[all-commits] [llvm/llvm-project] 6ce86a: AMDGPU: Ensure flat loads are broken into dword in...

[Matt Arsenault via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: 6ce86a7effea8e516a0f242aa06ad73e4557150d
      https://github.com/llvm/llvm-project/commit/6ce86a7effea8e516a0f242aa06ad73e4557150d
  Author: Matt Arsenault <Matthew.Arsenault at amd.com>
  Date:   2023-02-05 (Sun, 05 Feb 2023)

  Changed paths:
    M llvm/lib/Target/AMDGPU/AMDGPULegalizerInfo.cpp
    M llvm/lib/Target/AMDGPU/SIISelLowering.cpp
    M llvm/test/CodeGen/AMDGPU/GlobalISel/legalize-load-flat.mir
    M llvm/test/CodeGen/AMDGPU/GlobalISel/llvm.amdgcn.atomic.dec.ll
    M llvm/test/CodeGen/AMDGPU/GlobalISel/llvm.amdgcn.atomic.inc.ll
    M llvm/test/CodeGen/AMDGPU/flat-address-space.ll

  Log Message:
  -----------
  AMDGPU: Ensure flat loads are broken into dword in functions

We were assuming we could rely on the flat scratch init detection
to imply if there are possible flat addressed stack objects, which
doesn't work outside of a kernel. We should have a way to prove
if a given flat access can't access the stack.

We could use a not-stack parameter attribute to avoid
these splits.

Make the minimally correct change for GlobalISel; I'll address
this better in my larger patch to rewrite load and store legalization.

Fixes: SWDEV-218237