[SCEV] Use depth limit instead of local cache for SExt and ZExt

[mkazantsev]

Summary

In rL300494 there was an attempt to deal with excessive compile time on
invocations of getSign/ZeroExtExpr using local caching. This approach only
helps if we request the same SCEV multiple times throughout recursion. But
in the bug PR33431 we see a case where we request different values all the time,
so caching does not help and the size of the cache grows enormously.

In this patch we remove the local cache for this methods and add the recursion
depth limit instead, as we do for arithmetics. This gives us a guarantee that the
invocation sequence is limited and reasonably short.

[wmi]

Thanks for helping on fixing the bug!

lib/Analysis/ScalarEvolution.cpp

1581–1582

Do we want to pass Depth into getExtendAddRecStart? Since getZeroExtendExpr and getSignExtendExpr can be called recursively inside of getExtendAddRecStart.

[mkazantsev]

lib/Analysis/ScalarEvolution.cpp

1581–1582

Thanks for pointing out! I will go through all invocations here, maybe we are missing it somewhere else.

[mkazantsev]

lib/Analysis/ScalarEvolution.cpp

1581–1582

Wei, actually I failed to find a place where getAddRecExpr invokes either of them. I only see it invoing itself. We maybe need a separate depth limit for it, but it is out of scope of this patch.

If you know a way how getSign/ZeroExt can be invoked from getAddRecExpr, could you please point it out?

[wmi]

lib/Analysis/ScalarEvolution.cpp

1581–1582

It is not getAddRecExpr. What I said was getExtendAddRecStart.

[mkazantsev]

My bad, need to change this. :) Thanks!

[mkazantsev]

Added depth limitation passed through getExtendAddRecStart

[wmi]

lib/Analysis/ScalarEvolution.cpp

1430

Maybe also pass Depth to getPreStartForExtend? we have SE->*GetExtendExpr() called inside of getPreStartForExtend.

[mkazantsev]

Added depth passed through getPreStartForExtend.

include/llvm/Analysis/ScalarEvolution.h

public:
   /// Return a SCEV expression for the full generality of the specified
   /// expression.
   const SCEV *getSCEV(Value *V);
 
   const SCEV *getConstant(ConstantInt *V);
   const SCEV *getConstant(const APInt &Val);
   const SCEV *getConstant(Type *Ty, uint64_t V, bool isSigned = false);
   const SCEV *getTruncateExpr(const SCEV *Op, Type *Ty);
-
-  typedef SmallDenseMap<std::pair<const SCEV *, Type *>, const SCEV *, 8>
-      ExtendCacheTy;
-  const SCEV *getZeroExtendExpr(const SCEV *Op, Type *Ty);
-  const SCEV *getZeroExtendExprCached(const SCEV *Op, Type *Ty,
-                                      ExtendCacheTy &Cache);
-  const SCEV *getZeroExtendExprImpl(const SCEV *Op, Type *Ty,
-                                    ExtendCacheTy &Cache);
-
-  const SCEV *getSignExtendExpr(const SCEV *Op, Type *Ty);
-  const SCEV *getSignExtendExprCached(const SCEV *Op, Type *Ty,
-                                      ExtendCacheTy &Cache);
-  const SCEV *getSignExtendExprImpl(const SCEV *Op, Type *Ty,
-                                    ExtendCacheTy &Cache);
+  const SCEV *getZeroExtendExpr(const SCEV *Op, Type *Ty, unsigned Depth = 0);
+  const SCEV *getSignExtendExpr(const SCEV *Op, Type *Ty, unsigned Depth = 0);
   const SCEV *getAnyExtendExpr(const SCEV *Op, Type *Ty);
   const SCEV *getAddExpr(SmallVectorImpl<const SCEV *> &Ops,
                          SCEV::NoWrapFlags Flags = SCEV::FlagAnyWrap,
                          unsigned Depth = 0);
   const SCEV *getAddExpr(const SCEV *LHS, const SCEV *RHS,
                          SCEV::NoWrapFlags Flags = SCEV::FlagAnyWrap,
                          unsigned Depth = 0) {
     SmallVector<const SCEV *, 2> Ops = {LHS, RHS};

lib/Analysis/ScalarEvolution.cpp

static cl::opt<unsigned>
     MaxArithDepth("scalar-evolution-max-arith-depth", cl::Hidden,
                   cl::desc("Maximum depth of recursive arithmetics"),
                   cl::init(32));
 
 static cl::opt<unsigned> MaxConstantEvolvingDepth(
     "scalar-evolution-max-constant-evolving-depth", cl::Hidden,
     cl::desc("Maximum depth of recursive constant evolving"), cl::init(32));
 
+static cl::opt<unsigned>
+    MaxExtDepth("scalar-evolution-max-ext-depth", cl::Hidden,
+                cl::desc("Maximum depth of recursive SExt/ZExt"),
+                cl::init(8));
+
 //===----------------------------------------------------------------------===//
 //                           SCEV class definitions
 //===----------------------------------------------------------------------===//
 
 //===----------------------------------------------------------------------===//
 // Implementation of the SCEV class.
 //
 
static const SCEV *getUnsignedOverflowLimitForStep(const SCEV *Step,
 
   return SE->getConstant(APInt::getMinValue(BitWidth) -
                          SE->getUnsignedRangeMax(Step));
 }
 
 namespace {
 
 struct ExtendOpTraitsBase {
-  typedef const SCEV *(ScalarEvolution::*GetExtendExprTy)(
-      const SCEV *, Type *, ScalarEvolution::ExtendCacheTy &Cache);
+  typedef const SCEV *(ScalarEvolution::*GetExtendExprTy)(const SCEV *, Type *,
+                                                          unsigned);
 };
 
 // Used to make code generic over signed and unsigned overflow.
 template <typename ExtendOp> struct ExtendOpTraits {
   // Members present:
   //
   // static const SCEV::NoWrapFlags WrapType;
   //
struct ExtendOpTraits<SCEVSignExtendExpr> : public ExtendOpTraitsBase {
 
   static const SCEV *getOverflowLimitForStep(const SCEV *Step,
                                              ICmpInst::Predicate *Pred,
                                              ScalarEvolution *SE) {
     return getSignedOverflowLimitForStep(Step, Pred, SE);
   }
 };
 
