[llvm] [LoopPeel] Support min/max intrinsics in loop peeling (PR #93162)

Sergey Kachkov via llvm-commits llvm-commits at lists.llvm.org
Fri May 31 03:08:06 PDT 2024


https://github.com/skachkov-sc updated https://github.com/llvm/llvm-project/pull/93162

>From 46263ea59709ed7eda4e8d05991de0ff8130d1a2 Mon Sep 17 00:00:00 2001
From: Sergey Kachkov <sergey.kachkov at syntacore.com>
Date: Thu, 16 May 2024 18:34:52 +0300
Subject: [PATCH] [LoopPeel] Support min/max intrinsics in loop peeling

---
 llvm/lib/Transforms/Utils/LoopPeel.cpp        |  86 +++++++---
 .../peel-loop-min-max-intrinsics.ll           | 159 ++++++++++++++----
 2 files changed, 192 insertions(+), 53 deletions(-)

diff --git a/llvm/lib/Transforms/Utils/LoopPeel.cpp b/llvm/lib/Transforms/Utils/LoopPeel.cpp
index f76fa3bb6c611..e2516930d251b 100644
--- a/llvm/lib/Transforms/Utils/LoopPeel.cpp
+++ b/llvm/lib/Transforms/Utils/LoopPeel.cpp
@@ -351,6 +351,21 @@ static unsigned countToEliminateCompares(Loop &L, unsigned MaxPeelCount,
     MaxPeelCount =
         std::min((unsigned)SC->getAPInt().getLimitedValue() - 1, MaxPeelCount);
 
+  // Increase PeelCount while (IterVal Pred BoundSCEV) condition is satisfied;
+  // return true if inversed condition become known before reaching the
+  // MaxPeelCount limit.
+  auto PeelWhilePredicateIsKnown =
+      [&](unsigned &PeelCount, const SCEV *&IterVal, const SCEV *BoundSCEV,
+          const SCEV *Step, ICmpInst::Predicate Pred) {
+        while (PeelCount < MaxPeelCount &&
+               SE.isKnownPredicate(Pred, IterVal, BoundSCEV)) {
+          IterVal = SE.getAddExpr(IterVal, Step);
+          ++PeelCount;
+        }
+        return SE.isKnownPredicate(ICmpInst::getInversePredicate(Pred), IterVal,
+                                   BoundSCEV);
+      };
+
   const unsigned MaxDepth = 4;
   std::function<void(Value *, unsigned)> ComputePeelCount =
       [&](Value *Condition, unsigned Depth) -> void {
@@ -411,48 +426,73 @@ static unsigned countToEliminateCompares(Loop &L, unsigned MaxPeelCount,
       Pred = ICmpInst::getInversePredicate(Pred);
 
     const SCEV *Step = LeftAR->getStepRecurrence(SE);
-    const SCEV *NextIterVal = SE.getAddExpr(IterVal, Step);
-    auto PeelOneMoreIteration = [&IterVal, &NextIterVal, &SE, Step,
-                                 &NewPeelCount]() {
-      IterVal = NextIterVal;
-      NextIterVal = SE.getAddExpr(IterVal, Step);
-      NewPeelCount++;
-    };
-
-    auto CanPeelOneMoreIteration = [&NewPeelCount, &MaxPeelCount]() {
-      return NewPeelCount < MaxPeelCount;
-    };
-
-    while (CanPeelOneMoreIteration() &&
-           SE.isKnownPredicate(Pred, IterVal, RightSCEV))
-      PeelOneMoreIteration();
-
-    // With *that* peel count, does the predicate !Pred become known in the
-    // first iteration of the loop body after peeling?
-    if (!SE.isKnownPredicate(ICmpInst::getInversePredicate(Pred), IterVal,
-                             RightSCEV))
-      return; // If not, give up.
+    if (!PeelWhilePredicateIsKnown(NewPeelCount, IterVal, RightSCEV, Step,
+                                   Pred))
+      return;
 
     // However, for equality comparisons, that isn't always sufficient to
     // eliminate the comparsion in loop body, we may need to peel one more
     // iteration. See if that makes !Pred become unknown again.
