[llvm] 0468124 - [Matrix] Limit dot lowering to column major matrixes.

Florian Hahn via llvm-commits llvm-commits at lists.llvm.org
Wed Apr 5 07:49:24 PDT 2023


Author: Florian Hahn
Date: 2023-04-05T15:49:06+01:00
New Revision: 04681243b417d78bdefecf14158ffced5a6d2273

URL: https://github.com/llvm/llvm-project/commit/04681243b417d78bdefecf14158ffced5a6d2273
DIFF: https://github.com/llvm/llvm-project/commit/04681243b417d78bdefecf14158ffced5a6d2273.diff

LOG: [Matrix] Limit dot lowering to column major matrixes.

Limit to dot product lowering to column major matrixes for now. This
simplifies the code and reasoning for upcoming planned improvements.
Support for row-major matrixes can be added later as extension.

Added: 
    

Modified: 
    llvm/lib/Transforms/Scalar/LowerMatrixIntrinsics.cpp
    llvm/test/Transforms/LowerMatrixIntrinsics/dot-product-int-row-major.ll
    llvm/test/Transforms/LowerMatrixIntrinsics/dot-product-int.ll

Removed: 
    


################################################################################
diff  --git a/llvm/lib/Transforms/Scalar/LowerMatrixIntrinsics.cpp b/llvm/lib/Transforms/Scalar/LowerMatrixIntrinsics.cpp
index 1ef64eb427702..1147f018d5655 100644
--- a/llvm/lib/Transforms/Scalar/LowerMatrixIntrinsics.cpp
+++ b/llvm/lib/Transforms/Scalar/LowerMatrixIntrinsics.cpp
@@ -1324,6 +1324,8 @@ class LowerMatrixIntrinsics {
   void lowerDotProduct(CallInst *MatMul,
                        SmallPtrSet<Instruction *, 16> &FusedInsts,
                        FastMathFlags FMF) {
+    if (MatrixLayout != MatrixLayoutTy::ColumnMajor)
+      return;
     ShapeInfo LShape(MatMul->getArgOperand(2), MatMul->getArgOperand(3));
     ShapeInfo RShape(MatMul->getArgOperand(3), MatMul->getArgOperand(4));
 
@@ -1348,8 +1350,7 @@ class LowerMatrixIntrinsics {
                            m_Intrinsic<Intrinsic::matrix_column_major_load>(
                                m_Value(), m_SpecificInt(N)))));
     };
-    if (!IsSupportedArg(RHS, RShape.NumColumns) ||
-        !IsSupportedArg(LHS, LShape.NumColumns))
+    if (!IsSupportedArg(RHS, RShape.NumColumns))
       return;
 
     // We compare the costs of a vector.reduce.add to sequential add.
@@ -1389,7 +1390,6 @@ class LowerMatrixIntrinsics {
       return Op;
     };
     LHS = FlattenArg(LHS);
-    RHS = FlattenArg(RHS);
 
