[Bug fortran/82471] Reorder loop for unfavorable index ordering in DO CONCURRENT and FORALL

[tkoenig at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=82471

Thomas Koenig <tkoenig at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |ASSIGNED
         Depends on|                            |14741
           Assignee|unassigned at gcc dot gnu.org      |tkoenig at gcc dot gnu.org

--- Comment #9 from Thomas Koenig <tkoenig at gcc dot gnu.org> ---
I'm trying something.

More generally, I think I should try for some very simple
loop interchange so that we at least get the simple
case of matrix multiplication right. PR 14741 has been around
for 13 years, and I see no indication that it is going to be
fixed the "right" way, via the middle end, so we might as well
do something in the Fortran front end.


Referenced Bugs:

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=14741
[Bug 14741] graphite with loop blocking and interchanging doesn't optimize a
matrix multiplication loop