Hello,
I quickly browse through your output files. They are alright. We observe
the same failures on all the machines. The failures in the test suite you
are observing are due to singular matrix pencils that the LAPACK solver is
not supposed to be able to solve. We ought to remove those pencils at some
point ...
Best wishes,
Julien.
On Tue, 21 Apr 2009, Jorge Naranjo wrote:
Dear all
I have been trying to build lapack version 3.2.1 on Power PC platform
(ppc970) running SUSE Linux SLES9-SP3
with the IBM XL Fortran Advanced Edition V10.1 for Linux
Version: 10.01.0000.0000
The problem is the compilation seems to go smoothly and it is also quite
easy and straight forward but once built, it fails running some test and
I am not sure about the relevance of this failures...
I am sending you a file with the failed tests along with the make.inc
used in the compilation.
Could you please give some light on this issue??
I am not quite sure if I should address you for this matter... if not,
sorry if I am bothering you
Thank you so much in advance
sincerely,
--
|--------------------------------------------------------------.--.
|Jorge Naranjo Bouzas. |o_o |
|Barcelona Supercomputing Center (BSC) |:_/ |
|Centro Nacional de Supercomputacion (CNS) // \ \
|Campus Nord de la UPC, Edificio Capilla. (| | )
|C\ Jordi Girona, 31. Barcelona 08034 (Spain) /'\_ _/`\
|----------------------------------------------------------\___)=(___/
jorge.naranjo@Domain.Removed - www.bsc.es
(34)-934-054-229
|