Implement PR1777 by detecting dependent phis that

all compute the same value.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@43777 91177308-0d34-0410-b5e6-96231b3b80d8

diff --git a/lib/Transforms/Scalar/InstructionCombining.cpp b/lib/Transforms/Scalar/InstructionCombining.cpp
index 1a638c30884f41d9ef91109eee9ff3d059bc44e5..41485b203ecdacd398b672fee9072e083c55c7da 100644 (file)
--- a/lib/Transforms/Scalar/InstructionCombining.cpp
+++ b/lib/Transforms/Scalar/InstructionCombining.cpp
@@ -8551,6 +8551,34 @@ static bool DeadPHICycle(PHINode *PN,
   return false;
 }
 
+/// PHIsEqualValue - Return true if this phi node is always equal to
+/// NonPhiInVal.  This happens with mutually cyclic phi nodes like:
+///   z = some value; x = phi (y, z); y = phi (x, z)
+static bool PHIsEqualValue(PHINode *PN, Value *NonPhiInVal, 
+                           SmallPtrSet<PHINode*, 16> &ValueEqualPHIs) {
+  // See if we already saw this PHI node.
+  if (!ValueEqualPHIs.insert(PN))
+    return true;
+  
+  // Don't scan crazily complex things.
+  if (ValueEqualPHIs.size() == 16)
+    return false;
+ 
+  // Scan the operands to see if they are either phi nodes or are equal to
+  // the value.
+  for (unsigned i = 0, e = PN->getNumIncomingValues(); i != e; ++i) {
+    Value *Op = PN->getIncomingValue(i);
+    if (PHINode *OpPN = dyn_cast<PHINode>(Op)) {
+      if (!PHIsEqualValue(OpPN, NonPhiInVal, ValueEqualPHIs))
+        return false;
+    } else if (Op != NonPhiInVal)
+      return false;
+  }
+  
+  return true;
+}
+
+
 // PHINode simplification
 //
 Instruction *InstCombiner::visitPHINode(PHINode &PN) {
@@ -8592,6 +8620,40 @@ Instruction *InstCombiner::visitPHINode(PHINode &PN) {
     }
   }
 
+  // We sometimes end up with phi cycles that non-obviously end up being the
+  // same value, for example:
+  //   z = some value; x = phi (y, z); y = phi (x, z)
+  // where the phi nodes don't necessarily need to be in the same block.  Do a
+  // quick check to see if the PHI node only contains a single non-phi value, if
+  // so, scan to see if the phi cycle is actually equal to that value.
+  {
+    unsigned InValNo = 0, NumOperandVals = PN.getNumIncomingValues();
+    // Scan for the first non-phi operand.
+    while (InValNo != NumOperandVals && 
+           isa<PHINode>(PN.getIncomingValue(InValNo)))
+      ++InValNo;
+
+    if (InValNo != NumOperandVals) {
+      Value *NonPhiInVal = PN.getOperand(InValNo);
+      
+      // Scan the rest of the operands to see if there are any conflicts, if so
+      // there is no need to recursively scan other phis.
+      for (++InValNo; InValNo != NumOperandVals; ++InValNo) {
+        Value *OpVal = PN.getIncomingValue(InValNo);
+        if (OpVal != NonPhiInVal && !isa<PHINode>(OpVal))
+          break;
+      }
+      
+      // If we scanned over all operands, then we have one unique value plus
+      // phi values.  Scan PHI nodes to see if they all merge in each other or
+      // the value.
+      if (InValNo == NumOperandVals) {
+        SmallPtrSet<PHINode*, 16> ValueEqualPHIs;
+        if (PHIsEqualValue(&PN, NonPhiInVal, ValueEqualPHIs))
+          return ReplaceInstUsesWith(PN, NonPhiInVal);
+      }
+    }
+  }
   return 0;
 }
 

diff --git a/test/Transforms/InstCombine/phi-merge.ll b/test/Transforms/InstCombine/phi-merge.ll
new file mode 100644 (file)
index 0000000..daac412
--- /dev/null
+++ b/test/Transforms/InstCombine/phi-merge.ll
@@ -0,0 +1,31 @@
+; RUN: llvm-as < %s | opt -instcombine | llvm-dis | not grep {phi i32}
+; PR1777
+
+declare i1 @rrr()
+
+define i1 @zxcv() {
+entry:
+%a = alloca i32
+%i = ptrtoint i32* %a to i32
+%b = call i1 @rrr()
+br i1 %b, label %one, label %two
+
+one:
+%x = phi i32 [%i, %entry], [%y, %two]
+%c = call i1 @rrr()
+br i1 %c, label %two, label %end
+
+two:
+%y = phi i32 [%i, %entry], [%x, %one]
+%d = call i1 @rrr()
+br i1 %d, label %one, label %end
+
+end:
+%f = phi i32 [ %x, %one], [%y, %two]
+; Change the %f to %i, and the optimizer suddenly becomes a lot smarter
+; even though %f must equal %i at this point
+%g = inttoptr i32 %f to i32*
+store i32 10, i32* %g
+%z = call i1 @rrr()
+ret i1 %z
+}