Re: [Freepooma-devel] [PATCH] PathScale EKOPath compiler support

[Richard Guenther]

Bryan O'Sullivan wrote:
> On Thu, 2005-03-17 at 09:33 +0100, Richard Guenther wrote:
>
>
> > Well in all the shipped benchmarks you'll get output like
> > (BlitzLoops):
>
>
> > where the ultimate goal is to have CppTran and PoomaII numbers
> > be equal or better (higher - these are sort of MFLOPs) than
> > the C and C restrict numbers.  Here we assume that the compilers
> > are already very good at optimizing the C code, which is usually
> > true.
>
>
> Thanks for the descriptions.  I'll take a look and see what numbers we
> get.  What kind of hardware did you get those numbers on, by the way?
> We have a variety of stuff here, so I can try to reproduce your numbers.

The numbers were on a 2.8GHz Pentium4, but really only the comparison
between the C and POOMA numbers matter.

Richard.