From: Sjoerd Meijer <sjoerd.meijer@arm.com>
Date: Tue, 25 Aug 2020 12:39:55 +0000 (+0100)
Subject: [LV] get.active.lane.mask consuming tripcount instead of backedge-taken count
X-Git-Tag: llvmorg-13-init~13819
X-Git-Url: http://review.tizen.org/git/?a=commitdiff_plain;h=ae366479e8c67c18756a823e6ce1c7fcd86a208f;p=platform%2Fupstream%2Fllvm.git

[LV] get.active.lane.mask consuming tripcount instead of backedge-taken count

This adapts LV to the new semantics of get.active.lane.mask as discussed in
D86147, which means that the LV now emits intrinsic get.active.lane.mask with
the loop tripcount instead of the backedge-taken count as its second argument.
The motivation for this is described in D86147.

Differential Revision: https://reviews.llvm.org/D86304
---

diff --git a/llvm/lib/Transforms/Vectorize/LoopVectorize.cpp b/llvm/lib/Transforms/Vectorize/LoopVectorize.cpp
index ecc41db..884b157 100644
--- a/llvm/lib/Transforms/Vectorize/LoopVectorize.cpp
+++ b/llvm/lib/Transforms/Vectorize/LoopVectorize.cpp
@@ -7155,10 +7155,16 @@ VPValue *VPRecipeBuilder::createBlockInMask(BasicBlock *BB, VPlanPtr &Plan) {
     }
     VPValue *BTC = Plan->getOrCreateBackedgeTakenCount();
     bool TailFolded = !CM.isScalarEpilogueAllowed();
-    if (TailFolded && CM.TTI.emitGetActiveLaneMask())
-      BlockMask = Builder.createNaryOp(VPInstruction::ActiveLaneMask, {IV, BTC});
-    else
+
+    if (TailFolded && CM.TTI.emitGetActiveLaneMask()) {
+      // While ActiveLaneMask is a binary op that consumes the loop tripcount
+      // as a second argument, we only pass the IV here and extract the
+      // tripcount from the transform state where codegen of the VP instructions
+      // happen.
+      BlockMask = Builder.createNaryOp(VPInstruction::ActiveLaneMask, {IV});
+    } else {
       BlockMask = Builder.createNaryOp(VPInstruction::ICmpULE, {IV, BTC});
+    }
     return BlockMaskCache[BB] = BlockMask;
   }
 
diff --git a/llvm/lib/Transforms/Vectorize/VPlan.cpp b/llvm/lib/Transforms/Vectorize/VPlan.cpp
index 1358f9d3..a616de6 100644
--- a/llvm/lib/Transforms/Vectorize/VPlan.cpp
+++ b/llvm/lib/Transforms/Vectorize/VPlan.cpp
@@ -384,14 +384,14 @@ void VPInstruction::generateInstruction(VPTransformState &State,
   case VPInstruction::ActiveLaneMask: {
     // Get first lane of vector induction variable.
     Value *VIVElem0 = State.get(getOperand(0), {Part, 0});
-    // Get first lane of backedge-taken-count.
-    Value *ScalarBTC = State.get(getOperand(1), {Part, 0});
+    // Get the original loop tripcount.
+    Value *ScalarTC = State.TripCount;
 
     auto *Int1Ty = Type::getInt1Ty(Builder.getContext());
     auto *PredTy = FixedVectorType::get(Int1Ty, State.VF.Min);
     Instruction *Call = Builder.CreateIntrinsic(
-        Intrinsic::get_active_lane_mask, {PredTy, ScalarBTC->getType()},
-        {VIVElem0, ScalarBTC}, nullptr, "active.lane.mask");
+        Intrinsic::get_active_lane_mask, {PredTy, ScalarTC->getType()},
+        {VIVElem0, ScalarTC}, nullptr, "active.lane.mask");
     State.set(this, Call, Part);
     break;
   }
diff --git a/llvm/test/Transforms/LoopVectorize/ARM/mve-gather-scatter-tailpred.ll b/llvm/test/Transforms/LoopVectorize/ARM/mve-gather-scatter-tailpred.ll
index b52bccd..460b8b2 100644
--- a/llvm/test/Transforms/LoopVectorize/ARM/mve-gather-scatter-tailpred.ll
+++ b/llvm/test/Transforms/LoopVectorize/ARM/mve-gather-scatter-tailpred.ll
@@ -3,28 +3,53 @@
 
 define void @test_stride1_4i32(i32* readonly %data, i32* noalias nocapture %dst, i32 %n) {
 ; CHECK-LABEL: @test_stride1_4i32(
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    br i1 false, label [[SCALAR_PH:%.*]], label [[VECTOR_PH:%.*]]
+; CHECK:       vector.ph:
+; CHECK-NEXT:    [[N_RND_UP:%.*]] = add i32 [[N:%.*]], 3
+; CHECK-NEXT:    [[N_MOD_VF:%.*]] = urem i32 [[N_RND_UP]], 4
+; CHECK-NEXT:    [[N_VEC:%.*]] = sub i32 [[N_RND_UP]], [[N_MOD_VF]]
+; CHECK-NEXT:    br label [[VECTOR_BODY:%.*]]
 ; CHECK:       vector.body:
-; CHECK-NEXT:    [[INDEX:%.*]] = phi i32 [ 0, [[VECTOR_PH:%.*]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY:%.*]] ]
-; CHECK-NEXT:    [[BROADCAST_SPLATINSERT1:%.*]] = insertelement <4 x i32> undef, i32 [[INDEX]], i32 0
-; CHECK-NEXT:    [[BROADCAST_SPLAT2:%.*]] = shufflevector <4 x i32> [[BROADCAST_SPLATINSERT1:%.*]], <4 x i32> undef, <4 x i32> zeroinitializer
-; CHECK-NEXT:    [[INDUCTION:%.*]] = add <4 x i32> [[BROADCAST_SPLAT2]], <i32 0, i32 1, i32 2, i32 3>
+; CHECK-NEXT:    [[INDEX:%.*]] = phi i32 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
+; CHECK-NEXT:    [[BROADCAST_SPLATINSERT:%.*]] = insertelement <4 x i32> undef, i32 [[INDEX]], i32 0
+; CHECK-NEXT:    [[BROADCAST_SPLAT:%.*]] = shufflevector <4 x i32> [[BROADCAST_SPLATINSERT]], <4 x i32> undef, <4 x i32> zeroinitializer
+; CHECK-NEXT:    [[INDUCTION:%.*]] = add <4 x i32> [[BROADCAST_SPLAT]], <i32 0, i32 1, i32 2, i32 3>
 ; CHECK-NEXT:    [[TMP0:%.*]] = add i32 [[INDEX]], 0
 ; CHECK-NEXT:    [[TMP1:%.*]] = mul nuw nsw i32 [[TMP0]], 1
 ; CHECK-NEXT:    [[TMP2:%.*]] = add nuw nsw i32 [[TMP1]], 2
 ; CHECK-NEXT:    [[TMP3:%.*]] = getelementptr inbounds i32, i32* [[DATA:%.*]], i32 [[TMP2]]
-; CHECK-NEXT:    [[TMP4:%.*]] = extractelement <4 x i32> [[BROADCAST_SPLAT:%.*]], i32 0
-; CHECK-NEXT:    [[ACTIVE_LANE_MASK:%.*]] = call <4 x i1> @llvm.get.active.lane.mask.v4i1.i32(i32 [[TMP0]], i32 [[TMP4]])
-; CHECK-NEXT:    [[TMP5:%.*]] = getelementptr inbounds i32, i32* [[TMP3]], i32 0
-; CHECK-NEXT:    [[TMP6:%.*]] = bitcast i32* [[TMP5]] to <4 x i32>*
-; CHECK-NEXT:    [[WIDE_MASKED_LOAD:%.*]] = call <4 x i32> @llvm.masked.load.v4i32.p0v4i32(<4 x i32>* [[TMP6]], i32 4, <4 x i1> [[ACTIVE_LANE_MASK]], <4 x i32> undef)
-; CHECK-NEXT:    [[TMP7:%.*]] = add nsw <4 x i32> <i32 5, i32 5, i32 5, i32 5>, [[WIDE_MASKED_LOAD]]
-; CHECK-NEXT:    [[TMP8:%.*]] = getelementptr inbounds i32, i32* [[DST:%.*]], i32 [[TMP0]]
-; CHECK-NEXT:    [[TMP9:%.*]] = getelementptr inbounds i32, i32* [[TMP8]], i32 0
-; CHECK-NEXT:    [[TMP10:%.*]] = bitcast i32* [[TMP9]] to <4 x i32>*
-; CHECK-NEXT:    call void @llvm.masked.store.v4i32.p0v4i32(<4 x i32> [[TMP7]], <4 x i32>* [[TMP10]], i32 4, <4 x i1> [[ACTIVE_LANE_MASK]])
+; CHECK-NEXT:    [[ACTIVE_LANE_MASK:%.*]] = call <4 x i1> @llvm.get.active.lane.mask.v4i1.i32(i32 [[TMP0]], i32 [[N]])
+; CHECK-NEXT:    [[TMP4:%.*]] = getelementptr inbounds i32, i32* [[TMP3]], i32 0
+; CHECK-NEXT:    [[TMP5:%.*]] = bitcast i32* [[TMP4]] to <4 x i32>*
+; CHECK-NEXT:    [[WIDE_MASKED_LOAD:%.*]] = call <4 x i32> @llvm.masked.load.v4i32.p0v4i32(<4 x i32>* [[TMP5]], i32 4, <4 x i1> [[ACTIVE_LANE_MASK]], <4 x i32> undef)
+; CHECK-NEXT:    [[TMP6:%.*]] = add nsw <4 x i32> <i32 5, i32 5, i32 5, i32 5>, [[WIDE_MASKED_LOAD]]
+; CHECK-NEXT:    [[TMP7:%.*]] = getelementptr inbounds i32, i32* [[DST:%.*]], i32 [[TMP0]]
+; CHECK-NEXT:    [[TMP8:%.*]] = getelementptr inbounds i32, i32* [[TMP7]], i32 0
+; CHECK-NEXT:    [[TMP9:%.*]] = bitcast i32* [[TMP8]] to <4 x i32>*
+; CHECK-NEXT:    call void @llvm.masked.store.v4i32.p0v4i32(<4 x i32> [[TMP6]], <4 x i32>* [[TMP9]], i32 4, <4 x i1> [[ACTIVE_LANE_MASK]])
 ; CHECK-NEXT:    [[INDEX_NEXT]] = add i32 [[INDEX]], 4
-; CHECK-NEXT:    [[TMP11:%.*]] = icmp eq i32 [[INDEX_NEXT]], [[N_VEC:%.*]]
-; CHECK-NEXT:    br i1 [[TMP11]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop !0
+; CHECK-NEXT:    [[TMP10:%.*]] = icmp eq i32 [[INDEX_NEXT]], [[N_VEC]]
+; CHECK-NEXT:    br i1 [[TMP10]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], [[LOOP0:!llvm.loop !.*]]
+; CHECK:       middle.block:
+; CHECK-NEXT:    br i1 true, label [[END:%.*]], label [[SCALAR_PH]]
+; CHECK:       scalar.ph:
+; CHECK-NEXT:    [[BC_RESUME_VAL:%.*]] = phi i32 [ [[N_VEC]], [[MIDDLE_BLOCK]] ], [ 0, [[ENTRY:%.*]] ]
+; CHECK-NEXT:    br label [[FOR_BODY:%.*]]
+; CHECK:       for.body:
+; CHECK-NEXT:    [[I_023:%.*]] = phi i32 [ [[INC:%.*]], [[FOR_BODY]] ], [ [[BC_RESUME_VAL]], [[SCALAR_PH]] ]
+; CHECK-NEXT:    [[MUL:%.*]] = mul nuw nsw i32 [[I_023]], 1
+; CHECK-NEXT:    [[ADD5:%.*]] = add nuw nsw i32 [[MUL]], 2
+; CHECK-NEXT:    [[ARRAYIDX6:%.*]] = getelementptr inbounds i32, i32* [[DATA]], i32 [[ADD5]]
+; CHECK-NEXT:    [[TMP11:%.*]] = load i32, i32* [[ARRAYIDX6]], align 4
+; CHECK-NEXT:    [[ADD7:%.*]] = add nsw i32 5, [[TMP11]]
