diff --git a/llvm/lib/CodeGen/SelectionDAG/DAGCombiner.cpp b/llvm/lib/CodeGen/SelectionDAG/DAGCombiner.cpp
--- a/llvm/lib/CodeGen/SelectionDAG/DAGCombiner.cpp
+++ b/llvm/lib/CodeGen/SelectionDAG/DAGCombiner.cpp
@@ -434,6 +434,7 @@
     SDValue visitOR(SDNode *N);
     SDValue visitORLike(SDValue N0, SDValue N1, SDNode *N);
     SDValue visitXOR(SDNode *N);
+    SDValue SimplifyVCastOp(SDNode *N, const SDLoc &DL);
     SDValue SimplifyVBinOp(SDNode *N, const SDLoc &DL);
     SDValue visitSHL(SDNode *N);
     SDValue visitSRA(SDNode *N);
@@ -11777,6 +11778,10 @@
   EVT VT = N->getValueType(0);
   SDLoc DL(N);
 
+  if (VT.isVector())
+    if (SDValue FoldedVOp = SimplifyVCastOp(N, DL))
+      return FoldedVOp;
+
   if (SDValue Res = tryToFoldExtendOfConstant(N, TLI, DAG, LegalTypes))
     return Res;
 
@@ -12022,6 +12027,10 @@
   SDValue N0 = N->getOperand(0);
   EVT VT = N->getValueType(0);
 
+  if (VT.isVector())
+    if (SDValue FoldedVOp = SimplifyVCastOp(N, SDLoc(N)))
+      return FoldedVOp;
+
   if (SDValue Res = tryToFoldExtendOfConstant(N, TLI, DAG, LegalTypes))
     return Res;
 
@@ -15576,6 +15585,10 @@
   SDValue N0 = N->getOperand(0);
   EVT VT = N->getValueType(0);
 
+  if (VT.isVector())
+    if (SDValue FoldedVOp = SimplifyVCastOp(N, SDLoc(N)))
+      return FoldedVOp;
+
   // If this is fp_round(fpextend), don't fold it, allow ourselves to be folded.
   if (N->hasOneUse() &&
       N->use_begin()->getOpcode() == ISD::FP_ROUND)
@@ -23059,6 +23072,30 @@
   return DAG.getBuildVector(VT, DL, Ops);
 }
 
+/// Visit a vector cast operation, like FP_EXTEND.
+SDValue DAGCombiner::SimplifyVCastOp(SDNode *N, const SDLoc &DL) {
+  EVT VT = N->getValueType(0);
+  assert(VT.isVector() && "SimplifyVCastOp only works on vectors!");
+  EVT EltVT = VT.getVectorElementType();
+
+  SDValue N0 = N->getOperand(0);
+  EVT SrcVT = N0->getValueType(0);
+  unsigned Opcode = N->getOpcode();
+  EVT SrcEltVT = SrcVT.getVectorElementType();
+
+  // TODO promote operation might be also good here?
+  if (N0.getOpcode() == ISD::SPLAT_VECTOR &&
+      TLI.isOperationLegalOrCustom(Opcode, EltVT)) {
+    SDValue IndexC = DAG.getVectorIdxConstant(0, DL);
+    SDValue Elt =
+        DAG.getNode(ISD::EXTRACT_VECTOR_ELT, DL, SrcEltVT, N0, IndexC);
+    SDValue ScalarBO = DAG.getNode(Opcode, DL, EltVT, Elt, N->getFlags());
+    return DAG.getSplatVector(VT, DL, ScalarBO);
+  }
+
+  return SDValue();
+}
+
 /// Visit a binary vector operation, like ADD.
 SDValue DAGCombiner::SimplifyVBinOp(SDNode *N, const SDLoc &DL) {
   EVT VT = N->getValueType(0);
diff --git a/llvm/lib/Target/RISCV/RISCVInstrInfoVSDPatterns.td b/llvm/lib/Target/RISCV/RISCVInstrInfoVSDPatterns.td
--- a/llvm/lib/Target/RISCV/RISCVInstrInfoVSDPatterns.td
+++ b/llvm/lib/Target/RISCV/RISCVInstrInfoVSDPatterns.td
@@ -410,11 +410,6 @@
               (!cast<Instruction>(instruction_name#"_VV_"#vti.Vti.LMul.MX)
                  vti.Vti.RegClass:$rs2, vti.Vti.RegClass:$rs1,
                  vti.Vti.AVL, vti.Vti.Log2SEW)>;
-    def : Pat<(op (vti.Wti.Vector (fpext_oneuse (vti.Vti.Vector vti.Vti.RegClass:$rs2))),
-                  (vti.Wti.Vector (fpext_oneuse (vti.Vti.Vector (SplatFPOp vti.Vti.ScalarRegClass:$rs1))))),
-              (!cast<Instruction>(instruction_name#"_V"#vti.Vti.ScalarSuffix#"_"#vti.Vti.LMul.MX)
-                 vti.Vti.RegClass:$rs2, vti.Vti.ScalarRegClass:$rs1,
-                 vti.Vti.AVL, vti.Vti.Log2SEW)>;
     def : Pat<(op (vti.Wti.Vector (fpext_oneuse (vti.Vti.Vector vti.Vti.RegClass:$rs2))),
                   (vti.Wti.Vector (SplatFPOp (fpext_oneuse vti.Vti.ScalarRegClass:$rs1)))),
               (!cast<Instruction>(instruction_name#"_V"#vti.Vti.ScalarSuffix#"_"#vti.Vti.LMul.MX)
@@ -430,11 +425,6 @@
               (!cast<Instruction>(instruction_name#"_WV_"#vti.Vti.LMul.MX)
                  vti.Wti.RegClass:$rs2, vti.Vti.RegClass:$rs1,
                  vti.Vti.AVL, vti.Vti.Log2SEW)>;
-    def : Pat<(op (vti.Wti.Vector vti.Wti.RegClass:$rs2),
-                  (vti.Wti.Vector (fpext_oneuse (vti.Vti.Vector (SplatFPOp vti.Vti.ScalarRegClass:$rs1))))),
-              (!cast<Instruction>(instruction_name#"_W"#vti.Vti.ScalarSuffix#"_"#vti.Vti.LMul.MX)
-                 vti.Wti.RegClass:$rs2, vti.Vti.ScalarRegClass:$rs1,
-                 vti.Vti.AVL, vti.Vti.Log2SEW)>;
       def : Pat<(op (vti.Wti.Vector vti.Wti.RegClass:$rs2),
                   (vti.Wti.Vector (SplatFPOp (fpext_oneuse vti.Vti.ScalarRegClass:$rs1)))),
               (!cast<Instruction>(instruction_name#"_W"#vti.Vti.ScalarSuffix#"_"#vti.Vti.LMul.MX)
@@ -456,7 +446,7 @@
               (!cast<Instruction>(instruction_name#"_VV_"#vti.Vti.LMul.MX)
                  vti.Wti.RegClass:$rd, vti.Vti.RegClass:$rs1, vti.Vti.RegClass:$rs2,
                  vti.Vti.AVL, vti.Vti.Log2SEW, TAIL_AGNOSTIC)>;
-    def : Pat<(fma (vti.Wti.Vector (fpext_oneuse (vti.Vti.Vector (SplatFPOp vti.Vti.ScalarRegClass:$rs1)))),
+    def : Pat<(fma (vti.Wti.Vector (SplatFPOp (fpext_oneuse vti.Vti.ScalarRegClass:$rs1))),
                    (vti.Wti.Vector (fpext_oneuse (vti.Vti.Vector vti.Vti.RegClass:$rs2))),
                    (vti.Wti.Vector vti.Wti.RegClass:$rd)),
               (!cast<Instruction>(instruction_name#"_V"#vti.Vti.ScalarSuffix#"_"#vti.Vti.LMul.MX)
@@ -475,13 +465,13 @@
               (!cast<Instruction>(instruction_name#"_VV_"#vti.LMul.MX)
                  wti.RegClass:$rd, vti.RegClass:$rs1, vti.RegClass:$rs2,
                  vti.AVL, vti.Log2SEW, TAIL_AGNOSTIC)>;
-    def : Pat<(fma (fpext_oneuse (vti.Vector (SplatFPOp vti.ScalarRegClass:$rs1))),
+    def : Pat<(fma (wti.Vector (SplatFPOp (fpext_oneuse vti.ScalarRegClass:$rs1))),
                    (fneg (wti.Vector (fpext_oneuse (vti.Vector vti.RegClass:$rs2)))),
                    (fneg wti.RegClass:$rd)),
               (!cast<Instruction>(instruction_name#"_V"#vti.ScalarSuffix#"_"#vti.LMul.MX)
                  wti.RegClass:$rd, vti.ScalarRegClass:$rs1, vti.RegClass:$rs2,
                  vti.AVL, vti.Log2SEW, TAIL_AGNOSTIC)>;
-    def : Pat<(fma (fneg (wti.Vector (fpext_oneuse (vti.Vector (SplatFPOp vti.ScalarRegClass:$rs1))))),
+    def : Pat<(fma (fneg (wti.Vector (SplatFPOp (fpext_oneuse vti.ScalarRegClass:$rs1)))),
                    (fpext_oneuse (vti.Vector vti.RegClass:$rs2)),
                    (fneg wti.RegClass:$rd)),
               (!cast<Instruction>(instruction_name#"_V"#vti.ScalarSuffix#"_"#vti.LMul.MX)
@@ -500,7 +490,7 @@
               (!cast<Instruction>(instruction_name#"_VV_"#vti.LMul.MX)
                  wti.RegClass:$rd, vti.RegClass:$rs1, vti.RegClass:$rs2,
                  vti.AVL, vti.Log2SEW, TAIL_AGNOSTIC)>;
-    def : Pat<(fma (wti.Vector (fpext_oneuse (vti.Vector (SplatFPOp vti.ScalarRegClass:$rs1)))),
+    def : Pat<(fma (wti.Vector (SplatFPOp (fpext_oneuse vti.ScalarRegClass:$rs1))),
                    (fpext_oneuse (vti.Vector vti.RegClass:$rs2)),
                    (fneg wti.RegClass:$rd)),
               (!cast<Instruction>(instruction_name#"_V"#vti.ScalarSuffix#"_"#vti.LMul.MX)
@@ -519,13 +509,13 @@
               (!cast<Instruction>(instruction_name#"_VV_"#vti.LMul.MX)
                  wti.RegClass:$rd, vti.RegClass:$rs1, vti.RegClass:$rs2,
                  vti.AVL, vti.Log2SEW, TAIL_AGNOSTIC)>;
-    def : Pat<(fma (wti.Vector (fpext_oneuse (vti.Vector (SplatFPOp vti.ScalarRegClass:$rs1)))),
+    def : Pat<(fma (wti.Vector (SplatFPOp (fpext_oneuse vti.ScalarRegClass:$rs1))),
                    (fneg (wti.Vector (fpext_oneuse (vti.Vector vti.RegClass:$rs2)))),
                    wti.RegClass:$rd),
               (!cast<Instruction>(instruction_name#"_V"#vti.ScalarSuffix#"_"#vti.LMul.MX)
                  wti.RegClass:$rd, vti.ScalarRegClass:$rs1, vti.RegClass:$rs2,
                  vti.AVL, vti.Log2SEW, TAIL_AGNOSTIC)>;
-    def : Pat<(fma (fneg (wti.Vector (fpext_oneuse (vti.Vector (SplatFPOp vti.ScalarRegClass:$rs1))))),
+    def : Pat<(fma (fneg (wti.Vector (SplatFPOp (fpext_oneuse vti.ScalarRegClass:$rs1)))),
                    (fpext_oneuse (vti.Vector vti.RegClass:$rs2)),
                    wti.RegClass:$rd),
               (!cast<Instruction>(instruction_name#"_V"#vti.ScalarSuffix#"_"#vti.LMul.MX)
diff --git a/llvm/test/CodeGen/RISCV/rvv/vfcopysign-sdnode.ll b/llvm/test/CodeGen/RISCV/rvv/vfcopysign-sdnode.ll
--- a/llvm/test/CodeGen/RISCV/rvv/vfcopysign-sdnode.ll
+++ b/llvm/test/CodeGen/RISCV/rvv/vfcopysign-sdnode.ll
@@ -604,11 +604,9 @@
 define <vscale x 1 x float> @vfcopysign_exttrunc_vf_nxv1f32_nxv1f16(<vscale x 1 x float> %vm, half %s) {
 ; CHECK-LABEL: vfcopysign_exttrunc_vf_nxv1f32_nxv1f16:
 ; CHECK:       # %bb.0:
-; CHECK-NEXT:    vsetvli a0, zero, e16, mf4, ta, mu
-; CHECK-NEXT:    vfmv.v.f v9, fa0
-; CHECK-NEXT:    vfwcvt.f.f.v v10, v9
-; CHECK-NEXT:    vsetvli zero, zero, e32, mf2, ta, mu
-; CHECK-NEXT:    vfsgnj.vv v8, v8, v10
+; CHECK-NEXT:    fcvt.s.h ft0, fa0
+; CHECK-NEXT:    vsetvli a0, zero, e32, mf2, ta, mu
+; CHECK-NEXT:    vfsgnj.vf v8, v8, ft0
 ; CHECK-NEXT:    ret
   %head = insertelement <vscale x 1 x half> poison, half %s, i32 0
   %splat = shufflevector <vscale x 1 x half> %head, <vscale x 1 x half> poison, <vscale x 1 x i32> zeroinitializer
@@ -634,11 +632,9 @@
 define <vscale x 1 x float> @vfcopynsign_exttrunc_vf_nxv1f32_nxv1f16(<vscale x 1 x float> %vm, half %s) {
 ; CHECK-LABEL: vfcopynsign_exttrunc_vf_nxv1f32_nxv1f16:
 ; CHECK:       # %bb.0:
-; CHECK-NEXT:    vsetvli a0, zero, e16, mf4, ta, mu
-; CHECK-NEXT:    vfmv.v.f v9, fa0
-; CHECK-NEXT:    vfwcvt.f.f.v v10, v9
-; CHECK-NEXT:    vsetvli zero, zero, e32, mf2, ta, mu
-; CHECK-NEXT:    vfsgnjn.vv v8, v8, v10
+; CHECK-NEXT:    fcvt.s.h ft0, fa0
+; CHECK-NEXT:    vsetvli a0, zero, e32, mf2, ta, mu
+; CHECK-NEXT:    vfsgnjn.vf v8, v8, ft0
 ; CHECK-NEXT:    ret
   %head = insertelement <vscale x 1 x half> poison, half %s, i32 0
   %splat = shufflevector <vscale x 1 x half> %head, <vscale x 1 x half> poison, <vscale x 1 x i32> zeroinitializer
@@ -866,11 +862,9 @@
 define <vscale x 8 x float> @vfcopysign_exttrunc_vf_nxv8f32_nxv8f16(<vscale x 8 x float> %vm, half %s) {
 ; CHECK-LABEL: vfcopysign_exttrunc_vf_nxv8f32_nxv8f16:
 ; CHECK:       # %bb.0:
-; CHECK-NEXT:    vsetvli a0, zero, e16, m2, ta, mu
-; CHECK-NEXT:    vfmv.v.f v12, fa0
-; CHECK-NEXT:    vfwcvt.f.f.v v16, v12
-; CHECK-NEXT:    vsetvli zero, zero, e32, m4, ta, mu
-; CHECK-NEXT:    vfsgnj.vv v8, v8, v16
+; CHECK-NEXT:    fcvt.s.h ft0, fa0
+; CHECK-NEXT:    vsetvli a0, zero, e32, m4, ta, mu
+; CHECK-NEXT:    vfsgnj.vf v8, v8, ft0
 ; CHECK-NEXT:    ret
   %head = insertelement <vscale x 8 x half> poison, half %s, i32 0
   %splat = shufflevector <vscale x 8 x half> %head, <vscale x 8 x half> poison, <vscale x 8 x i32> zeroinitializer
@@ -896,11 +890,9 @@
 define <vscale x 8 x float> @vfcopynsign_exttrunc_vf_nxv8f32_nxv8f16(<vscale x 8 x float> %vm, half %s) {
 ; CHECK-LABEL: vfcopynsign_exttrunc_vf_nxv8f32_nxv8f16:
 ; CHECK:       # %bb.0:
-; CHECK-NEXT:    vsetvli a0, zero, e16, m2, ta, mu
-; CHECK-NEXT:    vfmv.v.f v12, fa0
-; CHECK-NEXT:    vfwcvt.f.f.v v16, v12
-; CHECK-NEXT:    vsetvli zero, zero, e32, m4, ta, mu
-; CHECK-NEXT:    vfsgnjn.vv v8, v8, v16
+; CHECK-NEXT:    fcvt.s.h ft0, fa0
+; CHECK-NEXT:    vsetvli a0, zero, e32, m4, ta, mu
+; CHECK-NEXT:    vfsgnjn.vf v8, v8, ft0
 ; CHECK-NEXT:    ret
   %head = insertelement <vscale x 8 x half> poison, half %s, i32 0
   %splat = shufflevector <vscale x 8 x half> %head, <vscale x 8 x half> poison, <vscale x 8 x i32> zeroinitializer
@@ -1082,13 +1074,9 @@
 define <vscale x 1 x double> @vfcopysign_exttrunc_vf_nxv1f64_nxv1f16(<vscale x 1 x double> %vm, half %s) {
 ; CHECK-LABEL: vfcopysign_exttrunc_vf_nxv1f64_nxv1f16:
 ; CHECK:       # %bb.0:
-; CHECK-NEXT:    vsetvli a0, zero, e16, mf4, ta, mu
-; CHECK-NEXT:    vfmv.v.f v9, fa0
-; CHECK-NEXT:    vfwcvt.f.f.v v10, v9
-; CHECK-NEXT:    vsetvli zero, zero, e32, mf2, ta, mu
-; CHECK-NEXT:    vfwcvt.f.f.v v9, v10
-; CHECK-NEXT:    vsetvli zero, zero, e64, m1, ta, mu
-; CHECK-NEXT:    vfsgnj.vv v8, v8, v9
+; CHECK-NEXT:    fcvt.d.h ft0, fa0
+; CHECK-NEXT:    vsetvli a0, zero, e64, m1, ta, mu
+; CHECK-NEXT:    vfsgnj.vf v8, v8, ft0
 ; CHECK-NEXT:    ret
   %head = insertelement <vscale x 1 x half> poison, half %s, i32 0
   %splat = shufflevector <vscale x 1 x half> %head, <vscale x 1 x half> poison, <vscale x 1 x i32> zeroinitializer
@@ -1116,13 +1104,9 @@
 define <vscale x 1 x double> @vfcopynsign_exttrunc_vf_nxv1f64_nxv1f16(<vscale x 1 x double> %vm, half %s) {
 ; CHECK-LABEL: vfcopynsign_exttrunc_vf_nxv1f64_nxv1f16:
 ; CHECK:       # %bb.0:
-; CHECK-NEXT:    vsetvli a0, zero, e16, mf4, ta, mu
-; CHECK-NEXT:    vfmv.v.f v9, fa0
-; CHECK-NEXT:    vfwcvt.f.f.v v10, v9
-; CHECK-NEXT:    vsetvli zero, zero, e32, mf2, ta, mu
-; CHECK-NEXT:    vfwcvt.f.f.v v9, v10
-; CHECK-NEXT:    vsetvli zero, zero, e64, m1, ta, mu
-; CHECK-NEXT:    vfsgnjn.vv v8, v8, v9
+; CHECK-NEXT:    fcvt.d.h ft0, fa0
+; CHECK-NEXT:    vsetvli a0, zero, e64, m1, ta, mu
+; CHECK-NEXT:    vfsgnjn.vf v8, v8, ft0
 ; CHECK-NEXT:    ret
   %head = insertelement <vscale x 1 x half> poison, half %s, i32 0
   %splat = shufflevector <vscale x 1 x half> %head, <vscale x 1 x half> poison, <vscale x 1 x i32> zeroinitializer
@@ -1148,11 +1132,9 @@
 define <vscale x 1 x double> @vfcopysign_exttrunc_vf_nxv1f64_nxv1f32(<vscale x 1 x double> %vm, float %s) {
 ; CHECK-LABEL: vfcopysign_exttrunc_vf_nxv1f64_nxv1f32:
 ; CHECK:       # %bb.0:
-; CHECK-NEXT:    vsetvli a0, zero, e32, mf2, ta, mu
-; CHECK-NEXT:    vfmv.v.f v9, fa0
-; CHECK-NEXT:    vfwcvt.f.f.v v10, v9
-; CHECK-NEXT:    vsetvli zero, zero, e64, m1, ta, mu
-; CHECK-NEXT:    vfsgnj.vv v8, v8, v10
+; CHECK-NEXT:    fcvt.d.s ft0, fa0
+; CHECK-NEXT:    vsetvli a0, zero, e64, m1, ta, mu
+; CHECK-NEXT:    vfsgnj.vf v8, v8, ft0
 ; CHECK-NEXT:    ret
   %head = insertelement <vscale x 1 x float> poison, float %s, i32 0
   %splat = shufflevector <vscale x 1 x float> %head, <vscale x 1 x float> poison, <vscale x 1 x i32> zeroinitializer
@@ -1178,11 +1160,9 @@
 define <vscale x 1 x double> @vfcopynsign_exttrunc_vf_nxv1f64_nxv1f32(<vscale x 1 x double> %vm, float %s) {
 ; CHECK-LABEL: vfcopynsign_exttrunc_vf_nxv1f64_nxv1f32:
 ; CHECK:       # %bb.0:
-; CHECK-NEXT:    vsetvli a0, zero, e32, mf2, ta, mu
-; CHECK-NEXT:    vfmv.v.f v9, fa0
-; CHECK-NEXT:    vfwcvt.f.f.v v10, v9
-; CHECK-NEXT:    vsetvli zero, zero, e64, m1, ta, mu
-; CHECK-NEXT:    vfsgnjn.vv v8, v8, v10
+; CHECK-NEXT:    fcvt.d.s ft0, fa0
+; CHECK-NEXT:    vsetvli a0, zero, e64, m1, ta, mu
+; CHECK-NEXT:    vfsgnjn.vf v8, v8, ft0
 ; CHECK-NEXT:    ret
   %head = insertelement <vscale x 1 x float> poison, float %s, i32 0
   %splat = shufflevector <vscale x 1 x float> %head, <vscale x 1 x float> poison, <vscale x 1 x i32> zeroinitializer
@@ -1354,13 +1334,9 @@
 define <vscale x 8 x double> @vfcopysign_exttrunc_vf_nxv8f64_nxv8f16(<vscale x 8 x double> %vm, half %s) {
 ; CHECK-LABEL: vfcopysign_exttrunc_vf_nxv8f64_nxv8f16:
 ; CHECK:       # %bb.0:
-; CHECK-NEXT:    vsetvli a0, zero, e16, m2, ta, mu
-; CHECK-NEXT:    vfmv.v.f v16, fa0
-; CHECK-NEXT:    vfwcvt.f.f.v v20, v16
-; CHECK-NEXT:    vsetvli zero, zero, e32, m4, ta, mu
-; CHECK-NEXT:    vfwcvt.f.f.v v24, v20
-; CHECK-NEXT:    vsetvli zero, zero, e64, m8, ta, mu
-; CHECK-NEXT:    vfsgnj.vv v8, v8, v24
+; CHECK-NEXT:    fcvt.d.h ft0, fa0
+; CHECK-NEXT:    vsetvli a0, zero, e64, m8, ta, mu
+; CHECK-NEXT:    vfsgnj.vf v8, v8, ft0
 ; CHECK-NEXT:    ret
   %head = insertelement <vscale x 8 x half> poison, half %s, i32 0
   %splat = shufflevector <vscale x 8 x half> %head, <vscale x 8 x half> poison, <vscale x 8 x i32> zeroinitializer
@@ -1388,13 +1364,9 @@
 define <vscale x 8 x double> @vfcopynsign_exttrunc_vf_nxv8f64_nxv8f16(<vscale x 8 x double> %vm, half %s) {
 ; CHECK-LABEL: vfcopynsign_exttrunc_vf_nxv8f64_nxv8f16:
 ; CHECK:       # %bb.0:
-; CHECK-NEXT:    vsetvli a0, zero, e16, m2, ta, mu
-; CHECK-NEXT:    vfmv.v.f v16, fa0
-; CHECK-NEXT:    vfwcvt.f.f.v v20, v16
-; CHECK-NEXT:    vsetvli zero, zero, e32, m4, ta, mu
-; CHECK-NEXT:    vfwcvt.f.f.v v24, v20
-; CHECK-NEXT:    vsetvli zero, zero, e64, m8, ta, mu
-; CHECK-NEXT:    vfsgnjn.vv v8, v8, v24
+; CHECK-NEXT:    fcvt.d.h ft0, fa0
+; CHECK-NEXT:    vsetvli a0, zero, e64, m8, ta, mu
+; CHECK-NEXT:    vfsgnjn.vf v8, v8, ft0
 ; CHECK-NEXT:    ret
   %head = insertelement <vscale x 8 x half> poison, half %s, i32 0
   %splat = shufflevector <vscale x 8 x half> %head, <vscale x 8 x half> poison, <vscale x 8 x i32> zeroinitializer
@@ -1420,11 +1392,9 @@
 define <vscale x 8 x double> @vfcopysign_exttrunc_vf_nxv8f64_nxv8f32(<vscale x 8 x double> %vm, float %s) {
 ; CHECK-LABEL: vfcopysign_exttrunc_vf_nxv8f64_nxv8f32:
 ; CHECK:       # %bb.0:
-; CHECK-NEXT:    vsetvli a0, zero, e32, m4, ta, mu
-; CHECK-NEXT:    vfmv.v.f v16, fa0
-; CHECK-NEXT:    vfwcvt.f.f.v v24, v16
-; CHECK-NEXT:    vsetvli zero, zero, e64, m8, ta, mu
-; CHECK-NEXT:    vfsgnj.vv v8, v8, v24
+; CHECK-NEXT:    fcvt.d.s ft0, fa0
+; CHECK-NEXT:    vsetvli a0, zero, e64, m8, ta, mu
+; CHECK-NEXT:    vfsgnj.vf v8, v8, ft0
 ; CHECK-NEXT:    ret
   %head = insertelement <vscale x 8 x float> poison, float %s, i32 0
   %splat = shufflevector <vscale x 8 x float> %head, <vscale x 8 x float> poison, <vscale x 8 x i32> zeroinitializer
@@ -1450,11 +1420,9 @@
 define <vscale x 8 x double> @vfcopynsign_exttrunc_vf_nxv8f64_nxv8f32(<vscale x 8 x double> %vm, float %s) {
 ; CHECK-LABEL: vfcopynsign_exttrunc_vf_nxv8f64_nxv8f32:
 ; CHECK:       # %bb.0:
-; CHECK-NEXT:    vsetvli a0, zero, e32, m4, ta, mu
-; CHECK-NEXT:    vfmv.v.f v16, fa0
-; CHECK-NEXT:    vfwcvt.f.f.v v24, v16
-; CHECK-NEXT:    vsetvli zero, zero, e64, m8, ta, mu
-; CHECK-NEXT:    vfsgnjn.vv v8, v8, v24
+; CHECK-NEXT:    fcvt.d.s ft0, fa0
+; CHECK-NEXT:    vsetvli a0, zero, e64, m8, ta, mu
+; CHECK-NEXT:    vfsgnjn.vf v8, v8, ft0
 ; CHECK-NEXT:    ret
   %head = insertelement <vscale x 8 x float> poison, float %s, i32 0
   %splat = shufflevector <vscale x 8 x float> %head, <vscale x 8 x float> poison, <vscale x 8 x i32> zeroinitializer
diff --git a/llvm/test/CodeGen/RISCV/rvv/vnsra-sdnode.ll b/llvm/test/CodeGen/RISCV/rvv/vnsra-sdnode.ll
--- a/llvm/test/CodeGen/RISCV/rvv/vnsra-sdnode.ll
+++ b/llvm/test/CodeGen/RISCV/rvv/vnsra-sdnode.ll
@@ -1,8 +1,8 @@
 ; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
 ; RUN: llc -mtriple=riscv32 -mattr=+v -target-abi=ilp32 \
-; RUN:     -verify-machineinstrs < %s | FileCheck %s
+; RUN:     -verify-machineinstrs < %s | FileCheck %s --check-prefixes=CHECK,RV32
 ; RUN: llc -mtriple=riscv64 -mattr=+v -target-abi=lp64 \
-; RUN:     -verify-machineinstrs < %s | FileCheck %s
+; RUN:     -verify-machineinstrs < %s | FileCheck %s --check-prefixes=CHECK,RV64
 
