[SciPy-user] linalg.expm() Illegal Instruction Error?

Dylan Gorman dgorman@berkeley....
Sun Jun 14 01:19:32 CDT 2009


David,

Interesting. If I compile an ATLAS build for each node, how would I  
let numpy know to use the particular ATLAS for the node it happens to  
be running on?

Regards,
Dylan

On Jun 13, 2009, at 10:02 PM, David Cournapeau wrote:

> On Sun, Jun 14, 2009 at 1:46 PM, Dylan Gorman<dgorman@berkeley.edu>  
> wrote:
>> Thanks, Robert.
>>
>> The situation is a little tricky. I installed numpy 1.3.0 from source
>> from, I guess, this sourceforge link: http://sourceforge.net/project/downloading.php?group_id=1369&filename=numpy-1.3.0.tar.gz&a=58155458
>>
>> I should perhaps elaborate. I have a small cluster, on which I have
>> non-root access. I've installed python 2.5.2, numpy 1.3.0, and scipy
>> 0.7rc1 to ~. ATLAS is already configured on the system, so prior to
>> installing these packages I set
>> export ATLAS=/usr/local/atlas
>> in ~/.bash_profile
>>
>> I'm just aiming to run non-parallel batch jobs on each node of the
>> cluster. The code works quite happily on a set of newer nodes in this
>> cluster, but on the older ones we are getting this Illegal  
>> instruction
>> error.
>
> The "optimal" solution is to build atlas for each node - this way, you
> will get an optimized ATLAS for each node. That's a maintenance
> nightmare.
>
> The other solution is to use ATLAS optimized on the machine with the
> lowest common denominator (say a machine with SSE only), and deploy
> this everywhere else.
>
> The easiest is to avoid atlas and use pure BLAS/LAPACK,
>
> David
> _______________________________________________
> SciPy-user mailing list
> SciPy-user@scipy.org
> http://mail.scipy.org/mailman/listinfo/scipy-user



More information about the SciPy-user mailing list