[RISCV][Clang] Add RVV vnsra, vnsrl and vwmul intrinsic functions.

[craig.topper]

LGTM

clang/include/clang/Basic/riscv_vector.td

foreach s_p = suffixes_prototypes in {
         defvar suffix = s_p[1];
         defvar prototype = s_p[2];
         def : RVVBuiltin<suffix, prototype, type_range>;
       }
     }
   }
 }
 
+// IntrinsicTypes is output, op0, op1 [-1, 0, 1]
+multiclass RVVOutOp0Op1BuiltinSet<string intrinsic_name, string type_range,
+                                  list<list<string>> suffixes_prototypes> {
+  defm NAME : RVVBuiltinSet<intrinsic_name, type_range, suffixes_prototypes,
+                            [-1, 0, 1]>;
+}
+
 // IntrinsicTypes is output, op1 [-1, 1]
 multiclass RVVOutOp1BuiltinSet<string intrinsic_name, string type_range,
                                list<list<string>> suffixes_prototypes> {
   defm NAME
       : RVVBuiltinSet<intrinsic_name, type_range, suffixes_prototypes, [-1, 1]>;
 }
 
 defvar TypeList = ["c","s","i","l","f","d"];

 defm vsrl : RVVOutOp1BuiltinSet<"vsrl", "csil",
                                 [["vv", "Uv", "UvUvUv"],
                                  ["vx", "Uv", "UvUvz"]]>;
 defm vsra : RVVOutOp1BuiltinSet<"vsra", "csil",
                                 [["vv", "v", "vvUv"],
                                  ["vx", "v", "vvz"]]>;
 
 // 12.7. Vector Narrowing Integer Right Shift Instructions
-// TODO
+let Log2LMUL = [-3, -2, -1, 0, 1, 2] in {
+defm vnsrl : RVVOutOp0Op1BuiltinSet<"vnsrl", "csi",
+                                    [["wv", "Uv", "UvUwUv"],
+                                     ["wx", "Uv", "UvUwz"]]>;
+defm vnsra : RVVOutOp0Op1BuiltinSet<"vnsra", "csi",
+                                    [["wv", "v", "vwUv"],
+                                     ["wx", "v", "vwz"]]>;
+}
 
 // 12.8. Vector Integer Comparison Instructions
 // TODO
 
 // 12.9. Vector Integer Min/Max Instructions
 defm vminu : RVVOutOp1BuiltinSet<"vminu", "csil",
                                  [["vv", "Uv", "UvUvUv"],
                                   ["vx", "Uv", "UvUvUe"]]>;

 defm vremu : RVVOutOp1BuiltinSet<"vremu", "csil",
                                  [["vv", "Uv", "UvUvUv"],
                                   ["vx", "Uv", "UvUvUe"]]>;
 defm vrem : RVVOutOp1BuiltinSet<"vrem", "csil",
                                 [["vv", "v", "vvv"],
                                  ["vx", "v", "vve"]]>;
 
 // 12.12. Vector Widening Integer Multiply Instructions
-// TODO
+let Log2LMUL = [-3, -2, -1, 0, 1, 2] in {
+defm vwmul : RVVOutOp0Op1BuiltinSet<"vwmul", "csi",
+                                    [["vv", "w", "wvv"],
+                                     ["vx", "w", "wve"]]>;
+defm vwmulu : RVVOutOp0Op1BuiltinSet<"vwmulu", "csi",
+                                     [["vv", "Uw", "UwUvUv"],
+                                      ["vx", "Uw", "UwUvUe"]]>;
+defm vwmulsu : RVVOutOp0Op1BuiltinSet<"vwmulsu", "csi",
+                                      [["vv", "w", "wvUv"],
+                                       ["vx", "w", "wvUe"]]>;
+}
 
 // 12.13. Vector Single-Width Integer Multiply-Add Instructions
 let HasMaskedOffOperand = false in {
 defm vmacc : RVVOutOp1BuiltinSet<"vmacc", "csil",
                                  [["vv", "v", "vvvv"],
                                   ["vx", "v", "vvev"],
                                   ["vv", "Uv", "UvUvUvUv"],
                                   ["vx", "Uv", "UvUvUeUv"]]>;

clang/test/CodeGen/RISCV/rvv-intrinsics-overloaded/vnsra.c

+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv32 -target-feature +experimental-v -disable-O0-optnone -emit-llvm %s -o - | opt -S -mem2reg | FileCheck --check-prefix=CHECK-RV32 %s
+// RUN: %clang_cc1 -triple riscv64 -target-feature +experimental-v -disable-O0-optnone -emit-llvm %s -o - | opt -S -mem2reg | FileCheck --check-prefix=CHECK-RV64 %s
+// RUN: %clang_cc1 -triple riscv64 -target-feature +m -target-feature +experimental-v -Werror -Wall -o - %s -S >/dev/null 2>&1 | FileCheck --check-prefix=ASM --allow-empty %s
+
+// ASM-NOT: warning
+#include <riscv_vector.h>
+
+// CHECK-RV32-LABEL: @test_vnsra_wv_i8mf8(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i8> @llvm.riscv.vnsra.nxv1i8.nxv1i16.nxv1i8.i32(<vscale x 1 x i16> [[OP1:%.*]], <vscale x 1 x i8> [[SHIFT:%.*]], i32 [[VL:%.*]]) [[ATTR7:#.*]]
+// CHECK-RV32-NEXT:    ret <vscale x 1 x i8> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsra_wv_i8mf8(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i8> @llvm.riscv.vnsra.nxv1i8.nxv1i16.nxv1i8.i64(<vscale x 1 x i16> [[OP1:%.*]], <vscale x 1 x i8> [[SHIFT:%.*]], i64 [[VL:%.*]]) [[ATTR7:#.*]]
+// CHECK-RV64-NEXT:    ret <vscale x 1 x i8> [[TMP0]]
+//
+vint8mf8_t test_vnsra_wv_i8mf8(vint16mf4_t op1, vuint8mf8_t shift, size_t vl) {
+  return vnsra(op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsra_wx_i8mf8(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i8> @llvm.riscv.vnsra.nxv1i8.nxv1i16.i32.i32(<vscale x 1 x i16> [[OP1:%.*]], i32 [[SHIFT:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 1 x i8> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsra_wx_i8mf8(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i8> @llvm.riscv.vnsra.nxv1i8.nxv1i16.i64.i64(<vscale x 1 x i16> [[OP1:%.*]], i64 [[SHIFT:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 1 x i8> [[TMP0]]
+//
+vint8mf8_t test_vnsra_wx_i8mf8(vint16mf4_t op1, size_t shift, size_t vl) {
+  return vnsra(op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsra_wv_i8mf4(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i8> @llvm.riscv.vnsra.nxv2i8.nxv2i16.nxv2i8.i32(<vscale x 2 x i16> [[OP1:%.*]], <vscale x 2 x i8> [[SHIFT:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 2 x i8> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsra_wv_i8mf4(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i8> @llvm.riscv.vnsra.nxv2i8.nxv2i16.nxv2i8.i64(<vscale x 2 x i16> [[OP1:%.*]], <vscale x 2 x i8> [[SHIFT:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 2 x i8> [[TMP0]]
+//
+vint8mf4_t test_vnsra_wv_i8mf4(vint16mf2_t op1, vuint8mf4_t shift, size_t vl) {
+  return vnsra(op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsra_wx_i8mf4(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i8> @llvm.riscv.vnsra.nxv2i8.nxv2i16.i32.i32(<vscale x 2 x i16> [[OP1:%.*]], i32 [[SHIFT:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 2 x i8> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsra_wx_i8mf4(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i8> @llvm.riscv.vnsra.nxv2i8.nxv2i16.i64.i64(<vscale x 2 x i16> [[OP1:%.*]], i64 [[SHIFT:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 2 x i8> [[TMP0]]
+//
+vint8mf4_t test_vnsra_wx_i8mf4(vint16mf2_t op1, size_t shift, size_t vl) {
+  return vnsra(op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsra_wv_i8mf2(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i8> @llvm.riscv.vnsra.nxv4i8.nxv4i16.nxv4i8.i32(<vscale x 4 x i16> [[OP1:%.*]], <vscale x 4 x i8> [[SHIFT:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 4 x i8> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsra_wv_i8mf2(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i8> @llvm.riscv.vnsra.nxv4i8.nxv4i16.nxv4i8.i64(<vscale x 4 x i16> [[OP1:%.*]], <vscale x 4 x i8> [[SHIFT:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 4 x i8> [[TMP0]]
+//
+vint8mf2_t test_vnsra_wv_i8mf2(vint16m1_t op1, vuint8mf2_t shift, size_t vl) {
+  return vnsra(op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsra_wx_i8mf2(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i8> @llvm.riscv.vnsra.nxv4i8.nxv4i16.i32.i32(<vscale x 4 x i16> [[OP1:%.*]], i32 [[SHIFT:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 4 x i8> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsra_wx_i8mf2(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i8> @llvm.riscv.vnsra.nxv4i8.nxv4i16.i64.i64(<vscale x 4 x i16> [[OP1:%.*]], i64 [[SHIFT:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 4 x i8> [[TMP0]]
+//
+vint8mf2_t test_vnsra_wx_i8mf2(vint16m1_t op1, size_t shift, size_t vl) {
+  return vnsra(op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsra_wv_i8m1(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i8> @llvm.riscv.vnsra.nxv8i8.nxv8i16.nxv8i8.i32(<vscale x 8 x i16> [[OP1:%.*]], <vscale x 8 x i8> [[SHIFT:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 8 x i8> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsra_wv_i8m1(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i8> @llvm.riscv.vnsra.nxv8i8.nxv8i16.nxv8i8.i64(<vscale x 8 x i16> [[OP1:%.*]], <vscale x 8 x i8> [[SHIFT:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 8 x i8> [[TMP0]]
+//
+vint8m1_t test_vnsra_wv_i8m1(vint16m2_t op1, vuint8m1_t shift, size_t vl) {
+  return vnsra(op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsra_wx_i8m1(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i8> @llvm.riscv.vnsra.nxv8i8.nxv8i16.i32.i32(<vscale x 8 x i16> [[OP1:%.*]], i32 [[SHIFT:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 8 x i8> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsra_wx_i8m1(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i8> @llvm.riscv.vnsra.nxv8i8.nxv8i16.i64.i64(<vscale x 8 x i16> [[OP1:%.*]], i64 [[SHIFT:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 8 x i8> [[TMP0]]
+//
+vint8m1_t test_vnsra_wx_i8m1(vint16m2_t op1, size_t shift, size_t vl) {
+  return vnsra(op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsra_wv_i8m2(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i8> @llvm.riscv.vnsra.nxv16i8.nxv16i16.nxv16i8.i32(<vscale x 16 x i16> [[OP1:%.*]], <vscale x 16 x i8> [[SHIFT:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 16 x i8> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsra_wv_i8m2(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i8> @llvm.riscv.vnsra.nxv16i8.nxv16i16.nxv16i8.i64(<vscale x 16 x i16> [[OP1:%.*]], <vscale x 16 x i8> [[SHIFT:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 16 x i8> [[TMP0]]
+//
+vint8m2_t test_vnsra_wv_i8m2(vint16m4_t op1, vuint8m2_t shift, size_t vl) {
+  return vnsra(op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsra_wx_i8m2(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i8> @llvm.riscv.vnsra.nxv16i8.nxv16i16.i32.i32(<vscale x 16 x i16> [[OP1:%.*]], i32 [[SHIFT:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 16 x i8> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsra_wx_i8m2(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i8> @llvm.riscv.vnsra.nxv16i8.nxv16i16.i64.i64(<vscale x 16 x i16> [[OP1:%.*]], i64 [[SHIFT:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 16 x i8> [[TMP0]]
+//
+vint8m2_t test_vnsra_wx_i8m2(vint16m4_t op1, size_t shift, size_t vl) {
+  return vnsra(op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsra_wv_i8m4(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i8> @llvm.riscv.vnsra.nxv32i8.nxv32i16.nxv32i8.i32(<vscale x 32 x i16> [[OP1:%.*]], <vscale x 32 x i8> [[SHIFT:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 32 x i8> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsra_wv_i8m4(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i8> @llvm.riscv.vnsra.nxv32i8.nxv32i16.nxv32i8.i64(<vscale x 32 x i16> [[OP1:%.*]], <vscale x 32 x i8> [[SHIFT:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 32 x i8> [[TMP0]]
+//
+vint8m4_t test_vnsra_wv_i8m4(vint16m8_t op1, vuint8m4_t shift, size_t vl) {
+  return vnsra(op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsra_wx_i8m4(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i8> @llvm.riscv.vnsra.nxv32i8.nxv32i16.i32.i32(<vscale x 32 x i16> [[OP1:%.*]], i32 [[SHIFT:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 32 x i8> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsra_wx_i8m4(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i8> @llvm.riscv.vnsra.nxv32i8.nxv32i16.i64.i64(<vscale x 32 x i16> [[OP1:%.*]], i64 [[SHIFT:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 32 x i8> [[TMP0]]
+//
+vint8m4_t test_vnsra_wx_i8m4(vint16m8_t op1, size_t shift, size_t vl) {
+  return vnsra(op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsra_wv_i16mf4(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vnsra.nxv1i16.nxv1i32.nxv1i16.i32(<vscale x 1 x i32> [[OP1:%.*]], <vscale x 1 x i16> [[SHIFT:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 1 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsra_wv_i16mf4(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vnsra.nxv1i16.nxv1i32.nxv1i16.i64(<vscale x 1 x i32> [[OP1:%.*]], <vscale x 1 x i16> [[SHIFT:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 1 x i16> [[TMP0]]
+//
+vint16mf4_t test_vnsra_wv_i16mf4(vint32mf2_t op1, vuint16mf4_t shift, size_t vl) {
+  return vnsra(op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsra_wx_i16mf4(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vnsra.nxv1i16.nxv1i32.i32.i32(<vscale x 1 x i32> [[OP1:%.*]], i32 [[SHIFT:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 1 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsra_wx_i16mf4(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vnsra.nxv1i16.nxv1i32.i64.i64(<vscale x 1 x i32> [[OP1:%.*]], i64 [[SHIFT:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 1 x i16> [[TMP0]]
+//
+vint16mf4_t test_vnsra_wx_i16mf4(vint32mf2_t op1, size_t shift, size_t vl) {
+  return vnsra(op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsra_wv_i16mf2(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vnsra.nxv2i16.nxv2i32.nxv2i16.i32(<vscale x 2 x i32> [[OP1:%.*]], <vscale x 2 x i16> [[SHIFT:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 2 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsra_wv_i16mf2(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vnsra.nxv2i16.nxv2i32.nxv2i16.i64(<vscale x 2 x i32> [[OP1:%.*]], <vscale x 2 x i16> [[SHIFT:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 2 x i16> [[TMP0]]
+//
+vint16mf2_t test_vnsra_wv_i16mf2(vint32m1_t op1, vuint16mf2_t shift, size_t vl) {
+  return vnsra(op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsra_wx_i16mf2(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vnsra.nxv2i16.nxv2i32.i32.i32(<vscale x 2 x i32> [[OP1:%.*]], i32 [[SHIFT:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 2 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsra_wx_i16mf2(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vnsra.nxv2i16.nxv2i32.i64.i64(<vscale x 2 x i32> [[OP1:%.*]], i64 [[SHIFT:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 2 x i16> [[TMP0]]
+//
+vint16mf2_t test_vnsra_wx_i16mf2(vint32m1_t op1, size_t shift, size_t vl) {
+  return vnsra(op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsra_wv_i16m1(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vnsra.nxv4i16.nxv4i32.nxv4i16.i32(<vscale x 4 x i32> [[OP1:%.*]], <vscale x 4 x i16> [[SHIFT:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 4 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsra_wv_i16m1(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vnsra.nxv4i16.nxv4i32.nxv4i16.i64(<vscale x 4 x i32> [[OP1:%.*]], <vscale x 4 x i16> [[SHIFT:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 4 x i16> [[TMP0]]
+//
+vint16m1_t test_vnsra_wv_i16m1(vint32m2_t op1, vuint16m1_t shift, size_t vl) {
+  return vnsra(op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsra_wx_i16m1(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vnsra.nxv4i16.nxv4i32.i32.i32(<vscale x 4 x i32> [[OP1:%.*]], i32 [[SHIFT:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 4 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsra_wx_i16m1(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vnsra.nxv4i16.nxv4i32.i64.i64(<vscale x 4 x i32> [[OP1:%.*]], i64 [[SHIFT:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 4 x i16> [[TMP0]]
+//
+vint16m1_t test_vnsra_wx_i16m1(vint32m2_t op1, size_t shift, size_t vl) {
+  return vnsra(op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsra_wv_i16m2(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vnsra.nxv8i16.nxv8i32.nxv8i16.i32(<vscale x 8 x i32> [[OP1:%.*]], <vscale x 8 x i16> [[SHIFT:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsra_wv_i16m2(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vnsra.nxv8i16.nxv8i32.nxv8i16.i64(<vscale x 8 x i32> [[OP1:%.*]], <vscale x 8 x i16> [[SHIFT:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
+//
+vint16m2_t test_vnsra_wv_i16m2(vint32m4_t op1, vuint16m2_t shift, size_t vl) {
+  return vnsra(op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsra_wx_i16m2(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vnsra.nxv8i16.nxv8i32.i32.i32(<vscale x 8 x i32> [[OP1:%.*]], i32 [[SHIFT:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsra_wx_i16m2(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vnsra.nxv8i16.nxv8i32.i64.i64(<vscale x 8 x i32> [[OP1:%.*]], i64 [[SHIFT:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
+//
+vint16m2_t test_vnsra_wx_i16m2(vint32m4_t op1, size_t shift, size_t vl) {
+  return vnsra(op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsra_wv_i16m4(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vnsra.nxv16i16.nxv16i32.nxv16i16.i32(<vscale x 16 x i32> [[OP1:%.*]], <vscale x 16 x i16> [[SHIFT:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 16 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsra_wv_i16m4(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vnsra.nxv16i16.nxv16i32.nxv16i16.i64(<vscale x 16 x i32> [[OP1:%.*]], <vscale x 16 x i16> [[SHIFT:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 16 x i16> [[TMP0]]
+//
+vint16m4_t test_vnsra_wv_i16m4(vint32m8_t op1, vuint16m4_t shift, size_t vl) {
+  return vnsra(op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsra_wx_i16m4(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vnsra.nxv16i16.nxv16i32.i32.i32(<vscale x 16 x i32> [[OP1:%.*]], i32 [[SHIFT:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 16 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsra_wx_i16m4(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vnsra.nxv16i16.nxv16i32.i64.i64(<vscale x 16 x i32> [[OP1:%.*]], i64 [[SHIFT:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 16 x i16> [[TMP0]]
+//
+vint16m4_t test_vnsra_wx_i16m4(vint32m8_t op1, size_t shift, size_t vl) {
+  return vnsra(op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsra_wv_i32mf2(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vnsra.nxv1i32.nxv1i64.nxv1i32.i32(<vscale x 1 x i64> [[OP1:%.*]], <vscale x 1 x i32> [[SHIFT:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsra_wv_i32mf2(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vnsra.nxv1i32.nxv1i64.nxv1i32.i64(<vscale x 1 x i64> [[OP1:%.*]], <vscale x 1 x i32> [[SHIFT:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
+//
+vint32mf2_t test_vnsra_wv_i32mf2(vint64m1_t op1, vuint32mf2_t shift, size_t vl) {
+  return vnsra(op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsra_wx_i32mf2(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vnsra.nxv1i32.nxv1i64.i32.i32(<vscale x 1 x i64> [[OP1:%.*]], i32 [[SHIFT:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsra_wx_i32mf2(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vnsra.nxv1i32.nxv1i64.i64.i64(<vscale x 1 x i64> [[OP1:%.*]], i64 [[SHIFT:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
+//
+vint32mf2_t test_vnsra_wx_i32mf2(vint64m1_t op1, size_t shift, size_t vl) {
+  return vnsra(op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsra_wv_i32m1(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vnsra.nxv2i32.nxv2i64.nxv2i32.i32(<vscale x 2 x i64> [[OP1:%.*]], <vscale x 2 x i32> [[SHIFT:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsra_wv_i32m1(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vnsra.nxv2i32.nxv2i64.nxv2i32.i64(<vscale x 2 x i64> [[OP1:%.*]], <vscale x 2 x i32> [[SHIFT:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
+//
+vint32m1_t test_vnsra_wv_i32m1(vint64m2_t op1, vuint32m1_t shift, size_t vl) {
+  return vnsra(op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsra_wx_i32m1(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vnsra.nxv2i32.nxv2i64.i32.i32(<vscale x 2 x i64> [[OP1:%.*]], i32 [[SHIFT:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsra_wx_i32m1(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vnsra.nxv2i32.nxv2i64.i64.i64(<vscale x 2 x i64> [[OP1:%.*]], i64 [[SHIFT:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
+//
+vint32m1_t test_vnsra_wx_i32m1(vint64m2_t op1, size_t shift, size_t vl) {
+  return vnsra(op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsra_wv_i32m2(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vnsra.nxv4i32.nxv4i64.nxv4i32.i32(<vscale x 4 x i64> [[OP1:%.*]], <vscale x 4 x i32> [[SHIFT:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsra_wv_i32m2(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vnsra.nxv4i32.nxv4i64.nxv4i32.i64(<vscale x 4 x i64> [[OP1:%.*]], <vscale x 4 x i32> [[SHIFT:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
+//
+vint32m2_t test_vnsra_wv_i32m2(vint64m4_t op1, vuint32m2_t shift, size_t vl) {
+  return vnsra(op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsra_wx_i32m2(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vnsra.nxv4i32.nxv4i64.i32.i32(<vscale x 4 x i64> [[OP1:%.*]], i32 [[SHIFT:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsra_wx_i32m2(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vnsra.nxv4i32.nxv4i64.i64.i64(<vscale x 4 x i64> [[OP1:%.*]], i64 [[SHIFT:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
+//
+vint32m2_t test_vnsra_wx_i32m2(vint64m4_t op1, size_t shift, size_t vl) {
+  return vnsra(op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsra_wv_i32m4(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vnsra.nxv8i32.nxv8i64.nxv8i32.i32(<vscale x 8 x i64> [[OP1:%.*]], <vscale x 8 x i32> [[SHIFT:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsra_wv_i32m4(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vnsra.nxv8i32.nxv8i64.nxv8i32.i64(<vscale x 8 x i64> [[OP1:%.*]], <vscale x 8 x i32> [[SHIFT:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
+//
+vint32m4_t test_vnsra_wv_i32m4(vint64m8_t op1, vuint32m4_t shift, size_t vl) {
+  return vnsra(op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsra_wx_i32m4(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vnsra.nxv8i32.nxv8i64.i32.i32(<vscale x 8 x i64> [[OP1:%.*]], i32 [[SHIFT:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsra_wx_i32m4(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vnsra.nxv8i32.nxv8i64.i64.i64(<vscale x 8 x i64> [[OP1:%.*]], i64 [[SHIFT:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
+//
+vint32m4_t test_vnsra_wx_i32m4(vint64m8_t op1, size_t shift, size_t vl) {
+  return vnsra(op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsra_wv_i8mf8_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i8> @llvm.riscv.vnsra.mask.nxv1i8.nxv1i16.nxv1i8.i32(<vscale x 1 x i8> [[MASKEDOFF:%.*]], <vscale x 1 x i16> [[OP1:%.*]], <vscale x 1 x i8> [[SHIFT:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 1 x i8> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsra_wv_i8mf8_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i8> @llvm.riscv.vnsra.mask.nxv1i8.nxv1i16.nxv1i8.i64(<vscale x 1 x i8> [[MASKEDOFF:%.*]], <vscale x 1 x i16> [[OP1:%.*]], <vscale x 1 x i8> [[SHIFT:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 1 x i8> [[TMP0]]
+//
+vint8mf8_t test_vnsra_wv_i8mf8_m(vbool64_t mask, vint8mf8_t maskedoff, vint16mf4_t op1, vuint8mf8_t shift, size_t vl) {
+  return vnsra(mask, maskedoff, op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsra_wx_i8mf8_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i8> @llvm.riscv.vnsra.mask.nxv1i8.nxv1i16.i32.i32(<vscale x 1 x i8> [[MASKEDOFF:%.*]], <vscale x 1 x i16> [[OP1:%.*]], i32 [[SHIFT:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 1 x i8> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsra_wx_i8mf8_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i8> @llvm.riscv.vnsra.mask.nxv1i8.nxv1i16.i64.i64(<vscale x 1 x i8> [[MASKEDOFF:%.*]], <vscale x 1 x i16> [[OP1:%.*]], i64 [[SHIFT:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 1 x i8> [[TMP0]]
+//
+vint8mf8_t test_vnsra_wx_i8mf8_m(vbool64_t mask, vint8mf8_t maskedoff, vint16mf4_t op1, size_t shift, size_t vl) {
+  return vnsra(mask, maskedoff, op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsra_wv_i8mf4_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i8> @llvm.riscv.vnsra.mask.nxv2i8.nxv2i16.nxv2i8.i32(<vscale x 2 x i8> [[MASKEDOFF:%.*]], <vscale x 2 x i16> [[OP1:%.*]], <vscale x 2 x i8> [[SHIFT:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 2 x i8> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsra_wv_i8mf4_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i8> @llvm.riscv.vnsra.mask.nxv2i8.nxv2i16.nxv2i8.i64(<vscale x 2 x i8> [[MASKEDOFF:%.*]], <vscale x 2 x i16> [[OP1:%.*]], <vscale x 2 x i8> [[SHIFT:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 2 x i8> [[TMP0]]
+//
+vint8mf4_t test_vnsra_wv_i8mf4_m(vbool32_t mask, vint8mf4_t maskedoff, vint16mf2_t op1, vuint8mf4_t shift, size_t vl) {
+  return vnsra(mask, maskedoff, op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsra_wx_i8mf4_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i8> @llvm.riscv.vnsra.mask.nxv2i8.nxv2i16.i32.i32(<vscale x 2 x i8> [[MASKEDOFF:%.*]], <vscale x 2 x i16> [[OP1:%.*]], i32 [[SHIFT:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 2 x i8> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsra_wx_i8mf4_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i8> @llvm.riscv.vnsra.mask.nxv2i8.nxv2i16.i64.i64(<vscale x 2 x i8> [[MASKEDOFF:%.*]], <vscale x 2 x i16> [[OP1:%.*]], i64 [[SHIFT:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 2 x i8> [[TMP0]]
+//
+vint8mf4_t test_vnsra_wx_i8mf4_m(vbool32_t mask, vint8mf4_t maskedoff, vint16mf2_t op1, size_t shift, size_t vl) {
+  return vnsra(mask, maskedoff, op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsra_wv_i8mf2_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i8> @llvm.riscv.vnsra.mask.nxv4i8.nxv4i16.nxv4i8.i32(<vscale x 4 x i8> [[MASKEDOFF:%.*]], <vscale x 4 x i16> [[OP1:%.*]], <vscale x 4 x i8> [[SHIFT:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 4 x i8> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsra_wv_i8mf2_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i8> @llvm.riscv.vnsra.mask.nxv4i8.nxv4i16.nxv4i8.i64(<vscale x 4 x i8> [[MASKEDOFF:%.*]], <vscale x 4 x i16> [[OP1:%.*]], <vscale x 4 x i8> [[SHIFT:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 4 x i8> [[TMP0]]
+//
+vint8mf2_t test_vnsra_wv_i8mf2_m(vbool16_t mask, vint8mf2_t maskedoff, vint16m1_t op1, vuint8mf2_t shift, size_t vl) {
+  return vnsra(mask, maskedoff, op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsra_wx_i8mf2_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i8> @llvm.riscv.vnsra.mask.nxv4i8.nxv4i16.i32.i32(<vscale x 4 x i8> [[MASKEDOFF:%.*]], <vscale x 4 x i16> [[OP1:%.*]], i32 [[SHIFT:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 4 x i8> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsra_wx_i8mf2_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i8> @llvm.riscv.vnsra.mask.nxv4i8.nxv4i16.i64.i64(<vscale x 4 x i8> [[MASKEDOFF:%.*]], <vscale x 4 x i16> [[OP1:%.*]], i64 [[SHIFT:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 4 x i8> [[TMP0]]
+//
+vint8mf2_t test_vnsra_wx_i8mf2_m(vbool16_t mask, vint8mf2_t maskedoff, vint16m1_t op1, size_t shift, size_t vl) {
+  return vnsra(mask, maskedoff, op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsra_wv_i8m1_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i8> @llvm.riscv.vnsra.mask.nxv8i8.nxv8i16.nxv8i8.i32(<vscale x 8 x i8> [[MASKEDOFF:%.*]], <vscale x 8 x i16> [[OP1:%.*]], <vscale x 8 x i8> [[SHIFT:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 8 x i8> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsra_wv_i8m1_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i8> @llvm.riscv.vnsra.mask.nxv8i8.nxv8i16.nxv8i8.i64(<vscale x 8 x i8> [[MASKEDOFF:%.*]], <vscale x 8 x i16> [[OP1:%.*]], <vscale x 8 x i8> [[SHIFT:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 8 x i8> [[TMP0]]
+//
+vint8m1_t test_vnsra_wv_i8m1_m(vbool8_t mask, vint8m1_t maskedoff, vint16m2_t op1, vuint8m1_t shift, size_t vl) {
+  return vnsra(mask, maskedoff, op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsra_wx_i8m1_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i8> @llvm.riscv.vnsra.mask.nxv8i8.nxv8i16.i32.i32(<vscale x 8 x i8> [[MASKEDOFF:%.*]], <vscale x 8 x i16> [[OP1:%.*]], i32 [[SHIFT:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 8 x i8> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsra_wx_i8m1_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i8> @llvm.riscv.vnsra.mask.nxv8i8.nxv8i16.i64.i64(<vscale x 8 x i8> [[MASKEDOFF:%.*]], <vscale x 8 x i16> [[OP1:%.*]], i64 [[SHIFT:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 8 x i8> [[TMP0]]
+//
+vint8m1_t test_vnsra_wx_i8m1_m(vbool8_t mask, vint8m1_t maskedoff, vint16m2_t op1, size_t shift, size_t vl) {
+  return vnsra(mask, maskedoff, op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsra_wv_i8m2_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i8> @llvm.riscv.vnsra.mask.nxv16i8.nxv16i16.nxv16i8.i32(<vscale x 16 x i8> [[MASKEDOFF:%.*]], <vscale x 16 x i16> [[OP1:%.*]], <vscale x 16 x i8> [[SHIFT:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 16 x i8> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsra_wv_i8m2_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i8> @llvm.riscv.vnsra.mask.nxv16i8.nxv16i16.nxv16i8.i64(<vscale x 16 x i8> [[MASKEDOFF:%.*]], <vscale x 16 x i16> [[OP1:%.*]], <vscale x 16 x i8> [[SHIFT:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 16 x i8> [[TMP0]]
+//
+vint8m2_t test_vnsra_wv_i8m2_m(vbool4_t mask, vint8m2_t maskedoff, vint16m4_t op1, vuint8m2_t shift, size_t vl) {
+  return vnsra(mask, maskedoff, op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsra_wx_i8m2_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i8> @llvm.riscv.vnsra.mask.nxv16i8.nxv16i16.i32.i32(<vscale x 16 x i8> [[MASKEDOFF:%.*]], <vscale x 16 x i16> [[OP1:%.*]], i32 [[SHIFT:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 16 x i8> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsra_wx_i8m2_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i8> @llvm.riscv.vnsra.mask.nxv16i8.nxv16i16.i64.i64(<vscale x 16 x i8> [[MASKEDOFF:%.*]], <vscale x 16 x i16> [[OP1:%.*]], i64 [[SHIFT:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 16 x i8> [[TMP0]]
+//
+vint8m2_t test_vnsra_wx_i8m2_m(vbool4_t mask, vint8m2_t maskedoff, vint16m4_t op1, size_t shift, size_t vl) {
+  return vnsra(mask, maskedoff, op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsra_wv_i8m4_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i8> @llvm.riscv.vnsra.mask.nxv32i8.nxv32i16.nxv32i8.i32(<vscale x 32 x i8> [[MASKEDOFF:%.*]], <vscale x 32 x i16> [[OP1:%.*]], <vscale x 32 x i8> [[SHIFT:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 32 x i8> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsra_wv_i8m4_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i8> @llvm.riscv.vnsra.mask.nxv32i8.nxv32i16.nxv32i8.i64(<vscale x 32 x i8> [[MASKEDOFF:%.*]], <vscale x 32 x i16> [[OP1:%.*]], <vscale x 32 x i8> [[SHIFT:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 32 x i8> [[TMP0]]
+//
+vint8m4_t test_vnsra_wv_i8m4_m(vbool2_t mask, vint8m4_t maskedoff, vint16m8_t op1, vuint8m4_t shift, size_t vl) {
+  return vnsra(mask, maskedoff, op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsra_wx_i8m4_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i8> @llvm.riscv.vnsra.mask.nxv32i8.nxv32i16.i32.i32(<vscale x 32 x i8> [[MASKEDOFF:%.*]], <vscale x 32 x i16> [[OP1:%.*]], i32 [[SHIFT:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 32 x i8> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsra_wx_i8m4_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i8> @llvm.riscv.vnsra.mask.nxv32i8.nxv32i16.i64.i64(<vscale x 32 x i8> [[MASKEDOFF:%.*]], <vscale x 32 x i16> [[OP1:%.*]], i64 [[SHIFT:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 32 x i8> [[TMP0]]
+//
+vint8m4_t test_vnsra_wx_i8m4_m(vbool2_t mask, vint8m4_t maskedoff, vint16m8_t op1, size_t shift, size_t vl) {
+  return vnsra(mask, maskedoff, op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsra_wv_i16mf4_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vnsra.mask.nxv1i16.nxv1i32.nxv1i16.i32(<vscale x 1 x i16> [[MASKEDOFF:%.*]], <vscale x 1 x i32> [[OP1:%.*]], <vscale x 1 x i16> [[SHIFT:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 1 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsra_wv_i16mf4_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vnsra.mask.nxv1i16.nxv1i32.nxv1i16.i64(<vscale x 1 x i16> [[MASKEDOFF:%.*]], <vscale x 1 x i32> [[OP1:%.*]], <vscale x 1 x i16> [[SHIFT:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 1 x i16> [[TMP0]]
+//
+vint16mf4_t test_vnsra_wv_i16mf4_m(vbool64_t mask, vint16mf4_t maskedoff, vint32mf2_t op1, vuint16mf4_t shift, size_t vl) {
+  return vnsra(mask, maskedoff, op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsra_wx_i16mf4_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vnsra.mask.nxv1i16.nxv1i32.i32.i32(<vscale x 1 x i16> [[MASKEDOFF:%.*]], <vscale x 1 x i32> [[OP1:%.*]], i32 [[SHIFT:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 1 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsra_wx_i16mf4_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vnsra.mask.nxv1i16.nxv1i32.i64.i64(<vscale x 1 x i16> [[MASKEDOFF:%.*]], <vscale x 1 x i32> [[OP1:%.*]], i64 [[SHIFT:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 1 x i16> [[TMP0]]
+//
+vint16mf4_t test_vnsra_wx_i16mf4_m(vbool64_t mask, vint16mf4_t maskedoff, vint32mf2_t op1, size_t shift, size_t vl) {
+  return vnsra(mask, maskedoff, op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsra_wv_i16mf2_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vnsra.mask.nxv2i16.nxv2i32.nxv2i16.i32(<vscale x 2 x i16> [[MASKEDOFF:%.*]], <vscale x 2 x i32> [[OP1:%.*]], <vscale x 2 x i16> [[SHIFT:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 2 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsra_wv_i16mf2_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vnsra.mask.nxv2i16.nxv2i32.nxv2i16.i64(<vscale x 2 x i16> [[MASKEDOFF:%.*]], <vscale x 2 x i32> [[OP1:%.*]], <vscale x 2 x i16> [[SHIFT:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 2 x i16> [[TMP0]]
+//
+vint16mf2_t test_vnsra_wv_i16mf2_m(vbool32_t mask, vint16mf2_t maskedoff, vint32m1_t op1, vuint16mf2_t shift, size_t vl) {
+  return vnsra(mask, maskedoff, op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsra_wx_i16mf2_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vnsra.mask.nxv2i16.nxv2i32.i32.i32(<vscale x 2 x i16> [[MASKEDOFF:%.*]], <vscale x 2 x i32> [[OP1:%.*]], i32 [[SHIFT:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 2 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsra_wx_i16mf2_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vnsra.mask.nxv2i16.nxv2i32.i64.i64(<vscale x 2 x i16> [[MASKEDOFF:%.*]], <vscale x 2 x i32> [[OP1:%.*]], i64 [[SHIFT:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 2 x i16> [[TMP0]]
+//
+vint16mf2_t test_vnsra_wx_i16mf2_m(vbool32_t mask, vint16mf2_t maskedoff, vint32m1_t op1, size_t shift, size_t vl) {
+  return vnsra(mask, maskedoff, op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsra_wv_i16m1_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vnsra.mask.nxv4i16.nxv4i32.nxv4i16.i32(<vscale x 4 x i16> [[MASKEDOFF:%.*]], <vscale x 4 x i32> [[OP1:%.*]], <vscale x 4 x i16> [[SHIFT:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 4 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsra_wv_i16m1_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vnsra.mask.nxv4i16.nxv4i32.nxv4i16.i64(<vscale x 4 x i16> [[MASKEDOFF:%.*]], <vscale x 4 x i32> [[OP1:%.*]], <vscale x 4 x i16> [[SHIFT:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 4 x i16> [[TMP0]]
+//
+vint16m1_t test_vnsra_wv_i16m1_m(vbool16_t mask, vint16m1_t maskedoff, vint32m2_t op1, vuint16m1_t shift, size_t vl) {
+  return vnsra(mask, maskedoff, op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsra_wx_i16m1_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vnsra.mask.nxv4i16.nxv4i32.i32.i32(<vscale x 4 x i16> [[MASKEDOFF:%.*]], <vscale x 4 x i32> [[OP1:%.*]], i32 [[SHIFT:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 4 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsra_wx_i16m1_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vnsra.mask.nxv4i16.nxv4i32.i64.i64(<vscale x 4 x i16> [[MASKEDOFF:%.*]], <vscale x 4 x i32> [[OP1:%.*]], i64 [[SHIFT:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 4 x i16> [[TMP0]]
+//
+vint16m1_t test_vnsra_wx_i16m1_m(vbool16_t mask, vint16m1_t maskedoff, vint32m2_t op1, size_t shift, size_t vl) {
+  return vnsra(mask, maskedoff, op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsra_wv_i16m2_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vnsra.mask.nxv8i16.nxv8i32.nxv8i16.i32(<vscale x 8 x i16> [[MASKEDOFF:%.*]], <vscale x 8 x i32> [[OP1:%.*]], <vscale x 8 x i16> [[SHIFT:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsra_wv_i16m2_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vnsra.mask.nxv8i16.nxv8i32.nxv8i16.i64(<vscale x 8 x i16> [[MASKEDOFF:%.*]], <vscale x 8 x i32> [[OP1:%.*]], <vscale x 8 x i16> [[SHIFT:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
+//
+vint16m2_t test_vnsra_wv_i16m2_m(vbool8_t mask, vint16m2_t maskedoff, vint32m4_t op1, vuint16m2_t shift, size_t vl) {
+  return vnsra(mask, maskedoff, op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsra_wx_i16m2_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vnsra.mask.nxv8i16.nxv8i32.i32.i32(<vscale x 8 x i16> [[MASKEDOFF:%.*]], <vscale x 8 x i32> [[OP1:%.*]], i32 [[SHIFT:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsra_wx_i16m2_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vnsra.mask.nxv8i16.nxv8i32.i64.i64(<vscale x 8 x i16> [[MASKEDOFF:%.*]], <vscale x 8 x i32> [[OP1:%.*]], i64 [[SHIFT:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
+//
+vint16m2_t test_vnsra_wx_i16m2_m(vbool8_t mask, vint16m2_t maskedoff, vint32m4_t op1, size_t shift, size_t vl) {
+  return vnsra(mask, maskedoff, op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsra_wv_i16m4_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vnsra.mask.nxv16i16.nxv16i32.nxv16i16.i32(<vscale x 16 x i16> [[MASKEDOFF:%.*]], <vscale x 16 x i32> [[OP1:%.*]], <vscale x 16 x i16> [[SHIFT:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 16 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsra_wv_i16m4_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vnsra.mask.nxv16i16.nxv16i32.nxv16i16.i64(<vscale x 16 x i16> [[MASKEDOFF:%.*]], <vscale x 16 x i32> [[OP1:%.*]], <vscale x 16 x i16> [[SHIFT:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 16 x i16> [[TMP0]]
+//
+vint16m4_t test_vnsra_wv_i16m4_m(vbool4_t mask, vint16m4_t maskedoff, vint32m8_t op1, vuint16m4_t shift, size_t vl) {
+  return vnsra(mask, maskedoff, op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsra_wx_i16m4_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vnsra.mask.nxv16i16.nxv16i32.i32.i32(<vscale x 16 x i16> [[MASKEDOFF:%.*]], <vscale x 16 x i32> [[OP1:%.*]], i32 [[SHIFT:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 16 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsra_wx_i16m4_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vnsra.mask.nxv16i16.nxv16i32.i64.i64(<vscale x 16 x i16> [[MASKEDOFF:%.*]], <vscale x 16 x i32> [[OP1:%.*]], i64 [[SHIFT:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 16 x i16> [[TMP0]]
+//
+vint16m4_t test_vnsra_wx_i16m4_m(vbool4_t mask, vint16m4_t maskedoff, vint32m8_t op1, size_t shift, size_t vl) {
+  return vnsra(mask, maskedoff, op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsra_wv_i32mf2_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vnsra.mask.nxv1i32.nxv1i64.nxv1i32.i32(<vscale x 1 x i32> [[MASKEDOFF:%.*]], <vscale x 1 x i64> [[OP1:%.*]], <vscale x 1 x i32> [[SHIFT:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsra_wv_i32mf2_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vnsra.mask.nxv1i32.nxv1i64.nxv1i32.i64(<vscale x 1 x i32> [[MASKEDOFF:%.*]], <vscale x 1 x i64> [[OP1:%.*]], <vscale x 1 x i32> [[SHIFT:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
+//
+vint32mf2_t test_vnsra_wv_i32mf2_m(vbool64_t mask, vint32mf2_t maskedoff, vint64m1_t op1, vuint32mf2_t shift, size_t vl) {
+  return vnsra(mask, maskedoff, op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsra_wx_i32mf2_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vnsra.mask.nxv1i32.nxv1i64.i32.i32(<vscale x 1 x i32> [[MASKEDOFF:%.*]], <vscale x 1 x i64> [[OP1:%.*]], i32 [[SHIFT:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsra_wx_i32mf2_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vnsra.mask.nxv1i32.nxv1i64.i64.i64(<vscale x 1 x i32> [[MASKEDOFF:%.*]], <vscale x 1 x i64> [[OP1:%.*]], i64 [[SHIFT:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
+//
+vint32mf2_t test_vnsra_wx_i32mf2_m(vbool64_t mask, vint32mf2_t maskedoff, vint64m1_t op1, size_t shift, size_t vl) {
+  return vnsra(mask, maskedoff, op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsra_wv_i32m1_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vnsra.mask.nxv2i32.nxv2i64.nxv2i32.i32(<vscale x 2 x i32> [[MASKEDOFF:%.*]], <vscale x 2 x i64> [[OP1:%.*]], <vscale x 2 x i32> [[SHIFT:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsra_wv_i32m1_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vnsra.mask.nxv2i32.nxv2i64.nxv2i32.i64(<vscale x 2 x i32> [[MASKEDOFF:%.*]], <vscale x 2 x i64> [[OP1:%.*]], <vscale x 2 x i32> [[SHIFT:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
+//
+vint32m1_t test_vnsra_wv_i32m1_m(vbool32_t mask, vint32m1_t maskedoff, vint64m2_t op1, vuint32m1_t shift, size_t vl) {
+  return vnsra(mask, maskedoff, op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsra_wx_i32m1_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vnsra.mask.nxv2i32.nxv2i64.i32.i32(<vscale x 2 x i32> [[MASKEDOFF:%.*]], <vscale x 2 x i64> [[OP1:%.*]], i32 [[SHIFT:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsra_wx_i32m1_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vnsra.mask.nxv2i32.nxv2i64.i64.i64(<vscale x 2 x i32> [[MASKEDOFF:%.*]], <vscale x 2 x i64> [[OP1:%.*]], i64 [[SHIFT:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
+//
+vint32m1_t test_vnsra_wx_i32m1_m(vbool32_t mask, vint32m1_t maskedoff, vint64m2_t op1, size_t shift, size_t vl) {
+  return vnsra(mask, maskedoff, op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsra_wv_i32m2_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vnsra.mask.nxv4i32.nxv4i64.nxv4i32.i32(<vscale x 4 x i32> [[MASKEDOFF:%.*]], <vscale x 4 x i64> [[OP1:%.*]], <vscale x 4 x i32> [[SHIFT:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsra_wv_i32m2_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vnsra.mask.nxv4i32.nxv4i64.nxv4i32.i64(<vscale x 4 x i32> [[MASKEDOFF:%.*]], <vscale x 4 x i64> [[OP1:%.*]], <vscale x 4 x i32> [[SHIFT:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
+//
+vint32m2_t test_vnsra_wv_i32m2_m(vbool16_t mask, vint32m2_t maskedoff, vint64m4_t op1, vuint32m2_t shift, size_t vl) {
+  return vnsra(mask, maskedoff, op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsra_wx_i32m2_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vnsra.mask.nxv4i32.nxv4i64.i32.i32(<vscale x 4 x i32> [[MASKEDOFF:%.*]], <vscale x 4 x i64> [[OP1:%.*]], i32 [[SHIFT:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsra_wx_i32m2_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vnsra.mask.nxv4i32.nxv4i64.i64.i64(<vscale x 4 x i32> [[MASKEDOFF:%.*]], <vscale x 4 x i64> [[OP1:%.*]], i64 [[SHIFT:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
+//
+vint32m2_t test_vnsra_wx_i32m2_m(vbool16_t mask, vint32m2_t maskedoff, vint64m4_t op1, size_t shift, size_t vl) {
+  return vnsra(mask, maskedoff, op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsra_wv_i32m4_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vnsra.mask.nxv8i32.nxv8i64.nxv8i32.i32(<vscale x 8 x i32> [[MASKEDOFF:%.*]], <vscale x 8 x i64> [[OP1:%.*]], <vscale x 8 x i32> [[SHIFT:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsra_wv_i32m4_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vnsra.mask.nxv8i32.nxv8i64.nxv8i32.i64(<vscale x 8 x i32> [[MASKEDOFF:%.*]], <vscale x 8 x i64> [[OP1:%.*]], <vscale x 8 x i32> [[SHIFT:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
+//
+vint32m4_t test_vnsra_wv_i32m4_m(vbool8_t mask, vint32m4_t maskedoff, vint64m8_t op1, vuint32m4_t shift, size_t vl) {
+  return vnsra(mask, maskedoff, op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsra_wx_i32m4_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vnsra.mask.nxv8i32.nxv8i64.i32.i32(<vscale x 8 x i32> [[MASKEDOFF:%.*]], <vscale x 8 x i64> [[OP1:%.*]], i32 [[SHIFT:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsra_wx_i32m4_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vnsra.mask.nxv8i32.nxv8i64.i64.i64(<vscale x 8 x i32> [[MASKEDOFF:%.*]], <vscale x 8 x i64> [[OP1:%.*]], i64 [[SHIFT:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
+//
+vint32m4_t test_vnsra_wx_i32m4_m(vbool8_t mask, vint32m4_t maskedoff, vint64m8_t op1, size_t shift, size_t vl) {
+  return vnsra(mask, maskedoff, op1, shift, vl);
+}