+    const SCEV *NextIterVal = SE.getAddExpr(IterVal, Step);
     if (ICmpInst::isEquality(Pred) &&
         !SE.isKnownPredicate(ICmpInst::getInversePredicate(Pred), NextIterVal,
                              RightSCEV) &&
         !SE.isKnownPredicate(Pred, IterVal, RightSCEV) &&
         SE.isKnownPredicate(Pred, NextIterVal, RightSCEV)) {
-      if (!CanPeelOneMoreIteration())
+      if (NewPeelCount >= MaxPeelCount)
         return; // Need to peel one more iteration, but can't. Give up.
-      PeelOneMoreIteration(); // Great!
+      ++NewPeelCount; // Great!
     }
 
     DesiredPeelCount = std::max(DesiredPeelCount, NewPeelCount);
   };
 
+  auto ComputePeelCountMinMax = [&](MinMaxIntrinsic *MinMax) {
+    if (!MinMax->getType()->isIntegerTy())
+      return;
+    Value *LHS = MinMax->getLHS(), *RHS = MinMax->getRHS();
+    const SCEV *BoundSCEV, *IterSCEV;
+    if (L.isLoopInvariant(LHS)) {
+      BoundSCEV = SE.getSCEV(LHS);
+      IterSCEV = SE.getSCEV(RHS);
+    } else if (L.isLoopInvariant(RHS)) {
+      BoundSCEV = SE.getSCEV(RHS);
+      IterSCEV = SE.getSCEV(LHS);
+    } else
+      return;
+    const auto *AddRec = dyn_cast<SCEVAddRecExpr>(IterSCEV);
+    // For simplicity, we support only affine recurrences.
+    if (!AddRec || !AddRec->isAffine() || AddRec->getLoop() != &L)
+      return;
+    const SCEV *Step = AddRec->getStepRecurrence(SE);
+    bool IsSigned = MinMax->isSigned();
+    // To minimize number of peeled iterations, we use strict relational
+    // predicates here.
+    ICmpInst::Predicate Pred;
+    if (SE.isKnownPositive(Step))
+      Pred = IsSigned ? ICmpInst::ICMP_SLT : ICmpInst::ICMP_ULT;
+    else if (SE.isKnownNegative(Step))
+      Pred = IsSigned ? ICmpInst::ICMP_SGT : ICmpInst::ICMP_UGT;
+    else
+      return;
+    // Check that AddRec is not wrapping.
+    if (!(IsSigned ? AddRec->hasNoSignedWrap() : AddRec->hasNoUnsignedWrap()))
+      return;
+    unsigned NewPeelCount = DesiredPeelCount;
+    const SCEV *IterVal = AddRec->evaluateAtIteration(
+        SE.getConstant(AddRec->getType(), NewPeelCount), SE);
+    if (!PeelWhilePredicateIsKnown(NewPeelCount, IterVal, BoundSCEV, Step,
+                                   Pred))
+      return;
+    DesiredPeelCount = NewPeelCount;
+  };
+
   for (BasicBlock *BB : L.blocks()) {
     for (Instruction &I : *BB) {
       if (SelectInst *SI = dyn_cast<SelectInst>(&I))
