AW: Linux-x86_64-CUDA version 2.8 on CentOS-5 x86_64 non local user issue?

From: Norman Geist (norman.geist_at_uni-greifswald.de)
Date: Mon Mar 26 2012 - 00:43:17 CDT

Ok so it seems to be related to namd. You could look if the problem persists
with 2.9b just to avoid searching a problem that is already solved, if not,
we need to look at the source.

Best regards

Norman Geist.

> -----Ursprüngliche Nachricht-----
> Von: owner-namd-l_at_ks.uiuc.edu [mailto:owner-namd-l_at_ks.uiuc.edu] Im
> Auftrag von Tru Huynh
> Gesendet: Sonntag, 25. März 2012 00:03
> An: Norman Geist
> Cc: Namd Mailing List
> Betreff: Re: namd-l: Linux-x86_64-CUDA version 2.8 on CentOS-5 x86_64
> non local user issue?
>
> On Fri, Mar 23, 2012 at 08:09:28AM +0100, Norman Geist wrote:
> > Tru,
> >
> > Unfortunately you still didn't executed a cuda binary, just a tool
> that
> > talks to the driver to get some data. Please run a program that does
> real
> > computation on the gpu. A tool that needs to alloc the gpu and uses
> it, so
> > we can clearly figure out if it is a namd problem or a general one.
> >
> /c5/shared/cuda/4.1.28/C/bin/linux/release/fastWalshTransform
> (for a user listed in /etc/passwd or non listed there)
>
> [fastWalshTransform] starting...
>
> Initializing data...
> ..allocating CPU memory
> ..allocating GPU memory
> ..generating data
> Data length: 8388608; kernel length: 128
> Running GPU dyadic convolution using Fast Walsh Transform...
> GPU time: 35.576000 ms; GOP/s: 8.134894
> Reading back GPU results...
> Running straightforward CPU dyadic convolution...
> Comparing the results...
> Shutting down...
> L2 norm: 1.021579E-07
> [fastWalshTransform] test results...
> PASSED
>
> Tru
>
> --
> Dr Tru Huynh | http://www.pasteur.fr/recherche/unites/Binfs/
> mailto:tru_at_pasteur.fr | tel/fax +33 1 45 68 87 37/19
> Institut Pasteur, 25-28 rue du Docteur Roux, 75724 Paris CEDEX 15
> France

This archive was generated by hypermail 2.1.6 : Tue Dec 31 2013 - 23:21:48 CST