[<prev] [next>] [thread-next>] [day] [month] [year] [list]
Message-ID: <loom.20140722T185913-342@post.gmane.org>
Date: Tue, 22 Jul 2014 17:07:51 +0000 (UTC)
From: Donovan <contact.newangels@...il.com>
To: john-users@...ts.openwall.com
Subject: OMP & Compilation issues on OSX
Hi All,
Usualy, i use Version 1.7.9_OMP, but would like to update my build as well,
to get more improuvement & new aglo's on....
I have just read couples of the last threads, where users seem got problem
who look similar than mine.
I am on MAC OSX mountain Lion.
I use usualy JTR version 1.7.9 Jumbo who are avaible on the custom builds, =
http://openwall.info/wiki/_media/john/john-1.7.9-jumbo-7-macosx-Intel-2.zip
i make some tests with the Magnum build =
https://github.com/magnumripper/JohnTheRipper
Because i would like to update....
& Unfortunately not got the expected improvement...
Look,
On the version 1.7.9
ver: 1.7.9-jumbo-7_omp [macosx-x86-64]
=
pc2:run1 xxxx$ ./john --format=des --test
Benchmarking: Traditional DES [128/128 BS SSE2-16]... (8xOMP) DONE
Many salts: 15925K c/s real, 1985K c/s virtual
Only one salt: 12845K c/s real, 1605K c/s virtual
****************
On the last version from Magnum Github, tested today
( i just do a classic compilation without touching any file or modify any
..., i just do $ ./configure & make, all is done without any errors )
version 1.8.0.2-bleeding-jumbo [darwin12.5.0 64-bit AVX-autoconf]
pc2:run xxxx$ ./john --format=descrypt --test
Benchmarking: descrypt, traditional crypt(3) [DES 128/128 AVX-16]... DONE
Many salts: 4702K c/s real, 4702K c/s virtual
Only one salt: 4483K c/s real, 4438K c/s virtual
pc2:run xxxxx$ ./john --format=descrypt-opencl --test
Device 0: Intel(R) Core(TM) i7-2720QM CPU @ 2.20GHz
Local worksize (LWS) 1, Global worksize (GWS) 16384
Benchmarking: descrypt-opencl, traditional crypt(3) [DES OpenCL]... DONE
Many salts: 2939K c/s real, 396687 c/s virtual
Only one salt: 2833K c/s real, 396187 c/s virtual
pc2:run xxx$ OMP_NUM_THREADS=8 ./john --format=descrypt --test
Benchmarking: descrypt, traditional crypt(3) [DES 128/128 AVX-16]... DONE
Many salts: 4697K c/s real, 4697K c/s virtual
Only one salt: 4480K c/s real, 4480K c/s virtual
**************
Like you can see, seem OMP is not enable as the other 1.7.9 version who can
up at 15925 K C/S, Even in GPU Mode we can't get the expected improuvement.
I try to make test with the addition of "--fork=N" option, but got an error
message, seem we can not Mix " --fork=N & --test""....can you confirm this
point please ?
Any help with be appreciated,
Thank you in advance,
Regards,
Donovan
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.
