llvm-project/llvm/lib/Target/ARM/ARMInstrMVE.td

//===-- ARMInstrMVE.td - MVE support for ARM ---------------*- tablegen -*-===//
//
// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
// See https://llvm.org/LICENSE.txt for license information.
// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
//
//===----------------------------------------------------------------------===//
//
// This file describes the ARM MVE instruction set.
//
//===----------------------------------------------------------------------===//

class ExpandImmAsmOp<string shift> : AsmOperandClass {
  let Name = !strconcat("ExpandImm", shift);
  let PredicateMethod = !strconcat("isExpImm<", shift, ">");
  let RenderMethod = "addImmOperands";
}
class InvertedExpandImmAsmOp<string shift, string size> : AsmOperandClass {
  let Name = !strconcat("InvertedExpandImm", shift, "_", size);
  let PredicateMethod = !strconcat("isInvertedExpImm<", shift, ",", size, ">");
  let RenderMethod = "addImmOperands";
}

class ExpandImm<string shift> : Operand<i32> {
  let ParserMatchClass = ExpandImmAsmOp<shift>;
  let EncoderMethod = !strconcat("getExpandedImmOpValue<",shift,",false>");
  let DecoderMethod = !strconcat("DecodeExpandedImmOperand<",shift,">");
  let PrintMethod = "printExpandedImmOperand";
}
class InvertedExpandImm<string shift, string size> : Operand<i32> {
  let ParserMatchClass = InvertedExpandImmAsmOp<shift, size>;
  let EncoderMethod = !strconcat("getExpandedImmOpValue<",shift,",true>");
  let PrintMethod = "printExpandedImmOperand";
  // No decoder method needed, because this operand type is only used
  // by aliases (VAND and VORN)
}

def expzero00 : ExpandImm<"0">;
def expzero08 : ExpandImm<"8">;
def expzero16 : ExpandImm<"16">;
def expzero24 : ExpandImm<"24">;

def expzero00inv16 : InvertedExpandImm<"0", "16">;
def expzero08inv16 : InvertedExpandImm<"8", "16">;

def expzero00inv32 : InvertedExpandImm<"0", "32">;
def expzero08inv32 : InvertedExpandImm<"8", "32">;
def expzero16inv32 : InvertedExpandImm<"16", "32">;
def expzero24inv32 : InvertedExpandImm<"24", "32">;

// VPT condition mask
def vpt_mask : Operand<i32> {
  let PrintMethod = "printVPTMask";
  let ParserMatchClass = it_mask_asmoperand;
  let EncoderMethod = "getVPTMaskOpValue";
  let DecoderMethod = "DecodeVPTMaskOperand";
}

// VPT/VCMP restricted predicate for sign invariant types
def pred_restricted_i_asmoperand : AsmOperandClass {
  let Name = "CondCodeRestrictedI";
  let RenderMethod = "addITCondCodeOperands";
  let PredicateMethod = "isITCondCodeRestrictedI";
  let ParserMethod = "parseITCondCode";
  let DiagnosticString = "condition code for sign-independent integer "#
                         "comparison must be EQ or NE";
}

// VPT/VCMP restricted predicate for signed types
def pred_restricted_s_asmoperand : AsmOperandClass {
  let Name = "CondCodeRestrictedS";
  let RenderMethod = "addITCondCodeOperands";
  let PredicateMethod = "isITCondCodeRestrictedS";
  let ParserMethod = "parseITCondCode";
  let DiagnosticString = "condition code for signed integer "#
                         "comparison must be EQ, NE, LT, GT, LE or GE";
}

// VPT/VCMP restricted predicate for unsigned types
def pred_restricted_u_asmoperand : AsmOperandClass {
  let Name = "CondCodeRestrictedU";
  let RenderMethod = "addITCondCodeOperands";
  let PredicateMethod = "isITCondCodeRestrictedU";
  let ParserMethod = "parseITCondCode";
  let DiagnosticString = "condition code for unsigned integer "#
                         "comparison must be EQ, NE, HS or HI";
}

// VPT/VCMP restricted predicate for floating point
def pred_restricted_fp_asmoperand : AsmOperandClass {
  let Name = "CondCodeRestrictedFP";
  let RenderMethod = "addITCondCodeOperands";
  let PredicateMethod = "isITCondCodeRestrictedFP";
  let ParserMethod = "parseITCondCode";
  let DiagnosticString = "condition code for floating-point "#
                         "comparison must be EQ, NE, LT, GT, LE or GE";
}

class VCMPPredicateOperand : Operand<i32>;

def pred_basic_i : VCMPPredicateOperand {
  let PrintMethod = "printMandatoryRestrictedPredicateOperand";
  let ParserMatchClass = pred_restricted_i_asmoperand;
  let DecoderMethod = "DecodeRestrictedIPredicateOperand";
  let EncoderMethod = "getRestrictedCondCodeOpValue";
}

def pred_basic_u : VCMPPredicateOperand {
  let PrintMethod = "printMandatoryRestrictedPredicateOperand";
  let ParserMatchClass = pred_restricted_u_asmoperand;
  let DecoderMethod = "DecodeRestrictedUPredicateOperand";
  let EncoderMethod = "getRestrictedCondCodeOpValue";
}

def pred_basic_s : VCMPPredicateOperand {
  let PrintMethod = "printMandatoryRestrictedPredicateOperand";
  let ParserMatchClass = pred_restricted_s_asmoperand;
  let DecoderMethod = "DecodeRestrictedSPredicateOperand";
  let EncoderMethod = "getRestrictedCondCodeOpValue";
}

def pred_basic_fp : VCMPPredicateOperand {
  let PrintMethod = "printMandatoryRestrictedPredicateOperand";
  let ParserMatchClass = pred_restricted_fp_asmoperand;
  let DecoderMethod = "DecodeRestrictedFPPredicateOperand";
  let EncoderMethod = "getRestrictedCondCodeOpValue";
}

// Register list operands for interleaving load/stores
def VecList2QAsmOperand : AsmOperandClass {
  let Name = "VecListTwoMQ";
  let ParserMethod = "parseVectorList";
  let RenderMethod = "addMVEVecListOperands";
  let DiagnosticString = "operand must be a list of two consecutive "#
                         "q-registers in range [q0,q7]";
}

def VecList2Q : RegisterOperand<QQPR, "printMVEVectorListTwoQ"> {
  let ParserMatchClass = VecList2QAsmOperand;
  let PrintMethod = "printMVEVectorList<2>";
}

def VecList4QAsmOperand : AsmOperandClass {
  let Name = "VecListFourMQ";
  let ParserMethod = "parseVectorList";
  let RenderMethod = "addMVEVecListOperands";
  let DiagnosticString = "operand must be a list of four consecutive "#
                         "q-registers in range [q0,q7]";
}

def VecList4Q : RegisterOperand<QQQQPR, "printMVEVectorListFourQ"> {
  let ParserMatchClass = VecList4QAsmOperand;
  let PrintMethod = "printMVEVectorList<4>";
}

// taddrmode_imm7  := reg[r0-r7] +/- (imm7 << shift)
class TMemImm7ShiftOffsetAsmOperand<int shift> : AsmOperandClass {
  let Name = "TMemImm7Shift"#shift#"Offset";
  let PredicateMethod = "isMemImm7ShiftedOffset<"#shift#",ARM::tGPRRegClassID>";
  let RenderMethod = "addMemImmOffsetOperands";
}

class taddrmode_imm7<int shift> : MemOperand,
    ComplexPattern<i32, 2, "SelectTAddrModeImm7<"#shift#">", []>  {
  let ParserMatchClass = TMemImm7ShiftOffsetAsmOperand<shift>;
  // They are printed the same way as the T2 imm8 version
  let PrintMethod = "printT2AddrModeImm8Operand<false>";
  // This can also be the same as the T2 version.
  let EncoderMethod = "getT2AddrModeImmOpValue<7,"#shift#">";
  let DecoderMethod = "DecodeTAddrModeImm7<"#shift#">";
  let MIOperandInfo = (ops tGPR:$base, i32imm:$offsimm);
}

// t2addrmode_imm7  := reg +/- (imm7)
class MemImm7ShiftOffsetAsmOperand<int shift> : AsmOperandClass {
  let Name = "MemImm7Shift"#shift#"Offset";
  let PredicateMethod = "isMemImm7ShiftedOffset<" # shift #
                        ",ARM::GPRnopcRegClassID>";
  let RenderMethod = "addMemImmOffsetOperands";
}

def MemImm7Shift0OffsetAsmOperand : MemImm7ShiftOffsetAsmOperand<0>;
def MemImm7Shift1OffsetAsmOperand : MemImm7ShiftOffsetAsmOperand<1>;
def MemImm7Shift2OffsetAsmOperand : MemImm7ShiftOffsetAsmOperand<2>;
class T2AddrMode_Imm7<int shift> : MemOperand,
      ComplexPattern<i32, 2, "SelectT2AddrModeImm7<"#shift#">", []> {
  let EncoderMethod = "getT2AddrModeImmOpValue<7,"#shift#">";
  let DecoderMethod = "DecodeT2AddrModeImm7<"#shift#", 0>";
  let ParserMatchClass =
    !cast<AsmOperandClass>("MemImm7Shift"#shift#"OffsetAsmOperand");
  let MIOperandInfo = (ops GPRnopc:$base, i32imm:$offsimm);
}

class t2addrmode_imm7<int shift> : T2AddrMode_Imm7<shift> {
  // They are printed the same way as the imm8 version
  let PrintMethod = "printT2AddrModeImm8Operand<false>";
}

class MemImm7ShiftOffsetWBAsmOperand<int shift> : AsmOperandClass {
  let Name = "MemImm7Shift"#shift#"OffsetWB";
  let PredicateMethod = "isMemImm7ShiftedOffset<" # shift #
                        ",ARM::rGPRRegClassID>";
  let RenderMethod = "addMemImmOffsetOperands";
}

def MemImm7Shift0OffsetWBAsmOperand : MemImm7ShiftOffsetWBAsmOperand<0>;
def MemImm7Shift1OffsetWBAsmOperand : MemImm7ShiftOffsetWBAsmOperand<1>;
def MemImm7Shift2OffsetWBAsmOperand : MemImm7ShiftOffsetWBAsmOperand<2>;

class t2addrmode_imm7_pre<int shift> : T2AddrMode_Imm7<shift> {
  // They are printed the same way as the imm8 version
  let PrintMethod = "printT2AddrModeImm8Operand<true>";
  let ParserMatchClass =
    !cast<AsmOperandClass>("MemImm7Shift"#shift#"OffsetWBAsmOperand");
  let DecoderMethod = "DecodeT2AddrModeImm7<"#shift#", 1>";
  let MIOperandInfo = (ops rGPR:$base, i32imm:$offsim);
}

class t2am_imm7shiftOffsetAsmOperand<int shift>
  : AsmOperandClass { let Name = "Imm7Shift"#shift; }
def t2am_imm7shift0OffsetAsmOperand : t2am_imm7shiftOffsetAsmOperand<0>;
def t2am_imm7shift1OffsetAsmOperand : t2am_imm7shiftOffsetAsmOperand<1>;
def t2am_imm7shift2OffsetAsmOperand : t2am_imm7shiftOffsetAsmOperand<2>;

class t2am_imm7_offset<int shift> : MemOperand,
      ComplexPattern<i32, 1, "SelectT2AddrModeImm7Offset<"#shift#">",
                     [], [SDNPWantRoot]> {
  // They are printed the same way as the imm8 version
  let PrintMethod = "printT2AddrModeImm8OffsetOperand";
  let ParserMatchClass =
    !cast<AsmOperandClass>("t2am_imm7shift"#shift#"OffsetAsmOperand");
  let EncoderMethod = "getT2ScaledImmOpValue<7,"#shift#">";
  let DecoderMethod = "DecodeT2Imm7<"#shift#">";
}

// Operands for gather/scatter loads of the form [Rbase, Qoffsets]
class MemRegRQOffsetAsmOperand<int shift> : AsmOperandClass {
  let Name = "MemRegRQS"#shift#"Offset";
  let PredicateMethod = "isMemRegRQOffset<"#shift#">";
  let RenderMethod = "addMemRegRQOffsetOperands";
}

def MemRegRQS0OffsetAsmOperand : MemRegRQOffsetAsmOperand<0>;
def MemRegRQS1OffsetAsmOperand : MemRegRQOffsetAsmOperand<1>;
def MemRegRQS2OffsetAsmOperand : MemRegRQOffsetAsmOperand<2>;
def MemRegRQS3OffsetAsmOperand : MemRegRQOffsetAsmOperand<3>;

// mve_addr_rq_shift  := reg + vreg{ << UXTW #shift}
class mve_addr_rq_shift<int shift> : MemOperand {
  let EncoderMethod = "getMveAddrModeRQOpValue";
  let PrintMethod = "printMveAddrModeRQOperand<"#shift#">";
  let ParserMatchClass =
    !cast<AsmOperandClass>("MemRegRQS"#shift#"OffsetAsmOperand");
  let DecoderMethod = "DecodeMveAddrModeRQ";
  let MIOperandInfo = (ops GPRnopc:$base, MQPR:$offsreg);
}

class MemRegQOffsetAsmOperand<int shift> : AsmOperandClass {
  let Name = "MemRegQS"#shift#"Offset";
  let PredicateMethod = "isMemRegQOffset<"#shift#">";
  let RenderMethod = "addMemImmOffsetOperands";
}

def MemRegQS2OffsetAsmOperand : MemRegQOffsetAsmOperand<2>;
def MemRegQS3OffsetAsmOperand : MemRegQOffsetAsmOperand<3>;

// mve_addr_q_shift  := vreg {+ #imm7s2/4}
class mve_addr_q_shift<int shift> : MemOperand {
  let EncoderMethod = "getMveAddrModeQOpValue<"#shift#">";
  // Can be printed same way as other reg + imm operands
  let PrintMethod = "printT2AddrModeImm8Operand<false>";
  let ParserMatchClass =
    !cast<AsmOperandClass>("MemRegQS"#shift#"OffsetAsmOperand");
  let DecoderMethod = "DecodeMveAddrModeQ<"#shift#">";
  let MIOperandInfo = (ops MQPR:$base, i32imm:$imm);
}

// A family of classes wrapping up information about the vector types
// used by MVE.
class MVEVectorVTInfo<ValueType vec, ValueType pred, bits<2> size,
                      string suffixletter, bit unsigned> {
  // The LLVM ValueType representing the vector, so we can use it in
  // ISel patterns.
  ValueType Vec = vec;

  // An LLVM ValueType representing a corresponding vector of
  // predicate bits, for use in ISel patterns that handle an IR
  // intrinsic describing the predicated form of the instruction.
  //
  // Usually, for a vector of N things, this will be vNi1. But for
  // vectors of 2 values, we make an exception, and use v4i1 instead
  // of v2i1. Rationale: MVE codegen doesn't support doing all the
  // auxiliary operations on v2i1 (vector shuffles etc), and also,
  // there's no MVE compare instruction that will _generate_ v2i1
  // directly.
  ValueType Pred = pred;

  // The most common representation of the vector element size in MVE
  // instruction encodings: a 2-bit value V representing an (8<<V)-bit
  // vector element.
  bits<2> Size = size;

  // For vectors explicitly mentioning a signedness of integers: 0 for
  // signed and 1 for unsigned. For anything else, undefined.
  bit Unsigned = unsigned;

  // The number of bits in a vector element, in integer form.
  int LaneBits = !shl(8, Size);

  // The suffix used in assembly language on an instruction operating
  // on this lane if it only cares about number of bits.
  string BitsSuffix = !cast<string>(LaneBits);

  // The suffix used on an instruction that mentions the whole type.
  string Suffix = suffixletter ## BitsSuffix;
}

// Integer vector types that don't treat signed and unsigned differently.
def MVE_v16i8 : MVEVectorVTInfo<v16i8, v16i1, 0b00, "i",  ?>;
def MVE_v8i16 : MVEVectorVTInfo<v8i16, v8i1,  0b01, "i", ?>;
def MVE_v4i32 : MVEVectorVTInfo<v4i32, v4i1,  0b10, "i", ?>;
def MVE_v2i64 : MVEVectorVTInfo<v2i64, v4i1,  0b11, "i", ?>;

// Explicitly signed and unsigned integer vectors. They map to the
// same set of LLVM ValueTypes as above, but are represented
// differently in assembly and instruction encodings.
def MVE_v16s8 : MVEVectorVTInfo<v16i8, v16i1, 0b00, "s",  0b0>;
def MVE_v8s16 : MVEVectorVTInfo<v8i16, v8i1,  0b01, "s", 0b0>;
def MVE_v4s32 : MVEVectorVTInfo<v4i32, v4i1,  0b10, "s", 0b0>;
def MVE_v2s64 : MVEVectorVTInfo<v2i64, v4i1,  0b11, "s", 0b0>;
def MVE_v16u8 : MVEVectorVTInfo<v16i8, v16i1, 0b00, "u",  0b1>;
def MVE_v8u16 : MVEVectorVTInfo<v8i16, v8i1,  0b01, "u", 0b1>;
def MVE_v4u32 : MVEVectorVTInfo<v4i32, v4i1,  0b10, "u", 0b1>;
def MVE_v2u64 : MVEVectorVTInfo<v2i64, v4i1,  0b11, "u", 0b1>;

// FP vector types.
def MVE_v8f16 : MVEVectorVTInfo<v8f16, v8i1,  0b01, "f", ?>;
def MVE_v4f32 : MVEVectorVTInfo<v4f32, v4i1,  0b10, "f", ?>;
def MVE_v2f64 : MVEVectorVTInfo<v2f64, v4i1,  0b11, "f", ?>;

// --------- Start of base classes for the instructions themselves

class MVE_MI<dag oops, dag iops, InstrItinClass itin, string asm,
             string ops, string cstr, list<dag> pattern>
  : Thumb2XI<oops, iops, AddrModeNone, 4, itin, !strconcat(asm, "\t", ops), cstr,
             pattern>,
    Requires<[HasMVEInt]> {
  let D = MVEDomain;
  let DecoderNamespace = "MVE";
}

// MVE_p is used for most predicated instructions, to add the cluster
// of input operands that provides the VPT suffix (none, T or E) and
// the input predicate register.
class MVE_p<dag oops, dag iops, InstrItinClass itin, string iname,
            string suffix, string ops, vpred_ops vpred, string cstr,
            list<dag> pattern=[]>
  : MVE_MI<oops, !con(iops, (ins vpred:$vp)), itin,
           // If the instruction has a suffix, like vadd.f32, then the
           // VPT predication suffix goes before the dot, so the full
           // name has to be "vadd${vp}.f32".
           !strconcat(iname, "${vp}",
                      !if(!eq(suffix, ""), "", !strconcat(".", suffix))),
           ops, !strconcat(cstr, vpred.vpred_constraint), pattern> {
  let Inst{31-29} = 0b111;
  let Inst{27-26} = 0b11;
}

class MVE_f<dag oops, dag iops, InstrItinClass itin, string iname,
            string suffix, string ops, vpred_ops vpred, string cstr,
            list<dag> pattern=[]>
  : MVE_p<oops, iops, itin, iname, suffix, ops, vpred, cstr, pattern> {
  let Predicates = [HasMVEFloat];
}

class MVE_MI_with_pred<dag oops, dag iops, InstrItinClass itin, string asm,
                       string ops, string cstr, list<dag> pattern>
  : Thumb2I<oops, iops, AddrModeNone, 4, itin, asm, !strconcat("\t", ops), cstr,
             pattern>,
    Requires<[HasV8_1MMainline, HasMVEInt]> {
  let D = MVEDomain;
  let DecoderNamespace = "MVE";
}

class MVE_VMOV_lane_base<dag oops, dag iops, InstrItinClass itin, string asm,
                         string suffix, string ops, string cstr,
                         list<dag> pattern>
  : Thumb2I<oops, iops, AddrModeNone, 4, itin, asm,
            !if(!eq(suffix, ""), "", "." # suffix) # "\t" # ops,
            cstr, pattern>,
    Requires<[HasV8_1MMainline, HasMVEInt]> {
  let D = MVEDomain;
  let DecoderNamespace = "MVE";
}

class MVE_ScalarShift<string iname, dag oops, dag iops, string asm, string cstr,
            list<dag> pattern=[]>
  : MVE_MI_with_pred<oops, iops, NoItinerary, iname, asm, cstr, pattern> {
  let Inst{31-20} = 0b111010100101;
  let Inst{8} = 0b1;

}

class MVE_ScalarShiftSingleReg<string iname, dag iops, string asm, string cstr,
                    list<dag> pattern=[]>
  : MVE_ScalarShift<iname, (outs rGPR:$RdaDest), iops, asm, cstr, pattern> {
  bits<4> RdaDest;

  let Inst{19-16} = RdaDest{3-0};
}

class MVE_ScalarShiftSRegImm<string iname, bits<2> op5_4>
  : MVE_ScalarShiftSingleReg<iname, (ins rGPR:$RdaSrc, long_shift:$imm),
                     "$RdaSrc, $imm", "$RdaDest = $RdaSrc",
                     [(set rGPR:$RdaDest,
                          (i32 (!cast<Intrinsic>("int_arm_mve_" # iname)
                                    (i32 rGPR:$RdaSrc), (i32 imm:$imm))))]> {
  bits<5> imm;

  let Inst{15} = 0b0;
  let Inst{14-12} = imm{4-2};
  let Inst{11-8} = 0b1111;
  let Inst{7-6} = imm{1-0};
  let Inst{5-4} = op5_4{1-0};
  let Inst{3-0} = 0b1111;
}

def MVE_SQSHL : MVE_ScalarShiftSRegImm<"sqshl", 0b11>;
def MVE_SRSHR : MVE_ScalarShiftSRegImm<"srshr", 0b10>;
def MVE_UQSHL : MVE_ScalarShiftSRegImm<"uqshl", 0b00>;
def MVE_URSHR : MVE_ScalarShiftSRegImm<"urshr", 0b01>;

class MVE_ScalarShiftSRegReg<string iname, bits<2> op5_4>
  : MVE_ScalarShiftSingleReg<iname, (ins rGPR:$RdaSrc, rGPR:$Rm),
                     "$RdaSrc, $Rm", "$RdaDest = $RdaSrc",
                     [(set rGPR:$RdaDest,
                         (i32 (!cast<Intrinsic>("int_arm_mve_" # iname)
                                   (i32 rGPR:$RdaSrc), (i32 rGPR:$Rm))))]> {
  bits<4> Rm;

  let Inst{15-12} = Rm{3-0};
  let Inst{11-8} = 0b1111;
  let Inst{7-6} = 0b00;
  let Inst{5-4} = op5_4{1-0};
  let Inst{3-0} = 0b1101;

  let Unpredictable{8-6} = 0b111;
}

def MVE_SQRSHR : MVE_ScalarShiftSRegReg<"sqrshr", 0b10>;
def MVE_UQRSHL : MVE_ScalarShiftSRegReg<"uqrshl", 0b00>;

class MVE_ScalarShiftDoubleReg<string iname, dag iops, string asm,
                               string cstr, list<dag> pattern=[]>
  : MVE_ScalarShift<iname, (outs tGPREven:$RdaLo, tGPROdd:$RdaHi),
                    iops, asm, cstr, pattern> {
  bits<4> RdaLo;
  bits<4> RdaHi;

  let Inst{19-17} = RdaLo{3-1};
  let Inst{11-9} = RdaHi{3-1};
}

class MVE_ScalarShiftDRegImm<string iname, bits<2> op5_4, bit op16,
                             list<dag> pattern=[]>
  : MVE_ScalarShiftDoubleReg<
      iname, (ins tGPREven:$RdaLo_src, tGPROdd:$RdaHi_src, long_shift:$imm),
      "$RdaLo, $RdaHi, $imm", "$RdaLo = $RdaLo_src,$RdaHi = $RdaHi_src",
      pattern> {
  bits<5> imm;

  let Inst{16} = op16;
  let Inst{15} = 0b0;
  let Inst{14-12} = imm{4-2};
  let Inst{7-6} = imm{1-0};
  let Inst{5-4} = op5_4{1-0};
  let Inst{3-0} = 0b1111;
}

class MVE_ScalarShiftDRegRegBase<string iname, dag iops, string asm,
                                 bit op5, bit op16, list<dag> pattern=[]>
  : MVE_ScalarShiftDoubleReg<
     iname, iops, asm, "@earlyclobber $RdaHi,@earlyclobber $RdaLo,"
                       "$RdaLo = $RdaLo_src,$RdaHi = $RdaHi_src",
     pattern> {
  bits<4> Rm;

  let Inst{16} = op16;
  let Inst{15-12} = Rm{3-0};
  let Inst{6} = 0b0;
  let Inst{5} = op5;
  let Inst{4} = 0b0;
  let Inst{3-0} = 0b1101;

  // Custom decoder method because of the following overlapping encodings:
  // ASRL and SQRSHR
  // LSLL and UQRSHL
  // SQRSHRL and SQRSHR
  // UQRSHLL and UQRSHL
  let DecoderMethod = "DecodeMVEOverlappingLongShift";
}

class MVE_ScalarShiftDRegReg<string iname, bit op5, list<dag> pattern=[]>
  : MVE_ScalarShiftDRegRegBase<
     iname, (ins tGPREven:$RdaLo_src, tGPROdd:$RdaHi_src, rGPR:$Rm),
     "$RdaLo, $RdaHi, $Rm", op5, 0b0, pattern> {

  let Inst{7} = 0b0;
}

class MVE_ScalarShiftDRegRegWithSat<string iname, bit op5, list<dag> pattern=[]>
  : MVE_ScalarShiftDRegRegBase<
     iname, (ins tGPREven:$RdaLo_src, tGPROdd:$RdaHi_src, rGPR:$Rm, saturateop:$sat),
     "$RdaLo, $RdaHi, $sat, $Rm", op5, 0b1, pattern> {
  bit sat;

  let Inst{7} = sat;
}

def MVE_ASRLr   : MVE_ScalarShiftDRegReg<"asrl",    0b1,  [(set tGPREven:$RdaLo, tGPROdd:$RdaHi,
                                        (ARMasrl tGPREven:$RdaLo_src,
                                        tGPROdd:$RdaHi_src, rGPR:$Rm))]>;
def MVE_ASRLi   : MVE_ScalarShiftDRegImm<"asrl",    0b10, ?, [(set tGPREven:$RdaLo, tGPROdd:$RdaHi,
                                        (ARMasrl tGPREven:$RdaLo_src,
                                        tGPROdd:$RdaHi_src, (i32 long_shift:$imm)))]>;
def MVE_LSLLr   : MVE_ScalarShiftDRegReg<"lsll",    0b0,  [(set tGPREven:$RdaLo, tGPROdd:$RdaHi,
                                        (ARMlsll tGPREven:$RdaLo_src,
                                        tGPROdd:$RdaHi_src, rGPR:$Rm))]>;
def MVE_LSLLi   : MVE_ScalarShiftDRegImm<"lsll",    0b00, ?, [(set tGPREven:$RdaLo, tGPROdd:$RdaHi,
                                        (ARMlsll tGPREven:$RdaLo_src,
                                        tGPROdd:$RdaHi_src, (i32 long_shift:$imm)))]>;
def MVE_LSRL    : MVE_ScalarShiftDRegImm<"lsrl",    0b01, ?, [(set tGPREven:$RdaLo, tGPROdd:$RdaHi,
                                        (ARMlsrl tGPREven:$RdaLo_src,
                                        tGPROdd:$RdaHi_src, (i32 long_shift:$imm)))]>;

def MVE_SQRSHRL : MVE_ScalarShiftDRegRegWithSat<"sqrshrl", 0b1>;
def MVE_SQSHLL  : MVE_ScalarShiftDRegImm<"sqshll",  0b11, 0b1>;
def MVE_SRSHRL  : MVE_ScalarShiftDRegImm<"srshrl",  0b10, 0b1>;

def MVE_UQRSHLL : MVE_ScalarShiftDRegRegWithSat<"uqrshll", 0b0>;
def MVE_UQSHLL  : MVE_ScalarShiftDRegImm<"uqshll",  0b00, 0b1>;
def MVE_URSHRL  : MVE_ScalarShiftDRegImm<"urshrl",  0b01, 0b1>;

// start of mve_rDest instructions

class MVE_rDest<dag oops, dag iops, InstrItinClass itin,
                string iname, string suffix,
                string ops, string cstr, list<dag> pattern=[]>
// Always use vpred_n and not vpred_r: with the output register being
// a GPR and not a vector register, there can't be any question of
// what to put in its inactive lanes.
  : MVE_p<oops, iops, itin, iname, suffix, ops, vpred_n, cstr, pattern> {

  let Inst{25-23} = 0b101;
  let Inst{11-9} = 0b111;
  let Inst{4} = 0b0;
}

class MVE_VABAV<string suffix, bit U, bits<2> size, list<dag> pattern=[]>
  : MVE_rDest<(outs rGPR:$Rda), (ins rGPR:$Rda_src, MQPR:$Qn, MQPR:$Qm),
              NoItinerary, "vabav", suffix, "$Rda, $Qn, $Qm", "$Rda = $Rda_src",
              pattern> {
  bits<4> Qm;
  bits<4> Qn;
  bits<4> Rda;

  let Inst{28} = U;
  let Inst{22} = 0b0;
  let Inst{21-20} = size{1-0};
  let Inst{19-17} = Qn{2-0};
  let Inst{16} = 0b0;
  let Inst{15-12} = Rda{3-0};
  let Inst{8} = 0b1;
  let Inst{7} = Qn{3};
  let Inst{6} = 0b0;
  let Inst{5} = Qm{3};
  let Inst{3-1} = Qm{2-0};
  let Inst{0} = 0b1;
}

def MVE_VABAVs8  : MVE_VABAV<"s8", 0b0, 0b00>;
def MVE_VABAVs16 : MVE_VABAV<"s16", 0b0, 0b01>;
def MVE_VABAVs32 : MVE_VABAV<"s32", 0b0, 0b10>;
def MVE_VABAVu8  : MVE_VABAV<"u8", 0b1, 0b00>;
def MVE_VABAVu16 : MVE_VABAV<"u16", 0b1, 0b01>;
def MVE_VABAVu32 : MVE_VABAV<"u32", 0b1, 0b10>;

class MVE_VADDV<string iname, string suffix, dag iops, string cstr,
              bit A, bit U, bits<2> size, list<dag> pattern=[]>
  : MVE_rDest<(outs tGPREven:$Rda), iops, NoItinerary,
              iname, suffix, "$Rda, $Qm", cstr, pattern> {
  bits<3> Qm;
  bits<4> Rda;

  let Inst{28} = U;
  let Inst{22-20} = 0b111;
  let Inst{19-18} = size{1-0};
  let Inst{17-16} = 0b01;
  let Inst{15-13} = Rda{3-1};
  let Inst{12} = 0b0;
  let Inst{8-6} = 0b100;
  let Inst{5} = A;
  let Inst{3-1} = Qm{2-0};
  let Inst{0} = 0b0;
}

multiclass MVE_VADDV_A<string suffix, bit U, bits<2> size,
                       list<dag> pattern=[]> {
  def acc    : MVE_VADDV<"vaddva", suffix,
                         (ins tGPREven:$Rda_src, MQPR:$Qm), "$Rda = $Rda_src",
                         0b1, U, size, pattern>;
  def no_acc : MVE_VADDV<"vaddv", suffix,
                         (ins MQPR:$Qm), "",
                         0b0, U, size, pattern>;
}

defm MVE_VADDVs8  : MVE_VADDV_A<"s8",  0b0, 0b00>;
defm MVE_VADDVs16 : MVE_VADDV_A<"s16", 0b0, 0b01>;
defm MVE_VADDVs32 : MVE_VADDV_A<"s32", 0b0, 0b10>;
defm MVE_VADDVu8  : MVE_VADDV_A<"u8",  0b1, 0b00>;
defm MVE_VADDVu16 : MVE_VADDV_A<"u16", 0b1, 0b01>;
defm MVE_VADDVu32 : MVE_VADDV_A<"u32", 0b1, 0b10>;

let Predicates = [HasMVEInt] in {
  def : Pat<(i32 (vecreduce_add (v4i32 MQPR:$src))), (i32 (MVE_VADDVu32no_acc $src))>;
  def : Pat<(i32 (vecreduce_add (v8i16 MQPR:$src))), (i32 (MVE_VADDVu16no_acc $src))>;
  def : Pat<(i32 (vecreduce_add (v16i8 MQPR:$src))), (i32 (MVE_VADDVu8no_acc $src))>;
  def : Pat<(i32 (add (i32 (vecreduce_add (v4i32 MQPR:$src1))), (i32 tGPR:$src2))),
            (i32 (MVE_VADDVu32acc $src2, $src1))>;
  def : Pat<(i32 (add (i32 (vecreduce_add (v8i16 MQPR:$src1))), (i32 tGPR:$src2))),
            (i32 (MVE_VADDVu16acc $src2, $src1))>;
  def : Pat<(i32 (add (i32 (vecreduce_add (v16i8 MQPR:$src1))), (i32 tGPR:$src2))),
            (i32 (MVE_VADDVu8acc $src2, $src1))>;

}

class MVE_VADDLV<string iname, string suffix, dag iops, string cstr,
               bit A, bit U, list<dag> pattern=[]>
  : MVE_rDest<(outs tGPREven:$RdaLo, tGPROdd:$RdaHi), iops, NoItinerary, iname,
              suffix, "$RdaLo, $RdaHi, $Qm", cstr, pattern> {
  bits<3> Qm;
  bits<4> RdaLo;
  bits<4> RdaHi;

  let Inst{28} = U;
  let Inst{22-20} = RdaHi{3-1};
  let Inst{19-18} = 0b10;
  let Inst{17-16} = 0b01;
  let Inst{15-13} = RdaLo{3-1};
  let Inst{12} = 0b0;
  let Inst{8-6} = 0b100;
  let Inst{5} = A;
  let Inst{3-1} = Qm{2-0};
  let Inst{0} = 0b0;
}

multiclass MVE_VADDLV_A<string suffix, bit U, list<dag> pattern=[]> {
  def acc    : MVE_VADDLV<"vaddlva", suffix,
                        (ins tGPREven:$RdaLo_src, tGPROdd:$RdaHi_src, MQPR:$Qm),
                        "$RdaLo = $RdaLo_src,$RdaHi = $RdaHi_src",
                        0b1, U, pattern>;
  def no_acc : MVE_VADDLV<"vaddlv", suffix,
                        (ins MQPR:$Qm), "",
                        0b0, U, pattern>;
}


defm MVE_VADDLVs32 : MVE_VADDLV_A<"s32", 0b0>;
defm MVE_VADDLVu32 : MVE_VADDLV_A<"u32", 0b1>;

class MVE_VMINMAXNMV<string iname, string suffix, bit sz,
                     bit bit_17, bit bit_7, list<dag> pattern=[]>
  : MVE_rDest<(outs rGPR:$RdaDest), (ins rGPR:$RdaSrc, MQPR:$Qm),
              NoItinerary, iname, suffix, "$RdaSrc, $Qm",
              "$RdaDest = $RdaSrc", pattern> {
  bits<3> Qm;
  bits<4> RdaDest;

  let Inst{28} = sz;
  let Inst{22-20} = 0b110;
  let Inst{19-18} = 0b11;
  let Inst{17} = bit_17;
  let Inst{16} = 0b0;
  let Inst{15-12} = RdaDest{3-0};
  let Inst{8} = 0b1;
  let Inst{7} = bit_7;
  let Inst{6-5} = 0b00;
  let Inst{3-1} = Qm{2-0};
  let Inst{0} = 0b0;

  let Predicates = [HasMVEFloat];
}

multiclass MVE_VMINMAXNMV_fty<string iname, bit bit_7, list<dag> pattern=[]> {
  def f32 : MVE_VMINMAXNMV<iname, "f32", 0b0, 0b1, bit_7, pattern>;
  def f16 : MVE_VMINMAXNMV<iname, "f16", 0b1, 0b1, bit_7, pattern>;
}

defm MVE_VMINNMV : MVE_VMINMAXNMV_fty<"vminnmv", 0b1>;
defm MVE_VMAXNMV : MVE_VMINMAXNMV_fty<"vmaxnmv", 0b0>;

multiclass MVE_VMINMAXNMAV_fty<string iname, bit bit_7, list<dag> pattern=[]> {
  def f32 : MVE_VMINMAXNMV<iname, "f32", 0b0, 0b0, bit_7, pattern>;
  def f16 : MVE_VMINMAXNMV<iname, "f16", 0b1, 0b0, bit_7, pattern>;
}

defm MVE_VMINNMAV : MVE_VMINMAXNMAV_fty<"vminnmav", 0b1>;
defm MVE_VMAXNMAV : MVE_VMINMAXNMAV_fty<"vmaxnmav", 0b0>;

class MVE_VMINMAXV<string iname, string suffix, bit U, bits<2> size,
                 bit bit_17, bit bit_7, list<dag> pattern=[]>
  : MVE_rDest<(outs rGPR:$RdaDest), (ins rGPR:$RdaSrc, MQPR:$Qm), NoItinerary,
              iname, suffix, "$RdaSrc, $Qm", "$RdaDest = $RdaSrc", pattern> {
  bits<3> Qm;
  bits<4> RdaDest;

  let Inst{28} = U;
  let Inst{22-20} = 0b110;
  let Inst{19-18} = size{1-0};
  let Inst{17} = bit_17;
  let Inst{16} = 0b0;
  let Inst{15-12} = RdaDest{3-0};
  let Inst{8} = 0b1;
  let Inst{7} = bit_7;
  let Inst{6-5} = 0b00;
  let Inst{3-1} = Qm{2-0};
  let Inst{0} = 0b0;
}

multiclass MVE_VMINMAXV_p<string iname, bit bit_17, bit bit_7,
                          MVEVectorVTInfo VTI, Intrinsic intr> {
  def "": MVE_VMINMAXV<iname, VTI.Suffix, VTI.Unsigned, VTI.Size,
                       bit_17, bit_7>;

  let Predicates = [HasMVEInt] in
  def _pat : Pat<(i32 (intr (i32 rGPR:$prev), (VTI.Vec MQPR:$vec))),
                 (i32 (!cast<Instruction>(NAME)
                           (i32 rGPR:$prev), (VTI.Vec MQPR:$vec)))>;
}

multiclass MVE_VMINMAXV_ty<string iname, bit bit_7,
                           Intrinsic intr_s, Intrinsic intr_u> {
  defm s8 : MVE_VMINMAXV_p<iname, 1, bit_7, MVE_v16s8, intr_s>;
  defm s16: MVE_VMINMAXV_p<iname, 1, bit_7, MVE_v8s16, intr_s>;
  defm s32: MVE_VMINMAXV_p<iname, 1, bit_7, MVE_v4s32, intr_s>;
  defm u8 : MVE_VMINMAXV_p<iname, 1, bit_7, MVE_v16u8, intr_u>;
  defm u16: MVE_VMINMAXV_p<iname, 1, bit_7, MVE_v8u16, intr_u>;
  defm u32: MVE_VMINMAXV_p<iname, 1, bit_7, MVE_v4u32, intr_u>;
}

defm MVE_VMINV : MVE_VMINMAXV_ty<
  "vminv", 0b1, int_arm_mve_minv_s, int_arm_mve_minv_u>;
defm MVE_VMAXV : MVE_VMINMAXV_ty<
  "vmaxv", 0b0, int_arm_mve_maxv_s, int_arm_mve_maxv_u>;

let Predicates = [HasMVEInt] in {
  def : Pat<(i32 (vecreduce_smax (v16i8 MQPR:$src))),
            (i32 (MVE_VMAXVs8 (t2MVNi (i32 127)), $src))>;
  def : Pat<(i32 (vecreduce_smax (v8i16 MQPR:$src))),
            (i32 (MVE_VMAXVs16 (t2MOVi32imm (i32 -32768)), $src))>;
  def : Pat<(i32 (vecreduce_smax (v4i32 MQPR:$src))),
            (i32 (MVE_VMAXVs32 (t2MOVi (i32 -2147483648)), $src))>;
  def : Pat<(i32 (vecreduce_umax (v16i8 MQPR:$src))),
            (i32 (MVE_VMAXVu8 (t2MOVi (i32 0)), $src))>;
  def : Pat<(i32 (vecreduce_umax (v8i16 MQPR:$src))),
            (i32 (MVE_VMAXVu16 (t2MOVi (i32 0)), $src))>;
  def : Pat<(i32 (vecreduce_umax (v4i32 MQPR:$src))),
            (i32 (MVE_VMAXVu32 (t2MOVi (i32 0)), $src))>;

  def : Pat<(i32 (vecreduce_smin (v16i8 MQPR:$src))),
            (i32 (MVE_VMINVs8 (t2MOVi (i32 127)), $src))>;
  def : Pat<(i32 (vecreduce_smin (v8i16 MQPR:$src))),
            (i32 (MVE_VMINVs16 (t2MOVi16 (i32 32767)), $src))>;
  def : Pat<(i32 (vecreduce_smin (v4i32 MQPR:$src))),
            (i32 (MVE_VMINVs32 (t2MVNi (i32 -2147483648)), $src))>;
  def : Pat<(i32 (vecreduce_umin (v16i8 MQPR:$src))),
            (i32 (MVE_VMINVu8 (t2MOVi (i32 255)), $src))>;
  def : Pat<(i32 (vecreduce_umin (v8i16 MQPR:$src))),
            (i32 (MVE_VMINVu16 (t2MOVi16 (i32 65535)), $src))>;
  def : Pat<(i32 (vecreduce_umin (v4i32 MQPR:$src))),
            (i32 (MVE_VMINVu32 (t2MOVi (i32 4294967295)), $src))>;

}

multiclass MVE_VMINMAXAV_ty<string iname, bit bit_7, list<dag> pattern=[]> {
  def s8  : MVE_VMINMAXV<iname, "s8",  0b0, 0b00, 0b0, bit_7>;
  def s16 : MVE_VMINMAXV<iname, "s16", 0b0, 0b01, 0b0, bit_7>;
  def s32 : MVE_VMINMAXV<iname, "s32", 0b0, 0b10, 0b0, bit_7>;
}

defm MVE_VMINAV : MVE_VMINMAXAV_ty<"vminav", 0b1>;
defm MVE_VMAXAV : MVE_VMINMAXAV_ty<"vmaxav", 0b0>;

class MVE_VMLAMLSDAV<string iname, string suffix, dag iops, string cstr,
                   bit sz, bit bit_28, bit A, bit X, bit bit_8, bit bit_0,
                   list<dag> pattern=[]>
  : MVE_rDest<(outs tGPREven:$RdaDest), iops, NoItinerary, iname, suffix,
              "$RdaDest, $Qn, $Qm", cstr, pattern> {
  bits<4> RdaDest;
  bits<3> Qm;
  bits<3> Qn;

  let Inst{28} = bit_28;
  let Inst{22-20} = 0b111;
  let Inst{19-17} = Qn{2-0};
  let Inst{16} = sz;
  let Inst{15-13} = RdaDest{3-1};
  let Inst{12} = X;
  let Inst{8} = bit_8;
  let Inst{7-6} = 0b00;
  let Inst{5} = A;
  let Inst{3-1} = Qm{2-0};
  let Inst{0} = bit_0;
}

multiclass MVE_VMLAMLSDAV_A<string iname, string x, string suffix,
                            bit sz, bit bit_28, bit X, bit bit_8, bit bit_0,
                            list<dag> pattern=[]> {
  def ""#x#suffix : MVE_VMLAMLSDAV<iname # x, suffix,
                                   (ins MQPR:$Qn, MQPR:$Qm), "",
                                   sz, bit_28, 0b0, X, bit_8, bit_0, pattern>;
  def "a"#x#suffix : MVE_VMLAMLSDAV<iname # "a" # x, suffix,
                                    (ins tGPREven:$RdaSrc, MQPR:$Qn, MQPR:$Qm),
                                    "$RdaDest = $RdaSrc",
                                    sz, bit_28, 0b1, X, bit_8, bit_0, pattern>;
}

multiclass MVE_VMLAMLSDAV_AX<string iname, string suffix, bit sz, bit bit_28,
                             bit bit_8, bit bit_0, list<dag> pattern=[]> {
  defm "" : MVE_VMLAMLSDAV_A<iname, "", suffix, sz, bit_28,
                             0b0, bit_8, bit_0, pattern>;
  defm "" : MVE_VMLAMLSDAV_A<iname, "x", suffix, sz, bit_28,
                             0b1, bit_8, bit_0, pattern>;
}

multiclass MVE_VMLADAV_multi<string suffix, bit sz, bit bit_8,
                             list<dag> pattern=[]> {
  defm "" : MVE_VMLAMLSDAV_AX<"vmladav", "s"#suffix,
                              sz, 0b0, bit_8, 0b0, pattern>;
  defm "" : MVE_VMLAMLSDAV_A<"vmladav", "", "u"#suffix,
                             sz, 0b1, 0b0, bit_8, 0b0, pattern>;
}

multiclass MVE_VMLSDAV_multi<string suffix, bit sz, bit bit_28,
                             list<dag> pattern=[]> {
  defm "" : MVE_VMLAMLSDAV_AX<"vmlsdav", "s"#suffix,
                              sz, bit_28, 0b0, 0b1, pattern>;
}

defm MVE_VMLADAV : MVE_VMLADAV_multi< "8", 0b0, 0b1>;
defm MVE_VMLADAV : MVE_VMLADAV_multi<"16", 0b0, 0b0>;
defm MVE_VMLADAV : MVE_VMLADAV_multi<"32", 0b1, 0b0>;

defm MVE_VMLSDAV : MVE_VMLSDAV_multi< "8", 0b0, 0b1>;
defm MVE_VMLSDAV : MVE_VMLSDAV_multi<"16", 0b0, 0b0>;
defm MVE_VMLSDAV : MVE_VMLSDAV_multi<"32", 0b1, 0b0>;

