[IRCE] Recognize loops with unsigned latch conditions

[mkazantsev]

Summary

This patch enables recognition of loops with ult/ugt latch conditions.

[mkazantsev]

lib/Transforms/Scalar/InductiveRangeCheckElimination.cpp

1020 ↗

(On Diff #106189)

This is actually a bug that also exists for start = SINT_MAX even without this patch. I'm going to fix it separately.

[mkazantsev]

Will rebase on top of the fix for the corner case problem.

[mkazantsev]

lib/Transforms/Scalar/InductiveRangeCheckElimination.cpp

1020 ↗

(On Diff #106189)

Fix is https://reviews.llvm.org/D35347

[mkazantsev]

Fixed an underlying bug with borders, test_12 have been re-enabled.

[anna]

Couple of minor changes. I haven't looked at the tests yet.

Do you mind separating out the ULT/ULE changes from the UGT/UGE just for ease of reviewing? Once the first part is reviewed, the second would be pretty straightforward.

lib/Transforms/Scalar/InductiveRangeCheckElimination.cpp

462 ↗

(On Diff #106395)

Pls rename this as isSignedPredicate.

846 ↗

(On Diff #106395)

Pls name this as the PredOpCode. I got confused with guards..

978 ↗

(On Diff #106395)

This to isSignedPredicate as well. Similarly for all other booleans.

[mkazantsev]

Reworked tests, made some refactoring, addressed TODOs.

[mkazantsev]

lib/Transforms/Scalar/InductiveRangeCheckElimination.cpp

846 ↗

(On Diff #106395)

Renamed to BoundPred to avoid confusion. Semantically it is a predicate we need to check to make sure that the indvar lies within bounds.

[anna]

Looks almost ready to go. Some comments regarding additional tests.

lib/Transforms/Scalar/InductiveRangeCheckElimination.cpp

1228 ↗

(On Diff #109068)

Nit: set to nullptr.

1248 ↗

(On Diff #109068)

set to nullptr.

test/Transforms/IRCE/unsigned_comparisons_ugt.ll

9 ↗

(On Diff #109068)

Please add some tests or checks for post loops as well. That code's not tested here.

test/Transforms/IRCE/unsigned_comparisons_ult.ll

10 ↗

(On Diff #109068)

Please add some tests or checks for post loops as well. That code's not tested here.

[mkazantsev]

Addressed nits, reworked tests so that now they check the structure of transformed loop, not only the fact that the transformation has happened.

[mkazantsev]

test/Transforms/IRCE/eq_ne.ll

18 ↗

(On Diff #109512)

Also need to update the tests in the same manner for this file. I will do it later in a separate patch.

[anna]

LGTM.

lib/Transforms/Scalar/InductiveRangeCheckElimination.cpp

1213 ↗

(On Diff #109512)

Set to nullptr here as well.

llvm/trunk/lib/Transforms/Scalar/InductiveRangeCheckElimination.cpp

struct LoopStructure {
   //
   // for (intN_ty iv = IndVarStart; predicate(iv, LoopExitAt); iv = IndVarNext)
   //   ... body ...
 
   Value *IndVarNext;
   Value *IndVarStart;
   Value *LoopExitAt;
   bool IndVarIncreasing;
+  bool IsSignedPredicate;
 
   LoopStructure()
       : Tag(""), Header(nullptr), Latch(nullptr), LatchBr(nullptr),
         LatchExit(nullptr), LatchBrExitIdx(-1), IndVarNext(nullptr),
-        IndVarStart(nullptr), LoopExitAt(nullptr), IndVarIncreasing(false) {}
+        IndVarStart(nullptr), LoopExitAt(nullptr), IndVarIncreasing(false),
+        IsSignedPredicate(true) {}
 
   template <typename M> LoopStructure map(M Map) const {
     LoopStructure Result;
     Result.Tag = Tag;
     Result.Header = cast<BasicBlock>(Map(Header));
     Result.Latch = cast<BasicBlock>(Map(Latch));
     Result.LatchBr = cast<BranchInst>(Map(LatchBr));
     Result.LatchExit = cast<BasicBlock>(Map(LatchExit));
     Result.LatchBrExitIdx = LatchBrExitIdx;
     Result.IndVarNext = Map(IndVarNext);
     Result.IndVarStart = Map(IndVarStart);
     Result.LoopExitAt = Map(LoopExitAt);
     Result.IndVarIncreasing = IndVarIncreasing;
+    Result.IsSignedPredicate = IsSignedPredicate;
     return Result;
   }
 
   static Optional<LoopStructure> parseLoopStructure(ScalarEvolution &,
                                                     BranchProbabilityInfo &BPI,
                                                     Loop &,
                                                     const char *&);
 };
class LoopConstrainer {
   // incoming block list with `ReplaceBy'.
   static void replacePHIBlock(PHINode *PN, BasicBlock *Block,
                               BasicBlock *ReplaceBy);
 
   // Compute a safe set of limits for the main loop to run in -- effectively the
   // intersection of `Range' and the iteration space of the original loop.
   // Return None if unable to compute the set of subranges.
   //
-  Optional<SubRanges> calculateSubRanges() const;
+  Optional<SubRanges> calculateSubRanges(bool IsSignedPredicate) const;
 
   // Clone `OriginalLoop' and return the result in CLResult.  The IR after
   // running `cloneLoop' is well formed except for the PHI nodes in CLResult --
   // the PHI nodes say that there is an incoming edge from `OriginalPreheader`
   // but there is no such edge.
   //
   void cloneLoop(ClonedLoop &CLResult, const char *Tag) const;
 

 
 void LoopConstrainer::replacePHIBlock(PHINode *PN, BasicBlock *Block,
                                       BasicBlock *ReplaceBy) {
   for (unsigned i = 0, e = PN->getNumIncomingValues(); i != e; ++i)
     if (PN->getIncomingBlock(i) == Block)
       PN->setIncomingBlock(i, ReplaceBy);
 }
 
-static bool CanBeSMax(ScalarEvolution &SE, const SCEV *S) {
-  APInt SMax =
-      APInt::getSignedMaxValue(cast<IntegerType>(S->getType())->getBitWidth());
-  return SE.getSignedRange(S).contains(SMax) &&
-         SE.getUnsignedRange(S).contains(SMax);
+static bool CanBeMax(ScalarEvolution &SE, const SCEV *S, bool Signed) {
+  APInt Max = Signed ?
+      APInt::getSignedMaxValue(cast<IntegerType>(S->getType())->getBitWidth()) :
+      APInt::getMaxValue(cast<IntegerType>(S->getType())->getBitWidth());
+  return SE.getSignedRange(S).contains(Max) &&
+         SE.getUnsignedRange(S).contains(Max);
 }
 
-static bool CanBeSMin(ScalarEvolution &SE, const SCEV *S) {
-  APInt SMin =
-      APInt::getSignedMinValue(cast<IntegerType>(S->getType())->getBitWidth());
-  return SE.getSignedRange(S).contains(SMin) &&
-         SE.getUnsignedRange(S).contains(SMin);
+static bool CanBeMin(ScalarEvolution &SE, const SCEV *S, bool Signed) {
+  APInt Min = Signed ?
+      APInt::getSignedMinValue(cast<IntegerType>(S->getType())->getBitWidth()) :
+      APInt::getMinValue(cast<IntegerType>(S->getType())->getBitWidth());
+  return SE.getSignedRange(S).contains(Min) &&
+         SE.getUnsignedRange(S).contains(Min);
 }
 
 Optional<LoopStructure>
-LoopStructure::parseLoopStructure(ScalarEvolution &SE, BranchProbabilityInfo &BPI,
+LoopStructure::parseLoopStructure(ScalarEvolution &SE,
+                                  BranchProbabilityInfo &BPI,
                                   Loop &L, const char *&FailureReason) {
   if (!L.isLoopSimplifyForm()) {
     FailureReason = "loop not in LoopSimplify form";
     return None;
   }
 
