[Matrix] Refine cost estimate for dot-product.

[fhahn]

Summary

Adjust lowerDotProduct cost estimate to include the cost benefits of:

• emitting a wide load
• emitting a wide multiply.

[thegameg]

LGTM.

llvm/lib/Transforms/Scalar/LowerMatrixIntrinsics.cpp

1346

llvm/test/Transforms/LowerMatrixIntrinsics/dot-product-int.ll

260–318

Wonder which part made this go back, the cost of the mul maybe?

[fhahn]

Rebase and fix a crash. I am planning to land this soon.

[fhahn]

llvm/lib/Transforms/Scalar/LowerMatrixIntrinsics.cpp

1346

I adjusted the code slightly to include an estimate of the cost of embedding the columns into a vector, so we can compute valid costs for all cases. I removed that part.

llvm/test/Transforms/LowerMatrixIntrinsics/dot-product-int.ll

260–318

Yep, I think vector integer multiply doesn't support i64 elements on ARM64.

llvm/lib/Transforms/Scalar/LowerMatrixIntrinsics.cpp

void lowerDotProduct(CallInst *MatMul,
 
     Type *ElementType = cast<VectorType>(LHS->getType())->getElementType();
     bool IsIntVec = ElementType->isIntegerTy();
 
     // Floating point reductions require reassocation.
     if (!IsIntVec && !FMF.allowReassoc())
       return;
 
-    auto IsSupportedArg = [](Value *Op, unsigned N) {
-      if (!isa<Instruction>(Op))
-        return true;
-      return match(
-          Op, m_OneUse(m_CombineOr(
+    auto CanBeFlattened = [](Value *Op) {
+      return match(Op, m_OneUse(m_CombineOr(
                            m_Load(m_Value()),
                            m_Intrinsic<Intrinsic::matrix_column_major_load>(
-                      m_Value(), m_SpecificInt(N)))));
+                               m_Value(), m_SpecificInt(1)))));
     };
-    if (!IsSupportedArg(LHS, RShape.NumColumns))
-      return;
+    // Returns the cost benefit of using \p Op with the dot product lowering. If
+    // the returned cost is < 0, the argument is cheaper to use in the
+    // dot-product lowering.
+    auto GetCostForArg = [this, &CanBeFlattened](Value *Op, unsigned N) {
+      if (!isa<Instruction>(Op))
+        return InstructionCost(0);
+
+      FixedVectorType *VecTy = cast<FixedVectorType>(Op->getType());
+      Type *EltTy = VecTy->getElementType();
+
+      if (CanBeFlattened(Op)) {
+        if (N == 1)
+          return InstructionCost(0);
+
+        return TTI.getMemoryOpCost(Instruction::Load, VecTy, Align(1), 0) -
+               N * TTI.getMemoryOpCost(Instruction::Load, EltTy, Align(1), 0);
+      }
+
+      InstructionCost EmbedCost(0);
+      // Roughly estimate the cost for embedding the columns into a vector.
+      for (unsigned I = 1; I < N; ++I)
+        EmbedCost +=
+            TTI.getShuffleCost(TTI::SK_Splice, FixedVectorType::get(EltTy, 1),
+                               std::nullopt, TTI::TCK_RecipThroughput);
+      return EmbedCost;
+    };
+    auto LHSCost = GetCostForArg(LHS, LShape.NumColumns);
 
     // We compare the costs of a vector.reduce.add to sequential add.
     int AddOpCode = IsIntVec ? Instruction::Add : Instruction::FAdd;
-    FastMathFlags FMFReassoc;
-    FMFReassoc.setAllowReassoc();
-    InstructionCost ReductionCost = TTI.getArithmeticReductionCost(
-        AddOpCode, cast<VectorType>(LHS->getType()), FMFReassoc);
+    int MulOpCode = IsIntVec ? Instruction::Mul : Instruction::FMul;
+    InstructionCost ReductionCost =
+        TTI.getArithmeticReductionCost(
+            AddOpCode, cast<VectorType>(LHS->getType()),
+            IsIntVec ? std::nullopt : std::optional(FMF)) +
+        TTI.getArithmeticInstrCost(MulOpCode, LHS->getType());
     InstructionCost SequentialAddCost =
         TTI.getArithmeticInstrCost(AddOpCode, ElementType) *
-        (LShape.NumColumns - 1);
-    if (ReductionCost >= SequentialAddCost)
+            (LShape.NumColumns - 1) +
+        TTI.getArithmeticInstrCost(MulOpCode, ElementType) *
+            (LShape.NumColumns);
+    if ((LHSCost + ReductionCost - SequentialAddCost) > InstructionCost(0))
       return;
 
     FusedInsts.insert(MatMul);
     IRBuilder<> Builder(MatMul);
-    auto FlattenArg = [&Builder, &FusedInsts](Value *Op) -> Value * {
+    auto FlattenArg = [&Builder, &FusedInsts,
+                       &CanBeFlattened](Value *Op) -> Value * {
       // Matmul must be the only user of loads because we don't use LowerLoad
       // for row vectors (LowerLoad results in scalar loads and shufflevectors
       // instead of single vector load).
-      if (!match(Op, m_CombineOr(
-                         m_Load(m_Value()),
-                         m_Intrinsic<Intrinsic::matrix_column_major_load>()))) {
+      if (!CanBeFlattened(Op))
         return Op;
-      }
-      FusedInsts.insert(cast<Instruction>(Op));
 
+      FusedInsts.insert(cast<Instruction>(Op));
       // If vector uses the builtin load, lower to a LoadInst
       Value *Ptr;
       if (match(Op, m_Intrinsic<Intrinsic::matrix_column_major_load>(
                         m_Value(Ptr)))) {
         auto *NewLoad = Builder.CreateLoad(Op->getType(), Ptr);
         Op->replaceAllUsesWith(NewLoad);
         cast<Instruction>(Op)->eraseFromParent();
         return NewLoad;

llvm/test/Transforms/LowerMatrixIntrinsics/dot-product-float.ll

entry:
   ret <1 x float> %c
 }
 
 declare <1 x float> @llvm.matrix.multiply.v1f32.v6f32.v6f32(<6 x float>, <6 x float>, i32, i32, i32)
 
 define <1 x float> @dotproduct_float_v1(<1 x float> %a, <1 x float> %b) {
 ; CHECK-LABEL: @dotproduct_float_v1(
 ; CHECK-NEXT:  entry:
-; CHECK-NEXT:    [[SPLIT:%.*]] = shufflevector <1 x float> [[A:%.*]], <1 x float> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[SPLIT1:%.*]] = shufflevector <1 x float> [[B:%.*]], <1 x float> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[BLOCK:%.*]] = shufflevector <1 x float> [[SPLIT]], <1 x float> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP0:%.*]] = extractelement <1 x float> [[SPLIT1]], i64 0
-; CHECK-NEXT:    [[SPLAT_SPLATINSERT:%.*]] = insertelement <1 x float> poison, float [[TMP0]], i64 0
-; CHECK-NEXT:    [[SPLAT_SPLAT:%.*]] = shufflevector <1 x float> [[SPLAT_SPLATINSERT]], <1 x float> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP1:%.*]] = fmul fast <1 x float> [[BLOCK]], [[SPLAT_SPLAT]]
-; CHECK-NEXT:    [[TMP2:%.*]] = shufflevector <1 x float> [[TMP1]], <1 x float> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP3:%.*]] = shufflevector <1 x float> undef, <1 x float> [[TMP2]], <1 x i32> <i32 1>
-; CHECK-NEXT:    ret <1 x float> [[TMP3]]
+; CHECK-NEXT:    [[TMP0:%.*]] = fmul <1 x float> [[A:%.*]], [[B:%.*]]
+; CHECK-NEXT:    [[TMP1:%.*]] = call fast float @llvm.vector.reduce.fadd.v1f32(float 0.000000e+00, <1 x float> [[TMP0]])
+; CHECK-NEXT:    [[TMP2:%.*]] = insertelement <1 x float> poison, float [[TMP1]], i64 0
+; CHECK-NEXT:    ret <1 x float> [[TMP2]]
 ;
 entry:
   %c = tail call fast <1 x float> @llvm.matrix.multiply.v1f32.v1f32.v1f32(<1 x float> %a, <1 x float> %b, i32 1, i32 1, i32 1)
   ret <1 x float> %c
 }
 
 declare <1 x float> @llvm.matrix.multiply.v1f32.v1f32.v1f32(<1 x float>, <1 x float>, i32, i32, i32)
 
 define <1 x float> @dotproduct_float_v3(<3 x float> %a, <3 x float> %b) {
 ; CHECK-LABEL: @dotproduct_float_v3(
 ; CHECK-NEXT:  entry:
-; CHECK-NEXT:    [[SPLIT:%.*]] = shufflevector <3 x float> [[A:%.*]], <3 x float> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[SPLIT1:%.*]] = shufflevector <3 x float> [[A]], <3 x float> poison, <1 x i32> <i32 1>
-; CHECK-NEXT:    [[SPLIT2:%.*]] = shufflevector <3 x float> [[A]], <3 x float> poison, <1 x i32> <i32 2>
-; CHECK-NEXT:    [[SPLIT3:%.*]] = shufflevector <3 x float> [[B:%.*]], <3 x float> poison, <3 x i32> <i32 0, i32 1, i32 2>
-; CHECK-NEXT:    [[BLOCK:%.*]] = shufflevector <1 x float> [[SPLIT]], <1 x float> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP0:%.*]] = extractelement <3 x float> [[SPLIT3]], i64 0
-; CHECK-NEXT:    [[SPLAT_SPLATINSERT:%.*]] = insertelement <1 x float> poison, float [[TMP0]], i64 0
-; CHECK-NEXT:    [[SPLAT_SPLAT:%.*]] = shufflevector <1 x float> [[SPLAT_SPLATINSERT]], <1 x float> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP1:%.*]] = fmul fast <1 x float> [[BLOCK]], [[SPLAT_SPLAT]]
-; CHECK-NEXT:    [[BLOCK4:%.*]] = shufflevector <1 x float> [[SPLIT1]], <1 x float> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP2:%.*]] = extractelement <3 x float> [[SPLIT3]], i64 1
-; CHECK-NEXT:    [[SPLAT_SPLATINSERT5:%.*]] = insertelement <1 x float> poison, float [[TMP2]], i64 0
-; CHECK-NEXT:    [[SPLAT_SPLAT6:%.*]] = shufflevector <1 x float> [[SPLAT_SPLATINSERT5]], <1 x float> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP3:%.*]] = call fast <1 x float> @llvm.fmuladd.v1f32(<1 x float> [[BLOCK4]], <1 x float> [[SPLAT_SPLAT6]], <1 x float> [[TMP1]])
-; CHECK-NEXT:    [[BLOCK7:%.*]] = shufflevector <1 x float> [[SPLIT2]], <1 x float> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP4:%.*]] = extractelement <3 x float> [[SPLIT3]], i64 2
-; CHECK-NEXT:    [[SPLAT_SPLATINSERT8:%.*]] = insertelement <1 x float> poison, float [[TMP4]], i64 0
-; CHECK-NEXT:    [[SPLAT_SPLAT9:%.*]] = shufflevector <1 x float> [[SPLAT_SPLATINSERT8]], <1 x float> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP5:%.*]] = call fast <1 x float> @llvm.fmuladd.v1f32(<1 x float> [[BLOCK7]], <1 x float> [[SPLAT_SPLAT9]], <1 x float> [[TMP3]])
-; CHECK-NEXT:    [[TMP6:%.*]] = shufflevector <1 x float> [[TMP5]], <1 x float> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP7:%.*]] = shufflevector <1 x float> undef, <1 x float> [[TMP6]], <1 x i32> <i32 1>
-; CHECK-NEXT:    ret <1 x float> [[TMP7]]
+; CHECK-NEXT:    [[TMP0:%.*]] = fmul <3 x float> [[A:%.*]], [[B:%.*]]
+; CHECK-NEXT:    [[TMP1:%.*]] = call fast float @llvm.vector.reduce.fadd.v3f32(float 0.000000e+00, <3 x float> [[TMP0]])
+; CHECK-NEXT:    [[TMP2:%.*]] = insertelement <1 x float> poison, float [[TMP1]], i64 0
+; CHECK-NEXT:    ret <1 x float> [[TMP2]]
 ;
 entry:
   %c = tail call fast <1 x float> @llvm.matrix.multiply.v1f32.v3f32.v3f32(<3 x float> %a, <3 x float> %b, i32 1, i32 3, i32 1)
   ret <1 x float> %c
 }
 
 declare <1 x float> @llvm.matrix.multiply.v1f32.v3f32.v3f32(<3 x float>, <3 x float>, i32, i32, i32)
 
 define <1 x float> @intrinsic_column_major_load_dot_product_float_v6(ptr %lhs_address, ptr %rhs_address) {
 ; CHECK-LABEL: @intrinsic_column_major_load_dot_product_float_v6(
 ; CHECK-NEXT:  entry:
-; CHECK-NEXT:    [[COL_LOAD:%.*]] = load <1 x float>, ptr [[LHS_ADDRESS:%.*]], align 4
-; CHECK-NEXT:    [[VEC_GEP:%.*]] = getelementptr float, ptr [[LHS_ADDRESS]], i64 1
-; CHECK-NEXT:    [[COL_LOAD1:%.*]] = load <1 x float>, ptr [[VEC_GEP]], align 4
-; CHECK-NEXT:    [[VEC_GEP2:%.*]] = getelementptr float, ptr [[LHS_ADDRESS]], i64 2
-; CHECK-NEXT:    [[COL_LOAD3:%.*]] = load <1 x float>, ptr [[VEC_GEP2]], align 4
-; CHECK-NEXT:    [[VEC_GEP4:%.*]] = getelementptr float, ptr [[LHS_ADDRESS]], i64 3
-; CHECK-NEXT:    [[COL_LOAD5:%.*]] = load <1 x float>, ptr [[VEC_GEP4]], align 4
-; CHECK-NEXT:    [[VEC_GEP6:%.*]] = getelementptr float, ptr [[LHS_ADDRESS]], i64 4
-; CHECK-NEXT:    [[COL_LOAD7:%.*]] = load <1 x float>, ptr [[VEC_GEP6]], align 4
-; CHECK-NEXT:    [[VEC_GEP8:%.*]] = getelementptr float, ptr [[LHS_ADDRESS]], i64 5
-; CHECK-NEXT:    [[COL_LOAD9:%.*]] = load <1 x float>, ptr [[VEC_GEP8]], align 4
-; CHECK-NEXT:    [[COL_LOAD10:%.*]] = load <6 x float>, ptr [[RHS_ADDRESS:%.*]], align 4
-; CHECK-NEXT:    [[BLOCK:%.*]] = shufflevector <1 x float> [[COL_LOAD]], <1 x float> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP0:%.*]] = extractelement <6 x float> [[COL_LOAD10]], i64 0
-; CHECK-NEXT:    [[SPLAT_SPLATINSERT:%.*]] = insertelement <1 x float> poison, float [[TMP0]], i64 0
-; CHECK-NEXT:    [[SPLAT_SPLAT:%.*]] = shufflevector <1 x float> [[SPLAT_SPLATINSERT]], <1 x float> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP1:%.*]] = fmul fast <1 x float> [[BLOCK]], [[SPLAT_SPLAT]]
-; CHECK-NEXT:    [[BLOCK11:%.*]] = shufflevector <1 x float> [[COL_LOAD1]], <1 x float> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP2:%.*]] = extractelement <6 x float> [[COL_LOAD10]], i64 1
-; CHECK-NEXT:    [[SPLAT_SPLATINSERT12:%.*]] = insertelement <1 x float> poison, float [[TMP2]], i64 0
-; CHECK-NEXT:    [[SPLAT_SPLAT13:%.*]] = shufflevector <1 x float> [[SPLAT_SPLATINSERT12]], <1 x float> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP3:%.*]] = call fast <1 x float> @llvm.fmuladd.v1f32(<1 x float> [[BLOCK11]], <1 x float> [[SPLAT_SPLAT13]], <1 x float> [[TMP1]])
-; CHECK-NEXT:    [[BLOCK14:%.*]] = shufflevector <1 x float> [[COL_LOAD3]], <1 x float> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP4:%.*]] = extractelement <6 x float> [[COL_LOAD10]], i64 2
-; CHECK-NEXT:    [[SPLAT_SPLATINSERT15:%.*]] = insertelement <1 x float> poison, float [[TMP4]], i64 0
-; CHECK-NEXT:    [[SPLAT_SPLAT16:%.*]] = shufflevector <1 x float> [[SPLAT_SPLATINSERT15]], <1 x float> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP5:%.*]] = call fast <1 x float> @llvm.fmuladd.v1f32(<1 x float> [[BLOCK14]], <1 x float> [[SPLAT_SPLAT16]], <1 x float> [[TMP3]])
-; CHECK-NEXT:    [[BLOCK17:%.*]] = shufflevector <1 x float> [[COL_LOAD5]], <1 x float> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP6:%.*]] = extractelement <6 x float> [[COL_LOAD10]], i64 3
-; CHECK-NEXT:    [[SPLAT_SPLATINSERT18:%.*]] = insertelement <1 x float> poison, float [[TMP6]], i64 0
-; CHECK-NEXT:    [[SPLAT_SPLAT19:%.*]] = shufflevector <1 x float> [[SPLAT_SPLATINSERT18]], <1 x float> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP7:%.*]] = call fast <1 x float> @llvm.fmuladd.v1f32(<1 x float> [[BLOCK17]], <1 x float> [[SPLAT_SPLAT19]], <1 x float> [[TMP5]])
-; CHECK-NEXT:    [[BLOCK20:%.*]] = shufflevector <1 x float> [[COL_LOAD7]], <1 x float> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP8:%.*]] = extractelement <6 x float> [[COL_LOAD10]], i64 4
-; CHECK-NEXT:    [[SPLAT_SPLATINSERT21:%.*]] = insertelement <1 x float> poison, float [[TMP8]], i64 0
-; CHECK-NEXT:    [[SPLAT_SPLAT22:%.*]] = shufflevector <1 x float> [[SPLAT_SPLATINSERT21]], <1 x float> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP9:%.*]] = call fast <1 x float> @llvm.fmuladd.v1f32(<1 x float> [[BLOCK20]], <1 x float> [[SPLAT_SPLAT22]], <1 x float> [[TMP7]])
-; CHECK-NEXT:    [[BLOCK23:%.*]] = shufflevector <1 x float> [[COL_LOAD9]], <1 x float> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP10:%.*]] = extractelement <6 x float> [[COL_LOAD10]], i64 5
-; CHECK-NEXT:    [[SPLAT_SPLATINSERT24:%.*]] = insertelement <1 x float> poison, float [[TMP10]], i64 0
-; CHECK-NEXT:    [[SPLAT_SPLAT25:%.*]] = shufflevector <1 x float> [[SPLAT_SPLATINSERT24]], <1 x float> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP11:%.*]] = call fast <1 x float> @llvm.fmuladd.v1f32(<1 x float> [[BLOCK23]], <1 x float> [[SPLAT_SPLAT25]], <1 x float> [[TMP9]])
-; CHECK-NEXT:    [[TMP12:%.*]] = shufflevector <1 x float> [[TMP11]], <1 x float> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP13:%.*]] = shufflevector <1 x float> undef, <1 x float> [[TMP12]], <1 x i32> <i32 1>
-; CHECK-NEXT:    ret <1 x float> [[TMP13]]
+; CHECK-NEXT:    [[COL_LOAD:%.*]] = load <6 x float>, ptr [[RHS_ADDRESS:%.*]], align 4
+; CHECK-NEXT:    [[TMP0:%.*]] = load <6 x float>, ptr [[LHS_ADDRESS:%.*]], align 32
+; CHECK-NEXT:    [[TMP1:%.*]] = fmul <6 x float> [[TMP0]], [[COL_LOAD]]
+; CHECK-NEXT:    [[TMP2:%.*]] = call fast float @llvm.vector.reduce.fadd.v6f32(float 0.000000e+00, <6 x float> [[TMP1]])
+; CHECK-NEXT:    [[TMP3:%.*]] = insertelement <1 x float> poison, float [[TMP2]], i64 0
+; CHECK-NEXT:    ret <1 x float> [[TMP3]]
 ;
 entry:
   %lhs = tail call fast <6 x float> @llvm.matrix.column.major.load.v6f32.i64(ptr nonnull align 4 %lhs_address, i64 1, i1 false, i32 1, i32 6)
   %rhs = tail call fast <6 x float> @llvm.matrix.column.major.load.v6f32.i64(ptr nonnull align 4 %rhs_address, i64 6, i1 false, i32 6, i32 1)
   %result = tail call fast <1 x float> @llvm.matrix.multiply.v1f32.v6f32.v6f32(<6 x float> %lhs, <6 x float> %rhs, i32 1, i32 6, i32 1)
   ret <1 x float> %result
 }
 
