[VP][RISCV] Add vp.nearbyint and RISC-V support.

[fakepaper56]

Summary

nearbyint has the property to execute without exception.
For not modifying fflags, the patch added new machine opcode
PseudoVFROUND_NOEXCEPT_V that expands vfcvt.x.f.v and vfcvt.f.x.v between a pair
of frflags and fsflags.

[fakepaper56]

Refine comment.

[craig.topper]

llvm/lib/Target/RISCV/RISCVISelLowering.cpp

10819

an -> a

10832

an -> a

[fakepaper56]

Fix typoes.

[fakepaper56]

Add nearbyint into VPIntrinsicTest.cpp

[craig.topper]

LGTM

llvm/docs/LangRef.rst

 type.
 
 Semantics:
 """"""""""
 
 This function returns the same values as the libm ``rint`` functions
 would, and handles error conditions in the same way.
 
+.. _int_nearbyint:
+
 '``llvm.nearbyint.*``' Intrinsic
 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
 
 Syntax:
 """""""
 
 This is an overloaded intrinsic. You can use ``llvm.nearbyint`` on any
 floating-point or vector of floating-point type. Not all targets support

 .. code-block:: llvm
 
       %r = call <4 x float> @llvm.vp.rint.v4f32(<4 x float> %a, <4 x i1> %mask, i32 %evl)
       ;; For all lanes below %evl, %r is lane-wise equivalent to %also.r
 
       %t = call <4 x float> @llvm.rint.v4f32(<4 x float> %a)
       %also.r = select <4 x i1> %mask, <4 x float> %t, <4 x float> poison
 
+.. _int_vp_nearbyint:
+
+'``llvm.vp.nearbyint.*``' Intrinsics
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+Syntax:
+"""""""
+This is an overloaded intrinsic.
+
+::
+
+      declare <16 x float>  @llvm.vp.nearbyint.v16f32 (<16 x float> <op>, <16 x i1> <mask>, i32 <vector_length>)
+      declare <vscale x 4 x float>  @llvm.vp.nearbyint.nxv4f32 (<vscale x 4 x float> <op>, <vscale x 4 x i1> <mask>, i32 <vector_length>)
+      declare <256 x double>  @llvm.vp.nearbyint.v256f64 (<256 x double> <op>, <256 x i1> <mask>, i32 <vector_length>)
+
+Overview:
+"""""""""
+
+Predicated floating-point nearbyint of a vector of floating-point values.
+
+
+Arguments:
+""""""""""
+
+The first operand and the result have the same vector of floating-point type.
+The second operand is the vector mask and has the same number of elements as the
+result vector type. The third operand is the explicit vector length of the
+operation.
+
+Semantics:
+""""""""""
+
+The '``llvm.vp.nearbyint``' intrinsic performs floating-point nearbyint
+(:ref:`nearbyint <int_nearbyint>`) of the first vector operand on each enabled lane.
+The result on disabled lanes is a :ref:`poison value <poisonvalues>`.
+
+Examples:
+"""""""""
+
+.. code-block:: llvm
+
+      %r = call <4 x float> @llvm.vp.nearbyint.v4f32(<4 x float> %a, <4 x i1> %mask, i32 %evl)
+      ;; For all lanes below %evl, %r is lane-wise equivalent to %also.r
+
+      %t = call <4 x float> @llvm.nearbyint.v4f32(<4 x float> %a)
+      %also.r = select <4 x i1> %mask, <4 x float> %t, <4 x float> poison
+
 .. _int_vp_round:
 
 '``llvm.vp.round.*``' Intrinsics
 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
 
 Syntax:
 """""""
 This is an overloaded intrinsic.

llvm/include/llvm/IR/Intrinsics.td

let IntrProperties = [IntrNoMem, IntrNoSync, IntrWillReturn] in {
   def int_vp_roundtozero : DefaultAttrsIntrinsic<[ llvm_anyvector_ty ],
                              [ LLVMMatchType<0>,
                                LLVMScalarOrSameVectorWidth<0, llvm_i1_ty>,
                                llvm_i32_ty]>;
   def int_vp_rint : DefaultAttrsIntrinsic<[ llvm_anyvector_ty ],
                              [ LLVMMatchType<0>,
                                LLVMScalarOrSameVectorWidth<0, llvm_i1_ty>,
                                llvm_i32_ty]>;
+  def int_vp_nearbyint : DefaultAttrsIntrinsic<[ llvm_anyvector_ty ],
+                             [ LLVMMatchType<0>,
+                               LLVMScalarOrSameVectorWidth<0, llvm_i1_ty>,
+                               llvm_i32_ty]>;
 
   // Casts
   def int_vp_trunc : DefaultAttrsIntrinsic<[ llvm_anyvector_ty ],
                              [ llvm_anyvector_ty,
                                LLVMScalarOrSameVectorWidth<0, llvm_i1_ty>,
                                llvm_i32_ty]>;
   def int_vp_zext : DefaultAttrsIntrinsic<[ llvm_anyvector_ty ],
                              [ llvm_anyvector_ty,

llvm/include/llvm/IR/VPIntrinsics.def

 // llvm.vp.roundtozero(x,mask,vlen)
 BEGIN_REGISTER_VP(vp_roundtozero, 1, 2, VP_FROUNDTOZERO, -1)
 END_REGISTER_VP(vp_roundtozero, VP_FROUNDTOZERO)
 
 // llvm.vp.rint(x,mask,vlen)
 BEGIN_REGISTER_VP(vp_rint, 1, 2, VP_FRINT, -1)
 END_REGISTER_VP(vp_rint, VP_FRINT)
 
+// llvm.vp.nearbyint(x,mask,vlen)
+BEGIN_REGISTER_VP(vp_nearbyint, 1, 2, VP_FNEARBYINT, -1)
+END_REGISTER_VP(vp_nearbyint, VP_FNEARBYINT)
+
 ///// } Floating-Point Arithmetic
 
 ///// Type Casts {
 // Specialized helper macro for type conversions.
 // <operation>(%x, %mask, %evl).
 #ifdef HELPER_REGISTER_FP_CAST_VP
 #error                                                                         \
     "The internal helper macro HELPER_REGISTER_FP_CAST_VP is already defined!"

llvm/lib/CodeGen/SelectionDAG/LegalizeVectorTypes.cpp

#endif
   case ISD::FEXP:
   case ISD::FEXP2:
   case ISD::FFLOOR:
   case ISD::VP_FFLOOR:
   case ISD::FLOG:
   case ISD::FLOG10:
   case ISD::FLOG2:
   case ISD::FNEARBYINT:
+  case ISD::VP_FNEARBYINT:
   case ISD::FNEG: case ISD::VP_FNEG:
   case ISD::FREEZE:
   case ISD::ARITH_FENCE:
   case ISD::FP_EXTEND:
   case ISD::VP_FP_EXTEND:
   case ISD::FP_ROUND:
   case ISD::VP_FP_ROUND:
   case ISD::FP_TO_SINT:
#include "llvm/IR/ConstrainedOps.def"
   case ISD::CTTZ:
   case ISD::CTTZ_ZERO_UNDEF:
   case ISD::FNEG: case ISD::VP_FNEG:
   case ISD::VP_FABS:
   case ISD::VP_SQRT:
   case ISD::VP_FCEIL:
   case ISD::VP_FFLOOR:
   case ISD::VP_FRINT:
+  case ISD::VP_FNEARBYINT:
   case ISD::VP_FROUND:
   case ISD::VP_FROUNDEVEN:
   case ISD::VP_FROUNDTOZERO:
   case ISD::FREEZE:
   case ISD::ARITH_FENCE:
   case ISD::FCANONICALIZE:
     Res = WidenVecRes_Unary(N);
     break;

llvm/lib/Target/RISCV/RISCVISelLowering.h

enum NodeType : unsigned {
   // Vector unary ops with a mask as a second operand and VL as a third operand.
   FNEG_VL,
   FABS_VL,
   FSQRT_VL,
   FCOPYSIGN_VL, // Has a merge operand
   VFCVT_RTZ_X_F_VL,
   VFCVT_RTZ_XU_F_VL,
   VFCVT_X_F_VL,
+  VFROUND_NOEXCEPT_VL,
   VFCVT_RM_X_F_VL, // Has a rounding mode operand.
   SINT_TO_FP_VL,
   UINT_TO_FP_VL,
   FP_ROUND_VL,
   FP_EXTEND_VL,
 
   // Vector FMA ops with a mask as a fourth operand and VL as a fifth operand.
   VFMADD_VL,

llvm/lib/Target/RISCV/RISCVISelLowering.cpp

static const unsigned FloatingPointVPOps[] = {
         ISD::VP_FDIV,        ISD::VP_FNEG,        ISD::VP_FABS,
         ISD::VP_FMA,         ISD::VP_REDUCE_FADD, ISD::VP_REDUCE_SEQ_FADD,
         ISD::VP_REDUCE_FMIN, ISD::VP_REDUCE_FMAX, ISD::VP_MERGE,
         ISD::VP_SELECT,      ISD::VP_SINT_TO_FP,  ISD::VP_UINT_TO_FP,
         ISD::VP_SETCC,       ISD::VP_FP_ROUND,    ISD::VP_FP_EXTEND,
         ISD::VP_SQRT,        ISD::VP_FMINNUM,     ISD::VP_FMAXNUM,
         ISD::VP_FCEIL,       ISD::VP_FFLOOR,      ISD::VP_FROUND,
         ISD::VP_FROUNDEVEN,  ISD::VP_FCOPYSIGN,   ISD::VP_FROUNDTOZERO,
-        ISD::VP_FRINT};
+        ISD::VP_FRINT,       ISD::VP_FNEARBYINT};
 
     static const unsigned IntegerVecReduceOps[] = {
         ISD::VECREDUCE_ADD,  ISD::VECREDUCE_AND,  ISD::VECREDUCE_OR,
         ISD::VECREDUCE_XOR,  ISD::VECREDUCE_SMAX, ISD::VECREDUCE_SMIN,
         ISD::VECREDUCE_UMAX, ISD::VECREDUCE_UMIN};
 
     static const unsigned FloatingPointVecReduceOps[] = {
         ISD::VECREDUCE_FADD, ISD::VECREDUCE_SEQ_FADD, ISD::VECREDUCE_FMIN,
static RISCVFPRndMode::RoundingMode matchRoundingOp(unsigned Opc) {
   case ISD::FRINT:
     return RISCVFPRndMode::DYN;
   }
 
   return RISCVFPRndMode::Invalid;
 }
 
 // Expand vector FTRUNC, FCEIL, FFLOOR, FROUND, VP_FCEIL, VP_FFLOOR, VP_FROUND
-// VP_FROUNDEVEN, VP_FROUNDTOZERO and VP_FRINT by converting to the integer
-// domain and back. Taking care to avoid converting values that are nan or
-// already correct.
+// VP_FROUNDEVEN, VP_FROUNDTOZERO, VP_FRINT and VP_FNEARBYINT by converting to
+// the integer domain and back. Taking care to avoid converting values that are
+// nan or already correct.
 static SDValue
 lowerVectorFTRUNC_FCEIL_FFLOOR_FROUND(SDValue Op, SelectionDAG &DAG,
                                       const RISCVSubtarget &Subtarget) {
   MVT VT = Op.getSimpleValueType();
   assert(VT.isVector() && "Unexpected type");
 
   SDLoc DL(Op);
 
lowerVectorFTRUNC_FCEIL_FFLOOR_FROUND(SDValue Op, SelectionDAG &DAG,
   }
   case ISD::FTRUNC:
     Truncated = DAG.getNode(RISCVISD::VFCVT_RTZ_X_F_VL, DL, IntVT, Src,
                             Mask, VL);
     break;
   case ISD::VP_FRINT:
     Truncated = DAG.getNode(RISCVISD::VFCVT_X_F_VL, DL, IntVT, Src, Mask, VL);
     break;
+  case ISD::VP_FNEARBYINT:
+    Truncated = DAG.getNode(RISCVISD::VFROUND_NOEXCEPT_VL, DL, ContainerVT, Src,
+                            Mask, VL);
+    break;
   }
 
+  // VFROUND_NOEXCEPT_VL includes SINT_TO_FP_VL.
+  if (Op.getOpcode() != ISD::VP_FNEARBYINT)
     Truncated = DAG.getNode(RISCVISD::SINT_TO_FP_VL, DL, ContainerVT, Truncated,
                             Mask, VL);
 
   // Restore the original sign so that -0.0 is preserved.
   Truncated = DAG.getNode(RISCVISD::FCOPYSIGN_VL, DL, ContainerVT, Truncated,
                           Src, Src, Mask, VL);
 
   if (!VT.isFixedLengthVector())
     return Truncated;
 
case ISD::VP_UMAX:
     return lowerVPOp(Op, DAG, RISCVISD::UMAX_VL, /*HasMergeOp*/ true);
   case ISD::EXPERIMENTAL_VP_STRIDED_LOAD:
     return lowerVPStridedLoad(Op, DAG);
   case ISD::EXPERIMENTAL_VP_STRIDED_STORE:
     return lowerVPStridedStore(Op, DAG);
   case ISD::VP_FCEIL:
   case ISD::VP_FFLOOR:
   case ISD::VP_FRINT:
+  case ISD::VP_FNEARBYINT:
   case ISD::VP_FROUND:
   case ISD::VP_FROUNDEVEN:
   case ISD::VP_FROUNDTOZERO:
     return lowerVectorFTRUNC_FCEIL_FFLOOR_FROUND(Op, DAG, Subtarget);
   }
 }
 
 static SDValue getTargetNode(GlobalAddressSDNode *N, SDLoc DL, EVT Ty,
emitVFCVT_RM_MASK(MachineInstr &MI, MachineBasicBlock *BB, unsigned Opcode) {
   BuildMI(*BB, MI, DL, TII.get(RISCV::WriteFRM))
       .addReg(SavedFRM, RegState::Kill);
 
   // Erase the pseudoinstruction.
   MI.eraseFromParent();
   return BB;
 }
 
+static MachineBasicBlock *emitVFROUND_NOEXCEPT_MASK(MachineInstr &MI,
+                                                    MachineBasicBlock *BB,
+                                                    unsigned CVTXOpc,
+                                                    unsigned CVTFOpc) {
+  DebugLoc DL = MI.getDebugLoc();
+
+  const TargetInstrInfo &TII = *BB->getParent()->getSubtarget().getInstrInfo();
+
+  MachineRegisterInfo &MRI = BB->getParent()->getRegInfo();
+  Register SavedFFLAGS = MRI.createVirtualRegister(&RISCV::GPRRegClass);
+
+  // Save the old value of FFLAGS.
+  BuildMI(*BB, MI, DL, TII.get(RISCV::ReadFFLAGS), SavedFFLAGS);
+
+  assert(MI.getNumOperands() == 7);
+
+  // Emit a VFCVT_X_F
+  const TargetRegisterInfo *TRI =
+      BB->getParent()->getSubtarget().getRegisterInfo();
+  const TargetRegisterClass *RC = MI.getRegClassConstraint(0, &TII, TRI);
+  Register Tmp = MRI.createVirtualRegister(RC);
+  BuildMI(*BB, MI, DL, TII.get(CVTXOpc), Tmp)
+      .add(MI.getOperand(1))
+      .add(MI.getOperand(2))
+      .add(MI.getOperand(3))
+      .add(MI.getOperand(4))
+      .add(MI.getOperand(5))
+      .add(MI.getOperand(6));
+
+  // Emit a VFCVT_F_X
+  BuildMI(*BB, MI, DL, TII.get(CVTFOpc))
+      .add(MI.getOperand(0))
+      .add(MI.getOperand(1))
+      .addReg(Tmp)
+      .add(MI.getOperand(3))
+      .add(MI.getOperand(4))
+      .add(MI.getOperand(5))
+      .add(MI.getOperand(6));
+
+  // Restore FFLAGS.
+  BuildMI(*BB, MI, DL, TII.get(RISCV::WriteFFLAGS))
+      .addReg(SavedFFLAGS, RegState::Kill);
+
+  // Erase the pseudoinstruction.
+  MI.eraseFromParent();
+  return BB;
+}
+
 static MachineBasicBlock *emitFROUND(MachineInstr &MI, MachineBasicBlock *MBB,
                                      const RISCVSubtarget &Subtarget) {
   unsigned CmpOpc, F2IOpc, I2FOpc, FSGNJOpc, FSGNJXOpc;
   const TargetRegisterClass *RC;
   switch (MI.getOpcode()) {
   default:
     llvm_unreachable("Unexpected opcode");
   case RISCV::PseudoFROUND_H:
RISCVTargetLowering::EmitInstrWithCustomInserter(MachineInstr &MI,
   case RISCV::PseudoVFCVT_RM_X_F_V_M4_MASK:
     return emitVFCVT_RM_MASK(MI, BB, RISCV::PseudoVFCVT_X_F_V_M4_MASK);
   case RISCV::PseudoVFCVT_RM_X_F_V_M8_MASK:
     return emitVFCVT_RM_MASK(MI, BB, RISCV::PseudoVFCVT_X_F_V_M8_MASK);
   case RISCV::PseudoVFCVT_RM_X_F_V_MF2_MASK:
     return emitVFCVT_RM_MASK(MI, BB, RISCV::PseudoVFCVT_X_F_V_MF2_MASK);
   case RISCV::PseudoVFCVT_RM_X_F_V_MF4_MASK:
     return emitVFCVT_RM_MASK(MI, BB, RISCV::PseudoVFCVT_X_F_V_MF4_MASK);
+  case RISCV::PseudoVFROUND_NOEXCEPT_V_M1_MASK:
+    return emitVFROUND_NOEXCEPT_MASK(MI, BB, RISCV::PseudoVFCVT_X_F_V_M1_MASK,
+                                     RISCV::PseudoVFCVT_F_X_V_M1_MASK);
+  case RISCV::PseudoVFROUND_NOEXCEPT_V_M2_MASK:
+    return emitVFROUND_NOEXCEPT_MASK(MI, BB, RISCV::PseudoVFCVT_X_F_V_M2_MASK,
+                                     RISCV::PseudoVFCVT_F_X_V_M2_MASK);
+  case RISCV::PseudoVFROUND_NOEXCEPT_V_M4_MASK:
+    return emitVFROUND_NOEXCEPT_MASK(MI, BB, RISCV::PseudoVFCVT_X_F_V_M4_MASK,
+                                     RISCV::PseudoVFCVT_F_X_V_M4_MASK);
+  case RISCV::PseudoVFROUND_NOEXCEPT_V_M8_MASK:
+    return emitVFROUND_NOEXCEPT_MASK(MI, BB, RISCV::PseudoVFCVT_X_F_V_M8_MASK,
+                                     RISCV::PseudoVFCVT_F_X_V_M8_MASK);
+  case RISCV::PseudoVFROUND_NOEXCEPT_V_MF2_MASK:
+    return emitVFROUND_NOEXCEPT_MASK(MI, BB, RISCV::PseudoVFCVT_X_F_V_MF2_MASK,
+                                     RISCV::PseudoVFCVT_F_X_V_MF2_MASK);
+  case RISCV::PseudoVFROUND_NOEXCEPT_V_MF4_MASK:
+    return emitVFROUND_NOEXCEPT_MASK(MI, BB, RISCV::PseudoVFCVT_X_F_V_MF4_MASK,
+                                     RISCV::PseudoVFCVT_F_X_V_MF4_MASK);
   case RISCV::PseudoFROUND_H:
   case RISCV::PseudoFROUND_S:
   case RISCV::PseudoFROUND_D:
     return emitFROUND(MI, BB, Subtarget);
   }
 }
 
 void RISCVTargetLowering::AdjustInstrPostInstrSelection(MachineInstr &MI,
#define NODE_NAME_CASE(NODE)                                                   \
   NODE_NAME_CASE(FMINNUM_VL)
   NODE_NAME_CASE(FMAXNUM_VL)
   NODE_NAME_CASE(MULHS_VL)
   NODE_NAME_CASE(MULHU_VL)
   NODE_NAME_CASE(VFCVT_RTZ_X_F_VL)
   NODE_NAME_CASE(VFCVT_RTZ_XU_F_VL)
   NODE_NAME_CASE(VFCVT_RM_X_F_VL)
   NODE_NAME_CASE(VFCVT_X_F_VL)
+  NODE_NAME_CASE(VFROUND_NOEXCEPT_VL)
   NODE_NAME_CASE(SINT_TO_FP_VL)
   NODE_NAME_CASE(UINT_TO_FP_VL)
   NODE_NAME_CASE(FP_EXTEND_VL)
   NODE_NAME_CASE(FP_ROUND_VL)
   NODE_NAME_CASE(VWMUL_VL)
   NODE_NAME_CASE(VWMULU_VL)
   NODE_NAME_CASE(VWMULSU_VL)
   NODE_NAME_CASE(VWADD_VL)

llvm/lib/Target/RISCV/RISCVInstrInfoVPseudos.td

class VPseudoUnaryMaskTA<VReg RetClass, VReg OpClass, string Constraint = ""> :
   let Constraints = Join<[Constraint, "$rd = $merge"], ",">.ret;
   let HasVLOp = 1;
   let HasSEWOp = 1;
   let HasMergeOp = 1;
   let HasVecPolicyOp = 1;
   let UsesMaskPolicy = 1;
 }
 
