[all-commits] [llvm/llvm-project] d7dd31: [SLP]Better analysis of the repeated instructions ...

[Alexey Bataev via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: d7dd31e41791d71ad81af9cc4e7a26b26d4cb27e
      https://github.com/llvm/llvm-project/commit/d7dd31e41791d71ad81af9cc4e7a26b26d4cb27e
  Author: Alexey Bataev <a.bataev at outlook.com>
  Date:   2024-09-24 (Tue, 24 Sep 2024)

  Changed paths:
    M llvm/lib/Transforms/Vectorize/SLPVectorizer.cpp
    M llvm/test/Transforms/SLPVectorizer/X86/splat-score-adjustment.ll

  Log Message:
  -----------
  [SLP]Better analysis of the repeated instructions during operands reordering

When doing the repeated instructions analysis, better to make the
reordering non-profitable, if the number of unique instructions is not
power-of-2. In this case better to keep power-of-2 elements as this
allows better vectorization.

Fixes https://github.com/llvm/llvm-project/issues/109725



To unsubscribe from these emails, change your notification settings at https://github.com/llvm/llvm-project/settings/notifications