+; CHECK-NEXT:    [[ARRAYIDX9:%.*]] = getelementptr inbounds i32, i32* [[DST]], i32 [[I_023]]
+; CHECK-NEXT:    store i32 [[ADD7]], i32* [[ARRAYIDX9]], align 4
+; CHECK-NEXT:    [[INC]] = add nuw nsw i32 [[I_023]], 1
+; CHECK-NEXT:    [[EXITCOND_NOT:%.*]] = icmp eq i32 [[INC]], [[N]]
+; CHECK-NEXT:    br i1 [[EXITCOND_NOT]], label [[END]], label [[FOR_BODY]], [[LOOP2:!llvm.loop !.*]]
+; CHECK:       end:
+; CHECK-NEXT:    ret void
 ;
 entry:
   br label %for.body
@@ -46,8 +71,28 @@ end:                                 ; preds = %end, %entry
 
 define void @test_stride-1_4i32(i32* readonly %data, i32* noalias nocapture %dst, i32 %n) {
 ; CHECK-LABEL: @test_stride-1_4i32(
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[MIN_ITERS_CHECK:%.*]] = icmp ult i32 [[N:%.*]], 4
+; CHECK-NEXT:    br i1 [[MIN_ITERS_CHECK]], label [[SCALAR_PH:%.*]], label [[VECTOR_SCEVCHECK:%.*]]
+; CHECK:       vector.scevcheck:
+; CHECK-NEXT:    [[TMP0:%.*]] = add i32 [[N]], -1
+; CHECK-NEXT:    [[MUL1:%.*]] = call { i32, i1 } @llvm.umul.with.overflow.i32(i32 1, i32 [[TMP0]])
+; CHECK-NEXT:    [[MUL_RESULT:%.*]] = extractvalue { i32, i1 } [[MUL1]], 0
+; CHECK-NEXT:    [[MUL_OVERFLOW:%.*]] = extractvalue { i32, i1 } [[MUL1]], 1
+; CHECK-NEXT:    [[TMP1:%.*]] = add i32 2, [[MUL_RESULT]]
+; CHECK-NEXT:    [[TMP2:%.*]] = sub i32 2, [[MUL_RESULT]]
+; CHECK-NEXT:    [[TMP3:%.*]] = icmp sgt i32 [[TMP2]], 2
+; CHECK-NEXT:    [[TMP4:%.*]] = icmp slt i32 [[TMP1]], 2
+; CHECK-NEXT:    [[TMP5:%.*]] = select i1 true, i1 [[TMP3]], i1 [[TMP4]]
+; CHECK-NEXT:    [[TMP6:%.*]] = or i1 [[TMP5]], [[MUL_OVERFLOW]]
+; CHECK-NEXT:    [[TMP7:%.*]] = or i1 false, [[TMP6]]
+; CHECK-NEXT:    br i1 [[TMP7]], label [[SCALAR_PH]], label [[VECTOR_PH:%.*]]
+; CHECK:       vector.ph:
+; CHECK-NEXT:    [[N_MOD_VF:%.*]] = urem i32 [[N]], 4
+; CHECK-NEXT:    [[N_VEC:%.*]] = sub i32 [[N]], [[N_MOD_VF]]
+; CHECK-NEXT:    br label [[VECTOR_BODY:%.*]]
 ; CHECK:       vector.body:
-; CHECK-NEXT:    [[INDEX:%.*]] = phi i32 [ 0, [[VECTOR_PH:%.*]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY:%.*]] ]
+; CHECK-NEXT:    [[INDEX:%.*]] = phi i32 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
 ; CHECK-NEXT:    [[TMP8:%.*]] = add i32 [[INDEX]], 0
 ; CHECK-NEXT:    [[TMP9:%.*]] = mul nuw nsw i32 [[TMP8]], -1
 ; CHECK-NEXT:    [[TMP10:%.*]] = add nuw nsw i32 [[TMP9]], 2
@@ -63,8 +108,28 @@ define void @test_stride-1_4i32(i32* readonly %data, i32* noalias nocapture %dst
 ; CHECK-NEXT:    [[TMP18:%.*]] = bitcast i32* [[TMP17]] to <4 x i32>*
 ; CHECK-NEXT:    store <4 x i32> [[TMP15]], <4 x i32>* [[TMP18]], align 4
 ; CHECK-NEXT:    [[INDEX_NEXT]] = add i32 [[INDEX]], 4
-; CHECK-NEXT:    [[TMP19:%.*]] = icmp eq i32 [[INDEX_NEXT]], [[N_VEC:%.*]]
-; CHECK-NEXT:    br i1 [[TMP19]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop !4
+; CHECK-NEXT:    [[TMP19:%.*]] = icmp eq i32 [[INDEX_NEXT]], [[N_VEC]]
+; CHECK-NEXT:    br i1 [[TMP19]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], [[LOOP4:!llvm.loop !.*]]
+; CHECK:       middle.block:
+; CHECK-NEXT:    [[CMP_N:%.*]] = icmp eq i32 [[N]], [[N_VEC]]
+; CHECK-NEXT:    br i1 [[CMP_N]], label [[END:%.*]], label [[SCALAR_PH]]
+; CHECK:       scalar.ph:
+; CHECK-NEXT:    [[BC_RESUME_VAL:%.*]] = phi i32 [ [[N_VEC]], [[MIDDLE_BLOCK]] ], [ 0, [[ENTRY:%.*]] ], [ 0, [[VECTOR_SCEVCHECK]] ]
+; CHECK-NEXT:    br label [[FOR_BODY:%.*]]
+; CHECK:       for.body:
+; CHECK-NEXT:    [[I_023:%.*]] = phi i32 [ [[INC:%.*]], [[FOR_BODY]] ], [ [[BC_RESUME_VAL]], [[SCALAR_PH]] ]
+; CHECK-NEXT:    [[MUL:%.*]] = mul nuw nsw i32 [[I_023]], -1
+; CHECK-NEXT:    [[ADD5:%.*]] = add nuw nsw i32 [[MUL]], 2
+; CHECK-NEXT:    [[ARRAYIDX6:%.*]] = getelementptr inbounds i32, i32* [[DATA]], i32 [[ADD5]]
+; CHECK-NEXT:    [[TMP20:%.*]] = load i32, i32* [[ARRAYIDX6]], align 4
+; CHECK-NEXT:    [[ADD7:%.*]] = add nsw i32 5, [[TMP20]]
+; CHECK-NEXT:    [[ARRAYIDX9:%.*]] = getelementptr inbounds i32, i32* [[DST]], i32 [[I_023]]
+; CHECK-NEXT:    store i32 [[ADD7]], i32* [[ARRAYIDX9]], align 4
+; CHECK-NEXT:    [[INC]] = add nuw nsw i32 [[I_023]], 1
+; CHECK-NEXT:    [[EXITCOND_NOT:%.*]] = icmp eq i32 [[INC]], [[N]]
+; CHECK-NEXT:    br i1 [[EXITCOND_NOT]], label [[END]], label [[FOR_BODY]], [[LOOP5:!llvm.loop !.*]]
+; CHECK:       end:
+; CHECK-NEXT:    ret void
 ;
 entry:
   br label %for.body
@@ -87,8 +152,17 @@ end:                                 ; preds = %end, %entry
 define void @test_stride2_4i32(i32* readonly %data, i32* noalias nocapture %dst, i32 %n) {
 ;
 ; CHECK-LABEL: @test_stride2_4i32(
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[MIN_ITERS_CHECK:%.*]] = icmp ule i32 [[N:%.*]], 4
+; CHECK-NEXT:    br i1 [[MIN_ITERS_CHECK]], label [[SCALAR_PH:%.*]], label [[VECTOR_PH:%.*]]
+; CHECK:       vector.ph:
+; CHECK-NEXT:    [[N_MOD_VF:%.*]] = urem i32 [[N]], 4
+; CHECK-NEXT:    [[TMP0:%.*]] = icmp eq i32 [[N_MOD_VF]], 0
+; CHECK-NEXT:    [[TMP1:%.*]] = select i1 [[TMP0]], i32 4, i32 [[N_MOD_VF]]
+; CHECK-NEXT:    [[N_VEC:%.*]] = sub i32 [[N]], [[TMP1]]
+; CHECK-NEXT:    br label [[VECTOR_BODY:%.*]]
 ; CHECK:       vector.body:
-; CHECK-NEXT:    [[INDEX:%.*]] = phi i32 [ 0, [[VECTOR_PH:%.*]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY:%.*]] ]
+; CHECK-NEXT:    [[INDEX:%.*]] = phi i32 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
 ; CHECK-NEXT:    [[TMP2:%.*]] = add i32 [[INDEX]], 0
 ; CHECK-NEXT:    [[TMP3:%.*]] = mul nuw nsw i32 [[TMP2]], 2
 ; CHECK-NEXT:    [[TMP4:%.*]] = add nuw nsw i32 [[TMP3]], 2
@@ -103,8 +177,28 @@ define void @test_stride2_4i32(i32* readonly %data, i32* noalias nocapture %dst,
 ; CHECK-NEXT:    [[TMP11:%.*]] = bitcast i32* [[TMP10]] to <4 x i32>*
 ; CHECK-NEXT:    store <4 x i32> [[TMP8]], <4 x i32>* [[TMP11]], align 4
 ; CHECK-NEXT:    [[INDEX_NEXT]] = add i32 [[INDEX]], 4
-; CHECK-NEXT:    [[TMP12:%.*]] = icmp eq i32 [[INDEX_NEXT]], [[N_VEC:%.*]]
-; CHECK-NEXT:    br i1 [[TMP12]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop !6
+; CHECK-NEXT:    [[TMP12:%.*]] = icmp eq i32 [[INDEX_NEXT]], [[N_VEC]]
+; CHECK-NEXT:    br i1 [[TMP12]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], [[LOOP6:!llvm.loop !.*]]
+; CHECK:       middle.block:
+; CHECK-NEXT:    [[CMP_N:%.*]] = icmp eq i32 [[N]], [[N_VEC]]
+; CHECK-NEXT:    br i1 [[CMP_N]], label [[END:%.*]], label [[SCALAR_PH]]
+; CHECK:       scalar.ph:
+; CHECK-NEXT:    [[BC_RESUME_VAL:%.*]] = phi i32 [ [[N_VEC]], [[MIDDLE_BLOCK]] ], [ 0, [[ENTRY:%.*]] ]
+; CHECK-NEXT:    br label [[FOR_BODY:%.*]]
+; CHECK:       for.body:
+; CHECK-NEXT:    [[I_023:%.*]] = phi i32 [ [[INC:%.*]], [[FOR_BODY]] ], [ [[BC_RESUME_VAL]], [[SCALAR_PH]] ]
+; CHECK-NEXT:    [[MUL:%.*]] = mul nuw nsw i32 [[I_023]], 2
+; CHECK-NEXT:    [[ADD5:%.*]] = add nuw nsw i32 [[MUL]], 2
+; CHECK-NEXT:    [[ARRAYIDX6:%.*]] = getelementptr inbounds i32, i32* [[DATA]], i32 [[ADD5]]
+; CHECK-NEXT:    [[TMP13:%.*]] = load i32, i32* [[ARRAYIDX6]], align 4
+; CHECK-NEXT:    [[ADD7:%.*]] = add nsw i32 5, [[TMP13]]
+; CHECK-NEXT:    [[ARRAYIDX9:%.*]] = getelementptr inbounds i32, i32* [[DST]], i32 [[I_023]]
+; CHECK-NEXT:    store i32 [[ADD7]], i32* [[ARRAYIDX9]], align 4
+; CHECK-NEXT:    [[INC]] = add nuw nsw i32 [[I_023]], 1
+; CHECK-NEXT:    [[EXITCOND_NOT:%.*]] = icmp eq i32 [[INC]], [[N]]
+; CHECK-NEXT:    br i1 [[EXITCOND_NOT]], label [[END]], label [[FOR_BODY]], [[LOOP7:!llvm.loop !.*]]
+; CHECK:       end:
+; CHECK-NEXT:    ret void
 ;
 entry:
   br label %for.body
@@ -126,8 +220,15 @@ end:                                 ; preds = %end, %entry
 
 define void @test_stride3_4i32(i32* readonly %data, i32* noalias nocapture %dst, i32 %n) {
 ; CHECK-LABEL: @test_stride3_4i32(
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    br i1 false, label [[SCALAR_PH:%.*]], label [[VECTOR_PH:%.*]]
+; CHECK:       vector.ph:
+; CHECK-NEXT:    [[N_RND_UP:%.*]] = add i32 [[N:%.*]], 3
+; CHECK-NEXT:    [[N_MOD_VF:%.*]] = urem i32 [[N_RND_UP]], 4
+; CHECK-NEXT:    [[N_VEC:%.*]] = sub i32 [[N_RND_UP]], [[N_MOD_VF]]
+; CHECK-NEXT:    br label [[VECTOR_BODY:%.*]]
 ; CHECK:       vector.body:
-; CHECK-NEXT:    [[INDEX:%.*]] = phi i32 [ 0, [[VECTOR_PH:%.*]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY:%.*]] ]
+; CHECK-NEXT:    [[INDEX:%.*]] = phi i32 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
