[all-commits] [llvm/llvm-project] c8ed36: [AMDGPU] Cast sub-dword elements to i32 in concat_...

[Stanislav Mekhanoshin via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: c8ed36281a929c34477ffc391a8a3e84d1a07a92
      https://github.com/llvm/llvm-project/commit/c8ed36281a929c34477ffc391a8a3e84d1a07a92
  Author: Stanislav Mekhanoshin <Stanislav.Mekhanoshin at amd.com>
  Date:   2023-01-09 (Mon, 09 Jan 2023)

  Changed paths:
    M llvm/lib/Target/AMDGPU/AMDGPUISelLowering.cpp
    M llvm/test/CodeGen/AMDGPU/vector_shuffle.packed.ll

  Log Message:
  -----------
  [AMDGPU] Cast sub-dword elements to i32 in concat_vectors

This produces better code by avoiding repacking in some cases.

Fixes: SWDEV-373436

Differential Revision: https://reviews.llvm.org/D141329