     // Insert mul/fmul and llvm.vector.reduce.fadd
     Value *Mul =

diff  --git a/llvm/test/Transforms/LowerMatrixIntrinsics/dot-product-int-row-major.ll b/llvm/test/Transforms/LowerMatrixIntrinsics/dot-product-int-row-major.ll
index 94144c908df8f..43ebd4aef64b1 100644
--- a/llvm/test/Transforms/LowerMatrixIntrinsics/dot-product-int-row-major.ll
+++ b/llvm/test/Transforms/LowerMatrixIntrinsics/dot-product-int-row-major.ll
@@ -5,10 +5,65 @@
 define <1 x i32> @dotproduct_i32_v8(<8 x i32> %a, <8 x i32> %b) {
 ; CHECK-LABEL: @dotproduct_i32_v8(
 ; CHECK-NEXT:  entry:
-; CHECK-NEXT:    [[TMP0:%.*]] = mul <8 x i32> [[A:%.*]], [[B:%.*]]
-; CHECK-NEXT:    [[TMP1:%.*]] = call i32 @llvm.vector.reduce.add.v8i32(<8 x i32> [[TMP0]])
-; CHECK-NEXT:    [[TMP2:%.*]] = insertelement <1 x i32> poison, i32 [[TMP1]], i64 0
-; CHECK-NEXT:    ret <1 x i32> [[TMP2]]
+; CHECK-NEXT:    [[SPLIT:%.*]] = shufflevector <8 x i32> [[A:%.*]], <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:    [[SPLIT1:%.*]] = shufflevector <8 x i32> [[B:%.*]], <8 x i32> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[SPLIT2:%.*]] = shufflevector <8 x i32> [[B]], <8 x i32> poison, <1 x i32> <i32 1>
+; CHECK-NEXT:    [[SPLIT3:%.*]] = shufflevector <8 x i32> [[B]], <8 x i32> poison, <1 x i32> <i32 2>
+; CHECK-NEXT:    [[SPLIT4:%.*]] = shufflevector <8 x i32> [[B]], <8 x i32> poison, <1 x i32> <i32 3>
+; CHECK-NEXT:    [[SPLIT5:%.*]] = shufflevector <8 x i32> [[B]], <8 x i32> poison, <1 x i32> <i32 4>
+; CHECK-NEXT:    [[SPLIT6:%.*]] = shufflevector <8 x i32> [[B]], <8 x i32> poison, <1 x i32> <i32 5>
+; CHECK-NEXT:    [[SPLIT7:%.*]] = shufflevector <8 x i32> [[B]], <8 x i32> poison, <1 x i32> <i32 6>
+; CHECK-NEXT:    [[SPLIT8:%.*]] = shufflevector <8 x i32> [[B]], <8 x i32> poison, <1 x i32> <i32 7>
+; CHECK-NEXT:    [[BLOCK:%.*]] = shufflevector <1 x i32> [[SPLIT1]], <1 x i32> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP0:%.*]] = extractelement <8 x i32> [[SPLIT]], 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> [[SPLAT_SPLAT]], [[BLOCK]]
+; CHECK-NEXT:    [[BLOCK9:%.*]] = shufflevector <1 x i32> [[SPLIT2]], <1 x i32> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP2:%.*]] = extractelement <8 x i32> [[SPLIT]], i64 1
+; CHECK-NEXT:    [[SPLAT_SPLATINSERT10:%.*]] = insertelement <1 x i32> poison, i32 [[TMP2]], i64 0
+; CHECK-NEXT:    [[SPLAT_SPLAT11:%.*]] = shufflevector <1 x i32> [[SPLAT_SPLATINSERT10]], <1 x i32> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP3:%.*]] = mul <1 x i32> [[SPLAT_SPLAT11]], [[BLOCK9]]
+; CHECK-NEXT:    [[TMP4:%.*]] = add <1 x i32> [[TMP1]], [[TMP3]]
+; CHECK-NEXT:    [[BLOCK12:%.*]] = shufflevector <1 x i32> [[SPLIT3]], <1 x i32> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP5:%.*]] = extractelement <8 x i32> [[SPLIT]], i64 2
+; CHECK-NEXT:    [[SPLAT_SPLATINSERT13:%.*]] = insertelement <1 x i32> poison, i32 [[TMP5]], i64 0
+; CHECK-NEXT:    [[SPLAT_SPLAT14:%.*]] = shufflevector <1 x i32> [[SPLAT_SPLATINSERT13]], <1 x i32> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP6:%.*]] = mul <1 x i32> [[SPLAT_SPLAT14]], [[BLOCK12]]
+; CHECK-NEXT:    [[TMP7:%.*]] = add <1 x i32> [[TMP4]], [[TMP6]]
+; CHECK-NEXT:    [[BLOCK15:%.*]] = shufflevector <1 x i32> [[SPLIT4]], <1 x i32> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP8:%.*]] = extractelement <8 x i32> [[SPLIT]], i64 3
+; CHECK-NEXT:    [[SPLAT_SPLATINSERT16:%.*]] = insertelement <1 x i32> poison, i32 [[TMP8]], i64 0
+; CHECK-NEXT:    [[SPLAT_SPLAT17:%.*]] = shufflevector <1 x i32> [[SPLAT_SPLATINSERT16]], <1 x i32> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP9:%.*]] = mul <1 x i32> [[SPLAT_SPLAT17]], [[BLOCK15]]
+; CHECK-NEXT:    [[TMP10:%.*]] = add <1 x i32> [[TMP7]], [[TMP9]]
+; CHECK-NEXT:    [[BLOCK18:%.*]] = shufflevector <1 x i32> [[SPLIT5]], <1 x i32> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP11:%.*]] = extractelement <8 x i32> [[SPLIT]], i64 4
+; CHECK-NEXT:    [[SPLAT_SPLATINSERT19:%.*]] = insertelement <1 x i32> poison, i32 [[TMP11]], i64 0
+; CHECK-NEXT:    [[SPLAT_SPLAT20:%.*]] = shufflevector <1 x i32> [[SPLAT_SPLATINSERT19]], <1 x i32> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP12:%.*]] = mul <1 x i32> [[SPLAT_SPLAT20]], [[BLOCK18]]
