diff --git a/include/dxc/DXIL/DxilInstructions.h b/include/dxc/DXIL/DxilInstructions.h
index 29cab28ce3..ffd4db167d 100644
--- a/include/dxc/DXIL/DxilInstructions.h
+++ b/include/dxc/DXIL/DxilInstructions.h
@@ -10867,7 +10867,7 @@ struct DxilInst_LinAlgMatrixAccumulateToMemory {
   // Validation support
   bool isAllowed() const { return true; }
   bool isArgumentListValid() const {
-    if (6 != llvm::dyn_cast<llvm::CallInst>(Instr)->getNumArgOperands())
+    if (7 != llvm::dyn_cast<llvm::CallInst>(Instr)->getNumArgOperands())
       return false;
     return true;
   }
@@ -10877,21 +10877,24 @@ struct DxilInst_LinAlgMatrixAccumulateToMemory {
   enum OperandIdx {
     arg_matrix = 1,
     arg_memory = 2,
-    arg_offset = 3,
-    arg_stride = 4,
-    arg_layout = 5,
+    arg_targetType = 3,
+    arg_offset = 4,
+    arg_stride = 5,
+    arg_layout = 6,
   };
   // Accessors
   llvm::Value *get_matrix() const { return Instr->getOperand(1); }
   void set_matrix(llvm::Value *val) { Instr->setOperand(1, val); }
   llvm::Value *get_memory() const { return Instr->getOperand(2); }
   void set_memory(llvm::Value *val) { Instr->setOperand(2, val); }
-  llvm::Value *get_offset() const { return Instr->getOperand(3); }
-  void set_offset(llvm::Value *val) { Instr->setOperand(3, val); }
-  llvm::Value *get_stride() const { return Instr->getOperand(4); }
-  void set_stride(llvm::Value *val) { Instr->setOperand(4, val); }
-  llvm::Value *get_layout() const { return Instr->getOperand(5); }
-  void set_layout(llvm::Value *val) { Instr->setOperand(5, val); }
+  llvm::Value *get_targetType() const { return Instr->getOperand(3); }
+  void set_targetType(llvm::Value *val) { Instr->setOperand(3, val); }
+  llvm::Value *get_offset() const { return Instr->getOperand(4); }
+  void set_offset(llvm::Value *val) { Instr->setOperand(4, val); }
+  llvm::Value *get_stride() const { return Instr->getOperand(5); }
+  void set_stride(llvm::Value *val) { Instr->setOperand(5, val); }
+  llvm::Value *get_layout() const { return Instr->getOperand(6); }
+  void set_layout(llvm::Value *val) { Instr->setOperand(6, val); }
 };
 
 /// This instruction Outer products an M sized vector and a N sized vector
diff --git a/lib/DXIL/DxilOperations.cpp b/lib/DXIL/DxilOperations.cpp
index 74a4959a86..21266be6f4 100644
--- a/lib/DXIL/DxilOperations.cpp
+++ b/lib/DXIL/DxilOperations.cpp
@@ -6669,6 +6669,7 @@ Function *OP::GetOpFunc(OpCode opCode, Type *pOverloadType) {
     A(pI32);
     A(pI32);
     A(pI32);
+    A(pI32);
     break;
   case OpCode::LinAlgMatrixOuterProduct:
     A(EXT(0));
diff --git a/lib/HLSL/HLOperationLower.cpp b/lib/HLSL/HLOperationLower.cpp
index 258ab5a152..780a51bdde 100644
--- a/lib/HLSL/HLOperationLower.cpp
+++ b/lib/HLSL/HLOperationLower.cpp
@@ -7115,7 +7115,7 @@ Value *TranslateLinAlgMatrixLoadFromMemory(
   return nullptr;
 }
 
-Value *TranslateLinAlgMatrixAccumStoreToMemory(
+Value *TranslateLinAlgMatrixStoreToMemory(
     CallInst *CI, IntrinsicOp IOP, OP::OpCode OpCode,
     HLOperationLowerHelper &Helper, HLObjectOperationLowerHelper *ObjHelper,
     bool &Translated) {
@@ -7139,6 +7139,31 @@ Value *TranslateLinAlgMatrixAccumStoreToMemory(
                             {OpArg, Matrix, ArrPtr, Offset, Stride, Layout});
 }
 
+Value *TranslateLinAlgMatrixAccumToMemory(
+    CallInst *CI, IntrinsicOp IOP, OP::OpCode OpCode,
+    HLOperationLowerHelper &Helper, HLObjectOperationLowerHelper *ObjHelper,
+    bool &Translated) {
+  hlsl::OP *HlslOp = &Helper.hlslOP;
+  IRBuilder<> Builder(CI);
+
+  Value *Matrix = CI->getArgOperand(1);
+  Value *Arr = CI->getArgOperand(2);
+  Value *TargetType = CI->getArgOperand(3);
+  Value *Offset = CI->getArgOperand(4);
+  Value *Stride = CI->getArgOperand(5);
+  Value *Layout = CI->getArgOperand(6);
+
+  Value *Zero = Builder.getInt32(0);
+  Value *ArrPtr = Builder.CreateGEP(Arr, {Zero, Zero});
+  Type *ArrEltTy = ArrPtr->getType()->getPointerElementType();
+
+  Constant *OpArg = HlslOp->GetU32Const((unsigned)OpCode);
+  Function *DxilFunc = HlslOp->GetOpFunc(OpCode, {Matrix->getType(), ArrEltTy});
+
+  return Builder.CreateCall(
+      DxilFunc, {OpArg, Matrix, ArrPtr, TargetType, Offset, Stride, Layout});
+}
+
 Value *TranslateLinAlgConvert(CallInst *CI, IntrinsicOp IOP, OP::OpCode OpCode,
                               HLOperationLowerHelper &Helper,
                               HLObjectOperationLowerHelper *ObjHelper,
@@ -7948,7 +7973,7 @@ constexpr IntrinsicLower gLowerTable[] = {
      TranslateLinAlgMatrixAccumStoreToDescriptor,
      DXIL::OpCode::LinAlgMatrixStoreToDescriptor},
     {IntrinsicOp::IOP___builtin_LinAlg_MatrixStoreToMemory,
-     TranslateLinAlgMatrixAccumStoreToMemory,
+     TranslateLinAlgMatrixStoreToMemory,
      DXIL::OpCode::LinAlgMatrixStoreToMemory},
     {IntrinsicOp::IOP___builtin_LinAlg_MatrixAccumulate,
      TranslateLinAlgMatrixAccumulate, DXIL::OpCode::LinAlgMatrixAccumulate},
@@ -7963,7 +7988,7 @@ constexpr IntrinsicLower gLowerTable[] = {
      TranslateLinAlgMatrixAccumStoreToDescriptor,
      DXIL::OpCode::LinAlgMatrixAccumulateToDescriptor},
     {IntrinsicOp::IOP___builtin_LinAlg_MatrixAccumulateToMemory,
-     TranslateLinAlgMatrixAccumStoreToMemory,
+     TranslateLinAlgMatrixAccumToMemory,
      DXIL::OpCode::LinAlgMatrixAccumulateToMemory},
     {IntrinsicOp::IOP___builtin_LinAlg_MatrixOuterProduct,
      TranslateLinAlgMatrixOuterProduct, DXIL::OpCode::LinAlgMatrixOuterProduct},
diff --git a/tools/clang/lib/Headers/hlsl/dx/linalg.h b/tools/clang/lib/Headers/hlsl/dx/linalg.h
index 0c9824b7c0..26df0571d4 100644
--- a/tools/clang/lib/Headers/hlsl/dx/linalg.h
+++ b/tools/clang/lib/Headers/hlsl/dx/linalg.h
@@ -354,8 +354,8 @@ class Matrix {
       void>::type
   InterlockedAccumulate(groupshared T Arr[Size], uint StartIdx, uint Stride,
                         MatrixLayoutEnum Layout) {
-    __builtin_LinAlg_MatrixAccumulateToMemory(__handle, Arr, StartIdx, Stride,
-                                              Layout);
+    __builtin_LinAlg_MatrixAccumulateToMemory(__handle, Arr, 0, StartIdx,
+                                              Stride, Layout);
   }
 
   template <ComponentEnum CompTy, MatrixUseEnum UseLocal = Use>
diff --git a/tools/clang/test/CodeGenDXIL/hlsl/linalg/api/matrix-class.hlsl b/tools/clang/test/CodeGenDXIL/hlsl/linalg/api/matrix-class.hlsl
index 38d0c43cbf..840a5ef8a4 100644
--- a/tools/clang/test/CodeGenDXIL/hlsl/linalg/api/matrix-class.hlsl
+++ b/tools/clang/test/CodeGenDXIL/hlsl/linalg/api/matrix-class.hlsl
@@ -138,8 +138,8 @@ void main(uint ID : SV_GroupID)
 // CHECK: call void @dx.op.linAlgMatrixAccumulateToMemory.mC9M4N4U2S1.f32(i32 -2147483620,
 // CHECK-SAME: %dx.types.LinAlgMatrixC9M4N4U2S1 %[[ACCUM0]],
 // CHECK-SAME: float addrspace(3)* getelementptr inbounds ([256 x float],
-// CHECK-SAME: [256 x float] addrspace(3)* @"\01?SharedArr@@3PAMA", i32 0, i32 0), i32 0, i32 16, i32 1)
-// CHECK-SAME: ; LinAlgMatrixAccumulateToMemory(matrix,memory,offset,stride,layout)
+// CHECK-SAME: [256 x float] addrspace(3)* @"\01?SharedArr@@3PAMA", i32 0, i32 0), i32 0, i32 0, i32 16, i32 1)
+// CHECK-SAME: ; LinAlgMatrixAccumulateToMemory(matrix,memory,targetType,offset,stride,layout)
   AccMat1.InterlockedAccumulate(SharedArr, 0, 16, MatrixLayoutEnum::ColMajor);
 
 // Matrix::Accumulate
diff --git a/tools/clang/test/CodeGenDXIL/hlsl/linalg/builtins/matrixaccumulatetomemory/nominal.hlsl b/tools/clang/test/CodeGenDXIL/hlsl/linalg/builtins/matrixaccumulatetomemory/nominal.hlsl
index 3adfac877c..824503cd9f 100644
--- a/tools/clang/test/CodeGenDXIL/hlsl/linalg/builtins/matrixaccumulatetomemory/nominal.hlsl
+++ b/tools/clang/test/CodeGenDXIL/hlsl/linalg/builtins/matrixaccumulatetomemory/nominal.hlsl
@@ -11,14 +11,14 @@ void main() {
 
