SCEV: Add MarkPendingLoopPredicates to avoid recursive isImpliedCond.

getUDivExpr attempts to simplify by checking for overflow.
isLoopEntryGuardedByCond then evaluates the loop predicate which
may lead to the same getUDivExpr causing endless recursion.

Fixes PR12868: clang 3.2 segmentation fault.

llvm-svn: 157092

diff --git a/llvm/include/llvm/Analysis/ScalarEvolution.h b/llvm/include/llvm/Analysis/ScalarEvolution.h
index 72408f7..8f87b58 100644 (file)
--- a/llvm/include/llvm/Analysis/ScalarEvolution.h
+++ b/llvm/include/llvm/Analysis/ScalarEvolution.h
@@ -30,7 +30,7 @@
 #include "llvm/Support/Allocator.h"
 #include "llvm/Support/ConstantRange.h"
 #include "llvm/ADT/FoldingSet.h"
-#include "llvm/ADT/DenseMap.h"
+#include "llvm/ADT/DenseSet.h"
 #include <map>
 
 namespace llvm {
@@ -250,6 +250,9 @@ namespace llvm {
     ///
     ValueExprMapType ValueExprMap;
 
+    /// Mark predicate values currently being processed by isImpliedCond.
+    DenseSet<Value*> PendingLoopPredicates;
+
     /// ExitLimit - Information about the number of loop iterations for
     /// which a loop exit's branch condition evaluates to the not-taken path.
     /// This is a temporary pair of exact and max expressions that are

diff --git a/llvm/lib/Analysis/ScalarEvolution.cpp b/llvm/lib/Analysis/ScalarEvolution.cpp
index d1ad8e0..4e3dc60 100644 (file)
--- a/llvm/lib/Analysis/ScalarEvolution.cpp
+++ b/llvm/lib/Analysis/ScalarEvolution.cpp
@@ -6039,12 +6039,34 @@ ScalarEvolution::isLoopEntryGuardedByCond(const Loop *L,
   return false;
 }
 
+/// RAII wrapper to prevent recursive application of isImpliedCond.
+/// ScalarEvolution's PendingLoopPredicates set must be empty unless we are
+/// currently evaluating isImpliedCond.
+struct MarkPendingLoopPredicate {
+  Value *Cond;
+  DenseSet<Value*> &LoopPreds;
+  bool Pending;
+
+  MarkPendingLoopPredicate(Value *C, DenseSet<Value*> &LP)
+    : Cond(C), LoopPreds(LP) {
+    Pending = !LoopPreds.insert(Cond).second;
+  }
+  ~MarkPendingLoopPredicate() {
+    if (!Pending)
+      LoopPreds.erase(Cond);
+  }
+};
+
 /// isImpliedCond - Test whether the condition described by Pred, LHS,
 /// and RHS is true whenever the given Cond value evaluates to true.
 bool ScalarEvolution::isImpliedCond(ICmpInst::Predicate Pred,
                                     const SCEV *LHS, const SCEV *RHS,
                                     Value *FoundCondValue,
                                     bool Inverse) {
+  MarkPendingLoopPredicate Mark(FoundCondValue, PendingLoopPredicates);
+  if (Mark.Pending)
+    return false;
+
   // Recursively handle And and Or conditions.
   if (BinaryOperator *BO = dyn_cast<BinaryOperator>(FoundCondValue)) {
     if (BO->getOpcode() == Instruction::And) {
@@ -6571,6 +6593,8 @@ void ScalarEvolution::releaseMemory() {
     I->second.clear();
   }
 
+  assert(PendingLoopPredicates.empty() && "isImpliedCond garbage");
+
   BackedgeTakenCounts.clear();
   ConstantEvolutionLoopExitValue.clear();
   ValuesAtScopes.clear();

diff --git a/llvm/test/Analysis/ScalarEvolution/2012-05-18-LoopPredRecurse.ll b/llvm/test/Analysis/ScalarEvolution/2012-05-18-LoopPredRecurse.ll
new file mode 100644 (file)
index 0000000..52e6683
--- /dev/null
+++ b/llvm/test/Analysis/ScalarEvolution/2012-05-18-LoopPredRecurse.ll
@@ -0,0 +1,30 @@
+; RUN: opt < %s -iv-users -S -disable-output
+;
+; PR12868: Infinite recursion:
+; getUDivExpr()->getZeroExtendExpr()->isLoopBackedgeGuardedBy()
+;
+; We actually want SCEV simplification to fail gracefully in this
+; case, so there's no output to check, just the absense of stack overflow.
+
+@c = common global i8 0, align 1
+
+define i32 @func() {
+entry:
+  br label %for.cond
+
+for.cond:                                         ; preds = %for.body, %entry
+  %storemerge = phi i8 [ -1, %entry ], [ %inc, %for.body ]
+  %ui.0 = phi i32 [ undef, %entry ], [ %div, %for.body ]
+  %tobool = icmp eq i8 %storemerge, 0
+  br i1 %tobool, label %for.end, label %for.body
+
+for.body:                                         ; preds = %for.cond
+  %conv = sext i8 %storemerge to i32
+  %div = lshr i32 %conv, 1
+  %tobool2 = icmp eq i32 %div, 0
+  %inc = add i8 %storemerge, 1
+  br i1 %tobool2, label %for.cond, label %for.end
+
+for.end:                                          ; preds = %for.body, %for.cond
+  ret i32 0
+}