[all-commits] [llvm/llvm-project] ffe86e: [AMDGPU] Update SIInsertHardClauses for GFX11

[Jay Foad via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: ffe86e3bdd77957245e13a48e341c38c511e9519
      https://github.com/llvm/llvm-project/commit/ffe86e3bdd77957245e13a48e341c38c511e9519
  Author: Jay Foad <jay.foad at amd.com>
  Date:   2022-06-09 (Thu, 09 Jun 2022)

  Changed paths:
    M llvm/lib/Target/AMDGPU/SIInsertHardClauses.cpp
    M llvm/test/CodeGen/AMDGPU/hard-clauses.mir

  Log Message:
  -----------
  [AMDGPU] Update SIInsertHardClauses for GFX11

Changes for GFX11:
- Clauses may not mix instructions of different types, and there are
  more types. For example image instructions with and without a sampler
  are now different types.
- The max size of a clause is explicitly documented as 63 instructions.
  Previously it was implicitly assumed to be 64. This is such a tiny
  difference that it does not seem worth making it conditional on the
  subtarget.
- It can be beneficial to clause stores as well as loads.

Differential Revision: https://reviews.llvm.org/D127391