// vmlav aliases vmladav
foreach acc = ["", "a"] in {
  foreach suffix = ["s8", "s16", "s32", "u8", "u16", "u32"] in {
    def : MVEInstAlias<"vmlav"#acc#"${vp}."#suffix#"\t$RdaDest, $Qn, $Qm",
                       (!cast<Instruction>("MVE_VMLADAV"#acc#suffix)
                        tGPREven:$RdaDest, MQPR:$Qn, MQPR:$Qm, vpred_n:$vp)>;
  }
}

// Base class for VMLALDAV and VMLSLDAV, VRMLALDAVH, VRMLSLDAVH
class MVE_VMLALDAVBase<string iname, string suffix, dag iops, string cstr,
                       bit sz, bit bit_28, bit A, bit X, bit bit_8, bit bit_0,
                       list<dag> pattern=[]>
  : MVE_rDest<(outs tGPREven:$RdaLoDest, tGPROdd:$RdaHiDest), iops, NoItinerary,
              iname, suffix, "$RdaLoDest, $RdaHiDest, $Qn, $Qm", cstr, pattern> {
  bits<4> RdaLoDest;
  bits<4> RdaHiDest;
  bits<3> Qm;
  bits<3> Qn;

  let Inst{28} = bit_28;
  let Inst{22-20} = RdaHiDest{3-1};
  let Inst{19-17} = Qn{2-0};
  let Inst{16} = sz;
  let Inst{15-13} = RdaLoDest{3-1};
  let Inst{12} = X;
  let Inst{8} = bit_8;
  let Inst{7-6} = 0b00;
  let Inst{5} = A;
  let Inst{3-1} = Qm{2-0};
  let Inst{0} = bit_0;
}

multiclass MVE_VMLALDAVBase_A<string iname, string x, string suffix,
                               bit sz, bit bit_28, bit X, bit bit_8, bit bit_0,
                               list<dag> pattern=[]> {
  def ""#x#suffix : MVE_VMLALDAVBase<
     iname # x, suffix, (ins MQPR:$Qn, MQPR:$Qm), "",
     sz, bit_28, 0b0, X, bit_8, bit_0, pattern>;
  def "a"#x#suffix : MVE_VMLALDAVBase<
     iname # "a" # x, suffix,
     (ins tGPREven:$RdaLoSrc, tGPROdd:$RdaHiSrc, MQPR:$Qn, MQPR:$Qm),
     "$RdaLoDest = $RdaLoSrc,$RdaHiDest = $RdaHiSrc",
     sz, bit_28, 0b1, X, bit_8, bit_0, pattern>;
}


multiclass MVE_VMLALDAVBase_AX<string iname, string suffix, bit sz, bit bit_28,
                               bit bit_8, bit bit_0, list<dag> pattern=[]> {
  defm "" : MVE_VMLALDAVBase_A<iname, "", suffix, sz,
                               bit_28, 0b0, bit_8, bit_0, pattern>;
  defm "" : MVE_VMLALDAVBase_A<iname, "x", suffix, sz,
                               bit_28, 0b1, bit_8, bit_0, pattern>;
}

multiclass MVE_VRMLALDAVH_multi<string suffix, list<dag> pattern=[]> {
  defm "" : MVE_VMLALDAVBase_AX<"vrmlaldavh", "s"#suffix,
                                0b0, 0b0, 0b1, 0b0, pattern>;
  defm "" : MVE_VMLALDAVBase_A<"vrmlaldavh", "", "u"#suffix,
                               0b0, 0b1, 0b0, 0b1, 0b0, pattern>;
}

defm MVE_VRMLALDAVH : MVE_VRMLALDAVH_multi<"32">;

// vrmlalvh aliases for vrmlaldavh
def : MVEInstAlias<"vrmlalvh${vp}.s32\t$RdaLo, $RdaHi, $Qn, $Qm",
                  (MVE_VRMLALDAVHs32
                   tGPREven:$RdaLo, tGPROdd:$RdaHi,
                   MQPR:$Qn, MQPR:$Qm, vpred_n:$vp)>;
def : MVEInstAlias<"vrmlalvha${vp}.s32\t$RdaLo, $RdaHi, $Qn, $Qm",
                  (MVE_VRMLALDAVHas32
                   tGPREven:$RdaLo, tGPROdd:$RdaHi,
                   MQPR:$Qn, MQPR:$Qm, vpred_n:$vp)>;
def : MVEInstAlias<"vrmlalvh${vp}.u32\t$RdaLo, $RdaHi, $Qn, $Qm",
                  (MVE_VRMLALDAVHu32
                   tGPREven:$RdaLo, tGPROdd:$RdaHi,
                   MQPR:$Qn, MQPR:$Qm, vpred_n:$vp)>;
def : MVEInstAlias<"vrmlalvha${vp}.u32\t$RdaLo, $RdaHi, $Qn, $Qm",
                  (MVE_VRMLALDAVHau32
                   tGPREven:$RdaLo, tGPROdd:$RdaHi,
                   MQPR:$Qn, MQPR:$Qm, vpred_n:$vp)>;

multiclass MVE_VMLALDAV_multi<string suffix, bit sz, list<dag> pattern=[]> {
  defm "" : MVE_VMLALDAVBase_AX<"vmlaldav", "s"#suffix, sz, 0b0, 0b0, 0b0, pattern>;
  defm "" : MVE_VMLALDAVBase_A<"vmlaldav", "", "u"#suffix,
                               sz, 0b1, 0b0, 0b0, 0b0, pattern>;
}

defm MVE_VMLALDAV : MVE_VMLALDAV_multi<"16", 0b0>;
defm MVE_VMLALDAV : MVE_VMLALDAV_multi<"32", 0b1>;

// vmlalv aliases vmlaldav
foreach acc = ["", "a"] in {
  foreach suffix = ["s16", "s32", "u16", "u32"] in {
    def : MVEInstAlias<"vmlalv" # acc # "${vp}." # suffix #
                          "\t$RdaLoDest, $RdaHiDest, $Qn, $Qm",
                       (!cast<Instruction>("MVE_VMLALDAV"#acc#suffix)
                       tGPREven:$RdaLoDest, tGPROdd:$RdaHiDest,
                       MQPR:$Qn, MQPR:$Qm, vpred_n:$vp)>;
  }
}

multiclass MVE_VMLSLDAV_multi<string iname, string suffix, bit sz,
                              bit bit_28, list<dag> pattern=[]> {
  defm "" : MVE_VMLALDAVBase_AX<iname, suffix, sz, bit_28, 0b0, 0b1, pattern>;
}

defm MVE_VMLSLDAV   : MVE_VMLSLDAV_multi<"vmlsldav", "s16", 0b0, 0b0>;
defm MVE_VMLSLDAV   : MVE_VMLSLDAV_multi<"vmlsldav", "s32", 0b1, 0b0>;
defm MVE_VRMLSLDAVH : MVE_VMLSLDAV_multi<"vrmlsldavh", "s32", 0b0, 0b1>;

// end of mve_rDest instructions

// start of mve_comp instructions

class MVE_comp<InstrItinClass itin, string iname, string suffix,
               string cstr, list<dag> pattern=[]>
  : MVE_p<(outs MQPR:$Qd), (ins MQPR:$Qn, MQPR:$Qm), itin, iname, suffix,
           "$Qd, $Qn, $Qm", vpred_r, cstr, pattern> {
  bits<4> Qd;
  bits<4> Qn;
  bits<4> Qm;

  let Inst{22} = Qd{3};
  let Inst{19-17} = Qn{2-0};
  let Inst{16} = 0b0;
  let Inst{15-13} = Qd{2-0};
  let Inst{12} = 0b0;
  let Inst{10-9} = 0b11;
  let Inst{7} = Qn{3};
  let Inst{5} = Qm{3};
  let Inst{3-1} = Qm{2-0};
  let Inst{0} = 0b0;
}

class MVE_VMINMAXNM<string iname, string suffix, bit sz, bit bit_21,
                list<dag> pattern=[]>
  : MVE_comp<NoItinerary, iname, suffix, "", pattern> {

  let Inst{28} = 0b1;
  let Inst{25-24} = 0b11;
  let Inst{23} = 0b0;
  let Inst{21} = bit_21;
  let Inst{20} = sz;
  let Inst{11} = 0b1;
  let Inst{8} = 0b1;
  let Inst{6} = 0b1;
  let Inst{4} = 0b1;

  let Predicates = [HasMVEFloat];
}

def MVE_VMAXNMf32 : MVE_VMINMAXNM<"vmaxnm", "f32", 0b0, 0b0>;
def MVE_VMAXNMf16 : MVE_VMINMAXNM<"vmaxnm", "f16", 0b1, 0b0>;

let Predicates = [HasMVEFloat] in {
  def : Pat<(v4f32 (fmaxnum (v4f32 MQPR:$val1), (v4f32 MQPR:$val2))),
            (v4f32 (MVE_VMAXNMf32 (v4f32 MQPR:$val1), (v4f32 MQPR:$val2)))>;
  def : Pat<(v8f16 (fmaxnum (v8f16 MQPR:$val1), (v8f16 MQPR:$val2))),
            (v8f16 (MVE_VMAXNMf16 (v8f16 MQPR:$val1), (v8f16 MQPR:$val2)))>;
}

def MVE_VMINNMf32 : MVE_VMINMAXNM<"vminnm", "f32", 0b0, 0b1>;
def MVE_VMINNMf16 : MVE_VMINMAXNM<"vminnm", "f16", 0b1, 0b1>;

let Predicates = [HasMVEFloat] in {
  def : Pat<(v4f32 (fminnum (v4f32 MQPR:$val1), (v4f32 MQPR:$val2))),
            (v4f32 (MVE_VMINNMf32 (v4f32 MQPR:$val1), (v4f32 MQPR:$val2)))>;
  def : Pat<(v8f16 (fminnum (v8f16 MQPR:$val1), (v8f16 MQPR:$val2))),
            (v8f16 (MVE_VMINNMf16 (v8f16 MQPR:$val1), (v8f16 MQPR:$val2)))>;
}


class MVE_VMINMAX<string iname, string suffix, bit U, bits<2> size,
              bit bit_4, list<dag> pattern=[]>
  : MVE_comp<NoItinerary, iname, suffix, "", pattern> {

  let Inst{28} = U;
  let Inst{25-24} = 0b11;
  let Inst{23} = 0b0;
  let Inst{21-20} = size{1-0};
  let Inst{11} = 0b0;
  let Inst{8} = 0b0;
  let Inst{6} = 0b1;
  let Inst{4} = bit_4;
}

multiclass MVE_VMINMAX_all_sizes<string iname, bit bit_4> {
  def s8  : MVE_VMINMAX<iname, "s8",  0b0, 0b00, bit_4>;
  def s16 : MVE_VMINMAX<iname, "s16", 0b0, 0b01, bit_4>;
  def s32 : MVE_VMINMAX<iname, "s32", 0b0, 0b10, bit_4>;
  def u8  : MVE_VMINMAX<iname, "u8",  0b1, 0b00, bit_4>;
  def u16 : MVE_VMINMAX<iname, "u16", 0b1, 0b01, bit_4>;
  def u32 : MVE_VMINMAX<iname, "u32", 0b1, 0b10, bit_4>;
}

defm MVE_VMAX : MVE_VMINMAX_all_sizes<"vmax", 0b0>;
defm MVE_VMIN : MVE_VMINMAX_all_sizes<"vmin", 0b1>;

let Predicates = [HasMVEInt] in {
  def : Pat<(v16i8 (smin (v16i8 MQPR:$val1), (v16i8 MQPR:$val2))),
            (v16i8 (MVE_VMINs8 (v16i8 MQPR:$val1), (v16i8 MQPR:$val2)))>;
  def : Pat<(v8i16 (smin (v8i16 MQPR:$val1), (v8i16 MQPR:$val2))),
            (v8i16 (MVE_VMINs16 (v8i16 MQPR:$val1), (v8i16 MQPR:$val2)))>;
  def : Pat<(v4i32 (smin (v4i32 MQPR:$val1), (v4i32 MQPR:$val2))),
            (v4i32 (MVE_VMINs32 (v4i32 MQPR:$val1), (v4i32 MQPR:$val2)))>;

  def : Pat<(v16i8 (smax (v16i8 MQPR:$val1), (v16i8 MQPR:$val2))),
            (v16i8 (MVE_VMAXs8 (v16i8 MQPR:$val1), (v16i8 MQPR:$val2)))>;
  def : Pat<(v8i16 (smax (v8i16 MQPR:$val1), (v8i16 MQPR:$val2))),
            (v8i16 (MVE_VMAXs16 (v8i16 MQPR:$val1), (v8i16 MQPR:$val2)))>;
  def : Pat<(v4i32 (smax (v4i32 MQPR:$val1), (v4i32 MQPR:$val2))),
            (v4i32 (MVE_VMAXs32 (v4i32 MQPR:$val1), (v4i32 MQPR:$val2)))>;

  def : Pat<(v16i8 (umin (v16i8 MQPR:$val1), (v16i8 MQPR:$val2))),
            (v16i8 (MVE_VMINu8 (v16i8 MQPR:$val1), (v16i8 MQPR:$val2)))>;
  def : Pat<(v8i16 (umin (v8i16 MQPR:$val1), (v8i16 MQPR:$val2))),
            (v8i16 (MVE_VMINu16 (v8i16 MQPR:$val1), (v8i16 MQPR:$val2)))>;
  def : Pat<(v4i32 (umin (v4i32 MQPR:$val1), (v4i32 MQPR:$val2))),
            (v4i32 (MVE_VMINu32 (v4i32 MQPR:$val1), (v4i32 MQPR:$val2)))>;

  def : Pat<(v16i8 (umax (v16i8 MQPR:$val1), (v16i8 MQPR:$val2))),
            (v16i8 (MVE_VMAXu8 (v16i8 MQPR:$val1), (v16i8 MQPR:$val2)))>;
  def : Pat<(v8i16 (umax (v8i16 MQPR:$val1), (v8i16 MQPR:$val2))),
            (v8i16 (MVE_VMAXu16 (v8i16 MQPR:$val1), (v8i16 MQPR:$val2)))>;
  def : Pat<(v4i32 (umax (v4i32 MQPR:$val1), (v4i32 MQPR:$val2))),
            (v4i32 (MVE_VMAXu32 (v4i32 MQPR:$val1), (v4i32 MQPR:$val2)))>;
}

// end of mve_comp instructions

// start of mve_bit instructions

class MVE_bit_arith<dag oops, dag iops, string iname, string suffix,
                    string ops, string cstr, list<dag> pattern=[]>
  : MVE_p<oops, iops, NoItinerary, iname, suffix, ops, vpred_r, cstr, pattern> {
  bits<4> Qd;
  bits<4> Qm;

  let Inst{22} = Qd{3};
  let Inst{15-13} = Qd{2-0};
  let Inst{5} = Qm{3};
  let Inst{3-1} = Qm{2-0};
}

def MVE_VBIC : MVE_bit_arith<(outs MQPR:$Qd), (ins MQPR:$Qn, MQPR:$Qm),
                             "vbic", "", "$Qd, $Qn, $Qm", ""> {
  bits<4> Qn;

  let Inst{28} = 0b0;
  let Inst{25-23} = 0b110;
  let Inst{21-20} = 0b01;
  let Inst{19-17} = Qn{2-0};
  let Inst{16} = 0b0;
  let Inst{12-8} = 0b00001;
  let Inst{7} = Qn{3};
  let Inst{6} = 0b1;
  let Inst{4} = 0b1;
  let Inst{0} = 0b0;
  let validForTailPredication = 1;
}

class MVE_VREV<string iname, string suffix, bits<2> size, bits<2> bit_8_7, string cstr="">
  : MVE_bit_arith<(outs MQPR:$Qd), (ins MQPR:$Qm), iname,
                  suffix, "$Qd, $Qm", cstr> {

  let Inst{28} = 0b1;
  let Inst{25-23} = 0b111;
  let Inst{21-20} = 0b11;
  let Inst{19-18} = size;
  let Inst{17-16} = 0b00;
  let Inst{12-9} = 0b0000;
  let Inst{8-7} = bit_8_7;
  let Inst{6} = 0b1;
  let Inst{4} = 0b0;
  let Inst{0} = 0b0;
}

def MVE_VREV64_8  : MVE_VREV<"vrev64", "8", 0b00, 0b00, "@earlyclobber $Qd">;
def MVE_VREV64_16 : MVE_VREV<"vrev64", "16", 0b01, 0b00, "@earlyclobber $Qd">;
def MVE_VREV64_32 : MVE_VREV<"vrev64", "32", 0b10, 0b00, "@earlyclobber $Qd">;

def MVE_VREV32_8  : MVE_VREV<"vrev32", "8", 0b00, 0b01>;
def MVE_VREV32_16 : MVE_VREV<"vrev32", "16", 0b01, 0b01>;

def MVE_VREV16_8  : MVE_VREV<"vrev16", "8", 0b00, 0b10>;

let Predicates = [HasMVEInt] in {
  def : Pat<(v8i16 (bswap (v8i16 MQPR:$src))),
            (v8i16 (MVE_VREV16_8 (v8i16 MQPR:$src)))>;
  def : Pat<(v4i32 (bswap (v4i32 MQPR:$src))),
            (v4i32 (MVE_VREV32_8 (v4i32 MQPR:$src)))>;
}

let Predicates = [HasMVEInt] in {
  def : Pat<(v4i32 (ARMvrev64 (v4i32 MQPR:$src))),
            (v4i32 (MVE_VREV64_32 (v4i32 MQPR:$src)))>;
  def : Pat<(v8i16 (ARMvrev64 (v8i16 MQPR:$src))),
            (v8i16 (MVE_VREV64_16 (v8i16 MQPR:$src)))>;
  def : Pat<(v16i8 (ARMvrev64 (v16i8 MQPR:$src))),
            (v16i8 (MVE_VREV64_8  (v16i8 MQPR:$src)))>;

  def : Pat<(v8i16 (ARMvrev32 (v8i16 MQPR:$src))),
            (v8i16 (MVE_VREV32_16 (v8i16 MQPR:$src)))>;
  def : Pat<(v16i8 (ARMvrev32 (v16i8 MQPR:$src))),
            (v16i8 (MVE_VREV32_8  (v16i8 MQPR:$src)))>;

  def : Pat<(v16i8 (ARMvrev16 (v16i8 MQPR:$src))),
            (v16i8 (MVE_VREV16_8  (v16i8 MQPR:$src)))>;

  def : Pat<(v4f32 (ARMvrev64 (v4f32 MQPR:$src))),
            (v4f32 (MVE_VREV64_32 (v4f32 MQPR:$src)))>;
  def : Pat<(v8f16 (ARMvrev64 (v8f16 MQPR:$src))),
            (v8f16 (MVE_VREV64_16 (v8f16 MQPR:$src)))>;
  def : Pat<(v8f16 (ARMvrev32 (v8f16 MQPR:$src))),
            (v8f16 (MVE_VREV32_16 (v8f16 MQPR:$src)))>;
}

def MVE_VMVN : MVE_bit_arith<(outs MQPR:$Qd), (ins MQPR:$Qm),
                             "vmvn", "", "$Qd, $Qm", ""> {
  let Inst{28} = 0b1;
  let Inst{25-23} = 0b111;
  let Inst{21-16} = 0b110000;
  let Inst{12-6} = 0b0010111;
  let Inst{4} = 0b0;
  let Inst{0} = 0b0;
  let validForTailPredication = 1;
}

let Predicates = [HasMVEInt] in {
  def : Pat<(v16i8 (vnotq  (v16i8 MQPR:$val1))),
            (v16i8 (MVE_VMVN (v16i8 MQPR:$val1)))>;
  def : Pat<(v8i16 (vnotq  (v8i16 MQPR:$val1))),
            (v8i16 (MVE_VMVN (v8i16 MQPR:$val1)))>;
  def : Pat<(v4i32 (vnotq  (v4i32 MQPR:$val1))),
            (v4i32 (MVE_VMVN (v4i32 MQPR:$val1)))>;
  def : Pat<(v2i64 (vnotq  (v2i64 MQPR:$val1))),
            (v2i64 (MVE_VMVN (v2i64 MQPR:$val1)))>;
}

class MVE_bit_ops<string iname, bits<2> bit_21_20, bit bit_28>
  : MVE_bit_arith<(outs MQPR:$Qd), (ins MQPR:$Qn, MQPR:$Qm),
                  iname, "", "$Qd, $Qn, $Qm", ""> {
  bits<4> Qn;

  let Inst{28} = bit_28;
  let Inst{25-23} = 0b110;
  let Inst{21-20} = bit_21_20;
  let Inst{19-17} = Qn{2-0};
  let Inst{16} = 0b0;
  let Inst{12-8} = 0b00001;
  let Inst{7} = Qn{3};
  let Inst{6} = 0b1;
  let Inst{4} = 0b1;
  let Inst{0} = 0b0;
  let validForTailPredication = 1;
}

def MVE_VEOR : MVE_bit_ops<"veor", 0b00, 0b1>;
def MVE_VORN : MVE_bit_ops<"vorn", 0b11, 0b0>;
def MVE_VORR : MVE_bit_ops<"vorr", 0b10, 0b0>;
def MVE_VAND : MVE_bit_ops<"vand", 0b00, 0b0>;

// add ignored suffixes as aliases

foreach s=["s8", "s16", "s32", "u8", "u16", "u32", "i8", "i16", "i32", "f16", "f32"] in {
  def : MVEInstAlias<"vbic${vp}." # s # "\t$QdSrc, $QnSrc, $QmSrc",
        (MVE_VBIC MQPR:$QdSrc, MQPR:$QnSrc, MQPR:$QmSrc, vpred_r:$vp)>;
  def : MVEInstAlias<"veor${vp}." # s # "\t$QdSrc, $QnSrc, $QmSrc",
        (MVE_VEOR MQPR:$QdSrc, MQPR:$QnSrc, MQPR:$QmSrc, vpred_r:$vp)>;
  def : MVEInstAlias<"vorn${vp}." # s # "\t$QdSrc, $QnSrc, $QmSrc",
        (MVE_VORN MQPR:$QdSrc, MQPR:$QnSrc, MQPR:$QmSrc, vpred_r:$vp)>;
  def : MVEInstAlias<"vorr${vp}." # s # "\t$QdSrc, $QnSrc, $QmSrc",
        (MVE_VORR MQPR:$QdSrc, MQPR:$QnSrc, MQPR:$QmSrc, vpred_r:$vp)>;
  def : MVEInstAlias<"vand${vp}." # s # "\t$QdSrc, $QnSrc, $QmSrc",
        (MVE_VAND MQPR:$QdSrc, MQPR:$QnSrc, MQPR:$QmSrc, vpred_r:$vp)>;
}

let Predicates = [HasMVEInt] in {
  def : Pat<(v16i8 (and (v16i8 MQPR:$val1), (v16i8 MQPR:$val2))),
            (v16i8 (MVE_VAND (v16i8 MQPR:$val1), (v16i8 MQPR:$val2)))>;
  def : Pat<(v8i16 (and (v8i16 MQPR:$val1), (v8i16 MQPR:$val2))),
            (v8i16 (MVE_VAND (v8i16 MQPR:$val1), (v8i16 MQPR:$val2)))>;
  def : Pat<(v4i32 (and (v4i32 MQPR:$val1), (v4i32 MQPR:$val2))),
            (v4i32 (MVE_VAND (v4i32 MQPR:$val1), (v4i32 MQPR:$val2)))>;
  def : Pat<(v2i64 (and (v2i64 MQPR:$val1), (v2i64 MQPR:$val2))),
            (v2i64 (MVE_VAND (v2i64 MQPR:$val1), (v2i64 MQPR:$val2)))>;

  def : Pat<(v16i8 (or (v16i8 MQPR:$val1), (v16i8 MQPR:$val2))),
            (v16i8 (MVE_VORR (v16i8 MQPR:$val1), (v16i8 MQPR:$val2)))>;
  def : Pat<(v8i16 (or (v8i16 MQPR:$val1), (v8i16 MQPR:$val2))),
            (v8i16 (MVE_VORR (v8i16 MQPR:$val1), (v8i16 MQPR:$val2)))>;
  def : Pat<(v4i32 (or (v4i32 MQPR:$val1), (v4i32 MQPR:$val2))),
            (v4i32 (MVE_VORR (v4i32 MQPR:$val1), (v4i32 MQPR:$val2)))>;
  def : Pat<(v2i64 (or (v2i64 MQPR:$val1), (v2i64 MQPR:$val2))),
            (v2i64 (MVE_VORR (v2i64 MQPR:$val1), (v2i64 MQPR:$val2)))>;

  def : Pat<(v16i8 (xor (v16i8 MQPR:$val1), (v16i8 MQPR:$val2))),
            (v16i8 (MVE_VEOR (v16i8 MQPR:$val1), (v16i8 MQPR:$val2)))>;
  def : Pat<(v8i16 (xor (v8i16 MQPR:$val1), (v8i16 MQPR:$val2))),
            (v8i16 (MVE_VEOR (v8i16 MQPR:$val1), (v8i16 MQPR:$val2)))>;
  def : Pat<(v4i32 (xor (v4i32 MQPR:$val1), (v4i32 MQPR:$val2))),
            (v4i32 (MVE_VEOR (v4i32 MQPR:$val1), (v4i32 MQPR:$val2)))>;
  def : Pat<(v2i64 (xor (v2i64 MQPR:$val1), (v2i64 MQPR:$val2))),
            (v2i64 (MVE_VEOR (v2i64 MQPR:$val1), (v2i64 MQPR:$val2)))>;

  def : Pat<(v16i8 (and (v16i8 MQPR:$val1), (vnotq MQPR:$val2))),
            (v16i8 (MVE_VBIC (v16i8 MQPR:$val1), (v16i8 MQPR:$val2)))>;
  def : Pat<(v8i16 (and (v8i16 MQPR:$val1), (vnotq MQPR:$val2))),
            (v8i16 (MVE_VBIC (v8i16 MQPR:$val1), (v8i16 MQPR:$val2)))>;
  def : Pat<(v4i32 (and (v4i32 MQPR:$val1), (vnotq MQPR:$val2))),
            (v4i32 (MVE_VBIC (v4i32 MQPR:$val1), (v4i32 MQPR:$val2)))>;
  def : Pat<(v2i64 (and (v2i64 MQPR:$val1), (vnotq MQPR:$val2))),
            (v2i64 (MVE_VBIC (v2i64 MQPR:$val1), (v2i64 MQPR:$val2)))>;

  def : Pat<(v16i8 (or (v16i8 MQPR:$val1), (vnotq MQPR:$val2))),
            (v16i8 (MVE_VORN (v16i8 MQPR:$val1), (v16i8 MQPR:$val2)))>;
  def : Pat<(v8i16 (or (v8i16 MQPR:$val1), (vnotq MQPR:$val2))),
            (v8i16 (MVE_VORN (v8i16 MQPR:$val1), (v8i16 MQPR:$val2)))>;
  def : Pat<(v4i32 (or (v4i32 MQPR:$val1), (vnotq MQPR:$val2))),
            (v4i32 (MVE_VORN (v4i32 MQPR:$val1), (v4i32 MQPR:$val2)))>;
  def : Pat<(v2i64 (or (v2i64 MQPR:$val1), (vnotq MQPR:$val2))),
            (v2i64 (MVE_VORN (v2i64 MQPR:$val1), (v2i64 MQPR:$val2)))>;
}

class MVE_bit_cmode<string iname, string suffix, bits<4> cmode, dag inOps>
  : MVE_p<(outs MQPR:$Qd), inOps, NoItinerary,
          iname, suffix, "$Qd, $imm", vpred_n, "$Qd = $Qd_src"> {
  bits<8> imm;
  bits<4> Qd;

  let Inst{28} = imm{7};
  let Inst{27-23} = 0b11111;
  let Inst{22} = Qd{3};
  let Inst{21-19} = 0b000;
  let Inst{18-16} = imm{6-4};
  let Inst{15-13} = Qd{2-0};
  let Inst{12} = 0b0;
  let Inst{11-8} = cmode;
  let Inst{7-6} = 0b01;
  let Inst{4} = 0b1;
  let Inst{3-0} = imm{3-0};
}

class MVE_VORR<string suffix, bits<4> cmode, ExpandImm imm_type>
  : MVE_bit_cmode<"vorr", suffix, cmode, (ins MQPR:$Qd_src, imm_type:$imm)> {
  let Inst{5} = 0b0;
  let validForTailPredication = 1;
}

def MVE_VORRIZ0v4i32  : MVE_VORR<"i32", 0b0001, expzero00>;
def MVE_VORRIZ0v8i16  : MVE_VORR<"i16", 0b1001, expzero00>;
def MVE_VORRIZ8v4i32  : MVE_VORR<"i32", 0b0011, expzero08>;
def MVE_VORRIZ8v8i16  : MVE_VORR<"i16", 0b1011, expzero08>;
def MVE_VORRIZ16v4i32 : MVE_VORR<"i32", 0b0101, expzero16>;
def MVE_VORRIZ24v4i32 : MVE_VORR<"i32", 0b0111, expzero24>;

def MVE_VORNIZ0v4i32 : MVEAsmPseudo<"vorn${vp}.i32\t$Qd, $imm",
    (ins MQPR:$Qd_src, expzero00inv32:$imm, vpred_n:$vp), (outs MQPR:$Qd)>;
def MVE_VORNIZ0v8i16 : MVEAsmPseudo<"vorn${vp}.i16\t$Qd, $imm",
    (ins MQPR:$Qd_src, expzero00inv16:$imm, vpred_n:$vp), (outs MQPR:$Qd)>;
def MVE_VORNIZ8v4i32 : MVEAsmPseudo<"vorn${vp}.i32\t$Qd, $imm",
    (ins MQPR:$Qd_src, expzero08inv32:$imm, vpred_n:$vp), (outs MQPR:$Qd)>;
def MVE_VORNIZ8v8i16 : MVEAsmPseudo<"vorn${vp}.i16\t$Qd, $imm",
    (ins MQPR:$Qd_src, expzero08inv16:$imm, vpred_n:$vp), (outs MQPR:$Qd)>;
def MVE_VORNIZ16v4i32 : MVEAsmPseudo<"vorn${vp}.i32\t$Qd, $imm",
    (ins MQPR:$Qd_src, expzero16inv32:$imm, vpred_n:$vp), (outs MQPR:$Qd)>;
def MVE_VORNIZ24v4i32 : MVEAsmPseudo<"vorn${vp}.i32\t$Qd, $imm",
    (ins MQPR:$Qd_src, expzero24inv32:$imm, vpred_n:$vp), (outs MQPR:$Qd)>;

def MVE_VMOV : MVEInstAlias<"vmov${vp}\t$Qd, $Qm",
    (MVE_VORR MQPR:$Qd, MQPR:$Qm, MQPR:$Qm, vpred_r:$vp)>;

class MVE_VBIC<string suffix, bits<4> cmode, ExpandImm imm_type>
  : MVE_bit_cmode<"vbic", suffix, cmode, (ins MQPR:$Qd_src, imm_type:$imm)> {
  let Inst{5} = 0b1;
  let validForTailPredication = 1;
}

def MVE_VBICIZ0v4i32  : MVE_VBIC<"i32", 0b0001, expzero00>;
def MVE_VBICIZ0v8i16  : MVE_VBIC<"i16", 0b1001, expzero00>;
def MVE_VBICIZ8v4i32  : MVE_VBIC<"i32", 0b0011, expzero08>;
def MVE_VBICIZ8v8i16  : MVE_VBIC<"i16", 0b1011, expzero08>;
def MVE_VBICIZ16v4i32 : MVE_VBIC<"i32", 0b0101, expzero16>;
def MVE_VBICIZ24v4i32 : MVE_VBIC<"i32", 0b0111, expzero24>;

def MVE_VANDIZ0v4i32 : MVEAsmPseudo<"vand${vp}.i32\t$Qda, $imm",
    (ins MQPR:$Qda_src, expzero00inv32:$imm, vpred_n:$vp), (outs MQPR:$Qda)>;
def MVE_VANDIZ0v8i16 : MVEAsmPseudo<"vand${vp}.i16\t$Qda, $imm",
    (ins MQPR:$Qda_src, expzero00inv16:$imm, vpred_n:$vp), (outs MQPR:$Qda)>;
def MVE_VANDIZ8v4i32 : MVEAsmPseudo<"vand${vp}.i32\t$Qda, $imm",
    (ins MQPR:$Qda_src, expzero08inv32:$imm, vpred_n:$vp), (outs MQPR:$Qda)>;
def MVE_VANDIZ8v8i16 : MVEAsmPseudo<"vand${vp}.i16\t$Qda, $imm",
    (ins MQPR:$Qda_src, expzero08inv16:$imm, vpred_n:$vp), (outs MQPR:$Qda)>;
def MVE_VANDIZ16v4i32 : MVEAsmPseudo<"vand${vp}.i32\t$Qda, $imm",
    (ins MQPR:$Qda_src, expzero16inv32:$imm, vpred_n:$vp), (outs MQPR:$Qda)>;
def MVE_VANDIZ24v4i32 : MVEAsmPseudo<"vand${vp}.i32\t$Qda, $imm",
    (ins MQPR:$Qda_src, expzero24inv32:$imm, vpred_n:$vp), (outs MQPR:$Qda)>;

class MVE_VMOV_lane_direction {
  bit bit_20;
  dag oops;
  dag iops;
  string ops;
  string cstr;
}
def MVE_VMOV_from_lane : MVE_VMOV_lane_direction {
  let bit_20 = 0b1;
  let oops = (outs rGPR:$Rt);
  let iops = (ins MQPR:$Qd);
  let ops = "$Rt, $Qd$Idx";
  let cstr = "";
}
def MVE_VMOV_to_lane : MVE_VMOV_lane_direction {
  let bit_20 = 0b0;
  let oops = (outs MQPR:$Qd);
  let iops = (ins MQPR:$Qd_src, rGPR:$Rt);
  let ops = "$Qd$Idx, $Rt";
  let cstr = "$Qd = $Qd_src";
}

class MVE_VMOV_lane<string suffix, bit U, dag indexop,
                    MVE_VMOV_lane_direction dir>
  : MVE_VMOV_lane_base<dir.oops, !con(dir.iops, indexop), NoItinerary,
                       "vmov", suffix, dir.ops, dir.cstr, []> {
  bits<4> Qd;
  bits<4> Rt;

  let Inst{31-24} = 0b11101110;
  let Inst{23} = U;
  let Inst{20} = dir.bit_20;
  let Inst{19-17} = Qd{2-0};
  let Inst{15-12} = Rt{3-0};
  let Inst{11-8} = 0b1011;
  let Inst{7} = Qd{3};
  let Inst{4-0} = 0b10000;
}

class MVE_VMOV_lane_32<MVE_VMOV_lane_direction dir>
    : MVE_VMOV_lane<"32", 0b0, (ins MVEVectorIndex<4>:$Idx), dir> {
  bits<2> Idx;
  let Inst{22} = 0b0;
  let Inst{6-5} = 0b00;
  let Inst{16} = Idx{1};
  let Inst{21} = Idx{0};

  let Predicates = [HasFPRegsV8_1M];
}

class MVE_VMOV_lane_16<string suffix, bit U, MVE_VMOV_lane_direction dir>
  : MVE_VMOV_lane<suffix, U, (ins MVEVectorIndex<8>:$Idx), dir> {
  bits<3> Idx;
  let Inst{22} = 0b0;
  let Inst{5} = 0b1;
  let Inst{16} = Idx{2};
  let Inst{21} = Idx{1};
  let Inst{6} = Idx{0};
}

class MVE_VMOV_lane_8<string suffix, bit U, MVE_VMOV_lane_direction dir>
  : MVE_VMOV_lane<suffix, U, (ins MVEVectorIndex<16>:$Idx), dir> {
  bits<4> Idx;
  let Inst{22} = 0b1;
  let Inst{16} = Idx{3};
  let Inst{21} = Idx{2};
  let Inst{6} = Idx{1};
  let Inst{5} = Idx{0};
}

def MVE_VMOV_from_lane_32  : MVE_VMOV_lane_32<            MVE_VMOV_from_lane>;
def MVE_VMOV_to_lane_32    : MVE_VMOV_lane_32<            MVE_VMOV_to_lane>;
def MVE_VMOV_from_lane_s16 : MVE_VMOV_lane_16<"s16", 0b0, MVE_VMOV_from_lane>;
def MVE_VMOV_from_lane_u16 : MVE_VMOV_lane_16<"u16", 0b1, MVE_VMOV_from_lane>;
def MVE_VMOV_to_lane_16    : MVE_VMOV_lane_16< "16", 0b0, MVE_VMOV_to_lane>;
def MVE_VMOV_from_lane_s8  : MVE_VMOV_lane_8 < "s8", 0b0, MVE_VMOV_from_lane>;
def MVE_VMOV_from_lane_u8  : MVE_VMOV_lane_8 < "u8", 0b1, MVE_VMOV_from_lane>;
def MVE_VMOV_to_lane_8     : MVE_VMOV_lane_8 <  "8", 0b0, MVE_VMOV_to_lane>;

let Predicates = [HasMVEInt] in {
  def : Pat<(extractelt (v2f64 MQPR:$src), imm:$lane),
            (f64 (EXTRACT_SUBREG MQPR:$src, (DSubReg_f64_reg imm:$lane)))>;
  def : Pat<(insertelt (v2f64 MQPR:$src1), DPR:$src2, imm:$lane),
            (INSERT_SUBREG (v2f64 (COPY_TO_REGCLASS MQPR:$src1, MQPR)), DPR:$src2, (DSubReg_f64_reg imm:$lane))>;

  def : Pat<(extractelt (v4i32 MQPR:$src), imm:$lane),
            (COPY_TO_REGCLASS
              (i32 (EXTRACT_SUBREG MQPR:$src, (SSubReg_f32_reg imm:$lane))), rGPR)>;
  def : Pat<(insertelt (v4i32 MQPR:$src1), rGPR:$src2, imm:$lane),
            (MVE_VMOV_to_lane_32 MQPR:$src1, rGPR:$src2, imm:$lane)>;

  def : Pat<(vector_insert (v16i8 MQPR:$src1), rGPR:$src2, imm:$lane),
            (MVE_VMOV_to_lane_8  MQPR:$src1, rGPR:$src2, imm:$lane)>;
  def : Pat<(vector_insert (v8i16 MQPR:$src1), rGPR:$src2, imm:$lane),
            (MVE_VMOV_to_lane_16 MQPR:$src1, rGPR:$src2, imm:$lane)>;

  def : Pat<(ARMvgetlanes (v16i8 MQPR:$src), imm:$lane),
            (MVE_VMOV_from_lane_s8 MQPR:$src, imm:$lane)>;
  def : Pat<(ARMvgetlanes (v8i16 MQPR:$src), imm:$lane),
            (MVE_VMOV_from_lane_s16 MQPR:$src, imm:$lane)>;
  def : Pat<(ARMvgetlaneu (v16i8 MQPR:$src), imm:$lane),
            (MVE_VMOV_from_lane_u8 MQPR:$src, imm:$lane)>;
  def : Pat<(ARMvgetlaneu (v8i16 MQPR:$src), imm:$lane),
            (MVE_VMOV_from_lane_u16 MQPR:$src, imm:$lane)>;

  def : Pat<(v16i8 (scalar_to_vector GPR:$src)),
            (MVE_VMOV_to_lane_8  (v16i8 (IMPLICIT_DEF)), rGPR:$src, (i32 0))>;
  def : Pat<(v8i16 (scalar_to_vector GPR:$src)),
            (MVE_VMOV_to_lane_16 (v8i16 (IMPLICIT_DEF)), rGPR:$src, (i32 0))>;
  def : Pat<(v4i32 (scalar_to_vector GPR:$src)),
            (MVE_VMOV_to_lane_32 (v4i32 (IMPLICIT_DEF)), rGPR:$src, (i32 0))>;

  // Floating point patterns, still enabled under HasMVEInt
  def : Pat<(extractelt (v4f32 MQPR:$src), imm:$lane),
            (COPY_TO_REGCLASS (f32 (EXTRACT_SUBREG MQPR:$src, (SSubReg_f32_reg imm:$lane))), SPR)>;
  def : Pat<(insertelt (v4f32 MQPR:$src1), (f32 SPR:$src2), imm:$lane),
            (INSERT_SUBREG (v4f32 (COPY_TO_REGCLASS MQPR:$src1, MQPR)), SPR:$src2, (SSubReg_f32_reg imm:$lane))>;

  def : Pat<(insertelt (v8f16 MQPR:$src1), HPR:$src2, imm:$lane),
            (MVE_VMOV_to_lane_16 MQPR:$src1, (COPY_TO_REGCLASS HPR:$src2, rGPR), imm:$lane)>;
  def : Pat<(extractelt (v8f16 MQPR:$src), imm_even:$lane),
            (EXTRACT_SUBREG MQPR:$src, (SSubReg_f16_reg imm_even:$lane))>;
  def : Pat<(extractelt (v8f16 MQPR:$src), imm_odd:$lane),
            (COPY_TO_REGCLASS
              (VMOVH (EXTRACT_SUBREG MQPR:$src, (SSubReg_f16_reg imm_odd:$lane))),
              HPR)>;

  def : Pat<(v4f32 (scalar_to_vector SPR:$src)),
            (INSERT_SUBREG (v4f32 (IMPLICIT_DEF)), SPR:$src, ssub_0)>;
  def : Pat<(v4f32 (scalar_to_vector GPR:$src)),
            (MVE_VMOV_to_lane_32 (v4f32 (IMPLICIT_DEF)), rGPR:$src, (i32 0))>;
  def : Pat<(v8f16 (scalar_to_vector HPR:$src)),
            (INSERT_SUBREG (v8f16 (IMPLICIT_DEF)), HPR:$src, ssub_0)>;
  def : Pat<(v8f16 (scalar_to_vector GPR:$src)),
            (MVE_VMOV_to_lane_16 (v8f16 (IMPLICIT_DEF)), rGPR:$src, (i32 0))>;
}

// end of mve_bit instructions

// start of MVE Integer instructions

class MVE_int<string iname, string suffix, bits<2> size, list<dag> pattern=[]>
  : MVE_p<(outs MQPR:$Qd), (ins MQPR:$Qn, MQPR:$Qm), NoItinerary,
          iname, suffix, "$Qd, $Qn, $Qm", vpred_r, "", pattern> {
  bits<4> Qd;
  bits<4> Qn;
  bits<4> Qm;

  let Inst{22} = Qd{3};
  let Inst{21-20} = size;
  let Inst{19-17} = Qn{2-0};
  let Inst{15-13} = Qd{2-0};
  let Inst{7} = Qn{3};
  let Inst{6} = 0b1;
  let Inst{5} = Qm{3};
  let Inst{3-1} = Qm{2-0};
}

class MVE_VMULt1<string iname, string suffix, bits<2> size,
                   list<dag> pattern=[]>
  : MVE_int<iname, suffix, size, pattern> {

  let Inst{28} = 0b0;
  let Inst{25-23} = 0b110;
  let Inst{16} = 0b0;
  let Inst{12-8} = 0b01001;
  let Inst{4} = 0b1;
  let Inst{0} = 0b0;
  let validForTailPredication = 1;
}

multiclass MVE_VMUL_m<string iname, MVEVectorVTInfo VTI,
                      SDNode unpred_op, Intrinsic pred_int> {
  def "" : MVE_VMULt1<iname, VTI.Suffix, VTI.Size>;

  let Predicates = [HasMVEInt] in {
    // Unpredicated multiply
    def : Pat<(VTI.Vec (unpred_op (VTI.Vec MQPR:$Qm), (VTI.Vec MQPR:$Qn))),
              (VTI.Vec (!cast<Instruction>(NAME)
                            (VTI.Vec MQPR:$Qm), (VTI.Vec MQPR:$Qn)))>;