clang/test/CodeGen/RISCV/rvv-intrinsics-overloaded/vnsrl.c

+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv32 -target-feature +experimental-v -disable-O0-optnone -emit-llvm %s -o - | opt -S -mem2reg | FileCheck --check-prefix=CHECK-RV32 %s
+// RUN: %clang_cc1 -triple riscv64 -target-feature +experimental-v -disable-O0-optnone -emit-llvm %s -o - | opt -S -mem2reg | FileCheck --check-prefix=CHECK-RV64 %s
+// RUN: %clang_cc1 -triple riscv64 -target-feature +m -target-feature +experimental-v -Werror -Wall -o - %s -S >/dev/null 2>&1 | FileCheck --check-prefix=ASM --allow-empty %s
+
+// ASM-NOT: warning
+#include <riscv_vector.h>
+
+// CHECK-RV32-LABEL: @test_vnsrl_wv_u8mf8(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i8> @llvm.riscv.vnsrl.nxv1i8.nxv1i16.nxv1i8.i32(<vscale x 1 x i16> [[OP1:%.*]], <vscale x 1 x i8> [[SHIFT:%.*]], i32 [[VL:%.*]]) [[ATTR7:#.*]]
+// CHECK-RV32-NEXT:    ret <vscale x 1 x i8> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsrl_wv_u8mf8(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i8> @llvm.riscv.vnsrl.nxv1i8.nxv1i16.nxv1i8.i64(<vscale x 1 x i16> [[OP1:%.*]], <vscale x 1 x i8> [[SHIFT:%.*]], i64 [[VL:%.*]]) [[ATTR7:#.*]]
+// CHECK-RV64-NEXT:    ret <vscale x 1 x i8> [[TMP0]]
+//
+vuint8mf8_t test_vnsrl_wv_u8mf8(vuint16mf4_t op1, vuint8mf8_t shift, size_t vl) {
+  return vnsrl(op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsrl_wx_u8mf8(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i8> @llvm.riscv.vnsrl.nxv1i8.nxv1i16.i32.i32(<vscale x 1 x i16> [[OP1:%.*]], i32 [[SHIFT:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 1 x i8> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsrl_wx_u8mf8(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i8> @llvm.riscv.vnsrl.nxv1i8.nxv1i16.i64.i64(<vscale x 1 x i16> [[OP1:%.*]], i64 [[SHIFT:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 1 x i8> [[TMP0]]
+//
+vuint8mf8_t test_vnsrl_wx_u8mf8(vuint16mf4_t op1, size_t shift, size_t vl) {
+  return vnsrl(op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsrl_wv_u8mf4(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i8> @llvm.riscv.vnsrl.nxv2i8.nxv2i16.nxv2i8.i32(<vscale x 2 x i16> [[OP1:%.*]], <vscale x 2 x i8> [[SHIFT:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 2 x i8> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsrl_wv_u8mf4(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i8> @llvm.riscv.vnsrl.nxv2i8.nxv2i16.nxv2i8.i64(<vscale x 2 x i16> [[OP1:%.*]], <vscale x 2 x i8> [[SHIFT:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 2 x i8> [[TMP0]]
+//
+vuint8mf4_t test_vnsrl_wv_u8mf4(vuint16mf2_t op1, vuint8mf4_t shift, size_t vl) {
+  return vnsrl(op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsrl_wx_u8mf4(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i8> @llvm.riscv.vnsrl.nxv2i8.nxv2i16.i32.i32(<vscale x 2 x i16> [[OP1:%.*]], i32 [[SHIFT:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 2 x i8> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsrl_wx_u8mf4(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i8> @llvm.riscv.vnsrl.nxv2i8.nxv2i16.i64.i64(<vscale x 2 x i16> [[OP1:%.*]], i64 [[SHIFT:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 2 x i8> [[TMP0]]
+//
+vuint8mf4_t test_vnsrl_wx_u8mf4(vuint16mf2_t op1, size_t shift, size_t vl) {
+  return vnsrl(op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsrl_wv_u8mf2(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i8> @llvm.riscv.vnsrl.nxv4i8.nxv4i16.nxv4i8.i32(<vscale x 4 x i16> [[OP1:%.*]], <vscale x 4 x i8> [[SHIFT:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 4 x i8> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsrl_wv_u8mf2(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i8> @llvm.riscv.vnsrl.nxv4i8.nxv4i16.nxv4i8.i64(<vscale x 4 x i16> [[OP1:%.*]], <vscale x 4 x i8> [[SHIFT:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 4 x i8> [[TMP0]]
+//
+vuint8mf2_t test_vnsrl_wv_u8mf2(vuint16m1_t op1, vuint8mf2_t shift, size_t vl) {
+  return vnsrl(op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsrl_wx_u8mf2(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i8> @llvm.riscv.vnsrl.nxv4i8.nxv4i16.i32.i32(<vscale x 4 x i16> [[OP1:%.*]], i32 [[SHIFT:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 4 x i8> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsrl_wx_u8mf2(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i8> @llvm.riscv.vnsrl.nxv4i8.nxv4i16.i64.i64(<vscale x 4 x i16> [[OP1:%.*]], i64 [[SHIFT:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 4 x i8> [[TMP0]]
+//
+vuint8mf2_t test_vnsrl_wx_u8mf2(vuint16m1_t op1, size_t shift, size_t vl) {
+  return vnsrl(op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsrl_wv_u8m1(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i8> @llvm.riscv.vnsrl.nxv8i8.nxv8i16.nxv8i8.i32(<vscale x 8 x i16> [[OP1:%.*]], <vscale x 8 x i8> [[SHIFT:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 8 x i8> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsrl_wv_u8m1(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i8> @llvm.riscv.vnsrl.nxv8i8.nxv8i16.nxv8i8.i64(<vscale x 8 x i16> [[OP1:%.*]], <vscale x 8 x i8> [[SHIFT:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 8 x i8> [[TMP0]]
+//
+vuint8m1_t test_vnsrl_wv_u8m1(vuint16m2_t op1, vuint8m1_t shift, size_t vl) {
+  return vnsrl(op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsrl_wx_u8m1(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i8> @llvm.riscv.vnsrl.nxv8i8.nxv8i16.i32.i32(<vscale x 8 x i16> [[OP1:%.*]], i32 [[SHIFT:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 8 x i8> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsrl_wx_u8m1(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i8> @llvm.riscv.vnsrl.nxv8i8.nxv8i16.i64.i64(<vscale x 8 x i16> [[OP1:%.*]], i64 [[SHIFT:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 8 x i8> [[TMP0]]
+//
+vuint8m1_t test_vnsrl_wx_u8m1(vuint16m2_t op1, size_t shift, size_t vl) {
+  return vnsrl(op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsrl_wv_u8m2(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i8> @llvm.riscv.vnsrl.nxv16i8.nxv16i16.nxv16i8.i32(<vscale x 16 x i16> [[OP1:%.*]], <vscale x 16 x i8> [[SHIFT:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 16 x i8> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsrl_wv_u8m2(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i8> @llvm.riscv.vnsrl.nxv16i8.nxv16i16.nxv16i8.i64(<vscale x 16 x i16> [[OP1:%.*]], <vscale x 16 x i8> [[SHIFT:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 16 x i8> [[TMP0]]
+//
+vuint8m2_t test_vnsrl_wv_u8m2(vuint16m4_t op1, vuint8m2_t shift, size_t vl) {
+  return vnsrl(op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsrl_wx_u8m2(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i8> @llvm.riscv.vnsrl.nxv16i8.nxv16i16.i32.i32(<vscale x 16 x i16> [[OP1:%.*]], i32 [[SHIFT:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 16 x i8> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsrl_wx_u8m2(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i8> @llvm.riscv.vnsrl.nxv16i8.nxv16i16.i64.i64(<vscale x 16 x i16> [[OP1:%.*]], i64 [[SHIFT:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 16 x i8> [[TMP0]]
+//
+vuint8m2_t test_vnsrl_wx_u8m2(vuint16m4_t op1, size_t shift, size_t vl) {
+  return vnsrl(op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsrl_wv_u8m4(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i8> @llvm.riscv.vnsrl.nxv32i8.nxv32i16.nxv32i8.i32(<vscale x 32 x i16> [[OP1:%.*]], <vscale x 32 x i8> [[SHIFT:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 32 x i8> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsrl_wv_u8m4(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i8> @llvm.riscv.vnsrl.nxv32i8.nxv32i16.nxv32i8.i64(<vscale x 32 x i16> [[OP1:%.*]], <vscale x 32 x i8> [[SHIFT:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 32 x i8> [[TMP0]]
+//
+vuint8m4_t test_vnsrl_wv_u8m4(vuint16m8_t op1, vuint8m4_t shift, size_t vl) {
+  return vnsrl(op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsrl_wx_u8m4(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i8> @llvm.riscv.vnsrl.nxv32i8.nxv32i16.i32.i32(<vscale x 32 x i16> [[OP1:%.*]], i32 [[SHIFT:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 32 x i8> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsrl_wx_u8m4(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i8> @llvm.riscv.vnsrl.nxv32i8.nxv32i16.i64.i64(<vscale x 32 x i16> [[OP1:%.*]], i64 [[SHIFT:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 32 x i8> [[TMP0]]
+//
+vuint8m4_t test_vnsrl_wx_u8m4(vuint16m8_t op1, size_t shift, size_t vl) {
+  return vnsrl(op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsrl_wv_u16mf4(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vnsrl.nxv1i16.nxv1i32.nxv1i16.i32(<vscale x 1 x i32> [[OP1:%.*]], <vscale x 1 x i16> [[SHIFT:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 1 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsrl_wv_u16mf4(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vnsrl.nxv1i16.nxv1i32.nxv1i16.i64(<vscale x 1 x i32> [[OP1:%.*]], <vscale x 1 x i16> [[SHIFT:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 1 x i16> [[TMP0]]
+//
+vuint16mf4_t test_vnsrl_wv_u16mf4(vuint32mf2_t op1, vuint16mf4_t shift, size_t vl) {
+  return vnsrl(op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsrl_wx_u16mf4(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vnsrl.nxv1i16.nxv1i32.i32.i32(<vscale x 1 x i32> [[OP1:%.*]], i32 [[SHIFT:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 1 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsrl_wx_u16mf4(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vnsrl.nxv1i16.nxv1i32.i64.i64(<vscale x 1 x i32> [[OP1:%.*]], i64 [[SHIFT:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 1 x i16> [[TMP0]]
+//
+vuint16mf4_t test_vnsrl_wx_u16mf4(vuint32mf2_t op1, size_t shift, size_t vl) {
+  return vnsrl(op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsrl_wv_u16mf2(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vnsrl.nxv2i16.nxv2i32.nxv2i16.i32(<vscale x 2 x i32> [[OP1:%.*]], <vscale x 2 x i16> [[SHIFT:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 2 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsrl_wv_u16mf2(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vnsrl.nxv2i16.nxv2i32.nxv2i16.i64(<vscale x 2 x i32> [[OP1:%.*]], <vscale x 2 x i16> [[SHIFT:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 2 x i16> [[TMP0]]
+//
+vuint16mf2_t test_vnsrl_wv_u16mf2(vuint32m1_t op1, vuint16mf2_t shift, size_t vl) {
+  return vnsrl(op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsrl_wx_u16mf2(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vnsrl.nxv2i16.nxv2i32.i32.i32(<vscale x 2 x i32> [[OP1:%.*]], i32 [[SHIFT:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 2 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsrl_wx_u16mf2(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vnsrl.nxv2i16.nxv2i32.i64.i64(<vscale x 2 x i32> [[OP1:%.*]], i64 [[SHIFT:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 2 x i16> [[TMP0]]
+//
+vuint16mf2_t test_vnsrl_wx_u16mf2(vuint32m1_t op1, size_t shift, size_t vl) {
+  return vnsrl(op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsrl_wv_u16m1(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vnsrl.nxv4i16.nxv4i32.nxv4i16.i32(<vscale x 4 x i32> [[OP1:%.*]], <vscale x 4 x i16> [[SHIFT:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 4 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsrl_wv_u16m1(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vnsrl.nxv4i16.nxv4i32.nxv4i16.i64(<vscale x 4 x i32> [[OP1:%.*]], <vscale x 4 x i16> [[SHIFT:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 4 x i16> [[TMP0]]
+//
+vuint16m1_t test_vnsrl_wv_u16m1(vuint32m2_t op1, vuint16m1_t shift, size_t vl) {
+  return vnsrl(op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsrl_wx_u16m1(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vnsrl.nxv4i16.nxv4i32.i32.i32(<vscale x 4 x i32> [[OP1:%.*]], i32 [[SHIFT:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 4 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsrl_wx_u16m1(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vnsrl.nxv4i16.nxv4i32.i64.i64(<vscale x 4 x i32> [[OP1:%.*]], i64 [[SHIFT:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 4 x i16> [[TMP0]]
+//
+vuint16m1_t test_vnsrl_wx_u16m1(vuint32m2_t op1, size_t shift, size_t vl) {
+  return vnsrl(op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsrl_wv_u16m2(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vnsrl.nxv8i16.nxv8i32.nxv8i16.i32(<vscale x 8 x i32> [[OP1:%.*]], <vscale x 8 x i16> [[SHIFT:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsrl_wv_u16m2(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vnsrl.nxv8i16.nxv8i32.nxv8i16.i64(<vscale x 8 x i32> [[OP1:%.*]], <vscale x 8 x i16> [[SHIFT:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
+//
+vuint16m2_t test_vnsrl_wv_u16m2(vuint32m4_t op1, vuint16m2_t shift, size_t vl) {
+  return vnsrl(op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsrl_wx_u16m2(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vnsrl.nxv8i16.nxv8i32.i32.i32(<vscale x 8 x i32> [[OP1:%.*]], i32 [[SHIFT:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsrl_wx_u16m2(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vnsrl.nxv8i16.nxv8i32.i64.i64(<vscale x 8 x i32> [[OP1:%.*]], i64 [[SHIFT:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
+//
+vuint16m2_t test_vnsrl_wx_u16m2(vuint32m4_t op1, size_t shift, size_t vl) {
+  return vnsrl(op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsrl_wv_u16m4(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vnsrl.nxv16i16.nxv16i32.nxv16i16.i32(<vscale x 16 x i32> [[OP1:%.*]], <vscale x 16 x i16> [[SHIFT:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 16 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsrl_wv_u16m4(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vnsrl.nxv16i16.nxv16i32.nxv16i16.i64(<vscale x 16 x i32> [[OP1:%.*]], <vscale x 16 x i16> [[SHIFT:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 16 x i16> [[TMP0]]
+//
+vuint16m4_t test_vnsrl_wv_u16m4(vuint32m8_t op1, vuint16m4_t shift, size_t vl) {
+  return vnsrl(op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsrl_wx_u16m4(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vnsrl.nxv16i16.nxv16i32.i32.i32(<vscale x 16 x i32> [[OP1:%.*]], i32 [[SHIFT:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 16 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsrl_wx_u16m4(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vnsrl.nxv16i16.nxv16i32.i64.i64(<vscale x 16 x i32> [[OP1:%.*]], i64 [[SHIFT:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 16 x i16> [[TMP0]]
+//
+vuint16m4_t test_vnsrl_wx_u16m4(vuint32m8_t op1, size_t shift, size_t vl) {
+  return vnsrl(op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsrl_wv_u32mf2(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vnsrl.nxv1i32.nxv1i64.nxv1i32.i32(<vscale x 1 x i64> [[OP1:%.*]], <vscale x 1 x i32> [[SHIFT:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsrl_wv_u32mf2(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vnsrl.nxv1i32.nxv1i64.nxv1i32.i64(<vscale x 1 x i64> [[OP1:%.*]], <vscale x 1 x i32> [[SHIFT:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
+//
+vuint32mf2_t test_vnsrl_wv_u32mf2(vuint64m1_t op1, vuint32mf2_t shift, size_t vl) {
+  return vnsrl(op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsrl_wx_u32mf2(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vnsrl.nxv1i32.nxv1i64.i32.i32(<vscale x 1 x i64> [[OP1:%.*]], i32 [[SHIFT:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsrl_wx_u32mf2(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vnsrl.nxv1i32.nxv1i64.i64.i64(<vscale x 1 x i64> [[OP1:%.*]], i64 [[SHIFT:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
+//
+vuint32mf2_t test_vnsrl_wx_u32mf2(vuint64m1_t op1, size_t shift, size_t vl) {
+  return vnsrl(op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsrl_wv_u32m1(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vnsrl.nxv2i32.nxv2i64.nxv2i32.i32(<vscale x 2 x i64> [[OP1:%.*]], <vscale x 2 x i32> [[SHIFT:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsrl_wv_u32m1(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vnsrl.nxv2i32.nxv2i64.nxv2i32.i64(<vscale x 2 x i64> [[OP1:%.*]], <vscale x 2 x i32> [[SHIFT:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
+//
+vuint32m1_t test_vnsrl_wv_u32m1(vuint64m2_t op1, vuint32m1_t shift, size_t vl) {
+  return vnsrl(op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsrl_wx_u32m1(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vnsrl.nxv2i32.nxv2i64.i32.i32(<vscale x 2 x i64> [[OP1:%.*]], i32 [[SHIFT:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsrl_wx_u32m1(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vnsrl.nxv2i32.nxv2i64.i64.i64(<vscale x 2 x i64> [[OP1:%.*]], i64 [[SHIFT:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
+//
+vuint32m1_t test_vnsrl_wx_u32m1(vuint64m2_t op1, size_t shift, size_t vl) {
+  return vnsrl(op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsrl_wv_u32m2(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vnsrl.nxv4i32.nxv4i64.nxv4i32.i32(<vscale x 4 x i64> [[OP1:%.*]], <vscale x 4 x i32> [[SHIFT:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsrl_wv_u32m2(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vnsrl.nxv4i32.nxv4i64.nxv4i32.i64(<vscale x 4 x i64> [[OP1:%.*]], <vscale x 4 x i32> [[SHIFT:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
+//
+vuint32m2_t test_vnsrl_wv_u32m2(vuint64m4_t op1, vuint32m2_t shift, size_t vl) {
+  return vnsrl(op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsrl_wx_u32m2(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vnsrl.nxv4i32.nxv4i64.i32.i32(<vscale x 4 x i64> [[OP1:%.*]], i32 [[SHIFT:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsrl_wx_u32m2(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vnsrl.nxv4i32.nxv4i64.i64.i64(<vscale x 4 x i64> [[OP1:%.*]], i64 [[SHIFT:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
+//
+vuint32m2_t test_vnsrl_wx_u32m2(vuint64m4_t op1, size_t shift, size_t vl) {
+  return vnsrl(op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsrl_wv_u32m4(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vnsrl.nxv8i32.nxv8i64.nxv8i32.i32(<vscale x 8 x i64> [[OP1:%.*]], <vscale x 8 x i32> [[SHIFT:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsrl_wv_u32m4(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vnsrl.nxv8i32.nxv8i64.nxv8i32.i64(<vscale x 8 x i64> [[OP1:%.*]], <vscale x 8 x i32> [[SHIFT:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
+//
+vuint32m4_t test_vnsrl_wv_u32m4(vuint64m8_t op1, vuint32m4_t shift, size_t vl) {
+  return vnsrl(op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsrl_wx_u32m4(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vnsrl.nxv8i32.nxv8i64.i32.i32(<vscale x 8 x i64> [[OP1:%.*]], i32 [[SHIFT:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsrl_wx_u32m4(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vnsrl.nxv8i32.nxv8i64.i64.i64(<vscale x 8 x i64> [[OP1:%.*]], i64 [[SHIFT:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
+//
+vuint32m4_t test_vnsrl_wx_u32m4(vuint64m8_t op1, size_t shift, size_t vl) {
+  return vnsrl(op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsrl_wv_u8mf8_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i8> @llvm.riscv.vnsrl.mask.nxv1i8.nxv1i16.nxv1i8.i32(<vscale x 1 x i8> [[MASKEDOFF:%.*]], <vscale x 1 x i16> [[OP1:%.*]], <vscale x 1 x i8> [[SHIFT:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 1 x i8> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsrl_wv_u8mf8_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i8> @llvm.riscv.vnsrl.mask.nxv1i8.nxv1i16.nxv1i8.i64(<vscale x 1 x i8> [[MASKEDOFF:%.*]], <vscale x 1 x i16> [[OP1:%.*]], <vscale x 1 x i8> [[SHIFT:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 1 x i8> [[TMP0]]
+//
+vuint8mf8_t test_vnsrl_wv_u8mf8_m(vbool64_t mask, vuint8mf8_t maskedoff, vuint16mf4_t op1, vuint8mf8_t shift, size_t vl) {
+  return vnsrl(mask, maskedoff, op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsrl_wx_u8mf8_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i8> @llvm.riscv.vnsrl.mask.nxv1i8.nxv1i16.i32.i32(<vscale x 1 x i8> [[MASKEDOFF:%.*]], <vscale x 1 x i16> [[OP1:%.*]], i32 [[SHIFT:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 1 x i8> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsrl_wx_u8mf8_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i8> @llvm.riscv.vnsrl.mask.nxv1i8.nxv1i16.i64.i64(<vscale x 1 x i8> [[MASKEDOFF:%.*]], <vscale x 1 x i16> [[OP1:%.*]], i64 [[SHIFT:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 1 x i8> [[TMP0]]
+//
+vuint8mf8_t test_vnsrl_wx_u8mf8_m(vbool64_t mask, vuint8mf8_t maskedoff, vuint16mf4_t op1, size_t shift, size_t vl) {
+  return vnsrl(mask, maskedoff, op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsrl_wv_u8mf4_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i8> @llvm.riscv.vnsrl.mask.nxv2i8.nxv2i16.nxv2i8.i32(<vscale x 2 x i8> [[MASKEDOFF:%.*]], <vscale x 2 x i16> [[OP1:%.*]], <vscale x 2 x i8> [[SHIFT:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 2 x i8> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsrl_wv_u8mf4_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i8> @llvm.riscv.vnsrl.mask.nxv2i8.nxv2i16.nxv2i8.i64(<vscale x 2 x i8> [[MASKEDOFF:%.*]], <vscale x 2 x i16> [[OP1:%.*]], <vscale x 2 x i8> [[SHIFT:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 2 x i8> [[TMP0]]
+//
+vuint8mf4_t test_vnsrl_wv_u8mf4_m(vbool32_t mask, vuint8mf4_t maskedoff, vuint16mf2_t op1, vuint8mf4_t shift, size_t vl) {
+  return vnsrl(mask, maskedoff, op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsrl_wx_u8mf4_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i8> @llvm.riscv.vnsrl.mask.nxv2i8.nxv2i16.i32.i32(<vscale x 2 x i8> [[MASKEDOFF:%.*]], <vscale x 2 x i16> [[OP1:%.*]], i32 [[SHIFT:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 2 x i8> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsrl_wx_u8mf4_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i8> @llvm.riscv.vnsrl.mask.nxv2i8.nxv2i16.i64.i64(<vscale x 2 x i8> [[MASKEDOFF:%.*]], <vscale x 2 x i16> [[OP1:%.*]], i64 [[SHIFT:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 2 x i8> [[TMP0]]
+//
+vuint8mf4_t test_vnsrl_wx_u8mf4_m(vbool32_t mask, vuint8mf4_t maskedoff, vuint16mf2_t op1, size_t shift, size_t vl) {
+  return vnsrl(mask, maskedoff, op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsrl_wv_u8mf2_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i8> @llvm.riscv.vnsrl.mask.nxv4i8.nxv4i16.nxv4i8.i32(<vscale x 4 x i8> [[MASKEDOFF:%.*]], <vscale x 4 x i16> [[OP1:%.*]], <vscale x 4 x i8> [[SHIFT:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 4 x i8> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsrl_wv_u8mf2_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i8> @llvm.riscv.vnsrl.mask.nxv4i8.nxv4i16.nxv4i8.i64(<vscale x 4 x i8> [[MASKEDOFF:%.*]], <vscale x 4 x i16> [[OP1:%.*]], <vscale x 4 x i8> [[SHIFT:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 4 x i8> [[TMP0]]
+//
+vuint8mf2_t test_vnsrl_wv_u8mf2_m(vbool16_t mask, vuint8mf2_t maskedoff, vuint16m1_t op1, vuint8mf2_t shift, size_t vl) {
+  return vnsrl(mask, maskedoff, op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsrl_wx_u8mf2_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i8> @llvm.riscv.vnsrl.mask.nxv4i8.nxv4i16.i32.i32(<vscale x 4 x i8> [[MASKEDOFF:%.*]], <vscale x 4 x i16> [[OP1:%.*]], i32 [[SHIFT:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 4 x i8> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsrl_wx_u8mf2_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i8> @llvm.riscv.vnsrl.mask.nxv4i8.nxv4i16.i64.i64(<vscale x 4 x i8> [[MASKEDOFF:%.*]], <vscale x 4 x i16> [[OP1:%.*]], i64 [[SHIFT:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 4 x i8> [[TMP0]]
+//
+vuint8mf2_t test_vnsrl_wx_u8mf2_m(vbool16_t mask, vuint8mf2_t maskedoff, vuint16m1_t op1, size_t shift, size_t vl) {
+  return vnsrl(mask, maskedoff, op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsrl_wv_u8m1_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i8> @llvm.riscv.vnsrl.mask.nxv8i8.nxv8i16.nxv8i8.i32(<vscale x 8 x i8> [[MASKEDOFF:%.*]], <vscale x 8 x i16> [[OP1:%.*]], <vscale x 8 x i8> [[SHIFT:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 8 x i8> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsrl_wv_u8m1_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i8> @llvm.riscv.vnsrl.mask.nxv8i8.nxv8i16.nxv8i8.i64(<vscale x 8 x i8> [[MASKEDOFF:%.*]], <vscale x 8 x i16> [[OP1:%.*]], <vscale x 8 x i8> [[SHIFT:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 8 x i8> [[TMP0]]
+//
+vuint8m1_t test_vnsrl_wv_u8m1_m(vbool8_t mask, vuint8m1_t maskedoff, vuint16m2_t op1, vuint8m1_t shift, size_t vl) {
+  return vnsrl(mask, maskedoff, op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsrl_wx_u8m1_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i8> @llvm.riscv.vnsrl.mask.nxv8i8.nxv8i16.i32.i32(<vscale x 8 x i8> [[MASKEDOFF:%.*]], <vscale x 8 x i16> [[OP1:%.*]], i32 [[SHIFT:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 8 x i8> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsrl_wx_u8m1_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i8> @llvm.riscv.vnsrl.mask.nxv8i8.nxv8i16.i64.i64(<vscale x 8 x i8> [[MASKEDOFF:%.*]], <vscale x 8 x i16> [[OP1:%.*]], i64 [[SHIFT:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 8 x i8> [[TMP0]]
+//
+vuint8m1_t test_vnsrl_wx_u8m1_m(vbool8_t mask, vuint8m1_t maskedoff, vuint16m2_t op1, size_t shift, size_t vl) {
+  return vnsrl(mask, maskedoff, op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsrl_wv_u8m2_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i8> @llvm.riscv.vnsrl.mask.nxv16i8.nxv16i16.nxv16i8.i32(<vscale x 16 x i8> [[MASKEDOFF:%.*]], <vscale x 16 x i16> [[OP1:%.*]], <vscale x 16 x i8> [[SHIFT:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 16 x i8> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsrl_wv_u8m2_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i8> @llvm.riscv.vnsrl.mask.nxv16i8.nxv16i16.nxv16i8.i64(<vscale x 16 x i8> [[MASKEDOFF:%.*]], <vscale x 16 x i16> [[OP1:%.*]], <vscale x 16 x i8> [[SHIFT:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 16 x i8> [[TMP0]]
+//
+vuint8m2_t test_vnsrl_wv_u8m2_m(vbool4_t mask, vuint8m2_t maskedoff, vuint16m4_t op1, vuint8m2_t shift, size_t vl) {
+  return vnsrl(mask, maskedoff, op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsrl_wx_u8m2_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i8> @llvm.riscv.vnsrl.mask.nxv16i8.nxv16i16.i32.i32(<vscale x 16 x i8> [[MASKEDOFF:%.*]], <vscale x 16 x i16> [[OP1:%.*]], i32 [[SHIFT:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 16 x i8> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsrl_wx_u8m2_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i8> @llvm.riscv.vnsrl.mask.nxv16i8.nxv16i16.i64.i64(<vscale x 16 x i8> [[MASKEDOFF:%.*]], <vscale x 16 x i16> [[OP1:%.*]], i64 [[SHIFT:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 16 x i8> [[TMP0]]
+//
+vuint8m2_t test_vnsrl_wx_u8m2_m(vbool4_t mask, vuint8m2_t maskedoff, vuint16m4_t op1, size_t shift, size_t vl) {
+  return vnsrl(mask, maskedoff, op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsrl_wv_u8m4_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i8> @llvm.riscv.vnsrl.mask.nxv32i8.nxv32i16.nxv32i8.i32(<vscale x 32 x i8> [[MASKEDOFF:%.*]], <vscale x 32 x i16> [[OP1:%.*]], <vscale x 32 x i8> [[SHIFT:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 32 x i8> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsrl_wv_u8m4_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i8> @llvm.riscv.vnsrl.mask.nxv32i8.nxv32i16.nxv32i8.i64(<vscale x 32 x i8> [[MASKEDOFF:%.*]], <vscale x 32 x i16> [[OP1:%.*]], <vscale x 32 x i8> [[SHIFT:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 32 x i8> [[TMP0]]
+//
+vuint8m4_t test_vnsrl_wv_u8m4_m(vbool2_t mask, vuint8m4_t maskedoff, vuint16m8_t op1, vuint8m4_t shift, size_t vl) {
+  return vnsrl(mask, maskedoff, op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsrl_wx_u8m4_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i8> @llvm.riscv.vnsrl.mask.nxv32i8.nxv32i16.i32.i32(<vscale x 32 x i8> [[MASKEDOFF:%.*]], <vscale x 32 x i16> [[OP1:%.*]], i32 [[SHIFT:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 32 x i8> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsrl_wx_u8m4_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i8> @llvm.riscv.vnsrl.mask.nxv32i8.nxv32i16.i64.i64(<vscale x 32 x i8> [[MASKEDOFF:%.*]], <vscale x 32 x i16> [[OP1:%.*]], i64 [[SHIFT:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 32 x i8> [[TMP0]]
+//
+vuint8m4_t test_vnsrl_wx_u8m4_m(vbool2_t mask, vuint8m4_t maskedoff, vuint16m8_t op1, size_t shift, size_t vl) {
+  return vnsrl(mask, maskedoff, op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsrl_wv_u16mf4_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vnsrl.mask.nxv1i16.nxv1i32.nxv1i16.i32(<vscale x 1 x i16> [[MASKEDOFF:%.*]], <vscale x 1 x i32> [[OP1:%.*]], <vscale x 1 x i16> [[SHIFT:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 1 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsrl_wv_u16mf4_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vnsrl.mask.nxv1i16.nxv1i32.nxv1i16.i64(<vscale x 1 x i16> [[MASKEDOFF:%.*]], <vscale x 1 x i32> [[OP1:%.*]], <vscale x 1 x i16> [[SHIFT:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 1 x i16> [[TMP0]]
+//
+vuint16mf4_t test_vnsrl_wv_u16mf4_m(vbool64_t mask, vuint16mf4_t maskedoff, vuint32mf2_t op1, vuint16mf4_t shift, size_t vl) {
+  return vnsrl(mask, maskedoff, op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsrl_wx_u16mf4_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vnsrl.mask.nxv1i16.nxv1i32.i32.i32(<vscale x 1 x i16> [[MASKEDOFF:%.*]], <vscale x 1 x i32> [[OP1:%.*]], i32 [[SHIFT:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 1 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsrl_wx_u16mf4_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vnsrl.mask.nxv1i16.nxv1i32.i64.i64(<vscale x 1 x i16> [[MASKEDOFF:%.*]], <vscale x 1 x i32> [[OP1:%.*]], i64 [[SHIFT:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 1 x i16> [[TMP0]]
+//
+vuint16mf4_t test_vnsrl_wx_u16mf4_m(vbool64_t mask, vuint16mf4_t maskedoff, vuint32mf2_t op1, size_t shift, size_t vl) {
+  return vnsrl(mask, maskedoff, op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsrl_wv_u16mf2_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vnsrl.mask.nxv2i16.nxv2i32.nxv2i16.i32(<vscale x 2 x i16> [[MASKEDOFF:%.*]], <vscale x 2 x i32> [[OP1:%.*]], <vscale x 2 x i16> [[SHIFT:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 2 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsrl_wv_u16mf2_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vnsrl.mask.nxv2i16.nxv2i32.nxv2i16.i64(<vscale x 2 x i16> [[MASKEDOFF:%.*]], <vscale x 2 x i32> [[OP1:%.*]], <vscale x 2 x i16> [[SHIFT:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 2 x i16> [[TMP0]]
+//
+vuint16mf2_t test_vnsrl_wv_u16mf2_m(vbool32_t mask, vuint16mf2_t maskedoff, vuint32m1_t op1, vuint16mf2_t shift, size_t vl) {
+  return vnsrl(mask, maskedoff, op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsrl_wx_u16mf2_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vnsrl.mask.nxv2i16.nxv2i32.i32.i32(<vscale x 2 x i16> [[MASKEDOFF:%.*]], <vscale x 2 x i32> [[OP1:%.*]], i32 [[SHIFT:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 2 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsrl_wx_u16mf2_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vnsrl.mask.nxv2i16.nxv2i32.i64.i64(<vscale x 2 x i16> [[MASKEDOFF:%.*]], <vscale x 2 x i32> [[OP1:%.*]], i64 [[SHIFT:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 2 x i16> [[TMP0]]
+//
+vuint16mf2_t test_vnsrl_wx_u16mf2_m(vbool32_t mask, vuint16mf2_t maskedoff, vuint32m1_t op1, size_t shift, size_t vl) {
+  return vnsrl(mask, maskedoff, op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsrl_wv_u16m1_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vnsrl.mask.nxv4i16.nxv4i32.nxv4i16.i32(<vscale x 4 x i16> [[MASKEDOFF:%.*]], <vscale x 4 x i32> [[OP1:%.*]], <vscale x 4 x i16> [[SHIFT:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 4 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsrl_wv_u16m1_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vnsrl.mask.nxv4i16.nxv4i32.nxv4i16.i64(<vscale x 4 x i16> [[MASKEDOFF:%.*]], <vscale x 4 x i32> [[OP1:%.*]], <vscale x 4 x i16> [[SHIFT:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 4 x i16> [[TMP0]]
+//
+vuint16m1_t test_vnsrl_wv_u16m1_m(vbool16_t mask, vuint16m1_t maskedoff, vuint32m2_t op1, vuint16m1_t shift, size_t vl) {
+  return vnsrl(mask, maskedoff, op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsrl_wx_u16m1_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vnsrl.mask.nxv4i16.nxv4i32.i32.i32(<vscale x 4 x i16> [[MASKEDOFF:%.*]], <vscale x 4 x i32> [[OP1:%.*]], i32 [[SHIFT:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 4 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsrl_wx_u16m1_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vnsrl.mask.nxv4i16.nxv4i32.i64.i64(<vscale x 4 x i16> [[MASKEDOFF:%.*]], <vscale x 4 x i32> [[OP1:%.*]], i64 [[SHIFT:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 4 x i16> [[TMP0]]
+//
+vuint16m1_t test_vnsrl_wx_u16m1_m(vbool16_t mask, vuint16m1_t maskedoff, vuint32m2_t op1, size_t shift, size_t vl) {
+  return vnsrl(mask, maskedoff, op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsrl_wv_u16m2_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vnsrl.mask.nxv8i16.nxv8i32.nxv8i16.i32(<vscale x 8 x i16> [[MASKEDOFF:%.*]], <vscale x 8 x i32> [[OP1:%.*]], <vscale x 8 x i16> [[SHIFT:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsrl_wv_u16m2_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vnsrl.mask.nxv8i16.nxv8i32.nxv8i16.i64(<vscale x 8 x i16> [[MASKEDOFF:%.*]], <vscale x 8 x i32> [[OP1:%.*]], <vscale x 8 x i16> [[SHIFT:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
+//
+vuint16m2_t test_vnsrl_wv_u16m2_m(vbool8_t mask, vuint16m2_t maskedoff, vuint32m4_t op1, vuint16m2_t shift, size_t vl) {
+  return vnsrl(mask, maskedoff, op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsrl_wx_u16m2_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vnsrl.mask.nxv8i16.nxv8i32.i32.i32(<vscale x 8 x i16> [[MASKEDOFF:%.*]], <vscale x 8 x i32> [[OP1:%.*]], i32 [[SHIFT:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsrl_wx_u16m2_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vnsrl.mask.nxv8i16.nxv8i32.i64.i64(<vscale x 8 x i16> [[MASKEDOFF:%.*]], <vscale x 8 x i32> [[OP1:%.*]], i64 [[SHIFT:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
+//
+vuint16m2_t test_vnsrl_wx_u16m2_m(vbool8_t mask, vuint16m2_t maskedoff, vuint32m4_t op1, size_t shift, size_t vl) {
+  return vnsrl(mask, maskedoff, op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsrl_wv_u16m4_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vnsrl.mask.nxv16i16.nxv16i32.nxv16i16.i32(<vscale x 16 x i16> [[MASKEDOFF:%.*]], <vscale x 16 x i32> [[OP1:%.*]], <vscale x 16 x i16> [[SHIFT:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 16 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsrl_wv_u16m4_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vnsrl.mask.nxv16i16.nxv16i32.nxv16i16.i64(<vscale x 16 x i16> [[MASKEDOFF:%.*]], <vscale x 16 x i32> [[OP1:%.*]], <vscale x 16 x i16> [[SHIFT:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 16 x i16> [[TMP0]]
+//
+vuint16m4_t test_vnsrl_wv_u16m4_m(vbool4_t mask, vuint16m4_t maskedoff, vuint32m8_t op1, vuint16m4_t shift, size_t vl) {
+  return vnsrl(mask, maskedoff, op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsrl_wx_u16m4_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vnsrl.mask.nxv16i16.nxv16i32.i32.i32(<vscale x 16 x i16> [[MASKEDOFF:%.*]], <vscale x 16 x i32> [[OP1:%.*]], i32 [[SHIFT:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 16 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsrl_wx_u16m4_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vnsrl.mask.nxv16i16.nxv16i32.i64.i64(<vscale x 16 x i16> [[MASKEDOFF:%.*]], <vscale x 16 x i32> [[OP1:%.*]], i64 [[SHIFT:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 16 x i16> [[TMP0]]
+//
+vuint16m4_t test_vnsrl_wx_u16m4_m(vbool4_t mask, vuint16m4_t maskedoff, vuint32m8_t op1, size_t shift, size_t vl) {
+  return vnsrl(mask, maskedoff, op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsrl_wv_u32mf2_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vnsrl.mask.nxv1i32.nxv1i64.nxv1i32.i32(<vscale x 1 x i32> [[MASKEDOFF:%.*]], <vscale x 1 x i64> [[OP1:%.*]], <vscale x 1 x i32> [[SHIFT:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsrl_wv_u32mf2_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vnsrl.mask.nxv1i32.nxv1i64.nxv1i32.i64(<vscale x 1 x i32> [[MASKEDOFF:%.*]], <vscale x 1 x i64> [[OP1:%.*]], <vscale x 1 x i32> [[SHIFT:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
+//
+vuint32mf2_t test_vnsrl_wv_u32mf2_m(vbool64_t mask, vuint32mf2_t maskedoff, vuint64m1_t op1, vuint32mf2_t shift, size_t vl) {
+  return vnsrl(mask, maskedoff, op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsrl_wx_u32mf2_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vnsrl.mask.nxv1i32.nxv1i64.i32.i32(<vscale x 1 x i32> [[MASKEDOFF:%.*]], <vscale x 1 x i64> [[OP1:%.*]], i32 [[SHIFT:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsrl_wx_u32mf2_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vnsrl.mask.nxv1i32.nxv1i64.i64.i64(<vscale x 1 x i32> [[MASKEDOFF:%.*]], <vscale x 1 x i64> [[OP1:%.*]], i64 [[SHIFT:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
+//
+vuint32mf2_t test_vnsrl_wx_u32mf2_m(vbool64_t mask, vuint32mf2_t maskedoff, vuint64m1_t op1, size_t shift, size_t vl) {
+  return vnsrl(mask, maskedoff, op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsrl_wv_u32m1_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vnsrl.mask.nxv2i32.nxv2i64.nxv2i32.i32(<vscale x 2 x i32> [[MASKEDOFF:%.*]], <vscale x 2 x i64> [[OP1:%.*]], <vscale x 2 x i32> [[SHIFT:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsrl_wv_u32m1_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vnsrl.mask.nxv2i32.nxv2i64.nxv2i32.i64(<vscale x 2 x i32> [[MASKEDOFF:%.*]], <vscale x 2 x i64> [[OP1:%.*]], <vscale x 2 x i32> [[SHIFT:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
+//
+vuint32m1_t test_vnsrl_wv_u32m1_m(vbool32_t mask, vuint32m1_t maskedoff, vuint64m2_t op1, vuint32m1_t shift, size_t vl) {
+  return vnsrl(mask, maskedoff, op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsrl_wx_u32m1_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vnsrl.mask.nxv2i32.nxv2i64.i32.i32(<vscale x 2 x i32> [[MASKEDOFF:%.*]], <vscale x 2 x i64> [[OP1:%.*]], i32 [[SHIFT:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsrl_wx_u32m1_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vnsrl.mask.nxv2i32.nxv2i64.i64.i64(<vscale x 2 x i32> [[MASKEDOFF:%.*]], <vscale x 2 x i64> [[OP1:%.*]], i64 [[SHIFT:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
+//
+vuint32m1_t test_vnsrl_wx_u32m1_m(vbool32_t mask, vuint32m1_t maskedoff, vuint64m2_t op1, size_t shift, size_t vl) {
+  return vnsrl(mask, maskedoff, op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsrl_wv_u32m2_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vnsrl.mask.nxv4i32.nxv4i64.nxv4i32.i32(<vscale x 4 x i32> [[MASKEDOFF:%.*]], <vscale x 4 x i64> [[OP1:%.*]], <vscale x 4 x i32> [[SHIFT:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsrl_wv_u32m2_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vnsrl.mask.nxv4i32.nxv4i64.nxv4i32.i64(<vscale x 4 x i32> [[MASKEDOFF:%.*]], <vscale x 4 x i64> [[OP1:%.*]], <vscale x 4 x i32> [[SHIFT:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
+//
+vuint32m2_t test_vnsrl_wv_u32m2_m(vbool16_t mask, vuint32m2_t maskedoff, vuint64m4_t op1, vuint32m2_t shift, size_t vl) {
+  return vnsrl(mask, maskedoff, op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsrl_wx_u32m2_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vnsrl.mask.nxv4i32.nxv4i64.i32.i32(<vscale x 4 x i32> [[MASKEDOFF:%.*]], <vscale x 4 x i64> [[OP1:%.*]], i32 [[SHIFT:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsrl_wx_u32m2_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vnsrl.mask.nxv4i32.nxv4i64.i64.i64(<vscale x 4 x i32> [[MASKEDOFF:%.*]], <vscale x 4 x i64> [[OP1:%.*]], i64 [[SHIFT:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
+//
+vuint32m2_t test_vnsrl_wx_u32m2_m(vbool16_t mask, vuint32m2_t maskedoff, vuint64m4_t op1, size_t shift, size_t vl) {
+  return vnsrl(mask, maskedoff, op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsrl_wv_u32m4_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vnsrl.mask.nxv8i32.nxv8i64.nxv8i32.i32(<vscale x 8 x i32> [[MASKEDOFF:%.*]], <vscale x 8 x i64> [[OP1:%.*]], <vscale x 8 x i32> [[SHIFT:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsrl_wv_u32m4_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vnsrl.mask.nxv8i32.nxv8i64.nxv8i32.i64(<vscale x 8 x i32> [[MASKEDOFF:%.*]], <vscale x 8 x i64> [[OP1:%.*]], <vscale x 8 x i32> [[SHIFT:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
+//
+vuint32m4_t test_vnsrl_wv_u32m4_m(vbool8_t mask, vuint32m4_t maskedoff, vuint64m8_t op1, vuint32m4_t shift, size_t vl) {
+  return vnsrl(mask, maskedoff, op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsrl_wx_u32m4_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vnsrl.mask.nxv8i32.nxv8i64.i32.i32(<vscale x 8 x i32> [[MASKEDOFF:%.*]], <vscale x 8 x i64> [[OP1:%.*]], i32 [[SHIFT:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsrl_wx_u32m4_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vnsrl.mask.nxv8i32.nxv8i64.i64.i64(<vscale x 8 x i32> [[MASKEDOFF:%.*]], <vscale x 8 x i64> [[OP1:%.*]], i64 [[SHIFT:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
+//
+vuint32m4_t test_vnsrl_wx_u32m4_m(vbool8_t mask, vuint32m4_t maskedoff, vuint64m8_t op1, size_t shift, size_t vl) {
+  return vnsrl(mask, maskedoff, op1, shift, vl);
+}