   // CHECK: call void @dx.op.linAlgMatrixAccumulateToMemory.mC4M5N4U1S2.f32(i32 -2147483620,
   // CHECK-SAME: %dx.types.LinAlgMatrixC4M5N4U1S2 %{{.*}}, float addrspace(3)* getelementptr inbounds ([64 x float],
-  // CHECK-SAME: [64 x float] addrspace(3)* @"\01?SharedArr@@3PAMA", i32 0, i32 0), i32 1, i32 2, i32 3)  
-  // CHECK-SAME: ; LinAlgMatrixAccumulateToMemory(matrix,memory,offset,stride,layout)
+  // CHECK-SAME: [64 x float] addrspace(3)* @"\01?SharedArr@@3PAMA", i32 0, i32 0), i32 9, i32 1, i32 2, i32 3)  
+  // CHECK-SAME: ; LinAlgMatrixAccumulateToMemory(matrix,memory,targetType,offset,stride,layout)
 
   // CHECK2: call void @"dx.hl.op..void (i32, %dx.types.LinAlgMatrixC4M5N4U1S2,
-  // CHECK2-SAME: [64 x float] addrspace(3)*, i32, i32, i32)"(i32 416, 
+  // CHECK2-SAME: [64 x float] addrspace(3)*, i32, i32, i32, i32)"(i32 416, 
   // CHECK2-SAME: %dx.types.LinAlgMatrixC4M5N4U1S2 %{{.*}}, [64 x float] addrspace(3)* @"\01?SharedArr@@3PAMA",
-  // CHECK2-SAME: i32 1, i32 2, i32 3)
+  // CHECK2-SAME: i32 9, i32 1, i32 2, i32 3)
   __builtin_LinAlgMatrix [[__LinAlgMatrix_Attributes(4, 5, 4, 1, 2)]] mat;
   __builtin_LinAlg_FillMatrix(mat, 1);
-  __builtin_LinAlg_MatrixAccumulateToMemory(mat, SharedArr, 1, 2, 3);
+  __builtin_LinAlg_MatrixAccumulateToMemory(mat, SharedArr, 9, 1, 2, 3);
 }
diff --git a/tools/clang/test/LitDXILValidation/LinAlgMatrix/linalgmatrix-as.ll b/tools/clang/test/LitDXILValidation/LinAlgMatrix/linalgmatrix-as.ll
index 0b0e008041..5a081b5c14 100644
--- a/tools/clang/test/LitDXILValidation/LinAlgMatrix/linalgmatrix-as.ll
+++ b/tools/clang/test/LitDXILValidation/LinAlgMatrix/linalgmatrix-as.ll
@@ -116,7 +116,7 @@ define void @mainAS() {
   call void @dx.op.linAlgMatrixStoreToDescriptor.mC4M5N4U0S2(i32 -2147483628, %dx.types.LinAlgMatrixC4M5N4U0S2 %v14, %dx.types.Handle %handle, i32 1, i32 2, i32 3, i32 4)  ; LinAlgMatrixStoreToDescriptor(matrix,handle,offset,stride,layout,align)
 
   ; dx.op.linAlgMatrixAccumulateToMemory
-  call void @dx.op.linAlgMatrixAccumulateToMemory.mC4M5N4U0S2.f32(i32 -2147483620, %dx.types.LinAlgMatrixC4M5N4U0S2 %v14, float addrspace(3)* getelementptr inbounds ([64 x float], [64 x float] addrspace(3)* @"\01?SharedArr@@3PAMA", i32 0, i32 0), i32 0, i32 0, i32 0)  ; LinAlgMatrixAccumulateToMemory(matrix,memory,offset,stride,layout)
+  call void @dx.op.linAlgMatrixAccumulateToMemory.mC4M5N4U0S2.f32(i32 -2147483620, %dx.types.LinAlgMatrixC4M5N4U0S2 %v14, float addrspace(3)* getelementptr inbounds ([64 x float], [64 x float] addrspace(3)* @"\01?SharedArr@@3PAMA", i32 0, i32 0), i32 0, i32 0, i32 0, i32 0)  ; LinAlgMatrixAccumulateToMemory(matrix,memory,targetType,offset,stride,layout)
 
   ; dx.op.linAlgMatrixLoadFromMemory
   %v15 = call %dx.types.LinAlgMatrixC4M5N4U0S2 @dx.op.linAlgMatrixLoadFromMemory.mC4M5N4U0S2.f32(i32 -2147483633, float addrspace(3)* getelementptr inbounds ([64 x float], [64 x float] addrspace(3)* @"\01?SharedArr@@3PAMA", i32 0, i32 0), i32 0, i32 0, i32 0)  ; LinAlgMatrixLoadFromMemory(memory,offset,stride,layout)
@@ -191,7 +191,7 @@ declare %dx.types.LinAlgMatrixC4M5N4U0S2 @dx.op.linAlgMatrixSetElement.mC4M5N4U0
 declare void @dx.op.linAlgMatrixStoreToMemory.mC4M5N4U0S2.f32(i32, %dx.types.LinAlgMatrixC4M5N4U0S2, float addrspace(3)*, i32, i32, i32) #0
 
 ; Function Attrs: nounwind
-declare void @dx.op.linAlgMatrixAccumulateToMemory.mC4M5N4U0S2.f32(i32, %dx.types.LinAlgMatrixC4M5N4U0S2, float addrspace(3)*, i32, i32, i32) #0
+declare void @dx.op.linAlgMatrixAccumulateToMemory.mC4M5N4U0S2.f32(i32, %dx.types.LinAlgMatrixC4M5N4U0S2, float addrspace(3)*, i32, i32, i32, i32) #0
 
 ; Function Attrs: nounwind
 declare %dx.types.LinAlgMatrixC4M5N4U0S2 @dx.op.linAlgMatrixLoadFromMemory.mC4M5N4U0S2.f32(i32, float addrspace(3)*, i32, i32, i32) #0
diff --git a/tools/clang/test/LitDXILValidation/LinAlgMatrix/linalgmatrix-cs.ll b/tools/clang/test/LitDXILValidation/LinAlgMatrix/linalgmatrix-cs.ll
index 7f93969b80..d06f8cb3e4 100644
--- a/tools/clang/test/LitDXILValidation/LinAlgMatrix/linalgmatrix-cs.ll
+++ b/tools/clang/test/LitDXILValidation/LinAlgMatrix/linalgmatrix-cs.ll
@@ -89,7 +89,7 @@ define void @mainCS() {
   call void @dx.op.linAlgMatrixStoreToDescriptor.mC4M5N4U0S2(i32 -2147483628, %dx.types.LinAlgMatrixC4M5N4U0S2 %v14, %dx.types.Handle %handle, i32 1, i32 2, i32 3, i32 4)  ; LinAlgMatrixStoreToDescriptor(matrix,handle,offset,stride,layout,align)
 
   ; dx.op.linAlgMatrixAccumulateToMemory
-  call void @dx.op.linAlgMatrixAccumulateToMemory.mC4M5N4U0S2.f32(i32 -2147483620, %dx.types.LinAlgMatrixC4M5N4U0S2 %v14, float addrspace(3)* getelementptr inbounds ([64 x float], [64 x float] addrspace(3)* @"\01?SharedArr@@3PAMA", i32 0, i32 0), i32 0, i32 0, i32 0)  ; LinAlgMatrixAccumulateToMemory(matrix,memory,offset,stride,layout)
+  call void @dx.op.linAlgMatrixAccumulateToMemory.mC4M5N4U0S2.f32(i32 -2147483620, %dx.types.LinAlgMatrixC4M5N4U0S2 %v14, float addrspace(3)* getelementptr inbounds ([64 x float], [64 x float] addrspace(3)* @"\01?SharedArr@@3PAMA", i32 0, i32 0), i32 0, i32 0, i32 0, i32 0)  ; LinAlgMatrixAccumulateToMemory(matrix,memory,targetType,offset,stride,layout)
 
   ; dx.op.linAlgMatrixLoadFromMemory
   %v15 = call %dx.types.LinAlgMatrixC4M5N4U0S2 @dx.op.linAlgMatrixLoadFromMemory.mC4M5N4U0S2.f32(i32 -2147483633, float addrspace(3)* getelementptr inbounds ([64 x float], [64 x float] addrspace(3)* @"\01?SharedArr@@3PAMA", i32 0, i32 0), i32 0, i32 0, i32 0)  ; LinAlgMatrixLoadFromMemory(memory,offset,stride,layout)
@@ -164,7 +164,7 @@ declare %dx.types.LinAlgMatrixC4M5N4U0S2 @dx.op.linAlgMatrixSetElement.mC4M5N4U0
 declare void @dx.op.linAlgMatrixStoreToMemory.mC4M5N4U0S2.f32(i32, %dx.types.LinAlgMatrixC4M5N4U0S2, float addrspace(3)*, i32, i32, i32) #0
 
 ; Function Attrs: nounwind
-declare void @dx.op.linAlgMatrixAccumulateToMemory.mC4M5N4U0S2.f32(i32, %dx.types.LinAlgMatrixC4M5N4U0S2, float addrspace(3)*, i32, i32, i32) #0
+declare void @dx.op.linAlgMatrixAccumulateToMemory.mC4M5N4U0S2.f32(i32, %dx.types.LinAlgMatrixC4M5N4U0S2, float addrspace(3)*, i32, i32, i32, i32) #0
 
 ; Function Attrs: nounwind
 declare %dx.types.LinAlgMatrixC4M5N4U0S2 @dx.op.linAlgMatrixLoadFromMemory.mC4M5N4U0S2.f32(i32, float addrspace(3)*, i32, i32, i32) #0
diff --git a/tools/clang/test/LitDXILValidation/LinAlgMatrix/linalgmatrix-ds.ll b/tools/clang/test/LitDXILValidation/LinAlgMatrix/linalgmatrix-ds.ll
index 0f375410b6..d932d6f394 100644
--- a/tools/clang/test/LitDXILValidation/LinAlgMatrix/linalgmatrix-ds.ll
+++ b/tools/clang/test/LitDXILValidation/LinAlgMatrix/linalgmatrix-ds.ll
@@ -117,7 +117,7 @@ define void @MainDS() {
   call void @dx.op.linAlgMatrixStoreToDescriptor.mC4M5N4U0S2(i32 -2147483628, %dx.types.LinAlgMatrixC4M5N4U0S2 %v14, %dx.types.Handle %handle, i32 1, i32 2, i32 3, i32 4)  ; LinAlgMatrixStoreToDescriptor(matrix,handle,offset,stride,layout,align)
 
