[ScalarEvolution] Ensure backedge-taken counts are not pointers.

[efriedma]

Summary

A backedge-taken count doesn't refer to memory; returning a pointer type is nonsense. So make sure we always return an integer.

The obvious way to do this would be to just convert the operands of the icmp to integers, but that doesn't quite work out at the moment: isLoopEntryGuardedByCond currently gets confused by ptrtoint operations. So we perform the ptrtoint conversion late for lt/gt operations.

The test changes are mostly innocuous. The most interesting changes are more complex SCEV expressions of the form "(-1 * (ptrtoint i8* %ptr to i64)) + %ptr)". This is expected: we can't fold this to zero because we need to preserve the pointer base.

The call to isLoopEntryGuardedByCond in howFarToZero is less precise because of ptrtoint operations; this shows up in the function pr46786_c26_char in ptrtoint.ll. Fixing it here would require more complex refactoring. It should eventually be fixed by future improvements to isImpliedCond.

See https://bugs.llvm.org/show_bug.cgi?id=46786 for context.

[lebedev.ri]

I have previously looked into something similar.
Are we not worried that this breaks SCEV for non-integral pointers/too small of an gep index sizes?

llvm/test/Analysis/ScalarEvolution/max-backedge-taken-count-guard-info.ll

835

Please precommit regeneration of check lines in this test file.

llvm/test/Transforms/PhaseOrdering/scev-custom-dl.ll

118

here

[lebedev.ri]

llvm/test/Transforms/IndVarSimplify/lftr-reuse.ll

32–33 ↗

(On Diff #349718)

I *think* this is not a non-integral pointer, and i think GEP index size (in DL) isn't too small,
so i'm not sure what is going on here?

[efriedma]

For pointers where the index type is smaller than the pointer type, coming up with the right result is a little tricky. As far as I can tell from the LangRef rules, icmp compares all the bits of the pointer, not just the index bits? So I guess we can special-case comparisons where both sides have the same pointer base: the non-index bits will be the same. If both sides have the same pointer base, and we've inferred some sort of nowrap, we can then convert the icmp to compare the index bits. I guess we can handle non-integral pointers the same way.

But I don't want to try to implement this logic without any real-world testing. As far as I know, no in-tree targets use this feature. So I'll leave it to someone who cares about such targets, and I don't think it should block this patch.

llvm/test/Transforms/IndVarSimplify/lftr-reuse.ll

32–33 ↗

(On Diff #349718)

Maybe something to do with the icmp in the preheader; I'll look further.

[lebedev.ri]

In D103656#2799450, @efriedma wrote:

For pointers where the index type is smaller than the pointer type, coming up with the right result is a little tricky. As far as I can tell from the LangRef rules, icmp compares all the bits of the pointer, not just the index bits? So I guess we can special-case comparisons where both sides have the same pointer base: the non-index bits will be the same. If both sides have the same pointer base, and we've inferred some sort of nowrap, we can then convert the icmp to compare the index bits. I guess we can handle non-integral pointers the same way.

But I don't want to try to implement this logic without any real-world testing. As far as I know, no in-tree targets use this feature. So I'll leave it to someone who cares about such targets, and I don't think it should block this patch.

So that is "yes, let's \"break\" SCEV for non-integral pointers and datalayouts with too small of an GEP index sizes".
I see. That does indeed allow us to fix SCEV :)

[efriedma]

Also perform ptrtoint conversion when analyzing implied conditions, for consistency with backedge taken analysis. This fixes some minor diffs in the previous version of the patch.

[efriedma]

Rebase

[reames]

The implementation and description of this patch seem badly out of sync.

Modifying the loop exit computation to always return an integer type *might* be reasonable. As has been raised already, there are concern with non-integral pointers, or cases where we don't *know* the size of a pointer, but those might be addressable. I have concerns, but they're potentially addressable concerns.

However, that doesn't seem to be what the implementation of this patch actually does? Having isKnownPredicate simply return false for a pointer argument which can't be cast is definitely not okay. Did you maybe upload the wrong diff?

[efriedma]

It's the right diff. The initial version only touched the backedge calculation, but it turned out to be sort of tied together with condition handling because the backedge calculation queries it. I'll update the description.

cases where we don't *know* the size of a pointer

There are no such cases. datalayout has been mandatory for a while.

Having isKnownPredicate simply return false for a pointer argument which can't be cast is definitely not okay.

The only pointers which can't be cast are non-integral pointers, and other exotic pointers where the index type is not equal to the pointer width. This doesn't apply to any in-tree target, as far as I know. I'm not sure why a minor optimization loss for out-of-tree targets is "definitely not okay". (We don't even have any test coverage for this besides the one loop in scev-custom-dl.ll.)

[efriedma]

Ping

[efriedma]

Ping.

[reames]

Eli,

Sorry for the silence. Partly busy, and partly having a hard time explaining something which seems obvious to me. Let me give this another try.

First, let's start with the interface question. You seem to be assuming that the only time we'll query facts about two SCEVs is when they're both trip counts. This is blatantly untrue. We will frequently query relations between SCEVs when e.g. discharging loop exits in IndVarSimplify. (And a bunch of other places.) Having those queries return unknown more often would be undesirable.

Second, the getLosslessPtrToIntExpr routine seems to not match what you're describing. It appears to have a bunch of cases where it does not succeed. Examples include min/max expressions, and udivs. (Going by code structure here. If I'm missing something, let me know.) Are you claiming that this routine is guaranteed to succeed? The existing code and comments don't seem to make this obvious.

If you want to achieve the stated goal, have you tried checking for lossless conversion of the *result* of the exit computation? If that worked, it would be a much cleaner and more obvious change. (e.g. never return an exit count which can't be converted to integer type) This would mean instrumenting the construction of ExitLimit to drop information if needed. Or maybe that part of your change with an assert in the ExitLimit constructor? Not sure there.

[efriedma]

Having those queries return unknown more often would be undesirable.

I agree.

Second, the getLosslessPtrToIntExpr routine seems to not match what you're describing. It appears to have a bunch of cases where it does not succeed. Examples include min/max expressions, and udivs. (Going by code structure here. If I'm missing something, let me know.) Are you claiming that this routine is guaranteed to succeed? The existing code and comments don't seem to make this obvious.

SCEVPtrToIntSinkingRewriter always succeeds; it inherits implementations for every kind of expression from SCEVRewriteVisitor. The explicit implementations of visitAddExpr and visitMulExpr are just there to preserve the flags. Not sure how you're reaching the conclusion that it fails for min/max expressions.

If you want to achieve the stated goal, have you tried checking for lossless conversion of the *result* of the exit computation? If that worked, it would be a much cleaner and more obvious change. (e.g. never return an exit count which can't be converted to integer type) This would mean instrumenting the construction of ExitLimit to drop information if needed. Or maybe that part of your change with an assert in the ExitLimit constructor? Not sure there.

This could work as an intermediate step. Digging a little, it wouldn't be that painful; there aren't that many places in howFarToZero and howManyLessThans/howManyGreaterThans that would need changes.

But we eventually want something like the current patch, I think. We don't really want to be doing computations like getMinusSCEV(LHS, RHS) where LHS and RHS are pointers. Currently, that returns a SCEV with pointer type, but it doesn't really make sense to treat it as a pointer, particularly in the context of an "icmp eq".

[reames]

In D103656#2821141, @efriedma wrote:

Having those queries return unknown more often would be undesirable.

I agree.

Second, the getLosslessPtrToIntExpr routine seems to not match what you're describing. It appears to have a bunch of cases where it does not succeed. Examples include min/max expressions, and udivs. (Going by code structure here. If I'm missing something, let me know.) Are you claiming that this routine is guaranteed to succeed? The existing code and comments don't seem to make this obvious.

SCEVPtrToIntSinkingRewriter always succeeds; it inherits implementations for every kind of expression from SCEVRewriteVisitor. The explicit implementations of visitAddExpr and visitMulExpr are just there to preserve the flags. Not sure how you're reaching the conclusion that it fails for min/max expressions.

Reading back through, I see you are correct. The only bailout (e.g. return of SCEVCouldNotCompute) is the effective type size check. Can you update/add some comments to the function to that effect?

If you want to achieve the stated goal, have you tried checking for lossless conversion of the *result* of the exit computation? If that worked, it would be a much cleaner and more obvious change. (e.g. never return an exit count which can't be converted to integer type) This would mean instrumenting the construction of ExitLimit to drop information if needed. Or maybe that part of your change with an assert in the ExitLimit constructor? Not sure there.

This could work as an intermediate step. Digging a little, it wouldn't be that painful; there aren't that many places in howFarToZero and howManyLessThans/howManyGreaterThans that would need changes.

But we eventually want something like the current patch, I think. We don't really want to be doing computations like getMinusSCEV(LHS, RHS) where LHS and RHS are pointers. Currently, that returns a SCEV with pointer type, but it doesn't really make sense to treat it as a pointer, particularly in the context of an "icmp eq".

Can I ask you to be incremental here? I'm not really bothered by your change in computeExitLimit, but I am very bothered by the accessor routine changes. This is complicated, and I don't think any of us fully understand it. I'd prefer a series of incremental patches with each one standing on it's own building towards the desired objective. In particular, I'm worried (mostly on your other patch admittedly) that this intersects with the discussion around pointer providence in subtle ways. Incrementalism, and thus time to catch mistakes early, seems very worthwhile here.

On the pointer subtraction point, that's a whole different can of worms. The vectorizers and various loop transforms - I think? - use that idiom for overlap checks. I agree it's very questionable, but I think that's a deeper issue than "just" SCEV.

[efriedma]

Updated to avoid the isImpliedCond changes, mostly. howFarToZero is slightly less powerful with this version of patch (see revised commit message), but hopefully the difference doesn't matter in practice.

[reames]

This looks really close to good to go.

One minor comment inline, but the main thing is we need to tweak getLosslessPtrToIntExpr to return CouldNotCompute for a non-integral type. As described in D104547, we can't insert new ptrtoints for non-integrals. We could move a single cast down to an operand, but we can do that later. A simple bailout should be fine for the moment. I don't think that really changes anything else in the patch.

llvm/lib/Analysis/ScalarEvolution.cpp

7513–7514

You've got the refactoring to reuse the constructor mixed in with functional changes. Would you mind splitting that out, landing, and rebasing this change?

[efriedma]

Address review comments.

[reames]

LGTM.

For the record, some of the test diffs are a bit concerning code quality wise, but I think we may need to accept them to achieve the goal. At least a couple of them have obvious possibilities for improvement in follow ups if needed, so I'm willing to have this go in. We may end up having to revert, fix a few things, then reapply though. We'll see.

Also, mostly for myself when reading through this later, this may regression exit counts for non-integral pointer tested loops, but that's - I think - okay because they're not idiomatic. Once (if?), we have a generic answer for pointer subtraction we can relax some of these and restore power while maintaining integer results.

llvm/lib/Analysis/ScalarEvolution.cpp

1085

Just for my context, they are? This is news to me. Do you have an example?

[efriedma]

llvm/lib/Analysis/ScalarEvolution.cpp

1085

I got this impression reading the patches that led up to adding the "indexing type" for pointers. I might be wrong, though.

llvm/lib/Analysis/ScalarEvolution.cpp

const SCEV *ScalarEvolution::getLosslessPtrToIntExpr(const SCEV *Op,
   ID.AddPointer(Op);
 
   void *IP = nullptr;
 
   // Is there already an expression for such a cast?
   if (const SCEV *S = UniqueSCEVs.FindNodeOrInsertPos(ID, IP))
     return S;
 
+  // It isn't legal for optimizations to construct new ptrtoint expressions
+  // for non-integral pointers.
+  if (getDataLayout().isNonIntegralPointerType(Op->getType()))
+    return getCouldNotCompute();
+
   Type *IntPtrTy = getDataLayout().getIntPtrType(Op->getType());
 
-  // We can only model ptrtoint if SCEV's effective (integer) type
+  // We can only trivially model ptrtoint if SCEV's effective (integer) type
   // is sufficiently wide to represent all possible pointer values.
+  // We could theoretically teach SCEV to truncate wider pointers, but
+  // that isn't implemented for now.
   if (getDataLayout().getTypeSizeInBits(getEffectiveSCEVType(Op->getType())) !=
       getDataLayout().getTypeSizeInBits(IntPtrTy))
     return getCouldNotCompute();
 
   // If not, is this expression something we can't reduce any further?
   if (auto *U = dyn_cast<SCEVUnknown>(Op)) {
     // Perform some basic constant folding. If the operand of the ptr2int cast
     // is a null pointer, don't create a ptr2int SCEV expression (that will be
     // left as-is), but produce a zero constant.
bool ScalarEvolution::BackedgeTakenInfo::isConstantMaxOrZero(
   };
   return MaxOrZero && !any_of(ExitNotTaken, PredicateNotAlwaysTrue);
 }
 
 bool ScalarEvolution::BackedgeTakenInfo::hasOperand(const SCEV *S) const {
   return Operands.contains(S);
 }
 
 ScalarEvolution::ExitLimit::ExitLimit(const SCEV *E)
     : ExitLimit(E, E, false, None) {
 }
 
 ScalarEvolution::ExitLimit::ExitLimit(
     const SCEV *E, const SCEV *M, bool MaxOrZero,
     ArrayRef<const SmallPtrSetImpl<const SCEVPredicate *> *> PredSetList)
     : ExactNotTaken(E), MaxNotTaken(M), MaxOrZero(MaxOrZero) {
   assert((isa<SCEVCouldNotCompute>(ExactNotTaken) ||
           !isa<SCEVCouldNotCompute>(MaxNotTaken)) &&
          "Exact is not allowed to be less precise than Max");
   assert((isa<SCEVCouldNotCompute>(MaxNotTaken) ||
           isa<SCEVConstant>(MaxNotTaken)) &&
          "No point in having a non-constant max backedge taken count!");
   for (auto *PredSet : PredSetList)
     for (auto *P : *PredSet)
       addPredicate(P);
+  assert((isa<SCEVCouldNotCompute>(E) || !E->getType()->isPointerTy()) &&
+         "Backedge count should be int");
+  assert((isa<SCEVCouldNotCompute>(M) || !M->getType()->isPointerTy()) &&
+         "Max backedge count should be int");
 }
 