clang/test/CodeGen/RISCV/rvv-intrinsics-overloaded/vwmul.c

+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv32 -target-feature +experimental-v -disable-O0-optnone -emit-llvm %s -o - | opt -S -mem2reg | FileCheck --check-prefix=CHECK-RV32 %s
+// RUN: %clang_cc1 -triple riscv64 -target-feature +experimental-v -disable-O0-optnone -emit-llvm %s -o - | opt -S -mem2reg | FileCheck --check-prefix=CHECK-RV64 %s
+// RUN: %clang_cc1 -triple riscv64 -target-feature +m -target-feature +experimental-v -Werror -Wall -o - %s -S >/dev/null 2>&1 | FileCheck --check-prefix=ASM --allow-empty %s
+
+// ASM-NOT: warning
+#include <riscv_vector.h>
+
+// CHECK-RV32-LABEL: @test_vwmul_vv_i16mf4(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vwmul.nxv1i16.nxv1i8.nxv1i8.i32(<vscale x 1 x i8> [[OP1:%.*]], <vscale x 1 x i8> [[OP2:%.*]], i32 [[VL:%.*]]) [[ATTR7:#.*]]
+// CHECK-RV32-NEXT:    ret <vscale x 1 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmul_vv_i16mf4(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vwmul.nxv1i16.nxv1i8.nxv1i8.i64(<vscale x 1 x i8> [[OP1:%.*]], <vscale x 1 x i8> [[OP2:%.*]], i64 [[VL:%.*]]) [[ATTR7:#.*]]
+// CHECK-RV64-NEXT:    ret <vscale x 1 x i16> [[TMP0]]
+//
+vint16mf4_t test_vwmul_vv_i16mf4(vint8mf8_t op1, vint8mf8_t op2, size_t vl) {
+  return vwmul(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmul_vx_i16mf4(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vwmul.nxv1i16.nxv1i8.i8.i32(<vscale x 1 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 1 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmul_vx_i16mf4(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vwmul.nxv1i16.nxv1i8.i8.i64(<vscale x 1 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 1 x i16> [[TMP0]]
+//
+vint16mf4_t test_vwmul_vx_i16mf4(vint8mf8_t op1, int8_t op2, size_t vl) {
+  return vwmul(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmul_vv_i16mf2(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vwmul.nxv2i16.nxv2i8.nxv2i8.i32(<vscale x 2 x i8> [[OP1:%.*]], <vscale x 2 x i8> [[OP2:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 2 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmul_vv_i16mf2(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vwmul.nxv2i16.nxv2i8.nxv2i8.i64(<vscale x 2 x i8> [[OP1:%.*]], <vscale x 2 x i8> [[OP2:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 2 x i16> [[TMP0]]
+//
+vint16mf2_t test_vwmul_vv_i16mf2(vint8mf4_t op1, vint8mf4_t op2, size_t vl) {
+  return vwmul(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmul_vx_i16mf2(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vwmul.nxv2i16.nxv2i8.i8.i32(<vscale x 2 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 2 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmul_vx_i16mf2(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vwmul.nxv2i16.nxv2i8.i8.i64(<vscale x 2 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 2 x i16> [[TMP0]]
+//
+vint16mf2_t test_vwmul_vx_i16mf2(vint8mf4_t op1, int8_t op2, size_t vl) {
+  return vwmul(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmul_vv_i16m1(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vwmul.nxv4i16.nxv4i8.nxv4i8.i32(<vscale x 4 x i8> [[OP1:%.*]], <vscale x 4 x i8> [[OP2:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 4 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmul_vv_i16m1(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vwmul.nxv4i16.nxv4i8.nxv4i8.i64(<vscale x 4 x i8> [[OP1:%.*]], <vscale x 4 x i8> [[OP2:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 4 x i16> [[TMP0]]
+//
+vint16m1_t test_vwmul_vv_i16m1(vint8mf2_t op1, vint8mf2_t op2, size_t vl) {
+  return vwmul(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmul_vx_i16m1(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vwmul.nxv4i16.nxv4i8.i8.i32(<vscale x 4 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 4 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmul_vx_i16m1(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vwmul.nxv4i16.nxv4i8.i8.i64(<vscale x 4 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 4 x i16> [[TMP0]]
+//
+vint16m1_t test_vwmul_vx_i16m1(vint8mf2_t op1, int8_t op2, size_t vl) {
+  return vwmul(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmul_vv_i16m2(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vwmul.nxv8i16.nxv8i8.nxv8i8.i32(<vscale x 8 x i8> [[OP1:%.*]], <vscale x 8 x i8> [[OP2:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmul_vv_i16m2(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vwmul.nxv8i16.nxv8i8.nxv8i8.i64(<vscale x 8 x i8> [[OP1:%.*]], <vscale x 8 x i8> [[OP2:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
+//
+vint16m2_t test_vwmul_vv_i16m2(vint8m1_t op1, vint8m1_t op2, size_t vl) {
+  return vwmul(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmul_vx_i16m2(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vwmul.nxv8i16.nxv8i8.i8.i32(<vscale x 8 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmul_vx_i16m2(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vwmul.nxv8i16.nxv8i8.i8.i64(<vscale x 8 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
+//
+vint16m2_t test_vwmul_vx_i16m2(vint8m1_t op1, int8_t op2, size_t vl) {
+  return vwmul(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmul_vv_i16m4(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vwmul.nxv16i16.nxv16i8.nxv16i8.i32(<vscale x 16 x i8> [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 16 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmul_vv_i16m4(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vwmul.nxv16i16.nxv16i8.nxv16i8.i64(<vscale x 16 x i8> [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 16 x i16> [[TMP0]]
+//
+vint16m4_t test_vwmul_vv_i16m4(vint8m2_t op1, vint8m2_t op2, size_t vl) {
+  return vwmul(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmul_vx_i16m4(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vwmul.nxv16i16.nxv16i8.i8.i32(<vscale x 16 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 16 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmul_vx_i16m4(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vwmul.nxv16i16.nxv16i8.i8.i64(<vscale x 16 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 16 x i16> [[TMP0]]
+//
+vint16m4_t test_vwmul_vx_i16m4(vint8m2_t op1, int8_t op2, size_t vl) {
+  return vwmul(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmul_vv_i16m8(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vwmul.nxv32i16.nxv32i8.nxv32i8.i32(<vscale x 32 x i8> [[OP1:%.*]], <vscale x 32 x i8> [[OP2:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 32 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmul_vv_i16m8(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vwmul.nxv32i16.nxv32i8.nxv32i8.i64(<vscale x 32 x i8> [[OP1:%.*]], <vscale x 32 x i8> [[OP2:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 32 x i16> [[TMP0]]
+//
+vint16m8_t test_vwmul_vv_i16m8(vint8m4_t op1, vint8m4_t op2, size_t vl) {
+  return vwmul(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmul_vx_i16m8(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vwmul.nxv32i16.nxv32i8.i8.i32(<vscale x 32 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 32 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmul_vx_i16m8(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vwmul.nxv32i16.nxv32i8.i8.i64(<vscale x 32 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 32 x i16> [[TMP0]]
+//
+vint16m8_t test_vwmul_vx_i16m8(vint8m4_t op1, int8_t op2, size_t vl) {
+  return vwmul(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmul_vv_i32mf2(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vwmul.nxv1i32.nxv1i16.nxv1i16.i32(<vscale x 1 x i16> [[OP1:%.*]], <vscale x 1 x i16> [[OP2:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmul_vv_i32mf2(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vwmul.nxv1i32.nxv1i16.nxv1i16.i64(<vscale x 1 x i16> [[OP1:%.*]], <vscale x 1 x i16> [[OP2:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
+//
+vint32mf2_t test_vwmul_vv_i32mf2(vint16mf4_t op1, vint16mf4_t op2, size_t vl) {
+  return vwmul(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmul_vx_i32mf2(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vwmul.nxv1i32.nxv1i16.i16.i32(<vscale x 1 x i16> [[OP1:%.*]], i16 [[OP2:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmul_vx_i32mf2(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vwmul.nxv1i32.nxv1i16.i16.i64(<vscale x 1 x i16> [[OP1:%.*]], i16 [[OP2:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
+//
+vint32mf2_t test_vwmul_vx_i32mf2(vint16mf4_t op1, int16_t op2, size_t vl) {
+  return vwmul(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmul_vv_i32m1(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vwmul.nxv2i32.nxv2i16.nxv2i16.i32(<vscale x 2 x i16> [[OP1:%.*]], <vscale x 2 x i16> [[OP2:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmul_vv_i32m1(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vwmul.nxv2i32.nxv2i16.nxv2i16.i64(<vscale x 2 x i16> [[OP1:%.*]], <vscale x 2 x i16> [[OP2:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
+//
+vint32m1_t test_vwmul_vv_i32m1(vint16mf2_t op1, vint16mf2_t op2, size_t vl) {
+  return vwmul(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmul_vx_i32m1(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vwmul.nxv2i32.nxv2i16.i16.i32(<vscale x 2 x i16> [[OP1:%.*]], i16 [[OP2:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmul_vx_i32m1(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vwmul.nxv2i32.nxv2i16.i16.i64(<vscale x 2 x i16> [[OP1:%.*]], i16 [[OP2:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
+//
+vint32m1_t test_vwmul_vx_i32m1(vint16mf2_t op1, int16_t op2, size_t vl) {
+  return vwmul(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmul_vv_i32m2(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vwmul.nxv4i32.nxv4i16.nxv4i16.i32(<vscale x 4 x i16> [[OP1:%.*]], <vscale x 4 x i16> [[OP2:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmul_vv_i32m2(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vwmul.nxv4i32.nxv4i16.nxv4i16.i64(<vscale x 4 x i16> [[OP1:%.*]], <vscale x 4 x i16> [[OP2:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
+//
+vint32m2_t test_vwmul_vv_i32m2(vint16m1_t op1, vint16m1_t op2, size_t vl) {
+  return vwmul(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmul_vx_i32m2(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vwmul.nxv4i32.nxv4i16.i16.i32(<vscale x 4 x i16> [[OP1:%.*]], i16 [[OP2:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmul_vx_i32m2(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vwmul.nxv4i32.nxv4i16.i16.i64(<vscale x 4 x i16> [[OP1:%.*]], i16 [[OP2:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
+//
+vint32m2_t test_vwmul_vx_i32m2(vint16m1_t op1, int16_t op2, size_t vl) {
+  return vwmul(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmul_vv_i32m4(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vwmul.nxv8i32.nxv8i16.nxv8i16.i32(<vscale x 8 x i16> [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmul_vv_i32m4(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vwmul.nxv8i32.nxv8i16.nxv8i16.i64(<vscale x 8 x i16> [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
+//
+vint32m4_t test_vwmul_vv_i32m4(vint16m2_t op1, vint16m2_t op2, size_t vl) {
+  return vwmul(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmul_vx_i32m4(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vwmul.nxv8i32.nxv8i16.i16.i32(<vscale x 8 x i16> [[OP1:%.*]], i16 [[OP2:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmul_vx_i32m4(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vwmul.nxv8i32.nxv8i16.i16.i64(<vscale x 8 x i16> [[OP1:%.*]], i16 [[OP2:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
+//
+vint32m4_t test_vwmul_vx_i32m4(vint16m2_t op1, int16_t op2, size_t vl) {
+  return vwmul(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmul_vv_i32m8(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vwmul.nxv16i32.nxv16i16.nxv16i16.i32(<vscale x 16 x i16> [[OP1:%.*]], <vscale x 16 x i16> [[OP2:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmul_vv_i32m8(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vwmul.nxv16i32.nxv16i16.nxv16i16.i64(<vscale x 16 x i16> [[OP1:%.*]], <vscale x 16 x i16> [[OP2:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
+//
+vint32m8_t test_vwmul_vv_i32m8(vint16m4_t op1, vint16m4_t op2, size_t vl) {
+  return vwmul(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmul_vx_i32m8(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vwmul.nxv16i32.nxv16i16.i16.i32(<vscale x 16 x i16> [[OP1:%.*]], i16 [[OP2:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmul_vx_i32m8(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vwmul.nxv16i32.nxv16i16.i16.i64(<vscale x 16 x i16> [[OP1:%.*]], i16 [[OP2:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
+//
+vint32m8_t test_vwmul_vx_i32m8(vint16m4_t op1, int16_t op2, size_t vl) {
+  return vwmul(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmul_vv_i64m1(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vwmul.nxv1i64.nxv1i32.nxv1i32.i32(<vscale x 1 x i32> [[OP1:%.*]], <vscale x 1 x i32> [[OP2:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 1 x i64> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmul_vv_i64m1(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vwmul.nxv1i64.nxv1i32.nxv1i32.i64(<vscale x 1 x i32> [[OP1:%.*]], <vscale x 1 x i32> [[OP2:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 1 x i64> [[TMP0]]
+//
+vint64m1_t test_vwmul_vv_i64m1(vint32mf2_t op1, vint32mf2_t op2, size_t vl) {
+  return vwmul(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmul_vx_i64m1(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vwmul.nxv1i64.nxv1i32.i32.i32(<vscale x 1 x i32> [[OP1:%.*]], i32 [[OP2:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 1 x i64> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmul_vx_i64m1(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vwmul.nxv1i64.nxv1i32.i32.i64(<vscale x 1 x i32> [[OP1:%.*]], i32 [[OP2:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 1 x i64> [[TMP0]]
+//
+vint64m1_t test_vwmul_vx_i64m1(vint32mf2_t op1, int32_t op2, size_t vl) {
+  return vwmul(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmul_vv_i64m2(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vwmul.nxv2i64.nxv2i32.nxv2i32.i32(<vscale x 2 x i32> [[OP1:%.*]], <vscale x 2 x i32> [[OP2:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 2 x i64> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmul_vv_i64m2(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vwmul.nxv2i64.nxv2i32.nxv2i32.i64(<vscale x 2 x i32> [[OP1:%.*]], <vscale x 2 x i32> [[OP2:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 2 x i64> [[TMP0]]
+//
+vint64m2_t test_vwmul_vv_i64m2(vint32m1_t op1, vint32m1_t op2, size_t vl) {
+  return vwmul(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmul_vx_i64m2(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vwmul.nxv2i64.nxv2i32.i32.i32(<vscale x 2 x i32> [[OP1:%.*]], i32 [[OP2:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 2 x i64> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmul_vx_i64m2(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vwmul.nxv2i64.nxv2i32.i32.i64(<vscale x 2 x i32> [[OP1:%.*]], i32 [[OP2:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 2 x i64> [[TMP0]]
+//
+vint64m2_t test_vwmul_vx_i64m2(vint32m1_t op1, int32_t op2, size_t vl) {
+  return vwmul(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmul_vv_i64m4(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vwmul.nxv4i64.nxv4i32.nxv4i32.i32(<vscale x 4 x i32> [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 4 x i64> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmul_vv_i64m4(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vwmul.nxv4i64.nxv4i32.nxv4i32.i64(<vscale x 4 x i32> [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 4 x i64> [[TMP0]]
+//
+vint64m4_t test_vwmul_vv_i64m4(vint32m2_t op1, vint32m2_t op2, size_t vl) {
+  return vwmul(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmul_vx_i64m4(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vwmul.nxv4i64.nxv4i32.i32.i32(<vscale x 4 x i32> [[OP1:%.*]], i32 [[OP2:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 4 x i64> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmul_vx_i64m4(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vwmul.nxv4i64.nxv4i32.i32.i64(<vscale x 4 x i32> [[OP1:%.*]], i32 [[OP2:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 4 x i64> [[TMP0]]
+//
+vint64m4_t test_vwmul_vx_i64m4(vint32m2_t op1, int32_t op2, size_t vl) {
+  return vwmul(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmul_vv_i64m8(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vwmul.nxv8i64.nxv8i32.nxv8i32.i32(<vscale x 8 x i32> [[OP1:%.*]], <vscale x 8 x i32> [[OP2:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 8 x i64> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmul_vv_i64m8(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vwmul.nxv8i64.nxv8i32.nxv8i32.i64(<vscale x 8 x i32> [[OP1:%.*]], <vscale x 8 x i32> [[OP2:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 8 x i64> [[TMP0]]
+//
+vint64m8_t test_vwmul_vv_i64m8(vint32m4_t op1, vint32m4_t op2, size_t vl) {
+  return vwmul(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmul_vx_i64m8(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vwmul.nxv8i64.nxv8i32.i32.i32(<vscale x 8 x i32> [[OP1:%.*]], i32 [[OP2:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 8 x i64> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmul_vx_i64m8(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vwmul.nxv8i64.nxv8i32.i32.i64(<vscale x 8 x i32> [[OP1:%.*]], i32 [[OP2:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 8 x i64> [[TMP0]]
+//
+vint64m8_t test_vwmul_vx_i64m8(vint32m4_t op1, int32_t op2, size_t vl) {
+  return vwmul(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulu_vv_u16mf4(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vwmulu.nxv1i16.nxv1i8.nxv1i8.i32(<vscale x 1 x i8> [[OP1:%.*]], <vscale x 1 x i8> [[OP2:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 1 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulu_vv_u16mf4(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vwmulu.nxv1i16.nxv1i8.nxv1i8.i64(<vscale x 1 x i8> [[OP1:%.*]], <vscale x 1 x i8> [[OP2:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 1 x i16> [[TMP0]]
+//
+vuint16mf4_t test_vwmulu_vv_u16mf4(vuint8mf8_t op1, vuint8mf8_t op2, size_t vl) {
+  return vwmulu(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulu_vx_u16mf4(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vwmulu.nxv1i16.nxv1i8.i8.i32(<vscale x 1 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 1 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulu_vx_u16mf4(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vwmulu.nxv1i16.nxv1i8.i8.i64(<vscale x 1 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 1 x i16> [[TMP0]]
+//
+vuint16mf4_t test_vwmulu_vx_u16mf4(vuint8mf8_t op1, uint8_t op2, size_t vl) {
+  return vwmulu(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulu_vv_u16mf2(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vwmulu.nxv2i16.nxv2i8.nxv2i8.i32(<vscale x 2 x i8> [[OP1:%.*]], <vscale x 2 x i8> [[OP2:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 2 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulu_vv_u16mf2(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vwmulu.nxv2i16.nxv2i8.nxv2i8.i64(<vscale x 2 x i8> [[OP1:%.*]], <vscale x 2 x i8> [[OP2:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 2 x i16> [[TMP0]]
+//
+vuint16mf2_t test_vwmulu_vv_u16mf2(vuint8mf4_t op1, vuint8mf4_t op2, size_t vl) {
+  return vwmulu(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulu_vx_u16mf2(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vwmulu.nxv2i16.nxv2i8.i8.i32(<vscale x 2 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 2 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulu_vx_u16mf2(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vwmulu.nxv2i16.nxv2i8.i8.i64(<vscale x 2 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 2 x i16> [[TMP0]]
+//
+vuint16mf2_t test_vwmulu_vx_u16mf2(vuint8mf4_t op1, uint8_t op2, size_t vl) {
+  return vwmulu(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulu_vv_u16m1(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vwmulu.nxv4i16.nxv4i8.nxv4i8.i32(<vscale x 4 x i8> [[OP1:%.*]], <vscale x 4 x i8> [[OP2:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 4 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulu_vv_u16m1(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vwmulu.nxv4i16.nxv4i8.nxv4i8.i64(<vscale x 4 x i8> [[OP1:%.*]], <vscale x 4 x i8> [[OP2:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 4 x i16> [[TMP0]]
+//
+vuint16m1_t test_vwmulu_vv_u16m1(vuint8mf2_t op1, vuint8mf2_t op2, size_t vl) {
+  return vwmulu(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulu_vx_u16m1(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vwmulu.nxv4i16.nxv4i8.i8.i32(<vscale x 4 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 4 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulu_vx_u16m1(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vwmulu.nxv4i16.nxv4i8.i8.i64(<vscale x 4 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 4 x i16> [[TMP0]]
+//
+vuint16m1_t test_vwmulu_vx_u16m1(vuint8mf2_t op1, uint8_t op2, size_t vl) {
+  return vwmulu(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulu_vv_u16m2(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vwmulu.nxv8i16.nxv8i8.nxv8i8.i32(<vscale x 8 x i8> [[OP1:%.*]], <vscale x 8 x i8> [[OP2:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulu_vv_u16m2(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vwmulu.nxv8i16.nxv8i8.nxv8i8.i64(<vscale x 8 x i8> [[OP1:%.*]], <vscale x 8 x i8> [[OP2:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
+//
+vuint16m2_t test_vwmulu_vv_u16m2(vuint8m1_t op1, vuint8m1_t op2, size_t vl) {
+  return vwmulu(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulu_vx_u16m2(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vwmulu.nxv8i16.nxv8i8.i8.i32(<vscale x 8 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulu_vx_u16m2(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vwmulu.nxv8i16.nxv8i8.i8.i64(<vscale x 8 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
+//
+vuint16m2_t test_vwmulu_vx_u16m2(vuint8m1_t op1, uint8_t op2, size_t vl) {
+  return vwmulu(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulu_vv_u16m4(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vwmulu.nxv16i16.nxv16i8.nxv16i8.i32(<vscale x 16 x i8> [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 16 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulu_vv_u16m4(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vwmulu.nxv16i16.nxv16i8.nxv16i8.i64(<vscale x 16 x i8> [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 16 x i16> [[TMP0]]
+//
+vuint16m4_t test_vwmulu_vv_u16m4(vuint8m2_t op1, vuint8m2_t op2, size_t vl) {
+  return vwmulu(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulu_vx_u16m4(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vwmulu.nxv16i16.nxv16i8.i8.i32(<vscale x 16 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 16 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulu_vx_u16m4(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vwmulu.nxv16i16.nxv16i8.i8.i64(<vscale x 16 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 16 x i16> [[TMP0]]
+//
+vuint16m4_t test_vwmulu_vx_u16m4(vuint8m2_t op1, uint8_t op2, size_t vl) {
+  return vwmulu(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulu_vv_u16m8(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vwmulu.nxv32i16.nxv32i8.nxv32i8.i32(<vscale x 32 x i8> [[OP1:%.*]], <vscale x 32 x i8> [[OP2:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 32 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulu_vv_u16m8(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vwmulu.nxv32i16.nxv32i8.nxv32i8.i64(<vscale x 32 x i8> [[OP1:%.*]], <vscale x 32 x i8> [[OP2:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 32 x i16> [[TMP0]]
+//
+vuint16m8_t test_vwmulu_vv_u16m8(vuint8m4_t op1, vuint8m4_t op2, size_t vl) {
+  return vwmulu(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulu_vx_u16m8(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vwmulu.nxv32i16.nxv32i8.i8.i32(<vscale x 32 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 32 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulu_vx_u16m8(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vwmulu.nxv32i16.nxv32i8.i8.i64(<vscale x 32 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 32 x i16> [[TMP0]]
+//
+vuint16m8_t test_vwmulu_vx_u16m8(vuint8m4_t op1, uint8_t op2, size_t vl) {
+  return vwmulu(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulu_vv_u32mf2(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vwmulu.nxv1i32.nxv1i16.nxv1i16.i32(<vscale x 1 x i16> [[OP1:%.*]], <vscale x 1 x i16> [[OP2:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulu_vv_u32mf2(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vwmulu.nxv1i32.nxv1i16.nxv1i16.i64(<vscale x 1 x i16> [[OP1:%.*]], <vscale x 1 x i16> [[OP2:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
+//
+vuint32mf2_t test_vwmulu_vv_u32mf2(vuint16mf4_t op1, vuint16mf4_t op2, size_t vl) {
+  return vwmulu(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulu_vx_u32mf2(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vwmulu.nxv1i32.nxv1i16.i16.i32(<vscale x 1 x i16> [[OP1:%.*]], i16 [[OP2:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulu_vx_u32mf2(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vwmulu.nxv1i32.nxv1i16.i16.i64(<vscale x 1 x i16> [[OP1:%.*]], i16 [[OP2:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
+//
+vuint32mf2_t test_vwmulu_vx_u32mf2(vuint16mf4_t op1, uint16_t op2, size_t vl) {
+  return vwmulu(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulu_vv_u32m1(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vwmulu.nxv2i32.nxv2i16.nxv2i16.i32(<vscale x 2 x i16> [[OP1:%.*]], <vscale x 2 x i16> [[OP2:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulu_vv_u32m1(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vwmulu.nxv2i32.nxv2i16.nxv2i16.i64(<vscale x 2 x i16> [[OP1:%.*]], <vscale x 2 x i16> [[OP2:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
+//
+vuint32m1_t test_vwmulu_vv_u32m1(vuint16mf2_t op1, vuint16mf2_t op2, size_t vl) {
+  return vwmulu(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulu_vx_u32m1(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vwmulu.nxv2i32.nxv2i16.i16.i32(<vscale x 2 x i16> [[OP1:%.*]], i16 [[OP2:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulu_vx_u32m1(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vwmulu.nxv2i32.nxv2i16.i16.i64(<vscale x 2 x i16> [[OP1:%.*]], i16 [[OP2:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
+//
+vuint32m1_t test_vwmulu_vx_u32m1(vuint16mf2_t op1, uint16_t op2, size_t vl) {
+  return vwmulu(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulu_vv_u32m2(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vwmulu.nxv4i32.nxv4i16.nxv4i16.i32(<vscale x 4 x i16> [[OP1:%.*]], <vscale x 4 x i16> [[OP2:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulu_vv_u32m2(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vwmulu.nxv4i32.nxv4i16.nxv4i16.i64(<vscale x 4 x i16> [[OP1:%.*]], <vscale x 4 x i16> [[OP2:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
+//
+vuint32m2_t test_vwmulu_vv_u32m2(vuint16m1_t op1, vuint16m1_t op2, size_t vl) {
+  return vwmulu(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulu_vx_u32m2(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vwmulu.nxv4i32.nxv4i16.i16.i32(<vscale x 4 x i16> [[OP1:%.*]], i16 [[OP2:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulu_vx_u32m2(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vwmulu.nxv4i32.nxv4i16.i16.i64(<vscale x 4 x i16> [[OP1:%.*]], i16 [[OP2:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
+//
+vuint32m2_t test_vwmulu_vx_u32m2(vuint16m1_t op1, uint16_t op2, size_t vl) {
+  return vwmulu(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulu_vv_u32m4(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vwmulu.nxv8i32.nxv8i16.nxv8i16.i32(<vscale x 8 x i16> [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulu_vv_u32m4(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vwmulu.nxv8i32.nxv8i16.nxv8i16.i64(<vscale x 8 x i16> [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
+//
+vuint32m4_t test_vwmulu_vv_u32m4(vuint16m2_t op1, vuint16m2_t op2, size_t vl) {
+  return vwmulu(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulu_vx_u32m4(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vwmulu.nxv8i32.nxv8i16.i16.i32(<vscale x 8 x i16> [[OP1:%.*]], i16 [[OP2:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulu_vx_u32m4(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vwmulu.nxv8i32.nxv8i16.i16.i64(<vscale x 8 x i16> [[OP1:%.*]], i16 [[OP2:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
+//
+vuint32m4_t test_vwmulu_vx_u32m4(vuint16m2_t op1, uint16_t op2, size_t vl) {
+  return vwmulu(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulu_vv_u32m8(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vwmulu.nxv16i32.nxv16i16.nxv16i16.i32(<vscale x 16 x i16> [[OP1:%.*]], <vscale x 16 x i16> [[OP2:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulu_vv_u32m8(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vwmulu.nxv16i32.nxv16i16.nxv16i16.i64(<vscale x 16 x i16> [[OP1:%.*]], <vscale x 16 x i16> [[OP2:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
+//
+vuint32m8_t test_vwmulu_vv_u32m8(vuint16m4_t op1, vuint16m4_t op2, size_t vl) {
+  return vwmulu(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulu_vx_u32m8(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vwmulu.nxv16i32.nxv16i16.i16.i32(<vscale x 16 x i16> [[OP1:%.*]], i16 [[OP2:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulu_vx_u32m8(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vwmulu.nxv16i32.nxv16i16.i16.i64(<vscale x 16 x i16> [[OP1:%.*]], i16 [[OP2:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
+//
+vuint32m8_t test_vwmulu_vx_u32m8(vuint16m4_t op1, uint16_t op2, size_t vl) {
+  return vwmulu(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulu_vv_u64m1(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vwmulu.nxv1i64.nxv1i32.nxv1i32.i32(<vscale x 1 x i32> [[OP1:%.*]], <vscale x 1 x i32> [[OP2:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 1 x i64> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulu_vv_u64m1(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vwmulu.nxv1i64.nxv1i32.nxv1i32.i64(<vscale x 1 x i32> [[OP1:%.*]], <vscale x 1 x i32> [[OP2:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 1 x i64> [[TMP0]]
+//
+vuint64m1_t test_vwmulu_vv_u64m1(vuint32mf2_t op1, vuint32mf2_t op2, size_t vl) {
+  return vwmulu(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulu_vx_u64m1(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vwmulu.nxv1i64.nxv1i32.i32.i32(<vscale x 1 x i32> [[OP1:%.*]], i32 [[OP2:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 1 x i64> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulu_vx_u64m1(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vwmulu.nxv1i64.nxv1i32.i32.i64(<vscale x 1 x i32> [[OP1:%.*]], i32 [[OP2:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 1 x i64> [[TMP0]]
+//
+vuint64m1_t test_vwmulu_vx_u64m1(vuint32mf2_t op1, uint32_t op2, size_t vl) {
+  return vwmulu(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulu_vv_u64m2(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vwmulu.nxv2i64.nxv2i32.nxv2i32.i32(<vscale x 2 x i32> [[OP1:%.*]], <vscale x 2 x i32> [[OP2:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 2 x i64> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulu_vv_u64m2(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vwmulu.nxv2i64.nxv2i32.nxv2i32.i64(<vscale x 2 x i32> [[OP1:%.*]], <vscale x 2 x i32> [[OP2:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 2 x i64> [[TMP0]]
+//
+vuint64m2_t test_vwmulu_vv_u64m2(vuint32m1_t op1, vuint32m1_t op2, size_t vl) {
+  return vwmulu(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulu_vx_u64m2(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vwmulu.nxv2i64.nxv2i32.i32.i32(<vscale x 2 x i32> [[OP1:%.*]], i32 [[OP2:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 2 x i64> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulu_vx_u64m2(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vwmulu.nxv2i64.nxv2i32.i32.i64(<vscale x 2 x i32> [[OP1:%.*]], i32 [[OP2:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 2 x i64> [[TMP0]]
+//
+vuint64m2_t test_vwmulu_vx_u64m2(vuint32m1_t op1, uint32_t op2, size_t vl) {
+  return vwmulu(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulu_vv_u64m4(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vwmulu.nxv4i64.nxv4i32.nxv4i32.i32(<vscale x 4 x i32> [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 4 x i64> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulu_vv_u64m4(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vwmulu.nxv4i64.nxv4i32.nxv4i32.i64(<vscale x 4 x i32> [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 4 x i64> [[TMP0]]
+//
+vuint64m4_t test_vwmulu_vv_u64m4(vuint32m2_t op1, vuint32m2_t op2, size_t vl) {
+  return vwmulu(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulu_vx_u64m4(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vwmulu.nxv4i64.nxv4i32.i32.i32(<vscale x 4 x i32> [[OP1:%.*]], i32 [[OP2:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 4 x i64> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulu_vx_u64m4(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vwmulu.nxv4i64.nxv4i32.i32.i64(<vscale x 4 x i32> [[OP1:%.*]], i32 [[OP2:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 4 x i64> [[TMP0]]
+//
+vuint64m4_t test_vwmulu_vx_u64m4(vuint32m2_t op1, uint32_t op2, size_t vl) {
+  return vwmulu(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulu_vv_u64m8(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vwmulu.nxv8i64.nxv8i32.nxv8i32.i32(<vscale x 8 x i32> [[OP1:%.*]], <vscale x 8 x i32> [[OP2:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 8 x i64> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulu_vv_u64m8(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vwmulu.nxv8i64.nxv8i32.nxv8i32.i64(<vscale x 8 x i32> [[OP1:%.*]], <vscale x 8 x i32> [[OP2:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 8 x i64> [[TMP0]]
+//
+vuint64m8_t test_vwmulu_vv_u64m8(vuint32m4_t op1, vuint32m4_t op2, size_t vl) {
+  return vwmulu(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulu_vx_u64m8(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vwmulu.nxv8i64.nxv8i32.i32.i32(<vscale x 8 x i32> [[OP1:%.*]], i32 [[OP2:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 8 x i64> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulu_vx_u64m8(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vwmulu.nxv8i64.nxv8i32.i32.i64(<vscale x 8 x i32> [[OP1:%.*]], i32 [[OP2:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 8 x i64> [[TMP0]]
+//
+vuint64m8_t test_vwmulu_vx_u64m8(vuint32m4_t op1, uint32_t op2, size_t vl) {
+  return vwmulu(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulsu_vv_i16mf4(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vwmulsu.nxv1i16.nxv1i8.nxv1i8.i32(<vscale x 1 x i8> [[OP1:%.*]], <vscale x 1 x i8> [[OP2:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 1 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulsu_vv_i16mf4(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vwmulsu.nxv1i16.nxv1i8.nxv1i8.i64(<vscale x 1 x i8> [[OP1:%.*]], <vscale x 1 x i8> [[OP2:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 1 x i16> [[TMP0]]
+//
+vint16mf4_t test_vwmulsu_vv_i16mf4(vint8mf8_t op1, vuint8mf8_t op2, size_t vl) {
+  return vwmulsu(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulsu_vx_i16mf4(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vwmulsu.nxv1i16.nxv1i8.i8.i32(<vscale x 1 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 1 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulsu_vx_i16mf4(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vwmulsu.nxv1i16.nxv1i8.i8.i64(<vscale x 1 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 1 x i16> [[TMP0]]
+//
+vint16mf4_t test_vwmulsu_vx_i16mf4(vint8mf8_t op1, uint8_t op2, size_t vl) {
+  return vwmulsu(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulsu_vv_i16mf2(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vwmulsu.nxv2i16.nxv2i8.nxv2i8.i32(<vscale x 2 x i8> [[OP1:%.*]], <vscale x 2 x i8> [[OP2:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 2 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulsu_vv_i16mf2(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vwmulsu.nxv2i16.nxv2i8.nxv2i8.i64(<vscale x 2 x i8> [[OP1:%.*]], <vscale x 2 x i8> [[OP2:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 2 x i16> [[TMP0]]
+//
+vint16mf2_t test_vwmulsu_vv_i16mf2(vint8mf4_t op1, vuint8mf4_t op2, size_t vl) {
+  return vwmulsu(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulsu_vx_i16mf2(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vwmulsu.nxv2i16.nxv2i8.i8.i32(<vscale x 2 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 2 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulsu_vx_i16mf2(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vwmulsu.nxv2i16.nxv2i8.i8.i64(<vscale x 2 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 2 x i16> [[TMP0]]
+//
+vint16mf2_t test_vwmulsu_vx_i16mf2(vint8mf4_t op1, uint8_t op2, size_t vl) {
+  return vwmulsu(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulsu_vv_i16m1(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vwmulsu.nxv4i16.nxv4i8.nxv4i8.i32(<vscale x 4 x i8> [[OP1:%.*]], <vscale x 4 x i8> [[OP2:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 4 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulsu_vv_i16m1(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vwmulsu.nxv4i16.nxv4i8.nxv4i8.i64(<vscale x 4 x i8> [[OP1:%.*]], <vscale x 4 x i8> [[OP2:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 4 x i16> [[TMP0]]
+//
+vint16m1_t test_vwmulsu_vv_i16m1(vint8mf2_t op1, vuint8mf2_t op2, size_t vl) {
+  return vwmulsu(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulsu_vx_i16m1(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vwmulsu.nxv4i16.nxv4i8.i8.i32(<vscale x 4 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 4 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulsu_vx_i16m1(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vwmulsu.nxv4i16.nxv4i8.i8.i64(<vscale x 4 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 4 x i16> [[TMP0]]
+//
+vint16m1_t test_vwmulsu_vx_i16m1(vint8mf2_t op1, uint8_t op2, size_t vl) {
+  return vwmulsu(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulsu_vv_i16m2(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vwmulsu.nxv8i16.nxv8i8.nxv8i8.i32(<vscale x 8 x i8> [[OP1:%.*]], <vscale x 8 x i8> [[OP2:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulsu_vv_i16m2(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vwmulsu.nxv8i16.nxv8i8.nxv8i8.i64(<vscale x 8 x i8> [[OP1:%.*]], <vscale x 8 x i8> [[OP2:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
+//
+vint16m2_t test_vwmulsu_vv_i16m2(vint8m1_t op1, vuint8m1_t op2, size_t vl) {
+  return vwmulsu(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulsu_vx_i16m2(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vwmulsu.nxv8i16.nxv8i8.i8.i32(<vscale x 8 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulsu_vx_i16m2(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vwmulsu.nxv8i16.nxv8i8.i8.i64(<vscale x 8 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
+//
+vint16m2_t test_vwmulsu_vx_i16m2(vint8m1_t op1, uint8_t op2, size_t vl) {
+  return vwmulsu(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulsu_vv_i16m4(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vwmulsu.nxv16i16.nxv16i8.nxv16i8.i32(<vscale x 16 x i8> [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 16 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulsu_vv_i16m4(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vwmulsu.nxv16i16.nxv16i8.nxv16i8.i64(<vscale x 16 x i8> [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 16 x i16> [[TMP0]]
+//
+vint16m4_t test_vwmulsu_vv_i16m4(vint8m2_t op1, vuint8m2_t op2, size_t vl) {
+  return vwmulsu(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulsu_vx_i16m4(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vwmulsu.nxv16i16.nxv16i8.i8.i32(<vscale x 16 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 16 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulsu_vx_i16m4(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vwmulsu.nxv16i16.nxv16i8.i8.i64(<vscale x 16 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 16 x i16> [[TMP0]]
+//
+vint16m4_t test_vwmulsu_vx_i16m4(vint8m2_t op1, uint8_t op2, size_t vl) {
+  return vwmulsu(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulsu_vv_i16m8(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vwmulsu.nxv32i16.nxv32i8.nxv32i8.i32(<vscale x 32 x i8> [[OP1:%.*]], <vscale x 32 x i8> [[OP2:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 32 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulsu_vv_i16m8(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vwmulsu.nxv32i16.nxv32i8.nxv32i8.i64(<vscale x 32 x i8> [[OP1:%.*]], <vscale x 32 x i8> [[OP2:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 32 x i16> [[TMP0]]
+//
+vint16m8_t test_vwmulsu_vv_i16m8(vint8m4_t op1, vuint8m4_t op2, size_t vl) {
+  return vwmulsu(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulsu_vx_i16m8(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vwmulsu.nxv32i16.nxv32i8.i8.i32(<vscale x 32 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 32 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulsu_vx_i16m8(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vwmulsu.nxv32i16.nxv32i8.i8.i64(<vscale x 32 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 32 x i16> [[TMP0]]
+//
+vint16m8_t test_vwmulsu_vx_i16m8(vint8m4_t op1, uint8_t op2, size_t vl) {
+  return vwmulsu(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulsu_vv_i32mf2(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vwmulsu.nxv1i32.nxv1i16.nxv1i16.i32(<vscale x 1 x i16> [[OP1:%.*]], <vscale x 1 x i16> [[OP2:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulsu_vv_i32mf2(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vwmulsu.nxv1i32.nxv1i16.nxv1i16.i64(<vscale x 1 x i16> [[OP1:%.*]], <vscale x 1 x i16> [[OP2:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
+//
+vint32mf2_t test_vwmulsu_vv_i32mf2(vint16mf4_t op1, vuint16mf4_t op2, size_t vl) {
+  return vwmulsu(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulsu_vx_i32mf2(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vwmulsu.nxv1i32.nxv1i16.i16.i32(<vscale x 1 x i16> [[OP1:%.*]], i16 [[OP2:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulsu_vx_i32mf2(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vwmulsu.nxv1i32.nxv1i16.i16.i64(<vscale x 1 x i16> [[OP1:%.*]], i16 [[OP2:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
+//
+vint32mf2_t test_vwmulsu_vx_i32mf2(vint16mf4_t op1, uint16_t op2, size_t vl) {
+  return vwmulsu(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulsu_vv_i32m1(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vwmulsu.nxv2i32.nxv2i16.nxv2i16.i32(<vscale x 2 x i16> [[OP1:%.*]], <vscale x 2 x i16> [[OP2:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulsu_vv_i32m1(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vwmulsu.nxv2i32.nxv2i16.nxv2i16.i64(<vscale x 2 x i16> [[OP1:%.*]], <vscale x 2 x i16> [[OP2:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
+//
+vint32m1_t test_vwmulsu_vv_i32m1(vint16mf2_t op1, vuint16mf2_t op2, size_t vl) {
+  return vwmulsu(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulsu_vx_i32m1(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vwmulsu.nxv2i32.nxv2i16.i16.i32(<vscale x 2 x i16> [[OP1:%.*]], i16 [[OP2:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulsu_vx_i32m1(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vwmulsu.nxv2i32.nxv2i16.i16.i64(<vscale x 2 x i16> [[OP1:%.*]], i16 [[OP2:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
+//
+vint32m1_t test_vwmulsu_vx_i32m1(vint16mf2_t op1, uint16_t op2, size_t vl) {
+  return vwmulsu(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulsu_vv_i32m2(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vwmulsu.nxv4i32.nxv4i16.nxv4i16.i32(<vscale x 4 x i16> [[OP1:%.*]], <vscale x 4 x i16> [[OP2:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulsu_vv_i32m2(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vwmulsu.nxv4i32.nxv4i16.nxv4i16.i64(<vscale x 4 x i16> [[OP1:%.*]], <vscale x 4 x i16> [[OP2:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
+//
+vint32m2_t test_vwmulsu_vv_i32m2(vint16m1_t op1, vuint16m1_t op2, size_t vl) {
+  return vwmulsu(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulsu_vx_i32m2(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vwmulsu.nxv4i32.nxv4i16.i16.i32(<vscale x 4 x i16> [[OP1:%.*]], i16 [[OP2:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulsu_vx_i32m2(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vwmulsu.nxv4i32.nxv4i16.i16.i64(<vscale x 4 x i16> [[OP1:%.*]], i16 [[OP2:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
+//
+vint32m2_t test_vwmulsu_vx_i32m2(vint16m1_t op1, uint16_t op2, size_t vl) {
+  return vwmulsu(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulsu_vv_i32m4(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vwmulsu.nxv8i32.nxv8i16.nxv8i16.i32(<vscale x 8 x i16> [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulsu_vv_i32m4(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vwmulsu.nxv8i32.nxv8i16.nxv8i16.i64(<vscale x 8 x i16> [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
+//
+vint32m4_t test_vwmulsu_vv_i32m4(vint16m2_t op1, vuint16m2_t op2, size_t vl) {
+  return vwmulsu(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulsu_vx_i32m4(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vwmulsu.nxv8i32.nxv8i16.i16.i32(<vscale x 8 x i16> [[OP1:%.*]], i16 [[OP2:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulsu_vx_i32m4(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vwmulsu.nxv8i32.nxv8i16.i16.i64(<vscale x 8 x i16> [[OP1:%.*]], i16 [[OP2:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
+//
+vint32m4_t test_vwmulsu_vx_i32m4(vint16m2_t op1, uint16_t op2, size_t vl) {
+  return vwmulsu(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulsu_vv_i32m8(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vwmulsu.nxv16i32.nxv16i16.nxv16i16.i32(<vscale x 16 x i16> [[OP1:%.*]], <vscale x 16 x i16> [[OP2:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulsu_vv_i32m8(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vwmulsu.nxv16i32.nxv16i16.nxv16i16.i64(<vscale x 16 x i16> [[OP1:%.*]], <vscale x 16 x i16> [[OP2:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
+//
+vint32m8_t test_vwmulsu_vv_i32m8(vint16m4_t op1, vuint16m4_t op2, size_t vl) {
+  return vwmulsu(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulsu_vx_i32m8(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vwmulsu.nxv16i32.nxv16i16.i16.i32(<vscale x 16 x i16> [[OP1:%.*]], i16 [[OP2:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulsu_vx_i32m8(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vwmulsu.nxv16i32.nxv16i16.i16.i64(<vscale x 16 x i16> [[OP1:%.*]], i16 [[OP2:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
+//
+vint32m8_t test_vwmulsu_vx_i32m8(vint16m4_t op1, uint16_t op2, size_t vl) {
+  return vwmulsu(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulsu_vv_i64m1(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vwmulsu.nxv1i64.nxv1i32.nxv1i32.i32(<vscale x 1 x i32> [[OP1:%.*]], <vscale x 1 x i32> [[OP2:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 1 x i64> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulsu_vv_i64m1(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vwmulsu.nxv1i64.nxv1i32.nxv1i32.i64(<vscale x 1 x i32> [[OP1:%.*]], <vscale x 1 x i32> [[OP2:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 1 x i64> [[TMP0]]
+//
+vint64m1_t test_vwmulsu_vv_i64m1(vint32mf2_t op1, vuint32mf2_t op2, size_t vl) {
+  return vwmulsu(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulsu_vx_i64m1(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vwmulsu.nxv1i64.nxv1i32.i32.i32(<vscale x 1 x i32> [[OP1:%.*]], i32 [[OP2:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 1 x i64> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulsu_vx_i64m1(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vwmulsu.nxv1i64.nxv1i32.i32.i64(<vscale x 1 x i32> [[OP1:%.*]], i32 [[OP2:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 1 x i64> [[TMP0]]
+//
+vint64m1_t test_vwmulsu_vx_i64m1(vint32mf2_t op1, uint32_t op2, size_t vl) {
+  return vwmulsu(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulsu_vv_i64m2(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vwmulsu.nxv2i64.nxv2i32.nxv2i32.i32(<vscale x 2 x i32> [[OP1:%.*]], <vscale x 2 x i32> [[OP2:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 2 x i64> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulsu_vv_i64m2(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vwmulsu.nxv2i64.nxv2i32.nxv2i32.i64(<vscale x 2 x i32> [[OP1:%.*]], <vscale x 2 x i32> [[OP2:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 2 x i64> [[TMP0]]
+//
+vint64m2_t test_vwmulsu_vv_i64m2(vint32m1_t op1, vuint32m1_t op2, size_t vl) {
+  return vwmulsu(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulsu_vx_i64m2(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vwmulsu.nxv2i64.nxv2i32.i32.i32(<vscale x 2 x i32> [[OP1:%.*]], i32 [[OP2:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 2 x i64> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulsu_vx_i64m2(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vwmulsu.nxv2i64.nxv2i32.i32.i64(<vscale x 2 x i32> [[OP1:%.*]], i32 [[OP2:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 2 x i64> [[TMP0]]
+//
+vint64m2_t test_vwmulsu_vx_i64m2(vint32m1_t op1, uint32_t op2, size_t vl) {
+  return vwmulsu(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulsu_vv_i64m4(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vwmulsu.nxv4i64.nxv4i32.nxv4i32.i32(<vscale x 4 x i32> [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 4 x i64> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulsu_vv_i64m4(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vwmulsu.nxv4i64.nxv4i32.nxv4i32.i64(<vscale x 4 x i32> [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 4 x i64> [[TMP0]]
+//
+vint64m4_t test_vwmulsu_vv_i64m4(vint32m2_t op1, vuint32m2_t op2, size_t vl) {
+  return vwmulsu(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulsu_vx_i64m4(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vwmulsu.nxv4i64.nxv4i32.i32.i32(<vscale x 4 x i32> [[OP1:%.*]], i32 [[OP2:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 4 x i64> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulsu_vx_i64m4(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vwmulsu.nxv4i64.nxv4i32.i32.i64(<vscale x 4 x i32> [[OP1:%.*]], i32 [[OP2:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 4 x i64> [[TMP0]]
+//
+vint64m4_t test_vwmulsu_vx_i64m4(vint32m2_t op1, uint32_t op2, size_t vl) {
+  return vwmulsu(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulsu_vv_i64m8(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vwmulsu.nxv8i64.nxv8i32.nxv8i32.i32(<vscale x 8 x i32> [[OP1:%.*]], <vscale x 8 x i32> [[OP2:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 8 x i64> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulsu_vv_i64m8(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vwmulsu.nxv8i64.nxv8i32.nxv8i32.i64(<vscale x 8 x i32> [[OP1:%.*]], <vscale x 8 x i32> [[OP2:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 8 x i64> [[TMP0]]
+//
+vint64m8_t test_vwmulsu_vv_i64m8(vint32m4_t op1, vuint32m4_t op2, size_t vl) {
+  return vwmulsu(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulsu_vx_i64m8(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vwmulsu.nxv8i64.nxv8i32.i32.i32(<vscale x 8 x i32> [[OP1:%.*]], i32 [[OP2:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 8 x i64> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulsu_vx_i64m8(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vwmulsu.nxv8i64.nxv8i32.i32.i64(<vscale x 8 x i32> [[OP1:%.*]], i32 [[OP2:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 8 x i64> [[TMP0]]
+//
+vint64m8_t test_vwmulsu_vx_i64m8(vint32m4_t op1, uint32_t op2, size_t vl) {
+  return vwmulsu(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmul_vv_i16mf4_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vwmul.mask.nxv1i16.nxv1i8.nxv1i8.i32(<vscale x 1 x i16> [[MASKEDOFF:%.*]], <vscale x 1 x i8> [[OP1:%.*]], <vscale x 1 x i8> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 1 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmul_vv_i16mf4_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vwmul.mask.nxv1i16.nxv1i8.nxv1i8.i64(<vscale x 1 x i16> [[MASKEDOFF:%.*]], <vscale x 1 x i8> [[OP1:%.*]], <vscale x 1 x i8> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 1 x i16> [[TMP0]]
+//
+vint16mf4_t test_vwmul_vv_i16mf4_m(vbool64_t mask, vint16mf4_t maskedoff, vint8mf8_t op1, vint8mf8_t op2, size_t vl) {
+  return vwmul(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmul_vx_i16mf4_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vwmul.mask.nxv1i16.nxv1i8.i8.i32(<vscale x 1 x i16> [[MASKEDOFF:%.*]], <vscale x 1 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 1 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmul_vx_i16mf4_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vwmul.mask.nxv1i16.nxv1i8.i8.i64(<vscale x 1 x i16> [[MASKEDOFF:%.*]], <vscale x 1 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 1 x i16> [[TMP0]]
+//
+vint16mf4_t test_vwmul_vx_i16mf4_m(vbool64_t mask, vint16mf4_t maskedoff, vint8mf8_t op1, int8_t op2, size_t vl) {
+  return vwmul(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmul_vv_i16mf2_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vwmul.mask.nxv2i16.nxv2i8.nxv2i8.i32(<vscale x 2 x i16> [[MASKEDOFF:%.*]], <vscale x 2 x i8> [[OP1:%.*]], <vscale x 2 x i8> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 2 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmul_vv_i16mf2_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vwmul.mask.nxv2i16.nxv2i8.nxv2i8.i64(<vscale x 2 x i16> [[MASKEDOFF:%.*]], <vscale x 2 x i8> [[OP1:%.*]], <vscale x 2 x i8> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 2 x i16> [[TMP0]]
+//
+vint16mf2_t test_vwmul_vv_i16mf2_m(vbool32_t mask, vint16mf2_t maskedoff, vint8mf4_t op1, vint8mf4_t op2, size_t vl) {
+  return vwmul(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmul_vx_i16mf2_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vwmul.mask.nxv2i16.nxv2i8.i8.i32(<vscale x 2 x i16> [[MASKEDOFF:%.*]], <vscale x 2 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 2 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmul_vx_i16mf2_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vwmul.mask.nxv2i16.nxv2i8.i8.i64(<vscale x 2 x i16> [[MASKEDOFF:%.*]], <vscale x 2 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 2 x i16> [[TMP0]]
+//
+vint16mf2_t test_vwmul_vx_i16mf2_m(vbool32_t mask, vint16mf2_t maskedoff, vint8mf4_t op1, int8_t op2, size_t vl) {
+  return vwmul(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmul_vv_i16m1_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vwmul.mask.nxv4i16.nxv4i8.nxv4i8.i32(<vscale x 4 x i16> [[MASKEDOFF:%.*]], <vscale x 4 x i8> [[OP1:%.*]], <vscale x 4 x i8> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 4 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmul_vv_i16m1_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vwmul.mask.nxv4i16.nxv4i8.nxv4i8.i64(<vscale x 4 x i16> [[MASKEDOFF:%.*]], <vscale x 4 x i8> [[OP1:%.*]], <vscale x 4 x i8> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 4 x i16> [[TMP0]]
+//
+vint16m1_t test_vwmul_vv_i16m1_m(vbool16_t mask, vint16m1_t maskedoff, vint8mf2_t op1, vint8mf2_t op2, size_t vl) {
+  return vwmul(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmul_vx_i16m1_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vwmul.mask.nxv4i16.nxv4i8.i8.i32(<vscale x 4 x i16> [[MASKEDOFF:%.*]], <vscale x 4 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 4 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmul_vx_i16m1_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vwmul.mask.nxv4i16.nxv4i8.i8.i64(<vscale x 4 x i16> [[MASKEDOFF:%.*]], <vscale x 4 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 4 x i16> [[TMP0]]
+//
+vint16m1_t test_vwmul_vx_i16m1_m(vbool16_t mask, vint16m1_t maskedoff, vint8mf2_t op1, int8_t op2, size_t vl) {
+  return vwmul(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmul_vv_i16m2_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vwmul.mask.nxv8i16.nxv8i8.nxv8i8.i32(<vscale x 8 x i16> [[MASKEDOFF:%.*]], <vscale x 8 x i8> [[OP1:%.*]], <vscale x 8 x i8> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmul_vv_i16m2_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vwmul.mask.nxv8i16.nxv8i8.nxv8i8.i64(<vscale x 8 x i16> [[MASKEDOFF:%.*]], <vscale x 8 x i8> [[OP1:%.*]], <vscale x 8 x i8> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
+//
+vint16m2_t test_vwmul_vv_i16m2_m(vbool8_t mask, vint16m2_t maskedoff, vint8m1_t op1, vint8m1_t op2, size_t vl) {