+class VPseudoUnaryMaskTA_NoExcept<VReg RetClass, VReg OpClass, string Constraint = ""> :
+        Pseudo<(outs GetVRegNoV0<RetClass>.R:$rd),
+               (ins GetVRegNoV0<RetClass>.R:$merge, OpClass:$rs2, VMaskOp:$vm,
+                    AVL:$vl, ixlenimm:$sew, ixlenimm:$policy), []> {
+  let mayLoad = 0;
+  let mayStore = 0;
+  let hasSideEffects = 0;
+  let Constraints = Join<[Constraint, "$rd = $merge"], ",">.ret;
+  let HasVLOp = 1;
+  let HasSEWOp = 1;
+  let HasMergeOp = 1;
+  let HasVecPolicyOp = 1;
+  let UsesMaskPolicy = 1;
+  let usesCustomInserter = 1;
+}
+
 class VPseudoUnaryMaskTA_FRM<VReg RetClass, VReg OpClass, string Constraint = ""> :
         Pseudo<(outs GetVRegNoV0<RetClass>.R:$rd),
                (ins GetVRegNoV0<RetClass>.R:$merge, OpClass:$rs2,
                     VMaskOp:$vm, ixlenimm:$frm, AVL:$vl, ixlenimm:$sew, ixlenimm:$policy), []> {
   let mayLoad = 0;
   let mayStore = 0;
   let hasSideEffects = 0;
   let Constraints = Join<[Constraint, "$rd = $merge"], ",">.ret;
multiclass VPseudoConversionRM<VReg RetClass,
                                LMULInfo MInfo,
                                string Constraint = ""> {
   let VLMul = MInfo.value in {
     def "_" # MInfo.MX # "_MASK" : VPseudoUnaryMaskTA_FRM<RetClass, Op1Class,
                                                           Constraint>;
   }
 }
 
+multiclass VPseudoConversionNoExcept<VReg RetClass,
+                                     VReg Op1Class,
+                                     LMULInfo MInfo,
+                                     string Constraint = ""> {
+  let VLMul = MInfo.value in {
+    def "_" # MInfo.MX # "_MASK" : VPseudoUnaryMaskTA_NoExcept<RetClass, Op1Class, Constraint>;
+  }
+}
+
 multiclass VPseudoVCVTI_V {
   foreach m = MxListF in
     defm _V : VPseudoConversion<m.vrclass, m.vrclass, m>,
               Sched<[WriteVFCvtFToIV, ReadVFCvtFToIV, ReadVMask]>;
 }
 
 multiclass VPseudoVCVTI_RM_V {
   foreach m = MxListF in
     defm _V : VPseudoConversionRM<m.vrclass, m.vrclass, m>,
               Sched<[WriteVFCvtFToIV, ReadVFCvtFToIV, ReadVMask]>;
 }
 
+multiclass VPseudoVFROUND_NOEXCEPT_V {
+  foreach m = MxListF in
+    defm _V : VPseudoConversionNoExcept<m.vrclass, m.vrclass, m>,
+              Sched<[WriteVFCvtFToIV, ReadVFCvtFToIV, ReadVMask]>;
+}
+
 multiclass VPseudoVCVTF_V {
   foreach m = MxListF in
     defm _V : VPseudoConversion<m.vrclass, m.vrclass, m>,
               Sched<[WriteVFCvtIToFV, ReadVFCvtIToFV, ReadVMask]>;
 }
 
 multiclass VPseudoConversionW_V {
   defvar constraint = "@earlyclobber $rd";

 let mayRaiseFPException = true in {
 let Uses = [FRM] in {
 defm PseudoVFCVT_XU_F : VPseudoVCVTI_V;
 defm PseudoVFCVT_X_F : VPseudoVCVTI_V;
 }
 defm PseudoVFCVT_RTZ_XU_F : VPseudoVCVTI_V;
 defm PseudoVFCVT_RTZ_X_F : VPseudoVCVTI_V;
 defm PseudoVFCVT_RM_X_F : VPseudoVCVTI_RM_V;
+defm PseudoVFROUND_NOEXCEPT : VPseudoVFROUND_NOEXCEPT_V;
 let Uses = [FRM] in {
 defm PseudoVFCVT_F_XU : VPseudoVCVTF_V;
 defm PseudoVFCVT_F_X : VPseudoVCVTF_V;
 }
 } // mayRaiseFPException = true
 
 //===----------------------------------------------------------------------===//
 // 14.18. Widening Floating-Point/Integer Type-Convert Instructions

llvm/lib/Target/RISCV/RISCVInstrInfoVVLPatterns.td

 def SDT_RISCVVecCvtX2FOp_VL : SDTypeProfile<1, 4, [
   SDTCisInt<0>, SDTCisFP<1>, SDTCisSameNumEltsAs<0, 1>,
   SDTCVecEltisVT<2, i1>, SDTCisSameNumEltsAs<1, 2>, SDTCisVT<3, XLenVT>,
   SDTCisVT<4, XLenVT>
 ]>;
 
 def riscv_vfcvt_rm_x_f_vl : SDNode<"RISCVISD::VFCVT_RM_X_F_VL", SDT_RISCVVecCvtX2FOp_VL>;
 def riscv_vfcvt_x_f_vl : SDNode<"RISCVISD::VFCVT_X_F_VL", SDT_RISCVFP2IOp_VL>;
+def riscv_vfround_noexcept_vl: SDNode<"RISCVISD::VFROUND_NOEXCEPT_VL", SDT_RISCVFPUnOp_VL>;
 
 def riscv_setcc_vl : SDNode<"RISCVISD::SETCC_VL",
                             SDTypeProfile<1, 6, [SDTCVecEltisVT<0, i1>,
                                                  SDTCisVec<1>,
                                                  SDTCisSameNumEltsAs<0, 1>,
                                                  SDTCisSameAs<1, 2>,
                                                  SDTCisVT<3, OtherVT>,
                                                  SDTCisSameAs<0, 4>,
def : Pat<(riscv_fcopysign_vl (vti.Vector vti.RegClass:$rs1),
                                 (SplatFPOp vti.ScalarRegClass:$rs2),
                                 vti.RegClass:$merge,
                                 (vti.Mask V0),
                                 VLOpFrag),
             (!cast<Instruction>("PseudoVFSGNJ_V"#vti.ScalarSuffix#"_"# vti.LMul.MX#"_MASK")
                  vti.RegClass:$merge, vti.RegClass:$rs1,
                  vti.ScalarRegClass:$rs2, (vti.Mask V0), GPR:$vl, vti.Log2SEW,
                  TAIL_AGNOSTIC)>;
+
+  // Rounding without exception to implement nearbyint.
+  def : Pat<(riscv_vfround_noexcept_vl (vti.Vector vti.RegClass:$rs1),
+                                       (vti.Mask V0), VLOpFrag),
+            (!cast<Instruction>("PseudoVFROUND_NOEXCEPT_V_" # vti.LMul.MX #"_MASK")
+                  (vti.Vector (IMPLICIT_DEF)), vti.RegClass:$rs1,
+                  (vti.Mask V0), GPR:$vl, vti.Log2SEW, TA_MA)>;
 }
 
 foreach fvti = AllFloatVectors in {
   // Floating-point vselects:
   // 12.15. Vector Integer Merge Instructions
   // 14.15. Vector Floating-Point Merge Instruction
   def : Pat<(fvti.Vector (riscv_vselect_vl (fvti.Mask V0),
                                            fvti.RegClass:$rs1,

llvm/lib/Target/RISCV/RISCVTargetTransformInfo.cpp

RISCVTTIImpl::getIntrinsicInstrCost(const IntrinsicCostAttributes &ICA,
   case Intrinsic::vp_rint: {
     // RISC-V target uses at least 5 instructions to lower rounding intrinsics.
     unsigned Cost = 5;
     auto LT = getTypeLegalizationCost(RetTy);
     if (TLI->isOperationCustom(ISD::VP_FRINT, LT.second))
       return Cost * LT.first;
     break;
   }
+  case Intrinsic::vp_nearbyint: {
+    // More one read and one write for fflags than vp_rint.
+    unsigned Cost = 7;
+    auto LT = getTypeLegalizationCost(RetTy);
+    if (TLI->isOperationCustom(ISD::VP_FRINT, LT.second))
+      return Cost * LT.first;
+    break;
+  }
   }
 
   if (ST->hasVInstructions() && RetTy->isVectorTy()) {
     auto LT = getTypeLegalizationCost(RetTy);
     if (const auto *Entry = CostTableLookup(VectorIntrinsicCostTable,
                                             ICA.getID(), LT.second))
       return LT.first * Entry->Cost;
   }

llvm/test/Analysis/CostModel/RISCV/fround.ll

;
   call <16 x double> @llvm.vp.rint.v16f64(<16 x double> undef, <16 x i1> undef, i32 undef)
   call <vscale x 1 x double> @llvm.vp.rint.nvx1f64(<vscale x 1 x double> undef, <vscale x 1 x i1> undef, i32 undef)
   call <vscale x 2 x double> @llvm.vp.rint.nvx2f64(<vscale x 2 x double> undef, <vscale x 2 x i1> undef, i32 undef)
   call <vscale x 4 x double> @llvm.vp.rint.nvx5f64(<vscale x 4 x double> undef, <vscale x 4 x i1> undef, i32 undef)
   call <vscale x 8 x double> @llvm.vp.rint.nvx8f64(<vscale x 8 x double> undef, <vscale x 8 x i1> undef, i32 undef)
   ret void
 }
 
+define void @vp_nearbyint() {
+; CHECK-LABEL: 'vp_nearbyint'
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %1 = call <2 x float> @llvm.vp.nearbyint.v2f32(<2 x float> undef, <2 x i1> undef, i32 undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %2 = call <4 x float> @llvm.vp.nearbyint.v4f32(<4 x float> undef, <4 x i1> undef, i32 undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %3 = call <8 x float> @llvm.vp.nearbyint.v8f32(<8 x float> undef, <8 x i1> undef, i32 undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %4 = call <16 x float> @llvm.vp.nearbyint.v16f32(<16 x float> undef, <16 x i1> undef, i32 undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %5 = call <vscale x 1 x float> @llvm.vp.nearbyint.nxv1f32(<vscale x 1 x float> undef, <vscale x 1 x i1> undef, i32 undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %6 = call <vscale x 2 x float> @llvm.vp.nearbyint.nxv2f32(<vscale x 2 x float> undef, <vscale x 2 x i1> undef, i32 undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %7 = call <vscale x 4 x float> @llvm.vp.nearbyint.nxv4f32(<vscale x 4 x float> undef, <vscale x 4 x i1> undef, i32 undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %8 = call <vscale x 8 x float> @llvm.vp.nearbyint.nxv8f32(<vscale x 8 x float> undef, <vscale x 8 x i1> undef, i32 undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %9 = call <vscale x 16 x float> @llvm.vp.nearbyint.nxv16f32(<vscale x 16 x float> undef, <vscale x 16 x i1> undef, i32 undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %10 = call <2 x double> @llvm.vp.nearbyint.v2f64(<2 x double> undef, <2 x i1> undef, i32 undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %11 = call <4 x double> @llvm.vp.nearbyint.v4f64(<4 x double> undef, <4 x i1> undef, i32 undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %12 = call <8 x double> @llvm.vp.nearbyint.v8f64(<8 x double> undef, <8 x i1> undef, i32 undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %13 = call <16 x double> @llvm.vp.nearbyint.v16f64(<16 x double> undef, <16 x i1> undef, i32 undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %14 = call <vscale x 1 x double> @llvm.vp.nearbyint.nxv1f64(<vscale x 1 x double> undef, <vscale x 1 x i1> undef, i32 undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %15 = call <vscale x 2 x double> @llvm.vp.nearbyint.nxv2f64(<vscale x 2 x double> undef, <vscale x 2 x i1> undef, i32 undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %16 = call <vscale x 4 x double> @llvm.vp.nearbyint.nxv4f64(<vscale x 4 x double> undef, <vscale x 4 x i1> undef, i32 undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %17 = call <vscale x 8 x double> @llvm.vp.nearbyint.nxv8f64(<vscale x 8 x double> undef, <vscale x 8 x i1> undef, i32 undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: ret void
+;
+  call <2 x float> @llvm.vp.nearbyint.v2f32(<2 x float> undef, <2 x i1> undef, i32 undef)
+  call <4 x float> @llvm.vp.nearbyint.v4f32(<4 x float> undef, <4 x i1> undef, i32 undef)
+  call <8 x float> @llvm.vp.nearbyint.v8f32(<8 x float> undef, <8 x i1> undef, i32 undef)
+  call <16 x float> @llvm.vp.nearbyint.v16f32(<16 x float> undef, <16 x i1> undef, i32 undef)
+  call <vscale x 1 x float> @llvm.vp.nearbyint.nvx1f32(<vscale x 1 x float> undef, <vscale x 1 x i1> undef, i32 undef)
+  call <vscale x 2 x float> @llvm.vp.nearbyint.nvx2f32(<vscale x 2 x float> undef, <vscale x 2 x i1> undef, i32 undef)
+  call <vscale x 4 x float> @llvm.vp.nearbyint.nvx4f32(<vscale x 4 x float> undef, <vscale x 4 x i1> undef, i32 undef)
+  call <vscale x 8 x float> @llvm.vp.nearbyint.nvx8f32(<vscale x 8 x float> undef, <vscale x 8 x i1> undef, i32 undef)
+  call <vscale x 16 x float> @llvm.vp.nearbyint.nvx16f32(<vscale x 16 x float> undef, <vscale x 16 x i1> undef, i32 undef)
+  call <2 x double> @llvm.vp.nearbyint.v2f64(<2 x double> undef, <2 x i1> undef, i32 undef)
+  call <4 x double> @llvm.vp.nearbyint.v4f64(<4 x double> undef, <4 x i1> undef, i32 undef)
+  call <8 x double> @llvm.vp.nearbyint.v8f64(<8 x double> undef, <8 x i1> undef, i32 undef)
+  call <16 x double> @llvm.vp.nearbyint.v16f64(<16 x double> undef, <16 x i1> undef, i32 undef)
+  call <vscale x 1 x double> @llvm.vp.nearbyint.nvx1f64(<vscale x 1 x double> undef, <vscale x 1 x i1> undef, i32 undef)
+  call <vscale x 2 x double> @llvm.vp.nearbyint.nvx2f64(<vscale x 2 x double> undef, <vscale x 2 x i1> undef, i32 undef)
+  call <vscale x 4 x double> @llvm.vp.nearbyint.nvx5f64(<vscale x 4 x double> undef, <vscale x 4 x i1> undef, i32 undef)
+  call <vscale x 8 x double> @llvm.vp.nearbyint.nvx8f64(<vscale x 8 x double> undef, <vscale x 8 x i1> undef, i32 undef)
+  ret void
+}
+
 declare float @llvm.floor.f32(float)
 declare <2 x float> @llvm.floor.v2f32(<2 x float>)
 declare <4 x float> @llvm.floor.v4f32(<4 x float>)
 declare <8 x float> @llvm.floor.v8f32(<8 x float>)
 declare <16 x float> @llvm.floor.v16f32(<16 x float>)
 declare <vscale x 1 x float> @llvm.floor.nvx1f32(<vscale x 1 x float>)
 declare <vscale x 2 x float> @llvm.floor.nvx2f32(<vscale x 2 x float>)
 declare <vscale x 4 x float> @llvm.floor.nvx4f32(<vscale x 4 x float>)