 ScalarEvolution::ExitLimit::ExitLimit(
     const SCEV *E, const SCEV *M, bool MaxOrZero,
     const SmallPtrSetImpl<const SCEVPredicate *> &PredSet)
     : ExitLimit(E, M, MaxOrZero, {&PredSet}) {
 }
 
if (const SCEVAddRecExpr *AddRec = dyn_cast<SCEVAddRecExpr>(LHS))
 
         const SCEV *Ret = AddRec->getNumIterationsInRange(CompRange, *this);
         if (!isa<SCEVCouldNotCompute>(Ret)) return Ret;
       }
 
   switch (Pred) {
   case ICmpInst::ICMP_NE: {                     // while (X != Y)
     // Convert to: while (X-Y != 0)
+    if (LHS->getType()->isPointerTy()) {
+      LHS = getLosslessPtrToIntExpr(LHS);
+      if (isa<SCEVCouldNotCompute>(LHS))
+        return LHS;
+    }
+    if (RHS->getType()->isPointerTy()) {
+      RHS = getLosslessPtrToIntExpr(RHS);
+      if (isa<SCEVCouldNotCompute>(RHS))
+        return RHS;
+    }
     ExitLimit EL = howFarToZero(getMinusSCEV(LHS, RHS), L, ControlsExit,
                                 AllowPredicates);
     if (EL.hasAnyInfo()) return EL;
     break;
   }
   case ICmpInst::ICMP_EQ: {                     // while (X == Y)
     // Convert to: while (X-Y == 0)
     ExitLimit EL = howFarToNonZero(getMinusSCEV(LHS, RHS), L);
if (!PositiveStride) {
     // count will not generate any unsigned overflow. Relaxed no-overflow
     // conditions exploit NoWrapFlags, allowing to optimize in presence of
     // undefined behaviors like the case of C language.
     if (canIVOverflowOnLT(RHS, Stride, IsSigned) && !isUBOnWrap())
       return getCouldNotCompute();
   }
 
   const SCEV *Start = IV->getStart();
+
+  // Preserve pointer-typed Start/RHS to pass to isLoopEntryGuardedByCond.
+  // Use integer-typed versions for actual computation.
+  const SCEV *OrigStart = Start;
+  const SCEV *OrigRHS = RHS;
+  if (Start->getType()->isPointerTy()) {
+    Start = getLosslessPtrToIntExpr(Start);
+    if (isa<SCEVCouldNotCompute>(Start))
+      return Start;
+  }
+  if (RHS->getType()->isPointerTy()) {
+    RHS = getLosslessPtrToIntExpr(RHS);
+    if (isa<SCEVCouldNotCompute>(RHS))
+      return RHS;
+  }
+
   const SCEV *End = RHS;
   // When the RHS is not invariant, we do not know the end bound of the loop and
   // cannot calculate the ExactBECount needed by ExitLimit. However, we can
   // calculate the MaxBECount, given the start, stride and max value for the end
   // bound of the loop (RHS), and the fact that IV does not overflow (which is
   // checked above).
   if (!isLoopInvariant(RHS, L)) {
     const SCEV *MaxBECount = computeMaxBECountForLT(
ScalarEvolution::howManyLessThans(const SCEV *LHS, const SCEV *RHS,
   // If the loop entry is guarded by the result of the backedge test of the
   // first loop iteration, then we know the backedge will be taken at least
   // once and so the backedge taken count is as above. If not then we use the
   // expression (max(End,Start)-Start)/Stride to describe the backedge count,
   // as if the backedge is taken at least once max(End,Start) is End and so the
   // result is as above, and if not max(End,Start) is Start so we get a backedge
   // count of zero.
   const SCEV *BECount;
-  if (isLoopEntryGuardedByCond(L, Cond, getMinusSCEV(Start, Stride), RHS))
+  if (isLoopEntryGuardedByCond(L, Cond, getMinusSCEV(OrigStart, Stride), OrigRHS))
     BECount = BECountIfBackedgeTaken;
   else {
     // If we know that RHS >= Start in the context of loop, then we know that
     // max(RHS, Start) = RHS at this point.
     if (isLoopEntryGuardedByCond(
-            L, IsSigned ? ICmpInst::ICMP_SGE : ICmpInst::ICMP_UGE, RHS, Start))
+            L, IsSigned ? ICmpInst::ICMP_SGE : ICmpInst::ICMP_UGE, OrigRHS, OrigStart))
       End = RHS;
     else
       End = IsSigned ? getSMaxExpr(RHS, Start) : getUMaxExpr(RHS, Start);
     BECount = computeBECount(getMinusSCEV(End, Start), Stride);
   }
 
   const SCEV *MaxBECount;
   bool MaxOrZero = false;
if (!isLoopEntryGuardedByCond(L, Cond, getAddExpr(Start, Stride), RHS)) {
     // min(RHS, Start) = RHS at this point.
     if (isLoopEntryGuardedByCond(
             L, IsSigned ? ICmpInst::ICMP_SGE : ICmpInst::ICMP_UGE, Start, RHS))
       End = RHS;
     else
       End = IsSigned ? getSMinExpr(RHS, Start) : getUMinExpr(RHS, Start);
   }
 
+  if (Start->getType()->isPointerTy()) {
+    Start = getLosslessPtrToIntExpr(Start);
+    if (isa<SCEVCouldNotCompute>(Start))
+      return Start;
+  }
+  if (End->getType()->isPointerTy()) {
+    End = getLosslessPtrToIntExpr(End);
+    if (isa<SCEVCouldNotCompute>(End))
+      return End;
+  }
+
   const SCEV *BECount = computeBECount(getMinusSCEV(Start, End), Stride);
 
   APInt MaxStart = IsSigned ? getSignedRangeMax(Start)
                             : getUnsignedRangeMax(Start);
 
   APInt MinStride = IsSigned ? getSignedRangeMin(Stride)
                              : getUnsignedRangeMin(Stride);
 

llvm/test/Analysis/ScalarEvolution/max-backedge-taken-count-guard-info.ll

 ; CHECK-LABEL: 'guard_pessimizes_analysis_step1'
 ; CHECK-NEXT:  Classifying expressions for: @guard_pessimizes_analysis_step1
 ; CHECK-NEXT:    %init = phi i32 [ 2, %entry ], [ 3, %bb1 ]
 ; CHECK-NEXT:    --> %init U: [2,4) S: [2,4)
 ; CHECK-NEXT:    %iv = phi i32 [ %iv.next, %loop ], [ %init, %loop.ph ]
 ; CHECK-NEXT:    --> {%init,+,1}<%loop> U: [2,11) S: [2,11) Exits: 9 LoopDispositions: { %loop: Computable }
 ; CHECK-NEXT:    %iv.next = add i32 %iv, 1
 ; CHECK-NEXT:    --> {(1 + %init)<nuw><nsw>,+,1}<%loop> U: [3,12) S: [3,12) Exits: 10 LoopDispositions: { %loop: Computable }
 ; CHECK-NEXT:  Determining loop execution counts for: @guard_pessimizes_analysis_step1
 ; CHECK-NEXT:  Loop %loop: backedge-taken count is (9 + (-1 * %init)<nsw>)<nsw>
 ; CHECK-NEXT:  Loop %loop: max backedge-taken count is 7
 ; CHECK-NEXT:  Loop %loop: Predicated backedge-taken count is (9 + (-1 * %init)<nsw>)<nsw>
 ; CHECK-NEXT:   Predicates:
 ; CHECK:       Loop %loop: Trip multiple is 1
 ;
 entry:
   br i1 %c, label %bb1, label %guard

 define void @crash(i8* %ptr) {
 ; CHECK-LABEL: 'crash'
 ; CHECK-NEXT:  Classifying expressions for: @crash
 ; CHECK-NEXT:    %text.addr.5 = phi i8* [ %incdec.ptr112, %while.cond111 ], [ null, %while.body ]
 ; CHECK-NEXT:    --> {null,+,-1}<nw><%while.cond111> U: full-set S: full-set Exits: <<Unknown>> LoopDispositions: { %while.cond111: Computable, %while.body: Variant }
 ; CHECK-NEXT:    %incdec.ptr112 = getelementptr inbounds i8, i8* %text.addr.5, i64 -1
 ; CHECK-NEXT:    --> {(-1 + null)<nuw><nsw>,+,-1}<nw><%while.cond111> U: full-set S: full-set Exits: <<Unknown>> LoopDispositions: { %while.cond111: Computable, %while.body: Variant }
 ; CHECK-NEXT:    %lastout.2271 = phi i8* [ %incdec.ptr126, %while.body125 ], [ %ptr, %while.end117 ]
-; CHECK-NEXT:    --> {%ptr,+,1}<nuw><%while.body125> U: full-set S: full-set Exits: {(-2 + null)<nuw><nsw>,+,-1}<nw><%while.cond111> LoopDispositions: { %while.body125: Computable }
+; CHECK-NEXT:    --> {%ptr,+,1}<nuw><%while.body125> U: full-set S: full-set Exits: {(-2 + (-1 * (ptrtoint i8* %ptr to i64)) + %ptr),+,-1}<nw><%while.cond111> LoopDispositions: { %while.body125: Computable }
 ; CHECK-NEXT:    %incdec.ptr126 = getelementptr inbounds i8, i8* %lastout.2271, i64 1
-; CHECK-NEXT:    --> {(1 + %ptr)<nuw>,+,1}<nuw><%while.body125> U: [1,0) S: [1,0) Exits: {(-1 + null)<nuw><nsw>,+,-1}<nw><%while.cond111> LoopDispositions: { %while.body125: Computable }
+; CHECK-NEXT:    --> {(1 + %ptr)<nuw>,+,1}<nuw><%while.body125> U: [1,0) S: [1,0) Exits: {(-1 + (-1 * (ptrtoint i8* %ptr to i64)) + %ptr),+,-1}<nw><%while.cond111> LoopDispositions: { %while.body125: Computable }
 ; CHECK-NEXT:  Determining loop execution counts for: @crash
-; CHECK-NEXT:  Loop %while.body125: backedge-taken count is {(-2 + (-1 * %ptr) + null),+,-1}<nw><%while.cond111>
+; CHECK-NEXT:  Loop %while.body125: backedge-taken count is {(-2 + (-1 * (ptrtoint i8* %ptr to i64))),+,-1}<nw><%while.cond111>
 ; CHECK-NEXT:  Loop %while.body125: max backedge-taken count is -2
-; CHECK-NEXT:  Loop %while.body125: Predicated backedge-taken count is {(-2 + (-1 * %ptr) + null),+,-1}<nw><%while.cond111>
+; CHECK-NEXT:  Loop %while.body125: Predicated backedge-taken count is {(-2 + (-1 * (ptrtoint i8* %ptr to i64))),+,-1}<nw><%while.cond111>
 ; CHECK-NEXT:   Predicates:
 ; CHECK:       Loop %while.body125: Trip multiple is 1
 ; CHECK-NEXT:  Loop %while.cond111: Unpredictable backedge-taken count.
 ; CHECK-NEXT:  Loop %while.cond111: Unpredictable max backedge-taken count.
 ; CHECK-NEXT:  Loop %while.cond111: Unpredictable predicated backedge-taken count.
 ; CHECK-NEXT:  Loop %while.body: Unpredictable backedge-taken count.
 ; CHECK-NEXT:  Loop %while.body: Unpredictable max backedge-taken count.
 ; CHECK-NEXT:  Loop %while.body: Unpredictable predicated backedge-taken count.

llvm/test/Analysis/ScalarEvolution/no-wrap-symbolic-becount.ll

loop:                                             ; preds = %backedge, %entry
   %indvars.iv.next = add nsw i64 %indvars.iv, -1
   br i1 %cond, label %exit, label %loop
 
 exit:                                             ; preds = %loop
   ret i32 0
 }
 
 define void @pointer_iv_nowrap(i8* %startptr, i8* %endptr) local_unnamed_addr {
-; CHECK-LABEL: Classifying expressions for: @pointer_iv_nowrap
-; CHECK-NEXT:  %init = getelementptr inbounds i8, i8* %startptr, i64 2000
-; CHECK-NEXT:  -->  (2000 + %startptr)<nuw> U: [2000,0) S: [2000,0)
-; CHECK-NEXT:  %iv = phi i8* [ %init, %entry ], [ %iv.next, %loop ]
-; CHECK-NEXT:  -->  {(2000 + %startptr)<nuw>,+,1}<nuw><%loop> U: [2000,0) S: [2000,0)
-; CHECK-NEXT:  %iv.next = getelementptr inbounds i8, i8* %iv, i64 1
-; CHECK-NEXT:  -->  {(2001 + %startptr)<nuw>,+,1}<nuw><%loop> U: [2001,0) S: [2001,0)
-
-; CHECK-NEXT:Determining loop execution counts for: @pointer_iv_nowrap
-; CHECK-NEXT:Loop %loop: backedge-taken count is (-2000 + (-1 * %startptr) + ((2000 + %startptr)<nuw> umax %endptr))
-; CHECK-NEXT:Loop %loop: max backedge-taken count is -2001
-; CHECK-NEXT:Loop %loop: Predicated backedge-taken count is (-2000 + (-1 * %startptr) + ((2000 + %startptr)<nuw> umax %endptr))
+; CHECK-LABEL: 'pointer_iv_nowrap'
+; CHECK-NEXT:  Classifying expressions for: @pointer_iv_nowrap
+; CHECK-NEXT:    %init = getelementptr inbounds i8, i8* %startptr, i64 2000
+; CHECK-NEXT:    --> (2000 + %startptr)<nuw> U: [2000,0) S: [2000,0)
+; CHECK-NEXT:    %iv = phi i8* [ %init, %entry ], [ %iv.next, %loop ]
+; CHECK-NEXT:    --> {(2000 + %startptr)<nuw>,+,1}<nuw><%loop> U: [2000,0) S: [2000,0) Exits: ((-1 * (ptrtoint i8* %startptr to i64)) + ((2000 + (ptrtoint i8* %startptr to i64))<nuw> umax (ptrtoint i8* %endptr to i64)) + %startptr) LoopDispositions: { %loop: Computable }
+; CHECK-NEXT:    %iv.next = getelementptr inbounds i8, i8* %iv, i64 1
+; CHECK-NEXT:    --> {(2001 + %startptr)<nuw>,+,1}<nuw><%loop> U: [2001,0) S: [2001,0) Exits: (1 + (-1 * (ptrtoint i8* %startptr to i64)) + ((2000 + (ptrtoint i8* %startptr to i64))<nuw> umax (ptrtoint i8* %endptr to i64)) + %startptr) LoopDispositions: { %loop: Computable }
+; CHECK-NEXT:  Determining loop execution counts for: @pointer_iv_nowrap
+; CHECK-NEXT:  Loop %loop: backedge-taken count is (-2000 + (-1 * (ptrtoint i8* %startptr to i64)) + ((2000 + (ptrtoint i8* %startptr to i64))<nuw> umax (ptrtoint i8* %endptr to i64)))
+; CHECK-NEXT:  Loop %loop: max backedge-taken count is -2001
+; CHECK-NEXT:  Loop %loop: Predicated backedge-taken count is (-2000 + (-1 * (ptrtoint i8* %startptr to i64)) + ((2000 + (ptrtoint i8* %startptr to i64))<nuw> umax (ptrtoint i8* %endptr to i64)))
+; CHECK-NEXT:   Predicates:
+; CHECK:       Loop %loop: Trip multiple is 1
 ;
 entry:
   %init = getelementptr inbounds i8, i8* %startptr, i64 2000
   br label %loop
 