   BasicBlock *Latch = L.getLoopLatch();
   assert(Latch && "Simplified loops only have one latch!");
auto IsInductionVar = [&](const SCEVAddRecExpr *AR, bool &IsIncreasing) {
     return false;
   };
 
   // `ICI` is interpreted as taking the backedge if the *next* value of the
   // induction variable satisfies some constraint.
 
   const SCEVAddRecExpr *IndVarNext = cast<SCEVAddRecExpr>(LeftSCEV);
   bool IsIncreasing = false;
+  bool IsSignedPredicate = true;
   if (!IsInductionVar(IndVarNext, IsIncreasing)) {
     FailureReason = "LHS in icmp not induction variable";
     return None;
   }
 
   const SCEV *StartNext = IndVarNext->getStart();
   const SCEV *Addend = SE.getNegativeSCEV(IndVarNext->getStepRecurrence(SE));
   const SCEV *IndVarStart = SE.getAddExpr(StartNext, Addend);
 
   ConstantInt *One = ConstantInt::get(IndVarTy, 1);
-  // TODO: generalize the predicates here to also match their unsigned variants.
   if (IsIncreasing) {
     bool DecreasedRightValueByOne = false;
     // Try to turn eq/ne predicates to those we can work with.
     if (Pred == ICmpInst::ICMP_NE && LatchBrExitIdx == 1)
       // while (++i != len) {         while (++i < len) {
       //   ...                 --->     ...
       // }                            }
+      // If both parts are known non-negative, it is profitable to use unsigned
+      // comparison in increasing loop. This allows us to make the comparison
+      // check against "RightSCEV + 1" more optimistic.
+      if (SE.isKnownNonNegative(IndVarStart) &&
+          SE.isKnownNonNegative(RightSCEV))
+        Pred = ICmpInst::ICMP_ULT;
+      else
         Pred = ICmpInst::ICMP_SLT;
     else if (Pred == ICmpInst::ICMP_EQ && LatchBrExitIdx == 0 &&
-             !CanBeSMin(SE, RightSCEV)) {
+             !CanBeMin(SE, RightSCEV, /* IsSignedPredicate */ true)) {
       // while (true) {               while (true) {
       //   if (++i == len)     --->     if (++i > len - 1)
       //     break;                       break;
       //   ...                          ...
       // }                            }
+      // TODO: Insert ICMP_UGT if both are non-negative?
       Pred = ICmpInst::ICMP_SGT;
       RightSCEV = SE.getMinusSCEV(RightSCEV, SE.getOne(RightSCEV->getType()));
       DecreasedRightValueByOne = true;
     }
 
+    bool LTPred = (Pred == ICmpInst::ICMP_SLT || Pred == ICmpInst::ICMP_ULT);
+    bool GTPred = (Pred == ICmpInst::ICMP_SGT || Pred == ICmpInst::ICMP_UGT);
     bool FoundExpectedPred =
-        (Pred == ICmpInst::ICMP_SLT && LatchBrExitIdx == 1) ||
-        (Pred == ICmpInst::ICMP_SGT && LatchBrExitIdx == 0);
+        (LTPred && LatchBrExitIdx == 1) || (GTPred && LatchBrExitIdx == 0);
 
     if (!FoundExpectedPred) {
       FailureReason = "expected icmp slt semantically, found something else";
       return None;
     }
 
+    IsSignedPredicate =
+        Pred == ICmpInst::ICMP_SLT || Pred == ICmpInst::ICMP_SGT;
+    // The predicate that we need to check that the induction variable lies
+    // within bounds.
+    ICmpInst::Predicate BoundPred =
+        IsSignedPredicate ? CmpInst::ICMP_SLT : CmpInst::ICMP_ULT;
+
     if (LatchBrExitIdx == 0) {
-      if (CanBeSMax(SE, RightSCEV)) {
+      if (CanBeMax(SE, RightSCEV, IsSignedPredicate)) {
         // TODO: this restriction is easily removable -- we just have to
         // remember that the icmp was an slt and not an sle.
-        FailureReason = "limit may overflow when coercing sle to slt";
+        FailureReason = "limit may overflow when coercing le to lt";
         return None;
       }
 
       if (!SE.isLoopEntryGuardedByCond(
-              &L, CmpInst::ICMP_SLT, IndVarStart,
+              &L, BoundPred, IndVarStart,
               SE.getAddExpr(RightSCEV, SE.getOne(RightSCEV->getType())))) {
         FailureReason = "Induction variable start not bounded by upper limit";
         return None;
       }
 
       // We need to increase the right value unless we have already decreased
       // it virtually when we replaced EQ with SGT.
       if (!DecreasedRightValueByOne) {
         IRBuilder<> B(Preheader->getTerminator());
         RightValue = B.CreateAdd(RightValue, One);
       }
     } else {
-      if (!SE.isLoopEntryGuardedByCond(&L, CmpInst::ICMP_SLT, IndVarStart,
-                                       RightSCEV)) {
+      if (!SE.isLoopEntryGuardedByCond(&L, BoundPred, IndVarStart, RightSCEV)) {
         FailureReason = "Induction variable start not bounded by upper limit";
         return None;
       }
       assert(!DecreasedRightValueByOne &&
              "Right value can be decreased only for LatchBrExitIdx == 0!");
     }
   } else {
     bool IncreasedRightValueByOne = false;
     // Try to turn eq/ne predicates to those we can work with.
     if (Pred == ICmpInst::ICMP_NE && LatchBrExitIdx == 1)
       // while (--i != len) {         while (--i > len) {
       //   ...                 --->     ...
       // }                            }
+      // We intentionally don't turn the predicate into UGT even if we know that
+      // both operands are non-negative, because it will only pessimize our
+      // check against "RightSCEV - 1".
       Pred = ICmpInst::ICMP_SGT;
     else if (Pred == ICmpInst::ICMP_EQ && LatchBrExitIdx == 0 &&
-             !CanBeSMax(SE, RightSCEV)) {
+             !CanBeMax(SE, RightSCEV, /* IsSignedPredicate */ true)) {
       // while (true) {               while (true) {
       //   if (--i == len)     --->     if (--i < len + 1)
       //     break;                       break;
       //   ...                          ...
       // }                            }
+      // TODO: Insert ICMP_ULT if both are non-negative?
       Pred = ICmpInst::ICMP_SLT;
       RightSCEV = SE.getAddExpr(RightSCEV, SE.getOne(RightSCEV->getType()));
       IncreasedRightValueByOne = true;
     }
 
+    bool LTPred = (Pred == ICmpInst::ICMP_SLT || Pred == ICmpInst::ICMP_ULT);
+    bool GTPred = (Pred == ICmpInst::ICMP_SGT || Pred == ICmpInst::ICMP_UGT);
+
     bool FoundExpectedPred =
-        (Pred == ICmpInst::ICMP_SGT && LatchBrExitIdx == 1) ||
-        (Pred == ICmpInst::ICMP_SLT && LatchBrExitIdx == 0);
+        (GTPred && LatchBrExitIdx == 1) || (LTPred && LatchBrExitIdx == 0);
 
     if (!FoundExpectedPred) {
       FailureReason = "expected icmp sgt semantically, found something else";
       return None;
     }
 
