[all-commits] [llvm/llvm-project] 788f94: AMDGPU: Don't use unreachable on stores to unhandl...

[Matt Arsenault via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: 788f94f731dc9ba514925ed0c5936c5e79f6abd8
      https://github.com/llvm/llvm-project/commit/788f94f731dc9ba514925ed0c5936c5e79f6abd8
  Author: Matt Arsenault <Matthew.Arsenault at amd.com>
  Date:   2022-04-12 (Tue, 12 Apr 2022)

  Changed paths:
    M llvm/lib/Target/AMDGPU/SIISelLowering.cpp
    A llvm/test/CodeGen/AMDGPU/store-to-constant-error.ll

  Log Message:
  -----------
  AMDGPU: Don't use unreachable on stores to unhandled address space

For stores to constant address space, this will now consistently hit a
selection error instead of hitting unreachable in an asserts build.

I'm not sure what we should really do here. We could either just
codegen as if it were global, delete the instruction, or declare the
IR invalid (we really should have a target IR verifier to enforce it).