+  return vwmul(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmul_vx_i16m2_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vwmul.mask.nxv8i16.nxv8i8.i8.i32(<vscale x 8 x i16> [[MASKEDOFF:%.*]], <vscale x 8 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmul_vx_i16m2_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vwmul.mask.nxv8i16.nxv8i8.i8.i64(<vscale x 8 x i16> [[MASKEDOFF:%.*]], <vscale x 8 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
+//
+vint16m2_t test_vwmul_vx_i16m2_m(vbool8_t mask, vint16m2_t maskedoff, vint8m1_t op1, int8_t op2, size_t vl) {
+  return vwmul(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmul_vv_i16m4_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vwmul.mask.nxv16i16.nxv16i8.nxv16i8.i32(<vscale x 16 x i16> [[MASKEDOFF:%.*]], <vscale x 16 x i8> [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 16 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmul_vv_i16m4_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vwmul.mask.nxv16i16.nxv16i8.nxv16i8.i64(<vscale x 16 x i16> [[MASKEDOFF:%.*]], <vscale x 16 x i8> [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 16 x i16> [[TMP0]]
+//
+vint16m4_t test_vwmul_vv_i16m4_m(vbool4_t mask, vint16m4_t maskedoff, vint8m2_t op1, vint8m2_t op2, size_t vl) {
+  return vwmul(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmul_vx_i16m4_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vwmul.mask.nxv16i16.nxv16i8.i8.i32(<vscale x 16 x i16> [[MASKEDOFF:%.*]], <vscale x 16 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 16 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmul_vx_i16m4_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vwmul.mask.nxv16i16.nxv16i8.i8.i64(<vscale x 16 x i16> [[MASKEDOFF:%.*]], <vscale x 16 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 16 x i16> [[TMP0]]
+//
+vint16m4_t test_vwmul_vx_i16m4_m(vbool4_t mask, vint16m4_t maskedoff, vint8m2_t op1, int8_t op2, size_t vl) {
+  return vwmul(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmul_vv_i16m8_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vwmul.mask.nxv32i16.nxv32i8.nxv32i8.i32(<vscale x 32 x i16> [[MASKEDOFF:%.*]], <vscale x 32 x i8> [[OP1:%.*]], <vscale x 32 x i8> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 32 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmul_vv_i16m8_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vwmul.mask.nxv32i16.nxv32i8.nxv32i8.i64(<vscale x 32 x i16> [[MASKEDOFF:%.*]], <vscale x 32 x i8> [[OP1:%.*]], <vscale x 32 x i8> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 32 x i16> [[TMP0]]
+//
+vint16m8_t test_vwmul_vv_i16m8_m(vbool2_t mask, vint16m8_t maskedoff, vint8m4_t op1, vint8m4_t op2, size_t vl) {
+  return vwmul(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmul_vx_i16m8_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vwmul.mask.nxv32i16.nxv32i8.i8.i32(<vscale x 32 x i16> [[MASKEDOFF:%.*]], <vscale x 32 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 32 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmul_vx_i16m8_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vwmul.mask.nxv32i16.nxv32i8.i8.i64(<vscale x 32 x i16> [[MASKEDOFF:%.*]], <vscale x 32 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 32 x i16> [[TMP0]]
+//
+vint16m8_t test_vwmul_vx_i16m8_m(vbool2_t mask, vint16m8_t maskedoff, vint8m4_t op1, int8_t op2, size_t vl) {
+  return vwmul(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmul_vv_i32mf2_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vwmul.mask.nxv1i32.nxv1i16.nxv1i16.i32(<vscale x 1 x i32> [[MASKEDOFF:%.*]], <vscale x 1 x i16> [[OP1:%.*]], <vscale x 1 x i16> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmul_vv_i32mf2_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vwmul.mask.nxv1i32.nxv1i16.nxv1i16.i64(<vscale x 1 x i32> [[MASKEDOFF:%.*]], <vscale x 1 x i16> [[OP1:%.*]], <vscale x 1 x i16> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
+//
+vint32mf2_t test_vwmul_vv_i32mf2_m(vbool64_t mask, vint32mf2_t maskedoff, vint16mf4_t op1, vint16mf4_t op2, size_t vl) {
+  return vwmul(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmul_vx_i32mf2_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vwmul.mask.nxv1i32.nxv1i16.i16.i32(<vscale x 1 x i32> [[MASKEDOFF:%.*]], <vscale x 1 x i16> [[OP1:%.*]], i16 [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmul_vx_i32mf2_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vwmul.mask.nxv1i32.nxv1i16.i16.i64(<vscale x 1 x i32> [[MASKEDOFF:%.*]], <vscale x 1 x i16> [[OP1:%.*]], i16 [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
+//
+vint32mf2_t test_vwmul_vx_i32mf2_m(vbool64_t mask, vint32mf2_t maskedoff, vint16mf4_t op1, int16_t op2, size_t vl) {
+  return vwmul(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmul_vv_i32m1_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vwmul.mask.nxv2i32.nxv2i16.nxv2i16.i32(<vscale x 2 x i32> [[MASKEDOFF:%.*]], <vscale x 2 x i16> [[OP1:%.*]], <vscale x 2 x i16> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmul_vv_i32m1_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vwmul.mask.nxv2i32.nxv2i16.nxv2i16.i64(<vscale x 2 x i32> [[MASKEDOFF:%.*]], <vscale x 2 x i16> [[OP1:%.*]], <vscale x 2 x i16> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
+//
+vint32m1_t test_vwmul_vv_i32m1_m(vbool32_t mask, vint32m1_t maskedoff, vint16mf2_t op1, vint16mf2_t op2, size_t vl) {
+  return vwmul(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmul_vx_i32m1_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vwmul.mask.nxv2i32.nxv2i16.i16.i32(<vscale x 2 x i32> [[MASKEDOFF:%.*]], <vscale x 2 x i16> [[OP1:%.*]], i16 [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmul_vx_i32m1_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vwmul.mask.nxv2i32.nxv2i16.i16.i64(<vscale x 2 x i32> [[MASKEDOFF:%.*]], <vscale x 2 x i16> [[OP1:%.*]], i16 [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
+//
+vint32m1_t test_vwmul_vx_i32m1_m(vbool32_t mask, vint32m1_t maskedoff, vint16mf2_t op1, int16_t op2, size_t vl) {
+  return vwmul(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmul_vv_i32m2_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vwmul.mask.nxv4i32.nxv4i16.nxv4i16.i32(<vscale x 4 x i32> [[MASKEDOFF:%.*]], <vscale x 4 x i16> [[OP1:%.*]], <vscale x 4 x i16> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmul_vv_i32m2_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vwmul.mask.nxv4i32.nxv4i16.nxv4i16.i64(<vscale x 4 x i32> [[MASKEDOFF:%.*]], <vscale x 4 x i16> [[OP1:%.*]], <vscale x 4 x i16> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
+//
+vint32m2_t test_vwmul_vv_i32m2_m(vbool16_t mask, vint32m2_t maskedoff, vint16m1_t op1, vint16m1_t op2, size_t vl) {
+  return vwmul(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmul_vx_i32m2_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vwmul.mask.nxv4i32.nxv4i16.i16.i32(<vscale x 4 x i32> [[MASKEDOFF:%.*]], <vscale x 4 x i16> [[OP1:%.*]], i16 [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmul_vx_i32m2_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vwmul.mask.nxv4i32.nxv4i16.i16.i64(<vscale x 4 x i32> [[MASKEDOFF:%.*]], <vscale x 4 x i16> [[OP1:%.*]], i16 [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
+//
+vint32m2_t test_vwmul_vx_i32m2_m(vbool16_t mask, vint32m2_t maskedoff, vint16m1_t op1, int16_t op2, size_t vl) {
+  return vwmul(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmul_vv_i32m4_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vwmul.mask.nxv8i32.nxv8i16.nxv8i16.i32(<vscale x 8 x i32> [[MASKEDOFF:%.*]], <vscale x 8 x i16> [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmul_vv_i32m4_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vwmul.mask.nxv8i32.nxv8i16.nxv8i16.i64(<vscale x 8 x i32> [[MASKEDOFF:%.*]], <vscale x 8 x i16> [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
+//
+vint32m4_t test_vwmul_vv_i32m4_m(vbool8_t mask, vint32m4_t maskedoff, vint16m2_t op1, vint16m2_t op2, size_t vl) {
+  return vwmul(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmul_vx_i32m4_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vwmul.mask.nxv8i32.nxv8i16.i16.i32(<vscale x 8 x i32> [[MASKEDOFF:%.*]], <vscale x 8 x i16> [[OP1:%.*]], i16 [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmul_vx_i32m4_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vwmul.mask.nxv8i32.nxv8i16.i16.i64(<vscale x 8 x i32> [[MASKEDOFF:%.*]], <vscale x 8 x i16> [[OP1:%.*]], i16 [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
+//
+vint32m4_t test_vwmul_vx_i32m4_m(vbool8_t mask, vint32m4_t maskedoff, vint16m2_t op1, int16_t op2, size_t vl) {
+  return vwmul(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmul_vv_i32m8_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vwmul.mask.nxv16i32.nxv16i16.nxv16i16.i32(<vscale x 16 x i32> [[MASKEDOFF:%.*]], <vscale x 16 x i16> [[OP1:%.*]], <vscale x 16 x i16> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmul_vv_i32m8_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vwmul.mask.nxv16i32.nxv16i16.nxv16i16.i64(<vscale x 16 x i32> [[MASKEDOFF:%.*]], <vscale x 16 x i16> [[OP1:%.*]], <vscale x 16 x i16> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
+//
+vint32m8_t test_vwmul_vv_i32m8_m(vbool4_t mask, vint32m8_t maskedoff, vint16m4_t op1, vint16m4_t op2, size_t vl) {
+  return vwmul(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmul_vx_i32m8_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vwmul.mask.nxv16i32.nxv16i16.i16.i32(<vscale x 16 x i32> [[MASKEDOFF:%.*]], <vscale x 16 x i16> [[OP1:%.*]], i16 [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmul_vx_i32m8_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vwmul.mask.nxv16i32.nxv16i16.i16.i64(<vscale x 16 x i32> [[MASKEDOFF:%.*]], <vscale x 16 x i16> [[OP1:%.*]], i16 [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
+//
+vint32m8_t test_vwmul_vx_i32m8_m(vbool4_t mask, vint32m8_t maskedoff, vint16m4_t op1, int16_t op2, size_t vl) {
+  return vwmul(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmul_vv_i64m1_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vwmul.mask.nxv1i64.nxv1i32.nxv1i32.i32(<vscale x 1 x i64> [[MASKEDOFF:%.*]], <vscale x 1 x i32> [[OP1:%.*]], <vscale x 1 x i32> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 1 x i64> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmul_vv_i64m1_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vwmul.mask.nxv1i64.nxv1i32.nxv1i32.i64(<vscale x 1 x i64> [[MASKEDOFF:%.*]], <vscale x 1 x i32> [[OP1:%.*]], <vscale x 1 x i32> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 1 x i64> [[TMP0]]
+//
+vint64m1_t test_vwmul_vv_i64m1_m(vbool64_t mask, vint64m1_t maskedoff, vint32mf2_t op1, vint32mf2_t op2, size_t vl) {
+  return vwmul(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmul_vx_i64m1_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vwmul.mask.nxv1i64.nxv1i32.i32.i32(<vscale x 1 x i64> [[MASKEDOFF:%.*]], <vscale x 1 x i32> [[OP1:%.*]], i32 [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 1 x i64> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmul_vx_i64m1_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vwmul.mask.nxv1i64.nxv1i32.i32.i64(<vscale x 1 x i64> [[MASKEDOFF:%.*]], <vscale x 1 x i32> [[OP1:%.*]], i32 [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 1 x i64> [[TMP0]]
+//
+vint64m1_t test_vwmul_vx_i64m1_m(vbool64_t mask, vint64m1_t maskedoff, vint32mf2_t op1, int32_t op2, size_t vl) {
+  return vwmul(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmul_vv_i64m2_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vwmul.mask.nxv2i64.nxv2i32.nxv2i32.i32(<vscale x 2 x i64> [[MASKEDOFF:%.*]], <vscale x 2 x i32> [[OP1:%.*]], <vscale x 2 x i32> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 2 x i64> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmul_vv_i64m2_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vwmul.mask.nxv2i64.nxv2i32.nxv2i32.i64(<vscale x 2 x i64> [[MASKEDOFF:%.*]], <vscale x 2 x i32> [[OP1:%.*]], <vscale x 2 x i32> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 2 x i64> [[TMP0]]
+//
+vint64m2_t test_vwmul_vv_i64m2_m(vbool32_t mask, vint64m2_t maskedoff, vint32m1_t op1, vint32m1_t op2, size_t vl) {
+  return vwmul(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmul_vx_i64m2_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vwmul.mask.nxv2i64.nxv2i32.i32.i32(<vscale x 2 x i64> [[MASKEDOFF:%.*]], <vscale x 2 x i32> [[OP1:%.*]], i32 [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 2 x i64> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmul_vx_i64m2_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vwmul.mask.nxv2i64.nxv2i32.i32.i64(<vscale x 2 x i64> [[MASKEDOFF:%.*]], <vscale x 2 x i32> [[OP1:%.*]], i32 [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 2 x i64> [[TMP0]]
+//
+vint64m2_t test_vwmul_vx_i64m2_m(vbool32_t mask, vint64m2_t maskedoff, vint32m1_t op1, int32_t op2, size_t vl) {
+  return vwmul(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmul_vv_i64m4_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vwmul.mask.nxv4i64.nxv4i32.nxv4i32.i32(<vscale x 4 x i64> [[MASKEDOFF:%.*]], <vscale x 4 x i32> [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 4 x i64> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmul_vv_i64m4_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vwmul.mask.nxv4i64.nxv4i32.nxv4i32.i64(<vscale x 4 x i64> [[MASKEDOFF:%.*]], <vscale x 4 x i32> [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 4 x i64> [[TMP0]]
+//
+vint64m4_t test_vwmul_vv_i64m4_m(vbool16_t mask, vint64m4_t maskedoff, vint32m2_t op1, vint32m2_t op2, size_t vl) {
+  return vwmul(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmul_vx_i64m4_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vwmul.mask.nxv4i64.nxv4i32.i32.i32(<vscale x 4 x i64> [[MASKEDOFF:%.*]], <vscale x 4 x i32> [[OP1:%.*]], i32 [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 4 x i64> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmul_vx_i64m4_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vwmul.mask.nxv4i64.nxv4i32.i32.i64(<vscale x 4 x i64> [[MASKEDOFF:%.*]], <vscale x 4 x i32> [[OP1:%.*]], i32 [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 4 x i64> [[TMP0]]
+//
+vint64m4_t test_vwmul_vx_i64m4_m(vbool16_t mask, vint64m4_t maskedoff, vint32m2_t op1, int32_t op2, size_t vl) {
+  return vwmul(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmul_vv_i64m8_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vwmul.mask.nxv8i64.nxv8i32.nxv8i32.i32(<vscale x 8 x i64> [[MASKEDOFF:%.*]], <vscale x 8 x i32> [[OP1:%.*]], <vscale x 8 x i32> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 8 x i64> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmul_vv_i64m8_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vwmul.mask.nxv8i64.nxv8i32.nxv8i32.i64(<vscale x 8 x i64> [[MASKEDOFF:%.*]], <vscale x 8 x i32> [[OP1:%.*]], <vscale x 8 x i32> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 8 x i64> [[TMP0]]
+//
+vint64m8_t test_vwmul_vv_i64m8_m(vbool8_t mask, vint64m8_t maskedoff, vint32m4_t op1, vint32m4_t op2, size_t vl) {
+  return vwmul(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmul_vx_i64m8_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vwmul.mask.nxv8i64.nxv8i32.i32.i32(<vscale x 8 x i64> [[MASKEDOFF:%.*]], <vscale x 8 x i32> [[OP1:%.*]], i32 [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 8 x i64> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmul_vx_i64m8_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vwmul.mask.nxv8i64.nxv8i32.i32.i64(<vscale x 8 x i64> [[MASKEDOFF:%.*]], <vscale x 8 x i32> [[OP1:%.*]], i32 [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 8 x i64> [[TMP0]]
+//
+vint64m8_t test_vwmul_vx_i64m8_m(vbool8_t mask, vint64m8_t maskedoff, vint32m4_t op1, int32_t op2, size_t vl) {
+  return vwmul(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulu_vv_u16mf4_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vwmulu.mask.nxv1i16.nxv1i8.nxv1i8.i32(<vscale x 1 x i16> [[MASKEDOFF:%.*]], <vscale x 1 x i8> [[OP1:%.*]], <vscale x 1 x i8> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 1 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulu_vv_u16mf4_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vwmulu.mask.nxv1i16.nxv1i8.nxv1i8.i64(<vscale x 1 x i16> [[MASKEDOFF:%.*]], <vscale x 1 x i8> [[OP1:%.*]], <vscale x 1 x i8> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 1 x i16> [[TMP0]]
+//
+vuint16mf4_t test_vwmulu_vv_u16mf4_m(vbool64_t mask, vuint16mf4_t maskedoff, vuint8mf8_t op1, vuint8mf8_t op2, size_t vl) {
+  return vwmulu(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulu_vx_u16mf4_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vwmulu.mask.nxv1i16.nxv1i8.i8.i32(<vscale x 1 x i16> [[MASKEDOFF:%.*]], <vscale x 1 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 1 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulu_vx_u16mf4_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vwmulu.mask.nxv1i16.nxv1i8.i8.i64(<vscale x 1 x i16> [[MASKEDOFF:%.*]], <vscale x 1 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 1 x i16> [[TMP0]]
+//
+vuint16mf4_t test_vwmulu_vx_u16mf4_m(vbool64_t mask, vuint16mf4_t maskedoff, vuint8mf8_t op1, uint8_t op2, size_t vl) {
+  return vwmulu(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulu_vv_u16mf2_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vwmulu.mask.nxv2i16.nxv2i8.nxv2i8.i32(<vscale x 2 x i16> [[MASKEDOFF:%.*]], <vscale x 2 x i8> [[OP1:%.*]], <vscale x 2 x i8> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 2 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulu_vv_u16mf2_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vwmulu.mask.nxv2i16.nxv2i8.nxv2i8.i64(<vscale x 2 x i16> [[MASKEDOFF:%.*]], <vscale x 2 x i8> [[OP1:%.*]], <vscale x 2 x i8> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 2 x i16> [[TMP0]]
+//
+vuint16mf2_t test_vwmulu_vv_u16mf2_m(vbool32_t mask, vuint16mf2_t maskedoff, vuint8mf4_t op1, vuint8mf4_t op2, size_t vl) {
+  return vwmulu(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulu_vx_u16mf2_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vwmulu.mask.nxv2i16.nxv2i8.i8.i32(<vscale x 2 x i16> [[MASKEDOFF:%.*]], <vscale x 2 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 2 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulu_vx_u16mf2_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vwmulu.mask.nxv2i16.nxv2i8.i8.i64(<vscale x 2 x i16> [[MASKEDOFF:%.*]], <vscale x 2 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 2 x i16> [[TMP0]]
+//
+vuint16mf2_t test_vwmulu_vx_u16mf2_m(vbool32_t mask, vuint16mf2_t maskedoff, vuint8mf4_t op1, uint8_t op2, size_t vl) {
+  return vwmulu(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulu_vv_u16m1_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vwmulu.mask.nxv4i16.nxv4i8.nxv4i8.i32(<vscale x 4 x i16> [[MASKEDOFF:%.*]], <vscale x 4 x i8> [[OP1:%.*]], <vscale x 4 x i8> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 4 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulu_vv_u16m1_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vwmulu.mask.nxv4i16.nxv4i8.nxv4i8.i64(<vscale x 4 x i16> [[MASKEDOFF:%.*]], <vscale x 4 x i8> [[OP1:%.*]], <vscale x 4 x i8> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 4 x i16> [[TMP0]]
+//
+vuint16m1_t test_vwmulu_vv_u16m1_m(vbool16_t mask, vuint16m1_t maskedoff, vuint8mf2_t op1, vuint8mf2_t op2, size_t vl) {
+  return vwmulu(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulu_vx_u16m1_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vwmulu.mask.nxv4i16.nxv4i8.i8.i32(<vscale x 4 x i16> [[MASKEDOFF:%.*]], <vscale x 4 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 4 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulu_vx_u16m1_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vwmulu.mask.nxv4i16.nxv4i8.i8.i64(<vscale x 4 x i16> [[MASKEDOFF:%.*]], <vscale x 4 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 4 x i16> [[TMP0]]
+//
+vuint16m1_t test_vwmulu_vx_u16m1_m(vbool16_t mask, vuint16m1_t maskedoff, vuint8mf2_t op1, uint8_t op2, size_t vl) {
+  return vwmulu(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulu_vv_u16m2_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vwmulu.mask.nxv8i16.nxv8i8.nxv8i8.i32(<vscale x 8 x i16> [[MASKEDOFF:%.*]], <vscale x 8 x i8> [[OP1:%.*]], <vscale x 8 x i8> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulu_vv_u16m2_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vwmulu.mask.nxv8i16.nxv8i8.nxv8i8.i64(<vscale x 8 x i16> [[MASKEDOFF:%.*]], <vscale x 8 x i8> [[OP1:%.*]], <vscale x 8 x i8> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
+//
+vuint16m2_t test_vwmulu_vv_u16m2_m(vbool8_t mask, vuint16m2_t maskedoff, vuint8m1_t op1, vuint8m1_t op2, size_t vl) {
+  return vwmulu(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulu_vx_u16m2_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vwmulu.mask.nxv8i16.nxv8i8.i8.i32(<vscale x 8 x i16> [[MASKEDOFF:%.*]], <vscale x 8 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulu_vx_u16m2_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vwmulu.mask.nxv8i16.nxv8i8.i8.i64(<vscale x 8 x i16> [[MASKEDOFF:%.*]], <vscale x 8 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
+//
+vuint16m2_t test_vwmulu_vx_u16m2_m(vbool8_t mask, vuint16m2_t maskedoff, vuint8m1_t op1, uint8_t op2, size_t vl) {
+  return vwmulu(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulu_vv_u16m4_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vwmulu.mask.nxv16i16.nxv16i8.nxv16i8.i32(<vscale x 16 x i16> [[MASKEDOFF:%.*]], <vscale x 16 x i8> [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 16 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulu_vv_u16m4_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vwmulu.mask.nxv16i16.nxv16i8.nxv16i8.i64(<vscale x 16 x i16> [[MASKEDOFF:%.*]], <vscale x 16 x i8> [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 16 x i16> [[TMP0]]
+//
+vuint16m4_t test_vwmulu_vv_u16m4_m(vbool4_t mask, vuint16m4_t maskedoff, vuint8m2_t op1, vuint8m2_t op2, size_t vl) {
+  return vwmulu(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulu_vx_u16m4_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vwmulu.mask.nxv16i16.nxv16i8.i8.i32(<vscale x 16 x i16> [[MASKEDOFF:%.*]], <vscale x 16 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 16 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulu_vx_u16m4_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vwmulu.mask.nxv16i16.nxv16i8.i8.i64(<vscale x 16 x i16> [[MASKEDOFF:%.*]], <vscale x 16 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 16 x i16> [[TMP0]]
+//
+vuint16m4_t test_vwmulu_vx_u16m4_m(vbool4_t mask, vuint16m4_t maskedoff, vuint8m2_t op1, uint8_t op2, size_t vl) {
+  return vwmulu(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulu_vv_u16m8_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vwmulu.mask.nxv32i16.nxv32i8.nxv32i8.i32(<vscale x 32 x i16> [[MASKEDOFF:%.*]], <vscale x 32 x i8> [[OP1:%.*]], <vscale x 32 x i8> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 32 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulu_vv_u16m8_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vwmulu.mask.nxv32i16.nxv32i8.nxv32i8.i64(<vscale x 32 x i16> [[MASKEDOFF:%.*]], <vscale x 32 x i8> [[OP1:%.*]], <vscale x 32 x i8> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 32 x i16> [[TMP0]]
+//
+vuint16m8_t test_vwmulu_vv_u16m8_m(vbool2_t mask, vuint16m8_t maskedoff, vuint8m4_t op1, vuint8m4_t op2, size_t vl) {
+  return vwmulu(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulu_vx_u16m8_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vwmulu.mask.nxv32i16.nxv32i8.i8.i32(<vscale x 32 x i16> [[MASKEDOFF:%.*]], <vscale x 32 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 32 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulu_vx_u16m8_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vwmulu.mask.nxv32i16.nxv32i8.i8.i64(<vscale x 32 x i16> [[MASKEDOFF:%.*]], <vscale x 32 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 32 x i16> [[TMP0]]
+//
+vuint16m8_t test_vwmulu_vx_u16m8_m(vbool2_t mask, vuint16m8_t maskedoff, vuint8m4_t op1, uint8_t op2, size_t vl) {
+  return vwmulu(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulu_vv_u32mf2_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vwmulu.mask.nxv1i32.nxv1i16.nxv1i16.i32(<vscale x 1 x i32> [[MASKEDOFF:%.*]], <vscale x 1 x i16> [[OP1:%.*]], <vscale x 1 x i16> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulu_vv_u32mf2_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vwmulu.mask.nxv1i32.nxv1i16.nxv1i16.i64(<vscale x 1 x i32> [[MASKEDOFF:%.*]], <vscale x 1 x i16> [[OP1:%.*]], <vscale x 1 x i16> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
+//
+vuint32mf2_t test_vwmulu_vv_u32mf2_m(vbool64_t mask, vuint32mf2_t maskedoff, vuint16mf4_t op1, vuint16mf4_t op2, size_t vl) {
+  return vwmulu(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulu_vx_u32mf2_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vwmulu.mask.nxv1i32.nxv1i16.i16.i32(<vscale x 1 x i32> [[MASKEDOFF:%.*]], <vscale x 1 x i16> [[OP1:%.*]], i16 [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulu_vx_u32mf2_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vwmulu.mask.nxv1i32.nxv1i16.i16.i64(<vscale x 1 x i32> [[MASKEDOFF:%.*]], <vscale x 1 x i16> [[OP1:%.*]], i16 [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
+//
+vuint32mf2_t test_vwmulu_vx_u32mf2_m(vbool64_t mask, vuint32mf2_t maskedoff, vuint16mf4_t op1, uint16_t op2, size_t vl) {
+  return vwmulu(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulu_vv_u32m1_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vwmulu.mask.nxv2i32.nxv2i16.nxv2i16.i32(<vscale x 2 x i32> [[MASKEDOFF:%.*]], <vscale x 2 x i16> [[OP1:%.*]], <vscale x 2 x i16> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulu_vv_u32m1_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vwmulu.mask.nxv2i32.nxv2i16.nxv2i16.i64(<vscale x 2 x i32> [[MASKEDOFF:%.*]], <vscale x 2 x i16> [[OP1:%.*]], <vscale x 2 x i16> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
+//
+vuint32m1_t test_vwmulu_vv_u32m1_m(vbool32_t mask, vuint32m1_t maskedoff, vuint16mf2_t op1, vuint16mf2_t op2, size_t vl) {
+  return vwmulu(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulu_vx_u32m1_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vwmulu.mask.nxv2i32.nxv2i16.i16.i32(<vscale x 2 x i32> [[MASKEDOFF:%.*]], <vscale x 2 x i16> [[OP1:%.*]], i16 [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulu_vx_u32m1_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vwmulu.mask.nxv2i32.nxv2i16.i16.i64(<vscale x 2 x i32> [[MASKEDOFF:%.*]], <vscale x 2 x i16> [[OP1:%.*]], i16 [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
+//
+vuint32m1_t test_vwmulu_vx_u32m1_m(vbool32_t mask, vuint32m1_t maskedoff, vuint16mf2_t op1, uint16_t op2, size_t vl) {
+  return vwmulu(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulu_vv_u32m2_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vwmulu.mask.nxv4i32.nxv4i16.nxv4i16.i32(<vscale x 4 x i32> [[MASKEDOFF:%.*]], <vscale x 4 x i16> [[OP1:%.*]], <vscale x 4 x i16> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulu_vv_u32m2_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vwmulu.mask.nxv4i32.nxv4i16.nxv4i16.i64(<vscale x 4 x i32> [[MASKEDOFF:%.*]], <vscale x 4 x i16> [[OP1:%.*]], <vscale x 4 x i16> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
+//
+vuint32m2_t test_vwmulu_vv_u32m2_m(vbool16_t mask, vuint32m2_t maskedoff, vuint16m1_t op1, vuint16m1_t op2, size_t vl) {
+  return vwmulu(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulu_vx_u32m2_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vwmulu.mask.nxv4i32.nxv4i16.i16.i32(<vscale x 4 x i32> [[MASKEDOFF:%.*]], <vscale x 4 x i16> [[OP1:%.*]], i16 [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulu_vx_u32m2_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vwmulu.mask.nxv4i32.nxv4i16.i16.i64(<vscale x 4 x i32> [[MASKEDOFF:%.*]], <vscale x 4 x i16> [[OP1:%.*]], i16 [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
+//
+vuint32m2_t test_vwmulu_vx_u32m2_m(vbool16_t mask, vuint32m2_t maskedoff, vuint16m1_t op1, uint16_t op2, size_t vl) {
+  return vwmulu(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulu_vv_u32m4_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vwmulu.mask.nxv8i32.nxv8i16.nxv8i16.i32(<vscale x 8 x i32> [[MASKEDOFF:%.*]], <vscale x 8 x i16> [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulu_vv_u32m4_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vwmulu.mask.nxv8i32.nxv8i16.nxv8i16.i64(<vscale x 8 x i32> [[MASKEDOFF:%.*]], <vscale x 8 x i16> [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
+//
+vuint32m4_t test_vwmulu_vv_u32m4_m(vbool8_t mask, vuint32m4_t maskedoff, vuint16m2_t op1, vuint16m2_t op2, size_t vl) {
+  return vwmulu(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulu_vx_u32m4_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vwmulu.mask.nxv8i32.nxv8i16.i16.i32(<vscale x 8 x i32> [[MASKEDOFF:%.*]], <vscale x 8 x i16> [[OP1:%.*]], i16 [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulu_vx_u32m4_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vwmulu.mask.nxv8i32.nxv8i16.i16.i64(<vscale x 8 x i32> [[MASKEDOFF:%.*]], <vscale x 8 x i16> [[OP1:%.*]], i16 [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
+//
+vuint32m4_t test_vwmulu_vx_u32m4_m(vbool8_t mask, vuint32m4_t maskedoff, vuint16m2_t op1, uint16_t op2, size_t vl) {
+  return vwmulu(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulu_vv_u32m8_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vwmulu.mask.nxv16i32.nxv16i16.nxv16i16.i32(<vscale x 16 x i32> [[MASKEDOFF:%.*]], <vscale x 16 x i16> [[OP1:%.*]], <vscale x 16 x i16> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulu_vv_u32m8_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vwmulu.mask.nxv16i32.nxv16i16.nxv16i16.i64(<vscale x 16 x i32> [[MASKEDOFF:%.*]], <vscale x 16 x i16> [[OP1:%.*]], <vscale x 16 x i16> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
+//
+vuint32m8_t test_vwmulu_vv_u32m8_m(vbool4_t mask, vuint32m8_t maskedoff, vuint16m4_t op1, vuint16m4_t op2, size_t vl) {
+  return vwmulu(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulu_vx_u32m8_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vwmulu.mask.nxv16i32.nxv16i16.i16.i32(<vscale x 16 x i32> [[MASKEDOFF:%.*]], <vscale x 16 x i16> [[OP1:%.*]], i16 [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulu_vx_u32m8_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vwmulu.mask.nxv16i32.nxv16i16.i16.i64(<vscale x 16 x i32> [[MASKEDOFF:%.*]], <vscale x 16 x i16> [[OP1:%.*]], i16 [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
+//
+vuint32m8_t test_vwmulu_vx_u32m8_m(vbool4_t mask, vuint32m8_t maskedoff, vuint16m4_t op1, uint16_t op2, size_t vl) {
+  return vwmulu(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulu_vv_u64m1_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vwmulu.mask.nxv1i64.nxv1i32.nxv1i32.i32(<vscale x 1 x i64> [[MASKEDOFF:%.*]], <vscale x 1 x i32> [[OP1:%.*]], <vscale x 1 x i32> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 1 x i64> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulu_vv_u64m1_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vwmulu.mask.nxv1i64.nxv1i32.nxv1i32.i64(<vscale x 1 x i64> [[MASKEDOFF:%.*]], <vscale x 1 x i32> [[OP1:%.*]], <vscale x 1 x i32> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 1 x i64> [[TMP0]]
+//
+vuint64m1_t test_vwmulu_vv_u64m1_m(vbool64_t mask, vuint64m1_t maskedoff, vuint32mf2_t op1, vuint32mf2_t op2, size_t vl) {
+  return vwmulu(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulu_vx_u64m1_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vwmulu.mask.nxv1i64.nxv1i32.i32.i32(<vscale x 1 x i64> [[MASKEDOFF:%.*]], <vscale x 1 x i32> [[OP1:%.*]], i32 [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 1 x i64> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulu_vx_u64m1_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vwmulu.mask.nxv1i64.nxv1i32.i32.i64(<vscale x 1 x i64> [[MASKEDOFF:%.*]], <vscale x 1 x i32> [[OP1:%.*]], i32 [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 1 x i64> [[TMP0]]
+//
+vuint64m1_t test_vwmulu_vx_u64m1_m(vbool64_t mask, vuint64m1_t maskedoff, vuint32mf2_t op1, uint32_t op2, size_t vl) {
+  return vwmulu(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulu_vv_u64m2_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vwmulu.mask.nxv2i64.nxv2i32.nxv2i32.i32(<vscale x 2 x i64> [[MASKEDOFF:%.*]], <vscale x 2 x i32> [[OP1:%.*]], <vscale x 2 x i32> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 2 x i64> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulu_vv_u64m2_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vwmulu.mask.nxv2i64.nxv2i32.nxv2i32.i64(<vscale x 2 x i64> [[MASKEDOFF:%.*]], <vscale x 2 x i32> [[OP1:%.*]], <vscale x 2 x i32> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 2 x i64> [[TMP0]]
+//
+vuint64m2_t test_vwmulu_vv_u64m2_m(vbool32_t mask, vuint64m2_t maskedoff, vuint32m1_t op1, vuint32m1_t op2, size_t vl) {
+  return vwmulu(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulu_vx_u64m2_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vwmulu.mask.nxv2i64.nxv2i32.i32.i32(<vscale x 2 x i64> [[MASKEDOFF:%.*]], <vscale x 2 x i32> [[OP1:%.*]], i32 [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 2 x i64> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulu_vx_u64m2_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vwmulu.mask.nxv2i64.nxv2i32.i32.i64(<vscale x 2 x i64> [[MASKEDOFF:%.*]], <vscale x 2 x i32> [[OP1:%.*]], i32 [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 2 x i64> [[TMP0]]
+//
+vuint64m2_t test_vwmulu_vx_u64m2_m(vbool32_t mask, vuint64m2_t maskedoff, vuint32m1_t op1, uint32_t op2, size_t vl) {
+  return vwmulu(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulu_vv_u64m4_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vwmulu.mask.nxv4i64.nxv4i32.nxv4i32.i32(<vscale x 4 x i64> [[MASKEDOFF:%.*]], <vscale x 4 x i32> [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 4 x i64> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulu_vv_u64m4_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vwmulu.mask.nxv4i64.nxv4i32.nxv4i32.i64(<vscale x 4 x i64> [[MASKEDOFF:%.*]], <vscale x 4 x i32> [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 4 x i64> [[TMP0]]
+//
+vuint64m4_t test_vwmulu_vv_u64m4_m(vbool16_t mask, vuint64m4_t maskedoff, vuint32m2_t op1, vuint32m2_t op2, size_t vl) {
+  return vwmulu(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulu_vx_u64m4_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vwmulu.mask.nxv4i64.nxv4i32.i32.i32(<vscale x 4 x i64> [[MASKEDOFF:%.*]], <vscale x 4 x i32> [[OP1:%.*]], i32 [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 4 x i64> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulu_vx_u64m4_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vwmulu.mask.nxv4i64.nxv4i32.i32.i64(<vscale x 4 x i64> [[MASKEDOFF:%.*]], <vscale x 4 x i32> [[OP1:%.*]], i32 [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 4 x i64> [[TMP0]]
+//
+vuint64m4_t test_vwmulu_vx_u64m4_m(vbool16_t mask, vuint64m4_t maskedoff, vuint32m2_t op1, uint32_t op2, size_t vl) {
+  return vwmulu(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulu_vv_u64m8_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vwmulu.mask.nxv8i64.nxv8i32.nxv8i32.i32(<vscale x 8 x i64> [[MASKEDOFF:%.*]], <vscale x 8 x i32> [[OP1:%.*]], <vscale x 8 x i32> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 8 x i64> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulu_vv_u64m8_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vwmulu.mask.nxv8i64.nxv8i32.nxv8i32.i64(<vscale x 8 x i64> [[MASKEDOFF:%.*]], <vscale x 8 x i32> [[OP1:%.*]], <vscale x 8 x i32> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 8 x i64> [[TMP0]]
+//
+vuint64m8_t test_vwmulu_vv_u64m8_m(vbool8_t mask, vuint64m8_t maskedoff, vuint32m4_t op1, vuint32m4_t op2, size_t vl) {
+  return vwmulu(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulu_vx_u64m8_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vwmulu.mask.nxv8i64.nxv8i32.i32.i32(<vscale x 8 x i64> [[MASKEDOFF:%.*]], <vscale x 8 x i32> [[OP1:%.*]], i32 [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 8 x i64> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulu_vx_u64m8_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vwmulu.mask.nxv8i64.nxv8i32.i32.i64(<vscale x 8 x i64> [[MASKEDOFF:%.*]], <vscale x 8 x i32> [[OP1:%.*]], i32 [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 8 x i64> [[TMP0]]
+//
+vuint64m8_t test_vwmulu_vx_u64m8_m(vbool8_t mask, vuint64m8_t maskedoff, vuint32m4_t op1, uint32_t op2, size_t vl) {
+  return vwmulu(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulsu_vv_i16mf4_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vwmulsu.mask.nxv1i16.nxv1i8.nxv1i8.i32(<vscale x 1 x i16> [[MASKEDOFF:%.*]], <vscale x 1 x i8> [[OP1:%.*]], <vscale x 1 x i8> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 1 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulsu_vv_i16mf4_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vwmulsu.mask.nxv1i16.nxv1i8.nxv1i8.i64(<vscale x 1 x i16> [[MASKEDOFF:%.*]], <vscale x 1 x i8> [[OP1:%.*]], <vscale x 1 x i8> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 1 x i16> [[TMP0]]
+//
+vint16mf4_t test_vwmulsu_vv_i16mf4_m(vbool64_t mask, vint16mf4_t maskedoff, vint8mf8_t op1, vuint8mf8_t op2, size_t vl) {
+  return vwmulsu(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulsu_vx_i16mf4_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vwmulsu.mask.nxv1i16.nxv1i8.i8.i32(<vscale x 1 x i16> [[MASKEDOFF:%.*]], <vscale x 1 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 1 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulsu_vx_i16mf4_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vwmulsu.mask.nxv1i16.nxv1i8.i8.i64(<vscale x 1 x i16> [[MASKEDOFF:%.*]], <vscale x 1 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 1 x i16> [[TMP0]]
+//
+vint16mf4_t test_vwmulsu_vx_i16mf4_m(vbool64_t mask, vint16mf4_t maskedoff, vint8mf8_t op1, uint8_t op2, size_t vl) {
+  return vwmulsu(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulsu_vv_i16mf2_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vwmulsu.mask.nxv2i16.nxv2i8.nxv2i8.i32(<vscale x 2 x i16> [[MASKEDOFF:%.*]], <vscale x 2 x i8> [[OP1:%.*]], <vscale x 2 x i8> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 2 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulsu_vv_i16mf2_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vwmulsu.mask.nxv2i16.nxv2i8.nxv2i8.i64(<vscale x 2 x i16> [[MASKEDOFF:%.*]], <vscale x 2 x i8> [[OP1:%.*]], <vscale x 2 x i8> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 2 x i16> [[TMP0]]
+//
+vint16mf2_t test_vwmulsu_vv_i16mf2_m(vbool32_t mask, vint16mf2_t maskedoff, vint8mf4_t op1, vuint8mf4_t op2, size_t vl) {
+  return vwmulsu(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulsu_vx_i16mf2_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vwmulsu.mask.nxv2i16.nxv2i8.i8.i32(<vscale x 2 x i16> [[MASKEDOFF:%.*]], <vscale x 2 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 2 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulsu_vx_i16mf2_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vwmulsu.mask.nxv2i16.nxv2i8.i8.i64(<vscale x 2 x i16> [[MASKEDOFF:%.*]], <vscale x 2 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 2 x i16> [[TMP0]]
+//
+vint16mf2_t test_vwmulsu_vx_i16mf2_m(vbool32_t mask, vint16mf2_t maskedoff, vint8mf4_t op1, uint8_t op2, size_t vl) {
+  return vwmulsu(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulsu_vv_i16m1_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vwmulsu.mask.nxv4i16.nxv4i8.nxv4i8.i32(<vscale x 4 x i16> [[MASKEDOFF:%.*]], <vscale x 4 x i8> [[OP1:%.*]], <vscale x 4 x i8> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 4 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulsu_vv_i16m1_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vwmulsu.mask.nxv4i16.nxv4i8.nxv4i8.i64(<vscale x 4 x i16> [[MASKEDOFF:%.*]], <vscale x 4 x i8> [[OP1:%.*]], <vscale x 4 x i8> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 4 x i16> [[TMP0]]
+//
+vint16m1_t test_vwmulsu_vv_i16m1_m(vbool16_t mask, vint16m1_t maskedoff, vint8mf2_t op1, vuint8mf2_t op2, size_t vl) {
+  return vwmulsu(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulsu_vx_i16m1_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vwmulsu.mask.nxv4i16.nxv4i8.i8.i32(<vscale x 4 x i16> [[MASKEDOFF:%.*]], <vscale x 4 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 4 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulsu_vx_i16m1_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vwmulsu.mask.nxv4i16.nxv4i8.i8.i64(<vscale x 4 x i16> [[MASKEDOFF:%.*]], <vscale x 4 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 4 x i16> [[TMP0]]
+//
+vint16m1_t test_vwmulsu_vx_i16m1_m(vbool16_t mask, vint16m1_t maskedoff, vint8mf2_t op1, uint8_t op2, size_t vl) {
+  return vwmulsu(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulsu_vv_i16m2_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vwmulsu.mask.nxv8i16.nxv8i8.nxv8i8.i32(<vscale x 8 x i16> [[MASKEDOFF:%.*]], <vscale x 8 x i8> [[OP1:%.*]], <vscale x 8 x i8> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulsu_vv_i16m2_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vwmulsu.mask.nxv8i16.nxv8i8.nxv8i8.i64(<vscale x 8 x i16> [[MASKEDOFF:%.*]], <vscale x 8 x i8> [[OP1:%.*]], <vscale x 8 x i8> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
+//
+vint16m2_t test_vwmulsu_vv_i16m2_m(vbool8_t mask, vint16m2_t maskedoff, vint8m1_t op1, vuint8m1_t op2, size_t vl) {
+  return vwmulsu(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulsu_vx_i16m2_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vwmulsu.mask.nxv8i16.nxv8i8.i8.i32(<vscale x 8 x i16> [[MASKEDOFF:%.*]], <vscale x 8 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulsu_vx_i16m2_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vwmulsu.mask.nxv8i16.nxv8i8.i8.i64(<vscale x 8 x i16> [[MASKEDOFF:%.*]], <vscale x 8 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
+//
+vint16m2_t test_vwmulsu_vx_i16m2_m(vbool8_t mask, vint16m2_t maskedoff, vint8m1_t op1, uint8_t op2, size_t vl) {
+  return vwmulsu(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulsu_vv_i16m4_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vwmulsu.mask.nxv16i16.nxv16i8.nxv16i8.i32(<vscale x 16 x i16> [[MASKEDOFF:%.*]], <vscale x 16 x i8> [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 16 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulsu_vv_i16m4_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vwmulsu.mask.nxv16i16.nxv16i8.nxv16i8.i64(<vscale x 16 x i16> [[MASKEDOFF:%.*]], <vscale x 16 x i8> [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 16 x i16> [[TMP0]]
+//
+vint16m4_t test_vwmulsu_vv_i16m4_m(vbool4_t mask, vint16m4_t maskedoff, vint8m2_t op1, vuint8m2_t op2, size_t vl) {
+  return vwmulsu(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulsu_vx_i16m4_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vwmulsu.mask.nxv16i16.nxv16i8.i8.i32(<vscale x 16 x i16> [[MASKEDOFF:%.*]], <vscale x 16 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 16 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulsu_vx_i16m4_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vwmulsu.mask.nxv16i16.nxv16i8.i8.i64(<vscale x 16 x i16> [[MASKEDOFF:%.*]], <vscale x 16 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 16 x i16> [[TMP0]]
+//
+vint16m4_t test_vwmulsu_vx_i16m4_m(vbool4_t mask, vint16m4_t maskedoff, vint8m2_t op1, uint8_t op2, size_t vl) {
+  return vwmulsu(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulsu_vv_i16m8_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vwmulsu.mask.nxv32i16.nxv32i8.nxv32i8.i32(<vscale x 32 x i16> [[MASKEDOFF:%.*]], <vscale x 32 x i8> [[OP1:%.*]], <vscale x 32 x i8> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 32 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulsu_vv_i16m8_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vwmulsu.mask.nxv32i16.nxv32i8.nxv32i8.i64(<vscale x 32 x i16> [[MASKEDOFF:%.*]], <vscale x 32 x i8> [[OP1:%.*]], <vscale x 32 x i8> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 32 x i16> [[TMP0]]
+//
+vint16m8_t test_vwmulsu_vv_i16m8_m(vbool2_t mask, vint16m8_t maskedoff, vint8m4_t op1, vuint8m4_t op2, size_t vl) {
+  return vwmulsu(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulsu_vx_i16m8_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vwmulsu.mask.nxv32i16.nxv32i8.i8.i32(<vscale x 32 x i16> [[MASKEDOFF:%.*]], <vscale x 32 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 32 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulsu_vx_i16m8_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vwmulsu.mask.nxv32i16.nxv32i8.i8.i64(<vscale x 32 x i16> [[MASKEDOFF:%.*]], <vscale x 32 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 32 x i16> [[TMP0]]
+//
+vint16m8_t test_vwmulsu_vx_i16m8_m(vbool2_t mask, vint16m8_t maskedoff, vint8m4_t op1, uint8_t op2, size_t vl) {
+  return vwmulsu(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulsu_vv_i32mf2_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vwmulsu.mask.nxv1i32.nxv1i16.nxv1i16.i32(<vscale x 1 x i32> [[MASKEDOFF:%.*]], <vscale x 1 x i16> [[OP1:%.*]], <vscale x 1 x i16> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulsu_vv_i32mf2_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vwmulsu.mask.nxv1i32.nxv1i16.nxv1i16.i64(<vscale x 1 x i32> [[MASKEDOFF:%.*]], <vscale x 1 x i16> [[OP1:%.*]], <vscale x 1 x i16> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
+//
+vint32mf2_t test_vwmulsu_vv_i32mf2_m(vbool64_t mask, vint32mf2_t maskedoff, vint16mf4_t op1, vuint16mf4_t op2, size_t vl) {
+  return vwmulsu(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulsu_vx_i32mf2_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vwmulsu.mask.nxv1i32.nxv1i16.i16.i32(<vscale x 1 x i32> [[MASKEDOFF:%.*]], <vscale x 1 x i16> [[OP1:%.*]], i16 [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulsu_vx_i32mf2_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vwmulsu.mask.nxv1i32.nxv1i16.i16.i64(<vscale x 1 x i32> [[MASKEDOFF:%.*]], <vscale x 1 x i16> [[OP1:%.*]], i16 [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
+//
+vint32mf2_t test_vwmulsu_vx_i32mf2_m(vbool64_t mask, vint32mf2_t maskedoff, vint16mf4_t op1, uint16_t op2, size_t vl) {
+  return vwmulsu(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulsu_vv_i32m1_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vwmulsu.mask.nxv2i32.nxv2i16.nxv2i16.i32(<vscale x 2 x i32> [[MASKEDOFF:%.*]], <vscale x 2 x i16> [[OP1:%.*]], <vscale x 2 x i16> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulsu_vv_i32m1_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vwmulsu.mask.nxv2i32.nxv2i16.nxv2i16.i64(<vscale x 2 x i32> [[MASKEDOFF:%.*]], <vscale x 2 x i16> [[OP1:%.*]], <vscale x 2 x i16> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
+//
+vint32m1_t test_vwmulsu_vv_i32m1_m(vbool32_t mask, vint32m1_t maskedoff, vint16mf2_t op1, vuint16mf2_t op2, size_t vl) {
+  return vwmulsu(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulsu_vx_i32m1_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vwmulsu.mask.nxv2i32.nxv2i16.i16.i32(<vscale x 2 x i32> [[MASKEDOFF:%.*]], <vscale x 2 x i16> [[OP1:%.*]], i16 [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulsu_vx_i32m1_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vwmulsu.mask.nxv2i32.nxv2i16.i16.i64(<vscale x 2 x i32> [[MASKEDOFF:%.*]], <vscale x 2 x i16> [[OP1:%.*]], i16 [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
+//
+vint32m1_t test_vwmulsu_vx_i32m1_m(vbool32_t mask, vint32m1_t maskedoff, vint16mf2_t op1, uint16_t op2, size_t vl) {
+  return vwmulsu(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulsu_vv_i32m2_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vwmulsu.mask.nxv4i32.nxv4i16.nxv4i16.i32(<vscale x 4 x i32> [[MASKEDOFF:%.*]], <vscale x 4 x i16> [[OP1:%.*]], <vscale x 4 x i16> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulsu_vv_i32m2_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vwmulsu.mask.nxv4i32.nxv4i16.nxv4i16.i64(<vscale x 4 x i32> [[MASKEDOFF:%.*]], <vscale x 4 x i16> [[OP1:%.*]], <vscale x 4 x i16> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
+//
+vint32m2_t test_vwmulsu_vv_i32m2_m(vbool16_t mask, vint32m2_t maskedoff, vint16m1_t op1, vuint16m1_t op2, size_t vl) {
+  return vwmulsu(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulsu_vx_i32m2_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vwmulsu.mask.nxv4i32.nxv4i16.i16.i32(<vscale x 4 x i32> [[MASKEDOFF:%.*]], <vscale x 4 x i16> [[OP1:%.*]], i16 [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulsu_vx_i32m2_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vwmulsu.mask.nxv4i32.nxv4i16.i16.i64(<vscale x 4 x i32> [[MASKEDOFF:%.*]], <vscale x 4 x i16> [[OP1:%.*]], i16 [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
+//
+vint32m2_t test_vwmulsu_vx_i32m2_m(vbool16_t mask, vint32m2_t maskedoff, vint16m1_t op1, uint16_t op2, size_t vl) {
+  return vwmulsu(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulsu_vv_i32m4_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vwmulsu.mask.nxv8i32.nxv8i16.nxv8i16.i32(<vscale x 8 x i32> [[MASKEDOFF:%.*]], <vscale x 8 x i16> [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulsu_vv_i32m4_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vwmulsu.mask.nxv8i32.nxv8i16.nxv8i16.i64(<vscale x 8 x i32> [[MASKEDOFF:%.*]], <vscale x 8 x i16> [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
+//
+vint32m4_t test_vwmulsu_vv_i32m4_m(vbool8_t mask, vint32m4_t maskedoff, vint16m2_t op1, vuint16m2_t op2, size_t vl) {
+  return vwmulsu(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulsu_vx_i32m4_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vwmulsu.mask.nxv8i32.nxv8i16.i16.i32(<vscale x 8 x i32> [[MASKEDOFF:%.*]], <vscale x 8 x i16> [[OP1:%.*]], i16 [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulsu_vx_i32m4_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vwmulsu.mask.nxv8i32.nxv8i16.i16.i64(<vscale x 8 x i32> [[MASKEDOFF:%.*]], <vscale x 8 x i16> [[OP1:%.*]], i16 [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
+//
+vint32m4_t test_vwmulsu_vx_i32m4_m(vbool8_t mask, vint32m4_t maskedoff, vint16m2_t op1, uint16_t op2, size_t vl) {
+  return vwmulsu(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulsu_vv_i32m8_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vwmulsu.mask.nxv16i32.nxv16i16.nxv16i16.i32(<vscale x 16 x i32> [[MASKEDOFF:%.*]], <vscale x 16 x i16> [[OP1:%.*]], <vscale x 16 x i16> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulsu_vv_i32m8_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vwmulsu.mask.nxv16i32.nxv16i16.nxv16i16.i64(<vscale x 16 x i32> [[MASKEDOFF:%.*]], <vscale x 16 x i16> [[OP1:%.*]], <vscale x 16 x i16> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
+//
+vint32m8_t test_vwmulsu_vv_i32m8_m(vbool4_t mask, vint32m8_t maskedoff, vint16m4_t op1, vuint16m4_t op2, size_t vl) {
+  return vwmulsu(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulsu_vx_i32m8_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vwmulsu.mask.nxv16i32.nxv16i16.i16.i32(<vscale x 16 x i32> [[MASKEDOFF:%.*]], <vscale x 16 x i16> [[OP1:%.*]], i16 [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulsu_vx_i32m8_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vwmulsu.mask.nxv16i32.nxv16i16.i16.i64(<vscale x 16 x i32> [[MASKEDOFF:%.*]], <vscale x 16 x i16> [[OP1:%.*]], i16 [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
+//
+vint32m8_t test_vwmulsu_vx_i32m8_m(vbool4_t mask, vint32m8_t maskedoff, vint16m4_t op1, uint16_t op2, size_t vl) {
+  return vwmulsu(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulsu_vv_i64m1_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vwmulsu.mask.nxv1i64.nxv1i32.nxv1i32.i32(<vscale x 1 x i64> [[MASKEDOFF:%.*]], <vscale x 1 x i32> [[OP1:%.*]], <vscale x 1 x i32> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 1 x i64> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulsu_vv_i64m1_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vwmulsu.mask.nxv1i64.nxv1i32.nxv1i32.i64(<vscale x 1 x i64> [[MASKEDOFF:%.*]], <vscale x 1 x i32> [[OP1:%.*]], <vscale x 1 x i32> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 1 x i64> [[TMP0]]
+//
+vint64m1_t test_vwmulsu_vv_i64m1_m(vbool64_t mask, vint64m1_t maskedoff, vint32mf2_t op1, vuint32mf2_t op2, size_t vl) {
+  return vwmulsu(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulsu_vx_i64m1_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vwmulsu.mask.nxv1i64.nxv1i32.i32.i32(<vscale x 1 x i64> [[MASKEDOFF:%.*]], <vscale x 1 x i32> [[OP1:%.*]], i32 [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 1 x i64> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulsu_vx_i64m1_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vwmulsu.mask.nxv1i64.nxv1i32.i32.i64(<vscale x 1 x i64> [[MASKEDOFF:%.*]], <vscale x 1 x i32> [[OP1:%.*]], i32 [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 1 x i64> [[TMP0]]
+//
+vint64m1_t test_vwmulsu_vx_i64m1_m(vbool64_t mask, vint64m1_t maskedoff, vint32mf2_t op1, uint32_t op2, size_t vl) {
+  return vwmulsu(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulsu_vv_i64m2_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vwmulsu.mask.nxv2i64.nxv2i32.nxv2i32.i32(<vscale x 2 x i64> [[MASKEDOFF:%.*]], <vscale x 2 x i32> [[OP1:%.*]], <vscale x 2 x i32> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 2 x i64> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulsu_vv_i64m2_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vwmulsu.mask.nxv2i64.nxv2i32.nxv2i32.i64(<vscale x 2 x i64> [[MASKEDOFF:%.*]], <vscale x 2 x i32> [[OP1:%.*]], <vscale x 2 x i32> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 2 x i64> [[TMP0]]
+//
+vint64m2_t test_vwmulsu_vv_i64m2_m(vbool32_t mask, vint64m2_t maskedoff, vint32m1_t op1, vuint32m1_t op2, size_t vl) {
+  return vwmulsu(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulsu_vx_i64m2_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vwmulsu.mask.nxv2i64.nxv2i32.i32.i32(<vscale x 2 x i64> [[MASKEDOFF:%.*]], <vscale x 2 x i32> [[OP1:%.*]], i32 [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 2 x i64> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulsu_vx_i64m2_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vwmulsu.mask.nxv2i64.nxv2i32.i32.i64(<vscale x 2 x i64> [[MASKEDOFF:%.*]], <vscale x 2 x i32> [[OP1:%.*]], i32 [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 2 x i64> [[TMP0]]
+//
+vint64m2_t test_vwmulsu_vx_i64m2_m(vbool32_t mask, vint64m2_t maskedoff, vint32m1_t op1, uint32_t op2, size_t vl) {
+  return vwmulsu(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulsu_vv_i64m4_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vwmulsu.mask.nxv4i64.nxv4i32.nxv4i32.i32(<vscale x 4 x i64> [[MASKEDOFF:%.*]], <vscale x 4 x i32> [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 4 x i64> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulsu_vv_i64m4_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vwmulsu.mask.nxv4i64.nxv4i32.nxv4i32.i64(<vscale x 4 x i64> [[MASKEDOFF:%.*]], <vscale x 4 x i32> [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 4 x i64> [[TMP0]]
+//
+vint64m4_t test_vwmulsu_vv_i64m4_m(vbool16_t mask, vint64m4_t maskedoff, vint32m2_t op1, vuint32m2_t op2, size_t vl) {
+  return vwmulsu(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulsu_vx_i64m4_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vwmulsu.mask.nxv4i64.nxv4i32.i32.i32(<vscale x 4 x i64> [[MASKEDOFF:%.*]], <vscale x 4 x i32> [[OP1:%.*]], i32 [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 4 x i64> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulsu_vx_i64m4_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vwmulsu.mask.nxv4i64.nxv4i32.i32.i64(<vscale x 4 x i64> [[MASKEDOFF:%.*]], <vscale x 4 x i32> [[OP1:%.*]], i32 [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 4 x i64> [[TMP0]]
+//
+vint64m4_t test_vwmulsu_vx_i64m4_m(vbool16_t mask, vint64m4_t maskedoff, vint32m2_t op1, uint32_t op2, size_t vl) {
+  return vwmulsu(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulsu_vv_i64m8_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vwmulsu.mask.nxv8i64.nxv8i32.nxv8i32.i32(<vscale x 8 x i64> [[MASKEDOFF:%.*]], <vscale x 8 x i32> [[OP1:%.*]], <vscale x 8 x i32> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 8 x i64> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulsu_vv_i64m8_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vwmulsu.mask.nxv8i64.nxv8i32.nxv8i32.i64(<vscale x 8 x i64> [[MASKEDOFF:%.*]], <vscale x 8 x i32> [[OP1:%.*]], <vscale x 8 x i32> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 8 x i64> [[TMP0]]
+//
+vint64m8_t test_vwmulsu_vv_i64m8_m(vbool8_t mask, vint64m8_t maskedoff, vint32m4_t op1, vuint32m4_t op2, size_t vl) {
+  return vwmulsu(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulsu_vx_i64m8_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vwmulsu.mask.nxv8i64.nxv8i32.i32.i32(<vscale x 8 x i64> [[MASKEDOFF:%.*]], <vscale x 8 x i32> [[OP1:%.*]], i32 [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i32 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV32-NEXT:    ret <vscale x 8 x i64> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulsu_vx_i64m8_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vwmulsu.mask.nxv8i64.nxv8i32.i32.i64(<vscale x 8 x i64> [[MASKEDOFF:%.*]], <vscale x 8 x i32> [[OP1:%.*]], i32 [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]]) [[ATTR7]]
+// CHECK-RV64-NEXT:    ret <vscale x 8 x i64> [[TMP0]]
+//
+vint64m8_t test_vwmulsu_vx_i64m8_m(vbool8_t mask, vint64m8_t maskedoff, vint32m4_t op1, uint32_t op2, size_t vl) {
+  return vwmulsu(mask, maskedoff, op1, op2, vl);
+}

