[Matrix] Fold the transpose into the matmul operand used to fetch scalars

[anemet]

Summary

For column-major this is:
  A * B^t
whereas for row-major:
  A^t * B

[fhahn]

llvm/lib/Transforms/Scalar/LowerMatrixIntrinsics.cpp

1018–1023

Can you mentioned scalar_matrix_transposed in the comment?

1027

Can you adjust the name to camel case so it matches the style elsewhere in the file (isTiled is also an odd one out)

1358

I think the crashes/test failures shown for the pre-merge checks may be related to not bailing out early if !DT.

1363

 Those changes seem quite isolated form the other code in the function. Might be worth to have them in a separate function and call it before LowerMatrixMultiplyFused? Otherwise I think it would be good to adjust the name/comment for the function.

llvm/test/Transforms/LowerMatrixIntrinsics/multiply-left-transpose-row-major.ll

4

no need to use the RM prefix?

[anemet]

Address Florian's comments. Also fix the testcase that I copied mine from re: --check-prefix=RM

llvm/lib/Transforms/Scalar/LowerMatrixIntrinsics.cpp

1363

I think it's good to have this central function to dispatch all fusing opportunities for multiply. Let me update the comment.

[fhahn]

LGTM, thanks!

llvm/lib/Transforms/Scalar/LowerMatrixIntrinsics.cpp

1363

Sounds good. I still think it might help to keep things readable if the code for this new transform would be in a separate function, called from there. Or at least Move the shape variables & co inside the if() block where they are used?

[anemet]

llvm/lib/Transforms/Scalar/LowerMatrixIntrinsics.cpp

1363

Thanks! I moved most of it under the if. I expect that the interface we want for these functions would have the matmul already cracked open (i.e. A, B, R, M, C) that is why I had this outside but that's probably too early to decide at this point.

llvm/lib/Transforms/Scalar/LowerMatrixIntrinsics.cpp

for (Instruction *Inst : MatrixInsts) {
         Changed |= VisitStore(cast<StoreInst>(Inst), Op1, Op2, Builder);
     }
 
     if (ORE) {
       RemarkGenerator RemarkGen(Inst2ColumnMatrix, *ORE, Func);
       RemarkGen.emitRemarks();
     }
 
-    for (Instruction *Inst : reverse(ToRemove))
+    // Delete the instructions backwards, as it has a reduced likelihood of
+    // having to update as many def-use and use-def chains.
+    for (auto *Inst : reverse(ToRemove)) {
+      if (!Inst->use_empty())
+        Inst->replaceAllUsesWith(UndefValue::get(Inst->getType()));
       Inst->eraseFromParent();
+    }
 
     return Changed;
   }
 
   /// Turns \p BasePtr into an elementwise pointer to \p EltType.
   Value *createElementPtr(Value *BasePtr, Type *EltType, IRBuilder<> &Builder) {
     unsigned AS = cast<PointerType>(BasePtr->getType())->getAddressSpace();
     Type *EltPtrType = PointerType::get(EltType, AS);
Value *createMulAdd(Value *Sum, Value *A, Value *B, bool UseFPOp,
     }
 
     NumComputeOps += getNumOps(A->getType());
     Value *Mul = Builder.CreateMul(A, B);
     return Builder.CreateAdd(Sum, Mul);
   }
 
   /// Cache \p Matrix as result of \p Inst and update the uses of \p Inst. For
-  /// users with shape information, there's nothing to do: the will use the
+  /// users with shape information, there's nothing to do: they will use the
   /// cached value when they are lowered. For other users, \p Matrix is
   /// flattened and the uses are updated to use it. Also marks \p Inst for
   /// deletion.
   void finalizeLowering(Instruction *Inst, MatrixTy Matrix,
                         IRBuilder<> &Builder) {
     Inst2ColumnMatrix.insert(std::make_pair(Inst, Matrix));
 
     ToRemove.push_back(Inst);
     Value *Flattened = nullptr;
     for (Use &U : llvm::make_early_inc_range(Inst->uses())) {
       if (ShapeMap.find(U.getUser()) == ShapeMap.end()) {
         if (!Flattened)
           Flattened = Matrix.embedInVector(Builder);
         U.set(Flattened);
       }
     }
   }
 
   /// Compute \p Result += \p A * \p B for input matrices with left-associating
   /// addition.
+  ///
+  /// We can fold a transpose into the operand that is used to extract scalars.
+  /// This is the first operands with row-major and the second with
+  /// column-major.  If \p IsScalarMatrixTransposed we assume the appropriate
+  /// operand is transposed.
   void emitMatrixMultiply(MatrixTy &Result, const MatrixTy &A,
                           const MatrixTy &B, bool AllowContraction,
-                          IRBuilder<> &Builder, bool isTiled) {
+                          IRBuilder<> &Builder, bool IsTiled,
+                          bool IsScalarMatrixTransposed) {
     const unsigned VF = std::max<unsigned>(
         TTI.getRegisterBitWidth(TargetTransformInfo::RGK_FixedWidthVector)
                 .getFixedSize() /
             Result.getElementType()->getPrimitiveSizeInBits().getFixedSize(),
         1U);
     unsigned R = Result.getNumRows();
     unsigned C = Result.getNumColumns();
     unsigned M = A.getNumColumns();
if (A.isColumnMajor()) {
         // If Result is zero, we don't need to accumulate in the K==0 iteration.
         bool isSumZero = isa<ConstantAggregateZero>(Result.getColumn(J));
 
         for (unsigned I = 0; I < R; I += BlockSize) {
           // Gradually lower the vectorization factor to cover the remainder.
           while (I + BlockSize > R)
             BlockSize /= 2;
 
-          Value *Sum = isTiled ? Result.extractVector(I, J, BlockSize, Builder)
+          Value *Sum = IsTiled ? Result.extractVector(I, J, BlockSize, Builder)
                                : nullptr;
           for (unsigned K = 0; K < M; ++K) {
             Value *L = A.extractVector(I, K, BlockSize, Builder);
-            Value *RH = Builder.CreateExtractElement(B.getColumn(J), K);
+            Value *RH = Builder.CreateExtractElement(
+                B.getColumn(IsScalarMatrixTransposed ? K : J),
+                IsScalarMatrixTransposed ? J : K);
             Value *Splat = Builder.CreateVectorSplat(BlockSize, RH, "splat");
             Sum = createMulAdd(isSumZero && K == 0 ? nullptr : Sum, L, Splat,
                                Result.getElementType()->isFloatingPointTy(),
                                Builder, AllowContraction, NumComputeOps);
           }
           Result.setVector(J,
                            insertVector(Result.getVector(J), I, Sum, Builder));
         }
       }
     } else {
       // Multiply rows from the second operand with scalars from the first
       // operand. Then move along the K axes and accumulate the rows.  With this
       // the adds can be vectorized without reassociation.
       for (unsigned I = 0; I < R; ++I) {
         unsigned BlockSize = VF;
         bool isSumZero = isa<ConstantAggregateZero>(Result.getRow(I));
         for (unsigned J = 0; J < C; J += BlockSize) {
           // Gradually lower the vectorization factor to cover the remainder.
           while (J + BlockSize > C)
             BlockSize /= 2;
 
           Value *Sum = nullptr;
           for (unsigned K = 0; K < M; ++K) {
             Value *R = B.extractVector(K, J, BlockSize, Builder);
-            Value *LH = Builder.CreateExtractElement(A.getVector(I), K);
+            Value *LH = Builder.CreateExtractElement(
+                A.getVector(IsScalarMatrixTransposed ? K : I),
+                IsScalarMatrixTransposed ? I : K);
             Value *Splat = Builder.CreateVectorSplat(BlockSize, LH, "splat");
             Sum = createMulAdd(isSumZero && K == 0 ? nullptr : Sum, Splat, R,
                                IsFP, Builder, AllowContraction, NumComputeOps);
           }
           Result.setVector(I,
                            insertVector(Result.getVector(I), J, Sum, Builder));
         }
       }
void createTiledLoops(CallInst *MatMul, Value *LPtr, ShapeInfo LShape,
     // Insert in the inner loop body, which computes
     //   Res += Load(CurrentRow, K) * Load(K, CurrentColumn)
     Builder.SetInsertPoint(InnerBody->getTerminator());
     // Load tiles of the operands.
     MatrixTy A = loadMatrix(LPtr, {}, false, LShape, TI.CurrentRow, TI.CurrentK,
                             {TileSize, TileSize}, EltType, Builder);
     MatrixTy B = loadMatrix(RPtr, {}, false, RShape, TI.CurrentK, TI.CurrentCol,
                             {TileSize, TileSize}, EltType, Builder);
-    emitMatrixMultiply(TileResult, A, B, AllowContract, Builder, true);
+    emitMatrixMultiply(TileResult, A, B, AllowContract, Builder, true, false);
     // Store result after the inner loop is done.
     Builder.SetInsertPoint(TI.RowLoopLatch->getTerminator());
     storeMatrix(TileResult, Store->getPointerOperand(), Store->getAlign(),
                 Store->isVolatile(), {LShape.NumRows, RShape.NumColumns},
                 TI.CurrentRow, TI.CurrentCol, EltType, Builder);
 
     for (unsigned I = 0; I < TileResult.getNumVectors(); I++)
       ColumnPhis[I]->addIncoming(TileResult.getVector(I), TI.InnerLoopLatch);
else {
             MatrixTy A =
                 loadMatrix(APtr, LoadOp0->getAlign(), LoadOp0->isVolatile(),
                            LShape, Builder.getInt64(I), Builder.getInt64(K),
                            {TileR, TileM}, EltType, Builder);
             MatrixTy B =
                 loadMatrix(BPtr, LoadOp1->getAlign(), LoadOp1->isVolatile(),
                            RShape, Builder.getInt64(K), Builder.getInt64(J),
                            {TileM, TileC}, EltType, Builder);
-            emitMatrixMultiply(Res, A, B, AllowContract, Builder, true);
+            emitMatrixMultiply(Res, A, B, AllowContract, Builder, true, false);
           }
           storeMatrix(Res, CPtr, Store->getAlign(), Store->isVolatile(), {R, M},
                       Builder.getInt64(I), Builder.getInt64(J), EltType,
                       Builder);
         }
     }
 
     // Mark eliminated instructions as fused and remove them.
     FusedInsts.insert(Store);
     FusedInsts.insert(MatMul);
     Store->eraseFromParent();
     MatMul->eraseFromParent();
     if (LoadOp0->hasNUses(0)) {
       FusedInsts.insert(LoadOp0);
       LoadOp0->eraseFromParent();
     }
     if (LoadOp1->hasNUses(0)) {
       FusedInsts.insert(LoadOp1);
       LoadOp1->eraseFromParent();
     }
   }
 
   /// Try to lower matrix multiply chains by fusing operations.
   ///
-  /// Currently we only lower {ld, ld} -> matmul -> st chains.
-  //
-  /// No need to return a MatrixTy object for the result of the operation, since
-  /// the single store user will be lowered as part of this. Instructions that
-  /// are completely eliminated by fusion are added to \p FusedInsts.
+  /// Call finalizeLowering on lowered instructions.  Instructions that are
+  /// completely eliminated by fusion are added to \p FusedInsts.
   void LowerMatrixMultiplyFused(CallInst *MatMul,
                                 SmallPtrSetImpl<Instruction *> &FusedInsts) {
-    if (!FuseMatrix || !MatMul->hasOneUse() ||
-        MatrixLayout != MatrixLayoutTy::ColumnMajor || !DT)
+    if (!FuseMatrix || !DT)
       return;
 
     assert(AA && LI && "Analyses should be available");
 