 loop:
   %iv = phi i8* [ %init, %entry ], [ %iv.next, %loop ]
   %iv.next = getelementptr inbounds i8, i8* %iv, i64 1
   %ec = icmp ugt i8* %iv.next, %endptr
   br i1 %ec, label %end, label %loop
 
 end:
   ret void
 }
+
+define i32 @dummy(i32 %start, i32* %p, i32* %q) {
+entry:
+  ret i32 0
+}

llvm/test/Analysis/ScalarEvolution/nsw.ll

 ; CHECK: %ptrincdec.i.i =
 ; CHECK: {(4 + %begin)<nuw>,+,4}<nuw><%for.body.i.i>
   %__first.addr.08.i.i = getelementptr inbounds i32, i32* %begin, i64 %indvar.i.i
 ; CHECK: %__first.addr.08.i.i
 ; CHECK: {%begin,+,4}<nuw><%for.body.i.i>
   store i32 0, i32* %__first.addr.08.i.i, align 4
   %cmp.i.i = icmp eq i32* %ptrincdec.i.i, %end
   br i1 %cmp.i.i, label %_ZSt4fillIPiiEvT_S1_RKT0_.exit, label %for.body.i.i
-; CHECK: Loop %for.body.i.i: backedge-taken count is ((-4 + (-1 * %begin) + %end) /u 4)
+; CHECK: Loop %for.body.i.i: backedge-taken count is ((-4 + (-1 * (ptrtoint i32* %begin to i64)) + (ptrtoint i32* %end to i64)) /u 4)
 ; CHECK: Loop %for.body.i.i: max backedge-taken count is 4611686018427387903
 _ZSt4fillIPiiEvT_S1_RKT0_.exit:                   ; preds = %for.body.i.i, %entry
   ret void
 }
 
 ; A single AddExpr exists for (%a + %b), which is not always <nsw>.
 ; CHECK: @addnsw
 ; CHECK-NOT: --> (%a + %b)<nsw>
bb1:                                              ; preds = %bb1, %bb
   %tmp6 = icmp ult i32* %tmp5, %tmp
   br i1 %tmp6, label %bb1, label %bb7
 
 bb7:                                              ; preds = %bb1
   ret i32 %tmp4
 }
 
 ; CHECK-LABEL: PR12376
-; CHECK: -->  {(4 + %arg)<nuw>,+,4}<nuw><%bb2>{{ U: [^ ]+ S: [^ ]+}}{{ *}}Exits: (4 + (4 * ((-1 + (-1 * %arg) + ((4 + %arg)<nuw> umax %arg1)) /u 4))<nuw> + %arg)
+; CHECK: -->  {(4 + %arg)<nuw>,+,4}<nuw><%bb2>{{ U: [^ ]+ S: [^ ]+}}{{ *}}Exits: (4 + (4 * ((-1 + (-1 * (ptrtoint i32* %arg to i64)) + ((4 + (ptrtoint i32* %arg to i64))<nuw> umax (ptrtoint i32* %arg1 to i64))) /u 4))<nuw> + %arg)
 define void @PR12376(i32* nocapture %arg, i32* nocapture %arg1)  {
 bb:
   br label %bb2
 
 bb2:                                              ; preds = %bb2, %bb
   %tmp = phi i32* [ %arg, %bb ], [ %tmp4, %bb2 ]
   %tmp4 = getelementptr inbounds i32, i32* %tmp, i64 1
   %tmp3 = icmp ult i32* %tmp4, %arg1

llvm/test/Analysis/ScalarEvolution/ptrtoint.ll

 ;     other[cur - start] += *cur;
 ; }
 define void @pr46786_c26_char(i8* %arg, i8* %arg1, i8* %arg2) {
 ; X64-LABEL: 'pr46786_c26_char'
 ; X64-NEXT:  Classifying expressions for: @pr46786_c26_char
 ; X64-NEXT:    %i4 = ptrtoint i8* %arg to i64
 ; X64-NEXT:    --> (ptrtoint i8* %arg to i64) U: full-set S: full-set
 ; X64-NEXT:    %i7 = phi i8* [ %arg, %bb3 ], [ %i14, %bb6 ]
-; X64-NEXT:    --> {%arg,+,1}<nuw><%bb6> U: full-set S: full-set Exits: (-1 + %arg1) LoopDispositions: { %bb6: Computable }
+; X64-NEXT:    --> {%arg,+,1}<nuw><%bb6> U: full-set S: full-set Exits: (-1 + (-1 * (ptrtoint i8* %arg to i64)) + (ptrtoint i8* %arg1 to i64) + %arg) LoopDispositions: { %bb6: Computable }
 ; X64-NEXT:    %i8 = load i8, i8* %i7, align 1
 ; X64-NEXT:    --> %i8 U: full-set S: full-set Exits: <<Unknown>> LoopDispositions: { %bb6: Variant }
 ; X64-NEXT:    %i9 = ptrtoint i8* %i7 to i64
-; X64-NEXT:    --> {(ptrtoint i8* %arg to i64),+,1}<nuw><%bb6> U: full-set S: full-set Exits: (-1 + (-1 * %arg) + (ptrtoint i8* %arg to i64) + %arg1) LoopDispositions: { %bb6: Computable }
+; X64-NEXT:    --> {(ptrtoint i8* %arg to i64),+,1}<nuw><%bb6> U: full-set S: full-set Exits: (-1 + (ptrtoint i8* %arg1 to i64)) LoopDispositions: { %bb6: Computable }
 ; X64-NEXT:    %i10 = sub i64 %i9, %i4
-; X64-NEXT:    --> {0,+,1}<nw><%bb6> U: [0,-1) S: [0,-1) Exits: (-1 + (-1 * %arg) + %arg1) LoopDispositions: { %bb6: Computable }
+; X64-NEXT:    --> {0,+,1}<nw><%bb6> U: full-set S: full-set Exits: (-1 + (-1 * (ptrtoint i8* %arg to i64)) + (ptrtoint i8* %arg1 to i64)) LoopDispositions: { %bb6: Computable }
 ; X64-NEXT:    %i11 = getelementptr inbounds i8, i8* %arg2, i64 %i10
-; X64-NEXT:    --> {%arg2,+,1}<nw><%bb6> U: full-set S: full-set Exits: (-1 + (-1 * %arg) + %arg1 + %arg2) LoopDispositions: { %bb6: Computable }
+; X64-NEXT:    --> {%arg2,+,1}<nw><%bb6> U: full-set S: full-set Exits: (-1 + (-1 * (ptrtoint i8* %arg to i64)) + (ptrtoint i8* %arg1 to i64) + %arg2) LoopDispositions: { %bb6: Computable }
 ; X64-NEXT:    %i12 = load i8, i8* %i11, align 1
 ; X64-NEXT:    --> %i12 U: full-set S: full-set Exits: <<Unknown>> LoopDispositions: { %bb6: Variant }
 ; X64-NEXT:    %i13 = add i8 %i12, %i8
 ; X64-NEXT:    --> (%i12 + %i8) U: full-set S: full-set Exits: <<Unknown>> LoopDispositions: { %bb6: Variant }
 ; X64-NEXT:    %i14 = getelementptr inbounds i8, i8* %i7, i64 1
-; X64-NEXT:    --> {(1 + %arg)<nuw>,+,1}<nuw><%bb6> U: [1,0) S: [1,0) Exits: %arg1 LoopDispositions: { %bb6: Computable }
+; X64-NEXT:    --> {(1 + %arg)<nuw>,+,1}<nuw><%bb6> U: [1,0) S: [1,0) Exits: ((-1 * (ptrtoint i8* %arg to i64)) + (ptrtoint i8* %arg1 to i64) + %arg) LoopDispositions: { %bb6: Computable }
 ; X64-NEXT:  Determining loop execution counts for: @pr46786_c26_char
-; X64-NEXT:  Loop %bb6: backedge-taken count is (-1 + (-1 * %arg) + %arg1)
-; X64-NEXT:  Loop %bb6: max backedge-taken count is -2
-; X64-NEXT:  Loop %bb6: Predicated backedge-taken count is (-1 + (-1 * %arg) + %arg1)
+; X64-NEXT:  Loop %bb6: backedge-taken count is (-1 + (-1 * (ptrtoint i8* %arg to i64)) + (ptrtoint i8* %arg1 to i64))
+; X64-NEXT:  Loop %bb6: max backedge-taken count is -1
+; X64-NEXT:  Loop %bb6: Predicated backedge-taken count is (-1 + (-1 * (ptrtoint i8* %arg to i64)) + (ptrtoint i8* %arg1 to i64))
 ; X64-NEXT:   Predicates:
 ; X64:       Loop %bb6: Trip multiple is 1
 ;
 ; X32-LABEL: 'pr46786_c26_char'
 ; X32-NEXT:  Classifying expressions for: @pr46786_c26_char
 ; X32-NEXT:    %i4 = ptrtoint i8* %arg to i64
 ; X32-NEXT:    --> (zext i32 (ptrtoint i8* %arg to i32) to i64) U: [0,4294967296) S: [0,4294967296)
 ; X32-NEXT:    %i7 = phi i8* [ %arg, %bb3 ], [ %i14, %bb6 ]
-; X32-NEXT:    --> {%arg,+,1}<nuw><%bb6> U: full-set S: full-set Exits: (-1 + %arg1) LoopDispositions: { %bb6: Computable }
+; X32-NEXT:    --> {%arg,+,1}<nuw><%bb6> U: full-set S: full-set Exits: (-1 + (-1 * (ptrtoint i8* %arg to i32)) + (ptrtoint i8* %arg1 to i32) + %arg) LoopDispositions: { %bb6: Computable }
 ; X32-NEXT:    %i8 = load i8, i8* %i7, align 1
 ; X32-NEXT:    --> %i8 U: full-set S: full-set Exits: <<Unknown>> LoopDispositions: { %bb6: Variant }
 ; X32-NEXT:    %i9 = ptrtoint i8* %i7 to i64
-; X32-NEXT:    --> {(zext i32 (ptrtoint i8* %arg to i32) to i64),+,1}<nuw><%bb6> U: [0,8589934590) S: [0,8589934590) Exits: ((zext i8* (-1 + (-1 * %arg) + %arg1) to i64) + (zext i32 (ptrtoint i8* %arg to i32) to i64)) LoopDispositions: { %bb6: Computable }
+; X32-NEXT:    --> {(zext i32 (ptrtoint i8* %arg to i32) to i64),+,1}<nuw><%bb6> U: [0,8589934591) S: [0,8589934591) Exits: ((zext i32 (-1 + (-1 * (ptrtoint i8* %arg to i32)) + (ptrtoint i8* %arg1 to i32)) to i64) + (zext i32 (ptrtoint i8* %arg to i32) to i64)) LoopDispositions: { %bb6: Computable }
 ; X32-NEXT:    %i10 = sub i64 %i9, %i4
-; X32-NEXT:    --> {0,+,1}<nw><%bb6> U: [0,4294967295) S: [0,4294967295) Exits: (zext i8* (-1 + (-1 * %arg) + %arg1) to i64) LoopDispositions: { %bb6: Computable }
+; X32-NEXT:    --> {0,+,1}<nw><%bb6> U: [0,4294967296) S: [0,4294967296) Exits: (zext i32 (-1 + (-1 * (ptrtoint i8* %arg to i32)) + (ptrtoint i8* %arg1 to i32)) to i64) LoopDispositions: { %bb6: Computable }
 ; X32-NEXT:    %i11 = getelementptr inbounds i8, i8* %arg2, i64 %i10
-; X32-NEXT:    --> {%arg2,+,1}<%bb6> U: full-set S: full-set Exits: (-1 + (-1 * %arg) + %arg1 + %arg2) LoopDispositions: { %bb6: Computable }
+; X32-NEXT:    --> {%arg2,+,1}<%bb6> U: full-set S: full-set Exits: (-1 + (-1 * (ptrtoint i8* %arg to i32)) + (ptrtoint i8* %arg1 to i32) + %arg2) LoopDispositions: { %bb6: Computable }
 ; X32-NEXT:    %i12 = load i8, i8* %i11, align 1
 ; X32-NEXT:    --> %i12 U: full-set S: full-set Exits: <<Unknown>> LoopDispositions: { %bb6: Variant }
 ; X32-NEXT:    %i13 = add i8 %i12, %i8
 ; X32-NEXT:    --> (%i12 + %i8) U: full-set S: full-set Exits: <<Unknown>> LoopDispositions: { %bb6: Variant }
 ; X32-NEXT:    %i14 = getelementptr inbounds i8, i8* %i7, i64 1
-; X32-NEXT:    --> {(1 + %arg)<nuw>,+,1}<nuw><%bb6> U: [1,0) S: [1,0) Exits: %arg1 LoopDispositions: { %bb6: Computable }
+; X32-NEXT:    --> {(1 + %arg)<nuw>,+,1}<nuw><%bb6> U: [1,0) S: [1,0) Exits: ((-1 * (ptrtoint i8* %arg to i32)) + (ptrtoint i8* %arg1 to i32) + %arg) LoopDispositions: { %bb6: Computable }
 ; X32-NEXT:  Determining loop execution counts for: @pr46786_c26_char
-; X32-NEXT:  Loop %bb6: backedge-taken count is (-1 + (-1 * %arg) + %arg1)
-; X32-NEXT:  Loop %bb6: max backedge-taken count is -2
-; X32-NEXT:  Loop %bb6: Predicated backedge-taken count is (-1 + (-1 * %arg) + %arg1)
+; X32-NEXT:  Loop %bb6: backedge-taken count is (-1 + (-1 * (ptrtoint i8* %arg to i32)) + (ptrtoint i8* %arg1 to i32))
+; X32-NEXT:  Loop %bb6: max backedge-taken count is -1
+; X32-NEXT:  Loop %bb6: Predicated backedge-taken count is (-1 + (-1 * (ptrtoint i8* %arg to i32)) + (ptrtoint i8* %arg1 to i32))
 ; X32-NEXT:   Predicates:
 ; X32:       Loop %bb6: Trip multiple is 1
 ;
   %i = icmp eq i8* %arg, %arg1
   br i1 %i, label %bb5, label %bb3
 
