[RISCV] Add intrinsics for vwmacc[u|su|us] instructions

[arcbbb]

Summary

This patch defines vwmacc[u|su|us] intrinsics and lower to V instructions.

We work with @rogfer01 from BSC to come out this patch.

Authored-by: Roger Ferrer Ibanez <[e-mail]>
Co-Authored-by: ShihPo Hung <[e-mail]>

[craig.topper]

LGTM

llvm/include/llvm/IR/IntrinsicsRISCV.td

351

Drop the space between llvm_anyint_ty and ']'

llvm/include/llvm/IR/IntrinsicsRISCV.td

let TargetPrefix = "riscv" in {
   }
   class RISCVTernaryAAXAMask
         : Intrinsic<[llvm_anyvector_ty],
                     [LLVMMatchType<0>, llvm_any_ty, LLVMMatchType<0>,
                      LLVMScalarOrSameVectorWidth<0, llvm_i1_ty>, llvm_anyint_ty],
                     [IntrNoMem]>, RISCVVIntrinsic {
     let ExtendOperand = 2;
   }
+  class RISCVTernaryWideNoMask
+        : Intrinsic< [llvm_anyvector_ty],
+                     [LLVMMatchType<0>, llvm_any_ty, llvm_anyvector_ty,
+                      llvm_anyint_ty],
+                     [IntrNoMem] >, RISCVVIntrinsic {
+    let ExtendOperand = 2;
+  }
+  class RISCVTernaryWideMask
+        : Intrinsic< [llvm_anyvector_ty],
+                     [LLVMMatchType<0>, llvm_any_ty, llvm_anyvector_ty,
+                      LLVMScalarOrSameVectorWidth<0, llvm_i1_ty>, llvm_anyint_ty],
+                     [IntrNoMem]>, RISCVVIntrinsic {
+    let ExtendOperand = 2;
+  }
 
   multiclass RISCVUSLoad {
     def "int_riscv_" # NAME : RISCVUSLoad;
     def "int_riscv_" # NAME # "_mask" : RISCVUSLoadMask;
   }
   multiclass RISCVSLoad {
     def "int_riscv_" # NAME : RISCVSLoad;
     def "int_riscv_" # NAME # "_mask" : RISCVSLoadMask;
let TargetPrefix = "riscv" in {
   multiclass RISCVTernaryAAXA {
     def "int_riscv_" # NAME : RISCVTernaryAAXANoMask;
     def "int_riscv_" # NAME # "_mask" : RISCVTernaryAAXAMask;
   }
   multiclass RISCVCompare {
     def "int_riscv_" # NAME : RISCVCompareNoMask;
     def "int_riscv_" # NAME # "_mask" : RISCVCompareMask;
   }
+  multiclass RISCVTernaryWide {
+    def "int_riscv_" # NAME : RISCVTernaryWideNoMask;
+    def "int_riscv_" # NAME # "_mask" : RISCVTernaryWideMask;
+  }
 
   defm vle : RISCVUSLoad;
   defm vleff : RISCVUSLoad;
   defm vse : RISCVUSStore;
   defm vlse: RISCVSLoad;
   defm vsse: RISCVSStore;
   defm vlxe: RISCVILoad;
   defm vsxe: RISCVIStore;
let TargetPrefix = "riscv" in {
   defm vwmulu : RISCVBinaryABX;
   defm vwmulsu : RISCVBinaryABX;
 
   defm vmacc : RISCVTernaryAAXA;
   defm vnmsac : RISCVTernaryAAXA;
   defm vmadd : RISCVTernaryAAXA;
   defm vnmsub : RISCVTernaryAAXA;
 
+  defm vwmaccu  : RISCVTernaryWide;
+  defm vwmacc   : RISCVTernaryWide;
+  defm vwmaccus : RISCVTernaryWide;
+  defm vwmaccsu : RISCVTernaryWide;
+
   defm vfadd : RISCVBinaryAAX;
   defm vfsub : RISCVBinaryAAX;
   defm vfrsub : RISCVBinaryAAX;
 
   defm vfwadd : RISCVBinaryABX;
   defm vfwsub : RISCVBinaryABX;
   defm vfwadd_w : RISCVBinaryAAX;
   defm vfwsub_w : RISCVBinaryAAX;

llvm/lib/Target/RISCV/RISCVInstrInfoVPseudos.td

foreach m = MxList.m in
     defm _VX : VPseudoTernary<m.vrclass, m.vrclass, GPR, m, Constraint>;
 }
 
 multiclass VPseudoTernaryV_VX_AAXA<string Constraint = ""> {
   foreach m = MxList.m in
     defm _VX : VPseudoTernary<m.vrclass, GPR, m.vrclass, m, Constraint>;
 }
 
+multiclass VPseudoTernaryW_VV {
+  defvar constraint = "@earlyclobber $rd";
+  foreach m = MxList.m in
+    defm _VV : VPseudoTernary<m.wvrclass, m.vrclass, m.vrclass, m, constraint>;
+}
+
+multiclass VPseudoTernaryW_VX {
+  defvar constraint = "@earlyclobber $rd";
+  foreach m = MxList.m in
+    defm _VX : VPseudoTernary<m.wvrclass, GPR, m.vrclass, m, constraint>;
+}
+
 multiclass VPseudoTernaryV_VI<Operand ImmType = simm5, string Constraint = ""> {
   foreach m = MxList.m in
     defm _VI : VPseudoTernary<m.vrclass, m.vrclass, ImmType, m, Constraint>;
 }
 
 multiclass VPseudoTernaryV_VV_VX_AAXA<string Constraint = ""> {
   defm "" : VPseudoTernaryV_VV<Constraint>;
   defm "" : VPseudoTernaryV_VX_AAXA<Constraint>;
 }
 
 multiclass VPseudoTernaryV_VX_VI<Operand ImmType = simm5, string Constraint = ""> {
   defm "" : VPseudoTernaryV_VX<Constraint>;
   defm "" : VPseudoTernaryV_VI<ImmType, Constraint>;
 }
 
+multiclass VPseudoTernaryW_VV_VX {
+  defm "" : VPseudoTernaryW_VV;
+  defm "" : VPseudoTernaryW_VX;
+}
+
 multiclass VPseudoBinaryM_VV_VX_VI {
   defm "" : VPseudoBinaryM_VV;
   defm "" : VPseudoBinaryM_VX</*IsFloat=*/0>;
   defm "" : VPseudoBinaryM_VI;
 }
 
 multiclass VPseudoBinaryM_VV_VX<bit IsFloat = 0> {
   defm "" : VPseudoBinaryM_VV;
multiclass VPatTernaryV_VI<string intrinsic, string instruction,
                            list<VTypeInfo> vtilist, Operand Imm_type> {
   foreach vti = vtilist in
     defm : VPatTernary<intrinsic, instruction, "VI",
                       vti.Vector, vti.Vector, XLenVT, vti.Mask,
                       vti.SEW, vti.LMul, vti.RegClass,
                       vti.RegClass, Imm_type>;
 }
 
+multiclass VPatTernaryW_VV<string intrinsic, string instruction,
+                           list<VTypeInfoToWide> vtilist> {
+  foreach vtiToWti = vtilist in {
+    defvar vti = vtiToWti.Vti;
+    defvar wti = vtiToWti.Wti;
+    defm : VPatTernary<intrinsic, instruction, "VV",
+                      wti.Vector, vti.Vector, vti.Vector,
+                      vti.Mask, vti.SEW, vti.LMul,
+                      wti.RegClass, vti.RegClass, vti.RegClass>;
+  }
+}
+
+multiclass VPatTernaryW_VX<string intrinsic, string instruction,
+                           list<VTypeInfoToWide> vtilist> {
+  foreach vtiToWti = vtilist in {
+    defvar vti = vtiToWti.Vti;
+    defvar wti = vtiToWti.Wti;
+    defm : VPatTernary<intrinsic, instruction, "VX",
+                      wti.Vector, XLenVT, vti.Vector,
+                      vti.Mask, vti.SEW, vti.LMul,
+                      wti.RegClass, vti.ScalarRegClass, vti.RegClass>;
+  }
+}
+
 multiclass VPatTernaryV_VV_VX_AAXA<string intrinsic, string instruction,
                               list<VTypeInfo> vtilist> {
   defm "" : VPatTernaryV_VV<intrinsic, instruction, vtilist>;
   defm "" : VPatTernaryV_VX_AAXA<intrinsic, instruction, vtilist>;
 }
 
 multiclass VPatTernaryV_VX_VI<string intrinsic, string instruction,
                               list<VTypeInfo> vtilist, Operand Imm_type = simm5> {
   defm "" : VPatTernaryV_VX<intrinsic, instruction, vtilist>;
   defm "" : VPatTernaryV_VI<intrinsic, instruction, vtilist, Imm_type>;
 }
 
 multiclass VPatBinaryM_VV_VX_VI<string intrinsic, string instruction,
                                 list<VTypeInfo> vtilist>
 {
   defm "" : VPatBinaryM_VV<intrinsic, instruction, vtilist>;
   defm "" : VPatBinaryM_VX<intrinsic, instruction, vtilist>;
   defm "" : VPatBinaryM_VI<intrinsic, instruction, vtilist>;
 }
 
+multiclass VPatTernaryW_VV_VX<string intrinsic, string instruction,
+                              list<VTypeInfoToWide> vtilist> {
+  defm "" : VPatTernaryW_VV<intrinsic, instruction, vtilist>;
+  defm "" : VPatTernaryW_VX<intrinsic, instruction, vtilist>;
+}
+
 multiclass VPatBinaryM_VV_VX<string intrinsic, string instruction,
                              list<VTypeInfo> vtilist>
 {
   defm "" : VPatBinaryM_VV<intrinsic, instruction, vtilist>;
   defm "" : VPatBinaryM_VX<intrinsic, instruction, vtilist>;
 }
 
 multiclass VPatBinaryM_VX_VI<string intrinsic, string instruction,

 // 12.13. Vector Single-Width Integer Multiply-Add Instructions
 //===----------------------------------------------------------------------===//
 defm PseudoVMACC       : VPseudoTernaryV_VV_VX_AAXA;
 defm PseudoVNMSAC      : VPseudoTernaryV_VV_VX_AAXA;
 defm PseudoVMADD       : VPseudoTernaryV_VV_VX_AAXA;
 defm PseudoVNMSUB      : VPseudoTernaryV_VV_VX_AAXA;
 
 //===----------------------------------------------------------------------===//
+// 12.14. Vector Widening Integer Multiply-Add Instructions
+//===----------------------------------------------------------------------===//
+defm PseudoVWMACCU     : VPseudoTernaryW_VV_VX;
+defm PseudoVWMACC      : VPseudoTernaryW_VV_VX;
+defm PseudoVWMACCSU    : VPseudoTernaryW_VV_VX;
+defm PseudoVWMACCUS    : VPseudoTernaryW_VX;
+
+//===----------------------------------------------------------------------===//
 // 12.17. Vector Integer Move Instructions
 //===----------------------------------------------------------------------===//
 
 defm PseudoVMV_V       : VPseudoUnaryV_V_X_I_NoDummyMask;
 
 //===----------------------------------------------------------------------===//
 // 13.1. Vector Single-Width Saturating Add and Subtract
 //===----------------------------------------------------------------------===//

 // 12.13. Vector Single-Width Integer Multiply-Add Instructions
 //===----------------------------------------------------------------------===//
 defm "" : VPatTernaryV_VV_VX_AAXA<"int_riscv_vmadd", "PseudoVMADD", AllIntegerVectors>;
 defm "" : VPatTernaryV_VV_VX_AAXA<"int_riscv_vnmsub", "PseudoVNMSUB", AllIntegerVectors>;
 defm "" : VPatTernaryV_VV_VX_AAXA<"int_riscv_vmacc", "PseudoVMACC", AllIntegerVectors>;
 defm "" : VPatTernaryV_VV_VX_AAXA<"int_riscv_vnmsac", "PseudoVNMSAC", AllIntegerVectors>;
 
 //===----------------------------------------------------------------------===//
+// 12.14. Vector Widening Integer Multiply-Add Instructions
+//===----------------------------------------------------------------------===//
+defm "" : VPatTernaryW_VV_VX<"int_riscv_vwmaccu", "PseudoVWMACCU", AllWidenableIntVectors>;
+defm "" : VPatTernaryW_VV_VX<"int_riscv_vwmacc", "PseudoVWMACC", AllWidenableIntVectors>;
+defm "" : VPatTernaryW_VV_VX<"int_riscv_vwmaccsu", "PseudoVWMACCSU", AllWidenableIntVectors>;
+defm "" : VPatTernaryW_VX<"int_riscv_vwmaccus", "PseudoVWMACCUS", AllWidenableIntVectors>;
+
+//===----------------------------------------------------------------------===//
 // 12.17. Vector Integer Move Instructions
 //===----------------------------------------------------------------------===//
 foreach vti = AllVectors in {
   def : Pat<(vti.Vector (int_riscv_vmv_v_v (vti.Vector vti.RegClass:$rs1),
                                            GPR:$vl)),
             (!cast<Instruction>("PseudoVMV_V_V_"#vti.LMul.MX)
              $rs1, (NoX0 GPR:$vl), vti.SEW)>;
 }

llvm/test/CodeGen/RISCV/rvv/vwmacc-rv32.ll

+; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
+; RUN: llc -mtriple=riscv32 -mattr=+experimental-v,+f -verify-machineinstrs \
+; RUN:   --riscv-no-aliases < %s | FileCheck %s
+declare <vscale x 1 x i16> @llvm.riscv.vwmacc.nxv1i16.nxv1i8(
+  <vscale x 1 x i16>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  i32);
+
+define <vscale x 1 x i16>  @intrinsic_vwmacc_vv_nxv1i16_nxv1i8_nxv1i8(<vscale x 1 x i16> %0, <vscale x 1 x i8> %1, <vscale x 1 x i8> %2, i32 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmacc_vv_nxv1i16_nxv1i8_nxv1i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e8,mf8,ta,mu
+; CHECK-NEXT:    vwmacc.vv v16, v17, v18
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i16> @llvm.riscv.vwmacc.nxv1i16.nxv1i8(
+    <vscale x 1 x i16> %0,
+    <vscale x 1 x i8> %1,
+    <vscale x 1 x i8> %2,
+    i32 %3)
+
+  ret <vscale x 1 x i16> %a
+}
+
+declare <vscale x 1 x i16> @llvm.riscv.vwmacc.mask.nxv1i16.nxv1i8(
+  <vscale x 1 x i16>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i1>,
+  i32);
+
+define <vscale x 1 x i16>  @intrinsic_vwmacc_mask_vv_nxv1i16_nxv1i8_nxv1i8(<vscale x 1 x i16> %0, <vscale x 1 x i8> %1, <vscale x 1 x i8> %2, <vscale x 1 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmacc_mask_vv_nxv1i16_nxv1i8_nxv1i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e8,mf8,ta,mu
+; CHECK-NEXT:    vwmacc.vv v16, v17, v18, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i16> @llvm.riscv.vwmacc.mask.nxv1i16.nxv1i8(
+    <vscale x 1 x i16> %0,
+    <vscale x 1 x i8> %1,
+    <vscale x 1 x i8> %2,
+    <vscale x 1 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 1 x i16> %a
+}
+
+declare <vscale x 2 x i16> @llvm.riscv.vwmacc.nxv2i16.nxv2i8(
+  <vscale x 2 x i16>,
+  <vscale x 2 x i8>,
+  <vscale x 2 x i8>,
+  i32);
+
+define <vscale x 2 x i16>  @intrinsic_vwmacc_vv_nxv2i16_nxv2i8_nxv2i8(<vscale x 2 x i16> %0, <vscale x 2 x i8> %1, <vscale x 2 x i8> %2, i32 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmacc_vv_nxv2i16_nxv2i8_nxv2i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e8,mf4,ta,mu
+; CHECK-NEXT:    vwmacc.vv v16, v17, v18
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i16> @llvm.riscv.vwmacc.nxv2i16.nxv2i8(
+    <vscale x 2 x i16> %0,
+    <vscale x 2 x i8> %1,
+    <vscale x 2 x i8> %2,
+    i32 %3)
+
+  ret <vscale x 2 x i16> %a
+}
+
+declare <vscale x 2 x i16> @llvm.riscv.vwmacc.mask.nxv2i16.nxv2i8(
+  <vscale x 2 x i16>,
+  <vscale x 2 x i8>,
+  <vscale x 2 x i8>,
+  <vscale x 2 x i1>,
+  i32);
+
+define <vscale x 2 x i16>  @intrinsic_vwmacc_mask_vv_nxv2i16_nxv2i8_nxv2i8(<vscale x 2 x i16> %0, <vscale x 2 x i8> %1, <vscale x 2 x i8> %2, <vscale x 2 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmacc_mask_vv_nxv2i16_nxv2i8_nxv2i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e8,mf4,ta,mu
+; CHECK-NEXT:    vwmacc.vv v16, v17, v18, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i16> @llvm.riscv.vwmacc.mask.nxv2i16.nxv2i8(
+    <vscale x 2 x i16> %0,
+    <vscale x 2 x i8> %1,
+    <vscale x 2 x i8> %2,
+    <vscale x 2 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 2 x i16> %a
+}
+
+declare <vscale x 4 x i16> @llvm.riscv.vwmacc.nxv4i16.nxv4i8(
+  <vscale x 4 x i16>,
+  <vscale x 4 x i8>,
+  <vscale x 4 x i8>,
+  i32);
+
+define <vscale x 4 x i16>  @intrinsic_vwmacc_vv_nxv4i16_nxv4i8_nxv4i8(<vscale x 4 x i16> %0, <vscale x 4 x i8> %1, <vscale x 4 x i8> %2, i32 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmacc_vv_nxv4i16_nxv4i8_nxv4i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e8,mf2,ta,mu
+; CHECK-NEXT:    vwmacc.vv v16, v17, v18
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i16> @llvm.riscv.vwmacc.nxv4i16.nxv4i8(
+    <vscale x 4 x i16> %0,
+    <vscale x 4 x i8> %1,
+    <vscale x 4 x i8> %2,
+    i32 %3)
+
+  ret <vscale x 4 x i16> %a
+}
+
+declare <vscale x 4 x i16> @llvm.riscv.vwmacc.mask.nxv4i16.nxv4i8(
+  <vscale x 4 x i16>,
+  <vscale x 4 x i8>,
+  <vscale x 4 x i8>,
+  <vscale x 4 x i1>,
+  i32);
+
+define <vscale x 4 x i16>  @intrinsic_vwmacc_mask_vv_nxv4i16_nxv4i8_nxv4i8(<vscale x 4 x i16> %0, <vscale x 4 x i8> %1, <vscale x 4 x i8> %2, <vscale x 4 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmacc_mask_vv_nxv4i16_nxv4i8_nxv4i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e8,mf2,ta,mu
+; CHECK-NEXT:    vwmacc.vv v16, v17, v18, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i16> @llvm.riscv.vwmacc.mask.nxv4i16.nxv4i8(
+    <vscale x 4 x i16> %0,
+    <vscale x 4 x i8> %1,
+    <vscale x 4 x i8> %2,
+    <vscale x 4 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 4 x i16> %a
+}
+
+declare <vscale x 8 x i16> @llvm.riscv.vwmacc.nxv8i16.nxv8i8(
+  <vscale x 8 x i16>,
+  <vscale x 8 x i8>,
+  <vscale x 8 x i8>,
+  i32);
+
+define <vscale x 8 x i16>  @intrinsic_vwmacc_vv_nxv8i16_nxv8i8_nxv8i8(<vscale x 8 x i16> %0, <vscale x 8 x i8> %1, <vscale x 8 x i8> %2, i32 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmacc_vv_nxv8i16_nxv8i8_nxv8i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e8,m1,ta,mu
+; CHECK-NEXT:    vwmacc.vv v16, v18, v19
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i16> @llvm.riscv.vwmacc.nxv8i16.nxv8i8(
+    <vscale x 8 x i16> %0,
+    <vscale x 8 x i8> %1,
+    <vscale x 8 x i8> %2,
+    i32 %3)
+
+  ret <vscale x 8 x i16> %a
+}
+
+declare <vscale x 8 x i16> @llvm.riscv.vwmacc.mask.nxv8i16.nxv8i8(
+  <vscale x 8 x i16>,
+  <vscale x 8 x i8>,
+  <vscale x 8 x i8>,
+  <vscale x 8 x i1>,
+  i32);
+
+define <vscale x 8 x i16>  @intrinsic_vwmacc_mask_vv_nxv8i16_nxv8i8_nxv8i8(<vscale x 8 x i16> %0, <vscale x 8 x i8> %1, <vscale x 8 x i8> %2, <vscale x 8 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmacc_mask_vv_nxv8i16_nxv8i8_nxv8i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e8,m1,ta,mu
+; CHECK-NEXT:    vwmacc.vv v16, v18, v19, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i16> @llvm.riscv.vwmacc.mask.nxv8i16.nxv8i8(
+    <vscale x 8 x i16> %0,
+    <vscale x 8 x i8> %1,
+    <vscale x 8 x i8> %2,
+    <vscale x 8 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 8 x i16> %a
+}
+
+declare <vscale x 16 x i16> @llvm.riscv.vwmacc.nxv16i16.nxv16i8(
+  <vscale x 16 x i16>,
+  <vscale x 16 x i8>,
+  <vscale x 16 x i8>,
+  i32);
+
+define <vscale x 16 x i16>  @intrinsic_vwmacc_vv_nxv16i16_nxv16i8_nxv16i8(<vscale x 16 x i16> %0, <vscale x 16 x i8> %1, <vscale x 16 x i8> %2, i32 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmacc_vv_nxv16i16_nxv16i8_nxv16i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e8,m2,ta,mu
+; CHECK-NEXT:    vwmacc.vv v16, v20, v22
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 16 x i16> @llvm.riscv.vwmacc.nxv16i16.nxv16i8(
+    <vscale x 16 x i16> %0,
+    <vscale x 16 x i8> %1,
+    <vscale x 16 x i8> %2,
+    i32 %3)
+
+  ret <vscale x 16 x i16> %a
+}
+
+declare <vscale x 16 x i16> @llvm.riscv.vwmacc.mask.nxv16i16.nxv16i8(
+  <vscale x 16 x i16>,
+  <vscale x 16 x i8>,
+  <vscale x 16 x i8>,
+  <vscale x 16 x i1>,
+  i32);
+
+define <vscale x 16 x i16>  @intrinsic_vwmacc_mask_vv_nxv16i16_nxv16i8_nxv16i8(<vscale x 16 x i16> %0, <vscale x 16 x i8> %1, <vscale x 16 x i8> %2, <vscale x 16 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmacc_mask_vv_nxv16i16_nxv16i8_nxv16i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e8,m2,ta,mu
+; CHECK-NEXT:    vwmacc.vv v16, v20, v22, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 16 x i16> @llvm.riscv.vwmacc.mask.nxv16i16.nxv16i8(
+    <vscale x 16 x i16> %0,
+    <vscale x 16 x i8> %1,
+    <vscale x 16 x i8> %2,
+    <vscale x 16 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 16 x i16> %a
+}
+
+declare <vscale x 32 x i16> @llvm.riscv.vwmacc.nxv32i16.nxv32i8(
+  <vscale x 32 x i16>,
+  <vscale x 32 x i8>,
+  <vscale x 32 x i8>,
+  i32);
+
+define <vscale x 32 x i16>  @intrinsic_vwmacc_vv_nxv32i16_nxv32i8_nxv32i8(<vscale x 32 x i16> %0, <vscale x 32 x i8> %1, <vscale x 32 x i8> %2, i32 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmacc_vv_nxv32i16_nxv32i8_nxv32i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a3, zero, e8,m4,ta,mu
+; CHECK-NEXT:    vle8.v v28, (a1)
+; CHECK-NEXT:    vle8.v v8, (a0)
+; CHECK-NEXT:    vsetvli a0, a2, e8,m4,ta,mu
+; CHECK-NEXT:    vwmacc.vv v16, v8, v28
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 32 x i16> @llvm.riscv.vwmacc.nxv32i16.nxv32i8(
+    <vscale x 32 x i16> %0,
+    <vscale x 32 x i8> %1,
+    <vscale x 32 x i8> %2,
+    i32 %3)
+
+  ret <vscale x 32 x i16> %a
+}
+
+declare <vscale x 32 x i16> @llvm.riscv.vwmacc.mask.nxv32i16.nxv32i8(
+  <vscale x 32 x i16>,
+  <vscale x 32 x i8>,
+  <vscale x 32 x i8>,
+  <vscale x 32 x i1>,
+  i32);
+
+define <vscale x 32 x i16>  @intrinsic_vwmacc_mask_vv_nxv32i16_nxv32i8_nxv32i8(<vscale x 32 x i16> %0, <vscale x 32 x i8> %1, <vscale x 32 x i8> %2, <vscale x 32 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmacc_mask_vv_nxv32i16_nxv32i8_nxv32i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a3, zero, e8,m4,ta,mu
+; CHECK-NEXT:    vle8.v v28, (a1)
+; CHECK-NEXT:    vle8.v v8, (a0)
+; CHECK-NEXT:    vsetvli a0, a2, e8,m4,ta,mu
+; CHECK-NEXT:    vwmacc.vv v16, v8, v28, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 32 x i16> @llvm.riscv.vwmacc.mask.nxv32i16.nxv32i8(
+    <vscale x 32 x i16> %0,
+    <vscale x 32 x i8> %1,
+    <vscale x 32 x i8> %2,
+    <vscale x 32 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 32 x i16> %a
+}
+
+declare <vscale x 1 x i32> @llvm.riscv.vwmacc.nxv1i32.nxv1i16(
+  <vscale x 1 x i32>,
+  <vscale x 1 x i16>,
+  <vscale x 1 x i16>,
+  i32);
+
+define <vscale x 1 x i32>  @intrinsic_vwmacc_vv_nxv1i32_nxv1i16_nxv1i16(<vscale x 1 x i32> %0, <vscale x 1 x i16> %1, <vscale x 1 x i16> %2, i32 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmacc_vv_nxv1i32_nxv1i16_nxv1i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e16,mf4,ta,mu
+; CHECK-NEXT:    vwmacc.vv v16, v17, v18
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i32> @llvm.riscv.vwmacc.nxv1i32.nxv1i16(
+    <vscale x 1 x i32> %0,
+    <vscale x 1 x i16> %1,
+    <vscale x 1 x i16> %2,
+    i32 %3)
+
+  ret <vscale x 1 x i32> %a
+}
+
+declare <vscale x 1 x i32> @llvm.riscv.vwmacc.mask.nxv1i32.nxv1i16(
+  <vscale x 1 x i32>,
+  <vscale x 1 x i16>,
+  <vscale x 1 x i16>,
+  <vscale x 1 x i1>,
+  i32);
+
+define <vscale x 1 x i32>  @intrinsic_vwmacc_mask_vv_nxv1i32_nxv1i16_nxv1i16(<vscale x 1 x i32> %0, <vscale x 1 x i16> %1, <vscale x 1 x i16> %2, <vscale x 1 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmacc_mask_vv_nxv1i32_nxv1i16_nxv1i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e16,mf4,ta,mu
+; CHECK-NEXT:    vwmacc.vv v16, v17, v18, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i32> @llvm.riscv.vwmacc.mask.nxv1i32.nxv1i16(
+    <vscale x 1 x i32> %0,
+    <vscale x 1 x i16> %1,
+    <vscale x 1 x i16> %2,
+    <vscale x 1 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 1 x i32> %a
+}
+
+declare <vscale x 2 x i32> @llvm.riscv.vwmacc.nxv2i32.nxv2i16(
+  <vscale x 2 x i32>,
+  <vscale x 2 x i16>,
+  <vscale x 2 x i16>,
+  i32);
+
+define <vscale x 2 x i32>  @intrinsic_vwmacc_vv_nxv2i32_nxv2i16_nxv2i16(<vscale x 2 x i32> %0, <vscale x 2 x i16> %1, <vscale x 2 x i16> %2, i32 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmacc_vv_nxv2i32_nxv2i16_nxv2i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e16,mf2,ta,mu
+; CHECK-NEXT:    vwmacc.vv v16, v17, v18
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i32> @llvm.riscv.vwmacc.nxv2i32.nxv2i16(
+    <vscale x 2 x i32> %0,
+    <vscale x 2 x i16> %1,
+    <vscale x 2 x i16> %2,
+    i32 %3)
+
+  ret <vscale x 2 x i32> %a
+}
+
+declare <vscale x 2 x i32> @llvm.riscv.vwmacc.mask.nxv2i32.nxv2i16(
+  <vscale x 2 x i32>,
+  <vscale x 2 x i16>,
+  <vscale x 2 x i16>,
+  <vscale x 2 x i1>,
+  i32);
+
+define <vscale x 2 x i32>  @intrinsic_vwmacc_mask_vv_nxv2i32_nxv2i16_nxv2i16(<vscale x 2 x i32> %0, <vscale x 2 x i16> %1, <vscale x 2 x i16> %2, <vscale x 2 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmacc_mask_vv_nxv2i32_nxv2i16_nxv2i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e16,mf2,ta,mu
+; CHECK-NEXT:    vwmacc.vv v16, v17, v18, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i32> @llvm.riscv.vwmacc.mask.nxv2i32.nxv2i16(
+    <vscale x 2 x i32> %0,
+    <vscale x 2 x i16> %1,
+    <vscale x 2 x i16> %2,
+    <vscale x 2 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 2 x i32> %a
+}
+
+declare <vscale x 4 x i32> @llvm.riscv.vwmacc.nxv4i32.nxv4i16(
+  <vscale x 4 x i32>,
+  <vscale x 4 x i16>,
+  <vscale x 4 x i16>,
+  i32);
+
+define <vscale x 4 x i32>  @intrinsic_vwmacc_vv_nxv4i32_nxv4i16_nxv4i16(<vscale x 4 x i32> %0, <vscale x 4 x i16> %1, <vscale x 4 x i16> %2, i32 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmacc_vv_nxv4i32_nxv4i16_nxv4i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e16,m1,ta,mu
+; CHECK-NEXT:    vwmacc.vv v16, v18, v19
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i32> @llvm.riscv.vwmacc.nxv4i32.nxv4i16(
+    <vscale x 4 x i32> %0,
+    <vscale x 4 x i16> %1,
+    <vscale x 4 x i16> %2,
+    i32 %3)
+
+  ret <vscale x 4 x i32> %a
+}
+
+declare <vscale x 4 x i32> @llvm.riscv.vwmacc.mask.nxv4i32.nxv4i16(
+  <vscale x 4 x i32>,
+  <vscale x 4 x i16>,
+  <vscale x 4 x i16>,
+  <vscale x 4 x i1>,
+  i32);
+
+define <vscale x 4 x i32>  @intrinsic_vwmacc_mask_vv_nxv4i32_nxv4i16_nxv4i16(<vscale x 4 x i32> %0, <vscale x 4 x i16> %1, <vscale x 4 x i16> %2, <vscale x 4 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmacc_mask_vv_nxv4i32_nxv4i16_nxv4i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e16,m1,ta,mu
+; CHECK-NEXT:    vwmacc.vv v16, v18, v19, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i32> @llvm.riscv.vwmacc.mask.nxv4i32.nxv4i16(
+    <vscale x 4 x i32> %0,
+    <vscale x 4 x i16> %1,
+    <vscale x 4 x i16> %2,
+    <vscale x 4 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 4 x i32> %a
+}
+
+declare <vscale x 8 x i32> @llvm.riscv.vwmacc.nxv8i32.nxv8i16(
+  <vscale x 8 x i32>,
+  <vscale x 8 x i16>,
+  <vscale x 8 x i16>,
+  i32);
+
+define <vscale x 8 x i32>  @intrinsic_vwmacc_vv_nxv8i32_nxv8i16_nxv8i16(<vscale x 8 x i32> %0, <vscale x 8 x i16> %1, <vscale x 8 x i16> %2, i32 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmacc_vv_nxv8i32_nxv8i16_nxv8i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e16,m2,ta,mu
+; CHECK-NEXT:    vwmacc.vv v16, v20, v22
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i32> @llvm.riscv.vwmacc.nxv8i32.nxv8i16(
+    <vscale x 8 x i32> %0,
+    <vscale x 8 x i16> %1,
+    <vscale x 8 x i16> %2,
+    i32 %3)
+
+  ret <vscale x 8 x i32> %a
+}
+
+declare <vscale x 8 x i32> @llvm.riscv.vwmacc.mask.nxv8i32.nxv8i16(
+  <vscale x 8 x i32>,
+  <vscale x 8 x i16>,
+  <vscale x 8 x i16>,
+  <vscale x 8 x i1>,
+  i32);
+
+define <vscale x 8 x i32>  @intrinsic_vwmacc_mask_vv_nxv8i32_nxv8i16_nxv8i16(<vscale x 8 x i32> %0, <vscale x 8 x i16> %1, <vscale x 8 x i16> %2, <vscale x 8 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmacc_mask_vv_nxv8i32_nxv8i16_nxv8i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e16,m2,ta,mu
+; CHECK-NEXT:    vwmacc.vv v16, v20, v22, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i32> @llvm.riscv.vwmacc.mask.nxv8i32.nxv8i16(
+    <vscale x 8 x i32> %0,
+    <vscale x 8 x i16> %1,
+    <vscale x 8 x i16> %2,
+    <vscale x 8 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 8 x i32> %a
+}
+
+declare <vscale x 16 x i32> @llvm.riscv.vwmacc.nxv16i32.nxv16i16(
+  <vscale x 16 x i32>,
+  <vscale x 16 x i16>,
+  <vscale x 16 x i16>,
+  i32);
+
+define <vscale x 16 x i32>  @intrinsic_vwmacc_vv_nxv16i32_nxv16i16_nxv16i16(<vscale x 16 x i32> %0, <vscale x 16 x i16> %1, <vscale x 16 x i16> %2, i32 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmacc_vv_nxv16i32_nxv16i16_nxv16i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a3, zero, e16,m4,ta,mu
+; CHECK-NEXT:    vle16.v v28, (a1)
+; CHECK-NEXT:    vle16.v v8, (a0)
+; CHECK-NEXT:    vsetvli a0, a2, e16,m4,ta,mu
+; CHECK-NEXT:    vwmacc.vv v16, v8, v28
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 16 x i32> @llvm.riscv.vwmacc.nxv16i32.nxv16i16(
+    <vscale x 16 x i32> %0,
+    <vscale x 16 x i16> %1,
+    <vscale x 16 x i16> %2,
+    i32 %3)
+
+  ret <vscale x 16 x i32> %a
+}
+
+declare <vscale x 16 x i32> @llvm.riscv.vwmacc.mask.nxv16i32.nxv16i16(
+  <vscale x 16 x i32>,
+  <vscale x 16 x i16>,
+  <vscale x 16 x i16>,
+  <vscale x 16 x i1>,
+  i32);
+
+define <vscale x 16 x i32>  @intrinsic_vwmacc_mask_vv_nxv16i32_nxv16i16_nxv16i16(<vscale x 16 x i32> %0, <vscale x 16 x i16> %1, <vscale x 16 x i16> %2, <vscale x 16 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmacc_mask_vv_nxv16i32_nxv16i16_nxv16i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a3, zero, e16,m4,ta,mu
+; CHECK-NEXT:    vle16.v v28, (a1)
+; CHECK-NEXT:    vle16.v v8, (a0)
+; CHECK-NEXT:    vsetvli a0, a2, e16,m4,ta,mu
+; CHECK-NEXT:    vwmacc.vv v16, v8, v28, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 16 x i32> @llvm.riscv.vwmacc.mask.nxv16i32.nxv16i16(
+    <vscale x 16 x i32> %0,
+    <vscale x 16 x i16> %1,
+    <vscale x 16 x i16> %2,
+    <vscale x 16 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 16 x i32> %a
+}
+
+declare <vscale x 1 x i16> @llvm.riscv.vwmacc.nxv1i16.i8(
+  <vscale x 1 x i16>,
+  i8,
+  <vscale x 1 x i8>,
+  i32);
+
+define <vscale x 1 x i16>  @intrinsic_vwmacc_vx_nxv1i16_i8_nxv1i8(<vscale x 1 x i16> %0, i8 %1, <vscale x 1 x i8> %2, i32 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmacc_vx_nxv1i16_i8_nxv1i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e8,mf8,ta,mu
+; CHECK-NEXT:    vwmacc.vx v16, a0, v17
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i16> @llvm.riscv.vwmacc.nxv1i16.i8(
+    <vscale x 1 x i16> %0,
+    i8 %1,
+    <vscale x 1 x i8> %2,
+    i32 %3)
+
+  ret <vscale x 1 x i16> %a
+}
+
+declare <vscale x 1 x i16> @llvm.riscv.vwmacc.mask.nxv1i16.i8(
+  <vscale x 1 x i16>,
+  i8,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i1>,
+  i32);
+
+define <vscale x 1 x i16> @intrinsic_vwmacc_mask_vx_nxv1i16_i8_nxv1i8(<vscale x 1 x i16> %0, i8 %1, <vscale x 1 x i8> %2, <vscale x 1 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmacc_mask_vx_nxv1i16_i8_nxv1i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e8,mf8,ta,mu
+; CHECK-NEXT:    vwmacc.vx v16, a0, v17, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i16> @llvm.riscv.vwmacc.mask.nxv1i16.i8(
+    <vscale x 1 x i16> %0,
+    i8 %1,
+    <vscale x 1 x i8> %2,
+    <vscale x 1 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 1 x i16> %a
+}
+
+declare <vscale x 2 x i16> @llvm.riscv.vwmacc.nxv2i16.i8(
+  <vscale x 2 x i16>,
+  i8,
+  <vscale x 2 x i8>,
+  i32);
+
+define <vscale x 2 x i16>  @intrinsic_vwmacc_vx_nxv2i16_i8_nxv2i8(<vscale x 2 x i16> %0, i8 %1, <vscale x 2 x i8> %2, i32 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmacc_vx_nxv2i16_i8_nxv2i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e8,mf4,ta,mu
+; CHECK-NEXT:    vwmacc.vx v16, a0, v17
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i16> @llvm.riscv.vwmacc.nxv2i16.i8(
+    <vscale x 2 x i16> %0,
+    i8 %1,
+    <vscale x 2 x i8> %2,
+    i32 %3)
+
+  ret <vscale x 2 x i16> %a
+}
+
+declare <vscale x 2 x i16> @llvm.riscv.vwmacc.mask.nxv2i16.i8(
+  <vscale x 2 x i16>,
+  i8,
+  <vscale x 2 x i8>,
+  <vscale x 2 x i1>,
+  i32);
+
+define <vscale x 2 x i16> @intrinsic_vwmacc_mask_vx_nxv2i16_i8_nxv2i8(<vscale x 2 x i16> %0, i8 %1, <vscale x 2 x i8> %2, <vscale x 2 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmacc_mask_vx_nxv2i16_i8_nxv2i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e8,mf4,ta,mu
+; CHECK-NEXT:    vwmacc.vx v16, a0, v17, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i16> @llvm.riscv.vwmacc.mask.nxv2i16.i8(
+    <vscale x 2 x i16> %0,
+    i8 %1,
+    <vscale x 2 x i8> %2,
+    <vscale x 2 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 2 x i16> %a
+}
+
+declare <vscale x 4 x i16> @llvm.riscv.vwmacc.nxv4i16.i8(
+  <vscale x 4 x i16>,
+  i8,
+  <vscale x 4 x i8>,
+  i32);
+
+define <vscale x 4 x i16>  @intrinsic_vwmacc_vx_nxv4i16_i8_nxv4i8(<vscale x 4 x i16> %0, i8 %1, <vscale x 4 x i8> %2, i32 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmacc_vx_nxv4i16_i8_nxv4i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e8,mf2,ta,mu
+; CHECK-NEXT:    vwmacc.vx v16, a0, v17
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i16> @llvm.riscv.vwmacc.nxv4i16.i8(
+    <vscale x 4 x i16> %0,
+    i8 %1,
+    <vscale x 4 x i8> %2,
+    i32 %3)
+
+  ret <vscale x 4 x i16> %a
+}
+
+declare <vscale x 4 x i16> @llvm.riscv.vwmacc.mask.nxv4i16.i8(
+  <vscale x 4 x i16>,
+  i8,
+  <vscale x 4 x i8>,
+  <vscale x 4 x i1>,
+  i32);
+
+define <vscale x 4 x i16> @intrinsic_vwmacc_mask_vx_nxv4i16_i8_nxv4i8(<vscale x 4 x i16> %0, i8 %1, <vscale x 4 x i8> %2, <vscale x 4 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmacc_mask_vx_nxv4i16_i8_nxv4i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e8,mf2,ta,mu
+; CHECK-NEXT:    vwmacc.vx v16, a0, v17, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i16> @llvm.riscv.vwmacc.mask.nxv4i16.i8(
+    <vscale x 4 x i16> %0,
+    i8 %1,
+    <vscale x 4 x i8> %2,
+    <vscale x 4 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 4 x i16> %a
+}
+
+declare <vscale x 8 x i16> @llvm.riscv.vwmacc.nxv8i16.i8(
+  <vscale x 8 x i16>,
+  i8,
+  <vscale x 8 x i8>,
+  i32);
+
+define <vscale x 8 x i16>  @intrinsic_vwmacc_vx_nxv8i16_i8_nxv8i8(<vscale x 8 x i16> %0, i8 %1, <vscale x 8 x i8> %2, i32 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmacc_vx_nxv8i16_i8_nxv8i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e8,m1,ta,mu
+; CHECK-NEXT:    vwmacc.vx v16, a0, v18
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i16> @llvm.riscv.vwmacc.nxv8i16.i8(
+    <vscale x 8 x i16> %0,
+    i8 %1,
+    <vscale x 8 x i8> %2,
+    i32 %3)
+
+  ret <vscale x 8 x i16> %a
+}
+
+declare <vscale x 8 x i16> @llvm.riscv.vwmacc.mask.nxv8i16.i8(
+  <vscale x 8 x i16>,
+  i8,
+  <vscale x 8 x i8>,
+  <vscale x 8 x i1>,
+  i32);
+
+define <vscale x 8 x i16> @intrinsic_vwmacc_mask_vx_nxv8i16_i8_nxv8i8(<vscale x 8 x i16> %0, i8 %1, <vscale x 8 x i8> %2, <vscale x 8 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmacc_mask_vx_nxv8i16_i8_nxv8i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e8,m1,ta,mu
+; CHECK-NEXT:    vwmacc.vx v16, a0, v18, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i16> @llvm.riscv.vwmacc.mask.nxv8i16.i8(
+    <vscale x 8 x i16> %0,
+    i8 %1,
+    <vscale x 8 x i8> %2,
+    <vscale x 8 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 8 x i16> %a
+}
+
+declare <vscale x 16 x i16> @llvm.riscv.vwmacc.nxv16i16.i8(
+  <vscale x 16 x i16>,
+  i8,
+  <vscale x 16 x i8>,
+  i32);
+
+define <vscale x 16 x i16>  @intrinsic_vwmacc_vx_nxv16i16_i8_nxv16i8(<vscale x 16 x i16> %0, i8 %1, <vscale x 16 x i8> %2, i32 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmacc_vx_nxv16i16_i8_nxv16i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e8,m2,ta,mu
+; CHECK-NEXT:    vwmacc.vx v16, a0, v20
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 16 x i16> @llvm.riscv.vwmacc.nxv16i16.i8(
+    <vscale x 16 x i16> %0,
+    i8 %1,
+    <vscale x 16 x i8> %2,
+    i32 %3)
+
+  ret <vscale x 16 x i16> %a
+}
+
+declare <vscale x 16 x i16> @llvm.riscv.vwmacc.mask.nxv16i16.i8(
+  <vscale x 16 x i16>,
+  i8,
+  <vscale x 16 x i8>,
+  <vscale x 16 x i1>,
+  i32);
+
+define <vscale x 16 x i16> @intrinsic_vwmacc_mask_vx_nxv16i16_i8_nxv16i8(<vscale x 16 x i16> %0, i8 %1, <vscale x 16 x i8> %2, <vscale x 16 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmacc_mask_vx_nxv16i16_i8_nxv16i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e8,m2,ta,mu
+; CHECK-NEXT:    vwmacc.vx v16, a0, v20, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 16 x i16> @llvm.riscv.vwmacc.mask.nxv16i16.i8(
+    <vscale x 16 x i16> %0,
+    i8 %1,
+    <vscale x 16 x i8> %2,
+    <vscale x 16 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 16 x i16> %a
+}
+
+declare <vscale x 32 x i16> @llvm.riscv.vwmacc.nxv32i16.i8(
+  <vscale x 32 x i16>,
+  i8,
+  <vscale x 32 x i8>,
+  i32);
+
+define <vscale x 32 x i16>  @intrinsic_vwmacc_vx_nxv32i16_i8_nxv32i8(<vscale x 32 x i16> %0, i8 %1, <vscale x 32 x i8> %2, i32 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmacc_vx_nxv32i16_i8_nxv32i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a3, zero, e8,m4,ta,mu
+; CHECK-NEXT:    vle8.v v28, (a1)
+; CHECK-NEXT:    vsetvli a1, a2, e8,m4,ta,mu
+; CHECK-NEXT:    vwmacc.vx v16, a0, v28
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 32 x i16> @llvm.riscv.vwmacc.nxv32i16.i8(
+    <vscale x 32 x i16> %0,
+    i8 %1,
+    <vscale x 32 x i8> %2,
+    i32 %3)
+
+  ret <vscale x 32 x i16> %a
+}
+
+declare <vscale x 32 x i16> @llvm.riscv.vwmacc.mask.nxv32i16.i8(
+  <vscale x 32 x i16>,
+  i8,
+  <vscale x 32 x i8>,
+  <vscale x 32 x i1>,
+  i32);
+
+define <vscale x 32 x i16> @intrinsic_vwmacc_mask_vx_nxv32i16_i8_nxv32i8(<vscale x 32 x i16> %0, i8 %1, <vscale x 32 x i8> %2, <vscale x 32 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmacc_mask_vx_nxv32i16_i8_nxv32i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a3, zero, e8,m4,ta,mu
+; CHECK-NEXT:    vle8.v v28, (a1)
+; CHECK-NEXT:    vsetvli a1, a2, e8,m4,ta,mu
+; CHECK-NEXT:    vwmacc.vx v16, a0, v28, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 32 x i16> @llvm.riscv.vwmacc.mask.nxv32i16.i8(
+    <vscale x 32 x i16> %0,
+    i8 %1,
+    <vscale x 32 x i8> %2,
+    <vscale x 32 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 32 x i16> %a
+}
+
+declare <vscale x 1 x i32> @llvm.riscv.vwmacc.nxv1i32.i16(
+  <vscale x 1 x i32>,
+  i16,
+  <vscale x 1 x i16>,
+  i32);
+
+define <vscale x 1 x i32>  @intrinsic_vwmacc_vx_nxv1i32_i16_nxv1i16(<vscale x 1 x i32> %0, i16 %1, <vscale x 1 x i16> %2, i32 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmacc_vx_nxv1i32_i16_nxv1i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e16,mf4,ta,mu
+; CHECK-NEXT:    vwmacc.vx v16, a0, v17
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i32> @llvm.riscv.vwmacc.nxv1i32.i16(
+    <vscale x 1 x i32> %0,
+    i16 %1,
+    <vscale x 1 x i16> %2,
+    i32 %3)
+
+  ret <vscale x 1 x i32> %a
+}
+
+declare <vscale x 1 x i32> @llvm.riscv.vwmacc.mask.nxv1i32.i16(
+  <vscale x 1 x i32>,
+  i16,
+  <vscale x 1 x i16>,
+  <vscale x 1 x i1>,
+  i32);
+
+define <vscale x 1 x i32> @intrinsic_vwmacc_mask_vx_nxv1i32_i16_nxv1i16(<vscale x 1 x i32> %0, i16 %1, <vscale x 1 x i16> %2, <vscale x 1 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmacc_mask_vx_nxv1i32_i16_nxv1i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e16,mf4,ta,mu
+; CHECK-NEXT:    vwmacc.vx v16, a0, v17, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i32> @llvm.riscv.vwmacc.mask.nxv1i32.i16(
+    <vscale x 1 x i32> %0,
+    i16 %1,
+    <vscale x 1 x i16> %2,
+    <vscale x 1 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 1 x i32> %a
+}
+
+declare <vscale x 2 x i32> @llvm.riscv.vwmacc.nxv2i32.i16(
+  <vscale x 2 x i32>,
+  i16,
+  <vscale x 2 x i16>,
+  i32);
+
+define <vscale x 2 x i32>  @intrinsic_vwmacc_vx_nxv2i32_i16_nxv2i16(<vscale x 2 x i32> %0, i16 %1, <vscale x 2 x i16> %2, i32 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmacc_vx_nxv2i32_i16_nxv2i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e16,mf2,ta,mu
+; CHECK-NEXT:    vwmacc.vx v16, a0, v17
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i32> @llvm.riscv.vwmacc.nxv2i32.i16(
+    <vscale x 2 x i32> %0,
+    i16 %1,
+    <vscale x 2 x i16> %2,
+    i32 %3)
+
+  ret <vscale x 2 x i32> %a
+}
+
+declare <vscale x 2 x i32> @llvm.riscv.vwmacc.mask.nxv2i32.i16(
+  <vscale x 2 x i32>,
+  i16,
+  <vscale x 2 x i16>,
+  <vscale x 2 x i1>,
+  i32);
+
+define <vscale x 2 x i32> @intrinsic_vwmacc_mask_vx_nxv2i32_i16_nxv2i16(<vscale x 2 x i32> %0, i16 %1, <vscale x 2 x i16> %2, <vscale x 2 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmacc_mask_vx_nxv2i32_i16_nxv2i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e16,mf2,ta,mu
+; CHECK-NEXT:    vwmacc.vx v16, a0, v17, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i32> @llvm.riscv.vwmacc.mask.nxv2i32.i16(
+    <vscale x 2 x i32> %0,
+    i16 %1,
+    <vscale x 2 x i16> %2,
+    <vscale x 2 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 2 x i32> %a
+}
+
+declare <vscale x 4 x i32> @llvm.riscv.vwmacc.nxv4i32.i16(
+  <vscale x 4 x i32>,
+  i16,
+  <vscale x 4 x i16>,
+  i32);
+
+define <vscale x 4 x i32>  @intrinsic_vwmacc_vx_nxv4i32_i16_nxv4i16(<vscale x 4 x i32> %0, i16 %1, <vscale x 4 x i16> %2, i32 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmacc_vx_nxv4i32_i16_nxv4i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e16,m1,ta,mu
+; CHECK-NEXT:    vwmacc.vx v16, a0, v18
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i32> @llvm.riscv.vwmacc.nxv4i32.i16(
+    <vscale x 4 x i32> %0,
+    i16 %1,
+    <vscale x 4 x i16> %2,
+    i32 %3)
+
+  ret <vscale x 4 x i32> %a
+}
+
+declare <vscale x 4 x i32> @llvm.riscv.vwmacc.mask.nxv4i32.i16(
+  <vscale x 4 x i32>,
+  i16,
+  <vscale x 4 x i16>,
+  <vscale x 4 x i1>,
+  i32);
+
+define <vscale x 4 x i32> @intrinsic_vwmacc_mask_vx_nxv4i32_i16_nxv4i16(<vscale x 4 x i32> %0, i16 %1, <vscale x 4 x i16> %2, <vscale x 4 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmacc_mask_vx_nxv4i32_i16_nxv4i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e16,m1,ta,mu
+; CHECK-NEXT:    vwmacc.vx v16, a0, v18, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i32> @llvm.riscv.vwmacc.mask.nxv4i32.i16(
+    <vscale x 4 x i32> %0,
+    i16 %1,
+    <vscale x 4 x i16> %2,
+    <vscale x 4 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 4 x i32> %a
+}
+
+declare <vscale x 8 x i32> @llvm.riscv.vwmacc.nxv8i32.i16(
+  <vscale x 8 x i32>,
+  i16,
+  <vscale x 8 x i16>,
+  i32);
+
+define <vscale x 8 x i32>  @intrinsic_vwmacc_vx_nxv8i32_i16_nxv8i16(<vscale x 8 x i32> %0, i16 %1, <vscale x 8 x i16> %2, i32 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmacc_vx_nxv8i32_i16_nxv8i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e16,m2,ta,mu
+; CHECK-NEXT:    vwmacc.vx v16, a0, v20
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i32> @llvm.riscv.vwmacc.nxv8i32.i16(
+    <vscale x 8 x i32> %0,
+    i16 %1,
+    <vscale x 8 x i16> %2,
+    i32 %3)
+
+  ret <vscale x 8 x i32> %a
+}
+
+declare <vscale x 8 x i32> @llvm.riscv.vwmacc.mask.nxv8i32.i16(
+  <vscale x 8 x i32>,
+  i16,
+  <vscale x 8 x i16>,
+  <vscale x 8 x i1>,
+  i32);
+
+define <vscale x 8 x i32> @intrinsic_vwmacc_mask_vx_nxv8i32_i16_nxv8i16(<vscale x 8 x i32> %0, i16 %1, <vscale x 8 x i16> %2, <vscale x 8 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmacc_mask_vx_nxv8i32_i16_nxv8i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e16,m2,ta,mu
+; CHECK-NEXT:    vwmacc.vx v16, a0, v20, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i32> @llvm.riscv.vwmacc.mask.nxv8i32.i16(
+    <vscale x 8 x i32> %0,
+    i16 %1,
+    <vscale x 8 x i16> %2,
+    <vscale x 8 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 8 x i32> %a
+}
+
+declare <vscale x 16 x i32> @llvm.riscv.vwmacc.nxv16i32.i16(
+  <vscale x 16 x i32>,
+  i16,
+  <vscale x 16 x i16>,
+  i32);
+
+define <vscale x 16 x i32>  @intrinsic_vwmacc_vx_nxv16i32_i16_nxv16i16(<vscale x 16 x i32> %0, i16 %1, <vscale x 16 x i16> %2, i32 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmacc_vx_nxv16i32_i16_nxv16i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a3, zero, e16,m4,ta,mu
+; CHECK-NEXT:    vle16.v v28, (a1)
+; CHECK-NEXT:    vsetvli a1, a2, e16,m4,ta,mu
+; CHECK-NEXT:    vwmacc.vx v16, a0, v28
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 16 x i32> @llvm.riscv.vwmacc.nxv16i32.i16(
+    <vscale x 16 x i32> %0,
+    i16 %1,
+    <vscale x 16 x i16> %2,
+    i32 %3)
+
+  ret <vscale x 16 x i32> %a
+}
+
+declare <vscale x 16 x i32> @llvm.riscv.vwmacc.mask.nxv16i32.i16(
+  <vscale x 16 x i32>,
+  i16,
+  <vscale x 16 x i16>,
+  <vscale x 16 x i1>,
+  i32);
+
+define <vscale x 16 x i32> @intrinsic_vwmacc_mask_vx_nxv16i32_i16_nxv16i16(<vscale x 16 x i32> %0, i16 %1, <vscale x 16 x i16> %2, <vscale x 16 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmacc_mask_vx_nxv16i32_i16_nxv16i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a3, zero, e16,m4,ta,mu
+; CHECK-NEXT:    vle16.v v28, (a1)
+; CHECK-NEXT:    vsetvli a1, a2, e16,m4,ta,mu
+; CHECK-NEXT:    vwmacc.vx v16, a0, v28, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 16 x i32> @llvm.riscv.vwmacc.mask.nxv16i32.i16(
+    <vscale x 16 x i32> %0,
+    i16 %1,
+    <vscale x 16 x i16> %2,
+    <vscale x 16 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 16 x i32> %a
+}