 declare <6 x float> @llvm.matrix.column.major.load.v6f32.i64(ptr nonnull align 4, i64, i1, i32, i32)
 
 define <1 x float> @LoadInst_dot_product_float_v7(ptr %lhs_address, ptr %rhs_address) {
 ; CHECK-LABEL: @LoadInst_dot_product_float_v7(
 ; CHECK-NEXT:  entry:
-; CHECK-NEXT:    [[COL_LOAD:%.*]] = load <1 x float>, ptr [[LHS_ADDRESS:%.*]], align 32
-; CHECK-NEXT:    [[VEC_GEP:%.*]] = getelementptr float, ptr [[LHS_ADDRESS]], i64 1
-; CHECK-NEXT:    [[COL_LOAD1:%.*]] = load <1 x float>, ptr [[VEC_GEP]], align 4
-; CHECK-NEXT:    [[VEC_GEP2:%.*]] = getelementptr float, ptr [[LHS_ADDRESS]], i64 2
-; CHECK-NEXT:    [[COL_LOAD3:%.*]] = load <1 x float>, ptr [[VEC_GEP2]], align 8
-; CHECK-NEXT:    [[VEC_GEP4:%.*]] = getelementptr float, ptr [[LHS_ADDRESS]], i64 3
-; CHECK-NEXT:    [[COL_LOAD5:%.*]] = load <1 x float>, ptr [[VEC_GEP4]], align 4
-; CHECK-NEXT:    [[VEC_GEP6:%.*]] = getelementptr float, ptr [[LHS_ADDRESS]], i64 4
-; CHECK-NEXT:    [[COL_LOAD7:%.*]] = load <1 x float>, ptr [[VEC_GEP6]], align 16
-; CHECK-NEXT:    [[VEC_GEP8:%.*]] = getelementptr float, ptr [[LHS_ADDRESS]], i64 5
-; CHECK-NEXT:    [[COL_LOAD9:%.*]] = load <1 x float>, ptr [[VEC_GEP8]], align 4
-; CHECK-NEXT:    [[VEC_GEP10:%.*]] = getelementptr float, ptr [[LHS_ADDRESS]], i64 6
-; CHECK-NEXT:    [[COL_LOAD11:%.*]] = load <1 x float>, ptr [[VEC_GEP10]], align 8
-; CHECK-NEXT:    [[COL_LOAD12:%.*]] = load <7 x float>, ptr [[RHS_ADDRESS:%.*]], align 32
-; CHECK-NEXT:    [[BLOCK:%.*]] = shufflevector <1 x float> [[COL_LOAD]], <1 x float> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP0:%.*]] = extractelement <7 x float> [[COL_LOAD12]], i64 0
-; CHECK-NEXT:    [[SPLAT_SPLATINSERT:%.*]] = insertelement <1 x float> poison, float [[TMP0]], i64 0
-; CHECK-NEXT:    [[SPLAT_SPLAT:%.*]] = shufflevector <1 x float> [[SPLAT_SPLATINSERT]], <1 x float> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP1:%.*]] = fmul fast <1 x float> [[BLOCK]], [[SPLAT_SPLAT]]
-; CHECK-NEXT:    [[BLOCK13:%.*]] = shufflevector <1 x float> [[COL_LOAD1]], <1 x float> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP2:%.*]] = extractelement <7 x float> [[COL_LOAD12]], i64 1
-; CHECK-NEXT:    [[SPLAT_SPLATINSERT14:%.*]] = insertelement <1 x float> poison, float [[TMP2]], i64 0
-; CHECK-NEXT:    [[SPLAT_SPLAT15:%.*]] = shufflevector <1 x float> [[SPLAT_SPLATINSERT14]], <1 x float> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP3:%.*]] = call fast <1 x float> @llvm.fmuladd.v1f32(<1 x float> [[BLOCK13]], <1 x float> [[SPLAT_SPLAT15]], <1 x float> [[TMP1]])
-; CHECK-NEXT:    [[BLOCK16:%.*]] = shufflevector <1 x float> [[COL_LOAD3]], <1 x float> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP4:%.*]] = extractelement <7 x float> [[COL_LOAD12]], i64 2
-; CHECK-NEXT:    [[SPLAT_SPLATINSERT17:%.*]] = insertelement <1 x float> poison, float [[TMP4]], i64 0
-; CHECK-NEXT:    [[SPLAT_SPLAT18:%.*]] = shufflevector <1 x float> [[SPLAT_SPLATINSERT17]], <1 x float> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP5:%.*]] = call fast <1 x float> @llvm.fmuladd.v1f32(<1 x float> [[BLOCK16]], <1 x float> [[SPLAT_SPLAT18]], <1 x float> [[TMP3]])
-; CHECK-NEXT:    [[BLOCK19:%.*]] = shufflevector <1 x float> [[COL_LOAD5]], <1 x float> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP6:%.*]] = extractelement <7 x float> [[COL_LOAD12]], i64 3
-; CHECK-NEXT:    [[SPLAT_SPLATINSERT20:%.*]] = insertelement <1 x float> poison, float [[TMP6]], i64 0
-; CHECK-NEXT:    [[SPLAT_SPLAT21:%.*]] = shufflevector <1 x float> [[SPLAT_SPLATINSERT20]], <1 x float> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP7:%.*]] = call fast <1 x float> @llvm.fmuladd.v1f32(<1 x float> [[BLOCK19]], <1 x float> [[SPLAT_SPLAT21]], <1 x float> [[TMP5]])
-; CHECK-NEXT:    [[BLOCK22:%.*]] = shufflevector <1 x float> [[COL_LOAD7]], <1 x float> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP8:%.*]] = extractelement <7 x float> [[COL_LOAD12]], i64 4
-; CHECK-NEXT:    [[SPLAT_SPLATINSERT23:%.*]] = insertelement <1 x float> poison, float [[TMP8]], i64 0
-; CHECK-NEXT:    [[SPLAT_SPLAT24:%.*]] = shufflevector <1 x float> [[SPLAT_SPLATINSERT23]], <1 x float> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP9:%.*]] = call fast <1 x float> @llvm.fmuladd.v1f32(<1 x float> [[BLOCK22]], <1 x float> [[SPLAT_SPLAT24]], <1 x float> [[TMP7]])
-; CHECK-NEXT:    [[BLOCK25:%.*]] = shufflevector <1 x float> [[COL_LOAD9]], <1 x float> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP10:%.*]] = extractelement <7 x float> [[COL_LOAD12]], i64 5
-; CHECK-NEXT:    [[SPLAT_SPLATINSERT26:%.*]] = insertelement <1 x float> poison, float [[TMP10]], i64 0
-; CHECK-NEXT:    [[SPLAT_SPLAT27:%.*]] = shufflevector <1 x float> [[SPLAT_SPLATINSERT26]], <1 x float> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP11:%.*]] = call fast <1 x float> @llvm.fmuladd.v1f32(<1 x float> [[BLOCK25]], <1 x float> [[SPLAT_SPLAT27]], <1 x float> [[TMP9]])
-; CHECK-NEXT:    [[BLOCK28:%.*]] = shufflevector <1 x float> [[COL_LOAD11]], <1 x float> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP12:%.*]] = extractelement <7 x float> [[COL_LOAD12]], i64 6
-; CHECK-NEXT:    [[SPLAT_SPLATINSERT29:%.*]] = insertelement <1 x float> poison, float [[TMP12]], i64 0
-; CHECK-NEXT:    [[SPLAT_SPLAT30:%.*]] = shufflevector <1 x float> [[SPLAT_SPLATINSERT29]], <1 x float> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP13:%.*]] = call fast <1 x float> @llvm.fmuladd.v1f32(<1 x float> [[BLOCK28]], <1 x float> [[SPLAT_SPLAT30]], <1 x float> [[TMP11]])
-; CHECK-NEXT:    [[TMP14:%.*]] = shufflevector <1 x float> [[TMP13]], <1 x float> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP15:%.*]] = shufflevector <1 x float> undef, <1 x float> [[TMP14]], <1 x i32> <i32 1>
-; CHECK-NEXT:    ret <1 x float> [[TMP15]]
+; CHECK-NEXT:    [[LHS:%.*]] = load <7 x float>, ptr [[LHS_ADDRESS:%.*]], align 32
+; CHECK-NEXT:    [[COL_LOAD:%.*]] = load <7 x float>, ptr [[RHS_ADDRESS:%.*]], align 32
+; CHECK-NEXT:    [[TMP0:%.*]] = fmul <7 x float> [[LHS]], [[COL_LOAD]]
+; CHECK-NEXT:    [[TMP1:%.*]] = call fast float @llvm.vector.reduce.fadd.v7f32(float 0.000000e+00, <7 x float> [[TMP0]])
+; CHECK-NEXT:    [[TMP2:%.*]] = insertelement <1 x float> poison, float [[TMP1]], i64 0
+; CHECK-NEXT:    ret <1 x float> [[TMP2]]
 ;
 entry:
   %lhs = load <7 x float>, ptr %lhs_address
   %rhs = load <7 x float>, ptr %rhs_address
   %c = tail call fast <1 x float> @llvm.matrix.multiply.v1f32.v7f32.v7f32(<7 x float> %lhs, <7 x float> %rhs, i32 1, i32 7, i32 1)
   ret <1 x float> %c
 }
 
entry:
   ret <1 x double> %result
 }
 
 declare <6 x double> @llvm.matrix.column.major.load.v6f64.i64(ptr nonnull align 4, i64, i1, i32, i32)
 