 bb3:
   %i4 = ptrtoint i8* %arg to i64

 ;
 ; FIXME: 4 * (%i10 EXACT/s 4) is just %i10
 define void @pr46786_c26_int(i32* %arg, i32* %arg1, i32* %arg2) {
 ; X64-LABEL: 'pr46786_c26_int'
 ; X64-NEXT:  Classifying expressions for: @pr46786_c26_int
 ; X64-NEXT:    %i4 = ptrtoint i32* %arg to i64
 ; X64-NEXT:    --> (ptrtoint i32* %arg to i64) U: full-set S: full-set
 ; X64-NEXT:    %i7 = phi i32* [ %arg, %bb3 ], [ %i15, %bb6 ]
-; X64-NEXT:    --> {%arg,+,4}<nuw><%bb6> U: full-set S: full-set Exits: ((4 * ((-4 + (-1 * %arg) + %arg1) /u 4))<nuw> + %arg) LoopDispositions: { %bb6: Computable }
+; X64-NEXT:    --> {%arg,+,4}<nuw><%bb6> U: full-set S: full-set Exits: ((4 * ((-4 + (-1 * (ptrtoint i32* %arg to i64)) + (ptrtoint i32* %arg1 to i64)) /u 4))<nuw> + %arg) LoopDispositions: { %bb6: Computable }
 ; X64-NEXT:    %i8 = load i32, i32* %i7, align 4
 ; X64-NEXT:    --> %i8 U: full-set S: full-set Exits: <<Unknown>> LoopDispositions: { %bb6: Variant }
 ; X64-NEXT:    %i9 = ptrtoint i32* %i7 to i64
-; X64-NEXT:    --> {(ptrtoint i32* %arg to i64),+,4}<nuw><%bb6> U: full-set S: full-set Exits: ((4 * ((-4 + (-1 * %arg) + %arg1) /u 4))<nuw> + (ptrtoint i32* %arg to i64)) LoopDispositions: { %bb6: Computable }
+; X64-NEXT:    --> {(ptrtoint i32* %arg to i64),+,4}<nuw><%bb6> U: full-set S: full-set Exits: ((4 * ((-4 + (-1 * (ptrtoint i32* %arg to i64)) + (ptrtoint i32* %arg1 to i64)) /u 4))<nuw> + (ptrtoint i32* %arg to i64)) LoopDispositions: { %bb6: Computable }
 ; X64-NEXT:    %i10 = sub i64 %i9, %i4
-; X64-NEXT:    --> {0,+,4}<nw><%bb6> U: [0,-3) S: [-9223372036854775808,9223372036854775805) Exits: (4 * ((-4 + (-1 * %arg) + %arg1) /u 4))<nuw> LoopDispositions: { %bb6: Computable }
+; X64-NEXT:    --> {0,+,4}<nw><%bb6> U: [0,-3) S: [-9223372036854775808,9223372036854775805) Exits: (4 * ((-4 + (-1 * (ptrtoint i32* %arg to i64)) + (ptrtoint i32* %arg1 to i64)) /u 4))<nuw> LoopDispositions: { %bb6: Computable }
 ; X64-NEXT:    %i11 = ashr exact i64 %i10, 2
 ; X64-NEXT:    --> %i11 U: [-2305843009213693952,2305843009213693952) S: [-2305843009213693952,2305843009213693952) Exits: <<Unknown>> LoopDispositions: { %bb6: Variant }
 ; X64-NEXT:    %i12 = getelementptr inbounds i32, i32* %arg2, i64 %i11
 ; X64-NEXT:    --> ((4 * %i11)<nsw> + %arg2) U: full-set S: full-set Exits: <<Unknown>> LoopDispositions: { %bb6: Variant }
 ; X64-NEXT:    %i13 = load i32, i32* %i12, align 4
 ; X64-NEXT:    --> %i13 U: full-set S: full-set Exits: <<Unknown>> LoopDispositions: { %bb6: Variant }
 ; X64-NEXT:    %i14 = add nsw i32 %i13, %i8
 ; X64-NEXT:    --> (%i13 + %i8) U: full-set S: full-set Exits: <<Unknown>> LoopDispositions: { %bb6: Variant }
 ; X64-NEXT:    %i15 = getelementptr inbounds i32, i32* %i7, i64 1
-; X64-NEXT:    --> {(4 + %arg)<nuw>,+,4}<nuw><%bb6> U: [4,0) S: [4,0) Exits: (4 + (4 * ((-4 + (-1 * %arg) + %arg1) /u 4))<nuw> + %arg) LoopDispositions: { %bb6: Computable }
+; X64-NEXT:    --> {(4 + %arg)<nuw>,+,4}<nuw><%bb6> U: [4,0) S: [4,0) Exits: (4 + (4 * ((-4 + (-1 * (ptrtoint i32* %arg to i64)) + (ptrtoint i32* %arg1 to i64)) /u 4))<nuw> + %arg) LoopDispositions: { %bb6: Computable }
 ; X64-NEXT:  Determining loop execution counts for: @pr46786_c26_int
-; X64-NEXT:  Loop %bb6: backedge-taken count is ((-4 + (-1 * %arg) + %arg1) /u 4)
+; X64-NEXT:  Loop %bb6: backedge-taken count is ((-4 + (-1 * (ptrtoint i32* %arg to i64)) + (ptrtoint i32* %arg1 to i64)) /u 4)
 ; X64-NEXT:  Loop %bb6: max backedge-taken count is 4611686018427387903
-; X64-NEXT:  Loop %bb6: Predicated backedge-taken count is ((-4 + (-1 * %arg) + %arg1) /u 4)
+; X64-NEXT:  Loop %bb6: Predicated backedge-taken count is ((-4 + (-1 * (ptrtoint i32* %arg to i64)) + (ptrtoint i32* %arg1 to i64)) /u 4)
 ; X64-NEXT:   Predicates:
 ; X64:       Loop %bb6: Trip multiple is 1
 ;
 ; X32-LABEL: 'pr46786_c26_int'
 ; X32-NEXT:  Classifying expressions for: @pr46786_c26_int
 ; X32-NEXT:    %i4 = ptrtoint i32* %arg to i64
 ; X32-NEXT:    --> (zext i32 (ptrtoint i32* %arg to i32) to i64) U: [0,4294967296) S: [0,4294967296)
 ; X32-NEXT:    %i7 = phi i32* [ %arg, %bb3 ], [ %i15, %bb6 ]
-; X32-NEXT:    --> {%arg,+,4}<nuw><%bb6> U: full-set S: full-set Exits: ((4 * ((-4 + (-1 * %arg) + %arg1) /u 4))<nuw> + %arg) LoopDispositions: { %bb6: Computable }
+; X32-NEXT:    --> {%arg,+,4}<nuw><%bb6> U: full-set S: full-set Exits: ((4 * ((-4 + (-1 * (ptrtoint i32* %arg to i32)) + (ptrtoint i32* %arg1 to i32)) /u 4))<nuw> + %arg) LoopDispositions: { %bb6: Computable }
 ; X32-NEXT:    %i8 = load i32, i32* %i7, align 4
 ; X32-NEXT:    --> %i8 U: full-set S: full-set Exits: <<Unknown>> LoopDispositions: { %bb6: Variant }
 ; X32-NEXT:    %i9 = ptrtoint i32* %i7 to i64
-; X32-NEXT:    --> {(zext i32 (ptrtoint i32* %arg to i32) to i64),+,4}<nuw><%bb6> U: [0,8589934588) S: [0,8589934588) Exits: ((zext i32 (ptrtoint i32* %arg to i32) to i64) + (4 * ((zext i32* (-4 + (-1 * %arg) + %arg1) to i64) /u 4))<nuw><nsw>) LoopDispositions: { %bb6: Computable }
+; X32-NEXT:    --> {(zext i32 (ptrtoint i32* %arg to i32) to i64),+,4}<nuw><%bb6> U: [0,8589934588) S: [0,8589934588) Exits: ((zext i32 (ptrtoint i32* %arg to i32) to i64) + (4 * ((zext i32 (-4 + (-1 * (ptrtoint i32* %arg to i32)) + (ptrtoint i32* %arg1 to i32)) to i64) /u 4))<nuw><nsw>) LoopDispositions: { %bb6: Computable }
 ; X32-NEXT:    %i10 = sub i64 %i9, %i4
-; X32-NEXT:    --> {0,+,4}<nw><%bb6> U: [0,4294967293) S: [0,4294967293) Exits: (4 * ((zext i32* (-4 + (-1 * %arg) + %arg1) to i64) /u 4))<nuw><nsw> LoopDispositions: { %bb6: Computable }
+; X32-NEXT:    --> {0,+,4}<nw><%bb6> U: [0,4294967293) S: [0,4294967293) Exits: (4 * ((zext i32 (-4 + (-1 * (ptrtoint i32* %arg to i32)) + (ptrtoint i32* %arg1 to i32)) to i64) /u 4))<nuw><nsw> LoopDispositions: { %bb6: Computable }
 ; X32-NEXT:    %i11 = ashr exact i64 %i10, 2
 ; X32-NEXT:    --> %i11 U: [-2147483648,2147483648) S: [-2147483648,2147483648) Exits: <<Unknown>> LoopDispositions: { %bb6: Variant }
 ; X32-NEXT:    %i12 = getelementptr inbounds i32, i32* %arg2, i64 %i11
 ; X32-NEXT:    --> ((4 * (trunc i64 %i11 to i32))<nsw> + %arg2) U: full-set S: full-set Exits: <<Unknown>> LoopDispositions: { %bb6: Variant }
 ; X32-NEXT:    %i13 = load i32, i32* %i12, align 4
 ; X32-NEXT:    --> %i13 U: full-set S: full-set Exits: <<Unknown>> LoopDispositions: { %bb6: Variant }
 ; X32-NEXT:    %i14 = add nsw i32 %i13, %i8
 ; X32-NEXT:    --> (%i13 + %i8) U: full-set S: full-set Exits: <<Unknown>> LoopDispositions: { %bb6: Variant }
 ; X32-NEXT:    %i15 = getelementptr inbounds i32, i32* %i7, i64 1
-; X32-NEXT:    --> {(4 + %arg)<nuw>,+,4}<nuw><%bb6> U: [4,0) S: [4,0) Exits: (4 + (4 * ((-4 + (-1 * %arg) + %arg1) /u 4))<nuw> + %arg) LoopDispositions: { %bb6: Computable }
+; X32-NEXT:    --> {(4 + %arg)<nuw>,+,4}<nuw><%bb6> U: [4,0) S: [4,0) Exits: (4 + (4 * ((-4 + (-1 * (ptrtoint i32* %arg to i32)) + (ptrtoint i32* %arg1 to i32)) /u 4))<nuw> + %arg) LoopDispositions: { %bb6: Computable }
 ; X32-NEXT:  Determining loop execution counts for: @pr46786_c26_int
-; X32-NEXT:  Loop %bb6: backedge-taken count is ((-4 + (-1 * %arg) + %arg1) /u 4)
+; X32-NEXT:  Loop %bb6: backedge-taken count is ((-4 + (-1 * (ptrtoint i32* %arg to i32)) + (ptrtoint i32* %arg1 to i32)) /u 4)
 ; X32-NEXT:  Loop %bb6: max backedge-taken count is 1073741823
-; X32-NEXT:  Loop %bb6: Predicated backedge-taken count is ((-4 + (-1 * %arg) + %arg1) /u 4)
+; X32-NEXT:  Loop %bb6: Predicated backedge-taken count is ((-4 + (-1 * (ptrtoint i32* %arg to i32)) + (ptrtoint i32* %arg1 to i32)) /u 4)
 ; X32-NEXT:   Predicates:
 ; X32:       Loop %bb6: Trip multiple is 1
 ;
   %i = icmp eq i32* %arg, %arg1
   br i1 %i, label %bb5, label %bb3
 
 bb3:
   %i4 = ptrtoint i32* %arg to i64

llvm/test/Transforms/IndVarSimplify/2011-11-01-lftrptr.ll

 ; NOTE: Assertions have been autogenerated by utils/update_test_checks.py
 ; RUN: opt < %s -indvars -S "-data-layout=e-p:64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:64:64-f32:32:32-f64:64:64-v64:64:64-v128:128:128-a0:0:64-s0:64:64-f80:128:128-n8:16:32:64" | FileCheck -check-prefix=PTR64 %s
 ; RUN: opt < %s -indvars -S "-data-layout=e-p:32:32:32-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:32:64-f32:32:32-f64:32:64-v64:64:64-v128:128:128-a0:0:64-f80:128:128-n8:16:32" | FileCheck -check-prefix=PTR32 %s
 ;
 ; PR11279: Assertion !IVLimit->getType()->isPointerTy()
 ;
 ; Test LinearFunctionTestReplace of a pointer-type loop counter. Note
 ; that BECount may or may not be a pointer type. A pointer type
 ; BECount doesn't really make sense, but that's what falls out of
 ; SCEV. Since it's an i8*, it has unit stride so we never adjust the
 ; SCEV expression in a way that would convert it to an integer type.
 