-    auto *LoadOp0 = dyn_cast<LoadInst>(MatMul->getOperand(0));
-    auto *LoadOp1 = dyn_cast<LoadInst>(MatMul->getOperand(1));
+    Value *A = MatMul->getArgOperand(0);
+    Value *B = MatMul->getArgOperand(1);
+
+    // We can fold the transpose into the operand that is used to fetch scalars.
+    Value *T;
+    if (MatrixLayout == MatrixLayoutTy::ColumnMajor
+            ? match(B, m_Intrinsic<Intrinsic::matrix_transpose>(m_Value(T)))
+            : match(A, m_Intrinsic<Intrinsic::matrix_transpose>(m_Value(T)))) {
+      IRBuilder<> Builder(MatMul);
+      auto *EltType = cast<VectorType>(MatMul->getType())->getElementType();
+      ShapeInfo LShape(MatMul->getArgOperand(2), MatMul->getArgOperand(3));
+      ShapeInfo RShape(MatMul->getArgOperand(3), MatMul->getArgOperand(4));
+      const unsigned R = LShape.NumRows;
+      const unsigned M = LShape.NumColumns;
+      const unsigned C = RShape.NumColumns;
+
+      MatrixTy MA;
+      MatrixTy MB;
+
+      Value *Transpose;
+      if (MatrixLayout == MatrixLayoutTy::ColumnMajor) {
+        MA = getMatrix(A, ShapeInfo(R, M), Builder);
+        MB = getMatrix(T, ShapeInfo(C, M), Builder);
+        Transpose = B;
+      } else {
+        MA = getMatrix(T, ShapeInfo(R, M), Builder);
+        MB = getMatrix(B, ShapeInfo(C, M), Builder);
+        Transpose = A;
+      }
+
+      // Initialize the output
+      MatrixTy Result(R, C, EltType);
+
+      bool AllowContract =
+          AllowContractEnabled ||
+          (isa<FPMathOperator>(MatMul) && MatMul->hasAllowContract());
+      emitMatrixMultiply(Result, MA, MB, AllowContract, Builder, false, true);
+
+      FusedInsts.insert(MatMul);
+      FusedInsts.insert(cast<Instruction>(Transpose));
+      if (Transpose->hasOneUse())
+        ToRemove.push_back(cast<Instruction>(Transpose));
+      finalizeLowering(MatMul, Result, Builder);
+      // TODO: add a fake entry for the folded instruction so that this is
+      // included in the expression in the remark.
+      Inst2ColumnMatrix[Transpose] = MatrixTy(M, C, EltType);
+      return;
+    }
+
+    if (!MatMul->hasOneUse() || MatrixLayout != MatrixLayoutTy::ColumnMajor)
+      return;
+
+    // Lower {ld, ld} -> matmul -> st chains.  No need to call finalizeLowering
+    // since the single store user will be lowered as part of this.
+    auto *LoadOp0 = dyn_cast<LoadInst>(A);
+    auto *LoadOp1 = dyn_cast<LoadInst>(B);
     auto *Store = dyn_cast<StoreInst>(*MatMul->user_begin());
     if (LoadOp0 && LoadOp1 && Store) {
       // The store address must dominate the MatMul instruction, otherwise
       // we create invalid IR.
       // FIXME: See if we can hoist the store address computation.
       auto *AddrI = dyn_cast<Instruction>(Store->getOperand(1));
       if (AddrI && (!DT->dominates(AddrI, MatMul)))
         return;
void LowerMultiply(CallInst *MatMul) {
 
     // Initialize the output
     MatrixTy Result(R, C, EltType);
     assert(Lhs.getElementType() == Result.getElementType() &&
            "Matrix multiply result element type does not match arguments.");
 
     bool AllowContract = AllowContractEnabled || (isa<FPMathOperator>(MatMul) &&
                                                   MatMul->hasAllowContract());
-    emitMatrixMultiply(Result, Lhs, Rhs, AllowContract, Builder, false);
+    emitMatrixMultiply(Result, Lhs, Rhs, AllowContract, Builder, false, false);
     finalizeLowering(MatMul, Result, Builder);
   }
 
   /// Lowers llvm.matrix.transpose.
   void LowerTranspose(CallInst *Inst) {
     MatrixTy Result;
     IRBuilder<> Builder(Inst);
     Value *InputVal = Inst->getArgOperand(0);

llvm/test/Transforms/LowerMatrixIntrinsics/multiply-double-row-major.ll

 ; NOTE: Assertions have been autogenerated by utils/update_test_checks.py
 
-; RUN: opt -lower-matrix-intrinsics -matrix-default-layout=row-major -S < %s | FileCheck --check-prefix=RM %s
-
-
+; RUN: opt -lower-matrix-intrinsics -matrix-default-layout=row-major -S < %s | FileCheck %s
 
 define <4 x double> @multiply_2x2(<4 x double> %a, <4 x double> %b) {
-; RM-LABEL: @multiply_2x2(
-; RM-NEXT:  entry:
-; RM-NEXT:    [[SPLIT:%.*]] = shufflevector <4 x double> [[A:%.*]], <4 x double> poison, <2 x i32> <i32 0, i32 1>
-; RM-NEXT:    [[SPLIT1:%.*]] = shufflevector <4 x double> [[A]], <4 x double> poison, <2 x i32> <i32 2, i32 3>
-; RM-NEXT:    [[SPLIT2:%.*]] = shufflevector <4 x double> [[B:%.*]], <4 x double> poison, <2 x i32> <i32 0, i32 1>
-; RM-NEXT:    [[SPLIT3:%.*]] = shufflevector <4 x double> [[B]], <4 x double> poison, <2 x i32> <i32 2, i32 3>
-; RM-NEXT:    [[BLOCK:%.*]] = shufflevector <2 x double> [[SPLIT2]], <2 x double> poison, <1 x i32> zeroinitializer
-; RM-NEXT:    [[TMP0:%.*]] = extractelement <2 x double> [[SPLIT]], i64 0
-; RM-NEXT:    [[SPLAT_SPLATINSERT:%.*]] = insertelement <1 x double> poison, double [[TMP0]], i32 0
-; RM-NEXT:    [[SPLAT_SPLAT:%.*]] = shufflevector <1 x double> [[SPLAT_SPLATINSERT]], <1 x double> poison, <1 x i32> zeroinitializer
-; RM-NEXT:    [[TMP1:%.*]] = fmul <1 x double> [[SPLAT_SPLAT]], [[BLOCK]]
-; RM-NEXT:    [[BLOCK4:%.*]] = shufflevector <2 x double> [[SPLIT3]], <2 x double> poison, <1 x i32> zeroinitializer
-; RM-NEXT:    [[TMP2:%.*]] = extractelement <2 x double> [[SPLIT]], i64 1
-; RM-NEXT:    [[SPLAT_SPLATINSERT5:%.*]] = insertelement <1 x double> poison, double [[TMP2]], i32 0
-; RM-NEXT:    [[SPLAT_SPLAT6:%.*]] = shufflevector <1 x double> [[SPLAT_SPLATINSERT5]], <1 x double> poison, <1 x i32> zeroinitializer
-; RM-NEXT:    [[TMP3:%.*]] = fmul <1 x double> [[SPLAT_SPLAT6]], [[BLOCK4]]
-; RM-NEXT:    [[TMP4:%.*]] = fadd <1 x double> [[TMP1]], [[TMP3]]
-; RM-NEXT:    [[TMP5:%.*]] = shufflevector <1 x double> [[TMP4]], <1 x double> poison, <2 x i32> <i32 0, i32 undef>
-; RM-NEXT:    [[TMP6:%.*]] = shufflevector <2 x double> undef, <2 x double> [[TMP5]], <2 x i32> <i32 2, i32 1>
-; RM-NEXT:    [[BLOCK7:%.*]] = shufflevector <2 x double> [[SPLIT2]], <2 x double> poison, <1 x i32> <i32 1>
-; RM-NEXT:    [[TMP7:%.*]] = extractelement <2 x double> [[SPLIT]], i64 0
-; RM-NEXT:    [[SPLAT_SPLATINSERT8:%.*]] = insertelement <1 x double> poison, double [[TMP7]], i32 0
-; RM-NEXT:    [[SPLAT_SPLAT9:%.*]] = shufflevector <1 x double> [[SPLAT_SPLATINSERT8]], <1 x double> poison, <1 x i32> zeroinitializer
-; RM-NEXT:    [[TMP8:%.*]] = fmul <1 x double> [[SPLAT_SPLAT9]], [[BLOCK7]]
-; RM-NEXT:    [[BLOCK10:%.*]] = shufflevector <2 x double> [[SPLIT3]], <2 x double> poison, <1 x i32> <i32 1>
-; RM-NEXT:    [[TMP9:%.*]] = extractelement <2 x double> [[SPLIT]], i64 1
-; RM-NEXT:    [[SPLAT_SPLATINSERT11:%.*]] = insertelement <1 x double> poison, double [[TMP9]], i32 0
-; RM-NEXT:    [[SPLAT_SPLAT12:%.*]] = shufflevector <1 x double> [[SPLAT_SPLATINSERT11]], <1 x double> poison, <1 x i32> zeroinitializer
-; RM-NEXT:    [[TMP10:%.*]] = fmul <1 x double> [[SPLAT_SPLAT12]], [[BLOCK10]]
-; RM-NEXT:    [[TMP11:%.*]] = fadd <1 x double> [[TMP8]], [[TMP10]]
-; RM-NEXT:    [[TMP12:%.*]] = shufflevector <1 x double> [[TMP11]], <1 x double> poison, <2 x i32> <i32 0, i32 undef>
-; RM-NEXT:    [[TMP13:%.*]] = shufflevector <2 x double> [[TMP6]], <2 x double> [[TMP12]], <2 x i32> <i32 0, i32 2>
-; RM-NEXT:    [[BLOCK13:%.*]] = shufflevector <2 x double> [[SPLIT2]], <2 x double> poison, <1 x i32> zeroinitializer
-; RM-NEXT:    [[TMP14:%.*]] = extractelement <2 x double> [[SPLIT1]], i64 0
-; RM-NEXT:    [[SPLAT_SPLATINSERT14:%.*]] = insertelement <1 x double> poison, double [[TMP14]], i32 0
-; RM-NEXT:    [[SPLAT_SPLAT15:%.*]] = shufflevector <1 x double> [[SPLAT_SPLATINSERT14]], <1 x double> poison, <1 x i32> zeroinitializer
-; RM-NEXT:    [[TMP15:%.*]] = fmul <1 x double> [[SPLAT_SPLAT15]], [[BLOCK13]]
-; RM-NEXT:    [[BLOCK16:%.*]] = shufflevector <2 x double> [[SPLIT3]], <2 x double> poison, <1 x i32> zeroinitializer
-; RM-NEXT:    [[TMP16:%.*]] = extractelement <2 x double> [[SPLIT1]], i64 1
-; RM-NEXT:    [[SPLAT_SPLATINSERT17:%.*]] = insertelement <1 x double> poison, double [[TMP16]], i32 0
-; RM-NEXT:    [[SPLAT_SPLAT18:%.*]] = shufflevector <1 x double> [[SPLAT_SPLATINSERT17]], <1 x double> poison, <1 x i32> zeroinitializer
-; RM-NEXT:    [[TMP17:%.*]] = fmul <1 x double> [[SPLAT_SPLAT18]], [[BLOCK16]]
-; RM-NEXT:    [[TMP18:%.*]] = fadd <1 x double> [[TMP15]], [[TMP17]]
-; RM-NEXT:    [[TMP19:%.*]] = shufflevector <1 x double> [[TMP18]], <1 x double> poison, <2 x i32> <i32 0, i32 undef>
-; RM-NEXT:    [[TMP20:%.*]] = shufflevector <2 x double> undef, <2 x double> [[TMP19]], <2 x i32> <i32 2, i32 1>
-; RM-NEXT:    [[BLOCK19:%.*]] = shufflevector <2 x double> [[SPLIT2]], <2 x double> poison, <1 x i32> <i32 1>
-; RM-NEXT:    [[TMP21:%.*]] = extractelement <2 x double> [[SPLIT1]], i64 0
-; RM-NEXT:    [[SPLAT_SPLATINSERT20:%.*]] = insertelement <1 x double> poison, double [[TMP21]], i32 0
-; RM-NEXT:    [[SPLAT_SPLAT21:%.*]] = shufflevector <1 x double> [[SPLAT_SPLATINSERT20]], <1 x double> poison, <1 x i32> zeroinitializer
-; RM-NEXT:    [[TMP22:%.*]] = fmul <1 x double> [[SPLAT_SPLAT21]], [[BLOCK19]]
-; RM-NEXT:    [[BLOCK22:%.*]] = shufflevector <2 x double> [[SPLIT3]], <2 x double> poison, <1 x i32> <i32 1>
-; RM-NEXT:    [[TMP23:%.*]] = extractelement <2 x double> [[SPLIT1]], i64 1
-; RM-NEXT:    [[SPLAT_SPLATINSERT23:%.*]] = insertelement <1 x double> poison, double [[TMP23]], i32 0
-; RM-NEXT:    [[SPLAT_SPLAT24:%.*]] = shufflevector <1 x double> [[SPLAT_SPLATINSERT23]], <1 x double> poison, <1 x i32> zeroinitializer
-; RM-NEXT:    [[TMP24:%.*]] = fmul <1 x double> [[SPLAT_SPLAT24]], [[BLOCK22]]
-; RM-NEXT:    [[TMP25:%.*]] = fadd <1 x double> [[TMP22]], [[TMP24]]
-; RM-NEXT:    [[TMP26:%.*]] = shufflevector <1 x double> [[TMP25]], <1 x double> poison, <2 x i32> <i32 0, i32 undef>
-; RM-NEXT:    [[TMP27:%.*]] = shufflevector <2 x double> [[TMP20]], <2 x double> [[TMP26]], <2 x i32> <i32 0, i32 2>
-; RM-NEXT:    [[TMP28:%.*]] = shufflevector <2 x double> [[TMP13]], <2 x double> [[TMP27]], <4 x i32> <i32 0, i32 1, i32 2, i32 3>
-; RM-NEXT:    ret <4 x double> [[TMP28]]
+; CHECK-LABEL: @multiply_2x2(
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[SPLIT:%.*]] = shufflevector <4 x double> [[A:%.*]], <4 x double> poison, <2 x i32> <i32 0, i32 1>
+; CHECK-NEXT:    [[SPLIT1:%.*]] = shufflevector <4 x double> [[A]], <4 x double> poison, <2 x i32> <i32 2, i32 3>
+; CHECK-NEXT:    [[SPLIT2:%.*]] = shufflevector <4 x double> [[B:%.*]], <4 x double> poison, <2 x i32> <i32 0, i32 1>
+; CHECK-NEXT:    [[SPLIT3:%.*]] = shufflevector <4 x double> [[B]], <4 x double> poison, <2 x i32> <i32 2, i32 3>
+; CHECK-NEXT:    [[BLOCK:%.*]] = shufflevector <2 x double> [[SPLIT2]], <2 x double> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP0:%.*]] = extractelement <2 x double> [[SPLIT]], i64 0
+; CHECK-NEXT:    [[SPLAT_SPLATINSERT:%.*]] = insertelement <1 x double> poison, double [[TMP0]], i32 0
+; CHECK-NEXT:    [[SPLAT_SPLAT:%.*]] = shufflevector <1 x double> [[SPLAT_SPLATINSERT]], <1 x double> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP1:%.*]] = fmul <1 x double> [[SPLAT_SPLAT]], [[BLOCK]]
+; CHECK-NEXT:    [[BLOCK4:%.*]] = shufflevector <2 x double> [[SPLIT3]], <2 x double> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP2:%.*]] = extractelement <2 x double> [[SPLIT]], i64 1
+; CHECK-NEXT:    [[SPLAT_SPLATINSERT5:%.*]] = insertelement <1 x double> poison, double [[TMP2]], i32 0
+; CHECK-NEXT:    [[SPLAT_SPLAT6:%.*]] = shufflevector <1 x double> [[SPLAT_SPLATINSERT5]], <1 x double> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP3:%.*]] = fmul <1 x double> [[SPLAT_SPLAT6]], [[BLOCK4]]
+; CHECK-NEXT:    [[TMP4:%.*]] = fadd <1 x double> [[TMP1]], [[TMP3]]
+; CHECK-NEXT:    [[TMP5:%.*]] = shufflevector <1 x double> [[TMP4]], <1 x double> poison, <2 x i32> <i32 0, i32 undef>
+; CHECK-NEXT:    [[TMP6:%.*]] = shufflevector <2 x double> undef, <2 x double> [[TMP5]], <2 x i32> <i32 2, i32 1>
+; CHECK-NEXT:    [[BLOCK7:%.*]] = shufflevector <2 x double> [[SPLIT2]], <2 x double> poison, <1 x i32> <i32 1>
+; CHECK-NEXT:    [[TMP7:%.*]] = extractelement <2 x double> [[SPLIT]], i64 0
+; CHECK-NEXT:    [[SPLAT_SPLATINSERT8:%.*]] = insertelement <1 x double> poison, double [[TMP7]], i32 0
+; CHECK-NEXT:    [[SPLAT_SPLAT9:%.*]] = shufflevector <1 x double> [[SPLAT_SPLATINSERT8]], <1 x double> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP8:%.*]] = fmul <1 x double> [[SPLAT_SPLAT9]], [[BLOCK7]]
+; CHECK-NEXT:    [[BLOCK10:%.*]] = shufflevector <2 x double> [[SPLIT3]], <2 x double> poison, <1 x i32> <i32 1>
+; CHECK-NEXT:    [[TMP9:%.*]] = extractelement <2 x double> [[SPLIT]], i64 1
+; CHECK-NEXT:    [[SPLAT_SPLATINSERT11:%.*]] = insertelement <1 x double> poison, double [[TMP9]], i32 0
+; CHECK-NEXT:    [[SPLAT_SPLAT12:%.*]] = shufflevector <1 x double> [[SPLAT_SPLATINSERT11]], <1 x double> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP10:%.*]] = fmul <1 x double> [[SPLAT_SPLAT12]], [[BLOCK10]]
+; CHECK-NEXT:    [[TMP11:%.*]] = fadd <1 x double> [[TMP8]], [[TMP10]]
+; CHECK-NEXT:    [[TMP12:%.*]] = shufflevector <1 x double> [[TMP11]], <1 x double> poison, <2 x i32> <i32 0, i32 undef>
+; CHECK-NEXT:    [[TMP13:%.*]] = shufflevector <2 x double> [[TMP6]], <2 x double> [[TMP12]], <2 x i32> <i32 0, i32 2>
+; CHECK-NEXT:    [[BLOCK13:%.*]] = shufflevector <2 x double> [[SPLIT2]], <2 x double> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP14:%.*]] = extractelement <2 x double> [[SPLIT1]], i64 0
+; CHECK-NEXT:    [[SPLAT_SPLATINSERT14:%.*]] = insertelement <1 x double> poison, double [[TMP14]], i32 0
+; CHECK-NEXT:    [[SPLAT_SPLAT15:%.*]] = shufflevector <1 x double> [[SPLAT_SPLATINSERT14]], <1 x double> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP15:%.*]] = fmul <1 x double> [[SPLAT_SPLAT15]], [[BLOCK13]]
+; CHECK-NEXT:    [[BLOCK16:%.*]] = shufflevector <2 x double> [[SPLIT3]], <2 x double> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP16:%.*]] = extractelement <2 x double> [[SPLIT1]], i64 1
+; CHECK-NEXT:    [[SPLAT_SPLATINSERT17:%.*]] = insertelement <1 x double> poison, double [[TMP16]], i32 0
+; CHECK-NEXT:    [[SPLAT_SPLAT18:%.*]] = shufflevector <1 x double> [[SPLAT_SPLATINSERT17]], <1 x double> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP17:%.*]] = fmul <1 x double> [[SPLAT_SPLAT18]], [[BLOCK16]]
+; CHECK-NEXT:    [[TMP18:%.*]] = fadd <1 x double> [[TMP15]], [[TMP17]]
+; CHECK-NEXT:    [[TMP19:%.*]] = shufflevector <1 x double> [[TMP18]], <1 x double> poison, <2 x i32> <i32 0, i32 undef>
+; CHECK-NEXT:    [[TMP20:%.*]] = shufflevector <2 x double> undef, <2 x double> [[TMP19]], <2 x i32> <i32 2, i32 1>
+; CHECK-NEXT:    [[BLOCK19:%.*]] = shufflevector <2 x double> [[SPLIT2]], <2 x double> poison, <1 x i32> <i32 1>
+; CHECK-NEXT:    [[TMP21:%.*]] = extractelement <2 x double> [[SPLIT1]], i64 0
+; CHECK-NEXT:    [[SPLAT_SPLATINSERT20:%.*]] = insertelement <1 x double> poison, double [[TMP21]], i32 0
+; CHECK-NEXT:    [[SPLAT_SPLAT21:%.*]] = shufflevector <1 x double> [[SPLAT_SPLATINSERT20]], <1 x double> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP22:%.*]] = fmul <1 x double> [[SPLAT_SPLAT21]], [[BLOCK19]]
+; CHECK-NEXT:    [[BLOCK22:%.*]] = shufflevector <2 x double> [[SPLIT3]], <2 x double> poison, <1 x i32> <i32 1>
+; CHECK-NEXT:    [[TMP23:%.*]] = extractelement <2 x double> [[SPLIT1]], i64 1
+; CHECK-NEXT:    [[SPLAT_SPLATINSERT23:%.*]] = insertelement <1 x double> poison, double [[TMP23]], i32 0
+; CHECK-NEXT:    [[SPLAT_SPLAT24:%.*]] = shufflevector <1 x double> [[SPLAT_SPLATINSERT23]], <1 x double> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP24:%.*]] = fmul <1 x double> [[SPLAT_SPLAT24]], [[BLOCK22]]
+; CHECK-NEXT:    [[TMP25:%.*]] = fadd <1 x double> [[TMP22]], [[TMP24]]
+; CHECK-NEXT:    [[TMP26:%.*]] = shufflevector <1 x double> [[TMP25]], <1 x double> poison, <2 x i32> <i32 0, i32 undef>
+; CHECK-NEXT:    [[TMP27:%.*]] = shufflevector <2 x double> [[TMP20]], <2 x double> [[TMP26]], <2 x i32> <i32 0, i32 2>
+; CHECK-NEXT:    [[TMP28:%.*]] = shufflevector <2 x double> [[TMP13]], <2 x double> [[TMP27]], <4 x i32> <i32 0, i32 1, i32 2, i32 3>
+; CHECK-NEXT:    ret <4 x double> [[TMP28]]
 ;
 entry:
   %c = call <4 x double> @llvm.matrix.multiply.v4f64.v4f64.v4f64(<4 x double> %a, <4 x double> %b, i32 2, i32 2, i32 2)
   ret <4 x double> %c
 }
 
