This is the mail archive of the
gcc@gcc.gnu.org
mailing list for the GCC project.
Re: graphite legacy-cloog/cloog-ppl benchmarks
- From: Jack Howarth <howarth at bromo dot med dot uc dot edu>
- To: gcc at gcc dot gnu dot org
- Date: Fri, 12 Nov 2010 22:29:07 -0500
- Subject: Re: graphite legacy-cloog/cloog-ppl benchmarks
A quick check with...
gfortran -ffast-math -funroll-loops -msse3 -O3 -fgraphite-identity -ftree-vectorizer-verbose=2 air.f90 -o air >& air_cloog_legacy.txt
grep "LOOP VECTORIZED" air_cloog_legacy.txt | wc -l
5
gfortran -ffast-math -funroll-loops -msse3 -O3 -fgraphite-identity -ftree-vectorizer-verbose=2 air.f90 -o air >& air_cloog_isl.txt
grep "LOOP VECTORIZED" air_cloog_isl.txt | wc -l
5
...for each build suggests that the 12% runtime performance penalty when using cloog-isl isn't due to missed vectorizations.
Jack