Remove LCSSA's bogus dependence on LoopSimplify and LoopSimplify's bogus
dependence on DominanceFrontier. Instead, add an explicit DominanceFrontier
pass in StandardPasses.h to ensure that it gets scheduled at the right
time.
Declare that loop unrolling preserves ScalarEvolution, and shuffle some
getAnalysisUsages.
This eliminates one LoopSimplify and one LCCSA run in the standard
compile opts sequence.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@109413 91177308-0d34-0410-b5e6-96231b3b80d8
diff --git a/lib/Transforms/Scalar/LoopDeletion.cpp b/lib/Transforms/Scalar/LoopDeletion.cpp
index 354b033..bd0e207 100644
--- a/lib/Transforms/Scalar/LoopDeletion.cpp
+++ b/lib/Transforms/Scalar/LoopDeletion.cpp
@@ -38,9 +38,9 @@
bool &Changed, BasicBlock *Preheader);
virtual void getAnalysisUsage(AnalysisUsage& AU) const {
- AU.addRequired<ScalarEvolution>();
AU.addRequired<DominatorTree>();
AU.addRequired<LoopInfo>();
+ AU.addRequired<ScalarEvolution>();
AU.addRequiredID(LoopSimplifyID);
AU.addRequiredID(LCSSAID);