 declare <4 x double> @llvm.matrix.multiply.v4f64.v4f64.v4f64(<4 x double>, <4 x double>, i32, i32, i32)
 
 define <4 x double> @multiply_1x2(<2 x double> %a, <2 x double> %b) {
-; RM-LABEL: @multiply_1x2(
-; RM-NEXT:  entry:
-; RM-NEXT:    [[SPLIT:%.*]] = shufflevector <2 x double> [[A:%.*]], <2 x double> poison, <1 x i32> zeroinitializer
-; RM-NEXT:    [[SPLIT1:%.*]] = shufflevector <2 x double> [[A]], <2 x double> poison, <1 x i32> <i32 1>
-; RM-NEXT:    [[SPLIT2:%.*]] = shufflevector <2 x double> [[B:%.*]], <2 x double> poison, <2 x i32> <i32 0, i32 1>
-; RM-NEXT:    [[BLOCK:%.*]] = shufflevector <2 x double> [[SPLIT2]], <2 x double> poison, <1 x i32> zeroinitializer
-; RM-NEXT:    [[TMP0:%.*]] = extractelement <1 x double> [[SPLIT]], i64 0
-; RM-NEXT:    [[SPLAT_SPLATINSERT:%.*]] = insertelement <1 x double> poison, double [[TMP0]], i32 0
-; RM-NEXT:    [[SPLAT_SPLAT:%.*]] = shufflevector <1 x double> [[SPLAT_SPLATINSERT]], <1 x double> poison, <1 x i32> zeroinitializer
-; RM-NEXT:    [[TMP1:%.*]] = fmul <1 x double> [[SPLAT_SPLAT]], [[BLOCK]]
-; RM-NEXT:    [[TMP2:%.*]] = shufflevector <1 x double> [[TMP1]], <1 x double> poison, <2 x i32> <i32 0, i32 undef>
-; RM-NEXT:    [[TMP3:%.*]] = shufflevector <2 x double> undef, <2 x double> [[TMP2]], <2 x i32> <i32 2, i32 1>
-; RM-NEXT:    [[BLOCK3:%.*]] = shufflevector <2 x double> [[SPLIT2]], <2 x double> poison, <1 x i32> <i32 1>
-; RM-NEXT:    [[TMP4:%.*]] = extractelement <1 x double> [[SPLIT]], i64 0
-; RM-NEXT:    [[SPLAT_SPLATINSERT4:%.*]] = insertelement <1 x double> poison, double [[TMP4]], i32 0
-; RM-NEXT:    [[SPLAT_SPLAT5:%.*]] = shufflevector <1 x double> [[SPLAT_SPLATINSERT4]], <1 x double> poison, <1 x i32> zeroinitializer
-; RM-NEXT:    [[TMP5:%.*]] = fmul <1 x double> [[SPLAT_SPLAT5]], [[BLOCK3]]
-; RM-NEXT:    [[TMP6:%.*]] = shufflevector <1 x double> [[TMP5]], <1 x double> poison, <2 x i32> <i32 0, i32 undef>
-; RM-NEXT:    [[TMP7:%.*]] = shufflevector <2 x double> [[TMP3]], <2 x double> [[TMP6]], <2 x i32> <i32 0, i32 2>
-; RM-NEXT:    [[BLOCK6:%.*]] = shufflevector <2 x double> [[SPLIT2]], <2 x double> poison, <1 x i32> zeroinitializer
-; RM-NEXT:    [[TMP8:%.*]] = extractelement <1 x double> [[SPLIT1]], i64 0
-; RM-NEXT:    [[SPLAT_SPLATINSERT7:%.*]] = insertelement <1 x double> poison, double [[TMP8]], i32 0
-; RM-NEXT:    [[SPLAT_SPLAT8:%.*]] = shufflevector <1 x double> [[SPLAT_SPLATINSERT7]], <1 x double> poison, <1 x i32> zeroinitializer
-; RM-NEXT:    [[TMP9:%.*]] = fmul <1 x double> [[SPLAT_SPLAT8]], [[BLOCK6]]
-; RM-NEXT:    [[TMP10:%.*]] = shufflevector <1 x double> [[TMP9]], <1 x double> poison, <2 x i32> <i32 0, i32 undef>
-; RM-NEXT:    [[TMP11:%.*]] = shufflevector <2 x double> undef, <2 x double> [[TMP10]], <2 x i32> <i32 2, i32 1>
-; RM-NEXT:    [[BLOCK9:%.*]] = shufflevector <2 x double> [[SPLIT2]], <2 x double> poison, <1 x i32> <i32 1>
-; RM-NEXT:    [[TMP12:%.*]] = extractelement <1 x double> [[SPLIT1]], i64 0
-; RM-NEXT:    [[SPLAT_SPLATINSERT10:%.*]] = insertelement <1 x double> poison, double [[TMP12]], i32 0
-; RM-NEXT:    [[SPLAT_SPLAT11:%.*]] = shufflevector <1 x double> [[SPLAT_SPLATINSERT10]], <1 x double> poison, <1 x i32> zeroinitializer
-; RM-NEXT:    [[TMP13:%.*]] = fmul <1 x double> [[SPLAT_SPLAT11]], [[BLOCK9]]
-; RM-NEXT:    [[TMP14:%.*]] = shufflevector <1 x double> [[TMP13]], <1 x double> poison, <2 x i32> <i32 0, i32 undef>
-; RM-NEXT:    [[TMP15:%.*]] = shufflevector <2 x double> [[TMP11]], <2 x double> [[TMP14]], <2 x i32> <i32 0, i32 2>
-; RM-NEXT:    [[TMP16:%.*]] = shufflevector <2 x double> [[TMP7]], <2 x double> [[TMP15]], <4 x i32> <i32 0, i32 1, i32 2, i32 3>
-; RM-NEXT:    ret <4 x double> [[TMP16]]
+; CHECK-LABEL: @multiply_1x2(
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[SPLIT:%.*]] = shufflevector <2 x double> [[A:%.*]], <2 x double> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[SPLIT1:%.*]] = shufflevector <2 x double> [[A]], <2 x double> poison, <1 x i32> <i32 1>
+; CHECK-NEXT:    [[SPLIT2:%.*]] = shufflevector <2 x double> [[B:%.*]], <2 x double> poison, <2 x i32> <i32 0, i32 1>
+; CHECK-NEXT:    [[BLOCK:%.*]] = shufflevector <2 x double> [[SPLIT2]], <2 x double> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP0:%.*]] = extractelement <1 x double> [[SPLIT]], i64 0
+; CHECK-NEXT:    [[SPLAT_SPLATINSERT:%.*]] = insertelement <1 x double> poison, double [[TMP0]], i32 0
+; CHECK-NEXT:    [[SPLAT_SPLAT:%.*]] = shufflevector <1 x double> [[SPLAT_SPLATINSERT]], <1 x double> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP1:%.*]] = fmul <1 x double> [[SPLAT_SPLAT]], [[BLOCK]]
+; CHECK-NEXT:    [[TMP2:%.*]] = shufflevector <1 x double> [[TMP1]], <1 x double> poison, <2 x i32> <i32 0, i32 undef>
+; CHECK-NEXT:    [[TMP3:%.*]] = shufflevector <2 x double> undef, <2 x double> [[TMP2]], <2 x i32> <i32 2, i32 1>
+; CHECK-NEXT:    [[BLOCK3:%.*]] = shufflevector <2 x double> [[SPLIT2]], <2 x double> poison, <1 x i32> <i32 1>
+; CHECK-NEXT:    [[TMP4:%.*]] = extractelement <1 x double> [[SPLIT]], i64 0
+; CHECK-NEXT:    [[SPLAT_SPLATINSERT4:%.*]] = insertelement <1 x double> poison, double [[TMP4]], i32 0
+; CHECK-NEXT:    [[SPLAT_SPLAT5:%.*]] = shufflevector <1 x double> [[SPLAT_SPLATINSERT4]], <1 x double> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP5:%.*]] = fmul <1 x double> [[SPLAT_SPLAT5]], [[BLOCK3]]
+; CHECK-NEXT:    [[TMP6:%.*]] = shufflevector <1 x double> [[TMP5]], <1 x double> poison, <2 x i32> <i32 0, i32 undef>
+; CHECK-NEXT:    [[TMP7:%.*]] = shufflevector <2 x double> [[TMP3]], <2 x double> [[TMP6]], <2 x i32> <i32 0, i32 2>
+; CHECK-NEXT:    [[BLOCK6:%.*]] = shufflevector <2 x double> [[SPLIT2]], <2 x double> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP8:%.*]] = extractelement <1 x double> [[SPLIT1]], i64 0
+; CHECK-NEXT:    [[SPLAT_SPLATINSERT7:%.*]] = insertelement <1 x double> poison, double [[TMP8]], i32 0
+; CHECK-NEXT:    [[SPLAT_SPLAT8:%.*]] = shufflevector <1 x double> [[SPLAT_SPLATINSERT7]], <1 x double> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP9:%.*]] = fmul <1 x double> [[SPLAT_SPLAT8]], [[BLOCK6]]
+; CHECK-NEXT:    [[TMP10:%.*]] = shufflevector <1 x double> [[TMP9]], <1 x double> poison, <2 x i32> <i32 0, i32 undef>
+; CHECK-NEXT:    [[TMP11:%.*]] = shufflevector <2 x double> undef, <2 x double> [[TMP10]], <2 x i32> <i32 2, i32 1>
+; CHECK-NEXT:    [[BLOCK9:%.*]] = shufflevector <2 x double> [[SPLIT2]], <2 x double> poison, <1 x i32> <i32 1>
+; CHECK-NEXT:    [[TMP12:%.*]] = extractelement <1 x double> [[SPLIT1]], i64 0
+; CHECK-NEXT:    [[SPLAT_SPLATINSERT10:%.*]] = insertelement <1 x double> poison, double [[TMP12]], i32 0
+; CHECK-NEXT:    [[SPLAT_SPLAT11:%.*]] = shufflevector <1 x double> [[SPLAT_SPLATINSERT10]], <1 x double> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP13:%.*]] = fmul <1 x double> [[SPLAT_SPLAT11]], [[BLOCK9]]
+; CHECK-NEXT:    [[TMP14:%.*]] = shufflevector <1 x double> [[TMP13]], <1 x double> poison, <2 x i32> <i32 0, i32 undef>
+; CHECK-NEXT:    [[TMP15:%.*]] = shufflevector <2 x double> [[TMP11]], <2 x double> [[TMP14]], <2 x i32> <i32 0, i32 2>
+; CHECK-NEXT:    [[TMP16:%.*]] = shufflevector <2 x double> [[TMP7]], <2 x double> [[TMP15]], <4 x i32> <i32 0, i32 1, i32 2, i32 3>
+; CHECK-NEXT:    ret <4 x double> [[TMP16]]
 ;
 entry:
   %c = call <4 x double> @llvm.matrix.multiply.v4f64.v2f64.v2f64(<2 x double> %a, <2 x double> %b, i32 2, i32 1, i32 2)
   ret <4 x double> %c
 }
 
 declare <4 x double> @llvm.matrix.multiply.v4f64.v2f64.v2f64(<2 x double>, <2 x double>, i32, i32, i32)
 