 define <vscale x 1 x i32> @vnsra_wv_nxv1i32_sext(<vscale x 1 x i64> %va, <vscale x 1 x i32> %vb) {
 ; CHECK-LABEL: vnsra_wv_nxv1i32_sext:
@@ -17,11 +17,20 @@
 }
 
 define <vscale x 1 x i32> @vnsra_wx_i32_nxv1i32_sext(<vscale x 1 x i64> %va, i32 %b) {
-; CHECK-LABEL: vnsra_wx_i32_nxv1i32_sext:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    vsetvli a1, zero, e32, mf2, ta, mu
-; CHECK-NEXT:    vnsra.wx v8, v8, a0
-; CHECK-NEXT:    ret
+; RV32-LABEL: vnsra_wx_i32_nxv1i32_sext:
+; RV32:       # %bb.0:
+; RV32-NEXT:    vsetvli a1, zero, e32, mf2, ta, mu
+; RV32-NEXT:    vnsra.wx v8, v8, a0
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: vnsra_wx_i32_nxv1i32_sext:
+; RV64:       # %bb.0:
+; RV64-NEXT:    sext.w a0, a0
+; RV64-NEXT:    vsetvli a1, zero, e64, m1, ta, mu
+; RV64-NEXT:    vsra.vx v8, v8, a0
+; RV64-NEXT:    vsetvli zero, zero, e32, mf2, ta, mu
+; RV64-NEXT:    vncvt.x.x.w v8, v8
+; RV64-NEXT:    ret
   %head = insertelement <vscale x 1 x i32> poison, i32 %b, i32 0
   %splat = shufflevector <vscale x 1 x i32> %head, <vscale x 1 x i32> poison, <vscale x 1 x i32> zeroinitializer
   %vb = sext <vscale x 1 x i32> %splat to <vscale x 1 x i64>
@@ -58,12 +67,21 @@
 }
 
 define <vscale x 2 x i32> @vnsra_wx_i32_nxv2i32_sext(<vscale x 2 x i64> %va, i32 %b) {
-; CHECK-LABEL: vnsra_wx_i32_nxv2i32_sext:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    vsetvli a1, zero, e32, m1, ta, mu
-; CHECK-NEXT:    vnsra.wx v10, v8, a0
-; CHECK-NEXT:    vmv.v.v v8, v10
-; CHECK-NEXT:    ret
+; RV32-LABEL: vnsra_wx_i32_nxv2i32_sext:
+; RV32:       # %bb.0:
+; RV32-NEXT:    vsetvli a1, zero, e32, m1, ta, mu
+; RV32-NEXT:    vnsra.wx v10, v8, a0
+; RV32-NEXT:    vmv.v.v v8, v10
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: vnsra_wx_i32_nxv2i32_sext:
+; RV64:       # %bb.0:
+; RV64-NEXT:    sext.w a0, a0
+; RV64-NEXT:    vsetvli a1, zero, e64, m2, ta, mu
+; RV64-NEXT:    vsra.vx v10, v8, a0
+; RV64-NEXT:    vsetvli zero, zero, e32, m1, ta, mu
+; RV64-NEXT:    vncvt.x.x.w v8, v10
+; RV64-NEXT:    ret
   %head = insertelement <vscale x 2 x i32> poison, i32 %b, i32 0
   %splat = shufflevector <vscale x 2 x i32> %head, <vscale x 2 x i32> poison, <vscale x 2 x i32> zeroinitializer
   %vb = sext <vscale x 2 x i32> %splat to <vscale x 2 x i64>
@@ -101,12 +119,21 @@
 }
 
 define <vscale x 4 x i32> @vnsra_wx_i32_nxv4i32_sext(<vscale x 4 x i64> %va, i32 %b) {
-; CHECK-LABEL: vnsra_wx_i32_nxv4i32_sext:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    vsetvli a1, zero, e32, m2, ta, mu
-; CHECK-NEXT:    vnsra.wx v12, v8, a0
-; CHECK-NEXT:    vmv.v.v v8, v12
-; CHECK-NEXT:    ret
+; RV32-LABEL: vnsra_wx_i32_nxv4i32_sext:
+; RV32:       # %bb.0:
+; RV32-NEXT:    vsetvli a1, zero, e32, m2, ta, mu
+; RV32-NEXT:    vnsra.wx v12, v8, a0
+; RV32-NEXT:    vmv.v.v v8, v12
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: vnsra_wx_i32_nxv4i32_sext:
+; RV64:       # %bb.0:
+; RV64-NEXT:    sext.w a0, a0
+; RV64-NEXT:    vsetvli a1, zero, e64, m4, ta, mu
+; RV64-NEXT:    vsra.vx v12, v8, a0
+; RV64-NEXT:    vsetvli zero, zero, e32, m2, ta, mu
+; RV64-NEXT:    vncvt.x.x.w v8, v12
+; RV64-NEXT:    ret
   %head = insertelement <vscale x 4 x i32> poison, i32 %b, i32 0
   %splat = shufflevector <vscale x 4 x i32> %head, <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer
   %vb = sext <vscale x 4 x i32> %splat to <vscale x 4 x i64>
@@ -144,12 +171,21 @@
 }
 
 define <vscale x 8 x i32> @vnsra_wx_i32_nxv8i32_sext(<vscale x 8 x i64> %va, i32 %b) {
-; CHECK-LABEL: vnsra_wx_i32_nxv8i32_sext:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    vsetvli a1, zero, e32, m4, ta, mu
-; CHECK-NEXT:    vnsra.wx v16, v8, a0
-; CHECK-NEXT:    vmv.v.v v8, v16
-; CHECK-NEXT:    ret
+; RV32-LABEL: vnsra_wx_i32_nxv8i32_sext:
+; RV32:       # %bb.0:
+; RV32-NEXT:    vsetvli a1, zero, e32, m4, ta, mu
+; RV32-NEXT:    vnsra.wx v16, v8, a0
+; RV32-NEXT:    vmv.v.v v8, v16
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: vnsra_wx_i32_nxv8i32_sext:
+; RV64:       # %bb.0:
+; RV64-NEXT:    sext.w a0, a0
+; RV64-NEXT:    vsetvli a1, zero, e64, m8, ta, mu
+; RV64-NEXT:    vsra.vx v16, v8, a0
+; RV64-NEXT:    vsetvli zero, zero, e32, m4, ta, mu
+; RV64-NEXT:    vncvt.x.x.w v8, v16
+; RV64-NEXT:    ret
   %head = insertelement <vscale x 8 x i32> poison, i32 %b, i32 0
   %splat = shufflevector <vscale x 8 x i32> %head, <vscale x 8 x i32> poison, <vscale x 8 x i32> zeroinitializer
   %vb = sext <vscale x 8 x i32> %splat to <vscale x 8 x i64>
@@ -186,11 +222,21 @@
 }
 
 define <vscale x 1 x i32> @vnsra_wx_i32_nxv1i32_zext(<vscale x 1 x i64> %va, i32 %b) {
-; CHECK-LABEL: vnsra_wx_i32_nxv1i32_zext:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    vsetvli a1, zero, e32, mf2, ta, mu
-; CHECK-NEXT:    vnsra.wx v8, v8, a0
-; CHECK-NEXT:    ret
+; RV32-LABEL: vnsra_wx_i32_nxv1i32_zext:
+; RV32:       # %bb.0:
+; RV32-NEXT:    vsetvli a1, zero, e32, mf2, ta, mu
+; RV32-NEXT:    vnsra.wx v8, v8, a0
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: vnsra_wx_i32_nxv1i32_zext:
+; RV64:       # %bb.0:
+; RV64-NEXT:    slli a0, a0, 32
+; RV64-NEXT:    srli a0, a0, 32
+; RV64-NEXT:    vsetvli a1, zero, e64, m1, ta, mu
+; RV64-NEXT:    vsra.vx v8, v8, a0
+; RV64-NEXT:    vsetvli zero, zero, e32, mf2, ta, mu
+; RV64-NEXT:    vncvt.x.x.w v8, v8
+; RV64-NEXT:    ret
   %head = insertelement <vscale x 1 x i32> poison, i32 %b, i32 0
   %splat = shufflevector <vscale x 1 x i32> %head, <vscale x 1 x i32> poison, <vscale x 1 x i32> zeroinitializer
   %vb = zext <vscale x 1 x i32> %splat to <vscale x 1 x i64>
@@ -227,12 +273,22 @@
 }
 
 define <vscale x 2 x i32> @vnsra_wx_i32_nxv2i32_zext(<vscale x 2 x i64> %va, i32 %b) {
-; CHECK-LABEL: vnsra_wx_i32_nxv2i32_zext:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    vsetvli a1, zero, e32, m1, ta, mu
-; CHECK-NEXT:    vnsra.wx v10, v8, a0
-; CHECK-NEXT:    vmv.v.v v8, v10
-; CHECK-NEXT:    ret
+; RV32-LABEL: vnsra_wx_i32_nxv2i32_zext:
+; RV32:       # %bb.0:
+; RV32-NEXT:    vsetvli a1, zero, e32, m1, ta, mu
+; RV32-NEXT:    vnsra.wx v10, v8, a0
+; RV32-NEXT:    vmv.v.v v8, v10
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: vnsra_wx_i32_nxv2i32_zext:
+; RV64:       # %bb.0:
+; RV64-NEXT:    slli a0, a0, 32
+; RV64-NEXT:    srli a0, a0, 32
+; RV64-NEXT:    vsetvli a1, zero, e64, m2, ta, mu
+; RV64-NEXT:    vsra.vx v10, v8, a0
+; RV64-NEXT:    vsetvli zero, zero, e32, m1, ta, mu
+; RV64-NEXT:    vncvt.x.x.w v8, v10
+; RV64-NEXT:    ret
   %head = insertelement <vscale x 2 x i32> poison, i32 %b, i32 0
   %splat = shufflevector <vscale x 2 x i32> %head, <vscale x 2 x i32> poison, <vscale x 2 x i32> zeroinitializer
   %vb = zext <vscale x 2 x i32> %splat to <vscale x 2 x i64>
@@ -270,12 +326,22 @@
 }
 
