[SCEV] Re-enable "Use nw flag and symbolic iteration count to sharpen ranges of AddRe...

We can sharpen the range of a AddRec if we know that it does not
self-wrap and know the symbolic iteration count in the loop. If we can
evaluate the value of AddRec on the last iteration and prove that at least
one its intermediate value lies between start and end, then no-wrap flag
allows us to conclude that all of them also lie between start and end. So
the estimate of range can be improved to union of ranges of start and end.

Switched off by default, can be turned on by flag.

Differential Revision: https://reviews.llvm.org/D89381
Reviewed By: lebedev.ri, nikic

diff --git a/llvm/include/llvm/Analysis/ScalarEvolution.h b/llvm/include/llvm/Analysis/ScalarEvolution.h
index 6dc3f2e..faf326d 100644 (file)
--- a/llvm/include/llvm/Analysis/ScalarEvolution.h
+++ b/llvm/include/llvm/Analysis/ScalarEvolution.h
@@ -1507,6 +1507,13 @@ private:
   ConstantRange getRangeForAffineAR(const SCEV *Start, const SCEV *Stop,
                                     const SCEV *MaxBECount, unsigned BitWidth);
 
+  /// Determines the range for the affine non-self-wrapping SCEVAddRecExpr {\p
+  /// Start,+,\p Stop}<nw>.
+  ConstantRange getRangeForAffineNoSelfWrappingAR(const SCEVAddRecExpr *AddRec,
+                                                  const SCEV *MaxBECount,
+                                                  unsigned BitWidth,
+                                                  RangeSignHint SignHint);
+
   /// Try to compute a range for the affine SCEVAddRecExpr {\p Start,+,\p
   /// Stop} by "factoring out" a ternary expression from the add recurrence.
   /// Helper called by \c getRange.

diff --git a/llvm/lib/Analysis/ScalarEvolution.cpp b/llvm/lib/Analysis/ScalarEvolution.cpp
index 17f9c12..17128fb 100644 (file)
--- a/llvm/lib/Analysis/ScalarEvolution.cpp
+++ b/llvm/lib/Analysis/ScalarEvolution.cpp
@@ -226,6 +226,11 @@ ClassifyExpressions("scalar-evolution-classify-expressions",
     cl::Hidden, cl::init(true),
     cl::desc("When printing analysis, include information on every instruction"));
 
+static cl::opt<bool> UseExpensiveRangeSharpening(
+    "scalar-evolution-use-expensive-range-sharpening", cl::Hidden,
+    cl::init(false),
+    cl::desc("Use more powerful methods of sharpening expression ranges. May "
+             "be costly in terms of compile time"));
 
 //===----------------------------------------------------------------------===//
 //                           SCEV class definitions
@@ -5527,6 +5532,20 @@ ScalarEvolution::getRangeRef(const SCEV *S,
         ConservativeResult =
             ConservativeResult.intersectWith(RangeFromFactoring, RangeType);
       }
+
+      // Now try symbolic BE count and more powerful methods.
+      if (UseExpensiveRangeSharpening) {
+        const SCEV *SymbolicMaxBECount =
+            getSymbolicMaxBackedgeTakenCount(AddRec->getLoop());
+        if (!isa<SCEVCouldNotCompute>(SymbolicMaxBECount) &&
+            getTypeSizeInBits(MaxBECount->getType()) <= BitWidth &&
+            AddRec->hasNoSelfWrap()) {
+          auto RangeFromAffineNew = getRangeForAffineNoSelfWrappingAR(
+              AddRec, SymbolicMaxBECount, BitWidth, SignHint);
+          ConservativeResult =
+              ConservativeResult.intersectWith(RangeFromAffineNew, RangeType);
+        }
+      }
     }
 
     return setRange(AddRec, SignHint, std::move(ConservativeResult));
@@ -5696,6 +5715,71 @@ ConstantRange ScalarEvolution::getRangeForAffineAR(const SCEV *Start,
   return SR.intersectWith(UR, ConstantRange::Smallest);
 }
 
