This is the mail archive of the
gcc-bugs@gcc.gnu.org
mailing list for the GCC project.
[Bug fortran/31067] MINLOC should sometimes be inlined (gas_dyn is sooooo sloooow)
- From: "burnus at gcc dot gnu dot org" <gcc-bugzilla at gcc dot gnu dot org>
- To: gcc-bugs at gcc dot gnu dot org
- Date: 27 Jul 2009 09:18:22 -0000
- Subject: [Bug fortran/31067] MINLOC should sometimes be inlined (gas_dyn is sooooo sloooow)
- References: <bug-31067-10259@http.gcc.gnu.org/bugzilla/>
- Reply-to: gcc-bugzilla at gcc dot gnu dot org
------- Comment #35 from burnus at gcc dot gnu dot org 2009-07-27 09:18 -------
(In reply to comment #34)
> Does ifort vectorize the exact same implemantion of minloc?
I tried to convert the minloc implementation into Fortran loops - and the
result is at
http://users.physik.fu-berlin.de/~tburnus/tmp/vect-PR31067/maxloc2.f90
$ ifort -O3 -xHost -diag-enable all maxloc2.f90
maxloc2.f90(25): (col. 5) remark: LOOP WAS VECTORIZED.
[timing: 0m1.384s]
$ gfortran -O3 -ffast-math -march=native -ftree-vectorize
-ftree-vectorizer-verbose=5 maxloc2.f90
maxloc2.f90:1: note: vectorized 0 loops in function.
[timing: 0m2.212s]
In case it helps: I put the ifort assembler output at:
http://users.physik.fu-berlin.de/~tburnus/tmp/vect-PR31067/maxloc2.s
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=31067