Follow @Openwall on Twitter for new release announcements and other news
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Date: Sat, 19 Jul 2014 12:37:55 +0200
From: magnum <>
Subject: Re: Help Compiling John on ARM w/ MPI & CUDA

On 2014-07-18 17:15, Pyrex wrote:
> When running it utilizing MPI you do not need to specify the device?
> ex: "--device=0"

With or without MPI, CUDA defaults to device 0. Actually the CUDA+MPI 
stuff is probably not complete in case you'd have some nodes with two or 
more devices. The OpenCL code handles that just fine but the CUDA 
version of that is probably not complete (I can't recall doing it). But 
as long as all nodes just have one CUDA device, it should work fine as is.

> Also, do I have to utilize OpenMPI or will it work with MPICH2?

It it supposed to work just fine with MPICH2 as well, or anything else 
that is OpenMPI compliant.


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.