clang/test/CodeGen/RISCV/rvv-intrinsics/vnsra.c

+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv32 -target-feature +experimental-v -disable-O0-optnone -emit-llvm %s -o - | opt -S -mem2reg | FileCheck --check-prefix=CHECK-RV32 %s
+// RUN: %clang_cc1 -triple riscv64 -target-feature +experimental-v -disable-O0-optnone -emit-llvm %s -o - | opt -S -mem2reg | FileCheck --check-prefix=CHECK-RV64 %s
+// RUN: %clang_cc1 -triple riscv64 -target-feature +m -target-feature +experimental-v -Werror -Wall -o - %s -S >/dev/null 2>&1 | FileCheck --check-prefix=ASM --allow-empty %s
+
+// ASM-NOT: warning
+#include <riscv_vector.h>
+
+// CHECK-RV32-LABEL: @test_vnsra_wv_i8mf8(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i8> @llvm.riscv.vnsra.nxv1i8.nxv1i16.nxv1i8.i32(<vscale x 1 x i16> [[OP1:%.*]], <vscale x 1 x i8> [[SHIFT:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 1 x i8> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsra_wv_i8mf8(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i8> @llvm.riscv.vnsra.nxv1i8.nxv1i16.nxv1i8.i64(<vscale x 1 x i16> [[OP1:%.*]], <vscale x 1 x i8> [[SHIFT:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 1 x i8> [[TMP0]]
+//
+vint8mf8_t test_vnsra_wv_i8mf8(vint16mf4_t op1, vuint8mf8_t shift, size_t vl) {
+  return vnsra_wv_i8mf8(op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsra_wx_i8mf8(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i8> @llvm.riscv.vnsra.nxv1i8.nxv1i16.i32.i32(<vscale x 1 x i16> [[OP1:%.*]], i32 [[SHIFT:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 1 x i8> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsra_wx_i8mf8(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i8> @llvm.riscv.vnsra.nxv1i8.nxv1i16.i64.i64(<vscale x 1 x i16> [[OP1:%.*]], i64 [[SHIFT:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 1 x i8> [[TMP0]]
+//
+vint8mf8_t test_vnsra_wx_i8mf8(vint16mf4_t op1, size_t shift, size_t vl) {
+  return vnsra_wx_i8mf8(op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsra_wv_i8mf4(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i8> @llvm.riscv.vnsra.nxv2i8.nxv2i16.nxv2i8.i32(<vscale x 2 x i16> [[OP1:%.*]], <vscale x 2 x i8> [[SHIFT:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 2 x i8> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsra_wv_i8mf4(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i8> @llvm.riscv.vnsra.nxv2i8.nxv2i16.nxv2i8.i64(<vscale x 2 x i16> [[OP1:%.*]], <vscale x 2 x i8> [[SHIFT:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 2 x i8> [[TMP0]]
+//
+vint8mf4_t test_vnsra_wv_i8mf4(vint16mf2_t op1, vuint8mf4_t shift, size_t vl) {
+  return vnsra_wv_i8mf4(op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsra_wx_i8mf4(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i8> @llvm.riscv.vnsra.nxv2i8.nxv2i16.i32.i32(<vscale x 2 x i16> [[OP1:%.*]], i32 [[SHIFT:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 2 x i8> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsra_wx_i8mf4(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i8> @llvm.riscv.vnsra.nxv2i8.nxv2i16.i64.i64(<vscale x 2 x i16> [[OP1:%.*]], i64 [[SHIFT:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 2 x i8> [[TMP0]]
+//
+vint8mf4_t test_vnsra_wx_i8mf4(vint16mf2_t op1, size_t shift, size_t vl) {
+  return vnsra_wx_i8mf4(op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsra_wv_i8mf2(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i8> @llvm.riscv.vnsra.nxv4i8.nxv4i16.nxv4i8.i32(<vscale x 4 x i16> [[OP1:%.*]], <vscale x 4 x i8> [[SHIFT:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 4 x i8> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsra_wv_i8mf2(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i8> @llvm.riscv.vnsra.nxv4i8.nxv4i16.nxv4i8.i64(<vscale x 4 x i16> [[OP1:%.*]], <vscale x 4 x i8> [[SHIFT:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 4 x i8> [[TMP0]]
+//
+vint8mf2_t test_vnsra_wv_i8mf2(vint16m1_t op1, vuint8mf2_t shift, size_t vl) {
+  return vnsra_wv_i8mf2(op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsra_wx_i8mf2(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i8> @llvm.riscv.vnsra.nxv4i8.nxv4i16.i32.i32(<vscale x 4 x i16> [[OP1:%.*]], i32 [[SHIFT:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 4 x i8> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsra_wx_i8mf2(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i8> @llvm.riscv.vnsra.nxv4i8.nxv4i16.i64.i64(<vscale x 4 x i16> [[OP1:%.*]], i64 [[SHIFT:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 4 x i8> [[TMP0]]
+//
+vint8mf2_t test_vnsra_wx_i8mf2(vint16m1_t op1, size_t shift, size_t vl) {
+  return vnsra_wx_i8mf2(op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsra_wv_i8m1(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i8> @llvm.riscv.vnsra.nxv8i8.nxv8i16.nxv8i8.i32(<vscale x 8 x i16> [[OP1:%.*]], <vscale x 8 x i8> [[SHIFT:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 8 x i8> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsra_wv_i8m1(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i8> @llvm.riscv.vnsra.nxv8i8.nxv8i16.nxv8i8.i64(<vscale x 8 x i16> [[OP1:%.*]], <vscale x 8 x i8> [[SHIFT:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 8 x i8> [[TMP0]]
+//
+vint8m1_t test_vnsra_wv_i8m1(vint16m2_t op1, vuint8m1_t shift, size_t vl) {
+  return vnsra_wv_i8m1(op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsra_wx_i8m1(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i8> @llvm.riscv.vnsra.nxv8i8.nxv8i16.i32.i32(<vscale x 8 x i16> [[OP1:%.*]], i32 [[SHIFT:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 8 x i8> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsra_wx_i8m1(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i8> @llvm.riscv.vnsra.nxv8i8.nxv8i16.i64.i64(<vscale x 8 x i16> [[OP1:%.*]], i64 [[SHIFT:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 8 x i8> [[TMP0]]
+//
+vint8m1_t test_vnsra_wx_i8m1(vint16m2_t op1, size_t shift, size_t vl) {
+  return vnsra_wx_i8m1(op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsra_wv_i8m2(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i8> @llvm.riscv.vnsra.nxv16i8.nxv16i16.nxv16i8.i32(<vscale x 16 x i16> [[OP1:%.*]], <vscale x 16 x i8> [[SHIFT:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 16 x i8> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsra_wv_i8m2(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i8> @llvm.riscv.vnsra.nxv16i8.nxv16i16.nxv16i8.i64(<vscale x 16 x i16> [[OP1:%.*]], <vscale x 16 x i8> [[SHIFT:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 16 x i8> [[TMP0]]
+//
+vint8m2_t test_vnsra_wv_i8m2(vint16m4_t op1, vuint8m2_t shift, size_t vl) {
+  return vnsra_wv_i8m2(op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsra_wx_i8m2(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i8> @llvm.riscv.vnsra.nxv16i8.nxv16i16.i32.i32(<vscale x 16 x i16> [[OP1:%.*]], i32 [[SHIFT:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 16 x i8> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsra_wx_i8m2(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i8> @llvm.riscv.vnsra.nxv16i8.nxv16i16.i64.i64(<vscale x 16 x i16> [[OP1:%.*]], i64 [[SHIFT:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 16 x i8> [[TMP0]]
+//
+vint8m2_t test_vnsra_wx_i8m2(vint16m4_t op1, size_t shift, size_t vl) {
+  return vnsra_wx_i8m2(op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsra_wv_i8m4(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i8> @llvm.riscv.vnsra.nxv32i8.nxv32i16.nxv32i8.i32(<vscale x 32 x i16> [[OP1:%.*]], <vscale x 32 x i8> [[SHIFT:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 32 x i8> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsra_wv_i8m4(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i8> @llvm.riscv.vnsra.nxv32i8.nxv32i16.nxv32i8.i64(<vscale x 32 x i16> [[OP1:%.*]], <vscale x 32 x i8> [[SHIFT:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 32 x i8> [[TMP0]]
+//
+vint8m4_t test_vnsra_wv_i8m4(vint16m8_t op1, vuint8m4_t shift, size_t vl) {
+  return vnsra_wv_i8m4(op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsra_wx_i8m4(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i8> @llvm.riscv.vnsra.nxv32i8.nxv32i16.i32.i32(<vscale x 32 x i16> [[OP1:%.*]], i32 [[SHIFT:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 32 x i8> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsra_wx_i8m4(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i8> @llvm.riscv.vnsra.nxv32i8.nxv32i16.i64.i64(<vscale x 32 x i16> [[OP1:%.*]], i64 [[SHIFT:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 32 x i8> [[TMP0]]
+//
+vint8m4_t test_vnsra_wx_i8m4(vint16m8_t op1, size_t shift, size_t vl) {
+  return vnsra_wx_i8m4(op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsra_wv_i16mf4(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vnsra.nxv1i16.nxv1i32.nxv1i16.i32(<vscale x 1 x i32> [[OP1:%.*]], <vscale x 1 x i16> [[SHIFT:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 1 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsra_wv_i16mf4(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vnsra.nxv1i16.nxv1i32.nxv1i16.i64(<vscale x 1 x i32> [[OP1:%.*]], <vscale x 1 x i16> [[SHIFT:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 1 x i16> [[TMP0]]
+//
+vint16mf4_t test_vnsra_wv_i16mf4(vint32mf2_t op1, vuint16mf4_t shift, size_t vl) {
+  return vnsra_wv_i16mf4(op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsra_wx_i16mf4(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vnsra.nxv1i16.nxv1i32.i32.i32(<vscale x 1 x i32> [[OP1:%.*]], i32 [[SHIFT:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 1 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsra_wx_i16mf4(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vnsra.nxv1i16.nxv1i32.i64.i64(<vscale x 1 x i32> [[OP1:%.*]], i64 [[SHIFT:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 1 x i16> [[TMP0]]
+//
+vint16mf4_t test_vnsra_wx_i16mf4(vint32mf2_t op1, size_t shift, size_t vl) {
+  return vnsra_wx_i16mf4(op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsra_wv_i16mf2(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vnsra.nxv2i16.nxv2i32.nxv2i16.i32(<vscale x 2 x i32> [[OP1:%.*]], <vscale x 2 x i16> [[SHIFT:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 2 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsra_wv_i16mf2(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vnsra.nxv2i16.nxv2i32.nxv2i16.i64(<vscale x 2 x i32> [[OP1:%.*]], <vscale x 2 x i16> [[SHIFT:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 2 x i16> [[TMP0]]
+//
+vint16mf2_t test_vnsra_wv_i16mf2(vint32m1_t op1, vuint16mf2_t shift, size_t vl) {
+  return vnsra_wv_i16mf2(op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsra_wx_i16mf2(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vnsra.nxv2i16.nxv2i32.i32.i32(<vscale x 2 x i32> [[OP1:%.*]], i32 [[SHIFT:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 2 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsra_wx_i16mf2(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vnsra.nxv2i16.nxv2i32.i64.i64(<vscale x 2 x i32> [[OP1:%.*]], i64 [[SHIFT:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 2 x i16> [[TMP0]]
+//
+vint16mf2_t test_vnsra_wx_i16mf2(vint32m1_t op1, size_t shift, size_t vl) {
+  return vnsra_wx_i16mf2(op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsra_wv_i16m1(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vnsra.nxv4i16.nxv4i32.nxv4i16.i32(<vscale x 4 x i32> [[OP1:%.*]], <vscale x 4 x i16> [[SHIFT:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 4 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsra_wv_i16m1(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vnsra.nxv4i16.nxv4i32.nxv4i16.i64(<vscale x 4 x i32> [[OP1:%.*]], <vscale x 4 x i16> [[SHIFT:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 4 x i16> [[TMP0]]
+//
+vint16m1_t test_vnsra_wv_i16m1(vint32m2_t op1, vuint16m1_t shift, size_t vl) {
+  return vnsra_wv_i16m1(op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsra_wx_i16m1(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vnsra.nxv4i16.nxv4i32.i32.i32(<vscale x 4 x i32> [[OP1:%.*]], i32 [[SHIFT:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 4 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsra_wx_i16m1(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vnsra.nxv4i16.nxv4i32.i64.i64(<vscale x 4 x i32> [[OP1:%.*]], i64 [[SHIFT:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 4 x i16> [[TMP0]]
+//
+vint16m1_t test_vnsra_wx_i16m1(vint32m2_t op1, size_t shift, size_t vl) {
+  return vnsra_wx_i16m1(op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsra_wv_i16m2(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vnsra.nxv8i16.nxv8i32.nxv8i16.i32(<vscale x 8 x i32> [[OP1:%.*]], <vscale x 8 x i16> [[SHIFT:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsra_wv_i16m2(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vnsra.nxv8i16.nxv8i32.nxv8i16.i64(<vscale x 8 x i32> [[OP1:%.*]], <vscale x 8 x i16> [[SHIFT:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
+//
+vint16m2_t test_vnsra_wv_i16m2(vint32m4_t op1, vuint16m2_t shift, size_t vl) {
+  return vnsra_wv_i16m2(op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsra_wx_i16m2(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vnsra.nxv8i16.nxv8i32.i32.i32(<vscale x 8 x i32> [[OP1:%.*]], i32 [[SHIFT:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsra_wx_i16m2(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vnsra.nxv8i16.nxv8i32.i64.i64(<vscale x 8 x i32> [[OP1:%.*]], i64 [[SHIFT:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
+//
+vint16m2_t test_vnsra_wx_i16m2(vint32m4_t op1, size_t shift, size_t vl) {
+  return vnsra_wx_i16m2(op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsra_wv_i16m4(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vnsra.nxv16i16.nxv16i32.nxv16i16.i32(<vscale x 16 x i32> [[OP1:%.*]], <vscale x 16 x i16> [[SHIFT:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 16 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsra_wv_i16m4(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vnsra.nxv16i16.nxv16i32.nxv16i16.i64(<vscale x 16 x i32> [[OP1:%.*]], <vscale x 16 x i16> [[SHIFT:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 16 x i16> [[TMP0]]
+//
+vint16m4_t test_vnsra_wv_i16m4(vint32m8_t op1, vuint16m4_t shift, size_t vl) {
+  return vnsra_wv_i16m4(op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsra_wx_i16m4(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vnsra.nxv16i16.nxv16i32.i32.i32(<vscale x 16 x i32> [[OP1:%.*]], i32 [[SHIFT:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 16 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsra_wx_i16m4(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vnsra.nxv16i16.nxv16i32.i64.i64(<vscale x 16 x i32> [[OP1:%.*]], i64 [[SHIFT:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 16 x i16> [[TMP0]]
+//
+vint16m4_t test_vnsra_wx_i16m4(vint32m8_t op1, size_t shift, size_t vl) {
+  return vnsra_wx_i16m4(op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsra_wv_i32mf2(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vnsra.nxv1i32.nxv1i64.nxv1i32.i32(<vscale x 1 x i64> [[OP1:%.*]], <vscale x 1 x i32> [[SHIFT:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsra_wv_i32mf2(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vnsra.nxv1i32.nxv1i64.nxv1i32.i64(<vscale x 1 x i64> [[OP1:%.*]], <vscale x 1 x i32> [[SHIFT:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
+//
+vint32mf2_t test_vnsra_wv_i32mf2(vint64m1_t op1, vuint32mf2_t shift, size_t vl) {
+  return vnsra_wv_i32mf2(op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsra_wx_i32mf2(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vnsra.nxv1i32.nxv1i64.i32.i32(<vscale x 1 x i64> [[OP1:%.*]], i32 [[SHIFT:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsra_wx_i32mf2(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vnsra.nxv1i32.nxv1i64.i64.i64(<vscale x 1 x i64> [[OP1:%.*]], i64 [[SHIFT:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
+//
+vint32mf2_t test_vnsra_wx_i32mf2(vint64m1_t op1, size_t shift, size_t vl) {
+  return vnsra_wx_i32mf2(op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsra_wv_i32m1(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vnsra.nxv2i32.nxv2i64.nxv2i32.i32(<vscale x 2 x i64> [[OP1:%.*]], <vscale x 2 x i32> [[SHIFT:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsra_wv_i32m1(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vnsra.nxv2i32.nxv2i64.nxv2i32.i64(<vscale x 2 x i64> [[OP1:%.*]], <vscale x 2 x i32> [[SHIFT:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
+//
+vint32m1_t test_vnsra_wv_i32m1(vint64m2_t op1, vuint32m1_t shift, size_t vl) {
+  return vnsra_wv_i32m1(op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsra_wx_i32m1(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vnsra.nxv2i32.nxv2i64.i32.i32(<vscale x 2 x i64> [[OP1:%.*]], i32 [[SHIFT:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsra_wx_i32m1(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vnsra.nxv2i32.nxv2i64.i64.i64(<vscale x 2 x i64> [[OP1:%.*]], i64 [[SHIFT:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
+//
+vint32m1_t test_vnsra_wx_i32m1(vint64m2_t op1, size_t shift, size_t vl) {
+  return vnsra_wx_i32m1(op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsra_wv_i32m2(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vnsra.nxv4i32.nxv4i64.nxv4i32.i32(<vscale x 4 x i64> [[OP1:%.*]], <vscale x 4 x i32> [[SHIFT:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsra_wv_i32m2(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vnsra.nxv4i32.nxv4i64.nxv4i32.i64(<vscale x 4 x i64> [[OP1:%.*]], <vscale x 4 x i32> [[SHIFT:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
+//
+vint32m2_t test_vnsra_wv_i32m2(vint64m4_t op1, vuint32m2_t shift, size_t vl) {
+  return vnsra_wv_i32m2(op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsra_wx_i32m2(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vnsra.nxv4i32.nxv4i64.i32.i32(<vscale x 4 x i64> [[OP1:%.*]], i32 [[SHIFT:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsra_wx_i32m2(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vnsra.nxv4i32.nxv4i64.i64.i64(<vscale x 4 x i64> [[OP1:%.*]], i64 [[SHIFT:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
+//
+vint32m2_t test_vnsra_wx_i32m2(vint64m4_t op1, size_t shift, size_t vl) {
+  return vnsra_wx_i32m2(op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsra_wv_i32m4(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vnsra.nxv8i32.nxv8i64.nxv8i32.i32(<vscale x 8 x i64> [[OP1:%.*]], <vscale x 8 x i32> [[SHIFT:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsra_wv_i32m4(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vnsra.nxv8i32.nxv8i64.nxv8i32.i64(<vscale x 8 x i64> [[OP1:%.*]], <vscale x 8 x i32> [[SHIFT:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
+//
+vint32m4_t test_vnsra_wv_i32m4(vint64m8_t op1, vuint32m4_t shift, size_t vl) {
+  return vnsra_wv_i32m4(op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsra_wx_i32m4(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vnsra.nxv8i32.nxv8i64.i32.i32(<vscale x 8 x i64> [[OP1:%.*]], i32 [[SHIFT:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsra_wx_i32m4(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vnsra.nxv8i32.nxv8i64.i64.i64(<vscale x 8 x i64> [[OP1:%.*]], i64 [[SHIFT:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
+//
+vint32m4_t test_vnsra_wx_i32m4(vint64m8_t op1, size_t shift, size_t vl) {
+  return vnsra_wx_i32m4(op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsra_wv_i8mf8_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i8> @llvm.riscv.vnsra.mask.nxv1i8.nxv1i16.nxv1i8.i32(<vscale x 1 x i8> [[MASKEDOFF:%.*]], <vscale x 1 x i16> [[OP1:%.*]], <vscale x 1 x i8> [[SHIFT:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 1 x i8> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsra_wv_i8mf8_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i8> @llvm.riscv.vnsra.mask.nxv1i8.nxv1i16.nxv1i8.i64(<vscale x 1 x i8> [[MASKEDOFF:%.*]], <vscale x 1 x i16> [[OP1:%.*]], <vscale x 1 x i8> [[SHIFT:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 1 x i8> [[TMP0]]
+//
+vint8mf8_t test_vnsra_wv_i8mf8_m(vbool64_t mask, vint8mf8_t maskedoff, vint16mf4_t op1, vuint8mf8_t shift, size_t vl) {
+  return vnsra_wv_i8mf8_m(mask, maskedoff, op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsra_wx_i8mf8_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i8> @llvm.riscv.vnsra.mask.nxv1i8.nxv1i16.i32.i32(<vscale x 1 x i8> [[MASKEDOFF:%.*]], <vscale x 1 x i16> [[OP1:%.*]], i32 [[SHIFT:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 1 x i8> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsra_wx_i8mf8_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i8> @llvm.riscv.vnsra.mask.nxv1i8.nxv1i16.i64.i64(<vscale x 1 x i8> [[MASKEDOFF:%.*]], <vscale x 1 x i16> [[OP1:%.*]], i64 [[SHIFT:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 1 x i8> [[TMP0]]
+//
+vint8mf8_t test_vnsra_wx_i8mf8_m(vbool64_t mask, vint8mf8_t maskedoff, vint16mf4_t op1, size_t shift, size_t vl) {
+  return vnsra_wx_i8mf8_m(mask, maskedoff, op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsra_wv_i8mf4_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i8> @llvm.riscv.vnsra.mask.nxv2i8.nxv2i16.nxv2i8.i32(<vscale x 2 x i8> [[MASKEDOFF:%.*]], <vscale x 2 x i16> [[OP1:%.*]], <vscale x 2 x i8> [[SHIFT:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 2 x i8> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsra_wv_i8mf4_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i8> @llvm.riscv.vnsra.mask.nxv2i8.nxv2i16.nxv2i8.i64(<vscale x 2 x i8> [[MASKEDOFF:%.*]], <vscale x 2 x i16> [[OP1:%.*]], <vscale x 2 x i8> [[SHIFT:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 2 x i8> [[TMP0]]
+//
+vint8mf4_t test_vnsra_wv_i8mf4_m(vbool32_t mask, vint8mf4_t maskedoff, vint16mf2_t op1, vuint8mf4_t shift, size_t vl) {
+  return vnsra_wv_i8mf4_m(mask, maskedoff, op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsra_wx_i8mf4_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i8> @llvm.riscv.vnsra.mask.nxv2i8.nxv2i16.i32.i32(<vscale x 2 x i8> [[MASKEDOFF:%.*]], <vscale x 2 x i16> [[OP1:%.*]], i32 [[SHIFT:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 2 x i8> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsra_wx_i8mf4_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i8> @llvm.riscv.vnsra.mask.nxv2i8.nxv2i16.i64.i64(<vscale x 2 x i8> [[MASKEDOFF:%.*]], <vscale x 2 x i16> [[OP1:%.*]], i64 [[SHIFT:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 2 x i8> [[TMP0]]
+//
+vint8mf4_t test_vnsra_wx_i8mf4_m(vbool32_t mask, vint8mf4_t maskedoff, vint16mf2_t op1, size_t shift, size_t vl) {
+  return vnsra_wx_i8mf4_m(mask, maskedoff, op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsra_wv_i8mf2_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i8> @llvm.riscv.vnsra.mask.nxv4i8.nxv4i16.nxv4i8.i32(<vscale x 4 x i8> [[MASKEDOFF:%.*]], <vscale x 4 x i16> [[OP1:%.*]], <vscale x 4 x i8> [[SHIFT:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 4 x i8> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsra_wv_i8mf2_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i8> @llvm.riscv.vnsra.mask.nxv4i8.nxv4i16.nxv4i8.i64(<vscale x 4 x i8> [[MASKEDOFF:%.*]], <vscale x 4 x i16> [[OP1:%.*]], <vscale x 4 x i8> [[SHIFT:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 4 x i8> [[TMP0]]
+//
+vint8mf2_t test_vnsra_wv_i8mf2_m(vbool16_t mask, vint8mf2_t maskedoff, vint16m1_t op1, vuint8mf2_t shift, size_t vl) {
+  return vnsra_wv_i8mf2_m(mask, maskedoff, op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsra_wx_i8mf2_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i8> @llvm.riscv.vnsra.mask.nxv4i8.nxv4i16.i32.i32(<vscale x 4 x i8> [[MASKEDOFF:%.*]], <vscale x 4 x i16> [[OP1:%.*]], i32 [[SHIFT:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 4 x i8> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsra_wx_i8mf2_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i8> @llvm.riscv.vnsra.mask.nxv4i8.nxv4i16.i64.i64(<vscale x 4 x i8> [[MASKEDOFF:%.*]], <vscale x 4 x i16> [[OP1:%.*]], i64 [[SHIFT:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 4 x i8> [[TMP0]]
+//
+vint8mf2_t test_vnsra_wx_i8mf2_m(vbool16_t mask, vint8mf2_t maskedoff, vint16m1_t op1, size_t shift, size_t vl) {
+  return vnsra_wx_i8mf2_m(mask, maskedoff, op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsra_wv_i8m1_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i8> @llvm.riscv.vnsra.mask.nxv8i8.nxv8i16.nxv8i8.i32(<vscale x 8 x i8> [[MASKEDOFF:%.*]], <vscale x 8 x i16> [[OP1:%.*]], <vscale x 8 x i8> [[SHIFT:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 8 x i8> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsra_wv_i8m1_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i8> @llvm.riscv.vnsra.mask.nxv8i8.nxv8i16.nxv8i8.i64(<vscale x 8 x i8> [[MASKEDOFF:%.*]], <vscale x 8 x i16> [[OP1:%.*]], <vscale x 8 x i8> [[SHIFT:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 8 x i8> [[TMP0]]
+//
+vint8m1_t test_vnsra_wv_i8m1_m(vbool8_t mask, vint8m1_t maskedoff, vint16m2_t op1, vuint8m1_t shift, size_t vl) {
+  return vnsra_wv_i8m1_m(mask, maskedoff, op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsra_wx_i8m1_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i8> @llvm.riscv.vnsra.mask.nxv8i8.nxv8i16.i32.i32(<vscale x 8 x i8> [[MASKEDOFF:%.*]], <vscale x 8 x i16> [[OP1:%.*]], i32 [[SHIFT:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 8 x i8> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsra_wx_i8m1_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i8> @llvm.riscv.vnsra.mask.nxv8i8.nxv8i16.i64.i64(<vscale x 8 x i8> [[MASKEDOFF:%.*]], <vscale x 8 x i16> [[OP1:%.*]], i64 [[SHIFT:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 8 x i8> [[TMP0]]
+//
+vint8m1_t test_vnsra_wx_i8m1_m(vbool8_t mask, vint8m1_t maskedoff, vint16m2_t op1, size_t shift, size_t vl) {
+  return vnsra_wx_i8m1_m(mask, maskedoff, op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsra_wv_i8m2_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i8> @llvm.riscv.vnsra.mask.nxv16i8.nxv16i16.nxv16i8.i32(<vscale x 16 x i8> [[MASKEDOFF:%.*]], <vscale x 16 x i16> [[OP1:%.*]], <vscale x 16 x i8> [[SHIFT:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 16 x i8> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsra_wv_i8m2_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i8> @llvm.riscv.vnsra.mask.nxv16i8.nxv16i16.nxv16i8.i64(<vscale x 16 x i8> [[MASKEDOFF:%.*]], <vscale x 16 x i16> [[OP1:%.*]], <vscale x 16 x i8> [[SHIFT:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 16 x i8> [[TMP0]]
+//
+vint8m2_t test_vnsra_wv_i8m2_m(vbool4_t mask, vint8m2_t maskedoff, vint16m4_t op1, vuint8m2_t shift, size_t vl) {
+  return vnsra_wv_i8m2_m(mask, maskedoff, op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsra_wx_i8m2_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i8> @llvm.riscv.vnsra.mask.nxv16i8.nxv16i16.i32.i32(<vscale x 16 x i8> [[MASKEDOFF:%.*]], <vscale x 16 x i16> [[OP1:%.*]], i32 [[SHIFT:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 16 x i8> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsra_wx_i8m2_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i8> @llvm.riscv.vnsra.mask.nxv16i8.nxv16i16.i64.i64(<vscale x 16 x i8> [[MASKEDOFF:%.*]], <vscale x 16 x i16> [[OP1:%.*]], i64 [[SHIFT:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 16 x i8> [[TMP0]]
+//
+vint8m2_t test_vnsra_wx_i8m2_m(vbool4_t mask, vint8m2_t maskedoff, vint16m4_t op1, size_t shift, size_t vl) {
+  return vnsra_wx_i8m2_m(mask, maskedoff, op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsra_wv_i8m4_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i8> @llvm.riscv.vnsra.mask.nxv32i8.nxv32i16.nxv32i8.i32(<vscale x 32 x i8> [[MASKEDOFF:%.*]], <vscale x 32 x i16> [[OP1:%.*]], <vscale x 32 x i8> [[SHIFT:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 32 x i8> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsra_wv_i8m4_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i8> @llvm.riscv.vnsra.mask.nxv32i8.nxv32i16.nxv32i8.i64(<vscale x 32 x i8> [[MASKEDOFF:%.*]], <vscale x 32 x i16> [[OP1:%.*]], <vscale x 32 x i8> [[SHIFT:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 32 x i8> [[TMP0]]
+//
+vint8m4_t test_vnsra_wv_i8m4_m(vbool2_t mask, vint8m4_t maskedoff, vint16m8_t op1, vuint8m4_t shift, size_t vl) {
+  return vnsra_wv_i8m4_m(mask, maskedoff, op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsra_wx_i8m4_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i8> @llvm.riscv.vnsra.mask.nxv32i8.nxv32i16.i32.i32(<vscale x 32 x i8> [[MASKEDOFF:%.*]], <vscale x 32 x i16> [[OP1:%.*]], i32 [[SHIFT:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 32 x i8> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsra_wx_i8m4_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i8> @llvm.riscv.vnsra.mask.nxv32i8.nxv32i16.i64.i64(<vscale x 32 x i8> [[MASKEDOFF:%.*]], <vscale x 32 x i16> [[OP1:%.*]], i64 [[SHIFT:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 32 x i8> [[TMP0]]
+//
+vint8m4_t test_vnsra_wx_i8m4_m(vbool2_t mask, vint8m4_t maskedoff, vint16m8_t op1, size_t shift, size_t vl) {
+  return vnsra_wx_i8m4_m(mask, maskedoff, op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsra_wv_i16mf4_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vnsra.mask.nxv1i16.nxv1i32.nxv1i16.i32(<vscale x 1 x i16> [[MASKEDOFF:%.*]], <vscale x 1 x i32> [[OP1:%.*]], <vscale x 1 x i16> [[SHIFT:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 1 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsra_wv_i16mf4_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vnsra.mask.nxv1i16.nxv1i32.nxv1i16.i64(<vscale x 1 x i16> [[MASKEDOFF:%.*]], <vscale x 1 x i32> [[OP1:%.*]], <vscale x 1 x i16> [[SHIFT:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 1 x i16> [[TMP0]]
+//
+vint16mf4_t test_vnsra_wv_i16mf4_m(vbool64_t mask, vint16mf4_t maskedoff, vint32mf2_t op1, vuint16mf4_t shift, size_t vl) {
+  return vnsra_wv_i16mf4_m(mask, maskedoff, op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsra_wx_i16mf4_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vnsra.mask.nxv1i16.nxv1i32.i32.i32(<vscale x 1 x i16> [[MASKEDOFF:%.*]], <vscale x 1 x i32> [[OP1:%.*]], i32 [[SHIFT:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 1 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsra_wx_i16mf4_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vnsra.mask.nxv1i16.nxv1i32.i64.i64(<vscale x 1 x i16> [[MASKEDOFF:%.*]], <vscale x 1 x i32> [[OP1:%.*]], i64 [[SHIFT:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 1 x i16> [[TMP0]]
+//
+vint16mf4_t test_vnsra_wx_i16mf4_m(vbool64_t mask, vint16mf4_t maskedoff, vint32mf2_t op1, size_t shift, size_t vl) {
+  return vnsra_wx_i16mf4_m(mask, maskedoff, op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsra_wv_i16mf2_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vnsra.mask.nxv2i16.nxv2i32.nxv2i16.i32(<vscale x 2 x i16> [[MASKEDOFF:%.*]], <vscale x 2 x i32> [[OP1:%.*]], <vscale x 2 x i16> [[SHIFT:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 2 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsra_wv_i16mf2_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vnsra.mask.nxv2i16.nxv2i32.nxv2i16.i64(<vscale x 2 x i16> [[MASKEDOFF:%.*]], <vscale x 2 x i32> [[OP1:%.*]], <vscale x 2 x i16> [[SHIFT:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 2 x i16> [[TMP0]]
+//
+vint16mf2_t test_vnsra_wv_i16mf2_m(vbool32_t mask, vint16mf2_t maskedoff, vint32m1_t op1, vuint16mf2_t shift, size_t vl) {
+  return vnsra_wv_i16mf2_m(mask, maskedoff, op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsra_wx_i16mf2_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vnsra.mask.nxv2i16.nxv2i32.i32.i32(<vscale x 2 x i16> [[MASKEDOFF:%.*]], <vscale x 2 x i32> [[OP1:%.*]], i32 [[SHIFT:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 2 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsra_wx_i16mf2_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vnsra.mask.nxv2i16.nxv2i32.i64.i64(<vscale x 2 x i16> [[MASKEDOFF:%.*]], <vscale x 2 x i32> [[OP1:%.*]], i64 [[SHIFT:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 2 x i16> [[TMP0]]
+//
+vint16mf2_t test_vnsra_wx_i16mf2_m(vbool32_t mask, vint16mf2_t maskedoff, vint32m1_t op1, size_t shift, size_t vl) {
+  return vnsra_wx_i16mf2_m(mask, maskedoff, op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsra_wv_i16m1_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vnsra.mask.nxv4i16.nxv4i32.nxv4i16.i32(<vscale x 4 x i16> [[MASKEDOFF:%.*]], <vscale x 4 x i32> [[OP1:%.*]], <vscale x 4 x i16> [[SHIFT:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 4 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsra_wv_i16m1_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vnsra.mask.nxv4i16.nxv4i32.nxv4i16.i64(<vscale x 4 x i16> [[MASKEDOFF:%.*]], <vscale x 4 x i32> [[OP1:%.*]], <vscale x 4 x i16> [[SHIFT:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 4 x i16> [[TMP0]]
+//
+vint16m1_t test_vnsra_wv_i16m1_m(vbool16_t mask, vint16m1_t maskedoff, vint32m2_t op1, vuint16m1_t shift, size_t vl) {
+  return vnsra_wv_i16m1_m(mask, maskedoff, op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsra_wx_i16m1_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vnsra.mask.nxv4i16.nxv4i32.i32.i32(<vscale x 4 x i16> [[MASKEDOFF:%.*]], <vscale x 4 x i32> [[OP1:%.*]], i32 [[SHIFT:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 4 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsra_wx_i16m1_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vnsra.mask.nxv4i16.nxv4i32.i64.i64(<vscale x 4 x i16> [[MASKEDOFF:%.*]], <vscale x 4 x i32> [[OP1:%.*]], i64 [[SHIFT:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 4 x i16> [[TMP0]]
+//
+vint16m1_t test_vnsra_wx_i16m1_m(vbool16_t mask, vint16m1_t maskedoff, vint32m2_t op1, size_t shift, size_t vl) {
+  return vnsra_wx_i16m1_m(mask, maskedoff, op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsra_wv_i16m2_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vnsra.mask.nxv8i16.nxv8i32.nxv8i16.i32(<vscale x 8 x i16> [[MASKEDOFF:%.*]], <vscale x 8 x i32> [[OP1:%.*]], <vscale x 8 x i16> [[SHIFT:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsra_wv_i16m2_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vnsra.mask.nxv8i16.nxv8i32.nxv8i16.i64(<vscale x 8 x i16> [[MASKEDOFF:%.*]], <vscale x 8 x i32> [[OP1:%.*]], <vscale x 8 x i16> [[SHIFT:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
+//
+vint16m2_t test_vnsra_wv_i16m2_m(vbool8_t mask, vint16m2_t maskedoff, vint32m4_t op1, vuint16m2_t shift, size_t vl) {
+  return vnsra_wv_i16m2_m(mask, maskedoff, op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsra_wx_i16m2_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vnsra.mask.nxv8i16.nxv8i32.i32.i32(<vscale x 8 x i16> [[MASKEDOFF:%.*]], <vscale x 8 x i32> [[OP1:%.*]], i32 [[SHIFT:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsra_wx_i16m2_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vnsra.mask.nxv8i16.nxv8i32.i64.i64(<vscale x 8 x i16> [[MASKEDOFF:%.*]], <vscale x 8 x i32> [[OP1:%.*]], i64 [[SHIFT:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
+//
+vint16m2_t test_vnsra_wx_i16m2_m(vbool8_t mask, vint16m2_t maskedoff, vint32m4_t op1, size_t shift, size_t vl) {
+  return vnsra_wx_i16m2_m(mask, maskedoff, op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsra_wv_i16m4_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vnsra.mask.nxv16i16.nxv16i32.nxv16i16.i32(<vscale x 16 x i16> [[MASKEDOFF:%.*]], <vscale x 16 x i32> [[OP1:%.*]], <vscale x 16 x i16> [[SHIFT:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 16 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsra_wv_i16m4_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vnsra.mask.nxv16i16.nxv16i32.nxv16i16.i64(<vscale x 16 x i16> [[MASKEDOFF:%.*]], <vscale x 16 x i32> [[OP1:%.*]], <vscale x 16 x i16> [[SHIFT:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 16 x i16> [[TMP0]]
+//
+vint16m4_t test_vnsra_wv_i16m4_m(vbool4_t mask, vint16m4_t maskedoff, vint32m8_t op1, vuint16m4_t shift, size_t vl) {
+  return vnsra_wv_i16m4_m(mask, maskedoff, op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsra_wx_i16m4_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vnsra.mask.nxv16i16.nxv16i32.i32.i32(<vscale x 16 x i16> [[MASKEDOFF:%.*]], <vscale x 16 x i32> [[OP1:%.*]], i32 [[SHIFT:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 16 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsra_wx_i16m4_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vnsra.mask.nxv16i16.nxv16i32.i64.i64(<vscale x 16 x i16> [[MASKEDOFF:%.*]], <vscale x 16 x i32> [[OP1:%.*]], i64 [[SHIFT:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 16 x i16> [[TMP0]]
+//
+vint16m4_t test_vnsra_wx_i16m4_m(vbool4_t mask, vint16m4_t maskedoff, vint32m8_t op1, size_t shift, size_t vl) {
+  return vnsra_wx_i16m4_m(mask, maskedoff, op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsra_wv_i32mf2_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vnsra.mask.nxv1i32.nxv1i64.nxv1i32.i32(<vscale x 1 x i32> [[MASKEDOFF:%.*]], <vscale x 1 x i64> [[OP1:%.*]], <vscale x 1 x i32> [[SHIFT:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsra_wv_i32mf2_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vnsra.mask.nxv1i32.nxv1i64.nxv1i32.i64(<vscale x 1 x i32> [[MASKEDOFF:%.*]], <vscale x 1 x i64> [[OP1:%.*]], <vscale x 1 x i32> [[SHIFT:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
+//
+vint32mf2_t test_vnsra_wv_i32mf2_m(vbool64_t mask, vint32mf2_t maskedoff, vint64m1_t op1, vuint32mf2_t shift, size_t vl) {
+  return vnsra_wv_i32mf2_m(mask, maskedoff, op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsra_wx_i32mf2_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vnsra.mask.nxv1i32.nxv1i64.i32.i32(<vscale x 1 x i32> [[MASKEDOFF:%.*]], <vscale x 1 x i64> [[OP1:%.*]], i32 [[SHIFT:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsra_wx_i32mf2_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vnsra.mask.nxv1i32.nxv1i64.i64.i64(<vscale x 1 x i32> [[MASKEDOFF:%.*]], <vscale x 1 x i64> [[OP1:%.*]], i64 [[SHIFT:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
+//
+vint32mf2_t test_vnsra_wx_i32mf2_m(vbool64_t mask, vint32mf2_t maskedoff, vint64m1_t op1, size_t shift, size_t vl) {
+  return vnsra_wx_i32mf2_m(mask, maskedoff, op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsra_wv_i32m1_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vnsra.mask.nxv2i32.nxv2i64.nxv2i32.i32(<vscale x 2 x i32> [[MASKEDOFF:%.*]], <vscale x 2 x i64> [[OP1:%.*]], <vscale x 2 x i32> [[SHIFT:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsra_wv_i32m1_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vnsra.mask.nxv2i32.nxv2i64.nxv2i32.i64(<vscale x 2 x i32> [[MASKEDOFF:%.*]], <vscale x 2 x i64> [[OP1:%.*]], <vscale x 2 x i32> [[SHIFT:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
+//
+vint32m1_t test_vnsra_wv_i32m1_m(vbool32_t mask, vint32m1_t maskedoff, vint64m2_t op1, vuint32m1_t shift, size_t vl) {
+  return vnsra_wv_i32m1_m(mask, maskedoff, op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsra_wx_i32m1_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vnsra.mask.nxv2i32.nxv2i64.i32.i32(<vscale x 2 x i32> [[MASKEDOFF:%.*]], <vscale x 2 x i64> [[OP1:%.*]], i32 [[SHIFT:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsra_wx_i32m1_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vnsra.mask.nxv2i32.nxv2i64.i64.i64(<vscale x 2 x i32> [[MASKEDOFF:%.*]], <vscale x 2 x i64> [[OP1:%.*]], i64 [[SHIFT:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
+//
+vint32m1_t test_vnsra_wx_i32m1_m(vbool32_t mask, vint32m1_t maskedoff, vint64m2_t op1, size_t shift, size_t vl) {
+  return vnsra_wx_i32m1_m(mask, maskedoff, op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsra_wv_i32m2_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vnsra.mask.nxv4i32.nxv4i64.nxv4i32.i32(<vscale x 4 x i32> [[MASKEDOFF:%.*]], <vscale x 4 x i64> [[OP1:%.*]], <vscale x 4 x i32> [[SHIFT:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsra_wv_i32m2_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vnsra.mask.nxv4i32.nxv4i64.nxv4i32.i64(<vscale x 4 x i32> [[MASKEDOFF:%.*]], <vscale x 4 x i64> [[OP1:%.*]], <vscale x 4 x i32> [[SHIFT:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
+//
+vint32m2_t test_vnsra_wv_i32m2_m(vbool16_t mask, vint32m2_t maskedoff, vint64m4_t op1, vuint32m2_t shift, size_t vl) {
+  return vnsra_wv_i32m2_m(mask, maskedoff, op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsra_wx_i32m2_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vnsra.mask.nxv4i32.nxv4i64.i32.i32(<vscale x 4 x i32> [[MASKEDOFF:%.*]], <vscale x 4 x i64> [[OP1:%.*]], i32 [[SHIFT:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsra_wx_i32m2_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vnsra.mask.nxv4i32.nxv4i64.i64.i64(<vscale x 4 x i32> [[MASKEDOFF:%.*]], <vscale x 4 x i64> [[OP1:%.*]], i64 [[SHIFT:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
+//
+vint32m2_t test_vnsra_wx_i32m2_m(vbool16_t mask, vint32m2_t maskedoff, vint64m4_t op1, size_t shift, size_t vl) {
+  return vnsra_wx_i32m2_m(mask, maskedoff, op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsra_wv_i32m4_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vnsra.mask.nxv8i32.nxv8i64.nxv8i32.i32(<vscale x 8 x i32> [[MASKEDOFF:%.*]], <vscale x 8 x i64> [[OP1:%.*]], <vscale x 8 x i32> [[SHIFT:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsra_wv_i32m4_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vnsra.mask.nxv8i32.nxv8i64.nxv8i32.i64(<vscale x 8 x i32> [[MASKEDOFF:%.*]], <vscale x 8 x i64> [[OP1:%.*]], <vscale x 8 x i32> [[SHIFT:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
+//
+vint32m4_t test_vnsra_wv_i32m4_m(vbool8_t mask, vint32m4_t maskedoff, vint64m8_t op1, vuint32m4_t shift, size_t vl) {
+  return vnsra_wv_i32m4_m(mask, maskedoff, op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsra_wx_i32m4_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vnsra.mask.nxv8i32.nxv8i64.i32.i32(<vscale x 8 x i32> [[MASKEDOFF:%.*]], <vscale x 8 x i64> [[OP1:%.*]], i32 [[SHIFT:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsra_wx_i32m4_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vnsra.mask.nxv8i32.nxv8i64.i64.i64(<vscale x 8 x i32> [[MASKEDOFF:%.*]], <vscale x 8 x i64> [[OP1:%.*]], i64 [[SHIFT:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
+//
+vint32m4_t test_vnsra_wx_i32m4_m(vbool8_t mask, vint32m4_t maskedoff, vint64m8_t op1, size_t shift, size_t vl) {
+  return vnsra_wx_i32m4_m(mask, maskedoff, op1, shift, vl);
+}