-const ExtendOpTraitsBase::GetExtendExprTy
-    ExtendOpTraits<SCEVSignExtendExpr>::GetExtendExpr =
-        &ScalarEvolution::getSignExtendExprCached;
+const ExtendOpTraitsBase::GetExtendExprTy ExtendOpTraits<
+    SCEVSignExtendExpr>::GetExtendExpr = &ScalarEvolution::getSignExtendExpr;
 
 template <>
 struct ExtendOpTraits<SCEVZeroExtendExpr> : public ExtendOpTraitsBase {
   static const SCEV::NoWrapFlags WrapType = SCEV::FlagNUW;
 
   static const GetExtendExprTy GetExtendExpr;
 
   static const SCEV *getOverflowLimitForStep(const SCEV *Step,
                                              ICmpInst::Predicate *Pred,
                                              ScalarEvolution *SE) {
     return getUnsignedOverflowLimitForStep(Step, Pred, SE);
   }
 };
 
-const ExtendOpTraitsBase::GetExtendExprTy
-    ExtendOpTraits<SCEVZeroExtendExpr>::GetExtendExpr =
-        &ScalarEvolution::getZeroExtendExprCached;
+const ExtendOpTraitsBase::GetExtendExprTy ExtendOpTraits<
+    SCEVZeroExtendExpr>::GetExtendExpr = &ScalarEvolution::getZeroExtendExpr;
 }
 
 // The recurrence AR has been shown to have no signed/unsigned wrap or something
 // close to it. Typically, if we can prove NSW/NUW for AR, then we can just as
 // easily prove NSW/NUW for its preincrement or postincrement sibling. This
 // allows normalizing a sign/zero extended AddRec as such: {sext/zext(Step +
 // Start),+,Step} => {(Step + sext/zext(Start),+,Step} As a result, the
 // expression "Step + sext/zext(PreIncAR)" is congruent with
 // "sext/zext(PostIncAR)"
 template <typename ExtendOpTy>
 static const SCEV *getPreStartForExtend(const SCEVAddRecExpr *AR, Type *Ty,
-                                        ScalarEvolution *SE,
-                                        ScalarEvolution::ExtendCacheTy &Cache) {
+                                        ScalarEvolution *SE) {
   auto WrapType = ExtendOpTraits<ExtendOpTy>::WrapType;
   auto GetExtendExpr = ExtendOpTraits<ExtendOpTy>::GetExtendExpr;
 
   const Loop *L = AR->getLoop();
   const SCEV *Start = AR->getStart();
   const SCEV *Step = AR->getStepRecurrence(*SE);
 
   // Check for a simple looking step prior to loop entry.
static const SCEV *getPreStartForExtend(const SCEVAddRecExpr *AR, Type *Ty,
   if (PreAR && PreAR->getNoWrapFlags(WrapType) &&
       !isa<SCEVCouldNotCompute>(BECount) && SE->isKnownPositive(BECount))
     return PreStart;
 
   // 2. Direct overflow check on the step operation's expression.
   unsigned BitWidth = SE->getTypeSizeInBits(AR->getType());
   Type *WideTy = IntegerType::get(SE->getContext(), BitWidth * 2);
   const SCEV *OperandExtendedStart =
-      SE->getAddExpr((SE->*GetExtendExpr)(PreStart, WideTy, Cache),
-                     (SE->*GetExtendExpr)(Step, WideTy, Cache));
-  if ((SE->*GetExtendExpr)(Start, WideTy, Cache) == OperandExtendedStart) {
+      SE->getAddExpr((SE->*GetExtendExpr)(PreStart, WideTy, 0),
+                     (SE->*GetExtendExpr)(Step, WideTy, 0));
+  if ((SE->*GetExtendExpr)(Start, WideTy, 0) == OperandExtendedStart) {
     if (PreAR && AR->getNoWrapFlags(WrapType)) {
       // If we know `AR` == {`PreStart`+`Step`,+,`Step`} is `WrapType` (FlagNSW
       // or FlagNUW) and that `PreStart` + `Step` is `WrapType` too, then
       // `PreAR` == {`PreStart`,+,`Step`} is also `WrapType`.  Cache this fact.
       const_cast<SCEVAddRecExpr *>(PreAR)->setNoWrapFlags(WrapType);
     }
     return PreStart;
   }
static const SCEV *getPreStartForExtend(const SCEVAddRecExpr *AR, Type *Ty,
 
   return nullptr;
 }
 
 // Get the normalized zero or sign extended expression for this AddRec's Start.
 template <typename ExtendOpTy>
 static const SCEV *getExtendAddRecStart(const SCEVAddRecExpr *AR, Type *Ty,
                                         ScalarEvolution *SE,
-                                        ScalarEvolution::ExtendCacheTy &Cache) {
+                                        unsigned Depth) {
   auto GetExtendExpr = ExtendOpTraits<ExtendOpTy>::GetExtendExpr;
 
-  const SCEV *PreStart = getPreStartForExtend<ExtendOpTy>(AR, Ty, SE, Cache);
+  const SCEV *PreStart = getPreStartForExtend<ExtendOpTy>(AR, Ty, SE);
   if (!PreStart)
-    return (SE->*GetExtendExpr)(AR->getStart(), Ty, Cache);
+    return (SE->*GetExtendExpr)(AR->getStart(), Ty, Depth);
 
-  return SE->getAddExpr(
-      (SE->*GetExtendExpr)(AR->getStepRecurrence(*SE), Ty, Cache),
-      (SE->*GetExtendExpr)(PreStart, Ty, Cache));
+  return SE->getAddExpr((SE->*GetExtendExpr)(AR->getStepRecurrence(*SE), Ty,
+                                             Depth),
+                        (SE->*GetExtendExpr)(PreStart, Ty, Depth));
 }
 
