[SLP]Improve splats vectorization.

[ABataev]

Summary

Replace insertelement instructions for splats with just single
insertelement + broadcast shuffle. Also, try to merge these instructions
if they come from the same/shuffled gather node.

[RKSimon]

llvm/lib/Transforms/Vectorize/SLPVectorizer.cpp

5535

What is the meaning behind this comment? Do you have examples where an identity (I guess <val0,undef,undef,undef> ?) hasn't been removed and has to be treated as a splat?

[ABataev]

llvm/lib/Transforms/Vectorize/SLPVectorizer.cpp

5535

I do not remember what does it mean ,most prpbably need to remove it.

[ABataev]

Rebase

[RKSimon]

LGTM - cheers

[fhahn]

LGTM, thanks for working on this!

llvm/lib/Transforms/Vectorize/SLPVectorizer.cpp

Value *BoUpSLP::vectorizeTree(ArrayRef<Value *> VL) {
   if (VL.size() > 2) {
     DenseMap<Value *, unsigned> UniquePositions;
     unsigned NumValues =
         std::distance(VL.begin(), find_if(reverse(VL), [](Value *V) {
                                     return !isa<UndefValue>(V);
                                   }).base());
     VF = std::max<unsigned>(VF, PowerOf2Ceil(NumValues));
     int UniqueVals = 0;
-    bool HasUndefs = false;
     for (Value *V : VL.drop_back(VL.size() - VF)) {
       if (isa<UndefValue>(V)) {
         ReuseShuffleIndicies.emplace_back(UndefMaskElem);
-        HasUndefs = true;
         continue;
       }
       if (isConstant(V)) {
         ReuseShuffleIndicies.emplace_back(UniqueValues.size());
         UniqueValues.emplace_back(V);
         continue;
       }
       auto Res = UniquePositions.try_emplace(V, UniqueValues.size());
       ReuseShuffleIndicies.emplace_back(Res.first->second);
       if (Res.second) {
         UniqueValues.emplace_back(V);
         ++UniqueVals;
       }
     }
-    if (HasUndefs && UniqueVals == 1 && UniqueValues.size() == 1) {
+    if (UniqueVals == 1 && UniqueValues.size() == 1) {
       // Emit pure splat vector.
       // FIXME: why it is not identified as an identity.
-      unsigned NumUndefs = count(ReuseShuffleIndicies, UndefMaskElem);
-      if (NumUndefs == ReuseShuffleIndicies.size() - 1)
-        ReuseShuffleIndicies.append(VF - ReuseShuffleIndicies.size(),
-                                    UndefMaskElem);
-      else
-        ReuseShuffleIndicies.assign(VF, 0);
     } else if (UniqueValues.size() >= VF - 1 || UniqueValues.size() <= 1) {
       ReuseShuffleIndicies.clear();
       UniqueValues.clear();
       UniqueValues.append(VL.begin(), std::next(VL.begin(), NumValues));
     }
     UniqueValues.append(VF - UniqueValues.size(),
                         PoisonValue::get(VL[0]->getType()));
     VL = UniqueValues;
assert(*I &&
            (I == CSEWorkList.begin() || !DT->dominates(*I, *std::prev(I))) &&
            "Worklist not sorted properly!");
     BasicBlock *BB = (*I)->getBlock();
     // For all instructions in blocks containing gather sequences:
     for (BasicBlock::iterator it = BB->begin(), e = BB->end(); it != e;) {
       Instruction *In = &*it++;
       if (isDeleted(In))
         continue;
-      if (!isa<InsertElementInst>(In) && !isa<ExtractElementInst>(In))
+      if (!isa<InsertElementInst>(In) && !isa<ExtractElementInst>(In) &&
+          !isa<ShuffleVectorInst>(In))
         continue;
 
       // Check if we can replace this instruction with any of the
       // visited instructions.
       for (Instruction *v : Visited) {
         if (In->isIdenticalTo(v) &&
             DT->dominates(v->getParent(), In->getParent())) {
           In->replaceAllUsesWith(v);

llvm/test/Transforms/SLPVectorizer/AArch64/loadi8.ll

 ; CHECK-NEXT:    [[ARRAYIDX2_1:%.*]] = getelementptr inbounds i8, i8* [[DST:%.*]], i64 1
 ; CHECK-NEXT:    [[ARRAYIDX_2:%.*]] = getelementptr inbounds i8, i8* [[SRC]], i64 2
 ; CHECK-NEXT:    [[ARRAYIDX2_2:%.*]] = getelementptr inbounds i8, i8* [[DST]], i64 2
 ; CHECK-NEXT:    [[ARRAYIDX_3:%.*]] = getelementptr inbounds i8, i8* [[SRC]], i64 3
 ; CHECK-NEXT:    [[TMP2:%.*]] = bitcast i8* [[SRC]] to <4 x i8>*
 ; CHECK-NEXT:    [[TMP3:%.*]] = load <4 x i8>, <4 x i8>* [[TMP2]], align 1
 ; CHECK-NEXT:    [[TMP4:%.*]] = zext <4 x i8> [[TMP3]] to <4 x i32>
 ; CHECK-NEXT:    [[TMP5:%.*]] = insertelement <4 x i32> poison, i32 [[TMP0]], i32 0
-; CHECK-NEXT:    [[TMP6:%.*]] = insertelement <4 x i32> [[TMP5]], i32 [[TMP0]], i32 1
-; CHECK-NEXT:    [[TMP7:%.*]] = insertelement <4 x i32> [[TMP6]], i32 [[TMP0]], i32 2
-; CHECK-NEXT:    [[TMP8:%.*]] = insertelement <4 x i32> [[TMP7]], i32 [[TMP0]], i32 3
-; CHECK-NEXT:    [[TMP9:%.*]] = mul nsw <4 x i32> [[TMP8]], [[TMP4]]
-; CHECK-NEXT:    [[TMP10:%.*]] = insertelement <4 x i32> poison, i32 [[TMP1]], i32 0
-; CHECK-NEXT:    [[TMP11:%.*]] = insertelement <4 x i32> [[TMP10]], i32 [[TMP1]], i32 1
-; CHECK-NEXT:    [[TMP12:%.*]] = insertelement <4 x i32> [[TMP11]], i32 [[TMP1]], i32 2
-; CHECK-NEXT:    [[TMP13:%.*]] = insertelement <4 x i32> [[TMP12]], i32 [[TMP1]], i32 3
-; CHECK-NEXT:    [[TMP14:%.*]] = add nsw <4 x i32> [[TMP9]], [[TMP13]]
-; CHECK-NEXT:    [[TMP15:%.*]] = icmp ult <4 x i32> [[TMP14]], <i32 256, i32 256, i32 256, i32 256>
-; CHECK-NEXT:    [[TMP16:%.*]] = icmp sgt <4 x i32> [[TMP14]], zeroinitializer
-; CHECK-NEXT:    [[TMP17:%.*]] = sext <4 x i1> [[TMP16]] to <4 x i32>
-; CHECK-NEXT:    [[TMP18:%.*]] = select <4 x i1> [[TMP15]], <4 x i32> [[TMP14]], <4 x i32> [[TMP17]]
-; CHECK-NEXT:    [[TMP19:%.*]] = trunc <4 x i32> [[TMP18]] to <4 x i8>
+; CHECK-NEXT:    [[SHUFFLE:%.*]] = shufflevector <4 x i32> [[TMP5]], <4 x i32> poison, <4 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP6:%.*]] = mul nsw <4 x i32> [[SHUFFLE]], [[TMP4]]
+; CHECK-NEXT:    [[TMP7:%.*]] = insertelement <4 x i32> poison, i32 [[TMP1]], i32 0
+; CHECK-NEXT:    [[SHUFFLE1:%.*]] = shufflevector <4 x i32> [[TMP7]], <4 x i32> poison, <4 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP8:%.*]] = add nsw <4 x i32> [[TMP6]], [[SHUFFLE1]]
+; CHECK-NEXT:    [[TMP9:%.*]] = icmp ult <4 x i32> [[TMP8]], <i32 256, i32 256, i32 256, i32 256>
+; CHECK-NEXT:    [[TMP10:%.*]] = icmp sgt <4 x i32> [[TMP8]], zeroinitializer
+; CHECK-NEXT:    [[TMP11:%.*]] = sext <4 x i1> [[TMP10]] to <4 x i32>
+; CHECK-NEXT:    [[TMP12:%.*]] = select <4 x i1> [[TMP9]], <4 x i32> [[TMP8]], <4 x i32> [[TMP11]]
+; CHECK-NEXT:    [[TMP13:%.*]] = trunc <4 x i32> [[TMP12]] to <4 x i8>
 ; CHECK-NEXT:    [[ARRAYIDX2_3:%.*]] = getelementptr inbounds i8, i8* [[DST]], i64 3
-; CHECK-NEXT:    [[TMP20:%.*]] = bitcast i8* [[DST]] to <4 x i8>*
-; CHECK-NEXT:    store <4 x i8> [[TMP19]], <4 x i8>* [[TMP20]], align 1
+; CHECK-NEXT:    [[TMP14:%.*]] = bitcast i8* [[DST]] to <4 x i8>*
+; CHECK-NEXT:    store <4 x i8> [[TMP13]], <4 x i8>* [[TMP14]], align 1
 ; CHECK-NEXT:    ret void
 ;
 entry:
   %scale = getelementptr inbounds %struct.weight_t, %struct.weight_t* %w, i64 0, i32 0
   %0 = load i32, i32* %scale, align 16
   %offset = getelementptr inbounds %struct.weight_t, %struct.weight_t* %w, i64 0, i32 1
   %1 = load i32, i32* %offset, align 4
   %2 = load i8, i8* %src, align 1

llvm/test/Transforms/SLPVectorizer/AArch64/vectorizable-selects-uniform-cmps.ll

 ; CHECK-NEXT:    [[GEP_4:%.*]] = getelementptr inbounds i8, i8* [[PTR]], i8 4
 ; CHECK-NEXT:    [[GEP_5:%.*]] = getelementptr inbounds i8, i8* [[PTR]], i8 5
 ; CHECK-NEXT:    [[GEP_6:%.*]] = getelementptr inbounds i8, i8* [[PTR]], i8 6
 ; CHECK-NEXT:    [[GEP_7:%.*]] = getelementptr inbounds i8, i8* [[PTR]], i8 7
 ; CHECK-NEXT:    [[TMP0:%.*]] = bitcast i8* [[PTR]] to <8 x i8>*
 ; CHECK-NEXT:    [[TMP1:%.*]] = load <8 x i8>, <8 x i8>* [[TMP0]], align 1
 ; CHECK-NEXT:    [[TMP2:%.*]] = icmp ugt <8 x i8> [[TMP1]], <i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1>
 ; CHECK-NEXT:    [[TMP3:%.*]] = insertelement <8 x i8> poison, i8 [[X:%.*]], i32 0
-; CHECK-NEXT:    [[TMP4:%.*]] = insertelement <8 x i8> [[TMP3]], i8 [[X]], i32 1
-; CHECK-NEXT:    [[TMP5:%.*]] = insertelement <8 x i8> [[TMP4]], i8 [[X]], i32 2
-; CHECK-NEXT:    [[TMP6:%.*]] = insertelement <8 x i8> [[TMP5]], i8 [[X]], i32 3
-; CHECK-NEXT:    [[TMP7:%.*]] = insertelement <8 x i8> [[TMP6]], i8 [[X]], i32 4
-; CHECK-NEXT:    [[TMP8:%.*]] = insertelement <8 x i8> [[TMP7]], i8 [[X]], i32 5
-; CHECK-NEXT:    [[TMP9:%.*]] = insertelement <8 x i8> [[TMP8]], i8 [[X]], i32 6
-; CHECK-NEXT:    [[TMP10:%.*]] = insertelement <8 x i8> [[TMP9]], i8 [[X]], i32 7
-; CHECK-NEXT:    [[TMP11:%.*]] = select <8 x i1> [[TMP2]], <8 x i8> [[TMP1]], <8 x i8> [[TMP10]]
-; CHECK-NEXT:    [[TMP12:%.*]] = bitcast i8* [[PTR]] to <8 x i8>*
-; CHECK-NEXT:    store <8 x i8> [[TMP11]], <8 x i8>* [[TMP12]], align 2
+; CHECK-NEXT:    [[SHUFFLE:%.*]] = shufflevector <8 x i8> [[TMP3]], <8 x i8> poison, <8 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP4:%.*]] = select <8 x i1> [[TMP2]], <8 x i8> [[TMP1]], <8 x i8> [[SHUFFLE]]
+; CHECK-NEXT:    [[TMP5:%.*]] = bitcast i8* [[PTR]] to <8 x i8>*
+; CHECK-NEXT:    store <8 x i8> [[TMP4]], <8 x i8>* [[TMP5]], align 2
 ; CHECK-NEXT:    ret void
 ;
 entry:
   %l.0 = load i8, i8* %ptr
   %cmp.0 = icmp ugt i8 %l.0, 16383
   %s.0 = select i1 %cmp.0, i8 %l.0, i8 %x
   store i8 %s.0, i8* %ptr, align 2
 

 ; CHECK-NEXT:    [[GEP_4:%.*]] = getelementptr inbounds i8, i8* [[PTR]], i8 4
 ; CHECK-NEXT:    [[GEP_5:%.*]] = getelementptr inbounds i8, i8* [[PTR]], i8 5
 ; CHECK-NEXT:    [[GEP_6:%.*]] = getelementptr inbounds i8, i8* [[PTR]], i8 6
 ; CHECK-NEXT:    [[GEP_7:%.*]] = getelementptr inbounds i8, i8* [[PTR]], i8 7
 ; CHECK-NEXT:    [[TMP0:%.*]] = bitcast i8* [[PTR]] to <8 x i8>*
 ; CHECK-NEXT:    [[TMP1:%.*]] = load <8 x i8>, <8 x i8>* [[TMP0]], align 1
 ; CHECK-NEXT:    [[TMP2:%.*]] = icmp ugt <8 x i8> [[TMP1]], <i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1>
 ; CHECK-NEXT:    [[TMP3:%.*]] = insertelement <8 x i8> poison, i8 [[X:%.*]], i32 0
-; CHECK-NEXT:    [[TMP4:%.*]] = insertelement <8 x i8> [[TMP3]], i8 [[X]], i32 1
-; CHECK-NEXT:    [[TMP5:%.*]] = insertelement <8 x i8> [[TMP4]], i8 [[X]], i32 2
-; CHECK-NEXT:    [[TMP6:%.*]] = insertelement <8 x i8> [[TMP5]], i8 [[X]], i32 3
-; CHECK-NEXT:    [[TMP7:%.*]] = insertelement <8 x i8> [[TMP6]], i8 [[X]], i32 4
-; CHECK-NEXT:    [[TMP8:%.*]] = insertelement <8 x i8> [[TMP7]], i8 [[X]], i32 5
-; CHECK-NEXT:    [[TMP9:%.*]] = insertelement <8 x i8> [[TMP8]], i8 [[X]], i32 6
-; CHECK-NEXT:    [[TMP10:%.*]] = insertelement <8 x i8> [[TMP9]], i8 [[X]], i32 7
-; CHECK-NEXT:    [[TMP11:%.*]] = select <8 x i1> [[TMP2]], <8 x i8> [[TMP1]], <8 x i8> [[TMP10]]
-; CHECK-NEXT:    [[TMP12:%.*]] = bitcast i8* [[PTR]] to <8 x i8>*
-; CHECK-NEXT:    store <8 x i8> [[TMP11]], <8 x i8>* [[TMP12]], align 2
+; CHECK-NEXT:    [[SHUFFLE:%.*]] = shufflevector <8 x i8> [[TMP3]], <8 x i8> poison, <8 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP4:%.*]] = select <8 x i1> [[TMP2]], <8 x i8> [[TMP1]], <8 x i8> [[SHUFFLE]]
+; CHECK-NEXT:    [[TMP5:%.*]] = bitcast i8* [[PTR]] to <8 x i8>*
+; CHECK-NEXT:    store <8 x i8> [[TMP4]], <8 x i8>* [[TMP5]], align 2
 ; CHECK-NEXT:    [[GEP_8:%.*]] = getelementptr inbounds i8, i8* [[PTR]], i8 8
 ; CHECK-NEXT:    [[L_8:%.*]] = load i8, i8* [[GEP_8]], align 1
 ; CHECK-NEXT:    [[CMP_8:%.*]] = icmp ugt i8 [[L_8]], -1
-; CHECK-NEXT:    [[TMP13:%.*]] = extractelement <8 x i8> [[TMP1]], i32 0
-; CHECK-NEXT:    [[S_8:%.*]] = select i1 [[CMP_8]], i8 [[TMP13]], i8 [[X]]
-; CHECK-NEXT:    [[TMP14:%.*]] = extractelement <8 x i8> [[TMP11]], i32 0
-; CHECK-NEXT:    store i8 [[TMP14]], i8* [[GEP_8]], align 2
+; CHECK-NEXT:    [[TMP6:%.*]] = extractelement <8 x i8> [[TMP1]], i32 0
+; CHECK-NEXT:    [[S_8:%.*]] = select i1 [[CMP_8]], i8 [[TMP6]], i8 [[X]]
+; CHECK-NEXT:    [[TMP7:%.*]] = extractelement <8 x i8> [[TMP4]], i32 0
+; CHECK-NEXT:    store i8 [[TMP7]], i8* [[GEP_8]], align 2
 ; CHECK-NEXT:    [[GEP_9:%.*]] = getelementptr inbounds i8, i8* [[PTR]], i8 9
 ; CHECK-NEXT:    [[L_9:%.*]] = load i8, i8* [[GEP_9]], align 1
 ; CHECK-NEXT:    [[CMP_9:%.*]] = icmp ugt i8 [[L_9]], -1
 ; CHECK-NEXT:    [[S_9:%.*]] = select i1 [[CMP_9]], i8 [[L_9]], i8 [[X]]
 ; CHECK-NEXT:    store i8 [[S_9]], i8* [[GEP_9]], align 2
 ; CHECK-NEXT:    [[GEP_10:%.*]] = getelementptr inbounds i8, i8* [[PTR]], i8 10
 ; CHECK-NEXT:    [[L_10:%.*]] = load i8, i8* [[GEP_10]], align 1
 ; CHECK-NEXT:    [[CMP_10:%.*]] = icmp ugt i8 [[L_10]], -1

 ; CHECK-NEXT:  entry:
 ; CHECK-NEXT:    [[GEP_1:%.*]] = getelementptr inbounds i16, i16* [[PTR:%.*]], i16 1
 ; CHECK-NEXT:    [[GEP_2:%.*]] = getelementptr inbounds i16, i16* [[PTR]], i16 2
 ; CHECK-NEXT:    [[GEP_3:%.*]] = getelementptr inbounds i16, i16* [[PTR]], i16 3
 ; CHECK-NEXT:    [[TMP0:%.*]] = bitcast i16* [[PTR]] to <4 x i16>*
 ; CHECK-NEXT:    [[TMP1:%.*]] = load <4 x i16>, <4 x i16>* [[TMP0]], align 2
 ; CHECK-NEXT:    [[TMP2:%.*]] = icmp ugt <4 x i16> [[TMP1]], <i16 16383, i16 16383, i16 16383, i16 16383>
 ; CHECK-NEXT:    [[TMP3:%.*]] = insertelement <4 x i16> poison, i16 [[X:%.*]], i32 0
-; CHECK-NEXT:    [[TMP4:%.*]] = insertelement <4 x i16> [[TMP3]], i16 [[X]], i32 1
-; CHECK-NEXT:    [[TMP5:%.*]] = insertelement <4 x i16> [[TMP4]], i16 [[X]], i32 2
-; CHECK-NEXT:    [[TMP6:%.*]] = insertelement <4 x i16> [[TMP5]], i16 [[X]], i32 3
-; CHECK-NEXT:    [[TMP7:%.*]] = select <4 x i1> [[TMP2]], <4 x i16> [[TMP1]], <4 x i16> [[TMP6]]
-; CHECK-NEXT:    [[TMP8:%.*]] = bitcast i16* [[PTR]] to <4 x i16>*
-; CHECK-NEXT:    store <4 x i16> [[TMP7]], <4 x i16>* [[TMP8]], align 2
+; CHECK-NEXT:    [[SHUFFLE:%.*]] = shufflevector <4 x i16> [[TMP3]], <4 x i16> poison, <4 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP4:%.*]] = select <4 x i1> [[TMP2]], <4 x i16> [[TMP1]], <4 x i16> [[SHUFFLE]]
+; CHECK-NEXT:    [[TMP5:%.*]] = bitcast i16* [[PTR]] to <4 x i16>*
+; CHECK-NEXT:    store <4 x i16> [[TMP4]], <4 x i16>* [[TMP5]], align 2
 ; CHECK-NEXT:    ret void
 ;
 entry:
   %l.0 = load i16, i16* %ptr
   %cmp.0 = icmp ugt i16 %l.0, 16383
   %s.0 = select i1 %cmp.0, i16 %l.0, i16 %x
   store i16 %s.0, i16* %ptr, align 2
 

 ; CHECK-NEXT:    [[GEP_4:%.*]] = getelementptr inbounds i16, i16* [[PTR]], i16 4
 ; CHECK-NEXT:    [[GEP_5:%.*]] = getelementptr inbounds i16, i16* [[PTR]], i16 5
 ; CHECK-NEXT:    [[GEP_6:%.*]] = getelementptr inbounds i16, i16* [[PTR]], i16 6
 ; CHECK-NEXT:    [[GEP_7:%.*]] = getelementptr inbounds i16, i16* [[PTR]], i16 7
 ; CHECK-NEXT:    [[TMP0:%.*]] = bitcast i16* [[PTR]] to <8 x i16>*
 ; CHECK-NEXT:    [[TMP1:%.*]] = load <8 x i16>, <8 x i16>* [[TMP0]], align 2
 ; CHECK-NEXT:    [[TMP2:%.*]] = icmp ult <8 x i16> [[TMP1]], <i16 16383, i16 16383, i16 16383, i16 16383, i16 16383, i16 16383, i16 16383, i16 16383>
 ; CHECK-NEXT:    [[TMP3:%.*]] = insertelement <8 x i16> poison, i16 [[X:%.*]], i32 0
-; CHECK-NEXT:    [[TMP4:%.*]] = insertelement <8 x i16> [[TMP3]], i16 [[X]], i32 1
-; CHECK-NEXT:    [[TMP5:%.*]] = insertelement <8 x i16> [[TMP4]], i16 [[X]], i32 2
-; CHECK-NEXT:    [[TMP6:%.*]] = insertelement <8 x i16> [[TMP5]], i16 [[X]], i32 3
-; CHECK-NEXT:    [[TMP7:%.*]] = insertelement <8 x i16> [[TMP6]], i16 [[X]], i32 4
-; CHECK-NEXT:    [[TMP8:%.*]] = insertelement <8 x i16> [[TMP7]], i16 [[X]], i32 5
-; CHECK-NEXT:    [[TMP9:%.*]] = insertelement <8 x i16> [[TMP8]], i16 [[X]], i32 6
-; CHECK-NEXT:    [[TMP10:%.*]] = insertelement <8 x i16> [[TMP9]], i16 [[X]], i32 7
-; CHECK-NEXT:    [[TMP11:%.*]] = select <8 x i1> [[TMP2]], <8 x i16> [[TMP1]], <8 x i16> [[TMP10]]
-; CHECK-NEXT:    [[TMP12:%.*]] = bitcast i16* [[PTR]] to <8 x i16>*
-; CHECK-NEXT:    store <8 x i16> [[TMP11]], <8 x i16>* [[TMP12]], align 2
+; CHECK-NEXT:    [[SHUFFLE:%.*]] = shufflevector <8 x i16> [[TMP3]], <8 x i16> poison, <8 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP4:%.*]] = select <8 x i1> [[TMP2]], <8 x i16> [[TMP1]], <8 x i16> [[SHUFFLE]]
+; CHECK-NEXT:    [[TMP5:%.*]] = bitcast i16* [[PTR]] to <8 x i16>*
+; CHECK-NEXT:    store <8 x i16> [[TMP4]], <8 x i16>* [[TMP5]], align 2
 ; CHECK-NEXT:    ret void
 ;
 entry:
   %l.0 = load i16, i16* %ptr
   %cmp.0 = icmp ult i16 %l.0, 16383
   %s.0 = select i1 %cmp.0, i16 %l.0, i16 %x
   store i16 %s.0, i16* %ptr, align 2
 

 ; CHECK-NEXT:  entry:
 ; CHECK-NEXT:    [[GEP_1:%.*]] = getelementptr inbounds i32, i32* [[PTR:%.*]], i32 1
 ; CHECK-NEXT:    [[GEP_2:%.*]] = getelementptr inbounds i32, i32* [[PTR]], i32 2
 ; CHECK-NEXT:    [[GEP_3:%.*]] = getelementptr inbounds i32, i32* [[PTR]], i32 3
 ; CHECK-NEXT:    [[TMP0:%.*]] = bitcast i32* [[PTR]] to <4 x i32>*
 ; CHECK-NEXT:    [[TMP1:%.*]] = load <4 x i32>, <4 x i32>* [[TMP0]], align 4
 ; CHECK-NEXT:    [[TMP2:%.*]] = icmp eq <4 x i32> [[TMP1]], <i32 16383, i32 16383, i32 16383, i32 16383>
 ; CHECK-NEXT:    [[TMP3:%.*]] = insertelement <4 x i32> poison, i32 [[X:%.*]], i32 0
-; CHECK-NEXT:    [[TMP4:%.*]] = insertelement <4 x i32> [[TMP3]], i32 [[X]], i32 1
-; CHECK-NEXT:    [[TMP5:%.*]] = insertelement <4 x i32> [[TMP4]], i32 [[X]], i32 2
-; CHECK-NEXT:    [[TMP6:%.*]] = insertelement <4 x i32> [[TMP5]], i32 [[X]], i32 3
-; CHECK-NEXT:    [[TMP7:%.*]] = select <4 x i1> [[TMP2]], <4 x i32> [[TMP1]], <4 x i32> [[TMP6]]
-; CHECK-NEXT:    [[TMP8:%.*]] = bitcast i32* [[PTR]] to <4 x i32>*
-; CHECK-NEXT:    store <4 x i32> [[TMP7]], <4 x i32>* [[TMP8]], align 2
+; CHECK-NEXT:    [[SHUFFLE:%.*]] = shufflevector <4 x i32> [[TMP3]], <4 x i32> poison, <4 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP4:%.*]] = select <4 x i1> [[TMP2]], <4 x i32> [[TMP1]], <4 x i32> [[SHUFFLE]]
+; CHECK-NEXT:    [[TMP5:%.*]] = bitcast i32* [[PTR]] to <4 x i32>*
+; CHECK-NEXT:    store <4 x i32> [[TMP4]], <4 x i32>* [[TMP5]], align 2
 ; CHECK-NEXT:    ret void
 ;
 entry:
   %l.0 = load i32, i32* %ptr
   %cmp.0 = icmp eq i32 %l.0, 16383
   %s.0 = select i1 %cmp.0, i32 %l.0, i32 %x
   store i32 %s.0, i32* %ptr, align 2
 

llvm/test/Transforms/SLPVectorizer/AArch64/vectorize-free-extracts-inserts.ll

 ; CHECK-NEXT:    [[TMP13:%.*]] = insertelement <8 x double> poison, double [[V1_LANE_6]], i32 0
 ; CHECK-NEXT:    [[TMP14:%.*]] = insertelement <8 x double> [[TMP13]], double [[V1_LANE_7]], i32 1
 ; CHECK-NEXT:    [[TMP15:%.*]] = insertelement <8 x double> [[TMP14]], double [[V1_LANE_8]], i32 2
 ; CHECK-NEXT:    [[TMP16:%.*]] = insertelement <8 x double> [[TMP15]], double [[V1_LANE_0]], i32 3
 ; CHECK-NEXT:    [[TMP17:%.*]] = insertelement <8 x double> [[TMP16]], double [[V1_LANE_1]], i32 4
 ; CHECK-NEXT:    [[TMP18:%.*]] = insertelement <8 x double> [[TMP17]], double [[V1_LANE_2]], i32 5
 ; CHECK-NEXT:    [[TMP19:%.*]] = insertelement <8 x double> [[TMP18]], double [[V1_LANE_3]], i32 6
 ; CHECK-NEXT:    [[TMP20:%.*]] = insertelement <8 x double> [[TMP19]], double [[V1_LANE_4]], i32 7
-; CHECK-NEXT:    [[SHUFFLE:%.*]] = shufflevector <8 x double> [[TMP10]], <8 x double> poison, <8 x i32> <i32 0, i32 1, i32 2, i32 1, i32 2, i32 0, i32 1, i32 2>
-; CHECK-NEXT:    [[TMP21:%.*]] = fmul <8 x double> [[TMP20]], [[SHUFFLE]]
+; CHECK-NEXT:    [[TMP21:%.*]] = fmul <8 x double> [[TMP20]], [[SHUFFLE2]]
 ; CHECK-NEXT:    [[B_LANE_8:%.*]] = fmul double [[V1_LANE_5]], [[V2_LANE_0]]
 ; CHECK-NEXT:    [[TMP22:%.*]] = shufflevector <8 x double> [[TMP21]], <8 x double> poison, <9 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7, i32 undef>
 ; CHECK-NEXT:    [[B_INS_71:%.*]] = shufflevector <9 x double> undef, <9 x double> [[TMP22]], <9 x i32> <i32 9, i32 10, i32 11, i32 12, i32 13, i32 14, i32 15, i32 16, i32 8>
 ; CHECK-NEXT:    [[B_INS_8:%.*]] = insertelement <9 x double> [[B_INS_71]], double [[B_LANE_8]], i32 8
 ; CHECK-NEXT:    [[RES:%.*]] = fsub <9 x double> [[A_INS_8]], [[B_INS_8]]
 ; CHECK-NEXT:    store <9 x double> [[RES]], <9 x double>* [[PTR_1]], align 8
 ; CHECK-NEXT:    ret void
 ;

llvm/test/Transforms/SLPVectorizer/X86/PR35628_2.ll

 ; NOTE: Assertions have been autogenerated by utils/update_test_checks.py
 ; RUN: opt -slp-vectorizer -slp-vectorize-hor -slp-vectorize-hor-store -S < %s -mtriple=x86_64-unknown-linux-gnu -mcpu=haswell | FileCheck %s
 target datalayout = "e-m:e-i64:64-f80:128-n8:16:32:64-S128-ni:1"
 
