[all-commits] [llvm/llvm-project] af2920: [TTI][X86] getArithmeticInstrCost - move opcode ca...

[Simon Pilgrim via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: af2920ec6f859e9b7e0ef519cdc601ee2567c379
      https://github.com/llvm/llvm-project/commit/af2920ec6f859e9b7e0ef519cdc601ee2567c379
  Author: Simon Pilgrim <llvm-dev at redking.me.uk>
  Date:   2021-08-30 (Mon, 30 Aug 2021)

  Changed paths:
    M llvm/lib/Target/X86/X86TargetTransformInfo.cpp

  Log Message:
  -----------
  [TTI][X86] getArithmeticInstrCost - move opcode canonicalization before all target-specific costs. NFCI.

The GLM/SLM special cases still get tested first but after the the MUL/DIV/REM pattern detection - this will be necessary for when we make the SLM vXi32 MUL canonicalization generic to improve PMULLW/PMULHW/PMADDDW cost support etc.