+; CHECK-NEXT:    [[TMP13:%.*]] = add <1 x i32> [[TMP10]], [[TMP12]]
+; CHECK-NEXT:    [[BLOCK21:%.*]] = shufflevector <1 x i32> [[SPLIT6]], <1 x i32> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP14:%.*]] = extractelement <8 x i32> [[SPLIT]], i64 5
+; CHECK-NEXT:    [[SPLAT_SPLATINSERT22:%.*]] = insertelement <1 x i32> poison, i32 [[TMP14]], i64 0
+; CHECK-NEXT:    [[SPLAT_SPLAT23:%.*]] = shufflevector <1 x i32> [[SPLAT_SPLATINSERT22]], <1 x i32> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP15:%.*]] = mul <1 x i32> [[SPLAT_SPLAT23]], [[BLOCK21]]
+; CHECK-NEXT:    [[TMP16:%.*]] = add <1 x i32> [[TMP13]], [[TMP15]]
+; CHECK-NEXT:    [[BLOCK24:%.*]] = shufflevector <1 x i32> [[SPLIT7]], <1 x i32> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP17:%.*]] = extractelement <8 x i32> [[SPLIT]], i64 6
+; CHECK-NEXT:    [[SPLAT_SPLATINSERT25:%.*]] = insertelement <1 x i32> poison, i32 [[TMP17]], i64 0
+; CHECK-NEXT:    [[SPLAT_SPLAT26:%.*]] = shufflevector <1 x i32> [[SPLAT_SPLATINSERT25]], <1 x i32> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP18:%.*]] = mul <1 x i32> [[SPLAT_SPLAT26]], [[BLOCK24]]
+; CHECK-NEXT:    [[TMP19:%.*]] = add <1 x i32> [[TMP16]], [[TMP18]]
+; CHECK-NEXT:    [[BLOCK27:%.*]] = shufflevector <1 x i32> [[SPLIT8]], <1 x i32> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP20:%.*]] = extractelement <8 x i32> [[SPLIT]], i64 7
+; CHECK-NEXT:    [[SPLAT_SPLATINSERT28:%.*]] = insertelement <1 x i32> poison, i32 [[TMP20]], i64 0
+; CHECK-NEXT:    [[SPLAT_SPLAT29:%.*]] = shufflevector <1 x i32> [[SPLAT_SPLATINSERT28]], <1 x i32> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP21:%.*]] = mul <1 x i32> [[SPLAT_SPLAT29]], [[BLOCK27]]
+; CHECK-NEXT:    [[TMP22:%.*]] = add <1 x i32> [[TMP19]], [[TMP21]]
+; CHECK-NEXT:    [[TMP23:%.*]] = shufflevector <1 x i32> [[TMP22]], <1 x i32> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP24:%.*]] = shufflevector <1 x i32> undef, <1 x i32> [[TMP23]], <1 x i32> <i32 1>
+; CHECK-NEXT:    ret <1 x i32> [[TMP24]]
 ;
 entry:
   %c = tail call <1 x i32> @llvm.matrix.multiply.v1i32.v8i32.v8i32(<8 x i32> %a, <8 x i32> %b, i32 1, i32 8, i32 1)
@@ -22,11 +77,66 @@ declare void @use(<1 x i32>)
 define <1 x i32> @dotproduct_i32_v8_result_used_by_inst(<8 x i32> %a, <8 x i32> %b) {
 ; CHECK-LABEL: @dotproduct_i32_v8_result_used_by_inst(
 ; CHECK-NEXT:  entry:
-; CHECK-NEXT:    [[TMP0:%.*]] = mul <8 x i32> [[A:%.*]], [[B:%.*]]
-; CHECK-NEXT:    [[TMP1:%.*]] = call i32 @llvm.vector.reduce.add.v8i32(<8 x i32> [[TMP0]])
-; CHECK-NEXT:    [[TMP2:%.*]] = insertelement <1 x i32> poison, i32 [[TMP1]], i64 0
-; CHECK-NEXT:    call void @use(<1 x i32> [[TMP2]])
-; CHECK-NEXT:    ret <1 x i32> [[TMP2]]
+; CHECK-NEXT:    [[SPLIT:%.*]] = shufflevector <8 x i32> [[A:%.*]], <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:    [[SPLIT1:%.*]] = shufflevector <8 x i32> [[B:%.*]], <8 x i32> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[SPLIT2:%.*]] = shufflevector <8 x i32> [[B]], <8 x i32> poison, <1 x i32> <i32 1>
+; CHECK-NEXT:    [[SPLIT3:%.*]] = shufflevector <8 x i32> [[B]], <8 x i32> poison, <1 x i32> <i32 2>
+; CHECK-NEXT:    [[SPLIT4:%.*]] = shufflevector <8 x i32> [[B]], <8 x i32> poison, <1 x i32> <i32 3>
+; CHECK-NEXT:    [[SPLIT5:%.*]] = shufflevector <8 x i32> [[B]], <8 x i32> poison, <1 x i32> <i32 4>
+; CHECK-NEXT:    [[SPLIT6:%.*]] = shufflevector <8 x i32> [[B]], <8 x i32> poison, <1 x i32> <i32 5>
+; CHECK-NEXT:    [[SPLIT7:%.*]] = shufflevector <8 x i32> [[B]], <8 x i32> poison, <1 x i32> <i32 6>
+; CHECK-NEXT:    [[SPLIT8:%.*]] = shufflevector <8 x i32> [[B]], <8 x i32> poison, <1 x i32> <i32 7>
+; CHECK-NEXT:    [[BLOCK:%.*]] = shufflevector <1 x i32> [[SPLIT1]], <1 x i32> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP0:%.*]] = extractelement <8 x i32> [[SPLIT]], 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> [[SPLAT_SPLAT]], [[BLOCK]]
+; CHECK-NEXT:    [[BLOCK9:%.*]] = shufflevector <1 x i32> [[SPLIT2]], <1 x i32> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP2:%.*]] = extractelement <8 x i32> [[SPLIT]], i64 1
+; CHECK-NEXT:    [[SPLAT_SPLATINSERT10:%.*]] = insertelement <1 x i32> poison, i32 [[TMP2]], i64 0