 declare <2 x double> @llvm.vp.rint.v2f64(<2 x double>, <2 x i1>, i32)
 declare <4 x double> @llvm.vp.rint.v4f64(<4 x double>, <4 x i1>, i32)
 declare <8 x double> @llvm.vp.rint.v8f64(<8 x double>, <8 x i1>, i32)
 declare <16 x double> @llvm.vp.rint.v16f64(<16 x double>, <16 x i1>, i32)
 declare <vscale x 1 x double> @llvm.vp.rint.nvx1f64(<vscale x 1 x double>, <vscale x 1 x i1>, i32)
 declare <vscale x 2 x double> @llvm.vp.rint.nvx2f64(<vscale x 2 x double>, <vscale x 2 x i1>, i32)
 declare <vscale x 4 x double> @llvm.vp.rint.nvx5f64(<vscale x 4 x double>, <vscale x 4 x i1>, i32)
 declare <vscale x 8 x double> @llvm.vp.rint.nvx8f64(<vscale x 8 x double>, <vscale x 8 x i1>, i32)
+
+declare <2 x float> @llvm.vp.nearbyint.v2f32(<2 x float>, <2 x i1>, i32)
+declare <4 x float> @llvm.vp.nearbyint.v4f32(<4 x float>, <4 x i1>, i32)
+declare <8 x float> @llvm.vp.nearbyint.v8f32(<8 x float>, <8 x i1>, i32)
+declare <16 x float> @llvm.vp.nearbyint.v16f32(<16 x float>, <16 x i1>, i32)
+declare <vscale x 1 x float> @llvm.vp.nearbyint.nvx1f32(<vscale x 1 x float>, <vscale x 1 x i1>, i32)
+declare <vscale x 2 x float> @llvm.vp.nearbyint.nvx2f32(<vscale x 2 x float>, <vscale x 2 x i1>, i32)
+declare <vscale x 4 x float> @llvm.vp.nearbyint.nvx4f32(<vscale x 4 x float>, <vscale x 4 x i1>, i32)
+declare <vscale x 8 x float> @llvm.vp.nearbyint.nvx8f32(<vscale x 8 x float>, <vscale x 8 x i1>, i32)
+declare <vscale x 16 x float> @llvm.vp.nearbyint.nvx16f32(<vscale x 16 x float>, <vscale x 16 x i1>, i32)
+declare double @llvm.vp.nearbyint.f64(double)
+declare <2 x double> @llvm.vp.nearbyint.v2f64(<2 x double>, <2 x i1>, i32)
+declare <4 x double> @llvm.vp.nearbyint.v4f64(<4 x double>, <4 x i1>, i32)
+declare <8 x double> @llvm.vp.nearbyint.v8f64(<8 x double>, <8 x i1>, i32)
+declare <16 x double> @llvm.vp.nearbyint.v16f64(<16 x double>, <16 x i1>, i32)
+declare <vscale x 1 x double> @llvm.vp.nearbyint.nvx1f64(<vscale x 1 x double>, <vscale x 1 x i1>, i32)
+declare <vscale x 2 x double> @llvm.vp.nearbyint.nvx2f64(<vscale x 2 x double>, <vscale x 2 x i1>, i32)
+declare <vscale x 4 x double> @llvm.vp.nearbyint.nvx5f64(<vscale x 4 x double>, <vscale x 4 x i1>, i32)
+declare <vscale x 8 x double> @llvm.vp.nearbyint.nvx8f64(<vscale x 8 x double>, <vscale x 8 x i1>, i32)