 // Try to prove away overflow by looking at "nearby" add recurrences.  A
 // motivating example for this rule: if we know `{0,+,4}` is `ult` `-1` and it
 // does not itself wrap then we can conclude that `{1,+,4}` is `nuw`.
 //
 // Formally:
 //
if (PreAR && PreAR->getNoWrapFlags(WrapType)) {  // proves (2)
       if (Limit && isKnownPredicate(Pred, PreAR, Limit))  // proves (1)
         return true;
     }
   }
 
   return false;
 }
 
-const SCEV *ScalarEvolution::getZeroExtendExpr(const SCEV *Op, Type *Ty) {
-  // Use the local cache to prevent exponential behavior of
-  // getZeroExtendExprImpl.
-  ExtendCacheTy Cache;
-  return getZeroExtendExprCached(Op, Ty, Cache);
-}
-
-/// Query \p Cache before calling getZeroExtendExprImpl. If there is no
-/// related entry in the \p Cache, call getZeroExtendExprImpl and save
-/// the result in the \p Cache.
-const SCEV *ScalarEvolution::getZeroExtendExprCached(const SCEV *Op, Type *Ty,
-                                                     ExtendCacheTy &Cache) {
-  auto It = Cache.find({Op, Ty});
-  if (It != Cache.end())
-    return It->second;
-  const SCEV *ZExt = getZeroExtendExprImpl(Op, Ty, Cache);
-  auto InsertResult = Cache.insert({{Op, Ty}, ZExt});
-  assert(InsertResult.second && "Expect the key was not in the cache");
-  (void)InsertResult;
-  return ZExt;
-}
-
-/// The real implementation of getZeroExtendExpr.
-const SCEV *ScalarEvolution::getZeroExtendExprImpl(const SCEV *Op, Type *Ty,
-                                                   ExtendCacheTy &Cache) {
+const SCEV *
+ScalarEvolution::getZeroExtendExpr(const SCEV *Op, Type *Ty, unsigned Depth) {
   assert(getTypeSizeInBits(Op->getType()) < getTypeSizeInBits(Ty) &&
          "This is not an extending conversion!");
   assert(isSCEVable(Ty) &&
          "This is not a conversion to a SCEVable type!");
   Ty = getEffectiveSCEVType(Ty);
 
   // Fold if the operand is constant.
   if (const SCEVConstant *SC = dyn_cast<SCEVConstant>(Op))
     return getConstant(
-        cast<ConstantInt>(ConstantExpr::getZExt(SC->getValue(), Ty)));
 
   // zext(zext(x)) --> zext(x)
   if (const SCEVZeroExtendExpr *SZ = dyn_cast<SCEVZeroExtendExpr>(Op))
-    return getZeroExtendExprCached(SZ->getOperand(), Ty, Cache);
+    return getZeroExtendExpr(SZ->getOperand(), Ty, Depth + 1);
 
   // Before doing any expensive analysis, check to see if we've already
   // computed a SCEV for this Op and Ty.
   FoldingSetNodeID ID;
   ID.AddInteger(scZeroExtend);
   ID.AddPointer(Op);
   ID.AddPointer(Ty);
   void *IP = nullptr;
   if (const SCEV *S = UniqueSCEVs.FindNodeOrInsertPos(ID, IP)) return S;
+  if (Depth > MaxExtDepth) {
+    SCEV *S = new (SCEVAllocator) SCEVZeroExtendExpr(ID.Intern(SCEVAllocator),
+                                                     Op, Ty);
+    UniqueSCEVs.InsertNode(S, IP);
+    return S;
+  }
 
   // zext(trunc(x)) --> zext(x) or x or trunc(x)
   if (const SCEVTruncateExpr *ST = dyn_cast<SCEVTruncateExpr>(Op)) {
     // It's possible the bits taken off by the truncate were all zero bits. If
     // so, we should be able to simplify this further.
     const SCEV *X = ST->getOperand();
     ConstantRange CR = getUnsignedRange(X);
     unsigned TruncBits = getTypeSizeInBits(ST->getType());
if (AR->isAffine()) {
         auto NewFlags = proveNoWrapViaConstantRanges(AR);
         const_cast<SCEVAddRecExpr *>(AR)->setNoWrapFlags(NewFlags);
       }
 
       // If we have special knowledge that this addrec won't overflow,
       // we don't need to do any further analysis.
       if (AR->hasNoUnsignedWrap())
         return getAddRecExpr(
-            getExtendAddRecStart<SCEVZeroExtendExpr>(AR, Ty, this, Cache),
-            getZeroExtendExprCached(Step, Ty, Cache), L, AR->getNoWrapFlags());
+            getExtendAddRecStart<SCEVZeroExtendExpr>(AR, Ty, this, Depth + 1),
+            getZeroExtendExpr(Step, Ty, Depth + 1), L, AR->getNoWrapFlags());
 
       // Check whether the backedge-taken count is SCEVCouldNotCompute.
       // Note that this serves two purposes: It filters out loops that are
       // simply not analyzable, and it covers the case where this code is
       // being called from within backedge-taken count analysis, such that
       // attempting to ask for the backedge-taken count would likely result
       // in infinite recursion. In the later case, the analysis code will
       // cope with a conservative value, and it will take care to purge
       // that value once it has finished.
       const SCEV *MaxBECount = getMaxBackedgeTakenCount(L);
       if (!isa<SCEVCouldNotCompute>(MaxBECount)) {
         // Manually compute the final value for AR, checking for
         // overflow.
 