    // Predicated multiply
    def : Pat<(VTI.Vec (pred_int (VTI.Vec MQPR:$Qm), (VTI.Vec MQPR:$Qn),
                            (VTI.Pred VCCR:$mask), (VTI.Vec MQPR:$inactive))),
              (VTI.Vec (!cast<Instruction>(NAME)
                            (VTI.Vec MQPR:$Qm), (VTI.Vec MQPR:$Qn),
                            (i32 1), (VTI.Pred VCCR:$mask),
                            (VTI.Vec MQPR:$inactive)))>;
  }
}

multiclass MVE_VMUL<MVEVectorVTInfo VTI>
  : MVE_VMUL_m<"vmul", VTI, mul, int_arm_mve_mul_predicated>;

defm MVE_VMULi8  : MVE_VMUL<MVE_v16i8>;
defm MVE_VMULi16 : MVE_VMUL<MVE_v8i16>;
defm MVE_VMULi32 : MVE_VMUL<MVE_v4i32>;

class MVE_VQxDMULH<string iname, string suffix, bits<2> size, bit rounding,
                  list<dag> pattern=[]>
  : MVE_int<iname, suffix, size, pattern> {

  let Inst{28} = rounding;
  let Inst{25-23} = 0b110;
  let Inst{16} = 0b0;
  let Inst{12-8} = 0b01011;
  let Inst{4} = 0b0;
  let Inst{0} = 0b0;
}

class MVE_VQDMULH<string suffix, bits<2> size, list<dag> pattern=[]>
  : MVE_VQxDMULH<"vqdmulh", suffix, size, 0b0, pattern>;
class MVE_VQRDMULH<string suffix, bits<2> size, list<dag> pattern=[]>
  : MVE_VQxDMULH<"vqrdmulh", suffix, size, 0b1, pattern>;

def MVE_VQDMULHi8   : MVE_VQDMULH<"s8",  0b00>;
def MVE_VQDMULHi16  : MVE_VQDMULH<"s16", 0b01>;
def MVE_VQDMULHi32  : MVE_VQDMULH<"s32", 0b10>;

def MVE_VQRDMULHi8  : MVE_VQRDMULH<"s8",  0b00>;
def MVE_VQRDMULHi16 : MVE_VQRDMULH<"s16", 0b01>;
def MVE_VQRDMULHi32 : MVE_VQRDMULH<"s32", 0b10>;

class MVE_VADDSUB<string iname, string suffix, bits<2> size, bit subtract,
                    list<dag> pattern=[]>
  : MVE_int<iname, suffix, size, pattern> {

  let Inst{28} = subtract;
  let Inst{25-23} = 0b110;
  let Inst{16} = 0b0;
  let Inst{12-8} = 0b01000;
  let Inst{4} = 0b0;
  let Inst{0} = 0b0;
  let validForTailPredication = 1;
}

multiclass MVE_VADDSUB_m<string iname, MVEVectorVTInfo VTI, bit subtract,
                         SDNode unpred_op, Intrinsic pred_int> {
  def "" : MVE_VADDSUB<iname, VTI.Suffix, VTI.Size, subtract>;

  let Predicates = [HasMVEInt] in {
    // Unpredicated add/subtract
    def : Pat<(VTI.Vec (unpred_op (VTI.Vec MQPR:$Qm), (VTI.Vec MQPR:$Qn))),
              (VTI.Vec (!cast<Instruction>(NAME)
                            (VTI.Vec MQPR:$Qm), (VTI.Vec MQPR:$Qn)))>;

    // Predicated add/subtract
    def : Pat<(VTI.Vec (pred_int (VTI.Vec MQPR:$Qm), (VTI.Vec MQPR:$Qn),
                            (VTI.Pred VCCR:$mask), (VTI.Vec MQPR:$inactive))),
              (VTI.Vec (!cast<Instruction>(NAME)
                            (VTI.Vec MQPR:$Qm), (VTI.Vec MQPR:$Qn),
                            (i32 1), (VTI.Pred VCCR:$mask),
                            (VTI.Vec MQPR:$inactive)))>;
  }
}

multiclass MVE_VADD<MVEVectorVTInfo VTI>
  : MVE_VADDSUB_m<"vadd", VTI, 0b0, add, int_arm_mve_add_predicated>;
multiclass MVE_VSUB<MVEVectorVTInfo VTI>
  : MVE_VADDSUB_m<"vsub", VTI, 0b1, sub, int_arm_mve_sub_predicated>;

defm MVE_VADDi8  : MVE_VADD<MVE_v16i8>;
defm MVE_VADDi16 : MVE_VADD<MVE_v8i16>;
defm MVE_VADDi32 : MVE_VADD<MVE_v4i32>;

defm MVE_VSUBi8  : MVE_VSUB<MVE_v16i8>;
defm MVE_VSUBi16 : MVE_VSUB<MVE_v8i16>;
defm MVE_VSUBi32 : MVE_VSUB<MVE_v4i32>;

class MVE_VQADDSUB<string iname, string suffix, bit U, bit subtract,
                   bits<2> size, ValueType vt>
  : MVE_int<iname, suffix, size, []> {

  let Inst{28} = U;
  let Inst{25-23} = 0b110;
  let Inst{16} = 0b0;
  let Inst{12-10} = 0b000;
  let Inst{9} = subtract;
  let Inst{8} = 0b0;
  let Inst{4} = 0b1;
  let Inst{0} = 0b0;
  let validForTailPredication = 1;

  ValueType VT = vt;
}

class MVE_VQADD<string suffix, bit U, bits<2> size, ValueType VT>
  : MVE_VQADDSUB<"vqadd", suffix, U, 0b0, size, VT>;
class MVE_VQSUB<string suffix, bit U, bits<2> size, ValueType VT>
  : MVE_VQADDSUB<"vqsub", suffix, U, 0b1, size, VT>;

def MVE_VQADDs8  : MVE_VQADD<"s8",  0b0, 0b00, v16i8>;
def MVE_VQADDs16 : MVE_VQADD<"s16", 0b0, 0b01, v8i16>;
def MVE_VQADDs32 : MVE_VQADD<"s32", 0b0, 0b10, v4i32>;
def MVE_VQADDu8  : MVE_VQADD<"u8",  0b1, 0b00, v16i8>;
def MVE_VQADDu16 : MVE_VQADD<"u16", 0b1, 0b01, v8i16>;
def MVE_VQADDu32 : MVE_VQADD<"u32", 0b1, 0b10, v4i32>;

def MVE_VQSUBs8  : MVE_VQSUB<"s8",  0b0, 0b00, v16i8>;
def MVE_VQSUBs16 : MVE_VQSUB<"s16", 0b0, 0b01, v8i16>;
def MVE_VQSUBs32 : MVE_VQSUB<"s32", 0b0, 0b10, v4i32>;
def MVE_VQSUBu8  : MVE_VQSUB<"u8",  0b1, 0b00, v16i8>;
def MVE_VQSUBu16 : MVE_VQSUB<"u16", 0b1, 0b01, v8i16>;
def MVE_VQSUBu32 : MVE_VQSUB<"u32", 0b1, 0b10, v4i32>;

let Predicates = [HasMVEInt] in {
  foreach instr = [MVE_VQADDu8, MVE_VQADDu16, MVE_VQADDu32] in
    foreach VT = [instr.VT] in
      def : Pat<(VT (uaddsat (VT MQPR:$Qm), (VT MQPR:$Qn))),
                (VT (instr (VT MQPR:$Qm), (VT MQPR:$Qn)))>;
  foreach instr = [MVE_VQADDs8, MVE_VQADDs16, MVE_VQADDs32] in
    foreach VT = [instr.VT] in
      def : Pat<(VT (saddsat (VT MQPR:$Qm), (VT MQPR:$Qn))),
                (VT (instr (VT MQPR:$Qm), (VT MQPR:$Qn)))>;
  foreach instr = [MVE_VQSUBu8, MVE_VQSUBu16, MVE_VQSUBu32] in
    foreach VT = [instr.VT] in
      def : Pat<(VT (usubsat (VT MQPR:$Qm), (VT MQPR:$Qn))),
                (VT (instr (VT MQPR:$Qm), (VT MQPR:$Qn)))>;
  foreach instr = [MVE_VQSUBs8, MVE_VQSUBs16, MVE_VQSUBs32] in
    foreach VT = [instr.VT] in
      def : Pat<(VT (ssubsat (VT MQPR:$Qm), (VT MQPR:$Qn))),
                (VT (instr (VT MQPR:$Qm), (VT MQPR:$Qn)))>;
}


class MVE_VABD_int<string suffix, bit U, bits<2> size,
                     list<dag> pattern=[]>
  : MVE_int<"vabd", suffix, size, pattern> {

  let Inst{28} = U;
  let Inst{25-23} = 0b110;
  let Inst{16} = 0b0;
  let Inst{12-8} = 0b00111;
  let Inst{4} = 0b0;
  let Inst{0} = 0b0;
  let validForTailPredication = 1;
}

multiclass MVE_VABD_m<MVEVectorVTInfo VTI,
                      Intrinsic unpred_int, Intrinsic pred_int> {
  def "" : MVE_VABD_int<VTI.Suffix, VTI.Unsigned, VTI.Size>;

  let Predicates = [HasMVEInt] in {
    // Unpredicated absolute difference
    def : Pat<(VTI.Vec (unpred_int (VTI.Vec MQPR:$Qm), (VTI.Vec MQPR:$Qn))),
              (VTI.Vec (!cast<Instruction>(NAME)
                            (VTI.Vec MQPR:$Qm), (VTI.Vec MQPR:$Qn)))>;

    // Predicated absolute difference
    def : Pat<(VTI.Vec (pred_int (VTI.Vec MQPR:$Qm), (VTI.Vec MQPR:$Qn),
                            (VTI.Pred VCCR:$mask), (VTI.Vec MQPR:$inactive))),
              (VTI.Vec (!cast<Instruction>(NAME)
                            (VTI.Vec MQPR:$Qm), (VTI.Vec MQPR:$Qn),
                            (i32 1), (VTI.Pred VCCR:$mask),
                            (VTI.Vec MQPR:$inactive)))>;
  }
}

multiclass MVE_VABD<MVEVectorVTInfo VTI>
  : MVE_VABD_m<VTI, int_arm_mve_vabd, int_arm_mve_abd_predicated>;

defm MVE_VABDs8  : MVE_VABD<MVE_v16s8>;
defm MVE_VABDs16 : MVE_VABD<MVE_v8s16>;
defm MVE_VABDs32 : MVE_VABD<MVE_v4s32>;
defm MVE_VABDu8  : MVE_VABD<MVE_v16u8>;
defm MVE_VABDu16 : MVE_VABD<MVE_v8u16>;
defm MVE_VABDu32 : MVE_VABD<MVE_v4u32>;

class MVE_VRHADD<string suffix, bit U, bits<2> size, list<dag> pattern=[]>
  : MVE_int<"vrhadd", suffix, size, pattern> {

  let Inst{28} = U;
  let Inst{25-23} = 0b110;
  let Inst{16} = 0b0;
  let Inst{12-8} = 0b00001;
  let Inst{4} = 0b0;
  let Inst{0} = 0b0;
  let validForTailPredication = 1;
}

def MVE_VRHADDs8  : MVE_VRHADD<"s8", 0b0, 0b00>;
def MVE_VRHADDs16 : MVE_VRHADD<"s16", 0b0, 0b01>;
def MVE_VRHADDs32 : MVE_VRHADD<"s32", 0b0, 0b10>;
def MVE_VRHADDu8  : MVE_VRHADD<"u8", 0b1, 0b00>;
def MVE_VRHADDu16 : MVE_VRHADD<"u16", 0b1, 0b01>;
def MVE_VRHADDu32 : MVE_VRHADD<"u32", 0b1, 0b10>;

class MVE_VHADDSUB<string iname, string suffix, bit U, bit subtract,
                   bits<2> size, list<dag> pattern=[]>
  : MVE_int<iname, suffix, size, pattern> {

  let Inst{28} = U;
  let Inst{25-23} = 0b110;
  let Inst{16} = 0b0;
  let Inst{12-10} = 0b000;
  let Inst{9} = subtract;
  let Inst{8} = 0b0;
  let Inst{4} = 0b0;
  let Inst{0} = 0b0;
  let validForTailPredication = 1;
}

class MVE_VHADD<string suffix, bit U, bits<2> size,
              list<dag> pattern=[]>
  : MVE_VHADDSUB<"vhadd", suffix, U, 0b0, size, pattern>;
class MVE_VHSUB<string suffix, bit U, bits<2> size,
              list<dag> pattern=[]>
  : MVE_VHADDSUB<"vhsub", suffix, U, 0b1, size, pattern>;

def MVE_VHADDs8  : MVE_VHADD<"s8",  0b0, 0b00>;
def MVE_VHADDs16 : MVE_VHADD<"s16", 0b0, 0b01>;
def MVE_VHADDs32 : MVE_VHADD<"s32", 0b0, 0b10>;
def MVE_VHADDu8  : MVE_VHADD<"u8",  0b1, 0b00>;
def MVE_VHADDu16 : MVE_VHADD<"u16", 0b1, 0b01>;
def MVE_VHADDu32 : MVE_VHADD<"u32", 0b1, 0b10>;

def MVE_VHSUBs8  : MVE_VHSUB<"s8",  0b0, 0b00>;
def MVE_VHSUBs16 : MVE_VHSUB<"s16", 0b0, 0b01>;
def MVE_VHSUBs32 : MVE_VHSUB<"s32", 0b0, 0b10>;
def MVE_VHSUBu8  : MVE_VHSUB<"u8",  0b1, 0b00>;
def MVE_VHSUBu16 : MVE_VHSUB<"u16", 0b1, 0b01>;
def MVE_VHSUBu32 : MVE_VHSUB<"u32", 0b1, 0b10>;

let Predicates = [HasMVEInt] in {
  def : Pat<(v16i8 (ARMvshrsImm
                     (v16i8 (add (v16i8 MQPR:$v1), (v16i8 MQPR:$v2))), 1)),
            (v16i8 (MVE_VHADDs8
                     (v16i8 MQPR:$v1), (v16i8 MQPR:$v2)))>;
  def : Pat<(v8i16 (ARMvshrsImm
                     (v8i16 (add (v8i16 MQPR:$v1), (v8i16 MQPR:$v2))), 1)),
            (v8i16 (MVE_VHADDs16
                     (v8i16 MQPR:$v1), (v8i16 MQPR:$v2)))>;
  def : Pat<(v4i32 (ARMvshrsImm
                     (v4i32 (add (v4i32 MQPR:$v1), (v4i32 MQPR:$v2))), 1)),
            (v4i32 (MVE_VHADDs32
                     (v4i32 MQPR:$v1), (v4i32 MQPR:$v2)))>;

  def : Pat<(v16i8 (ARMvshruImm
                     (v16i8 (add (v16i8 MQPR:$v1), (v16i8 MQPR:$v2))), 1)),
            (v16i8 (MVE_VHADDu8
                     (v16i8 MQPR:$v1), (v16i8 MQPR:$v2)))>;
  def : Pat<(v8i16 (ARMvshruImm
                     (v8i16 (add (v8i16 MQPR:$v1), (v8i16 MQPR:$v2))), 1)),
            (v8i16 (MVE_VHADDu16
                     (v8i16 MQPR:$v1), (v8i16 MQPR:$v2)))>;
  def : Pat<(v4i32 (ARMvshruImm
                     (v4i32 (add (v4i32 MQPR:$v1), (v4i32 MQPR:$v2))), 1)),
            (v4i32 (MVE_VHADDu32
                     (v4i32 MQPR:$v1), (v4i32 MQPR:$v2)))>;

  def : Pat<(v16i8 (ARMvshrsImm
                     (v16i8 (sub (v16i8 MQPR:$v1), (v16i8 MQPR:$v2))), 1)),
            (v16i8 (MVE_VHSUBs8
                     (v16i8 MQPR:$v1), (v16i8 MQPR:$v2)))>;
  def : Pat<(v8i16 (ARMvshrsImm
                     (v8i16 (sub (v8i16 MQPR:$v1), (v8i16 MQPR:$v2))), 1)),
            (v8i16 (MVE_VHSUBs16
                     (v8i16 MQPR:$v1), (v8i16 MQPR:$v2)))>;
  def : Pat<(v4i32 (ARMvshrsImm
                     (v4i32 (sub (v4i32 MQPR:$v1), (v4i32 MQPR:$v2))), 1)),
            (v4i32 (MVE_VHSUBs32
                     (v4i32 MQPR:$v1), (v4i32 MQPR:$v2)))>;

  def : Pat<(v16i8 (ARMvshruImm
                     (v16i8 (sub (v16i8 MQPR:$v1), (v16i8 MQPR:$v2))), 1)),
            (v16i8 (MVE_VHSUBu8
                     (v16i8 MQPR:$v1), (v16i8 MQPR:$v2)))>;
  def : Pat<(v8i16 (ARMvshruImm
                     (v8i16 (sub (v8i16 MQPR:$v1), (v8i16 MQPR:$v2))), 1)),
            (v8i16 (MVE_VHSUBu16
                     (v8i16 MQPR:$v1), (v8i16 MQPR:$v2)))>;
  def : Pat<(v4i32 (ARMvshruImm
                     (v4i32 (sub (v4i32 MQPR:$v1), (v4i32 MQPR:$v2))), 1)),
            (v4i32 (MVE_VHSUBu32
                     (v4i32 MQPR:$v1), (v4i32 MQPR:$v2)))>;
}

class MVE_VDUP<string suffix, bit B, bit E, list<dag> pattern=[]>
  : MVE_p<(outs MQPR:$Qd), (ins rGPR:$Rt), NoItinerary,
          "vdup", suffix, "$Qd, $Rt", vpred_r, "", pattern> {
  bits<4> Qd;
  bits<4> Rt;

  let Inst{28} = 0b0;
  let Inst{25-23} = 0b101;
  let Inst{22} = B;
  let Inst{21-20} = 0b10;
  let Inst{19-17} = Qd{2-0};
  let Inst{16} = 0b0;
  let Inst{15-12} = Rt;
  let Inst{11-8} = 0b1011;
  let Inst{7} = Qd{3};
  let Inst{6} = 0b0;
  let Inst{5} = E;
  let Inst{4-0} = 0b10000;
  let validForTailPredication = 1;
}

def MVE_VDUP32 : MVE_VDUP<"32", 0b0, 0b0>;
def MVE_VDUP16 : MVE_VDUP<"16", 0b0, 0b1>;
def MVE_VDUP8  : MVE_VDUP<"8",  0b1, 0b0>;

let Predicates = [HasMVEInt] in {
  def : Pat<(v16i8 (ARMvdup (i32 rGPR:$elem))),
            (MVE_VDUP8  rGPR:$elem)>;
  def : Pat<(v8i16 (ARMvdup (i32 rGPR:$elem))),
            (MVE_VDUP16 rGPR:$elem)>;
  def : Pat<(v4i32 (ARMvdup (i32 rGPR:$elem))),
            (MVE_VDUP32 rGPR:$elem)>;

  def : Pat<(v4i32 (ARMvduplane (v4i32 MQPR:$src), imm:$lane)),
            (MVE_VDUP32 (MVE_VMOV_from_lane_32 MQPR:$src, imm:$lane))>;
  // For the 16-bit and 8-bit vduplanes we don't care about the signedness
  // of the lane move operation as we only want the lowest 8/16 bits anyway.
  def : Pat<(v8i16 (ARMvduplane (v8i16 MQPR:$src), imm:$lane)),
            (MVE_VDUP16 (MVE_VMOV_from_lane_u16 MQPR:$src, imm:$lane))>;
  def : Pat<(v16i8 (ARMvduplane (v16i8 MQPR:$src), imm:$lane)),
            (MVE_VDUP8  (MVE_VMOV_from_lane_u8 MQPR:$src, imm:$lane))>;

  def : Pat<(v4f32 (ARMvdup (f32 SPR:$elem))),
            (v4f32 (MVE_VDUP32 (i32 (COPY_TO_REGCLASS (f32 SPR:$elem), rGPR))))>;
  def : Pat<(v8f16 (ARMvdup (f16 HPR:$elem))),
            (v8f16 (MVE_VDUP16 (i32 (COPY_TO_REGCLASS (f16 HPR:$elem), rGPR))))>;

  def : Pat<(v4f32 (ARMvduplane (v4f32 MQPR:$src), imm:$lane)),
            (MVE_VDUP32 (MVE_VMOV_from_lane_32 MQPR:$src, imm:$lane))>;
  def : Pat<(v8f16 (ARMvduplane (v8f16 MQPR:$src), imm:$lane)),
            (MVE_VDUP16 (MVE_VMOV_from_lane_u16 MQPR:$src, imm:$lane))>;
}


class MVEIntSingleSrc<string iname, string suffix, bits<2> size,
                         list<dag> pattern=[]>
  : MVE_p<(outs MQPR:$Qd), (ins MQPR:$Qm), NoItinerary,
          iname, suffix, "$Qd, $Qm", vpred_r, "", pattern> {
  bits<4> Qd;
  bits<4> Qm;

  let Inst{22} = Qd{3};
  let Inst{19-18} = size{1-0};
  let Inst{15-13} = Qd{2-0};
  let Inst{5} = Qm{3};
  let Inst{3-1} = Qm{2-0};
}

class MVE_VCLSCLZ<string iname, string suffix, bits<2> size,
                   bit count_zeroes, list<dag> pattern=[]>
  : MVEIntSingleSrc<iname, suffix, size, pattern> {

  let Inst{28} = 0b1;
  let Inst{25-23} = 0b111;
  let Inst{21-20} = 0b11;
  let Inst{17-16} = 0b00;
  let Inst{12-8} = 0b00100;
  let Inst{7} = count_zeroes;
  let Inst{6} = 0b1;
  let Inst{4} = 0b0;
  let Inst{0} = 0b0;
  let validForTailPredication = 1;
}

def MVE_VCLSs8  : MVE_VCLSCLZ<"vcls", "s8",  0b00, 0b0>;
def MVE_VCLSs16 : MVE_VCLSCLZ<"vcls", "s16", 0b01, 0b0>;
def MVE_VCLSs32 : MVE_VCLSCLZ<"vcls", "s32", 0b10, 0b0>;

def MVE_VCLZs8  : MVE_VCLSCLZ<"vclz", "i8",  0b00, 0b1>;
def MVE_VCLZs16 : MVE_VCLSCLZ<"vclz", "i16", 0b01, 0b1>;
def MVE_VCLZs32 : MVE_VCLSCLZ<"vclz", "i32", 0b10, 0b1>;

let Predicates = [HasMVEInt] in {
  def : Pat<(v16i8 ( ctlz (v16i8 MQPR:$val1))),
            (v16i8 ( MVE_VCLZs8 (v16i8 MQPR:$val1)))>;
  def : Pat<(v4i32 ( ctlz (v4i32 MQPR:$val1))),
            (v4i32 ( MVE_VCLZs32 (v4i32 MQPR:$val1)))>;
  def : Pat<(v8i16 ( ctlz (v8i16 MQPR:$val1))),
            (v8i16 ( MVE_VCLZs16 (v8i16 MQPR:$val1)))>;
}

class MVE_VABSNEG_int<string iname, string suffix, bits<2> size, bit negate,
                      list<dag> pattern=[]>
  : MVEIntSingleSrc<iname, suffix, size, pattern> {

  let Inst{28} = 0b1;
  let Inst{25-23} = 0b111;
  let Inst{21-20} = 0b11;
  let Inst{17-16} = 0b01;
  let Inst{12-8} = 0b00011;
  let Inst{7} = negate;
  let Inst{6} = 0b1;
  let Inst{4} = 0b0;
  let Inst{0} = 0b0;
  let validForTailPredication = 1;
}

def MVE_VABSs8  : MVE_VABSNEG_int<"vabs", "s8",  0b00, 0b0>;
def MVE_VABSs16 : MVE_VABSNEG_int<"vabs", "s16", 0b01, 0b0>;
def MVE_VABSs32 : MVE_VABSNEG_int<"vabs", "s32", 0b10, 0b0>;

let Predicates = [HasMVEInt] in {
  def : Pat<(v16i8 (abs (v16i8 MQPR:$v))),
            (v16i8 (MVE_VABSs8 $v))>;
  def : Pat<(v8i16 (abs (v8i16 MQPR:$v))),
            (v8i16 (MVE_VABSs16 $v))>;
  def : Pat<(v4i32 (abs (v4i32 MQPR:$v))),
            (v4i32 (MVE_VABSs32 $v))>;
}

def MVE_VNEGs8  : MVE_VABSNEG_int<"vneg", "s8",  0b00, 0b1>;
def MVE_VNEGs16 : MVE_VABSNEG_int<"vneg", "s16", 0b01, 0b1>;
def MVE_VNEGs32 : MVE_VABSNEG_int<"vneg", "s32", 0b10, 0b1>;

let Predicates = [HasMVEInt] in {
  def : Pat<(v16i8 (vnegq (v16i8 MQPR:$v))),
            (v16i8 (MVE_VNEGs8 $v))>;
  def : Pat<(v8i16 (vnegq (v8i16 MQPR:$v))),
            (v8i16 (MVE_VNEGs16 $v))>;
  def : Pat<(v4i32 (vnegq (v4i32 MQPR:$v))),
            (v4i32 (MVE_VNEGs32 $v))>;
}

class MVE_VQABSNEG<string iname, string suffix, bits<2> size,
                   bit negate, list<dag> pattern=[]>
  : MVEIntSingleSrc<iname, suffix, size, pattern> {

  let Inst{28} = 0b1;
  let Inst{25-23} = 0b111;
  let Inst{21-20} = 0b11;
  let Inst{17-16} = 0b00;
  let Inst{12-8} = 0b00111;
  let Inst{7} = negate;
  let Inst{6} = 0b1;
  let Inst{4} = 0b0;
  let Inst{0} = 0b0;
  let validForTailPredication = 1;
}

def MVE_VQABSs8  : MVE_VQABSNEG<"vqabs", "s8",  0b00, 0b0>;
def MVE_VQABSs16 : MVE_VQABSNEG<"vqabs", "s16", 0b01, 0b0>;
def MVE_VQABSs32 : MVE_VQABSNEG<"vqabs", "s32", 0b10, 0b0>;

def MVE_VQNEGs8  : MVE_VQABSNEG<"vqneg", "s8",  0b00, 0b1>;
def MVE_VQNEGs16 : MVE_VQABSNEG<"vqneg", "s16", 0b01, 0b1>;
def MVE_VQNEGs32 : MVE_VQABSNEG<"vqneg", "s32", 0b10, 0b1>;

// int_min/int_max: vector containing INT_MIN/INT_MAX VTI.Size times
// zero_vec: v4i32-initialized zero vector, potentially wrapped in a bitconvert
multiclass vqabsneg_pattern<MVEVectorVTInfo VTI, dag int_min, dag int_max,
                         dag zero_vec,  MVE_VQABSNEG vqabs_instruction,
                         MVE_VQABSNEG vqneg_instruction> {
  let Predicates = [HasMVEInt] in {
    // The below tree can be replaced by a vqabs instruction, as it represents
    // the following vectorized expression (r being the value in $reg):
    // r > 0 ? r : (r == INT_MIN ? INT_MAX : -r)
    def : Pat<(VTI.Vec (vselect
                      (VTI.Pred (ARMvcmpz (VTI.Vec MQPR:$reg), (i32 12))),
                      (VTI.Vec MQPR:$reg),
                      (VTI.Vec (vselect
                                (VTI.Pred (ARMvcmp (VTI.Vec MQPR:$reg), int_min, (i32 0))),
                                int_max,
                                (sub (VTI.Vec zero_vec), (VTI.Vec MQPR:$reg)))))),
            (VTI.Vec (vqabs_instruction (VTI.Vec MQPR:$reg)))>;
    // Similarly, this tree represents vqneg, i.e. the following vectorized expression:
    // r == INT_MIN ? INT_MAX : -r
    def : Pat<(VTI.Vec (vselect
                        (VTI.Pred (ARMvcmp (VTI.Vec MQPR:$reg), int_min, (i32 0))),
                        int_max,
                        (sub (VTI.Vec zero_vec), (VTI.Vec MQPR:$reg)))),
               (VTI.Vec (vqneg_instruction (VTI.Vec MQPR:$reg)))>;
  }
}

defm MVE_VQABSNEG_Ps8  : vqabsneg_pattern<MVE_v16i8,
                                    (v16i8 (ARMvmovImm (i32 3712))),
                                    (v16i8 (ARMvmovImm (i32 3711))),
                                    (bitconvert (v4i32 (ARMvmovImm (i32 0)))),
                                    MVE_VQABSs8, MVE_VQNEGs8>;
defm MVE_VQABSNEG_Ps16 : vqabsneg_pattern<MVE_v8i16,
                                    (v8i16 (ARMvmovImm (i32 2688))),
                                    (v8i16 (ARMvmvnImm (i32 2688))),
                                    (bitconvert (v4i32 (ARMvmovImm (i32 0)))),
                                    MVE_VQABSs16, MVE_VQNEGs16>;
defm MVE_VQABSNEG_Ps32 : vqabsneg_pattern<MVE_v4i32,
                                    (v4i32 (ARMvmovImm (i32 1664))),
                                    (v4i32 (ARMvmvnImm (i32 1664))),
                                    (ARMvmovImm (i32 0)),
                                    MVE_VQABSs32, MVE_VQNEGs32>;

class MVE_mod_imm<string iname, string suffix, bits<4> cmode, bit op,
                  dag iops, list<dag> pattern=[]>
  : MVE_p<(outs MQPR:$Qd), iops, NoItinerary, iname, suffix, "$Qd, $imm",
          vpred_r, "", pattern> {
  bits<13> imm;
  bits<4> Qd;

  let Inst{28} = imm{7};
  let Inst{25-23} = 0b111;
  let Inst{22} = Qd{3};
  let Inst{21-19} = 0b000;
  let Inst{18-16} = imm{6-4};
  let Inst{15-13} = Qd{2-0};
  let Inst{12} = 0b0;
  let Inst{11-8} = cmode{3-0};
  let Inst{7-6} = 0b01;
  let Inst{5} = op;
  let Inst{4} = 0b1;
  let Inst{3-0} = imm{3-0};

  let DecoderMethod = "DecodeMVEModImmInstruction";
  let validForTailPredication = 1;
}

let isReMaterializable = 1 in {
let isAsCheapAsAMove = 1 in {
def MVE_VMOVimmi8  : MVE_mod_imm<"vmov", "i8",  {1,1,1,0}, 0b0, (ins nImmSplatI8:$imm)>;
def MVE_VMOVimmi16 : MVE_mod_imm<"vmov", "i16", {1,0,?,0}, 0b0, (ins nImmSplatI16:$imm)> {
  let Inst{9} = imm{9};
}
def MVE_VMOVimmi32 : MVE_mod_imm<"vmov", "i32", {?,?,?,?}, 0b0, (ins nImmVMOVI32:$imm)> {
  let Inst{11-8} = imm{11-8};
}
def MVE_VMOVimmi64 : MVE_mod_imm<"vmov", "i64", {1,1,1,0}, 0b1, (ins nImmSplatI64:$imm)>;
def MVE_VMOVimmf32 : MVE_mod_imm<"vmov", "f32", {1,1,1,1}, 0b0, (ins nImmVMOVF32:$imm)>;
} // let isAsCheapAsAMove = 1

def MVE_VMVNimmi16 : MVE_mod_imm<"vmvn", "i16", {1,0,?,0}, 0b1, (ins nImmSplatI16:$imm)> {
  let Inst{9} = imm{9};
}
def MVE_VMVNimmi32 : MVE_mod_imm<"vmvn", "i32", {?,?,?,?}, 0b1, (ins nImmVMOVI32:$imm)> {
  let Inst{11-8} = imm{11-8};
}
} // let isReMaterializable = 1

let Predicates = [HasMVEInt] in {
  def : Pat<(v16i8 (ARMvmovImm timm:$simm)),
            (v16i8 (MVE_VMOVimmi8  nImmSplatI8:$simm))>;
  def : Pat<(v8i16 (ARMvmovImm timm:$simm)),
            (v8i16 (MVE_VMOVimmi16 nImmSplatI16:$simm))>;
  def : Pat<(v4i32 (ARMvmovImm timm:$simm)),
            (v4i32 (MVE_VMOVimmi32 nImmVMOVI32:$simm))>;

  def : Pat<(v8i16 (ARMvmvnImm timm:$simm)),
            (v8i16 (MVE_VMVNimmi16 nImmSplatI16:$simm))>;
  def : Pat<(v4i32 (ARMvmvnImm timm:$simm)),
            (v4i32 (MVE_VMVNimmi32 nImmVMOVI32:$simm))>;

  def : Pat<(v4f32 (ARMvmovFPImm timm:$simm)),
            (v4f32 (MVE_VMOVimmf32 nImmVMOVF32:$simm))>;
}

class MVE_VMINMAXA<string iname, string suffix, bits<2> size,
                   bit bit_12, list<dag> pattern=[]>
  : MVE_p<(outs MQPR:$Qd), (ins MQPR:$Qd_src, MQPR:$Qm),
          NoItinerary, iname, suffix, "$Qd, $Qm", vpred_n, "$Qd = $Qd_src",
          pattern> {
  bits<4> Qd;
  bits<4> Qm;

  let Inst{28} = 0b0;
  let Inst{25-23} = 0b100;
  let Inst{22} = Qd{3};
  let Inst{21-20} = 0b11;
  let Inst{19-18} = size;
  let Inst{17-16} = 0b11;
  let Inst{15-13} = Qd{2-0};
  let Inst{12} = bit_12;
  let Inst{11-6} = 0b111010;
  let Inst{5} = Qm{3};
  let Inst{4} = 0b0;
  let Inst{3-1} = Qm{2-0};
  let Inst{0} = 0b1;
}

def MVE_VMAXAs8  : MVE_VMINMAXA<"vmaxa", "s8",  0b00, 0b0>;
def MVE_VMAXAs16 : MVE_VMINMAXA<"vmaxa", "s16", 0b01, 0b0>;
def MVE_VMAXAs32 : MVE_VMINMAXA<"vmaxa", "s32", 0b10, 0b0>;

def MVE_VMINAs8  : MVE_VMINMAXA<"vmina", "s8",  0b00, 0b1>;
def MVE_VMINAs16 : MVE_VMINMAXA<"vmina", "s16", 0b01, 0b1>;
def MVE_VMINAs32 : MVE_VMINMAXA<"vmina", "s32", 0b10, 0b1>;

// end of MVE Integer instructions

// start of mve_imm_shift instructions

def MVE_VSHLC : MVE_p<(outs rGPR:$RdmDest, MQPR:$Qd),
                      (ins MQPR:$QdSrc, rGPR:$RdmSrc, long_shift:$imm),
                      NoItinerary, "vshlc", "", "$QdSrc, $RdmSrc, $imm",
                      vpred_n, "$RdmDest = $RdmSrc,$Qd = $QdSrc"> {
  bits<5> imm;
  bits<4> Qd;
  bits<4> RdmDest;

  let Inst{28} = 0b0;
  let Inst{25-23} = 0b101;
  let Inst{22} = Qd{3};
  let Inst{21} = 0b1;
  let Inst{20-16} = imm{4-0};
  let Inst{15-13} = Qd{2-0};
  let Inst{12-4} = 0b011111100;
  let Inst{3-0} = RdmDest{3-0};
}

class MVE_shift_imm<dag oops, dag iops, string iname, string suffix,
                    string ops, vpred_ops vpred, string cstr,
                    list<dag> pattern=[]>
  : MVE_p<oops, iops, NoItinerary, iname, suffix, ops, vpred, cstr, pattern> {
  bits<4> Qd;
  bits<4> Qm;

  let Inst{22} = Qd{3};
  let Inst{15-13} = Qd{2-0};
  let Inst{5} = Qm{3};
  let Inst{3-1} = Qm{2-0};
}

class MVE_VMOVL<string iname, string suffix, bits<2> sz, bit U,
              list<dag> pattern=[]>
  : MVE_shift_imm<(outs MQPR:$Qd), (ins MQPR:$Qm),
                  iname, suffix, "$Qd, $Qm", vpred_r, "",
                  pattern> {
  let Inst{28} = U;
  let Inst{25-23} = 0b101;
  let Inst{21} = 0b1;
  let Inst{20-19} = sz{1-0};
  let Inst{18-16} = 0b000;
  let Inst{11-6} = 0b111101;
  let Inst{4} = 0b0;
  let Inst{0} = 0b0;
}

multiclass MVE_VMOVL_shift_half<string iname, string suffix, bits<2> sz, bit U,
                                list<dag> pattern=[]> {
  def bh : MVE_VMOVL<!strconcat(iname, "b"), suffix, sz, U, pattern> {
    let Inst{12} = 0b0;
  }
  def th : MVE_VMOVL<!strconcat(iname, "t"), suffix, sz, U, pattern> {
    let Inst{12} = 0b1;
  }
}

defm MVE_VMOVLs8 : MVE_VMOVL_shift_half<"vmovl", "s8", 0b01, 0b0>;
defm MVE_VMOVLu8 : MVE_VMOVL_shift_half<"vmovl", "u8", 0b01, 0b1>;
defm MVE_VMOVLs16 : MVE_VMOVL_shift_half<"vmovl", "s16", 0b10, 0b0>;
defm MVE_VMOVLu16 : MVE_VMOVL_shift_half<"vmovl", "u16", 0b10, 0b1>;

let Predicates = [HasMVEInt] in {
  def : Pat<(sext_inreg (v4i32 MQPR:$src), v4i16),
            (MVE_VMOVLs16bh MQPR:$src)>;
  def : Pat<(sext_inreg (v8i16 MQPR:$src), v8i8),
            (MVE_VMOVLs8bh MQPR:$src)>;
  def : Pat<(sext_inreg (v4i32 MQPR:$src), v4i8),
            (MVE_VMOVLs16bh (MVE_VMOVLs8bh MQPR:$src))>;

  // zext_inreg 16 -> 32
  def : Pat<(and (v4i32 MQPR:$src), (v4i32 (ARMvmovImm (i32 0xCFF)))),
            (MVE_VMOVLu16bh MQPR:$src)>;
  // zext_inreg 8 -> 16
  def : Pat<(and (v8i16 MQPR:$src), (v8i16 (ARMvmovImm (i32 0x8FF)))),
            (MVE_VMOVLu8bh MQPR:$src)>;
}


class MVE_VSHLL_imm<string iname, string suffix, bit U, bit th,
                    dag immops, list<dag> pattern=[]>
  : MVE_shift_imm<(outs MQPR:$Qd), !con((ins MQPR:$Qm), immops),
                  iname, suffix, "$Qd, $Qm, $imm", vpred_r, "", pattern> {
  let Inst{28} = U;
  let Inst{25-23} = 0b101;
  let Inst{21} = 0b1;
  let Inst{12} = th;
  let Inst{11-6} = 0b111101;
  let Inst{4} = 0b0;
  let Inst{0} = 0b0;
}

// The immediate VSHLL instructions accept shift counts from 1 up to
// the lane width (8 or 16), but the full-width shifts have an
// entirely separate encoding, given below with 'lw' in the name.

class MVE_VSHLL_imm8<string iname, string suffix,
                     bit U, bit th, list<dag> pattern=[]>
  : MVE_VSHLL_imm<iname, suffix, U, th, (ins mve_shift_imm1_7:$imm), pattern> {
  bits<3> imm;
  let Inst{20-19} = 0b01;
  let Inst{18-16} = imm;
}

class MVE_VSHLL_imm16<string iname, string suffix,
                      bit U, bit th, list<dag> pattern=[]>
  : MVE_VSHLL_imm<iname, suffix, U, th, (ins mve_shift_imm1_15:$imm), pattern> {
  bits<4> imm;
  let Inst{20} = 0b1;
  let Inst{19-16} = imm;
}

def MVE_VSHLL_imms8bh  : MVE_VSHLL_imm8 <"vshllb", "s8", 0b0, 0b0>;
def MVE_VSHLL_imms8th  : MVE_VSHLL_imm8 <"vshllt", "s8", 0b0, 0b1>;
def MVE_VSHLL_immu8bh  : MVE_VSHLL_imm8 <"vshllb", "u8", 0b1, 0b0>;
def MVE_VSHLL_immu8th  : MVE_VSHLL_imm8 <"vshllt", "u8", 0b1, 0b1>;
def MVE_VSHLL_imms16bh : MVE_VSHLL_imm16<"vshllb", "s16", 0b0, 0b0>;
def MVE_VSHLL_imms16th : MVE_VSHLL_imm16<"vshllt", "s16", 0b0, 0b1>;
def MVE_VSHLL_immu16bh : MVE_VSHLL_imm16<"vshllb", "u16", 0b1, 0b0>;
def MVE_VSHLL_immu16th : MVE_VSHLL_imm16<"vshllt", "u16", 0b1, 0b1>;

class MVE_VSHLL_by_lane_width<string iname, string suffix, bits<2> size,
                              bit U, string ops, list<dag> pattern=[]>
  : MVE_shift_imm<(outs MQPR:$Qd), (ins MQPR:$Qm),
                  iname, suffix, ops, vpred_r, "", pattern> {
  let Inst{28} = U;
  let Inst{25-23} = 0b100;
  let Inst{21-20} = 0b11;
  let Inst{19-18} = size{1-0};
  let Inst{17-16} = 0b01;
  let Inst{11-6} = 0b111000;
  let Inst{4} = 0b0;
  let Inst{0} = 0b1;
}

multiclass MVE_VSHLL_lw<string iname, string suffix, bits<2> sz, bit U,
                              string ops, list<dag> pattern=[]> {
  def bh : MVE_VSHLL_by_lane_width<iname#"b", suffix, sz, U, ops, pattern> {
    let Inst{12} = 0b0;
  }
  def th : MVE_VSHLL_by_lane_width<iname#"t", suffix, sz, U, ops, pattern> {
    let Inst{12} = 0b1;
  }
}

defm MVE_VSHLL_lws8  : MVE_VSHLL_lw<"vshll", "s8",  0b00, 0b0, "$Qd, $Qm, #8">;
defm MVE_VSHLL_lws16 : MVE_VSHLL_lw<"vshll", "s16", 0b01, 0b0, "$Qd, $Qm, #16">;
defm MVE_VSHLL_lwu8  : MVE_VSHLL_lw<"vshll", "u8",  0b00, 0b1, "$Qd, $Qm, #8">;
defm MVE_VSHLL_lwu16 : MVE_VSHLL_lw<"vshll", "u16", 0b01, 0b1, "$Qd, $Qm, #16">;

class MVE_VxSHRN<string iname, string suffix, bit bit_12, bit bit_28,
               dag immops, list<dag> pattern=[]>
  : MVE_shift_imm<(outs MQPR:$Qd), !con((ins MQPR:$QdSrc, MQPR:$Qm), immops),
                  iname, suffix, "$Qd, $Qm, $imm", vpred_n, "$Qd = $QdSrc",
                  pattern> {
  bits<5> imm;

  let Inst{28} = bit_28;
  let Inst{25-23} = 0b101;
  let Inst{21} = 0b0;
  let Inst{20-16} = imm{4-0};
  let Inst{12} = bit_12;
  let Inst{11-6} = 0b111111;
  let Inst{4} = 0b0;
  let Inst{0} = 0b1;
}

def MVE_VRSHRNi16bh : MVE_VxSHRN<
    "vrshrnb", "i16", 0b0, 0b1, (ins shr_imm8:$imm)> {
  let Inst{20-19} = 0b01;
}
def MVE_VRSHRNi16th : MVE_VxSHRN<
    "vrshrnt", "i16", 0b1, 0b1,(ins shr_imm8:$imm)> {
  let Inst{20-19} = 0b01;
}
def MVE_VRSHRNi32bh : MVE_VxSHRN<
    "vrshrnb", "i32", 0b0, 0b1, (ins shr_imm16:$imm)> {
  let Inst{20} = 0b1;
}
def MVE_VRSHRNi32th : MVE_VxSHRN<
    "vrshrnt", "i32", 0b1, 0b1, (ins shr_imm16:$imm)> {
  let Inst{20} = 0b1;
}

def MVE_VSHRNi16bh : MVE_VxSHRN<
    "vshrnb", "i16", 0b0, 0b0, (ins shr_imm8:$imm)> {
  let Inst{20-19} = 0b01;
}
def MVE_VSHRNi16th : MVE_VxSHRN<
    "vshrnt", "i16", 0b1, 0b0, (ins shr_imm8:$imm)> {
  let Inst{20-19} = 0b01;
}
def MVE_VSHRNi32bh : MVE_VxSHRN<
    "vshrnb", "i32", 0b0, 0b0, (ins shr_imm16:$imm)> {
  let Inst{20} = 0b1;
}
def MVE_VSHRNi32th : MVE_VxSHRN<
    "vshrnt", "i32", 0b1, 0b0, (ins shr_imm16:$imm)> {
  let Inst{20} = 0b1;
}

class MVE_VxQRSHRUN<string iname, string suffix, bit bit_28, bit bit_12, dag immops,
                 list<dag> pattern=[]>
  : MVE_shift_imm<(outs MQPR:$Qd), !con((ins MQPR:$QdSrc, MQPR:$Qm), immops),
                  iname, suffix, "$Qd, $Qm, $imm", vpred_n, "$Qd = $QdSrc",
                  pattern> {
  bits<5> imm;

  let Inst{28} = bit_28;
  let Inst{25-23} = 0b101;
  let Inst{21} = 0b0;
  let Inst{20-16} = imm{4-0};
  let Inst{12} = bit_12;
  let Inst{11-6} = 0b111111;
  let Inst{4} = 0b0;
  let Inst{0} = 0b0;
}

def MVE_VQRSHRUNs16bh : MVE_VxQRSHRUN<
    "vqrshrunb", "s16", 0b1, 0b0, (ins shr_imm8:$imm)> {
  let Inst{20-19} = 0b01;
}
def MVE_VQRSHRUNs16th : MVE_VxQRSHRUN<
    "vqrshrunt", "s16", 0b1, 0b1, (ins shr_imm8:$imm)> {
  let Inst{20-19} = 0b01;
}
def MVE_VQRSHRUNs32bh : MVE_VxQRSHRUN<
    "vqrshrunb", "s32", 0b1, 0b0, (ins shr_imm16:$imm)> {
  let Inst{20} = 0b1;
}
def MVE_VQRSHRUNs32th : MVE_VxQRSHRUN<
    "vqrshrunt", "s32", 0b1, 0b1, (ins shr_imm16:$imm)> {
  let Inst{20} = 0b1;
}

def MVE_VQSHRUNs16bh : MVE_VxQRSHRUN<
    "vqshrunb", "s16", 0b0, 0b0, (ins shr_imm8:$imm)> {
  let Inst{20-19} = 0b01;
}
def MVE_VQSHRUNs16th : MVE_VxQRSHRUN<
    "vqshrunt", "s16", 0b0, 0b1, (ins shr_imm8:$imm)> {
  let Inst{20-19} = 0b01;
}
def MVE_VQSHRUNs32bh : MVE_VxQRSHRUN<
    "vqshrunb", "s32", 0b0, 0b0, (ins shr_imm16:$imm)> {
  let Inst{20} = 0b1;
}
def MVE_VQSHRUNs32th : MVE_VxQRSHRUN<
    "vqshrunt", "s32", 0b0, 0b1, (ins shr_imm16:$imm)> {
  let Inst{20} = 0b1;
}

class MVE_VxQRSHRN<string iname, string suffix, bit bit_0, bit bit_12,
                   dag immops, list<dag> pattern=[]>
  : MVE_shift_imm<(outs MQPR:$Qd), !con((ins MQPR:$QdSrc, MQPR:$Qm), immops),
                  iname, suffix, "$Qd, $Qm, $imm", vpred_n, "$Qd = $QdSrc",
                  pattern> {
  bits<5> imm;

  let Inst{25-23} = 0b101;
  let Inst{21} = 0b0;
  let Inst{20-16} = imm{4-0};
  let Inst{12} = bit_12;
  let Inst{11-6} = 0b111101;
  let Inst{4} = 0b0;
  let Inst{0} = bit_0;
}

multiclass MVE_VxQRSHRN_types<string iname, bit bit_0, bit bit_12> {
  def s16 : MVE_VxQRSHRN<iname, "s16", bit_0, bit_12, (ins shr_imm8:$imm)> {
    let Inst{28} = 0b0;
    let Inst{20-19} = 0b01;
  }
  def u16 : MVE_VxQRSHRN<iname, "u16", bit_0, bit_12, (ins shr_imm8:$imm)> {
    let Inst{28} = 0b1;
    let Inst{20-19} = 0b01;
  }
  def s32 : MVE_VxQRSHRN<iname, "s32", bit_0, bit_12, (ins shr_imm16:$imm)> {
    let Inst{28} = 0b0;
    let Inst{20} = 0b1;
  }
  def u32 : MVE_VxQRSHRN<iname, "u32", bit_0, bit_12, (ins shr_imm16:$imm)> {
    let Inst{28} = 0b1;
    let Inst{20} = 0b1;
  }
}

defm MVE_VQRSHRNbh : MVE_VxQRSHRN_types<"vqrshrnb", 0b1, 0b0>;
defm MVE_VQRSHRNth : MVE_VxQRSHRN_types<"vqrshrnt", 0b1, 0b1>;
defm MVE_VQSHRNbh  : MVE_VxQRSHRN_types<"vqshrnb", 0b0, 0b0>;
defm MVE_VQSHRNth  : MVE_VxQRSHRN_types<"vqshrnt", 0b0, 0b1>;

