[all-commits] [llvm/llvm-project] 2fb4bf: [AMDGPU][True16] Fix ISel for A16 Image Instructions

[Joe Nash via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: 2fb4bfa5ba972bf074bea8a89590171979055a41
      https://github.com/llvm/llvm-project/commit/2fb4bfa5ba972bf074bea8a89590171979055a41
  Author: Joe Nash <Joseph.Nash at amd.com>
  Date:   2023-08-11 (Fri, 11 Aug 2023)

  Changed paths:
    M llvm/lib/Target/AMDGPU/AMDGPULegalizerInfo.cpp
    M llvm/test/CodeGen/AMDGPU/GlobalISel/legalize-llvm.amdgcn.image.atomic.dim.a16.ll
    M llvm/test/CodeGen/AMDGPU/GlobalISel/legalize-llvm.amdgcn.image.dim.a16.ll
    M llvm/test/CodeGen/AMDGPU/GlobalISel/legalize-llvm.amdgcn.image.sample.a16.ll

  Log Message:
  -----------
  [AMDGPU][True16] Fix ISel for A16 Image Instructions

The 16-bit VAddr arguments to A16 image instructions are packed into
legal VGPR_32 operands in AMDGPULegalizerInfo::legalizeImageIntrinsic on
all subtargets. With True16, we also need to pack if the number of VAddr is one
because VGPR_16 is not a legal argument to those Image instructions.

No change to emitted code intended on subtargets pre-GFX11, and none on GFX11
until True16 is active.

Reviewed By: foad

Differential Revision: https://reviews.llvm.org/D157426