 define void @test() #0 {
 ; CHECK-LABEL: @test(
 ; CHECK-NEXT:  entry:
 ; CHECK-NEXT:    br label [[LOOP:%.*]]
 ; CHECK:       loop:
 ; CHECK-NEXT:    [[DUMMY_PHI:%.*]] = phi i64 [ 1, [[ENTRY:%.*]] ], [ [[OP_EXTRA1:%.*]], [[LOOP]] ]
-; CHECK-NEXT:    [[TMP0:%.*]] = phi i64 [ 2, [[ENTRY]] ], [ [[TMP6:%.*]], [[LOOP]] ]
+; CHECK-NEXT:    [[TMP0:%.*]] = phi i64 [ 2, [[ENTRY]] ], [ [[TMP3:%.*]], [[LOOP]] ]
 ; CHECK-NEXT:    [[DUMMY_ADD:%.*]] = add i16 0, 0
 ; CHECK-NEXT:    [[TMP1:%.*]] = insertelement <4 x i64> poison, i64 [[TMP0]], i32 0
-; CHECK-NEXT:    [[TMP2:%.*]] = insertelement <4 x i64> [[TMP1]], i64 [[TMP0]], i32 1
-; CHECK-NEXT:    [[TMP3:%.*]] = insertelement <4 x i64> [[TMP2]], i64 [[TMP0]], i32 2
-; CHECK-NEXT:    [[TMP4:%.*]] = insertelement <4 x i64> [[TMP3]], i64 [[TMP0]], i32 3
-; CHECK-NEXT:    [[TMP5:%.*]] = add <4 x i64> [[TMP4]], <i64 3, i64 2, i64 1, i64 0>
-; CHECK-NEXT:    [[TMP6]] = extractelement <4 x i64> [[TMP5]], i32 3
-; CHECK-NEXT:    [[TMP7:%.*]] = extractelement <4 x i64> [[TMP5]], i32 0
-; CHECK-NEXT:    [[DUMMY_SHL:%.*]] = shl i64 [[TMP7]], 32
-; CHECK-NEXT:    [[TMP8:%.*]] = add <4 x i64> <i64 1, i64 1, i64 1, i64 1>, [[TMP5]]
-; CHECK-NEXT:    [[TMP9:%.*]] = ashr exact <4 x i64> [[TMP8]], <i64 32, i64 32, i64 32, i64 32>
-; CHECK-NEXT:    [[TMP10:%.*]] = call i64 @llvm.vector.reduce.add.v4i64(<4 x i64> [[TMP9]])
-; CHECK-NEXT:    [[OP_EXTRA:%.*]] = add i64 [[TMP10]], 0
-; CHECK-NEXT:    [[OP_EXTRA1]] = add i64 [[OP_EXTRA]], [[TMP6]]
+; CHECK-NEXT:    [[SHUFFLE:%.*]] = shufflevector <4 x i64> [[TMP1]], <4 x i64> poison, <4 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP2:%.*]] = add <4 x i64> [[SHUFFLE]], <i64 3, i64 2, i64 1, i64 0>
+; CHECK-NEXT:    [[TMP3]] = extractelement <4 x i64> [[TMP2]], i32 3
+; CHECK-NEXT:    [[TMP4:%.*]] = extractelement <4 x i64> [[TMP2]], i32 0
+; CHECK-NEXT:    [[DUMMY_SHL:%.*]] = shl i64 [[TMP4]], 32
+; CHECK-NEXT:    [[TMP5:%.*]] = add <4 x i64> <i64 1, i64 1, i64 1, i64 1>, [[TMP2]]
+; CHECK-NEXT:    [[TMP6:%.*]] = ashr exact <4 x i64> [[TMP5]], <i64 32, i64 32, i64 32, i64 32>
+; CHECK-NEXT:    [[TMP7:%.*]] = call i64 @llvm.vector.reduce.add.v4i64(<4 x i64> [[TMP6]])
+; CHECK-NEXT:    [[OP_EXTRA:%.*]] = add i64 [[TMP7]], 0
+; CHECK-NEXT:    [[OP_EXTRA1]] = add i64 [[OP_EXTRA]], [[TMP3]]
 ; CHECK-NEXT:    br label [[LOOP]]
 ;
 entry:
   br label %loop
 
 loop:
   %dummy_phi = phi i64 [ 1, %entry ], [ %last, %loop ]
   %0 = phi i64 [ 2, %entry ], [ %fork, %loop ]

llvm/test/Transforms/SLPVectorizer/X86/bad-reduction.ll

 @output = dso_local local_unnamed_addr global [8 x i32] zeroinitializer, align 16
 
 define void @PR47450(i16* nocapture readonly %p) {
 ; CHECK-LABEL: @PR47450(
 ; CHECK-NEXT:    [[X:%.*]] = load i16, i16* [[P:%.*]], align 2
 ; CHECK-NEXT:    [[Z:%.*]] = zext i16 [[X]] to i32
 ; CHECK-NEXT:    [[S:%.*]] = shl nuw nsw i32 [[Z]], 1
 ; CHECK-NEXT:    [[TMP1:%.*]] = insertelement <4 x i32> poison, i32 [[S]], i32 0
-; CHECK-NEXT:    [[TMP2:%.*]] = insertelement <4 x i32> [[TMP1]], i32 [[S]], i32 1
-; CHECK-NEXT:    [[TMP3:%.*]] = insertelement <4 x i32> [[TMP2]], i32 [[S]], i32 2
-; CHECK-NEXT:    [[TMP4:%.*]] = insertelement <4 x i32> [[TMP3]], i32 [[S]], i32 3
-; CHECK-NEXT:    store <4 x i32> [[TMP4]], <4 x i32>* bitcast ([8 x i32]* @output to <4 x i32>*), align 16
+; CHECK-NEXT:    [[SHUFFLE:%.*]] = shufflevector <4 x i32> [[TMP1]], <4 x i32> poison, <4 x i32> zeroinitializer
+; CHECK-NEXT:    store <4 x i32> [[SHUFFLE]], <4 x i32>* bitcast ([8 x i32]* @output to <4 x i32>*), align 16
 ; CHECK-NEXT:    ret void
 ;
   %x = load i16, i16* %p, align 2
   %z = zext i16 %x to i32
   %s = shl nuw nsw i32 %z, 1
   store i32 %s, i32* getelementptr inbounds ([8 x i32], [8 x i32]* @output, i64 0, i64 0), align 16
   store i32 %s, i32* getelementptr inbounds ([8 x i32], [8 x i32]* @output, i64 0, i64 1), align 4
   store i32 %s, i32* getelementptr inbounds ([8 x i32], [8 x i32]* @output, i64 0, i64 2), align 8
   store i32 %s, i32* getelementptr inbounds ([8 x i32], [8 x i32]* @output, i64 0, i64 3), align 4
   ret void
 }

llvm/test/Transforms/SLPVectorizer/X86/barriercall.ll

 ; NOTE: Assertions have been autogenerated by utils/update_test_checks.py
 ; RUN: opt < %s -basic-aa -slp-vectorizer -dce -S -mtriple=x86_64-apple-macosx10.8.0 -mcpu=corei7-avx | FileCheck %s
 
 target datalayout = "e-p:64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:64:64-f32:32:32-f64:64:64-v64:64:64-v128:128:128-a0:0:64-s0:64:64-f80:128:128-n8:16:32:64-S128"
 target triple = "x86_64-apple-macosx10.8.0"
 
 define i32 @foo(i32* nocapture %A, i32 %n) {
 ; CHECK-LABEL: @foo(
 ; CHECK-NEXT:  entry:
 ; CHECK-NEXT:    [[CALL:%.*]] = tail call i32 (...) @bar()
 ; CHECK-NEXT:    [[TMP0:%.*]] = insertelement <4 x i32> poison, i32 [[N:%.*]], i32 0
-; CHECK-NEXT:    [[TMP1:%.*]] = insertelement <4 x i32> [[TMP0]], i32 [[N]], i32 1
-; CHECK-NEXT:    [[TMP2:%.*]] = insertelement <4 x i32> [[TMP1]], i32 [[N]], i32 2
-; CHECK-NEXT:    [[TMP3:%.*]] = insertelement <4 x i32> [[TMP2]], i32 [[N]], i32 3
-; CHECK-NEXT:    [[TMP4:%.*]] = mul nsw <4 x i32> [[TMP3]], <i32 5, i32 9, i32 3, i32 10>
-; CHECK-NEXT:    [[TMP5:%.*]] = shl <4 x i32> [[TMP3]], <i32 5, i32 9, i32 3, i32 10>
-; CHECK-NEXT:    [[TMP6:%.*]] = shufflevector <4 x i32> [[TMP4]], <4 x i32> [[TMP5]], <4 x i32> <i32 0, i32 1, i32 6, i32 3>
-; CHECK-NEXT:    [[TMP7:%.*]] = add nsw <4 x i32> [[TMP6]], <i32 9, i32 9, i32 9, i32 9>
-; CHECK-NEXT:    [[TMP8:%.*]] = bitcast i32* [[A:%.*]] to <4 x i32>*
-; CHECK-NEXT:    store <4 x i32> [[TMP7]], <4 x i32>* [[TMP8]], align 4
+; CHECK-NEXT:    [[SHUFFLE:%.*]] = shufflevector <4 x i32> [[TMP0]], <4 x i32> poison, <4 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP1:%.*]] = mul nsw <4 x i32> [[SHUFFLE]], <i32 5, i32 9, i32 3, i32 10>
+; CHECK-NEXT:    [[TMP2:%.*]] = shl <4 x i32> [[SHUFFLE]], <i32 5, i32 9, i32 3, i32 10>
+; CHECK-NEXT:    [[TMP3:%.*]] = shufflevector <4 x i32> [[TMP1]], <4 x i32> [[TMP2]], <4 x i32> <i32 0, i32 1, i32 6, i32 3>
+; CHECK-NEXT:    [[TMP4:%.*]] = add nsw <4 x i32> [[TMP3]], <i32 9, i32 9, i32 9, i32 9>
+; CHECK-NEXT:    [[TMP5:%.*]] = bitcast i32* [[A:%.*]] to <4 x i32>*
+; CHECK-NEXT:    store <4 x i32> [[TMP4]], <4 x i32>* [[TMP5]], align 4
 ; CHECK-NEXT:    ret i32 undef
 ;
 entry:
   %call = tail call i32 (...) @bar() #2
   %mul = mul nsw i32 %n, 5
   %add = add nsw i32 %mul, 9
   store i32 %add, i32* %A, align 4
   %mul1 = mul nsw i32 %n, 9

llvm/test/Transforms/SLPVectorizer/X86/broadcast.ll

 define void @bcast_vals(i64 *%A, i64 *%B, i64 *%S) {
 ; CHECK-LABEL: @bcast_vals(
 ; CHECK-NEXT:  entry:
 ; CHECK-NEXT:    [[A0:%.*]] = load i64, i64* [[A:%.*]], align 8
 ; CHECK-NEXT:    [[B0:%.*]] = load i64, i64* [[B:%.*]], align 8
 ; CHECK-NEXT:    [[V1:%.*]] = sub i64 [[A0]], 1
 ; CHECK-NEXT:    [[V2:%.*]] = sub i64 [[B0]], 1
 ; CHECK-NEXT:    [[TMP0:%.*]] = insertelement <4 x i64> poison, i64 [[V1]], i32 0
-; CHECK-NEXT:    [[TMP1:%.*]] = insertelement <4 x i64> [[TMP0]], i64 [[V1]], i32 1
-; CHECK-NEXT:    [[TMP2:%.*]] = insertelement <4 x i64> [[TMP1]], i64 [[V1]], i32 2
-; CHECK-NEXT:    [[TMP3:%.*]] = insertelement <4 x i64> [[TMP2]], i64 [[V1]], i32 3
-; CHECK-NEXT:    [[TMP4:%.*]] = insertelement <4 x i64> poison, i64 [[V2]], i32 0
-; CHECK-NEXT:    [[TMP5:%.*]] = insertelement <4 x i64> [[TMP4]], i64 [[V2]], i32 1
-; CHECK-NEXT:    [[TMP6:%.*]] = insertelement <4 x i64> [[TMP5]], i64 [[V2]], i32 2
-; CHECK-NEXT:    [[TMP7:%.*]] = insertelement <4 x i64> [[TMP6]], i64 [[V2]], i32 3
-; CHECK-NEXT:    [[TMP8:%.*]] = add <4 x i64> [[TMP3]], [[TMP7]]
+; CHECK-NEXT:    [[SHUFFLE:%.*]] = shufflevector <4 x i64> [[TMP0]], <4 x i64> poison, <4 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP1:%.*]] = insertelement <4 x i64> poison, i64 [[V2]], i32 0
+; CHECK-NEXT:    [[SHUFFLE1:%.*]] = shufflevector <4 x i64> [[TMP1]], <4 x i64> poison, <4 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP2:%.*]] = add <4 x i64> [[SHUFFLE]], [[SHUFFLE1]]
 ; CHECK-NEXT:    [[IDXS0:%.*]] = getelementptr inbounds i64, i64* [[S:%.*]], i64 0
 ; CHECK-NEXT:    [[IDXS1:%.*]] = getelementptr inbounds i64, i64* [[S]], i64 1
 ; CHECK-NEXT:    [[IDXS2:%.*]] = getelementptr inbounds i64, i64* [[S]], i64 2
 ; CHECK-NEXT:    [[IDXS3:%.*]] = getelementptr inbounds i64, i64* [[S]], i64 3
-; CHECK-NEXT:    [[TMP9:%.*]] = bitcast i64* [[IDXS0]] to <4 x i64>*
-; CHECK-NEXT:    store <4 x i64> [[TMP8]], <4 x i64>* [[TMP9]], align 8
+; CHECK-NEXT:    [[TMP3:%.*]] = bitcast i64* [[IDXS0]] to <4 x i64>*
+; CHECK-NEXT:    store <4 x i64> [[TMP2]], <4 x i64>* [[TMP3]], align 8
 ; CHECK-NEXT:    ret void
 ;
 entry:
   %A0 = load i64, i64 *%A, align 8
   %B0 = load i64, i64 *%B, align 8
 
   %v1 = sub i64 %A0, 1
   %v2 = sub i64 %B0, 1

 ; CHECK-NEXT:    [[E0:%.*]] = load i16, i16* [[E:%.*]], align 8
 ; CHECK-NEXT:    [[V1:%.*]] = sext i16 [[A0]] to i32
 ; CHECK-NEXT:    [[TMP0:%.*]] = insertelement <4 x i16> poison, i16 [[B0]], i32 0
 ; CHECK-NEXT:    [[TMP1:%.*]] = insertelement <4 x i16> [[TMP0]], i16 [[C0]], i32 1
 ; CHECK-NEXT:    [[TMP2:%.*]] = insertelement <4 x i16> [[TMP1]], i16 [[E0]], i32 2
 ; CHECK-NEXT:    [[TMP3:%.*]] = insertelement <4 x i16> [[TMP2]], i16 [[D0]], i32 3
 ; CHECK-NEXT:    [[TMP4:%.*]] = sext <4 x i16> [[TMP3]] to <4 x i32>
 ; CHECK-NEXT:    [[TMP5:%.*]] = insertelement <4 x i32> poison, i32 [[V1]], i32 0
-; CHECK-NEXT:    [[TMP6:%.*]] = insertelement <4 x i32> [[TMP5]], i32 [[V1]], i32 1
-; CHECK-NEXT:    [[TMP7:%.*]] = insertelement <4 x i32> [[TMP6]], i32 [[V1]], i32 2
-; CHECK-NEXT:    [[TMP8:%.*]] = insertelement <4 x i32> [[TMP7]], i32 [[V1]], i32 3
-; CHECK-NEXT:    [[TMP9:%.*]] = add <4 x i32> [[TMP8]], [[TMP4]]
+; CHECK-NEXT:    [[SHUFFLE:%.*]] = shufflevector <4 x i32> [[TMP5]], <4 x i32> poison, <4 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP6:%.*]] = add <4 x i32> [[SHUFFLE]], [[TMP4]]
 ; CHECK-NEXT:    [[IDXS0:%.*]] = getelementptr inbounds i32, i32* [[S:%.*]], i64 0
 ; CHECK-NEXT:    [[IDXS1:%.*]] = getelementptr inbounds i32, i32* [[S]], i64 1
 ; CHECK-NEXT:    [[IDXS2:%.*]] = getelementptr inbounds i32, i32* [[S]], i64 2
 ; CHECK-NEXT:    [[IDXS3:%.*]] = getelementptr inbounds i32, i32* [[S]], i64 3
-; CHECK-NEXT:    [[TMP10:%.*]] = bitcast i32* [[IDXS0]] to <4 x i32>*
-; CHECK-NEXT:    store <4 x i32> [[TMP9]], <4 x i32>* [[TMP10]], align 8
+; CHECK-NEXT:    [[TMP7:%.*]] = bitcast i32* [[IDXS0]] to <4 x i32>*
+; CHECK-NEXT:    store <4 x i32> [[TMP6]], <4 x i32>* [[TMP7]], align 8
 ; CHECK-NEXT:    ret void
 ;
 entry:
   %A0 = load i16, i16 *%A, align 8
   %B0 = load i16, i16 *%B, align 8
   %C0 = load i16, i16 *%C, align 8
   %D0 = load i16, i16 *%D, align 8
   %E0 = load i16, i16 *%E, align 8

llvm/test/Transforms/SLPVectorizer/X86/commutativity.ll

 ; SSE-NEXT:    [[TMP15:%.*]] = xor i8 [[A]], [[C]]
 ; SSE-NEXT:    store i8 [[TMP15]], i8* getelementptr inbounds ([32 x i8], [32 x i8]* @cle, i64 0, i64 14), align 1
 ; SSE-NEXT:    [[TMP16:%.*]] = xor i8 [[A]], [[C]]
 ; SSE-NEXT:    store i8 [[TMP16]], i8* getelementptr inbounds ([32 x i8], [32 x i8]* @cle, i64 0, i64 15), align 1
 ; SSE-NEXT:    ret void
 ;
 ; AVX-LABEL: @splat(
 ; AVX-NEXT:    [[TMP1:%.*]] = insertelement <16 x i8> poison, i8 [[C:%.*]], i32 0
-; AVX-NEXT:    [[TMP2:%.*]] = insertelement <16 x i8> [[TMP1]], i8 [[C]], i32 1
-; AVX-NEXT:    [[TMP3:%.*]] = insertelement <16 x i8> [[TMP2]], i8 [[C]], i32 2
-; AVX-NEXT:    [[TMP4:%.*]] = insertelement <16 x i8> [[TMP3]], i8 [[C]], i32 3
-; AVX-NEXT:    [[TMP5:%.*]] = insertelement <16 x i8> [[TMP4]], i8 [[C]], i32 4
-; AVX-NEXT:    [[TMP6:%.*]] = insertelement <16 x i8> [[TMP5]], i8 [[C]], i32 5
-; AVX-NEXT:    [[TMP7:%.*]] = insertelement <16 x i8> [[TMP6]], i8 [[C]], i32 6
-; AVX-NEXT:    [[TMP8:%.*]] = insertelement <16 x i8> [[TMP7]], i8 [[C]], i32 7
-; AVX-NEXT:    [[TMP9:%.*]] = insertelement <16 x i8> [[TMP8]], i8 [[C]], i32 8
-; AVX-NEXT:    [[TMP10:%.*]] = insertelement <16 x i8> [[TMP9]], i8 [[C]], i32 9
-; AVX-NEXT:    [[TMP11:%.*]] = insertelement <16 x i8> [[TMP10]], i8 [[C]], i32 10
-; AVX-NEXT:    [[TMP12:%.*]] = insertelement <16 x i8> [[TMP11]], i8 [[C]], i32 11
-; AVX-NEXT:    [[TMP13:%.*]] = insertelement <16 x i8> [[TMP12]], i8 [[C]], i32 12
-; AVX-NEXT:    [[TMP14:%.*]] = insertelement <16 x i8> [[TMP13]], i8 [[C]], i32 13
-; AVX-NEXT:    [[TMP15:%.*]] = insertelement <16 x i8> [[TMP14]], i8 [[C]], i32 14
-; AVX-NEXT:    [[TMP16:%.*]] = insertelement <16 x i8> [[TMP15]], i8 [[C]], i32 15
-; AVX-NEXT:    [[TMP17:%.*]] = insertelement <16 x i8> poison, i8 [[A:%.*]], i32 0
-; AVX-NEXT:    [[TMP18:%.*]] = insertelement <16 x i8> [[TMP17]], i8 [[B:%.*]], i32 1
-; AVX-NEXT:    [[SHUFFLE:%.*]] = shufflevector <16 x i8> [[TMP18]], <16 x i8> poison, <16 x i32> <i32 0, i32 0, i32 0, i32 0, i32 0, i32 1, i32 0, i32 1, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0>
-; AVX-NEXT:    [[TMP19:%.*]] = xor <16 x i8> [[TMP16]], [[SHUFFLE]]
-; AVX-NEXT:    store <16 x i8> [[TMP19]], <16 x i8>* bitcast ([32 x i8]* @cle to <16 x i8>*), align 16
+; AVX-NEXT:    [[SHUFFLE:%.*]] = shufflevector <16 x i8> [[TMP1]], <16 x i8> poison, <16 x i32> zeroinitializer
+; AVX-NEXT:    [[TMP2:%.*]] = insertelement <16 x i8> poison, i8 [[A:%.*]], i32 0
+; AVX-NEXT:    [[TMP3:%.*]] = insertelement <16 x i8> [[TMP2]], i8 [[B:%.*]], i32 1
+; AVX-NEXT:    [[SHUFFLE1:%.*]] = shufflevector <16 x i8> [[TMP3]], <16 x i8> poison, <16 x i32> <i32 0, i32 0, i32 0, i32 0, i32 0, i32 1, i32 0, i32 1, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0>
+; AVX-NEXT:    [[TMP4:%.*]] = xor <16 x i8> [[SHUFFLE]], [[SHUFFLE1]]
+; AVX-NEXT:    store <16 x i8> [[TMP4]], <16 x i8>* bitcast ([32 x i8]* @cle to <16 x i8>*), align 16
 ; AVX-NEXT:    ret void
 ;
   %1 = xor i8 %c, %a
   store i8 %1, i8* getelementptr inbounds ([32 x i8], [32 x i8]* @cle, i64 0, i64 0), align 16
   %2 = xor i8 %a, %c
   store i8 %2, i8* getelementptr inbounds ([32 x i8], [32 x i8]* @cle, i64 0, i64 1)
   %3 = xor i8 %a, %c
   store i8 %3, i8* getelementptr inbounds ([32 x i8], [32 x i8]* @cle, i64 0, i64 2)

 ; SSE-NEXT:    [[TMP3:%.*]] = xor i32 [[C]], [[ADD3]]
 ; SSE-NEXT:    store i32 [[TMP3]], i32* getelementptr inbounds ([32 x i32], [32 x i32]* @cle32, i64 0, i64 2), align 4
 ; SSE-NEXT:    [[TMP4:%.*]] = xor i32 [[A]], [[ADD4]]
 ; SSE-NEXT:    store i32 [[TMP4]], i32* getelementptr inbounds ([32 x i32], [32 x i32]* @cle32, i64 0, i64 3), align 4
 ; SSE-NEXT:    ret void
 ;
 ; AVX-LABEL: @same_opcode_on_one_side(
 ; AVX-NEXT:    [[TMP1:%.*]] = insertelement <4 x i32> poison, i32 [[C:%.*]], i32 0
-; AVX-NEXT:    [[TMP2:%.*]] = insertelement <4 x i32> [[TMP1]], i32 [[C]], i32 1
-; AVX-NEXT:    [[TMP3:%.*]] = insertelement <4 x i32> [[TMP2]], i32 [[C]], i32 2
-; AVX-NEXT:    [[TMP4:%.*]] = insertelement <4 x i32> [[TMP3]], i32 [[C]], i32 3
-; AVX-NEXT:    [[TMP5:%.*]] = insertelement <4 x i32> poison, i32 [[A:%.*]], i32 0
-; AVX-NEXT:    [[TMP6:%.*]] = insertelement <4 x i32> [[TMP5]], i32 [[A]], i32 1
-; AVX-NEXT:    [[TMP7:%.*]] = insertelement <4 x i32> [[TMP6]], i32 [[A]], i32 2
-; AVX-NEXT:    [[TMP8:%.*]] = insertelement <4 x i32> [[TMP7]], i32 [[A]], i32 3
-; AVX-NEXT:    [[TMP9:%.*]] = add <4 x i32> [[TMP4]], [[TMP8]]
-; AVX-NEXT:    [[TMP10:%.*]] = insertelement <4 x i32> [[TMP5]], i32 [[B:%.*]], i32 1
-; AVX-NEXT:    [[TMP11:%.*]] = insertelement <4 x i32> [[TMP10]], i32 [[C]], i32 2
-; AVX-NEXT:    [[TMP12:%.*]] = insertelement <4 x i32> [[TMP11]], i32 [[A]], i32 3
-; AVX-NEXT:    [[TMP13:%.*]] = xor <4 x i32> [[TMP9]], [[TMP12]]
-; AVX-NEXT:    store <4 x i32> [[TMP13]], <4 x i32>* bitcast ([32 x i32]* @cle32 to <4 x i32>*), align 16
+; AVX-NEXT:    [[SHUFFLE:%.*]] = shufflevector <4 x i32> [[TMP1]], <4 x i32> poison, <4 x i32> zeroinitializer
+; AVX-NEXT:    [[TMP2:%.*]] = insertelement <4 x i32> poison, i32 [[A:%.*]], i32 0
+; AVX-NEXT:    [[SHUFFLE1:%.*]] = shufflevector <4 x i32> [[TMP2]], <4 x i32> poison, <4 x i32> zeroinitializer
+; AVX-NEXT:    [[TMP3:%.*]] = add <4 x i32> [[SHUFFLE]], [[SHUFFLE1]]
+; AVX-NEXT:    [[TMP4:%.*]] = insertelement <4 x i32> [[TMP2]], i32 [[B:%.*]], i32 1
+; AVX-NEXT:    [[TMP5:%.*]] = insertelement <4 x i32> [[TMP4]], i32 [[C]], i32 2
+; AVX-NEXT:    [[TMP6:%.*]] = insertelement <4 x i32> [[TMP5]], i32 [[A]], i32 3
+; AVX-NEXT:    [[TMP7:%.*]] = xor <4 x i32> [[TMP3]], [[TMP6]]
+; AVX-NEXT:    store <4 x i32> [[TMP7]], <4 x i32>* bitcast ([32 x i32]* @cle32 to <4 x i32>*), align 16
 ; AVX-NEXT:    ret void
 ;
   %add1 = add i32 %c, %a
   %add2 = add i32 %c, %a
   %add3 = add i32 %a, %c
   %add4 = add i32 %c, %a
   %1 = xor i32 %add1, %a
   store i32 %1, i32* getelementptr inbounds ([32 x i32], [32 x i32]* @cle32, i64 0, i64 0), align 16
   %2 = xor i32 %b, %add2
   store i32 %2, i32* getelementptr inbounds ([32 x i32], [32 x i32]* @cle32, i64 0, i64 1)
   %3 = xor i32 %c, %add3
   store i32 %3, i32* getelementptr inbounds ([32 x i32], [32 x i32]* @cle32, i64 0, i64 2)
   %4 = xor i32 %a, %add4
   store i32 %4, i32* getelementptr inbounds ([32 x i32], [32 x i32]* @cle32, i64 0, i64 3)
   ret void
 }

llvm/test/Transforms/SLPVectorizer/X86/cse.ll

 ; CHECK-NEXT:    [[CONV:%.*]] = sitofp i32 [[N:%.*]] to double
 ; CHECK-NEXT:    [[ARRAYIDX3:%.*]] = getelementptr inbounds double, double* [[A:%.*]], i64 1
 ; CHECK-NEXT:    [[ARRAYIDX9:%.*]] = getelementptr inbounds double, double* [[A]], i64 2
 ; CHECK-NEXT:    [[ARRAYIDX15:%.*]] = getelementptr inbounds double, double* [[A]], i64 3
 ; CHECK-NEXT:    [[TMP0:%.*]] = bitcast double* [[A]] to <4 x double>*
 ; CHECK-NEXT:    [[TMP1:%.*]] = load <4 x double>, <4 x double>* [[TMP0]], align 8
 ; CHECK-NEXT:    [[TMP2:%.*]] = fmul <4 x double> [[TMP1]], <double 7.900000e+00, double 7.700000e+00, double 7.600000e+00, double 7.400000e+00>
 ; CHECK-NEXT:    [[TMP3:%.*]] = insertelement <4 x double> poison, double [[CONV]], i32 0
-; CHECK-NEXT:    [[TMP4:%.*]] = insertelement <4 x double> [[TMP3]], double [[CONV]], i32 1
-; CHECK-NEXT:    [[TMP5:%.*]] = insertelement <4 x double> [[TMP4]], double [[CONV]], i32 2
-; CHECK-NEXT:    [[TMP6:%.*]] = insertelement <4 x double> [[TMP5]], double [[CONV]], i32 3
-; CHECK-NEXT:    [[TMP7:%.*]] = fmul <4 x double> [[TMP6]], [[TMP2]]
-; CHECK-NEXT:    [[TMP8:%.*]] = fadd <4 x double> [[TMP7]], <double 6.000000e+00, double 2.000000e+00, double 3.000000e+00, double 4.000000e+00>
-; CHECK-NEXT:    [[TMP9:%.*]] = bitcast double* [[A]] to <4 x double>*
-; CHECK-NEXT:    store <4 x double> [[TMP8]], <4 x double>* [[TMP9]], align 8
+; CHECK-NEXT:    [[SHUFFLE:%.*]] = shufflevector <4 x double> [[TMP3]], <4 x double> poison, <4 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP4:%.*]] = fmul <4 x double> [[SHUFFLE]], [[TMP2]]
+; CHECK-NEXT:    [[TMP5:%.*]] = fadd <4 x double> [[TMP4]], <double 6.000000e+00, double 2.000000e+00, double 3.000000e+00, double 4.000000e+00>
+; CHECK-NEXT:    [[TMP6:%.*]] = bitcast double* [[A]] to <4 x double>*
+; CHECK-NEXT:    store <4 x double> [[TMP5]], <4 x double>* [[TMP6]], align 8
 ; CHECK-NEXT:    ret i32 undef
 ;
 entry:
   %0 = load double, double* %A, align 8
   %mul = fmul double %0, 7.900000e+00
   %conv = sitofp i32 %n to double
   %mul1 = fmul double %conv, %mul
   %add = fadd double %mul1, 6.000000e+00