 define <9 x double> @multiply_2x3(<6 x double> %a, <6 x double> %b) {
-; RM-LABEL: @multiply_2x3(
-; RM-NEXT:  entry:
-; RM-NEXT:    [[SPLIT:%.*]] = shufflevector <6 x double> [[A:%.*]], <6 x double> poison, <2 x i32> <i32 0, i32 1>
-; RM-NEXT:    [[SPLIT1:%.*]] = shufflevector <6 x double> [[A]], <6 x double> poison, <2 x i32> <i32 2, i32 3>
-; RM-NEXT:    [[SPLIT2:%.*]] = shufflevector <6 x double> [[A]], <6 x double> poison, <2 x i32> <i32 4, i32 5>
-; RM-NEXT:    [[SPLIT3:%.*]] = shufflevector <6 x double> [[B:%.*]], <6 x double> poison, <3 x i32> <i32 0, i32 1, i32 2>
-; RM-NEXT:    [[SPLIT4:%.*]] = shufflevector <6 x double> [[B]], <6 x double> poison, <3 x i32> <i32 3, i32 4, i32 5>
-; RM-NEXT:    [[BLOCK:%.*]] = shufflevector <3 x double> [[SPLIT3]], <3 x double> poison, <1 x i32> zeroinitializer
-; RM-NEXT:    [[TMP0:%.*]] = extractelement <2 x double> [[SPLIT]], i64 0
-; RM-NEXT:    [[SPLAT_SPLATINSERT:%.*]] = insertelement <1 x double> poison, double [[TMP0]], i32 0
-; RM-NEXT:    [[SPLAT_SPLAT:%.*]] = shufflevector <1 x double> [[SPLAT_SPLATINSERT]], <1 x double> poison, <1 x i32> zeroinitializer
-; RM-NEXT:    [[TMP1:%.*]] = fmul <1 x double> [[SPLAT_SPLAT]], [[BLOCK]]
-; RM-NEXT:    [[BLOCK5:%.*]] = shufflevector <3 x double> [[SPLIT4]], <3 x double> poison, <1 x i32> zeroinitializer
-; RM-NEXT:    [[TMP2:%.*]] = extractelement <2 x double> [[SPLIT]], i64 1
-; RM-NEXT:    [[SPLAT_SPLATINSERT6:%.*]] = insertelement <1 x double> poison, double [[TMP2]], i32 0
-; RM-NEXT:    [[SPLAT_SPLAT7:%.*]] = shufflevector <1 x double> [[SPLAT_SPLATINSERT6]], <1 x double> poison, <1 x i32> zeroinitializer
-; RM-NEXT:    [[TMP3:%.*]] = fmul <1 x double> [[SPLAT_SPLAT7]], [[BLOCK5]]
-; RM-NEXT:    [[TMP4:%.*]] = fadd <1 x double> [[TMP1]], [[TMP3]]
-; RM-NEXT:    [[TMP5:%.*]] = shufflevector <1 x double> [[TMP4]], <1 x double> poison, <3 x i32> <i32 0, i32 undef, i32 undef>
-; RM-NEXT:    [[TMP6:%.*]] = shufflevector <3 x double> undef, <3 x double> [[TMP5]], <3 x i32> <i32 3, i32 1, i32 2>
-; RM-NEXT:    [[BLOCK8:%.*]] = shufflevector <3 x double> [[SPLIT3]], <3 x double> poison, <1 x i32> <i32 1>
-; RM-NEXT:    [[TMP7:%.*]] = extractelement <2 x double> [[SPLIT]], i64 0
-; RM-NEXT:    [[SPLAT_SPLATINSERT9:%.*]] = insertelement <1 x double> poison, double [[TMP7]], i32 0
-; RM-NEXT:    [[SPLAT_SPLAT10:%.*]] = shufflevector <1 x double> [[SPLAT_SPLATINSERT9]], <1 x double> poison, <1 x i32> zeroinitializer
-; RM-NEXT:    [[TMP8:%.*]] = fmul <1 x double> [[SPLAT_SPLAT10]], [[BLOCK8]]
-; RM-NEXT:    [[BLOCK11:%.*]] = shufflevector <3 x double> [[SPLIT4]], <3 x double> poison, <1 x i32> <i32 1>
-; RM-NEXT:    [[TMP9:%.*]] = extractelement <2 x double> [[SPLIT]], i64 1
-; RM-NEXT:    [[SPLAT_SPLATINSERT12:%.*]] = insertelement <1 x double> poison, double [[TMP9]], i32 0
-; RM-NEXT:    [[SPLAT_SPLAT13:%.*]] = shufflevector <1 x double> [[SPLAT_SPLATINSERT12]], <1 x double> poison, <1 x i32> zeroinitializer
-; RM-NEXT:    [[TMP10:%.*]] = fmul <1 x double> [[SPLAT_SPLAT13]], [[BLOCK11]]
-; RM-NEXT:    [[TMP11:%.*]] = fadd <1 x double> [[TMP8]], [[TMP10]]
-; RM-NEXT:    [[TMP12:%.*]] = shufflevector <1 x double> [[TMP11]], <1 x double> poison, <3 x i32> <i32 0, i32 undef, i32 undef>
-; RM-NEXT:    [[TMP13:%.*]] = shufflevector <3 x double> [[TMP6]], <3 x double> [[TMP12]], <3 x i32> <i32 0, i32 3, i32 2>
-; RM-NEXT:    [[BLOCK14:%.*]] = shufflevector <3 x double> [[SPLIT3]], <3 x double> poison, <1 x i32> <i32 2>
-; RM-NEXT:    [[TMP14:%.*]] = extractelement <2 x double> [[SPLIT]], i64 0
-; RM-NEXT:    [[SPLAT_SPLATINSERT15:%.*]] = insertelement <1 x double> poison, double [[TMP14]], i32 0
-; RM-NEXT:    [[SPLAT_SPLAT16:%.*]] = shufflevector <1 x double> [[SPLAT_SPLATINSERT15]], <1 x double> poison, <1 x i32> zeroinitializer
-; RM-NEXT:    [[TMP15:%.*]] = fmul <1 x double> [[SPLAT_SPLAT16]], [[BLOCK14]]
-; RM-NEXT:    [[BLOCK17:%.*]] = shufflevector <3 x double> [[SPLIT4]], <3 x double> poison, <1 x i32> <i32 2>
-; RM-NEXT:    [[TMP16:%.*]] = extractelement <2 x double> [[SPLIT]], i64 1
-; RM-NEXT:    [[SPLAT_SPLATINSERT18:%.*]] = insertelement <1 x double> poison, double [[TMP16]], i32 0
-; RM-NEXT:    [[SPLAT_SPLAT19:%.*]] = shufflevector <1 x double> [[SPLAT_SPLATINSERT18]], <1 x double> poison, <1 x i32> zeroinitializer
-; RM-NEXT:    [[TMP17:%.*]] = fmul <1 x double> [[SPLAT_SPLAT19]], [[BLOCK17]]
-; RM-NEXT:    [[TMP18:%.*]] = fadd <1 x double> [[TMP15]], [[TMP17]]
-; RM-NEXT:    [[TMP19:%.*]] = shufflevector <1 x double> [[TMP18]], <1 x double> poison, <3 x i32> <i32 0, i32 undef, i32 undef>
-; RM-NEXT:    [[TMP20:%.*]] = shufflevector <3 x double> [[TMP13]], <3 x double> [[TMP19]], <3 x i32> <i32 0, i32 1, i32 3>
-; RM-NEXT:    [[BLOCK20:%.*]] = shufflevector <3 x double> [[SPLIT3]], <3 x double> poison, <1 x i32> zeroinitializer
-; RM-NEXT:    [[TMP21:%.*]] = extractelement <2 x double> [[SPLIT1]], i64 0
-; RM-NEXT:    [[SPLAT_SPLATINSERT21:%.*]] = insertelement <1 x double> poison, double [[TMP21]], i32 0
-; RM-NEXT:    [[SPLAT_SPLAT22:%.*]] = shufflevector <1 x double> [[SPLAT_SPLATINSERT21]], <1 x double> poison, <1 x i32> zeroinitializer
-; RM-NEXT:    [[TMP22:%.*]] = fmul <1 x double> [[SPLAT_SPLAT22]], [[BLOCK20]]
-; RM-NEXT:    [[BLOCK23:%.*]] = shufflevector <3 x double> [[SPLIT4]], <3 x double> poison, <1 x i32> zeroinitializer
-; RM-NEXT:    [[TMP23:%.*]] = extractelement <2 x double> [[SPLIT1]], i64 1
-; RM-NEXT:    [[SPLAT_SPLATINSERT24:%.*]] = insertelement <1 x double> poison, double [[TMP23]], i32 0
-; RM-NEXT:    [[SPLAT_SPLAT25:%.*]] = shufflevector <1 x double> [[SPLAT_SPLATINSERT24]], <1 x double> poison, <1 x i32> zeroinitializer
-; RM-NEXT:    [[TMP24:%.*]] = fmul <1 x double> [[SPLAT_SPLAT25]], [[BLOCK23]]
-; RM-NEXT:    [[TMP25:%.*]] = fadd <1 x double> [[TMP22]], [[TMP24]]
-; RM-NEXT:    [[TMP26:%.*]] = shufflevector <1 x double> [[TMP25]], <1 x double> poison, <3 x i32> <i32 0, i32 undef, i32 undef>
-; RM-NEXT:    [[TMP27:%.*]] = shufflevector <3 x double> undef, <3 x double> [[TMP26]], <3 x i32> <i32 3, i32 1, i32 2>
-; RM-NEXT:    [[BLOCK26:%.*]] = shufflevector <3 x double> [[SPLIT3]], <3 x double> poison, <1 x i32> <i32 1>
-; RM-NEXT:    [[TMP28:%.*]] = extractelement <2 x double> [[SPLIT1]], i64 0
-; RM-NEXT:    [[SPLAT_SPLATINSERT27:%.*]] = insertelement <1 x double> poison, double [[TMP28]], i32 0
-; RM-NEXT:    [[SPLAT_SPLAT28:%.*]] = shufflevector <1 x double> [[SPLAT_SPLATINSERT27]], <1 x double> poison, <1 x i32> zeroinitializer
-; RM-NEXT:    [[TMP29:%.*]] = fmul <1 x double> [[SPLAT_SPLAT28]], [[BLOCK26]]
-; RM-NEXT:    [[BLOCK29:%.*]] = shufflevector <3 x double> [[SPLIT4]], <3 x double> poison, <1 x i32> <i32 1>
-; RM-NEXT:    [[TMP30:%.*]] = extractelement <2 x double> [[SPLIT1]], i64 1
-; RM-NEXT:    [[SPLAT_SPLATINSERT30:%.*]] = insertelement <1 x double> poison, double [[TMP30]], i32 0
-; RM-NEXT:    [[SPLAT_SPLAT31:%.*]] = shufflevector <1 x double> [[SPLAT_SPLATINSERT30]], <1 x double> poison, <1 x i32> zeroinitializer
-; RM-NEXT:    [[TMP31:%.*]] = fmul <1 x double> [[SPLAT_SPLAT31]], [[BLOCK29]]
-; RM-NEXT:    [[TMP32:%.*]] = fadd <1 x double> [[TMP29]], [[TMP31]]
-; RM-NEXT:    [[TMP33:%.*]] = shufflevector <1 x double> [[TMP32]], <1 x double> poison, <3 x i32> <i32 0, i32 undef, i32 undef>
-; RM-NEXT:    [[TMP34:%.*]] = shufflevector <3 x double> [[TMP27]], <3 x double> [[TMP33]], <3 x i32> <i32 0, i32 3, i32 2>
-; RM-NEXT:    [[BLOCK32:%.*]] = shufflevector <3 x double> [[SPLIT3]], <3 x double> poison, <1 x i32> <i32 2>
-; RM-NEXT:    [[TMP35:%.*]] = extractelement <2 x double> [[SPLIT1]], i64 0
-; RM-NEXT:    [[SPLAT_SPLATINSERT33:%.*]] = insertelement <1 x double> poison, double [[TMP35]], i32 0
-; RM-NEXT:    [[SPLAT_SPLAT34:%.*]] = shufflevector <1 x double> [[SPLAT_SPLATINSERT33]], <1 x double> poison, <1 x i32> zeroinitializer
-; RM-NEXT:    [[TMP36:%.*]] = fmul <1 x double> [[SPLAT_SPLAT34]], [[BLOCK32]]
-; RM-NEXT:    [[BLOCK35:%.*]] = shufflevector <3 x double> [[SPLIT4]], <3 x double> poison, <1 x i32> <i32 2>
-; RM-NEXT:    [[TMP37:%.*]] = extractelement <2 x double> [[SPLIT1]], i64 1
-; RM-NEXT:    [[SPLAT_SPLATINSERT36:%.*]] = insertelement <1 x double> poison, double [[TMP37]], i32 0
-; RM-NEXT:    [[SPLAT_SPLAT37:%.*]] = shufflevector <1 x double> [[SPLAT_SPLATINSERT36]], <1 x double> poison, <1 x i32> zeroinitializer
-; RM-NEXT:    [[TMP38:%.*]] = fmul <1 x double> [[SPLAT_SPLAT37]], [[BLOCK35]]
-; RM-NEXT:    [[TMP39:%.*]] = fadd <1 x double> [[TMP36]], [[TMP38]]
-; RM-NEXT:    [[TMP40:%.*]] = shufflevector <1 x double> [[TMP39]], <1 x double> poison, <3 x i32> <i32 0, i32 undef, i32 undef>
-; RM-NEXT:    [[TMP41:%.*]] = shufflevector <3 x double> [[TMP34]], <3 x double> [[TMP40]], <3 x i32> <i32 0, i32 1, i32 3>
-; RM-NEXT:    [[BLOCK38:%.*]] = shufflevector <3 x double> [[SPLIT3]], <3 x double> poison, <1 x i32> zeroinitializer
-; RM-NEXT:    [[TMP42:%.*]] = extractelement <2 x double> [[SPLIT2]], i64 0
-; RM-NEXT:    [[SPLAT_SPLATINSERT39:%.*]] = insertelement <1 x double> poison, double [[TMP42]], i32 0
