https://bugs.llvm.org/show_bug.cgi?id=45823
Mikael Holmén <mikael.hol...@ericsson.com> changed:
What |Removed |Added
----------------------------------------------------------------------------
Resolution|--- |FIXED
Status|NEW |RESOLVED
--- Comment #2 from Mikael Holmén <mikael.hol...@ericsson.com> ---
It seems that with commit aa00b1d76364:
[LV] Try to sink users recursively for first-order recurrences.
Update isFirstOrderRecurrence to explore all uses of a recurrence phi
and check if we can sink them. If there are multiple users to sink, they
are all mapped to the previous instruction.
Fixes PR44286 (and another PR or two).
Reviewed By: Ayal
Differential Revision: https://reviews.llvm.org/D84951
the vectorizer doesn't bail out on the loop anymore.
Now we instead get
LV: Vectorization is possible but not beneficial.
and if adding e.g.
-force-vector-width=2
the loop indeed gets vectorized.
Thanks!
--
You are receiving this mail because:
You are on the CC list for the bug._______________________________________________
llvm-bugs mailing list
llvm-bugs@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-bugs