+ConstantRange ScalarEvolution::getRangeForAffineNoSelfWrappingAR(
+    const SCEVAddRecExpr *AddRec, const SCEV *MaxBECount, unsigned BitWidth,
+    ScalarEvolution::RangeSignHint SignHint) {
+  assert(AddRec->isAffine() && "Non-affine AddRecs are not suppored!\n");
+  assert(AddRec->hasNoSelfWrap() &&
+         "This only works for non-self-wrapping AddRecs!");
+  const bool IsSigned = SignHint == HINT_RANGE_SIGNED;
+  const SCEV *Step = AddRec->getStepRecurrence(*this);
+  // Only deal with constant step to save compile time.
+  if (!isa<SCEVConstant>(Step))
+    return ConstantRange::getFull(BitWidth);
+  // Let's make sure that we can prove that we do not self-wrap during
+  // MaxBECount iterations. We need this because MaxBECount is a maximum
+  // iteration count estimate, and we might infer nw from some exit for which we
+  // do not know max exit count (or any other side reasoning).
+  // TODO: Turn into assert at some point.
+  MaxBECount = getNoopOrZeroExtend(MaxBECount, AddRec->getType());
+  const SCEV *RangeWidth = getMinusOne(AddRec->getType());
+  const SCEV *StepAbs = getUMinExpr(Step, getNegativeSCEV(Step));
+  const SCEV *MaxItersWithoutWrap = getUDivExpr(RangeWidth, StepAbs);
+  if (!isKnownPredicateViaConstantRanges(ICmpInst::ICMP_ULE, MaxBECount,
+                                         MaxItersWithoutWrap))
+    return ConstantRange::getFull(BitWidth);
+
+  ICmpInst::Predicate LEPred =
+      IsSigned ? ICmpInst::ICMP_SLE : ICmpInst::ICMP_ULE;
+  ICmpInst::Predicate GEPred =
+      IsSigned ? ICmpInst::ICMP_SGE : ICmpInst::ICMP_UGE;
+  const SCEV *End = AddRec->evaluateAtIteration(MaxBECount, *this);
+
+  // We know that there is no self-wrap. Let's take Start and End values and
+  // look at all intermediate values V1, V2, ..., Vn that IndVar takes during
+  // the iteration. They either lie inside the range [Min(Start, End),
+  // Max(Start, End)] or outside it:
+  //
+  // Case 1:   RangeMin    ...    Start V1 ... VN End ...           RangeMax;
+  // Case 2:   RangeMin Vk ... V1 Start    ...    End Vn ... Vk + 1 RangeMax;
+  //
+  // No self wrap flag guarantees that the intermediate values cannot be BOTH
+  // outside and inside the range [Min(Start, End), Max(Start, End)]. Using that
+  // knowledge, let's try to prove that we are dealing with Case 1. It is so if
+  // Start <= End and step is positive, or Start >= End and step is negative.
+  const SCEV *Start = AddRec->getStart();
+  ConstantRange StartRange = getRangeRef(Start, SignHint);
+  ConstantRange EndRange = getRangeRef(End, SignHint);
+  ConstantRange RangeBetween = StartRange.unionWith(EndRange);
+  // If they already cover full iteration space, we will know nothing useful
+  // even if we prove what we want to prove.
+  if (RangeBetween.isFullSet())
+    return RangeBetween;
+  // Only deal with ranges that do not wrap (i.e. RangeMin < RangeMax).
+  bool IsWrappedSet = IsSigned ? RangeBetween.isSignWrappedSet()
+                               : RangeBetween.isWrappedSet();
+  if (IsWrappedSet)
+    return ConstantRange::getFull(BitWidth);
+
+  if (isKnownPositive(Step) &&
+      isKnownPredicateViaConstantRanges(LEPred, Start, End))
+    return RangeBetween;
+  else if (isKnownNegative(Step) &&
+           isKnownPredicateViaConstantRanges(GEPred, Start, End))
+    return RangeBetween;
+  return ConstantRange::getFull(BitWidth);
+}
+
 ConstantRange ScalarEvolution::getRangeViaFactoring(const SCEV *Start,
                                                     const SCEV *Step,
                                                     const SCEV *MaxBECount,