-; RM-NEXT:    [[SPLAT_SPLAT40:%.*]] = shufflevector <1 x double> [[SPLAT_SPLATINSERT39]], <1 x double> poison, <1 x i32> zeroinitializer
-; RM-NEXT:    [[TMP43:%.*]] = fmul <1 x double> [[SPLAT_SPLAT40]], [[BLOCK38]]
-; RM-NEXT:    [[BLOCK41:%.*]] = shufflevector <3 x double> [[SPLIT4]], <3 x double> poison, <1 x i32> zeroinitializer
-; RM-NEXT:    [[TMP44:%.*]] = extractelement <2 x double> [[SPLIT2]], i64 1
-; RM-NEXT:    [[SPLAT_SPLATINSERT42:%.*]] = insertelement <1 x double> poison, double [[TMP44]], i32 0
-; RM-NEXT:    [[SPLAT_SPLAT43:%.*]] = shufflevector <1 x double> [[SPLAT_SPLATINSERT42]], <1 x double> poison, <1 x i32> zeroinitializer
-; RM-NEXT:    [[TMP45:%.*]] = fmul <1 x double> [[SPLAT_SPLAT43]], [[BLOCK41]]
-; RM-NEXT:    [[TMP46:%.*]] = fadd <1 x double> [[TMP43]], [[TMP45]]
-; RM-NEXT:    [[TMP47:%.*]] = shufflevector <1 x double> [[TMP46]], <1 x double> poison, <3 x i32> <i32 0, i32 undef, i32 undef>
-; RM-NEXT:    [[TMP48:%.*]] = shufflevector <3 x double> undef, <3 x double> [[TMP47]], <3 x i32> <i32 3, i32 1, i32 2>
-; RM-NEXT:    [[BLOCK44:%.*]] = shufflevector <3 x double> [[SPLIT3]], <3 x double> poison, <1 x i32> <i32 1>
-; RM-NEXT:    [[TMP49:%.*]] = extractelement <2 x double> [[SPLIT2]], i64 0
-; RM-NEXT:    [[SPLAT_SPLATINSERT45:%.*]] = insertelement <1 x double> poison, double [[TMP49]], i32 0
-; RM-NEXT:    [[SPLAT_SPLAT46:%.*]] = shufflevector <1 x double> [[SPLAT_SPLATINSERT45]], <1 x double> poison, <1 x i32> zeroinitializer
-; RM-NEXT:    [[TMP50:%.*]] = fmul <1 x double> [[SPLAT_SPLAT46]], [[BLOCK44]]
-; RM-NEXT:    [[BLOCK47:%.*]] = shufflevector <3 x double> [[SPLIT4]], <3 x double> poison, <1 x i32> <i32 1>
-; RM-NEXT:    [[TMP51:%.*]] = extractelement <2 x double> [[SPLIT2]], i64 1
-; RM-NEXT:    [[SPLAT_SPLATINSERT48:%.*]] = insertelement <1 x double> poison, double [[TMP51]], i32 0
-; RM-NEXT:    [[SPLAT_SPLAT49:%.*]] = shufflevector <1 x double> [[SPLAT_SPLATINSERT48]], <1 x double> poison, <1 x i32> zeroinitializer
-; RM-NEXT:    [[TMP52:%.*]] = fmul <1 x double> [[SPLAT_SPLAT49]], [[BLOCK47]]
-; RM-NEXT:    [[TMP53:%.*]] = fadd <1 x double> [[TMP50]], [[TMP52]]
-; RM-NEXT:    [[TMP54:%.*]] = shufflevector <1 x double> [[TMP53]], <1 x double> poison, <3 x i32> <i32 0, i32 undef, i32 undef>
-; RM-NEXT:    [[TMP55:%.*]] = shufflevector <3 x double> [[TMP48]], <3 x double> [[TMP54]], <3 x i32> <i32 0, i32 3, i32 2>
-; RM-NEXT:    [[BLOCK50:%.*]] = shufflevector <3 x double> [[SPLIT3]], <3 x double> poison, <1 x i32> <i32 2>
-; RM-NEXT:    [[TMP56:%.*]] = extractelement <2 x double> [[SPLIT2]], i64 0
-; RM-NEXT:    [[SPLAT_SPLATINSERT51:%.*]] = insertelement <1 x double> poison, double [[TMP56]], i32 0
-; RM-NEXT:    [[SPLAT_SPLAT52:%.*]] = shufflevector <1 x double> [[SPLAT_SPLATINSERT51]], <1 x double> poison, <1 x i32> zeroinitializer
-; RM-NEXT:    [[TMP57:%.*]] = fmul <1 x double> [[SPLAT_SPLAT52]], [[BLOCK50]]
-; RM-NEXT:    [[BLOCK53:%.*]] = shufflevector <3 x double> [[SPLIT4]], <3 x double> poison, <1 x i32> <i32 2>
-; RM-NEXT:    [[TMP58:%.*]] = extractelement <2 x double> [[SPLIT2]], i64 1
-; RM-NEXT:    [[SPLAT_SPLATINSERT54:%.*]] = insertelement <1 x double> poison, double [[TMP58]], i32 0
-; RM-NEXT:    [[SPLAT_SPLAT55:%.*]] = shufflevector <1 x double> [[SPLAT_SPLATINSERT54]], <1 x double> poison, <1 x i32> zeroinitializer
-; RM-NEXT:    [[TMP59:%.*]] = fmul <1 x double> [[SPLAT_SPLAT55]], [[BLOCK53]]
-; RM-NEXT:    [[TMP60:%.*]] = fadd <1 x double> [[TMP57]], [[TMP59]]
-; RM-NEXT:    [[TMP61:%.*]] = shufflevector <1 x double> [[TMP60]], <1 x double> poison, <3 x i32> <i32 0, i32 undef, i32 undef>
-; RM-NEXT:    [[TMP62:%.*]] = shufflevector <3 x double> [[TMP55]], <3 x double> [[TMP61]], <3 x i32> <i32 0, i32 1, i32 3>
-; RM-NEXT:    [[TMP63:%.*]] = shufflevector <3 x double> [[TMP20]], <3 x double> [[TMP41]], <6 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5>
-; RM-NEXT:    [[TMP64:%.*]] = shufflevector <3 x double> [[TMP62]], <3 x double> poison, <6 x i32> <i32 0, i32 1, i32 2, i32 undef, i32 undef, i32 undef>
-; RM-NEXT:    [[TMP65:%.*]] = shufflevector <6 x double> [[TMP63]], <6 x double> [[TMP64]], <9 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7, i32 8>
-; RM-NEXT:    ret <9 x double> [[TMP65]]
+; CHECK-LABEL: @multiply_2x3(
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[SPLIT:%.*]] = shufflevector <6 x double> [[A:%.*]], <6 x double> poison, <2 x i32> <i32 0, i32 1>
+; CHECK-NEXT:    [[SPLIT1:%.*]] = shufflevector <6 x double> [[A]], <6 x double> poison, <2 x i32> <i32 2, i32 3>
+; CHECK-NEXT:    [[SPLIT2:%.*]] = shufflevector <6 x double> [[A]], <6 x double> poison, <2 x i32> <i32 4, i32 5>
+; CHECK-NEXT:    [[SPLIT3:%.*]] = shufflevector <6 x double> [[B:%.*]], <6 x double> poison, <3 x i32> <i32 0, i32 1, i32 2>
+; CHECK-NEXT:    [[SPLIT4:%.*]] = shufflevector <6 x double> [[B]], <6 x double> poison, <3 x i32> <i32 3, i32 4, i32 5>
+; CHECK-NEXT:    [[BLOCK:%.*]] = shufflevector <3 x double> [[SPLIT3]], <3 x double> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP0:%.*]] = extractelement <2 x double> [[SPLIT]], i64 0
+; CHECK-NEXT:    [[SPLAT_SPLATINSERT:%.*]] = insertelement <1 x double> poison, double [[TMP0]], i32 0
+; CHECK-NEXT:    [[SPLAT_SPLAT:%.*]] = shufflevector <1 x double> [[SPLAT_SPLATINSERT]], <1 x double> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP1:%.*]] = fmul <1 x double> [[SPLAT_SPLAT]], [[BLOCK]]
+; CHECK-NEXT:    [[BLOCK5:%.*]] = shufflevector <3 x double> [[SPLIT4]], <3 x double> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP2:%.*]] = extractelement <2 x double> [[SPLIT]], i64 1
+; CHECK-NEXT:    [[SPLAT_SPLATINSERT6:%.*]] = insertelement <1 x double> poison, double [[TMP2]], i32 0
+; CHECK-NEXT:    [[SPLAT_SPLAT7:%.*]] = shufflevector <1 x double> [[SPLAT_SPLATINSERT6]], <1 x double> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP3:%.*]] = fmul <1 x double> [[SPLAT_SPLAT7]], [[BLOCK5]]
+; CHECK-NEXT:    [[TMP4:%.*]] = fadd <1 x double> [[TMP1]], [[TMP3]]
+; CHECK-NEXT:    [[TMP5:%.*]] = shufflevector <1 x double> [[TMP4]], <1 x double> poison, <3 x i32> <i32 0, i32 undef, i32 undef>
+; CHECK-NEXT:    [[TMP6:%.*]] = shufflevector <3 x double> undef, <3 x double> [[TMP5]], <3 x i32> <i32 3, i32 1, i32 2>
+; CHECK-NEXT:    [[BLOCK8:%.*]] = shufflevector <3 x double> [[SPLIT3]], <3 x double> poison, <1 x i32> <i32 1>
+; CHECK-NEXT:    [[TMP7:%.*]] = extractelement <2 x double> [[SPLIT]], i64 0
+; CHECK-NEXT:    [[SPLAT_SPLATINSERT9:%.*]] = insertelement <1 x double> poison, double [[TMP7]], i32 0
+; CHECK-NEXT:    [[SPLAT_SPLAT10:%.*]] = shufflevector <1 x double> [[SPLAT_SPLATINSERT9]], <1 x double> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP8:%.*]] = fmul <1 x double> [[SPLAT_SPLAT10]], [[BLOCK8]]
+; CHECK-NEXT:    [[BLOCK11:%.*]] = shufflevector <3 x double> [[SPLIT4]], <3 x double> poison, <1 x i32> <i32 1>
+; CHECK-NEXT:    [[TMP9:%.*]] = extractelement <2 x double> [[SPLIT]], i64 1
+; CHECK-NEXT:    [[SPLAT_SPLATINSERT12:%.*]] = insertelement <1 x double> poison, double [[TMP9]], i32 0
+; CHECK-NEXT:    [[SPLAT_SPLAT13:%.*]] = shufflevector <1 x double> [[SPLAT_SPLATINSERT12]], <1 x double> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP10:%.*]] = fmul <1 x double> [[SPLAT_SPLAT13]], [[BLOCK11]]
+; CHECK-NEXT:    [[TMP11:%.*]] = fadd <1 x double> [[TMP8]], [[TMP10]]
+; CHECK-NEXT:    [[TMP12:%.*]] = shufflevector <1 x double> [[TMP11]], <1 x double> poison, <3 x i32> <i32 0, i32 undef, i32 undef>
+; CHECK-NEXT:    [[TMP13:%.*]] = shufflevector <3 x double> [[TMP6]], <3 x double> [[TMP12]], <3 x i32> <i32 0, i32 3, i32 2>
+; CHECK-NEXT:    [[BLOCK14:%.*]] = shufflevector <3 x double> [[SPLIT3]], <3 x double> poison, <1 x i32> <i32 2>
+; CHECK-NEXT:    [[TMP14:%.*]] = extractelement <2 x double> [[SPLIT]], i64 0
+; CHECK-NEXT:    [[SPLAT_SPLATINSERT15:%.*]] = insertelement <1 x double> poison, double [[TMP14]], i32 0
+; CHECK-NEXT:    [[SPLAT_SPLAT16:%.*]] = shufflevector <1 x double> [[SPLAT_SPLATINSERT15]], <1 x double> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP15:%.*]] = fmul <1 x double> [[SPLAT_SPLAT16]], [[BLOCK14]]
+; CHECK-NEXT:    [[BLOCK17:%.*]] = shufflevector <3 x double> [[SPLIT4]], <3 x double> poison, <1 x i32> <i32 2>
+; CHECK-NEXT:    [[TMP16:%.*]] = extractelement <2 x double> [[SPLIT]], i64 1
+; CHECK-NEXT:    [[SPLAT_SPLATINSERT18:%.*]] = insertelement <1 x double> poison, double [[TMP16]], i32 0
+; CHECK-NEXT:    [[SPLAT_SPLAT19:%.*]] = shufflevector <1 x double> [[SPLAT_SPLATINSERT18]], <1 x double> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP17:%.*]] = fmul <1 x double> [[SPLAT_SPLAT19]], [[BLOCK17]]
+; CHECK-NEXT:    [[TMP18:%.*]] = fadd <1 x double> [[TMP15]], [[TMP17]]
+; CHECK-NEXT:    [[TMP19:%.*]] = shufflevector <1 x double> [[TMP18]], <1 x double> poison, <3 x i32> <i32 0, i32 undef, i32 undef>
+; CHECK-NEXT:    [[TMP20:%.*]] = shufflevector <3 x double> [[TMP13]], <3 x double> [[TMP19]], <3 x i32> <i32 0, i32 1, i32 3>