+; CHECK-NEXT:    [[SPLAT_SPLAT11:%.*]] = shufflevector <1 x i32> [[SPLAT_SPLATINSERT10]], <1 x i32> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP3:%.*]] = mul <1 x i32> [[SPLAT_SPLAT11]], [[BLOCK9]]
+; CHECK-NEXT:    [[TMP4:%.*]] = add <1 x i32> [[TMP1]], [[TMP3]]
+; CHECK-NEXT:    [[BLOCK12:%.*]] = shufflevector <1 x i32> [[SPLIT3]], <1 x i32> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP5:%.*]] = extractelement <8 x i32> [[SPLIT]], i64 2
+; CHECK-NEXT:    [[SPLAT_SPLATINSERT13:%.*]] = insertelement <1 x i32> poison, i32 [[TMP5]], i64 0
+; CHECK-NEXT:    [[SPLAT_SPLAT14:%.*]] = shufflevector <1 x i32> [[SPLAT_SPLATINSERT13]], <1 x i32> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP6:%.*]] = mul <1 x i32> [[SPLAT_SPLAT14]], [[BLOCK12]]
+; CHECK-NEXT:    [[TMP7:%.*]] = add <1 x i32> [[TMP4]], [[TMP6]]
+; CHECK-NEXT:    [[BLOCK15:%.*]] = shufflevector <1 x i32> [[SPLIT4]], <1 x i32> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP8:%.*]] = extractelement <8 x i32> [[SPLIT]], i64 3
+; CHECK-NEXT:    [[SPLAT_SPLATINSERT16:%.*]] = insertelement <1 x i32> poison, i32 [[TMP8]], i64 0
+; CHECK-NEXT:    [[SPLAT_SPLAT17:%.*]] = shufflevector <1 x i32> [[SPLAT_SPLATINSERT16]], <1 x i32> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP9:%.*]] = mul <1 x i32> [[SPLAT_SPLAT17]], [[BLOCK15]]
+; CHECK-NEXT:    [[TMP10:%.*]] = add <1 x i32> [[TMP7]], [[TMP9]]
+; CHECK-NEXT:    [[BLOCK18:%.*]] = shufflevector <1 x i32> [[SPLIT5]], <1 x i32> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP11:%.*]] = extractelement <8 x i32> [[SPLIT]], i64 4
+; CHECK-NEXT:    [[SPLAT_SPLATINSERT19:%.*]] = insertelement <1 x i32> poison, i32 [[TMP11]], i64 0
+; CHECK-NEXT:    [[SPLAT_SPLAT20:%.*]] = shufflevector <1 x i32> [[SPLAT_SPLATINSERT19]], <1 x i32> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP12:%.*]] = mul <1 x i32> [[SPLAT_SPLAT20]], [[BLOCK18]]
+; CHECK-NEXT:    [[TMP13:%.*]] = add <1 x i32> [[TMP10]], [[TMP12]]
+; CHECK-NEXT:    [[BLOCK21:%.*]] = shufflevector <1 x i32> [[SPLIT6]], <1 x i32> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP14:%.*]] = extractelement <8 x i32> [[SPLIT]], i64 5
+; CHECK-NEXT:    [[SPLAT_SPLATINSERT22:%.*]] = insertelement <1 x i32> poison, i32 [[TMP14]], i64 0
+; CHECK-NEXT:    [[SPLAT_SPLAT23:%.*]] = shufflevector <1 x i32> [[SPLAT_SPLATINSERT22]], <1 x i32> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP15:%.*]] = mul <1 x i32> [[SPLAT_SPLAT23]], [[BLOCK21]]
+; CHECK-NEXT:    [[TMP16:%.*]] = add <1 x i32> [[TMP13]], [[TMP15]]
+; CHECK-NEXT:    [[BLOCK24:%.*]] = shufflevector <1 x i32> [[SPLIT7]], <1 x i32> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP17:%.*]] = extractelement <8 x i32> [[SPLIT]], i64 6
+; CHECK-NEXT:    [[SPLAT_SPLATINSERT25:%.*]] = insertelement <1 x i32> poison, i32 [[TMP17]], i64 0
+; CHECK-NEXT:    [[SPLAT_SPLAT26:%.*]] = shufflevector <1 x i32> [[SPLAT_SPLATINSERT25]], <1 x i32> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP18:%.*]] = mul <1 x i32> [[SPLAT_SPLAT26]], [[BLOCK24]]
+; CHECK-NEXT:    [[TMP19:%.*]] = add <1 x i32> [[TMP16]], [[TMP18]]
+; CHECK-NEXT:    [[BLOCK27:%.*]] = shufflevector <1 x i32> [[SPLIT8]], <1 x i32> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP20:%.*]] = extractelement <8 x i32> [[SPLIT]], i64 7
+; CHECK-NEXT:    [[SPLAT_SPLATINSERT28:%.*]] = insertelement <1 x i32> poison, i32 [[TMP20]], i64 0
+; CHECK-NEXT:    [[SPLAT_SPLAT29:%.*]] = shufflevector <1 x i32> [[SPLAT_SPLATINSERT28]], <1 x i32> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP21:%.*]] = mul <1 x i32> [[SPLAT_SPLAT29]], [[BLOCK27]]
+; CHECK-NEXT:    [[TMP22:%.*]] = add <1 x i32> [[TMP19]], [[TMP21]]
+; CHECK-NEXT:    [[TMP23:%.*]] = shufflevector <1 x i32> [[TMP22]], <1 x i32> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP24:%.*]] = shufflevector <1 x i32> undef, <1 x i32> [[TMP23]], <1 x i32> <i32 1>
+; CHECK-NEXT:    call void @use(<1 x i32> [[TMP24]])
+; CHECK-NEXT:    ret <1 x i32> [[TMP24]]
 ;
 entry:
   %c = tail call <1 x i32> @llvm.matrix.multiply.v1i32.v8i32.v8i32(<8 x i32> %a, <8 x i32> %b, i32 1, i32 8, i32 1)
@@ -38,10 +148,49 @@ entry:
 define <1 x i32> @dotproduct_i32_v8_constvector(<8 x i32> %a) {
 ; CHECK-LABEL: @dotproduct_i32_v8_constvector(
 ; CHECK-NEXT:  entry:
-; CHECK-NEXT:    [[TMP0:%.*]] = mul <8 x i32> [[A:%.*]], <i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7, i32 8>