 define <vscale x 4 x i32> @vnsra_wx_i32_nxv4i32_zext(<vscale x 4 x i64> %va, i32 %b) {
-; CHECK-LABEL: vnsra_wx_i32_nxv4i32_zext:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    vsetvli a1, zero, e32, m2, ta, mu
-; CHECK-NEXT:    vnsra.wx v12, v8, a0
-; CHECK-NEXT:    vmv.v.v v8, v12
-; CHECK-NEXT:    ret
+; RV32-LABEL: vnsra_wx_i32_nxv4i32_zext:
+; RV32:       # %bb.0:
+; RV32-NEXT:    vsetvli a1, zero, e32, m2, ta, mu
+; RV32-NEXT:    vnsra.wx v12, v8, a0
+; RV32-NEXT:    vmv.v.v v8, v12
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: vnsra_wx_i32_nxv4i32_zext:
+; RV64:       # %bb.0:
+; RV64-NEXT:    slli a0, a0, 32
+; RV64-NEXT:    srli a0, a0, 32
+; RV64-NEXT:    vsetvli a1, zero, e64, m4, ta, mu
+; RV64-NEXT:    vsra.vx v12, v8, a0
+; RV64-NEXT:    vsetvli zero, zero, e32, m2, ta, mu
+; RV64-NEXT:    vncvt.x.x.w v8, v12
+; RV64-NEXT:    ret
   %head = insertelement <vscale x 4 x i32> poison, i32 %b, i32 0
   %splat = shufflevector <vscale x 4 x i32> %head, <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer
   %vb = zext <vscale x 4 x i32> %splat to <vscale x 4 x i64>
@@ -313,12 +379,22 @@
 }
 
 define <vscale x 8 x i32> @vnsra_wx_i32_nxv8i32_zext(<vscale x 8 x i64> %va, i32 %b) {
-; CHECK-LABEL: vnsra_wx_i32_nxv8i32_zext:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    vsetvli a1, zero, e32, m4, ta, mu
-; CHECK-NEXT:    vnsra.wx v16, v8, a0
-; CHECK-NEXT:    vmv.v.v v8, v16
-; CHECK-NEXT:    ret
+; RV32-LABEL: vnsra_wx_i32_nxv8i32_zext:
+; RV32:       # %bb.0:
+; RV32-NEXT:    vsetvli a1, zero, e32, m4, ta, mu
+; RV32-NEXT:    vnsra.wx v16, v8, a0
+; RV32-NEXT:    vmv.v.v v8, v16
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: vnsra_wx_i32_nxv8i32_zext:
+; RV64:       # %bb.0:
+; RV64-NEXT:    slli a0, a0, 32
+; RV64-NEXT:    srli a0, a0, 32
+; RV64-NEXT:    vsetvli a1, zero, e64, m8, ta, mu
+; RV64-NEXT:    vsra.vx v16, v8, a0
+; RV64-NEXT:    vsetvli zero, zero, e32, m4, ta, mu
+; RV64-NEXT:    vncvt.x.x.w v8, v16
+; RV64-NEXT:    ret
   %head = insertelement <vscale x 8 x i32> poison, i32 %b, i32 0
   %splat = shufflevector <vscale x 8 x i32> %head, <vscale x 8 x i32> poison, <vscale x 8 x i32> zeroinitializer
   %vb = zext <vscale x 8 x i32> %splat to <vscale x 8 x i64>
diff --git a/llvm/test/CodeGen/RISCV/rvv/vnsrl-sdnode.ll b/llvm/test/CodeGen/RISCV/rvv/vnsrl-sdnode.ll
--- a/llvm/test/CodeGen/RISCV/rvv/vnsrl-sdnode.ll
+++ b/llvm/test/CodeGen/RISCV/rvv/vnsrl-sdnode.ll
@@ -1,8 +1,8 @@
 ; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
 ; RUN: llc -mtriple=riscv32 -mattr=+v -target-abi=ilp32 \
-; RUN:     -verify-machineinstrs < %s | FileCheck %s
+; RUN:     -verify-machineinstrs < %s | FileCheck %s --check-prefixes=CHECK,RV32
 ; RUN: llc -mtriple=riscv64 -mattr=+v -target-abi=lp64 \
-; RUN:     -verify-machineinstrs < %s | FileCheck %s
+; RUN:     -verify-machineinstrs < %s | FileCheck %s --check-prefixes=CHECK,RV64
 
 define <vscale x 1 x i32> @vnsrl_wv_nxv1i32_sext(<vscale x 1 x i64> %va, <vscale x 1 x i32> %vb) {
 ; CHECK-LABEL: vnsrl_wv_nxv1i32_sext:
@@ -17,11 +17,20 @@
 }
 
 define <vscale x 1 x i32> @vnsrl_wx_i32_nxv1i32_sext(<vscale x 1 x i64> %va, i32 %b) {
-; CHECK-LABEL: vnsrl_wx_i32_nxv1i32_sext:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    vsetvli a1, zero, e32, mf2, ta, mu
-; CHECK-NEXT:    vnsrl.wx v8, v8, a0
-; CHECK-NEXT:    ret
+; RV32-LABEL: vnsrl_wx_i32_nxv1i32_sext:
+; RV32:       # %bb.0:
+; RV32-NEXT:    vsetvli a1, zero, e32, mf2, ta, mu
+; RV32-NEXT:    vnsrl.wx v8, v8, a0
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: vnsrl_wx_i32_nxv1i32_sext:
+; RV64:       # %bb.0:
+; RV64-NEXT:    sext.w a0, a0
+; RV64-NEXT:    vsetvli a1, zero, e64, m1, ta, mu
+; RV64-NEXT:    vsrl.vx v8, v8, a0
+; RV64-NEXT:    vsetvli zero, zero, e32, mf2, ta, mu
+; RV64-NEXT:    vncvt.x.x.w v8, v8
+; RV64-NEXT:    ret
   %head = insertelement <vscale x 1 x i32> poison, i32 %b, i32 0
   %splat = shufflevector <vscale x 1 x i32> %head, <vscale x 1 x i32> poison, <vscale x 1 x i32> zeroinitializer
   %vb = sext <vscale x 1 x i32> %splat to <vscale x 1 x i64>
@@ -58,12 +67,21 @@
 }
 
 define <vscale x 2 x i32> @vnsrl_wx_i32_nxv2i32_sext(<vscale x 2 x i64> %va, i32 %b) {
-; CHECK-LABEL: vnsrl_wx_i32_nxv2i32_sext:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    vsetvli a1, zero, e32, m1, ta, mu
-; CHECK-NEXT:    vnsrl.wx v10, v8, a0
-; CHECK-NEXT:    vmv.v.v v8, v10
-; CHECK-NEXT:    ret
+; RV32-LABEL: vnsrl_wx_i32_nxv2i32_sext:
+; RV32:       # %bb.0:
+; RV32-NEXT:    vsetvli a1, zero, e32, m1, ta, mu
+; RV32-NEXT:    vnsrl.wx v10, v8, a0
+; RV32-NEXT:    vmv.v.v v8, v10
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: vnsrl_wx_i32_nxv2i32_sext:
+; RV64:       # %bb.0:
+; RV64-NEXT:    sext.w a0, a0
+; RV64-NEXT:    vsetvli a1, zero, e64, m2, ta, mu
+; RV64-NEXT:    vsrl.vx v10, v8, a0
+; RV64-NEXT:    vsetvli zero, zero, e32, m1, ta, mu
+; RV64-NEXT:    vncvt.x.x.w v8, v10
+; RV64-NEXT:    ret
   %head = insertelement <vscale x 2 x i32> poison, i32 %b, i32 0
   %splat = shufflevector <vscale x 2 x i32> %head, <vscale x 2 x i32> poison, <vscale x 2 x i32> zeroinitializer
   %vb = sext <vscale x 2 x i32> %splat to <vscale x 2 x i64>
@@ -101,12 +119,21 @@
 }
 
 define <vscale x 4 x i32> @vnsrl_wx_i32_nxv4i32_sext(<vscale x 4 x i64> %va, i32 %b) {
-; CHECK-LABEL: vnsrl_wx_i32_nxv4i32_sext:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    vsetvli a1, zero, e32, m2, ta, mu
-; CHECK-NEXT:    vnsrl.wx v12, v8, a0
-; CHECK-NEXT:    vmv.v.v v8, v12
-; CHECK-NEXT:    ret
+; RV32-LABEL: vnsrl_wx_i32_nxv4i32_sext:
+; RV32:       # %bb.0:
+; RV32-NEXT:    vsetvli a1, zero, e32, m2, ta, mu
+; RV32-NEXT:    vnsrl.wx v12, v8, a0
+; RV32-NEXT:    vmv.v.v v8, v12
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: vnsrl_wx_i32_nxv4i32_sext:
+; RV64:       # %bb.0:
+; RV64-NEXT:    sext.w a0, a0
+; RV64-NEXT:    vsetvli a1, zero, e64, m4, ta, mu
+; RV64-NEXT:    vsrl.vx v12, v8, a0
+; RV64-NEXT:    vsetvli zero, zero, e32, m2, ta, mu
+; RV64-NEXT:    vncvt.x.x.w v8, v12
+; RV64-NEXT:    ret
   %head = insertelement <vscale x 4 x i32> poison, i32 %b, i32 0
   %splat = shufflevector <vscale x 4 x i32> %head, <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer
   %vb = sext <vscale x 4 x i32> %splat to <vscale x 4 x i64>
@@ -144,12 +171,21 @@
 }
 
 define <vscale x 8 x i32> @vnsrl_wx_i32_nxv8i32_sext(<vscale x 8 x i64> %va, i32 %b) {
-; CHECK-LABEL: vnsrl_wx_i32_nxv8i32_sext:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    vsetvli a1, zero, e32, m4, ta, mu
-; CHECK-NEXT:    vnsrl.wx v16, v8, a0
-; CHECK-NEXT:    vmv.v.v v8, v16
-; CHECK-NEXT:    ret
+; RV32-LABEL: vnsrl_wx_i32_nxv8i32_sext:
+; RV32:       # %bb.0:
+; RV32-NEXT:    vsetvli a1, zero, e32, m4, ta, mu
+; RV32-NEXT:    vnsrl.wx v16, v8, a0
+; RV32-NEXT:    vmv.v.v v8, v16
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: vnsrl_wx_i32_nxv8i32_sext:
+; RV64:       # %bb.0:
+; RV64-NEXT:    sext.w a0, a0
+; RV64-NEXT:    vsetvli a1, zero, e64, m8, ta, mu
+; RV64-NEXT:    vsrl.vx v16, v8, a0
+; RV64-NEXT:    vsetvli zero, zero, e32, m4, ta, mu
+; RV64-NEXT:    vncvt.x.x.w v8, v16
+; RV64-NEXT:    ret
   %head = insertelement <vscale x 8 x i32> poison, i32 %b, i32 0
   %splat = shufflevector <vscale x 8 x i32> %head, <vscale x 8 x i32> poison, <vscale x 8 x i32> zeroinitializer
   %vb = sext <vscale x 8 x i32> %splat to <vscale x 8 x i64>
@@ -186,11 +222,21 @@
 }
 
 define <vscale x 1 x i32> @vnsrl_wx_i32_nxv1i32_zext(<vscale x 1 x i64> %va, i32 %b) {
-; CHECK-LABEL: vnsrl_wx_i32_nxv1i32_zext:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    vsetvli a1, zero, e32, mf2, ta, mu
-; CHECK-NEXT:    vnsrl.wx v8, v8, a0
-; CHECK-NEXT:    ret
+; RV32-LABEL: vnsrl_wx_i32_nxv1i32_zext:
+; RV32:       # %bb.0:
+; RV32-NEXT:    vsetvli a1, zero, e32, mf2, ta, mu
+; RV32-NEXT:    vnsrl.wx v8, v8, a0
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: vnsrl_wx_i32_nxv1i32_zext:
+; RV64:       # %bb.0:
+; RV64-NEXT:    slli a0, a0, 32
+; RV64-NEXT:    srli a0, a0, 32
+; RV64-NEXT:    vsetvli a1, zero, e64, m1, ta, mu
+; RV64-NEXT:    vsrl.vx v8, v8, a0
+; RV64-NEXT:    vsetvli zero, zero, e32, mf2, ta, mu
+; RV64-NEXT:    vncvt.x.x.w v8, v8
+; RV64-NEXT:    ret
   %head = insertelement <vscale x 1 x i32> poison, i32 %b, i32 0
   %splat = shufflevector <vscale x 1 x i32> %head, <vscale x 1 x i32> poison, <vscale x 1 x i32> zeroinitializer
   %vb = zext <vscale x 1 x i32> %splat to <vscale x 1 x i64>
@@ -227,12 +273,22 @@
 }
 
 define <vscale x 2 x i32> @vnsrl_wx_i32_nxv2i32_zext(<vscale x 2 x i64> %va, i32 %b) {
-; CHECK-LABEL: vnsrl_wx_i32_nxv2i32_zext:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    vsetvli a1, zero, e32, m1, ta, mu
-; CHECK-NEXT:    vnsrl.wx v10, v8, a0
-; CHECK-NEXT:    vmv.v.v v8, v10
-; CHECK-NEXT:    ret
+; RV32-LABEL: vnsrl_wx_i32_nxv2i32_zext:
+; RV32:       # %bb.0:
+; RV32-NEXT:    vsetvli a1, zero, e32, m1, ta, mu
+; RV32-NEXT:    vnsrl.wx v10, v8, a0
+; RV32-NEXT:    vmv.v.v v8, v10
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: vnsrl_wx_i32_nxv2i32_zext:
+; RV64:       # %bb.0:
+; RV64-NEXT:    slli a0, a0, 32
+; RV64-NEXT:    srli a0, a0, 32
+; RV64-NEXT:    vsetvli a1, zero, e64, m2, ta, mu
+; RV64-NEXT:    vsrl.vx v10, v8, a0
+; RV64-NEXT:    vsetvli zero, zero, e32, m1, ta, mu
+; RV64-NEXT:    vncvt.x.x.w v8, v10
+; RV64-NEXT:    ret
   %head = insertelement <vscale x 2 x i32> poison, i32 %b, i32 0
   %splat = shufflevector <vscale x 2 x i32> %head, <vscale x 2 x i32> poison, <vscale x 2 x i32> zeroinitializer
   %vb = zext <vscale x 2 x i32> %splat to <vscale x 2 x i64>
@@ -270,12 +326,22 @@
 }
 
 define <vscale x 4 x i32> @vnsrl_wx_i32_nxv4i32_zext(<vscale x 4 x i64> %va, i32 %b) {
-; CHECK-LABEL: vnsrl_wx_i32_nxv4i32_zext:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    vsetvli a1, zero, e32, m2, ta, mu
-; CHECK-NEXT:    vnsrl.wx v12, v8, a0
-; CHECK-NEXT:    vmv.v.v v8, v12
-; CHECK-NEXT:    ret
+; RV32-LABEL: vnsrl_wx_i32_nxv4i32_zext:
+; RV32:       # %bb.0:
+; RV32-NEXT:    vsetvli a1, zero, e32, m2, ta, mu
+; RV32-NEXT:    vnsrl.wx v12, v8, a0
+; RV32-NEXT:    vmv.v.v v8, v12
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: vnsrl_wx_i32_nxv4i32_zext:
+; RV64:       # %bb.0:
+; RV64-NEXT:    slli a0, a0, 32
+; RV64-NEXT:    srli a0, a0, 32
+; RV64-NEXT:    vsetvli a1, zero, e64, m4, ta, mu
+; RV64-NEXT:    vsrl.vx v12, v8, a0
+; RV64-NEXT:    vsetvli zero, zero, e32, m2, ta, mu
+; RV64-NEXT:    vncvt.x.x.w v8, v12
+; RV64-NEXT:    ret
   %head = insertelement <vscale x 4 x i32> poison, i32 %b, i32 0
   %splat = shufflevector <vscale x 4 x i32> %head, <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer
   %vb = zext <vscale x 4 x i32> %splat to <vscale x 4 x i64>
@@ -313,12 +379,22 @@
 }
 
 define <vscale x 8 x i32> @vnsrl_wx_i32_nxv8i32_zext(<vscale x 8 x i64> %va, i32 %b) {
-; CHECK-LABEL: vnsrl_wx_i32_nxv8i32_zext:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    vsetvli a1, zero, e32, m4, ta, mu
-; CHECK-NEXT:    vnsrl.wx v16, v8, a0
-; CHECK-NEXT:    vmv.v.v v8, v16
-; CHECK-NEXT:    ret
+; RV32-LABEL: vnsrl_wx_i32_nxv8i32_zext:
+; RV32:       # %bb.0:
+; RV32-NEXT:    vsetvli a1, zero, e32, m4, ta, mu
+; RV32-NEXT:    vnsrl.wx v16, v8, a0
+; RV32-NEXT:    vmv.v.v v8, v16
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: vnsrl_wx_i32_nxv8i32_zext:
+; RV64:       # %bb.0:
+; RV64-NEXT:    slli a0, a0, 32
+; RV64-NEXT:    srli a0, a0, 32
+; RV64-NEXT:    vsetvli a1, zero, e64, m8, ta, mu
+; RV64-NEXT:    vsrl.vx v16, v8, a0
+; RV64-NEXT:    vsetvli zero, zero, e32, m4, ta, mu
+; RV64-NEXT:    vncvt.x.x.w v8, v16
+; RV64-NEXT:    ret
   %head = insertelement <vscale x 8 x i32> poison, i32 %b, i32 0
   %splat = shufflevector <vscale x 8 x i32> %head, <vscale x 8 x i32> poison, <vscale x 8 x i32> zeroinitializer
   %vb = zext <vscale x 8 x i32> %splat to <vscale x 8 x i64>
diff --git a/llvm/test/CodeGen/RISCV/rvv/vwadd-sdnode.ll b/llvm/test/CodeGen/RISCV/rvv/vwadd-sdnode.ll
--- a/llvm/test/CodeGen/RISCV/rvv/vwadd-sdnode.ll
+++ b/llvm/test/CodeGen/RISCV/rvv/vwadd-sdnode.ll
@@ -1,6 +1,6 @@
 ; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
-; RUN: llc -mtriple=riscv32 -mattr=+v -verify-machineinstrs < %s | FileCheck %s
-; RUN: llc -mtriple=riscv64 -mattr=+v -verify-machineinstrs < %s | FileCheck %s
+; RUN: llc -mtriple=riscv32 -mattr=+v -verify-machineinstrs < %s | FileCheck %s --check-prefixes=CHECK,RV32
+; RUN: llc -mtriple=riscv64 -mattr=+v -verify-machineinstrs < %s | FileCheck %s --check-prefixes=CHECK,RV64
 
 define <vscale x 1 x i64> @vwadd_vv_nxv1i64(<vscale x 1 x i32> %va, <vscale x 1 x i32> %vb) {
 ; CHECK-LABEL: vwadd_vv_nxv1i64:
@@ -29,12 +29,20 @@
 }
 