llvm/test/CodeGen/RISCV/rvv/vwmacc-rv64.ll

+; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
+; RUN: llc -mtriple=riscv64 -mattr=+experimental-v,+d -verify-machineinstrs \
+; RUN:   --riscv-no-aliases < %s | FileCheck %s
+declare <vscale x 1 x i16> @llvm.riscv.vwmacc.nxv1i16.nxv1i8(
+  <vscale x 1 x i16>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  i64);
+
+define <vscale x 1 x i16>  @intrinsic_vwmacc_vv_nxv1i16_nxv1i8_nxv1i8(<vscale x 1 x i16> %0, <vscale x 1 x i8> %1, <vscale x 1 x i8> %2, i64 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmacc_vv_nxv1i16_nxv1i8_nxv1i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e8,mf8,ta,mu
+; CHECK-NEXT:    vwmacc.vv v16, v17, v18
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i16> @llvm.riscv.vwmacc.nxv1i16.nxv1i8(
+    <vscale x 1 x i16> %0,
+    <vscale x 1 x i8> %1,
+    <vscale x 1 x i8> %2,
+    i64 %3)
+
+  ret <vscale x 1 x i16> %a
+}
+
+declare <vscale x 1 x i16> @llvm.riscv.vwmacc.mask.nxv1i16.nxv1i8(
+  <vscale x 1 x i16>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i1>,
+  i64);
+
+define <vscale x 1 x i16>  @intrinsic_vwmacc_mask_vv_nxv1i16_nxv1i8_nxv1i8(<vscale x 1 x i16> %0, <vscale x 1 x i8> %1, <vscale x 1 x i8> %2, <vscale x 1 x i1> %3, i64 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmacc_mask_vv_nxv1i16_nxv1i8_nxv1i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e8,mf8,ta,mu
+; CHECK-NEXT:    vwmacc.vv v16, v17, v18, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i16> @llvm.riscv.vwmacc.mask.nxv1i16.nxv1i8(
+    <vscale x 1 x i16> %0,
+    <vscale x 1 x i8> %1,
+    <vscale x 1 x i8> %2,
+    <vscale x 1 x i1> %3,
+    i64 %4)
+
+  ret <vscale x 1 x i16> %a
+}
+
+declare <vscale x 2 x i16> @llvm.riscv.vwmacc.nxv2i16.nxv2i8(
+  <vscale x 2 x i16>,
+  <vscale x 2 x i8>,
+  <vscale x 2 x i8>,
+  i64);
+
+define <vscale x 2 x i16>  @intrinsic_vwmacc_vv_nxv2i16_nxv2i8_nxv2i8(<vscale x 2 x i16> %0, <vscale x 2 x i8> %1, <vscale x 2 x i8> %2, i64 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmacc_vv_nxv2i16_nxv2i8_nxv2i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e8,mf4,ta,mu
+; CHECK-NEXT:    vwmacc.vv v16, v17, v18
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i16> @llvm.riscv.vwmacc.nxv2i16.nxv2i8(
+    <vscale x 2 x i16> %0,
+    <vscale x 2 x i8> %1,
+    <vscale x 2 x i8> %2,
+    i64 %3)
+
+  ret <vscale x 2 x i16> %a
+}
+
+declare <vscale x 2 x i16> @llvm.riscv.vwmacc.mask.nxv2i16.nxv2i8(
+  <vscale x 2 x i16>,
+  <vscale x 2 x i8>,
+  <vscale x 2 x i8>,
+  <vscale x 2 x i1>,
+  i64);
+
+define <vscale x 2 x i16>  @intrinsic_vwmacc_mask_vv_nxv2i16_nxv2i8_nxv2i8(<vscale x 2 x i16> %0, <vscale x 2 x i8> %1, <vscale x 2 x i8> %2, <vscale x 2 x i1> %3, i64 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmacc_mask_vv_nxv2i16_nxv2i8_nxv2i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e8,mf4,ta,mu
+; CHECK-NEXT:    vwmacc.vv v16, v17, v18, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i16> @llvm.riscv.vwmacc.mask.nxv2i16.nxv2i8(
+    <vscale x 2 x i16> %0,
+    <vscale x 2 x i8> %1,
+    <vscale x 2 x i8> %2,
+    <vscale x 2 x i1> %3,
+    i64 %4)
+
+  ret <vscale x 2 x i16> %a
+}
+
+declare <vscale x 4 x i16> @llvm.riscv.vwmacc.nxv4i16.nxv4i8(
+  <vscale x 4 x i16>,
+  <vscale x 4 x i8>,
+  <vscale x 4 x i8>,
+  i64);
+
+define <vscale x 4 x i16>  @intrinsic_vwmacc_vv_nxv4i16_nxv4i8_nxv4i8(<vscale x 4 x i16> %0, <vscale x 4 x i8> %1, <vscale x 4 x i8> %2, i64 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmacc_vv_nxv4i16_nxv4i8_nxv4i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e8,mf2,ta,mu
+; CHECK-NEXT:    vwmacc.vv v16, v17, v18
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i16> @llvm.riscv.vwmacc.nxv4i16.nxv4i8(
+    <vscale x 4 x i16> %0,
+    <vscale x 4 x i8> %1,
+    <vscale x 4 x i8> %2,
+    i64 %3)
+
+  ret <vscale x 4 x i16> %a
+}
+
+declare <vscale x 4 x i16> @llvm.riscv.vwmacc.mask.nxv4i16.nxv4i8(
+  <vscale x 4 x i16>,
+  <vscale x 4 x i8>,
+  <vscale x 4 x i8>,
+  <vscale x 4 x i1>,
+  i64);
+
+define <vscale x 4 x i16>  @intrinsic_vwmacc_mask_vv_nxv4i16_nxv4i8_nxv4i8(<vscale x 4 x i16> %0, <vscale x 4 x i8> %1, <vscale x 4 x i8> %2, <vscale x 4 x i1> %3, i64 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmacc_mask_vv_nxv4i16_nxv4i8_nxv4i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e8,mf2,ta,mu
+; CHECK-NEXT:    vwmacc.vv v16, v17, v18, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i16> @llvm.riscv.vwmacc.mask.nxv4i16.nxv4i8(
+    <vscale x 4 x i16> %0,
+    <vscale x 4 x i8> %1,
+    <vscale x 4 x i8> %2,
+    <vscale x 4 x i1> %3,
+    i64 %4)
+
+  ret <vscale x 4 x i16> %a
+}
+
+declare <vscale x 8 x i16> @llvm.riscv.vwmacc.nxv8i16.nxv8i8(
+  <vscale x 8 x i16>,
+  <vscale x 8 x i8>,
+  <vscale x 8 x i8>,
+  i64);
+
+define <vscale x 8 x i16>  @intrinsic_vwmacc_vv_nxv8i16_nxv8i8_nxv8i8(<vscale x 8 x i16> %0, <vscale x 8 x i8> %1, <vscale x 8 x i8> %2, i64 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmacc_vv_nxv8i16_nxv8i8_nxv8i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e8,m1,ta,mu
+; CHECK-NEXT:    vwmacc.vv v16, v18, v19
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i16> @llvm.riscv.vwmacc.nxv8i16.nxv8i8(
+    <vscale x 8 x i16> %0,
+    <vscale x 8 x i8> %1,
+    <vscale x 8 x i8> %2,
+    i64 %3)
+
+  ret <vscale x 8 x i16> %a
+}
+
+declare <vscale x 8 x i16> @llvm.riscv.vwmacc.mask.nxv8i16.nxv8i8(
+  <vscale x 8 x i16>,
+  <vscale x 8 x i8>,
+  <vscale x 8 x i8>,
+  <vscale x 8 x i1>,
+  i64);
+
+define <vscale x 8 x i16>  @intrinsic_vwmacc_mask_vv_nxv8i16_nxv8i8_nxv8i8(<vscale x 8 x i16> %0, <vscale x 8 x i8> %1, <vscale x 8 x i8> %2, <vscale x 8 x i1> %3, i64 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmacc_mask_vv_nxv8i16_nxv8i8_nxv8i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e8,m1,ta,mu
+; CHECK-NEXT:    vwmacc.vv v16, v18, v19, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i16> @llvm.riscv.vwmacc.mask.nxv8i16.nxv8i8(
+    <vscale x 8 x i16> %0,
+    <vscale x 8 x i8> %1,
+    <vscale x 8 x i8> %2,
+    <vscale x 8 x i1> %3,
+    i64 %4)
+
+  ret <vscale x 8 x i16> %a
+}
+
+declare <vscale x 16 x i16> @llvm.riscv.vwmacc.nxv16i16.nxv16i8(
+  <vscale x 16 x i16>,
+  <vscale x 16 x i8>,
+  <vscale x 16 x i8>,
+  i64);
+
+define <vscale x 16 x i16>  @intrinsic_vwmacc_vv_nxv16i16_nxv16i8_nxv16i8(<vscale x 16 x i16> %0, <vscale x 16 x i8> %1, <vscale x 16 x i8> %2, i64 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmacc_vv_nxv16i16_nxv16i8_nxv16i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e8,m2,ta,mu
+; CHECK-NEXT:    vwmacc.vv v16, v20, v22
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 16 x i16> @llvm.riscv.vwmacc.nxv16i16.nxv16i8(
+    <vscale x 16 x i16> %0,
+    <vscale x 16 x i8> %1,
+    <vscale x 16 x i8> %2,
+    i64 %3)
+
+  ret <vscale x 16 x i16> %a
+}
+
+declare <vscale x 16 x i16> @llvm.riscv.vwmacc.mask.nxv16i16.nxv16i8(
+  <vscale x 16 x i16>,
+  <vscale x 16 x i8>,
+  <vscale x 16 x i8>,
+  <vscale x 16 x i1>,
+  i64);
+
+define <vscale x 16 x i16>  @intrinsic_vwmacc_mask_vv_nxv16i16_nxv16i8_nxv16i8(<vscale x 16 x i16> %0, <vscale x 16 x i8> %1, <vscale x 16 x i8> %2, <vscale x 16 x i1> %3, i64 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmacc_mask_vv_nxv16i16_nxv16i8_nxv16i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e8,m2,ta,mu
+; CHECK-NEXT:    vwmacc.vv v16, v20, v22, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 16 x i16> @llvm.riscv.vwmacc.mask.nxv16i16.nxv16i8(
+    <vscale x 16 x i16> %0,
+    <vscale x 16 x i8> %1,
+    <vscale x 16 x i8> %2,
+    <vscale x 16 x i1> %3,
+    i64 %4)
+
+  ret <vscale x 16 x i16> %a
+}
+
+declare <vscale x 32 x i16> @llvm.riscv.vwmacc.nxv32i16.nxv32i8(
+  <vscale x 32 x i16>,
+  <vscale x 32 x i8>,
+  <vscale x 32 x i8>,
+  i64);
+
+define <vscale x 32 x i16>  @intrinsic_vwmacc_vv_nxv32i16_nxv32i8_nxv32i8(<vscale x 32 x i16> %0, <vscale x 32 x i8> %1, <vscale x 32 x i8> %2, i64 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmacc_vv_nxv32i16_nxv32i8_nxv32i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a3, zero, e8,m4,ta,mu
+; CHECK-NEXT:    vle8.v v28, (a1)
+; CHECK-NEXT:    vle8.v v8, (a0)
+; CHECK-NEXT:    vsetvli a0, a2, e8,m4,ta,mu
+; CHECK-NEXT:    vwmacc.vv v16, v8, v28
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 32 x i16> @llvm.riscv.vwmacc.nxv32i16.nxv32i8(
+    <vscale x 32 x i16> %0,
+    <vscale x 32 x i8> %1,
+    <vscale x 32 x i8> %2,
+    i64 %3)
+
+  ret <vscale x 32 x i16> %a
+}
+
+declare <vscale x 32 x i16> @llvm.riscv.vwmacc.mask.nxv32i16.nxv32i8(
+  <vscale x 32 x i16>,
+  <vscale x 32 x i8>,
+  <vscale x 32 x i8>,
+  <vscale x 32 x i1>,
+  i64);
+
+define <vscale x 32 x i16>  @intrinsic_vwmacc_mask_vv_nxv32i16_nxv32i8_nxv32i8(<vscale x 32 x i16> %0, <vscale x 32 x i8> %1, <vscale x 32 x i8> %2, <vscale x 32 x i1> %3, i64 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmacc_mask_vv_nxv32i16_nxv32i8_nxv32i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a3, zero, e8,m4,ta,mu
+; CHECK-NEXT:    vle8.v v28, (a1)
+; CHECK-NEXT:    vle8.v v8, (a0)
+; CHECK-NEXT:    vsetvli a0, a2, e8,m4,ta,mu
+; CHECK-NEXT:    vwmacc.vv v16, v8, v28, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 32 x i16> @llvm.riscv.vwmacc.mask.nxv32i16.nxv32i8(
+    <vscale x 32 x i16> %0,
+    <vscale x 32 x i8> %1,
+    <vscale x 32 x i8> %2,
+    <vscale x 32 x i1> %3,
+    i64 %4)
+
+  ret <vscale x 32 x i16> %a
+}
+
+declare <vscale x 1 x i32> @llvm.riscv.vwmacc.nxv1i32.nxv1i16(
+  <vscale x 1 x i32>,
+  <vscale x 1 x i16>,
+  <vscale x 1 x i16>,
+  i64);
+
+define <vscale x 1 x i32>  @intrinsic_vwmacc_vv_nxv1i32_nxv1i16_nxv1i16(<vscale x 1 x i32> %0, <vscale x 1 x i16> %1, <vscale x 1 x i16> %2, i64 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmacc_vv_nxv1i32_nxv1i16_nxv1i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e16,mf4,ta,mu
+; CHECK-NEXT:    vwmacc.vv v16, v17, v18
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i32> @llvm.riscv.vwmacc.nxv1i32.nxv1i16(
+    <vscale x 1 x i32> %0,
+    <vscale x 1 x i16> %1,
+    <vscale x 1 x i16> %2,
+    i64 %3)
+
+  ret <vscale x 1 x i32> %a
+}
+
+declare <vscale x 1 x i32> @llvm.riscv.vwmacc.mask.nxv1i32.nxv1i16(
+  <vscale x 1 x i32>,
+  <vscale x 1 x i16>,
+  <vscale x 1 x i16>,
+  <vscale x 1 x i1>,
+  i64);
+
+define <vscale x 1 x i32>  @intrinsic_vwmacc_mask_vv_nxv1i32_nxv1i16_nxv1i16(<vscale x 1 x i32> %0, <vscale x 1 x i16> %1, <vscale x 1 x i16> %2, <vscale x 1 x i1> %3, i64 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmacc_mask_vv_nxv1i32_nxv1i16_nxv1i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e16,mf4,ta,mu
+; CHECK-NEXT:    vwmacc.vv v16, v17, v18, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i32> @llvm.riscv.vwmacc.mask.nxv1i32.nxv1i16(
+    <vscale x 1 x i32> %0,
+    <vscale x 1 x i16> %1,
+    <vscale x 1 x i16> %2,
+    <vscale x 1 x i1> %3,
+    i64 %4)
+
+  ret <vscale x 1 x i32> %a
+}
+
+declare <vscale x 2 x i32> @llvm.riscv.vwmacc.nxv2i32.nxv2i16(
+  <vscale x 2 x i32>,
+  <vscale x 2 x i16>,
+  <vscale x 2 x i16>,
+  i64);
+
+define <vscale x 2 x i32>  @intrinsic_vwmacc_vv_nxv2i32_nxv2i16_nxv2i16(<vscale x 2 x i32> %0, <vscale x 2 x i16> %1, <vscale x 2 x i16> %2, i64 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmacc_vv_nxv2i32_nxv2i16_nxv2i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e16,mf2,ta,mu
+; CHECK-NEXT:    vwmacc.vv v16, v17, v18
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i32> @llvm.riscv.vwmacc.nxv2i32.nxv2i16(
+    <vscale x 2 x i32> %0,
+    <vscale x 2 x i16> %1,
+    <vscale x 2 x i16> %2,
+    i64 %3)
+
+  ret <vscale x 2 x i32> %a
+}
+
+declare <vscale x 2 x i32> @llvm.riscv.vwmacc.mask.nxv2i32.nxv2i16(
+  <vscale x 2 x i32>,
+  <vscale x 2 x i16>,
+  <vscale x 2 x i16>,
+  <vscale x 2 x i1>,
+  i64);
+
+define <vscale x 2 x i32>  @intrinsic_vwmacc_mask_vv_nxv2i32_nxv2i16_nxv2i16(<vscale x 2 x i32> %0, <vscale x 2 x i16> %1, <vscale x 2 x i16> %2, <vscale x 2 x i1> %3, i64 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmacc_mask_vv_nxv2i32_nxv2i16_nxv2i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e16,mf2,ta,mu
+; CHECK-NEXT:    vwmacc.vv v16, v17, v18, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i32> @llvm.riscv.vwmacc.mask.nxv2i32.nxv2i16(
+    <vscale x 2 x i32> %0,
+    <vscale x 2 x i16> %1,
+    <vscale x 2 x i16> %2,
+    <vscale x 2 x i1> %3,
+    i64 %4)
+
+  ret <vscale x 2 x i32> %a
+}
+
+declare <vscale x 4 x i32> @llvm.riscv.vwmacc.nxv4i32.nxv4i16(
+  <vscale x 4 x i32>,
+  <vscale x 4 x i16>,
+  <vscale x 4 x i16>,
+  i64);
+
+define <vscale x 4 x i32>  @intrinsic_vwmacc_vv_nxv4i32_nxv4i16_nxv4i16(<vscale x 4 x i32> %0, <vscale x 4 x i16> %1, <vscale x 4 x i16> %2, i64 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmacc_vv_nxv4i32_nxv4i16_nxv4i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e16,m1,ta,mu
+; CHECK-NEXT:    vwmacc.vv v16, v18, v19
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i32> @llvm.riscv.vwmacc.nxv4i32.nxv4i16(
+    <vscale x 4 x i32> %0,
+    <vscale x 4 x i16> %1,
+    <vscale x 4 x i16> %2,
+    i64 %3)
+
+  ret <vscale x 4 x i32> %a
+}
+
+declare <vscale x 4 x i32> @llvm.riscv.vwmacc.mask.nxv4i32.nxv4i16(
+  <vscale x 4 x i32>,
+  <vscale x 4 x i16>,
+  <vscale x 4 x i16>,
+  <vscale x 4 x i1>,
+  i64);
+
+define <vscale x 4 x i32>  @intrinsic_vwmacc_mask_vv_nxv4i32_nxv4i16_nxv4i16(<vscale x 4 x i32> %0, <vscale x 4 x i16> %1, <vscale x 4 x i16> %2, <vscale x 4 x i1> %3, i64 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmacc_mask_vv_nxv4i32_nxv4i16_nxv4i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e16,m1,ta,mu
+; CHECK-NEXT:    vwmacc.vv v16, v18, v19, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i32> @llvm.riscv.vwmacc.mask.nxv4i32.nxv4i16(
+    <vscale x 4 x i32> %0,
+    <vscale x 4 x i16> %1,
+    <vscale x 4 x i16> %2,
+    <vscale x 4 x i1> %3,
+    i64 %4)
+
+  ret <vscale x 4 x i32> %a
+}
+
+declare <vscale x 8 x i32> @llvm.riscv.vwmacc.nxv8i32.nxv8i16(
+  <vscale x 8 x i32>,
+  <vscale x 8 x i16>,
+  <vscale x 8 x i16>,
+  i64);
+
+define <vscale x 8 x i32>  @intrinsic_vwmacc_vv_nxv8i32_nxv8i16_nxv8i16(<vscale x 8 x i32> %0, <vscale x 8 x i16> %1, <vscale x 8 x i16> %2, i64 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmacc_vv_nxv8i32_nxv8i16_nxv8i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e16,m2,ta,mu
+; CHECK-NEXT:    vwmacc.vv v16, v20, v22
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i32> @llvm.riscv.vwmacc.nxv8i32.nxv8i16(
+    <vscale x 8 x i32> %0,
+    <vscale x 8 x i16> %1,
+    <vscale x 8 x i16> %2,
+    i64 %3)
+
+  ret <vscale x 8 x i32> %a
+}
+
+declare <vscale x 8 x i32> @llvm.riscv.vwmacc.mask.nxv8i32.nxv8i16(
+  <vscale x 8 x i32>,
+  <vscale x 8 x i16>,
+  <vscale x 8 x i16>,
+  <vscale x 8 x i1>,
+  i64);
+
+define <vscale x 8 x i32>  @intrinsic_vwmacc_mask_vv_nxv8i32_nxv8i16_nxv8i16(<vscale x 8 x i32> %0, <vscale x 8 x i16> %1, <vscale x 8 x i16> %2, <vscale x 8 x i1> %3, i64 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmacc_mask_vv_nxv8i32_nxv8i16_nxv8i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e16,m2,ta,mu
+; CHECK-NEXT:    vwmacc.vv v16, v20, v22, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i32> @llvm.riscv.vwmacc.mask.nxv8i32.nxv8i16(
+    <vscale x 8 x i32> %0,
+    <vscale x 8 x i16> %1,
+    <vscale x 8 x i16> %2,
+    <vscale x 8 x i1> %3,
+    i64 %4)
+
+  ret <vscale x 8 x i32> %a
+}
+
+declare <vscale x 16 x i32> @llvm.riscv.vwmacc.nxv16i32.nxv16i16(
+  <vscale x 16 x i32>,
+  <vscale x 16 x i16>,
+  <vscale x 16 x i16>,
+  i64);
+
+define <vscale x 16 x i32>  @intrinsic_vwmacc_vv_nxv16i32_nxv16i16_nxv16i16(<vscale x 16 x i32> %0, <vscale x 16 x i16> %1, <vscale x 16 x i16> %2, i64 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmacc_vv_nxv16i32_nxv16i16_nxv16i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a3, zero, e16,m4,ta,mu
+; CHECK-NEXT:    vle16.v v28, (a1)
+; CHECK-NEXT:    vle16.v v8, (a0)
+; CHECK-NEXT:    vsetvli a0, a2, e16,m4,ta,mu
+; CHECK-NEXT:    vwmacc.vv v16, v8, v28
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 16 x i32> @llvm.riscv.vwmacc.nxv16i32.nxv16i16(
+    <vscale x 16 x i32> %0,
+    <vscale x 16 x i16> %1,
+    <vscale x 16 x i16> %2,
+    i64 %3)
+
+  ret <vscale x 16 x i32> %a
+}
+
+declare <vscale x 16 x i32> @llvm.riscv.vwmacc.mask.nxv16i32.nxv16i16(
+  <vscale x 16 x i32>,
+  <vscale x 16 x i16>,
+  <vscale x 16 x i16>,
+  <vscale x 16 x i1>,
+  i64);
+
+define <vscale x 16 x i32>  @intrinsic_vwmacc_mask_vv_nxv16i32_nxv16i16_nxv16i16(<vscale x 16 x i32> %0, <vscale x 16 x i16> %1, <vscale x 16 x i16> %2, <vscale x 16 x i1> %3, i64 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmacc_mask_vv_nxv16i32_nxv16i16_nxv16i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a3, zero, e16,m4,ta,mu
+; CHECK-NEXT:    vle16.v v28, (a1)
+; CHECK-NEXT:    vle16.v v8, (a0)
+; CHECK-NEXT:    vsetvli a0, a2, e16,m4,ta,mu
+; CHECK-NEXT:    vwmacc.vv v16, v8, v28, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 16 x i32> @llvm.riscv.vwmacc.mask.nxv16i32.nxv16i16(
+    <vscale x 16 x i32> %0,
+    <vscale x 16 x i16> %1,
+    <vscale x 16 x i16> %2,
+    <vscale x 16 x i1> %3,
+    i64 %4)
+
+  ret <vscale x 16 x i32> %a
+}
+
+declare <vscale x 1 x i64> @llvm.riscv.vwmacc.nxv1i64.nxv1i32(
+  <vscale x 1 x i64>,
+  <vscale x 1 x i32>,
+  <vscale x 1 x i32>,
+  i64);
+
+define <vscale x 1 x i64>  @intrinsic_vwmacc_vv_nxv1i64_nxv1i32_nxv1i32(<vscale x 1 x i64> %0, <vscale x 1 x i32> %1, <vscale x 1 x i32> %2, i64 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmacc_vv_nxv1i64_nxv1i32_nxv1i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e32,mf2,ta,mu
+; CHECK-NEXT:    vwmacc.vv v16, v17, v18
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i64> @llvm.riscv.vwmacc.nxv1i64.nxv1i32(
+    <vscale x 1 x i64> %0,
+    <vscale x 1 x i32> %1,
+    <vscale x 1 x i32> %2,
+    i64 %3)
+
+  ret <vscale x 1 x i64> %a
+}
+
+declare <vscale x 1 x i64> @llvm.riscv.vwmacc.mask.nxv1i64.nxv1i32(
+  <vscale x 1 x i64>,
+  <vscale x 1 x i32>,
+  <vscale x 1 x i32>,
+  <vscale x 1 x i1>,
+  i64);
+
+define <vscale x 1 x i64>  @intrinsic_vwmacc_mask_vv_nxv1i64_nxv1i32_nxv1i32(<vscale x 1 x i64> %0, <vscale x 1 x i32> %1, <vscale x 1 x i32> %2, <vscale x 1 x i1> %3, i64 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmacc_mask_vv_nxv1i64_nxv1i32_nxv1i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e32,mf2,ta,mu
+; CHECK-NEXT:    vwmacc.vv v16, v17, v18, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i64> @llvm.riscv.vwmacc.mask.nxv1i64.nxv1i32(
+    <vscale x 1 x i64> %0,
+    <vscale x 1 x i32> %1,
+    <vscale x 1 x i32> %2,
+    <vscale x 1 x i1> %3,
+    i64 %4)
+
+  ret <vscale x 1 x i64> %a
+}
+
+declare <vscale x 2 x i64> @llvm.riscv.vwmacc.nxv2i64.nxv2i32(
+  <vscale x 2 x i64>,
+  <vscale x 2 x i32>,
+  <vscale x 2 x i32>,
+  i64);
+
+define <vscale x 2 x i64>  @intrinsic_vwmacc_vv_nxv2i64_nxv2i32_nxv2i32(<vscale x 2 x i64> %0, <vscale x 2 x i32> %1, <vscale x 2 x i32> %2, i64 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmacc_vv_nxv2i64_nxv2i32_nxv2i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e32,m1,ta,mu
+; CHECK-NEXT:    vwmacc.vv v16, v18, v19
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i64> @llvm.riscv.vwmacc.nxv2i64.nxv2i32(
+    <vscale x 2 x i64> %0,
+    <vscale x 2 x i32> %1,
+    <vscale x 2 x i32> %2,
+    i64 %3)
+
+  ret <vscale x 2 x i64> %a
+}
+
+declare <vscale x 2 x i64> @llvm.riscv.vwmacc.mask.nxv2i64.nxv2i32(
+  <vscale x 2 x i64>,
+  <vscale x 2 x i32>,
+  <vscale x 2 x i32>,
+  <vscale x 2 x i1>,
+  i64);
+
+define <vscale x 2 x i64>  @intrinsic_vwmacc_mask_vv_nxv2i64_nxv2i32_nxv2i32(<vscale x 2 x i64> %0, <vscale x 2 x i32> %1, <vscale x 2 x i32> %2, <vscale x 2 x i1> %3, i64 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmacc_mask_vv_nxv2i64_nxv2i32_nxv2i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e32,m1,ta,mu
+; CHECK-NEXT:    vwmacc.vv v16, v18, v19, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i64> @llvm.riscv.vwmacc.mask.nxv2i64.nxv2i32(
+    <vscale x 2 x i64> %0,
+    <vscale x 2 x i32> %1,
+    <vscale x 2 x i32> %2,
+    <vscale x 2 x i1> %3,
+    i64 %4)
+
+  ret <vscale x 2 x i64> %a
+}
+
+declare <vscale x 4 x i64> @llvm.riscv.vwmacc.nxv4i64.nxv4i32(
+  <vscale x 4 x i64>,
+  <vscale x 4 x i32>,
+  <vscale x 4 x i32>,
+  i64);
+
+define <vscale x 4 x i64>  @intrinsic_vwmacc_vv_nxv4i64_nxv4i32_nxv4i32(<vscale x 4 x i64> %0, <vscale x 4 x i32> %1, <vscale x 4 x i32> %2, i64 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmacc_vv_nxv4i64_nxv4i32_nxv4i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e32,m2,ta,mu
+; CHECK-NEXT:    vwmacc.vv v16, v20, v22
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i64> @llvm.riscv.vwmacc.nxv4i64.nxv4i32(
+    <vscale x 4 x i64> %0,
+    <vscale x 4 x i32> %1,
+    <vscale x 4 x i32> %2,
+    i64 %3)
+
+  ret <vscale x 4 x i64> %a
+}
+
+declare <vscale x 4 x i64> @llvm.riscv.vwmacc.mask.nxv4i64.nxv4i32(
+  <vscale x 4 x i64>,
+  <vscale x 4 x i32>,
+  <vscale x 4 x i32>,
+  <vscale x 4 x i1>,
+  i64);
+
+define <vscale x 4 x i64>  @intrinsic_vwmacc_mask_vv_nxv4i64_nxv4i32_nxv4i32(<vscale x 4 x i64> %0, <vscale x 4 x i32> %1, <vscale x 4 x i32> %2, <vscale x 4 x i1> %3, i64 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmacc_mask_vv_nxv4i64_nxv4i32_nxv4i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e32,m2,ta,mu
+; CHECK-NEXT:    vwmacc.vv v16, v20, v22, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i64> @llvm.riscv.vwmacc.mask.nxv4i64.nxv4i32(
+    <vscale x 4 x i64> %0,
+    <vscale x 4 x i32> %1,
+    <vscale x 4 x i32> %2,
+    <vscale x 4 x i1> %3,
+    i64 %4)
+
+  ret <vscale x 4 x i64> %a
+}
+
+declare <vscale x 8 x i64> @llvm.riscv.vwmacc.nxv8i64.nxv8i32(
+  <vscale x 8 x i64>,
+  <vscale x 8 x i32>,
+  <vscale x 8 x i32>,
+  i64);
+
+define <vscale x 8 x i64>  @intrinsic_vwmacc_vv_nxv8i64_nxv8i32_nxv8i32(<vscale x 8 x i64> %0, <vscale x 8 x i32> %1, <vscale x 8 x i32> %2, i64 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmacc_vv_nxv8i64_nxv8i32_nxv8i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a3, zero, e32,m4,ta,mu
+; CHECK-NEXT:    vle32.v v28, (a1)
+; CHECK-NEXT:    vle32.v v8, (a0)
+; CHECK-NEXT:    vsetvli a0, a2, e32,m4,ta,mu
+; CHECK-NEXT:    vwmacc.vv v16, v8, v28
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i64> @llvm.riscv.vwmacc.nxv8i64.nxv8i32(
+    <vscale x 8 x i64> %0,
+    <vscale x 8 x i32> %1,
+    <vscale x 8 x i32> %2,
+    i64 %3)
+
+  ret <vscale x 8 x i64> %a
+}
+
+declare <vscale x 8 x i64> @llvm.riscv.vwmacc.mask.nxv8i64.nxv8i32(
+  <vscale x 8 x i64>,
+  <vscale x 8 x i32>,
+  <vscale x 8 x i32>,
+  <vscale x 8 x i1>,
+  i64);
+
+define <vscale x 8 x i64>  @intrinsic_vwmacc_mask_vv_nxv8i64_nxv8i32_nxv8i32(<vscale x 8 x i64> %0, <vscale x 8 x i32> %1, <vscale x 8 x i32> %2, <vscale x 8 x i1> %3, i64 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmacc_mask_vv_nxv8i64_nxv8i32_nxv8i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a3, zero, e32,m4,ta,mu
+; CHECK-NEXT:    vle32.v v28, (a1)
+; CHECK-NEXT:    vle32.v v8, (a0)
+; CHECK-NEXT:    vsetvli a0, a2, e32,m4,ta,mu
+; CHECK-NEXT:    vwmacc.vv v16, v8, v28, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i64> @llvm.riscv.vwmacc.mask.nxv8i64.nxv8i32(
+    <vscale x 8 x i64> %0,
+    <vscale x 8 x i32> %1,
+    <vscale x 8 x i32> %2,
+    <vscale x 8 x i1> %3,
+    i64 %4)
+
+  ret <vscale x 8 x i64> %a
+}
+
+declare <vscale x 1 x i16> @llvm.riscv.vwmacc.nxv1i16.i8(
+  <vscale x 1 x i16>,
+  i8,
+  <vscale x 1 x i8>,
+  i64);
+
+define <vscale x 1 x i16>  @intrinsic_vwmacc_vx_nxv1i16_i8_nxv1i8(<vscale x 1 x i16> %0, i8 %1, <vscale x 1 x i8> %2, i64 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmacc_vx_nxv1i16_i8_nxv1i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e8,mf8,ta,mu
+; CHECK-NEXT:    vwmacc.vx v16, a0, v17
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i16> @llvm.riscv.vwmacc.nxv1i16.i8(
+    <vscale x 1 x i16> %0,
+    i8 %1,
+    <vscale x 1 x i8> %2,
+    i64 %3)
+
+  ret <vscale x 1 x i16> %a
+}
+
+declare <vscale x 1 x i16> @llvm.riscv.vwmacc.mask.nxv1i16.i8(
+  <vscale x 1 x i16>,
+  i8,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i1>,
+  i64);
+
+define <vscale x 1 x i16> @intrinsic_vwmacc_mask_vx_nxv1i16_i8_nxv1i8(<vscale x 1 x i16> %0, i8 %1, <vscale x 1 x i8> %2, <vscale x 1 x i1> %3, i64 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmacc_mask_vx_nxv1i16_i8_nxv1i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e8,mf8,ta,mu
+; CHECK-NEXT:    vwmacc.vx v16, a0, v17, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i16> @llvm.riscv.vwmacc.mask.nxv1i16.i8(
+    <vscale x 1 x i16> %0,
+    i8 %1,
+    <vscale x 1 x i8> %2,
+    <vscale x 1 x i1> %3,
+    i64 %4)
+
+  ret <vscale x 1 x i16> %a
+}
+
+declare <vscale x 2 x i16> @llvm.riscv.vwmacc.nxv2i16.i8(
+  <vscale x 2 x i16>,
+  i8,
+  <vscale x 2 x i8>,
+  i64);
+
+define <vscale x 2 x i16>  @intrinsic_vwmacc_vx_nxv2i16_i8_nxv2i8(<vscale x 2 x i16> %0, i8 %1, <vscale x 2 x i8> %2, i64 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmacc_vx_nxv2i16_i8_nxv2i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e8,mf4,ta,mu
+; CHECK-NEXT:    vwmacc.vx v16, a0, v17
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i16> @llvm.riscv.vwmacc.nxv2i16.i8(
+    <vscale x 2 x i16> %0,
+    i8 %1,
+    <vscale x 2 x i8> %2,
+    i64 %3)
+
+  ret <vscale x 2 x i16> %a
+}
+
+declare <vscale x 2 x i16> @llvm.riscv.vwmacc.mask.nxv2i16.i8(
+  <vscale x 2 x i16>,
+  i8,
+  <vscale x 2 x i8>,
+  <vscale x 2 x i1>,
+  i64);
+
+define <vscale x 2 x i16> @intrinsic_vwmacc_mask_vx_nxv2i16_i8_nxv2i8(<vscale x 2 x i16> %0, i8 %1, <vscale x 2 x i8> %2, <vscale x 2 x i1> %3, i64 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmacc_mask_vx_nxv2i16_i8_nxv2i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e8,mf4,ta,mu
+; CHECK-NEXT:    vwmacc.vx v16, a0, v17, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i16> @llvm.riscv.vwmacc.mask.nxv2i16.i8(
+    <vscale x 2 x i16> %0,
+    i8 %1,
+    <vscale x 2 x i8> %2,
+    <vscale x 2 x i1> %3,
+    i64 %4)
+
+  ret <vscale x 2 x i16> %a
+}
+
+declare <vscale x 4 x i16> @llvm.riscv.vwmacc.nxv4i16.i8(
+  <vscale x 4 x i16>,
+  i8,
+  <vscale x 4 x i8>,
+  i64);
+
+define <vscale x 4 x i16>  @intrinsic_vwmacc_vx_nxv4i16_i8_nxv4i8(<vscale x 4 x i16> %0, i8 %1, <vscale x 4 x i8> %2, i64 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmacc_vx_nxv4i16_i8_nxv4i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e8,mf2,ta,mu
+; CHECK-NEXT:    vwmacc.vx v16, a0, v17
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i16> @llvm.riscv.vwmacc.nxv4i16.i8(
+    <vscale x 4 x i16> %0,
+    i8 %1,
+    <vscale x 4 x i8> %2,
+    i64 %3)
+
+  ret <vscale x 4 x i16> %a
+}
+
+declare <vscale x 4 x i16> @llvm.riscv.vwmacc.mask.nxv4i16.i8(
+  <vscale x 4 x i16>,
+  i8,
+  <vscale x 4 x i8>,
+  <vscale x 4 x i1>,
+  i64);
+
+define <vscale x 4 x i16> @intrinsic_vwmacc_mask_vx_nxv4i16_i8_nxv4i8(<vscale x 4 x i16> %0, i8 %1, <vscale x 4 x i8> %2, <vscale x 4 x i1> %3, i64 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmacc_mask_vx_nxv4i16_i8_nxv4i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e8,mf2,ta,mu
+; CHECK-NEXT:    vwmacc.vx v16, a0, v17, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i16> @llvm.riscv.vwmacc.mask.nxv4i16.i8(
+    <vscale x 4 x i16> %0,
+    i8 %1,
+    <vscale x 4 x i8> %2,
+    <vscale x 4 x i1> %3,
+    i64 %4)
+
+  ret <vscale x 4 x i16> %a
+}
+
+declare <vscale x 8 x i16> @llvm.riscv.vwmacc.nxv8i16.i8(
+  <vscale x 8 x i16>,
+  i8,
+  <vscale x 8 x i8>,
+  i64);
+
+define <vscale x 8 x i16>  @intrinsic_vwmacc_vx_nxv8i16_i8_nxv8i8(<vscale x 8 x i16> %0, i8 %1, <vscale x 8 x i8> %2, i64 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmacc_vx_nxv8i16_i8_nxv8i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e8,m1,ta,mu
+; CHECK-NEXT:    vwmacc.vx v16, a0, v18
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i16> @llvm.riscv.vwmacc.nxv8i16.i8(
+    <vscale x 8 x i16> %0,
+    i8 %1,
+    <vscale x 8 x i8> %2,
+    i64 %3)
+
+  ret <vscale x 8 x i16> %a
+}
+
+declare <vscale x 8 x i16> @llvm.riscv.vwmacc.mask.nxv8i16.i8(
+  <vscale x 8 x i16>,
+  i8,
+  <vscale x 8 x i8>,
+  <vscale x 8 x i1>,
+  i64);
+
+define <vscale x 8 x i16> @intrinsic_vwmacc_mask_vx_nxv8i16_i8_nxv8i8(<vscale x 8 x i16> %0, i8 %1, <vscale x 8 x i8> %2, <vscale x 8 x i1> %3, i64 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmacc_mask_vx_nxv8i16_i8_nxv8i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e8,m1,ta,mu
+; CHECK-NEXT:    vwmacc.vx v16, a0, v18, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i16> @llvm.riscv.vwmacc.mask.nxv8i16.i8(
+    <vscale x 8 x i16> %0,
+    i8 %1,
+    <vscale x 8 x i8> %2,
+    <vscale x 8 x i1> %3,
+    i64 %4)
+
+  ret <vscale x 8 x i16> %a
+}
+
+declare <vscale x 16 x i16> @llvm.riscv.vwmacc.nxv16i16.i8(
+  <vscale x 16 x i16>,
+  i8,
+  <vscale x 16 x i8>,
+  i64);
+
+define <vscale x 16 x i16>  @intrinsic_vwmacc_vx_nxv16i16_i8_nxv16i8(<vscale x 16 x i16> %0, i8 %1, <vscale x 16 x i8> %2, i64 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmacc_vx_nxv16i16_i8_nxv16i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e8,m2,ta,mu
+; CHECK-NEXT:    vwmacc.vx v16, a0, v20
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 16 x i16> @llvm.riscv.vwmacc.nxv16i16.i8(
+    <vscale x 16 x i16> %0,
+    i8 %1,
+    <vscale x 16 x i8> %2,
+    i64 %3)
+
+  ret <vscale x 16 x i16> %a
+}
+
+declare <vscale x 16 x i16> @llvm.riscv.vwmacc.mask.nxv16i16.i8(
+  <vscale x 16 x i16>,
+  i8,
+  <vscale x 16 x i8>,
+  <vscale x 16 x i1>,
+  i64);
+
+define <vscale x 16 x i16> @intrinsic_vwmacc_mask_vx_nxv16i16_i8_nxv16i8(<vscale x 16 x i16> %0, i8 %1, <vscale x 16 x i8> %2, <vscale x 16 x i1> %3, i64 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmacc_mask_vx_nxv16i16_i8_nxv16i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e8,m2,ta,mu
+; CHECK-NEXT:    vwmacc.vx v16, a0, v20, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 16 x i16> @llvm.riscv.vwmacc.mask.nxv16i16.i8(
+    <vscale x 16 x i16> %0,
+    i8 %1,
+    <vscale x 16 x i8> %2,
+    <vscale x 16 x i1> %3,
+    i64 %4)
+
+  ret <vscale x 16 x i16> %a
+}
+
+declare <vscale x 32 x i16> @llvm.riscv.vwmacc.nxv32i16.i8(
+  <vscale x 32 x i16>,
+  i8,
+  <vscale x 32 x i8>,
+  i64);
+
+define <vscale x 32 x i16>  @intrinsic_vwmacc_vx_nxv32i16_i8_nxv32i8(<vscale x 32 x i16> %0, i8 %1, <vscale x 32 x i8> %2, i64 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmacc_vx_nxv32i16_i8_nxv32i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a3, zero, e8,m4,ta,mu
+; CHECK-NEXT:    vle8.v v28, (a1)
+; CHECK-NEXT:    vsetvli a1, a2, e8,m4,ta,mu
+; CHECK-NEXT:    vwmacc.vx v16, a0, v28
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 32 x i16> @llvm.riscv.vwmacc.nxv32i16.i8(
+    <vscale x 32 x i16> %0,
+    i8 %1,
+    <vscale x 32 x i8> %2,
+    i64 %3)
+
+  ret <vscale x 32 x i16> %a
+}
+
+declare <vscale x 32 x i16> @llvm.riscv.vwmacc.mask.nxv32i16.i8(
+  <vscale x 32 x i16>,
+  i8,
+  <vscale x 32 x i8>,
+  <vscale x 32 x i1>,
+  i64);
+
+define <vscale x 32 x i16> @intrinsic_vwmacc_mask_vx_nxv32i16_i8_nxv32i8(<vscale x 32 x i16> %0, i8 %1, <vscale x 32 x i8> %2, <vscale x 32 x i1> %3, i64 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmacc_mask_vx_nxv32i16_i8_nxv32i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a3, zero, e8,m4,ta,mu
+; CHECK-NEXT:    vle8.v v28, (a1)
+; CHECK-NEXT:    vsetvli a1, a2, e8,m4,ta,mu
+; CHECK-NEXT:    vwmacc.vx v16, a0, v28, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 32 x i16> @llvm.riscv.vwmacc.mask.nxv32i16.i8(
+    <vscale x 32 x i16> %0,
+    i8 %1,
+    <vscale x 32 x i8> %2,
+    <vscale x 32 x i1> %3,
+    i64 %4)
+
+  ret <vscale x 32 x i16> %a
+}
+
+declare <vscale x 1 x i32> @llvm.riscv.vwmacc.nxv1i32.i16(
+  <vscale x 1 x i32>,
+  i16,
+  <vscale x 1 x i16>,
+  i64);
+
+define <vscale x 1 x i32>  @intrinsic_vwmacc_vx_nxv1i32_i16_nxv1i16(<vscale x 1 x i32> %0, i16 %1, <vscale x 1 x i16> %2, i64 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmacc_vx_nxv1i32_i16_nxv1i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e16,mf4,ta,mu
+; CHECK-NEXT:    vwmacc.vx v16, a0, v17
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i32> @llvm.riscv.vwmacc.nxv1i32.i16(
+    <vscale x 1 x i32> %0,
+    i16 %1,
+    <vscale x 1 x i16> %2,
+    i64 %3)
+
+  ret <vscale x 1 x i32> %a
+}
+
+declare <vscale x 1 x i32> @llvm.riscv.vwmacc.mask.nxv1i32.i16(
+  <vscale x 1 x i32>,
+  i16,
+  <vscale x 1 x i16>,
+  <vscale x 1 x i1>,
+  i64);
+
+define <vscale x 1 x i32> @intrinsic_vwmacc_mask_vx_nxv1i32_i16_nxv1i16(<vscale x 1 x i32> %0, i16 %1, <vscale x 1 x i16> %2, <vscale x 1 x i1> %3, i64 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmacc_mask_vx_nxv1i32_i16_nxv1i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e16,mf4,ta,mu
+; CHECK-NEXT:    vwmacc.vx v16, a0, v17, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i32> @llvm.riscv.vwmacc.mask.nxv1i32.i16(
+    <vscale x 1 x i32> %0,
+    i16 %1,
+    <vscale x 1 x i16> %2,
+    <vscale x 1 x i1> %3,
+    i64 %4)
+
+  ret <vscale x 1 x i32> %a
+}
+
+declare <vscale x 2 x i32> @llvm.riscv.vwmacc.nxv2i32.i16(
+  <vscale x 2 x i32>,
+  i16,
+  <vscale x 2 x i16>,
+  i64);
+
+define <vscale x 2 x i32>  @intrinsic_vwmacc_vx_nxv2i32_i16_nxv2i16(<vscale x 2 x i32> %0, i16 %1, <vscale x 2 x i16> %2, i64 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmacc_vx_nxv2i32_i16_nxv2i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e16,mf2,ta,mu
+; CHECK-NEXT:    vwmacc.vx v16, a0, v17
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i32> @llvm.riscv.vwmacc.nxv2i32.i16(
+    <vscale x 2 x i32> %0,
+    i16 %1,
+    <vscale x 2 x i16> %2,
+    i64 %3)
+
+  ret <vscale x 2 x i32> %a
+}
+
+declare <vscale x 2 x i32> @llvm.riscv.vwmacc.mask.nxv2i32.i16(
+  <vscale x 2 x i32>,
+  i16,
+  <vscale x 2 x i16>,
+  <vscale x 2 x i1>,
+  i64);
+
+define <vscale x 2 x i32> @intrinsic_vwmacc_mask_vx_nxv2i32_i16_nxv2i16(<vscale x 2 x i32> %0, i16 %1, <vscale x 2 x i16> %2, <vscale x 2 x i1> %3, i64 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmacc_mask_vx_nxv2i32_i16_nxv2i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e16,mf2,ta,mu
+; CHECK-NEXT:    vwmacc.vx v16, a0, v17, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i32> @llvm.riscv.vwmacc.mask.nxv2i32.i16(
+    <vscale x 2 x i32> %0,
+    i16 %1,
+    <vscale x 2 x i16> %2,
+    <vscale x 2 x i1> %3,
+    i64 %4)
+
+  ret <vscale x 2 x i32> %a
+}
+
+declare <vscale x 4 x i32> @llvm.riscv.vwmacc.nxv4i32.i16(
+  <vscale x 4 x i32>,
+  i16,
+  <vscale x 4 x i16>,
+  i64);
+
+define <vscale x 4 x i32>  @intrinsic_vwmacc_vx_nxv4i32_i16_nxv4i16(<vscale x 4 x i32> %0, i16 %1, <vscale x 4 x i16> %2, i64 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmacc_vx_nxv4i32_i16_nxv4i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e16,m1,ta,mu
+; CHECK-NEXT:    vwmacc.vx v16, a0, v18
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i32> @llvm.riscv.vwmacc.nxv4i32.i16(
+    <vscale x 4 x i32> %0,
+    i16 %1,
+    <vscale x 4 x i16> %2,
+    i64 %3)
+
+  ret <vscale x 4 x i32> %a
+}
+
+declare <vscale x 4 x i32> @llvm.riscv.vwmacc.mask.nxv4i32.i16(
+  <vscale x 4 x i32>,
+  i16,
+  <vscale x 4 x i16>,
+  <vscale x 4 x i1>,
+  i64);
+
+define <vscale x 4 x i32> @intrinsic_vwmacc_mask_vx_nxv4i32_i16_nxv4i16(<vscale x 4 x i32> %0, i16 %1, <vscale x 4 x i16> %2, <vscale x 4 x i1> %3, i64 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmacc_mask_vx_nxv4i32_i16_nxv4i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e16,m1,ta,mu
+; CHECK-NEXT:    vwmacc.vx v16, a0, v18, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i32> @llvm.riscv.vwmacc.mask.nxv4i32.i16(
+    <vscale x 4 x i32> %0,
+    i16 %1,
+    <vscale x 4 x i16> %2,
+    <vscale x 4 x i1> %3,
+    i64 %4)
+
+  ret <vscale x 4 x i32> %a
+}
+
+declare <vscale x 8 x i32> @llvm.riscv.vwmacc.nxv8i32.i16(
+  <vscale x 8 x i32>,
+  i16,
+  <vscale x 8 x i16>,
+  i64);
+
+define <vscale x 8 x i32>  @intrinsic_vwmacc_vx_nxv8i32_i16_nxv8i16(<vscale x 8 x i32> %0, i16 %1, <vscale x 8 x i16> %2, i64 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmacc_vx_nxv8i32_i16_nxv8i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e16,m2,ta,mu
+; CHECK-NEXT:    vwmacc.vx v16, a0, v20
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i32> @llvm.riscv.vwmacc.nxv8i32.i16(
+    <vscale x 8 x i32> %0,
+    i16 %1,
+    <vscale x 8 x i16> %2,
+    i64 %3)
+
+  ret <vscale x 8 x i32> %a
+}
+
+declare <vscale x 8 x i32> @llvm.riscv.vwmacc.mask.nxv8i32.i16(
+  <vscale x 8 x i32>,
+  i16,
+  <vscale x 8 x i16>,
+  <vscale x 8 x i1>,
+  i64);
+
+define <vscale x 8 x i32> @intrinsic_vwmacc_mask_vx_nxv8i32_i16_nxv8i16(<vscale x 8 x i32> %0, i16 %1, <vscale x 8 x i16> %2, <vscale x 8 x i1> %3, i64 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmacc_mask_vx_nxv8i32_i16_nxv8i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e16,m2,ta,mu
+; CHECK-NEXT:    vwmacc.vx v16, a0, v20, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i32> @llvm.riscv.vwmacc.mask.nxv8i32.i16(
+    <vscale x 8 x i32> %0,
+    i16 %1,
+    <vscale x 8 x i16> %2,
+    <vscale x 8 x i1> %3,
+    i64 %4)
+
+  ret <vscale x 8 x i32> %a
+}
+
+declare <vscale x 16 x i32> @llvm.riscv.vwmacc.nxv16i32.i16(
+  <vscale x 16 x i32>,
+  i16,
+  <vscale x 16 x i16>,
+  i64);
+
+define <vscale x 16 x i32>  @intrinsic_vwmacc_vx_nxv16i32_i16_nxv16i16(<vscale x 16 x i32> %0, i16 %1, <vscale x 16 x i16> %2, i64 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmacc_vx_nxv16i32_i16_nxv16i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a3, zero, e16,m4,ta,mu
+; CHECK-NEXT:    vle16.v v28, (a1)
+; CHECK-NEXT:    vsetvli a1, a2, e16,m4,ta,mu
+; CHECK-NEXT:    vwmacc.vx v16, a0, v28
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 16 x i32> @llvm.riscv.vwmacc.nxv16i32.i16(
+    <vscale x 16 x i32> %0,
+    i16 %1,
+    <vscale x 16 x i16> %2,
+    i64 %3)
+
+  ret <vscale x 16 x i32> %a
+}
+
+declare <vscale x 16 x i32> @llvm.riscv.vwmacc.mask.nxv16i32.i16(
+  <vscale x 16 x i32>,
+  i16,
+  <vscale x 16 x i16>,
+  <vscale x 16 x i1>,
+  i64);
+
+define <vscale x 16 x i32> @intrinsic_vwmacc_mask_vx_nxv16i32_i16_nxv16i16(<vscale x 16 x i32> %0, i16 %1, <vscale x 16 x i16> %2, <vscale x 16 x i1> %3, i64 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmacc_mask_vx_nxv16i32_i16_nxv16i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a3, zero, e16,m4,ta,mu
+; CHECK-NEXT:    vle16.v v28, (a1)
+; CHECK-NEXT:    vsetvli a1, a2, e16,m4,ta,mu
+; CHECK-NEXT:    vwmacc.vx v16, a0, v28, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 16 x i32> @llvm.riscv.vwmacc.mask.nxv16i32.i16(
+    <vscale x 16 x i32> %0,
+    i16 %1,
+    <vscale x 16 x i16> %2,
+    <vscale x 16 x i1> %3,
+    i64 %4)
+
+  ret <vscale x 16 x i32> %a
+}
+
+declare <vscale x 1 x i64> @llvm.riscv.vwmacc.nxv1i64.i32(
+  <vscale x 1 x i64>,
+  i32,
+  <vscale x 1 x i32>,
+  i64);
+
+define <vscale x 1 x i64>  @intrinsic_vwmacc_vx_nxv1i64_i32_nxv1i32(<vscale x 1 x i64> %0, i32 %1, <vscale x 1 x i32> %2, i64 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmacc_vx_nxv1i64_i32_nxv1i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e32,mf2,ta,mu
+; CHECK-NEXT:    vwmacc.vx v16, a0, v17
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i64> @llvm.riscv.vwmacc.nxv1i64.i32(
+    <vscale x 1 x i64> %0,
+    i32 %1,
+    <vscale x 1 x i32> %2,
+    i64 %3)
+
+  ret <vscale x 1 x i64> %a
+}
+
+declare <vscale x 1 x i64> @llvm.riscv.vwmacc.mask.nxv1i64.i32(
+  <vscale x 1 x i64>,
+  i32,
+  <vscale x 1 x i32>,
+  <vscale x 1 x i1>,
+  i64);
+
+define <vscale x 1 x i64> @intrinsic_vwmacc_mask_vx_nxv1i64_i32_nxv1i32(<vscale x 1 x i64> %0, i32 %1, <vscale x 1 x i32> %2, <vscale x 1 x i1> %3, i64 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmacc_mask_vx_nxv1i64_i32_nxv1i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e32,mf2,ta,mu
+; CHECK-NEXT:    vwmacc.vx v16, a0, v17, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i64> @llvm.riscv.vwmacc.mask.nxv1i64.i32(
+    <vscale x 1 x i64> %0,
+    i32 %1,
+    <vscale x 1 x i32> %2,
+    <vscale x 1 x i1> %3,
+    i64 %4)
+
+  ret <vscale x 1 x i64> %a
+}
+
+declare <vscale x 2 x i64> @llvm.riscv.vwmacc.nxv2i64.i32(
+  <vscale x 2 x i64>,
+  i32,
+  <vscale x 2 x i32>,
+  i64);
+
+define <vscale x 2 x i64>  @intrinsic_vwmacc_vx_nxv2i64_i32_nxv2i32(<vscale x 2 x i64> %0, i32 %1, <vscale x 2 x i32> %2, i64 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmacc_vx_nxv2i64_i32_nxv2i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e32,m1,ta,mu
+; CHECK-NEXT:    vwmacc.vx v16, a0, v18
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i64> @llvm.riscv.vwmacc.nxv2i64.i32(
+    <vscale x 2 x i64> %0,
+    i32 %1,
+    <vscale x 2 x i32> %2,
+    i64 %3)
+
+  ret <vscale x 2 x i64> %a
+}
+
+declare <vscale x 2 x i64> @llvm.riscv.vwmacc.mask.nxv2i64.i32(
+  <vscale x 2 x i64>,
+  i32,
+  <vscale x 2 x i32>,
+  <vscale x 2 x i1>,
+  i64);
+
+define <vscale x 2 x i64> @intrinsic_vwmacc_mask_vx_nxv2i64_i32_nxv2i32(<vscale x 2 x i64> %0, i32 %1, <vscale x 2 x i32> %2, <vscale x 2 x i1> %3, i64 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmacc_mask_vx_nxv2i64_i32_nxv2i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e32,m1,ta,mu
+; CHECK-NEXT:    vwmacc.vx v16, a0, v18, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i64> @llvm.riscv.vwmacc.mask.nxv2i64.i32(
+    <vscale x 2 x i64> %0,
+    i32 %1,
+    <vscale x 2 x i32> %2,
+    <vscale x 2 x i1> %3,
+    i64 %4)
+
+  ret <vscale x 2 x i64> %a
+}
+
+declare <vscale x 4 x i64> @llvm.riscv.vwmacc.nxv4i64.i32(
+  <vscale x 4 x i64>,
+  i32,
+  <vscale x 4 x i32>,
+  i64);
+
+define <vscale x 4 x i64>  @intrinsic_vwmacc_vx_nxv4i64_i32_nxv4i32(<vscale x 4 x i64> %0, i32 %1, <vscale x 4 x i32> %2, i64 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmacc_vx_nxv4i64_i32_nxv4i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e32,m2,ta,mu
+; CHECK-NEXT:    vwmacc.vx v16, a0, v20
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i64> @llvm.riscv.vwmacc.nxv4i64.i32(
+    <vscale x 4 x i64> %0,
+    i32 %1,
+    <vscale x 4 x i32> %2,
+    i64 %3)
+
+  ret <vscale x 4 x i64> %a
+}
+
+declare <vscale x 4 x i64> @llvm.riscv.vwmacc.mask.nxv4i64.i32(
+  <vscale x 4 x i64>,
+  i32,
+  <vscale x 4 x i32>,
+  <vscale x 4 x i1>,
+  i64);
+
+define <vscale x 4 x i64> @intrinsic_vwmacc_mask_vx_nxv4i64_i32_nxv4i32(<vscale x 4 x i64> %0, i32 %1, <vscale x 4 x i32> %2, <vscale x 4 x i1> %3, i64 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmacc_mask_vx_nxv4i64_i32_nxv4i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e32,m2,ta,mu
+; CHECK-NEXT:    vwmacc.vx v16, a0, v20, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i64> @llvm.riscv.vwmacc.mask.nxv4i64.i32(
+    <vscale x 4 x i64> %0,
+    i32 %1,
+    <vscale x 4 x i32> %2,
+    <vscale x 4 x i1> %3,
+    i64 %4)
+
+  ret <vscale x 4 x i64> %a
+}
+
+declare <vscale x 8 x i64> @llvm.riscv.vwmacc.nxv8i64.i32(
+  <vscale x 8 x i64>,
+  i32,
+  <vscale x 8 x i32>,
+  i64);
+
+define <vscale x 8 x i64>  @intrinsic_vwmacc_vx_nxv8i64_i32_nxv8i32(<vscale x 8 x i64> %0, i32 %1, <vscale x 8 x i32> %2, i64 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmacc_vx_nxv8i64_i32_nxv8i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a3, zero, e32,m4,ta,mu
+; CHECK-NEXT:    vle32.v v28, (a1)
+; CHECK-NEXT:    vsetvli a1, a2, e32,m4,ta,mu
+; CHECK-NEXT:    vwmacc.vx v16, a0, v28
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i64> @llvm.riscv.vwmacc.nxv8i64.i32(
+    <vscale x 8 x i64> %0,
+    i32 %1,
+    <vscale x 8 x i32> %2,
+    i64 %3)
+
+  ret <vscale x 8 x i64> %a
+}
+
+declare <vscale x 8 x i64> @llvm.riscv.vwmacc.mask.nxv8i64.i32(
+  <vscale x 8 x i64>,
+  i32,
+  <vscale x 8 x i32>,
+  <vscale x 8 x i1>,
+  i64);
+
+define <vscale x 8 x i64> @intrinsic_vwmacc_mask_vx_nxv8i64_i32_nxv8i32(<vscale x 8 x i64> %0, i32 %1, <vscale x 8 x i32> %2, <vscale x 8 x i1> %3, i64 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmacc_mask_vx_nxv8i64_i32_nxv8i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a3, zero, e32,m4,ta,mu
+; CHECK-NEXT:    vle32.v v28, (a1)
+; CHECK-NEXT:    vsetvli a1, a2, e32,m4,ta,mu
+; CHECK-NEXT:    vwmacc.vx v16, a0, v28, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i64> @llvm.riscv.vwmacc.mask.nxv8i64.i32(
+    <vscale x 8 x i64> %0,
+    i32 %1,
+    <vscale x 8 x i32> %2,
+    <vscale x 8 x i1> %3,
+    i64 %4)
+
+  ret <vscale x 8 x i64> %a
+}