 define i8 @testnullptrptr(i8* %buf, i8* %end) nounwind {
 ; PTR64-LABEL: @testnullptrptr(
+; PTR64-NEXT:    [[END1:%.*]] = ptrtoint i8* [[END:%.*]] to i64
 ; PTR64-NEXT:    br label [[LOOPGUARD:%.*]]
 ; PTR64:       loopguard:
-; PTR64-NEXT:    [[GUARD:%.*]] = icmp ult i8* null, [[END:%.*]]
+; PTR64-NEXT:    [[GUARD:%.*]] = icmp ult i8* null, [[END]]
 ; PTR64-NEXT:    br i1 [[GUARD]], label [[PREHEADER:%.*]], label [[EXIT:%.*]]
 ; PTR64:       preheader:
+; PTR64-NEXT:    [[SCEVGEP:%.*]] = getelementptr i8, i8* null, i64 [[END1]]
 ; PTR64-NEXT:    br label [[LOOP:%.*]]
 ; PTR64:       loop:
 ; PTR64-NEXT:    [[P_01_US_US:%.*]] = phi i8* [ null, [[PREHEADER]] ], [ [[GEP:%.*]], [[LOOP]] ]
 ; PTR64-NEXT:    [[GEP]] = getelementptr inbounds i8, i8* [[P_01_US_US]], i64 1
 ; PTR64-NEXT:    [[SNEXT:%.*]] = load i8, i8* [[GEP]], align 1
-; PTR64-NEXT:    [[EXITCOND:%.*]] = icmp ne i8* [[GEP]], [[END]]
+; PTR64-NEXT:    [[EXITCOND:%.*]] = icmp ne i8* [[GEP]], [[SCEVGEP]]
 ; PTR64-NEXT:    br i1 [[EXITCOND]], label [[LOOP]], label [[EXIT_LOOPEXIT:%.*]]
 ; PTR64:       exit.loopexit:
 ; PTR64-NEXT:    [[SNEXT_LCSSA:%.*]] = phi i8 [ [[SNEXT]], [[LOOP]] ]
 ; PTR64-NEXT:    br label [[EXIT]]
 ; PTR64:       exit:
 ; PTR64-NEXT:    [[RET:%.*]] = phi i8 [ 0, [[LOOPGUARD]] ], [ [[SNEXT_LCSSA]], [[EXIT_LOOPEXIT]] ]
 ; PTR64-NEXT:    ret i8 [[RET]]
 ;
 ; PTR32-LABEL: @testnullptrptr(
+; PTR32-NEXT:    [[END1:%.*]] = ptrtoint i8* [[END:%.*]] to i32
 ; PTR32-NEXT:    br label [[LOOPGUARD:%.*]]
 ; PTR32:       loopguard:
-; PTR32-NEXT:    [[GUARD:%.*]] = icmp ult i8* null, [[END:%.*]]
+; PTR32-NEXT:    [[GUARD:%.*]] = icmp ult i8* null, [[END]]
 ; PTR32-NEXT:    br i1 [[GUARD]], label [[PREHEADER:%.*]], label [[EXIT:%.*]]
 ; PTR32:       preheader:
+; PTR32-NEXT:    [[SCEVGEP:%.*]] = getelementptr i8, i8* null, i32 [[END1]]
 ; PTR32-NEXT:    br label [[LOOP:%.*]]
 ; PTR32:       loop:
 ; PTR32-NEXT:    [[P_01_US_US:%.*]] = phi i8* [ null, [[PREHEADER]] ], [ [[GEP:%.*]], [[LOOP]] ]
 ; PTR32-NEXT:    [[GEP]] = getelementptr inbounds i8, i8* [[P_01_US_US]], i64 1
 ; PTR32-NEXT:    [[SNEXT:%.*]] = load i8, i8* [[GEP]], align 1
-; PTR32-NEXT:    [[EXITCOND:%.*]] = icmp ne i8* [[GEP]], [[END]]
+; PTR32-NEXT:    [[EXITCOND:%.*]] = icmp ne i8* [[GEP]], [[SCEVGEP]]
 ; PTR32-NEXT:    br i1 [[EXITCOND]], label [[LOOP]], label [[EXIT_LOOPEXIT:%.*]]
 ; PTR32:       exit.loopexit:
 ; PTR32-NEXT:    [[SNEXT_LCSSA:%.*]] = phi i8 [ [[SNEXT]], [[LOOP]] ]
 ; PTR32-NEXT:    br label [[EXIT]]
 ; PTR32:       exit:
 ; PTR32-NEXT:    [[RET:%.*]] = phi i8 [ 0, [[LOOPGUARD]] ], [ [[SNEXT_LCSSA]], [[EXIT_LOOPEXIT]] ]
 ; PTR32-NEXT:    ret i8 [[RET]]
 ;

 
 exit:
   %ret = phi i8 [0, %loopguard], [%snext, %loop]
   ret i8 %ret
 }
 
 define i8 @testptrptr(i8* %buf, i8* %end) nounwind {
 ; PTR64-LABEL: @testptrptr(
+; PTR64-NEXT:    [[BUF2:%.*]] = ptrtoint i8* [[BUF:%.*]] to i64
+; PTR64-NEXT:    [[END1:%.*]] = ptrtoint i8* [[END:%.*]] to i64
 ; PTR64-NEXT:    br label [[LOOPGUARD:%.*]]
 ; PTR64:       loopguard:
-; PTR64-NEXT:    [[GUARD:%.*]] = icmp ult i8* [[BUF:%.*]], [[END:%.*]]
+; PTR64-NEXT:    [[GUARD:%.*]] = icmp ult i8* [[BUF]], [[END]]
 ; PTR64-NEXT:    br i1 [[GUARD]], label [[PREHEADER:%.*]], label [[EXIT:%.*]]
 ; PTR64:       preheader:
+; PTR64-NEXT:    [[TMP1:%.*]] = sub i64 [[END1]], [[BUF2]]
+; PTR64-NEXT:    [[SCEVGEP:%.*]] = getelementptr i8, i8* [[BUF]], i64 [[TMP1]]
 ; PTR64-NEXT:    br label [[LOOP:%.*]]
 ; PTR64:       loop:
 ; PTR64-NEXT:    [[P_01_US_US:%.*]] = phi i8* [ [[BUF]], [[PREHEADER]] ], [ [[GEP:%.*]], [[LOOP]] ]
 ; PTR64-NEXT:    [[GEP]] = getelementptr inbounds i8, i8* [[P_01_US_US]], i64 1
 ; PTR64-NEXT:    [[SNEXT:%.*]] = load i8, i8* [[GEP]], align 1
-; PTR64-NEXT:    [[EXITCOND:%.*]] = icmp ne i8* [[GEP]], [[END]]
+; PTR64-NEXT:    [[EXITCOND:%.*]] = icmp ne i8* [[GEP]], [[SCEVGEP]]
 ; PTR64-NEXT:    br i1 [[EXITCOND]], label [[LOOP]], label [[EXIT_LOOPEXIT:%.*]]
 ; PTR64:       exit.loopexit:
 ; PTR64-NEXT:    [[SNEXT_LCSSA:%.*]] = phi i8 [ [[SNEXT]], [[LOOP]] ]
 ; PTR64-NEXT:    br label [[EXIT]]
 ; PTR64:       exit:
 ; PTR64-NEXT:    [[RET:%.*]] = phi i8 [ 0, [[LOOPGUARD]] ], [ [[SNEXT_LCSSA]], [[EXIT_LOOPEXIT]] ]
 ; PTR64-NEXT:    ret i8 [[RET]]
 ;
 ; PTR32-LABEL: @testptrptr(
+; PTR32-NEXT:    [[BUF2:%.*]] = ptrtoint i8* [[BUF:%.*]] to i32
+; PTR32-NEXT:    [[END1:%.*]] = ptrtoint i8* [[END:%.*]] to i32
 ; PTR32-NEXT:    br label [[LOOPGUARD:%.*]]
 ; PTR32:       loopguard:
-; PTR32-NEXT:    [[GUARD:%.*]] = icmp ult i8* [[BUF:%.*]], [[END:%.*]]
+; PTR32-NEXT:    [[GUARD:%.*]] = icmp ult i8* [[BUF]], [[END]]
 ; PTR32-NEXT:    br i1 [[GUARD]], label [[PREHEADER:%.*]], label [[EXIT:%.*]]
 ; PTR32:       preheader:
+; PTR32-NEXT:    [[TMP1:%.*]] = sub i32 [[END1]], [[BUF2]]
+; PTR32-NEXT:    [[SCEVGEP:%.*]] = getelementptr i8, i8* [[BUF]], i32 [[TMP1]]
 ; PTR32-NEXT:    br label [[LOOP:%.*]]
 ; PTR32:       loop:
 ; PTR32-NEXT:    [[P_01_US_US:%.*]] = phi i8* [ [[BUF]], [[PREHEADER]] ], [ [[GEP:%.*]], [[LOOP]] ]
 ; PTR32-NEXT:    [[GEP]] = getelementptr inbounds i8, i8* [[P_01_US_US]], i64 1
 ; PTR32-NEXT:    [[SNEXT:%.*]] = load i8, i8* [[GEP]], align 1
-; PTR32-NEXT:    [[EXITCOND:%.*]] = icmp ne i8* [[GEP]], [[END]]
+; PTR32-NEXT:    [[EXITCOND:%.*]] = icmp ne i8* [[GEP]], [[SCEVGEP]]
 ; PTR32-NEXT:    br i1 [[EXITCOND]], label [[LOOP]], label [[EXIT_LOOPEXIT:%.*]]
 ; PTR32:       exit.loopexit:
 ; PTR32-NEXT:    [[SNEXT_LCSSA:%.*]] = phi i8 [ [[SNEXT]], [[LOOP]] ]
 ; PTR32-NEXT:    br label [[EXIT]]
 ; PTR32:       exit:
 ; PTR32-NEXT:    [[RET:%.*]] = phi i8 [ 0, [[LOOPGUARD]] ], [ [[SNEXT_LCSSA]], [[EXIT_LOOPEXIT]] ]
 ; PTR32-NEXT:    ret i8 [[RET]]
 ;

 ; PTR64-NEXT:    [[ADD_PTR1603:%.*]] = getelementptr [512 x i8], [512 x i8]* [[BASE:%.*]], i64 0, i64 512
 ; PTR64-NEXT:    br label [[PREHEADER:%.*]]
 ; PTR64:       preheader:
 ; PTR64-NEXT:    [[CMP1604192:%.*]] = icmp ult i8* undef, [[ADD_PTR1603]]
 ; PTR64-NEXT:    br i1 [[CMP1604192]], label [[FOR_BODY_PREHEADER:%.*]], label [[FOR_END1609:%.*]]
 ; PTR64:       for.body.preheader:
 ; PTR64-NEXT:    br label [[FOR_BODY:%.*]]
 ; PTR64:       for.body:
-; PTR64-NEXT:    [[R_17193:%.*]] = phi i8* [ [[INCDEC_PTR1608:%.*]], [[FOR_BODY]] ], [ null, [[FOR_BODY_PREHEADER]] ]
-; PTR64-NEXT:    [[INCDEC_PTR1608]] = getelementptr i8, i8* [[R_17193]], i64 1
-; PTR64-NEXT:    [[CMP1604:%.*]] = icmp ult i8* [[INCDEC_PTR1608]], [[ADD_PTR1603]]
-; PTR64-NEXT:    br i1 [[CMP1604]], label [[FOR_BODY]], label [[FOR_END1609_LOOPEXIT:%.*]]
+; PTR64-NEXT:    br i1 false, label [[FOR_BODY]], label [[FOR_END1609_LOOPEXIT:%.*]]
 ; PTR64:       for.end1609.loopexit:
 ; PTR64-NEXT:    br label [[FOR_END1609]]
 ; PTR64:       for.end1609:
 ; PTR64-NEXT:    unreachable
 ;
 ; PTR32-LABEL: @testnullptr(
 ; PTR32-NEXT:  entry:
 ; PTR32-NEXT:    [[ADD_PTR1603:%.*]] = getelementptr [512 x i8], [512 x i8]* [[BASE:%.*]], i64 0, i64 512
 ; PTR32-NEXT:    br label [[PREHEADER:%.*]]
 ; PTR32:       preheader:
 ; PTR32-NEXT:    [[CMP1604192:%.*]] = icmp ult i8* undef, [[ADD_PTR1603]]
 ; PTR32-NEXT:    br i1 [[CMP1604192]], label [[FOR_BODY_PREHEADER:%.*]], label [[FOR_END1609:%.*]]
 ; PTR32:       for.body.preheader:
 ; PTR32-NEXT:    br label [[FOR_BODY:%.*]]
 ; PTR32:       for.body:
-; PTR32-NEXT:    [[R_17193:%.*]] = phi i8* [ [[INCDEC_PTR1608:%.*]], [[FOR_BODY]] ], [ null, [[FOR_BODY_PREHEADER]] ]
-; PTR32-NEXT:    [[INCDEC_PTR1608]] = getelementptr i8, i8* [[R_17193]], i64 1
-; PTR32-NEXT:    [[CMP1604:%.*]] = icmp ult i8* [[INCDEC_PTR1608]], [[ADD_PTR1603]]
-; PTR32-NEXT:    br i1 [[CMP1604]], label [[FOR_BODY]], label [[FOR_END1609_LOOPEXIT:%.*]]
+; PTR32-NEXT:    br i1 false, label [[FOR_BODY]], label [[FOR_END1609_LOOPEXIT:%.*]]
 ; PTR32:       for.end1609.loopexit:
 ; PTR32-NEXT:    br label [[FOR_END1609]]
 ; PTR32:       for.end1609:
 ; PTR32-NEXT:    unreachable
 ;
 entry:
   %add.ptr1603 = getelementptr [512 x i8], [512 x i8]* %base, i64 0, i64 512
   br label %preheader

llvm/test/Transforms/IndVarSimplify/eliminate-exit-no-dl.ll

 ; NOTE: Assertions have been autogenerated by utils/update_test_checks.py
 ; RUN: opt -indvars -S < %s | FileCheck %s
 
 ; Check the case where one exit has a pointer EC, and the other doesn't.
 ; Note that this test case is really really fragile.  Removing any
 ; instruction in the below causes the result to differ.  Note that the lack
 ; of a data layout (with pointer size info) is critical to getting a pointer
 ; EC returned by SCEV.
 