         ComputePeelCount(SI->getCondition(), 0);
+      if (MinMaxIntrinsic *MinMax = dyn_cast<MinMaxIntrinsic>(&I))
+        ComputePeelCountMinMax(MinMax);
     }
 
     auto *BI = dyn_cast<BranchInst>(BB->getTerminator());
diff --git a/llvm/test/Transforms/LoopUnroll/peel-loop-min-max-intrinsics.ll b/llvm/test/Transforms/LoopUnroll/peel-loop-min-max-intrinsics.ll
index 3a40eec7624e6..c699450afb6e5 100644
--- a/llvm/test/Transforms/LoopUnroll/peel-loop-min-max-intrinsics.ll
+++ b/llvm/test/Transforms/LoopUnroll/peel-loop-min-max-intrinsics.ll
@@ -11,14 +11,37 @@ define void @test_umin(i32 %N) {
 ; CHECK-NEXT:    [[CMP5_NOT:%.*]] = icmp eq i32 [[N]], 0
 ; CHECK-NEXT:    br i1 [[CMP5_NOT]], label [[FOR_COND_CLEANUP:%.*]], label [[FOR_BODY_PREHEADER:%.*]]
 ; CHECK:       for.body.preheader:
+; CHECK-NEXT:    br label [[FOR_BODY_PEEL2:%.*]]
+; CHECK:       for.body.peel.begin:
+; CHECK-NEXT:    br label [[FOR_BODY_PEEL:%.*]]
+; CHECK:       for.body.peel:
+; CHECK-NEXT:    [[COND_PEEL:%.*]] = tail call i32 @llvm.umin.i32(i32 0, i32 2)
+; CHECK-NEXT:    tail call void @foo(i32 [[COND_PEEL]])
+; CHECK-NEXT:    [[INC_PEEL:%.*]] = add nuw i32 0, 1
+; CHECK-NEXT:    [[EXITCOND_NOT_PEEL:%.*]] = icmp eq i32 [[INC_PEEL]], [[N]]
+; CHECK-NEXT:    br i1 [[EXITCOND_NOT_PEEL]], label [[FOR_COND_CLEANUP_LOOPEXIT:%.*]], label [[FOR_BODY_PEEL_NEXT:%.*]]
+; CHECK:       for.body.peel.next:
+; CHECK-NEXT:    br label [[FOR_BODY_PEEL3:%.*]]
+; CHECK:       for.body.peel2:
+; CHECK-NEXT:    [[COND_PEEL3:%.*]] = tail call i32 @llvm.umin.i32(i32 [[INC_PEEL]], i32 2)
+; CHECK-NEXT:    tail call void @foo(i32 [[COND_PEEL3]])
+; CHECK-NEXT:    [[INC_PEEL4:%.*]] = add nuw i32 [[INC_PEEL]], 1
+; CHECK-NEXT:    [[EXITCOND_NOT_PEEL5:%.*]] = icmp eq i32 [[INC_PEEL4]], [[N]]
+; CHECK-NEXT:    br i1 [[EXITCOND_NOT_PEEL5]], label [[FOR_COND_CLEANUP_LOOPEXIT]], label [[FOR_BODY_PEEL_NEXT1:%.*]]
+; CHECK:       for.body.peel.next1:
+; CHECK-NEXT:    br label [[FOR_BODY_PEEL_NEXT6:%.*]]
+; CHECK:       for.body.peel.next6:
+; CHECK-NEXT:    br label [[FOR_BODY_PREHEADER_PEEL_NEWPH:%.*]]
+; CHECK:       for.body.preheader.peel.newph:
 ; CHECK-NEXT:    br label [[FOR_BODY:%.*]]
 ; CHECK:       for.body:
-; CHECK-NEXT:    [[I_06:%.*]] = phi i32 [ [[INC:%.*]], [[FOR_BODY]] ], [ 0, [[FOR_BODY_PREHEADER]] ]
-; CHECK-NEXT:    [[COND:%.*]] = tail call i32 @llvm.umin.i32(i32 [[I_06]], i32 2)