 define <1 x double> @LoadInst_dot_product_double_v7(ptr %lhs_address, ptr %rhs_address) {
 ; CHECK-LABEL: @LoadInst_dot_product_double_v7(
 ; CHECK-NEXT:  entry:
-; CHECK-NEXT:    [[COL_LOAD:%.*]] = load <1 x double>, ptr [[LHS_ADDRESS:%.*]], align 64
-; CHECK-NEXT:    [[VEC_GEP:%.*]] = getelementptr double, ptr [[LHS_ADDRESS]], i64 1
-; CHECK-NEXT:    [[COL_LOAD1:%.*]] = load <1 x double>, ptr [[VEC_GEP]], align 8
-; CHECK-NEXT:    [[VEC_GEP2:%.*]] = getelementptr double, ptr [[LHS_ADDRESS]], i64 2
-; CHECK-NEXT:    [[COL_LOAD3:%.*]] = load <1 x double>, ptr [[VEC_GEP2]], align 16
-; CHECK-NEXT:    [[VEC_GEP4:%.*]] = getelementptr double, ptr [[LHS_ADDRESS]], i64 3
-; CHECK-NEXT:    [[COL_LOAD5:%.*]] = load <1 x double>, ptr [[VEC_GEP4]], align 8
-; CHECK-NEXT:    [[VEC_GEP6:%.*]] = getelementptr double, ptr [[LHS_ADDRESS]], i64 4
-; CHECK-NEXT:    [[COL_LOAD7:%.*]] = load <1 x double>, ptr [[VEC_GEP6]], align 32
-; CHECK-NEXT:    [[VEC_GEP8:%.*]] = getelementptr double, ptr [[LHS_ADDRESS]], i64 5
-; CHECK-NEXT:    [[COL_LOAD9:%.*]] = load <1 x double>, ptr [[VEC_GEP8]], align 8
-; CHECK-NEXT:    [[VEC_GEP10:%.*]] = getelementptr double, ptr [[LHS_ADDRESS]], i64 6
-; CHECK-NEXT:    [[COL_LOAD11:%.*]] = load <1 x double>, ptr [[VEC_GEP10]], align 16
-; CHECK-NEXT:    [[COL_LOAD12:%.*]] = load <7 x double>, ptr [[RHS_ADDRESS:%.*]], align 64
-; CHECK-NEXT:    [[BLOCK:%.*]] = shufflevector <1 x double> [[COL_LOAD]], <1 x double> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP0:%.*]] = extractelement <7 x double> [[COL_LOAD12]], i64 0
-; CHECK-NEXT:    [[SPLAT_SPLATINSERT:%.*]] = insertelement <1 x double> poison, double [[TMP0]], i64 0
-; CHECK-NEXT:    [[SPLAT_SPLAT:%.*]] = shufflevector <1 x double> [[SPLAT_SPLATINSERT]], <1 x double> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP1:%.*]] = fmul fast <1 x double> [[BLOCK]], [[SPLAT_SPLAT]]
-; CHECK-NEXT:    [[BLOCK13:%.*]] = shufflevector <1 x double> [[COL_LOAD1]], <1 x double> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP2:%.*]] = extractelement <7 x double> [[COL_LOAD12]], i64 1
-; CHECK-NEXT:    [[SPLAT_SPLATINSERT14:%.*]] = insertelement <1 x double> poison, double [[TMP2]], i64 0
-; CHECK-NEXT:    [[SPLAT_SPLAT15:%.*]] = shufflevector <1 x double> [[SPLAT_SPLATINSERT14]], <1 x double> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP3:%.*]] = call fast <1 x double> @llvm.fmuladd.v1f64(<1 x double> [[BLOCK13]], <1 x double> [[SPLAT_SPLAT15]], <1 x double> [[TMP1]])
-; CHECK-NEXT:    [[BLOCK16:%.*]] = shufflevector <1 x double> [[COL_LOAD3]], <1 x double> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP4:%.*]] = extractelement <7 x double> [[COL_LOAD12]], i64 2
-; CHECK-NEXT:    [[SPLAT_SPLATINSERT17:%.*]] = insertelement <1 x double> poison, double [[TMP4]], i64 0
-; CHECK-NEXT:    [[SPLAT_SPLAT18:%.*]] = shufflevector <1 x double> [[SPLAT_SPLATINSERT17]], <1 x double> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP5:%.*]] = call fast <1 x double> @llvm.fmuladd.v1f64(<1 x double> [[BLOCK16]], <1 x double> [[SPLAT_SPLAT18]], <1 x double> [[TMP3]])
-; CHECK-NEXT:    [[BLOCK19:%.*]] = shufflevector <1 x double> [[COL_LOAD5]], <1 x double> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP6:%.*]] = extractelement <7 x double> [[COL_LOAD12]], i64 3
-; CHECK-NEXT:    [[SPLAT_SPLATINSERT20:%.*]] = insertelement <1 x double> poison, double [[TMP6]], i64 0
-; CHECK-NEXT:    [[SPLAT_SPLAT21:%.*]] = shufflevector <1 x double> [[SPLAT_SPLATINSERT20]], <1 x double> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP7:%.*]] = call fast <1 x double> @llvm.fmuladd.v1f64(<1 x double> [[BLOCK19]], <1 x double> [[SPLAT_SPLAT21]], <1 x double> [[TMP5]])
-; CHECK-NEXT:    [[BLOCK22:%.*]] = shufflevector <1 x double> [[COL_LOAD7]], <1 x double> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP8:%.*]] = extractelement <7 x double> [[COL_LOAD12]], i64 4
-; CHECK-NEXT:    [[SPLAT_SPLATINSERT23:%.*]] = insertelement <1 x double> poison, double [[TMP8]], i64 0
-; CHECK-NEXT:    [[SPLAT_SPLAT24:%.*]] = shufflevector <1 x double> [[SPLAT_SPLATINSERT23]], <1 x double> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP9:%.*]] = call fast <1 x double> @llvm.fmuladd.v1f64(<1 x double> [[BLOCK22]], <1 x double> [[SPLAT_SPLAT24]], <1 x double> [[TMP7]])
-; CHECK-NEXT:    [[BLOCK25:%.*]] = shufflevector <1 x double> [[COL_LOAD9]], <1 x double> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP10:%.*]] = extractelement <7 x double> [[COL_LOAD12]], i64 5
-; CHECK-NEXT:    [[SPLAT_SPLATINSERT26:%.*]] = insertelement <1 x double> poison, double [[TMP10]], i64 0
-; CHECK-NEXT:    [[SPLAT_SPLAT27:%.*]] = shufflevector <1 x double> [[SPLAT_SPLATINSERT26]], <1 x double> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP11:%.*]] = call fast <1 x double> @llvm.fmuladd.v1f64(<1 x double> [[BLOCK25]], <1 x double> [[SPLAT_SPLAT27]], <1 x double> [[TMP9]])
-; CHECK-NEXT:    [[BLOCK28:%.*]] = shufflevector <1 x double> [[COL_LOAD11]], <1 x double> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP12:%.*]] = extractelement <7 x double> [[COL_LOAD12]], i64 6
-; CHECK-NEXT:    [[SPLAT_SPLATINSERT29:%.*]] = insertelement <1 x double> poison, double [[TMP12]], i64 0
-; CHECK-NEXT:    [[SPLAT_SPLAT30:%.*]] = shufflevector <1 x double> [[SPLAT_SPLATINSERT29]], <1 x double> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP13:%.*]] = call fast <1 x double> @llvm.fmuladd.v1f64(<1 x double> [[BLOCK28]], <1 x double> [[SPLAT_SPLAT30]], <1 x double> [[TMP11]])
-; CHECK-NEXT:    [[TMP14:%.*]] = shufflevector <1 x double> [[TMP13]], <1 x double> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP15:%.*]] = shufflevector <1 x double> undef, <1 x double> [[TMP14]], <1 x i32> <i32 1>
-; CHECK-NEXT:    ret <1 x double> [[TMP15]]
+; CHECK-NEXT:    [[LHS:%.*]] = load <7 x double>, ptr [[LHS_ADDRESS:%.*]], align 64
+; CHECK-NEXT:    [[COL_LOAD:%.*]] = load <7 x double>, ptr [[RHS_ADDRESS:%.*]], align 64
+; CHECK-NEXT:    [[TMP0:%.*]] = fmul <7 x double> [[LHS]], [[COL_LOAD]]
+; CHECK-NEXT:    [[TMP1:%.*]] = call fast double @llvm.vector.reduce.fadd.v7f64(double 0.000000e+00, <7 x double> [[TMP0]])
+; CHECK-NEXT:    [[TMP2:%.*]] = insertelement <1 x double> poison, double [[TMP1]], i64 0
+; CHECK-NEXT:    ret <1 x double> [[TMP2]]
 ;
 entry:
   %lhs = load <7 x double>, ptr %lhs_address
   %rhs = load <7 x double>, ptr %rhs_address
   %c = tail call fast <1 x double> @llvm.matrix.multiply.v1f64.v7f64.v7f64(<7 x double> %lhs, <7 x double> %rhs, i32 1, i32 7, i32 1)
   ret <1 x double> %c
 }
 
 declare <1 x double> @llvm.matrix.multiply.v1f64.v7f64.v7f64(<7 x double>, <7 x double>, i32, i32, i32)

llvm/test/Transforms/LowerMatrixIntrinsics/dot-product-int.ll

 ; CHECK-NEXT:    [[TMP0:%.*]] = add <1 x i32> [[SPLIT]], [[SPLIT8]]
 ; CHECK-NEXT:    [[TMP1:%.*]] = add <1 x i32> [[SPLIT1]], [[SPLIT9]]
 ; CHECK-NEXT:    [[TMP2:%.*]] = add <1 x i32> [[SPLIT2]], [[SPLIT10]]
 ; CHECK-NEXT:    [[TMP3:%.*]] = add <1 x i32> [[SPLIT3]], [[SPLIT11]]
 ; CHECK-NEXT:    [[TMP4:%.*]] = add <1 x i32> [[SPLIT4]], [[SPLIT12]]
 ; CHECK-NEXT:    [[TMP5:%.*]] = add <1 x i32> [[SPLIT5]], [[SPLIT13]]
 ; CHECK-NEXT:    [[TMP6:%.*]] = add <1 x i32> [[SPLIT6]], [[SPLIT14]]
 ; CHECK-NEXT:    [[TMP7:%.*]] = add <1 x i32> [[SPLIT7]], [[SPLIT15]]
-; CHECK-NEXT:    [[SPLIT16:%.*]] = shufflevector <8 x i32> [[C:%.*]], <8 x i32> poison, <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7>
-; CHECK-NEXT:    [[BLOCK:%.*]] = shufflevector <1 x i32> [[TMP0]], <1 x i32> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP8:%.*]] = extractelement <8 x i32> [[SPLIT16]], i64 0
-; CHECK-NEXT:    [[SPLAT_SPLATINSERT:%.*]] = insertelement <1 x i32> poison, i32 [[TMP8]], i64 0
-; CHECK-NEXT:    [[SPLAT_SPLAT:%.*]] = shufflevector <1 x i32> [[SPLAT_SPLATINSERT]], <1 x i32> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP9:%.*]] = mul <1 x i32> [[BLOCK]], [[SPLAT_SPLAT]]
-; CHECK-NEXT:    [[BLOCK17:%.*]] = shufflevector <1 x i32> [[TMP1]], <1 x i32> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP10:%.*]] = extractelement <8 x i32> [[SPLIT16]], i64 1
-; CHECK-NEXT:    [[SPLAT_SPLATINSERT18:%.*]] = insertelement <1 x i32> poison, i32 [[TMP10]], i64 0
-; CHECK-NEXT:    [[SPLAT_SPLAT19:%.*]] = shufflevector <1 x i32> [[SPLAT_SPLATINSERT18]], <1 x i32> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP11:%.*]] = mul <1 x i32> [[BLOCK17]], [[SPLAT_SPLAT19]]
-; CHECK-NEXT:    [[TMP12:%.*]] = add <1 x i32> [[TMP9]], [[TMP11]]
-; CHECK-NEXT:    [[BLOCK20:%.*]] = shufflevector <1 x i32> [[TMP2]], <1 x i32> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP13:%.*]] = extractelement <8 x i32> [[SPLIT16]], i64 2
-; CHECK-NEXT:    [[SPLAT_SPLATINSERT21:%.*]] = insertelement <1 x i32> poison, i32 [[TMP13]], i64 0
-; CHECK-NEXT:    [[SPLAT_SPLAT22:%.*]] = shufflevector <1 x i32> [[SPLAT_SPLATINSERT21]], <1 x i32> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP14:%.*]] = mul <1 x i32> [[BLOCK20]], [[SPLAT_SPLAT22]]
-; CHECK-NEXT:    [[TMP15:%.*]] = add <1 x i32> [[TMP12]], [[TMP14]]
-; CHECK-NEXT:    [[BLOCK23:%.*]] = shufflevector <1 x i32> [[TMP3]], <1 x i32> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP16:%.*]] = extractelement <8 x i32> [[SPLIT16]], i64 3
-; CHECK-NEXT:    [[SPLAT_SPLATINSERT24:%.*]] = insertelement <1 x i32> poison, i32 [[TMP16]], i64 0
-; CHECK-NEXT:    [[SPLAT_SPLAT25:%.*]] = shufflevector <1 x i32> [[SPLAT_SPLATINSERT24]], <1 x i32> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP17:%.*]] = mul <1 x i32> [[BLOCK23]], [[SPLAT_SPLAT25]]
-; CHECK-NEXT:    [[TMP18:%.*]] = add <1 x i32> [[TMP15]], [[TMP17]]
-; CHECK-NEXT:    [[BLOCK26:%.*]] = shufflevector <1 x i32> [[TMP4]], <1 x i32> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP19:%.*]] = extractelement <8 x i32> [[SPLIT16]], i64 4
-; CHECK-NEXT:    [[SPLAT_SPLATINSERT27:%.*]] = insertelement <1 x i32> poison, i32 [[TMP19]], i64 0
-; CHECK-NEXT:    [[SPLAT_SPLAT28:%.*]] = shufflevector <1 x i32> [[SPLAT_SPLATINSERT27]], <1 x i32> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP20:%.*]] = mul <1 x i32> [[BLOCK26]], [[SPLAT_SPLAT28]]
-; CHECK-NEXT:    [[TMP21:%.*]] = add <1 x i32> [[TMP18]], [[TMP20]]
-; CHECK-NEXT:    [[BLOCK29:%.*]] = shufflevector <1 x i32> [[TMP5]], <1 x i32> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP22:%.*]] = extractelement <8 x i32> [[SPLIT16]], i64 5
-; CHECK-NEXT:    [[SPLAT_SPLATINSERT30:%.*]] = insertelement <1 x i32> poison, i32 [[TMP22]], i64 0
-; CHECK-NEXT:    [[SPLAT_SPLAT31:%.*]] = shufflevector <1 x i32> [[SPLAT_SPLATINSERT30]], <1 x i32> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP23:%.*]] = mul <1 x i32> [[BLOCK29]], [[SPLAT_SPLAT31]]
-; CHECK-NEXT:    [[TMP24:%.*]] = add <1 x i32> [[TMP21]], [[TMP23]]
-; CHECK-NEXT:    [[BLOCK32:%.*]] = shufflevector <1 x i32> [[TMP6]], <1 x i32> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP25:%.*]] = extractelement <8 x i32> [[SPLIT16]], i64 6
-; CHECK-NEXT:    [[SPLAT_SPLATINSERT33:%.*]] = insertelement <1 x i32> poison, i32 [[TMP25]], i64 0
-; CHECK-NEXT:    [[SPLAT_SPLAT34:%.*]] = shufflevector <1 x i32> [[SPLAT_SPLATINSERT33]], <1 x i32> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP26:%.*]] = mul <1 x i32> [[BLOCK32]], [[SPLAT_SPLAT34]]
-; CHECK-NEXT:    [[TMP27:%.*]] = add <1 x i32> [[TMP24]], [[TMP26]]
-; CHECK-NEXT:    [[BLOCK35:%.*]] = shufflevector <1 x i32> [[TMP7]], <1 x i32> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP28:%.*]] = extractelement <8 x i32> [[SPLIT16]], i64 7
-; CHECK-NEXT:    [[SPLAT_SPLATINSERT36:%.*]] = insertelement <1 x i32> poison, i32 [[TMP28]], i64 0
-; CHECK-NEXT:    [[SPLAT_SPLAT37:%.*]] = shufflevector <1 x i32> [[SPLAT_SPLATINSERT36]], <1 x i32> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP29:%.*]] = mul <1 x i32> [[BLOCK35]], [[SPLAT_SPLAT37]]
-; CHECK-NEXT:    [[TMP30:%.*]] = add <1 x i32> [[TMP27]], [[TMP29]]
-; CHECK-NEXT:    [[TMP31:%.*]] = shufflevector <1 x i32> [[TMP30]], <1 x i32> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP32:%.*]] = shufflevector <1 x i32> undef, <1 x i32> [[TMP31]], <1 x i32> <i32 1>
-; CHECK-NEXT:    ret <1 x i32> [[TMP32]]
+; CHECK-NEXT:    [[TMP8:%.*]] = shufflevector <1 x i32> [[TMP0]], <1 x i32> [[TMP1]], <2 x i32> <i32 0, i32 1>
+; CHECK-NEXT:    [[TMP9:%.*]] = shufflevector <1 x i32> [[TMP2]], <1 x i32> [[TMP3]], <2 x i32> <i32 0, i32 1>
+; CHECK-NEXT:    [[TMP10:%.*]] = shufflevector <1 x i32> [[TMP4]], <1 x i32> [[TMP5]], <2 x i32> <i32 0, i32 1>
+; CHECK-NEXT:    [[TMP11:%.*]] = shufflevector <1 x i32> [[TMP6]], <1 x i32> [[TMP7]], <2 x i32> <i32 0, i32 1>
+; CHECK-NEXT:    [[TMP12:%.*]] = shufflevector <2 x i32> [[TMP8]], <2 x i32> [[TMP9]], <4 x i32> <i32 0, i32 1, i32 2, i32 3>
+; CHECK-NEXT:    [[TMP13:%.*]] = shufflevector <2 x i32> [[TMP10]], <2 x i32> [[TMP11]], <4 x i32> <i32 0, i32 1, i32 2, i32 3>
+; CHECK-NEXT:    [[TMP14:%.*]] = shufflevector <4 x i32> [[TMP12]], <4 x i32> [[TMP13]], <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7>
+; CHECK-NEXT:    [[TMP15:%.*]] = mul <8 x i32> [[TMP14]], [[C:%.*]]
+; CHECK-NEXT:    [[TMP16:%.*]] = call i32 @llvm.vector.reduce.add.v8i32(<8 x i32> [[TMP15]])
+; CHECK-NEXT:    [[TMP17:%.*]] = insertelement <1 x i32> poison, i32 [[TMP16]], i64 0
+; CHECK-NEXT:    ret <1 x i32> [[TMP17]]
 ;
 entry:
   %add = add <8 x i32> %a, %b
   %res = tail call <1 x i32> @llvm.matrix.multiply.v1i32.v8i32.v8i32(<8 x i32> %add, <8 x i32> %c, i32 1, i32 8, i32 1)
   ret <1 x i32> %res
 }
 
 define <1 x i32> @add_feeding_dotproduct_i32_v8_2(<8 x i32> %a, <8 x i32> %b, <8 x i32> %c) {

