Polyhedron benchmark on Opteron

Dominique Dhumieres dominiq@lps.ens.fr
Sat Sep 30 09:35:00 GMT 2006


May I ping the following post?

http://gcc.gnu.org/ml/fortran/2006-07/msg00095.html

The timing regression on the fatigue test is still there
on OSX(G5). As far as I can tell I don't see it on AMD.

My experience with the polyhedron tests is that they are
bad on OSX when using a lot of malloc (higher penalty than
under Linux).

Dominique



More information about the Fortran mailing list