// end of mve_imm_shift instructions

// start of mve_shift instructions

class MVE_shift_by_vec<string iname, string suffix, bit U,
                       bits<2> size, bit bit_4, bit bit_8>
  : MVE_p<(outs MQPR:$Qd), (ins MQPR:$Qm, MQPR:$Qn), NoItinerary,
           iname, suffix, "$Qd, $Qm, $Qn", vpred_r, "", []> {
  // Shift instructions which take a vector of shift counts
  bits<4> Qd;
  bits<4> Qm;
  bits<4> Qn;

  let Inst{28} = U;
  let Inst{25-24} = 0b11;
  let Inst{23} = 0b0;
  let Inst{22} = Qd{3};
  let Inst{21-20} = size;
  let Inst{19-17} = Qn{2-0};
  let Inst{16} = 0b0;
  let Inst{15-13} = Qd{2-0};
  let Inst{12-9} = 0b0010;
  let Inst{8} = bit_8;
  let Inst{7} = Qn{3};
  let Inst{6} = 0b1;
  let Inst{5} = Qm{3};
  let Inst{4} = bit_4;
  let Inst{3-1} = Qm{2-0};
  let Inst{0} = 0b0;
  let validForTailPredication = 1;
}

multiclass mve_shift_by_vec_multi<string iname, bit bit_4, bit bit_8> {
  def s8  : MVE_shift_by_vec<iname, "s8", 0b0, 0b00, bit_4, bit_8>;
  def s16 : MVE_shift_by_vec<iname, "s16", 0b0, 0b01, bit_4, bit_8>;
  def s32 : MVE_shift_by_vec<iname, "s32", 0b0, 0b10, bit_4, bit_8>;
  def u8  : MVE_shift_by_vec<iname, "u8", 0b1, 0b00, bit_4, bit_8>;
  def u16 : MVE_shift_by_vec<iname, "u16", 0b1, 0b01, bit_4, bit_8>;
  def u32 : MVE_shift_by_vec<iname, "u32", 0b1, 0b10, bit_4, bit_8>;
}

defm MVE_VSHL_by_vec   : mve_shift_by_vec_multi<"vshl",   0b0, 0b0>;
defm MVE_VQSHL_by_vec  : mve_shift_by_vec_multi<"vqshl",  0b1, 0b0>;
defm MVE_VQRSHL_by_vec : mve_shift_by_vec_multi<"vqrshl", 0b1, 0b1>;
defm MVE_VRSHL_by_vec  : mve_shift_by_vec_multi<"vrshl",  0b0, 0b1>;

let Predicates = [HasMVEInt] in {
  def : Pat<(v4i32 (ARMvshlu (v4i32 MQPR:$Qm), (v4i32 MQPR:$Qn))),
            (v4i32 (MVE_VSHL_by_vecu32 (v4i32 MQPR:$Qm), (v4i32 MQPR:$Qn)))>;
  def : Pat<(v8i16 (ARMvshlu (v8i16 MQPR:$Qm), (v8i16 MQPR:$Qn))),
            (v8i16 (MVE_VSHL_by_vecu16 (v8i16 MQPR:$Qm), (v8i16 MQPR:$Qn)))>;
  def : Pat<(v16i8 (ARMvshlu (v16i8 MQPR:$Qm), (v16i8 MQPR:$Qn))),
            (v16i8 (MVE_VSHL_by_vecu8 (v16i8 MQPR:$Qm), (v16i8 MQPR:$Qn)))>;

  def : Pat<(v4i32 (ARMvshls (v4i32 MQPR:$Qm), (v4i32 MQPR:$Qn))),
            (v4i32 (MVE_VSHL_by_vecs32 (v4i32 MQPR:$Qm), (v4i32 MQPR:$Qn)))>;
  def : Pat<(v8i16 (ARMvshls (v8i16 MQPR:$Qm), (v8i16 MQPR:$Qn))),
            (v8i16 (MVE_VSHL_by_vecs16 (v8i16 MQPR:$Qm), (v8i16 MQPR:$Qn)))>;
  def : Pat<(v16i8 (ARMvshls (v16i8 MQPR:$Qm), (v16i8 MQPR:$Qn))),
            (v16i8 (MVE_VSHL_by_vecs8 (v16i8 MQPR:$Qm), (v16i8 MQPR:$Qn)))>;
}

class MVE_shift_with_imm<string iname, string suffix, dag oops, dag iops,
                         string ops, vpred_ops vpred, string cstr,
                         list<dag> pattern=[]>
  : MVE_p<oops, iops, NoItinerary, iname, suffix, ops, vpred, cstr, pattern> {
  bits<4> Qd;
  bits<4> Qm;

  let Inst{23} = 0b1;
  let Inst{22} = Qd{3};
  let Inst{15-13} = Qd{2-0};
  let Inst{12-11} = 0b00;
  let Inst{7-6} = 0b01;
  let Inst{5} = Qm{3};
  let Inst{4} = 0b1;
  let Inst{3-1} = Qm{2-0};
  let Inst{0} = 0b0;
  let validForTailPredication = 1;
}

class MVE_VSxI_imm<string iname, string suffix, bit bit_8, dag imm>
  : MVE_shift_with_imm<iname, suffix, (outs MQPR:$Qd),
                       !con((ins MQPR:$Qd_src, MQPR:$Qm), imm),
                       "$Qd, $Qm, $imm", vpred_n, "$Qd = $Qd_src"> {
  bits<6> imm;
  let Inst{28} = 0b1;
  let Inst{25-24} = 0b11;
  let Inst{21-16} = imm;
  let Inst{10-9} = 0b10;
  let Inst{8} = bit_8;
  let validForTailPredication = 1;
}

def MVE_VSRIimm8 : MVE_VSxI_imm<"vsri", "8", 0b0, (ins shr_imm8:$imm)> {
  let Inst{21-19} = 0b001;
}

def MVE_VSRIimm16 : MVE_VSxI_imm<"vsri", "16", 0b0, (ins shr_imm16:$imm)> {
  let Inst{21-20} = 0b01;
}

def MVE_VSRIimm32 : MVE_VSxI_imm<"vsri", "32", 0b0, (ins shr_imm32:$imm)> {
  let Inst{21} = 0b1;
}

def MVE_VSLIimm8 : MVE_VSxI_imm<"vsli", "8", 0b1, (ins imm0_7:$imm)> {
  let Inst{21-19} = 0b001;
}

def MVE_VSLIimm16 : MVE_VSxI_imm<"vsli", "16", 0b1, (ins imm0_15:$imm)> {
  let Inst{21-20} = 0b01;
}

def MVE_VSLIimm32 : MVE_VSxI_imm<"vsli", "32", 0b1,(ins imm0_31:$imm)> {
  let Inst{21} = 0b1;
}

class MVE_VQSHL_imm<string suffix, dag imm>
  : MVE_shift_with_imm<"vqshl", suffix, (outs MQPR:$Qd),
                       !con((ins MQPR:$Qm), imm), "$Qd, $Qm, $imm",
                       vpred_r, ""> {
  bits<6> imm;

  let Inst{25-24} = 0b11;
  let Inst{21-16} = imm;
  let Inst{10-8} = 0b111;
}

def MVE_VSLIimms8 : MVE_VQSHL_imm<"s8", (ins imm0_7:$imm)> {
  let Inst{28} = 0b0;
  let Inst{21-19} = 0b001;
}

def MVE_VSLIimmu8 : MVE_VQSHL_imm<"u8", (ins imm0_7:$imm)> {
  let Inst{28} = 0b1;
  let Inst{21-19} = 0b001;
}

def MVE_VSLIimms16 : MVE_VQSHL_imm<"s16", (ins imm0_15:$imm)> {
  let Inst{28} = 0b0;
  let Inst{21-20} = 0b01;
}

def MVE_VSLIimmu16 : MVE_VQSHL_imm<"u16", (ins imm0_15:$imm)> {
  let Inst{28} = 0b1;
  let Inst{21-20} = 0b01;
}

def MVE_VSLIimms32 : MVE_VQSHL_imm<"s32", (ins imm0_31:$imm)> {
  let Inst{28} = 0b0;
  let Inst{21} = 0b1;
}

def MVE_VSLIimmu32 : MVE_VQSHL_imm<"u32", (ins imm0_31:$imm)> {
  let Inst{28} = 0b1;
  let Inst{21} = 0b1;
}

class MVE_VQSHLU_imm<string suffix, dag imm>
  : MVE_shift_with_imm<"vqshlu", suffix, (outs MQPR:$Qd),
                       !con((ins MQPR:$Qm), imm), "$Qd, $Qm, $imm",
                       vpred_r, ""> {
  bits<6> imm;

  let Inst{28} = 0b1;
  let Inst{25-24} = 0b11;
  let Inst{21-16} = imm;
  let Inst{10-8} = 0b110;
}

def MVE_VQSHLU_imms8 : MVE_VQSHLU_imm<"s8", (ins imm0_7:$imm)> {
  let Inst{21-19} = 0b001;
}

def MVE_VQSHLU_imms16 : MVE_VQSHLU_imm<"s16", (ins imm0_15:$imm)> {
  let Inst{21-20} = 0b01;
}

def MVE_VQSHLU_imms32 : MVE_VQSHLU_imm<"s32", (ins imm0_31:$imm)> {
  let Inst{21} = 0b1;
}

class MVE_VRSHR_imm<string suffix, dag imm>
  : MVE_shift_with_imm<"vrshr", suffix, (outs MQPR:$Qd),
                       !con((ins MQPR:$Qm), imm), "$Qd, $Qm, $imm",
                       vpred_r, ""> {
  bits<6> imm;

  let Inst{25-24} = 0b11;
  let Inst{21-16} = imm;
  let Inst{10-8} = 0b010;
}

def MVE_VRSHR_imms8 : MVE_VRSHR_imm<"s8", (ins shr_imm8:$imm)> {
  let Inst{28} = 0b0;
  let Inst{21-19} = 0b001;
}

def MVE_VRSHR_immu8 : MVE_VRSHR_imm<"u8", (ins shr_imm8:$imm)> {
  let Inst{28} = 0b1;
  let Inst{21-19} = 0b001;
}

def MVE_VRSHR_imms16 : MVE_VRSHR_imm<"s16", (ins shr_imm16:$imm)> {
  let Inst{28} = 0b0;
  let Inst{21-20} = 0b01;
}

def MVE_VRSHR_immu16 : MVE_VRSHR_imm<"u16", (ins shr_imm16:$imm)> {
  let Inst{28} = 0b1;
  let Inst{21-20} = 0b01;
}

def MVE_VRSHR_imms32 : MVE_VRSHR_imm<"s32", (ins shr_imm32:$imm)> {
  let Inst{28} = 0b0;
  let Inst{21} = 0b1;
}

def MVE_VRSHR_immu32 : MVE_VRSHR_imm<"u32", (ins shr_imm32:$imm)> {
  let Inst{28} = 0b1;
  let Inst{21} = 0b1;
}

class MVE_VSHR_imm<string suffix, dag imm>
  : MVE_shift_with_imm<"vshr", suffix, (outs MQPR:$Qd),
                       !con((ins MQPR:$Qm), imm), "$Qd, $Qm, $imm",
                       vpred_r, ""> {
  bits<6> imm;

  let Inst{25-24} = 0b11;
  let Inst{21-16} = imm;
  let Inst{10-8} = 0b000;
}

def MVE_VSHR_imms8 : MVE_VSHR_imm<"s8", (ins shr_imm8:$imm)> {
  let Inst{28} = 0b0;
  let Inst{21-19} = 0b001;
}

def MVE_VSHR_immu8 : MVE_VSHR_imm<"u8", (ins shr_imm8:$imm)> {
  let Inst{28} = 0b1;
  let Inst{21-19} = 0b001;
}

def MVE_VSHR_imms16 : MVE_VSHR_imm<"s16", (ins shr_imm16:$imm)> {
  let Inst{28} = 0b0;
  let Inst{21-20} = 0b01;
}

def MVE_VSHR_immu16 : MVE_VSHR_imm<"u16", (ins shr_imm16:$imm)> {
  let Inst{28} = 0b1;
  let Inst{21-20} = 0b01;
}

def MVE_VSHR_imms32 : MVE_VSHR_imm<"s32", (ins shr_imm32:$imm)> {
  let Inst{28} = 0b0;
  let Inst{21} = 0b1;
}

def MVE_VSHR_immu32 : MVE_VSHR_imm<"u32", (ins shr_imm32:$imm)> {
  let Inst{28} = 0b1;
  let Inst{21} = 0b1;
}

class MVE_VSHL_imm<string suffix, dag imm>
  : MVE_shift_with_imm<"vshl", suffix, (outs MQPR:$Qd),
                       !con((ins MQPR:$Qm), imm), "$Qd, $Qm, $imm",
                       vpred_r, ""> {
  bits<6> imm;

  let Inst{28} = 0b0;
  let Inst{25-24} = 0b11;
  let Inst{21-16} = imm;
  let Inst{10-8} = 0b101;
}

def MVE_VSHL_immi8 : MVE_VSHL_imm<"i8", (ins imm0_7:$imm)> {
  let Inst{21-19} = 0b001;
}

def MVE_VSHL_immi16 : MVE_VSHL_imm<"i16", (ins imm0_15:$imm)> {
  let Inst{21-20} = 0b01;
}

def MVE_VSHL_immi32 : MVE_VSHL_imm<"i32", (ins imm0_31:$imm)> {
  let Inst{21} = 0b1;
}

let Predicates = [HasMVEInt] in {
  def : Pat<(v4i32 (ARMvshlImm (v4i32 MQPR:$src), imm0_31:$imm)),
            (v4i32 (MVE_VSHL_immi32 (v4i32 MQPR:$src), imm0_31:$imm))>;
  def : Pat<(v8i16 (ARMvshlImm (v8i16 MQPR:$src), imm0_15:$imm)),
            (v8i16 (MVE_VSHL_immi16 (v8i16 MQPR:$src), imm0_15:$imm))>;
  def : Pat<(v16i8 (ARMvshlImm (v16i8 MQPR:$src), imm0_7:$imm)),
            (v16i8 (MVE_VSHL_immi8 (v16i8 MQPR:$src), imm0_7:$imm))>;

  def : Pat<(v4i32 (ARMvshruImm (v4i32 MQPR:$src), imm0_31:$imm)),
            (v4i32 (MVE_VSHR_immu32 (v4i32 MQPR:$src), imm0_31:$imm))>;
  def : Pat<(v8i16 (ARMvshruImm (v8i16 MQPR:$src), imm0_15:$imm)),
            (v8i16 (MVE_VSHR_immu16 (v8i16 MQPR:$src), imm0_15:$imm))>;
  def : Pat<(v16i8 (ARMvshruImm (v16i8 MQPR:$src), imm0_7:$imm)),
            (v16i8 (MVE_VSHR_immu8 (v16i8 MQPR:$src), imm0_7:$imm))>;

  def : Pat<(v4i32 (ARMvshrsImm (v4i32 MQPR:$src), imm0_31:$imm)),
            (v4i32 (MVE_VSHR_imms32 (v4i32 MQPR:$src), imm0_31:$imm))>;
  def : Pat<(v8i16 (ARMvshrsImm (v8i16 MQPR:$src), imm0_15:$imm)),
            (v8i16 (MVE_VSHR_imms16 (v8i16 MQPR:$src), imm0_15:$imm))>;
  def : Pat<(v16i8 (ARMvshrsImm (v16i8 MQPR:$src), imm0_7:$imm)),
            (v16i8 (MVE_VSHR_imms8 (v16i8 MQPR:$src), imm0_7:$imm))>;
}

// end of mve_shift instructions

// start of MVE Floating Point instructions

class MVE_float<string iname, string suffix, dag oops, dag iops, string ops,
                vpred_ops vpred, string cstr, list<dag> pattern=[]>
  : MVE_f<oops, iops, NoItinerary, iname, suffix, ops, vpred, cstr, pattern> {
  bits<4> Qm;

  let Inst{12} = 0b0;
  let Inst{6} = 0b1;
  let Inst{5} = Qm{3};
  let Inst{3-1} = Qm{2-0};
  let Inst{0} = 0b0;
}

class MVE_VRINT<string rmode, bits<3> op, string suffix, bits<2> size,
                list<dag> pattern=[]>
  : MVE_float<!strconcat("vrint", rmode), suffix, (outs MQPR:$Qd),
              (ins MQPR:$Qm), "$Qd, $Qm", vpred_r, "", pattern> {
  bits<4> Qd;

  let Inst{28} = 0b1;
  let Inst{25-23} = 0b111;
  let Inst{22} = Qd{3};
  let Inst{21-20} = 0b11;
  let Inst{19-18} = size;
  let Inst{17-16} = 0b10;
  let Inst{15-13} = Qd{2-0};
  let Inst{11-10} = 0b01;
  let Inst{9-7} = op{2-0};
  let Inst{4} = 0b0;
  let validForTailPredication = 1;

}

multiclass MVE_VRINT_ops<string suffix, bits<2> size, list<dag> pattern=[]> {
  def N : MVE_VRINT<"n", 0b000, suffix, size, pattern>;
  def X : MVE_VRINT<"x", 0b001, suffix, size, pattern>;
  def A : MVE_VRINT<"a", 0b010, suffix, size, pattern>;
  def Z : MVE_VRINT<"z", 0b011, suffix, size, pattern>;
  def M : MVE_VRINT<"m", 0b101, suffix, size, pattern>;
  def P : MVE_VRINT<"p", 0b111, suffix, size, pattern>;
}

defm MVE_VRINTf16 : MVE_VRINT_ops<"f16", 0b01>;
defm MVE_VRINTf32 : MVE_VRINT_ops<"f32", 0b10>;

let Predicates = [HasMVEFloat] in {
  def : Pat<(v4f32 (frint (v4f32 MQPR:$val1))),
            (v4f32 (MVE_VRINTf32X (v4f32 MQPR:$val1)))>;
  def : Pat<(v8f16 (frint (v8f16 MQPR:$val1))),
            (v8f16 (MVE_VRINTf16X (v8f16 MQPR:$val1)))>;
  def : Pat<(v4f32 (fround (v4f32 MQPR:$val1))),
            (v4f32 (MVE_VRINTf32A (v4f32 MQPR:$val1)))>;
  def : Pat<(v8f16 (fround (v8f16 MQPR:$val1))),
            (v8f16 (MVE_VRINTf16A (v8f16 MQPR:$val1)))>;
  def : Pat<(v4f32 (ftrunc (v4f32 MQPR:$val1))),
            (v4f32 (MVE_VRINTf32Z (v4f32 MQPR:$val1)))>;
  def : Pat<(v8f16 (ftrunc (v8f16 MQPR:$val1))),
            (v8f16 (MVE_VRINTf16Z (v8f16 MQPR:$val1)))>;
  def : Pat<(v4f32 (ffloor (v4f32 MQPR:$val1))),
            (v4f32 (MVE_VRINTf32M (v4f32 MQPR:$val1)))>;
  def : Pat<(v8f16 (ffloor (v8f16 MQPR:$val1))),
            (v8f16 (MVE_VRINTf16M (v8f16 MQPR:$val1)))>;
  def : Pat<(v4f32 (fceil (v4f32 MQPR:$val1))),
            (v4f32 (MVE_VRINTf32P (v4f32 MQPR:$val1)))>;
  def : Pat<(v8f16 (fceil (v8f16 MQPR:$val1))),
            (v8f16 (MVE_VRINTf16P (v8f16 MQPR:$val1)))>;
}

class MVEFloatArithNeon<string iname, string suffix, bit size,
                           dag oops, dag iops, string ops,
                           vpred_ops vpred, string cstr, list<dag> pattern=[]>
  : MVE_float<iname, suffix, oops, iops, ops, vpred, cstr, pattern> {
  let Inst{20} = size;
  let Inst{16} = 0b0;
}

class MVE_VMUL_fp<string iname, string suffix, bit size, list<dag> pattern=[]>
  : MVEFloatArithNeon<iname, suffix, size, (outs MQPR:$Qd),
                      (ins MQPR:$Qn, MQPR:$Qm), "$Qd, $Qn, $Qm", vpred_r, "",
                      pattern> {
  bits<4> Qd;
  bits<4> Qn;

  let Inst{28} = 0b1;
  let Inst{25-23} = 0b110;
  let Inst{22} = Qd{3};
  let Inst{21} = 0b0;
  let Inst{19-17} = Qn{2-0};
  let Inst{15-13} = Qd{2-0};
  let Inst{12-8} = 0b01101;
  let Inst{7} = Qn{3};
  let Inst{4} = 0b1;
  let validForTailPredication = 1;
}

multiclass MVE_VMULT_fp_m<string iname, bit bit_21, MVEVectorVTInfo VTI,
                            SDNode unpred_op, Intrinsic pred_int> {
  def "" : MVE_VMUL_fp<iname, VTI.Suffix, VTI.Size{0}>;

  let Predicates = [HasMVEFloat] in {
    def : Pat<(VTI.Vec (unpred_op (VTI.Vec MQPR:$Qm), (VTI.Vec MQPR:$Qn))),
              (VTI.Vec (!cast<Instruction>(NAME)
                            (VTI.Vec MQPR:$Qm), (VTI.Vec MQPR:$Qn)))>;
    def : Pat<(VTI.Vec (pred_int (VTI.Vec MQPR:$Qm), (VTI.Vec MQPR:$Qn),
                            (VTI.Pred VCCR:$mask), (VTI.Vec MQPR:$inactive))),
              (VTI.Vec (!cast<Instruction>(NAME)
                            (VTI.Vec MQPR:$Qm), (VTI.Vec MQPR:$Qn),
                            (i32 1), (VTI.Pred VCCR:$mask),
                            (VTI.Vec MQPR:$inactive)))>;
  }
}

multiclass MVE_VMUL_fp_m<MVEVectorVTInfo VTI>
  : MVE_VMULT_fp_m<"vmul", 0, VTI, fmul, int_arm_mve_mul_predicated>;

defm MVE_VMULf32 : MVE_VMUL_fp_m<MVE_v4f32>;
defm MVE_VMULf16 : MVE_VMUL_fp_m<MVE_v8f16>;

class MVE_VCMLA<string suffix, bit size, list<dag> pattern=[]>
  : MVEFloatArithNeon<"vcmla", suffix, size, (outs MQPR:$Qd),
                         (ins MQPR:$Qd_src, MQPR:$Qn, MQPR:$Qm, complexrotateop:$rot),
                         "$Qd, $Qn, $Qm, $rot", vpred_n, "$Qd = $Qd_src", pattern> {
  bits<4> Qd;
  bits<4> Qn;
  bits<2> rot;

  let Inst{28} = 0b1;
  let Inst{25} = 0b0;
  let Inst{24-23} = rot;
  let Inst{22} = Qd{3};
  let Inst{21} = 0b1;
  let Inst{19-17} = Qn{2-0};
  let Inst{15-13} = Qd{2-0};
  let Inst{12-8} = 0b01000;
  let Inst{7} = Qn{3};
  let Inst{4} = 0b0;
}

def MVE_VCMLAf16 : MVE_VCMLA<"f16", 0b0>;
def MVE_VCMLAf32 : MVE_VCMLA<"f32", 0b1>;

class MVE_VADDSUBFMA_fp<string iname, string suffix, bit size, bit bit_4,
                        bit bit_8, bit bit_21, dag iops=(ins),
                        vpred_ops vpred=vpred_r, string cstr="",
                        list<dag> pattern=[]>
  : MVEFloatArithNeon<iname, suffix, size, (outs MQPR:$Qd),
                      !con(iops, (ins MQPR:$Qn, MQPR:$Qm)), "$Qd, $Qn, $Qm",
                      vpred, cstr, pattern> {
  bits<4> Qd;
  bits<4> Qn;

  let Inst{28} = 0b0;
  let Inst{25-23} = 0b110;
  let Inst{22} = Qd{3};
  let Inst{21} = bit_21;
  let Inst{19-17} = Qn{2-0};
  let Inst{15-13} = Qd{2-0};
  let Inst{11-9} = 0b110;
  let Inst{8} = bit_8;
  let Inst{7} = Qn{3};
  let Inst{4} = bit_4;
}

def MVE_VFMAf32 : MVE_VADDSUBFMA_fp<"vfma", "f32", 0b0, 0b1, 0b0, 0b0,
    (ins MQPR:$Qd_src), vpred_n, "$Qd = $Qd_src">;
def MVE_VFMAf16 : MVE_VADDSUBFMA_fp<"vfma", "f16", 0b1, 0b1, 0b0, 0b0,
    (ins MQPR:$Qd_src), vpred_n, "$Qd = $Qd_src">;

def MVE_VFMSf32 : MVE_VADDSUBFMA_fp<"vfms", "f32", 0b0, 0b1, 0b0, 0b1,
    (ins MQPR:$Qd_src), vpred_n, "$Qd = $Qd_src">;
def MVE_VFMSf16 : MVE_VADDSUBFMA_fp<"vfms", "f16", 0b1, 0b1, 0b0, 0b1,
    (ins MQPR:$Qd_src), vpred_n, "$Qd = $Qd_src">;

let Predicates = [HasMVEFloat] in {
  def : Pat<(v8f16 (fma (v8f16 MQPR:$src1), (v8f16 MQPR:$src2), (v8f16 MQPR:$src3))),
            (v8f16 (MVE_VFMAf16 $src3, $src1, $src2))>;
  def : Pat<(v4f32 (fma (v4f32 MQPR:$src1), (v4f32 MQPR:$src2), (v4f32 MQPR:$src3))),
            (v4f32 (MVE_VFMAf32 $src3, $src1, $src2))>;
  def : Pat<(v8f16 (fma (fneg (v8f16 MQPR:$src1)), (v8f16 MQPR:$src2), (v8f16 MQPR:$src3))),
            (v8f16 (MVE_VFMSf16 $src3, $src1, $src2))>;
  def : Pat<(v4f32 (fma (fneg (v4f32 MQPR:$src1)), (v4f32 MQPR:$src2), (v4f32 MQPR:$src3))),
            (v4f32 (MVE_VFMSf32 $src3, $src1, $src2))>;
}

multiclass MVE_VADDSUB_fp_m<string iname, bit bit_21, MVEVectorVTInfo VTI,
                            SDNode unpred_op, Intrinsic pred_int> {
  def "" : MVE_VADDSUBFMA_fp<iname, VTI.Suffix, VTI.Size{0}, 0, 1, bit_21> {
    let validForTailPredication = 1;
  }

  let Predicates = [HasMVEFloat] in {
    def : Pat<(VTI.Vec (unpred_op (VTI.Vec MQPR:$Qm), (VTI.Vec MQPR:$Qn))),
              (VTI.Vec (!cast<Instruction>(NAME)
                            (VTI.Vec MQPR:$Qm), (VTI.Vec MQPR:$Qn)))>;
    def : Pat<(VTI.Vec (pred_int (VTI.Vec MQPR:$Qm), (VTI.Vec MQPR:$Qn),
                            (VTI.Pred VCCR:$mask), (VTI.Vec MQPR:$inactive))),
              (VTI.Vec (!cast<Instruction>(NAME)
                            (VTI.Vec MQPR:$Qm), (VTI.Vec MQPR:$Qn),
                            (i32 1), (VTI.Pred VCCR:$mask),
                            (VTI.Vec MQPR:$inactive)))>;
  }
}

multiclass MVE_VADD_fp_m<MVEVectorVTInfo VTI>
  : MVE_VADDSUB_fp_m<"vadd", 0, VTI, fadd, int_arm_mve_add_predicated>;
multiclass MVE_VSUB_fp_m<MVEVectorVTInfo VTI>
  : MVE_VADDSUB_fp_m<"vsub", 1, VTI, fsub, int_arm_mve_sub_predicated>;

defm MVE_VADDf32 : MVE_VADD_fp_m<MVE_v4f32>;
defm MVE_VADDf16 : MVE_VADD_fp_m<MVE_v8f16>;

defm MVE_VSUBf32 : MVE_VSUB_fp_m<MVE_v4f32>;
defm MVE_VSUBf16 : MVE_VSUB_fp_m<MVE_v8f16>;

class MVE_VCADD<string suffix, bit size, string cstr="", list<dag> pattern=[]>
  : MVEFloatArithNeon<"vcadd", suffix, size, (outs MQPR:$Qd),
                         (ins MQPR:$Qn, MQPR:$Qm, complexrotateopodd:$rot),
                         "$Qd, $Qn, $Qm, $rot", vpred_r, cstr, pattern> {
  bits<4> Qd;
  bits<4> Qn;
  bit rot;

  let Inst{28} = 0b1;
  let Inst{25} = 0b0;
  let Inst{24} = rot;
  let Inst{23} = 0b1;
  let Inst{22} = Qd{3};
  let Inst{21} = 0b0;
  let Inst{19-17} = Qn{2-0};
  let Inst{15-13} = Qd{2-0};
  let Inst{12-8} = 0b01000;
  let Inst{7} = Qn{3};
  let Inst{4} = 0b0;
}

def MVE_VCADDf16 : MVE_VCADD<"f16", 0b0>;
def MVE_VCADDf32 : MVE_VCADD<"f32", 0b1, "@earlyclobber $Qd">;

class MVE_VABD_fp<string suffix, bit size>
  : MVE_float<"vabd", suffix, (outs MQPR:$Qd), (ins MQPR:$Qn, MQPR:$Qm),
              "$Qd, $Qn, $Qm", vpred_r, ""> {
  bits<4> Qd;
  bits<4> Qn;

  let Inst{28} = 0b1;
  let Inst{25-23} = 0b110;
  let Inst{22} = Qd{3};
  let Inst{21} = 0b1;
  let Inst{20} = size;
  let Inst{19-17} = Qn{2-0};
  let Inst{16} = 0b0;
  let Inst{15-13} = Qd{2-0};
  let Inst{11-8} = 0b1101;
  let Inst{7} = Qn{3};
  let Inst{4} = 0b0;
  let validForTailPredication = 1;
}

multiclass MVE_VABDT_fp_m<MVEVectorVTInfo VTI,
                            Intrinsic unpred_int, Intrinsic pred_int> {
  def "" : MVE_VABD_fp<VTI.Suffix, VTI.Size{0}>;

  let Predicates = [HasMVEFloat] in {
    def : Pat<(VTI.Vec (unpred_int (VTI.Vec MQPR:$Qm), (VTI.Vec MQPR:$Qn))),
              (VTI.Vec (!cast<Instruction>(NAME)
                            (VTI.Vec MQPR:$Qm), (VTI.Vec MQPR:$Qn)))>;
    def : Pat<(VTI.Vec (pred_int (VTI.Vec MQPR:$Qm), (VTI.Vec MQPR:$Qn),
                            (VTI.Pred VCCR:$mask), (VTI.Vec MQPR:$inactive))),
              (VTI.Vec (!cast<Instruction>(NAME)
                            (VTI.Vec MQPR:$Qm), (VTI.Vec MQPR:$Qn),
                            (i32 1), (VTI.Pred VCCR:$mask),
                            (VTI.Vec MQPR:$inactive)))>;
  }
}

multiclass MVE_VABD_fp_m<MVEVectorVTInfo VTI>
  : MVE_VABDT_fp_m<VTI, int_arm_mve_vabd, int_arm_mve_abd_predicated>;

defm MVE_VABDf32 : MVE_VABD_fp_m<MVE_v4f32>;
defm MVE_VABDf16 : MVE_VABD_fp_m<MVE_v8f16>;

class MVE_VCVT_fix<string suffix, bit fsi, bit U, bit op,
                   Operand imm_operand_type, list<dag> pattern=[]>
  : MVE_float<"vcvt", suffix,
              (outs MQPR:$Qd), (ins MQPR:$Qm, imm_operand_type:$imm6),
              "$Qd, $Qm, $imm6", vpred_r, "", pattern> {
  bits<4> Qd;
  bits<6> imm6;

  let Inst{28} = U;
  let Inst{25-23} = 0b111;
  let Inst{22} = Qd{3};
  let Inst{21} = 0b1;
  let Inst{19-16} = imm6{3-0};
  let Inst{15-13} = Qd{2-0};
  let Inst{11-10} = 0b11;
  let Inst{9} = fsi;
  let Inst{8} = op;
  let Inst{7} = 0b0;
  let Inst{4} = 0b1;

  let DecoderMethod = "DecodeMVEVCVTt1fp";
  let validForTailPredication = 1;
}

class MVE_VCVT_imm_asmop<int Bits> : AsmOperandClass {
  let PredicateMethod = "isImmediate<1," # Bits # ">";
  let DiagnosticString =
      "MVE fixed-point immediate operand must be between 1 and " # Bits;
  let Name = "MVEVcvtImm" # Bits;
  let RenderMethod = "addImmOperands";
}
class MVE_VCVT_imm<int Bits>: Operand<i32> {
  let ParserMatchClass = MVE_VCVT_imm_asmop<Bits>;
  let EncoderMethod = "getNEONVcvtImm32OpValue";
  let DecoderMethod = "DecodeVCVTImmOperand";
}

class MVE_VCVT_fix_f32<string suffix, bit U, bit op>
    : MVE_VCVT_fix<suffix, 0b1, U, op, MVE_VCVT_imm<32>> {
  let Inst{20} = imm6{4};
}
class MVE_VCVT_fix_f16<string suffix, bit U, bit op>
    : MVE_VCVT_fix<suffix, 0b0, U, op, MVE_VCVT_imm<16>> {
  let Inst{20} = 0b1;
}

def MVE_VCVTf16s16_fix : MVE_VCVT_fix_f16<"f16.s16", 0b0, 0b0>;
def MVE_VCVTs16f16_fix : MVE_VCVT_fix_f16<"s16.f16", 0b0, 0b1>;
def MVE_VCVTf16u16_fix : MVE_VCVT_fix_f16<"f16.u16", 0b1, 0b0>;
def MVE_VCVTu16f16_fix : MVE_VCVT_fix_f16<"u16.f16", 0b1, 0b1>;
def MVE_VCVTf32s32_fix : MVE_VCVT_fix_f32<"f32.s32", 0b0, 0b0>;
def MVE_VCVTs32f32_fix : MVE_VCVT_fix_f32<"s32.f32", 0b0, 0b1>;
def MVE_VCVTf32u32_fix : MVE_VCVT_fix_f32<"f32.u32", 0b1, 0b0>;
def MVE_VCVTu32f32_fix : MVE_VCVT_fix_f32<"u32.f32", 0b1, 0b1>;

class MVE_VCVT_fp_int_anpm<string suffix, bits<2> size, bit op, string anpm,
                bits<2> rm, list<dag> pattern=[]>
  : MVE_float<!strconcat("vcvt", anpm), suffix, (outs MQPR:$Qd),
              (ins MQPR:$Qm), "$Qd, $Qm", vpred_r, "", pattern> {
  bits<4> Qd;

  let Inst{28} = 0b1;
  let Inst{25-23} = 0b111;
  let Inst{22} = Qd{3};
  let Inst{21-20} = 0b11;
  let Inst{19-18} = size;
  let Inst{17-16} = 0b11;
  let Inst{15-13} = Qd{2-0};
  let Inst{12-10} = 0b000;
  let Inst{9-8} = rm;
  let Inst{7} = op;
  let Inst{4} = 0b0;
  let validForTailPredication = 1;
}

multiclass MVE_VCVT_fp_int_anpm_multi<string suffix, bits<2> size, bit op,
                                list<dag> pattern=[]> {
  def a : MVE_VCVT_fp_int_anpm<suffix, size, op, "a", 0b00>;
  def n : MVE_VCVT_fp_int_anpm<suffix, size, op, "n", 0b01>;
  def p : MVE_VCVT_fp_int_anpm<suffix, size, op, "p", 0b10>;
  def m : MVE_VCVT_fp_int_anpm<suffix, size, op, "m", 0b11>;
}

// This defines instructions such as MVE_VCVTu16f16a, with an explicit
// rounding-mode suffix on the mnemonic. The class below will define
// the bare MVE_VCVTu16f16 (with implied rounding toward zero).
defm MVE_VCVTs16f16 : MVE_VCVT_fp_int_anpm_multi<"s16.f16", 0b01, 0b0>;
defm MVE_VCVTu16f16 : MVE_VCVT_fp_int_anpm_multi<"u16.f16", 0b01, 0b1>;
defm MVE_VCVTs32f32 : MVE_VCVT_fp_int_anpm_multi<"s32.f32", 0b10, 0b0>;
defm MVE_VCVTu32f32 : MVE_VCVT_fp_int_anpm_multi<"u32.f32", 0b10, 0b1>;

class MVE_VCVT_fp_int<string suffix, bits<2> size, bits<2> op,
                      list<dag> pattern=[]>
  : MVE_float<"vcvt", suffix, (outs MQPR:$Qd),
              (ins MQPR:$Qm), "$Qd, $Qm", vpred_r, "", pattern> {
  bits<4> Qd;

  let Inst{28} = 0b1;
  let Inst{25-23} = 0b111;
  let Inst{22} = Qd{3};
  let Inst{21-20} = 0b11;
  let Inst{19-18} = size;
  let Inst{17-16} = 0b11;
  let Inst{15-13} = Qd{2-0};
  let Inst{12-9} = 0b0011;
  let Inst{8-7} = op;
  let Inst{4} = 0b0;
  let validForTailPredication = 1;
}