 ; CHECK-NEXT:    [[TMP0:%.*]] = sub <1 x i32> [[SPLIT]], [[SPLIT8]]
 ; CHECK-NEXT:    [[TMP1:%.*]] = sub <1 x i32> [[SPLIT1]], [[SPLIT9]]
 ; CHECK-NEXT:    [[TMP2:%.*]] = sub <1 x i32> [[SPLIT2]], [[SPLIT10]]
 ; CHECK-NEXT:    [[TMP3:%.*]] = sub <1 x i32> [[SPLIT3]], [[SPLIT11]]
 ; CHECK-NEXT:    [[TMP4:%.*]] = sub <1 x i32> [[SPLIT4]], [[SPLIT12]]
 ; CHECK-NEXT:    [[TMP5:%.*]] = sub <1 x i32> [[SPLIT5]], [[SPLIT13]]
 ; CHECK-NEXT:    [[TMP6:%.*]] = sub <1 x i32> [[SPLIT6]], [[SPLIT14]]
 ; CHECK-NEXT:    [[TMP7:%.*]] = sub <1 x i32> [[SPLIT7]], [[SPLIT15]]
-; CHECK-NEXT:    [[SPLIT16:%.*]] = shufflevector <8 x i32> [[C:%.*]], <8 x i32> poison, <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7>
-; CHECK-NEXT:    [[BLOCK:%.*]] = shufflevector <1 x i32> [[TMP0]], <1 x i32> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP8:%.*]] = extractelement <8 x i32> [[SPLIT16]], i64 0
-; CHECK-NEXT:    [[SPLAT_SPLATINSERT:%.*]] = insertelement <1 x i32> poison, i32 [[TMP8]], i64 0
-; CHECK-NEXT:    [[SPLAT_SPLAT:%.*]] = shufflevector <1 x i32> [[SPLAT_SPLATINSERT]], <1 x i32> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP9:%.*]] = mul <1 x i32> [[BLOCK]], [[SPLAT_SPLAT]]
-; CHECK-NEXT:    [[BLOCK17:%.*]] = shufflevector <1 x i32> [[TMP1]], <1 x i32> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP10:%.*]] = extractelement <8 x i32> [[SPLIT16]], i64 1
-; CHECK-NEXT:    [[SPLAT_SPLATINSERT18:%.*]] = insertelement <1 x i32> poison, i32 [[TMP10]], i64 0
-; CHECK-NEXT:    [[SPLAT_SPLAT19:%.*]] = shufflevector <1 x i32> [[SPLAT_SPLATINSERT18]], <1 x i32> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP11:%.*]] = mul <1 x i32> [[BLOCK17]], [[SPLAT_SPLAT19]]
-; CHECK-NEXT:    [[TMP12:%.*]] = add <1 x i32> [[TMP9]], [[TMP11]]
-; CHECK-NEXT:    [[BLOCK20:%.*]] = shufflevector <1 x i32> [[TMP2]], <1 x i32> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP13:%.*]] = extractelement <8 x i32> [[SPLIT16]], i64 2
-; CHECK-NEXT:    [[SPLAT_SPLATINSERT21:%.*]] = insertelement <1 x i32> poison, i32 [[TMP13]], i64 0
-; CHECK-NEXT:    [[SPLAT_SPLAT22:%.*]] = shufflevector <1 x i32> [[SPLAT_SPLATINSERT21]], <1 x i32> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP14:%.*]] = mul <1 x i32> [[BLOCK20]], [[SPLAT_SPLAT22]]
-; CHECK-NEXT:    [[TMP15:%.*]] = add <1 x i32> [[TMP12]], [[TMP14]]
-; CHECK-NEXT:    [[BLOCK23:%.*]] = shufflevector <1 x i32> [[TMP3]], <1 x i32> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP16:%.*]] = extractelement <8 x i32> [[SPLIT16]], i64 3
-; CHECK-NEXT:    [[SPLAT_SPLATINSERT24:%.*]] = insertelement <1 x i32> poison, i32 [[TMP16]], i64 0
-; CHECK-NEXT:    [[SPLAT_SPLAT25:%.*]] = shufflevector <1 x i32> [[SPLAT_SPLATINSERT24]], <1 x i32> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP17:%.*]] = mul <1 x i32> [[BLOCK23]], [[SPLAT_SPLAT25]]
-; CHECK-NEXT:    [[TMP18:%.*]] = add <1 x i32> [[TMP15]], [[TMP17]]
-; CHECK-NEXT:    [[BLOCK26:%.*]] = shufflevector <1 x i32> [[TMP4]], <1 x i32> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP19:%.*]] = extractelement <8 x i32> [[SPLIT16]], i64 4
-; CHECK-NEXT:    [[SPLAT_SPLATINSERT27:%.*]] = insertelement <1 x i32> poison, i32 [[TMP19]], i64 0
-; CHECK-NEXT:    [[SPLAT_SPLAT28:%.*]] = shufflevector <1 x i32> [[SPLAT_SPLATINSERT27]], <1 x i32> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP20:%.*]] = mul <1 x i32> [[BLOCK26]], [[SPLAT_SPLAT28]]
-; CHECK-NEXT:    [[TMP21:%.*]] = add <1 x i32> [[TMP18]], [[TMP20]]
-; CHECK-NEXT:    [[BLOCK29:%.*]] = shufflevector <1 x i32> [[TMP5]], <1 x i32> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP22:%.*]] = extractelement <8 x i32> [[SPLIT16]], i64 5
-; CHECK-NEXT:    [[SPLAT_SPLATINSERT30:%.*]] = insertelement <1 x i32> poison, i32 [[TMP22]], i64 0
-; CHECK-NEXT:    [[SPLAT_SPLAT31:%.*]] = shufflevector <1 x i32> [[SPLAT_SPLATINSERT30]], <1 x i32> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP23:%.*]] = mul <1 x i32> [[BLOCK29]], [[SPLAT_SPLAT31]]
-; CHECK-NEXT:    [[TMP24:%.*]] = add <1 x i32> [[TMP21]], [[TMP23]]
-; CHECK-NEXT:    [[BLOCK32:%.*]] = shufflevector <1 x i32> [[TMP6]], <1 x i32> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP25:%.*]] = extractelement <8 x i32> [[SPLIT16]], i64 6
-; CHECK-NEXT:    [[SPLAT_SPLATINSERT33:%.*]] = insertelement <1 x i32> poison, i32 [[TMP25]], i64 0
-; CHECK-NEXT:    [[SPLAT_SPLAT34:%.*]] = shufflevector <1 x i32> [[SPLAT_SPLATINSERT33]], <1 x i32> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP26:%.*]] = mul <1 x i32> [[BLOCK32]], [[SPLAT_SPLAT34]]
-; CHECK-NEXT:    [[TMP27:%.*]] = add <1 x i32> [[TMP24]], [[TMP26]]
-; CHECK-NEXT:    [[BLOCK35:%.*]] = shufflevector <1 x i32> [[TMP7]], <1 x i32> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP28:%.*]] = extractelement <8 x i32> [[SPLIT16]], i64 7
-; CHECK-NEXT:    [[SPLAT_SPLATINSERT36:%.*]] = insertelement <1 x i32> poison, i32 [[TMP28]], i64 0
-; CHECK-NEXT:    [[SPLAT_SPLAT37:%.*]] = shufflevector <1 x i32> [[SPLAT_SPLATINSERT36]], <1 x i32> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP29:%.*]] = mul <1 x i32> [[BLOCK35]], [[SPLAT_SPLAT37]]
-; CHECK-NEXT:    [[TMP30:%.*]] = add <1 x i32> [[TMP27]], [[TMP29]]
-; CHECK-NEXT:    [[TMP31:%.*]] = shufflevector <1 x i32> [[TMP30]], <1 x i32> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP32:%.*]] = shufflevector <1 x i32> undef, <1 x i32> [[TMP31]], <1 x i32> <i32 1>
-; CHECK-NEXT:    ret <1 x i32> [[TMP32]]
+; CHECK-NEXT:    [[TMP8:%.*]] = shufflevector <1 x i32> [[TMP0]], <1 x i32> [[TMP1]], <2 x i32> <i32 0, i32 1>
+; CHECK-NEXT:    [[TMP9:%.*]] = shufflevector <1 x i32> [[TMP2]], <1 x i32> [[TMP3]], <2 x i32> <i32 0, i32 1>
+; CHECK-NEXT:    [[TMP10:%.*]] = shufflevector <1 x i32> [[TMP4]], <1 x i32> [[TMP5]], <2 x i32> <i32 0, i32 1>
+; CHECK-NEXT:    [[TMP11:%.*]] = shufflevector <1 x i32> [[TMP6]], <1 x i32> [[TMP7]], <2 x i32> <i32 0, i32 1>
+; CHECK-NEXT:    [[TMP12:%.*]] = shufflevector <2 x i32> [[TMP8]], <2 x i32> [[TMP9]], <4 x i32> <i32 0, i32 1, i32 2, i32 3>
+; CHECK-NEXT:    [[TMP13:%.*]] = shufflevector <2 x i32> [[TMP10]], <2 x i32> [[TMP11]], <4 x i32> <i32 0, i32 1, i32 2, i32 3>
+; CHECK-NEXT:    [[TMP14:%.*]] = shufflevector <4 x i32> [[TMP12]], <4 x i32> [[TMP13]], <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7>
+; CHECK-NEXT:    [[TMP15:%.*]] = mul <8 x i32> [[TMP14]], [[C:%.*]]
+; CHECK-NEXT:    [[TMP16:%.*]] = call i32 @llvm.vector.reduce.add.v8i32(<8 x i32> [[TMP15]])
+; CHECK-NEXT:    [[TMP17:%.*]] = insertelement <1 x i32> poison, i32 [[TMP16]], i64 0
+; CHECK-NEXT:    ret <1 x i32> [[TMP17]]
 ;
 entry:
   %sub = sub <8 x i32> %a, %b
   %res = tail call <1 x i32> @llvm.matrix.multiply.v1i32.v8i32.v8i32(<8 x i32> %sub, <8 x i32> %c, i32 1, i32 8, i32 1)
   ret <1 x i32> %res
 }
 
 define <1 x i32> @sub_feeding_dotproduct_i32_v8_2(<8 x i32> %a, <8 x i32> %b, <8 x i32> %c) {

 ; CHECK-NEXT:    [[TMP8:%.*]] = add <1 x i32> [[TMP0]], [[SPLIT16]]
 ; CHECK-NEXT:    [[TMP9:%.*]] = add <1 x i32> [[TMP1]], [[SPLIT17]]
 ; CHECK-NEXT:    [[TMP10:%.*]] = add <1 x i32> [[TMP2]], [[SPLIT18]]
 ; CHECK-NEXT:    [[TMP11:%.*]] = add <1 x i32> [[TMP3]], [[SPLIT19]]
 ; CHECK-NEXT:    [[TMP12:%.*]] = add <1 x i32> [[TMP4]], [[SPLIT20]]
 ; CHECK-NEXT:    [[TMP13:%.*]] = add <1 x i32> [[TMP5]], [[SPLIT21]]
 ; CHECK-NEXT:    [[TMP14:%.*]] = add <1 x i32> [[TMP6]], [[SPLIT22]]
 ; CHECK-NEXT:    [[TMP15:%.*]] = add <1 x i32> [[TMP7]], [[SPLIT23]]
-; CHECK-NEXT:    [[SPLIT24:%.*]] = shufflevector <8 x i32> [[D:%.*]], <8 x i32> poison, <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7>
-; CHECK-NEXT:    [[BLOCK:%.*]] = shufflevector <1 x i32> [[TMP8]], <1 x i32> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP16:%.*]] = extractelement <8 x i32> [[SPLIT24]], i64 0
-; CHECK-NEXT:    [[SPLAT_SPLATINSERT:%.*]] = insertelement <1 x i32> poison, i32 [[TMP16]], i64 0
-; CHECK-NEXT:    [[SPLAT_SPLAT:%.*]] = shufflevector <1 x i32> [[SPLAT_SPLATINSERT]], <1 x i32> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP17:%.*]] = mul <1 x i32> [[BLOCK]], [[SPLAT_SPLAT]]
-; CHECK-NEXT:    [[BLOCK25:%.*]] = shufflevector <1 x i32> [[TMP9]], <1 x i32> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP18:%.*]] = extractelement <8 x i32> [[SPLIT24]], i64 1
-; CHECK-NEXT:    [[SPLAT_SPLATINSERT26:%.*]] = insertelement <1 x i32> poison, i32 [[TMP18]], i64 0
-; CHECK-NEXT:    [[SPLAT_SPLAT27:%.*]] = shufflevector <1 x i32> [[SPLAT_SPLATINSERT26]], <1 x i32> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP19:%.*]] = mul <1 x i32> [[BLOCK25]], [[SPLAT_SPLAT27]]
-; CHECK-NEXT:    [[TMP20:%.*]] = add <1 x i32> [[TMP17]], [[TMP19]]
-; CHECK-NEXT:    [[BLOCK28:%.*]] = shufflevector <1 x i32> [[TMP10]], <1 x i32> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP21:%.*]] = extractelement <8 x i32> [[SPLIT24]], i64 2
-; CHECK-NEXT:    [[SPLAT_SPLATINSERT29:%.*]] = insertelement <1 x i32> poison, i32 [[TMP21]], i64 0
-; CHECK-NEXT:    [[SPLAT_SPLAT30:%.*]] = shufflevector <1 x i32> [[SPLAT_SPLATINSERT29]], <1 x i32> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP22:%.*]] = mul <1 x i32> [[BLOCK28]], [[SPLAT_SPLAT30]]
-; CHECK-NEXT:    [[TMP23:%.*]] = add <1 x i32> [[TMP20]], [[TMP22]]
-; CHECK-NEXT:    [[BLOCK31:%.*]] = shufflevector <1 x i32> [[TMP11]], <1 x i32> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP24:%.*]] = extractelement <8 x i32> [[SPLIT24]], i64 3
-; CHECK-NEXT:    [[SPLAT_SPLATINSERT32:%.*]] = insertelement <1 x i32> poison, i32 [[TMP24]], i64 0
-; CHECK-NEXT:    [[SPLAT_SPLAT33:%.*]] = shufflevector <1 x i32> [[SPLAT_SPLATINSERT32]], <1 x i32> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP25:%.*]] = mul <1 x i32> [[BLOCK31]], [[SPLAT_SPLAT33]]
-; CHECK-NEXT:    [[TMP26:%.*]] = add <1 x i32> [[TMP23]], [[TMP25]]
-; CHECK-NEXT:    [[BLOCK34:%.*]] = shufflevector <1 x i32> [[TMP12]], <1 x i32> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP27:%.*]] = extractelement <8 x i32> [[SPLIT24]], i64 4
-; CHECK-NEXT:    [[SPLAT_SPLATINSERT35:%.*]] = insertelement <1 x i32> poison, i32 [[TMP27]], i64 0
-; CHECK-NEXT:    [[SPLAT_SPLAT36:%.*]] = shufflevector <1 x i32> [[SPLAT_SPLATINSERT35]], <1 x i32> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP28:%.*]] = mul <1 x i32> [[BLOCK34]], [[SPLAT_SPLAT36]]
-; CHECK-NEXT:    [[TMP29:%.*]] = add <1 x i32> [[TMP26]], [[TMP28]]
-; CHECK-NEXT:    [[BLOCK37:%.*]] = shufflevector <1 x i32> [[TMP13]], <1 x i32> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP30:%.*]] = extractelement <8 x i32> [[SPLIT24]], i64 5
-; CHECK-NEXT:    [[SPLAT_SPLATINSERT38:%.*]] = insertelement <1 x i32> poison, i32 [[TMP30]], i64 0
-; CHECK-NEXT:    [[SPLAT_SPLAT39:%.*]] = shufflevector <1 x i32> [[SPLAT_SPLATINSERT38]], <1 x i32> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP31:%.*]] = mul <1 x i32> [[BLOCK37]], [[SPLAT_SPLAT39]]
-; CHECK-NEXT:    [[TMP32:%.*]] = add <1 x i32> [[TMP29]], [[TMP31]]
-; CHECK-NEXT:    [[BLOCK40:%.*]] = shufflevector <1 x i32> [[TMP14]], <1 x i32> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP33:%.*]] = extractelement <8 x i32> [[SPLIT24]], i64 6
-; CHECK-NEXT:    [[SPLAT_SPLATINSERT41:%.*]] = insertelement <1 x i32> poison, i32 [[TMP33]], i64 0
-; CHECK-NEXT:    [[SPLAT_SPLAT42:%.*]] = shufflevector <1 x i32> [[SPLAT_SPLATINSERT41]], <1 x i32> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP34:%.*]] = mul <1 x i32> [[BLOCK40]], [[SPLAT_SPLAT42]]
-; CHECK-NEXT:    [[TMP35:%.*]] = add <1 x i32> [[TMP32]], [[TMP34]]
-; CHECK-NEXT:    [[BLOCK43:%.*]] = shufflevector <1 x i32> [[TMP15]], <1 x i32> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP36:%.*]] = extractelement <8 x i32> [[SPLIT24]], i64 7
-; CHECK-NEXT:    [[SPLAT_SPLATINSERT44:%.*]] = insertelement <1 x i32> poison, i32 [[TMP36]], i64 0
-; CHECK-NEXT:    [[SPLAT_SPLAT45:%.*]] = shufflevector <1 x i32> [[SPLAT_SPLATINSERT44]], <1 x i32> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP37:%.*]] = mul <1 x i32> [[BLOCK43]], [[SPLAT_SPLAT45]]
-; CHECK-NEXT:    [[TMP38:%.*]] = add <1 x i32> [[TMP35]], [[TMP37]]
-; CHECK-NEXT:    [[TMP39:%.*]] = shufflevector <1 x i32> [[TMP38]], <1 x i32> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP40:%.*]] = shufflevector <1 x i32> undef, <1 x i32> [[TMP39]], <1 x i32> <i32 1>
-; CHECK-NEXT:    ret <1 x i32> [[TMP40]]
+; CHECK-NEXT:    [[TMP16:%.*]] = shufflevector <1 x i32> [[TMP8]], <1 x i32> [[TMP9]], <2 x i32> <i32 0, i32 1>
+; CHECK-NEXT:    [[TMP17:%.*]] = shufflevector <1 x i32> [[TMP10]], <1 x i32> [[TMP11]], <2 x i32> <i32 0, i32 1>
+; CHECK-NEXT:    [[TMP18:%.*]] = shufflevector <1 x i32> [[TMP12]], <1 x i32> [[TMP13]], <2 x i32> <i32 0, i32 1>
+; CHECK-NEXT:    [[TMP19:%.*]] = shufflevector <1 x i32> [[TMP14]], <1 x i32> [[TMP15]], <2 x i32> <i32 0, i32 1>
+; CHECK-NEXT:    [[TMP20:%.*]] = shufflevector <2 x i32> [[TMP16]], <2 x i32> [[TMP17]], <4 x i32> <i32 0, i32 1, i32 2, i32 3>
+; CHECK-NEXT:    [[TMP21:%.*]] = shufflevector <2 x i32> [[TMP18]], <2 x i32> [[TMP19]], <4 x i32> <i32 0, i32 1, i32 2, i32 3>
+; CHECK-NEXT:    [[TMP22:%.*]] = shufflevector <4 x i32> [[TMP20]], <4 x i32> [[TMP21]], <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7>
+; CHECK-NEXT:    [[TMP23:%.*]] = mul <8 x i32> [[TMP22]], [[D:%.*]]
+; CHECK-NEXT:    [[TMP24:%.*]] = call i32 @llvm.vector.reduce.add.v8i32(<8 x i32> [[TMP23]])
+; CHECK-NEXT:    [[TMP25:%.*]] = insertelement <1 x i32> poison, i32 [[TMP24]], i64 0
+; CHECK-NEXT:    ret <1 x i32> [[TMP25]]
 ;
 entry:
   %add.1 = add <8 x i32> %a, %b
   %add.2 = add <8 x i32> %add.1, %c
   %res = tail call <1 x i32> @llvm.matrix.multiply.v1i32.v8i32.v8i32(<8 x i32> %add.2, <8 x i32> %d, i32 1, i32 8, i32 1)
   ret <1 x i32> %res
 }
 