 @global = external global [0 x i8], align 1
 
 define void @foo() {
 ; CHECK-LABEL: @foo(
 ; CHECK-NEXT:  bb:
+; CHECK-NEXT:    [[UMIN:%.*]] = call i64 @llvm.umin.i64(i64 add (i64 ptrtoint ([0 x i8]* @global to i64), i64 500), i64 add (i64 ptrtoint ([0 x i8]* @global to i64), i64 1))
+; CHECK-NEXT:    [[TMP0:%.*]] = sub i64 add (i64 ptrtoint ([0 x i8]* @global to i64), i64 1), [[UMIN]]
+; CHECK-NEXT:    [[TMP1:%.*]] = icmp ne i64 [[TMP0]], 0
 ; CHECK-NEXT:    br label [[BB3:%.*]]
 ; CHECK:       bb3:
 ; CHECK-NEXT:    [[TMP:%.*]] = phi i8* [ [[TMP4:%.*]], [[BB7:%.*]] ], [ getelementptr inbounds ([0 x i8], [0 x i8]* @global, i64 0, i64 2), [[BB:%.*]] ]
 ; CHECK-NEXT:    [[TMP4]] = getelementptr inbounds i8, i8* [[TMP]], i64 -1
 ; CHECK-NEXT:    [[TMP6:%.*]] = load i8, i8* [[TMP4]], align 1
-; CHECK-NEXT:    [[TMP5:%.*]] = icmp ugt i8* [[TMP4]], getelementptr inbounds ([0 x i8], [0 x i8]* @global, i64 0, i64 500)
-; CHECK-NEXT:    br i1 [[TMP5]], label [[BB7]], label [[BB11:%.*]]
+; CHECK-NEXT:    br i1 [[TMP1]], label [[BB7]], label [[BB11:%.*]]
 ; CHECK:       bb7:
 ; CHECK-NEXT:    [[TMP8:%.*]] = zext i8 [[TMP6]] to i64
 ; CHECK-NEXT:    br i1 true, label [[BB11]], label [[BB3]]
 ; CHECK:       bb11:
 ; CHECK-NEXT:    ret void
 ;
 bb:
   br label %bb3

llvm/test/Transforms/LoopIdiom/memset-debugify-remarks.ll

 ;     *begin = value;
 ; }
 
 ; CHECK: remark: <stdin>:4:1: Transformed loop-strided store in _Z15my_basic_memsetPcS_c function into a call to llvm.memset.p0i8.i64() intrinsic
 
 define void @_Z15my_basic_memsetPcS_c(i8* %ptr, i8* %end, i8 %value) {
 ; CHECK-LABEL: @_Z15my_basic_memsetPcS_c(
 ; CHECK-NEXT:  entry:
-; CHECK-NEXT:    [[PTR1:%.*]] = ptrtoint i8* [[PTR:%.*]] to i64, !dbg [[DBG15:![0-9]+]]
-; CHECK-NEXT:    [[CMP3:%.*]] = icmp eq i8* [[PTR]], [[END:%.*]], !dbg [[DBG15]]
+; CHECK-NEXT:    [[PTR2:%.*]] = ptrtoint i8* [[PTR:%.*]] to i64, !dbg [[DBG15:![0-9]+]]
+; CHECK-NEXT:    [[END1:%.*]] = ptrtoint i8* [[END:%.*]] to i64, !dbg [[DBG15]]
+; CHECK-NEXT:    [[CMP3:%.*]] = icmp eq i8* [[PTR]], [[END]], !dbg [[DBG15]]
 ; CHECK-NEXT:    call void @llvm.dbg.value(metadata i1 [[CMP3]], metadata [[META9:![0-9]+]], metadata !DIExpression()), !dbg [[DBG15]]
 ; CHECK-NEXT:    br i1 [[CMP3]], label [[FOR_END:%.*]], label [[FOR_BODY_PREHEADER:%.*]], !dbg [[DBG16:![0-9]+]]
 ; CHECK:       for.body.preheader:
-; CHECK-NEXT:    [[TMP0:%.*]] = sub i64 0, [[PTR1]], !dbg [[DBG17:![0-9]+]]
-; CHECK-NEXT:    [[SCEVGEP:%.*]] = getelementptr i8, i8* [[END]], i64 [[TMP0]], !dbg [[DBG17]]
-; CHECK-NEXT:    [[SCEVGEP2:%.*]] = ptrtoint i8* [[SCEVGEP]] to i64, !dbg [[DBG17]]
-; CHECK-NEXT:    call void @llvm.memset.p0i8.i64(i8* align 1 [[PTR]], i8 [[VALUE:%.*]], i64 [[SCEVGEP2]], i1 false), !dbg [[DBG18:![0-9]+]]
+; CHECK-NEXT:    [[TMP0:%.*]] = sub i64 [[END1]], [[PTR2]], !dbg [[DBG17:![0-9]+]]
+; CHECK-NEXT:    call void @llvm.memset.p0i8.i64(i8* align 1 [[PTR]], i8 [[VALUE:%.*]], i64 [[TMP0]], i1 false), !dbg [[DBG18:![0-9]+]]
 ; CHECK-NEXT:    br label [[FOR_BODY:%.*]], !dbg [[DBG17]]
 ; CHECK:       for.body:
 ; CHECK-NEXT:    [[PTR_ADDR_04:%.*]] = phi i8* [ [[INCDEC_PTR:%.*]], [[FOR_BODY]] ], [ [[PTR]], [[FOR_BODY_PREHEADER]] ], !dbg [[DBG19:![0-9]+]]
 ; CHECK-NEXT:    call void @llvm.dbg.value(metadata i8* [[PTR_ADDR_04]], metadata [[META11:![0-9]+]], metadata !DIExpression()), !dbg [[DBG19]]
 ; CHECK-NEXT:    [[INCDEC_PTR]] = getelementptr inbounds i8, i8* [[PTR_ADDR_04]], i64 1, !dbg [[DBG20:![0-9]+]]
 ; CHECK-NEXT:    call void @llvm.dbg.value(metadata i8* [[INCDEC_PTR]], metadata [[META13:![0-9]+]], metadata !DIExpression()), !dbg [[DBG20]]
 ; CHECK-NEXT:    [[CMP:%.*]] = icmp eq i8* [[INCDEC_PTR]], [[END]], !dbg [[DBG21:![0-9]+]]
 ; CHECK-NEXT:    call void @llvm.dbg.value(metadata i1 [[CMP]], metadata [[META14:![0-9]+]], metadata !DIExpression()), !dbg [[DBG21]]

llvm/test/Transforms/LoopReroll/ptrindvar.ll

 ; RUN: opt -S  -loop-reroll   %s | FileCheck %s
 target triple = "aarch64--linux-gnu"
 
 define i32 @test(i32* readonly %buf, i32* readnone %end) #0 {
 entry:
   %cmp.9 = icmp eq i32* %buf, %end
   br i1 %cmp.9, label %while.end, label %while.body.preheader
 
 while.body.preheader:
   br label %while.body
 
 while.body:
 ;CHECK-LABEL: while.body:
 ;CHECK-NEXT:    %indvar = phi i64 [ %indvar.next, %while.body ], [ 0, %while.body.preheader ]
 ;CHECK-NEXT:    %S.011 = phi i32 [ %add, %while.body ], [ undef, %while.body.preheader ]
 ;CHECK-NEXT:    %scevgep = getelementptr i32, i32* %buf, i64 %indvar
-;CHECK-NEXT:    %4 = load i32, i32* %scevgep, align 4
-;CHECK-NEXT:    %add = add nsw i32 %4, %S.011
+;CHECK-NEXT:    %5 = load i32, i32* %scevgep, align 4
+;CHECK-NEXT:    %add = add nsw i32 %5, %S.011
 ;CHECK-NEXT:    %indvar.next = add i64 %indvar, 1
-;CHECK-NEXT:    %exitcond = icmp eq i64 %indvar, %3
+;CHECK-NEXT:    %exitcond = icmp eq i64 %indvar, %4
 ;CHECK-NEXT:    br i1 %exitcond, label %while.end.loopexit, label %while.body
 
   %S.011 = phi i32 [ %add2, %while.body ], [ undef, %while.body.preheader ]
   %buf.addr.010 = phi i32* [ %add.ptr, %while.body ], [ %buf, %while.body.preheader ]
   %0 = load i32, i32* %buf.addr.010, align 4
   %add = add nsw i32 %0, %S.011
   %arrayidx1 = getelementptr inbounds i32, i32* %buf.addr.010, i64 1
   %1 = load i32, i32* %arrayidx1, align 4

 
 while.body.preheader:
   br label %while.body
 
 while.body:
 ;CHECK-LABEL: while.body:
 ;CHECK-NEXT:    %indvar = phi i64 [ %indvar.next, %while.body ], [ 0, %while.body.preheader ]
 ;CHECK-NEXT:    %S.011 = phi i32 [ %add, %while.body ], [ undef, %while.body.preheader ]
-;CHECK-NEXT:    %4 = mul nsw i64 %indvar, -1
-;CHECK-NEXT:    %scevgep = getelementptr i32, i32* %buf, i64 %4
-;CHECK-NEXT:    %5 = load i32, i32* %scevgep, align 4
-;CHECK-NEXT:    %add = add nsw i32 %5, %S.011
+;CHECK-NEXT:    %5 = mul nsw i64 %indvar, -1
+;CHECK-NEXT:    %scevgep = getelementptr i32, i32* %buf, i64 %5
+;CHECK-NEXT:    %6 = load i32, i32* %scevgep, align 4
+;CHECK-NEXT:    %add = add nsw i32 %6, %S.011
 ;CHECK-NEXT:    %indvar.next = add i64 %indvar, 1
-;CHECK-NEXT:    %exitcond = icmp eq i64 %indvar, %3
+;CHECK-NEXT:    %exitcond = icmp eq i64 %indvar, %4
 ;CHECK-NEXT:    br i1 %exitcond, label %while.end.loopexit, label %while.body
 
   %S.011 = phi i32 [ %add2, %while.body ], [ undef, %while.body.preheader ]
   %buf.addr.010 = phi i32* [ %add.ptr, %while.body ], [ %buf, %while.body.preheader ]
   %0 = load i32, i32* %buf.addr.010, align 4
   %add = add nsw i32 %0, %S.011
   %arrayidx1 = getelementptr inbounds i32, i32* %buf.addr.010, i64 -1
   %1 = load i32, i32* %arrayidx1, align 4

llvm/test/Transforms/LoopVectorize/X86/cost-model-assert.ll

 target datalayout = "e-m:w-p270:32:32-p271:32:32-p272:64:64-i64:64-f80:128-n8:16:32:64-S128"
 target triple = "x86_64-w64-windows-gnu"
 
