[all-commits] [llvm/llvm-project] 953c13: [AMDGPU][PromoteAlloca] Whole-function alloca prom...

[Pierre van Houtryve via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: 953c13b5c90bed1e24fe95e90137c4e226ac2d09
      https://github.com/llvm/llvm-project/commit/953c13b5c90bed1e24fe95e90137c4e226ac2d09
  Author: Pierre van Houtryve <pierre.vanhoutryve at amd.com>
  Date:   2024-03-19 (Tue, 19 Mar 2024)

  Changed paths:
    M llvm/lib/Target/AMDGPU/AMDGPUPromoteAlloca.cpp
    M llvm/test/CodeGen/AMDGPU/llc-pipeline.ll
    A llvm/test/CodeGen/AMDGPU/promote-alloca-scoring.ll

  Log Message:
  -----------
  [AMDGPU][PromoteAlloca] Whole-function alloca promotion to vector (#84735)

Update PromoteAllocaToVector so it considers the whole function before promoting allocas.
Allocas are scored & sorted so the highest value ones are seen first. The budget is now per function instead of per alloca.

Passed internal performance testing.



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