 define <vscale x 1 x i64> @vwadd_vx_nxv1i64(<vscale x 1 x i32> %va, i32 %b) {
-; CHECK-LABEL: vwadd_vx_nxv1i64:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    vsetvli a1, zero, e32, mf2, ta, mu
-; CHECK-NEXT:    vwadd.vx v9, v8, a0
-; CHECK-NEXT:    vmv1r.v v8, v9
-; CHECK-NEXT:    ret
+; RV32-LABEL: vwadd_vx_nxv1i64:
+; RV32:       # %bb.0:
+; RV32-NEXT:    vsetvli a1, zero, e32, mf2, ta, mu
+; RV32-NEXT:    vwadd.vx v9, v8, a0
+; RV32-NEXT:    vmv1r.v v8, v9
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: vwadd_vx_nxv1i64:
+; RV64:       # %bb.0:
+; RV64-NEXT:    vsetvli a1, zero, e64, m1, ta, mu
+; RV64-NEXT:    vsext.vf2 v9, v8
+; RV64-NEXT:    sext.w a0, a0
+; RV64-NEXT:    vadd.vx v8, v9, a0
+; RV64-NEXT:    ret
   %head = insertelement <vscale x 1 x i32> poison, i32 %b, i32 0
   %splat = shufflevector <vscale x 1 x i32> %head, <vscale x 1 x i32> poison, <vscale x 1 x i32> zeroinitializer
   %vc = sext <vscale x 1 x i32> %va to <vscale x 1 x i64>
@@ -44,12 +52,21 @@
 }
 
 define <vscale x 1 x i64> @vwaddu_vx_nxv1i64(<vscale x 1 x i32> %va, i32 %b) {
-; CHECK-LABEL: vwaddu_vx_nxv1i64:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    vsetvli a1, zero, e32, mf2, ta, mu
-; CHECK-NEXT:    vwaddu.vx v9, v8, a0
-; CHECK-NEXT:    vmv1r.v v8, v9
-; CHECK-NEXT:    ret
+; RV32-LABEL: vwaddu_vx_nxv1i64:
+; RV32:       # %bb.0:
+; RV32-NEXT:    vsetvli a1, zero, e32, mf2, ta, mu
+; RV32-NEXT:    vwaddu.vx v9, v8, a0
+; RV32-NEXT:    vmv1r.v v8, v9
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: vwaddu_vx_nxv1i64:
+; RV64:       # %bb.0:
+; RV64-NEXT:    vsetvli a1, zero, e64, m1, ta, mu
+; RV64-NEXT:    vzext.vf2 v9, v8
+; RV64-NEXT:    slli a0, a0, 32
+; RV64-NEXT:    srli a0, a0, 32
+; RV64-NEXT:    vadd.vx v8, v9, a0
+; RV64-NEXT:    ret
   %head = insertelement <vscale x 1 x i32> poison, i32 %b, i32 0
   %splat = shufflevector <vscale x 1 x i32> %head, <vscale x 1 x i32> poison, <vscale x 1 x i32> zeroinitializer
   %vc = zext <vscale x 1 x i32> %va to <vscale x 1 x i64>
@@ -83,11 +100,18 @@
 }
 
 define <vscale x 1 x i64> @vwadd_wx_nxv1i64(<vscale x 1 x i64> %va, i32 %b) {
-; CHECK-LABEL: vwadd_wx_nxv1i64:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    vsetvli a1, zero, e32, mf2, ta, mu
-; CHECK-NEXT:    vwadd.wx v8, v8, a0
-; CHECK-NEXT:    ret
+; RV32-LABEL: vwadd_wx_nxv1i64:
+; RV32:       # %bb.0:
+; RV32-NEXT:    vsetvli a1, zero, e32, mf2, ta, mu
+; RV32-NEXT:    vwadd.wx v8, v8, a0
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: vwadd_wx_nxv1i64:
+; RV64:       # %bb.0:
+; RV64-NEXT:    sext.w a0, a0
+; RV64-NEXT:    vsetvli a1, zero, e64, m1, ta, mu
+; RV64-NEXT:    vadd.vx v8, v8, a0
+; RV64-NEXT:    ret
   %head = insertelement <vscale x 1 x i32> poison, i32 %b, i32 0
   %splat = shufflevector <vscale x 1 x i32> %head, <vscale x 1 x i32> poison, <vscale x 1 x i32> zeroinitializer
   %vb = sext <vscale x 1 x i32> %splat to <vscale x 1 x i64>
@@ -96,11 +120,19 @@
 }
 
 define <vscale x 1 x i64> @vwaddu_wx_nxv1i64(<vscale x 1 x i64> %va, i32 %b) {
-; CHECK-LABEL: vwaddu_wx_nxv1i64:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    vsetvli a1, zero, e32, mf2, ta, mu
-; CHECK-NEXT:    vwaddu.wx v8, v8, a0
-; CHECK-NEXT:    ret
+; RV32-LABEL: vwaddu_wx_nxv1i64:
+; RV32:       # %bb.0:
+; RV32-NEXT:    vsetvli a1, zero, e32, mf2, ta, mu
+; RV32-NEXT:    vwaddu.wx v8, v8, a0
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: vwaddu_wx_nxv1i64:
+; RV64:       # %bb.0:
+; RV64-NEXT:    slli a0, a0, 32
+; RV64-NEXT:    srli a0, a0, 32
+; RV64-NEXT:    vsetvli a1, zero, e64, m1, ta, mu
+; RV64-NEXT:    vadd.vx v8, v8, a0
+; RV64-NEXT:    ret
   %head = insertelement <vscale x 1 x i32> poison, i32 %b, i32 0
   %splat = shufflevector <vscale x 1 x i32> %head, <vscale x 1 x i32> poison, <vscale x 1 x i32> zeroinitializer
   %vb = zext <vscale x 1 x i32> %splat to <vscale x 1 x i64>
@@ -135,12 +167,20 @@
 }
 
 define <vscale x 2 x i64> @vwadd_vx_nxv2i64(<vscale x 2 x i32> %va, i32 %b) {
-; CHECK-LABEL: vwadd_vx_nxv2i64:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    vsetvli a1, zero, e32, m1, ta, mu
-; CHECK-NEXT:    vwadd.vx v10, v8, a0
-; CHECK-NEXT:    vmv2r.v v8, v10
-; CHECK-NEXT:    ret
+; RV32-LABEL: vwadd_vx_nxv2i64:
+; RV32:       # %bb.0:
+; RV32-NEXT:    vsetvli a1, zero, e32, m1, ta, mu
+; RV32-NEXT:    vwadd.vx v10, v8, a0
+; RV32-NEXT:    vmv2r.v v8, v10
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: vwadd_vx_nxv2i64:
+; RV64:       # %bb.0:
+; RV64-NEXT:    vsetvli a1, zero, e64, m2, ta, mu
+; RV64-NEXT:    vsext.vf2 v10, v8
+; RV64-NEXT:    sext.w a0, a0
+; RV64-NEXT:    vadd.vx v8, v10, a0
+; RV64-NEXT:    ret
   %head = insertelement <vscale x 2 x i32> poison, i32 %b, i32 0
   %splat = shufflevector <vscale x 2 x i32> %head, <vscale x 2 x i32> poison, <vscale x 2 x i32> zeroinitializer
   %vc = sext <vscale x 2 x i32> %va to <vscale x 2 x i64>
@@ -150,12 +190,21 @@
 }
 
 define <vscale x 2 x i64> @vwaddu_vx_nxv2i64(<vscale x 2 x i32> %va, i32 %b) {
-; CHECK-LABEL: vwaddu_vx_nxv2i64:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    vsetvli a1, zero, e32, m1, ta, mu
-; CHECK-NEXT:    vwaddu.vx v10, v8, a0
-; CHECK-NEXT:    vmv2r.v v8, v10
-; CHECK-NEXT:    ret
+; RV32-LABEL: vwaddu_vx_nxv2i64:
+; RV32:       # %bb.0:
+; RV32-NEXT:    vsetvli a1, zero, e32, m1, ta, mu
+; RV32-NEXT:    vwaddu.vx v10, v8, a0
+; RV32-NEXT:    vmv2r.v v8, v10
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: vwaddu_vx_nxv2i64:
+; RV64:       # %bb.0:
+; RV64-NEXT:    vsetvli a1, zero, e64, m2, ta, mu
+; RV64-NEXT:    vzext.vf2 v10, v8
+; RV64-NEXT:    slli a0, a0, 32
+; RV64-NEXT:    srli a0, a0, 32
+; RV64-NEXT:    vadd.vx v8, v10, a0
+; RV64-NEXT:    ret
   %head = insertelement <vscale x 2 x i32> poison, i32 %b, i32 0
   %splat = shufflevector <vscale x 2 x i32> %head, <vscale x 2 x i32> poison, <vscale x 2 x i32> zeroinitializer
   %vc = zext <vscale x 2 x i32> %va to <vscale x 2 x i64>
@@ -189,11 +238,18 @@
 }
 
 define <vscale x 2 x i64> @vwadd_wx_nxv2i64(<vscale x 2 x i64> %va, i32 %b) {
-; CHECK-LABEL: vwadd_wx_nxv2i64:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    vsetvli a1, zero, e32, m1, ta, mu
-; CHECK-NEXT:    vwadd.wx v8, v8, a0
-; CHECK-NEXT:    ret
+; RV32-LABEL: vwadd_wx_nxv2i64:
+; RV32:       # %bb.0:
+; RV32-NEXT:    vsetvli a1, zero, e32, m1, ta, mu
+; RV32-NEXT:    vwadd.wx v8, v8, a0
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: vwadd_wx_nxv2i64:
+; RV64:       # %bb.0:
+; RV64-NEXT:    sext.w a0, a0
+; RV64-NEXT:    vsetvli a1, zero, e64, m2, ta, mu
+; RV64-NEXT:    vadd.vx v8, v8, a0
+; RV64-NEXT:    ret
   %head = insertelement <vscale x 2 x i32> poison, i32 %b, i32 0
   %splat = shufflevector <vscale x 2 x i32> %head, <vscale x 2 x i32> poison, <vscale x 2 x i32> zeroinitializer
   %vb = sext <vscale x 2 x i32> %splat to <vscale x 2 x i64>
@@ -202,11 +258,19 @@
 }
 
 define <vscale x 2 x i64> @vwaddu_wx_nxv2i64(<vscale x 2 x i64> %va, i32 %b) {
-; CHECK-LABEL: vwaddu_wx_nxv2i64:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    vsetvli a1, zero, e32, m1, ta, mu
-; CHECK-NEXT:    vwaddu.wx v8, v8, a0
-; CHECK-NEXT:    ret
+; RV32-LABEL: vwaddu_wx_nxv2i64:
+; RV32:       # %bb.0:
+; RV32-NEXT:    vsetvli a1, zero, e32, m1, ta, mu
+; RV32-NEXT:    vwaddu.wx v8, v8, a0
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: vwaddu_wx_nxv2i64:
+; RV64:       # %bb.0:
+; RV64-NEXT:    slli a0, a0, 32
+; RV64-NEXT:    srli a0, a0, 32
+; RV64-NEXT:    vsetvli a1, zero, e64, m2, ta, mu
+; RV64-NEXT:    vadd.vx v8, v8, a0
+; RV64-NEXT:    ret
   %head = insertelement <vscale x 2 x i32> poison, i32 %b, i32 0
   %splat = shufflevector <vscale x 2 x i32> %head, <vscale x 2 x i32> poison, <vscale x 2 x i32> zeroinitializer
   %vb = zext <vscale x 2 x i32> %splat to <vscale x 2 x i64>
@@ -241,12 +305,20 @@
 }
 
 define <vscale x 4 x i64> @vwadd_vx_nxv4i64(<vscale x 4 x i32> %va, i32 %b) {
-; CHECK-LABEL: vwadd_vx_nxv4i64:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    vsetvli a1, zero, e32, m2, ta, mu
-; CHECK-NEXT:    vwadd.vx v12, v8, a0
-; CHECK-NEXT:    vmv4r.v v8, v12
-; CHECK-NEXT:    ret
+; RV32-LABEL: vwadd_vx_nxv4i64:
+; RV32:       # %bb.0:
+; RV32-NEXT:    vsetvli a1, zero, e32, m2, ta, mu
+; RV32-NEXT:    vwadd.vx v12, v8, a0
+; RV32-NEXT:    vmv4r.v v8, v12
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: vwadd_vx_nxv4i64:
+; RV64:       # %bb.0:
+; RV64-NEXT:    vsetvli a1, zero, e64, m4, ta, mu
+; RV64-NEXT:    vsext.vf2 v12, v8
+; RV64-NEXT:    sext.w a0, a0
+; RV64-NEXT:    vadd.vx v8, v12, a0
+; RV64-NEXT:    ret
   %head = insertelement <vscale x 4 x i32> poison, i32 %b, i32 0
   %splat = shufflevector <vscale x 4 x i32> %head, <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer
   %vc = sext <vscale x 4 x i32> %va to <vscale x 4 x i64>
@@ -256,12 +328,21 @@
 }
 
 define <vscale x 4 x i64> @vwaddu_vx_nxv4i64(<vscale x 4 x i32> %va, i32 %b) {
-; CHECK-LABEL: vwaddu_vx_nxv4i64:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    vsetvli a1, zero, e32, m2, ta, mu
-; CHECK-NEXT:    vwaddu.vx v12, v8, a0
-; CHECK-NEXT:    vmv4r.v v8, v12
-; CHECK-NEXT:    ret
+; RV32-LABEL: vwaddu_vx_nxv4i64:
+; RV32:       # %bb.0:
+; RV32-NEXT:    vsetvli a1, zero, e32, m2, ta, mu
+; RV32-NEXT:    vwaddu.vx v12, v8, a0
+; RV32-NEXT:    vmv4r.v v8, v12
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: vwaddu_vx_nxv4i64:
+; RV64:       # %bb.0:
+; RV64-NEXT:    vsetvli a1, zero, e64, m4, ta, mu
+; RV64-NEXT:    vzext.vf2 v12, v8
+; RV64-NEXT:    slli a0, a0, 32
+; RV64-NEXT:    srli a0, a0, 32
+; RV64-NEXT:    vadd.vx v8, v12, a0
+; RV64-NEXT:    ret
   %head = insertelement <vscale x 4 x i32> poison, i32 %b, i32 0
   %splat = shufflevector <vscale x 4 x i32> %head, <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer
   %vc = zext <vscale x 4 x i32> %va to <vscale x 4 x i64>
@@ -295,11 +376,18 @@
 }
 
 define <vscale x 4 x i64> @vwadd_wx_nxv4i64(<vscale x 4 x i64> %va, i32 %b) {
-; CHECK-LABEL: vwadd_wx_nxv4i64:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    vsetvli a1, zero, e32, m2, ta, mu
-; CHECK-NEXT:    vwadd.wx v8, v8, a0
-; CHECK-NEXT:    ret
+; RV32-LABEL: vwadd_wx_nxv4i64:
+; RV32:       # %bb.0:
+; RV32-NEXT:    vsetvli a1, zero, e32, m2, ta, mu
+; RV32-NEXT:    vwadd.wx v8, v8, a0
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: vwadd_wx_nxv4i64:
+; RV64:       # %bb.0:
+; RV64-NEXT:    sext.w a0, a0
+; RV64-NEXT:    vsetvli a1, zero, e64, m4, ta, mu
+; RV64-NEXT:    vadd.vx v8, v8, a0
+; RV64-NEXT:    ret
   %head = insertelement <vscale x 4 x i32> poison, i32 %b, i32 0
   %splat = shufflevector <vscale x 4 x i32> %head, <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer
   %vb = sext <vscale x 4 x i32> %splat to <vscale x 4 x i64>
@@ -308,11 +396,19 @@
 }
 
 define <vscale x 4 x i64> @vwaddu_wx_nxv4i64(<vscale x 4 x i64> %va, i32 %b) {
-; CHECK-LABEL: vwaddu_wx_nxv4i64:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    vsetvli a1, zero, e32, m2, ta, mu
-; CHECK-NEXT:    vwaddu.wx v8, v8, a0
-; CHECK-NEXT:    ret
+; RV32-LABEL: vwaddu_wx_nxv4i64:
+; RV32:       # %bb.0:
+; RV32-NEXT:    vsetvli a1, zero, e32, m2, ta, mu
+; RV32-NEXT:    vwaddu.wx v8, v8, a0
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: vwaddu_wx_nxv4i64:
+; RV64:       # %bb.0:
+; RV64-NEXT:    slli a0, a0, 32
+; RV64-NEXT:    srli a0, a0, 32
+; RV64-NEXT:    vsetvli a1, zero, e64, m4, ta, mu
+; RV64-NEXT:    vadd.vx v8, v8, a0
+; RV64-NEXT:    ret
   %head = insertelement <vscale x 4 x i32> poison, i32 %b, i32 0
   %splat = shufflevector <vscale x 4 x i32> %head, <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer
   %vb = zext <vscale x 4 x i32> %splat to <vscale x 4 x i64>
@@ -347,12 +443,20 @@
 }
 
 define <vscale x 8 x i64> @vwadd_vx_nxv8i64(<vscale x 8 x i32> %va, i32 %b) {
-; CHECK-LABEL: vwadd_vx_nxv8i64:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    vsetvli a1, zero, e32, m4, ta, mu
-; CHECK-NEXT:    vwadd.vx v16, v8, a0
-; CHECK-NEXT:    vmv8r.v v8, v16
-; CHECK-NEXT:    ret
+; RV32-LABEL: vwadd_vx_nxv8i64:
+; RV32:       # %bb.0:
+; RV32-NEXT:    vsetvli a1, zero, e32, m4, ta, mu
+; RV32-NEXT:    vwadd.vx v16, v8, a0
+; RV32-NEXT:    vmv8r.v v8, v16
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: vwadd_vx_nxv8i64:
+; RV64:       # %bb.0:
+; RV64-NEXT:    vsetvli a1, zero, e64, m8, ta, mu
+; RV64-NEXT:    vsext.vf2 v16, v8
+; RV64-NEXT:    sext.w a0, a0
+; RV64-NEXT:    vadd.vx v8, v16, a0
+; RV64-NEXT:    ret
   %head = insertelement <vscale x 8 x i32> poison, i32 %b, i32 0
   %splat = shufflevector <vscale x 8 x i32> %head, <vscale x 8 x i32> poison, <vscale x 8 x i32> zeroinitializer
   %vc = sext <vscale x 8 x i32> %va to <vscale x 8 x i64>
@@ -362,12 +466,21 @@
 }
 
