On Tue, Dec 13, 2011 at 13:16, Bogdan Dita <bogdan at lmn.pub.ro> wrote:
> So I reconfigured PETSc as advised, the result was that the time obtained > from the new version of PETSc Optimised version(with COPTFLAGS) was about > the same as the time from the Debug version(+/- 2 sec). > You have to send output from -log_summary in both cases if you want us to help explain observed performance. > Although is not > what I expected this is still an intriguing result for me: on one hand > there was definitely something about the flags that the optimised PETSc > didn't like, exactly how Mr. J. Brown predicted, but on the other hand > this raises new questions. > > This is the output from 'mpicc -show'. > for the cluster: > gcc -m64 -O2 -fPIC -Wl,-z,noexecstack -I/usr/include/mpich2-x86_64 > -L/usr/lib64/mpich2/lib -lmpich > This one is optimized by default. > > and from my laptop: > gcc -I/usr/lib64/mpich2/include -L/usr/lib64/mpich2/lib > -Wl,-rpath,/usr/lib64/mpich2/lib -lmpich -lopa -lmpl -lrt -lpthread > This one is not. PETSc folks, should we fight MPICH2 by explicitly adding -O0 for debug builds in which CFLAGS are not passed by the user? -------------- next part -------------- An HTML attachment was scrubbed... URL: <http://lists.mcs.anl.gov/pipermail/petsc-users/attachments/20111213/6eb15ebc/attachment.htm>