llvm/test/CodeGen/RISCV/rvv/vwmaccsu-rv32.ll

+; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
+; RUN: llc -mtriple=riscv32 -mattr=+experimental-v,+f -verify-machineinstrs \
+; RUN:   --riscv-no-aliases < %s | FileCheck %s
+declare <vscale x 1 x i16> @llvm.riscv.vwmaccsu.nxv1i16.nxv1i8(
+  <vscale x 1 x i16>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  i32);
+
+define <vscale x 1 x i16>  @intrinsic_vwmaccsu_vv_nxv1i16_nxv1i8_nxv1i8(<vscale x 1 x i16> %0, <vscale x 1 x i8> %1, <vscale x 1 x i8> %2, i32 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccsu_vv_nxv1i16_nxv1i8_nxv1i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e8,mf8,ta,mu
+; CHECK-NEXT:    vwmaccsu.vv v16, v17, v18
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i16> @llvm.riscv.vwmaccsu.nxv1i16.nxv1i8(
+    <vscale x 1 x i16> %0,
+    <vscale x 1 x i8> %1,
+    <vscale x 1 x i8> %2,
+    i32 %3)
+
+  ret <vscale x 1 x i16> %a
+}
+
+declare <vscale x 1 x i16> @llvm.riscv.vwmaccsu.mask.nxv1i16.nxv1i8(
+  <vscale x 1 x i16>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i1>,
+  i32);
+
+define <vscale x 1 x i16>  @intrinsic_vwmaccsu_mask_vv_nxv1i16_nxv1i8_nxv1i8(<vscale x 1 x i16> %0, <vscale x 1 x i8> %1, <vscale x 1 x i8> %2, <vscale x 1 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccsu_mask_vv_nxv1i16_nxv1i8_nxv1i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e8,mf8,ta,mu
+; CHECK-NEXT:    vwmaccsu.vv v16, v17, v18, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i16> @llvm.riscv.vwmaccsu.mask.nxv1i16.nxv1i8(
+    <vscale x 1 x i16> %0,
+    <vscale x 1 x i8> %1,
+    <vscale x 1 x i8> %2,
+    <vscale x 1 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 1 x i16> %a
+}
+
+declare <vscale x 2 x i16> @llvm.riscv.vwmaccsu.nxv2i16.nxv2i8(
+  <vscale x 2 x i16>,
+  <vscale x 2 x i8>,
+  <vscale x 2 x i8>,
+  i32);
+
+define <vscale x 2 x i16>  @intrinsic_vwmaccsu_vv_nxv2i16_nxv2i8_nxv2i8(<vscale x 2 x i16> %0, <vscale x 2 x i8> %1, <vscale x 2 x i8> %2, i32 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccsu_vv_nxv2i16_nxv2i8_nxv2i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e8,mf4,ta,mu
+; CHECK-NEXT:    vwmaccsu.vv v16, v17, v18
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i16> @llvm.riscv.vwmaccsu.nxv2i16.nxv2i8(
+    <vscale x 2 x i16> %0,
+    <vscale x 2 x i8> %1,
+    <vscale x 2 x i8> %2,
+    i32 %3)
+
+  ret <vscale x 2 x i16> %a
+}
+
+declare <vscale x 2 x i16> @llvm.riscv.vwmaccsu.mask.nxv2i16.nxv2i8(
+  <vscale x 2 x i16>,
+  <vscale x 2 x i8>,
+  <vscale x 2 x i8>,
+  <vscale x 2 x i1>,
+  i32);
+
+define <vscale x 2 x i16>  @intrinsic_vwmaccsu_mask_vv_nxv2i16_nxv2i8_nxv2i8(<vscale x 2 x i16> %0, <vscale x 2 x i8> %1, <vscale x 2 x i8> %2, <vscale x 2 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccsu_mask_vv_nxv2i16_nxv2i8_nxv2i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e8,mf4,ta,mu
+; CHECK-NEXT:    vwmaccsu.vv v16, v17, v18, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i16> @llvm.riscv.vwmaccsu.mask.nxv2i16.nxv2i8(
+    <vscale x 2 x i16> %0,
+    <vscale x 2 x i8> %1,
+    <vscale x 2 x i8> %2,
+    <vscale x 2 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 2 x i16> %a
+}
+
+declare <vscale x 4 x i16> @llvm.riscv.vwmaccsu.nxv4i16.nxv4i8(
+  <vscale x 4 x i16>,
+  <vscale x 4 x i8>,
+  <vscale x 4 x i8>,
+  i32);
+
+define <vscale x 4 x i16>  @intrinsic_vwmaccsu_vv_nxv4i16_nxv4i8_nxv4i8(<vscale x 4 x i16> %0, <vscale x 4 x i8> %1, <vscale x 4 x i8> %2, i32 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccsu_vv_nxv4i16_nxv4i8_nxv4i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e8,mf2,ta,mu
+; CHECK-NEXT:    vwmaccsu.vv v16, v17, v18
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i16> @llvm.riscv.vwmaccsu.nxv4i16.nxv4i8(
+    <vscale x 4 x i16> %0,
+    <vscale x 4 x i8> %1,
+    <vscale x 4 x i8> %2,
+    i32 %3)
+
+  ret <vscale x 4 x i16> %a
+}
+
+declare <vscale x 4 x i16> @llvm.riscv.vwmaccsu.mask.nxv4i16.nxv4i8(
+  <vscale x 4 x i16>,
+  <vscale x 4 x i8>,
+  <vscale x 4 x i8>,
+  <vscale x 4 x i1>,
+  i32);
+
+define <vscale x 4 x i16>  @intrinsic_vwmaccsu_mask_vv_nxv4i16_nxv4i8_nxv4i8(<vscale x 4 x i16> %0, <vscale x 4 x i8> %1, <vscale x 4 x i8> %2, <vscale x 4 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccsu_mask_vv_nxv4i16_nxv4i8_nxv4i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e8,mf2,ta,mu
+; CHECK-NEXT:    vwmaccsu.vv v16, v17, v18, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i16> @llvm.riscv.vwmaccsu.mask.nxv4i16.nxv4i8(
+    <vscale x 4 x i16> %0,
+    <vscale x 4 x i8> %1,
+    <vscale x 4 x i8> %2,
+    <vscale x 4 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 4 x i16> %a
+}
+
+declare <vscale x 8 x i16> @llvm.riscv.vwmaccsu.nxv8i16.nxv8i8(
+  <vscale x 8 x i16>,
+  <vscale x 8 x i8>,
+  <vscale x 8 x i8>,
+  i32);
+
+define <vscale x 8 x i16>  @intrinsic_vwmaccsu_vv_nxv8i16_nxv8i8_nxv8i8(<vscale x 8 x i16> %0, <vscale x 8 x i8> %1, <vscale x 8 x i8> %2, i32 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccsu_vv_nxv8i16_nxv8i8_nxv8i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e8,m1,ta,mu
+; CHECK-NEXT:    vwmaccsu.vv v16, v18, v19
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i16> @llvm.riscv.vwmaccsu.nxv8i16.nxv8i8(
+    <vscale x 8 x i16> %0,
+    <vscale x 8 x i8> %1,
+    <vscale x 8 x i8> %2,
+    i32 %3)
+
+  ret <vscale x 8 x i16> %a
+}
+
+declare <vscale x 8 x i16> @llvm.riscv.vwmaccsu.mask.nxv8i16.nxv8i8(
+  <vscale x 8 x i16>,
+  <vscale x 8 x i8>,
+  <vscale x 8 x i8>,
+  <vscale x 8 x i1>,
+  i32);
+
+define <vscale x 8 x i16>  @intrinsic_vwmaccsu_mask_vv_nxv8i16_nxv8i8_nxv8i8(<vscale x 8 x i16> %0, <vscale x 8 x i8> %1, <vscale x 8 x i8> %2, <vscale x 8 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccsu_mask_vv_nxv8i16_nxv8i8_nxv8i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e8,m1,ta,mu
+; CHECK-NEXT:    vwmaccsu.vv v16, v18, v19, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i16> @llvm.riscv.vwmaccsu.mask.nxv8i16.nxv8i8(
+    <vscale x 8 x i16> %0,
+    <vscale x 8 x i8> %1,
+    <vscale x 8 x i8> %2,
+    <vscale x 8 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 8 x i16> %a
+}
+
+declare <vscale x 16 x i16> @llvm.riscv.vwmaccsu.nxv16i16.nxv16i8(
+  <vscale x 16 x i16>,
+  <vscale x 16 x i8>,
+  <vscale x 16 x i8>,
+  i32);
+
+define <vscale x 16 x i16>  @intrinsic_vwmaccsu_vv_nxv16i16_nxv16i8_nxv16i8(<vscale x 16 x i16> %0, <vscale x 16 x i8> %1, <vscale x 16 x i8> %2, i32 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccsu_vv_nxv16i16_nxv16i8_nxv16i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e8,m2,ta,mu
+; CHECK-NEXT:    vwmaccsu.vv v16, v20, v22
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 16 x i16> @llvm.riscv.vwmaccsu.nxv16i16.nxv16i8(
+    <vscale x 16 x i16> %0,
+    <vscale x 16 x i8> %1,
+    <vscale x 16 x i8> %2,
+    i32 %3)
+
+  ret <vscale x 16 x i16> %a
+}
+
+declare <vscale x 16 x i16> @llvm.riscv.vwmaccsu.mask.nxv16i16.nxv16i8(
+  <vscale x 16 x i16>,
+  <vscale x 16 x i8>,
+  <vscale x 16 x i8>,
+  <vscale x 16 x i1>,
+  i32);
+
+define <vscale x 16 x i16>  @intrinsic_vwmaccsu_mask_vv_nxv16i16_nxv16i8_nxv16i8(<vscale x 16 x i16> %0, <vscale x 16 x i8> %1, <vscale x 16 x i8> %2, <vscale x 16 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccsu_mask_vv_nxv16i16_nxv16i8_nxv16i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e8,m2,ta,mu
+; CHECK-NEXT:    vwmaccsu.vv v16, v20, v22, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 16 x i16> @llvm.riscv.vwmaccsu.mask.nxv16i16.nxv16i8(
+    <vscale x 16 x i16> %0,
+    <vscale x 16 x i8> %1,
+    <vscale x 16 x i8> %2,
+    <vscale x 16 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 16 x i16> %a
+}
+
+declare <vscale x 32 x i16> @llvm.riscv.vwmaccsu.nxv32i16.nxv32i8(
+  <vscale x 32 x i16>,
+  <vscale x 32 x i8>,
+  <vscale x 32 x i8>,
+  i32);
+
+define <vscale x 32 x i16>  @intrinsic_vwmaccsu_vv_nxv32i16_nxv32i8_nxv32i8(<vscale x 32 x i16> %0, <vscale x 32 x i8> %1, <vscale x 32 x i8> %2, i32 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccsu_vv_nxv32i16_nxv32i8_nxv32i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a3, zero, e8,m4,ta,mu
+; CHECK-NEXT:    vle8.v v28, (a1)
+; CHECK-NEXT:    vle8.v v8, (a0)
+; CHECK-NEXT:    vsetvli a0, a2, e8,m4,ta,mu
+; CHECK-NEXT:    vwmaccsu.vv v16, v8, v28
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 32 x i16> @llvm.riscv.vwmaccsu.nxv32i16.nxv32i8(
+    <vscale x 32 x i16> %0,
+    <vscale x 32 x i8> %1,
+    <vscale x 32 x i8> %2,
+    i32 %3)
+
+  ret <vscale x 32 x i16> %a
+}
+
+declare <vscale x 32 x i16> @llvm.riscv.vwmaccsu.mask.nxv32i16.nxv32i8(
+  <vscale x 32 x i16>,
+  <vscale x 32 x i8>,
+  <vscale x 32 x i8>,
+  <vscale x 32 x i1>,
+  i32);
+
+define <vscale x 32 x i16>  @intrinsic_vwmaccsu_mask_vv_nxv32i16_nxv32i8_nxv32i8(<vscale x 32 x i16> %0, <vscale x 32 x i8> %1, <vscale x 32 x i8> %2, <vscale x 32 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccsu_mask_vv_nxv32i16_nxv32i8_nxv32i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a3, zero, e8,m4,ta,mu
+; CHECK-NEXT:    vle8.v v28, (a1)
+; CHECK-NEXT:    vle8.v v8, (a0)
+; CHECK-NEXT:    vsetvli a0, a2, e8,m4,ta,mu
+; CHECK-NEXT:    vwmaccsu.vv v16, v8, v28, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 32 x i16> @llvm.riscv.vwmaccsu.mask.nxv32i16.nxv32i8(
+    <vscale x 32 x i16> %0,
+    <vscale x 32 x i8> %1,
+    <vscale x 32 x i8> %2,
+    <vscale x 32 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 32 x i16> %a
+}
+
+declare <vscale x 1 x i32> @llvm.riscv.vwmaccsu.nxv1i32.nxv1i16(
+  <vscale x 1 x i32>,
+  <vscale x 1 x i16>,
+  <vscale x 1 x i16>,
+  i32);
+
+define <vscale x 1 x i32>  @intrinsic_vwmaccsu_vv_nxv1i32_nxv1i16_nxv1i16(<vscale x 1 x i32> %0, <vscale x 1 x i16> %1, <vscale x 1 x i16> %2, i32 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccsu_vv_nxv1i32_nxv1i16_nxv1i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e16,mf4,ta,mu
+; CHECK-NEXT:    vwmaccsu.vv v16, v17, v18
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i32> @llvm.riscv.vwmaccsu.nxv1i32.nxv1i16(
+    <vscale x 1 x i32> %0,
+    <vscale x 1 x i16> %1,
+    <vscale x 1 x i16> %2,
+    i32 %3)
+
+  ret <vscale x 1 x i32> %a
+}
+
+declare <vscale x 1 x i32> @llvm.riscv.vwmaccsu.mask.nxv1i32.nxv1i16(
+  <vscale x 1 x i32>,
+  <vscale x 1 x i16>,
+  <vscale x 1 x i16>,
+  <vscale x 1 x i1>,
+  i32);
+
+define <vscale x 1 x i32>  @intrinsic_vwmaccsu_mask_vv_nxv1i32_nxv1i16_nxv1i16(<vscale x 1 x i32> %0, <vscale x 1 x i16> %1, <vscale x 1 x i16> %2, <vscale x 1 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccsu_mask_vv_nxv1i32_nxv1i16_nxv1i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e16,mf4,ta,mu
+; CHECK-NEXT:    vwmaccsu.vv v16, v17, v18, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i32> @llvm.riscv.vwmaccsu.mask.nxv1i32.nxv1i16(
+    <vscale x 1 x i32> %0,
+    <vscale x 1 x i16> %1,
+    <vscale x 1 x i16> %2,
+    <vscale x 1 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 1 x i32> %a
+}
+
+declare <vscale x 2 x i32> @llvm.riscv.vwmaccsu.nxv2i32.nxv2i16(
+  <vscale x 2 x i32>,
+  <vscale x 2 x i16>,
+  <vscale x 2 x i16>,
+  i32);
+
+define <vscale x 2 x i32>  @intrinsic_vwmaccsu_vv_nxv2i32_nxv2i16_nxv2i16(<vscale x 2 x i32> %0, <vscale x 2 x i16> %1, <vscale x 2 x i16> %2, i32 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccsu_vv_nxv2i32_nxv2i16_nxv2i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e16,mf2,ta,mu
+; CHECK-NEXT:    vwmaccsu.vv v16, v17, v18
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i32> @llvm.riscv.vwmaccsu.nxv2i32.nxv2i16(
+    <vscale x 2 x i32> %0,
+    <vscale x 2 x i16> %1,
+    <vscale x 2 x i16> %2,
+    i32 %3)
+
+  ret <vscale x 2 x i32> %a
+}
+
+declare <vscale x 2 x i32> @llvm.riscv.vwmaccsu.mask.nxv2i32.nxv2i16(
+  <vscale x 2 x i32>,
+  <vscale x 2 x i16>,
+  <vscale x 2 x i16>,
+  <vscale x 2 x i1>,
+  i32);
+
+define <vscale x 2 x i32>  @intrinsic_vwmaccsu_mask_vv_nxv2i32_nxv2i16_nxv2i16(<vscale x 2 x i32> %0, <vscale x 2 x i16> %1, <vscale x 2 x i16> %2, <vscale x 2 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccsu_mask_vv_nxv2i32_nxv2i16_nxv2i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e16,mf2,ta,mu
+; CHECK-NEXT:    vwmaccsu.vv v16, v17, v18, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i32> @llvm.riscv.vwmaccsu.mask.nxv2i32.nxv2i16(
+    <vscale x 2 x i32> %0,
+    <vscale x 2 x i16> %1,
+    <vscale x 2 x i16> %2,
+    <vscale x 2 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 2 x i32> %a
+}
+
+declare <vscale x 4 x i32> @llvm.riscv.vwmaccsu.nxv4i32.nxv4i16(
+  <vscale x 4 x i32>,
+  <vscale x 4 x i16>,
+  <vscale x 4 x i16>,
+  i32);
+
+define <vscale x 4 x i32>  @intrinsic_vwmaccsu_vv_nxv4i32_nxv4i16_nxv4i16(<vscale x 4 x i32> %0, <vscale x 4 x i16> %1, <vscale x 4 x i16> %2, i32 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccsu_vv_nxv4i32_nxv4i16_nxv4i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e16,m1,ta,mu
+; CHECK-NEXT:    vwmaccsu.vv v16, v18, v19
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i32> @llvm.riscv.vwmaccsu.nxv4i32.nxv4i16(
+    <vscale x 4 x i32> %0,
+    <vscale x 4 x i16> %1,
+    <vscale x 4 x i16> %2,
+    i32 %3)
+
+  ret <vscale x 4 x i32> %a
+}
+
+declare <vscale x 4 x i32> @llvm.riscv.vwmaccsu.mask.nxv4i32.nxv4i16(
+  <vscale x 4 x i32>,
+  <vscale x 4 x i16>,
+  <vscale x 4 x i16>,
+  <vscale x 4 x i1>,
+  i32);
+
+define <vscale x 4 x i32>  @intrinsic_vwmaccsu_mask_vv_nxv4i32_nxv4i16_nxv4i16(<vscale x 4 x i32> %0, <vscale x 4 x i16> %1, <vscale x 4 x i16> %2, <vscale x 4 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccsu_mask_vv_nxv4i32_nxv4i16_nxv4i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e16,m1,ta,mu
+; CHECK-NEXT:    vwmaccsu.vv v16, v18, v19, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i32> @llvm.riscv.vwmaccsu.mask.nxv4i32.nxv4i16(
+    <vscale x 4 x i32> %0,
+    <vscale x 4 x i16> %1,
+    <vscale x 4 x i16> %2,
+    <vscale x 4 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 4 x i32> %a
+}
+
+declare <vscale x 8 x i32> @llvm.riscv.vwmaccsu.nxv8i32.nxv8i16(
+  <vscale x 8 x i32>,
+  <vscale x 8 x i16>,
+  <vscale x 8 x i16>,
+  i32);
+
+define <vscale x 8 x i32>  @intrinsic_vwmaccsu_vv_nxv8i32_nxv8i16_nxv8i16(<vscale x 8 x i32> %0, <vscale x 8 x i16> %1, <vscale x 8 x i16> %2, i32 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccsu_vv_nxv8i32_nxv8i16_nxv8i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e16,m2,ta,mu
+; CHECK-NEXT:    vwmaccsu.vv v16, v20, v22
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i32> @llvm.riscv.vwmaccsu.nxv8i32.nxv8i16(
+    <vscale x 8 x i32> %0,
+    <vscale x 8 x i16> %1,
+    <vscale x 8 x i16> %2,
+    i32 %3)
+
+  ret <vscale x 8 x i32> %a
+}
+
+declare <vscale x 8 x i32> @llvm.riscv.vwmaccsu.mask.nxv8i32.nxv8i16(
+  <vscale x 8 x i32>,
+  <vscale x 8 x i16>,
+  <vscale x 8 x i16>,
+  <vscale x 8 x i1>,
+  i32);
+
+define <vscale x 8 x i32>  @intrinsic_vwmaccsu_mask_vv_nxv8i32_nxv8i16_nxv8i16(<vscale x 8 x i32> %0, <vscale x 8 x i16> %1, <vscale x 8 x i16> %2, <vscale x 8 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccsu_mask_vv_nxv8i32_nxv8i16_nxv8i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e16,m2,ta,mu
+; CHECK-NEXT:    vwmaccsu.vv v16, v20, v22, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i32> @llvm.riscv.vwmaccsu.mask.nxv8i32.nxv8i16(
+    <vscale x 8 x i32> %0,
+    <vscale x 8 x i16> %1,
+    <vscale x 8 x i16> %2,
+    <vscale x 8 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 8 x i32> %a
+}
+
+declare <vscale x 16 x i32> @llvm.riscv.vwmaccsu.nxv16i32.nxv16i16(
+  <vscale x 16 x i32>,
+  <vscale x 16 x i16>,
+  <vscale x 16 x i16>,
+  i32);
+
+define <vscale x 16 x i32>  @intrinsic_vwmaccsu_vv_nxv16i32_nxv16i16_nxv16i16(<vscale x 16 x i32> %0, <vscale x 16 x i16> %1, <vscale x 16 x i16> %2, i32 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccsu_vv_nxv16i32_nxv16i16_nxv16i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a3, zero, e16,m4,ta,mu
+; CHECK-NEXT:    vle16.v v28, (a1)
+; CHECK-NEXT:    vle16.v v8, (a0)
+; CHECK-NEXT:    vsetvli a0, a2, e16,m4,ta,mu
+; CHECK-NEXT:    vwmaccsu.vv v16, v8, v28
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 16 x i32> @llvm.riscv.vwmaccsu.nxv16i32.nxv16i16(
+    <vscale x 16 x i32> %0,
+    <vscale x 16 x i16> %1,
+    <vscale x 16 x i16> %2,
+    i32 %3)
+
+  ret <vscale x 16 x i32> %a
+}
+
+declare <vscale x 16 x i32> @llvm.riscv.vwmaccsu.mask.nxv16i32.nxv16i16(
+  <vscale x 16 x i32>,
+  <vscale x 16 x i16>,
+  <vscale x 16 x i16>,
+  <vscale x 16 x i1>,
+  i32);
+
+define <vscale x 16 x i32>  @intrinsic_vwmaccsu_mask_vv_nxv16i32_nxv16i16_nxv16i16(<vscale x 16 x i32> %0, <vscale x 16 x i16> %1, <vscale x 16 x i16> %2, <vscale x 16 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccsu_mask_vv_nxv16i32_nxv16i16_nxv16i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a3, zero, e16,m4,ta,mu
+; CHECK-NEXT:    vle16.v v28, (a1)
+; CHECK-NEXT:    vle16.v v8, (a0)
+; CHECK-NEXT:    vsetvli a0, a2, e16,m4,ta,mu
+; CHECK-NEXT:    vwmaccsu.vv v16, v8, v28, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 16 x i32> @llvm.riscv.vwmaccsu.mask.nxv16i32.nxv16i16(
+    <vscale x 16 x i32> %0,
+    <vscale x 16 x i16> %1,
+    <vscale x 16 x i16> %2,
+    <vscale x 16 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 16 x i32> %a
+}
+
+declare <vscale x 1 x i16> @llvm.riscv.vwmaccsu.nxv1i16.i8(
+  <vscale x 1 x i16>,
+  i8,
+  <vscale x 1 x i8>,
+  i32);
+
+define <vscale x 1 x i16>  @intrinsic_vwmaccsu_vx_nxv1i16_i8_nxv1i8(<vscale x 1 x i16> %0, i8 %1, <vscale x 1 x i8> %2, i32 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccsu_vx_nxv1i16_i8_nxv1i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e8,mf8,ta,mu
+; CHECK-NEXT:    vwmaccsu.vx v16, a0, v17
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i16> @llvm.riscv.vwmaccsu.nxv1i16.i8(
+    <vscale x 1 x i16> %0,
+    i8 %1,
+    <vscale x 1 x i8> %2,
+    i32 %3)
+
+  ret <vscale x 1 x i16> %a
+}
+
+declare <vscale x 1 x i16> @llvm.riscv.vwmaccsu.mask.nxv1i16.i8(
+  <vscale x 1 x i16>,
+  i8,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i1>,
+  i32);
+
+define <vscale x 1 x i16> @intrinsic_vwmaccsu_mask_vx_nxv1i16_i8_nxv1i8(<vscale x 1 x i16> %0, i8 %1, <vscale x 1 x i8> %2, <vscale x 1 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccsu_mask_vx_nxv1i16_i8_nxv1i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e8,mf8,ta,mu
+; CHECK-NEXT:    vwmaccsu.vx v16, a0, v17, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i16> @llvm.riscv.vwmaccsu.mask.nxv1i16.i8(
+    <vscale x 1 x i16> %0,
+    i8 %1,
+    <vscale x 1 x i8> %2,
+    <vscale x 1 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 1 x i16> %a
+}
+
+declare <vscale x 2 x i16> @llvm.riscv.vwmaccsu.nxv2i16.i8(
+  <vscale x 2 x i16>,
+  i8,
+  <vscale x 2 x i8>,
+  i32);
+
+define <vscale x 2 x i16>  @intrinsic_vwmaccsu_vx_nxv2i16_i8_nxv2i8(<vscale x 2 x i16> %0, i8 %1, <vscale x 2 x i8> %2, i32 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccsu_vx_nxv2i16_i8_nxv2i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e8,mf4,ta,mu
+; CHECK-NEXT:    vwmaccsu.vx v16, a0, v17
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i16> @llvm.riscv.vwmaccsu.nxv2i16.i8(
+    <vscale x 2 x i16> %0,
+    i8 %1,
+    <vscale x 2 x i8> %2,
+    i32 %3)
+
+  ret <vscale x 2 x i16> %a
+}
+
+declare <vscale x 2 x i16> @llvm.riscv.vwmaccsu.mask.nxv2i16.i8(
+  <vscale x 2 x i16>,
+  i8,
+  <vscale x 2 x i8>,
+  <vscale x 2 x i1>,
+  i32);
+
+define <vscale x 2 x i16> @intrinsic_vwmaccsu_mask_vx_nxv2i16_i8_nxv2i8(<vscale x 2 x i16> %0, i8 %1, <vscale x 2 x i8> %2, <vscale x 2 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccsu_mask_vx_nxv2i16_i8_nxv2i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e8,mf4,ta,mu
+; CHECK-NEXT:    vwmaccsu.vx v16, a0, v17, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i16> @llvm.riscv.vwmaccsu.mask.nxv2i16.i8(
+    <vscale x 2 x i16> %0,
+    i8 %1,
+    <vscale x 2 x i8> %2,
+    <vscale x 2 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 2 x i16> %a
+}
+
+declare <vscale x 4 x i16> @llvm.riscv.vwmaccsu.nxv4i16.i8(
+  <vscale x 4 x i16>,
+  i8,
+  <vscale x 4 x i8>,
+  i32);
+
+define <vscale x 4 x i16>  @intrinsic_vwmaccsu_vx_nxv4i16_i8_nxv4i8(<vscale x 4 x i16> %0, i8 %1, <vscale x 4 x i8> %2, i32 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccsu_vx_nxv4i16_i8_nxv4i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e8,mf2,ta,mu
+; CHECK-NEXT:    vwmaccsu.vx v16, a0, v17
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i16> @llvm.riscv.vwmaccsu.nxv4i16.i8(
+    <vscale x 4 x i16> %0,
+    i8 %1,
+    <vscale x 4 x i8> %2,
+    i32 %3)
+
+  ret <vscale x 4 x i16> %a
+}
+
+declare <vscale x 4 x i16> @llvm.riscv.vwmaccsu.mask.nxv4i16.i8(
+  <vscale x 4 x i16>,
+  i8,
+  <vscale x 4 x i8>,
+  <vscale x 4 x i1>,
+  i32);
+
+define <vscale x 4 x i16> @intrinsic_vwmaccsu_mask_vx_nxv4i16_i8_nxv4i8(<vscale x 4 x i16> %0, i8 %1, <vscale x 4 x i8> %2, <vscale x 4 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccsu_mask_vx_nxv4i16_i8_nxv4i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e8,mf2,ta,mu
+; CHECK-NEXT:    vwmaccsu.vx v16, a0, v17, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i16> @llvm.riscv.vwmaccsu.mask.nxv4i16.i8(
+    <vscale x 4 x i16> %0,
+    i8 %1,
+    <vscale x 4 x i8> %2,
+    <vscale x 4 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 4 x i16> %a
+}
+
+declare <vscale x 8 x i16> @llvm.riscv.vwmaccsu.nxv8i16.i8(
+  <vscale x 8 x i16>,
+  i8,
+  <vscale x 8 x i8>,
+  i32);
+
+define <vscale x 8 x i16>  @intrinsic_vwmaccsu_vx_nxv8i16_i8_nxv8i8(<vscale x 8 x i16> %0, i8 %1, <vscale x 8 x i8> %2, i32 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccsu_vx_nxv8i16_i8_nxv8i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e8,m1,ta,mu
+; CHECK-NEXT:    vwmaccsu.vx v16, a0, v18
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i16> @llvm.riscv.vwmaccsu.nxv8i16.i8(
+    <vscale x 8 x i16> %0,
+    i8 %1,
+    <vscale x 8 x i8> %2,
+    i32 %3)
+
+  ret <vscale x 8 x i16> %a
+}
+
+declare <vscale x 8 x i16> @llvm.riscv.vwmaccsu.mask.nxv8i16.i8(
+  <vscale x 8 x i16>,
+  i8,
+  <vscale x 8 x i8>,
+  <vscale x 8 x i1>,
+  i32);
+
+define <vscale x 8 x i16> @intrinsic_vwmaccsu_mask_vx_nxv8i16_i8_nxv8i8(<vscale x 8 x i16> %0, i8 %1, <vscale x 8 x i8> %2, <vscale x 8 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccsu_mask_vx_nxv8i16_i8_nxv8i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e8,m1,ta,mu
+; CHECK-NEXT:    vwmaccsu.vx v16, a0, v18, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i16> @llvm.riscv.vwmaccsu.mask.nxv8i16.i8(
+    <vscale x 8 x i16> %0,
+    i8 %1,
+    <vscale x 8 x i8> %2,
+    <vscale x 8 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 8 x i16> %a
+}
+
+declare <vscale x 16 x i16> @llvm.riscv.vwmaccsu.nxv16i16.i8(
+  <vscale x 16 x i16>,
+  i8,
+  <vscale x 16 x i8>,
+  i32);
+
+define <vscale x 16 x i16>  @intrinsic_vwmaccsu_vx_nxv16i16_i8_nxv16i8(<vscale x 16 x i16> %0, i8 %1, <vscale x 16 x i8> %2, i32 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccsu_vx_nxv16i16_i8_nxv16i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e8,m2,ta,mu
+; CHECK-NEXT:    vwmaccsu.vx v16, a0, v20
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 16 x i16> @llvm.riscv.vwmaccsu.nxv16i16.i8(
+    <vscale x 16 x i16> %0,
+    i8 %1,
+    <vscale x 16 x i8> %2,
+    i32 %3)
+
+  ret <vscale x 16 x i16> %a
+}
+
+declare <vscale x 16 x i16> @llvm.riscv.vwmaccsu.mask.nxv16i16.i8(
+  <vscale x 16 x i16>,
+  i8,
+  <vscale x 16 x i8>,
+  <vscale x 16 x i1>,
+  i32);
+
+define <vscale x 16 x i16> @intrinsic_vwmaccsu_mask_vx_nxv16i16_i8_nxv16i8(<vscale x 16 x i16> %0, i8 %1, <vscale x 16 x i8> %2, <vscale x 16 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccsu_mask_vx_nxv16i16_i8_nxv16i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e8,m2,ta,mu
+; CHECK-NEXT:    vwmaccsu.vx v16, a0, v20, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 16 x i16> @llvm.riscv.vwmaccsu.mask.nxv16i16.i8(
+    <vscale x 16 x i16> %0,
+    i8 %1,
+    <vscale x 16 x i8> %2,
+    <vscale x 16 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 16 x i16> %a
+}
+
+declare <vscale x 32 x i16> @llvm.riscv.vwmaccsu.nxv32i16.i8(
+  <vscale x 32 x i16>,
+  i8,
+  <vscale x 32 x i8>,
+  i32);
+
+define <vscale x 32 x i16>  @intrinsic_vwmaccsu_vx_nxv32i16_i8_nxv32i8(<vscale x 32 x i16> %0, i8 %1, <vscale x 32 x i8> %2, i32 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccsu_vx_nxv32i16_i8_nxv32i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a3, zero, e8,m4,ta,mu
+; CHECK-NEXT:    vle8.v v28, (a1)
+; CHECK-NEXT:    vsetvli a1, a2, e8,m4,ta,mu
+; CHECK-NEXT:    vwmaccsu.vx v16, a0, v28
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 32 x i16> @llvm.riscv.vwmaccsu.nxv32i16.i8(
+    <vscale x 32 x i16> %0,
+    i8 %1,
+    <vscale x 32 x i8> %2,
+    i32 %3)
+
+  ret <vscale x 32 x i16> %a
+}
+
+declare <vscale x 32 x i16> @llvm.riscv.vwmaccsu.mask.nxv32i16.i8(
+  <vscale x 32 x i16>,
+  i8,
+  <vscale x 32 x i8>,
+  <vscale x 32 x i1>,
+  i32);
+
+define <vscale x 32 x i16> @intrinsic_vwmaccsu_mask_vx_nxv32i16_i8_nxv32i8(<vscale x 32 x i16> %0, i8 %1, <vscale x 32 x i8> %2, <vscale x 32 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccsu_mask_vx_nxv32i16_i8_nxv32i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a3, zero, e8,m4,ta,mu
+; CHECK-NEXT:    vle8.v v28, (a1)
+; CHECK-NEXT:    vsetvli a1, a2, e8,m4,ta,mu
+; CHECK-NEXT:    vwmaccsu.vx v16, a0, v28, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 32 x i16> @llvm.riscv.vwmaccsu.mask.nxv32i16.i8(
+    <vscale x 32 x i16> %0,
+    i8 %1,
+    <vscale x 32 x i8> %2,
+    <vscale x 32 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 32 x i16> %a
+}
+
+declare <vscale x 1 x i32> @llvm.riscv.vwmaccsu.nxv1i32.i16(
+  <vscale x 1 x i32>,
+  i16,
+  <vscale x 1 x i16>,
+  i32);
+
+define <vscale x 1 x i32>  @intrinsic_vwmaccsu_vx_nxv1i32_i16_nxv1i16(<vscale x 1 x i32> %0, i16 %1, <vscale x 1 x i16> %2, i32 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccsu_vx_nxv1i32_i16_nxv1i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e16,mf4,ta,mu
+; CHECK-NEXT:    vwmaccsu.vx v16, a0, v17
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i32> @llvm.riscv.vwmaccsu.nxv1i32.i16(
+    <vscale x 1 x i32> %0,
+    i16 %1,
+    <vscale x 1 x i16> %2,
+    i32 %3)
+
+  ret <vscale x 1 x i32> %a
+}
+
+declare <vscale x 1 x i32> @llvm.riscv.vwmaccsu.mask.nxv1i32.i16(
+  <vscale x 1 x i32>,
+  i16,
+  <vscale x 1 x i16>,
+  <vscale x 1 x i1>,
+  i32);
+
+define <vscale x 1 x i32> @intrinsic_vwmaccsu_mask_vx_nxv1i32_i16_nxv1i16(<vscale x 1 x i32> %0, i16 %1, <vscale x 1 x i16> %2, <vscale x 1 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccsu_mask_vx_nxv1i32_i16_nxv1i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e16,mf4,ta,mu
+; CHECK-NEXT:    vwmaccsu.vx v16, a0, v17, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i32> @llvm.riscv.vwmaccsu.mask.nxv1i32.i16(
+    <vscale x 1 x i32> %0,
+    i16 %1,
+    <vscale x 1 x i16> %2,
+    <vscale x 1 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 1 x i32> %a
+}
+
+declare <vscale x 2 x i32> @llvm.riscv.vwmaccsu.nxv2i32.i16(
+  <vscale x 2 x i32>,
+  i16,
+  <vscale x 2 x i16>,
+  i32);
+
+define <vscale x 2 x i32>  @intrinsic_vwmaccsu_vx_nxv2i32_i16_nxv2i16(<vscale x 2 x i32> %0, i16 %1, <vscale x 2 x i16> %2, i32 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccsu_vx_nxv2i32_i16_nxv2i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e16,mf2,ta,mu
+; CHECK-NEXT:    vwmaccsu.vx v16, a0, v17
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i32> @llvm.riscv.vwmaccsu.nxv2i32.i16(
+    <vscale x 2 x i32> %0,
+    i16 %1,
+    <vscale x 2 x i16> %2,
+    i32 %3)
+
+  ret <vscale x 2 x i32> %a
+}
+
+declare <vscale x 2 x i32> @llvm.riscv.vwmaccsu.mask.nxv2i32.i16(
+  <vscale x 2 x i32>,
+  i16,
+  <vscale x 2 x i16>,
+  <vscale x 2 x i1>,
+  i32);
+
+define <vscale x 2 x i32> @intrinsic_vwmaccsu_mask_vx_nxv2i32_i16_nxv2i16(<vscale x 2 x i32> %0, i16 %1, <vscale x 2 x i16> %2, <vscale x 2 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccsu_mask_vx_nxv2i32_i16_nxv2i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e16,mf2,ta,mu
+; CHECK-NEXT:    vwmaccsu.vx v16, a0, v17, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i32> @llvm.riscv.vwmaccsu.mask.nxv2i32.i16(
+    <vscale x 2 x i32> %0,
+    i16 %1,
+    <vscale x 2 x i16> %2,
+    <vscale x 2 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 2 x i32> %a
+}
+
+declare <vscale x 4 x i32> @llvm.riscv.vwmaccsu.nxv4i32.i16(
+  <vscale x 4 x i32>,
+  i16,
+  <vscale x 4 x i16>,
+  i32);
+
+define <vscale x 4 x i32>  @intrinsic_vwmaccsu_vx_nxv4i32_i16_nxv4i16(<vscale x 4 x i32> %0, i16 %1, <vscale x 4 x i16> %2, i32 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccsu_vx_nxv4i32_i16_nxv4i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e16,m1,ta,mu
+; CHECK-NEXT:    vwmaccsu.vx v16, a0, v18
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i32> @llvm.riscv.vwmaccsu.nxv4i32.i16(
+    <vscale x 4 x i32> %0,
+    i16 %1,
+    <vscale x 4 x i16> %2,
+    i32 %3)
+
+  ret <vscale x 4 x i32> %a
+}
+
+declare <vscale x 4 x i32> @llvm.riscv.vwmaccsu.mask.nxv4i32.i16(
+  <vscale x 4 x i32>,
+  i16,
+  <vscale x 4 x i16>,
+  <vscale x 4 x i1>,
+  i32);
+
+define <vscale x 4 x i32> @intrinsic_vwmaccsu_mask_vx_nxv4i32_i16_nxv4i16(<vscale x 4 x i32> %0, i16 %1, <vscale x 4 x i16> %2, <vscale x 4 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccsu_mask_vx_nxv4i32_i16_nxv4i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e16,m1,ta,mu
+; CHECK-NEXT:    vwmaccsu.vx v16, a0, v18, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i32> @llvm.riscv.vwmaccsu.mask.nxv4i32.i16(
+    <vscale x 4 x i32> %0,
+    i16 %1,
+    <vscale x 4 x i16> %2,
+    <vscale x 4 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 4 x i32> %a
+}
+
+declare <vscale x 8 x i32> @llvm.riscv.vwmaccsu.nxv8i32.i16(
+  <vscale x 8 x i32>,
+  i16,
+  <vscale x 8 x i16>,
+  i32);
+
+define <vscale x 8 x i32>  @intrinsic_vwmaccsu_vx_nxv8i32_i16_nxv8i16(<vscale x 8 x i32> %0, i16 %1, <vscale x 8 x i16> %2, i32 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccsu_vx_nxv8i32_i16_nxv8i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e16,m2,ta,mu
+; CHECK-NEXT:    vwmaccsu.vx v16, a0, v20
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i32> @llvm.riscv.vwmaccsu.nxv8i32.i16(
+    <vscale x 8 x i32> %0,
+    i16 %1,
+    <vscale x 8 x i16> %2,
+    i32 %3)
+
+  ret <vscale x 8 x i32> %a
+}
+
+declare <vscale x 8 x i32> @llvm.riscv.vwmaccsu.mask.nxv8i32.i16(
+  <vscale x 8 x i32>,
+  i16,
+  <vscale x 8 x i16>,
+  <vscale x 8 x i1>,
+  i32);
+
+define <vscale x 8 x i32> @intrinsic_vwmaccsu_mask_vx_nxv8i32_i16_nxv8i16(<vscale x 8 x i32> %0, i16 %1, <vscale x 8 x i16> %2, <vscale x 8 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccsu_mask_vx_nxv8i32_i16_nxv8i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e16,m2,ta,mu
+; CHECK-NEXT:    vwmaccsu.vx v16, a0, v20, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i32> @llvm.riscv.vwmaccsu.mask.nxv8i32.i16(
+    <vscale x 8 x i32> %0,
+    i16 %1,
+    <vscale x 8 x i16> %2,
+    <vscale x 8 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 8 x i32> %a
+}
+
+declare <vscale x 16 x i32> @llvm.riscv.vwmaccsu.nxv16i32.i16(
+  <vscale x 16 x i32>,
+  i16,
+  <vscale x 16 x i16>,
+  i32);
+
+define <vscale x 16 x i32>  @intrinsic_vwmaccsu_vx_nxv16i32_i16_nxv16i16(<vscale x 16 x i32> %0, i16 %1, <vscale x 16 x i16> %2, i32 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccsu_vx_nxv16i32_i16_nxv16i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a3, zero, e16,m4,ta,mu
+; CHECK-NEXT:    vle16.v v28, (a1)
+; CHECK-NEXT:    vsetvli a1, a2, e16,m4,ta,mu
+; CHECK-NEXT:    vwmaccsu.vx v16, a0, v28
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 16 x i32> @llvm.riscv.vwmaccsu.nxv16i32.i16(
+    <vscale x 16 x i32> %0,
+    i16 %1,
+    <vscale x 16 x i16> %2,
+    i32 %3)
+
+  ret <vscale x 16 x i32> %a
+}
+
+declare <vscale x 16 x i32> @llvm.riscv.vwmaccsu.mask.nxv16i32.i16(
+  <vscale x 16 x i32>,
+  i16,
+  <vscale x 16 x i16>,
+  <vscale x 16 x i1>,
+  i32);
+
+define <vscale x 16 x i32> @intrinsic_vwmaccsu_mask_vx_nxv16i32_i16_nxv16i16(<vscale x 16 x i32> %0, i16 %1, <vscale x 16 x i16> %2, <vscale x 16 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccsu_mask_vx_nxv16i32_i16_nxv16i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a3, zero, e16,m4,ta,mu
+; CHECK-NEXT:    vle16.v v28, (a1)
+; CHECK-NEXT:    vsetvli a1, a2, e16,m4,ta,mu
+; CHECK-NEXT:    vwmaccsu.vx v16, a0, v28, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 16 x i32> @llvm.riscv.vwmaccsu.mask.nxv16i32.i16(
+    <vscale x 16 x i32> %0,
+    i16 %1,
+    <vscale x 16 x i16> %2,
+    <vscale x 16 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 16 x i32> %a
+}