entry:
   %add.2 = add <8 x i32> %add.1, %c
   %res = tail call <1 x i32> @llvm.matrix.multiply.v1i32.v8i32.v8i32(<8 x i32> %d, <8 x i32> %add.2, i32 1, i32 8, i32 1)
   ret <1 x i32> %res
 }
 
 define <1 x i64> @dotproduct_i64_v8(<8 x i64> %a, <8 x i64> %b) {
 ; CHECK-LABEL: @dotproduct_i64_v8(
 ; CHECK-NEXT:  entry:
-; CHECK-NEXT:    [[TMP0:%.*]] = mul <8 x i64> [[A:%.*]], [[B:%.*]]
-; CHECK-NEXT:    [[TMP1:%.*]] = call i64 @llvm.vector.reduce.add.v8i64(<8 x i64> [[TMP0]])
-; CHECK-NEXT:    [[TMP2:%.*]] = insertelement <1 x i64> poison, i64 [[TMP1]], i64 0
-; CHECK-NEXT:    ret <1 x i64> [[TMP2]]
+; CHECK-NEXT:    [[SPLIT:%.*]] = shufflevector <8 x i64> [[A:%.*]], <8 x i64> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[SPLIT1:%.*]] = shufflevector <8 x i64> [[A]], <8 x i64> poison, <1 x i32> <i32 1>
+; CHECK-NEXT:    [[SPLIT2:%.*]] = shufflevector <8 x i64> [[A]], <8 x i64> poison, <1 x i32> <i32 2>
+; CHECK-NEXT:    [[SPLIT3:%.*]] = shufflevector <8 x i64> [[A]], <8 x i64> poison, <1 x i32> <i32 3>
+; CHECK-NEXT:    [[SPLIT4:%.*]] = shufflevector <8 x i64> [[A]], <8 x i64> poison, <1 x i32> <i32 4>
+; CHECK-NEXT:    [[SPLIT5:%.*]] = shufflevector <8 x i64> [[A]], <8 x i64> poison, <1 x i32> <i32 5>
+; CHECK-NEXT:    [[SPLIT6:%.*]] = shufflevector <8 x i64> [[A]], <8 x i64> poison, <1 x i32> <i32 6>
+; CHECK-NEXT:    [[SPLIT7:%.*]] = shufflevector <8 x i64> [[A]], <8 x i64> poison, <1 x i32> <i32 7>
+; CHECK-NEXT:    [[SPLIT8:%.*]] = shufflevector <8 x i64> [[B:%.*]], <8 x i64> poison, <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7>
+; CHECK-NEXT:    [[BLOCK:%.*]] = shufflevector <1 x i64> [[SPLIT]], <1 x i64> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP0:%.*]] = extractelement <8 x i64> [[SPLIT8]], i64 0
+; CHECK-NEXT:    [[SPLAT_SPLATINSERT:%.*]] = insertelement <1 x i64> poison, i64 [[TMP0]], i64 0
+; CHECK-NEXT:    [[SPLAT_SPLAT:%.*]] = shufflevector <1 x i64> [[SPLAT_SPLATINSERT]], <1 x i64> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP1:%.*]] = mul <1 x i64> [[BLOCK]], [[SPLAT_SPLAT]]
+; CHECK-NEXT:    [[BLOCK9:%.*]] = shufflevector <1 x i64> [[SPLIT1]], <1 x i64> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP2:%.*]] = extractelement <8 x i64> [[SPLIT8]], i64 1
+; CHECK-NEXT:    [[SPLAT_SPLATINSERT10:%.*]] = insertelement <1 x i64> poison, i64 [[TMP2]], i64 0
+; CHECK-NEXT:    [[SPLAT_SPLAT11:%.*]] = shufflevector <1 x i64> [[SPLAT_SPLATINSERT10]], <1 x i64> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP3:%.*]] = mul <1 x i64> [[BLOCK9]], [[SPLAT_SPLAT11]]
+; CHECK-NEXT:    [[TMP4:%.*]] = add <1 x i64> [[TMP1]], [[TMP3]]
+; CHECK-NEXT:    [[BLOCK12:%.*]] = shufflevector <1 x i64> [[SPLIT2]], <1 x i64> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP5:%.*]] = extractelement <8 x i64> [[SPLIT8]], i64 2
+; CHECK-NEXT:    [[SPLAT_SPLATINSERT13:%.*]] = insertelement <1 x i64> poison, i64 [[TMP5]], i64 0
+; CHECK-NEXT:    [[SPLAT_SPLAT14:%.*]] = shufflevector <1 x i64> [[SPLAT_SPLATINSERT13]], <1 x i64> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP6:%.*]] = mul <1 x i64> [[BLOCK12]], [[SPLAT_SPLAT14]]
+; CHECK-NEXT:    [[TMP7:%.*]] = add <1 x i64> [[TMP4]], [[TMP6]]
+; CHECK-NEXT:    [[BLOCK15:%.*]] = shufflevector <1 x i64> [[SPLIT3]], <1 x i64> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP8:%.*]] = extractelement <8 x i64> [[SPLIT8]], i64 3
+; CHECK-NEXT:    [[SPLAT_SPLATINSERT16:%.*]] = insertelement <1 x i64> poison, i64 [[TMP8]], i64 0
+; CHECK-NEXT:    [[SPLAT_SPLAT17:%.*]] = shufflevector <1 x i64> [[SPLAT_SPLATINSERT16]], <1 x i64> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP9:%.*]] = mul <1 x i64> [[BLOCK15]], [[SPLAT_SPLAT17]]
+; CHECK-NEXT:    [[TMP10:%.*]] = add <1 x i64> [[TMP7]], [[TMP9]]
+; CHECK-NEXT:    [[BLOCK18:%.*]] = shufflevector <1 x i64> [[SPLIT4]], <1 x i64> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP11:%.*]] = extractelement <8 x i64> [[SPLIT8]], i64 4
+; CHECK-NEXT:    [[SPLAT_SPLATINSERT19:%.*]] = insertelement <1 x i64> poison, i64 [[TMP11]], i64 0
+; CHECK-NEXT:    [[SPLAT_SPLAT20:%.*]] = shufflevector <1 x i64> [[SPLAT_SPLATINSERT19]], <1 x i64> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP12:%.*]] = mul <1 x i64> [[BLOCK18]], [[SPLAT_SPLAT20]]
+; CHECK-NEXT:    [[TMP13:%.*]] = add <1 x i64> [[TMP10]], [[TMP12]]
+; CHECK-NEXT:    [[BLOCK21:%.*]] = shufflevector <1 x i64> [[SPLIT5]], <1 x i64> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP14:%.*]] = extractelement <8 x i64> [[SPLIT8]], i64 5
+; CHECK-NEXT:    [[SPLAT_SPLATINSERT22:%.*]] = insertelement <1 x i64> poison, i64 [[TMP14]], i64 0
+; CHECK-NEXT:    [[SPLAT_SPLAT23:%.*]] = shufflevector <1 x i64> [[SPLAT_SPLATINSERT22]], <1 x i64> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP15:%.*]] = mul <1 x i64> [[BLOCK21]], [[SPLAT_SPLAT23]]
+; CHECK-NEXT:    [[TMP16:%.*]] = add <1 x i64> [[TMP13]], [[TMP15]]
+; CHECK-NEXT:    [[BLOCK24:%.*]] = shufflevector <1 x i64> [[SPLIT6]], <1 x i64> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP17:%.*]] = extractelement <8 x i64> [[SPLIT8]], i64 6
+; CHECK-NEXT:    [[SPLAT_SPLATINSERT25:%.*]] = insertelement <1 x i64> poison, i64 [[TMP17]], i64 0
+; CHECK-NEXT:    [[SPLAT_SPLAT26:%.*]] = shufflevector <1 x i64> [[SPLAT_SPLATINSERT25]], <1 x i64> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP18:%.*]] = mul <1 x i64> [[BLOCK24]], [[SPLAT_SPLAT26]]
+; CHECK-NEXT:    [[TMP19:%.*]] = add <1 x i64> [[TMP16]], [[TMP18]]
+; CHECK-NEXT:    [[BLOCK27:%.*]] = shufflevector <1 x i64> [[SPLIT7]], <1 x i64> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP20:%.*]] = extractelement <8 x i64> [[SPLIT8]], i64 7
+; CHECK-NEXT:    [[SPLAT_SPLATINSERT28:%.*]] = insertelement <1 x i64> poison, i64 [[TMP20]], i64 0
+; CHECK-NEXT:    [[SPLAT_SPLAT29:%.*]] = shufflevector <1 x i64> [[SPLAT_SPLATINSERT28]], <1 x i64> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP21:%.*]] = mul <1 x i64> [[BLOCK27]], [[SPLAT_SPLAT29]]
+; CHECK-NEXT:    [[TMP22:%.*]] = add <1 x i64> [[TMP19]], [[TMP21]]
+; CHECK-NEXT:    [[TMP23:%.*]] = shufflevector <1 x i64> [[TMP22]], <1 x i64> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP24:%.*]] = shufflevector <1 x i64> undef, <1 x i64> [[TMP23]], <1 x i32> <i32 1>
+; CHECK-NEXT:    ret <1 x i64> [[TMP24]]
 ;
 entry:
   %c = tail call <1 x i64> @llvm.matrix.multiply.v1i64.v8i64.v8i64(<8 x i64> %a, <8 x i64> %b, i32 1, i32 8, i32 1)
   ret <1 x i64> %c
 }
 
 declare <1 x i64> @llvm.matrix.multiply.v1i64.v8i64.v8i64(<8 x i64>, <8 x i64>, i32, i32, i32)
 

 ; CHECK-NEXT:  entry:
 ; CHECK-NEXT:    [[COL_LOAD:%.*]] = load <1 x i32>, ptr [[LHS_ADDRESS:%.*]], align 4
 ; CHECK-NEXT:    [[VEC_GEP:%.*]] = getelementptr i32, ptr [[LHS_ADDRESS]], i64 4
 ; CHECK-NEXT:    [[COL_LOAD1:%.*]] = load <1 x i32>, ptr [[VEC_GEP]], align 4
 ; CHECK-NEXT:    [[VEC_GEP2:%.*]] = getelementptr i32, ptr [[LHS_ADDRESS]], i64 8
 ; CHECK-NEXT:    [[COL_LOAD3:%.*]] = load <1 x i32>, ptr [[VEC_GEP2]], align 4
 ; CHECK-NEXT:    [[VEC_GEP4:%.*]] = getelementptr i32, ptr [[LHS_ADDRESS]], i64 12
 ; CHECK-NEXT:    [[COL_LOAD5:%.*]] = load <1 x i32>, ptr [[VEC_GEP4]], align 4
+; CHECK-NEXT:    [[TMP0:%.*]] = shufflevector <1 x i32> [[COL_LOAD]], <1 x i32> [[COL_LOAD1]], <2 x i32> <i32 0, i32 1>
+; CHECK-NEXT:    [[TMP1:%.*]] = shufflevector <1 x i32> [[COL_LOAD3]], <1 x i32> [[COL_LOAD5]], <2 x i32> <i32 0, i32 1>
+; CHECK-NEXT:    [[TMP2:%.*]] = shufflevector <2 x i32> [[TMP0]], <2 x i32> [[TMP1]], <4 x i32> <i32 0, i32 1, i32 2, i32 3>
 ; CHECK-NEXT:    [[COL_LOAD6:%.*]] = load <4 x i32>, ptr [[RHS_ADDRESS:%.*]], align 4
-; CHECK-NEXT:    [[BLOCK:%.*]] = shufflevector <1 x i32> [[COL_LOAD]], <1 x i32> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP0:%.*]] = extractelement <4 x i32> [[COL_LOAD6]], i64 0
-; CHECK-NEXT:    [[SPLAT_SPLATINSERT:%.*]] = insertelement <1 x i32> poison, i32 [[TMP0]], i64 0
-; CHECK-NEXT:    [[SPLAT_SPLAT:%.*]] = shufflevector <1 x i32> [[SPLAT_SPLATINSERT]], <1 x i32> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP1:%.*]] = mul <1 x i32> [[BLOCK]], [[SPLAT_SPLAT]]
-; CHECK-NEXT:    [[BLOCK7:%.*]] = shufflevector <1 x i32> [[COL_LOAD1]], <1 x i32> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP2:%.*]] = extractelement <4 x i32> [[COL_LOAD6]], i64 1
-; CHECK-NEXT:    [[SPLAT_SPLATINSERT8:%.*]] = insertelement <1 x i32> poison, i32 [[TMP2]], i64 0
-; CHECK-NEXT:    [[SPLAT_SPLAT9:%.*]] = shufflevector <1 x i32> [[SPLAT_SPLATINSERT8]], <1 x i32> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP3:%.*]] = mul <1 x i32> [[BLOCK7]], [[SPLAT_SPLAT9]]
-; CHECK-NEXT:    [[TMP4:%.*]] = add <1 x i32> [[TMP1]], [[TMP3]]
-; CHECK-NEXT:    [[BLOCK10:%.*]] = shufflevector <1 x i32> [[COL_LOAD3]], <1 x i32> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP5:%.*]] = extractelement <4 x i32> [[COL_LOAD6]], i64 2
-; CHECK-NEXT:    [[SPLAT_SPLATINSERT11:%.*]] = insertelement <1 x i32> poison, i32 [[TMP5]], i64 0
-; CHECK-NEXT:    [[SPLAT_SPLAT12:%.*]] = shufflevector <1 x i32> [[SPLAT_SPLATINSERT11]], <1 x i32> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP6:%.*]] = mul <1 x i32> [[BLOCK10]], [[SPLAT_SPLAT12]]
-; CHECK-NEXT:    [[TMP7:%.*]] = add <1 x i32> [[TMP4]], [[TMP6]]
-; CHECK-NEXT:    [[BLOCK13:%.*]] = shufflevector <1 x i32> [[COL_LOAD5]], <1 x i32> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP8:%.*]] = extractelement <4 x i32> [[COL_LOAD6]], i64 3
-; CHECK-NEXT:    [[SPLAT_SPLATINSERT14:%.*]] = insertelement <1 x i32> poison, i32 [[TMP8]], i64 0
-; CHECK-NEXT:    [[SPLAT_SPLAT15:%.*]] = shufflevector <1 x i32> [[SPLAT_SPLATINSERT14]], <1 x i32> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP9:%.*]] = mul <1 x i32> [[BLOCK13]], [[SPLAT_SPLAT15]]
-; CHECK-NEXT:    [[TMP10:%.*]] = add <1 x i32> [[TMP7]], [[TMP9]]
-; CHECK-NEXT:    [[TMP11:%.*]] = shufflevector <1 x i32> [[TMP10]], <1 x i32> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP12:%.*]] = shufflevector <1 x i32> undef, <1 x i32> [[TMP11]], <1 x i32> <i32 1>
-; CHECK-NEXT:    ret <1 x i32> [[TMP12]]
+; CHECK-NEXT:    [[TMP3:%.*]] = mul <4 x i32> [[TMP2]], [[COL_LOAD6]]
+; CHECK-NEXT:    [[TMP4:%.*]] = call i32 @llvm.vector.reduce.add.v4i32(<4 x i32> [[TMP3]])
+; CHECK-NEXT:    [[TMP5:%.*]] = insertelement <1 x i32> poison, i32 [[TMP4]], i64 0
+; CHECK-NEXT:    ret <1 x i32> [[TMP5]]
 ;
 entry:
   %lhs = tail call <4 x i32> @llvm.matrix.column.major.load.v4i32.i64(ptr nonnull align 4 %lhs_address, i64 4, i1 false, i32 1, i32 4)
   %rhs = tail call <4 x i32> @llvm.matrix.column.major.load.v4i32.i64(ptr nonnull align 4 %rhs_address, i64 4, i1 false, i32 4, i32 1)
   %result = tail call <1 x i32> @llvm.matrix.multiply.v1i32.v4i32.v4i32(<4 x i32> %lhs, <4 x i32> %rhs, i32 1, i32 4, i32 1)
   ret <1 x i32> %result
 }
 