-; CHECK-NEXT:    [[TMP1:%.*]] = call i32 @llvm.vector.reduce.add.v8i32(<8 x i32> [[TMP0]])
-; CHECK-NEXT:    [[TMP2:%.*]] = insertelement <1 x i32> poison, i32 [[TMP1]], i64 0
-; CHECK-NEXT:    ret <1 x i32> [[TMP2]]
+; CHECK-NEXT:    [[SPLIT:%.*]] = shufflevector <8 x i32> [[A:%.*]], <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:    [[TMP0:%.*]] = extractelement <8 x i32> [[SPLIT]], 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> [[SPLAT_SPLAT]], <i32 1>
+; CHECK-NEXT:    [[TMP2:%.*]] = extractelement <8 x i32> [[SPLIT]], i64 1
+; CHECK-NEXT:    [[SPLAT_SPLATINSERT1:%.*]] = insertelement <1 x i32> poison, i32 [[TMP2]], i64 0
+; CHECK-NEXT:    [[SPLAT_SPLAT2:%.*]] = shufflevector <1 x i32> [[SPLAT_SPLATINSERT1]], <1 x i32> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP3:%.*]] = mul <1 x i32> [[SPLAT_SPLAT2]], <i32 2>
+; CHECK-NEXT:    [[TMP4:%.*]] = add <1 x i32> [[TMP1]], [[TMP3]]
+; CHECK-NEXT:    [[TMP5:%.*]] = extractelement <8 x i32> [[SPLIT]], i64 2
+; CHECK-NEXT:    [[SPLAT_SPLATINSERT3:%.*]] = insertelement <1 x i32> poison, i32 [[TMP5]], i64 0
+; CHECK-NEXT:    [[SPLAT_SPLAT4:%.*]] = shufflevector <1 x i32> [[SPLAT_SPLATINSERT3]], <1 x i32> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP6:%.*]] = mul <1 x i32> [[SPLAT_SPLAT4]], <i32 3>
+; CHECK-NEXT:    [[TMP7:%.*]] = add <1 x i32> [[TMP4]], [[TMP6]]
+; CHECK-NEXT:    [[TMP8:%.*]] = extractelement <8 x i32> [[SPLIT]], i64 3
+; CHECK-NEXT:    [[SPLAT_SPLATINSERT5:%.*]] = insertelement <1 x i32> poison, i32 [[TMP8]], i64 0
+; CHECK-NEXT:    [[SPLAT_SPLAT6:%.*]] = shufflevector <1 x i32> [[SPLAT_SPLATINSERT5]], <1 x i32> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP9:%.*]] = mul <1 x i32> [[SPLAT_SPLAT6]], <i32 4>
+; CHECK-NEXT:    [[TMP10:%.*]] = add <1 x i32> [[TMP7]], [[TMP9]]
+; CHECK-NEXT:    [[TMP11:%.*]] = extractelement <8 x i32> [[SPLIT]], i64 4
+; 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> [[SPLAT_SPLAT8]], <i32 5>
+; CHECK-NEXT:    [[TMP13:%.*]] = add <1 x i32> [[TMP10]], [[TMP12]]
+; CHECK-NEXT:    [[TMP14:%.*]] = extractelement <8 x i32> [[SPLIT]], i64 5
+; CHECK-NEXT:    [[SPLAT_SPLATINSERT9:%.*]] = insertelement <1 x i32> poison, i32 [[TMP14]], i64 0
+; CHECK-NEXT:    [[SPLAT_SPLAT10:%.*]] = shufflevector <1 x i32> [[SPLAT_SPLATINSERT9]], <1 x i32> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP15:%.*]] = mul <1 x i32> [[SPLAT_SPLAT10]], <i32 6>
+; CHECK-NEXT:    [[TMP16:%.*]] = add <1 x i32> [[TMP13]], [[TMP15]]
+; CHECK-NEXT:    [[TMP17:%.*]] = extractelement <8 x i32> [[SPLIT]], i64 6
+; CHECK-NEXT:    [[SPLAT_SPLATINSERT11:%.*]] = insertelement <1 x i32> poison, i32 [[TMP17]], i64 0
+; CHECK-NEXT:    [[SPLAT_SPLAT12:%.*]] = shufflevector <1 x i32> [[SPLAT_SPLATINSERT11]], <1 x i32> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP18:%.*]] = mul <1 x i32> [[SPLAT_SPLAT12]], <i32 7>
+; CHECK-NEXT:    [[TMP19:%.*]] = add <1 x i32> [[TMP16]], [[TMP18]]
+; CHECK-NEXT:    [[TMP20:%.*]] = extractelement <8 x i32> [[SPLIT]], i64 7
+; CHECK-NEXT:    [[SPLAT_SPLATINSERT13:%.*]] = insertelement <1 x i32> poison, i32 [[TMP20]], i64 0
+; CHECK-NEXT:    [[SPLAT_SPLAT14:%.*]] = shufflevector <1 x i32> [[SPLAT_SPLATINSERT13]], <1 x i32> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP21:%.*]] = mul <1 x i32> [[SPLAT_SPLAT14]], <i32 8>
+; CHECK-NEXT:    [[TMP22:%.*]] = add <1 x i32> [[TMP19]], [[TMP21]]
+; CHECK-NEXT:    [[TMP23:%.*]] = shufflevector <1 x i32> [[TMP22]], <1 x i32> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP24:%.*]] = shufflevector <1 x i32> undef, <1 x i32> [[TMP23]], <1 x i32> <i32 1>
+; CHECK-NEXT:    ret <1 x i32> [[TMP24]]
 ;
 entry:
   %c = tail call <1 x i32> @llvm.matrix.multiply.v1i32.v8i32.v8i32(<8 x i32> %a, <8 x i32> <i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7, i32 8>, i32 1, i32 8, i32 1)
@@ -539,10 +688,65 @@ entry:
 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, <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7>
+; CHECK-NEXT:    [[SPLIT1:%.*]] = shufflevector <8 x i64> [[B:%.*]], <8 x i64> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[SPLIT2:%.*]] = shufflevector <8 x i64> [[B]], <8 x i64> poison, <1 x i32> <i32 1>