+    IsSignedPredicate =
+        Pred == ICmpInst::ICMP_SLT || Pred == ICmpInst::ICMP_SGT;
+    // The predicate that we need to check that the induction variable lies
+    // within bounds.
+    ICmpInst::Predicate BoundPred =
+        IsSignedPredicate ? CmpInst::ICMP_SGT : CmpInst::ICMP_UGT;
+
     if (LatchBrExitIdx == 0) {
-      if (CanBeSMin(SE, RightSCEV)) {
+      if (CanBeMin(SE, RightSCEV, IsSignedPredicate)) {
         // TODO: this restriction is easily removable -- we just have to
         // remember that the icmp was an sgt and not an sge.
-        FailureReason = "limit may overflow when coercing sge to sgt";
+        FailureReason = "limit may overflow when coercing ge to gt";
         return None;
       }
 
       if (!SE.isLoopEntryGuardedByCond(
-              &L, CmpInst::ICMP_SGT, IndVarStart,
+              &L, BoundPred, IndVarStart,
               SE.getMinusSCEV(RightSCEV, SE.getOne(RightSCEV->getType())))) {
         FailureReason = "Induction variable start not bounded by lower limit";
         return None;
       }
 
       // We need to decrease the right value unless we have already increased
       // it virtually when we replaced EQ with SLT.
       if (!IncreasedRightValueByOne) {
         IRBuilder<> B(Preheader->getTerminator());
         RightValue = B.CreateSub(RightValue, One);
       }
     } else {
-      if (!SE.isLoopEntryGuardedByCond(&L, CmpInst::ICMP_SGT, IndVarStart,
-                                       RightSCEV)) {
+      if (!SE.isLoopEntryGuardedByCond(&L, BoundPred, IndVarStart, RightSCEV)) {
         FailureReason = "Induction variable start not bounded by lower limit";
         return None;
       }
       assert(!IncreasedRightValueByOne &&
              "Right value can be increased only for LatchBrExitIdx == 0!");
     }
   }
-
   BasicBlock *LatchExit = LatchBr->getSuccessor(LatchBrExitIdx);
 
   assert(SE.getLoopDisposition(LatchCount, &L) ==
              ScalarEvolution::LoopInvariant &&
          "loop variant exit count doesn't make sense!");
 
   assert(!L.contains(LatchExit) && "expected an exit block!");
   const DataLayout &DL = Preheader->getModule()->getDataLayout();
LoopStructure::parseLoopStructure(ScalarEvolution &SE,
   Result.Latch = Latch;
   Result.LatchBr = LatchBr;
   Result.LatchExit = LatchExit;
   Result.LatchBrExitIdx = LatchBrExitIdx;
   Result.IndVarStart = IndVarStartV;
   Result.IndVarNext = LeftValue;
   Result.IndVarIncreasing = IsIncreasing;
   Result.LoopExitAt = RightValue;
+  Result.IsSignedPredicate = IsSignedPredicate;
 
   FailureReason = nullptr;
 
   return Result;
 }
 
 Optional<LoopConstrainer::SubRanges>
-LoopConstrainer::calculateSubRanges() const {
+LoopConstrainer::calculateSubRanges(bool IsSignedPredicate) const {
   IntegerType *Ty = cast<IntegerType>(LatchTakenCount->getType());
 
   if (Range.getType() != Ty)
     return None;
 
   LoopConstrainer::SubRanges Result;
 
   // I think we can be more aggressive here and make this nuw / nsw if the
if (Increasing) {
     //    will be an empty range.  Returning an empty range is always safe.
     //
 
     Smallest = SE.getAddExpr(End, One);
     Greatest = SE.getAddExpr(Start, One);
     GreatestSeen = Start;
   }
 
-  auto Clamp = [this, Smallest, Greatest](const SCEV *S) {
-    return SE.getSMaxExpr(Smallest, SE.getSMinExpr(Greatest, S));
+  auto Clamp = [this, Smallest, Greatest, IsSignedPredicate](const SCEV *S) {
+    return IsSignedPredicate
+               ? SE.getSMaxExpr(Smallest, SE.getSMinExpr(Greatest, S))
+               : SE.getUMaxExpr(Smallest, SE.getUMinExpr(Greatest, S));
   };
 
-  // In some cases we can prove that we don't need a pre or post loop
+  // In some cases we can prove that we don't need a pre or post loop.
+  ICmpInst::Predicate PredLE =
+      IsSignedPredicate ? ICmpInst::ICMP_SLE : ICmpInst::ICMP_ULE;
+  ICmpInst::Predicate PredLT =
+      IsSignedPredicate ? ICmpInst::ICMP_SLT : ICmpInst::ICMP_ULT;
 
   bool ProvablyNoPreloop =
-      SE.isKnownPredicate(ICmpInst::ICMP_SLE, Range.getBegin(), Smallest);
+      SE.isKnownPredicate(PredLE, Range.getBegin(), Smallest);
   if (!ProvablyNoPreloop)
     Result.LowLimit = Clamp(Range.getBegin());
 
   bool ProvablyNoPostLoop =
-      SE.isKnownPredicate(ICmpInst::ICMP_SLT, GreatestSeen, Range.getEnd());
+      SE.isKnownPredicate(PredLT, GreatestSeen, Range.getEnd());
   if (!ProvablyNoPostLoop)
     Result.HighLimit = Clamp(Range.getEnd());
 
   return Result;
 }
 
 void LoopConstrainer::cloneLoop(LoopConstrainer::ClonedLoop &Result,
                                 const char *Tag) const {
LoopConstrainer::RewrittenRangeInfo LoopConstrainer::changeIterationSpaceEnd(
   BasicBlock *BBInsertLocation = LS.Latch->getNextNode();
   RRI.ExitSelector = BasicBlock::Create(Ctx, Twine(LS.Tag) + ".exit.selector",
                                         &F, BBInsertLocation);
   RRI.PseudoExit = BasicBlock::Create(Ctx, Twine(LS.Tag) + ".pseudo.exit", &F,
                                       BBInsertLocation);
 
   BranchInst *PreheaderJump = cast<BranchInst>(Preheader->getTerminator());
   bool Increasing = LS.IndVarIncreasing;
+  bool IsSignedPredicate = LS.IsSignedPredicate;
 
   IRBuilder<> B(PreheaderJump);
 
   // EnterLoopCond - is it okay to start executing this `LS'?
-  Value *EnterLoopCond = Increasing
-                             ? B.CreateICmpSLT(LS.IndVarStart, ExitSubloopAt)
-                             : B.CreateICmpSGT(LS.IndVarStart, ExitSubloopAt);
+  Value *EnterLoopCond = nullptr;
+  if (Increasing)
+    EnterLoopCond = IsSignedPredicate
+                        : B.CreateICmpULT(LS.IndVarStart, ExitSubloopAt);
+  else
+    EnterLoopCond = IsSignedPredicate
+                        ? B.CreateICmpSGT(LS.IndVarStart, ExitSubloopAt)
+                        : B.CreateICmpUGT(LS.IndVarStart, ExitSubloopAt);
 
   B.CreateCondBr(EnterLoopCond, LS.Header, RRI.PseudoExit);
   PreheaderJump->eraseFromParent();
 
   LS.LatchBr->setSuccessor(LS.LatchBrExitIdx, RRI.ExitSelector);
   B.SetInsertPoint(LS.LatchBr);
-  Value *TakeBackedgeLoopCond =
-      Increasing ? B.CreateICmpSLT(LS.IndVarNext, ExitSubloopAt)
-                 : B.CreateICmpSGT(LS.IndVarNext, ExitSubloopAt);
+  Value *TakeBackedgeLoopCond = nullptr;
+  if (Increasing)
+    TakeBackedgeLoopCond = IsSignedPredicate
+                        ? B.CreateICmpSLT(LS.IndVarNext, ExitSubloopAt)
+                        : B.CreateICmpULT(LS.IndVarNext, ExitSubloopAt);
+  else
+    TakeBackedgeLoopCond = IsSignedPredicate
+                        ? B.CreateICmpSGT(LS.IndVarNext, ExitSubloopAt)
+                        : B.CreateICmpUGT(LS.IndVarNext, ExitSubloopAt);
   Value *CondForBranch = LS.LatchBrExitIdx == 1
                              ? TakeBackedgeLoopCond
                              : B.CreateNot(TakeBackedgeLoopCond);
 