 ; CHECK-NEXT:    [[CONV:%.*]] = sitofp i32 [[N:%.*]] to double
 ; CHECK-NEXT:    [[ARRAYIDX3:%.*]] = getelementptr inbounds double, double* [[A:%.*]], i64 1
 ; CHECK-NEXT:    [[ARRAYIDX9:%.*]] = getelementptr inbounds double, double* [[A]], i64 2
 ; CHECK-NEXT:    [[ARRAYIDX15:%.*]] = getelementptr inbounds double, double* [[A]], i64 3
 ; CHECK-NEXT:    [[TMP0:%.*]] = bitcast double* [[A]] to <4 x double>*
 ; CHECK-NEXT:    [[TMP1:%.*]] = load <4 x double>, <4 x double>* [[TMP0]], align 8
 ; CHECK-NEXT:    [[TMP2:%.*]] = fmul <4 x double> [[TMP1]], <double 7.900000e+00, double 7.900000e+00, double 7.900000e+00, double 7.900000e+00>
 ; CHECK-NEXT:    [[TMP3:%.*]] = insertelement <4 x double> poison, double [[CONV]], i32 0
-; CHECK-NEXT:    [[TMP4:%.*]] = insertelement <4 x double> [[TMP3]], double [[CONV]], i32 1
-; CHECK-NEXT:    [[TMP5:%.*]] = insertelement <4 x double> [[TMP4]], double [[CONV]], i32 2
-; CHECK-NEXT:    [[TMP6:%.*]] = insertelement <4 x double> [[TMP5]], double [[CONV]], i32 3
-; CHECK-NEXT:    [[TMP7:%.*]] = fmul <4 x double> [[TMP6]], [[TMP2]]
-; CHECK-NEXT:    [[TMP8:%.*]] = fadd <4 x double> [[TMP7]], <double 6.000000e+00, double 6.000000e+00, double 6.000000e+00, double 6.000000e+00>
-; CHECK-NEXT:    [[TMP9:%.*]] = bitcast double* [[A]] to <4 x double>*
-; CHECK-NEXT:    store <4 x double> [[TMP8]], <4 x double>* [[TMP9]], align 8
+; CHECK-NEXT:    [[SHUFFLE:%.*]] = shufflevector <4 x double> [[TMP3]], <4 x double> poison, <4 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP4:%.*]] = fmul <4 x double> [[SHUFFLE]], [[TMP2]]
+; CHECK-NEXT:    [[TMP5:%.*]] = fadd <4 x double> [[TMP4]], <double 6.000000e+00, double 6.000000e+00, double 6.000000e+00, double 6.000000e+00>
+; CHECK-NEXT:    [[TMP6:%.*]] = bitcast double* [[A]] to <4 x double>*
+; CHECK-NEXT:    store <4 x double> [[TMP5]], <4 x double>* [[TMP6]], align 8
 ; CHECK-NEXT:    ret i32 undef
 ;
 entry:
   %0 = load double, double* %A, align 8
   %mul = fmul double %0, 7.900000e+00
   %conv = sitofp i32 %n to double
   %mul1 = fmul double %conv, %mul
   %add = fadd double %mul1, 6.000000e+00

llvm/test/Transforms/SLPVectorizer/X86/diamond.ll

 ; CHECK-NEXT:    [[ARRAYIDX4:%.*]] = getelementptr inbounds i32, i32* [[A:%.*]], i64 1
 ; CHECK-NEXT:    [[ARRAYIDX9:%.*]] = getelementptr inbounds i32, i32* [[B:%.*]], i64 1
 ; CHECK-NEXT:    [[ARRAYIDX10:%.*]] = getelementptr inbounds i32, i32* [[A]], i64 2
 ; CHECK-NEXT:    [[ARRAYIDX15:%.*]] = getelementptr inbounds i32, i32* [[B]], i64 2
 ; CHECK-NEXT:    [[ARRAYIDX16:%.*]] = getelementptr inbounds i32, i32* [[A]], i64 3
 ; CHECK-NEXT:    [[TMP0:%.*]] = bitcast i32* [[A]] to <4 x i32>*
 ; CHECK-NEXT:    [[TMP1:%.*]] = load <4 x i32>, <4 x i32>* [[TMP0]], align 4
 ; CHECK-NEXT:    [[TMP2:%.*]] = insertelement <4 x i32> poison, i32 [[MUL238]], i32 0
-; CHECK-NEXT:    [[TMP3:%.*]] = insertelement <4 x i32> [[TMP2]], i32 [[MUL238]], i32 1
-; CHECK-NEXT:    [[TMP4:%.*]] = insertelement <4 x i32> [[TMP3]], i32 [[MUL238]], i32 2
-; CHECK-NEXT:    [[TMP5:%.*]] = insertelement <4 x i32> [[TMP4]], i32 [[MUL238]], i32 3
-; CHECK-NEXT:    [[TMP6:%.*]] = mul <4 x i32> [[TMP1]], [[TMP5]]
+; CHECK-NEXT:    [[SHUFFLE:%.*]] = shufflevector <4 x i32> [[TMP2]], <4 x i32> poison, <4 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP3:%.*]] = mul <4 x i32> [[TMP1]], [[SHUFFLE]]
 ; CHECK-NEXT:    [[ARRAYIDX21:%.*]] = getelementptr inbounds i32, i32* [[B]], i64 3
-; CHECK-NEXT:    [[TMP7:%.*]] = bitcast i32* [[B]] to <4 x i32>*
-; CHECK-NEXT:    store <4 x i32> [[TMP6]], <4 x i32>* [[TMP7]], align 4
+; CHECK-NEXT:    [[TMP4:%.*]] = bitcast i32* [[B]] to <4 x i32>*
+; CHECK-NEXT:    store <4 x i32> [[TMP3]], <4 x i32>* [[TMP4]], align 4
 ; CHECK-NEXT:    ret i32 0
 ;
 entry:
   %0 = load i32, i32* %A, align 4
   %mul238 = add i32 %m, %n
   %add = mul i32 %0, %mul238
   store i32 %add, i32* %B, align 4
   %arrayidx4 = getelementptr inbounds i32, i32* %A, i64 1

 ; CHECK-NEXT:    [[ARRAYIDX4:%.*]] = getelementptr inbounds i32, i32* [[A:%.*]], i64 1
 ; CHECK-NEXT:    [[ARRAYIDX9:%.*]] = getelementptr inbounds i32, i32* [[B:%.*]], i64 1
 ; CHECK-NEXT:    [[ARRAYIDX10:%.*]] = getelementptr inbounds i32, i32* [[A]], i64 2
 ; CHECK-NEXT:    [[ARRAYIDX15:%.*]] = getelementptr inbounds i32, i32* [[B]], i64 2
 ; CHECK-NEXT:    [[ARRAYIDX16:%.*]] = getelementptr inbounds i32, i32* [[A]], i64 3
 ; CHECK-NEXT:    [[TMP0:%.*]] = bitcast i32* [[A]] to <4 x i32>*
 ; CHECK-NEXT:    [[TMP1:%.*]] = load <4 x i32>, <4 x i32>* [[TMP0]], align 4
 ; CHECK-NEXT:    [[TMP2:%.*]] = insertelement <4 x i32> poison, i32 [[MUL238]], i32 0
-; CHECK-NEXT:    [[TMP3:%.*]] = insertelement <4 x i32> [[TMP2]], i32 [[MUL238]], i32 1
-; CHECK-NEXT:    [[TMP4:%.*]] = insertelement <4 x i32> [[TMP3]], i32 [[MUL238]], i32 2
-; CHECK-NEXT:    [[TMP5:%.*]] = insertelement <4 x i32> [[TMP4]], i32 [[MUL238]], i32 3
-; CHECK-NEXT:    [[TMP6:%.*]] = mul <4 x i32> [[TMP1]], [[TMP5]]
+; CHECK-NEXT:    [[SHUFFLE:%.*]] = shufflevector <4 x i32> [[TMP2]], <4 x i32> poison, <4 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP3:%.*]] = mul <4 x i32> [[TMP1]], [[SHUFFLE]]
 ; CHECK-NEXT:    [[ARRAYIDX21:%.*]] = getelementptr inbounds i32, i32* [[B]], i64 3
-; CHECK-NEXT:    [[TMP7:%.*]] = bitcast i32* [[B]] to <4 x i32>*
-; CHECK-NEXT:    store <4 x i32> [[TMP6]], <4 x i32>* [[TMP7]], align 4
-; CHECK-NEXT:    [[TMP8:%.*]] = extractelement <4 x i32> [[TMP1]], i32 0
-; CHECK-NEXT:    ret i32 [[TMP8]]
+; CHECK-NEXT:    [[TMP4:%.*]] = bitcast i32* [[B]] to <4 x i32>*
+; CHECK-NEXT:    store <4 x i32> [[TMP3]], <4 x i32>* [[TMP4]], align 4
+; CHECK-NEXT:    [[TMP5:%.*]] = extractelement <4 x i32> [[TMP1]], i32 0
+; CHECK-NEXT:    ret i32 [[TMP5]]
 ;
 entry:
   %0 = load i32, i32* %A, align 4
   %mul238 = add i32 %m, %n
   %add = mul i32 %0, %mul238
   store i32 %add, i32* %B, align 4
   %arrayidx4 = getelementptr inbounds i32, i32* %A, i64 1
   %1 = load i32, i32* %arrayidx4, align 4

 ; CHECK-NEXT:    [[ARRAYIDX4:%.*]] = getelementptr inbounds i32, i32* [[A:%.*]], i64 1
 ; CHECK-NEXT:    [[ARRAYIDX9:%.*]] = getelementptr inbounds i32, i32* [[B:%.*]], i64 1
 ; CHECK-NEXT:    [[ARRAYIDX10:%.*]] = getelementptr inbounds i32, i32* [[A]], i64 2
 ; CHECK-NEXT:    [[ARRAYIDX15:%.*]] = getelementptr inbounds i32, i32* [[B]], i64 2
 ; CHECK-NEXT:    [[ARRAYIDX16:%.*]] = getelementptr inbounds i32, i32* [[A]], i64 3
 ; CHECK-NEXT:    [[TMP0:%.*]] = bitcast i32* [[A]] to <4 x i32>*
 ; CHECK-NEXT:    [[TMP1:%.*]] = load <4 x i32>, <4 x i32>* [[TMP0]], align 4
 ; CHECK-NEXT:    [[TMP2:%.*]] = insertelement <4 x i32> poison, i32 [[MUL238]], i32 0
-; CHECK-NEXT:    [[TMP3:%.*]] = insertelement <4 x i32> [[TMP2]], i32 [[MUL238]], i32 1
-; CHECK-NEXT:    [[TMP4:%.*]] = insertelement <4 x i32> [[TMP3]], i32 [[MUL238]], i32 2
-; CHECK-NEXT:    [[TMP5:%.*]] = insertelement <4 x i32> [[TMP4]], i32 [[MUL238]], i32 3
-; CHECK-NEXT:    [[TMP6:%.*]] = mul <4 x i32> [[TMP1]], [[TMP5]]
+; CHECK-NEXT:    [[SHUFFLE:%.*]] = shufflevector <4 x i32> [[TMP2]], <4 x i32> poison, <4 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP3:%.*]] = mul <4 x i32> [[TMP1]], [[SHUFFLE]]
 ; CHECK-NEXT:    [[ARRAYIDX21:%.*]] = getelementptr inbounds i32, i32* [[B]], i64 3
-; CHECK-NEXT:    [[TMP7:%.*]] = bitcast i32* [[B]] to <4 x i32>*
-; CHECK-NEXT:    store <4 x i32> [[TMP6]], <4 x i32>* [[TMP7]], align 4
-; CHECK-NEXT:    [[TMP8:%.*]] = extractelement <4 x i32> [[TMP1]], i32 1
-; CHECK-NEXT:    ret i32 [[TMP8]]
+; CHECK-NEXT:    [[TMP4:%.*]] = bitcast i32* [[B]] to <4 x i32>*
+; CHECK-NEXT:    store <4 x i32> [[TMP3]], <4 x i32>* [[TMP4]], align 4
+; CHECK-NEXT:    [[TMP5:%.*]] = extractelement <4 x i32> [[TMP1]], i32 1
+; CHECK-NEXT:    ret i32 [[TMP5]]
 ;
 entry:
   %0 = load i32, i32* %A, align 4
   %mul238 = add i32 %m, %n
   %add = mul i32 %0, %mul238
   store i32 %add, i32* %B, align 4
   %arrayidx4 = getelementptr inbounds i32, i32* %A, i64 1
   %1 = load i32, i32* %arrayidx4, align 4

llvm/test/Transforms/SLPVectorizer/X86/diamond_broadcast.ll

 ; NOTE: Assertions have been autogenerated by utils/update_test_checks.py
 ; RUN: opt < %s -slp-vectorizer -S -mtriple=x86_64-unknown-linux -slp-threshold=-1 | FileCheck %s
 
 define i32 @diamond_broadcast(i32* noalias nocapture %B, i32* noalias nocapture %A) {
 ; CHECK-LABEL: @diamond_broadcast(
 ; CHECK-NEXT:  entry:
 ; CHECK-NEXT:    [[LD:%.*]] = load i32, i32* [[A:%.*]], align 4
 ; CHECK-NEXT:    [[ARRAYIDX9:%.*]] = getelementptr inbounds i32, i32* [[B:%.*]], i64 1
 ; CHECK-NEXT:    [[ARRAYIDX15:%.*]] = getelementptr inbounds i32, i32* [[B]], i64 2
 ; CHECK-NEXT:    [[TMP0:%.*]] = insertelement <4 x i32> poison, i32 [[LD]], i32 0
-; CHECK-NEXT:    [[TMP1:%.*]] = insertelement <4 x i32> [[TMP0]], i32 [[LD]], i32 1
-; CHECK-NEXT:    [[TMP2:%.*]] = insertelement <4 x i32> [[TMP1]], i32 [[LD]], i32 2
-; CHECK-NEXT:    [[TMP3:%.*]] = insertelement <4 x i32> [[TMP2]], i32 [[LD]], i32 3
-; CHECK-NEXT:    [[TMP4:%.*]] = mul <4 x i32> [[TMP3]], [[TMP3]]
+; CHECK-NEXT:    [[SHUFFLE:%.*]] = shufflevector <4 x i32> [[TMP0]], <4 x i32> poison, <4 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP1:%.*]] = mul <4 x i32> [[SHUFFLE]], [[SHUFFLE]]
 ; CHECK-NEXT:    [[ARRAYIDX21:%.*]] = getelementptr inbounds i32, i32* [[B]], i64 3
-; CHECK-NEXT:    [[TMP5:%.*]] = bitcast i32* [[B]] to <4 x i32>*
-; CHECK-NEXT:    store <4 x i32> [[TMP4]], <4 x i32>* [[TMP5]], align 4
+; CHECK-NEXT:    [[TMP2:%.*]] = bitcast i32* [[B]] to <4 x i32>*
+; CHECK-NEXT:    store <4 x i32> [[TMP1]], <4 x i32>* [[TMP2]], align 4
 ; CHECK-NEXT:    ret i32 0
 ;
 entry:
   %ld = load i32, i32* %A, align 4
   %mul = mul i32 %ld, %ld
   store i32 %mul, i32* %B, align 4
   %mul8 = mul i32 %ld, %ld
   %arrayidx9 = getelementptr inbounds i32, i32* %B, i64 1

llvm/test/Transforms/SLPVectorizer/X86/extractcost.ll

 ; NOTE: Assertions have been autogenerated by utils/update_test_checks.py
 ; RUN: opt < %s -basic-aa -slp-vectorizer -dce -S -mtriple=x86_64-apple-macosx10.8.0 -mcpu=corei7-avx | FileCheck %s
 
 target datalayout = "e-p:64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:64:64-f32:32:32-f64:64:64-v64:64:64-v128:128:128-a0:0:64-s0:64:64-f80:128:128-n8:16:32:64-S128"
 target triple = "x86_64-apple-macosx10.8.0"
 
 define i32 @foo(i32* nocapture %A, i32 %n, i32 %m) {
 ; CHECK-LABEL: @foo(
 ; CHECK-NEXT:  entry:
 ; CHECK-NEXT:    [[TMP0:%.*]] = insertelement <4 x i32> poison, i32 [[N:%.*]], i32 0
-; CHECK-NEXT:    [[TMP1:%.*]] = insertelement <4 x i32> [[TMP0]], i32 [[N]], i32 1
-; CHECK-NEXT:    [[TMP2:%.*]] = insertelement <4 x i32> [[TMP1]], i32 [[N]], i32 2
-; CHECK-NEXT:    [[TMP3:%.*]] = insertelement <4 x i32> [[TMP2]], i32 [[N]], i32 3
-; CHECK-NEXT:    [[TMP4:%.*]] = mul nsw <4 x i32> [[TMP3]], <i32 5, i32 9, i32 3, i32 10>
-; CHECK-NEXT:    [[TMP5:%.*]] = shl <4 x i32> [[TMP3]], <i32 5, i32 9, i32 3, i32 10>
-; CHECK-NEXT:    [[TMP6:%.*]] = shufflevector <4 x i32> [[TMP4]], <4 x i32> [[TMP5]], <4 x i32> <i32 0, i32 1, i32 6, i32 3>
-; CHECK-NEXT:    [[TMP7:%.*]] = add nsw <4 x i32> [[TMP6]], <i32 9, i32 9, i32 9, i32 9>
-; CHECK-NEXT:    [[TMP8:%.*]] = bitcast i32* [[A:%.*]] to <4 x i32>*
-; CHECK-NEXT:    store <4 x i32> [[TMP7]], <4 x i32>* [[TMP8]], align 4
-; CHECK-NEXT:    [[TMP9:%.*]] = extractelement <4 x i32> [[TMP7]], i32 0
-; CHECK-NEXT:    [[EXTERNALUSE1:%.*]] = add nsw i32 [[TMP9]], [[M:%.*]]
-; CHECK-NEXT:    [[EXTERNALUSE2:%.*]] = mul nsw i32 [[TMP9]], [[M]]
+; CHECK-NEXT:    [[SHUFFLE:%.*]] = shufflevector <4 x i32> [[TMP0]], <4 x i32> poison, <4 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP1:%.*]] = mul nsw <4 x i32> [[SHUFFLE]], <i32 5, i32 9, i32 3, i32 10>
+; CHECK-NEXT:    [[TMP2:%.*]] = shl <4 x i32> [[SHUFFLE]], <i32 5, i32 9, i32 3, i32 10>
+; CHECK-NEXT:    [[TMP3:%.*]] = shufflevector <4 x i32> [[TMP1]], <4 x i32> [[TMP2]], <4 x i32> <i32 0, i32 1, i32 6, i32 3>
+; CHECK-NEXT:    [[TMP4:%.*]] = add nsw <4 x i32> [[TMP3]], <i32 9, i32 9, i32 9, i32 9>
+; CHECK-NEXT:    [[TMP5:%.*]] = bitcast i32* [[A:%.*]] to <4 x i32>*
+; CHECK-NEXT:    store <4 x i32> [[TMP4]], <4 x i32>* [[TMP5]], align 4
+; CHECK-NEXT:    [[TMP6:%.*]] = extractelement <4 x i32> [[TMP4]], i32 0
+; CHECK-NEXT:    [[EXTERNALUSE1:%.*]] = add nsw i32 [[TMP6]], [[M:%.*]]
+; CHECK-NEXT:    [[EXTERNALUSE2:%.*]] = mul nsw i32 [[TMP6]], [[M]]
 ; CHECK-NEXT:    [[ADD10:%.*]] = add nsw i32 [[EXTERNALUSE1]], [[EXTERNALUSE2]]
 ; CHECK-NEXT:    ret i32 [[ADD10]]
 ;
 entry:
   %mul = mul nsw i32 %n, 5
   %add = add nsw i32 %mul, 9
   store i32 %add, i32* %A, align 4
   %mul1 = mul nsw i32 %n, 9

llvm/test/Transforms/SLPVectorizer/X86/horizontal-list.ll

;
   %add5 = fadd fast float %add4.2, %a
   ret float %add5
 }
 
 define i32 @wobble(i32 %arg, i32 %bar) {
 ; CHECK-LABEL: @wobble(
 ; CHECK-NEXT:  bb:
 ; CHECK-NEXT:    [[TMP0:%.*]] = insertelement <4 x i32> poison, i32 [[ARG:%.*]], i32 0
-; CHECK-NEXT:    [[TMP1:%.*]] = insertelement <4 x i32> [[TMP0]], i32 [[ARG]], i32 1
-; CHECK-NEXT:    [[TMP2:%.*]] = insertelement <4 x i32> [[TMP1]], i32 [[ARG]], i32 2
-; CHECK-NEXT:    [[TMP3:%.*]] = insertelement <4 x i32> [[TMP2]], i32 [[ARG]], i32 3
-; CHECK-NEXT:    [[TMP4:%.*]] = insertelement <4 x i32> poison, i32 [[BAR:%.*]], i32 0
-; CHECK-NEXT:    [[TMP5:%.*]] = insertelement <4 x i32> [[TMP4]], i32 [[BAR]], i32 1
-; CHECK-NEXT:    [[TMP6:%.*]] = insertelement <4 x i32> [[TMP5]], i32 [[BAR]], i32 2
-; CHECK-NEXT:    [[TMP7:%.*]] = insertelement <4 x i32> [[TMP6]], i32 [[BAR]], i32 3
-; CHECK-NEXT:    [[TMP8:%.*]] = xor <4 x i32> [[TMP3]], [[TMP7]]
-; CHECK-NEXT:    [[TMP9:%.*]] = extractelement <4 x i32> [[TMP8]], i32 3
-; CHECK-NEXT:    [[TMP10:%.*]] = icmp eq <4 x i32> [[TMP8]], zeroinitializer
-; CHECK-NEXT:    [[TMP11:%.*]] = sext <4 x i1> [[TMP10]] to <4 x i32>
-; CHECK-NEXT:    [[TMP12:%.*]] = call i32 @llvm.vector.reduce.add.v4i32(<4 x i32> [[TMP11]])
-; CHECK-NEXT:    [[OP_EXTRA:%.*]] = add nuw i32 [[TMP12]], [[ARG]]
-; CHECK-NEXT:    [[OP_EXTRA1:%.*]] = add nsw i32 [[OP_EXTRA]], [[TMP9]]
-; CHECK-NEXT:    ret i32 [[OP_EXTRA1]]
+; CHECK-NEXT:    [[SHUFFLE:%.*]] = shufflevector <4 x i32> [[TMP0]], <4 x i32> poison, <4 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP1:%.*]] = insertelement <4 x i32> poison, i32 [[BAR:%.*]], i32 0
+; CHECK-NEXT:    [[SHUFFLE1:%.*]] = shufflevector <4 x i32> [[TMP1]], <4 x i32> poison, <4 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP2:%.*]] = xor <4 x i32> [[SHUFFLE]], [[SHUFFLE1]]
+; CHECK-NEXT:    [[TMP3:%.*]] = extractelement <4 x i32> [[TMP2]], i32 3
+; CHECK-NEXT:    [[TMP4:%.*]] = icmp eq <4 x i32> [[TMP2]], zeroinitializer
+; CHECK-NEXT:    [[TMP5:%.*]] = sext <4 x i1> [[TMP4]] to <4 x i32>
+; CHECK-NEXT:    [[TMP6:%.*]] = call i32 @llvm.vector.reduce.add.v4i32(<4 x i32> [[TMP5]])
+; CHECK-NEXT:    [[OP_EXTRA:%.*]] = add nuw i32 [[TMP6]], [[ARG]]
+; CHECK-NEXT:    [[OP_EXTRA2:%.*]] = add nsw i32 [[OP_EXTRA]], [[TMP3]]
+; CHECK-NEXT:    ret i32 [[OP_EXTRA2]]
 ;
 ; THRESHOLD-LABEL: @wobble(
 ; THRESHOLD-NEXT:  bb:
 ; THRESHOLD-NEXT:    [[TMP0:%.*]] = insertelement <4 x i32> poison, i32 [[ARG:%.*]], i32 0
-; THRESHOLD-NEXT:    [[TMP1:%.*]] = insertelement <4 x i32> [[TMP0]], i32 [[ARG]], i32 1
-; THRESHOLD-NEXT:    [[TMP2:%.*]] = insertelement <4 x i32> [[TMP1]], i32 [[ARG]], i32 2
-; THRESHOLD-NEXT:    [[TMP3:%.*]] = insertelement <4 x i32> [[TMP2]], i32 [[ARG]], i32 3
-; THRESHOLD-NEXT:    [[TMP4:%.*]] = insertelement <4 x i32> poison, i32 [[BAR:%.*]], i32 0
-; THRESHOLD-NEXT:    [[TMP5:%.*]] = insertelement <4 x i32> [[TMP4]], i32 [[BAR]], i32 1
-; THRESHOLD-NEXT:    [[TMP6:%.*]] = insertelement <4 x i32> [[TMP5]], i32 [[BAR]], i32 2
-; THRESHOLD-NEXT:    [[TMP7:%.*]] = insertelement <4 x i32> [[TMP6]], i32 [[BAR]], i32 3
-; THRESHOLD-NEXT:    [[TMP8:%.*]] = xor <4 x i32> [[TMP3]], [[TMP7]]
-; THRESHOLD-NEXT:    [[TMP9:%.*]] = extractelement <4 x i32> [[TMP8]], i32 3
-; THRESHOLD-NEXT:    [[TMP10:%.*]] = icmp eq <4 x i32> [[TMP8]], zeroinitializer
-; THRESHOLD-NEXT:    [[TMP11:%.*]] = sext <4 x i1> [[TMP10]] to <4 x i32>
-; THRESHOLD-NEXT:    [[TMP12:%.*]] = call i32 @llvm.vector.reduce.add.v4i32(<4 x i32> [[TMP11]])
-; THRESHOLD-NEXT:    [[OP_EXTRA:%.*]] = add nuw i32 [[TMP12]], [[ARG]]
-; THRESHOLD-NEXT:    [[OP_EXTRA1:%.*]] = add nsw i32 [[OP_EXTRA]], [[TMP9]]
-; THRESHOLD-NEXT:    ret i32 [[OP_EXTRA1]]
+; THRESHOLD-NEXT:    [[SHUFFLE:%.*]] = shufflevector <4 x i32> [[TMP0]], <4 x i32> poison, <4 x i32> zeroinitializer
+; THRESHOLD-NEXT:    [[TMP1:%.*]] = insertelement <4 x i32> poison, i32 [[BAR:%.*]], i32 0
+; THRESHOLD-NEXT:    [[SHUFFLE1:%.*]] = shufflevector <4 x i32> [[TMP1]], <4 x i32> poison, <4 x i32> zeroinitializer
+; THRESHOLD-NEXT:    [[TMP2:%.*]] = xor <4 x i32> [[SHUFFLE]], [[SHUFFLE1]]
+; THRESHOLD-NEXT:    [[TMP3:%.*]] = extractelement <4 x i32> [[TMP2]], i32 3
+; THRESHOLD-NEXT:    [[TMP4:%.*]] = icmp eq <4 x i32> [[TMP2]], zeroinitializer
+; THRESHOLD-NEXT:    [[TMP5:%.*]] = sext <4 x i1> [[TMP4]] to <4 x i32>
+; THRESHOLD-NEXT:    [[TMP6:%.*]] = call i32 @llvm.vector.reduce.add.v4i32(<4 x i32> [[TMP5]])
+; THRESHOLD-NEXT:    [[OP_EXTRA:%.*]] = add nuw i32 [[TMP6]], [[ARG]]
+; THRESHOLD-NEXT:    [[OP_EXTRA2:%.*]] = add nsw i32 [[OP_EXTRA]], [[TMP3]]
+; THRESHOLD-NEXT:    ret i32 [[OP_EXTRA2]]
 ;
   bb:
   %x1 = xor i32 %arg, %bar
   %i1 = icmp eq i32 %x1, 0
   %s1 = sext i1 %i1 to i32
   %x2 = xor i32 %arg, %bar
   %i2 = icmp eq i32 %x2, 0
   %s2 = sext i1 %i2 to i32

