This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: GCC-4.5.0 comparison with previous releases and LLVM-2.7 on SPEC2000 for x86/x86_64
- From: Jan Hubicka <hubicka at ucw dot cz>
- To: Xinliang David Li <davidxl at google dot com>, gcc-patches at gcc dot gnu dot org
- Date: Thu, 29 Apr 2010 23:20:44 +0200
- Subject: Re: GCC-4.5.0 comparison with previous releases and LLVM-2.7 on SPEC2000 for x86/x86_64
- References: <4BD9B2EB.9060207@redhat.com> <w2v522e93241004291117n7eb214ah55245e0209f3297e@mail.gmail.com>
> This matches our previous observation that to bring the best out of
> LTO, FDO is also needed. (As a reference, LIPO improves over plain FDO
> by ~4.5%, vortex improves 23%). You will probably see even smaller
> improvement in SPEC2006.
There is FDO tester at http://gcc.opensuse.org/SPEC/CINT/sb-frescobaldi.suse.de-fdo-64-FDO/index.html
I switched it to LTO+FDO some time ago. For some reason vortex is not passing there, I will investigate (tehre is some vortex aliasing fix needed that probably
was not applied to that machine).
Yes FDO+LTO combine well. Still SPEC2000 is not the best suite for this,
since it contains only benchmarks that was optimized for LTO less compilers.
More modern benchmarks are more fun.
Honza
>
> It would be great if there is number collected comparing LTO + FDO vs
> plain FDO in the same setup.
>
> Thanks,
>
> David
>
>
>
>
> >
> > ?Currently Graphite gives small improvements on x86 (one exception is
> > 2% for peak x86 SPECFP2000) and mostly degradation on x86_64 (with
> > maximum one more than 10% for SPECFP2000 because of big degradations
> > on mgrid and swim). ?So further work is needed on the project because
> > it seems not mature yet.
> >
> > ?As for LLVM, LLVM became slower (e.g. in comparison with llvm-2.5 on
> > 15%-50% for x86-64). ?So the gap between compilation speed of GCC and
> > LLVM decreased and sometimes achieves 4% on x86_64 and 8% on x86 (both
> > for SPECInt2000 in -O2 mode). ?May be I am wrong but I don't think
> > CLANG will improve this situation significantly (in -O2 and -O3 mode)
> > because optimizations still take most of time of any serious
> > optimizing compiler.
> >
> > ?LLVM did a progress in code performance especially for floating
> > point benchmarks. ?But the gap between LLVM-2.7 and GCC-4.5 in peak
> > performance (not including GCC LTO and Graphite) still 6-7% on
> > SPECInt200 and 13-17% on SPECFP2000.
> >
> > ?In general, IMHO GCC-4.5.0 is a good and promising release.
> >
> >