llvm-project/llvm/lib/Target/X86/X86ScheduleBtVer2.td

//=- X86ScheduleBtVer2.td - X86 BtVer2 (Jaguar) Scheduling ---*- tablegen -*-=//
//
//                     The LLVM Compiler Infrastructure
//
// This file is distributed under the University of Illinois Open Source
// License. See LICENSE.TXT for details.
//
//===----------------------------------------------------------------------===//
//
// This file defines the machine model for AMD btver2 (Jaguar) to support
// instruction scheduling and other instruction cost heuristics. Based off AMD Software
// Optimization Guide for AMD Family 16h Processors & Instruction Latency appendix.
//
//===----------------------------------------------------------------------===//

def BtVer2Model : SchedMachineModel {
  // All x86 instructions are modeled as a single micro-op, and btver2 can
  // decode 2 instructions per cycle.
  let IssueWidth = 2;
  let MicroOpBufferSize = 64; // Retire Control Unit
  let LoadLatency = 5; // FPU latency (worse case cf Integer 3 cycle latency)
  let HighLatency = 25;
  let MispredictPenalty = 14; // Minimum branch misdirection penalty
  let PostRAScheduler = 1;

  // FIXME: SSE4/AVX is unimplemented. This flag is set to allow
  // the scheduler to assign a default model to unrecognized opcodes.
  let CompleteModel = 0;
}

let SchedModel = BtVer2Model in {

// Jaguar can issue up to 6 micro-ops in one cycle
def JALU0 : ProcResource<1>; // Integer Pipe0: integer ALU0 (also handle FP->INT jam)
def JALU1 : ProcResource<1>; // Integer Pipe1: integer ALU1/MUL/DIV
def JLAGU : ProcResource<1>; // Integer Pipe2: LAGU
def JSAGU : ProcResource<1>; // Integer Pipe3: SAGU (also handles 3-operand LEA)
def JFPU0 : ProcResource<1>; // Vector/FPU Pipe0: VALU0/VIMUL/FPA
def JFPU1 : ProcResource<1>; // Vector/FPU Pipe1: VALU1/STC/FPM

// The Integer PRF for Jaguar is 64 entries, and it holds the architectural and
// speculative version of the 64-bit integer registers.
// Reference: www.realworldtech.com/jaguar/4/
def IntegerPRF : RegisterFile<64, [GR8, GR16, GR32, GR64, CCR]>;

// The Jaguar FP Retire Queue renames SIMD and FP uOps onto a pool of 72 SSE
// registers. Operations on 256-bit data types are cracked into two COPs.
// Reference: www.realworldtech.com/jaguar/4/
def FpuPRF: RegisterFile<72, [VR64, VR128, VR256], [1, 1, 2]>;

// The retire control unit (RCU) can track up to 64 macro-ops in-flight. It can
// retire up to two macro-ops per cycle.
// Reference: "Software Optimization Guide for AMD Family 16h Processors"
def RCU : RetireControlUnit<64, 2>;

// Integer Pipe Scheduler
def JALU01 : ProcResGroup<[JALU0, JALU1]> {
  let BufferSize=20;
}

// AGU Pipe Scheduler
def JLSAGU : ProcResGroup<[JLAGU, JSAGU]> {
  let BufferSize=12;
}

// Fpu Pipe Scheduler
def JFPU01 : ProcResGroup<[JFPU0, JFPU1]> {
  let BufferSize=18;
}

// Functional units
def JDiv    : ProcResource<1>; // integer division
def JMul    : ProcResource<1>; // integer multiplication
def JVALU0  : ProcResource<1>; // vector integer
def JVALU1  : ProcResource<1>; // vector integer
def JVIMUL  : ProcResource<1>; // vector integer multiplication
def JSTC    : ProcResource<1>; // vector store/convert
def JFPM    : ProcResource<1>; // FP multiplication
def JFPA    : ProcResource<1>; // FP addition

