[all-commits] [llvm/llvm-project] 687e0e: AMDGPU: Create alloca wide load/store with explici...

[Matt Arsenault via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: 687e0e205e5e95ff157aa5977164c155fb424f23
      https://github.com/llvm/llvm-project/commit/687e0e205e5e95ff157aa5977164c155fb424f23
  Author: Matt Arsenault <Matthew.Arsenault at amd.com>
  Date:   2023-01-03 (Tue, 03 Jan 2023)

  Changed paths:
    M llvm/lib/Target/AMDGPU/AMDGPUPromoteAlloca.cpp
    M llvm/test/CodeGen/AMDGPU/promote-alloca-array-aggregate.ll
    M llvm/test/CodeGen/AMDGPU/promote-alloca-pointer-array.ll

  Log Message:
  -----------
  AMDGPU: Create alloca wide load/store with explicit alignment

This was introducing transient UB by using the default alignment of a
larger vector type.