llvm/test/CodeGen/RISCV/rvv/fixed-vectors-nearbyint-vp.ll

+; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
+; RUN: llc -mtriple=riscv32 -mattr=+d,+zfh,+experimental-zvfh,+v -target-abi=ilp32d -riscv-v-vector-bits-min=128 \
+; RUN:   -verify-machineinstrs < %s | FileCheck %s
+; RUN: llc -mtriple=riscv64 -mattr=+d,+zfh,+experimental-zvfh,+v -target-abi=lp64d -riscv-v-vector-bits-min=128 \
+; RUN:   -verify-machineinstrs < %s | FileCheck %s
+
+declare <2 x half> @llvm.vp.nearbyint.v2f16(<2 x half>, <2 x i1>, i32)
+
+define <2 x half> @vp_nearbyint_v2f16(<2 x half> %va, <2 x i1> %m, i32 zeroext %evl) {
+; CHECK-LABEL: vp_nearbyint_v2f16:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    lui a1, %hi(.LCPI0_0)
+; CHECK-NEXT:    flh ft0, %lo(.LCPI0_0)(a1)
+; CHECK-NEXT:    vsetvli zero, a0, e16, mf4, ta, ma
+; CHECK-NEXT:    vfabs.v v9, v8, v0.t
+; CHECK-NEXT:    vsetvli zero, zero, e16, mf4, ta, mu
+; CHECK-NEXT:    vmflt.vf v0, v9, ft0, v0.t
+; CHECK-NEXT:    frflags a0
+; CHECK-NEXT:    vsetvli zero, zero, e16, mf4, ta, ma
+; CHECK-NEXT:    vfcvt.x.f.v v9, v8, v0.t
+; CHECK-NEXT:    vfcvt.f.x.v v9, v9, v0.t
+; CHECK-NEXT:    fsflags a0
+; CHECK-NEXT:    vsetvli zero, zero, e16, mf4, ta, mu
+; CHECK-NEXT:    vfsgnj.vv v8, v9, v8, v0.t
+; CHECK-NEXT:    ret
+  %v = call <2 x half> @llvm.vp.nearbyint.v2f16(<2 x half> %va, <2 x i1> %m, i32 %evl)
+  ret <2 x half> %v
+}
+
+define <2 x half> @vp_nearbyint_v2f16_unmasked(<2 x half> %va, i32 zeroext %evl) {
+; CHECK-LABEL: vp_nearbyint_v2f16_unmasked:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    lui a1, %hi(.LCPI1_0)
+; CHECK-NEXT:    flh ft0, %lo(.LCPI1_0)(a1)
+; CHECK-NEXT:    vsetivli zero, 2, e8, mf8, ta, ma
+; CHECK-NEXT:    vmset.m v0
+; CHECK-NEXT:    vsetvli zero, a0, e16, mf4, ta, ma
+; CHECK-NEXT:    vfabs.v v9, v8, v0.t
+; CHECK-NEXT:    vsetvli zero, zero, e16, mf4, ta, mu
+; CHECK-NEXT:    vmflt.vf v0, v9, ft0, v0.t
+; CHECK-NEXT:    frflags a0
+; CHECK-NEXT:    vsetvli zero, zero, e16, mf4, ta, ma
+; CHECK-NEXT:    vfcvt.x.f.v v9, v8, v0.t
+; CHECK-NEXT:    vfcvt.f.x.v v9, v9, v0.t
+; CHECK-NEXT:    fsflags a0
+; CHECK-NEXT:    vsetvli zero, zero, e16, mf4, ta, mu
+; CHECK-NEXT:    vfsgnj.vv v8, v9, v8, v0.t
+; CHECK-NEXT:    ret
+  %head = insertelement <2 x i1> poison, i1 true, i32 0
+  %m = shufflevector <2 x i1> %head, <2 x i1> poison, <2 x i32> zeroinitializer
+  %v = call <2 x half> @llvm.vp.nearbyint.v2f16(<2 x half> %va, <2 x i1> %m, i32 %evl)
+  ret <2 x half> %v
+}
+
+declare <4 x half> @llvm.vp.nearbyint.v4f16(<4 x half>, <4 x i1>, i32)
+
+define <4 x half> @vp_nearbyint_v4f16(<4 x half> %va, <4 x i1> %m, i32 zeroext %evl) {
+; CHECK-LABEL: vp_nearbyint_v4f16:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    lui a1, %hi(.LCPI2_0)
+; CHECK-NEXT:    flh ft0, %lo(.LCPI2_0)(a1)
+; CHECK-NEXT:    vsetvli zero, a0, e16, mf2, ta, ma
+; CHECK-NEXT:    vfabs.v v9, v8, v0.t
+; CHECK-NEXT:    vsetvli zero, zero, e16, mf2, ta, mu
+; CHECK-NEXT:    vmflt.vf v0, v9, ft0, v0.t
+; CHECK-NEXT:    frflags a0
+; CHECK-NEXT:    vsetvli zero, zero, e16, mf2, ta, ma
+; CHECK-NEXT:    vfcvt.x.f.v v9, v8, v0.t
+; CHECK-NEXT:    vfcvt.f.x.v v9, v9, v0.t
+; CHECK-NEXT:    fsflags a0
+; CHECK-NEXT:    vsetvli zero, zero, e16, mf2, ta, mu
+; CHECK-NEXT:    vfsgnj.vv v8, v9, v8, v0.t
+; CHECK-NEXT:    ret
+  %v = call <4 x half> @llvm.vp.nearbyint.v4f16(<4 x half> %va, <4 x i1> %m, i32 %evl)
+  ret <4 x half> %v
+}
+
+define <4 x half> @vp_nearbyint_v4f16_unmasked(<4 x half> %va, i32 zeroext %evl) {
+; CHECK-LABEL: vp_nearbyint_v4f16_unmasked:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    lui a1, %hi(.LCPI3_0)
+; CHECK-NEXT:    flh ft0, %lo(.LCPI3_0)(a1)
+; CHECK-NEXT:    vsetivli zero, 4, e8, mf4, ta, ma
+; CHECK-NEXT:    vmset.m v0
+; CHECK-NEXT:    vsetvli zero, a0, e16, mf2, ta, ma
+; CHECK-NEXT:    vfabs.v v9, v8, v0.t
+; CHECK-NEXT:    vsetvli zero, zero, e16, mf2, ta, mu
+; CHECK-NEXT:    vmflt.vf v0, v9, ft0, v0.t
+; CHECK-NEXT:    frflags a0
+; CHECK-NEXT:    vsetvli zero, zero, e16, mf2, ta, ma
+; CHECK-NEXT:    vfcvt.x.f.v v9, v8, v0.t
+; CHECK-NEXT:    vfcvt.f.x.v v9, v9, v0.t
+; CHECK-NEXT:    fsflags a0
+; CHECK-NEXT:    vsetvli zero, zero, e16, mf2, ta, mu
+; CHECK-NEXT:    vfsgnj.vv v8, v9, v8, v0.t
+; CHECK-NEXT:    ret
+  %head = insertelement <4 x i1> poison, i1 true, i32 0
+  %m = shufflevector <4 x i1> %head, <4 x i1> poison, <4 x i32> zeroinitializer
+  %v = call <4 x half> @llvm.vp.nearbyint.v4f16(<4 x half> %va, <4 x i1> %m, i32 %evl)
+  ret <4 x half> %v
+}
+
+declare <8 x half> @llvm.vp.nearbyint.v8f16(<8 x half>, <8 x i1>, i32)
+
+define <8 x half> @vp_nearbyint_v8f16(<8 x half> %va, <8 x i1> %m, i32 zeroext %evl) {
+; CHECK-LABEL: vp_nearbyint_v8f16:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    lui a1, %hi(.LCPI4_0)
+; CHECK-NEXT:    flh ft0, %lo(.LCPI4_0)(a1)
+; CHECK-NEXT:    vsetvli zero, a0, e16, m1, ta, ma
+; CHECK-NEXT:    vfabs.v v9, v8, v0.t
+; CHECK-NEXT:    vsetvli zero, zero, e16, m1, ta, mu
+; CHECK-NEXT:    vmflt.vf v0, v9, ft0, v0.t
+; CHECK-NEXT:    frflags a0
+; CHECK-NEXT:    vsetvli zero, zero, e16, m1, ta, ma
+; CHECK-NEXT:    vfcvt.x.f.v v9, v8, v0.t
+; CHECK-NEXT:    vfcvt.f.x.v v9, v9, v0.t
+; CHECK-NEXT:    fsflags a0
+; CHECK-NEXT:    vsetvli zero, zero, e16, m1, ta, mu
+; CHECK-NEXT:    vfsgnj.vv v8, v9, v8, v0.t
+; CHECK-NEXT:    ret
+  %v = call <8 x half> @llvm.vp.nearbyint.v8f16(<8 x half> %va, <8 x i1> %m, i32 %evl)
+  ret <8 x half> %v
+}
+
+define <8 x half> @vp_nearbyint_v8f16_unmasked(<8 x half> %va, i32 zeroext %evl) {
+; CHECK-LABEL: vp_nearbyint_v8f16_unmasked:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    lui a1, %hi(.LCPI5_0)
+; CHECK-NEXT:    flh ft0, %lo(.LCPI5_0)(a1)
+; CHECK-NEXT:    vsetivli zero, 8, e8, mf2, ta, ma
+; CHECK-NEXT:    vmset.m v0
+; CHECK-NEXT:    vsetvli zero, a0, e16, m1, ta, ma
+; CHECK-NEXT:    vfabs.v v9, v8, v0.t
+; CHECK-NEXT:    vsetvli zero, zero, e16, m1, ta, mu
+; CHECK-NEXT:    vmflt.vf v0, v9, ft0, v0.t
+; CHECK-NEXT:    frflags a0
+; CHECK-NEXT:    vsetvli zero, zero, e16, m1, ta, ma
+; CHECK-NEXT:    vfcvt.x.f.v v9, v8, v0.t
+; CHECK-NEXT:    vfcvt.f.x.v v9, v9, v0.t
+; CHECK-NEXT:    fsflags a0
+; CHECK-NEXT:    vsetvli zero, zero, e16, m1, ta, mu
+; CHECK-NEXT:    vfsgnj.vv v8, v9, v8, v0.t
+; CHECK-NEXT:    ret
+  %head = insertelement <8 x i1> poison, i1 true, i32 0
+  %m = shufflevector <8 x i1> %head, <8 x i1> poison, <8 x i32> zeroinitializer
+  %v = call <8 x half> @llvm.vp.nearbyint.v8f16(<8 x half> %va, <8 x i1> %m, i32 %evl)
+  ret <8 x half> %v
+}
+
+declare <16 x half> @llvm.vp.nearbyint.v16f16(<16 x half>, <16 x i1>, i32)
+
+define <16 x half> @vp_nearbyint_v16f16(<16 x half> %va, <16 x i1> %m, i32 zeroext %evl) {
+; CHECK-LABEL: vp_nearbyint_v16f16:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vmv1r.v v10, v0
+; CHECK-NEXT:    lui a1, %hi(.LCPI6_0)
+; CHECK-NEXT:    flh ft0, %lo(.LCPI6_0)(a1)
+; CHECK-NEXT:    vsetvli zero, a0, e16, m2, ta, ma
+; CHECK-NEXT:    vfabs.v v12, v8, v0.t
+; CHECK-NEXT:    vsetvli zero, zero, e16, m2, ta, mu
+; CHECK-NEXT:    vmflt.vf v10, v12, ft0, v0.t
+; CHECK-NEXT:    frflags a0
+; CHECK-NEXT:    vsetvli zero, zero, e16, m2, ta, ma
+; CHECK-NEXT:    vmv1r.v v0, v10
+; CHECK-NEXT:    vfcvt.x.f.v v12, v8, v0.t
+; CHECK-NEXT:    vfcvt.f.x.v v12, v12, v0.t
+; CHECK-NEXT:    fsflags a0
+; CHECK-NEXT:    vsetvli zero, zero, e16, m2, ta, mu
+; CHECK-NEXT:    vfsgnj.vv v8, v12, v8, v0.t
+; CHECK-NEXT:    ret
+  %v = call <16 x half> @llvm.vp.nearbyint.v16f16(<16 x half> %va, <16 x i1> %m, i32 %evl)
+  ret <16 x half> %v
+}
+
+define <16 x half> @vp_nearbyint_v16f16_unmasked(<16 x half> %va, i32 zeroext %evl) {
+; CHECK-LABEL: vp_nearbyint_v16f16_unmasked:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    lui a1, %hi(.LCPI7_0)
+; CHECK-NEXT:    flh ft0, %lo(.LCPI7_0)(a1)
+; CHECK-NEXT:    vsetivli zero, 16, e8, m1, ta, ma
+; CHECK-NEXT:    vmset.m v10
+; CHECK-NEXT:    vsetvli zero, a0, e16, m2, ta, ma
+; CHECK-NEXT:    vmv1r.v v0, v10
+; CHECK-NEXT:    vfabs.v v12, v8, v0.t
+; CHECK-NEXT:    vsetvli zero, zero, e16, m2, ta, mu
+; CHECK-NEXT:    vmflt.vf v10, v12, ft0, v0.t
+; CHECK-NEXT:    frflags a0
+; CHECK-NEXT:    vsetvli zero, zero, e16, m2, ta, ma
+; CHECK-NEXT:    vmv1r.v v0, v10
+; CHECK-NEXT:    vfcvt.x.f.v v12, v8, v0.t
+; CHECK-NEXT:    vfcvt.f.x.v v12, v12, v0.t
+; CHECK-NEXT:    fsflags a0
+; CHECK-NEXT:    vsetvli zero, zero, e16, m2, ta, mu
+; CHECK-NEXT:    vfsgnj.vv v8, v12, v8, v0.t
+; CHECK-NEXT:    ret
+  %head = insertelement <16 x i1> poison, i1 true, i32 0
+  %m = shufflevector <16 x i1> %head, <16 x i1> poison, <16 x i32> zeroinitializer
+  %v = call <16 x half> @llvm.vp.nearbyint.v16f16(<16 x half> %va, <16 x i1> %m, i32 %evl)
+  ret <16 x half> %v
+}
+
+declare <2 x float> @llvm.vp.nearbyint.v2f32(<2 x float>, <2 x i1>, i32)
+
+define <2 x float> @vp_nearbyint_v2f32(<2 x float> %va, <2 x i1> %m, i32 zeroext %evl) {
+; CHECK-LABEL: vp_nearbyint_v2f32:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    lui a1, %hi(.LCPI8_0)
+; CHECK-NEXT:    flw ft0, %lo(.LCPI8_0)(a1)
+; CHECK-NEXT:    vsetvli zero, a0, e32, mf2, ta, ma
+; CHECK-NEXT:    vfabs.v v9, v8, v0.t
+; CHECK-NEXT:    vsetvli zero, zero, e32, mf2, ta, mu
+; CHECK-NEXT:    vmflt.vf v0, v9, ft0, v0.t
+; CHECK-NEXT:    frflags a0
+; CHECK-NEXT:    vsetvli zero, zero, e32, mf2, ta, ma
+; CHECK-NEXT:    vfcvt.x.f.v v9, v8, v0.t
+; CHECK-NEXT:    vfcvt.f.x.v v9, v9, v0.t
+; CHECK-NEXT:    fsflags a0
+; CHECK-NEXT:    vsetvli zero, zero, e32, mf2, ta, mu
+; CHECK-NEXT:    vfsgnj.vv v8, v9, v8, v0.t
+; CHECK-NEXT:    ret
+  %v = call <2 x float> @llvm.vp.nearbyint.v2f32(<2 x float> %va, <2 x i1> %m, i32 %evl)
+  ret <2 x float> %v
+}
+
+define <2 x float> @vp_nearbyint_v2f32_unmasked(<2 x float> %va, i32 zeroext %evl) {
+; CHECK-LABEL: vp_nearbyint_v2f32_unmasked:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    lui a1, %hi(.LCPI9_0)
+; CHECK-NEXT:    flw ft0, %lo(.LCPI9_0)(a1)
+; CHECK-NEXT:    vsetivli zero, 2, e8, mf8, ta, ma
+; CHECK-NEXT:    vmset.m v0
+; CHECK-NEXT:    vsetvli zero, a0, e32, mf2, ta, ma
+; CHECK-NEXT:    vfabs.v v9, v8, v0.t
+; CHECK-NEXT:    vsetvli zero, zero, e32, mf2, ta, mu
+; CHECK-NEXT:    vmflt.vf v0, v9, ft0, v0.t
+; CHECK-NEXT:    frflags a0
+; CHECK-NEXT:    vsetvli zero, zero, e32, mf2, ta, ma
+; CHECK-NEXT:    vfcvt.x.f.v v9, v8, v0.t
+; CHECK-NEXT:    vfcvt.f.x.v v9, v9, v0.t
+; CHECK-NEXT:    fsflags a0
+; CHECK-NEXT:    vsetvli zero, zero, e32, mf2, ta, mu
+; CHECK-NEXT:    vfsgnj.vv v8, v9, v8, v0.t
+; CHECK-NEXT:    ret
+  %head = insertelement <2 x i1> poison, i1 true, i32 0
+  %m = shufflevector <2 x i1> %head, <2 x i1> poison, <2 x i32> zeroinitializer
+  %v = call <2 x float> @llvm.vp.nearbyint.v2f32(<2 x float> %va, <2 x i1> %m, i32 %evl)
+  ret <2 x float> %v
+}
+
+declare <4 x float> @llvm.vp.nearbyint.v4f32(<4 x float>, <4 x i1>, i32)
+
+define <4 x float> @vp_nearbyint_v4f32(<4 x float> %va, <4 x i1> %m, i32 zeroext %evl) {
+; CHECK-LABEL: vp_nearbyint_v4f32:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    lui a1, %hi(.LCPI10_0)
+; CHECK-NEXT:    flw ft0, %lo(.LCPI10_0)(a1)
+; CHECK-NEXT:    vsetvli zero, a0, e32, m1, ta, ma
+; CHECK-NEXT:    vfabs.v v9, v8, v0.t
+; CHECK-NEXT:    vsetvli zero, zero, e32, m1, ta, mu
+; CHECK-NEXT:    vmflt.vf v0, v9, ft0, v0.t
+; CHECK-NEXT:    frflags a0
+; CHECK-NEXT:    vsetvli zero, zero, e32, m1, ta, ma
+; CHECK-NEXT:    vfcvt.x.f.v v9, v8, v0.t
+; CHECK-NEXT:    vfcvt.f.x.v v9, v9, v0.t
+; CHECK-NEXT:    fsflags a0
+; CHECK-NEXT:    vsetvli zero, zero, e32, m1, ta, mu
+; CHECK-NEXT:    vfsgnj.vv v8, v9, v8, v0.t
+; CHECK-NEXT:    ret
+  %v = call <4 x float> @llvm.vp.nearbyint.v4f32(<4 x float> %va, <4 x i1> %m, i32 %evl)
+  ret <4 x float> %v
+}
+
+define <4 x float> @vp_nearbyint_v4f32_unmasked(<4 x float> %va, i32 zeroext %evl) {
+; CHECK-LABEL: vp_nearbyint_v4f32_unmasked:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    lui a1, %hi(.LCPI11_0)
+; CHECK-NEXT:    flw ft0, %lo(.LCPI11_0)(a1)
+; CHECK-NEXT:    vsetivli zero, 4, e8, mf4, ta, ma
+; CHECK-NEXT:    vmset.m v0
+; CHECK-NEXT:    vsetvli zero, a0, e32, m1, ta, ma
+; CHECK-NEXT:    vfabs.v v9, v8, v0.t
+; CHECK-NEXT:    vsetvli zero, zero, e32, m1, ta, mu
+; CHECK-NEXT:    vmflt.vf v0, v9, ft0, v0.t
+; CHECK-NEXT:    frflags a0
+; CHECK-NEXT:    vsetvli zero, zero, e32, m1, ta, ma
+; CHECK-NEXT:    vfcvt.x.f.v v9, v8, v0.t
+; CHECK-NEXT:    vfcvt.f.x.v v9, v9, v0.t
+; CHECK-NEXT:    fsflags a0
+; CHECK-NEXT:    vsetvli zero, zero, e32, m1, ta, mu
+; CHECK-NEXT:    vfsgnj.vv v8, v9, v8, v0.t
+; CHECK-NEXT:    ret
+  %head = insertelement <4 x i1> poison, i1 true, i32 0
+  %m = shufflevector <4 x i1> %head, <4 x i1> poison, <4 x i32> zeroinitializer
+  %v = call <4 x float> @llvm.vp.nearbyint.v4f32(<4 x float> %va, <4 x i1> %m, i32 %evl)
+  ret <4 x float> %v
+}
+
+declare <8 x float> @llvm.vp.nearbyint.v8f32(<8 x float>, <8 x i1>, i32)
+
+define <8 x float> @vp_nearbyint_v8f32(<8 x float> %va, <8 x i1> %m, i32 zeroext %evl) {
+; CHECK-LABEL: vp_nearbyint_v8f32:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vmv1r.v v10, v0
+; CHECK-NEXT:    lui a1, %hi(.LCPI12_0)
+; CHECK-NEXT:    flw ft0, %lo(.LCPI12_0)(a1)
+; CHECK-NEXT:    vsetvli zero, a0, e32, m2, ta, ma
+; CHECK-NEXT:    vfabs.v v12, v8, v0.t
+; CHECK-NEXT:    vsetvli zero, zero, e32, m2, ta, mu
+; CHECK-NEXT:    vmflt.vf v10, v12, ft0, v0.t
+; CHECK-NEXT:    frflags a0
+; CHECK-NEXT:    vsetvli zero, zero, e32, m2, ta, ma
+; CHECK-NEXT:    vmv1r.v v0, v10
+; CHECK-NEXT:    vfcvt.x.f.v v12, v8, v0.t
+; CHECK-NEXT:    vfcvt.f.x.v v12, v12, v0.t
+; CHECK-NEXT:    fsflags a0
+; CHECK-NEXT:    vsetvli zero, zero, e32, m2, ta, mu
+; CHECK-NEXT:    vfsgnj.vv v8, v12, v8, v0.t
+; CHECK-NEXT:    ret
+  %v = call <8 x float> @llvm.vp.nearbyint.v8f32(<8 x float> %va, <8 x i1> %m, i32 %evl)
+  ret <8 x float> %v
+}
+
+define <8 x float> @vp_nearbyint_v8f32_unmasked(<8 x float> %va, i32 zeroext %evl) {
+; CHECK-LABEL: vp_nearbyint_v8f32_unmasked:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    lui a1, %hi(.LCPI13_0)
+; CHECK-NEXT:    flw ft0, %lo(.LCPI13_0)(a1)
+; CHECK-NEXT:    vsetivli zero, 8, e8, mf2, ta, ma
+; CHECK-NEXT:    vmset.m v10
+; CHECK-NEXT:    vsetvli zero, a0, e32, m2, ta, ma
+; CHECK-NEXT:    vmv1r.v v0, v10
+; CHECK-NEXT:    vfabs.v v12, v8, v0.t
+; CHECK-NEXT:    vsetvli zero, zero, e32, m2, ta, mu
+; CHECK-NEXT:    vmflt.vf v10, v12, ft0, v0.t
+; CHECK-NEXT:    frflags a0
+; CHECK-NEXT:    vsetvli zero, zero, e32, m2, ta, ma
+; CHECK-NEXT:    vmv1r.v v0, v10
+; CHECK-NEXT:    vfcvt.x.f.v v12, v8, v0.t
+; CHECK-NEXT:    vfcvt.f.x.v v12, v12, v0.t
+; CHECK-NEXT:    fsflags a0
+; CHECK-NEXT:    vsetvli zero, zero, e32, m2, ta, mu
+; CHECK-NEXT:    vfsgnj.vv v8, v12, v8, v0.t
+; CHECK-NEXT:    ret
+  %head = insertelement <8 x i1> poison, i1 true, i32 0
+  %m = shufflevector <8 x i1> %head, <8 x i1> poison, <8 x i32> zeroinitializer
+  %v = call <8 x float> @llvm.vp.nearbyint.v8f32(<8 x float> %va, <8 x i1> %m, i32 %evl)
+  ret <8 x float> %v
+}
+
+declare <16 x float> @llvm.vp.nearbyint.v16f32(<16 x float>, <16 x i1>, i32)
+
+define <16 x float> @vp_nearbyint_v16f32(<16 x float> %va, <16 x i1> %m, i32 zeroext %evl) {
+; CHECK-LABEL: vp_nearbyint_v16f32:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vmv1r.v v12, v0
+; CHECK-NEXT:    lui a1, %hi(.LCPI14_0)
+; CHECK-NEXT:    flw ft0, %lo(.LCPI14_0)(a1)
+; CHECK-NEXT:    vsetvli zero, a0, e32, m4, ta, ma
+; CHECK-NEXT:    vfabs.v v16, v8, v0.t
+; CHECK-NEXT:    vsetvli zero, zero, e32, m4, ta, mu
+; CHECK-NEXT:    vmflt.vf v12, v16, ft0, v0.t
+; CHECK-NEXT:    frflags a0
+; CHECK-NEXT:    vsetvli zero, zero, e32, m4, ta, ma
+; CHECK-NEXT:    vmv1r.v v0, v12
+; CHECK-NEXT:    vfcvt.x.f.v v16, v8, v0.t
+; CHECK-NEXT:    vfcvt.f.x.v v16, v16, v0.t
+; CHECK-NEXT:    fsflags a0
+; CHECK-NEXT:    vsetvli zero, zero, e32, m4, ta, mu
+; CHECK-NEXT:    vfsgnj.vv v8, v16, v8, v0.t
+; CHECK-NEXT:    ret
+  %v = call <16 x float> @llvm.vp.nearbyint.v16f32(<16 x float> %va, <16 x i1> %m, i32 %evl)
+  ret <16 x float> %v
+}
+
+define <16 x float> @vp_nearbyint_v16f32_unmasked(<16 x float> %va, i32 zeroext %evl) {