 ; CHECK-NEXT:    [[VEC_IND:%.*]] = phi <4 x i32> [ <i32 0, i32 1, i32 2, i32 3>, [[VECTOR_PH]] ], [ [[VEC_IND_NEXT:%.*]], [[VECTOR_BODY]] ]
 ; CHECK-NEXT:    [[TMP0:%.*]] = add i32 [[INDEX]], 0
 ; CHECK-NEXT:    [[TMP1:%.*]] = add i32 [[INDEX]], 1
@@ -136,18 +237,36 @@ define void @test_stride3_4i32(i32* readonly %data, i32* noalias nocapture %dst,
 ; CHECK-NEXT:    [[TMP4:%.*]] = mul nuw nsw <4 x i32> [[VEC_IND]], <i32 3, i32 3, i32 3, i32 3>
 ; CHECK-NEXT:    [[TMP5:%.*]] = add nuw nsw <4 x i32> [[TMP4]], <i32 2, i32 2, i32 2, i32 2>
 ; CHECK-NEXT:    [[TMP6:%.*]] = getelementptr inbounds i32, i32* [[DATA:%.*]], <4 x i32> [[TMP5]]
-; CHECK-NEXT:    [[TMP7:%.*]] = extractelement <4 x i32> [[BROADCAST_SPLAT:%.*]], i32 0
-; CHECK-NEXT:    [[ACTIVE_LANE_MASK:%.*]] = call <4 x i1> @llvm.get.active.lane.mask.v4i1.i32(i32 [[TMP0]], i32 [[TMP7]])
+; CHECK-NEXT:    [[ACTIVE_LANE_MASK:%.*]] = call <4 x i1> @llvm.get.active.lane.mask.v4i1.i32(i32 [[TMP0]], i32 [[N]])
 ; CHECK-NEXT:    [[WIDE_MASKED_GATHER:%.*]] = call <4 x i32> @llvm.masked.gather.v4i32.v4p0i32(<4 x i32*> [[TMP6]], i32 4, <4 x i1> [[ACTIVE_LANE_MASK]], <4 x i32> undef)
-; CHECK-NEXT:    [[TMP8:%.*]] = add nsw <4 x i32> <i32 5, i32 5, i32 5, i32 5>, [[WIDE_MASKED_GATHER]]
-; CHECK-NEXT:    [[TMP9:%.*]] = getelementptr inbounds i32, i32* [[DST:%.*]], i32 [[TMP0]]
-; CHECK-NEXT:    [[TMP10:%.*]] = getelementptr inbounds i32, i32* [[TMP9]], i32 0
-; CHECK-NEXT:    [[TMP11:%.*]] = bitcast i32* [[TMP10]] to <4 x i32>*
-; CHECK-NEXT:    call void @llvm.masked.store.v4i32.p0v4i32(<4 x i32> [[TMP8]], <4 x i32>* [[TMP11]], i32 4, <4 x i1> [[ACTIVE_LANE_MASK]])
+; CHECK-NEXT:    [[TMP7:%.*]] = add nsw <4 x i32> <i32 5, i32 5, i32 5, i32 5>, [[WIDE_MASKED_GATHER]]
+; CHECK-NEXT:    [[TMP8:%.*]] = getelementptr inbounds i32, i32* [[DST:%.*]], i32 [[TMP0]]
+; CHECK-NEXT:    [[TMP9:%.*]] = getelementptr inbounds i32, i32* [[TMP8]], i32 0
+; CHECK-NEXT:    [[TMP10:%.*]] = bitcast i32* [[TMP9]] to <4 x i32>*
+; CHECK-NEXT:    call void @llvm.masked.store.v4i32.p0v4i32(<4 x i32> [[TMP7]], <4 x i32>* [[TMP10]], i32 4, <4 x i1> [[ACTIVE_LANE_MASK]])
 ; CHECK-NEXT:    [[INDEX_NEXT]] = add i32 [[INDEX]], 4
 ; CHECK-NEXT:    [[VEC_IND_NEXT]] = add <4 x i32> [[VEC_IND]], <i32 4, i32 4, i32 4, i32 4>
-; CHECK-NEXT:    [[TMP12:%.*]] = icmp eq i32 [[INDEX_NEXT]], [[N_VEC:%.*]]
-; CHECK-NEXT:    br i1 [[TMP12]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop !8
+; CHECK-NEXT:    [[TMP11:%.*]] = icmp eq i32 [[INDEX_NEXT]], [[N_VEC]]
+; CHECK-NEXT:    br i1 [[TMP11]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], [[LOOP8:!llvm.loop !.*]]
+; CHECK:       middle.block:
+; CHECK-NEXT:    br i1 true, label [[END:%.*]], label [[SCALAR_PH]]
+; CHECK:       scalar.ph:
+; CHECK-NEXT:    [[BC_RESUME_VAL:%.*]] = phi i32 [ [[N_VEC]], [[MIDDLE_BLOCK]] ], [ 0, [[ENTRY:%.*]] ]
+; CHECK-NEXT:    br label [[FOR_BODY:%.*]]
+; CHECK:       for.body:
+; CHECK-NEXT:    [[I_023:%.*]] = phi i32 [ [[INC:%.*]], [[FOR_BODY]] ], [ [[BC_RESUME_VAL]], [[SCALAR_PH]] ]
+; CHECK-NEXT:    [[MUL:%.*]] = mul nuw nsw i32 [[I_023]], 3
+; CHECK-NEXT:    [[ADD5:%.*]] = add nuw nsw i32 [[MUL]], 2
+; CHECK-NEXT:    [[ARRAYIDX6:%.*]] = getelementptr inbounds i32, i32* [[DATA]], i32 [[ADD5]]
+; CHECK-NEXT:    [[TMP12:%.*]] = load i32, i32* [[ARRAYIDX6]], align 4
+; CHECK-NEXT:    [[ADD7:%.*]] = add nsw i32 5, [[TMP12]]
+; CHECK-NEXT:    [[ARRAYIDX9:%.*]] = getelementptr inbounds i32, i32* [[DST]], i32 [[I_023]]
+; CHECK-NEXT:    store i32 [[ADD7]], i32* [[ARRAYIDX9]], align 4
+; CHECK-NEXT:    [[INC]] = add nuw nsw i32 [[I_023]], 1
+; CHECK-NEXT:    [[EXITCOND_NOT:%.*]] = icmp eq i32 [[INC]], [[N]]
+; CHECK-NEXT:    br i1 [[EXITCOND_NOT]], label [[END]], label [[FOR_BODY]], [[LOOP9:!llvm.loop !.*]]
+; CHECK:       end:
+; CHECK-NEXT:    ret void
 ;
 entry:
   br label %for.body
@@ -169,8 +288,15 @@ end:                                 ; preds = %end, %entry
 
 define void @test_stride4_4i32(i32* readonly %data, i32* noalias nocapture %dst, i32 %n) {
 ; CHECK-LABEL: @test_stride4_4i32(
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    br i1 false, label [[SCALAR_PH:%.*]], label [[VECTOR_PH:%.*]]
+; CHECK:       vector.ph:
+; CHECK-NEXT:    [[N_RND_UP:%.*]] = add i32 [[N:%.*]], 3
+; CHECK-NEXT:    [[N_MOD_VF:%.*]] = urem i32 [[N_RND_UP]], 4
+; CHECK-NEXT:    [[N_VEC:%.*]] = sub i32 [[N_RND_UP]], [[N_MOD_VF]]
+; CHECK-NEXT:    br label [[VECTOR_BODY:%.*]]
 ; CHECK:       vector.body:
-; CHECK-NEXT:    [[INDEX:%.*]] = phi i32 [ 0, [[VECTOR_PH:%.*]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY:%.*]] ]
+; CHECK-NEXT:    [[INDEX:%.*]] = phi i32 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
 ; CHECK-NEXT:    [[VEC_IND:%.*]] = phi <4 x i32> [ <i32 0, i32 1, i32 2, i32 3>, [[VECTOR_PH]] ], [ [[VEC_IND_NEXT:%.*]], [[VECTOR_BODY]] ]
 ; CHECK-NEXT:    [[TMP0:%.*]] = add i32 [[INDEX]], 0
 ; CHECK-NEXT:    [[TMP1:%.*]] = add i32 [[INDEX]], 1
@@ -179,18 +305,36 @@ define void @test_stride4_4i32(i32* readonly %data, i32* noalias nocapture %dst,
 ; CHECK-NEXT:    [[TMP4:%.*]] = mul nuw nsw <4 x i32> [[VEC_IND]], <i32 4, i32 4, i32 4, i32 4>
 ; CHECK-NEXT:    [[TMP5:%.*]] = add nuw nsw <4 x i32> [[TMP4]], <i32 2, i32 2, i32 2, i32 2>
 ; CHECK-NEXT:    [[TMP6:%.*]] = getelementptr inbounds i32, i32* [[DATA:%.*]], <4 x i32> [[TMP5]]
-; CHECK-NEXT:    [[TMP7:%.*]] = extractelement <4 x i32> [[BROADCAST_SPLAT:%.*]], i32 0
-; CHECK-NEXT:    [[ACTIVE_LANE_MASK:%.*]] = call <4 x i1> @llvm.get.active.lane.mask.v4i1.i32(i32 [[TMP0]], i32 [[TMP7]])
+; CHECK-NEXT:    [[ACTIVE_LANE_MASK:%.*]] = call <4 x i1> @llvm.get.active.lane.mask.v4i1.i32(i32 [[TMP0]], i32 [[N]])
 ; CHECK-NEXT:    [[WIDE_MASKED_GATHER:%.*]] = call <4 x i32> @llvm.masked.gather.v4i32.v4p0i32(<4 x i32*> [[TMP6]], i32 4, <4 x i1> [[ACTIVE_LANE_MASK]], <4 x i32> undef)
-; CHECK-NEXT:    [[TMP8:%.*]] = add nsw <4 x i32> <i32 5, i32 5, i32 5, i32 5>, [[WIDE_MASKED_GATHER]]
-; CHECK-NEXT:    [[TMP9:%.*]] = getelementptr inbounds i32, i32* [[DST:%.*]], i32 [[TMP0]]
-; CHECK-NEXT:    [[TMP10:%.*]] = getelementptr inbounds i32, i32* [[TMP9]], i32 0
-; CHECK-NEXT:    [[TMP11:%.*]] = bitcast i32* [[TMP10]] to <4 x i32>*
-; CHECK-NEXT:    call void @llvm.masked.store.v4i32.p0v4i32(<4 x i32> [[TMP8]], <4 x i32>* [[TMP11]], i32 4, <4 x i1> [[ACTIVE_LANE_MASK]])
+; CHECK-NEXT:    [[TMP7:%.*]] = add nsw <4 x i32> <i32 5, i32 5, i32 5, i32 5>, [[WIDE_MASKED_GATHER]]
+; CHECK-NEXT:    [[TMP8:%.*]] = getelementptr inbounds i32, i32* [[DST:%.*]], i32 [[TMP0]]
+; CHECK-NEXT:    [[TMP9:%.*]] = getelementptr inbounds i32, i32* [[TMP8]], i32 0
+; CHECK-NEXT:    [[TMP10:%.*]] = bitcast i32* [[TMP9]] to <4 x i32>*
+; CHECK-NEXT:    call void @llvm.masked.store.v4i32.p0v4i32(<4 x i32> [[TMP7]], <4 x i32>* [[TMP10]], i32 4, <4 x i1> [[ACTIVE_LANE_MASK]])
 ; CHECK-NEXT:    [[INDEX_NEXT]] = add i32 [[INDEX]], 4
 ; CHECK-NEXT:    [[VEC_IND_NEXT]] = add <4 x i32> [[VEC_IND]], <i32 4, i32 4, i32 4, i32 4>
-; CHECK-NEXT:    [[TMP12:%.*]] = icmp eq i32 [[INDEX_NEXT]], [[N_VEC:%.*]]
-; CHECK-NEXT:    br i1 [[TMP12]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], [[LOOP10:!llvm.loop !.*]]
+; CHECK-NEXT:    [[TMP11:%.*]] = icmp eq i32 [[INDEX_NEXT]], [[N_VEC]]
+; CHECK-NEXT:    br i1 [[TMP11]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], [[LOOP10:!llvm.loop !.*]]
+; CHECK:       middle.block:
+; CHECK-NEXT:    br i1 true, label [[END:%.*]], label [[SCALAR_PH]]
+; CHECK:       scalar.ph:
+; CHECK-NEXT:    [[BC_RESUME_VAL:%.*]] = phi i32 [ [[N_VEC]], [[MIDDLE_BLOCK]] ], [ 0, [[ENTRY:%.*]] ]
+; CHECK-NEXT:    br label [[FOR_BODY:%.*]]
+; CHECK:       for.body:
+; CHECK-NEXT:    [[I_023:%.*]] = phi i32 [ [[INC:%.*]], [[FOR_BODY]] ], [ [[BC_RESUME_VAL]], [[SCALAR_PH]] ]
+; CHECK-NEXT:    [[MUL:%.*]] = mul nuw nsw i32 [[I_023]], 4
+; CHECK-NEXT:    [[ADD5:%.*]] = add nuw nsw i32 [[MUL]], 2
+; CHECK-NEXT:    [[ARRAYIDX6:%.*]] = getelementptr inbounds i32, i32* [[DATA]], i32 [[ADD5]]
+; CHECK-NEXT:    [[TMP12:%.*]] = load i32, i32* [[ARRAYIDX6]], align 4
+; CHECK-NEXT:    [[ADD7:%.*]] = add nsw i32 5, [[TMP12]]
+; CHECK-NEXT:    [[ARRAYIDX9:%.*]] = getelementptr inbounds i32, i32* [[DST]], i32 [[I_023]]
+; CHECK-NEXT:    store i32 [[ADD7]], i32* [[ARRAYIDX9]], align 4
