[RISCV] Starting fixing issues that prevent us from testing vXi64 intrinsics on RV32.

[craig.topper]

Summary

Currently we crash in type legalization any time an intrinsic
uses a scalar i64 on RV32.

This patch adds support for type legalizing this to prevent
crashing. I don't promise that it uses the best possible codegen
just that it is functional.

This first version handles 3 cases. vmv.v.x intrinsic, vmv.s.x
intrinsic and intrinsics that take a scalar input, splat it and
then do some operation.

For vmv.v.x we'll either rely on hardware sign extension for
constants or we'll convert it to multiple splats and bit
manipulation.

For vmv.s.x we use a really unoptimal sequence inspired by what
we do for an INSERT_VECTOR_ELT.

For the third case we'll either try to use the .vi form for
constants or convert to a complicated splat and bitmanip and use
the .vv form of the operation.

I've renamed the ExtendOperand field to SplatOperand now use it
specifically for the third case. The first two cases are handled
by custom lowering specifically for those intrinsics.

I haven't updated all tests yet, but I tried to cover a subset
that includes single-width, widening, and narrowing.

[craig.topper]

Fix a typo and clang-format.

[craig.topper]

llvm/lib/Target/RISCV/RISCVISelLowering.cpp

2479

I was thinking maybe we just need two slide1ups using SEW=32 with VL set to 2 so that we don't slide anything but the scalars we're inserting.

llvm/lib/Target/RISCV/RISCVISelLowering.h

102

I think @frasercrmck has a patch that adds VMV_S_XF_VL which I think I can migrate to once its committed.

[frasercrmck]

llvm/lib/Target/RISCV/RISCVISelLowering.cpp

2479

Crafty; I like it. Doing that later along with INSERT_VECTOR_ELT would be my preferred way to go.

llvm/lib/Target/RISCV/RISCVISelLowering.h

102

Yeah you should be good to go with that now.

[craig.topper]

llvm/lib/Target/RISCV/RISCVISelLowering.cpp

2479

I was also wondering if we could do something like this for the splat

vmv.v.x vX, hi // using SEW=64
vsll.vx vX, vX, /*32*/ // clear the lower 32 bits like we're doing now.
vsetvli e32 // same vl with half the lmul of the 64 bit type.
vaddu.wx vX, vX, lo // should zero extend the lo value to 64 bits by zero extending. Since we cleared the lower 32 bits above this is equivalent to OR.
vsetvli e64 // mask to original sew/lmul

The nice advantage it has is that it can be done in one physical register or physical register group. The current sequence requires two.

[craig.topper]

Remove custom node now that Fraser's patch is in

[frasercrmck]

llvm/lib/Target/RISCV/RISCVISelLowering.cpp

2479

Sounds like that'd work, yep.

Off the top of my head, could you not also extend the INSERT_VECTOR_ELT sequence with a vrgather.vi vd, vs2, 0 to splat the first element? Perhaps fewer instructions but you'd still that second register due to the non-overlap constraint. I'm not sure which is better: perhaps it's situational.

[craig.topper]

llvm/lib/Target/RISCV/RISCVISelLowering.cpp

2479

Off the top of my head, could you not also extend the INSERT_VECTOR_ELT sequence with a vrgather.vi vd, vs2, 0 to splat the first element? Perhaps fewer instructions but you'd still that second register due to the non-overlap constraint. I'm not sure which is better: perhaps it's situational.

I'm not following. INSERT_ELEMENT has a scalar register input input. How did it get to element 0 for vrgather.vi vd, vs2, 0?

[frasercrmck]

llvm/lib/Target/RISCV/RISCVISelLowering.cpp

2479

Ah sorry I might have skipped a step. I was proposing extending the two slide1up sequence you proposed earlier to get it from element zero to all elements.

[craig.topper]

llvm/lib/Target/RISCV/RISCVISelLowering.cpp

2479

Ok I get it now. When I wrote my slide1up suggestion I was only thinking about this case where element 0 was the final location.

[frasercrmck]

Anyway, LGTM.

llvm/include/llvm/IR/IntrinsicsRISCV.td

 //===----------------------------------------------------------------------===//
 // Vectors
 
 class RISCVVIntrinsic {
   // These intrinsics may accept illegal integer values in their llvm_any_ty
   // operand, so they have to be extended. If set to zero then the intrinsic
   // does not have any operand that must be extended.
   Intrinsic IntrinsicID = !cast<Intrinsic>(NAME);
-  bits<4> ExtendOperand = 0;
+  bits<4> SplatOperand = 0;
 }
 
 let TargetPrefix = "riscv" in {
   // We use anyint here but we only support XLen.
   def int_riscv_vsetvli   : Intrinsic<[llvm_anyint_ty],
                            /* AVL */  [LLVMMatchType<0>,
                            /* VSEW */  LLVMMatchType<0>,
                            /* VLMUL */ LLVMMatchType<0>],
class RISCVGatherVXMask
                    [IntrNoMem]>, RISCVVIntrinsic {
   }
   // For destination vector type is the same as first source vector.
   // Input: (vector_in, vector_in/scalar_in, vl)
   class RISCVBinaryAAXNoMask
         : Intrinsic<[llvm_anyvector_ty],
                     [LLVMMatchType<0>, llvm_any_ty, llvm_anyint_ty],
                     [IntrNoMem]>, RISCVVIntrinsic {
-    let ExtendOperand = 2;
+    let SplatOperand = 2;
   }
   // For destination vector type is the same as first source vector (with mask).
   // Input: (maskedoff, vector_in, vector_in/scalar_in, mask, vl)
   class RISCVBinaryAAXMask
        : Intrinsic<[llvm_anyvector_ty],
                    [LLVMMatchType<0>, LLVMMatchType<0>, llvm_any_ty,
                     LLVMScalarOrSameVectorWidth<0, llvm_i1_ty>, llvm_anyint_ty],
                    [IntrNoMem]>, RISCVVIntrinsic {
-    let ExtendOperand = 3;
+    let SplatOperand = 3;
   }
   // For destination vector type is NOT the same as first source vector.
   // Input: (vector_in, vector_in/scalar_in, vl)
   class RISCVBinaryABXNoMask
         : Intrinsic<[llvm_anyvector_ty],
                     [llvm_anyvector_ty, llvm_any_ty, llvm_anyint_ty],
                     [IntrNoMem]>, RISCVVIntrinsic {
-    let ExtendOperand = 2;
+    let SplatOperand = 2;
   }
   // For destination vector type is NOT the same as first source vector (with mask).
   // Input: (maskedoff, vector_in, vector_in/scalar_in, mask, vl)
   class RISCVBinaryABXMask
         : Intrinsic<[llvm_anyvector_ty],
                     [LLVMMatchType<0>, llvm_anyvector_ty, llvm_any_ty,
                      LLVMScalarOrSameVectorWidth<0, llvm_i1_ty>, llvm_anyint_ty],
                     [IntrNoMem]>, RISCVVIntrinsic {
-    let ExtendOperand = 3;
+    let SplatOperand = 3;
   }
   // For binary operations with V0 as input.
   // Input: (vector_in, vector_in/scalar_in, V0, vl)
   class RISCVBinaryWithV0
         : Intrinsic<[llvm_anyvector_ty],
                     [LLVMMatchType<0>, llvm_any_ty,
                      LLVMScalarOrSameVectorWidth<0, llvm_i1_ty>,
                      llvm_anyint_ty],
                     [IntrNoMem]>, RISCVVIntrinsic {
-    let ExtendOperand = 2;
+    let SplatOperand = 2;
   }
   // For binary operations with mask type output and V0 as input.
   // Output: (mask type output)
   // Input: (vector_in, vector_in/scalar_in, V0, vl)
   class RISCVBinaryMOutWithV0
         :Intrinsic<[LLVMScalarOrSameVectorWidth<0, llvm_i1_ty>],
                    [llvm_anyvector_ty, llvm_any_ty,
                     LLVMScalarOrSameVectorWidth<0, llvm_i1_ty>,
                     llvm_anyint_ty],
                    [IntrNoMem]>, RISCVVIntrinsic {
-    let ExtendOperand = 2;
+    let SplatOperand = 2;
   }
   // For binary operations with mask type output.
   // Output: (mask type output)
   // Input: (vector_in, vector_in/scalar_in, vl)
   class RISCVBinaryMOut
         : Intrinsic<[LLVMScalarOrSameVectorWidth<0, llvm_i1_ty>],
                     [llvm_anyvector_ty, llvm_any_ty, llvm_anyint_ty],
                     [IntrNoMem]>, RISCVVIntrinsic {
-    let ExtendOperand = 2;
+    let SplatOperand = 2;
   }
   // For binary operations with mask type output without mask.
   // Output: (mask type output)
   // Input: (vector_in, vector_in/scalar_in, vl)
   class RISCVCompareNoMask
         : Intrinsic<[LLVMScalarOrSameVectorWidth<0, llvm_i1_ty>],
                     [llvm_anyvector_ty, llvm_any_ty, llvm_anyint_ty],
                     [IntrNoMem]>, RISCVVIntrinsic {
-    let ExtendOperand = 2;
+    let SplatOperand = 2;
   }
   // For binary operations with mask type output with mask.
   // Output: (mask type output)
   // Input: (maskedoff, vector_in, vector_in/scalar_in, mask, vl)
   class RISCVCompareMask
         : Intrinsic<[LLVMScalarOrSameVectorWidth<0, llvm_i1_ty>],
                     [LLVMScalarOrSameVectorWidth<0, llvm_i1_ty>,
                      llvm_anyvector_ty, llvm_any_ty,
                      LLVMScalarOrSameVectorWidth<0, llvm_i1_ty>, llvm_anyint_ty],
                     [IntrNoMem]>, RISCVVIntrinsic {
-    let ExtendOperand = 3;
+    let SplatOperand = 3;
   }
   // For FP classify operations.
   // Output: (bit mask type output)
   // Input: (vector_in, vl)
   class RISCVClassifyNoMask
         : Intrinsic<[LLVMVectorOfBitcastsToInt<0>],
                     [llvm_anyvector_ty, llvm_anyint_ty],
                     [IntrNoMem]>, RISCVVIntrinsic;
   // For FP classify operations with mask.
   // Output: (bit mask type output)
   // Input: (maskedoff, vector_in, mask, vl)
   class RISCVClassifyMask
         : Intrinsic<[LLVMVectorOfBitcastsToInt<0>],
                     [LLVMVectorOfBitcastsToInt<0>, llvm_anyvector_ty,
                      LLVMScalarOrSameVectorWidth<0, llvm_i1_ty>, llvm_anyint_ty],
                     [IntrNoMem]>, RISCVVIntrinsic;
   // For Saturating binary operations.
   // The destination vector type is the same as first source vector.
   // Input: (vector_in, vector_in/scalar_in, vl)
   class RISCVSaturatingBinaryAAXNoMask
         : Intrinsic<[llvm_anyvector_ty],
                     [LLVMMatchType<0>, llvm_any_ty, llvm_anyint_ty],
                     [IntrNoMem, IntrHasSideEffects]>, RISCVVIntrinsic {
-    let ExtendOperand = 2;
+    let SplatOperand = 2;
   }
   // For Saturating binary operations with mask.
   // The destination vector type is the same as first source vector.
   // Input: (maskedoff, vector_in, vector_in/scalar_in, mask, vl)
   class RISCVSaturatingBinaryAAXMask
         : Intrinsic<[llvm_anyvector_ty],
                     [LLVMMatchType<0>, LLVMMatchType<0>, llvm_any_ty,
                      LLVMScalarOrSameVectorWidth<0, llvm_i1_ty>, llvm_anyint_ty],
                     [IntrNoMem, IntrHasSideEffects]>, RISCVVIntrinsic {
-    let ExtendOperand = 3;
+    let SplatOperand = 3;
   }
   // For Saturating binary operations.
   // The destination vector type is NOT the same as first source vector.
   // Input: (vector_in, vector_in/scalar_in, vl)
   class RISCVSaturatingBinaryABXNoMask
         : Intrinsic<[llvm_anyvector_ty],
                     [llvm_anyvector_ty, llvm_any_ty, llvm_anyint_ty],
                     [IntrNoMem, IntrHasSideEffects]>, RISCVVIntrinsic {
-    let ExtendOperand = 2;
+    let SplatOperand = 2;
   }
   // For Saturating binary operations with mask.
   // The destination vector type is NOT the same as first source vector (with mask).
   // Input: (maskedoff, vector_in, vector_in/scalar_in, mask, vl)
   class RISCVSaturatingBinaryABXMask
         : Intrinsic<[llvm_anyvector_ty],
                     [LLVMMatchType<0>, llvm_anyvector_ty, llvm_any_ty,
                      LLVMScalarOrSameVectorWidth<0, llvm_i1_ty>, llvm_anyint_ty],
                     [IntrNoMem, IntrHasSideEffects]>, RISCVVIntrinsic {
-    let ExtendOperand = 3;
+    let SplatOperand = 3;
   }
   class RISCVTernaryAAAXNoMask
         : Intrinsic<[llvm_anyvector_ty],
                     [LLVMMatchType<0>, LLVMMatchType<0>, llvm_anyint_ty,
                      LLVMMatchType<1>],
                     [IntrNoMem]>, RISCVVIntrinsic;
   class RISCVTernaryAAAXMask
         : Intrinsic<[llvm_anyvector_ty],
                     [LLVMMatchType<0>, LLVMMatchType<0>, llvm_anyint_ty,
                      LLVMScalarOrSameVectorWidth<0, llvm_i1_ty>, LLVMMatchType<1>],
                     [IntrNoMem]>, RISCVVIntrinsic;
   class RISCVTernaryAAXANoMask
         : Intrinsic<[llvm_anyvector_ty],
                     [LLVMMatchType<0>, llvm_any_ty, LLVMMatchType<0>,
                      llvm_anyint_ty],
                     [IntrNoMem]>, RISCVVIntrinsic {
-    let ExtendOperand = 2;
+    let SplatOperand = 2;
   }
   class RISCVTernaryAAXAMask
         : Intrinsic<[llvm_anyvector_ty],
                     [LLVMMatchType<0>, llvm_any_ty, LLVMMatchType<0>,
                      LLVMScalarOrSameVectorWidth<0, llvm_i1_ty>, llvm_anyint_ty],
                     [IntrNoMem]>, RISCVVIntrinsic {
-    let ExtendOperand = 2;
+    let SplatOperand = 2;
   }
   class RISCVTernaryWideNoMask
         : Intrinsic< [llvm_anyvector_ty],
                      [LLVMMatchType<0>, llvm_any_ty, llvm_anyvector_ty,
                       llvm_anyint_ty],
                      [IntrNoMem] >, RISCVVIntrinsic {
-    let ExtendOperand = 2;
+    let SplatOperand = 2;
   }
   class RISCVTernaryWideMask
         : Intrinsic< [llvm_anyvector_ty],
                      [LLVMMatchType<0>, llvm_any_ty, llvm_anyvector_ty,
                       LLVMScalarOrSameVectorWidth<0, llvm_i1_ty>, llvm_anyint_ty],
                      [IntrNoMem]>, RISCVVIntrinsic {
-    let ExtendOperand = 2;
+    let SplatOperand = 2;
   }
   // For Reduction ternary operations.
   // For destination vector type is the same as first and third source vector.
   // Input: (vector_in, vector_in, vector_in, vl)
   class RISCVReductionNoMask
         : Intrinsic<[llvm_anyvector_ty],
                     [LLVMMatchType<0>, llvm_anyvector_ty, LLVMMatchType<0>,
                      llvm_anyint_ty],
let TargetPrefix = "riscv" in {
 
   def int_riscv_vmerge : RISCVBinaryWithV0;
 
   def int_riscv_vmv_v_v : Intrinsic<[llvm_anyvector_ty],
                                     [LLVMMatchType<0>, llvm_anyint_ty],
                                     [IntrNoMem]>, RISCVVIntrinsic;
   def int_riscv_vmv_v_x : Intrinsic<[llvm_anyint_ty],
                                     [LLVMVectorElementType<0>, llvm_anyint_ty],
-                                    [IntrNoMem]>, RISCVVIntrinsic {
-    let ExtendOperand = 1;
-  }
+                                    [IntrNoMem]>, RISCVVIntrinsic;
   def int_riscv_vfmv_v_f : Intrinsic<[llvm_anyfloat_ty],
                                      [LLVMVectorElementType<0>, llvm_anyint_ty],
                                      [IntrNoMem]>, RISCVVIntrinsic;
 
   def int_riscv_vmv_x_s : Intrinsic<[LLVMVectorElementType<0>],
                                     [llvm_anyint_ty],
                                     [IntrNoMem]>, RISCVVIntrinsic;
   def int_riscv_vmv_s_x : Intrinsic<[llvm_anyint_ty],
                                     [LLVMMatchType<0>, LLVMVectorElementType<0>,
                                      llvm_anyint_ty],
-                                    [IntrNoMem]>, RISCVVIntrinsic {
-    let ExtendOperand = 2;
-  }
+                                    [IntrNoMem]>, RISCVVIntrinsic;
 
   def int_riscv_vfmv_f_s : Intrinsic<[LLVMVectorElementType<0>],
                                      [llvm_anyfloat_ty],
                                      [IntrNoMem]>, RISCVVIntrinsic;
   def int_riscv_vfmv_s_f : Intrinsic<[llvm_anyfloat_ty],
                                      [LLVMMatchType<0>, LLVMVectorElementType<0>,
                                       llvm_anyint_ty],
                                      [IntrNoMem]>, RISCVVIntrinsic;

llvm/lib/Target/RISCV/RISCVISelLowering.h

enum NodeType : unsigned {
   // VMV_V_X_VL matches the semantics of vmv.v.x but includes an extra operand
   // for the VL value to be used for the operation.
   VMV_V_X_VL,
   // VFMV_V_F_VL matches the semantics of vfmv.v.f but includes an extra operand
   // for the VL value to be used for the operation.
   VFMV_V_F_VL,
   // VMV_X_S matches the semantics of vmv.x.s. The result is always XLenVT sign
   // extended from the vector element size.
   VMV_X_S,
   // VMV_S_XF_VL matches the semantics of vmv.s.x/vmv.s.f, depending on the
   // types of its operands. It carries a VL operand.
   VMV_S_XF_VL,
   // Splats an i64 scalar to a vector type (with element type i64) where the
   // scalar is a sign-extended i32.
   SPLAT_VECTOR_I64,
   // Read VLENB CSR
   READ_VLENB,

 // We use 64 bits as the known part in the scalable vector types.
 static constexpr unsigned RVVBitsPerBlock = 64;
 } // namespace RISCV
 
 namespace RISCVVIntrinsicsTable {
 
 struct RISCVVIntrinsicInfo {
   unsigned IntrinsicID;
-  uint8_t ExtendedOperand;
+  uint8_t SplatOperand;
 };
 
 using namespace RISCV;
 
 #define GET_RISCVVIntrinsicsTable_DECL
 #include "RISCVGenSearchableTables.inc"
 
 } // end namespace RISCVVIntrinsicsTable
 
 } // end namespace llvm
 
 #endif

llvm/lib/Target/RISCV/RISCVISelLowering.cpp

if (Subtarget.hasStdExtV()) {
     // RVV intrinsics may have illegal operands.
     // We also need to custom legalize vmv.x.s.
     setOperationAction(ISD::INTRINSIC_WO_CHAIN, MVT::i8, Custom);
     setOperationAction(ISD::INTRINSIC_WO_CHAIN, MVT::i16, Custom);
     setOperationAction(ISD::INTRINSIC_W_CHAIN, MVT::i8, Custom);
     setOperationAction(ISD::INTRINSIC_W_CHAIN, MVT::i16, Custom);
     setOperationAction(ISD::INTRINSIC_WO_CHAIN, MVT::i32, Custom);
     setOperationAction(ISD::INTRINSIC_W_CHAIN, MVT::i32, Custom);
+    setOperationAction(ISD::INTRINSIC_WO_CHAIN, MVT::i64, Custom);
+    setOperationAction(ISD::INTRINSIC_W_CHAIN, MVT::i64, Custom);
 
     setOperationAction(ISD::INTRINSIC_W_CHAIN, MVT::Other, Custom);
 
-    if (Subtarget.is64Bit()) {
-      setOperationAction(ISD::INTRINSIC_WO_CHAIN, MVT::i64, Custom);
-      setOperationAction(ISD::INTRINSIC_W_CHAIN, MVT::i64, Custom);
-    } else {
+    if (!Subtarget.is64Bit()) {
       // We must custom-lower certain vXi64 operations on RV32 due to the vector
       // element type being illegal.
       setOperationAction(ISD::SPLAT_VECTOR, MVT::i64, Custom);
       setOperationAction(ISD::INSERT_VECTOR_ELT, MVT::i64, Custom);
       setOperationAction(ISD::EXTRACT_VECTOR_ELT, MVT::i64, Custom);
 
       setOperationAction(ISD::VECREDUCE_ADD, MVT::i64, Custom);
       setOperationAction(ISD::VECREDUCE_AND, MVT::i64, Custom);
if (!EltVT.isInteger()) {
     return DAG.getNode(ISD::EXTRACT_VECTOR_ELT, DL, EltVT, Vec,
                        DAG.getConstant(0, DL, XLenVT));
   }
 
   SDValue Elt0 = DAG.getNode(RISCVISD::VMV_X_S, DL, XLenVT, Vec);
   return DAG.getNode(ISD::TRUNCATE, DL, EltVT, Elt0);
 }
 
-SDValue RISCVTargetLowering::LowerINTRINSIC_WO_CHAIN(SDValue Op,
-                                                     SelectionDAG &DAG) const {
-  unsigned IntNo = cast<ConstantSDNode>(Op.getOperand(0))->getZExtValue();
-  SDLoc DL(Op);
+// Called by type legalization to handle splat of i64 on RV32.
+// FIXME: We can optimize this when the type has sign or zero bits in one
+// of the halves.
+static SDValue splatSplitI64WithVL(const SDLoc &DL, MVT VT, SDValue Scalar,
+                                   SDValue VL, SelectionDAG &DAG) {
+  SDValue ThirtyTwoV = DAG.getConstant(32, DL, VT);
+  SDValue Lo = DAG.getNode(ISD::EXTRACT_ELEMENT, DL, MVT::i32, Scalar,
+                           DAG.getConstant(0, DL, MVT::i32));
+  SDValue Hi = DAG.getNode(ISD::EXTRACT_ELEMENT, DL, MVT::i32, Scalar,
+                           DAG.getConstant(1, DL, MVT::i32));
+
+  // vmv.v.x vX, hi
+  // vsll.vx vX, vX, /*32*/
+  // vmv.v.x vY, lo
+  // vsll.vx vY, vY, /*32*/
+  // vsrl.vx vY, vY, /*32*/
+  // vor.vv vX, vX, vY
+  MVT MaskVT = MVT::getVectorVT(MVT::i1, VT.getVectorElementCount());
+  SDValue Mask = DAG.getNode(RISCVISD::VMSET_VL, DL, MaskVT, VL);
+  Lo = DAG.getNode(RISCVISD::VMV_V_X_VL, DL, VT, Lo, VL);
+  Lo = DAG.getNode(RISCVISD::SHL_VL, DL, VT, Lo, ThirtyTwoV, Mask, VL);
+  Lo = DAG.getNode(RISCVISD::SRL_VL, DL, VT, Lo, ThirtyTwoV, Mask, VL);
+
+  Hi = DAG.getNode(RISCVISD::VMV_V_X_VL, DL, VT, Hi, VL);
+  Hi = DAG.getNode(RISCVISD::SHL_VL, DL, VT, Hi, ThirtyTwoV, Mask, VL);
+
+  return DAG.getNode(RISCVISD::OR_VL, DL, VT, Lo, Hi, Mask, VL);
+}
 
-  if (Subtarget.hasStdExtV()) {
-    // Some RVV intrinsics may claim that they want an integer operand to be
-    // extended.
-    if (const RISCVVIntrinsicsTable::RISCVVIntrinsicInfo *II =
-            RISCVVIntrinsicsTable::getRISCVVIntrinsicInfo(IntNo)) {
-      if (II->ExtendedOperand) {
-        assert(II->ExtendedOperand < Op.getNumOperands());
-        SmallVector<SDValue, 8> Operands(Op->op_begin(), Op->op_end());
-        SDValue &ScalarOp = Operands[II->ExtendedOperand];
-        EVT OpVT = ScalarOp.getValueType();
-        if (OpVT == MVT::i8 || OpVT == MVT::i16 ||
-            (OpVT == MVT::i32 && Subtarget.is64Bit())) {
-          // If the operand is a constant, sign extend to increase our chances
-          // of being able to use a .vi instruction. ANY_EXTEND would become a
-          // a zero extend and the simm5 check in isel would fail.
-          // FIXME: Should we ignore the upper bits in isel instead?
-          unsigned ExtOpc = isa<ConstantSDNode>(ScalarOp) ? ISD::SIGN_EXTEND
-                                                          : ISD::ANY_EXTEND;
-          ScalarOp = DAG.getNode(ExtOpc, DL, Subtarget.getXLenVT(), ScalarOp);
-          return DAG.getNode(ISD::INTRINSIC_WO_CHAIN, DL, Op.getValueType(),
-                             Operands);
-        }
-      }
-    }
-  }
+// promoted or expanded.
+static SDValue lowerVectorIntrinsicSplats(SDValue Op, SelectionDAG &DAG,
+                                          const RISCVSubtarget &Subtarget) {
+  assert((Op.getOpcode() == ISD::INTRINSIC_WO_CHAIN ||
+          Op.getOpcode() == ISD::INTRINSIC_W_CHAIN) &&
+         "Unexpected opcode");
+
+  if (!Subtarget.hasStdExtV())
+    return SDValue();
+
+  bool HasChain = Op.getOpcode() == ISD::INTRINSIC_W_CHAIN;
+  unsigned IntNo = Op.getConstantOperandVal(HasChain ? 1 : 0);
+  SDLoc DL(Op);
+
+  const RISCVVIntrinsicsTable::RISCVVIntrinsicInfo *II =
+      RISCVVIntrinsicsTable::getRISCVVIntrinsicInfo(IntNo);
+  if (!II || !II->SplatOperand)
+    return SDValue();
+
+  unsigned SplatOp = II->SplatOperand + HasChain;
+  assert(SplatOp < Op.getNumOperands());
+
+  SDValue &ScalarOp = Operands[SplatOp];
+  MVT OpVT = ScalarOp.getSimpleValueType();
+  MVT VT = Op.getSimpleValueType();
+  MVT XLenVT = Subtarget.getXLenVT();
+
+  // If this isn't a scalar, or its type is XLenVT we're done.
+  if (!OpVT.isScalarInteger() || OpVT == XLenVT)
+    return SDValue();
+
+  // Simplest case is that the operand needs to be promoted to XLenVT.
+  if (OpVT.bitsLT(XLenVT)) {
+    unsigned ExtOpc =
+        isa<ConstantSDNode>(ScalarOp) ? ISD::SIGN_EXTEND : ISD::ANY_EXTEND;
+    ScalarOp = DAG.getNode(ExtOpc, DL, XLenVT, ScalarOp);
+    return DAG.getNode(Op->getOpcode(), DL, Op->getVTList(), Operands);
+  }
+
+  // The more complex case is when the scalar is larger than XLenVT.
+  assert(XLenVT == MVT::i32 && OpVT == MVT::i64 &&
+         VT.getVectorElementType() == MVT::i64 && "Unexpected VTs!");
+
+  // If this is a sign-extended 32-bit constant, we can truncate it and rely
+  // on the instruction to sign-extend since SEW>XLEN.
+  if (auto *CVal = dyn_cast<ConstantSDNode>(ScalarOp)) {
+    if (isInt<32>(CVal->getSExtValue())) {
+      ScalarOp = DAG.getConstant(CVal->getSExtValue(), DL, MVT::i32);
+      return DAG.getNode(Op->getOpcode(), DL, Op->getVTList(), Operands);
+
+  // We need to convert the scalar to a splat vector.
+  // FIXME: Can we implicitly truncate the scalar if it is known to
+  // be sign extended?
+  // VL should be the last operand.
+  SDValue VL = Op.getOperand(Op.getNumOperands() - 1);
+  assert(VL.getValueType() == XLenVT);
+  ScalarOp = splatSplitI64WithVL(DL, VT, ScalarOp, VL, DAG);
+  return DAG.getNode(Op->getOpcode(), DL, Op->getVTList(), Operands);
 
+SDValue RISCVTargetLowering::LowerINTRINSIC_WO_CHAIN(SDValue Op,
+                                                     SelectionDAG &DAG) const {
+  unsigned IntNo = Op.getConstantOperandVal(0);
+  SDLoc DL(Op);
+  MVT XLenVT = Subtarget.getXLenVT();
+
   switch (IntNo) {
   default:
-    return SDValue();    // Don't custom lower most intrinsics.
+    break; // Don't custom lower most intrinsics.
   case Intrinsic::thread_pointer: {
     EVT PtrVT = getPointerTy(DAG.getDataLayout());
     return DAG.getRegister(RISCV::X4, PtrVT);
   }
   case Intrinsic::riscv_vmv_x_s:
-    assert(Op.getValueType() == Subtarget.getXLenVT() && "Unexpected VT!");
+    assert(Op.getValueType() == XLenVT && "Unexpected VT!");
     return DAG.getNode(RISCVISD::VMV_X_S, DL, Op.getValueType(),
                        Op.getOperand(1));
   case Intrinsic::riscv_vmv_v_x: {
-    SDValue Scalar = DAG.getNode(ISD::ANY_EXTEND, DL, Subtarget.getXLenVT(),
-                                 Op.getOperand(1));
+    SDValue Scalar = Op.getOperand(1);
+    if (Scalar.getValueType().bitsLE(XLenVT)) {
+      unsigned ExtOpc =
+          isa<ConstantSDNode>(Scalar) ? ISD::SIGN_EXTEND : ISD::ANY_EXTEND;
+      Scalar = DAG.getNode(ExtOpc, DL, XLenVT, Scalar);
+      return DAG.getNode(RISCVISD::VMV_V_X_VL, DL, Op.getValueType(), Scalar,
+                         Op.getOperand(2));
+    }
+
+    assert(Scalar.getValueType() == MVT::i64 && "Unexpected scalar VT!");
+
+    // If this is a sign-extended 32-bit constant, we can truncate it and rely
+    // on the instruction to sign-extend since SEW>XLEN.
+    if (auto *CVal = dyn_cast<ConstantSDNode>(Scalar)) {
+      if (isInt<32>(CVal->getSExtValue()))
         return DAG.getNode(RISCVISD::VMV_V_X_VL, DL, Op.getValueType(),
-                       Scalar, Op.getOperand(2));
+                           DAG.getConstant(CVal->getSExtValue(), DL, MVT::i32),
+                           Op.getOperand(2));
+    }
+
+    // Otherwise use the more complicated splatting algorithm.
+    return splatSplitI64WithVL(DL, Op.getSimpleValueType(), Scalar,
+                               Op.getOperand(2), DAG);
   }
   case Intrinsic::riscv_vfmv_v_f:
     return DAG.getNode(RISCVISD::VFMV_V_F_VL, DL, Op.getValueType(),
                        Op.getOperand(1), Op.getOperand(2));
-  }
-}
+  case Intrinsic::riscv_vmv_s_x: {
+    SDValue Scalar = Op.getOperand(2);
 
-SDValue RISCVTargetLowering::LowerINTRINSIC_W_CHAIN(SDValue Op,
-                                                    SelectionDAG &DAG) const {
-  unsigned IntNo = cast<ConstantSDNode>(Op.getOperand(1))->getZExtValue();
-  SDLoc DL(Op);
+    if (Scalar.getValueType().bitsLE(XLenVT)) {
+      Scalar = DAG.getNode(ISD::ANY_EXTEND, DL, XLenVT, Scalar);
+      return DAG.getNode(RISCVISD::VMV_S_XF_VL, DL, Op.getValueType(),
+                         Op.getOperand(1), Scalar, Op.getOperand(3));
+    }
+
+    assert(Scalar.getValueType() == MVT::i64 && "Unexpected scalar VT!");
+
+    // This is an i64 value that lives in two scalar registers. We have to
+    // insert this in a convoluted way. First we build vXi64 splat containing
+    // the/ two values that we assemble using some bit math. Next we'll use
+    // vid.v and vmseq to build a mask with bit 0 set. Then we'll use that mask
+    // to merge element 0 from our splat into the source vector.
+    // FIXME: This is probably not the best way to do this, but it is
+    // consistent with INSERT_VECTOR_ELT lowering so it is a good starting
+    // point.
+    //   vmv.v.x vX, hi
+    //   vsll.vx vX, vX, /*32*/
+    //   vmv.v.x vY, lo
+    //   vsll.vx vY, vY, /*32*/
+    //   vsrl.vx vY, vY, /*32*/
+    //   vor.vv vX, vX, vY
+    //
+    //   vid.v      vVid
+    //   vmseq.vx   mMask, vVid, 0
+    //   vmerge.vvm vDest, vSrc, vVal, mMask
+    MVT VT = Op.getSimpleValueType();
+    SDValue Vec = Op.getOperand(1);
+    SDValue VL = Op.getOperand(3);
 
-  if (Subtarget.hasStdExtV()) {
-    // Some RVV intrinsics may claim that they want an integer operand to be
-    // extended.
-    if (const RISCVVIntrinsicsTable::RISCVVIntrinsicInfo *II =
-            RISCVVIntrinsicsTable::getRISCVVIntrinsicInfo(IntNo)) {
-      if (II->ExtendedOperand) {
-        // The operands start from the second argument in INTRINSIC_W_CHAIN.
-        unsigned ExtendOp = II->ExtendedOperand + 1;
-        assert(ExtendOp < Op.getNumOperands());
-        SmallVector<SDValue, 8> Operands(Op->op_begin(), Op->op_end());
-        SDValue &ScalarOp = Operands[ExtendOp];
-        EVT OpVT = ScalarOp.getValueType();
-        if (OpVT == MVT::i8 || OpVT == MVT::i16 ||
-            (OpVT == MVT::i32 && Subtarget.is64Bit())) {
-          // If the operand is a constant, sign extend to increase our chances
-          // of being able to use a .vi instruction. ANY_EXTEND would become a
-          // a zero extend and the simm5 check in isel would fail.
-          // FIXME: Should we ignore the upper bits in isel instead?
-          unsigned ExtOpc = isa<ConstantSDNode>(ScalarOp) ? ISD::SIGN_EXTEND
-                                                          : ISD::ANY_EXTEND;
-          ScalarOp = DAG.getNode(ExtOpc, DL, Subtarget.getXLenVT(), ScalarOp);
-          return DAG.getNode(ISD::INTRINSIC_W_CHAIN, DL, Op->getVTList(),
-                             Operands);
-        }
-      }
-    }
-  }
+    SDValue SplattedVal = splatSplitI64WithVL(DL, VT, Scalar, VL, DAG);
+    SDValue SplattedIdx = DAG.getNode(RISCVISD::VMV_V_X_VL, DL, VT,
+                                      DAG.getConstant(0, DL, MVT::i32), VL);
+
+    MVT MaskVT = MVT::getVectorVT(MVT::i1, VT.getVectorElementCount());
+    SDValue Mask = DAG.getNode(RISCVISD::VMSET_VL, DL, MaskVT, VL);
+    SDValue VID = DAG.getNode(RISCVISD::VID_VL, DL, VT, Mask, VL);
+    SDValue SelectCond =
+        DAG.getNode(RISCVISD::SETCC_VL, DL, MaskVT, VID, SplattedIdx,
+                    DAG.getCondCode(ISD::SETEQ), Mask, VL);
+    return DAG.getNode(RISCVISD::VSELECT_VL, DL, VT, SelectCond, SplattedVal,
+                       Vec, VL);
+
+  return lowerVectorIntrinsicSplats(Op, DAG, Subtarget);
 
-  return SDValue(); // Don't custom lower most intrinsics.
+SDValue RISCVTargetLowering::LowerINTRINSIC_W_CHAIN(SDValue Op,
+                                                    SelectionDAG &DAG) const {
+  return lowerVectorIntrinsicSplats(Op, DAG, Subtarget);
 }
 
 static MVT getLMUL1VT(MVT VT) {
   assert(VT.getVectorElementType().getSizeInBits() <= 64 &&
          "Unexpected vector MVT");
   return MVT::getScalableVectorVT(
       VT.getVectorElementType(),
       RISCV::RVVBitsPerBlock / VT.getVectorElementType().getSizeInBits());

llvm/lib/Target/RISCV/RISCVInstrInfoVPseudos.td

def RISCVVPseudosTable : GenericTable {
   let PrimaryKey = [ "Pseudo" ];
   let PrimaryKeyName = "getPseudoInfo";
   let PrimaryKeyEarlyOut = true;
 }
 
 def RISCVVIntrinsicsTable : GenericTable {
   let FilterClass = "RISCVVIntrinsic";
   let CppTypeName = "RISCVVIntrinsicInfo";
-  let Fields = ["IntrinsicID", "ExtendOperand"];
+  let Fields = ["IntrinsicID", "SplatOperand"];
   let PrimaryKey = ["IntrinsicID"];
   let PrimaryKeyName = "getRISCVVIntrinsicInfo";
 }
 
 class RISCVVLE<bit M, bit Str, bit F, bits<7> S, bits<3> L> {
   bits<1> Masked = M;
   bits<1> Strided = Str;
   bits<1> FF = F;

 //===----------------------------------------------------------------------===//
 // 17.1. Integer Scalar Move Instructions
 //===----------------------------------------------------------------------===//
 
 let Predicates = [HasStdExtV] in {
 foreach vti = AllIntegerVectors in {
   def : Pat<(riscv_vmv_x_s (vti.Vector vti.RegClass:$rs2)),
             (!cast<Instruction>("PseudoVMV_X_S_" # vti.LMul.MX) $rs2, vti.SEW)>;
-  def : Pat<(vti.Vector (int_riscv_vmv_s_x (vti.Vector vti.RegClass:$rs1),
-                                           GPR:$rs2, (XLenVT (VLOp GPR:$vl)))),
-            (!cast<Instruction>("PseudoVMV_S_X_" # vti.LMul.MX)
-             (vti.Vector $rs1), $rs2, GPR:$vl, vti.SEW)>;
+  // vmv.s.x is handled with a custom node in RISCVInstrInfoVVLPatterns.td
 }
 } // Predicates = [HasStdExtV]
 
 //===----------------------------------------------------------------------===//
 // 17.2. Floating-Point Scalar Move Instructions
 //===----------------------------------------------------------------------===//
 
 let Predicates = [HasStdExtV, HasStdExtF] in {

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

%a = call <vscale x 16 x i32> @llvm.riscv.vaadd.mask.nxv16i32.nxv16i32(
     <vscale x 16 x i32> %1,
     <vscale x 16 x i32> %2,
     <vscale x 16 x i1> %3,
     i32 %4)
 
   ret <vscale x 16 x i32> %a
 }
 
+declare <vscale x 1 x i64> @llvm.riscv.vaadd.nxv1i64.nxv1i64(
+  <vscale x 1 x i64>,
+  <vscale x 1 x i64>,
+  i32);
+
+define <vscale x 1 x i64> @intrinsic_vaadd_vv_nxv1i64_nxv1i64_nxv1i64(<vscale x 1 x i64> %0, <vscale x 1 x i64> %1, i32 %2) nounwind {
+; CHECK-LABEL: intrinsic_vaadd_vv_nxv1i64_nxv1i64_nxv1i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e64,m1,ta,mu
+; CHECK-NEXT:    vaadd.vv v8, v8, v9
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i64> @llvm.riscv.vaadd.nxv1i64.nxv1i64(
+    <vscale x 1 x i64> %0,
+    <vscale x 1 x i64> %1,
+    i32 %2)
+
+  ret <vscale x 1 x i64> %a
+}
+
+declare <vscale x 1 x i64> @llvm.riscv.vaadd.mask.nxv1i64.nxv1i64(
+  <vscale x 1 x i64>,
+  <vscale x 1 x i64>,
+  <vscale x 1 x i64>,
+  <vscale x 1 x i1>,
+  i32);
+
+define <vscale x 1 x i64> @intrinsic_vaadd_mask_vv_nxv1i64_nxv1i64_nxv1i64(<vscale x 1 x i64> %0, <vscale x 1 x i64> %1, <vscale x 1 x i64> %2, <vscale x 1 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vaadd_mask_vv_nxv1i64_nxv1i64_nxv1i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e64,m1,tu,mu
+; CHECK-NEXT:    vaadd.vv v8, v9, v10, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i64> @llvm.riscv.vaadd.mask.nxv1i64.nxv1i64(
+    <vscale x 1 x i64> %0,
+    <vscale x 1 x i64> %1,
+    <vscale x 1 x i64> %2,
+    <vscale x 1 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 1 x i64> %a
+}
+
+declare <vscale x 2 x i64> @llvm.riscv.vaadd.nxv2i64.nxv2i64(
+  <vscale x 2 x i64>,
+  <vscale x 2 x i64>,
+  i32);
+
+define <vscale x 2 x i64> @intrinsic_vaadd_vv_nxv2i64_nxv2i64_nxv2i64(<vscale x 2 x i64> %0, <vscale x 2 x i64> %1, i32 %2) nounwind {
+; CHECK-LABEL: intrinsic_vaadd_vv_nxv2i64_nxv2i64_nxv2i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e64,m2,ta,mu
+; CHECK-NEXT:    vaadd.vv v8, v8, v10
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i64> @llvm.riscv.vaadd.nxv2i64.nxv2i64(
+    <vscale x 2 x i64> %0,
+    <vscale x 2 x i64> %1,
+    i32 %2)
+
+  ret <vscale x 2 x i64> %a
+}
+
+declare <vscale x 2 x i64> @llvm.riscv.vaadd.mask.nxv2i64.nxv2i64(
+  <vscale x 2 x i64>,
+  <vscale x 2 x i64>,
+  <vscale x 2 x i64>,
+  <vscale x 2 x i1>,
+  i32);
+
+define <vscale x 2 x i64> @intrinsic_vaadd_mask_vv_nxv2i64_nxv2i64_nxv2i64(<vscale x 2 x i64> %0, <vscale x 2 x i64> %1, <vscale x 2 x i64> %2, <vscale x 2 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vaadd_mask_vv_nxv2i64_nxv2i64_nxv2i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e64,m2,tu,mu
+; CHECK-NEXT:    vaadd.vv v8, v10, v12, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i64> @llvm.riscv.vaadd.mask.nxv2i64.nxv2i64(
+    <vscale x 2 x i64> %0,
+    <vscale x 2 x i64> %1,
+    <vscale x 2 x i64> %2,
+    <vscale x 2 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 2 x i64> %a
+}
+
+declare <vscale x 4 x i64> @llvm.riscv.vaadd.nxv4i64.nxv4i64(
+  <vscale x 4 x i64>,
+  <vscale x 4 x i64>,
+  i32);
+
+define <vscale x 4 x i64> @intrinsic_vaadd_vv_nxv4i64_nxv4i64_nxv4i64(<vscale x 4 x i64> %0, <vscale x 4 x i64> %1, i32 %2) nounwind {
+; CHECK-LABEL: intrinsic_vaadd_vv_nxv4i64_nxv4i64_nxv4i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e64,m4,ta,mu
+; CHECK-NEXT:    vaadd.vv v8, v8, v12
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i64> @llvm.riscv.vaadd.nxv4i64.nxv4i64(
+    <vscale x 4 x i64> %0,
+    <vscale x 4 x i64> %1,
+    i32 %2)
+
+  ret <vscale x 4 x i64> %a
+}
+
+declare <vscale x 4 x i64> @llvm.riscv.vaadd.mask.nxv4i64.nxv4i64(
+  <vscale x 4 x i64>,
+  <vscale x 4 x i64>,
+  <vscale x 4 x i64>,
+  <vscale x 4 x i1>,
+  i32);
+
+define <vscale x 4 x i64> @intrinsic_vaadd_mask_vv_nxv4i64_nxv4i64_nxv4i64(<vscale x 4 x i64> %0, <vscale x 4 x i64> %1, <vscale x 4 x i64> %2, <vscale x 4 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vaadd_mask_vv_nxv4i64_nxv4i64_nxv4i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e64,m4,tu,mu
+; CHECK-NEXT:    vaadd.vv v8, v12, v16, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i64> @llvm.riscv.vaadd.mask.nxv4i64.nxv4i64(
+    <vscale x 4 x i64> %0,
+    <vscale x 4 x i64> %1,
+    <vscale x 4 x i64> %2,
+    <vscale x 4 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 4 x i64> %a
+}
+
+declare <vscale x 8 x i64> @llvm.riscv.vaadd.nxv8i64.nxv8i64(
+  <vscale x 8 x i64>,
+  <vscale x 8 x i64>,
+  i32);
+
+define <vscale x 8 x i64> @intrinsic_vaadd_vv_nxv8i64_nxv8i64_nxv8i64(<vscale x 8 x i64> %0, <vscale x 8 x i64> %1, i32 %2) nounwind {
+; CHECK-LABEL: intrinsic_vaadd_vv_nxv8i64_nxv8i64_nxv8i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e64,m8,ta,mu
+; CHECK-NEXT:    vaadd.vv v8, v8, v16
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i64> @llvm.riscv.vaadd.nxv8i64.nxv8i64(
+    <vscale x 8 x i64> %0,
+    <vscale x 8 x i64> %1,
+    i32 %2)
+
+  ret <vscale x 8 x i64> %a
+}
+
+declare <vscale x 8 x i64> @llvm.riscv.vaadd.mask.nxv8i64.nxv8i64(
+  <vscale x 8 x i64>,
+  <vscale x 8 x i64>,
+  <vscale x 8 x i64>,
+  <vscale x 8 x i1>,
+  i32);
+
+define <vscale x 8 x i64> @intrinsic_vaadd_mask_vv_nxv8i64_nxv8i64_nxv8i64(<vscale x 8 x i64> %0, <vscale x 8 x i64> %1, <vscale x 8 x i64> %2, <vscale x 8 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vaadd_mask_vv_nxv8i64_nxv8i64_nxv8i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vl8re64.v v24, (a0)
+; CHECK-NEXT:    vsetvli a0, a1, e64,m8,tu,mu
+; CHECK-NEXT:    vaadd.vv v8, v16, v24, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i64> @llvm.riscv.vaadd.mask.nxv8i64.nxv8i64(
+    <vscale x 8 x i64> %0,
+    <vscale x 8 x i64> %1,
+    <vscale x 8 x i64> %2,
+    <vscale x 8 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 8 x i64> %a
+}
+
 declare <vscale x 1 x i8> @llvm.riscv.vaadd.nxv1i8.i8(
   <vscale x 1 x i8>,
   i8,
   i32);
 
 define <vscale x 1 x i8> @intrinsic_vaadd_vx_nxv1i8_nxv1i8_i8(<vscale x 1 x i8> %0, i8 %1, i32 %2) nounwind {
 ; CHECK-LABEL: intrinsic_vaadd_vx_nxv1i8_nxv1i8_i8:
 ; CHECK:       # %bb.0: # %entry
%a = call <vscale x 16 x i32> @llvm.riscv.vaadd.mask.nxv16i32.i32(
     <vscale x 16 x i32> %0,
     <vscale x 16 x i32> %1,
     i32 %2,
     <vscale x 16 x i1> %3,
     i32 %4)
 