 declare <4 x i32> @llvm.matrix.column.major.load.v4i32.i64(ptr nonnull align 4, i64, i1, i32, i32)
 declare <1 x i32> @llvm.matrix.multiply.v1i32.v4i32.v4i32(<4 x i32>, <4 x i32>, i32, i32, i32)
 
 define <1 x i32> @intrinsic_column_major_load_dot_product_i32_stride_too_large_2(ptr %lhs_address, ptr %rhs_address) {
 ; CHECK-LABEL: @intrinsic_column_major_load_dot_product_i32_stride_too_large_2(
 ; CHECK-NEXT:  entry:
 ; CHECK-NEXT:    [[COL_LOAD:%.*]] = load <1 x i32>, ptr [[LHS_ADDRESS:%.*]], align 4
 ; CHECK-NEXT:    [[VEC_GEP:%.*]] = getelementptr i32, ptr [[LHS_ADDRESS]], i64 5
 ; CHECK-NEXT:    [[COL_LOAD1:%.*]] = load <1 x i32>, ptr [[VEC_GEP]], align 4
 ; CHECK-NEXT:    [[VEC_GEP2:%.*]] = getelementptr i32, ptr [[LHS_ADDRESS]], i64 10
 ; CHECK-NEXT:    [[COL_LOAD3:%.*]] = load <1 x i32>, ptr [[VEC_GEP2]], align 4
 ; CHECK-NEXT:    [[VEC_GEP4:%.*]] = getelementptr i32, ptr [[LHS_ADDRESS]], i64 15
 ; CHECK-NEXT:    [[COL_LOAD5:%.*]] = load <1 x i32>, ptr [[VEC_GEP4]], align 4
+; CHECK-NEXT:    [[TMP0:%.*]] = shufflevector <1 x i32> [[COL_LOAD]], <1 x i32> [[COL_LOAD1]], <2 x i32> <i32 0, i32 1>
+; CHECK-NEXT:    [[TMP1:%.*]] = shufflevector <1 x i32> [[COL_LOAD3]], <1 x i32> [[COL_LOAD5]], <2 x i32> <i32 0, i32 1>
+; CHECK-NEXT:    [[TMP2:%.*]] = shufflevector <2 x i32> [[TMP0]], <2 x i32> [[TMP1]], <4 x i32> <i32 0, i32 1, i32 2, i32 3>
 ; CHECK-NEXT:    [[COL_LOAD6:%.*]] = load <4 x i32>, ptr [[RHS_ADDRESS:%.*]], align 4
-; CHECK-NEXT:    [[BLOCK:%.*]] = shufflevector <1 x i32> [[COL_LOAD]], <1 x i32> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP0:%.*]] = extractelement <4 x i32> [[COL_LOAD6]], i64 0
-; CHECK-NEXT:    [[SPLAT_SPLATINSERT:%.*]] = insertelement <1 x i32> poison, i32 [[TMP0]], i64 0
-; CHECK-NEXT:    [[SPLAT_SPLAT:%.*]] = shufflevector <1 x i32> [[SPLAT_SPLATINSERT]], <1 x i32> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP1:%.*]] = mul <1 x i32> [[BLOCK]], [[SPLAT_SPLAT]]
-; CHECK-NEXT:    [[BLOCK7:%.*]] = shufflevector <1 x i32> [[COL_LOAD1]], <1 x i32> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP2:%.*]] = extractelement <4 x i32> [[COL_LOAD6]], i64 1
-; CHECK-NEXT:    [[SPLAT_SPLATINSERT8:%.*]] = insertelement <1 x i32> poison, i32 [[TMP2]], i64 0
-; CHECK-NEXT:    [[SPLAT_SPLAT9:%.*]] = shufflevector <1 x i32> [[SPLAT_SPLATINSERT8]], <1 x i32> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP3:%.*]] = mul <1 x i32> [[BLOCK7]], [[SPLAT_SPLAT9]]
-; CHECK-NEXT:    [[TMP4:%.*]] = add <1 x i32> [[TMP1]], [[TMP3]]
-; CHECK-NEXT:    [[BLOCK10:%.*]] = shufflevector <1 x i32> [[COL_LOAD3]], <1 x i32> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP5:%.*]] = extractelement <4 x i32> [[COL_LOAD6]], i64 2
-; CHECK-NEXT:    [[SPLAT_SPLATINSERT11:%.*]] = insertelement <1 x i32> poison, i32 [[TMP5]], i64 0
-; CHECK-NEXT:    [[SPLAT_SPLAT12:%.*]] = shufflevector <1 x i32> [[SPLAT_SPLATINSERT11]], <1 x i32> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP6:%.*]] = mul <1 x i32> [[BLOCK10]], [[SPLAT_SPLAT12]]
-; CHECK-NEXT:    [[TMP7:%.*]] = add <1 x i32> [[TMP4]], [[TMP6]]
-; CHECK-NEXT:    [[BLOCK13:%.*]] = shufflevector <1 x i32> [[COL_LOAD5]], <1 x i32> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP8:%.*]] = extractelement <4 x i32> [[COL_LOAD6]], i64 3
-; CHECK-NEXT:    [[SPLAT_SPLATINSERT14:%.*]] = insertelement <1 x i32> poison, i32 [[TMP8]], i64 0
-; CHECK-NEXT:    [[SPLAT_SPLAT15:%.*]] = shufflevector <1 x i32> [[SPLAT_SPLATINSERT14]], <1 x i32> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP9:%.*]] = mul <1 x i32> [[BLOCK13]], [[SPLAT_SPLAT15]]
-; CHECK-NEXT:    [[TMP10:%.*]] = add <1 x i32> [[TMP7]], [[TMP9]]
-; CHECK-NEXT:    [[TMP11:%.*]] = shufflevector <1 x i32> [[TMP10]], <1 x i32> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP12:%.*]] = shufflevector <1 x i32> undef, <1 x i32> [[TMP11]], <1 x i32> <i32 1>
-; CHECK-NEXT:    ret <1 x i32> [[TMP12]]
+; CHECK-NEXT:    [[TMP3:%.*]] = mul <4 x i32> [[TMP2]], [[COL_LOAD6]]
+; CHECK-NEXT:    [[TMP4:%.*]] = call i32 @llvm.vector.reduce.add.v4i32(<4 x i32> [[TMP3]])
+; CHECK-NEXT:    [[TMP5:%.*]] = insertelement <1 x i32> poison, i32 [[TMP4]], i64 0
+; CHECK-NEXT:    ret <1 x i32> [[TMP5]]
 ;
 entry:
   %lhs = tail call <4 x i32> @llvm.matrix.column.major.load.v4i32.i64(ptr nonnull align 4 %lhs_address, i64 5, i1 false, i32 1, i32 4)
   %rhs = tail call <4 x i32> @llvm.matrix.column.major.load.v4i32.i64(ptr nonnull align 4 %rhs_address, i64 4, i1 false, i32 4, i32 1)
   %result = tail call <1 x i32> @llvm.matrix.multiply.v1i32.v4i32.v4i32(<4 x i32> %lhs, <4 x i32> %rhs, i32 1, i32 4, i32 1)
   ret <1 x i32> %result
 }
 

 ; CHECK-NEXT:    [[BLOCK3:%.*]] = shufflevector <2 x i32> [[SPLIT]], <2 x i32> poison, <2 x i32> <i32 0, i32 1>
 ; CHECK-NEXT:    [[TMP5:%.*]] = mul <2 x i32> [[BLOCK3]], zeroinitializer
 ; CHECK-NEXT:    [[BLOCK4:%.*]] = shufflevector <2 x i32> [[SPLIT1]], <2 x i32> poison, <2 x i32> <i32 0, i32 1>
 ; CHECK-NEXT:    [[TMP6:%.*]] = mul <2 x i32> [[BLOCK4]], zeroinitializer
 ; CHECK-NEXT:    [[TMP7:%.*]] = add <2 x i32> [[TMP5]], [[TMP6]]
 ; CHECK-NEXT:    [[TMP8:%.*]] = shufflevector <2 x i32> [[TMP7]], <2 x i32> poison, <2 x i32> <i32 0, i32 1>
 ; CHECK-NEXT:    [[TMP9:%.*]] = shufflevector <2 x i32> undef, <2 x i32> [[TMP8]], <2 x i32> <i32 2, i32 3>
 ; CHECK-NEXT:    [[TMP10:%.*]] = shufflevector <2 x i32> [[TMP4]], <2 x i32> [[TMP9]], <4 x i32> <i32 0, i32 1, i32 2, i32 3>
-; CHECK-NEXT:    [[SPLIT5:%.*]] = shufflevector <4 x i32> [[TMP10]], <4 x i32> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[SPLIT6:%.*]] = shufflevector <4 x i32> [[TMP10]], <4 x i32> poison, <1 x i32> <i32 1>
-; CHECK-NEXT:    [[SPLIT7:%.*]] = shufflevector <4 x i32> [[TMP10]], <4 x i32> poison, <1 x i32> <i32 2>
-; CHECK-NEXT:    [[SPLIT8:%.*]] = shufflevector <4 x i32> [[TMP10]], <4 x i32> poison, <1 x i32> <i32 3>
-; CHECK-NEXT:    [[BLOCK9:%.*]] = shufflevector <1 x i32> [[SPLIT5]], <1 x i32> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP11:%.*]] = mul <1 x i32> [[BLOCK9]], zeroinitializer
-; CHECK-NEXT:    [[BLOCK10:%.*]] = shufflevector <1 x i32> [[SPLIT6]], <1 x i32> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP12:%.*]] = mul <1 x i32> [[BLOCK10]], zeroinitializer
-; CHECK-NEXT:    [[TMP13:%.*]] = add <1 x i32> [[TMP11]], [[TMP12]]
-; CHECK-NEXT:    [[BLOCK11:%.*]] = shufflevector <1 x i32> [[SPLIT7]], <1 x i32> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP14:%.*]] = mul <1 x i32> [[BLOCK11]], zeroinitializer
-; CHECK-NEXT:    [[TMP15:%.*]] = add <1 x i32> [[TMP13]], [[TMP14]]
-; CHECK-NEXT:    [[BLOCK12:%.*]] = shufflevector <1 x i32> [[SPLIT8]], <1 x i32> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP16:%.*]] = mul <1 x i32> [[BLOCK12]], zeroinitializer
-; CHECK-NEXT:    [[TMP17:%.*]] = add <1 x i32> [[TMP15]], [[TMP16]]
-; CHECK-NEXT:    [[TMP18:%.*]] = shufflevector <1 x i32> [[TMP17]], <1 x i32> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP19:%.*]] = shufflevector <1 x i32> undef, <1 x i32> [[TMP18]], <1 x i32> <i32 1>
+; CHECK-NEXT:    [[TMP11:%.*]] = mul <4 x i32> [[TMP10]], zeroinitializer
+; CHECK-NEXT:    [[TMP12:%.*]] = call i32 @llvm.vector.reduce.add.v4i32(<4 x i32> [[TMP11]])
+; CHECK-NEXT:    [[TMP13:%.*]] = insertelement <1 x i32> poison, i32 [[TMP12]], i64 0
 ; CHECK-NEXT:    ret void
 ;
 entry:
   %scalar.splat.splat.i.i.i.i = shufflevector <4 x i32> zeroinitializer, <4 x i32> zeroinitializer, <4 x i32> zeroinitializer
   %0 = call <4 x i32> @llvm.matrix.multiply.v4i32.v4i32.v4i32(<4 x i32> zeroinitializer, <4 x i32> %scalar.splat.splat.i.i.i.i, i32 2, i32 2, i32 2)
   %1 = call <4 x i32> @llvm.matrix.transpose.v4i32(<4 x i32> %0, i32 4, i32 1)
   %2 = call <1 x i32> @llvm.matrix.multiply.v1i32.v4i32.v4i32(<4 x i32> %1, <4 x i32> zeroinitializer, i32 1, i32 4, i32 1)
   ret void
 }
 
 declare <4 x i32> @llvm.matrix.transpose.v4i32(<4 x i32>, i32 immarg, i32 immarg)
 
 declare <4 x i32> @llvm.matrix.multiply.v4i32.v4i32.v4i32(<4 x i32>, <4 x i32>, i32 immarg, i32 immarg, i32 immarg)
 