   LS.LatchBr->setCondition(CondForBranch);
 
   B.SetInsertPoint(RRI.ExitSelector);
 
   // IterationsLeft - are there any more iterations left, given the original
   // upper bound on the induction variable?  If not, we branch to the "real"
   // exit.
-  Value *IterationsLeft = Increasing
-                              ? B.CreateICmpSLT(LS.IndVarNext, LS.LoopExitAt)
-                              : B.CreateICmpSGT(LS.IndVarNext, LS.LoopExitAt);
+  Value *IterationsLeft = nullptr;
+  if (Increasing)
+    IterationsLeft = IsSignedPredicate
+                         : B.CreateICmpULT(LS.IndVarNext, LS.LoopExitAt);
+  else
+    IterationsLeft = IsSignedPredicate
+                         ? B.CreateICmpSGT(LS.IndVarNext, LS.LoopExitAt)
+                         : B.CreateICmpUGT(LS.IndVarNext, LS.LoopExitAt);
   B.CreateCondBr(IterationsLeft, RRI.PseudoExit, LS.LatchExit);
 
   BranchInst *BranchToContinuation =
       BranchInst::Create(ContinuationBlock, RRI.PseudoExit);
 
   // We emit PHI nodes into `RRI.PseudoExit' that compute the "latest" value of
   // each of the PHI nodes in the loop header.  This feeds into the initial
   // value of the same PHI nodes if/when we continue execution.
bool LoopConstrainer::run() {
   LatchTakenCount = SE.getExitCount(&OriginalLoop, MainLoopStructure.Latch);
   Preheader = OriginalLoop.getLoopPreheader();
   assert(!isa<SCEVCouldNotCompute>(LatchTakenCount) && Preheader != nullptr &&
          "preconditions!");
 
   OriginalPreheader = Preheader;
   MainLoopPreheader = Preheader;
 
-  Optional<SubRanges> MaybeSR = calculateSubRanges();
+  bool IsSignedPredicate = MainLoopStructure.IsSignedPredicate;
+  Optional<SubRanges> MaybeSR = calculateSubRanges(IsSignedPredicate);
   if (!MaybeSR.hasValue()) {
     DEBUG(dbgs() << "irce: could not compute subranges\n");
     return false;
   }
 
   SubRanges SR = MaybeSR.getValue();
   bool Increasing = MainLoopStructure.IndVarIncreasing;
   IntegerType *IVTy =
const SCEVConstant *MinusOneS =
       cast<SCEVConstant>(SE.getConstant(IVTy, -1, true /* isSigned */));
 
   if (NeedsPreLoop) {
     const SCEV *ExitPreLoopAtSCEV = nullptr;
 
     if (Increasing)
       ExitPreLoopAtSCEV = *SR.LowLimit;
     else {
-      if (CanBeSMin(SE, *SR.HighLimit)) {
+      if (CanBeMin(SE, *SR.HighLimit, IsSignedPredicate)) {
         DEBUG(dbgs() << "irce: could not prove no-overflow when computing "
                      << "preloop exit limit.  HighLimit = " << *(*SR.HighLimit)
                      << "\n");
         return false;
       }
       ExitPreLoopAtSCEV = SE.getAddExpr(*SR.HighLimit, MinusOneS);
     }
 
     ExitPreLoopAt = Expander.expandCodeFor(ExitPreLoopAtSCEV, IVTy, InsertPt);
     ExitPreLoopAt->setName("exit.preloop.at");
   }
 
   if (NeedsPostLoop) {
     const SCEV *ExitMainLoopAtSCEV = nullptr;
 
     if (Increasing)
       ExitMainLoopAtSCEV = *SR.HighLimit;
     else {
-      if (CanBeSMin(SE, *SR.LowLimit)) {
+      if (CanBeMin(SE, *SR.LowLimit, IsSignedPredicate)) {
         DEBUG(dbgs() << "irce: could not prove no-overflow when computing "
                      << "mainloop exit limit.  LowLimit = " << *(*SR.LowLimit)
                      << "\n");
         return false;
       }
       ExitMainLoopAtSCEV = SE.getAddExpr(*SR.LowLimit, MinusOneS);
     }
 

llvm/trunk/test/Transforms/IRCE/eq_ne.ll

 ; RUN: opt -verify-loop-info -irce-print-changed-loops -irce -S < %s 2>&1 | FileCheck %s
 
 ; CHECK: irce: in function test_01: constrained Loop at depth 1 containing: %loop<header><exiting>,%in.bounds<latch><exiting>
+; CHECK: irce: in function test_01u: constrained Loop at depth 1 containing: %loop<header><exiting>,%in.bounds<latch><exiting>
 ; CHECK-NOT: irce: in function test_02: constrained Loop at depth 1 containing: %loop<header><exiting>,%in.bounds<latch><exiting>
 ; CHECK: irce: in function test_03: constrained Loop at depth 1 containing: %loop<header><exiting>,%in.bounds<latch><exiting>
 ; CHECK-NOT: irce: in function test_04: constrained Loop at depth 1 containing: %loop<header><exiting>,%in.bounds<latch><exiting>
 ; CHECK: irce: in function test_05: constrained Loop at depth 1 containing: %loop<header><exiting>,%in.bounds<latch><exiting>
 ; CHECK-NOT: irce: in function test_06: constrained Loop at depth 1 containing: %loop<header><exiting>,%in.bounds<latch><exiting>
 ; CHECK: irce: in function test_07: constrained Loop at depth 1 containing: %loop<header><exiting>,%in.bounds<latch><exiting>
 ; CHECK-NOT: irce: in function test_08: constrained Loop at depth 1 containing: %loop<header><exiting>,%in.bounds<latch><exiting>
 
-; Show that IRCE can turn 'ne' condition to 'slt' in increasing IV.
+; Show that IRCE can turn 'ne' condition to 'slt' in increasing IV when the IV
+; can be negative at some point.
 define void @test_01(i32* %arr, i32* %a_len_ptr) #0 {
 
 ; CHECK:      test_01
 ; CHECK:        main.exit.selector:
 ; CHECK-NEXT:     [[PSEUDO_PHI:%[^ ]+]] = phi i32 [ %idx.next, %in.bounds ]
 ; CHECK-NEXT:     [[COND:%[^ ]+]] = icmp slt i32 [[PSEUDO_PHI]], 100
 ; CHECK-NEXT:     br i1 [[COND]]
 
 entry:
   %len = load i32, i32* %a_len_ptr, !range !0
   br label %loop
 
 loop:
+  %idx = phi i32 [ -3, %entry ], [ %idx.next, %in.bounds ]
+  %idx.next = add i32 %idx, 1
+  %abc = icmp slt i32 %idx, %len
+  br i1 %abc, label %in.bounds, label %out.of.bounds
+
+in.bounds:
+  %addr = getelementptr i32, i32* %arr, i32 %idx
+  store i32 0, i32* %addr
+  %next = icmp ne i32 %idx.next, 100
+  br i1 %next, label %loop, label %exit
+
+out.of.bounds:
+  ret void
+
+exit:
+  ret void
+}
+
+; Show that IRCE can turn 'ne' condition to 'ult' in increasing IV when IV is
+; non-negative.
+define void @test_01u(i32* %arr, i32* %a_len_ptr) #0 {
+
+; CHECK:      test_01u
+; CHECK:        main.exit.selector:
+; CHECK-NEXT:     [[PSEUDO_PHI:%[^ ]+]] = phi i32 [ %idx.next, %in.bounds ]
+; CHECK-NEXT:     [[COND:%[^ ]+]] = icmp ult i32 [[PSEUDO_PHI]], 100
+; CHECK-NEXT:     br i1 [[COND]]
+
+entry:
+  %len = load i32, i32* %a_len_ptr, !range !0
+  br label %loop
+
+loop:
   %idx = phi i32 [ 0, %entry ], [ %idx.next, %in.bounds ]
   %idx.next = add i32 %idx, 1
   %abc = icmp slt i32 %idx, %len
   br i1 %abc, label %in.bounds, label %out.of.bounds
 
