Fix PR404: Loop simplify is really slow on 252.eon

This eliminates an N*N*logN algorithm from the loop simplify pass, replacing
it with a much simpler and faster alternative.  In a debug build, this reduces
gccas time on eon from 85s to 42s.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@14851 91177308-0d34-0410-b5e6-96231b3b80d8

diff --git a/lib/Transforms/Utils/LoopSimplify.cpp b/lib/Transforms/Utils/LoopSimplify.cpp
index b2752089d44614531eb019186060aa1b26e6865b..9f8373e5450e0dd8e82948a563dbec368a00f243 100644 (file)
--- a/lib/Transforms/Utils/LoopSimplify.cpp
+++ b/lib/Transforms/Utils/LoopSimplify.cpp
@@ -147,6 +147,9 @@ bool LoopSimplify::ProcessLoop(Loop *L) {
     Changed = true;
   }
 
+
+  DominatorSet &DS = getAnalysis<DominatorSet>();  // Update dominator info
+
   // Next, check to make sure that all exit nodes of the loop only have
   // predecessors that are inside of the loop.  This check guarantees that the
   // loop preheader/header will dominate the exit blocks.  If the exit block has
@@ -155,19 +158,16 @@ bool LoopSimplify::ProcessLoop(Loop *L) {
   L->getExitBlocks(ExitBlocks);
   for (unsigned i = 0, e = ExitBlocks.size(); i != e; ++i) {
     BasicBlock *ExitBlock = ExitBlocks[i];
-    for (pred_iterator PI = pred_begin(ExitBlock), PE = pred_end(ExitBlock);
-         PI != PE; ++PI)
-      if (!L->contains(*PI)) {
-        BasicBlock *NewBB = RewriteLoopExitBlock(L, ExitBlock);
-        for (unsigned j = i; j != ExitBlocks.size(); ++j)
-          if (ExitBlocks[j] == ExitBlock)
-            ExitBlocks[j] = NewBB;
-
-        NumInserted++;
-        Changed = true;
-        break;
-      }
+    if (!DS.dominates(L->getHeader(), ExitBlock)) {
+      BasicBlock *NewBB = RewriteLoopExitBlock(L, ExitBlock);
+      for (unsigned j = i; j != ExitBlocks.size(); ++j)
+        if (ExitBlocks[j] == ExitBlock)
+          ExitBlocks[j] = NewBB;
+      
+      NumInserted++;
+      Changed = true;
     }
+  }
 
   // If the header has more than two predecessors at this point (from the
   // preheader and from multiple backedges), we must adjust the loop.