[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <f51505133095aba4e42bdbdc374237ee@smtp.hushmail.com>
Date: Fri, 3 May 2013 14:26:43 +0200
From: magnum <john.magnum@...hmail.com>
To: john-dev@...ts.openwall.com
Subject: Re: MPI vs. --fork
On 3 May, 2013, at 2:52 , magnum <john.magnum@...hmail.com> wrote:
> On 3 May, 2013, at 0:20 , magnum <john.magnum@...hmail.com> wrote:
>> Currently it's not finished and MPI session save/resume is busted. A warning is printed about that if applicable. I am not aware of any problems with non-MPI builds of bleeding though: All half-baked code is #ifdef HAVE_MPI.
>
> For the record I believe the currently committed version works fine now. The only thing missing is hijacking the -fork line in the session file (using it for MPI as well).
This is in place now too. Works like a champ. Start it as MPI, resume it without MPI and it will -fork instead - or the other way round. My main goal with this was testing but it can be useful in real life too.
$ ../run/john -fork=4 test -ext:keyboard -max-run=5 -ses:test
Loaded 1 password hash (Traditional DES [128/128 BS SSE2-16])
Node numbers 1-4 of 4 (fork)
1 0g 0:00:00:05 0g/s 1081Kp/s 1081Kc/s 1081KC/s l/.lplkj..l/.l,jyt
3 0g 0:00:00:05 0g/s 1030Kp/s 1030Kc/s 1030KC/s kuytgyjm..kuyuyty6
Waiting for 3 children to terminate
2 0g 0:00:00:05 0g/s 1043Kp/s 1043Kc/s 1043KC/s kop9ijik..kop0[0o9
4 0g 0:00:00:05 0g/s 1070Kp/s 1070Kc/s 1070KC/s lil./li8..lil.lkio
Session stopped (max run-time reached)
$ mpirun -np 4 ../run/john -res:test
Loaded 1 password hash (Traditional DES [128/128 BS SSE2-16])
Node numbers 1-4 of 4 (MPI)
0g 0:00:00:11 0g/s 899141p/s 899141c/s 899141C/s 5r3e3wdf..5r3e45tg
Node 1 finished at 0:00:00:11.
0g 0:00:00:11 0g/s 939101p/s 939101c/s 939101C/s 6ukjmkl;..6ukjnbhy
Node 3 finished at 0:00:00:11.
0g 0:00:00:11 0g/s 885282p/s 885282c/s 885282C/s 5tfgrtrt..5tfgr3e2
Node 2 finished at 0:00:00:11.
0g 0:00:00:11 0g/s 934225p/s 934225c/s 934225C/s 6uyu8ilo..6uyu8uhb
Node 0 finished at 0:00:00:11.
Session stopped (max run-time reached)
$ mpirun -np 4 ../run/john test -ext:keyboard -max-run=5 -ses:test
Loaded 1 password hash (Traditional DES [128/128 BS SSE2-16])
Node numbers 1-4 of 4 (MPI)
0g 0:00:00:05 0g/s 1033Kp/s 1033Kc/s 1033KC/s ku7yhtgf..ku7ygnmn
Node 0 finished at 0:00:00:05.
0g 0:00:00:05 0g/s 1021Kp/s 1021Kc/s 1021KC/s kjy7y56t..kjyjhu8u
Node 1 finished at 0:00:00:05.
0g 0:00:00:05 0g/s 1056Kp/s 1056Kc/s 1056KC/s kmjukjhn..kmjujyjk
Node 3 finished at 0:00:00:05.
0g 0:00:00:05 0g/s 1019Kp/s 1019Kc/s 1019KC/s kjhgfrfv..kjhghnjm
Node 2 finished at 0:00:00:05.
Session stopped (max run-time reached)
$ ../run/john -res:test
Loaded 1 password hash (Traditional DES [128/128 BS SSE2-16])
Node numbers 1-4 of 4 (fork)
1 0g 0:00:00:11 0g/s 940893p/s 940893c/s 940893C/s 6uk./.ko..6uk./;l/
Waiting for 3 children to terminate
2 0g 0:00:00:11 0g/s 916165p/s 916165c/s 916165C/s 67i767iu..67i76uhy
3 0g 0:00:00:11 0g/s 901829p/s 901829c/s 901829C/s 6567yjhn..6567yhyu
4 0g 0:00:00:11 0g/s 943418p/s 943418c/s 943418C/s 6uilijuk..6uiloijk
Session stopped (max run-time reached)
magnum
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.