   ret <vscale x 16 x i32> %a
 }
+
+declare <vscale x 1 x i64> @llvm.riscv.vaadd.nxv1i64.i64(
+  <vscale x 1 x i64>,
+  i64,
+  i32);
+
+define <vscale x 1 x i64> @intrinsic_vaadd_vx_nxv1i64_nxv1i64_i64(<vscale x 1 x i64> %0, i64 %1, i32 %2) nounwind {
+; CHECK-LABEL: intrinsic_vaadd_vx_nxv1i64_nxv1i64_i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a2, a2, e64,m1,ta,mu
+; CHECK-NEXT:    vmv.v.x v25, a1
+; CHECK-NEXT:    addi a1, zero, 32
+; CHECK-NEXT:    vsll.vx v25, v25, a1
+; CHECK-NEXT:    vmv.v.x v26, a0
+; CHECK-NEXT:    vsll.vx v26, v26, a1
+; CHECK-NEXT:    vsrl.vx v26, v26, a1
+; CHECK-NEXT:    vor.vv v25, v26, v25
+; CHECK-NEXT:    vaadd.vv v8, v8, v25
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i64> @llvm.riscv.vaadd.nxv1i64.i64(
+    <vscale x 1 x i64> %0,
+    i64 %1,
+    i32 %2)
+
+  ret <vscale x 1 x i64> %a
+}
+
+declare <vscale x 1 x i64> @llvm.riscv.vaadd.mask.nxv1i64.i64(
+  <vscale x 1 x i64>,
+  <vscale x 1 x i64>,
+  i64,
+  <vscale x 1 x i1>,
+  i32);
+
+define <vscale x 1 x i64> @intrinsic_vaadd_mask_vx_nxv1i64_nxv1i64_i64(<vscale x 1 x i64> %0, <vscale x 1 x i64> %1, i64 %2, <vscale x 1 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vaadd_mask_vx_nxv1i64_nxv1i64_i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a3, a2, e64,m1,ta,mu
+; CHECK-NEXT:    vmv.v.x v25, a1
+; CHECK-NEXT:    addi a1, zero, 32
+; CHECK-NEXT:    vsll.vx v25, v25, a1
+; CHECK-NEXT:    vmv.v.x v26, a0
+; CHECK-NEXT:    vsll.vx v26, v26, a1
+; CHECK-NEXT:    vsrl.vx v26, v26, a1
+; CHECK-NEXT:    vor.vv v25, v26, v25
+; CHECK-NEXT:    vsetvli a0, a2, e64,m1,tu,mu
+; CHECK-NEXT:    vaadd.vv v8, v9, v25, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i64> @llvm.riscv.vaadd.mask.nxv1i64.i64(
+    <vscale x 1 x i64> %0,
+    <vscale x 1 x i64> %1,
+    i64 %2,
+    <vscale x 1 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 1 x i64> %a
+}
+
+declare <vscale x 2 x i64> @llvm.riscv.vaadd.nxv2i64.i64(
+  <vscale x 2 x i64>,
+  i64,
+  i32);
+
+define <vscale x 2 x i64> @intrinsic_vaadd_vx_nxv2i64_nxv2i64_i64(<vscale x 2 x i64> %0, i64 %1, i32 %2) nounwind {
+; CHECK-LABEL: intrinsic_vaadd_vx_nxv2i64_nxv2i64_i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a2, a2, e64,m2,ta,mu
+; CHECK-NEXT:    vmv.v.x v26, a1
+; CHECK-NEXT:    addi a1, zero, 32
+; CHECK-NEXT:    vsll.vx v26, v26, a1
+; CHECK-NEXT:    vmv.v.x v28, a0
+; CHECK-NEXT:    vsll.vx v28, v28, a1
+; CHECK-NEXT:    vsrl.vx v28, v28, a1
+; CHECK-NEXT:    vor.vv v26, v28, v26
+; CHECK-NEXT:    vaadd.vv v8, v8, v26
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i64> @llvm.riscv.vaadd.nxv2i64.i64(
+    <vscale x 2 x i64> %0,
+    i64 %1,
+    i32 %2)
+
+  ret <vscale x 2 x i64> %a
+}
+
+declare <vscale x 2 x i64> @llvm.riscv.vaadd.mask.nxv2i64.i64(
+  <vscale x 2 x i64>,
+  <vscale x 2 x i64>,
+  i64,
+  <vscale x 2 x i1>,
+  i32);
+
+define <vscale x 2 x i64> @intrinsic_vaadd_mask_vx_nxv2i64_nxv2i64_i64(<vscale x 2 x i64> %0, <vscale x 2 x i64> %1, i64 %2, <vscale x 2 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vaadd_mask_vx_nxv2i64_nxv2i64_i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a3, a2, e64,m2,ta,mu
+; CHECK-NEXT:    vmv.v.x v26, a1
+; CHECK-NEXT:    addi a1, zero, 32
+; CHECK-NEXT:    vsll.vx v26, v26, a1
+; CHECK-NEXT:    vmv.v.x v28, a0
+; CHECK-NEXT:    vsll.vx v28, v28, a1
+; CHECK-NEXT:    vsrl.vx v28, v28, a1
+; CHECK-NEXT:    vor.vv v26, v28, v26
+; CHECK-NEXT:    vsetvli a0, a2, e64,m2,tu,mu
+; CHECK-NEXT:    vaadd.vv v8, v10, v26, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i64> @llvm.riscv.vaadd.mask.nxv2i64.i64(
+    <vscale x 2 x i64> %0,
+    <vscale x 2 x i64> %1,
+    i64 %2,
+    <vscale x 2 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 2 x i64> %a
+}
+
+declare <vscale x 4 x i64> @llvm.riscv.vaadd.nxv4i64.i64(
+  <vscale x 4 x i64>,
+  i64,
+  i32);
+
+define <vscale x 4 x i64> @intrinsic_vaadd_vx_nxv4i64_nxv4i64_i64(<vscale x 4 x i64> %0, i64 %1, i32 %2) nounwind {
+; CHECK-LABEL: intrinsic_vaadd_vx_nxv4i64_nxv4i64_i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a2, a2, e64,m4,ta,mu
+; CHECK-NEXT:    vmv.v.x v28, a1
+; CHECK-NEXT:    addi a1, zero, 32
+; CHECK-NEXT:    vsll.vx v28, v28, a1
+; CHECK-NEXT:    vmv.v.x v12, a0
+; CHECK-NEXT:    vsll.vx v12, v12, a1
+; CHECK-NEXT:    vsrl.vx v12, v12, a1
+; CHECK-NEXT:    vor.vv v28, v12, v28
+; CHECK-NEXT:    vaadd.vv v8, v8, v28
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i64> @llvm.riscv.vaadd.nxv4i64.i64(
+    <vscale x 4 x i64> %0,
+    i64 %1,
+    i32 %2)
+
+  ret <vscale x 4 x i64> %a
+}
+
+declare <vscale x 4 x i64> @llvm.riscv.vaadd.mask.nxv4i64.i64(
+  <vscale x 4 x i64>,
+  <vscale x 4 x i64>,
+  i64,
+  <vscale x 4 x i1>,
+  i32);
+
+define <vscale x 4 x i64> @intrinsic_vaadd_mask_vx_nxv4i64_nxv4i64_i64(<vscale x 4 x i64> %0, <vscale x 4 x i64> %1, i64 %2, <vscale x 4 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vaadd_mask_vx_nxv4i64_nxv4i64_i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a3, a2, e64,m4,ta,mu
+; CHECK-NEXT:    vmv.v.x v28, a1
+; CHECK-NEXT:    addi a1, zero, 32
+; CHECK-NEXT:    vsll.vx v28, v28, a1
+; CHECK-NEXT:    vmv.v.x v16, a0
+; CHECK-NEXT:    vsll.vx v16, v16, a1
+; CHECK-NEXT:    vsrl.vx v16, v16, a1
+; CHECK-NEXT:    vor.vv v28, v16, v28
+; CHECK-NEXT:    vsetvli a0, a2, e64,m4,tu,mu
+; CHECK-NEXT:    vaadd.vv v8, v12, v28, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i64> @llvm.riscv.vaadd.mask.nxv4i64.i64(
+    <vscale x 4 x i64> %0,
+    <vscale x 4 x i64> %1,
+    i64 %2,
+    <vscale x 4 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 4 x i64> %a
+}
+
+declare <vscale x 8 x i64> @llvm.riscv.vaadd.nxv8i64.i64(
+  <vscale x 8 x i64>,
+  i64,
+  i32);
+
+define <vscale x 8 x i64> @intrinsic_vaadd_vx_nxv8i64_nxv8i64_i64(<vscale x 8 x i64> %0, i64 %1, i32 %2) nounwind {
+; CHECK-LABEL: intrinsic_vaadd_vx_nxv8i64_nxv8i64_i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a2, a2, e64,m8,ta,mu
+; CHECK-NEXT:    vmv.v.x v16, a1
+; CHECK-NEXT:    addi a1, zero, 32
+; CHECK-NEXT:    vsll.vx v16, v16, a1
+; CHECK-NEXT:    vmv.v.x v24, a0
+; CHECK-NEXT:    vsll.vx v24, v24, a1
+; CHECK-NEXT:    vsrl.vx v24, v24, a1
+; CHECK-NEXT:    vor.vv v16, v24, v16
+; CHECK-NEXT:    vaadd.vv v8, v8, v16
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i64> @llvm.riscv.vaadd.nxv8i64.i64(
+    <vscale x 8 x i64> %0,
+    i64 %1,
+    i32 %2)
+
+  ret <vscale x 8 x i64> %a
+}
+
+declare <vscale x 8 x i64> @llvm.riscv.vaadd.mask.nxv8i64.i64(
+  <vscale x 8 x i64>,
+  <vscale x 8 x i64>,
+  i64,
+  <vscale x 8 x i1>,
+  i32);
+
+define <vscale x 8 x i64> @intrinsic_vaadd_mask_vx_nxv8i64_nxv8i64_i64(<vscale x 8 x i64> %0, <vscale x 8 x i64> %1, i64 %2, <vscale x 8 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vaadd_mask_vx_nxv8i64_nxv8i64_i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    csrrs a3, vlenb, zero
+; CHECK-NEXT:    sub sp, sp, a3
+; CHECK-NEXT:    vs1r.v v0, (sp) # Unknown-size Folded Spill
+; CHECK-NEXT:    vsetvli a3, a2, e64,m8,ta,mu
+; CHECK-NEXT:    vmv.v.x v24, a1
+; CHECK-NEXT:    addi a1, zero, 32
+; CHECK-NEXT:    vsll.vx v0, v24, a1
+; CHECK-NEXT:    vmv.v.x v24, a0
+; CHECK-NEXT:    vsll.vx v24, v24, a1
+; CHECK-NEXT:    vsrl.vx v24, v24, a1
+; CHECK-NEXT:    vor.vv v24, v24, v0
+; CHECK-NEXT:    vsetvli a0, a2, e64,m8,tu,mu
+; CHECK-NEXT:    vl1re8.v v0, (sp) # Unknown-size Folded Reload
+; CHECK-NEXT:    vaadd.vv v8, v16, v24, v0.t
+; CHECK-NEXT:    csrrs a0, vlenb, zero
+; CHECK-NEXT:    add sp, sp, a0
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i64> @llvm.riscv.vaadd.mask.nxv8i64.i64(
+    <vscale x 8 x i64> %0,
+    <vscale x 8 x i64> %1,
+    i64 %2,
+    <vscale x 8 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 8 x i64> %a
+}

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

%a = call <vscale x 16 x i32> @llvm.riscv.vaaddu.mask.nxv16i32.nxv16i32(
     <vscale x 16 x i32> %1,
     <vscale x 16 x i32> %2,
     <vscale x 16 x i1> %3,
     i32 %4)
 
   ret <vscale x 16 x i32> %a
 }
 
+declare <vscale x 1 x i64> @llvm.riscv.vaaddu.nxv1i64.nxv1i64(
+  <vscale x 1 x i64>,
+  <vscale x 1 x i64>,
+  i32);
+
+define <vscale x 1 x i64> @intrinsic_vaaddu_vv_nxv1i64_nxv1i64_nxv1i64(<vscale x 1 x i64> %0, <vscale x 1 x i64> %1, i32 %2) nounwind {
+; CHECK-LABEL: intrinsic_vaaddu_vv_nxv1i64_nxv1i64_nxv1i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e64,m1,ta,mu
+; CHECK-NEXT:    vaaddu.vv v8, v8, v9
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i64> @llvm.riscv.vaaddu.nxv1i64.nxv1i64(
+    <vscale x 1 x i64> %0,
+    <vscale x 1 x i64> %1,
+    i32 %2)
+
+  ret <vscale x 1 x i64> %a
+}
+
+declare <vscale x 1 x i64> @llvm.riscv.vaaddu.mask.nxv1i64.nxv1i64(
+  <vscale x 1 x i64>,
+  <vscale x 1 x i64>,
+  <vscale x 1 x i64>,
+  <vscale x 1 x i1>,
+  i32);
+
+define <vscale x 1 x i64> @intrinsic_vaaddu_mask_vv_nxv1i64_nxv1i64_nxv1i64(<vscale x 1 x i64> %0, <vscale x 1 x i64> %1, <vscale x 1 x i64> %2, <vscale x 1 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vaaddu_mask_vv_nxv1i64_nxv1i64_nxv1i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e64,m1,tu,mu
+; CHECK-NEXT:    vaaddu.vv v8, v9, v10, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i64> @llvm.riscv.vaaddu.mask.nxv1i64.nxv1i64(
+    <vscale x 1 x i64> %0,
+    <vscale x 1 x i64> %1,
+    <vscale x 1 x i64> %2,
+    <vscale x 1 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 1 x i64> %a
+}
+
+declare <vscale x 2 x i64> @llvm.riscv.vaaddu.nxv2i64.nxv2i64(
+  <vscale x 2 x i64>,
+  <vscale x 2 x i64>,
+  i32);
+
+define <vscale x 2 x i64> @intrinsic_vaaddu_vv_nxv2i64_nxv2i64_nxv2i64(<vscale x 2 x i64> %0, <vscale x 2 x i64> %1, i32 %2) nounwind {
+; CHECK-LABEL: intrinsic_vaaddu_vv_nxv2i64_nxv2i64_nxv2i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e64,m2,ta,mu
+; CHECK-NEXT:    vaaddu.vv v8, v8, v10
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i64> @llvm.riscv.vaaddu.nxv2i64.nxv2i64(
+    <vscale x 2 x i64> %0,
+    <vscale x 2 x i64> %1,
+    i32 %2)
+
+  ret <vscale x 2 x i64> %a
+}
+
+declare <vscale x 2 x i64> @llvm.riscv.vaaddu.mask.nxv2i64.nxv2i64(
+  <vscale x 2 x i64>,
+  <vscale x 2 x i64>,
+  <vscale x 2 x i64>,
+  <vscale x 2 x i1>,
+  i32);
+
+define <vscale x 2 x i64> @intrinsic_vaaddu_mask_vv_nxv2i64_nxv2i64_nxv2i64(<vscale x 2 x i64> %0, <vscale x 2 x i64> %1, <vscale x 2 x i64> %2, <vscale x 2 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vaaddu_mask_vv_nxv2i64_nxv2i64_nxv2i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e64,m2,tu,mu
+; CHECK-NEXT:    vaaddu.vv v8, v10, v12, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i64> @llvm.riscv.vaaddu.mask.nxv2i64.nxv2i64(
+    <vscale x 2 x i64> %0,
+    <vscale x 2 x i64> %1,
+    <vscale x 2 x i64> %2,
+    <vscale x 2 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 2 x i64> %a
+}
+
+declare <vscale x 4 x i64> @llvm.riscv.vaaddu.nxv4i64.nxv4i64(
+  <vscale x 4 x i64>,
+  <vscale x 4 x i64>,
+  i32);
+
+define <vscale x 4 x i64> @intrinsic_vaaddu_vv_nxv4i64_nxv4i64_nxv4i64(<vscale x 4 x i64> %0, <vscale x 4 x i64> %1, i32 %2) nounwind {
+; CHECK-LABEL: intrinsic_vaaddu_vv_nxv4i64_nxv4i64_nxv4i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e64,m4,ta,mu
+; CHECK-NEXT:    vaaddu.vv v8, v8, v12
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i64> @llvm.riscv.vaaddu.nxv4i64.nxv4i64(
+    <vscale x 4 x i64> %0,
+    <vscale x 4 x i64> %1,
+    i32 %2)
+
+  ret <vscale x 4 x i64> %a
+}
+
+declare <vscale x 4 x i64> @llvm.riscv.vaaddu.mask.nxv4i64.nxv4i64(
+  <vscale x 4 x i64>,
+  <vscale x 4 x i64>,
+  <vscale x 4 x i64>,
+  <vscale x 4 x i1>,
+  i32);
+
+define <vscale x 4 x i64> @intrinsic_vaaddu_mask_vv_nxv4i64_nxv4i64_nxv4i64(<vscale x 4 x i64> %0, <vscale x 4 x i64> %1, <vscale x 4 x i64> %2, <vscale x 4 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vaaddu_mask_vv_nxv4i64_nxv4i64_nxv4i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e64,m4,tu,mu
+; CHECK-NEXT:    vaaddu.vv v8, v12, v16, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i64> @llvm.riscv.vaaddu.mask.nxv4i64.nxv4i64(
+    <vscale x 4 x i64> %0,
+    <vscale x 4 x i64> %1,
+    <vscale x 4 x i64> %2,
+    <vscale x 4 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 4 x i64> %a
+}
+
+declare <vscale x 8 x i64> @llvm.riscv.vaaddu.nxv8i64.nxv8i64(
+  <vscale x 8 x i64>,
+  <vscale x 8 x i64>,
+  i32);
+
+define <vscale x 8 x i64> @intrinsic_vaaddu_vv_nxv8i64_nxv8i64_nxv8i64(<vscale x 8 x i64> %0, <vscale x 8 x i64> %1, i32 %2) nounwind {
+; CHECK-LABEL: intrinsic_vaaddu_vv_nxv8i64_nxv8i64_nxv8i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e64,m8,ta,mu
+; CHECK-NEXT:    vaaddu.vv v8, v8, v16
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i64> @llvm.riscv.vaaddu.nxv8i64.nxv8i64(
+    <vscale x 8 x i64> %0,
+    <vscale x 8 x i64> %1,
+    i32 %2)
+
+  ret <vscale x 8 x i64> %a
+}
+
+declare <vscale x 8 x i64> @llvm.riscv.vaaddu.mask.nxv8i64.nxv8i64(
+  <vscale x 8 x i64>,
+  <vscale x 8 x i64>,
+  <vscale x 8 x i64>,
+  <vscale x 8 x i1>,
+  i32);
+
+define <vscale x 8 x i64> @intrinsic_vaaddu_mask_vv_nxv8i64_nxv8i64_nxv8i64(<vscale x 8 x i64> %0, <vscale x 8 x i64> %1, <vscale x 8 x i64> %2, <vscale x 8 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vaaddu_mask_vv_nxv8i64_nxv8i64_nxv8i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vl8re64.v v24, (a0)
+; CHECK-NEXT:    vsetvli a0, a1, e64,m8,tu,mu
+; CHECK-NEXT:    vaaddu.vv v8, v16, v24, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i64> @llvm.riscv.vaaddu.mask.nxv8i64.nxv8i64(
+    <vscale x 8 x i64> %0,
+    <vscale x 8 x i64> %1,
+    <vscale x 8 x i64> %2,
+    <vscale x 8 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 8 x i64> %a
+}
+
 declare <vscale x 1 x i8> @llvm.riscv.vaaddu.nxv1i8.i8(
   <vscale x 1 x i8>,
   i8,
   i32);
 
 define <vscale x 1 x i8> @intrinsic_vaaddu_vx_nxv1i8_nxv1i8_i8(<vscale x 1 x i8> %0, i8 %1, i32 %2) nounwind {
 ; CHECK-LABEL: intrinsic_vaaddu_vx_nxv1i8_nxv1i8_i8:
 ; CHECK:       # %bb.0: # %entry
%a = call <vscale x 16 x i32> @llvm.riscv.vaaddu.mask.nxv16i32.i32(
     <vscale x 16 x i32> %0,
     <vscale x 16 x i32> %1,
     i32 %2,
     <vscale x 16 x i1> %3,
     i32 %4)
 
   ret <vscale x 16 x i32> %a
 }
+
+declare <vscale x 1 x i64> @llvm.riscv.vaaddu.nxv1i64.i64(
+  <vscale x 1 x i64>,
+  i64,
+  i32);
+
+define <vscale x 1 x i64> @intrinsic_vaaddu_vx_nxv1i64_nxv1i64_i64(<vscale x 1 x i64> %0, i64 %1, i32 %2) nounwind {
+; CHECK-LABEL: intrinsic_vaaddu_vx_nxv1i64_nxv1i64_i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a2, a2, e64,m1,ta,mu
+; CHECK-NEXT:    vmv.v.x v25, a1
+; CHECK-NEXT:    addi a1, zero, 32
+; CHECK-NEXT:    vsll.vx v25, v25, a1
+; CHECK-NEXT:    vmv.v.x v26, a0
+; CHECK-NEXT:    vsll.vx v26, v26, a1
+; CHECK-NEXT:    vsrl.vx v26, v26, a1
+; CHECK-NEXT:    vor.vv v25, v26, v25
+; CHECK-NEXT:    vaaddu.vv v8, v8, v25
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i64> @llvm.riscv.vaaddu.nxv1i64.i64(
+    <vscale x 1 x i64> %0,
+    i64 %1,
+    i32 %2)
+
+  ret <vscale x 1 x i64> %a
+}
+
+declare <vscale x 1 x i64> @llvm.riscv.vaaddu.mask.nxv1i64.i64(
+  <vscale x 1 x i64>,
+  <vscale x 1 x i64>,
+  i64,
+  <vscale x 1 x i1>,
+  i32);
+
+define <vscale x 1 x i64> @intrinsic_vaaddu_mask_vx_nxv1i64_nxv1i64_i64(<vscale x 1 x i64> %0, <vscale x 1 x i64> %1, i64 %2, <vscale x 1 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vaaddu_mask_vx_nxv1i64_nxv1i64_i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a3, a2, e64,m1,ta,mu
+; CHECK-NEXT:    vmv.v.x v25, a1
+; CHECK-NEXT:    addi a1, zero, 32
+; CHECK-NEXT:    vsll.vx v25, v25, a1
+; CHECK-NEXT:    vmv.v.x v26, a0
+; CHECK-NEXT:    vsll.vx v26, v26, a1
+; CHECK-NEXT:    vsrl.vx v26, v26, a1
+; CHECK-NEXT:    vor.vv v25, v26, v25
+; CHECK-NEXT:    vsetvli a0, a2, e64,m1,tu,mu
+; CHECK-NEXT:    vaaddu.vv v8, v9, v25, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i64> @llvm.riscv.vaaddu.mask.nxv1i64.i64(
+    <vscale x 1 x i64> %0,
+    <vscale x 1 x i64> %1,
+    i64 %2,
+    <vscale x 1 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 1 x i64> %a
+}
+
+declare <vscale x 2 x i64> @llvm.riscv.vaaddu.nxv2i64.i64(
+  <vscale x 2 x i64>,
+  i64,
+  i32);
+
+define <vscale x 2 x i64> @intrinsic_vaaddu_vx_nxv2i64_nxv2i64_i64(<vscale x 2 x i64> %0, i64 %1, i32 %2) nounwind {
+; CHECK-LABEL: intrinsic_vaaddu_vx_nxv2i64_nxv2i64_i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a2, a2, e64,m2,ta,mu
+; CHECK-NEXT:    vmv.v.x v26, a1
+; CHECK-NEXT:    addi a1, zero, 32
+; CHECK-NEXT:    vsll.vx v26, v26, a1
+; CHECK-NEXT:    vmv.v.x v28, a0
+; CHECK-NEXT:    vsll.vx v28, v28, a1
+; CHECK-NEXT:    vsrl.vx v28, v28, a1
+; CHECK-NEXT:    vor.vv v26, v28, v26
+; CHECK-NEXT:    vaaddu.vv v8, v8, v26
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i64> @llvm.riscv.vaaddu.nxv2i64.i64(
+    <vscale x 2 x i64> %0,
+    i64 %1,
+    i32 %2)
+
+  ret <vscale x 2 x i64> %a
+}
+
+declare <vscale x 2 x i64> @llvm.riscv.vaaddu.mask.nxv2i64.i64(
+  <vscale x 2 x i64>,
+  <vscale x 2 x i64>,
+  i64,
+  <vscale x 2 x i1>,
+  i32);
+
+define <vscale x 2 x i64> @intrinsic_vaaddu_mask_vx_nxv2i64_nxv2i64_i64(<vscale x 2 x i64> %0, <vscale x 2 x i64> %1, i64 %2, <vscale x 2 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vaaddu_mask_vx_nxv2i64_nxv2i64_i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a3, a2, e64,m2,ta,mu
+; CHECK-NEXT:    vmv.v.x v26, a1
+; CHECK-NEXT:    addi a1, zero, 32
+; CHECK-NEXT:    vsll.vx v26, v26, a1
+; CHECK-NEXT:    vmv.v.x v28, a0
+; CHECK-NEXT:    vsll.vx v28, v28, a1
+; CHECK-NEXT:    vsrl.vx v28, v28, a1
+; CHECK-NEXT:    vor.vv v26, v28, v26
+; CHECK-NEXT:    vsetvli a0, a2, e64,m2,tu,mu
+; CHECK-NEXT:    vaaddu.vv v8, v10, v26, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i64> @llvm.riscv.vaaddu.mask.nxv2i64.i64(
+    <vscale x 2 x i64> %0,
+    <vscale x 2 x i64> %1,
+    i64 %2,
+    <vscale x 2 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 2 x i64> %a
+}
+
+declare <vscale x 4 x i64> @llvm.riscv.vaaddu.nxv4i64.i64(
+  <vscale x 4 x i64>,
+  i64,
+  i32);
+
+define <vscale x 4 x i64> @intrinsic_vaaddu_vx_nxv4i64_nxv4i64_i64(<vscale x 4 x i64> %0, i64 %1, i32 %2) nounwind {
+; CHECK-LABEL: intrinsic_vaaddu_vx_nxv4i64_nxv4i64_i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a2, a2, e64,m4,ta,mu
+; CHECK-NEXT:    vmv.v.x v28, a1
+; CHECK-NEXT:    addi a1, zero, 32
+; CHECK-NEXT:    vsll.vx v28, v28, a1
+; CHECK-NEXT:    vmv.v.x v12, a0
+; CHECK-NEXT:    vsll.vx v12, v12, a1
+; CHECK-NEXT:    vsrl.vx v12, v12, a1
+; CHECK-NEXT:    vor.vv v28, v12, v28
+; CHECK-NEXT:    vaaddu.vv v8, v8, v28
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i64> @llvm.riscv.vaaddu.nxv4i64.i64(
+    <vscale x 4 x i64> %0,
+    i64 %1,
+    i32 %2)
+
+  ret <vscale x 4 x i64> %a
+}
+
+declare <vscale x 4 x i64> @llvm.riscv.vaaddu.mask.nxv4i64.i64(
+  <vscale x 4 x i64>,
+  <vscale x 4 x i64>,
+  i64,
+  <vscale x 4 x i1>,
+  i32);
+
+define <vscale x 4 x i64> @intrinsic_vaaddu_mask_vx_nxv4i64_nxv4i64_i64(<vscale x 4 x i64> %0, <vscale x 4 x i64> %1, i64 %2, <vscale x 4 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vaaddu_mask_vx_nxv4i64_nxv4i64_i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a3, a2, e64,m4,ta,mu
+; CHECK-NEXT:    vmv.v.x v28, a1
+; CHECK-NEXT:    addi a1, zero, 32
+; CHECK-NEXT:    vsll.vx v28, v28, a1
+; CHECK-NEXT:    vmv.v.x v16, a0
+; CHECK-NEXT:    vsll.vx v16, v16, a1
+; CHECK-NEXT:    vsrl.vx v16, v16, a1
+; CHECK-NEXT:    vor.vv v28, v16, v28
+; CHECK-NEXT:    vsetvli a0, a2, e64,m4,tu,mu
+; CHECK-NEXT:    vaaddu.vv v8, v12, v28, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i64> @llvm.riscv.vaaddu.mask.nxv4i64.i64(
+    <vscale x 4 x i64> %0,
+    <vscale x 4 x i64> %1,
+    i64 %2,
+    <vscale x 4 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 4 x i64> %a
+}
+
+declare <vscale x 8 x i64> @llvm.riscv.vaaddu.nxv8i64.i64(
+  <vscale x 8 x i64>,
+  i64,
+  i32);
+
+define <vscale x 8 x i64> @intrinsic_vaaddu_vx_nxv8i64_nxv8i64_i64(<vscale x 8 x i64> %0, i64 %1, i32 %2) nounwind {
+; CHECK-LABEL: intrinsic_vaaddu_vx_nxv8i64_nxv8i64_i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a2, a2, e64,m8,ta,mu
+; CHECK-NEXT:    vmv.v.x v16, a1
+; CHECK-NEXT:    addi a1, zero, 32
+; CHECK-NEXT:    vsll.vx v16, v16, a1
+; CHECK-NEXT:    vmv.v.x v24, a0
+; CHECK-NEXT:    vsll.vx v24, v24, a1
+; CHECK-NEXT:    vsrl.vx v24, v24, a1
+; CHECK-NEXT:    vor.vv v16, v24, v16
+; CHECK-NEXT:    vaaddu.vv v8, v8, v16
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i64> @llvm.riscv.vaaddu.nxv8i64.i64(
+    <vscale x 8 x i64> %0,
+    i64 %1,
+    i32 %2)
+
+  ret <vscale x 8 x i64> %a
+}
+
+declare <vscale x 8 x i64> @llvm.riscv.vaaddu.mask.nxv8i64.i64(
+  <vscale x 8 x i64>,
+  <vscale x 8 x i64>,
+  i64,
+  <vscale x 8 x i1>,
+  i32);
+
+define <vscale x 8 x i64> @intrinsic_vaaddu_mask_vx_nxv8i64_nxv8i64_i64(<vscale x 8 x i64> %0, <vscale x 8 x i64> %1, i64 %2, <vscale x 8 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vaaddu_mask_vx_nxv8i64_nxv8i64_i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    csrrs a3, vlenb, zero
+; CHECK-NEXT:    sub sp, sp, a3
+; CHECK-NEXT:    vs1r.v v0, (sp) # Unknown-size Folded Spill
+; CHECK-NEXT:    vsetvli a3, a2, e64,m8,ta,mu
+; CHECK-NEXT:    vmv.v.x v24, a1
+; CHECK-NEXT:    addi a1, zero, 32
+; CHECK-NEXT:    vsll.vx v0, v24, a1
+; CHECK-NEXT:    vmv.v.x v24, a0
+; CHECK-NEXT:    vsll.vx v24, v24, a1
+; CHECK-NEXT:    vsrl.vx v24, v24, a1
+; CHECK-NEXT:    vor.vv v24, v24, v0
+; CHECK-NEXT:    vsetvli a0, a2, e64,m8,tu,mu
+; CHECK-NEXT:    vl1re8.v v0, (sp) # Unknown-size Folded Reload
+; CHECK-NEXT:    vaaddu.vv v8, v16, v24, v0.t
+; CHECK-NEXT:    csrrs a0, vlenb, zero
+; CHECK-NEXT:    add sp, sp, a0
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i64> @llvm.riscv.vaaddu.mask.nxv8i64.i64(
+    <vscale x 8 x i64> %0,
+    <vscale x 8 x i64> %1,
+    i64 %2,
+    <vscale x 8 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 8 x i64> %a
+}

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

%a = call <vscale x 16 x i32> @llvm.riscv.vadc.nxv16i32.nxv16i32(
     <vscale x 16 x i32> %0,
     <vscale x 16 x i32> %1,
     <vscale x 16 x i1> %2,
     i32 %3)
 
   ret <vscale x 16 x i32> %a
 }
 
+declare <vscale x 1 x i64> @llvm.riscv.vadc.nxv1i64.nxv1i64(
+  <vscale x 1 x i64>,
+  <vscale x 1 x i64>,
+  <vscale x 1 x i1>,
+  i32);
+
+define <vscale x 1 x i64> @intrinsic_vadc_vvm_nxv1i64_nxv1i64_nxv1i64(<vscale x 1 x i64> %0, <vscale x 1 x i64> %1, <vscale x 1 x i1> %2, i32 %3) nounwind {
+; CHECK-LABEL: intrinsic_vadc_vvm_nxv1i64_nxv1i64_nxv1i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e64,m1,ta,mu
+; CHECK-NEXT:    vadc.vvm v8, v8, v9, v0
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i64> @llvm.riscv.vadc.nxv1i64.nxv1i64(
+    <vscale x 1 x i64> %0,
+    <vscale x 1 x i64> %1,
+    <vscale x 1 x i1> %2,
+    i32 %3)
+
+  ret <vscale x 1 x i64> %a
+}
+
+declare <vscale x 2 x i64> @llvm.riscv.vadc.nxv2i64.nxv2i64(
+  <vscale x 2 x i64>,
+  <vscale x 2 x i64>,
+  <vscale x 2 x i1>,
+  i32);
+
+define <vscale x 2 x i64> @intrinsic_vadc_vvm_nxv2i64_nxv2i64_nxv2i64(<vscale x 2 x i64> %0, <vscale x 2 x i64> %1, <vscale x 2 x i1> %2, i32 %3) nounwind {
+; CHECK-LABEL: intrinsic_vadc_vvm_nxv2i64_nxv2i64_nxv2i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e64,m2,ta,mu
+; CHECK-NEXT:    vadc.vvm v8, v8, v10, v0
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i64> @llvm.riscv.vadc.nxv2i64.nxv2i64(
+    <vscale x 2 x i64> %0,
+    <vscale x 2 x i64> %1,
+    <vscale x 2 x i1> %2,
+    i32 %3)
+
+  ret <vscale x 2 x i64> %a
+}
+
+declare <vscale x 4 x i64> @llvm.riscv.vadc.nxv4i64.nxv4i64(
+  <vscale x 4 x i64>,
+  <vscale x 4 x i64>,
+  <vscale x 4 x i1>,
+  i32);
+
+define <vscale x 4 x i64> @intrinsic_vadc_vvm_nxv4i64_nxv4i64_nxv4i64(<vscale x 4 x i64> %0, <vscale x 4 x i64> %1, <vscale x 4 x i1> %2, i32 %3) nounwind {
+; CHECK-LABEL: intrinsic_vadc_vvm_nxv4i64_nxv4i64_nxv4i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e64,m4,ta,mu
+; CHECK-NEXT:    vadc.vvm v8, v8, v12, v0
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i64> @llvm.riscv.vadc.nxv4i64.nxv4i64(
+    <vscale x 4 x i64> %0,
+    <vscale x 4 x i64> %1,
+    <vscale x 4 x i1> %2,
+    i32 %3)
+
+  ret <vscale x 4 x i64> %a
+}
+
+declare <vscale x 8 x i64> @llvm.riscv.vadc.nxv8i64.nxv8i64(
+  <vscale x 8 x i64>,
+  <vscale x 8 x i64>,
+  <vscale x 8 x i1>,
+  i32);
+
+define <vscale x 8 x i64> @intrinsic_vadc_vvm_nxv8i64_nxv8i64_nxv8i64(<vscale x 8 x i64> %0, <vscale x 8 x i64> %1, <vscale x 8 x i1> %2, i32 %3) nounwind {
+; CHECK-LABEL: intrinsic_vadc_vvm_nxv8i64_nxv8i64_nxv8i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e64,m8,ta,mu
+; CHECK-NEXT:    vadc.vvm v8, v8, v16, v0
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i64> @llvm.riscv.vadc.nxv8i64.nxv8i64(
+    <vscale x 8 x i64> %0,
+    <vscale x 8 x i64> %1,
+    <vscale x 8 x i1> %2,
+    i32 %3)
+
+  ret <vscale x 8 x i64> %a
+}
+
 declare <vscale x 1 x i8> @llvm.riscv.vadc.nxv1i8.i8(
   <vscale x 1 x i8>,
   i8,
   <vscale x 1 x i1>,
   i32);
 
 define <vscale x 1 x i8> @intrinsic_vadc_vxm_nxv1i8_nxv1i8_i8(<vscale x 1 x i8> %0, i8 %1, <vscale x 1 x i1> %2, i32 %3) nounwind {
 ; CHECK-LABEL: intrinsic_vadc_vxm_nxv1i8_nxv1i8_i8:
%a = call <vscale x 16 x i32> @llvm.riscv.vadc.nxv16i32.i32(
     <vscale x 16 x i32> %0,
     i32 %1,
     <vscale x 16 x i1> %2,
     i32 %3)
 
   ret <vscale x 16 x i32> %a
 }
 
+declare <vscale x 1 x i64> @llvm.riscv.vadc.nxv1i64.i64(
+  <vscale x 1 x i64>,
+  i64,
+  <vscale x 1 x i1>,
+  i32);
+
+define <vscale x 1 x i64> @intrinsic_vadc_vxm_nxv1i64_nxv1i64_i64(<vscale x 1 x i64> %0, i64 %1, <vscale x 1 x i1> %2, i32 %3) nounwind {
+; CHECK-LABEL: intrinsic_vadc_vxm_nxv1i64_nxv1i64_i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a2, a2, e64,m1,ta,mu
+; CHECK-NEXT:    vmv.v.x v25, a1
+; CHECK-NEXT:    addi a1, zero, 32
+; CHECK-NEXT:    vsll.vx v25, v25, a1
+; CHECK-NEXT:    vmv.v.x v26, a0
+; CHECK-NEXT:    vsll.vx v26, v26, a1
+; CHECK-NEXT:    vsrl.vx v26, v26, a1
+; CHECK-NEXT:    vor.vv v25, v26, v25
+; CHECK-NEXT:    vadc.vvm v8, v8, v25, v0
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i64> @llvm.riscv.vadc.nxv1i64.i64(
+    <vscale x 1 x i64> %0,
+    i64 %1,
+    <vscale x 1 x i1> %2,
+    i32 %3)
+
+  ret <vscale x 1 x i64> %a
+}
+
+declare <vscale x 2 x i64> @llvm.riscv.vadc.nxv2i64.i64(
+  <vscale x 2 x i64>,
+  i64,
+  <vscale x 2 x i1>,
+  i32);
+
+define <vscale x 2 x i64> @intrinsic_vadc_vxm_nxv2i64_nxv2i64_i64(<vscale x 2 x i64> %0, i64 %1, <vscale x 2 x i1> %2, i32 %3) nounwind {
+; CHECK-LABEL: intrinsic_vadc_vxm_nxv2i64_nxv2i64_i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a2, a2, e64,m2,ta,mu
+; CHECK-NEXT:    vmv.v.x v26, a1
+; CHECK-NEXT:    addi a1, zero, 32
+; CHECK-NEXT:    vsll.vx v26, v26, a1
+; CHECK-NEXT:    vmv.v.x v28, a0
+; CHECK-NEXT:    vsll.vx v28, v28, a1
+; CHECK-NEXT:    vsrl.vx v28, v28, a1
+; CHECK-NEXT:    vor.vv v26, v28, v26
+; CHECK-NEXT:    vadc.vvm v8, v8, v26, v0
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i64> @llvm.riscv.vadc.nxv2i64.i64(
+    <vscale x 2 x i64> %0,
+    i64 %1,
+    <vscale x 2 x i1> %2,
+    i32 %3)
+
+  ret <vscale x 2 x i64> %a
+}
+
+declare <vscale x 4 x i64> @llvm.riscv.vadc.nxv4i64.i64(
+  <vscale x 4 x i64>,
+  i64,
+  <vscale x 4 x i1>,
+  i32);
+
+define <vscale x 4 x i64> @intrinsic_vadc_vxm_nxv4i64_nxv4i64_i64(<vscale x 4 x i64> %0, i64 %1, <vscale x 4 x i1> %2, i32 %3) nounwind {
+; CHECK-LABEL: intrinsic_vadc_vxm_nxv4i64_nxv4i64_i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a2, a2, e64,m4,ta,mu
+; CHECK-NEXT:    vmv.v.x v28, a1
+; CHECK-NEXT:    addi a1, zero, 32
+; CHECK-NEXT:    vsll.vx v28, v28, a1
+; CHECK-NEXT:    vmv.v.x v12, a0
+; CHECK-NEXT:    vsll.vx v12, v12, a1
+; CHECK-NEXT:    vsrl.vx v12, v12, a1
+; CHECK-NEXT:    vor.vv v28, v12, v28
+; CHECK-NEXT:    vadc.vvm v8, v8, v28, v0
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i64> @llvm.riscv.vadc.nxv4i64.i64(
+    <vscale x 4 x i64> %0,
+    i64 %1,
+    <vscale x 4 x i1> %2,
+    i32 %3)
+
+  ret <vscale x 4 x i64> %a
+}
+
+declare <vscale x 8 x i64> @llvm.riscv.vadc.nxv8i64.i64(
+  <vscale x 8 x i64>,
+  i64,
+  <vscale x 8 x i1>,
+  i32);
+
+define <vscale x 8 x i64> @intrinsic_vadc_vxm_nxv8i64_nxv8i64_i64(<vscale x 8 x i64> %0, i64 %1, <vscale x 8 x i1> %2, i32 %3) nounwind {
+; CHECK-LABEL: intrinsic_vadc_vxm_nxv8i64_nxv8i64_i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a2, a2, e64,m8,ta,mu
+; CHECK-NEXT:    vmv.v.x v16, a1
+; CHECK-NEXT:    addi a1, zero, 32
+; CHECK-NEXT:    vsll.vx v16, v16, a1
+; CHECK-NEXT:    vmv.v.x v24, a0
+; CHECK-NEXT:    vsll.vx v24, v24, a1
+; CHECK-NEXT:    vsrl.vx v24, v24, a1
+; CHECK-NEXT:    vor.vv v16, v24, v16
+; CHECK-NEXT:    vadc.vvm v8, v8, v16, v0
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i64> @llvm.riscv.vadc.nxv8i64.i64(
+    <vscale x 8 x i64> %0,
+    i64 %1,
+    <vscale x 8 x i1> %2,
+    i32 %3)
+
+  ret <vscale x 8 x i64> %a
+}
+
 define <vscale x 1 x i8> @intrinsic_vadc_vim_nxv1i8_nxv1i8_i8(<vscale x 1 x i8> %0, <vscale x 1 x i1> %1, i32 %2) nounwind {
 ; CHECK-LABEL: intrinsic_vadc_vim_nxv1i8_nxv1i8_i8:
 ; CHECK:       # %bb.0: # %entry
 ; CHECK-NEXT:    vsetvli a0, a0, e8,mf8,ta,mu
 ; CHECK-NEXT:    vadc.vim v8, v8, -9, v0
 ; CHECK-NEXT:    jalr zero, 0(ra)
 entry:
   %a = call <vscale x 1 x i8> @llvm.riscv.vadc.nxv1i8.i8(
entry:
   %a = call <vscale x 16 x i32> @llvm.riscv.vadc.nxv16i32.i32(
     <vscale x 16 x i32> %0,
     i32 9,
     <vscale x 16 x i1> %1,
     i32 %2)
 
   ret <vscale x 16 x i32> %a
 }
+
+define <vscale x 1 x i64> @intrinsic_vadc_vim_nxv1i64_nxv1i64_i64(<vscale x 1 x i64> %0, <vscale x 1 x i1> %1, i32 %2) nounwind {
+; CHECK-LABEL: intrinsic_vadc_vim_nxv1i64_nxv1i64_i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e64,m1,ta,mu
+; CHECK-NEXT:    vadc.vim v8, v8, 9, v0
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i64> @llvm.riscv.vadc.nxv1i64.i64(
+    <vscale x 1 x i64> %0,
+    i64 9,
+    <vscale x 1 x i1> %1,
+    i32 %2)
+
+  ret <vscale x 1 x i64> %a
+}
+
+define <vscale x 2 x i64> @intrinsic_vadc_vim_nxv2i64_nxv2i64_i64(<vscale x 2 x i64> %0, <vscale x 2 x i1> %1, i32 %2) nounwind {
+; CHECK-LABEL: intrinsic_vadc_vim_nxv2i64_nxv2i64_i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e64,m2,ta,mu
+; CHECK-NEXT:    vadc.vim v8, v8, -9, v0
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i64> @llvm.riscv.vadc.nxv2i64.i64(
+    <vscale x 2 x i64> %0,
+    i64 -9,
+    <vscale x 2 x i1> %1,
+    i32 %2)
+
+  ret <vscale x 2 x i64> %a
+}
+
+define <vscale x 4 x i64> @intrinsic_vadc_vim_nxv4i64_nxv4i64_i64(<vscale x 4 x i64> %0, <vscale x 4 x i1> %1, i32 %2) nounwind {
+; CHECK-LABEL: intrinsic_vadc_vim_nxv4i64_nxv4i64_i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e64,m4,ta,mu
+; CHECK-NEXT:    vadc.vim v8, v8, 9, v0
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i64> @llvm.riscv.vadc.nxv4i64.i64(
+    <vscale x 4 x i64> %0,
+    i64 9,
+    <vscale x 4 x i1> %1,
+    i32 %2)
+
+  ret <vscale x 4 x i64> %a
+}
+
+define <vscale x 8 x i64> @intrinsic_vadc_vim_nxv8i64_nxv8i64_i64(<vscale x 8 x i64> %0, <vscale x 8 x i1> %1, i32 %2) nounwind {
+; CHECK-LABEL: intrinsic_vadc_vim_nxv8i64_nxv8i64_i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e64,m8,ta,mu
+; CHECK-NEXT:    vadc.vim v8, v8, -9, v0
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i64> @llvm.riscv.vadc.nxv8i64.i64(
+    <vscale x 8 x i64> %0,
+    i64 -9,
+    <vscale x 8 x i1> %1,
+    i32 %2)
+
+  ret <vscale x 8 x i64> %a
+}