   ; dx.op.linAlgMatrixAccumulateToMemory
-  call void @dx.op.linAlgMatrixAccumulateToMemory.mC4M5N4U0S2.f32(i32 -2147483620, %dx.types.LinAlgMatrixC4M5N4U0S2 %v14, float addrspace(3)* getelementptr inbounds ([64 x float], [64 x float] addrspace(3)* @"\01?SharedArr@@3PAMA", i32 0, i32 0), i32 0, i32 0, i32 0)  ; LinAlgMatrixAccumulateToMemory(matrix,memory,offset,stride,layout)
+  call void @dx.op.linAlgMatrixAccumulateToMemory.mC4M5N4U0S2.f32(i32 -2147483620, %dx.types.LinAlgMatrixC4M5N4U0S2 %v14, float addrspace(3)* getelementptr inbounds ([64 x float], [64 x float] addrspace(3)* @"\01?SharedArr@@3PAMA", i32 0, i32 0), i32 0, i32 0, i32 0, i32 0)  ; LinAlgMatrixAccumulateToMemory(matrix,memory,targetType,offset,stride,layout)
 
   ; dx.op.linAlgMatrixLoadFromMemory
   %v15 = call %dx.types.LinAlgMatrixC4M5N4U0S2 @dx.op.linAlgMatrixLoadFromMemory.mC4M5N4U0S2.f32(i32 -2147483633, float addrspace(3)* getelementptr inbounds ([64 x float], [64 x float] addrspace(3)* @"\01?SharedArr@@3PAMA", i32 0, i32 0), i32 0, i32 0, i32 0)  ; LinAlgMatrixLoadFromMemory(memory,offset,stride,layout)
@@ -197,7 +197,7 @@ declare %dx.types.LinAlgMatrixC4M5N4U0S2 @dx.op.linAlgMatrixSetElement.mC4M5N4U0
 declare void @dx.op.linAlgMatrixStoreToMemory.mC4M5N4U0S2.f32(i32, %dx.types.LinAlgMatrixC4M5N4U0S2, float addrspace(3)*, i32, i32, i32) #0
 
 ; Function Attrs: nounwind
-declare void @dx.op.linAlgMatrixAccumulateToMemory.mC4M5N4U0S2.f32(i32, %dx.types.LinAlgMatrixC4M5N4U0S2, float addrspace(3)*, i32, i32, i32) #0
+declare void @dx.op.linAlgMatrixAccumulateToMemory.mC4M5N4U0S2.f32(i32, %dx.types.LinAlgMatrixC4M5N4U0S2, float addrspace(3)*, i32, i32, i32, i32) #0
 
 ; Function Attrs: nounwind
 declare %dx.types.LinAlgMatrixC4M5N4U0S2 @dx.op.linAlgMatrixLoadFromMemory.mC4M5N4U0S2.f32(i32, float addrspace(3)*, i32, i32, i32) #0
diff --git a/tools/clang/test/LitDXILValidation/LinAlgMatrix/linalgmatrix-gs.ll b/tools/clang/test/LitDXILValidation/LinAlgMatrix/linalgmatrix-gs.ll
index 9f8b88d6a8..a564d90c45 100644
--- a/tools/clang/test/LitDXILValidation/LinAlgMatrix/linalgmatrix-gs.ll
+++ b/tools/clang/test/LitDXILValidation/LinAlgMatrix/linalgmatrix-gs.ll
@@ -117,7 +117,7 @@ define void @MainGS() {
   call void @dx.op.linAlgMatrixStoreToDescriptor.mC4M5N4U0S2(i32 -2147483628, %dx.types.LinAlgMatrixC4M5N4U0S2 %v14, %dx.types.Handle %handle, i32 1, i32 2, i32 3, i32 4)  ; LinAlgMatrixStoreToDescriptor(matrix,handle,offset,stride,layout,align)
 
     ; dx.op.linAlgMatrixAccumulateToMemory
-  call void @dx.op.linAlgMatrixAccumulateToMemory.mC4M5N4U0S2.f32(i32 -2147483620, %dx.types.LinAlgMatrixC4M5N4U0S2 %v14, float addrspace(3)* getelementptr inbounds ([64 x float], [64 x float] addrspace(3)* @"\01?SharedArr@@3PAMA", i32 0, i32 0), i32 0, i32 0, i32 0)  ; LinAlgMatrixAccumulateToMemory(matrix,memory,offset,stride,layout)
+  call void @dx.op.linAlgMatrixAccumulateToMemory.mC4M5N4U0S2.f32(i32 -2147483620, %dx.types.LinAlgMatrixC4M5N4U0S2 %v14, float addrspace(3)* getelementptr inbounds ([64 x float], [64 x float] addrspace(3)* @"\01?SharedArr@@3PAMA", i32 0, i32 0), i32 0, i32 0, i32 0, i32 0)  ; LinAlgMatrixAccumulateToMemory(matrix,memory,targetType,offset,stride,layout)
 
   ; dx.op.linAlgMatrixLoadFromMemory
   %v15 = call %dx.types.LinAlgMatrixC4M5N4U0S2 @dx.op.linAlgMatrixLoadFromMemory.mC4M5N4U0S2.f32(i32 -2147483633, float addrspace(3)* getelementptr inbounds ([64 x float], [64 x float] addrspace(3)* @"\01?SharedArr@@3PAMA", i32 0, i32 0), i32 0, i32 0, i32 0)  ; LinAlgMatrixLoadFromMemory(memory,offset,stride,layout)
@@ -196,7 +196,7 @@ declare %dx.types.LinAlgMatrixC4M5N4U0S2 @dx.op.linAlgMatrixSetElement.mC4M5N4U0
 declare void @dx.op.linAlgMatrixStoreToMemory.mC4M5N4U0S2.f32(i32, %dx.types.LinAlgMatrixC4M5N4U0S2, float addrspace(3)*, i32, i32, i32) #0
 
 ; Function Attrs: nounwind
-declare void @dx.op.linAlgMatrixAccumulateToMemory.mC4M5N4U0S2.f32(i32, %dx.types.LinAlgMatrixC4M5N4U0S2, float addrspace(3)*, i32, i32, i32) #0
+declare void @dx.op.linAlgMatrixAccumulateToMemory.mC4M5N4U0S2.f32(i32, %dx.types.LinAlgMatrixC4M5N4U0S2, float addrspace(3)*, i32, i32, i32, i32) #0
 
 ; Function Attrs: nounwind
 declare %dx.types.LinAlgMatrixC4M5N4U0S2 @dx.op.linAlgMatrixLoadFromMemory.mC4M5N4U0S2.f32(i32, float addrspace(3)*, i32, i32, i32) #0
diff --git a/tools/clang/test/LitDXILValidation/LinAlgMatrix/linalgmatrix-hs.ll b/tools/clang/test/LitDXILValidation/LinAlgMatrix/linalgmatrix-hs.ll
index 7e2af0057d..05e535f9c8 100644
--- a/tools/clang/test/LitDXILValidation/LinAlgMatrix/linalgmatrix-hs.ll
+++ b/tools/clang/test/LitDXILValidation/LinAlgMatrix/linalgmatrix-hs.ll
@@ -116,7 +116,7 @@ define void @MainHS() {
   call void @dx.op.linAlgMatrixStoreToDescriptor.mC4M5N4U0S2(i32 -2147483628, %dx.types.LinAlgMatrixC4M5N4U0S2 %v14, %dx.types.Handle %handle, i32 1, i32 2, i32 3, i32 4)  ; LinAlgMatrixStoreToDescriptor(matrix,handle,offset,stride,layout,align)
 
   ; dx.op.linAlgMatrixAccumulateToMemory
-  call void @dx.op.linAlgMatrixAccumulateToMemory.mC4M5N4U0S2.f32(i32 -2147483620, %dx.types.LinAlgMatrixC4M5N4U0S2 %v14, float addrspace(3)* getelementptr inbounds ([64 x float], [64 x float] addrspace(3)* @"\01?SharedArr@@3PAMA", i32 0, i32 0), i32 0, i32 0, i32 0)  ; LinAlgMatrixAccumulateToMemory(matrix,memory,offset,stride,layout)
+  call void @dx.op.linAlgMatrixAccumulateToMemory.mC4M5N4U0S2.f32(i32 -2147483620, %dx.types.LinAlgMatrixC4M5N4U0S2 %v14, float addrspace(3)* getelementptr inbounds ([64 x float], [64 x float] addrspace(3)* @"\01?SharedArr@@3PAMA", i32 0, i32 0), i32 0, i32 0, i32 0, i32 0)  ; LinAlgMatrixAccumulateToMemory(matrix,memory,targetType,offset,stride,layout)
 
   ; dx.op.linAlgMatrixLoadFromMemory
   %v15 = call %dx.types.LinAlgMatrixC4M5N4U0S2 @dx.op.linAlgMatrixLoadFromMemory.mC4M5N4U0S2.f32(i32 -2147483633, float addrspace(3)* getelementptr inbounds ([64 x float], [64 x float] addrspace(3)* @"\01?SharedArr@@3PAMA", i32 0, i32 0), i32 0, i32 0, i32 0)  ; LinAlgMatrixLoadFromMemory(memory,offset,stride,layout)
@@ -201,7 +201,7 @@ declare %dx.types.LinAlgMatrixC4M5N4U0S2 @dx.op.linAlgMatrixSetElement.mC4M5N4U0
 declare void @dx.op.linAlgMatrixStoreToMemory.mC4M5N4U0S2.f32(i32, %dx.types.LinAlgMatrixC4M5N4U0S2, float addrspace(3)*, i32, i32, i32) #0
 
 ; Function Attrs: nounwind
-declare void @dx.op.linAlgMatrixAccumulateToMemory.mC4M5N4U0S2.f32(i32, %dx.types.LinAlgMatrixC4M5N4U0S2, float addrspace(3)*, i32, i32, i32) #0
+declare void @dx.op.linAlgMatrixAccumulateToMemory.mC4M5N4U0S2.f32(i32, %dx.types.LinAlgMatrixC4M5N4U0S2, float addrspace(3)*, i32, i32, i32, i32) #0
 