 in.bounds:
   %addr = getelementptr i32, i32* %arr, i32 %idx
   store i32 0, i32* %addr

llvm/trunk/test/Transforms/IRCE/unsigned_comparisons_ugt.ll

+; RUN: opt -verify-loop-info -irce-print-changed-loops -irce -S < %s 2>&1 | FileCheck %s
+
+; CHECK: irce: in function test_01: constrained Loop at depth 1 containing: %loop<header><exiting>,%in.bounds<latch><exiting>
+; CHECK: irce: in function test_02: constrained Loop at depth 1 containing: %loop<header><exiting>,%in.bounds<latch><exiting>
+; CHECK: irce: in function test_03: constrained Loop at depth 1 containing: %loop<header><exiting>,%in.bounds<latch><exiting>
+; CHECK: irce: in function test_04: constrained Loop at depth 1 containing: %loop<header><exiting>,%in.bounds<latch><exiting>
+; CHECK-NOT: irce: in function test_05: constrained Loop at depth 1 containing: %loop<header><exiting>,%in.bounds<latch><exiting>
+; CHECK: irce: in function test_06: constrained Loop at depth 1 containing: %loop<header><exiting>,%in.bounds<latch><exiting>
+
+; UGT condition for increasing loop.
+define void @test_01(i32* %arr, i32* %a_len_ptr) #0 {
+
+; CHECK: test_01(
+; CHECK:        entry:
+; CHECK-NEXT:     %exit.mainloop.at = load i32, i32* %a_len_ptr, !range !0
+; CHECK-NEXT:     [[COND:%[^ ]+]] = icmp ult i32 0, %exit.mainloop.at
+; CHECK-NEXT:     br i1 [[COND]], label %loop.preheader, label %main.pseudo.exit
+; CHECK:        loop:
+; CHECK-NEXT:     %idx = phi i32 [ %idx.next, %in.bounds ], [ 0, %loop.preheader ]
+; CHECK-NEXT:     %idx.next = add nuw nsw i32 %idx, 1
+; CHECK-NEXT:     %abc = icmp ult i32 %idx, %exit.mainloop.at
+; CHECK-NEXT:     br i1 true, label %in.bounds, label %out.of.bounds.loopexit1
+; CHECK-NOT:    loop.preloop:
+; CHECK:        loop.postloop:
+; CHECK-NEXT:     %idx.postloop = phi i32 [ %idx.copy, %postloop ], [ %idx.next.postloop, %in.bounds.postloop ]
+; CHECK-NEXT:     %idx.next.postloop = add nuw nsw i32 %idx.postloop, 1
+; CHECK-NEXT:     %abc.postloop = icmp ult i32 %idx.postloop, %exit.mainloop.at
+; CHECK-NEXT:     br i1 %abc.postloop, label %in.bounds.postloop, label %out.of.bounds.loopexit
+
+entry:
+  %len = load i32, i32* %a_len_ptr, !range !0
+  br label %loop
+
+loop:
+  %idx = phi i32 [ 0, %entry ], [ %idx.next, %in.bounds ]
+  %idx.next = add nsw nuw i32 %idx, 1
+  %abc = icmp ult i32 %idx, %len
+  br i1 %abc, label %in.bounds, label %out.of.bounds
+
+in.bounds:
+  %addr = getelementptr i32, i32* %arr, i32 %idx
+  store i32 0, i32* %addr
+  %next = icmp ugt i32 %idx.next, 100
+  br i1 %next, label %exit, label %loop
+
+out.of.bounds:
+  ret void
+
+exit:
+  ret void
+}
+
+; UGT condition for decreasing loop.
+define void @test_02(i32* %arr, i32* %a_len_ptr) #0 {
+
+; CHECK: test_02(
+; CHECK:        entry:
+; CHECK-NEXT:     %len = load i32, i32* %a_len_ptr, !range !0
+; CHECK-NEXT:     [[COND1:%[^ ]+]] = icmp ugt i32 %len, 1
+; CHECK-NEXT:     %umax = select i1 [[COND1]], i32 %len, i32 1
+; CHECK-NEXT:     %exit.preloop.at = add i32 %umax, -1
+; CHECK-NEXT:     [[COND2:%[^ ]+]] = icmp ugt i32 100, %exit.preloop.at
+; CHECK-NEXT:     br i1 [[COND2]], label %loop.preloop.preheader, label %preloop.pseudo.exit
+; CHECK:        mainloop:
+; CHECK-NEXT:     br label %loop
+; CHECK:        loop:
+; CHECK-NEXT:     %idx = phi i32 [ %idx.preloop.copy, %mainloop ], [ %idx.next, %in.bounds ]
+; CHECK-NEXT:     %idx.next = add i32 %idx, -1
+; CHECK-NEXT:     %abc = icmp ult i32 %idx, %len
+; CHECK-NEXT:     br i1 true, label %in.bounds, label %out.of.bounds.loopexit1
+; CHECK-NOT:    loop.postloop:
+; CHECK:        loop.preloop:
+; CHECK-NEXT:     %idx.preloop = phi i32 [ %idx.next.preloop, %in.bounds.preloop ], [ 100, %loop.preloop.preheader ]
+; CHECK-NEXT:     %idx.next.preloop = add i32 %idx.preloop, -1
+; CHECK-NEXT:     %abc.preloop = icmp ult i32 %idx.preloop, %len
+; CHECK-NEXT:     br i1 %abc.preloop, label %in.bounds.preloop, label %out.of.bounds.loopexit
+
+entry:
+  %len = load i32, i32* %a_len_ptr, !range !0
+  br label %loop
+
+loop:
+  %idx = phi i32 [ 100, %entry ], [ %idx.next, %in.bounds ]
+  %idx.next = add i32 %idx, -1
+  %abc = icmp ult i32 %idx, %len
+  br i1 %abc, label %in.bounds, label %out.of.bounds
+
+in.bounds:
+  %addr = getelementptr i32, i32* %arr, i32 %idx
+  store i32 0, i32* %addr
+  %next = icmp ugt i32 %idx.next, 0
+  br i1 %next, label %loop, label %exit
+
+out.of.bounds:
+  ret void
+
+exit:
+  ret void
+}
+
+; Check SINT_MAX + 1, test is similar to test_01.
+define void @test_03(i32* %arr, i32* %a_len_ptr) #0 {
+
+; CHECK: test_03(
+; CHECK:        entry:
+; CHECK-NEXT:     %exit.mainloop.at = load i32, i32* %a_len_ptr, !range !0
+; CHECK-NEXT:     [[COND:%[^ ]+]] = icmp ult i32 0, %exit.mainloop.at
+; CHECK-NEXT:     br i1 [[COND]], label %loop.preheader, label %main.pseudo.exit
+; CHECK:        loop:
+; CHECK-NEXT:     %idx = phi i32 [ %idx.next, %in.bounds ], [ 0, %loop.preheader ]
+; CHECK-NEXT:     %idx.next = add nuw nsw i32 %idx, 1
+; CHECK-NEXT:     %abc = icmp ult i32 %idx, %exit.mainloop.at