// Functional unit groups
def JFPX  : ProcResGroup<[JFPA, JFPM]>;
def JVALU : ProcResGroup<[JVALU0, JVALU1]>;

// Integer loads are 3 cycles, so ReadAfterLd registers needn't be available until 3
// cycles after the memory operand.
def : ReadAdvance<ReadAfterLd, 3>;

// Many SchedWrites are defined in pairs with and without a folded load.
// Instructions with folded loads are usually micro-fused, so they only appear
// as two micro-ops when dispatched by the schedulers.
// This multiclass defines the resource usage for variants with and without
// folded loads.
multiclass JWriteResIntPair<X86FoldableSchedWrite SchedRW,
                            list<ProcResourceKind> ExePorts,
                            int Lat, list<int> Res = [], int UOps = 1> {
  // Register variant is using a single cycle on ExePort.
  def : WriteRes<SchedRW, ExePorts> {
    let Latency = Lat;
    let ResourceCycles = Res;
    let NumMicroOps = UOps;
  }

  // Memory variant also uses a cycle on JLAGU and adds 3 cycles to the
  // latency.
  def : WriteRes<SchedRW.Folded, !listconcat([JLAGU], ExePorts)> {
    let Latency = !add(Lat, 3);
    let ResourceCycles = !if(!empty(Res), [], !listconcat([1], Res));
    let NumMicroOps = UOps;
  }
}

multiclass JWriteResFpuPair<X86FoldableSchedWrite SchedRW,
                            list<ProcResourceKind> ExePorts,
                            int Lat, list<int> Res = [], int UOps = 1> {
  // Register variant is using a single cycle on ExePort.
  def : WriteRes<SchedRW, ExePorts> {
    let Latency = Lat;
    let ResourceCycles = Res;
    let NumMicroOps = UOps;
  }

  // Memory variant also uses a cycle on JLAGU and adds 5 cycles to the
  // latency.
  def : WriteRes<SchedRW.Folded, !listconcat([JLAGU], ExePorts)> {
    let Latency = !add(Lat, 5);
    let ResourceCycles = !if(!empty(Res), [], !listconcat([1], Res));
    let NumMicroOps = UOps;
  }
}

multiclass JWriteResYMMPair<X86FoldableSchedWrite SchedRW,
                            list<ProcResourceKind> ExePorts,
                            int Lat, list<int> Res = [2], int UOps = 2> {
  // Register variant is using a single cycle on ExePort.
  def : WriteRes<SchedRW, ExePorts> {
    let Latency = Lat;
    let ResourceCycles = Res;
    let NumMicroOps = UOps;
  }

  // Memory variant also uses 2 cycles on JLAGU and adds 5 cycles to the
  // latency.
  def : WriteRes<SchedRW.Folded, !listconcat([JLAGU], ExePorts)> {
    let Latency = !add(Lat, 5);
    let ResourceCycles = !listconcat([2], Res);
    let NumMicroOps = UOps;
  }
}

// A folded store needs a cycle on the SAGU for the store data.
def : WriteRes<WriteRMW, [JSAGU]>;

////////////////////////////////////////////////////////////////////////////////
// Arithmetic.
////////////////////////////////////////////////////////////////////////////////

defm : JWriteResIntPair<WriteALU,   [JALU01], 1>;
defm : JWriteResIntPair<WriteIMul,  [JALU1, JMul], 3, [1, 1], 2>; // i8/i16/i32 multiplication
defm : JWriteResIntPair<WriteIDiv,  [JALU1, JDiv], 41, [1, 41], 2>; // Worst case (i64 division)
defm : JWriteResIntPair<WriteCRC32, [JALU01], 3, [4], 3>;

defm : JWriteResIntPair<WriteCMOV,  [JALU01], 1>; // Conditional move.
def  : WriteRes<WriteSETCC, [JALU01]>; // Setcc.
def  : WriteRes<WriteSETCCStore, [JALU01,JSAGU]>;

def  : WriteRes<WriteIMulH, [JALU1]> {
  let Latency = 6;
  let ResourceCycles = [4];
}