 define void @cff_index_load_offsets(i1 %cond, i8 %x, i8* %p) #0 {
 ; CHECK-LABEL: @cff_index_load_offsets(
 ; CHECK-NEXT:  entry:
 ; CHECK-NEXT:    br i1 [[COND:%.*]], label [[IF_THEN:%.*]], label [[EXIT:%.*]]
 ; CHECK:       if.then:
-; CHECK-NEXT:    br i1 true, label [[SCALAR_PH:%.*]], label [[VECTOR_PH:%.*]]
+; CHECK-NEXT:    [[UMAX:%.*]] = call i64 @llvm.umax.i64(i64 undef, i64 4)
+; CHECK-NEXT:    [[TMP0:%.*]] = add i64 [[UMAX]], -1
+; CHECK-NEXT:    [[TMP1:%.*]] = lshr i64 [[TMP0]], 2
+; CHECK-NEXT:    [[TMP2:%.*]] = add nuw nsw i64 [[TMP1]], 1
+; CHECK-NEXT:    [[MIN_ITERS_CHECK:%.*]] = icmp ult i64 [[TMP2]], 8
+; CHECK-NEXT:    br i1 [[MIN_ITERS_CHECK]], label [[SCALAR_PH:%.*]], label [[VECTOR_PH:%.*]]
 ; CHECK:       vector.ph:
+; CHECK-NEXT:    [[N_MOD_VF:%.*]] = urem i64 [[TMP2]], 8
+; CHECK-NEXT:    [[N_VEC:%.*]] = sub i64 [[TMP2]], [[N_MOD_VF]]
+; CHECK-NEXT:    [[TMP3:%.*]] = mul i64 [[N_VEC]], 4
+; CHECK-NEXT:    [[IND_END:%.*]] = getelementptr i8, i8* null, i64 [[TMP3]]
 ; CHECK-NEXT:    [[BROADCAST_SPLATINSERT:%.*]] = insertelement <4 x i8> poison, i8 [[X:%.*]], i32 0
 ; CHECK-NEXT:    [[BROADCAST_SPLAT:%.*]] = shufflevector <4 x i8> [[BROADCAST_SPLATINSERT]], <4 x i8> poison, <4 x i32> zeroinitializer
 ; CHECK-NEXT:    [[BROADCAST_SPLATINSERT2:%.*]] = insertelement <4 x i8> poison, i8 [[X]], i32 0
 ; CHECK-NEXT:    [[BROADCAST_SPLAT3:%.*]] = shufflevector <4 x i8> [[BROADCAST_SPLATINSERT2]], <4 x i8> poison, <4 x i32> zeroinitializer
 ; CHECK-NEXT:    br label [[VECTOR_BODY:%.*]]
 ; CHECK:       vector.body:
 ; CHECK-NEXT:    [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
-; CHECK-NEXT:    [[TMP0:%.*]] = add i64 [[INDEX]], 0
-; CHECK-NEXT:    [[TMP1:%.*]] = mul i64 [[TMP0]], 4
-; CHECK-NEXT:    [[NEXT_GEP:%.*]] = getelementptr i8, i8* null, i64 [[TMP1]]
-; CHECK-NEXT:    [[TMP2:%.*]] = add i64 [[INDEX]], 4
-; CHECK-NEXT:    [[TMP3:%.*]] = mul i64 [[TMP2]], 4
-; CHECK-NEXT:    [[NEXT_GEP1:%.*]] = getelementptr i8, i8* null, i64 [[TMP3]]
-; CHECK-NEXT:    [[TMP4:%.*]] = zext <4 x i8> [[BROADCAST_SPLAT]] to <4 x i32>
-; CHECK-NEXT:    [[TMP5:%.*]] = zext <4 x i8> [[BROADCAST_SPLAT3]] to <4 x i32>
-; CHECK-NEXT:    [[TMP6:%.*]] = shl nuw <4 x i32> [[TMP4]], <i32 24, i32 24, i32 24, i32 24>
-; CHECK-NEXT:    [[TMP7:%.*]] = shl nuw <4 x i32> [[TMP5]], <i32 24, i32 24, i32 24, i32 24>
-; CHECK-NEXT:    [[TMP8:%.*]] = load i8, i8* [[P:%.*]], align 1, !tbaa [[TBAA1:![0-9]+]]
-; CHECK-NEXT:    [[BROADCAST_SPLATINSERT4:%.*]] = insertelement <4 x i8> poison, i8 [[TMP8]], i32 0
+; CHECK-NEXT:    [[TMP4:%.*]] = add i64 [[INDEX]], 0
+; CHECK-NEXT:    [[TMP5:%.*]] = mul i64 [[TMP4]], 4
+; CHECK-NEXT:    [[NEXT_GEP:%.*]] = getelementptr i8, i8* null, i64 [[TMP5]]
+; CHECK-NEXT:    [[TMP6:%.*]] = add i64 [[INDEX]], 4
+; CHECK-NEXT:    [[TMP7:%.*]] = mul i64 [[TMP6]], 4
+; CHECK-NEXT:    [[NEXT_GEP1:%.*]] = getelementptr i8, i8* null, i64 [[TMP7]]
+; CHECK-NEXT:    [[TMP8:%.*]] = zext <4 x i8> [[BROADCAST_SPLAT]] to <4 x i32>
+; CHECK-NEXT:    [[TMP9:%.*]] = zext <4 x i8> [[BROADCAST_SPLAT3]] to <4 x i32>
+; CHECK-NEXT:    [[TMP10:%.*]] = shl nuw <4 x i32> [[TMP8]], <i32 24, i32 24, i32 24, i32 24>
+; CHECK-NEXT:    [[TMP11:%.*]] = shl nuw <4 x i32> [[TMP9]], <i32 24, i32 24, i32 24, i32 24>
+; CHECK-NEXT:    [[TMP12:%.*]] = load i8, i8* [[P:%.*]], align 1, !tbaa [[TBAA1:![0-9]+]]
+; CHECK-NEXT:    [[BROADCAST_SPLATINSERT4:%.*]] = insertelement <4 x i8> poison, i8 [[TMP12]], i32 0
 ; CHECK-NEXT:    [[BROADCAST_SPLAT5:%.*]] = shufflevector <4 x i8> [[BROADCAST_SPLATINSERT4]], <4 x i8> poison, <4 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP9:%.*]] = load i8, i8* [[P]], align 1, !tbaa [[TBAA1]]
-; CHECK-NEXT:    [[BROADCAST_SPLATINSERT6:%.*]] = insertelement <4 x i8> poison, i8 [[TMP9]], i32 0
+; CHECK-NEXT:    [[TMP13:%.*]] = load i8, i8* [[P]], align 1, !tbaa [[TBAA1]]
+; CHECK-NEXT:    [[BROADCAST_SPLATINSERT6:%.*]] = insertelement <4 x i8> poison, i8 [[TMP13]], i32 0
 ; CHECK-NEXT:    [[BROADCAST_SPLAT7:%.*]] = shufflevector <4 x i8> [[BROADCAST_SPLATINSERT6]], <4 x i8> poison, <4 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP10:%.*]] = zext <4 x i8> [[BROADCAST_SPLAT5]] to <4 x i32>
-; CHECK-NEXT:    [[TMP11:%.*]] = zext <4 x i8> [[BROADCAST_SPLAT7]] to <4 x i32>
-; CHECK-NEXT:    [[TMP12:%.*]] = shl nuw nsw <4 x i32> [[TMP10]], <i32 16, i32 16, i32 16, i32 16>
-; CHECK-NEXT:    [[TMP13:%.*]] = shl nuw nsw <4 x i32> [[TMP11]], <i32 16, i32 16, i32 16, i32 16>
-; CHECK-NEXT:    [[TMP14:%.*]] = or <4 x i32> [[TMP12]], [[TMP6]]
-; CHECK-NEXT:    [[TMP15:%.*]] = or <4 x i32> [[TMP13]], [[TMP7]]
-; CHECK-NEXT:    [[TMP16:%.*]] = load i8, i8* undef, align 1, !tbaa [[TBAA1]]
-; CHECK-NEXT:    [[TMP17:%.*]] = load i8, i8* undef, align 1, !tbaa [[TBAA1]]
-; CHECK-NEXT:    [[TMP18:%.*]] = or <4 x i32> [[TMP14]], zeroinitializer
-; CHECK-NEXT:    [[TMP19:%.*]] = or <4 x i32> [[TMP15]], zeroinitializer
-; CHECK-NEXT:    [[TMP20:%.*]] = or <4 x i32> [[TMP18]], zeroinitializer
-; CHECK-NEXT:    [[TMP21:%.*]] = or <4 x i32> [[TMP19]], zeroinitializer
-; CHECK-NEXT:    [[TMP22:%.*]] = extractelement <4 x i32> [[TMP20]], i32 0
-; CHECK-NEXT:    store i32 [[TMP22]], i32* undef, align 4, !tbaa [[TBAA4:![0-9]+]]
-; CHECK-NEXT:    [[TMP23:%.*]] = extractelement <4 x i32> [[TMP20]], i32 1
-; CHECK-NEXT:    store i32 [[TMP23]], i32* undef, align 4, !tbaa [[TBAA4]]
-; CHECK-NEXT:    [[TMP24:%.*]] = extractelement <4 x i32> [[TMP20]], i32 2
-; CHECK-NEXT:    store i32 [[TMP24]], i32* undef, align 4, !tbaa [[TBAA4]]
-; CHECK-NEXT:    [[TMP25:%.*]] = extractelement <4 x i32> [[TMP20]], i32 3
-; CHECK-NEXT:    store i32 [[TMP25]], i32* undef, align 4, !tbaa [[TBAA4]]
-; CHECK-NEXT:    [[TMP26:%.*]] = extractelement <4 x i32> [[TMP21]], i32 0
-; CHECK-NEXT:    store i32 [[TMP26]], i32* undef, align 4, !tbaa [[TBAA4]]
-; CHECK-NEXT:    [[TMP27:%.*]] = extractelement <4 x i32> [[TMP21]], i32 1
+; CHECK-NEXT:    [[TMP14:%.*]] = zext <4 x i8> [[BROADCAST_SPLAT5]] to <4 x i32>
+; CHECK-NEXT:    [[TMP15:%.*]] = zext <4 x i8> [[BROADCAST_SPLAT7]] to <4 x i32>
+; CHECK-NEXT:    [[TMP16:%.*]] = shl nuw nsw <4 x i32> [[TMP14]], <i32 16, i32 16, i32 16, i32 16>
+; CHECK-NEXT:    [[TMP17:%.*]] = shl nuw nsw <4 x i32> [[TMP15]], <i32 16, i32 16, i32 16, i32 16>
+; CHECK-NEXT:    [[TMP18:%.*]] = or <4 x i32> [[TMP16]], [[TMP10]]
+; CHECK-NEXT:    [[TMP19:%.*]] = or <4 x i32> [[TMP17]], [[TMP11]]
+; CHECK-NEXT:    [[TMP20:%.*]] = load i8, i8* undef, align 1, !tbaa [[TBAA1]]
+; CHECK-NEXT:    [[TMP21:%.*]] = load i8, i8* undef, align 1, !tbaa [[TBAA1]]
+; CHECK-NEXT:    [[TMP22:%.*]] = or <4 x i32> [[TMP18]], zeroinitializer
+; CHECK-NEXT:    [[TMP23:%.*]] = or <4 x i32> [[TMP19]], zeroinitializer
+; CHECK-NEXT:    [[TMP24:%.*]] = or <4 x i32> [[TMP22]], zeroinitializer
+; CHECK-NEXT:    [[TMP25:%.*]] = or <4 x i32> [[TMP23]], zeroinitializer
+; CHECK-NEXT:    [[TMP26:%.*]] = extractelement <4 x i32> [[TMP24]], i32 0
+; CHECK-NEXT:    store i32 [[TMP26]], i32* undef, align 4, !tbaa [[TBAA4:![0-9]+]]
+; CHECK-NEXT:    [[TMP27:%.*]] = extractelement <4 x i32> [[TMP24]], i32 1
 ; CHECK-NEXT:    store i32 [[TMP27]], i32* undef, align 4, !tbaa [[TBAA4]]
-; CHECK-NEXT:    [[TMP28:%.*]] = extractelement <4 x i32> [[TMP21]], i32 2
+; CHECK-NEXT:    [[TMP28:%.*]] = extractelement <4 x i32> [[TMP24]], i32 2
 ; CHECK-NEXT:    store i32 [[TMP28]], i32* undef, align 4, !tbaa [[TBAA4]]
-; CHECK-NEXT:    [[TMP29:%.*]] = extractelement <4 x i32> [[TMP21]], i32 3
+; CHECK-NEXT:    [[TMP29:%.*]] = extractelement <4 x i32> [[TMP24]], i32 3
 ; CHECK-NEXT:    store i32 [[TMP29]], i32* undef, align 4, !tbaa [[TBAA4]]
+; CHECK-NEXT:    [[TMP30:%.*]] = extractelement <4 x i32> [[TMP25]], i32 0
+; CHECK-NEXT:    store i32 [[TMP30]], i32* undef, align 4, !tbaa [[TBAA4]]
+; CHECK-NEXT:    [[TMP31:%.*]] = extractelement <4 x i32> [[TMP25]], i32 1
+; CHECK-NEXT:    store i32 [[TMP31]], i32* undef, align 4, !tbaa [[TBAA4]]
+; CHECK-NEXT:    [[TMP32:%.*]] = extractelement <4 x i32> [[TMP25]], i32 2
+; CHECK-NEXT:    store i32 [[TMP32]], i32* undef, align 4, !tbaa [[TBAA4]]
+; CHECK-NEXT:    [[TMP33:%.*]] = extractelement <4 x i32> [[TMP25]], i32 3
+; CHECK-NEXT:    store i32 [[TMP33]], i32* undef, align 4, !tbaa [[TBAA4]]
 ; CHECK-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], 8
-; CHECK-NEXT:    [[TMP30:%.*]] = icmp eq i64 [[INDEX_NEXT]], 0
-; CHECK-NEXT:    br i1 [[TMP30]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP6:![0-9]+]]
+; CHECK-NEXT:    [[TMP34:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]]
+; CHECK-NEXT:    br i1 [[TMP34]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP6:![0-9]+]]
 ; CHECK:       middle.block:
-; CHECK-NEXT:    [[CMP_N:%.*]] = icmp eq i64 1, 0
+; CHECK-NEXT:    [[CMP_N:%.*]] = icmp eq i64 [[TMP2]], [[N_VEC]]
 ; CHECK-NEXT:    br i1 [[CMP_N]], label [[SW_EPILOG:%.*]], label [[SCALAR_PH]]
 ; CHECK:       scalar.ph:
-; CHECK-NEXT:    [[BC_RESUME_VAL:%.*]] = phi i8* [ null, [[MIDDLE_BLOCK]] ], [ null, [[IF_THEN]] ]
+; CHECK-NEXT:    [[BC_RESUME_VAL:%.*]] = phi i8* [ [[IND_END]], [[MIDDLE_BLOCK]] ], [ null, [[IF_THEN]] ]
 ; CHECK-NEXT:    br label [[FOR_BODY68:%.*]]
 ; CHECK:       for.body68:
 ; CHECK-NEXT:    [[P_359:%.*]] = phi i8* [ [[ADD_PTR86:%.*]], [[FOR_BODY68]] ], [ [[BC_RESUME_VAL]], [[SCALAR_PH]] ]
 ; CHECK-NEXT:    [[CONV70:%.*]] = zext i8 [[X]] to i32
 ; CHECK-NEXT:    [[SHL71:%.*]] = shl nuw i32 [[CONV70]], 24
-; CHECK-NEXT:    [[TMP31:%.*]] = load i8, i8* [[P]], align 1, !tbaa [[TBAA1]]
-; CHECK-NEXT:    [[CONV73:%.*]] = zext i8 [[TMP31]] to i32
+; CHECK-NEXT:    [[TMP35:%.*]] = load i8, i8* [[P]], align 1, !tbaa [[TBAA1]]
+; CHECK-NEXT:    [[CONV73:%.*]] = zext i8 [[TMP35]] to i32
 ; CHECK-NEXT:    [[SHL74:%.*]] = shl nuw nsw i32 [[CONV73]], 16
 ; CHECK-NEXT:    [[OR75:%.*]] = or i32 [[SHL74]], [[SHL71]]
-; CHECK-NEXT:    [[TMP32:%.*]] = load i8, i8* undef, align 1, !tbaa [[TBAA1]]
+; CHECK-NEXT:    [[TMP36:%.*]] = load i8, i8* undef, align 1, !tbaa [[TBAA1]]
 ; CHECK-NEXT:    [[SHL78:%.*]] = shl nuw nsw i32 undef, 8
 ; CHECK-NEXT:    [[OR79:%.*]] = or i32 [[OR75]], [[SHL78]]
 ; CHECK-NEXT:    [[CONV81:%.*]] = zext i8 undef to i32
 ; CHECK-NEXT:    [[OR83:%.*]] = or i32 [[OR79]], [[CONV81]]
 ; CHECK-NEXT:    store i32 [[OR83]], i32* undef, align 4, !tbaa [[TBAA4]]
 ; CHECK-NEXT:    [[ADD_PTR86]] = getelementptr inbounds i8, i8* [[P_359]], i64 4
 ; CHECK-NEXT:    [[CMP66:%.*]] = icmp ult i8* [[ADD_PTR86]], undef
 ; CHECK-NEXT:    br i1 [[CMP66]], label [[FOR_BODY68]], label [[SW_EPILOG]], !llvm.loop [[LOOP8:![0-9]+]]

llvm/test/Transforms/LoopVectorize/pointer-induction.ll

 ; NOTE: Assertions have been autogenerated by utils/update_test_checks.py
 ; RUN: opt < %s  -loop-vectorize -force-vector-interleave=1 -force-vector-width=4 -S | FileCheck %s
 target datalayout = "e-m:e-p270:32:32-p271:32:32-p272:64:64-i64:64-f80:128-n8:16:32:64-S128"
 
 
 ; Function Attrs: nofree norecurse nounwind
 define void @a(i8* readnone %b) {
 ; CHECK-LABEL: @a(
 ; CHECK-NEXT:  entry:
 ; CHECK-NEXT:    [[B1:%.*]] = ptrtoint i8* [[B:%.*]] to i64
 ; CHECK-NEXT:    [[CMP_NOT4:%.*]] = icmp eq i8* [[B]], null
 ; CHECK-NEXT:    br i1 [[CMP_NOT4]], label [[FOR_COND_CLEANUP:%.*]], label [[FOR_BODY_PREHEADER:%.*]]
 ; CHECK:       for.body.preheader:
 ; CHECK-NEXT:    [[TMP0:%.*]] = sub i64 0, [[B1]]
-; CHECK-NEXT:    [[SCEVGEP:%.*]] = getelementptr i8, i8* null, i64 [[TMP0]]
-; CHECK-NEXT:    [[EXITCOUNT_PTRCNT_TO_INT:%.*]] = ptrtoint i8* [[SCEVGEP]] to i64
-; CHECK-NEXT:    [[MIN_ITERS_CHECK:%.*]] = icmp ult i64 [[EXITCOUNT_PTRCNT_TO_INT]], 4
+; CHECK-NEXT:    [[MIN_ITERS_CHECK:%.*]] = icmp ult i64 [[TMP0]], 4
 ; CHECK-NEXT:    br i1 [[MIN_ITERS_CHECK]], label [[SCALAR_PH:%.*]], label [[VECTOR_PH:%.*]]
 ; CHECK:       vector.ph:
-; CHECK-NEXT:    [[N_MOD_VF:%.*]] = urem i64 [[EXITCOUNT_PTRCNT_TO_INT]], 4
-; CHECK-NEXT:    [[N_VEC:%.*]] = sub i64 [[EXITCOUNT_PTRCNT_TO_INT]], [[N_MOD_VF]]
+; CHECK-NEXT:    [[N_MOD_VF:%.*]] = urem i64 [[TMP0]], 4
+; CHECK-NEXT:    [[N_VEC:%.*]] = sub i64 [[TMP0]], [[N_MOD_VF]]
 ; CHECK-NEXT:    [[TMP1:%.*]] = mul i64 [[N_VEC]], -1
 ; CHECK-NEXT:    [[IND_END:%.*]] = getelementptr i8, i8* null, i64 [[TMP1]]
 ; CHECK-NEXT:    br label [[VECTOR_BODY:%.*]]
 ; CHECK:       vector.body:
 ; CHECK-NEXT:    [[POINTER_PHI:%.*]] = phi i8* [ null, [[VECTOR_PH]] ], [ [[PTR_IND:%.*]], [[PRED_STORE_CONTINUE7:%.*]] ]
 ; CHECK-NEXT:    [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[PRED_STORE_CONTINUE7]] ]
 ; CHECK-NEXT:    [[TMP2:%.*]] = getelementptr i8, i8* [[POINTER_PHI]], <4 x i64> <i64 0, i64 -1, i64 -2, i64 -3>
 ; CHECK-NEXT:    [[TMP3:%.*]] = getelementptr inbounds i8, <4 x i8*> [[TMP2]], i64 -1