// The unsuffixed VCVT for float->int implicitly rounds toward zero,
// which I reflect here in the llvm instruction names
def MVE_VCVTs16f16z : MVE_VCVT_fp_int<"s16.f16", 0b01, 0b10>;
def MVE_VCVTu16f16z : MVE_VCVT_fp_int<"u16.f16", 0b01, 0b11>;
def MVE_VCVTs32f32z : MVE_VCVT_fp_int<"s32.f32", 0b10, 0b10>;
def MVE_VCVTu32f32z : MVE_VCVT_fp_int<"u32.f32", 0b10, 0b11>;
// Whereas VCVT for int->float rounds to nearest
def MVE_VCVTf16s16n : MVE_VCVT_fp_int<"f16.s16", 0b01, 0b00>;
def MVE_VCVTf16u16n : MVE_VCVT_fp_int<"f16.u16", 0b01, 0b01>;
def MVE_VCVTf32s32n : MVE_VCVT_fp_int<"f32.s32", 0b10, 0b00>;
def MVE_VCVTf32u32n : MVE_VCVT_fp_int<"f32.u32", 0b10, 0b01>;

let Predicates = [HasMVEFloat] in {
  def : Pat<(v4i32 (fp_to_sint (v4f32 MQPR:$src))),
            (v4i32 (MVE_VCVTs32f32z (v4f32 MQPR:$src)))>;
  def : Pat<(v4i32 (fp_to_uint (v4f32 MQPR:$src))),
            (v4i32 (MVE_VCVTu32f32z (v4f32 MQPR:$src)))>;
  def : Pat<(v8i16 (fp_to_sint (v8f16 MQPR:$src))),
            (v8i16 (MVE_VCVTs16f16z (v8f16 MQPR:$src)))>;
  def : Pat<(v8i16 (fp_to_uint (v8f16 MQPR:$src))),
            (v8i16 (MVE_VCVTu16f16z (v8f16 MQPR:$src)))>;
  def : Pat<(v4f32 (sint_to_fp (v4i32 MQPR:$src))),
            (v4f32 (MVE_VCVTf32s32n (v4i32 MQPR:$src)))>;
  def : Pat<(v4f32 (uint_to_fp (v4i32 MQPR:$src))),
            (v4f32 (MVE_VCVTf32u32n (v4i32 MQPR:$src)))>;
  def : Pat<(v8f16 (sint_to_fp (v8i16 MQPR:$src))),
            (v8f16 (MVE_VCVTf16s16n (v8i16 MQPR:$src)))>;
  def : Pat<(v8f16 (uint_to_fp (v8i16 MQPR:$src))),
            (v8f16 (MVE_VCVTf16u16n (v8i16 MQPR:$src)))>;
}

class MVE_VABSNEG_fp<string iname, string suffix, bits<2> size, bit negate,
                   list<dag> pattern=[]>
  : MVE_float<iname, suffix, (outs MQPR:$Qd),
              (ins MQPR:$Qm), "$Qd, $Qm", vpred_r, "", pattern> {
  bits<4> Qd;

  let Inst{28} = 0b1;
  let Inst{25-23} = 0b111;
  let Inst{22} = Qd{3};
  let Inst{21-20} = 0b11;
  let Inst{19-18} = size;
  let Inst{17-16} = 0b01;
  let Inst{15-13} = Qd{2-0};
  let Inst{11-8} = 0b0111;
  let Inst{7} = negate;
  let Inst{4} = 0b0;
  let validForTailPredication = 1;
}

def MVE_VABSf16 : MVE_VABSNEG_fp<"vabs", "f16", 0b01, 0b0>;
def MVE_VABSf32 : MVE_VABSNEG_fp<"vabs", "f32", 0b10, 0b0>;

let Predicates = [HasMVEFloat] in {
  def : Pat<(v8f16 (fabs MQPR:$src)),
            (MVE_VABSf16 MQPR:$src)>;
  def : Pat<(v4f32 (fabs MQPR:$src)),
            (MVE_VABSf32 MQPR:$src)>;
}

def MVE_VNEGf16 : MVE_VABSNEG_fp<"vneg", "f16", 0b01, 0b1>;
def MVE_VNEGf32 : MVE_VABSNEG_fp<"vneg", "f32", 0b10, 0b1>;

let Predicates = [HasMVEFloat] in {
  def : Pat<(v8f16 (fneg MQPR:$src)),
            (MVE_VNEGf16 MQPR:$src)>;
  def : Pat<(v4f32 (fneg MQPR:$src)),
            (MVE_VNEGf32 MQPR:$src)>;
}

class MVE_VMAXMINNMA<string iname, string suffix, bit size, bit bit_12,
                     list<dag> pattern=[]>
  : MVE_f<(outs MQPR:$Qd), (ins MQPR:$Qd_src, MQPR:$Qm),
          NoItinerary, iname, suffix, "$Qd, $Qm", vpred_n, "$Qd = $Qd_src",
          pattern> {
  bits<4> Qd;
  bits<4> Qm;

  let Inst{28} = size;
  let Inst{25-23} = 0b100;
  let Inst{22} = Qd{3};
  let Inst{21-16} = 0b111111;
  let Inst{15-13} = Qd{2-0};
  let Inst{12} = bit_12;
  let Inst{11-6} = 0b111010;
  let Inst{5} = Qm{3};
  let Inst{4} = 0b0;
  let Inst{3-1} = Qm{2-0};
  let Inst{0} = 0b1;
}

def MVE_VMAXNMAf32 : MVE_VMAXMINNMA<"vmaxnma", "f32", 0b0, 0b0>;
def MVE_VMAXNMAf16 : MVE_VMAXMINNMA<"vmaxnma", "f16", 0b1, 0b0>;

def MVE_VMINNMAf32 : MVE_VMAXMINNMA<"vminnma", "f32", 0b0, 0b1>;
def MVE_VMINNMAf16 : MVE_VMAXMINNMA<"vminnma", "f16", 0b1, 0b1>;

// end of MVE Floating Point instructions

// start of MVE compares

class MVE_VCMPqq<string suffix, bit bit_28, bits<2> bits_21_20,
                 VCMPPredicateOperand predtype, list<dag> pattern=[]>
  : MVE_p<(outs VCCR:$P0), (ins MQPR:$Qn, MQPR:$Qm, predtype:$fc),
           NoItinerary, "vcmp", suffix, "$fc, $Qn, $Qm", vpred_n, "", pattern> {
  // Base class for comparing two vector registers
  bits<3> fc;
  bits<4> Qn;
  bits<4> Qm;

  let Inst{28} = bit_28;
  let Inst{25-22} = 0b1000;
  let Inst{21-20} = bits_21_20;
  let Inst{19-17} = Qn{2-0};
  let Inst{16-13} = 0b1000;
  let Inst{12} = fc{2};
  let Inst{11-8} = 0b1111;
  let Inst{7} = fc{0};
  let Inst{6} = 0b0;
  let Inst{5} = Qm{3};
  let Inst{4} = 0b0;
  let Inst{3-1} = Qm{2-0};
  let Inst{0} = fc{1};

  let Constraints = "";

  // We need a custom decoder method for these instructions because of
  // the output VCCR operand, which isn't encoded in the instruction
  // bits anywhere (there is only one choice for it) but has to be
  // included in the MC operands so that codegen will be able to track
  // its data flow between instructions, spill/reload it when
  // necessary, etc. There seems to be no way to get the Tablegen
  // decoder to emit an operand that isn't affected by any instruction
  // bit.
  let DecoderMethod = "DecodeMVEVCMP<false," # predtype.DecoderMethod # ">";
  let validForTailPredication = 1;
}

class MVE_VCMPqqf<string suffix, bit size>
    : MVE_VCMPqq<suffix, size, 0b11, pred_basic_fp> {
  let Predicates = [HasMVEFloat];
}

class MVE_VCMPqqi<string suffix, bits<2> size>
    : MVE_VCMPqq<suffix, 0b1, size, pred_basic_i> {
  let Inst{12} = 0b0;
  let Inst{0} = 0b0;
}

class MVE_VCMPqqu<string suffix, bits<2> size>
    : MVE_VCMPqq<suffix, 0b1, size, pred_basic_u> {
  let Inst{12} = 0b0;
  let Inst{0} = 0b1;
}

class MVE_VCMPqqs<string suffix, bits<2> size>
    : MVE_VCMPqq<suffix, 0b1, size, pred_basic_s> {
  let Inst{12} = 0b1;
}

def MVE_VCMPf32 : MVE_VCMPqqf<"f32", 0b0>;
def MVE_VCMPf16 : MVE_VCMPqqf<"f16", 0b1>;

def MVE_VCMPi8  : MVE_VCMPqqi<"i8",  0b00>;
def MVE_VCMPi16 : MVE_VCMPqqi<"i16", 0b01>;
def MVE_VCMPi32 : MVE_VCMPqqi<"i32", 0b10>;

def MVE_VCMPu8  : MVE_VCMPqqu<"u8",  0b00>;
def MVE_VCMPu16 : MVE_VCMPqqu<"u16", 0b01>;
def MVE_VCMPu32 : MVE_VCMPqqu<"u32", 0b10>;

def MVE_VCMPs8  : MVE_VCMPqqs<"s8",  0b00>;
def MVE_VCMPs16 : MVE_VCMPqqs<"s16", 0b01>;
def MVE_VCMPs32 : MVE_VCMPqqs<"s32", 0b10>;

class MVE_VCMPqr<string suffix, bit bit_28, bits<2> bits_21_20,
                 VCMPPredicateOperand predtype, list<dag> pattern=[]>
  : MVE_p<(outs VCCR:$P0), (ins MQPR:$Qn, GPRwithZR:$Rm, predtype:$fc),
           NoItinerary, "vcmp", suffix, "$fc, $Qn, $Rm", vpred_n, "", pattern> {
  // Base class for comparing a vector register with a scalar
  bits<3> fc;
  bits<4> Qn;
  bits<4> Rm;

  let Inst{28} = bit_28;
  let Inst{25-22} = 0b1000;
  let Inst{21-20} = bits_21_20;
  let Inst{19-17} = Qn{2-0};
  let Inst{16-13} = 0b1000;
  let Inst{12} = fc{2};
  let Inst{11-8} = 0b1111;
  let Inst{7} = fc{0};
  let Inst{6} = 0b1;
  let Inst{5} = fc{1};
  let Inst{4} = 0b0;
  let Inst{3-0} = Rm{3-0};

  let Constraints = "";
  // Custom decoder method, for the same reason as MVE_VCMPqq
  let DecoderMethod = "DecodeMVEVCMP<true," # predtype.DecoderMethod # ">";
  let validForTailPredication = 1;
}

class MVE_VCMPqrf<string suffix, bit size>
    : MVE_VCMPqr<suffix, size, 0b11, pred_basic_fp> {
  let Predicates = [HasMVEFloat];
}

class MVE_VCMPqri<string suffix, bits<2> size>
    : MVE_VCMPqr<suffix, 0b1, size, pred_basic_i> {
  let Inst{12} = 0b0;
  let Inst{5} = 0b0;
}

class MVE_VCMPqru<string suffix, bits<2> size>
    : MVE_VCMPqr<suffix, 0b1, size, pred_basic_u> {
  let Inst{12} = 0b0;
  let Inst{5} = 0b1;
}

class MVE_VCMPqrs<string suffix, bits<2> size>
    : MVE_VCMPqr<suffix, 0b1, size, pred_basic_s> {
  let Inst{12} = 0b1;
}

def MVE_VCMPf32r : MVE_VCMPqrf<"f32", 0b0>;
def MVE_VCMPf16r : MVE_VCMPqrf<"f16", 0b1>;

def MVE_VCMPi8r  : MVE_VCMPqri<"i8",  0b00>;
def MVE_VCMPi16r : MVE_VCMPqri<"i16", 0b01>;
def MVE_VCMPi32r : MVE_VCMPqri<"i32", 0b10>;

def MVE_VCMPu8r  : MVE_VCMPqru<"u8",  0b00>;
def MVE_VCMPu16r : MVE_VCMPqru<"u16", 0b01>;
def MVE_VCMPu32r : MVE_VCMPqru<"u32", 0b10>;

def MVE_VCMPs8r  : MVE_VCMPqrs<"s8",  0b00>;
def MVE_VCMPs16r : MVE_VCMPqrs<"s16", 0b01>;
def MVE_VCMPs32r : MVE_VCMPqrs<"s32", 0b10>;

multiclass unpred_vcmp_z<string suffix, int fc> {
  def i8  : Pat<(v16i1 (ARMvcmpz (v16i8 MQPR:$v1), (i32 fc))),
                (v16i1 (!cast<Instruction>("MVE_VCMP"#suffix#"8r") (v16i8 MQPR:$v1), ZR, fc))>;
  def i16 : Pat<(v8i1 (ARMvcmpz (v8i16 MQPR:$v1), (i32 fc))),
                (v8i1 (!cast<Instruction>("MVE_VCMP"#suffix#"16r") (v8i16 MQPR:$v1), ZR, fc))>;
  def i32 : Pat<(v4i1 (ARMvcmpz (v4i32 MQPR:$v1), (i32 fc))),
                (v4i1 (!cast<Instruction>("MVE_VCMP"#suffix#"32r") (v4i32 MQPR:$v1), ZR, fc))>;

  def : Pat<(v16i1 (and (v16i1 VCCR:$p1), (v16i1 (ARMvcmpz (v16i8 MQPR:$v1), (i32 fc))))),
            (v16i1 (!cast<Instruction>("MVE_VCMP"#suffix#"8r") (v16i8 MQPR:$v1), ZR, fc, 1, VCCR:$p1))>;
  def : Pat<(v8i1 (and (v8i1 VCCR:$p1), (v8i1 (ARMvcmpz (v8i16 MQPR:$v1), (i32 fc))))),
            (v8i1 (!cast<Instruction>("MVE_VCMP"#suffix#"16r") (v8i16 MQPR:$v1), ZR, fc, 1, VCCR:$p1))>;
  def : Pat<(v4i1 (and (v4i1 VCCR:$p1), (v4i1 (ARMvcmpz (v4i32 MQPR:$v1), (i32 fc))))),
            (v4i1 (!cast<Instruction>("MVE_VCMP"#suffix#"32r") (v4i32 MQPR:$v1), ZR, fc, 1, VCCR:$p1))>;
}

multiclass unpred_vcmp_r<string suffix, int fc> {
  def i8  : Pat<(v16i1 (ARMvcmp (v16i8 MQPR:$v1), (v16i8 MQPR:$v2), (i32 fc))),
                (v16i1 (!cast<Instruction>("MVE_VCMP"#suffix#"8") (v16i8 MQPR:$v1), (v16i8 MQPR:$v2), fc))>;
  def i16 : Pat<(v8i1 (ARMvcmp (v8i16 MQPR:$v1), (v8i16 MQPR:$v2), (i32 fc))),
                (v8i1 (!cast<Instruction>("MVE_VCMP"#suffix#"16") (v8i16 MQPR:$v1), (v8i16 MQPR:$v2), fc))>;
  def i32 : Pat<(v4i1 (ARMvcmp (v4i32 MQPR:$v1), (v4i32 MQPR:$v2), (i32 fc))),
                (v4i1 (!cast<Instruction>("MVE_VCMP"#suffix#"32") (v4i32 MQPR:$v1), (v4i32 MQPR:$v2), fc))>;

  def i8r  : Pat<(v16i1 (ARMvcmp (v16i8 MQPR:$v1), (v16i8 (ARMvdup GPR:$v2)), (i32 fc))),
                 (v16i1 (!cast<Instruction>("MVE_VCMP"#suffix#"8r") (v16i8 MQPR:$v1), (i32 GPR:$v2), fc))>;
  def i16r : Pat<(v8i1 (ARMvcmp (v8i16 MQPR:$v1), (v8i16 (ARMvdup GPR:$v2)), (i32 fc))),
                 (v8i1 (!cast<Instruction>("MVE_VCMP"#suffix#"16r") (v8i16 MQPR:$v1), (i32 GPR:$v2), fc))>;
  def i32r : Pat<(v4i1 (ARMvcmp (v4i32 MQPR:$v1), (v4i32 (ARMvdup GPR:$v2)), (i32 fc))),
                 (v4i1 (!cast<Instruction>("MVE_VCMP"#suffix#"32r") (v4i32 MQPR:$v1), (i32 GPR:$v2), fc))>;

  def : Pat<(v16i1 (and (v16i1 VCCR:$p1), (v16i1 (ARMvcmp (v16i8 MQPR:$v1), (v16i8 MQPR:$v2), (i32 fc))))),
            (v16i1 (!cast<Instruction>("MVE_VCMP"#suffix#"8") (v16i8 MQPR:$v1), (v16i8 MQPR:$v2), fc, 1, VCCR:$p1))>;
  def : Pat<(v8i1 (and (v8i1 VCCR:$p1), (v8i1 (ARMvcmp (v8i16 MQPR:$v1), (v8i16 MQPR:$v2), (i32 fc))))),
            (v8i1 (!cast<Instruction>("MVE_VCMP"#suffix#"16") (v8i16 MQPR:$v1), (v8i16 MQPR:$v2), fc, 1, VCCR:$p1))>;
  def : Pat<(v4i1 (and (v4i1 VCCR:$p1), (v4i1 (ARMvcmp (v4i32 MQPR:$v1), (v4i32 MQPR:$v2), (i32 fc))))),
            (v4i1 (!cast<Instruction>("MVE_VCMP"#suffix#"32") (v4i32 MQPR:$v1), (v4i32 MQPR:$v2), fc, 1, VCCR:$p1))>;

  def : Pat<(v16i1 (and (v16i1 VCCR:$p1), (v16i1 (ARMvcmp (v16i8 MQPR:$v1), (v16i8 (ARMvdup GPR:$v2)), (i32 fc))))),
            (v16i1 (!cast<Instruction>("MVE_VCMP"#suffix#"8r") (v16i8 MQPR:$v1), (i32 GPR:$v2), fc, 1, VCCR:$p1))>;
  def : Pat<(v8i1 (and (v8i1 VCCR:$p1), (v8i1 (ARMvcmp (v8i16 MQPR:$v1), (v8i16 (ARMvdup GPR:$v2)), (i32 fc))))),
            (v8i1 (!cast<Instruction>("MVE_VCMP"#suffix#"16r") (v8i16 MQPR:$v1), (i32 GPR:$v2), fc, 1, VCCR:$p1))>;
  def : Pat<(v4i1 (and (v4i1 VCCR:$p1), (v4i1 (ARMvcmp (v4i32 MQPR:$v1), (v4i32 (ARMvdup GPR:$v2)), (i32 fc))))),
            (v4i1 (!cast<Instruction>("MVE_VCMP"#suffix#"32r") (v4i32 MQPR:$v1), (i32 GPR:$v2), fc, 1, VCCR:$p1))>;
}

multiclass unpred_vcmp_r_reversible<string suffix, int fc, int fcReversed> {
  defm "":  unpred_vcmp_r<suffix, fc>;

  // Additional patterns that match the vector/scalar comparisons the
  // opposite way round, with the ARMvdup in the first operand of the
  // ARMvcmp. These will usually need a different condition code
  // (except for the symmetric conditions EQ and NE). They're in a
  // separate multiclass because the unsigned CS and HI comparisons
  // don't have reversed forms.

  def : Pat<(v16i1 (ARMvcmp (v16i8 (ARMvdup GPR:$v1)), (v16i8 MQPR:$v2), (i32 fc))),
            (v16i1 (!cast<Instruction>("MVE_VCMP"#suffix#"8r") (v16i8 MQPR:$v2), (i32 GPR:$v1), fcReversed))>;
  def : Pat<(v8i1 (ARMvcmp (v8i16 (ARMvdup GPR:$v1)), (v8i16 MQPR:$v2), (i32 fc))),
            (v8i1 (!cast<Instruction>("MVE_VCMP"#suffix#"16r") (v8i16 MQPR:$v2), (i32 GPR:$v1), fcReversed))>;
  def : Pat<(v4i1 (ARMvcmp (v4i32 (ARMvdup GPR:$v1)), (v4i32 MQPR:$v2), (i32 fc))),
            (v4i1 (!cast<Instruction>("MVE_VCMP"#suffix#"32r") (v4i32 MQPR:$v2), (i32 GPR:$v1), fcReversed))>;

  def : Pat<(v16i1 (and (v16i1 VCCR:$p1), (v16i1 (ARMvcmp (v16i8 (ARMvdup GPR:$v1)), (v16i8 MQPR:$v2), (i32 fc))))),
            (v16i1 (!cast<Instruction>("MVE_VCMP"#suffix#"8r") (v16i8 MQPR:$v2), (i32 GPR:$v1), fcReversed, 1, VCCR:$p1))>;
  def : Pat<(v8i1 (and (v8i1 VCCR:$p1), (v8i1 (ARMvcmp (v8i16 (ARMvdup GPR:$v1)), (v8i16 MQPR:$v2), (i32 fc))))),
            (v8i1 (!cast<Instruction>("MVE_VCMP"#suffix#"16r") (v8i16 MQPR:$v2), (i32 GPR:$v1), fcReversed, 1, VCCR:$p1))>;
  def : Pat<(v4i1 (and (v4i1 VCCR:$p1), (v4i1 (ARMvcmp (v4i32 (ARMvdup GPR:$v1)), (v4i32 MQPR:$v2), (i32 fc))))),
            (v4i1 (!cast<Instruction>("MVE_VCMP"#suffix#"32r") (v4i32 MQPR:$v2), (i32 GPR:$v1), fcReversed, 1, VCCR:$p1))>;
}

multiclass unpred_vcmpf_z<int fc> {
  def f16 : Pat<(v8i1 (ARMvcmpz (v8f16 MQPR:$v1), (i32 fc))),
                (v8i1 (MVE_VCMPf16r (v8f16 MQPR:$v1), ZR, fc))>;
  def f32 : Pat<(v4i1 (ARMvcmpz (v4f32 MQPR:$v1), (i32 fc))),
                (v4i1 (MVE_VCMPf32r (v4f32 MQPR:$v1), ZR, fc))>;

  def : Pat<(v8i1 (and (v8i1 VCCR:$p1), (v8i1 (ARMvcmpz (v8f16 MQPR:$v1), (i32 fc))))),
            (v8i1 (MVE_VCMPf32r (v8f16 MQPR:$v1), ZR, fc, 1, VCCR:$p1))>;
  def : Pat<(v4i1 (and (v4i1 VCCR:$p1), (v4i1 (ARMvcmpz (v4f32 MQPR:$v1), (i32 fc))))),
            (v4i1 (MVE_VCMPf32r (v4f32 MQPR:$v1), ZR, fc, 1, VCCR:$p1))>;
}

multiclass unpred_vcmpf_r<int fc, int fcReversed> {
  def f16 : Pat<(v8i1 (ARMvcmp (v8f16 MQPR:$v1), (v8f16 MQPR:$v2), (i32 fc))),
                (v8i1 (MVE_VCMPf16 (v8f16 MQPR:$v1), (v8f16 MQPR:$v2), fc))>;
  def f32 : Pat<(v4i1 (ARMvcmp (v4f32 MQPR:$v1), (v4f32 MQPR:$v2), (i32 fc))),
                (v4i1 (MVE_VCMPf32 (v4f32 MQPR:$v1), (v4f32 MQPR:$v2), fc))>;

  def f16r : Pat<(v8i1 (ARMvcmp (v8f16 MQPR:$v1), (v8f16 (ARMvdup HPR:$v2)), (i32 fc))),
                 (v8i1 (MVE_VCMPf16r (v8f16 MQPR:$v1), (i32 (COPY_TO_REGCLASS (f16 HPR:$v2), rGPR)), fc))>;
  def f32r : Pat<(v4i1 (ARMvcmp (v4f32 MQPR:$v1), (v4f32 (ARMvdup SPR:$v2)), (i32 fc))),
                 (v4i1 (MVE_VCMPf32r (v4f32 MQPR:$v1), (i32 (COPY_TO_REGCLASS (f32 SPR:$v2), rGPR)), fc))>;

  def      : Pat<(v8i1 (ARMvcmp (v8f16 (ARMvdup HPR:$v1)), (v8f16 MQPR:$v2), (i32 fc))),
                 (v8i1 (MVE_VCMPf16r (v8f16 MQPR:$v2), (i32 (COPY_TO_REGCLASS (f16 HPR:$v1), rGPR)), fcReversed))>;
  def      : Pat<(v4i1 (ARMvcmp (v4f32 (ARMvdup SPR:$v1)), (v4f32 MQPR:$v2), (i32 fc))),
                 (v4i1 (MVE_VCMPf32r (v4f32 MQPR:$v2), (i32 (COPY_TO_REGCLASS (f32 SPR:$v1), rGPR)), fcReversed))>;

  def : Pat<(v8i1 (and (v8i1 VCCR:$p1), (v8i1 (ARMvcmp (v8f16 MQPR:$v1), (v8f16 MQPR:$v2), (i32 fc))))),
            (v8i1 (MVE_VCMPf16 (v8f16 MQPR:$v1), (v8f16 MQPR:$v2), fc, 1, VCCR:$p1))>;
  def : Pat<(v4i1 (and (v4i1 VCCR:$p1), (v4i1 (ARMvcmp (v4f32 MQPR:$v1), (v4f32 MQPR:$v2), (i32 fc))))),
            (v4i1 (MVE_VCMPf32 (v4f32 MQPR:$v1), (v4f32 MQPR:$v2), fc, 1, VCCR:$p1))>;

  def : Pat<(v8i1 (and (v8i1 VCCR:$p1), (v8i1 (ARMvcmp (v8f16 MQPR:$v1), (v8f16 (ARMvdup HPR:$v2)), (i32 fc))))),
            (v8i1 (MVE_VCMPf16r (v8f16 MQPR:$v1), (i32 (COPY_TO_REGCLASS (f16 HPR:$v2), rGPR)), fc, 1, VCCR:$p1))>;
  def : Pat<(v4i1 (and (v4i1 VCCR:$p1), (v4i1 (ARMvcmp (v4f32 MQPR:$v1), (v4f32 (ARMvdup SPR:$v2)), (i32 fc))))),
            (v4i1 (MVE_VCMPf32r (v4f32 MQPR:$v1), (i32 (COPY_TO_REGCLASS (f32 SPR:$v2), rGPR)), fc, 1, VCCR:$p1))>;

  def : Pat<(v8i1 (and (v8i1 VCCR:$p1), (v8i1 (ARMvcmp (v8f16 (ARMvdup HPR:$v1)), (v8f16 MQPR:$v2), (i32 fc))))),
            (v8i1 (MVE_VCMPf16r (v8f16 MQPR:$v2), (i32 (COPY_TO_REGCLASS (f16 HPR:$v1), rGPR)), fcReversed, 1, VCCR:$p1))>;
  def : Pat<(v4i1 (and (v4i1 VCCR:$p1), (v4i1 (ARMvcmp (v4f32 (ARMvdup SPR:$v1)), (v4f32 MQPR:$v2), (i32 fc))))),
            (v4i1 (MVE_VCMPf32r (v4f32 MQPR:$v2), (i32 (COPY_TO_REGCLASS (f32 SPR:$v1), rGPR)), fcReversed, 1, VCCR:$p1))>;
}

let Predicates = [HasMVEInt] in {
  defm MVE_VCEQZ  : unpred_vcmp_z<"i", 0>;
  defm MVE_VCNEZ  : unpred_vcmp_z<"i", 1>;
  defm MVE_VCGEZ  : unpred_vcmp_z<"s", 10>;
  defm MVE_VCLTZ  : unpred_vcmp_z<"s", 11>;
  defm MVE_VCGTZ  : unpred_vcmp_z<"s", 12>;
  defm MVE_VCLEZ  : unpred_vcmp_z<"s", 13>;
  defm MVE_VCGTUZ : unpred_vcmp_z<"u", 8>;
  defm MVE_VCGEUZ : unpred_vcmp_z<"u", 2>;

  defm MVE_VCEQ   : unpred_vcmp_r_reversible<"i", 0, 0>;
  defm MVE_VCNE   : unpred_vcmp_r_reversible<"i", 1, 1>;
  defm MVE_VCGE   : unpred_vcmp_r_reversible<"s", 10, 13>;
  defm MVE_VCLT   : unpred_vcmp_r_reversible<"s", 11, 12>;
  defm MVE_VCGT   : unpred_vcmp_r_reversible<"s", 12, 11>;
  defm MVE_VCLE   : unpred_vcmp_r_reversible<"s", 13, 10>;
  defm MVE_VCGTU  : unpred_vcmp_r<"u", 8>;
  defm MVE_VCGEU  : unpred_vcmp_r<"u", 2>;
}

let Predicates = [HasMVEFloat] in {
  defm MVE_VFCEQZ  : unpred_vcmpf_z<0>;
  defm MVE_VFCNEZ  : unpred_vcmpf_z<1>;
  defm MVE_VFCGEZ  : unpred_vcmpf_z<10>;
  defm MVE_VFCLTZ  : unpred_vcmpf_z<11>;
  defm MVE_VFCGTZ  : unpred_vcmpf_z<12>;
  defm MVE_VFCLEZ  : unpred_vcmpf_z<13>;

  defm MVE_VFCEQ   : unpred_vcmpf_r<0, 0>;
  defm MVE_VFCNE   : unpred_vcmpf_r<1, 1>;
  defm MVE_VFCGE   : unpred_vcmpf_r<10, 13>;
  defm MVE_VFCLT   : unpred_vcmpf_r<11, 12>;
  defm MVE_VFCGT   : unpred_vcmpf_r<12, 11>;
  defm MVE_VFCLE   : unpred_vcmpf_r<13, 10>;
}


// Extra "worst case" and/or/xor partterns, going into and out of GRP
multiclass two_predops<SDPatternOperator opnode, Instruction insn> {
  def v16i1 : Pat<(v16i1 (opnode (v16i1 VCCR:$p1), (v16i1 VCCR:$p2))),
                  (v16i1 (COPY_TO_REGCLASS
                           (insn (i32 (COPY_TO_REGCLASS (v16i1 VCCR:$p1), rGPR)),
                                 (i32 (COPY_TO_REGCLASS (v16i1 VCCR:$p2), rGPR))),
                           VCCR))>;
  def v8i1  : Pat<(v8i1 (opnode (v8i1 VCCR:$p1), (v8i1 VCCR:$p2))),
                  (v8i1 (COPY_TO_REGCLASS
                          (insn (i32 (COPY_TO_REGCLASS (v8i1 VCCR:$p1), rGPR)),
                                (i32 (COPY_TO_REGCLASS (v8i1 VCCR:$p2), rGPR))),
                          VCCR))>;
  def v4i1  : Pat<(v4i1 (opnode (v4i1 VCCR:$p1), (v4i1 VCCR:$p2))),
                  (v4i1 (COPY_TO_REGCLASS
                          (insn (i32 (COPY_TO_REGCLASS (v4i1 VCCR:$p1), rGPR)),
                                (i32 (COPY_TO_REGCLASS (v4i1 VCCR:$p2), rGPR))),
                          VCCR))>;
}

let Predicates = [HasMVEInt] in {
  defm POR    : two_predops<or,  t2ORRrr>;
  defm PAND   : two_predops<and, t2ANDrr>;
  defm PEOR   : two_predops<xor, t2EORrr>;
}

// Occasionally we need to cast between a i32 and a boolean vector, for
// example when moving between rGPR and VPR.P0 as part of predicate vector
// shuffles. We also sometimes need to cast between different predicate
// vector types (v4i1<>v8i1, etc.) also as part of lowering vector shuffles.

def predicate_cast : SDNode<"ARMISD::PREDICATE_CAST", SDTUnaryOp>;

let Predicates = [HasMVEInt] in {
  foreach VT = [ v4i1, v8i1, v16i1 ] in {
    def : Pat<(i32 (predicate_cast (VT VCCR:$src))),
              (i32 (COPY_TO_REGCLASS (VT VCCR:$src), VCCR))>;
    def : Pat<(VT  (predicate_cast (i32 VCCR:$src))),
              (VT  (COPY_TO_REGCLASS (i32 VCCR:$src), VCCR))>;

    foreach VT2 = [ v4i1, v8i1, v16i1 ] in
      def : Pat<(VT  (predicate_cast (VT2 VCCR:$src))),
                (VT  (COPY_TO_REGCLASS (VT2 VCCR:$src), VCCR))>;
  }
}

// end of MVE compares

// start of MVE_qDest_qSrc

class MVE_qDest_qSrc<string iname, string suffix, dag oops, dag iops,
                     string ops, vpred_ops vpred, string cstr,
                     list<dag> pattern=[]>
  : MVE_p<oops, iops, NoItinerary, iname, suffix,
          ops, vpred, cstr, pattern> {
  bits<4> Qd;
  bits<4> Qm;

  let Inst{25-23} = 0b100;
  let Inst{22} = Qd{3};
  let Inst{15-13} = Qd{2-0};
  let Inst{11-9} = 0b111;
  let Inst{6} = 0b0;
  let Inst{5} = Qm{3};
  let Inst{4} = 0b0;
  let Inst{3-1} = Qm{2-0};
}

class MVE_VQxDMLxDH<string iname, bit exch, bit round, bit subtract,
                    string suffix, bits<2> size, string cstr="", list<dag> pattern=[]>
  : MVE_qDest_qSrc<iname, suffix, (outs MQPR:$Qd),
                   (ins MQPR:$Qd_src, MQPR:$Qn, MQPR:$Qm), "$Qd, $Qn, $Qm",
                   vpred_n, "$Qd = $Qd_src"#cstr, pattern> {
  bits<4> Qn;

  let Inst{28} = subtract;
  let Inst{21-20} = size;
  let Inst{19-17} = Qn{2-0};
  let Inst{16} = 0b0;
  let Inst{12} = exch;
  let Inst{8} = 0b0;
  let Inst{7} = Qn{3};
  let Inst{0} = round;
}

multiclass MVE_VQxDMLxDH_multi<string iname, bit exch,
                               bit round, bit subtract> {
  def s8  : MVE_VQxDMLxDH<iname, exch, round, subtract, "s8",  0b00>;
  def s16 : MVE_VQxDMLxDH<iname, exch, round, subtract, "s16", 0b01>;
  def s32 : MVE_VQxDMLxDH<iname, exch, round, subtract, "s32", 0b10, ",@earlyclobber $Qd">;
}

defm MVE_VQDMLADH   : MVE_VQxDMLxDH_multi<"vqdmladh",   0b0, 0b0, 0b0>;
defm MVE_VQDMLADHX  : MVE_VQxDMLxDH_multi<"vqdmladhx",  0b1, 0b0, 0b0>;
defm MVE_VQRDMLADH  : MVE_VQxDMLxDH_multi<"vqrdmladh",  0b0, 0b1, 0b0>;
defm MVE_VQRDMLADHX : MVE_VQxDMLxDH_multi<"vqrdmladhx", 0b1, 0b1, 0b0>;
defm MVE_VQDMLSDH   : MVE_VQxDMLxDH_multi<"vqdmlsdh",   0b0, 0b0, 0b1>;
defm MVE_VQDMLSDHX  : MVE_VQxDMLxDH_multi<"vqdmlsdhx",  0b1, 0b0, 0b1>;
defm MVE_VQRDMLSDH  : MVE_VQxDMLxDH_multi<"vqrdmlsdh",  0b0, 0b1, 0b1>;
defm MVE_VQRDMLSDHX : MVE_VQxDMLxDH_multi<"vqrdmlsdhx", 0b1, 0b1, 0b1>;

class MVE_VCMUL<string iname, string suffix, bit size, string cstr="", list<dag> pattern=[]>
  : MVE_qDest_qSrc<iname, suffix, (outs MQPR:$Qd),
                   (ins MQPR:$Qn, MQPR:$Qm, complexrotateop:$rot),
                   "$Qd, $Qn, $Qm, $rot", vpred_r, cstr, pattern> {
  bits<4> Qn;
  bits<2> rot;

  let Inst{28} = size;
  let Inst{21-20} = 0b11;
  let Inst{19-17} = Qn{2-0};
  let Inst{16} = 0b0;
  let Inst{12} = rot{1};
  let Inst{8} = 0b0;
  let Inst{7} = Qn{3};
  let Inst{0} = rot{0};

  let Predicates = [HasMVEFloat];
}

def MVE_VCMULf16 : MVE_VCMUL<"vcmul", "f16", 0b0>;
def MVE_VCMULf32 : MVE_VCMUL<"vcmul", "f32", 0b1, "@earlyclobber $Qd">;

class MVE_VMULL<string iname, string suffix, bit bit_28, bits<2> bits_21_20,
                bit T, string cstr, list<dag> pattern=[]>
  : MVE_qDest_qSrc<iname, suffix, (outs MQPR:$Qd),
                   (ins MQPR:$Qn, MQPR:$Qm), "$Qd, $Qn, $Qm",
                   vpred_r, cstr, pattern> {
  bits<4> Qd;
  bits<4> Qn;
  bits<4> Qm;

  let Inst{28} = bit_28;
  let Inst{21-20} = bits_21_20;
  let Inst{19-17} = Qn{2-0};
  let Inst{16} = 0b1;
  let Inst{12} = T;
  let Inst{8} = 0b0;
  let Inst{7} = Qn{3};
  let Inst{0} = 0b0;
}

multiclass MVE_VMULL_multi<string iname, string suffix,
                           bit bit_28, bits<2> bits_21_20, string cstr=""> {
  def bh : MVE_VMULL<iname # "b", suffix, bit_28, bits_21_20, 0b0, cstr>;
  def th : MVE_VMULL<iname # "t", suffix, bit_28, bits_21_20, 0b1, cstr>;
}

// For integer multiplies, bits 21:20 encode size, and bit 28 signedness.
// For polynomial multiplies, bits 21:20 take the unused value 0b11, and
// bit 28 switches to encoding the size.

defm MVE_VMULLs8  : MVE_VMULL_multi<"vmull", "s8",  0b0, 0b00>;
defm MVE_VMULLs16 : MVE_VMULL_multi<"vmull", "s16", 0b0, 0b01>;
defm MVE_VMULLs32 : MVE_VMULL_multi<"vmull", "s32", 0b0, 0b10, "@earlyclobber $Qd">;
defm MVE_VMULLu8  : MVE_VMULL_multi<"vmull", "u8",  0b1, 0b00>;
defm MVE_VMULLu16 : MVE_VMULL_multi<"vmull", "u16", 0b1, 0b01>;
defm MVE_VMULLu32 : MVE_VMULL_multi<"vmull", "u32", 0b1, 0b10, "@earlyclobber $Qd">;
defm MVE_VMULLp8  : MVE_VMULL_multi<"vmull", "p8",  0b0, 0b11>;
defm MVE_VMULLp16 : MVE_VMULL_multi<"vmull", "p16", 0b1, 0b11>;

class MVE_VxMULH<string iname, string suffix, bit U, bits<2> size,
                 bit round, list<dag> pattern=[]>
  : MVE_qDest_qSrc<iname, suffix, (outs MQPR:$Qd),
                   (ins MQPR:$Qn, MQPR:$Qm), "$Qd, $Qn, $Qm",
                   vpred_r, "", pattern> {
  bits<4> Qn;

  let Inst{28} = U;
  let Inst{21-20} = size;
  let Inst{19-17} = Qn{2-0};
  let Inst{16} = 0b1;
  let Inst{12} = round;
  let Inst{8} = 0b0;
  let Inst{7} = Qn{3};
  let Inst{0} = 0b1;
}

def MVE_VMULHs8   : MVE_VxMULH<"vmulh",  "s8",  0b0, 0b00, 0b0>;
def MVE_VMULHs16  : MVE_VxMULH<"vmulh",  "s16", 0b0, 0b01, 0b0>;
def MVE_VMULHs32  : MVE_VxMULH<"vmulh",  "s32", 0b0, 0b10, 0b0>;
def MVE_VMULHu8   : MVE_VxMULH<"vmulh",  "u8",  0b1, 0b00, 0b0>;
def MVE_VMULHu16  : MVE_VxMULH<"vmulh",  "u16", 0b1, 0b01, 0b0>;
def MVE_VMULHu32  : MVE_VxMULH<"vmulh",  "u32", 0b1, 0b10, 0b0>;

def MVE_VRMULHs8  : MVE_VxMULH<"vrmulh", "s8",  0b0, 0b00, 0b1>;
def MVE_VRMULHs16 : MVE_VxMULH<"vrmulh", "s16", 0b0, 0b01, 0b1>;
def MVE_VRMULHs32 : MVE_VxMULH<"vrmulh", "s32", 0b0, 0b10, 0b1>;
def MVE_VRMULHu8  : MVE_VxMULH<"vrmulh", "u8",  0b1, 0b00, 0b1>;
def MVE_VRMULHu16 : MVE_VxMULH<"vrmulh", "u16", 0b1, 0b01, 0b1>;
def MVE_VRMULHu32 : MVE_VxMULH<"vrmulh", "u32", 0b1, 0b10, 0b1>;

class MVE_VxMOVxN<string iname, string suffix, bit bit_28, bit bit_17,
                  bits<2> size, bit T, list<dag> pattern=[]>
  : MVE_qDest_qSrc<iname, suffix, (outs MQPR:$Qd),
                   (ins MQPR:$Qd_src, MQPR:$Qm), "$Qd, $Qm",
                   vpred_n, "$Qd = $Qd_src", pattern> {

  let Inst{28} = bit_28;
  let Inst{21-20} = 0b11;
  let Inst{19-18} = size;
  let Inst{17} = bit_17;
  let Inst{16} = 0b1;
  let Inst{12} = T;
  let Inst{8} = 0b0;
  let Inst{7} = !if(!eq(bit_17, 0), 1, 0);
  let Inst{0} = 0b1;
}

multiclass MVE_VxMOVxN_halves<string iname, string suffix,
                              bit bit_28, bit bit_17, bits<2> size> {
  def bh : MVE_VxMOVxN<iname # "b", suffix, bit_28, bit_17, size, 0b0>;
  def th : MVE_VxMOVxN<iname # "t", suffix, bit_28, bit_17, size, 0b1>;
}

defm MVE_VMOVNi16   : MVE_VxMOVxN_halves<"vmovn",   "i16", 0b1, 0b0, 0b00>;
defm MVE_VMOVNi32   : MVE_VxMOVxN_halves<"vmovn",   "i32", 0b1, 0b0, 0b01>;
defm MVE_VQMOVNs16  : MVE_VxMOVxN_halves<"vqmovn",  "s16", 0b0, 0b1, 0b00>;
defm MVE_VQMOVNs32  : MVE_VxMOVxN_halves<"vqmovn",  "s32", 0b0, 0b1, 0b01>;
defm MVE_VQMOVNu16  : MVE_VxMOVxN_halves<"vqmovn",  "u16", 0b1, 0b1, 0b00>;
defm MVE_VQMOVNu32  : MVE_VxMOVxN_halves<"vqmovn",  "u32", 0b1, 0b1, 0b01>;
defm MVE_VQMOVUNs16 : MVE_VxMOVxN_halves<"vqmovun", "s16", 0b0, 0b0, 0b00>;
defm MVE_VQMOVUNs32 : MVE_VxMOVxN_halves<"vqmovun", "s32", 0b0, 0b0, 0b01>;

def MVEvmovn       : SDNode<"ARMISD::VMOVN", SDTARMVEXT>;
let Predicates = [HasMVEInt] in {
  def : Pat<(v8i16 (MVEvmovn (v8i16 MQPR:$Qd_src), (v8i16 MQPR:$Qm), (i32 0))),
            (v8i16 (MVE_VMOVNi32bh (v8i16 MQPR:$Qd_src), (v8i16 MQPR:$Qm)))>;
  def : Pat<(v8i16 (MVEvmovn (v8i16 MQPR:$Qd_src), (v8i16 MQPR:$Qm), (i32 1))),
            (v8i16 (MVE_VMOVNi32th (v8i16 MQPR:$Qd_src), (v8i16 MQPR:$Qm)))>;
  def : Pat<(v16i8 (MVEvmovn (v16i8 MQPR:$Qd_src), (v16i8 MQPR:$Qm), (i32 0))),
            (v16i8 (MVE_VMOVNi16bh (v16i8 MQPR:$Qd_src), (v16i8 MQPR:$Qm)))>;
  def : Pat<(v16i8 (MVEvmovn (v16i8 MQPR:$Qd_src), (v16i8 MQPR:$Qm), (i32 1))),
            (v16i8 (MVE_VMOVNi16th (v16i8 MQPR:$Qd_src), (v16i8 MQPR:$Qm)))>;
}

class MVE_VCVT_ff<string iname, string suffix, bit op, bit T,
                  list<dag> pattern=[]>
  : MVE_qDest_qSrc<iname, suffix, (outs MQPR:$Qd), (ins MQPR:$Qd_src, MQPR:$Qm),
                   "$Qd, $Qm", vpred_n, "$Qd = $Qd_src", pattern> {
  let Inst{28} = op;
  let Inst{21-16} = 0b111111;
  let Inst{12} = T;
  let Inst{8-7} = 0b00;
  let Inst{0} = 0b1;

  let Predicates = [HasMVEFloat];
}

multiclass MVE_VCVT_f2h_m<string iname, int half> {
  def "": MVE_VCVT_ff<iname, "f16.f32", 0b0, half>;

  let Predicates = [HasMVEFloat] in {
    def : Pat<(v8f16 (int_arm_mve_vcvt_narrow
                         (v8f16 MQPR:$Qd_src), (v4f32 MQPR:$Qm), (i32 half))),
              (v8f16 (!cast<Instruction>(NAME)
                         (v8f16 MQPR:$Qd_src), (v4f32 MQPR:$Qm)))>;
    def : Pat<(v8f16 (int_arm_mve_vcvt_narrow_predicated
                         (v8f16 MQPR:$Qd_src), (v4f32 MQPR:$Qm), (i32 half),
                         (v4i1 VCCR:$mask))),
              (v8f16 (!cast<Instruction>(NAME)
                         (v8f16 MQPR:$Qd_src), (v4f32 MQPR:$Qm),
                         (i32 1), (v4i1 VCCR:$mask)))>;
  }
}

multiclass MVE_VCVT_h2f_m<string iname, int half> {
  def "": MVE_VCVT_ff<iname, "f32.f16", 0b1, half>;
}

defm MVE_VCVTf16f32bh : MVE_VCVT_f2h_m<"vcvtb", 0b0>;
defm MVE_VCVTf16f32th : MVE_VCVT_f2h_m<"vcvtt", 0b1>;
defm MVE_VCVTf32f16bh : MVE_VCVT_h2f_m<"vcvtb", 0b0>;
defm MVE_VCVTf32f16th : MVE_VCVT_h2f_m<"vcvtt", 0b1>;

class MVE_VxCADD<string iname, string suffix, bits<2> size, bit halve,
                 string cstr="", list<dag> pattern=[]>
  : MVE_qDest_qSrc<iname, suffix, (outs MQPR:$Qd),
                   (ins MQPR:$Qn, MQPR:$Qm, complexrotateopodd:$rot),
                   "$Qd, $Qn, $Qm, $rot", vpred_r, cstr, pattern> {
  bits<4> Qn;
  bit rot;

  let Inst{28} = halve;
  let Inst{21-20} = size;
  let Inst{19-17} = Qn{2-0};
  let Inst{16} = 0b0;
  let Inst{12} = rot;
  let Inst{8} = 0b1;
  let Inst{7} = Qn{3};
  let Inst{0} = 0b0;
}

def MVE_VCADDi8   : MVE_VxCADD<"vcadd", "i8", 0b00, 0b1>;
def MVE_VCADDi16  : MVE_VxCADD<"vcadd", "i16", 0b01, 0b1>;
def MVE_VCADDi32  : MVE_VxCADD<"vcadd", "i32", 0b10, 0b1, "@earlyclobber $Qd">;

def MVE_VHCADDs8  : MVE_VxCADD<"vhcadd", "s8", 0b00, 0b0>;
def MVE_VHCADDs16 : MVE_VxCADD<"vhcadd", "s16", 0b01, 0b0>;
def MVE_VHCADDs32 : MVE_VxCADD<"vhcadd", "s32", 0b10, 0b0, "@earlyclobber $Qd">;

class MVE_VADCSBC<string iname, bit I, bit subtract,
                  dag carryin, list<dag> pattern=[]>
  : MVE_qDest_qSrc<iname, "i32", (outs MQPR:$Qd, cl_FPSCR_NZCV:$carryout),
                   !con((ins MQPR:$Qn, MQPR:$Qm), carryin),
                   "$Qd, $Qn, $Qm", vpred_r, "", pattern> {
  bits<4> Qn;

  let Inst{28} = subtract;
  let Inst{21-20} = 0b11;
  let Inst{19-17} = Qn{2-0};
  let Inst{16} = 0b0;
  let Inst{12} = I;
  let Inst{8} = 0b1;
  let Inst{7} = Qn{3};
  let Inst{0} = 0b0;