 define <1 x i32> @test_load_multiuse(ptr %src, <4 x i32> %b) {
 ; CHECK-LABEL: @test_load_multiuse(
 ; CHECK-NEXT:    [[COL_LOAD:%.*]] = load <1 x i32>, ptr [[SRC:%.*]], align 16
 ; CHECK-NEXT:    [[VEC_GEP:%.*]] = getelementptr i32, ptr [[SRC]], i64 1
 ; CHECK-NEXT:    [[COL_LOAD1:%.*]] = load <1 x i32>, ptr [[VEC_GEP]], align 4
 ; CHECK-NEXT:    [[VEC_GEP2:%.*]] = getelementptr i32, ptr [[SRC]], i64 2
 ; CHECK-NEXT:    [[COL_LOAD3:%.*]] = load <1 x i32>, ptr [[VEC_GEP2]], align 8
 ; CHECK-NEXT:    [[VEC_GEP4:%.*]] = getelementptr i32, ptr [[SRC]], i64 3
 ; CHECK-NEXT:    [[COL_LOAD5:%.*]] = load <1 x i32>, ptr [[VEC_GEP4]], align 4
-; CHECK-NEXT:    [[TMP1:%.*]] = extractelement <1 x i32> [[COL_LOAD]], i64 0
-; CHECK-NEXT:    [[TMP2:%.*]] = insertelement <4 x i32> poison, i32 [[TMP1]], i64 0
-; CHECK-NEXT:    [[TMP3:%.*]] = extractelement <1 x i32> [[COL_LOAD1]], i64 0
-; CHECK-NEXT:    [[TMP4:%.*]] = insertelement <4 x i32> [[TMP2]], i32 [[TMP3]], i64 1
-; CHECK-NEXT:    [[TMP5:%.*]] = extractelement <1 x i32> [[COL_LOAD3]], i64 0
-; CHECK-NEXT:    [[TMP6:%.*]] = insertelement <4 x i32> [[TMP4]], i32 [[TMP5]], i64 2
-; CHECK-NEXT:    [[TMP7:%.*]] = extractelement <1 x i32> [[COL_LOAD5]], i64 0
-; CHECK-NEXT:    [[TMP8:%.*]] = insertelement <4 x i32> [[TMP6]], i32 [[TMP7]], i64 3
-; CHECK-NEXT:    call void @use.v4i32(<4 x i32> [[TMP8]])
-; CHECK-NEXT:    [[SPLIT:%.*]] = shufflevector <4 x i32> [[B:%.*]], <4 x i32> poison, <4 x i32> <i32 0, i32 1, i32 2, i32 3>
-; CHECK-NEXT:    [[BLOCK:%.*]] = shufflevector <1 x i32> [[COL_LOAD]], <1 x i32> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP9:%.*]] = extractelement <4 x i32> [[SPLIT]], i64 0
-; CHECK-NEXT:    [[SPLAT_SPLATINSERT:%.*]] = insertelement <1 x i32> poison, i32 [[TMP9]], i64 0
-; CHECK-NEXT:    [[SPLAT_SPLAT:%.*]] = shufflevector <1 x i32> [[SPLAT_SPLATINSERT]], <1 x i32> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP10:%.*]] = mul <1 x i32> [[BLOCK]], [[SPLAT_SPLAT]]
-; CHECK-NEXT:    [[BLOCK6:%.*]] = shufflevector <1 x i32> [[COL_LOAD1]], <1 x i32> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP11:%.*]] = extractelement <4 x i32> [[SPLIT]], i64 1
-; CHECK-NEXT:    [[SPLAT_SPLATINSERT7:%.*]] = insertelement <1 x i32> poison, i32 [[TMP11]], i64 0
-; CHECK-NEXT:    [[SPLAT_SPLAT8:%.*]] = shufflevector <1 x i32> [[SPLAT_SPLATINSERT7]], <1 x i32> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP12:%.*]] = mul <1 x i32> [[BLOCK6]], [[SPLAT_SPLAT8]]
-; CHECK-NEXT:    [[TMP13:%.*]] = add <1 x i32> [[TMP10]], [[TMP12]]
-; CHECK-NEXT:    [[BLOCK9:%.*]] = shufflevector <1 x i32> [[COL_LOAD3]], <1 x i32> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP14:%.*]] = extractelement <4 x i32> [[SPLIT]], i64 2
-; CHECK-NEXT:    [[SPLAT_SPLATINSERT10:%.*]] = insertelement <1 x i32> poison, i32 [[TMP14]], i64 0
-; CHECK-NEXT:    [[SPLAT_SPLAT11:%.*]] = shufflevector <1 x i32> [[SPLAT_SPLATINSERT10]], <1 x i32> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP15:%.*]] = mul <1 x i32> [[BLOCK9]], [[SPLAT_SPLAT11]]
-; CHECK-NEXT:    [[TMP16:%.*]] = add <1 x i32> [[TMP13]], [[TMP15]]
-; CHECK-NEXT:    [[BLOCK12:%.*]] = shufflevector <1 x i32> [[COL_LOAD5]], <1 x i32> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP17:%.*]] = extractelement <4 x i32> [[SPLIT]], i64 3
-; CHECK-NEXT:    [[SPLAT_SPLATINSERT13:%.*]] = insertelement <1 x i32> poison, i32 [[TMP17]], i64 0
-; CHECK-NEXT:    [[SPLAT_SPLAT14:%.*]] = shufflevector <1 x i32> [[SPLAT_SPLATINSERT13]], <1 x i32> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP18:%.*]] = mul <1 x i32> [[BLOCK12]], [[SPLAT_SPLAT14]]
-; CHECK-NEXT:    [[TMP19:%.*]] = add <1 x i32> [[TMP16]], [[TMP18]]
-; CHECK-NEXT:    [[TMP20:%.*]] = shufflevector <1 x i32> [[TMP19]], <1 x i32> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP21:%.*]] = shufflevector <1 x i32> undef, <1 x i32> [[TMP20]], <1 x i32> <i32 1>
-; CHECK-NEXT:    ret <1 x i32> [[TMP21]]
+; CHECK-NEXT:    [[TMP1:%.*]] = shufflevector <1 x i32> [[COL_LOAD]], <1 x i32> [[COL_LOAD1]], <2 x i32> <i32 0, i32 1>
+; CHECK-NEXT:    [[TMP2:%.*]] = shufflevector <1 x i32> [[COL_LOAD3]], <1 x i32> [[COL_LOAD5]], <2 x i32> <i32 0, i32 1>
+; CHECK-NEXT:    [[TMP3:%.*]] = shufflevector <2 x i32> [[TMP1]], <2 x i32> [[TMP2]], <4 x i32> <i32 0, i32 1, i32 2, i32 3>
+; CHECK-NEXT:    [[TMP4:%.*]] = extractelement <1 x i32> [[COL_LOAD]], i64 0
+; CHECK-NEXT:    [[TMP5:%.*]] = insertelement <4 x i32> poison, i32 [[TMP4]], i64 0
+; CHECK-NEXT:    [[TMP6:%.*]] = extractelement <1 x i32> [[COL_LOAD1]], i64 0
+; CHECK-NEXT:    [[TMP7:%.*]] = insertelement <4 x i32> [[TMP5]], i32 [[TMP6]], i64 1
+; CHECK-NEXT:    [[TMP8:%.*]] = extractelement <1 x i32> [[COL_LOAD3]], i64 0
+; CHECK-NEXT:    [[TMP9:%.*]] = insertelement <4 x i32> [[TMP7]], i32 [[TMP8]], i64 2
+; CHECK-NEXT:    [[TMP10:%.*]] = extractelement <1 x i32> [[COL_LOAD5]], i64 0
+; CHECK-NEXT:    [[TMP11:%.*]] = insertelement <4 x i32> [[TMP9]], i32 [[TMP10]], i64 3
+; CHECK-NEXT:    call void @use.v4i32(<4 x i32> [[TMP11]])
+; CHECK-NEXT:    [[TMP12:%.*]] = mul <4 x i32> [[TMP3]], [[B:%.*]]
+; CHECK-NEXT:    [[TMP13:%.*]] = call i32 @llvm.vector.reduce.add.v4i32(<4 x i32> [[TMP12]])
+; CHECK-NEXT:    [[TMP14:%.*]] = insertelement <1 x i32> poison, i32 [[TMP13]], i64 0
+; CHECK-NEXT:    ret <1 x i32> [[TMP14]]
 ;
   %l = load <4 x i32>, ptr %src
   %t = call <4 x i32> @llvm.matrix.transpose.v4i32(<4 x i32> %l, i32 1, i32 4)
   call void @use.v4i32(<4 x i32> %t)
   %res = call <1 x i32> @llvm.matrix.multiply.v1i32.v4i32.v4i32(<4 x i32> %l, <4 x i32> %b, i32 1, i32 4, i32 1)
   ret <1 x i32> %res
 }
 define <1 x i32> @test_builtin_column_major_load_multiuse(ptr %src, <4 x i32> %b) {
 ; CHECK-LABEL: @test_builtin_column_major_load_multiuse(
 ; CHECK-NEXT:    [[COL_LOAD:%.*]] = load <1 x i32>, ptr [[SRC:%.*]], align 4
 ; CHECK-NEXT:    [[VEC_GEP:%.*]] = getelementptr i32, ptr [[SRC]], i64 1
 ; CHECK-NEXT:    [[COL_LOAD1:%.*]] = load <1 x i32>, ptr [[VEC_GEP]], align 4
 ; CHECK-NEXT:    [[VEC_GEP2:%.*]] = getelementptr i32, ptr [[SRC]], i64 2
 ; CHECK-NEXT:    [[COL_LOAD3:%.*]] = load <1 x i32>, ptr [[VEC_GEP2]], align 4
 ; CHECK-NEXT:    [[VEC_GEP4:%.*]] = getelementptr i32, ptr [[SRC]], i64 3
 ; CHECK-NEXT:    [[COL_LOAD5:%.*]] = load <1 x i32>, ptr [[VEC_GEP4]], align 4
-; CHECK-NEXT:    [[TMP1:%.*]] = extractelement <1 x i32> [[COL_LOAD]], i64 0
-; CHECK-NEXT:    [[TMP2:%.*]] = insertelement <4 x i32> poison, i32 [[TMP1]], i64 0
-; CHECK-NEXT:    [[TMP3:%.*]] = extractelement <1 x i32> [[COL_LOAD1]], i64 0
-; CHECK-NEXT:    [[TMP4:%.*]] = insertelement <4 x i32> [[TMP2]], i32 [[TMP3]], i64 1
-; CHECK-NEXT:    [[TMP5:%.*]] = extractelement <1 x i32> [[COL_LOAD3]], i64 0
-; CHECK-NEXT:    [[TMP6:%.*]] = insertelement <4 x i32> [[TMP4]], i32 [[TMP5]], i64 2
-; CHECK-NEXT:    [[TMP7:%.*]] = extractelement <1 x i32> [[COL_LOAD5]], i64 0
-; CHECK-NEXT:    [[TMP8:%.*]] = insertelement <4 x i32> [[TMP6]], i32 [[TMP7]], i64 3
-; CHECK-NEXT:    call void @use.v4i32(<4 x i32> [[TMP8]])
-; CHECK-NEXT:    [[SPLIT:%.*]] = shufflevector <4 x i32> [[B:%.*]], <4 x i32> poison, <4 x i32> <i32 0, i32 1, i32 2, i32 3>
-; CHECK-NEXT:    [[BLOCK:%.*]] = shufflevector <1 x i32> [[COL_LOAD]], <1 x i32> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP9:%.*]] = extractelement <4 x i32> [[SPLIT]], i64 0
-; CHECK-NEXT:    [[SPLAT_SPLATINSERT:%.*]] = insertelement <1 x i32> poison, i32 [[TMP9]], i64 0
-; CHECK-NEXT:    [[SPLAT_SPLAT:%.*]] = shufflevector <1 x i32> [[SPLAT_SPLATINSERT]], <1 x i32> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP10:%.*]] = mul <1 x i32> [[BLOCK]], [[SPLAT_SPLAT]]
-; CHECK-NEXT:    [[BLOCK6:%.*]] = shufflevector <1 x i32> [[COL_LOAD1]], <1 x i32> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP11:%.*]] = extractelement <4 x i32> [[SPLIT]], i64 1
-; CHECK-NEXT:    [[SPLAT_SPLATINSERT7:%.*]] = insertelement <1 x i32> poison, i32 [[TMP11]], i64 0
-; CHECK-NEXT:    [[SPLAT_SPLAT8:%.*]] = shufflevector <1 x i32> [[SPLAT_SPLATINSERT7]], <1 x i32> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP12:%.*]] = mul <1 x i32> [[BLOCK6]], [[SPLAT_SPLAT8]]
-; CHECK-NEXT:    [[TMP13:%.*]] = add <1 x i32> [[TMP10]], [[TMP12]]
-; CHECK-NEXT:    [[BLOCK9:%.*]] = shufflevector <1 x i32> [[COL_LOAD3]], <1 x i32> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP14:%.*]] = extractelement <4 x i32> [[SPLIT]], i64 2
-; CHECK-NEXT:    [[SPLAT_SPLATINSERT10:%.*]] = insertelement <1 x i32> poison, i32 [[TMP14]], i64 0
-; CHECK-NEXT:    [[SPLAT_SPLAT11:%.*]] = shufflevector <1 x i32> [[SPLAT_SPLATINSERT10]], <1 x i32> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP15:%.*]] = mul <1 x i32> [[BLOCK9]], [[SPLAT_SPLAT11]]
-; CHECK-NEXT:    [[TMP16:%.*]] = add <1 x i32> [[TMP13]], [[TMP15]]
-; CHECK-NEXT:    [[BLOCK12:%.*]] = shufflevector <1 x i32> [[COL_LOAD5]], <1 x i32> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP17:%.*]] = extractelement <4 x i32> [[SPLIT]], i64 3
-; CHECK-NEXT:    [[SPLAT_SPLATINSERT13:%.*]] = insertelement <1 x i32> poison, i32 [[TMP17]], i64 0
-; CHECK-NEXT:    [[SPLAT_SPLAT14:%.*]] = shufflevector <1 x i32> [[SPLAT_SPLATINSERT13]], <1 x i32> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP18:%.*]] = mul <1 x i32> [[BLOCK12]], [[SPLAT_SPLAT14]]
-; CHECK-NEXT:    [[TMP19:%.*]] = add <1 x i32> [[TMP16]], [[TMP18]]
-; CHECK-NEXT:    [[TMP20:%.*]] = shufflevector <1 x i32> [[TMP19]], <1 x i32> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP21:%.*]] = shufflevector <1 x i32> undef, <1 x i32> [[TMP20]], <1 x i32> <i32 1>
-; CHECK-NEXT:    ret <1 x i32> [[TMP21]]
+; CHECK-NEXT:    [[TMP1:%.*]] = shufflevector <1 x i32> [[COL_LOAD]], <1 x i32> [[COL_LOAD1]], <2 x i32> <i32 0, i32 1>
+; CHECK-NEXT:    [[TMP2:%.*]] = shufflevector <1 x i32> [[COL_LOAD3]], <1 x i32> [[COL_LOAD5]], <2 x i32> <i32 0, i32 1>
+; CHECK-NEXT:    [[TMP3:%.*]] = shufflevector <2 x i32> [[TMP1]], <2 x i32> [[TMP2]], <4 x i32> <i32 0, i32 1, i32 2, i32 3>
+; CHECK-NEXT:    [[TMP4:%.*]] = extractelement <1 x i32> [[COL_LOAD]], i64 0
+; CHECK-NEXT:    [[TMP5:%.*]] = insertelement <4 x i32> poison, i32 [[TMP4]], i64 0
+; CHECK-NEXT:    [[TMP6:%.*]] = extractelement <1 x i32> [[COL_LOAD1]], i64 0
+; CHECK-NEXT:    [[TMP7:%.*]] = insertelement <4 x i32> [[TMP5]], i32 [[TMP6]], i64 1
+; CHECK-NEXT:    [[TMP8:%.*]] = extractelement <1 x i32> [[COL_LOAD3]], i64 0
+; CHECK-NEXT:    [[TMP9:%.*]] = insertelement <4 x i32> [[TMP7]], i32 [[TMP8]], i64 2
+; CHECK-NEXT:    [[TMP10:%.*]] = extractelement <1 x i32> [[COL_LOAD5]], i64 0
+; CHECK-NEXT:    [[TMP11:%.*]] = insertelement <4 x i32> [[TMP9]], i32 [[TMP10]], i64 3
+; CHECK-NEXT:    call void @use.v4i32(<4 x i32> [[TMP11]])
+; CHECK-NEXT:    [[TMP12:%.*]] = mul <4 x i32> [[TMP3]], [[B:%.*]]
+; CHECK-NEXT:    [[TMP13:%.*]] = call i32 @llvm.vector.reduce.add.v4i32(<4 x i32> [[TMP12]])
+; CHECK-NEXT:    [[TMP14:%.*]] = insertelement <1 x i32> poison, i32 [[TMP13]], i64 0
+; CHECK-NEXT:    ret <1 x i32> [[TMP14]]
 ;
   %l = call <4 x i32> @llvm.matrix.column.major.load.v4i32.i64(ptr %src, i64 1, i1 false, i32 1, i32 4)
   %t = call <4 x i32> @llvm.matrix.transpose.v4i32(<4 x i32> %l, i32 1, i32 4)
   call void @use.v4i32(<4 x i32> %t)
   %res = call <1 x i32> @llvm.matrix.multiply.v1i32.v4i32.v4i32(<4 x i32> %l, <4 x i32> %b, i32 1, i32 4, i32 1)
   ret <1 x i32> %res
 }
 
 declare void @use.v4i32(<4 x i32>)
 