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

%a = call <vscale x 16 x i32> @llvm.riscv.vadd.mask.nxv16i32.nxv16i32(
     <vscale x 16 x i32> %1,
     <vscale x 16 x i32> %2,
     <vscale x 16 x i1> %3,
     i32 %4)
 
   ret <vscale x 16 x i32> %a
 }
 
+declare <vscale x 1 x i64> @llvm.riscv.vadd.nxv1i64.nxv1i64(
+  <vscale x 1 x i64>,
+  <vscale x 1 x i64>,
+  i32);
+
+define <vscale x 1 x i64> @intrinsic_vadd_vv_nxv1i64_nxv1i64_nxv1i64(<vscale x 1 x i64> %0, <vscale x 1 x i64> %1, i32 %2) nounwind {
+; CHECK-LABEL: intrinsic_vadd_vv_nxv1i64_nxv1i64_nxv1i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e64,m1,ta,mu
+; CHECK-NEXT:    vadd.vv v8, v8, v9
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i64> @llvm.riscv.vadd.nxv1i64.nxv1i64(
+    <vscale x 1 x i64> %0,
+    <vscale x 1 x i64> %1,
+    i32 %2)
+
+  ret <vscale x 1 x i64> %a
+}
+
+declare <vscale x 1 x i64> @llvm.riscv.vadd.mask.nxv1i64.nxv1i64(
+  <vscale x 1 x i64>,
+  <vscale x 1 x i64>,
+  <vscale x 1 x i64>,
+  <vscale x 1 x i1>,
+  i32);
+
+define <vscale x 1 x i64> @intrinsic_vadd_mask_vv_nxv1i64_nxv1i64_nxv1i64(<vscale x 1 x i64> %0, <vscale x 1 x i64> %1, <vscale x 1 x i64> %2, <vscale x 1 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vadd_mask_vv_nxv1i64_nxv1i64_nxv1i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e64,m1,tu,mu
+; CHECK-NEXT:    vadd.vv v8, v9, v10, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i64> @llvm.riscv.vadd.mask.nxv1i64.nxv1i64(
+    <vscale x 1 x i64> %0,
+    <vscale x 1 x i64> %1,
+    <vscale x 1 x i64> %2,
+    <vscale x 1 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 1 x i64> %a
+}
+
+declare <vscale x 2 x i64> @llvm.riscv.vadd.nxv2i64.nxv2i64(
+  <vscale x 2 x i64>,
+  <vscale x 2 x i64>,
+  i32);
+
+define <vscale x 2 x i64> @intrinsic_vadd_vv_nxv2i64_nxv2i64_nxv2i64(<vscale x 2 x i64> %0, <vscale x 2 x i64> %1, i32 %2) nounwind {
+; CHECK-LABEL: intrinsic_vadd_vv_nxv2i64_nxv2i64_nxv2i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e64,m2,ta,mu
+; CHECK-NEXT:    vadd.vv v8, v8, v10
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i64> @llvm.riscv.vadd.nxv2i64.nxv2i64(
+    <vscale x 2 x i64> %0,
+    <vscale x 2 x i64> %1,
+    i32 %2)
+
+  ret <vscale x 2 x i64> %a
+}
+
+declare <vscale x 2 x i64> @llvm.riscv.vadd.mask.nxv2i64.nxv2i64(
+  <vscale x 2 x i64>,
+  <vscale x 2 x i64>,
+  <vscale x 2 x i64>,
+  <vscale x 2 x i1>,
+  i32);
+
+define <vscale x 2 x i64> @intrinsic_vadd_mask_vv_nxv2i64_nxv2i64_nxv2i64(<vscale x 2 x i64> %0, <vscale x 2 x i64> %1, <vscale x 2 x i64> %2, <vscale x 2 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vadd_mask_vv_nxv2i64_nxv2i64_nxv2i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e64,m2,tu,mu
+; CHECK-NEXT:    vadd.vv v8, v10, v12, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i64> @llvm.riscv.vadd.mask.nxv2i64.nxv2i64(
+    <vscale x 2 x i64> %0,
+    <vscale x 2 x i64> %1,
+    <vscale x 2 x i64> %2,
+    <vscale x 2 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 2 x i64> %a
+}
+
+declare <vscale x 4 x i64> @llvm.riscv.vadd.nxv4i64.nxv4i64(
+  <vscale x 4 x i64>,
+  <vscale x 4 x i64>,
+  i32);
+
+define <vscale x 4 x i64> @intrinsic_vadd_vv_nxv4i64_nxv4i64_nxv4i64(<vscale x 4 x i64> %0, <vscale x 4 x i64> %1, i32 %2) nounwind {
+; CHECK-LABEL: intrinsic_vadd_vv_nxv4i64_nxv4i64_nxv4i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e64,m4,ta,mu
+; CHECK-NEXT:    vadd.vv v8, v8, v12
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i64> @llvm.riscv.vadd.nxv4i64.nxv4i64(
+    <vscale x 4 x i64> %0,
+    <vscale x 4 x i64> %1,
+    i32 %2)
+
+  ret <vscale x 4 x i64> %a
+}
+
+declare <vscale x 4 x i64> @llvm.riscv.vadd.mask.nxv4i64.nxv4i64(
+  <vscale x 4 x i64>,
+  <vscale x 4 x i64>,
+  <vscale x 4 x i64>,
+  <vscale x 4 x i1>,
+  i32);
+
+define <vscale x 4 x i64> @intrinsic_vadd_mask_vv_nxv4i64_nxv4i64_nxv4i64(<vscale x 4 x i64> %0, <vscale x 4 x i64> %1, <vscale x 4 x i64> %2, <vscale x 4 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vadd_mask_vv_nxv4i64_nxv4i64_nxv4i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e64,m4,tu,mu
+; CHECK-NEXT:    vadd.vv v8, v12, v16, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i64> @llvm.riscv.vadd.mask.nxv4i64.nxv4i64(
+    <vscale x 4 x i64> %0,
+    <vscale x 4 x i64> %1,
+    <vscale x 4 x i64> %2,
+    <vscale x 4 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 4 x i64> %a
+}
+
+declare <vscale x 8 x i64> @llvm.riscv.vadd.nxv8i64.nxv8i64(
+  <vscale x 8 x i64>,
+  <vscale x 8 x i64>,
+  i32);
+
+define <vscale x 8 x i64> @intrinsic_vadd_vv_nxv8i64_nxv8i64_nxv8i64(<vscale x 8 x i64> %0, <vscale x 8 x i64> %1, i32 %2) nounwind {
+; CHECK-LABEL: intrinsic_vadd_vv_nxv8i64_nxv8i64_nxv8i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e64,m8,ta,mu
+; CHECK-NEXT:    vadd.vv v8, v8, v16
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i64> @llvm.riscv.vadd.nxv8i64.nxv8i64(
+    <vscale x 8 x i64> %0,
+    <vscale x 8 x i64> %1,
+    i32 %2)
+
+  ret <vscale x 8 x i64> %a
+}
+
+declare <vscale x 8 x i64> @llvm.riscv.vadd.mask.nxv8i64.nxv8i64(
+  <vscale x 8 x i64>,
+  <vscale x 8 x i64>,
+  <vscale x 8 x i64>,
+  <vscale x 8 x i1>,
+  i32);
+
+define <vscale x 8 x i64> @intrinsic_vadd_mask_vv_nxv8i64_nxv8i64_nxv8i64(<vscale x 8 x i64> %0, <vscale x 8 x i64> %1, <vscale x 8 x i64> %2, <vscale x 8 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vadd_mask_vv_nxv8i64_nxv8i64_nxv8i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vl8re64.v v24, (a0)
+; CHECK-NEXT:    vsetvli a0, a1, e64,m8,tu,mu
+; CHECK-NEXT:    vadd.vv v8, v16, v24, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i64> @llvm.riscv.vadd.mask.nxv8i64.nxv8i64(
+    <vscale x 8 x i64> %0,
+    <vscale x 8 x i64> %1,
+    <vscale x 8 x i64> %2,
+    <vscale x 8 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 8 x i64> %a
+}
+
 declare <vscale x 1 x i8> @llvm.riscv.vadd.nxv1i8.i8(
   <vscale x 1 x i8>,
   i8,
   i32);
 
 define <vscale x 1 x i8> @intrinsic_vadd_vx_nxv1i8_nxv1i8_i8(<vscale x 1 x i8> %0, i8 %1, i32 %2) nounwind {
 ; CHECK-LABEL: intrinsic_vadd_vx_nxv1i8_nxv1i8_i8:
 ; CHECK:       # %bb.0: # %entry
%a = call <vscale x 16 x i32> @llvm.riscv.vadd.mask.nxv16i32.i32(
     <vscale x 16 x i32> %1,
     i32 %2,
     <vscale x 16 x i1> %3,
     i32 %4)
 
   ret <vscale x 16 x i32> %a
 }
 
+declare <vscale x 1 x i64> @llvm.riscv.vadd.nxv1i64.i64(
+  <vscale x 1 x i64>,
+  i64,
+  i32);
+
+define <vscale x 1 x i64> @intrinsic_vadd_vx_nxv1i64_nxv1i64_i64(<vscale x 1 x i64> %0, i64 %1, i32 %2) nounwind {
+; CHECK-LABEL: intrinsic_vadd_vx_nxv1i64_nxv1i64_i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a2, a2, e64,m1,ta,mu
+; CHECK-NEXT:    vmv.v.x v25, a1
+; CHECK-NEXT:    addi a1, zero, 32
+; CHECK-NEXT:    vsll.vx v25, v25, a1
+; CHECK-NEXT:    vmv.v.x v26, a0
+; CHECK-NEXT:    vsll.vx v26, v26, a1
+; CHECK-NEXT:    vsrl.vx v26, v26, a1
+; CHECK-NEXT:    vor.vv v25, v26, v25
+; CHECK-NEXT:    vadd.vv v8, v8, v25
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i64> @llvm.riscv.vadd.nxv1i64.i64(
+    <vscale x 1 x i64> %0,
+    i64 %1,
+    i32 %2)
+
+  ret <vscale x 1 x i64> %a
+}
+
+declare <vscale x 1 x i64> @llvm.riscv.vadd.mask.nxv1i64.i64(
+  <vscale x 1 x i64>,
+  <vscale x 1 x i64>,
+  i64,
+  <vscale x 1 x i1>,
+  i32);
+
+define <vscale x 1 x i64> @intrinsic_vadd_mask_vx_nxv1i64_nxv1i64_i64(<vscale x 1 x i64> %0, <vscale x 1 x i64> %1, i64 %2, <vscale x 1 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vadd_mask_vx_nxv1i64_nxv1i64_i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a3, a2, e64,m1,ta,mu
+; CHECK-NEXT:    vmv.v.x v25, a1
+; CHECK-NEXT:    addi a1, zero, 32
+; CHECK-NEXT:    vsll.vx v25, v25, a1
+; CHECK-NEXT:    vmv.v.x v26, a0
+; CHECK-NEXT:    vsll.vx v26, v26, a1
+; CHECK-NEXT:    vsrl.vx v26, v26, a1
+; CHECK-NEXT:    vor.vv v25, v26, v25
+; CHECK-NEXT:    vsetvli a0, a2, e64,m1,tu,mu
+; CHECK-NEXT:    vadd.vv v8, v9, v25, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i64> @llvm.riscv.vadd.mask.nxv1i64.i64(
+    <vscale x 1 x i64> %0,
+    <vscale x 1 x i64> %1,
+    i64 %2,
+    <vscale x 1 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 1 x i64> %a
+}
+
+declare <vscale x 2 x i64> @llvm.riscv.vadd.nxv2i64.i64(
+  <vscale x 2 x i64>,
+  i64,
+  i32);
+
+define <vscale x 2 x i64> @intrinsic_vadd_vx_nxv2i64_nxv2i64_i64(<vscale x 2 x i64> %0, i64 %1, i32 %2) nounwind {
+; CHECK-LABEL: intrinsic_vadd_vx_nxv2i64_nxv2i64_i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a2, a2, e64,m2,ta,mu
+; CHECK-NEXT:    vmv.v.x v26, a1
+; CHECK-NEXT:    addi a1, zero, 32
+; CHECK-NEXT:    vsll.vx v26, v26, a1
+; CHECK-NEXT:    vmv.v.x v28, a0
+; CHECK-NEXT:    vsll.vx v28, v28, a1
+; CHECK-NEXT:    vsrl.vx v28, v28, a1
+; CHECK-NEXT:    vor.vv v26, v28, v26
+; CHECK-NEXT:    vadd.vv v8, v8, v26
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i64> @llvm.riscv.vadd.nxv2i64.i64(
+    <vscale x 2 x i64> %0,
+    i64 %1,
+    i32 %2)
+
+  ret <vscale x 2 x i64> %a
+}
+
+declare <vscale x 2 x i64> @llvm.riscv.vadd.mask.nxv2i64.i64(
+  <vscale x 2 x i64>,
+  <vscale x 2 x i64>,
+  i64,
+  <vscale x 2 x i1>,
+  i32);
+
+define <vscale x 2 x i64> @intrinsic_vadd_mask_vx_nxv2i64_nxv2i64_i64(<vscale x 2 x i64> %0, <vscale x 2 x i64> %1, i64 %2, <vscale x 2 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vadd_mask_vx_nxv2i64_nxv2i64_i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a3, a2, e64,m2,ta,mu
+; CHECK-NEXT:    vmv.v.x v26, a1
+; CHECK-NEXT:    addi a1, zero, 32
+; CHECK-NEXT:    vsll.vx v26, v26, a1
+; CHECK-NEXT:    vmv.v.x v28, a0
+; CHECK-NEXT:    vsll.vx v28, v28, a1
+; CHECK-NEXT:    vsrl.vx v28, v28, a1
+; CHECK-NEXT:    vor.vv v26, v28, v26
+; CHECK-NEXT:    vsetvli a0, a2, e64,m2,tu,mu
+; CHECK-NEXT:    vadd.vv v8, v10, v26, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i64> @llvm.riscv.vadd.mask.nxv2i64.i64(
+    <vscale x 2 x i64> %0,
+    <vscale x 2 x i64> %1,
+    i64 %2,
+    <vscale x 2 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 2 x i64> %a
+}
+
+declare <vscale x 4 x i64> @llvm.riscv.vadd.nxv4i64.i64(
+  <vscale x 4 x i64>,
+  i64,
+  i32);
+
+define <vscale x 4 x i64> @intrinsic_vadd_vx_nxv4i64_nxv4i64_i64(<vscale x 4 x i64> %0, i64 %1, i32 %2) nounwind {
+; CHECK-LABEL: intrinsic_vadd_vx_nxv4i64_nxv4i64_i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a2, a2, e64,m4,ta,mu
+; CHECK-NEXT:    vmv.v.x v28, a1
+; CHECK-NEXT:    addi a1, zero, 32
+; CHECK-NEXT:    vsll.vx v28, v28, a1
+; CHECK-NEXT:    vmv.v.x v12, a0
+; CHECK-NEXT:    vsll.vx v12, v12, a1
+; CHECK-NEXT:    vsrl.vx v12, v12, a1
+; CHECK-NEXT:    vor.vv v28, v12, v28
+; CHECK-NEXT:    vadd.vv v8, v8, v28
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i64> @llvm.riscv.vadd.nxv4i64.i64(
+    <vscale x 4 x i64> %0,
+    i64 %1,
+    i32 %2)
+
+  ret <vscale x 4 x i64> %a
+}
+
+declare <vscale x 4 x i64> @llvm.riscv.vadd.mask.nxv4i64.i64(
+  <vscale x 4 x i64>,
+  <vscale x 4 x i64>,
+  i64,
+  <vscale x 4 x i1>,
+  i32);
+
+define <vscale x 4 x i64> @intrinsic_vadd_mask_vx_nxv4i64_nxv4i64_i64(<vscale x 4 x i64> %0, <vscale x 4 x i64> %1, i64 %2, <vscale x 4 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vadd_mask_vx_nxv4i64_nxv4i64_i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a3, a2, e64,m4,ta,mu
+; CHECK-NEXT:    vmv.v.x v28, a1
+; CHECK-NEXT:    addi a1, zero, 32
+; CHECK-NEXT:    vsll.vx v28, v28, a1
+; CHECK-NEXT:    vmv.v.x v16, a0
+; CHECK-NEXT:    vsll.vx v16, v16, a1
+; CHECK-NEXT:    vsrl.vx v16, v16, a1
+; CHECK-NEXT:    vor.vv v28, v16, v28
+; CHECK-NEXT:    vsetvli a0, a2, e64,m4,tu,mu
+; CHECK-NEXT:    vadd.vv v8, v12, v28, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i64> @llvm.riscv.vadd.mask.nxv4i64.i64(
+    <vscale x 4 x i64> %0,
+    <vscale x 4 x i64> %1,
+    i64 %2,
+    <vscale x 4 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 4 x i64> %a
+}
+
+declare <vscale x 8 x i64> @llvm.riscv.vadd.nxv8i64.i64(
+  <vscale x 8 x i64>,
+  i64,
+  i32);
+
+define <vscale x 8 x i64> @intrinsic_vadd_vx_nxv8i64_nxv8i64_i64(<vscale x 8 x i64> %0, i64 %1, i32 %2) nounwind {
+; CHECK-LABEL: intrinsic_vadd_vx_nxv8i64_nxv8i64_i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a2, a2, e64,m8,ta,mu
+; CHECK-NEXT:    vmv.v.x v16, a1
+; CHECK-NEXT:    addi a1, zero, 32
+; CHECK-NEXT:    vsll.vx v16, v16, a1
+; CHECK-NEXT:    vmv.v.x v24, a0
+; CHECK-NEXT:    vsll.vx v24, v24, a1
+; CHECK-NEXT:    vsrl.vx v24, v24, a1
+; CHECK-NEXT:    vor.vv v16, v24, v16
+; CHECK-NEXT:    vadd.vv v8, v8, v16
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i64> @llvm.riscv.vadd.nxv8i64.i64(
+    <vscale x 8 x i64> %0,
+    i64 %1,
+    i32 %2)
+
+  ret <vscale x 8 x i64> %a
+}
+
+declare <vscale x 8 x i64> @llvm.riscv.vadd.mask.nxv8i64.i64(
+  <vscale x 8 x i64>,
+  <vscale x 8 x i64>,
+  i64,
+  <vscale x 8 x i1>,
+  i32);
+
+define <vscale x 8 x i64> @intrinsic_vadd_mask_vx_nxv8i64_nxv8i64_i64(<vscale x 8 x i64> %0, <vscale x 8 x i64> %1, i64 %2, <vscale x 8 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vadd_mask_vx_nxv8i64_nxv8i64_i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    csrrs a3, vlenb, zero
+; CHECK-NEXT:    sub sp, sp, a3
+; CHECK-NEXT:    vs1r.v v0, (sp) # Unknown-size Folded Spill
+; CHECK-NEXT:    vsetvli a3, a2, e64,m8,ta,mu
+; CHECK-NEXT:    vmv.v.x v24, a1
+; CHECK-NEXT:    addi a1, zero, 32
+; CHECK-NEXT:    vsll.vx v0, v24, a1
+; CHECK-NEXT:    vmv.v.x v24, a0
+; CHECK-NEXT:    vsll.vx v24, v24, a1
+; CHECK-NEXT:    vsrl.vx v24, v24, a1
+; CHECK-NEXT:    vor.vv v24, v24, v0
+; CHECK-NEXT:    vsetvli a0, a2, e64,m8,tu,mu
+; CHECK-NEXT:    vl1re8.v v0, (sp) # Unknown-size Folded Reload
+; CHECK-NEXT:    vadd.vv v8, v16, v24, v0.t
+; CHECK-NEXT:    csrrs a0, vlenb, zero
+; CHECK-NEXT:    add sp, sp, a0
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i64> @llvm.riscv.vadd.mask.nxv8i64.i64(
+    <vscale x 8 x i64> %0,
+    <vscale x 8 x i64> %1,
+    i64 %2,
+    <vscale x 8 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 8 x i64> %a
+}
+
 define <vscale x 1 x i8> @intrinsic_vadd_vi_nxv1i8_nxv1i8_i8(<vscale x 1 x i8> %0, i32 %1) nounwind {
 ; CHECK-LABEL: intrinsic_vadd_vi_nxv1i8_nxv1i8_i8:
 ; CHECK:       # %bb.0: # %entry
 ; CHECK-NEXT:    vsetvli a0, a0, e8,mf8,ta,mu
 ; CHECK-NEXT:    vadd.vi v8, v8, 9
 ; CHECK-NEXT:    jalr zero, 0(ra)
 entry:
   %a = call <vscale x 1 x i8> @llvm.riscv.vadd.nxv1i8.i8(
%a = call <vscale x 16 x i32> @llvm.riscv.vadd.mask.nxv16i32.i32(
     <vscale x 16 x i32> %0,
     <vscale x 16 x i32> %1,
     i32 -9,
     <vscale x 16 x i1> %2,
     i32 %3)
 
   ret <vscale x 16 x i32> %a
 }
+
+define <vscale x 1 x i64> @intrinsic_vadd_vi_nxv1i64_nxv1i64_i64(<vscale x 1 x i64> %0, i32 %1) nounwind {
+; CHECK-LABEL: intrinsic_vadd_vi_nxv1i64_nxv1i64_i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e64,m1,ta,mu
+; CHECK-NEXT:    vadd.vi v8, v8, 9
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i64> @llvm.riscv.vadd.nxv1i64.i64(
+    <vscale x 1 x i64> %0,
+    i64 9,
+    i32 %1)
+
+  ret <vscale x 1 x i64> %a
+}
+
+define <vscale x 1 x i64> @intrinsic_vadd_mask_vi_nxv1i64_nxv1i64_i64(<vscale x 1 x i64> %0, <vscale x 1 x i64> %1, <vscale x 1 x i1> %2, i32 %3) nounwind {
+; CHECK-LABEL: intrinsic_vadd_mask_vi_nxv1i64_nxv1i64_i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e64,m1,tu,mu
+; CHECK-NEXT:    vadd.vi v8, v9, 9, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i64> @llvm.riscv.vadd.mask.nxv1i64.i64(
+    <vscale x 1 x i64> %0,
+    <vscale x 1 x i64> %1,
+    i64 9,
+    <vscale x 1 x i1> %2,
+    i32 %3)
+
+  ret <vscale x 1 x i64> %a
+}
+
+define <vscale x 2 x i64> @intrinsic_vadd_vi_nxv2i64_nxv2i64_i64(<vscale x 2 x i64> %0, i32 %1) nounwind {
+; CHECK-LABEL: intrinsic_vadd_vi_nxv2i64_nxv2i64_i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e64,m2,ta,mu
+; CHECK-NEXT:    vadd.vi v8, v8, 9
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i64> @llvm.riscv.vadd.nxv2i64.i64(
+    <vscale x 2 x i64> %0,
+    i64 9,
+    i32 %1)
+
+  ret <vscale x 2 x i64> %a
+}
+
+define <vscale x 2 x i64> @intrinsic_vadd_mask_vi_nxv2i64_nxv2i64_i64(<vscale x 2 x i64> %0, <vscale x 2 x i64> %1, <vscale x 2 x i1> %2, i32 %3) nounwind {
+; CHECK-LABEL: intrinsic_vadd_mask_vi_nxv2i64_nxv2i64_i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e64,m2,tu,mu
+; CHECK-NEXT:    vadd.vi v8, v10, 9, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i64> @llvm.riscv.vadd.mask.nxv2i64.i64(
+    <vscale x 2 x i64> %0,
+    <vscale x 2 x i64> %1,
+    i64 9,
+    <vscale x 2 x i1> %2,
+    i32 %3)
+
+  ret <vscale x 2 x i64> %a
+}
+
+define <vscale x 4 x i64> @intrinsic_vadd_vi_nxv4i64_nxv4i64_i64(<vscale x 4 x i64> %0, i32 %1) nounwind {
+; CHECK-LABEL: intrinsic_vadd_vi_nxv4i64_nxv4i64_i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e64,m4,ta,mu
+; CHECK-NEXT:    vadd.vi v8, v8, 9
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i64> @llvm.riscv.vadd.nxv4i64.i64(
+    <vscale x 4 x i64> %0,
+    i64 9,
+    i32 %1)
+
+  ret <vscale x 4 x i64> %a
+}
+
+define <vscale x 4 x i64> @intrinsic_vadd_mask_vi_nxv4i64_nxv4i64_i64(<vscale x 4 x i64> %0, <vscale x 4 x i64> %1, <vscale x 4 x i1> %2, i32 %3) nounwind {
+; CHECK-LABEL: intrinsic_vadd_mask_vi_nxv4i64_nxv4i64_i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e64,m4,tu,mu
+; CHECK-NEXT:    vadd.vi v8, v12, 9, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i64> @llvm.riscv.vadd.mask.nxv4i64.i64(
+    <vscale x 4 x i64> %0,
+    <vscale x 4 x i64> %1,
+    i64 9,
+    <vscale x 4 x i1> %2,
+    i32 %3)
+
+  ret <vscale x 4 x i64> %a
+}
+
+define <vscale x 8 x i64> @intrinsic_vadd_vi_nxv8i64_nxv8i64_i64(<vscale x 8 x i64> %0, i32 %1) nounwind {
+; CHECK-LABEL: intrinsic_vadd_vi_nxv8i64_nxv8i64_i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e64,m8,ta,mu
+; CHECK-NEXT:    vadd.vi v8, v8, 9
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i64> @llvm.riscv.vadd.nxv8i64.i64(
+    <vscale x 8 x i64> %0,
+    i64 9,
+    i32 %1)
+
+  ret <vscale x 8 x i64> %a
+}
+
+define <vscale x 8 x i64> @intrinsic_vadd_mask_vi_nxv8i64_nxv8i64_i64(<vscale x 8 x i64> %0, <vscale x 8 x i64> %1, <vscale x 8 x i1> %2, i32 %3) nounwind {
+; CHECK-LABEL: intrinsic_vadd_mask_vi_nxv8i64_nxv8i64_i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e64,m8,tu,mu
+; CHECK-NEXT:    vadd.vi v8, v16, 9, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i64> @llvm.riscv.vadd.mask.nxv8i64.i64(
+    <vscale x 8 x i64> %0,
+    <vscale x 8 x i64> %1,
+    i64 9,
+    <vscale x 8 x i1> %2,
+    i32 %3)
+
+  ret <vscale x 8 x i64> %a
+}

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

%a = call <vscale x 16 x i32> @llvm.riscv.vand.mask.nxv16i32.nxv16i32(
     <vscale x 16 x i32> %1,
     <vscale x 16 x i32> %2,
     <vscale x 16 x i1> %3,
     i32 %4)
 
   ret <vscale x 16 x i32> %a
 }
 
+declare <vscale x 1 x i64> @llvm.riscv.vand.nxv1i64.nxv1i64(
+  <vscale x 1 x i64>,
+  <vscale x 1 x i64>,
+  i32);
+
+define <vscale x 1 x i64> @intrinsic_vand_vv_nxv1i64_nxv1i64_nxv1i64(<vscale x 1 x i64> %0, <vscale x 1 x i64> %1, i32 %2) nounwind {
+; CHECK-LABEL: intrinsic_vand_vv_nxv1i64_nxv1i64_nxv1i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e64,m1,ta,mu
+; CHECK-NEXT:    vand.vv v8, v8, v9
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i64> @llvm.riscv.vand.nxv1i64.nxv1i64(
+    <vscale x 1 x i64> %0,
+    <vscale x 1 x i64> %1,
+    i32 %2)
+
+  ret <vscale x 1 x i64> %a
+}
+
+declare <vscale x 1 x i64> @llvm.riscv.vand.mask.nxv1i64.nxv1i64(
+  <vscale x 1 x i64>,
+  <vscale x 1 x i64>,
+  <vscale x 1 x i64>,
+  <vscale x 1 x i1>,
+  i32);
+
+define <vscale x 1 x i64> @intrinsic_vand_mask_vv_nxv1i64_nxv1i64_nxv1i64(<vscale x 1 x i64> %0, <vscale x 1 x i64> %1, <vscale x 1 x i64> %2, <vscale x 1 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vand_mask_vv_nxv1i64_nxv1i64_nxv1i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e64,m1,tu,mu
+; CHECK-NEXT:    vand.vv v8, v9, v10, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i64> @llvm.riscv.vand.mask.nxv1i64.nxv1i64(
+    <vscale x 1 x i64> %0,
+    <vscale x 1 x i64> %1,
+    <vscale x 1 x i64> %2,
+    <vscale x 1 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 1 x i64> %a
+}
+
+declare <vscale x 2 x i64> @llvm.riscv.vand.nxv2i64.nxv2i64(
+  <vscale x 2 x i64>,
+  <vscale x 2 x i64>,
+  i32);
+
+define <vscale x 2 x i64> @intrinsic_vand_vv_nxv2i64_nxv2i64_nxv2i64(<vscale x 2 x i64> %0, <vscale x 2 x i64> %1, i32 %2) nounwind {
+; CHECK-LABEL: intrinsic_vand_vv_nxv2i64_nxv2i64_nxv2i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e64,m2,ta,mu
+; CHECK-NEXT:    vand.vv v8, v8, v10
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i64> @llvm.riscv.vand.nxv2i64.nxv2i64(
+    <vscale x 2 x i64> %0,
+    <vscale x 2 x i64> %1,
+    i32 %2)
+
+  ret <vscale x 2 x i64> %a
+}
+
+declare <vscale x 2 x i64> @llvm.riscv.vand.mask.nxv2i64.nxv2i64(
+  <vscale x 2 x i64>,
+  <vscale x 2 x i64>,
+  <vscale x 2 x i64>,
+  <vscale x 2 x i1>,
+  i32);
+
+define <vscale x 2 x i64> @intrinsic_vand_mask_vv_nxv2i64_nxv2i64_nxv2i64(<vscale x 2 x i64> %0, <vscale x 2 x i64> %1, <vscale x 2 x i64> %2, <vscale x 2 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vand_mask_vv_nxv2i64_nxv2i64_nxv2i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e64,m2,tu,mu
+; CHECK-NEXT:    vand.vv v8, v10, v12, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i64> @llvm.riscv.vand.mask.nxv2i64.nxv2i64(
+    <vscale x 2 x i64> %0,
+    <vscale x 2 x i64> %1,
+    <vscale x 2 x i64> %2,
+    <vscale x 2 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 2 x i64> %a
+}
+
+declare <vscale x 4 x i64> @llvm.riscv.vand.nxv4i64.nxv4i64(
+  <vscale x 4 x i64>,
+  <vscale x 4 x i64>,
+  i32);
+
+define <vscale x 4 x i64> @intrinsic_vand_vv_nxv4i64_nxv4i64_nxv4i64(<vscale x 4 x i64> %0, <vscale x 4 x i64> %1, i32 %2) nounwind {
+; CHECK-LABEL: intrinsic_vand_vv_nxv4i64_nxv4i64_nxv4i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e64,m4,ta,mu
+; CHECK-NEXT:    vand.vv v8, v8, v12
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i64> @llvm.riscv.vand.nxv4i64.nxv4i64(
+    <vscale x 4 x i64> %0,
+    <vscale x 4 x i64> %1,
+    i32 %2)
+
+  ret <vscale x 4 x i64> %a
+}
+
+declare <vscale x 4 x i64> @llvm.riscv.vand.mask.nxv4i64.nxv4i64(
+  <vscale x 4 x i64>,
+  <vscale x 4 x i64>,
+  <vscale x 4 x i64>,
+  <vscale x 4 x i1>,
+  i32);
+
+define <vscale x 4 x i64> @intrinsic_vand_mask_vv_nxv4i64_nxv4i64_nxv4i64(<vscale x 4 x i64> %0, <vscale x 4 x i64> %1, <vscale x 4 x i64> %2, <vscale x 4 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vand_mask_vv_nxv4i64_nxv4i64_nxv4i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e64,m4,tu,mu
+; CHECK-NEXT:    vand.vv v8, v12, v16, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i64> @llvm.riscv.vand.mask.nxv4i64.nxv4i64(
+    <vscale x 4 x i64> %0,
+    <vscale x 4 x i64> %1,
+    <vscale x 4 x i64> %2,
+    <vscale x 4 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 4 x i64> %a
+}
+
+declare <vscale x 8 x i64> @llvm.riscv.vand.nxv8i64.nxv8i64(
+  <vscale x 8 x i64>,
+  <vscale x 8 x i64>,
+  i32);
+
+define <vscale x 8 x i64> @intrinsic_vand_vv_nxv8i64_nxv8i64_nxv8i64(<vscale x 8 x i64> %0, <vscale x 8 x i64> %1, i32 %2) nounwind {
+; CHECK-LABEL: intrinsic_vand_vv_nxv8i64_nxv8i64_nxv8i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e64,m8,ta,mu
+; CHECK-NEXT:    vand.vv v8, v8, v16
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i64> @llvm.riscv.vand.nxv8i64.nxv8i64(
+    <vscale x 8 x i64> %0,
+    <vscale x 8 x i64> %1,
+    i32 %2)
+
+  ret <vscale x 8 x i64> %a
+}
+
+declare <vscale x 8 x i64> @llvm.riscv.vand.mask.nxv8i64.nxv8i64(
+  <vscale x 8 x i64>,
+  <vscale x 8 x i64>,
+  <vscale x 8 x i64>,
+  <vscale x 8 x i1>,
+  i32);
+
+define <vscale x 8 x i64> @intrinsic_vand_mask_vv_nxv8i64_nxv8i64_nxv8i64(<vscale x 8 x i64> %0, <vscale x 8 x i64> %1, <vscale x 8 x i64> %2, <vscale x 8 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vand_mask_vv_nxv8i64_nxv8i64_nxv8i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vl8re64.v v24, (a0)
+; CHECK-NEXT:    vsetvli a0, a1, e64,m8,tu,mu
+; CHECK-NEXT:    vand.vv v8, v16, v24, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i64> @llvm.riscv.vand.mask.nxv8i64.nxv8i64(
+    <vscale x 8 x i64> %0,
+    <vscale x 8 x i64> %1,
+    <vscale x 8 x i64> %2,
+    <vscale x 8 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 8 x i64> %a
+}
+
 declare <vscale x 1 x i8> @llvm.riscv.vand.nxv1i8.i8(
   <vscale x 1 x i8>,
   i8,
   i32);
 
 define <vscale x 1 x i8> @intrinsic_vand_vx_nxv1i8_nxv1i8_i8(<vscale x 1 x i8> %0, i8 %1, i32 %2) nounwind {
 ; CHECK-LABEL: intrinsic_vand_vx_nxv1i8_nxv1i8_i8:
 ; CHECK:       # %bb.0: # %entry
%a = call <vscale x 16 x i32> @llvm.riscv.vand.mask.nxv16i32.i32(
     <vscale x 16 x i32> %1,
     i32 %2,
     <vscale x 16 x i1> %3,
     i32 %4)
 
   ret <vscale x 16 x i32> %a
 }
 
+declare <vscale x 1 x i64> @llvm.riscv.vand.nxv1i64.i64(
+  <vscale x 1 x i64>,
+  i64,
+  i32);
+
+define <vscale x 1 x i64> @intrinsic_vand_vx_nxv1i64_nxv1i64_i64(<vscale x 1 x i64> %0, i64 %1, i32 %2) nounwind {
+; CHECK-LABEL: intrinsic_vand_vx_nxv1i64_nxv1i64_i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a2, a2, e64,m1,ta,mu
+; CHECK-NEXT:    vmv.v.x v25, a1
+; CHECK-NEXT:    addi a1, zero, 32
+; CHECK-NEXT:    vsll.vx v25, v25, a1
+; CHECK-NEXT:    vmv.v.x v26, a0
+; CHECK-NEXT:    vsll.vx v26, v26, a1
+; CHECK-NEXT:    vsrl.vx v26, v26, a1
+; CHECK-NEXT:    vor.vv v25, v26, v25
+; CHECK-NEXT:    vand.vv v8, v8, v25
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i64> @llvm.riscv.vand.nxv1i64.i64(
+    <vscale x 1 x i64> %0,
+    i64 %1,
+    i32 %2)
+
+  ret <vscale x 1 x i64> %a
+}
+
+declare <vscale x 1 x i64> @llvm.riscv.vand.mask.nxv1i64.i64(
+  <vscale x 1 x i64>,
+  <vscale x 1 x i64>,
+  i64,
+  <vscale x 1 x i1>,
+  i32);
+
+define <vscale x 1 x i64> @intrinsic_vand_mask_vx_nxv1i64_nxv1i64_i64(<vscale x 1 x i64> %0, <vscale x 1 x i64> %1, i64 %2, <vscale x 1 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vand_mask_vx_nxv1i64_nxv1i64_i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a3, a2, e64,m1,ta,mu
+; CHECK-NEXT:    vmv.v.x v25, a1
+; CHECK-NEXT:    addi a1, zero, 32
+; CHECK-NEXT:    vsll.vx v25, v25, a1
+; CHECK-NEXT:    vmv.v.x v26, a0
+; CHECK-NEXT:    vsll.vx v26, v26, a1
+; CHECK-NEXT:    vsrl.vx v26, v26, a1
+; CHECK-NEXT:    vor.vv v25, v26, v25
+; CHECK-NEXT:    vsetvli a0, a2, e64,m1,tu,mu
+; CHECK-NEXT:    vand.vv v8, v9, v25, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i64> @llvm.riscv.vand.mask.nxv1i64.i64(
+    <vscale x 1 x i64> %0,
+    <vscale x 1 x i64> %1,
+    i64 %2,
+    <vscale x 1 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 1 x i64> %a
+}
+
+declare <vscale x 2 x i64> @llvm.riscv.vand.nxv2i64.i64(
+  <vscale x 2 x i64>,
+  i64,
+  i32);
+
+define <vscale x 2 x i64> @intrinsic_vand_vx_nxv2i64_nxv2i64_i64(<vscale x 2 x i64> %0, i64 %1, i32 %2) nounwind {
+; CHECK-LABEL: intrinsic_vand_vx_nxv2i64_nxv2i64_i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a2, a2, e64,m2,ta,mu
+; CHECK-NEXT:    vmv.v.x v26, a1
+; CHECK-NEXT:    addi a1, zero, 32
+; CHECK-NEXT:    vsll.vx v26, v26, a1
+; CHECK-NEXT:    vmv.v.x v28, a0
+; CHECK-NEXT:    vsll.vx v28, v28, a1
+; CHECK-NEXT:    vsrl.vx v28, v28, a1
+; CHECK-NEXT:    vor.vv v26, v28, v26
+; CHECK-NEXT:    vand.vv v8, v8, v26
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i64> @llvm.riscv.vand.nxv2i64.i64(
+    <vscale x 2 x i64> %0,
+    i64 %1,
+    i32 %2)
+
+  ret <vscale x 2 x i64> %a
+}
+
+declare <vscale x 2 x i64> @llvm.riscv.vand.mask.nxv2i64.i64(
+  <vscale x 2 x i64>,
+  <vscale x 2 x i64>,
+  i64,
+  <vscale x 2 x i1>,
+  i32);
+
+define <vscale x 2 x i64> @intrinsic_vand_mask_vx_nxv2i64_nxv2i64_i64(<vscale x 2 x i64> %0, <vscale x 2 x i64> %1, i64 %2, <vscale x 2 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vand_mask_vx_nxv2i64_nxv2i64_i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a3, a2, e64,m2,ta,mu
+; CHECK-NEXT:    vmv.v.x v26, a1
+; CHECK-NEXT:    addi a1, zero, 32
+; CHECK-NEXT:    vsll.vx v26, v26, a1
+; CHECK-NEXT:    vmv.v.x v28, a0
+; CHECK-NEXT:    vsll.vx v28, v28, a1
+; CHECK-NEXT:    vsrl.vx v28, v28, a1
+; CHECK-NEXT:    vor.vv v26, v28, v26
+; CHECK-NEXT:    vsetvli a0, a2, e64,m2,tu,mu
+; CHECK-NEXT:    vand.vv v8, v10, v26, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i64> @llvm.riscv.vand.mask.nxv2i64.i64(
+    <vscale x 2 x i64> %0,
+    <vscale x 2 x i64> %1,
+    i64 %2,
+    <vscale x 2 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 2 x i64> %a
+}
+
+declare <vscale x 4 x i64> @llvm.riscv.vand.nxv4i64.i64(
+  <vscale x 4 x i64>,
+  i64,
+  i32);
+
+define <vscale x 4 x i64> @intrinsic_vand_vx_nxv4i64_nxv4i64_i64(<vscale x 4 x i64> %0, i64 %1, i32 %2) nounwind {
+; CHECK-LABEL: intrinsic_vand_vx_nxv4i64_nxv4i64_i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a2, a2, e64,m4,ta,mu
+; CHECK-NEXT:    vmv.v.x v28, a1
+; CHECK-NEXT:    addi a1, zero, 32
+; CHECK-NEXT:    vsll.vx v28, v28, a1
+; CHECK-NEXT:    vmv.v.x v12, a0
+; CHECK-NEXT:    vsll.vx v12, v12, a1
+; CHECK-NEXT:    vsrl.vx v12, v12, a1
+; CHECK-NEXT:    vor.vv v28, v12, v28
+; CHECK-NEXT:    vand.vv v8, v8, v28
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i64> @llvm.riscv.vand.nxv4i64.i64(
+    <vscale x 4 x i64> %0,
+    i64 %1,
+    i32 %2)
+
+  ret <vscale x 4 x i64> %a
+}
+
+declare <vscale x 4 x i64> @llvm.riscv.vand.mask.nxv4i64.i64(
+  <vscale x 4 x i64>,
+  <vscale x 4 x i64>,
+  i64,
+  <vscale x 4 x i1>,
+  i32);
+
+define <vscale x 4 x i64> @intrinsic_vand_mask_vx_nxv4i64_nxv4i64_i64(<vscale x 4 x i64> %0, <vscale x 4 x i64> %1, i64 %2, <vscale x 4 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vand_mask_vx_nxv4i64_nxv4i64_i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a3, a2, e64,m4,ta,mu
+; CHECK-NEXT:    vmv.v.x v28, a1
+; CHECK-NEXT:    addi a1, zero, 32
+; CHECK-NEXT:    vsll.vx v28, v28, a1
+; CHECK-NEXT:    vmv.v.x v16, a0
+; CHECK-NEXT:    vsll.vx v16, v16, a1
+; CHECK-NEXT:    vsrl.vx v16, v16, a1
+; CHECK-NEXT:    vor.vv v28, v16, v28
+; CHECK-NEXT:    vsetvli a0, a2, e64,m4,tu,mu
+; CHECK-NEXT:    vand.vv v8, v12, v28, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i64> @llvm.riscv.vand.mask.nxv4i64.i64(
+    <vscale x 4 x i64> %0,
+    <vscale x 4 x i64> %1,
+    i64 %2,
+    <vscale x 4 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 4 x i64> %a
+}
+
+declare <vscale x 8 x i64> @llvm.riscv.vand.nxv8i64.i64(
+  <vscale x 8 x i64>,
+  i64,
+  i32);
+
+define <vscale x 8 x i64> @intrinsic_vand_vx_nxv8i64_nxv8i64_i64(<vscale x 8 x i64> %0, i64 %1, i32 %2) nounwind {
+; CHECK-LABEL: intrinsic_vand_vx_nxv8i64_nxv8i64_i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a2, a2, e64,m8,ta,mu
+; CHECK-NEXT:    vmv.v.x v16, a1
+; CHECK-NEXT:    addi a1, zero, 32
+; CHECK-NEXT:    vsll.vx v16, v16, a1
+; CHECK-NEXT:    vmv.v.x v24, a0
+; CHECK-NEXT:    vsll.vx v24, v24, a1
+; CHECK-NEXT:    vsrl.vx v24, v24, a1
+; CHECK-NEXT:    vor.vv v16, v24, v16
+; CHECK-NEXT:    vand.vv v8, v8, v16
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i64> @llvm.riscv.vand.nxv8i64.i64(
+    <vscale x 8 x i64> %0,
+    i64 %1,
+    i32 %2)
+
+  ret <vscale x 8 x i64> %a
+}
+
+declare <vscale x 8 x i64> @llvm.riscv.vand.mask.nxv8i64.i64(
+  <vscale x 8 x i64>,
+  <vscale x 8 x i64>,
+  i64,
+  <vscale x 8 x i1>,
+  i32);
+
+define <vscale x 8 x i64> @intrinsic_vand_mask_vx_nxv8i64_nxv8i64_i64(<vscale x 8 x i64> %0, <vscale x 8 x i64> %1, i64 %2, <vscale x 8 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vand_mask_vx_nxv8i64_nxv8i64_i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    csrrs a3, vlenb, zero
+; CHECK-NEXT:    sub sp, sp, a3
+; CHECK-NEXT:    vs1r.v v0, (sp) # Unknown-size Folded Spill
+; CHECK-NEXT:    vsetvli a3, a2, e64,m8,ta,mu
+; CHECK-NEXT:    vmv.v.x v24, a1
+; CHECK-NEXT:    addi a1, zero, 32
+; CHECK-NEXT:    vsll.vx v0, v24, a1
+; CHECK-NEXT:    vmv.v.x v24, a0
+; CHECK-NEXT:    vsll.vx v24, v24, a1
+; CHECK-NEXT:    vsrl.vx v24, v24, a1
+; CHECK-NEXT:    vor.vv v24, v24, v0
+; CHECK-NEXT:    vsetvli a0, a2, e64,m8,tu,mu
+; CHECK-NEXT:    vl1re8.v v0, (sp) # Unknown-size Folded Reload
+; CHECK-NEXT:    vand.vv v8, v16, v24, v0.t
+; CHECK-NEXT:    csrrs a0, vlenb, zero
+; CHECK-NEXT:    add sp, sp, a0
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i64> @llvm.riscv.vand.mask.nxv8i64.i64(
+    <vscale x 8 x i64> %0,
+    <vscale x 8 x i64> %1,
+    i64 %2,
+    <vscale x 8 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 8 x i64> %a
+}
+
 define <vscale x 1 x i8> @intrinsic_vand_vi_nxv1i8_nxv1i8_i8(<vscale x 1 x i8> %0, i32 %1) nounwind {
 ; CHECK-LABEL: intrinsic_vand_vi_nxv1i8_nxv1i8_i8:
 ; CHECK:       # %bb.0: # %entry
 ; CHECK-NEXT:    vsetvli a0, a0, e8,mf8,ta,mu
 ; CHECK-NEXT:    vand.vi v8, v8, 9
 ; CHECK-NEXT:    jalr zero, 0(ra)
 entry:
   %a = call <vscale x 1 x i8> @llvm.riscv.vand.nxv1i8.i8(
%a = call <vscale x 16 x i32> @llvm.riscv.vand.mask.nxv16i32.i32(
     <vscale x 16 x i32> %0,
     <vscale x 16 x i32> %1,
     i32 9,
     <vscale x 16 x i1> %2,
     i32 %3)
 
   ret <vscale x 16 x i32> %a
 }
+
+define <vscale x 1 x i64> @intrinsic_vand_vi_nxv1i64_nxv1i64_i64(<vscale x 1 x i64> %0, i32 %1) nounwind {
+; CHECK-LABEL: intrinsic_vand_vi_nxv1i64_nxv1i64_i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e64,m1,ta,mu
+; CHECK-NEXT:    vand.vi v8, v8, 9
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i64> @llvm.riscv.vand.nxv1i64.i64(
+    <vscale x 1 x i64> %0,
+    i64 9,
+    i32 %1)
+
+  ret <vscale x 1 x i64> %a
+}
+
+define <vscale x 1 x i64> @intrinsic_vand_mask_vi_nxv1i64_nxv1i64_i64(<vscale x 1 x i64> %0, <vscale x 1 x i64> %1, <vscale x 1 x i1> %2, i32 %3) nounwind {
+; CHECK-LABEL: intrinsic_vand_mask_vi_nxv1i64_nxv1i64_i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e64,m1,tu,mu
+; CHECK-NEXT:    vand.vi v8, v9, 9, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i64> @llvm.riscv.vand.mask.nxv1i64.i64(
+    <vscale x 1 x i64> %0,
+    <vscale x 1 x i64> %1,
+    i64 9,
+    <vscale x 1 x i1> %2,
+    i32 %3)
+
+  ret <vscale x 1 x i64> %a
+}
+
+define <vscale x 2 x i64> @intrinsic_vand_vi_nxv2i64_nxv2i64_i64(<vscale x 2 x i64> %0, i32 %1) nounwind {
+; CHECK-LABEL: intrinsic_vand_vi_nxv2i64_nxv2i64_i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e64,m2,ta,mu
+; CHECK-NEXT:    vand.vi v8, v8, 9
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i64> @llvm.riscv.vand.nxv2i64.i64(
+    <vscale x 2 x i64> %0,
+    i64 9,
+    i32 %1)
+
+  ret <vscale x 2 x i64> %a
+}
+
+define <vscale x 2 x i64> @intrinsic_vand_mask_vi_nxv2i64_nxv2i64_i64(<vscale x 2 x i64> %0, <vscale x 2 x i64> %1, <vscale x 2 x i1> %2, i32 %3) nounwind {
+; CHECK-LABEL: intrinsic_vand_mask_vi_nxv2i64_nxv2i64_i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e64,m2,tu,mu
+; CHECK-NEXT:    vand.vi v8, v10, 9, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i64> @llvm.riscv.vand.mask.nxv2i64.i64(
+    <vscale x 2 x i64> %0,
+    <vscale x 2 x i64> %1,
+    i64 9,
+    <vscale x 2 x i1> %2,
+    i32 %3)
+
+  ret <vscale x 2 x i64> %a
+}
+
+define <vscale x 4 x i64> @intrinsic_vand_vi_nxv4i64_nxv4i64_i64(<vscale x 4 x i64> %0, i32 %1) nounwind {
+; CHECK-LABEL: intrinsic_vand_vi_nxv4i64_nxv4i64_i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e64,m4,ta,mu
+; CHECK-NEXT:    vand.vi v8, v8, 9
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i64> @llvm.riscv.vand.nxv4i64.i64(
+    <vscale x 4 x i64> %0,
+    i64 9,
+    i32 %1)
+
+  ret <vscale x 4 x i64> %a
+}
+
+define <vscale x 4 x i64> @intrinsic_vand_mask_vi_nxv4i64_nxv4i64_i64(<vscale x 4 x i64> %0, <vscale x 4 x i64> %1, <vscale x 4 x i1> %2, i32 %3) nounwind {
+; CHECK-LABEL: intrinsic_vand_mask_vi_nxv4i64_nxv4i64_i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e64,m4,tu,mu
+; CHECK-NEXT:    vand.vi v8, v12, 9, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i64> @llvm.riscv.vand.mask.nxv4i64.i64(
+    <vscale x 4 x i64> %0,
+    <vscale x 4 x i64> %1,
+    i64 9,
+    <vscale x 4 x i1> %2,
+    i32 %3)
+
+  ret <vscale x 4 x i64> %a
+}
+
+define <vscale x 8 x i64> @intrinsic_vand_vi_nxv8i64_nxv8i64_i64(<vscale x 8 x i64> %0, i32 %1) nounwind {
+; CHECK-LABEL: intrinsic_vand_vi_nxv8i64_nxv8i64_i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e64,m8,ta,mu
+; CHECK-NEXT:    vand.vi v8, v8, 9
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i64> @llvm.riscv.vand.nxv8i64.i64(
+    <vscale x 8 x i64> %0,
+    i64 9,
+    i32 %1)
+
+  ret <vscale x 8 x i64> %a
+}
+
+define <vscale x 8 x i64> @intrinsic_vand_mask_vi_nxv8i64_nxv8i64_i64(<vscale x 8 x i64> %0, <vscale x 8 x i64> %1, <vscale x 8 x i1> %2, i32 %3) nounwind {
+; CHECK-LABEL: intrinsic_vand_mask_vi_nxv8i64_nxv8i64_i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e64,m8,tu,mu
+; CHECK-NEXT:    vand.vi v8, v16, 9, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i64> @llvm.riscv.vand.mask.nxv8i64.i64(
+    <vscale x 8 x i64> %0,
+    <vscale x 8 x i64> %1,
+    i64 9,
+    <vscale x 8 x i1> %2,
+    i32 %3)
+
+  ret <vscale x 8 x i64> %a
+}

