Openwall GNU/*/Linux - a small security-enhanced Linux distro for servers
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Date: Wed, 23 May 2012 10:58:37 -0300
From: Claudio André <claudioandre.br@...il.com>
To: john-dev@...ts.openwall.com
Subject: Re: Nvidia compiler bug

Em 22-05-2012 16:39, magnum escreveu:
> So this made the compiler actually use local memory? Weird. I'll try 
> that in RAR. 

The compiler was using local memory, but it wasn't showing the memory 
usage on its verbosity (It can't on compile time using 1st syntax). And 
my fault, i glanced at it and start thinking the wrong way and i tried 
to fight against the compiler. I failed miserably.

So, I started compiling for linux-x86-64-gpu, and i paid attention on 
what cuda, all other formats, are doing.

If you have doubts about possibilities, you can try it again. But you 
probably toke care better on details than me.
But, it is not a bad idea to reopen your mind and try again LWS+KPC 
impact on your code (fine tunning if necessary).

The key to me was: pay attention on details. BTW: was a pleasure to work 
on GTX and see CUDA compiler working. Can't say the same about "other" 
hardware manufacture. More about this some time later.

> Thanks, I'll have a look at this some time later. magnum 

Thanks, but it seems to be just one more weirdness to be beaten.

For instance, I was thinking about put fire on my GPU and its unpleasant 
software. At least, it is not a strong decision anymore. In this 
situation, we can pay the price of any slowness necessary to get this 
working.

Powered by blists - more mailing lists

Your e-mail address:

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