-; CHECK-NEXT:    tail call void @foo(i32 [[COND]])
+; CHECK-NEXT:    [[I_06:%.*]] = phi i32 [ [[INC:%.*]], [[FOR_BODY]] ], [ [[INC_PEEL4]], [[FOR_BODY_PREHEADER_PEEL_NEWPH]] ]
+; CHECK-NEXT:    tail call void @foo(i32 2)
 ; CHECK-NEXT:    [[INC]] = add nuw i32 [[I_06]], 1
 ; CHECK-NEXT:    [[EXITCOND_NOT:%.*]] = icmp eq i32 [[INC]], [[N]]
-; CHECK-NEXT:    br i1 [[EXITCOND_NOT]], label [[FOR_COND_CLEANUP_LOOPEXIT:%.*]], label [[FOR_BODY]]
+; CHECK-NEXT:    br i1 [[EXITCOND_NOT]], label [[FOR_COND_CLEANUP_LOOPEXIT_LOOPEXIT:%.*]], label [[FOR_BODY]], !llvm.loop [[LOOP0:![0-9]+]]
+; CHECK:       for.cond.cleanup.loopexit.loopexit:
+; CHECK-NEXT:    br label [[FOR_COND_CLEANUP_LOOPEXIT]]
 ; CHECK:       for.cond.cleanup.loopexit:
 ; CHECK-NEXT:    br label [[FOR_COND_CLEANUP]]
 ; CHECK:       for.cond.cleanup:
@@ -47,14 +70,37 @@ define void @test_umax(i32 %N) {
 ; CHECK-NEXT:    [[CMP5_NOT:%.*]] = icmp eq i32 [[N]], 0
 ; CHECK-NEXT:    br i1 [[CMP5_NOT]], label [[FOR_COND_CLEANUP:%.*]], label [[FOR_BODY_PREHEADER:%.*]]
 ; CHECK:       for.body.preheader:
+; CHECK-NEXT:    br label [[FOR_BODY_PEEL2:%.*]]
+; CHECK:       for.body.peel.begin:
+; CHECK-NEXT:    br label [[FOR_BODY_PEEL:%.*]]
+; CHECK:       for.body.peel:
+; CHECK-NEXT:    [[COND_PEEL:%.*]] = tail call i32 @llvm.umax.i32(i32 0, i32 2)
+; CHECK-NEXT:    tail call void @foo(i32 [[COND_PEEL]])
+; CHECK-NEXT:    [[INC_PEEL:%.*]] = add nuw i32 0, 1
+; CHECK-NEXT:    [[EXITCOND_NOT_PEEL:%.*]] = icmp eq i32 [[INC_PEEL]], [[N]]
+; CHECK-NEXT:    br i1 [[EXITCOND_NOT_PEEL]], label [[FOR_COND_CLEANUP_LOOPEXIT:%.*]], label [[FOR_BODY_PEEL_NEXT:%.*]]
+; CHECK:       for.body.peel.next:
+; CHECK-NEXT:    br label [[FOR_BODY_PEEL3:%.*]]
+; CHECK:       for.body.peel2:
+; CHECK-NEXT:    [[COND_PEEL3:%.*]] = tail call i32 @llvm.umax.i32(i32 [[INC_PEEL]], i32 2)
+; CHECK-NEXT:    tail call void @foo(i32 [[COND_PEEL3]])
+; CHECK-NEXT:    [[INC_PEEL4:%.*]] = add nuw i32 [[INC_PEEL]], 1
+; CHECK-NEXT:    [[EXITCOND_NOT_PEEL5:%.*]] = icmp eq i32 [[INC_PEEL4]], [[N]]
+; CHECK-NEXT:    br i1 [[EXITCOND_NOT_PEEL5]], label [[FOR_COND_CLEANUP_LOOPEXIT]], label [[FOR_BODY_PEEL_NEXT1:%.*]]
+; CHECK:       for.body.peel.next1:
+; CHECK-NEXT:    br label [[FOR_BODY_PEEL_NEXT6:%.*]]
+; CHECK:       for.body.peel.next6:
+; CHECK-NEXT:    br label [[FOR_BODY_PREHEADER_PEEL_NEWPH:%.*]]
+; CHECK:       for.body.preheader.peel.newph:
