[llvm] [SLP]Initial support for non-power-of-2 vectorization (PR #194189)

via llvm-commits llvm-commits at lists.llvm.org
Wed May 13 05:06:47 PDT 2026

Previous message: [llvm] [AArch64][SelectionDAG] Fold late FCMP+CSEL patterns to FMAXNM/FMINNM during AArch64 DAG-to-DAG preprocessing. (PR #197429)
Next message: [llvm] [AArch64] Select `CMP WZR, reg, shift #amount` (PR #197424)
Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]

asmok-g wrote:

Note: We saw some regressions ~2-3% in [Blur_benchmark](https://github.com/llvm/llvm-test-suite/tree/main/MicroBenchmarks/ImageProcessing/Blur) because of this commit. I don't think it is serious, but thought I should notify.

https://github.com/llvm/llvm-project/pull/194189

Previous message: [llvm] [AArch64][SelectionDAG] Fold late FCMP+CSEL patterns to FMAXNM/FMINNM during AArch64 DAG-to-DAG preprocessing. (PR #197429)
Next message: [llvm] [AArch64] Select `CMP WZR, reg, shift #amount` (PR #197424)
Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]

More information about the llvm-commits mailing list