[AMDGPUUnifyDivergentExitNodes] Add NewPM support
authorAnshil Gandhi <Anshil.Gandhi@amd.com>
Sat, 25 Mar 2023 19:42:31 +0000 (13:42 -0600)
committerAnshil Gandhi <gandhi21299@gmail.com>
Sat, 25 Mar 2023 20:04:36 +0000 (14:04 -0600)
commitb48e7c2d01a329977da32ef7d5f0feba95d201f7
tree7841cc39fc26b0ff41d1459379b824361c0913e8
parent731264b0c2af7aa46bd39625202a99e06cfccff9
[AMDGPUUnifyDivergentExitNodes] Add NewPM support

Meanwhile, use UniformityAnalysis instead of LegacyDivergenceAnalysis to collect divergence info.

Reviewed By: arsenm, sameerds

Differential Revision: https://reviews.llvm.org/D141355
llvm/lib/Target/AMDGPU/AMDGPUTargetMachine.cpp
llvm/lib/Target/AMDGPU/AMDGPUUnifyDivergentExitNodes.cpp
llvm/lib/Target/AMDGPU/AMDGPUUnifyDivergentExitNodes.h [new file with mode: 0644]
llvm/test/CodeGen/AMDGPU/si-annotate-nested-control-flows.ll