[llvm] [LoopVectorizer] Add support for partial reductions (PR #92418)

Sam Tebbs via llvm-commits llvm-commits at lists.llvm.org
Fri Dec 20 03:56:37 PST 2024


SamTebbs33 wrote:

I've figured it out. The reduction was scaled by 8, so during the execution of the reduction recipe with a VF of 8, it was divided by 8 to 1, causing the rest of the function to think the VF was scalar.

https://github.com/llvm/llvm-project/pull/92418


More information about the llvm-commits mailing list