llvm/test/CodeGen/RISCV/rvv/vwmaccsu-rv64.ll

+; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
+; RUN: llc -mtriple=riscv64 -mattr=+experimental-v,+d -verify-machineinstrs \
+; RUN:   --riscv-no-aliases < %s | FileCheck %s
+declare <vscale x 1 x i16> @llvm.riscv.vwmaccsu.nxv1i16.nxv1i8(
+  <vscale x 1 x i16>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  i64);
+
+define <vscale x 1 x i16>  @intrinsic_vwmaccsu_vv_nxv1i16_nxv1i8_nxv1i8(<vscale x 1 x i16> %0, <vscale x 1 x i8> %1, <vscale x 1 x i8> %2, i64 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccsu_vv_nxv1i16_nxv1i8_nxv1i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e8,mf8,ta,mu
+; CHECK-NEXT:    vwmaccsu.vv v16, v17, v18
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i16> @llvm.riscv.vwmaccsu.nxv1i16.nxv1i8(
+    <vscale x 1 x i16> %0,
+    <vscale x 1 x i8> %1,
+    <vscale x 1 x i8> %2,
+    i64 %3)
+
+  ret <vscale x 1 x i16> %a
+}
+
+declare <vscale x 1 x i16> @llvm.riscv.vwmaccsu.mask.nxv1i16.nxv1i8(
+  <vscale x 1 x i16>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i1>,
+  i64);
+
+define <vscale x 1 x i16>  @intrinsic_vwmaccsu_mask_vv_nxv1i16_nxv1i8_nxv1i8(<vscale x 1 x i16> %0, <vscale x 1 x i8> %1, <vscale x 1 x i8> %2, <vscale x 1 x i1> %3, i64 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccsu_mask_vv_nxv1i16_nxv1i8_nxv1i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e8,mf8,ta,mu
+; CHECK-NEXT:    vwmaccsu.vv v16, v17, v18, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i16> @llvm.riscv.vwmaccsu.mask.nxv1i16.nxv1i8(
+    <vscale x 1 x i16> %0,
+    <vscale x 1 x i8> %1,
+    <vscale x 1 x i8> %2,
+    <vscale x 1 x i1> %3,
+    i64 %4)
+
+  ret <vscale x 1 x i16> %a
+}
+
+declare <vscale x 2 x i16> @llvm.riscv.vwmaccsu.nxv2i16.nxv2i8(
+  <vscale x 2 x i16>,
+  <vscale x 2 x i8>,
+  <vscale x 2 x i8>,
+  i64);
+
+define <vscale x 2 x i16>  @intrinsic_vwmaccsu_vv_nxv2i16_nxv2i8_nxv2i8(<vscale x 2 x i16> %0, <vscale x 2 x i8> %1, <vscale x 2 x i8> %2, i64 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccsu_vv_nxv2i16_nxv2i8_nxv2i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e8,mf4,ta,mu
+; CHECK-NEXT:    vwmaccsu.vv v16, v17, v18
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i16> @llvm.riscv.vwmaccsu.nxv2i16.nxv2i8(
+    <vscale x 2 x i16> %0,
+    <vscale x 2 x i8> %1,
+    <vscale x 2 x i8> %2,
+    i64 %3)
+
+  ret <vscale x 2 x i16> %a
+}
+
+declare <vscale x 2 x i16> @llvm.riscv.vwmaccsu.mask.nxv2i16.nxv2i8(
+  <vscale x 2 x i16>,
+  <vscale x 2 x i8>,
+  <vscale x 2 x i8>,
+  <vscale x 2 x i1>,
+  i64);
+
+define <vscale x 2 x i16>  @intrinsic_vwmaccsu_mask_vv_nxv2i16_nxv2i8_nxv2i8(<vscale x 2 x i16> %0, <vscale x 2 x i8> %1, <vscale x 2 x i8> %2, <vscale x 2 x i1> %3, i64 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccsu_mask_vv_nxv2i16_nxv2i8_nxv2i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e8,mf4,ta,mu
+; CHECK-NEXT:    vwmaccsu.vv v16, v17, v18, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i16> @llvm.riscv.vwmaccsu.mask.nxv2i16.nxv2i8(
+    <vscale x 2 x i16> %0,
+    <vscale x 2 x i8> %1,
+    <vscale x 2 x i8> %2,
+    <vscale x 2 x i1> %3,
+    i64 %4)
+
+  ret <vscale x 2 x i16> %a
+}
+
+declare <vscale x 4 x i16> @llvm.riscv.vwmaccsu.nxv4i16.nxv4i8(
+  <vscale x 4 x i16>,
+  <vscale x 4 x i8>,
+  <vscale x 4 x i8>,
+  i64);
+
+define <vscale x 4 x i16>  @intrinsic_vwmaccsu_vv_nxv4i16_nxv4i8_nxv4i8(<vscale x 4 x i16> %0, <vscale x 4 x i8> %1, <vscale x 4 x i8> %2, i64 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccsu_vv_nxv4i16_nxv4i8_nxv4i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e8,mf2,ta,mu
+; CHECK-NEXT:    vwmaccsu.vv v16, v17, v18
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i16> @llvm.riscv.vwmaccsu.nxv4i16.nxv4i8(
+    <vscale x 4 x i16> %0,
+    <vscale x 4 x i8> %1,
+    <vscale x 4 x i8> %2,
+    i64 %3)
+
+  ret <vscale x 4 x i16> %a
+}
+
+declare <vscale x 4 x i16> @llvm.riscv.vwmaccsu.mask.nxv4i16.nxv4i8(
+  <vscale x 4 x i16>,
+  <vscale x 4 x i8>,
+  <vscale x 4 x i8>,
+  <vscale x 4 x i1>,
+  i64);
+
+define <vscale x 4 x i16>  @intrinsic_vwmaccsu_mask_vv_nxv4i16_nxv4i8_nxv4i8(<vscale x 4 x i16> %0, <vscale x 4 x i8> %1, <vscale x 4 x i8> %2, <vscale x 4 x i1> %3, i64 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccsu_mask_vv_nxv4i16_nxv4i8_nxv4i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e8,mf2,ta,mu
+; CHECK-NEXT:    vwmaccsu.vv v16, v17, v18, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i16> @llvm.riscv.vwmaccsu.mask.nxv4i16.nxv4i8(
+    <vscale x 4 x i16> %0,
+    <vscale x 4 x i8> %1,
+    <vscale x 4 x i8> %2,
+    <vscale x 4 x i1> %3,
+    i64 %4)
+
+  ret <vscale x 4 x i16> %a
+}
+
+declare <vscale x 8 x i16> @llvm.riscv.vwmaccsu.nxv8i16.nxv8i8(
+  <vscale x 8 x i16>,
+  <vscale x 8 x i8>,
+  <vscale x 8 x i8>,
+  i64);
+
+define <vscale x 8 x i16>  @intrinsic_vwmaccsu_vv_nxv8i16_nxv8i8_nxv8i8(<vscale x 8 x i16> %0, <vscale x 8 x i8> %1, <vscale x 8 x i8> %2, i64 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccsu_vv_nxv8i16_nxv8i8_nxv8i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e8,m1,ta,mu
+; CHECK-NEXT:    vwmaccsu.vv v16, v18, v19
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i16> @llvm.riscv.vwmaccsu.nxv8i16.nxv8i8(
+    <vscale x 8 x i16> %0,
+    <vscale x 8 x i8> %1,
+    <vscale x 8 x i8> %2,
+    i64 %3)
+
+  ret <vscale x 8 x i16> %a
+}
+
+declare <vscale x 8 x i16> @llvm.riscv.vwmaccsu.mask.nxv8i16.nxv8i8(
+  <vscale x 8 x i16>,
+  <vscale x 8 x i8>,
+  <vscale x 8 x i8>,
+  <vscale x 8 x i1>,
+  i64);
+
+define <vscale x 8 x i16>  @intrinsic_vwmaccsu_mask_vv_nxv8i16_nxv8i8_nxv8i8(<vscale x 8 x i16> %0, <vscale x 8 x i8> %1, <vscale x 8 x i8> %2, <vscale x 8 x i1> %3, i64 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccsu_mask_vv_nxv8i16_nxv8i8_nxv8i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e8,m1,ta,mu
+; CHECK-NEXT:    vwmaccsu.vv v16, v18, v19, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i16> @llvm.riscv.vwmaccsu.mask.nxv8i16.nxv8i8(
+    <vscale x 8 x i16> %0,
+    <vscale x 8 x i8> %1,
+    <vscale x 8 x i8> %2,
+    <vscale x 8 x i1> %3,
+    i64 %4)
+
+  ret <vscale x 8 x i16> %a
+}
+
+declare <vscale x 16 x i16> @llvm.riscv.vwmaccsu.nxv16i16.nxv16i8(
+  <vscale x 16 x i16>,
+  <vscale x 16 x i8>,
+  <vscale x 16 x i8>,
+  i64);
+
+define <vscale x 16 x i16>  @intrinsic_vwmaccsu_vv_nxv16i16_nxv16i8_nxv16i8(<vscale x 16 x i16> %0, <vscale x 16 x i8> %1, <vscale x 16 x i8> %2, i64 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccsu_vv_nxv16i16_nxv16i8_nxv16i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e8,m2,ta,mu
+; CHECK-NEXT:    vwmaccsu.vv v16, v20, v22
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 16 x i16> @llvm.riscv.vwmaccsu.nxv16i16.nxv16i8(
+    <vscale x 16 x i16> %0,
+    <vscale x 16 x i8> %1,
+    <vscale x 16 x i8> %2,
+    i64 %3)
+
+  ret <vscale x 16 x i16> %a
+}
+
+declare <vscale x 16 x i16> @llvm.riscv.vwmaccsu.mask.nxv16i16.nxv16i8(
+  <vscale x 16 x i16>,
+  <vscale x 16 x i8>,
+  <vscale x 16 x i8>,
+  <vscale x 16 x i1>,
+  i64);
+
+define <vscale x 16 x i16>  @intrinsic_vwmaccsu_mask_vv_nxv16i16_nxv16i8_nxv16i8(<vscale x 16 x i16> %0, <vscale x 16 x i8> %1, <vscale x 16 x i8> %2, <vscale x 16 x i1> %3, i64 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccsu_mask_vv_nxv16i16_nxv16i8_nxv16i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e8,m2,ta,mu
+; CHECK-NEXT:    vwmaccsu.vv v16, v20, v22, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 16 x i16> @llvm.riscv.vwmaccsu.mask.nxv16i16.nxv16i8(
+    <vscale x 16 x i16> %0,
+    <vscale x 16 x i8> %1,
+    <vscale x 16 x i8> %2,
+    <vscale x 16 x i1> %3,
+    i64 %4)
+
+  ret <vscale x 16 x i16> %a
+}
+
+declare <vscale x 32 x i16> @llvm.riscv.vwmaccsu.nxv32i16.nxv32i8(
+  <vscale x 32 x i16>,
+  <vscale x 32 x i8>,
+  <vscale x 32 x i8>,
+  i64);
+
+define <vscale x 32 x i16>  @intrinsic_vwmaccsu_vv_nxv32i16_nxv32i8_nxv32i8(<vscale x 32 x i16> %0, <vscale x 32 x i8> %1, <vscale x 32 x i8> %2, i64 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccsu_vv_nxv32i16_nxv32i8_nxv32i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a3, zero, e8,m4,ta,mu
+; CHECK-NEXT:    vle8.v v28, (a1)
+; CHECK-NEXT:    vle8.v v8, (a0)
+; CHECK-NEXT:    vsetvli a0, a2, e8,m4,ta,mu
+; CHECK-NEXT:    vwmaccsu.vv v16, v8, v28
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 32 x i16> @llvm.riscv.vwmaccsu.nxv32i16.nxv32i8(
+    <vscale x 32 x i16> %0,
+    <vscale x 32 x i8> %1,
+    <vscale x 32 x i8> %2,
+    i64 %3)
+
+  ret <vscale x 32 x i16> %a
+}
+
+declare <vscale x 32 x i16> @llvm.riscv.vwmaccsu.mask.nxv32i16.nxv32i8(
+  <vscale x 32 x i16>,
+  <vscale x 32 x i8>,
+  <vscale x 32 x i8>,
+  <vscale x 32 x i1>,
+  i64);
+
+define <vscale x 32 x i16>  @intrinsic_vwmaccsu_mask_vv_nxv32i16_nxv32i8_nxv32i8(<vscale x 32 x i16> %0, <vscale x 32 x i8> %1, <vscale x 32 x i8> %2, <vscale x 32 x i1> %3, i64 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccsu_mask_vv_nxv32i16_nxv32i8_nxv32i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a3, zero, e8,m4,ta,mu
+; CHECK-NEXT:    vle8.v v28, (a1)
+; CHECK-NEXT:    vle8.v v8, (a0)
+; CHECK-NEXT:    vsetvli a0, a2, e8,m4,ta,mu
+; CHECK-NEXT:    vwmaccsu.vv v16, v8, v28, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 32 x i16> @llvm.riscv.vwmaccsu.mask.nxv32i16.nxv32i8(
+    <vscale x 32 x i16> %0,
+    <vscale x 32 x i8> %1,
+    <vscale x 32 x i8> %2,
+    <vscale x 32 x i1> %3,
+    i64 %4)
+
+  ret <vscale x 32 x i16> %a
+}
+
+declare <vscale x 1 x i32> @llvm.riscv.vwmaccsu.nxv1i32.nxv1i16(
+  <vscale x 1 x i32>,
+  <vscale x 1 x i16>,
+  <vscale x 1 x i16>,
+  i64);
+
+define <vscale x 1 x i32>  @intrinsic_vwmaccsu_vv_nxv1i32_nxv1i16_nxv1i16(<vscale x 1 x i32> %0, <vscale x 1 x i16> %1, <vscale x 1 x i16> %2, i64 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccsu_vv_nxv1i32_nxv1i16_nxv1i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e16,mf4,ta,mu
+; CHECK-NEXT:    vwmaccsu.vv v16, v17, v18
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i32> @llvm.riscv.vwmaccsu.nxv1i32.nxv1i16(
+    <vscale x 1 x i32> %0,
+    <vscale x 1 x i16> %1,
+    <vscale x 1 x i16> %2,
+    i64 %3)
+
+  ret <vscale x 1 x i32> %a
+}
+
+declare <vscale x 1 x i32> @llvm.riscv.vwmaccsu.mask.nxv1i32.nxv1i16(
+  <vscale x 1 x i32>,
+  <vscale x 1 x i16>,
+  <vscale x 1 x i16>,
+  <vscale x 1 x i1>,
+  i64);
+
+define <vscale x 1 x i32>  @intrinsic_vwmaccsu_mask_vv_nxv1i32_nxv1i16_nxv1i16(<vscale x 1 x i32> %0, <vscale x 1 x i16> %1, <vscale x 1 x i16> %2, <vscale x 1 x i1> %3, i64 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccsu_mask_vv_nxv1i32_nxv1i16_nxv1i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e16,mf4,ta,mu
+; CHECK-NEXT:    vwmaccsu.vv v16, v17, v18, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i32> @llvm.riscv.vwmaccsu.mask.nxv1i32.nxv1i16(
+    <vscale x 1 x i32> %0,
+    <vscale x 1 x i16> %1,
+    <vscale x 1 x i16> %2,
+    <vscale x 1 x i1> %3,
+    i64 %4)
+
+  ret <vscale x 1 x i32> %a
+}
+
+declare <vscale x 2 x i32> @llvm.riscv.vwmaccsu.nxv2i32.nxv2i16(
+  <vscale x 2 x i32>,
+  <vscale x 2 x i16>,
+  <vscale x 2 x i16>,
+  i64);
+
+define <vscale x 2 x i32>  @intrinsic_vwmaccsu_vv_nxv2i32_nxv2i16_nxv2i16(<vscale x 2 x i32> %0, <vscale x 2 x i16> %1, <vscale x 2 x i16> %2, i64 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccsu_vv_nxv2i32_nxv2i16_nxv2i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e16,mf2,ta,mu
+; CHECK-NEXT:    vwmaccsu.vv v16, v17, v18
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i32> @llvm.riscv.vwmaccsu.nxv2i32.nxv2i16(
+    <vscale x 2 x i32> %0,
+    <vscale x 2 x i16> %1,
+    <vscale x 2 x i16> %2,
+    i64 %3)
+
+  ret <vscale x 2 x i32> %a
+}
+
+declare <vscale x 2 x i32> @llvm.riscv.vwmaccsu.mask.nxv2i32.nxv2i16(
+  <vscale x 2 x i32>,
+  <vscale x 2 x i16>,
+  <vscale x 2 x i16>,
+  <vscale x 2 x i1>,
+  i64);
+
+define <vscale x 2 x i32>  @intrinsic_vwmaccsu_mask_vv_nxv2i32_nxv2i16_nxv2i16(<vscale x 2 x i32> %0, <vscale x 2 x i16> %1, <vscale x 2 x i16> %2, <vscale x 2 x i1> %3, i64 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccsu_mask_vv_nxv2i32_nxv2i16_nxv2i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e16,mf2,ta,mu
+; CHECK-NEXT:    vwmaccsu.vv v16, v17, v18, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i32> @llvm.riscv.vwmaccsu.mask.nxv2i32.nxv2i16(
+    <vscale x 2 x i32> %0,
+    <vscale x 2 x i16> %1,
+    <vscale x 2 x i16> %2,
+    <vscale x 2 x i1> %3,
+    i64 %4)
+
+  ret <vscale x 2 x i32> %a
+}
+
+declare <vscale x 4 x i32> @llvm.riscv.vwmaccsu.nxv4i32.nxv4i16(
+  <vscale x 4 x i32>,
+  <vscale x 4 x i16>,
+  <vscale x 4 x i16>,
+  i64);
+
+define <vscale x 4 x i32>  @intrinsic_vwmaccsu_vv_nxv4i32_nxv4i16_nxv4i16(<vscale x 4 x i32> %0, <vscale x 4 x i16> %1, <vscale x 4 x i16> %2, i64 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccsu_vv_nxv4i32_nxv4i16_nxv4i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e16,m1,ta,mu
+; CHECK-NEXT:    vwmaccsu.vv v16, v18, v19
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i32> @llvm.riscv.vwmaccsu.nxv4i32.nxv4i16(
+    <vscale x 4 x i32> %0,
+    <vscale x 4 x i16> %1,
+    <vscale x 4 x i16> %2,
+    i64 %3)
+
+  ret <vscale x 4 x i32> %a
+}
+
+declare <vscale x 4 x i32> @llvm.riscv.vwmaccsu.mask.nxv4i32.nxv4i16(
+  <vscale x 4 x i32>,
+  <vscale x 4 x i16>,
+  <vscale x 4 x i16>,
+  <vscale x 4 x i1>,
+  i64);
+
+define <vscale x 4 x i32>  @intrinsic_vwmaccsu_mask_vv_nxv4i32_nxv4i16_nxv4i16(<vscale x 4 x i32> %0, <vscale x 4 x i16> %1, <vscale x 4 x i16> %2, <vscale x 4 x i1> %3, i64 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccsu_mask_vv_nxv4i32_nxv4i16_nxv4i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e16,m1,ta,mu
+; CHECK-NEXT:    vwmaccsu.vv v16, v18, v19, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i32> @llvm.riscv.vwmaccsu.mask.nxv4i32.nxv4i16(
+    <vscale x 4 x i32> %0,
+    <vscale x 4 x i16> %1,
+    <vscale x 4 x i16> %2,
+    <vscale x 4 x i1> %3,
+    i64 %4)
+
+  ret <vscale x 4 x i32> %a
+}
+
+declare <vscale x 8 x i32> @llvm.riscv.vwmaccsu.nxv8i32.nxv8i16(
+  <vscale x 8 x i32>,
+  <vscale x 8 x i16>,
+  <vscale x 8 x i16>,
+  i64);
+
+define <vscale x 8 x i32>  @intrinsic_vwmaccsu_vv_nxv8i32_nxv8i16_nxv8i16(<vscale x 8 x i32> %0, <vscale x 8 x i16> %1, <vscale x 8 x i16> %2, i64 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccsu_vv_nxv8i32_nxv8i16_nxv8i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e16,m2,ta,mu
+; CHECK-NEXT:    vwmaccsu.vv v16, v20, v22
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i32> @llvm.riscv.vwmaccsu.nxv8i32.nxv8i16(
+    <vscale x 8 x i32> %0,
+    <vscale x 8 x i16> %1,
+    <vscale x 8 x i16> %2,
+    i64 %3)
+
+  ret <vscale x 8 x i32> %a
+}
+
+declare <vscale x 8 x i32> @llvm.riscv.vwmaccsu.mask.nxv8i32.nxv8i16(
+  <vscale x 8 x i32>,
+  <vscale x 8 x i16>,
+  <vscale x 8 x i16>,
+  <vscale x 8 x i1>,
+  i64);
+
+define <vscale x 8 x i32>  @intrinsic_vwmaccsu_mask_vv_nxv8i32_nxv8i16_nxv8i16(<vscale x 8 x i32> %0, <vscale x 8 x i16> %1, <vscale x 8 x i16> %2, <vscale x 8 x i1> %3, i64 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccsu_mask_vv_nxv8i32_nxv8i16_nxv8i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e16,m2,ta,mu
+; CHECK-NEXT:    vwmaccsu.vv v16, v20, v22, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i32> @llvm.riscv.vwmaccsu.mask.nxv8i32.nxv8i16(
+    <vscale x 8 x i32> %0,
+    <vscale x 8 x i16> %1,
+    <vscale x 8 x i16> %2,
+    <vscale x 8 x i1> %3,
+    i64 %4)
+
+  ret <vscale x 8 x i32> %a
+}
+
+declare <vscale x 16 x i32> @llvm.riscv.vwmaccsu.nxv16i32.nxv16i16(
+  <vscale x 16 x i32>,
+  <vscale x 16 x i16>,
+  <vscale x 16 x i16>,
+  i64);
+
+define <vscale x 16 x i32>  @intrinsic_vwmaccsu_vv_nxv16i32_nxv16i16_nxv16i16(<vscale x 16 x i32> %0, <vscale x 16 x i16> %1, <vscale x 16 x i16> %2, i64 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccsu_vv_nxv16i32_nxv16i16_nxv16i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a3, zero, e16,m4,ta,mu
+; CHECK-NEXT:    vle16.v v28, (a1)
+; CHECK-NEXT:    vle16.v v8, (a0)
+; CHECK-NEXT:    vsetvli a0, a2, e16,m4,ta,mu
+; CHECK-NEXT:    vwmaccsu.vv v16, v8, v28
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 16 x i32> @llvm.riscv.vwmaccsu.nxv16i32.nxv16i16(
+    <vscale x 16 x i32> %0,
+    <vscale x 16 x i16> %1,
+    <vscale x 16 x i16> %2,
+    i64 %3)
+
+  ret <vscale x 16 x i32> %a
+}
+
+declare <vscale x 16 x i32> @llvm.riscv.vwmaccsu.mask.nxv16i32.nxv16i16(
+  <vscale x 16 x i32>,
+  <vscale x 16 x i16>,
+  <vscale x 16 x i16>,
+  <vscale x 16 x i1>,
+  i64);
+
+define <vscale x 16 x i32>  @intrinsic_vwmaccsu_mask_vv_nxv16i32_nxv16i16_nxv16i16(<vscale x 16 x i32> %0, <vscale x 16 x i16> %1, <vscale x 16 x i16> %2, <vscale x 16 x i1> %3, i64 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccsu_mask_vv_nxv16i32_nxv16i16_nxv16i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a3, zero, e16,m4,ta,mu
+; CHECK-NEXT:    vle16.v v28, (a1)
+; CHECK-NEXT:    vle16.v v8, (a0)
+; CHECK-NEXT:    vsetvli a0, a2, e16,m4,ta,mu
+; CHECK-NEXT:    vwmaccsu.vv v16, v8, v28, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 16 x i32> @llvm.riscv.vwmaccsu.mask.nxv16i32.nxv16i16(
+    <vscale x 16 x i32> %0,
+    <vscale x 16 x i16> %1,
+    <vscale x 16 x i16> %2,
+    <vscale x 16 x i1> %3,
+    i64 %4)
+
+  ret <vscale x 16 x i32> %a
+}
+
+declare <vscale x 1 x i64> @llvm.riscv.vwmaccsu.nxv1i64.nxv1i32(
+  <vscale x 1 x i64>,
+  <vscale x 1 x i32>,
+  <vscale x 1 x i32>,
+  i64);
+
+define <vscale x 1 x i64>  @intrinsic_vwmaccsu_vv_nxv1i64_nxv1i32_nxv1i32(<vscale x 1 x i64> %0, <vscale x 1 x i32> %1, <vscale x 1 x i32> %2, i64 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccsu_vv_nxv1i64_nxv1i32_nxv1i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e32,mf2,ta,mu
+; CHECK-NEXT:    vwmaccsu.vv v16, v17, v18
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i64> @llvm.riscv.vwmaccsu.nxv1i64.nxv1i32(
+    <vscale x 1 x i64> %0,
+    <vscale x 1 x i32> %1,
+    <vscale x 1 x i32> %2,
+    i64 %3)
+
+  ret <vscale x 1 x i64> %a
+}
+
+declare <vscale x 1 x i64> @llvm.riscv.vwmaccsu.mask.nxv1i64.nxv1i32(
+  <vscale x 1 x i64>,
+  <vscale x 1 x i32>,
+  <vscale x 1 x i32>,
+  <vscale x 1 x i1>,
+  i64);
+
+define <vscale x 1 x i64>  @intrinsic_vwmaccsu_mask_vv_nxv1i64_nxv1i32_nxv1i32(<vscale x 1 x i64> %0, <vscale x 1 x i32> %1, <vscale x 1 x i32> %2, <vscale x 1 x i1> %3, i64 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccsu_mask_vv_nxv1i64_nxv1i32_nxv1i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e32,mf2,ta,mu
+; CHECK-NEXT:    vwmaccsu.vv v16, v17, v18, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i64> @llvm.riscv.vwmaccsu.mask.nxv1i64.nxv1i32(
+    <vscale x 1 x i64> %0,
+    <vscale x 1 x i32> %1,
+    <vscale x 1 x i32> %2,
+    <vscale x 1 x i1> %3,
+    i64 %4)
+
+  ret <vscale x 1 x i64> %a
+}
+
+declare <vscale x 2 x i64> @llvm.riscv.vwmaccsu.nxv2i64.nxv2i32(
+  <vscale x 2 x i64>,
+  <vscale x 2 x i32>,
+  <vscale x 2 x i32>,
+  i64);
+
+define <vscale x 2 x i64>  @intrinsic_vwmaccsu_vv_nxv2i64_nxv2i32_nxv2i32(<vscale x 2 x i64> %0, <vscale x 2 x i32> %1, <vscale x 2 x i32> %2, i64 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccsu_vv_nxv2i64_nxv2i32_nxv2i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e32,m1,ta,mu
+; CHECK-NEXT:    vwmaccsu.vv v16, v18, v19
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i64> @llvm.riscv.vwmaccsu.nxv2i64.nxv2i32(
+    <vscale x 2 x i64> %0,
+    <vscale x 2 x i32> %1,
+    <vscale x 2 x i32> %2,
+    i64 %3)
+
+  ret <vscale x 2 x i64> %a
+}
+
+declare <vscale x 2 x i64> @llvm.riscv.vwmaccsu.mask.nxv2i64.nxv2i32(
+  <vscale x 2 x i64>,
+  <vscale x 2 x i32>,
+  <vscale x 2 x i32>,
+  <vscale x 2 x i1>,
+  i64);
+
+define <vscale x 2 x i64>  @intrinsic_vwmaccsu_mask_vv_nxv2i64_nxv2i32_nxv2i32(<vscale x 2 x i64> %0, <vscale x 2 x i32> %1, <vscale x 2 x i32> %2, <vscale x 2 x i1> %3, i64 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccsu_mask_vv_nxv2i64_nxv2i32_nxv2i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e32,m1,ta,mu
+; CHECK-NEXT:    vwmaccsu.vv v16, v18, v19, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i64> @llvm.riscv.vwmaccsu.mask.nxv2i64.nxv2i32(
+    <vscale x 2 x i64> %0,
+    <vscale x 2 x i32> %1,
+    <vscale x 2 x i32> %2,
+    <vscale x 2 x i1> %3,
+    i64 %4)
+
+  ret <vscale x 2 x i64> %a
+}
+
+declare <vscale x 4 x i64> @llvm.riscv.vwmaccsu.nxv4i64.nxv4i32(
+  <vscale x 4 x i64>,
+  <vscale x 4 x i32>,
+  <vscale x 4 x i32>,
+  i64);
+
+define <vscale x 4 x i64>  @intrinsic_vwmaccsu_vv_nxv4i64_nxv4i32_nxv4i32(<vscale x 4 x i64> %0, <vscale x 4 x i32> %1, <vscale x 4 x i32> %2, i64 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccsu_vv_nxv4i64_nxv4i32_nxv4i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e32,m2,ta,mu
+; CHECK-NEXT:    vwmaccsu.vv v16, v20, v22
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i64> @llvm.riscv.vwmaccsu.nxv4i64.nxv4i32(
+    <vscale x 4 x i64> %0,
+    <vscale x 4 x i32> %1,
+    <vscale x 4 x i32> %2,
+    i64 %3)
+
+  ret <vscale x 4 x i64> %a
+}
+
+declare <vscale x 4 x i64> @llvm.riscv.vwmaccsu.mask.nxv4i64.nxv4i32(
+  <vscale x 4 x i64>,
+  <vscale x 4 x i32>,
+  <vscale x 4 x i32>,
+  <vscale x 4 x i1>,
+  i64);
+
+define <vscale x 4 x i64>  @intrinsic_vwmaccsu_mask_vv_nxv4i64_nxv4i32_nxv4i32(<vscale x 4 x i64> %0, <vscale x 4 x i32> %1, <vscale x 4 x i32> %2, <vscale x 4 x i1> %3, i64 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccsu_mask_vv_nxv4i64_nxv4i32_nxv4i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e32,m2,ta,mu
+; CHECK-NEXT:    vwmaccsu.vv v16, v20, v22, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i64> @llvm.riscv.vwmaccsu.mask.nxv4i64.nxv4i32(
+    <vscale x 4 x i64> %0,
+    <vscale x 4 x i32> %1,
+    <vscale x 4 x i32> %2,
+    <vscale x 4 x i1> %3,
+    i64 %4)
+
+  ret <vscale x 4 x i64> %a
+}
+
+declare <vscale x 8 x i64> @llvm.riscv.vwmaccsu.nxv8i64.nxv8i32(
+  <vscale x 8 x i64>,
+  <vscale x 8 x i32>,
+  <vscale x 8 x i32>,
+  i64);
+
+define <vscale x 8 x i64>  @intrinsic_vwmaccsu_vv_nxv8i64_nxv8i32_nxv8i32(<vscale x 8 x i64> %0, <vscale x 8 x i32> %1, <vscale x 8 x i32> %2, i64 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccsu_vv_nxv8i64_nxv8i32_nxv8i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a3, zero, e32,m4,ta,mu
+; CHECK-NEXT:    vle32.v v28, (a1)
+; CHECK-NEXT:    vle32.v v8, (a0)
+; CHECK-NEXT:    vsetvli a0, a2, e32,m4,ta,mu
+; CHECK-NEXT:    vwmaccsu.vv v16, v8, v28