+; CHECK-NEXT:     br i1 true, label %in.bounds, label %out.of.bounds.loopexit1
+; CHECK-NOT:    loop.preloop:
+; CHECK:        loop.postloop:
+; CHECK-NEXT:     %idx.postloop = phi i32 [ %idx.copy, %postloop ], [ %idx.next.postloop, %in.bounds.postloop ]
+; CHECK-NEXT:     %idx.next.postloop = add nuw nsw i32 %idx.postloop, 1
+; CHECK-NEXT:     %abc.postloop = icmp ult i32 %idx.postloop, %exit.mainloop.at
+; CHECK-NEXT:     br i1 %abc.postloop, label %in.bounds.postloop, label %out.of.bounds.loopexit
+
+entry:
+  %len = load i32, i32* %a_len_ptr, !range !0
+  br label %loop
+
+loop:
+  %idx = phi i32 [ 0, %entry ], [ %idx.next, %in.bounds ]
+  %idx.next = add nsw nuw i32 %idx, 1
+  %abc = icmp ult i32 %idx, %len
+  br i1 %abc, label %in.bounds, label %out.of.bounds
+
+in.bounds:
+  %addr = getelementptr i32, i32* %arr, i32 %idx
+  store i32 0, i32* %addr
+  %next = icmp ugt i32 %idx.next, 2147483648
+  br i1 %next, label %exit, label %loop
+
+out.of.bounds:
+  ret void
+
+exit:
+  ret void
+}
+
+; Check SINT_MAX + 1, test is similar to test_02.
+define void @test_04(i32* %arr, i32* %a_len_ptr) #0 {
+
+; CHECK: test_04(
+; CHECK:        entry:
+; CHECK-NEXT:     %len = load i32, i32* %a_len_ptr, !range !0
+; CHECK-NEXT:     [[COND1:%[^ ]+]] = icmp ugt i32 %len, 1
+; CHECK-NEXT:     %umax = select i1 [[COND1]], i32 %len, i32 1
+; CHECK-NEXT:     %exit.preloop.at = add i32 %umax, -1
+; CHECK-NEXT:     [[COND2:%[^ ]+]] = icmp ugt i32 -2147483648, %exit.preloop.at
+; CHECK-NEXT:     br i1 [[COND2]], label %loop.preloop.preheader, label %preloop.pseudo.exit
+; CHECK:        mainloop:
+; CHECK-NEXT:     br label %loop
+; CHECK:        loop:
+; CHECK-NEXT:     %idx = phi i32 [ %idx.preloop.copy, %mainloop ], [ %idx.next, %in.bounds ]
+; CHECK-NEXT:     %idx.next = add i32 %idx, -1
+; CHECK-NEXT:     %abc = icmp ult i32 %idx, %len
+; CHECK-NEXT:     br i1 true, label %in.bounds, label %out.of.bounds.loopexit1
+; CHECK-NOT:    loop.postloop:
+; CHECK:        loop.preloop:
+; CHECK-NEXT:     %idx.preloop = phi i32 [ %idx.next.preloop, %in.bounds.preloop ], [ -2147483648, %loop.preloop.preheader ]
+; CHECK-NEXT:     %idx.next.preloop = add i32 %idx.preloop, -1
+; CHECK-NEXT:     %abc.preloop = icmp ult i32 %idx.preloop, %len
+; CHECK-NEXT:     br i1 %abc.preloop, label %in.bounds.preloop, label %out.of.bounds.loopexit
+
+entry:
+  %len = load i32, i32* %a_len_ptr, !range !0
+  br label %loop
+
+loop:
+  %idx = phi i32 [ 2147483648, %entry ], [ %idx.next, %in.bounds ]
+  %idx.next = add i32 %idx, -1
+  %abc = icmp ult i32 %idx, %len
+  br i1 %abc, label %in.bounds, label %out.of.bounds
+
+in.bounds:
+  %addr = getelementptr i32, i32* %arr, i32 %idx
+  store i32 0, i32* %addr
+  %next = icmp ugt i32 %idx.next, 0
+  br i1 %next, label %loop, label %exit
+
+out.of.bounds:
+  ret void
+
+exit:
+  ret void
+}
+
+; Increasing loop, UINT_MAX. Negative test: we cannot add 1 to UINT_MAX.
+define void @test_05(i32* %arr, i32* %a_len_ptr) #0 {
+
+; CHECK: test_05(
+; CHECK-NOT:    loop.preloop:
+; CHECK-NOT:    loop.postloop:
+
+entry:
+  %len = load i32, i32* %a_len_ptr, !range !0
+  br label %loop
+
+loop:
+  %idx = phi i32 [ 0, %entry ], [ %idx.next, %in.bounds ]
+  %idx.next = add nsw nuw i32 %idx, 1
+  %abc = icmp ult i32 %idx, %len
+  br i1 %abc, label %in.bounds, label %out.of.bounds
+
+in.bounds:
+  %addr = getelementptr i32, i32* %arr, i32 %idx
+  store i32 0, i32* %addr
+  %next = icmp ugt i32 %idx.next, 4294967295
+  br i1 %next, label %exit, label %loop
+
+out.of.bounds:
+  ret void
+
+exit:
+  ret void
+}
+
+; Decreasing loop, UINT_MAX. Positive test.
+define void @test_06(i32* %arr, i32* %a_len_ptr) #0 {
+
+; CHECK: test_06(
+; CHECK:        mainloop:
+; CHECK-NEXT:     br label %loop
+; CHECK:        loop:
+; CHECK-NEXT:     %idx = phi i32 [ %idx.preloop.copy, %mainloop ], [ %idx.next, %in.bounds ]
+; CHECK-NEXT:     %idx.next = add nuw i32 %idx, -1
+; CHECK-NEXT:     %abc = icmp ult i32 %idx, %len
+; CHECK-NEXT:     br i1 true, label %in.bounds, label %out.of.bounds.loopexit1
+; CHECK-NOT:    loop.postloop:
+; CHECK:        loop.preloop:
+; CHECK-NEXT:     %idx.preloop = phi i32 [ %idx.next.preloop, %in.bounds.preloop ], [ -1, %loop.preloop.preheader ]
+; CHECK-NEXT:     %idx.next.preloop = add nuw i32 %idx.preloop, -1
+; CHECK-NEXT:     %abc.preloop = icmp ult i32 %idx.preloop, %len
+; CHECK-NEXT:     br i1 %abc.preloop, label %in.bounds.preloop, label %out.of.bounds.loopexit
+
+entry:
+  %len = load i32, i32* %a_len_ptr, !range !0
+  br label %loop
+
+loop:
+  %idx = phi i32 [ 4294967295, %entry ], [ %idx.next, %in.bounds ]
+  %idx.next = add nuw i32 %idx, -1
+  %abc = icmp ult i32 %idx, %len
+  br i1 %abc, label %in.bounds, label %out.of.bounds
+
+in.bounds:
+  %addr = getelementptr i32, i32* %arr, i32 %idx
+  store i32 0, i32* %addr
+  %next = icmp ugt i32 %idx.next, 0
+  br i1 %next, label %loop, label %exit
+
+out.of.bounds:
+  ret void
+
+exit:
+  ret void
+}
+
+!0 = !{i32 0, i32 50}