+; CHECK-NEXT:    [[INC]] = add nuw nsw i32 [[I_023]], 1
+; CHECK-NEXT:    [[EXITCOND_NOT:%.*]] = icmp eq i32 [[INC]], [[N]]
+; CHECK-NEXT:    br i1 [[EXITCOND_NOT]], label [[END]], label [[FOR_BODY]], [[LOOP11:!llvm.loop !.*]]
+; CHECK:       end:
+; CHECK-NEXT:    ret void
 ;
 entry:
   br label %for.body
@@ -212,28 +356,57 @@ end:                                 ; preds = %end, %entry
 
 define void @test_stride_loopinvar_4i32(i32* readonly %data, i32* noalias nocapture %dst, i32 %n, i32 %stride) {
 ; CHECK-LABEL: @test_stride_loopinvar_4i32(
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    br i1 false, label [[SCALAR_PH:%.*]], label [[VECTOR_SCEVCHECK:%.*]]
+; CHECK:       vector.scevcheck:
+; CHECK-NEXT:    [[IDENT_CHECK:%.*]] = icmp ne i32 [[STRIDE:%.*]], 1
+; CHECK-NEXT:    [[TMP0:%.*]] = or i1 false, [[IDENT_CHECK]]
+; CHECK-NEXT:    br i1 [[TMP0]], label [[SCALAR_PH]], label [[VECTOR_PH:%.*]]
+; CHECK:       vector.ph:
+; CHECK-NEXT:    [[N_RND_UP:%.*]] = add i32 [[N:%.*]], 3
+; CHECK-NEXT:    [[N_MOD_VF:%.*]] = urem i32 [[N_RND_UP]], 4
+; CHECK-NEXT:    [[N_VEC:%.*]] = sub i32 [[N_RND_UP]], [[N_MOD_VF]]
+; CHECK-NEXT:    br label [[VECTOR_BODY:%.*]]
 ; CHECK:       vector.body:
-; CHECK-NEXT:    [[INDEX:%.*]] = phi i32 [ 0, [[VECTOR_PH:%.*]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY:%.*]] ]
-; CHECK-NEXT:    [[BROADCAST_SPLATINSERT1:%.*]] = insertelement <4 x i32> undef, i32 [[INDEX]], i32 0
-; CHECK-NEXT:    [[BROADCAST_SPLAT2:%.*]] = shufflevector <4 x i32> [[BROADCAST_SPLATINSERT1]], <4 x i32> undef, <4 x i32> zeroinitializer
-; CHECK-NEXT:    [[INDUCTION:%.*]] = add <4 x i32> [[BROADCAST_SPLAT2]], <i32 0, i32 1, i32 2, i32 3>
+; CHECK-NEXT:    [[INDEX:%.*]] = phi i32 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
+; CHECK-NEXT:    [[BROADCAST_SPLATINSERT:%.*]] = insertelement <4 x i32> undef, i32 [[INDEX]], i32 0
+; CHECK-NEXT:    [[BROADCAST_SPLAT:%.*]] = shufflevector <4 x i32> [[BROADCAST_SPLATINSERT]], <4 x i32> undef, <4 x i32> zeroinitializer
+; CHECK-NEXT:    [[INDUCTION:%.*]] = add <4 x i32> [[BROADCAST_SPLAT]], <i32 0, i32 1, i32 2, i32 3>
 ; CHECK-NEXT:    [[TMP1:%.*]] = add i32 [[INDEX]], 0
-; CHECK-NEXT:    [[TMP2:%.*]] = mul nuw nsw i32 [[TMP1]], [[STRIDE:%.*]]
+; CHECK-NEXT:    [[TMP2:%.*]] = mul nuw nsw i32 [[TMP1]], [[STRIDE]]
 ; CHECK-NEXT:    [[TMP3:%.*]] = add nuw nsw i32 [[TMP2]], 2
 ; CHECK-NEXT:    [[TMP4:%.*]] = getelementptr inbounds i32, i32* [[DATA:%.*]], i32 [[TMP3]]
-; CHECK-NEXT:    [[TMP5:%.*]] = extractelement <4 x i32> [[BROADCAST_SPLAT]], i32 0
-; CHECK-NEXT:    [[ACTIVE_LANE_MASK:%.*]] = call <4 x i1> @llvm.get.active.lane.mask.v4i1.i32(i32 [[TMP1]], i32 [[TMP5]])
-; CHECK-NEXT:    [[TMP6:%.*]] = getelementptr inbounds i32, i32* [[TMP4]], i32 0
-; CHECK-NEXT:    [[TMP7:%.*]] = bitcast i32* [[TMP6]] to <4 x i32>*
-; CHECK-NEXT:    [[WIDE_MASKED_LOAD:%.*]] = call <4 x i32> @llvm.masked.load.v4i32.p0v4i32(<4 x i32>* [[TMP7]], i32 4, <4 x i1> [[ACTIVE_LANE_MASK]], <4 x i32> undef)
-; CHECK-NEXT:    [[TMP8:%.*]] = add nsw <4 x i32> <i32 5, i32 5, i32 5, i32 5>, [[WIDE_MASKED_LOAD]]
-; CHECK-NEXT:    [[TMP9:%.*]] = getelementptr inbounds i32, i32* [[DST:%.*]], i32 [[TMP1]]
-; CHECK-NEXT:    [[TMP10:%.*]] = getelementptr inbounds i32, i32* [[TMP9]], i32 0
-; CHECK-NEXT:    [[TMP11:%.*]] = bitcast i32* [[TMP10]] to <4 x i32>*
-; CHECK-NEXT:    call void @llvm.masked.store.v4i32.p0v4i32(<4 x i32> [[TMP8]], <4 x i32>* [[TMP11]], i32 4, <4 x i1> [[ACTIVE_LANE_MASK]])
+; CHECK-NEXT:    [[ACTIVE_LANE_MASK:%.*]] = call <4 x i1> @llvm.get.active.lane.mask.v4i1.i32(i32 [[TMP1]], i32 [[N]])
+; CHECK-NEXT:    [[TMP5:%.*]] = getelementptr inbounds i32, i32* [[TMP4]], i32 0
+; CHECK-NEXT:    [[TMP6:%.*]] = bitcast i32* [[TMP5]] to <4 x i32>*
+; CHECK-NEXT:    [[WIDE_MASKED_LOAD:%.*]] = call <4 x i32> @llvm.masked.load.v4i32.p0v4i32(<4 x i32>* [[TMP6]], i32 4, <4 x i1> [[ACTIVE_LANE_MASK]], <4 x i32> undef)
+; CHECK-NEXT:    [[TMP7:%.*]] = add nsw <4 x i32> <i32 5, i32 5, i32 5, i32 5>, [[WIDE_MASKED_LOAD]]
+; CHECK-NEXT:    [[TMP8:%.*]] = getelementptr inbounds i32, i32* [[DST:%.*]], i32 [[TMP1]]
+; CHECK-NEXT:    [[TMP9:%.*]] = getelementptr inbounds i32, i32* [[TMP8]], i32 0
+; CHECK-NEXT:    [[TMP10:%.*]] = bitcast i32* [[TMP9]] to <4 x i32>*
+; CHECK-NEXT:    call void @llvm.masked.store.v4i32.p0v4i32(<4 x i32> [[TMP7]], <4 x i32>* [[TMP10]], i32 4, <4 x i1> [[ACTIVE_LANE_MASK]])
 ; CHECK-NEXT:    [[INDEX_NEXT]] = add i32 [[INDEX]], 4
-; CHECK-NEXT:    [[TMP12:%.*]] = icmp eq i32 [[INDEX_NEXT]], [[N_VEC:%.*]]
-; CHECK-NEXT:    br i1 [[TMP12]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], [[LOOP12:!llvm.loop !.*]]
+; CHECK-NEXT:    [[TMP11:%.*]] = icmp eq i32 [[INDEX_NEXT]], [[N_VEC]]
+; CHECK-NEXT:    br i1 [[TMP11]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], [[LOOP12:!llvm.loop !.*]]
+; CHECK:       middle.block:
+; CHECK-NEXT:    br i1 true, label [[END:%.*]], label [[SCALAR_PH]]
+; CHECK:       scalar.ph:
+; CHECK-NEXT:    [[BC_RESUME_VAL:%.*]] = phi i32 [ [[N_VEC]], [[MIDDLE_BLOCK]] ], [ 0, [[ENTRY:%.*]] ], [ 0, [[VECTOR_SCEVCHECK]] ]
+; CHECK-NEXT:    br label [[FOR_BODY:%.*]]
+; CHECK:       for.body:
+; CHECK-NEXT:    [[I_023:%.*]] = phi i32 [ [[INC:%.*]], [[FOR_BODY]] ], [ [[BC_RESUME_VAL]], [[SCALAR_PH]] ]
+; CHECK-NEXT:    [[MUL:%.*]] = mul nuw nsw i32 [[I_023]], [[STRIDE]]
+; CHECK-NEXT:    [[ADD5:%.*]] = add nuw nsw i32 [[MUL]], 2
+; CHECK-NEXT:    [[ARRAYIDX6:%.*]] = getelementptr inbounds i32, i32* [[DATA]], i32 [[ADD5]]
+; CHECK-NEXT:    [[TMP12:%.*]] = load i32, i32* [[ARRAYIDX6]], align 4
+; CHECK-NEXT:    [[ADD7:%.*]] = add nsw i32 5, [[TMP12]]
+; CHECK-NEXT:    [[ARRAYIDX9:%.*]] = getelementptr inbounds i32, i32* [[DST]], i32 [[I_023]]
+; CHECK-NEXT:    store i32 [[ADD7]], i32* [[ARRAYIDX9]], align 4
+; CHECK-NEXT:    [[INC]] = add nuw nsw i32 [[I_023]], 1
+; CHECK-NEXT:    [[EXITCOND_NOT:%.*]] = icmp eq i32 [[INC]], [[N]]
+; CHECK-NEXT:    br i1 [[EXITCOND_NOT]], label [[END]], label [[FOR_BODY]], [[LOOP13:!llvm.loop !.*]]
+; CHECK:       end:
+; CHECK-NEXT:    ret void
 ;
 entry:
   br label %for.body
@@ -255,8 +428,17 @@ end:                                 ; preds = %end, %entry
 
 define void @test_stride_noninvar_4i32(i32* readonly %data, i32* noalias nocapture %dst, i32 %n) {
 ; CHECK-LABEL: @test_stride_noninvar_4i32(
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[MIN_ITERS_CHECK:%.*]] = icmp ult i32 [[N:%.*]], 4
+; CHECK-NEXT:    br i1 [[MIN_ITERS_CHECK]], label [[SCALAR_PH:%.*]], label [[VECTOR_PH:%.*]]
+; CHECK:       vector.ph:
+; CHECK-NEXT:    [[N_MOD_VF:%.*]] = urem i32 [[N]], 4
+; CHECK-NEXT:    [[N_VEC:%.*]] = sub i32 [[N]], [[N_MOD_VF]]
+; CHECK-NEXT:    [[TMP0:%.*]] = mul i32 [[N_VEC]], 8
+; CHECK-NEXT:    [[IND_END:%.*]] = add i32 3, [[TMP0]]
+; CHECK-NEXT:    br label [[VECTOR_BODY:%.*]]
 ; CHECK:       vector.body:
-; CHECK-NEXT:    [[INDEX:%.*]] = phi i32 [ 0, [[VECTOR_PH:%.*]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY:%.*]] ]
+; CHECK-NEXT:    [[INDEX:%.*]] = phi i32 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
 ; CHECK-NEXT:    [[VEC_IND:%.*]] = phi <4 x i32> [ <i32 0, i32 1, i32 2, i32 3>, [[VECTOR_PH]] ], [ [[VEC_IND_NEXT:%.*]], [[VECTOR_BODY]] ]
 ; CHECK-NEXT:    [[VEC_IND2:%.*]] = phi <4 x i32> [ <i32 3, i32 11, i32 19, i32 27>, [[VECTOR_PH]] ], [ [[VEC_IND_NEXT3:%.*]], [[VECTOR_BODY]] ]
 ; CHECK-NEXT:    [[TMP1:%.*]] = add i32 [[INDEX]], 0
@@ -275,8 +457,31 @@ define void @test_stride_noninvar_4i32(i32* readonly %data, i32* noalias nocaptu
 ; CHECK-NEXT:    [[INDEX_NEXT]] = add i32 [[INDEX]], 4
 ; CHECK-NEXT:    [[VEC_IND_NEXT]] = add <4 x i32> [[VEC_IND]], <i32 4, i32 4, i32 4, i32 4>
 ; CHECK-NEXT:    [[VEC_IND_NEXT3]] = add <4 x i32> [[VEC_IND2]], <i32 32, i32 32, i32 32, i32 32>
-; CHECK-NEXT:    [[TMP12:%.*]] = icmp eq i32 [[INDEX_NEXT]], [[N_VEC:%.*]]
+; CHECK-NEXT:    [[TMP12:%.*]] = icmp eq i32 [[INDEX_NEXT]], [[N_VEC]]
 ; CHECK-NEXT:    br i1 [[TMP12]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], [[LOOP14:!llvm.loop !.*]]
+; CHECK:       middle.block:
+; CHECK-NEXT:    [[CMP_N:%.*]] = icmp eq i32 [[N]], [[N_VEC]]
+; CHECK-NEXT:    br i1 [[CMP_N]], label [[END:%.*]], label [[SCALAR_PH]]
+; CHECK:       scalar.ph:
