[<prev] [next>] [day] [month] [year] [list]
Date: Thu, 28 Jun 2012 08:08:37 +0200
From: Frank Dittrich <frank_dittrich@...mail.com>
To: john-dev@...ts.openwall.com
Subject: Some extremely long running self tests for --test=0 (linux-x86-sse2i-OMP)
A few self tests are extremely slow for linux-x86-sse2i-OMP, even when I
use --test=0.
At least for keepass I verified that the run time is normal for non-OMP.
$ for f in `./john --list=formats|sed 's#,# #g'`;do ( time ./john
--test=0 --format=$f ) 2>&1 ; done|tee test0_runtime.txt
$ grep -B 5 -A 2 "^real" test0_runtime.txt |grep -E -B 5 -A 2
"^real(\s[1-9]|\s0m[0-9][0-9])"
Warning: doing quick benchmarking - the performance numbers will be
inaccurate
Benchmarking: KeePass SHA-256 AES [32/32]... (2xOMP) DONE
Raw: 10.0 c/s real, 5.0 c/s virtual
real 0m42.676s
user 1m23.719s
sys 0m0.047s
--
Warning: doing quick benchmarking - the performance numbers will be
inaccurate
Benchmarking: Django PBKDF2-HMAC-SHA-256 (x10000) [32/32]... (2xOMP) DONE
Raw: 10.4 c/s real, 5.3 c/s virtual
real 0m40.332s
user 1m18.925s
sys 0m0.041s
Plus a few one that take less than 10 seconds, but more than 2:
$ grep -B 5 -A 2 "^real" test0_runtime.txt |grep -B 5 -A 2
"^real\s0m[2-9]\."
Warning: doing quick benchmarking - the performance numbers will be
inaccurate
Benchmarking: Mac OS X Keychain PBKDF2-HMAC-SHA-1 3DES [32/32]...
(2xOMP) DONE
Raw: 203 c/s real, 104 c/s virtual
real 0m2.156s
user 0m4.168s
sys 0m0.021s
--
Warning: doing quick benchmarking - the performance numbers will be
inaccurate
Benchmarking: Office 2007/2010 SHA-1/AES [32/32]... (2xOMP) DONE
Raw: 7.8 c/s real, 3.9 c/s virtual
real 0m3.059s
user 0m5.778s
sys 0m0.013s
--
Warning: doing quick benchmarking - the performance numbers will be
inaccurate
Benchmarking: Drupal 7 $S$ SHA-512 (x16385) [32/32]... (2xOMP) DONE
Raw: 14.2 c/s real, 7.1 c/s virtual
real 0m4.500s
user 0m8.550s
sys 0m0.011s
--
Warning: doing quick benchmarking - the performance numbers will be
inaccurate
Benchmarking: sha512crypt (rounds=5000) [32/32]... (2xOMP) DONE
Raw: 44.4 c/s real, 22.3 c/s virtual
real 0m2.930s
user 0m5.642s
sys 0m0.005s
I don't mind run time of a few seconds, but more than 40 seconds?
This means
$ ./john --test=0|grep FAIL
will not run fast enough do do it regularly.
Frank
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.