+; CHECK-NEXT:    [[SPLIT3:%.*]] = shufflevector <8 x i64> [[B]], <8 x i64> poison, <1 x i32> <i32 2>
+; CHECK-NEXT:    [[SPLIT4:%.*]] = shufflevector <8 x i64> [[B]], <8 x i64> poison, <1 x i32> <i32 3>
+; CHECK-NEXT:    [[SPLIT5:%.*]] = shufflevector <8 x i64> [[B]], <8 x i64> poison, <1 x i32> <i32 4>
+; CHECK-NEXT:    [[SPLIT6:%.*]] = shufflevector <8 x i64> [[B]], <8 x i64> poison, <1 x i32> <i32 5>
+; CHECK-NEXT:    [[SPLIT7:%.*]] = shufflevector <8 x i64> [[B]], <8 x i64> poison, <1 x i32> <i32 6>
+; CHECK-NEXT:    [[SPLIT8:%.*]] = shufflevector <8 x i64> [[B]], <8 x i64> poison, <1 x i32> <i32 7>
+; CHECK-NEXT:    [[BLOCK:%.*]] = shufflevector <1 x i64> [[SPLIT1]], <1 x i64> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP0:%.*]] = extractelement <8 x i64> [[SPLIT]], 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> [[SPLAT_SPLAT]], [[BLOCK]]
+; CHECK-NEXT:    [[BLOCK9:%.*]] = shufflevector <1 x i64> [[SPLIT2]], <1 x i64> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP2:%.*]] = extractelement <8 x i64> [[SPLIT]], 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> [[SPLAT_SPLAT11]], [[BLOCK9]]
+; CHECK-NEXT:    [[TMP4:%.*]] = add <1 x i64> [[TMP1]], [[TMP3]]
+; CHECK-NEXT:    [[BLOCK12:%.*]] = shufflevector <1 x i64> [[SPLIT3]], <1 x i64> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP5:%.*]] = extractelement <8 x i64> [[SPLIT]], 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> [[SPLAT_SPLAT14]], [[BLOCK12]]
+; CHECK-NEXT:    [[TMP7:%.*]] = add <1 x i64> [[TMP4]], [[TMP6]]
+; CHECK-NEXT:    [[BLOCK15:%.*]] = shufflevector <1 x i64> [[SPLIT4]], <1 x i64> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP8:%.*]] = extractelement <8 x i64> [[SPLIT]], 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> [[SPLAT_SPLAT17]], [[BLOCK15]]
+; CHECK-NEXT:    [[TMP10:%.*]] = add <1 x i64> [[TMP7]], [[TMP9]]
+; CHECK-NEXT:    [[BLOCK18:%.*]] = shufflevector <1 x i64> [[SPLIT5]], <1 x i64> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP11:%.*]] = extractelement <8 x i64> [[SPLIT]], 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> [[SPLAT_SPLAT20]], [[BLOCK18]]
+; CHECK-NEXT:    [[TMP13:%.*]] = add <1 x i64> [[TMP10]], [[TMP12]]
+; CHECK-NEXT:    [[BLOCK21:%.*]] = shufflevector <1 x i64> [[SPLIT6]], <1 x i64> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP14:%.*]] = extractelement <8 x i64> [[SPLIT]], 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> [[SPLAT_SPLAT23]], [[BLOCK21]]
+; CHECK-NEXT:    [[TMP16:%.*]] = add <1 x i64> [[TMP13]], [[TMP15]]
+; CHECK-NEXT:    [[BLOCK24:%.*]] = shufflevector <1 x i64> [[SPLIT7]], <1 x i64> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP17:%.*]] = extractelement <8 x i64> [[SPLIT]], 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> [[SPLAT_SPLAT26]], [[BLOCK24]]
+; CHECK-NEXT:    [[TMP19:%.*]] = add <1 x i64> [[TMP16]], [[TMP18]]
+; CHECK-NEXT:    [[BLOCK27:%.*]] = shufflevector <1 x i64> [[SPLIT8]], <1 x i64> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP20:%.*]] = extractelement <8 x i64> [[SPLIT]], 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> [[SPLAT_SPLAT29]], [[BLOCK27]]
+; 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)
@@ -554,12 +758,56 @@ declare <1 x i64> @llvm.matrix.multiply.v1i64.v8i64.v8i64(<8 x i64>, <8 x i64>,
 define <1 x i16> @LoadInst_dot_product_i16_v6(ptr %lhs_address, ptr %rhs_address) {
 ; CHECK-LABEL: @LoadInst_dot_product_i16_v6(
 ; CHECK-NEXT:  entry:
-; CHECK-NEXT:    [[LHS:%.*]] = load <6 x i16>, ptr [[LHS_ADDRESS:%.*]], align 16
-; CHECK-NEXT:    [[RHS:%.*]] = load <6 x i16>, ptr [[RHS_ADDRESS:%.*]], align 16
-; CHECK-NEXT:    [[TMP0:%.*]] = mul <6 x i16> [[LHS]], [[RHS]]
-; CHECK-NEXT:    [[TMP1:%.*]] = call i16 @llvm.vector.reduce.add.v6i16(<6 x i16> [[TMP0]])
-; CHECK-NEXT:    [[TMP2:%.*]] = insertelement <1 x i16> poison, i16 [[TMP1]], i64 0
-; CHECK-NEXT:    ret <1 x i16> [[TMP2]]
+; CHECK-NEXT:    [[COL_LOAD:%.*]] = load <6 x i16>, ptr [[LHS_ADDRESS:%.*]], align 16
+; CHECK-NEXT:    [[COL_LOAD1:%.*]] = load <1 x i16>, ptr [[RHS_ADDRESS:%.*]], align 16
+; CHECK-NEXT:    [[VEC_GEP:%.*]] = getelementptr i16, ptr [[RHS_ADDRESS]], i64 1
+; CHECK-NEXT:    [[COL_LOAD2:%.*]] = load <1 x i16>, ptr [[VEC_GEP]], align 2
+; CHECK-NEXT:    [[VEC_GEP3:%.*]] = getelementptr i16, ptr [[RHS_ADDRESS]], i64 2