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

%a = call <vscale x 16 x i32> @llvm.riscv.vasub.mask.nxv16i32.nxv16i32(
     <vscale x 16 x i32> %1,
     <vscale x 16 x i32> %2,
     <vscale x 16 x i1> %3,
     i32 %4)
 
   ret <vscale x 16 x i32> %a
 }
 
+declare <vscale x 1 x i64> @llvm.riscv.vasub.nxv1i64.nxv1i64(
+  <vscale x 1 x i64>,
+  <vscale x 1 x i64>,
+  i32);
+
+define <vscale x 1 x i64> @intrinsic_vasub_vv_nxv1i64_nxv1i64_nxv1i64(<vscale x 1 x i64> %0, <vscale x 1 x i64> %1, i32 %2) nounwind {
+; CHECK-LABEL: intrinsic_vasub_vv_nxv1i64_nxv1i64_nxv1i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e64,m1,ta,mu
+; CHECK-NEXT:    vasub.vv v8, v8, v9
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i64> @llvm.riscv.vasub.nxv1i64.nxv1i64(
+    <vscale x 1 x i64> %0,
+    <vscale x 1 x i64> %1,
+    i32 %2)
+
+  ret <vscale x 1 x i64> %a
+}
+
+declare <vscale x 1 x i64> @llvm.riscv.vasub.mask.nxv1i64.nxv1i64(
+  <vscale x 1 x i64>,
+  <vscale x 1 x i64>,
+  <vscale x 1 x i64>,
+  <vscale x 1 x i1>,
+  i32);
+
+define <vscale x 1 x i64> @intrinsic_vasub_mask_vv_nxv1i64_nxv1i64_nxv1i64(<vscale x 1 x i64> %0, <vscale x 1 x i64> %1, <vscale x 1 x i64> %2, <vscale x 1 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vasub_mask_vv_nxv1i64_nxv1i64_nxv1i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e64,m1,tu,mu
+; CHECK-NEXT:    vasub.vv v8, v9, v10, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i64> @llvm.riscv.vasub.mask.nxv1i64.nxv1i64(
+    <vscale x 1 x i64> %0,
+    <vscale x 1 x i64> %1,
+    <vscale x 1 x i64> %2,
+    <vscale x 1 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 1 x i64> %a
+}
+
+declare <vscale x 2 x i64> @llvm.riscv.vasub.nxv2i64.nxv2i64(
+  <vscale x 2 x i64>,
+  <vscale x 2 x i64>,
+  i32);
+
+define <vscale x 2 x i64> @intrinsic_vasub_vv_nxv2i64_nxv2i64_nxv2i64(<vscale x 2 x i64> %0, <vscale x 2 x i64> %1, i32 %2) nounwind {
+; CHECK-LABEL: intrinsic_vasub_vv_nxv2i64_nxv2i64_nxv2i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e64,m2,ta,mu
+; CHECK-NEXT:    vasub.vv v8, v8, v10
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i64> @llvm.riscv.vasub.nxv2i64.nxv2i64(
+    <vscale x 2 x i64> %0,
+    <vscale x 2 x i64> %1,
+    i32 %2)
+
+  ret <vscale x 2 x i64> %a
+}
+
+declare <vscale x 2 x i64> @llvm.riscv.vasub.mask.nxv2i64.nxv2i64(
+  <vscale x 2 x i64>,
+  <vscale x 2 x i64>,
+  <vscale x 2 x i64>,
+  <vscale x 2 x i1>,
+  i32);
+
+define <vscale x 2 x i64> @intrinsic_vasub_mask_vv_nxv2i64_nxv2i64_nxv2i64(<vscale x 2 x i64> %0, <vscale x 2 x i64> %1, <vscale x 2 x i64> %2, <vscale x 2 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vasub_mask_vv_nxv2i64_nxv2i64_nxv2i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e64,m2,tu,mu
+; CHECK-NEXT:    vasub.vv v8, v10, v12, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i64> @llvm.riscv.vasub.mask.nxv2i64.nxv2i64(
+    <vscale x 2 x i64> %0,
+    <vscale x 2 x i64> %1,
+    <vscale x 2 x i64> %2,
+    <vscale x 2 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 2 x i64> %a
+}
+
+declare <vscale x 4 x i64> @llvm.riscv.vasub.nxv4i64.nxv4i64(
+  <vscale x 4 x i64>,
+  <vscale x 4 x i64>,
+  i32);
+
+define <vscale x 4 x i64> @intrinsic_vasub_vv_nxv4i64_nxv4i64_nxv4i64(<vscale x 4 x i64> %0, <vscale x 4 x i64> %1, i32 %2) nounwind {
+; CHECK-LABEL: intrinsic_vasub_vv_nxv4i64_nxv4i64_nxv4i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e64,m4,ta,mu
+; CHECK-NEXT:    vasub.vv v8, v8, v12
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i64> @llvm.riscv.vasub.nxv4i64.nxv4i64(
+    <vscale x 4 x i64> %0,
+    <vscale x 4 x i64> %1,
+    i32 %2)
+
+  ret <vscale x 4 x i64> %a
+}
+
+declare <vscale x 4 x i64> @llvm.riscv.vasub.mask.nxv4i64.nxv4i64(
+  <vscale x 4 x i64>,
+  <vscale x 4 x i64>,
+  <vscale x 4 x i64>,
+  <vscale x 4 x i1>,
+  i32);
+
+define <vscale x 4 x i64> @intrinsic_vasub_mask_vv_nxv4i64_nxv4i64_nxv4i64(<vscale x 4 x i64> %0, <vscale x 4 x i64> %1, <vscale x 4 x i64> %2, <vscale x 4 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vasub_mask_vv_nxv4i64_nxv4i64_nxv4i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e64,m4,tu,mu
+; CHECK-NEXT:    vasub.vv v8, v12, v16, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i64> @llvm.riscv.vasub.mask.nxv4i64.nxv4i64(
+    <vscale x 4 x i64> %0,
+    <vscale x 4 x i64> %1,
+    <vscale x 4 x i64> %2,
+    <vscale x 4 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 4 x i64> %a
+}
+
+declare <vscale x 8 x i64> @llvm.riscv.vasub.nxv8i64.nxv8i64(
+  <vscale x 8 x i64>,
+  <vscale x 8 x i64>,
+  i32);
+
+define <vscale x 8 x i64> @intrinsic_vasub_vv_nxv8i64_nxv8i64_nxv8i64(<vscale x 8 x i64> %0, <vscale x 8 x i64> %1, i32 %2) nounwind {
+; CHECK-LABEL: intrinsic_vasub_vv_nxv8i64_nxv8i64_nxv8i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e64,m8,ta,mu
+; CHECK-NEXT:    vasub.vv v8, v8, v16
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i64> @llvm.riscv.vasub.nxv8i64.nxv8i64(
+    <vscale x 8 x i64> %0,
+    <vscale x 8 x i64> %1,
+    i32 %2)
+
+  ret <vscale x 8 x i64> %a
+}
+
+declare <vscale x 8 x i64> @llvm.riscv.vasub.mask.nxv8i64.nxv8i64(
+  <vscale x 8 x i64>,
+  <vscale x 8 x i64>,
+  <vscale x 8 x i64>,
+  <vscale x 8 x i1>,
+  i32);
+
+define <vscale x 8 x i64> @intrinsic_vasub_mask_vv_nxv8i64_nxv8i64_nxv8i64(<vscale x 8 x i64> %0, <vscale x 8 x i64> %1, <vscale x 8 x i64> %2, <vscale x 8 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vasub_mask_vv_nxv8i64_nxv8i64_nxv8i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vl8re64.v v24, (a0)
+; CHECK-NEXT:    vsetvli a0, a1, e64,m8,tu,mu
+; CHECK-NEXT:    vasub.vv v8, v16, v24, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i64> @llvm.riscv.vasub.mask.nxv8i64.nxv8i64(
+    <vscale x 8 x i64> %0,
+    <vscale x 8 x i64> %1,
+    <vscale x 8 x i64> %2,
+    <vscale x 8 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 8 x i64> %a
+}
+
 declare <vscale x 1 x i8> @llvm.riscv.vasub.nxv1i8.i8(
   <vscale x 1 x i8>,
   i8,
   i32);
 
 define <vscale x 1 x i8> @intrinsic_vasub_vx_nxv1i8_nxv1i8_i8(<vscale x 1 x i8> %0, i8 %1, i32 %2) nounwind {
 ; CHECK-LABEL: intrinsic_vasub_vx_nxv1i8_nxv1i8_i8:
 ; CHECK:       # %bb.0: # %entry
%a = call <vscale x 16 x i32> @llvm.riscv.vasub.mask.nxv16i32.i32(
     <vscale x 16 x i32> %0,
     <vscale x 16 x i32> %1,
     i32 %2,
     <vscale x 16 x i1> %3,
     i32 %4)
 
   ret <vscale x 16 x i32> %a
 }
+
+declare <vscale x 1 x i64> @llvm.riscv.vasub.nxv1i64.i64(
+  <vscale x 1 x i64>,
+  i64,
+  i32);
+
+define <vscale x 1 x i64> @intrinsic_vasub_vx_nxv1i64_nxv1i64_i64(<vscale x 1 x i64> %0, i64 %1, i32 %2) nounwind {
+; CHECK-LABEL: intrinsic_vasub_vx_nxv1i64_nxv1i64_i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a2, a2, e64,m1,ta,mu
+; CHECK-NEXT:    vmv.v.x v25, a1
+; CHECK-NEXT:    addi a1, zero, 32
+; CHECK-NEXT:    vsll.vx v25, v25, a1
+; CHECK-NEXT:    vmv.v.x v26, a0
+; CHECK-NEXT:    vsll.vx v26, v26, a1
+; CHECK-NEXT:    vsrl.vx v26, v26, a1
+; CHECK-NEXT:    vor.vv v25, v26, v25
+; CHECK-NEXT:    vasub.vv v8, v8, v25
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i64> @llvm.riscv.vasub.nxv1i64.i64(
+    <vscale x 1 x i64> %0,
+    i64 %1,
+    i32 %2)
+
+  ret <vscale x 1 x i64> %a
+}
+
+declare <vscale x 1 x i64> @llvm.riscv.vasub.mask.nxv1i64.i64(
+  <vscale x 1 x i64>,
+  <vscale x 1 x i64>,
+  i64,
+  <vscale x 1 x i1>,
+  i32);
+
+define <vscale x 1 x i64> @intrinsic_vasub_mask_vx_nxv1i64_nxv1i64_i64(<vscale x 1 x i64> %0, <vscale x 1 x i64> %1, i64 %2, <vscale x 1 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vasub_mask_vx_nxv1i64_nxv1i64_i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a3, a2, e64,m1,ta,mu
+; CHECK-NEXT:    vmv.v.x v25, a1
+; CHECK-NEXT:    addi a1, zero, 32
+; CHECK-NEXT:    vsll.vx v25, v25, a1
+; CHECK-NEXT:    vmv.v.x v26, a0
+; CHECK-NEXT:    vsll.vx v26, v26, a1
+; CHECK-NEXT:    vsrl.vx v26, v26, a1
+; CHECK-NEXT:    vor.vv v25, v26, v25
+; CHECK-NEXT:    vsetvli a0, a2, e64,m1,tu,mu
+; CHECK-NEXT:    vasub.vv v8, v9, v25, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i64> @llvm.riscv.vasub.mask.nxv1i64.i64(
+    <vscale x 1 x i64> %0,
+    <vscale x 1 x i64> %1,
+    i64 %2,
+    <vscale x 1 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 1 x i64> %a
+}
+
+declare <vscale x 2 x i64> @llvm.riscv.vasub.nxv2i64.i64(
+  <vscale x 2 x i64>,
+  i64,
+  i32);
+
+define <vscale x 2 x i64> @intrinsic_vasub_vx_nxv2i64_nxv2i64_i64(<vscale x 2 x i64> %0, i64 %1, i32 %2) nounwind {
+; CHECK-LABEL: intrinsic_vasub_vx_nxv2i64_nxv2i64_i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a2, a2, e64,m2,ta,mu
+; CHECK-NEXT:    vmv.v.x v26, a1
+; CHECK-NEXT:    addi a1, zero, 32
+; CHECK-NEXT:    vsll.vx v26, v26, a1
+; CHECK-NEXT:    vmv.v.x v28, a0
+; CHECK-NEXT:    vsll.vx v28, v28, a1
+; CHECK-NEXT:    vsrl.vx v28, v28, a1
+; CHECK-NEXT:    vor.vv v26, v28, v26
+; CHECK-NEXT:    vasub.vv v8, v8, v26
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i64> @llvm.riscv.vasub.nxv2i64.i64(
+    <vscale x 2 x i64> %0,
+    i64 %1,
+    i32 %2)
+
+  ret <vscale x 2 x i64> %a
+}
+
+declare <vscale x 2 x i64> @llvm.riscv.vasub.mask.nxv2i64.i64(
+  <vscale x 2 x i64>,
+  <vscale x 2 x i64>,
+  i64,
+  <vscale x 2 x i1>,
+  i32);
+
+define <vscale x 2 x i64> @intrinsic_vasub_mask_vx_nxv2i64_nxv2i64_i64(<vscale x 2 x i64> %0, <vscale x 2 x i64> %1, i64 %2, <vscale x 2 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vasub_mask_vx_nxv2i64_nxv2i64_i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a3, a2, e64,m2,ta,mu
+; CHECK-NEXT:    vmv.v.x v26, a1
+; CHECK-NEXT:    addi a1, zero, 32
+; CHECK-NEXT:    vsll.vx v26, v26, a1
+; CHECK-NEXT:    vmv.v.x v28, a0
+; CHECK-NEXT:    vsll.vx v28, v28, a1
+; CHECK-NEXT:    vsrl.vx v28, v28, a1
+; CHECK-NEXT:    vor.vv v26, v28, v26
+; CHECK-NEXT:    vsetvli a0, a2, e64,m2,tu,mu
+; CHECK-NEXT:    vasub.vv v8, v10, v26, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i64> @llvm.riscv.vasub.mask.nxv2i64.i64(
+    <vscale x 2 x i64> %0,
+    <vscale x 2 x i64> %1,
+    i64 %2,
+    <vscale x 2 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 2 x i64> %a
+}
+
+declare <vscale x 4 x i64> @llvm.riscv.vasub.nxv4i64.i64(
+  <vscale x 4 x i64>,
+  i64,
+  i32);
+
+define <vscale x 4 x i64> @intrinsic_vasub_vx_nxv4i64_nxv4i64_i64(<vscale x 4 x i64> %0, i64 %1, i32 %2) nounwind {
+; CHECK-LABEL: intrinsic_vasub_vx_nxv4i64_nxv4i64_i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a2, a2, e64,m4,ta,mu
+; CHECK-NEXT:    vmv.v.x v28, a1
+; CHECK-NEXT:    addi a1, zero, 32
+; CHECK-NEXT:    vsll.vx v28, v28, a1
+; CHECK-NEXT:    vmv.v.x v12, a0
+; CHECK-NEXT:    vsll.vx v12, v12, a1
+; CHECK-NEXT:    vsrl.vx v12, v12, a1
+; CHECK-NEXT:    vor.vv v28, v12, v28
+; CHECK-NEXT:    vasub.vv v8, v8, v28
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i64> @llvm.riscv.vasub.nxv4i64.i64(
+    <vscale x 4 x i64> %0,
+    i64 %1,
+    i32 %2)
+
+  ret <vscale x 4 x i64> %a
+}
+
+declare <vscale x 4 x i64> @llvm.riscv.vasub.mask.nxv4i64.i64(
+  <vscale x 4 x i64>,
+  <vscale x 4 x i64>,
+  i64,
+  <vscale x 4 x i1>,
+  i32);
+
+define <vscale x 4 x i64> @intrinsic_vasub_mask_vx_nxv4i64_nxv4i64_i64(<vscale x 4 x i64> %0, <vscale x 4 x i64> %1, i64 %2, <vscale x 4 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vasub_mask_vx_nxv4i64_nxv4i64_i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a3, a2, e64,m4,ta,mu
+; CHECK-NEXT:    vmv.v.x v28, a1
+; CHECK-NEXT:    addi a1, zero, 32
+; CHECK-NEXT:    vsll.vx v28, v28, a1
+; CHECK-NEXT:    vmv.v.x v16, a0
+; CHECK-NEXT:    vsll.vx v16, v16, a1
+; CHECK-NEXT:    vsrl.vx v16, v16, a1
+; CHECK-NEXT:    vor.vv v28, v16, v28
+; CHECK-NEXT:    vsetvli a0, a2, e64,m4,tu,mu
+; CHECK-NEXT:    vasub.vv v8, v12, v28, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i64> @llvm.riscv.vasub.mask.nxv4i64.i64(
+    <vscale x 4 x i64> %0,
+    <vscale x 4 x i64> %1,
+    i64 %2,
+    <vscale x 4 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 4 x i64> %a
+}
+
+declare <vscale x 8 x i64> @llvm.riscv.vasub.nxv8i64.i64(
+  <vscale x 8 x i64>,
+  i64,
+  i32);
+
+define <vscale x 8 x i64> @intrinsic_vasub_vx_nxv8i64_nxv8i64_i64(<vscale x 8 x i64> %0, i64 %1, i32 %2) nounwind {
+; CHECK-LABEL: intrinsic_vasub_vx_nxv8i64_nxv8i64_i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a2, a2, e64,m8,ta,mu
+; CHECK-NEXT:    vmv.v.x v16, a1
+; CHECK-NEXT:    addi a1, zero, 32
+; CHECK-NEXT:    vsll.vx v16, v16, a1
+; CHECK-NEXT:    vmv.v.x v24, a0
+; CHECK-NEXT:    vsll.vx v24, v24, a1
+; CHECK-NEXT:    vsrl.vx v24, v24, a1
+; CHECK-NEXT:    vor.vv v16, v24, v16
+; CHECK-NEXT:    vasub.vv v8, v8, v16
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i64> @llvm.riscv.vasub.nxv8i64.i64(
+    <vscale x 8 x i64> %0,
+    i64 %1,
+    i32 %2)
+
+  ret <vscale x 8 x i64> %a
+}
+
+declare <vscale x 8 x i64> @llvm.riscv.vasub.mask.nxv8i64.i64(
+  <vscale x 8 x i64>,
+  <vscale x 8 x i64>,
+  i64,
+  <vscale x 8 x i1>,
+  i32);
+
+define <vscale x 8 x i64> @intrinsic_vasub_mask_vx_nxv8i64_nxv8i64_i64(<vscale x 8 x i64> %0, <vscale x 8 x i64> %1, i64 %2, <vscale x 8 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vasub_mask_vx_nxv8i64_nxv8i64_i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    csrrs a3, vlenb, zero
+; CHECK-NEXT:    sub sp, sp, a3
+; CHECK-NEXT:    vs1r.v v0, (sp) # Unknown-size Folded Spill
+; CHECK-NEXT:    vsetvli a3, a2, e64,m8,ta,mu
+; CHECK-NEXT:    vmv.v.x v24, a1
+; CHECK-NEXT:    addi a1, zero, 32
+; CHECK-NEXT:    vsll.vx v0, v24, a1
+; CHECK-NEXT:    vmv.v.x v24, a0
+; CHECK-NEXT:    vsll.vx v24, v24, a1
+; CHECK-NEXT:    vsrl.vx v24, v24, a1
+; CHECK-NEXT:    vor.vv v24, v24, v0
+; CHECK-NEXT:    vsetvli a0, a2, e64,m8,tu,mu
+; CHECK-NEXT:    vl1re8.v v0, (sp) # Unknown-size Folded Reload
+; CHECK-NEXT:    vasub.vv v8, v16, v24, v0.t
+; CHECK-NEXT:    csrrs a0, vlenb, zero
+; CHECK-NEXT:    add sp, sp, a0
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i64> @llvm.riscv.vasub.mask.nxv8i64.i64(
+    <vscale x 8 x i64> %0,
+    <vscale x 8 x i64> %1,
+    i64 %2,
+    <vscale x 8 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 8 x i64> %a
+}

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

%a = call <vscale x 16 x i32> @llvm.riscv.vasubu.mask.nxv16i32.nxv16i32(
     <vscale x 16 x i32> %1,
     <vscale x 16 x i32> %2,
     <vscale x 16 x i1> %3,
     i32 %4)
 
   ret <vscale x 16 x i32> %a
 }
 
+declare <vscale x 1 x i64> @llvm.riscv.vasubu.nxv1i64.nxv1i64(
+  <vscale x 1 x i64>,
+  <vscale x 1 x i64>,
+  i32);
+
+define <vscale x 1 x i64> @intrinsic_vasubu_vv_nxv1i64_nxv1i64_nxv1i64(<vscale x 1 x i64> %0, <vscale x 1 x i64> %1, i32 %2) nounwind {
+; CHECK-LABEL: intrinsic_vasubu_vv_nxv1i64_nxv1i64_nxv1i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e64,m1,ta,mu
+; CHECK-NEXT:    vasubu.vv v8, v8, v9
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i64> @llvm.riscv.vasubu.nxv1i64.nxv1i64(
+    <vscale x 1 x i64> %0,
+    <vscale x 1 x i64> %1,
+    i32 %2)
+
+  ret <vscale x 1 x i64> %a
+}
+
+declare <vscale x 1 x i64> @llvm.riscv.vasubu.mask.nxv1i64.nxv1i64(
+  <vscale x 1 x i64>,
+  <vscale x 1 x i64>,
+  <vscale x 1 x i64>,
+  <vscale x 1 x i1>,
+  i32);
+
+define <vscale x 1 x i64> @intrinsic_vasubu_mask_vv_nxv1i64_nxv1i64_nxv1i64(<vscale x 1 x i64> %0, <vscale x 1 x i64> %1, <vscale x 1 x i64> %2, <vscale x 1 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vasubu_mask_vv_nxv1i64_nxv1i64_nxv1i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e64,m1,tu,mu
+; CHECK-NEXT:    vasubu.vv v8, v9, v10, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i64> @llvm.riscv.vasubu.mask.nxv1i64.nxv1i64(
+    <vscale x 1 x i64> %0,
+    <vscale x 1 x i64> %1,
+    <vscale x 1 x i64> %2,
+    <vscale x 1 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 1 x i64> %a
+}
+
+declare <vscale x 2 x i64> @llvm.riscv.vasubu.nxv2i64.nxv2i64(
+  <vscale x 2 x i64>,
+  <vscale x 2 x i64>,
+  i32);
+
+define <vscale x 2 x i64> @intrinsic_vasubu_vv_nxv2i64_nxv2i64_nxv2i64(<vscale x 2 x i64> %0, <vscale x 2 x i64> %1, i32 %2) nounwind {
+; CHECK-LABEL: intrinsic_vasubu_vv_nxv2i64_nxv2i64_nxv2i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e64,m2,ta,mu
+; CHECK-NEXT:    vasubu.vv v8, v8, v10
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i64> @llvm.riscv.vasubu.nxv2i64.nxv2i64(
+    <vscale x 2 x i64> %0,
+    <vscale x 2 x i64> %1,
+    i32 %2)
+
+  ret <vscale x 2 x i64> %a
+}
+
+declare <vscale x 2 x i64> @llvm.riscv.vasubu.mask.nxv2i64.nxv2i64(
+  <vscale x 2 x i64>,
+  <vscale x 2 x i64>,
+  <vscale x 2 x i64>,
+  <vscale x 2 x i1>,
+  i32);
+
+define <vscale x 2 x i64> @intrinsic_vasubu_mask_vv_nxv2i64_nxv2i64_nxv2i64(<vscale x 2 x i64> %0, <vscale x 2 x i64> %1, <vscale x 2 x i64> %2, <vscale x 2 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vasubu_mask_vv_nxv2i64_nxv2i64_nxv2i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e64,m2,tu,mu
+; CHECK-NEXT:    vasubu.vv v8, v10, v12, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i64> @llvm.riscv.vasubu.mask.nxv2i64.nxv2i64(
+    <vscale x 2 x i64> %0,
+    <vscale x 2 x i64> %1,
+    <vscale x 2 x i64> %2,
+    <vscale x 2 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 2 x i64> %a
+}
+
+declare <vscale x 4 x i64> @llvm.riscv.vasubu.nxv4i64.nxv4i64(
+  <vscale x 4 x i64>,
+  <vscale x 4 x i64>,
+  i32);
+
+define <vscale x 4 x i64> @intrinsic_vasubu_vv_nxv4i64_nxv4i64_nxv4i64(<vscale x 4 x i64> %0, <vscale x 4 x i64> %1, i32 %2) nounwind {
+; CHECK-LABEL: intrinsic_vasubu_vv_nxv4i64_nxv4i64_nxv4i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e64,m4,ta,mu
+; CHECK-NEXT:    vasubu.vv v8, v8, v12
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i64> @llvm.riscv.vasubu.nxv4i64.nxv4i64(
+    <vscale x 4 x i64> %0,
+    <vscale x 4 x i64> %1,
+    i32 %2)
+
+  ret <vscale x 4 x i64> %a
+}
+
+declare <vscale x 4 x i64> @llvm.riscv.vasubu.mask.nxv4i64.nxv4i64(
+  <vscale x 4 x i64>,
+  <vscale x 4 x i64>,
+  <vscale x 4 x i64>,
+  <vscale x 4 x i1>,
+  i32);
+
+define <vscale x 4 x i64> @intrinsic_vasubu_mask_vv_nxv4i64_nxv4i64_nxv4i64(<vscale x 4 x i64> %0, <vscale x 4 x i64> %1, <vscale x 4 x i64> %2, <vscale x 4 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vasubu_mask_vv_nxv4i64_nxv4i64_nxv4i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e64,m4,tu,mu
+; CHECK-NEXT:    vasubu.vv v8, v12, v16, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i64> @llvm.riscv.vasubu.mask.nxv4i64.nxv4i64(
+    <vscale x 4 x i64> %0,
+    <vscale x 4 x i64> %1,
+    <vscale x 4 x i64> %2,
+    <vscale x 4 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 4 x i64> %a
+}
+
+declare <vscale x 8 x i64> @llvm.riscv.vasubu.nxv8i64.nxv8i64(
+  <vscale x 8 x i64>,
+  <vscale x 8 x i64>,
+  i32);
+
+define <vscale x 8 x i64> @intrinsic_vasubu_vv_nxv8i64_nxv8i64_nxv8i64(<vscale x 8 x i64> %0, <vscale x 8 x i64> %1, i32 %2) nounwind {
+; CHECK-LABEL: intrinsic_vasubu_vv_nxv8i64_nxv8i64_nxv8i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e64,m8,ta,mu
+; CHECK-NEXT:    vasubu.vv v8, v8, v16
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i64> @llvm.riscv.vasubu.nxv8i64.nxv8i64(
+    <vscale x 8 x i64> %0,
+    <vscale x 8 x i64> %1,
+    i32 %2)
+
+  ret <vscale x 8 x i64> %a
+}
+
+declare <vscale x 8 x i64> @llvm.riscv.vasubu.mask.nxv8i64.nxv8i64(
+  <vscale x 8 x i64>,
+  <vscale x 8 x i64>,
+  <vscale x 8 x i64>,
+  <vscale x 8 x i1>,
+  i32);
+
+define <vscale x 8 x i64> @intrinsic_vasubu_mask_vv_nxv8i64_nxv8i64_nxv8i64(<vscale x 8 x i64> %0, <vscale x 8 x i64> %1, <vscale x 8 x i64> %2, <vscale x 8 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vasubu_mask_vv_nxv8i64_nxv8i64_nxv8i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vl8re64.v v24, (a0)
+; CHECK-NEXT:    vsetvli a0, a1, e64,m8,tu,mu
+; CHECK-NEXT:    vasubu.vv v8, v16, v24, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i64> @llvm.riscv.vasubu.mask.nxv8i64.nxv8i64(
+    <vscale x 8 x i64> %0,
+    <vscale x 8 x i64> %1,
+    <vscale x 8 x i64> %2,
+    <vscale x 8 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 8 x i64> %a
+}
+
 declare <vscale x 1 x i8> @llvm.riscv.vasubu.nxv1i8.i8(
   <vscale x 1 x i8>,
   i8,
   i32);
 
 define <vscale x 1 x i8> @intrinsic_vasubu_vx_nxv1i8_nxv1i8_i8(<vscale x 1 x i8> %0, i8 %1, i32 %2) nounwind {
 ; CHECK-LABEL: intrinsic_vasubu_vx_nxv1i8_nxv1i8_i8:
 ; CHECK:       # %bb.0: # %entry
%a = call <vscale x 16 x i32> @llvm.riscv.vasubu.mask.nxv16i32.i32(
     <vscale x 16 x i32> %0,
     <vscale x 16 x i32> %1,
     i32 %2,
     <vscale x 16 x i1> %3,
     i32 %4)
 
   ret <vscale x 16 x i32> %a
 }
+
+declare <vscale x 1 x i64> @llvm.riscv.vasubu.nxv1i64.i64(
+  <vscale x 1 x i64>,
+  i64,
+  i32);
+
+define <vscale x 1 x i64> @intrinsic_vasubu_vx_nxv1i64_nxv1i64_i64(<vscale x 1 x i64> %0, i64 %1, i32 %2) nounwind {
+; CHECK-LABEL: intrinsic_vasubu_vx_nxv1i64_nxv1i64_i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a2, a2, e64,m1,ta,mu
+; CHECK-NEXT:    vmv.v.x v25, a1
+; CHECK-NEXT:    addi a1, zero, 32
+; CHECK-NEXT:    vsll.vx v25, v25, a1
+; CHECK-NEXT:    vmv.v.x v26, a0
+; CHECK-NEXT:    vsll.vx v26, v26, a1
+; CHECK-NEXT:    vsrl.vx v26, v26, a1
+; CHECK-NEXT:    vor.vv v25, v26, v25
+; CHECK-NEXT:    vasubu.vv v8, v8, v25
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i64> @llvm.riscv.vasubu.nxv1i64.i64(
+    <vscale x 1 x i64> %0,
+    i64 %1,
+    i32 %2)
+
+  ret <vscale x 1 x i64> %a
+}
+
+declare <vscale x 1 x i64> @llvm.riscv.vasubu.mask.nxv1i64.i64(
+  <vscale x 1 x i64>,
+  <vscale x 1 x i64>,
+  i64,
+  <vscale x 1 x i1>,
+  i32);
+
+define <vscale x 1 x i64> @intrinsic_vasubu_mask_vx_nxv1i64_nxv1i64_i64(<vscale x 1 x i64> %0, <vscale x 1 x i64> %1, i64 %2, <vscale x 1 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vasubu_mask_vx_nxv1i64_nxv1i64_i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a3, a2, e64,m1,ta,mu
+; CHECK-NEXT:    vmv.v.x v25, a1
+; CHECK-NEXT:    addi a1, zero, 32
+; CHECK-NEXT:    vsll.vx v25, v25, a1
+; CHECK-NEXT:    vmv.v.x v26, a0
+; CHECK-NEXT:    vsll.vx v26, v26, a1
+; CHECK-NEXT:    vsrl.vx v26, v26, a1
+; CHECK-NEXT:    vor.vv v25, v26, v25
+; CHECK-NEXT:    vsetvli a0, a2, e64,m1,tu,mu
+; CHECK-NEXT:    vasubu.vv v8, v9, v25, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i64> @llvm.riscv.vasubu.mask.nxv1i64.i64(
+    <vscale x 1 x i64> %0,
+    <vscale x 1 x i64> %1,
+    i64 %2,
+    <vscale x 1 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 1 x i64> %a
+}
+
+declare <vscale x 2 x i64> @llvm.riscv.vasubu.nxv2i64.i64(
+  <vscale x 2 x i64>,
+  i64,
+  i32);
+
+define <vscale x 2 x i64> @intrinsic_vasubu_vx_nxv2i64_nxv2i64_i64(<vscale x 2 x i64> %0, i64 %1, i32 %2) nounwind {
+; CHECK-LABEL: intrinsic_vasubu_vx_nxv2i64_nxv2i64_i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a2, a2, e64,m2,ta,mu
+; CHECK-NEXT:    vmv.v.x v26, a1
+; CHECK-NEXT:    addi a1, zero, 32
+; CHECK-NEXT:    vsll.vx v26, v26, a1
+; CHECK-NEXT:    vmv.v.x v28, a0
+; CHECK-NEXT:    vsll.vx v28, v28, a1
+; CHECK-NEXT:    vsrl.vx v28, v28, a1
+; CHECK-NEXT:    vor.vv v26, v28, v26
+; CHECK-NEXT:    vasubu.vv v8, v8, v26
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i64> @llvm.riscv.vasubu.nxv2i64.i64(
+    <vscale x 2 x i64> %0,
+    i64 %1,
+    i32 %2)
+
+  ret <vscale x 2 x i64> %a
+}
+
+declare <vscale x 2 x i64> @llvm.riscv.vasubu.mask.nxv2i64.i64(
+  <vscale x 2 x i64>,
+  <vscale x 2 x i64>,
+  i64,
+  <vscale x 2 x i1>,
+  i32);
+
+define <vscale x 2 x i64> @intrinsic_vasubu_mask_vx_nxv2i64_nxv2i64_i64(<vscale x 2 x i64> %0, <vscale x 2 x i64> %1, i64 %2, <vscale x 2 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vasubu_mask_vx_nxv2i64_nxv2i64_i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a3, a2, e64,m2,ta,mu
+; CHECK-NEXT:    vmv.v.x v26, a1
+; CHECK-NEXT:    addi a1, zero, 32
+; CHECK-NEXT:    vsll.vx v26, v26, a1
+; CHECK-NEXT:    vmv.v.x v28, a0
+; CHECK-NEXT:    vsll.vx v28, v28, a1
+; CHECK-NEXT:    vsrl.vx v28, v28, a1
+; CHECK-NEXT:    vor.vv v26, v28, v26
+; CHECK-NEXT:    vsetvli a0, a2, e64,m2,tu,mu
+; CHECK-NEXT:    vasubu.vv v8, v10, v26, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i64> @llvm.riscv.vasubu.mask.nxv2i64.i64(
+    <vscale x 2 x i64> %0,
+    <vscale x 2 x i64> %1,
+    i64 %2,
+    <vscale x 2 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 2 x i64> %a
+}
+
+declare <vscale x 4 x i64> @llvm.riscv.vasubu.nxv4i64.i64(
+  <vscale x 4 x i64>,
+  i64,
+  i32);
+
+define <vscale x 4 x i64> @intrinsic_vasubu_vx_nxv4i64_nxv4i64_i64(<vscale x 4 x i64> %0, i64 %1, i32 %2) nounwind {
+; CHECK-LABEL: intrinsic_vasubu_vx_nxv4i64_nxv4i64_i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a2, a2, e64,m4,ta,mu
+; CHECK-NEXT:    vmv.v.x v28, a1
+; CHECK-NEXT:    addi a1, zero, 32
+; CHECK-NEXT:    vsll.vx v28, v28, a1
+; CHECK-NEXT:    vmv.v.x v12, a0
+; CHECK-NEXT:    vsll.vx v12, v12, a1
+; CHECK-NEXT:    vsrl.vx v12, v12, a1
+; CHECK-NEXT:    vor.vv v28, v12, v28
+; CHECK-NEXT:    vasubu.vv v8, v8, v28
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i64> @llvm.riscv.vasubu.nxv4i64.i64(
+    <vscale x 4 x i64> %0,
+    i64 %1,
+    i32 %2)
+
+  ret <vscale x 4 x i64> %a
+}
+
+declare <vscale x 4 x i64> @llvm.riscv.vasubu.mask.nxv4i64.i64(
+  <vscale x 4 x i64>,
+  <vscale x 4 x i64>,
+  i64,
+  <vscale x 4 x i1>,
+  i32);
+
+define <vscale x 4 x i64> @intrinsic_vasubu_mask_vx_nxv4i64_nxv4i64_i64(<vscale x 4 x i64> %0, <vscale x 4 x i64> %1, i64 %2, <vscale x 4 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vasubu_mask_vx_nxv4i64_nxv4i64_i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a3, a2, e64,m4,ta,mu
+; CHECK-NEXT:    vmv.v.x v28, a1
+; CHECK-NEXT:    addi a1, zero, 32
+; CHECK-NEXT:    vsll.vx v28, v28, a1
+; CHECK-NEXT:    vmv.v.x v16, a0
+; CHECK-NEXT:    vsll.vx v16, v16, a1
+; CHECK-NEXT:    vsrl.vx v16, v16, a1
+; CHECK-NEXT:    vor.vv v28, v16, v28
+; CHECK-NEXT:    vsetvli a0, a2, e64,m4,tu,mu
+; CHECK-NEXT:    vasubu.vv v8, v12, v28, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i64> @llvm.riscv.vasubu.mask.nxv4i64.i64(
+    <vscale x 4 x i64> %0,
+    <vscale x 4 x i64> %1,
+    i64 %2,
+    <vscale x 4 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 4 x i64> %a
+}
+
+declare <vscale x 8 x i64> @llvm.riscv.vasubu.nxv8i64.i64(
+  <vscale x 8 x i64>,
+  i64,
+  i32);
+
+define <vscale x 8 x i64> @intrinsic_vasubu_vx_nxv8i64_nxv8i64_i64(<vscale x 8 x i64> %0, i64 %1, i32 %2) nounwind {
+; CHECK-LABEL: intrinsic_vasubu_vx_nxv8i64_nxv8i64_i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a2, a2, e64,m8,ta,mu
+; CHECK-NEXT:    vmv.v.x v16, a1
+; CHECK-NEXT:    addi a1, zero, 32
+; CHECK-NEXT:    vsll.vx v16, v16, a1
+; CHECK-NEXT:    vmv.v.x v24, a0
+; CHECK-NEXT:    vsll.vx v24, v24, a1
+; CHECK-NEXT:    vsrl.vx v24, v24, a1
+; CHECK-NEXT:    vor.vv v16, v24, v16
+; CHECK-NEXT:    vasubu.vv v8, v8, v16
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i64> @llvm.riscv.vasubu.nxv8i64.i64(
+    <vscale x 8 x i64> %0,
+    i64 %1,
+    i32 %2)
+
+  ret <vscale x 8 x i64> %a
+}
+
+declare <vscale x 8 x i64> @llvm.riscv.vasubu.mask.nxv8i64.i64(
+  <vscale x 8 x i64>,
+  <vscale x 8 x i64>,
+  i64,
+  <vscale x 8 x i1>,
+  i32);
+
+define <vscale x 8 x i64> @intrinsic_vasubu_mask_vx_nxv8i64_nxv8i64_i64(<vscale x 8 x i64> %0, <vscale x 8 x i64> %1, i64 %2, <vscale x 8 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vasubu_mask_vx_nxv8i64_nxv8i64_i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    csrrs a3, vlenb, zero
+; CHECK-NEXT:    sub sp, sp, a3
+; CHECK-NEXT:    vs1r.v v0, (sp) # Unknown-size Folded Spill
+; CHECK-NEXT:    vsetvli a3, a2, e64,m8,ta,mu
+; CHECK-NEXT:    vmv.v.x v24, a1
+; CHECK-NEXT:    addi a1, zero, 32
+; CHECK-NEXT:    vsll.vx v0, v24, a1
+; CHECK-NEXT:    vmv.v.x v24, a0
+; CHECK-NEXT:    vsll.vx v24, v24, a1
+; CHECK-NEXT:    vsrl.vx v24, v24, a1
+; CHECK-NEXT:    vor.vv v24, v24, v0
+; CHECK-NEXT:    vsetvli a0, a2, e64,m8,tu,mu
+; CHECK-NEXT:    vl1re8.v v0, (sp) # Unknown-size Folded Reload
+; CHECK-NEXT:    vasubu.vv v8, v16, v24, v0.t
+; CHECK-NEXT:    csrrs a0, vlenb, zero
+; CHECK-NEXT:    add sp, sp, a0
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i64> @llvm.riscv.vasubu.mask.nxv8i64.i64(
+    <vscale x 8 x i64> %0,
+    <vscale x 8 x i64> %1,
+    i64 %2,
+    <vscale x 8 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 8 x i64> %a
+}

llvm/test/CodeGen/RISCV/rvv/vmv.s.x-rv32.ll

 ; CHECK:       # %bb.0: # %entry
 ; CHECK-NEXT:    vsetvli a1, a1, e32,m8,ta,mu
 ; CHECK-NEXT:    vmv.s.x v8, a0
 ; CHECK-NEXT:    ret
 entry:
   %a = call <vscale x 16 x i32> @llvm.riscv.vmv.s.x.nxv16i32(<vscale x 16 x i32> %0, i32 %1, i32 %2)
   ret <vscale x 16 x i32> %a
 }
+
+declare <vscale x 1 x i64> @llvm.riscv.vmv.s.x.nxv1i64(<vscale x 1 x i64>, i64, i32);
+
+define <vscale x 1 x i64> @intrinsic_vmv.s.x_x_nxv1i64(<vscale x 1 x i64> %0, i64 %1, i32 %2) nounwind {
+; CHECK-LABEL: intrinsic_vmv.s.x_x_nxv1i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a2, a2, e64,m1,ta,mu
+; CHECK-NEXT:    vmv.v.x v25, a1
+; CHECK-NEXT:    addi a1, zero, 32
+; CHECK-NEXT:    vsll.vx v25, v25, a1
+; CHECK-NEXT:    vmv.v.x v26, a0
+; CHECK-NEXT:    vsll.vx v26, v26, a1
+; CHECK-NEXT:    vsrl.vx v26, v26, a1
+; CHECK-NEXT:    vor.vv v25, v26, v25
+; CHECK-NEXT:    vid.v v26
+; CHECK-NEXT:    vmseq.vi v0, v26, 0
+; CHECK-NEXT:    vmerge.vvm v8, v8, v25, v0
+; CHECK-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i64> @llvm.riscv.vmv.s.x.nxv1i64(<vscale x 1 x i64> %0, i64 %1, i32 %2)
+  ret <vscale x 1 x i64> %a
+}
+
+declare <vscale x 2 x i64> @llvm.riscv.vmv.s.x.nxv2i64(<vscale x 2 x i64>, i64, i32);
+
+define <vscale x 2 x i64> @intrinsic_vmv.s.x_x_nxv2i64(<vscale x 2 x i64> %0, i64 %1, i32 %2) nounwind {
+; CHECK-LABEL: intrinsic_vmv.s.x_x_nxv2i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a2, a2, e64,m2,ta,mu
+; CHECK-NEXT:    vmv.v.x v26, a1
+; CHECK-NEXT:    addi a1, zero, 32
+; CHECK-NEXT:    vsll.vx v26, v26, a1
+; CHECK-NEXT:    vmv.v.x v28, a0
+; CHECK-NEXT:    vsll.vx v28, v28, a1
+; CHECK-NEXT:    vsrl.vx v28, v28, a1
+; CHECK-NEXT:    vor.vv v26, v28, v26
+; CHECK-NEXT:    vid.v v28
+; CHECK-NEXT:    vmseq.vi v0, v28, 0
+; CHECK-NEXT:    vmerge.vvm v8, v8, v26, v0
+; CHECK-NEXT:    ret
+entry:
+  %a = call <vscale x 2 x i64> @llvm.riscv.vmv.s.x.nxv2i64(<vscale x 2 x i64> %0, i64 %1, i32 %2)
+  ret <vscale x 2 x i64> %a
+}
+
+declare <vscale x 4 x i64> @llvm.riscv.vmv.s.x.nxv4i64(<vscale x 4 x i64>, i64, i32);
+
+define <vscale x 4 x i64> @intrinsic_vmv.s.x_x_nxv4i64(<vscale x 4 x i64> %0, i64 %1, i32 %2) nounwind {
+; CHECK-LABEL: intrinsic_vmv.s.x_x_nxv4i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a2, a2, e64,m4,ta,mu
+; CHECK-NEXT:    vmv.v.x v28, a1
+; CHECK-NEXT:    addi a1, zero, 32
+; CHECK-NEXT:    vsll.vx v28, v28, a1
+; CHECK-NEXT:    vmv.v.x v12, a0
+; CHECK-NEXT:    vsll.vx v12, v12, a1
+; CHECK-NEXT:    vsrl.vx v12, v12, a1
+; CHECK-NEXT:    vor.vv v28, v12, v28
+; CHECK-NEXT:    vid.v v12
+; CHECK-NEXT:    vmseq.vi v0, v12, 0
+; CHECK-NEXT:    vmerge.vvm v8, v8, v28, v0
+; CHECK-NEXT:    ret
+entry:
+  %a = call <vscale x 4 x i64> @llvm.riscv.vmv.s.x.nxv4i64(<vscale x 4 x i64> %0, i64 %1, i32 %2)
+  ret <vscale x 4 x i64> %a
+}
+
+declare <vscale x 8 x i64> @llvm.riscv.vmv.s.x.nxv8i64(<vscale x 8 x i64>, i64, i32);
+
+define <vscale x 8 x i64> @intrinsic_vmv.s.x_x_nxv8i64(<vscale x 8 x i64> %0, i64 %1, i32 %2) nounwind {
+; CHECK-LABEL: intrinsic_vmv.s.x_x_nxv8i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a2, a2, e64,m8,ta,mu
+; CHECK-NEXT:    vmv.v.x v16, a1
+; CHECK-NEXT:    addi a1, zero, 32
+; CHECK-NEXT:    vsll.vx v16, v16, a1
+; CHECK-NEXT:    vmv.v.x v24, a0
+; CHECK-NEXT:    vsll.vx v24, v24, a1
+; CHECK-NEXT:    vsrl.vx v24, v24, a1
+; CHECK-NEXT:    vor.vv v16, v24, v16
+; CHECK-NEXT:    vid.v v24
+; CHECK-NEXT:    vmseq.vi v0, v24, 0
+; CHECK-NEXT:    vmerge.vvm v8, v8, v16, v0
+; CHECK-NEXT:    ret
+entry:
+  %a = call <vscale x 8 x i64> @llvm.riscv.vmv.s.x.nxv8i64(<vscale x 8 x i64> %0, i64 %1, i32 %2)
+  ret <vscale x 8 x i64> %a
+}

llvm/test/CodeGen/RISCV/rvv/vmv.v.x-rv32.ll

 entry:
   %a = call <vscale x 16 x i32> @llvm.riscv.vmv.v.x.nxv16i32(
     i32 %0,
     i32 %1)
 
   ret <vscale x 16 x i32> %a
 }
 
+declare <vscale x 1 x i64> @llvm.riscv.vmv.v.x.nxv1i64(
+  i64,
+  i32);
+
+define <vscale x 1 x i64> @intrinsic_vmv.v.x_x_nxv1i64(i64 %0, i32 %1) nounwind {
+; CHECK-LABEL: intrinsic_vmv.v.x_x_nxv1i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a2, a2, e64,m1,ta,mu
+; CHECK-NEXT:    vmv.v.x v25, a1
+; CHECK-NEXT:    addi a1, zero, 32
+; CHECK-NEXT:    vsll.vx v25, v25, a1
+; CHECK-NEXT:    vmv.v.x v26, a0
+; CHECK-NEXT:    vsll.vx v26, v26, a1
+; CHECK-NEXT:    vsrl.vx v26, v26, a1
+; CHECK-NEXT:    vor.vv v8, v26, v25
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i64> @llvm.riscv.vmv.v.x.nxv1i64(
+    i64 %0,
+    i32 %1)
+
+  ret <vscale x 1 x i64> %a
+}
+
+declare <vscale x 2 x i64> @llvm.riscv.vmv.v.x.nxv2i64(
+  i64,
+  i32);
+
+define <vscale x 2 x i64> @intrinsic_vmv.v.x_x_nxv2i64(i64 %0, i32 %1) nounwind {
+; CHECK-LABEL: intrinsic_vmv.v.x_x_nxv2i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a2, a2, e64,m2,ta,mu
+; CHECK-NEXT:    vmv.v.x v26, a1
+; CHECK-NEXT:    addi a1, zero, 32
+; CHECK-NEXT:    vsll.vx v26, v26, a1
+; CHECK-NEXT:    vmv.v.x v28, a0
+; CHECK-NEXT:    vsll.vx v28, v28, a1
+; CHECK-NEXT:    vsrl.vx v28, v28, a1
+; CHECK-NEXT:    vor.vv v8, v28, v26
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i64> @llvm.riscv.vmv.v.x.nxv2i64(
+    i64 %0,
+    i32 %1)
+
+  ret <vscale x 2 x i64> %a
+}
+
+declare <vscale x 4 x i64> @llvm.riscv.vmv.v.x.nxv4i64(
+  i64,
+  i32);
+
+define <vscale x 4 x i64> @intrinsic_vmv.v.x_x_nxv4i64(i64 %0, i32 %1) nounwind {
+; CHECK-LABEL: intrinsic_vmv.v.x_x_nxv4i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a2, a2, e64,m4,ta,mu
+; CHECK-NEXT:    vmv.v.x v28, a1
+; CHECK-NEXT:    addi a1, zero, 32
+; CHECK-NEXT:    vsll.vx v28, v28, a1
+; CHECK-NEXT:    vmv.v.x v8, a0
+; CHECK-NEXT:    vsll.vx v8, v8, a1
+; CHECK-NEXT:    vsrl.vx v8, v8, a1
+; CHECK-NEXT:    vor.vv v8, v8, v28
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i64> @llvm.riscv.vmv.v.x.nxv4i64(
+    i64 %0,
+    i32 %1)
+
+  ret <vscale x 4 x i64> %a
+}
+
+declare <vscale x 8 x i64> @llvm.riscv.vmv.v.x.nxv8i64(
+  i64,
+  i32);
+
+define <vscale x 8 x i64> @intrinsic_vmv.v.x_x_nxv8i64(i64 %0, i32 %1) nounwind {
+; CHECK-LABEL: intrinsic_vmv.v.x_x_nxv8i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a2, a2, e64,m8,ta,mu
+; CHECK-NEXT:    vmv.v.x v8, a1
+; CHECK-NEXT:    addi a1, zero, 32
+; CHECK-NEXT:    vsll.vx v8, v8, a1
+; CHECK-NEXT:    vmv.v.x v16, a0
+; CHECK-NEXT:    vsll.vx v16, v16, a1
+; CHECK-NEXT:    vsrl.vx v16, v16, a1
+; CHECK-NEXT:    vor.vv v8, v16, v8
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i64> @llvm.riscv.vmv.v.x.nxv8i64(
+    i64 %0,
+    i32 %1)
+
+  ret <vscale x 8 x i64> %a
+}
+
 define <vscale x 1 x i8> @intrinsic_vmv.v.x_i_nxv1i8(i32 %0) nounwind {
 ; CHECK-LABEL: intrinsic_vmv.v.x_i_nxv1i8:
 ; CHECK:       # %bb.0: # %entry
 ; CHECK-NEXT:    vsetvli a0, a0, e8,mf8,ta,mu
 ; CHECK-NEXT:    vmv.v.i v8, 9
 ; CHECK-NEXT:    jalr zero, 0(ra)
 entry:
   %a = call <vscale x 1 x i8> @llvm.riscv.vmv.v.x.nxv1i8(

 ; CHECK-NEXT:    jalr zero, 0(ra)
 entry:
   %a = call <vscale x 16 x i32> @llvm.riscv.vmv.v.x.nxv16i32(
     i32 9,
     i32 %0)
 
   ret <vscale x 16 x i32> %a
 }
+
+define <vscale x 1 x i64> @intrinsic_vmv.v.x_i_nxv1i64(i32 %0) nounwind {
+; CHECK-LABEL: intrinsic_vmv.v.x_i_nxv1i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e64,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v8, 9
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i64> @llvm.riscv.vmv.v.x.nxv1i64(
+    i64 9,
+    i32 %0)
+
+  ret <vscale x 1 x i64> %a
+}
+
+define <vscale x 2 x i64> @intrinsic_vmv.v.x_i_nxv2i64(i32 %0) nounwind {
+; CHECK-LABEL: intrinsic_vmv.v.x_i_nxv2i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e64,m2,ta,mu
+; CHECK-NEXT:    vmv.v.i v8, 9
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i64> @llvm.riscv.vmv.v.x.nxv2i64(
+    i64 9,
+    i32 %0)
+
+  ret <vscale x 2 x i64> %a
+}
+
+define <vscale x 4 x i64> @intrinsic_vmv.v.x_i_nxv4i64(i32 %0) nounwind {
+; CHECK-LABEL: intrinsic_vmv.v.x_i_nxv4i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e64,m4,ta,mu
+; CHECK-NEXT:    vmv.v.i v8, 9
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i64> @llvm.riscv.vmv.v.x.nxv4i64(
+    i64 9,
+    i32 %0)
+
+  ret <vscale x 4 x i64> %a
+}
+
+define <vscale x 8 x i64> @intrinsic_vmv.v.x_i_nxv8i64(i32 %0) nounwind {
+; CHECK-LABEL: intrinsic_vmv.v.x_i_nxv8i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e64,m8,ta,mu
+; CHECK-NEXT:    vmv.v.i v8, 9
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i64> @llvm.riscv.vmv.v.x.nxv8i64(
+    i64 9,
+    i32 %0)
+
+  ret <vscale x 8 x i64> %a
+}