llvm/test/Transforms/SLPVectorizer/X86/insert-after-bundle.ll

entry:
   %retval.0 = trunc i32 %b.a to i8
   ret i8 %retval.0
 }
 
 define void @bar(i8* noalias nocapture readonly %a, i8* noalias nocapture readonly %b, i8* noalias nocapture readonly %c, i8* noalias nocapture readonly %d, i8* noalias nocapture %e, i32 %w) local_unnamed_addr #1 {
 ; SSE-LABEL: @bar(
 ; SSE-NEXT:  entry:
 ; SSE-NEXT:    [[TMP0:%.*]] = insertelement <4 x i32> poison, i32 [[W:%.*]], i32 0
-; SSE-NEXT:    [[TMP1:%.*]] = insertelement <4 x i32> [[TMP0]], i32 [[W]], i32 1
-; SSE-NEXT:    [[TMP2:%.*]] = insertelement <4 x i32> [[TMP1]], i32 [[W]], i32 2
-; SSE-NEXT:    [[TMP3:%.*]] = insertelement <4 x i32> [[TMP2]], i32 [[W]], i32 3
-; SSE-NEXT:    [[TMP4:%.*]] = insertelement <4 x i32> poison, i32 [[W]], i32 0
-; SSE-NEXT:    [[TMP5:%.*]] = insertelement <4 x i32> [[TMP4]], i32 [[W]], i32 1
-; SSE-NEXT:    [[TMP6:%.*]] = insertelement <4 x i32> [[TMP5]], i32 [[W]], i32 2
-; SSE-NEXT:    [[TMP7:%.*]] = insertelement <4 x i32> [[TMP6]], i32 [[W]], i32 3
-; SSE-NEXT:    [[TMP8:%.*]] = insertelement <4 x i32> poison, i32 [[W]], i32 0
-; SSE-NEXT:    [[TMP9:%.*]] = insertelement <4 x i32> [[TMP8]], i32 [[W]], i32 1
-; SSE-NEXT:    [[TMP10:%.*]] = insertelement <4 x i32> [[TMP9]], i32 [[W]], i32 2
-; SSE-NEXT:    [[TMP11:%.*]] = insertelement <4 x i32> [[TMP10]], i32 [[W]], i32 3
-; SSE-NEXT:    [[TMP12:%.*]] = insertelement <4 x i32> poison, i32 [[W]], i32 0
-; SSE-NEXT:    [[TMP13:%.*]] = insertelement <4 x i32> [[TMP12]], i32 [[W]], i32 1
-; SSE-NEXT:    [[TMP14:%.*]] = insertelement <4 x i32> [[TMP13]], i32 [[W]], i32 2
-; SSE-NEXT:    [[TMP15:%.*]] = insertelement <4 x i32> [[TMP14]], i32 [[W]], i32 3
+; SSE-NEXT:    [[SHUFFLE:%.*]] = shufflevector <4 x i32> [[TMP0]], <4 x i32> poison, <4 x i32> zeroinitializer
+; SSE-NEXT:    [[TMP1:%.*]] = insertelement <4 x i32> poison, i32 [[W]], i32 0
+; SSE-NEXT:    [[SHUFFLE1:%.*]] = shufflevector <4 x i32> [[TMP1]], <4 x i32> poison, <4 x i32> zeroinitializer
+; SSE-NEXT:    [[TMP2:%.*]] = insertelement <4 x i32> poison, i32 [[W]], i32 0
+; SSE-NEXT:    [[SHUFFLE2:%.*]] = shufflevector <4 x i32> [[TMP2]], <4 x i32> poison, <4 x i32> zeroinitializer
+; SSE-NEXT:    [[TMP3:%.*]] = insertelement <4 x i32> poison, i32 [[W]], i32 0
+; SSE-NEXT:    [[SHUFFLE3:%.*]] = shufflevector <4 x i32> [[TMP3]], <4 x i32> poison, <4 x i32> zeroinitializer
 ; SSE-NEXT:    br label [[FOR_BODY:%.*]]
 ; SSE:       for.body:
 ; SSE-NEXT:    [[I_0356:%.*]] = phi i32 [ 0, [[ENTRY:%.*]] ], [ [[INC:%.*]], [[FOR_BODY]] ]
 ; SSE-NEXT:    [[A_ADDR_0355:%.*]] = phi i8* [ [[A:%.*]], [[ENTRY]] ], [ [[ADD_PTR:%.*]], [[FOR_BODY]] ]
 ; SSE-NEXT:    [[E_ADDR_0354:%.*]] = phi i8* [ [[E:%.*]], [[ENTRY]] ], [ [[ADD_PTR192:%.*]], [[FOR_BODY]] ]
 ; SSE-NEXT:    [[D_ADDR_0353:%.*]] = phi i8* [ [[D:%.*]], [[ENTRY]] ], [ [[ADD_PTR191:%.*]], [[FOR_BODY]] ]
 ; SSE-NEXT:    [[C_ADDR_0352:%.*]] = phi i8* [ [[C:%.*]], [[ENTRY]] ], [ [[ADD_PTR190:%.*]], [[FOR_BODY]] ]
 ; SSE-NEXT:    [[B_ADDR_0351:%.*]] = phi i8* [ [[B:%.*]], [[ENTRY]] ], [ [[ADD_PTR189:%.*]], [[FOR_BODY]] ]
 ; SSE-NEXT:    [[ARRAYIDX9:%.*]] = getelementptr inbounds i8, i8* [[C_ADDR_0352]], i64 1
 ; SSE-NEXT:    [[ARRAYIDX11:%.*]] = getelementptr inbounds i8, i8* [[D_ADDR_0353]], i64 1
 ; SSE-NEXT:    [[ARRAYIDX13:%.*]] = getelementptr inbounds i8, i8* [[A_ADDR_0355]], i64 1
 ; SSE-NEXT:    [[ARRAYIDX16:%.*]] = getelementptr inbounds i8, i8* [[B_ADDR_0351]], i64 1
 ; SSE-NEXT:    [[ARRAYIDX20:%.*]] = getelementptr inbounds i8, i8* [[E_ADDR_0354]], i64 1
 ; SSE-NEXT:    [[ARRAYIDX21:%.*]] = getelementptr inbounds i8, i8* [[C_ADDR_0352]], i64 2
 ; SSE-NEXT:    [[ARRAYIDX23:%.*]] = getelementptr inbounds i8, i8* [[D_ADDR_0353]], i64 2
 ; SSE-NEXT:    [[ARRAYIDX25:%.*]] = getelementptr inbounds i8, i8* [[A_ADDR_0355]], i64 2
 ; SSE-NEXT:    [[ARRAYIDX28:%.*]] = getelementptr inbounds i8, i8* [[B_ADDR_0351]], i64 2
 ; SSE-NEXT:    [[ARRAYIDX32:%.*]] = getelementptr inbounds i8, i8* [[E_ADDR_0354]], i64 2
 ; SSE-NEXT:    [[ARRAYIDX33:%.*]] = getelementptr inbounds i8, i8* [[C_ADDR_0352]], i64 3
-; SSE-NEXT:    [[TMP16:%.*]] = bitcast i8* [[C_ADDR_0352]] to <4 x i8>*
-; SSE-NEXT:    [[TMP17:%.*]] = load <4 x i8>, <4 x i8>* [[TMP16]], align 1
+; SSE-NEXT:    [[TMP4:%.*]] = bitcast i8* [[C_ADDR_0352]] to <4 x i8>*
+; SSE-NEXT:    [[TMP5:%.*]] = load <4 x i8>, <4 x i8>* [[TMP4]], align 1
 ; SSE-NEXT:    [[ARRAYIDX35:%.*]] = getelementptr inbounds i8, i8* [[D_ADDR_0353]], i64 3
-; SSE-NEXT:    [[TMP18:%.*]] = bitcast i8* [[D_ADDR_0353]] to <4 x i8>*
-; SSE-NEXT:    [[TMP19:%.*]] = load <4 x i8>, <4 x i8>* [[TMP18]], align 1
+; SSE-NEXT:    [[TMP6:%.*]] = bitcast i8* [[D_ADDR_0353]] to <4 x i8>*
+; SSE-NEXT:    [[TMP7:%.*]] = load <4 x i8>, <4 x i8>* [[TMP6]], align 1
 ; SSE-NEXT:    [[ARRAYIDX37:%.*]] = getelementptr inbounds i8, i8* [[A_ADDR_0355]], i64 3
-; SSE-NEXT:    [[TMP20:%.*]] = bitcast i8* [[A_ADDR_0355]] to <4 x i8>*
-; SSE-NEXT:    [[TMP21:%.*]] = load <4 x i8>, <4 x i8>* [[TMP20]], align 1
+; SSE-NEXT:    [[TMP8:%.*]] = bitcast i8* [[A_ADDR_0355]] to <4 x i8>*
+; SSE-NEXT:    [[TMP9:%.*]] = load <4 x i8>, <4 x i8>* [[TMP8]], align 1
 ; SSE-NEXT:    [[ARRAYIDX40:%.*]] = getelementptr inbounds i8, i8* [[B_ADDR_0351]], i64 3
-; SSE-NEXT:    [[TMP22:%.*]] = bitcast i8* [[B_ADDR_0351]] to <4 x i8>*
-; SSE-NEXT:    [[TMP23:%.*]] = load <4 x i8>, <4 x i8>* [[TMP22]], align 1
-; SSE-NEXT:    [[TMP24:%.*]] = icmp ult <4 x i8> [[TMP17]], [[TMP19]]
-; SSE-NEXT:    [[TMP25:%.*]] = select <4 x i1> [[TMP24]], <4 x i8> [[TMP23]], <4 x i8> [[TMP21]]
-; SSE-NEXT:    [[TMP26:%.*]] = zext <4 x i8> [[TMP25]] to <4 x i32>
-; SSE-NEXT:    [[TMP27:%.*]] = mul <4 x i32> [[TMP26]], [[TMP3]]
-; SSE-NEXT:    [[TMP28:%.*]] = trunc <4 x i32> [[TMP27]] to <4 x i8>
+; SSE-NEXT:    [[TMP10:%.*]] = bitcast i8* [[B_ADDR_0351]] to <4 x i8>*
+; SSE-NEXT:    [[TMP11:%.*]] = load <4 x i8>, <4 x i8>* [[TMP10]], align 1
+; SSE-NEXT:    [[TMP12:%.*]] = icmp ult <4 x i8> [[TMP5]], [[TMP7]]
+; SSE-NEXT:    [[TMP13:%.*]] = select <4 x i1> [[TMP12]], <4 x i8> [[TMP11]], <4 x i8> [[TMP9]]
+; SSE-NEXT:    [[TMP14:%.*]] = zext <4 x i8> [[TMP13]] to <4 x i32>
+; SSE-NEXT:    [[TMP15:%.*]] = mul <4 x i32> [[TMP14]], [[SHUFFLE]]
+; SSE-NEXT:    [[TMP16:%.*]] = trunc <4 x i32> [[TMP15]] to <4 x i8>
 ; SSE-NEXT:    [[ARRAYIDX44:%.*]] = getelementptr inbounds i8, i8* [[E_ADDR_0354]], i64 3
-; SSE-NEXT:    [[TMP29:%.*]] = bitcast i8* [[E_ADDR_0354]] to <4 x i8>*
-; SSE-NEXT:    store <4 x i8> [[TMP28]], <4 x i8>* [[TMP29]], align 1
+; SSE-NEXT:    [[TMP17:%.*]] = bitcast i8* [[E_ADDR_0354]] to <4 x i8>*
+; SSE-NEXT:    store <4 x i8> [[TMP16]], <4 x i8>* [[TMP17]], align 1
 ; SSE-NEXT:    [[ARRAYIDX45:%.*]] = getelementptr inbounds i8, i8* [[C_ADDR_0352]], i64 4
 ; SSE-NEXT:    [[ARRAYIDX47:%.*]] = getelementptr inbounds i8, i8* [[D_ADDR_0353]], i64 4
 ; SSE-NEXT:    [[ARRAYIDX49:%.*]] = getelementptr inbounds i8, i8* [[A_ADDR_0355]], i64 4
 ; SSE-NEXT:    [[ARRAYIDX52:%.*]] = getelementptr inbounds i8, i8* [[B_ADDR_0351]], i64 4
 ; SSE-NEXT:    [[ARRAYIDX56:%.*]] = getelementptr inbounds i8, i8* [[E_ADDR_0354]], i64 4
 ; SSE-NEXT:    [[ARRAYIDX57:%.*]] = getelementptr inbounds i8, i8* [[C_ADDR_0352]], i64 5
 ; SSE-NEXT:    [[ARRAYIDX59:%.*]] = getelementptr inbounds i8, i8* [[D_ADDR_0353]], i64 5
 ; SSE-NEXT:    [[ARRAYIDX61:%.*]] = getelementptr inbounds i8, i8* [[A_ADDR_0355]], i64 5
 ; SSE-NEXT:    [[ARRAYIDX64:%.*]] = getelementptr inbounds i8, i8* [[B_ADDR_0351]], i64 5
 ; SSE-NEXT:    [[ARRAYIDX68:%.*]] = getelementptr inbounds i8, i8* [[E_ADDR_0354]], i64 5
 ; SSE-NEXT:    [[ARRAYIDX69:%.*]] = getelementptr inbounds i8, i8* [[C_ADDR_0352]], i64 6
 ; SSE-NEXT:    [[ARRAYIDX71:%.*]] = getelementptr inbounds i8, i8* [[D_ADDR_0353]], i64 6
 ; SSE-NEXT:    [[ARRAYIDX73:%.*]] = getelementptr inbounds i8, i8* [[A_ADDR_0355]], i64 6
 ; SSE-NEXT:    [[ARRAYIDX76:%.*]] = getelementptr inbounds i8, i8* [[B_ADDR_0351]], i64 6
 ; SSE-NEXT:    [[ARRAYIDX80:%.*]] = getelementptr inbounds i8, i8* [[E_ADDR_0354]], i64 6
 ; SSE-NEXT:    [[ARRAYIDX81:%.*]] = getelementptr inbounds i8, i8* [[C_ADDR_0352]], i64 7
-; SSE-NEXT:    [[TMP30:%.*]] = bitcast i8* [[ARRAYIDX45]] to <4 x i8>*
-; SSE-NEXT:    [[TMP31:%.*]] = load <4 x i8>, <4 x i8>* [[TMP30]], align 1
+; SSE-NEXT:    [[TMP18:%.*]] = bitcast i8* [[ARRAYIDX45]] to <4 x i8>*
+; SSE-NEXT:    [[TMP19:%.*]] = load <4 x i8>, <4 x i8>* [[TMP18]], align 1
 ; SSE-NEXT:    [[ARRAYIDX83:%.*]] = getelementptr inbounds i8, i8* [[D_ADDR_0353]], i64 7
-; SSE-NEXT:    [[TMP32:%.*]] = bitcast i8* [[ARRAYIDX47]] to <4 x i8>*
-; SSE-NEXT:    [[TMP33:%.*]] = load <4 x i8>, <4 x i8>* [[TMP32]], align 1
+; SSE-NEXT:    [[TMP20:%.*]] = bitcast i8* [[ARRAYIDX47]] to <4 x i8>*
+; SSE-NEXT:    [[TMP21:%.*]] = load <4 x i8>, <4 x i8>* [[TMP20]], align 1
 ; SSE-NEXT:    [[ARRAYIDX85:%.*]] = getelementptr inbounds i8, i8* [[A_ADDR_0355]], i64 7
-; SSE-NEXT:    [[TMP34:%.*]] = bitcast i8* [[ARRAYIDX49]] to <4 x i8>*
-; SSE-NEXT:    [[TMP35:%.*]] = load <4 x i8>, <4 x i8>* [[TMP34]], align 1
+; SSE-NEXT:    [[TMP22:%.*]] = bitcast i8* [[ARRAYIDX49]] to <4 x i8>*
+; SSE-NEXT:    [[TMP23:%.*]] = load <4 x i8>, <4 x i8>* [[TMP22]], align 1
 ; SSE-NEXT:    [[ARRAYIDX88:%.*]] = getelementptr inbounds i8, i8* [[B_ADDR_0351]], i64 7
-; SSE-NEXT:    [[TMP36:%.*]] = bitcast i8* [[ARRAYIDX52]] to <4 x i8>*
-; SSE-NEXT:    [[TMP37:%.*]] = load <4 x i8>, <4 x i8>* [[TMP36]], align 1
-; SSE-NEXT:    [[TMP38:%.*]] = icmp ult <4 x i8> [[TMP31]], [[TMP33]]
-; SSE-NEXT:    [[TMP39:%.*]] = select <4 x i1> [[TMP38]], <4 x i8> [[TMP37]], <4 x i8> [[TMP35]]
-; SSE-NEXT:    [[TMP40:%.*]] = zext <4 x i8> [[TMP39]] to <4 x i32>
-; SSE-NEXT:    [[TMP41:%.*]] = mul <4 x i32> [[TMP40]], [[TMP7]]
-; SSE-NEXT:    [[TMP42:%.*]] = trunc <4 x i32> [[TMP41]] to <4 x i8>
+; SSE-NEXT:    [[TMP24:%.*]] = bitcast i8* [[ARRAYIDX52]] to <4 x i8>*
+; SSE-NEXT:    [[TMP25:%.*]] = load <4 x i8>, <4 x i8>* [[TMP24]], align 1
+; SSE-NEXT:    [[TMP26:%.*]] = icmp ult <4 x i8> [[TMP19]], [[TMP21]]
+; SSE-NEXT:    [[TMP27:%.*]] = select <4 x i1> [[TMP26]], <4 x i8> [[TMP25]], <4 x i8> [[TMP23]]
+; SSE-NEXT:    [[TMP28:%.*]] = zext <4 x i8> [[TMP27]] to <4 x i32>
+; SSE-NEXT:    [[TMP29:%.*]] = mul <4 x i32> [[TMP28]], [[SHUFFLE1]]
+; SSE-NEXT:    [[TMP30:%.*]] = trunc <4 x i32> [[TMP29]] to <4 x i8>
 ; SSE-NEXT:    [[ARRAYIDX92:%.*]] = getelementptr inbounds i8, i8* [[E_ADDR_0354]], i64 7
-; SSE-NEXT:    [[TMP43:%.*]] = bitcast i8* [[ARRAYIDX56]] to <4 x i8>*
-; SSE-NEXT:    store <4 x i8> [[TMP42]], <4 x i8>* [[TMP43]], align 1
+; SSE-NEXT:    [[TMP31:%.*]] = bitcast i8* [[ARRAYIDX56]] to <4 x i8>*
+; SSE-NEXT:    store <4 x i8> [[TMP30]], <4 x i8>* [[TMP31]], align 1
 ; SSE-NEXT:    [[ARRAYIDX93:%.*]] = getelementptr inbounds i8, i8* [[C_ADDR_0352]], i64 8
 ; SSE-NEXT:    [[ARRAYIDX95:%.*]] = getelementptr inbounds i8, i8* [[D_ADDR_0353]], i64 8
 ; SSE-NEXT:    [[ARRAYIDX97:%.*]] = getelementptr inbounds i8, i8* [[A_ADDR_0355]], i64 8
 ; SSE-NEXT:    [[ARRAYIDX100:%.*]] = getelementptr inbounds i8, i8* [[B_ADDR_0351]], i64 8
 ; SSE-NEXT:    [[ARRAYIDX104:%.*]] = getelementptr inbounds i8, i8* [[E_ADDR_0354]], i64 8
 ; SSE-NEXT:    [[ARRAYIDX105:%.*]] = getelementptr inbounds i8, i8* [[C_ADDR_0352]], i64 9
 ; SSE-NEXT:    [[ARRAYIDX107:%.*]] = getelementptr inbounds i8, i8* [[D_ADDR_0353]], i64 9
 ; SSE-NEXT:    [[ARRAYIDX109:%.*]] = getelementptr inbounds i8, i8* [[A_ADDR_0355]], i64 9
 ; SSE-NEXT:    [[ARRAYIDX112:%.*]] = getelementptr inbounds i8, i8* [[B_ADDR_0351]], i64 9
 ; SSE-NEXT:    [[ARRAYIDX116:%.*]] = getelementptr inbounds i8, i8* [[E_ADDR_0354]], i64 9
 ; SSE-NEXT:    [[ARRAYIDX117:%.*]] = getelementptr inbounds i8, i8* [[C_ADDR_0352]], i64 10
 ; SSE-NEXT:    [[ARRAYIDX119:%.*]] = getelementptr inbounds i8, i8* [[D_ADDR_0353]], i64 10
 ; SSE-NEXT:    [[ARRAYIDX121:%.*]] = getelementptr inbounds i8, i8* [[A_ADDR_0355]], i64 10
 ; SSE-NEXT:    [[ARRAYIDX124:%.*]] = getelementptr inbounds i8, i8* [[B_ADDR_0351]], i64 10
 ; SSE-NEXT:    [[ARRAYIDX128:%.*]] = getelementptr inbounds i8, i8* [[E_ADDR_0354]], i64 10
 ; SSE-NEXT:    [[ARRAYIDX129:%.*]] = getelementptr inbounds i8, i8* [[C_ADDR_0352]], i64 11
-; SSE-NEXT:    [[TMP44:%.*]] = bitcast i8* [[ARRAYIDX93]] to <4 x i8>*
-; SSE-NEXT:    [[TMP45:%.*]] = load <4 x i8>, <4 x i8>* [[TMP44]], align 1
+; SSE-NEXT:    [[TMP32:%.*]] = bitcast i8* [[ARRAYIDX93]] to <4 x i8>*
+; SSE-NEXT:    [[TMP33:%.*]] = load <4 x i8>, <4 x i8>* [[TMP32]], align 1
 ; SSE-NEXT:    [[ARRAYIDX131:%.*]] = getelementptr inbounds i8, i8* [[D_ADDR_0353]], i64 11
-; SSE-NEXT:    [[TMP46:%.*]] = bitcast i8* [[ARRAYIDX95]] to <4 x i8>*
-; SSE-NEXT:    [[TMP47:%.*]] = load <4 x i8>, <4 x i8>* [[TMP46]], align 1
+; SSE-NEXT:    [[TMP34:%.*]] = bitcast i8* [[ARRAYIDX95]] to <4 x i8>*
+; SSE-NEXT:    [[TMP35:%.*]] = load <4 x i8>, <4 x i8>* [[TMP34]], align 1
 ; SSE-NEXT:    [[ARRAYIDX133:%.*]] = getelementptr inbounds i8, i8* [[A_ADDR_0355]], i64 11
-; SSE-NEXT:    [[TMP48:%.*]] = bitcast i8* [[ARRAYIDX97]] to <4 x i8>*
-; SSE-NEXT:    [[TMP49:%.*]] = load <4 x i8>, <4 x i8>* [[TMP48]], align 1
+; SSE-NEXT:    [[TMP36:%.*]] = bitcast i8* [[ARRAYIDX97]] to <4 x i8>*
+; SSE-NEXT:    [[TMP37:%.*]] = load <4 x i8>, <4 x i8>* [[TMP36]], align 1
 ; SSE-NEXT:    [[ARRAYIDX136:%.*]] = getelementptr inbounds i8, i8* [[B_ADDR_0351]], i64 11
-; SSE-NEXT:    [[TMP50:%.*]] = bitcast i8* [[ARRAYIDX100]] to <4 x i8>*
-; SSE-NEXT:    [[TMP51:%.*]] = load <4 x i8>, <4 x i8>* [[TMP50]], align 1
-; SSE-NEXT:    [[TMP52:%.*]] = icmp ult <4 x i8> [[TMP45]], [[TMP47]]
-; SSE-NEXT:    [[TMP53:%.*]] = select <4 x i1> [[TMP52]], <4 x i8> [[TMP51]], <4 x i8> [[TMP49]]
-; SSE-NEXT:    [[TMP54:%.*]] = zext <4 x i8> [[TMP53]] to <4 x i32>
-; SSE-NEXT:    [[TMP55:%.*]] = mul <4 x i32> [[TMP54]], [[TMP11]]
-; SSE-NEXT:    [[TMP56:%.*]] = trunc <4 x i32> [[TMP55]] to <4 x i8>
+; SSE-NEXT:    [[TMP38:%.*]] = bitcast i8* [[ARRAYIDX100]] to <4 x i8>*
+; SSE-NEXT:    [[TMP39:%.*]] = load <4 x i8>, <4 x i8>* [[TMP38]], align 1
+; SSE-NEXT:    [[TMP40:%.*]] = icmp ult <4 x i8> [[TMP33]], [[TMP35]]
+; SSE-NEXT:    [[TMP41:%.*]] = select <4 x i1> [[TMP40]], <4 x i8> [[TMP39]], <4 x i8> [[TMP37]]
+; SSE-NEXT:    [[TMP42:%.*]] = zext <4 x i8> [[TMP41]] to <4 x i32>
+; SSE-NEXT:    [[TMP43:%.*]] = mul <4 x i32> [[TMP42]], [[SHUFFLE2]]
+; SSE-NEXT:    [[TMP44:%.*]] = trunc <4 x i32> [[TMP43]] to <4 x i8>
 ; SSE-NEXT:    [[ARRAYIDX140:%.*]] = getelementptr inbounds i8, i8* [[E_ADDR_0354]], i64 11
-; SSE-NEXT:    [[TMP57:%.*]] = bitcast i8* [[ARRAYIDX104]] to <4 x i8>*
-; SSE-NEXT:    store <4 x i8> [[TMP56]], <4 x i8>* [[TMP57]], align 1
+; SSE-NEXT:    [[TMP45:%.*]] = bitcast i8* [[ARRAYIDX104]] to <4 x i8>*
+; SSE-NEXT:    store <4 x i8> [[TMP44]], <4 x i8>* [[TMP45]], align 1
 ; SSE-NEXT:    [[ARRAYIDX141:%.*]] = getelementptr inbounds i8, i8* [[C_ADDR_0352]], i64 12
 ; SSE-NEXT:    [[ARRAYIDX143:%.*]] = getelementptr inbounds i8, i8* [[D_ADDR_0353]], i64 12
 ; SSE-NEXT:    [[ARRAYIDX145:%.*]] = getelementptr inbounds i8, i8* [[A_ADDR_0355]], i64 12
 ; SSE-NEXT:    [[ARRAYIDX148:%.*]] = getelementptr inbounds i8, i8* [[B_ADDR_0351]], i64 12
 ; SSE-NEXT:    [[ARRAYIDX152:%.*]] = getelementptr inbounds i8, i8* [[E_ADDR_0354]], i64 12
 ; SSE-NEXT:    [[ARRAYIDX153:%.*]] = getelementptr inbounds i8, i8* [[C_ADDR_0352]], i64 13
 ; SSE-NEXT:    [[ARRAYIDX155:%.*]] = getelementptr inbounds i8, i8* [[D_ADDR_0353]], i64 13
 ; SSE-NEXT:    [[ARRAYIDX157:%.*]] = getelementptr inbounds i8, i8* [[A_ADDR_0355]], i64 13
 ; SSE-NEXT:    [[ARRAYIDX160:%.*]] = getelementptr inbounds i8, i8* [[B_ADDR_0351]], i64 13
 ; SSE-NEXT:    [[ARRAYIDX164:%.*]] = getelementptr inbounds i8, i8* [[E_ADDR_0354]], i64 13
 ; SSE-NEXT:    [[ARRAYIDX165:%.*]] = getelementptr inbounds i8, i8* [[C_ADDR_0352]], i64 14
 ; SSE-NEXT:    [[ARRAYIDX167:%.*]] = getelementptr inbounds i8, i8* [[D_ADDR_0353]], i64 14
 ; SSE-NEXT:    [[ARRAYIDX169:%.*]] = getelementptr inbounds i8, i8* [[A_ADDR_0355]], i64 14
 ; SSE-NEXT:    [[ARRAYIDX172:%.*]] = getelementptr inbounds i8, i8* [[B_ADDR_0351]], i64 14
 ; SSE-NEXT:    [[ARRAYIDX176:%.*]] = getelementptr inbounds i8, i8* [[E_ADDR_0354]], i64 14
 ; SSE-NEXT:    [[ARRAYIDX177:%.*]] = getelementptr inbounds i8, i8* [[C_ADDR_0352]], i64 15
-; SSE-NEXT:    [[TMP58:%.*]] = bitcast i8* [[ARRAYIDX141]] to <4 x i8>*
-; SSE-NEXT:    [[TMP59:%.*]] = load <4 x i8>, <4 x i8>* [[TMP58]], align 1
+; SSE-NEXT:    [[TMP46:%.*]] = bitcast i8* [[ARRAYIDX141]] to <4 x i8>*
+; SSE-NEXT:    [[TMP47:%.*]] = load <4 x i8>, <4 x i8>* [[TMP46]], align 1
 ; SSE-NEXT:    [[ARRAYIDX179:%.*]] = getelementptr inbounds i8, i8* [[D_ADDR_0353]], i64 15
-; SSE-NEXT:    [[TMP60:%.*]] = bitcast i8* [[ARRAYIDX143]] to <4 x i8>*
-; SSE-NEXT:    [[TMP61:%.*]] = load <4 x i8>, <4 x i8>* [[TMP60]], align 1
+; SSE-NEXT:    [[TMP48:%.*]] = bitcast i8* [[ARRAYIDX143]] to <4 x i8>*
+; SSE-NEXT:    [[TMP49:%.*]] = load <4 x i8>, <4 x i8>* [[TMP48]], align 1
 ; SSE-NEXT:    [[ARRAYIDX181:%.*]] = getelementptr inbounds i8, i8* [[A_ADDR_0355]], i64 15
-; SSE-NEXT:    [[TMP62:%.*]] = bitcast i8* [[ARRAYIDX145]] to <4 x i8>*
-; SSE-NEXT:    [[TMP63:%.*]] = load <4 x i8>, <4 x i8>* [[TMP62]], align 1
+; SSE-NEXT:    [[TMP50:%.*]] = bitcast i8* [[ARRAYIDX145]] to <4 x i8>*
+; SSE-NEXT:    [[TMP51:%.*]] = load <4 x i8>, <4 x i8>* [[TMP50]], align 1
 ; SSE-NEXT:    [[ARRAYIDX184:%.*]] = getelementptr inbounds i8, i8* [[B_ADDR_0351]], i64 15
-; SSE-NEXT:    [[TMP64:%.*]] = bitcast i8* [[ARRAYIDX148]] to <4 x i8>*
-; SSE-NEXT:    [[TMP65:%.*]] = load <4 x i8>, <4 x i8>* [[TMP64]], align 1
-; SSE-NEXT:    [[TMP66:%.*]] = icmp ult <4 x i8> [[TMP59]], [[TMP61]]
-; SSE-NEXT:    [[TMP67:%.*]] = select <4 x i1> [[TMP66]], <4 x i8> [[TMP65]], <4 x i8> [[TMP63]]
-; SSE-NEXT:    [[TMP68:%.*]] = zext <4 x i8> [[TMP67]] to <4 x i32>
-; SSE-NEXT:    [[TMP69:%.*]] = mul <4 x i32> [[TMP68]], [[TMP15]]
-; SSE-NEXT:    [[TMP70:%.*]] = trunc <4 x i32> [[TMP69]] to <4 x i8>
+; SSE-NEXT:    [[TMP52:%.*]] = bitcast i8* [[ARRAYIDX148]] to <4 x i8>*
+; SSE-NEXT:    [[TMP53:%.*]] = load <4 x i8>, <4 x i8>* [[TMP52]], align 1
+; SSE-NEXT:    [[TMP54:%.*]] = icmp ult <4 x i8> [[TMP47]], [[TMP49]]
+; SSE-NEXT:    [[TMP55:%.*]] = select <4 x i1> [[TMP54]], <4 x i8> [[TMP53]], <4 x i8> [[TMP51]]
+; SSE-NEXT:    [[TMP56:%.*]] = zext <4 x i8> [[TMP55]] to <4 x i32>
+; SSE-NEXT:    [[TMP57:%.*]] = mul <4 x i32> [[TMP56]], [[SHUFFLE3]]
+; SSE-NEXT:    [[TMP58:%.*]] = trunc <4 x i32> [[TMP57]] to <4 x i8>
 ; SSE-NEXT:    [[ARRAYIDX188:%.*]] = getelementptr inbounds i8, i8* [[E_ADDR_0354]], i64 15
-; SSE-NEXT:    [[TMP71:%.*]] = bitcast i8* [[ARRAYIDX152]] to <4 x i8>*
-; SSE-NEXT:    store <4 x i8> [[TMP70]], <4 x i8>* [[TMP71]], align 1
+; SSE-NEXT:    [[TMP59:%.*]] = bitcast i8* [[ARRAYIDX152]] to <4 x i8>*
+; SSE-NEXT:    store <4 x i8> [[TMP58]], <4 x i8>* [[TMP59]], align 1
 ; SSE-NEXT:    [[INC]] = add nuw nsw i32 [[I_0356]], 1
 ; SSE-NEXT:    [[ADD_PTR]] = getelementptr inbounds i8, i8* [[A_ADDR_0355]], i64 16
 ; SSE-NEXT:    [[ADD_PTR189]] = getelementptr inbounds i8, i8* [[B_ADDR_0351]], i64 16
 ; SSE-NEXT:    [[ADD_PTR190]] = getelementptr inbounds i8, i8* [[C_ADDR_0352]], i64 16
 ; SSE-NEXT:    [[ADD_PTR191]] = getelementptr inbounds i8, i8* [[D_ADDR_0353]], i64 16
 ; SSE-NEXT:    [[ADD_PTR192]] = getelementptr inbounds i8, i8* [[E_ADDR_0354]], i64 16
 ; SSE-NEXT:    [[EXITCOND:%.*]] = icmp eq i32 [[INC]], 8
 ; SSE-NEXT:    br i1 [[EXITCOND]], label [[FOR_END:%.*]], label [[FOR_BODY]]
 ; SSE:       for.end:
 ; SSE-NEXT:    ret void
 ;
 ; AVX512-LABEL: @bar(
 ; AVX512-NEXT:  entry:
 ; AVX512-NEXT:    [[TMP0:%.*]] = insertelement <16 x i32> poison, i32 [[W:%.*]], i32 0
-; AVX512-NEXT:    [[TMP1:%.*]] = insertelement <16 x i32> [[TMP0]], i32 [[W]], i32 1
-; AVX512-NEXT:    [[TMP2:%.*]] = insertelement <16 x i32> [[TMP1]], i32 [[W]], i32 2
-; AVX512-NEXT:    [[TMP3:%.*]] = insertelement <16 x i32> [[TMP2]], i32 [[W]], i32 3
-; AVX512-NEXT:    [[TMP4:%.*]] = insertelement <16 x i32> [[TMP3]], i32 [[W]], i32 4
-; AVX512-NEXT:    [[TMP5:%.*]] = insertelement <16 x i32> [[TMP4]], i32 [[W]], i32 5
-; AVX512-NEXT:    [[TMP6:%.*]] = insertelement <16 x i32> [[TMP5]], i32 [[W]], i32 6
-; AVX512-NEXT:    [[TMP7:%.*]] = insertelement <16 x i32> [[TMP6]], i32 [[W]], i32 7
-; AVX512-NEXT:    [[TMP8:%.*]] = insertelement <16 x i32> [[TMP7]], i32 [[W]], i32 8
-; AVX512-NEXT:    [[TMP9:%.*]] = insertelement <16 x i32> [[TMP8]], i32 [[W]], i32 9
-; AVX512-NEXT:    [[TMP10:%.*]] = insertelement <16 x i32> [[TMP9]], i32 [[W]], i32 10
-; AVX512-NEXT:    [[TMP11:%.*]] = insertelement <16 x i32> [[TMP10]], i32 [[W]], i32 11
-; AVX512-NEXT:    [[TMP12:%.*]] = insertelement <16 x i32> [[TMP11]], i32 [[W]], i32 12
-; AVX512-NEXT:    [[TMP13:%.*]] = insertelement <16 x i32> [[TMP12]], i32 [[W]], i32 13
-; AVX512-NEXT:    [[TMP14:%.*]] = insertelement <16 x i32> [[TMP13]], i32 [[W]], i32 14
-; AVX512-NEXT:    [[TMP15:%.*]] = insertelement <16 x i32> [[TMP14]], i32 [[W]], i32 15
+; AVX512-NEXT:    [[SHUFFLE:%.*]] = shufflevector <16 x i32> [[TMP0]], <16 x i32> poison, <16 x i32> zeroinitializer
 ; AVX512-NEXT:    br label [[FOR_BODY:%.*]]
 ; AVX512:       for.body:
 ; AVX512-NEXT:    [[I_0356:%.*]] = phi i32 [ 0, [[ENTRY:%.*]] ], [ [[INC:%.*]], [[FOR_BODY]] ]
 ; AVX512-NEXT:    [[A_ADDR_0355:%.*]] = phi i8* [ [[A:%.*]], [[ENTRY]] ], [ [[ADD_PTR:%.*]], [[FOR_BODY]] ]
 ; AVX512-NEXT:    [[E_ADDR_0354:%.*]] = phi i8* [ [[E:%.*]], [[ENTRY]] ], [ [[ADD_PTR192:%.*]], [[FOR_BODY]] ]
 ; AVX512-NEXT:    [[D_ADDR_0353:%.*]] = phi i8* [ [[D:%.*]], [[ENTRY]] ], [ [[ADD_PTR191:%.*]], [[FOR_BODY]] ]
 ; AVX512-NEXT:    [[C_ADDR_0352:%.*]] = phi i8* [ [[C:%.*]], [[ENTRY]] ], [ [[ADD_PTR190:%.*]], [[FOR_BODY]] ]
 ; AVX512-NEXT:    [[B_ADDR_0351:%.*]] = phi i8* [ [[B:%.*]], [[ENTRY]] ], [ [[ADD_PTR189:%.*]], [[FOR_BODY]] ]

