[all-commits] [llvm/llvm-project] 145b4a: [AMDGPU][CodeGenPrepare] Narrow 64 bit math to 32 ...

[Shoreshen via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: 145b4a39504b88a695f1f85f4d9da991bb9a2656
      https://github.com/llvm/llvm-project/commit/145b4a39504b88a695f1f85f4d9da991bb9a2656
  Author: Shoreshen <372660931 at qq.com>
  Date:   2025-04-01 (Tue, 01 Apr 2025)

  Changed paths:
    M llvm/lib/Target/AMDGPU/AMDGPUCodeGenPrepare.cpp
    M llvm/test/CodeGen/AMDGPU/amdgpu-codegenprepare-mul24.ll
    M llvm/test/CodeGen/AMDGPU/atomic_optimizations_global_pointer.ll
    A llvm/test/CodeGen/AMDGPU/narrow_math_for_and.ll
    M llvm/test/CodeGen/AMDGPU/widen-smrd-loads.ll

  Log Message:
  -----------
  [AMDGPU][CodeGenPrepare] Narrow 64 bit math to 32 bit if profitable (#130577)

For Add, Sub, Mul with Int64 type, if profitable, then do:
1. Trunc operands to Int32 type
2. Apply 32 bit Add/Sub/Mul
3. Zext to Int64 type



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