diff --git a/llvm/test/Analysis/ScalarEvolution/no-wrap-symbolic-becount.ll b/llvm/test/Analysis/ScalarEvolution/no-wrap-symbolic-becount.ll
index af3092e..403a0c9 100644 (file)
--- a/llvm/test/Analysis/ScalarEvolution/no-wrap-symbolic-becount.ll
+++ b/llvm/test/Analysis/ScalarEvolution/no-wrap-symbolic-becount.ll
@@ -1,6 +1,6 @@
 ; NOTE: Assertions have been autogenerated by utils/update_analyze_test_checks.py
-; RUN: opt < %s -S -analyze -scalar-evolution -enable-new-pm=0 | FileCheck %s
-; RUN: opt < %s -S -passes='print<scalar-evolution>' 2>&1 | FileCheck %s
+; RUN: opt < %s -S -scalar-evolution-use-expensive-range-sharpening -analyze -scalar-evolution -enable-new-pm=0 | FileCheck %s
+; RUN: opt < %s -S -scalar-evolution-use-expensive-range-sharpening -passes='print<scalar-evolution>' 2>&1 | FileCheck %s
 
 define i32 @test_01(i32 %start, i32* %p, i32* %q) {
 ; CHECK-LABEL: 'test_01'
@@ -8,7 +8,7 @@ define i32 @test_01(i32 %start, i32* %p, i32* %q) {
 ; CHECK-NEXT:    %0 = zext i32 %start to i64
 ; CHECK-NEXT:    --> (zext i32 %start to i64) U: [0,4294967296) S: [0,4294967296)
 ; CHECK-NEXT:    %indvars.iv = phi i64 [ %indvars.iv.next, %backedge ], [ %0, %entry ]
-; CHECK-NEXT:    --> {(zext i32 %start to i64),+,-1}<nsw><%loop> U: [-4294967295,4294967296) S: [-4294967295,4294967296) Exits: <<Unknown>> LoopDispositions: { %loop: Computable }
+; CHECK-NEXT:    --> {(zext i32 %start to i64),+,-1}<nsw><%loop> U: [0,4294967296) S: [0,4294967296) Exits: <<Unknown>> LoopDispositions: { %loop: Computable }
 ; CHECK-NEXT:    %iv = phi i32 [ %start, %entry ], [ %iv.next, %backedge ]
 ; CHECK-NEXT:    --> {%start,+,-1}<%loop> U: full-set S: full-set Exits: <<Unknown>> LoopDispositions: { %loop: Computable }
 ; CHECK-NEXT:    %iv.next = add i32 %iv, -1
@@ -22,7 +22,7 @@ define i32 @test_01(i32 %start, i32* %p, i32* %q) {
 ; CHECK-NEXT:    %stop = load i32, i32* %load.addr, align 4
 ; CHECK-NEXT:    --> %stop U: full-set S: full-set Exits: <<Unknown>> LoopDispositions: { %loop: Variant }
 ; CHECK-NEXT:    %indvars.iv.next = add nsw i64 %indvars.iv, -1
-; CHECK-NEXT:    --> {(-1 + (zext i32 %start to i64))<nsw>,+,-1}<nsw><%loop> U: [-4294967296,4294967295) S: [-4294967296,4294967295) Exits: <<Unknown>> LoopDispositions: { %loop: Computable }
+; CHECK-NEXT:    --> {(-1 + (zext i32 %start to i64))<nsw>,+,-1}<nsw><%loop> U: [-4294967296,4294967295) S: [-1,4294967295) Exits: <<Unknown>> LoopDispositions: { %loop: Computable }
 ; CHECK-NEXT:  Determining loop execution counts for: @test_01
 ; CHECK-NEXT:  Loop %loop: <multiple exits> Unpredictable backedge-taken count.
 ; CHECK-NEXT:    exit count for loop: (zext i32 %start to i64)