This is the mail archive of the
fortran@gcc.gnu.org
mailing list for the GNU Fortran project.
Re: Recent polyhedron regressions.
- From: Paul Thomas <paulthomas2 at wanadoo dot fr>
- To: Richard Guenther <rguenther at suse dot de>
- Cc: fortran at gcc dot gnu dot org
- Date: Wed, 08 Mar 2006 18:01:37 +0100
- Subject: Re: Recent polyhedron regressions.
- References: <Pine.LNX.4.63.0603081026140.5325@t148.fhfr.qr>
Richard,
Many thanks for posting this message about these regressions and for
pinpointing the revisions that might be responsible.
They are, in fact, visible in the results posted to the list:
http://gcc.gnu.org/ml/fortran/2006-03/msg00056.html
http://gcc.gnu.org/ml/fortran/2006-03/msg00064.html
Changes that could have affected nf are revisions 111593:111631 (dot
product and some where changes).
I'll take a look see if I can figure out which it is.
Changes that could have affected channel are revisions 111702:111728,
* dependency.c (gfc_check_element_vs_element): Revert last change.
(no middle-end changes that can be accounted for the change).
http://www.suse.de/~rguenther/polyhedron/
Yes, as remarked in messages associated with the above thread, the
reason for this one is clear.
Roger Sayle is working on a replacement for the second patch, which will
claw back the regression.
Cheers
Paul