+; CHECK-NEXT:    [[BC_RESUME_VAL:%.*]] = phi i32 [ [[N_VEC]], [[MIDDLE_BLOCK]] ], [ 0, [[ENTRY:%.*]] ]
+; CHECK-NEXT:    [[BC_RESUME_VAL1:%.*]] = phi i32 [ [[IND_END]], [[MIDDLE_BLOCK]] ], [ 3, [[ENTRY]] ]
+; CHECK-NEXT:    br label [[FOR_BODY:%.*]]
+; CHECK:       for.body:
+; CHECK-NEXT:    [[I_023:%.*]] = phi i32 [ [[INC:%.*]], [[FOR_BODY]] ], [ [[BC_RESUME_VAL]], [[SCALAR_PH]] ]
+; CHECK-NEXT:    [[STRIDE:%.*]] = phi i32 [ [[NEXT_STRIDE:%.*]], [[FOR_BODY]] ], [ [[BC_RESUME_VAL1]], [[SCALAR_PH]] ]
+; CHECK-NEXT:    [[MUL:%.*]] = mul nuw nsw i32 [[I_023]], [[STRIDE]]
+; CHECK-NEXT:    [[ADD5:%.*]] = add nuw nsw i32 [[MUL]], 2
+; CHECK-NEXT:    [[ARRAYIDX6:%.*]] = getelementptr inbounds i32, i32* [[DATA]], i32 [[ADD5]]
+; CHECK-NEXT:    [[TMP13:%.*]] = load i32, i32* [[ARRAYIDX6]], align 4
+; CHECK-NEXT:    [[ADD7:%.*]] = add nsw i32 5, [[TMP13]]
+; CHECK-NEXT:    [[ARRAYIDX9:%.*]] = getelementptr inbounds i32, i32* [[DST]], i32 [[I_023]]
+; CHECK-NEXT:    store i32 [[ADD7]], i32* [[ARRAYIDX9]], align 4
+; CHECK-NEXT:    [[INC]] = add nuw nsw i32 [[I_023]], 1
+; CHECK-NEXT:    [[NEXT_STRIDE]] = add nuw nsw i32 [[STRIDE]], 8
+; CHECK-NEXT:    [[EXITCOND_NOT:%.*]] = icmp eq i32 [[INC]], [[N]]
+; CHECK-NEXT:    br i1 [[EXITCOND_NOT]], label [[END]], label [[FOR_BODY]], [[LOOP15:!llvm.loop !.*]]
+; CHECK:       end:
+; CHECK-NEXT:    ret void
 ;
 entry:
   br label %for.body
@@ -341,10 +546,26 @@ end:                                 ; preds = %end, %entry
 
 define void @test_stride_noninvar3_4i32(i32* readonly %data, i32* noalias nocapture %dst, i32 %n, i32 %x) {
 ; CHECK-LABEL: @test_stride_noninvar3_4i32(
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[MIN_ITERS_CHECK:%.*]] = icmp ult i32 [[N:%.*]], 4
+; CHECK-NEXT:    br i1 [[MIN_ITERS_CHECK]], label [[SCALAR_PH:%.*]], label [[VECTOR_PH:%.*]]
+; CHECK:       vector.ph:
+; CHECK-NEXT:    [[N_MOD_VF:%.*]] = urem i32 [[N]], 4
+; CHECK-NEXT:    [[N_VEC:%.*]] = sub i32 [[N]], [[N_MOD_VF]]
+; CHECK-NEXT:    [[TMP0:%.*]] = mul i32 [[N_VEC]], [[X:%.*]]
+; CHECK-NEXT:    [[IND_END:%.*]] = add i32 3, [[TMP0]]
+; CHECK-NEXT:    [[DOTSPLATINSERT:%.*]] = insertelement <4 x i32> undef, i32 [[X]], i32 0
+; CHECK-NEXT:    [[DOTSPLAT:%.*]] = shufflevector <4 x i32> [[DOTSPLATINSERT]], <4 x i32> undef, <4 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP1:%.*]] = mul <4 x i32> <i32 0, i32 1, i32 2, i32 3>, [[DOTSPLAT]]
+; CHECK-NEXT:    [[INDUCTION:%.*]] = add <4 x i32> <i32 3, i32 3, i32 3, i32 3>, [[TMP1]]
+; CHECK-NEXT:    [[TMP2:%.*]] = mul i32 [[X]], 4
+; CHECK-NEXT:    [[DOTSPLATINSERT2:%.*]] = insertelement <4 x i32> undef, i32 [[TMP2]], i32 0
+; CHECK-NEXT:    [[DOTSPLAT3:%.*]] = shufflevector <4 x i32> [[DOTSPLATINSERT2]], <4 x i32> undef, <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:    [[INDEX:%.*]] = phi i32 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
 ; CHECK-NEXT:    [[VEC_IND:%.*]] = phi <4 x i32> [ <i32 0, i32 1, i32 2, i32 3>, [[VECTOR_PH]] ], [ [[VEC_IND_NEXT:%.*]], [[VECTOR_BODY]] ]
-; CHECK-NEXT:    [[VEC_IND4:%.*]] = phi <4 x i32> [ [[INDUCTION:%.*]], [[VECTOR_PH]] ], [ [[VEC_IND_NEXT5:%.*]], [[VECTOR_BODY]] ]
+; CHECK-NEXT:    [[VEC_IND4:%.*]] = phi <4 x i32> [ [[INDUCTION]], [[VECTOR_PH]] ], [ [[VEC_IND_NEXT5:%.*]], [[VECTOR_BODY]] ]
 ; CHECK-NEXT:    [[TMP3:%.*]] = add i32 [[INDEX]], 0
 ; CHECK-NEXT:    [[TMP4:%.*]] = add i32 [[INDEX]], 1
 ; CHECK-NEXT:    [[TMP5:%.*]] = add i32 [[INDEX]], 2
@@ -360,9 +581,32 @@ define void @test_stride_noninvar3_4i32(i32* readonly %data, i32* noalias nocapt
 ; CHECK-NEXT:    store <4 x i32> [[TMP10]], <4 x i32>* [[TMP13]], align 4
 ; CHECK-NEXT:    [[INDEX_NEXT]] = add i32 [[INDEX]], 4
 ; CHECK-NEXT:    [[VEC_IND_NEXT]] = add <4 x i32> [[VEC_IND]], <i32 4, i32 4, i32 4, i32 4>
-; CHECK-NEXT:    [[VEC_IND_NEXT5]] = add <4 x i32> [[VEC_IND4]], [[DOTSPLAT3:%.*]]
-; CHECK-NEXT:    [[TMP14:%.*]] = icmp eq i32 [[INDEX_NEXT]], [[N_VEC:%.*]]
+; CHECK-NEXT:    [[VEC_IND_NEXT5]] = add <4 x i32> [[VEC_IND4]], [[DOTSPLAT3]]
+; CHECK-NEXT:    [[TMP14:%.*]] = icmp eq i32 [[INDEX_NEXT]], [[N_VEC]]
 ; CHECK-NEXT:    br i1 [[TMP14]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], [[LOOP16:!llvm.loop !.*]]
+; CHECK:       middle.block:
+; CHECK-NEXT:    [[CMP_N:%.*]] = icmp eq i32 [[N]], [[N_VEC]]
+; CHECK-NEXT:    br i1 [[CMP_N]], label [[END:%.*]], label [[SCALAR_PH]]
+; CHECK:       scalar.ph:
+; CHECK-NEXT:    [[BC_RESUME_VAL:%.*]] = phi i32 [ [[N_VEC]], [[MIDDLE_BLOCK]] ], [ 0, [[ENTRY:%.*]] ]
+; CHECK-NEXT:    [[BC_RESUME_VAL1:%.*]] = phi i32 [ [[IND_END]], [[MIDDLE_BLOCK]] ], [ 3, [[ENTRY]] ]
+; CHECK-NEXT:    br label [[FOR_BODY:%.*]]
+; CHECK:       for.body:
+; CHECK-NEXT:    [[I_023:%.*]] = phi i32 [ [[INC:%.*]], [[FOR_BODY]] ], [ [[BC_RESUME_VAL]], [[SCALAR_PH]] ]
+; CHECK-NEXT:    [[STRIDE:%.*]] = phi i32 [ [[NEXT_STRIDE:%.*]], [[FOR_BODY]] ], [ [[BC_RESUME_VAL1]], [[SCALAR_PH]] ]
+; CHECK-NEXT:    [[MUL:%.*]] = mul nuw nsw i32 [[I_023]], [[STRIDE]]
+; CHECK-NEXT:    [[ADD5:%.*]] = add nuw nsw i32 [[MUL]], 2
+; CHECK-NEXT:    [[ARRAYIDX6:%.*]] = getelementptr inbounds i32, i32* [[DATA]], i32 [[ADD5]]
+; CHECK-NEXT:    [[TMP15:%.*]] = load i32, i32* [[ARRAYIDX6]], align 4
+; CHECK-NEXT:    [[ADD7:%.*]] = add nsw i32 5, [[TMP15]]
+; CHECK-NEXT:    [[ARRAYIDX9:%.*]] = getelementptr inbounds i32, i32* [[DST]], i32 [[I_023]]
+; CHECK-NEXT:    store i32 [[ADD7]], i32* [[ARRAYIDX9]], align 4
+; CHECK-NEXT:    [[INC]] = add nuw nsw i32 [[I_023]], 1
+; CHECK-NEXT:    [[NEXT_STRIDE]] = add nuw nsw i32 [[STRIDE]], [[X]]
+; CHECK-NEXT:    [[EXITCOND_NOT:%.*]] = icmp eq i32 [[INC]], [[N]]
+; CHECK-NEXT:    br i1 [[EXITCOND_NOT]], label [[END]], label [[FOR_BODY]], [[LOOP17:!llvm.loop !.*]]
+; CHECK:       end:
+; CHECK-NEXT:    ret void
 ;
 entry:
   br label %for.body
diff --git a/llvm/test/Transforms/LoopVectorize/ARM/mve-reduction-predselect.ll b/llvm/test/Transforms/LoopVectorize/ARM/mve-reduction-predselect.ll
index cf7bf00..b7dc26e 100644
--- a/llvm/test/Transforms/LoopVectorize/ARM/mve-reduction-predselect.ll
+++ b/llvm/test/Transforms/LoopVectorize/ARM/mve-reduction-predselect.ll
@@ -14,21 +14,21 @@ define i32 @reduction_sum_single(i32* noalias nocapture %A) {
 ; CHECK-NEXT:    [[INDEX:%.*]] = phi i32 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
 ; CHECK-NEXT:    [[VEC_PHI:%.*]] = phi <4 x i32> [ zeroinitializer, [[VECTOR_PH]] ], [ [[TMP3:%.*]], [[VECTOR_BODY]] ]
 ; CHECK-NEXT:    [[TMP0:%.*]] = getelementptr inbounds i32, i32* [[A:%.*]], i32 [[INDEX]]
-; CHECK-NEXT:    [[ACTIVE_LANE_MASK:%.*]] = call <4 x i1> @llvm.get.active.lane.mask.v4i1.i32(i32 [[INDEX]], i32 256)
+; CHECK-NEXT:    [[ACTIVE_LANE_MASK:%.*]] = call <4 x i1> @llvm.get.active.lane.mask.v4i1.i32(i32 [[INDEX]], i32 257)
 ; CHECK-NEXT:    [[TMP1:%.*]] = bitcast i32* [[TMP0]] to <4 x i32>*
 ; CHECK-NEXT:    [[WIDE_MASKED_LOAD:%.*]] = call <4 x i32> @llvm.masked.load.v4i32.p0v4i32(<4 x i32>* [[TMP1]], i32 4, <4 x i1> [[ACTIVE_LANE_MASK]], <4 x i32> undef)
 ; CHECK-NEXT:    [[TMP2:%.*]] = select <4 x i1> [[ACTIVE_LANE_MASK]], <4 x i32> [[WIDE_MASKED_LOAD]], <4 x i32> zeroinitializer
 ; CHECK-NEXT:    [[TMP3]] = add <4 x i32> [[VEC_PHI]], [[TMP2]]
 ; CHECK-NEXT:    [[INDEX_NEXT]] = add i32 [[INDEX]], 4
 ; CHECK-NEXT:    [[TMP4:%.*]] = icmp eq i32 [[INDEX_NEXT]], 260
-; CHECK-NEXT:    br i1 [[TMP4]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop !0
+; CHECK-NEXT:    br i1 [[TMP4]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], [[LOOP0:!llvm.loop !.*]]
 ; CHECK:       middle.block:
 ; CHECK-NEXT:    [[TMP5:%.*]] = call i32 @llvm.experimental.vector.reduce.add.v4i32(<4 x i32> [[TMP3]])
 ; CHECK-NEXT:    br i1 true, label [[DOT_CRIT_EDGE:%.*]], label [[SCALAR_PH]]
 ; CHECK:       scalar.ph:
 ; CHECK-NEXT:    br label [[DOTLR_PH:%.*]]
 ; CHECK:       .lr.ph:
-; CHECK-NEXT:    br i1 undef, label [[DOT_CRIT_EDGE]], label [[DOTLR_PH]], !llvm.loop !2
+; CHECK-NEXT:    br i1 undef, label [[DOT_CRIT_EDGE]], label [[DOTLR_PH]], [[LOOP2:!llvm.loop !.*]]
 ; CHECK:       ._crit_edge:
 ; CHECK-NEXT:    [[SUM_0_LCSSA:%.*]] = phi i32 [ undef, [[DOTLR_PH]] ], [ [[TMP5]], [[MIDDLE_BLOCK]] ]
 ; CHECK-NEXT:    ret i32 [[SUM_0_LCSSA]]
@@ -62,7 +62,7 @@ define i32 @reduction_sum(i32* noalias nocapture %A, i32* noalias nocapture %B)
