[all-commits] [llvm/llvm-project] 8bc0f8: [AMDGPU][True16][CodeGen] D16 LDS load/store pseud...

[Brox Chen via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: 8bc0f879a05228c58235ded510360da2220f0afd
      https://github.com/llvm/llvm-project/commit/8bc0f879a05228c58235ded510360da2220f0afd
  Author: Brox Chen <guochen2 at amd.com>
  Date:   2025-03-17 (Mon, 17 Mar 2025)

  Changed paths:
    M llvm/lib/Target/AMDGPU/DSInstructions.td
    M llvm/test/CodeGen/AMDGPU/atomic_load_local.ll
    M llvm/test/CodeGen/AMDGPU/atomic_store_local.ll
    M llvm/test/CodeGen/AMDGPU/chain-hi-to-lo.ll
    M llvm/test/CodeGen/AMDGPU/ds-sub-offset.ll
    M llvm/test/CodeGen/AMDGPU/integer-mad-patterns.ll

  Log Message:
  -----------
  [AMDGPU][True16][CodeGen] D16 LDS load/store pseudo instructions in true16 (#131427)

Implement new pseudos with the suffix _t16 which have VGPR_16 as the
store src or load dst. This affects LDS 8 and 16-bit loads and stores.
Lower the pseudos to the existing real Hi/Lo instructions in MC inst
layer with VGPR_32 src or dst

---------

Co-authored-by: Abhinav <abhinav.garg at amd.com>



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