 define <vscale x 8 x i64> @vwaddu_vx_nxv8i64(<vscale x 8 x i32> %va, i32 %b) {
-; CHECK-LABEL: vwaddu_vx_nxv8i64:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    vsetvli a1, zero, e32, m4, ta, mu
-; CHECK-NEXT:    vwaddu.vx v16, v8, a0
-; CHECK-NEXT:    vmv8r.v v8, v16
-; CHECK-NEXT:    ret
+; RV32-LABEL: vwaddu_vx_nxv8i64:
+; RV32:       # %bb.0:
+; RV32-NEXT:    vsetvli a1, zero, e32, m4, ta, mu
+; RV32-NEXT:    vwaddu.vx v16, v8, a0
+; RV32-NEXT:    vmv8r.v v8, v16
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: vwaddu_vx_nxv8i64:
+; RV64:       # %bb.0:
+; RV64-NEXT:    vsetvli a1, zero, e64, m8, ta, mu
+; RV64-NEXT:    vzext.vf2 v16, v8
+; RV64-NEXT:    slli a0, a0, 32
+; RV64-NEXT:    srli a0, a0, 32
+; RV64-NEXT:    vadd.vx v8, v16, a0
+; RV64-NEXT:    ret
   %head = insertelement <vscale x 8 x i32> poison, i32 %b, i32 0
   %splat = shufflevector <vscale x 8 x i32> %head, <vscale x 8 x i32> poison, <vscale x 8 x i32> zeroinitializer
   %vc = zext <vscale x 8 x i32> %va to <vscale x 8 x i64>
@@ -401,11 +514,18 @@
 }
 
 define <vscale x 8 x i64> @vwadd_wx_nxv8i64(<vscale x 8 x i64> %va, i32 %b) {
-; CHECK-LABEL: vwadd_wx_nxv8i64:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    vsetvli a1, zero, e32, m4, ta, mu
-; CHECK-NEXT:    vwadd.wx v8, v8, a0
-; CHECK-NEXT:    ret
+; RV32-LABEL: vwadd_wx_nxv8i64:
+; RV32:       # %bb.0:
+; RV32-NEXT:    vsetvli a1, zero, e32, m4, ta, mu
+; RV32-NEXT:    vwadd.wx v8, v8, a0
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: vwadd_wx_nxv8i64:
+; RV64:       # %bb.0:
+; RV64-NEXT:    sext.w a0, a0
+; RV64-NEXT:    vsetvli a1, zero, e64, m8, ta, mu
+; RV64-NEXT:    vadd.vx v8, v8, a0
+; RV64-NEXT:    ret
   %head = insertelement <vscale x 8 x i32> poison, i32 %b, i32 0
   %splat = shufflevector <vscale x 8 x i32> %head, <vscale x 8 x i32> poison, <vscale x 8 x i32> zeroinitializer
   %vb = sext <vscale x 8 x i32> %splat to <vscale x 8 x i64>
@@ -414,11 +534,19 @@
 }
 
 define <vscale x 8 x i64> @vwaddu_wx_nxv8i64(<vscale x 8 x i64> %va, i32 %b) {
-; CHECK-LABEL: vwaddu_wx_nxv8i64:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    vsetvli a1, zero, e32, m4, ta, mu
-; CHECK-NEXT:    vwaddu.wx v8, v8, a0
-; CHECK-NEXT:    ret
+; RV32-LABEL: vwaddu_wx_nxv8i64:
+; RV32:       # %bb.0:
+; RV32-NEXT:    vsetvli a1, zero, e32, m4, ta, mu
+; RV32-NEXT:    vwaddu.wx v8, v8, a0
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: vwaddu_wx_nxv8i64:
+; RV64:       # %bb.0:
+; RV64-NEXT:    slli a0, a0, 32
+; RV64-NEXT:    srli a0, a0, 32
+; RV64-NEXT:    vsetvli a1, zero, e64, m8, ta, mu
+; RV64-NEXT:    vadd.vx v8, v8, a0
+; RV64-NEXT:    ret
   %head = insertelement <vscale x 8 x i32> poison, i32 %b, i32 0
   %splat = shufflevector <vscale x 8 x i32> %head, <vscale x 8 x i32> poison, <vscale x 8 x i32> zeroinitializer
   %vb = zext <vscale x 8 x i32> %splat to <vscale x 8 x i64>
diff --git a/llvm/test/CodeGen/RISCV/rvv/vwmacc-sdnode.ll b/llvm/test/CodeGen/RISCV/rvv/vwmacc-sdnode.ll
--- a/llvm/test/CodeGen/RISCV/rvv/vwmacc-sdnode.ll
+++ b/llvm/test/CodeGen/RISCV/rvv/vwmacc-sdnode.ll
@@ -1,8 +1,8 @@
 ; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
 ; RUN: llc -mtriple=riscv32 -mattr=+v -target-abi=ilp32 \
-; RUN:     -verify-machineinstrs < %s | FileCheck %s
+; RUN:     -verify-machineinstrs < %s | FileCheck %s --check-prefixes=CHECK,RV32
 ; RUN: llc -mtriple=riscv64 -mattr=+v -target-abi=lp64 \
-; RUN:     -verify-machineinstrs < %s | FileCheck %s
+; RUN:     -verify-machineinstrs < %s | FileCheck %s --check-prefixes=CHECK,RV64
 
 define <vscale x 1 x i64> @vwmacc_vv_nxv1i32(<vscale x 1 x i32> %va, <vscale x 1 x i32> %vb, <vscale x 1 x i64> %vc) {
 ; CHECK-LABEL: vwmacc_vv_nxv1i32:
@@ -20,12 +20,21 @@
 }
 
 define <vscale x 1 x i64> @vwmacc_vx_nxv1i32(<vscale x 1 x i32> %va, i32 %b, <vscale x 1 x i64> %vc) {
-; CHECK-LABEL: vwmacc_vx_nxv1i32:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    vsetvli a1, zero, e32, mf2, ta, mu
-; CHECK-NEXT:    vwmacc.vx v9, a0, v8
-; CHECK-NEXT:    vmv1r.v v8, v9
-; CHECK-NEXT:    ret
+; RV32-LABEL: vwmacc_vx_nxv1i32:
+; RV32:       # %bb.0:
+; RV32-NEXT:    vsetvli a1, zero, e32, mf2, ta, mu
+; RV32-NEXT:    vwmacc.vx v9, a0, v8
+; RV32-NEXT:    vmv1r.v v8, v9
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: vwmacc_vx_nxv1i32:
+; RV64:       # %bb.0:
+; RV64-NEXT:    vsetvli a1, zero, e64, m1, ta, mu
+; RV64-NEXT:    vsext.vf2 v10, v8
+; RV64-NEXT:    sext.w a0, a0
+; RV64-NEXT:    vmadd.vx v10, a0, v9
+; RV64-NEXT:    vmv.v.v v8, v10
+; RV64-NEXT:    ret
   %head = insertelement <vscale x 1 x i32> poison, i32 %b, i32 0
   %splat = shufflevector <vscale x 1 x i32> %head, <vscale x 1 x i32> poison, <vscale x 1 x i32> zeroinitializer
   %vd = sext <vscale x 1 x i32> %va to <vscale x 1 x i64>
@@ -52,12 +61,22 @@
 }
 
 define <vscale x 1 x i64> @vwmaccu_vx_nxv1i32(<vscale x 1 x i32> %va, i32 %b, <vscale x 1 x i64> %vc) {
-; CHECK-LABEL: vwmaccu_vx_nxv1i32:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    vsetvli a1, zero, e32, mf2, ta, mu
-; CHECK-NEXT:    vwmaccu.vx v9, a0, v8
-; CHECK-NEXT:    vmv1r.v v8, v9
-; CHECK-NEXT:    ret
+; RV32-LABEL: vwmaccu_vx_nxv1i32:
+; RV32:       # %bb.0:
+; RV32-NEXT:    vsetvli a1, zero, e32, mf2, ta, mu
+; RV32-NEXT:    vwmaccu.vx v9, a0, v8
+; RV32-NEXT:    vmv1r.v v8, v9
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: vwmaccu_vx_nxv1i32:
+; RV64:       # %bb.0:
+; RV64-NEXT:    vsetvli a1, zero, e64, m1, ta, mu
+; RV64-NEXT:    vzext.vf2 v10, v8
+; RV64-NEXT:    slli a0, a0, 32
+; RV64-NEXT:    srli a0, a0, 32
+; RV64-NEXT:    vmadd.vx v10, a0, v9
+; RV64-NEXT:    vmv.v.v v8, v10
+; RV64-NEXT:    ret
   %head = insertelement <vscale x 1 x i32> poison, i32 %b, i32 0
   %splat = shufflevector <vscale x 1 x i32> %head, <vscale x 1 x i32> poison, <vscale x 1 x i32> zeroinitializer
   %vd = zext <vscale x 1 x i32> %va to <vscale x 1 x i64>
@@ -84,12 +103,21 @@
 }
 
 define <vscale x 1 x i64> @vwmaccsu_vx_nxv1i32(<vscale x 1 x i32> %va, i32 %b, <vscale x 1 x i64> %vc) {
-; CHECK-LABEL: vwmaccsu_vx_nxv1i32:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    vsetvli a1, zero, e32, mf2, ta, mu
-; CHECK-NEXT:    vwmaccsu.vx v9, a0, v8
-; CHECK-NEXT:    vmv1r.v v8, v9
-; CHECK-NEXT:    ret
+; RV32-LABEL: vwmaccsu_vx_nxv1i32:
+; RV32:       # %bb.0:
+; RV32-NEXT:    vsetvli a1, zero, e32, mf2, ta, mu
+; RV32-NEXT:    vwmaccsu.vx v9, a0, v8
+; RV32-NEXT:    vmv1r.v v8, v9
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: vwmaccsu_vx_nxv1i32:
+; RV64:       # %bb.0:
+; RV64-NEXT:    vsetvli a1, zero, e64, m1, ta, mu
+; RV64-NEXT:    vzext.vf2 v10, v8
+; RV64-NEXT:    sext.w a0, a0
+; RV64-NEXT:    vmadd.vx v10, a0, v9
+; RV64-NEXT:    vmv.v.v v8, v10
+; RV64-NEXT:    ret
   %head = insertelement <vscale x 1 x i32> poison, i32 %b, i32 0
   %splat = shufflevector <vscale x 1 x i32> %head, <vscale x 1 x i32> poison, <vscale x 1 x i32> zeroinitializer
   %vd = zext <vscale x 1 x i32> %va to <vscale x 1 x i64>
@@ -101,12 +129,22 @@
 }
 
 define <vscale x 1 x i64> @vwmaccus_vx_nxv1i32(<vscale x 1 x i32> %va, i32 %b, <vscale x 1 x i64> %vc) {
-; CHECK-LABEL: vwmaccus_vx_nxv1i32:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    vsetvli a1, zero, e32, mf2, ta, mu
-; CHECK-NEXT:    vwmaccus.vx v9, a0, v8
-; CHECK-NEXT:    vmv1r.v v8, v9
-; CHECK-NEXT:    ret
+; RV32-LABEL: vwmaccus_vx_nxv1i32:
+; RV32:       # %bb.0:
+; RV32-NEXT:    vsetvli a1, zero, e32, mf2, ta, mu
+; RV32-NEXT:    vwmaccus.vx v9, a0, v8
+; RV32-NEXT:    vmv1r.v v8, v9
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: vwmaccus_vx_nxv1i32:
+; RV64:       # %bb.0:
+; RV64-NEXT:    vsetvli a1, zero, e64, m1, ta, mu
+; RV64-NEXT:    vsext.vf2 v10, v8
+; RV64-NEXT:    slli a0, a0, 32
+; RV64-NEXT:    srli a0, a0, 32
+; RV64-NEXT:    vmadd.vx v10, a0, v9
+; RV64-NEXT:    vmv.v.v v8, v10
+; RV64-NEXT:    ret
   %head = insertelement <vscale x 1 x i32> poison, i32 %b, i32 0
   %splat = shufflevector <vscale x 1 x i32> %head, <vscale x 1 x i32> poison, <vscale x 1 x i32> zeroinitializer
   %vd = sext <vscale x 1 x i32> %va to <vscale x 1 x i64>
@@ -133,12 +171,21 @@
 }
 
 define <vscale x 2 x i64> @vwmacc_vx_nxv2i32(<vscale x 2 x i32> %va, i32 %b, <vscale x 2 x i64> %vc) {
-; CHECK-LABEL: vwmacc_vx_nxv2i32:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    vsetvli a1, zero, e32, m1, ta, mu
-; CHECK-NEXT:    vwmacc.vx v10, a0, v8
-; CHECK-NEXT:    vmv2r.v v8, v10
-; CHECK-NEXT:    ret
+; RV32-LABEL: vwmacc_vx_nxv2i32:
+; RV32:       # %bb.0:
+; RV32-NEXT:    vsetvli a1, zero, e32, m1, ta, mu
+; RV32-NEXT:    vwmacc.vx v10, a0, v8
+; RV32-NEXT:    vmv2r.v v8, v10
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: vwmacc_vx_nxv2i32:
+; RV64:       # %bb.0:
+; RV64-NEXT:    vsetvli a1, zero, e64, m2, ta, mu
+; RV64-NEXT:    vsext.vf2 v12, v8
+; RV64-NEXT:    sext.w a0, a0
+; RV64-NEXT:    vmadd.vx v12, a0, v10
+; RV64-NEXT:    vmv.v.v v8, v12
+; RV64-NEXT:    ret
   %head = insertelement <vscale x 2 x i32> poison, i32 %b, i32 0
   %splat = shufflevector <vscale x 2 x i32> %head, <vscale x 2 x i32> poison, <vscale x 2 x i32> zeroinitializer
   %vd = sext <vscale x 2 x i32> %va to <vscale x 2 x i64>
@@ -165,12 +212,22 @@
 }
 
 define <vscale x 2 x i64> @vwmaccu_vx_nxv2i32(<vscale x 2 x i32> %va, i32 %b, <vscale x 2 x i64> %vc) {
-; CHECK-LABEL: vwmaccu_vx_nxv2i32:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    vsetvli a1, zero, e32, m1, ta, mu
-; CHECK-NEXT:    vwmaccu.vx v10, a0, v8
-; CHECK-NEXT:    vmv2r.v v8, v10
-; CHECK-NEXT:    ret
+; RV32-LABEL: vwmaccu_vx_nxv2i32:
+; RV32:       # %bb.0:
+; RV32-NEXT:    vsetvli a1, zero, e32, m1, ta, mu
+; RV32-NEXT:    vwmaccu.vx v10, a0, v8
+; RV32-NEXT:    vmv2r.v v8, v10
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: vwmaccu_vx_nxv2i32:
+; RV64:       # %bb.0:
+; RV64-NEXT:    vsetvli a1, zero, e64, m2, ta, mu
+; RV64-NEXT:    vzext.vf2 v12, v8
+; RV64-NEXT:    slli a0, a0, 32
+; RV64-NEXT:    srli a0, a0, 32
+; RV64-NEXT:    vmadd.vx v12, a0, v10
+; RV64-NEXT:    vmv.v.v v8, v12
+; RV64-NEXT:    ret
   %head = insertelement <vscale x 2 x i32> poison, i32 %b, i32 0
   %splat = shufflevector <vscale x 2 x i32> %head, <vscale x 2 x i32> poison, <vscale x 2 x i32> zeroinitializer
   %vd = zext <vscale x 2 x i32> %va to <vscale x 2 x i64>
@@ -197,12 +254,21 @@
 }
 
 define <vscale x 2 x i64> @vwmaccsu_vx_nxv2i32(<vscale x 2 x i32> %va, i32 %b, <vscale x 2 x i64> %vc) {
-; CHECK-LABEL: vwmaccsu_vx_nxv2i32:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    vsetvli a1, zero, e32, m1, ta, mu
-; CHECK-NEXT:    vwmaccsu.vx v10, a0, v8
-; CHECK-NEXT:    vmv2r.v v8, v10
-; CHECK-NEXT:    ret
+; RV32-LABEL: vwmaccsu_vx_nxv2i32:
+; RV32:       # %bb.0:
+; RV32-NEXT:    vsetvli a1, zero, e32, m1, ta, mu
+; RV32-NEXT:    vwmaccsu.vx v10, a0, v8
+; RV32-NEXT:    vmv2r.v v8, v10
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: vwmaccsu_vx_nxv2i32:
+; RV64:       # %bb.0:
+; RV64-NEXT:    vsetvli a1, zero, e64, m2, ta, mu
+; RV64-NEXT:    vzext.vf2 v12, v8
+; RV64-NEXT:    sext.w a0, a0
+; RV64-NEXT:    vmadd.vx v12, a0, v10
+; RV64-NEXT:    vmv.v.v v8, v12
+; RV64-NEXT:    ret
   %head = insertelement <vscale x 2 x i32> poison, i32 %b, i32 0
   %splat = shufflevector <vscale x 2 x i32> %head, <vscale x 2 x i32> poison, <vscale x 2 x i32> zeroinitializer
   %vd = zext <vscale x 2 x i32> %va to <vscale x 2 x i64>
@@ -214,12 +280,22 @@
 }
 
 define <vscale x 2 x i64> @vwmaccus_vx_nxv2i32(<vscale x 2 x i32> %va, i32 %b, <vscale x 2 x i64> %vc) {
-; CHECK-LABEL: vwmaccus_vx_nxv2i32:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    vsetvli a1, zero, e32, m1, ta, mu
-; CHECK-NEXT:    vwmaccus.vx v10, a0, v8
-; CHECK-NEXT:    vmv2r.v v8, v10
-; CHECK-NEXT:    ret
+; RV32-LABEL: vwmaccus_vx_nxv2i32:
+; RV32:       # %bb.0:
+; RV32-NEXT:    vsetvli a1, zero, e32, m1, ta, mu
+; RV32-NEXT:    vwmaccus.vx v10, a0, v8
+; RV32-NEXT:    vmv2r.v v8, v10
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: vwmaccus_vx_nxv2i32:
+; RV64:       # %bb.0:
+; RV64-NEXT:    vsetvli a1, zero, e64, m2, ta, mu
+; RV64-NEXT:    vsext.vf2 v12, v8
+; RV64-NEXT:    slli a0, a0, 32
+; RV64-NEXT:    srli a0, a0, 32
+; RV64-NEXT:    vmadd.vx v12, a0, v10
+; RV64-NEXT:    vmv.v.v v8, v12
+; RV64-NEXT:    ret
   %head = insertelement <vscale x 2 x i32> poison, i32 %b, i32 0
   %splat = shufflevector <vscale x 2 x i32> %head, <vscale x 2 x i32> poison, <vscale x 2 x i32> zeroinitializer
   %vd = sext <vscale x 2 x i32> %va to <vscale x 2 x i64>
@@ -246,12 +322,21 @@
 }
 
 define <vscale x 4 x i64> @vwmacc_vx_nxv4i32(<vscale x 4 x i32> %va, i32 %b, <vscale x 4 x i64> %vc) {
-; CHECK-LABEL: vwmacc_vx_nxv4i32:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    vsetvli a1, zero, e32, m2, ta, mu
-; CHECK-NEXT:    vwmacc.vx v12, a0, v8
-; CHECK-NEXT:    vmv4r.v v8, v12
-; CHECK-NEXT:    ret
+; RV32-LABEL: vwmacc_vx_nxv4i32:
+; RV32:       # %bb.0:
+; RV32-NEXT:    vsetvli a1, zero, e32, m2, ta, mu
+; RV32-NEXT:    vwmacc.vx v12, a0, v8
+; RV32-NEXT:    vmv4r.v v8, v12
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: vwmacc_vx_nxv4i32:
+; RV64:       # %bb.0:
+; RV64-NEXT:    vsetvli a1, zero, e64, m4, ta, mu
+; RV64-NEXT:    vsext.vf2 v16, v8
+; RV64-NEXT:    sext.w a0, a0
+; RV64-NEXT:    vmadd.vx v16, a0, v12
+; RV64-NEXT:    vmv.v.v v8, v16
+; RV64-NEXT:    ret
   %head = insertelement <vscale x 4 x i32> poison, i32 %b, i32 0
   %splat = shufflevector <vscale x 4 x i32> %head, <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer
   %vd = sext <vscale x 4 x i32> %va to <vscale x 4 x i64>
@@ -278,12 +363,22 @@
 }
 