 ; Function Attrs: nounwind
 declare %dx.types.LinAlgMatrixC4M5N4U0S2 @dx.op.linAlgMatrixLoadFromMemory.mC4M5N4U0S2.f32(i32, float addrspace(3)*, i32, i32, i32) #0
diff --git a/tools/clang/test/LitDXILValidation/LinAlgMatrix/linalgmatrix-ms.ll b/tools/clang/test/LitDXILValidation/LinAlgMatrix/linalgmatrix-ms.ll
index d6d590e4e0..3fd26a40e3 100644
--- a/tools/clang/test/LitDXILValidation/LinAlgMatrix/linalgmatrix-ms.ll
+++ b/tools/clang/test/LitDXILValidation/LinAlgMatrix/linalgmatrix-ms.ll
@@ -116,7 +116,7 @@ define void @mainMeS() {
   call void @dx.op.linAlgMatrixStoreToDescriptor.mC4M5N4U0S2(i32 -2147483628, %dx.types.LinAlgMatrixC4M5N4U0S2 %v14, %dx.types.Handle %handle, i32 1, i32 2, i32 3, i32 4)  ; LinAlgMatrixStoreToDescriptor(matrix,handle,offset,stride,layout,align)
 
   ; dx.op.linAlgMatrixAccumulateToMemory
-  call void @dx.op.linAlgMatrixAccumulateToMemory.mC4M5N4U0S2.f32(i32 -2147483620, %dx.types.LinAlgMatrixC4M5N4U0S2 %v14, float addrspace(3)* getelementptr inbounds ([64 x float], [64 x float] addrspace(3)* @"\01?SharedArr@@3PAMA", i32 0, i32 0), i32 0, i32 0, i32 0)  ; LinAlgMatrixAccumulateToMemory(matrix,memory,offset,stride,layout)
+  call void @dx.op.linAlgMatrixAccumulateToMemory.mC4M5N4U0S2.f32(i32 -2147483620, %dx.types.LinAlgMatrixC4M5N4U0S2 %v14, float addrspace(3)* getelementptr inbounds ([64 x float], [64 x float] addrspace(3)* @"\01?SharedArr@@3PAMA", i32 0, i32 0), i32 0, i32 0, i32 0, i32 0)  ; LinAlgMatrixAccumulateToMemory(matrix,memory,targetType,offset,stride,layout)
 
   ; dx.op.linAlgMatrixLoadFromMemory
   %v15 = call %dx.types.LinAlgMatrixC4M5N4U0S2 @dx.op.linAlgMatrixLoadFromMemory.mC4M5N4U0S2.f32(i32 -2147483633, float addrspace(3)* getelementptr inbounds ([64 x float], [64 x float] addrspace(3)* @"\01?SharedArr@@3PAMA", i32 0, i32 0), i32 0, i32 0, i32 0)  ; LinAlgMatrixLoadFromMemory(memory,offset,stride,layout)
@@ -194,7 +194,7 @@ declare %dx.types.LinAlgMatrixC4M5N4U0S2 @dx.op.linAlgMatrixSetElement.mC4M5N4U0
 declare void @dx.op.linAlgMatrixStoreToMemory.mC4M5N4U0S2.f32(i32, %dx.types.LinAlgMatrixC4M5N4U0S2, float addrspace(3)*, i32, i32, i32) #0
 
 ; Function Attrs: nounwind
-declare void @dx.op.linAlgMatrixAccumulateToMemory.mC4M5N4U0S2.f32(i32, %dx.types.LinAlgMatrixC4M5N4U0S2, float addrspace(3)*, i32, i32, i32) #0
+declare void @dx.op.linAlgMatrixAccumulateToMemory.mC4M5N4U0S2.f32(i32, %dx.types.LinAlgMatrixC4M5N4U0S2, float addrspace(3)*, i32, i32, i32, i32) #0
 
 ; Function Attrs: nounwind
 declare %dx.types.LinAlgMatrixC4M5N4U0S2 @dx.op.linAlgMatrixLoadFromMemory.mC4M5N4U0S2.f32(i32, float addrspace(3)*, i32, i32, i32) #0
diff --git a/tools/clang/test/LitDXILValidation/LinAlgMatrix/linalgmatrix-node.ll b/tools/clang/test/LitDXILValidation/LinAlgMatrix/linalgmatrix-node.ll
index f1850a5051..def67f1a3d 100644
--- a/tools/clang/test/LitDXILValidation/LinAlgMatrix/linalgmatrix-node.ll
+++ b/tools/clang/test/LitDXILValidation/LinAlgMatrix/linalgmatrix-node.ll
@@ -117,7 +117,7 @@ define void @mainNS() {
   call void @dx.op.linAlgMatrixStoreToDescriptor.mC4M5N4U0S2(i32 -2147483628, %dx.types.LinAlgMatrixC4M5N4U0S2 %v14, %dx.types.Handle %handle, i32 1, i32 2, i32 3, i32 4)  ; LinAlgMatrixStoreToDescriptor(matrix,handle,offset,stride,layout,align)
 
   ; dx.op.linAlgMatrixAccumulateToMemory
-  call void @dx.op.linAlgMatrixAccumulateToMemory.mC4M5N4U0S2.f32(i32 -2147483620, %dx.types.LinAlgMatrixC4M5N4U0S2 %v14, float addrspace(3)* getelementptr inbounds ([64 x float], [64 x float] addrspace(3)* @"\01?SharedArr@@3PAMA", i32 0, i32 0), i32 0, i32 0, i32 0)  ; LinAlgMatrixAccumulateToMemory(matrix,memory,offset,stride,layout)
+  call void @dx.op.linAlgMatrixAccumulateToMemory.mC4M5N4U0S2.f32(i32 -2147483620, %dx.types.LinAlgMatrixC4M5N4U0S2 %v14, float addrspace(3)* getelementptr inbounds ([64 x float], [64 x float] addrspace(3)* @"\01?SharedArr@@3PAMA", i32 0, i32 0), i32 0, i32 0, i32 0, i32 0)  ; LinAlgMatrixAccumulateToMemory(matrix,memory,targetType,offset,stride,layout)
 
   ; dx.op.linAlgMatrixLoadFromMemory
   %v15 = call %dx.types.LinAlgMatrixC4M5N4U0S2 @dx.op.linAlgMatrixLoadFromMemory.mC4M5N4U0S2.f32(i32 -2147483633, float addrspace(3)* getelementptr inbounds ([64 x float], [64 x float] addrspace(3)* @"\01?SharedArr@@3PAMA", i32 0, i32 0), i32 0, i32 0, i32 0)  ; LinAlgMatrixLoadFromMemory(memory,offset,stride,layout)
@@ -189,7 +189,7 @@ declare %dx.types.LinAlgMatrixC4M5N4U0S2 @dx.op.linAlgMatrixSetElement.mC4M5N4U0
 declare void @dx.op.linAlgMatrixStoreToMemory.mC4M5N4U0S2.f32(i32, %dx.types.LinAlgMatrixC4M5N4U0S2, float addrspace(3)*, i32, i32, i32) #0
 
 ; Function Attrs: nounwind
-declare void @dx.op.linAlgMatrixAccumulateToMemory.mC4M5N4U0S2.f32(i32, %dx.types.LinAlgMatrixC4M5N4U0S2, float addrspace(3)*, i32, i32, i32) #0
+declare void @dx.op.linAlgMatrixAccumulateToMemory.mC4M5N4U0S2.f32(i32, %dx.types.LinAlgMatrixC4M5N4U0S2, float addrspace(3)*, i32, i32, i32, i32) #0
 
 ; Function Attrs: nounwind
 declare %dx.types.LinAlgMatrixC4M5N4U0S2 @dx.op.linAlgMatrixLoadFromMemory.mC4M5N4U0S2.f32(i32, float addrspace(3)*, i32, i32, i32) #0
diff --git a/tools/clang/test/LitDXILValidation/LinAlgMatrix/linalgmatrix-ps.ll b/tools/clang/test/LitDXILValidation/LinAlgMatrix/linalgmatrix-ps.ll
index 67a7e4669c..e1d531118c 100644
--- a/tools/clang/test/LitDXILValidation/LinAlgMatrix/linalgmatrix-ps.ll
+++ b/tools/clang/test/LitDXILValidation/LinAlgMatrix/linalgmatrix-ps.ll
@@ -116,7 +116,7 @@ define void @mainPS() {
   call void @dx.op.linAlgMatrixStoreToDescriptor.mC4M5N4U0S2(i32 -2147483628, %dx.types.LinAlgMatrixC4M5N4U0S2 %v14, %dx.types.Handle %handle, i32 1, i32 2, i32 3, i32 4)  ; LinAlgMatrixStoreToDescriptor(matrix,handle,offset,stride,layout,align)
 
   ; dx.op.linAlgMatrixAccumulateToMemory
-  call void @dx.op.linAlgMatrixAccumulateToMemory.mC4M5N4U0S2.f32(i32 -2147483620, %dx.types.LinAlgMatrixC4M5N4U0S2 %v14, float addrspace(3)* getelementptr inbounds ([64 x float], [64 x float] addrspace(3)* @"\01?SharedArr@@3PAMA", i32 0, i32 0), i32 0, i32 0, i32 0)  ; LinAlgMatrixAccumulateToMemory(matrix,memory,offset,stride,layout)
+  call void @dx.op.linAlgMatrixAccumulateToMemory.mC4M5N4U0S2.f32(i32 -2147483620, %dx.types.LinAlgMatrixC4M5N4U0S2 %v14, float addrspace(3)* getelementptr inbounds ([64 x float], [64 x float] addrspace(3)* @"\01?SharedArr@@3PAMA", i32 0, i32 0), i32 0, i32 0, i32 0, i32 0)  ; LinAlgMatrixAccumulateToMemory(matrix,memory,targetType,offset,stride,layout)
 
   ; dx.op.linAlgMatrixLoadFromMemory
   %v15 = call %dx.types.LinAlgMatrixC4M5N4U0S2 @dx.op.linAlgMatrixLoadFromMemory.mC4M5N4U0S2.f32(i32 -2147483633, float addrspace(3)* getelementptr inbounds ([64 x float], [64 x float] addrspace(3)* @"\01?SharedArr@@3PAMA", i32 0, i32 0), i32 0, i32 0, i32 0)  ; LinAlgMatrixLoadFromMemory(memory,offset,stride,layout)
@@ -193,7 +193,7 @@ declare %dx.types.LinAlgMatrixC4M5N4U0S2 @dx.op.linAlgMatrixSetElement.mC4M5N4U0
 declare void @dx.op.linAlgMatrixStoreToMemory.mC4M5N4U0S2.f32(i32, %dx.types.LinAlgMatrixC4M5N4U0S2, float addrspace(3)*, i32, i32, i32) #0
 