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i64> @llvm.riscv.vwmaccsu.nxv8i64.nxv8i32(
+    <vscale x 8 x i64> %0,
+    <vscale x 8 x i32> %1,
+    <vscale x 8 x i32> %2,
+    i64 %3)
+
+  ret <vscale x 8 x i64> %a
+}
+
+declare <vscale x 8 x i64> @llvm.riscv.vwmaccsu.mask.nxv8i64.nxv8i32(
+  <vscale x 8 x i64>,
+  <vscale x 8 x i32>,
+  <vscale x 8 x i32>,
+  <vscale x 8 x i1>,
+  i64);
+
+define <vscale x 8 x i64>  @intrinsic_vwmaccsu_mask_vv_nxv8i64_nxv8i32_nxv8i32(<vscale x 8 x i64> %0, <vscale x 8 x i32> %1, <vscale x 8 x i32> %2, <vscale x 8 x i1> %3, i64 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccsu_mask_vv_nxv8i64_nxv8i32_nxv8i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a3, zero, e32,m4,ta,mu
+; CHECK-NEXT:    vle32.v v28, (a1)
+; CHECK-NEXT:    vle32.v v8, (a0)
+; CHECK-NEXT:    vsetvli a0, a2, e32,m4,ta,mu
+; CHECK-NEXT:    vwmaccsu.vv v16, v8, v28, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i64> @llvm.riscv.vwmaccsu.mask.nxv8i64.nxv8i32(
+    <vscale x 8 x i64> %0,
+    <vscale x 8 x i32> %1,
+    <vscale x 8 x i32> %2,
+    <vscale x 8 x i1> %3,
+    i64 %4)
+
+  ret <vscale x 8 x i64> %a
+}
+
+declare <vscale x 1 x i16> @llvm.riscv.vwmaccsu.nxv1i16.i8(
+  <vscale x 1 x i16>,
+  i8,
+  <vscale x 1 x i8>,
+  i64);
+
+define <vscale x 1 x i16>  @intrinsic_vwmaccsu_vx_nxv1i16_i8_nxv1i8(<vscale x 1 x i16> %0, i8 %1, <vscale x 1 x i8> %2, i64 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccsu_vx_nxv1i16_i8_nxv1i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e8,mf8,ta,mu
+; CHECK-NEXT:    vwmaccsu.vx v16, a0, v17
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i16> @llvm.riscv.vwmaccsu.nxv1i16.i8(
+    <vscale x 1 x i16> %0,
+    i8 %1,
+    <vscale x 1 x i8> %2,
+    i64 %3)
+
+  ret <vscale x 1 x i16> %a
+}
+
+declare <vscale x 1 x i16> @llvm.riscv.vwmaccsu.mask.nxv1i16.i8(
+  <vscale x 1 x i16>,
+  i8,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i1>,
+  i64);
+
+define <vscale x 1 x i16> @intrinsic_vwmaccsu_mask_vx_nxv1i16_i8_nxv1i8(<vscale x 1 x i16> %0, i8 %1, <vscale x 1 x i8> %2, <vscale x 1 x i1> %3, i64 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccsu_mask_vx_nxv1i16_i8_nxv1i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e8,mf8,ta,mu
+; CHECK-NEXT:    vwmaccsu.vx v16, a0, v17, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i16> @llvm.riscv.vwmaccsu.mask.nxv1i16.i8(
+    <vscale x 1 x i16> %0,
+    i8 %1,
+    <vscale x 1 x i8> %2,
+    <vscale x 1 x i1> %3,
+    i64 %4)
+
+  ret <vscale x 1 x i16> %a
+}
+
+declare <vscale x 2 x i16> @llvm.riscv.vwmaccsu.nxv2i16.i8(
+  <vscale x 2 x i16>,
+  i8,
+  <vscale x 2 x i8>,
+  i64);
+
+define <vscale x 2 x i16>  @intrinsic_vwmaccsu_vx_nxv2i16_i8_nxv2i8(<vscale x 2 x i16> %0, i8 %1, <vscale x 2 x i8> %2, i64 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccsu_vx_nxv2i16_i8_nxv2i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e8,mf4,ta,mu
+; CHECK-NEXT:    vwmaccsu.vx v16, a0, v17
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i16> @llvm.riscv.vwmaccsu.nxv2i16.i8(
+    <vscale x 2 x i16> %0,
+    i8 %1,
+    <vscale x 2 x i8> %2,
+    i64 %3)
+
+  ret <vscale x 2 x i16> %a
+}
+
+declare <vscale x 2 x i16> @llvm.riscv.vwmaccsu.mask.nxv2i16.i8(
+  <vscale x 2 x i16>,
+  i8,
+  <vscale x 2 x i8>,
+  <vscale x 2 x i1>,
+  i64);
+
+define <vscale x 2 x i16> @intrinsic_vwmaccsu_mask_vx_nxv2i16_i8_nxv2i8(<vscale x 2 x i16> %0, i8 %1, <vscale x 2 x i8> %2, <vscale x 2 x i1> %3, i64 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccsu_mask_vx_nxv2i16_i8_nxv2i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e8,mf4,ta,mu
+; CHECK-NEXT:    vwmaccsu.vx v16, a0, v17, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i16> @llvm.riscv.vwmaccsu.mask.nxv2i16.i8(
+    <vscale x 2 x i16> %0,
+    i8 %1,
+    <vscale x 2 x i8> %2,
+    <vscale x 2 x i1> %3,
+    i64 %4)
+
+  ret <vscale x 2 x i16> %a
+}
+
+declare <vscale x 4 x i16> @llvm.riscv.vwmaccsu.nxv4i16.i8(
+  <vscale x 4 x i16>,
+  i8,
+  <vscale x 4 x i8>,
+  i64);
+
+define <vscale x 4 x i16>  @intrinsic_vwmaccsu_vx_nxv4i16_i8_nxv4i8(<vscale x 4 x i16> %0, i8 %1, <vscale x 4 x i8> %2, i64 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccsu_vx_nxv4i16_i8_nxv4i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e8,mf2,ta,mu
+; CHECK-NEXT:    vwmaccsu.vx v16, a0, v17
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i16> @llvm.riscv.vwmaccsu.nxv4i16.i8(
+    <vscale x 4 x i16> %0,
+    i8 %1,
+    <vscale x 4 x i8> %2,
+    i64 %3)
+
+  ret <vscale x 4 x i16> %a
+}
+
+declare <vscale x 4 x i16> @llvm.riscv.vwmaccsu.mask.nxv4i16.i8(
+  <vscale x 4 x i16>,
+  i8,
+  <vscale x 4 x i8>,
+  <vscale x 4 x i1>,
+  i64);
+
+define <vscale x 4 x i16> @intrinsic_vwmaccsu_mask_vx_nxv4i16_i8_nxv4i8(<vscale x 4 x i16> %0, i8 %1, <vscale x 4 x i8> %2, <vscale x 4 x i1> %3, i64 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccsu_mask_vx_nxv4i16_i8_nxv4i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e8,mf2,ta,mu
+; CHECK-NEXT:    vwmaccsu.vx v16, a0, v17, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i16> @llvm.riscv.vwmaccsu.mask.nxv4i16.i8(
+    <vscale x 4 x i16> %0,
+    i8 %1,
+    <vscale x 4 x i8> %2,
+    <vscale x 4 x i1> %3,
+    i64 %4)
+
+  ret <vscale x 4 x i16> %a
+}
+
+declare <vscale x 8 x i16> @llvm.riscv.vwmaccsu.nxv8i16.i8(
+  <vscale x 8 x i16>,
+  i8,
+  <vscale x 8 x i8>,
+  i64);
+
+define <vscale x 8 x i16>  @intrinsic_vwmaccsu_vx_nxv8i16_i8_nxv8i8(<vscale x 8 x i16> %0, i8 %1, <vscale x 8 x i8> %2, i64 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccsu_vx_nxv8i16_i8_nxv8i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e8,m1,ta,mu
+; CHECK-NEXT:    vwmaccsu.vx v16, a0, v18
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i16> @llvm.riscv.vwmaccsu.nxv8i16.i8(
+    <vscale x 8 x i16> %0,
+    i8 %1,
+    <vscale x 8 x i8> %2,
+    i64 %3)
+
+  ret <vscale x 8 x i16> %a
+}
+
+declare <vscale x 8 x i16> @llvm.riscv.vwmaccsu.mask.nxv8i16.i8(
+  <vscale x 8 x i16>,
+  i8,
+  <vscale x 8 x i8>,
+  <vscale x 8 x i1>,
+  i64);
+
+define <vscale x 8 x i16> @intrinsic_vwmaccsu_mask_vx_nxv8i16_i8_nxv8i8(<vscale x 8 x i16> %0, i8 %1, <vscale x 8 x i8> %2, <vscale x 8 x i1> %3, i64 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccsu_mask_vx_nxv8i16_i8_nxv8i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e8,m1,ta,mu
+; CHECK-NEXT:    vwmaccsu.vx v16, a0, v18, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i16> @llvm.riscv.vwmaccsu.mask.nxv8i16.i8(
+    <vscale x 8 x i16> %0,
+    i8 %1,
+    <vscale x 8 x i8> %2,
+    <vscale x 8 x i1> %3,
+    i64 %4)
+
+  ret <vscale x 8 x i16> %a
+}
+
+declare <vscale x 16 x i16> @llvm.riscv.vwmaccsu.nxv16i16.i8(
+  <vscale x 16 x i16>,
+  i8,
+  <vscale x 16 x i8>,
+  i64);
+
+define <vscale x 16 x i16>  @intrinsic_vwmaccsu_vx_nxv16i16_i8_nxv16i8(<vscale x 16 x i16> %0, i8 %1, <vscale x 16 x i8> %2, i64 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccsu_vx_nxv16i16_i8_nxv16i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e8,m2,ta,mu
+; CHECK-NEXT:    vwmaccsu.vx v16, a0, v20
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 16 x i16> @llvm.riscv.vwmaccsu.nxv16i16.i8(
+    <vscale x 16 x i16> %0,
+    i8 %1,
+    <vscale x 16 x i8> %2,
+    i64 %3)
+
+  ret <vscale x 16 x i16> %a
+}
+
+declare <vscale x 16 x i16> @llvm.riscv.vwmaccsu.mask.nxv16i16.i8(
+  <vscale x 16 x i16>,
+  i8,
+  <vscale x 16 x i8>,
+  <vscale x 16 x i1>,
+  i64);
+
+define <vscale x 16 x i16> @intrinsic_vwmaccsu_mask_vx_nxv16i16_i8_nxv16i8(<vscale x 16 x i16> %0, i8 %1, <vscale x 16 x i8> %2, <vscale x 16 x i1> %3, i64 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccsu_mask_vx_nxv16i16_i8_nxv16i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e8,m2,ta,mu
+; CHECK-NEXT:    vwmaccsu.vx v16, a0, v20, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 16 x i16> @llvm.riscv.vwmaccsu.mask.nxv16i16.i8(
+    <vscale x 16 x i16> %0,
+    i8 %1,
+    <vscale x 16 x i8> %2,
+    <vscale x 16 x i1> %3,
+    i64 %4)
+
+  ret <vscale x 16 x i16> %a
+}
+
+declare <vscale x 32 x i16> @llvm.riscv.vwmaccsu.nxv32i16.i8(
+  <vscale x 32 x i16>,
+  i8,
+  <vscale x 32 x i8>,
+  i64);
+
+define <vscale x 32 x i16>  @intrinsic_vwmaccsu_vx_nxv32i16_i8_nxv32i8(<vscale x 32 x i16> %0, i8 %1, <vscale x 32 x i8> %2, i64 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccsu_vx_nxv32i16_i8_nxv32i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a3, zero, e8,m4,ta,mu
+; CHECK-NEXT:    vle8.v v28, (a1)
+; CHECK-NEXT:    vsetvli a1, a2, e8,m4,ta,mu
+; CHECK-NEXT:    vwmaccsu.vx v16, a0, v28
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 32 x i16> @llvm.riscv.vwmaccsu.nxv32i16.i8(
+    <vscale x 32 x i16> %0,
+    i8 %1,
+    <vscale x 32 x i8> %2,
+    i64 %3)
+
+  ret <vscale x 32 x i16> %a
+}
+
+declare <vscale x 32 x i16> @llvm.riscv.vwmaccsu.mask.nxv32i16.i8(
+  <vscale x 32 x i16>,
+  i8,
+  <vscale x 32 x i8>,
+  <vscale x 32 x i1>,
+  i64);
+
+define <vscale x 32 x i16> @intrinsic_vwmaccsu_mask_vx_nxv32i16_i8_nxv32i8(<vscale x 32 x i16> %0, i8 %1, <vscale x 32 x i8> %2, <vscale x 32 x i1> %3, i64 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccsu_mask_vx_nxv32i16_i8_nxv32i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a3, zero, e8,m4,ta,mu
+; CHECK-NEXT:    vle8.v v28, (a1)
+; CHECK-NEXT:    vsetvli a1, a2, e8,m4,ta,mu
+; CHECK-NEXT:    vwmaccsu.vx v16, a0, v28, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 32 x i16> @llvm.riscv.vwmaccsu.mask.nxv32i16.i8(
+    <vscale x 32 x i16> %0,
+    i8 %1,
+    <vscale x 32 x i8> %2,
+    <vscale x 32 x i1> %3,
+    i64 %4)
+
+  ret <vscale x 32 x i16> %a
+}
+
+declare <vscale x 1 x i32> @llvm.riscv.vwmaccsu.nxv1i32.i16(
+  <vscale x 1 x i32>,
+  i16,
+  <vscale x 1 x i16>,
+  i64);
+
+define <vscale x 1 x i32>  @intrinsic_vwmaccsu_vx_nxv1i32_i16_nxv1i16(<vscale x 1 x i32> %0, i16 %1, <vscale x 1 x i16> %2, i64 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccsu_vx_nxv1i32_i16_nxv1i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e16,mf4,ta,mu
+; CHECK-NEXT:    vwmaccsu.vx v16, a0, v17
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i32> @llvm.riscv.vwmaccsu.nxv1i32.i16(
+    <vscale x 1 x i32> %0,
+    i16 %1,
+    <vscale x 1 x i16> %2,
+    i64 %3)
+
+  ret <vscale x 1 x i32> %a
+}
+
+declare <vscale x 1 x i32> @llvm.riscv.vwmaccsu.mask.nxv1i32.i16(
+  <vscale x 1 x i32>,
+  i16,
+  <vscale x 1 x i16>,
+  <vscale x 1 x i1>,
+  i64);
+
+define <vscale x 1 x i32> @intrinsic_vwmaccsu_mask_vx_nxv1i32_i16_nxv1i16(<vscale x 1 x i32> %0, i16 %1, <vscale x 1 x i16> %2, <vscale x 1 x i1> %3, i64 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccsu_mask_vx_nxv1i32_i16_nxv1i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e16,mf4,ta,mu
+; CHECK-NEXT:    vwmaccsu.vx v16, a0, v17, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i32> @llvm.riscv.vwmaccsu.mask.nxv1i32.i16(
+    <vscale x 1 x i32> %0,
+    i16 %1,
+    <vscale x 1 x i16> %2,
+    <vscale x 1 x i1> %3,
+    i64 %4)
+
+  ret <vscale x 1 x i32> %a
+}
+
+declare <vscale x 2 x i32> @llvm.riscv.vwmaccsu.nxv2i32.i16(
+  <vscale x 2 x i32>,
+  i16,
+  <vscale x 2 x i16>,
+  i64);
+
+define <vscale x 2 x i32>  @intrinsic_vwmaccsu_vx_nxv2i32_i16_nxv2i16(<vscale x 2 x i32> %0, i16 %1, <vscale x 2 x i16> %2, i64 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccsu_vx_nxv2i32_i16_nxv2i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e16,mf2,ta,mu
+; CHECK-NEXT:    vwmaccsu.vx v16, a0, v17
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i32> @llvm.riscv.vwmaccsu.nxv2i32.i16(
+    <vscale x 2 x i32> %0,
+    i16 %1,
+    <vscale x 2 x i16> %2,
+    i64 %3)
+
+  ret <vscale x 2 x i32> %a
+}
+
+declare <vscale x 2 x i32> @llvm.riscv.vwmaccsu.mask.nxv2i32.i16(
+  <vscale x 2 x i32>,
+  i16,
+  <vscale x 2 x i16>,
+  <vscale x 2 x i1>,
+  i64);
+
+define <vscale x 2 x i32> @intrinsic_vwmaccsu_mask_vx_nxv2i32_i16_nxv2i16(<vscale x 2 x i32> %0, i16 %1, <vscale x 2 x i16> %2, <vscale x 2 x i1> %3, i64 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccsu_mask_vx_nxv2i32_i16_nxv2i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e16,mf2,ta,mu
+; CHECK-NEXT:    vwmaccsu.vx v16, a0, v17, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i32> @llvm.riscv.vwmaccsu.mask.nxv2i32.i16(
+    <vscale x 2 x i32> %0,
+    i16 %1,
+    <vscale x 2 x i16> %2,
+    <vscale x 2 x i1> %3,
+    i64 %4)
+
+  ret <vscale x 2 x i32> %a
+}
+
+declare <vscale x 4 x i32> @llvm.riscv.vwmaccsu.nxv4i32.i16(
+  <vscale x 4 x i32>,
+  i16,
+  <vscale x 4 x i16>,
+  i64);
+
+define <vscale x 4 x i32>  @intrinsic_vwmaccsu_vx_nxv4i32_i16_nxv4i16(<vscale x 4 x i32> %0, i16 %1, <vscale x 4 x i16> %2, i64 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccsu_vx_nxv4i32_i16_nxv4i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e16,m1,ta,mu
+; CHECK-NEXT:    vwmaccsu.vx v16, a0, v18
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i32> @llvm.riscv.vwmaccsu.nxv4i32.i16(
+    <vscale x 4 x i32> %0,
+    i16 %1,
+    <vscale x 4 x i16> %2,
+    i64 %3)
+
+  ret <vscale x 4 x i32> %a
+}
+
+declare <vscale x 4 x i32> @llvm.riscv.vwmaccsu.mask.nxv4i32.i16(
+  <vscale x 4 x i32>,
+  i16,
+  <vscale x 4 x i16>,
+  <vscale x 4 x i1>,
+  i64);
+
+define <vscale x 4 x i32> @intrinsic_vwmaccsu_mask_vx_nxv4i32_i16_nxv4i16(<vscale x 4 x i32> %0, i16 %1, <vscale x 4 x i16> %2, <vscale x 4 x i1> %3, i64 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccsu_mask_vx_nxv4i32_i16_nxv4i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e16,m1,ta,mu
+; CHECK-NEXT:    vwmaccsu.vx v16, a0, v18, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i32> @llvm.riscv.vwmaccsu.mask.nxv4i32.i16(
+    <vscale x 4 x i32> %0,
+    i16 %1,
+    <vscale x 4 x i16> %2,
+    <vscale x 4 x i1> %3,
+    i64 %4)
+
+  ret <vscale x 4 x i32> %a
+}
+
+declare <vscale x 8 x i32> @llvm.riscv.vwmaccsu.nxv8i32.i16(
+  <vscale x 8 x i32>,
+  i16,
+  <vscale x 8 x i16>,
+  i64);
+
+define <vscale x 8 x i32>  @intrinsic_vwmaccsu_vx_nxv8i32_i16_nxv8i16(<vscale x 8 x i32> %0, i16 %1, <vscale x 8 x i16> %2, i64 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccsu_vx_nxv8i32_i16_nxv8i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e16,m2,ta,mu
+; CHECK-NEXT:    vwmaccsu.vx v16, a0, v20
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i32> @llvm.riscv.vwmaccsu.nxv8i32.i16(
+    <vscale x 8 x i32> %0,
+    i16 %1,
+    <vscale x 8 x i16> %2,
+    i64 %3)
+
+  ret <vscale x 8 x i32> %a
+}
+
+declare <vscale x 8 x i32> @llvm.riscv.vwmaccsu.mask.nxv8i32.i16(
+  <vscale x 8 x i32>,
+  i16,
+  <vscale x 8 x i16>,
+  <vscale x 8 x i1>,
+  i64);
+
+define <vscale x 8 x i32> @intrinsic_vwmaccsu_mask_vx_nxv8i32_i16_nxv8i16(<vscale x 8 x i32> %0, i16 %1, <vscale x 8 x i16> %2, <vscale x 8 x i1> %3, i64 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccsu_mask_vx_nxv8i32_i16_nxv8i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e16,m2,ta,mu
+; CHECK-NEXT:    vwmaccsu.vx v16, a0, v20, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i32> @llvm.riscv.vwmaccsu.mask.nxv8i32.i16(
+    <vscale x 8 x i32> %0,
+    i16 %1,
+    <vscale x 8 x i16> %2,
+    <vscale x 8 x i1> %3,
+    i64 %4)
+
+  ret <vscale x 8 x i32> %a
+}
+
+declare <vscale x 16 x i32> @llvm.riscv.vwmaccsu.nxv16i32.i16(
+  <vscale x 16 x i32>,
+  i16,
+  <vscale x 16 x i16>,
+  i64);
+
+define <vscale x 16 x i32>  @intrinsic_vwmaccsu_vx_nxv16i32_i16_nxv16i16(<vscale x 16 x i32> %0, i16 %1, <vscale x 16 x i16> %2, i64 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccsu_vx_nxv16i32_i16_nxv16i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a3, zero, e16,m4,ta,mu
+; CHECK-NEXT:    vle16.v v28, (a1)
+; CHECK-NEXT:    vsetvli a1, a2, e16,m4,ta,mu
+; CHECK-NEXT:    vwmaccsu.vx v16, a0, v28
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 16 x i32> @llvm.riscv.vwmaccsu.nxv16i32.i16(
+    <vscale x 16 x i32> %0,
+    i16 %1,
+    <vscale x 16 x i16> %2,
+    i64 %3)
+
+  ret <vscale x 16 x i32> %a
+}
+
+declare <vscale x 16 x i32> @llvm.riscv.vwmaccsu.mask.nxv16i32.i16(
+  <vscale x 16 x i32>,
+  i16,
+  <vscale x 16 x i16>,
+  <vscale x 16 x i1>,
+  i64);
+
+define <vscale x 16 x i32> @intrinsic_vwmaccsu_mask_vx_nxv16i32_i16_nxv16i16(<vscale x 16 x i32> %0, i16 %1, <vscale x 16 x i16> %2, <vscale x 16 x i1> %3, i64 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccsu_mask_vx_nxv16i32_i16_nxv16i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a3, zero, e16,m4,ta,mu
+; CHECK-NEXT:    vle16.v v28, (a1)
+; CHECK-NEXT:    vsetvli a1, a2, e16,m4,ta,mu
+; CHECK-NEXT:    vwmaccsu.vx v16, a0, v28, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 16 x i32> @llvm.riscv.vwmaccsu.mask.nxv16i32.i16(
+    <vscale x 16 x i32> %0,
+    i16 %1,
+    <vscale x 16 x i16> %2,
+    <vscale x 16 x i1> %3,
+    i64 %4)
+
+  ret <vscale x 16 x i32> %a
+}
+
+declare <vscale x 1 x i64> @llvm.riscv.vwmaccsu.nxv1i64.i32(
+  <vscale x 1 x i64>,
+  i32,
+  <vscale x 1 x i32>,
+  i64);
+
+define <vscale x 1 x i64>  @intrinsic_vwmaccsu_vx_nxv1i64_i32_nxv1i32(<vscale x 1 x i64> %0, i32 %1, <vscale x 1 x i32> %2, i64 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccsu_vx_nxv1i64_i32_nxv1i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e32,mf2,ta,mu
+; CHECK-NEXT:    vwmaccsu.vx v16, a0, v17
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i64> @llvm.riscv.vwmaccsu.nxv1i64.i32(
+    <vscale x 1 x i64> %0,
+    i32 %1,
+    <vscale x 1 x i32> %2,
+    i64 %3)
+
+  ret <vscale x 1 x i64> %a
+}
+
+declare <vscale x 1 x i64> @llvm.riscv.vwmaccsu.mask.nxv1i64.i32(
+  <vscale x 1 x i64>,
+  i32,
+  <vscale x 1 x i32>,
+  <vscale x 1 x i1>,
+  i64);
+
+define <vscale x 1 x i64> @intrinsic_vwmaccsu_mask_vx_nxv1i64_i32_nxv1i32(<vscale x 1 x i64> %0, i32 %1, <vscale x 1 x i32> %2, <vscale x 1 x i1> %3, i64 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccsu_mask_vx_nxv1i64_i32_nxv1i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e32,mf2,ta,mu
+; CHECK-NEXT:    vwmaccsu.vx v16, a0, v17, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i64> @llvm.riscv.vwmaccsu.mask.nxv1i64.i32(
+    <vscale x 1 x i64> %0,
+    i32 %1,
+    <vscale x 1 x i32> %2,
+    <vscale x 1 x i1> %3,
+    i64 %4)
+
+  ret <vscale x 1 x i64> %a
+}
+
+declare <vscale x 2 x i64> @llvm.riscv.vwmaccsu.nxv2i64.i32(
+  <vscale x 2 x i64>,
+  i32,
+  <vscale x 2 x i32>,
+  i64);
+
+define <vscale x 2 x i64>  @intrinsic_vwmaccsu_vx_nxv2i64_i32_nxv2i32(<vscale x 2 x i64> %0, i32 %1, <vscale x 2 x i32> %2, i64 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccsu_vx_nxv2i64_i32_nxv2i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e32,m1,ta,mu
+; CHECK-NEXT:    vwmaccsu.vx v16, a0, v18
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i64> @llvm.riscv.vwmaccsu.nxv2i64.i32(
+    <vscale x 2 x i64> %0,
+    i32 %1,
+    <vscale x 2 x i32> %2,
+    i64 %3)
+
+  ret <vscale x 2 x i64> %a
+}
+
+declare <vscale x 2 x i64> @llvm.riscv.vwmaccsu.mask.nxv2i64.i32(
+  <vscale x 2 x i64>,
+  i32,
+  <vscale x 2 x i32>,
+  <vscale x 2 x i1>,
+  i64);
+
+define <vscale x 2 x i64> @intrinsic_vwmaccsu_mask_vx_nxv2i64_i32_nxv2i32(<vscale x 2 x i64> %0, i32 %1, <vscale x 2 x i32> %2, <vscale x 2 x i1> %3, i64 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccsu_mask_vx_nxv2i64_i32_nxv2i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e32,m1,ta,mu
+; CHECK-NEXT:    vwmaccsu.vx v16, a0, v18, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i64> @llvm.riscv.vwmaccsu.mask.nxv2i64.i32(
+    <vscale x 2 x i64> %0,
+    i32 %1,
+    <vscale x 2 x i32> %2,
+    <vscale x 2 x i1> %3,
+    i64 %4)
+
+  ret <vscale x 2 x i64> %a
+}
+
+declare <vscale x 4 x i64> @llvm.riscv.vwmaccsu.nxv4i64.i32(
+  <vscale x 4 x i64>,
+  i32,
+  <vscale x 4 x i32>,
+  i64);
+
+define <vscale x 4 x i64>  @intrinsic_vwmaccsu_vx_nxv4i64_i32_nxv4i32(<vscale x 4 x i64> %0, i32 %1, <vscale x 4 x i32> %2, i64 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccsu_vx_nxv4i64_i32_nxv4i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e32,m2,ta,mu
+; CHECK-NEXT:    vwmaccsu.vx v16, a0, v20
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i64> @llvm.riscv.vwmaccsu.nxv4i64.i32(
+    <vscale x 4 x i64> %0,
+    i32 %1,
+    <vscale x 4 x i32> %2,
+    i64 %3)
+
+  ret <vscale x 4 x i64> %a
+}
+
+declare <vscale x 4 x i64> @llvm.riscv.vwmaccsu.mask.nxv4i64.i32(
+  <vscale x 4 x i64>,
+  i32,
+  <vscale x 4 x i32>,
+  <vscale x 4 x i1>,
+  i64);
+
+define <vscale x 4 x i64> @intrinsic_vwmaccsu_mask_vx_nxv4i64_i32_nxv4i32(<vscale x 4 x i64> %0, i32 %1, <vscale x 4 x i32> %2, <vscale x 4 x i1> %3, i64 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccsu_mask_vx_nxv4i64_i32_nxv4i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e32,m2,ta,mu
+; CHECK-NEXT:    vwmaccsu.vx v16, a0, v20, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i64> @llvm.riscv.vwmaccsu.mask.nxv4i64.i32(
+    <vscale x 4 x i64> %0,
+    i32 %1,
+    <vscale x 4 x i32> %2,
+    <vscale x 4 x i1> %3,
+    i64 %4)
+
+  ret <vscale x 4 x i64> %a
+}
+
+declare <vscale x 8 x i64> @llvm.riscv.vwmaccsu.nxv8i64.i32(
+  <vscale x 8 x i64>,
+  i32,
+  <vscale x 8 x i32>,
+  i64);
+
+define <vscale x 8 x i64>  @intrinsic_vwmaccsu_vx_nxv8i64_i32_nxv8i32(<vscale x 8 x i64> %0, i32 %1, <vscale x 8 x i32> %2, i64 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccsu_vx_nxv8i64_i32_nxv8i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a3, zero, e32,m4,ta,mu
+; CHECK-NEXT:    vle32.v v28, (a1)
+; CHECK-NEXT:    vsetvli a1, a2, e32,m4,ta,mu
+; CHECK-NEXT:    vwmaccsu.vx v16, a0, v28
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i64> @llvm.riscv.vwmaccsu.nxv8i64.i32(
+    <vscale x 8 x i64> %0,
+    i32 %1,
+    <vscale x 8 x i32> %2,
+    i64 %3)
+
+  ret <vscale x 8 x i64> %a
+}
+
+declare <vscale x 8 x i64> @llvm.riscv.vwmaccsu.mask.nxv8i64.i32(
+  <vscale x 8 x i64>,
+  i32,
+  <vscale x 8 x i32>,
+  <vscale x 8 x i1>,
+  i64);
+
+define <vscale x 8 x i64> @intrinsic_vwmaccsu_mask_vx_nxv8i64_i32_nxv8i32(<vscale x 8 x i64> %0, i32 %1, <vscale x 8 x i32> %2, <vscale x 8 x i1> %3, i64 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccsu_mask_vx_nxv8i64_i32_nxv8i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a3, zero, e32,m4,ta,mu
+; CHECK-NEXT:    vle32.v v28, (a1)
+; CHECK-NEXT:    vsetvli a1, a2, e32,m4,ta,mu
+; CHECK-NEXT:    vwmaccsu.vx v16, a0, v28, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i64> @llvm.riscv.vwmaccsu.mask.nxv8i64.i32(
+    <vscale x 8 x i64> %0,
+    i32 %1,
+    <vscale x 8 x i32> %2,
+    <vscale x 8 x i1> %3,
+    i64 %4)
+
+  ret <vscale x 8 x i64> %a
+}