 define <1 x i32> @test_builtin_column_major_variable_stride(ptr %src, <4 x i32> %a, i64 %stride) {
 ; CHECK-LABEL: @test_builtin_column_major_variable_stride(
 ; CHECK-NEXT:    [[VEC_START:%.*]] = mul i64 0, [[STRIDE:%.*]]
 ; CHECK-NEXT:    [[VEC_GEP:%.*]] = getelementptr i32, ptr [[SRC:%.*]], i64 [[VEC_START]]
 ; CHECK-NEXT:    [[COL_LOAD:%.*]] = load <1 x i32>, ptr [[VEC_GEP]], align 4
 ; CHECK-NEXT:    [[VEC_START1:%.*]] = mul i64 1, [[STRIDE]]
 ; CHECK-NEXT:    [[VEC_GEP2:%.*]] = getelementptr i32, ptr [[SRC]], i64 [[VEC_START1]]
 ; CHECK-NEXT:    [[COL_LOAD3:%.*]] = load <1 x i32>, ptr [[VEC_GEP2]], align 4
 ; CHECK-NEXT:    [[VEC_START4:%.*]] = mul i64 2, [[STRIDE]]
 ; CHECK-NEXT:    [[VEC_GEP5:%.*]] = getelementptr i32, ptr [[SRC]], i64 [[VEC_START4]]
 ; CHECK-NEXT:    [[COL_LOAD6:%.*]] = load <1 x i32>, ptr [[VEC_GEP5]], align 4
 ; CHECK-NEXT:    [[VEC_START7:%.*]] = mul i64 3, [[STRIDE]]
 ; CHECK-NEXT:    [[VEC_GEP8:%.*]] = getelementptr i32, ptr [[SRC]], i64 [[VEC_START7]]
 ; CHECK-NEXT:    [[COL_LOAD9:%.*]] = load <1 x i32>, ptr [[VEC_GEP8]], align 4
-; CHECK-NEXT:    [[SPLIT:%.*]] = shufflevector <4 x i32> [[A:%.*]], <4 x i32> poison, <4 x i32> <i32 0, i32 1, i32 2, i32 3>
-; CHECK-NEXT:    [[BLOCK:%.*]] = shufflevector <1 x i32> [[COL_LOAD]], <1 x i32> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP1:%.*]] = extractelement <4 x i32> [[SPLIT]], i64 0
-; CHECK-NEXT:    [[SPLAT_SPLATINSERT:%.*]] = insertelement <1 x i32> poison, i32 [[TMP1]], i64 0
-; CHECK-NEXT:    [[SPLAT_SPLAT:%.*]] = shufflevector <1 x i32> [[SPLAT_SPLATINSERT]], <1 x i32> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP2:%.*]] = mul <1 x i32> [[BLOCK]], [[SPLAT_SPLAT]]
-; CHECK-NEXT:    [[BLOCK10:%.*]] = shufflevector <1 x i32> [[COL_LOAD3]], <1 x i32> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP3:%.*]] = extractelement <4 x i32> [[SPLIT]], i64 1
-; CHECK-NEXT:    [[SPLAT_SPLATINSERT11:%.*]] = insertelement <1 x i32> poison, i32 [[TMP3]], i64 0
-; CHECK-NEXT:    [[SPLAT_SPLAT12:%.*]] = shufflevector <1 x i32> [[SPLAT_SPLATINSERT11]], <1 x i32> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP4:%.*]] = mul <1 x i32> [[BLOCK10]], [[SPLAT_SPLAT12]]
-; CHECK-NEXT:    [[TMP5:%.*]] = add <1 x i32> [[TMP2]], [[TMP4]]
-; CHECK-NEXT:    [[BLOCK13:%.*]] = shufflevector <1 x i32> [[COL_LOAD6]], <1 x i32> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP6:%.*]] = extractelement <4 x i32> [[SPLIT]], i64 2
-; CHECK-NEXT:    [[SPLAT_SPLATINSERT14:%.*]] = insertelement <1 x i32> poison, i32 [[TMP6]], i64 0
-; CHECK-NEXT:    [[SPLAT_SPLAT15:%.*]] = shufflevector <1 x i32> [[SPLAT_SPLATINSERT14]], <1 x i32> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP7:%.*]] = mul <1 x i32> [[BLOCK13]], [[SPLAT_SPLAT15]]
-; CHECK-NEXT:    [[TMP8:%.*]] = add <1 x i32> [[TMP5]], [[TMP7]]
-; CHECK-NEXT:    [[BLOCK16:%.*]] = shufflevector <1 x i32> [[COL_LOAD9]], <1 x i32> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP9:%.*]] = extractelement <4 x i32> [[SPLIT]], i64 3
-; CHECK-NEXT:    [[SPLAT_SPLATINSERT17:%.*]] = insertelement <1 x i32> poison, i32 [[TMP9]], i64 0
-; CHECK-NEXT:    [[SPLAT_SPLAT18:%.*]] = shufflevector <1 x i32> [[SPLAT_SPLATINSERT17]], <1 x i32> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP10:%.*]] = mul <1 x i32> [[BLOCK16]], [[SPLAT_SPLAT18]]
-; CHECK-NEXT:    [[TMP11:%.*]] = add <1 x i32> [[TMP8]], [[TMP10]]
-; CHECK-NEXT:    [[TMP12:%.*]] = shufflevector <1 x i32> [[TMP11]], <1 x i32> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP13:%.*]] = shufflevector <1 x i32> undef, <1 x i32> [[TMP12]], <1 x i32> <i32 1>
-; CHECK-NEXT:    ret <1 x i32> [[TMP13]]
+; CHECK-NEXT:    [[TMP1:%.*]] = shufflevector <1 x i32> [[COL_LOAD]], <1 x i32> [[COL_LOAD3]], <2 x i32> <i32 0, i32 1>
+; CHECK-NEXT:    [[TMP2:%.*]] = shufflevector <1 x i32> [[COL_LOAD6]], <1 x i32> [[COL_LOAD9]], <2 x i32> <i32 0, i32 1>
+; CHECK-NEXT:    [[TMP3:%.*]] = shufflevector <2 x i32> [[TMP1]], <2 x i32> [[TMP2]], <4 x i32> <i32 0, i32 1, i32 2, i32 3>
+; CHECK-NEXT:    [[TMP4:%.*]] = mul <4 x i32> [[TMP3]], [[A:%.*]]
+; CHECK-NEXT:    [[TMP5:%.*]] = call i32 @llvm.vector.reduce.add.v4i32(<4 x i32> [[TMP4]])
+; CHECK-NEXT:    [[TMP6:%.*]] = insertelement <1 x i32> poison, i32 [[TMP5]], i64 0
+; CHECK-NEXT:    ret <1 x i32> [[TMP6]]
 ;
   %l = call <4 x i32> @llvm.matrix.column.major.load.v4i32.i64(ptr %src, i64 %stride, i1 false, i32 1, i32 4)
   %r = call <1 x i32> @llvm.matrix.multiply.v1i32.v4i32.v4i32(<4 x i32> %l, <4 x i32> %a, i32 1, i32 4, i32 1)
   ret <1 x i32> %r
 }
 
 define <1 x i32> @test_builtin_column_major_variable_stride_multiuse(ptr %src, <5 x i32> %a, i64 %stride) {
 ; CHECK-LABEL: @test_builtin_column_major_variable_stride_multiuse(
 ; CHECK-NEXT:    [[VEC_START:%.*]] = mul i64 0, [[STRIDE:%.*]]
 ; CHECK-NEXT:    [[VEC_GEP:%.*]] = getelementptr i32, ptr [[SRC:%.*]], i64 [[VEC_START]]
 ; CHECK-NEXT:    [[COL_LOAD:%.*]] = load <1 x i32>, ptr [[VEC_GEP]], align 4
 ; CHECK-NEXT:    [[VEC_START1:%.*]] = mul i64 1, [[STRIDE]]
 ; CHECK-NEXT:    [[VEC_GEP2:%.*]] = getelementptr i32, ptr [[SRC]], i64 [[VEC_START1]]
 ; CHECK-NEXT:    [[COL_LOAD3:%.*]] = load <1 x i32>, ptr [[VEC_GEP2]], align 4
 ; CHECK-NEXT:    [[VEC_START4:%.*]] = mul i64 2, [[STRIDE]]
 ; CHECK-NEXT:    [[VEC_GEP5:%.*]] = getelementptr i32, ptr [[SRC]], i64 [[VEC_START4]]
 ; CHECK-NEXT:    [[COL_LOAD6:%.*]] = load <1 x i32>, ptr [[VEC_GEP5]], align 4
 ; CHECK-NEXT:    [[VEC_START7:%.*]] = mul i64 3, [[STRIDE]]
 ; CHECK-NEXT:    [[VEC_GEP8:%.*]] = getelementptr i32, ptr [[SRC]], i64 [[VEC_START7]]
 ; CHECK-NEXT:    [[COL_LOAD9:%.*]] = load <1 x i32>, ptr [[VEC_GEP8]], align 4
 ; CHECK-NEXT:    [[VEC_START10:%.*]] = mul i64 4, [[STRIDE]]
 ; CHECK-NEXT:    [[VEC_GEP11:%.*]] = getelementptr i32, ptr [[SRC]], i64 [[VEC_START10]]
 ; CHECK-NEXT:    [[COL_LOAD12:%.*]] = load <1 x i32>, ptr [[VEC_GEP11]], align 4
-; CHECK-NEXT:    [[TMP1:%.*]] = extractelement <1 x i32> [[COL_LOAD]], i64 0
-; CHECK-NEXT:    [[TMP2:%.*]] = insertelement <5 x i32> poison, i32 [[TMP1]], i64 0
-; CHECK-NEXT:    [[TMP3:%.*]] = extractelement <1 x i32> [[COL_LOAD3]], i64 0
-; CHECK-NEXT:    [[TMP4:%.*]] = insertelement <5 x i32> [[TMP2]], i32 [[TMP3]], i64 1
-; CHECK-NEXT:    [[TMP5:%.*]] = extractelement <1 x i32> [[COL_LOAD6]], i64 0
-; CHECK-NEXT:    [[TMP6:%.*]] = insertelement <5 x i32> [[TMP4]], i32 [[TMP5]], i64 2
-; CHECK-NEXT:    [[TMP7:%.*]] = extractelement <1 x i32> [[COL_LOAD9]], i64 0
-; CHECK-NEXT:    [[TMP8:%.*]] = insertelement <5 x i32> [[TMP6]], i32 [[TMP7]], i64 3
-; CHECK-NEXT:    [[TMP9:%.*]] = extractelement <1 x i32> [[COL_LOAD12]], i64 0
-; CHECK-NEXT:    [[TMP10:%.*]] = insertelement <5 x i32> [[TMP8]], i32 [[TMP9]], i64 4
-; CHECK-NEXT:    call void @use.v5i32(<5 x i32> [[TMP10]])
-; CHECK-NEXT:    [[SPLIT:%.*]] = shufflevector <5 x i32> [[A:%.*]], <5 x i32> poison, <5 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4>
-; CHECK-NEXT:    [[BLOCK:%.*]] = shufflevector <1 x i32> [[COL_LOAD]], <1 x i32> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP11:%.*]] = extractelement <5 x i32> [[SPLIT]], i64 0
-; CHECK-NEXT:    [[SPLAT_SPLATINSERT:%.*]] = insertelement <1 x i32> poison, i32 [[TMP11]], i64 0
-; CHECK-NEXT:    [[SPLAT_SPLAT:%.*]] = shufflevector <1 x i32> [[SPLAT_SPLATINSERT]], <1 x i32> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP12:%.*]] = mul <1 x i32> [[BLOCK]], [[SPLAT_SPLAT]]
-; CHECK-NEXT:    [[BLOCK13:%.*]] = shufflevector <1 x i32> [[COL_LOAD3]], <1 x i32> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP13:%.*]] = extractelement <5 x i32> [[SPLIT]], i64 1
-; CHECK-NEXT:    [[SPLAT_SPLATINSERT14:%.*]] = insertelement <1 x i32> poison, i32 [[TMP13]], i64 0
-; CHECK-NEXT:    [[SPLAT_SPLAT15:%.*]] = shufflevector <1 x i32> [[SPLAT_SPLATINSERT14]], <1 x i32> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP14:%.*]] = mul <1 x i32> [[BLOCK13]], [[SPLAT_SPLAT15]]
-; CHECK-NEXT:    [[TMP15:%.*]] = add <1 x i32> [[TMP12]], [[TMP14]]
-; CHECK-NEXT:    [[BLOCK16:%.*]] = shufflevector <1 x i32> [[COL_LOAD6]], <1 x i32> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP16:%.*]] = extractelement <5 x i32> [[SPLIT]], i64 2
-; CHECK-NEXT:    [[SPLAT_SPLATINSERT17:%.*]] = insertelement <1 x i32> poison, i32 [[TMP16]], i64 0
-; CHECK-NEXT:    [[SPLAT_SPLAT18:%.*]] = shufflevector <1 x i32> [[SPLAT_SPLATINSERT17]], <1 x i32> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP17:%.*]] = mul <1 x i32> [[BLOCK16]], [[SPLAT_SPLAT18]]
-; CHECK-NEXT:    [[TMP18:%.*]] = add <1 x i32> [[TMP15]], [[TMP17]]
-; CHECK-NEXT:    [[BLOCK19:%.*]] = shufflevector <1 x i32> [[COL_LOAD9]], <1 x i32> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP19:%.*]] = extractelement <5 x i32> [[SPLIT]], i64 3
-; CHECK-NEXT:    [[SPLAT_SPLATINSERT20:%.*]] = insertelement <1 x i32> poison, i32 [[TMP19]], i64 0
-; CHECK-NEXT:    [[SPLAT_SPLAT21:%.*]] = shufflevector <1 x i32> [[SPLAT_SPLATINSERT20]], <1 x i32> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP20:%.*]] = mul <1 x i32> [[BLOCK19]], [[SPLAT_SPLAT21]]
-; CHECK-NEXT:    [[TMP21:%.*]] = add <1 x i32> [[TMP18]], [[TMP20]]
-; CHECK-NEXT:    [[BLOCK22:%.*]] = shufflevector <1 x i32> [[COL_LOAD12]], <1 x i32> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP22:%.*]] = extractelement <5 x i32> [[SPLIT]], i64 4
-; CHECK-NEXT:    [[SPLAT_SPLATINSERT23:%.*]] = insertelement <1 x i32> poison, i32 [[TMP22]], i64 0
-; CHECK-NEXT:    [[SPLAT_SPLAT24:%.*]] = shufflevector <1 x i32> [[SPLAT_SPLATINSERT23]], <1 x i32> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP23:%.*]] = mul <1 x i32> [[BLOCK22]], [[SPLAT_SPLAT24]]
-; CHECK-NEXT:    [[TMP24:%.*]] = add <1 x i32> [[TMP21]], [[TMP23]]
-; CHECK-NEXT:    [[TMP25:%.*]] = shufflevector <1 x i32> [[TMP24]], <1 x i32> poison, <1 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP26:%.*]] = shufflevector <1 x i32> undef, <1 x i32> [[TMP25]], <1 x i32> <i32 1>
-; CHECK-NEXT:    ret <1 x i32> [[TMP26]]
+; CHECK-NEXT:    [[TMP1:%.*]] = shufflevector <1 x i32> [[COL_LOAD]], <1 x i32> [[COL_LOAD3]], <2 x i32> <i32 0, i32 1>
+; CHECK-NEXT:    [[TMP2:%.*]] = shufflevector <1 x i32> [[COL_LOAD6]], <1 x i32> [[COL_LOAD9]], <2 x i32> <i32 0, i32 1>
+; CHECK-NEXT:    [[TMP3:%.*]] = shufflevector <2 x i32> [[TMP1]], <2 x i32> [[TMP2]], <4 x i32> <i32 0, i32 1, i32 2, i32 3>
+; CHECK-NEXT:    [[TMP4:%.*]] = shufflevector <1 x i32> [[COL_LOAD12]], <1 x i32> poison, <4 x i32> <i32 0, i32 undef, i32 undef, i32 undef>
+; CHECK-NEXT:    [[TMP5:%.*]] = shufflevector <4 x i32> [[TMP3]], <4 x i32> [[TMP4]], <5 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4>
+; CHECK-NEXT:    [[TMP6:%.*]] = extractelement <1 x i32> [[COL_LOAD]], i64 0
+; CHECK-NEXT:    [[TMP7:%.*]] = insertelement <5 x i32> poison, i32 [[TMP6]], i64 0
+; CHECK-NEXT:    [[TMP8:%.*]] = extractelement <1 x i32> [[COL_LOAD3]], i64 0
+; CHECK-NEXT:    [[TMP9:%.*]] = insertelement <5 x i32> [[TMP7]], i32 [[TMP8]], i64 1
+; CHECK-NEXT:    [[TMP10:%.*]] = extractelement <1 x i32> [[COL_LOAD6]], i64 0
+; CHECK-NEXT:    [[TMP11:%.*]] = insertelement <5 x i32> [[TMP9]], i32 [[TMP10]], i64 2
+; CHECK-NEXT:    [[TMP12:%.*]] = extractelement <1 x i32> [[COL_LOAD9]], i64 0
+; CHECK-NEXT:    [[TMP13:%.*]] = insertelement <5 x i32> [[TMP11]], i32 [[TMP12]], i64 3
+; CHECK-NEXT:    [[TMP14:%.*]] = extractelement <1 x i32> [[COL_LOAD12]], i64 0
+; CHECK-NEXT:    [[TMP15:%.*]] = insertelement <5 x i32> [[TMP13]], i32 [[TMP14]], i64 4
+; CHECK-NEXT:    call void @use.v5i32(<5 x i32> [[TMP15]])
+; CHECK-NEXT:    [[TMP16:%.*]] = mul <5 x i32> [[TMP5]], [[A:%.*]]
+; CHECK-NEXT:    [[TMP17:%.*]] = call i32 @llvm.vector.reduce.add.v5i32(<5 x i32> [[TMP16]])
+; CHECK-NEXT:    [[TMP18:%.*]] = insertelement <1 x i32> poison, i32 [[TMP17]], i64 0
+; CHECK-NEXT:    ret <1 x i32> [[TMP18]]
 ;
   %l = call <5 x i32> @llvm.matrix.column.major.load.v5i32.i64(ptr %src, i64 %stride, i1 false, i32 1, i32 5)
   %t = call <5 x i32> @llvm.matrix.transpose.v5i32(<5 x i32> %l, i32 1, i32 5)
   call void @use.v5i32(<5 x i32> %t)
   %r = call <1 x i32> @llvm.matrix.multiply.v1i32.v5i32.v5i32(<5 x i32> %l, <5 x i32> %a, i32 1, i32 5, i32 1)
   ret <1 x i32> %r
 }
 
 declare void @use.v5i32(<5 x i32>)
 
 declare <1 x i32> @llvm.matrix.multiply.v1i32.v5i32.v5i32(<5 x i32>, <5 x i32>, i32 immarg, i32 immarg, i32 immarg) #0
 
 declare <5 x i32> @llvm.matrix.column.major.load.v5i32.i64(ptr nocapture, i64, i1 immarg, i32 immarg, i32 immarg) #1
 
 declare <5 x i32> @llvm.matrix.transpose.v5i32(<5 x i32>, i32 immarg, i32 immarg) #0