This is the mail archive of the
fortran@gcc.gnu.org
mailing list for the GNU Fortran project.
matmul problem, gcc-4.2 snapshot 060617
- From: Tim Prince <timothyprince at sbcglobal dot net>
- To: gfortran <fortran at gcc dot gnu dot org>
- Date: Tue, 20 Jun 2006 06:19:43 -0700
- Subject: matmul problem, gcc-4.2 snapshot 060617
- Reply-to: tprince at myrealbox dot com
In my cygwin build of this snapshot, something has gone wrong in matmul,
for a case where the first argument is a rank 1 array of stride 5. The
stride is seen correctly in matmul, but it appears to choose an inner
loop built for stride 1, and the time taken is about 5 times too long.
I guess I'll have to download an older source and see what changed. I
believe it was still working a week earlier.
I have been having a great deal of problems just getting snapshots to
build for various targets lately, so I can't check this one on linux.