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

%a = call <vscale x 16 x i16> @llvm.riscv.vnsra.mask.nxv16i16.nxv16i32.nxv16i16(
     <vscale x 16 x i32> %1,
     <vscale x 16 x i16> %2,
     <vscale x 16 x i1> %3,
     i32 %4)
 
   ret <vscale x 16 x i16> %a
 }
 
+declare <vscale x 1 x i32> @llvm.riscv.vnsra.nxv1i32.nxv1i64.nxv1i32(
+  <vscale x 1 x i64>,
+  <vscale x 1 x i32>,
+  i32);
+
+define <vscale x 1 x i32> @intrinsic_vnsra_wv_nxv1i32_nxv1i64_nxv1i32(<vscale x 1 x i64> %0, <vscale x 1 x i32> %1, i32 %2) nounwind {
+; CHECK-LABEL: intrinsic_vnsra_wv_nxv1i32_nxv1i64_nxv1i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e32,mf2,ta,mu
+; CHECK-NEXT:    vnsra.wv v25, v8, v9
+; CHECK-NEXT:    vmv1r.v v8, v25
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i32> @llvm.riscv.vnsra.nxv1i32.nxv1i64.nxv1i32(
+    <vscale x 1 x i64> %0,
+    <vscale x 1 x i32> %1,
+    i32 %2)
+
+  ret <vscale x 1 x i32> %a
+}
+
+declare <vscale x 1 x i32> @llvm.riscv.vnsra.mask.nxv1i32.nxv1i64.nxv1i32(
+  <vscale x 1 x i32>,
+  <vscale x 1 x i64>,
+  <vscale x 1 x i32>,
+  <vscale x 1 x i1>,
+  i32);
+
+define <vscale x 1 x i32> @intrinsic_vnsra_mask_wv_nxv1i32_nxv1i64_nxv1i32(<vscale x 1 x i32> %0, <vscale x 1 x i64> %1, <vscale x 1 x i32> %2, <vscale x 1 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vnsra_mask_wv_nxv1i32_nxv1i64_nxv1i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e32,mf2,tu,mu
+; CHECK-NEXT:    vnsra.wv v8, v9, v10, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i32> @llvm.riscv.vnsra.mask.nxv1i32.nxv1i64.nxv1i32(
+    <vscale x 1 x i32> %0,
+    <vscale x 1 x i64> %1,
+    <vscale x 1 x i32> %2,
+    <vscale x 1 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 1 x i32> %a
+}
+
+declare <vscale x 2 x i32> @llvm.riscv.vnsra.nxv2i32.nxv2i64.nxv2i32(
+  <vscale x 2 x i64>,
+  <vscale x 2 x i32>,
+  i32);
+
+define <vscale x 2 x i32> @intrinsic_vnsra_wv_nxv2i32_nxv2i64_nxv2i32(<vscale x 2 x i64> %0, <vscale x 2 x i32> %1, i32 %2) nounwind {
+; CHECK-LABEL: intrinsic_vnsra_wv_nxv2i32_nxv2i64_nxv2i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e32,m1,ta,mu
+; CHECK-NEXT:    vnsra.wv v25, v8, v10
+; CHECK-NEXT:    vmv1r.v v8, v25
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i32> @llvm.riscv.vnsra.nxv2i32.nxv2i64.nxv2i32(
+    <vscale x 2 x i64> %0,
+    <vscale x 2 x i32> %1,
+    i32 %2)
+
+  ret <vscale x 2 x i32> %a
+}
+
+declare <vscale x 2 x i32> @llvm.riscv.vnsra.mask.nxv2i32.nxv2i64.nxv2i32(
+  <vscale x 2 x i32>,
+  <vscale x 2 x i64>,
+  <vscale x 2 x i32>,
+  <vscale x 2 x i1>,
+  i32);
+
+define <vscale x 2 x i32> @intrinsic_vnsra_mask_wv_nxv2i32_nxv2i64_nxv2i32(<vscale x 2 x i32> %0, <vscale x 2 x i64> %1, <vscale x 2 x i32> %2, <vscale x 2 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vnsra_mask_wv_nxv2i32_nxv2i64_nxv2i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e32,m1,tu,mu
+; CHECK-NEXT:    vnsra.wv v8, v10, v9, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i32> @llvm.riscv.vnsra.mask.nxv2i32.nxv2i64.nxv2i32(
+    <vscale x 2 x i32> %0,
+    <vscale x 2 x i64> %1,
+    <vscale x 2 x i32> %2,
+    <vscale x 2 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 2 x i32> %a
+}
+
+declare <vscale x 4 x i32> @llvm.riscv.vnsra.nxv4i32.nxv4i64.nxv4i32(
+  <vscale x 4 x i64>,
+  <vscale x 4 x i32>,
+  i32);
+
+define <vscale x 4 x i32> @intrinsic_vnsra_wv_nxv4i32_nxv4i64_nxv4i32(<vscale x 4 x i64> %0, <vscale x 4 x i32> %1, i32 %2) nounwind {
+; CHECK-LABEL: intrinsic_vnsra_wv_nxv4i32_nxv4i64_nxv4i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e32,m2,ta,mu
+; CHECK-NEXT:    vnsra.wv v26, v8, v12
+; CHECK-NEXT:    vmv2r.v v8, v26
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i32> @llvm.riscv.vnsra.nxv4i32.nxv4i64.nxv4i32(
+    <vscale x 4 x i64> %0,
+    <vscale x 4 x i32> %1,
+    i32 %2)
+
+  ret <vscale x 4 x i32> %a
+}
+
+declare <vscale x 4 x i32> @llvm.riscv.vnsra.mask.nxv4i32.nxv4i64.nxv4i32(
+  <vscale x 4 x i32>,
+  <vscale x 4 x i64>,
+  <vscale x 4 x i32>,
+  <vscale x 4 x i1>,
+  i32);
+
+define <vscale x 4 x i32> @intrinsic_vnsra_mask_wv_nxv4i32_nxv4i64_nxv4i32(<vscale x 4 x i32> %0, <vscale x 4 x i64> %1, <vscale x 4 x i32> %2, <vscale x 4 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vnsra_mask_wv_nxv4i32_nxv4i64_nxv4i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e32,m2,tu,mu
+; CHECK-NEXT:    vnsra.wv v8, v12, v10, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i32> @llvm.riscv.vnsra.mask.nxv4i32.nxv4i64.nxv4i32(
+    <vscale x 4 x i32> %0,
+    <vscale x 4 x i64> %1,
+    <vscale x 4 x i32> %2,
+    <vscale x 4 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 4 x i32> %a
+}
+
+declare <vscale x 8 x i32> @llvm.riscv.vnsra.nxv8i32.nxv8i64.nxv8i32(
+  <vscale x 8 x i64>,
+  <vscale x 8 x i32>,
+  i32);
+
+define <vscale x 8 x i32> @intrinsic_vnsra_wv_nxv8i32_nxv8i64_nxv8i32(<vscale x 8 x i64> %0, <vscale x 8 x i32> %1, i32 %2) nounwind {
+; CHECK-LABEL: intrinsic_vnsra_wv_nxv8i32_nxv8i64_nxv8i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e32,m4,ta,mu
+; CHECK-NEXT:    vnsra.wv v28, v8, v16
+; CHECK-NEXT:    vmv4r.v v8, v28
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i32> @llvm.riscv.vnsra.nxv8i32.nxv8i64.nxv8i32(
+    <vscale x 8 x i64> %0,
+    <vscale x 8 x i32> %1,
+    i32 %2)
+
+  ret <vscale x 8 x i32> %a
+}
+
+declare <vscale x 8 x i32> @llvm.riscv.vnsra.mask.nxv8i32.nxv8i64.nxv8i32(
+  <vscale x 8 x i32>,
+  <vscale x 8 x i64>,
+  <vscale x 8 x i32>,
+  <vscale x 8 x i1>,
+  i32);
+
+define <vscale x 8 x i32> @intrinsic_vnsra_mask_wv_nxv8i32_nxv8i64_nxv8i32(<vscale x 8 x i32> %0, <vscale x 8 x i64> %1, <vscale x 8 x i32> %2, <vscale x 8 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vnsra_mask_wv_nxv8i32_nxv8i64_nxv8i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e32,m4,tu,mu
+; CHECK-NEXT:    vnsra.wv v8, v16, v12, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i32> @llvm.riscv.vnsra.mask.nxv8i32.nxv8i64.nxv8i32(
+    <vscale x 8 x i32> %0,
+    <vscale x 8 x i64> %1,
+    <vscale x 8 x i32> %2,
+    <vscale x 8 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 8 x i32> %a
+}
+
 declare <vscale x 1 x i8> @llvm.riscv.vnsra.nxv1i8.nxv1i16.i8(
   <vscale x 1 x i16>,
   i8,
   i32);
 
 define <vscale x 1 x i8> @intrinsic_vnsra_wx_nxv1i8_nxv1i16_i8(<vscale x 1 x i16> %0, i8 %1, i32 %2) nounwind {
 ; CHECK-LABEL: intrinsic_vnsra_wx_nxv1i8_nxv1i16_i8:
 ; CHECK:       # %bb.0: # %entry
%a = call <vscale x 16 x i16> @llvm.riscv.vnsra.mask.nxv16i16.nxv16i32.i16(
     <vscale x 16 x i32> %1,
     i16 %2,
     <vscale x 16 x i1> %3,
     i32 %4)
 
   ret <vscale x 16 x i16> %a
 }
 
+declare <vscale x 1 x i32> @llvm.riscv.vnsra.nxv1i32.nxv1i64.i32(
+  <vscale x 1 x i64>,
+  i32,
+  i32);
+
+define <vscale x 1 x i32> @intrinsic_vnsra_wx_nxv1i32_nxv1i64_i32(<vscale x 1 x i64> %0, i32 %1, i32 %2) nounwind {
+; CHECK-LABEL: intrinsic_vnsra_wx_nxv1i32_nxv1i64_i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e32,mf2,ta,mu
+; CHECK-NEXT:    vnsra.wx v25, v8, a0
+; CHECK-NEXT:    vmv1r.v v8, v25
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i32> @llvm.riscv.vnsra.nxv1i32.nxv1i64.i32(
+    <vscale x 1 x i64> %0,
+    i32 %1,
+    i32 %2)
+
+  ret <vscale x 1 x i32> %a
+}
+
+declare <vscale x 1 x i32> @llvm.riscv.vnsra.mask.nxv1i32.nxv1i64.i32(
+  <vscale x 1 x i32>,
+  <vscale x 1 x i64>,
+  i32,
+  <vscale x 1 x i1>,
+  i32);
+
+define <vscale x 1 x i32> @intrinsic_vnsra_mask_wx_nxv1i32_nxv1i64_i32(<vscale x 1 x i32> %0, <vscale x 1 x i64> %1, i32 %2, <vscale x 1 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vnsra_mask_wx_nxv1i32_nxv1i64_i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e32,mf2,tu,mu
+; CHECK-NEXT:    vnsra.wx v8, v9, a0, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i32> @llvm.riscv.vnsra.mask.nxv1i32.nxv1i64.i32(
+    <vscale x 1 x i32> %0,
+    <vscale x 1 x i64> %1,
+    i32 %2,
+    <vscale x 1 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 1 x i32> %a
+}
+
+declare <vscale x 2 x i32> @llvm.riscv.vnsra.nxv2i32.nxv2i64.i32(
+  <vscale x 2 x i64>,
+  i32,
+  i32);
+
+define <vscale x 2 x i32> @intrinsic_vnsra_wx_nxv2i32_nxv2i64_i32(<vscale x 2 x i64> %0, i32 %1, i32 %2) nounwind {
+; CHECK-LABEL: intrinsic_vnsra_wx_nxv2i32_nxv2i64_i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e32,m1,ta,mu
+; CHECK-NEXT:    vnsra.wx v25, v8, a0
+; CHECK-NEXT:    vmv1r.v v8, v25
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i32> @llvm.riscv.vnsra.nxv2i32.nxv2i64.i32(
+    <vscale x 2 x i64> %0,
+    i32 %1,
+    i32 %2)
+
+  ret <vscale x 2 x i32> %a
+}
+
+declare <vscale x 2 x i32> @llvm.riscv.vnsra.mask.nxv2i32.nxv2i64.i32(
+  <vscale x 2 x i32>,
+  <vscale x 2 x i64>,
+  i32,
+  <vscale x 2 x i1>,
+  i32);
+
+define <vscale x 2 x i32> @intrinsic_vnsra_mask_wx_nxv2i32_nxv2i64_i32(<vscale x 2 x i32> %0, <vscale x 2 x i64> %1, i32 %2, <vscale x 2 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vnsra_mask_wx_nxv2i32_nxv2i64_i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e32,m1,tu,mu
+; CHECK-NEXT:    vnsra.wx v8, v10, a0, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i32> @llvm.riscv.vnsra.mask.nxv2i32.nxv2i64.i32(
+    <vscale x 2 x i32> %0,
+    <vscale x 2 x i64> %1,
+    i32 %2,
+    <vscale x 2 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 2 x i32> %a
+}
+
+declare <vscale x 4 x i32> @llvm.riscv.vnsra.nxv4i32.nxv4i64.i32(
+  <vscale x 4 x i64>,
+  i32,
+  i32);
+
+define <vscale x 4 x i32> @intrinsic_vnsra_wx_nxv4i32_nxv4i64_i32(<vscale x 4 x i64> %0, i32 %1, i32 %2) nounwind {
+; CHECK-LABEL: intrinsic_vnsra_wx_nxv4i32_nxv4i64_i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e32,m2,ta,mu
+; CHECK-NEXT:    vnsra.wx v26, v8, a0
+; CHECK-NEXT:    vmv2r.v v8, v26
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i32> @llvm.riscv.vnsra.nxv4i32.nxv4i64.i32(
+    <vscale x 4 x i64> %0,
+    i32 %1,
+    i32 %2)
+
+  ret <vscale x 4 x i32> %a
+}
+
+declare <vscale x 4 x i32> @llvm.riscv.vnsra.mask.nxv4i32.nxv4i64.i32(
+  <vscale x 4 x i32>,
+  <vscale x 4 x i64>,
+  i32,
+  <vscale x 4 x i1>,
+  i32);
+
+define <vscale x 4 x i32> @intrinsic_vnsra_mask_wx_nxv4i32_nxv4i64_i32(<vscale x 4 x i32> %0, <vscale x 4 x i64> %1, i32 %2, <vscale x 4 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vnsra_mask_wx_nxv4i32_nxv4i64_i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e32,m2,tu,mu
+; CHECK-NEXT:    vnsra.wx v8, v12, a0, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i32> @llvm.riscv.vnsra.mask.nxv4i32.nxv4i64.i32(
+    <vscale x 4 x i32> %0,
+    <vscale x 4 x i64> %1,
+    i32 %2,
+    <vscale x 4 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 4 x i32> %a
+}
+
+declare <vscale x 8 x i32> @llvm.riscv.vnsra.nxv8i32.nxv8i64.i32(
+  <vscale x 8 x i64>,
+  i32,
+  i32);
+
+define <vscale x 8 x i32> @intrinsic_vnsra_wx_nxv8i32_nxv8i64_i32(<vscale x 8 x i64> %0, i32 %1, i32 %2) nounwind {
+; CHECK-LABEL: intrinsic_vnsra_wx_nxv8i32_nxv8i64_i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e32,m4,ta,mu
+; CHECK-NEXT:    vnsra.wx v28, v8, a0
+; CHECK-NEXT:    vmv4r.v v8, v28
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i32> @llvm.riscv.vnsra.nxv8i32.nxv8i64.i32(
+    <vscale x 8 x i64> %0,
+    i32 %1,
+    i32 %2)
+
+  ret <vscale x 8 x i32> %a
+}
+
+declare <vscale x 8 x i32> @llvm.riscv.vnsra.mask.nxv8i32.nxv8i64.i32(
+  <vscale x 8 x i32>,
+  <vscale x 8 x i64>,
+  i32,
+  <vscale x 8 x i1>,
+  i32);
+
+define <vscale x 8 x i32> @intrinsic_vnsra_mask_wx_nxv8i32_nxv8i64_i32(<vscale x 8 x i32> %0, <vscale x 8 x i64> %1, i32 %2, <vscale x 8 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vnsra_mask_wx_nxv8i32_nxv8i64_i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e32,m4,tu,mu
+; CHECK-NEXT:    vnsra.wx v8, v16, a0, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i32> @llvm.riscv.vnsra.mask.nxv8i32.nxv8i64.i32(
+    <vscale x 8 x i32> %0,
+    <vscale x 8 x i64> %1,
+    i32 %2,
+    <vscale x 8 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 8 x i32> %a
+}
+
 define <vscale x 1 x i8> @intrinsic_vnsra_wi_nxv1i8_nxv1i16_i8(<vscale x 1 x i16> %0, i32 %1) nounwind {
 ; CHECK-LABEL: intrinsic_vnsra_wi_nxv1i8_nxv1i16_i8:
 ; CHECK:       # %bb.0: # %entry
 ; CHECK-NEXT:    vsetvli a0, a0, e8,mf8,ta,mu
 ; CHECK-NEXT:    vnsra.wi v25, v8, 9
 ; CHECK-NEXT:    vmv1r.v v8, v25
 ; CHECK-NEXT:    jalr zero, 0(ra)
 entry:
%a = call <vscale x 16 x i16> @llvm.riscv.vnsra.mask.nxv16i16.nxv16i32.i16(
     <vscale x 16 x i16> %0,
     <vscale x 16 x i32> %1,
     i16 9,
     <vscale x 16 x i1> %2,
     i32 %3)
 
   ret <vscale x 16 x i16> %a
 }
+
+define <vscale x 1 x i32> @intrinsic_vnsra_wi_nxv1i32_nxv1i64_i32(<vscale x 1 x i64> %0, i32 %1) nounwind {
+; CHECK-LABEL: intrinsic_vnsra_wi_nxv1i32_nxv1i64_i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e32,mf2,ta,mu
+; CHECK-NEXT:    vnsra.wi v25, v8, 9
+; CHECK-NEXT:    vmv1r.v v8, v25
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i32> @llvm.riscv.vnsra.nxv1i32.nxv1i64.i32(
+    <vscale x 1 x i64> %0,
+    i32 9,
+    i32 %1)
+
+  ret <vscale x 1 x i32> %a
+}
+
+define <vscale x 1 x i32> @intrinsic_vnsra_mask_wi_nxv1i32_nxv1i64_i32(<vscale x 1 x i32> %0, <vscale x 1 x i64> %1, <vscale x 1 x i1> %2, i32 %3) nounwind {
+; CHECK-LABEL: intrinsic_vnsra_mask_wi_nxv1i32_nxv1i64_i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e32,mf2,tu,mu
+; CHECK-NEXT:    vnsra.wi v8, v9, 9, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i32> @llvm.riscv.vnsra.mask.nxv1i32.nxv1i64.i32(
+    <vscale x 1 x i32> %0,
+    <vscale x 1 x i64> %1,
+    i32 9,
+    <vscale x 1 x i1> %2,
+    i32 %3)
+
+  ret <vscale x 1 x i32> %a
+}
+
+define <vscale x 2 x i32> @intrinsic_vnsra_wi_nxv2i32_nxv2i64_i32(<vscale x 2 x i64> %0, i32 %1) nounwind {
+; CHECK-LABEL: intrinsic_vnsra_wi_nxv2i32_nxv2i64_i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e32,m1,ta,mu
+; CHECK-NEXT:    vnsra.wi v25, v8, 9
+; CHECK-NEXT:    vmv1r.v v8, v25
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i32> @llvm.riscv.vnsra.nxv2i32.nxv2i64.i32(
+    <vscale x 2 x i64> %0,
+    i32 9,
+    i32 %1)
+
+  ret <vscale x 2 x i32> %a
+}
+
+define <vscale x 2 x i32> @intrinsic_vnsra_mask_wi_nxv2i32_nxv2i64_i32(<vscale x 2 x i32> %0, <vscale x 2 x i64> %1, <vscale x 2 x i1> %2, i32 %3) nounwind {
+; CHECK-LABEL: intrinsic_vnsra_mask_wi_nxv2i32_nxv2i64_i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e32,m1,tu,mu
+; CHECK-NEXT:    vnsra.wi v8, v10, 9, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i32> @llvm.riscv.vnsra.mask.nxv2i32.nxv2i64.i32(
+    <vscale x 2 x i32> %0,
+    <vscale x 2 x i64> %1,
+    i32 9,
+    <vscale x 2 x i1> %2,
+    i32 %3)
+
+  ret <vscale x 2 x i32> %a
+}
+
+define <vscale x 4 x i32> @intrinsic_vnsra_wi_nxv4i32_nxv4i64_i32(<vscale x 4 x i64> %0, i32 %1) nounwind {
+; CHECK-LABEL: intrinsic_vnsra_wi_nxv4i32_nxv4i64_i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e32,m2,ta,mu
+; CHECK-NEXT:    vnsra.wi v26, v8, 9
+; CHECK-NEXT:    vmv2r.v v8, v26
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i32> @llvm.riscv.vnsra.nxv4i32.nxv4i64.i32(
+    <vscale x 4 x i64> %0,
+    i32 9,
+    i32 %1)
+
+  ret <vscale x 4 x i32> %a
+}
+
+define <vscale x 4 x i32> @intrinsic_vnsra_mask_wi_nxv4i32_nxv4i64_i32(<vscale x 4 x i32> %0, <vscale x 4 x i64> %1, <vscale x 4 x i1> %2, i32 %3) nounwind {
+; CHECK-LABEL: intrinsic_vnsra_mask_wi_nxv4i32_nxv4i64_i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e32,m2,tu,mu
+; CHECK-NEXT:    vnsra.wi v8, v12, 9, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i32> @llvm.riscv.vnsra.mask.nxv4i32.nxv4i64.i32(
+    <vscale x 4 x i32> %0,
+    <vscale x 4 x i64> %1,
+    i32 9,
+    <vscale x 4 x i1> %2,
+    i32 %3)
+
+  ret <vscale x 4 x i32> %a
+}
+
+define <vscale x 8 x i32> @intrinsic_vnsra_wi_nxv8i32_nxv8i64_i32(<vscale x 8 x i64> %0, i32 %1) nounwind {
+; CHECK-LABEL: intrinsic_vnsra_wi_nxv8i32_nxv8i64_i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e32,m4,ta,mu
+; CHECK-NEXT:    vnsra.wi v28, v8, 9
+; CHECK-NEXT:    vmv4r.v v8, v28
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i32> @llvm.riscv.vnsra.nxv8i32.nxv8i64.i32(
+    <vscale x 8 x i64> %0,
+    i32 9,
+    i32 %1)
+
+  ret <vscale x 8 x i32> %a
+}
+
+define <vscale x 8 x i32> @intrinsic_vnsra_mask_wi_nxv8i32_nxv8i64_i32(<vscale x 8 x i32> %0, <vscale x 8 x i64> %1, <vscale x 8 x i1> %2, i32 %3) nounwind {
+; CHECK-LABEL: intrinsic_vnsra_mask_wi_nxv8i32_nxv8i64_i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e32,m4,tu,mu
+; CHECK-NEXT:    vnsra.wi v8, v16, 9, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i32> @llvm.riscv.vnsra.mask.nxv8i32.nxv8i64.i32(
+    <vscale x 8 x i32> %0,
+    <vscale x 8 x i64> %1,
+    i32 9,
+    <vscale x 8 x i1> %2,
+    i32 %3)
+
+  ret <vscale x 8 x i32> %a
+}

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

%a = call <vscale x 16 x i16> @llvm.riscv.vnsrl.mask.nxv16i16.nxv16i32.nxv16i16(
     <vscale x 16 x i32> %1,
     <vscale x 16 x i16> %2,
     <vscale x 16 x i1> %3,
     i32 %4)
 
   ret <vscale x 16 x i16> %a
 }
 
+declare <vscale x 1 x i32> @llvm.riscv.vnsrl.nxv1i32.nxv1i64.nxv1i32(
+  <vscale x 1 x i64>,
+  <vscale x 1 x i32>,
+  i32);
+
+define <vscale x 1 x i32> @intrinsic_vnsrl_wv_nxv1i32_nxv1i64_nxv1i32(<vscale x 1 x i64> %0, <vscale x 1 x i32> %1, i32 %2) nounwind {
+; CHECK-LABEL: intrinsic_vnsrl_wv_nxv1i32_nxv1i64_nxv1i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e32,mf2,ta,mu
+; CHECK-NEXT:    vnsrl.wv v25, v8, v9
+; CHECK-NEXT:    vmv1r.v v8, v25
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i32> @llvm.riscv.vnsrl.nxv1i32.nxv1i64.nxv1i32(
+    <vscale x 1 x i64> %0,
+    <vscale x 1 x i32> %1,
+    i32 %2)
+
+  ret <vscale x 1 x i32> %a
+}
+
+declare <vscale x 1 x i32> @llvm.riscv.vnsrl.mask.nxv1i32.nxv1i64.nxv1i32(
+  <vscale x 1 x i32>,
+  <vscale x 1 x i64>,
+  <vscale x 1 x i32>,
+  <vscale x 1 x i1>,
+  i32);
+
+define <vscale x 1 x i32> @intrinsic_vnsrl_mask_wv_nxv1i32_nxv1i64_nxv1i32(<vscale x 1 x i32> %0, <vscale x 1 x i64> %1, <vscale x 1 x i32> %2, <vscale x 1 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vnsrl_mask_wv_nxv1i32_nxv1i64_nxv1i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e32,mf2,tu,mu
+; CHECK-NEXT:    vnsrl.wv v8, v9, v10, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i32> @llvm.riscv.vnsrl.mask.nxv1i32.nxv1i64.nxv1i32(
+    <vscale x 1 x i32> %0,
+    <vscale x 1 x i64> %1,
+    <vscale x 1 x i32> %2,
+    <vscale x 1 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 1 x i32> %a
+}
+
+declare <vscale x 2 x i32> @llvm.riscv.vnsrl.nxv2i32.nxv2i64.nxv2i32(
+  <vscale x 2 x i64>,
+  <vscale x 2 x i32>,
+  i32);
+
+define <vscale x 2 x i32> @intrinsic_vnsrl_wv_nxv2i32_nxv2i64_nxv2i32(<vscale x 2 x i64> %0, <vscale x 2 x i32> %1, i32 %2) nounwind {
+; CHECK-LABEL: intrinsic_vnsrl_wv_nxv2i32_nxv2i64_nxv2i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e32,m1,ta,mu
+; CHECK-NEXT:    vnsrl.wv v25, v8, v10
+; CHECK-NEXT:    vmv1r.v v8, v25
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i32> @llvm.riscv.vnsrl.nxv2i32.nxv2i64.nxv2i32(
+    <vscale x 2 x i64> %0,
+    <vscale x 2 x i32> %1,
+    i32 %2)
+
+  ret <vscale x 2 x i32> %a
+}
+
+declare <vscale x 2 x i32> @llvm.riscv.vnsrl.mask.nxv2i32.nxv2i64.nxv2i32(
+  <vscale x 2 x i32>,
+  <vscale x 2 x i64>,
+  <vscale x 2 x i32>,
+  <vscale x 2 x i1>,
+  i32);
+
+define <vscale x 2 x i32> @intrinsic_vnsrl_mask_wv_nxv2i32_nxv2i64_nxv2i32(<vscale x 2 x i32> %0, <vscale x 2 x i64> %1, <vscale x 2 x i32> %2, <vscale x 2 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vnsrl_mask_wv_nxv2i32_nxv2i64_nxv2i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e32,m1,tu,mu
+; CHECK-NEXT:    vnsrl.wv v8, v10, v9, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i32> @llvm.riscv.vnsrl.mask.nxv2i32.nxv2i64.nxv2i32(
+    <vscale x 2 x i32> %0,
+    <vscale x 2 x i64> %1,
+    <vscale x 2 x i32> %2,
+    <vscale x 2 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 2 x i32> %a
+}
+
+declare <vscale x 4 x i32> @llvm.riscv.vnsrl.nxv4i32.nxv4i64.nxv4i32(
+  <vscale x 4 x i64>,
+  <vscale x 4 x i32>,
+  i32);
+
+define <vscale x 4 x i32> @intrinsic_vnsrl_wv_nxv4i32_nxv4i64_nxv4i32(<vscale x 4 x i64> %0, <vscale x 4 x i32> %1, i32 %2) nounwind {
+; CHECK-LABEL: intrinsic_vnsrl_wv_nxv4i32_nxv4i64_nxv4i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e32,m2,ta,mu
+; CHECK-NEXT:    vnsrl.wv v26, v8, v12
+; CHECK-NEXT:    vmv2r.v v8, v26
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i32> @llvm.riscv.vnsrl.nxv4i32.nxv4i64.nxv4i32(
+    <vscale x 4 x i64> %0,
+    <vscale x 4 x i32> %1,
+    i32 %2)
+
+  ret <vscale x 4 x i32> %a
+}
+
+declare <vscale x 4 x i32> @llvm.riscv.vnsrl.mask.nxv4i32.nxv4i64.nxv4i32(
+  <vscale x 4 x i32>,
+  <vscale x 4 x i64>,
+  <vscale x 4 x i32>,
+  <vscale x 4 x i1>,
+  i32);
+
+define <vscale x 4 x i32> @intrinsic_vnsrl_mask_wv_nxv4i32_nxv4i64_nxv4i32(<vscale x 4 x i32> %0, <vscale x 4 x i64> %1, <vscale x 4 x i32> %2, <vscale x 4 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vnsrl_mask_wv_nxv4i32_nxv4i64_nxv4i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e32,m2,tu,mu
+; CHECK-NEXT:    vnsrl.wv v8, v12, v10, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i32> @llvm.riscv.vnsrl.mask.nxv4i32.nxv4i64.nxv4i32(
+    <vscale x 4 x i32> %0,
+    <vscale x 4 x i64> %1,
+    <vscale x 4 x i32> %2,
+    <vscale x 4 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 4 x i32> %a
+}
+
+declare <vscale x 8 x i32> @llvm.riscv.vnsrl.nxv8i32.nxv8i64.nxv8i32(
+  <vscale x 8 x i64>,
+  <vscale x 8 x i32>,
+  i32);
+
+define <vscale x 8 x i32> @intrinsic_vnsrl_wv_nxv8i32_nxv8i64_nxv8i32(<vscale x 8 x i64> %0, <vscale x 8 x i32> %1, i32 %2) nounwind {
+; CHECK-LABEL: intrinsic_vnsrl_wv_nxv8i32_nxv8i64_nxv8i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e32,m4,ta,mu
+; CHECK-NEXT:    vnsrl.wv v28, v8, v16
+; CHECK-NEXT:    vmv4r.v v8, v28
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i32> @llvm.riscv.vnsrl.nxv8i32.nxv8i64.nxv8i32(
+    <vscale x 8 x i64> %0,
+    <vscale x 8 x i32> %1,
+    i32 %2)
+
+  ret <vscale x 8 x i32> %a
+}
+
+declare <vscale x 8 x i32> @llvm.riscv.vnsrl.mask.nxv8i32.nxv8i64.nxv8i32(
+  <vscale x 8 x i32>,
+  <vscale x 8 x i64>,
+  <vscale x 8 x i32>,
+  <vscale x 8 x i1>,
+  i32);
+
+define <vscale x 8 x i32> @intrinsic_vnsrl_mask_wv_nxv8i32_nxv8i64_nxv8i32(<vscale x 8 x i32> %0, <vscale x 8 x i64> %1, <vscale x 8 x i32> %2, <vscale x 8 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vnsrl_mask_wv_nxv8i32_nxv8i64_nxv8i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e32,m4,tu,mu
+; CHECK-NEXT:    vnsrl.wv v8, v16, v12, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i32> @llvm.riscv.vnsrl.mask.nxv8i32.nxv8i64.nxv8i32(
+    <vscale x 8 x i32> %0,
+    <vscale x 8 x i64> %1,
+    <vscale x 8 x i32> %2,
+    <vscale x 8 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 8 x i32> %a
+}
+
 declare <vscale x 1 x i8> @llvm.riscv.vnsrl.nxv1i8.nxv1i16.i8(
   <vscale x 1 x i16>,
   i8,
   i32);
 
 define <vscale x 1 x i8> @intrinsic_vnsrl_wx_nxv1i8_nxv1i16_i8(<vscale x 1 x i16> %0, i8 %1, i32 %2) nounwind {
 ; CHECK-LABEL: intrinsic_vnsrl_wx_nxv1i8_nxv1i16_i8:
 ; CHECK:       # %bb.0: # %entry
%a = call <vscale x 16 x i16> @llvm.riscv.vnsrl.mask.nxv16i16.nxv16i32.i16(
     <vscale x 16 x i32> %1,
     i16 %2,
     <vscale x 16 x i1> %3,
     i32 %4)
 
   ret <vscale x 16 x i16> %a
 }
 
+declare <vscale x 1 x i32> @llvm.riscv.vnsrl.nxv1i32.nxv1i64.i32(
+  <vscale x 1 x i64>,
+  i32,
+  i32);
+
+define <vscale x 1 x i32> @intrinsic_vnsrl_wx_nxv1i32_nxv1i64_i32(<vscale x 1 x i64> %0, i32 %1, i32 %2) nounwind {
+; CHECK-LABEL: intrinsic_vnsrl_wx_nxv1i32_nxv1i64_i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e32,mf2,ta,mu
+; CHECK-NEXT:    vnsrl.wx v25, v8, a0
+; CHECK-NEXT:    vmv1r.v v8, v25
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i32> @llvm.riscv.vnsrl.nxv1i32.nxv1i64.i32(
+    <vscale x 1 x i64> %0,
+    i32 %1,
+    i32 %2)
+
+  ret <vscale x 1 x i32> %a
+}
+
+declare <vscale x 1 x i32> @llvm.riscv.vnsrl.mask.nxv1i32.nxv1i64.i32(
+  <vscale x 1 x i32>,
+  <vscale x 1 x i64>,
+  i32,
+  <vscale x 1 x i1>,
+  i32);
+
+define <vscale x 1 x i32> @intrinsic_vnsrl_mask_wx_nxv1i32_nxv1i64_i32(<vscale x 1 x i32> %0, <vscale x 1 x i64> %1, i32 %2, <vscale x 1 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vnsrl_mask_wx_nxv1i32_nxv1i64_i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e32,mf2,tu,mu
+; CHECK-NEXT:    vnsrl.wx v8, v9, a0, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i32> @llvm.riscv.vnsrl.mask.nxv1i32.nxv1i64.i32(
+    <vscale x 1 x i32> %0,
+    <vscale x 1 x i64> %1,
+    i32 %2,
+    <vscale x 1 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 1 x i32> %a
+}
+
+declare <vscale x 2 x i32> @llvm.riscv.vnsrl.nxv2i32.nxv2i64.i32(
+  <vscale x 2 x i64>,
+  i32,
+  i32);
+
+define <vscale x 2 x i32> @intrinsic_vnsrl_wx_nxv2i32_nxv2i64_i32(<vscale x 2 x i64> %0, i32 %1, i32 %2) nounwind {
+; CHECK-LABEL: intrinsic_vnsrl_wx_nxv2i32_nxv2i64_i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e32,m1,ta,mu
+; CHECK-NEXT:    vnsrl.wx v25, v8, a0
+; CHECK-NEXT:    vmv1r.v v8, v25
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i32> @llvm.riscv.vnsrl.nxv2i32.nxv2i64.i32(
+    <vscale x 2 x i64> %0,
+    i32 %1,
+    i32 %2)
+
+  ret <vscale x 2 x i32> %a
+}
+
+declare <vscale x 2 x i32> @llvm.riscv.vnsrl.mask.nxv2i32.nxv2i64.i32(
+  <vscale x 2 x i32>,
+  <vscale x 2 x i64>,
+  i32,
+  <vscale x 2 x i1>,
+  i32);
+
+define <vscale x 2 x i32> @intrinsic_vnsrl_mask_wx_nxv2i32_nxv2i64_i32(<vscale x 2 x i32> %0, <vscale x 2 x i64> %1, i32 %2, <vscale x 2 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vnsrl_mask_wx_nxv2i32_nxv2i64_i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e32,m1,tu,mu
+; CHECK-NEXT:    vnsrl.wx v8, v10, a0, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i32> @llvm.riscv.vnsrl.mask.nxv2i32.nxv2i64.i32(
+    <vscale x 2 x i32> %0,
+    <vscale x 2 x i64> %1,
+    i32 %2,
+    <vscale x 2 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 2 x i32> %a
+}
+
+declare <vscale x 4 x i32> @llvm.riscv.vnsrl.nxv4i32.nxv4i64.i32(
+  <vscale x 4 x i64>,
+  i32,
+  i32);
+
+define <vscale x 4 x i32> @intrinsic_vnsrl_wx_nxv4i32_nxv4i64_i32(<vscale x 4 x i64> %0, i32 %1, i32 %2) nounwind {
+; CHECK-LABEL: intrinsic_vnsrl_wx_nxv4i32_nxv4i64_i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e32,m2,ta,mu
+; CHECK-NEXT:    vnsrl.wx v26, v8, a0
+; CHECK-NEXT:    vmv2r.v v8, v26
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i32> @llvm.riscv.vnsrl.nxv4i32.nxv4i64.i32(
+    <vscale x 4 x i64> %0,
+    i32 %1,
+    i32 %2)
+
+  ret <vscale x 4 x i32> %a
+}
+
+declare <vscale x 4 x i32> @llvm.riscv.vnsrl.mask.nxv4i32.nxv4i64.i32(
+  <vscale x 4 x i32>,
+  <vscale x 4 x i64>,
+  i32,
+  <vscale x 4 x i1>,
+  i32);
+
+define <vscale x 4 x i32> @intrinsic_vnsrl_mask_wx_nxv4i32_nxv4i64_i32(<vscale x 4 x i32> %0, <vscale x 4 x i64> %1, i32 %2, <vscale x 4 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vnsrl_mask_wx_nxv4i32_nxv4i64_i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e32,m2,tu,mu
+; CHECK-NEXT:    vnsrl.wx v8, v12, a0, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i32> @llvm.riscv.vnsrl.mask.nxv4i32.nxv4i64.i32(
+    <vscale x 4 x i32> %0,
+    <vscale x 4 x i64> %1,
+    i32 %2,
+    <vscale x 4 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 4 x i32> %a
+}
+
+declare <vscale x 8 x i32> @llvm.riscv.vnsrl.nxv8i32.nxv8i64.i32(
+  <vscale x 8 x i64>,
+  i32,
+  i32);
+
+define <vscale x 8 x i32> @intrinsic_vnsrl_wx_nxv8i32_nxv8i64_i32(<vscale x 8 x i64> %0, i32 %1, i32 %2) nounwind {
+; CHECK-LABEL: intrinsic_vnsrl_wx_nxv8i32_nxv8i64_i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e32,m4,ta,mu
+; CHECK-NEXT:    vnsrl.wx v28, v8, a0
+; CHECK-NEXT:    vmv4r.v v8, v28
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i32> @llvm.riscv.vnsrl.nxv8i32.nxv8i64.i32(
+    <vscale x 8 x i64> %0,
+    i32 %1,
+    i32 %2)
+
+  ret <vscale x 8 x i32> %a
+}
+
+declare <vscale x 8 x i32> @llvm.riscv.vnsrl.mask.nxv8i32.nxv8i64.i32(
+  <vscale x 8 x i32>,
+  <vscale x 8 x i64>,
+  i32,
+  <vscale x 8 x i1>,
+  i32);
+
+define <vscale x 8 x i32> @intrinsic_vnsrl_mask_wx_nxv8i32_nxv8i64_i32(<vscale x 8 x i32> %0, <vscale x 8 x i64> %1, i32 %2, <vscale x 8 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vnsrl_mask_wx_nxv8i32_nxv8i64_i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e32,m4,tu,mu
+; CHECK-NEXT:    vnsrl.wx v8, v16, a0, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i32> @llvm.riscv.vnsrl.mask.nxv8i32.nxv8i64.i32(
+    <vscale x 8 x i32> %0,
+    <vscale x 8 x i64> %1,
+    i32 %2,
+    <vscale x 8 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 8 x i32> %a
+}
+
 define <vscale x 1 x i8> @intrinsic_vnsrl_wi_nxv1i8_nxv1i16_i8(<vscale x 1 x i16> %0, i32 %1) nounwind {
 ; CHECK-LABEL: intrinsic_vnsrl_wi_nxv1i8_nxv1i16_i8:
 ; CHECK:       # %bb.0: # %entry
 ; CHECK-NEXT:    vsetvli a0, a0, e8,mf8,ta,mu
 ; CHECK-NEXT:    vnsrl.wi v25, v8, 9
 ; CHECK-NEXT:    vmv1r.v v8, v25
 ; CHECK-NEXT:    jalr zero, 0(ra)
 entry:
%a = call <vscale x 16 x i16> @llvm.riscv.vnsrl.mask.nxv16i16.nxv16i32.i16(
     <vscale x 16 x i16> %0,
     <vscale x 16 x i32> %1,
     i16 9,
     <vscale x 16 x i1> %2,
     i32 %3)
 
   ret <vscale x 16 x i16> %a
 }
+
+define <vscale x 1 x i32> @intrinsic_vnsrl_wi_nxv1i32_nxv1i64_i32(<vscale x 1 x i64> %0, i32 %1) nounwind {
+; CHECK-LABEL: intrinsic_vnsrl_wi_nxv1i32_nxv1i64_i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e32,mf2,ta,mu
+; CHECK-NEXT:    vnsrl.wi v25, v8, 9
+; CHECK-NEXT:    vmv1r.v v8, v25
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i32> @llvm.riscv.vnsrl.nxv1i32.nxv1i64.i32(
+    <vscale x 1 x i64> %0,
+    i32 9,
+    i32 %1)
+
+  ret <vscale x 1 x i32> %a
+}
+
+define <vscale x 1 x i32> @intrinsic_vnsrl_mask_wi_nxv1i32_nxv1i64_i32(<vscale x 1 x i32> %0, <vscale x 1 x i64> %1, <vscale x 1 x i1> %2, i32 %3) nounwind {
+; CHECK-LABEL: intrinsic_vnsrl_mask_wi_nxv1i32_nxv1i64_i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e32,mf2,tu,mu
+; CHECK-NEXT:    vnsrl.wi v8, v9, 9, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i32> @llvm.riscv.vnsrl.mask.nxv1i32.nxv1i64.i32(
+    <vscale x 1 x i32> %0,
+    <vscale x 1 x i64> %1,
+    i32 9,
+    <vscale x 1 x i1> %2,
+    i32 %3)
+
+  ret <vscale x 1 x i32> %a
+}
+
+define <vscale x 2 x i32> @intrinsic_vnsrl_wi_nxv2i32_nxv2i64_i32(<vscale x 2 x i64> %0, i32 %1) nounwind {
+; CHECK-LABEL: intrinsic_vnsrl_wi_nxv2i32_nxv2i64_i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e32,m1,ta,mu
+; CHECK-NEXT:    vnsrl.wi v25, v8, 9
+; CHECK-NEXT:    vmv1r.v v8, v25
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i32> @llvm.riscv.vnsrl.nxv2i32.nxv2i64.i32(
+    <vscale x 2 x i64> %0,
+    i32 9,
+    i32 %1)
+
+  ret <vscale x 2 x i32> %a
+}
+
+define <vscale x 2 x i32> @intrinsic_vnsrl_mask_wi_nxv2i32_nxv2i64_i32(<vscale x 2 x i32> %0, <vscale x 2 x i64> %1, <vscale x 2 x i1> %2, i32 %3) nounwind {
+; CHECK-LABEL: intrinsic_vnsrl_mask_wi_nxv2i32_nxv2i64_i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e32,m1,tu,mu
+; CHECK-NEXT:    vnsrl.wi v8, v10, 9, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i32> @llvm.riscv.vnsrl.mask.nxv2i32.nxv2i64.i32(
+    <vscale x 2 x i32> %0,
+    <vscale x 2 x i64> %1,
+    i32 9,
+    <vscale x 2 x i1> %2,
+    i32 %3)
+
+  ret <vscale x 2 x i32> %a
+}
+
+define <vscale x 4 x i32> @intrinsic_vnsrl_wi_nxv4i32_nxv4i64_i32(<vscale x 4 x i64> %0, i32 %1) nounwind {
+; CHECK-LABEL: intrinsic_vnsrl_wi_nxv4i32_nxv4i64_i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e32,m2,ta,mu
+; CHECK-NEXT:    vnsrl.wi v26, v8, 9
+; CHECK-NEXT:    vmv2r.v v8, v26
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i32> @llvm.riscv.vnsrl.nxv4i32.nxv4i64.i32(
+    <vscale x 4 x i64> %0,
+    i32 9,
+    i32 %1)
+
+  ret <vscale x 4 x i32> %a
+}
+
+define <vscale x 4 x i32> @intrinsic_vnsrl_mask_wi_nxv4i32_nxv4i64_i32(<vscale x 4 x i32> %0, <vscale x 4 x i64> %1, <vscale x 4 x i1> %2, i32 %3) nounwind {
+; CHECK-LABEL: intrinsic_vnsrl_mask_wi_nxv4i32_nxv4i64_i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e32,m2,tu,mu
+; CHECK-NEXT:    vnsrl.wi v8, v12, 9, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i32> @llvm.riscv.vnsrl.mask.nxv4i32.nxv4i64.i32(
+    <vscale x 4 x i32> %0,
+    <vscale x 4 x i64> %1,
+    i32 9,
+    <vscale x 4 x i1> %2,
+    i32 %3)
+
+  ret <vscale x 4 x i32> %a
+}
+
+define <vscale x 8 x i32> @intrinsic_vnsrl_wi_nxv8i32_nxv8i64_i32(<vscale x 8 x i64> %0, i32 %1) nounwind {
+; CHECK-LABEL: intrinsic_vnsrl_wi_nxv8i32_nxv8i64_i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e32,m4,ta,mu
+; CHECK-NEXT:    vnsrl.wi v28, v8, 9
+; CHECK-NEXT:    vmv4r.v v8, v28
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i32> @llvm.riscv.vnsrl.nxv8i32.nxv8i64.i32(
+    <vscale x 8 x i64> %0,
+    i32 9,
+    i32 %1)
+
+  ret <vscale x 8 x i32> %a
+}
+
+define <vscale x 8 x i32> @intrinsic_vnsrl_mask_wi_nxv8i32_nxv8i64_i32(<vscale x 8 x i32> %0, <vscale x 8 x i64> %1, <vscale x 8 x i1> %2, i32 %3) nounwind {
+; CHECK-LABEL: intrinsic_vnsrl_mask_wi_nxv8i32_nxv8i64_i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e32,m4,tu,mu
+; CHECK-NEXT:    vnsrl.wi v8, v16, 9, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i32> @llvm.riscv.vnsrl.mask.nxv8i32.nxv8i64.i32(
+    <vscale x 8 x i32> %0,
+    <vscale x 8 x i64> %1,
+    i32 9,
+    <vscale x 8 x i1> %2,
+    i32 %3)
+
+  ret <vscale x 8 x i32> %a
+}