 ; AVX512-NEXT:    [[ARRAYIDX160:%.*]] = getelementptr inbounds i8, i8* [[B_ADDR_0351]], i64 13
 ; AVX512-NEXT:    [[ARRAYIDX164:%.*]] = getelementptr inbounds i8, i8* [[E_ADDR_0354]], i64 13
 ; AVX512-NEXT:    [[ARRAYIDX165:%.*]] = getelementptr inbounds i8, i8* [[C_ADDR_0352]], i64 14
 ; AVX512-NEXT:    [[ARRAYIDX167:%.*]] = getelementptr inbounds i8, i8* [[D_ADDR_0353]], i64 14
 ; AVX512-NEXT:    [[ARRAYIDX169:%.*]] = getelementptr inbounds i8, i8* [[A_ADDR_0355]], i64 14
 ; AVX512-NEXT:    [[ARRAYIDX172:%.*]] = getelementptr inbounds i8, i8* [[B_ADDR_0351]], i64 14
 ; AVX512-NEXT:    [[ARRAYIDX176:%.*]] = getelementptr inbounds i8, i8* [[E_ADDR_0354]], i64 14
 ; AVX512-NEXT:    [[ARRAYIDX177:%.*]] = getelementptr inbounds i8, i8* [[C_ADDR_0352]], i64 15
-; AVX512-NEXT:    [[TMP16:%.*]] = bitcast i8* [[C_ADDR_0352]] to <16 x i8>*
-; AVX512-NEXT:    [[TMP17:%.*]] = load <16 x i8>, <16 x i8>* [[TMP16]], align 1
+; AVX512-NEXT:    [[TMP1:%.*]] = bitcast i8* [[C_ADDR_0352]] to <16 x i8>*
+; AVX512-NEXT:    [[TMP2:%.*]] = load <16 x i8>, <16 x i8>* [[TMP1]], align 1
 ; AVX512-NEXT:    [[ARRAYIDX179:%.*]] = getelementptr inbounds i8, i8* [[D_ADDR_0353]], i64 15
-; AVX512-NEXT:    [[TMP18:%.*]] = bitcast i8* [[D_ADDR_0353]] to <16 x i8>*
-; AVX512-NEXT:    [[TMP19:%.*]] = load <16 x i8>, <16 x i8>* [[TMP18]], align 1
+; AVX512-NEXT:    [[TMP3:%.*]] = bitcast i8* [[D_ADDR_0353]] to <16 x i8>*
+; AVX512-NEXT:    [[TMP4:%.*]] = load <16 x i8>, <16 x i8>* [[TMP3]], align 1
 ; AVX512-NEXT:    [[ARRAYIDX181:%.*]] = getelementptr inbounds i8, i8* [[A_ADDR_0355]], i64 15
-; AVX512-NEXT:    [[TMP20:%.*]] = bitcast i8* [[A_ADDR_0355]] to <16 x i8>*
-; AVX512-NEXT:    [[TMP21:%.*]] = load <16 x i8>, <16 x i8>* [[TMP20]], align 1
+; AVX512-NEXT:    [[TMP5:%.*]] = bitcast i8* [[A_ADDR_0355]] to <16 x i8>*
+; AVX512-NEXT:    [[TMP6:%.*]] = load <16 x i8>, <16 x i8>* [[TMP5]], align 1
 ; AVX512-NEXT:    [[ARRAYIDX184:%.*]] = getelementptr inbounds i8, i8* [[B_ADDR_0351]], i64 15
-; AVX512-NEXT:    [[TMP22:%.*]] = bitcast i8* [[B_ADDR_0351]] to <16 x i8>*
-; AVX512-NEXT:    [[TMP23:%.*]] = load <16 x i8>, <16 x i8>* [[TMP22]], align 1
-; AVX512-NEXT:    [[TMP24:%.*]] = icmp ult <16 x i8> [[TMP17]], [[TMP19]]
-; AVX512-NEXT:    [[TMP25:%.*]] = select <16 x i1> [[TMP24]], <16 x i8> [[TMP23]], <16 x i8> [[TMP21]]
-; AVX512-NEXT:    [[TMP26:%.*]] = zext <16 x i8> [[TMP25]] to <16 x i32>
-; AVX512-NEXT:    [[TMP27:%.*]] = mul <16 x i32> [[TMP26]], [[TMP15]]
-; AVX512-NEXT:    [[TMP28:%.*]] = trunc <16 x i32> [[TMP27]] to <16 x i8>
+; AVX512-NEXT:    [[TMP7:%.*]] = bitcast i8* [[B_ADDR_0351]] to <16 x i8>*
+; AVX512-NEXT:    [[TMP8:%.*]] = load <16 x i8>, <16 x i8>* [[TMP7]], align 1
+; AVX512-NEXT:    [[TMP9:%.*]] = icmp ult <16 x i8> [[TMP2]], [[TMP4]]
+; AVX512-NEXT:    [[TMP10:%.*]] = select <16 x i1> [[TMP9]], <16 x i8> [[TMP8]], <16 x i8> [[TMP6]]
+; AVX512-NEXT:    [[TMP11:%.*]] = zext <16 x i8> [[TMP10]] to <16 x i32>
+; AVX512-NEXT:    [[TMP12:%.*]] = mul <16 x i32> [[TMP11]], [[SHUFFLE]]
+; AVX512-NEXT:    [[TMP13:%.*]] = trunc <16 x i32> [[TMP12]] to <16 x i8>
 ; AVX512-NEXT:    [[ARRAYIDX188:%.*]] = getelementptr inbounds i8, i8* [[E_ADDR_0354]], i64 15
-; AVX512-NEXT:    [[TMP29:%.*]] = bitcast i8* [[E_ADDR_0354]] to <16 x i8>*
-; AVX512-NEXT:    store <16 x i8> [[TMP28]], <16 x i8>* [[TMP29]], align 1
+; AVX512-NEXT:    [[TMP14:%.*]] = bitcast i8* [[E_ADDR_0354]] to <16 x i8>*
+; AVX512-NEXT:    store <16 x i8> [[TMP13]], <16 x i8>* [[TMP14]], align 1
 ; AVX512-NEXT:    [[INC]] = add nuw nsw i32 [[I_0356]], 1
 ; AVX512-NEXT:    [[ADD_PTR]] = getelementptr inbounds i8, i8* [[A_ADDR_0355]], i64 16
 ; AVX512-NEXT:    [[ADD_PTR189]] = getelementptr inbounds i8, i8* [[B_ADDR_0351]], i64 16
 ; AVX512-NEXT:    [[ADD_PTR190]] = getelementptr inbounds i8, i8* [[C_ADDR_0352]], i64 16
 ; AVX512-NEXT:    [[ADD_PTR191]] = getelementptr inbounds i8, i8* [[D_ADDR_0353]], i64 16
 ; AVX512-NEXT:    [[ADD_PTR192]] = getelementptr inbounds i8, i8* [[E_ADDR_0354]], i64 16
 ; AVX512-NEXT:    [[EXITCOND:%.*]] = icmp eq i32 [[INC]], 8
 ; AVX512-NEXT:    br i1 [[EXITCOND]], label [[FOR_END:%.*]], label [[FOR_BODY]]

llvm/test/Transforms/SLPVectorizer/X86/loopinvariant.ll

 ; CHECK-NEXT:    [[ARRAYIDX20:%.*]] = getelementptr inbounds i32, i32* [[A]], i64 [[TMP4]]
 ; CHECK-NEXT:    [[TMP5:%.*]] = or i64 [[INDVARS_IV]], 6
 ; CHECK-NEXT:    [[ARRAYIDX24:%.*]] = getelementptr inbounds i32, i32* [[A]], i64 [[TMP5]]
 ; CHECK-NEXT:    [[TMP6:%.*]] = or i64 [[INDVARS_IV]], 7
 ; CHECK-NEXT:    [[ARRAYIDX28:%.*]] = getelementptr inbounds i32, i32* [[A]], i64 [[TMP6]]
 ; CHECK-NEXT:    [[TMP7:%.*]] = bitcast i32* [[ARRAYIDX]] to <8 x i32>*
 ; CHECK-NEXT:    [[TMP8:%.*]] = load <8 x i32>, <8 x i32>* [[TMP7]], align 4
 ; CHECK-NEXT:    [[TMP9:%.*]] = insertelement <8 x i32> poison, i32 [[N]], i32 0
-; CHECK-NEXT:    [[TMP10:%.*]] = insertelement <8 x i32> [[TMP9]], i32 [[N]], i32 1
-; CHECK-NEXT:    [[TMP11:%.*]] = insertelement <8 x i32> [[TMP10]], i32 [[N]], i32 2
-; CHECK-NEXT:    [[TMP12:%.*]] = insertelement <8 x i32> [[TMP11]], i32 [[N]], i32 3
-; CHECK-NEXT:    [[TMP13:%.*]] = insertelement <8 x i32> [[TMP12]], i32 [[N]], i32 4
-; CHECK-NEXT:    [[TMP14:%.*]] = insertelement <8 x i32> [[TMP13]], i32 [[N]], i32 5
-; CHECK-NEXT:    [[TMP15:%.*]] = insertelement <8 x i32> [[TMP14]], i32 [[N]], i32 6
-; CHECK-NEXT:    [[TMP16:%.*]] = insertelement <8 x i32> [[TMP15]], i32 [[N]], i32 7
-; CHECK-NEXT:    [[TMP17:%.*]] = add nsw <8 x i32> [[TMP8]], [[TMP16]]
-; CHECK-NEXT:    [[TMP18:%.*]] = bitcast i32* [[ARRAYIDX]] to <8 x i32>*
-; CHECK-NEXT:    store <8 x i32> [[TMP17]], <8 x i32>* [[TMP18]], align 4
+; CHECK-NEXT:    [[SHUFFLE:%.*]] = shufflevector <8 x i32> [[TMP9]], <8 x i32> poison, <8 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP10:%.*]] = add nsw <8 x i32> [[TMP8]], [[SHUFFLE]]
+; CHECK-NEXT:    [[TMP11:%.*]] = bitcast i32* [[ARRAYIDX]] to <8 x i32>*
+; CHECK-NEXT:    store <8 x i32> [[TMP10]], <8 x i32>* [[TMP11]], align 4
 ; CHECK-NEXT:    [[INDVARS_IV_NEXT]] = add i64 [[INDVARS_IV]], 8
-; CHECK-NEXT:    [[TMP19:%.*]] = trunc i64 [[INDVARS_IV_NEXT]] to i32
-; CHECK-NEXT:    [[CMP:%.*]] = icmp slt i32 [[TMP19]], [[N]]
+; CHECK-NEXT:    [[TMP12:%.*]] = trunc i64 [[INDVARS_IV_NEXT]] to i32
+; CHECK-NEXT:    [[CMP:%.*]] = icmp slt i32 [[TMP12]], [[N]]
 ; CHECK-NEXT:    br i1 [[CMP]], label [[FOR_BODY]], label [[FOR_END]]
 ; CHECK:       for.end:
 ; CHECK-NEXT:    ret i32 undef
 ;
 entry:
   %cmp62 = icmp sgt i32 %n, 0
   br i1 %cmp62, label %for.body, label %for.end
 

llvm/test/Transforms/SLPVectorizer/X86/multi_user.ll

 ;  A[3] += n * 5 + 10;
 ;  A[4] += n * 5 + 11;
 ;}
 
 define i32 @foo(i32* nocapture %A, i32 %n) {
 ; CHECK-LABEL: @foo(
 ; CHECK-NEXT:    [[TMP1:%.*]] = mul nsw i32 [[N:%.*]], 5
 ; CHECK-NEXT:    [[TMP2:%.*]] = insertelement <4 x i32> poison, i32 [[TMP1]], i32 0
-; CHECK-NEXT:    [[TMP3:%.*]] = insertelement <4 x i32> [[TMP2]], i32 [[TMP1]], i32 1
-; CHECK-NEXT:    [[TMP4:%.*]] = insertelement <4 x i32> [[TMP3]], i32 [[TMP1]], i32 2
-; CHECK-NEXT:    [[TMP5:%.*]] = insertelement <4 x i32> [[TMP4]], i32 [[TMP1]], i32 3
-; CHECK-NEXT:    [[TMP6:%.*]] = add nsw <4 x i32> [[TMP5]], <i32 7, i32 8, i32 9, i32 10>
-; CHECK-NEXT:    [[TMP7:%.*]] = bitcast i32* [[A:%.*]] to <4 x i32>*
-; CHECK-NEXT:    [[TMP8:%.*]] = load <4 x i32>, <4 x i32>* [[TMP7]], align 4
-; CHECK-NEXT:    [[TMP9:%.*]] = add nsw <4 x i32> [[TMP6]], [[TMP8]]
-; CHECK-NEXT:    [[TMP10:%.*]] = bitcast i32* [[A]] to <4 x i32>*
-; CHECK-NEXT:    store <4 x i32> [[TMP9]], <4 x i32>* [[TMP10]], align 4
-; CHECK-NEXT:    [[TMP11:%.*]] = add nsw i32 [[TMP1]], 11
-; CHECK-NEXT:    [[TMP12:%.*]] = getelementptr inbounds i32, i32* [[A]], i64 4
-; CHECK-NEXT:    [[TMP13:%.*]] = load i32, i32* [[TMP12]], align 4
-; CHECK-NEXT:    [[TMP14:%.*]] = add nsw i32 [[TMP11]], [[TMP13]]
-; CHECK-NEXT:    store i32 [[TMP14]], i32* [[TMP12]], align 4
+; CHECK-NEXT:    [[SHUFFLE:%.*]] = shufflevector <4 x i32> [[TMP2]], <4 x i32> poison, <4 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP3:%.*]] = add nsw <4 x i32> [[SHUFFLE]], <i32 7, i32 8, i32 9, i32 10>
+; CHECK-NEXT:    [[TMP4:%.*]] = bitcast i32* [[A:%.*]] to <4 x i32>*
+; CHECK-NEXT:    [[TMP5:%.*]] = load <4 x i32>, <4 x i32>* [[TMP4]], align 4
+; CHECK-NEXT:    [[TMP6:%.*]] = add nsw <4 x i32> [[TMP3]], [[TMP5]]
+; CHECK-NEXT:    [[TMP7:%.*]] = bitcast i32* [[A]] to <4 x i32>*
+; CHECK-NEXT:    store <4 x i32> [[TMP6]], <4 x i32>* [[TMP7]], align 4
+; CHECK-NEXT:    [[TMP8:%.*]] = add nsw i32 [[TMP1]], 11
+; CHECK-NEXT:    [[TMP9:%.*]] = getelementptr inbounds i32, i32* [[A]], i64 4
+; CHECK-NEXT:    [[TMP10:%.*]] = load i32, i32* [[TMP9]], align 4
+; CHECK-NEXT:    [[TMP11:%.*]] = add nsw i32 [[TMP8]], [[TMP10]]
+; CHECK-NEXT:    store i32 [[TMP11]], i32* [[TMP9]], align 4
 ; CHECK-NEXT:    ret i32 undef
 ;
   %1 = mul nsw i32 %n, 5
   %2 = add nsw i32 %1, 7
   %3 = load i32, i32* %A, align 4
   %4 = add nsw i32 %2, %3
   store i32 %4, i32* %A, align 4
   %5 = add nsw i32 %1, 8

llvm/test/Transforms/SLPVectorizer/X86/remark_extract_broadcast.ll

 ; CHECK-NEXT:    [[TMP0:%.*]] = insertelement <8 x i16> poison, i16 [[V0]], i32 0
 ; CHECK-NEXT:    [[TMP1:%.*]] = insertelement <8 x i16> [[TMP0]], i16 [[V1]], i32 1
 ; CHECK-NEXT:    [[TMP2:%.*]] = insertelement <8 x i16> [[TMP1]], i16 [[V2]], i32 2
 ; CHECK-NEXT:    [[TMP3:%.*]] = insertelement <8 x i16> [[TMP2]], i16 [[V3]], i32 3
 ; CHECK-NEXT:    [[TMP4:%.*]] = insertelement <8 x i16> [[TMP3]], i16 [[V4]], i32 4
 ; CHECK-NEXT:    [[TMP5:%.*]] = insertelement <8 x i16> [[TMP4]], i16 [[V5]], i32 5
 ; CHECK-NEXT:    [[TMP6:%.*]] = insertelement <8 x i16> [[TMP5]], i16 [[V6]], i32 6
 ; CHECK-NEXT:    [[TMP7:%.*]] = insertelement <8 x i16> [[TMP6]], i16 [[V7]], i32 7
-; CHECK-NEXT:    [[SHUFFLE:%.*]] = shufflevector <8 x i16> [[TMP0]], <8 x i16> poison, <8 x i32> zeroinitializer
+; CHECK-NEXT:    [[SHUFFLE:%.*]] = shufflevector <8 x i16> [[TMP0]], <8 x i16> poison, <8 x i32> <i32 0, i32 undef, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0>
 ; CHECK-NEXT:    [[TMP8:%.*]] = add <8 x i16> [[TMP7]], [[SHUFFLE]]
 ; CHECK-NEXT:    [[TMP9:%.*]] = bitcast i16* [[P0]] to <8 x i16>*
 ; CHECK-NEXT:    store <8 x i16> [[TMP8]], <8 x i16>* [[TMP9]], align 2
 ; CHECK-NEXT:    ret void
 ;
 ; YAML:      Pass:            slp-vectorizer
 ; YAML-NEXT: Name:            StoresVectorized
 ; YAML-NEXT: Function:        fextr

llvm/test/Transforms/SLPVectorizer/X86/resched.ll

 ; CHECK:       if.then22.i:
 ; CHECK-NEXT:    [[SUB_I:%.*]] = add nsw i32 undef, -1
 ; CHECK-NEXT:    [[CONV31_I:%.*]] = and i32 undef, [[SUB_I]]
 ; CHECK-NEXT:    [[TMP0:%.*]] = getelementptr inbounds %"struct.std::array", %"struct.std::array"* undef, i64 0, i32 0, i64 0
 ; CHECK-NEXT:    [[ARRAYIDX_I_I7_1_I_I:%.*]] = getelementptr inbounds %"struct.std::array", %"struct.std::array"* undef, i64 0, i32 0, i64 1
 ; CHECK-NEXT:    [[ARRAYIDX_I_I7_2_I_I:%.*]] = getelementptr inbounds %"struct.std::array", %"struct.std::array"* undef, i64 0, i32 0, i64 2
 ; CHECK-NEXT:    [[ARRAYIDX_I_I7_3_I_I:%.*]] = getelementptr inbounds %"struct.std::array", %"struct.std::array"* undef, i64 0, i32 0, i64 3
 ; CHECK-NEXT:    [[TMP1:%.*]] = insertelement <4 x i32> poison, i32 [[CONV31_I]], i32 0
-; CHECK-NEXT:    [[TMP2:%.*]] = insertelement <4 x i32> [[TMP1]], i32 [[CONV31_I]], i32 1
-; CHECK-NEXT:    [[TMP3:%.*]] = insertelement <4 x i32> [[TMP2]], i32 [[CONV31_I]], i32 2
-; CHECK-NEXT:    [[TMP4:%.*]] = insertelement <4 x i32> [[TMP3]], i32 [[CONV31_I]], i32 3
-; CHECK-NEXT:    [[TMP5:%.*]] = lshr <4 x i32> [[TMP4]], <i32 1, i32 2, i32 3, i32 4>
+; CHECK-NEXT:    [[SHUFFLE:%.*]] = shufflevector <4 x i32> [[TMP1]], <4 x i32> poison, <4 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP2:%.*]] = lshr <4 x i32> [[SHUFFLE]], <i32 1, i32 2, i32 3, i32 4>
 ; CHECK-NEXT:    [[ARRAYIDX_I_I7_4_I_I:%.*]] = getelementptr inbounds %"struct.std::array", %"struct.std::array"* undef, i64 0, i32 0, i64 4
 ; CHECK-NEXT:    [[SHR_4_I_I:%.*]] = lshr i32 [[CONV31_I]], 5
 ; CHECK-NEXT:    [[ARRAYIDX_I_I7_5_I_I:%.*]] = getelementptr inbounds %"struct.std::array", %"struct.std::array"* undef, i64 0, i32 0, i64 5
 ; CHECK-NEXT:    [[ARRAYIDX_I_I7_6_I_I:%.*]] = getelementptr inbounds %"struct.std::array", %"struct.std::array"* undef, i64 0, i32 0, i64 6
-; CHECK-NEXT:    [[TMP6:%.*]] = insertelement <2 x i32> poison, i32 [[CONV31_I]], i32 0
-; CHECK-NEXT:    [[TMP7:%.*]] = insertelement <2 x i32> [[TMP6]], i32 [[CONV31_I]], i32 1
-; CHECK-NEXT:    [[TMP8:%.*]] = lshr <2 x i32> [[TMP7]], <i32 6, i32 7>
+; CHECK-NEXT:    [[TMP3:%.*]] = insertelement <2 x i32> poison, i32 [[CONV31_I]], i32 0
+; CHECK-NEXT:    [[TMP4:%.*]] = insertelement <2 x i32> [[TMP3]], i32 [[CONV31_I]], i32 1
+; CHECK-NEXT:    [[TMP5:%.*]] = lshr <2 x i32> [[TMP4]], <i32 6, i32 7>
 ; CHECK-NEXT:    [[ARRAYIDX_I_I7_7_I_I:%.*]] = getelementptr inbounds %"struct.std::array", %"struct.std::array"* undef, i64 0, i32 0, i64 7
 ; CHECK-NEXT:    [[ARRAYIDX_I_I7_8_I_I:%.*]] = getelementptr inbounds %"struct.std::array", %"struct.std::array"* undef, i64 0, i32 0, i64 8
 ; CHECK-NEXT:    [[ARRAYIDX_I_I7_9_I_I:%.*]] = getelementptr inbounds %"struct.std::array", %"struct.std::array"* undef, i64 0, i32 0, i64 9
 ; CHECK-NEXT:    [[ARRAYIDX_I_I7_10_I_I:%.*]] = getelementptr inbounds %"struct.std::array", %"struct.std::array"* undef, i64 0, i32 0, i64 10
-; CHECK-NEXT:    [[TMP9:%.*]] = lshr <4 x i32> [[TMP4]], <i32 8, i32 9, i32 10, i32 11>
+; CHECK-NEXT:    [[TMP6:%.*]] = lshr <4 x i32> [[SHUFFLE]], <i32 8, i32 9, i32 10, i32 11>
 ; CHECK-NEXT:    [[ARRAYIDX_I_I7_11_I_I:%.*]] = getelementptr inbounds %"struct.std::array", %"struct.std::array"* undef, i64 0, i32 0, i64 11
 ; CHECK-NEXT:    [[ARRAYIDX_I_I7_12_I_I:%.*]] = getelementptr inbounds %"struct.std::array", %"struct.std::array"* undef, i64 0, i32 0, i64 12
 ; CHECK-NEXT:    [[ARRAYIDX_I_I7_13_I_I:%.*]] = getelementptr inbounds %"struct.std::array", %"struct.std::array"* undef, i64 0, i32 0, i64 13
 ; CHECK-NEXT:    [[ARRAYIDX_I_I7_14_I_I:%.*]] = getelementptr inbounds %"struct.std::array", %"struct.std::array"* undef, i64 0, i32 0, i64 14
-; CHECK-NEXT:    [[TMP10:%.*]] = lshr <4 x i32> [[TMP4]], <i32 12, i32 13, i32 14, i32 15>
-; CHECK-NEXT:    [[TMP11:%.*]] = extractelement <4 x i32> [[TMP10]], i32 3
-; CHECK-NEXT:    [[TMP12:%.*]] = insertelement <16 x i32> poison, i32 [[SUB_I]], i32 0
-; CHECK-NEXT:    [[TMP13:%.*]] = shufflevector <4 x i32> [[TMP5]], <4 x i32> poison, <16 x i32> <i32 0, i32 1, i32 2, i32 3, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef>
-; CHECK-NEXT:    [[TMP14:%.*]] = shufflevector <16 x i32> [[TMP12]], <16 x i32> [[TMP13]], <16 x i32> <i32 0, i32 16, i32 17, i32 18, i32 19, i32 5, i32 6, i32 7, i32 8, i32 9, i32 10, i32 11, i32 12, i32 13, i32 14, i32 15>
-; CHECK-NEXT:    [[TMP15:%.*]] = insertelement <16 x i32> [[TMP14]], i32 [[SHR_4_I_I]], i32 5
-; CHECK-NEXT:    [[TMP16:%.*]] = shufflevector <2 x i32> [[TMP8]], <2 x i32> poison, <16 x i32> <i32 0, i32 1, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef>
-; CHECK-NEXT:    [[TMP17:%.*]] = shufflevector <16 x i32> [[TMP15]], <16 x i32> [[TMP16]], <16 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 16, i32 17, i32 8, i32 9, i32 10, i32 11, i32 12, i32 13, i32 14, i32 15>
-; CHECK-NEXT:    [[TMP18:%.*]] = shufflevector <4 x i32> [[TMP9]], <4 x i32> poison, <16 x i32> <i32 0, i32 1, i32 2, i32 3, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef>
-; CHECK-NEXT:    [[TMP19:%.*]] = shufflevector <16 x i32> [[TMP17]], <16 x i32> [[TMP18]], <16 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7, i32 16, i32 17, i32 18, i32 19, i32 12, i32 13, i32 14, i32 15>
-; CHECK-NEXT:    [[TMP20:%.*]] = shufflevector <4 x i32> [[TMP10]], <4 x i32> poison, <16 x i32> <i32 0, i32 1, i32 2, i32 3, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef>
-; CHECK-NEXT:    [[TMP21:%.*]] = shufflevector <16 x i32> [[TMP19]], <16 x i32> [[TMP20]], <16 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7, i32 8, i32 9, i32 10, i32 11, i32 16, i32 17, i32 18, i32 19>
-; CHECK-NEXT:    [[TMP22:%.*]] = trunc <16 x i32> [[TMP21]] to <16 x i8>
-; CHECK-NEXT:    [[TMP23:%.*]] = extractelement <4 x i32> [[TMP10]], i32 2
-; CHECK-NEXT:    [[TMP24:%.*]] = extractelement <4 x i32> [[TMP10]], i32 1
-; CHECK-NEXT:    [[TMP25:%.*]] = extractelement <4 x i32> [[TMP10]], i32 0
-; CHECK-NEXT:    [[TMP26:%.*]] = extractelement <4 x i32> [[TMP9]], i32 3
-; CHECK-NEXT:    [[TMP27:%.*]] = extractelement <4 x i32> [[TMP9]], i32 2
-; CHECK-NEXT:    [[TMP28:%.*]] = extractelement <4 x i32> [[TMP9]], i32 1
-; CHECK-NEXT:    [[TMP29:%.*]] = extractelement <4 x i32> [[TMP9]], i32 0
-; CHECK-NEXT:    [[TMP30:%.*]] = extractelement <2 x i32> [[TMP8]], i32 1
-; CHECK-NEXT:    [[TMP31:%.*]] = extractelement <2 x i32> [[TMP8]], i32 0
-; CHECK-NEXT:    [[TMP32:%.*]] = extractelement <4 x i32> [[TMP5]], i32 3
-; CHECK-NEXT:    [[TMP33:%.*]] = extractelement <4 x i32> [[TMP5]], i32 2
-; CHECK-NEXT:    [[TMP34:%.*]] = extractelement <4 x i32> [[TMP5]], i32 1
-; CHECK-NEXT:    [[TMP35:%.*]] = extractelement <4 x i32> [[TMP5]], i32 0
-; CHECK-NEXT:    [[TMP36:%.*]] = and <16 x i8> [[TMP22]], <i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1>
+; CHECK-NEXT:    [[TMP7:%.*]] = lshr <4 x i32> [[SHUFFLE]], <i32 12, i32 13, i32 14, i32 15>
+; CHECK-NEXT:    [[TMP8:%.*]] = extractelement <4 x i32> [[TMP7]], i32 3
+; CHECK-NEXT:    [[TMP9:%.*]] = insertelement <16 x i32> poison, i32 [[SUB_I]], i32 0
+; CHECK-NEXT:    [[TMP10:%.*]] = shufflevector <4 x i32> [[TMP2]], <4 x i32> poison, <16 x i32> <i32 0, i32 1, i32 2, i32 3, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef>
+; CHECK-NEXT:    [[TMP11:%.*]] = shufflevector <16 x i32> [[TMP9]], <16 x i32> [[TMP10]], <16 x i32> <i32 0, i32 16, i32 17, i32 18, i32 19, i32 5, i32 6, i32 7, i32 8, i32 9, i32 10, i32 11, i32 12, i32 13, i32 14, i32 15>
+; CHECK-NEXT:    [[TMP12:%.*]] = insertelement <16 x i32> [[TMP11]], i32 [[SHR_4_I_I]], i32 5
+; CHECK-NEXT:    [[TMP13:%.*]] = shufflevector <2 x i32> [[TMP5]], <2 x i32> poison, <16 x i32> <i32 0, i32 1, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef>
+; CHECK-NEXT:    [[TMP14:%.*]] = shufflevector <16 x i32> [[TMP12]], <16 x i32> [[TMP13]], <16 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 16, i32 17, i32 8, i32 9, i32 10, i32 11, i32 12, i32 13, i32 14, i32 15>
+; CHECK-NEXT:    [[TMP15:%.*]] = shufflevector <4 x i32> [[TMP6]], <4 x i32> poison, <16 x i32> <i32 0, i32 1, i32 2, i32 3, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef>
+; CHECK-NEXT:    [[TMP16:%.*]] = shufflevector <16 x i32> [[TMP14]], <16 x i32> [[TMP15]], <16 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7, i32 16, i32 17, i32 18, i32 19, i32 12, i32 13, i32 14, i32 15>
+; CHECK-NEXT:    [[TMP17:%.*]] = shufflevector <4 x i32> [[TMP7]], <4 x i32> poison, <16 x i32> <i32 0, i32 1, i32 2, i32 3, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef>
+; CHECK-NEXT:    [[TMP18:%.*]] = shufflevector <16 x i32> [[TMP16]], <16 x i32> [[TMP17]], <16 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7, i32 8, i32 9, i32 10, i32 11, i32 16, i32 17, i32 18, i32 19>
+; CHECK-NEXT:    [[TMP19:%.*]] = trunc <16 x i32> [[TMP18]] to <16 x i8>
+; CHECK-NEXT:    [[TMP20:%.*]] = extractelement <4 x i32> [[TMP7]], i32 2
+; CHECK-NEXT:    [[TMP21:%.*]] = extractelement <4 x i32> [[TMP7]], i32 1
+; CHECK-NEXT:    [[TMP22:%.*]] = extractelement <4 x i32> [[TMP7]], i32 0
+; CHECK-NEXT:    [[TMP23:%.*]] = extractelement <4 x i32> [[TMP6]], i32 3
+; CHECK-NEXT:    [[TMP24:%.*]] = extractelement <4 x i32> [[TMP6]], i32 2
+; CHECK-NEXT:    [[TMP25:%.*]] = extractelement <4 x i32> [[TMP6]], i32 1
+; CHECK-NEXT:    [[TMP26:%.*]] = extractelement <4 x i32> [[TMP6]], i32 0
+; CHECK-NEXT:    [[TMP27:%.*]] = extractelement <2 x i32> [[TMP5]], i32 1
+; CHECK-NEXT:    [[TMP28:%.*]] = extractelement <2 x i32> [[TMP5]], i32 0
+; CHECK-NEXT:    [[TMP29:%.*]] = extractelement <4 x i32> [[TMP2]], i32 3
+; CHECK-NEXT:    [[TMP30:%.*]] = extractelement <4 x i32> [[TMP2]], i32 2
+; CHECK-NEXT:    [[TMP31:%.*]] = extractelement <4 x i32> [[TMP2]], i32 1
+; CHECK-NEXT:    [[TMP32:%.*]] = extractelement <4 x i32> [[TMP2]], i32 0
+; CHECK-NEXT:    [[TMP33:%.*]] = and <16 x i8> [[TMP19]], <i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1>
 ; CHECK-NEXT:    [[ARRAYIDX_I_I7_15_I_I:%.*]] = getelementptr inbounds %"struct.std::array", %"struct.std::array"* undef, i64 0, i32 0, i64 15
-; CHECK-NEXT:    [[TMP37:%.*]] = bitcast i8* [[TMP0]] to <16 x i8>*
-; CHECK-NEXT:    store <16 x i8> [[TMP36]], <16 x i8>* [[TMP37]], align 1
+; CHECK-NEXT:    [[TMP34:%.*]] = bitcast i8* [[TMP0]] to <16 x i8>*
+; CHECK-NEXT:    store <16 x i8> [[TMP33]], <16 x i8>* [[TMP34]], align 1
 ; CHECK-NEXT:    unreachable
 ; CHECK:       if.end50.i:
 ; CHECK-NEXT:    ret void
 ;
 entry:
   br i1 undef, label %if.end50.i, label %if.then22.i
 