+; CHECK-NEXT:    [[COL_LOAD4:%.*]] = load <1 x i16>, ptr [[VEC_GEP3]], align 4
+; CHECK-NEXT:    [[VEC_GEP5:%.*]] = getelementptr i16, ptr [[RHS_ADDRESS]], i64 3
+; CHECK-NEXT:    [[COL_LOAD6:%.*]] = load <1 x i16>, ptr [[VEC_GEP5]], align 2
+; CHECK-NEXT:    [[VEC_GEP7:%.*]] = getelementptr i16, ptr [[RHS_ADDRESS]], i64 4
+; CHECK-NEXT:    [[COL_LOAD8:%.*]] = load <1 x i16>, ptr [[VEC_GEP7]], align 8
+; CHECK-NEXT:    [[VEC_GEP9:%.*]] = getelementptr i16, ptr [[RHS_ADDRESS]], i64 5
+; CHECK-NEXT:    [[COL_LOAD10:%.*]] = load <1 x i16>, ptr [[VEC_GEP9]], align 2
+; CHECK-NEXT:    [[BLOCK:%.*]] = shufflevector <1 x i16> [[COL_LOAD1]], <1 x i16> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP0:%.*]] = extractelement <6 x i16> [[COL_LOAD]], i64 0
+; CHECK-NEXT:    [[SPLAT_SPLATINSERT:%.*]] = insertelement <1 x i16> poison, i16 [[TMP0]], i64 0
+; CHECK-NEXT:    [[SPLAT_SPLAT:%.*]] = shufflevector <1 x i16> [[SPLAT_SPLATINSERT]], <1 x i16> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP1:%.*]] = mul <1 x i16> [[SPLAT_SPLAT]], [[BLOCK]]
+; CHECK-NEXT:    [[BLOCK11:%.*]] = shufflevector <1 x i16> [[COL_LOAD2]], <1 x i16> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP2:%.*]] = extractelement <6 x i16> [[COL_LOAD]], i64 1
+; CHECK-NEXT:    [[SPLAT_SPLATINSERT12:%.*]] = insertelement <1 x i16> poison, i16 [[TMP2]], i64 0
+; CHECK-NEXT:    [[SPLAT_SPLAT13:%.*]] = shufflevector <1 x i16> [[SPLAT_SPLATINSERT12]], <1 x i16> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP3:%.*]] = mul <1 x i16> [[SPLAT_SPLAT13]], [[BLOCK11]]
+; CHECK-NEXT:    [[TMP4:%.*]] = add <1 x i16> [[TMP1]], [[TMP3]]
+; CHECK-NEXT:    [[BLOCK14:%.*]] = shufflevector <1 x i16> [[COL_LOAD4]], <1 x i16> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP5:%.*]] = extractelement <6 x i16> [[COL_LOAD]], i64 2
+; CHECK-NEXT:    [[SPLAT_SPLATINSERT15:%.*]] = insertelement <1 x i16> poison, i16 [[TMP5]], i64 0
+; CHECK-NEXT:    [[SPLAT_SPLAT16:%.*]] = shufflevector <1 x i16> [[SPLAT_SPLATINSERT15]], <1 x i16> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP6:%.*]] = mul <1 x i16> [[SPLAT_SPLAT16]], [[BLOCK14]]
+; CHECK-NEXT:    [[TMP7:%.*]] = add <1 x i16> [[TMP4]], [[TMP6]]
+; CHECK-NEXT:    [[BLOCK17:%.*]] = shufflevector <1 x i16> [[COL_LOAD6]], <1 x i16> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP8:%.*]] = extractelement <6 x i16> [[COL_LOAD]], i64 3
+; CHECK-NEXT:    [[SPLAT_SPLATINSERT18:%.*]] = insertelement <1 x i16> poison, i16 [[TMP8]], i64 0
+; CHECK-NEXT:    [[SPLAT_SPLAT19:%.*]] = shufflevector <1 x i16> [[SPLAT_SPLATINSERT18]], <1 x i16> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP9:%.*]] = mul <1 x i16> [[SPLAT_SPLAT19]], [[BLOCK17]]
+; CHECK-NEXT:    [[TMP10:%.*]] = add <1 x i16> [[TMP7]], [[TMP9]]
+; CHECK-NEXT:    [[BLOCK20:%.*]] = shufflevector <1 x i16> [[COL_LOAD8]], <1 x i16> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP11:%.*]] = extractelement <6 x i16> [[COL_LOAD]], i64 4
+; CHECK-NEXT:    [[SPLAT_SPLATINSERT21:%.*]] = insertelement <1 x i16> poison, i16 [[TMP11]], i64 0
+; CHECK-NEXT:    [[SPLAT_SPLAT22:%.*]] = shufflevector <1 x i16> [[SPLAT_SPLATINSERT21]], <1 x i16> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP12:%.*]] = mul <1 x i16> [[SPLAT_SPLAT22]], [[BLOCK20]]
+; CHECK-NEXT:    [[TMP13:%.*]] = add <1 x i16> [[TMP10]], [[TMP12]]
+; CHECK-NEXT:    [[BLOCK23:%.*]] = shufflevector <1 x i16> [[COL_LOAD10]], <1 x i16> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP14:%.*]] = extractelement <6 x i16> [[COL_LOAD]], i64 5
+; CHECK-NEXT:    [[SPLAT_SPLATINSERT24:%.*]] = insertelement <1 x i16> poison, i16 [[TMP14]], i64 0
+; CHECK-NEXT:    [[SPLAT_SPLAT25:%.*]] = shufflevector <1 x i16> [[SPLAT_SPLATINSERT24]], <1 x i16> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP15:%.*]] = mul <1 x i16> [[SPLAT_SPLAT25]], [[BLOCK23]]
+; CHECK-NEXT:    [[TMP16:%.*]] = add <1 x i16> [[TMP13]], [[TMP15]]
+; CHECK-NEXT:    [[TMP17:%.*]] = shufflevector <1 x i16> [[TMP16]], <1 x i16> poison, <1 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP18:%.*]] = shufflevector <1 x i16> undef, <1 x i16> [[TMP17]], <1 x i32> <i32 1>
+; CHECK-NEXT:    ret <1 x i16> [[TMP18]]
 ;
 entry:
   %lhs = load <6 x i16>, ptr %lhs_address