 ; CHECK-NEXT:    [[VEC_IND:%.*]] = phi <4 x i32> [ <i32 0, i32 1, i32 2, i32 3>, [[VECTOR_PH]] ], [ [[VEC_IND_NEXT:%.*]], [[VECTOR_BODY]] ]
 ; CHECK-NEXT:    [[VEC_PHI:%.*]] = phi <4 x i32> [ zeroinitializer, [[VECTOR_PH]] ], [ [[TMP7:%.*]], [[VECTOR_BODY]] ]
 ; CHECK-NEXT:    [[TMP0:%.*]] = getelementptr inbounds i32, i32* [[A:%.*]], i32 [[INDEX]]
-; CHECK-NEXT:    [[ACTIVE_LANE_MASK:%.*]] = call <4 x i1> @llvm.get.active.lane.mask.v4i1.i32(i32 [[INDEX]], i32 256)
+; CHECK-NEXT:    [[ACTIVE_LANE_MASK:%.*]] = call <4 x i1> @llvm.get.active.lane.mask.v4i1.i32(i32 [[INDEX]], i32 257)
 ; CHECK-NEXT:    [[TMP1:%.*]] = bitcast i32* [[TMP0]] to <4 x i32>*
 ; CHECK-NEXT:    [[WIDE_MASKED_LOAD:%.*]] = call <4 x i32> @llvm.masked.load.v4i32.p0v4i32(<4 x i32>* [[TMP1]], i32 4, <4 x i1> [[ACTIVE_LANE_MASK]], <4 x i32> undef)
 ; CHECK-NEXT:    [[TMP2:%.*]] = getelementptr inbounds i32, i32* [[B:%.*]], i32 [[INDEX]]
@@ -75,14 +75,14 @@ define i32 @reduction_sum(i32* noalias nocapture %A, i32* noalias nocapture %B)
 ; CHECK-NEXT:    [[INDEX_NEXT]] = add i32 [[INDEX]], 4
 ; CHECK-NEXT:    [[VEC_IND_NEXT]] = add <4 x i32> [[VEC_IND]], <i32 4, i32 4, i32 4, i32 4>
 ; CHECK-NEXT:    [[TMP8:%.*]] = icmp eq i32 [[INDEX_NEXT]], 260
-; CHECK-NEXT:    br i1 [[TMP8]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop !4
+; CHECK-NEXT:    br i1 [[TMP8]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], [[LOOP4:!llvm.loop !.*]]
 ; CHECK:       middle.block:
 ; CHECK-NEXT:    [[TMP9:%.*]] = call i32 @llvm.experimental.vector.reduce.add.v4i32(<4 x i32> [[TMP7]])
 ; CHECK-NEXT:    br i1 true, label [[DOT_CRIT_EDGE:%.*]], label [[SCALAR_PH]]
 ; CHECK:       scalar.ph:
 ; CHECK-NEXT:    br label [[DOTLR_PH:%.*]]
 ; CHECK:       .lr.ph:
-; CHECK-NEXT:    br i1 undef, label [[DOT_CRIT_EDGE]], label [[DOTLR_PH]], !llvm.loop !5
+; CHECK-NEXT:    br i1 undef, label [[DOT_CRIT_EDGE]], label [[DOTLR_PH]], [[LOOP5:!llvm.loop !.*]]
 ; CHECK:       ._crit_edge:
 ; CHECK-NEXT:    [[SUM_0_LCSSA:%.*]] = phi i32 [ undef, [[DOTLR_PH]] ], [ [[TMP9]], [[MIDDLE_BLOCK]] ]
 ; CHECK-NEXT:    ret i32 [[SUM_0_LCSSA]]
@@ -119,7 +119,7 @@ define i32 @reduction_prod(i32* noalias nocapture %A, i32* noalias nocapture %B)
 ; CHECK-NEXT:    [[INDEX:%.*]] = phi i32 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
 ; CHECK-NEXT:    [[VEC_PHI:%.*]] = phi <4 x i32> [ <i32 1, i32 1, i32 1, i32 1>, [[VECTOR_PH]] ], [ [[TMP6:%.*]], [[VECTOR_BODY]] ]
 ; CHECK-NEXT:    [[TMP0:%.*]] = getelementptr inbounds i32, i32* [[A:%.*]], i32 [[INDEX]]
-; CHECK-NEXT:    [[ACTIVE_LANE_MASK:%.*]] = call <4 x i1> @llvm.get.active.lane.mask.v4i1.i32(i32 [[INDEX]], i32 256)
+; CHECK-NEXT:    [[ACTIVE_LANE_MASK:%.*]] = call <4 x i1> @llvm.get.active.lane.mask.v4i1.i32(i32 [[INDEX]], i32 257)
 ; CHECK-NEXT:    [[TMP1:%.*]] = bitcast i32* [[TMP0]] to <4 x i32>*
 ; CHECK-NEXT:    [[WIDE_MASKED_LOAD:%.*]] = call <4 x i32> @llvm.masked.load.v4i32.p0v4i32(<4 x i32>* [[TMP1]], i32 4, <4 x i1> [[ACTIVE_LANE_MASK]], <4 x i32> undef)
 ; CHECK-NEXT:    [[TMP2:%.*]] = getelementptr inbounds i32, i32* [[B:%.*]], i32 [[INDEX]]
@@ -130,14 +130,14 @@ define i32 @reduction_prod(i32* noalias nocapture %A, i32* noalias nocapture %B)
 ; CHECK-NEXT:    [[TMP6]] = select <4 x i1> [[ACTIVE_LANE_MASK]], <4 x i32> [[TMP5]], <4 x i32> [[VEC_PHI]]
 ; CHECK-NEXT:    [[INDEX_NEXT]] = add i32 [[INDEX]], 4
 ; CHECK-NEXT:    [[TMP7:%.*]] = icmp eq i32 [[INDEX_NEXT]], 260
-; CHECK-NEXT:    br i1 [[TMP7]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop !6
+; CHECK-NEXT:    br i1 [[TMP7]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], [[LOOP6:!llvm.loop !.*]]
 ; CHECK:       middle.block:
 ; CHECK-NEXT:    [[TMP8:%.*]] = call i32 @llvm.experimental.vector.reduce.mul.v4i32(<4 x i32> [[TMP6]])
 ; CHECK-NEXT:    br i1 true, label [[DOT_CRIT_EDGE:%.*]], label [[SCALAR_PH]]
 ; CHECK:       scalar.ph:
 ; CHECK-NEXT:    br label [[DOTLR_PH:%.*]]
 ; CHECK:       .lr.ph:
-; CHECK-NEXT:    br i1 undef, label [[DOT_CRIT_EDGE]], label [[DOTLR_PH]], !llvm.loop !7
+; CHECK-NEXT:    br i1 undef, label [[DOT_CRIT_EDGE]], label [[DOTLR_PH]], [[LOOP7:!llvm.loop !.*]]
 ; CHECK:       ._crit_edge:
 ; CHECK-NEXT:    [[PROD_0_LCSSA:%.*]] = phi i32 [ undef, [[DOTLR_PH]] ], [ [[TMP8]], [[MIDDLE_BLOCK]] ]
 ; CHECK-NEXT:    ret i32 [[PROD_0_LCSSA]]
@@ -173,7 +173,7 @@ define i32 @reduction_and(i32* nocapture %A, i32* nocapture %B) {
 ; CHECK-NEXT:    [[INDEX:%.*]] = phi i32 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
 ; CHECK-NEXT:    [[VEC_PHI:%.*]] = phi <4 x i32> [ <i32 -1, i32 -1, i32 -1, i32 -1>, [[VECTOR_PH]] ], [ [[TMP6:%.*]], [[VECTOR_BODY]] ]
 ; CHECK-NEXT:    [[TMP0:%.*]] = getelementptr inbounds i32, i32* [[A:%.*]], i32 [[INDEX]]
-; CHECK-NEXT:    [[ACTIVE_LANE_MASK:%.*]] = call <4 x i1> @llvm.get.active.lane.mask.v4i1.i32(i32 [[INDEX]], i32 256)
+; CHECK-NEXT:    [[ACTIVE_LANE_MASK:%.*]] = call <4 x i1> @llvm.get.active.lane.mask.v4i1.i32(i32 [[INDEX]], i32 257)
 ; CHECK-NEXT:    [[TMP1:%.*]] = bitcast i32* [[TMP0]] to <4 x i32>*
 ; CHECK-NEXT:    [[WIDE_MASKED_LOAD:%.*]] = call <4 x i32> @llvm.masked.load.v4i32.p0v4i32(<4 x i32>* [[TMP1]], i32 4, <4 x i1> [[ACTIVE_LANE_MASK]], <4 x i32> undef)
 ; CHECK-NEXT:    [[TMP2:%.*]] = getelementptr inbounds i32, i32* [[B:%.*]], i32 [[INDEX]]
@@ -184,14 +184,14 @@ define i32 @reduction_and(i32* nocapture %A, i32* nocapture %B) {
 ; CHECK-NEXT:    [[TMP6]] = select <4 x i1> [[ACTIVE_LANE_MASK]], <4 x i32> [[TMP5]], <4 x i32> [[VEC_PHI]]
 ; CHECK-NEXT:    [[INDEX_NEXT]] = add i32 [[INDEX]], 4
 ; CHECK-NEXT:    [[TMP7:%.*]] = icmp eq i32 [[INDEX_NEXT]], 260
-; CHECK-NEXT:    br i1 [[TMP7]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop !8
+; CHECK-NEXT:    br i1 [[TMP7]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], [[LOOP8:!llvm.loop !.*]]
 ; CHECK:       middle.block:
 ; CHECK-NEXT:    [[TMP8:%.*]] = call i32 @llvm.experimental.vector.reduce.and.v4i32(<4 x i32> [[TMP6]])
 ; CHECK-NEXT:    br i1 true, label [[FOR_END:%.*]], label [[SCALAR_PH]]
 ; CHECK:       scalar.ph:
 ; CHECK-NEXT:    br label [[FOR_BODY:%.*]]
 ; CHECK:       for.body:
-; CHECK-NEXT:    br i1 undef, label [[FOR_END]], label [[FOR_BODY]], !llvm.loop !9
+; CHECK-NEXT:    br i1 undef, label [[FOR_END]], label [[FOR_BODY]], [[LOOP9:!llvm.loop !.*]]
 ; CHECK:       for.end:
 ; CHECK-NEXT:    [[RESULT_0_LCSSA:%.*]] = phi i32 [ undef, [[FOR_BODY]] ], [ [[TMP8]], [[MIDDLE_BLOCK]] ]
 ; CHECK-NEXT:    ret i32 [[RESULT_0_LCSSA]]
@@ -227,7 +227,7 @@ define i32 @reduction_or(i32* nocapture %A, i32* nocapture %B) {
 ; CHECK-NEXT:    [[INDEX:%.*]] = phi i32 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
 ; CHECK-NEXT:    [[VEC_PHI:%.*]] = phi <4 x i32> [ zeroinitializer, [[VECTOR_PH]] ], [ [[TMP6:%.*]], [[VECTOR_BODY]] ]
 ; CHECK-NEXT:    [[TMP0:%.*]] = getelementptr inbounds i32, i32* [[A:%.*]], i32 [[INDEX]]
-; CHECK-NEXT:    [[ACTIVE_LANE_MASK:%.*]] = call <4 x i1> @llvm.get.active.lane.mask.v4i1.i32(i32 [[INDEX]], i32 256)
+; CHECK-NEXT:    [[ACTIVE_LANE_MASK:%.*]] = call <4 x i1> @llvm.get.active.lane.mask.v4i1.i32(i32 [[INDEX]], i32 257)
 ; CHECK-NEXT:    [[TMP1:%.*]] = bitcast i32* [[TMP0]] to <4 x i32>*
 ; CHECK-NEXT:    [[WIDE_MASKED_LOAD:%.*]] = call <4 x i32> @llvm.masked.load.v4i32.p0v4i32(<4 x i32>* [[TMP1]], i32 4, <4 x i1> [[ACTIVE_LANE_MASK]], <4 x i32> undef)
 ; CHECK-NEXT:    [[TMP2:%.*]] = getelementptr inbounds i32, i32* [[B:%.*]], i32 [[INDEX]]
@@ -238,14 +238,14 @@ define i32 @reduction_or(i32* nocapture %A, i32* nocapture %B) {
 ; CHECK-NEXT:    [[TMP6]] = or <4 x i32> [[VEC_PHI]], [[TMP5]]
 ; CHECK-NEXT:    [[INDEX_NEXT]] = add i32 [[INDEX]], 4
 ; CHECK-NEXT:    [[TMP7:%.*]] = icmp eq i32 [[INDEX_NEXT]], 260
-; CHECK-NEXT:    br i1 [[TMP7]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop !10
+; CHECK-NEXT:    br i1 [[TMP7]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], [[LOOP10:!llvm.loop !.*]]
 ; CHECK:       middle.block:
 ; CHECK-NEXT:    [[TMP8:%.*]] = call i32 @llvm.experimental.vector.reduce.or.v4i32(<4 x i32> [[TMP6]])
 ; CHECK-NEXT:    br i1 true, label [[FOR_END:%.*]], label [[SCALAR_PH]]
 ; CHECK:       scalar.ph:
 ; CHECK-NEXT:    br label [[FOR_BODY:%.*]]
 ; CHECK:       for.body:
-; CHECK-NEXT:    br i1 undef, label [[FOR_END]], label [[FOR_BODY]], !llvm.loop !11