         // Check whether the backedge-taken count can be losslessly casted to
         // the addrec's type. The count is always unsigned.
         const SCEV *CastedMaxBECount =
           getTruncateOrZeroExtend(MaxBECount, Start->getType());
         const SCEV *RecastedMaxBECount =
           getTruncateOrZeroExtend(CastedMaxBECount, MaxBECount->getType());
         if (MaxBECount == RecastedMaxBECount) {
           Type *WideTy = IntegerType::get(getContext(), BitWidth * 2);
           // Check whether Start+Step*MaxBECount has no unsigned overflow.
-          const SCEV *ZMul = getMulExpr(CastedMaxBECount, Step);
-          const SCEV *ZAdd =
-              getZeroExtendExprCached(getAddExpr(Start, ZMul), WideTy, Cache);
-          const SCEV *WideStart = getZeroExtendExprCached(Start, WideTy, Cache);
+          const SCEV *ZMul = getMulExpr(CastedMaxBECount, Step,
+                                        SCEV::FlagAnyWrap, Depth + 1);
+          const SCEV *ZAdd = getZeroExtendExpr(getAddExpr(Start, ZMul,
+                                                          SCEV::FlagAnyWrap,
+                                                          Depth + 1),
+                                               WideTy, Depth + 1);
+          const SCEV *WideStart = getZeroExtendExpr(Start, WideTy, Depth + 1);
           const SCEV *WideMaxBECount =
-              getZeroExtendExprCached(CastedMaxBECount, WideTy, Cache);
-          const SCEV *OperandExtendedAdd = getAddExpr(
-              WideStart, getMulExpr(WideMaxBECount, getZeroExtendExprCached(
-                                                        Step, WideTy, Cache)));
+            getZeroExtendExpr(CastedMaxBECount, WideTy, Depth + 1);
+          const SCEV *OperandExtendedAdd =
+            getAddExpr(WideStart,
+                       getMulExpr(WideMaxBECount,
+                                  getZeroExtendExpr(Step, WideTy, Depth + 1),
+                                  SCEV::FlagAnyWrap, Depth + 1),
+                       SCEV::FlagAnyWrap, Depth + 1);
           if (ZAdd == OperandExtendedAdd) {
             // Cache knowledge of AR NUW, which is propagated to this AddRec.
             const_cast<SCEVAddRecExpr *>(AR)->setNoWrapFlags(SCEV::FlagNUW);
             // Return the expression with the addrec on the outside.
             return getAddRecExpr(
-                getExtendAddRecStart<SCEVZeroExtendExpr>(AR, Ty, this, Cache),
-                getZeroExtendExprCached(Step, Ty, Cache), L,
+                getExtendAddRecStart<SCEVZeroExtendExpr>(AR, Ty, this,
+                                                         Depth + 1),
+                getZeroExtendExpr(Step, Ty, Depth + 1), L,
                 AR->getNoWrapFlags());
           }
           // Similar to above, only this time treat the step value as signed.
           // This covers loops that count down.
           OperandExtendedAdd =
             getAddExpr(WideStart,
                        getMulExpr(WideMaxBECount,
-                                  getSignExtendExpr(Step, WideTy)));
+                                  getSignExtendExpr(Step, WideTy, Depth + 1),
+                                  SCEV::FlagAnyWrap, Depth + 1),
+                       SCEV::FlagAnyWrap, Depth + 1);
           if (ZAdd == OperandExtendedAdd) {
             // Cache knowledge of AR NW, which is propagated to this AddRec.
             // Negative step causes unsigned wrap, but it still can't self-wrap.
             const_cast<SCEVAddRecExpr *>(AR)->setNoWrapFlags(SCEV::FlagNW);
             // Return the expression with the addrec on the outside.
             return getAddRecExpr(
-                getExtendAddRecStart<SCEVZeroExtendExpr>(AR, Ty, this, Cache),
-                getSignExtendExpr(Step, Ty), L, AR->getNoWrapFlags());
+                getExtendAddRecStart<SCEVZeroExtendExpr>(AR, Ty, this,
+                                                         Depth + 1),
+                getSignExtendExpr(Step, Ty, Depth + 1), L,
+                AR->getNoWrapFlags());
           }
         }
       }
 
       // Normally, in the cases we can prove no-overflow via a
       // backedge guarding condition, we can also compute a backedge
       // taken count for the loop.  The exceptions are assumptions and
       // guards present in the loop -- SCEV is not great at exploiting
if (AR->isAffine()) {
               (isLoopEntryGuardedByCond(L, ICmpInst::ICMP_ULT, Start, N) &&
                isLoopBackedgeGuardedByCond(L, ICmpInst::ICMP_ULT,
                                            AR->getPostIncExpr(*this), N))) {
             // Cache knowledge of AR NUW, which is propagated to this
             // AddRec.
             const_cast<SCEVAddRecExpr *>(AR)->setNoWrapFlags(SCEV::FlagNUW);
             // Return the expression with the addrec on the outside.
             return getAddRecExpr(
-                getExtendAddRecStart<SCEVZeroExtendExpr>(AR, Ty, this, Cache),
-                getZeroExtendExprCached(Step, Ty, Cache), L,
+                getExtendAddRecStart<SCEVZeroExtendExpr>(AR, Ty, this,
+                                                         Depth + 1),
+                getZeroExtendExpr(Step, Ty, Depth + 1), L,
                 AR->getNoWrapFlags());
           }
         } else if (isKnownNegative(Step)) {
           const SCEV *N = getConstant(APInt::getMaxValue(BitWidth) -
                                       getSignedRangeMin(Step));
           if (isLoopBackedgeGuardedByCond(L, ICmpInst::ICMP_UGT, AR, N) ||
               (isLoopEntryGuardedByCond(L, ICmpInst::ICMP_UGT, Start, N) &&
                isLoopBackedgeGuardedByCond(L, ICmpInst::ICMP_UGT,
                                            AR->getPostIncExpr(*this), N))) {
             // Cache knowledge of AR NW, which is propagated to this
             // AddRec.  Negative step causes unsigned wrap, but it
             // still can't self-wrap.
             const_cast<SCEVAddRecExpr *>(AR)->setNoWrapFlags(SCEV::FlagNW);
             // Return the expression with the addrec on the outside.
             return getAddRecExpr(
-                getExtendAddRecStart<SCEVZeroExtendExpr>(AR, Ty, this, Cache),
-                getSignExtendExpr(Step, Ty), L, AR->getNoWrapFlags());
+                getExtendAddRecStart<SCEVZeroExtendExpr>(AR, Ty, this,
+                                                         Depth + 1),
+                getSignExtendExpr(Step, Ty, Depth + 1), L,
+                AR->getNoWrapFlags());
           }
         }
       }
 