 ; CHECK-NEXT:    br label [[FOR_BODY:%.*]]
 ; CHECK:       for.body:
-; CHECK-NEXT:    [[I_06:%.*]] = phi i32 [ [[INC:%.*]], [[FOR_BODY]] ], [ 0, [[FOR_BODY_PREHEADER]] ]
-; CHECK-NEXT:    [[COND:%.*]] = tail call i32 @llvm.umax.i32(i32 [[I_06]], i32 2)
-; CHECK-NEXT:    tail call void @foo(i32 [[COND]])
+; CHECK-NEXT:    [[I_06:%.*]] = phi i32 [ [[INC:%.*]], [[FOR_BODY]] ], [ [[INC_PEEL4]], [[FOR_BODY_PREHEADER_PEEL_NEWPH]] ]
+; CHECK-NEXT:    tail call void @foo(i32 [[I_06]])
 ; CHECK-NEXT:    [[INC]] = add nuw i32 [[I_06]], 1
 ; CHECK-NEXT:    [[EXITCOND_NOT:%.*]] = icmp eq i32 [[INC]], [[N]]
-; CHECK-NEXT:    br i1 [[EXITCOND_NOT]], label [[FOR_COND_CLEANUP_LOOPEXIT:%.*]], label [[FOR_BODY]]
+; CHECK-NEXT:    br i1 [[EXITCOND_NOT]], label [[FOR_COND_CLEANUP_LOOPEXIT_LOOPEXIT:%.*]], label [[FOR_BODY]], !llvm.loop [[LOOP2:![0-9]+]]
+; CHECK:       for.cond.cleanup.loopexit.loopexit:
+; CHECK-NEXT:    br label [[FOR_COND_CLEANUP_LOOPEXIT]]
 ; CHECK:       for.cond.cleanup.loopexit:
 ; CHECK-NEXT:    br label [[FOR_COND_CLEANUP]]
 ; CHECK:       for.cond.cleanup:
@@ -83,14 +129,37 @@ define void @test_smax(i32 %N) {
 ; CHECK-NEXT:    [[CMP5:%.*]] = icmp slt i32 [[N]], 0
 ; CHECK-NEXT:    br i1 [[CMP5]], label [[FOR_BODY_PREHEADER:%.*]], label [[FOR_COND_CLEANUP:%.*]]
 ; CHECK:       for.body.preheader:
+; CHECK-NEXT:    br label [[FOR_BODY_PEEL2:%.*]]
+; CHECK:       for.body.peel.begin:
+; CHECK-NEXT:    br label [[FOR_BODY_PEEL:%.*]]
+; CHECK:       for.body.peel:
+; CHECK-NEXT:    [[COND_PEEL:%.*]] = tail call i32 @llvm.smax.i32(i32 0, i32 -2)
+; CHECK-NEXT:    tail call void @foo(i32 [[COND_PEEL]])
+; CHECK-NEXT:    [[DEC_PEEL:%.*]] = add nsw i32 0, -1
+; CHECK-NEXT:    [[CMP_PEEL:%.*]] = icmp sgt i32 [[DEC_PEEL]], [[N]]
+; CHECK-NEXT:    br i1 [[CMP_PEEL]], label [[FOR_BODY_PEEL_NEXT:%.*]], label [[FOR_COND_CLEANUP_LOOPEXIT:%.*]]
+; CHECK:       for.body.peel.next:
+; CHECK-NEXT:    br label [[FOR_BODY_PEEL3:%.*]]
+; CHECK:       for.body.peel2:
+; CHECK-NEXT:    [[COND_PEEL3:%.*]] = tail call i32 @llvm.smax.i32(i32 [[DEC_PEEL]], i32 -2)
+; CHECK-NEXT:    tail call void @foo(i32 [[COND_PEEL3]])
+; CHECK-NEXT:    [[DEC_PEEL4:%.*]] = add nsw i32 [[DEC_PEEL]], -1
+; CHECK-NEXT:    [[CMP_PEEL5:%.*]] = icmp sgt i32 [[DEC_PEEL4]], [[N]]
+; CHECK-NEXT:    br i1 [[CMP_PEEL5]], label [[FOR_BODY_PEEL_NEXT1:%.*]], label [[FOR_COND_CLEANUP_LOOPEXIT]]