clang/test/CodeGen/RISCV/rvv-intrinsics/vnsrl.c

+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv32 -target-feature +experimental-v -disable-O0-optnone -emit-llvm %s -o - | opt -S -mem2reg | FileCheck --check-prefix=CHECK-RV32 %s
+// RUN: %clang_cc1 -triple riscv64 -target-feature +experimental-v -disable-O0-optnone -emit-llvm %s -o - | opt -S -mem2reg | FileCheck --check-prefix=CHECK-RV64 %s
+// RUN: %clang_cc1 -triple riscv64 -target-feature +m -target-feature +experimental-v -Werror -Wall -o - %s -S >/dev/null 2>&1 | FileCheck --check-prefix=ASM --allow-empty %s
+
+// ASM-NOT: warning
+#include <riscv_vector.h>
+
+// CHECK-RV32-LABEL: @test_vnsrl_wv_u8mf8(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i8> @llvm.riscv.vnsrl.nxv1i8.nxv1i16.nxv1i8.i32(<vscale x 1 x i16> [[OP1:%.*]], <vscale x 1 x i8> [[SHIFT:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 1 x i8> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsrl_wv_u8mf8(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i8> @llvm.riscv.vnsrl.nxv1i8.nxv1i16.nxv1i8.i64(<vscale x 1 x i16> [[OP1:%.*]], <vscale x 1 x i8> [[SHIFT:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 1 x i8> [[TMP0]]
+//
+vuint8mf8_t test_vnsrl_wv_u8mf8(vuint16mf4_t op1, vuint8mf8_t shift, size_t vl) {
+  return vnsrl_wv_u8mf8(op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsrl_wx_u8mf8(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i8> @llvm.riscv.vnsrl.nxv1i8.nxv1i16.i32.i32(<vscale x 1 x i16> [[OP1:%.*]], i32 [[SHIFT:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 1 x i8> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsrl_wx_u8mf8(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i8> @llvm.riscv.vnsrl.nxv1i8.nxv1i16.i64.i64(<vscale x 1 x i16> [[OP1:%.*]], i64 [[SHIFT:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 1 x i8> [[TMP0]]
+//
+vuint8mf8_t test_vnsrl_wx_u8mf8(vuint16mf4_t op1, size_t shift, size_t vl) {
+  return vnsrl_wx_u8mf8(op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsrl_wv_u8mf4(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i8> @llvm.riscv.vnsrl.nxv2i8.nxv2i16.nxv2i8.i32(<vscale x 2 x i16> [[OP1:%.*]], <vscale x 2 x i8> [[SHIFT:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 2 x i8> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsrl_wv_u8mf4(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i8> @llvm.riscv.vnsrl.nxv2i8.nxv2i16.nxv2i8.i64(<vscale x 2 x i16> [[OP1:%.*]], <vscale x 2 x i8> [[SHIFT:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 2 x i8> [[TMP0]]
+//
+vuint8mf4_t test_vnsrl_wv_u8mf4(vuint16mf2_t op1, vuint8mf4_t shift, size_t vl) {
+  return vnsrl_wv_u8mf4(op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsrl_wx_u8mf4(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i8> @llvm.riscv.vnsrl.nxv2i8.nxv2i16.i32.i32(<vscale x 2 x i16> [[OP1:%.*]], i32 [[SHIFT:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 2 x i8> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsrl_wx_u8mf4(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i8> @llvm.riscv.vnsrl.nxv2i8.nxv2i16.i64.i64(<vscale x 2 x i16> [[OP1:%.*]], i64 [[SHIFT:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 2 x i8> [[TMP0]]
+//
+vuint8mf4_t test_vnsrl_wx_u8mf4(vuint16mf2_t op1, size_t shift, size_t vl) {
+  return vnsrl_wx_u8mf4(op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsrl_wv_u8mf2(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i8> @llvm.riscv.vnsrl.nxv4i8.nxv4i16.nxv4i8.i32(<vscale x 4 x i16> [[OP1:%.*]], <vscale x 4 x i8> [[SHIFT:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 4 x i8> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsrl_wv_u8mf2(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i8> @llvm.riscv.vnsrl.nxv4i8.nxv4i16.nxv4i8.i64(<vscale x 4 x i16> [[OP1:%.*]], <vscale x 4 x i8> [[SHIFT:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 4 x i8> [[TMP0]]
+//
+vuint8mf2_t test_vnsrl_wv_u8mf2(vuint16m1_t op1, vuint8mf2_t shift, size_t vl) {
+  return vnsrl_wv_u8mf2(op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsrl_wx_u8mf2(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i8> @llvm.riscv.vnsrl.nxv4i8.nxv4i16.i32.i32(<vscale x 4 x i16> [[OP1:%.*]], i32 [[SHIFT:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 4 x i8> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsrl_wx_u8mf2(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i8> @llvm.riscv.vnsrl.nxv4i8.nxv4i16.i64.i64(<vscale x 4 x i16> [[OP1:%.*]], i64 [[SHIFT:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 4 x i8> [[TMP0]]
+//
+vuint8mf2_t test_vnsrl_wx_u8mf2(vuint16m1_t op1, size_t shift, size_t vl) {
+  return vnsrl_wx_u8mf2(op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsrl_wv_u8m1(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i8> @llvm.riscv.vnsrl.nxv8i8.nxv8i16.nxv8i8.i32(<vscale x 8 x i16> [[OP1:%.*]], <vscale x 8 x i8> [[SHIFT:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 8 x i8> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsrl_wv_u8m1(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i8> @llvm.riscv.vnsrl.nxv8i8.nxv8i16.nxv8i8.i64(<vscale x 8 x i16> [[OP1:%.*]], <vscale x 8 x i8> [[SHIFT:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 8 x i8> [[TMP0]]
+//
+vuint8m1_t test_vnsrl_wv_u8m1(vuint16m2_t op1, vuint8m1_t shift, size_t vl) {
+  return vnsrl_wv_u8m1(op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsrl_wx_u8m1(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i8> @llvm.riscv.vnsrl.nxv8i8.nxv8i16.i32.i32(<vscale x 8 x i16> [[OP1:%.*]], i32 [[SHIFT:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 8 x i8> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsrl_wx_u8m1(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i8> @llvm.riscv.vnsrl.nxv8i8.nxv8i16.i64.i64(<vscale x 8 x i16> [[OP1:%.*]], i64 [[SHIFT:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 8 x i8> [[TMP0]]
+//
+vuint8m1_t test_vnsrl_wx_u8m1(vuint16m2_t op1, size_t shift, size_t vl) {
+  return vnsrl_wx_u8m1(op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsrl_wv_u8m2(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i8> @llvm.riscv.vnsrl.nxv16i8.nxv16i16.nxv16i8.i32(<vscale x 16 x i16> [[OP1:%.*]], <vscale x 16 x i8> [[SHIFT:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 16 x i8> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsrl_wv_u8m2(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i8> @llvm.riscv.vnsrl.nxv16i8.nxv16i16.nxv16i8.i64(<vscale x 16 x i16> [[OP1:%.*]], <vscale x 16 x i8> [[SHIFT:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 16 x i8> [[TMP0]]
+//
+vuint8m2_t test_vnsrl_wv_u8m2(vuint16m4_t op1, vuint8m2_t shift, size_t vl) {
+  return vnsrl_wv_u8m2(op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsrl_wx_u8m2(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i8> @llvm.riscv.vnsrl.nxv16i8.nxv16i16.i32.i32(<vscale x 16 x i16> [[OP1:%.*]], i32 [[SHIFT:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 16 x i8> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsrl_wx_u8m2(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i8> @llvm.riscv.vnsrl.nxv16i8.nxv16i16.i64.i64(<vscale x 16 x i16> [[OP1:%.*]], i64 [[SHIFT:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 16 x i8> [[TMP0]]
+//
+vuint8m2_t test_vnsrl_wx_u8m2(vuint16m4_t op1, size_t shift, size_t vl) {
+  return vnsrl_wx_u8m2(op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsrl_wv_u8m4(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i8> @llvm.riscv.vnsrl.nxv32i8.nxv32i16.nxv32i8.i32(<vscale x 32 x i16> [[OP1:%.*]], <vscale x 32 x i8> [[SHIFT:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 32 x i8> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsrl_wv_u8m4(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i8> @llvm.riscv.vnsrl.nxv32i8.nxv32i16.nxv32i8.i64(<vscale x 32 x i16> [[OP1:%.*]], <vscale x 32 x i8> [[SHIFT:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 32 x i8> [[TMP0]]
+//
+vuint8m4_t test_vnsrl_wv_u8m4(vuint16m8_t op1, vuint8m4_t shift, size_t vl) {
+  return vnsrl_wv_u8m4(op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsrl_wx_u8m4(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i8> @llvm.riscv.vnsrl.nxv32i8.nxv32i16.i32.i32(<vscale x 32 x i16> [[OP1:%.*]], i32 [[SHIFT:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 32 x i8> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsrl_wx_u8m4(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i8> @llvm.riscv.vnsrl.nxv32i8.nxv32i16.i64.i64(<vscale x 32 x i16> [[OP1:%.*]], i64 [[SHIFT:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 32 x i8> [[TMP0]]
+//
+vuint8m4_t test_vnsrl_wx_u8m4(vuint16m8_t op1, size_t shift, size_t vl) {
+  return vnsrl_wx_u8m4(op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsrl_wv_u16mf4(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vnsrl.nxv1i16.nxv1i32.nxv1i16.i32(<vscale x 1 x i32> [[OP1:%.*]], <vscale x 1 x i16> [[SHIFT:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 1 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsrl_wv_u16mf4(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vnsrl.nxv1i16.nxv1i32.nxv1i16.i64(<vscale x 1 x i32> [[OP1:%.*]], <vscale x 1 x i16> [[SHIFT:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 1 x i16> [[TMP0]]
+//
+vuint16mf4_t test_vnsrl_wv_u16mf4(vuint32mf2_t op1, vuint16mf4_t shift, size_t vl) {
+  return vnsrl_wv_u16mf4(op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsrl_wx_u16mf4(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vnsrl.nxv1i16.nxv1i32.i32.i32(<vscale x 1 x i32> [[OP1:%.*]], i32 [[SHIFT:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 1 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsrl_wx_u16mf4(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vnsrl.nxv1i16.nxv1i32.i64.i64(<vscale x 1 x i32> [[OP1:%.*]], i64 [[SHIFT:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 1 x i16> [[TMP0]]
+//
+vuint16mf4_t test_vnsrl_wx_u16mf4(vuint32mf2_t op1, size_t shift, size_t vl) {
+  return vnsrl_wx_u16mf4(op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsrl_wv_u16mf2(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vnsrl.nxv2i16.nxv2i32.nxv2i16.i32(<vscale x 2 x i32> [[OP1:%.*]], <vscale x 2 x i16> [[SHIFT:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 2 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsrl_wv_u16mf2(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vnsrl.nxv2i16.nxv2i32.nxv2i16.i64(<vscale x 2 x i32> [[OP1:%.*]], <vscale x 2 x i16> [[SHIFT:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 2 x i16> [[TMP0]]
+//
+vuint16mf2_t test_vnsrl_wv_u16mf2(vuint32m1_t op1, vuint16mf2_t shift, size_t vl) {
+  return vnsrl_wv_u16mf2(op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsrl_wx_u16mf2(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vnsrl.nxv2i16.nxv2i32.i32.i32(<vscale x 2 x i32> [[OP1:%.*]], i32 [[SHIFT:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 2 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsrl_wx_u16mf2(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vnsrl.nxv2i16.nxv2i32.i64.i64(<vscale x 2 x i32> [[OP1:%.*]], i64 [[SHIFT:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 2 x i16> [[TMP0]]
+//
+vuint16mf2_t test_vnsrl_wx_u16mf2(vuint32m1_t op1, size_t shift, size_t vl) {
+  return vnsrl_wx_u16mf2(op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsrl_wv_u16m1(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vnsrl.nxv4i16.nxv4i32.nxv4i16.i32(<vscale x 4 x i32> [[OP1:%.*]], <vscale x 4 x i16> [[SHIFT:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 4 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsrl_wv_u16m1(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vnsrl.nxv4i16.nxv4i32.nxv4i16.i64(<vscale x 4 x i32> [[OP1:%.*]], <vscale x 4 x i16> [[SHIFT:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 4 x i16> [[TMP0]]
+//
+vuint16m1_t test_vnsrl_wv_u16m1(vuint32m2_t op1, vuint16m1_t shift, size_t vl) {
+  return vnsrl_wv_u16m1(op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsrl_wx_u16m1(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vnsrl.nxv4i16.nxv4i32.i32.i32(<vscale x 4 x i32> [[OP1:%.*]], i32 [[SHIFT:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 4 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsrl_wx_u16m1(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vnsrl.nxv4i16.nxv4i32.i64.i64(<vscale x 4 x i32> [[OP1:%.*]], i64 [[SHIFT:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 4 x i16> [[TMP0]]
+//
+vuint16m1_t test_vnsrl_wx_u16m1(vuint32m2_t op1, size_t shift, size_t vl) {
+  return vnsrl_wx_u16m1(op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsrl_wv_u16m2(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vnsrl.nxv8i16.nxv8i32.nxv8i16.i32(<vscale x 8 x i32> [[OP1:%.*]], <vscale x 8 x i16> [[SHIFT:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsrl_wv_u16m2(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vnsrl.nxv8i16.nxv8i32.nxv8i16.i64(<vscale x 8 x i32> [[OP1:%.*]], <vscale x 8 x i16> [[SHIFT:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
+//
+vuint16m2_t test_vnsrl_wv_u16m2(vuint32m4_t op1, vuint16m2_t shift, size_t vl) {
+  return vnsrl_wv_u16m2(op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsrl_wx_u16m2(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vnsrl.nxv8i16.nxv8i32.i32.i32(<vscale x 8 x i32> [[OP1:%.*]], i32 [[SHIFT:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsrl_wx_u16m2(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vnsrl.nxv8i16.nxv8i32.i64.i64(<vscale x 8 x i32> [[OP1:%.*]], i64 [[SHIFT:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
+//
+vuint16m2_t test_vnsrl_wx_u16m2(vuint32m4_t op1, size_t shift, size_t vl) {
+  return vnsrl_wx_u16m2(op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsrl_wv_u16m4(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vnsrl.nxv16i16.nxv16i32.nxv16i16.i32(<vscale x 16 x i32> [[OP1:%.*]], <vscale x 16 x i16> [[SHIFT:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 16 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsrl_wv_u16m4(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vnsrl.nxv16i16.nxv16i32.nxv16i16.i64(<vscale x 16 x i32> [[OP1:%.*]], <vscale x 16 x i16> [[SHIFT:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 16 x i16> [[TMP0]]
+//
+vuint16m4_t test_vnsrl_wv_u16m4(vuint32m8_t op1, vuint16m4_t shift, size_t vl) {
+  return vnsrl_wv_u16m4(op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsrl_wx_u16m4(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vnsrl.nxv16i16.nxv16i32.i32.i32(<vscale x 16 x i32> [[OP1:%.*]], i32 [[SHIFT:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 16 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsrl_wx_u16m4(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vnsrl.nxv16i16.nxv16i32.i64.i64(<vscale x 16 x i32> [[OP1:%.*]], i64 [[SHIFT:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 16 x i16> [[TMP0]]
+//
+vuint16m4_t test_vnsrl_wx_u16m4(vuint32m8_t op1, size_t shift, size_t vl) {
+  return vnsrl_wx_u16m4(op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsrl_wv_u32mf2(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vnsrl.nxv1i32.nxv1i64.nxv1i32.i32(<vscale x 1 x i64> [[OP1:%.*]], <vscale x 1 x i32> [[SHIFT:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsrl_wv_u32mf2(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vnsrl.nxv1i32.nxv1i64.nxv1i32.i64(<vscale x 1 x i64> [[OP1:%.*]], <vscale x 1 x i32> [[SHIFT:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
+//
+vuint32mf2_t test_vnsrl_wv_u32mf2(vuint64m1_t op1, vuint32mf2_t shift, size_t vl) {
+  return vnsrl_wv_u32mf2(op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsrl_wx_u32mf2(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vnsrl.nxv1i32.nxv1i64.i32.i32(<vscale x 1 x i64> [[OP1:%.*]], i32 [[SHIFT:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsrl_wx_u32mf2(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vnsrl.nxv1i32.nxv1i64.i64.i64(<vscale x 1 x i64> [[OP1:%.*]], i64 [[SHIFT:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
+//
+vuint32mf2_t test_vnsrl_wx_u32mf2(vuint64m1_t op1, size_t shift, size_t vl) {
+  return vnsrl_wx_u32mf2(op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsrl_wv_u32m1(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vnsrl.nxv2i32.nxv2i64.nxv2i32.i32(<vscale x 2 x i64> [[OP1:%.*]], <vscale x 2 x i32> [[SHIFT:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsrl_wv_u32m1(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vnsrl.nxv2i32.nxv2i64.nxv2i32.i64(<vscale x 2 x i64> [[OP1:%.*]], <vscale x 2 x i32> [[SHIFT:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
+//
+vuint32m1_t test_vnsrl_wv_u32m1(vuint64m2_t op1, vuint32m1_t shift, size_t vl) {
+  return vnsrl_wv_u32m1(op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsrl_wx_u32m1(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vnsrl.nxv2i32.nxv2i64.i32.i32(<vscale x 2 x i64> [[OP1:%.*]], i32 [[SHIFT:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsrl_wx_u32m1(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vnsrl.nxv2i32.nxv2i64.i64.i64(<vscale x 2 x i64> [[OP1:%.*]], i64 [[SHIFT:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
+//
+vuint32m1_t test_vnsrl_wx_u32m1(vuint64m2_t op1, size_t shift, size_t vl) {
+  return vnsrl_wx_u32m1(op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsrl_wv_u32m2(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vnsrl.nxv4i32.nxv4i64.nxv4i32.i32(<vscale x 4 x i64> [[OP1:%.*]], <vscale x 4 x i32> [[SHIFT:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsrl_wv_u32m2(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vnsrl.nxv4i32.nxv4i64.nxv4i32.i64(<vscale x 4 x i64> [[OP1:%.*]], <vscale x 4 x i32> [[SHIFT:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
+//
+vuint32m2_t test_vnsrl_wv_u32m2(vuint64m4_t op1, vuint32m2_t shift, size_t vl) {
+  return vnsrl_wv_u32m2(op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsrl_wx_u32m2(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vnsrl.nxv4i32.nxv4i64.i32.i32(<vscale x 4 x i64> [[OP1:%.*]], i32 [[SHIFT:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsrl_wx_u32m2(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vnsrl.nxv4i32.nxv4i64.i64.i64(<vscale x 4 x i64> [[OP1:%.*]], i64 [[SHIFT:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
+//
+vuint32m2_t test_vnsrl_wx_u32m2(vuint64m4_t op1, size_t shift, size_t vl) {
+  return vnsrl_wx_u32m2(op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsrl_wv_u32m4(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vnsrl.nxv8i32.nxv8i64.nxv8i32.i32(<vscale x 8 x i64> [[OP1:%.*]], <vscale x 8 x i32> [[SHIFT:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsrl_wv_u32m4(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vnsrl.nxv8i32.nxv8i64.nxv8i32.i64(<vscale x 8 x i64> [[OP1:%.*]], <vscale x 8 x i32> [[SHIFT:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
+//
+vuint32m4_t test_vnsrl_wv_u32m4(vuint64m8_t op1, vuint32m4_t shift, size_t vl) {
+  return vnsrl_wv_u32m4(op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsrl_wx_u32m4(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vnsrl.nxv8i32.nxv8i64.i32.i32(<vscale x 8 x i64> [[OP1:%.*]], i32 [[SHIFT:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsrl_wx_u32m4(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vnsrl.nxv8i32.nxv8i64.i64.i64(<vscale x 8 x i64> [[OP1:%.*]], i64 [[SHIFT:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
+//
+vuint32m4_t test_vnsrl_wx_u32m4(vuint64m8_t op1, size_t shift, size_t vl) {
+  return vnsrl_wx_u32m4(op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsrl_wv_u8mf8_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i8> @llvm.riscv.vnsrl.mask.nxv1i8.nxv1i16.nxv1i8.i32(<vscale x 1 x i8> [[MASKEDOFF:%.*]], <vscale x 1 x i16> [[OP1:%.*]], <vscale x 1 x i8> [[SHIFT:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 1 x i8> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsrl_wv_u8mf8_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i8> @llvm.riscv.vnsrl.mask.nxv1i8.nxv1i16.nxv1i8.i64(<vscale x 1 x i8> [[MASKEDOFF:%.*]], <vscale x 1 x i16> [[OP1:%.*]], <vscale x 1 x i8> [[SHIFT:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 1 x i8> [[TMP0]]
+//
+vuint8mf8_t test_vnsrl_wv_u8mf8_m(vbool64_t mask, vuint8mf8_t maskedoff, vuint16mf4_t op1, vuint8mf8_t shift, size_t vl) {
+  return vnsrl_wv_u8mf8_m(mask, maskedoff, op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsrl_wx_u8mf8_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i8> @llvm.riscv.vnsrl.mask.nxv1i8.nxv1i16.i32.i32(<vscale x 1 x i8> [[MASKEDOFF:%.*]], <vscale x 1 x i16> [[OP1:%.*]], i32 [[SHIFT:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 1 x i8> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsrl_wx_u8mf8_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i8> @llvm.riscv.vnsrl.mask.nxv1i8.nxv1i16.i64.i64(<vscale x 1 x i8> [[MASKEDOFF:%.*]], <vscale x 1 x i16> [[OP1:%.*]], i64 [[SHIFT:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 1 x i8> [[TMP0]]
+//
+vuint8mf8_t test_vnsrl_wx_u8mf8_m(vbool64_t mask, vuint8mf8_t maskedoff, vuint16mf4_t op1, size_t shift, size_t vl) {
+  return vnsrl_wx_u8mf8_m(mask, maskedoff, op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsrl_wv_u8mf4_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i8> @llvm.riscv.vnsrl.mask.nxv2i8.nxv2i16.nxv2i8.i32(<vscale x 2 x i8> [[MASKEDOFF:%.*]], <vscale x 2 x i16> [[OP1:%.*]], <vscale x 2 x i8> [[SHIFT:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 2 x i8> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsrl_wv_u8mf4_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i8> @llvm.riscv.vnsrl.mask.nxv2i8.nxv2i16.nxv2i8.i64(<vscale x 2 x i8> [[MASKEDOFF:%.*]], <vscale x 2 x i16> [[OP1:%.*]], <vscale x 2 x i8> [[SHIFT:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 2 x i8> [[TMP0]]
+//
+vuint8mf4_t test_vnsrl_wv_u8mf4_m(vbool32_t mask, vuint8mf4_t maskedoff, vuint16mf2_t op1, vuint8mf4_t shift, size_t vl) {
+  return vnsrl_wv_u8mf4_m(mask, maskedoff, op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsrl_wx_u8mf4_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i8> @llvm.riscv.vnsrl.mask.nxv2i8.nxv2i16.i32.i32(<vscale x 2 x i8> [[MASKEDOFF:%.*]], <vscale x 2 x i16> [[OP1:%.*]], i32 [[SHIFT:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 2 x i8> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsrl_wx_u8mf4_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i8> @llvm.riscv.vnsrl.mask.nxv2i8.nxv2i16.i64.i64(<vscale x 2 x i8> [[MASKEDOFF:%.*]], <vscale x 2 x i16> [[OP1:%.*]], i64 [[SHIFT:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 2 x i8> [[TMP0]]
+//
+vuint8mf4_t test_vnsrl_wx_u8mf4_m(vbool32_t mask, vuint8mf4_t maskedoff, vuint16mf2_t op1, size_t shift, size_t vl) {
+  return vnsrl_wx_u8mf4_m(mask, maskedoff, op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsrl_wv_u8mf2_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i8> @llvm.riscv.vnsrl.mask.nxv4i8.nxv4i16.nxv4i8.i32(<vscale x 4 x i8> [[MASKEDOFF:%.*]], <vscale x 4 x i16> [[OP1:%.*]], <vscale x 4 x i8> [[SHIFT:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 4 x i8> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsrl_wv_u8mf2_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i8> @llvm.riscv.vnsrl.mask.nxv4i8.nxv4i16.nxv4i8.i64(<vscale x 4 x i8> [[MASKEDOFF:%.*]], <vscale x 4 x i16> [[OP1:%.*]], <vscale x 4 x i8> [[SHIFT:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 4 x i8> [[TMP0]]
+//
+vuint8mf2_t test_vnsrl_wv_u8mf2_m(vbool16_t mask, vuint8mf2_t maskedoff, vuint16m1_t op1, vuint8mf2_t shift, size_t vl) {
+  return vnsrl_wv_u8mf2_m(mask, maskedoff, op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsrl_wx_u8mf2_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i8> @llvm.riscv.vnsrl.mask.nxv4i8.nxv4i16.i32.i32(<vscale x 4 x i8> [[MASKEDOFF:%.*]], <vscale x 4 x i16> [[OP1:%.*]], i32 [[SHIFT:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 4 x i8> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsrl_wx_u8mf2_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i8> @llvm.riscv.vnsrl.mask.nxv4i8.nxv4i16.i64.i64(<vscale x 4 x i8> [[MASKEDOFF:%.*]], <vscale x 4 x i16> [[OP1:%.*]], i64 [[SHIFT:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 4 x i8> [[TMP0]]
+//
+vuint8mf2_t test_vnsrl_wx_u8mf2_m(vbool16_t mask, vuint8mf2_t maskedoff, vuint16m1_t op1, size_t shift, size_t vl) {
+  return vnsrl_wx_u8mf2_m(mask, maskedoff, op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsrl_wv_u8m1_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i8> @llvm.riscv.vnsrl.mask.nxv8i8.nxv8i16.nxv8i8.i32(<vscale x 8 x i8> [[MASKEDOFF:%.*]], <vscale x 8 x i16> [[OP1:%.*]], <vscale x 8 x i8> [[SHIFT:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 8 x i8> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsrl_wv_u8m1_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i8> @llvm.riscv.vnsrl.mask.nxv8i8.nxv8i16.nxv8i8.i64(<vscale x 8 x i8> [[MASKEDOFF:%.*]], <vscale x 8 x i16> [[OP1:%.*]], <vscale x 8 x i8> [[SHIFT:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 8 x i8> [[TMP0]]
+//
+vuint8m1_t test_vnsrl_wv_u8m1_m(vbool8_t mask, vuint8m1_t maskedoff, vuint16m2_t op1, vuint8m1_t shift, size_t vl) {
+  return vnsrl_wv_u8m1_m(mask, maskedoff, op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsrl_wx_u8m1_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i8> @llvm.riscv.vnsrl.mask.nxv8i8.nxv8i16.i32.i32(<vscale x 8 x i8> [[MASKEDOFF:%.*]], <vscale x 8 x i16> [[OP1:%.*]], i32 [[SHIFT:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 8 x i8> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsrl_wx_u8m1_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i8> @llvm.riscv.vnsrl.mask.nxv8i8.nxv8i16.i64.i64(<vscale x 8 x i8> [[MASKEDOFF:%.*]], <vscale x 8 x i16> [[OP1:%.*]], i64 [[SHIFT:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 8 x i8> [[TMP0]]
+//
+vuint8m1_t test_vnsrl_wx_u8m1_m(vbool8_t mask, vuint8m1_t maskedoff, vuint16m2_t op1, size_t shift, size_t vl) {
+  return vnsrl_wx_u8m1_m(mask, maskedoff, op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsrl_wv_u8m2_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i8> @llvm.riscv.vnsrl.mask.nxv16i8.nxv16i16.nxv16i8.i32(<vscale x 16 x i8> [[MASKEDOFF:%.*]], <vscale x 16 x i16> [[OP1:%.*]], <vscale x 16 x i8> [[SHIFT:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 16 x i8> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsrl_wv_u8m2_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i8> @llvm.riscv.vnsrl.mask.nxv16i8.nxv16i16.nxv16i8.i64(<vscale x 16 x i8> [[MASKEDOFF:%.*]], <vscale x 16 x i16> [[OP1:%.*]], <vscale x 16 x i8> [[SHIFT:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 16 x i8> [[TMP0]]
+//
+vuint8m2_t test_vnsrl_wv_u8m2_m(vbool4_t mask, vuint8m2_t maskedoff, vuint16m4_t op1, vuint8m2_t shift, size_t vl) {
+  return vnsrl_wv_u8m2_m(mask, maskedoff, op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsrl_wx_u8m2_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i8> @llvm.riscv.vnsrl.mask.nxv16i8.nxv16i16.i32.i32(<vscale x 16 x i8> [[MASKEDOFF:%.*]], <vscale x 16 x i16> [[OP1:%.*]], i32 [[SHIFT:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 16 x i8> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsrl_wx_u8m2_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i8> @llvm.riscv.vnsrl.mask.nxv16i8.nxv16i16.i64.i64(<vscale x 16 x i8> [[MASKEDOFF:%.*]], <vscale x 16 x i16> [[OP1:%.*]], i64 [[SHIFT:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 16 x i8> [[TMP0]]
+//
+vuint8m2_t test_vnsrl_wx_u8m2_m(vbool4_t mask, vuint8m2_t maskedoff, vuint16m4_t op1, size_t shift, size_t vl) {
+  return vnsrl_wx_u8m2_m(mask, maskedoff, op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsrl_wv_u8m4_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i8> @llvm.riscv.vnsrl.mask.nxv32i8.nxv32i16.nxv32i8.i32(<vscale x 32 x i8> [[MASKEDOFF:%.*]], <vscale x 32 x i16> [[OP1:%.*]], <vscale x 32 x i8> [[SHIFT:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 32 x i8> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsrl_wv_u8m4_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i8> @llvm.riscv.vnsrl.mask.nxv32i8.nxv32i16.nxv32i8.i64(<vscale x 32 x i8> [[MASKEDOFF:%.*]], <vscale x 32 x i16> [[OP1:%.*]], <vscale x 32 x i8> [[SHIFT:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 32 x i8> [[TMP0]]
+//
+vuint8m4_t test_vnsrl_wv_u8m4_m(vbool2_t mask, vuint8m4_t maskedoff, vuint16m8_t op1, vuint8m4_t shift, size_t vl) {
+  return vnsrl_wv_u8m4_m(mask, maskedoff, op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsrl_wx_u8m4_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i8> @llvm.riscv.vnsrl.mask.nxv32i8.nxv32i16.i32.i32(<vscale x 32 x i8> [[MASKEDOFF:%.*]], <vscale x 32 x i16> [[OP1:%.*]], i32 [[SHIFT:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 32 x i8> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsrl_wx_u8m4_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i8> @llvm.riscv.vnsrl.mask.nxv32i8.nxv32i16.i64.i64(<vscale x 32 x i8> [[MASKEDOFF:%.*]], <vscale x 32 x i16> [[OP1:%.*]], i64 [[SHIFT:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 32 x i8> [[TMP0]]
+//
+vuint8m4_t test_vnsrl_wx_u8m4_m(vbool2_t mask, vuint8m4_t maskedoff, vuint16m8_t op1, size_t shift, size_t vl) {
+  return vnsrl_wx_u8m4_m(mask, maskedoff, op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsrl_wv_u16mf4_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vnsrl.mask.nxv1i16.nxv1i32.nxv1i16.i32(<vscale x 1 x i16> [[MASKEDOFF:%.*]], <vscale x 1 x i32> [[OP1:%.*]], <vscale x 1 x i16> [[SHIFT:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 1 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsrl_wv_u16mf4_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vnsrl.mask.nxv1i16.nxv1i32.nxv1i16.i64(<vscale x 1 x i16> [[MASKEDOFF:%.*]], <vscale x 1 x i32> [[OP1:%.*]], <vscale x 1 x i16> [[SHIFT:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 1 x i16> [[TMP0]]
+//
+vuint16mf4_t test_vnsrl_wv_u16mf4_m(vbool64_t mask, vuint16mf4_t maskedoff, vuint32mf2_t op1, vuint16mf4_t shift, size_t vl) {
+  return vnsrl_wv_u16mf4_m(mask, maskedoff, op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsrl_wx_u16mf4_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vnsrl.mask.nxv1i16.nxv1i32.i32.i32(<vscale x 1 x i16> [[MASKEDOFF:%.*]], <vscale x 1 x i32> [[OP1:%.*]], i32 [[SHIFT:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 1 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsrl_wx_u16mf4_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vnsrl.mask.nxv1i16.nxv1i32.i64.i64(<vscale x 1 x i16> [[MASKEDOFF:%.*]], <vscale x 1 x i32> [[OP1:%.*]], i64 [[SHIFT:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 1 x i16> [[TMP0]]
+//
+vuint16mf4_t test_vnsrl_wx_u16mf4_m(vbool64_t mask, vuint16mf4_t maskedoff, vuint32mf2_t op1, size_t shift, size_t vl) {
+  return vnsrl_wx_u16mf4_m(mask, maskedoff, op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsrl_wv_u16mf2_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vnsrl.mask.nxv2i16.nxv2i32.nxv2i16.i32(<vscale x 2 x i16> [[MASKEDOFF:%.*]], <vscale x 2 x i32> [[OP1:%.*]], <vscale x 2 x i16> [[SHIFT:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 2 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsrl_wv_u16mf2_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vnsrl.mask.nxv2i16.nxv2i32.nxv2i16.i64(<vscale x 2 x i16> [[MASKEDOFF:%.*]], <vscale x 2 x i32> [[OP1:%.*]], <vscale x 2 x i16> [[SHIFT:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 2 x i16> [[TMP0]]
+//
+vuint16mf2_t test_vnsrl_wv_u16mf2_m(vbool32_t mask, vuint16mf2_t maskedoff, vuint32m1_t op1, vuint16mf2_t shift, size_t vl) {
+  return vnsrl_wv_u16mf2_m(mask, maskedoff, op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsrl_wx_u16mf2_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vnsrl.mask.nxv2i16.nxv2i32.i32.i32(<vscale x 2 x i16> [[MASKEDOFF:%.*]], <vscale x 2 x i32> [[OP1:%.*]], i32 [[SHIFT:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 2 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsrl_wx_u16mf2_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vnsrl.mask.nxv2i16.nxv2i32.i64.i64(<vscale x 2 x i16> [[MASKEDOFF:%.*]], <vscale x 2 x i32> [[OP1:%.*]], i64 [[SHIFT:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 2 x i16> [[TMP0]]
+//
+vuint16mf2_t test_vnsrl_wx_u16mf2_m(vbool32_t mask, vuint16mf2_t maskedoff, vuint32m1_t op1, size_t shift, size_t vl) {
+  return vnsrl_wx_u16mf2_m(mask, maskedoff, op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsrl_wv_u16m1_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vnsrl.mask.nxv4i16.nxv4i32.nxv4i16.i32(<vscale x 4 x i16> [[MASKEDOFF:%.*]], <vscale x 4 x i32> [[OP1:%.*]], <vscale x 4 x i16> [[SHIFT:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 4 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsrl_wv_u16m1_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vnsrl.mask.nxv4i16.nxv4i32.nxv4i16.i64(<vscale x 4 x i16> [[MASKEDOFF:%.*]], <vscale x 4 x i32> [[OP1:%.*]], <vscale x 4 x i16> [[SHIFT:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 4 x i16> [[TMP0]]
+//
+vuint16m1_t test_vnsrl_wv_u16m1_m(vbool16_t mask, vuint16m1_t maskedoff, vuint32m2_t op1, vuint16m1_t shift, size_t vl) {
+  return vnsrl_wv_u16m1_m(mask, maskedoff, op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsrl_wx_u16m1_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vnsrl.mask.nxv4i16.nxv4i32.i32.i32(<vscale x 4 x i16> [[MASKEDOFF:%.*]], <vscale x 4 x i32> [[OP1:%.*]], i32 [[SHIFT:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 4 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsrl_wx_u16m1_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vnsrl.mask.nxv4i16.nxv4i32.i64.i64(<vscale x 4 x i16> [[MASKEDOFF:%.*]], <vscale x 4 x i32> [[OP1:%.*]], i64 [[SHIFT:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 4 x i16> [[TMP0]]
+//
+vuint16m1_t test_vnsrl_wx_u16m1_m(vbool16_t mask, vuint16m1_t maskedoff, vuint32m2_t op1, size_t shift, size_t vl) {
+  return vnsrl_wx_u16m1_m(mask, maskedoff, op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsrl_wv_u16m2_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vnsrl.mask.nxv8i16.nxv8i32.nxv8i16.i32(<vscale x 8 x i16> [[MASKEDOFF:%.*]], <vscale x 8 x i32> [[OP1:%.*]], <vscale x 8 x i16> [[SHIFT:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsrl_wv_u16m2_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vnsrl.mask.nxv8i16.nxv8i32.nxv8i16.i64(<vscale x 8 x i16> [[MASKEDOFF:%.*]], <vscale x 8 x i32> [[OP1:%.*]], <vscale x 8 x i16> [[SHIFT:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
+//
+vuint16m2_t test_vnsrl_wv_u16m2_m(vbool8_t mask, vuint16m2_t maskedoff, vuint32m4_t op1, vuint16m2_t shift, size_t vl) {
+  return vnsrl_wv_u16m2_m(mask, maskedoff, op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsrl_wx_u16m2_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vnsrl.mask.nxv8i16.nxv8i32.i32.i32(<vscale x 8 x i16> [[MASKEDOFF:%.*]], <vscale x 8 x i32> [[OP1:%.*]], i32 [[SHIFT:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsrl_wx_u16m2_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vnsrl.mask.nxv8i16.nxv8i32.i64.i64(<vscale x 8 x i16> [[MASKEDOFF:%.*]], <vscale x 8 x i32> [[OP1:%.*]], i64 [[SHIFT:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
+//
+vuint16m2_t test_vnsrl_wx_u16m2_m(vbool8_t mask, vuint16m2_t maskedoff, vuint32m4_t op1, size_t shift, size_t vl) {
+  return vnsrl_wx_u16m2_m(mask, maskedoff, op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsrl_wv_u16m4_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vnsrl.mask.nxv16i16.nxv16i32.nxv16i16.i32(<vscale x 16 x i16> [[MASKEDOFF:%.*]], <vscale x 16 x i32> [[OP1:%.*]], <vscale x 16 x i16> [[SHIFT:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 16 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsrl_wv_u16m4_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vnsrl.mask.nxv16i16.nxv16i32.nxv16i16.i64(<vscale x 16 x i16> [[MASKEDOFF:%.*]], <vscale x 16 x i32> [[OP1:%.*]], <vscale x 16 x i16> [[SHIFT:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 16 x i16> [[TMP0]]
+//
+vuint16m4_t test_vnsrl_wv_u16m4_m(vbool4_t mask, vuint16m4_t maskedoff, vuint32m8_t op1, vuint16m4_t shift, size_t vl) {
+  return vnsrl_wv_u16m4_m(mask, maskedoff, op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsrl_wx_u16m4_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vnsrl.mask.nxv16i16.nxv16i32.i32.i32(<vscale x 16 x i16> [[MASKEDOFF:%.*]], <vscale x 16 x i32> [[OP1:%.*]], i32 [[SHIFT:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 16 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsrl_wx_u16m4_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vnsrl.mask.nxv16i16.nxv16i32.i64.i64(<vscale x 16 x i16> [[MASKEDOFF:%.*]], <vscale x 16 x i32> [[OP1:%.*]], i64 [[SHIFT:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 16 x i16> [[TMP0]]
+//
+vuint16m4_t test_vnsrl_wx_u16m4_m(vbool4_t mask, vuint16m4_t maskedoff, vuint32m8_t op1, size_t shift, size_t vl) {
+  return vnsrl_wx_u16m4_m(mask, maskedoff, op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsrl_wv_u32mf2_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vnsrl.mask.nxv1i32.nxv1i64.nxv1i32.i32(<vscale x 1 x i32> [[MASKEDOFF:%.*]], <vscale x 1 x i64> [[OP1:%.*]], <vscale x 1 x i32> [[SHIFT:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsrl_wv_u32mf2_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vnsrl.mask.nxv1i32.nxv1i64.nxv1i32.i64(<vscale x 1 x i32> [[MASKEDOFF:%.*]], <vscale x 1 x i64> [[OP1:%.*]], <vscale x 1 x i32> [[SHIFT:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
+//
+vuint32mf2_t test_vnsrl_wv_u32mf2_m(vbool64_t mask, vuint32mf2_t maskedoff, vuint64m1_t op1, vuint32mf2_t shift, size_t vl) {
+  return vnsrl_wv_u32mf2_m(mask, maskedoff, op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsrl_wx_u32mf2_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vnsrl.mask.nxv1i32.nxv1i64.i32.i32(<vscale x 1 x i32> [[MASKEDOFF:%.*]], <vscale x 1 x i64> [[OP1:%.*]], i32 [[SHIFT:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsrl_wx_u32mf2_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vnsrl.mask.nxv1i32.nxv1i64.i64.i64(<vscale x 1 x i32> [[MASKEDOFF:%.*]], <vscale x 1 x i64> [[OP1:%.*]], i64 [[SHIFT:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
+//
+vuint32mf2_t test_vnsrl_wx_u32mf2_m(vbool64_t mask, vuint32mf2_t maskedoff, vuint64m1_t op1, size_t shift, size_t vl) {
+  return vnsrl_wx_u32mf2_m(mask, maskedoff, op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsrl_wv_u32m1_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vnsrl.mask.nxv2i32.nxv2i64.nxv2i32.i32(<vscale x 2 x i32> [[MASKEDOFF:%.*]], <vscale x 2 x i64> [[OP1:%.*]], <vscale x 2 x i32> [[SHIFT:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsrl_wv_u32m1_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vnsrl.mask.nxv2i32.nxv2i64.nxv2i32.i64(<vscale x 2 x i32> [[MASKEDOFF:%.*]], <vscale x 2 x i64> [[OP1:%.*]], <vscale x 2 x i32> [[SHIFT:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
+//
+vuint32m1_t test_vnsrl_wv_u32m1_m(vbool32_t mask, vuint32m1_t maskedoff, vuint64m2_t op1, vuint32m1_t shift, size_t vl) {
+  return vnsrl_wv_u32m1_m(mask, maskedoff, op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsrl_wx_u32m1_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vnsrl.mask.nxv2i32.nxv2i64.i32.i32(<vscale x 2 x i32> [[MASKEDOFF:%.*]], <vscale x 2 x i64> [[OP1:%.*]], i32 [[SHIFT:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsrl_wx_u32m1_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vnsrl.mask.nxv2i32.nxv2i64.i64.i64(<vscale x 2 x i32> [[MASKEDOFF:%.*]], <vscale x 2 x i64> [[OP1:%.*]], i64 [[SHIFT:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
+//
+vuint32m1_t test_vnsrl_wx_u32m1_m(vbool32_t mask, vuint32m1_t maskedoff, vuint64m2_t op1, size_t shift, size_t vl) {
+  return vnsrl_wx_u32m1_m(mask, maskedoff, op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsrl_wv_u32m2_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vnsrl.mask.nxv4i32.nxv4i64.nxv4i32.i32(<vscale x 4 x i32> [[MASKEDOFF:%.*]], <vscale x 4 x i64> [[OP1:%.*]], <vscale x 4 x i32> [[SHIFT:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsrl_wv_u32m2_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vnsrl.mask.nxv4i32.nxv4i64.nxv4i32.i64(<vscale x 4 x i32> [[MASKEDOFF:%.*]], <vscale x 4 x i64> [[OP1:%.*]], <vscale x 4 x i32> [[SHIFT:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
+//
+vuint32m2_t test_vnsrl_wv_u32m2_m(vbool16_t mask, vuint32m2_t maskedoff, vuint64m4_t op1, vuint32m2_t shift, size_t vl) {
+  return vnsrl_wv_u32m2_m(mask, maskedoff, op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsrl_wx_u32m2_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vnsrl.mask.nxv4i32.nxv4i64.i32.i32(<vscale x 4 x i32> [[MASKEDOFF:%.*]], <vscale x 4 x i64> [[OP1:%.*]], i32 [[SHIFT:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsrl_wx_u32m2_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vnsrl.mask.nxv4i32.nxv4i64.i64.i64(<vscale x 4 x i32> [[MASKEDOFF:%.*]], <vscale x 4 x i64> [[OP1:%.*]], i64 [[SHIFT:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
+//
+vuint32m2_t test_vnsrl_wx_u32m2_m(vbool16_t mask, vuint32m2_t maskedoff, vuint64m4_t op1, size_t shift, size_t vl) {
+  return vnsrl_wx_u32m2_m(mask, maskedoff, op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsrl_wv_u32m4_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vnsrl.mask.nxv8i32.nxv8i64.nxv8i32.i32(<vscale x 8 x i32> [[MASKEDOFF:%.*]], <vscale x 8 x i64> [[OP1:%.*]], <vscale x 8 x i32> [[SHIFT:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsrl_wv_u32m4_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vnsrl.mask.nxv8i32.nxv8i64.nxv8i32.i64(<vscale x 8 x i32> [[MASKEDOFF:%.*]], <vscale x 8 x i64> [[OP1:%.*]], <vscale x 8 x i32> [[SHIFT:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
+//
+vuint32m4_t test_vnsrl_wv_u32m4_m(vbool8_t mask, vuint32m4_t maskedoff, vuint64m8_t op1, vuint32m4_t shift, size_t vl) {
+  return vnsrl_wv_u32m4_m(mask, maskedoff, op1, shift, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vnsrl_wx_u32m4_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vnsrl.mask.nxv8i32.nxv8i64.i32.i32(<vscale x 8 x i32> [[MASKEDOFF:%.*]], <vscale x 8 x i64> [[OP1:%.*]], i32 [[SHIFT:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vnsrl_wx_u32m4_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vnsrl.mask.nxv8i32.nxv8i64.i64.i64(<vscale x 8 x i32> [[MASKEDOFF:%.*]], <vscale x 8 x i64> [[OP1:%.*]], i64 [[SHIFT:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
+//
+vuint32m4_t test_vnsrl_wx_u32m4_m(vbool8_t mask, vuint32m4_t maskedoff, vuint64m8_t op1, size_t shift, size_t vl) {
+  return vnsrl_wx_u32m4_m(mask, maskedoff, op1, shift, vl);
+}