llvm/test/CodeGen/RISCV/rvv/vwmaccu-rv32.ll

+; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
+; RUN: llc -mtriple=riscv32 -mattr=+experimental-v,+f -verify-machineinstrs \
+; RUN:   --riscv-no-aliases < %s | FileCheck %s
+declare <vscale x 1 x i16> @llvm.riscv.vwmaccu.nxv1i16.nxv1i8(
+  <vscale x 1 x i16>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  i32);
+
+define <vscale x 1 x i16>  @intrinsic_vwmaccu_vv_nxv1i16_nxv1i8_nxv1i8(<vscale x 1 x i16> %0, <vscale x 1 x i8> %1, <vscale x 1 x i8> %2, i32 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccu_vv_nxv1i16_nxv1i8_nxv1i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e8,mf8,ta,mu
+; CHECK-NEXT:    vwmaccu.vv v16, v17, v18
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i16> @llvm.riscv.vwmaccu.nxv1i16.nxv1i8(
+    <vscale x 1 x i16> %0,
+    <vscale x 1 x i8> %1,
+    <vscale x 1 x i8> %2,
+    i32 %3)
+
+  ret <vscale x 1 x i16> %a
+}
+
+declare <vscale x 1 x i16> @llvm.riscv.vwmaccu.mask.nxv1i16.nxv1i8(
+  <vscale x 1 x i16>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i1>,
+  i32);
+
+define <vscale x 1 x i16>  @intrinsic_vwmaccu_mask_vv_nxv1i16_nxv1i8_nxv1i8(<vscale x 1 x i16> %0, <vscale x 1 x i8> %1, <vscale x 1 x i8> %2, <vscale x 1 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccu_mask_vv_nxv1i16_nxv1i8_nxv1i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e8,mf8,ta,mu
+; CHECK-NEXT:    vwmaccu.vv v16, v17, v18, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i16> @llvm.riscv.vwmaccu.mask.nxv1i16.nxv1i8(
+    <vscale x 1 x i16> %0,
+    <vscale x 1 x i8> %1,
+    <vscale x 1 x i8> %2,
+    <vscale x 1 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 1 x i16> %a
+}
+
+declare <vscale x 2 x i16> @llvm.riscv.vwmaccu.nxv2i16.nxv2i8(
+  <vscale x 2 x i16>,
+  <vscale x 2 x i8>,
+  <vscale x 2 x i8>,
+  i32);
+
+define <vscale x 2 x i16>  @intrinsic_vwmaccu_vv_nxv2i16_nxv2i8_nxv2i8(<vscale x 2 x i16> %0, <vscale x 2 x i8> %1, <vscale x 2 x i8> %2, i32 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccu_vv_nxv2i16_nxv2i8_nxv2i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e8,mf4,ta,mu
+; CHECK-NEXT:    vwmaccu.vv v16, v17, v18
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i16> @llvm.riscv.vwmaccu.nxv2i16.nxv2i8(
+    <vscale x 2 x i16> %0,
+    <vscale x 2 x i8> %1,
+    <vscale x 2 x i8> %2,
+    i32 %3)
+
+  ret <vscale x 2 x i16> %a
+}
+
+declare <vscale x 2 x i16> @llvm.riscv.vwmaccu.mask.nxv2i16.nxv2i8(
+  <vscale x 2 x i16>,
+  <vscale x 2 x i8>,
+  <vscale x 2 x i8>,
+  <vscale x 2 x i1>,
+  i32);
+
+define <vscale x 2 x i16>  @intrinsic_vwmaccu_mask_vv_nxv2i16_nxv2i8_nxv2i8(<vscale x 2 x i16> %0, <vscale x 2 x i8> %1, <vscale x 2 x i8> %2, <vscale x 2 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccu_mask_vv_nxv2i16_nxv2i8_nxv2i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e8,mf4,ta,mu
+; CHECK-NEXT:    vwmaccu.vv v16, v17, v18, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i16> @llvm.riscv.vwmaccu.mask.nxv2i16.nxv2i8(
+    <vscale x 2 x i16> %0,
+    <vscale x 2 x i8> %1,
+    <vscale x 2 x i8> %2,
+    <vscale x 2 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 2 x i16> %a
+}
+
+declare <vscale x 4 x i16> @llvm.riscv.vwmaccu.nxv4i16.nxv4i8(
+  <vscale x 4 x i16>,
+  <vscale x 4 x i8>,
+  <vscale x 4 x i8>,
+  i32);
+
+define <vscale x 4 x i16>  @intrinsic_vwmaccu_vv_nxv4i16_nxv4i8_nxv4i8(<vscale x 4 x i16> %0, <vscale x 4 x i8> %1, <vscale x 4 x i8> %2, i32 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccu_vv_nxv4i16_nxv4i8_nxv4i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e8,mf2,ta,mu
+; CHECK-NEXT:    vwmaccu.vv v16, v17, v18
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i16> @llvm.riscv.vwmaccu.nxv4i16.nxv4i8(
+    <vscale x 4 x i16> %0,
+    <vscale x 4 x i8> %1,
+    <vscale x 4 x i8> %2,
+    i32 %3)
+
+  ret <vscale x 4 x i16> %a
+}
+
+declare <vscale x 4 x i16> @llvm.riscv.vwmaccu.mask.nxv4i16.nxv4i8(
+  <vscale x 4 x i16>,
+  <vscale x 4 x i8>,
+  <vscale x 4 x i8>,
+  <vscale x 4 x i1>,
+  i32);
+
+define <vscale x 4 x i16>  @intrinsic_vwmaccu_mask_vv_nxv4i16_nxv4i8_nxv4i8(<vscale x 4 x i16> %0, <vscale x 4 x i8> %1, <vscale x 4 x i8> %2, <vscale x 4 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccu_mask_vv_nxv4i16_nxv4i8_nxv4i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e8,mf2,ta,mu
+; CHECK-NEXT:    vwmaccu.vv v16, v17, v18, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i16> @llvm.riscv.vwmaccu.mask.nxv4i16.nxv4i8(
+    <vscale x 4 x i16> %0,
+    <vscale x 4 x i8> %1,
+    <vscale x 4 x i8> %2,
+    <vscale x 4 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 4 x i16> %a
+}
+
+declare <vscale x 8 x i16> @llvm.riscv.vwmaccu.nxv8i16.nxv8i8(
+  <vscale x 8 x i16>,
+  <vscale x 8 x i8>,
+  <vscale x 8 x i8>,
+  i32);
+
+define <vscale x 8 x i16>  @intrinsic_vwmaccu_vv_nxv8i16_nxv8i8_nxv8i8(<vscale x 8 x i16> %0, <vscale x 8 x i8> %1, <vscale x 8 x i8> %2, i32 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccu_vv_nxv8i16_nxv8i8_nxv8i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e8,m1,ta,mu
+; CHECK-NEXT:    vwmaccu.vv v16, v18, v19
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i16> @llvm.riscv.vwmaccu.nxv8i16.nxv8i8(
+    <vscale x 8 x i16> %0,
+    <vscale x 8 x i8> %1,
+    <vscale x 8 x i8> %2,
+    i32 %3)
+
+  ret <vscale x 8 x i16> %a
+}
+
+declare <vscale x 8 x i16> @llvm.riscv.vwmaccu.mask.nxv8i16.nxv8i8(
+  <vscale x 8 x i16>,
+  <vscale x 8 x i8>,
+  <vscale x 8 x i8>,
+  <vscale x 8 x i1>,
+  i32);
+
+define <vscale x 8 x i16>  @intrinsic_vwmaccu_mask_vv_nxv8i16_nxv8i8_nxv8i8(<vscale x 8 x i16> %0, <vscale x 8 x i8> %1, <vscale x 8 x i8> %2, <vscale x 8 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccu_mask_vv_nxv8i16_nxv8i8_nxv8i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e8,m1,ta,mu
+; CHECK-NEXT:    vwmaccu.vv v16, v18, v19, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i16> @llvm.riscv.vwmaccu.mask.nxv8i16.nxv8i8(
+    <vscale x 8 x i16> %0,
+    <vscale x 8 x i8> %1,
+    <vscale x 8 x i8> %2,
+    <vscale x 8 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 8 x i16> %a
+}
+
+declare <vscale x 16 x i16> @llvm.riscv.vwmaccu.nxv16i16.nxv16i8(
+  <vscale x 16 x i16>,
+  <vscale x 16 x i8>,
+  <vscale x 16 x i8>,
+  i32);
+
+define <vscale x 16 x i16>  @intrinsic_vwmaccu_vv_nxv16i16_nxv16i8_nxv16i8(<vscale x 16 x i16> %0, <vscale x 16 x i8> %1, <vscale x 16 x i8> %2, i32 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccu_vv_nxv16i16_nxv16i8_nxv16i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e8,m2,ta,mu
+; CHECK-NEXT:    vwmaccu.vv v16, v20, v22
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 16 x i16> @llvm.riscv.vwmaccu.nxv16i16.nxv16i8(
+    <vscale x 16 x i16> %0,
+    <vscale x 16 x i8> %1,
+    <vscale x 16 x i8> %2,
+    i32 %3)
+
+  ret <vscale x 16 x i16> %a
+}
+
+declare <vscale x 16 x i16> @llvm.riscv.vwmaccu.mask.nxv16i16.nxv16i8(
+  <vscale x 16 x i16>,
+  <vscale x 16 x i8>,
+  <vscale x 16 x i8>,
+  <vscale x 16 x i1>,
+  i32);
+
+define <vscale x 16 x i16>  @intrinsic_vwmaccu_mask_vv_nxv16i16_nxv16i8_nxv16i8(<vscale x 16 x i16> %0, <vscale x 16 x i8> %1, <vscale x 16 x i8> %2, <vscale x 16 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccu_mask_vv_nxv16i16_nxv16i8_nxv16i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e8,m2,ta,mu
+; CHECK-NEXT:    vwmaccu.vv v16, v20, v22, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 16 x i16> @llvm.riscv.vwmaccu.mask.nxv16i16.nxv16i8(
+    <vscale x 16 x i16> %0,
+    <vscale x 16 x i8> %1,
+    <vscale x 16 x i8> %2,
+    <vscale x 16 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 16 x i16> %a
+}
+
+declare <vscale x 32 x i16> @llvm.riscv.vwmaccu.nxv32i16.nxv32i8(
+  <vscale x 32 x i16>,
+  <vscale x 32 x i8>,
+  <vscale x 32 x i8>,
+  i32);
+
+define <vscale x 32 x i16>  @intrinsic_vwmaccu_vv_nxv32i16_nxv32i8_nxv32i8(<vscale x 32 x i16> %0, <vscale x 32 x i8> %1, <vscale x 32 x i8> %2, i32 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccu_vv_nxv32i16_nxv32i8_nxv32i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a3, zero, e8,m4,ta,mu
+; CHECK-NEXT:    vle8.v v28, (a1)
+; CHECK-NEXT:    vle8.v v8, (a0)
+; CHECK-NEXT:    vsetvli a0, a2, e8,m4,ta,mu
+; CHECK-NEXT:    vwmaccu.vv v16, v8, v28
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 32 x i16> @llvm.riscv.vwmaccu.nxv32i16.nxv32i8(
+    <vscale x 32 x i16> %0,
+    <vscale x 32 x i8> %1,
+    <vscale x 32 x i8> %2,
+    i32 %3)
+
+  ret <vscale x 32 x i16> %a
+}
+
+declare <vscale x 32 x i16> @llvm.riscv.vwmaccu.mask.nxv32i16.nxv32i8(
+  <vscale x 32 x i16>,
+  <vscale x 32 x i8>,
+  <vscale x 32 x i8>,
+  <vscale x 32 x i1>,
+  i32);
+
+define <vscale x 32 x i16>  @intrinsic_vwmaccu_mask_vv_nxv32i16_nxv32i8_nxv32i8(<vscale x 32 x i16> %0, <vscale x 32 x i8> %1, <vscale x 32 x i8> %2, <vscale x 32 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccu_mask_vv_nxv32i16_nxv32i8_nxv32i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a3, zero, e8,m4,ta,mu
+; CHECK-NEXT:    vle8.v v28, (a1)
+; CHECK-NEXT:    vle8.v v8, (a0)
+; CHECK-NEXT:    vsetvli a0, a2, e8,m4,ta,mu
+; CHECK-NEXT:    vwmaccu.vv v16, v8, v28, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 32 x i16> @llvm.riscv.vwmaccu.mask.nxv32i16.nxv32i8(
+    <vscale x 32 x i16> %0,
+    <vscale x 32 x i8> %1,
+    <vscale x 32 x i8> %2,
+    <vscale x 32 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 32 x i16> %a
+}
+
+declare <vscale x 1 x i32> @llvm.riscv.vwmaccu.nxv1i32.nxv1i16(
+  <vscale x 1 x i32>,
+  <vscale x 1 x i16>,
+  <vscale x 1 x i16>,
+  i32);
+
+define <vscale x 1 x i32>  @intrinsic_vwmaccu_vv_nxv1i32_nxv1i16_nxv1i16(<vscale x 1 x i32> %0, <vscale x 1 x i16> %1, <vscale x 1 x i16> %2, i32 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccu_vv_nxv1i32_nxv1i16_nxv1i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e16,mf4,ta,mu
+; CHECK-NEXT:    vwmaccu.vv v16, v17, v18
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i32> @llvm.riscv.vwmaccu.nxv1i32.nxv1i16(
+    <vscale x 1 x i32> %0,
+    <vscale x 1 x i16> %1,
+    <vscale x 1 x i16> %2,
+    i32 %3)
+
+  ret <vscale x 1 x i32> %a
+}
+
+declare <vscale x 1 x i32> @llvm.riscv.vwmaccu.mask.nxv1i32.nxv1i16(
+  <vscale x 1 x i32>,
+  <vscale x 1 x i16>,
+  <vscale x 1 x i16>,
+  <vscale x 1 x i1>,
+  i32);
+
+define <vscale x 1 x i32>  @intrinsic_vwmaccu_mask_vv_nxv1i32_nxv1i16_nxv1i16(<vscale x 1 x i32> %0, <vscale x 1 x i16> %1, <vscale x 1 x i16> %2, <vscale x 1 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccu_mask_vv_nxv1i32_nxv1i16_nxv1i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e16,mf4,ta,mu
+; CHECK-NEXT:    vwmaccu.vv v16, v17, v18, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i32> @llvm.riscv.vwmaccu.mask.nxv1i32.nxv1i16(
+    <vscale x 1 x i32> %0,
+    <vscale x 1 x i16> %1,
+    <vscale x 1 x i16> %2,
+    <vscale x 1 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 1 x i32> %a
+}
+
+declare <vscale x 2 x i32> @llvm.riscv.vwmaccu.nxv2i32.nxv2i16(
+  <vscale x 2 x i32>,
+  <vscale x 2 x i16>,
+  <vscale x 2 x i16>,
+  i32);
+
+define <vscale x 2 x i32>  @intrinsic_vwmaccu_vv_nxv2i32_nxv2i16_nxv2i16(<vscale x 2 x i32> %0, <vscale x 2 x i16> %1, <vscale x 2 x i16> %2, i32 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccu_vv_nxv2i32_nxv2i16_nxv2i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e16,mf2,ta,mu
+; CHECK-NEXT:    vwmaccu.vv v16, v17, v18
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i32> @llvm.riscv.vwmaccu.nxv2i32.nxv2i16(
+    <vscale x 2 x i32> %0,
+    <vscale x 2 x i16> %1,
+    <vscale x 2 x i16> %2,
+    i32 %3)
+
+  ret <vscale x 2 x i32> %a
+}
+
+declare <vscale x 2 x i32> @llvm.riscv.vwmaccu.mask.nxv2i32.nxv2i16(
+  <vscale x 2 x i32>,
+  <vscale x 2 x i16>,
+  <vscale x 2 x i16>,
+  <vscale x 2 x i1>,
+  i32);
+
+define <vscale x 2 x i32>  @intrinsic_vwmaccu_mask_vv_nxv2i32_nxv2i16_nxv2i16(<vscale x 2 x i32> %0, <vscale x 2 x i16> %1, <vscale x 2 x i16> %2, <vscale x 2 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccu_mask_vv_nxv2i32_nxv2i16_nxv2i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e16,mf2,ta,mu
+; CHECK-NEXT:    vwmaccu.vv v16, v17, v18, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i32> @llvm.riscv.vwmaccu.mask.nxv2i32.nxv2i16(
+    <vscale x 2 x i32> %0,
+    <vscale x 2 x i16> %1,
+    <vscale x 2 x i16> %2,
+    <vscale x 2 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 2 x i32> %a
+}
+
+declare <vscale x 4 x i32> @llvm.riscv.vwmaccu.nxv4i32.nxv4i16(
+  <vscale x 4 x i32>,
+  <vscale x 4 x i16>,
+  <vscale x 4 x i16>,
+  i32);
+
+define <vscale x 4 x i32>  @intrinsic_vwmaccu_vv_nxv4i32_nxv4i16_nxv4i16(<vscale x 4 x i32> %0, <vscale x 4 x i16> %1, <vscale x 4 x i16> %2, i32 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccu_vv_nxv4i32_nxv4i16_nxv4i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e16,m1,ta,mu
+; CHECK-NEXT:    vwmaccu.vv v16, v18, v19
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i32> @llvm.riscv.vwmaccu.nxv4i32.nxv4i16(
+    <vscale x 4 x i32> %0,
+    <vscale x 4 x i16> %1,
+    <vscale x 4 x i16> %2,
+    i32 %3)
+
+  ret <vscale x 4 x i32> %a
+}
+
+declare <vscale x 4 x i32> @llvm.riscv.vwmaccu.mask.nxv4i32.nxv4i16(
+  <vscale x 4 x i32>,
+  <vscale x 4 x i16>,
+  <vscale x 4 x i16>,
+  <vscale x 4 x i1>,
+  i32);
+
+define <vscale x 4 x i32>  @intrinsic_vwmaccu_mask_vv_nxv4i32_nxv4i16_nxv4i16(<vscale x 4 x i32> %0, <vscale x 4 x i16> %1, <vscale x 4 x i16> %2, <vscale x 4 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccu_mask_vv_nxv4i32_nxv4i16_nxv4i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e16,m1,ta,mu
+; CHECK-NEXT:    vwmaccu.vv v16, v18, v19, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i32> @llvm.riscv.vwmaccu.mask.nxv4i32.nxv4i16(
+    <vscale x 4 x i32> %0,
+    <vscale x 4 x i16> %1,
+    <vscale x 4 x i16> %2,
+    <vscale x 4 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 4 x i32> %a
+}
+
+declare <vscale x 8 x i32> @llvm.riscv.vwmaccu.nxv8i32.nxv8i16(
+  <vscale x 8 x i32>,
+  <vscale x 8 x i16>,
+  <vscale x 8 x i16>,
+  i32);
+
+define <vscale x 8 x i32>  @intrinsic_vwmaccu_vv_nxv8i32_nxv8i16_nxv8i16(<vscale x 8 x i32> %0, <vscale x 8 x i16> %1, <vscale x 8 x i16> %2, i32 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccu_vv_nxv8i32_nxv8i16_nxv8i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e16,m2,ta,mu
+; CHECK-NEXT:    vwmaccu.vv v16, v20, v22
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i32> @llvm.riscv.vwmaccu.nxv8i32.nxv8i16(
+    <vscale x 8 x i32> %0,
+    <vscale x 8 x i16> %1,
+    <vscale x 8 x i16> %2,
+    i32 %3)
+
+  ret <vscale x 8 x i32> %a
+}
+
+declare <vscale x 8 x i32> @llvm.riscv.vwmaccu.mask.nxv8i32.nxv8i16(
+  <vscale x 8 x i32>,
+  <vscale x 8 x i16>,
+  <vscale x 8 x i16>,
+  <vscale x 8 x i1>,
+  i32);
+
+define <vscale x 8 x i32>  @intrinsic_vwmaccu_mask_vv_nxv8i32_nxv8i16_nxv8i16(<vscale x 8 x i32> %0, <vscale x 8 x i16> %1, <vscale x 8 x i16> %2, <vscale x 8 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccu_mask_vv_nxv8i32_nxv8i16_nxv8i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e16,m2,ta,mu
+; CHECK-NEXT:    vwmaccu.vv v16, v20, v22, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i32> @llvm.riscv.vwmaccu.mask.nxv8i32.nxv8i16(
+    <vscale x 8 x i32> %0,
+    <vscale x 8 x i16> %1,
+    <vscale x 8 x i16> %2,
+    <vscale x 8 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 8 x i32> %a
+}
+
+declare <vscale x 16 x i32> @llvm.riscv.vwmaccu.nxv16i32.nxv16i16(
+  <vscale x 16 x i32>,
+  <vscale x 16 x i16>,
+  <vscale x 16 x i16>,
+  i32);
+
+define <vscale x 16 x i32>  @intrinsic_vwmaccu_vv_nxv16i32_nxv16i16_nxv16i16(<vscale x 16 x i32> %0, <vscale x 16 x i16> %1, <vscale x 16 x i16> %2, i32 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccu_vv_nxv16i32_nxv16i16_nxv16i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a3, zero, e16,m4,ta,mu
+; CHECK-NEXT:    vle16.v v28, (a1)
+; CHECK-NEXT:    vle16.v v8, (a0)
+; CHECK-NEXT:    vsetvli a0, a2, e16,m4,ta,mu
+; CHECK-NEXT:    vwmaccu.vv v16, v8, v28
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 16 x i32> @llvm.riscv.vwmaccu.nxv16i32.nxv16i16(
+    <vscale x 16 x i32> %0,
+    <vscale x 16 x i16> %1,
+    <vscale x 16 x i16> %2,
+    i32 %3)
+
+  ret <vscale x 16 x i32> %a
+}
+
+declare <vscale x 16 x i32> @llvm.riscv.vwmaccu.mask.nxv16i32.nxv16i16(
+  <vscale x 16 x i32>,
+  <vscale x 16 x i16>,
+  <vscale x 16 x i16>,
+  <vscale x 16 x i1>,
+  i32);
+
+define <vscale x 16 x i32>  @intrinsic_vwmaccu_mask_vv_nxv16i32_nxv16i16_nxv16i16(<vscale x 16 x i32> %0, <vscale x 16 x i16> %1, <vscale x 16 x i16> %2, <vscale x 16 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccu_mask_vv_nxv16i32_nxv16i16_nxv16i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a3, zero, e16,m4,ta,mu
+; CHECK-NEXT:    vle16.v v28, (a1)
+; CHECK-NEXT:    vle16.v v8, (a0)
+; CHECK-NEXT:    vsetvli a0, a2, e16,m4,ta,mu
+; CHECK-NEXT:    vwmaccu.vv v16, v8, v28, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 16 x i32> @llvm.riscv.vwmaccu.mask.nxv16i32.nxv16i16(
+    <vscale x 16 x i32> %0,
+    <vscale x 16 x i16> %1,
+    <vscale x 16 x i16> %2,
+    <vscale x 16 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 16 x i32> %a
+}
+
+declare <vscale x 1 x i16> @llvm.riscv.vwmaccu.nxv1i16.i8(
+  <vscale x 1 x i16>,
+  i8,
+  <vscale x 1 x i8>,
+  i32);
+
+define <vscale x 1 x i16>  @intrinsic_vwmaccu_vx_nxv1i16_i8_nxv1i8(<vscale x 1 x i16> %0, i8 %1, <vscale x 1 x i8> %2, i32 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccu_vx_nxv1i16_i8_nxv1i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e8,mf8,ta,mu
+; CHECK-NEXT:    vwmaccu.vx v16, a0, v17
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i16> @llvm.riscv.vwmaccu.nxv1i16.i8(
+    <vscale x 1 x i16> %0,
+    i8 %1,
+    <vscale x 1 x i8> %2,
+    i32 %3)
+
+  ret <vscale x 1 x i16> %a
+}
+
+declare <vscale x 1 x i16> @llvm.riscv.vwmaccu.mask.nxv1i16.i8(
+  <vscale x 1 x i16>,
+  i8,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i1>,
+  i32);
+
+define <vscale x 1 x i16> @intrinsic_vwmaccu_mask_vx_nxv1i16_i8_nxv1i8(<vscale x 1 x i16> %0, i8 %1, <vscale x 1 x i8> %2, <vscale x 1 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccu_mask_vx_nxv1i16_i8_nxv1i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e8,mf8,ta,mu
+; CHECK-NEXT:    vwmaccu.vx v16, a0, v17, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i16> @llvm.riscv.vwmaccu.mask.nxv1i16.i8(
+    <vscale x 1 x i16> %0,
+    i8 %1,
+    <vscale x 1 x i8> %2,
+    <vscale x 1 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 1 x i16> %a
+}
+
+declare <vscale x 2 x i16> @llvm.riscv.vwmaccu.nxv2i16.i8(
+  <vscale x 2 x i16>,
+  i8,
+  <vscale x 2 x i8>,
+  i32);
+
+define <vscale x 2 x i16>  @intrinsic_vwmaccu_vx_nxv2i16_i8_nxv2i8(<vscale x 2 x i16> %0, i8 %1, <vscale x 2 x i8> %2, i32 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccu_vx_nxv2i16_i8_nxv2i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e8,mf4,ta,mu
+; CHECK-NEXT:    vwmaccu.vx v16, a0, v17
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i16> @llvm.riscv.vwmaccu.nxv2i16.i8(
+    <vscale x 2 x i16> %0,
+    i8 %1,
+    <vscale x 2 x i8> %2,
+    i32 %3)
+
+  ret <vscale x 2 x i16> %a
+}
+
+declare <vscale x 2 x i16> @llvm.riscv.vwmaccu.mask.nxv2i16.i8(
+  <vscale x 2 x i16>,
+  i8,
+  <vscale x 2 x i8>,
+  <vscale x 2 x i1>,
+  i32);
+
+define <vscale x 2 x i16> @intrinsic_vwmaccu_mask_vx_nxv2i16_i8_nxv2i8(<vscale x 2 x i16> %0, i8 %1, <vscale x 2 x i8> %2, <vscale x 2 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccu_mask_vx_nxv2i16_i8_nxv2i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e8,mf4,ta,mu
+; CHECK-NEXT:    vwmaccu.vx v16, a0, v17, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i16> @llvm.riscv.vwmaccu.mask.nxv2i16.i8(
+    <vscale x 2 x i16> %0,
+    i8 %1,
+    <vscale x 2 x i8> %2,
+    <vscale x 2 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 2 x i16> %a
+}
+
+declare <vscale x 4 x i16> @llvm.riscv.vwmaccu.nxv4i16.i8(
+  <vscale x 4 x i16>,
+  i8,
+  <vscale x 4 x i8>,
+  i32);
+
+define <vscale x 4 x i16>  @intrinsic_vwmaccu_vx_nxv4i16_i8_nxv4i8(<vscale x 4 x i16> %0, i8 %1, <vscale x 4 x i8> %2, i32 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccu_vx_nxv4i16_i8_nxv4i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e8,mf2,ta,mu
+; CHECK-NEXT:    vwmaccu.vx v16, a0, v17
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i16> @llvm.riscv.vwmaccu.nxv4i16.i8(
+    <vscale x 4 x i16> %0,
+    i8 %1,
+    <vscale x 4 x i8> %2,
+    i32 %3)
+
+  ret <vscale x 4 x i16> %a
+}
+
+declare <vscale x 4 x i16> @llvm.riscv.vwmaccu.mask.nxv4i16.i8(
+  <vscale x 4 x i16>,
+  i8,
+  <vscale x 4 x i8>,
+  <vscale x 4 x i1>,
+  i32);
+
+define <vscale x 4 x i16> @intrinsic_vwmaccu_mask_vx_nxv4i16_i8_nxv4i8(<vscale x 4 x i16> %0, i8 %1, <vscale x 4 x i8> %2, <vscale x 4 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccu_mask_vx_nxv4i16_i8_nxv4i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e8,mf2,ta,mu
+; CHECK-NEXT:    vwmaccu.vx v16, a0, v17, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i16> @llvm.riscv.vwmaccu.mask.nxv4i16.i8(
+    <vscale x 4 x i16> %0,
+    i8 %1,
+    <vscale x 4 x i8> %2,
+    <vscale x 4 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 4 x i16> %a
+}
+
+declare <vscale x 8 x i16> @llvm.riscv.vwmaccu.nxv8i16.i8(
+  <vscale x 8 x i16>,
+  i8,
+  <vscale x 8 x i8>,
+  i32);
+
+define <vscale x 8 x i16>  @intrinsic_vwmaccu_vx_nxv8i16_i8_nxv8i8(<vscale x 8 x i16> %0, i8 %1, <vscale x 8 x i8> %2, i32 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccu_vx_nxv8i16_i8_nxv8i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e8,m1,ta,mu
+; CHECK-NEXT:    vwmaccu.vx v16, a0, v18
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i16> @llvm.riscv.vwmaccu.nxv8i16.i8(
+    <vscale x 8 x i16> %0,
+    i8 %1,
+    <vscale x 8 x i8> %2,
+    i32 %3)
+
+  ret <vscale x 8 x i16> %a
+}
+
+declare <vscale x 8 x i16> @llvm.riscv.vwmaccu.mask.nxv8i16.i8(
+  <vscale x 8 x i16>,
+  i8,
+  <vscale x 8 x i8>,
+  <vscale x 8 x i1>,
+  i32);
+
+define <vscale x 8 x i16> @intrinsic_vwmaccu_mask_vx_nxv8i16_i8_nxv8i8(<vscale x 8 x i16> %0, i8 %1, <vscale x 8 x i8> %2, <vscale x 8 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccu_mask_vx_nxv8i16_i8_nxv8i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e8,m1,ta,mu
+; CHECK-NEXT:    vwmaccu.vx v16, a0, v18, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i16> @llvm.riscv.vwmaccu.mask.nxv8i16.i8(
+    <vscale x 8 x i16> %0,
+    i8 %1,
+    <vscale x 8 x i8> %2,
+    <vscale x 8 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 8 x i16> %a
+}
+
+declare <vscale x 16 x i16> @llvm.riscv.vwmaccu.nxv16i16.i8(
+  <vscale x 16 x i16>,
+  i8,
+  <vscale x 16 x i8>,
+  i32);
+
+define <vscale x 16 x i16>  @intrinsic_vwmaccu_vx_nxv16i16_i8_nxv16i8(<vscale x 16 x i16> %0, i8 %1, <vscale x 16 x i8> %2, i32 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccu_vx_nxv16i16_i8_nxv16i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e8,m2,ta,mu
+; CHECK-NEXT:    vwmaccu.vx v16, a0, v20
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 16 x i16> @llvm.riscv.vwmaccu.nxv16i16.i8(
+    <vscale x 16 x i16> %0,
+    i8 %1,
+    <vscale x 16 x i8> %2,
+    i32 %3)
+
+  ret <vscale x 16 x i16> %a
+}
+
+declare <vscale x 16 x i16> @llvm.riscv.vwmaccu.mask.nxv16i16.i8(
+  <vscale x 16 x i16>,
+  i8,
+  <vscale x 16 x i8>,
+  <vscale x 16 x i1>,
+  i32);
+
+define <vscale x 16 x i16> @intrinsic_vwmaccu_mask_vx_nxv16i16_i8_nxv16i8(<vscale x 16 x i16> %0, i8 %1, <vscale x 16 x i8> %2, <vscale x 16 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccu_mask_vx_nxv16i16_i8_nxv16i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e8,m2,ta,mu
+; CHECK-NEXT:    vwmaccu.vx v16, a0, v20, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 16 x i16> @llvm.riscv.vwmaccu.mask.nxv16i16.i8(
+    <vscale x 16 x i16> %0,
+    i8 %1,
+    <vscale x 16 x i8> %2,
+    <vscale x 16 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 16 x i16> %a
+}
+
+declare <vscale x 32 x i16> @llvm.riscv.vwmaccu.nxv32i16.i8(
+  <vscale x 32 x i16>,
+  i8,
+  <vscale x 32 x i8>,
+  i32);
+
+define <vscale x 32 x i16>  @intrinsic_vwmaccu_vx_nxv32i16_i8_nxv32i8(<vscale x 32 x i16> %0, i8 %1, <vscale x 32 x i8> %2, i32 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccu_vx_nxv32i16_i8_nxv32i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a3, zero, e8,m4,ta,mu
+; CHECK-NEXT:    vle8.v v28, (a1)
+; CHECK-NEXT:    vsetvli a1, a2, e8,m4,ta,mu
+; CHECK-NEXT:    vwmaccu.vx v16, a0, v28
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 32 x i16> @llvm.riscv.vwmaccu.nxv32i16.i8(
+    <vscale x 32 x i16> %0,
+    i8 %1,
+    <vscale x 32 x i8> %2,
+    i32 %3)
+
+  ret <vscale x 32 x i16> %a
+}
+
+declare <vscale x 32 x i16> @llvm.riscv.vwmaccu.mask.nxv32i16.i8(
+  <vscale x 32 x i16>,
+  i8,
+  <vscale x 32 x i8>,
+  <vscale x 32 x i1>,
+  i32);
+
+define <vscale x 32 x i16> @intrinsic_vwmaccu_mask_vx_nxv32i16_i8_nxv32i8(<vscale x 32 x i16> %0, i8 %1, <vscale x 32 x i8> %2, <vscale x 32 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccu_mask_vx_nxv32i16_i8_nxv32i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a3, zero, e8,m4,ta,mu
+; CHECK-NEXT:    vle8.v v28, (a1)
+; CHECK-NEXT:    vsetvli a1, a2, e8,m4,ta,mu
+; CHECK-NEXT:    vwmaccu.vx v16, a0, v28, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 32 x i16> @llvm.riscv.vwmaccu.mask.nxv32i16.i8(
+    <vscale x 32 x i16> %0,
+    i8 %1,
+    <vscale x 32 x i8> %2,
+    <vscale x 32 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 32 x i16> %a
+}
+
+declare <vscale x 1 x i32> @llvm.riscv.vwmaccu.nxv1i32.i16(
+  <vscale x 1 x i32>,
+  i16,
+  <vscale x 1 x i16>,
+  i32);
+
+define <vscale x 1 x i32>  @intrinsic_vwmaccu_vx_nxv1i32_i16_nxv1i16(<vscale x 1 x i32> %0, i16 %1, <vscale x 1 x i16> %2, i32 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccu_vx_nxv1i32_i16_nxv1i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e16,mf4,ta,mu
+; CHECK-NEXT:    vwmaccu.vx v16, a0, v17
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i32> @llvm.riscv.vwmaccu.nxv1i32.i16(
+    <vscale x 1 x i32> %0,
+    i16 %1,
+    <vscale x 1 x i16> %2,
+    i32 %3)
+
+  ret <vscale x 1 x i32> %a
+}
+
+declare <vscale x 1 x i32> @llvm.riscv.vwmaccu.mask.nxv1i32.i16(
+  <vscale x 1 x i32>,
+  i16,
+  <vscale x 1 x i16>,
+  <vscale x 1 x i1>,
+  i32);
+
+define <vscale x 1 x i32> @intrinsic_vwmaccu_mask_vx_nxv1i32_i16_nxv1i16(<vscale x 1 x i32> %0, i16 %1, <vscale x 1 x i16> %2, <vscale x 1 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccu_mask_vx_nxv1i32_i16_nxv1i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e16,mf4,ta,mu
+; CHECK-NEXT:    vwmaccu.vx v16, a0, v17, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i32> @llvm.riscv.vwmaccu.mask.nxv1i32.i16(
+    <vscale x 1 x i32> %0,
+    i16 %1,
+    <vscale x 1 x i16> %2,
+    <vscale x 1 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 1 x i32> %a
+}
+
+declare <vscale x 2 x i32> @llvm.riscv.vwmaccu.nxv2i32.i16(
+  <vscale x 2 x i32>,
+  i16,
+  <vscale x 2 x i16>,
+  i32);
+
+define <vscale x 2 x i32>  @intrinsic_vwmaccu_vx_nxv2i32_i16_nxv2i16(<vscale x 2 x i32> %0, i16 %1, <vscale x 2 x i16> %2, i32 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccu_vx_nxv2i32_i16_nxv2i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e16,mf2,ta,mu
+; CHECK-NEXT:    vwmaccu.vx v16, a0, v17
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i32> @llvm.riscv.vwmaccu.nxv2i32.i16(
+    <vscale x 2 x i32> %0,
+    i16 %1,
+    <vscale x 2 x i16> %2,
+    i32 %3)
+
+  ret <vscale x 2 x i32> %a
+}
+
+declare <vscale x 2 x i32> @llvm.riscv.vwmaccu.mask.nxv2i32.i16(
+  <vscale x 2 x i32>,
+  i16,
+  <vscale x 2 x i16>,
+  <vscale x 2 x i1>,
+  i32);
+
+define <vscale x 2 x i32> @intrinsic_vwmaccu_mask_vx_nxv2i32_i16_nxv2i16(<vscale x 2 x i32> %0, i16 %1, <vscale x 2 x i16> %2, <vscale x 2 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccu_mask_vx_nxv2i32_i16_nxv2i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e16,mf2,ta,mu
+; CHECK-NEXT:    vwmaccu.vx v16, a0, v17, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i32> @llvm.riscv.vwmaccu.mask.nxv2i32.i16(
+    <vscale x 2 x i32> %0,
+    i16 %1,
+    <vscale x 2 x i16> %2,
+    <vscale x 2 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 2 x i32> %a
+}
+
+declare <vscale x 4 x i32> @llvm.riscv.vwmaccu.nxv4i32.i16(
+  <vscale x 4 x i32>,
+  i16,
+  <vscale x 4 x i16>,
+  i32);
+
+define <vscale x 4 x i32>  @intrinsic_vwmaccu_vx_nxv4i32_i16_nxv4i16(<vscale x 4 x i32> %0, i16 %1, <vscale x 4 x i16> %2, i32 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccu_vx_nxv4i32_i16_nxv4i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e16,m1,ta,mu
+; CHECK-NEXT:    vwmaccu.vx v16, a0, v18
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i32> @llvm.riscv.vwmaccu.nxv4i32.i16(
+    <vscale x 4 x i32> %0,
+    i16 %1,
+    <vscale x 4 x i16> %2,
+    i32 %3)
+
+  ret <vscale x 4 x i32> %a
+}
+
+declare <vscale x 4 x i32> @llvm.riscv.vwmaccu.mask.nxv4i32.i16(
+  <vscale x 4 x i32>,
+  i16,
+  <vscale x 4 x i16>,
+  <vscale x 4 x i1>,
+  i32);
+
+define <vscale x 4 x i32> @intrinsic_vwmaccu_mask_vx_nxv4i32_i16_nxv4i16(<vscale x 4 x i32> %0, i16 %1, <vscale x 4 x i16> %2, <vscale x 4 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccu_mask_vx_nxv4i32_i16_nxv4i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e16,m1,ta,mu
+; CHECK-NEXT:    vwmaccu.vx v16, a0, v18, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i32> @llvm.riscv.vwmaccu.mask.nxv4i32.i16(
+    <vscale x 4 x i32> %0,
+    i16 %1,
+    <vscale x 4 x i16> %2,
+    <vscale x 4 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 4 x i32> %a
+}
+
+declare <vscale x 8 x i32> @llvm.riscv.vwmaccu.nxv8i32.i16(
+  <vscale x 8 x i32>,
+  i16,
+  <vscale x 8 x i16>,
+  i32);
+
+define <vscale x 8 x i32>  @intrinsic_vwmaccu_vx_nxv8i32_i16_nxv8i16(<vscale x 8 x i32> %0, i16 %1, <vscale x 8 x i16> %2, i32 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccu_vx_nxv8i32_i16_nxv8i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e16,m2,ta,mu
+; CHECK-NEXT:    vwmaccu.vx v16, a0, v20
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i32> @llvm.riscv.vwmaccu.nxv8i32.i16(
+    <vscale x 8 x i32> %0,
+    i16 %1,
+    <vscale x 8 x i16> %2,
+    i32 %3)
+
+  ret <vscale x 8 x i32> %a
+}
+
+declare <vscale x 8 x i32> @llvm.riscv.vwmaccu.mask.nxv8i32.i16(
+  <vscale x 8 x i32>,
+  i16,
+  <vscale x 8 x i16>,
+  <vscale x 8 x i1>,
+  i32);
+
+define <vscale x 8 x i32> @intrinsic_vwmaccu_mask_vx_nxv8i32_i16_nxv8i16(<vscale x 8 x i32> %0, i16 %1, <vscale x 8 x i16> %2, <vscale x 8 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccu_mask_vx_nxv8i32_i16_nxv8i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e16,m2,ta,mu
+; CHECK-NEXT:    vwmaccu.vx v16, a0, v20, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i32> @llvm.riscv.vwmaccu.mask.nxv8i32.i16(
+    <vscale x 8 x i32> %0,
+    i16 %1,
+    <vscale x 8 x i16> %2,
+    <vscale x 8 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 8 x i32> %a
+}
+
+declare <vscale x 16 x i32> @llvm.riscv.vwmaccu.nxv16i32.i16(
+  <vscale x 16 x i32>,
+  i16,
+  <vscale x 16 x i16>,
+  i32);
+
+define <vscale x 16 x i32>  @intrinsic_vwmaccu_vx_nxv16i32_i16_nxv16i16(<vscale x 16 x i32> %0, i16 %1, <vscale x 16 x i16> %2, i32 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccu_vx_nxv16i32_i16_nxv16i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a3, zero, e16,m4,ta,mu
+; CHECK-NEXT:    vle16.v v28, (a1)
+; CHECK-NEXT:    vsetvli a1, a2, e16,m4,ta,mu
+; CHECK-NEXT:    vwmaccu.vx v16, a0, v28
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 16 x i32> @llvm.riscv.vwmaccu.nxv16i32.i16(
+    <vscale x 16 x i32> %0,
+    i16 %1,
+    <vscale x 16 x i16> %2,
+    i32 %3)
+
+  ret <vscale x 16 x i32> %a
+}
+
+declare <vscale x 16 x i32> @llvm.riscv.vwmaccu.mask.nxv16i32.i16(
+  <vscale x 16 x i32>,
+  i16,
+  <vscale x 16 x i16>,
+  <vscale x 16 x i1>,
+  i32);
+
+define <vscale x 16 x i32> @intrinsic_vwmaccu_mask_vx_nxv16i32_i16_nxv16i16(<vscale x 16 x i32> %0, i16 %1, <vscale x 16 x i16> %2, <vscale x 16 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccu_mask_vx_nxv16i32_i16_nxv16i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a3, zero, e16,m4,ta,mu
+; CHECK-NEXT:    vle16.v v28, (a1)
+; CHECK-NEXT:    vsetvli a1, a2, e16,m4,ta,mu
+; CHECK-NEXT:    vwmaccu.vx v16, a0, v28, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 16 x i32> @llvm.riscv.vwmaccu.mask.nxv16i32.i16(
+    <vscale x 16 x i32> %0,
+    i16 %1,
+    <vscale x 16 x i16> %2,
+    <vscale x 16 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 16 x i32> %a
+}

