[all-commits] [llvm/llvm-project] d921bf: [AMDGPU] Extend promotion of alloca to vectors (#1...

[Carl Ritson via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: d921bf233c4fd3840953b1a4b5bb35ad594da773
      https://github.com/llvm/llvm-project/commit/d921bf233c4fd3840953b1a4b5bb35ad594da773
  Author: Carl Ritson <carl.ritson at amd.com>
  Date:   2025-03-12 (Wed, 12 Mar 2025)

  Changed paths:
    M llvm/docs/AMDGPUUsage.rst
    M llvm/lib/Target/AMDGPU/AMDGPUPromoteAlloca.cpp
    M llvm/test/CodeGen/AMDGPU/amdgpu.private-memory.ll
    M llvm/test/CodeGen/AMDGPU/array-ptr-calc-i32.ll
    A llvm/test/CodeGen/AMDGPU/promote-alloca-max-regs.ll
    M llvm/test/CodeGen/AMDGPU/promote-alloca-memset.ll
    A llvm/test/CodeGen/AMDGPU/promote-alloca-multidim.ll
    M llvm/test/CodeGen/AMDGPU/promote-alloca-no-opts.ll
    M llvm/test/CodeGen/AMDGPU/promote-alloca-subvecs.ll
    A llvm/test/CodeGen/AMDGPU/promote-alloca-vgpr-ratio.ll
    M llvm/test/CodeGen/AMDGPU/vector-alloca-limits.ll

  Log Message:
  -----------
  [AMDGPU] Extend promotion of alloca to vectors (#127973)

* Add multi dimensional array support
* Make maximum vector size tunable
* Make ratio of VGPRs used for vector promotion tunable
* Maximum array size now based on VGPR count (32b) instead of element count



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