https://gcc.gnu.org/bugzilla/show_bug.cgi?id=104015
Kewen Lin <linkw at gcc dot gnu.org> changed: What |Removed |Added ---------------------------------------------------------------------------- Status|UNCONFIRMED |NEW Ever confirmed|0 |1 CC| |avieira at gcc dot gnu.org, | |linkw at gcc dot gnu.org Last reconfirmed| |2022-01-14 --- Comment #1 from Kewen Lin <linkw at gcc dot gnu.org> --- I think it's caused by r12-6240, since I only cherry-picked r12-6523 onto it, the issue can be reproduced. It's likely duplicated of PR103997. On Power9, we don't support partial vector so vectorizer should not try to use the same vector mode for the epilogue? gcc/testsuite/gcc.dg/vect/slp-perm-9.c:17:17: note: ***** Analysis succeeded with vector mode V8HI gcc/testsuite/gcc.dg/vect/slp-perm-9.c:17:17: note: ***** Choosing vector mode V8HI gcc/testsuite/gcc.dg/vect/slp-perm-9.c:17:17: note: ***** Re-trying epilogue analysis with vector mode V8HI