[PATCH] D141355: [AMDGPUUnifyDivergentExitNodes] Add NewPM support

[Anshil Gandhi via Phabricator via llvm-commits]

gandhi21299 added inline comments.


================
Comment at: llvm/lib/Target/AMDGPU/AMDGPUUnifyDivergentExitNodes.cpp:162
 
-bool AMDGPUUnifyDivergentExitNodes::runOnFunction(Function &F) {
-  DominatorTree *DT = nullptr;
-  if (RequireAndPreserveDomTree)
-    DT = &getAnalysis<DominatorTreeWrapperPass>().getDomTree();
-
-  auto &PDT = getAnalysis<PostDominatorTreeWrapperPass>().getPostDomTree();
-  if (PDT.root_size() == 0 ||
-      (PDT.root_size() == 1 &&
-       !isa<BranchInst>(PDT.getRoot()->getTerminator())))
+bool AMDGPUUnifyDivergentExitNodesImpl::run(Function &F, DominatorTree *DT,
+                                            PostDominatorTree *PDT,
----------------
sameerds wrote:
> Either declare parameters as references, or implement checks for nullptr in the function body. Also, can any of these be const?
`DominatorTree` may be updated but `PostDominatorTree` and `UniformityInfo` are preserved.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D141355/new/

https://reviews.llvm.org/D141355