[PATCH] D24216: AMDGPU: Partially fix control flow at -O0

[Matt Arsenault via llvm-commits]

arsenm created this revision.
arsenm added subscribers: llvm-commits, AMDGPU.
Herald added a reviewer: tstellarAMD.
Herald added subscribers: nhaehnle, wdng, arsenm, qcolombet, MatzeB.

Fixes to allow spilling all registers at the end of the block
work with exec modifications. Don't emit s_and_saveexec_b64 for
if lowering, and instead emit copies. Mark control flow mask
instructions as terminators to get correct spill code placement
with fast regalloc, and then have a separate optimization pass
form the saveexec.
    
This should work if SGPRs are spilled to VGPRs, but
will likely fail in the case that an SGPR spills to memory
and no workitem takes a divergent branch.

https://reviews.llvm.org/D24216

Files:
  lib/Target/AMDGPU/AMDGPU.h
  lib/Target/AMDGPU/AMDGPUTargetMachine.cpp
  lib/Target/AMDGPU/CMakeLists.txt
  lib/Target/AMDGPU/SIInstrInfo.cpp
  lib/Target/AMDGPU/SIInstructions.td
  lib/Target/AMDGPU/SILowerControlFlow.cpp
  lib/Target/AMDGPU/SIOptimizeExecMasking.cpp
  test/CodeGen/AMDGPU/control-flow-fastregalloc.ll
  test/CodeGen/MIR/AMDGPU/optimize-if-exec-masking.mir

-------------- next part --------------
A non-text attachment was scrubbed...
Name: D24216.70270.patch
Type: text/x-patch
Size: 54755 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20160903/d4b22bed/attachment.bin>