  // Custom decoder method in order to add the FPSCR operand(s), which
  // Tablegen won't do right
  let DecoderMethod = "DecodeMVEVADCInstruction";
}

def MVE_VADC  : MVE_VADCSBC<"vadc",  0b0, 0b0, (ins cl_FPSCR_NZCV:$carryin)>;
def MVE_VADCI : MVE_VADCSBC<"vadci", 0b1, 0b0, (ins)>;

def MVE_VSBC  : MVE_VADCSBC<"vsbc",  0b0, 0b1, (ins cl_FPSCR_NZCV:$carryin)>;
def MVE_VSBCI : MVE_VADCSBC<"vsbci", 0b1, 0b1, (ins)>;

class MVE_VQDMULL<string iname, string suffix, bit size, bit T,
                  string cstr="", list<dag> pattern=[]>
  : MVE_qDest_qSrc<iname, suffix, (outs MQPR:$Qd),
                   (ins MQPR:$Qn, MQPR:$Qm), "$Qd, $Qn, $Qm",
                   vpred_r, cstr, pattern> {
  bits<4> Qn;

  let Inst{28} = size;
  let Inst{21-20} = 0b11;
  let Inst{19-17} = Qn{2-0};
  let Inst{16} = 0b0;
  let Inst{12} = T;
  let Inst{8} = 0b1;
  let Inst{7} = Qn{3};
  let Inst{0} = 0b1;
}

multiclass MVE_VQDMULL_halves<string suffix, bit size, string cstr=""> {
  def bh : MVE_VQDMULL<"vqdmullb", suffix, size, 0b0, cstr>;
  def th : MVE_VQDMULL<"vqdmullt", suffix, size, 0b1, cstr>;
}

defm MVE_VQDMULLs16 : MVE_VQDMULL_halves<"s16", 0b0>;
defm MVE_VQDMULLs32 : MVE_VQDMULL_halves<"s32", 0b1, "@earlyclobber $Qd">;

// end of mve_qDest_qSrc

// start of mve_qDest_rSrc

class MVE_qr_base<dag oops, dag iops, InstrItinClass itin, string iname,
                  string suffix, string ops, vpred_ops vpred, string cstr,
                  list<dag> pattern=[]>
   : MVE_p<oops, iops, NoItinerary, iname, suffix, ops, vpred, cstr, pattern> {
  bits<4> Qd;
  bits<4> Qn;
  bits<4> Rm;

  let Inst{25-23} = 0b100;
  let Inst{22} = Qd{3};
  let Inst{19-17} = Qn{2-0};
  let Inst{15-13} = Qd{2-0};
  let Inst{11-9} = 0b111;
  let Inst{7} = Qn{3};
  let Inst{6} = 0b1;
  let Inst{4} = 0b0;
  let Inst{3-0} = Rm{3-0};
}

class MVE_qDest_rSrc<string iname, string suffix, string cstr="", list<dag> pattern=[]>
  : MVE_qr_base<(outs MQPR:$Qd), (ins MQPR:$Qn, rGPR:$Rm),
          NoItinerary, iname, suffix, "$Qd, $Qn, $Rm", vpred_r, cstr,
           pattern>;

class MVE_qDestSrc_rSrc<string iname, string suffix, list<dag> pattern=[]>
  : MVE_qr_base<(outs MQPR:$Qd), (ins MQPR:$Qd_src, MQPR:$Qn, rGPR:$Rm),
          NoItinerary, iname, suffix, "$Qd, $Qn, $Rm", vpred_n, "$Qd = $Qd_src",
           pattern>;

class MVE_qDest_single_rSrc<string iname, string suffix, list<dag> pattern=[]>
  : MVE_p<(outs MQPR:$Qd), (ins MQPR:$Qd_src, rGPR:$Rm), NoItinerary, iname,
          suffix, "$Qd, $Rm", vpred_n, "$Qd = $Qd_src", pattern> {
  bits<4> Qd;
  bits<4> Rm;

  let Inst{22} = Qd{3};
  let Inst{15-13} = Qd{2-0};
  let Inst{3-0} = Rm{3-0};
}

class MVE_VADDSUB_qr<string iname, string suffix, bits<2> size,
                     bit bit_5, bit bit_12, bit bit_16,
                     bit bit_28, list<dag> pattern=[]>
  : MVE_qDest_rSrc<iname, suffix, "", pattern> {

  let Inst{28} = bit_28;
  let Inst{21-20} = size;
  let Inst{16} = bit_16;
  let Inst{12} = bit_12;
  let Inst{8} = 0b1;
  let Inst{5} = bit_5;
  let validForTailPredication = 1;
}

multiclass MVE_VADDSUB_qr_sizes<string iname, string suffix,
                                bit bit_5, bit bit_12, bit bit_16,
                                bit bit_28, list<dag> pattern=[]> {
  def "8"  : MVE_VADDSUB_qr<iname, suffix#"8",  0b00,
                            bit_5, bit_12, bit_16, bit_28>;
  def "16" : MVE_VADDSUB_qr<iname, suffix#"16", 0b01,
                            bit_5, bit_12, bit_16, bit_28>;
  def "32" : MVE_VADDSUB_qr<iname, suffix#"32", 0b10,
                            bit_5, bit_12, bit_16, bit_28>;
}

defm MVE_VADD_qr_i  : MVE_VADDSUB_qr_sizes<"vadd",  "i", 0b0, 0b0, 0b1, 0b0>;
defm MVE_VQADD_qr_s : MVE_VADDSUB_qr_sizes<"vqadd", "s", 0b1, 0b0, 0b0, 0b0>;
defm MVE_VQADD_qr_u : MVE_VADDSUB_qr_sizes<"vqadd", "u", 0b1, 0b0, 0b0, 0b1>;

defm MVE_VSUB_qr_i  : MVE_VADDSUB_qr_sizes<"vsub",  "i", 0b0, 0b1, 0b1, 0b0>;
defm MVE_VQSUB_qr_s : MVE_VADDSUB_qr_sizes<"vqsub", "s", 0b1, 0b1, 0b0, 0b0>;
defm MVE_VQSUB_qr_u : MVE_VADDSUB_qr_sizes<"vqsub", "u", 0b1, 0b1, 0b0, 0b1>;

let Predicates = [HasMVEInt] in {
  def : Pat<(v16i8 (add (v16i8 MQPR:$val1), (v16i8 (ARMvdup GPR:$val2)))),
            (v16i8 (MVE_VADD_qr_i8 (v16i8 MQPR:$val1), (i32 GPR:$val2)))>;
  def : Pat<(v8i16 (add (v8i16 MQPR:$val1), (v8i16 (ARMvdup GPR:$val2)))),
            (v8i16 (MVE_VADD_qr_i16 (v8i16 MQPR:$val1), (i32 GPR:$val2)))>;
  def : Pat<(v4i32 (add (v4i32 MQPR:$val1), (v4i32 (ARMvdup GPR:$val2)))),
            (v4i32 (MVE_VADD_qr_i32 (v4i32 MQPR:$val1), (i32 GPR:$val2)))>;
}

let Predicates = [HasMVEInt] in {
  def : Pat<(v16i8 (sub (v16i8 MQPR:$val1), (v16i8 (ARMvdup GPR:$val2)))),
            (v16i8 (MVE_VSUB_qr_i8 (v16i8 MQPR:$val1), (i32 GPR:$val2)))>;
  def : Pat<(v8i16 (sub (v8i16 MQPR:$val1), (v8i16 (ARMvdup GPR:$val2)))),
            (v8i16 (MVE_VSUB_qr_i16 (v8i16 MQPR:$val1), (i32 GPR:$val2)))>;
  def : Pat<(v4i32 (sub (v4i32 MQPR:$val1), (v4i32 (ARMvdup GPR:$val2)))),
            (v4i32 (MVE_VSUB_qr_i32 (v4i32 MQPR:$val1), (i32 GPR:$val2)))>;
}

class MVE_VQDMULL_qr<string iname, string suffix, bit size,
                     bit T, string cstr="", list<dag> pattern=[]>
  : MVE_qDest_rSrc<iname, suffix, cstr, pattern> {

  let Inst{28} = size;
  let Inst{21-20} = 0b11;
  let Inst{16} = 0b0;
  let Inst{12} = T;
  let Inst{8} = 0b1;
  let Inst{5} = 0b1;
}

multiclass MVE_VQDMULL_qr_halves<string suffix, bit size, string cstr=""> {
  def bh : MVE_VQDMULL_qr<"vqdmullb", suffix, size, 0b0, cstr>;
  def th : MVE_VQDMULL_qr<"vqdmullt", suffix, size, 0b1, cstr>;
}

defm MVE_VQDMULL_qr_s16 : MVE_VQDMULL_qr_halves<"s16", 0b0>;
defm MVE_VQDMULL_qr_s32 : MVE_VQDMULL_qr_halves<"s32", 0b1, "@earlyclobber $Qd">;

class MVE_VxADDSUB_qr<string iname, string suffix,
                      bit bit_28, bits<2> bits_21_20, bit subtract,
                      list<dag> pattern=[]>
  : MVE_qDest_rSrc<iname, suffix, "", pattern> {

  let Inst{28} = bit_28;
  let Inst{21-20} = bits_21_20;
  let Inst{16} = 0b0;
  let Inst{12} = subtract;
  let Inst{8} = 0b1;
  let Inst{5} = 0b0;
  let validForTailPredication = 1;
}

def MVE_VHADD_qr_s8   : MVE_VxADDSUB_qr<"vhadd", "s8",  0b0, 0b00, 0b0>;
def MVE_VHADD_qr_s16  : MVE_VxADDSUB_qr<"vhadd", "s16", 0b0, 0b01, 0b0>;
def MVE_VHADD_qr_s32  : MVE_VxADDSUB_qr<"vhadd", "s32", 0b0, 0b10, 0b0>;
def MVE_VHADD_qr_u8   : MVE_VxADDSUB_qr<"vhadd", "u8",  0b1, 0b00, 0b0>;
def MVE_VHADD_qr_u16  : MVE_VxADDSUB_qr<"vhadd", "u16", 0b1, 0b01, 0b0>;
def MVE_VHADD_qr_u32  : MVE_VxADDSUB_qr<"vhadd", "u32", 0b1, 0b10, 0b0>;

def MVE_VHSUB_qr_s8   : MVE_VxADDSUB_qr<"vhsub", "s8",  0b0, 0b00, 0b1>;
def MVE_VHSUB_qr_s16  : MVE_VxADDSUB_qr<"vhsub", "s16", 0b0, 0b01, 0b1>;
def MVE_VHSUB_qr_s32  : MVE_VxADDSUB_qr<"vhsub", "s32", 0b0, 0b10, 0b1>;
def MVE_VHSUB_qr_u8   : MVE_VxADDSUB_qr<"vhsub", "u8",  0b1, 0b00, 0b1>;
def MVE_VHSUB_qr_u16  : MVE_VxADDSUB_qr<"vhsub", "u16", 0b1, 0b01, 0b1>;
def MVE_VHSUB_qr_u32  : MVE_VxADDSUB_qr<"vhsub", "u32", 0b1, 0b10, 0b1>;

let Predicates = [HasMVEFloat] in {
  def MVE_VADD_qr_f32 : MVE_VxADDSUB_qr<"vadd",  "f32", 0b0, 0b11, 0b0>;
  def MVE_VADD_qr_f16 : MVE_VxADDSUB_qr<"vadd",  "f16", 0b1, 0b11, 0b0>;

  def MVE_VSUB_qr_f32 : MVE_VxADDSUB_qr<"vsub",  "f32", 0b0, 0b11, 0b1>;
  def MVE_VSUB_qr_f16 : MVE_VxADDSUB_qr<"vsub",  "f16", 0b1, 0b11, 0b1>;
}

class MVE_VxSHL_qr<string iname, string suffix, bit U, bits<2> size,
                   bit bit_7, bit bit_17, list<dag> pattern=[]>
  : MVE_qDest_single_rSrc<iname, suffix, pattern> {

  let Inst{28} = U;
  let Inst{25-23} = 0b100;
  let Inst{21-20} = 0b11;
  let Inst{19-18} = size;
  let Inst{17} = bit_17;
  let Inst{16} = 0b1;
  let Inst{12-8} = 0b11110;
  let Inst{7} = bit_7;
  let Inst{6-4} = 0b110;
  let validForTailPredication = 1;
}

multiclass MVE_VxSHL_qr_types<string iname, bit bit_7, bit bit_17> {
  def s8  : MVE_VxSHL_qr<iname, "s8", 0b0, 0b00, bit_7, bit_17>;
  def s16 : MVE_VxSHL_qr<iname, "s16", 0b0, 0b01, bit_7, bit_17>;
  def s32 : MVE_VxSHL_qr<iname, "s32", 0b0, 0b10, bit_7, bit_17>;
  def u8  : MVE_VxSHL_qr<iname, "u8", 0b1, 0b00, bit_7, bit_17>;
  def u16 : MVE_VxSHL_qr<iname, "u16", 0b1, 0b01, bit_7, bit_17>;
  def u32 : MVE_VxSHL_qr<iname, "u32", 0b1, 0b10, bit_7, bit_17>;
}

defm MVE_VSHL_qr   : MVE_VxSHL_qr_types<"vshl",   0b0, 0b0>;
defm MVE_VRSHL_qr  : MVE_VxSHL_qr_types<"vrshl",  0b0, 0b1>;
defm MVE_VQSHL_qr  : MVE_VxSHL_qr_types<"vqshl",  0b1, 0b0>;
defm MVE_VQRSHL_qr : MVE_VxSHL_qr_types<"vqrshl", 0b1, 0b1>;

let Predicates = [HasMVEInt] in {
  def : Pat<(v4i32 (ARMvshlu (v4i32 MQPR:$Qm), (v4i32 (ARMvdup GPR:$Rm)))),
            (v4i32 (MVE_VSHL_qru32 (v4i32 MQPR:$Qm), GPR:$Rm))>;
  def : Pat<(v8i16 (ARMvshlu (v8i16 MQPR:$Qm), (v8i16 (ARMvdup GPR:$Rm)))),
            (v8i16 (MVE_VSHL_qru16 (v8i16 MQPR:$Qm), GPR:$Rm))>;
  def : Pat<(v16i8 (ARMvshlu (v16i8 MQPR:$Qm), (v16i8 (ARMvdup GPR:$Rm)))),
            (v16i8 (MVE_VSHL_qru8 (v16i8 MQPR:$Qm), GPR:$Rm))>;

  def : Pat<(v4i32 (ARMvshls (v4i32 MQPR:$Qm), (v4i32 (ARMvdup GPR:$Rm)))),
            (v4i32 (MVE_VSHL_qrs32 (v4i32 MQPR:$Qm), GPR:$Rm))>;
  def : Pat<(v8i16 (ARMvshls (v8i16 MQPR:$Qm), (v8i16 (ARMvdup GPR:$Rm)))),
            (v8i16 (MVE_VSHL_qrs16 (v8i16 MQPR:$Qm), GPR:$Rm))>;
  def : Pat<(v16i8 (ARMvshls (v16i8 MQPR:$Qm), (v16i8 (ARMvdup GPR:$Rm)))),
            (v16i8 (MVE_VSHL_qrs8 (v16i8 MQPR:$Qm), GPR:$Rm))>;
}

class MVE_VBRSR<string iname, string suffix, bits<2> size, list<dag> pattern=[]>
  : MVE_qDest_rSrc<iname, suffix, "", pattern> {

  let Inst{28} = 0b1;
  let Inst{21-20} = size;
  let Inst{16} = 0b1;
  let Inst{12} = 0b1;
  let Inst{8} = 0b0;
  let Inst{5} = 0b1;
  let validForTailPredication = 1;
}

def MVE_VBRSR8  : MVE_VBRSR<"vbrsr", "8", 0b00>;
def MVE_VBRSR16 : MVE_VBRSR<"vbrsr", "16", 0b01>;
def MVE_VBRSR32 : MVE_VBRSR<"vbrsr", "32", 0b10>;

let Predicates = [HasMVEInt] in {
  def : Pat<(v16i8 ( bitreverse (v16i8 MQPR:$val1))),
            (v16i8 ( MVE_VBRSR8 (v16i8 MQPR:$val1), (t2MOVi (i32 8)) ))>;

  def : Pat<(v4i32 ( bitreverse (v4i32 MQPR:$val1))),
            (v4i32 ( MVE_VBRSR32 (v4i32 MQPR:$val1), (t2MOVi (i32 32)) ))>;

  def : Pat<(v8i16 ( bitreverse (v8i16 MQPR:$val1))),
            (v8i16 ( MVE_VBRSR16 (v8i16 MQPR:$val1), (t2MOVi (i32 16)) ))>;
}

class MVE_VMUL_qr_int<string iname, string suffix,
                      bits<2> size, list<dag> pattern=[]>
  : MVE_qDest_rSrc<iname, suffix, "", pattern> {

  let Inst{28} = 0b0;
  let Inst{21-20} = size;
  let Inst{16} = 0b1;
  let Inst{12} = 0b1;
  let Inst{8} = 0b0;
  let Inst{5} = 0b1;
  let validForTailPredication = 1;
}

def MVE_VMUL_qr_i8  : MVE_VMUL_qr_int<"vmul", "i8",  0b00>;
def MVE_VMUL_qr_i16 : MVE_VMUL_qr_int<"vmul", "i16", 0b01>;
def MVE_VMUL_qr_i32 : MVE_VMUL_qr_int<"vmul", "i32", 0b10>;

let Predicates = [HasMVEInt] in {
  def : Pat<(v16i8 (mul (v16i8 MQPR:$val1), (v16i8 (ARMvdup GPR:$val2)))),
            (v16i8 (MVE_VMUL_qr_i8 (v16i8 MQPR:$val1), (i32 GPR:$val2)))>;
  def : Pat<(v8i16 (mul (v8i16 MQPR:$val1), (v8i16 (ARMvdup GPR:$val2)))),
            (v8i16 (MVE_VMUL_qr_i16 (v8i16 MQPR:$val1), (i32 GPR:$val2)))>;
  def : Pat<(v4i32 (mul (v4i32 MQPR:$val1), (v4i32 (ARMvdup GPR:$val2)))),
            (v4i32 (MVE_VMUL_qr_i32 (v4i32 MQPR:$val1), (i32 GPR:$val2)))>;
}

class MVE_VxxMUL_qr<string iname, string suffix,
                    bit bit_28, bits<2> bits_21_20, list<dag> pattern=[]>
  : MVE_qDest_rSrc<iname, suffix, "", pattern> {

  let Inst{28} = bit_28;
  let Inst{21-20} = bits_21_20;
  let Inst{16} = 0b1;
  let Inst{12} = 0b0;
  let Inst{8} = 0b0;
  let Inst{5} = 0b1;
}

def MVE_VQDMULH_qr_s8   : MVE_VxxMUL_qr<"vqdmulh",  "s8",  0b0, 0b00>;
def MVE_VQDMULH_qr_s16  : MVE_VxxMUL_qr<"vqdmulh",  "s16", 0b0, 0b01>;
def MVE_VQDMULH_qr_s32  : MVE_VxxMUL_qr<"vqdmulh",  "s32", 0b0, 0b10>;

def MVE_VQRDMULH_qr_s8  : MVE_VxxMUL_qr<"vqrdmulh", "s8",  0b1, 0b00>;
def MVE_VQRDMULH_qr_s16 : MVE_VxxMUL_qr<"vqrdmulh", "s16", 0b1, 0b01>;
def MVE_VQRDMULH_qr_s32 : MVE_VxxMUL_qr<"vqrdmulh", "s32", 0b1, 0b10>;

let Predicates = [HasMVEFloat], validForTailPredication = 1 in {
  def MVE_VMUL_qr_f16   : MVE_VxxMUL_qr<"vmul", "f16", 0b1, 0b11>;
  def MVE_VMUL_qr_f32   : MVE_VxxMUL_qr<"vmul", "f32", 0b0, 0b11>;
}

class MVE_VFMAMLA_qr<string iname, string suffix,
                     bit bit_28, bits<2> bits_21_20, bit S,
                     list<dag> pattern=[]>
  : MVE_qDestSrc_rSrc<iname, suffix, pattern> {

  let Inst{28} = bit_28;
  let Inst{21-20} = bits_21_20;
  let Inst{16} = 0b1;
  let Inst{12} = S;
  let Inst{8} = 0b0;
  let Inst{5} = 0b0;
  let validForTailPredication = 1;
}

def MVE_VMLA_qr_s8     : MVE_VFMAMLA_qr<"vmla",  "s8",  0b0, 0b00, 0b0>;
def MVE_VMLA_qr_s16    : MVE_VFMAMLA_qr<"vmla",  "s16", 0b0, 0b01, 0b0>;
def MVE_VMLA_qr_s32    : MVE_VFMAMLA_qr<"vmla",  "s32", 0b0, 0b10, 0b0>;
def MVE_VMLA_qr_u8     : MVE_VFMAMLA_qr<"vmla",  "u8",  0b1, 0b00, 0b0>;
def MVE_VMLA_qr_u16    : MVE_VFMAMLA_qr<"vmla",  "u16", 0b1, 0b01, 0b0>;
def MVE_VMLA_qr_u32    : MVE_VFMAMLA_qr<"vmla",  "u32", 0b1, 0b10, 0b0>;

def MVE_VMLAS_qr_s8    : MVE_VFMAMLA_qr<"vmlas", "s8",  0b0, 0b00, 0b1>;
def MVE_VMLAS_qr_s16   : MVE_VFMAMLA_qr<"vmlas", "s16", 0b0, 0b01, 0b1>;
def MVE_VMLAS_qr_s32   : MVE_VFMAMLA_qr<"vmlas", "s32", 0b0, 0b10, 0b1>;
def MVE_VMLAS_qr_u8    : MVE_VFMAMLA_qr<"vmlas", "u8",  0b1, 0b00, 0b1>;
def MVE_VMLAS_qr_u16   : MVE_VFMAMLA_qr<"vmlas", "u16", 0b1, 0b01, 0b1>;
def MVE_VMLAS_qr_u32   : MVE_VFMAMLA_qr<"vmlas", "u32", 0b1, 0b10, 0b1>;

let Predicates = [HasMVEInt] in {
  def : Pat<(v4i32 (add (v4i32 MQPR:$src1),
                        (v4i32 (mul (v4i32 MQPR:$src2),
                                    (v4i32 (ARMvdup (i32 rGPR:$x))))))),
            (v4i32 (MVE_VMLA_qr_u32 $src1, $src2, $x))>;
  def : Pat<(v8i16 (add (v8i16 MQPR:$src1),
                        (v8i16 (mul (v8i16 MQPR:$src2),
                                    (v8i16 (ARMvdup (i32 rGPR:$x))))))),
            (v8i16 (MVE_VMLA_qr_u16 $src1, $src2, $x))>;
  def : Pat<(v16i8 (add (v16i8 MQPR:$src1),
                        (v16i8 (mul (v16i8 MQPR:$src2),
                                    (v16i8 (ARMvdup (i32 rGPR:$x))))))),
            (v16i8 (MVE_VMLA_qr_u8 $src1, $src2, $x))>;
}

let Predicates = [HasMVEFloat] in {
  def MVE_VFMA_qr_f16  : MVE_VFMAMLA_qr<"vfma",  "f16", 0b1, 0b11, 0b0>;
  def MVE_VFMA_qr_f32  : MVE_VFMAMLA_qr<"vfma",  "f32", 0b0, 0b11, 0b0>;
  def MVE_VFMA_qr_Sf16 : MVE_VFMAMLA_qr<"vfmas", "f16", 0b1, 0b11, 0b1>;
  def MVE_VFMA_qr_Sf32 : MVE_VFMAMLA_qr<"vfmas", "f32", 0b0, 0b11, 0b1>;
}

class MVE_VQDMLAH_qr<string iname, string suffix, bit U, bits<2> size,
                     bit bit_5, bit bit_12, list<dag> pattern=[]>
  : MVE_qDestSrc_rSrc<iname, suffix, pattern> {

  let Inst{28} = U;
  let Inst{21-20} = size;
  let Inst{16} = 0b0;
  let Inst{12} = bit_12;
  let Inst{8} = 0b0;
  let Inst{5} = bit_5;
}

multiclass MVE_VQDMLAH_qr_types<string iname, bit bit_5, bit bit_12> {
  def s8  : MVE_VQDMLAH_qr<iname, "s8",  0b0, 0b00, bit_5, bit_12>;
  def s16 : MVE_VQDMLAH_qr<iname, "s16", 0b0, 0b01, bit_5, bit_12>;
  def s32 : MVE_VQDMLAH_qr<iname, "s32", 0b0, 0b10, bit_5, bit_12>;
}

defm MVE_VQDMLAH_qr   : MVE_VQDMLAH_qr_types<"vqdmlah",   0b1, 0b0>;
defm MVE_VQRDMLAH_qr  : MVE_VQDMLAH_qr_types<"vqrdmlah",  0b0, 0b0>;
defm MVE_VQDMLASH_qr  : MVE_VQDMLAH_qr_types<"vqdmlash",  0b1, 0b1>;
defm MVE_VQRDMLASH_qr : MVE_VQDMLAH_qr_types<"vqrdmlash", 0b0, 0b1>;

class MVE_VxDUP<string iname, string suffix, bits<2> size, bit bit_12,
              list<dag> pattern=[]>
  : MVE_p<(outs MQPR:$Qd, tGPREven:$Rn),
          (ins tGPREven:$Rn_src, MVE_VIDUP_imm:$imm), NoItinerary,
          iname, suffix, "$Qd, $Rn, $imm", vpred_r, "$Rn = $Rn_src",
          pattern> {
  bits<4> Qd;
  bits<4> Rn;
  bits<2> imm;

  let Inst{28} = 0b0;
  let Inst{25-23} = 0b100;
  let Inst{22} = Qd{3};
  let Inst{21-20} = size;
  let Inst{19-17} = Rn{3-1};
  let Inst{16} = 0b1;
  let Inst{15-13} = Qd{2-0};
  let Inst{12} = bit_12;
  let Inst{11-8} = 0b1111;
  let Inst{7} = imm{1};
  let Inst{6-1} = 0b110111;
  let Inst{0} = imm{0};
  let validForTailPredication = 1;
}

def MVE_VIDUPu8  : MVE_VxDUP<"vidup", "u8",  0b00, 0b0>;
def MVE_VIDUPu16 : MVE_VxDUP<"vidup", "u16", 0b01, 0b0>;
def MVE_VIDUPu32 : MVE_VxDUP<"vidup", "u32", 0b10, 0b0>;

def MVE_VDDUPu8  : MVE_VxDUP<"vddup", "u8",  0b00, 0b1>;
def MVE_VDDUPu16 : MVE_VxDUP<"vddup", "u16", 0b01, 0b1>;
def MVE_VDDUPu32 : MVE_VxDUP<"vddup", "u32", 0b10, 0b1>;

class MVE_VxWDUP<string iname, string suffix, bits<2> size, bit bit_12,
                 list<dag> pattern=[]>
  : MVE_p<(outs MQPR:$Qd, tGPREven:$Rn),
          (ins tGPREven:$Rn_src, tGPROdd:$Rm, MVE_VIDUP_imm:$imm), NoItinerary,
          iname, suffix, "$Qd, $Rn, $Rm, $imm", vpred_r, "$Rn = $Rn_src",
          pattern> {
  bits<4> Qd;
  bits<4> Rm;
  bits<4> Rn;
  bits<2> imm;

  let Inst{28} = 0b0;
  let Inst{25-23} = 0b100;
  let Inst{22} = Qd{3};
  let Inst{21-20} = size;
  let Inst{19-17} = Rn{3-1};
  let Inst{16} = 0b1;
  let Inst{15-13} = Qd{2-0};
  let Inst{12} = bit_12;
  let Inst{11-8} = 0b1111;
  let Inst{7} = imm{1};
  let Inst{6-4} = 0b110;
  let Inst{3-1} = Rm{3-1};
  let Inst{0} = imm{0};
  let validForTailPredication = 1;
}

def MVE_VIWDUPu8  : MVE_VxWDUP<"viwdup", "u8",  0b00, 0b0>;
def MVE_VIWDUPu16 : MVE_VxWDUP<"viwdup", "u16", 0b01, 0b0>;
def MVE_VIWDUPu32 : MVE_VxWDUP<"viwdup", "u32", 0b10, 0b0>;

def MVE_VDWDUPu8  : MVE_VxWDUP<"vdwdup", "u8",  0b00, 0b1>;
def MVE_VDWDUPu16 : MVE_VxWDUP<"vdwdup", "u16", 0b01, 0b1>;
def MVE_VDWDUPu32 : MVE_VxWDUP<"vdwdup", "u32", 0b10, 0b1>;

let hasSideEffects = 1 in
class MVE_VCTP<string suffix, bits<2> size, list<dag> pattern=[]>
  : MVE_p<(outs VCCR:$P0), (ins rGPR:$Rn), NoItinerary, "vctp", suffix,
          "$Rn", vpred_n, "", pattern> {
  bits<4> Rn;

  let Inst{28-27} = 0b10;
  let Inst{26-22} = 0b00000;
  let Inst{21-20} = size;
  let Inst{19-16} = Rn{3-0};
  let Inst{15-11} = 0b11101;
  let Inst{10-0}  = 0b00000000001;
  let Unpredictable{10-0} = 0b11111111111;

  let Constraints = "";
  let DecoderMethod = "DecodeMveVCTP";
  let validForTailPredication = 1;
}

def MVE_VCTP8  : MVE_VCTP<"8",  0b00>;
def MVE_VCTP16 : MVE_VCTP<"16", 0b01>;
def MVE_VCTP32 : MVE_VCTP<"32", 0b10>;
def MVE_VCTP64 : MVE_VCTP<"64", 0b11>;

let Predicates = [HasMVEInt] in {
  def : Pat<(int_arm_vctp8 rGPR:$Rn),
            (v16i1 (MVE_VCTP8 rGPR:$Rn))>;
  def : Pat<(int_arm_vctp16 rGPR:$Rn),
            (v8i1 (MVE_VCTP16 rGPR:$Rn))>;
  def : Pat<(int_arm_vctp32 rGPR:$Rn),
            (v4i1 (MVE_VCTP32 rGPR:$Rn))>;
}

// end of mve_qDest_rSrc

// start of coproc mov

class MVE_VMOV_64bit<dag oops, dag iops, bit to_qreg, string ops, string cstr>
  : MVE_VMOV_lane_base<oops, !con(iops, (ins MVEPairVectorIndex2:$idx,
                                             MVEPairVectorIndex0:$idx2)),
                       NoItinerary, "vmov", "", ops, cstr, []> {
  bits<5> Rt;
  bits<5> Rt2;
  bits<4> Qd;
  bit idx;
  bit idx2;

  let Inst{31-23} = 0b111011000;
  let Inst{22} = Qd{3};
  let Inst{21} = 0b0;
  let Inst{20} = to_qreg;
  let Inst{19-16} = Rt2{3-0};
  let Inst{15-13} = Qd{2-0};
  let Inst{12-5} = 0b01111000;
  let Inst{4} = idx2;
  let Inst{3-0} = Rt{3-0};
}

// The assembly syntax for these instructions mentions the vector
// register name twice, e.g.
//
//    vmov q2[2], q2[0], r0, r1
//    vmov r0, r1, q2[2], q2[0]
//
// which needs a bit of juggling with MC operand handling.
//
// For the move _into_ a vector register, the MC operand list also has
// to mention the register name twice: once as the output, and once as
// an extra input to represent where the unchanged half of the output
// register comes from (when this instruction is used in code
// generation). So we arrange that the first mention of the vector reg
// in the instruction is considered by the AsmMatcher to be the output
// ($Qd), and the second one is the input ($QdSrc). Binding them
// together with the existing 'tie' constraint is enough to enforce at
// register allocation time that they have to be the same register.
//
// For the move _from_ a vector register, there's no way to get round
// the fact that both instances of that register name have to be
// inputs. They have to be the same register again, but this time, we
// can't use a tie constraint, because that has to be between an
// output and an input operand. So this time, we have to arrange that
// the q-reg appears just once in the MC operand list, in spite of
// being mentioned twice in the asm syntax - which needs a custom
// AsmMatchConverter.

def MVE_VMOV_q_rr : MVE_VMOV_64bit<(outs MQPR:$Qd),
                                   (ins MQPR:$QdSrc, rGPR:$Rt, rGPR:$Rt2),
                                   0b1, "$Qd$idx, $QdSrc$idx2, $Rt, $Rt2",
                                   "$Qd = $QdSrc"> {
  let DecoderMethod = "DecodeMVEVMOVDRegtoQ";
}

def MVE_VMOV_rr_q : MVE_VMOV_64bit<(outs rGPR:$Rt, rGPR:$Rt2), (ins MQPR:$Qd),
                                   0b0, "$Rt, $Rt2, $Qd$idx, $Qd$idx2", ""> {
  let DecoderMethod = "DecodeMVEVMOVQtoDReg";
  let AsmMatchConverter = "cvtMVEVMOVQtoDReg";
}

// end of coproc mov

// start of MVE interleaving load/store

// Base class for the family of interleaving/deinterleaving
// load/stores with names like VLD20.8 and VST43.32.
class MVE_vldst24_base<bit writeback, bit fourregs, bits<2> stage, bits<2> size,
                       bit load, dag Oops, dag loadIops, dag wbIops,
                       string iname, string ops,
                       string cstr, list<dag> pattern=[]>
  : MVE_MI<Oops, !con(loadIops, wbIops), NoItinerary, iname, ops, cstr, pattern> {
  bits<4> VQd;
  bits<4> Rn;

  let Inst{31-22} = 0b1111110010;
  let Inst{21} = writeback;
  let Inst{20} = load;
  let Inst{19-16} = Rn;
  let Inst{15-13} = VQd{2-0};
  let Inst{12-9} = 0b1111;
  let Inst{8-7} = size;
  let Inst{6-5} = stage;
  let Inst{4-1} = 0b0000;
  let Inst{0} = fourregs;

  let mayLoad = load;
  let mayStore = !eq(load,0);
}

// A parameter class used to encapsulate all the ways the writeback
// variants of VLD20 and friends differ from the non-writeback ones.
class MVE_vldst24_writeback<bit b, dag Oo, dag Io,
                            string sy="", string c="", string n=""> {
  bit writeback = b;
  dag Oops = Oo;
  dag Iops = Io;
  string syntax = sy;
  string cstr = c;
  string id_suffix = n;
}

// Another parameter class that encapsulates the differences between VLD2x
// and VLD4x.
class MVE_vldst24_nvecs<int n, list<int> s, bit b, RegisterOperand vl> {
  int nvecs = n;
  list<int> stages = s;
  bit bit0 = b;
  RegisterOperand VecList = vl;
}

// A third parameter class that distinguishes VLDnn.8 from .16 from .32.
class MVE_vldst24_lanesize<int i, bits<2> b> {
  int lanesize = i;
  bits<2> sizebits = b;
}

// A base class for each direction of transfer: one for load, one for
// store. I can't make these a fourth independent parametric tuple
// class, because they have to take the nvecs tuple class as a
// parameter, in order to find the right VecList operand type.

class MVE_vld24_base<MVE_vldst24_nvecs n, bits<2> pat, bits<2> size,
                     MVE_vldst24_writeback wb, string iname,
                     list<dag> pattern=[]>
  : MVE_vldst24_base<wb.writeback, n.bit0, pat, size, 1,
                     !con((outs n.VecList:$VQd), wb.Oops),
                     (ins n.VecList:$VQdSrc), wb.Iops,
                     iname, "$VQd, $Rn" # wb.syntax,
                     wb.cstr # ",$VQdSrc = $VQd", pattern>;

class MVE_vst24_base<MVE_vldst24_nvecs n, bits<2> pat, bits<2> size,
                     MVE_vldst24_writeback wb, string iname,
                     list<dag> pattern=[]>
  : MVE_vldst24_base<wb.writeback, n.bit0, pat, size, 0,
                     wb.Oops, (ins n.VecList:$VQd), wb.Iops,
                     iname, "$VQd, $Rn" # wb.syntax,
                     wb.cstr, pattern>;

// Actually define all the interleaving loads and stores, by a series
// of nested foreaches over number of vectors (VLD2/VLD4); stage
// within one of those series (VLDx0/VLDx1/VLDx2/VLDx3); size of
// vector lane; writeback or no writeback.
foreach n = [MVE_vldst24_nvecs<2, [0,1],     0, VecList2Q>,
             MVE_vldst24_nvecs<4, [0,1,2,3], 1, VecList4Q>] in
foreach stage = n.stages in
foreach s = [MVE_vldst24_lanesize< 8, 0b00>,
             MVE_vldst24_lanesize<16, 0b01>,
             MVE_vldst24_lanesize<32, 0b10>] in
foreach wb = [MVE_vldst24_writeback<
                1, (outs rGPR:$wb), (ins t2_nosp_addr_offset_none:$Rn),
                "!", "$Rn.base = $wb", "_wb">,
              MVE_vldst24_writeback<0, (outs), (ins t2_addr_offset_none:$Rn)>] in {

  // For each case within all of those foreaches, define the actual
  // instructions. The def names are made by gluing together pieces
  // from all the parameter classes, and will end up being things like
  // MVE_VLD20_8 and MVE_VST43_16_wb.

  def "MVE_VLD" # n.nvecs # stage # "_" # s.lanesize # wb.id_suffix
    : MVE_vld24_base<n, stage, s.sizebits, wb,
                     "vld" # n.nvecs # stage # "." # s.lanesize>;

  def "MVE_VST" # n.nvecs # stage # "_" # s.lanesize # wb.id_suffix
    : MVE_vst24_base<n, stage, s.sizebits, wb,
                     "vst" # n.nvecs # stage # "." # s.lanesize>;
}

multiclass MVE_vst24_patterns<int lanesize, ValueType VT> {
  foreach stage = [0,1] in
    def : Pat<(int_arm_mve_vst2q i32:$addr,
               (VT MQPR:$v0), (VT MQPR:$v1), (i32 stage)),
              (!cast<Instruction>("MVE_VST2"#stage#"_"#lanesize)
               (REG_SEQUENCE QQPR, VT:$v0, qsub_0, VT:$v1, qsub_1),
               t2_addr_offset_none:$addr)>;

  foreach stage = [0,1,2,3] in
    def : Pat<(int_arm_mve_vst4q i32:$addr,
               (VT MQPR:$v0), (VT MQPR:$v1),
               (VT MQPR:$v2), (VT MQPR:$v3), (i32 stage)),
              (!cast<Instruction>("MVE_VST4"#stage#"_"#lanesize)
               (REG_SEQUENCE QQQQPR, VT:$v0, qsub_0, VT:$v1, qsub_1,
                                     VT:$v2, qsub_2, VT:$v3, qsub_3),
               t2_addr_offset_none:$addr)>;
}
defm : MVE_vst24_patterns<8, v16i8>;
defm : MVE_vst24_patterns<16, v8i16>;
defm : MVE_vst24_patterns<32, v4i32>;
defm : MVE_vst24_patterns<16, v8f16>;
defm : MVE_vst24_patterns<32, v4f32>;

// end of MVE interleaving load/store

// start of MVE predicable load/store

// A parameter class for the direction of transfer.
class MVE_ldst_direction<bit b, dag Oo, dag Io, string c=""> {
  bit load = b;
  dag Oops = Oo;
  dag Iops = Io;
  string cstr = c;
}
def MVE_ld: MVE_ldst_direction<1, (outs MQPR:$Qd), (ins), ",@earlyclobber $Qd">;
def MVE_st: MVE_ldst_direction<0, (outs), (ins MQPR:$Qd)>;

// A parameter class for the size of memory access in a load.
class MVE_memsz<bits<2> e, int s, AddrMode m, string mn, list<string> types> {
  bits<2> encoding = e;         // opcode bit(s) for encoding
  int shift = s;                // shift applied to immediate load offset
  AddrMode AM = m;

  // For instruction aliases: define the complete list of type
  // suffixes at this size, and the canonical ones for loads and
  // stores.
  string MnemonicLetter = mn;
  int TypeBits = !shl(8, s);
  string CanonLoadSuffix = ".u" # TypeBits;
  string CanonStoreSuffix = "." # TypeBits;
  list<string> suffixes = !foreach(letter, types, "." # letter # TypeBits);
}

// Instances of MVE_memsz.
//
// (memD doesn't need an AddrMode, because those are only for
// contiguous loads, and memD is only used by gather/scatters.)
def MVE_memB: MVE_memsz<0b00, 0, AddrModeT2_i7,   "b", ["", "u", "s"]>;
def MVE_memH: MVE_memsz<0b01, 1, AddrModeT2_i7s2, "h", ["", "u", "s", "f"]>;
def MVE_memW: MVE_memsz<0b10, 2, AddrModeT2_i7s4, "w", ["", "u", "s", "f"]>;
def MVE_memD: MVE_memsz<0b11, 3, ?,               "d", ["", "u", "s", "f"]>;

// This is the base class for all the MVE loads and stores other than
// the interleaving ones. All the non-interleaving loads/stores share
// the characteristic that they operate on just one vector register,
// so they are VPT-predicable.
//
// The predication operand is vpred_n, for both loads and stores. For
// store instructions, the reason is obvious: if there is no output
// register, there can't be a need for an input parameter giving the
// output register's previous value. Load instructions also don't need
// that input parameter, because unlike MVE data processing
// instructions, predicated loads are defined to set the inactive
// lanes of the output register to zero, instead of preserving their
// input values.
class MVE_VLDRSTR_base<MVE_ldst_direction dir, bit U, bit P, bit W, bit opc,
                       dag oops, dag iops, string asm, string suffix,
                       string ops, string cstr, list<dag> pattern=[]>
 : MVE_p<oops, iops, NoItinerary, asm, suffix, ops, vpred_n, cstr, pattern> {
  bits<3> Qd;

  let Inst{28} = U;
  let Inst{25} = 0b0;
  let Inst{24} = P;
  let Inst{22} = 0b0;
  let Inst{21} = W;
  let Inst{20} = dir.load;
  let Inst{15-13} = Qd{2-0};
  let Inst{12} = opc;
  let Inst{11-9} = 0b111;

  let mayLoad = dir.load;
  let mayStore = !eq(dir.load,0);
  let validForTailPredication = 1;
}

// Contiguous load and store instructions. These come in two main
// categories: same-size loads/stores in which 128 bits of vector
// register is transferred to or from 128 bits of memory in the most
// obvious way, and widening loads / narrowing stores, in which the
// size of memory accessed is less than the size of a vector register,
// so the load instructions sign- or zero-extend each memory value
// into a wider vector lane, and the store instructions truncate
// correspondingly.
//
// The instruction mnemonics for these two classes look reasonably
// similar, but the actual encodings are different enough to need two
// separate base classes.

