[all-commits] [llvm/llvm-project] 0c9c92: [X86][XOP] Tidyup VPHADD/VPHSUB unary horizontal o...

[Simon Pilgrim via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: 0c9c92ffc0acde32376b7ba8dcbe7fec90de3ff3
      https://github.com/llvm/llvm-project/commit/0c9c92ffc0acde32376b7ba8dcbe7fec90de3ff3
  Author: Simon Pilgrim <llvm-dev at redking.me.uk>
  Date:   2022-03-03 (Thu, 03 Mar 2022)

  Changed paths:
    M llvm/lib/Target/X86/X86InstrXOP.td
    M llvm/test/tools/llvm-mca/X86/Generic/resources-xop.s

  Log Message:
  -----------
  [X86][XOP] Tidyup VPHADD/VPHSUB unary horizontal ops default schedule class

Based off Agner and AMD SoG tables, the XOP VPHADD/VPHSUB unary horizontal ops are as fast as basic arithmetic ops, not the slower SSSE3 binary horizontal add/sub ops. This also matches what the bdver2 model already lists.

Noticed while investigating reduction add optimizations.