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

%a = call <vscale x 16 x i32> @llvm.riscv.vssra.mask.nxv16i32.nxv16i32(
     <vscale x 16 x i32> %1,
     <vscale x 16 x i32> %2,
     <vscale x 16 x i1> %3,
     i32 %4)
 
   ret <vscale x 16 x i32> %a
 }
 
+declare <vscale x 1 x i64> @llvm.riscv.vssra.nxv1i64.nxv1i64(
+  <vscale x 1 x i64>,
+  <vscale x 1 x i64>,
+  i32);
+
+define <vscale x 1 x i64> @intrinsic_vssra_vv_nxv1i64_nxv1i64_nxv1i64(<vscale x 1 x i64> %0, <vscale x 1 x i64> %1, i32 %2) nounwind {
+; CHECK-LABEL: intrinsic_vssra_vv_nxv1i64_nxv1i64_nxv1i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e64,m1,ta,mu
+; CHECK-NEXT:    vssra.vv v8, v8, v9
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i64> @llvm.riscv.vssra.nxv1i64.nxv1i64(
+    <vscale x 1 x i64> %0,
+    <vscale x 1 x i64> %1,
+    i32 %2)
+
+  ret <vscale x 1 x i64> %a
+}
+
+declare <vscale x 1 x i64> @llvm.riscv.vssra.mask.nxv1i64.nxv1i64(
+  <vscale x 1 x i64>,
+  <vscale x 1 x i64>,
+  <vscale x 1 x i64>,
+  <vscale x 1 x i1>,
+  i32);
+
+define <vscale x 1 x i64> @intrinsic_vssra_mask_vv_nxv1i64_nxv1i64_nxv1i64(<vscale x 1 x i64> %0, <vscale x 1 x i64> %1, <vscale x 1 x i64> %2, <vscale x 1 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vssra_mask_vv_nxv1i64_nxv1i64_nxv1i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e64,m1,tu,mu
+; CHECK-NEXT:    vssra.vv v8, v9, v10, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i64> @llvm.riscv.vssra.mask.nxv1i64.nxv1i64(
+    <vscale x 1 x i64> %0,
+    <vscale x 1 x i64> %1,
+    <vscale x 1 x i64> %2,
+    <vscale x 1 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 1 x i64> %a
+}
+
+declare <vscale x 2 x i64> @llvm.riscv.vssra.nxv2i64.nxv2i64(
+  <vscale x 2 x i64>,
+  <vscale x 2 x i64>,
+  i32);
+
+define <vscale x 2 x i64> @intrinsic_vssra_vv_nxv2i64_nxv2i64_nxv2i64(<vscale x 2 x i64> %0, <vscale x 2 x i64> %1, i32 %2) nounwind {
+; CHECK-LABEL: intrinsic_vssra_vv_nxv2i64_nxv2i64_nxv2i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e64,m2,ta,mu
+; CHECK-NEXT:    vssra.vv v8, v8, v10
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i64> @llvm.riscv.vssra.nxv2i64.nxv2i64(
+    <vscale x 2 x i64> %0,
+    <vscale x 2 x i64> %1,
+    i32 %2)
+
+  ret <vscale x 2 x i64> %a
+}
+
+declare <vscale x 2 x i64> @llvm.riscv.vssra.mask.nxv2i64.nxv2i64(
+  <vscale x 2 x i64>,
+  <vscale x 2 x i64>,
+  <vscale x 2 x i64>,
+  <vscale x 2 x i1>,
+  i32);
+
+define <vscale x 2 x i64> @intrinsic_vssra_mask_vv_nxv2i64_nxv2i64_nxv2i64(<vscale x 2 x i64> %0, <vscale x 2 x i64> %1, <vscale x 2 x i64> %2, <vscale x 2 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vssra_mask_vv_nxv2i64_nxv2i64_nxv2i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e64,m2,tu,mu
+; CHECK-NEXT:    vssra.vv v8, v10, v12, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i64> @llvm.riscv.vssra.mask.nxv2i64.nxv2i64(
+    <vscale x 2 x i64> %0,
+    <vscale x 2 x i64> %1,
+    <vscale x 2 x i64> %2,
+    <vscale x 2 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 2 x i64> %a
+}
+
+declare <vscale x 4 x i64> @llvm.riscv.vssra.nxv4i64.nxv4i64(
+  <vscale x 4 x i64>,
+  <vscale x 4 x i64>,
+  i32);
+
+define <vscale x 4 x i64> @intrinsic_vssra_vv_nxv4i64_nxv4i64_nxv4i64(<vscale x 4 x i64> %0, <vscale x 4 x i64> %1, i32 %2) nounwind {
+; CHECK-LABEL: intrinsic_vssra_vv_nxv4i64_nxv4i64_nxv4i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e64,m4,ta,mu
+; CHECK-NEXT:    vssra.vv v8, v8, v12
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i64> @llvm.riscv.vssra.nxv4i64.nxv4i64(
+    <vscale x 4 x i64> %0,
+    <vscale x 4 x i64> %1,
+    i32 %2)
+
+  ret <vscale x 4 x i64> %a
+}
+
+declare <vscale x 4 x i64> @llvm.riscv.vssra.mask.nxv4i64.nxv4i64(
+  <vscale x 4 x i64>,
+  <vscale x 4 x i64>,
+  <vscale x 4 x i64>,
+  <vscale x 4 x i1>,
+  i32);
+
+define <vscale x 4 x i64> @intrinsic_vssra_mask_vv_nxv4i64_nxv4i64_nxv4i64(<vscale x 4 x i64> %0, <vscale x 4 x i64> %1, <vscale x 4 x i64> %2, <vscale x 4 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vssra_mask_vv_nxv4i64_nxv4i64_nxv4i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e64,m4,tu,mu
+; CHECK-NEXT:    vssra.vv v8, v12, v16, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i64> @llvm.riscv.vssra.mask.nxv4i64.nxv4i64(
+    <vscale x 4 x i64> %0,
+    <vscale x 4 x i64> %1,
+    <vscale x 4 x i64> %2,
+    <vscale x 4 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 4 x i64> %a
+}
+
+declare <vscale x 8 x i64> @llvm.riscv.vssra.nxv8i64.nxv8i64(
+  <vscale x 8 x i64>,
+  <vscale x 8 x i64>,
+  i32);
+
+define <vscale x 8 x i64> @intrinsic_vssra_vv_nxv8i64_nxv8i64_nxv8i64(<vscale x 8 x i64> %0, <vscale x 8 x i64> %1, i32 %2) nounwind {
+; CHECK-LABEL: intrinsic_vssra_vv_nxv8i64_nxv8i64_nxv8i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e64,m8,ta,mu
+; CHECK-NEXT:    vssra.vv v8, v8, v16
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i64> @llvm.riscv.vssra.nxv8i64.nxv8i64(
+    <vscale x 8 x i64> %0,
+    <vscale x 8 x i64> %1,
+    i32 %2)
+
+  ret <vscale x 8 x i64> %a
+}
+
+declare <vscale x 8 x i64> @llvm.riscv.vssra.mask.nxv8i64.nxv8i64(
+  <vscale x 8 x i64>,
+  <vscale x 8 x i64>,
+  <vscale x 8 x i64>,
+  <vscale x 8 x i1>,
+  i32);
+
+define <vscale x 8 x i64> @intrinsic_vssra_mask_vv_nxv8i64_nxv8i64_nxv8i64(<vscale x 8 x i64> %0, <vscale x 8 x i64> %1, <vscale x 8 x i64> %2, <vscale x 8 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vssra_mask_vv_nxv8i64_nxv8i64_nxv8i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vl8re64.v v24, (a0)
+; CHECK-NEXT:    vsetvli a0, a1, e64,m8,tu,mu
+; CHECK-NEXT:    vssra.vv v8, v16, v24, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i64> @llvm.riscv.vssra.mask.nxv8i64.nxv8i64(
+    <vscale x 8 x i64> %0,
+    <vscale x 8 x i64> %1,
+    <vscale x 8 x i64> %2,
+    <vscale x 8 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 8 x i64> %a
+}
+
 declare <vscale x 1 x i8> @llvm.riscv.vssra.nxv1i8.i8(
   <vscale x 1 x i8>,
   i8,
   i32);
 
 define <vscale x 1 x i8> @intrinsic_vssra_vx_nxv1i8_nxv1i8_i8(<vscale x 1 x i8> %0, i8 %1, i32 %2) nounwind {
 ; CHECK-LABEL: intrinsic_vssra_vx_nxv1i8_nxv1i8_i8:
 ; CHECK:       # %bb.0: # %entry
%a = call <vscale x 16 x i32> @llvm.riscv.vssra.mask.nxv16i32.i32(
     <vscale x 16 x i32> %1,
     i32 %2,
     <vscale x 16 x i1> %3,
     i32 %4)
 
   ret <vscale x 16 x i32> %a
 }
 
+declare <vscale x 1 x i64> @llvm.riscv.vssra.nxv1i64.i64(
+  <vscale x 1 x i64>,
+  i64,
+  i32);
+
+define <vscale x 1 x i64> @intrinsic_vssra_vx_nxv1i64_nxv1i64_i64(<vscale x 1 x i64> %0, i64 %1, i32 %2) nounwind {
+; CHECK-LABEL: intrinsic_vssra_vx_nxv1i64_nxv1i64_i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a2, a2, e64,m1,ta,mu
+; CHECK-NEXT:    vmv.v.x v25, a1
+; CHECK-NEXT:    addi a1, zero, 32
+; CHECK-NEXT:    vsll.vx v25, v25, a1
+; CHECK-NEXT:    vmv.v.x v26, a0
+; CHECK-NEXT:    vsll.vx v26, v26, a1
+; CHECK-NEXT:    vsrl.vx v26, v26, a1
+; CHECK-NEXT:    vor.vv v25, v26, v25
+; CHECK-NEXT:    vssra.vv v8, v8, v25
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i64> @llvm.riscv.vssra.nxv1i64.i64(
+    <vscale x 1 x i64> %0,
+    i64 %1,
+    i32 %2)
+
+  ret <vscale x 1 x i64> %a
+}
+
+declare <vscale x 1 x i64> @llvm.riscv.vssra.mask.nxv1i64.i64(
+  <vscale x 1 x i64>,
+  <vscale x 1 x i64>,
+  i64,
+  <vscale x 1 x i1>,
+  i32);
+
+define <vscale x 1 x i64> @intrinsic_vssra_mask_vx_nxv1i64_nxv1i64_i64(<vscale x 1 x i64> %0, <vscale x 1 x i64> %1, i64 %2, <vscale x 1 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vssra_mask_vx_nxv1i64_nxv1i64_i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a3, a2, e64,m1,ta,mu
+; CHECK-NEXT:    vmv.v.x v25, a1
+; CHECK-NEXT:    addi a1, zero, 32
+; CHECK-NEXT:    vsll.vx v25, v25, a1
+; CHECK-NEXT:    vmv.v.x v26, a0
+; CHECK-NEXT:    vsll.vx v26, v26, a1
+; CHECK-NEXT:    vsrl.vx v26, v26, a1
+; CHECK-NEXT:    vor.vv v25, v26, v25
+; CHECK-NEXT:    vsetvli a0, a2, e64,m1,tu,mu
+; CHECK-NEXT:    vssra.vv v8, v9, v25, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i64> @llvm.riscv.vssra.mask.nxv1i64.i64(
+    <vscale x 1 x i64> %0,
+    <vscale x 1 x i64> %1,
+    i64 %2,
+    <vscale x 1 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 1 x i64> %a
+}
+
+declare <vscale x 2 x i64> @llvm.riscv.vssra.nxv2i64.i64(
+  <vscale x 2 x i64>,
+  i64,
+  i32);
+
+define <vscale x 2 x i64> @intrinsic_vssra_vx_nxv2i64_nxv2i64_i64(<vscale x 2 x i64> %0, i64 %1, i32 %2) nounwind {
+; CHECK-LABEL: intrinsic_vssra_vx_nxv2i64_nxv2i64_i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a2, a2, e64,m2,ta,mu
+; CHECK-NEXT:    vmv.v.x v26, a1
+; CHECK-NEXT:    addi a1, zero, 32
+; CHECK-NEXT:    vsll.vx v26, v26, a1
+; CHECK-NEXT:    vmv.v.x v28, a0
+; CHECK-NEXT:    vsll.vx v28, v28, a1
+; CHECK-NEXT:    vsrl.vx v28, v28, a1
+; CHECK-NEXT:    vor.vv v26, v28, v26
+; CHECK-NEXT:    vssra.vv v8, v8, v26
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i64> @llvm.riscv.vssra.nxv2i64.i64(
+    <vscale x 2 x i64> %0,
+    i64 %1,
+    i32 %2)
+
+  ret <vscale x 2 x i64> %a
+}
+
+declare <vscale x 2 x i64> @llvm.riscv.vssra.mask.nxv2i64.i64(
+  <vscale x 2 x i64>,
+  <vscale x 2 x i64>,
+  i64,
+  <vscale x 2 x i1>,
+  i32);
+
+define <vscale x 2 x i64> @intrinsic_vssra_mask_vx_nxv2i64_nxv2i64_i64(<vscale x 2 x i64> %0, <vscale x 2 x i64> %1, i64 %2, <vscale x 2 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vssra_mask_vx_nxv2i64_nxv2i64_i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a3, a2, e64,m2,ta,mu
+; CHECK-NEXT:    vmv.v.x v26, a1
+; CHECK-NEXT:    addi a1, zero, 32
+; CHECK-NEXT:    vsll.vx v26, v26, a1
+; CHECK-NEXT:    vmv.v.x v28, a0
+; CHECK-NEXT:    vsll.vx v28, v28, a1
+; CHECK-NEXT:    vsrl.vx v28, v28, a1
+; CHECK-NEXT:    vor.vv v26, v28, v26
+; CHECK-NEXT:    vsetvli a0, a2, e64,m2,tu,mu
+; CHECK-NEXT:    vssra.vv v8, v10, v26, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i64> @llvm.riscv.vssra.mask.nxv2i64.i64(
+    <vscale x 2 x i64> %0,
+    <vscale x 2 x i64> %1,
+    i64 %2,
+    <vscale x 2 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 2 x i64> %a
+}
+
+declare <vscale x 4 x i64> @llvm.riscv.vssra.nxv4i64.i64(
+  <vscale x 4 x i64>,
+  i64,
+  i32);
+
+define <vscale x 4 x i64> @intrinsic_vssra_vx_nxv4i64_nxv4i64_i64(<vscale x 4 x i64> %0, i64 %1, i32 %2) nounwind {
+; CHECK-LABEL: intrinsic_vssra_vx_nxv4i64_nxv4i64_i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a2, a2, e64,m4,ta,mu
+; CHECK-NEXT:    vmv.v.x v28, a1
+; CHECK-NEXT:    addi a1, zero, 32
+; CHECK-NEXT:    vsll.vx v28, v28, a1
+; CHECK-NEXT:    vmv.v.x v12, a0
+; CHECK-NEXT:    vsll.vx v12, v12, a1
+; CHECK-NEXT:    vsrl.vx v12, v12, a1
+; CHECK-NEXT:    vor.vv v28, v12, v28
+; CHECK-NEXT:    vssra.vv v8, v8, v28
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i64> @llvm.riscv.vssra.nxv4i64.i64(
+    <vscale x 4 x i64> %0,
+    i64 %1,
+    i32 %2)
+
+  ret <vscale x 4 x i64> %a
+}
+
+declare <vscale x 4 x i64> @llvm.riscv.vssra.mask.nxv4i64.i64(
+  <vscale x 4 x i64>,
+  <vscale x 4 x i64>,
+  i64,
+  <vscale x 4 x i1>,
+  i32);
+
+define <vscale x 4 x i64> @intrinsic_vssra_mask_vx_nxv4i64_nxv4i64_i64(<vscale x 4 x i64> %0, <vscale x 4 x i64> %1, i64 %2, <vscale x 4 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vssra_mask_vx_nxv4i64_nxv4i64_i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a3, a2, e64,m4,ta,mu
+; CHECK-NEXT:    vmv.v.x v28, a1
+; CHECK-NEXT:    addi a1, zero, 32
+; CHECK-NEXT:    vsll.vx v28, v28, a1
+; CHECK-NEXT:    vmv.v.x v16, a0
+; CHECK-NEXT:    vsll.vx v16, v16, a1
+; CHECK-NEXT:    vsrl.vx v16, v16, a1
+; CHECK-NEXT:    vor.vv v28, v16, v28
+; CHECK-NEXT:    vsetvli a0, a2, e64,m4,tu,mu
+; CHECK-NEXT:    vssra.vv v8, v12, v28, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i64> @llvm.riscv.vssra.mask.nxv4i64.i64(
+    <vscale x 4 x i64> %0,
+    <vscale x 4 x i64> %1,
+    i64 %2,
+    <vscale x 4 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 4 x i64> %a
+}
+
+declare <vscale x 8 x i64> @llvm.riscv.vssra.nxv8i64.i64(
+  <vscale x 8 x i64>,
+  i64,
+  i32);
+
+define <vscale x 8 x i64> @intrinsic_vssra_vx_nxv8i64_nxv8i64_i64(<vscale x 8 x i64> %0, i64 %1, i32 %2) nounwind {
+; CHECK-LABEL: intrinsic_vssra_vx_nxv8i64_nxv8i64_i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a2, a2, e64,m8,ta,mu
+; CHECK-NEXT:    vmv.v.x v16, a1
+; CHECK-NEXT:    addi a1, zero, 32
+; CHECK-NEXT:    vsll.vx v16, v16, a1
+; CHECK-NEXT:    vmv.v.x v24, a0
+; CHECK-NEXT:    vsll.vx v24, v24, a1
+; CHECK-NEXT:    vsrl.vx v24, v24, a1
+; CHECK-NEXT:    vor.vv v16, v24, v16
+; CHECK-NEXT:    vssra.vv v8, v8, v16
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i64> @llvm.riscv.vssra.nxv8i64.i64(
+    <vscale x 8 x i64> %0,
+    i64 %1,
+    i32 %2)
+
+  ret <vscale x 8 x i64> %a
+}
+
+declare <vscale x 8 x i64> @llvm.riscv.vssra.mask.nxv8i64.i64(
+  <vscale x 8 x i64>,
+  <vscale x 8 x i64>,
+  i64,
+  <vscale x 8 x i1>,
+  i32);
+
+define <vscale x 8 x i64> @intrinsic_vssra_mask_vx_nxv8i64_nxv8i64_i64(<vscale x 8 x i64> %0, <vscale x 8 x i64> %1, i64 %2, <vscale x 8 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vssra_mask_vx_nxv8i64_nxv8i64_i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    csrrs a3, vlenb, zero
+; CHECK-NEXT:    sub sp, sp, a3
+; CHECK-NEXT:    vs1r.v v0, (sp) # Unknown-size Folded Spill
+; CHECK-NEXT:    vsetvli a3, a2, e64,m8,ta,mu
+; CHECK-NEXT:    vmv.v.x v24, a1
+; CHECK-NEXT:    addi a1, zero, 32
+; CHECK-NEXT:    vsll.vx v0, v24, a1
+; CHECK-NEXT:    vmv.v.x v24, a0
+; CHECK-NEXT:    vsll.vx v24, v24, a1
+; CHECK-NEXT:    vsrl.vx v24, v24, a1
+; CHECK-NEXT:    vor.vv v24, v24, v0
+; CHECK-NEXT:    vsetvli a0, a2, e64,m8,tu,mu
+; CHECK-NEXT:    vl1re8.v v0, (sp) # Unknown-size Folded Reload
+; CHECK-NEXT:    vssra.vv v8, v16, v24, v0.t
+; CHECK-NEXT:    csrrs a0, vlenb, zero
+; CHECK-NEXT:    add sp, sp, a0
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i64> @llvm.riscv.vssra.mask.nxv8i64.i64(
+    <vscale x 8 x i64> %0,
+    <vscale x 8 x i64> %1,
+    i64 %2,
+    <vscale x 8 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 8 x i64> %a
+}
+
 define <vscale x 1 x i8> @intrinsic_vssra_vi_nxv1i8_nxv1i8_i8(<vscale x 1 x i8> %0, i32 %1) nounwind {
 ; CHECK-LABEL: intrinsic_vssra_vi_nxv1i8_nxv1i8_i8:
 ; CHECK:       # %bb.0: # %entry
 ; CHECK-NEXT:    vsetvli a0, a0, e8,mf8,ta,mu
 ; CHECK-NEXT:    vssra.vi v8, v8, 9
 ; CHECK-NEXT:    jalr zero, 0(ra)
 entry:
   %a = call <vscale x 1 x i8> @llvm.riscv.vssra.nxv1i8.i8(
%a = call <vscale x 16 x i32> @llvm.riscv.vssra.mask.nxv16i32.i32(
     <vscale x 16 x i32> %0,
     <vscale x 16 x i32> %1,
     i32 9,
     <vscale x 16 x i1> %2,
     i32 %3)
 
   ret <vscale x 16 x i32> %a
 }
+
+define <vscale x 1 x i64> @intrinsic_vssra_vi_nxv1i64_nxv1i64_i64(<vscale x 1 x i64> %0, i32 %1) nounwind {
+; CHECK-LABEL: intrinsic_vssra_vi_nxv1i64_nxv1i64_i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e64,m1,ta,mu
+; CHECK-NEXT:    vssra.vi v8, v8, 9
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i64> @llvm.riscv.vssra.nxv1i64.i64(
+    <vscale x 1 x i64> %0,
+    i64 9,
+    i32 %1)
+
+  ret <vscale x 1 x i64> %a
+}
+
+define <vscale x 1 x i64> @intrinsic_vssra_mask_vi_nxv1i64_nxv1i64_i64(<vscale x 1 x i64> %0, <vscale x 1 x i64> %1, <vscale x 1 x i1> %2, i32 %3) nounwind {
+; CHECK-LABEL: intrinsic_vssra_mask_vi_nxv1i64_nxv1i64_i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e64,m1,tu,mu
+; CHECK-NEXT:    vssra.vi v8, v9, 9, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i64> @llvm.riscv.vssra.mask.nxv1i64.i64(
+    <vscale x 1 x i64> %0,
+    <vscale x 1 x i64> %1,
+    i64 9,
+    <vscale x 1 x i1> %2,
+    i32 %3)
+
+  ret <vscale x 1 x i64> %a
+}
+
+define <vscale x 2 x i64> @intrinsic_vssra_vi_nxv2i64_nxv2i64_i64(<vscale x 2 x i64> %0, i32 %1) nounwind {
+; CHECK-LABEL: intrinsic_vssra_vi_nxv2i64_nxv2i64_i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e64,m2,ta,mu
+; CHECK-NEXT:    vssra.vi v8, v8, 9
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i64> @llvm.riscv.vssra.nxv2i64.i64(
+    <vscale x 2 x i64> %0,
+    i64 9,
+    i32 %1)
+
+  ret <vscale x 2 x i64> %a
+}
+
+define <vscale x 2 x i64> @intrinsic_vssra_mask_vi_nxv2i64_nxv2i64_i64(<vscale x 2 x i64> %0, <vscale x 2 x i64> %1, <vscale x 2 x i1> %2, i32 %3) nounwind {
+; CHECK-LABEL: intrinsic_vssra_mask_vi_nxv2i64_nxv2i64_i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e64,m2,tu,mu
+; CHECK-NEXT:    vssra.vi v8, v10, 9, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i64> @llvm.riscv.vssra.mask.nxv2i64.i64(
+    <vscale x 2 x i64> %0,
+    <vscale x 2 x i64> %1,
+    i64 9,
+    <vscale x 2 x i1> %2,
+    i32 %3)
+
+  ret <vscale x 2 x i64> %a
+}
+
+define <vscale x 4 x i64> @intrinsic_vssra_vi_nxv4i64_nxv4i64_i64(<vscale x 4 x i64> %0, i32 %1) nounwind {
+; CHECK-LABEL: intrinsic_vssra_vi_nxv4i64_nxv4i64_i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e64,m4,ta,mu
+; CHECK-NEXT:    vssra.vi v8, v8, 9
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i64> @llvm.riscv.vssra.nxv4i64.i64(
+    <vscale x 4 x i64> %0,
+    i64 9,
+    i32 %1)
+
+  ret <vscale x 4 x i64> %a
+}
+
+define <vscale x 4 x i64> @intrinsic_vssra_mask_vi_nxv4i64_nxv4i64_i64(<vscale x 4 x i64> %0, <vscale x 4 x i64> %1, <vscale x 4 x i1> %2, i32 %3) nounwind {
+; CHECK-LABEL: intrinsic_vssra_mask_vi_nxv4i64_nxv4i64_i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e64,m4,tu,mu
+; CHECK-NEXT:    vssra.vi v8, v12, 9, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i64> @llvm.riscv.vssra.mask.nxv4i64.i64(
+    <vscale x 4 x i64> %0,
+    <vscale x 4 x i64> %1,
+    i64 9,
+    <vscale x 4 x i1> %2,
+    i32 %3)
+
+  ret <vscale x 4 x i64> %a
+}
+
+define <vscale x 8 x i64> @intrinsic_vssra_vi_nxv8i64_nxv8i64_i64(<vscale x 8 x i64> %0, i32 %1) nounwind {
+; CHECK-LABEL: intrinsic_vssra_vi_nxv8i64_nxv8i64_i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e64,m8,ta,mu
+; CHECK-NEXT:    vssra.vi v8, v8, 9
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i64> @llvm.riscv.vssra.nxv8i64.i64(
+    <vscale x 8 x i64> %0,
+    i64 9,
+    i32 %1)
+
+  ret <vscale x 8 x i64> %a
+}
+
+define <vscale x 8 x i64> @intrinsic_vssra_mask_vi_nxv8i64_nxv8i64_i64(<vscale x 8 x i64> %0, <vscale x 8 x i64> %1, <vscale x 8 x i1> %2, i32 %3) nounwind {
+; CHECK-LABEL: intrinsic_vssra_mask_vi_nxv8i64_nxv8i64_i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e64,m8,tu,mu
+; CHECK-NEXT:    vssra.vi v8, v16, 9, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i64> @llvm.riscv.vssra.mask.nxv8i64.i64(
+    <vscale x 8 x i64> %0,
+    <vscale x 8 x i64> %1,
+    i64 9,
+    <vscale x 8 x i1> %2,
+    i32 %3)
+
+  ret <vscale x 8 x i64> %a
+}

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

%a = call <vscale x 16 x i32> @llvm.riscv.vssrl.mask.nxv16i32.nxv16i32(
     <vscale x 16 x i32> %1,
     <vscale x 16 x i32> %2,
     <vscale x 16 x i1> %3,
     i32 %4)
 
   ret <vscale x 16 x i32> %a
 }
 
+declare <vscale x 1 x i64> @llvm.riscv.vssrl.nxv1i64.nxv1i64(
+  <vscale x 1 x i64>,
+  <vscale x 1 x i64>,
+  i32);
+
+define <vscale x 1 x i64> @intrinsic_vssrl_vv_nxv1i64_nxv1i64_nxv1i64(<vscale x 1 x i64> %0, <vscale x 1 x i64> %1, i32 %2) nounwind {
+; CHECK-LABEL: intrinsic_vssrl_vv_nxv1i64_nxv1i64_nxv1i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e64,m1,ta,mu
+; CHECK-NEXT:    vssrl.vv v8, v8, v9
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i64> @llvm.riscv.vssrl.nxv1i64.nxv1i64(
+    <vscale x 1 x i64> %0,
+    <vscale x 1 x i64> %1,
+    i32 %2)
+
+  ret <vscale x 1 x i64> %a
+}
+
+declare <vscale x 1 x i64> @llvm.riscv.vssrl.mask.nxv1i64.nxv1i64(
+  <vscale x 1 x i64>,
+  <vscale x 1 x i64>,
+  <vscale x 1 x i64>,
+  <vscale x 1 x i1>,
+  i32);
+
+define <vscale x 1 x i64> @intrinsic_vssrl_mask_vv_nxv1i64_nxv1i64_nxv1i64(<vscale x 1 x i64> %0, <vscale x 1 x i64> %1, <vscale x 1 x i64> %2, <vscale x 1 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vssrl_mask_vv_nxv1i64_nxv1i64_nxv1i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e64,m1,tu,mu
+; CHECK-NEXT:    vssrl.vv v8, v9, v10, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i64> @llvm.riscv.vssrl.mask.nxv1i64.nxv1i64(
+    <vscale x 1 x i64> %0,
+    <vscale x 1 x i64> %1,
+    <vscale x 1 x i64> %2,
+    <vscale x 1 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 1 x i64> %a
+}
+
+declare <vscale x 2 x i64> @llvm.riscv.vssrl.nxv2i64.nxv2i64(
+  <vscale x 2 x i64>,
+  <vscale x 2 x i64>,
+  i32);
+
+define <vscale x 2 x i64> @intrinsic_vssrl_vv_nxv2i64_nxv2i64_nxv2i64(<vscale x 2 x i64> %0, <vscale x 2 x i64> %1, i32 %2) nounwind {
+; CHECK-LABEL: intrinsic_vssrl_vv_nxv2i64_nxv2i64_nxv2i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e64,m2,ta,mu
+; CHECK-NEXT:    vssrl.vv v8, v8, v10
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i64> @llvm.riscv.vssrl.nxv2i64.nxv2i64(
+    <vscale x 2 x i64> %0,
+    <vscale x 2 x i64> %1,
+    i32 %2)
+
+  ret <vscale x 2 x i64> %a
+}
+
+declare <vscale x 2 x i64> @llvm.riscv.vssrl.mask.nxv2i64.nxv2i64(
+  <vscale x 2 x i64>,
+  <vscale x 2 x i64>,
+  <vscale x 2 x i64>,
+  <vscale x 2 x i1>,
+  i32);
+
+define <vscale x 2 x i64> @intrinsic_vssrl_mask_vv_nxv2i64_nxv2i64_nxv2i64(<vscale x 2 x i64> %0, <vscale x 2 x i64> %1, <vscale x 2 x i64> %2, <vscale x 2 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vssrl_mask_vv_nxv2i64_nxv2i64_nxv2i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e64,m2,tu,mu
+; CHECK-NEXT:    vssrl.vv v8, v10, v12, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i64> @llvm.riscv.vssrl.mask.nxv2i64.nxv2i64(
+    <vscale x 2 x i64> %0,
+    <vscale x 2 x i64> %1,
+    <vscale x 2 x i64> %2,
+    <vscale x 2 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 2 x i64> %a
+}
+
+declare <vscale x 4 x i64> @llvm.riscv.vssrl.nxv4i64.nxv4i64(
+  <vscale x 4 x i64>,
+  <vscale x 4 x i64>,
+  i32);
+
+define <vscale x 4 x i64> @intrinsic_vssrl_vv_nxv4i64_nxv4i64_nxv4i64(<vscale x 4 x i64> %0, <vscale x 4 x i64> %1, i32 %2) nounwind {
+; CHECK-LABEL: intrinsic_vssrl_vv_nxv4i64_nxv4i64_nxv4i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e64,m4,ta,mu
+; CHECK-NEXT:    vssrl.vv v8, v8, v12
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i64> @llvm.riscv.vssrl.nxv4i64.nxv4i64(
+    <vscale x 4 x i64> %0,
+    <vscale x 4 x i64> %1,
+    i32 %2)
+
+  ret <vscale x 4 x i64> %a
+}
+
+declare <vscale x 4 x i64> @llvm.riscv.vssrl.mask.nxv4i64.nxv4i64(
+  <vscale x 4 x i64>,
+  <vscale x 4 x i64>,
+  <vscale x 4 x i64>,
+  <vscale x 4 x i1>,
+  i32);
+
+define <vscale x 4 x i64> @intrinsic_vssrl_mask_vv_nxv4i64_nxv4i64_nxv4i64(<vscale x 4 x i64> %0, <vscale x 4 x i64> %1, <vscale x 4 x i64> %2, <vscale x 4 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vssrl_mask_vv_nxv4i64_nxv4i64_nxv4i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e64,m4,tu,mu
+; CHECK-NEXT:    vssrl.vv v8, v12, v16, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i64> @llvm.riscv.vssrl.mask.nxv4i64.nxv4i64(
+    <vscale x 4 x i64> %0,
+    <vscale x 4 x i64> %1,
+    <vscale x 4 x i64> %2,
+    <vscale x 4 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 4 x i64> %a
+}
+
+declare <vscale x 8 x i64> @llvm.riscv.vssrl.nxv8i64.nxv8i64(
+  <vscale x 8 x i64>,
+  <vscale x 8 x i64>,
+  i32);
+
+define <vscale x 8 x i64> @intrinsic_vssrl_vv_nxv8i64_nxv8i64_nxv8i64(<vscale x 8 x i64> %0, <vscale x 8 x i64> %1, i32 %2) nounwind {
+; CHECK-LABEL: intrinsic_vssrl_vv_nxv8i64_nxv8i64_nxv8i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e64,m8,ta,mu
+; CHECK-NEXT:    vssrl.vv v8, v8, v16
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i64> @llvm.riscv.vssrl.nxv8i64.nxv8i64(
+    <vscale x 8 x i64> %0,
+    <vscale x 8 x i64> %1,
+    i32 %2)
+
+  ret <vscale x 8 x i64> %a
+}
+
+declare <vscale x 8 x i64> @llvm.riscv.vssrl.mask.nxv8i64.nxv8i64(
+  <vscale x 8 x i64>,
+  <vscale x 8 x i64>,
+  <vscale x 8 x i64>,
+  <vscale x 8 x i1>,
+  i32);
+
+define <vscale x 8 x i64> @intrinsic_vssrl_mask_vv_nxv8i64_nxv8i64_nxv8i64(<vscale x 8 x i64> %0, <vscale x 8 x i64> %1, <vscale x 8 x i64> %2, <vscale x 8 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vssrl_mask_vv_nxv8i64_nxv8i64_nxv8i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vl8re64.v v24, (a0)
+; CHECK-NEXT:    vsetvli a0, a1, e64,m8,tu,mu
+; CHECK-NEXT:    vssrl.vv v8, v16, v24, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i64> @llvm.riscv.vssrl.mask.nxv8i64.nxv8i64(
+    <vscale x 8 x i64> %0,
+    <vscale x 8 x i64> %1,
+    <vscale x 8 x i64> %2,
+    <vscale x 8 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 8 x i64> %a
+}
+
 declare <vscale x 1 x i8> @llvm.riscv.vssrl.nxv1i8.i8(
   <vscale x 1 x i8>,
   i8,
   i32);
 
 define <vscale x 1 x i8> @intrinsic_vssrl_vx_nxv1i8_nxv1i8_i8(<vscale x 1 x i8> %0, i8 %1, i32 %2) nounwind {
 ; CHECK-LABEL: intrinsic_vssrl_vx_nxv1i8_nxv1i8_i8:
 ; CHECK:       # %bb.0: # %entry
%a = call <vscale x 16 x i32> @llvm.riscv.vssrl.mask.nxv16i32.i32(
     <vscale x 16 x i32> %1,
     i32 %2,
     <vscale x 16 x i1> %3,
     i32 %4)
 
   ret <vscale x 16 x i32> %a
 }
 
+declare <vscale x 1 x i64> @llvm.riscv.vssrl.nxv1i64.i64(
+  <vscale x 1 x i64>,
+  i64,
+  i32);
+
+define <vscale x 1 x i64> @intrinsic_vssrl_vx_nxv1i64_nxv1i64_i64(<vscale x 1 x i64> %0, i64 %1, i32 %2) nounwind {
+; CHECK-LABEL: intrinsic_vssrl_vx_nxv1i64_nxv1i64_i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a2, a2, e64,m1,ta,mu
+; CHECK-NEXT:    vmv.v.x v25, a1
+; CHECK-NEXT:    addi a1, zero, 32
+; CHECK-NEXT:    vsll.vx v25, v25, a1
+; CHECK-NEXT:    vmv.v.x v26, a0
+; CHECK-NEXT:    vsll.vx v26, v26, a1
+; CHECK-NEXT:    vsrl.vx v26, v26, a1
+; CHECK-NEXT:    vor.vv v25, v26, v25
+; CHECK-NEXT:    vssrl.vv v8, v8, v25
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i64> @llvm.riscv.vssrl.nxv1i64.i64(
+    <vscale x 1 x i64> %0,
+    i64 %1,
+    i32 %2)
+
+  ret <vscale x 1 x i64> %a
+}
+
+declare <vscale x 1 x i64> @llvm.riscv.vssrl.mask.nxv1i64.i64(
+  <vscale x 1 x i64>,
+  <vscale x 1 x i64>,
+  i64,
+  <vscale x 1 x i1>,
+  i32);
+
+define <vscale x 1 x i64> @intrinsic_vssrl_mask_vx_nxv1i64_nxv1i64_i64(<vscale x 1 x i64> %0, <vscale x 1 x i64> %1, i64 %2, <vscale x 1 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vssrl_mask_vx_nxv1i64_nxv1i64_i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a3, a2, e64,m1,ta,mu
+; CHECK-NEXT:    vmv.v.x v25, a1
+; CHECK-NEXT:    addi a1, zero, 32
+; CHECK-NEXT:    vsll.vx v25, v25, a1
+; CHECK-NEXT:    vmv.v.x v26, a0
+; CHECK-NEXT:    vsll.vx v26, v26, a1
+; CHECK-NEXT:    vsrl.vx v26, v26, a1
+; CHECK-NEXT:    vor.vv v25, v26, v25
+; CHECK-NEXT:    vsetvli a0, a2, e64,m1,tu,mu
+; CHECK-NEXT:    vssrl.vv v8, v9, v25, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i64> @llvm.riscv.vssrl.mask.nxv1i64.i64(
+    <vscale x 1 x i64> %0,
+    <vscale x 1 x i64> %1,
+    i64 %2,
+    <vscale x 1 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 1 x i64> %a
+}
+
+declare <vscale x 2 x i64> @llvm.riscv.vssrl.nxv2i64.i64(
+  <vscale x 2 x i64>,
+  i64,
+  i32);
+
+define <vscale x 2 x i64> @intrinsic_vssrl_vx_nxv2i64_nxv2i64_i64(<vscale x 2 x i64> %0, i64 %1, i32 %2) nounwind {
+; CHECK-LABEL: intrinsic_vssrl_vx_nxv2i64_nxv2i64_i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a2, a2, e64,m2,ta,mu
+; CHECK-NEXT:    vmv.v.x v26, a1
+; CHECK-NEXT:    addi a1, zero, 32
+; CHECK-NEXT:    vsll.vx v26, v26, a1
+; CHECK-NEXT:    vmv.v.x v28, a0
+; CHECK-NEXT:    vsll.vx v28, v28, a1
+; CHECK-NEXT:    vsrl.vx v28, v28, a1
+; CHECK-NEXT:    vor.vv v26, v28, v26
+; CHECK-NEXT:    vssrl.vv v8, v8, v26
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i64> @llvm.riscv.vssrl.nxv2i64.i64(
+    <vscale x 2 x i64> %0,
+    i64 %1,
+    i32 %2)
+
+  ret <vscale x 2 x i64> %a
+}
+
+declare <vscale x 2 x i64> @llvm.riscv.vssrl.mask.nxv2i64.i64(
+  <vscale x 2 x i64>,
+  <vscale x 2 x i64>,
+  i64,
+  <vscale x 2 x i1>,
+  i32);
+
+define <vscale x 2 x i64> @intrinsic_vssrl_mask_vx_nxv2i64_nxv2i64_i64(<vscale x 2 x i64> %0, <vscale x 2 x i64> %1, i64 %2, <vscale x 2 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vssrl_mask_vx_nxv2i64_nxv2i64_i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a3, a2, e64,m2,ta,mu
+; CHECK-NEXT:    vmv.v.x v26, a1
+; CHECK-NEXT:    addi a1, zero, 32
+; CHECK-NEXT:    vsll.vx v26, v26, a1
+; CHECK-NEXT:    vmv.v.x v28, a0
+; CHECK-NEXT:    vsll.vx v28, v28, a1
+; CHECK-NEXT:    vsrl.vx v28, v28, a1
+; CHECK-NEXT:    vor.vv v26, v28, v26
+; CHECK-NEXT:    vsetvli a0, a2, e64,m2,tu,mu
+; CHECK-NEXT:    vssrl.vv v8, v10, v26, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i64> @llvm.riscv.vssrl.mask.nxv2i64.i64(
+    <vscale x 2 x i64> %0,
+    <vscale x 2 x i64> %1,
+    i64 %2,
+    <vscale x 2 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 2 x i64> %a
+}
+
+declare <vscale x 4 x i64> @llvm.riscv.vssrl.nxv4i64.i64(
+  <vscale x 4 x i64>,
+  i64,
+  i32);
+
+define <vscale x 4 x i64> @intrinsic_vssrl_vx_nxv4i64_nxv4i64_i64(<vscale x 4 x i64> %0, i64 %1, i32 %2) nounwind {
+; CHECK-LABEL: intrinsic_vssrl_vx_nxv4i64_nxv4i64_i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a2, a2, e64,m4,ta,mu
+; CHECK-NEXT:    vmv.v.x v28, a1
+; CHECK-NEXT:    addi a1, zero, 32
+; CHECK-NEXT:    vsll.vx v28, v28, a1
+; CHECK-NEXT:    vmv.v.x v12, a0
+; CHECK-NEXT:    vsll.vx v12, v12, a1
+; CHECK-NEXT:    vsrl.vx v12, v12, a1
+; CHECK-NEXT:    vor.vv v28, v12, v28
+; CHECK-NEXT:    vssrl.vv v8, v8, v28
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i64> @llvm.riscv.vssrl.nxv4i64.i64(
+    <vscale x 4 x i64> %0,
+    i64 %1,
+    i32 %2)
+
+  ret <vscale x 4 x i64> %a
+}
+
+declare <vscale x 4 x i64> @llvm.riscv.vssrl.mask.nxv4i64.i64(
+  <vscale x 4 x i64>,
+  <vscale x 4 x i64>,
+  i64,
+  <vscale x 4 x i1>,
+  i32);
+
+define <vscale x 4 x i64> @intrinsic_vssrl_mask_vx_nxv4i64_nxv4i64_i64(<vscale x 4 x i64> %0, <vscale x 4 x i64> %1, i64 %2, <vscale x 4 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vssrl_mask_vx_nxv4i64_nxv4i64_i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a3, a2, e64,m4,ta,mu
+; CHECK-NEXT:    vmv.v.x v28, a1
+; CHECK-NEXT:    addi a1, zero, 32
+; CHECK-NEXT:    vsll.vx v28, v28, a1
+; CHECK-NEXT:    vmv.v.x v16, a0
+; CHECK-NEXT:    vsll.vx v16, v16, a1
+; CHECK-NEXT:    vsrl.vx v16, v16, a1
+; CHECK-NEXT:    vor.vv v28, v16, v28
+; CHECK-NEXT:    vsetvli a0, a2, e64,m4,tu,mu
+; CHECK-NEXT:    vssrl.vv v8, v12, v28, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i64> @llvm.riscv.vssrl.mask.nxv4i64.i64(
+    <vscale x 4 x i64> %0,
+    <vscale x 4 x i64> %1,
+    i64 %2,
+    <vscale x 4 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 4 x i64> %a
+}
+
+declare <vscale x 8 x i64> @llvm.riscv.vssrl.nxv8i64.i64(
+  <vscale x 8 x i64>,
+  i64,
+  i32);
+
+define <vscale x 8 x i64> @intrinsic_vssrl_vx_nxv8i64_nxv8i64_i64(<vscale x 8 x i64> %0, i64 %1, i32 %2) nounwind {
+; CHECK-LABEL: intrinsic_vssrl_vx_nxv8i64_nxv8i64_i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a2, a2, e64,m8,ta,mu
+; CHECK-NEXT:    vmv.v.x v16, a1
+; CHECK-NEXT:    addi a1, zero, 32
+; CHECK-NEXT:    vsll.vx v16, v16, a1
+; CHECK-NEXT:    vmv.v.x v24, a0
+; CHECK-NEXT:    vsll.vx v24, v24, a1
+; CHECK-NEXT:    vsrl.vx v24, v24, a1
+; CHECK-NEXT:    vor.vv v16, v24, v16
+; CHECK-NEXT:    vssrl.vv v8, v8, v16
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i64> @llvm.riscv.vssrl.nxv8i64.i64(
+    <vscale x 8 x i64> %0,
+    i64 %1,
+    i32 %2)
+
+  ret <vscale x 8 x i64> %a
+}
+
+declare <vscale x 8 x i64> @llvm.riscv.vssrl.mask.nxv8i64.i64(
+  <vscale x 8 x i64>,
+  <vscale x 8 x i64>,
+  i64,
+  <vscale x 8 x i1>,
+  i32);
+
+define <vscale x 8 x i64> @intrinsic_vssrl_mask_vx_nxv8i64_nxv8i64_i64(<vscale x 8 x i64> %0, <vscale x 8 x i64> %1, i64 %2, <vscale x 8 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vssrl_mask_vx_nxv8i64_nxv8i64_i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    csrrs a3, vlenb, zero
+; CHECK-NEXT:    sub sp, sp, a3
+; CHECK-NEXT:    vs1r.v v0, (sp) # Unknown-size Folded Spill
+; CHECK-NEXT:    vsetvli a3, a2, e64,m8,ta,mu
+; CHECK-NEXT:    vmv.v.x v24, a1
+; CHECK-NEXT:    addi a1, zero, 32
+; CHECK-NEXT:    vsll.vx v0, v24, a1
+; CHECK-NEXT:    vmv.v.x v24, a0
+; CHECK-NEXT:    vsll.vx v24, v24, a1
+; CHECK-NEXT:    vsrl.vx v24, v24, a1
+; CHECK-NEXT:    vor.vv v24, v24, v0
+; CHECK-NEXT:    vsetvli a0, a2, e64,m8,tu,mu
+; CHECK-NEXT:    vl1re8.v v0, (sp) # Unknown-size Folded Reload
+; CHECK-NEXT:    vssrl.vv v8, v16, v24, v0.t
+; CHECK-NEXT:    csrrs a0, vlenb, zero
+; CHECK-NEXT:    add sp, sp, a0
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i64> @llvm.riscv.vssrl.mask.nxv8i64.i64(
+    <vscale x 8 x i64> %0,
+    <vscale x 8 x i64> %1,
+    i64 %2,
+    <vscale x 8 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 8 x i64> %a
+}
+
 define <vscale x 1 x i8> @intrinsic_vssrl_vi_nxv1i8_nxv1i8_i8(<vscale x 1 x i8> %0, i32 %1) nounwind {
 ; CHECK-LABEL: intrinsic_vssrl_vi_nxv1i8_nxv1i8_i8:
 ; CHECK:       # %bb.0: # %entry
 ; CHECK-NEXT:    vsetvli a0, a0, e8,mf8,ta,mu
 ; CHECK-NEXT:    vssrl.vi v8, v8, 9
 ; CHECK-NEXT:    jalr zero, 0(ra)
 entry:
   %a = call <vscale x 1 x i8> @llvm.riscv.vssrl.nxv1i8.i8(
%a = call <vscale x 16 x i32> @llvm.riscv.vssrl.mask.nxv16i32.i32(
     <vscale x 16 x i32> %0,
     <vscale x 16 x i32> %1,
     i32 9,
     <vscale x 16 x i1> %2,
     i32 %3)
 
   ret <vscale x 16 x i32> %a
 }
+
+define <vscale x 1 x i64> @intrinsic_vssrl_vi_nxv1i64_nxv1i64_i64(<vscale x 1 x i64> %0, i32 %1) nounwind {
+; CHECK-LABEL: intrinsic_vssrl_vi_nxv1i64_nxv1i64_i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e64,m1,ta,mu
+; CHECK-NEXT:    vssrl.vi v8, v8, 9
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i64> @llvm.riscv.vssrl.nxv1i64.i64(
+    <vscale x 1 x i64> %0,
+    i64 9,
+    i32 %1)
+
+  ret <vscale x 1 x i64> %a
+}
+
+define <vscale x 1 x i64> @intrinsic_vssrl_mask_vi_nxv1i64_nxv1i64_i64(<vscale x 1 x i64> %0, <vscale x 1 x i64> %1, <vscale x 1 x i1> %2, i32 %3) nounwind {
+; CHECK-LABEL: intrinsic_vssrl_mask_vi_nxv1i64_nxv1i64_i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e64,m1,tu,mu
+; CHECK-NEXT:    vssrl.vi v8, v9, 9, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i64> @llvm.riscv.vssrl.mask.nxv1i64.i64(
+    <vscale x 1 x i64> %0,
+    <vscale x 1 x i64> %1,
+    i64 9,
+    <vscale x 1 x i1> %2,
+    i32 %3)
+
+  ret <vscale x 1 x i64> %a
+}
+
+define <vscale x 2 x i64> @intrinsic_vssrl_vi_nxv2i64_nxv2i64_i64(<vscale x 2 x i64> %0, i32 %1) nounwind {
+; CHECK-LABEL: intrinsic_vssrl_vi_nxv2i64_nxv2i64_i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e64,m2,ta,mu
+; CHECK-NEXT:    vssrl.vi v8, v8, 9
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i64> @llvm.riscv.vssrl.nxv2i64.i64(
+    <vscale x 2 x i64> %0,
+    i64 9,
+    i32 %1)
+
+  ret <vscale x 2 x i64> %a
+}
+
+define <vscale x 2 x i64> @intrinsic_vssrl_mask_vi_nxv2i64_nxv2i64_i64(<vscale x 2 x i64> %0, <vscale x 2 x i64> %1, <vscale x 2 x i1> %2, i32 %3) nounwind {
+; CHECK-LABEL: intrinsic_vssrl_mask_vi_nxv2i64_nxv2i64_i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e64,m2,tu,mu
+; CHECK-NEXT:    vssrl.vi v8, v10, 9, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i64> @llvm.riscv.vssrl.mask.nxv2i64.i64(
+    <vscale x 2 x i64> %0,
+    <vscale x 2 x i64> %1,
+    i64 9,
+    <vscale x 2 x i1> %2,
+    i32 %3)
+
+  ret <vscale x 2 x i64> %a
+}
+
+define <vscale x 4 x i64> @intrinsic_vssrl_vi_nxv4i64_nxv4i64_i64(<vscale x 4 x i64> %0, i32 %1) nounwind {
+; CHECK-LABEL: intrinsic_vssrl_vi_nxv4i64_nxv4i64_i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e64,m4,ta,mu
+; CHECK-NEXT:    vssrl.vi v8, v8, 9
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i64> @llvm.riscv.vssrl.nxv4i64.i64(
+    <vscale x 4 x i64> %0,
+    i64 9,
+    i32 %1)
+
+  ret <vscale x 4 x i64> %a
+}
+
+define <vscale x 4 x i64> @intrinsic_vssrl_mask_vi_nxv4i64_nxv4i64_i64(<vscale x 4 x i64> %0, <vscale x 4 x i64> %1, <vscale x 4 x i1> %2, i32 %3) nounwind {
+; CHECK-LABEL: intrinsic_vssrl_mask_vi_nxv4i64_nxv4i64_i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e64,m4,tu,mu
+; CHECK-NEXT:    vssrl.vi v8, v12, 9, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i64> @llvm.riscv.vssrl.mask.nxv4i64.i64(
+    <vscale x 4 x i64> %0,
+    <vscale x 4 x i64> %1,
+    i64 9,
+    <vscale x 4 x i1> %2,
+    i32 %3)
+
+  ret <vscale x 4 x i64> %a
+}
+
+define <vscale x 8 x i64> @intrinsic_vssrl_vi_nxv8i64_nxv8i64_i64(<vscale x 8 x i64> %0, i32 %1) nounwind {
+; CHECK-LABEL: intrinsic_vssrl_vi_nxv8i64_nxv8i64_i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e64,m8,ta,mu
+; CHECK-NEXT:    vssrl.vi v8, v8, 9
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i64> @llvm.riscv.vssrl.nxv8i64.i64(
+    <vscale x 8 x i64> %0,
+    i64 9,
+    i32 %1)
+
+  ret <vscale x 8 x i64> %a
+}
+
+define <vscale x 8 x i64> @intrinsic_vssrl_mask_vi_nxv8i64_nxv8i64_i64(<vscale x 8 x i64> %0, <vscale x 8 x i64> %1, <vscale x 8 x i1> %2, i32 %3) nounwind {
+; CHECK-LABEL: intrinsic_vssrl_mask_vi_nxv8i64_nxv8i64_i64:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e64,m8,tu,mu
+; CHECK-NEXT:    vssrl.vi v8, v16, 9, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i64> @llvm.riscv.vssrl.mask.nxv8i64.i64(
+    <vscale x 8 x i64> %0,
+    <vscale x 8 x i64> %1,
+    i64 9,
+    <vscale x 8 x i1> %2,
+    i32 %3)
+
+  ret <vscale x 8 x i64> %a
+}

