[all-commits] [llvm/llvm-project] a305d8: [X86] Adjust fsetcc/fmin/fmax costs to match SoG (...

[Simon Pilgrim via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: a305d8f44ec8e91739c1f1e714cbfd19acc80489
      https://github.com/llvm/llvm-project/commit/a305d8f44ec8e91739c1f1e714cbfd19acc80489
  Author: Simon Pilgrim <llvm-dev at redking.me.uk>
  Date:   2022-04-14 (Thu, 14 Apr 2022)

  Changed paths:
    M llvm/lib/Target/X86/X86ScheduleZnver1.td
    M llvm/lib/Target/X86/X86ScheduleZnver2.td
    M llvm/test/tools/llvm-mca/X86/Znver1/resources-avx1.s
    M llvm/test/tools/llvm-mca/X86/Znver1/resources-sse1.s
    M llvm/test/tools/llvm-mca/X86/Znver1/resources-sse2.s
    M llvm/test/tools/llvm-mca/X86/Znver2/resources-avx1.s
    M llvm/test/tools/llvm-mca/X86/Znver2/resources-sse1.s
    M llvm/test/tools/llvm-mca/X86/Znver2/resources-sse2.s

  Log Message:
  -----------
  [X86] Adjust fsetcc/fmin/fmax costs to match SoG (Issue #54889)

znver1/2 models were incorrectly modelling these as 3 cycle latency instructions on the wrong pipe and znver1 ymm variants also require double pumping.

Now matches AMD SoG, Agner and instlatx64 numbers.

Thanks to @fabian-r for the report