Openwall GNU/*/Linux - a small security-enhanced Linux distro for servers
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Date: Wed, 4 Jan 2012 22:32:03 +0400
From: Solar Designer <solar@...nwall.com>
To: john-dev@...ts.openwall.com
Subject: Re: getting the CUDA & OpenCL stuff into -jumbo

On Wed, Jan 04, 2012 at 06:57:23PM +0100, magnum wrote:
> As soon as there is a patch that supposedly doesn't break stuff, I'll 
> include it in the git tree.

Great.

> I would appreciate some sort of "how to" (if 
> that's not included already), for getting whatever is needed for a Linux 
> system to build and run OpenCL stuff.

Yes, I'd appreciate this kind of content being added to the wiki or/and
put in a text file under doc.

There are some CUDA links here:

http://openwall.info/wiki/john/GPU#CUDA-development-environment

but this is insufficient (I'd like some how-to right on our wiki) and
there's nothing for OpenCL on our wiki yet.

BTW, this has some links to OpenCL implementations for CPUs:

http://www.opennet.ru/opennews/art.shtml?num=32092

Specifically:

https://launchpad.net/pocl
http://code.google.com/p/freeocl/

Maybe we could use these for making our OpenCL code more portable and
for testing it on machines without GPUs.

Alexander

Powered by blists - more mailing lists

Your e-mail address:

Powered by Openwall GNU/*/Linux - Powered by OpenVZ