+; CHECK-LABEL: vp_nearbyint_v16f32_unmasked:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    lui a1, %hi(.LCPI15_0)
+; CHECK-NEXT:    flw ft0, %lo(.LCPI15_0)(a1)
+; CHECK-NEXT:    vsetivli zero, 16, e8, m1, ta, ma
+; CHECK-NEXT:    vmset.m v12
+; CHECK-NEXT:    vsetvli zero, a0, e32, m4, ta, ma
+; CHECK-NEXT:    vmv1r.v v0, v12
+; CHECK-NEXT:    vfabs.v v16, v8, v0.t
+; CHECK-NEXT:    vsetvli zero, zero, e32, m4, ta, mu
+; CHECK-NEXT:    vmflt.vf v12, v16, ft0, v0.t
+; CHECK-NEXT:    frflags a0
+; CHECK-NEXT:    vsetvli zero, zero, e32, m4, ta, ma
+; CHECK-NEXT:    vmv1r.v v0, v12
+; CHECK-NEXT:    vfcvt.x.f.v v16, v8, v0.t
+; CHECK-NEXT:    vfcvt.f.x.v v16, v16, v0.t
+; CHECK-NEXT:    fsflags a0
+; CHECK-NEXT:    vsetvli zero, zero, e32, m4, ta, mu
+; CHECK-NEXT:    vfsgnj.vv v8, v16, v8, v0.t
+; CHECK-NEXT:    ret
+  %head = insertelement <16 x i1> poison, i1 true, i32 0
+  %m = shufflevector <16 x i1> %head, <16 x i1> poison, <16 x i32> zeroinitializer
+  %v = call <16 x float> @llvm.vp.nearbyint.v16f32(<16 x float> %va, <16 x i1> %m, i32 %evl)
+  ret <16 x float> %v
+}
+
+declare <2 x double> @llvm.vp.nearbyint.v2f64(<2 x double>, <2 x i1>, i32)
+
+define <2 x double> @vp_nearbyint_v2f64(<2 x double> %va, <2 x i1> %m, i32 zeroext %evl) {
+; CHECK-LABEL: vp_nearbyint_v2f64:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    lui a1, %hi(.LCPI16_0)
+; CHECK-NEXT:    fld ft0, %lo(.LCPI16_0)(a1)
+; CHECK-NEXT:    vsetvli zero, a0, e64, m1, ta, ma
+; CHECK-NEXT:    vfabs.v v9, v8, v0.t
+; CHECK-NEXT:    vsetvli zero, zero, e64, m1, ta, mu
+; CHECK-NEXT:    vmflt.vf v0, v9, ft0, v0.t
+; CHECK-NEXT:    frflags a0
+; CHECK-NEXT:    vsetvli zero, zero, e64, m1, ta, ma
+; CHECK-NEXT:    vfcvt.x.f.v v9, v8, v0.t
+; CHECK-NEXT:    vfcvt.f.x.v v9, v9, v0.t
+; CHECK-NEXT:    fsflags a0
+; CHECK-NEXT:    vsetvli zero, zero, e64, m1, ta, mu
+; CHECK-NEXT:    vfsgnj.vv v8, v9, v8, v0.t
+; CHECK-NEXT:    ret
+  %v = call <2 x double> @llvm.vp.nearbyint.v2f64(<2 x double> %va, <2 x i1> %m, i32 %evl)
+  ret <2 x double> %v
+}
+
+define <2 x double> @vp_nearbyint_v2f64_unmasked(<2 x double> %va, i32 zeroext %evl) {
+; CHECK-LABEL: vp_nearbyint_v2f64_unmasked:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    lui a1, %hi(.LCPI17_0)
+; CHECK-NEXT:    fld ft0, %lo(.LCPI17_0)(a1)
+; CHECK-NEXT:    vsetivli zero, 2, e8, mf8, ta, ma
+; CHECK-NEXT:    vmset.m v0
+; CHECK-NEXT:    vsetvli zero, a0, e64, m1, ta, ma
+; CHECK-NEXT:    vfabs.v v9, v8, v0.t
+; CHECK-NEXT:    vsetvli zero, zero, e64, m1, ta, mu
+; CHECK-NEXT:    vmflt.vf v0, v9, ft0, v0.t
+; CHECK-NEXT:    frflags a0
+; CHECK-NEXT:    vsetvli zero, zero, e64, m1, ta, ma
+; CHECK-NEXT:    vfcvt.x.f.v v9, v8, v0.t
+; CHECK-NEXT:    vfcvt.f.x.v v9, v9, v0.t
+; CHECK-NEXT:    fsflags a0
+; CHECK-NEXT:    vsetvli zero, zero, e64, m1, ta, mu
+; CHECK-NEXT:    vfsgnj.vv v8, v9, v8, v0.t
+; CHECK-NEXT:    ret
+  %head = insertelement <2 x i1> poison, i1 true, i32 0
+  %m = shufflevector <2 x i1> %head, <2 x i1> poison, <2 x i32> zeroinitializer
+  %v = call <2 x double> @llvm.vp.nearbyint.v2f64(<2 x double> %va, <2 x i1> %m, i32 %evl)
+  ret <2 x double> %v
+}
+
+declare <4 x double> @llvm.vp.nearbyint.v4f64(<4 x double>, <4 x i1>, i32)
+
+define <4 x double> @vp_nearbyint_v4f64(<4 x double> %va, <4 x i1> %m, i32 zeroext %evl) {
+; CHECK-LABEL: vp_nearbyint_v4f64:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vmv1r.v v10, v0
+; CHECK-NEXT:    lui a1, %hi(.LCPI18_0)
+; CHECK-NEXT:    fld ft0, %lo(.LCPI18_0)(a1)
+; CHECK-NEXT:    vsetvli zero, a0, e64, m2, ta, ma
+; CHECK-NEXT:    vfabs.v v12, v8, v0.t
+; CHECK-NEXT:    vsetvli zero, zero, e64, m2, ta, mu
+; CHECK-NEXT:    vmflt.vf v10, v12, ft0, v0.t
+; CHECK-NEXT:    frflags a0
+; CHECK-NEXT:    vsetvli zero, zero, e64, m2, ta, ma
+; CHECK-NEXT:    vmv1r.v v0, v10
+; CHECK-NEXT:    vfcvt.x.f.v v12, v8, v0.t
+; CHECK-NEXT:    vfcvt.f.x.v v12, v12, v0.t
+; CHECK-NEXT:    fsflags a0
+; CHECK-NEXT:    vsetvli zero, zero, e64, m2, ta, mu
+; CHECK-NEXT:    vfsgnj.vv v8, v12, v8, v0.t
+; CHECK-NEXT:    ret
+  %v = call <4 x double> @llvm.vp.nearbyint.v4f64(<4 x double> %va, <4 x i1> %m, i32 %evl)
+  ret <4 x double> %v
+}
+
+define <4 x double> @vp_nearbyint_v4f64_unmasked(<4 x double> %va, i32 zeroext %evl) {
+; CHECK-LABEL: vp_nearbyint_v4f64_unmasked:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    lui a1, %hi(.LCPI19_0)
+; CHECK-NEXT:    fld ft0, %lo(.LCPI19_0)(a1)
+; CHECK-NEXT:    vsetivli zero, 4, e8, mf4, ta, ma
+; CHECK-NEXT:    vmset.m v10
+; CHECK-NEXT:    vsetvli zero, a0, e64, m2, ta, ma
+; CHECK-NEXT:    vmv1r.v v0, v10
+; CHECK-NEXT:    vfabs.v v12, v8, v0.t
+; CHECK-NEXT:    vsetvli zero, zero, e64, m2, ta, mu
+; CHECK-NEXT:    vmflt.vf v10, v12, ft0, v0.t
+; CHECK-NEXT:    frflags a0
+; CHECK-NEXT:    vsetvli zero, zero, e64, m2, ta, ma
+; CHECK-NEXT:    vmv1r.v v0, v10
+; CHECK-NEXT:    vfcvt.x.f.v v12, v8, v0.t
+; CHECK-NEXT:    vfcvt.f.x.v v12, v12, v0.t
+; CHECK-NEXT:    fsflags a0
+; CHECK-NEXT:    vsetvli zero, zero, e64, m2, ta, mu
+; CHECK-NEXT:    vfsgnj.vv v8, v12, v8, v0.t
+; CHECK-NEXT:    ret
+  %head = insertelement <4 x i1> poison, i1 true, i32 0
+  %m = shufflevector <4 x i1> %head, <4 x i1> poison, <4 x i32> zeroinitializer
+  %v = call <4 x double> @llvm.vp.nearbyint.v4f64(<4 x double> %va, <4 x i1> %m, i32 %evl)
+  ret <4 x double> %v
+}
+
+declare <8 x double> @llvm.vp.nearbyint.v8f64(<8 x double>, <8 x i1>, i32)
+
+define <8 x double> @vp_nearbyint_v8f64(<8 x double> %va, <8 x i1> %m, i32 zeroext %evl) {
+; CHECK-LABEL: vp_nearbyint_v8f64:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vmv1r.v v12, v0
+; CHECK-NEXT:    lui a1, %hi(.LCPI20_0)
+; CHECK-NEXT:    fld ft0, %lo(.LCPI20_0)(a1)
+; CHECK-NEXT:    vsetvli zero, a0, e64, m4, ta, ma
+; CHECK-NEXT:    vfabs.v v16, v8, v0.t
+; CHECK-NEXT:    vsetvli zero, zero, e64, m4, ta, mu
+; CHECK-NEXT:    vmflt.vf v12, v16, ft0, v0.t
+; CHECK-NEXT:    frflags a0
+; CHECK-NEXT:    vsetvli zero, zero, e64, m4, ta, ma
+; CHECK-NEXT:    vmv1r.v v0, v12
+; CHECK-NEXT:    vfcvt.x.f.v v16, v8, v0.t
+; CHECK-NEXT:    vfcvt.f.x.v v16, v16, v0.t
+; CHECK-NEXT:    fsflags a0
+; CHECK-NEXT:    vsetvli zero, zero, e64, m4, ta, mu
+; CHECK-NEXT:    vfsgnj.vv v8, v16, v8, v0.t
+; CHECK-NEXT:    ret
+  %v = call <8 x double> @llvm.vp.nearbyint.v8f64(<8 x double> %va, <8 x i1> %m, i32 %evl)
+  ret <8 x double> %v
+}
+
+define <8 x double> @vp_nearbyint_v8f64_unmasked(<8 x double> %va, i32 zeroext %evl) {
+; CHECK-LABEL: vp_nearbyint_v8f64_unmasked:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    lui a1, %hi(.LCPI21_0)
+; CHECK-NEXT:    fld ft0, %lo(.LCPI21_0)(a1)
+; CHECK-NEXT:    vsetivli zero, 8, e8, mf2, ta, ma
+; CHECK-NEXT:    vmset.m v12
+; CHECK-NEXT:    vsetvli zero, a0, e64, m4, ta, ma
+; CHECK-NEXT:    vmv1r.v v0, v12
+; CHECK-NEXT:    vfabs.v v16, v8, v0.t
+; CHECK-NEXT:    vsetvli zero, zero, e64, m4, ta, mu
+; CHECK-NEXT:    vmflt.vf v12, v16, ft0, v0.t
+; CHECK-NEXT:    frflags a0
+; CHECK-NEXT:    vsetvli zero, zero, e64, m4, ta, ma
+; CHECK-NEXT:    vmv1r.v v0, v12
+; CHECK-NEXT:    vfcvt.x.f.v v16, v8, v0.t
+; CHECK-NEXT:    vfcvt.f.x.v v16, v16, v0.t
+; CHECK-NEXT:    fsflags a0
+; CHECK-NEXT:    vsetvli zero, zero, e64, m4, ta, mu
+; CHECK-NEXT:    vfsgnj.vv v8, v16, v8, v0.t
+; CHECK-NEXT:    ret
+  %head = insertelement <8 x i1> poison, i1 true, i32 0
+  %m = shufflevector <8 x i1> %head, <8 x i1> poison, <8 x i32> zeroinitializer
+  %v = call <8 x double> @llvm.vp.nearbyint.v8f64(<8 x double> %va, <8 x i1> %m, i32 %evl)
+  ret <8 x double> %v
+}
+
+declare <15 x double> @llvm.vp.nearbyint.v15f64(<15 x double>, <15 x i1>, i32)
+
+define <15 x double> @vp_nearbyint_v15f64(<15 x double> %va, <15 x i1> %m, i32 zeroext %evl) {
+; CHECK-LABEL: vp_nearbyint_v15f64:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vmv1r.v v16, v0
+; CHECK-NEXT:    lui a1, %hi(.LCPI22_0)
+; CHECK-NEXT:    fld ft0, %lo(.LCPI22_0)(a1)
+; CHECK-NEXT:    vsetvli zero, a0, e64, m8, ta, ma
+; CHECK-NEXT:    vfabs.v v24, v8, v0.t
+; CHECK-NEXT:    vsetvli zero, zero, e64, m8, ta, mu
+; CHECK-NEXT:    vmflt.vf v16, v24, ft0, v0.t
+; CHECK-NEXT:    frflags a0
+; CHECK-NEXT:    vsetvli zero, zero, e64, m8, ta, ma
+; CHECK-NEXT:    vmv1r.v v0, v16
+; CHECK-NEXT:    vfcvt.x.f.v v24, v8, v0.t
+; CHECK-NEXT:    vfcvt.f.x.v v24, v24, v0.t
+; CHECK-NEXT:    fsflags a0
+; CHECK-NEXT:    vsetvli zero, zero, e64, m8, ta, mu
+; CHECK-NEXT:    vfsgnj.vv v8, v24, v8, v0.t
+; CHECK-NEXT:    ret
+  %v = call <15 x double> @llvm.vp.nearbyint.v15f64(<15 x double> %va, <15 x i1> %m, i32 %evl)
+  ret <15 x double> %v
+}
+
+define <15 x double> @vp_nearbyint_v15f64_unmasked(<15 x double> %va, i32 zeroext %evl) {
+; CHECK-LABEL: vp_nearbyint_v15f64_unmasked:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    lui a1, %hi(.LCPI23_0)
+; CHECK-NEXT:    fld ft0, %lo(.LCPI23_0)(a1)
+; CHECK-NEXT:    vsetivli zero, 16, e8, m1, ta, ma
+; CHECK-NEXT:    vmset.m v16
+; CHECK-NEXT:    vsetvli zero, a0, e64, m8, ta, ma
+; CHECK-NEXT:    vmv1r.v v0, v16
+; CHECK-NEXT:    vfabs.v v24, v8, v0.t
+; CHECK-NEXT:    vsetvli zero, zero, e64, m8, ta, mu
+; CHECK-NEXT:    vmflt.vf v16, v24, ft0, v0.t
+; CHECK-NEXT:    frflags a0
+; CHECK-NEXT:    vsetvli zero, zero, e64, m8, ta, ma
+; CHECK-NEXT:    vmv1r.v v0, v16
+; CHECK-NEXT:    vfcvt.x.f.v v24, v8, v0.t
+; CHECK-NEXT:    vfcvt.f.x.v v24, v24, v0.t
+; CHECK-NEXT:    fsflags a0
+; CHECK-NEXT:    vsetvli zero, zero, e64, m8, ta, mu
+; CHECK-NEXT:    vfsgnj.vv v8, v24, v8, v0.t
+; CHECK-NEXT:    ret
+  %head = insertelement <15 x i1> poison, i1 true, i32 0
+  %m = shufflevector <15 x i1> %head, <15 x i1> poison, <15 x i32> zeroinitializer
+  %v = call <15 x double> @llvm.vp.nearbyint.v15f64(<15 x double> %va, <15 x i1> %m, i32 %evl)
+  ret <15 x double> %v
+}
+
+declare <16 x double> @llvm.vp.nearbyint.v16f64(<16 x double>, <16 x i1>, i32)
+
+define <16 x double> @vp_nearbyint_v16f64(<16 x double> %va, <16 x i1> %m, i32 zeroext %evl) {
+; CHECK-LABEL: vp_nearbyint_v16f64:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vmv1r.v v16, v0
+; CHECK-NEXT:    lui a1, %hi(.LCPI24_0)
+; CHECK-NEXT:    fld ft0, %lo(.LCPI24_0)(a1)
+; CHECK-NEXT:    vsetvli zero, a0, e64, m8, ta, ma
+; CHECK-NEXT:    vfabs.v v24, v8, v0.t
+; CHECK-NEXT:    vsetvli zero, zero, e64, m8, ta, mu
+; CHECK-NEXT:    vmflt.vf v16, v24, ft0, v0.t
+; CHECK-NEXT:    frflags a0
+; CHECK-NEXT:    vsetvli zero, zero, e64, m8, ta, ma
+; CHECK-NEXT:    vmv1r.v v0, v16
+; CHECK-NEXT:    vfcvt.x.f.v v24, v8, v0.t
+; CHECK-NEXT:    vfcvt.f.x.v v24, v24, v0.t
+; CHECK-NEXT:    fsflags a0
+; CHECK-NEXT:    vsetvli zero, zero, e64, m8, ta, mu
+; CHECK-NEXT:    vfsgnj.vv v8, v24, v8, v0.t
+; CHECK-NEXT:    ret
+  %v = call <16 x double> @llvm.vp.nearbyint.v16f64(<16 x double> %va, <16 x i1> %m, i32 %evl)
+  ret <16 x double> %v
+}
+
+define <16 x double> @vp_nearbyint_v16f64_unmasked(<16 x double> %va, i32 zeroext %evl) {
+; CHECK-LABEL: vp_nearbyint_v16f64_unmasked:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    lui a1, %hi(.LCPI25_0)
+; CHECK-NEXT:    fld ft0, %lo(.LCPI25_0)(a1)
+; CHECK-NEXT:    vsetivli zero, 16, e8, m1, ta, ma
+; CHECK-NEXT:    vmset.m v16
+; CHECK-NEXT:    vsetvli zero, a0, e64, m8, ta, ma
+; CHECK-NEXT:    vmv1r.v v0, v16
+; CHECK-NEXT:    vfabs.v v24, v8, v0.t
+; CHECK-NEXT:    vsetvli zero, zero, e64, m8, ta, mu
+; CHECK-NEXT:    vmflt.vf v16, v24, ft0, v0.t
+; CHECK-NEXT:    frflags a0
+; CHECK-NEXT:    vsetvli zero, zero, e64, m8, ta, ma
+; CHECK-NEXT:    vmv1r.v v0, v16
+; CHECK-NEXT:    vfcvt.x.f.v v24, v8, v0.t
+; CHECK-NEXT:    vfcvt.f.x.v v24, v24, v0.t
+; CHECK-NEXT:    fsflags a0
+; CHECK-NEXT:    vsetvli zero, zero, e64, m8, ta, mu
+; CHECK-NEXT:    vfsgnj.vv v8, v24, v8, v0.t
+; CHECK-NEXT:    ret
+  %head = insertelement <16 x i1> poison, i1 true, i32 0
+  %m = shufflevector <16 x i1> %head, <16 x i1> poison, <16 x i32> zeroinitializer
+  %v = call <16 x double> @llvm.vp.nearbyint.v16f64(<16 x double> %va, <16 x i1> %m, i32 %evl)
+  ret <16 x double> %v
+}
+
+declare <32 x double> @llvm.vp.nearbyint.v32f64(<32 x double>, <32 x i1>, i32)
+
+define <32 x double> @vp_nearbyint_v32f64(<32 x double> %va, <32 x i1> %m, i32 zeroext %evl) {
+; CHECK-LABEL: vp_nearbyint_v32f64:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vmv1r.v v2, v0
+; CHECK-NEXT:    vsetivli zero, 2, e8, mf4, ta, ma
+; CHECK-NEXT:    li a2, 16
+; CHECK-NEXT:    vslidedown.vi v1, v0, 2
+; CHECK-NEXT:    mv a1, a0
+; CHECK-NEXT:    bltu a0, a2, .LBB26_2
+; CHECK-NEXT:  # %bb.1:
+; CHECK-NEXT:    li a1, 16
+; CHECK-NEXT:  .LBB26_2:
+; CHECK-NEXT:    lui a2, %hi(.LCPI26_0)
+; CHECK-NEXT:    fld ft0, %lo(.LCPI26_0)(a2)
+; CHECK-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
+; CHECK-NEXT:    vmv1r.v v0, v2
+; CHECK-NEXT:    vfabs.v v24, v8, v0.t
+; CHECK-NEXT:    vsetvli zero, zero, e64, m8, ta, mu
+; CHECK-NEXT:    vmflt.vf v2, v24, ft0, v0.t
+; CHECK-NEXT:    frflags a1
+; CHECK-NEXT:    vsetvli zero, zero, e64, m8, ta, ma
+; CHECK-NEXT:    vmv1r.v v0, v2
+; CHECK-NEXT:    vfcvt.x.f.v v24, v8, v0.t
+; CHECK-NEXT:    vfcvt.f.x.v v24, v24, v0.t
+; CHECK-NEXT:    fsflags a1
+; CHECK-NEXT:    vsetvli zero, zero, e64, m8, ta, mu
+; CHECK-NEXT:    vfsgnj.vv v8, v24, v8, v0.t
+; CHECK-NEXT:    addi a1, a0, -16
+; CHECK-NEXT:    sltu a0, a0, a1
+; CHECK-NEXT:    addi a0, a0, -1
+; CHECK-NEXT:    and a0, a0, a1
+; CHECK-NEXT:    vsetvli zero, a0, e64, m8, ta, ma
+; CHECK-NEXT:    vmv1r.v v0, v1
+; CHECK-NEXT:    vfabs.v v24, v16, v0.t
+; CHECK-NEXT:    vsetvli zero, zero, e64, m8, ta, mu
+; CHECK-NEXT:    vmflt.vf v1, v24, ft0, v0.t
+; CHECK-NEXT:    frflags a0
+; CHECK-NEXT:    vsetvli zero, zero, e64, m8, ta, ma
+; CHECK-NEXT:    vmv1r.v v0, v1
+; CHECK-NEXT:    vfcvt.x.f.v v24, v16, v0.t
+; CHECK-NEXT:    vfcvt.f.x.v v24, v24, v0.t
+; CHECK-NEXT:    fsflags a0
+; CHECK-NEXT:    vsetvli zero, zero, e64, m8, ta, mu
+; CHECK-NEXT:    vfsgnj.vv v16, v24, v16, v0.t
+; CHECK-NEXT:    ret
+  %v = call <32 x double> @llvm.vp.nearbyint.v32f64(<32 x double> %va, <32 x i1> %m, i32 %evl)
+  ret <32 x double> %v
+}
+
+define <32 x double> @vp_nearbyint_v32f64_unmasked(<32 x double> %va, i32 zeroext %evl) {
+; CHECK-LABEL: vp_nearbyint_v32f64_unmasked:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli zero, 16, e8, m1, ta, ma
+; CHECK-NEXT:    li a2, 16
+; CHECK-NEXT:    vmset.m v1
+; CHECK-NEXT:    mv a1, a0
+; CHECK-NEXT:    bltu a0, a2, .LBB27_2
+; CHECK-NEXT:  # %bb.1:
+; CHECK-NEXT:    li a1, 16
+; CHECK-NEXT:  .LBB27_2:
+; CHECK-NEXT:    lui a2, %hi(.LCPI27_0)
+; CHECK-NEXT:    fld ft0, %lo(.LCPI27_0)(a2)
+; CHECK-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
+; CHECK-NEXT:    vmv1r.v v0, v1
+; CHECK-NEXT:    vfabs.v v24, v8, v0.t
+; CHECK-NEXT:    vsetvli zero, zero, e64, m8, ta, mu
+; CHECK-NEXT:    vmv1r.v v2, v1
+; CHECK-NEXT:    vmflt.vf v2, v24, ft0, v0.t
+; CHECK-NEXT:    frflags a1
+; CHECK-NEXT:    vsetvli zero, zero, e64, m8, ta, ma
+; CHECK-NEXT:    vmv1r.v v0, v2
+; CHECK-NEXT:    vfcvt.x.f.v v24, v8, v0.t
+; CHECK-NEXT:    vfcvt.f.x.v v24, v24, v0.t
+; CHECK-NEXT:    fsflags a1
+; CHECK-NEXT:    vsetvli zero, zero, e64, m8, ta, mu
+; CHECK-NEXT:    vfsgnj.vv v8, v24, v8, v0.t
+; CHECK-NEXT:    addi a1, a0, -16
+; CHECK-NEXT:    sltu a0, a0, a1
+; CHECK-NEXT:    addi a0, a0, -1
+; CHECK-NEXT:    and a0, a0, a1
+; CHECK-NEXT:    vsetvli zero, a0, e64, m8, ta, ma
+; CHECK-NEXT:    vmv1r.v v0, v1
+; CHECK-NEXT:    vfabs.v v24, v16, v0.t
+; CHECK-NEXT:    vsetvli zero, zero, e64, m8, ta, mu
+; CHECK-NEXT:    vmflt.vf v1, v24, ft0, v0.t
+; CHECK-NEXT:    frflags a0
+; CHECK-NEXT:    vsetvli zero, zero, e64, m8, ta, ma
+; CHECK-NEXT:    vmv1r.v v0, v1
+; CHECK-NEXT:    vfcvt.x.f.v v24, v16, v0.t
+; CHECK-NEXT:    vfcvt.f.x.v v24, v24, v0.t
+; CHECK-NEXT:    fsflags a0
+; CHECK-NEXT:    vsetvli zero, zero, e64, m8, ta, mu
+; CHECK-NEXT:    vfsgnj.vv v16, v24, v16, v0.t
+; CHECK-NEXT:    ret
+  %head = insertelement <32 x i1> poison, i1 true, i32 0
+  %m = shufflevector <32 x i1> %head, <32 x i1> poison, <32 x i32> zeroinitializer
+  %v = call <32 x double> @llvm.vp.nearbyint.v32f64(<32 x double> %va, <32 x i1> %m, i32 %evl)
+  ret <32 x double> %v
+}