+; CHECK-NEXT:    br i1 undef, label [[FOR_END]], label [[FOR_BODY]], [[LOOP11:!llvm.loop !.*]]
 ; CHECK:       for.end:
 ; CHECK-NEXT:    [[RESULT_0_LCSSA:%.*]] = phi i32 [ undef, [[FOR_BODY]] ], [ [[TMP8]], [[MIDDLE_BLOCK]] ]
 ; CHECK-NEXT:    ret i32 [[RESULT_0_LCSSA]]
@@ -281,7 +281,7 @@ define i32 @reduction_xor(i32* nocapture %A, i32* nocapture %B) {
 ; CHECK-NEXT:    [[INDEX:%.*]] = phi i32 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
 ; CHECK-NEXT:    [[VEC_PHI:%.*]] = phi <4 x i32> [ zeroinitializer, [[VECTOR_PH]] ], [ [[TMP6:%.*]], [[VECTOR_BODY]] ]
 ; CHECK-NEXT:    [[TMP0:%.*]] = getelementptr inbounds i32, i32* [[A:%.*]], i32 [[INDEX]]
-; CHECK-NEXT:    [[ACTIVE_LANE_MASK:%.*]] = call <4 x i1> @llvm.get.active.lane.mask.v4i1.i32(i32 [[INDEX]], i32 256)
+; CHECK-NEXT:    [[ACTIVE_LANE_MASK:%.*]] = call <4 x i1> @llvm.get.active.lane.mask.v4i1.i32(i32 [[INDEX]], i32 257)
 ; CHECK-NEXT:    [[TMP1:%.*]] = bitcast i32* [[TMP0]] to <4 x i32>*
 ; CHECK-NEXT:    [[WIDE_MASKED_LOAD:%.*]] = call <4 x i32> @llvm.masked.load.v4i32.p0v4i32(<4 x i32>* [[TMP1]], i32 4, <4 x i1> [[ACTIVE_LANE_MASK]], <4 x i32> undef)
 ; CHECK-NEXT:    [[TMP2:%.*]] = getelementptr inbounds i32, i32* [[B:%.*]], i32 [[INDEX]]
@@ -292,14 +292,14 @@ define i32 @reduction_xor(i32* nocapture %A, i32* nocapture %B) {
 ; CHECK-NEXT:    [[TMP6]] = xor <4 x i32> [[VEC_PHI]], [[TMP5]]
 ; CHECK-NEXT:    [[INDEX_NEXT]] = add i32 [[INDEX]], 4
 ; CHECK-NEXT:    [[TMP7:%.*]] = icmp eq i32 [[INDEX_NEXT]], 260
-; CHECK-NEXT:    br i1 [[TMP7]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop !12
+; CHECK-NEXT:    br i1 [[TMP7]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], [[LOOP12:!llvm.loop !.*]]
 ; CHECK:       middle.block:
 ; CHECK-NEXT:    [[TMP8:%.*]] = call i32 @llvm.experimental.vector.reduce.xor.v4i32(<4 x i32> [[TMP6]])
 ; CHECK-NEXT:    br i1 true, label [[FOR_END:%.*]], label [[SCALAR_PH]]
 ; CHECK:       scalar.ph:
 ; CHECK-NEXT:    br label [[FOR_BODY:%.*]]
 ; CHECK:       for.body:
-; CHECK-NEXT:    br i1 undef, label [[FOR_END]], label [[FOR_BODY]], !llvm.loop !13
+; CHECK-NEXT:    br i1 undef, label [[FOR_END]], label [[FOR_BODY]], [[LOOP13:!llvm.loop !.*]]
 ; CHECK:       for.end:
 ; CHECK-NEXT:    [[RESULT_0_LCSSA:%.*]] = phi i32 [ undef, [[FOR_BODY]] ], [ [[TMP8]], [[MIDDLE_BLOCK]] ]
 ; CHECK-NEXT:    ret i32 [[RESULT_0_LCSSA]]
@@ -335,7 +335,7 @@ define float @reduction_fadd(float* nocapture %A, float* nocapture %B) {
 ; CHECK-NEXT:    [[INDEX:%.*]] = phi i32 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
 ; CHECK-NEXT:    [[VEC_PHI:%.*]] = phi <4 x float> [ zeroinitializer, [[VECTOR_PH]] ], [ [[TMP6:%.*]], [[VECTOR_BODY]] ]
 ; CHECK-NEXT:    [[TMP0:%.*]] = getelementptr inbounds float, float* [[A:%.*]], i32 [[INDEX]]
-; CHECK-NEXT:    [[ACTIVE_LANE_MASK:%.*]] = call <4 x i1> @llvm.get.active.lane.mask.v4i1.i32(i32 [[INDEX]], i32 256)
+; CHECK-NEXT:    [[ACTIVE_LANE_MASK:%.*]] = call <4 x i1> @llvm.get.active.lane.mask.v4i1.i32(i32 [[INDEX]], i32 257)
 ; CHECK-NEXT:    [[TMP1:%.*]] = bitcast float* [[TMP0]] to <4 x float>*
 ; CHECK-NEXT:    [[WIDE_MASKED_LOAD:%.*]] = call <4 x float> @llvm.masked.load.v4f32.p0v4f32(<4 x float>* [[TMP1]], i32 4, <4 x i1> [[ACTIVE_LANE_MASK]], <4 x float> undef)
 ; CHECK-NEXT:    [[TMP2:%.*]] = getelementptr inbounds float, float* [[B:%.*]], i32 [[INDEX]]
@@ -346,14 +346,14 @@ define float @reduction_fadd(float* nocapture %A, float* nocapture %B) {
 ; CHECK-NEXT:    [[TMP6]] = select <4 x i1> [[ACTIVE_LANE_MASK]], <4 x float> [[TMP5]], <4 x float> [[VEC_PHI]]
 ; CHECK-NEXT:    [[INDEX_NEXT]] = add i32 [[INDEX]], 4
 ; CHECK-NEXT:    [[TMP7:%.*]] = icmp eq i32 [[INDEX_NEXT]], 260
-; CHECK-NEXT:    br i1 [[TMP7]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop !14
+; CHECK-NEXT:    br i1 [[TMP7]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], [[LOOP14:!llvm.loop !.*]]
 ; CHECK:       middle.block:
 ; CHECK-NEXT:    [[TMP8:%.*]] = call fast float @llvm.experimental.vector.reduce.v2.fadd.f32.v4f32(float 0.000000e+00, <4 x float> [[TMP6]])
 ; CHECK-NEXT:    br i1 true, label [[FOR_END:%.*]], label [[SCALAR_PH]]
 ; CHECK:       scalar.ph:
 ; CHECK-NEXT:    br label [[FOR_BODY:%.*]]
 ; CHECK:       for.body:
-; CHECK-NEXT:    br i1 undef, label [[FOR_END]], label [[FOR_BODY]], !llvm.loop !15
+; CHECK-NEXT:    br i1 undef, label [[FOR_END]], label [[FOR_BODY]], [[LOOP15:!llvm.loop !.*]]
 ; CHECK:       for.end:
 ; CHECK-NEXT:    [[RESULT_0_LCSSA:%.*]] = phi float [ undef, [[FOR_BODY]] ], [ [[TMP8]], [[MIDDLE_BLOCK]] ]
 ; CHECK-NEXT:    ret float [[RESULT_0_LCSSA]]
@@ -389,7 +389,7 @@ define float @reduction_fmul(float* nocapture %A, float* nocapture %B) {
 ; CHECK-NEXT:    [[INDEX:%.*]] = phi i32 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
 ; CHECK-NEXT:    [[VEC_PHI:%.*]] = phi <4 x float> [ <float 0.000000e+00, float 1.000000e+00, float 1.000000e+00, float 1.000000e+00>, [[VECTOR_PH]] ], [ [[TMP6:%.*]], [[VECTOR_BODY]] ]
 ; CHECK-NEXT:    [[TMP0:%.*]] = getelementptr inbounds float, float* [[A:%.*]], i32 [[INDEX]]
-; CHECK-NEXT:    [[ACTIVE_LANE_MASK:%.*]] = call <4 x i1> @llvm.get.active.lane.mask.v4i1.i32(i32 [[INDEX]], i32 256)
+; CHECK-NEXT:    [[ACTIVE_LANE_MASK:%.*]] = call <4 x i1> @llvm.get.active.lane.mask.v4i1.i32(i32 [[INDEX]], i32 257)
 ; CHECK-NEXT:    [[TMP1:%.*]] = bitcast float* [[TMP0]] to <4 x float>*
 ; CHECK-NEXT:    [[WIDE_MASKED_LOAD:%.*]] = call <4 x float> @llvm.masked.load.v4f32.p0v4f32(<4 x float>* [[TMP1]], i32 4, <4 x i1> [[ACTIVE_LANE_MASK]], <4 x float> undef)
 ; CHECK-NEXT:    [[TMP2:%.*]] = getelementptr inbounds float, float* [[B:%.*]], i32 [[INDEX]]
@@ -400,14 +400,14 @@ define float @reduction_fmul(float* nocapture %A, float* nocapture %B) {
 ; CHECK-NEXT:    [[TMP6]] = select <4 x i1> [[ACTIVE_LANE_MASK]], <4 x float> [[TMP5]], <4 x float> [[VEC_PHI]]
 ; CHECK-NEXT:    [[INDEX_NEXT]] = add i32 [[INDEX]], 4
 ; CHECK-NEXT:    [[TMP7:%.*]] = icmp eq i32 [[INDEX_NEXT]], 260
-; CHECK-NEXT:    br i1 [[TMP7]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop !16
+; CHECK-NEXT:    br i1 [[TMP7]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], [[LOOP16:!llvm.loop !.*]]
 ; CHECK:       middle.block:
 ; CHECK-NEXT:    [[TMP8:%.*]] = call fast float @llvm.experimental.vector.reduce.v2.fmul.f32.v4f32(float 1.000000e+00, <4 x float> [[TMP6]])
 ; CHECK-NEXT:    br i1 true, label [[FOR_END:%.*]], label [[SCALAR_PH]]
 ; CHECK:       scalar.ph:
 ; CHECK-NEXT:    br label [[FOR_BODY:%.*]]
 ; CHECK:       for.body:
-; CHECK-NEXT:    br i1 undef, label [[FOR_END]], label [[FOR_BODY]], !llvm.loop !17
+; CHECK-NEXT:    br i1 undef, label [[FOR_END]], label [[FOR_BODY]], [[LOOP17:!llvm.loop !.*]]
 ; CHECK:       for.end:
 ; CHECK-NEXT:    [[RESULT_0_LCSSA:%.*]] = phi float [ undef, [[FOR_BODY]] ], [ [[TMP8]], [[MIDDLE_BLOCK]] ]
 ; CHECK-NEXT:    ret float [[RESULT_0_LCSSA]]
@@ -449,7 +449,7 @@ define i32 @reduction_min(i32* nocapture %A, i32* nocapture %B) {
 ; CHECK-NEXT:    [[TMP3]] = select <4 x i1> [[TMP2]], <4 x i32> [[VEC_PHI]], <4 x i32> [[WIDE_LOAD]]
 ; CHECK-NEXT:    [[INDEX_NEXT]] = add i32 [[INDEX]], 4
 ; CHECK-NEXT:    [[TMP4:%.*]] = icmp eq i32 [[INDEX_NEXT]], 256
-; CHECK-NEXT:    br i1 [[TMP4]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop !18
+; CHECK-NEXT:    br i1 [[TMP4]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], [[LOOP18:!llvm.loop !.*]]
 ; CHECK:       middle.block:
 ; CHECK-NEXT:    [[TMP5:%.*]] = call i32 @llvm.experimental.vector.reduce.smin.v4i32(<4 x i32> [[TMP3]])
 ; CHECK-NEXT:    br i1 false, label [[FOR_END:%.*]], label [[SCALAR_PH]]
@@ -466,7 +466,7 @@ define i32 @reduction_min(i32* nocapture %A, i32* nocapture %B) {
 ; CHECK-NEXT:    [[V0]] = select i1 [[C0]], i32 [[RESULT_08]], i32 [[L0]]
 ; CHECK-NEXT:    [[INDVARS_IV_NEXT]] = add i32 [[INDVARS_IV]], 1
 ; CHECK-NEXT:    [[EXITCOND:%.*]] = icmp eq i32 [[INDVARS_IV_NEXT]], 257
-; CHECK-NEXT:    br i1 [[EXITCOND]], label [[FOR_END]], label [[FOR_BODY]], !llvm.loop !19
+; CHECK-NEXT:    br i1 [[EXITCOND]], label [[FOR_END]], label [[FOR_BODY]], [[LOOP19:!llvm.loop !.*]]
 ; CHECK:       for.end:
 ; CHECK-NEXT:    [[RESULT_0_LCSSA:%.*]] = phi i32 [ [[V0]], [[FOR_BODY]] ], [ [[TMP5]], [[MIDDLE_BLOCK]] ]
