This is the mail archive of the
gcc@gcc.gnu.org
mailing list for the GCC project.
Re: Little C-benchmarks of gcc-2.95.4-ss,gcc-3.3.4-ss,gcc-3.4.1-ss & gcc-3.5.0-ss.
- From: Steven Bosscher <stevenb at suse dot de>
- To: jc-nospam at jr-pizarro dot jazztel dot es
- Cc: gcc at gcc dot gnu dot org
- Date: Tue, 1 Jun 2004 18:22:12 +0200
- Subject: Re: Little C-benchmarks of gcc-2.95.4-ss,gcc-3.3.4-ss,gcc-3.4.1-ss & gcc-3.5.0-ss.
- Organization: SUSE Labs
- References: <1086105219.40bca683e5503@webmail.jazznet.es> <m33c5fs1j6.fsf@gromit.moeb>
On Tuesday 01 June 2004 18:09, Andreas Jaeger wrote:
> jc-nospam@jr-pizarro.jazztel.es writes:
> > My favorite architecture for GCC is i486 because it is simple pipeline
> >? non-superscalar with cache's instructions and fairly good for the
> >? combination of GCC's optimizer and the complex microprocessor AthlonXP.
Of course your benchmarks are completely meaningless if you compile
for a i486 and run it on an Athlon.
> > * CFLAGS_only_for_gcc35="$CFLAGS -ftree-ccp -ftree-pre -ftree-dce
> > -ftree-dse \ -ftree-lrs -ftree-combine-temps -ftree-ter -ftree-copyrename
> > \
> > -ftree-dominator-opts -ftree-ch -ftree-sra"
>
> Aren't those enabled by default?
At -O2 they all are enabled by default.
> > For HPC (High Performance Computing), between gcc-2.95.4, gcc-3.3.3,
> > gcc-3.4.0 and gcc-3.5.0-ss, i recommend gcc-3.4.0 as the
> > default stable compiler because gcc-3.5.0-ss is very big, unstable and
> > slow due to many SSA's optimizations that it doesn't match very good
> > results.
To the best of (at least) my knowledge, the SSA optimizations are
not slow. You can see what component of the compiler takes too
much compile time by looking at the output of -ftime-report (and
file bugs when an SSA optimizer eats up more than, say, 10% of the
total compile time).
Did you build the 3.5 snapshot with --disable-checking?
Gr.
Steven