[llvm] AMDGPU: Mark the SCC def dead in the waterfall loop expansion (PR #226300)

[via llvm-commits]

llvmorg-github-actions[bot] wrote:


<!--LLVM PR SUMMARY COMMENT-->

@llvm/pr-subscribers-backend-amdgpu

Author: Matt Arsenault (arsenm)

<details>
<summary>Changes</summary>

The exec-mask manipulation in emitLoadScalarOpsFromVGPRLoop clobbers SCC,
but nothing in the generated loop reads it. Avoid depending on later fresh
computation of dead flags.

Co-Authored-By: Claude Opus 5 <noreply@<!-- -->anthropic.com>

---
Full diff: https://github.com/llvm/llvm-project/pull/226300.diff


24 Files Affected:

- (modified) llvm/lib/Target/AMDGPU/SIInstrInfo.cpp (+10-5) 
- (modified) llvm/test/CodeGen/AMDGPU/legalize-amdgcn.raw.buffer.load.format.f16.ll (+20-20) 
- (modified) llvm/test/CodeGen/AMDGPU/legalize-amdgcn.raw.buffer.load.format.ll (+24-24) 
- (modified) llvm/test/CodeGen/AMDGPU/legalize-amdgcn.raw.buffer.load.ll (+138-138) 
- (modified) llvm/test/CodeGen/AMDGPU/legalize-amdgcn.raw.buffer.store.format.f16.ll (+42-42) 
- (modified) llvm/test/CodeGen/AMDGPU/legalize-amdgcn.raw.buffer.store.format.f32.ll (+46-46) 
- (modified) llvm/test/CodeGen/AMDGPU/legalize-amdgcn.raw.buffer.store.ll (+130-130) 
- (modified) llvm/test/CodeGen/AMDGPU/legalize-amdgcn.raw.ptr.buffer.load.format.f16.ll (+20-20) 
- (modified) llvm/test/CodeGen/AMDGPU/legalize-amdgcn.raw.ptr.buffer.load.format.ll (+24-24) 
- (modified) llvm/test/CodeGen/AMDGPU/legalize-amdgcn.raw.ptr.buffer.load.ll (+177-177) 
- (modified) llvm/test/CodeGen/AMDGPU/legalize-amdgcn.raw.ptr.buffer.store.format.f16.ll (+42-42) 
- (modified) llvm/test/CodeGen/AMDGPU/legalize-amdgcn.raw.ptr.buffer.store.format.f32.ll (+46-46) 
- (modified) llvm/test/CodeGen/AMDGPU/legalize-amdgcn.raw.ptr.buffer.store.ll (+130-130) 
- (modified) llvm/test/CodeGen/AMDGPU/legalize-amdgcn.raw.ptr.tbuffer.load.f16.ll (+32-32) 
- (modified) llvm/test/CodeGen/AMDGPU/legalize-amdgcn.raw.ptr.tbuffer.load.ll (+36-36) 
- (modified) llvm/test/CodeGen/AMDGPU/legalize-amdgcn.raw.ptr.tbuffer.store.f16.ll (+40-40) 
- (modified) llvm/test/CodeGen/AMDGPU/legalize-amdgcn.raw.ptr.tbuffer.store.ll (+98-98) 
- (modified) llvm/test/CodeGen/AMDGPU/legalize-amdgcn.raw.tbuffer.load.f16.ll (+32-32) 
- (modified) llvm/test/CodeGen/AMDGPU/legalize-amdgcn.raw.tbuffer.load.ll (+36-36) 
- (modified) llvm/test/CodeGen/AMDGPU/legalize-amdgcn.raw.tbuffer.store.f16.ll (+40-40) 
- (modified) llvm/test/CodeGen/AMDGPU/legalize-amdgcn.raw.tbuffer.store.ll (+98-98) 
- (modified) llvm/test/CodeGen/AMDGPU/legalize-soffset-mbuf.ll (+64-64) 
- (modified) llvm/test/CodeGen/AMDGPU/move-to-valu-vimage-vsample.ll (+6-6) 
- (modified) llvm/test/CodeGen/AMDGPU/mubuf-legalize-operands.mir (+20-20) 


``````````diff
The server is unavailable at this time. Please wait a few minutes before you try again.
``````````

</details>


https://github.com/llvm/llvm-project/pull/226300