       if (proveNoWrapByVaryingStart<SCEVZeroExtendExpr>(Start, Step, L)) {
         const_cast<SCEVAddRecExpr *>(AR)->setNoWrapFlags(SCEV::FlagNUW);
         return getAddRecExpr(
-            getExtendAddRecStart<SCEVZeroExtendExpr>(AR, Ty, this, Cache),
-            getZeroExtendExprCached(Step, Ty, Cache), L, AR->getNoWrapFlags());
+            getExtendAddRecStart<SCEVZeroExtendExpr>(AR, Ty, this, Depth + 1),
+            getZeroExtendExpr(Step, Ty, Depth + 1), L, AR->getNoWrapFlags());
       }
     }
 
   if (auto *SA = dyn_cast<SCEVAddExpr>(Op)) {
     // zext((A + B + ...)<nuw>) --> (zext(A) + zext(B) + ...)<nuw>
     if (SA->hasNoUnsignedWrap()) {
       // If the addition does not unsign overflow then we can, by definition,
       // commute the zero extension with the addition operation.
       SmallVector<const SCEV *, 4> Ops;
       for (const auto *Op : SA->operands())
-        Ops.push_back(getZeroExtendExprCached(Op, Ty, Cache));
-      return getAddExpr(Ops, SCEV::FlagNUW);
+        Ops.push_back(getZeroExtendExpr(Op, Ty, Depth + 1));
+      return getAddExpr(Ops, SCEV::FlagNUW, Depth + 1);
     }
   }
 
   // The cast wasn't folded; create an explicit cast node.
   // Recompute the insert position, as it may have been invalidated.
   if (const SCEV *S = UniqueSCEVs.FindNodeOrInsertPos(ID, IP)) return S;
   SCEV *S = new (SCEVAllocator) SCEVZeroExtendExpr(ID.Intern(SCEVAllocator),
                                                    Op, Ty);
   UniqueSCEVs.InsertNode(S, IP);
   return S;
 }
 
-const SCEV *ScalarEvolution::getSignExtendExpr(const SCEV *Op, Type *Ty) {
-  // Use the local cache to prevent exponential behavior of
-  // getSignExtendExprImpl.
-  ExtendCacheTy Cache;
-  return getSignExtendExprCached(Op, Ty, Cache);
-}
-
-/// Query \p Cache before calling getSignExtendExprImpl. If there is no
-/// related entry in the \p Cache, call getSignExtendExprImpl and save
-/// the result in the \p Cache.
-const SCEV *ScalarEvolution::getSignExtendExprCached(const SCEV *Op, Type *Ty,
-                                                     ExtendCacheTy &Cache) {
-  auto It = Cache.find({Op, Ty});
-  if (It != Cache.end())
-    return It->second;
-  const SCEV *SExt = getSignExtendExprImpl(Op, Ty, Cache);
-  auto InsertResult = Cache.insert({{Op, Ty}, SExt});
-  assert(InsertResult.second && "Expect the key was not in the cache");
-  (void)InsertResult;
-  return SExt;
-}
-
-/// The real implementation of getSignExtendExpr.
-const SCEV *ScalarEvolution::getSignExtendExprImpl(const SCEV *Op, Type *Ty,
-                                                   ExtendCacheTy &Cache) {
+const SCEV *
+ScalarEvolution::getSignExtendExpr(const SCEV *Op, Type *Ty, unsigned Depth) {
   assert(getTypeSizeInBits(Op->getType()) < getTypeSizeInBits(Ty) &&
          "This is not an extending conversion!");
   assert(isSCEVable(Ty) &&
          "This is not a conversion to a SCEVable type!");
   Ty = getEffectiveSCEVType(Ty);
 
   // Fold if the operand is constant.
   if (const SCEVConstant *SC = dyn_cast<SCEVConstant>(Op))
     return getConstant(
-        cast<ConstantInt>(ConstantExpr::getSExt(SC->getValue(), Ty)));
 
   // sext(sext(x)) --> sext(x)
   if (const SCEVSignExtendExpr *SS = dyn_cast<SCEVSignExtendExpr>(Op))
-    return getSignExtendExprCached(SS->getOperand(), Ty, Cache);
+    return getSignExtendExpr(SS->getOperand(), Ty, Depth + 1);
 
   // sext(zext(x)) --> zext(x)
   if (const SCEVZeroExtendExpr *SZ = dyn_cast<SCEVZeroExtendExpr>(Op))
-    return getZeroExtendExpr(SZ->getOperand(), Ty);
+    return getZeroExtendExpr(SZ->getOperand(), Ty, Depth + 1);
 
   // Before doing any expensive analysis, check to see if we've already
   // computed a SCEV for this Op and Ty.
   FoldingSetNodeID ID;
   ID.AddInteger(scSignExtend);
   ID.AddPointer(Op);
   ID.AddPointer(Ty);
   void *IP = nullptr;
   if (const SCEV *S = UniqueSCEVs.FindNodeOrInsertPos(ID, IP)) return S;
+  // Limit recursion depth.
+  if (Depth > MaxExtDepth) {
+    SCEV *S = new (SCEVAllocator) SCEVSignExtendExpr(ID.Intern(SCEVAllocator),
+                                                     Op, Ty);
+    UniqueSCEVs.InsertNode(S, IP);
+    return S;
+  }
 