+; CHECK-NEXT:    [[BLOCK20:%.*]] = shufflevector <3 x double> [[SPLIT3]], <3 x double> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP21:%.*]] = extractelement <2 x double> [[SPLIT1]], i64 0
+; CHECK-NEXT:    [[SPLAT_SPLATINSERT21:%.*]] = insertelement <1 x double> poison, double [[TMP21]], i32 0
+; CHECK-NEXT:    [[SPLAT_SPLAT22:%.*]] = shufflevector <1 x double> [[SPLAT_SPLATINSERT21]], <1 x double> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP22:%.*]] = fmul <1 x double> [[SPLAT_SPLAT22]], [[BLOCK20]]
+; CHECK-NEXT:    [[BLOCK23:%.*]] = shufflevector <3 x double> [[SPLIT4]], <3 x double> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP23:%.*]] = extractelement <2 x double> [[SPLIT1]], i64 1
+; CHECK-NEXT:    [[SPLAT_SPLATINSERT24:%.*]] = insertelement <1 x double> poison, double [[TMP23]], i32 0
+; CHECK-NEXT:    [[SPLAT_SPLAT25:%.*]] = shufflevector <1 x double> [[SPLAT_SPLATINSERT24]], <1 x double> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP24:%.*]] = fmul <1 x double> [[SPLAT_SPLAT25]], [[BLOCK23]]
+; CHECK-NEXT:    [[TMP25:%.*]] = fadd <1 x double> [[TMP22]], [[TMP24]]
+; CHECK-NEXT:    [[TMP26:%.*]] = shufflevector <1 x double> [[TMP25]], <1 x double> poison, <3 x i32> <i32 0, i32 undef, i32 undef>
+; CHECK-NEXT:    [[TMP27:%.*]] = shufflevector <3 x double> undef, <3 x double> [[TMP26]], <3 x i32> <i32 3, i32 1, i32 2>
+; CHECK-NEXT:    [[BLOCK26:%.*]] = shufflevector <3 x double> [[SPLIT3]], <3 x double> poison, <1 x i32> <i32 1>
+; CHECK-NEXT:    [[TMP28:%.*]] = extractelement <2 x double> [[SPLIT1]], i64 0
+; CHECK-NEXT:    [[SPLAT_SPLATINSERT27:%.*]] = insertelement <1 x double> poison, double [[TMP28]], i32 0
+; CHECK-NEXT:    [[SPLAT_SPLAT28:%.*]] = shufflevector <1 x double> [[SPLAT_SPLATINSERT27]], <1 x double> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP29:%.*]] = fmul <1 x double> [[SPLAT_SPLAT28]], [[BLOCK26]]
+; CHECK-NEXT:    [[BLOCK29:%.*]] = shufflevector <3 x double> [[SPLIT4]], <3 x double> poison, <1 x i32> <i32 1>
+; CHECK-NEXT:    [[TMP30:%.*]] = extractelement <2 x double> [[SPLIT1]], i64 1
+; CHECK-NEXT:    [[SPLAT_SPLATINSERT30:%.*]] = insertelement <1 x double> poison, double [[TMP30]], i32 0
+; CHECK-NEXT:    [[SPLAT_SPLAT31:%.*]] = shufflevector <1 x double> [[SPLAT_SPLATINSERT30]], <1 x double> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP31:%.*]] = fmul <1 x double> [[SPLAT_SPLAT31]], [[BLOCK29]]
+; CHECK-NEXT:    [[TMP32:%.*]] = fadd <1 x double> [[TMP29]], [[TMP31]]
+; CHECK-NEXT:    [[TMP33:%.*]] = shufflevector <1 x double> [[TMP32]], <1 x double> poison, <3 x i32> <i32 0, i32 undef, i32 undef>
+; CHECK-NEXT:    [[TMP34:%.*]] = shufflevector <3 x double> [[TMP27]], <3 x double> [[TMP33]], <3 x i32> <i32 0, i32 3, i32 2>
+; CHECK-NEXT:    [[BLOCK32:%.*]] = shufflevector <3 x double> [[SPLIT3]], <3 x double> poison, <1 x i32> <i32 2>
+; CHECK-NEXT:    [[TMP35:%.*]] = extractelement <2 x double> [[SPLIT1]], i64 0
+; CHECK-NEXT:    [[SPLAT_SPLATINSERT33:%.*]] = insertelement <1 x double> poison, double [[TMP35]], i32 0
+; CHECK-NEXT:    [[SPLAT_SPLAT34:%.*]] = shufflevector <1 x double> [[SPLAT_SPLATINSERT33]], <1 x double> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP36:%.*]] = fmul <1 x double> [[SPLAT_SPLAT34]], [[BLOCK32]]
+; CHECK-NEXT:    [[BLOCK35:%.*]] = shufflevector <3 x double> [[SPLIT4]], <3 x double> poison, <1 x i32> <i32 2>
+; CHECK-NEXT:    [[TMP37:%.*]] = extractelement <2 x double> [[SPLIT1]], i64 1
+; CHECK-NEXT:    [[SPLAT_SPLATINSERT36:%.*]] = insertelement <1 x double> poison, double [[TMP37]], i32 0
+; CHECK-NEXT:    [[SPLAT_SPLAT37:%.*]] = shufflevector <1 x double> [[SPLAT_SPLATINSERT36]], <1 x double> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP38:%.*]] = fmul <1 x double> [[SPLAT_SPLAT37]], [[BLOCK35]]
+; CHECK-NEXT:    [[TMP39:%.*]] = fadd <1 x double> [[TMP36]], [[TMP38]]
+; CHECK-NEXT:    [[TMP40:%.*]] = shufflevector <1 x double> [[TMP39]], <1 x double> poison, <3 x i32> <i32 0, i32 undef, i32 undef>
+; CHECK-NEXT:    [[TMP41:%.*]] = shufflevector <3 x double> [[TMP34]], <3 x double> [[TMP40]], <3 x i32> <i32 0, i32 1, i32 3>
+; CHECK-NEXT:    [[BLOCK38:%.*]] = shufflevector <3 x double> [[SPLIT3]], <3 x double> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP42:%.*]] = extractelement <2 x double> [[SPLIT2]], i64 0
+; CHECK-NEXT:    [[SPLAT_SPLATINSERT39:%.*]] = insertelement <1 x double> poison, double [[TMP42]], i32 0
+; CHECK-NEXT:    [[SPLAT_SPLAT40:%.*]] = shufflevector <1 x double> [[SPLAT_SPLATINSERT39]], <1 x double> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP43:%.*]] = fmul <1 x double> [[SPLAT_SPLAT40]], [[BLOCK38]]
+; CHECK-NEXT:    [[BLOCK41:%.*]] = shufflevector <3 x double> [[SPLIT4]], <3 x double> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP44:%.*]] = extractelement <2 x double> [[SPLIT2]], i64 1
+; CHECK-NEXT:    [[SPLAT_SPLATINSERT42:%.*]] = insertelement <1 x double> poison, double [[TMP44]], i32 0
+; CHECK-NEXT:    [[SPLAT_SPLAT43:%.*]] = shufflevector <1 x double> [[SPLAT_SPLATINSERT42]], <1 x double> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP45:%.*]] = fmul <1 x double> [[SPLAT_SPLAT43]], [[BLOCK41]]
+; CHECK-NEXT:    [[TMP46:%.*]] = fadd <1 x double> [[TMP43]], [[TMP45]]
+; CHECK-NEXT:    [[TMP47:%.*]] = shufflevector <1 x double> [[TMP46]], <1 x double> poison, <3 x i32> <i32 0, i32 undef, i32 undef>
+; CHECK-NEXT:    [[TMP48:%.*]] = shufflevector <3 x double> undef, <3 x double> [[TMP47]], <3 x i32> <i32 3, i32 1, i32 2>
+; CHECK-NEXT:    [[BLOCK44:%.*]] = shufflevector <3 x double> [[SPLIT3]], <3 x double> poison, <1 x i32> <i32 1>
+; CHECK-NEXT:    [[TMP49:%.*]] = extractelement <2 x double> [[SPLIT2]], i64 0
+; CHECK-NEXT:    [[SPLAT_SPLATINSERT45:%.*]] = insertelement <1 x double> poison, double [[TMP49]], i32 0
+; CHECK-NEXT:    [[SPLAT_SPLAT46:%.*]] = shufflevector <1 x double> [[SPLAT_SPLATINSERT45]], <1 x double> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP50:%.*]] = fmul <1 x double> [[SPLAT_SPLAT46]], [[BLOCK44]]
+; CHECK-NEXT:    [[BLOCK47:%.*]] = shufflevector <3 x double> [[SPLIT4]], <3 x double> poison, <1 x i32> <i32 1>
+; CHECK-NEXT:    [[TMP51:%.*]] = extractelement <2 x double> [[SPLIT2]], i64 1
+; CHECK-NEXT:    [[SPLAT_SPLATINSERT48:%.*]] = insertelement <1 x double> poison, double [[TMP51]], i32 0
+; CHECK-NEXT:    [[SPLAT_SPLAT49:%.*]] = shufflevector <1 x double> [[SPLAT_SPLATINSERT48]], <1 x double> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP52:%.*]] = fmul <1 x double> [[SPLAT_SPLAT49]], [[BLOCK47]]
+; CHECK-NEXT:    [[TMP53:%.*]] = fadd <1 x double> [[TMP50]], [[TMP52]]
+; CHECK-NEXT:    [[TMP54:%.*]] = shufflevector <1 x double> [[TMP53]], <1 x double> poison, <3 x i32> <i32 0, i32 undef, i32 undef>
+; CHECK-NEXT:    [[TMP55:%.*]] = shufflevector <3 x double> [[TMP48]], <3 x double> [[TMP54]], <3 x i32> <i32 0, i32 3, i32 2>
+; CHECK-NEXT:    [[BLOCK50:%.*]] = shufflevector <3 x double> [[SPLIT3]], <3 x double> poison, <1 x i32> <i32 2>
+; CHECK-NEXT:    [[TMP56:%.*]] = extractelement <2 x double> [[SPLIT2]], i64 0
+; CHECK-NEXT:    [[SPLAT_SPLATINSERT51:%.*]] = insertelement <1 x double> poison, double [[TMP56]], i32 0
+; CHECK-NEXT:    [[SPLAT_SPLAT52:%.*]] = shufflevector <1 x double> [[SPLAT_SPLATINSERT51]], <1 x double> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP57:%.*]] = fmul <1 x double> [[SPLAT_SPLAT52]], [[BLOCK50]]
+; CHECK-NEXT:    [[BLOCK53:%.*]] = shufflevector <3 x double> [[SPLIT4]], <3 x double> poison, <1 x i32> <i32 2>
+; CHECK-NEXT:    [[TMP58:%.*]] = extractelement <2 x double> [[SPLIT2]], i64 1
+; CHECK-NEXT:    [[SPLAT_SPLATINSERT54:%.*]] = insertelement <1 x double> poison, double [[TMP58]], i32 0
+; CHECK-NEXT:    [[SPLAT_SPLAT55:%.*]] = shufflevector <1 x double> [[SPLAT_SPLATINSERT54]], <1 x double> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP59:%.*]] = fmul <1 x double> [[SPLAT_SPLAT55]], [[BLOCK53]]
+; CHECK-NEXT:    [[TMP60:%.*]] = fadd <1 x double> [[TMP57]], [[TMP59]]
+; CHECK-NEXT:    [[TMP61:%.*]] = shufflevector <1 x double> [[TMP60]], <1 x double> poison, <3 x i32> <i32 0, i32 undef, i32 undef>
+; CHECK-NEXT:    [[TMP62:%.*]] = shufflevector <3 x double> [[TMP55]], <3 x double> [[TMP61]], <3 x i32> <i32 0, i32 1, i32 3>
+; CHECK-NEXT:    [[TMP63:%.*]] = shufflevector <3 x double> [[TMP20]], <3 x double> [[TMP41]], <6 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5>
+; CHECK-NEXT:    [[TMP64:%.*]] = shufflevector <3 x double> [[TMP62]], <3 x double> poison, <6 x i32> <i32 0, i32 1, i32 2, i32 undef, i32 undef, i32 undef>
+; CHECK-NEXT:    [[TMP65:%.*]] = shufflevector <6 x double> [[TMP63]], <6 x double> [[TMP64]], <9 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7, i32 8>
+; CHECK-NEXT:    ret <9 x double> [[TMP65]]
 ;
 entry:
   %c = call <9 x double> @llvm.matrix.multiply.v6f64.v6f64.v6f64(<6 x double> %a, <6 x double> %b, i32 3, i32 2, i32 3)
   ret <9 x double> %c
 }
 