 ; CHECK-NEXT:    ret i32 [[RESULT_0_LCSSA]]
@@ -506,7 +506,7 @@ define i32 @reduction_max(i32* nocapture %A, i32* nocapture %B) {
 ; CHECK-NEXT:    [[TMP3]] = select <4 x i1> [[TMP2]], <4 x i32> [[VEC_PHI]], <4 x i32> [[WIDE_LOAD]]
 ; CHECK-NEXT:    [[INDEX_NEXT]] = add i32 [[INDEX]], 4
 ; CHECK-NEXT:    [[TMP4:%.*]] = icmp eq i32 [[INDEX_NEXT]], 256
-; CHECK-NEXT:    br i1 [[TMP4]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop !20
+; CHECK-NEXT:    br i1 [[TMP4]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], [[LOOP20:!llvm.loop !.*]]
 ; CHECK:       middle.block:
 ; CHECK-NEXT:    [[TMP5:%.*]] = call i32 @llvm.experimental.vector.reduce.umax.v4i32(<4 x i32> [[TMP3]])
 ; CHECK-NEXT:    br i1 false, label [[FOR_END:%.*]], label [[SCALAR_PH]]
@@ -523,7 +523,7 @@ define i32 @reduction_max(i32* nocapture %A, i32* nocapture %B) {
 ; CHECK-NEXT:    [[V0]] = select i1 [[C0]], i32 [[RESULT_08]], i32 [[L0]]
 ; CHECK-NEXT:    [[INDVARS_IV_NEXT]] = add i32 [[INDVARS_IV]], 1
 ; CHECK-NEXT:    [[EXITCOND:%.*]] = icmp eq i32 [[INDVARS_IV_NEXT]], 257
-; CHECK-NEXT:    br i1 [[EXITCOND]], label [[FOR_END]], label [[FOR_BODY]], !llvm.loop !21
+; CHECK-NEXT:    br i1 [[EXITCOND]], label [[FOR_END]], label [[FOR_BODY]], [[LOOP21:!llvm.loop !.*]]
 ; CHECK:       for.end:
 ; CHECK-NEXT:    [[RESULT_0_LCSSA:%.*]] = phi i32 [ [[V0]], [[FOR_BODY]] ], [ [[TMP5]], [[MIDDLE_BLOCK]] ]
 ; CHECK-NEXT:    ret i32 [[RESULT_0_LCSSA]]
diff --git a/llvm/test/Transforms/LoopVectorize/ARM/prefer-tail-loop-folding.ll b/llvm/test/Transforms/LoopVectorize/ARM/prefer-tail-loop-folding.ll
index 9473d4e..f4961e4 100644
--- a/llvm/test/Transforms/LoopVectorize/ARM/prefer-tail-loop-folding.ll
+++ b/llvm/test/Transforms/LoopVectorize/ARM/prefer-tail-loop-folding.ll
@@ -47,7 +47,7 @@ define void @prefer_folding(i32* noalias nocapture %A, i32* noalias nocapture re
 ; PREFER-FOLDING: vector.body:
 ; PREFER-FOLDING: %index = phi i32 [ 0, %vector.ph ], [ %index.next, %vector.body ]
 ; PREFER-FOLDING: %[[VIVELEM0:.*]] = add i32 %index, 0
-; PREFER-FOLDING: %active.lane.mask = call <4 x i1> @llvm.get.active.lane.mask.v4i1.i32(i32 %[[VIVELEM0]], i32 430)
+; PREFER-FOLDING: %active.lane.mask = call <4 x i1> @llvm.get.active.lane.mask.v4i1.i32(i32 %[[VIVELEM0]], i32 431)
 ; PREFER-FOLDING: call <4 x i32> @llvm.masked.load.v4i32.p0v4i32({{.*}}, <4 x i1> %active.lane.mask,
 ; PREFER-FOLDING: call <4 x i32> @llvm.masked.load.v4i32.p0v4i32({{.*}}, <4 x i1> %active.lane.mask,
 ; PREFER-FOLDING: call void @llvm.masked.store.v4i32.p0v4i32({{.*}}, <4 x i1> %active.lane.mask
@@ -331,7 +331,7 @@ define void @float(float* noalias nocapture %A, float* noalias nocapture readonl
 ; PREFER-FOLDING: vector.body:
 ; PREFER-FOLDING: %index = phi i32 [ 0, %vector.ph ], [ %index.next, %vector.body ]
 ; PREFER-FOLDING: %[[VIVELEM0:.*]] = add i32 %index, 0
-; PREFER-FOLDING: %active.lane.mask = call <4 x i1> @llvm.get.active.lane.mask.v4i1.i32(i32 %[[VIVELEM0]], i32 430)
+; PREFER-FOLDING: %active.lane.mask = call <4 x i1> @llvm.get.active.lane.mask.v4i1.i32(i32 %[[VIVELEM0]], i32 431)
 ; PREFER-FOLDING: call <4 x float> @llvm.masked.load.v4f32.p0v4f32({{.*}}%active.lane.mask
 ; PREFER-FOLDING: call <4 x float> @llvm.masked.load.v4f32.p0v4f32({{.*}}%active.lane.mask
 ; PREFER-FOLDING: call void @llvm.masked.store.v4f32.p0v4f32({{.*}}%active.lane.mask
diff --git a/llvm/test/Transforms/LoopVectorize/ARM/tail-folding-counting-down.ll b/llvm/test/Transforms/LoopVectorize/ARM/tail-folding-counting-down.ll
index 8a327f1..493c540 100644
--- a/llvm/test/Transforms/LoopVectorize/ARM/tail-folding-counting-down.ll
+++ b/llvm/test/Transforms/LoopVectorize/ARM/tail-folding-counting-down.ll
@@ -17,8 +17,7 @@ define dso_local void @sgt_loopguard(i8* noalias nocapture readonly %a, i8* noal
 ; COMMON:       vector.body:
 
 ; CHECK-TF:     %[[VIVELEM0:.*]] = extractelement <16 x i32> %vec.iv, i32 0
-; CHECK-TF:     %[[SCALARBTC:.*]] = extractelement <16 x i32> %broadcast.splat, i32 0
-; CHECK-TF:     %active.lane.mask = call <16 x i1> @llvm.get.active.lane.mask.v16i1.i32(i32 %[[VIVELEM0]], i32 %[[SCALARBTC]])
+; CHECK-TF:     %active.lane.mask = call <16 x i1> @llvm.get.active.lane.mask.v16i1.i32(i32 %[[VIVELEM0]], i32 %N)
 ; CHECK-TF:     llvm.masked.load.v16i8.p0v16i8(<16 x i8>* %{{.*}}, i32 1, <16 x i1> %active.lane.mask
 ; CHECK-TF:     llvm.masked.load.v16i8.p0v16i8(<16 x i8>* %{{.*}}, i32 1, <16 x i1> %active.lane.mask
 ; CHECK-TF:     llvm.masked.store.v16i8.p0v16i8(<16 x i8> %{{.*}}, <16 x i8>* %{{.*}}, i32 1, <16 x i1> %active.lane.mask)
diff --git a/llvm/test/Transforms/LoopVectorize/ARM/tail-folding-loop-hint.ll b/llvm/test/Transforms/LoopVectorize/ARM/tail-folding-loop-hint.ll
index 2fa7390..be29b74 100644
--- a/llvm/test/Transforms/LoopVectorize/ARM/tail-folding-loop-hint.ll
+++ b/llvm/test/Transforms/LoopVectorize/ARM/tail-folding-loop-hint.ll
@@ -41,7 +41,7 @@ define dso_local void @predicate_loop_hint(i32* noalias nocapture %A, i32* noali
 ; CHECK:       vector.body:
 ; CHECK:         %index = phi i64 [ 0, %vector.ph ], [ %index.next, %vector.body ]
 ; CHECK:         %[[ELEM0:.*]] = add i64 %index, 0
-; CHECK:         %active.lane.mask = call <4 x i1> @llvm.get.active.lane.mask.v4i1.i64(i64 %[[ELEM0]], i64 429)
+; CHECK:         %active.lane.mask = call <4 x i1> @llvm.get.active.lane.mask.v4i1.i64(i64 %[[ELEM0]], i64 430)
 ; CHECK:         %[[WML1:.*]] = call <4 x i32> @llvm.masked.load.v4i32.p0v4i32({{.*}}<4 x i1> %active.lane.mask
 ; CHECK:         %[[WML2:.*]] = call <4 x i32> @llvm.masked.load.v4i32.p0v4i32({{.*}}<4 x i1> %active.lane.mask
 ; CHECK:         %[[ADD:.*]] = add nsw <4 x i32> %[[WML2]], %[[WML1]]
diff --git a/llvm/test/Transforms/LoopVectorize/ARM/tail-folding-prefer-flag.ll b/llvm/test/Transforms/LoopVectorize/ARM/tail-folding-prefer-flag.ll
index 3261d7a..f6eacf2 100644
--- a/llvm/test/Transforms/LoopVectorize/ARM/tail-folding-prefer-flag.ll
+++ b/llvm/test/Transforms/LoopVectorize/ARM/tail-folding-prefer-flag.ll
@@ -19,7 +19,7 @@ define dso_local void @flag_overrules_hint(i32* noalias nocapture %A, i32* noali
 ; PREDFLAG:  vector.body:
 ; PREDFLAG:  %index = phi i64 [ 0, %vector.ph ], [ %index.next, %vector.body ]
 ; PREDFLAG:  %[[ELEM0:.*]] = add i64 %index, 0
-; PREDFLAG:  %active.lane.mask = call <4 x i1> @llvm.get.active.lane.mask.v4i1.i64(i64 %[[ELEM0]], i64 429)
+; PREDFLAG:  %active.lane.mask = call <4 x i1> @llvm.get.active.lane.mask.v4i1.i64(i64 %[[ELEM0]], i64 430)
 ; PREDFLAG:  %wide.masked.load = call <4 x i32> @llvm.masked.load.v4i32.p0v4i32({{.*}}, <4 x i1> %active.lane.mask
 ; PREDFLAG:  %wide.masked.load1 = call <4 x i32> @llvm.masked.load.v4i32.p0v4i32({{.*}}, <4 x i1> %active.lane.mask
 ; PREDFLAG:  %{{.*}} = add nsw <4 x i32> %wide.masked.load1, %wide.masked.load
@@ -53,11 +53,10 @@ define dso_local void @interleave4(i32* noalias nocapture %A, i32* noalias nocap
 ; PREDFLAG:  %[[ADD2:.*]] = add i32 %index, 4
 ; PREDFLAG:  %[[ADD3:.*]] = add i32 %index, 8
 ; PREDFLAG:  %[[ADD4:.*]] = add i32 %index, 12
-; PREDFLAG:  %[[BTC:.*]] = extractelement <4 x i32> %broadcast.splat, i32 0
-; PREDFLAG:  %[[ALM1:active.lane.mask.*]] = call <4 x i1> @llvm.get.active.lane.mask.v4i1.i32(i32 %[[ADD1]], i32 %[[BTC]])
-; PREDFLAG:  %[[ALM2:active.lane.mask.*]] = call <4 x i1> @llvm.get.active.lane.mask.v4i1.i32(i32 %[[ADD2]], i32 %[[BTC]])
-; PREDFLAG:  %[[ALM3:active.lane.mask.*]] = call <4 x i1> @llvm.get.active.lane.mask.v4i1.i32(i32 %[[ADD3]], i32 %[[BTC]])
-; PREDFLAG:  %[[ALM4:active.lane.mask.*]] = call <4 x i1> @llvm.get.active.lane.mask.v4i1.i32(i32 %[[ADD4]], i32 %[[BTC]])
+; PREDFLAG:  %[[ALM1:active.lane.mask.*]] = call <4 x i1> @llvm.get.active.lane.mask.v4i1.i32(i32 %[[ADD1]], i32 %N)
+; PREDFLAG:  %[[ALM2:active.lane.mask.*]] = call <4 x i1> @llvm.get.active.lane.mask.v4i1.i32(i32 %[[ADD2]], i32 %N)
+; PREDFLAG:  %[[ALM3:active.lane.mask.*]] = call <4 x i1> @llvm.get.active.lane.mask.v4i1.i32(i32 %[[ADD3]], i32 %N)
+; PREDFLAG:  %[[ALM4:active.lane.mask.*]] = call <4 x i1> @llvm.get.active.lane.mask.v4i1.i32(i32 %[[ADD4]], i32 %N)
 ;
 ; PREDFLAG:  call <4 x i32> @llvm.masked.load.v4i32.p0v4i32({{.*}}, <4 x i1> %[[ALM1]],{{.*}}
 ; PREDFLAG:  call <4 x i32> @llvm.masked.load.v4i32.p0v4i32({{.*}}, <4 x i1> %[[ALM2]],{{.*}}