llvm/test/CodeGen/RISCV/rvv/vwmaccu-rv64.ll

+; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
+; RUN: llc -mtriple=riscv64 -mattr=+experimental-v,+d -verify-machineinstrs \
+; RUN:   --riscv-no-aliases < %s | FileCheck %s
+declare <vscale x 1 x i16> @llvm.riscv.vwmaccu.nxv1i16.nxv1i8(
+  <vscale x 1 x i16>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  i64);
+
+define <vscale x 1 x i16>  @intrinsic_vwmaccu_vv_nxv1i16_nxv1i8_nxv1i8(<vscale x 1 x i16> %0, <vscale x 1 x i8> %1, <vscale x 1 x i8> %2, i64 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccu_vv_nxv1i16_nxv1i8_nxv1i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e8,mf8,ta,mu
+; CHECK-NEXT:    vwmaccu.vv v16, v17, v18
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i16> @llvm.riscv.vwmaccu.nxv1i16.nxv1i8(
+    <vscale x 1 x i16> %0,
+    <vscale x 1 x i8> %1,
+    <vscale x 1 x i8> %2,
+    i64 %3)
+
+  ret <vscale x 1 x i16> %a
+}
+
+declare <vscale x 1 x i16> @llvm.riscv.vwmaccu.mask.nxv1i16.nxv1i8(
+  <vscale x 1 x i16>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i1>,
+  i64);
+
+define <vscale x 1 x i16>  @intrinsic_vwmaccu_mask_vv_nxv1i16_nxv1i8_nxv1i8(<vscale x 1 x i16> %0, <vscale x 1 x i8> %1, <vscale x 1 x i8> %2, <vscale x 1 x i1> %3, i64 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccu_mask_vv_nxv1i16_nxv1i8_nxv1i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e8,mf8,ta,mu
+; CHECK-NEXT:    vwmaccu.vv v16, v17, v18, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i16> @llvm.riscv.vwmaccu.mask.nxv1i16.nxv1i8(
+    <vscale x 1 x i16> %0,
+    <vscale x 1 x i8> %1,
+    <vscale x 1 x i8> %2,
+    <vscale x 1 x i1> %3,
+    i64 %4)
+
+  ret <vscale x 1 x i16> %a
+}
+
+declare <vscale x 2 x i16> @llvm.riscv.vwmaccu.nxv2i16.nxv2i8(
+  <vscale x 2 x i16>,
+  <vscale x 2 x i8>,
+  <vscale x 2 x i8>,
+  i64);
+
+define <vscale x 2 x i16>  @intrinsic_vwmaccu_vv_nxv2i16_nxv2i8_nxv2i8(<vscale x 2 x i16> %0, <vscale x 2 x i8> %1, <vscale x 2 x i8> %2, i64 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccu_vv_nxv2i16_nxv2i8_nxv2i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e8,mf4,ta,mu
+; CHECK-NEXT:    vwmaccu.vv v16, v17, v18
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i16> @llvm.riscv.vwmaccu.nxv2i16.nxv2i8(
+    <vscale x 2 x i16> %0,
+    <vscale x 2 x i8> %1,
+    <vscale x 2 x i8> %2,
+    i64 %3)
+
+  ret <vscale x 2 x i16> %a
+}
+
+declare <vscale x 2 x i16> @llvm.riscv.vwmaccu.mask.nxv2i16.nxv2i8(
+  <vscale x 2 x i16>,
+  <vscale x 2 x i8>,
+  <vscale x 2 x i8>,
+  <vscale x 2 x i1>,
+  i64);
+
+define <vscale x 2 x i16>  @intrinsic_vwmaccu_mask_vv_nxv2i16_nxv2i8_nxv2i8(<vscale x 2 x i16> %0, <vscale x 2 x i8> %1, <vscale x 2 x i8> %2, <vscale x 2 x i1> %3, i64 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccu_mask_vv_nxv2i16_nxv2i8_nxv2i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e8,mf4,ta,mu
+; CHECK-NEXT:    vwmaccu.vv v16, v17, v18, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i16> @llvm.riscv.vwmaccu.mask.nxv2i16.nxv2i8(
+    <vscale x 2 x i16> %0,
+    <vscale x 2 x i8> %1,
+    <vscale x 2 x i8> %2,
+    <vscale x 2 x i1> %3,
+    i64 %4)
+
+  ret <vscale x 2 x i16> %a
+}
+
+declare <vscale x 4 x i16> @llvm.riscv.vwmaccu.nxv4i16.nxv4i8(
+  <vscale x 4 x i16>,
+  <vscale x 4 x i8>,
+  <vscale x 4 x i8>,
+  i64);
+
+define <vscale x 4 x i16>  @intrinsic_vwmaccu_vv_nxv4i16_nxv4i8_nxv4i8(<vscale x 4 x i16> %0, <vscale x 4 x i8> %1, <vscale x 4 x i8> %2, i64 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccu_vv_nxv4i16_nxv4i8_nxv4i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e8,mf2,ta,mu
+; CHECK-NEXT:    vwmaccu.vv v16, v17, v18
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i16> @llvm.riscv.vwmaccu.nxv4i16.nxv4i8(
+    <vscale x 4 x i16> %0,
+    <vscale x 4 x i8> %1,
+    <vscale x 4 x i8> %2,
+    i64 %3)
+
+  ret <vscale x 4 x i16> %a
+}
+
+declare <vscale x 4 x i16> @llvm.riscv.vwmaccu.mask.nxv4i16.nxv4i8(
+  <vscale x 4 x i16>,
+  <vscale x 4 x i8>,
+  <vscale x 4 x i8>,
+  <vscale x 4 x i1>,
+  i64);
+
+define <vscale x 4 x i16>  @intrinsic_vwmaccu_mask_vv_nxv4i16_nxv4i8_nxv4i8(<vscale x 4 x i16> %0, <vscale x 4 x i8> %1, <vscale x 4 x i8> %2, <vscale x 4 x i1> %3, i64 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccu_mask_vv_nxv4i16_nxv4i8_nxv4i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e8,mf2,ta,mu
+; CHECK-NEXT:    vwmaccu.vv v16, v17, v18, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i16> @llvm.riscv.vwmaccu.mask.nxv4i16.nxv4i8(
+    <vscale x 4 x i16> %0,
+    <vscale x 4 x i8> %1,
+    <vscale x 4 x i8> %2,
+    <vscale x 4 x i1> %3,
+    i64 %4)
+
+  ret <vscale x 4 x i16> %a
+}
+
+declare <vscale x 8 x i16> @llvm.riscv.vwmaccu.nxv8i16.nxv8i8(
+  <vscale x 8 x i16>,
+  <vscale x 8 x i8>,
+  <vscale x 8 x i8>,
+  i64);
+
+define <vscale x 8 x i16>  @intrinsic_vwmaccu_vv_nxv8i16_nxv8i8_nxv8i8(<vscale x 8 x i16> %0, <vscale x 8 x i8> %1, <vscale x 8 x i8> %2, i64 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccu_vv_nxv8i16_nxv8i8_nxv8i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e8,m1,ta,mu
+; CHECK-NEXT:    vwmaccu.vv v16, v18, v19
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i16> @llvm.riscv.vwmaccu.nxv8i16.nxv8i8(
+    <vscale x 8 x i16> %0,
+    <vscale x 8 x i8> %1,
+    <vscale x 8 x i8> %2,
+    i64 %3)
+
+  ret <vscale x 8 x i16> %a
+}
+
+declare <vscale x 8 x i16> @llvm.riscv.vwmaccu.mask.nxv8i16.nxv8i8(
+  <vscale x 8 x i16>,
+  <vscale x 8 x i8>,
+  <vscale x 8 x i8>,
+  <vscale x 8 x i1>,
+  i64);
+
+define <vscale x 8 x i16>  @intrinsic_vwmaccu_mask_vv_nxv8i16_nxv8i8_nxv8i8(<vscale x 8 x i16> %0, <vscale x 8 x i8> %1, <vscale x 8 x i8> %2, <vscale x 8 x i1> %3, i64 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccu_mask_vv_nxv8i16_nxv8i8_nxv8i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e8,m1,ta,mu
+; CHECK-NEXT:    vwmaccu.vv v16, v18, v19, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i16> @llvm.riscv.vwmaccu.mask.nxv8i16.nxv8i8(
+    <vscale x 8 x i16> %0,
+    <vscale x 8 x i8> %1,
+    <vscale x 8 x i8> %2,
+    <vscale x 8 x i1> %3,
+    i64 %4)
+
+  ret <vscale x 8 x i16> %a
+}
+
+declare <vscale x 16 x i16> @llvm.riscv.vwmaccu.nxv16i16.nxv16i8(
+  <vscale x 16 x i16>,
+  <vscale x 16 x i8>,
+  <vscale x 16 x i8>,
+  i64);
+
+define <vscale x 16 x i16>  @intrinsic_vwmaccu_vv_nxv16i16_nxv16i8_nxv16i8(<vscale x 16 x i16> %0, <vscale x 16 x i8> %1, <vscale x 16 x i8> %2, i64 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccu_vv_nxv16i16_nxv16i8_nxv16i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e8,m2,ta,mu
+; CHECK-NEXT:    vwmaccu.vv v16, v20, v22
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 16 x i16> @llvm.riscv.vwmaccu.nxv16i16.nxv16i8(
+    <vscale x 16 x i16> %0,
+    <vscale x 16 x i8> %1,
+    <vscale x 16 x i8> %2,
+    i64 %3)
+
+  ret <vscale x 16 x i16> %a
+}
+
+declare <vscale x 16 x i16> @llvm.riscv.vwmaccu.mask.nxv16i16.nxv16i8(
+  <vscale x 16 x i16>,
+  <vscale x 16 x i8>,
+  <vscale x 16 x i8>,
+  <vscale x 16 x i1>,
+  i64);
+
+define <vscale x 16 x i16>  @intrinsic_vwmaccu_mask_vv_nxv16i16_nxv16i8_nxv16i8(<vscale x 16 x i16> %0, <vscale x 16 x i8> %1, <vscale x 16 x i8> %2, <vscale x 16 x i1> %3, i64 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccu_mask_vv_nxv16i16_nxv16i8_nxv16i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e8,m2,ta,mu
+; CHECK-NEXT:    vwmaccu.vv v16, v20, v22, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 16 x i16> @llvm.riscv.vwmaccu.mask.nxv16i16.nxv16i8(
+    <vscale x 16 x i16> %0,
+    <vscale x 16 x i8> %1,
+    <vscale x 16 x i8> %2,
+    <vscale x 16 x i1> %3,
+    i64 %4)
+
+  ret <vscale x 16 x i16> %a
+}
+
+declare <vscale x 32 x i16> @llvm.riscv.vwmaccu.nxv32i16.nxv32i8(
+  <vscale x 32 x i16>,
+  <vscale x 32 x i8>,
+  <vscale x 32 x i8>,
+  i64);
+
+define <vscale x 32 x i16>  @intrinsic_vwmaccu_vv_nxv32i16_nxv32i8_nxv32i8(<vscale x 32 x i16> %0, <vscale x 32 x i8> %1, <vscale x 32 x i8> %2, i64 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccu_vv_nxv32i16_nxv32i8_nxv32i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a3, zero, e8,m4,ta,mu
+; CHECK-NEXT:    vle8.v v28, (a1)
+; CHECK-NEXT:    vle8.v v8, (a0)
+; CHECK-NEXT:    vsetvli a0, a2, e8,m4,ta,mu
+; CHECK-NEXT:    vwmaccu.vv v16, v8, v28
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 32 x i16> @llvm.riscv.vwmaccu.nxv32i16.nxv32i8(
+    <vscale x 32 x i16> %0,
+    <vscale x 32 x i8> %1,
+    <vscale x 32 x i8> %2,
+    i64 %3)
+
+  ret <vscale x 32 x i16> %a
+}
+
+declare <vscale x 32 x i16> @llvm.riscv.vwmaccu.mask.nxv32i16.nxv32i8(
+  <vscale x 32 x i16>,
+  <vscale x 32 x i8>,
+  <vscale x 32 x i8>,
+  <vscale x 32 x i1>,
+  i64);
+
+define <vscale x 32 x i16>  @intrinsic_vwmaccu_mask_vv_nxv32i16_nxv32i8_nxv32i8(<vscale x 32 x i16> %0, <vscale x 32 x i8> %1, <vscale x 32 x i8> %2, <vscale x 32 x i1> %3, i64 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccu_mask_vv_nxv32i16_nxv32i8_nxv32i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a3, zero, e8,m4,ta,mu
+; CHECK-NEXT:    vle8.v v28, (a1)
+; CHECK-NEXT:    vle8.v v8, (a0)
+; CHECK-NEXT:    vsetvli a0, a2, e8,m4,ta,mu
+; CHECK-NEXT:    vwmaccu.vv v16, v8, v28, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 32 x i16> @llvm.riscv.vwmaccu.mask.nxv32i16.nxv32i8(
+    <vscale x 32 x i16> %0,
+    <vscale x 32 x i8> %1,
+    <vscale x 32 x i8> %2,
+    <vscale x 32 x i1> %3,
+    i64 %4)
+
+  ret <vscale x 32 x i16> %a
+}
+
+declare <vscale x 1 x i32> @llvm.riscv.vwmaccu.nxv1i32.nxv1i16(
+  <vscale x 1 x i32>,
+  <vscale x 1 x i16>,
+  <vscale x 1 x i16>,
+  i64);
+
+define <vscale x 1 x i32>  @intrinsic_vwmaccu_vv_nxv1i32_nxv1i16_nxv1i16(<vscale x 1 x i32> %0, <vscale x 1 x i16> %1, <vscale x 1 x i16> %2, i64 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccu_vv_nxv1i32_nxv1i16_nxv1i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e16,mf4,ta,mu
+; CHECK-NEXT:    vwmaccu.vv v16, v17, v18
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i32> @llvm.riscv.vwmaccu.nxv1i32.nxv1i16(
+    <vscale x 1 x i32> %0,
+    <vscale x 1 x i16> %1,
+    <vscale x 1 x i16> %2,
+    i64 %3)
+
+  ret <vscale x 1 x i32> %a
+}
+
+declare <vscale x 1 x i32> @llvm.riscv.vwmaccu.mask.nxv1i32.nxv1i16(
+  <vscale x 1 x i32>,
+  <vscale x 1 x i16>,
+  <vscale x 1 x i16>,
+  <vscale x 1 x i1>,
+  i64);
+
+define <vscale x 1 x i32>  @intrinsic_vwmaccu_mask_vv_nxv1i32_nxv1i16_nxv1i16(<vscale x 1 x i32> %0, <vscale x 1 x i16> %1, <vscale x 1 x i16> %2, <vscale x 1 x i1> %3, i64 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccu_mask_vv_nxv1i32_nxv1i16_nxv1i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e16,mf4,ta,mu
+; CHECK-NEXT:    vwmaccu.vv v16, v17, v18, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i32> @llvm.riscv.vwmaccu.mask.nxv1i32.nxv1i16(
+    <vscale x 1 x i32> %0,
+    <vscale x 1 x i16> %1,
+    <vscale x 1 x i16> %2,
+    <vscale x 1 x i1> %3,
+    i64 %4)
+
+  ret <vscale x 1 x i32> %a
+}
+
+declare <vscale x 2 x i32> @llvm.riscv.vwmaccu.nxv2i32.nxv2i16(
+  <vscale x 2 x i32>,
+  <vscale x 2 x i16>,
+  <vscale x 2 x i16>,
+  i64);
+
+define <vscale x 2 x i32>  @intrinsic_vwmaccu_vv_nxv2i32_nxv2i16_nxv2i16(<vscale x 2 x i32> %0, <vscale x 2 x i16> %1, <vscale x 2 x i16> %2, i64 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccu_vv_nxv2i32_nxv2i16_nxv2i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e16,mf2,ta,mu
+; CHECK-NEXT:    vwmaccu.vv v16, v17, v18
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i32> @llvm.riscv.vwmaccu.nxv2i32.nxv2i16(
+    <vscale x 2 x i32> %0,
+    <vscale x 2 x i16> %1,
+    <vscale x 2 x i16> %2,
+    i64 %3)
+
+  ret <vscale x 2 x i32> %a
+}
+
+declare <vscale x 2 x i32> @llvm.riscv.vwmaccu.mask.nxv2i32.nxv2i16(
+  <vscale x 2 x i32>,
+  <vscale x 2 x i16>,
+  <vscale x 2 x i16>,
+  <vscale x 2 x i1>,
+  i64);
+
+define <vscale x 2 x i32>  @intrinsic_vwmaccu_mask_vv_nxv2i32_nxv2i16_nxv2i16(<vscale x 2 x i32> %0, <vscale x 2 x i16> %1, <vscale x 2 x i16> %2, <vscale x 2 x i1> %3, i64 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccu_mask_vv_nxv2i32_nxv2i16_nxv2i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e16,mf2,ta,mu
+; CHECK-NEXT:    vwmaccu.vv v16, v17, v18, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i32> @llvm.riscv.vwmaccu.mask.nxv2i32.nxv2i16(
+    <vscale x 2 x i32> %0,
+    <vscale x 2 x i16> %1,
+    <vscale x 2 x i16> %2,
+    <vscale x 2 x i1> %3,
+    i64 %4)
+
+  ret <vscale x 2 x i32> %a
+}
+
+declare <vscale x 4 x i32> @llvm.riscv.vwmaccu.nxv4i32.nxv4i16(
+  <vscale x 4 x i32>,
+  <vscale x 4 x i16>,
+  <vscale x 4 x i16>,
+  i64);
+
+define <vscale x 4 x i32>  @intrinsic_vwmaccu_vv_nxv4i32_nxv4i16_nxv4i16(<vscale x 4 x i32> %0, <vscale x 4 x i16> %1, <vscale x 4 x i16> %2, i64 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccu_vv_nxv4i32_nxv4i16_nxv4i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e16,m1,ta,mu
+; CHECK-NEXT:    vwmaccu.vv v16, v18, v19
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i32> @llvm.riscv.vwmaccu.nxv4i32.nxv4i16(
+    <vscale x 4 x i32> %0,
+    <vscale x 4 x i16> %1,
+    <vscale x 4 x i16> %2,
+    i64 %3)
+
+  ret <vscale x 4 x i32> %a
+}
+
+declare <vscale x 4 x i32> @llvm.riscv.vwmaccu.mask.nxv4i32.nxv4i16(
+  <vscale x 4 x i32>,
+  <vscale x 4 x i16>,
+  <vscale x 4 x i16>,
+  <vscale x 4 x i1>,
+  i64);
+
+define <vscale x 4 x i32>  @intrinsic_vwmaccu_mask_vv_nxv4i32_nxv4i16_nxv4i16(<vscale x 4 x i32> %0, <vscale x 4 x i16> %1, <vscale x 4 x i16> %2, <vscale x 4 x i1> %3, i64 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccu_mask_vv_nxv4i32_nxv4i16_nxv4i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e16,m1,ta,mu
+; CHECK-NEXT:    vwmaccu.vv v16, v18, v19, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i32> @llvm.riscv.vwmaccu.mask.nxv4i32.nxv4i16(
+    <vscale x 4 x i32> %0,
+    <vscale x 4 x i16> %1,
+    <vscale x 4 x i16> %2,
+    <vscale x 4 x i1> %3,
+    i64 %4)
+
+  ret <vscale x 4 x i32> %a
+}
+
+declare <vscale x 8 x i32> @llvm.riscv.vwmaccu.nxv8i32.nxv8i16(
+  <vscale x 8 x i32>,
+  <vscale x 8 x i16>,
+  <vscale x 8 x i16>,
+  i64);
+
+define <vscale x 8 x i32>  @intrinsic_vwmaccu_vv_nxv8i32_nxv8i16_nxv8i16(<vscale x 8 x i32> %0, <vscale x 8 x i16> %1, <vscale x 8 x i16> %2, i64 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccu_vv_nxv8i32_nxv8i16_nxv8i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e16,m2,ta,mu
+; CHECK-NEXT:    vwmaccu.vv v16, v20, v22
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i32> @llvm.riscv.vwmaccu.nxv8i32.nxv8i16(
+    <vscale x 8 x i32> %0,
+    <vscale x 8 x i16> %1,
+    <vscale x 8 x i16> %2,
+    i64 %3)
+
+  ret <vscale x 8 x i32> %a
+}
+
+declare <vscale x 8 x i32> @llvm.riscv.vwmaccu.mask.nxv8i32.nxv8i16(
+  <vscale x 8 x i32>,
+  <vscale x 8 x i16>,
+  <vscale x 8 x i16>,
+  <vscale x 8 x i1>,
+  i64);
+
+define <vscale x 8 x i32>  @intrinsic_vwmaccu_mask_vv_nxv8i32_nxv8i16_nxv8i16(<vscale x 8 x i32> %0, <vscale x 8 x i16> %1, <vscale x 8 x i16> %2, <vscale x 8 x i1> %3, i64 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccu_mask_vv_nxv8i32_nxv8i16_nxv8i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e16,m2,ta,mu
+; CHECK-NEXT:    vwmaccu.vv v16, v20, v22, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i32> @llvm.riscv.vwmaccu.mask.nxv8i32.nxv8i16(
+    <vscale x 8 x i32> %0,
+    <vscale x 8 x i16> %1,
+    <vscale x 8 x i16> %2,
+    <vscale x 8 x i1> %3,
+    i64 %4)
+
+  ret <vscale x 8 x i32> %a
+}
+
+declare <vscale x 16 x i32> @llvm.riscv.vwmaccu.nxv16i32.nxv16i16(
+  <vscale x 16 x i32>,
+  <vscale x 16 x i16>,
+  <vscale x 16 x i16>,
+  i64);
+
+define <vscale x 16 x i32>  @intrinsic_vwmaccu_vv_nxv16i32_nxv16i16_nxv16i16(<vscale x 16 x i32> %0, <vscale x 16 x i16> %1, <vscale x 16 x i16> %2, i64 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccu_vv_nxv16i32_nxv16i16_nxv16i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a3, zero, e16,m4,ta,mu
+; CHECK-NEXT:    vle16.v v28, (a1)
+; CHECK-NEXT:    vle16.v v8, (a0)
+; CHECK-NEXT:    vsetvli a0, a2, e16,m4,ta,mu
+; CHECK-NEXT:    vwmaccu.vv v16, v8, v28
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 16 x i32> @llvm.riscv.vwmaccu.nxv16i32.nxv16i16(
+    <vscale x 16 x i32> %0,
+    <vscale x 16 x i16> %1,
+    <vscale x 16 x i16> %2,
+    i64 %3)
+
+  ret <vscale x 16 x i32> %a
+}
+
+declare <vscale x 16 x i32> @llvm.riscv.vwmaccu.mask.nxv16i32.nxv16i16(
+  <vscale x 16 x i32>,
+  <vscale x 16 x i16>,
+  <vscale x 16 x i16>,
+  <vscale x 16 x i1>,
+  i64);
+
+define <vscale x 16 x i32>  @intrinsic_vwmaccu_mask_vv_nxv16i32_nxv16i16_nxv16i16(<vscale x 16 x i32> %0, <vscale x 16 x i16> %1, <vscale x 16 x i16> %2, <vscale x 16 x i1> %3, i64 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccu_mask_vv_nxv16i32_nxv16i16_nxv16i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a3, zero, e16,m4,ta,mu
+; CHECK-NEXT:    vle16.v v28, (a1)
+; CHECK-NEXT:    vle16.v v8, (a0)
+; CHECK-NEXT:    vsetvli a0, a2, e16,m4,ta,mu
+; CHECK-NEXT:    vwmaccu.vv v16, v8, v28, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 16 x i32> @llvm.riscv.vwmaccu.mask.nxv16i32.nxv16i16(
+    <vscale x 16 x i32> %0,
+    <vscale x 16 x i16> %1,
+    <vscale x 16 x i16> %2,
+    <vscale x 16 x i1> %3,
+    i64 %4)
+
+  ret <vscale x 16 x i32> %a
+}
+
+declare <vscale x 1 x i64> @llvm.riscv.vwmaccu.nxv1i64.nxv1i32(
+  <vscale x 1 x i64>,
+  <vscale x 1 x i32>,
+  <vscale x 1 x i32>,
+  i64);
+
+define <vscale x 1 x i64>  @intrinsic_vwmaccu_vv_nxv1i64_nxv1i32_nxv1i32(<vscale x 1 x i64> %0, <vscale x 1 x i32> %1, <vscale x 1 x i32> %2, i64 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccu_vv_nxv1i64_nxv1i32_nxv1i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e32,mf2,ta,mu
+; CHECK-NEXT:    vwmaccu.vv v16, v17, v18
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i64> @llvm.riscv.vwmaccu.nxv1i64.nxv1i32(
+    <vscale x 1 x i64> %0,
+    <vscale x 1 x i32> %1,
+    <vscale x 1 x i32> %2,
+    i64 %3)
+
+  ret <vscale x 1 x i64> %a
+}
+
+declare <vscale x 1 x i64> @llvm.riscv.vwmaccu.mask.nxv1i64.nxv1i32(
+  <vscale x 1 x i64>,
+  <vscale x 1 x i32>,
+  <vscale x 1 x i32>,
+  <vscale x 1 x i1>,
+  i64);
+
+define <vscale x 1 x i64>  @intrinsic_vwmaccu_mask_vv_nxv1i64_nxv1i32_nxv1i32(<vscale x 1 x i64> %0, <vscale x 1 x i32> %1, <vscale x 1 x i32> %2, <vscale x 1 x i1> %3, i64 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccu_mask_vv_nxv1i64_nxv1i32_nxv1i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e32,mf2,ta,mu
+; CHECK-NEXT:    vwmaccu.vv v16, v17, v18, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i64> @llvm.riscv.vwmaccu.mask.nxv1i64.nxv1i32(
+    <vscale x 1 x i64> %0,
+    <vscale x 1 x i32> %1,
+    <vscale x 1 x i32> %2,
+    <vscale x 1 x i1> %3,
+    i64 %4)
+
+  ret <vscale x 1 x i64> %a
+}
+
+declare <vscale x 2 x i64> @llvm.riscv.vwmaccu.nxv2i64.nxv2i32(
+  <vscale x 2 x i64>,
+  <vscale x 2 x i32>,
+  <vscale x 2 x i32>,
+  i64);
+
+define <vscale x 2 x i64>  @intrinsic_vwmaccu_vv_nxv2i64_nxv2i32_nxv2i32(<vscale x 2 x i64> %0, <vscale x 2 x i32> %1, <vscale x 2 x i32> %2, i64 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccu_vv_nxv2i64_nxv2i32_nxv2i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e32,m1,ta,mu
+; CHECK-NEXT:    vwmaccu.vv v16, v18, v19
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i64> @llvm.riscv.vwmaccu.nxv2i64.nxv2i32(
+    <vscale x 2 x i64> %0,
+    <vscale x 2 x i32> %1,
+    <vscale x 2 x i32> %2,
+    i64 %3)
+
+  ret <vscale x 2 x i64> %a
+}
+
+declare <vscale x 2 x i64> @llvm.riscv.vwmaccu.mask.nxv2i64.nxv2i32(
+  <vscale x 2 x i64>,
+  <vscale x 2 x i32>,
+  <vscale x 2 x i32>,
+  <vscale x 2 x i1>,
+  i64);
+
+define <vscale x 2 x i64>  @intrinsic_vwmaccu_mask_vv_nxv2i64_nxv2i32_nxv2i32(<vscale x 2 x i64> %0, <vscale x 2 x i32> %1, <vscale x 2 x i32> %2, <vscale x 2 x i1> %3, i64 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccu_mask_vv_nxv2i64_nxv2i32_nxv2i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e32,m1,ta,mu
+; CHECK-NEXT:    vwmaccu.vv v16, v18, v19, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i64> @llvm.riscv.vwmaccu.mask.nxv2i64.nxv2i32(
+    <vscale x 2 x i64> %0,
+    <vscale x 2 x i32> %1,
+    <vscale x 2 x i32> %2,
+    <vscale x 2 x i1> %3,
+    i64 %4)
+
+  ret <vscale x 2 x i64> %a
+}
+
+declare <vscale x 4 x i64> @llvm.riscv.vwmaccu.nxv4i64.nxv4i32(
+  <vscale x 4 x i64>,
+  <vscale x 4 x i32>,
+  <vscale x 4 x i32>,
+  i64);
+
+define <vscale x 4 x i64>  @intrinsic_vwmaccu_vv_nxv4i64_nxv4i32_nxv4i32(<vscale x 4 x i64> %0, <vscale x 4 x i32> %1, <vscale x 4 x i32> %2, i64 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccu_vv_nxv4i64_nxv4i32_nxv4i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e32,m2,ta,mu
+; CHECK-NEXT:    vwmaccu.vv v16, v20, v22
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i64> @llvm.riscv.vwmaccu.nxv4i64.nxv4i32(
+    <vscale x 4 x i64> %0,
+    <vscale x 4 x i32> %1,
+    <vscale x 4 x i32> %2,
+    i64 %3)
+
+  ret <vscale x 4 x i64> %a
+}
+
+declare <vscale x 4 x i64> @llvm.riscv.vwmaccu.mask.nxv4i64.nxv4i32(
+  <vscale x 4 x i64>,
+  <vscale x 4 x i32>,
+  <vscale x 4 x i32>,
+  <vscale x 4 x i1>,
+  i64);
+
+define <vscale x 4 x i64>  @intrinsic_vwmaccu_mask_vv_nxv4i64_nxv4i32_nxv4i32(<vscale x 4 x i64> %0, <vscale x 4 x i32> %1, <vscale x 4 x i32> %2, <vscale x 4 x i1> %3, i64 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccu_mask_vv_nxv4i64_nxv4i32_nxv4i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e32,m2,ta,mu
+; CHECK-NEXT:    vwmaccu.vv v16, v20, v22, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i64> @llvm.riscv.vwmaccu.mask.nxv4i64.nxv4i32(
+    <vscale x 4 x i64> %0,
+    <vscale x 4 x i32> %1,
+    <vscale x 4 x i32> %2,
+    <vscale x 4 x i1> %3,
+    i64 %4)
+
+  ret <vscale x 4 x i64> %a
+}
+
+declare <vscale x 8 x i64> @llvm.riscv.vwmaccu.nxv8i64.nxv8i32(
+  <vscale x 8 x i64>,
+  <vscale x 8 x i32>,
+  <vscale x 8 x i32>,
+  i64);
+
+define <vscale x 8 x i64>  @intrinsic_vwmaccu_vv_nxv8i64_nxv8i32_nxv8i32(<vscale x 8 x i64> %0, <vscale x 8 x i32> %1, <vscale x 8 x i32> %2, i64 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccu_vv_nxv8i64_nxv8i32_nxv8i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a3, zero, e32,m4,ta,mu
+; CHECK-NEXT:    vle32.v v28, (a1)
+; CHECK-NEXT:    vle32.v v8, (a0)
+; CHECK-NEXT:    vsetvli a0, a2, e32,m4,ta,mu
+; CHECK-NEXT:    vwmaccu.vv v16, v8, v28
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i64> @llvm.riscv.vwmaccu.nxv8i64.nxv8i32(
+    <vscale x 8 x i64> %0,
+    <vscale x 8 x i32> %1,
+    <vscale x 8 x i32> %2,
+    i64 %3)
+
+  ret <vscale x 8 x i64> %a
+}
+
+declare <vscale x 8 x i64> @llvm.riscv.vwmaccu.mask.nxv8i64.nxv8i32(
+  <vscale x 8 x i64>,
+  <vscale x 8 x i32>,
+  <vscale x 8 x i32>,
+  <vscale x 8 x i1>,
+  i64);
+
+define <vscale x 8 x i64>  @intrinsic_vwmaccu_mask_vv_nxv8i64_nxv8i32_nxv8i32(<vscale x 8 x i64> %0, <vscale x 8 x i32> %1, <vscale x 8 x i32> %2, <vscale x 8 x i1> %3, i64 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccu_mask_vv_nxv8i64_nxv8i32_nxv8i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a3, zero, e32,m4,ta,mu
+; CHECK-NEXT:    vle32.v v28, (a1)
+; CHECK-NEXT:    vle32.v v8, (a0)
+; CHECK-NEXT:    vsetvli a0, a2, e32,m4,ta,mu
+; CHECK-NEXT:    vwmaccu.vv v16, v8, v28, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i64> @llvm.riscv.vwmaccu.mask.nxv8i64.nxv8i32(
+    <vscale x 8 x i64> %0,
+    <vscale x 8 x i32> %1,
+    <vscale x 8 x i32> %2,
+    <vscale x 8 x i1> %3,
+    i64 %4)
+
+  ret <vscale x 8 x i64> %a
+}
+
+declare <vscale x 1 x i16> @llvm.riscv.vwmaccu.nxv1i16.i8(
+  <vscale x 1 x i16>,
+  i8,
+  <vscale x 1 x i8>,
+  i64);
+
+define <vscale x 1 x i16>  @intrinsic_vwmaccu_vx_nxv1i16_i8_nxv1i8(<vscale x 1 x i16> %0, i8 %1, <vscale x 1 x i8> %2, i64 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccu_vx_nxv1i16_i8_nxv1i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e8,mf8,ta,mu
+; CHECK-NEXT:    vwmaccu.vx v16, a0, v17
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i16> @llvm.riscv.vwmaccu.nxv1i16.i8(
+    <vscale x 1 x i16> %0,
+    i8 %1,
+    <vscale x 1 x i8> %2,
+    i64 %3)
+
+  ret <vscale x 1 x i16> %a
+}
+
+declare <vscale x 1 x i16> @llvm.riscv.vwmaccu.mask.nxv1i16.i8(
+  <vscale x 1 x i16>,
+  i8,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i1>,
+  i64);
+
+define <vscale x 1 x i16> @intrinsic_vwmaccu_mask_vx_nxv1i16_i8_nxv1i8(<vscale x 1 x i16> %0, i8 %1, <vscale x 1 x i8> %2, <vscale x 1 x i1> %3, i64 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccu_mask_vx_nxv1i16_i8_nxv1i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e8,mf8,ta,mu
+; CHECK-NEXT:    vwmaccu.vx v16, a0, v17, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i16> @llvm.riscv.vwmaccu.mask.nxv1i16.i8(
+    <vscale x 1 x i16> %0,
+    i8 %1,
+    <vscale x 1 x i8> %2,
+    <vscale x 1 x i1> %3,
+    i64 %4)
+
+  ret <vscale x 1 x i16> %a
+}
+
+declare <vscale x 2 x i16> @llvm.riscv.vwmaccu.nxv2i16.i8(
+  <vscale x 2 x i16>,
+  i8,
+  <vscale x 2 x i8>,
+  i64);
+
+define <vscale x 2 x i16>  @intrinsic_vwmaccu_vx_nxv2i16_i8_nxv2i8(<vscale x 2 x i16> %0, i8 %1, <vscale x 2 x i8> %2, i64 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccu_vx_nxv2i16_i8_nxv2i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e8,mf4,ta,mu
+; CHECK-NEXT:    vwmaccu.vx v16, a0, v17
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i16> @llvm.riscv.vwmaccu.nxv2i16.i8(
+    <vscale x 2 x i16> %0,
+    i8 %1,
+    <vscale x 2 x i8> %2,
+    i64 %3)
+
+  ret <vscale x 2 x i16> %a
+}
+
+declare <vscale x 2 x i16> @llvm.riscv.vwmaccu.mask.nxv2i16.i8(
+  <vscale x 2 x i16>,
+  i8,
+  <vscale x 2 x i8>,
+  <vscale x 2 x i1>,
+  i64);
+
+define <vscale x 2 x i16> @intrinsic_vwmaccu_mask_vx_nxv2i16_i8_nxv2i8(<vscale x 2 x i16> %0, i8 %1, <vscale x 2 x i8> %2, <vscale x 2 x i1> %3, i64 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccu_mask_vx_nxv2i16_i8_nxv2i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e8,mf4,ta,mu
+; CHECK-NEXT:    vwmaccu.vx v16, a0, v17, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i16> @llvm.riscv.vwmaccu.mask.nxv2i16.i8(
+    <vscale x 2 x i16> %0,
+    i8 %1,
+    <vscale x 2 x i8> %2,
+    <vscale x 2 x i1> %3,
+    i64 %4)
+
+  ret <vscale x 2 x i16> %a
+}
+
+declare <vscale x 4 x i16> @llvm.riscv.vwmaccu.nxv4i16.i8(
+  <vscale x 4 x i16>,
+  i8,
+  <vscale x 4 x i8>,
+  i64);
+
+define <vscale x 4 x i16>  @intrinsic_vwmaccu_vx_nxv4i16_i8_nxv4i8(<vscale x 4 x i16> %0, i8 %1, <vscale x 4 x i8> %2, i64 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccu_vx_nxv4i16_i8_nxv4i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e8,mf2,ta,mu
+; CHECK-NEXT:    vwmaccu.vx v16, a0, v17
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i16> @llvm.riscv.vwmaccu.nxv4i16.i8(
+    <vscale x 4 x i16> %0,
+    i8 %1,
+    <vscale x 4 x i8> %2,
+    i64 %3)
+
+  ret <vscale x 4 x i16> %a
+}
+
+declare <vscale x 4 x i16> @llvm.riscv.vwmaccu.mask.nxv4i16.i8(
+  <vscale x 4 x i16>,
+  i8,
+  <vscale x 4 x i8>,
+  <vscale x 4 x i1>,
+  i64);
+
+define <vscale x 4 x i16> @intrinsic_vwmaccu_mask_vx_nxv4i16_i8_nxv4i8(<vscale x 4 x i16> %0, i8 %1, <vscale x 4 x i8> %2, <vscale x 4 x i1> %3, i64 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccu_mask_vx_nxv4i16_i8_nxv4i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e8,mf2,ta,mu
+; CHECK-NEXT:    vwmaccu.vx v16, a0, v17, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i16> @llvm.riscv.vwmaccu.mask.nxv4i16.i8(
+    <vscale x 4 x i16> %0,
+    i8 %1,
+    <vscale x 4 x i8> %2,
+    <vscale x 4 x i1> %3,
+    i64 %4)
+
+  ret <vscale x 4 x i16> %a
+}
+
+declare <vscale x 8 x i16> @llvm.riscv.vwmaccu.nxv8i16.i8(
+  <vscale x 8 x i16>,
+  i8,
+  <vscale x 8 x i8>,
+  i64);
+
+define <vscale x 8 x i16>  @intrinsic_vwmaccu_vx_nxv8i16_i8_nxv8i8(<vscale x 8 x i16> %0, i8 %1, <vscale x 8 x i8> %2, i64 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccu_vx_nxv8i16_i8_nxv8i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e8,m1,ta,mu
+; CHECK-NEXT:    vwmaccu.vx v16, a0, v18
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i16> @llvm.riscv.vwmaccu.nxv8i16.i8(
+    <vscale x 8 x i16> %0,
+    i8 %1,
+    <vscale x 8 x i8> %2,
+    i64 %3)
+
+  ret <vscale x 8 x i16> %a
+}
+
+declare <vscale x 8 x i16> @llvm.riscv.vwmaccu.mask.nxv8i16.i8(
+  <vscale x 8 x i16>,
+  i8,
+  <vscale x 8 x i8>,
+  <vscale x 8 x i1>,
+  i64);
+
+define <vscale x 8 x i16> @intrinsic_vwmaccu_mask_vx_nxv8i16_i8_nxv8i8(<vscale x 8 x i16> %0, i8 %1, <vscale x 8 x i8> %2, <vscale x 8 x i1> %3, i64 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccu_mask_vx_nxv8i16_i8_nxv8i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e8,m1,ta,mu
+; CHECK-NEXT:    vwmaccu.vx v16, a0, v18, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i16> @llvm.riscv.vwmaccu.mask.nxv8i16.i8(
+    <vscale x 8 x i16> %0,
+    i8 %1,
+    <vscale x 8 x i8> %2,
+    <vscale x 8 x i1> %3,
+    i64 %4)
+
+  ret <vscale x 8 x i16> %a
+}
+
+declare <vscale x 16 x i16> @llvm.riscv.vwmaccu.nxv16i16.i8(
+  <vscale x 16 x i16>,
+  i8,
+  <vscale x 16 x i8>,
+  i64);
+
+define <vscale x 16 x i16>  @intrinsic_vwmaccu_vx_nxv16i16_i8_nxv16i8(<vscale x 16 x i16> %0, i8 %1, <vscale x 16 x i8> %2, i64 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccu_vx_nxv16i16_i8_nxv16i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e8,m2,ta,mu
+; CHECK-NEXT:    vwmaccu.vx v16, a0, v20
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 16 x i16> @llvm.riscv.vwmaccu.nxv16i16.i8(
+    <vscale x 16 x i16> %0,
+    i8 %1,
+    <vscale x 16 x i8> %2,
+    i64 %3)
+
+  ret <vscale x 16 x i16> %a
+}
+
+declare <vscale x 16 x i16> @llvm.riscv.vwmaccu.mask.nxv16i16.i8(
+  <vscale x 16 x i16>,
+  i8,
+  <vscale x 16 x i8>,
+  <vscale x 16 x i1>,
+  i64);
+
+define <vscale x 16 x i16> @intrinsic_vwmaccu_mask_vx_nxv16i16_i8_nxv16i8(<vscale x 16 x i16> %0, i8 %1, <vscale x 16 x i8> %2, <vscale x 16 x i1> %3, i64 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccu_mask_vx_nxv16i16_i8_nxv16i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e8,m2,ta,mu
+; CHECK-NEXT:    vwmaccu.vx v16, a0, v20, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 16 x i16> @llvm.riscv.vwmaccu.mask.nxv16i16.i8(
+    <vscale x 16 x i16> %0,
+    i8 %1,
+    <vscale x 16 x i8> %2,
+    <vscale x 16 x i1> %3,
+    i64 %4)
+
+  ret <vscale x 16 x i16> %a
+}
+
+declare <vscale x 32 x i16> @llvm.riscv.vwmaccu.nxv32i16.i8(
+  <vscale x 32 x i16>,
+  i8,
+  <vscale x 32 x i8>,
+  i64);
+
+define <vscale x 32 x i16>  @intrinsic_vwmaccu_vx_nxv32i16_i8_nxv32i8(<vscale x 32 x i16> %0, i8 %1, <vscale x 32 x i8> %2, i64 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccu_vx_nxv32i16_i8_nxv32i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a3, zero, e8,m4,ta,mu
+; CHECK-NEXT:    vle8.v v28, (a1)
+; CHECK-NEXT:    vsetvli a1, a2, e8,m4,ta,mu
+; CHECK-NEXT:    vwmaccu.vx v16, a0, v28
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 32 x i16> @llvm.riscv.vwmaccu.nxv32i16.i8(
+    <vscale x 32 x i16> %0,
+    i8 %1,
+    <vscale x 32 x i8> %2,
+    i64 %3)
+
+  ret <vscale x 32 x i16> %a
+}
+
+declare <vscale x 32 x i16> @llvm.riscv.vwmaccu.mask.nxv32i16.i8(
+  <vscale x 32 x i16>,
+  i8,
+  <vscale x 32 x i8>,
+  <vscale x 32 x i1>,
+  i64);
+
+define <vscale x 32 x i16> @intrinsic_vwmaccu_mask_vx_nxv32i16_i8_nxv32i8(<vscale x 32 x i16> %0, i8 %1, <vscale x 32 x i8> %2, <vscale x 32 x i1> %3, i64 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccu_mask_vx_nxv32i16_i8_nxv32i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a3, zero, e8,m4,ta,mu
+; CHECK-NEXT:    vle8.v v28, (a1)
+; CHECK-NEXT:    vsetvli a1, a2, e8,m4,ta,mu
+; CHECK-NEXT:    vwmaccu.vx v16, a0, v28, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 32 x i16> @llvm.riscv.vwmaccu.mask.nxv32i16.i8(
+    <vscale x 32 x i16> %0,
+    i8 %1,
+    <vscale x 32 x i8> %2,
+    <vscale x 32 x i1> %3,
+    i64 %4)
+
+  ret <vscale x 32 x i16> %a
+}
+
+declare <vscale x 1 x i32> @llvm.riscv.vwmaccu.nxv1i32.i16(
+  <vscale x 1 x i32>,
+  i16,
+  <vscale x 1 x i16>,
+  i64);
+
+define <vscale x 1 x i32>  @intrinsic_vwmaccu_vx_nxv1i32_i16_nxv1i16(<vscale x 1 x i32> %0, i16 %1, <vscale x 1 x i16> %2, i64 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccu_vx_nxv1i32_i16_nxv1i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e16,mf4,ta,mu
+; CHECK-NEXT:    vwmaccu.vx v16, a0, v17
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i32> @llvm.riscv.vwmaccu.nxv1i32.i16(
+    <vscale x 1 x i32> %0,
+    i16 %1,
+    <vscale x 1 x i16> %2,
+    i64 %3)
+
+  ret <vscale x 1 x i32> %a
+}
+
+declare <vscale x 1 x i32> @llvm.riscv.vwmaccu.mask.nxv1i32.i16(
+  <vscale x 1 x i32>,
+  i16,
+  <vscale x 1 x i16>,
+  <vscale x 1 x i1>,
+  i64);
+
+define <vscale x 1 x i32> @intrinsic_vwmaccu_mask_vx_nxv1i32_i16_nxv1i16(<vscale x 1 x i32> %0, i16 %1, <vscale x 1 x i16> %2, <vscale x 1 x i1> %3, i64 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccu_mask_vx_nxv1i32_i16_nxv1i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e16,mf4,ta,mu
+; CHECK-NEXT:    vwmaccu.vx v16, a0, v17, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i32> @llvm.riscv.vwmaccu.mask.nxv1i32.i16(
+    <vscale x 1 x i32> %0,
+    i16 %1,
+    <vscale x 1 x i16> %2,
+    <vscale x 1 x i1> %3,
+    i64 %4)
+
+  ret <vscale x 1 x i32> %a
+}
+
+declare <vscale x 2 x i32> @llvm.riscv.vwmaccu.nxv2i32.i16(
+  <vscale x 2 x i32>,
+  i16,
+  <vscale x 2 x i16>,
+  i64);
+
+define <vscale x 2 x i32>  @intrinsic_vwmaccu_vx_nxv2i32_i16_nxv2i16(<vscale x 2 x i32> %0, i16 %1, <vscale x 2 x i16> %2, i64 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccu_vx_nxv2i32_i16_nxv2i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e16,mf2,ta,mu
+; CHECK-NEXT:    vwmaccu.vx v16, a0, v17
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i32> @llvm.riscv.vwmaccu.nxv2i32.i16(
+    <vscale x 2 x i32> %0,
+    i16 %1,
+    <vscale x 2 x i16> %2,
+    i64 %3)
+
+  ret <vscale x 2 x i32> %a
+}
+
+declare <vscale x 2 x i32> @llvm.riscv.vwmaccu.mask.nxv2i32.i16(
+  <vscale x 2 x i32>,
+  i16,
+  <vscale x 2 x i16>,
+  <vscale x 2 x i1>,
+  i64);
+
+define <vscale x 2 x i32> @intrinsic_vwmaccu_mask_vx_nxv2i32_i16_nxv2i16(<vscale x 2 x i32> %0, i16 %1, <vscale x 2 x i16> %2, <vscale x 2 x i1> %3, i64 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccu_mask_vx_nxv2i32_i16_nxv2i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e16,mf2,ta,mu
+; CHECK-NEXT:    vwmaccu.vx v16, a0, v17, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i32> @llvm.riscv.vwmaccu.mask.nxv2i32.i16(
+    <vscale x 2 x i32> %0,
+    i16 %1,
+    <vscale x 2 x i16> %2,
+    <vscale x 2 x i1> %3,
+    i64 %4)
+
+  ret <vscale x 2 x i32> %a
+}
+
+declare <vscale x 4 x i32> @llvm.riscv.vwmaccu.nxv4i32.i16(
+  <vscale x 4 x i32>,
+  i16,
+  <vscale x 4 x i16>,
+  i64);
+
+define <vscale x 4 x i32>  @intrinsic_vwmaccu_vx_nxv4i32_i16_nxv4i16(<vscale x 4 x i32> %0, i16 %1, <vscale x 4 x i16> %2, i64 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccu_vx_nxv4i32_i16_nxv4i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e16,m1,ta,mu
+; CHECK-NEXT:    vwmaccu.vx v16, a0, v18
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i32> @llvm.riscv.vwmaccu.nxv4i32.i16(
+    <vscale x 4 x i32> %0,
+    i16 %1,
+    <vscale x 4 x i16> %2,
+    i64 %3)
+
+  ret <vscale x 4 x i32> %a
+}
+
+declare <vscale x 4 x i32> @llvm.riscv.vwmaccu.mask.nxv4i32.i16(
+  <vscale x 4 x i32>,
+  i16,
+  <vscale x 4 x i16>,
+  <vscale x 4 x i1>,
+  i64);
+
+define <vscale x 4 x i32> @intrinsic_vwmaccu_mask_vx_nxv4i32_i16_nxv4i16(<vscale x 4 x i32> %0, i16 %1, <vscale x 4 x i16> %2, <vscale x 4 x i1> %3, i64 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccu_mask_vx_nxv4i32_i16_nxv4i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e16,m1,ta,mu
+; CHECK-NEXT:    vwmaccu.vx v16, a0, v18, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i32> @llvm.riscv.vwmaccu.mask.nxv4i32.i16(
+    <vscale x 4 x i32> %0,
+    i16 %1,
+    <vscale x 4 x i16> %2,
+    <vscale x 4 x i1> %3,
+    i64 %4)
+
+  ret <vscale x 4 x i32> %a
+}
+
+declare <vscale x 8 x i32> @llvm.riscv.vwmaccu.nxv8i32.i16(
+  <vscale x 8 x i32>,
+  i16,
+  <vscale x 8 x i16>,
+  i64);
+
+define <vscale x 8 x i32>  @intrinsic_vwmaccu_vx_nxv8i32_i16_nxv8i16(<vscale x 8 x i32> %0, i16 %1, <vscale x 8 x i16> %2, i64 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccu_vx_nxv8i32_i16_nxv8i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e16,m2,ta,mu
+; CHECK-NEXT:    vwmaccu.vx v16, a0, v20
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i32> @llvm.riscv.vwmaccu.nxv8i32.i16(
+    <vscale x 8 x i32> %0,
+    i16 %1,
+    <vscale x 8 x i16> %2,
+    i64 %3)
+
+  ret <vscale x 8 x i32> %a
+}
+
+declare <vscale x 8 x i32> @llvm.riscv.vwmaccu.mask.nxv8i32.i16(
+  <vscale x 8 x i32>,
+  i16,
+  <vscale x 8 x i16>,
+  <vscale x 8 x i1>,
+  i64);
+
+define <vscale x 8 x i32> @intrinsic_vwmaccu_mask_vx_nxv8i32_i16_nxv8i16(<vscale x 8 x i32> %0, i16 %1, <vscale x 8 x i16> %2, <vscale x 8 x i1> %3, i64 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccu_mask_vx_nxv8i32_i16_nxv8i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e16,m2,ta,mu
+; CHECK-NEXT:    vwmaccu.vx v16, a0, v20, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i32> @llvm.riscv.vwmaccu.mask.nxv8i32.i16(
+    <vscale x 8 x i32> %0,
+    i16 %1,
+    <vscale x 8 x i16> %2,
+    <vscale x 8 x i1> %3,
+    i64 %4)
+
+  ret <vscale x 8 x i32> %a
+}
+
+declare <vscale x 16 x i32> @llvm.riscv.vwmaccu.nxv16i32.i16(
+  <vscale x 16 x i32>,
+  i16,
+  <vscale x 16 x i16>,
+  i64);
+
+define <vscale x 16 x i32>  @intrinsic_vwmaccu_vx_nxv16i32_i16_nxv16i16(<vscale x 16 x i32> %0, i16 %1, <vscale x 16 x i16> %2, i64 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccu_vx_nxv16i32_i16_nxv16i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a3, zero, e16,m4,ta,mu
+; CHECK-NEXT:    vle16.v v28, (a1)
+; CHECK-NEXT:    vsetvli a1, a2, e16,m4,ta,mu
+; CHECK-NEXT:    vwmaccu.vx v16, a0, v28
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 16 x i32> @llvm.riscv.vwmaccu.nxv16i32.i16(
+    <vscale x 16 x i32> %0,
+    i16 %1,
+    <vscale x 16 x i16> %2,
+    i64 %3)
+
+  ret <vscale x 16 x i32> %a
+}
+
+declare <vscale x 16 x i32> @llvm.riscv.vwmaccu.mask.nxv16i32.i16(
+  <vscale x 16 x i32>,
+  i16,
+  <vscale x 16 x i16>,
+  <vscale x 16 x i1>,
+  i64);
+
+define <vscale x 16 x i32> @intrinsic_vwmaccu_mask_vx_nxv16i32_i16_nxv16i16(<vscale x 16 x i32> %0, i16 %1, <vscale x 16 x i16> %2, <vscale x 16 x i1> %3, i64 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccu_mask_vx_nxv16i32_i16_nxv16i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a3, zero, e16,m4,ta,mu
+; CHECK-NEXT:    vle16.v v28, (a1)
+; CHECK-NEXT:    vsetvli a1, a2, e16,m4,ta,mu
+; CHECK-NEXT:    vwmaccu.vx v16, a0, v28, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 16 x i32> @llvm.riscv.vwmaccu.mask.nxv16i32.i16(
+    <vscale x 16 x i32> %0,
+    i16 %1,
+    <vscale x 16 x i16> %2,
+    <vscale x 16 x i1> %3,
+    i64 %4)
+
+  ret <vscale x 16 x i32> %a
+}
+
+declare <vscale x 1 x i64> @llvm.riscv.vwmaccu.nxv1i64.i32(
+  <vscale x 1 x i64>,
+  i32,
+  <vscale x 1 x i32>,
+  i64);
+
+define <vscale x 1 x i64>  @intrinsic_vwmaccu_vx_nxv1i64_i32_nxv1i32(<vscale x 1 x i64> %0, i32 %1, <vscale x 1 x i32> %2, i64 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccu_vx_nxv1i64_i32_nxv1i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e32,mf2,ta,mu
+; CHECK-NEXT:    vwmaccu.vx v16, a0, v17
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i64> @llvm.riscv.vwmaccu.nxv1i64.i32(
+    <vscale x 1 x i64> %0,
+    i32 %1,
+    <vscale x 1 x i32> %2,
+    i64 %3)
+
+  ret <vscale x 1 x i64> %a
+}
+
+declare <vscale x 1 x i64> @llvm.riscv.vwmaccu.mask.nxv1i64.i32(
+  <vscale x 1 x i64>,
+  i32,
+  <vscale x 1 x i32>,
+  <vscale x 1 x i1>,
+  i64);
+
+define <vscale x 1 x i64> @intrinsic_vwmaccu_mask_vx_nxv1i64_i32_nxv1i32(<vscale x 1 x i64> %0, i32 %1, <vscale x 1 x i32> %2, <vscale x 1 x i1> %3, i64 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccu_mask_vx_nxv1i64_i32_nxv1i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e32,mf2,ta,mu
+; CHECK-NEXT:    vwmaccu.vx v16, a0, v17, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i64> @llvm.riscv.vwmaccu.mask.nxv1i64.i32(
+    <vscale x 1 x i64> %0,
+    i32 %1,
+    <vscale x 1 x i32> %2,
+    <vscale x 1 x i1> %3,
+    i64 %4)
+
+  ret <vscale x 1 x i64> %a
+}
+
+declare <vscale x 2 x i64> @llvm.riscv.vwmaccu.nxv2i64.i32(
+  <vscale x 2 x i64>,
+  i32,
+  <vscale x 2 x i32>,
+  i64);
+
+define <vscale x 2 x i64>  @intrinsic_vwmaccu_vx_nxv2i64_i32_nxv2i32(<vscale x 2 x i64> %0, i32 %1, <vscale x 2 x i32> %2, i64 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccu_vx_nxv2i64_i32_nxv2i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e32,m1,ta,mu
+; CHECK-NEXT:    vwmaccu.vx v16, a0, v18
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i64> @llvm.riscv.vwmaccu.nxv2i64.i32(
+    <vscale x 2 x i64> %0,
+    i32 %1,
+    <vscale x 2 x i32> %2,
+    i64 %3)
+
+  ret <vscale x 2 x i64> %a
+}
+
+declare <vscale x 2 x i64> @llvm.riscv.vwmaccu.mask.nxv2i64.i32(
+  <vscale x 2 x i64>,
+  i32,
+  <vscale x 2 x i32>,
+  <vscale x 2 x i1>,
+  i64);
+
+define <vscale x 2 x i64> @intrinsic_vwmaccu_mask_vx_nxv2i64_i32_nxv2i32(<vscale x 2 x i64> %0, i32 %1, <vscale x 2 x i32> %2, <vscale x 2 x i1> %3, i64 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccu_mask_vx_nxv2i64_i32_nxv2i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e32,m1,ta,mu
+; CHECK-NEXT:    vwmaccu.vx v16, a0, v18, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i64> @llvm.riscv.vwmaccu.mask.nxv2i64.i32(
+    <vscale x 2 x i64> %0,
+    i32 %1,
+    <vscale x 2 x i32> %2,
+    <vscale x 2 x i1> %3,
+    i64 %4)
+
+  ret <vscale x 2 x i64> %a
+}
+
+declare <vscale x 4 x i64> @llvm.riscv.vwmaccu.nxv4i64.i32(
+  <vscale x 4 x i64>,
+  i32,
+  <vscale x 4 x i32>,
+  i64);
+
+define <vscale x 4 x i64>  @intrinsic_vwmaccu_vx_nxv4i64_i32_nxv4i32(<vscale x 4 x i64> %0, i32 %1, <vscale x 4 x i32> %2, i64 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccu_vx_nxv4i64_i32_nxv4i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e32,m2,ta,mu
+; CHECK-NEXT:    vwmaccu.vx v16, a0, v20
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i64> @llvm.riscv.vwmaccu.nxv4i64.i32(
+    <vscale x 4 x i64> %0,
+    i32 %1,
+    <vscale x 4 x i32> %2,
+    i64 %3)
+
+  ret <vscale x 4 x i64> %a
+}
+
+declare <vscale x 4 x i64> @llvm.riscv.vwmaccu.mask.nxv4i64.i32(
+  <vscale x 4 x i64>,
+  i32,
+  <vscale x 4 x i32>,
+  <vscale x 4 x i1>,
+  i64);
+
+define <vscale x 4 x i64> @intrinsic_vwmaccu_mask_vx_nxv4i64_i32_nxv4i32(<vscale x 4 x i64> %0, i32 %1, <vscale x 4 x i32> %2, <vscale x 4 x i1> %3, i64 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccu_mask_vx_nxv4i64_i32_nxv4i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e32,m2,ta,mu
+; CHECK-NEXT:    vwmaccu.vx v16, a0, v20, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i64> @llvm.riscv.vwmaccu.mask.nxv4i64.i32(
+    <vscale x 4 x i64> %0,
+    i32 %1,
+    <vscale x 4 x i32> %2,
+    <vscale x 4 x i1> %3,
+    i64 %4)
+
+  ret <vscale x 4 x i64> %a
+}
+
+declare <vscale x 8 x i64> @llvm.riscv.vwmaccu.nxv8i64.i32(
+  <vscale x 8 x i64>,
+  i32,
+  <vscale x 8 x i32>,
+  i64);
+
+define <vscale x 8 x i64>  @intrinsic_vwmaccu_vx_nxv8i64_i32_nxv8i32(<vscale x 8 x i64> %0, i32 %1, <vscale x 8 x i32> %2, i64 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccu_vx_nxv8i64_i32_nxv8i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a3, zero, e32,m4,ta,mu
+; CHECK-NEXT:    vle32.v v28, (a1)
+; CHECK-NEXT:    vsetvli a1, a2, e32,m4,ta,mu
+; CHECK-NEXT:    vwmaccu.vx v16, a0, v28
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i64> @llvm.riscv.vwmaccu.nxv8i64.i32(
+    <vscale x 8 x i64> %0,
+    i32 %1,
+    <vscale x 8 x i32> %2,
+    i64 %3)
+
+  ret <vscale x 8 x i64> %a
+}
+
+declare <vscale x 8 x i64> @llvm.riscv.vwmaccu.mask.nxv8i64.i32(
+  <vscale x 8 x i64>,
+  i32,
+  <vscale x 8 x i32>,
+  <vscale x 8 x i1>,
+  i64);
+
+define <vscale x 8 x i64> @intrinsic_vwmaccu_mask_vx_nxv8i64_i32_nxv8i32(<vscale x 8 x i64> %0, i32 %1, <vscale x 8 x i32> %2, <vscale x 8 x i1> %3, i64 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccu_mask_vx_nxv8i64_i32_nxv8i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a3, zero, e32,m4,ta,mu
+; CHECK-NEXT:    vle32.v v28, (a1)
+; CHECK-NEXT:    vsetvli a1, a2, e32,m4,ta,mu
+; CHECK-NEXT:    vwmaccu.vx v16, a0, v28, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i64> @llvm.riscv.vwmaccu.mask.nxv8i64.i32(
+    <vscale x 8 x i64> %0,
+    i32 %1,
+    <vscale x 8 x i32> %2,
+    <vscale x 8 x i1> %3,
+    i64 %4)
+
+  ret <vscale x 8 x i64> %a
+}