 declare <9 x double> @llvm.matrix.multiply.v6f64.v6f64.v6f64(<6 x double>, <6 x double>, i32, i32, i32)

llvm/test/Transforms/LowerMatrixIntrinsics/multiply-left-transpose-row-major.ll

+; NOTE: Assertions have been autogenerated by utils/update_test_checks.py
+
+; RUN: opt -lower-matrix-intrinsics -matrix-default-layout=row-major -S < %s | FileCheck %s
+
+define <4 x double> @multiply_left_transpose_2x2(<4 x double> %a, <4 x double> %b) {
+; CHECK-LABEL: @multiply_left_transpose_2x2(
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[SPLIT:%.*]] = shufflevector <4 x double> [[A:%.*]], <4 x double> poison, <2 x i32> <i32 0, i32 1>
+; CHECK-NEXT:    [[SPLIT1:%.*]] = shufflevector <4 x double> [[A]], <4 x double> poison, <2 x i32> <i32 2, i32 3>
+; CHECK-NEXT:    [[SPLIT2:%.*]] = shufflevector <4 x double> [[B:%.*]], <4 x double> poison, <2 x i32> <i32 0, i32 1>
+; CHECK-NEXT:    [[SPLIT3:%.*]] = shufflevector <4 x double> [[B]], <4 x double> poison, <2 x i32> <i32 2, i32 3>
+; CHECK-NEXT:    [[BLOCK:%.*]] = shufflevector <2 x double> [[SPLIT2]], <2 x double> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP0:%.*]] = extractelement <2 x double> [[SPLIT]], i64 0
+; CHECK-NEXT:    [[SPLAT_SPLATINSERT:%.*]] = insertelement <1 x double> poison, double [[TMP0]], i32 0
+; CHECK-NEXT:    [[SPLAT_SPLAT:%.*]] = shufflevector <1 x double> [[SPLAT_SPLATINSERT]], <1 x double> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP1:%.*]] = fmul <1 x double> [[SPLAT_SPLAT]], [[BLOCK]]
+; CHECK-NEXT:    [[BLOCK4:%.*]] = shufflevector <2 x double> [[SPLIT3]], <2 x double> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP2:%.*]] = extractelement <2 x double> [[SPLIT1]], i64 0
+; CHECK-NEXT:    [[SPLAT_SPLATINSERT5:%.*]] = insertelement <1 x double> poison, double [[TMP2]], i32 0
+; CHECK-NEXT:    [[SPLAT_SPLAT6:%.*]] = shufflevector <1 x double> [[SPLAT_SPLATINSERT5]], <1 x double> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP3:%.*]] = fmul <1 x double> [[SPLAT_SPLAT6]], [[BLOCK4]]
+; CHECK-NEXT:    [[TMP4:%.*]] = fadd <1 x double> [[TMP1]], [[TMP3]]
+; CHECK-NEXT:    [[TMP5:%.*]] = shufflevector <1 x double> [[TMP4]], <1 x double> poison, <2 x i32> <i32 0, i32 undef>
+; CHECK-NEXT:    [[TMP6:%.*]] = shufflevector <2 x double> undef, <2 x double> [[TMP5]], <2 x i32> <i32 2, i32 1>
+; CHECK-NEXT:    [[BLOCK7:%.*]] = shufflevector <2 x double> [[SPLIT2]], <2 x double> poison, <1 x i32> <i32 1>
+; CHECK-NEXT:    [[TMP7:%.*]] = extractelement <2 x double> [[SPLIT]], i64 0
+; CHECK-NEXT:    [[SPLAT_SPLATINSERT8:%.*]] = insertelement <1 x double> poison, double [[TMP7]], i32 0
+; CHECK-NEXT:    [[SPLAT_SPLAT9:%.*]] = shufflevector <1 x double> [[SPLAT_SPLATINSERT8]], <1 x double> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP8:%.*]] = fmul <1 x double> [[SPLAT_SPLAT9]], [[BLOCK7]]
+; CHECK-NEXT:    [[BLOCK10:%.*]] = shufflevector <2 x double> [[SPLIT3]], <2 x double> poison, <1 x i32> <i32 1>
+; CHECK-NEXT:    [[TMP9:%.*]] = extractelement <2 x double> [[SPLIT1]], i64 0
+; CHECK-NEXT:    [[SPLAT_SPLATINSERT11:%.*]] = insertelement <1 x double> poison, double [[TMP9]], i32 0
+; CHECK-NEXT:    [[SPLAT_SPLAT12:%.*]] = shufflevector <1 x double> [[SPLAT_SPLATINSERT11]], <1 x double> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP10:%.*]] = fmul <1 x double> [[SPLAT_SPLAT12]], [[BLOCK10]]
+; CHECK-NEXT:    [[TMP11:%.*]] = fadd <1 x double> [[TMP8]], [[TMP10]]
+; CHECK-NEXT:    [[TMP12:%.*]] = shufflevector <1 x double> [[TMP11]], <1 x double> poison, <2 x i32> <i32 0, i32 undef>
+; CHECK-NEXT:    [[TMP13:%.*]] = shufflevector <2 x double> [[TMP6]], <2 x double> [[TMP12]], <2 x i32> <i32 0, i32 2>
+; CHECK-NEXT:    [[BLOCK13:%.*]] = shufflevector <2 x double> [[SPLIT2]], <2 x double> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP14:%.*]] = extractelement <2 x double> [[SPLIT]], i64 1
+; CHECK-NEXT:    [[SPLAT_SPLATINSERT14:%.*]] = insertelement <1 x double> poison, double [[TMP14]], i32 0
+; CHECK-NEXT:    [[SPLAT_SPLAT15:%.*]] = shufflevector <1 x double> [[SPLAT_SPLATINSERT14]], <1 x double> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP15:%.*]] = fmul <1 x double> [[SPLAT_SPLAT15]], [[BLOCK13]]
+; CHECK-NEXT:    [[BLOCK16:%.*]] = shufflevector <2 x double> [[SPLIT3]], <2 x double> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP16:%.*]] = extractelement <2 x double> [[SPLIT1]], i64 1
+; CHECK-NEXT:    [[SPLAT_SPLATINSERT17:%.*]] = insertelement <1 x double> poison, double [[TMP16]], i32 0
+; CHECK-NEXT:    [[SPLAT_SPLAT18:%.*]] = shufflevector <1 x double> [[SPLAT_SPLATINSERT17]], <1 x double> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP17:%.*]] = fmul <1 x double> [[SPLAT_SPLAT18]], [[BLOCK16]]
+; CHECK-NEXT:    [[TMP18:%.*]] = fadd <1 x double> [[TMP15]], [[TMP17]]
+; CHECK-NEXT:    [[TMP19:%.*]] = shufflevector <1 x double> [[TMP18]], <1 x double> poison, <2 x i32> <i32 0, i32 undef>
+; CHECK-NEXT:    [[TMP20:%.*]] = shufflevector <2 x double> undef, <2 x double> [[TMP19]], <2 x i32> <i32 2, i32 1>
+; CHECK-NEXT:    [[BLOCK19:%.*]] = shufflevector <2 x double> [[SPLIT2]], <2 x double> poison, <1 x i32> <i32 1>
+; CHECK-NEXT:    [[TMP21:%.*]] = extractelement <2 x double> [[SPLIT]], i64 1
+; CHECK-NEXT:    [[SPLAT_SPLATINSERT20:%.*]] = insertelement <1 x double> poison, double [[TMP21]], i32 0
+; CHECK-NEXT:    [[SPLAT_SPLAT21:%.*]] = shufflevector <1 x double> [[SPLAT_SPLATINSERT20]], <1 x double> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP22:%.*]] = fmul <1 x double> [[SPLAT_SPLAT21]], [[BLOCK19]]
+; CHECK-NEXT:    [[BLOCK22:%.*]] = shufflevector <2 x double> [[SPLIT3]], <2 x double> poison, <1 x i32> <i32 1>
+; CHECK-NEXT:    [[TMP23:%.*]] = extractelement <2 x double> [[SPLIT1]], i64 1
+; CHECK-NEXT:    [[SPLAT_SPLATINSERT23:%.*]] = insertelement <1 x double> poison, double [[TMP23]], i32 0
+; CHECK-NEXT:    [[SPLAT_SPLAT24:%.*]] = shufflevector <1 x double> [[SPLAT_SPLATINSERT23]], <1 x double> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP24:%.*]] = fmul <1 x double> [[SPLAT_SPLAT24]], [[BLOCK22]]
+; CHECK-NEXT:    [[TMP25:%.*]] = fadd <1 x double> [[TMP22]], [[TMP24]]
+; CHECK-NEXT:    [[TMP26:%.*]] = shufflevector <1 x double> [[TMP25]], <1 x double> poison, <2 x i32> <i32 0, i32 undef>
+; CHECK-NEXT:    [[TMP27:%.*]] = shufflevector <2 x double> [[TMP20]], <2 x double> [[TMP26]], <2 x i32> <i32 0, i32 2>
+; CHECK-NEXT:    [[TMP28:%.*]] = shufflevector <2 x double> [[TMP13]], <2 x double> [[TMP27]], <4 x i32> <i32 0, i32 1, i32 2, i32 3>
+; CHECK-NEXT:    ret <4 x double> [[TMP28]]
+;
+entry:
+  %at = call <4 x double> @llvm.matrix.transpose.v4f64(<4 x double> %a, i32 2, i32 2)
+  %c = call <4 x double> @llvm.matrix.multiply.v4f64.v4f64.v4f64(<4 x double> %at, <4 x double> %b, i32 2, i32 2, i32 2)
+  ret <4 x double> %c
+}
+
+declare <4 x double> @llvm.matrix.multiply.v4f64.v4f64.v4f64(<4 x double>, <4 x double>, i32, i32, i32)
+declare <4 x double> @llvm.matrix.transpose.v4f64(<4 x double>, i32, i32)