 ; Function Attrs: nounwind
-declare void @dx.op.linAlgMatrixAccumulateToMemory.mC4M5N4U0S2.f32(i32, %dx.types.LinAlgMatrixC4M5N4U0S2, float addrspace(3)*, i32, i32, i32) #0
+declare void @dx.op.linAlgMatrixAccumulateToMemory.mC4M5N4U0S2.f32(i32, %dx.types.LinAlgMatrixC4M5N4U0S2, float addrspace(3)*, i32, i32, i32, i32) #0
 
 ; Function Attrs: nounwind
 declare %dx.types.LinAlgMatrixC4M5N4U0S2 @dx.op.linAlgMatrixLoadFromMemory.mC4M5N4U0S2.f32(i32, float addrspace(3)*, i32, i32, i32) #0
diff --git a/tools/clang/test/LitDXILValidation/LinAlgMatrix/linalgmatrix-raytracing.ll b/tools/clang/test/LitDXILValidation/LinAlgMatrix/linalgmatrix-raytracing.ll
index f8bb658abe..e34bf177e7 100644
--- a/tools/clang/test/LitDXILValidation/LinAlgMatrix/linalgmatrix-raytracing.ll
+++ b/tools/clang/test/LitDXILValidation/LinAlgMatrix/linalgmatrix-raytracing.ll
@@ -245,7 +245,7 @@ define void @"\01?MainRG@@YAXXZ"() #0 {
   %v13 = call %dx.types.LinAlgMatrixC4M5N4U2S2 @dx.op.linAlgMatrixMultiplyAccumulate.mC4M5N4U2S2.mC4M5N4U0S2.mC4M4N5U1S2.mC4M5N4U2S2(i32 -2147483637, %dx.types.LinAlgMatrixC4M5N4U0S2 %v9, %dx.types.LinAlgMatrixC4M4N5U1S2 %v8, %dx.types.LinAlgMatrixC4M5N4U2S2 %v12)  ; LinAlgMatrixMultiplyAccumulate(matrixA,matrixB,matrixC)
   %v14 = call %dx.types.LinAlgMatrixC4M5N4U0S2 @dx.op.linAlgMatrixSetElement.mC4M5N4U0S2.mC4M5N4U0S2.i32(i32 -2147483629, %dx.types.LinAlgMatrixC4M5N4U0S2 %v9, i32 1, i32 1)  ; LinAlgMatrixSetElement(matrix,threadLocalIndex,value)
   call void @dx.op.linAlgMatrixStoreToDescriptor.mC4M5N4U0S2(i32 -2147483628, %dx.types.LinAlgMatrixC4M5N4U0S2 %v14, %dx.types.Handle %handle, i32 1, i32 2, i32 3, i32 4)  ; LinAlgMatrixStoreToDescriptor(matrix,handle,offset,stride,layout)
-  call void @dx.op.linAlgMatrixAccumulateToMemory.mC4M5N4U0S2.f32(i32 -2147483620, %dx.types.LinAlgMatrixC4M5N4U0S2 %v14, float addrspace(3)* getelementptr inbounds ([64 x float], [64 x float] addrspace(3)* @"\01?SharedArr@@3PAMA", i32 0, i32 0), i32 0, i32 0, i32 0)  ; LinAlgMatrixAccumulateToMemory(matrix,memory,offset,stride,layout)
+  call void @dx.op.linAlgMatrixAccumulateToMemory.mC4M5N4U0S2.f32(i32 -2147483620, %dx.types.LinAlgMatrixC4M5N4U0S2 %v14, float addrspace(3)* getelementptr inbounds ([64 x float], [64 x float] addrspace(3)* @"\01?SharedArr@@3PAMA", i32 0, i32 0), i32 0, i32 0, i32 0, i32 0)  ; LinAlgMatrixAccumulateToMemory(matrix,memory,targetType,offset,stride,layout)
   %v15 = call %dx.types.LinAlgMatrixC4M5N4U0S2 @dx.op.linAlgMatrixLoadFromMemory.mC4M5N4U0S2.f32(i32 -2147483633, float addrspace(3)* getelementptr inbounds ([64 x float], [64 x float] addrspace(3)* @"\01?SharedArr@@3PAMA", i32 0, i32 0), i32 0, i32 0, i32 0)  ; LinAlgMatrixLoadFromMemory(memory,offset,stride,layout)
   call void @dx.op.linAlgMatrixStoreToMemory.mC4M5N4U0S2.f32(i32 -2147483627, %dx.types.LinAlgMatrixC4M5N4U0S2 %v15, float addrspace(3)* getelementptr inbounds ([64 x float], [64 x float] addrspace(3)* @"\01?SharedArr@@3PAMA", i32 0, i32 0), i32 0, i32 0, i32 0)  ; LinAlgMatrixStoreToMemory(matrix,memory,offset,stride,layout)
 
@@ -285,7 +285,7 @@ define void @"\01?MainIS@@YAXXZ"() #0 {
   %v13 = call %dx.types.LinAlgMatrixC4M5N4U2S2 @dx.op.linAlgMatrixMultiplyAccumulate.mC4M5N4U2S2.mC4M5N4U0S2.mC4M4N5U1S2.mC4M5N4U2S2(i32 -2147483637, %dx.types.LinAlgMatrixC4M5N4U0S2 %v9, %dx.types.LinAlgMatrixC4M4N5U1S2 %v8, %dx.types.LinAlgMatrixC4M5N4U2S2 %v12)  ; LinAlgMatrixMultiplyAccumulate(matrixA,matrixB,matrixC)
   %v14 = call %dx.types.LinAlgMatrixC4M5N4U0S2 @dx.op.linAlgMatrixSetElement.mC4M5N4U0S2.mC4M5N4U0S2.i32(i32 -2147483629, %dx.types.LinAlgMatrixC4M5N4U0S2 %v9, i32 1, i32 1)  ; LinAlgMatrixSetElement(matrix,threadLocalIndex,value)
   call void @dx.op.linAlgMatrixStoreToDescriptor.mC4M5N4U0S2(i32 -2147483628, %dx.types.LinAlgMatrixC4M5N4U0S2 %v14, %dx.types.Handle %handle, i32 1, i32 2, i32 3, i32 4)  ; LinAlgMatrixStoreToDescriptor(matrix,handle,offset,stride,layout)
-  call void @dx.op.linAlgMatrixAccumulateToMemory.mC4M5N4U0S2.f32(i32 -2147483620, %dx.types.LinAlgMatrixC4M5N4U0S2 %v14, float addrspace(3)* getelementptr inbounds ([64 x float], [64 x float] addrspace(3)* @"\01?SharedArr@@3PAMA", i32 0, i32 0), i32 0, i32 0, i32 0)  ; LinAlgMatrixAccumulateToMemory(matrix,memory,offset,stride,layout)
+  call void @dx.op.linAlgMatrixAccumulateToMemory.mC4M5N4U0S2.f32(i32 -2147483620, %dx.types.LinAlgMatrixC4M5N4U0S2 %v14, float addrspace(3)* getelementptr inbounds ([64 x float], [64 x float] addrspace(3)* @"\01?SharedArr@@3PAMA", i32 0, i32 0), i32 0, i32 0, i32 0, i32 0)  ; LinAlgMatrixAccumulateToMemory(matrix,memory,targetType,offset,stride,layout)
   %v15 = call %dx.types.LinAlgMatrixC4M5N4U0S2 @dx.op.linAlgMatrixLoadFromMemory.mC4M5N4U0S2.f32(i32 -2147483633, float addrspace(3)* getelementptr inbounds ([64 x float], [64 x float] addrspace(3)* @"\01?SharedArr@@3PAMA", i32 0, i32 0), i32 0, i32 0, i32 0)  ; LinAlgMatrixLoadFromMemory(memory,offset,stride,layout)
   call void @dx.op.linAlgMatrixStoreToMemory.mC4M5N4U0S2.f32(i32 -2147483627, %dx.types.LinAlgMatrixC4M5N4U0S2 %v15, float addrspace(3)* getelementptr inbounds ([64 x float], [64 x float] addrspace(3)* @"\01?SharedArr@@3PAMA", i32 0, i32 0), i32 0, i32 0, i32 0)  ; LinAlgMatrixStoreToMemory(matrix,memory,offset,stride,layout)
 
@@ -325,7 +325,7 @@ define void @"\01?MainCL@@YAXUAttribs@@@Z"(%struct.Attribs* noalias nocapture %a
   %v13 = call %dx.types.LinAlgMatrixC4M5N4U2S2 @dx.op.linAlgMatrixMultiplyAccumulate.mC4M5N4U2S2.mC4M5N4U0S2.mC4M4N5U1S2.mC4M5N4U2S2(i32 -2147483637, %dx.types.LinAlgMatrixC4M5N4U0S2 %v9, %dx.types.LinAlgMatrixC4M4N5U1S2 %v8, %dx.types.LinAlgMatrixC4M5N4U2S2 %v12)  ; LinAlgMatrixMultiplyAccumulate(matrixA,matrixB,matrixC)
   %v14 = call %dx.types.LinAlgMatrixC4M5N4U0S2 @dx.op.linAlgMatrixSetElement.mC4M5N4U0S2.mC4M5N4U0S2.i32(i32 -2147483629, %dx.types.LinAlgMatrixC4M5N4U0S2 %v9, i32 1, i32 1)  ; LinAlgMatrixSetElement(matrix,threadLocalIndex,value)
   call void @dx.op.linAlgMatrixStoreToDescriptor.mC4M5N4U0S2(i32 -2147483628, %dx.types.LinAlgMatrixC4M5N4U0S2 %v14, %dx.types.Handle %handle, i32 1, i32 2, i32 3, i32 4)  ; LinAlgMatrixStoreToDescriptor(matrix,handle,offset,stride,layout)
-  call void @dx.op.linAlgMatrixAccumulateToMemory.mC4M5N4U0S2.f32(i32 -2147483620, %dx.types.LinAlgMatrixC4M5N4U0S2 %v14, float addrspace(3)* getelementptr inbounds ([64 x float], [64 x float] addrspace(3)* @"\01?SharedArr@@3PAMA", i32 0, i32 0), i32 0, i32 0, i32 0)  ; LinAlgMatrixAccumulateToMemory(matrix,memory,offset,stride,layout)
+  call void @dx.op.linAlgMatrixAccumulateToMemory.mC4M5N4U0S2.f32(i32 -2147483620, %dx.types.LinAlgMatrixC4M5N4U0S2 %v14, float addrspace(3)* getelementptr inbounds ([64 x float], [64 x float] addrspace(3)* @"\01?SharedArr@@3PAMA", i32 0, i32 0), i32 0, i32 0, i32 0, i32 0)  ; LinAlgMatrixAccumulateToMemory(matrix,memory,targetType,offset,stride,layout)
   %v15 = call %dx.types.LinAlgMatrixC4M5N4U0S2 @dx.op.linAlgMatrixLoadFromMemory.mC4M5N4U0S2.f32(i32 -2147483633, float addrspace(3)* getelementptr inbounds ([64 x float], [64 x float] addrspace(3)* @"\01?SharedArr@@3PAMA", i32 0, i32 0), i32 0, i32 0, i32 0)  ; LinAlgMatrixLoadFromMemory(memory,offset,stride,layout)
   call void @dx.op.linAlgMatrixStoreToMemory.mC4M5N4U0S2.f32(i32 -2147483627, %dx.types.LinAlgMatrixC4M5N4U0S2 %v15, float addrspace(3)* getelementptr inbounds ([64 x float], [64 x float] addrspace(3)* @"\01?SharedArr@@3PAMA", i32 0, i32 0), i32 0, i32 0, i32 0)  ; LinAlgMatrixStoreToMemory(matrix,memory,offset,stride,layout)
 