   // sext(trunc(x)) --> sext(x) or x or trunc(x)
   if (const SCEVTruncateExpr *ST = dyn_cast<SCEVTruncateExpr>(Op)) {
     // It's possible the bits taken off by the truncate were all sign bits. If
     // so, we should be able to simplify this further.
     const SCEV *X = ST->getOperand();
     ConstantRange CR = getSignedRange(X);
     unsigned TruncBits = getTypeSizeInBits(ST->getType());
if (SA->getNumOperands() == 2) {
       auto *SC1 = dyn_cast<SCEVConstant>(SA->getOperand(0));
       auto *SMul = dyn_cast<SCEVMulExpr>(SA->getOperand(1));
       if (SMul && SC1) {
         if (auto *SC2 = dyn_cast<SCEVConstant>(SMul->getOperand(0))) {
           const APInt &C1 = SC1->getAPInt();
           const APInt &C2 = SC2->getAPInt();
           if (C1.isStrictlyPositive() && C2.isStrictlyPositive() &&
               C2.ugt(C1) && C2.isPowerOf2())
-            return getAddExpr(getSignExtendExprCached(SC1, Ty, Cache),
-                              getSignExtendExprCached(SMul, Ty, Cache));
+            return getAddExpr(getSignExtendExpr(SC1, Ty, Depth + 1),
+                              getSignExtendExpr(SMul, Ty, Depth + 1),
+                              SCEV::FlagAnyWrap, Depth + 1);
         }
       }
     }
 
     // sext((A + B + ...)<nsw>) --> (sext(A) + sext(B) + ...)<nsw>
     if (SA->hasNoSignedWrap()) {
       // If the addition does not sign overflow then we can, by definition,
       // commute the sign extension with the addition operation.
       SmallVector<const SCEV *, 4> Ops;
       for (const auto *Op : SA->operands())
-        Ops.push_back(getSignExtendExprCached(Op, Ty, Cache));
-      return getAddExpr(Ops, SCEV::FlagNSW);
+        Ops.push_back(getSignExtendExpr(Op, Ty, Depth + 1));
+      return getAddExpr(Ops, SCEV::FlagNSW, Depth + 1);
     }
   }
   // If the input value is a chrec scev, and we can prove that the value
   // did not overflow the old, smaller, value, we can sign extend all of the
   // operands (often constants).  This allows analysis of something like
   // this:  for (signed char X = 0; X < 100; ++X) { int Y = X; }
   if (const SCEVAddRecExpr *AR = dyn_cast<SCEVAddRecExpr>(Op))
     if (AR->isAffine()) {
       const SCEV *Start = AR->getStart();
       const SCEV *Step = AR->getStepRecurrence(*this);
       unsigned BitWidth = getTypeSizeInBits(AR->getType());
       const Loop *L = AR->getLoop();
 
       if (!AR->hasNoSignedWrap()) {
         auto NewFlags = proveNoWrapViaConstantRanges(AR);
         const_cast<SCEVAddRecExpr *>(AR)->setNoWrapFlags(NewFlags);
       }
 
       // If we have special knowledge that this addrec won't overflow,
       // we don't need to do any further analysis.
       if (AR->hasNoSignedWrap())
         return getAddRecExpr(
-            getExtendAddRecStart<SCEVSignExtendExpr>(AR, Ty, this, Cache),
-            getSignExtendExprCached(Step, Ty, Cache), L, SCEV::FlagNSW);
+            getExtendAddRecStart<SCEVSignExtendExpr>(AR, Ty, this, Depth + 1),
+            getSignExtendExpr(Step, Ty, Depth + 1), L, SCEV::FlagNSW);
 
       // Check whether the backedge-taken count is SCEVCouldNotCompute.
       // Note that this serves two purposes: It filters out loops that are
       // simply not analyzable, and it covers the case where this code is
       // being called from within backedge-taken count analysis, such that
       // attempting to ask for the backedge-taken count would likely result
       // in infinite recursion. In the later case, the analysis code will
       // cope with a conservative value, and it will take care to purge
       // that value once it has finished.
       const SCEV *MaxBECount = getMaxBackedgeTakenCount(L);
       if (!isa<SCEVCouldNotCompute>(MaxBECount)) {
         // Manually compute the final value for AR, checking for
         // overflow.
 
         // Check whether the backedge-taken count can be losslessly casted to
         // the addrec's type. The count is always unsigned.
         const SCEV *CastedMaxBECount =
           getTruncateOrZeroExtend(MaxBECount, Start->getType());
         const SCEV *RecastedMaxBECount =
           getTruncateOrZeroExtend(CastedMaxBECount, MaxBECount->getType());
         if (MaxBECount == RecastedMaxBECount) {
           Type *WideTy = IntegerType::get(getContext(), BitWidth * 2);
           // Check whether Start+Step*MaxBECount has no signed overflow.
-          const SCEV *SMul = getMulExpr(CastedMaxBECount, Step);
-          const SCEV *SAdd =
-              getSignExtendExprCached(getAddExpr(Start, SMul), WideTy, Cache);
-          const SCEV *WideStart = getSignExtendExprCached(Start, WideTy, Cache);
+          const SCEV *SMul = getMulExpr(CastedMaxBECount, Step,
+                                        SCEV::FlagAnyWrap, Depth + 1);
+          const SCEV *SAdd = getSignExtendExpr(getAddExpr(Start, SMul,
+                                                          SCEV::FlagAnyWrap,
+                                                          Depth + 1),
+                                               WideTy, Depth + 1);
+          const SCEV *WideStart = getSignExtendExpr(Start, WideTy, Depth + 1);
           const SCEV *WideMaxBECount =
-              getZeroExtendExpr(CastedMaxBECount, WideTy);
-          const SCEV *OperandExtendedAdd = getAddExpr(
-              WideStart, getMulExpr(WideMaxBECount, getSignExtendExprCached(
-                                                        Step, WideTy, Cache)));
+            getZeroExtendExpr(CastedMaxBECount, WideTy, Depth + 1);
+          const SCEV *OperandExtendedAdd =
+            getAddExpr(WideStart,
+                       getMulExpr(WideMaxBECount,
+                                  getSignExtendExpr(Step, WideTy, Depth + 1),
+                                  SCEV::FlagAnyWrap, Depth + 1),
+                       SCEV::FlagAnyWrap, Depth + 1);
           if (SAdd == OperandExtendedAdd) {
             // Cache knowledge of AR NSW, which is propagated to this AddRec.
             const_cast<SCEVAddRecExpr *>(AR)->setNoWrapFlags(SCEV::FlagNSW);
             // Return the expression with the addrec on the outside.
             return getAddRecExpr(
-                getExtendAddRecStart<SCEVSignExtendExpr>(AR, Ty, this, Cache),
-                getSignExtendExprCached(Step, Ty, Cache), L,
+                getExtendAddRecStart<SCEVSignExtendExpr>(AR, Ty, this,
+                                                         Depth + 1),
+                getSignExtendExpr(Step, Ty, Depth + 1), L,
                 AR->getNoWrapFlags());
           }
           // Similar to above, only this time treat the step value as unsigned.
           // This covers loops that count up with an unsigned step.
           OperandExtendedAdd =
             getAddExpr(WideStart,
                        getMulExpr(WideMaxBECount,
-                                  getZeroExtendExpr(Step, WideTy)));
+                                  getZeroExtendExpr(Step, WideTy, Depth + 1),
+                                  SCEV::FlagAnyWrap, Depth + 1),
+                       SCEV::FlagAnyWrap, Depth + 1);
           if (SAdd == OperandExtendedAdd) {
             // If AR wraps around then
             //
             //    abs(Step) * MaxBECount > unsigned-max(AR->getType())
             // => SAdd != OperandExtendedAdd
             //
             // Thus (AR is not NW => SAdd != OperandExtendedAdd) <=>
             // (SAdd == OperandExtendedAdd => AR is NW)
 