llvm/test/Transforms/LowerMatrixIntrinsics/multiply-right-transpose.ll

+; NOTE: Assertions have been autogenerated by utils/update_test_checks.py
+; RUN: opt -lower-matrix-intrinsics -S < %s | FileCheck %s
+; RUN: opt -passes='lower-matrix-intrinsics' -S < %s | FileCheck %s
+
+define <4 x double> @multiply_right_transpose_2x2(<4 x double> %a, <4 x double> %b) {
+; CHECK-LABEL: @multiply_right_transpose_2x2(
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[SPLIT:%.*]] = shufflevector <4 x double> [[A:%.*]], <4 x double> poison, <2 x i32> <i32 0, i32 1>
+; CHECK-NEXT:    [[SPLIT1:%.*]] = shufflevector <4 x double> [[A]], <4 x double> poison, <2 x i32> <i32 2, i32 3>
+; CHECK-NEXT:    [[SPLIT2:%.*]] = shufflevector <4 x double> [[B:%.*]], <4 x double> poison, <2 x i32> <i32 0, i32 1>
+; CHECK-NEXT:    [[SPLIT3:%.*]] = shufflevector <4 x double> [[B]], <4 x double> poison, <2 x i32> <i32 2, i32 3>
+; CHECK-NEXT:    [[BLOCK:%.*]] = shufflevector <2 x double> [[SPLIT]], <2 x double> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP0:%.*]] = extractelement <2 x double> [[SPLIT2]], i64 0
+; CHECK-NEXT:    [[SPLAT_SPLATINSERT:%.*]] = insertelement <1 x double> poison, double [[TMP0]], i32 0
+; CHECK-NEXT:    [[SPLAT_SPLAT:%.*]] = shufflevector <1 x double> [[SPLAT_SPLATINSERT]], <1 x double> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP1:%.*]] = fmul <1 x double> [[BLOCK]], [[SPLAT_SPLAT]]
+; CHECK-NEXT:    [[BLOCK4:%.*]] = shufflevector <2 x double> [[SPLIT1]], <2 x double> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP2:%.*]] = extractelement <2 x double> [[SPLIT3]], i64 0
+; CHECK-NEXT:    [[SPLAT_SPLATINSERT5:%.*]] = insertelement <1 x double> poison, double [[TMP2]], i32 0
+; CHECK-NEXT:    [[SPLAT_SPLAT6:%.*]] = shufflevector <1 x double> [[SPLAT_SPLATINSERT5]], <1 x double> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP3:%.*]] = fmul <1 x double> [[BLOCK4]], [[SPLAT_SPLAT6]]
+; CHECK-NEXT:    [[TMP4:%.*]] = fadd <1 x double> [[TMP1]], [[TMP3]]
+; CHECK-NEXT:    [[TMP5:%.*]] = shufflevector <1 x double> [[TMP4]], <1 x double> poison, <2 x i32> <i32 0, i32 undef>
+; CHECK-NEXT:    [[TMP6:%.*]] = shufflevector <2 x double> undef, <2 x double> [[TMP5]], <2 x i32> <i32 2, i32 1>
+; CHECK-NEXT:    [[BLOCK7:%.*]] = shufflevector <2 x double> [[SPLIT]], <2 x double> poison, <1 x i32> <i32 1>
+; CHECK-NEXT:    [[TMP7:%.*]] = extractelement <2 x double> [[SPLIT2]], i64 0
+; CHECK-NEXT:    [[SPLAT_SPLATINSERT8:%.*]] = insertelement <1 x double> poison, double [[TMP7]], i32 0
+; CHECK-NEXT:    [[SPLAT_SPLAT9:%.*]] = shufflevector <1 x double> [[SPLAT_SPLATINSERT8]], <1 x double> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP8:%.*]] = fmul <1 x double> [[BLOCK7]], [[SPLAT_SPLAT9]]
+; CHECK-NEXT:    [[BLOCK10:%.*]] = shufflevector <2 x double> [[SPLIT1]], <2 x double> poison, <1 x i32> <i32 1>
+; CHECK-NEXT:    [[TMP9:%.*]] = extractelement <2 x double> [[SPLIT3]], i64 0
+; CHECK-NEXT:    [[SPLAT_SPLATINSERT11:%.*]] = insertelement <1 x double> poison, double [[TMP9]], i32 0
+; CHECK-NEXT:    [[SPLAT_SPLAT12:%.*]] = shufflevector <1 x double> [[SPLAT_SPLATINSERT11]], <1 x double> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP10:%.*]] = fmul <1 x double> [[BLOCK10]], [[SPLAT_SPLAT12]]
+; CHECK-NEXT:    [[TMP11:%.*]] = fadd <1 x double> [[TMP8]], [[TMP10]]
+; CHECK-NEXT:    [[TMP12:%.*]] = shufflevector <1 x double> [[TMP11]], <1 x double> poison, <2 x i32> <i32 0, i32 undef>
+; CHECK-NEXT:    [[TMP13:%.*]] = shufflevector <2 x double> [[TMP6]], <2 x double> [[TMP12]], <2 x i32> <i32 0, i32 2>
+; CHECK-NEXT:    [[BLOCK13:%.*]] = shufflevector <2 x double> [[SPLIT]], <2 x double> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP14:%.*]] = extractelement <2 x double> [[SPLIT2]], i64 1
+; CHECK-NEXT:    [[SPLAT_SPLATINSERT14:%.*]] = insertelement <1 x double> poison, double [[TMP14]], i32 0
+; CHECK-NEXT:    [[SPLAT_SPLAT15:%.*]] = shufflevector <1 x double> [[SPLAT_SPLATINSERT14]], <1 x double> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP15:%.*]] = fmul <1 x double> [[BLOCK13]], [[SPLAT_SPLAT15]]
+; CHECK-NEXT:    [[BLOCK16:%.*]] = shufflevector <2 x double> [[SPLIT1]], <2 x double> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP16:%.*]] = extractelement <2 x double> [[SPLIT3]], i64 1
+; CHECK-NEXT:    [[SPLAT_SPLATINSERT17:%.*]] = insertelement <1 x double> poison, double [[TMP16]], i32 0
+; CHECK-NEXT:    [[SPLAT_SPLAT18:%.*]] = shufflevector <1 x double> [[SPLAT_SPLATINSERT17]], <1 x double> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP17:%.*]] = fmul <1 x double> [[BLOCK16]], [[SPLAT_SPLAT18]]
+; CHECK-NEXT:    [[TMP18:%.*]] = fadd <1 x double> [[TMP15]], [[TMP17]]
+; CHECK-NEXT:    [[TMP19:%.*]] = shufflevector <1 x double> [[TMP18]], <1 x double> poison, <2 x i32> <i32 0, i32 undef>
+; CHECK-NEXT:    [[TMP20:%.*]] = shufflevector <2 x double> undef, <2 x double> [[TMP19]], <2 x i32> <i32 2, i32 1>
+; CHECK-NEXT:    [[BLOCK19:%.*]] = shufflevector <2 x double> [[SPLIT]], <2 x double> poison, <1 x i32> <i32 1>
+; CHECK-NEXT:    [[TMP21:%.*]] = extractelement <2 x double> [[SPLIT2]], i64 1
+; CHECK-NEXT:    [[SPLAT_SPLATINSERT20:%.*]] = insertelement <1 x double> poison, double [[TMP21]], i32 0
+; CHECK-NEXT:    [[SPLAT_SPLAT21:%.*]] = shufflevector <1 x double> [[SPLAT_SPLATINSERT20]], <1 x double> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP22:%.*]] = fmul <1 x double> [[BLOCK19]], [[SPLAT_SPLAT21]]
+; CHECK-NEXT:    [[BLOCK22:%.*]] = shufflevector <2 x double> [[SPLIT1]], <2 x double> poison, <1 x i32> <i32 1>
+; CHECK-NEXT:    [[TMP23:%.*]] = extractelement <2 x double> [[SPLIT3]], i64 1
+; CHECK-NEXT:    [[SPLAT_SPLATINSERT23:%.*]] = insertelement <1 x double> poison, double [[TMP23]], i32 0
+; CHECK-NEXT:    [[SPLAT_SPLAT24:%.*]] = shufflevector <1 x double> [[SPLAT_SPLATINSERT23]], <1 x double> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP24:%.*]] = fmul <1 x double> [[BLOCK22]], [[SPLAT_SPLAT24]]
+; CHECK-NEXT:    [[TMP25:%.*]] = fadd <1 x double> [[TMP22]], [[TMP24]]
+; CHECK-NEXT:    [[TMP26:%.*]] = shufflevector <1 x double> [[TMP25]], <1 x double> poison, <2 x i32> <i32 0, i32 undef>
+; CHECK-NEXT:    [[TMP27:%.*]] = shufflevector <2 x double> [[TMP20]], <2 x double> [[TMP26]], <2 x i32> <i32 0, i32 2>
+; CHECK-NEXT:    [[TMP28:%.*]] = shufflevector <2 x double> [[TMP13]], <2 x double> [[TMP27]], <4 x i32> <i32 0, i32 1, i32 2, i32 3>
+; CHECK-NEXT:    ret <4 x double> [[TMP28]]
+;
+entry:
+  %bt = call <4 x double> @llvm.matrix.transpose.v4f64(<4 x double> %b, i32 2, i32 2)
+  %c = call <4 x double> @llvm.matrix.multiply.v4f64.v4f64.v4f64(<4 x double> %a, <4 x double> %bt, i32 2, i32 2, i32 2)
+  ret <4 x double> %c
+}
+
+declare <4 x double> @llvm.matrix.multiply.v4f64.v4f64.v4f64(<4 x double>, <4 x double>, i32, i32, i32)
+declare <4 x double> @llvm.matrix.transpose.v4f64(<4 x double>, i32, i32)