diff  --git a/llvm/test/Transforms/LowerMatrixIntrinsics/dot-product-int.ll b/llvm/test/Transforms/LowerMatrixIntrinsics/dot-product-int.ll
index 9347504279f81..0129dff311216 100644
--- a/llvm/test/Transforms/LowerMatrixIntrinsics/dot-product-int.ll
+++ b/llvm/test/Transforms/LowerMatrixIntrinsics/dot-product-int.ll
@@ -75,57 +75,17 @@ define <1 x i32> @add_feeding_dotproduct_i32_v8_1(<8 x i32> %a, <8 x i32> %b, <8
 ; 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
@@ -231,57 +191,17 @@ define <1 x i32> @sub_feeding_dotproduct_i32_v8_1(<8 x i32> %a, <8 x i32> %b, <8
 ; 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
@@ -403,57 +323,17 @@ define <1 x i32> @add_chain_feeding_dotproduct_i32_v8_1(<8 x i32> %a, <8 x i32>
 ; 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
@@ -730,8 +610,8 @@ define <1 x i16> @LoadInst_dot_product_i16_v6(ptr %lhs_address, ptr %rhs_address
 ; CHECK-LABEL: @LoadInst_dot_product_i16_v6(
 ; CHECK-NEXT:  entry:
 ; CHECK-NEXT:    [[LHS:%.*]] = load <6 x i16>, ptr [[LHS_ADDRESS:%.*]], align 16
-; CHECK-NEXT:    [[RHS:%.*]] = load <6 x i16>, ptr [[RHS_ADDRESS:%.*]], align 16
-; CHECK-NEXT:    [[TMP0:%.*]] = mul <6 x i16> [[LHS]], [[RHS]]
+; CHECK-NEXT:    [[COL_LOAD:%.*]] = load <6 x i16>, ptr [[RHS_ADDRESS:%.*]], align 16
+; CHECK-NEXT:    [[TMP0:%.*]] = mul <6 x i16> [[LHS]], [[COL_LOAD]]
 ; CHECK-NEXT:    [[TMP1:%.*]] = call i16 @llvm.vector.reduce.add.v6i16(<6 x i16> [[TMP0]])
 ; CHECK-NEXT:    [[TMP2:%.*]] = insertelement <1 x i16> poison, i16 [[TMP1]], i64 0
 ; CHECK-NEXT:    ret <1 x i16> [[TMP2]]


        


More information about the llvm-commits mailing list