[all-commits] [llvm/llvm-project] f23246: [LV] Directly add fast-math flags to select recipe...

[Florian Hahn via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: f23246a0bb403c2a6f206f6e190496f45fad4ef9
      https://github.com/llvm/llvm-project/commit/f23246a0bb403c2a6f206f6e190496f45fad4ef9
  Author: Florian Hahn <flo at fhahn.com>
  Date:   2023-09-21 (Thu, 21 Sep 2023)

  Changed paths:
    M llvm/lib/Transforms/Vectorize/LoopVectorize.cpp
    M llvm/lib/Transforms/Vectorize/VPlanRecipes.cpp

  Log Message:
  -----------
  [LV] Directly add fast-math flags to select recipe (NFC).

Now that VPInstruction can manage fast math flags via
VPRecipeWithIRFlags, use them directly to model the fast-math flags of
the select created for the final reduction value instead of adding them
late.