[all-commits] [llvm/llvm-project] 36fc29: [VPlan] Implement VPBlendRecipe::computeCost.

[Florian Hahn via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: 36fc291b6ec6d4c8e3d956d6855631ab2db7b05a
      https://github.com/llvm/llvm-project/commit/36fc291b6ec6d4c8e3d956d6855631ab2db7b05a
  Author: Florian Hahn <flo at fhahn.com>
  Date:   2024-10-08 (Tue, 08 Oct 2024)

  Changed paths:
    M llvm/lib/Transforms/Vectorize/LoopVectorize.cpp
    M llvm/lib/Transforms/Vectorize/VPlan.h
    M llvm/lib/Transforms/Vectorize/VPlanRecipes.cpp
    A llvm/test/Transforms/LoopVectorize/AArch64/blend-costs.ll
    M llvm/test/Transforms/LoopVectorize/X86/replicate-uniform-call.ll

  Log Message:
  -----------
  [VPlan] Implement VPBlendRecipe::computeCost.

Implement VPBlendRecipe::computeCost. VPBlendRecipe is currently is also
used if only the first lane is used.

This also requires pre-computing costs for forced scalars and
instructions considered profitable to scalarize. For those, the cost
will be computed separately in the legacy cost model. This will also be
needed when implementing VPReplicateRecipe::computeCost.



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