llvm/test/CodeGen/RISCV/rvv/nearbyint-vp.ll

+; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
+; RUN: llc -mtriple=riscv32 -mattr=+d,+zfh,+experimental-zvfh,+v -target-abi=ilp32d \
+; RUN:     -verify-machineinstrs < %s | FileCheck %s
+; RUN: llc -mtriple=riscv64 -mattr=+d,+zfh,+experimental-zvfh,+v -target-abi=lp64d \
+; RUN:     -verify-machineinstrs < %s | FileCheck %s
+
+declare <vscale x 1 x half> @llvm.vp.nearbyint.nxv1f16(<vscale x 1 x half>, <vscale x 1 x i1>, i32)
+
+define <vscale x 1 x half> @vp_nearbyint_nxv1f16(<vscale x 1 x half> %va, <vscale x 1 x i1> %m, i32 zeroext %evl) {
+; CHECK-LABEL: vp_nearbyint_nxv1f16:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    lui a1, %hi(.LCPI0_0)
+; CHECK-NEXT:    flh ft0, %lo(.LCPI0_0)(a1)
+; CHECK-NEXT:    vsetvli zero, a0, e16, mf4, ta, ma
+; CHECK-NEXT:    vfabs.v v9, v8, v0.t
+; CHECK-NEXT:    vsetvli zero, zero, e16, mf4, ta, mu
+; CHECK-NEXT:    vmflt.vf v0, v9, ft0, v0.t
+; CHECK-NEXT:    frflags a0
+; CHECK-NEXT:    vsetvli zero, zero, e16, mf4, ta, ma
+; CHECK-NEXT:    vfcvt.x.f.v v9, v8, v0.t
+; CHECK-NEXT:    vfcvt.f.x.v v9, v9, v0.t
+; CHECK-NEXT:    fsflags a0
+; CHECK-NEXT:    vsetvli zero, zero, e16, mf4, ta, mu
+; CHECK-NEXT:    vfsgnj.vv v8, v9, v8, v0.t
+; CHECK-NEXT:    ret
+  %v = call <vscale x 1 x half> @llvm.vp.nearbyint.nxv1f16(<vscale x 1 x half> %va, <vscale x 1 x i1> %m, i32 %evl)
+  ret <vscale x 1 x half> %v
+}
+
+define <vscale x 1 x half> @vp_nearbyint_nxv1f16_unmasked(<vscale x 1 x half> %va, i32 zeroext %evl) {
+; CHECK-LABEL: vp_nearbyint_nxv1f16_unmasked:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    lui a1, %hi(.LCPI1_0)
+; CHECK-NEXT:    flh ft0, %lo(.LCPI1_0)(a1)
+; CHECK-NEXT:    vsetvli zero, a0, e16, mf4, ta, ma
+; CHECK-NEXT:    vfabs.v v9, v8
+; CHECK-NEXT:    vmflt.vf v0, v9, ft0
+; CHECK-NEXT:    frflags a0
+; CHECK-NEXT:    vfcvt.x.f.v v9, v8, v0.t
+; CHECK-NEXT:    vfcvt.f.x.v v9, v9, v0.t
+; CHECK-NEXT:    fsflags a0
+; CHECK-NEXT:    vsetvli zero, zero, e16, mf4, ta, mu
+; CHECK-NEXT:    vfsgnj.vv v8, v9, v8, v0.t
+; CHECK-NEXT:    ret
+  %head = insertelement <vscale x 1 x i1> poison, i1 true, i32 0
+  %m = shufflevector <vscale x 1 x i1> %head, <vscale x 1 x i1> poison, <vscale x 1 x i32> zeroinitializer
+  %v = call <vscale x 1 x half> @llvm.vp.nearbyint.nxv1f16(<vscale x 1 x half> %va, <vscale x 1 x i1> %m, i32 %evl)
+  ret <vscale x 1 x half> %v
+}
+
+declare <vscale x 2 x half> @llvm.vp.nearbyint.nxv2f16(<vscale x 2 x half>, <vscale x 2 x i1>, i32)
+
+define <vscale x 2 x half> @vp_nearbyint_nxv2f16(<vscale x 2 x half> %va, <vscale x 2 x i1> %m, i32 zeroext %evl) {
+; CHECK-LABEL: vp_nearbyint_nxv2f16:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    lui a1, %hi(.LCPI2_0)
+; CHECK-NEXT:    flh ft0, %lo(.LCPI2_0)(a1)
+; CHECK-NEXT:    vsetvli zero, a0, e16, mf2, ta, ma
+; CHECK-NEXT:    vfabs.v v9, v8, v0.t
+; CHECK-NEXT:    vsetvli zero, zero, e16, mf2, ta, mu
+; CHECK-NEXT:    vmflt.vf v0, v9, ft0, v0.t
+; CHECK-NEXT:    frflags a0
+; CHECK-NEXT:    vsetvli zero, zero, e16, mf2, ta, ma
+; CHECK-NEXT:    vfcvt.x.f.v v9, v8, v0.t
+; CHECK-NEXT:    vfcvt.f.x.v v9, v9, v0.t
+; CHECK-NEXT:    fsflags a0
+; CHECK-NEXT:    vsetvli zero, zero, e16, mf2, ta, mu
+; CHECK-NEXT:    vfsgnj.vv v8, v9, v8, v0.t
+; CHECK-NEXT:    ret
+  %v = call <vscale x 2 x half> @llvm.vp.nearbyint.nxv2f16(<vscale x 2 x half> %va, <vscale x 2 x i1> %m, i32 %evl)
+  ret <vscale x 2 x half> %v
+}
+
+define <vscale x 2 x half> @vp_nearbyint_nxv2f16_unmasked(<vscale x 2 x half> %va, i32 zeroext %evl) {
+; CHECK-LABEL: vp_nearbyint_nxv2f16_unmasked:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    lui a1, %hi(.LCPI3_0)
+; CHECK-NEXT:    flh ft0, %lo(.LCPI3_0)(a1)
+; CHECK-NEXT:    vsetvli zero, a0, e16, mf2, ta, ma
+; CHECK-NEXT:    vfabs.v v9, v8
+; CHECK-NEXT:    vmflt.vf v0, v9, ft0
+; CHECK-NEXT:    frflags a0
+; CHECK-NEXT:    vfcvt.x.f.v v9, v8, v0.t
+; CHECK-NEXT:    vfcvt.f.x.v v9, v9, v0.t
+; CHECK-NEXT:    fsflags a0
+; CHECK-NEXT:    vsetvli zero, zero, e16, mf2, ta, mu
+; CHECK-NEXT:    vfsgnj.vv v8, v9, v8, v0.t
+; CHECK-NEXT:    ret
+  %head = insertelement <vscale x 2 x i1> poison, i1 true, i32 0
+  %m = shufflevector <vscale x 2 x i1> %head, <vscale x 2 x i1> poison, <vscale x 2 x i32> zeroinitializer
+  %v = call <vscale x 2 x half> @llvm.vp.nearbyint.nxv2f16(<vscale x 2 x half> %va, <vscale x 2 x i1> %m, i32 %evl)
+  ret <vscale x 2 x half> %v
+}
+
+declare <vscale x 4 x half> @llvm.vp.nearbyint.nxv4f16(<vscale x 4 x half>, <vscale x 4 x i1>, i32)
+
+define <vscale x 4 x half> @vp_nearbyint_nxv4f16(<vscale x 4 x half> %va, <vscale x 4 x i1> %m, i32 zeroext %evl) {
+; CHECK-LABEL: vp_nearbyint_nxv4f16:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    lui a1, %hi(.LCPI4_0)
+; CHECK-NEXT:    flh ft0, %lo(.LCPI4_0)(a1)
+; CHECK-NEXT:    vsetvli zero, a0, e16, m1, ta, ma
+; CHECK-NEXT:    vfabs.v v9, v8, v0.t
+; CHECK-NEXT:    vsetvli zero, zero, e16, m1, ta, mu
+; CHECK-NEXT:    vmflt.vf v0, v9, ft0, v0.t
+; CHECK-NEXT:    frflags a0
+; CHECK-NEXT:    vsetvli zero, zero, e16, m1, ta, ma
+; CHECK-NEXT:    vfcvt.x.f.v v9, v8, v0.t
+; CHECK-NEXT:    vfcvt.f.x.v v9, v9, v0.t
+; CHECK-NEXT:    fsflags a0
+; CHECK-NEXT:    vsetvli zero, zero, e16, m1, ta, mu
+; CHECK-NEXT:    vfsgnj.vv v8, v9, v8, v0.t
+; CHECK-NEXT:    ret
+  %v = call <vscale x 4 x half> @llvm.vp.nearbyint.nxv4f16(<vscale x 4 x half> %va, <vscale x 4 x i1> %m, i32 %evl)
+  ret <vscale x 4 x half> %v
+}
+
+define <vscale x 4 x half> @vp_nearbyint_nxv4f16_unmasked(<vscale x 4 x half> %va, i32 zeroext %evl) {
+; CHECK-LABEL: vp_nearbyint_nxv4f16_unmasked:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    lui a1, %hi(.LCPI5_0)
+; CHECK-NEXT:    flh ft0, %lo(.LCPI5_0)(a1)
+; CHECK-NEXT:    vsetvli zero, a0, e16, m1, ta, ma
+; CHECK-NEXT:    vfabs.v v9, v8
+; CHECK-NEXT:    vmflt.vf v0, v9, ft0
+; CHECK-NEXT:    frflags a0
+; CHECK-NEXT:    vfcvt.x.f.v v9, v8, v0.t
+; CHECK-NEXT:    vfcvt.f.x.v v9, v9, v0.t
+; CHECK-NEXT:    fsflags a0
+; CHECK-NEXT:    vsetvli zero, zero, e16, m1, ta, mu
+; CHECK-NEXT:    vfsgnj.vv v8, v9, v8, v0.t
+; CHECK-NEXT:    ret
+  %head = insertelement <vscale x 4 x i1> poison, i1 true, i32 0
+  %m = shufflevector <vscale x 4 x i1> %head, <vscale x 4 x i1> poison, <vscale x 4 x i32> zeroinitializer
+  %v = call <vscale x 4 x half> @llvm.vp.nearbyint.nxv4f16(<vscale x 4 x half> %va, <vscale x 4 x i1> %m, i32 %evl)
+  ret <vscale x 4 x half> %v
+}
+
+declare <vscale x 8 x half> @llvm.vp.nearbyint.nxv8f16(<vscale x 8 x half>, <vscale x 8 x i1>, i32)
+
+define <vscale x 8 x half> @vp_nearbyint_nxv8f16(<vscale x 8 x half> %va, <vscale x 8 x i1> %m, i32 zeroext %evl) {
+; CHECK-LABEL: vp_nearbyint_nxv8f16:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vmv1r.v v10, v0
+; CHECK-NEXT:    lui a1, %hi(.LCPI6_0)
+; CHECK-NEXT:    flh ft0, %lo(.LCPI6_0)(a1)
+; CHECK-NEXT:    vsetvli zero, a0, e16, m2, ta, ma
+; CHECK-NEXT:    vfabs.v v12, v8, v0.t
+; CHECK-NEXT:    vsetvli zero, zero, e16, m2, ta, mu
+; CHECK-NEXT:    vmflt.vf v10, v12, ft0, v0.t
+; CHECK-NEXT:    frflags a0
+; CHECK-NEXT:    vsetvli zero, zero, e16, m2, ta, ma
+; CHECK-NEXT:    vmv1r.v v0, v10
+; CHECK-NEXT:    vfcvt.x.f.v v12, v8, v0.t
+; CHECK-NEXT:    vfcvt.f.x.v v12, v12, v0.t
+; CHECK-NEXT:    fsflags a0
+; CHECK-NEXT:    vsetvli zero, zero, e16, m2, ta, mu
+; CHECK-NEXT:    vfsgnj.vv v8, v12, v8, v0.t
+; CHECK-NEXT:    ret
+  %v = call <vscale x 8 x half> @llvm.vp.nearbyint.nxv8f16(<vscale x 8 x half> %va, <vscale x 8 x i1> %m, i32 %evl)
+  ret <vscale x 8 x half> %v
+}
+
+define <vscale x 8 x half> @vp_nearbyint_nxv8f16_unmasked(<vscale x 8 x half> %va, i32 zeroext %evl) {
+; CHECK-LABEL: vp_nearbyint_nxv8f16_unmasked:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    lui a1, %hi(.LCPI7_0)
+; CHECK-NEXT:    flh ft0, %lo(.LCPI7_0)(a1)
+; CHECK-NEXT:    vsetvli zero, a0, e16, m2, ta, ma
+; CHECK-NEXT:    vfabs.v v10, v8
+; CHECK-NEXT:    vmflt.vf v0, v10, ft0
+; CHECK-NEXT:    frflags a0
+; CHECK-NEXT:    vfcvt.x.f.v v10, v8, v0.t
+; CHECK-NEXT:    vfcvt.f.x.v v10, v10, v0.t
+; CHECK-NEXT:    fsflags a0
+; CHECK-NEXT:    vsetvli zero, zero, e16, m2, ta, mu
+; CHECK-NEXT:    vfsgnj.vv v8, v10, v8, v0.t
+; CHECK-NEXT:    ret
+  %head = insertelement <vscale x 8 x i1> poison, i1 true, i32 0
+  %m = shufflevector <vscale x 8 x i1> %head, <vscale x 8 x i1> poison, <vscale x 8 x i32> zeroinitializer
+  %v = call <vscale x 8 x half> @llvm.vp.nearbyint.nxv8f16(<vscale x 8 x half> %va, <vscale x 8 x i1> %m, i32 %evl)
+  ret <vscale x 8 x half> %v
+}
+
+declare <vscale x 16 x half> @llvm.vp.nearbyint.nxv16f16(<vscale x 16 x half>, <vscale x 16 x i1>, i32)
+
+define <vscale x 16 x half> @vp_nearbyint_nxv16f16(<vscale x 16 x half> %va, <vscale x 16 x i1> %m, i32 zeroext %evl) {
+; CHECK-LABEL: vp_nearbyint_nxv16f16:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vmv1r.v v12, v0
+; CHECK-NEXT:    lui a1, %hi(.LCPI8_0)
+; CHECK-NEXT:    flh ft0, %lo(.LCPI8_0)(a1)
+; CHECK-NEXT:    vsetvli zero, a0, e16, m4, ta, ma
+; CHECK-NEXT:    vfabs.v v16, v8, v0.t
+; CHECK-NEXT:    vsetvli zero, zero, e16, m4, ta, mu
+; CHECK-NEXT:    vmflt.vf v12, v16, ft0, v0.t
+; CHECK-NEXT:    frflags a0
+; CHECK-NEXT:    vsetvli zero, zero, e16, m4, ta, ma
+; CHECK-NEXT:    vmv1r.v v0, v12
+; CHECK-NEXT:    vfcvt.x.f.v v16, v8, v0.t
+; CHECK-NEXT:    vfcvt.f.x.v v16, v16, v0.t
+; CHECK-NEXT:    fsflags a0
+; CHECK-NEXT:    vsetvli zero, zero, e16, m4, ta, mu
+; CHECK-NEXT:    vfsgnj.vv v8, v16, v8, v0.t
+; CHECK-NEXT:    ret
+  %v = call <vscale x 16 x half> @llvm.vp.nearbyint.nxv16f16(<vscale x 16 x half> %va, <vscale x 16 x i1> %m, i32 %evl)
+  ret <vscale x 16 x half> %v
+}
+
+define <vscale x 16 x half> @vp_nearbyint_nxv16f16_unmasked(<vscale x 16 x half> %va, i32 zeroext %evl) {
+; CHECK-LABEL: vp_nearbyint_nxv16f16_unmasked:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    lui a1, %hi(.LCPI9_0)
+; CHECK-NEXT:    flh ft0, %lo(.LCPI9_0)(a1)
+; CHECK-NEXT:    vsetvli zero, a0, e16, m4, ta, ma
+; CHECK-NEXT:    vfabs.v v12, v8
+; CHECK-NEXT:    vmflt.vf v0, v12, ft0
+; CHECK-NEXT:    frflags a0
+; CHECK-NEXT:    vfcvt.x.f.v v12, v8, v0.t
+; CHECK-NEXT:    vfcvt.f.x.v v12, v12, v0.t
+; CHECK-NEXT:    fsflags a0
+; CHECK-NEXT:    vsetvli zero, zero, e16, m4, ta, mu
+; CHECK-NEXT:    vfsgnj.vv v8, v12, v8, v0.t
+; CHECK-NEXT:    ret
+  %head = insertelement <vscale x 16 x i1> poison, i1 true, i32 0
+  %m = shufflevector <vscale x 16 x i1> %head, <vscale x 16 x i1> poison, <vscale x 16 x i32> zeroinitializer
+  %v = call <vscale x 16 x half> @llvm.vp.nearbyint.nxv16f16(<vscale x 16 x half> %va, <vscale x 16 x i1> %m, i32 %evl)
+  ret <vscale x 16 x half> %v
+}
+
+declare <vscale x 32 x half> @llvm.vp.nearbyint.nxv32f16(<vscale x 32 x half>, <vscale x 32 x i1>, i32)
+
+define <vscale x 32 x half> @vp_nearbyint_nxv32f16(<vscale x 32 x half> %va, <vscale x 32 x i1> %m, i32 zeroext %evl) {
+; CHECK-LABEL: vp_nearbyint_nxv32f16:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vmv1r.v v16, v0
+; CHECK-NEXT:    lui a1, %hi(.LCPI10_0)
+; CHECK-NEXT:    flh ft0, %lo(.LCPI10_0)(a1)
+; CHECK-NEXT:    vsetvli zero, a0, e16, m8, ta, ma
+; CHECK-NEXT:    vfabs.v v24, v8, v0.t
+; CHECK-NEXT:    vsetvli zero, zero, e16, m8, ta, mu
+; CHECK-NEXT:    vmflt.vf v16, v24, ft0, v0.t
+; CHECK-NEXT:    frflags a0
+; CHECK-NEXT:    vsetvli zero, zero, e16, m8, ta, ma
+; CHECK-NEXT:    vmv1r.v v0, v16
+; CHECK-NEXT:    vfcvt.x.f.v v24, v8, v0.t
+; CHECK-NEXT:    vfcvt.f.x.v v24, v24, v0.t
+; CHECK-NEXT:    fsflags a0
+; CHECK-NEXT:    vsetvli zero, zero, e16, m8, ta, mu
+; CHECK-NEXT:    vfsgnj.vv v8, v24, v8, v0.t
+; CHECK-NEXT:    ret
+  %v = call <vscale x 32 x half> @llvm.vp.nearbyint.nxv32f16(<vscale x 32 x half> %va, <vscale x 32 x i1> %m, i32 %evl)
+  ret <vscale x 32 x half> %v
+}
+
+define <vscale x 32 x half> @vp_nearbyint_nxv32f16_unmasked(<vscale x 32 x half> %va, i32 zeroext %evl) {
+; CHECK-LABEL: vp_nearbyint_nxv32f16_unmasked:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    lui a1, %hi(.LCPI11_0)
+; CHECK-NEXT:    flh ft0, %lo(.LCPI11_0)(a1)
+; CHECK-NEXT:    vsetvli zero, a0, e16, m8, ta, ma
+; CHECK-NEXT:    vfabs.v v16, v8
+; CHECK-NEXT:    vmflt.vf v0, v16, ft0
+; CHECK-NEXT:    frflags a0
+; CHECK-NEXT:    vfcvt.x.f.v v16, v8, v0.t
+; CHECK-NEXT:    vfcvt.f.x.v v16, v16, v0.t
+; CHECK-NEXT:    fsflags a0
+; CHECK-NEXT:    vsetvli zero, zero, e16, m8, ta, mu
+; CHECK-NEXT:    vfsgnj.vv v8, v16, v8, v0.t
+; CHECK-NEXT:    ret
+  %head = insertelement <vscale x 32 x i1> poison, i1 true, i32 0
+  %m = shufflevector <vscale x 32 x i1> %head, <vscale x 32 x i1> poison, <vscale x 32 x i32> zeroinitializer
+  %v = call <vscale x 32 x half> @llvm.vp.nearbyint.nxv32f16(<vscale x 32 x half> %va, <vscale x 32 x i1> %m, i32 %evl)
+  ret <vscale x 32 x half> %v
+}
+
+declare <vscale x 1 x float> @llvm.vp.nearbyint.nxv1f32(<vscale x 1 x float>, <vscale x 1 x i1>, i32)