 ; CHECK-NEXT:    store i8 95, i8* [[TMP17]], align 1
 ; CHECK-NEXT:    br label [[PRED_STORE_CONTINUE7]]
 ; CHECK:       pred.store.continue7:
 ; CHECK-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], 4
 ; CHECK-NEXT:    [[TMP18:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]]
 ; CHECK-NEXT:    [[PTR_IND]] = getelementptr i8, i8* [[POINTER_PHI]], i64 -4
 ; CHECK-NEXT:    br i1 [[TMP18]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP0:![0-9]+]]
 ; CHECK:       middle.block:
-; CHECK-NEXT:    [[CMP_N:%.*]] = icmp eq i64 [[EXITCOUNT_PTRCNT_TO_INT]], [[N_VEC]]
+; CHECK-NEXT:    [[CMP_N:%.*]] = icmp eq i64 [[TMP0]], [[N_VEC]]
 ; CHECK-NEXT:    br i1 [[CMP_N]], label [[FOR_COND_CLEANUP_LOOPEXIT:%.*]], label [[SCALAR_PH]]
 ; CHECK:       scalar.ph:
 ; CHECK-NEXT:    [[BC_RESUME_VAL:%.*]] = phi i8* [ [[IND_END]], [[MIDDLE_BLOCK]] ], [ null, [[FOR_BODY_PREHEADER]] ]
 ; CHECK-NEXT:    br label [[FOR_BODY:%.*]]
 ; CHECK:       for.cond.cleanup.loopexit:
 ; CHECK-NEXT:    br label [[FOR_COND_CLEANUP]]
 ; CHECK:       for.cond.cleanup:
 ; CHECK-NEXT:    ret void

llvm/test/Transforms/LoopVectorize/pr45259.ll

 ; NOTE: Assertions have been autogenerated by utils/update_test_checks.py
 ; RUN: opt < %s -S -loop-vectorize -force-vector-width=4 -force-vector-interleave=1 | FileCheck %s
 
 ; Check that we can vectorize this loop without crashing.
 
 define i8 @widget(i8* %arr, i8 %t9) {
 ; CHECK-LABEL: @widget(
 ; CHECK-NEXT:  bb:
-; CHECK-NEXT:    [[ARR1:%.*]] = ptrtoint i8* [[ARR:%.*]] to i64
+; CHECK-NEXT:    [[ARR2:%.*]] = ptrtoint i8* [[ARR:%.*]] to i64
 ; CHECK-NEXT:    br label [[BB6:%.*]]
 ; CHECK:       bb6:
 ; CHECK-NEXT:    [[T1_0:%.*]] = phi i8* [ [[ARR]], [[BB:%.*]] ], [ null, [[BB6]] ]
 ; CHECK-NEXT:    [[C:%.*]] = call i1 @cond()
 ; CHECK-NEXT:    br i1 [[C]], label [[FOR_PREHEADER:%.*]], label [[BB6]]
 ; CHECK:       for.preheader:
 ; CHECK-NEXT:    [[T1_0_LCSSA:%.*]] = phi i8* [ [[T1_0]], [[BB6]] ]
-; CHECK-NEXT:    [[T1_0_LCSSA3:%.*]] = ptrtoint i8* [[T1_0_LCSSA]] to i64
-; CHECK-NEXT:    [[TMP0:%.*]] = trunc i64 [[T1_0_LCSSA3]] to i32
-; CHECK-NEXT:    [[TMP1:%.*]] = trunc i64 [[ARR1]] to i32
+; CHECK-NEXT:    [[T1_0_LCSSA1:%.*]] = ptrtoint i8* [[T1_0_LCSSA]] to i64
+; CHECK-NEXT:    [[TMP0:%.*]] = trunc i64 [[T1_0_LCSSA1]] to i32
+; CHECK-NEXT:    [[TMP1:%.*]] = trunc i64 [[ARR2]] to i32
 ; CHECK-NEXT:    [[TMP2:%.*]] = sub i32 [[TMP0]], [[TMP1]]
 ; CHECK-NEXT:    [[MIN_ITERS_CHECK:%.*]] = icmp ult i32 [[TMP2]], 4
 ; CHECK-NEXT:    br i1 [[MIN_ITERS_CHECK]], label [[SCALAR_PH:%.*]], label [[VECTOR_SCEVCHECK:%.*]]
 ; CHECK:       vector.scevcheck:
-; CHECK-NEXT:    [[TMP3:%.*]] = sub i64 -1, [[ARR1]]
-; CHECK-NEXT:    [[SCEVGEP:%.*]] = getelementptr i8, i8* [[T1_0_LCSSA]], i64 [[TMP3]]
-; CHECK-NEXT:    [[SCEVGEP2:%.*]] = ptrtoint i8* [[SCEVGEP]] to i64
-; CHECK-NEXT:    [[TMP4:%.*]] = trunc i64 [[SCEVGEP2]] to i8
-; CHECK-NEXT:    [[MUL:%.*]] = call { i8, i1 } @llvm.umul.with.overflow.i8(i8 1, i8 [[TMP4]])
+; CHECK-NEXT:    [[TMP3:%.*]] = add i64 [[T1_0_LCSSA1]], -1
+; CHECK-NEXT:    [[TMP4:%.*]] = sub i64 [[TMP3]], [[ARR2]]
+; CHECK-NEXT:    [[TMP5:%.*]] = trunc i64 [[TMP4]] to i8
+; CHECK-NEXT:    [[MUL:%.*]] = call { i8, i1 } @llvm.umul.with.overflow.i8(i8 1, i8 [[TMP5]])
 ; CHECK-NEXT:    [[MUL_RESULT:%.*]] = extractvalue { i8, i1 } [[MUL]], 0
 ; CHECK-NEXT:    [[MUL_OVERFLOW:%.*]] = extractvalue { i8, i1 } [[MUL]], 1
-; CHECK-NEXT:    [[TMP5:%.*]] = add i8 1, [[MUL_RESULT]]
-; CHECK-NEXT:    [[TMP6:%.*]] = sub i8 1, [[MUL_RESULT]]
-; CHECK-NEXT:    [[TMP7:%.*]] = icmp sgt i8 [[TMP6]], 1
-; CHECK-NEXT:    [[TMP8:%.*]] = icmp slt i8 [[TMP5]], 1
-; CHECK-NEXT:    [[TMP9:%.*]] = select i1 false, i1 [[TMP7]], i1 [[TMP8]]
-; CHECK-NEXT:    [[TMP10:%.*]] = icmp ugt i64 [[SCEVGEP2]], 255
-; CHECK-NEXT:    [[TMP11:%.*]] = or i1 [[TMP9]], [[TMP10]]
-; CHECK-NEXT:    [[TMP12:%.*]] = or i1 [[TMP11]], [[MUL_OVERFLOW]]
-; CHECK-NEXT:    [[TMP13:%.*]] = or i1 false, [[TMP12]]
-; CHECK-NEXT:    br i1 [[TMP13]], label [[SCALAR_PH]], label [[VECTOR_PH:%.*]]
+; CHECK-NEXT:    [[TMP6:%.*]] = add i8 1, [[MUL_RESULT]]
+; CHECK-NEXT:    [[TMP7:%.*]] = sub i8 1, [[MUL_RESULT]]
+; CHECK-NEXT:    [[TMP8:%.*]] = icmp sgt i8 [[TMP7]], 1
+; CHECK-NEXT:    [[TMP9:%.*]] = icmp slt i8 [[TMP6]], 1
+; CHECK-NEXT:    [[TMP10:%.*]] = select i1 false, i1 [[TMP8]], i1 [[TMP9]]
+; CHECK-NEXT:    [[TMP11:%.*]] = icmp ugt i64 [[TMP4]], 255
+; CHECK-NEXT:    [[TMP12:%.*]] = or i1 [[TMP10]], [[TMP11]]
+; CHECK-NEXT:    [[TMP13:%.*]] = or i1 [[TMP12]], [[MUL_OVERFLOW]]
+; CHECK-NEXT:    [[TMP14:%.*]] = or i1 false, [[TMP13]]
+; CHECK-NEXT:    br i1 [[TMP14]], label [[SCALAR_PH]], label [[VECTOR_PH:%.*]]
 ; CHECK:       vector.ph:
 ; CHECK-NEXT:    [[N_MOD_VF:%.*]] = urem i32 [[TMP2]], 4
 ; CHECK-NEXT:    [[N_VEC:%.*]] = sub i32 [[TMP2]], [[N_MOD_VF]]
 ; CHECK-NEXT:    [[IND_END:%.*]] = trunc i32 [[N_VEC]] to i8
 ; CHECK-NEXT:    [[BROADCAST_SPLATINSERT:%.*]] = insertelement <4 x i8> poison, i8 [[T9:%.*]], i32 0
 ; CHECK-NEXT:    [[BROADCAST_SPLAT:%.*]] = shufflevector <4 x i8> [[BROADCAST_SPLATINSERT]], <4 x i8> poison, <4 x i32> zeroinitializer
 ; CHECK-NEXT:    br label [[VECTOR_BODY:%.*]]
 ; CHECK:       vector.body:
 ; CHECK-NEXT:    [[INDEX:%.*]] = phi i32 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
 ; CHECK-NEXT:    [[VEC_IND:%.*]] = phi <4 x i8> [ <i8 0, i8 1, i8 2, i8 3>, [[VECTOR_PH]] ], [ [[VEC_IND_NEXT:%.*]], [[VECTOR_BODY]] ]
-; CHECK-NEXT:    [[TMP14:%.*]] = add <4 x i8> [[VEC_IND]], <i8 1, i8 1, i8 1, i8 1>
-; CHECK-NEXT:    [[TMP15:%.*]] = extractelement <4 x i8> [[TMP14]], i32 0
-; CHECK-NEXT:    [[TMP16:%.*]] = getelementptr inbounds i8, i8* [[ARR]], i8 [[TMP15]]
-; CHECK-NEXT:    [[TMP17:%.*]] = icmp slt <4 x i8> [[TMP14]], [[BROADCAST_SPLAT]]
-; CHECK-NEXT:    [[TMP18:%.*]] = zext <4 x i1> [[TMP17]] to <4 x i8>
-; CHECK-NEXT:    [[TMP19:%.*]] = getelementptr inbounds i8, i8* [[TMP16]], i32 0
-; CHECK-NEXT:    [[TMP20:%.*]] = bitcast i8* [[TMP19]] to <4 x i8>*
-; CHECK-NEXT:    store <4 x i8> [[TMP18]], <4 x i8>* [[TMP20]], align 1
+; CHECK-NEXT:    [[TMP15:%.*]] = add <4 x i8> [[VEC_IND]], <i8 1, i8 1, i8 1, i8 1>
+; CHECK-NEXT:    [[TMP16:%.*]] = extractelement <4 x i8> [[TMP15]], i32 0
+; CHECK-NEXT:    [[TMP17:%.*]] = getelementptr inbounds i8, i8* [[ARR]], i8 [[TMP16]]
+; CHECK-NEXT:    [[TMP18:%.*]] = icmp slt <4 x i8> [[TMP15]], [[BROADCAST_SPLAT]]
+; CHECK-NEXT:    [[TMP19:%.*]] = zext <4 x i1> [[TMP18]] to <4 x i8>
+; CHECK-NEXT:    [[TMP20:%.*]] = getelementptr inbounds i8, i8* [[TMP17]], i32 0
+; CHECK-NEXT:    [[TMP21:%.*]] = bitcast i8* [[TMP20]] to <4 x i8>*
+; CHECK-NEXT:    store <4 x i8> [[TMP19]], <4 x i8>* [[TMP21]], align 1
 ; CHECK-NEXT:    [[INDEX_NEXT]] = add nuw i32 [[INDEX]], 4
 ; CHECK-NEXT:    [[VEC_IND_NEXT]] = add <4 x i8> [[VEC_IND]], <i8 4, i8 4, i8 4, i8 4>
-; CHECK-NEXT:    [[TMP21:%.*]] = icmp eq i32 [[INDEX_NEXT]], [[N_VEC]]
-; CHECK-NEXT:    br i1 [[TMP21]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP0:![0-9]+]]
+; CHECK-NEXT:    [[TMP22:%.*]] = icmp eq i32 [[INDEX_NEXT]], [[N_VEC]]
+; CHECK-NEXT:    br i1 [[TMP22]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP0:![0-9]+]]
 ; CHECK:       middle.block:
 ; CHECK-NEXT:    [[CMP_N:%.*]] = icmp eq i32 [[TMP2]], [[N_VEC]]
 ; CHECK-NEXT:    br i1 [[CMP_N]], label [[FOR_EXIT:%.*]], label [[SCALAR_PH]]
 ; CHECK:       scalar.ph:
 ; CHECK-NEXT:    [[BC_RESUME_VAL:%.*]] = phi i8 [ [[IND_END]], [[MIDDLE_BLOCK]] ], [ 0, [[FOR_PREHEADER]] ], [ 0, [[VECTOR_SCEVCHECK]] ]
 ; CHECK-NEXT:    br label [[FOR_BODY:%.*]]
 ; CHECK:       for.body:
 ; CHECK-NEXT:    [[IV:%.*]] = phi i8 [ [[IV_NEXT:%.*]], [[FOR_BODY]] ], [ [[BC_RESUME_VAL]], [[SCALAR_PH]] ]

llvm/test/Transforms/PhaseOrdering/scev-custom-dl.ll

bb2:    ; preds = %bb1
   ret i32 %5
 }
 
 declare void @use(i1)
 
 declare void @llvm.experimental.guard(i1, ...)
 
 ; This tests getRangeRef acts as intended with different idx size.
-; CHECK: max backedge-taken count is 41
+; CHECK: Loop %loop: Unpredictable max backedge-taken count.
 define void @test_range_ref1(i8 %t) {
  entry:
   %t.ptr = inttoptr i8 %t to i8*
   %p.42 = inttoptr i8 42 to i8*
   %cmp1 = icmp slt i8* %t.ptr, %p.42
   call void(i1, ...) @llvm.experimental.guard(i1 %cmp1) [ "deopt"() ]
   br label %loop