Polyhedron benchmark results on AMD64

[Steven Bosscher]

On Friday 09 February 2007 18:04, Dorit Nuzman wrote:
> > Contrary to you I used -ftree-vectorize which seems to make things
> > faster, but breaks two g95 tests.
>
> which tests? is there a PR open for this?

You probably have already fixed those PRs somewhen.  G95 uses the already
ancient GCC 4.0 as its back end.  With gfortran, there are no vectorizer
failures, so the bugs that trigger in the gcc 4.0 based compiler are most
likely fixed already in GCC 4.3 (which is, of course, what gfortran is
based on ;-))

Gr.
Steven