             const_cast<SCEVAddRecExpr *>(AR)->setNoWrapFlags(SCEV::FlagNW);
 
             // Return the expression with the addrec on the outside.
             return getAddRecExpr(
-                getExtendAddRecStart<SCEVSignExtendExpr>(AR, Ty, this, Cache),
-                getZeroExtendExpr(Step, Ty), L, AR->getNoWrapFlags());
+                getExtendAddRecStart<SCEVSignExtendExpr>(AR, Ty, this,
+                                                         Depth + 1),
+                getZeroExtendExpr(Step, Ty, Depth + 1), L,
+                AR->getNoWrapFlags());
           }
         }
       }
 
       // Normally, in the cases we can prove no-overflow via a
       // backedge guarding condition, we can also compute a backedge
       // taken count for the loop.  The exceptions are assumptions and
       // guards present in the loop -- SCEV is not great at exploiting
if (AR->isAffine()) {
         if (OverflowLimit &&
             (isLoopBackedgeGuardedByCond(L, Pred, AR, OverflowLimit) ||
              (isLoopEntryGuardedByCond(L, Pred, Start, OverflowLimit) &&
               isLoopBackedgeGuardedByCond(L, Pred, AR->getPostIncExpr(*this),
                                           OverflowLimit)))) {
           // Cache knowledge of AR NSW, then propagate NSW to the wide AddRec.
           const_cast<SCEVAddRecExpr *>(AR)->setNoWrapFlags(SCEV::FlagNSW);
           return getAddRecExpr(
-              getExtendAddRecStart<SCEVSignExtendExpr>(AR, Ty, this, Cache),
-              getSignExtendExprCached(Step, Ty, Cache), L,
-              AR->getNoWrapFlags());
+              getExtendAddRecStart<SCEVSignExtendExpr>(AR, Ty, this, Depth + 1),
+              getSignExtendExpr(Step, Ty, Depth + 1), L, AR->getNoWrapFlags());
         }
       }
 
       // If Start and Step are constants, check if we can apply this
       // transformation:
       // sext{C1,+,C2} --> C1 + sext{0,+,C2} if C1 < C2
       auto *SC1 = dyn_cast<SCEVConstant>(Start);
       auto *SC2 = dyn_cast<SCEVConstant>(Step);
       if (SC1 && SC2) {
         const APInt &C1 = SC1->getAPInt();
         const APInt &C2 = SC2->getAPInt();
         if (C1.isStrictlyPositive() && C2.isStrictlyPositive() && C2.ugt(C1) &&
             C2.isPowerOf2()) {
-          Start = getSignExtendExprCached(Start, Ty, Cache);
+          Start = getSignExtendExpr(Start, Ty, Depth + 1);
           const SCEV *NewAR = getAddRecExpr(getZero(AR->getType()), Step, L,
                                             AR->getNoWrapFlags());
-          return getAddExpr(Start, getSignExtendExprCached(NewAR, Ty, Cache));
+          return getAddExpr(Start, getSignExtendExpr(NewAR, Ty, Depth + 1),
+                            SCEV::FlagAnyWrap, Depth + 1);
         }
       }
 
       if (proveNoWrapByVaryingStart<SCEVSignExtendExpr>(Start, Step, L)) {
         const_cast<SCEVAddRecExpr *>(AR)->setNoWrapFlags(SCEV::FlagNSW);
         return getAddRecExpr(
-            getExtendAddRecStart<SCEVSignExtendExpr>(AR, Ty, this, Cache),
-            getSignExtendExprCached(Step, Ty, Cache), L, AR->getNoWrapFlags());
+            getExtendAddRecStart<SCEVSignExtendExpr>(AR, Ty, this, Depth + 1),
+            getSignExtendExpr(Step, Ty, Depth + 1), L, AR->getNoWrapFlags());
       }
     }
 
   // If the input value is provably positive and we could not simplify
   // away the sext build a zext instead.
   if (isKnownNonNegative(Op))
-    return getZeroExtendExpr(Op, Ty);
+    return getZeroExtendExpr(Op, Ty, Depth + 1);
 