+; CHECK:       for.body.peel.next1:
+; CHECK-NEXT:    br label [[FOR_BODY_PEEL_NEXT6:%.*]]
+; CHECK:       for.body.peel.next6:
+; CHECK-NEXT:    br label [[FOR_BODY_PREHEADER_PEEL_NEWPH:%.*]]
+; CHECK:       for.body.preheader.peel.newph:
 ; CHECK-NEXT:    br label [[FOR_BODY:%.*]]
 ; CHECK:       for.body:
-; CHECK-NEXT:    [[I_06:%.*]] = phi i32 [ [[DEC:%.*]], [[FOR_BODY]] ], [ 0, [[FOR_BODY_PREHEADER]] ]
-; CHECK-NEXT:    [[COND:%.*]] = tail call i32 @llvm.smax.i32(i32 [[I_06]], i32 -2)
-; CHECK-NEXT:    tail call void @foo(i32 [[COND]])
+; CHECK-NEXT:    [[I_06:%.*]] = phi i32 [ [[DEC:%.*]], [[FOR_BODY]] ], [ [[DEC_PEEL4]], [[FOR_BODY_PREHEADER_PEEL_NEWPH]] ]
+; CHECK-NEXT:    tail call void @foo(i32 -2)
 ; CHECK-NEXT:    [[DEC]] = add nsw i32 [[I_06]], -1
 ; CHECK-NEXT:    [[CMP:%.*]] = icmp sgt i32 [[DEC]], [[N]]
-; CHECK-NEXT:    br i1 [[CMP]], label [[FOR_BODY]], label [[FOR_COND_CLEANUP_LOOPEXIT:%.*]]
+; CHECK-NEXT:    br i1 [[CMP]], label [[FOR_BODY]], label [[FOR_COND_CLEANUP_LOOPEXIT_LOOPEXIT:%.*]], !llvm.loop [[LOOP3:![0-9]+]]
+; CHECK:       for.cond.cleanup.loopexit.loopexit:
+; CHECK-NEXT:    br label [[FOR_COND_CLEANUP_LOOPEXIT]]
 ; CHECK:       for.cond.cleanup.loopexit:
 ; CHECK-NEXT:    br label [[FOR_COND_CLEANUP]]
 ; CHECK:       for.cond.cleanup:
@@ -119,14 +188,37 @@ define void @test_smin(i32 %N) {
 ; CHECK-NEXT:    [[CMP5:%.*]] = icmp slt i32 [[N]], 0
 ; CHECK-NEXT:    br i1 [[CMP5]], label [[FOR_BODY_PREHEADER:%.*]], label [[FOR_COND_CLEANUP:%.*]]
 ; CHECK:       for.body.preheader:
+; CHECK-NEXT:    br label [[FOR_BODY_PEEL2:%.*]]
+; CHECK:       for.body.peel.begin:
+; CHECK-NEXT:    br label [[FOR_BODY_PEEL:%.*]]
+; CHECK:       for.body.peel:
+; CHECK-NEXT:    [[COND_PEEL:%.*]] = tail call i32 @llvm.smin.i32(i32 0, i32 -2)
+; CHECK-NEXT:    tail call void @foo(i32 noundef signext [[COND_PEEL]])
+; CHECK-NEXT:    [[DEC_PEEL:%.*]] = add nsw i32 0, -1
+; CHECK-NEXT:    [[CMP_PEEL:%.*]] = icmp sgt i32 [[DEC_PEEL]], [[N]]
+; CHECK-NEXT:    br i1 [[CMP_PEEL]], label [[FOR_BODY_PEEL_NEXT:%.*]], label [[FOR_COND_CLEANUP_LOOPEXIT:%.*]]
+; CHECK:       for.body.peel.next:
+; CHECK-NEXT:    br label [[FOR_BODY_PEEL3:%.*]]
+; CHECK:       for.body.peel2:
+; CHECK-NEXT:    [[COND_PEEL3:%.*]] = tail call i32 @llvm.smin.i32(i32 [[DEC_PEEL]], i32 -2)
+; CHECK-NEXT:    tail call void @foo(i32 noundef signext [[COND_PEEL3]])