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

%a = call <vscale x 16 x i32> @llvm.riscv.vwadd.mask.nxv16i32.nxv16i16.nxv16i16(
     <vscale x 16 x i16> %1,
     <vscale x 16 x i16> %2,
     <vscale x 16 x i1> %3,
     i32 %4)
 
   ret <vscale x 16 x i32> %a
 }
 
+declare <vscale x 1 x i64> @llvm.riscv.vwadd.nxv1i64.nxv1i32.nxv1i32(
+  <vscale x 1 x i32>,
+  <vscale x 1 x i32>,
+  i32);
+
+define <vscale x 1 x i64> @intrinsic_vwadd_vv_nxv1i64_nxv1i32_nxv1i32(<vscale x 1 x i32> %0, <vscale x 1 x i32> %1, i32 %2) nounwind {
+; CHECK-LABEL: intrinsic_vwadd_vv_nxv1i64_nxv1i32_nxv1i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e32,mf2,ta,mu
+; CHECK-NEXT:    vwadd.vv v25, v8, v9
+; CHECK-NEXT:    vmv1r.v v8, v25
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i64> @llvm.riscv.vwadd.nxv1i64.nxv1i32.nxv1i32(
+    <vscale x 1 x i32> %0,
+    <vscale x 1 x i32> %1,
+    i32 %2)
+
+  ret <vscale x 1 x i64> %a
+}
+
+declare <vscale x 1 x i64> @llvm.riscv.vwadd.mask.nxv1i64.nxv1i32.nxv1i32(
+  <vscale x 1 x i64>,
+  <vscale x 1 x i32>,
+  <vscale x 1 x i32>,
+  <vscale x 1 x i1>,
+  i32);
+
+define <vscale x 1 x i64> @intrinsic_vwadd_mask_vv_nxv1i64_nxv1i32_nxv1i32(<vscale x 1 x i64> %0, <vscale x 1 x i32> %1, <vscale x 1 x i32> %2, <vscale x 1 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwadd_mask_vv_nxv1i64_nxv1i32_nxv1i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e32,mf2,tu,mu
+; CHECK-NEXT:    vwadd.vv v8, v9, v10, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i64> @llvm.riscv.vwadd.mask.nxv1i64.nxv1i32.nxv1i32(
+    <vscale x 1 x i64> %0,
+    <vscale x 1 x i32> %1,
+    <vscale x 1 x i32> %2,
+    <vscale x 1 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 1 x i64> %a
+}
+
+declare <vscale x 2 x i64> @llvm.riscv.vwadd.nxv2i64.nxv2i32.nxv2i32(
+  <vscale x 2 x i32>,
+  <vscale x 2 x i32>,
+  i32);
+
+define <vscale x 2 x i64> @intrinsic_vwadd_vv_nxv2i64_nxv2i32_nxv2i32(<vscale x 2 x i32> %0, <vscale x 2 x i32> %1, i32 %2) nounwind {
+; CHECK-LABEL: intrinsic_vwadd_vv_nxv2i64_nxv2i32_nxv2i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e32,m1,ta,mu
+; CHECK-NEXT:    vwadd.vv v26, v8, v9
+; CHECK-NEXT:    vmv2r.v v8, v26
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i64> @llvm.riscv.vwadd.nxv2i64.nxv2i32.nxv2i32(
+    <vscale x 2 x i32> %0,
+    <vscale x 2 x i32> %1,
+    i32 %2)
+
+  ret <vscale x 2 x i64> %a
+}
+
+declare <vscale x 2 x i64> @llvm.riscv.vwadd.mask.nxv2i64.nxv2i32.nxv2i32(
+  <vscale x 2 x i64>,
+  <vscale x 2 x i32>,
+  <vscale x 2 x i32>,
+  <vscale x 2 x i1>,
+  i32);
+
+define <vscale x 2 x i64> @intrinsic_vwadd_mask_vv_nxv2i64_nxv2i32_nxv2i32(<vscale x 2 x i64> %0, <vscale x 2 x i32> %1, <vscale x 2 x i32> %2, <vscale x 2 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwadd_mask_vv_nxv2i64_nxv2i32_nxv2i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e32,m1,tu,mu
+; CHECK-NEXT:    vwadd.vv v8, v10, v11, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i64> @llvm.riscv.vwadd.mask.nxv2i64.nxv2i32.nxv2i32(
+    <vscale x 2 x i64> %0,
+    <vscale x 2 x i32> %1,
+    <vscale x 2 x i32> %2,
+    <vscale x 2 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 2 x i64> %a
+}
+
+declare <vscale x 4 x i64> @llvm.riscv.vwadd.nxv4i64.nxv4i32.nxv4i32(
+  <vscale x 4 x i32>,
+  <vscale x 4 x i32>,
+  i32);
+
+define <vscale x 4 x i64> @intrinsic_vwadd_vv_nxv4i64_nxv4i32_nxv4i32(<vscale x 4 x i32> %0, <vscale x 4 x i32> %1, i32 %2) nounwind {
+; CHECK-LABEL: intrinsic_vwadd_vv_nxv4i64_nxv4i32_nxv4i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e32,m2,ta,mu
+; CHECK-NEXT:    vwadd.vv v28, v8, v10
+; CHECK-NEXT:    vmv4r.v v8, v28
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i64> @llvm.riscv.vwadd.nxv4i64.nxv4i32.nxv4i32(
+    <vscale x 4 x i32> %0,
+    <vscale x 4 x i32> %1,
+    i32 %2)
+
+  ret <vscale x 4 x i64> %a
+}
+
+declare <vscale x 4 x i64> @llvm.riscv.vwadd.mask.nxv4i64.nxv4i32.nxv4i32(
+  <vscale x 4 x i64>,
+  <vscale x 4 x i32>,
+  <vscale x 4 x i32>,
+  <vscale x 4 x i1>,
+  i32);
+
+define <vscale x 4 x i64> @intrinsic_vwadd_mask_vv_nxv4i64_nxv4i32_nxv4i32(<vscale x 4 x i64> %0, <vscale x 4 x i32> %1, <vscale x 4 x i32> %2, <vscale x 4 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwadd_mask_vv_nxv4i64_nxv4i32_nxv4i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e32,m2,tu,mu
+; CHECK-NEXT:    vwadd.vv v8, v12, v14, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i64> @llvm.riscv.vwadd.mask.nxv4i64.nxv4i32.nxv4i32(
+    <vscale x 4 x i64> %0,
+    <vscale x 4 x i32> %1,
+    <vscale x 4 x i32> %2,
+    <vscale x 4 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 4 x i64> %a
+}
+
+declare <vscale x 8 x i64> @llvm.riscv.vwadd.nxv8i64.nxv8i32.nxv8i32(
+  <vscale x 8 x i32>,
+  <vscale x 8 x i32>,
+  i32);
+
+define <vscale x 8 x i64> @intrinsic_vwadd_vv_nxv8i64_nxv8i32_nxv8i32(<vscale x 8 x i32> %0, <vscale x 8 x i32> %1, i32 %2) nounwind {
+; CHECK-LABEL: intrinsic_vwadd_vv_nxv8i64_nxv8i32_nxv8i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e32,m4,ta,mu
+; CHECK-NEXT:    vwadd.vv v16, v8, v12
+; CHECK-NEXT:    vmv8r.v v8, v16
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i64> @llvm.riscv.vwadd.nxv8i64.nxv8i32.nxv8i32(
+    <vscale x 8 x i32> %0,
+    <vscale x 8 x i32> %1,
+    i32 %2)
+
+  ret <vscale x 8 x i64> %a
+}
+
+declare <vscale x 8 x i64> @llvm.riscv.vwadd.mask.nxv8i64.nxv8i32.nxv8i32(
+  <vscale x 8 x i64>,
+  <vscale x 8 x i32>,
+  <vscale x 8 x i32>,
+  <vscale x 8 x i1>,
+  i32);
+
+define <vscale x 8 x i64> @intrinsic_vwadd_mask_vv_nxv8i64_nxv8i32_nxv8i32(<vscale x 8 x i64> %0, <vscale x 8 x i32> %1, <vscale x 8 x i32> %2, <vscale x 8 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwadd_mask_vv_nxv8i64_nxv8i32_nxv8i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e32,m4,tu,mu
+; CHECK-NEXT:    vwadd.vv v8, v16, v20, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i64> @llvm.riscv.vwadd.mask.nxv8i64.nxv8i32.nxv8i32(
+    <vscale x 8 x i64> %0,
+    <vscale x 8 x i32> %1,
+    <vscale x 8 x i32> %2,
+    <vscale x 8 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 8 x i64> %a
+}
+
 declare <vscale x 1 x i16> @llvm.riscv.vwadd.nxv1i16.nxv1i8.i8(
   <vscale x 1 x i8>,
   i8,
   i32);
 
 define <vscale x 1 x i16> @intrinsic_vwadd_vx_nxv1i16_nxv1i8_i8(<vscale x 1 x i8> %0, i8 %1, i32 %2) nounwind {
 ; CHECK-LABEL: intrinsic_vwadd_vx_nxv1i16_nxv1i8_i8:
 ; CHECK:       # %bb.0: # %entry
%a = call <vscale x 16 x i32> @llvm.riscv.vwadd.mask.nxv16i32.nxv16i16.i16(
     <vscale x 16 x i32> %0,
     <vscale x 16 x i16> %1,
     i16 %2,
     <vscale x 16 x i1> %3,
     i32 %4)
 
   ret <vscale x 16 x i32> %a
 }
+
+declare <vscale x 1 x i64> @llvm.riscv.vwadd.nxv1i64.nxv1i32.i32(
+  <vscale x 1 x i32>,
+  i32,
+  i32);
+
+define <vscale x 1 x i64> @intrinsic_vwadd_vx_nxv1i64_nxv1i32_i32(<vscale x 1 x i32> %0, i32 %1, i32 %2) nounwind {
+; CHECK-LABEL: intrinsic_vwadd_vx_nxv1i64_nxv1i32_i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e32,mf2,ta,mu
+; CHECK-NEXT:    vwadd.vx v25, v8, a0
+; CHECK-NEXT:    vmv1r.v v8, v25
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i64> @llvm.riscv.vwadd.nxv1i64.nxv1i32.i32(
+    <vscale x 1 x i32> %0,
+    i32 %1,
+    i32 %2)
+
+  ret <vscale x 1 x i64> %a
+}
+
+declare <vscale x 1 x i64> @llvm.riscv.vwadd.mask.nxv1i64.nxv1i32.i32(
+  <vscale x 1 x i64>,
+  <vscale x 1 x i32>,
+  i32,
+  <vscale x 1 x i1>,
+  i32);
+
+define <vscale x 1 x i64> @intrinsic_vwadd_mask_vx_nxv1i64_nxv1i32_i32(<vscale x 1 x i64> %0, <vscale x 1 x i32> %1, i32 %2, <vscale x 1 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwadd_mask_vx_nxv1i64_nxv1i32_i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e32,mf2,tu,mu
+; CHECK-NEXT:    vwadd.vx v8, v9, a0, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i64> @llvm.riscv.vwadd.mask.nxv1i64.nxv1i32.i32(
+    <vscale x 1 x i64> %0,
+    <vscale x 1 x i32> %1,
+    i32 %2,
+    <vscale x 1 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 1 x i64> %a
+}
+
+declare <vscale x 2 x i64> @llvm.riscv.vwadd.nxv2i64.nxv2i32.i32(
+  <vscale x 2 x i32>,
+  i32,
+  i32);
+
+define <vscale x 2 x i64> @intrinsic_vwadd_vx_nxv2i64_nxv2i32_i32(<vscale x 2 x i32> %0, i32 %1, i32 %2) nounwind {
+; CHECK-LABEL: intrinsic_vwadd_vx_nxv2i64_nxv2i32_i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e32,m1,ta,mu
+; CHECK-NEXT:    vwadd.vx v26, v8, a0
+; CHECK-NEXT:    vmv2r.v v8, v26
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i64> @llvm.riscv.vwadd.nxv2i64.nxv2i32.i32(
+    <vscale x 2 x i32> %0,
+    i32 %1,
+    i32 %2)
+
+  ret <vscale x 2 x i64> %a
+}
+
+declare <vscale x 2 x i64> @llvm.riscv.vwadd.mask.nxv2i64.nxv2i32.i32(
+  <vscale x 2 x i64>,
+  <vscale x 2 x i32>,
+  i32,
+  <vscale x 2 x i1>,
+  i32);
+
+define <vscale x 2 x i64> @intrinsic_vwadd_mask_vx_nxv2i64_nxv2i32_i32(<vscale x 2 x i64> %0, <vscale x 2 x i32> %1, i32 %2, <vscale x 2 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwadd_mask_vx_nxv2i64_nxv2i32_i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e32,m1,tu,mu
+; CHECK-NEXT:    vwadd.vx v8, v10, a0, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i64> @llvm.riscv.vwadd.mask.nxv2i64.nxv2i32.i32(
+    <vscale x 2 x i64> %0,
+    <vscale x 2 x i32> %1,
+    i32 %2,
+    <vscale x 2 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 2 x i64> %a
+}
+
+declare <vscale x 4 x i64> @llvm.riscv.vwadd.nxv4i64.nxv4i32.i32(
+  <vscale x 4 x i32>,
+  i32,
+  i32);
+
+define <vscale x 4 x i64> @intrinsic_vwadd_vx_nxv4i64_nxv4i32_i32(<vscale x 4 x i32> %0, i32 %1, i32 %2) nounwind {
+; CHECK-LABEL: intrinsic_vwadd_vx_nxv4i64_nxv4i32_i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e32,m2,ta,mu
+; CHECK-NEXT:    vwadd.vx v28, v8, a0
+; CHECK-NEXT:    vmv4r.v v8, v28
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i64> @llvm.riscv.vwadd.nxv4i64.nxv4i32.i32(
+    <vscale x 4 x i32> %0,
+    i32 %1,
+    i32 %2)
+
+  ret <vscale x 4 x i64> %a
+}
+
+declare <vscale x 4 x i64> @llvm.riscv.vwadd.mask.nxv4i64.nxv4i32.i32(
+  <vscale x 4 x i64>,
+  <vscale x 4 x i32>,
+  i32,
+  <vscale x 4 x i1>,
+  i32);
+
+define <vscale x 4 x i64> @intrinsic_vwadd_mask_vx_nxv4i64_nxv4i32_i32(<vscale x 4 x i64> %0, <vscale x 4 x i32> %1, i32 %2, <vscale x 4 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwadd_mask_vx_nxv4i64_nxv4i32_i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e32,m2,tu,mu
+; CHECK-NEXT:    vwadd.vx v8, v12, a0, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i64> @llvm.riscv.vwadd.mask.nxv4i64.nxv4i32.i32(
+    <vscale x 4 x i64> %0,
+    <vscale x 4 x i32> %1,
+    i32 %2,
+    <vscale x 4 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 4 x i64> %a
+}
+
+declare <vscale x 8 x i64> @llvm.riscv.vwadd.nxv8i64.nxv8i32.i32(
+  <vscale x 8 x i32>,
+  i32,
+  i32);
+
+define <vscale x 8 x i64> @intrinsic_vwadd_vx_nxv8i64_nxv8i32_i32(<vscale x 8 x i32> %0, i32 %1, i32 %2) nounwind {
+; CHECK-LABEL: intrinsic_vwadd_vx_nxv8i64_nxv8i32_i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e32,m4,ta,mu
+; CHECK-NEXT:    vwadd.vx v16, v8, a0
+; CHECK-NEXT:    vmv8r.v v8, v16
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i64> @llvm.riscv.vwadd.nxv8i64.nxv8i32.i32(
+    <vscale x 8 x i32> %0,
+    i32 %1,
+    i32 %2)
+
+  ret <vscale x 8 x i64> %a
+}
+
+declare <vscale x 8 x i64> @llvm.riscv.vwadd.mask.nxv8i64.nxv8i32.i32(
+  <vscale x 8 x i64>,
+  <vscale x 8 x i32>,
+  i32,
+  <vscale x 8 x i1>,
+  i32);
+
+define <vscale x 8 x i64> @intrinsic_vwadd_mask_vx_nxv8i64_nxv8i32_i32(<vscale x 8 x i64> %0, <vscale x 8 x i32> %1, i32 %2, <vscale x 8 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwadd_mask_vx_nxv8i64_nxv8i32_i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e32,m4,tu,mu
+; CHECK-NEXT:    vwadd.vx v8, v16, a0, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i64> @llvm.riscv.vwadd.mask.nxv8i64.nxv8i32.i32(
+    <vscale x 8 x i64> %0,
+    <vscale x 8 x i32> %1,
+    i32 %2,
+    <vscale x 8 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 8 x i64> %a
+}

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

%a = call <vscale x 16 x i32> @llvm.riscv.vwaddu.mask.nxv16i32.nxv16i16.nxv16i16(
     <vscale x 16 x i16> %1,
     <vscale x 16 x i16> %2,
     <vscale x 16 x i1> %3,
     i32 %4)
 
   ret <vscale x 16 x i32> %a
 }
 
+declare <vscale x 1 x i64> @llvm.riscv.vwaddu.nxv1i64.nxv1i32.nxv1i32(
+  <vscale x 1 x i32>,
+  <vscale x 1 x i32>,
+  i32);
+
+define <vscale x 1 x i64> @intrinsic_vwaddu_vv_nxv1i64_nxv1i32_nxv1i32(<vscale x 1 x i32> %0, <vscale x 1 x i32> %1, i32 %2) nounwind {
+; CHECK-LABEL: intrinsic_vwaddu_vv_nxv1i64_nxv1i32_nxv1i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e32,mf2,ta,mu
+; CHECK-NEXT:    vwaddu.vv v25, v8, v9
+; CHECK-NEXT:    vmv1r.v v8, v25
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i64> @llvm.riscv.vwaddu.nxv1i64.nxv1i32.nxv1i32(
+    <vscale x 1 x i32> %0,
+    <vscale x 1 x i32> %1,
+    i32 %2)
+
+  ret <vscale x 1 x i64> %a
+}
+
+declare <vscale x 1 x i64> @llvm.riscv.vwaddu.mask.nxv1i64.nxv1i32.nxv1i32(
+  <vscale x 1 x i64>,
+  <vscale x 1 x i32>,
+  <vscale x 1 x i32>,
+  <vscale x 1 x i1>,
+  i32);
+
+define <vscale x 1 x i64> @intrinsic_vwaddu_mask_vv_nxv1i64_nxv1i32_nxv1i32(<vscale x 1 x i64> %0, <vscale x 1 x i32> %1, <vscale x 1 x i32> %2, <vscale x 1 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwaddu_mask_vv_nxv1i64_nxv1i32_nxv1i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e32,mf2,tu,mu
+; CHECK-NEXT:    vwaddu.vv v8, v9, v10, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i64> @llvm.riscv.vwaddu.mask.nxv1i64.nxv1i32.nxv1i32(
+    <vscale x 1 x i64> %0,
+    <vscale x 1 x i32> %1,
+    <vscale x 1 x i32> %2,
+    <vscale x 1 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 1 x i64> %a
+}
+
+declare <vscale x 2 x i64> @llvm.riscv.vwaddu.nxv2i64.nxv2i32.nxv2i32(
+  <vscale x 2 x i32>,
+  <vscale x 2 x i32>,
+  i32);
+
+define <vscale x 2 x i64> @intrinsic_vwaddu_vv_nxv2i64_nxv2i32_nxv2i32(<vscale x 2 x i32> %0, <vscale x 2 x i32> %1, i32 %2) nounwind {
+; CHECK-LABEL: intrinsic_vwaddu_vv_nxv2i64_nxv2i32_nxv2i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e32,m1,ta,mu
+; CHECK-NEXT:    vwaddu.vv v26, v8, v9
+; CHECK-NEXT:    vmv2r.v v8, v26
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i64> @llvm.riscv.vwaddu.nxv2i64.nxv2i32.nxv2i32(
+    <vscale x 2 x i32> %0,
+    <vscale x 2 x i32> %1,
+    i32 %2)
+
+  ret <vscale x 2 x i64> %a
+}
+
+declare <vscale x 2 x i64> @llvm.riscv.vwaddu.mask.nxv2i64.nxv2i32.nxv2i32(
+  <vscale x 2 x i64>,
+  <vscale x 2 x i32>,
+  <vscale x 2 x i32>,
+  <vscale x 2 x i1>,
+  i32);
+
+define <vscale x 2 x i64> @intrinsic_vwaddu_mask_vv_nxv2i64_nxv2i32_nxv2i32(<vscale x 2 x i64> %0, <vscale x 2 x i32> %1, <vscale x 2 x i32> %2, <vscale x 2 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwaddu_mask_vv_nxv2i64_nxv2i32_nxv2i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e32,m1,tu,mu
+; CHECK-NEXT:    vwaddu.vv v8, v10, v11, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i64> @llvm.riscv.vwaddu.mask.nxv2i64.nxv2i32.nxv2i32(
+    <vscale x 2 x i64> %0,
+    <vscale x 2 x i32> %1,
+    <vscale x 2 x i32> %2,
+    <vscale x 2 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 2 x i64> %a
+}
+
+declare <vscale x 4 x i64> @llvm.riscv.vwaddu.nxv4i64.nxv4i32.nxv4i32(
+  <vscale x 4 x i32>,
+  <vscale x 4 x i32>,
+  i32);
+
+define <vscale x 4 x i64> @intrinsic_vwaddu_vv_nxv4i64_nxv4i32_nxv4i32(<vscale x 4 x i32> %0, <vscale x 4 x i32> %1, i32 %2) nounwind {
+; CHECK-LABEL: intrinsic_vwaddu_vv_nxv4i64_nxv4i32_nxv4i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e32,m2,ta,mu
+; CHECK-NEXT:    vwaddu.vv v28, v8, v10
+; CHECK-NEXT:    vmv4r.v v8, v28
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i64> @llvm.riscv.vwaddu.nxv4i64.nxv4i32.nxv4i32(
+    <vscale x 4 x i32> %0,
+    <vscale x 4 x i32> %1,
+    i32 %2)
+
+  ret <vscale x 4 x i64> %a
+}
+
+declare <vscale x 4 x i64> @llvm.riscv.vwaddu.mask.nxv4i64.nxv4i32.nxv4i32(
+  <vscale x 4 x i64>,
+  <vscale x 4 x i32>,
+  <vscale x 4 x i32>,
+  <vscale x 4 x i1>,
+  i32);
+
+define <vscale x 4 x i64> @intrinsic_vwaddu_mask_vv_nxv4i64_nxv4i32_nxv4i32(<vscale x 4 x i64> %0, <vscale x 4 x i32> %1, <vscale x 4 x i32> %2, <vscale x 4 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwaddu_mask_vv_nxv4i64_nxv4i32_nxv4i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e32,m2,tu,mu
+; CHECK-NEXT:    vwaddu.vv v8, v12, v14, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i64> @llvm.riscv.vwaddu.mask.nxv4i64.nxv4i32.nxv4i32(
+    <vscale x 4 x i64> %0,
+    <vscale x 4 x i32> %1,
+    <vscale x 4 x i32> %2,
+    <vscale x 4 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 4 x i64> %a
+}
+
+declare <vscale x 8 x i64> @llvm.riscv.vwaddu.nxv8i64.nxv8i32.nxv8i32(
+  <vscale x 8 x i32>,
+  <vscale x 8 x i32>,
+  i32);
+
+define <vscale x 8 x i64> @intrinsic_vwaddu_vv_nxv8i64_nxv8i32_nxv8i32(<vscale x 8 x i32> %0, <vscale x 8 x i32> %1, i32 %2) nounwind {
+; CHECK-LABEL: intrinsic_vwaddu_vv_nxv8i64_nxv8i32_nxv8i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e32,m4,ta,mu
+; CHECK-NEXT:    vwaddu.vv v16, v8, v12
+; CHECK-NEXT:    vmv8r.v v8, v16
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i64> @llvm.riscv.vwaddu.nxv8i64.nxv8i32.nxv8i32(
+    <vscale x 8 x i32> %0,
+    <vscale x 8 x i32> %1,
+    i32 %2)
+
+  ret <vscale x 8 x i64> %a
+}
+
+declare <vscale x 8 x i64> @llvm.riscv.vwaddu.mask.nxv8i64.nxv8i32.nxv8i32(
+  <vscale x 8 x i64>,
+  <vscale x 8 x i32>,
+  <vscale x 8 x i32>,
+  <vscale x 8 x i1>,
+  i32);
+
+define <vscale x 8 x i64> @intrinsic_vwaddu_mask_vv_nxv8i64_nxv8i32_nxv8i32(<vscale x 8 x i64> %0, <vscale x 8 x i32> %1, <vscale x 8 x i32> %2, <vscale x 8 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwaddu_mask_vv_nxv8i64_nxv8i32_nxv8i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e32,m4,tu,mu
+; CHECK-NEXT:    vwaddu.vv v8, v16, v20, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i64> @llvm.riscv.vwaddu.mask.nxv8i64.nxv8i32.nxv8i32(
+    <vscale x 8 x i64> %0,
+    <vscale x 8 x i32> %1,
+    <vscale x 8 x i32> %2,
+    <vscale x 8 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 8 x i64> %a
+}
+
 declare <vscale x 1 x i16> @llvm.riscv.vwaddu.nxv1i16.nxv1i8.i8(
   <vscale x 1 x i8>,
   i8,
   i32);
 
 define <vscale x 1 x i16> @intrinsic_vwaddu_vx_nxv1i16_nxv1i8_i8(<vscale x 1 x i8> %0, i8 %1, i32 %2) nounwind {
 ; CHECK-LABEL: intrinsic_vwaddu_vx_nxv1i16_nxv1i8_i8:
 ; CHECK:       # %bb.0: # %entry
%a = call <vscale x 16 x i32> @llvm.riscv.vwaddu.mask.nxv16i32.nxv16i16.i16(
     <vscale x 16 x i32> %0,
     <vscale x 16 x i16> %1,
     i16 %2,
     <vscale x 16 x i1> %3,
     i32 %4)
 
   ret <vscale x 16 x i32> %a
 }
+
+declare <vscale x 1 x i64> @llvm.riscv.vwaddu.nxv1i64.nxv1i32.i32(
+  <vscale x 1 x i32>,
+  i32,
+  i32);
+
+define <vscale x 1 x i64> @intrinsic_vwaddu_vx_nxv1i64_nxv1i32_i32(<vscale x 1 x i32> %0, i32 %1, i32 %2) nounwind {
+; CHECK-LABEL: intrinsic_vwaddu_vx_nxv1i64_nxv1i32_i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e32,mf2,ta,mu
+; CHECK-NEXT:    vwaddu.vx v25, v8, a0
+; CHECK-NEXT:    vmv1r.v v8, v25
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i64> @llvm.riscv.vwaddu.nxv1i64.nxv1i32.i32(
+    <vscale x 1 x i32> %0,
+    i32 %1,
+    i32 %2)
+
+  ret <vscale x 1 x i64> %a
+}
+
+declare <vscale x 1 x i64> @llvm.riscv.vwaddu.mask.nxv1i64.nxv1i32.i32(
+  <vscale x 1 x i64>,
+  <vscale x 1 x i32>,
+  i32,
+  <vscale x 1 x i1>,
+  i32);
+
+define <vscale x 1 x i64> @intrinsic_vwaddu_mask_vx_nxv1i64_nxv1i32_i32(<vscale x 1 x i64> %0, <vscale x 1 x i32> %1, i32 %2, <vscale x 1 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwaddu_mask_vx_nxv1i64_nxv1i32_i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e32,mf2,tu,mu
+; CHECK-NEXT:    vwaddu.vx v8, v9, a0, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i64> @llvm.riscv.vwaddu.mask.nxv1i64.nxv1i32.i32(
+    <vscale x 1 x i64> %0,
+    <vscale x 1 x i32> %1,
+    i32 %2,
+    <vscale x 1 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 1 x i64> %a
+}
+
+declare <vscale x 2 x i64> @llvm.riscv.vwaddu.nxv2i64.nxv2i32.i32(
+  <vscale x 2 x i32>,
+  i32,
+  i32);
+
+define <vscale x 2 x i64> @intrinsic_vwaddu_vx_nxv2i64_nxv2i32_i32(<vscale x 2 x i32> %0, i32 %1, i32 %2) nounwind {
+; CHECK-LABEL: intrinsic_vwaddu_vx_nxv2i64_nxv2i32_i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e32,m1,ta,mu
+; CHECK-NEXT:    vwaddu.vx v26, v8, a0
+; CHECK-NEXT:    vmv2r.v v8, v26
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i64> @llvm.riscv.vwaddu.nxv2i64.nxv2i32.i32(
+    <vscale x 2 x i32> %0,
+    i32 %1,
+    i32 %2)
+
+  ret <vscale x 2 x i64> %a
+}
+
+declare <vscale x 2 x i64> @llvm.riscv.vwaddu.mask.nxv2i64.nxv2i32.i32(
+  <vscale x 2 x i64>,
+  <vscale x 2 x i32>,
+  i32,
+  <vscale x 2 x i1>,
+  i32);
+
+define <vscale x 2 x i64> @intrinsic_vwaddu_mask_vx_nxv2i64_nxv2i32_i32(<vscale x 2 x i64> %0, <vscale x 2 x i32> %1, i32 %2, <vscale x 2 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwaddu_mask_vx_nxv2i64_nxv2i32_i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e32,m1,tu,mu
+; CHECK-NEXT:    vwaddu.vx v8, v10, a0, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i64> @llvm.riscv.vwaddu.mask.nxv2i64.nxv2i32.i32(
+    <vscale x 2 x i64> %0,
+    <vscale x 2 x i32> %1,
+    i32 %2,
+    <vscale x 2 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 2 x i64> %a
+}
+
+declare <vscale x 4 x i64> @llvm.riscv.vwaddu.nxv4i64.nxv4i32.i32(
+  <vscale x 4 x i32>,
+  i32,
+  i32);
+
+define <vscale x 4 x i64> @intrinsic_vwaddu_vx_nxv4i64_nxv4i32_i32(<vscale x 4 x i32> %0, i32 %1, i32 %2) nounwind {
+; CHECK-LABEL: intrinsic_vwaddu_vx_nxv4i64_nxv4i32_i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e32,m2,ta,mu
+; CHECK-NEXT:    vwaddu.vx v28, v8, a0
+; CHECK-NEXT:    vmv4r.v v8, v28
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i64> @llvm.riscv.vwaddu.nxv4i64.nxv4i32.i32(
+    <vscale x 4 x i32> %0,
+    i32 %1,
+    i32 %2)
+
+  ret <vscale x 4 x i64> %a
+}
+
+declare <vscale x 4 x i64> @llvm.riscv.vwaddu.mask.nxv4i64.nxv4i32.i32(
+  <vscale x 4 x i64>,
+  <vscale x 4 x i32>,
+  i32,
+  <vscale x 4 x i1>,
+  i32);
+
+define <vscale x 4 x i64> @intrinsic_vwaddu_mask_vx_nxv4i64_nxv4i32_i32(<vscale x 4 x i64> %0, <vscale x 4 x i32> %1, i32 %2, <vscale x 4 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwaddu_mask_vx_nxv4i64_nxv4i32_i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e32,m2,tu,mu
+; CHECK-NEXT:    vwaddu.vx v8, v12, a0, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i64> @llvm.riscv.vwaddu.mask.nxv4i64.nxv4i32.i32(
+    <vscale x 4 x i64> %0,
+    <vscale x 4 x i32> %1,
+    i32 %2,
+    <vscale x 4 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 4 x i64> %a
+}
+
+declare <vscale x 8 x i64> @llvm.riscv.vwaddu.nxv8i64.nxv8i32.i32(
+  <vscale x 8 x i32>,
+  i32,
+  i32);
+
+define <vscale x 8 x i64> @intrinsic_vwaddu_vx_nxv8i64_nxv8i32_i32(<vscale x 8 x i32> %0, i32 %1, i32 %2) nounwind {
+; CHECK-LABEL: intrinsic_vwaddu_vx_nxv8i64_nxv8i32_i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e32,m4,ta,mu
+; CHECK-NEXT:    vwaddu.vx v16, v8, a0
+; CHECK-NEXT:    vmv8r.v v8, v16
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i64> @llvm.riscv.vwaddu.nxv8i64.nxv8i32.i32(
+    <vscale x 8 x i32> %0,
+    i32 %1,
+    i32 %2)
+
+  ret <vscale x 8 x i64> %a
+}
+
+declare <vscale x 8 x i64> @llvm.riscv.vwaddu.mask.nxv8i64.nxv8i32.i32(
+  <vscale x 8 x i64>,
+  <vscale x 8 x i32>,
+  i32,
+  <vscale x 8 x i1>,
+  i32);
+
+define <vscale x 8 x i64> @intrinsic_vwaddu_mask_vx_nxv8i64_nxv8i32_i32(<vscale x 8 x i64> %0, <vscale x 8 x i32> %1, i32 %2, <vscale x 8 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwaddu_mask_vx_nxv8i64_nxv8i32_i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e32,m4,tu,mu
+; CHECK-NEXT:    vwaddu.vx v8, v16, a0, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i64> @llvm.riscv.vwaddu.mask.nxv8i64.nxv8i32.i32(
+    <vscale x 8 x i64> %0,
+    <vscale x 8 x i32> %1,
+    i32 %2,
+    <vscale x 8 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 8 x i64> %a
+}

llvm/test/CodeGen/RISCV/rvv/vwaddu.w-rv32.ll

%a = call <vscale x 16 x i32> @llvm.riscv.vwaddu.w.mask.nxv16i32.nxv16i16(
     <vscale x 16 x i32> %1,
     <vscale x 16 x i16> %2,
     <vscale x 16 x i1> %3,
     i32 %4)
 
   ret <vscale x 16 x i32> %a
 }
 
+declare <vscale x 1 x i64> @llvm.riscv.vwaddu.w.nxv1i64.nxv1i32(
+  <vscale x 1 x i64>,
+  <vscale x 1 x i32>,
+  i32);
+
+define <vscale x 1 x i64> @intrinsic_vwaddu.w_wv_nxv1i64_nxv1i64_nxv1i32(<vscale x 1 x i64> %0, <vscale x 1 x i32> %1, i32 %2) nounwind {
+; CHECK-LABEL: intrinsic_vwaddu.w_wv_nxv1i64_nxv1i64_nxv1i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e32,mf2,ta,mu
+; CHECK-NEXT:    vwaddu.wv v25, v8, v9
+; CHECK-NEXT:    vmv1r.v v8, v25
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i64> @llvm.riscv.vwaddu.w.nxv1i64.nxv1i32(
+    <vscale x 1 x i64> %0,
+    <vscale x 1 x i32> %1,
+    i32 %2)
+
+  ret <vscale x 1 x i64> %a
+}
+
+declare <vscale x 1 x i64> @llvm.riscv.vwaddu.w.mask.nxv1i64.nxv1i32(
+  <vscale x 1 x i64>,
+  <vscale x 1 x i64>,
+  <vscale x 1 x i32>,
+  <vscale x 1 x i1>,
+  i32);
+
+define <vscale x 1 x i64> @intrinsic_vwaddu.w_mask_wv_nxv1i64_nxv1i64_nxv1i32(<vscale x 1 x i64> %0, <vscale x 1 x i64> %1, <vscale x 1 x i32> %2, <vscale x 1 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwaddu.w_mask_wv_nxv1i64_nxv1i64_nxv1i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e32,mf2,tu,mu
+; CHECK-NEXT:    vwaddu.wv v8, v9, v10, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i64> @llvm.riscv.vwaddu.w.mask.nxv1i64.nxv1i32(
+    <vscale x 1 x i64> %0,
+    <vscale x 1 x i64> %1,
+    <vscale x 1 x i32> %2,
+    <vscale x 1 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 1 x i64> %a
+}
+
+declare <vscale x 2 x i64> @llvm.riscv.vwaddu.w.nxv2i64.nxv2i32(
+  <vscale x 2 x i64>,
+  <vscale x 2 x i32>,
+  i32);
+
+define <vscale x 2 x i64> @intrinsic_vwaddu.w_wv_nxv2i64_nxv2i64_nxv2i32(<vscale x 2 x i64> %0, <vscale x 2 x i32> %1, i32 %2) nounwind {
+; CHECK-LABEL: intrinsic_vwaddu.w_wv_nxv2i64_nxv2i64_nxv2i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e32,m1,ta,mu
+; CHECK-NEXT:    vwaddu.wv v26, v8, v10
+; CHECK-NEXT:    vmv2r.v v8, v26
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i64> @llvm.riscv.vwaddu.w.nxv2i64.nxv2i32(
+    <vscale x 2 x i64> %0,
+    <vscale x 2 x i32> %1,
+    i32 %2)
+
+  ret <vscale x 2 x i64> %a
+}
+
+declare <vscale x 2 x i64> @llvm.riscv.vwaddu.w.mask.nxv2i64.nxv2i32(
+  <vscale x 2 x i64>,
+  <vscale x 2 x i64>,
+  <vscale x 2 x i32>,
+  <vscale x 2 x i1>,
+  i32);
+
+define <vscale x 2 x i64> @intrinsic_vwaddu.w_mask_wv_nxv2i64_nxv2i64_nxv2i32(<vscale x 2 x i64> %0, <vscale x 2 x i64> %1, <vscale x 2 x i32> %2, <vscale x 2 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwaddu.w_mask_wv_nxv2i64_nxv2i64_nxv2i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e32,m1,tu,mu
+; CHECK-NEXT:    vwaddu.wv v8, v10, v12, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i64> @llvm.riscv.vwaddu.w.mask.nxv2i64.nxv2i32(
+    <vscale x 2 x i64> %0,
+    <vscale x 2 x i64> %1,
+    <vscale x 2 x i32> %2,
+    <vscale x 2 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 2 x i64> %a
+}
+
+declare <vscale x 4 x i64> @llvm.riscv.vwaddu.w.nxv4i64.nxv4i32(
+  <vscale x 4 x i64>,
+  <vscale x 4 x i32>,
+  i32);
+
+define <vscale x 4 x i64> @intrinsic_vwaddu.w_wv_nxv4i64_nxv4i64_nxv4i32(<vscale x 4 x i64> %0, <vscale x 4 x i32> %1, i32 %2) nounwind {
+; CHECK-LABEL: intrinsic_vwaddu.w_wv_nxv4i64_nxv4i64_nxv4i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e32,m2,ta,mu
+; CHECK-NEXT:    vwaddu.wv v28, v8, v12
+; CHECK-NEXT:    vmv4r.v v8, v28
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i64> @llvm.riscv.vwaddu.w.nxv4i64.nxv4i32(
+    <vscale x 4 x i64> %0,
+    <vscale x 4 x i32> %1,
+    i32 %2)
+
+  ret <vscale x 4 x i64> %a
+}
+
+declare <vscale x 4 x i64> @llvm.riscv.vwaddu.w.mask.nxv4i64.nxv4i32(
+  <vscale x 4 x i64>,
+  <vscale x 4 x i64>,
+  <vscale x 4 x i32>,
+  <vscale x 4 x i1>,
+  i32);
+
+define <vscale x 4 x i64> @intrinsic_vwaddu.w_mask_wv_nxv4i64_nxv4i64_nxv4i32(<vscale x 4 x i64> %0, <vscale x 4 x i64> %1, <vscale x 4 x i32> %2, <vscale x 4 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwaddu.w_mask_wv_nxv4i64_nxv4i64_nxv4i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e32,m2,tu,mu
+; CHECK-NEXT:    vwaddu.wv v8, v12, v16, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i64> @llvm.riscv.vwaddu.w.mask.nxv4i64.nxv4i32(
+    <vscale x 4 x i64> %0,
+    <vscale x 4 x i64> %1,
+    <vscale x 4 x i32> %2,
+    <vscale x 4 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 4 x i64> %a
+}
+
+declare <vscale x 8 x i64> @llvm.riscv.vwaddu.w.nxv8i64.nxv8i32(
+  <vscale x 8 x i64>,
+  <vscale x 8 x i32>,
+  i32);
+
+define <vscale x 8 x i64> @intrinsic_vwaddu.w_wv_nxv8i64_nxv8i64_nxv8i32(<vscale x 8 x i64> %0, <vscale x 8 x i32> %1, i32 %2) nounwind {
+; CHECK-LABEL: intrinsic_vwaddu.w_wv_nxv8i64_nxv8i64_nxv8i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e32,m4,ta,mu
+; CHECK-NEXT:    vwaddu.wv v24, v8, v16
+; CHECK-NEXT:    vmv8r.v v8, v24
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i64> @llvm.riscv.vwaddu.w.nxv8i64.nxv8i32(
+    <vscale x 8 x i64> %0,
+    <vscale x 8 x i32> %1,
+    i32 %2)
+
+  ret <vscale x 8 x i64> %a
+}
+
+declare <vscale x 8 x i64> @llvm.riscv.vwaddu.w.mask.nxv8i64.nxv8i32(
+  <vscale x 8 x i64>,
+  <vscale x 8 x i64>,
+  <vscale x 8 x i32>,
+  <vscale x 8 x i1>,
+  i32);
+
+define <vscale x 8 x i64> @intrinsic_vwaddu.w_mask_wv_nxv8i64_nxv8i64_nxv8i32(<vscale x 8 x i64> %0, <vscale x 8 x i64> %1, <vscale x 8 x i32> %2, <vscale x 8 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwaddu.w_mask_wv_nxv8i64_nxv8i64_nxv8i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vl4re32.v v28, (a0)
+; CHECK-NEXT:    vsetvli a0, a1, e32,m4,tu,mu
+; CHECK-NEXT:    vwaddu.wv v8, v16, v28, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i64> @llvm.riscv.vwaddu.w.mask.nxv8i64.nxv8i32(
+    <vscale x 8 x i64> %0,
+    <vscale x 8 x i64> %1,
+    <vscale x 8 x i32> %2,
+    <vscale x 8 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 8 x i64> %a
+}
+
 declare <vscale x 1 x i16> @llvm.riscv.vwaddu.w.nxv1i16.i8(
   <vscale x 1 x i16>,
   i8,
   i32);
 
 define <vscale x 1 x i16> @intrinsic_vwaddu.w_wx_nxv1i16_nxv1i16_i8(<vscale x 1 x i16> %0, i8 %1, i32 %2) nounwind {
 ; CHECK-LABEL: intrinsic_vwaddu.w_wx_nxv1i16_nxv1i16_i8:
 ; CHECK:       # %bb.0: # %entry
%a = call <vscale x 16 x i32> @llvm.riscv.vwaddu.w.mask.nxv16i32.i16(
     <vscale x 16 x i32> %0,
     <vscale x 16 x i32> %1,
     i16 %2,
     <vscale x 16 x i1> %3,
     i32 %4)
 
   ret <vscale x 16 x i32> %a
 }
+
+declare <vscale x 1 x i64> @llvm.riscv.vwaddu.w.nxv1i64.i32(
+  <vscale x 1 x i64>,
+  i32,
+  i32);
+
+define <vscale x 1 x i64> @intrinsic_vwaddu.w_wx_nxv1i64_nxv1i64_i32(<vscale x 1 x i64> %0, i32 %1, i32 %2) nounwind {
+; CHECK-LABEL: intrinsic_vwaddu.w_wx_nxv1i64_nxv1i64_i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e32,mf2,ta,mu
+; CHECK-NEXT:    vwaddu.wx v25, v8, a0
+; CHECK-NEXT:    vmv1r.v v8, v25
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i64> @llvm.riscv.vwaddu.w.nxv1i64.i32(
+    <vscale x 1 x i64> %0,
+    i32 %1,
+    i32 %2)
+
+  ret <vscale x 1 x i64> %a
+}
+
+declare <vscale x 1 x i64> @llvm.riscv.vwaddu.w.mask.nxv1i64.i32(
+  <vscale x 1 x i64>,
+  <vscale x 1 x i64>,
+  i32,
+  <vscale x 1 x i1>,
+  i32);
+
+define <vscale x 1 x i64> @intrinsic_vwaddu.w_mask_wx_nxv1i64_nxv1i64_i32(<vscale x 1 x i64> %0, <vscale x 1 x i64> %1, i32 %2, <vscale x 1 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwaddu.w_mask_wx_nxv1i64_nxv1i64_i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e32,mf2,tu,mu
+; CHECK-NEXT:    vwaddu.wx v8, v9, a0, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i64> @llvm.riscv.vwaddu.w.mask.nxv1i64.i32(
+    <vscale x 1 x i64> %0,
+    <vscale x 1 x i64> %1,
+    i32 %2,
+    <vscale x 1 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 1 x i64> %a
+}
+
+declare <vscale x 2 x i64> @llvm.riscv.vwaddu.w.nxv2i64.i32(
+  <vscale x 2 x i64>,
+  i32,
+  i32);
+
+define <vscale x 2 x i64> @intrinsic_vwaddu.w_wx_nxv2i64_nxv2i64_i32(<vscale x 2 x i64> %0, i32 %1, i32 %2) nounwind {
+; CHECK-LABEL: intrinsic_vwaddu.w_wx_nxv2i64_nxv2i64_i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e32,m1,ta,mu
+; CHECK-NEXT:    vwaddu.wx v26, v8, a0
+; CHECK-NEXT:    vmv2r.v v8, v26
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i64> @llvm.riscv.vwaddu.w.nxv2i64.i32(
+    <vscale x 2 x i64> %0,
+    i32 %1,
+    i32 %2)
+
+  ret <vscale x 2 x i64> %a
+}
+
+declare <vscale x 2 x i64> @llvm.riscv.vwaddu.w.mask.nxv2i64.i32(
+  <vscale x 2 x i64>,
+  <vscale x 2 x i64>,
+  i32,
+  <vscale x 2 x i1>,
+  i32);
+
+define <vscale x 2 x i64> @intrinsic_vwaddu.w_mask_wx_nxv2i64_nxv2i64_i32(<vscale x 2 x i64> %0, <vscale x 2 x i64> %1, i32 %2, <vscale x 2 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwaddu.w_mask_wx_nxv2i64_nxv2i64_i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e32,m1,tu,mu
+; CHECK-NEXT:    vwaddu.wx v8, v10, a0, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i64> @llvm.riscv.vwaddu.w.mask.nxv2i64.i32(
+    <vscale x 2 x i64> %0,
+    <vscale x 2 x i64> %1,
+    i32 %2,
+    <vscale x 2 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 2 x i64> %a
+}
+
+declare <vscale x 4 x i64> @llvm.riscv.vwaddu.w.nxv4i64.i32(
+  <vscale x 4 x i64>,
+  i32,
+  i32);
+
+define <vscale x 4 x i64> @intrinsic_vwaddu.w_wx_nxv4i64_nxv4i64_i32(<vscale x 4 x i64> %0, i32 %1, i32 %2) nounwind {
+; CHECK-LABEL: intrinsic_vwaddu.w_wx_nxv4i64_nxv4i64_i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e32,m2,ta,mu
+; CHECK-NEXT:    vwaddu.wx v28, v8, a0
+; CHECK-NEXT:    vmv4r.v v8, v28
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i64> @llvm.riscv.vwaddu.w.nxv4i64.i32(
+    <vscale x 4 x i64> %0,
+    i32 %1,
+    i32 %2)
+
+  ret <vscale x 4 x i64> %a
+}
+
+declare <vscale x 4 x i64> @llvm.riscv.vwaddu.w.mask.nxv4i64.i32(
+  <vscale x 4 x i64>,
+  <vscale x 4 x i64>,
+  i32,
+  <vscale x 4 x i1>,
+  i32);
+
+define <vscale x 4 x i64> @intrinsic_vwaddu.w_mask_wx_nxv4i64_nxv4i64_i32(<vscale x 4 x i64> %0, <vscale x 4 x i64> %1, i32 %2, <vscale x 4 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwaddu.w_mask_wx_nxv4i64_nxv4i64_i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e32,m2,tu,mu
+; CHECK-NEXT:    vwaddu.wx v8, v12, a0, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i64> @llvm.riscv.vwaddu.w.mask.nxv4i64.i32(
+    <vscale x 4 x i64> %0,
+    <vscale x 4 x i64> %1,
+    i32 %2,
+    <vscale x 4 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 4 x i64> %a
+}
+
+declare <vscale x 8 x i64> @llvm.riscv.vwaddu.w.nxv8i64.i32(
+  <vscale x 8 x i64>,
+  i32,
+  i32);
+
+define <vscale x 8 x i64> @intrinsic_vwaddu.w_wx_nxv8i64_nxv8i64_i32(<vscale x 8 x i64> %0, i32 %1, i32 %2) nounwind {
+; CHECK-LABEL: intrinsic_vwaddu.w_wx_nxv8i64_nxv8i64_i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e32,m4,ta,mu
+; CHECK-NEXT:    vwaddu.wx v16, v8, a0
+; CHECK-NEXT:    vmv8r.v v8, v16
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i64> @llvm.riscv.vwaddu.w.nxv8i64.i32(
+    <vscale x 8 x i64> %0,
+    i32 %1,
+    i32 %2)
+
+  ret <vscale x 8 x i64> %a
+}
+
+declare <vscale x 8 x i64> @llvm.riscv.vwaddu.w.mask.nxv8i64.i32(
+  <vscale x 8 x i64>,
+  <vscale x 8 x i64>,
+  i32,
+  <vscale x 8 x i1>,
+  i32);
+
+define <vscale x 8 x i64> @intrinsic_vwaddu.w_mask_wx_nxv8i64_nxv8i64_i32(<vscale x 8 x i64> %0, <vscale x 8 x i64> %1, i32 %2, <vscale x 8 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwaddu.w_mask_wx_nxv8i64_nxv8i64_i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e32,m4,tu,mu
+; CHECK-NEXT:    vwaddu.wx v8, v16, a0, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i64> @llvm.riscv.vwaddu.w.mask.nxv8i64.i32(
+    <vscale x 8 x i64> %0,
+    <vscale x 8 x i64> %1,
+    i32 %2,
+    <vscale x 8 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 8 x i64> %a
+}