+
+define <vscale x 1 x float> @vp_nearbyint_nxv1f32(<vscale x 1 x float> %va, <vscale x 1 x i1> %m, i32 zeroext %evl) {
+; CHECK-LABEL: vp_nearbyint_nxv1f32:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    lui a1, %hi(.LCPI12_0)
+; CHECK-NEXT:    flw ft0, %lo(.LCPI12_0)(a1)
+; CHECK-NEXT:    vsetvli zero, a0, e32, mf2, ta, ma
+; CHECK-NEXT:    vfabs.v v9, v8, v0.t
+; CHECK-NEXT:    vsetvli zero, zero, e32, mf2, ta, mu
+; CHECK-NEXT:    vmflt.vf v0, v9, ft0, v0.t
+; CHECK-NEXT:    frflags a0
+; CHECK-NEXT:    vsetvli zero, zero, e32, mf2, ta, ma
+; CHECK-NEXT:    vfcvt.x.f.v v9, v8, v0.t
+; CHECK-NEXT:    vfcvt.f.x.v v9, v9, v0.t
+; CHECK-NEXT:    fsflags a0
+; CHECK-NEXT:    vsetvli zero, zero, e32, mf2, ta, mu
+; CHECK-NEXT:    vfsgnj.vv v8, v9, v8, v0.t
+; CHECK-NEXT:    ret
+  %v = call <vscale x 1 x float> @llvm.vp.nearbyint.nxv1f32(<vscale x 1 x float> %va, <vscale x 1 x i1> %m, i32 %evl)
+  ret <vscale x 1 x float> %v
+}
+
+define <vscale x 1 x float> @vp_nearbyint_nxv1f32_unmasked(<vscale x 1 x float> %va, i32 zeroext %evl) {
+; CHECK-LABEL: vp_nearbyint_nxv1f32_unmasked:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    lui a1, %hi(.LCPI13_0)
+; CHECK-NEXT:    flw ft0, %lo(.LCPI13_0)(a1)
+; CHECK-NEXT:    vsetvli zero, a0, e32, mf2, ta, ma
+; CHECK-NEXT:    vfabs.v v9, v8
+; CHECK-NEXT:    vmflt.vf v0, v9, ft0
+; CHECK-NEXT:    frflags a0
+; CHECK-NEXT:    vfcvt.x.f.v v9, v8, v0.t
+; CHECK-NEXT:    vfcvt.f.x.v v9, v9, v0.t
+; CHECK-NEXT:    fsflags a0
+; CHECK-NEXT:    vsetvli zero, zero, e32, mf2, ta, mu
+; CHECK-NEXT:    vfsgnj.vv v8, v9, v8, v0.t
+; CHECK-NEXT:    ret
+  %head = insertelement <vscale x 1 x i1> poison, i1 true, i32 0
+  %m = shufflevector <vscale x 1 x i1> %head, <vscale x 1 x i1> poison, <vscale x 1 x i32> zeroinitializer
+  %v = call <vscale x 1 x float> @llvm.vp.nearbyint.nxv1f32(<vscale x 1 x float> %va, <vscale x 1 x i1> %m, i32 %evl)
+  ret <vscale x 1 x float> %v
+}
+
+declare <vscale x 2 x float> @llvm.vp.nearbyint.nxv2f32(<vscale x 2 x float>, <vscale x 2 x i1>, i32)
+
+define <vscale x 2 x float> @vp_nearbyint_nxv2f32(<vscale x 2 x float> %va, <vscale x 2 x i1> %m, i32 zeroext %evl) {
+; CHECK-LABEL: vp_nearbyint_nxv2f32:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    lui a1, %hi(.LCPI14_0)
+; CHECK-NEXT:    flw ft0, %lo(.LCPI14_0)(a1)
+; CHECK-NEXT:    vsetvli zero, a0, e32, m1, ta, ma
+; CHECK-NEXT:    vfabs.v v9, v8, v0.t
+; CHECK-NEXT:    vsetvli zero, zero, e32, m1, ta, mu
+; CHECK-NEXT:    vmflt.vf v0, v9, ft0, v0.t
+; CHECK-NEXT:    frflags a0
+; CHECK-NEXT:    vsetvli zero, zero, e32, m1, ta, ma
+; CHECK-NEXT:    vfcvt.x.f.v v9, v8, v0.t
+; CHECK-NEXT:    vfcvt.f.x.v v9, v9, v0.t
+; CHECK-NEXT:    fsflags a0
+; CHECK-NEXT:    vsetvli zero, zero, e32, m1, ta, mu
+; CHECK-NEXT:    vfsgnj.vv v8, v9, v8, v0.t
+; CHECK-NEXT:    ret
+  %v = call <vscale x 2 x float> @llvm.vp.nearbyint.nxv2f32(<vscale x 2 x float> %va, <vscale x 2 x i1> %m, i32 %evl)
+  ret <vscale x 2 x float> %v
+}
+
+define <vscale x 2 x float> @vp_nearbyint_nxv2f32_unmasked(<vscale x 2 x float> %va, i32 zeroext %evl) {
+; CHECK-LABEL: vp_nearbyint_nxv2f32_unmasked:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    lui a1, %hi(.LCPI15_0)
+; CHECK-NEXT:    flw ft0, %lo(.LCPI15_0)(a1)
+; CHECK-NEXT:    vsetvli zero, a0, e32, m1, ta, ma
+; CHECK-NEXT:    vfabs.v v9, v8
+; CHECK-NEXT:    vmflt.vf v0, v9, ft0
+; CHECK-NEXT:    frflags a0
+; CHECK-NEXT:    vfcvt.x.f.v v9, v8, v0.t
+; CHECK-NEXT:    vfcvt.f.x.v v9, v9, v0.t
+; CHECK-NEXT:    fsflags a0
+; CHECK-NEXT:    vsetvli zero, zero, e32, m1, ta, mu
+; CHECK-NEXT:    vfsgnj.vv v8, v9, v8, v0.t
+; CHECK-NEXT:    ret
+  %head = insertelement <vscale x 2 x i1> poison, i1 true, i32 0
+  %m = shufflevector <vscale x 2 x i1> %head, <vscale x 2 x i1> poison, <vscale x 2 x i32> zeroinitializer
+  %v = call <vscale x 2 x float> @llvm.vp.nearbyint.nxv2f32(<vscale x 2 x float> %va, <vscale x 2 x i1> %m, i32 %evl)
+  ret <vscale x 2 x float> %v
+}
+
+declare <vscale x 4 x float> @llvm.vp.nearbyint.nxv4f32(<vscale x 4 x float>, <vscale x 4 x i1>, i32)
+
+define <vscale x 4 x float> @vp_nearbyint_nxv4f32(<vscale x 4 x float> %va, <vscale x 4 x i1> %m, i32 zeroext %evl) {
+; CHECK-LABEL: vp_nearbyint_nxv4f32:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vmv1r.v v10, v0
+; CHECK-NEXT:    lui a1, %hi(.LCPI16_0)
+; CHECK-NEXT:    flw ft0, %lo(.LCPI16_0)(a1)
+; CHECK-NEXT:    vsetvli zero, a0, e32, m2, ta, ma
+; CHECK-NEXT:    vfabs.v v12, v8, v0.t
+; CHECK-NEXT:    vsetvli zero, zero, e32, m2, ta, mu
+; CHECK-NEXT:    vmflt.vf v10, v12, ft0, v0.t
+; CHECK-NEXT:    frflags a0
+; CHECK-NEXT:    vsetvli zero, zero, e32, m2, ta, ma
+; CHECK-NEXT:    vmv1r.v v0, v10
+; CHECK-NEXT:    vfcvt.x.f.v v12, v8, v0.t
+; CHECK-NEXT:    vfcvt.f.x.v v12, v12, v0.t
+; CHECK-NEXT:    fsflags a0
+; CHECK-NEXT:    vsetvli zero, zero, e32, m2, ta, mu
+; CHECK-NEXT:    vfsgnj.vv v8, v12, v8, v0.t
+; CHECK-NEXT:    ret
+  %v = call <vscale x 4 x float> @llvm.vp.nearbyint.nxv4f32(<vscale x 4 x float> %va, <vscale x 4 x i1> %m, i32 %evl)
+  ret <vscale x 4 x float> %v
+}
+
+define <vscale x 4 x float> @vp_nearbyint_nxv4f32_unmasked(<vscale x 4 x float> %va, i32 zeroext %evl) {
+; CHECK-LABEL: vp_nearbyint_nxv4f32_unmasked:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    lui a1, %hi(.LCPI17_0)
+; CHECK-NEXT:    flw ft0, %lo(.LCPI17_0)(a1)
+; CHECK-NEXT:    vsetvli zero, a0, e32, m2, ta, ma
+; CHECK-NEXT:    vfabs.v v10, v8
+; CHECK-NEXT:    vmflt.vf v0, v10, ft0
+; CHECK-NEXT:    frflags a0
+; CHECK-NEXT:    vfcvt.x.f.v v10, v8, v0.t
+; CHECK-NEXT:    vfcvt.f.x.v v10, v10, v0.t
+; CHECK-NEXT:    fsflags a0
+; CHECK-NEXT:    vsetvli zero, zero, e32, m2, ta, mu
+; CHECK-NEXT:    vfsgnj.vv v8, v10, v8, v0.t
+; CHECK-NEXT:    ret
+  %head = insertelement <vscale x 4 x i1> poison, i1 true, i32 0
+  %m = shufflevector <vscale x 4 x i1> %head, <vscale x 4 x i1> poison, <vscale x 4 x i32> zeroinitializer
+  %v = call <vscale x 4 x float> @llvm.vp.nearbyint.nxv4f32(<vscale x 4 x float> %va, <vscale x 4 x i1> %m, i32 %evl)
+  ret <vscale x 4 x float> %v
+}
+
+declare <vscale x 8 x float> @llvm.vp.nearbyint.nxv8f32(<vscale x 8 x float>, <vscale x 8 x i1>, i32)
+
+define <vscale x 8 x float> @vp_nearbyint_nxv8f32(<vscale x 8 x float> %va, <vscale x 8 x i1> %m, i32 zeroext %evl) {
+; CHECK-LABEL: vp_nearbyint_nxv8f32:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vmv1r.v v12, v0
+; CHECK-NEXT:    lui a1, %hi(.LCPI18_0)
+; CHECK-NEXT:    flw ft0, %lo(.LCPI18_0)(a1)
+; CHECK-NEXT:    vsetvli zero, a0, e32, m4, ta, ma
+; CHECK-NEXT:    vfabs.v v16, v8, v0.t
+; CHECK-NEXT:    vsetvli zero, zero, e32, m4, ta, mu
+; CHECK-NEXT:    vmflt.vf v12, v16, ft0, v0.t
+; CHECK-NEXT:    frflags a0
+; CHECK-NEXT:    vsetvli zero, zero, e32, m4, ta, ma
+; CHECK-NEXT:    vmv1r.v v0, v12
+; CHECK-NEXT:    vfcvt.x.f.v v16, v8, v0.t
+; CHECK-NEXT:    vfcvt.f.x.v v16, v16, v0.t
+; CHECK-NEXT:    fsflags a0
+; CHECK-NEXT:    vsetvli zero, zero, e32, m4, ta, mu
+; CHECK-NEXT:    vfsgnj.vv v8, v16, v8, v0.t
+; CHECK-NEXT:    ret
+  %v = call <vscale x 8 x float> @llvm.vp.nearbyint.nxv8f32(<vscale x 8 x float> %va, <vscale x 8 x i1> %m, i32 %evl)
+  ret <vscale x 8 x float> %v
+}
+
+define <vscale x 8 x float> @vp_nearbyint_nxv8f32_unmasked(<vscale x 8 x float> %va, i32 zeroext %evl) {
+; CHECK-LABEL: vp_nearbyint_nxv8f32_unmasked:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    lui a1, %hi(.LCPI19_0)
+; CHECK-NEXT:    flw ft0, %lo(.LCPI19_0)(a1)
+; CHECK-NEXT:    vsetvli zero, a0, e32, m4, ta, ma
+; CHECK-NEXT:    vfabs.v v12, v8
+; CHECK-NEXT:    vmflt.vf v0, v12, ft0
+; CHECK-NEXT:    frflags a0
+; CHECK-NEXT:    vfcvt.x.f.v v12, v8, v0.t
+; CHECK-NEXT:    vfcvt.f.x.v v12, v12, v0.t
+; CHECK-NEXT:    fsflags a0
+; CHECK-NEXT:    vsetvli zero, zero, e32, m4, ta, mu
+; CHECK-NEXT:    vfsgnj.vv v8, v12, v8, v0.t
+; CHECK-NEXT:    ret
+  %head = insertelement <vscale x 8 x i1> poison, i1 true, i32 0
+  %m = shufflevector <vscale x 8 x i1> %head, <vscale x 8 x i1> poison, <vscale x 8 x i32> zeroinitializer
+  %v = call <vscale x 8 x float> @llvm.vp.nearbyint.nxv8f32(<vscale x 8 x float> %va, <vscale x 8 x i1> %m, i32 %evl)
+  ret <vscale x 8 x float> %v
+}
+
+declare <vscale x 16 x float> @llvm.vp.nearbyint.nxv16f32(<vscale x 16 x float>, <vscale x 16 x i1>, i32)
+
+define <vscale x 16 x float> @vp_nearbyint_nxv16f32(<vscale x 16 x float> %va, <vscale x 16 x i1> %m, i32 zeroext %evl) {
+; CHECK-LABEL: vp_nearbyint_nxv16f32:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vmv1r.v v16, v0
+; CHECK-NEXT:    lui a1, %hi(.LCPI20_0)
+; CHECK-NEXT:    flw ft0, %lo(.LCPI20_0)(a1)
+; CHECK-NEXT:    vsetvli zero, a0, e32, m8, ta, ma
+; CHECK-NEXT:    vfabs.v v24, v8, v0.t
+; CHECK-NEXT:    vsetvli zero, zero, e32, m8, ta, mu
+; CHECK-NEXT:    vmflt.vf v16, v24, ft0, v0.t
+; CHECK-NEXT:    frflags a0
+; CHECK-NEXT:    vsetvli zero, zero, e32, m8, ta, ma
+; CHECK-NEXT:    vmv1r.v v0, v16
+; CHECK-NEXT:    vfcvt.x.f.v v24, v8, v0.t
+; CHECK-NEXT:    vfcvt.f.x.v v24, v24, v0.t
+; CHECK-NEXT:    fsflags a0
+; CHECK-NEXT:    vsetvli zero, zero, e32, m8, ta, mu
+; CHECK-NEXT:    vfsgnj.vv v8, v24, v8, v0.t
+; CHECK-NEXT:    ret
+  %v = call <vscale x 16 x float> @llvm.vp.nearbyint.nxv16f32(<vscale x 16 x float> %va, <vscale x 16 x i1> %m, i32 %evl)
+  ret <vscale x 16 x float> %v
+}
+
+define <vscale x 16 x float> @vp_nearbyint_nxv16f32_unmasked(<vscale x 16 x float> %va, i32 zeroext %evl) {
+; CHECK-LABEL: vp_nearbyint_nxv16f32_unmasked:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    lui a1, %hi(.LCPI21_0)
+; CHECK-NEXT:    flw ft0, %lo(.LCPI21_0)(a1)
+; CHECK-NEXT:    vsetvli zero, a0, e32, m8, ta, ma
+; CHECK-NEXT:    vfabs.v v16, v8
+; CHECK-NEXT:    vmflt.vf v0, v16, ft0
+; CHECK-NEXT:    frflags a0
+; CHECK-NEXT:    vfcvt.x.f.v v16, v8, v0.t
+; CHECK-NEXT:    vfcvt.f.x.v v16, v16, v0.t
+; CHECK-NEXT:    fsflags a0
+; CHECK-NEXT:    vsetvli zero, zero, e32, m8, ta, mu
+; CHECK-NEXT:    vfsgnj.vv v8, v16, v8, v0.t
+; CHECK-NEXT:    ret
+  %head = insertelement <vscale x 16 x i1> poison, i1 true, i32 0
+  %m = shufflevector <vscale x 16 x i1> %head, <vscale x 16 x i1> poison, <vscale x 16 x i32> zeroinitializer
+  %v = call <vscale x 16 x float> @llvm.vp.nearbyint.nxv16f32(<vscale x 16 x float> %va, <vscale x 16 x i1> %m, i32 %evl)
+  ret <vscale x 16 x float> %v
+}
+
+declare <vscale x 1 x double> @llvm.vp.nearbyint.nxv1f64(<vscale x 1 x double>, <vscale x 1 x i1>, i32)
+
+define <vscale x 1 x double> @vp_nearbyint_nxv1f64(<vscale x 1 x double> %va, <vscale x 1 x i1> %m, i32 zeroext %evl) {
+; CHECK-LABEL: vp_nearbyint_nxv1f64:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    lui a1, %hi(.LCPI22_0)
+; CHECK-NEXT:    fld ft0, %lo(.LCPI22_0)(a1)
+; CHECK-NEXT:    vsetvli zero, a0, e64, m1, ta, ma
+; CHECK-NEXT:    vfabs.v v9, v8, v0.t
+; CHECK-NEXT:    vsetvli zero, zero, e64, m1, ta, mu
+; CHECK-NEXT:    vmflt.vf v0, v9, ft0, v0.t
+; CHECK-NEXT:    frflags a0
+; CHECK-NEXT:    vsetvli zero, zero, e64, m1, ta, ma
+; CHECK-NEXT:    vfcvt.x.f.v v9, v8, v0.t
+; CHECK-NEXT:    vfcvt.f.x.v v9, v9, v0.t
+; CHECK-NEXT:    fsflags a0
+; CHECK-NEXT:    vsetvli zero, zero, e64, m1, ta, mu
+; CHECK-NEXT:    vfsgnj.vv v8, v9, v8, v0.t
+; CHECK-NEXT:    ret
+  %v = call <vscale x 1 x double> @llvm.vp.nearbyint.nxv1f64(<vscale x 1 x double> %va, <vscale x 1 x i1> %m, i32 %evl)
+  ret <vscale x 1 x double> %v
+}
+
+define <vscale x 1 x double> @vp_nearbyint_nxv1f64_unmasked(<vscale x 1 x double> %va, i32 zeroext %evl) {
+; CHECK-LABEL: vp_nearbyint_nxv1f64_unmasked:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    lui a1, %hi(.LCPI23_0)
+; CHECK-NEXT:    fld ft0, %lo(.LCPI23_0)(a1)
+; CHECK-NEXT:    vsetvli zero, a0, e64, m1, ta, ma
+; CHECK-NEXT:    vfabs.v v9, v8
+; CHECK-NEXT:    vmflt.vf v0, v9, ft0
+; CHECK-NEXT:    frflags a0
+; CHECK-NEXT:    vfcvt.x.f.v v9, v8, v0.t
+; CHECK-NEXT:    vfcvt.f.x.v v9, v9, v0.t
+; CHECK-NEXT:    fsflags a0
+; CHECK-NEXT:    vsetvli zero, zero, e64, m1, ta, mu
+; CHECK-NEXT:    vfsgnj.vv v8, v9, v8, v0.t
+; CHECK-NEXT:    ret
+  %head = insertelement <vscale x 1 x i1> poison, i1 true, i32 0
+  %m = shufflevector <vscale x 1 x i1> %head, <vscale x 1 x i1> poison, <vscale x 1 x i32> zeroinitializer
+  %v = call <vscale x 1 x double> @llvm.vp.nearbyint.nxv1f64(<vscale x 1 x double> %va, <vscale x 1 x i1> %m, i32 %evl)
+  ret <vscale x 1 x double> %v
+}
+
+declare <vscale x 2 x double> @llvm.vp.nearbyint.nxv2f64(<vscale x 2 x double>, <vscale x 2 x i1>, i32)
+
+define <vscale x 2 x double> @vp_nearbyint_nxv2f64(<vscale x 2 x double> %va, <vscale x 2 x i1> %m, i32 zeroext %evl) {
+; CHECK-LABEL: vp_nearbyint_nxv2f64:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vmv1r.v v10, v0
+; CHECK-NEXT:    lui a1, %hi(.LCPI24_0)
+; CHECK-NEXT:    fld ft0, %lo(.LCPI24_0)(a1)