+; CHECK-NEXT:    [[DEC_PEEL4:%.*]] = add nsw i32 [[DEC_PEEL]], -1
+; CHECK-NEXT:    [[CMP_PEEL5:%.*]] = icmp sgt i32 [[DEC_PEEL4]], [[N]]
+; CHECK-NEXT:    br i1 [[CMP_PEEL5]], label [[FOR_BODY_PEEL_NEXT1:%.*]], label [[FOR_COND_CLEANUP_LOOPEXIT]]
+; CHECK:       for.body.peel.next1:
+; CHECK-NEXT:    br label [[FOR_BODY_PEEL_NEXT6:%.*]]
+; CHECK:       for.body.peel.next6:
+; CHECK-NEXT:    br label [[FOR_BODY_PREHEADER_PEEL_NEWPH:%.*]]
+; CHECK:       for.body.preheader.peel.newph:
 ; CHECK-NEXT:    br label [[FOR_BODY:%.*]]
 ; CHECK:       for.body:
-; CHECK-NEXT:    [[I_06:%.*]] = phi i32 [ [[DEC:%.*]], [[FOR_BODY]] ], [ 0, [[FOR_BODY_PREHEADER]] ]
-; CHECK-NEXT:    [[COND:%.*]] = tail call i32 @llvm.smin.i32(i32 [[I_06]], i32 -2)
-; CHECK-NEXT:    tail call void @foo(i32 noundef signext [[COND]])
+; CHECK-NEXT:    [[I_06:%.*]] = phi i32 [ [[DEC:%.*]], [[FOR_BODY]] ], [ [[DEC_PEEL4]], [[FOR_BODY_PREHEADER_PEEL_NEWPH]] ]
+; CHECK-NEXT:    tail call void @foo(i32 noundef signext [[I_06]])
 ; CHECK-NEXT:    [[DEC]] = add nsw i32 [[I_06]], -1
 ; CHECK-NEXT:    [[CMP:%.*]] = icmp sgt i32 [[DEC]], [[N]]
-; CHECK-NEXT:    br i1 [[CMP]], label [[FOR_BODY]], label [[FOR_COND_CLEANUP_LOOPEXIT:%.*]]
+; CHECK-NEXT:    br i1 [[CMP]], label [[FOR_BODY]], label [[FOR_COND_CLEANUP_LOOPEXIT_LOOPEXIT:%.*]], !llvm.loop [[LOOP4:![0-9]+]]
+; CHECK:       for.cond.cleanup.loopexit.loopexit:
+; CHECK-NEXT:    br label [[FOR_COND_CLEANUP_LOOPEXIT]]
 ; CHECK:       for.cond.cleanup.loopexit:
 ; CHECK-NEXT:    br label [[FOR_COND_CLEANUP]]
 ; CHECK:       for.cond.cleanup:
@@ -191,14 +283,14 @@ define void @test_max_count_threshold(i32 %N) {
 ; CHECK-NEXT:    [[CMP5_NOT:%.*]] = icmp eq i32 [[N]], 0
 ; CHECK-NEXT:    br i1 [[CMP5_NOT]], label [[FOR_COND_CLEANUP:%.*]], label [[FOR_BODY_PREHEADER:%.*]]
 ; CHECK:       for.body.preheader:
-; CHECK-NEXT:    br label [[FOR_BODY:%.*]]
+; CHECK-NEXT:    br label [[FOR_BODY1:%.*]]
 ; CHECK:       for.body:
-; CHECK-NEXT:    [[I_06:%.*]] = phi i32 [ [[INC:%.*]], [[FOR_BODY]] ], [ 0, [[FOR_BODY_PREHEADER]] ]
-; CHECK-NEXT:    [[COND:%.*]] = tail call i32 @llvm.umin.i32(i32 [[I_06]], i32 5)
-; CHECK-NEXT:    tail call void @foo(i32 [[COND]])
-; CHECK-NEXT:    [[INC]] = add nuw i32 [[I_06]], 1
-; CHECK-NEXT:    [[EXITCOND_NOT:%.*]] = icmp eq i32 [[INC]], [[N]]
-; CHECK-NEXT:    br i1 [[EXITCOND_NOT]], label [[FOR_COND_CLEANUP_LOOPEXIT:%.*]], label [[FOR_BODY]]
