benchmarking: FreeBSD MD5 [32/64 X2]... DONE
Raw:	80325 c/s real, 14051 c/s virtual

Benchmarking: OpenBSD Blowfish (x32) [32/64 X2]... DONE
Raw:	5367 c/s real, 920 c/s virtual

Benchmarking: Kerberos AFS DES [48/64 4K]... DONE
Short:	527104 c/s real, 527104 c/s virtual
Long:	1486K c/s real, 1486K c/s virtual

Benchmarking: LM DES [128/128 BS SSE2-16]... DONE
Raw:	60707K c/s real, 10220K c/s virtual

Benchmarking: generic crypt(3) [?/64]... DONE
Many salts:	1501K c/s real, 250622 c/s virtual
Only one salt:	1417K c/s real, 237153 c/s virtual

Benchmarking: Tripcode DES [48/64 4K]... DONE
Raw:	414822 c/s real, 414822 c/s virtual

Benchmarking: dummy [N/A]... DONE
Raw:	112167K c/s real, 112167K c/s virtual




> Brad -
>
> On Tue, Nov 29, 2011 at 04:03:48PM -0500, Brad Tilley wrote:
>> I have one of these CPUs (AMD x6 1090T). I'd be glad to test if that would
>> be helpful. Let me know.
>
> I'm quite sure the problem is not CPU type specific, but it'd be nice if
you run benchmarks of 1.7.9 built with and without OpenMP and submit
your results to the wiki:
>
> http://openwall.info/wiki/john/benchmarks
>
> As to SHA-crypt, there's currently no column in those tables for it
(because JtR does not include its own code for it, using the system's),
but you may post such results to this thread.  I'd expect something like
2000 c/s with glibc's crypt_r(3) code on a 6-core (for a 64-bit build).
>
> Thanks,
>
> Alexander

Powered by blists - more mailing lists

Your e-mail address:

Confused about mailing lists and their use? Read about mailing lists on Wikipedia and check out these guidelines on proper formatting of your messages.