// Contiguous, same size
class MVE_VLDRSTR_cs<MVE_ldst_direction dir, MVE_memsz memsz, bit P, bit W,
                     dag oops, dag iops, string asm, string suffix,
                     IndexMode im, string ops, string cstr>
  : MVE_VLDRSTR_base<dir, 0, P, W, 1, oops, iops, asm, suffix, ops, cstr> {
  bits<12> addr;
  let Inst{23} = addr{7};
  let Inst{19-16} = addr{11-8};
  let Inst{8-7} = memsz.encoding;
  let Inst{6-0} = addr{6-0};
}

// Contiguous, widening/narrowing
class MVE_VLDRSTR_cw<MVE_ldst_direction dir, MVE_memsz memsz, bit U,
                     bit P, bit W, bits<2> size, dag oops, dag iops,
                     string asm, string suffix, IndexMode im,
                     string ops, string cstr>
  : MVE_VLDRSTR_base<dir, U, P, W, 0, oops, iops, asm, suffix, ops, cstr> {
  bits<11> addr;
  let Inst{23} = addr{7};
  let Inst{19} = memsz.encoding{0}; // enough to tell 16- from 32-bit
  let Inst{18-16} = addr{10-8};
  let Inst{8-7} = size;
  let Inst{6-0} = addr{6-0};

  let IM = im;
}

// Multiclass wrapper on each of the _cw and _cs base classes, to
// generate three writeback modes (none, preindex, postindex).

multiclass MVE_VLDRSTR_cw_m<MVE_ldst_direction dir, MVE_memsz memsz,
                            string asm, string suffix, bit U, bits<2> size> {
  let AM = memsz.AM in {
    def "" : MVE_VLDRSTR_cw<
        dir, memsz, U, 1, 0, size,
        dir.Oops, !con(dir.Iops, (ins taddrmode_imm7<memsz.shift>:$addr)),
        asm, suffix, IndexModeNone, "$Qd, $addr", "">;

    def _pre : MVE_VLDRSTR_cw<
        dir, memsz, U, 1, 1, size,
        !con((outs tGPR:$wb), dir.Oops),
        !con(dir.Iops, (ins taddrmode_imm7<memsz.shift>:$addr)),
        asm, suffix, IndexModePre, "$Qd, $addr!", "$addr.base = $wb"> {
      let DecoderMethod = "DecodeMVE_MEM_1_pre<"#memsz.shift#">";
    }

    def _post : MVE_VLDRSTR_cw<
        dir, memsz, U, 0, 1, size,
        !con((outs tGPR:$wb), dir.Oops),
        !con(dir.Iops, (ins t_addr_offset_none:$Rn,
                            t2am_imm7_offset<memsz.shift>:$addr)),
        asm, suffix, IndexModePost, "$Qd, $Rn$addr", "$Rn.base = $wb"> {
      bits<4> Rn;
      let Inst{18-16} = Rn{2-0};
    }
  }
}

multiclass MVE_VLDRSTR_cs_m<MVE_ldst_direction dir, MVE_memsz memsz,
                            string asm, string suffix> {
  let AM = memsz.AM in {
    def "" : MVE_VLDRSTR_cs<
        dir, memsz, 1, 0,
        dir.Oops, !con(dir.Iops, (ins t2addrmode_imm7<memsz.shift>:$addr)),
        asm, suffix, IndexModeNone, "$Qd, $addr", "">;

    def _pre : MVE_VLDRSTR_cs<
        dir, memsz, 1, 1,
        !con((outs rGPR:$wb), dir.Oops),
        !con(dir.Iops, (ins t2addrmode_imm7_pre<memsz.shift>:$addr)),
        asm, suffix, IndexModePre, "$Qd, $addr!", "$addr.base = $wb"> {
      let DecoderMethod = "DecodeMVE_MEM_2_pre<"#memsz.shift#">";
    }

    def _post : MVE_VLDRSTR_cs<
        dir, memsz, 0, 1,
        !con((outs rGPR:$wb), dir.Oops),
        // We need an !if here to select the base register class,
        // because it's legal to write back to SP in a load of this
        // type, but not in a store.
        !con(dir.Iops, (ins !if(dir.load, t2_addr_offset_none,
                                          t2_nosp_addr_offset_none):$Rn,
                            t2am_imm7_offset<memsz.shift>:$addr)),
        asm, suffix, IndexModePost, "$Qd, $Rn$addr", "$Rn.base = $wb"> {
      bits<4> Rn;
      let Inst{19-16} = Rn{3-0};
    }
  }
}

// Now actually declare all the contiguous load/stores, via those
// multiclasses. The instruction ids coming out of this are the bare
// names shown in the defm, with _pre or _post appended for writeback,
// e.g. MVE_VLDRBS16, MVE_VSTRB16_pre, MVE_VSTRHU16_post.

defm MVE_VLDRBS16: MVE_VLDRSTR_cw_m<MVE_ld, MVE_memB, "vldrb", "s16", 0, 0b01>;
defm MVE_VLDRBS32: MVE_VLDRSTR_cw_m<MVE_ld, MVE_memB, "vldrb", "s32", 0, 0b10>;
defm MVE_VLDRBU16: MVE_VLDRSTR_cw_m<MVE_ld, MVE_memB, "vldrb", "u16", 1, 0b01>;
defm MVE_VLDRBU32: MVE_VLDRSTR_cw_m<MVE_ld, MVE_memB, "vldrb", "u32", 1, 0b10>;
defm MVE_VLDRHS32: MVE_VLDRSTR_cw_m<MVE_ld, MVE_memH, "vldrh", "s32", 0, 0b10>;
defm MVE_VLDRHU32: MVE_VLDRSTR_cw_m<MVE_ld, MVE_memH, "vldrh", "u32", 1, 0b10>;

defm MVE_VLDRBU8:  MVE_VLDRSTR_cs_m<MVE_ld, MVE_memB, "vldrb", "u8">;
defm MVE_VLDRHU16: MVE_VLDRSTR_cs_m<MVE_ld, MVE_memH, "vldrh", "u16">;
defm MVE_VLDRWU32: MVE_VLDRSTR_cs_m<MVE_ld, MVE_memW, "vldrw", "u32">;

defm MVE_VSTRB16:  MVE_VLDRSTR_cw_m<MVE_st, MVE_memB, "vstrb", "16",  0, 0b01>;
defm MVE_VSTRB32:  MVE_VLDRSTR_cw_m<MVE_st, MVE_memB, "vstrb", "32",  0, 0b10>;
defm MVE_VSTRH32:  MVE_VLDRSTR_cw_m<MVE_st, MVE_memH, "vstrh", "32",  0, 0b10>;

defm MVE_VSTRBU8 : MVE_VLDRSTR_cs_m<MVE_st, MVE_memB, "vstrb", "8">;
defm MVE_VSTRHU16: MVE_VLDRSTR_cs_m<MVE_st, MVE_memH, "vstrh", "16">;
defm MVE_VSTRWU32: MVE_VLDRSTR_cs_m<MVE_st, MVE_memW, "vstrw", "32">;

// Gather loads / scatter stores whose address operand is of the form
// [Rn,Qm], i.e. a single GPR as the common base address, plus a
// vector of offset from it. ('Load/store this sequence of elements of
// the same array.')
//
// Like the contiguous family, these loads and stores can widen the
// loaded values / truncate the stored ones, or they can just
// load/store the same size of memory and vector lane. But unlike the
// contiguous family, there's no particular difference in encoding
// between those two cases.
//
// This family also comes with the option to scale the offset values
// in Qm by the size of the loaded memory (i.e. to treat them as array
// indices), or not to scale them (to treat them as plain byte offsets
// in memory, so that perhaps the loaded values are unaligned). The
// scaled instructions' address operand in assembly looks like
// [Rn,Qm,UXTW #2] or similar.

// Base class.
class MVE_VLDRSTR_rq<MVE_ldst_direction dir, MVE_memsz memsz, bit U,
                     bits<2> size, bit os, string asm, string suffix, int shift>
  : MVE_VLDRSTR_base<dir, U, 0b0, 0b0, 0, dir.Oops,
                     !con(dir.Iops, (ins mve_addr_rq_shift<shift>:$addr)),
                     asm, suffix, "$Qd, $addr", dir.cstr> {
  bits<7> addr;
  let Inst{23} = 0b1;
  let Inst{19-16} = addr{6-3};
  let Inst{8-7} = size;
  let Inst{6} = memsz.encoding{1};
  let Inst{5} = 0;
  let Inst{4} = memsz.encoding{0};
  let Inst{3-1} = addr{2-0};
  let Inst{0} = os;
}

// Multiclass that defines the scaled and unscaled versions of an
// instruction, when the memory size is wider than a byte. The scaled
// version gets the default name like MVE_VLDRBU16_rq; the unscaled /
// potentially unaligned version gets a "_u" suffix, e.g.
// MVE_VLDRBU16_rq_u.
multiclass MVE_VLDRSTR_rq_w<MVE_ldst_direction dir, MVE_memsz memsz,
                            string asm, string suffix, bit U, bits<2> size> {
  def _u : MVE_VLDRSTR_rq<dir, memsz, U, size, 0, asm, suffix, 0>;
  def "" : MVE_VLDRSTR_rq<dir, memsz, U, size, 1, asm, suffix, memsz.shift>;
}

// Subclass of MVE_VLDRSTR_rq with the same API as that multiclass,
// for use when the memory size is one byte, so there's no 'scaled'
// version of the instruction at all. (This is encoded as if it were
// unscaled, but named in the default way with no _u suffix.)
class MVE_VLDRSTR_rq_b<MVE_ldst_direction dir, MVE_memsz memsz,
                       string asm, string suffix, bit U, bits<2> size>
  : MVE_VLDRSTR_rq<dir, memsz, U, size, 0, asm, suffix, 0>;

// Multiclasses wrapping that to add ISel patterns for intrinsics.
multiclass MVE_VLDR_rq_w<MVE_memsz memsz, list<MVEVectorVTInfo> VTIs> {
  defm "": MVE_VLDRSTR_rq_w<MVE_ld, memsz, "vldr" # memsz.MnemonicLetter,
                            VTIs[0].Suffix, VTIs[0].Unsigned, VTIs[0].Size>;
  foreach VTI = VTIs in
  foreach UnsignedFlag = !if(!eq(VTI.Size, memsz.encoding),
                             [0,1], [VTI.Unsigned]) in {
    def : Pat<(VTI.Vec (int_arm_mve_vldr_gather_offset GPR:$base, (VTIs[0].Vec MQPR:$offsets), memsz.TypeBits, 0, UnsignedFlag)),
              (VTI.Vec (!cast<Instruction>(NAME#"_u") GPR:$base, MQPR:$offsets))>;
    def : Pat<(VTI.Vec (int_arm_mve_vldr_gather_offset GPR:$base, (VTIs[0].Vec MQPR:$offsets), memsz.TypeBits, memsz.shift, UnsignedFlag)),
              (VTI.Vec (!cast<Instruction>(NAME) GPR:$base, MQPR:$offsets))>;
    def : Pat<(VTI.Vec (int_arm_mve_vldr_gather_offset_predicated GPR:$base, (VTIs[0].Vec MQPR:$offsets), memsz.TypeBits, 0, UnsignedFlag, (VTI.Pred VCCR:$pred))),
              (VTI.Vec (!cast<Instruction>(NAME#"_u") GPR:$base, MQPR:$offsets, 1, VCCR:$pred))>;
    def : Pat<(VTI.Vec (int_arm_mve_vldr_gather_offset_predicated GPR:$base, (VTIs[0].Vec MQPR:$offsets), memsz.TypeBits, memsz.shift, UnsignedFlag, (VTI.Pred VCCR:$pred))),
              (VTI.Vec (!cast<Instruction>(NAME) GPR:$base, MQPR:$offsets, 1, VCCR:$pred))>;
  }
}
multiclass MVE_VLDR_rq_b<list<MVEVectorVTInfo> VTIs> {
  def "": MVE_VLDRSTR_rq_b<MVE_ld, MVE_memB, "vldrb",
                           VTIs[0].Suffix, VTIs[0].Unsigned, VTIs[0].Size>;
  foreach VTI = VTIs in {
    def : Pat<(VTI.Vec (int_arm_mve_vldr_gather_offset GPR:$base, (VTIs[0].Vec MQPR:$offsets), 8, 0, VTI.Unsigned)),
              (VTI.Vec (!cast<Instruction>(NAME) GPR:$base, MQPR:$offsets))>;
    def : Pat<(VTI.Vec (int_arm_mve_vldr_gather_offset_predicated GPR:$base, (VTIs[0].Vec MQPR:$offsets), 8, 0, VTI.Unsigned, (VTI.Pred VCCR:$pred))),
              (VTI.Vec (!cast<Instruction>(NAME) GPR:$base, MQPR:$offsets, 1, VCCR:$pred))>;
  }
}
multiclass MVE_VSTR_rq_w<MVE_memsz memsz, list<MVEVectorVTInfo> VTIs> {
  defm "": MVE_VLDRSTR_rq_w<MVE_st, memsz, "vstr" # memsz.MnemonicLetter,
                            VTIs[0].BitsSuffix, 0, VTIs[0].Size>;
  foreach VTI = VTIs in {
    def : Pat<(int_arm_mve_vstr_scatter_offset GPR:$base, (VTIs[0].Vec MQPR:$offsets), (VTI.Vec MQPR:$data), memsz.TypeBits, 0),
              (!cast<Instruction>(NAME#"_u") MQPR:$data, GPR:$base, MQPR:$offsets)>;
    def : Pat<(int_arm_mve_vstr_scatter_offset GPR:$base, (VTIs[0].Vec MQPR:$offsets), (VTI.Vec MQPR:$data), memsz.TypeBits, memsz.shift),
              (!cast<Instruction>(NAME) MQPR:$data, GPR:$base, MQPR:$offsets)>;
    def : Pat<(int_arm_mve_vstr_scatter_offset_predicated GPR:$base, (VTIs[0].Vec MQPR:$offsets), (VTI.Vec MQPR:$data), memsz.TypeBits, 0, (VTI.Pred VCCR:$pred)),
              (!cast<Instruction>(NAME#"_u") MQPR:$data, GPR:$base, MQPR:$offsets, 1, VCCR:$pred)>;
    def : Pat<(int_arm_mve_vstr_scatter_offset_predicated GPR:$base, (VTIs[0].Vec MQPR:$offsets), (VTI.Vec MQPR:$data), memsz.TypeBits, memsz.shift, (VTI.Pred VCCR:$pred)),
              (!cast<Instruction>(NAME) MQPR:$data, GPR:$base, MQPR:$offsets, 1, VCCR:$pred)>;
  }
}
multiclass MVE_VSTR_rq_b<list<MVEVectorVTInfo> VTIs> {
  def "": MVE_VLDRSTR_rq_b<MVE_st, MVE_memB, "vstrb",
                           VTIs[0].BitsSuffix, 0, VTIs[0].Size>;
  foreach VTI = VTIs in {
    def : Pat<(int_arm_mve_vstr_scatter_offset GPR:$base, (VTIs[0].Vec MQPR:$offsets), (VTI.Vec MQPR:$data), 8, 0),
              (!cast<Instruction>(NAME) MQPR:$data, GPR:$base, MQPR:$offsets)>;
    def : Pat<(int_arm_mve_vstr_scatter_offset_predicated GPR:$base, (VTIs[0].Vec MQPR:$offsets), (VTI.Vec MQPR:$data), 8, 0, (VTI.Pred VCCR:$pred)),
              (!cast<Instruction>(NAME) MQPR:$data, GPR:$base, MQPR:$offsets, 1, VCCR:$pred)>;
  }
}

// Actually define all the loads and stores in this family.

defm MVE_VLDRBU8_rq : MVE_VLDR_rq_b<[MVE_v16u8,MVE_v16s8]>;
defm MVE_VLDRBU16_rq: MVE_VLDR_rq_b<[MVE_v8u16]>;
defm MVE_VLDRBS16_rq: MVE_VLDR_rq_b<[MVE_v8s16]>;
defm MVE_VLDRBU32_rq: MVE_VLDR_rq_b<[MVE_v4u32]>;
defm MVE_VLDRBS32_rq: MVE_VLDR_rq_b<[MVE_v4s32]>;

defm MVE_VLDRHU16_rq: MVE_VLDR_rq_w<MVE_memH, [MVE_v8u16,MVE_v8s16,MVE_v8f16]>;
defm MVE_VLDRHU32_rq: MVE_VLDR_rq_w<MVE_memH, [MVE_v4u32]>;
defm MVE_VLDRHS32_rq: MVE_VLDR_rq_w<MVE_memH, [MVE_v4s32]>;
defm MVE_VLDRWU32_rq: MVE_VLDR_rq_w<MVE_memW, [MVE_v4u32,MVE_v4s32,MVE_v4f32]>;
defm MVE_VLDRDU64_rq: MVE_VLDR_rq_w<MVE_memD, [MVE_v2u64,MVE_v2s64]>;

defm MVE_VSTRB8_rq  : MVE_VSTR_rq_b<[MVE_v16i8]>;
defm MVE_VSTRB16_rq : MVE_VSTR_rq_b<[MVE_v8i16]>;
defm MVE_VSTRB32_rq : MVE_VSTR_rq_b<[MVE_v4i32]>;

defm MVE_VSTRH16_rq : MVE_VSTR_rq_w<MVE_memH, [MVE_v8i16,MVE_v8f16]>;
defm MVE_VSTRH32_rq : MVE_VSTR_rq_w<MVE_memH, [MVE_v4i32]>;
defm MVE_VSTRW32_rq : MVE_VSTR_rq_w<MVE_memW, [MVE_v4i32,MVE_v4f32]>;
defm MVE_VSTRD64_rq : MVE_VSTR_rq_w<MVE_memD, [MVE_v2i64]>;

// Gather loads / scatter stores whose address operand is of the form
// [Qm,#imm], i.e. a vector containing a full base address for each
// loaded item, plus an immediate offset applied consistently to all
// of them. ('Load/store the same field from this vector of pointers
// to a structure type.')
//
// This family requires the vector lane size to be at least 32 bits
// (so there's room for an address in each lane at all). It has no
// widening/narrowing variants. But it does support preindex
// writeback, in which the address vector is updated to hold the
// addresses actually loaded from.

// Base class.
class MVE_VLDRSTR_qi<MVE_ldst_direction dir, MVE_memsz memsz, bit W, dag wbops,
                     string asm, string wbAsm, string suffix, string cstr = "">
  : MVE_VLDRSTR_base<dir, 1, 1, W, 1, !con(wbops, dir.Oops),
                     !con(dir.Iops, (ins mve_addr_q_shift<memsz.shift>:$addr)),
                     asm, suffix, "$Qd, $addr" # wbAsm, cstr # dir.cstr> {
  bits<11> addr;
  let Inst{23} = addr{7};
  let Inst{19-17} = addr{10-8};
  let Inst{16} = 0;
  let Inst{8} = memsz.encoding{0}; // enough to distinguish 32- from 64-bit
  let Inst{7} = 0;
  let Inst{6-0} = addr{6-0};
}

// Multiclass that generates the non-writeback and writeback variants.
multiclass MVE_VLDRSTR_qi_m<MVE_ldst_direction dir, MVE_memsz memsz,
                            string asm, string suffix> {
  def ""   : MVE_VLDRSTR_qi<dir, memsz, 0, (outs),          asm, "",  suffix>;
  def _pre : MVE_VLDRSTR_qi<dir, memsz, 1, (outs MQPR:$wb), asm, "!", suffix,
                            "$addr.base = $wb"> {
    let DecoderMethod="DecodeMVE_MEM_3_pre<"#memsz.shift#">";
  }
}

// Multiclasses wrapping that one, adding selection patterns for the
// non-writeback loads and all the stores. (The writeback loads must
// deliver multiple output values, so they have to be selected by C++
// code.)
multiclass MVE_VLDR_qi<MVE_memsz memsz, MVEVectorVTInfo AVTI,
                       list<MVEVectorVTInfo> DVTIs> {
  defm "" : MVE_VLDRSTR_qi_m<MVE_ld, memsz, "vldr" # memsz.MnemonicLetter,
                             "u" # memsz.TypeBits>;

  foreach DVTI = DVTIs in {
    def : Pat<(DVTI.Vec (int_arm_mve_vldr_gather_base
                 (AVTI.Vec MQPR:$addr), (i32 imm:$offset))),
              (DVTI.Vec (!cast<Instruction>(NAME)
                 (AVTI.Vec MQPR:$addr), (i32 imm:$offset)))>;
    def : Pat<(DVTI.Vec (int_arm_mve_vldr_gather_base_predicated
                 (AVTI.Vec MQPR:$addr), (i32 imm:$offset), (AVTI.Pred VCCR:$pred))),
              (DVTI.Vec (!cast<Instruction>(NAME)
                 (AVTI.Vec MQPR:$addr), (i32 imm:$offset), 1, VCCR:$pred))>;
  }
}
multiclass MVE_VSTR_qi<MVE_memsz memsz, MVEVectorVTInfo AVTI,
                       list<MVEVectorVTInfo> DVTIs> {
  defm "" : MVE_VLDRSTR_qi_m<MVE_st, memsz, "vstr" # memsz.MnemonicLetter,
                             !cast<string>(memsz.TypeBits)>;

  foreach DVTI = DVTIs in {
    def : Pat<(int_arm_mve_vstr_scatter_base
                (AVTI.Vec MQPR:$addr), (i32 imm:$offset), (DVTI.Vec MQPR:$data)),
              (!cast<Instruction>(NAME)
               (DVTI.Vec MQPR:$data), (AVTI.Vec MQPR:$addr), (i32 imm:$offset))>;
    def : Pat<(int_arm_mve_vstr_scatter_base_predicated
                (AVTI.Vec MQPR:$addr), (i32 imm:$offset), (DVTI.Vec MQPR:$data), (AVTI.Pred VCCR:$pred)),
              (!cast<Instruction>(NAME)
                (DVTI.Vec MQPR:$data), (AVTI.Vec MQPR:$addr), (i32 imm:$offset), 1, VCCR:$pred)>;
    def : Pat<(AVTI.Vec (int_arm_mve_vstr_scatter_base_wb
                (AVTI.Vec MQPR:$addr), (i32 imm:$offset), (DVTI.Vec MQPR:$data))),
              (AVTI.Vec (!cast<Instruction>(NAME # "_pre")
                 (DVTI.Vec MQPR:$data), (AVTI.Vec MQPR:$addr), (i32 imm:$offset)))>;
    def : Pat<(AVTI.Vec (int_arm_mve_vstr_scatter_base_wb_predicated
                (AVTI.Vec MQPR:$addr), (i32 imm:$offset), (DVTI.Vec MQPR:$data), (AVTI.Pred VCCR:$pred))),
              (AVTI.Vec (!cast<Instruction>(NAME # "_pre")
                 (DVTI.Vec MQPR:$data), (AVTI.Vec MQPR:$addr), (i32 imm:$offset), 1, VCCR:$pred))>;
  }
}

// Actual instruction definitions.
defm MVE_VLDRWU32_qi: MVE_VLDR_qi<MVE_memW, MVE_v4i32, [MVE_v4i32,MVE_v4f32]>;
defm MVE_VLDRDU64_qi: MVE_VLDR_qi<MVE_memD, MVE_v2i64, [MVE_v2i64,MVE_v2f64]>;
defm MVE_VSTRW32_qi:  MVE_VSTR_qi<MVE_memW, MVE_v4i32, [MVE_v4i32,MVE_v4f32]>;
defm MVE_VSTRD64_qi:  MVE_VSTR_qi<MVE_memD, MVE_v2i64, [MVE_v2i64,MVE_v2f64]>;

// Define aliases for all the instructions where memory size and
// vector lane size are the same. These are mnemonic aliases, so they
// apply consistently across all of the above families - contiguous
// loads, and both the rq and qi types of gather/scatter.
//
// Rationale: As long as you're loading (for example) 16-bit memory
// values into 16-bit vector lanes, you can think of them as signed or
// unsigned integers, fp16 or just raw 16-bit blobs and it makes no
// difference. So we permit all of vldrh.16, vldrh.u16, vldrh.s16,
// vldrh.f16 and treat them all as equivalent to the canonical
// spelling (which happens to be .u16 for loads, and just .16 for
// stores).

foreach vpt_cond = ["", "t", "e"] in
foreach memsz = [MVE_memB, MVE_memH, MVE_memW, MVE_memD] in
foreach suffix = memsz.suffixes in {

  // These foreaches are conceptually ifs, implemented by iterating a
  // dummy variable over a list with 0 or 1 elements depending on the
  // condition. The idea is to iterate over _nearly_ all the suffixes
  // in memsz.suffixes, but omit the one we want all the others to alias.

  foreach _ = !if(!ne(suffix, memsz.CanonLoadSuffix), [1], []<int>) in
  def : MnemonicAlias<
    "vldr" # memsz.MnemonicLetter # vpt_cond # suffix,
    "vldr" # memsz.MnemonicLetter # vpt_cond # memsz.CanonLoadSuffix>;

  foreach _ = !if(!ne(suffix, memsz.CanonStoreSuffix), [1], []<int>) in
  def : MnemonicAlias<
    "vstr" # memsz.MnemonicLetter # vpt_cond # suffix,
    "vstr" # memsz.MnemonicLetter # vpt_cond # memsz.CanonStoreSuffix>;
}

// end of MVE predicable load/store

class MVE_VPT<string suffix, bits<2> size, dag iops, string asm, list<dag> pattern=[]>
  : MVE_MI<(outs ), iops, NoItinerary, !strconcat("vpt", "${Mk}", ".", suffix), asm, "", pattern> {
  bits<3> fc;
  bits<4> Mk;
  bits<3> Qn;

  let Inst{31-23} = 0b111111100;
  let Inst{22} = Mk{3};
  let Inst{21-20} = size;
  let Inst{19-17} = Qn{2-0};
  let Inst{16} = 0b1;
  let Inst{15-13} = Mk{2-0};
  let Inst{12} = fc{2};
  let Inst{11-8} = 0b1111;
  let Inst{7} = fc{0};
  let Inst{4} = 0b0;

  let Defs = [VPR];
  let validForTailPredication = 1;
}

class MVE_VPTt1<string suffix, bits<2> size, dag iops>
  : MVE_VPT<suffix, size, iops, "$fc, $Qn, $Qm"> {
  bits<4> Qm;
  bits<4> Mk;

  let Inst{6} = 0b0;
  let Inst{5} = Qm{3};
  let Inst{3-1} = Qm{2-0};
  let Inst{0} = fc{1};
  let validForTailPredication = 1;
}

class MVE_VPTt1i<string suffix, bits<2> size>
 : MVE_VPTt1<suffix, size,
           (ins vpt_mask:$Mk, MQPR:$Qn, MQPR:$Qm, pred_basic_i:$fc)> {
  let Inst{12} = 0b0;
  let Inst{0} = 0b0;
}

def MVE_VPTv4i32 : MVE_VPTt1i<"i32", 0b10>;
def MVE_VPTv8i16 : MVE_VPTt1i<"i16", 0b01>;
def MVE_VPTv16i8 : MVE_VPTt1i<"i8", 0b00>;

class MVE_VPTt1u<string suffix, bits<2> size>
 : MVE_VPTt1<suffix, size,
           (ins vpt_mask:$Mk, MQPR:$Qn, MQPR:$Qm, pred_basic_u:$fc)> {
  let Inst{12} = 0b0;
  let Inst{0} = 0b1;
}

def MVE_VPTv4u32 : MVE_VPTt1u<"u32", 0b10>;
def MVE_VPTv8u16 : MVE_VPTt1u<"u16", 0b01>;
def MVE_VPTv16u8 : MVE_VPTt1u<"u8", 0b00>;

class MVE_VPTt1s<string suffix, bits<2> size>
 : MVE_VPTt1<suffix, size,
           (ins vpt_mask:$Mk, MQPR:$Qn, MQPR:$Qm, pred_basic_s:$fc)> {
  let Inst{12} = 0b1;
}

def MVE_VPTv4s32 : MVE_VPTt1s<"s32", 0b10>;
def MVE_VPTv8s16 : MVE_VPTt1s<"s16", 0b01>;
def MVE_VPTv16s8 : MVE_VPTt1s<"s8", 0b00>;

class MVE_VPTt2<string suffix, bits<2> size, dag iops>
  : MVE_VPT<suffix, size, iops,
          "$fc, $Qn, $Rm"> {
  bits<4> Rm;
  bits<3> fc;
  bits<4> Mk;

  let Inst{6} = 0b1;
  let Inst{5} = fc{1};
  let Inst{3-0} = Rm{3-0};
}

class MVE_VPTt2i<string suffix, bits<2> size>
  : MVE_VPTt2<suffix, size,
            (ins vpt_mask:$Mk, MQPR:$Qn, GPRwithZR:$Rm, pred_basic_i:$fc)> {
  let Inst{12} = 0b0;
  let Inst{5} = 0b0;
}

def MVE_VPTv4i32r : MVE_VPTt2i<"i32", 0b10>;
def MVE_VPTv8i16r : MVE_VPTt2i<"i16", 0b01>;
def MVE_VPTv16i8r : MVE_VPTt2i<"i8", 0b00>;

class MVE_VPTt2u<string suffix, bits<2> size>
  : MVE_VPTt2<suffix, size,
            (ins vpt_mask:$Mk, MQPR:$Qn, GPRwithZR:$Rm, pred_basic_u:$fc)> {
  let Inst{12} = 0b0;
  let Inst{5} = 0b1;
}

def MVE_VPTv4u32r : MVE_VPTt2u<"u32", 0b10>;
def MVE_VPTv8u16r : MVE_VPTt2u<"u16", 0b01>;
def MVE_VPTv16u8r : MVE_VPTt2u<"u8", 0b00>;

class MVE_VPTt2s<string suffix, bits<2> size>
  : MVE_VPTt2<suffix, size,
            (ins vpt_mask:$Mk, MQPR:$Qn, GPRwithZR:$Rm, pred_basic_s:$fc)> {
  let Inst{12} = 0b1;
}

def MVE_VPTv4s32r : MVE_VPTt2s<"s32", 0b10>;
def MVE_VPTv8s16r : MVE_VPTt2s<"s16", 0b01>;
def MVE_VPTv16s8r : MVE_VPTt2s<"s8", 0b00>;


class MVE_VPTf<string suffix, bit size, dag iops, string asm, list<dag> pattern=[]>
  : MVE_MI<(outs ), iops, NoItinerary, !strconcat("vpt", "${Mk}", ".", suffix), asm,
            "", pattern> {
  bits<3> fc;
  bits<4> Mk;
  bits<3> Qn;

  let Inst{31-29} = 0b111;
  let Inst{28} = size;
  let Inst{27-23} = 0b11100;
  let Inst{22} = Mk{3};
  let Inst{21-20} = 0b11;
  let Inst{19-17} = Qn{2-0};
  let Inst{16} = 0b1;
  let Inst{15-13} = Mk{2-0};
  let Inst{12} = fc{2};
  let Inst{11-8} = 0b1111;
  let Inst{7} = fc{0};
  let Inst{4} = 0b0;

  let Defs = [VPR];
  let Predicates = [HasMVEFloat];
  let validForTailPredication = 1;
}

class MVE_VPTft1<string suffix, bit size>
  : MVE_VPTf<suffix, size, (ins vpt_mask:$Mk, MQPR:$Qn, MQPR:$Qm, pred_basic_fp:$fc),
          "$fc, $Qn, $Qm"> {
  bits<3> fc;
  bits<4> Qm;

  let Inst{6} = 0b0;
  let Inst{5} = Qm{3};
  let Inst{3-1} = Qm{2-0};
  let Inst{0} = fc{1};
}

def MVE_VPTv4f32         : MVE_VPTft1<"f32", 0b0>;
def MVE_VPTv8f16         : MVE_VPTft1<"f16", 0b1>;

class MVE_VPTft2<string suffix, bit size>
  : MVE_VPTf<suffix, size, (ins vpt_mask:$Mk, MQPR:$Qn, GPRwithZR:$Rm, pred_basic_fp:$fc),
          "$fc, $Qn, $Rm"> {
  bits<3> fc;
  bits<4> Rm;

  let Inst{6} = 0b1;
  let Inst{5} = fc{1};
  let Inst{3-0} = Rm{3-0};
}

def MVE_VPTv4f32r        : MVE_VPTft2<"f32", 0b0>;
def MVE_VPTv8f16r        : MVE_VPTft2<"f16", 0b1>;

def MVE_VPST : MVE_MI<(outs ), (ins vpt_mask:$Mk), NoItinerary,
       !strconcat("vpst", "${Mk}"), "", "", []> {
  bits<4> Mk;

  let Inst{31-23} = 0b111111100;
  let Inst{22} = Mk{3};
  let Inst{21-16} = 0b110001;
  let Inst{15-13} = Mk{2-0};
  let Inst{12-0} = 0b0111101001101;
  let Unpredictable{12} = 0b1;
  let Unpredictable{7} = 0b1;
  let Unpredictable{5} = 0b1;

  let Uses = [VPR];
  let validForTailPredication = 1;
}

def MVE_VPSEL : MVE_p<(outs MQPR:$Qd), (ins MQPR:$Qn, MQPR:$Qm), NoItinerary,
                      "vpsel", "", "$Qd, $Qn, $Qm", vpred_n, "", []> {
  bits<4> Qn;
  bits<4> Qd;
  bits<4> Qm;

  let Inst{28} = 0b1;
  let Inst{25-23} = 0b100;
  let Inst{22} = Qd{3};
  let Inst{21-20} = 0b11;
  let Inst{19-17} = Qn{2-0};
  let Inst{16} = 0b1;
  let Inst{15-13} = Qd{2-0};
  let Inst{12-9} = 0b0111;
  let Inst{8} = 0b1;
  let Inst{7} = Qn{3};
  let Inst{6} = 0b0;
  let Inst{5} = Qm{3};
  let Inst{4} = 0b0;
  let Inst{3-1} = Qm{2-0};
  let Inst{0} = 0b1;
  let validForTailPredication = 1;
}

foreach suffix = ["s8", "s16", "s32", "u8", "u16", "u32",
                  "i8", "i16", "i32",       "f16", "f32"] in
def : MVEInstAlias<"vpsel${vp}." # suffix # "\t$Qd, $Qn, $Qm",
                   (MVE_VPSEL MQPR:$Qd, MQPR:$Qn, MQPR:$Qm, vpred_n:$vp)>;

let Predicates = [HasMVEInt] in {
  def : Pat<(v16i8 (vselect (v16i1 VCCR:$pred), (v16i8 MQPR:$v1), (v16i8 MQPR:$v2))),
            (v16i8 (MVE_VPSEL MQPR:$v1, MQPR:$v2, 0, VCCR:$pred))>;
  def : Pat<(v8i16 (vselect (v8i1 VCCR:$pred), (v8i16 MQPR:$v1), (v8i16 MQPR:$v2))),
            (v8i16 (MVE_VPSEL MQPR:$v1, MQPR:$v2, 0, VCCR:$pred))>;
  def : Pat<(v4i32 (vselect (v4i1 VCCR:$pred), (v4i32 MQPR:$v1), (v4i32 MQPR:$v2))),
            (v4i32 (MVE_VPSEL MQPR:$v1, MQPR:$v2, 0, VCCR:$pred))>;

  def : Pat<(v8f16 (vselect (v8i1 VCCR:$pred), (v8f16 MQPR:$v1), (v8f16 MQPR:$v2))),
            (v8f16 (MVE_VPSEL MQPR:$v1, MQPR:$v2, 0, VCCR:$pred))>;
  def : Pat<(v4f32 (vselect (v4i1 VCCR:$pred), (v4f32 MQPR:$v1), (v4f32 MQPR:$v2))),
            (v4f32 (MVE_VPSEL MQPR:$v1, MQPR:$v2, 0, VCCR:$pred))>;

  def : Pat<(v16i8 (vselect (v16i8 MQPR:$pred), (v16i8 MQPR:$v1), (v16i8 MQPR:$v2))),
            (v16i8 (MVE_VPSEL MQPR:$v1, MQPR:$v2, 0,
                              (MVE_VCMPi8 (v16i8 MQPR:$pred), (MVE_VMOVimmi8 0), 1)))>;
  def : Pat<(v8i16 (vselect (v8i16 MQPR:$pred), (v8i16 MQPR:$v1), (v8i16 MQPR:$v2))),
            (v8i16 (MVE_VPSEL MQPR:$v1, MQPR:$v2, 0,
                              (MVE_VCMPi16 (v8i16 MQPR:$pred), (MVE_VMOVimmi16 0), 1)))>;
  def : Pat<(v4i32 (vselect (v4i32 MQPR:$pred), (v4i32 MQPR:$v1), (v4i32 MQPR:$v2))),
            (v4i32 (MVE_VPSEL MQPR:$v1, MQPR:$v2, 0,
                              (MVE_VCMPi32 (v4i32 MQPR:$pred), (MVE_VMOVimmi32 0), 1)))>;

  def : Pat<(v8f16 (vselect (v8i16 MQPR:$pred), (v8f16 MQPR:$v1), (v8f16 MQPR:$v2))),
            (v8f16 (MVE_VPSEL MQPR:$v1, MQPR:$v2, 0,
                              (MVE_VCMPi16 (v8i16 MQPR:$pred), (MVE_VMOVimmi16 0), 1)))>;
  def : Pat<(v4f32 (vselect (v4i32 MQPR:$pred), (v4f32 MQPR:$v1), (v4f32 MQPR:$v2))),
            (v4f32 (MVE_VPSEL MQPR:$v1, MQPR:$v2, 0,
                              (MVE_VCMPi32 (v4i32 MQPR:$pred), (MVE_VMOVimmi32 0), 1)))>;

  // Pred <-> Int
  def : Pat<(v16i8 (zext  (v16i1 VCCR:$pred))),
            (v16i8 (MVE_VPSEL (MVE_VMOVimmi8 1), (MVE_VMOVimmi8 0), 0, VCCR:$pred))>;
  def : Pat<(v8i16 (zext  (v8i1  VCCR:$pred))),
            (v8i16 (MVE_VPSEL (MVE_VMOVimmi16 1), (MVE_VMOVimmi16 0), 0, VCCR:$pred))>;
  def : Pat<(v4i32 (zext  (v4i1  VCCR:$pred))),
            (v4i32 (MVE_VPSEL (MVE_VMOVimmi32 1), (MVE_VMOVimmi32 0), 0, VCCR:$pred))>;

  def : Pat<(v16i8 (sext  (v16i1 VCCR:$pred))),
            (v16i8 (MVE_VPSEL (MVE_VMOVimmi8 255), (MVE_VMOVimmi8 0), 0, VCCR:$pred))>;
  def : Pat<(v8i16 (sext  (v8i1  VCCR:$pred))),
            (v8i16 (MVE_VPSEL (MVE_VMOVimmi8 255), (MVE_VMOVimmi16 0), 0, VCCR:$pred))>;
  def : Pat<(v4i32 (sext  (v4i1  VCCR:$pred))),
            (v4i32 (MVE_VPSEL (MVE_VMOVimmi8 255), (MVE_VMOVimmi32 0), 0, VCCR:$pred))>;

  def : Pat<(v16i8 (anyext  (v16i1 VCCR:$pred))),
            (v16i8 (MVE_VPSEL (MVE_VMOVimmi8 1), (MVE_VMOVimmi8 0), 0, VCCR:$pred))>;
  def : Pat<(v8i16 (anyext  (v8i1  VCCR:$pred))),
            (v8i16 (MVE_VPSEL (MVE_VMOVimmi16 1), (MVE_VMOVimmi16 0), 0, VCCR:$pred))>;
  def : Pat<(v4i32 (anyext  (v4i1  VCCR:$pred))),
            (v4i32 (MVE_VPSEL (MVE_VMOVimmi32 1), (MVE_VMOVimmi32 0), 0, VCCR:$pred))>;

  def : Pat<(v16i1 (trunc (v16i8 MQPR:$v1))),
            (v16i1 (MVE_VCMPi32r (v16i8 MQPR:$v1), ZR, 1))>;
  def : Pat<(v8i1 (trunc (v8i16  MQPR:$v1))),
            (v8i1 (MVE_VCMPi32r (v8i16 MQPR:$v1), ZR, 1))>;
  def : Pat<(v4i1 (trunc (v4i32  MQPR:$v1))),
            (v4i1 (MVE_VCMPi32r (v4i32 MQPR:$v1), ZR, 1))>;
}

let Predicates = [HasMVEFloat] in {
  // Pred <-> Float
  // 112 is 1.0 in float
  def : Pat<(v4f32 (uint_to_fp (v4i1 VCCR:$pred))),
            (v4f32 (MVE_VPSEL (v4f32 (MVE_VMOVimmf32 112)), (v4f32 (MVE_VMOVimmi32 0)), 0, VCCR:$pred))>;
  // 2620 in 1.0 in half
  def : Pat<(v8f16 (uint_to_fp (v8i1 VCCR:$pred))),
            (v8f16 (MVE_VPSEL (v8f16 (MVE_VMOVimmi16 2620)), (v8f16 (MVE_VMOVimmi16 0)), 0, VCCR:$pred))>;
  // 240 is -1.0 in float
  def : Pat<(v4f32 (sint_to_fp (v4i1 VCCR:$pred))),
            (v4f32 (MVE_VPSEL (v4f32 (MVE_VMOVimmf32 240)), (v4f32 (MVE_VMOVimmi32 0)), 0, VCCR:$pred))>;
  // 2748 is -1.0 in half
  def : Pat<(v8f16 (sint_to_fp (v8i1 VCCR:$pred))),
            (v8f16 (MVE_VPSEL (v8f16 (MVE_VMOVimmi16 2748)), (v8f16 (MVE_VMOVimmi16 0)), 0, VCCR:$pred))>;

  def : Pat<(v4i1 (fp_to_uint (v4f32 MQPR:$v1))),
            (v4i1 (MVE_VCMPf32r (v4f32 MQPR:$v1), ZR, 1))>;
  def : Pat<(v8i1 (fp_to_uint (v8f16 MQPR:$v1))),
            (v8i1 (MVE_VCMPf16r (v8f16 MQPR:$v1), ZR, 1))>;
  def : Pat<(v4i1 (fp_to_sint (v4f32 MQPR:$v1))),
            (v4i1 (MVE_VCMPf32r (v4f32 MQPR:$v1), ZR, 1))>;
  def : Pat<(v8i1 (fp_to_sint (v8f16 MQPR:$v1))),
            (v8i1 (MVE_VCMPf16r (v8f16 MQPR:$v1), ZR, 1))>;
}

def MVE_VPNOT : MVE_p<(outs VCCR:$P0), (ins VCCR:$P0_in), NoItinerary,
                      "vpnot", "", "", vpred_n, "", []> {
  let Inst{31-0} = 0b11111110001100010000111101001101;
  let Unpredictable{19-17} = 0b111;
  let Unpredictable{12} = 0b1;
  let Unpredictable{7} = 0b1;
  let Unpredictable{5} = 0b1;

  let Constraints = "";
  let DecoderMethod = "DecodeMVEVPNOT";
}

let Predicates = [HasMVEInt] in {
  def : Pat<(v4i1 (xor (v4i1 VCCR:$pred), (v4i1 (predicate_cast (i32 65535))))),
            (v4i1 (MVE_VPNOT (v4i1 VCCR:$pred)))>;
  def : Pat<(v8i1 (xor (v8i1 VCCR:$pred), (v8i1 (predicate_cast (i32 65535))))),
            (v8i1 (MVE_VPNOT (v8i1 VCCR:$pred)))>;
  def : Pat<(v16i1 (xor (v16i1 VCCR:$pred), (v16i1 (predicate_cast (i32 65535))))),
            (v16i1 (MVE_VPNOT (v16i1 VCCR:$pred)))>;
}


class MVE_loltp_start<dag iops, string asm, string ops, bits<2> size>
  : t2LOL<(outs GPRlr:$LR), iops, asm, ops> {
  bits<4> Rn;
  let Predicates = [HasMVEInt];
  let Inst{22} = 0b0;
  let Inst{21-20} = size;
  let Inst{19-16} = Rn{3-0};
  let Inst{12} = 0b0;
}

class MVE_DLSTP<string asm, bits<2> size>
  : MVE_loltp_start<(ins rGPR:$Rn), asm, "$LR, $Rn", size> {
  let Inst{13} = 0b1;
  let Inst{11-1} = 0b00000000000;
  let Unpredictable{10-1} = 0b1111111111;
}

class MVE_WLSTP<string asm, bits<2> size>
  : MVE_loltp_start<(ins rGPR:$Rn, wlslabel_u11:$label),
                    asm, "$LR, $Rn, $label", size> {
  bits<11> label;
  let Inst{13} = 0b0;
  let Inst{11} = label{0};
  let Inst{10-1} = label{10-1};
}

def MVE_DLSTP_8  : MVE_DLSTP<"dlstp.8",  0b00>;
def MVE_DLSTP_16 : MVE_DLSTP<"dlstp.16", 0b01>;
def MVE_DLSTP_32 : MVE_DLSTP<"dlstp.32", 0b10>;
def MVE_DLSTP_64 : MVE_DLSTP<"dlstp.64", 0b11>;

def MVE_WLSTP_8  : MVE_WLSTP<"wlstp.8",  0b00>;
def MVE_WLSTP_16 : MVE_WLSTP<"wlstp.16", 0b01>;
def MVE_WLSTP_32 : MVE_WLSTP<"wlstp.32", 0b10>;
def MVE_WLSTP_64 : MVE_WLSTP<"wlstp.64", 0b11>;

class MVE_loltp_end<dag oops, dag iops, string asm, string ops>
  : t2LOL<oops, iops, asm, ops> {
  let Predicates = [HasMVEInt];
  let Inst{22-21} = 0b00;
  let Inst{19-16} = 0b1111;
  let Inst{12} = 0b0;
}

def MVE_LETP : MVE_loltp_end<(outs GPRlr:$LRout),
                             (ins GPRlr:$LRin, lelabel_u11:$label),
                             "letp", "$LRin, $label"> {
  bits<11> label;
  let Inst{20} = 0b1;
  let Inst{13} = 0b0;
  let Inst{11} = label{0};
  let Inst{10-1} = label{10-1};
}

def MVE_LCTP : MVE_loltp_end<(outs), (ins pred:$p), "lctp${p}", ""> {
  let Inst{20} = 0b0;
  let Inst{13} = 0b1;
  let Inst{11-1} = 0b00000000000;
  let Unpredictable{21-20} = 0b11;
  let Unpredictable{11-1} = 0b11111111111;
}