+; CHECK-NEXT:    [[I_6:%.*]] = phi i32 [ [[INC1:%.*]], [[FOR_BODY1]] ], [ 0, [[FOR_BODY_PREHEADER]] ]
+; CHECK-NEXT:    [[COND1:%.*]] = tail call i32 @llvm.umin.i32(i32 [[I_6]], i32 5)
+; CHECK-NEXT:    tail call void @foo(i32 [[COND1]])
+; CHECK-NEXT:    [[INC1]] = add nuw i32 [[I_6]], 1
+; CHECK-NEXT:    [[EXITCOND_NOT1:%.*]] = icmp eq i32 [[INC1]], [[N]]
+; CHECK-NEXT:    br i1 [[EXITCOND_NOT1]], label [[FOR_COND_CLEANUP_LOOPEXIT:%.*]], label [[FOR_BODY1]]
 ; CHECK:       for.cond.cleanup.loopexit:
 ; CHECK-NEXT:    br label [[FOR_COND_CLEANUP]]
 ; CHECK:       for.cond.cleanup:
@@ -227,14 +319,14 @@ define void @test_wrap(i8 %N) {
 ; CHECK-NEXT:    [[CMP5_NOT:%.*]] = icmp eq i8 [[N]], 0
 ; CHECK-NEXT:    br i1 [[CMP5_NOT]], label [[FOR_COND_CLEANUP:%.*]], label [[FOR_BODY_PREHEADER:%.*]]
 ; CHECK:       for.body.preheader:
-; CHECK-NEXT:    br label [[FOR_BODY1:%.*]]
+; CHECK-NEXT:    br label [[FOR_BODY:%.*]]
 ; CHECK:       for.body:
-; CHECK-NEXT:    [[I_06:%.*]] = phi i8 [ [[INC1:%.*]], [[FOR_BODY1]] ], [ 0, [[FOR_BODY_PREHEADER]] ]
-; CHECK-NEXT:    [[COND1:%.*]] = tail call i8 @llvm.umin.i8(i8 [[I_06]], i8 -2)
-; CHECK-NEXT:    tail call void @bar(i8 [[COND1]])
-; CHECK-NEXT:    [[INC1]] = add i8 [[I_06]], 127
-; CHECK-NEXT:    [[EXITCOND_NOT1:%.*]] = icmp eq i8 [[INC1]], [[N]]
-; CHECK-NEXT:    br i1 [[EXITCOND_NOT1]], label [[FOR_COND_CLEANUP_LOOPEXIT:%.*]], label [[FOR_BODY1]]
+; CHECK-NEXT:    [[I_06:%.*]] = phi i8 [ [[INC:%.*]], [[FOR_BODY]] ], [ 0, [[FOR_BODY_PREHEADER]] ]
+; CHECK-NEXT:    [[COND:%.*]] = tail call i8 @llvm.umin.i8(i8 [[I_06]], i8 -2)
+; CHECK-NEXT:    tail call void @bar(i8 [[COND]])
+; CHECK-NEXT:    [[INC]] = add i8 [[I_06]], 127
+; CHECK-NEXT:    [[EXITCOND_NOT:%.*]] = icmp eq i8 [[INC]], [[N]]
+; CHECK-NEXT:    br i1 [[EXITCOND_NOT]], label [[FOR_COND_CLEANUP_LOOPEXIT:%.*]], label [[FOR_BODY]]
 ; CHECK:       for.cond.cleanup.loopexit:
 ; CHECK-NEXT:    br label [[FOR_COND_CLEANUP]]
 ; CHECK:       for.cond.cleanup:
@@ -255,3 +347,10 @@ for.body:
 for.cond.cleanup:
   ret void
 }
+;.
+; CHECK: [[LOOP0]] = distinct !{[[LOOP0]], [[META1:![0-9]+]]}
+; CHECK: [[META1]] = !{!"llvm.loop.peeled.count", i32 2}
+; CHECK: [[LOOP2]] = distinct !{[[LOOP2]], [[META1]]}
+; CHECK: [[LOOP3]] = distinct !{[[LOOP3]], [[META1]]}
+; CHECK: [[LOOP4]] = distinct !{[[LOOP4]], [[META1]]}
+;.



More information about the llvm-commits mailing list