 if.then22.i:                                      ; preds = %entry

llvm/test/Transforms/SLPVectorizer/X86/saxpy.ll

 ; NOTE: Assertions have been autogenerated by utils/update_test_checks.py
 ; RUN: opt < %s -basic-aa -slp-vectorizer -dce -S -mtriple=x86_64-apple-macosx10.8.0 -mcpu=corei7-avx | FileCheck %s
 
 target datalayout = "e-p:64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:64:64-f32:32:32-f64:64:64-v64:64:64-v128:128:128-a0:0:64-s0:64:64-f80:128:128-n8:16:32:64-S128"
 target triple = "x86_64-apple-macosx10.8.0"
 
 ; SLP vectorization example from http://cs.stanford.edu/people/eschkufz/research/asplos291-schkufza.pdf
 define void @SAXPY(i32* noalias nocapture %x, i32* noalias nocapture %y, i32 %a, i64 %i) {
 ; CHECK-LABEL: @SAXPY(
 ; CHECK-NEXT:    [[TMP1:%.*]] = getelementptr inbounds i32, i32* [[X:%.*]], i64 [[I:%.*]]
 ; CHECK-NEXT:    [[TMP2:%.*]] = getelementptr inbounds i32, i32* [[Y:%.*]], i64 [[I]]
 ; CHECK-NEXT:    [[TMP3:%.*]] = bitcast i32* [[TMP1]] to <4 x i32>*
 ; CHECK-NEXT:    [[TMP4:%.*]] = load <4 x i32>, <4 x i32>* [[TMP3]], align 4
 ; CHECK-NEXT:    [[TMP5:%.*]] = insertelement <4 x i32> poison, i32 [[A:%.*]], i32 0
-; CHECK-NEXT:    [[TMP6:%.*]] = insertelement <4 x i32> [[TMP5]], i32 [[A]], i32 1
-; CHECK-NEXT:    [[TMP7:%.*]] = insertelement <4 x i32> [[TMP6]], i32 [[A]], i32 2
-; CHECK-NEXT:    [[TMP8:%.*]] = insertelement <4 x i32> [[TMP7]], i32 [[A]], i32 3
-; CHECK-NEXT:    [[TMP9:%.*]] = mul nsw <4 x i32> [[TMP4]], [[TMP8]]
-; CHECK-NEXT:    [[TMP10:%.*]] = bitcast i32* [[TMP2]] to <4 x i32>*
-; CHECK-NEXT:    [[TMP11:%.*]] = load <4 x i32>, <4 x i32>* [[TMP10]], align 4
-; CHECK-NEXT:    [[TMP12:%.*]] = add nsw <4 x i32> [[TMP9]], [[TMP11]]
-; CHECK-NEXT:    [[TMP13:%.*]] = bitcast i32* [[TMP1]] to <4 x i32>*
-; CHECK-NEXT:    store <4 x i32> [[TMP12]], <4 x i32>* [[TMP13]], align 4
+; CHECK-NEXT:    [[SHUFFLE:%.*]] = shufflevector <4 x i32> [[TMP5]], <4 x i32> poison, <4 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP6:%.*]] = mul nsw <4 x i32> [[TMP4]], [[SHUFFLE]]
+; CHECK-NEXT:    [[TMP7:%.*]] = bitcast i32* [[TMP2]] to <4 x i32>*
+; CHECK-NEXT:    [[TMP8:%.*]] = load <4 x i32>, <4 x i32>* [[TMP7]], align 4
+; CHECK-NEXT:    [[TMP9:%.*]] = add nsw <4 x i32> [[TMP6]], [[TMP8]]
+; CHECK-NEXT:    [[TMP10:%.*]] = bitcast i32* [[TMP1]] to <4 x i32>*
+; CHECK-NEXT:    store <4 x i32> [[TMP9]], <4 x i32>* [[TMP10]], align 4
 ; CHECK-NEXT:    ret void
 ;
   %1 = getelementptr inbounds i32, i32* %x, i64 %i
   %2 = load i32, i32* %1, align 4
   %3 = mul nsw i32 %2, %a
   %4 = getelementptr inbounds i32, i32* %y, i64 %i
   %5 = load i32, i32* %4, align 4
   %6 = add nsw i32 %3, %5

llvm/test/Transforms/SLPVectorizer/X86/tiny-tree.ll

 
 define void @store_splat(float*, float) {
 ; CHECK-LABEL: @store_splat(
 ; CHECK-NEXT:    [[TMP3:%.*]] = getelementptr inbounds float, float* [[TMP0:%.*]], i64 0
 ; CHECK-NEXT:    [[TMP4:%.*]] = getelementptr inbounds float, float* [[TMP0]], i64 1
 ; CHECK-NEXT:    [[TMP5:%.*]] = getelementptr inbounds float, float* [[TMP0]], i64 2
 ; CHECK-NEXT:    [[TMP6:%.*]] = getelementptr inbounds float, float* [[TMP0]], i64 3
 ; CHECK-NEXT:    [[TMP7:%.*]] = insertelement <4 x float> poison, float [[TMP1:%.*]], i32 0
-; CHECK-NEXT:    [[TMP8:%.*]] = insertelement <4 x float> [[TMP7]], float [[TMP1]], i32 1
-; CHECK-NEXT:    [[TMP9:%.*]] = insertelement <4 x float> [[TMP8]], float [[TMP1]], i32 2
-; CHECK-NEXT:    [[TMP10:%.*]] = insertelement <4 x float> [[TMP9]], float [[TMP1]], i32 3
-; CHECK-NEXT:    [[TMP11:%.*]] = bitcast float* [[TMP3]] to <4 x float>*
-; CHECK-NEXT:    store <4 x float> [[TMP10]], <4 x float>* [[TMP11]], align 4
+; CHECK-NEXT:    [[SHUFFLE:%.*]] = shufflevector <4 x float> [[TMP7]], <4 x float> poison, <4 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP8:%.*]] = bitcast float* [[TMP3]] to <4 x float>*
+; CHECK-NEXT:    store <4 x float> [[SHUFFLE]], <4 x float>* [[TMP8]], align 4
 ; CHECK-NEXT:    ret void
 ;
   %3 = getelementptr inbounds float, float* %0, i64 0
   store float %1, float* %3, align 4
   %4 = getelementptr inbounds float, float* %0, i64 1
   store float %1, float* %4, align 4
   %5 = getelementptr inbounds float, float* %0, i64 2
   store float %1, float* %5, align 4