@@ -365,7 +365,7 @@ define void @"\01?MainAH@@YAXURayPayload@@UAttribs@@@Z"(%struct.RayPayload* noal
   %v13 = call %dx.types.LinAlgMatrixC4M5N4U2S2 @dx.op.linAlgMatrixMultiplyAccumulate.mC4M5N4U2S2.mC4M5N4U0S2.mC4M4N5U1S2.mC4M5N4U2S2(i32 -2147483637, %dx.types.LinAlgMatrixC4M5N4U0S2 %v9, %dx.types.LinAlgMatrixC4M4N5U1S2 %v8, %dx.types.LinAlgMatrixC4M5N4U2S2 %v12)  ; LinAlgMatrixMultiplyAccumulate(matrixA,matrixB,matrixC)
   %v14 = call %dx.types.LinAlgMatrixC4M5N4U0S2 @dx.op.linAlgMatrixSetElement.mC4M5N4U0S2.mC4M5N4U0S2.i32(i32 -2147483629, %dx.types.LinAlgMatrixC4M5N4U0S2 %v9, i32 1, i32 1)  ; LinAlgMatrixSetElement(matrix,threadLocalIndex,value)
   call void @dx.op.linAlgMatrixStoreToDescriptor.mC4M5N4U0S2(i32 -2147483628, %dx.types.LinAlgMatrixC4M5N4U0S2 %v14, %dx.types.Handle %handle, i32 1, i32 2, i32 3, i32 4)  ; LinAlgMatrixStoreToDescriptor(matrix,handle,offset,stride,layout)
-  call void @dx.op.linAlgMatrixAccumulateToMemory.mC4M5N4U0S2.f32(i32 -2147483620, %dx.types.LinAlgMatrixC4M5N4U0S2 %v14, float addrspace(3)* getelementptr inbounds ([64 x float], [64 x float] addrspace(3)* @"\01?SharedArr@@3PAMA", i32 0, i32 0), i32 0, i32 0, i32 0)  ; LinAlgMatrixAccumulateToMemory(matrix,memory,offset,stride,layout)
+  call void @dx.op.linAlgMatrixAccumulateToMemory.mC4M5N4U0S2.f32(i32 -2147483620, %dx.types.LinAlgMatrixC4M5N4U0S2 %v14, float addrspace(3)* getelementptr inbounds ([64 x float], [64 x float] addrspace(3)* @"\01?SharedArr@@3PAMA", i32 0, i32 0), i32 0, i32 0, i32 0, i32 0)  ; LinAlgMatrixAccumulateToMemory(matrix,memory,targetType,offset,stride,layout)
   %v15 = call %dx.types.LinAlgMatrixC4M5N4U0S2 @dx.op.linAlgMatrixLoadFromMemory.mC4M5N4U0S2.f32(i32 -2147483633, float addrspace(3)* getelementptr inbounds ([64 x float], [64 x float] addrspace(3)* @"\01?SharedArr@@3PAMA", i32 0, i32 0), i32 0, i32 0, i32 0)  ; LinAlgMatrixLoadFromMemory(memory,offset,stride,layout)
   call void @dx.op.linAlgMatrixStoreToMemory.mC4M5N4U0S2.f32(i32 -2147483627, %dx.types.LinAlgMatrixC4M5N4U0S2 %v15, float addrspace(3)* getelementptr inbounds ([64 x float], [64 x float] addrspace(3)* @"\01?SharedArr@@3PAMA", i32 0, i32 0), i32 0, i32 0, i32 0)  ; LinAlgMatrixStoreToMemory(matrix,memory,offset,stride,layout)
 
@@ -405,7 +405,7 @@ define void @"\01?MainCH@@YAXURayPayload@@UAttribs@@@Z"(%struct.RayPayload* noal
   %v13 = call %dx.types.LinAlgMatrixC4M5N4U2S2 @dx.op.linAlgMatrixMultiplyAccumulate.mC4M5N4U2S2.mC4M5N4U0S2.mC4M4N5U1S2.mC4M5N4U2S2(i32 -2147483637, %dx.types.LinAlgMatrixC4M5N4U0S2 %v9, %dx.types.LinAlgMatrixC4M4N5U1S2 %v8, %dx.types.LinAlgMatrixC4M5N4U2S2 %v12)  ; LinAlgMatrixMultiplyAccumulate(matrixA,matrixB,matrixC)
   %v14 = call %dx.types.LinAlgMatrixC4M5N4U0S2 @dx.op.linAlgMatrixSetElement.mC4M5N4U0S2.mC4M5N4U0S2.i32(i32 -2147483629, %dx.types.LinAlgMatrixC4M5N4U0S2 %v9, i32 1, i32 1)  ; LinAlgMatrixSetElement(matrix,threadLocalIndex,value)
   call void @dx.op.linAlgMatrixStoreToDescriptor.mC4M5N4U0S2(i32 -2147483628, %dx.types.LinAlgMatrixC4M5N4U0S2 %v14, %dx.types.Handle %handle, i32 1, i32 2, i32 3, i32 4)  ; LinAlgMatrixStoreToDescriptor(matrix,handle,offset,stride,layout)
-  call void @dx.op.linAlgMatrixAccumulateToMemory.mC4M5N4U0S2.f32(i32 -2147483620, %dx.types.LinAlgMatrixC4M5N4U0S2 %v14, float addrspace(3)* getelementptr inbounds ([64 x float], [64 x float] addrspace(3)* @"\01?SharedArr@@3PAMA", i32 0, i32 0), i32 0, i32 0, i32 0)  ; LinAlgMatrixAccumulateToMemory(matrix,memory,offset,stride,layout)
+  call void @dx.op.linAlgMatrixAccumulateToMemory.mC4M5N4U0S2.f32(i32 -2147483620, %dx.types.LinAlgMatrixC4M5N4U0S2 %v14, float addrspace(3)* getelementptr inbounds ([64 x float], [64 x float] addrspace(3)* @"\01?SharedArr@@3PAMA", i32 0, i32 0), i32 0, i32 0, i32 0, i32 0)  ; LinAlgMatrixAccumulateToMemory(matrix,memory,targetType,offset,stride,layout)
   %v15 = call %dx.types.LinAlgMatrixC4M5N4U0S2 @dx.op.linAlgMatrixLoadFromMemory.mC4M5N4U0S2.f32(i32 -2147483633, float addrspace(3)* getelementptr inbounds ([64 x float], [64 x float] addrspace(3)* @"\01?SharedArr@@3PAMA", i32 0, i32 0), i32 0, i32 0, i32 0)  ; LinAlgMatrixLoadFromMemory(memory,offset,stride,layout)
   call void @dx.op.linAlgMatrixStoreToMemory.mC4M5N4U0S2.f32(i32 -2147483627, %dx.types.LinAlgMatrixC4M5N4U0S2 %v15, float addrspace(3)* getelementptr inbounds ([64 x float], [64 x float] addrspace(3)* @"\01?SharedArr@@3PAMA", i32 0, i32 0), i32 0, i32 0, i32 0)  ; LinAlgMatrixStoreToMemory(matrix,memory,offset,stride,layout)
 
@@ -445,7 +445,7 @@ define void @"\01?MainMS@@YAXURayPayload@@@Z"(%struct.RayPayload* noalias nocapt
   %v13 = call %dx.types.LinAlgMatrixC4M5N4U2S2 @dx.op.linAlgMatrixMultiplyAccumulate.mC4M5N4U2S2.mC4M5N4U0S2.mC4M4N5U1S2.mC4M5N4U2S2(i32 -2147483637, %dx.types.LinAlgMatrixC4M5N4U0S2 %v9, %dx.types.LinAlgMatrixC4M4N5U1S2 %v8, %dx.types.LinAlgMatrixC4M5N4U2S2 %v12)  ; LinAlgMatrixMultiplyAccumulate(matrixA,matrixB,matrixC)
   %v14 = call %dx.types.LinAlgMatrixC4M5N4U0S2 @dx.op.linAlgMatrixSetElement.mC4M5N4U0S2.mC4M5N4U0S2.i32(i32 -2147483629, %dx.types.LinAlgMatrixC4M5N4U0S2 %v9, i32 1, i32 1)  ; LinAlgMatrixSetElement(matrix,threadLocalIndex,value)
   call void @dx.op.linAlgMatrixStoreToDescriptor.mC4M5N4U0S2(i32 -2147483628, %dx.types.LinAlgMatrixC4M5N4U0S2 %v14, %dx.types.Handle %handle, i32 1, i32 2, i32 3, i32 4)  ; LinAlgMatrixStoreToDescriptor(matrix,handle,offset,stride,layout)
-  call void @dx.op.linAlgMatrixAccumulateToMemory.mC4M5N4U0S2.f32(i32 -2147483620, %dx.types.LinAlgMatrixC4M5N4U0S2 %v14, float addrspace(3)* getelementptr inbounds ([64 x float], [64 x float] addrspace(3)* @"\01?SharedArr@@3PAMA", i32 0, i32 0), i32 0, i32 0, i32 0)  ; LinAlgMatrixAccumulateToMemory(matrix,memory,offset,stride,layout)
+  call void @dx.op.linAlgMatrixAccumulateToMemory.mC4M5N4U0S2.f32(i32 -2147483620, %dx.types.LinAlgMatrixC4M5N4U0S2 %v14, float addrspace(3)* getelementptr inbounds ([64 x float], [64 x float] addrspace(3)* @"\01?SharedArr@@3PAMA", i32 0, i32 0), i32 0, i32 0, i32 0, i32 0)  ; LinAlgMatrixAccumulateToMemory(matrix,memory,targetType,offset,stride,layout)
   %v15 = call %dx.types.LinAlgMatrixC4M5N4U0S2 @dx.op.linAlgMatrixLoadFromMemory.mC4M5N4U0S2.f32(i32 -2147483633, float addrspace(3)* getelementptr inbounds ([64 x float], [64 x float] addrspace(3)* @"\01?SharedArr@@3PAMA", i32 0, i32 0), i32 0, i32 0, i32 0)  ; LinAlgMatrixLoadFromMemory(memory,offset,stride,layout)
   call void @dx.op.linAlgMatrixStoreToMemory.mC4M5N4U0S2.f32(i32 -2147483627, %dx.types.LinAlgMatrixC4M5N4U0S2 %v15, float addrspace(3)* getelementptr inbounds ([64 x float], [64 x float] addrspace(3)* @"\01?SharedArr@@3PAMA", i32 0, i32 0), i32 0, i32 0, i32 0)  ; LinAlgMatrixStoreToMemory(matrix,memory,offset,stride,layout)
 