llvm/trunk/test/Transforms/IRCE/unsigned_comparisons_ult.ll

+; RUN: opt -verify-loop-info -irce-print-changed-loops -irce -S < %s 2>&1 | FileCheck %s
+
+; CHECK: irce: in function test_01: constrained Loop at depth 1 containing: %loop<header><exiting>,%in.bounds<latch><exiting>
+; CHECK: irce: in function test_02: constrained Loop at depth 1 containing: %loop<header><exiting>,%in.bounds<latch><exiting>
+; CHECK: irce: in function test_03: constrained Loop at depth 1 containing: %loop<header><exiting>,%in.bounds<latch><exiting>
+; CHECK: irce: in function test_04: constrained Loop at depth 1 containing: %loop<header><exiting>,%in.bounds<latch><exiting>
+; CHECK: irce: in function test_05: constrained Loop at depth 1 containing: %loop<header><exiting>,%in.bounds<latch><exiting>
+; CHECK: irce: in function test_06: constrained Loop at depth 1 containing: %loop<header><exiting>,%in.bounds<latch><exiting>
+; CHECK-NOT: irce: in function test_07: constrained Loop at depth 1 containing: %loop<header><exiting>,%in.bounds<latch><exiting>
+
+; ULT condition for increasing loop.
+define void @test_01(i32* %arr, i32* %a_len_ptr) #0 {
+
+; CHECK:      test_01
+; CHECK:        entry:
+; CHECK-NEXT:     %exit.mainloop.at = load i32, i32* %a_len_ptr, !range !0
+; CHECK-NEXT:     [[COND:%[^ ]+]] = icmp ult i32 0, %exit.mainloop.at
+; CHECK-NEXT:     br i1 [[COND]], label %loop.preheader, label %main.pseudo.exit
+; CHECK:        loop:
+; CHECK-NEXT:     %idx = phi i32 [ %idx.next, %in.bounds ], [ 0, %loop.preheader ]
+; CHECK-NEXT:     %idx.next = add nuw nsw i32 %idx, 1
+; CHECK-NEXT:     %abc = icmp ult i32 %idx, %exit.mainloop.at
+; CHECK-NEXT:     br i1 true, label %in.bounds, label %out.of.bounds.loopexit1
+; CHECK-NOT:    loop.preloop:
+; CHECK:        loop.postloop:
+; CHECK-NEXT:     %idx.postloop = phi i32 [ %idx.copy, %postloop ], [ %idx.next.postloop, %in.bounds.postloop ]
+; CHECK-NEXT:     %idx.next.postloop = add nuw nsw i32 %idx.postloop, 1
+; CHECK-NEXT:     %abc.postloop = icmp ult i32 %idx.postloop, %exit.mainloop.at
+; CHECK-NEXT:     br i1 %abc.postloop, label %in.bounds.postloop, label %out.of.bounds.loopexit
+
+entry:
+  %len = load i32, i32* %a_len_ptr, !range !0
+  br label %loop
+
+loop:
+  %idx = phi i32 [ 0, %entry ], [ %idx.next, %in.bounds ]
+  %idx.next = add nsw nuw i32 %idx, 1
+  %abc = icmp ult i32 %idx, %len
+  br i1 %abc, label %in.bounds, label %out.of.bounds
+
+in.bounds:
+  %addr = getelementptr i32, i32* %arr, i32 %idx
+  store i32 0, i32* %addr
+  %next = icmp ult i32 %idx.next, 100
+  br i1 %next, label %loop, label %exit
+
+out.of.bounds:
+  ret void
+
+exit:
+  ret void
+}
+
+; ULT condition for decreasing loops.
+define void @test_02(i32* %arr, i32* %a_len_ptr) #0 {
+
+; CHECK: test_02(
+; CHECK:        entry:
+; CHECK-NEXT:     %len = load i32, i32* %a_len_ptr, !range !0
+; CHECK-NEXT:     [[COND1:%[^ ]+]] = icmp ugt i32 %len, 1
+; CHECK-NEXT:     %umax = select i1 [[COND1]], i32 %len, i32 1
+; CHECK-NEXT:     %exit.preloop.at = add i32 %umax, -1
+; CHECK-NEXT:     [[COND2:%[^ ]+]] = icmp ugt i32 100, %exit.preloop.at
+; CHECK-NEXT:     br i1 [[COND2]], label %loop.preloop.preheader, label %preloop.pseudo.exit
+; CHECK:        mainloop:
+; CHECK-NEXT:     br label %loop
+; CHECK:        loop:
+; CHECK-NEXT:     %idx = phi i32 [ %idx.preloop.copy, %mainloop ], [ %idx.next, %in.bounds ]
+; CHECK-NEXT:     %idx.next = add i32 %idx, -1
+; CHECK-NEXT:     %abc = icmp ult i32 %idx, %len
+; CHECK-NEXT:     br i1 true, label %in.bounds, label %out.of.bounds.loopexit1
+; CHECK-NOT:    loop.postloop:
+; CHECK:        loop.preloop:
+; CHECK-NEXT:     %idx.preloop = phi i32 [ %idx.next.preloop, %in.bounds.preloop ], [ 100, %loop.preloop.preheader ]
+; CHECK-NEXT:     %idx.next.preloop = add i32 %idx.preloop, -1
+; CHECK-NEXT:     %abc.preloop = icmp ult i32 %idx.preloop, %len
+; CHECK-NEXT:     br i1 %abc.preloop, label %in.bounds.preloop, label %out.of.bounds.loopexit
+
+entry:
+  %len = load i32, i32* %a_len_ptr, !range !0
+  br label %loop
+
+loop:
+  %idx = phi i32 [ 100, %entry ], [ %idx.next, %in.bounds ]
+  %idx.next = add i32 %idx, -1
+  %abc = icmp ult i32 %idx, %len
+  br i1 %abc, label %in.bounds, label %out.of.bounds
+
+in.bounds:
+  %addr = getelementptr i32, i32* %arr, i32 %idx
+  store i32 0, i32* %addr
+  %next = icmp ult i32 %idx.next, 1
+  br i1 %next, label %exit, label %loop
+
+out.of.bounds:
+  ret void
+
+exit:
+  ret void
+}
+
+; Check SINT_MAX.
+define void @test_03(i32* %arr, i32* %a_len_ptr) #0 {
+
+; CHECK:      test_03
+; CHECK:        entry:
+; CHECK-NEXT:     %exit.mainloop.at = load i32, i32* %a_len_ptr, !range !0
+; CHECK-NEXT:     [[COND:%[^ ]+]] = icmp ult i32 0, %exit.mainloop.at
+; CHECK-NEXT:     br i1 [[COND]], label %loop.preheader, label %main.pseudo.exit
+; CHECK:        loop:
+; CHECK-NEXT:     %idx = phi i32 [ %idx.next, %in.bounds ], [ 0, %loop.preheader ]
+; CHECK-NEXT:     %idx.next = add nuw nsw i32 %idx, 1
+; CHECK-NEXT:     %abc = icmp ult i32 %idx, %exit.mainloop.at
+; CHECK-NEXT:     br i1 true, label %in.bounds, label %out.of.bounds.loopexit1
+; CHECK-NOT:    loop.preloop:
+; CHECK:        loop.postloop:
+; CHECK-NEXT:     %idx.postloop = phi i32 [ %idx.copy, %postloop ], [ %idx.next.postloop, %in.bounds.postloop ]
+; CHECK-NEXT:     %idx.next.postloop = add nuw nsw i32 %idx.postloop, 1
+; CHECK-NEXT:     %abc.postloop = icmp ult i32 %idx.postloop, %exit.mainloop.at
+; CHECK-NEXT:     br i1 %abc.postloop, label %in.bounds.postloop, label %out.of.bounds.loopexit
+
+entry:
+  %len = load i32, i32* %a_len_ptr, !range !0
+  br label %loop
+
+loop:
+  %idx = phi i32 [ 0, %entry ], [ %idx.next, %in.bounds ]
+  %idx.next = add nsw nuw i32 %idx, 1
+  %abc = icmp ult i32 %idx, %len
+  br i1 %abc, label %in.bounds, label %out.of.bounds
+
+in.bounds:
+  %addr = getelementptr i32, i32* %arr, i32 %idx
+  store i32 0, i32* %addr
+  %next = icmp ult i32 %idx.next, 2147483647