 define <vscale x 4 x i64> @vwmaccu_vx_nxv4i32(<vscale x 4 x i32> %va, i32 %b, <vscale x 4 x i64> %vc) {
-; CHECK-LABEL: vwmaccu_vx_nxv4i32:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    vsetvli a1, zero, e32, m2, ta, mu
-; CHECK-NEXT:    vwmaccu.vx v12, a0, v8
-; CHECK-NEXT:    vmv4r.v v8, v12
-; CHECK-NEXT:    ret
+; RV32-LABEL: vwmaccu_vx_nxv4i32:
+; RV32:       # %bb.0:
+; RV32-NEXT:    vsetvli a1, zero, e32, m2, ta, mu
+; RV32-NEXT:    vwmaccu.vx v12, a0, v8
+; RV32-NEXT:    vmv4r.v v8, v12
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: vwmaccu_vx_nxv4i32:
+; RV64:       # %bb.0:
+; RV64-NEXT:    vsetvli a1, zero, e64, m4, ta, mu
+; RV64-NEXT:    vzext.vf2 v16, v8
+; RV64-NEXT:    slli a0, a0, 32
+; RV64-NEXT:    srli a0, a0, 32
+; RV64-NEXT:    vmadd.vx v16, a0, v12
+; RV64-NEXT:    vmv.v.v v8, v16
+; RV64-NEXT:    ret
   %head = insertelement <vscale x 4 x i32> poison, i32 %b, i32 0
   %splat = shufflevector <vscale x 4 x i32> %head, <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer
   %vd = zext <vscale x 4 x i32> %va to <vscale x 4 x i64>
@@ -310,12 +405,21 @@
 }
 
 define <vscale x 4 x i64> @vwmaccsu_vx_nxv4i32(<vscale x 4 x i32> %va, i32 %b, <vscale x 4 x i64> %vc) {
-; CHECK-LABEL: vwmaccsu_vx_nxv4i32:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    vsetvli a1, zero, e32, m2, ta, mu
-; CHECK-NEXT:    vwmaccsu.vx v12, a0, v8
-; CHECK-NEXT:    vmv4r.v v8, v12
-; CHECK-NEXT:    ret
+; RV32-LABEL: vwmaccsu_vx_nxv4i32:
+; RV32:       # %bb.0:
+; RV32-NEXT:    vsetvli a1, zero, e32, m2, ta, mu
+; RV32-NEXT:    vwmaccsu.vx v12, a0, v8
+; RV32-NEXT:    vmv4r.v v8, v12
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: vwmaccsu_vx_nxv4i32:
+; RV64:       # %bb.0:
+; RV64-NEXT:    vsetvli a1, zero, e64, m4, ta, mu
+; RV64-NEXT:    vzext.vf2 v16, v8
+; RV64-NEXT:    sext.w a0, a0
+; RV64-NEXT:    vmadd.vx v16, a0, v12
+; RV64-NEXT:    vmv.v.v v8, v16
+; RV64-NEXT:    ret
   %head = insertelement <vscale x 4 x i32> poison, i32 %b, i32 0
   %splat = shufflevector <vscale x 4 x i32> %head, <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer
   %vd = zext <vscale x 4 x i32> %va to <vscale x 4 x i64>
@@ -327,12 +431,22 @@
 }
 
 define <vscale x 4 x i64> @vwmaccus_vx_nxv4i32(<vscale x 4 x i32> %va, i32 %b, <vscale x 4 x i64> %vc) {
-; CHECK-LABEL: vwmaccus_vx_nxv4i32:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    vsetvli a1, zero, e32, m2, ta, mu
-; CHECK-NEXT:    vwmaccus.vx v12, a0, v8
-; CHECK-NEXT:    vmv4r.v v8, v12
-; CHECK-NEXT:    ret
+; RV32-LABEL: vwmaccus_vx_nxv4i32:
+; RV32:       # %bb.0:
+; RV32-NEXT:    vsetvli a1, zero, e32, m2, ta, mu
+; RV32-NEXT:    vwmaccus.vx v12, a0, v8
+; RV32-NEXT:    vmv4r.v v8, v12
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: vwmaccus_vx_nxv4i32:
+; RV64:       # %bb.0:
+; RV64-NEXT:    vsetvli a1, zero, e64, m4, ta, mu
+; RV64-NEXT:    vsext.vf2 v16, v8
+; RV64-NEXT:    slli a0, a0, 32
+; RV64-NEXT:    srli a0, a0, 32
+; RV64-NEXT:    vmadd.vx v16, a0, v12
+; RV64-NEXT:    vmv.v.v v8, v16
+; RV64-NEXT:    ret
   %head = insertelement <vscale x 4 x i32> poison, i32 %b, i32 0
   %splat = shufflevector <vscale x 4 x i32> %head, <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer
   %vd = sext <vscale x 4 x i32> %va to <vscale x 4 x i64>
@@ -359,12 +473,21 @@
 }
 
 define <vscale x 8 x i64> @vwmacc_vx_nxv8i32(<vscale x 8 x i32> %va, i32 %b, <vscale x 8 x i64> %vc) {
-; CHECK-LABEL: vwmacc_vx_nxv8i32:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    vsetvli a1, zero, e32, m4, ta, mu
-; CHECK-NEXT:    vwmacc.vx v16, a0, v8
-; CHECK-NEXT:    vmv8r.v v8, v16
-; CHECK-NEXT:    ret
+; RV32-LABEL: vwmacc_vx_nxv8i32:
+; RV32:       # %bb.0:
+; RV32-NEXT:    vsetvli a1, zero, e32, m4, ta, mu
+; RV32-NEXT:    vwmacc.vx v16, a0, v8
+; RV32-NEXT:    vmv8r.v v8, v16
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: vwmacc_vx_nxv8i32:
+; RV64:       # %bb.0:
+; RV64-NEXT:    vsetvli a1, zero, e64, m8, ta, mu
+; RV64-NEXT:    vsext.vf2 v24, v8
+; RV64-NEXT:    sext.w a0, a0
+; RV64-NEXT:    vmadd.vx v24, a0, v16
+; RV64-NEXT:    vmv.v.v v8, v24
+; RV64-NEXT:    ret
   %head = insertelement <vscale x 8 x i32> poison, i32 %b, i32 0
   %splat = shufflevector <vscale x 8 x i32> %head, <vscale x 8 x i32> poison, <vscale x 8 x i32> zeroinitializer
   %vd = sext <vscale x 8 x i32> %va to <vscale x 8 x i64>
@@ -391,12 +514,22 @@
 }
 
 define <vscale x 8 x i64> @vwmaccu_vx_nxv8i32(<vscale x 8 x i32> %va, i32 %b, <vscale x 8 x i64> %vc) {
-; CHECK-LABEL: vwmaccu_vx_nxv8i32:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    vsetvli a1, zero, e32, m4, ta, mu
-; CHECK-NEXT:    vwmaccu.vx v16, a0, v8
-; CHECK-NEXT:    vmv8r.v v8, v16
-; CHECK-NEXT:    ret
+; RV32-LABEL: vwmaccu_vx_nxv8i32:
+; RV32:       # %bb.0:
+; RV32-NEXT:    vsetvli a1, zero, e32, m4, ta, mu
+; RV32-NEXT:    vwmaccu.vx v16, a0, v8
+; RV32-NEXT:    vmv8r.v v8, v16
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: vwmaccu_vx_nxv8i32:
+; RV64:       # %bb.0:
+; RV64-NEXT:    vsetvli a1, zero, e64, m8, ta, mu
+; RV64-NEXT:    vzext.vf2 v24, v8
+; RV64-NEXT:    slli a0, a0, 32
+; RV64-NEXT:    srli a0, a0, 32
+; RV64-NEXT:    vmadd.vx v24, a0, v16
+; RV64-NEXT:    vmv.v.v v8, v24
+; RV64-NEXT:    ret
   %head = insertelement <vscale x 8 x i32> poison, i32 %b, i32 0
   %splat = shufflevector <vscale x 8 x i32> %head, <vscale x 8 x i32> poison, <vscale x 8 x i32> zeroinitializer
   %vd = zext <vscale x 8 x i32> %va to <vscale x 8 x i64>
@@ -423,12 +556,21 @@
 }
 
 define <vscale x 8 x i64> @vwmaccsu_vx_nxv8i32(<vscale x 8 x i32> %va, i32 %b, <vscale x 8 x i64> %vc) {
-; CHECK-LABEL: vwmaccsu_vx_nxv8i32:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    vsetvli a1, zero, e32, m4, ta, mu
-; CHECK-NEXT:    vwmaccsu.vx v16, a0, v8
-; CHECK-NEXT:    vmv8r.v v8, v16
-; CHECK-NEXT:    ret
+; RV32-LABEL: vwmaccsu_vx_nxv8i32:
+; RV32:       # %bb.0:
+; RV32-NEXT:    vsetvli a1, zero, e32, m4, ta, mu
+; RV32-NEXT:    vwmaccsu.vx v16, a0, v8
+; RV32-NEXT:    vmv8r.v v8, v16
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: vwmaccsu_vx_nxv8i32:
+; RV64:       # %bb.0:
+; RV64-NEXT:    vsetvli a1, zero, e64, m8, ta, mu
+; RV64-NEXT:    vzext.vf2 v24, v8
+; RV64-NEXT:    sext.w a0, a0
+; RV64-NEXT:    vmadd.vx v24, a0, v16
+; RV64-NEXT:    vmv.v.v v8, v24
+; RV64-NEXT:    ret
   %head = insertelement <vscale x 8 x i32> poison, i32 %b, i32 0
   %splat = shufflevector <vscale x 8 x i32> %head, <vscale x 8 x i32> poison, <vscale x 8 x i32> zeroinitializer
   %vd = zext <vscale x 8 x i32> %va to <vscale x 8 x i64>
@@ -440,12 +582,22 @@
 }
 
 define <vscale x 8 x i64> @vwmaccus_vx_nxv8i32(<vscale x 8 x i32> %va, i32 %b, <vscale x 8 x i64> %vc) {
-; CHECK-LABEL: vwmaccus_vx_nxv8i32:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    vsetvli a1, zero, e32, m4, ta, mu
-; CHECK-NEXT:    vwmaccus.vx v16, a0, v8
-; CHECK-NEXT:    vmv8r.v v8, v16
-; CHECK-NEXT:    ret
+; RV32-LABEL: vwmaccus_vx_nxv8i32:
+; RV32:       # %bb.0:
+; RV32-NEXT:    vsetvli a1, zero, e32, m4, ta, mu
+; RV32-NEXT:    vwmaccus.vx v16, a0, v8
+; RV32-NEXT:    vmv8r.v v8, v16
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: vwmaccus_vx_nxv8i32:
+; RV64:       # %bb.0:
+; RV64-NEXT:    vsetvli a1, zero, e64, m8, ta, mu
+; RV64-NEXT:    vsext.vf2 v24, v8
+; RV64-NEXT:    slli a0, a0, 32
+; RV64-NEXT:    srli a0, a0, 32
+; RV64-NEXT:    vmadd.vx v24, a0, v16
+; RV64-NEXT:    vmv.v.v v8, v24
+; RV64-NEXT:    ret
   %head = insertelement <vscale x 8 x i32> poison, i32 %b, i32 0
   %splat = shufflevector <vscale x 8 x i32> %head, <vscale x 8 x i32> poison, <vscale x 8 x i32> zeroinitializer
   %vd = sext <vscale x 8 x i32> %va to <vscale x 8 x i64>
diff --git a/llvm/test/CodeGen/RISCV/rvv/vwmul-sdnode.ll b/llvm/test/CodeGen/RISCV/rvv/vwmul-sdnode.ll
--- a/llvm/test/CodeGen/RISCV/rvv/vwmul-sdnode.ll
+++ b/llvm/test/CodeGen/RISCV/rvv/vwmul-sdnode.ll
@@ -1,6 +1,6 @@
 ; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
-; RUN: llc -mtriple=riscv32 -mattr=+v -verify-machineinstrs < %s | FileCheck %s
-; RUN: llc -mtriple=riscv64 -mattr=+v -verify-machineinstrs < %s | FileCheck %s
+; RUN: llc -mtriple=riscv32 -mattr=+v -verify-machineinstrs < %s | FileCheck %s --check-prefixes=CHECK,RV32
+; RUN: llc -mtriple=riscv64 -mattr=+v -verify-machineinstrs < %s | FileCheck %s --check-prefixes=CHECK,RV64
 
 define <vscale x 1 x i64> @vwmul_vv_nxv1i64(<vscale x 1 x i32> %va, <vscale x 1 x i32> %vb) {
 ; CHECK-LABEL: vwmul_vv_nxv1i64:
@@ -42,12 +42,20 @@
 }
 
 define <vscale x 1 x i64> @vwmul_vx_nxv1i64(<vscale x 1 x i32> %va, i32 %b) {
-; CHECK-LABEL: vwmul_vx_nxv1i64:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    vsetvli a1, zero, e32, mf2, ta, mu
-; CHECK-NEXT:    vwmul.vx v9, v8, a0
-; CHECK-NEXT:    vmv1r.v v8, v9
-; CHECK-NEXT:    ret
+; RV32-LABEL: vwmul_vx_nxv1i64:
+; RV32:       # %bb.0:
+; RV32-NEXT:    vsetvli a1, zero, e32, mf2, ta, mu
+; RV32-NEXT:    vwmul.vx v9, v8, a0
+; RV32-NEXT:    vmv1r.v v8, v9
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: vwmul_vx_nxv1i64:
+; RV64:       # %bb.0:
+; RV64-NEXT:    vsetvli a1, zero, e64, m1, ta, mu
+; RV64-NEXT:    vsext.vf2 v9, v8
+; RV64-NEXT:    sext.w a0, a0
+; RV64-NEXT:    vmul.vx v8, v9, a0
+; RV64-NEXT:    ret
   %head = insertelement <vscale x 1 x i32> undef, i32 %b, i32 0
   %splat = shufflevector <vscale x 1 x i32> %head, <vscale x 1 x i32> undef, <vscale x 1 x i32> zeroinitializer
   %vc = sext <vscale x 1 x i32> %va to <vscale x 1 x i64>
@@ -57,12 +65,21 @@
 }
 
 define <vscale x 1 x i64> @vwmulu_vx_nxv1i64(<vscale x 1 x i32> %va, i32 %b) {
-; CHECK-LABEL: vwmulu_vx_nxv1i64:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    vsetvli a1, zero, e32, mf2, ta, mu
-; CHECK-NEXT:    vwmulu.vx v9, v8, a0
-; CHECK-NEXT:    vmv1r.v v8, v9
-; CHECK-NEXT:    ret
+; RV32-LABEL: vwmulu_vx_nxv1i64:
+; RV32:       # %bb.0:
+; RV32-NEXT:    vsetvli a1, zero, e32, mf2, ta, mu
+; RV32-NEXT:    vwmulu.vx v9, v8, a0
+; RV32-NEXT:    vmv1r.v v8, v9
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: vwmulu_vx_nxv1i64:
+; RV64:       # %bb.0:
+; RV64-NEXT:    vsetvli a1, zero, e64, m1, ta, mu
+; RV64-NEXT:    vzext.vf2 v9, v8
+; RV64-NEXT:    slli a0, a0, 32
+; RV64-NEXT:    srli a0, a0, 32
+; RV64-NEXT:    vmul.vx v8, v9, a0
+; RV64-NEXT:    ret
   %head = insertelement <vscale x 1 x i32> undef, i32 %b, i32 0
   %splat = shufflevector <vscale x 1 x i32> %head, <vscale x 1 x i32> undef, <vscale x 1 x i32> zeroinitializer
   %vc = zext <vscale x 1 x i32> %va to <vscale x 1 x i64>
@@ -72,12 +89,21 @@
 }
 
 define <vscale x 1 x i64> @vwmulsu_vx_nxv1i64(<vscale x 1 x i32> %va, i32 %b) {
-; CHECK-LABEL: vwmulsu_vx_nxv1i64:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    vsetvli a1, zero, e32, mf2, ta, mu
-; CHECK-NEXT:    vwmulsu.vx v9, v8, a0
-; CHECK-NEXT:    vmv1r.v v8, v9
-; CHECK-NEXT:    ret
+; RV32-LABEL: vwmulsu_vx_nxv1i64:
+; RV32:       # %bb.0:
+; RV32-NEXT:    vsetvli a1, zero, e32, mf2, ta, mu
+; RV32-NEXT:    vwmulsu.vx v9, v8, a0
+; RV32-NEXT:    vmv1r.v v8, v9
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: vwmulsu_vx_nxv1i64:
+; RV64:       # %bb.0:
+; RV64-NEXT:    vsetvli a1, zero, e64, m1, ta, mu
+; RV64-NEXT:    vsext.vf2 v9, v8
+; RV64-NEXT:    slli a0, a0, 32
+; RV64-NEXT:    srli a0, a0, 32
+; RV64-NEXT:    vmul.vx v8, v9, a0
+; RV64-NEXT:    ret
   %head = insertelement <vscale x 1 x i32> undef, i32 %b, i32 0
   %splat = shufflevector <vscale x 1 x i32> %head, <vscale x 1 x i32> undef, <vscale x 1 x i32> zeroinitializer
   %vc = sext <vscale x 1 x i32> %va to <vscale x 1 x i64>
@@ -126,12 +152,20 @@
 }
 
 define <vscale x 2 x i64> @vwmul_vx_nxv2i64(<vscale x 2 x i32> %va, i32 %b) {
-; CHECK-LABEL: vwmul_vx_nxv2i64:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    vsetvli a1, zero, e32, m1, ta, mu
-; CHECK-NEXT:    vwmul.vx v10, v8, a0
-; CHECK-NEXT:    vmv2r.v v8, v10
-; CHECK-NEXT:    ret
+; RV32-LABEL: vwmul_vx_nxv2i64:
+; RV32:       # %bb.0:
+; RV32-NEXT:    vsetvli a1, zero, e32, m1, ta, mu
+; RV32-NEXT:    vwmul.vx v10, v8, a0
+; RV32-NEXT:    vmv2r.v v8, v10
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: vwmul_vx_nxv2i64:
+; RV64:       # %bb.0:
+; RV64-NEXT:    vsetvli a1, zero, e64, m2, ta, mu
+; RV64-NEXT:    vsext.vf2 v10, v8
+; RV64-NEXT:    sext.w a0, a0
+; RV64-NEXT:    vmul.vx v8, v10, a0
+; RV64-NEXT:    ret
   %head = insertelement <vscale x 2 x i32> undef, i32 %b, i32 0
   %splat = shufflevector <vscale x 2 x i32> %head, <vscale x 2 x i32> undef, <vscale x 2 x i32> zeroinitializer
   %vc = sext <vscale x 2 x i32> %va to <vscale x 2 x i64>
@@ -141,12 +175,21 @@
 }
 