@@ -513,7 +513,7 @@ declare %dx.types.LinAlgMatrixC4M5N4U0S2 @dx.op.linAlgMatrixSetElement.mC4M5N4U0
 declare void @dx.op.linAlgMatrixStoreToMemory.mC4M5N4U0S2.f32(i32, %dx.types.LinAlgMatrixC4M5N4U0S2, float addrspace(3)*, i32, i32, i32) #0
 
 ; Function Attrs: nounwind
-declare void @dx.op.linAlgMatrixAccumulateToMemory.mC4M5N4U0S2.f32(i32, %dx.types.LinAlgMatrixC4M5N4U0S2, float addrspace(3)*, i32, i32, i32) #0
+declare void @dx.op.linAlgMatrixAccumulateToMemory.mC4M5N4U0S2.f32(i32, %dx.types.LinAlgMatrixC4M5N4U0S2, float addrspace(3)*, i32, i32, i32, i32) #0
 
 ; Function Attrs: nounwind
 declare %dx.types.LinAlgMatrixC4M5N4U0S2 @dx.op.linAlgMatrixLoadFromMemory.mC4M5N4U0S2.f32(i32, float addrspace(3)*, i32, i32, i32) #0
diff --git a/tools/clang/test/LitDXILValidation/LinAlgMatrix/linalgmatrix-vs.ll b/tools/clang/test/LitDXILValidation/LinAlgMatrix/linalgmatrix-vs.ll
index bd894fa6d8..7998e1857d 100644
--- a/tools/clang/test/LitDXILValidation/LinAlgMatrix/linalgmatrix-vs.ll
+++ b/tools/clang/test/LitDXILValidation/LinAlgMatrix/linalgmatrix-vs.ll
@@ -62,7 +62,7 @@ define void @mainVS() {
   %v1 = call %dx.types.LinAlgMatrixC4M5N4U2S2 @dx.op.linAlgMatrixAccumulate.mC4M5N4U2S2.mC4M5N4U0S2.mC4M4N5U1S2(i32 -2147483624, %dx.types.LinAlgMatrixC4M5N4U0S2 %mC4M5N4U0S2, %dx.types.LinAlgMatrixC4M4N5U1S2 %mC4M4N5U1S2)  ; LinAlgMatrixAccumulate(matrixLHS,matrixRHS)
 
   ; dx.op.linAlgMatrixAccumulateToDescriptor
-  call void @dx.op.linAlgMatrixAccumulateToDescriptor.mC4M5N4U0S2(i32 -2147483621, %dx.types.LinAlgMatrixC4M5N4U0S2 %mC4M5N4U0S2, %dx.types.Handle %handle, i32 1, i32 2, i32 3, i32 4)  ; LinAlgMatrixAccumulateToDescriptor(matrix,handle,offset,stride,layout)
+  call void @dx.op.linAlgMatrixAccumulateToDescriptor.mC4M5N4U0S2(i32 -2147483621, %dx.types.LinAlgMatrixC4M5N4U0S2 %mC4M5N4U0S2, %dx.types.Handle %handle, i32 1, i32 2, i32 3, i32 4)  ; LinAlgMatrixAccumulateToDescriptor(matrix,handle,targetType,offset,stride,layout)
 
   ; dx.op.linAlgMatrixLength
   %v2 = call i32 @dx.op.linAlgMatrixLength.mC4M5N4U0S2(i32 -2147483632, %dx.types.LinAlgMatrixC4M5N4U0S2 %mC4M5N4U0S2)  ; LinAlgMatrixLength(matrix)
@@ -117,7 +117,7 @@ define void @mainVS() {
   call void @dx.op.linAlgMatrixStoreToDescriptor.mC4M5N4U0S2(i32 -2147483628, %dx.types.LinAlgMatrixC4M5N4U0S2 %v14, %dx.types.Handle %handle, i32 1, i32 2, i32 3, i32 4)  ; LinAlgMatrixStoreToDescriptor(matrix,handle,offset,stride,layout)
 
   ; dx.op.linAlgMatrixAccumulateToMemory
-  call void @dx.op.linAlgMatrixAccumulateToMemory.mC4M5N4U0S2.f32(i32 -2147483620, %dx.types.LinAlgMatrixC4M5N4U0S2 %v14, float addrspace(3)* getelementptr inbounds ([64 x float], [64 x float] addrspace(3)* @"\01?SharedArr@@3PAMA", i32 0, i32 0), i32 0, i32 0, i32 0)  ; LinAlgMatrixAccumulateToMemory(matrix,memory,offset,stride,layout)
+  call void @dx.op.linAlgMatrixAccumulateToMemory.mC4M5N4U0S2.f32(i32 -2147483620, %dx.types.LinAlgMatrixC4M5N4U0S2 %v14, float addrspace(3)* getelementptr inbounds ([64 x float], [64 x float] addrspace(3)* @"\01?SharedArr@@3PAMA", i32 0, i32 0), i32 0, i32 0, i32 0, i32 0)  ; LinAlgMatrixAccumulateToMemory(matrix,memory,targetType,offset,stride,layout)
 
   ; dx.op.linAlgMatrixLoadFromMemory
   %v15 = call %dx.types.LinAlgMatrixC4M5N4U0S2 @dx.op.linAlgMatrixLoadFromMemory.mC4M5N4U0S2.f32(i32 -2147483633, float addrspace(3)* getelementptr inbounds ([64 x float], [64 x float] addrspace(3)* @"\01?SharedArr@@3PAMA", i32 0, i32 0), i32 0, i32 0, i32 0)  ; LinAlgMatrixLoadFromMemory(memory,offset,stride,layout)
@@ -194,7 +194,7 @@ declare %dx.types.LinAlgMatrixC4M5N4U0S2 @dx.op.linAlgMatrixSetElement.mC4M5N4U0
 declare void @dx.op.linAlgMatrixStoreToMemory.mC4M5N4U0S2.f32(i32, %dx.types.LinAlgMatrixC4M5N4U0S2, float addrspace(3)*, i32, i32, i32) #0
 
 ; Function Attrs: nounwind
-declare void @dx.op.linAlgMatrixAccumulateToMemory.mC4M5N4U0S2.f32(i32, %dx.types.LinAlgMatrixC4M5N4U0S2, float addrspace(3)*, i32, i32, i32) #0
+declare void @dx.op.linAlgMatrixAccumulateToMemory.mC4M5N4U0S2.f32(i32, %dx.types.LinAlgMatrixC4M5N4U0S2, float addrspace(3)*, i32, i32, i32, i32) #0
 
 ; Function Attrs: nounwind
 declare %dx.types.LinAlgMatrixC4M5N4U0S2 @dx.op.linAlgMatrixLoadFromMemory.mC4M5N4U0S2.f32(i32, float addrspace(3)*, i32, i32, i32) #0
diff --git a/tools/clang/test/SemaHLSL/hlsl/linalg/builtins/matrix-builtins-ast.hlsl b/tools/clang/test/SemaHLSL/hlsl/linalg/builtins/matrix-builtins-ast.hlsl
index c8ddb896ad..e6797756bc 100644
--- a/tools/clang/test/SemaHLSL/hlsl/linalg/builtins/matrix-builtins-ast.hlsl
+++ b/tools/clang/test/SemaHLSL/hlsl/linalg/builtins/matrix-builtins-ast.hlsl
@@ -45,15 +45,16 @@ void main() {
 // CHECK-NEXT: AvailabilityAttr {{.*}} Implicit  6.10 0 0 ""
   __builtin_LinAlg_MatrixAccumulateToDescriptor(mat1, Buf, 1, 2, 3, 4);
 
-// CHECK: FunctionDecl {{.*}} implicit used __builtin_LinAlg_MatrixAccumulateToMemory 'void (__builtin_LinAlgMatrix {{.*}}, float const __attribute__((address_space(3))) (&)[64], unsigned int, unsigned int, unsigned int)' extern
+// CHECK: FunctionDecl {{.*}} implicit used __builtin_LinAlg_MatrixAccumulateToMemory 'void (__builtin_LinAlgMatrix {{.*}}, float const __attribute__((address_space(3))) (&)[64], unsigned int, unsigned int, unsigned int, unsigned int)' extern
 // CHECK-NEXT: ParmVarDecl {{.*}} matrix '__builtin_LinAlgMatrix {{.*}}'
 // CHECK-NEXT: ParmVarDecl {{.*}} memory 'float const __attribute__((address_space(3))) (&)[64]'
+// CHECK-NEXT: ParmVarDecl {{.*}} targetType 'unsigned int'
 // CHECK-NEXT: ParmVarDecl {{.*}} offset 'unsigned int'
 // CHECK-NEXT: ParmVarDecl {{.*}} stride 'unsigned int'
 // CHECK-NEXT: ParmVarDecl {{.*}} layout 'unsigned int'
 // CHECK-NEXT: HLSLIntrinsicAttr {{.*}} Implicit "op" "" 416
 // CHECK-NEXT: AvailabilityAttr {{.*}} Implicit  6.10 0 0 ""
-  __builtin_LinAlg_MatrixAccumulateToMemory(mat1, SharedArr, 0, 0, 0);
+  __builtin_LinAlg_MatrixAccumulateToMemory(mat1, SharedArr, 0, 0, 0, 0);
 