clang/test/CodeGen/RISCV/rvv-intrinsics/vwmul.c

+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv32 -target-feature +experimental-v -disable-O0-optnone -emit-llvm %s -o - | opt -S -mem2reg | FileCheck --check-prefix=CHECK-RV32 %s
+// RUN: %clang_cc1 -triple riscv64 -target-feature +experimental-v -disable-O0-optnone -emit-llvm %s -o - | opt -S -mem2reg | FileCheck --check-prefix=CHECK-RV64 %s
+// RUN: %clang_cc1 -triple riscv64 -target-feature +m -target-feature +experimental-v -Werror -Wall -o - %s -S >/dev/null 2>&1 | FileCheck --check-prefix=ASM --allow-empty %s
+
+// ASM-NOT: warning
+#include <riscv_vector.h>
+
+// CHECK-RV32-LABEL: @test_vwmul_vv_i16mf4(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vwmul.nxv1i16.nxv1i8.nxv1i8.i32(<vscale x 1 x i8> [[OP1:%.*]], <vscale x 1 x i8> [[OP2:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 1 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmul_vv_i16mf4(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vwmul.nxv1i16.nxv1i8.nxv1i8.i64(<vscale x 1 x i8> [[OP1:%.*]], <vscale x 1 x i8> [[OP2:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 1 x i16> [[TMP0]]
+//
+vint16mf4_t test_vwmul_vv_i16mf4(vint8mf8_t op1, vint8mf8_t op2, size_t vl) {
+  return vwmul_vv_i16mf4(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmul_vx_i16mf4(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vwmul.nxv1i16.nxv1i8.i8.i32(<vscale x 1 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 1 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmul_vx_i16mf4(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vwmul.nxv1i16.nxv1i8.i8.i64(<vscale x 1 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 1 x i16> [[TMP0]]
+//
+vint16mf4_t test_vwmul_vx_i16mf4(vint8mf8_t op1, int8_t op2, size_t vl) {
+  return vwmul_vx_i16mf4(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmul_vv_i16mf2(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vwmul.nxv2i16.nxv2i8.nxv2i8.i32(<vscale x 2 x i8> [[OP1:%.*]], <vscale x 2 x i8> [[OP2:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 2 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmul_vv_i16mf2(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vwmul.nxv2i16.nxv2i8.nxv2i8.i64(<vscale x 2 x i8> [[OP1:%.*]], <vscale x 2 x i8> [[OP2:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 2 x i16> [[TMP0]]
+//
+vint16mf2_t test_vwmul_vv_i16mf2(vint8mf4_t op1, vint8mf4_t op2, size_t vl) {
+  return vwmul_vv_i16mf2(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmul_vx_i16mf2(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vwmul.nxv2i16.nxv2i8.i8.i32(<vscale x 2 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 2 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmul_vx_i16mf2(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vwmul.nxv2i16.nxv2i8.i8.i64(<vscale x 2 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 2 x i16> [[TMP0]]
+//
+vint16mf2_t test_vwmul_vx_i16mf2(vint8mf4_t op1, int8_t op2, size_t vl) {
+  return vwmul_vx_i16mf2(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmul_vv_i16m1(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vwmul.nxv4i16.nxv4i8.nxv4i8.i32(<vscale x 4 x i8> [[OP1:%.*]], <vscale x 4 x i8> [[OP2:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 4 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmul_vv_i16m1(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vwmul.nxv4i16.nxv4i8.nxv4i8.i64(<vscale x 4 x i8> [[OP1:%.*]], <vscale x 4 x i8> [[OP2:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 4 x i16> [[TMP0]]
+//
+vint16m1_t test_vwmul_vv_i16m1(vint8mf2_t op1, vint8mf2_t op2, size_t vl) {
+  return vwmul_vv_i16m1(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmul_vx_i16m1(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vwmul.nxv4i16.nxv4i8.i8.i32(<vscale x 4 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 4 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmul_vx_i16m1(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vwmul.nxv4i16.nxv4i8.i8.i64(<vscale x 4 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 4 x i16> [[TMP0]]
+//
+vint16m1_t test_vwmul_vx_i16m1(vint8mf2_t op1, int8_t op2, size_t vl) {
+  return vwmul_vx_i16m1(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmul_vv_i16m2(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vwmul.nxv8i16.nxv8i8.nxv8i8.i32(<vscale x 8 x i8> [[OP1:%.*]], <vscale x 8 x i8> [[OP2:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmul_vv_i16m2(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vwmul.nxv8i16.nxv8i8.nxv8i8.i64(<vscale x 8 x i8> [[OP1:%.*]], <vscale x 8 x i8> [[OP2:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
+//
+vint16m2_t test_vwmul_vv_i16m2(vint8m1_t op1, vint8m1_t op2, size_t vl) {
+  return vwmul_vv_i16m2(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmul_vx_i16m2(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vwmul.nxv8i16.nxv8i8.i8.i32(<vscale x 8 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmul_vx_i16m2(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vwmul.nxv8i16.nxv8i8.i8.i64(<vscale x 8 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
+//
+vint16m2_t test_vwmul_vx_i16m2(vint8m1_t op1, int8_t op2, size_t vl) {
+  return vwmul_vx_i16m2(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmul_vv_i16m4(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vwmul.nxv16i16.nxv16i8.nxv16i8.i32(<vscale x 16 x i8> [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 16 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmul_vv_i16m4(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vwmul.nxv16i16.nxv16i8.nxv16i8.i64(<vscale x 16 x i8> [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 16 x i16> [[TMP0]]
+//
+vint16m4_t test_vwmul_vv_i16m4(vint8m2_t op1, vint8m2_t op2, size_t vl) {
+  return vwmul_vv_i16m4(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmul_vx_i16m4(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vwmul.nxv16i16.nxv16i8.i8.i32(<vscale x 16 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 16 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmul_vx_i16m4(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vwmul.nxv16i16.nxv16i8.i8.i64(<vscale x 16 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 16 x i16> [[TMP0]]
+//
+vint16m4_t test_vwmul_vx_i16m4(vint8m2_t op1, int8_t op2, size_t vl) {
+  return vwmul_vx_i16m4(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmul_vv_i16m8(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vwmul.nxv32i16.nxv32i8.nxv32i8.i32(<vscale x 32 x i8> [[OP1:%.*]], <vscale x 32 x i8> [[OP2:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 32 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmul_vv_i16m8(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vwmul.nxv32i16.nxv32i8.nxv32i8.i64(<vscale x 32 x i8> [[OP1:%.*]], <vscale x 32 x i8> [[OP2:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 32 x i16> [[TMP0]]
+//
+vint16m8_t test_vwmul_vv_i16m8(vint8m4_t op1, vint8m4_t op2, size_t vl) {
+  return vwmul_vv_i16m8(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmul_vx_i16m8(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vwmul.nxv32i16.nxv32i8.i8.i32(<vscale x 32 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 32 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmul_vx_i16m8(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vwmul.nxv32i16.nxv32i8.i8.i64(<vscale x 32 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 32 x i16> [[TMP0]]
+//
+vint16m8_t test_vwmul_vx_i16m8(vint8m4_t op1, int8_t op2, size_t vl) {
+  return vwmul_vx_i16m8(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmul_vv_i32mf2(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vwmul.nxv1i32.nxv1i16.nxv1i16.i32(<vscale x 1 x i16> [[OP1:%.*]], <vscale x 1 x i16> [[OP2:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmul_vv_i32mf2(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vwmul.nxv1i32.nxv1i16.nxv1i16.i64(<vscale x 1 x i16> [[OP1:%.*]], <vscale x 1 x i16> [[OP2:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
+//
+vint32mf2_t test_vwmul_vv_i32mf2(vint16mf4_t op1, vint16mf4_t op2, size_t vl) {
+  return vwmul_vv_i32mf2(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmul_vx_i32mf2(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vwmul.nxv1i32.nxv1i16.i16.i32(<vscale x 1 x i16> [[OP1:%.*]], i16 [[OP2:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmul_vx_i32mf2(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vwmul.nxv1i32.nxv1i16.i16.i64(<vscale x 1 x i16> [[OP1:%.*]], i16 [[OP2:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
+//
+vint32mf2_t test_vwmul_vx_i32mf2(vint16mf4_t op1, int16_t op2, size_t vl) {
+  return vwmul_vx_i32mf2(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmul_vv_i32m1(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vwmul.nxv2i32.nxv2i16.nxv2i16.i32(<vscale x 2 x i16> [[OP1:%.*]], <vscale x 2 x i16> [[OP2:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmul_vv_i32m1(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vwmul.nxv2i32.nxv2i16.nxv2i16.i64(<vscale x 2 x i16> [[OP1:%.*]], <vscale x 2 x i16> [[OP2:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
+//
+vint32m1_t test_vwmul_vv_i32m1(vint16mf2_t op1, vint16mf2_t op2, size_t vl) {
+  return vwmul_vv_i32m1(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmul_vx_i32m1(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vwmul.nxv2i32.nxv2i16.i16.i32(<vscale x 2 x i16> [[OP1:%.*]], i16 [[OP2:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmul_vx_i32m1(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vwmul.nxv2i32.nxv2i16.i16.i64(<vscale x 2 x i16> [[OP1:%.*]], i16 [[OP2:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
+//
+vint32m1_t test_vwmul_vx_i32m1(vint16mf2_t op1, int16_t op2, size_t vl) {
+  return vwmul_vx_i32m1(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmul_vv_i32m2(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vwmul.nxv4i32.nxv4i16.nxv4i16.i32(<vscale x 4 x i16> [[OP1:%.*]], <vscale x 4 x i16> [[OP2:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmul_vv_i32m2(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vwmul.nxv4i32.nxv4i16.nxv4i16.i64(<vscale x 4 x i16> [[OP1:%.*]], <vscale x 4 x i16> [[OP2:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
+//
+vint32m2_t test_vwmul_vv_i32m2(vint16m1_t op1, vint16m1_t op2, size_t vl) {
+  return vwmul_vv_i32m2(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmul_vx_i32m2(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vwmul.nxv4i32.nxv4i16.i16.i32(<vscale x 4 x i16> [[OP1:%.*]], i16 [[OP2:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmul_vx_i32m2(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vwmul.nxv4i32.nxv4i16.i16.i64(<vscale x 4 x i16> [[OP1:%.*]], i16 [[OP2:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
+//
+vint32m2_t test_vwmul_vx_i32m2(vint16m1_t op1, int16_t op2, size_t vl) {
+  return vwmul_vx_i32m2(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmul_vv_i32m4(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vwmul.nxv8i32.nxv8i16.nxv8i16.i32(<vscale x 8 x i16> [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmul_vv_i32m4(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vwmul.nxv8i32.nxv8i16.nxv8i16.i64(<vscale x 8 x i16> [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
+//
+vint32m4_t test_vwmul_vv_i32m4(vint16m2_t op1, vint16m2_t op2, size_t vl) {
+  return vwmul_vv_i32m4(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmul_vx_i32m4(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vwmul.nxv8i32.nxv8i16.i16.i32(<vscale x 8 x i16> [[OP1:%.*]], i16 [[OP2:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmul_vx_i32m4(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vwmul.nxv8i32.nxv8i16.i16.i64(<vscale x 8 x i16> [[OP1:%.*]], i16 [[OP2:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
+//
+vint32m4_t test_vwmul_vx_i32m4(vint16m2_t op1, int16_t op2, size_t vl) {
+  return vwmul_vx_i32m4(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmul_vv_i32m8(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vwmul.nxv16i32.nxv16i16.nxv16i16.i32(<vscale x 16 x i16> [[OP1:%.*]], <vscale x 16 x i16> [[OP2:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmul_vv_i32m8(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vwmul.nxv16i32.nxv16i16.nxv16i16.i64(<vscale x 16 x i16> [[OP1:%.*]], <vscale x 16 x i16> [[OP2:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
+//
+vint32m8_t test_vwmul_vv_i32m8(vint16m4_t op1, vint16m4_t op2, size_t vl) {
+  return vwmul_vv_i32m8(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmul_vx_i32m8(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vwmul.nxv16i32.nxv16i16.i16.i32(<vscale x 16 x i16> [[OP1:%.*]], i16 [[OP2:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmul_vx_i32m8(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vwmul.nxv16i32.nxv16i16.i16.i64(<vscale x 16 x i16> [[OP1:%.*]], i16 [[OP2:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
+//
+vint32m8_t test_vwmul_vx_i32m8(vint16m4_t op1, int16_t op2, size_t vl) {
+  return vwmul_vx_i32m8(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmul_vv_i64m1(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vwmul.nxv1i64.nxv1i32.nxv1i32.i32(<vscale x 1 x i32> [[OP1:%.*]], <vscale x 1 x i32> [[OP2:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 1 x i64> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmul_vv_i64m1(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vwmul.nxv1i64.nxv1i32.nxv1i32.i64(<vscale x 1 x i32> [[OP1:%.*]], <vscale x 1 x i32> [[OP2:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 1 x i64> [[TMP0]]
+//
+vint64m1_t test_vwmul_vv_i64m1(vint32mf2_t op1, vint32mf2_t op2, size_t vl) {
+  return vwmul_vv_i64m1(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmul_vx_i64m1(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vwmul.nxv1i64.nxv1i32.i32.i32(<vscale x 1 x i32> [[OP1:%.*]], i32 [[OP2:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 1 x i64> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmul_vx_i64m1(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vwmul.nxv1i64.nxv1i32.i32.i64(<vscale x 1 x i32> [[OP1:%.*]], i32 [[OP2:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 1 x i64> [[TMP0]]
+//
+vint64m1_t test_vwmul_vx_i64m1(vint32mf2_t op1, int32_t op2, size_t vl) {
+  return vwmul_vx_i64m1(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmul_vv_i64m2(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vwmul.nxv2i64.nxv2i32.nxv2i32.i32(<vscale x 2 x i32> [[OP1:%.*]], <vscale x 2 x i32> [[OP2:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 2 x i64> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmul_vv_i64m2(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vwmul.nxv2i64.nxv2i32.nxv2i32.i64(<vscale x 2 x i32> [[OP1:%.*]], <vscale x 2 x i32> [[OP2:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 2 x i64> [[TMP0]]
+//
+vint64m2_t test_vwmul_vv_i64m2(vint32m1_t op1, vint32m1_t op2, size_t vl) {
+  return vwmul_vv_i64m2(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmul_vx_i64m2(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vwmul.nxv2i64.nxv2i32.i32.i32(<vscale x 2 x i32> [[OP1:%.*]], i32 [[OP2:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 2 x i64> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmul_vx_i64m2(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vwmul.nxv2i64.nxv2i32.i32.i64(<vscale x 2 x i32> [[OP1:%.*]], i32 [[OP2:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 2 x i64> [[TMP0]]
+//
+vint64m2_t test_vwmul_vx_i64m2(vint32m1_t op1, int32_t op2, size_t vl) {
+  return vwmul_vx_i64m2(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmul_vv_i64m4(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vwmul.nxv4i64.nxv4i32.nxv4i32.i32(<vscale x 4 x i32> [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 4 x i64> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmul_vv_i64m4(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vwmul.nxv4i64.nxv4i32.nxv4i32.i64(<vscale x 4 x i32> [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 4 x i64> [[TMP0]]
+//
+vint64m4_t test_vwmul_vv_i64m4(vint32m2_t op1, vint32m2_t op2, size_t vl) {
+  return vwmul_vv_i64m4(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmul_vx_i64m4(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vwmul.nxv4i64.nxv4i32.i32.i32(<vscale x 4 x i32> [[OP1:%.*]], i32 [[OP2:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 4 x i64> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmul_vx_i64m4(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vwmul.nxv4i64.nxv4i32.i32.i64(<vscale x 4 x i32> [[OP1:%.*]], i32 [[OP2:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 4 x i64> [[TMP0]]
+//
+vint64m4_t test_vwmul_vx_i64m4(vint32m2_t op1, int32_t op2, size_t vl) {
+  return vwmul_vx_i64m4(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmul_vv_i64m8(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vwmul.nxv8i64.nxv8i32.nxv8i32.i32(<vscale x 8 x i32> [[OP1:%.*]], <vscale x 8 x i32> [[OP2:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 8 x i64> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmul_vv_i64m8(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vwmul.nxv8i64.nxv8i32.nxv8i32.i64(<vscale x 8 x i32> [[OP1:%.*]], <vscale x 8 x i32> [[OP2:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 8 x i64> [[TMP0]]
+//
+vint64m8_t test_vwmul_vv_i64m8(vint32m4_t op1, vint32m4_t op2, size_t vl) {
+  return vwmul_vv_i64m8(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmul_vx_i64m8(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vwmul.nxv8i64.nxv8i32.i32.i32(<vscale x 8 x i32> [[OP1:%.*]], i32 [[OP2:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 8 x i64> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmul_vx_i64m8(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vwmul.nxv8i64.nxv8i32.i32.i64(<vscale x 8 x i32> [[OP1:%.*]], i32 [[OP2:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 8 x i64> [[TMP0]]
+//
+vint64m8_t test_vwmul_vx_i64m8(vint32m4_t op1, int32_t op2, size_t vl) {
+  return vwmul_vx_i64m8(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulu_vv_u16mf4(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vwmulu.nxv1i16.nxv1i8.nxv1i8.i32(<vscale x 1 x i8> [[OP1:%.*]], <vscale x 1 x i8> [[OP2:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 1 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulu_vv_u16mf4(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vwmulu.nxv1i16.nxv1i8.nxv1i8.i64(<vscale x 1 x i8> [[OP1:%.*]], <vscale x 1 x i8> [[OP2:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 1 x i16> [[TMP0]]
+//
+vuint16mf4_t test_vwmulu_vv_u16mf4(vuint8mf8_t op1, vuint8mf8_t op2, size_t vl) {
+  return vwmulu_vv_u16mf4(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulu_vx_u16mf4(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vwmulu.nxv1i16.nxv1i8.i8.i32(<vscale x 1 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 1 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulu_vx_u16mf4(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vwmulu.nxv1i16.nxv1i8.i8.i64(<vscale x 1 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 1 x i16> [[TMP0]]
+//
+vuint16mf4_t test_vwmulu_vx_u16mf4(vuint8mf8_t op1, uint8_t op2, size_t vl) {
+  return vwmulu_vx_u16mf4(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulu_vv_u16mf2(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vwmulu.nxv2i16.nxv2i8.nxv2i8.i32(<vscale x 2 x i8> [[OP1:%.*]], <vscale x 2 x i8> [[OP2:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 2 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulu_vv_u16mf2(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vwmulu.nxv2i16.nxv2i8.nxv2i8.i64(<vscale x 2 x i8> [[OP1:%.*]], <vscale x 2 x i8> [[OP2:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 2 x i16> [[TMP0]]
+//
+vuint16mf2_t test_vwmulu_vv_u16mf2(vuint8mf4_t op1, vuint8mf4_t op2, size_t vl) {
+  return vwmulu_vv_u16mf2(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulu_vx_u16mf2(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vwmulu.nxv2i16.nxv2i8.i8.i32(<vscale x 2 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 2 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulu_vx_u16mf2(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vwmulu.nxv2i16.nxv2i8.i8.i64(<vscale x 2 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 2 x i16> [[TMP0]]
+//
+vuint16mf2_t test_vwmulu_vx_u16mf2(vuint8mf4_t op1, uint8_t op2, size_t vl) {
+  return vwmulu_vx_u16mf2(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulu_vv_u16m1(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vwmulu.nxv4i16.nxv4i8.nxv4i8.i32(<vscale x 4 x i8> [[OP1:%.*]], <vscale x 4 x i8> [[OP2:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 4 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulu_vv_u16m1(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vwmulu.nxv4i16.nxv4i8.nxv4i8.i64(<vscale x 4 x i8> [[OP1:%.*]], <vscale x 4 x i8> [[OP2:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 4 x i16> [[TMP0]]
+//
+vuint16m1_t test_vwmulu_vv_u16m1(vuint8mf2_t op1, vuint8mf2_t op2, size_t vl) {
+  return vwmulu_vv_u16m1(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulu_vx_u16m1(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vwmulu.nxv4i16.nxv4i8.i8.i32(<vscale x 4 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 4 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulu_vx_u16m1(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vwmulu.nxv4i16.nxv4i8.i8.i64(<vscale x 4 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 4 x i16> [[TMP0]]
+//
+vuint16m1_t test_vwmulu_vx_u16m1(vuint8mf2_t op1, uint8_t op2, size_t vl) {
+  return vwmulu_vx_u16m1(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulu_vv_u16m2(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vwmulu.nxv8i16.nxv8i8.nxv8i8.i32(<vscale x 8 x i8> [[OP1:%.*]], <vscale x 8 x i8> [[OP2:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulu_vv_u16m2(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vwmulu.nxv8i16.nxv8i8.nxv8i8.i64(<vscale x 8 x i8> [[OP1:%.*]], <vscale x 8 x i8> [[OP2:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
+//
+vuint16m2_t test_vwmulu_vv_u16m2(vuint8m1_t op1, vuint8m1_t op2, size_t vl) {
+  return vwmulu_vv_u16m2(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulu_vx_u16m2(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vwmulu.nxv8i16.nxv8i8.i8.i32(<vscale x 8 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulu_vx_u16m2(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vwmulu.nxv8i16.nxv8i8.i8.i64(<vscale x 8 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
+//
+vuint16m2_t test_vwmulu_vx_u16m2(vuint8m1_t op1, uint8_t op2, size_t vl) {
+  return vwmulu_vx_u16m2(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulu_vv_u16m4(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vwmulu.nxv16i16.nxv16i8.nxv16i8.i32(<vscale x 16 x i8> [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 16 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulu_vv_u16m4(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vwmulu.nxv16i16.nxv16i8.nxv16i8.i64(<vscale x 16 x i8> [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 16 x i16> [[TMP0]]
+//
+vuint16m4_t test_vwmulu_vv_u16m4(vuint8m2_t op1, vuint8m2_t op2, size_t vl) {
+  return vwmulu_vv_u16m4(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulu_vx_u16m4(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vwmulu.nxv16i16.nxv16i8.i8.i32(<vscale x 16 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 16 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulu_vx_u16m4(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vwmulu.nxv16i16.nxv16i8.i8.i64(<vscale x 16 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 16 x i16> [[TMP0]]
+//
+vuint16m4_t test_vwmulu_vx_u16m4(vuint8m2_t op1, uint8_t op2, size_t vl) {
+  return vwmulu_vx_u16m4(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulu_vv_u16m8(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vwmulu.nxv32i16.nxv32i8.nxv32i8.i32(<vscale x 32 x i8> [[OP1:%.*]], <vscale x 32 x i8> [[OP2:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 32 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulu_vv_u16m8(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vwmulu.nxv32i16.nxv32i8.nxv32i8.i64(<vscale x 32 x i8> [[OP1:%.*]], <vscale x 32 x i8> [[OP2:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 32 x i16> [[TMP0]]
+//
+vuint16m8_t test_vwmulu_vv_u16m8(vuint8m4_t op1, vuint8m4_t op2, size_t vl) {
+  return vwmulu_vv_u16m8(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulu_vx_u16m8(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vwmulu.nxv32i16.nxv32i8.i8.i32(<vscale x 32 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 32 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulu_vx_u16m8(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vwmulu.nxv32i16.nxv32i8.i8.i64(<vscale x 32 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 32 x i16> [[TMP0]]
+//
+vuint16m8_t test_vwmulu_vx_u16m8(vuint8m4_t op1, uint8_t op2, size_t vl) {
+  return vwmulu_vx_u16m8(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulu_vv_u32mf2(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vwmulu.nxv1i32.nxv1i16.nxv1i16.i32(<vscale x 1 x i16> [[OP1:%.*]], <vscale x 1 x i16> [[OP2:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulu_vv_u32mf2(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vwmulu.nxv1i32.nxv1i16.nxv1i16.i64(<vscale x 1 x i16> [[OP1:%.*]], <vscale x 1 x i16> [[OP2:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
+//
+vuint32mf2_t test_vwmulu_vv_u32mf2(vuint16mf4_t op1, vuint16mf4_t op2, size_t vl) {
+  return vwmulu_vv_u32mf2(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulu_vx_u32mf2(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vwmulu.nxv1i32.nxv1i16.i16.i32(<vscale x 1 x i16> [[OP1:%.*]], i16 [[OP2:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulu_vx_u32mf2(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vwmulu.nxv1i32.nxv1i16.i16.i64(<vscale x 1 x i16> [[OP1:%.*]], i16 [[OP2:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
+//
+vuint32mf2_t test_vwmulu_vx_u32mf2(vuint16mf4_t op1, uint16_t op2, size_t vl) {
+  return vwmulu_vx_u32mf2(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulu_vv_u32m1(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vwmulu.nxv2i32.nxv2i16.nxv2i16.i32(<vscale x 2 x i16> [[OP1:%.*]], <vscale x 2 x i16> [[OP2:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulu_vv_u32m1(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vwmulu.nxv2i32.nxv2i16.nxv2i16.i64(<vscale x 2 x i16> [[OP1:%.*]], <vscale x 2 x i16> [[OP2:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
+//
+vuint32m1_t test_vwmulu_vv_u32m1(vuint16mf2_t op1, vuint16mf2_t op2, size_t vl) {
+  return vwmulu_vv_u32m1(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulu_vx_u32m1(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vwmulu.nxv2i32.nxv2i16.i16.i32(<vscale x 2 x i16> [[OP1:%.*]], i16 [[OP2:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulu_vx_u32m1(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vwmulu.nxv2i32.nxv2i16.i16.i64(<vscale x 2 x i16> [[OP1:%.*]], i16 [[OP2:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
+//
+vuint32m1_t test_vwmulu_vx_u32m1(vuint16mf2_t op1, uint16_t op2, size_t vl) {
+  return vwmulu_vx_u32m1(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulu_vv_u32m2(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vwmulu.nxv4i32.nxv4i16.nxv4i16.i32(<vscale x 4 x i16> [[OP1:%.*]], <vscale x 4 x i16> [[OP2:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulu_vv_u32m2(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vwmulu.nxv4i32.nxv4i16.nxv4i16.i64(<vscale x 4 x i16> [[OP1:%.*]], <vscale x 4 x i16> [[OP2:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
+//
+vuint32m2_t test_vwmulu_vv_u32m2(vuint16m1_t op1, vuint16m1_t op2, size_t vl) {
+  return vwmulu_vv_u32m2(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulu_vx_u32m2(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vwmulu.nxv4i32.nxv4i16.i16.i32(<vscale x 4 x i16> [[OP1:%.*]], i16 [[OP2:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulu_vx_u32m2(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vwmulu.nxv4i32.nxv4i16.i16.i64(<vscale x 4 x i16> [[OP1:%.*]], i16 [[OP2:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
+//
+vuint32m2_t test_vwmulu_vx_u32m2(vuint16m1_t op1, uint16_t op2, size_t vl) {
+  return vwmulu_vx_u32m2(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulu_vv_u32m4(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vwmulu.nxv8i32.nxv8i16.nxv8i16.i32(<vscale x 8 x i16> [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulu_vv_u32m4(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vwmulu.nxv8i32.nxv8i16.nxv8i16.i64(<vscale x 8 x i16> [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
+//
+vuint32m4_t test_vwmulu_vv_u32m4(vuint16m2_t op1, vuint16m2_t op2, size_t vl) {
+  return vwmulu_vv_u32m4(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulu_vx_u32m4(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vwmulu.nxv8i32.nxv8i16.i16.i32(<vscale x 8 x i16> [[OP1:%.*]], i16 [[OP2:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulu_vx_u32m4(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vwmulu.nxv8i32.nxv8i16.i16.i64(<vscale x 8 x i16> [[OP1:%.*]], i16 [[OP2:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
+//
+vuint32m4_t test_vwmulu_vx_u32m4(vuint16m2_t op1, uint16_t op2, size_t vl) {
+  return vwmulu_vx_u32m4(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulu_vv_u32m8(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vwmulu.nxv16i32.nxv16i16.nxv16i16.i32(<vscale x 16 x i16> [[OP1:%.*]], <vscale x 16 x i16> [[OP2:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulu_vv_u32m8(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vwmulu.nxv16i32.nxv16i16.nxv16i16.i64(<vscale x 16 x i16> [[OP1:%.*]], <vscale x 16 x i16> [[OP2:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
+//
+vuint32m8_t test_vwmulu_vv_u32m8(vuint16m4_t op1, vuint16m4_t op2, size_t vl) {
+  return vwmulu_vv_u32m8(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulu_vx_u32m8(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vwmulu.nxv16i32.nxv16i16.i16.i32(<vscale x 16 x i16> [[OP1:%.*]], i16 [[OP2:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulu_vx_u32m8(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vwmulu.nxv16i32.nxv16i16.i16.i64(<vscale x 16 x i16> [[OP1:%.*]], i16 [[OP2:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
+//
+vuint32m8_t test_vwmulu_vx_u32m8(vuint16m4_t op1, uint16_t op2, size_t vl) {
+  return vwmulu_vx_u32m8(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulu_vv_u64m1(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vwmulu.nxv1i64.nxv1i32.nxv1i32.i32(<vscale x 1 x i32> [[OP1:%.*]], <vscale x 1 x i32> [[OP2:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 1 x i64> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulu_vv_u64m1(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vwmulu.nxv1i64.nxv1i32.nxv1i32.i64(<vscale x 1 x i32> [[OP1:%.*]], <vscale x 1 x i32> [[OP2:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 1 x i64> [[TMP0]]
+//
+vuint64m1_t test_vwmulu_vv_u64m1(vuint32mf2_t op1, vuint32mf2_t op2, size_t vl) {
+  return vwmulu_vv_u64m1(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulu_vx_u64m1(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vwmulu.nxv1i64.nxv1i32.i32.i32(<vscale x 1 x i32> [[OP1:%.*]], i32 [[OP2:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 1 x i64> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulu_vx_u64m1(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vwmulu.nxv1i64.nxv1i32.i32.i64(<vscale x 1 x i32> [[OP1:%.*]], i32 [[OP2:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 1 x i64> [[TMP0]]
+//
+vuint64m1_t test_vwmulu_vx_u64m1(vuint32mf2_t op1, uint32_t op2, size_t vl) {
+  return vwmulu_vx_u64m1(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulu_vv_u64m2(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vwmulu.nxv2i64.nxv2i32.nxv2i32.i32(<vscale x 2 x i32> [[OP1:%.*]], <vscale x 2 x i32> [[OP2:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 2 x i64> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulu_vv_u64m2(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vwmulu.nxv2i64.nxv2i32.nxv2i32.i64(<vscale x 2 x i32> [[OP1:%.*]], <vscale x 2 x i32> [[OP2:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 2 x i64> [[TMP0]]
+//
+vuint64m2_t test_vwmulu_vv_u64m2(vuint32m1_t op1, vuint32m1_t op2, size_t vl) {
+  return vwmulu_vv_u64m2(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulu_vx_u64m2(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vwmulu.nxv2i64.nxv2i32.i32.i32(<vscale x 2 x i32> [[OP1:%.*]], i32 [[OP2:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 2 x i64> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulu_vx_u64m2(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vwmulu.nxv2i64.nxv2i32.i32.i64(<vscale x 2 x i32> [[OP1:%.*]], i32 [[OP2:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 2 x i64> [[TMP0]]
+//
+vuint64m2_t test_vwmulu_vx_u64m2(vuint32m1_t op1, uint32_t op2, size_t vl) {
+  return vwmulu_vx_u64m2(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulu_vv_u64m4(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vwmulu.nxv4i64.nxv4i32.nxv4i32.i32(<vscale x 4 x i32> [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 4 x i64> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulu_vv_u64m4(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vwmulu.nxv4i64.nxv4i32.nxv4i32.i64(<vscale x 4 x i32> [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 4 x i64> [[TMP0]]
+//
+vuint64m4_t test_vwmulu_vv_u64m4(vuint32m2_t op1, vuint32m2_t op2, size_t vl) {
+  return vwmulu_vv_u64m4(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulu_vx_u64m4(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vwmulu.nxv4i64.nxv4i32.i32.i32(<vscale x 4 x i32> [[OP1:%.*]], i32 [[OP2:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 4 x i64> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulu_vx_u64m4(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vwmulu.nxv4i64.nxv4i32.i32.i64(<vscale x 4 x i32> [[OP1:%.*]], i32 [[OP2:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 4 x i64> [[TMP0]]
+//
+vuint64m4_t test_vwmulu_vx_u64m4(vuint32m2_t op1, uint32_t op2, size_t vl) {
+  return vwmulu_vx_u64m4(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulu_vv_u64m8(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vwmulu.nxv8i64.nxv8i32.nxv8i32.i32(<vscale x 8 x i32> [[OP1:%.*]], <vscale x 8 x i32> [[OP2:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 8 x i64> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulu_vv_u64m8(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vwmulu.nxv8i64.nxv8i32.nxv8i32.i64(<vscale x 8 x i32> [[OP1:%.*]], <vscale x 8 x i32> [[OP2:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 8 x i64> [[TMP0]]
+//
+vuint64m8_t test_vwmulu_vv_u64m8(vuint32m4_t op1, vuint32m4_t op2, size_t vl) {
+  return vwmulu_vv_u64m8(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulu_vx_u64m8(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vwmulu.nxv8i64.nxv8i32.i32.i32(<vscale x 8 x i32> [[OP1:%.*]], i32 [[OP2:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 8 x i64> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulu_vx_u64m8(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vwmulu.nxv8i64.nxv8i32.i32.i64(<vscale x 8 x i32> [[OP1:%.*]], i32 [[OP2:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 8 x i64> [[TMP0]]
+//
+vuint64m8_t test_vwmulu_vx_u64m8(vuint32m4_t op1, uint32_t op2, size_t vl) {
+  return vwmulu_vx_u64m8(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulsu_vv_i16mf4(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vwmulsu.nxv1i16.nxv1i8.nxv1i8.i32(<vscale x 1 x i8> [[OP1:%.*]], <vscale x 1 x i8> [[OP2:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 1 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulsu_vv_i16mf4(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vwmulsu.nxv1i16.nxv1i8.nxv1i8.i64(<vscale x 1 x i8> [[OP1:%.*]], <vscale x 1 x i8> [[OP2:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 1 x i16> [[TMP0]]
+//
+vint16mf4_t test_vwmulsu_vv_i16mf4(vint8mf8_t op1, vuint8mf8_t op2, size_t vl) {
+  return vwmulsu_vv_i16mf4(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulsu_vx_i16mf4(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vwmulsu.nxv1i16.nxv1i8.i8.i32(<vscale x 1 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 1 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulsu_vx_i16mf4(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vwmulsu.nxv1i16.nxv1i8.i8.i64(<vscale x 1 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 1 x i16> [[TMP0]]
+//
+vint16mf4_t test_vwmulsu_vx_i16mf4(vint8mf8_t op1, uint8_t op2, size_t vl) {
+  return vwmulsu_vx_i16mf4(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulsu_vv_i16mf2(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vwmulsu.nxv2i16.nxv2i8.nxv2i8.i32(<vscale x 2 x i8> [[OP1:%.*]], <vscale x 2 x i8> [[OP2:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 2 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulsu_vv_i16mf2(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vwmulsu.nxv2i16.nxv2i8.nxv2i8.i64(<vscale x 2 x i8> [[OP1:%.*]], <vscale x 2 x i8> [[OP2:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 2 x i16> [[TMP0]]
+//
+vint16mf2_t test_vwmulsu_vv_i16mf2(vint8mf4_t op1, vuint8mf4_t op2, size_t vl) {
+  return vwmulsu_vv_i16mf2(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulsu_vx_i16mf2(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vwmulsu.nxv2i16.nxv2i8.i8.i32(<vscale x 2 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 2 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulsu_vx_i16mf2(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vwmulsu.nxv2i16.nxv2i8.i8.i64(<vscale x 2 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 2 x i16> [[TMP0]]
+//
+vint16mf2_t test_vwmulsu_vx_i16mf2(vint8mf4_t op1, uint8_t op2, size_t vl) {
+  return vwmulsu_vx_i16mf2(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulsu_vv_i16m1(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vwmulsu.nxv4i16.nxv4i8.nxv4i8.i32(<vscale x 4 x i8> [[OP1:%.*]], <vscale x 4 x i8> [[OP2:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 4 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulsu_vv_i16m1(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vwmulsu.nxv4i16.nxv4i8.nxv4i8.i64(<vscale x 4 x i8> [[OP1:%.*]], <vscale x 4 x i8> [[OP2:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 4 x i16> [[TMP0]]
+//
+vint16m1_t test_vwmulsu_vv_i16m1(vint8mf2_t op1, vuint8mf2_t op2, size_t vl) {
+  return vwmulsu_vv_i16m1(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulsu_vx_i16m1(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vwmulsu.nxv4i16.nxv4i8.i8.i32(<vscale x 4 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 4 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulsu_vx_i16m1(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vwmulsu.nxv4i16.nxv4i8.i8.i64(<vscale x 4 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 4 x i16> [[TMP0]]
+//
+vint16m1_t test_vwmulsu_vx_i16m1(vint8mf2_t op1, uint8_t op2, size_t vl) {
+  return vwmulsu_vx_i16m1(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulsu_vv_i16m2(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vwmulsu.nxv8i16.nxv8i8.nxv8i8.i32(<vscale x 8 x i8> [[OP1:%.*]], <vscale x 8 x i8> [[OP2:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulsu_vv_i16m2(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vwmulsu.nxv8i16.nxv8i8.nxv8i8.i64(<vscale x 8 x i8> [[OP1:%.*]], <vscale x 8 x i8> [[OP2:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
+//
+vint16m2_t test_vwmulsu_vv_i16m2(vint8m1_t op1, vuint8m1_t op2, size_t vl) {
+  return vwmulsu_vv_i16m2(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulsu_vx_i16m2(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vwmulsu.nxv8i16.nxv8i8.i8.i32(<vscale x 8 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulsu_vx_i16m2(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vwmulsu.nxv8i16.nxv8i8.i8.i64(<vscale x 8 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
+//
+vint16m2_t test_vwmulsu_vx_i16m2(vint8m1_t op1, uint8_t op2, size_t vl) {
+  return vwmulsu_vx_i16m2(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulsu_vv_i16m4(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vwmulsu.nxv16i16.nxv16i8.nxv16i8.i32(<vscale x 16 x i8> [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 16 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulsu_vv_i16m4(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vwmulsu.nxv16i16.nxv16i8.nxv16i8.i64(<vscale x 16 x i8> [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 16 x i16> [[TMP0]]
+//
+vint16m4_t test_vwmulsu_vv_i16m4(vint8m2_t op1, vuint8m2_t op2, size_t vl) {
+  return vwmulsu_vv_i16m4(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulsu_vx_i16m4(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vwmulsu.nxv16i16.nxv16i8.i8.i32(<vscale x 16 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 16 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulsu_vx_i16m4(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vwmulsu.nxv16i16.nxv16i8.i8.i64(<vscale x 16 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 16 x i16> [[TMP0]]
+//
+vint16m4_t test_vwmulsu_vx_i16m4(vint8m2_t op1, uint8_t op2, size_t vl) {
+  return vwmulsu_vx_i16m4(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulsu_vv_i16m8(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vwmulsu.nxv32i16.nxv32i8.nxv32i8.i32(<vscale x 32 x i8> [[OP1:%.*]], <vscale x 32 x i8> [[OP2:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 32 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulsu_vv_i16m8(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vwmulsu.nxv32i16.nxv32i8.nxv32i8.i64(<vscale x 32 x i8> [[OP1:%.*]], <vscale x 32 x i8> [[OP2:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 32 x i16> [[TMP0]]
+//
+vint16m8_t test_vwmulsu_vv_i16m8(vint8m4_t op1, vuint8m4_t op2, size_t vl) {
+  return vwmulsu_vv_i16m8(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulsu_vx_i16m8(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vwmulsu.nxv32i16.nxv32i8.i8.i32(<vscale x 32 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 32 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulsu_vx_i16m8(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vwmulsu.nxv32i16.nxv32i8.i8.i64(<vscale x 32 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 32 x i16> [[TMP0]]
+//
+vint16m8_t test_vwmulsu_vx_i16m8(vint8m4_t op1, uint8_t op2, size_t vl) {
+  return vwmulsu_vx_i16m8(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulsu_vv_i32mf2(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vwmulsu.nxv1i32.nxv1i16.nxv1i16.i32(<vscale x 1 x i16> [[OP1:%.*]], <vscale x 1 x i16> [[OP2:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulsu_vv_i32mf2(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vwmulsu.nxv1i32.nxv1i16.nxv1i16.i64(<vscale x 1 x i16> [[OP1:%.*]], <vscale x 1 x i16> [[OP2:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
+//
+vint32mf2_t test_vwmulsu_vv_i32mf2(vint16mf4_t op1, vuint16mf4_t op2, size_t vl) {
+  return vwmulsu_vv_i32mf2(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulsu_vx_i32mf2(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vwmulsu.nxv1i32.nxv1i16.i16.i32(<vscale x 1 x i16> [[OP1:%.*]], i16 [[OP2:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulsu_vx_i32mf2(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vwmulsu.nxv1i32.nxv1i16.i16.i64(<vscale x 1 x i16> [[OP1:%.*]], i16 [[OP2:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
+//
+vint32mf2_t test_vwmulsu_vx_i32mf2(vint16mf4_t op1, uint16_t op2, size_t vl) {
+  return vwmulsu_vx_i32mf2(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulsu_vv_i32m1(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vwmulsu.nxv2i32.nxv2i16.nxv2i16.i32(<vscale x 2 x i16> [[OP1:%.*]], <vscale x 2 x i16> [[OP2:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulsu_vv_i32m1(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vwmulsu.nxv2i32.nxv2i16.nxv2i16.i64(<vscale x 2 x i16> [[OP1:%.*]], <vscale x 2 x i16> [[OP2:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
+//
+vint32m1_t test_vwmulsu_vv_i32m1(vint16mf2_t op1, vuint16mf2_t op2, size_t vl) {
+  return vwmulsu_vv_i32m1(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulsu_vx_i32m1(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vwmulsu.nxv2i32.nxv2i16.i16.i32(<vscale x 2 x i16> [[OP1:%.*]], i16 [[OP2:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulsu_vx_i32m1(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vwmulsu.nxv2i32.nxv2i16.i16.i64(<vscale x 2 x i16> [[OP1:%.*]], i16 [[OP2:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
+//
+vint32m1_t test_vwmulsu_vx_i32m1(vint16mf2_t op1, uint16_t op2, size_t vl) {
+  return vwmulsu_vx_i32m1(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulsu_vv_i32m2(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vwmulsu.nxv4i32.nxv4i16.nxv4i16.i32(<vscale x 4 x i16> [[OP1:%.*]], <vscale x 4 x i16> [[OP2:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulsu_vv_i32m2(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vwmulsu.nxv4i32.nxv4i16.nxv4i16.i64(<vscale x 4 x i16> [[OP1:%.*]], <vscale x 4 x i16> [[OP2:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
+//
+vint32m2_t test_vwmulsu_vv_i32m2(vint16m1_t op1, vuint16m1_t op2, size_t vl) {
+  return vwmulsu_vv_i32m2(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulsu_vx_i32m2(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vwmulsu.nxv4i32.nxv4i16.i16.i32(<vscale x 4 x i16> [[OP1:%.*]], i16 [[OP2:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulsu_vx_i32m2(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vwmulsu.nxv4i32.nxv4i16.i16.i64(<vscale x 4 x i16> [[OP1:%.*]], i16 [[OP2:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
+//
+vint32m2_t test_vwmulsu_vx_i32m2(vint16m1_t op1, uint16_t op2, size_t vl) {
+  return vwmulsu_vx_i32m2(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulsu_vv_i32m4(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vwmulsu.nxv8i32.nxv8i16.nxv8i16.i32(<vscale x 8 x i16> [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulsu_vv_i32m4(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vwmulsu.nxv8i32.nxv8i16.nxv8i16.i64(<vscale x 8 x i16> [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
+//
+vint32m4_t test_vwmulsu_vv_i32m4(vint16m2_t op1, vuint16m2_t op2, size_t vl) {
+  return vwmulsu_vv_i32m4(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulsu_vx_i32m4(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vwmulsu.nxv8i32.nxv8i16.i16.i32(<vscale x 8 x i16> [[OP1:%.*]], i16 [[OP2:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulsu_vx_i32m4(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vwmulsu.nxv8i32.nxv8i16.i16.i64(<vscale x 8 x i16> [[OP1:%.*]], i16 [[OP2:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
+//
+vint32m4_t test_vwmulsu_vx_i32m4(vint16m2_t op1, uint16_t op2, size_t vl) {
+  return vwmulsu_vx_i32m4(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulsu_vv_i32m8(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vwmulsu.nxv16i32.nxv16i16.nxv16i16.i32(<vscale x 16 x i16> [[OP1:%.*]], <vscale x 16 x i16> [[OP2:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulsu_vv_i32m8(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vwmulsu.nxv16i32.nxv16i16.nxv16i16.i64(<vscale x 16 x i16> [[OP1:%.*]], <vscale x 16 x i16> [[OP2:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
+//
+vint32m8_t test_vwmulsu_vv_i32m8(vint16m4_t op1, vuint16m4_t op2, size_t vl) {
+  return vwmulsu_vv_i32m8(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulsu_vx_i32m8(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vwmulsu.nxv16i32.nxv16i16.i16.i32(<vscale x 16 x i16> [[OP1:%.*]], i16 [[OP2:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulsu_vx_i32m8(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vwmulsu.nxv16i32.nxv16i16.i16.i64(<vscale x 16 x i16> [[OP1:%.*]], i16 [[OP2:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
+//
+vint32m8_t test_vwmulsu_vx_i32m8(vint16m4_t op1, uint16_t op2, size_t vl) {
+  return vwmulsu_vx_i32m8(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulsu_vv_i64m1(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vwmulsu.nxv1i64.nxv1i32.nxv1i32.i32(<vscale x 1 x i32> [[OP1:%.*]], <vscale x 1 x i32> [[OP2:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 1 x i64> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulsu_vv_i64m1(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vwmulsu.nxv1i64.nxv1i32.nxv1i32.i64(<vscale x 1 x i32> [[OP1:%.*]], <vscale x 1 x i32> [[OP2:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 1 x i64> [[TMP0]]
+//
+vint64m1_t test_vwmulsu_vv_i64m1(vint32mf2_t op1, vuint32mf2_t op2, size_t vl) {
+  return vwmulsu_vv_i64m1(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulsu_vx_i64m1(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vwmulsu.nxv1i64.nxv1i32.i32.i32(<vscale x 1 x i32> [[OP1:%.*]], i32 [[OP2:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 1 x i64> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulsu_vx_i64m1(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vwmulsu.nxv1i64.nxv1i32.i32.i64(<vscale x 1 x i32> [[OP1:%.*]], i32 [[OP2:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 1 x i64> [[TMP0]]
+//
+vint64m1_t test_vwmulsu_vx_i64m1(vint32mf2_t op1, uint32_t op2, size_t vl) {
+  return vwmulsu_vx_i64m1(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulsu_vv_i64m2(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vwmulsu.nxv2i64.nxv2i32.nxv2i32.i32(<vscale x 2 x i32> [[OP1:%.*]], <vscale x 2 x i32> [[OP2:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 2 x i64> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulsu_vv_i64m2(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vwmulsu.nxv2i64.nxv2i32.nxv2i32.i64(<vscale x 2 x i32> [[OP1:%.*]], <vscale x 2 x i32> [[OP2:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 2 x i64> [[TMP0]]
+//
+vint64m2_t test_vwmulsu_vv_i64m2(vint32m1_t op1, vuint32m1_t op2, size_t vl) {
+  return vwmulsu_vv_i64m2(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulsu_vx_i64m2(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vwmulsu.nxv2i64.nxv2i32.i32.i32(<vscale x 2 x i32> [[OP1:%.*]], i32 [[OP2:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 2 x i64> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulsu_vx_i64m2(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vwmulsu.nxv2i64.nxv2i32.i32.i64(<vscale x 2 x i32> [[OP1:%.*]], i32 [[OP2:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 2 x i64> [[TMP0]]
+//
+vint64m2_t test_vwmulsu_vx_i64m2(vint32m1_t op1, uint32_t op2, size_t vl) {
+  return vwmulsu_vx_i64m2(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulsu_vv_i64m4(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vwmulsu.nxv4i64.nxv4i32.nxv4i32.i32(<vscale x 4 x i32> [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 4 x i64> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulsu_vv_i64m4(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vwmulsu.nxv4i64.nxv4i32.nxv4i32.i64(<vscale x 4 x i32> [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 4 x i64> [[TMP0]]
+//
+vint64m4_t test_vwmulsu_vv_i64m4(vint32m2_t op1, vuint32m2_t op2, size_t vl) {
+  return vwmulsu_vv_i64m4(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulsu_vx_i64m4(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vwmulsu.nxv4i64.nxv4i32.i32.i32(<vscale x 4 x i32> [[OP1:%.*]], i32 [[OP2:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 4 x i64> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulsu_vx_i64m4(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vwmulsu.nxv4i64.nxv4i32.i32.i64(<vscale x 4 x i32> [[OP1:%.*]], i32 [[OP2:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 4 x i64> [[TMP0]]
+//
+vint64m4_t test_vwmulsu_vx_i64m4(vint32m2_t op1, uint32_t op2, size_t vl) {
+  return vwmulsu_vx_i64m4(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulsu_vv_i64m8(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vwmulsu.nxv8i64.nxv8i32.nxv8i32.i32(<vscale x 8 x i32> [[OP1:%.*]], <vscale x 8 x i32> [[OP2:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 8 x i64> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulsu_vv_i64m8(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vwmulsu.nxv8i64.nxv8i32.nxv8i32.i64(<vscale x 8 x i32> [[OP1:%.*]], <vscale x 8 x i32> [[OP2:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 8 x i64> [[TMP0]]
+//
+vint64m8_t test_vwmulsu_vv_i64m8(vint32m4_t op1, vuint32m4_t op2, size_t vl) {
+  return vwmulsu_vv_i64m8(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulsu_vx_i64m8(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vwmulsu.nxv8i64.nxv8i32.i32.i32(<vscale x 8 x i32> [[OP1:%.*]], i32 [[OP2:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 8 x i64> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulsu_vx_i64m8(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vwmulsu.nxv8i64.nxv8i32.i32.i64(<vscale x 8 x i32> [[OP1:%.*]], i32 [[OP2:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 8 x i64> [[TMP0]]
+//
+vint64m8_t test_vwmulsu_vx_i64m8(vint32m4_t op1, uint32_t op2, size_t vl) {
+  return vwmulsu_vx_i64m8(op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmul_vv_i16mf4_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vwmul.mask.nxv1i16.nxv1i8.nxv1i8.i32(<vscale x 1 x i16> [[MASKEDOFF:%.*]], <vscale x 1 x i8> [[OP1:%.*]], <vscale x 1 x i8> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 1 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmul_vv_i16mf4_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vwmul.mask.nxv1i16.nxv1i8.nxv1i8.i64(<vscale x 1 x i16> [[MASKEDOFF:%.*]], <vscale x 1 x i8> [[OP1:%.*]], <vscale x 1 x i8> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 1 x i16> [[TMP0]]
+//
+vint16mf4_t test_vwmul_vv_i16mf4_m(vbool64_t mask, vint16mf4_t maskedoff, vint8mf8_t op1, vint8mf8_t op2, size_t vl) {
+  return vwmul_vv_i16mf4_m(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmul_vx_i16mf4_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vwmul.mask.nxv1i16.nxv1i8.i8.i32(<vscale x 1 x i16> [[MASKEDOFF:%.*]], <vscale x 1 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 1 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmul_vx_i16mf4_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vwmul.mask.nxv1i16.nxv1i8.i8.i64(<vscale x 1 x i16> [[MASKEDOFF:%.*]], <vscale x 1 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 1 x i16> [[TMP0]]
+//
+vint16mf4_t test_vwmul_vx_i16mf4_m(vbool64_t mask, vint16mf4_t maskedoff, vint8mf8_t op1, int8_t op2, size_t vl) {
+  return vwmul_vx_i16mf4_m(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmul_vv_i16mf2_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vwmul.mask.nxv2i16.nxv2i8.nxv2i8.i32(<vscale x 2 x i16> [[MASKEDOFF:%.*]], <vscale x 2 x i8> [[OP1:%.*]], <vscale x 2 x i8> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 2 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmul_vv_i16mf2_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vwmul.mask.nxv2i16.nxv2i8.nxv2i8.i64(<vscale x 2 x i16> [[MASKEDOFF:%.*]], <vscale x 2 x i8> [[OP1:%.*]], <vscale x 2 x i8> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 2 x i16> [[TMP0]]
+//
+vint16mf2_t test_vwmul_vv_i16mf2_m(vbool32_t mask, vint16mf2_t maskedoff, vint8mf4_t op1, vint8mf4_t op2, size_t vl) {
+  return vwmul_vv_i16mf2_m(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmul_vx_i16mf2_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vwmul.mask.nxv2i16.nxv2i8.i8.i32(<vscale x 2 x i16> [[MASKEDOFF:%.*]], <vscale x 2 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 2 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmul_vx_i16mf2_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vwmul.mask.nxv2i16.nxv2i8.i8.i64(<vscale x 2 x i16> [[MASKEDOFF:%.*]], <vscale x 2 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 2 x i16> [[TMP0]]
+//
+vint16mf2_t test_vwmul_vx_i16mf2_m(vbool32_t mask, vint16mf2_t maskedoff, vint8mf4_t op1, int8_t op2, size_t vl) {
+  return vwmul_vx_i16mf2_m(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmul_vv_i16m1_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vwmul.mask.nxv4i16.nxv4i8.nxv4i8.i32(<vscale x 4 x i16> [[MASKEDOFF:%.*]], <vscale x 4 x i8> [[OP1:%.*]], <vscale x 4 x i8> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 4 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmul_vv_i16m1_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vwmul.mask.nxv4i16.nxv4i8.nxv4i8.i64(<vscale x 4 x i16> [[MASKEDOFF:%.*]], <vscale x 4 x i8> [[OP1:%.*]], <vscale x 4 x i8> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 4 x i16> [[TMP0]]
+//
+vint16m1_t test_vwmul_vv_i16m1_m(vbool16_t mask, vint16m1_t maskedoff, vint8mf2_t op1, vint8mf2_t op2, size_t vl) {
+  return vwmul_vv_i16m1_m(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmul_vx_i16m1_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vwmul.mask.nxv4i16.nxv4i8.i8.i32(<vscale x 4 x i16> [[MASKEDOFF:%.*]], <vscale x 4 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 4 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmul_vx_i16m1_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vwmul.mask.nxv4i16.nxv4i8.i8.i64(<vscale x 4 x i16> [[MASKEDOFF:%.*]], <vscale x 4 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 4 x i16> [[TMP0]]
+//
+vint16m1_t test_vwmul_vx_i16m1_m(vbool16_t mask, vint16m1_t maskedoff, vint8mf2_t op1, int8_t op2, size_t vl) {
+  return vwmul_vx_i16m1_m(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmul_vv_i16m2_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vwmul.mask.nxv8i16.nxv8i8.nxv8i8.i32(<vscale x 8 x i16> [[MASKEDOFF:%.*]], <vscale x 8 x i8> [[OP1:%.*]], <vscale x 8 x i8> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmul_vv_i16m2_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vwmul.mask.nxv8i16.nxv8i8.nxv8i8.i64(<vscale x 8 x i16> [[MASKEDOFF:%.*]], <vscale x 8 x i8> [[OP1:%.*]], <vscale x 8 x i8> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
+//
+vint16m2_t test_vwmul_vv_i16m2_m(vbool8_t mask, vint16m2_t maskedoff, vint8m1_t op1, vint8m1_t op2, size_t vl) {
+  return vwmul_vv_i16m2_m(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmul_vx_i16m2_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vwmul.mask.nxv8i16.nxv8i8.i8.i32(<vscale x 8 x i16> [[MASKEDOFF:%.*]], <vscale x 8 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmul_vx_i16m2_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vwmul.mask.nxv8i16.nxv8i8.i8.i64(<vscale x 8 x i16> [[MASKEDOFF:%.*]], <vscale x 8 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