 define <vscale x 2 x i64> @vwmulu_vx_nxv2i64(<vscale x 2 x i32> %va, i32 %b) {
-; CHECK-LABEL: vwmulu_vx_nxv2i64:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    vsetvli a1, zero, e32, m1, ta, mu
-; CHECK-NEXT:    vwmulu.vx v10, v8, a0
-; CHECK-NEXT:    vmv2r.v v8, v10
-; CHECK-NEXT:    ret
+; RV32-LABEL: vwmulu_vx_nxv2i64:
+; RV32:       # %bb.0:
+; RV32-NEXT:    vsetvli a1, zero, e32, m1, ta, mu
+; RV32-NEXT:    vwmulu.vx v10, v8, a0
+; RV32-NEXT:    vmv2r.v v8, v10
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: vwmulu_vx_nxv2i64:
+; RV64:       # %bb.0:
+; RV64-NEXT:    vsetvli a1, zero, e64, m2, ta, mu
+; RV64-NEXT:    vzext.vf2 v10, v8
+; RV64-NEXT:    slli a0, a0, 32
+; RV64-NEXT:    srli a0, a0, 32
+; RV64-NEXT:    vmul.vx v8, v10, a0
+; RV64-NEXT:    ret
   %head = insertelement <vscale x 2 x i32> undef, i32 %b, i32 0
   %splat = shufflevector <vscale x 2 x i32> %head, <vscale x 2 x i32> undef, <vscale x 2 x i32> zeroinitializer
   %vc = zext <vscale x 2 x i32> %va to <vscale x 2 x i64>
@@ -156,12 +199,21 @@
 }
 
 define <vscale x 2 x i64> @vwmulsu_vx_nxv2i64(<vscale x 2 x i32> %va, i32 %b) {
-; CHECK-LABEL: vwmulsu_vx_nxv2i64:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    vsetvli a1, zero, e32, m1, ta, mu
-; CHECK-NEXT:    vwmulsu.vx v10, v8, a0
-; CHECK-NEXT:    vmv2r.v v8, v10
-; CHECK-NEXT:    ret
+; RV32-LABEL: vwmulsu_vx_nxv2i64:
+; RV32:       # %bb.0:
+; RV32-NEXT:    vsetvli a1, zero, e32, m1, ta, mu
+; RV32-NEXT:    vwmulsu.vx v10, v8, a0
+; RV32-NEXT:    vmv2r.v v8, v10
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: vwmulsu_vx_nxv2i64:
+; RV64:       # %bb.0:
+; RV64-NEXT:    vsetvli a1, zero, e64, m2, ta, mu
+; RV64-NEXT:    vsext.vf2 v10, v8
+; RV64-NEXT:    slli a0, a0, 32
+; RV64-NEXT:    srli a0, a0, 32
+; RV64-NEXT:    vmul.vx v8, v10, a0
+; RV64-NEXT:    ret
   %head = insertelement <vscale x 2 x i32> undef, i32 %b, i32 0
   %splat = shufflevector <vscale x 2 x i32> %head, <vscale x 2 x i32> undef, <vscale x 2 x i32> zeroinitializer
   %vc = sext <vscale x 2 x i32> %va to <vscale x 2 x i64>
@@ -210,12 +262,20 @@
 }
 
 define <vscale x 4 x i64> @vwmul_vx_nxv4i64(<vscale x 4 x i32> %va, i32 %b) {
-; CHECK-LABEL: vwmul_vx_nxv4i64:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    vsetvli a1, zero, e32, m2, ta, mu
-; CHECK-NEXT:    vwmul.vx v12, v8, a0
-; CHECK-NEXT:    vmv4r.v v8, v12
-; CHECK-NEXT:    ret
+; RV32-LABEL: vwmul_vx_nxv4i64:
+; RV32:       # %bb.0:
+; RV32-NEXT:    vsetvli a1, zero, e32, m2, ta, mu
+; RV32-NEXT:    vwmul.vx v12, v8, a0
+; RV32-NEXT:    vmv4r.v v8, v12
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: vwmul_vx_nxv4i64:
+; RV64:       # %bb.0:
+; RV64-NEXT:    vsetvli a1, zero, e64, m4, ta, mu
+; RV64-NEXT:    vsext.vf2 v12, v8
+; RV64-NEXT:    sext.w a0, a0
+; RV64-NEXT:    vmul.vx v8, v12, a0
+; RV64-NEXT:    ret
   %head = insertelement <vscale x 4 x i32> undef, i32 %b, i32 0
   %splat = shufflevector <vscale x 4 x i32> %head, <vscale x 4 x i32> undef, <vscale x 4 x i32> zeroinitializer
   %vc = sext <vscale x 4 x i32> %va to <vscale x 4 x i64>
@@ -225,12 +285,21 @@
 }
 
 define <vscale x 4 x i64> @vwmulu_vx_nxv4i64(<vscale x 4 x i32> %va, i32 %b) {
-; CHECK-LABEL: vwmulu_vx_nxv4i64:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    vsetvli a1, zero, e32, m2, ta, mu
-; CHECK-NEXT:    vwmulu.vx v12, v8, a0
-; CHECK-NEXT:    vmv4r.v v8, v12
-; CHECK-NEXT:    ret
+; RV32-LABEL: vwmulu_vx_nxv4i64:
+; RV32:       # %bb.0:
+; RV32-NEXT:    vsetvli a1, zero, e32, m2, ta, mu
+; RV32-NEXT:    vwmulu.vx v12, v8, a0
+; RV32-NEXT:    vmv4r.v v8, v12
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: vwmulu_vx_nxv4i64:
+; RV64:       # %bb.0:
+; RV64-NEXT:    vsetvli a1, zero, e64, m4, ta, mu
+; RV64-NEXT:    vzext.vf2 v12, v8
+; RV64-NEXT:    slli a0, a0, 32
+; RV64-NEXT:    srli a0, a0, 32
+; RV64-NEXT:    vmul.vx v8, v12, a0
+; RV64-NEXT:    ret
   %head = insertelement <vscale x 4 x i32> undef, i32 %b, i32 0
   %splat = shufflevector <vscale x 4 x i32> %head, <vscale x 4 x i32> undef, <vscale x 4 x i32> zeroinitializer
   %vc = zext <vscale x 4 x i32> %va to <vscale x 4 x i64>
@@ -240,12 +309,21 @@
 }
 
 define <vscale x 4 x i64> @vwmulsu_vx_nxv4i64(<vscale x 4 x i32> %va, i32 %b) {
-; CHECK-LABEL: vwmulsu_vx_nxv4i64:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    vsetvli a1, zero, e32, m2, ta, mu
-; CHECK-NEXT:    vwmulsu.vx v12, v8, a0
-; CHECK-NEXT:    vmv4r.v v8, v12
-; CHECK-NEXT:    ret
+; RV32-LABEL: vwmulsu_vx_nxv4i64:
+; RV32:       # %bb.0:
+; RV32-NEXT:    vsetvli a1, zero, e32, m2, ta, mu
+; RV32-NEXT:    vwmulsu.vx v12, v8, a0
+; RV32-NEXT:    vmv4r.v v8, v12
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: vwmulsu_vx_nxv4i64:
+; RV64:       # %bb.0:
+; RV64-NEXT:    vsetvli a1, zero, e64, m4, ta, mu
+; RV64-NEXT:    vsext.vf2 v12, v8
+; RV64-NEXT:    slli a0, a0, 32
+; RV64-NEXT:    srli a0, a0, 32
+; RV64-NEXT:    vmul.vx v8, v12, a0
+; RV64-NEXT:    ret
   %head = insertelement <vscale x 4 x i32> undef, i32 %b, i32 0
   %splat = shufflevector <vscale x 4 x i32> %head, <vscale x 4 x i32> undef, <vscale x 4 x i32> zeroinitializer
   %vc = sext <vscale x 4 x i32> %va to <vscale x 4 x i64>
@@ -294,12 +372,20 @@
 }
 
 define <vscale x 8 x i64> @vwmul_vx_nxv8i64(<vscale x 8 x i32> %va, i32 %b) {
-; CHECK-LABEL: vwmul_vx_nxv8i64:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    vsetvli a1, zero, e32, m4, ta, mu
-; CHECK-NEXT:    vwmul.vx v16, v8, a0
-; CHECK-NEXT:    vmv8r.v v8, v16
-; CHECK-NEXT:    ret
+; RV32-LABEL: vwmul_vx_nxv8i64:
+; RV32:       # %bb.0:
+; RV32-NEXT:    vsetvli a1, zero, e32, m4, ta, mu
+; RV32-NEXT:    vwmul.vx v16, v8, a0
+; RV32-NEXT:    vmv8r.v v8, v16
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: vwmul_vx_nxv8i64:
+; RV64:       # %bb.0:
+; RV64-NEXT:    vsetvli a1, zero, e64, m8, ta, mu
+; RV64-NEXT:    vsext.vf2 v16, v8
+; RV64-NEXT:    sext.w a0, a0
+; RV64-NEXT:    vmul.vx v8, v16, a0
+; RV64-NEXT:    ret
   %head = insertelement <vscale x 8 x i32> undef, i32 %b, i32 0
   %splat = shufflevector <vscale x 8 x i32> %head, <vscale x 8 x i32> undef, <vscale x 8 x i32> zeroinitializer
   %vc = sext <vscale x 8 x i32> %va to <vscale x 8 x i64>
@@ -309,12 +395,21 @@
 }
 
 define <vscale x 8 x i64> @vwmulu_vx_nxv8i64(<vscale x 8 x i32> %va, i32 %b) {
-; CHECK-LABEL: vwmulu_vx_nxv8i64:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    vsetvli a1, zero, e32, m4, ta, mu
-; CHECK-NEXT:    vwmulu.vx v16, v8, a0
-; CHECK-NEXT:    vmv8r.v v8, v16
-; CHECK-NEXT:    ret
+; RV32-LABEL: vwmulu_vx_nxv8i64:
+; RV32:       # %bb.0:
+; RV32-NEXT:    vsetvli a1, zero, e32, m4, ta, mu
+; RV32-NEXT:    vwmulu.vx v16, v8, a0
+; RV32-NEXT:    vmv8r.v v8, v16
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: vwmulu_vx_nxv8i64:
+; RV64:       # %bb.0:
+; RV64-NEXT:    vsetvli a1, zero, e64, m8, ta, mu
+; RV64-NEXT:    vzext.vf2 v16, v8
+; RV64-NEXT:    slli a0, a0, 32
+; RV64-NEXT:    srli a0, a0, 32
+; RV64-NEXT:    vmul.vx v8, v16, a0
+; RV64-NEXT:    ret
   %head = insertelement <vscale x 8 x i32> undef, i32 %b, i32 0
   %splat = shufflevector <vscale x 8 x i32> %head, <vscale x 8 x i32> undef, <vscale x 8 x i32> zeroinitializer
   %vc = zext <vscale x 8 x i32> %va to <vscale x 8 x i64>
@@ -324,12 +419,21 @@
 }
 
 define <vscale x 8 x i64> @vwmulsu_vx_nxv8i64(<vscale x 8 x i32> %va, i32 %b) {
-; CHECK-LABEL: vwmulsu_vx_nxv8i64:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    vsetvli a1, zero, e32, m4, ta, mu
-; CHECK-NEXT:    vwmulsu.vx v16, v8, a0
-; CHECK-NEXT:    vmv8r.v v8, v16
-; CHECK-NEXT:    ret
+; RV32-LABEL: vwmulsu_vx_nxv8i64:
+; RV32:       # %bb.0:
+; RV32-NEXT:    vsetvli a1, zero, e32, m4, ta, mu
+; RV32-NEXT:    vwmulsu.vx v16, v8, a0
+; RV32-NEXT:    vmv8r.v v8, v16
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: vwmulsu_vx_nxv8i64:
+; RV64:       # %bb.0:
+; RV64-NEXT:    vsetvli a1, zero, e64, m8, ta, mu
+; RV64-NEXT:    vsext.vf2 v16, v8
+; RV64-NEXT:    slli a0, a0, 32
+; RV64-NEXT:    srli a0, a0, 32
+; RV64-NEXT:    vmul.vx v8, v16, a0
+; RV64-NEXT:    ret
   %head = insertelement <vscale x 8 x i32> undef, i32 %b, i32 0
   %splat = shufflevector <vscale x 8 x i32> %head, <vscale x 8 x i32> undef, <vscale x 8 x i32> zeroinitializer
   %vc = sext <vscale x 8 x i32> %va to <vscale x 8 x i64>
diff --git a/llvm/test/CodeGen/RISCV/rvv/vwsub-sdnode.ll b/llvm/test/CodeGen/RISCV/rvv/vwsub-sdnode.ll
--- a/llvm/test/CodeGen/RISCV/rvv/vwsub-sdnode.ll
+++ b/llvm/test/CodeGen/RISCV/rvv/vwsub-sdnode.ll
@@ -1,6 +1,6 @@
 ; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
-; RUN: llc -mtriple=riscv32 -mattr=+v -verify-machineinstrs < %s | FileCheck %s
-; RUN: llc -mtriple=riscv64 -mattr=+v -verify-machineinstrs < %s | FileCheck %s
+; RUN: llc -mtriple=riscv32 -mattr=+v -verify-machineinstrs < %s | FileCheck %s --check-prefixes=CHECK,CHECK32
+; RUN: llc -mtriple=riscv64 -mattr=+v -verify-machineinstrs < %s | FileCheck %s --check-prefixes=CHECK,CHECK64
 
 define <vscale x 1 x i64> @vwsub_vv_nxv1i64(<vscale x 1 x i32> %va, <vscale x 1 x i32> %vb) {
 ; CHECK-LABEL: vwsub_vv_nxv1i64:
@@ -29,12 +29,20 @@
 }
 
 define <vscale x 1 x i64> @vwsub_vx_nxv1i64(<vscale x 1 x i32> %va, i32 %b) {
-; CHECK-LABEL: vwsub_vx_nxv1i64:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    vsetvli a1, zero, e32, mf2, ta, mu
-; CHECK-NEXT:    vwsub.vx v9, v8, a0
-; CHECK-NEXT:    vmv1r.v v8, v9
-; CHECK-NEXT:    ret
+; CHECK32-LABEL: vwsub_vx_nxv1i64:
+; CHECK32:       # %bb.0:
+; CHECK32-NEXT:    vsetvli a1, zero, e32, mf2, ta, mu
+; CHECK32-NEXT:    vwsub.vx v9, v8, a0
+; CHECK32-NEXT:    vmv1r.v v8, v9
+; CHECK32-NEXT:    ret
+;
+; CHECK64-LABEL: vwsub_vx_nxv1i64:
+; CHECK64:       # %bb.0:
+; CHECK64-NEXT:    vsetvli a1, zero, e64, m1, ta, mu
+; CHECK64-NEXT:    vsext.vf2 v9, v8
+; CHECK64-NEXT:    sext.w a0, a0
+; CHECK64-NEXT:    vsub.vx v8, v9, a0
+; CHECK64-NEXT:    ret
   %head = insertelement <vscale x 1 x i32> poison, i32 %b, i32 0
   %splat = shufflevector <vscale x 1 x i32> %head, <vscale x 1 x i32> poison, <vscale x 1 x i32> zeroinitializer
   %vc = sext <vscale x 1 x i32> %va to <vscale x 1 x i64>
@@ -44,12 +52,21 @@
 }
 
 define <vscale x 1 x i64> @vwsubu_vx_nxv1i64(<vscale x 1 x i32> %va, i32 %b) {
-; CHECK-LABEL: vwsubu_vx_nxv1i64:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    vsetvli a1, zero, e32, mf2, ta, mu
-; CHECK-NEXT:    vwsubu.vx v9, v8, a0
-; CHECK-NEXT:    vmv1r.v v8, v9
-; CHECK-NEXT:    ret
+; CHECK32-LABEL: vwsubu_vx_nxv1i64:
+; CHECK32:       # %bb.0:
+; CHECK32-NEXT:    vsetvli a1, zero, e32, mf2, ta, mu
+; CHECK32-NEXT:    vwsubu.vx v9, v8, a0
+; CHECK32-NEXT:    vmv1r.v v8, v9
+; CHECK32-NEXT:    ret
+;
+; CHECK64-LABEL: vwsubu_vx_nxv1i64:
+; CHECK64:       # %bb.0:
+; CHECK64-NEXT:    vsetvli a1, zero, e64, m1, ta, mu
+; CHECK64-NEXT:    vzext.vf2 v9, v8
+; CHECK64-NEXT:    slli a0, a0, 32
+; CHECK64-NEXT:    srli a0, a0, 32
+; CHECK64-NEXT:    vsub.vx v8, v9, a0
+; CHECK64-NEXT:    ret
   %head = insertelement <vscale x 1 x i32> poison, i32 %b, i32 0
   %splat = shufflevector <vscale x 1 x i32> %head, <vscale x 1 x i32> poison, <vscale x 1 x i32> zeroinitializer
   %vc = zext <vscale x 1 x i32> %va to <vscale x 1 x i64>
@@ -83,11 +100,18 @@
 }
 
 define <vscale x 1 x i64> @vwsub_wx_nxv1i64(<vscale x 1 x i64> %va, i32 %b) {
-; CHECK-LABEL: vwsub_wx_nxv1i64:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    vsetvli a1, zero, e32, mf2, ta, mu
-; CHECK-NEXT:    vwsub.wx v8, v8, a0
-; CHECK-NEXT:    ret
+; CHECK32-LABEL: vwsub_wx_nxv1i64:
+; CHECK32:       # %bb.0:
+; CHECK32-NEXT:    vsetvli a1, zero, e32, mf2, ta, mu
+; CHECK32-NEXT:    vwsub.wx v8, v8, a0
+; CHECK32-NEXT:    ret
+;
+; CHECK64-LABEL: vwsub_wx_nxv1i64:
+; CHECK64:       # %bb.0:
+; CHECK64-NEXT:    sext.w a0, a0
+; CHECK64-NEXT:    vsetvli a1, zero, e64, m1, ta, mu
+; CHECK64-NEXT:    vsub.vx v8, v8, a0
+; CHECK64-NEXT:    ret
   %head = insertelement <vscale x 1 x i32> poison, i32 %b, i32 0
   %splat = shufflevector <vscale x 1 x i32> %head, <vscale x 1 x i32> poison, <vscale x 1 x i32> zeroinitializer
   %vb = sext <vscale x 1 x i32> %splat to <vscale x 1 x i64>
@@ -96,11 +120,19 @@
 }
 
 define <vscale x 1 x i64> @vwsubu_wx_nxv1i64(<vscale x 1 x i64> %va, i32 %b) {
-; CHECK-LABEL: vwsubu_wx_nxv1i64:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    vsetvli a1, zero, e32, mf2, ta, mu
-; CHECK-NEXT:    vwsubu.wx v8, v8, a0
-; CHECK-NEXT:    ret
+; CHECK32-LABEL: vwsubu_wx_nxv1i64:
+; CHECK32:       # %bb.0:
+; CHECK32-NEXT:    vsetvli a1, zero, e32, mf2, ta, mu
+; CHECK32-NEXT:    vwsubu.wx v8, v8, a0
+; CHECK32-NEXT:    ret
+;
+; CHECK64-LABEL: vwsubu_wx_nxv1i64:
+; CHECK64:       # %bb.0:
+; CHECK64-NEXT:    slli a0, a0, 32
+; CHECK64-NEXT:    srli a0, a0, 32
+; CHECK64-NEXT:    vsetvli a1, zero, e64, m1, ta, mu
+; CHECK64-NEXT:    vsub.vx v8, v8, a0
+; CHECK64-NEXT:    ret
   %head = insertelement <vscale x 1 x i32> poison, i32 %b, i32 0
   %splat = shufflevector <vscale x 1 x i32> %head, <vscale x 1 x i32> poison, <vscale x 1 x i32> zeroinitializer
   %vb = zext <vscale x 1 x i32> %splat to <vscale x 1 x i64>
@@ -135,12 +167,20 @@
 }
 