llvm/test/Transforms/SLPVectorizer/X86/used-reduced-op.ll

 ; CHECK-NEXT:    [[TMP28:%.*]] = load i32, i32* getelementptr inbounds ([8 x [4 x i32]], [8 x [4 x i32]]* @k, i64 0, i64 7, i64 3), align 4
 ; CHECK-NEXT:    br label [[FOR_COND:%.*]]
 ; CHECK:       for.cond:
 ; CHECK-NEXT:    [[INDVARS_IV:%.*]] = phi i64 [ [[INDVARS_IV_NEXT:%.*]], [[FOR_COND]] ], [ 0, [[ENTRY:%.*]] ]
 ; CHECK-NEXT:    [[B_0:%.*]] = phi i32 [ [[SPEC_SELECT8_3_7:%.*]], [[FOR_COND]] ], [ undef, [[ENTRY]] ]
 ; CHECK-NEXT:    [[TMP29:%.*]] = trunc i64 [[INDVARS_IV]] to i32
 ; CHECK-NEXT:    [[TMP30:%.*]] = add i32 [[TMP29]], -183
 ; CHECK-NEXT:    [[TMP31:%.*]] = insertelement <4 x i32> poison, i32 [[TMP30]], i32 0
-; CHECK-NEXT:    [[TMP32:%.*]] = insertelement <4 x i32> [[TMP31]], i32 [[TMP30]], i32 1
-; CHECK-NEXT:    [[TMP33:%.*]] = insertelement <4 x i32> [[TMP32]], i32 [[TMP30]], i32 2
-; CHECK-NEXT:    [[TMP34:%.*]] = insertelement <4 x i32> [[TMP33]], i32 [[TMP30]], i32 3
-; CHECK-NEXT:    [[TMP35:%.*]] = sub <4 x i32> [[TMP34]], [[TMP0]]
-; CHECK-NEXT:    [[TMP36:%.*]] = icmp slt <4 x i32> [[TMP35]], zeroinitializer
-; CHECK-NEXT:    [[TMP37:%.*]] = sub nsw <4 x i32> zeroinitializer, [[TMP35]]
-; CHECK-NEXT:    [[TMP38:%.*]] = select <4 x i1> [[TMP36]], <4 x i32> [[TMP37]], <4 x i32> [[TMP35]]
-; CHECK-NEXT:    [[TMP39:%.*]] = call i32 @llvm.vector.reduce.smin.v4i32(<4 x i32> [[TMP38]])
-; CHECK-NEXT:    [[OP_EXTRA:%.*]] = icmp slt i32 [[TMP39]], [[B_0]]
-; CHECK-NEXT:    [[OP_EXTRA1:%.*]] = select i1 [[OP_EXTRA]], i32 [[TMP39]], i32 [[B_0]]
+; CHECK-NEXT:    [[SHUFFLE:%.*]] = shufflevector <4 x i32> [[TMP31]], <4 x i32> poison, <4 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP32:%.*]] = sub <4 x i32> [[SHUFFLE]], [[TMP0]]
+; CHECK-NEXT:    [[TMP33:%.*]] = icmp slt <4 x i32> [[TMP32]], zeroinitializer
+; CHECK-NEXT:    [[TMP34:%.*]] = sub nsw <4 x i32> zeroinitializer, [[TMP32]]
+; CHECK-NEXT:    [[TMP35:%.*]] = select <4 x i1> [[TMP33]], <4 x i32> [[TMP34]], <4 x i32> [[TMP32]]
+; CHECK-NEXT:    [[TMP36:%.*]] = call i32 @llvm.vector.reduce.smin.v4i32(<4 x i32> [[TMP35]])
+; CHECK-NEXT:    [[OP_EXTRA:%.*]] = icmp slt i32 [[TMP36]], [[B_0]]
+; CHECK-NEXT:    [[OP_EXTRA1:%.*]] = select i1 [[OP_EXTRA]], i32 [[TMP36]], i32 [[B_0]]
 ; CHECK-NEXT:    [[SUB_116:%.*]] = sub i32 [[TMP30]], [[TMP1]]
-; CHECK-NEXT:    [[TMP40:%.*]] = icmp slt i32 [[SUB_116]], 0
+; CHECK-NEXT:    [[TMP37:%.*]] = icmp slt i32 [[SUB_116]], 0
 ; CHECK-NEXT:    [[NEG_117:%.*]] = sub nsw i32 0, [[SUB_116]]
-; CHECK-NEXT:    [[TMP41:%.*]] = select i1 [[TMP40]], i32 [[NEG_117]], i32 [[SUB_116]]
-; CHECK-NEXT:    [[CMP12_118:%.*]] = icmp slt i32 [[TMP41]], [[OP_EXTRA1]]
-; CHECK-NEXT:    [[SPEC_SELECT8_120:%.*]] = select i1 [[CMP12_118]], i32 [[TMP41]], i32 [[OP_EXTRA1]]
+; CHECK-NEXT:    [[TMP38:%.*]] = select i1 [[TMP37]], i32 [[NEG_117]], i32 [[SUB_116]]
+; CHECK-NEXT:    [[CMP12_118:%.*]] = icmp slt i32 [[TMP38]], [[OP_EXTRA1]]
+; CHECK-NEXT:    [[SPEC_SELECT8_120:%.*]] = select i1 [[CMP12_118]], i32 [[TMP38]], i32 [[OP_EXTRA1]]
 ; CHECK-NEXT:    [[SUB_1_1:%.*]] = sub i32 [[TMP30]], [[TMP2]]
-; CHECK-NEXT:    [[TMP42:%.*]] = icmp slt i32 [[SUB_1_1]], 0
+; CHECK-NEXT:    [[TMP39:%.*]] = icmp slt i32 [[SUB_1_1]], 0
 ; CHECK-NEXT:    [[NEG_1_1:%.*]] = sub nsw i32 0, [[SUB_1_1]]
-; CHECK-NEXT:    [[TMP43:%.*]] = select i1 [[TMP42]], i32 [[NEG_1_1]], i32 [[SUB_1_1]]
-; CHECK-NEXT:    [[CMP12_1_1:%.*]] = icmp slt i32 [[TMP43]], [[SPEC_SELECT8_120]]
+; CHECK-NEXT:    [[TMP40:%.*]] = select i1 [[TMP39]], i32 [[NEG_1_1]], i32 [[SUB_1_1]]
+; CHECK-NEXT:    [[CMP12_1_1:%.*]] = icmp slt i32 [[TMP40]], [[SPEC_SELECT8_120]]
 ; CHECK-NEXT:    [[NARROW:%.*]] = or i1 [[CMP12_1_1]], [[CMP12_118]]
-; CHECK-NEXT:    [[SPEC_SELECT8_1_1:%.*]] = select i1 [[CMP12_1_1]], i32 [[TMP43]], i32 [[SPEC_SELECT8_120]]
+; CHECK-NEXT:    [[SPEC_SELECT8_1_1:%.*]] = select i1 [[CMP12_1_1]], i32 [[TMP40]], i32 [[SPEC_SELECT8_120]]
 ; CHECK-NEXT:    [[SUB_2_1:%.*]] = sub i32 [[TMP30]], [[TMP3]]
-; CHECK-NEXT:    [[TMP44:%.*]] = icmp slt i32 [[SUB_2_1]], 0
+; CHECK-NEXT:    [[TMP41:%.*]] = icmp slt i32 [[SUB_2_1]], 0
 ; CHECK-NEXT:    [[NEG_2_1:%.*]] = sub nsw i32 0, [[SUB_2_1]]
-; CHECK-NEXT:    [[TMP45:%.*]] = select i1 [[TMP44]], i32 [[NEG_2_1]], i32 [[SUB_2_1]]
-; CHECK-NEXT:    [[CMP12_2_1:%.*]] = icmp slt i32 [[TMP45]], [[SPEC_SELECT8_1_1]]
+; CHECK-NEXT:    [[TMP42:%.*]] = select i1 [[TMP41]], i32 [[NEG_2_1]], i32 [[SUB_2_1]]
+; CHECK-NEXT:    [[CMP12_2_1:%.*]] = icmp slt i32 [[TMP42]], [[SPEC_SELECT8_1_1]]
 ; CHECK-NEXT:    [[NARROW34:%.*]] = or i1 [[CMP12_2_1]], [[NARROW]]
-; CHECK-NEXT:    [[SPEC_SELECT8_2_1:%.*]] = select i1 [[CMP12_2_1]], i32 [[TMP45]], i32 [[SPEC_SELECT8_1_1]]
+; CHECK-NEXT:    [[SPEC_SELECT8_2_1:%.*]] = select i1 [[CMP12_2_1]], i32 [[TMP42]], i32 [[SPEC_SELECT8_1_1]]
 ; CHECK-NEXT:    [[SUB_3_1:%.*]] = sub i32 [[TMP30]], [[TMP4]]
-; CHECK-NEXT:    [[TMP46:%.*]] = icmp slt i32 [[SUB_3_1]], 0
+; CHECK-NEXT:    [[TMP43:%.*]] = icmp slt i32 [[SUB_3_1]], 0
 ; CHECK-NEXT:    [[NEG_3_1:%.*]] = sub nsw i32 0, [[SUB_3_1]]
-; CHECK-NEXT:    [[TMP47:%.*]] = select i1 [[TMP46]], i32 [[NEG_3_1]], i32 [[SUB_3_1]]
-; CHECK-NEXT:    [[CMP12_3_1:%.*]] = icmp slt i32 [[TMP47]], [[SPEC_SELECT8_2_1]]
+; CHECK-NEXT:    [[TMP44:%.*]] = select i1 [[TMP43]], i32 [[NEG_3_1]], i32 [[SUB_3_1]]
+; CHECK-NEXT:    [[CMP12_3_1:%.*]] = icmp slt i32 [[TMP44]], [[SPEC_SELECT8_2_1]]
 ; CHECK-NEXT:    [[NARROW35:%.*]] = or i1 [[CMP12_3_1]], [[NARROW34]]
 ; CHECK-NEXT:    [[SPEC_SELECT_3_1:%.*]] = zext i1 [[NARROW35]] to i32
-; CHECK-NEXT:    [[SPEC_SELECT8_3_1:%.*]] = select i1 [[CMP12_3_1]], i32 [[TMP47]], i32 [[SPEC_SELECT8_2_1]]
+; CHECK-NEXT:    [[SPEC_SELECT8_3_1:%.*]] = select i1 [[CMP12_3_1]], i32 [[TMP44]], i32 [[SPEC_SELECT8_2_1]]
 ; CHECK-NEXT:    [[SUB_222:%.*]] = sub i32 [[TMP30]], [[TMP5]]
-; CHECK-NEXT:    [[TMP48:%.*]] = icmp slt i32 [[SUB_222]], 0
+; CHECK-NEXT:    [[TMP45:%.*]] = icmp slt i32 [[SUB_222]], 0
 ; CHECK-NEXT:    [[NEG_223:%.*]] = sub nsw i32 0, [[SUB_222]]
-; CHECK-NEXT:    [[TMP49:%.*]] = select i1 [[TMP48]], i32 [[NEG_223]], i32 [[SUB_222]]
-; CHECK-NEXT:    [[CMP12_224:%.*]] = icmp slt i32 [[TMP49]], [[SPEC_SELECT8_3_1]]
-; CHECK-NEXT:    [[SPEC_SELECT8_226:%.*]] = select i1 [[CMP12_224]], i32 [[TMP49]], i32 [[SPEC_SELECT8_3_1]]
+; CHECK-NEXT:    [[TMP46:%.*]] = select i1 [[TMP45]], i32 [[NEG_223]], i32 [[SUB_222]]
+; CHECK-NEXT:    [[CMP12_224:%.*]] = icmp slt i32 [[TMP46]], [[SPEC_SELECT8_3_1]]
+; CHECK-NEXT:    [[SPEC_SELECT8_226:%.*]] = select i1 [[CMP12_224]], i32 [[TMP46]], i32 [[SPEC_SELECT8_3_1]]
 ; CHECK-NEXT:    [[SUB_1_2:%.*]] = sub i32 [[TMP30]], [[TMP6]]
-; CHECK-NEXT:    [[TMP50:%.*]] = icmp slt i32 [[SUB_1_2]], 0
+; CHECK-NEXT:    [[TMP47:%.*]] = icmp slt i32 [[SUB_1_2]], 0
 ; CHECK-NEXT:    [[NEG_1_2:%.*]] = sub nsw i32 0, [[SUB_1_2]]
-; CHECK-NEXT:    [[TMP51:%.*]] = select i1 [[TMP50]], i32 [[NEG_1_2]], i32 [[SUB_1_2]]
-; CHECK-NEXT:    [[CMP12_1_2:%.*]] = icmp slt i32 [[TMP51]], [[SPEC_SELECT8_226]]
-; CHECK-NEXT:    [[TMP52:%.*]] = or i1 [[CMP12_1_2]], [[CMP12_224]]
-; CHECK-NEXT:    [[SPEC_SELECT8_1_2:%.*]] = select i1 [[CMP12_1_2]], i32 [[TMP51]], i32 [[SPEC_SELECT8_226]]
+; CHECK-NEXT:    [[TMP48:%.*]] = select i1 [[TMP47]], i32 [[NEG_1_2]], i32 [[SUB_1_2]]
+; CHECK-NEXT:    [[CMP12_1_2:%.*]] = icmp slt i32 [[TMP48]], [[SPEC_SELECT8_226]]
+; CHECK-NEXT:    [[TMP49:%.*]] = or i1 [[CMP12_1_2]], [[CMP12_224]]
+; CHECK-NEXT:    [[SPEC_SELECT8_1_2:%.*]] = select i1 [[CMP12_1_2]], i32 [[TMP48]], i32 [[SPEC_SELECT8_226]]
 ; CHECK-NEXT:    [[SUB_2_2:%.*]] = sub i32 [[TMP30]], [[TMP7]]
-; CHECK-NEXT:    [[TMP53:%.*]] = icmp slt i32 [[SUB_2_2]], 0
+; CHECK-NEXT:    [[TMP50:%.*]] = icmp slt i32 [[SUB_2_2]], 0
 ; CHECK-NEXT:    [[NEG_2_2:%.*]] = sub nsw i32 0, [[SUB_2_2]]
-; CHECK-NEXT:    [[TMP54:%.*]] = select i1 [[TMP53]], i32 [[NEG_2_2]], i32 [[SUB_2_2]]
-; CHECK-NEXT:    [[CMP12_2_2:%.*]] = icmp slt i32 [[TMP54]], [[SPEC_SELECT8_1_2]]
-; CHECK-NEXT:    [[TMP55:%.*]] = or i1 [[CMP12_2_2]], [[TMP52]]
-; CHECK-NEXT:    [[SPEC_SELECT8_2_2:%.*]] = select i1 [[CMP12_2_2]], i32 [[TMP54]], i32 [[SPEC_SELECT8_1_2]]
+; CHECK-NEXT:    [[TMP51:%.*]] = select i1 [[TMP50]], i32 [[NEG_2_2]], i32 [[SUB_2_2]]
+; CHECK-NEXT:    [[CMP12_2_2:%.*]] = icmp slt i32 [[TMP51]], [[SPEC_SELECT8_1_2]]
+; CHECK-NEXT:    [[TMP52:%.*]] = or i1 [[CMP12_2_2]], [[TMP49]]
+; CHECK-NEXT:    [[SPEC_SELECT8_2_2:%.*]] = select i1 [[CMP12_2_2]], i32 [[TMP51]], i32 [[SPEC_SELECT8_1_2]]
 ; CHECK-NEXT:    [[SUB_3_2:%.*]] = sub i32 [[TMP30]], [[TMP8]]
-; CHECK-NEXT:    [[TMP56:%.*]] = icmp slt i32 [[SUB_3_2]], 0
+; CHECK-NEXT:    [[TMP53:%.*]] = icmp slt i32 [[SUB_3_2]], 0
 ; CHECK-NEXT:    [[NEG_3_2:%.*]] = sub nsw i32 0, [[SUB_3_2]]
-; CHECK-NEXT:    [[TMP57:%.*]] = select i1 [[TMP56]], i32 [[NEG_3_2]], i32 [[SUB_3_2]]
-; CHECK-NEXT:    [[CMP12_3_2:%.*]] = icmp slt i32 [[TMP57]], [[SPEC_SELECT8_2_2]]
-; CHECK-NEXT:    [[TMP58:%.*]] = or i1 [[CMP12_3_2]], [[TMP55]]
-; CHECK-NEXT:    [[SPEC_SELECT_3_2:%.*]] = select i1 [[TMP58]], i32 2, i32 [[SPEC_SELECT_3_1]]
-; CHECK-NEXT:    [[SPEC_SELECT8_3_2:%.*]] = select i1 [[CMP12_3_2]], i32 [[TMP57]], i32 [[SPEC_SELECT8_2_2]]
+; CHECK-NEXT:    [[TMP54:%.*]] = select i1 [[TMP53]], i32 [[NEG_3_2]], i32 [[SUB_3_2]]
+; CHECK-NEXT:    [[CMP12_3_2:%.*]] = icmp slt i32 [[TMP54]], [[SPEC_SELECT8_2_2]]
+; CHECK-NEXT:    [[TMP55:%.*]] = or i1 [[CMP12_3_2]], [[TMP52]]
+; CHECK-NEXT:    [[SPEC_SELECT_3_2:%.*]] = select i1 [[TMP55]], i32 2, i32 [[SPEC_SELECT_3_1]]
+; CHECK-NEXT:    [[SPEC_SELECT8_3_2:%.*]] = select i1 [[CMP12_3_2]], i32 [[TMP54]], i32 [[SPEC_SELECT8_2_2]]
 ; CHECK-NEXT:    [[SUB_328:%.*]] = sub i32 [[TMP30]], [[TMP9]]
-; CHECK-NEXT:    [[TMP59:%.*]] = icmp slt i32 [[SUB_328]], 0
+; CHECK-NEXT:    [[TMP56:%.*]] = icmp slt i32 [[SUB_328]], 0
 ; CHECK-NEXT:    [[NEG_329:%.*]] = sub nsw i32 0, [[SUB_328]]
-; CHECK-NEXT:    [[TMP60:%.*]] = select i1 [[TMP59]], i32 [[NEG_329]], i32 [[SUB_328]]
-; CHECK-NEXT:    [[CMP12_330:%.*]] = icmp slt i32 [[TMP60]], [[SPEC_SELECT8_3_2]]
-; CHECK-NEXT:    [[SPEC_SELECT8_332:%.*]] = select i1 [[CMP12_330]], i32 [[TMP60]], i32 [[SPEC_SELECT8_3_2]]
+; CHECK-NEXT:    [[TMP57:%.*]] = select i1 [[TMP56]], i32 [[NEG_329]], i32 [[SUB_328]]
+; CHECK-NEXT:    [[CMP12_330:%.*]] = icmp slt i32 [[TMP57]], [[SPEC_SELECT8_3_2]]
+; CHECK-NEXT:    [[SPEC_SELECT8_332:%.*]] = select i1 [[CMP12_330]], i32 [[TMP57]], i32 [[SPEC_SELECT8_3_2]]
 ; CHECK-NEXT:    [[SUB_1_3:%.*]] = sub i32 [[TMP30]], [[TMP10]]
-; CHECK-NEXT:    [[TMP61:%.*]] = icmp slt i32 [[SUB_1_3]], 0
+; CHECK-NEXT:    [[TMP58:%.*]] = icmp slt i32 [[SUB_1_3]], 0
 ; CHECK-NEXT:    [[NEG_1_3:%.*]] = sub nsw i32 0, [[SUB_1_3]]
-; CHECK-NEXT:    [[TMP62:%.*]] = select i1 [[TMP61]], i32 [[NEG_1_3]], i32 [[SUB_1_3]]
-; CHECK-NEXT:    [[CMP12_1_3:%.*]] = icmp slt i32 [[TMP62]], [[SPEC_SELECT8_332]]
-; CHECK-NEXT:    [[TMP63:%.*]] = or i1 [[CMP12_1_3]], [[CMP12_330]]
-; CHECK-NEXT:    [[SPEC_SELECT8_1_3:%.*]] = select i1 [[CMP12_1_3]], i32 [[TMP62]], i32 [[SPEC_SELECT8_332]]
+; CHECK-NEXT:    [[TMP59:%.*]] = select i1 [[TMP58]], i32 [[NEG_1_3]], i32 [[SUB_1_3]]
+; CHECK-NEXT:    [[CMP12_1_3:%.*]] = icmp slt i32 [[TMP59]], [[SPEC_SELECT8_332]]
+; CHECK-NEXT:    [[TMP60:%.*]] = or i1 [[CMP12_1_3]], [[CMP12_330]]
+; CHECK-NEXT:    [[SPEC_SELECT8_1_3:%.*]] = select i1 [[CMP12_1_3]], i32 [[TMP59]], i32 [[SPEC_SELECT8_332]]
 ; CHECK-NEXT:    [[SUB_2_3:%.*]] = sub i32 [[TMP30]], [[TMP11]]
-; CHECK-NEXT:    [[TMP64:%.*]] = icmp slt i32 [[SUB_2_3]], 0
+; CHECK-NEXT:    [[TMP61:%.*]] = icmp slt i32 [[SUB_2_3]], 0
 ; CHECK-NEXT:    [[NEG_2_3:%.*]] = sub nsw i32 0, [[SUB_2_3]]
-; CHECK-NEXT:    [[TMP65:%.*]] = select i1 [[TMP64]], i32 [[NEG_2_3]], i32 [[SUB_2_3]]
-; CHECK-NEXT:    [[CMP12_2_3:%.*]] = icmp slt i32 [[TMP65]], [[SPEC_SELECT8_1_3]]
-; CHECK-NEXT:    [[TMP66:%.*]] = or i1 [[CMP12_2_3]], [[TMP63]]
-; CHECK-NEXT:    [[SPEC_SELECT8_2_3:%.*]] = select i1 [[CMP12_2_3]], i32 [[TMP65]], i32 [[SPEC_SELECT8_1_3]]
+; CHECK-NEXT:    [[TMP62:%.*]] = select i1 [[TMP61]], i32 [[NEG_2_3]], i32 [[SUB_2_3]]
+; CHECK-NEXT:    [[CMP12_2_3:%.*]] = icmp slt i32 [[TMP62]], [[SPEC_SELECT8_1_3]]
+; CHECK-NEXT:    [[TMP63:%.*]] = or i1 [[CMP12_2_3]], [[TMP60]]
+; CHECK-NEXT:    [[SPEC_SELECT8_2_3:%.*]] = select i1 [[CMP12_2_3]], i32 [[TMP62]], i32 [[SPEC_SELECT8_1_3]]
 ; CHECK-NEXT:    [[SUB_3_3:%.*]] = sub i32 [[TMP30]], [[TMP12]]
-; CHECK-NEXT:    [[TMP67:%.*]] = icmp slt i32 [[SUB_3_3]], 0
+; CHECK-NEXT:    [[TMP64:%.*]] = icmp slt i32 [[SUB_3_3]], 0
 ; CHECK-NEXT:    [[NEG_3_3:%.*]] = sub nsw i32 0, [[SUB_3_3]]
-; CHECK-NEXT:    [[TMP68:%.*]] = select i1 [[TMP67]], i32 [[NEG_3_3]], i32 [[SUB_3_3]]
-; CHECK-NEXT:    [[CMP12_3_3:%.*]] = icmp slt i32 [[TMP68]], [[SPEC_SELECT8_2_3]]
-; CHECK-NEXT:    [[TMP69:%.*]] = or i1 [[CMP12_3_3]], [[TMP66]]
-; CHECK-NEXT:    [[SPEC_SELECT_3_3:%.*]] = select i1 [[TMP69]], i32 3, i32 [[SPEC_SELECT_3_2]]
-; CHECK-NEXT:    [[SPEC_SELECT8_3_3:%.*]] = select i1 [[CMP12_3_3]], i32 [[TMP68]], i32 [[SPEC_SELECT8_2_3]]
+; CHECK-NEXT:    [[TMP65:%.*]] = select i1 [[TMP64]], i32 [[NEG_3_3]], i32 [[SUB_3_3]]
+; CHECK-NEXT:    [[CMP12_3_3:%.*]] = icmp slt i32 [[TMP65]], [[SPEC_SELECT8_2_3]]
+; CHECK-NEXT:    [[TMP66:%.*]] = or i1 [[CMP12_3_3]], [[TMP63]]
+; CHECK-NEXT:    [[SPEC_SELECT_3_3:%.*]] = select i1 [[TMP66]], i32 3, i32 [[SPEC_SELECT_3_2]]
+; CHECK-NEXT:    [[SPEC_SELECT8_3_3:%.*]] = select i1 [[CMP12_3_3]], i32 [[TMP65]], i32 [[SPEC_SELECT8_2_3]]
 ; CHECK-NEXT:    [[SUB_4:%.*]] = sub i32 [[TMP30]], [[TMP13]]
-; CHECK-NEXT:    [[TMP70:%.*]] = icmp slt i32 [[SUB_4]], 0
+; CHECK-NEXT:    [[TMP67:%.*]] = icmp slt i32 [[SUB_4]], 0
 ; CHECK-NEXT:    [[NEG_4:%.*]] = sub nsw i32 0, [[SUB_4]]
-; CHECK-NEXT:    [[TMP71:%.*]] = select i1 [[TMP70]], i32 [[NEG_4]], i32 [[SUB_4]]
-; CHECK-NEXT:    [[CMP12_4:%.*]] = icmp slt i32 [[TMP71]], [[SPEC_SELECT8_3_3]]
-; CHECK-NEXT:    [[SPEC_SELECT8_4:%.*]] = select i1 [[CMP12_4]], i32 [[TMP71]], i32 [[SPEC_SELECT8_3_3]]
+; CHECK-NEXT:    [[TMP68:%.*]] = select i1 [[TMP67]], i32 [[NEG_4]], i32 [[SUB_4]]
+; CHECK-NEXT:    [[CMP12_4:%.*]] = icmp slt i32 [[TMP68]], [[SPEC_SELECT8_3_3]]
+; CHECK-NEXT:    [[SPEC_SELECT8_4:%.*]] = select i1 [[CMP12_4]], i32 [[TMP68]], i32 [[SPEC_SELECT8_3_3]]
 ; CHECK-NEXT:    [[SUB_1_4:%.*]] = sub i32 [[TMP30]], [[TMP14]]
-; CHECK-NEXT:    [[TMP72:%.*]] = icmp slt i32 [[SUB_1_4]], 0
+; CHECK-NEXT:    [[TMP69:%.*]] = icmp slt i32 [[SUB_1_4]], 0
 ; CHECK-NEXT:    [[NEG_1_4:%.*]] = sub nsw i32 0, [[SUB_1_4]]
-; CHECK-NEXT:    [[TMP73:%.*]] = select i1 [[TMP72]], i32 [[NEG_1_4]], i32 [[SUB_1_4]]
-; CHECK-NEXT:    [[CMP12_1_4:%.*]] = icmp slt i32 [[TMP73]], [[SPEC_SELECT8_4]]
-; CHECK-NEXT:    [[TMP74:%.*]] = or i1 [[CMP12_1_4]], [[CMP12_4]]
-; CHECK-NEXT:    [[SPEC_SELECT8_1_4:%.*]] = select i1 [[CMP12_1_4]], i32 [[TMP73]], i32 [[SPEC_SELECT8_4]]
+; CHECK-NEXT:    [[TMP70:%.*]] = select i1 [[TMP69]], i32 [[NEG_1_4]], i32 [[SUB_1_4]]
+; CHECK-NEXT:    [[CMP12_1_4:%.*]] = icmp slt i32 [[TMP70]], [[SPEC_SELECT8_4]]
+; CHECK-NEXT:    [[TMP71:%.*]] = or i1 [[CMP12_1_4]], [[CMP12_4]]
+; CHECK-NEXT:    [[SPEC_SELECT8_1_4:%.*]] = select i1 [[CMP12_1_4]], i32 [[TMP70]], i32 [[SPEC_SELECT8_4]]
 ; CHECK-NEXT:    [[SUB_2_4:%.*]] = sub i32 [[TMP30]], [[TMP15]]
-; CHECK-NEXT:    [[TMP75:%.*]] = icmp slt i32 [[SUB_2_4]], 0
+; CHECK-NEXT:    [[TMP72:%.*]] = icmp slt i32 [[SUB_2_4]], 0
 ; CHECK-NEXT:    [[NEG_2_4:%.*]] = sub nsw i32 0, [[SUB_2_4]]
-; CHECK-NEXT:    [[TMP76:%.*]] = select i1 [[TMP75]], i32 [[NEG_2_4]], i32 [[SUB_2_4]]
-; CHECK-NEXT:    [[CMP12_2_4:%.*]] = icmp slt i32 [[TMP76]], [[SPEC_SELECT8_1_4]]
-; CHECK-NEXT:    [[TMP77:%.*]] = or i1 [[CMP12_2_4]], [[TMP74]]
-; CHECK-NEXT:    [[SPEC_SELECT8_2_4:%.*]] = select i1 [[CMP12_2_4]], i32 [[TMP76]], i32 [[SPEC_SELECT8_1_4]]
+; CHECK-NEXT:    [[TMP73:%.*]] = select i1 [[TMP72]], i32 [[NEG_2_4]], i32 [[SUB_2_4]]
+; CHECK-NEXT:    [[CMP12_2_4:%.*]] = icmp slt i32 [[TMP73]], [[SPEC_SELECT8_1_4]]
+; CHECK-NEXT:    [[TMP74:%.*]] = or i1 [[CMP12_2_4]], [[TMP71]]
+; CHECK-NEXT:    [[SPEC_SELECT8_2_4:%.*]] = select i1 [[CMP12_2_4]], i32 [[TMP73]], i32 [[SPEC_SELECT8_1_4]]
 ; CHECK-NEXT:    [[SUB_3_4:%.*]] = sub i32 [[TMP30]], [[TMP16]]
-; CHECK-NEXT:    [[TMP78:%.*]] = icmp slt i32 [[SUB_3_4]], 0
+; CHECK-NEXT:    [[TMP75:%.*]] = icmp slt i32 [[SUB_3_4]], 0
 ; CHECK-NEXT:    [[NEG_3_4:%.*]] = sub nsw i32 0, [[SUB_3_4]]
-; CHECK-NEXT:    [[TMP79:%.*]] = select i1 [[TMP78]], i32 [[NEG_3_4]], i32 [[SUB_3_4]]
-; CHECK-NEXT:    [[CMP12_3_4:%.*]] = icmp slt i32 [[TMP79]], [[SPEC_SELECT8_2_4]]
-; CHECK-NEXT:    [[TMP80:%.*]] = or i1 [[CMP12_3_4]], [[TMP77]]
-; CHECK-NEXT:    [[SPEC_SELECT_3_4:%.*]] = select i1 [[TMP80]], i32 4, i32 [[SPEC_SELECT_3_3]]
-; CHECK-NEXT:    [[SPEC_SELECT8_3_4:%.*]] = select i1 [[CMP12_3_4]], i32 [[TMP79]], i32 [[SPEC_SELECT8_2_4]]
+; CHECK-NEXT:    [[TMP76:%.*]] = select i1 [[TMP75]], i32 [[NEG_3_4]], i32 [[SUB_3_4]]
+; CHECK-NEXT:    [[CMP12_3_4:%.*]] = icmp slt i32 [[TMP76]], [[SPEC_SELECT8_2_4]]
+; CHECK-NEXT:    [[TMP77:%.*]] = or i1 [[CMP12_3_4]], [[TMP74]]
+; CHECK-NEXT:    [[SPEC_SELECT_3_4:%.*]] = select i1 [[TMP77]], i32 4, i32 [[SPEC_SELECT_3_3]]
+; CHECK-NEXT:    [[SPEC_SELECT8_3_4:%.*]] = select i1 [[CMP12_3_4]], i32 [[TMP76]], i32 [[SPEC_SELECT8_2_4]]
 ; CHECK-NEXT:    [[SUB_5:%.*]] = sub i32 [[TMP30]], [[TMP17]]
-; CHECK-NEXT:    [[TMP81:%.*]] = icmp slt i32 [[SUB_5]], 0
+; CHECK-NEXT:    [[TMP78:%.*]] = icmp slt i32 [[SUB_5]], 0
 ; CHECK-NEXT:    [[NEG_5:%.*]] = sub nsw i32 0, [[SUB_5]]
-; CHECK-NEXT:    [[TMP82:%.*]] = select i1 [[TMP81]], i32 [[NEG_5]], i32 [[SUB_5]]
-; CHECK-NEXT:    [[CMP12_5:%.*]] = icmp slt i32 [[TMP82]], [[SPEC_SELECT8_3_4]]
-; CHECK-NEXT:    [[SPEC_SELECT8_5:%.*]] = select i1 [[CMP12_5]], i32 [[TMP82]], i32 [[SPEC_SELECT8_3_4]]
+; CHECK-NEXT:    [[TMP79:%.*]] = select i1 [[TMP78]], i32 [[NEG_5]], i32 [[SUB_5]]
+; CHECK-NEXT:    [[CMP12_5:%.*]] = icmp slt i32 [[TMP79]], [[SPEC_SELECT8_3_4]]
+; CHECK-NEXT:    [[SPEC_SELECT8_5:%.*]] = select i1 [[CMP12_5]], i32 [[TMP79]], i32 [[SPEC_SELECT8_3_4]]
 ; CHECK-NEXT:    [[SUB_1_5:%.*]] = sub i32 [[TMP30]], [[TMP18]]
-; CHECK-NEXT:    [[TMP83:%.*]] = icmp slt i32 [[SUB_1_5]], 0
+; CHECK-NEXT:    [[TMP80:%.*]] = icmp slt i32 [[SUB_1_5]], 0
 ; CHECK-NEXT:    [[NEG_1_5:%.*]] = sub nsw i32 0, [[SUB_1_5]]
-; CHECK-NEXT:    [[TMP84:%.*]] = select i1 [[TMP83]], i32 [[NEG_1_5]], i32 [[SUB_1_5]]
-; CHECK-NEXT:    [[CMP12_1_5:%.*]] = icmp slt i32 [[TMP84]], [[SPEC_SELECT8_5]]
-; CHECK-NEXT:    [[TMP85:%.*]] = or i1 [[CMP12_1_5]], [[CMP12_5]]
-; CHECK-NEXT:    [[SPEC_SELECT8_1_5:%.*]] = select i1 [[CMP12_1_5]], i32 [[TMP84]], i32 [[SPEC_SELECT8_5]]
+; CHECK-NEXT:    [[TMP81:%.*]] = select i1 [[TMP80]], i32 [[NEG_1_5]], i32 [[SUB_1_5]]
+; CHECK-NEXT:    [[CMP12_1_5:%.*]] = icmp slt i32 [[TMP81]], [[SPEC_SELECT8_5]]
+; CHECK-NEXT:    [[TMP82:%.*]] = or i1 [[CMP12_1_5]], [[CMP12_5]]
+; CHECK-NEXT:    [[SPEC_SELECT8_1_5:%.*]] = select i1 [[CMP12_1_5]], i32 [[TMP81]], i32 [[SPEC_SELECT8_5]]
 ; CHECK-NEXT:    [[SUB_2_5:%.*]] = sub i32 [[TMP30]], [[TMP19]]
-; CHECK-NEXT:    [[TMP86:%.*]] = icmp slt i32 [[SUB_2_5]], 0
+; CHECK-NEXT:    [[TMP83:%.*]] = icmp slt i32 [[SUB_2_5]], 0
 ; CHECK-NEXT:    [[NEG_2_5:%.*]] = sub nsw i32 0, [[SUB_2_5]]
-; CHECK-NEXT:    [[TMP87:%.*]] = select i1 [[TMP86]], i32 [[NEG_2_5]], i32 [[SUB_2_5]]
-; CHECK-NEXT:    [[CMP12_2_5:%.*]] = icmp slt i32 [[TMP87]], [[SPEC_SELECT8_1_5]]
-; CHECK-NEXT:    [[TMP88:%.*]] = or i1 [[CMP12_2_5]], [[TMP85]]
-; CHECK-NEXT:    [[SPEC_SELECT8_2_5:%.*]] = select i1 [[CMP12_2_5]], i32 [[TMP87]], i32 [[SPEC_SELECT8_1_5]]
+; CHECK-NEXT:    [[TMP84:%.*]] = select i1 [[TMP83]], i32 [[NEG_2_5]], i32 [[SUB_2_5]]
+; CHECK-NEXT:    [[CMP12_2_5:%.*]] = icmp slt i32 [[TMP84]], [[SPEC_SELECT8_1_5]]
+; CHECK-NEXT:    [[TMP85:%.*]] = or i1 [[CMP12_2_5]], [[TMP82]]
+; CHECK-NEXT:    [[SPEC_SELECT8_2_5:%.*]] = select i1 [[CMP12_2_5]], i32 [[TMP84]], i32 [[SPEC_SELECT8_1_5]]
 ; CHECK-NEXT:    [[SUB_3_5:%.*]] = sub i32 [[TMP30]], [[TMP20]]
-; CHECK-NEXT:    [[TMP89:%.*]] = icmp slt i32 [[SUB_3_5]], 0
+; CHECK-NEXT:    [[TMP86:%.*]] = icmp slt i32 [[SUB_3_5]], 0
 ; CHECK-NEXT:    [[NEG_3_5:%.*]] = sub nsw i32 0, [[SUB_3_5]]
-; CHECK-NEXT:    [[TMP90:%.*]] = select i1 [[TMP89]], i32 [[NEG_3_5]], i32 [[SUB_3_5]]
-; CHECK-NEXT:    [[CMP12_3_5:%.*]] = icmp slt i32 [[TMP90]], [[SPEC_SELECT8_2_5]]
-; CHECK-NEXT:    [[TMP91:%.*]] = or i1 [[CMP12_3_5]], [[TMP88]]
-; CHECK-NEXT:    [[SPEC_SELECT_3_5:%.*]] = select i1 [[TMP91]], i32 5, i32 [[SPEC_SELECT_3_4]]
-; CHECK-NEXT:    [[SPEC_SELECT8_3_5:%.*]] = select i1 [[CMP12_3_5]], i32 [[TMP90]], i32 [[SPEC_SELECT8_2_5]]
+; CHECK-NEXT:    [[TMP87:%.*]] = select i1 [[TMP86]], i32 [[NEG_3_5]], i32 [[SUB_3_5]]
+; CHECK-NEXT:    [[CMP12_3_5:%.*]] = icmp slt i32 [[TMP87]], [[SPEC_SELECT8_2_5]]
+; CHECK-NEXT:    [[TMP88:%.*]] = or i1 [[CMP12_3_5]], [[TMP85]]
+; CHECK-NEXT:    [[SPEC_SELECT_3_5:%.*]] = select i1 [[TMP88]], i32 5, i32 [[SPEC_SELECT_3_4]]
+; CHECK-NEXT:    [[SPEC_SELECT8_3_5:%.*]] = select i1 [[CMP12_3_5]], i32 [[TMP87]], i32 [[SPEC_SELECT8_2_5]]
 ; CHECK-NEXT:    [[SUB_6:%.*]] = sub i32 [[TMP30]], [[TMP21]]
-; CHECK-NEXT:    [[TMP92:%.*]] = icmp slt i32 [[SUB_6]], 0
+; CHECK-NEXT:    [[TMP89:%.*]] = icmp slt i32 [[SUB_6]], 0
 ; CHECK-NEXT:    [[NEG_6:%.*]] = sub nsw i32 0, [[SUB_6]]
-; CHECK-NEXT:    [[TMP93:%.*]] = select i1 [[TMP92]], i32 [[NEG_6]], i32 [[SUB_6]]
-; CHECK-NEXT:    [[CMP12_6:%.*]] = icmp slt i32 [[TMP93]], [[SPEC_SELECT8_3_5]]
-; CHECK-NEXT:    [[SPEC_SELECT8_6:%.*]] = select i1 [[CMP12_6]], i32 [[TMP93]], i32 [[SPEC_SELECT8_3_5]]
+; CHECK-NEXT:    [[TMP90:%.*]] = select i1 [[TMP89]], i32 [[NEG_6]], i32 [[SUB_6]]
+; CHECK-NEXT:    [[CMP12_6:%.*]] = icmp slt i32 [[TMP90]], [[SPEC_SELECT8_3_5]]
+; CHECK-NEXT:    [[SPEC_SELECT8_6:%.*]] = select i1 [[CMP12_6]], i32 [[TMP90]], i32 [[SPEC_SELECT8_3_5]]
 ; CHECK-NEXT:    [[SUB_1_6:%.*]] = sub i32 [[TMP30]], [[TMP22]]
-; CHECK-NEXT:    [[TMP94:%.*]] = icmp slt i32 [[SUB_1_6]], 0
+; CHECK-NEXT:    [[TMP91:%.*]] = icmp slt i32 [[SUB_1_6]], 0
 ; CHECK-NEXT:    [[NEG_1_6:%.*]] = sub nsw i32 0, [[SUB_1_6]]
-; CHECK-NEXT:    [[TMP95:%.*]] = select i1 [[TMP94]], i32 [[NEG_1_6]], i32 [[SUB_1_6]]
-; CHECK-NEXT:    [[CMP12_1_6:%.*]] = icmp slt i32 [[TMP95]], [[SPEC_SELECT8_6]]
-; CHECK-NEXT:    [[TMP96:%.*]] = or i1 [[CMP12_1_6]], [[CMP12_6]]
-; CHECK-NEXT:    [[SPEC_SELECT8_1_6:%.*]] = select i1 [[CMP12_1_6]], i32 [[TMP95]], i32 [[SPEC_SELECT8_6]]
+; CHECK-NEXT:    [[TMP92:%.*]] = select i1 [[TMP91]], i32 [[NEG_1_6]], i32 [[SUB_1_6]]
+; CHECK-NEXT:    [[CMP12_1_6:%.*]] = icmp slt i32 [[TMP92]], [[SPEC_SELECT8_6]]
+; CHECK-NEXT:    [[TMP93:%.*]] = or i1 [[CMP12_1_6]], [[CMP12_6]]
+; CHECK-NEXT:    [[SPEC_SELECT8_1_6:%.*]] = select i1 [[CMP12_1_6]], i32 [[TMP92]], i32 [[SPEC_SELECT8_6]]
 ; CHECK-NEXT:    [[SUB_2_6:%.*]] = sub i32 [[TMP30]], [[TMP23]]
-; CHECK-NEXT:    [[TMP97:%.*]] = icmp slt i32 [[SUB_2_6]], 0
+; CHECK-NEXT:    [[TMP94:%.*]] = icmp slt i32 [[SUB_2_6]], 0
 ; CHECK-NEXT:    [[NEG_2_6:%.*]] = sub nsw i32 0, [[SUB_2_6]]
-; CHECK-NEXT:    [[TMP98:%.*]] = select i1 [[TMP97]], i32 [[NEG_2_6]], i32 [[SUB_2_6]]
-; CHECK-NEXT:    [[CMP12_2_6:%.*]] = icmp slt i32 [[TMP98]], [[SPEC_SELECT8_1_6]]
-; CHECK-NEXT:    [[TMP99:%.*]] = or i1 [[CMP12_2_6]], [[TMP96]]
-; CHECK-NEXT:    [[SPEC_SELECT8_2_6:%.*]] = select i1 [[CMP12_2_6]], i32 [[TMP98]], i32 [[SPEC_SELECT8_1_6]]
+; CHECK-NEXT:    [[TMP95:%.*]] = select i1 [[TMP94]], i32 [[NEG_2_6]], i32 [[SUB_2_6]]
+; CHECK-NEXT:    [[CMP12_2_6:%.*]] = icmp slt i32 [[TMP95]], [[SPEC_SELECT8_1_6]]
+; CHECK-NEXT:    [[TMP96:%.*]] = or i1 [[CMP12_2_6]], [[TMP93]]
+; CHECK-NEXT:    [[SPEC_SELECT8_2_6:%.*]] = select i1 [[CMP12_2_6]], i32 [[TMP95]], i32 [[SPEC_SELECT8_1_6]]
 ; CHECK-NEXT:    [[SUB_3_6:%.*]] = sub i32 [[TMP30]], [[TMP24]]
-; CHECK-NEXT:    [[TMP100:%.*]] = icmp slt i32 [[SUB_3_6]], 0
+; CHECK-NEXT:    [[TMP97:%.*]] = icmp slt i32 [[SUB_3_6]], 0
 ; CHECK-NEXT:    [[NEG_3_6:%.*]] = sub nsw i32 0, [[SUB_3_6]]
-; CHECK-NEXT:    [[TMP101:%.*]] = select i1 [[TMP100]], i32 [[NEG_3_6]], i32 [[SUB_3_6]]
-; CHECK-NEXT:    [[CMP12_3_6:%.*]] = icmp slt i32 [[TMP101]], [[SPEC_SELECT8_2_6]]
-; CHECK-NEXT:    [[TMP102:%.*]] = or i1 [[CMP12_3_6]], [[TMP99]]
-; CHECK-NEXT:    [[SPEC_SELECT_3_6:%.*]] = select i1 [[TMP102]], i32 6, i32 [[SPEC_SELECT_3_5]]
-; CHECK-NEXT:    [[SPEC_SELECT8_3_6:%.*]] = select i1 [[CMP12_3_6]], i32 [[TMP101]], i32 [[SPEC_SELECT8_2_6]]
+; CHECK-NEXT:    [[TMP98:%.*]] = select i1 [[TMP97]], i32 [[NEG_3_6]], i32 [[SUB_3_6]]
+; CHECK-NEXT:    [[CMP12_3_6:%.*]] = icmp slt i32 [[TMP98]], [[SPEC_SELECT8_2_6]]
+; CHECK-NEXT:    [[TMP99:%.*]] = or i1 [[CMP12_3_6]], [[TMP96]]
+; CHECK-NEXT:    [[SPEC_SELECT_3_6:%.*]] = select i1 [[TMP99]], i32 6, i32 [[SPEC_SELECT_3_5]]
+; CHECK-NEXT:    [[SPEC_SELECT8_3_6:%.*]] = select i1 [[CMP12_3_6]], i32 [[TMP98]], i32 [[SPEC_SELECT8_2_6]]
 ; CHECK-NEXT:    [[SUB_7:%.*]] = sub i32 [[TMP30]], [[TMP25]]
-; CHECK-NEXT:    [[TMP103:%.*]] = icmp slt i32 [[SUB_7]], 0
+; CHECK-NEXT:    [[TMP100:%.*]] = icmp slt i32 [[SUB_7]], 0
 ; CHECK-NEXT:    [[NEG_7:%.*]] = sub nsw i32 0, [[SUB_7]]
-; CHECK-NEXT:    [[TMP104:%.*]] = select i1 [[TMP103]], i32 [[NEG_7]], i32 [[SUB_7]]
-; CHECK-NEXT:    [[CMP12_7:%.*]] = icmp slt i32 [[TMP104]], [[SPEC_SELECT8_3_6]]
-; CHECK-NEXT:    [[SPEC_SELECT8_7:%.*]] = select i1 [[CMP12_7]], i32 [[TMP104]], i32 [[SPEC_SELECT8_3_6]]
+; CHECK-NEXT:    [[TMP101:%.*]] = select i1 [[TMP100]], i32 [[NEG_7]], i32 [[SUB_7]]
+; CHECK-NEXT:    [[CMP12_7:%.*]] = icmp slt i32 [[TMP101]], [[SPEC_SELECT8_3_6]]
+; CHECK-NEXT:    [[SPEC_SELECT8_7:%.*]] = select i1 [[CMP12_7]], i32 [[TMP101]], i32 [[SPEC_SELECT8_3_6]]
 ; CHECK-NEXT:    [[SUB_1_7:%.*]] = sub i32 [[TMP30]], [[TMP26]]
-; CHECK-NEXT:    [[TMP105:%.*]] = icmp slt i32 [[SUB_1_7]], 0
+; CHECK-NEXT:    [[TMP102:%.*]] = icmp slt i32 [[SUB_1_7]], 0
 ; CHECK-NEXT:    [[NEG_1_7:%.*]] = sub nsw i32 0, [[SUB_1_7]]
-; CHECK-NEXT:    [[TMP106:%.*]] = select i1 [[TMP105]], i32 [[NEG_1_7]], i32 [[SUB_1_7]]
-; CHECK-NEXT:    [[CMP12_1_7:%.*]] = icmp slt i32 [[TMP106]], [[SPEC_SELECT8_7]]
-; CHECK-NEXT:    [[TMP107:%.*]] = or i1 [[CMP12_1_7]], [[CMP12_7]]
-; CHECK-NEXT:    [[SPEC_SELECT8_1_7:%.*]] = select i1 [[CMP12_1_7]], i32 [[TMP106]], i32 [[SPEC_SELECT8_7]]
+; CHECK-NEXT:    [[TMP103:%.*]] = select i1 [[TMP102]], i32 [[NEG_1_7]], i32 [[SUB_1_7]]
+; CHECK-NEXT:    [[CMP12_1_7:%.*]] = icmp slt i32 [[TMP103]], [[SPEC_SELECT8_7]]
+; CHECK-NEXT:    [[TMP104:%.*]] = or i1 [[CMP12_1_7]], [[CMP12_7]]
+; CHECK-NEXT:    [[SPEC_SELECT8_1_7:%.*]] = select i1 [[CMP12_1_7]], i32 [[TMP103]], i32 [[SPEC_SELECT8_7]]
 ; CHECK-NEXT:    [[SUB_2_7:%.*]] = sub i32 [[TMP30]], [[TMP27]]
-; CHECK-NEXT:    [[TMP108:%.*]] = icmp slt i32 [[SUB_2_7]], 0
+; CHECK-NEXT:    [[TMP105:%.*]] = icmp slt i32 [[SUB_2_7]], 0
 ; CHECK-NEXT:    [[NEG_2_7:%.*]] = sub nsw i32 0, [[SUB_2_7]]
-; CHECK-NEXT:    [[TMP109:%.*]] = select i1 [[TMP108]], i32 [[NEG_2_7]], i32 [[SUB_2_7]]
-; CHECK-NEXT:    [[CMP12_2_7:%.*]] = icmp slt i32 [[TMP109]], [[SPEC_SELECT8_1_7]]
-; CHECK-NEXT:    [[TMP110:%.*]] = or i1 [[CMP12_2_7]], [[TMP107]]
-; CHECK-NEXT:    [[SPEC_SELECT8_2_7:%.*]] = select i1 [[CMP12_2_7]], i32 [[TMP109]], i32 [[SPEC_SELECT8_1_7]]
+; CHECK-NEXT:    [[TMP106:%.*]] = select i1 [[TMP105]], i32 [[NEG_2_7]], i32 [[SUB_2_7]]
+; CHECK-NEXT:    [[CMP12_2_7:%.*]] = icmp slt i32 [[TMP106]], [[SPEC_SELECT8_1_7]]
+; CHECK-NEXT:    [[TMP107:%.*]] = or i1 [[CMP12_2_7]], [[TMP104]]
+; CHECK-NEXT:    [[SPEC_SELECT8_2_7:%.*]] = select i1 [[CMP12_2_7]], i32 [[TMP106]], i32 [[SPEC_SELECT8_1_7]]
 ; CHECK-NEXT:    [[SUB_3_7:%.*]] = sub i32 [[TMP30]], [[TMP28]]
-; CHECK-NEXT:    [[TMP111:%.*]] = icmp slt i32 [[SUB_3_7]], 0
+; CHECK-NEXT:    [[TMP108:%.*]] = icmp slt i32 [[SUB_3_7]], 0
 ; CHECK-NEXT:    [[NEG_3_7:%.*]] = sub nsw i32 0, [[SUB_3_7]]
-; CHECK-NEXT:    [[TMP112:%.*]] = select i1 [[TMP111]], i32 [[NEG_3_7]], i32 [[SUB_3_7]]
-; CHECK-NEXT:    [[CMP12_3_7:%.*]] = icmp slt i32 [[TMP112]], [[SPEC_SELECT8_2_7]]
-; CHECK-NEXT:    [[TMP113:%.*]] = or i1 [[CMP12_3_7]], [[TMP110]]
-; CHECK-NEXT:    [[SPEC_SELECT_3_7:%.*]] = select i1 [[TMP113]], i32 7, i32 [[SPEC_SELECT_3_6]]
-; CHECK-NEXT:    [[SPEC_SELECT8_3_7]] = select i1 [[CMP12_3_7]], i32 [[TMP112]], i32 [[SPEC_SELECT8_2_7]]
+; CHECK-NEXT:    [[TMP109:%.*]] = select i1 [[TMP108]], i32 [[NEG_3_7]], i32 [[SUB_3_7]]
+; CHECK-NEXT:    [[CMP12_3_7:%.*]] = icmp slt i32 [[TMP109]], [[SPEC_SELECT8_2_7]]
+; CHECK-NEXT:    [[TMP110:%.*]] = or i1 [[CMP12_3_7]], [[TMP107]]
+; CHECK-NEXT:    [[SPEC_SELECT_3_7:%.*]] = select i1 [[TMP110]], i32 7, i32 [[SPEC_SELECT_3_6]]
+; CHECK-NEXT:    [[SPEC_SELECT8_3_7]] = select i1 [[CMP12_3_7]], i32 [[TMP109]], i32 [[SPEC_SELECT8_2_7]]
 ; CHECK-NEXT:    [[K:%.*]] = getelementptr inbounds [366 x i32], [366 x i32]* @l, i64 0, i64 [[INDVARS_IV]]
 ; CHECK-NEXT:    store i32 [[SPEC_SELECT_3_7]], i32* [[K]], align 4
 ; CHECK-NEXT:    [[INDVARS_IV_NEXT]] = add i64 [[INDVARS_IV]], 1
 ; CHECK-NEXT:    br label [[FOR_COND]]
 ;
 entry:
   %0 = load i32, i32* getelementptr inbounds ([8 x [4 x i32]], [8 x [4 x i32]]* @k, i64 0, i64 0, i64 0), align 16
   %1 = load i32, i32* getelementptr inbounds ([8 x [4 x i32]], [8 x [4 x i32]]* @k, i64 0, i64 0, i64 1), align 4