 // CHECK: FunctionDecl {{.*}} implicit used __builtin_LinAlg_MatrixGetCoordinate 'vector<uint, 2> (__builtin_LinAlgMatrix {{.*}}, unsigned int)' extern
 // CHECK-NEXT: ParmVarDecl {{.*}} matrix '__builtin_LinAlgMatrix {{.*}}'
diff --git a/tools/clang/test/SemaHLSL/hlsl/linalg/builtins/matrixaccumulatetomemory/ast.hlsl b/tools/clang/test/SemaHLSL/hlsl/linalg/builtins/matrixaccumulatetomemory/ast.hlsl
index 4cae1086bf..fcf8e5199f 100644
--- a/tools/clang/test/SemaHLSL/hlsl/linalg/builtins/matrixaccumulatetomemory/ast.hlsl
+++ b/tools/clang/test/SemaHLSL/hlsl/linalg/builtins/matrixaccumulatetomemory/ast.hlsl
@@ -1,9 +1,10 @@
 // REQUIRES: dxil-1-10
 // RUN: %dxc -T lib_6_10 -E main %s -ast-dump-implicit | FileCheck %s
 
-// CHECK: FunctionDecl {{.*}} implicit used __builtin_LinAlg_MatrixAccumulateToMemory 'void (__builtin_LinAlgMatrix {{.*}}, float const __attribute__((address_space(3))) (&)[64], unsigned int, unsigned int, unsigned int)' extern
+// CHECK: FunctionDecl {{.*}} implicit used __builtin_LinAlg_MatrixAccumulateToMemory 'void (__builtin_LinAlgMatrix {{.*}}, float const __attribute__((address_space(3))) (&)[64], unsigned int, unsigned int, unsigned int, unsigned int)' extern
 // CHECK-NEXT: ParmVarDecl {{.*}} matrix '__builtin_LinAlgMatrix {{.*}}'
 // CHECK-NEXT: ParmVarDecl {{.*}} memory 'float const __attribute__((address_space(3))) (&)[64]'
+// CHECK-NEXT: ParmVarDecl {{.*}} targetType 'unsigned int'
 // CHECK-NEXT: ParmVarDecl {{.*}} offset 'unsigned int'
 // CHECK-NEXT: ParmVarDecl {{.*}} stride 'unsigned int'
 // CHECK-NEXT: ParmVarDecl {{.*}} layout 'unsigned int'
@@ -16,5 +17,5 @@ groupshared float SharedArr[64];
 [numthreads(1,1,1)]
 void main() {
   __builtin_LinAlgMatrix [[__LinAlgMatrix_Attributes(4, 5, 4, 1, 2)]] mat;
-  __builtin_LinAlg_MatrixAccumulateToMemory(mat, SharedArr, 0, 0, 0);
+  __builtin_LinAlg_MatrixAccumulateToMemory(mat, SharedArr, 0, 0, 0, 0);
 }
diff --git a/tools/clang/test/SemaHLSL/hlsl/linalg/builtins/matrixaccumulatetomemory/unavailable_pre_sm610.hlsl b/tools/clang/test/SemaHLSL/hlsl/linalg/builtins/matrixaccumulatetomemory/unavailable_pre_sm610.hlsl
index e5a9ea4895..265ac0ac42 100644
--- a/tools/clang/test/SemaHLSL/hlsl/linalg/builtins/matrixaccumulatetomemory/unavailable_pre_sm610.hlsl
+++ b/tools/clang/test/SemaHLSL/hlsl/linalg/builtins/matrixaccumulatetomemory/unavailable_pre_sm610.hlsl
@@ -7,5 +7,5 @@ void main() {
   __builtin_LinAlgMatrix [[__LinAlgMatrix_Attributes(4, 5, 4, 1, 2)]] mat;
 
   // expected-error@+1{{intrinsic __builtin_LinAlg_MatrixAccumulateToMemory potentially used by ''main'' requires shader model 6.10 or greater}}
-  __builtin_LinAlg_MatrixAccumulateToMemory(mat, SharedArr, 0, 0, 0);
+  __builtin_LinAlg_MatrixAccumulateToMemory(mat, SharedArr, 0, 0, 0, 0);
 }
diff --git a/tools/clang/test/SemaHLSL/hlsl/linalg/builtins/stage-errors.hlsl b/tools/clang/test/SemaHLSL/hlsl/linalg/builtins/stage-errors.hlsl
index 9f33b32ded..b2b72279ac 100644
--- a/tools/clang/test/SemaHLSL/hlsl/linalg/builtins/stage-errors.hlsl
+++ b/tools/clang/test/SemaHLSL/hlsl/linalg/builtins/stage-errors.hlsl
@@ -75,7 +75,7 @@ void CallFunction()
 #endif
 
 #ifdef MATRIX_ACCUMULATE_TO_MEMORY
-  #define DO_FUNC __builtin_LinAlg_MatrixAccumulateToMemory(mat1, gs_arr, 0, 0, 0);
+  #define DO_FUNC __builtin_LinAlg_MatrixAccumulateToMemory(mat1, gs_arr, 0, 0, 0, 0);
 #endif
 
   // The builtins below are allowed in all stages, if they raise an error
diff --git a/tools/clang/test/SemaHLSL/hlsl/linalg/builtins/unavailable_pre_sm610.hlsl b/tools/clang/test/SemaHLSL/hlsl/linalg/builtins/unavailable_pre_sm610.hlsl
index a6e7b8234c..85bb28e70b 100644
--- a/tools/clang/test/SemaHLSL/hlsl/linalg/builtins/unavailable_pre_sm610.hlsl
+++ b/tools/clang/test/SemaHLSL/hlsl/linalg/builtins/unavailable_pre_sm610.hlsl
@@ -61,7 +61,7 @@ void main() {
   __builtin_LinAlg_MatrixVectorMultiplyAdd(result, mat, true, vec1, 1, vec2, 0);
 
   // expected-error@+1{{intrinsic __builtin_LinAlg_MatrixAccumulateToMemory potentially used by ''main'' requires shader model 6.10 or greater}}
-  __builtin_LinAlg_MatrixAccumulateToMemory(mat, SharedArr, 0, 0, 0);
+  __builtin_LinAlg_MatrixAccumulateToMemory(mat, SharedArr, 0, 0, 0, 0);
 
   // expected-error@+1{{intrinsic __builtin_LinAlg_MatrixLoadFromMemory potentially used by ''main'' requires shader model 6.10 or greater}}
   __builtin_LinAlg_MatrixLoadFromMemory(mat, SharedArr, 0, 0, 0);
diff --git a/tools/clang/unittests/HLSLExec/LinAlgTests.cpp b/tools/clang/unittests/HLSLExec/LinAlgTests.cpp
index 8a5b0c49f1..9fe6c91d71 100644
--- a/tools/clang/unittests/HLSLExec/LinAlgTests.cpp
+++ b/tools/clang/unittests/HLSLExec/LinAlgTests.cpp
@@ -1742,7 +1742,7 @@ static const char AccumulateMemoryShader[] = R"(
     __builtin_LinAlg_FillMatrix(Mat, FILL_VALUE);
 
     __builtin_LinAlg_MatrixAccumulateToMemory(
-      Mat, GsData, OFFSET / ELEM_SIZE, STRIDE / ELEM_SIZE, LAYOUT);
+      Mat, GsData, COMP_TYPE, OFFSET / ELEM_SIZE, STRIDE / ELEM_SIZE, LAYOUT);
 
     for (uint I = 0; I < M_DIM*N_DIM; ++I) {
       Output.Store<ELEM_TYPE>(I*ELEM_SIZE, GsData[I]);
diff --git a/utils/hct/gen_intrin_main.txt b/utils/hct/gen_intrin_main.txt
index e8a5cb431d..49d2038393 100644
--- a/utils/hct/gen_intrin_main.txt
+++ b/utils/hct/gen_intrin_main.txt
@@ -407,7 +407,7 @@ void [[min_sm=6.10]] __builtin_LinAlg_MatrixAccumulate(ref LinAlgMatrix matrixC,
 void [[min_sm=6.10]] __builtin_LinAlg_MatrixVectorMultiply(out LinAlg<c> ret, in LinAlgMatrix mat, in bool isOutputSigned, in LinAlg<c2> input, in uint inputInterp);
 void [[min_sm=6.10]] __builtin_LinAlg_MatrixVectorMultiplyAdd(out LinAlg<c> ret, in LinAlgMatrix mat, in bool isOutputSigned, in LinAlg<c2> input, in uint inputInterp, in LinAlg<c> bias, in uint biasInterp);
 void [[min_sm=6.10]] __builtin_LinAlg_MatrixAccumulateToDescriptor(in LinAlgMatrix matrix, in RWByteAddressBuffer buf, in uint offset, in uint stride, in uint layout, in uint align);
-void [[min_sm=6.10]] __builtin_LinAlg_MatrixAccumulateToMemory(in LinAlgMatrix matrix, groupshared LinAlg[] memory, in uint offset, in uint stride, in uint layout);
+void [[min_sm=6.10]] __builtin_LinAlg_MatrixAccumulateToMemory(in LinAlgMatrix matrix, groupshared LinAlg[] memory, in uint targetType, in uint offset, in uint stride, in uint layout);
 void [[min_sm=6.10]] __builtin_LinAlg_MatrixOuterProduct(out LinAlgMatrix ret, in LinAlg<c> vecA, in LinAlg<c2> vecB);
 void [[min_sm=6.10]] __builtin_LinAlg_Convert(out LinAlg<c> ret, in LinAlg<c2> vec, in uint input_interp, in uint output_interp);
 void [[min_sm=6.10]] __builtin_LinAlg_VectorAccumulateToDescriptor(in RWByteAddressBuffer buf, in uint offset, in uint align, in LinAlg<> vec);
diff --git a/utils/hct/hctdb.py b/utils/hct/hctdb.py
index 224c9efeba..4a1fa14571 100644
--- a/utils/hct/hctdb.py
+++ b/utils/hct/hctdb.py
@@ -6549,14 +6549,15 @@ def populate_ExperimentalOps(self):
                 db_dxil_param(
                     3, "$x_gs1", "memory", "groupshared array to accumulate into"
                 ),
-                db_dxil_param(4, "i32", "offset", "starting offset in the array"),
+                db_dxil_param(4, "i32", "targetType", "data type of the array"),
+                db_dxil_param(5, "i32", "offset", "starting offset in the array"),
                 db_dxil_param(
-                    5,
+                    6,
                     "i32",
                     "stride",
                     "number of bytes between the start of each row or column",
                 ),
-                db_dxil_param(6, "i32", "layout", "memory layout of matrix elements"),
+                db_dxil_param(7, "i32", "layout", "memory layout of matrix elements"),
             ],
         )