+; CHECK-NEXT:    vsetvli zero, a0, e64, m2, ta, ma
+; CHECK-NEXT:    vfabs.v v12, v8, v0.t
+; CHECK-NEXT:    vsetvli zero, zero, e64, m2, ta, mu
+; CHECK-NEXT:    vmflt.vf v10, v12, ft0, v0.t
+; CHECK-NEXT:    frflags a0
+; CHECK-NEXT:    vsetvli zero, zero, e64, m2, ta, ma
+; CHECK-NEXT:    vmv1r.v v0, v10
+; CHECK-NEXT:    vfcvt.x.f.v v12, v8, v0.t
+; CHECK-NEXT:    vfcvt.f.x.v v12, v12, v0.t
+; CHECK-NEXT:    fsflags a0
+; CHECK-NEXT:    vsetvli zero, zero, e64, m2, ta, mu
+; CHECK-NEXT:    vfsgnj.vv v8, v12, v8, v0.t
+; CHECK-NEXT:    ret
+  %v = call <vscale x 2 x double> @llvm.vp.nearbyint.nxv2f64(<vscale x 2 x double> %va, <vscale x 2 x i1> %m, i32 %evl)
+  ret <vscale x 2 x double> %v
+}
+
+define <vscale x 2 x double> @vp_nearbyint_nxv2f64_unmasked(<vscale x 2 x double> %va, i32 zeroext %evl) {
+; CHECK-LABEL: vp_nearbyint_nxv2f64_unmasked:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    lui a1, %hi(.LCPI25_0)
+; CHECK-NEXT:    fld ft0, %lo(.LCPI25_0)(a1)
+; CHECK-NEXT:    vsetvli zero, a0, e64, m2, ta, ma
+; CHECK-NEXT:    vfabs.v v10, v8
+; CHECK-NEXT:    vmflt.vf v0, v10, ft0
+; CHECK-NEXT:    frflags a0
+; CHECK-NEXT:    vfcvt.x.f.v v10, v8, v0.t
+; CHECK-NEXT:    vfcvt.f.x.v v10, v10, v0.t
+; CHECK-NEXT:    fsflags a0
+; CHECK-NEXT:    vsetvli zero, zero, e64, m2, ta, mu
+; CHECK-NEXT:    vfsgnj.vv v8, v10, v8, v0.t
+; CHECK-NEXT:    ret
+  %head = insertelement <vscale x 2 x i1> poison, i1 true, i32 0
+  %m = shufflevector <vscale x 2 x i1> %head, <vscale x 2 x i1> poison, <vscale x 2 x i32> zeroinitializer
+  %v = call <vscale x 2 x double> @llvm.vp.nearbyint.nxv2f64(<vscale x 2 x double> %va, <vscale x 2 x i1> %m, i32 %evl)
+  ret <vscale x 2 x double> %v
+}
+
+declare <vscale x 4 x double> @llvm.vp.nearbyint.nxv4f64(<vscale x 4 x double>, <vscale x 4 x i1>, i32)
+
+define <vscale x 4 x double> @vp_nearbyint_nxv4f64(<vscale x 4 x double> %va, <vscale x 4 x i1> %m, i32 zeroext %evl) {
+; CHECK-LABEL: vp_nearbyint_nxv4f64:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vmv1r.v v12, v0
+; CHECK-NEXT:    lui a1, %hi(.LCPI26_0)
+; CHECK-NEXT:    fld ft0, %lo(.LCPI26_0)(a1)
+; CHECK-NEXT:    vsetvli zero, a0, e64, m4, ta, ma
+; CHECK-NEXT:    vfabs.v v16, v8, v0.t
+; CHECK-NEXT:    vsetvli zero, zero, e64, m4, ta, mu
+; CHECK-NEXT:    vmflt.vf v12, v16, ft0, v0.t
+; CHECK-NEXT:    frflags a0
+; CHECK-NEXT:    vsetvli zero, zero, e64, m4, ta, ma
+; CHECK-NEXT:    vmv1r.v v0, v12
+; CHECK-NEXT:    vfcvt.x.f.v v16, v8, v0.t
+; CHECK-NEXT:    vfcvt.f.x.v v16, v16, v0.t
+; CHECK-NEXT:    fsflags a0
+; CHECK-NEXT:    vsetvli zero, zero, e64, m4, ta, mu
+; CHECK-NEXT:    vfsgnj.vv v8, v16, v8, v0.t
+; CHECK-NEXT:    ret
+  %v = call <vscale x 4 x double> @llvm.vp.nearbyint.nxv4f64(<vscale x 4 x double> %va, <vscale x 4 x i1> %m, i32 %evl)
+  ret <vscale x 4 x double> %v
+}
+
+define <vscale x 4 x double> @vp_nearbyint_nxv4f64_unmasked(<vscale x 4 x double> %va, i32 zeroext %evl) {
+; CHECK-LABEL: vp_nearbyint_nxv4f64_unmasked:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    lui a1, %hi(.LCPI27_0)
+; CHECK-NEXT:    fld ft0, %lo(.LCPI27_0)(a1)
+; CHECK-NEXT:    vsetvli zero, a0, e64, m4, ta, ma
+; CHECK-NEXT:    vfabs.v v12, v8
+; CHECK-NEXT:    vmflt.vf v0, v12, ft0
+; CHECK-NEXT:    frflags a0
+; CHECK-NEXT:    vfcvt.x.f.v v12, v8, v0.t
+; CHECK-NEXT:    vfcvt.f.x.v v12, v12, v0.t
+; CHECK-NEXT:    fsflags a0
+; CHECK-NEXT:    vsetvli zero, zero, e64, m4, ta, mu
+; CHECK-NEXT:    vfsgnj.vv v8, v12, v8, v0.t
+; CHECK-NEXT:    ret
+  %head = insertelement <vscale x 4 x i1> poison, i1 true, i32 0
+  %m = shufflevector <vscale x 4 x i1> %head, <vscale x 4 x i1> poison, <vscale x 4 x i32> zeroinitializer
+  %v = call <vscale x 4 x double> @llvm.vp.nearbyint.nxv4f64(<vscale x 4 x double> %va, <vscale x 4 x i1> %m, i32 %evl)
+  ret <vscale x 4 x double> %v
+}
+
+declare <vscale x 7 x double> @llvm.vp.nearbyint.nxv7f64(<vscale x 7 x double>, <vscale x 7 x i1>, i32)
+
+define <vscale x 7 x double> @vp_nearbyint_nxv7f64(<vscale x 7 x double> %va, <vscale x 7 x i1> %m, i32 zeroext %evl) {
+; CHECK-LABEL: vp_nearbyint_nxv7f64:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vmv1r.v v16, v0
+; CHECK-NEXT:    lui a1, %hi(.LCPI28_0)
+; CHECK-NEXT:    fld ft0, %lo(.LCPI28_0)(a1)
+; CHECK-NEXT:    vsetvli zero, a0, e64, m8, ta, ma
+; CHECK-NEXT:    vfabs.v v24, v8, v0.t
+; CHECK-NEXT:    vsetvli zero, zero, e64, m8, ta, mu
+; CHECK-NEXT:    vmflt.vf v16, v24, ft0, v0.t
+; CHECK-NEXT:    frflags a0
+; CHECK-NEXT:    vsetvli zero, zero, e64, m8, ta, ma
+; CHECK-NEXT:    vmv1r.v v0, v16
+; CHECK-NEXT:    vfcvt.x.f.v v24, v8, v0.t
+; CHECK-NEXT:    vfcvt.f.x.v v24, v24, v0.t
+; CHECK-NEXT:    fsflags a0
+; CHECK-NEXT:    vsetvli zero, zero, e64, m8, ta, mu
+; CHECK-NEXT:    vfsgnj.vv v8, v24, v8, v0.t
+; CHECK-NEXT:    ret
+  %v = call <vscale x 7 x double> @llvm.vp.nearbyint.nxv7f64(<vscale x 7 x double> %va, <vscale x 7 x i1> %m, i32 %evl)
+  ret <vscale x 7 x double> %v
+}
+
+define <vscale x 7 x double> @vp_nearbyint_nxv7f64_unmasked(<vscale x 7 x double> %va, i32 zeroext %evl) {
+; CHECK-LABEL: vp_nearbyint_nxv7f64_unmasked:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    lui a1, %hi(.LCPI29_0)
+; CHECK-NEXT:    fld ft0, %lo(.LCPI29_0)(a1)
+; CHECK-NEXT:    vsetvli zero, a0, e64, m8, ta, ma
+; CHECK-NEXT:    vfabs.v v16, v8
+; CHECK-NEXT:    vmflt.vf v0, v16, ft0
+; CHECK-NEXT:    frflags a0
+; CHECK-NEXT:    vfcvt.x.f.v v16, v8, v0.t
+; CHECK-NEXT:    vfcvt.f.x.v v16, v16, v0.t
+; CHECK-NEXT:    fsflags a0
+; CHECK-NEXT:    vsetvli zero, zero, e64, m8, ta, mu
+; CHECK-NEXT:    vfsgnj.vv v8, v16, v8, v0.t
+; CHECK-NEXT:    ret
+  %head = insertelement <vscale x 7 x i1> poison, i1 true, i32 0
+  %m = shufflevector <vscale x 7 x i1> %head, <vscale x 7 x i1> poison, <vscale x 7 x i32> zeroinitializer
+  %v = call <vscale x 7 x double> @llvm.vp.nearbyint.nxv7f64(<vscale x 7 x double> %va, <vscale x 7 x i1> %m, i32 %evl)
+  ret <vscale x 7 x double> %v
+}
+
+declare <vscale x 8 x double> @llvm.vp.nearbyint.nxv8f64(<vscale x 8 x double>, <vscale x 8 x i1>, i32)
+
+define <vscale x 8 x double> @vp_nearbyint_nxv8f64(<vscale x 8 x double> %va, <vscale x 8 x i1> %m, i32 zeroext %evl) {
+; CHECK-LABEL: vp_nearbyint_nxv8f64:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vmv1r.v v16, v0
+; CHECK-NEXT:    lui a1, %hi(.LCPI30_0)
+; CHECK-NEXT:    fld ft0, %lo(.LCPI30_0)(a1)
+; CHECK-NEXT:    vsetvli zero, a0, e64, m8, ta, ma
+; CHECK-NEXT:    vfabs.v v24, v8, v0.t
+; CHECK-NEXT:    vsetvli zero, zero, e64, m8, ta, mu
+; CHECK-NEXT:    vmflt.vf v16, v24, ft0, v0.t
+; CHECK-NEXT:    frflags a0
+; CHECK-NEXT:    vsetvli zero, zero, e64, m8, ta, ma
+; CHECK-NEXT:    vmv1r.v v0, v16
+; CHECK-NEXT:    vfcvt.x.f.v v24, v8, v0.t
+; CHECK-NEXT:    vfcvt.f.x.v v24, v24, v0.t
+; CHECK-NEXT:    fsflags a0
+; CHECK-NEXT:    vsetvli zero, zero, e64, m8, ta, mu
+; CHECK-NEXT:    vfsgnj.vv v8, v24, v8, v0.t
+; CHECK-NEXT:    ret
+  %v = call <vscale x 8 x double> @llvm.vp.nearbyint.nxv8f64(<vscale x 8 x double> %va, <vscale x 8 x i1> %m, i32 %evl)
+  ret <vscale x 8 x double> %v
+}
+
+define <vscale x 8 x double> @vp_nearbyint_nxv8f64_unmasked(<vscale x 8 x double> %va, i32 zeroext %evl) {
+; CHECK-LABEL: vp_nearbyint_nxv8f64_unmasked:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    lui a1, %hi(.LCPI31_0)
+; CHECK-NEXT:    fld ft0, %lo(.LCPI31_0)(a1)
+; CHECK-NEXT:    vsetvli zero, a0, e64, m8, ta, ma
+; CHECK-NEXT:    vfabs.v v16, v8
+; CHECK-NEXT:    vmflt.vf v0, v16, ft0
+; CHECK-NEXT:    frflags a0
+; CHECK-NEXT:    vfcvt.x.f.v v16, v8, v0.t
+; CHECK-NEXT:    vfcvt.f.x.v v16, v16, v0.t
+; CHECK-NEXT:    fsflags a0
+; CHECK-NEXT:    vsetvli zero, zero, e64, m8, ta, mu
+; CHECK-NEXT:    vfsgnj.vv v8, v16, v8, v0.t
+; CHECK-NEXT:    ret
+  %head = insertelement <vscale x 8 x i1> poison, i1 true, i32 0
+  %m = shufflevector <vscale x 8 x i1> %head, <vscale x 8 x i1> poison, <vscale x 8 x i32> zeroinitializer
+  %v = call <vscale x 8 x double> @llvm.vp.nearbyint.nxv8f64(<vscale x 8 x double> %va, <vscale x 8 x i1> %m, i32 %evl)
+  ret <vscale x 8 x double> %v
+}
+
+; Test splitting.
+declare <vscale x 16 x double> @llvm.vp.nearbyint.nxv16f64(<vscale x 16 x double>, <vscale x 16 x i1>, i32)
+
+define <vscale x 16 x double> @vp_nearbyint_nxv16f64(<vscale x 16 x double> %va, <vscale x 16 x i1> %m, i32 zeroext %evl) {
+; CHECK-LABEL: vp_nearbyint_nxv16f64:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vmv1r.v v1, v0
+; CHECK-NEXT:    csrr a1, vlenb
+; CHECK-NEXT:    srli a2, a1, 3
+; CHECK-NEXT:    vsetvli a3, zero, e8, mf4, ta, ma
+; CHECK-NEXT:    vslidedown.vx v2, v0, a2
+; CHECK-NEXT:    sub a2, a0, a1
+; CHECK-NEXT:    sltu a3, a0, a2
+; CHECK-NEXT:    addi a3, a3, -1
+; CHECK-NEXT:    and a2, a3, a2
+; CHECK-NEXT:    lui a3, %hi(.LCPI32_0)
+; CHECK-NEXT:    fld ft0, %lo(.LCPI32_0)(a3)
+; CHECK-NEXT:    vsetvli zero, a2, e64, m8, ta, ma
+; CHECK-NEXT:    vmv1r.v v0, v2
+; CHECK-NEXT:    vfabs.v v24, v16, v0.t
+; CHECK-NEXT:    vsetvli zero, zero, e64, m8, ta, mu
+; CHECK-NEXT:    vmflt.vf v2, v24, ft0, v0.t
+; CHECK-NEXT:    frflags a2
+; CHECK-NEXT:    vsetvli zero, zero, e64, m8, ta, ma
+; CHECK-NEXT:    vmv1r.v v0, v2
+; CHECK-NEXT:    vfcvt.x.f.v v24, v16, v0.t
+; CHECK-NEXT:    vfcvt.f.x.v v24, v24, v0.t
+; CHECK-NEXT:    fsflags a2
+; CHECK-NEXT:    vsetvli zero, zero, e64, m8, ta, mu
+; CHECK-NEXT:    vfsgnj.vv v16, v24, v16, v0.t
+; CHECK-NEXT:    bltu a0, a1, .LBB32_2
+; CHECK-NEXT:  # %bb.1:
+; CHECK-NEXT:    mv a0, a1
+; CHECK-NEXT:  .LBB32_2:
+; CHECK-NEXT:    vsetvli zero, a0, e64, m8, ta, ma
+; CHECK-NEXT:    vmv1r.v v0, v1
+; CHECK-NEXT:    vfabs.v v24, v8, v0.t
+; CHECK-NEXT:    vsetvli zero, zero, e64, m8, ta, mu
+; CHECK-NEXT:    vmflt.vf v1, v24, ft0, v0.t
+; CHECK-NEXT:    frflags a0
+; CHECK-NEXT:    vsetvli zero, zero, e64, m8, ta, ma
+; CHECK-NEXT:    vmv1r.v v0, v1
+; CHECK-NEXT:    vfcvt.x.f.v v24, v8, v0.t
+; CHECK-NEXT:    vfcvt.f.x.v v24, v24, v0.t
+; CHECK-NEXT:    fsflags a0
+; CHECK-NEXT:    vsetvli zero, zero, e64, m8, ta, mu
+; CHECK-NEXT:    vfsgnj.vv v8, v24, v8, v0.t
+; CHECK-NEXT:    ret
+  %v = call <vscale x 16 x double> @llvm.vp.nearbyint.nxv16f64(<vscale x 16 x double> %va, <vscale x 16 x i1> %m, i32 %evl)
+  ret <vscale x 16 x double> %v
+}
+
+define <vscale x 16 x double> @vp_nearbyint_nxv16f64_unmasked(<vscale x 16 x double> %va, i32 zeroext %evl) {
+; CHECK-LABEL: vp_nearbyint_nxv16f64_unmasked:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    csrr a1, vlenb
+; CHECK-NEXT:    sub a2, a0, a1
+; CHECK-NEXT:    lui a3, %hi(.LCPI33_0)
+; CHECK-NEXT:    fld ft0, %lo(.LCPI33_0)(a3)
+; CHECK-NEXT:    sltu a3, a0, a2
+; CHECK-NEXT:    addi a3, a3, -1
+; CHECK-NEXT:    and a2, a3, a2
+; CHECK-NEXT:    vsetvli zero, a2, e64, m8, ta, ma
+; CHECK-NEXT:    vfabs.v v24, v16
+; CHECK-NEXT:    vmflt.vf v0, v24, ft0
+; CHECK-NEXT:    frflags a2
+; CHECK-NEXT:    vfcvt.x.f.v v24, v16, v0.t
+; CHECK-NEXT:    vfcvt.f.x.v v24, v24, v0.t
+; CHECK-NEXT:    fsflags a2
+; CHECK-NEXT:    vsetvli zero, zero, e64, m8, ta, mu
+; CHECK-NEXT:    vfsgnj.vv v16, v24, v16, v0.t
+; CHECK-NEXT:    bltu a0, a1, .LBB33_2
+; CHECK-NEXT:  # %bb.1:
+; CHECK-NEXT:    mv a0, a1
+; CHECK-NEXT:  .LBB33_2:
+; CHECK-NEXT:    vsetvli zero, a0, e64, m8, ta, ma
+; CHECK-NEXT:    vfabs.v v24, v8
+; CHECK-NEXT:    vmflt.vf v0, v24, ft0
+; CHECK-NEXT:    frflags a0
+; CHECK-NEXT:    vfcvt.x.f.v v24, v8, v0.t
+; CHECK-NEXT:    vfcvt.f.x.v v24, v24, v0.t
+; CHECK-NEXT:    fsflags a0
+; CHECK-NEXT:    vsetvli zero, zero, e64, m8, ta, mu
+; CHECK-NEXT:    vfsgnj.vv v8, v24, v8, v0.t
+; CHECK-NEXT:    ret
+  %head = insertelement <vscale x 16 x i1> poison, i1 true, i32 0
+  %m = shufflevector <vscale x 16 x i1> %head, <vscale x 16 x i1> poison, <vscale x 16 x i32> zeroinitializer
+  %v = call <vscale x 16 x double> @llvm.vp.nearbyint.nxv16f64(<vscale x 16 x double> %va, <vscale x 16 x i1> %m, i32 %evl)
+  ret <vscale x 16 x double> %v
+}

llvm/unittests/IR/VPIntrinsicTest.cpp

std::unique_ptr<Module> createVPDeclarationModule() {
     Str << " declare <8 x float> @llvm.vp.roundeven.v8f32(<8 x float>, <8 x "
            "i1>, "
            "i32)";
     Str << " declare <8 x float> @llvm.vp.roundtozero.v8f32(<8 x float>, <8 x "
            "i1>, "
            "i32)";
     Str << " declare <8 x float> @llvm.vp.rint.v8f32(<8 x float>, <8 x i1>, "
            "i32)";
+    Str << " declare <8 x float> @llvm.vp.nearbyint.v8f32(<8 x float>, <8 x "
+           "i1>, "
+           "i32)";
     Str << " declare <8 x float> @llvm.vp.ceil.v8f32(<8 x float>, <8 x i1>, "
            "i32)";
     Str << " declare <8 x float> @llvm.vp.fneg.v8f32(<8 x float>, <8 x i1>, "
            "i32)";
     Str << " declare <8 x float> @llvm.vp.fabs.v8f32(<8 x float>, <8 x i1>, "
            "i32)";
     Str << " declare <8 x float> @llvm.vp.sqrt.v8f32(<8 x float>, <8 x i1>, "
            "i32)";