This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: [graphite] Fix for pr38492 and pr38498
On Tue, Jan 06, 2009 at 10:42:36PM -0600, Sebastian Pop wrote:
> On Tue, Jan 6, 2009 at 9:03 PM, Jack Howarth <howarth@bromo.med.uc.edu> wrote:
> > Jan,
> > FYI, this patch also eliminates the remaining wrong result failures in the
> > capacita and protein polyhedron 2005 benchmarks. With this patch applied to gcc
> > trunk on i686-apple-darwin9, I am now able to build the polyhedron benchmarks with
> > "-ffast-math -funroll-loops -msse3 -O3 -fgraphite-identity" and the entire benchmark
> > builds and passes. Nice.
>
> Thanks Jack for reporting these results: it's good to hear this.
> Indeed Jan did a good job here. I will also try a bootstrap with
> BOOT_CFLAGS="-O2 -fgraphite-identity" to see if it passes now.
>
> Thanks again,
> Sebastian
Sebastian,
Interestingly, if I add -floop-block to "-ffast-math -funroll-loops -msse3 -O3 -fgraphite-identity",
I find that the capacita benchmark fails again with the same NaN errors as before Jan's patch. Is it
possible that Jan's patch needs to extended to handle -floop-block?
Jack