llvm/test/CodeGen/RISCV/rvv/vwmaccus-rv32.ll

+; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
+; RUN: llc -mtriple=riscv32 -mattr=+experimental-v,+f -verify-machineinstrs \
+; RUN:   --riscv-no-aliases < %s | FileCheck %s
+declare <vscale x 1 x i16> @llvm.riscv.vwmaccus.nxv1i16.i8(
+  <vscale x 1 x i16>,
+  i8,
+  <vscale x 1 x i8>,
+  i32);
+
+define <vscale x 1 x i16>  @intrinsic_vwmaccus_vx_nxv1i16_i8_nxv1i8(<vscale x 1 x i16> %0, i8 %1, <vscale x 1 x i8> %2, i32 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccus_vx_nxv1i16_i8_nxv1i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e8,mf8,ta,mu
+; CHECK-NEXT:    vwmaccus.vx v16, a0, v17
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i16> @llvm.riscv.vwmaccus.nxv1i16.i8(
+    <vscale x 1 x i16> %0,
+    i8 %1,
+    <vscale x 1 x i8> %2,
+    i32 %3)
+
+  ret <vscale x 1 x i16> %a
+}
+
+declare <vscale x 1 x i16> @llvm.riscv.vwmaccus.mask.nxv1i16.i8(
+  <vscale x 1 x i16>,
+  i8,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i1>,
+  i32);
+
+define <vscale x 1 x i16> @intrinsic_vwmaccus_mask_vx_nxv1i16_i8_nxv1i8(<vscale x 1 x i16> %0, i8 %1, <vscale x 1 x i8> %2, <vscale x 1 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccus_mask_vx_nxv1i16_i8_nxv1i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e8,mf8,ta,mu
+; CHECK-NEXT:    vwmaccus.vx v16, a0, v17, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i16> @llvm.riscv.vwmaccus.mask.nxv1i16.i8(
+    <vscale x 1 x i16> %0,
+    i8 %1,
+    <vscale x 1 x i8> %2,
+    <vscale x 1 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 1 x i16> %a
+}
+
+declare <vscale x 2 x i16> @llvm.riscv.vwmaccus.nxv2i16.i8(
+  <vscale x 2 x i16>,
+  i8,
+  <vscale x 2 x i8>,
+  i32);
+
+define <vscale x 2 x i16>  @intrinsic_vwmaccus_vx_nxv2i16_i8_nxv2i8(<vscale x 2 x i16> %0, i8 %1, <vscale x 2 x i8> %2, i32 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccus_vx_nxv2i16_i8_nxv2i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e8,mf4,ta,mu
+; CHECK-NEXT:    vwmaccus.vx v16, a0, v17
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i16> @llvm.riscv.vwmaccus.nxv2i16.i8(
+    <vscale x 2 x i16> %0,
+    i8 %1,
+    <vscale x 2 x i8> %2,
+    i32 %3)
+
+  ret <vscale x 2 x i16> %a
+}
+
+declare <vscale x 2 x i16> @llvm.riscv.vwmaccus.mask.nxv2i16.i8(
+  <vscale x 2 x i16>,
+  i8,
+  <vscale x 2 x i8>,
+  <vscale x 2 x i1>,
+  i32);
+
+define <vscale x 2 x i16> @intrinsic_vwmaccus_mask_vx_nxv2i16_i8_nxv2i8(<vscale x 2 x i16> %0, i8 %1, <vscale x 2 x i8> %2, <vscale x 2 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccus_mask_vx_nxv2i16_i8_nxv2i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e8,mf4,ta,mu
+; CHECK-NEXT:    vwmaccus.vx v16, a0, v17, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i16> @llvm.riscv.vwmaccus.mask.nxv2i16.i8(
+    <vscale x 2 x i16> %0,
+    i8 %1,
+    <vscale x 2 x i8> %2,
+    <vscale x 2 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 2 x i16> %a
+}
+
+declare <vscale x 4 x i16> @llvm.riscv.vwmaccus.nxv4i16.i8(
+  <vscale x 4 x i16>,
+  i8,
+  <vscale x 4 x i8>,
+  i32);
+
+define <vscale x 4 x i16>  @intrinsic_vwmaccus_vx_nxv4i16_i8_nxv4i8(<vscale x 4 x i16> %0, i8 %1, <vscale x 4 x i8> %2, i32 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccus_vx_nxv4i16_i8_nxv4i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e8,mf2,ta,mu
+; CHECK-NEXT:    vwmaccus.vx v16, a0, v17
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i16> @llvm.riscv.vwmaccus.nxv4i16.i8(
+    <vscale x 4 x i16> %0,
+    i8 %1,
+    <vscale x 4 x i8> %2,
+    i32 %3)
+
+  ret <vscale x 4 x i16> %a
+}
+
+declare <vscale x 4 x i16> @llvm.riscv.vwmaccus.mask.nxv4i16.i8(
+  <vscale x 4 x i16>,
+  i8,
+  <vscale x 4 x i8>,
+  <vscale x 4 x i1>,
+  i32);
+
+define <vscale x 4 x i16> @intrinsic_vwmaccus_mask_vx_nxv4i16_i8_nxv4i8(<vscale x 4 x i16> %0, i8 %1, <vscale x 4 x i8> %2, <vscale x 4 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccus_mask_vx_nxv4i16_i8_nxv4i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e8,mf2,ta,mu
+; CHECK-NEXT:    vwmaccus.vx v16, a0, v17, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i16> @llvm.riscv.vwmaccus.mask.nxv4i16.i8(
+    <vscale x 4 x i16> %0,
+    i8 %1,
+    <vscale x 4 x i8> %2,
+    <vscale x 4 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 4 x i16> %a
+}
+
+declare <vscale x 8 x i16> @llvm.riscv.vwmaccus.nxv8i16.i8(
+  <vscale x 8 x i16>,
+  i8,
+  <vscale x 8 x i8>,
+  i32);
+
+define <vscale x 8 x i16>  @intrinsic_vwmaccus_vx_nxv8i16_i8_nxv8i8(<vscale x 8 x i16> %0, i8 %1, <vscale x 8 x i8> %2, i32 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccus_vx_nxv8i16_i8_nxv8i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e8,m1,ta,mu
+; CHECK-NEXT:    vwmaccus.vx v16, a0, v18
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i16> @llvm.riscv.vwmaccus.nxv8i16.i8(
+    <vscale x 8 x i16> %0,
+    i8 %1,
+    <vscale x 8 x i8> %2,
+    i32 %3)
+
+  ret <vscale x 8 x i16> %a
+}
+
+declare <vscale x 8 x i16> @llvm.riscv.vwmaccus.mask.nxv8i16.i8(
+  <vscale x 8 x i16>,
+  i8,
+  <vscale x 8 x i8>,
+  <vscale x 8 x i1>,
+  i32);
+
+define <vscale x 8 x i16> @intrinsic_vwmaccus_mask_vx_nxv8i16_i8_nxv8i8(<vscale x 8 x i16> %0, i8 %1, <vscale x 8 x i8> %2, <vscale x 8 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccus_mask_vx_nxv8i16_i8_nxv8i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e8,m1,ta,mu
+; CHECK-NEXT:    vwmaccus.vx v16, a0, v18, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i16> @llvm.riscv.vwmaccus.mask.nxv8i16.i8(
+    <vscale x 8 x i16> %0,
+    i8 %1,
+    <vscale x 8 x i8> %2,
+    <vscale x 8 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 8 x i16> %a
+}
+
+declare <vscale x 16 x i16> @llvm.riscv.vwmaccus.nxv16i16.i8(
+  <vscale x 16 x i16>,
+  i8,
+  <vscale x 16 x i8>,
+  i32);
+
+define <vscale x 16 x i16>  @intrinsic_vwmaccus_vx_nxv16i16_i8_nxv16i8(<vscale x 16 x i16> %0, i8 %1, <vscale x 16 x i8> %2, i32 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccus_vx_nxv16i16_i8_nxv16i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e8,m2,ta,mu
+; CHECK-NEXT:    vwmaccus.vx v16, a0, v20
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 16 x i16> @llvm.riscv.vwmaccus.nxv16i16.i8(
+    <vscale x 16 x i16> %0,
+    i8 %1,
+    <vscale x 16 x i8> %2,
+    i32 %3)
+
+  ret <vscale x 16 x i16> %a
+}
+
+declare <vscale x 16 x i16> @llvm.riscv.vwmaccus.mask.nxv16i16.i8(
+  <vscale x 16 x i16>,
+  i8,
+  <vscale x 16 x i8>,
+  <vscale x 16 x i1>,
+  i32);
+
+define <vscale x 16 x i16> @intrinsic_vwmaccus_mask_vx_nxv16i16_i8_nxv16i8(<vscale x 16 x i16> %0, i8 %1, <vscale x 16 x i8> %2, <vscale x 16 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccus_mask_vx_nxv16i16_i8_nxv16i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e8,m2,ta,mu
+; CHECK-NEXT:    vwmaccus.vx v16, a0, v20, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 16 x i16> @llvm.riscv.vwmaccus.mask.nxv16i16.i8(
+    <vscale x 16 x i16> %0,
+    i8 %1,
+    <vscale x 16 x i8> %2,
+    <vscale x 16 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 16 x i16> %a
+}
+
+declare <vscale x 32 x i16> @llvm.riscv.vwmaccus.nxv32i16.i8(
+  <vscale x 32 x i16>,
+  i8,
+  <vscale x 32 x i8>,
+  i32);
+
+define <vscale x 32 x i16>  @intrinsic_vwmaccus_vx_nxv32i16_i8_nxv32i8(<vscale x 32 x i16> %0, i8 %1, <vscale x 32 x i8> %2, i32 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccus_vx_nxv32i16_i8_nxv32i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a3, zero, e8,m4,ta,mu
+; CHECK-NEXT:    vle8.v v28, (a1)
+; CHECK-NEXT:    vsetvli a1, a2, e8,m4,ta,mu
+; CHECK-NEXT:    vwmaccus.vx v16, a0, v28
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 32 x i16> @llvm.riscv.vwmaccus.nxv32i16.i8(
+    <vscale x 32 x i16> %0,
+    i8 %1,
+    <vscale x 32 x i8> %2,
+    i32 %3)
+
+  ret <vscale x 32 x i16> %a
+}
+
+declare <vscale x 32 x i16> @llvm.riscv.vwmaccus.mask.nxv32i16.i8(
+  <vscale x 32 x i16>,
+  i8,
+  <vscale x 32 x i8>,
+  <vscale x 32 x i1>,
+  i32);
+
+define <vscale x 32 x i16> @intrinsic_vwmaccus_mask_vx_nxv32i16_i8_nxv32i8(<vscale x 32 x i16> %0, i8 %1, <vscale x 32 x i8> %2, <vscale x 32 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccus_mask_vx_nxv32i16_i8_nxv32i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a3, zero, e8,m4,ta,mu
+; CHECK-NEXT:    vle8.v v28, (a1)
+; CHECK-NEXT:    vsetvli a1, a2, e8,m4,ta,mu
+; CHECK-NEXT:    vwmaccus.vx v16, a0, v28, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 32 x i16> @llvm.riscv.vwmaccus.mask.nxv32i16.i8(
+    <vscale x 32 x i16> %0,
+    i8 %1,
+    <vscale x 32 x i8> %2,
+    <vscale x 32 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 32 x i16> %a
+}
+
+declare <vscale x 1 x i32> @llvm.riscv.vwmaccus.nxv1i32.i16(
+  <vscale x 1 x i32>,
+  i16,
+  <vscale x 1 x i16>,
+  i32);
+
+define <vscale x 1 x i32>  @intrinsic_vwmaccus_vx_nxv1i32_i16_nxv1i16(<vscale x 1 x i32> %0, i16 %1, <vscale x 1 x i16> %2, i32 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccus_vx_nxv1i32_i16_nxv1i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e16,mf4,ta,mu
+; CHECK-NEXT:    vwmaccus.vx v16, a0, v17
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i32> @llvm.riscv.vwmaccus.nxv1i32.i16(
+    <vscale x 1 x i32> %0,
+    i16 %1,
+    <vscale x 1 x i16> %2,
+    i32 %3)
+
+  ret <vscale x 1 x i32> %a
+}
+
+declare <vscale x 1 x i32> @llvm.riscv.vwmaccus.mask.nxv1i32.i16(
+  <vscale x 1 x i32>,
+  i16,
+  <vscale x 1 x i16>,
+  <vscale x 1 x i1>,
+  i32);
+
+define <vscale x 1 x i32> @intrinsic_vwmaccus_mask_vx_nxv1i32_i16_nxv1i16(<vscale x 1 x i32> %0, i16 %1, <vscale x 1 x i16> %2, <vscale x 1 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccus_mask_vx_nxv1i32_i16_nxv1i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e16,mf4,ta,mu
+; CHECK-NEXT:    vwmaccus.vx v16, a0, v17, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i32> @llvm.riscv.vwmaccus.mask.nxv1i32.i16(
+    <vscale x 1 x i32> %0,
+    i16 %1,
+    <vscale x 1 x i16> %2,
+    <vscale x 1 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 1 x i32> %a
+}
+
+declare <vscale x 2 x i32> @llvm.riscv.vwmaccus.nxv2i32.i16(
+  <vscale x 2 x i32>,
+  i16,
+  <vscale x 2 x i16>,
+  i32);
+
+define <vscale x 2 x i32>  @intrinsic_vwmaccus_vx_nxv2i32_i16_nxv2i16(<vscale x 2 x i32> %0, i16 %1, <vscale x 2 x i16> %2, i32 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccus_vx_nxv2i32_i16_nxv2i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e16,mf2,ta,mu
+; CHECK-NEXT:    vwmaccus.vx v16, a0, v17
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i32> @llvm.riscv.vwmaccus.nxv2i32.i16(
+    <vscale x 2 x i32> %0,
+    i16 %1,
+    <vscale x 2 x i16> %2,
+    i32 %3)
+
+  ret <vscale x 2 x i32> %a
+}
+
+declare <vscale x 2 x i32> @llvm.riscv.vwmaccus.mask.nxv2i32.i16(
+  <vscale x 2 x i32>,
+  i16,
+  <vscale x 2 x i16>,
+  <vscale x 2 x i1>,
+  i32);
+
+define <vscale x 2 x i32> @intrinsic_vwmaccus_mask_vx_nxv2i32_i16_nxv2i16(<vscale x 2 x i32> %0, i16 %1, <vscale x 2 x i16> %2, <vscale x 2 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccus_mask_vx_nxv2i32_i16_nxv2i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e16,mf2,ta,mu
+; CHECK-NEXT:    vwmaccus.vx v16, a0, v17, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i32> @llvm.riscv.vwmaccus.mask.nxv2i32.i16(
+    <vscale x 2 x i32> %0,
+    i16 %1,
+    <vscale x 2 x i16> %2,
+    <vscale x 2 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 2 x i32> %a
+}
+
+declare <vscale x 4 x i32> @llvm.riscv.vwmaccus.nxv4i32.i16(
+  <vscale x 4 x i32>,
+  i16,
+  <vscale x 4 x i16>,
+  i32);
+
+define <vscale x 4 x i32>  @intrinsic_vwmaccus_vx_nxv4i32_i16_nxv4i16(<vscale x 4 x i32> %0, i16 %1, <vscale x 4 x i16> %2, i32 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccus_vx_nxv4i32_i16_nxv4i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e16,m1,ta,mu
+; CHECK-NEXT:    vwmaccus.vx v16, a0, v18
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i32> @llvm.riscv.vwmaccus.nxv4i32.i16(
+    <vscale x 4 x i32> %0,
+    i16 %1,
+    <vscale x 4 x i16> %2,
+    i32 %3)
+
+  ret <vscale x 4 x i32> %a
+}
+
+declare <vscale x 4 x i32> @llvm.riscv.vwmaccus.mask.nxv4i32.i16(
+  <vscale x 4 x i32>,
+  i16,
+  <vscale x 4 x i16>,
+  <vscale x 4 x i1>,
+  i32);
+
+define <vscale x 4 x i32> @intrinsic_vwmaccus_mask_vx_nxv4i32_i16_nxv4i16(<vscale x 4 x i32> %0, i16 %1, <vscale x 4 x i16> %2, <vscale x 4 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccus_mask_vx_nxv4i32_i16_nxv4i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e16,m1,ta,mu
+; CHECK-NEXT:    vwmaccus.vx v16, a0, v18, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i32> @llvm.riscv.vwmaccus.mask.nxv4i32.i16(
+    <vscale x 4 x i32> %0,
+    i16 %1,
+    <vscale x 4 x i16> %2,
+    <vscale x 4 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 4 x i32> %a
+}
+
+declare <vscale x 8 x i32> @llvm.riscv.vwmaccus.nxv8i32.i16(
+  <vscale x 8 x i32>,
+  i16,
+  <vscale x 8 x i16>,
+  i32);
+
+define <vscale x 8 x i32>  @intrinsic_vwmaccus_vx_nxv8i32_i16_nxv8i16(<vscale x 8 x i32> %0, i16 %1, <vscale x 8 x i16> %2, i32 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccus_vx_nxv8i32_i16_nxv8i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e16,m2,ta,mu
+; CHECK-NEXT:    vwmaccus.vx v16, a0, v20
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i32> @llvm.riscv.vwmaccus.nxv8i32.i16(
+    <vscale x 8 x i32> %0,
+    i16 %1,
+    <vscale x 8 x i16> %2,
+    i32 %3)
+
+  ret <vscale x 8 x i32> %a
+}
+
+declare <vscale x 8 x i32> @llvm.riscv.vwmaccus.mask.nxv8i32.i16(
+  <vscale x 8 x i32>,
+  i16,
+  <vscale x 8 x i16>,
+  <vscale x 8 x i1>,
+  i32);
+
+define <vscale x 8 x i32> @intrinsic_vwmaccus_mask_vx_nxv8i32_i16_nxv8i16(<vscale x 8 x i32> %0, i16 %1, <vscale x 8 x i16> %2, <vscale x 8 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccus_mask_vx_nxv8i32_i16_nxv8i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e16,m2,ta,mu
+; CHECK-NEXT:    vwmaccus.vx v16, a0, v20, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i32> @llvm.riscv.vwmaccus.mask.nxv8i32.i16(
+    <vscale x 8 x i32> %0,
+    i16 %1,
+    <vscale x 8 x i16> %2,
+    <vscale x 8 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 8 x i32> %a
+}
+
+declare <vscale x 16 x i32> @llvm.riscv.vwmaccus.nxv16i32.i16(
+  <vscale x 16 x i32>,
+  i16,
+  <vscale x 16 x i16>,
+  i32);
+
+define <vscale x 16 x i32>  @intrinsic_vwmaccus_vx_nxv16i32_i16_nxv16i16(<vscale x 16 x i32> %0, i16 %1, <vscale x 16 x i16> %2, i32 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccus_vx_nxv16i32_i16_nxv16i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a3, zero, e16,m4,ta,mu
+; CHECK-NEXT:    vle16.v v28, (a1)
+; CHECK-NEXT:    vsetvli a1, a2, e16,m4,ta,mu
+; CHECK-NEXT:    vwmaccus.vx v16, a0, v28
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 16 x i32> @llvm.riscv.vwmaccus.nxv16i32.i16(
+    <vscale x 16 x i32> %0,
+    i16 %1,
+    <vscale x 16 x i16> %2,
+    i32 %3)
+
+  ret <vscale x 16 x i32> %a
+}
+
+declare <vscale x 16 x i32> @llvm.riscv.vwmaccus.mask.nxv16i32.i16(
+  <vscale x 16 x i32>,
+  i16,
+  <vscale x 16 x i16>,
+  <vscale x 16 x i1>,
+  i32);
+
+define <vscale x 16 x i32> @intrinsic_vwmaccus_mask_vx_nxv16i32_i16_nxv16i16(<vscale x 16 x i32> %0, i16 %1, <vscale x 16 x i16> %2, <vscale x 16 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccus_mask_vx_nxv16i32_i16_nxv16i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a3, zero, e16,m4,ta,mu
+; CHECK-NEXT:    vle16.v v28, (a1)
+; CHECK-NEXT:    vsetvli a1, a2, e16,m4,ta,mu
+; CHECK-NEXT:    vwmaccus.vx v16, a0, v28, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 16 x i32> @llvm.riscv.vwmaccus.mask.nxv16i32.i16(
+    <vscale x 16 x i32> %0,
+    i16 %1,
+    <vscale x 16 x i16> %2,
+    <vscale x 16 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 16 x i32> %a
+}

llvm/test/CodeGen/RISCV/rvv/vwmaccus-rv64.ll

+; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
+; RUN: llc -mtriple=riscv64 -mattr=+experimental-v,+d -verify-machineinstrs \
+; RUN:   --riscv-no-aliases < %s | FileCheck %s
+declare <vscale x 1 x i16> @llvm.riscv.vwmaccus.nxv1i16.i8(
+  <vscale x 1 x i16>,
+  i8,
+  <vscale x 1 x i8>,
+  i64);
+
+define <vscale x 1 x i16>  @intrinsic_vwmaccus_vx_nxv1i16_i8_nxv1i8(<vscale x 1 x i16> %0, i8 %1, <vscale x 1 x i8> %2, i64 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccus_vx_nxv1i16_i8_nxv1i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e8,mf8,ta,mu
+; CHECK-NEXT:    vwmaccus.vx v16, a0, v17
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i16> @llvm.riscv.vwmaccus.nxv1i16.i8(
+    <vscale x 1 x i16> %0,
+    i8 %1,
+    <vscale x 1 x i8> %2,
+    i64 %3)
+
+  ret <vscale x 1 x i16> %a
+}
+
+declare <vscale x 1 x i16> @llvm.riscv.vwmaccus.mask.nxv1i16.i8(
+  <vscale x 1 x i16>,
+  i8,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i1>,
+  i64);
+
+define <vscale x 1 x i16> @intrinsic_vwmaccus_mask_vx_nxv1i16_i8_nxv1i8(<vscale x 1 x i16> %0, i8 %1, <vscale x 1 x i8> %2, <vscale x 1 x i1> %3, i64 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccus_mask_vx_nxv1i16_i8_nxv1i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e8,mf8,ta,mu
+; CHECK-NEXT:    vwmaccus.vx v16, a0, v17, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i16> @llvm.riscv.vwmaccus.mask.nxv1i16.i8(
+    <vscale x 1 x i16> %0,
+    i8 %1,
+    <vscale x 1 x i8> %2,
+    <vscale x 1 x i1> %3,
+    i64 %4)
+
+  ret <vscale x 1 x i16> %a
+}
+
+declare <vscale x 2 x i16> @llvm.riscv.vwmaccus.nxv2i16.i8(
+  <vscale x 2 x i16>,
+  i8,
+  <vscale x 2 x i8>,
+  i64);
+
+define <vscale x 2 x i16>  @intrinsic_vwmaccus_vx_nxv2i16_i8_nxv2i8(<vscale x 2 x i16> %0, i8 %1, <vscale x 2 x i8> %2, i64 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccus_vx_nxv2i16_i8_nxv2i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e8,mf4,ta,mu
+; CHECK-NEXT:    vwmaccus.vx v16, a0, v17
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i16> @llvm.riscv.vwmaccus.nxv2i16.i8(
+    <vscale x 2 x i16> %0,
+    i8 %1,
+    <vscale x 2 x i8> %2,
+    i64 %3)
+
+  ret <vscale x 2 x i16> %a
+}
+
+declare <vscale x 2 x i16> @llvm.riscv.vwmaccus.mask.nxv2i16.i8(
+  <vscale x 2 x i16>,
+  i8,
+  <vscale x 2 x i8>,
+  <vscale x 2 x i1>,
+  i64);
+
+define <vscale x 2 x i16> @intrinsic_vwmaccus_mask_vx_nxv2i16_i8_nxv2i8(<vscale x 2 x i16> %0, i8 %1, <vscale x 2 x i8> %2, <vscale x 2 x i1> %3, i64 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccus_mask_vx_nxv2i16_i8_nxv2i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e8,mf4,ta,mu
+; CHECK-NEXT:    vwmaccus.vx v16, a0, v17, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i16> @llvm.riscv.vwmaccus.mask.nxv2i16.i8(
+    <vscale x 2 x i16> %0,
+    i8 %1,
+    <vscale x 2 x i8> %2,
+    <vscale x 2 x i1> %3,
+    i64 %4)
+
+  ret <vscale x 2 x i16> %a
+}
+
+declare <vscale x 4 x i16> @llvm.riscv.vwmaccus.nxv4i16.i8(
+  <vscale x 4 x i16>,
+  i8,
+  <vscale x 4 x i8>,
+  i64);
+
+define <vscale x 4 x i16>  @intrinsic_vwmaccus_vx_nxv4i16_i8_nxv4i8(<vscale x 4 x i16> %0, i8 %1, <vscale x 4 x i8> %2, i64 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccus_vx_nxv4i16_i8_nxv4i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e8,mf2,ta,mu
+; CHECK-NEXT:    vwmaccus.vx v16, a0, v17
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i16> @llvm.riscv.vwmaccus.nxv4i16.i8(
+    <vscale x 4 x i16> %0,
+    i8 %1,
+    <vscale x 4 x i8> %2,
+    i64 %3)
+
+  ret <vscale x 4 x i16> %a
+}
+
+declare <vscale x 4 x i16> @llvm.riscv.vwmaccus.mask.nxv4i16.i8(
+  <vscale x 4 x i16>,
+  i8,
+  <vscale x 4 x i8>,
+  <vscale x 4 x i1>,
+  i64);
+
+define <vscale x 4 x i16> @intrinsic_vwmaccus_mask_vx_nxv4i16_i8_nxv4i8(<vscale x 4 x i16> %0, i8 %1, <vscale x 4 x i8> %2, <vscale x 4 x i1> %3, i64 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccus_mask_vx_nxv4i16_i8_nxv4i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e8,mf2,ta,mu
+; CHECK-NEXT:    vwmaccus.vx v16, a0, v17, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i16> @llvm.riscv.vwmaccus.mask.nxv4i16.i8(
+    <vscale x 4 x i16> %0,
+    i8 %1,
+    <vscale x 4 x i8> %2,
+    <vscale x 4 x i1> %3,
+    i64 %4)
+
+  ret <vscale x 4 x i16> %a
+}
+
+declare <vscale x 8 x i16> @llvm.riscv.vwmaccus.nxv8i16.i8(
+  <vscale x 8 x i16>,
+  i8,
+  <vscale x 8 x i8>,
+  i64);
+
+define <vscale x 8 x i16>  @intrinsic_vwmaccus_vx_nxv8i16_i8_nxv8i8(<vscale x 8 x i16> %0, i8 %1, <vscale x 8 x i8> %2, i64 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccus_vx_nxv8i16_i8_nxv8i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e8,m1,ta,mu
+; CHECK-NEXT:    vwmaccus.vx v16, a0, v18
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i16> @llvm.riscv.vwmaccus.nxv8i16.i8(
+    <vscale x 8 x i16> %0,
+    i8 %1,
+    <vscale x 8 x i8> %2,
+    i64 %3)
+
+  ret <vscale x 8 x i16> %a
+}
+
+declare <vscale x 8 x i16> @llvm.riscv.vwmaccus.mask.nxv8i16.i8(
+  <vscale x 8 x i16>,
+  i8,
+  <vscale x 8 x i8>,
+  <vscale x 8 x i1>,
+  i64);
+
+define <vscale x 8 x i16> @intrinsic_vwmaccus_mask_vx_nxv8i16_i8_nxv8i8(<vscale x 8 x i16> %0, i8 %1, <vscale x 8 x i8> %2, <vscale x 8 x i1> %3, i64 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccus_mask_vx_nxv8i16_i8_nxv8i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e8,m1,ta,mu
+; CHECK-NEXT:    vwmaccus.vx v16, a0, v18, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i16> @llvm.riscv.vwmaccus.mask.nxv8i16.i8(
+    <vscale x 8 x i16> %0,
+    i8 %1,
+    <vscale x 8 x i8> %2,
+    <vscale x 8 x i1> %3,
+    i64 %4)
+
+  ret <vscale x 8 x i16> %a
+}
+
+declare <vscale x 16 x i16> @llvm.riscv.vwmaccus.nxv16i16.i8(
+  <vscale x 16 x i16>,
+  i8,
+  <vscale x 16 x i8>,
+  i64);
+
+define <vscale x 16 x i16>  @intrinsic_vwmaccus_vx_nxv16i16_i8_nxv16i8(<vscale x 16 x i16> %0, i8 %1, <vscale x 16 x i8> %2, i64 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccus_vx_nxv16i16_i8_nxv16i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e8,m2,ta,mu
+; CHECK-NEXT:    vwmaccus.vx v16, a0, v20
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 16 x i16> @llvm.riscv.vwmaccus.nxv16i16.i8(
+    <vscale x 16 x i16> %0,
+    i8 %1,
+    <vscale x 16 x i8> %2,
+    i64 %3)
+
+  ret <vscale x 16 x i16> %a
+}
+
+declare <vscale x 16 x i16> @llvm.riscv.vwmaccus.mask.nxv16i16.i8(
+  <vscale x 16 x i16>,
+  i8,
+  <vscale x 16 x i8>,
+  <vscale x 16 x i1>,
+  i64);
+
+define <vscale x 16 x i16> @intrinsic_vwmaccus_mask_vx_nxv16i16_i8_nxv16i8(<vscale x 16 x i16> %0, i8 %1, <vscale x 16 x i8> %2, <vscale x 16 x i1> %3, i64 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccus_mask_vx_nxv16i16_i8_nxv16i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e8,m2,ta,mu
+; CHECK-NEXT:    vwmaccus.vx v16, a0, v20, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 16 x i16> @llvm.riscv.vwmaccus.mask.nxv16i16.i8(
+    <vscale x 16 x i16> %0,
+    i8 %1,
+    <vscale x 16 x i8> %2,
+    <vscale x 16 x i1> %3,
+    i64 %4)
+
+  ret <vscale x 16 x i16> %a
+}
+
+declare <vscale x 32 x i16> @llvm.riscv.vwmaccus.nxv32i16.i8(
+  <vscale x 32 x i16>,
+  i8,
+  <vscale x 32 x i8>,
+  i64);
+
+define <vscale x 32 x i16>  @intrinsic_vwmaccus_vx_nxv32i16_i8_nxv32i8(<vscale x 32 x i16> %0, i8 %1, <vscale x 32 x i8> %2, i64 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccus_vx_nxv32i16_i8_nxv32i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a3, zero, e8,m4,ta,mu
+; CHECK-NEXT:    vle8.v v28, (a1)
+; CHECK-NEXT:    vsetvli a1, a2, e8,m4,ta,mu
+; CHECK-NEXT:    vwmaccus.vx v16, a0, v28
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 32 x i16> @llvm.riscv.vwmaccus.nxv32i16.i8(
+    <vscale x 32 x i16> %0,
+    i8 %1,
+    <vscale x 32 x i8> %2,
+    i64 %3)
+
+  ret <vscale x 32 x i16> %a
+}
+
+declare <vscale x 32 x i16> @llvm.riscv.vwmaccus.mask.nxv32i16.i8(
+  <vscale x 32 x i16>,
+  i8,
+  <vscale x 32 x i8>,
+  <vscale x 32 x i1>,
+  i64);
+
+define <vscale x 32 x i16> @intrinsic_vwmaccus_mask_vx_nxv32i16_i8_nxv32i8(<vscale x 32 x i16> %0, i8 %1, <vscale x 32 x i8> %2, <vscale x 32 x i1> %3, i64 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccus_mask_vx_nxv32i16_i8_nxv32i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a3, zero, e8,m4,ta,mu
+; CHECK-NEXT:    vle8.v v28, (a1)
+; CHECK-NEXT:    vsetvli a1, a2, e8,m4,ta,mu
+; CHECK-NEXT:    vwmaccus.vx v16, a0, v28, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 32 x i16> @llvm.riscv.vwmaccus.mask.nxv32i16.i8(
+    <vscale x 32 x i16> %0,
+    i8 %1,
+    <vscale x 32 x i8> %2,
+    <vscale x 32 x i1> %3,
+    i64 %4)
+
+  ret <vscale x 32 x i16> %a
+}
+
+declare <vscale x 1 x i32> @llvm.riscv.vwmaccus.nxv1i32.i16(
+  <vscale x 1 x i32>,
+  i16,
+  <vscale x 1 x i16>,
+  i64);
+
+define <vscale x 1 x i32>  @intrinsic_vwmaccus_vx_nxv1i32_i16_nxv1i16(<vscale x 1 x i32> %0, i16 %1, <vscale x 1 x i16> %2, i64 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccus_vx_nxv1i32_i16_nxv1i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e16,mf4,ta,mu
+; CHECK-NEXT:    vwmaccus.vx v16, a0, v17
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i32> @llvm.riscv.vwmaccus.nxv1i32.i16(
+    <vscale x 1 x i32> %0,
+    i16 %1,
+    <vscale x 1 x i16> %2,
+    i64 %3)
+
+  ret <vscale x 1 x i32> %a
+}
+
+declare <vscale x 1 x i32> @llvm.riscv.vwmaccus.mask.nxv1i32.i16(
+  <vscale x 1 x i32>,
+  i16,
+  <vscale x 1 x i16>,
+  <vscale x 1 x i1>,
+  i64);
+
+define <vscale x 1 x i32> @intrinsic_vwmaccus_mask_vx_nxv1i32_i16_nxv1i16(<vscale x 1 x i32> %0, i16 %1, <vscale x 1 x i16> %2, <vscale x 1 x i1> %3, i64 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccus_mask_vx_nxv1i32_i16_nxv1i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e16,mf4,ta,mu
+; CHECK-NEXT:    vwmaccus.vx v16, a0, v17, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i32> @llvm.riscv.vwmaccus.mask.nxv1i32.i16(
+    <vscale x 1 x i32> %0,
+    i16 %1,
+    <vscale x 1 x i16> %2,
+    <vscale x 1 x i1> %3,
+    i64 %4)
+
+  ret <vscale x 1 x i32> %a
+}
+
+declare <vscale x 2 x i32> @llvm.riscv.vwmaccus.nxv2i32.i16(
+  <vscale x 2 x i32>,
+  i16,
+  <vscale x 2 x i16>,
+  i64);
+
+define <vscale x 2 x i32>  @intrinsic_vwmaccus_vx_nxv2i32_i16_nxv2i16(<vscale x 2 x i32> %0, i16 %1, <vscale x 2 x i16> %2, i64 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccus_vx_nxv2i32_i16_nxv2i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e16,mf2,ta,mu
+; CHECK-NEXT:    vwmaccus.vx v16, a0, v17
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i32> @llvm.riscv.vwmaccus.nxv2i32.i16(
+    <vscale x 2 x i32> %0,
+    i16 %1,
+    <vscale x 2 x i16> %2,
+    i64 %3)
+
+  ret <vscale x 2 x i32> %a
+}
+
+declare <vscale x 2 x i32> @llvm.riscv.vwmaccus.mask.nxv2i32.i16(
+  <vscale x 2 x i32>,
+  i16,
+  <vscale x 2 x i16>,
+  <vscale x 2 x i1>,
+  i64);
+
+define <vscale x 2 x i32> @intrinsic_vwmaccus_mask_vx_nxv2i32_i16_nxv2i16(<vscale x 2 x i32> %0, i16 %1, <vscale x 2 x i16> %2, <vscale x 2 x i1> %3, i64 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccus_mask_vx_nxv2i32_i16_nxv2i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e16,mf2,ta,mu
+; CHECK-NEXT:    vwmaccus.vx v16, a0, v17, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i32> @llvm.riscv.vwmaccus.mask.nxv2i32.i16(
+    <vscale x 2 x i32> %0,
+    i16 %1,
+    <vscale x 2 x i16> %2,
+    <vscale x 2 x i1> %3,
+    i64 %4)
+
+  ret <vscale x 2 x i32> %a
+}
+
+declare <vscale x 4 x i32> @llvm.riscv.vwmaccus.nxv4i32.i16(
+  <vscale x 4 x i32>,
+  i16,
+  <vscale x 4 x i16>,
+  i64);
+
+define <vscale x 4 x i32>  @intrinsic_vwmaccus_vx_nxv4i32_i16_nxv4i16(<vscale x 4 x i32> %0, i16 %1, <vscale x 4 x i16> %2, i64 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccus_vx_nxv4i32_i16_nxv4i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e16,m1,ta,mu
+; CHECK-NEXT:    vwmaccus.vx v16, a0, v18
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i32> @llvm.riscv.vwmaccus.nxv4i32.i16(
+    <vscale x 4 x i32> %0,
+    i16 %1,
+    <vscale x 4 x i16> %2,
+    i64 %3)
+
+  ret <vscale x 4 x i32> %a
+}
+
+declare <vscale x 4 x i32> @llvm.riscv.vwmaccus.mask.nxv4i32.i16(
+  <vscale x 4 x i32>,
+  i16,
+  <vscale x 4 x i16>,
+  <vscale x 4 x i1>,
+  i64);
+
+define <vscale x 4 x i32> @intrinsic_vwmaccus_mask_vx_nxv4i32_i16_nxv4i16(<vscale x 4 x i32> %0, i16 %1, <vscale x 4 x i16> %2, <vscale x 4 x i1> %3, i64 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccus_mask_vx_nxv4i32_i16_nxv4i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e16,m1,ta,mu
+; CHECK-NEXT:    vwmaccus.vx v16, a0, v18, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i32> @llvm.riscv.vwmaccus.mask.nxv4i32.i16(
+    <vscale x 4 x i32> %0,
+    i16 %1,
+    <vscale x 4 x i16> %2,
+    <vscale x 4 x i1> %3,
+    i64 %4)
+
+  ret <vscale x 4 x i32> %a
+}
+
+declare <vscale x 8 x i32> @llvm.riscv.vwmaccus.nxv8i32.i16(
+  <vscale x 8 x i32>,
+  i16,
+  <vscale x 8 x i16>,
+  i64);
+
+define <vscale x 8 x i32>  @intrinsic_vwmaccus_vx_nxv8i32_i16_nxv8i16(<vscale x 8 x i32> %0, i16 %1, <vscale x 8 x i16> %2, i64 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccus_vx_nxv8i32_i16_nxv8i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e16,m2,ta,mu
+; CHECK-NEXT:    vwmaccus.vx v16, a0, v20
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i32> @llvm.riscv.vwmaccus.nxv8i32.i16(
+    <vscale x 8 x i32> %0,
+    i16 %1,
+    <vscale x 8 x i16> %2,
+    i64 %3)
+
+  ret <vscale x 8 x i32> %a
+}
+
+declare <vscale x 8 x i32> @llvm.riscv.vwmaccus.mask.nxv8i32.i16(
+  <vscale x 8 x i32>,
+  i16,
+  <vscale x 8 x i16>,
+  <vscale x 8 x i1>,
+  i64);
+
+define <vscale x 8 x i32> @intrinsic_vwmaccus_mask_vx_nxv8i32_i16_nxv8i16(<vscale x 8 x i32> %0, i16 %1, <vscale x 8 x i16> %2, <vscale x 8 x i1> %3, i64 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccus_mask_vx_nxv8i32_i16_nxv8i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e16,m2,ta,mu
+; CHECK-NEXT:    vwmaccus.vx v16, a0, v20, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i32> @llvm.riscv.vwmaccus.mask.nxv8i32.i16(
+    <vscale x 8 x i32> %0,
+    i16 %1,
+    <vscale x 8 x i16> %2,
+    <vscale x 8 x i1> %3,
+    i64 %4)
+
+  ret <vscale x 8 x i32> %a
+}
+
+declare <vscale x 16 x i32> @llvm.riscv.vwmaccus.nxv16i32.i16(
+  <vscale x 16 x i32>,
+  i16,
+  <vscale x 16 x i16>,
+  i64);
+
+define <vscale x 16 x i32>  @intrinsic_vwmaccus_vx_nxv16i32_i16_nxv16i16(<vscale x 16 x i32> %0, i16 %1, <vscale x 16 x i16> %2, i64 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccus_vx_nxv16i32_i16_nxv16i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a3, zero, e16,m4,ta,mu
+; CHECK-NEXT:    vle16.v v28, (a1)
+; CHECK-NEXT:    vsetvli a1, a2, e16,m4,ta,mu
+; CHECK-NEXT:    vwmaccus.vx v16, a0, v28
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 16 x i32> @llvm.riscv.vwmaccus.nxv16i32.i16(
+    <vscale x 16 x i32> %0,
+    i16 %1,
+    <vscale x 16 x i16> %2,
+    i64 %3)
+
+  ret <vscale x 16 x i32> %a
+}
+
+declare <vscale x 16 x i32> @llvm.riscv.vwmaccus.mask.nxv16i32.i16(
+  <vscale x 16 x i32>,
+  i16,
+  <vscale x 16 x i16>,
+  <vscale x 16 x i1>,
+  i64);
+
+define <vscale x 16 x i32> @intrinsic_vwmaccus_mask_vx_nxv16i32_i16_nxv16i16(<vscale x 16 x i32> %0, i16 %1, <vscale x 16 x i16> %2, <vscale x 16 x i1> %3, i64 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccus_mask_vx_nxv16i32_i16_nxv16i16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a3, zero, e16,m4,ta,mu
+; CHECK-NEXT:    vle16.v v28, (a1)
+; CHECK-NEXT:    vsetvli a1, a2, e16,m4,ta,mu
+; CHECK-NEXT:    vwmaccus.vx v16, a0, v28, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 16 x i32> @llvm.riscv.vwmaccus.mask.nxv16i32.i16(
+    <vscale x 16 x i32> %0,
+    i16 %1,
+    <vscale x 16 x i16> %2,
+    <vscale x 16 x i1> %3,
+    i64 %4)
+
+  ret <vscale x 16 x i32> %a
+}
+
+declare <vscale x 1 x i64> @llvm.riscv.vwmaccus.nxv1i64.i32(
+  <vscale x 1 x i64>,
+  i32,
+  <vscale x 1 x i32>,
+  i64);
+
+define <vscale x 1 x i64>  @intrinsic_vwmaccus_vx_nxv1i64_i32_nxv1i32(<vscale x 1 x i64> %0, i32 %1, <vscale x 1 x i32> %2, i64 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccus_vx_nxv1i64_i32_nxv1i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e32,mf2,ta,mu
+; CHECK-NEXT:    vwmaccus.vx v16, a0, v17
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i64> @llvm.riscv.vwmaccus.nxv1i64.i32(
+    <vscale x 1 x i64> %0,
+    i32 %1,
+    <vscale x 1 x i32> %2,
+    i64 %3)
+
+  ret <vscale x 1 x i64> %a
+}
+
+declare <vscale x 1 x i64> @llvm.riscv.vwmaccus.mask.nxv1i64.i32(
+  <vscale x 1 x i64>,
+  i32,
+  <vscale x 1 x i32>,
+  <vscale x 1 x i1>,
+  i64);
+
+define <vscale x 1 x i64> @intrinsic_vwmaccus_mask_vx_nxv1i64_i32_nxv1i32(<vscale x 1 x i64> %0, i32 %1, <vscale x 1 x i32> %2, <vscale x 1 x i1> %3, i64 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccus_mask_vx_nxv1i64_i32_nxv1i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e32,mf2,ta,mu
+; CHECK-NEXT:    vwmaccus.vx v16, a0, v17, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i64> @llvm.riscv.vwmaccus.mask.nxv1i64.i32(
+    <vscale x 1 x i64> %0,
+    i32 %1,
+    <vscale x 1 x i32> %2,
+    <vscale x 1 x i1> %3,
+    i64 %4)
+
+  ret <vscale x 1 x i64> %a
+}
+
+declare <vscale x 2 x i64> @llvm.riscv.vwmaccus.nxv2i64.i32(
+  <vscale x 2 x i64>,
+  i32,
+  <vscale x 2 x i32>,
+  i64);
+
+define <vscale x 2 x i64>  @intrinsic_vwmaccus_vx_nxv2i64_i32_nxv2i32(<vscale x 2 x i64> %0, i32 %1, <vscale x 2 x i32> %2, i64 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccus_vx_nxv2i64_i32_nxv2i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e32,m1,ta,mu
+; CHECK-NEXT:    vwmaccus.vx v16, a0, v18
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i64> @llvm.riscv.vwmaccus.nxv2i64.i32(
+    <vscale x 2 x i64> %0,
+    i32 %1,
+    <vscale x 2 x i32> %2,
+    i64 %3)
+
+  ret <vscale x 2 x i64> %a
+}
+
+declare <vscale x 2 x i64> @llvm.riscv.vwmaccus.mask.nxv2i64.i32(
+  <vscale x 2 x i64>,
+  i32,
+  <vscale x 2 x i32>,
+  <vscale x 2 x i1>,
+  i64);
+
+define <vscale x 2 x i64> @intrinsic_vwmaccus_mask_vx_nxv2i64_i32_nxv2i32(<vscale x 2 x i64> %0, i32 %1, <vscale x 2 x i32> %2, <vscale x 2 x i1> %3, i64 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccus_mask_vx_nxv2i64_i32_nxv2i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e32,m1,ta,mu
+; CHECK-NEXT:    vwmaccus.vx v16, a0, v18, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i64> @llvm.riscv.vwmaccus.mask.nxv2i64.i32(
+    <vscale x 2 x i64> %0,
+    i32 %1,
+    <vscale x 2 x i32> %2,
+    <vscale x 2 x i1> %3,
+    i64 %4)
+
+  ret <vscale x 2 x i64> %a
+}
+
+declare <vscale x 4 x i64> @llvm.riscv.vwmaccus.nxv4i64.i32(
+  <vscale x 4 x i64>,
+  i32,
+  <vscale x 4 x i32>,
+  i64);
+
+define <vscale x 4 x i64>  @intrinsic_vwmaccus_vx_nxv4i64_i32_nxv4i32(<vscale x 4 x i64> %0, i32 %1, <vscale x 4 x i32> %2, i64 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccus_vx_nxv4i64_i32_nxv4i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e32,m2,ta,mu
+; CHECK-NEXT:    vwmaccus.vx v16, a0, v20
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i64> @llvm.riscv.vwmaccus.nxv4i64.i32(
+    <vscale x 4 x i64> %0,
+    i32 %1,
+    <vscale x 4 x i32> %2,
+    i64 %3)
+
+  ret <vscale x 4 x i64> %a
+}
+
+declare <vscale x 4 x i64> @llvm.riscv.vwmaccus.mask.nxv4i64.i32(
+  <vscale x 4 x i64>,
+  i32,
+  <vscale x 4 x i32>,
+  <vscale x 4 x i1>,
+  i64);
+
+define <vscale x 4 x i64> @intrinsic_vwmaccus_mask_vx_nxv4i64_i32_nxv4i32(<vscale x 4 x i64> %0, i32 %1, <vscale x 4 x i32> %2, <vscale x 4 x i1> %3, i64 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccus_mask_vx_nxv4i64_i32_nxv4i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e32,m2,ta,mu
+; CHECK-NEXT:    vwmaccus.vx v16, a0, v20, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i64> @llvm.riscv.vwmaccus.mask.nxv4i64.i32(
+    <vscale x 4 x i64> %0,
+    i32 %1,
+    <vscale x 4 x i32> %2,
+    <vscale x 4 x i1> %3,
+    i64 %4)
+
+  ret <vscale x 4 x i64> %a
+}
+
+declare <vscale x 8 x i64> @llvm.riscv.vwmaccus.nxv8i64.i32(
+  <vscale x 8 x i64>,
+  i32,
+  <vscale x 8 x i32>,
+  i64);
+
+define <vscale x 8 x i64>  @intrinsic_vwmaccus_vx_nxv8i64_i32_nxv8i32(<vscale x 8 x i64> %0, i32 %1, <vscale x 8 x i32> %2, i64 %3) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccus_vx_nxv8i64_i32_nxv8i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a3, zero, e32,m4,ta,mu
+; CHECK-NEXT:    vle32.v v28, (a1)
+; CHECK-NEXT:    vsetvli a1, a2, e32,m4,ta,mu
+; CHECK-NEXT:    vwmaccus.vx v16, a0, v28
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i64> @llvm.riscv.vwmaccus.nxv8i64.i32(
+    <vscale x 8 x i64> %0,
+    i32 %1,
+    <vscale x 8 x i32> %2,
+    i64 %3)
+
+  ret <vscale x 8 x i64> %a
+}
+
+declare <vscale x 8 x i64> @llvm.riscv.vwmaccus.mask.nxv8i64.i32(
+  <vscale x 8 x i64>,
+  i32,
+  <vscale x 8 x i32>,
+  <vscale x 8 x i1>,
+  i64);
+
+define <vscale x 8 x i64> @intrinsic_vwmaccus_mask_vx_nxv8i64_i32_nxv8i32(<vscale x 8 x i64> %0, i32 %1, <vscale x 8 x i32> %2, <vscale x 8 x i1> %3, i64 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwmaccus_mask_vx_nxv8i64_i32_nxv8i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a3, zero, e32,m4,ta,mu
+; CHECK-NEXT:    vle32.v v28, (a1)
+; CHECK-NEXT:    vsetvli a1, a2, e32,m4,ta,mu
+; CHECK-NEXT:    vwmaccus.vx v16, a0, v28, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i64> @llvm.riscv.vwmaccus.mask.nxv8i64.i32(
+    <vscale x 8 x i64> %0,
+    i32 %1,
+    <vscale x 8 x i32> %2,
+    <vscale x 8 x i1> %3,
+    i64 %4)
+
+  ret <vscale x 8 x i64> %a
+}