+  br i1 %next, label %loop, label %exit
+
+out.of.bounds:
+  ret void
+
+exit:
+  ret void
+}
+
+; Check SINT_MAX + 1, test is similar to test_01.
+define void @test_04(i32* %arr, i32* %a_len_ptr) #0 {
+
+; CHECK:      test_04
+; CHECK:        entry:
+; CHECK-NEXT:     %exit.mainloop.at = load i32, i32* %a_len_ptr, !range !0
+; CHECK-NEXT:     [[COND:%[^ ]+]] = icmp ult i32 0, %exit.mainloop.at
+; CHECK-NEXT:     br i1 [[COND]], label %loop.preheader, label %main.pseudo.exit
+; CHECK:        loop:
+; CHECK-NEXT:     %idx = phi i32 [ %idx.next, %in.bounds ], [ 0, %loop.preheader ]
+; CHECK-NEXT:     %idx.next = add nuw nsw i32 %idx, 1
+; CHECK-NEXT:     %abc = icmp ult i32 %idx, %exit.mainloop.at
+; CHECK-NEXT:     br i1 true, label %in.bounds, label %out.of.bounds.loopexit1
+; CHECK-NOT:    loop.preloop:
+; CHECK:        loop.postloop:
+; CHECK-NEXT:     %idx.postloop = phi i32 [ %idx.copy, %postloop ], [ %idx.next.postloop, %in.bounds.postloop ]
+; CHECK-NEXT:     %idx.next.postloop = add nuw nsw i32 %idx.postloop, 1
+; CHECK-NEXT:     %abc.postloop = icmp ult i32 %idx.postloop, %exit.mainloop.at
+; CHECK-NEXT:     br i1 %abc.postloop, label %in.bounds.postloop, label %out.of.bounds.loopexit
+
+entry:
+  %len = load i32, i32* %a_len_ptr, !range !0
+  br label %loop
+
+loop:
+  %idx = phi i32 [ 0, %entry ], [ %idx.next, %in.bounds ]
+  %idx.next = add nsw nuw i32 %idx, 1
+  %abc = icmp ult i32 %idx, %len
+  br i1 %abc, label %in.bounds, label %out.of.bounds
+
+in.bounds:
+  %addr = getelementptr i32, i32* %arr, i32 %idx
+  store i32 0, i32* %addr
+  %next = icmp ult i32 %idx.next, 2147483648
+  br i1 %next, label %loop, label %exit
+
+out.of.bounds:
+  ret void
+
+exit:
+  ret void
+}
+
+; Check SINT_MAX + 1, test is similar to test_02.
+define void @test_05(i32* %arr, i32* %a_len_ptr) #0 {
+; CHECK: test_05(
+; CHECK:        entry:
+; CHECK-NEXT:     %len = load i32, i32* %a_len_ptr, !range !0
+; CHECK-NEXT:     [[COND1:%[^ ]+]] = icmp ugt i32 %len, 1
+; CHECK-NEXT:     %umax = select i1 [[COND1]], i32 %len, i32 1
+; CHECK-NEXT:     %exit.preloop.at = add i32 %umax, -1
+; CHECK-NEXT:     [[COND2:%[^ ]+]] = icmp ugt i32 -2147483648, %exit.preloop.at
+; CHECK-NEXT:     br i1 [[COND2]], label %loop.preloop.preheader, label %preloop.pseudo.exit
+; CHECK:        mainloop:
+; CHECK-NEXT:     br label %loop
+; CHECK:        loop:
+; CHECK-NEXT:     %idx = phi i32 [ %idx.preloop.copy, %mainloop ], [ %idx.next, %in.bounds ]
+; CHECK-NEXT:     %idx.next = add i32 %idx, -1
+; CHECK-NEXT:     %abc = icmp ult i32 %idx, %len
+; CHECK-NEXT:     br i1 true, label %in.bounds, label %out.of.bounds.loopexit1
+; CHECK-NOT:    loop.postloop:
+; CHECK:        loop.preloop:
+; CHECK-NEXT:     %idx.preloop = phi i32 [ %idx.next.preloop, %in.bounds.preloop ], [ -2147483648, %loop.preloop.preheader ]
+; CHECK-NEXT:     %idx.next.preloop = add i32 %idx.preloop, -1
+; CHECK-NEXT:     %abc.preloop = icmp ult i32 %idx.preloop, %len
+; CHECK-NEXT:     br i1 %abc.preloop, label %in.bounds.preloop, label %out.of.bounds.loopexit
+
+entry:
+  %len = load i32, i32* %a_len_ptr, !range !0
+  br label %loop
+
+loop:
+  %idx = phi i32 [ 2147483648, %entry ], [ %idx.next, %in.bounds ]
+  %idx.next = add i32 %idx, -1
+  %abc = icmp ult i32 %idx, %len
+  br i1 %abc, label %in.bounds, label %out.of.bounds
+
+in.bounds:
+  %addr = getelementptr i32, i32* %arr, i32 %idx
+  store i32 0, i32* %addr
+  %next = icmp ult i32 %idx.next, 1
+  br i1 %next, label %exit, label %loop
+
+out.of.bounds:
+  ret void
+
+exit:
+  ret void
+}
+
+; Increasing loop, UINT_MAX. Positive test.
+define void @test_06(i32* %arr, i32* %a_len_ptr) #0 {
+
+; CHECK:      test_06
+; CHECK:        entry:
+; CHECK-NEXT:     %exit.mainloop.at = load i32, i32* %a_len_ptr, !range !0
+; CHECK-NEXT:     [[COND:%[^ ]+]] = icmp ult i32 0, %exit.mainloop.at
+; CHECK-NEXT:     br i1 [[COND]], label %loop.preheader, label %main.pseudo.exit
+; CHECK:        loop:
+; CHECK-NEXT:     %idx = phi i32 [ %idx.next, %in.bounds ], [ 0, %loop.preheader ]
+; CHECK-NEXT:     %idx.next = add nuw nsw i32 %idx, 1
+; CHECK-NEXT:     %abc = icmp ult i32 %idx, %exit.mainloop.at
+; CHECK-NEXT:     br i1 true, label %in.bounds, label %out.of.bounds.loopexit1
+; CHECK-NOT:    loop.preloop:
+; CHECK:        loop.postloop:
+; CHECK-NEXT:     %idx.postloop = phi i32 [ %idx.copy, %postloop ], [ %idx.next.postloop, %in.bounds.postloop ]
+; CHECK-NEXT:     %idx.next.postloop = add nuw nsw i32 %idx.postloop, 1
+; CHECK-NEXT:     %abc.postloop = icmp ult i32 %idx.postloop, %exit.mainloop.at
+; CHECK-NEXT:     br i1 %abc.postloop, label %in.bounds.postloop, label %out.of.bounds.loopexit
+
+entry:
+  %len = load i32, i32* %a_len_ptr, !range !0
+  br label %loop
+
+loop:
+  %idx = phi i32 [ 0, %entry ], [ %idx.next, %in.bounds ]
+  %idx.next = add nsw nuw i32 %idx, 1
+  %abc = icmp ult i32 %idx, %len
+  br i1 %abc, label %in.bounds, label %out.of.bounds
+
+in.bounds:
+  %addr = getelementptr i32, i32* %arr, i32 %idx
+  store i32 0, i32* %addr
+  %next = icmp ult i32 %idx.next, 4294967295
+  br i1 %next, label %loop, label %exit
+
+out.of.bounds:
+  ret void
+
+exit:
+  ret void
+}
+
+; Decreasing loop, UINT_MAX. Negative test: we cannot substract -1 from 0.
+define void @test_07(i32* %arr, i32* %a_len_ptr) #0 {
+
+; CHECK: test_07(
+; CHECK-NOT:    loop.preloop:
+; CHECK-NOT:    loop.postloop:
+
+entry:
+  %len = load i32, i32* %a_len_ptr, !range !0
+  br label %loop
+
+loop:
+  %idx = phi i32 [ 4294967295, %entry ], [ %idx.next, %in.bounds ]
+  %idx.next = add nuw i32 %idx, -1
+  %abc = icmp ult i32 %idx, %len
+  br i1 %abc, label %in.bounds, label %out.of.bounds
+
+in.bounds:
+  %addr = getelementptr i32, i32* %arr, i32 %idx
+  store i32 0, i32* %addr
+  %next = icmp ult i32 %idx.next, 0
+  br i1 %next, label %exit, label %loop
+
+out.of.bounds:
+  ret void
+
+exit:
+  ret void
+}
+
+!0 = !{i32 0, i32 50}