[all-commits] [llvm/llvm-project] c22297: AMDGPU/GlobalISel: Stop using NarrowScalar/FewerEl...

[Matt Arsenault via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: c22297244209c6bea27fe54cd5eca15d689cf8ad
      https://github.com/llvm/llvm-project/commit/c22297244209c6bea27fe54cd5eca15d689cf8ad
  Author: Matt Arsenault <Matthew.Arsenault at amd.com>
  Date:   2021-12-20 (Mon, 20 Dec 2021)

  Changed paths:
    M llvm/lib/Target/AMDGPU/AMDGPULegalizerInfo.cpp
    M llvm/test/CodeGen/AMDGPU/GlobalISel/cvt_f32_ubyte.ll
    M llvm/test/CodeGen/AMDGPU/GlobalISel/legalize-load-constant-32bit.mir
    M llvm/test/CodeGen/AMDGPU/GlobalISel/legalize-load-constant.mir
    M llvm/test/CodeGen/AMDGPU/GlobalISel/legalize-load-flat.mir
    M llvm/test/CodeGen/AMDGPU/GlobalISel/legalize-load-global.mir
    M llvm/test/CodeGen/AMDGPU/GlobalISel/legalize-load-local.mir
    M llvm/test/CodeGen/AMDGPU/GlobalISel/legalize-load-private.mir
    M llvm/test/CodeGen/AMDGPU/GlobalISel/legalize-store-global.mir
    M llvm/test/CodeGen/AMDGPU/GlobalISel/legalize-store.mir
    M llvm/test/CodeGen/AMDGPU/GlobalISel/llvm.memcpy.ll
    M llvm/test/CodeGen/AMDGPU/GlobalISel/load-constant.96.ll
    M llvm/test/CodeGen/AMDGPU/GlobalISel/load-local.128.ll
    M llvm/test/CodeGen/AMDGPU/GlobalISel/load-local.96.ll
    M llvm/test/CodeGen/AMDGPU/GlobalISel/load-unaligned.ll
    M llvm/test/CodeGen/AMDGPU/GlobalISel/store-local.128.ll
    M llvm/test/CodeGen/AMDGPU/GlobalISel/store-local.96.ll
    M llvm/test/CodeGen/AMDGPU/GlobalISel/widen-i8-i16-scalar-loads.ll
    M llvm/test/CodeGen/AMDGPU/cttz_zero_undef.ll
    M llvm/test/CodeGen/AMDGPU/ds-alignment.ll

  Log Message:
  -----------
  AMDGPU/GlobalISel: Stop using NarrowScalar/FewerElements for unaligned splitting

These actions should only be used for adjusting the register types
(and the memory type as needed to satisfy the register
type). Unaligned accesses should be split as a type of lowering.

This has the effect of improving the code in many cases since now we
produce zextloads instead of separate loads with ands. The load/store
legality rules still seem far more complicated than necessary though.