llvm/test/Transforms/SLPVectorizer/slp-max-phi-size.ll

 ; MAX256-NEXT:  bb:
 ; MAX256-NEXT:    br label [[BB1:%.*]]
 ; MAX256:       bb1:
 ; MAX256-NEXT:    [[I:%.*]] = fpext half [[HVAL:%.*]] to float
 ; MAX256-NEXT:    [[I3:%.*]] = fpext half [[HVAL]] to float
 ; MAX256-NEXT:    [[I6:%.*]] = fpext half [[HVAL]] to float
 ; MAX256-NEXT:    [[I9:%.*]] = fpext half [[HVAL]] to float
 ; MAX256-NEXT:    [[TMP0:%.*]] = insertelement <8 x float> poison, float [[I]], i32 0
-; MAX256-NEXT:    [[TMP1:%.*]] = insertelement <8 x float> [[TMP0]], float [[I]], i32 1
-; MAX256-NEXT:    [[TMP2:%.*]] = insertelement <8 x float> [[TMP1]], float [[I]], i32 2
-; MAX256-NEXT:    [[TMP3:%.*]] = insertelement <8 x float> [[TMP2]], float [[I]], i32 3
-; MAX256-NEXT:    [[TMP4:%.*]] = insertelement <8 x float> [[TMP3]], float [[I]], i32 4
-; MAX256-NEXT:    [[TMP5:%.*]] = insertelement <8 x float> [[TMP4]], float [[I]], i32 5
-; MAX256-NEXT:    [[TMP6:%.*]] = insertelement <8 x float> [[TMP5]], float [[I]], i32 6
-; MAX256-NEXT:    [[TMP7:%.*]] = insertelement <8 x float> [[TMP6]], float [[I]], i32 7
-; MAX256-NEXT:    [[TMP8:%.*]] = insertelement <8 x float> poison, float [[FVAL:%.*]], i32 0
-; MAX256-NEXT:    [[TMP9:%.*]] = insertelement <8 x float> [[TMP8]], float [[FVAL]], i32 1
-; MAX256-NEXT:    [[TMP10:%.*]] = insertelement <8 x float> [[TMP9]], float [[FVAL]], i32 2
-; MAX256-NEXT:    [[TMP11:%.*]] = insertelement <8 x float> [[TMP10]], float [[FVAL]], i32 3
-; MAX256-NEXT:    [[TMP12:%.*]] = insertelement <8 x float> [[TMP11]], float [[FVAL]], i32 4
-; MAX256-NEXT:    [[TMP13:%.*]] = insertelement <8 x float> [[TMP12]], float [[FVAL]], i32 5
-; MAX256-NEXT:    [[TMP14:%.*]] = insertelement <8 x float> [[TMP13]], float [[FVAL]], i32 6
-; MAX256-NEXT:    [[TMP15:%.*]] = insertelement <8 x float> [[TMP14]], float [[FVAL]], i32 7
-; MAX256-NEXT:    [[TMP16:%.*]] = fmul <8 x float> [[TMP7]], [[TMP15]]
-; MAX256-NEXT:    [[TMP17:%.*]] = fadd <8 x float> zeroinitializer, [[TMP16]]
-; MAX256-NEXT:    [[TMP18:%.*]] = insertelement <8 x float> poison, float [[I3]], i32 0
-; MAX256-NEXT:    [[TMP19:%.*]] = insertelement <8 x float> [[TMP18]], float [[I3]], i32 1
-; MAX256-NEXT:    [[TMP20:%.*]] = insertelement <8 x float> [[TMP19]], float [[I3]], i32 2
-; MAX256-NEXT:    [[TMP21:%.*]] = insertelement <8 x float> [[TMP20]], float [[I3]], i32 3
-; MAX256-NEXT:    [[TMP22:%.*]] = insertelement <8 x float> [[TMP21]], float [[I3]], i32 4
-; MAX256-NEXT:    [[TMP23:%.*]] = insertelement <8 x float> [[TMP22]], float [[I3]], i32 5
-; MAX256-NEXT:    [[TMP24:%.*]] = insertelement <8 x float> [[TMP23]], float [[I3]], i32 6
-; MAX256-NEXT:    [[TMP25:%.*]] = insertelement <8 x float> [[TMP24]], float [[I3]], i32 7
-; MAX256-NEXT:    [[TMP26:%.*]] = fmul <8 x float> [[TMP25]], [[TMP15]]
-; MAX256-NEXT:    [[TMP27:%.*]] = fadd <8 x float> zeroinitializer, [[TMP26]]
-; MAX256-NEXT:    [[TMP28:%.*]] = insertelement <8 x float> poison, float [[I6]], i32 0
-; MAX256-NEXT:    [[TMP29:%.*]] = insertelement <8 x float> [[TMP28]], float [[I6]], i32 1
-; MAX256-NEXT:    [[TMP30:%.*]] = insertelement <8 x float> [[TMP29]], float [[I6]], i32 2
-; MAX256-NEXT:    [[TMP31:%.*]] = insertelement <8 x float> [[TMP30]], float [[I6]], i32 3
-; MAX256-NEXT:    [[TMP32:%.*]] = insertelement <8 x float> [[TMP31]], float [[I6]], i32 4
-; MAX256-NEXT:    [[TMP33:%.*]] = insertelement <8 x float> [[TMP32]], float [[I6]], i32 5
-; MAX256-NEXT:    [[TMP34:%.*]] = insertelement <8 x float> [[TMP33]], float [[I6]], i32 6
-; MAX256-NEXT:    [[TMP35:%.*]] = insertelement <8 x float> [[TMP34]], float [[I6]], i32 7
-; MAX256-NEXT:    [[TMP36:%.*]] = fmul <8 x float> [[TMP35]], [[TMP15]]
-; MAX256-NEXT:    [[TMP37:%.*]] = fadd <8 x float> zeroinitializer, [[TMP36]]
-; MAX256-NEXT:    [[TMP38:%.*]] = insertelement <8 x float> poison, float [[I9]], i32 0
-; MAX256-NEXT:    [[TMP39:%.*]] = insertelement <8 x float> [[TMP38]], float [[I9]], i32 1
-; MAX256-NEXT:    [[TMP40:%.*]] = insertelement <8 x float> [[TMP39]], float [[I9]], i32 2
-; MAX256-NEXT:    [[TMP41:%.*]] = insertelement <8 x float> [[TMP40]], float [[I9]], i32 3
-; MAX256-NEXT:    [[TMP42:%.*]] = insertelement <8 x float> [[TMP41]], float [[I9]], i32 4
-; MAX256-NEXT:    [[TMP43:%.*]] = insertelement <8 x float> [[TMP42]], float [[I9]], i32 5
-; MAX256-NEXT:    [[TMP44:%.*]] = insertelement <8 x float> [[TMP43]], float [[I9]], i32 6
-; MAX256-NEXT:    [[TMP45:%.*]] = insertelement <8 x float> [[TMP44]], float [[I9]], i32 7
-; MAX256-NEXT:    [[TMP46:%.*]] = fmul <8 x float> [[TMP45]], [[TMP15]]
-; MAX256-NEXT:    [[TMP47:%.*]] = fadd <8 x float> zeroinitializer, [[TMP46]]
+; MAX256-NEXT:    [[SHUFFLE11:%.*]] = shufflevector <8 x float> [[TMP0]], <8 x float> poison, <8 x i32> zeroinitializer
+; MAX256-NEXT:    [[TMP1:%.*]] = insertelement <8 x float> poison, float [[FVAL:%.*]], i32 0
+; MAX256-NEXT:    [[SHUFFLE12:%.*]] = shufflevector <8 x float> [[TMP1]], <8 x float> poison, <8 x i32> zeroinitializer
+; MAX256-NEXT:    [[TMP2:%.*]] = fmul <8 x float> [[SHUFFLE11]], [[SHUFFLE12]]
+; MAX256-NEXT:    [[TMP3:%.*]] = fadd <8 x float> zeroinitializer, [[TMP2]]
+; MAX256-NEXT:    [[TMP4:%.*]] = insertelement <8 x float> poison, float [[I3]], i32 0
+; MAX256-NEXT:    [[SHUFFLE:%.*]] = shufflevector <8 x float> [[TMP4]], <8 x float> poison, <8 x i32> zeroinitializer
+; MAX256-NEXT:    [[TMP5:%.*]] = fmul <8 x float> [[SHUFFLE]], [[SHUFFLE12]]
+; MAX256-NEXT:    [[TMP6:%.*]] = fadd <8 x float> zeroinitializer, [[TMP5]]
+; MAX256-NEXT:    [[TMP7:%.*]] = insertelement <8 x float> poison, float [[I6]], i32 0
+; MAX256-NEXT:    [[SHUFFLE5:%.*]] = shufflevector <8 x float> [[TMP7]], <8 x float> poison, <8 x i32> zeroinitializer
+; MAX256-NEXT:    [[TMP8:%.*]] = fmul <8 x float> [[SHUFFLE5]], [[SHUFFLE12]]
+; MAX256-NEXT:    [[TMP9:%.*]] = fadd <8 x float> zeroinitializer, [[TMP8]]
+; MAX256-NEXT:    [[TMP10:%.*]] = insertelement <8 x float> poison, float [[I9]], i32 0
+; MAX256-NEXT:    [[SHUFFLE8:%.*]] = shufflevector <8 x float> [[TMP10]], <8 x float> poison, <8 x i32> zeroinitializer
+; MAX256-NEXT:    [[TMP11:%.*]] = fmul <8 x float> [[SHUFFLE8]], [[SHUFFLE12]]
+; MAX256-NEXT:    [[TMP12:%.*]] = fadd <8 x float> zeroinitializer, [[TMP11]]
 ; MAX256-NEXT:    switch i32 undef, label [[BB5:%.*]] [
 ; MAX256-NEXT:    i32 0, label [[BB2:%.*]]
 ; MAX256-NEXT:    i32 1, label [[BB3:%.*]]
 ; MAX256-NEXT:    i32 2, label [[BB4:%.*]]
 ; MAX256-NEXT:    ]
 ; MAX256:       bb3:
 ; MAX256-NEXT:    br label [[BB2]]
 ; MAX256:       bb4:
 ; MAX256-NEXT:    br label [[BB2]]
 ; MAX256:       bb5:
 ; MAX256-NEXT:    br label [[BB2]]
 ; MAX256:       bb2:
-; MAX256-NEXT:    [[TMP48:%.*]] = phi <8 x float> [ [[TMP27]], [[BB3]] ], [ [[TMP15]], [[BB4]] ], [ [[TMP15]], [[BB5]] ], [ [[TMP15]], [[BB1]] ]
-; MAX256-NEXT:    [[TMP49:%.*]] = phi <8 x float> [ [[TMP37]], [[BB3]] ], [ [[TMP15]], [[BB4]] ], [ [[TMP37]], [[BB5]] ], [ [[TMP37]], [[BB1]] ]
-; MAX256-NEXT:    [[TMP50:%.*]] = phi <8 x float> [ [[TMP47]], [[BB3]] ], [ [[TMP47]], [[BB4]] ], [ [[TMP15]], [[BB5]] ], [ [[TMP47]], [[BB1]] ]
-; MAX256-NEXT:    [[TMP51:%.*]] = phi <8 x float> [ [[TMP17]], [[BB3]] ], [ [[TMP17]], [[BB4]] ], [ [[TMP17]], [[BB5]] ], [ [[TMP15]], [[BB1]] ]
-; MAX256-NEXT:    [[TMP52:%.*]] = extractelement <8 x float> [[TMP49]], i32 7
-; MAX256-NEXT:    store float [[TMP52]], float* undef, align 4
+; MAX256-NEXT:    [[TMP13:%.*]] = phi <8 x float> [ [[TMP6]], [[BB3]] ], [ [[SHUFFLE12]], [[BB4]] ], [ [[SHUFFLE12]], [[BB5]] ], [ [[SHUFFLE12]], [[BB1]] ]
+; MAX256-NEXT:    [[TMP14:%.*]] = phi <8 x float> [ [[TMP9]], [[BB3]] ], [ [[SHUFFLE12]], [[BB4]] ], [ [[TMP9]], [[BB5]] ], [ [[TMP9]], [[BB1]] ]
+; MAX256-NEXT:    [[TMP15:%.*]] = phi <8 x float> [ [[TMP12]], [[BB3]] ], [ [[TMP12]], [[BB4]] ], [ [[SHUFFLE12]], [[BB5]] ], [ [[TMP12]], [[BB1]] ]
+; MAX256-NEXT:    [[TMP16:%.*]] = phi <8 x float> [ [[TMP3]], [[BB3]] ], [ [[TMP3]], [[BB4]] ], [ [[TMP3]], [[BB5]] ], [ [[SHUFFLE12]], [[BB1]] ]
+; MAX256-NEXT:    [[TMP17:%.*]] = extractelement <8 x float> [[TMP14]], i32 7
+; MAX256-NEXT:    store float [[TMP17]], float* undef, align 4
 ; MAX256-NEXT:    ret void
 ;
 ; MAX1024-LABEL: @phi_float32(
 ; MAX1024-NEXT:  bb:
 ; MAX1024-NEXT:    br label [[BB1:%.*]]
 ; MAX1024:       bb1:
 ; MAX1024-NEXT:    [[I:%.*]] = fpext half [[HVAL:%.*]] to float
 ; MAX1024-NEXT:    [[I3:%.*]] = fpext half [[HVAL]] to float
 ; MAX1024-NEXT:    [[I6:%.*]] = fpext half [[HVAL]] to float
 ; MAX1024-NEXT:    [[I9:%.*]] = fpext half [[HVAL]] to float
 ; MAX1024-NEXT:    [[TMP0:%.*]] = insertelement <8 x float> poison, float [[I]], i32 0
-; MAX1024-NEXT:    [[TMP1:%.*]] = insertelement <8 x float> [[TMP0]], float [[I]], i32 1
-; MAX1024-NEXT:    [[TMP2:%.*]] = insertelement <8 x float> [[TMP1]], float [[I]], i32 2
-; MAX1024-NEXT:    [[TMP3:%.*]] = insertelement <8 x float> [[TMP2]], float [[I]], i32 3
-; MAX1024-NEXT:    [[TMP4:%.*]] = insertelement <8 x float> [[TMP3]], float [[I]], i32 4
-; MAX1024-NEXT:    [[TMP5:%.*]] = insertelement <8 x float> [[TMP4]], float [[I]], i32 5
-; MAX1024-NEXT:    [[TMP6:%.*]] = insertelement <8 x float> [[TMP5]], float [[I]], i32 6
-; MAX1024-NEXT:    [[TMP7:%.*]] = insertelement <8 x float> [[TMP6]], float [[I]], i32 7
-; MAX1024-NEXT:    [[TMP8:%.*]] = insertelement <8 x float> poison, float [[FVAL:%.*]], i32 0
-; MAX1024-NEXT:    [[TMP9:%.*]] = insertelement <8 x float> [[TMP8]], float [[FVAL]], i32 1
-; MAX1024-NEXT:    [[TMP10:%.*]] = insertelement <8 x float> [[TMP9]], float [[FVAL]], i32 2
-; MAX1024-NEXT:    [[TMP11:%.*]] = insertelement <8 x float> [[TMP10]], float [[FVAL]], i32 3
-; MAX1024-NEXT:    [[TMP12:%.*]] = insertelement <8 x float> [[TMP11]], float [[FVAL]], i32 4
-; MAX1024-NEXT:    [[TMP13:%.*]] = insertelement <8 x float> [[TMP12]], float [[FVAL]], i32 5
-; MAX1024-NEXT:    [[TMP14:%.*]] = insertelement <8 x float> [[TMP13]], float [[FVAL]], i32 6
-; MAX1024-NEXT:    [[TMP15:%.*]] = insertelement <8 x float> [[TMP14]], float [[FVAL]], i32 7
-; MAX1024-NEXT:    [[TMP16:%.*]] = fmul <8 x float> [[TMP7]], [[TMP15]]
-; MAX1024-NEXT:    [[TMP17:%.*]] = fadd <8 x float> zeroinitializer, [[TMP16]]
-; MAX1024-NEXT:    [[TMP18:%.*]] = insertelement <8 x float> poison, float [[I3]], i32 0
-; MAX1024-NEXT:    [[TMP19:%.*]] = insertelement <8 x float> [[TMP18]], float [[I3]], i32 1
-; MAX1024-NEXT:    [[TMP20:%.*]] = insertelement <8 x float> [[TMP19]], float [[I3]], i32 2
-; MAX1024-NEXT:    [[TMP21:%.*]] = insertelement <8 x float> [[TMP20]], float [[I3]], i32 3
-; MAX1024-NEXT:    [[TMP22:%.*]] = insertelement <8 x float> [[TMP21]], float [[I3]], i32 4
-; MAX1024-NEXT:    [[TMP23:%.*]] = insertelement <8 x float> [[TMP22]], float [[I3]], i32 5
-; MAX1024-NEXT:    [[TMP24:%.*]] = insertelement <8 x float> [[TMP23]], float [[I3]], i32 6
-; MAX1024-NEXT:    [[TMP25:%.*]] = insertelement <8 x float> [[TMP24]], float [[I3]], i32 7
-; MAX1024-NEXT:    [[TMP26:%.*]] = fmul <8 x float> [[TMP25]], [[TMP15]]
-; MAX1024-NEXT:    [[TMP27:%.*]] = fadd <8 x float> zeroinitializer, [[TMP26]]
-; MAX1024-NEXT:    [[TMP28:%.*]] = insertelement <8 x float> poison, float [[I6]], i32 0
-; MAX1024-NEXT:    [[TMP29:%.*]] = insertelement <8 x float> [[TMP28]], float [[I6]], i32 1
-; MAX1024-NEXT:    [[TMP30:%.*]] = insertelement <8 x float> [[TMP29]], float [[I6]], i32 2
-; MAX1024-NEXT:    [[TMP31:%.*]] = insertelement <8 x float> [[TMP30]], float [[I6]], i32 3
-; MAX1024-NEXT:    [[TMP32:%.*]] = insertelement <8 x float> [[TMP31]], float [[I6]], i32 4
-; MAX1024-NEXT:    [[TMP33:%.*]] = insertelement <8 x float> [[TMP32]], float [[I6]], i32 5
-; MAX1024-NEXT:    [[TMP34:%.*]] = insertelement <8 x float> [[TMP33]], float [[I6]], i32 6
-; MAX1024-NEXT:    [[TMP35:%.*]] = insertelement <8 x float> [[TMP34]], float [[I6]], i32 7
-; MAX1024-NEXT:    [[TMP36:%.*]] = fmul <8 x float> [[TMP35]], [[TMP15]]
-; MAX1024-NEXT:    [[TMP37:%.*]] = fadd <8 x float> zeroinitializer, [[TMP36]]
-; MAX1024-NEXT:    [[TMP38:%.*]] = insertelement <8 x float> poison, float [[I9]], i32 0
-; MAX1024-NEXT:    [[TMP39:%.*]] = insertelement <8 x float> [[TMP38]], float [[I9]], i32 1
-; MAX1024-NEXT:    [[TMP40:%.*]] = insertelement <8 x float> [[TMP39]], float [[I9]], i32 2
-; MAX1024-NEXT:    [[TMP41:%.*]] = insertelement <8 x float> [[TMP40]], float [[I9]], i32 3
-; MAX1024-NEXT:    [[TMP42:%.*]] = insertelement <8 x float> [[TMP41]], float [[I9]], i32 4
-; MAX1024-NEXT:    [[TMP43:%.*]] = insertelement <8 x float> [[TMP42]], float [[I9]], i32 5
-; MAX1024-NEXT:    [[TMP44:%.*]] = insertelement <8 x float> [[TMP43]], float [[I9]], i32 6
-; MAX1024-NEXT:    [[TMP45:%.*]] = insertelement <8 x float> [[TMP44]], float [[I9]], i32 7
-; MAX1024-NEXT:    [[TMP46:%.*]] = fmul <8 x float> [[TMP45]], [[TMP15]]
-; MAX1024-NEXT:    [[TMP47:%.*]] = fadd <8 x float> zeroinitializer, [[TMP46]]
+; MAX1024-NEXT:    [[SHUFFLE11:%.*]] = shufflevector <8 x float> [[TMP0]], <8 x float> poison, <8 x i32> zeroinitializer
+; MAX1024-NEXT:    [[TMP1:%.*]] = insertelement <8 x float> poison, float [[FVAL:%.*]], i32 0
+; MAX1024-NEXT:    [[SHUFFLE12:%.*]] = shufflevector <8 x float> [[TMP1]], <8 x float> poison, <8 x i32> zeroinitializer
+; MAX1024-NEXT:    [[TMP2:%.*]] = fmul <8 x float> [[SHUFFLE11]], [[SHUFFLE12]]
+; MAX1024-NEXT:    [[TMP3:%.*]] = fadd <8 x float> zeroinitializer, [[TMP2]]
+; MAX1024-NEXT:    [[TMP4:%.*]] = insertelement <8 x float> poison, float [[I3]], i32 0
+; MAX1024-NEXT:    [[SHUFFLE:%.*]] = shufflevector <8 x float> [[TMP4]], <8 x float> poison, <8 x i32> zeroinitializer
+; MAX1024-NEXT:    [[TMP5:%.*]] = fmul <8 x float> [[SHUFFLE]], [[SHUFFLE12]]
+; MAX1024-NEXT:    [[TMP6:%.*]] = fadd <8 x float> zeroinitializer, [[TMP5]]
+; MAX1024-NEXT:    [[TMP7:%.*]] = insertelement <8 x float> poison, float [[I6]], i32 0
+; MAX1024-NEXT:    [[SHUFFLE5:%.*]] = shufflevector <8 x float> [[TMP7]], <8 x float> poison, <8 x i32> zeroinitializer
+; MAX1024-NEXT:    [[TMP8:%.*]] = fmul <8 x float> [[SHUFFLE5]], [[SHUFFLE12]]
+; MAX1024-NEXT:    [[TMP9:%.*]] = fadd <8 x float> zeroinitializer, [[TMP8]]
+; MAX1024-NEXT:    [[TMP10:%.*]] = insertelement <8 x float> poison, float [[I9]], i32 0
+; MAX1024-NEXT:    [[SHUFFLE8:%.*]] = shufflevector <8 x float> [[TMP10]], <8 x float> poison, <8 x i32> zeroinitializer
+; MAX1024-NEXT:    [[TMP11:%.*]] = fmul <8 x float> [[SHUFFLE8]], [[SHUFFLE12]]
+; MAX1024-NEXT:    [[TMP12:%.*]] = fadd <8 x float> zeroinitializer, [[TMP11]]
 ; MAX1024-NEXT:    switch i32 undef, label [[BB5:%.*]] [
 ; MAX1024-NEXT:    i32 0, label [[BB2:%.*]]
 ; MAX1024-NEXT:    i32 1, label [[BB3:%.*]]
 ; MAX1024-NEXT:    i32 2, label [[BB4:%.*]]
 ; MAX1024-NEXT:    ]
 ; MAX1024:       bb3:
 ; MAX1024-NEXT:    br label [[BB2]]
 ; MAX1024:       bb4:
 ; MAX1024-NEXT:    br label [[BB2]]
 ; MAX1024:       bb5:
 ; MAX1024-NEXT:    br label [[BB2]]
 ; MAX1024:       bb2:
-; MAX1024-NEXT:    [[TMP48:%.*]] = phi <8 x float> [ [[TMP27]], [[BB3]] ], [ [[TMP15]], [[BB4]] ], [ [[TMP15]], [[BB5]] ], [ [[TMP15]], [[BB1]] ]
-; MAX1024-NEXT:    [[TMP49:%.*]] = phi <8 x float> [ [[TMP37]], [[BB3]] ], [ [[TMP15]], [[BB4]] ], [ [[TMP37]], [[BB5]] ], [ [[TMP37]], [[BB1]] ]
-; MAX1024-NEXT:    [[TMP50:%.*]] = phi <8 x float> [ [[TMP47]], [[BB3]] ], [ [[TMP47]], [[BB4]] ], [ [[TMP15]], [[BB5]] ], [ [[TMP47]], [[BB1]] ]
-; MAX1024-NEXT:    [[TMP51:%.*]] = phi <8 x float> [ [[TMP17]], [[BB3]] ], [ [[TMP17]], [[BB4]] ], [ [[TMP17]], [[BB5]] ], [ [[TMP15]], [[BB1]] ]
-; MAX1024-NEXT:    [[TMP52:%.*]] = extractelement <8 x float> [[TMP49]], i32 7
-; MAX1024-NEXT:    store float [[TMP52]], float* undef, align 4
+; MAX1024-NEXT:    [[TMP13:%.*]] = phi <8 x float> [ [[TMP6]], [[BB3]] ], [ [[SHUFFLE12]], [[BB4]] ], [ [[SHUFFLE12]], [[BB5]] ], [ [[SHUFFLE12]], [[BB1]] ]
+; MAX1024-NEXT:    [[TMP14:%.*]] = phi <8 x float> [ [[TMP9]], [[BB3]] ], [ [[SHUFFLE12]], [[BB4]] ], [ [[TMP9]], [[BB5]] ], [ [[TMP9]], [[BB1]] ]
+; MAX1024-NEXT:    [[TMP15:%.*]] = phi <8 x float> [ [[TMP12]], [[BB3]] ], [ [[TMP12]], [[BB4]] ], [ [[SHUFFLE12]], [[BB5]] ], [ [[TMP12]], [[BB1]] ]
+; MAX1024-NEXT:    [[TMP16:%.*]] = phi <8 x float> [ [[TMP3]], [[BB3]] ], [ [[TMP3]], [[BB4]] ], [ [[TMP3]], [[BB5]] ], [ [[SHUFFLE12]], [[BB1]] ]
+; MAX1024-NEXT:    [[TMP17:%.*]] = extractelement <8 x float> [[TMP14]], i32 7
+; MAX1024-NEXT:    store float [[TMP17]], float* undef, align 4
 ; MAX1024-NEXT:    ret void
 ;
 bb:
   br label %bb1
 
 bb1:
   %i = fpext half %hval to float
   %i1 = fmul float %i, %fval