 define <vscale x 2 x i64> @vwsub_vx_nxv2i64(<vscale x 2 x i32> %va, i32 %b) {
-; CHECK-LABEL: vwsub_vx_nxv2i64:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    vsetvli a1, zero, e32, m1, ta, mu
-; CHECK-NEXT:    vwsub.vx v10, v8, a0
-; CHECK-NEXT:    vmv2r.v v8, v10
-; CHECK-NEXT:    ret
+; CHECK32-LABEL: vwsub_vx_nxv2i64:
+; CHECK32:       # %bb.0:
+; CHECK32-NEXT:    vsetvli a1, zero, e32, m1, ta, mu
+; CHECK32-NEXT:    vwsub.vx v10, v8, a0
+; CHECK32-NEXT:    vmv2r.v v8, v10
+; CHECK32-NEXT:    ret
+;
+; CHECK64-LABEL: vwsub_vx_nxv2i64:
+; CHECK64:       # %bb.0:
+; CHECK64-NEXT:    vsetvli a1, zero, e64, m2, ta, mu
+; CHECK64-NEXT:    vsext.vf2 v10, v8
+; CHECK64-NEXT:    sext.w a0, a0
+; CHECK64-NEXT:    vsub.vx v8, v10, a0
+; CHECK64-NEXT:    ret
   %head = insertelement <vscale x 2 x i32> poison, i32 %b, i32 0
   %splat = shufflevector <vscale x 2 x i32> %head, <vscale x 2 x i32> poison, <vscale x 2 x i32> zeroinitializer
   %vc = sext <vscale x 2 x i32> %va to <vscale x 2 x i64>
@@ -150,12 +190,21 @@
 }
 
 define <vscale x 2 x i64> @vwsubu_vx_nxv2i64(<vscale x 2 x i32> %va, i32 %b) {
-; CHECK-LABEL: vwsubu_vx_nxv2i64:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    vsetvli a1, zero, e32, m1, ta, mu
-; CHECK-NEXT:    vwsubu.vx v10, v8, a0
-; CHECK-NEXT:    vmv2r.v v8, v10
-; CHECK-NEXT:    ret
+; CHECK32-LABEL: vwsubu_vx_nxv2i64:
+; CHECK32:       # %bb.0:
+; CHECK32-NEXT:    vsetvli a1, zero, e32, m1, ta, mu
+; CHECK32-NEXT:    vwsubu.vx v10, v8, a0
+; CHECK32-NEXT:    vmv2r.v v8, v10
+; CHECK32-NEXT:    ret
+;
+; CHECK64-LABEL: vwsubu_vx_nxv2i64:
+; CHECK64:       # %bb.0:
+; CHECK64-NEXT:    vsetvli a1, zero, e64, m2, ta, mu
+; CHECK64-NEXT:    vzext.vf2 v10, v8
+; CHECK64-NEXT:    slli a0, a0, 32
+; CHECK64-NEXT:    srli a0, a0, 32
+; CHECK64-NEXT:    vsub.vx v8, v10, a0
+; CHECK64-NEXT:    ret
   %head = insertelement <vscale x 2 x i32> poison, i32 %b, i32 0
   %splat = shufflevector <vscale x 2 x i32> %head, <vscale x 2 x i32> poison, <vscale x 2 x i32> zeroinitializer
   %vc = zext <vscale x 2 x i32> %va to <vscale x 2 x i64>
@@ -189,11 +238,18 @@
 }
 
 define <vscale x 2 x i64> @vwsub_wx_nxv2i64(<vscale x 2 x i64> %va, i32 %b) {
-; CHECK-LABEL: vwsub_wx_nxv2i64:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    vsetvli a1, zero, e32, m1, ta, mu
-; CHECK-NEXT:    vwsub.wx v8, v8, a0
-; CHECK-NEXT:    ret
+; CHECK32-LABEL: vwsub_wx_nxv2i64:
+; CHECK32:       # %bb.0:
+; CHECK32-NEXT:    vsetvli a1, zero, e32, m1, ta, mu
+; CHECK32-NEXT:    vwsub.wx v8, v8, a0
+; CHECK32-NEXT:    ret
+;
+; CHECK64-LABEL: vwsub_wx_nxv2i64:
+; CHECK64:       # %bb.0:
+; CHECK64-NEXT:    sext.w a0, a0
+; CHECK64-NEXT:    vsetvli a1, zero, e64, m2, ta, mu
+; CHECK64-NEXT:    vsub.vx v8, v8, a0
+; CHECK64-NEXT:    ret
   %head = insertelement <vscale x 2 x i32> poison, i32 %b, i32 0
   %splat = shufflevector <vscale x 2 x i32> %head, <vscale x 2 x i32> poison, <vscale x 2 x i32> zeroinitializer
   %vb = sext <vscale x 2 x i32> %splat to <vscale x 2 x i64>
@@ -202,11 +258,19 @@
 }
 
 define <vscale x 2 x i64> @vwsubu_wx_nxv2i64(<vscale x 2 x i64> %va, i32 %b) {
-; CHECK-LABEL: vwsubu_wx_nxv2i64:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    vsetvli a1, zero, e32, m1, ta, mu
-; CHECK-NEXT:    vwsubu.wx v8, v8, a0
-; CHECK-NEXT:    ret
+; CHECK32-LABEL: vwsubu_wx_nxv2i64:
+; CHECK32:       # %bb.0:
+; CHECK32-NEXT:    vsetvli a1, zero, e32, m1, ta, mu
+; CHECK32-NEXT:    vwsubu.wx v8, v8, a0
+; CHECK32-NEXT:    ret
+;
+; CHECK64-LABEL: vwsubu_wx_nxv2i64:
+; CHECK64:       # %bb.0:
+; CHECK64-NEXT:    slli a0, a0, 32
+; CHECK64-NEXT:    srli a0, a0, 32
+; CHECK64-NEXT:    vsetvli a1, zero, e64, m2, ta, mu
+; CHECK64-NEXT:    vsub.vx v8, v8, a0
+; CHECK64-NEXT:    ret
   %head = insertelement <vscale x 2 x i32> poison, i32 %b, i32 0
   %splat = shufflevector <vscale x 2 x i32> %head, <vscale x 2 x i32> poison, <vscale x 2 x i32> zeroinitializer
   %vb = zext <vscale x 2 x i32> %splat to <vscale x 2 x i64>
@@ -241,12 +305,20 @@
 }
 
 define <vscale x 4 x i64> @vwsub_vx_nxv4i64(<vscale x 4 x i32> %va, i32 %b) {
-; CHECK-LABEL: vwsub_vx_nxv4i64:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    vsetvli a1, zero, e32, m2, ta, mu
-; CHECK-NEXT:    vwsub.vx v12, v8, a0
-; CHECK-NEXT:    vmv4r.v v8, v12
-; CHECK-NEXT:    ret
+; CHECK32-LABEL: vwsub_vx_nxv4i64:
+; CHECK32:       # %bb.0:
+; CHECK32-NEXT:    vsetvli a1, zero, e32, m2, ta, mu
+; CHECK32-NEXT:    vwsub.vx v12, v8, a0
+; CHECK32-NEXT:    vmv4r.v v8, v12
+; CHECK32-NEXT:    ret
+;
+; CHECK64-LABEL: vwsub_vx_nxv4i64:
+; CHECK64:       # %bb.0:
+; CHECK64-NEXT:    vsetvli a1, zero, e64, m4, ta, mu
+; CHECK64-NEXT:    vsext.vf2 v12, v8
+; CHECK64-NEXT:    sext.w a0, a0
+; CHECK64-NEXT:    vsub.vx v8, v12, a0
+; CHECK64-NEXT:    ret
   %head = insertelement <vscale x 4 x i32> poison, i32 %b, i32 0
   %splat = shufflevector <vscale x 4 x i32> %head, <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer
   %vc = sext <vscale x 4 x i32> %va to <vscale x 4 x i64>
@@ -256,12 +328,21 @@
 }
 
 define <vscale x 4 x i64> @vwsubu_vx_nxv4i64(<vscale x 4 x i32> %va, i32 %b) {
-; CHECK-LABEL: vwsubu_vx_nxv4i64:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    vsetvli a1, zero, e32, m2, ta, mu
-; CHECK-NEXT:    vwsubu.vx v12, v8, a0
-; CHECK-NEXT:    vmv4r.v v8, v12
-; CHECK-NEXT:    ret
+; CHECK32-LABEL: vwsubu_vx_nxv4i64:
+; CHECK32:       # %bb.0:
+; CHECK32-NEXT:    vsetvli a1, zero, e32, m2, ta, mu
+; CHECK32-NEXT:    vwsubu.vx v12, v8, a0
+; CHECK32-NEXT:    vmv4r.v v8, v12
+; CHECK32-NEXT:    ret
+;
+; CHECK64-LABEL: vwsubu_vx_nxv4i64:
+; CHECK64:       # %bb.0:
+; CHECK64-NEXT:    vsetvli a1, zero, e64, m4, ta, mu
+; CHECK64-NEXT:    vzext.vf2 v12, v8
+; CHECK64-NEXT:    slli a0, a0, 32
+; CHECK64-NEXT:    srli a0, a0, 32
+; CHECK64-NEXT:    vsub.vx v8, v12, a0
+; CHECK64-NEXT:    ret
   %head = insertelement <vscale x 4 x i32> poison, i32 %b, i32 0
   %splat = shufflevector <vscale x 4 x i32> %head, <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer
   %vc = zext <vscale x 4 x i32> %va to <vscale x 4 x i64>
@@ -295,11 +376,18 @@
 }
 
 define <vscale x 4 x i64> @vwsub_wx_nxv4i64(<vscale x 4 x i64> %va, i32 %b) {
-; CHECK-LABEL: vwsub_wx_nxv4i64:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    vsetvli a1, zero, e32, m2, ta, mu
-; CHECK-NEXT:    vwsub.wx v8, v8, a0
-; CHECK-NEXT:    ret
+; CHECK32-LABEL: vwsub_wx_nxv4i64:
+; CHECK32:       # %bb.0:
+; CHECK32-NEXT:    vsetvli a1, zero, e32, m2, ta, mu
+; CHECK32-NEXT:    vwsub.wx v8, v8, a0
+; CHECK32-NEXT:    ret
+;
+; CHECK64-LABEL: vwsub_wx_nxv4i64:
+; CHECK64:       # %bb.0:
+; CHECK64-NEXT:    sext.w a0, a0
+; CHECK64-NEXT:    vsetvli a1, zero, e64, m4, ta, mu
+; CHECK64-NEXT:    vsub.vx v8, v8, a0
+; CHECK64-NEXT:    ret
   %head = insertelement <vscale x 4 x i32> poison, i32 %b, i32 0
   %splat = shufflevector <vscale x 4 x i32> %head, <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer
   %vb = sext <vscale x 4 x i32> %splat to <vscale x 4 x i64>
@@ -308,11 +396,19 @@
 }
 
 define <vscale x 4 x i64> @vwsubu_wx_nxv4i64(<vscale x 4 x i64> %va, i32 %b) {
-; CHECK-LABEL: vwsubu_wx_nxv4i64:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    vsetvli a1, zero, e32, m2, ta, mu
-; CHECK-NEXT:    vwsubu.wx v8, v8, a0
-; CHECK-NEXT:    ret
+; CHECK32-LABEL: vwsubu_wx_nxv4i64:
+; CHECK32:       # %bb.0:
+; CHECK32-NEXT:    vsetvli a1, zero, e32, m2, ta, mu
+; CHECK32-NEXT:    vwsubu.wx v8, v8, a0
+; CHECK32-NEXT:    ret
+;
+; CHECK64-LABEL: vwsubu_wx_nxv4i64:
+; CHECK64:       # %bb.0:
+; CHECK64-NEXT:    slli a0, a0, 32
+; CHECK64-NEXT:    srli a0, a0, 32
+; CHECK64-NEXT:    vsetvli a1, zero, e64, m4, ta, mu
+; CHECK64-NEXT:    vsub.vx v8, v8, a0
+; CHECK64-NEXT:    ret
   %head = insertelement <vscale x 4 x i32> poison, i32 %b, i32 0
   %splat = shufflevector <vscale x 4 x i32> %head, <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer
   %vb = zext <vscale x 4 x i32> %splat to <vscale x 4 x i64>
@@ -347,12 +443,20 @@
 }
 
 define <vscale x 8 x i64> @vwsub_vx_nxv8i64(<vscale x 8 x i32> %va, i32 %b) {
-; CHECK-LABEL: vwsub_vx_nxv8i64:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    vsetvli a1, zero, e32, m4, ta, mu
-; CHECK-NEXT:    vwsub.vx v16, v8, a0
-; CHECK-NEXT:    vmv8r.v v8, v16
-; CHECK-NEXT:    ret
+; CHECK32-LABEL: vwsub_vx_nxv8i64:
+; CHECK32:       # %bb.0:
+; CHECK32-NEXT:    vsetvli a1, zero, e32, m4, ta, mu
+; CHECK32-NEXT:    vwsub.vx v16, v8, a0
+; CHECK32-NEXT:    vmv8r.v v8, v16
+; CHECK32-NEXT:    ret
+;
+; CHECK64-LABEL: vwsub_vx_nxv8i64:
+; CHECK64:       # %bb.0:
+; CHECK64-NEXT:    vsetvli a1, zero, e64, m8, ta, mu
+; CHECK64-NEXT:    vsext.vf2 v16, v8
+; CHECK64-NEXT:    sext.w a0, a0
+; CHECK64-NEXT:    vsub.vx v8, v16, a0
+; CHECK64-NEXT:    ret
   %head = insertelement <vscale x 8 x i32> poison, i32 %b, i32 0
   %splat = shufflevector <vscale x 8 x i32> %head, <vscale x 8 x i32> poison, <vscale x 8 x i32> zeroinitializer
   %vc = sext <vscale x 8 x i32> %va to <vscale x 8 x i64>
@@ -362,12 +466,21 @@
 }
 
 define <vscale x 8 x i64> @vwsubu_vx_nxv8i64(<vscale x 8 x i32> %va, i32 %b) {
-; CHECK-LABEL: vwsubu_vx_nxv8i64:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    vsetvli a1, zero, e32, m4, ta, mu
-; CHECK-NEXT:    vwsubu.vx v16, v8, a0
-; CHECK-NEXT:    vmv8r.v v8, v16
-; CHECK-NEXT:    ret
+; CHECK32-LABEL: vwsubu_vx_nxv8i64:
+; CHECK32:       # %bb.0:
+; CHECK32-NEXT:    vsetvli a1, zero, e32, m4, ta, mu
+; CHECK32-NEXT:    vwsubu.vx v16, v8, a0
+; CHECK32-NEXT:    vmv8r.v v8, v16
+; CHECK32-NEXT:    ret
+;
+; CHECK64-LABEL: vwsubu_vx_nxv8i64:
+; CHECK64:       # %bb.0:
+; CHECK64-NEXT:    vsetvli a1, zero, e64, m8, ta, mu
+; CHECK64-NEXT:    vzext.vf2 v16, v8
+; CHECK64-NEXT:    slli a0, a0, 32
+; CHECK64-NEXT:    srli a0, a0, 32
+; CHECK64-NEXT:    vsub.vx v8, v16, a0
+; CHECK64-NEXT:    ret
   %head = insertelement <vscale x 8 x i32> poison, i32 %b, i32 0
   %splat = shufflevector <vscale x 8 x i32> %head, <vscale x 8 x i32> poison, <vscale x 8 x i32> zeroinitializer
   %vc = zext <vscale x 8 x i32> %va to <vscale x 8 x i64>
@@ -401,11 +514,18 @@
 }
 
 define <vscale x 8 x i64> @vwsub_wx_nxv8i64(<vscale x 8 x i64> %va, i32 %b) {
-; CHECK-LABEL: vwsub_wx_nxv8i64:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    vsetvli a1, zero, e32, m4, ta, mu
-; CHECK-NEXT:    vwsub.wx v8, v8, a0
-; CHECK-NEXT:    ret
+; CHECK32-LABEL: vwsub_wx_nxv8i64:
+; CHECK32:       # %bb.0:
+; CHECK32-NEXT:    vsetvli a1, zero, e32, m4, ta, mu
+; CHECK32-NEXT:    vwsub.wx v8, v8, a0
+; CHECK32-NEXT:    ret
+;
+; CHECK64-LABEL: vwsub_wx_nxv8i64:
+; CHECK64:       # %bb.0:
+; CHECK64-NEXT:    sext.w a0, a0
+; CHECK64-NEXT:    vsetvli a1, zero, e64, m8, ta, mu
+; CHECK64-NEXT:    vsub.vx v8, v8, a0
+; CHECK64-NEXT:    ret
   %head = insertelement <vscale x 8 x i32> poison, i32 %b, i32 0
   %splat = shufflevector <vscale x 8 x i32> %head, <vscale x 8 x i32> poison, <vscale x 8 x i32> zeroinitializer
   %vb = sext <vscale x 8 x i32> %splat to <vscale x 8 x i64>
@@ -414,11 +534,19 @@
 }
 
 define <vscale x 8 x i64> @vwsubu_wx_nxv8i64(<vscale x 8 x i64> %va, i32 %b) {
-; CHECK-LABEL: vwsubu_wx_nxv8i64:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    vsetvli a1, zero, e32, m4, ta, mu
-; CHECK-NEXT:    vwsubu.wx v8, v8, a0
-; CHECK-NEXT:    ret
+; CHECK32-LABEL: vwsubu_wx_nxv8i64:
+; CHECK32:       # %bb.0:
+; CHECK32-NEXT:    vsetvli a1, zero, e32, m4, ta, mu
+; CHECK32-NEXT:    vwsubu.wx v8, v8, a0
+; CHECK32-NEXT:    ret
+;
+; CHECK64-LABEL: vwsubu_wx_nxv8i64:
+; CHECK64:       # %bb.0:
+; CHECK64-NEXT:    slli a0, a0, 32
+; CHECK64-NEXT:    srli a0, a0, 32
+; CHECK64-NEXT:    vsetvli a1, zero, e64, m8, ta, mu
+; CHECK64-NEXT:    vsub.vx v8, v8, a0
+; CHECK64-NEXT:    ret
   %head = insertelement <vscale x 8 x i32> poison, i32 %b, i32 0
   %splat = shufflevector <vscale x 8 x i32> %head, <vscale x 8 x i32> poison, <vscale x 8 x i32> zeroinitializer
   %vb = zext <vscale x 8 x i32> %splat to <vscale x 8 x i64>