   // The cast wasn't folded; create an explicit cast node.
   // Recompute the insert position, as it may have been invalidated.
   if (const SCEV *S = UniqueSCEVs.FindNodeOrInsertPos(ID, IP)) return S;
   SCEV *S = new (SCEVAllocator) SCEVSignExtendExpr(ID.Intern(SCEVAllocator),
                                                    Op, Ty);
   UniqueSCEVs.InsertNode(S, IP);
   return S;

lib/Transforms/Utils/SimplifyIndVar.cpp

 bool SimplifyIndvar::eliminateOverflowIntrinsic(CallInst *CI) {
   auto *F = CI->getCalledFunction();
   if (!F)
     return false;
 
   typedef const SCEV *(ScalarEvolution::*OperationFunctionTy)(
       const SCEV *, const SCEV *, SCEV::NoWrapFlags, unsigned);
   typedef const SCEV *(ScalarEvolution::*ExtensionFunctionTy)(
-      const SCEV *, Type *);
+      const SCEV *, Type *, unsigned);
 
   OperationFunctionTy Operation;
   ExtensionFunctionTy Extension;
 
   Instruction::BinaryOps RawOp;
 
   // We always have exactly one of nsw or nuw.  If NoSignedOverflow is false, we
   // have nuw.
bool SimplifyIndvar::eliminateOverflowIntrinsic(CallInst *CI) {
   const SCEV *LHS = SE->getSCEV(CI->getArgOperand(0));
   const SCEV *RHS = SE->getSCEV(CI->getArgOperand(1));
 
   auto *NarrowTy = cast<IntegerType>(LHS->getType());
   auto *WideTy =
     IntegerType::get(NarrowTy->getContext(), NarrowTy->getBitWidth() * 2);
 
   const SCEV *A =
-      (SE->*Extension)((SE->*Operation)(LHS, RHS, SCEV::FlagAnyWrap, 0u),
-                       WideTy);
+      (SE->*Extension)((SE->*Operation)(LHS, RHS, SCEV::FlagAnyWrap, 0),
+                       WideTy, 0);
   const SCEV *B =
-      (SE->*Operation)((SE->*Extension)(LHS, WideTy),
-                       (SE->*Extension)(RHS, WideTy), SCEV::FlagAnyWrap, 0u);
+      (SE->*Operation)((SE->*Extension)(LHS, WideTy, 0),
+                       (SE->*Extension)(RHS, WideTy, 0), SCEV::FlagAnyWrap, 0);
 
   if (A != B)
     return false;
 
   // Proved no overflow, nuke the overflow check and, if possible, the overflow
   // intrinsic as well.
 
   BinaryOperator *NewResult = BinaryOperator::Create(

test/Analysis/ScalarEvolution/limit-depth.ll

-; RUN: opt -scalar-evolution-max-arith-depth=0 -analyze -scalar-evolution < %s | FileCheck %s
+; RUN: opt -scalar-evolution-max-arith-depth=0 -scalar-evolution-max-ext-depth=0 -analyze -scalar-evolution < %s | FileCheck %s
 
 ; Check that depth set to 0 prevents getAddExpr and getMulExpr from making
 ; transformations in SCEV. We expect the result to be very straightforward.
 
 define void @test_add(i32 %a, i32 %b, i32 %c, i32 %d, i32 %e, i32 %f) {
 ; CHECK-LABEL: @test_add
 ; CHECK:       %s2 = add i32 %s1, %p3
 ; CHECK-NEXT:   -->  (%a + %a + %b + %b + %c + %c + %d + %d + %e + %e + %f + %f)
; CHECK-NEXT:  -->  (2 * 3 * 4 * 5 * 6 * 7 * %a * %b * %c * %d * %e * %f)
   %p1 = mul i32 %tmp0, %tmp3
   %p2 = mul i32 %tmp1, %tmp4
   %p3 = mul i32 %tmp2, %tmp5
 
   %s1 = mul i32 %p1, %p2
   %s2 = mul i32 %s1, %p3
   ret void
 }
+
+define void @test_sext(i32 %a, i32 %b, i32 %c, i32 %d, i32 %e, i32 %f) {
+; CHECK-LABEL: @test_sext
+; CHECK:        %se2 = sext i64 %iv2.inc to i128
+; CHECK-NEXT:   -->  {(1 + (sext i64 {(sext i32 (1 + %a) to i64),+,1}<nsw><%loop> to i128))<nsw>,+,1}<nsw><%loop2>
+entry:
+  br label %loop
+
+loop:
+  %iv = phi i32 [ %a, %entry ], [ %iv.inc, %loop ]
+  %iv.inc = add nsw i32 %iv, 1
+  %cond = icmp sle i32 %iv.inc, 50
+  br i1 %cond, label %loop, label %between
+
+between:
+  %se = sext i32 %iv.inc to i64
+  br label %loop2
+
+loop2:
+  %iv2 = phi i64 [ %se, %between ], [ %iv2.inc, %loop2 ]
+  %iv2.inc = add nsw i64 %iv2, 1
+  %cond2 = icmp sle i64 %iv2.inc, 50
+  br i1 %cond2, label %loop2, label %exit
+
+exit:
+  %se2 = sext i64 %iv2.inc to i128
+  ret void
+}
+
+define void @test_zext(i32 %a, i32 %b, i32 %c, i32 %d, i32 %e, i32 %f) {
+; CHECK-LABEL: @test_zext
+; CHECK:          %ze2 = zext i64 %iv2.inc to i128
+; CHECK-NEXT:     -->  {(1 + (zext i64 {7,+,1}<nuw><nsw><%loop> to i128))<nuw><nsw>,+,1}<nuw><%loop2>
+entry:
+  br label %loop
+
+loop:
+  %iv = phi i32 [ 6, %entry ], [ %iv.inc, %loop ]
+  %iv.inc = add nsw i32 %iv, 1
+  %cond = icmp sle i32 %iv.inc, 50
+  br i1 %cond, label %loop, label %between
+
+between:
+  %ze = zext i32 %iv.inc to i64
+  br label %loop2
+
+loop2:
+  %iv2 = phi i64 [ %ze, %between ], [ %iv2.inc, %loop2 ]
+  %iv2.inc = add nuw i64 %iv2, 1
+  %cond2 = icmp sle i64 %iv2.inc, 50
+  br i1 %cond2, label %loop2, label %exit
+
+exit:
+  %ze2 = zext i64 %iv2.inc to i128
+  ret void
+}