+//
+vint16m2_t test_vwmul_vx_i16m2_m(vbool8_t mask, vint16m2_t maskedoff, vint8m1_t op1, int8_t op2, size_t vl) {
+  return vwmul_vx_i16m2_m(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmul_vv_i16m4_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vwmul.mask.nxv16i16.nxv16i8.nxv16i8.i32(<vscale x 16 x i16> [[MASKEDOFF:%.*]], <vscale x 16 x i8> [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 16 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmul_vv_i16m4_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vwmul.mask.nxv16i16.nxv16i8.nxv16i8.i64(<vscale x 16 x i16> [[MASKEDOFF:%.*]], <vscale x 16 x i8> [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 16 x i16> [[TMP0]]
+//
+vint16m4_t test_vwmul_vv_i16m4_m(vbool4_t mask, vint16m4_t maskedoff, vint8m2_t op1, vint8m2_t op2, size_t vl) {
+  return vwmul_vv_i16m4_m(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmul_vx_i16m4_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vwmul.mask.nxv16i16.nxv16i8.i8.i32(<vscale x 16 x i16> [[MASKEDOFF:%.*]], <vscale x 16 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 16 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmul_vx_i16m4_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vwmul.mask.nxv16i16.nxv16i8.i8.i64(<vscale x 16 x i16> [[MASKEDOFF:%.*]], <vscale x 16 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 16 x i16> [[TMP0]]
+//
+vint16m4_t test_vwmul_vx_i16m4_m(vbool4_t mask, vint16m4_t maskedoff, vint8m2_t op1, int8_t op2, size_t vl) {
+  return vwmul_vx_i16m4_m(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmul_vv_i16m8_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vwmul.mask.nxv32i16.nxv32i8.nxv32i8.i32(<vscale x 32 x i16> [[MASKEDOFF:%.*]], <vscale x 32 x i8> [[OP1:%.*]], <vscale x 32 x i8> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 32 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmul_vv_i16m8_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vwmul.mask.nxv32i16.nxv32i8.nxv32i8.i64(<vscale x 32 x i16> [[MASKEDOFF:%.*]], <vscale x 32 x i8> [[OP1:%.*]], <vscale x 32 x i8> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 32 x i16> [[TMP0]]
+//
+vint16m8_t test_vwmul_vv_i16m8_m(vbool2_t mask, vint16m8_t maskedoff, vint8m4_t op1, vint8m4_t op2, size_t vl) {
+  return vwmul_vv_i16m8_m(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmul_vx_i16m8_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vwmul.mask.nxv32i16.nxv32i8.i8.i32(<vscale x 32 x i16> [[MASKEDOFF:%.*]], <vscale x 32 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 32 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmul_vx_i16m8_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vwmul.mask.nxv32i16.nxv32i8.i8.i64(<vscale x 32 x i16> [[MASKEDOFF:%.*]], <vscale x 32 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 32 x i16> [[TMP0]]
+//
+vint16m8_t test_vwmul_vx_i16m8_m(vbool2_t mask, vint16m8_t maskedoff, vint8m4_t op1, int8_t op2, size_t vl) {
+  return vwmul_vx_i16m8_m(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmul_vv_i32mf2_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vwmul.mask.nxv1i32.nxv1i16.nxv1i16.i32(<vscale x 1 x i32> [[MASKEDOFF:%.*]], <vscale x 1 x i16> [[OP1:%.*]], <vscale x 1 x i16> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmul_vv_i32mf2_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vwmul.mask.nxv1i32.nxv1i16.nxv1i16.i64(<vscale x 1 x i32> [[MASKEDOFF:%.*]], <vscale x 1 x i16> [[OP1:%.*]], <vscale x 1 x i16> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
+//
+vint32mf2_t test_vwmul_vv_i32mf2_m(vbool64_t mask, vint32mf2_t maskedoff, vint16mf4_t op1, vint16mf4_t op2, size_t vl) {
+  return vwmul_vv_i32mf2_m(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmul_vx_i32mf2_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vwmul.mask.nxv1i32.nxv1i16.i16.i32(<vscale x 1 x i32> [[MASKEDOFF:%.*]], <vscale x 1 x i16> [[OP1:%.*]], i16 [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmul_vx_i32mf2_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vwmul.mask.nxv1i32.nxv1i16.i16.i64(<vscale x 1 x i32> [[MASKEDOFF:%.*]], <vscale x 1 x i16> [[OP1:%.*]], i16 [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
+//
+vint32mf2_t test_vwmul_vx_i32mf2_m(vbool64_t mask, vint32mf2_t maskedoff, vint16mf4_t op1, int16_t op2, size_t vl) {
+  return vwmul_vx_i32mf2_m(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmul_vv_i32m1_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vwmul.mask.nxv2i32.nxv2i16.nxv2i16.i32(<vscale x 2 x i32> [[MASKEDOFF:%.*]], <vscale x 2 x i16> [[OP1:%.*]], <vscale x 2 x i16> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmul_vv_i32m1_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vwmul.mask.nxv2i32.nxv2i16.nxv2i16.i64(<vscale x 2 x i32> [[MASKEDOFF:%.*]], <vscale x 2 x i16> [[OP1:%.*]], <vscale x 2 x i16> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
+//
+vint32m1_t test_vwmul_vv_i32m1_m(vbool32_t mask, vint32m1_t maskedoff, vint16mf2_t op1, vint16mf2_t op2, size_t vl) {
+  return vwmul_vv_i32m1_m(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmul_vx_i32m1_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vwmul.mask.nxv2i32.nxv2i16.i16.i32(<vscale x 2 x i32> [[MASKEDOFF:%.*]], <vscale x 2 x i16> [[OP1:%.*]], i16 [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmul_vx_i32m1_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vwmul.mask.nxv2i32.nxv2i16.i16.i64(<vscale x 2 x i32> [[MASKEDOFF:%.*]], <vscale x 2 x i16> [[OP1:%.*]], i16 [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
+//
+vint32m1_t test_vwmul_vx_i32m1_m(vbool32_t mask, vint32m1_t maskedoff, vint16mf2_t op1, int16_t op2, size_t vl) {
+  return vwmul_vx_i32m1_m(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmul_vv_i32m2_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vwmul.mask.nxv4i32.nxv4i16.nxv4i16.i32(<vscale x 4 x i32> [[MASKEDOFF:%.*]], <vscale x 4 x i16> [[OP1:%.*]], <vscale x 4 x i16> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmul_vv_i32m2_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vwmul.mask.nxv4i32.nxv4i16.nxv4i16.i64(<vscale x 4 x i32> [[MASKEDOFF:%.*]], <vscale x 4 x i16> [[OP1:%.*]], <vscale x 4 x i16> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
+//
+vint32m2_t test_vwmul_vv_i32m2_m(vbool16_t mask, vint32m2_t maskedoff, vint16m1_t op1, vint16m1_t op2, size_t vl) {
+  return vwmul_vv_i32m2_m(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmul_vx_i32m2_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vwmul.mask.nxv4i32.nxv4i16.i16.i32(<vscale x 4 x i32> [[MASKEDOFF:%.*]], <vscale x 4 x i16> [[OP1:%.*]], i16 [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmul_vx_i32m2_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vwmul.mask.nxv4i32.nxv4i16.i16.i64(<vscale x 4 x i32> [[MASKEDOFF:%.*]], <vscale x 4 x i16> [[OP1:%.*]], i16 [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
+//
+vint32m2_t test_vwmul_vx_i32m2_m(vbool16_t mask, vint32m2_t maskedoff, vint16m1_t op1, int16_t op2, size_t vl) {
+  return vwmul_vx_i32m2_m(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmul_vv_i32m4_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vwmul.mask.nxv8i32.nxv8i16.nxv8i16.i32(<vscale x 8 x i32> [[MASKEDOFF:%.*]], <vscale x 8 x i16> [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmul_vv_i32m4_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vwmul.mask.nxv8i32.nxv8i16.nxv8i16.i64(<vscale x 8 x i32> [[MASKEDOFF:%.*]], <vscale x 8 x i16> [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
+//
+vint32m4_t test_vwmul_vv_i32m4_m(vbool8_t mask, vint32m4_t maskedoff, vint16m2_t op1, vint16m2_t op2, size_t vl) {
+  return vwmul_vv_i32m4_m(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmul_vx_i32m4_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vwmul.mask.nxv8i32.nxv8i16.i16.i32(<vscale x 8 x i32> [[MASKEDOFF:%.*]], <vscale x 8 x i16> [[OP1:%.*]], i16 [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmul_vx_i32m4_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vwmul.mask.nxv8i32.nxv8i16.i16.i64(<vscale x 8 x i32> [[MASKEDOFF:%.*]], <vscale x 8 x i16> [[OP1:%.*]], i16 [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
+//
+vint32m4_t test_vwmul_vx_i32m4_m(vbool8_t mask, vint32m4_t maskedoff, vint16m2_t op1, int16_t op2, size_t vl) {
+  return vwmul_vx_i32m4_m(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmul_vv_i32m8_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vwmul.mask.nxv16i32.nxv16i16.nxv16i16.i32(<vscale x 16 x i32> [[MASKEDOFF:%.*]], <vscale x 16 x i16> [[OP1:%.*]], <vscale x 16 x i16> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmul_vv_i32m8_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vwmul.mask.nxv16i32.nxv16i16.nxv16i16.i64(<vscale x 16 x i32> [[MASKEDOFF:%.*]], <vscale x 16 x i16> [[OP1:%.*]], <vscale x 16 x i16> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
+//
+vint32m8_t test_vwmul_vv_i32m8_m(vbool4_t mask, vint32m8_t maskedoff, vint16m4_t op1, vint16m4_t op2, size_t vl) {
+  return vwmul_vv_i32m8_m(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmul_vx_i32m8_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vwmul.mask.nxv16i32.nxv16i16.i16.i32(<vscale x 16 x i32> [[MASKEDOFF:%.*]], <vscale x 16 x i16> [[OP1:%.*]], i16 [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmul_vx_i32m8_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vwmul.mask.nxv16i32.nxv16i16.i16.i64(<vscale x 16 x i32> [[MASKEDOFF:%.*]], <vscale x 16 x i16> [[OP1:%.*]], i16 [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
+//
+vint32m8_t test_vwmul_vx_i32m8_m(vbool4_t mask, vint32m8_t maskedoff, vint16m4_t op1, int16_t op2, size_t vl) {
+  return vwmul_vx_i32m8_m(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmul_vv_i64m1_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vwmul.mask.nxv1i64.nxv1i32.nxv1i32.i32(<vscale x 1 x i64> [[MASKEDOFF:%.*]], <vscale x 1 x i32> [[OP1:%.*]], <vscale x 1 x i32> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 1 x i64> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmul_vv_i64m1_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vwmul.mask.nxv1i64.nxv1i32.nxv1i32.i64(<vscale x 1 x i64> [[MASKEDOFF:%.*]], <vscale x 1 x i32> [[OP1:%.*]], <vscale x 1 x i32> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 1 x i64> [[TMP0]]
+//
+vint64m1_t test_vwmul_vv_i64m1_m(vbool64_t mask, vint64m1_t maskedoff, vint32mf2_t op1, vint32mf2_t op2, size_t vl) {
+  return vwmul_vv_i64m1_m(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmul_vx_i64m1_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vwmul.mask.nxv1i64.nxv1i32.i32.i32(<vscale x 1 x i64> [[MASKEDOFF:%.*]], <vscale x 1 x i32> [[OP1:%.*]], i32 [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 1 x i64> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmul_vx_i64m1_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vwmul.mask.nxv1i64.nxv1i32.i32.i64(<vscale x 1 x i64> [[MASKEDOFF:%.*]], <vscale x 1 x i32> [[OP1:%.*]], i32 [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 1 x i64> [[TMP0]]
+//
+vint64m1_t test_vwmul_vx_i64m1_m(vbool64_t mask, vint64m1_t maskedoff, vint32mf2_t op1, int32_t op2, size_t vl) {
+  return vwmul_vx_i64m1_m(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmul_vv_i64m2_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vwmul.mask.nxv2i64.nxv2i32.nxv2i32.i32(<vscale x 2 x i64> [[MASKEDOFF:%.*]], <vscale x 2 x i32> [[OP1:%.*]], <vscale x 2 x i32> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 2 x i64> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmul_vv_i64m2_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vwmul.mask.nxv2i64.nxv2i32.nxv2i32.i64(<vscale x 2 x i64> [[MASKEDOFF:%.*]], <vscale x 2 x i32> [[OP1:%.*]], <vscale x 2 x i32> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 2 x i64> [[TMP0]]
+//
+vint64m2_t test_vwmul_vv_i64m2_m(vbool32_t mask, vint64m2_t maskedoff, vint32m1_t op1, vint32m1_t op2, size_t vl) {
+  return vwmul_vv_i64m2_m(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmul_vx_i64m2_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vwmul.mask.nxv2i64.nxv2i32.i32.i32(<vscale x 2 x i64> [[MASKEDOFF:%.*]], <vscale x 2 x i32> [[OP1:%.*]], i32 [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 2 x i64> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmul_vx_i64m2_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vwmul.mask.nxv2i64.nxv2i32.i32.i64(<vscale x 2 x i64> [[MASKEDOFF:%.*]], <vscale x 2 x i32> [[OP1:%.*]], i32 [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 2 x i64> [[TMP0]]
+//
+vint64m2_t test_vwmul_vx_i64m2_m(vbool32_t mask, vint64m2_t maskedoff, vint32m1_t op1, int32_t op2, size_t vl) {
+  return vwmul_vx_i64m2_m(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmul_vv_i64m4_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vwmul.mask.nxv4i64.nxv4i32.nxv4i32.i32(<vscale x 4 x i64> [[MASKEDOFF:%.*]], <vscale x 4 x i32> [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 4 x i64> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmul_vv_i64m4_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vwmul.mask.nxv4i64.nxv4i32.nxv4i32.i64(<vscale x 4 x i64> [[MASKEDOFF:%.*]], <vscale x 4 x i32> [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 4 x i64> [[TMP0]]
+//
+vint64m4_t test_vwmul_vv_i64m4_m(vbool16_t mask, vint64m4_t maskedoff, vint32m2_t op1, vint32m2_t op2, size_t vl) {
+  return vwmul_vv_i64m4_m(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmul_vx_i64m4_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vwmul.mask.nxv4i64.nxv4i32.i32.i32(<vscale x 4 x i64> [[MASKEDOFF:%.*]], <vscale x 4 x i32> [[OP1:%.*]], i32 [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 4 x i64> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmul_vx_i64m4_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vwmul.mask.nxv4i64.nxv4i32.i32.i64(<vscale x 4 x i64> [[MASKEDOFF:%.*]], <vscale x 4 x i32> [[OP1:%.*]], i32 [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 4 x i64> [[TMP0]]
+//
+vint64m4_t test_vwmul_vx_i64m4_m(vbool16_t mask, vint64m4_t maskedoff, vint32m2_t op1, int32_t op2, size_t vl) {
+  return vwmul_vx_i64m4_m(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmul_vv_i64m8_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vwmul.mask.nxv8i64.nxv8i32.nxv8i32.i32(<vscale x 8 x i64> [[MASKEDOFF:%.*]], <vscale x 8 x i32> [[OP1:%.*]], <vscale x 8 x i32> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 8 x i64> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmul_vv_i64m8_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vwmul.mask.nxv8i64.nxv8i32.nxv8i32.i64(<vscale x 8 x i64> [[MASKEDOFF:%.*]], <vscale x 8 x i32> [[OP1:%.*]], <vscale x 8 x i32> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 8 x i64> [[TMP0]]
+//
+vint64m8_t test_vwmul_vv_i64m8_m(vbool8_t mask, vint64m8_t maskedoff, vint32m4_t op1, vint32m4_t op2, size_t vl) {
+  return vwmul_vv_i64m8_m(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmul_vx_i64m8_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vwmul.mask.nxv8i64.nxv8i32.i32.i32(<vscale x 8 x i64> [[MASKEDOFF:%.*]], <vscale x 8 x i32> [[OP1:%.*]], i32 [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 8 x i64> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmul_vx_i64m8_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vwmul.mask.nxv8i64.nxv8i32.i32.i64(<vscale x 8 x i64> [[MASKEDOFF:%.*]], <vscale x 8 x i32> [[OP1:%.*]], i32 [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 8 x i64> [[TMP0]]
+//
+vint64m8_t test_vwmul_vx_i64m8_m(vbool8_t mask, vint64m8_t maskedoff, vint32m4_t op1, int32_t op2, size_t vl) {
+  return vwmul_vx_i64m8_m(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulu_vv_u16mf4_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vwmulu.mask.nxv1i16.nxv1i8.nxv1i8.i32(<vscale x 1 x i16> [[MASKEDOFF:%.*]], <vscale x 1 x i8> [[OP1:%.*]], <vscale x 1 x i8> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 1 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulu_vv_u16mf4_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vwmulu.mask.nxv1i16.nxv1i8.nxv1i8.i64(<vscale x 1 x i16> [[MASKEDOFF:%.*]], <vscale x 1 x i8> [[OP1:%.*]], <vscale x 1 x i8> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 1 x i16> [[TMP0]]
+//
+vuint16mf4_t test_vwmulu_vv_u16mf4_m(vbool64_t mask, vuint16mf4_t maskedoff, vuint8mf8_t op1, vuint8mf8_t op2, size_t vl) {
+  return vwmulu_vv_u16mf4_m(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulu_vx_u16mf4_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vwmulu.mask.nxv1i16.nxv1i8.i8.i32(<vscale x 1 x i16> [[MASKEDOFF:%.*]], <vscale x 1 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 1 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulu_vx_u16mf4_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vwmulu.mask.nxv1i16.nxv1i8.i8.i64(<vscale x 1 x i16> [[MASKEDOFF:%.*]], <vscale x 1 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 1 x i16> [[TMP0]]
+//
+vuint16mf4_t test_vwmulu_vx_u16mf4_m(vbool64_t mask, vuint16mf4_t maskedoff, vuint8mf8_t op1, uint8_t op2, size_t vl) {
+  return vwmulu_vx_u16mf4_m(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulu_vv_u16mf2_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vwmulu.mask.nxv2i16.nxv2i8.nxv2i8.i32(<vscale x 2 x i16> [[MASKEDOFF:%.*]], <vscale x 2 x i8> [[OP1:%.*]], <vscale x 2 x i8> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 2 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulu_vv_u16mf2_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vwmulu.mask.nxv2i16.nxv2i8.nxv2i8.i64(<vscale x 2 x i16> [[MASKEDOFF:%.*]], <vscale x 2 x i8> [[OP1:%.*]], <vscale x 2 x i8> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 2 x i16> [[TMP0]]
+//
+vuint16mf2_t test_vwmulu_vv_u16mf2_m(vbool32_t mask, vuint16mf2_t maskedoff, vuint8mf4_t op1, vuint8mf4_t op2, size_t vl) {
+  return vwmulu_vv_u16mf2_m(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulu_vx_u16mf2_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vwmulu.mask.nxv2i16.nxv2i8.i8.i32(<vscale x 2 x i16> [[MASKEDOFF:%.*]], <vscale x 2 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 2 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulu_vx_u16mf2_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vwmulu.mask.nxv2i16.nxv2i8.i8.i64(<vscale x 2 x i16> [[MASKEDOFF:%.*]], <vscale x 2 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 2 x i16> [[TMP0]]
+//
+vuint16mf2_t test_vwmulu_vx_u16mf2_m(vbool32_t mask, vuint16mf2_t maskedoff, vuint8mf4_t op1, uint8_t op2, size_t vl) {
+  return vwmulu_vx_u16mf2_m(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulu_vv_u16m1_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vwmulu.mask.nxv4i16.nxv4i8.nxv4i8.i32(<vscale x 4 x i16> [[MASKEDOFF:%.*]], <vscale x 4 x i8> [[OP1:%.*]], <vscale x 4 x i8> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 4 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulu_vv_u16m1_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vwmulu.mask.nxv4i16.nxv4i8.nxv4i8.i64(<vscale x 4 x i16> [[MASKEDOFF:%.*]], <vscale x 4 x i8> [[OP1:%.*]], <vscale x 4 x i8> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 4 x i16> [[TMP0]]
+//
+vuint16m1_t test_vwmulu_vv_u16m1_m(vbool16_t mask, vuint16m1_t maskedoff, vuint8mf2_t op1, vuint8mf2_t op2, size_t vl) {
+  return vwmulu_vv_u16m1_m(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulu_vx_u16m1_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vwmulu.mask.nxv4i16.nxv4i8.i8.i32(<vscale x 4 x i16> [[MASKEDOFF:%.*]], <vscale x 4 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 4 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulu_vx_u16m1_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vwmulu.mask.nxv4i16.nxv4i8.i8.i64(<vscale x 4 x i16> [[MASKEDOFF:%.*]], <vscale x 4 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 4 x i16> [[TMP0]]
+//
+vuint16m1_t test_vwmulu_vx_u16m1_m(vbool16_t mask, vuint16m1_t maskedoff, vuint8mf2_t op1, uint8_t op2, size_t vl) {
+  return vwmulu_vx_u16m1_m(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulu_vv_u16m2_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vwmulu.mask.nxv8i16.nxv8i8.nxv8i8.i32(<vscale x 8 x i16> [[MASKEDOFF:%.*]], <vscale x 8 x i8> [[OP1:%.*]], <vscale x 8 x i8> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulu_vv_u16m2_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vwmulu.mask.nxv8i16.nxv8i8.nxv8i8.i64(<vscale x 8 x i16> [[MASKEDOFF:%.*]], <vscale x 8 x i8> [[OP1:%.*]], <vscale x 8 x i8> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
+//
+vuint16m2_t test_vwmulu_vv_u16m2_m(vbool8_t mask, vuint16m2_t maskedoff, vuint8m1_t op1, vuint8m1_t op2, size_t vl) {
+  return vwmulu_vv_u16m2_m(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulu_vx_u16m2_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vwmulu.mask.nxv8i16.nxv8i8.i8.i32(<vscale x 8 x i16> [[MASKEDOFF:%.*]], <vscale x 8 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulu_vx_u16m2_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vwmulu.mask.nxv8i16.nxv8i8.i8.i64(<vscale x 8 x i16> [[MASKEDOFF:%.*]], <vscale x 8 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
+//
+vuint16m2_t test_vwmulu_vx_u16m2_m(vbool8_t mask, vuint16m2_t maskedoff, vuint8m1_t op1, uint8_t op2, size_t vl) {
+  return vwmulu_vx_u16m2_m(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulu_vv_u16m4_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vwmulu.mask.nxv16i16.nxv16i8.nxv16i8.i32(<vscale x 16 x i16> [[MASKEDOFF:%.*]], <vscale x 16 x i8> [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 16 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulu_vv_u16m4_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vwmulu.mask.nxv16i16.nxv16i8.nxv16i8.i64(<vscale x 16 x i16> [[MASKEDOFF:%.*]], <vscale x 16 x i8> [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 16 x i16> [[TMP0]]
+//
+vuint16m4_t test_vwmulu_vv_u16m4_m(vbool4_t mask, vuint16m4_t maskedoff, vuint8m2_t op1, vuint8m2_t op2, size_t vl) {
+  return vwmulu_vv_u16m4_m(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulu_vx_u16m4_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vwmulu.mask.nxv16i16.nxv16i8.i8.i32(<vscale x 16 x i16> [[MASKEDOFF:%.*]], <vscale x 16 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 16 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulu_vx_u16m4_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vwmulu.mask.nxv16i16.nxv16i8.i8.i64(<vscale x 16 x i16> [[MASKEDOFF:%.*]], <vscale x 16 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 16 x i16> [[TMP0]]
+//
+vuint16m4_t test_vwmulu_vx_u16m4_m(vbool4_t mask, vuint16m4_t maskedoff, vuint8m2_t op1, uint8_t op2, size_t vl) {
+  return vwmulu_vx_u16m4_m(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulu_vv_u16m8_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vwmulu.mask.nxv32i16.nxv32i8.nxv32i8.i32(<vscale x 32 x i16> [[MASKEDOFF:%.*]], <vscale x 32 x i8> [[OP1:%.*]], <vscale x 32 x i8> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 32 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulu_vv_u16m8_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vwmulu.mask.nxv32i16.nxv32i8.nxv32i8.i64(<vscale x 32 x i16> [[MASKEDOFF:%.*]], <vscale x 32 x i8> [[OP1:%.*]], <vscale x 32 x i8> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 32 x i16> [[TMP0]]
+//
+vuint16m8_t test_vwmulu_vv_u16m8_m(vbool2_t mask, vuint16m8_t maskedoff, vuint8m4_t op1, vuint8m4_t op2, size_t vl) {
+  return vwmulu_vv_u16m8_m(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulu_vx_u16m8_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vwmulu.mask.nxv32i16.nxv32i8.i8.i32(<vscale x 32 x i16> [[MASKEDOFF:%.*]], <vscale x 32 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 32 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulu_vx_u16m8_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vwmulu.mask.nxv32i16.nxv32i8.i8.i64(<vscale x 32 x i16> [[MASKEDOFF:%.*]], <vscale x 32 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 32 x i16> [[TMP0]]
+//
+vuint16m8_t test_vwmulu_vx_u16m8_m(vbool2_t mask, vuint16m8_t maskedoff, vuint8m4_t op1, uint8_t op2, size_t vl) {
+  return vwmulu_vx_u16m8_m(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulu_vv_u32mf2_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vwmulu.mask.nxv1i32.nxv1i16.nxv1i16.i32(<vscale x 1 x i32> [[MASKEDOFF:%.*]], <vscale x 1 x i16> [[OP1:%.*]], <vscale x 1 x i16> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulu_vv_u32mf2_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vwmulu.mask.nxv1i32.nxv1i16.nxv1i16.i64(<vscale x 1 x i32> [[MASKEDOFF:%.*]], <vscale x 1 x i16> [[OP1:%.*]], <vscale x 1 x i16> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
+//
+vuint32mf2_t test_vwmulu_vv_u32mf2_m(vbool64_t mask, vuint32mf2_t maskedoff, vuint16mf4_t op1, vuint16mf4_t op2, size_t vl) {
+  return vwmulu_vv_u32mf2_m(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulu_vx_u32mf2_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vwmulu.mask.nxv1i32.nxv1i16.i16.i32(<vscale x 1 x i32> [[MASKEDOFF:%.*]], <vscale x 1 x i16> [[OP1:%.*]], i16 [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulu_vx_u32mf2_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vwmulu.mask.nxv1i32.nxv1i16.i16.i64(<vscale x 1 x i32> [[MASKEDOFF:%.*]], <vscale x 1 x i16> [[OP1:%.*]], i16 [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
+//
+vuint32mf2_t test_vwmulu_vx_u32mf2_m(vbool64_t mask, vuint32mf2_t maskedoff, vuint16mf4_t op1, uint16_t op2, size_t vl) {
+  return vwmulu_vx_u32mf2_m(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulu_vv_u32m1_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vwmulu.mask.nxv2i32.nxv2i16.nxv2i16.i32(<vscale x 2 x i32> [[MASKEDOFF:%.*]], <vscale x 2 x i16> [[OP1:%.*]], <vscale x 2 x i16> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulu_vv_u32m1_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vwmulu.mask.nxv2i32.nxv2i16.nxv2i16.i64(<vscale x 2 x i32> [[MASKEDOFF:%.*]], <vscale x 2 x i16> [[OP1:%.*]], <vscale x 2 x i16> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
+//
+vuint32m1_t test_vwmulu_vv_u32m1_m(vbool32_t mask, vuint32m1_t maskedoff, vuint16mf2_t op1, vuint16mf2_t op2, size_t vl) {
+  return vwmulu_vv_u32m1_m(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulu_vx_u32m1_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vwmulu.mask.nxv2i32.nxv2i16.i16.i32(<vscale x 2 x i32> [[MASKEDOFF:%.*]], <vscale x 2 x i16> [[OP1:%.*]], i16 [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulu_vx_u32m1_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vwmulu.mask.nxv2i32.nxv2i16.i16.i64(<vscale x 2 x i32> [[MASKEDOFF:%.*]], <vscale x 2 x i16> [[OP1:%.*]], i16 [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
+//
+vuint32m1_t test_vwmulu_vx_u32m1_m(vbool32_t mask, vuint32m1_t maskedoff, vuint16mf2_t op1, uint16_t op2, size_t vl) {
+  return vwmulu_vx_u32m1_m(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulu_vv_u32m2_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vwmulu.mask.nxv4i32.nxv4i16.nxv4i16.i32(<vscale x 4 x i32> [[MASKEDOFF:%.*]], <vscale x 4 x i16> [[OP1:%.*]], <vscale x 4 x i16> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulu_vv_u32m2_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vwmulu.mask.nxv4i32.nxv4i16.nxv4i16.i64(<vscale x 4 x i32> [[MASKEDOFF:%.*]], <vscale x 4 x i16> [[OP1:%.*]], <vscale x 4 x i16> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
+//
+vuint32m2_t test_vwmulu_vv_u32m2_m(vbool16_t mask, vuint32m2_t maskedoff, vuint16m1_t op1, vuint16m1_t op2, size_t vl) {
+  return vwmulu_vv_u32m2_m(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulu_vx_u32m2_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vwmulu.mask.nxv4i32.nxv4i16.i16.i32(<vscale x 4 x i32> [[MASKEDOFF:%.*]], <vscale x 4 x i16> [[OP1:%.*]], i16 [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulu_vx_u32m2_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vwmulu.mask.nxv4i32.nxv4i16.i16.i64(<vscale x 4 x i32> [[MASKEDOFF:%.*]], <vscale x 4 x i16> [[OP1:%.*]], i16 [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
+//
+vuint32m2_t test_vwmulu_vx_u32m2_m(vbool16_t mask, vuint32m2_t maskedoff, vuint16m1_t op1, uint16_t op2, size_t vl) {
+  return vwmulu_vx_u32m2_m(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulu_vv_u32m4_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vwmulu.mask.nxv8i32.nxv8i16.nxv8i16.i32(<vscale x 8 x i32> [[MASKEDOFF:%.*]], <vscale x 8 x i16> [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulu_vv_u32m4_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vwmulu.mask.nxv8i32.nxv8i16.nxv8i16.i64(<vscale x 8 x i32> [[MASKEDOFF:%.*]], <vscale x 8 x i16> [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
+//
+vuint32m4_t test_vwmulu_vv_u32m4_m(vbool8_t mask, vuint32m4_t maskedoff, vuint16m2_t op1, vuint16m2_t op2, size_t vl) {
+  return vwmulu_vv_u32m4_m(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulu_vx_u32m4_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vwmulu.mask.nxv8i32.nxv8i16.i16.i32(<vscale x 8 x i32> [[MASKEDOFF:%.*]], <vscale x 8 x i16> [[OP1:%.*]], i16 [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulu_vx_u32m4_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vwmulu.mask.nxv8i32.nxv8i16.i16.i64(<vscale x 8 x i32> [[MASKEDOFF:%.*]], <vscale x 8 x i16> [[OP1:%.*]], i16 [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
+//
+vuint32m4_t test_vwmulu_vx_u32m4_m(vbool8_t mask, vuint32m4_t maskedoff, vuint16m2_t op1, uint16_t op2, size_t vl) {
+  return vwmulu_vx_u32m4_m(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulu_vv_u32m8_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vwmulu.mask.nxv16i32.nxv16i16.nxv16i16.i32(<vscale x 16 x i32> [[MASKEDOFF:%.*]], <vscale x 16 x i16> [[OP1:%.*]], <vscale x 16 x i16> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulu_vv_u32m8_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vwmulu.mask.nxv16i32.nxv16i16.nxv16i16.i64(<vscale x 16 x i32> [[MASKEDOFF:%.*]], <vscale x 16 x i16> [[OP1:%.*]], <vscale x 16 x i16> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
+//
+vuint32m8_t test_vwmulu_vv_u32m8_m(vbool4_t mask, vuint32m8_t maskedoff, vuint16m4_t op1, vuint16m4_t op2, size_t vl) {
+  return vwmulu_vv_u32m8_m(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulu_vx_u32m8_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vwmulu.mask.nxv16i32.nxv16i16.i16.i32(<vscale x 16 x i32> [[MASKEDOFF:%.*]], <vscale x 16 x i16> [[OP1:%.*]], i16 [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulu_vx_u32m8_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vwmulu.mask.nxv16i32.nxv16i16.i16.i64(<vscale x 16 x i32> [[MASKEDOFF:%.*]], <vscale x 16 x i16> [[OP1:%.*]], i16 [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
+//
+vuint32m8_t test_vwmulu_vx_u32m8_m(vbool4_t mask, vuint32m8_t maskedoff, vuint16m4_t op1, uint16_t op2, size_t vl) {
+  return vwmulu_vx_u32m8_m(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulu_vv_u64m1_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vwmulu.mask.nxv1i64.nxv1i32.nxv1i32.i32(<vscale x 1 x i64> [[MASKEDOFF:%.*]], <vscale x 1 x i32> [[OP1:%.*]], <vscale x 1 x i32> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 1 x i64> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulu_vv_u64m1_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vwmulu.mask.nxv1i64.nxv1i32.nxv1i32.i64(<vscale x 1 x i64> [[MASKEDOFF:%.*]], <vscale x 1 x i32> [[OP1:%.*]], <vscale x 1 x i32> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 1 x i64> [[TMP0]]
+//
+vuint64m1_t test_vwmulu_vv_u64m1_m(vbool64_t mask, vuint64m1_t maskedoff, vuint32mf2_t op1, vuint32mf2_t op2, size_t vl) {
+  return vwmulu_vv_u64m1_m(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulu_vx_u64m1_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vwmulu.mask.nxv1i64.nxv1i32.i32.i32(<vscale x 1 x i64> [[MASKEDOFF:%.*]], <vscale x 1 x i32> [[OP1:%.*]], i32 [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 1 x i64> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulu_vx_u64m1_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vwmulu.mask.nxv1i64.nxv1i32.i32.i64(<vscale x 1 x i64> [[MASKEDOFF:%.*]], <vscale x 1 x i32> [[OP1:%.*]], i32 [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 1 x i64> [[TMP0]]
+//
+vuint64m1_t test_vwmulu_vx_u64m1_m(vbool64_t mask, vuint64m1_t maskedoff, vuint32mf2_t op1, uint32_t op2, size_t vl) {
+  return vwmulu_vx_u64m1_m(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulu_vv_u64m2_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vwmulu.mask.nxv2i64.nxv2i32.nxv2i32.i32(<vscale x 2 x i64> [[MASKEDOFF:%.*]], <vscale x 2 x i32> [[OP1:%.*]], <vscale x 2 x i32> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 2 x i64> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulu_vv_u64m2_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vwmulu.mask.nxv2i64.nxv2i32.nxv2i32.i64(<vscale x 2 x i64> [[MASKEDOFF:%.*]], <vscale x 2 x i32> [[OP1:%.*]], <vscale x 2 x i32> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 2 x i64> [[TMP0]]
+//
+vuint64m2_t test_vwmulu_vv_u64m2_m(vbool32_t mask, vuint64m2_t maskedoff, vuint32m1_t op1, vuint32m1_t op2, size_t vl) {
+  return vwmulu_vv_u64m2_m(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulu_vx_u64m2_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vwmulu.mask.nxv2i64.nxv2i32.i32.i32(<vscale x 2 x i64> [[MASKEDOFF:%.*]], <vscale x 2 x i32> [[OP1:%.*]], i32 [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 2 x i64> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulu_vx_u64m2_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vwmulu.mask.nxv2i64.nxv2i32.i32.i64(<vscale x 2 x i64> [[MASKEDOFF:%.*]], <vscale x 2 x i32> [[OP1:%.*]], i32 [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 2 x i64> [[TMP0]]
+//
+vuint64m2_t test_vwmulu_vx_u64m2_m(vbool32_t mask, vuint64m2_t maskedoff, vuint32m1_t op1, uint32_t op2, size_t vl) {
+  return vwmulu_vx_u64m2_m(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulu_vv_u64m4_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vwmulu.mask.nxv4i64.nxv4i32.nxv4i32.i32(<vscale x 4 x i64> [[MASKEDOFF:%.*]], <vscale x 4 x i32> [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 4 x i64> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulu_vv_u64m4_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vwmulu.mask.nxv4i64.nxv4i32.nxv4i32.i64(<vscale x 4 x i64> [[MASKEDOFF:%.*]], <vscale x 4 x i32> [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 4 x i64> [[TMP0]]
+//
+vuint64m4_t test_vwmulu_vv_u64m4_m(vbool16_t mask, vuint64m4_t maskedoff, vuint32m2_t op1, vuint32m2_t op2, size_t vl) {
+  return vwmulu_vv_u64m4_m(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulu_vx_u64m4_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vwmulu.mask.nxv4i64.nxv4i32.i32.i32(<vscale x 4 x i64> [[MASKEDOFF:%.*]], <vscale x 4 x i32> [[OP1:%.*]], i32 [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 4 x i64> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulu_vx_u64m4_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vwmulu.mask.nxv4i64.nxv4i32.i32.i64(<vscale x 4 x i64> [[MASKEDOFF:%.*]], <vscale x 4 x i32> [[OP1:%.*]], i32 [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 4 x i64> [[TMP0]]
+//
+vuint64m4_t test_vwmulu_vx_u64m4_m(vbool16_t mask, vuint64m4_t maskedoff, vuint32m2_t op1, uint32_t op2, size_t vl) {
+  return vwmulu_vx_u64m4_m(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulu_vv_u64m8_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vwmulu.mask.nxv8i64.nxv8i32.nxv8i32.i32(<vscale x 8 x i64> [[MASKEDOFF:%.*]], <vscale x 8 x i32> [[OP1:%.*]], <vscale x 8 x i32> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 8 x i64> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulu_vv_u64m8_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vwmulu.mask.nxv8i64.nxv8i32.nxv8i32.i64(<vscale x 8 x i64> [[MASKEDOFF:%.*]], <vscale x 8 x i32> [[OP1:%.*]], <vscale x 8 x i32> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 8 x i64> [[TMP0]]
+//
+vuint64m8_t test_vwmulu_vv_u64m8_m(vbool8_t mask, vuint64m8_t maskedoff, vuint32m4_t op1, vuint32m4_t op2, size_t vl) {
+  return vwmulu_vv_u64m8_m(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulu_vx_u64m8_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vwmulu.mask.nxv8i64.nxv8i32.i32.i32(<vscale x 8 x i64> [[MASKEDOFF:%.*]], <vscale x 8 x i32> [[OP1:%.*]], i32 [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 8 x i64> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulu_vx_u64m8_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vwmulu.mask.nxv8i64.nxv8i32.i32.i64(<vscale x 8 x i64> [[MASKEDOFF:%.*]], <vscale x 8 x i32> [[OP1:%.*]], i32 [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 8 x i64> [[TMP0]]
+//
+vuint64m8_t test_vwmulu_vx_u64m8_m(vbool8_t mask, vuint64m8_t maskedoff, vuint32m4_t op1, uint32_t op2, size_t vl) {
+  return vwmulu_vx_u64m8_m(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulsu_vv_i16mf4_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vwmulsu.mask.nxv1i16.nxv1i8.nxv1i8.i32(<vscale x 1 x i16> [[MASKEDOFF:%.*]], <vscale x 1 x i8> [[OP1:%.*]], <vscale x 1 x i8> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 1 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulsu_vv_i16mf4_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vwmulsu.mask.nxv1i16.nxv1i8.nxv1i8.i64(<vscale x 1 x i16> [[MASKEDOFF:%.*]], <vscale x 1 x i8> [[OP1:%.*]], <vscale x 1 x i8> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 1 x i16> [[TMP0]]
+//
+vint16mf4_t test_vwmulsu_vv_i16mf4_m(vbool64_t mask, vint16mf4_t maskedoff, vint8mf8_t op1, vuint8mf8_t op2, size_t vl) {
+  return vwmulsu_vv_i16mf4_m(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulsu_vx_i16mf4_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vwmulsu.mask.nxv1i16.nxv1i8.i8.i32(<vscale x 1 x i16> [[MASKEDOFF:%.*]], <vscale x 1 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 1 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulsu_vx_i16mf4_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i16> @llvm.riscv.vwmulsu.mask.nxv1i16.nxv1i8.i8.i64(<vscale x 1 x i16> [[MASKEDOFF:%.*]], <vscale x 1 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 1 x i16> [[TMP0]]
+//
+vint16mf4_t test_vwmulsu_vx_i16mf4_m(vbool64_t mask, vint16mf4_t maskedoff, vint8mf8_t op1, uint8_t op2, size_t vl) {
+  return vwmulsu_vx_i16mf4_m(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulsu_vv_i16mf2_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vwmulsu.mask.nxv2i16.nxv2i8.nxv2i8.i32(<vscale x 2 x i16> [[MASKEDOFF:%.*]], <vscale x 2 x i8> [[OP1:%.*]], <vscale x 2 x i8> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 2 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulsu_vv_i16mf2_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vwmulsu.mask.nxv2i16.nxv2i8.nxv2i8.i64(<vscale x 2 x i16> [[MASKEDOFF:%.*]], <vscale x 2 x i8> [[OP1:%.*]], <vscale x 2 x i8> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 2 x i16> [[TMP0]]
+//
+vint16mf2_t test_vwmulsu_vv_i16mf2_m(vbool32_t mask, vint16mf2_t maskedoff, vint8mf4_t op1, vuint8mf4_t op2, size_t vl) {
+  return vwmulsu_vv_i16mf2_m(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulsu_vx_i16mf2_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vwmulsu.mask.nxv2i16.nxv2i8.i8.i32(<vscale x 2 x i16> [[MASKEDOFF:%.*]], <vscale x 2 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 2 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulsu_vx_i16mf2_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i16> @llvm.riscv.vwmulsu.mask.nxv2i16.nxv2i8.i8.i64(<vscale x 2 x i16> [[MASKEDOFF:%.*]], <vscale x 2 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 2 x i16> [[TMP0]]
+//
+vint16mf2_t test_vwmulsu_vx_i16mf2_m(vbool32_t mask, vint16mf2_t maskedoff, vint8mf4_t op1, uint8_t op2, size_t vl) {
+  return vwmulsu_vx_i16mf2_m(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulsu_vv_i16m1_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vwmulsu.mask.nxv4i16.nxv4i8.nxv4i8.i32(<vscale x 4 x i16> [[MASKEDOFF:%.*]], <vscale x 4 x i8> [[OP1:%.*]], <vscale x 4 x i8> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 4 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulsu_vv_i16m1_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vwmulsu.mask.nxv4i16.nxv4i8.nxv4i8.i64(<vscale x 4 x i16> [[MASKEDOFF:%.*]], <vscale x 4 x i8> [[OP1:%.*]], <vscale x 4 x i8> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 4 x i16> [[TMP0]]
+//
+vint16m1_t test_vwmulsu_vv_i16m1_m(vbool16_t mask, vint16m1_t maskedoff, vint8mf2_t op1, vuint8mf2_t op2, size_t vl) {
+  return vwmulsu_vv_i16m1_m(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulsu_vx_i16m1_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vwmulsu.mask.nxv4i16.nxv4i8.i8.i32(<vscale x 4 x i16> [[MASKEDOFF:%.*]], <vscale x 4 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 4 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulsu_vx_i16m1_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i16> @llvm.riscv.vwmulsu.mask.nxv4i16.nxv4i8.i8.i64(<vscale x 4 x i16> [[MASKEDOFF:%.*]], <vscale x 4 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 4 x i16> [[TMP0]]
+//
+vint16m1_t test_vwmulsu_vx_i16m1_m(vbool16_t mask, vint16m1_t maskedoff, vint8mf2_t op1, uint8_t op2, size_t vl) {
+  return vwmulsu_vx_i16m1_m(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulsu_vv_i16m2_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vwmulsu.mask.nxv8i16.nxv8i8.nxv8i8.i32(<vscale x 8 x i16> [[MASKEDOFF:%.*]], <vscale x 8 x i8> [[OP1:%.*]], <vscale x 8 x i8> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulsu_vv_i16m2_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vwmulsu.mask.nxv8i16.nxv8i8.nxv8i8.i64(<vscale x 8 x i16> [[MASKEDOFF:%.*]], <vscale x 8 x i8> [[OP1:%.*]], <vscale x 8 x i8> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
+//
+vint16m2_t test_vwmulsu_vv_i16m2_m(vbool8_t mask, vint16m2_t maskedoff, vint8m1_t op1, vuint8m1_t op2, size_t vl) {
+  return vwmulsu_vv_i16m2_m(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulsu_vx_i16m2_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vwmulsu.mask.nxv8i16.nxv8i8.i8.i32(<vscale x 8 x i16> [[MASKEDOFF:%.*]], <vscale x 8 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulsu_vx_i16m2_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i16> @llvm.riscv.vwmulsu.mask.nxv8i16.nxv8i8.i8.i64(<vscale x 8 x i16> [[MASKEDOFF:%.*]], <vscale x 8 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
+//
+vint16m2_t test_vwmulsu_vx_i16m2_m(vbool8_t mask, vint16m2_t maskedoff, vint8m1_t op1, uint8_t op2, size_t vl) {
+  return vwmulsu_vx_i16m2_m(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulsu_vv_i16m4_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vwmulsu.mask.nxv16i16.nxv16i8.nxv16i8.i32(<vscale x 16 x i16> [[MASKEDOFF:%.*]], <vscale x 16 x i8> [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 16 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulsu_vv_i16m4_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vwmulsu.mask.nxv16i16.nxv16i8.nxv16i8.i64(<vscale x 16 x i16> [[MASKEDOFF:%.*]], <vscale x 16 x i8> [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 16 x i16> [[TMP0]]
+//
+vint16m4_t test_vwmulsu_vv_i16m4_m(vbool4_t mask, vint16m4_t maskedoff, vint8m2_t op1, vuint8m2_t op2, size_t vl) {
+  return vwmulsu_vv_i16m4_m(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulsu_vx_i16m4_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vwmulsu.mask.nxv16i16.nxv16i8.i8.i32(<vscale x 16 x i16> [[MASKEDOFF:%.*]], <vscale x 16 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 16 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulsu_vx_i16m4_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i16> @llvm.riscv.vwmulsu.mask.nxv16i16.nxv16i8.i8.i64(<vscale x 16 x i16> [[MASKEDOFF:%.*]], <vscale x 16 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 16 x i16> [[TMP0]]
+//
+vint16m4_t test_vwmulsu_vx_i16m4_m(vbool4_t mask, vint16m4_t maskedoff, vint8m2_t op1, uint8_t op2, size_t vl) {
+  return vwmulsu_vx_i16m4_m(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulsu_vv_i16m8_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vwmulsu.mask.nxv32i16.nxv32i8.nxv32i8.i32(<vscale x 32 x i16> [[MASKEDOFF:%.*]], <vscale x 32 x i8> [[OP1:%.*]], <vscale x 32 x i8> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 32 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulsu_vv_i16m8_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vwmulsu.mask.nxv32i16.nxv32i8.nxv32i8.i64(<vscale x 32 x i16> [[MASKEDOFF:%.*]], <vscale x 32 x i8> [[OP1:%.*]], <vscale x 32 x i8> [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 32 x i16> [[TMP0]]
+//
+vint16m8_t test_vwmulsu_vv_i16m8_m(vbool2_t mask, vint16m8_t maskedoff, vint8m4_t op1, vuint8m4_t op2, size_t vl) {
+  return vwmulsu_vv_i16m8_m(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulsu_vx_i16m8_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vwmulsu.mask.nxv32i16.nxv32i8.i8.i32(<vscale x 32 x i16> [[MASKEDOFF:%.*]], <vscale x 32 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 32 x i16> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulsu_vx_i16m8_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.vwmulsu.mask.nxv32i16.nxv32i8.i8.i64(<vscale x 32 x i16> [[MASKEDOFF:%.*]], <vscale x 32 x i8> [[OP1:%.*]], i8 [[OP2:%.*]], <vscale x 32 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 32 x i16> [[TMP0]]
+//
+vint16m8_t test_vwmulsu_vx_i16m8_m(vbool2_t mask, vint16m8_t maskedoff, vint8m4_t op1, uint8_t op2, size_t vl) {
+  return vwmulsu_vx_i16m8_m(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulsu_vv_i32mf2_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vwmulsu.mask.nxv1i32.nxv1i16.nxv1i16.i32(<vscale x 1 x i32> [[MASKEDOFF:%.*]], <vscale x 1 x i16> [[OP1:%.*]], <vscale x 1 x i16> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulsu_vv_i32mf2_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vwmulsu.mask.nxv1i32.nxv1i16.nxv1i16.i64(<vscale x 1 x i32> [[MASKEDOFF:%.*]], <vscale x 1 x i16> [[OP1:%.*]], <vscale x 1 x i16> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
+//
+vint32mf2_t test_vwmulsu_vv_i32mf2_m(vbool64_t mask, vint32mf2_t maskedoff, vint16mf4_t op1, vuint16mf4_t op2, size_t vl) {
+  return vwmulsu_vv_i32mf2_m(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulsu_vx_i32mf2_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vwmulsu.mask.nxv1i32.nxv1i16.i16.i32(<vscale x 1 x i32> [[MASKEDOFF:%.*]], <vscale x 1 x i16> [[OP1:%.*]], i16 [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulsu_vx_i32mf2_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i32> @llvm.riscv.vwmulsu.mask.nxv1i32.nxv1i16.i16.i64(<vscale x 1 x i32> [[MASKEDOFF:%.*]], <vscale x 1 x i16> [[OP1:%.*]], i16 [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 1 x i32> [[TMP0]]
+//
+vint32mf2_t test_vwmulsu_vx_i32mf2_m(vbool64_t mask, vint32mf2_t maskedoff, vint16mf4_t op1, uint16_t op2, size_t vl) {
+  return vwmulsu_vx_i32mf2_m(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulsu_vv_i32m1_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vwmulsu.mask.nxv2i32.nxv2i16.nxv2i16.i32(<vscale x 2 x i32> [[MASKEDOFF:%.*]], <vscale x 2 x i16> [[OP1:%.*]], <vscale x 2 x i16> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulsu_vv_i32m1_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vwmulsu.mask.nxv2i32.nxv2i16.nxv2i16.i64(<vscale x 2 x i32> [[MASKEDOFF:%.*]], <vscale x 2 x i16> [[OP1:%.*]], <vscale x 2 x i16> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
+//
+vint32m1_t test_vwmulsu_vv_i32m1_m(vbool32_t mask, vint32m1_t maskedoff, vint16mf2_t op1, vuint16mf2_t op2, size_t vl) {
+  return vwmulsu_vv_i32m1_m(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulsu_vx_i32m1_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vwmulsu.mask.nxv2i32.nxv2i16.i16.i32(<vscale x 2 x i32> [[MASKEDOFF:%.*]], <vscale x 2 x i16> [[OP1:%.*]], i16 [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulsu_vx_i32m1_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i32> @llvm.riscv.vwmulsu.mask.nxv2i32.nxv2i16.i16.i64(<vscale x 2 x i32> [[MASKEDOFF:%.*]], <vscale x 2 x i16> [[OP1:%.*]], i16 [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 2 x i32> [[TMP0]]
+//
+vint32m1_t test_vwmulsu_vx_i32m1_m(vbool32_t mask, vint32m1_t maskedoff, vint16mf2_t op1, uint16_t op2, size_t vl) {
+  return vwmulsu_vx_i32m1_m(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulsu_vv_i32m2_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vwmulsu.mask.nxv4i32.nxv4i16.nxv4i16.i32(<vscale x 4 x i32> [[MASKEDOFF:%.*]], <vscale x 4 x i16> [[OP1:%.*]], <vscale x 4 x i16> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulsu_vv_i32m2_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vwmulsu.mask.nxv4i32.nxv4i16.nxv4i16.i64(<vscale x 4 x i32> [[MASKEDOFF:%.*]], <vscale x 4 x i16> [[OP1:%.*]], <vscale x 4 x i16> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
+//
+vint32m2_t test_vwmulsu_vv_i32m2_m(vbool16_t mask, vint32m2_t maskedoff, vint16m1_t op1, vuint16m1_t op2, size_t vl) {
+  return vwmulsu_vv_i32m2_m(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulsu_vx_i32m2_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vwmulsu.mask.nxv4i32.nxv4i16.i16.i32(<vscale x 4 x i32> [[MASKEDOFF:%.*]], <vscale x 4 x i16> [[OP1:%.*]], i16 [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulsu_vx_i32m2_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i32> @llvm.riscv.vwmulsu.mask.nxv4i32.nxv4i16.i16.i64(<vscale x 4 x i32> [[MASKEDOFF:%.*]], <vscale x 4 x i16> [[OP1:%.*]], i16 [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
+//
+vint32m2_t test_vwmulsu_vx_i32m2_m(vbool16_t mask, vint32m2_t maskedoff, vint16m1_t op1, uint16_t op2, size_t vl) {
+  return vwmulsu_vx_i32m2_m(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulsu_vv_i32m4_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vwmulsu.mask.nxv8i32.nxv8i16.nxv8i16.i32(<vscale x 8 x i32> [[MASKEDOFF:%.*]], <vscale x 8 x i16> [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulsu_vv_i32m4_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vwmulsu.mask.nxv8i32.nxv8i16.nxv8i16.i64(<vscale x 8 x i32> [[MASKEDOFF:%.*]], <vscale x 8 x i16> [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
+//
+vint32m4_t test_vwmulsu_vv_i32m4_m(vbool8_t mask, vint32m4_t maskedoff, vint16m2_t op1, vuint16m2_t op2, size_t vl) {
+  return vwmulsu_vv_i32m4_m(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulsu_vx_i32m4_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vwmulsu.mask.nxv8i32.nxv8i16.i16.i32(<vscale x 8 x i32> [[MASKEDOFF:%.*]], <vscale x 8 x i16> [[OP1:%.*]], i16 [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulsu_vx_i32m4_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i32> @llvm.riscv.vwmulsu.mask.nxv8i32.nxv8i16.i16.i64(<vscale x 8 x i32> [[MASKEDOFF:%.*]], <vscale x 8 x i16> [[OP1:%.*]], i16 [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 8 x i32> [[TMP0]]
+//
+vint32m4_t test_vwmulsu_vx_i32m4_m(vbool8_t mask, vint32m4_t maskedoff, vint16m2_t op1, uint16_t op2, size_t vl) {
+  return vwmulsu_vx_i32m4_m(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulsu_vv_i32m8_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vwmulsu.mask.nxv16i32.nxv16i16.nxv16i16.i32(<vscale x 16 x i32> [[MASKEDOFF:%.*]], <vscale x 16 x i16> [[OP1:%.*]], <vscale x 16 x i16> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulsu_vv_i32m8_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vwmulsu.mask.nxv16i32.nxv16i16.nxv16i16.i64(<vscale x 16 x i32> [[MASKEDOFF:%.*]], <vscale x 16 x i16> [[OP1:%.*]], <vscale x 16 x i16> [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
+//
+vint32m8_t test_vwmulsu_vv_i32m8_m(vbool4_t mask, vint32m8_t maskedoff, vint16m4_t op1, vuint16m4_t op2, size_t vl) {
+  return vwmulsu_vv_i32m8_m(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulsu_vx_i32m8_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vwmulsu.mask.nxv16i32.nxv16i16.i16.i32(<vscale x 16 x i32> [[MASKEDOFF:%.*]], <vscale x 16 x i16> [[OP1:%.*]], i16 [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulsu_vx_i32m8_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vwmulsu.mask.nxv16i32.nxv16i16.i16.i64(<vscale x 16 x i32> [[MASKEDOFF:%.*]], <vscale x 16 x i16> [[OP1:%.*]], i16 [[OP2:%.*]], <vscale x 16 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
+//
+vint32m8_t test_vwmulsu_vx_i32m8_m(vbool4_t mask, vint32m8_t maskedoff, vint16m4_t op1, uint16_t op2, size_t vl) {
+  return vwmulsu_vx_i32m8_m(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulsu_vv_i64m1_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vwmulsu.mask.nxv1i64.nxv1i32.nxv1i32.i32(<vscale x 1 x i64> [[MASKEDOFF:%.*]], <vscale x 1 x i32> [[OP1:%.*]], <vscale x 1 x i32> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 1 x i64> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulsu_vv_i64m1_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vwmulsu.mask.nxv1i64.nxv1i32.nxv1i32.i64(<vscale x 1 x i64> [[MASKEDOFF:%.*]], <vscale x 1 x i32> [[OP1:%.*]], <vscale x 1 x i32> [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 1 x i64> [[TMP0]]
+//
+vint64m1_t test_vwmulsu_vv_i64m1_m(vbool64_t mask, vint64m1_t maskedoff, vint32mf2_t op1, vuint32mf2_t op2, size_t vl) {
+  return vwmulsu_vv_i64m1_m(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulsu_vx_i64m1_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vwmulsu.mask.nxv1i64.nxv1i32.i32.i32(<vscale x 1 x i64> [[MASKEDOFF:%.*]], <vscale x 1 x i32> [[OP1:%.*]], i32 [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 1 x i64> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulsu_vx_i64m1_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x i64> @llvm.riscv.vwmulsu.mask.nxv1i64.nxv1i32.i32.i64(<vscale x 1 x i64> [[MASKEDOFF:%.*]], <vscale x 1 x i32> [[OP1:%.*]], i32 [[OP2:%.*]], <vscale x 1 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 1 x i64> [[TMP0]]
+//
+vint64m1_t test_vwmulsu_vx_i64m1_m(vbool64_t mask, vint64m1_t maskedoff, vint32mf2_t op1, uint32_t op2, size_t vl) {
+  return vwmulsu_vx_i64m1_m(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulsu_vv_i64m2_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vwmulsu.mask.nxv2i64.nxv2i32.nxv2i32.i32(<vscale x 2 x i64> [[MASKEDOFF:%.*]], <vscale x 2 x i32> [[OP1:%.*]], <vscale x 2 x i32> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 2 x i64> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulsu_vv_i64m2_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vwmulsu.mask.nxv2i64.nxv2i32.nxv2i32.i64(<vscale x 2 x i64> [[MASKEDOFF:%.*]], <vscale x 2 x i32> [[OP1:%.*]], <vscale x 2 x i32> [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 2 x i64> [[TMP0]]
+//
+vint64m2_t test_vwmulsu_vv_i64m2_m(vbool32_t mask, vint64m2_t maskedoff, vint32m1_t op1, vuint32m1_t op2, size_t vl) {
+  return vwmulsu_vv_i64m2_m(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulsu_vx_i64m2_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vwmulsu.mask.nxv2i64.nxv2i32.i32.i32(<vscale x 2 x i64> [[MASKEDOFF:%.*]], <vscale x 2 x i32> [[OP1:%.*]], i32 [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 2 x i64> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulsu_vx_i64m2_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x i64> @llvm.riscv.vwmulsu.mask.nxv2i64.nxv2i32.i32.i64(<vscale x 2 x i64> [[MASKEDOFF:%.*]], <vscale x 2 x i32> [[OP1:%.*]], i32 [[OP2:%.*]], <vscale x 2 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 2 x i64> [[TMP0]]
+//
+vint64m2_t test_vwmulsu_vx_i64m2_m(vbool32_t mask, vint64m2_t maskedoff, vint32m1_t op1, uint32_t op2, size_t vl) {
+  return vwmulsu_vx_i64m2_m(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulsu_vv_i64m4_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vwmulsu.mask.nxv4i64.nxv4i32.nxv4i32.i32(<vscale x 4 x i64> [[MASKEDOFF:%.*]], <vscale x 4 x i32> [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 4 x i64> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulsu_vv_i64m4_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vwmulsu.mask.nxv4i64.nxv4i32.nxv4i32.i64(<vscale x 4 x i64> [[MASKEDOFF:%.*]], <vscale x 4 x i32> [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 4 x i64> [[TMP0]]
+//
+vint64m4_t test_vwmulsu_vv_i64m4_m(vbool16_t mask, vint64m4_t maskedoff, vint32m2_t op1, vuint32m2_t op2, size_t vl) {
+  return vwmulsu_vv_i64m4_m(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulsu_vx_i64m4_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vwmulsu.mask.nxv4i64.nxv4i32.i32.i32(<vscale x 4 x i64> [[MASKEDOFF:%.*]], <vscale x 4 x i32> [[OP1:%.*]], i32 [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 4 x i64> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulsu_vx_i64m4_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.riscv.vwmulsu.mask.nxv4i64.nxv4i32.i32.i64(<vscale x 4 x i64> [[MASKEDOFF:%.*]], <vscale x 4 x i32> [[OP1:%.*]], i32 [[OP2:%.*]], <vscale x 4 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 4 x i64> [[TMP0]]
+//
+vint64m4_t test_vwmulsu_vx_i64m4_m(vbool16_t mask, vint64m4_t maskedoff, vint32m2_t op1, uint32_t op2, size_t vl) {
+  return vwmulsu_vx_i64m4_m(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulsu_vv_i64m8_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vwmulsu.mask.nxv8i64.nxv8i32.nxv8i32.i32(<vscale x 8 x i64> [[MASKEDOFF:%.*]], <vscale x 8 x i32> [[OP1:%.*]], <vscale x 8 x i32> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 8 x i64> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulsu_vv_i64m8_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vwmulsu.mask.nxv8i64.nxv8i32.nxv8i32.i64(<vscale x 8 x i64> [[MASKEDOFF:%.*]], <vscale x 8 x i32> [[OP1:%.*]], <vscale x 8 x i32> [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 8 x i64> [[TMP0]]
+//
+vint64m8_t test_vwmulsu_vv_i64m8_m(vbool8_t mask, vint64m8_t maskedoff, vint32m4_t op1, vuint32m4_t op2, size_t vl) {
+  return vwmulsu_vv_i64m8_m(mask, maskedoff, op1, op2, vl);
+}
+
+// CHECK-RV32-LABEL: @test_vwmulsu_vx_i64m8_m(
+// CHECK-RV32-NEXT:  entry:
+// CHECK-RV32-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vwmulsu.mask.nxv8i64.nxv8i32.i32.i32(<vscale x 8 x i64> [[MASKEDOFF:%.*]], <vscale x 8 x i32> [[OP1:%.*]], i32 [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i32 [[VL:%.*]])
+// CHECK-RV32-NEXT:    ret <vscale x 8 x i64> [[TMP0]]
+//
+// CHECK-RV64-LABEL: @test_vwmulsu_vx_i64m8_m(
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.vwmulsu.mask.nxv8i64.nxv8i32.i32.i64(<vscale x 8 x i64> [[MASKEDOFF:%.*]], <vscale x 8 x i32> [[OP1:%.*]], i32 [[OP2:%.*]], <vscale x 8 x i1> [[MASK:%.*]], i64 [[VL:%.*]])
+// CHECK-RV64-NEXT:    ret <vscale x 8 x i64> [[TMP0]]
+//
+vint64m8_t test_vwmulsu_vx_i64m8_m(vbool8_t mask, vint64m8_t maskedoff, vint32m4_t op1, uint32_t op2, size_t vl) {
+  return vwmulsu_vx_i64m8_m(mask, maskedoff, op1, op2, vl);
+}