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

%a = call <vscale x 16 x i32> @llvm.riscv.vwsub.mask.nxv16i32.nxv16i16.nxv16i16(
     <vscale x 16 x i16> %1,
     <vscale x 16 x i16> %2,
     <vscale x 16 x i1> %3,
     i32 %4)
 
   ret <vscale x 16 x i32> %a
 }
 
+declare <vscale x 1 x i64> @llvm.riscv.vwsub.nxv1i64.nxv1i32.nxv1i32(
+  <vscale x 1 x i32>,
+  <vscale x 1 x i32>,
+  i32);
+
+define <vscale x 1 x i64> @intrinsic_vwsub_vv_nxv1i64_nxv1i32_nxv1i32(<vscale x 1 x i32> %0, <vscale x 1 x i32> %1, i32 %2) nounwind {
+; CHECK-LABEL: intrinsic_vwsub_vv_nxv1i64_nxv1i32_nxv1i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e32,mf2,ta,mu
+; CHECK-NEXT:    vwsub.vv v25, v8, v9
+; CHECK-NEXT:    vmv1r.v v8, v25
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i64> @llvm.riscv.vwsub.nxv1i64.nxv1i32.nxv1i32(
+    <vscale x 1 x i32> %0,
+    <vscale x 1 x i32> %1,
+    i32 %2)
+
+  ret <vscale x 1 x i64> %a
+}
+
+declare <vscale x 1 x i64> @llvm.riscv.vwsub.mask.nxv1i64.nxv1i32.nxv1i32(
+  <vscale x 1 x i64>,
+  <vscale x 1 x i32>,
+  <vscale x 1 x i32>,
+  <vscale x 1 x i1>,
+  i32);
+
+define <vscale x 1 x i64> @intrinsic_vwsub_mask_vv_nxv1i64_nxv1i32_nxv1i32(<vscale x 1 x i64> %0, <vscale x 1 x i32> %1, <vscale x 1 x i32> %2, <vscale x 1 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwsub_mask_vv_nxv1i64_nxv1i32_nxv1i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e32,mf2,tu,mu
+; CHECK-NEXT:    vwsub.vv v8, v9, v10, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i64> @llvm.riscv.vwsub.mask.nxv1i64.nxv1i32.nxv1i32(
+    <vscale x 1 x i64> %0,
+    <vscale x 1 x i32> %1,
+    <vscale x 1 x i32> %2,
+    <vscale x 1 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 1 x i64> %a
+}
+
+declare <vscale x 2 x i64> @llvm.riscv.vwsub.nxv2i64.nxv2i32.nxv2i32(
+  <vscale x 2 x i32>,
+  <vscale x 2 x i32>,
+  i32);
+
+define <vscale x 2 x i64> @intrinsic_vwsub_vv_nxv2i64_nxv2i32_nxv2i32(<vscale x 2 x i32> %0, <vscale x 2 x i32> %1, i32 %2) nounwind {
+; CHECK-LABEL: intrinsic_vwsub_vv_nxv2i64_nxv2i32_nxv2i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e32,m1,ta,mu
+; CHECK-NEXT:    vwsub.vv v26, v8, v9
+; CHECK-NEXT:    vmv2r.v v8, v26
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i64> @llvm.riscv.vwsub.nxv2i64.nxv2i32.nxv2i32(
+    <vscale x 2 x i32> %0,
+    <vscale x 2 x i32> %1,
+    i32 %2)
+
+  ret <vscale x 2 x i64> %a
+}
+
+declare <vscale x 2 x i64> @llvm.riscv.vwsub.mask.nxv2i64.nxv2i32.nxv2i32(
+  <vscale x 2 x i64>,
+  <vscale x 2 x i32>,
+  <vscale x 2 x i32>,
+  <vscale x 2 x i1>,
+  i32);
+
+define <vscale x 2 x i64> @intrinsic_vwsub_mask_vv_nxv2i64_nxv2i32_nxv2i32(<vscale x 2 x i64> %0, <vscale x 2 x i32> %1, <vscale x 2 x i32> %2, <vscale x 2 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwsub_mask_vv_nxv2i64_nxv2i32_nxv2i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e32,m1,tu,mu
+; CHECK-NEXT:    vwsub.vv v8, v10, v11, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i64> @llvm.riscv.vwsub.mask.nxv2i64.nxv2i32.nxv2i32(
+    <vscale x 2 x i64> %0,
+    <vscale x 2 x i32> %1,
+    <vscale x 2 x i32> %2,
+    <vscale x 2 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 2 x i64> %a
+}
+
+declare <vscale x 4 x i64> @llvm.riscv.vwsub.nxv4i64.nxv4i32.nxv4i32(
+  <vscale x 4 x i32>,
+  <vscale x 4 x i32>,
+  i32);
+
+define <vscale x 4 x i64> @intrinsic_vwsub_vv_nxv4i64_nxv4i32_nxv4i32(<vscale x 4 x i32> %0, <vscale x 4 x i32> %1, i32 %2) nounwind {
+; CHECK-LABEL: intrinsic_vwsub_vv_nxv4i64_nxv4i32_nxv4i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e32,m2,ta,mu
+; CHECK-NEXT:    vwsub.vv v28, v8, v10
+; CHECK-NEXT:    vmv4r.v v8, v28
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i64> @llvm.riscv.vwsub.nxv4i64.nxv4i32.nxv4i32(
+    <vscale x 4 x i32> %0,
+    <vscale x 4 x i32> %1,
+    i32 %2)
+
+  ret <vscale x 4 x i64> %a
+}
+
+declare <vscale x 4 x i64> @llvm.riscv.vwsub.mask.nxv4i64.nxv4i32.nxv4i32(
+  <vscale x 4 x i64>,
+  <vscale x 4 x i32>,
+  <vscale x 4 x i32>,
+  <vscale x 4 x i1>,
+  i32);
+
+define <vscale x 4 x i64> @intrinsic_vwsub_mask_vv_nxv4i64_nxv4i32_nxv4i32(<vscale x 4 x i64> %0, <vscale x 4 x i32> %1, <vscale x 4 x i32> %2, <vscale x 4 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwsub_mask_vv_nxv4i64_nxv4i32_nxv4i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e32,m2,tu,mu
+; CHECK-NEXT:    vwsub.vv v8, v12, v14, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i64> @llvm.riscv.vwsub.mask.nxv4i64.nxv4i32.nxv4i32(
+    <vscale x 4 x i64> %0,
+    <vscale x 4 x i32> %1,
+    <vscale x 4 x i32> %2,
+    <vscale x 4 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 4 x i64> %a
+}
+
+declare <vscale x 8 x i64> @llvm.riscv.vwsub.nxv8i64.nxv8i32.nxv8i32(
+  <vscale x 8 x i32>,
+  <vscale x 8 x i32>,
+  i32);
+
+define <vscale x 8 x i64> @intrinsic_vwsub_vv_nxv8i64_nxv8i32_nxv8i32(<vscale x 8 x i32> %0, <vscale x 8 x i32> %1, i32 %2) nounwind {
+; CHECK-LABEL: intrinsic_vwsub_vv_nxv8i64_nxv8i32_nxv8i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e32,m4,ta,mu
+; CHECK-NEXT:    vwsub.vv v16, v8, v12
+; CHECK-NEXT:    vmv8r.v v8, v16
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i64> @llvm.riscv.vwsub.nxv8i64.nxv8i32.nxv8i32(
+    <vscale x 8 x i32> %0,
+    <vscale x 8 x i32> %1,
+    i32 %2)
+
+  ret <vscale x 8 x i64> %a
+}
+
+declare <vscale x 8 x i64> @llvm.riscv.vwsub.mask.nxv8i64.nxv8i32.nxv8i32(
+  <vscale x 8 x i64>,
+  <vscale x 8 x i32>,
+  <vscale x 8 x i32>,
+  <vscale x 8 x i1>,
+  i32);
+
+define <vscale x 8 x i64> @intrinsic_vwsub_mask_vv_nxv8i64_nxv8i32_nxv8i32(<vscale x 8 x i64> %0, <vscale x 8 x i32> %1, <vscale x 8 x i32> %2, <vscale x 8 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwsub_mask_vv_nxv8i64_nxv8i32_nxv8i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e32,m4,tu,mu
+; CHECK-NEXT:    vwsub.vv v8, v16, v20, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i64> @llvm.riscv.vwsub.mask.nxv8i64.nxv8i32.nxv8i32(
+    <vscale x 8 x i64> %0,
+    <vscale x 8 x i32> %1,
+    <vscale x 8 x i32> %2,
+    <vscale x 8 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 8 x i64> %a
+}
+
 declare <vscale x 1 x i16> @llvm.riscv.vwsub.nxv1i16.nxv1i8.i8(
   <vscale x 1 x i8>,
   i8,
   i32);
 
 define <vscale x 1 x i16> @intrinsic_vwsub_vx_nxv1i16_nxv1i8_i8(<vscale x 1 x i8> %0, i8 %1, i32 %2) nounwind {
 ; CHECK-LABEL: intrinsic_vwsub_vx_nxv1i16_nxv1i8_i8:
 ; CHECK:       # %bb.0: # %entry
%a = call <vscale x 16 x i32> @llvm.riscv.vwsub.mask.nxv16i32.nxv16i16.i16(
     <vscale x 16 x i32> %0,
     <vscale x 16 x i16> %1,
     i16 %2,
     <vscale x 16 x i1> %3,
     i32 %4)
 
   ret <vscale x 16 x i32> %a
 }
+
+declare <vscale x 1 x i64> @llvm.riscv.vwsub.nxv1i64.nxv1i32.i32(
+  <vscale x 1 x i32>,
+  i32,
+  i32);
+
+define <vscale x 1 x i64> @intrinsic_vwsub_vx_nxv1i64_nxv1i32_i32(<vscale x 1 x i32> %0, i32 %1, i32 %2) nounwind {
+; CHECK-LABEL: intrinsic_vwsub_vx_nxv1i64_nxv1i32_i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e32,mf2,ta,mu
+; CHECK-NEXT:    vwsub.vx v25, v8, a0
+; CHECK-NEXT:    vmv1r.v v8, v25
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i64> @llvm.riscv.vwsub.nxv1i64.nxv1i32.i32(
+    <vscale x 1 x i32> %0,
+    i32 %1,
+    i32 %2)
+
+  ret <vscale x 1 x i64> %a
+}
+
+declare <vscale x 1 x i64> @llvm.riscv.vwsub.mask.nxv1i64.nxv1i32.i32(
+  <vscale x 1 x i64>,
+  <vscale x 1 x i32>,
+  i32,
+  <vscale x 1 x i1>,
+  i32);
+
+define <vscale x 1 x i64> @intrinsic_vwsub_mask_vx_nxv1i64_nxv1i32_i32(<vscale x 1 x i64> %0, <vscale x 1 x i32> %1, i32 %2, <vscale x 1 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwsub_mask_vx_nxv1i64_nxv1i32_i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e32,mf2,tu,mu
+; CHECK-NEXT:    vwsub.vx v8, v9, a0, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i64> @llvm.riscv.vwsub.mask.nxv1i64.nxv1i32.i32(
+    <vscale x 1 x i64> %0,
+    <vscale x 1 x i32> %1,
+    i32 %2,
+    <vscale x 1 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 1 x i64> %a
+}
+
+declare <vscale x 2 x i64> @llvm.riscv.vwsub.nxv2i64.nxv2i32.i32(
+  <vscale x 2 x i32>,
+  i32,
+  i32);
+
+define <vscale x 2 x i64> @intrinsic_vwsub_vx_nxv2i64_nxv2i32_i32(<vscale x 2 x i32> %0, i32 %1, i32 %2) nounwind {
+; CHECK-LABEL: intrinsic_vwsub_vx_nxv2i64_nxv2i32_i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e32,m1,ta,mu
+; CHECK-NEXT:    vwsub.vx v26, v8, a0
+; CHECK-NEXT:    vmv2r.v v8, v26
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i64> @llvm.riscv.vwsub.nxv2i64.nxv2i32.i32(
+    <vscale x 2 x i32> %0,
+    i32 %1,
+    i32 %2)
+
+  ret <vscale x 2 x i64> %a
+}
+
+declare <vscale x 2 x i64> @llvm.riscv.vwsub.mask.nxv2i64.nxv2i32.i32(
+  <vscale x 2 x i64>,
+  <vscale x 2 x i32>,
+  i32,
+  <vscale x 2 x i1>,
+  i32);
+
+define <vscale x 2 x i64> @intrinsic_vwsub_mask_vx_nxv2i64_nxv2i32_i32(<vscale x 2 x i64> %0, <vscale x 2 x i32> %1, i32 %2, <vscale x 2 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwsub_mask_vx_nxv2i64_nxv2i32_i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e32,m1,tu,mu
+; CHECK-NEXT:    vwsub.vx v8, v10, a0, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i64> @llvm.riscv.vwsub.mask.nxv2i64.nxv2i32.i32(
+    <vscale x 2 x i64> %0,
+    <vscale x 2 x i32> %1,
+    i32 %2,
+    <vscale x 2 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 2 x i64> %a
+}
+
+declare <vscale x 4 x i64> @llvm.riscv.vwsub.nxv4i64.nxv4i32.i32(
+  <vscale x 4 x i32>,
+  i32,
+  i32);
+
+define <vscale x 4 x i64> @intrinsic_vwsub_vx_nxv4i64_nxv4i32_i32(<vscale x 4 x i32> %0, i32 %1, i32 %2) nounwind {
+; CHECK-LABEL: intrinsic_vwsub_vx_nxv4i64_nxv4i32_i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e32,m2,ta,mu
+; CHECK-NEXT:    vwsub.vx v28, v8, a0
+; CHECK-NEXT:    vmv4r.v v8, v28
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i64> @llvm.riscv.vwsub.nxv4i64.nxv4i32.i32(
+    <vscale x 4 x i32> %0,
+    i32 %1,
+    i32 %2)
+
+  ret <vscale x 4 x i64> %a
+}
+
+declare <vscale x 4 x i64> @llvm.riscv.vwsub.mask.nxv4i64.nxv4i32.i32(
+  <vscale x 4 x i64>,
+  <vscale x 4 x i32>,
+  i32,
+  <vscale x 4 x i1>,
+  i32);
+
+define <vscale x 4 x i64> @intrinsic_vwsub_mask_vx_nxv4i64_nxv4i32_i32(<vscale x 4 x i64> %0, <vscale x 4 x i32> %1, i32 %2, <vscale x 4 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwsub_mask_vx_nxv4i64_nxv4i32_i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e32,m2,tu,mu
+; CHECK-NEXT:    vwsub.vx v8, v12, a0, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i64> @llvm.riscv.vwsub.mask.nxv4i64.nxv4i32.i32(
+    <vscale x 4 x i64> %0,
+    <vscale x 4 x i32> %1,
+    i32 %2,
+    <vscale x 4 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 4 x i64> %a
+}
+
+declare <vscale x 8 x i64> @llvm.riscv.vwsub.nxv8i64.nxv8i32.i32(
+  <vscale x 8 x i32>,
+  i32,
+  i32);
+
+define <vscale x 8 x i64> @intrinsic_vwsub_vx_nxv8i64_nxv8i32_i32(<vscale x 8 x i32> %0, i32 %1, i32 %2) nounwind {
+; CHECK-LABEL: intrinsic_vwsub_vx_nxv8i64_nxv8i32_i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e32,m4,ta,mu
+; CHECK-NEXT:    vwsub.vx v16, v8, a0
+; CHECK-NEXT:    vmv8r.v v8, v16
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i64> @llvm.riscv.vwsub.nxv8i64.nxv8i32.i32(
+    <vscale x 8 x i32> %0,
+    i32 %1,
+    i32 %2)
+
+  ret <vscale x 8 x i64> %a
+}
+
+declare <vscale x 8 x i64> @llvm.riscv.vwsub.mask.nxv8i64.nxv8i32.i32(
+  <vscale x 8 x i64>,
+  <vscale x 8 x i32>,
+  i32,
+  <vscale x 8 x i1>,
+  i32);
+
+define <vscale x 8 x i64> @intrinsic_vwsub_mask_vx_nxv8i64_nxv8i32_i32(<vscale x 8 x i64> %0, <vscale x 8 x i32> %1, i32 %2, <vscale x 8 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwsub_mask_vx_nxv8i64_nxv8i32_i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e32,m4,tu,mu
+; CHECK-NEXT:    vwsub.vx v8, v16, a0, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i64> @llvm.riscv.vwsub.mask.nxv8i64.nxv8i32.i32(
+    <vscale x 8 x i64> %0,
+    <vscale x 8 x i32> %1,
+    i32 %2,
+    <vscale x 8 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 8 x i64> %a
+}

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

%a = call <vscale x 16 x i32> @llvm.riscv.vwsubu.mask.nxv16i32.nxv16i16.nxv16i16(
     <vscale x 16 x i16> %1,
     <vscale x 16 x i16> %2,
     <vscale x 16 x i1> %3,
     i32 %4)
 
   ret <vscale x 16 x i32> %a
 }
 
+declare <vscale x 1 x i64> @llvm.riscv.vwsubu.nxv1i64.nxv1i32.nxv1i32(
+  <vscale x 1 x i32>,
+  <vscale x 1 x i32>,
+  i32);
+
+define <vscale x 1 x i64> @intrinsic_vwsubu_vv_nxv1i64_nxv1i32_nxv1i32(<vscale x 1 x i32> %0, <vscale x 1 x i32> %1, i32 %2) nounwind {
+; CHECK-LABEL: intrinsic_vwsubu_vv_nxv1i64_nxv1i32_nxv1i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e32,mf2,ta,mu
+; CHECK-NEXT:    vwsubu.vv v25, v8, v9
+; CHECK-NEXT:    vmv1r.v v8, v25
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i64> @llvm.riscv.vwsubu.nxv1i64.nxv1i32.nxv1i32(
+    <vscale x 1 x i32> %0,
+    <vscale x 1 x i32> %1,
+    i32 %2)
+
+  ret <vscale x 1 x i64> %a
+}
+
+declare <vscale x 1 x i64> @llvm.riscv.vwsubu.mask.nxv1i64.nxv1i32.nxv1i32(
+  <vscale x 1 x i64>,
+  <vscale x 1 x i32>,
+  <vscale x 1 x i32>,
+  <vscale x 1 x i1>,
+  i32);
+
+define <vscale x 1 x i64> @intrinsic_vwsubu_mask_vv_nxv1i64_nxv1i32_nxv1i32(<vscale x 1 x i64> %0, <vscale x 1 x i32> %1, <vscale x 1 x i32> %2, <vscale x 1 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwsubu_mask_vv_nxv1i64_nxv1i32_nxv1i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e32,mf2,tu,mu
+; CHECK-NEXT:    vwsubu.vv v8, v9, v10, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i64> @llvm.riscv.vwsubu.mask.nxv1i64.nxv1i32.nxv1i32(
+    <vscale x 1 x i64> %0,
+    <vscale x 1 x i32> %1,
+    <vscale x 1 x i32> %2,
+    <vscale x 1 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 1 x i64> %a
+}
+
+declare <vscale x 2 x i64> @llvm.riscv.vwsubu.nxv2i64.nxv2i32.nxv2i32(
+  <vscale x 2 x i32>,
+  <vscale x 2 x i32>,
+  i32);
+
+define <vscale x 2 x i64> @intrinsic_vwsubu_vv_nxv2i64_nxv2i32_nxv2i32(<vscale x 2 x i32> %0, <vscale x 2 x i32> %1, i32 %2) nounwind {
+; CHECK-LABEL: intrinsic_vwsubu_vv_nxv2i64_nxv2i32_nxv2i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e32,m1,ta,mu
+; CHECK-NEXT:    vwsubu.vv v26, v8, v9
+; CHECK-NEXT:    vmv2r.v v8, v26
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i64> @llvm.riscv.vwsubu.nxv2i64.nxv2i32.nxv2i32(
+    <vscale x 2 x i32> %0,
+    <vscale x 2 x i32> %1,
+    i32 %2)
+
+  ret <vscale x 2 x i64> %a
+}
+
+declare <vscale x 2 x i64> @llvm.riscv.vwsubu.mask.nxv2i64.nxv2i32.nxv2i32(
+  <vscale x 2 x i64>,
+  <vscale x 2 x i32>,
+  <vscale x 2 x i32>,
+  <vscale x 2 x i1>,
+  i32);
+
+define <vscale x 2 x i64> @intrinsic_vwsubu_mask_vv_nxv2i64_nxv2i32_nxv2i32(<vscale x 2 x i64> %0, <vscale x 2 x i32> %1, <vscale x 2 x i32> %2, <vscale x 2 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwsubu_mask_vv_nxv2i64_nxv2i32_nxv2i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e32,m1,tu,mu
+; CHECK-NEXT:    vwsubu.vv v8, v10, v11, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i64> @llvm.riscv.vwsubu.mask.nxv2i64.nxv2i32.nxv2i32(
+    <vscale x 2 x i64> %0,
+    <vscale x 2 x i32> %1,
+    <vscale x 2 x i32> %2,
+    <vscale x 2 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 2 x i64> %a
+}
+
+declare <vscale x 4 x i64> @llvm.riscv.vwsubu.nxv4i64.nxv4i32.nxv4i32(
+  <vscale x 4 x i32>,
+  <vscale x 4 x i32>,
+  i32);
+
+define <vscale x 4 x i64> @intrinsic_vwsubu_vv_nxv4i64_nxv4i32_nxv4i32(<vscale x 4 x i32> %0, <vscale x 4 x i32> %1, i32 %2) nounwind {
+; CHECK-LABEL: intrinsic_vwsubu_vv_nxv4i64_nxv4i32_nxv4i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e32,m2,ta,mu
+; CHECK-NEXT:    vwsubu.vv v28, v8, v10
+; CHECK-NEXT:    vmv4r.v v8, v28
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i64> @llvm.riscv.vwsubu.nxv4i64.nxv4i32.nxv4i32(
+    <vscale x 4 x i32> %0,
+    <vscale x 4 x i32> %1,
+    i32 %2)
+
+  ret <vscale x 4 x i64> %a
+}
+
+declare <vscale x 4 x i64> @llvm.riscv.vwsubu.mask.nxv4i64.nxv4i32.nxv4i32(
+  <vscale x 4 x i64>,
+  <vscale x 4 x i32>,
+  <vscale x 4 x i32>,
+  <vscale x 4 x i1>,
+  i32);
+
+define <vscale x 4 x i64> @intrinsic_vwsubu_mask_vv_nxv4i64_nxv4i32_nxv4i32(<vscale x 4 x i64> %0, <vscale x 4 x i32> %1, <vscale x 4 x i32> %2, <vscale x 4 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwsubu_mask_vv_nxv4i64_nxv4i32_nxv4i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e32,m2,tu,mu
+; CHECK-NEXT:    vwsubu.vv v8, v12, v14, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i64> @llvm.riscv.vwsubu.mask.nxv4i64.nxv4i32.nxv4i32(
+    <vscale x 4 x i64> %0,
+    <vscale x 4 x i32> %1,
+    <vscale x 4 x i32> %2,
+    <vscale x 4 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 4 x i64> %a
+}
+
+declare <vscale x 8 x i64> @llvm.riscv.vwsubu.nxv8i64.nxv8i32.nxv8i32(
+  <vscale x 8 x i32>,
+  <vscale x 8 x i32>,
+  i32);
+
+define <vscale x 8 x i64> @intrinsic_vwsubu_vv_nxv8i64_nxv8i32_nxv8i32(<vscale x 8 x i32> %0, <vscale x 8 x i32> %1, i32 %2) nounwind {
+; CHECK-LABEL: intrinsic_vwsubu_vv_nxv8i64_nxv8i32_nxv8i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e32,m4,ta,mu
+; CHECK-NEXT:    vwsubu.vv v16, v8, v12
+; CHECK-NEXT:    vmv8r.v v8, v16
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i64> @llvm.riscv.vwsubu.nxv8i64.nxv8i32.nxv8i32(
+    <vscale x 8 x i32> %0,
+    <vscale x 8 x i32> %1,
+    i32 %2)
+
+  ret <vscale x 8 x i64> %a
+}
+
+declare <vscale x 8 x i64> @llvm.riscv.vwsubu.mask.nxv8i64.nxv8i32.nxv8i32(
+  <vscale x 8 x i64>,
+  <vscale x 8 x i32>,
+  <vscale x 8 x i32>,
+  <vscale x 8 x i1>,
+  i32);
+
+define <vscale x 8 x i64> @intrinsic_vwsubu_mask_vv_nxv8i64_nxv8i32_nxv8i32(<vscale x 8 x i64> %0, <vscale x 8 x i32> %1, <vscale x 8 x i32> %2, <vscale x 8 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwsubu_mask_vv_nxv8i64_nxv8i32_nxv8i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e32,m4,tu,mu
+; CHECK-NEXT:    vwsubu.vv v8, v16, v20, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i64> @llvm.riscv.vwsubu.mask.nxv8i64.nxv8i32.nxv8i32(
+    <vscale x 8 x i64> %0,
+    <vscale x 8 x i32> %1,
+    <vscale x 8 x i32> %2,
+    <vscale x 8 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 8 x i64> %a
+}
+
 declare <vscale x 1 x i16> @llvm.riscv.vwsubu.nxv1i16.nxv1i8.i8(
   <vscale x 1 x i8>,
   i8,
   i32);
 
 define <vscale x 1 x i16> @intrinsic_vwsubu_vx_nxv1i16_nxv1i8_i8(<vscale x 1 x i8> %0, i8 %1, i32 %2) nounwind {
 ; CHECK-LABEL: intrinsic_vwsubu_vx_nxv1i16_nxv1i8_i8:
 ; CHECK:       # %bb.0: # %entry
%a = call <vscale x 16 x i32> @llvm.riscv.vwsubu.mask.nxv16i32.nxv16i16.i16(
     <vscale x 16 x i32> %0,
     <vscale x 16 x i16> %1,
     i16 %2,
     <vscale x 16 x i1> %3,
     i32 %4)
 
   ret <vscale x 16 x i32> %a
 }
+
+declare <vscale x 1 x i64> @llvm.riscv.vwsubu.nxv1i64.nxv1i32.i32(
+  <vscale x 1 x i32>,
+  i32,
+  i32);
+
+define <vscale x 1 x i64> @intrinsic_vwsubu_vx_nxv1i64_nxv1i32_i32(<vscale x 1 x i32> %0, i32 %1, i32 %2) nounwind {
+; CHECK-LABEL: intrinsic_vwsubu_vx_nxv1i64_nxv1i32_i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e32,mf2,ta,mu
+; CHECK-NEXT:    vwsubu.vx v25, v8, a0
+; CHECK-NEXT:    vmv1r.v v8, v25
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i64> @llvm.riscv.vwsubu.nxv1i64.nxv1i32.i32(
+    <vscale x 1 x i32> %0,
+    i32 %1,
+    i32 %2)
+
+  ret <vscale x 1 x i64> %a
+}
+
+declare <vscale x 1 x i64> @llvm.riscv.vwsubu.mask.nxv1i64.nxv1i32.i32(
+  <vscale x 1 x i64>,
+  <vscale x 1 x i32>,
+  i32,
+  <vscale x 1 x i1>,
+  i32);
+
+define <vscale x 1 x i64> @intrinsic_vwsubu_mask_vx_nxv1i64_nxv1i32_i32(<vscale x 1 x i64> %0, <vscale x 1 x i32> %1, i32 %2, <vscale x 1 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwsubu_mask_vx_nxv1i64_nxv1i32_i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e32,mf2,tu,mu
+; CHECK-NEXT:    vwsubu.vx v8, v9, a0, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i64> @llvm.riscv.vwsubu.mask.nxv1i64.nxv1i32.i32(
+    <vscale x 1 x i64> %0,
+    <vscale x 1 x i32> %1,
+    i32 %2,
+    <vscale x 1 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 1 x i64> %a
+}
+
+declare <vscale x 2 x i64> @llvm.riscv.vwsubu.nxv2i64.nxv2i32.i32(
+  <vscale x 2 x i32>,
+  i32,
+  i32);
+
+define <vscale x 2 x i64> @intrinsic_vwsubu_vx_nxv2i64_nxv2i32_i32(<vscale x 2 x i32> %0, i32 %1, i32 %2) nounwind {
+; CHECK-LABEL: intrinsic_vwsubu_vx_nxv2i64_nxv2i32_i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e32,m1,ta,mu
+; CHECK-NEXT:    vwsubu.vx v26, v8, a0
+; CHECK-NEXT:    vmv2r.v v8, v26
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i64> @llvm.riscv.vwsubu.nxv2i64.nxv2i32.i32(
+    <vscale x 2 x i32> %0,
+    i32 %1,
+    i32 %2)
+
+  ret <vscale x 2 x i64> %a
+}
+
+declare <vscale x 2 x i64> @llvm.riscv.vwsubu.mask.nxv2i64.nxv2i32.i32(
+  <vscale x 2 x i64>,
+  <vscale x 2 x i32>,
+  i32,
+  <vscale x 2 x i1>,
+  i32);
+
+define <vscale x 2 x i64> @intrinsic_vwsubu_mask_vx_nxv2i64_nxv2i32_i32(<vscale x 2 x i64> %0, <vscale x 2 x i32> %1, i32 %2, <vscale x 2 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwsubu_mask_vx_nxv2i64_nxv2i32_i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e32,m1,tu,mu
+; CHECK-NEXT:    vwsubu.vx v8, v10, a0, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i64> @llvm.riscv.vwsubu.mask.nxv2i64.nxv2i32.i32(
+    <vscale x 2 x i64> %0,
+    <vscale x 2 x i32> %1,
+    i32 %2,
+    <vscale x 2 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 2 x i64> %a
+}
+
+declare <vscale x 4 x i64> @llvm.riscv.vwsubu.nxv4i64.nxv4i32.i32(
+  <vscale x 4 x i32>,
+  i32,
+  i32);
+
+define <vscale x 4 x i64> @intrinsic_vwsubu_vx_nxv4i64_nxv4i32_i32(<vscale x 4 x i32> %0, i32 %1, i32 %2) nounwind {
+; CHECK-LABEL: intrinsic_vwsubu_vx_nxv4i64_nxv4i32_i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e32,m2,ta,mu
+; CHECK-NEXT:    vwsubu.vx v28, v8, a0
+; CHECK-NEXT:    vmv4r.v v8, v28
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i64> @llvm.riscv.vwsubu.nxv4i64.nxv4i32.i32(
+    <vscale x 4 x i32> %0,
+    i32 %1,
+    i32 %2)
+
+  ret <vscale x 4 x i64> %a
+}
+
+declare <vscale x 4 x i64> @llvm.riscv.vwsubu.mask.nxv4i64.nxv4i32.i32(
+  <vscale x 4 x i64>,
+  <vscale x 4 x i32>,
+  i32,
+  <vscale x 4 x i1>,
+  i32);
+
+define <vscale x 4 x i64> @intrinsic_vwsubu_mask_vx_nxv4i64_nxv4i32_i32(<vscale x 4 x i64> %0, <vscale x 4 x i32> %1, i32 %2, <vscale x 4 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwsubu_mask_vx_nxv4i64_nxv4i32_i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e32,m2,tu,mu
+; CHECK-NEXT:    vwsubu.vx v8, v12, a0, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i64> @llvm.riscv.vwsubu.mask.nxv4i64.nxv4i32.i32(
+    <vscale x 4 x i64> %0,
+    <vscale x 4 x i32> %1,
+    i32 %2,
+    <vscale x 4 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 4 x i64> %a
+}
+
+declare <vscale x 8 x i64> @llvm.riscv.vwsubu.nxv8i64.nxv8i32.i32(
+  <vscale x 8 x i32>,
+  i32,
+  i32);
+
+define <vscale x 8 x i64> @intrinsic_vwsubu_vx_nxv8i64_nxv8i32_i32(<vscale x 8 x i32> %0, i32 %1, i32 %2) nounwind {
+; CHECK-LABEL: intrinsic_vwsubu_vx_nxv8i64_nxv8i32_i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e32,m4,ta,mu
+; CHECK-NEXT:    vwsubu.vx v16, v8, a0
+; CHECK-NEXT:    vmv8r.v v8, v16
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i64> @llvm.riscv.vwsubu.nxv8i64.nxv8i32.i32(
+    <vscale x 8 x i32> %0,
+    i32 %1,
+    i32 %2)
+
+  ret <vscale x 8 x i64> %a
+}
+
+declare <vscale x 8 x i64> @llvm.riscv.vwsubu.mask.nxv8i64.nxv8i32.i32(
+  <vscale x 8 x i64>,
+  <vscale x 8 x i32>,
+  i32,
+  <vscale x 8 x i1>,
+  i32);
+
+define <vscale x 8 x i64> @intrinsic_vwsubu_mask_vx_nxv8i64_nxv8i32_i32(<vscale x 8 x i64> %0, <vscale x 8 x i32> %1, i32 %2, <vscale x 8 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwsubu_mask_vx_nxv8i64_nxv8i32_i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e32,m4,tu,mu
+; CHECK-NEXT:    vwsubu.vx v8, v16, a0, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i64> @llvm.riscv.vwsubu.mask.nxv8i64.nxv8i32.i32(
+    <vscale x 8 x i64> %0,
+    <vscale x 8 x i32> %1,
+    i32 %2,
+    <vscale x 8 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 8 x i64> %a
+}

llvm/test/CodeGen/RISCV/rvv/vwsubu.w-rv32.ll

%a = call <vscale x 16 x i32> @llvm.riscv.vwsubu.w.mask.nxv16i32.nxv16i16(
     <vscale x 16 x i32> %1,
     <vscale x 16 x i16> %2,
     <vscale x 16 x i1> %3,
     i32 %4)
 
   ret <vscale x 16 x i32> %a
 }
 
+declare <vscale x 1 x i64> @llvm.riscv.vwsubu.w.nxv1i64.nxv1i32(
+  <vscale x 1 x i64>,
+  <vscale x 1 x i32>,
+  i32);
+
+define <vscale x 1 x i64> @intrinsic_vwsubu.w_wv_nxv1i64_nxv1i64_nxv1i32(<vscale x 1 x i64> %0, <vscale x 1 x i32> %1, i32 %2) nounwind {
+; CHECK-LABEL: intrinsic_vwsubu.w_wv_nxv1i64_nxv1i64_nxv1i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e32,mf2,ta,mu
+; CHECK-NEXT:    vwsubu.wv v25, v8, v9
+; CHECK-NEXT:    vmv1r.v v8, v25
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i64> @llvm.riscv.vwsubu.w.nxv1i64.nxv1i32(
+    <vscale x 1 x i64> %0,
+    <vscale x 1 x i32> %1,
+    i32 %2)
+
+  ret <vscale x 1 x i64> %a
+}
+
+declare <vscale x 1 x i64> @llvm.riscv.vwsubu.w.mask.nxv1i64.nxv1i32(
+  <vscale x 1 x i64>,
+  <vscale x 1 x i64>,
+  <vscale x 1 x i32>,
+  <vscale x 1 x i1>,
+  i32);
+
+define <vscale x 1 x i64> @intrinsic_vwsubu.w_mask_wv_nxv1i64_nxv1i64_nxv1i32(<vscale x 1 x i64> %0, <vscale x 1 x i64> %1, <vscale x 1 x i32> %2, <vscale x 1 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwsubu.w_mask_wv_nxv1i64_nxv1i64_nxv1i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e32,mf2,tu,mu
+; CHECK-NEXT:    vwsubu.wv v8, v9, v10, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i64> @llvm.riscv.vwsubu.w.mask.nxv1i64.nxv1i32(
+    <vscale x 1 x i64> %0,
+    <vscale x 1 x i64> %1,
+    <vscale x 1 x i32> %2,
+    <vscale x 1 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 1 x i64> %a
+}
+
+declare <vscale x 2 x i64> @llvm.riscv.vwsubu.w.nxv2i64.nxv2i32(
+  <vscale x 2 x i64>,
+  <vscale x 2 x i32>,
+  i32);
+
+define <vscale x 2 x i64> @intrinsic_vwsubu.w_wv_nxv2i64_nxv2i64_nxv2i32(<vscale x 2 x i64> %0, <vscale x 2 x i32> %1, i32 %2) nounwind {
+; CHECK-LABEL: intrinsic_vwsubu.w_wv_nxv2i64_nxv2i64_nxv2i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e32,m1,ta,mu
+; CHECK-NEXT:    vwsubu.wv v26, v8, v10
+; CHECK-NEXT:    vmv2r.v v8, v26
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i64> @llvm.riscv.vwsubu.w.nxv2i64.nxv2i32(
+    <vscale x 2 x i64> %0,
+    <vscale x 2 x i32> %1,
+    i32 %2)
+
+  ret <vscale x 2 x i64> %a
+}
+
+declare <vscale x 2 x i64> @llvm.riscv.vwsubu.w.mask.nxv2i64.nxv2i32(
+  <vscale x 2 x i64>,
+  <vscale x 2 x i64>,
+  <vscale x 2 x i32>,
+  <vscale x 2 x i1>,
+  i32);
+
+define <vscale x 2 x i64> @intrinsic_vwsubu.w_mask_wv_nxv2i64_nxv2i64_nxv2i32(<vscale x 2 x i64> %0, <vscale x 2 x i64> %1, <vscale x 2 x i32> %2, <vscale x 2 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwsubu.w_mask_wv_nxv2i64_nxv2i64_nxv2i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e32,m1,tu,mu
+; CHECK-NEXT:    vwsubu.wv v8, v10, v12, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i64> @llvm.riscv.vwsubu.w.mask.nxv2i64.nxv2i32(
+    <vscale x 2 x i64> %0,
+    <vscale x 2 x i64> %1,
+    <vscale x 2 x i32> %2,
+    <vscale x 2 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 2 x i64> %a
+}
+
+declare <vscale x 4 x i64> @llvm.riscv.vwsubu.w.nxv4i64.nxv4i32(
+  <vscale x 4 x i64>,
+  <vscale x 4 x i32>,
+  i32);
+
+define <vscale x 4 x i64> @intrinsic_vwsubu.w_wv_nxv4i64_nxv4i64_nxv4i32(<vscale x 4 x i64> %0, <vscale x 4 x i32> %1, i32 %2) nounwind {
+; CHECK-LABEL: intrinsic_vwsubu.w_wv_nxv4i64_nxv4i64_nxv4i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e32,m2,ta,mu
+; CHECK-NEXT:    vwsubu.wv v28, v8, v12
+; CHECK-NEXT:    vmv4r.v v8, v28
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i64> @llvm.riscv.vwsubu.w.nxv4i64.nxv4i32(
+    <vscale x 4 x i64> %0,
+    <vscale x 4 x i32> %1,
+    i32 %2)
+
+  ret <vscale x 4 x i64> %a
+}
+
+declare <vscale x 4 x i64> @llvm.riscv.vwsubu.w.mask.nxv4i64.nxv4i32(
+  <vscale x 4 x i64>,
+  <vscale x 4 x i64>,
+  <vscale x 4 x i32>,
+  <vscale x 4 x i1>,
+  i32);
+
+define <vscale x 4 x i64> @intrinsic_vwsubu.w_mask_wv_nxv4i64_nxv4i64_nxv4i32(<vscale x 4 x i64> %0, <vscale x 4 x i64> %1, <vscale x 4 x i32> %2, <vscale x 4 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwsubu.w_mask_wv_nxv4i64_nxv4i64_nxv4i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e32,m2,tu,mu
+; CHECK-NEXT:    vwsubu.wv v8, v12, v16, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i64> @llvm.riscv.vwsubu.w.mask.nxv4i64.nxv4i32(
+    <vscale x 4 x i64> %0,
+    <vscale x 4 x i64> %1,
+    <vscale x 4 x i32> %2,
+    <vscale x 4 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 4 x i64> %a
+}
+
+declare <vscale x 8 x i64> @llvm.riscv.vwsubu.w.nxv8i64.nxv8i32(
+  <vscale x 8 x i64>,
+  <vscale x 8 x i32>,
+  i32);
+
+define <vscale x 8 x i64> @intrinsic_vwsubu.w_wv_nxv8i64_nxv8i64_nxv8i32(<vscale x 8 x i64> %0, <vscale x 8 x i32> %1, i32 %2) nounwind {
+; CHECK-LABEL: intrinsic_vwsubu.w_wv_nxv8i64_nxv8i64_nxv8i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a0, a0, e32,m4,ta,mu
+; CHECK-NEXT:    vwsubu.wv v24, v8, v16
+; CHECK-NEXT:    vmv8r.v v8, v24
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i64> @llvm.riscv.vwsubu.w.nxv8i64.nxv8i32(
+    <vscale x 8 x i64> %0,
+    <vscale x 8 x i32> %1,
+    i32 %2)
+
+  ret <vscale x 8 x i64> %a
+}
+
+declare <vscale x 8 x i64> @llvm.riscv.vwsubu.w.mask.nxv8i64.nxv8i32(
+  <vscale x 8 x i64>,
+  <vscale x 8 x i64>,
+  <vscale x 8 x i32>,
+  <vscale x 8 x i1>,
+  i32);
+
+define <vscale x 8 x i64> @intrinsic_vwsubu.w_mask_wv_nxv8i64_nxv8i64_nxv8i32(<vscale x 8 x i64> %0, <vscale x 8 x i64> %1, <vscale x 8 x i32> %2, <vscale x 8 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwsubu.w_mask_wv_nxv8i64_nxv8i64_nxv8i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vl4re32.v v28, (a0)
+; CHECK-NEXT:    vsetvli a0, a1, e32,m4,tu,mu
+; CHECK-NEXT:    vwsubu.wv v8, v16, v28, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i64> @llvm.riscv.vwsubu.w.mask.nxv8i64.nxv8i32(
+    <vscale x 8 x i64> %0,
+    <vscale x 8 x i64> %1,
+    <vscale x 8 x i32> %2,
+    <vscale x 8 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 8 x i64> %a
+}
+
 declare <vscale x 1 x i16> @llvm.riscv.vwsubu.w.nxv1i16.i8(
   <vscale x 1 x i16>,
   i8,
   i32);
 
 define <vscale x 1 x i16> @intrinsic_vwsubu.w_wx_nxv1i16_nxv1i16_i8(<vscale x 1 x i16> %0, i8 %1, i32 %2) nounwind {
 ; CHECK-LABEL: intrinsic_vwsubu.w_wx_nxv1i16_nxv1i16_i8:
 ; CHECK:       # %bb.0: # %entry
%a = call <vscale x 16 x i32> @llvm.riscv.vwsubu.w.mask.nxv16i32.i16(
     <vscale x 16 x i32> %0,
     <vscale x 16 x i32> %1,
     i16 %2,
     <vscale x 16 x i1> %3,
     i32 %4)
 
   ret <vscale x 16 x i32> %a
 }
+
+declare <vscale x 1 x i64> @llvm.riscv.vwsubu.w.nxv1i64.i32(
+  <vscale x 1 x i64>,
+  i32,
+  i32);
+
+define <vscale x 1 x i64> @intrinsic_vwsubu.w_wx_nxv1i64_nxv1i64_i32(<vscale x 1 x i64> %0, i32 %1, i32 %2) nounwind {
+; CHECK-LABEL: intrinsic_vwsubu.w_wx_nxv1i64_nxv1i64_i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e32,mf2,ta,mu
+; CHECK-NEXT:    vwsubu.wx v25, v8, a0
+; CHECK-NEXT:    vmv1r.v v8, v25
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i64> @llvm.riscv.vwsubu.w.nxv1i64.i32(
+    <vscale x 1 x i64> %0,
+    i32 %1,
+    i32 %2)
+
+  ret <vscale x 1 x i64> %a
+}
+
+declare <vscale x 1 x i64> @llvm.riscv.vwsubu.w.mask.nxv1i64.i32(
+  <vscale x 1 x i64>,
+  <vscale x 1 x i64>,
+  i32,
+  <vscale x 1 x i1>,
+  i32);
+
+define <vscale x 1 x i64> @intrinsic_vwsubu.w_mask_wx_nxv1i64_nxv1i64_i32(<vscale x 1 x i64> %0, <vscale x 1 x i64> %1, i32 %2, <vscale x 1 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwsubu.w_mask_wx_nxv1i64_nxv1i64_i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e32,mf2,tu,mu
+; CHECK-NEXT:    vwsubu.wx v8, v9, a0, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 1 x i64> @llvm.riscv.vwsubu.w.mask.nxv1i64.i32(
+    <vscale x 1 x i64> %0,
+    <vscale x 1 x i64> %1,
+    i32 %2,
+    <vscale x 1 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 1 x i64> %a
+}
+
+declare <vscale x 2 x i64> @llvm.riscv.vwsubu.w.nxv2i64.i32(
+  <vscale x 2 x i64>,
+  i32,
+  i32);
+
+define <vscale x 2 x i64> @intrinsic_vwsubu.w_wx_nxv2i64_nxv2i64_i32(<vscale x 2 x i64> %0, i32 %1, i32 %2) nounwind {
+; CHECK-LABEL: intrinsic_vwsubu.w_wx_nxv2i64_nxv2i64_i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e32,m1,ta,mu
+; CHECK-NEXT:    vwsubu.wx v26, v8, a0
+; CHECK-NEXT:    vmv2r.v v8, v26
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i64> @llvm.riscv.vwsubu.w.nxv2i64.i32(
+    <vscale x 2 x i64> %0,
+    i32 %1,
+    i32 %2)
+
+  ret <vscale x 2 x i64> %a
+}
+
+declare <vscale x 2 x i64> @llvm.riscv.vwsubu.w.mask.nxv2i64.i32(
+  <vscale x 2 x i64>,
+  <vscale x 2 x i64>,
+  i32,
+  <vscale x 2 x i1>,
+  i32);
+
+define <vscale x 2 x i64> @intrinsic_vwsubu.w_mask_wx_nxv2i64_nxv2i64_i32(<vscale x 2 x i64> %0, <vscale x 2 x i64> %1, i32 %2, <vscale x 2 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwsubu.w_mask_wx_nxv2i64_nxv2i64_i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e32,m1,tu,mu
+; CHECK-NEXT:    vwsubu.wx v8, v10, a0, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 2 x i64> @llvm.riscv.vwsubu.w.mask.nxv2i64.i32(
+    <vscale x 2 x i64> %0,
+    <vscale x 2 x i64> %1,
+    i32 %2,
+    <vscale x 2 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 2 x i64> %a
+}
+
+declare <vscale x 4 x i64> @llvm.riscv.vwsubu.w.nxv4i64.i32(
+  <vscale x 4 x i64>,
+  i32,
+  i32);
+
+define <vscale x 4 x i64> @intrinsic_vwsubu.w_wx_nxv4i64_nxv4i64_i32(<vscale x 4 x i64> %0, i32 %1, i32 %2) nounwind {
+; CHECK-LABEL: intrinsic_vwsubu.w_wx_nxv4i64_nxv4i64_i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e32,m2,ta,mu
+; CHECK-NEXT:    vwsubu.wx v28, v8, a0
+; CHECK-NEXT:    vmv4r.v v8, v28
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i64> @llvm.riscv.vwsubu.w.nxv4i64.i32(
+    <vscale x 4 x i64> %0,
+    i32 %1,
+    i32 %2)
+
+  ret <vscale x 4 x i64> %a
+}
+
+declare <vscale x 4 x i64> @llvm.riscv.vwsubu.w.mask.nxv4i64.i32(
+  <vscale x 4 x i64>,
+  <vscale x 4 x i64>,
+  i32,
+  <vscale x 4 x i1>,
+  i32);
+
+define <vscale x 4 x i64> @intrinsic_vwsubu.w_mask_wx_nxv4i64_nxv4i64_i32(<vscale x 4 x i64> %0, <vscale x 4 x i64> %1, i32 %2, <vscale x 4 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwsubu.w_mask_wx_nxv4i64_nxv4i64_i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e32,m2,tu,mu
+; CHECK-NEXT:    vwsubu.wx v8, v12, a0, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 4 x i64> @llvm.riscv.vwsubu.w.mask.nxv4i64.i32(
+    <vscale x 4 x i64> %0,
+    <vscale x 4 x i64> %1,
+    i32 %2,
+    <vscale x 4 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 4 x i64> %a
+}
+
+declare <vscale x 8 x i64> @llvm.riscv.vwsubu.w.nxv8i64.i32(
+  <vscale x 8 x i64>,
+  i32,
+  i32);
+
+define <vscale x 8 x i64> @intrinsic_vwsubu.w_wx_nxv8i64_nxv8i64_i32(<vscale x 8 x i64> %0, i32 %1, i32 %2) nounwind {
+; CHECK-LABEL: intrinsic_vwsubu.w_wx_nxv8i64_nxv8i64_i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e32,m4,ta,mu
+; CHECK-NEXT:    vwsubu.wx v16, v8, a0
+; CHECK-NEXT:    vmv8r.v v8, v16
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i64> @llvm.riscv.vwsubu.w.nxv8i64.i32(
+    <vscale x 8 x i64> %0,
+    i32 %1,
+    i32 %2)
+
+  ret <vscale x 8 x i64> %a
+}
+
+declare <vscale x 8 x i64> @llvm.riscv.vwsubu.w.mask.nxv8i64.i32(
+  <vscale x 8 x i64>,
+  <vscale x 8 x i64>,
+  i32,
+  <vscale x 8 x i1>,
+  i32);
+
+define <vscale x 8 x i64> @intrinsic_vwsubu.w_mask_wx_nxv8i64_nxv8i64_i32(<vscale x 8 x i64> %0, <vscale x 8 x i64> %1, i32 %2, <vscale x 8 x i1> %3, i32 %4) nounwind {
+; CHECK-LABEL: intrinsic_vwsubu.w_mask_wx_nxv8i64_nxv8i64_i32:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a1, a1, e32,m4,tu,mu
+; CHECK-NEXT:    vwsubu.wx v8, v16, a0, v0.t
+; CHECK-NEXT:    jalr zero, 0(ra)
+entry:
+  %a = call <vscale x 8 x i64> @llvm.riscv.vwsubu.w.mask.nxv8i64.i32(
+    <vscale x 8 x i64> %0,
+    <vscale x 8 x i64> %1,
+    i32 %2,
+    <vscale x 8 x i1> %3,
+    i32 %4)
+
+  ret <vscale x 8 x i64> %a
+}