john-users - Re: Is JTR MPIrun can be optimized for more cores ?

Openwall wordlists collection for password cracking (20+ languages)

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Date: Sun, 14 Mar 2010 19:38:27 +0300
From: Solar Designer <solar@...nwall.com>
To: john-users@...ts.openwall.com
Subject: Re: Is JTR MPIrun can be optimized for more cores ?

On Sat, Mar 13, 2010 at 11:23:21AM -0700, RB wrote:
> On Sat, Mar 13, 2010 at 09:23, Magnum, P.I. <rawsmooth@...dband.net> wrote:
> > I didn't change one single line of code, just recompiled it with openmpi
> > libs.
> 
> This is similar to what I was seeing; I developed and tested with
> OpenMPI.  My tests (without the jumbo patches) were simply to run
> 'john --test' for a few key ciphers and showed similarly significant
> hits in c/s rates over 10-15 runs each.

Here's a guess: maybe the gcc optimization flags are somehow lost (not
passed on to gcc) when you build with the MPI compiler wrapper?  Maybe
try running your build under strace and see?

Alexander