https://bugs.llvm.org/show_bug.cgi?id=43578
Sanjay Patel <spatel+l...@rotateright.com> changed:
What |Removed |Added
----------------------------------------------------------------------------
Fixed By Commit(s)| |r374183
Status|NEW |RESOLVED
Resolution|--- |FIXED
--- Comment #2 from Sanjay Patel <spatel+l...@rotateright.com> ---
I didn't step through the full SPEC example to see what happens, but changing
SLP to use smaller groups of GEPs seems to have restored the perf by preventing
the vectorization entirely in SetupFastFullPelSearch():
https://reviews.llvm.org/rL374183
--
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