//===----------------------------------------------------------------------===//
// Patterns
//===----------------------------------------------------------------------===//

// PatFrags for loads and stores. Often trying to keep semi-consistent names.

def aligned32_pre_store : PatFrag<(ops node:$val, node:$ptr, node:$offset),
                                  (pre_store node:$val, node:$ptr, node:$offset), [{
  return cast<StoreSDNode>(N)->getAlignment() >= 4;
}]>;
def aligned32_post_store : PatFrag<(ops node:$val, node:$ptr, node:$offset),
                                   (post_store node:$val, node:$ptr, node:$offset), [{
  return cast<StoreSDNode>(N)->getAlignment() >= 4;
}]>;
def aligned16_pre_store : PatFrag<(ops node:$val, node:$ptr, node:$offset),
                                  (pre_store node:$val, node:$ptr, node:$offset), [{
  return cast<StoreSDNode>(N)->getAlignment() >= 2;
}]>;
def aligned16_post_store : PatFrag<(ops node:$val, node:$ptr, node:$offset),
                                   (post_store node:$val, node:$ptr, node:$offset), [{
  return cast<StoreSDNode>(N)->getAlignment() >= 2;
}]>;


def aligned_maskedloadvi8 : PatFrag<(ops node:$ptr, node:$pred, node:$passthru),
                                    (masked_ld node:$ptr, undef, node:$pred, node:$passthru), [{
  auto *Ld = cast<MaskedLoadSDNode>(N);
  return Ld->getMemoryVT().getScalarType() == MVT::i8;
}]>;
def aligned_sextmaskedloadvi8 : PatFrag<(ops node:$ptr, node:$pred, node:$passthru),
                                        (aligned_maskedloadvi8 node:$ptr, node:$pred, node:$passthru), [{
  return cast<MaskedLoadSDNode>(N)->getExtensionType() == ISD::SEXTLOAD;
}]>;
def aligned_zextmaskedloadvi8 : PatFrag<(ops node:$ptr, node:$pred, node:$passthru),
                                        (aligned_maskedloadvi8 node:$ptr, node:$pred, node:$passthru), [{
  return cast<MaskedLoadSDNode>(N)->getExtensionType() == ISD::ZEXTLOAD;
}]>;
def aligned_extmaskedloadvi8 : PatFrag<(ops node:$ptr, node:$pred, node:$passthru),
                                       (aligned_maskedloadvi8 node:$ptr, node:$pred, node:$passthru), [{
  auto *Ld = cast<MaskedLoadSDNode>(N);
  EVT ScalarVT = Ld->getMemoryVT().getScalarType();
  return ScalarVT.isInteger() && Ld->getExtensionType() == ISD::EXTLOAD;
}]>;
def aligned_maskedloadvi16: PatFrag<(ops node:$ptr, node:$pred, node:$passthru),
                                    (masked_ld node:$ptr, undef, node:$pred, node:$passthru), [{
  auto *Ld = cast<MaskedLoadSDNode>(N);
  EVT ScalarVT = Ld->getMemoryVT().getScalarType();
  return (ScalarVT == MVT::i16 || ScalarVT == MVT::f16) && Ld->getAlignment() >= 2;
}]>;
def aligned_sextmaskedloadvi16 : PatFrag<(ops node:$ptr, node:$pred, node:$passthru),
                                         (aligned_maskedloadvi16 node:$ptr, node:$pred, node:$passthru), [{
  return cast<MaskedLoadSDNode>(N)->getExtensionType() == ISD::SEXTLOAD;
}]>;
def aligned_zextmaskedloadvi16 : PatFrag<(ops node:$ptr, node:$pred, node:$passthru),
                                         (aligned_maskedloadvi16 node:$ptr, node:$pred, node:$passthru), [{
  return cast<MaskedLoadSDNode>(N)->getExtensionType() == ISD::ZEXTLOAD;
}]>;
def aligned_extmaskedloadvi16 : PatFrag<(ops node:$ptr, node:$pred, node:$passthru),
                                        (aligned_maskedloadvi16 node:$ptr, node:$pred, node:$passthru), [{
  auto *Ld = cast<MaskedLoadSDNode>(N);
  EVT ScalarVT = Ld->getMemoryVT().getScalarType();
  return ScalarVT.isInteger() && Ld->getExtensionType() == ISD::EXTLOAD;
}]>;
def aligned_maskedloadvi32: PatFrag<(ops node:$ptr, node:$pred, node:$passthru),
                                    (masked_ld node:$ptr, undef, node:$pred, node:$passthru), [{
  auto *Ld = cast<MaskedLoadSDNode>(N);
  EVT ScalarVT = Ld->getMemoryVT().getScalarType();
  return (ScalarVT == MVT::i32 || ScalarVT == MVT::f32) && Ld->getAlignment() >= 4;
}]>;

def aligned_maskedstvi8 : PatFrag<(ops node:$val, node:$ptr, node:$pred),
                                  (masked_st node:$val, node:$ptr, undef, node:$pred), [{
  return cast<MaskedStoreSDNode>(N)->getMemoryVT().getScalarType() == MVT::i8;
}]>;
def aligned_maskedstvi16 : PatFrag<(ops node:$val, node:$ptr, node:$pred),
                                   (masked_st node:$val, node:$ptr, undef, node:$pred), [{
  auto *St = cast<MaskedStoreSDNode>(N);
  EVT ScalarVT = St->getMemoryVT().getScalarType();
  return (ScalarVT == MVT::i16 || ScalarVT == MVT::f16) && St->getAlignment() >= 2;
}]>;
def aligned_maskedstvi32 : PatFrag<(ops node:$val, node:$ptr, node:$pred),
                                   (masked_st node:$val, node:$ptr, undef, node:$pred), [{
  auto *St = cast<MaskedStoreSDNode>(N);
  EVT ScalarVT = St->getMemoryVT().getScalarType();
  return (ScalarVT == MVT::i32 || ScalarVT == MVT::f32) && St->getAlignment() >= 4;
}]>;

def pre_maskedstore : PatFrag<(ops node:$val, node:$base, node:$offset, node:$mask),
                              (masked_st node:$val, node:$base, node:$offset, node:$mask), [{
  ISD::MemIndexedMode AM = cast<MaskedStoreSDNode>(N)->getAddressingMode();
  return AM == ISD::PRE_INC || AM == ISD::PRE_DEC;
}]>;
def post_maskedstore : PatFrag<(ops node:$val, node:$base, node:$offset, node:$mask),
                               (masked_st node:$val, node:$base, node:$offset, node:$mask), [{
  ISD::MemIndexedMode AM = cast<MaskedStoreSDNode>(N)->getAddressingMode();
  return AM == ISD::POST_INC || AM == ISD::POST_DEC;
}]>;
def aligned_pre_maskedstorevi8 : PatFrag<(ops node:$val, node:$ptr, node:$offset, node:$mask),
                                         (pre_maskedstore node:$val, node:$ptr, node:$offset, node:$mask), [{
  return cast<MaskedStoreSDNode>(N)->getMemoryVT().getScalarType() == MVT::i8;
}]>;
def aligned_post_maskedstorevi8 : PatFrag<(ops node:$val, node:$ptr, node:$offset, node:$mask),
                                          (post_maskedstore node:$val, node:$ptr, node:$offset, node:$mask), [{
  return cast<MaskedStoreSDNode>(N)->getMemoryVT().getScalarType() == MVT::i8;
}]>;
def aligned_pre_maskedstorevi16 : PatFrag<(ops node:$val, node:$ptr, node:$offset, node:$mask),
                                          (pre_maskedstore node:$val, node:$ptr, node:$offset, node:$mask), [{
  auto *St = cast<MaskedStoreSDNode>(N);
  EVT ScalarVT = St->getMemoryVT().getScalarType();
  return (ScalarVT == MVT::i16 || ScalarVT == MVT::f16) && St->getAlignment() >= 2;
}]>;
def aligned_post_maskedstorevi16 : PatFrag<(ops node:$val, node:$ptr, node:$offset, node:$mask),
                                           (post_maskedstore node:$val, node:$ptr, node:$offset, node:$mask), [{
  auto *St = cast<MaskedStoreSDNode>(N);
  EVT ScalarVT = St->getMemoryVT().getScalarType();
  return (ScalarVT == MVT::i16 || ScalarVT == MVT::f16) && St->getAlignment() >= 2;
}]>;
def aligned_pre_maskedstorevi32 : PatFrag<(ops node:$val, node:$ptr, node:$offset, node:$mask),
                                          (pre_maskedstore node:$val, node:$ptr, node:$offset, node:$mask), [{
  auto *St = cast<MaskedStoreSDNode>(N);
  EVT ScalarVT = St->getMemoryVT().getScalarType();
  return (ScalarVT == MVT::i32 || ScalarVT == MVT::f32) && St->getAlignment() >= 4;
}]>;
def aligned_post_maskedstorevi32 : PatFrag<(ops node:$val, node:$ptr, node:$offset, node:$mask),
                                           (post_maskedstore node:$val, node:$ptr, node:$offset, node:$mask), [{
  auto *St = cast<MaskedStoreSDNode>(N);
  EVT ScalarVT = St->getMemoryVT().getScalarType();
  return (ScalarVT == MVT::i32 || ScalarVT == MVT::f32) && St->getAlignment() >= 4;
}]>;


// PatFrags for "Aligned" extending / truncating

def aligned_extloadvi8  : PatFrag<(ops node:$ptr), (extloadvi8 node:$ptr)>;
def aligned_sextloadvi8 : PatFrag<(ops node:$ptr), (sextloadvi8 node:$ptr)>;
def aligned_zextloadvi8 : PatFrag<(ops node:$ptr), (zextloadvi8 node:$ptr)>;

def aligned_truncstvi8 : PatFrag<(ops node:$val, node:$ptr),
                                 (truncstorevi8 node:$val, node:$ptr)>;
def aligned_post_truncstvi8 : PatFrag<(ops node:$val, node:$base, node:$offset),
                                      (post_truncstvi8 node:$val, node:$base, node:$offset)>;
def aligned_pre_truncstvi8 : PatFrag<(ops node:$val, node:$base, node:$offset),
                                     (pre_truncstvi8 node:$val, node:$base, node:$offset)>;

let MinAlignment = 2 in {
  def aligned_extloadvi16  : PatFrag<(ops node:$ptr), (extloadvi16 node:$ptr)>;
  def aligned_sextloadvi16 : PatFrag<(ops node:$ptr), (sextloadvi16 node:$ptr)>;
  def aligned_zextloadvi16 : PatFrag<(ops node:$ptr), (zextloadvi16 node:$ptr)>;

  def aligned_truncstvi16 : PatFrag<(ops node:$val, node:$ptr),
                                    (truncstorevi16 node:$val, node:$ptr)>;
  def aligned_post_truncstvi16 : PatFrag<(ops node:$val, node:$base, node:$offset),
                                         (post_truncstvi16 node:$val, node:$base, node:$offset)>;
  def aligned_pre_truncstvi16 : PatFrag<(ops node:$val, node:$base, node:$offset),
                                        (pre_truncstvi16 node:$val, node:$base, node:$offset)>;
}

def truncmaskedst : PatFrag<(ops node:$val, node:$base, node:$pred),
                            (masked_st node:$val, node:$base, undef, node:$pred), [{
  return cast<MaskedStoreSDNode>(N)->isTruncatingStore();
}]>;
def aligned_truncmaskedstvi8 : PatFrag<(ops node:$val, node:$base, node:$pred),
                                       (truncmaskedst node:$val, node:$base, node:$pred), [{
  return cast<MaskedStoreSDNode>(N)->getMemoryVT().getScalarType() == MVT::i8;
}]>;
def aligned_truncmaskedstvi16 : PatFrag<(ops node:$val, node:$base, node:$pred),
                                        (truncmaskedst node:$val, node:$base, node:$pred), [{
  auto *St = cast<MaskedStoreSDNode>(N);
  EVT ScalarVT = St->getMemoryVT().getScalarType();
  return (ScalarVT == MVT::i16 || ScalarVT == MVT::f16) && St->getAlignment() >= 2;
}]>;
def pre_truncmaskedst : PatFrag<(ops node:$val, node:$base, node:$offset, node:$pred),
                                (masked_st node:$val, node:$base, node:$offset, node:$pred), [{
  ISD::MemIndexedMode AM = cast<MaskedStoreSDNode>(N)->getAddressingMode();
  return cast<MaskedStoreSDNode>(N)->isTruncatingStore() && (AM == ISD::PRE_INC || AM == ISD::PRE_DEC);
}]>;
def aligned_pre_truncmaskedstvi8 : PatFrag<(ops node:$val, node:$base, node:$offset, node:$pred),
                                           (pre_truncmaskedst node:$val, node:$base, node:$offset, node:$pred), [{
  return cast<MaskedStoreSDNode>(N)->getMemoryVT().getScalarType() == MVT::i8;
}]>;
def aligned_pre_truncmaskedstvi16 : PatFrag<(ops node:$val, node:$base, node:$offset, node:$pred),
                                            (pre_truncmaskedst node:$val, node:$base, node:$offset, node:$pred), [{
  auto *St = cast<MaskedStoreSDNode>(N);
  EVT ScalarVT = St->getMemoryVT().getScalarType();
  return (ScalarVT == MVT::i16 || ScalarVT == MVT::f16) && St->getAlignment() >= 2;
}]>;
def post_truncmaskedst : PatFrag<(ops node:$val, node:$base, node:$offset, node:$postd),
                                 (masked_st node:$val, node:$base, node:$offset, node:$postd), [{
  ISD::MemIndexedMode AM = cast<MaskedStoreSDNode>(N)->getAddressingMode();
  return cast<MaskedStoreSDNode>(N)->isTruncatingStore() && (AM == ISD::POST_INC || AM == ISD::POST_DEC);
}]>;
def aligned_post_truncmaskedstvi8 : PatFrag<(ops node:$val, node:$base, node:$offset, node:$postd),
                                            (post_truncmaskedst node:$val, node:$base, node:$offset, node:$postd), [{
  return cast<MaskedStoreSDNode>(N)->getMemoryVT().getScalarType() == MVT::i8;
}]>;
def aligned_post_truncmaskedstvi16 : PatFrag<(ops node:$val, node:$base, node:$offset, node:$postd),
                                             (post_truncmaskedst node:$val, node:$base, node:$offset, node:$postd), [{
  auto *St = cast<MaskedStoreSDNode>(N);
  EVT ScalarVT = St->getMemoryVT().getScalarType();
  return (ScalarVT == MVT::i16 || ScalarVT == MVT::f16) && St->getAlignment() >= 2;
}]>;

// Load/store patterns

class MVE_vector_store_typed<ValueType Ty, Instruction RegImmInst,
                             PatFrag StoreKind, int shift>
  : Pat<(StoreKind (Ty MQPR:$val), t2addrmode_imm7<shift>:$addr),
        (RegImmInst (Ty MQPR:$val), t2addrmode_imm7<shift>:$addr)>;

class MVE_vector_maskedstore_typed<ValueType Ty, Instruction RegImmInst,
                                   PatFrag StoreKind, int shift>
  : Pat<(StoreKind (Ty MQPR:$val), t2addrmode_imm7<shift>:$addr, VCCR:$pred),
        (RegImmInst (Ty MQPR:$val), t2addrmode_imm7<shift>:$addr, (i32 1), VCCR:$pred)>;

multiclass MVE_vector_store<Instruction RegImmInst, PatFrag StoreKind,
                            int shift> {
  def : MVE_vector_store_typed<v16i8, RegImmInst, StoreKind, shift>;
  def : MVE_vector_store_typed<v8i16, RegImmInst, StoreKind, shift>;
  def : MVE_vector_store_typed<v8f16, RegImmInst, StoreKind, shift>;
  def : MVE_vector_store_typed<v4i32, RegImmInst, StoreKind, shift>;
  def : MVE_vector_store_typed<v4f32, RegImmInst, StoreKind, shift>;
  def : MVE_vector_store_typed<v2i64, RegImmInst, StoreKind, shift>;
  def : MVE_vector_store_typed<v2f64, RegImmInst, StoreKind, shift>;
}

class MVE_vector_load_typed<ValueType Ty, Instruction RegImmInst,
                            PatFrag LoadKind, int shift>
  : Pat<(Ty (LoadKind t2addrmode_imm7<shift>:$addr)),
        (Ty (RegImmInst t2addrmode_imm7<shift>:$addr))>;

class MVE_vector_maskedload_typed<ValueType Ty, Instruction RegImmInst,
                                  PatFrag LoadKind, int shift>
  : Pat<(Ty (LoadKind t2addrmode_imm7<shift>:$addr, VCCR:$pred, (Ty NEONimmAllZerosV))),
        (Ty (RegImmInst t2addrmode_imm7<shift>:$addr, (i32 1), VCCR:$pred))>;

multiclass MVE_vector_load<Instruction RegImmInst, PatFrag LoadKind,
                           int shift> {
  def : MVE_vector_load_typed<v16i8, RegImmInst, LoadKind, shift>;
  def : MVE_vector_load_typed<v8i16, RegImmInst, LoadKind, shift>;
  def : MVE_vector_load_typed<v8f16, RegImmInst, LoadKind, shift>;
  def : MVE_vector_load_typed<v4i32, RegImmInst, LoadKind, shift>;
  def : MVE_vector_load_typed<v4f32, RegImmInst, LoadKind, shift>;
  def : MVE_vector_load_typed<v2i64, RegImmInst, LoadKind, shift>;
  def : MVE_vector_load_typed<v2f64, RegImmInst, LoadKind, shift>;
}

class MVE_vector_offset_store_typed<ValueType Ty, Instruction Opcode,
                                    PatFrag StoreKind, int shift>
  : Pat<(StoreKind (Ty MQPR:$Rt), tGPR:$Rn, t2am_imm7_offset<shift>:$addr),
        (Opcode MQPR:$Rt, tGPR:$Rn, t2am_imm7_offset<shift>:$addr)>;

class MVE_vector_offset_maskedstore_typed<ValueType Ty, Instruction Opcode,
                                          PatFrag StoreKind, int shift>
  : Pat<(StoreKind (Ty MQPR:$Rt), tGPR:$Rn, t2am_imm7_offset<shift>:$addr, VCCR:$pred),
        (Opcode MQPR:$Rt, tGPR:$Rn, t2am_imm7_offset<shift>:$addr, (i32 1), VCCR:$pred)>;

multiclass MVE_vector_offset_store<Instruction RegImmInst, PatFrag StoreKind,
                                   int shift> {
  def : MVE_vector_offset_store_typed<v16i8, RegImmInst, StoreKind, shift>;
  def : MVE_vector_offset_store_typed<v8i16, RegImmInst, StoreKind, shift>;
  def : MVE_vector_offset_store_typed<v8f16, RegImmInst, StoreKind, shift>;
  def : MVE_vector_offset_store_typed<v4i32, RegImmInst, StoreKind, shift>;
  def : MVE_vector_offset_store_typed<v4f32, RegImmInst, StoreKind, shift>;
  def : MVE_vector_offset_store_typed<v2i64, RegImmInst, StoreKind, shift>;
  def : MVE_vector_offset_store_typed<v2f64, RegImmInst, StoreKind, shift>;
}


let Predicates = [HasMVEInt, IsLE] in {
  // Stores
  defm : MVE_vector_store<MVE_VSTRBU8, byte_alignedstore, 0>;
  defm : MVE_vector_store<MVE_VSTRHU16, hword_alignedstore, 1>;
  defm : MVE_vector_store<MVE_VSTRWU32, alignedstore32, 2>;

  // Loads
  defm : MVE_vector_load<MVE_VLDRBU8, byte_alignedload, 0>;
  defm : MVE_vector_load<MVE_VLDRHU16, hword_alignedload, 1>;
  defm : MVE_vector_load<MVE_VLDRWU32, alignedload32, 2>;

  // Pre/post inc stores
  defm : MVE_vector_offset_store<MVE_VSTRBU8_pre, pre_store, 0>;
  defm : MVE_vector_offset_store<MVE_VSTRBU8_post, post_store, 0>;
  defm : MVE_vector_offset_store<MVE_VSTRHU16_pre, aligned16_pre_store, 1>;
  defm : MVE_vector_offset_store<MVE_VSTRHU16_post, aligned16_post_store, 1>;
  defm : MVE_vector_offset_store<MVE_VSTRWU32_pre, aligned32_pre_store, 2>;
  defm : MVE_vector_offset_store<MVE_VSTRWU32_post, aligned32_post_store, 2>;
}

let Predicates = [HasMVEInt, IsBE] in {
  // Aligned Stores
  def : MVE_vector_store_typed<v16i8, MVE_VSTRBU8, store, 0>;
  def : MVE_vector_store_typed<v8i16, MVE_VSTRHU16, alignedstore16, 1>;
  def : MVE_vector_store_typed<v8f16, MVE_VSTRHU16, alignedstore16, 1>;
  def : MVE_vector_store_typed<v4i32, MVE_VSTRWU32, alignedstore32, 2>;
  def : MVE_vector_store_typed<v4f32, MVE_VSTRWU32, alignedstore32, 2>;

  // Aligned Loads
  def : MVE_vector_load_typed<v16i8, MVE_VLDRBU8, load, 0>;
  def : MVE_vector_load_typed<v8i16, MVE_VLDRHU16, alignedload16, 1>;
  def : MVE_vector_load_typed<v8f16, MVE_VLDRHU16, alignedload16, 1>;
  def : MVE_vector_load_typed<v4i32, MVE_VLDRWU32, alignedload32, 2>;
  def : MVE_vector_load_typed<v4f32, MVE_VLDRWU32, alignedload32, 2>;

  // Other unaligned loads/stores need to go though a VREV
  def : Pat<(v2f64 (load t2addrmode_imm7<0>:$addr)),
            (v2f64 (MVE_VREV64_8 (MVE_VLDRBU8 t2addrmode_imm7<0>:$addr)))>;
  def : Pat<(v2i64 (load t2addrmode_imm7<0>:$addr)),
            (v2i64 (MVE_VREV64_8 (MVE_VLDRBU8 t2addrmode_imm7<0>:$addr)))>;
  def : Pat<(v4i32 (load t2addrmode_imm7<0>:$addr)),
            (v4i32 (MVE_VREV32_8 (MVE_VLDRBU8 t2addrmode_imm7<0>:$addr)))>;
  def : Pat<(v4f32 (load t2addrmode_imm7<0>:$addr)),
            (v4f32 (MVE_VREV32_8 (MVE_VLDRBU8 t2addrmode_imm7<0>:$addr)))>;
  def : Pat<(v8i16 (load t2addrmode_imm7<0>:$addr)),
            (v8i16 (MVE_VREV16_8 (MVE_VLDRBU8 t2addrmode_imm7<0>:$addr)))>;
  def : Pat<(v8f16 (load t2addrmode_imm7<0>:$addr)),
            (v8f16 (MVE_VREV16_8 (MVE_VLDRBU8 t2addrmode_imm7<0>:$addr)))>;
  def : Pat<(store (v2f64 MQPR:$val), t2addrmode_imm7<0>:$addr),
            (MVE_VSTRBU8 (MVE_VREV64_8 MQPR:$val), t2addrmode_imm7<0>:$addr)>;
  def : Pat<(store (v2i64 MQPR:$val), t2addrmode_imm7<0>:$addr),
            (MVE_VSTRBU8 (MVE_VREV64_8 MQPR:$val), t2addrmode_imm7<0>:$addr)>;
  def : Pat<(store (v4i32 MQPR:$val), t2addrmode_imm7<0>:$addr),
            (MVE_VSTRBU8 (MVE_VREV32_8 MQPR:$val), t2addrmode_imm7<0>:$addr)>;
  def : Pat<(store (v4f32 MQPR:$val), t2addrmode_imm7<0>:$addr),
            (MVE_VSTRBU8 (MVE_VREV32_8 MQPR:$val), t2addrmode_imm7<0>:$addr)>;
  def : Pat<(store (v8i16 MQPR:$val), t2addrmode_imm7<0>:$addr),
            (MVE_VSTRBU8 (MVE_VREV16_8 MQPR:$val), t2addrmode_imm7<0>:$addr)>;
  def : Pat<(store (v8f16 MQPR:$val), t2addrmode_imm7<0>:$addr),
            (MVE_VSTRBU8 (MVE_VREV16_8 MQPR:$val), t2addrmode_imm7<0>:$addr)>;

  // Pre/Post inc stores
  def : MVE_vector_offset_store_typed<v16i8, MVE_VSTRBU8_pre, pre_store, 0>;
  def : MVE_vector_offset_store_typed<v16i8, MVE_VSTRBU8_post, post_store, 0>;
  def : MVE_vector_offset_store_typed<v8i16, MVE_VSTRHU16_pre, aligned16_pre_store, 1>;
  def : MVE_vector_offset_store_typed<v8i16, MVE_VSTRHU16_post, aligned16_post_store, 1>;
  def : MVE_vector_offset_store_typed<v8f16, MVE_VSTRHU16_pre, aligned16_pre_store, 1>;
  def : MVE_vector_offset_store_typed<v8f16, MVE_VSTRHU16_post, aligned16_post_store, 1>;
  def : MVE_vector_offset_store_typed<v4i32, MVE_VSTRWU32_pre, aligned32_pre_store, 2>;
  def : MVE_vector_offset_store_typed<v4i32, MVE_VSTRWU32_post, aligned32_post_store, 2>;
  def : MVE_vector_offset_store_typed<v4f32, MVE_VSTRWU32_pre, aligned32_pre_store, 2>;
  def : MVE_vector_offset_store_typed<v4f32, MVE_VSTRWU32_post, aligned32_post_store, 2>;
}

let Predicates = [HasMVEInt] in {
  // Aligned masked store, shared between LE and BE
  def : MVE_vector_maskedstore_typed<v16i8, MVE_VSTRBU8, aligned_maskedstvi8, 0>;
  def : MVE_vector_maskedstore_typed<v8i16, MVE_VSTRHU16, aligned_maskedstvi16, 1>;
  def : MVE_vector_maskedstore_typed<v8f16, MVE_VSTRHU16, aligned_maskedstvi16, 1>;
  def : MVE_vector_maskedstore_typed<v4i32, MVE_VSTRWU32, aligned_maskedstvi32, 2>;
  def : MVE_vector_maskedstore_typed<v4f32, MVE_VSTRWU32, aligned_maskedstvi32, 2>;

  // Pre/Post inc masked stores
  def : MVE_vector_offset_maskedstore_typed<v16i8, MVE_VSTRBU8_pre, aligned_pre_maskedstorevi8, 0>;
  def : MVE_vector_offset_maskedstore_typed<v16i8, MVE_VSTRBU8_post, aligned_post_maskedstorevi8, 0>;
  def : MVE_vector_offset_maskedstore_typed<v8i16, MVE_VSTRHU16_pre, aligned_pre_maskedstorevi16, 1>;
  def : MVE_vector_offset_maskedstore_typed<v8i16, MVE_VSTRHU16_post, aligned_post_maskedstorevi16, 1>;
  def : MVE_vector_offset_maskedstore_typed<v8f16, MVE_VSTRHU16_pre, aligned_pre_maskedstorevi16, 1>;
  def : MVE_vector_offset_maskedstore_typed<v8f16, MVE_VSTRHU16_post, aligned_post_maskedstorevi16, 1>;
  def : MVE_vector_offset_maskedstore_typed<v4i32, MVE_VSTRWU32_pre, aligned_pre_maskedstorevi32, 2>;
  def : MVE_vector_offset_maskedstore_typed<v4i32, MVE_VSTRWU32_post, aligned_post_maskedstorevi32, 2>;
  def : MVE_vector_offset_maskedstore_typed<v4f32, MVE_VSTRWU32_pre, aligned_pre_maskedstorevi32, 2>;
  def : MVE_vector_offset_maskedstore_typed<v4f32, MVE_VSTRWU32_post, aligned_post_maskedstorevi32, 2>;

  // Aligned masked loads
  def : MVE_vector_maskedload_typed<v16i8, MVE_VLDRBU8, aligned_maskedloadvi8, 0>;
  def : MVE_vector_maskedload_typed<v8i16, MVE_VLDRHU16, aligned_maskedloadvi16, 1>;
  def : MVE_vector_maskedload_typed<v8f16, MVE_VLDRHU16, aligned_maskedloadvi16, 1>;
  def : MVE_vector_maskedload_typed<v4i32, MVE_VLDRWU32, aligned_maskedloadvi32, 2>;
  def : MVE_vector_maskedload_typed<v4f32, MVE_VLDRWU32, aligned_maskedloadvi32, 2>;
}

// Widening/Narrowing Loads/Stores

multiclass MVEExtLoadStore<Instruction LoadSInst, Instruction LoadUInst, string StoreInst,
                         string Amble, ValueType VT, int Shift> {
  // Trunc stores
  def : Pat<(!cast<PatFrag>("aligned_truncst"#Amble) (VT MQPR:$val), taddrmode_imm7<Shift>:$addr),
            (!cast<Instruction>(StoreInst) MQPR:$val, taddrmode_imm7<Shift>:$addr)>;
  def : Pat<(!cast<PatFrag>("aligned_post_truncst"#Amble) (VT MQPR:$Rt), tGPR:$Rn, t2am_imm7_offset<Shift>:$addr),
            (!cast<Instruction>(StoreInst#"_post") MQPR:$Rt, tGPR:$Rn, t2am_imm7_offset<Shift>:$addr)>;
  def : Pat<(!cast<PatFrag>("aligned_pre_truncst"#Amble) (VT MQPR:$Rt), tGPR:$Rn, t2am_imm7_offset<Shift>:$addr),
            (!cast<Instruction>(StoreInst#"_pre") MQPR:$Rt, tGPR:$Rn, t2am_imm7_offset<Shift>:$addr)>;

  // Masked trunc stores
  def : Pat<(!cast<PatFrag>("aligned_truncmaskedst"#Amble) (VT MQPR:$val), taddrmode_imm7<Shift>:$addr, VCCR:$pred),
            (!cast<Instruction>(StoreInst) MQPR:$val, taddrmode_imm7<Shift>:$addr, (i32 1), VCCR:$pred)>;
  def : Pat<(!cast<PatFrag>("aligned_post_truncmaskedst"#Amble) (VT MQPR:$Rt), tGPR:$Rn, t2am_imm7_offset<Shift>:$addr, VCCR:$pred),
            (!cast<Instruction>(StoreInst#"_post") MQPR:$Rt, tGPR:$Rn, t2am_imm7_offset<Shift>:$addr, (i32 1), VCCR:$pred)>;
  def : Pat<(!cast<PatFrag>("aligned_pre_truncmaskedst"#Amble) (VT MQPR:$Rt), tGPR:$Rn, t2am_imm7_offset<Shift>:$addr, VCCR:$pred),
            (!cast<Instruction>(StoreInst#"_pre") MQPR:$Rt, tGPR:$Rn, t2am_imm7_offset<Shift>:$addr, (i32 1), VCCR:$pred)>;

  // Ext loads
  def : Pat<(VT (!cast<PatFrag>("aligned_extload"#Amble) taddrmode_imm7<Shift>:$addr)),
            (VT (LoadUInst taddrmode_imm7<Shift>:$addr))>;
  def : Pat<(VT (!cast<PatFrag>("aligned_sextload"#Amble) taddrmode_imm7<Shift>:$addr)),
            (VT (LoadSInst taddrmode_imm7<Shift>:$addr))>;
  def : Pat<(VT (!cast<PatFrag>("aligned_zextload"#Amble) taddrmode_imm7<Shift>:$addr)),
            (VT (LoadUInst taddrmode_imm7<Shift>:$addr))>;

  // Masked ext loads
  def : Pat<(VT (!cast<PatFrag>("aligned_extmaskedload"#Amble) taddrmode_imm7<Shift>:$addr, VCCR:$pred, (VT NEONimmAllZerosV))),
            (VT (LoadUInst taddrmode_imm7<Shift>:$addr, (i32 1), VCCR:$pred))>;
  def : Pat<(VT (!cast<PatFrag>("aligned_sextmaskedload"#Amble) taddrmode_imm7<Shift>:$addr, VCCR:$pred, (VT NEONimmAllZerosV))),
            (VT (LoadSInst taddrmode_imm7<Shift>:$addr, (i32 1), VCCR:$pred))>;
  def : Pat<(VT (!cast<PatFrag>("aligned_zextmaskedload"#Amble) taddrmode_imm7<Shift>:$addr, VCCR:$pred, (VT NEONimmAllZerosV))),
            (VT (LoadUInst taddrmode_imm7<Shift>:$addr, (i32 1), VCCR:$pred))>;
}

let Predicates = [HasMVEInt] in {
  defm : MVEExtLoadStore<MVE_VLDRBS16, MVE_VLDRBU16, "MVE_VSTRB16", "vi8", v8i16, 0>;
  defm : MVEExtLoadStore<MVE_VLDRBS32, MVE_VLDRBU32, "MVE_VSTRB32", "vi8", v4i32, 0>;
  defm : MVEExtLoadStore<MVE_VLDRHS32, MVE_VLDRHU32, "MVE_VSTRH32", "vi16", v4i32, 1>;
}


// Bit convert patterns

let Predicates = [HasMVEInt] in {
  def : Pat<(v2f64 (bitconvert (v2i64 MQPR:$src))), (v2f64 MQPR:$src)>;
  def : Pat<(v2i64 (bitconvert (v2f64 MQPR:$src))), (v2i64 MQPR:$src)>;

  def : Pat<(v4i32 (bitconvert (v4f32 MQPR:$src))), (v4i32 MQPR:$src)>;
  def : Pat<(v4f32 (bitconvert (v4i32 MQPR:$src))), (v4f32 MQPR:$src)>;

  def : Pat<(v8i16 (bitconvert (v8f16 MQPR:$src))), (v8i16  MQPR:$src)>;
  def : Pat<(v8f16 (bitconvert (v8i16 MQPR:$src))), (v8f16  MQPR:$src)>;
}

let Predicates = [IsLE,HasMVEInt] in {
  def : Pat<(v2f64 (bitconvert (v4f32 MQPR:$src))), (v2f64 MQPR:$src)>;
  def : Pat<(v2f64 (bitconvert (v4i32 MQPR:$src))), (v2f64 MQPR:$src)>;
  def : Pat<(v2f64 (bitconvert (v8f16 MQPR:$src))), (v2f64 MQPR:$src)>;
  def : Pat<(v2f64 (bitconvert (v8i16 MQPR:$src))), (v2f64 MQPR:$src)>;
  def : Pat<(v2f64 (bitconvert (v16i8 MQPR:$src))), (v2f64 MQPR:$src)>;

  def : Pat<(v2i64 (bitconvert (v4f32 MQPR:$src))), (v2i64 MQPR:$src)>;
  def : Pat<(v2i64 (bitconvert (v4i32 MQPR:$src))), (v2i64 MQPR:$src)>;
  def : Pat<(v2i64 (bitconvert (v8f16 MQPR:$src))), (v2i64 MQPR:$src)>;
  def : Pat<(v2i64 (bitconvert (v8i16 MQPR:$src))), (v2i64 MQPR:$src)>;
  def : Pat<(v2i64 (bitconvert (v16i8 MQPR:$src))), (v2i64 MQPR:$src)>;

  def : Pat<(v4f32 (bitconvert (v2f64 MQPR:$src))), (v4f32 MQPR:$src)>;
  def : Pat<(v4f32 (bitconvert (v2i64 MQPR:$src))), (v4f32 MQPR:$src)>;
  def : Pat<(v4f32 (bitconvert (v8f16 MQPR:$src))), (v4f32 MQPR:$src)>;
  def : Pat<(v4f32 (bitconvert (v8i16 MQPR:$src))), (v4f32 MQPR:$src)>;
  def : Pat<(v4f32 (bitconvert (v16i8 MQPR:$src))), (v4f32 MQPR:$src)>;

  def : Pat<(v4i32 (bitconvert (v2f64 MQPR:$src))), (v4i32 MQPR:$src)>;
  def : Pat<(v4i32 (bitconvert (v2i64 MQPR:$src))), (v4i32 MQPR:$src)>;
  def : Pat<(v4i32 (bitconvert (v8f16 MQPR:$src))), (v4i32 MQPR:$src)>;
  def : Pat<(v4i32 (bitconvert (v8i16 MQPR:$src))), (v4i32 MQPR:$src)>;
  def : Pat<(v4i32 (bitconvert (v16i8 MQPR:$src))), (v4i32 MQPR:$src)>;

  def : Pat<(v8f16 (bitconvert (v2f64 MQPR:$src))), (v8f16 MQPR:$src)>;
  def : Pat<(v8f16 (bitconvert (v2i64 MQPR:$src))), (v8f16 MQPR:$src)>;
  def : Pat<(v8f16 (bitconvert (v4f32 MQPR:$src))), (v8f16 MQPR:$src)>;
  def : Pat<(v8f16 (bitconvert (v4i32 MQPR:$src))), (v8f16 MQPR:$src)>;
  def : Pat<(v8f16 (bitconvert (v16i8 MQPR:$src))), (v8f16 MQPR:$src)>;

  def : Pat<(v8i16 (bitconvert (v2f64 MQPR:$src))), (v8i16 MQPR:$src)>;
  def : Pat<(v8i16 (bitconvert (v2i64 MQPR:$src))), (v8i16 MQPR:$src)>;
  def : Pat<(v8i16 (bitconvert (v4f32 MQPR:$src))), (v8i16 MQPR:$src)>;
  def : Pat<(v8i16 (bitconvert (v4i32 MQPR:$src))), (v8i16 MQPR:$src)>;
  def : Pat<(v8i16 (bitconvert (v16i8 MQPR:$src))), (v8i16 MQPR:$src)>;

  def : Pat<(v16i8 (bitconvert (v2f64 MQPR:$src))), (v16i8 MQPR:$src)>;
  def : Pat<(v16i8 (bitconvert (v2i64 MQPR:$src))), (v16i8 MQPR:$src)>;
  def : Pat<(v16i8 (bitconvert (v4f32 MQPR:$src))), (v16i8 MQPR:$src)>;
  def : Pat<(v16i8 (bitconvert (v4i32 MQPR:$src))), (v16i8 MQPR:$src)>;
  def : Pat<(v16i8 (bitconvert (v8f16 MQPR:$src))), (v16i8 MQPR:$src)>;
  def : Pat<(v16i8 (bitconvert (v8i16 MQPR:$src))), (v16i8 MQPR:$src)>;
}

let Predicates = [IsBE,HasMVEInt] in {
  def : Pat<(v2f64 (bitconvert (v4f32 MQPR:$src))), (v2f64 (MVE_VREV64_32 MQPR:$src))>;
  def : Pat<(v2f64 (bitconvert (v4i32 MQPR:$src))), (v2f64 (MVE_VREV64_32 MQPR:$src))>;
  def : Pat<(v2f64 (bitconvert (v8f16 MQPR:$src))), (v2f64 (MVE_VREV64_16 MQPR:$src))>;
  def : Pat<(v2f64 (bitconvert (v8i16 MQPR:$src))), (v2f64 (MVE_VREV64_16 MQPR:$src))>;
  def : Pat<(v2f64 (bitconvert (v16i8 MQPR:$src))), (v2f64 (MVE_VREV64_8 MQPR:$src))>;

  def : Pat<(v2i64 (bitconvert (v4f32 MQPR:$src))), (v2i64 (MVE_VREV64_32 MQPR:$src))>;
  def : Pat<(v2i64 (bitconvert (v4i32 MQPR:$src))), (v2i64 (MVE_VREV64_32 MQPR:$src))>;
  def : Pat<(v2i64 (bitconvert (v8f16 MQPR:$src))), (v2i64 (MVE_VREV64_16 MQPR:$src))>;
  def : Pat<(v2i64 (bitconvert (v8i16 MQPR:$src))), (v2i64 (MVE_VREV64_16 MQPR:$src))>;
  def : Pat<(v2i64 (bitconvert (v16i8 MQPR:$src))), (v2i64 (MVE_VREV64_8 MQPR:$src))>;

  def : Pat<(v4f32 (bitconvert (v2f64 MQPR:$src))), (v4f32 (MVE_VREV64_32 MQPR:$src))>;
  def : Pat<(v4f32 (bitconvert (v2i64 MQPR:$src))), (v4f32 (MVE_VREV64_32 MQPR:$src))>;
  def : Pat<(v4f32 (bitconvert (v8f16 MQPR:$src))), (v4f32 (MVE_VREV32_16 MQPR:$src))>;
  def : Pat<(v4f32 (bitconvert (v8i16 MQPR:$src))), (v4f32 (MVE_VREV32_16 MQPR:$src))>;
  def : Pat<(v4f32 (bitconvert (v16i8 MQPR:$src))), (v4f32 (MVE_VREV32_8 MQPR:$src))>;

  def : Pat<(v4i32 (bitconvert (v2f64 MQPR:$src))), (v4i32 (MVE_VREV64_32 MQPR:$src))>;
  def : Pat<(v4i32 (bitconvert (v2i64 MQPR:$src))), (v4i32 (MVE_VREV64_32 MQPR:$src))>;
  def : Pat<(v4i32 (bitconvert (v8f16 MQPR:$src))), (v4i32 (MVE_VREV32_16 MQPR:$src))>;
  def : Pat<(v4i32 (bitconvert (v8i16 MQPR:$src))), (v4i32 (MVE_VREV32_16 MQPR:$src))>;
  def : Pat<(v4i32 (bitconvert (v16i8 MQPR:$src))), (v4i32 (MVE_VREV32_8 MQPR:$src))>;

  def : Pat<(v8f16 (bitconvert (v2f64 MQPR:$src))), (v8f16 (MVE_VREV64_16 MQPR:$src))>;
  def : Pat<(v8f16 (bitconvert (v2i64 MQPR:$src))), (v8f16 (MVE_VREV64_16 MQPR:$src))>;
  def : Pat<(v8f16 (bitconvert (v4f32 MQPR:$src))), (v8f16 (MVE_VREV32_16 MQPR:$src))>;
  def : Pat<(v8f16 (bitconvert (v4i32 MQPR:$src))), (v8f16 (MVE_VREV32_16 MQPR:$src))>;
  def : Pat<(v8f16 (bitconvert (v16i8 MQPR:$src))), (v8f16 (MVE_VREV16_8 MQPR:$src))>;

  def : Pat<(v8i16 (bitconvert (v2f64 MQPR:$src))), (v8i16 (MVE_VREV64_16 MQPR:$src))>;
  def : Pat<(v8i16 (bitconvert (v2i64 MQPR:$src))), (v8i16 (MVE_VREV64_16 MQPR:$src))>;
  def : Pat<(v8i16 (bitconvert (v4f32 MQPR:$src))), (v8i16 (MVE_VREV32_16 MQPR:$src))>;
  def : Pat<(v8i16 (bitconvert (v4i32 MQPR:$src))), (v8i16 (MVE_VREV32_16 MQPR:$src))>;
  def : Pat<(v8i16 (bitconvert (v16i8 MQPR:$src))), (v8i16 (MVE_VREV16_8 MQPR:$src))>;

  def : Pat<(v16i8 (bitconvert (v2f64 MQPR:$src))), (v16i8 (MVE_VREV64_8 MQPR:$src))>;
  def : Pat<(v16i8 (bitconvert (v2i64 MQPR:$src))), (v16i8 (MVE_VREV64_8 MQPR:$src))>;
  def : Pat<(v16i8 (bitconvert (v4f32 MQPR:$src))), (v16i8 (MVE_VREV32_8 MQPR:$src))>;
  def : Pat<(v16i8 (bitconvert (v4i32 MQPR:$src))), (v16i8 (MVE_VREV32_8 MQPR:$src))>;
  def : Pat<(v16i8 (bitconvert (v8f16 MQPR:$src))), (v16i8 (MVE_VREV16_8 MQPR:$src))>;
  def : Pat<(v16i8 (bitconvert (v8i16 MQPR:$src))), (v16i8 (MVE_VREV16_8 MQPR:$src))>;
}