// This is for simple LEAs with one or two input operands.
// FIXME: SAGU 3-operand LEA
def : WriteRes<WriteLEA, [JALU01]>;

// Bit counts.
defm : JWriteResIntPair<WriteBitScan, [JALU01], 5, [4], 8>;
defm : JWriteResIntPair<WritePOPCNT,  [JALU01], 1>;
defm : JWriteResIntPair<WriteLZCNT,   [JALU01], 1>;
defm : JWriteResIntPair<WriteTZCNT,   [JALU01], 2, [2]>;

// BMI1 BEXTR, BMI2 BZHI
defm : JWriteResIntPair<WriteBEXTR, [JALU01], 1>;
defm : JWriteResIntPair<WriteBZHI, [JALU01], 1>; // NOTE: Doesn't exist on Jaguar.

def JWriteIMul64 : SchedWriteRes<[JALU1, JMul]> {
  let Latency = 6;
  let ResourceCycles = [1, 4];
  let NumMicroOps = 2;
}
def JWriteIMul64Ld : SchedWriteRes<[JLAGU, JALU1, JMul]> {
  let Latency = 9;
  let ResourceCycles = [1, 1, 4];
  let NumMicroOps = 2;
}
def : InstRW<[JWriteIMul64], (instrs MUL64r, IMUL64r)>;
def : InstRW<[JWriteIMul64Ld], (instrs MUL64m, IMUL64m)>;

def JWriteIDiv8 : SchedWriteRes<[JALU1, JDiv]> {
  let Latency = 12;
  let ResourceCycles = [1, 12];
}
def JWriteIDiv8Ld : SchedWriteRes<[JLAGU, JALU1, JDiv]> {
  let Latency = 15;
  let ResourceCycles = [1, 1, 12];
}
def : InstRW<[JWriteIDiv8], (instrs DIV8r, IDIV8r)>;
def : InstRW<[JWriteIDiv8Ld], (instrs DIV8m, IDIV8m)>;

def JWriteIDiv16 : SchedWriteRes<[JALU1, JDiv]> {
  let Latency = 17;
  let ResourceCycles = [1, 17];
  let NumMicroOps = 2;
}
def JWriteIDiv16Ld : SchedWriteRes<[JLAGU, JALU1, JDiv]> {
  let Latency = 20;
  let ResourceCycles = [1, 1, 17];
  let NumMicroOps = 2;
}
def : InstRW<[JWriteIDiv16], (instrs DIV16r, IDIV16r)>;
def : InstRW<[JWriteIDiv16Ld], (instrs DIV16m, IDIV16m)>;

def JWriteIDiv32 : SchedWriteRes<[JALU1, JDiv]> {
  let Latency = 25;
  let ResourceCycles = [1, 25];
  let NumMicroOps = 2;
}
def JWriteIDiv32Ld : SchedWriteRes<[JLAGU, JALU1, JDiv]> {
  let Latency = 28;
  let ResourceCycles = [1, 1, 25];
  let NumMicroOps = 2;
}
def : InstRW<[JWriteIDiv32], (instrs DIV32r, IDIV32r)>;
def : InstRW<[JWriteIDiv32Ld], (instrs DIV32m, IDIV32m)>;

////////////////////////////////////////////////////////////////////////////////
// Integer shifts and rotates.
////////////////////////////////////////////////////////////////////////////////

defm : JWriteResIntPair<WriteShift, [JALU01], 1>;

def JWriteSHLDrri : SchedWriteRes<[JALU01]> {
  let Latency = 3;
  let ResourceCycles = [6];
  let NumMicroOps = 6;
}
def: InstRW<[JWriteSHLDrri], (instrs SHLD16rri8, SHLD32rri8, SHLD64rri8,
                                     SHRD16rri8, SHRD32rri8, SHRD64rri8)>;

def JWriteSHLDrrCL : SchedWriteRes<[JALU01]> {
  let Latency = 4;
  let ResourceCycles = [8];
  let NumMicroOps = 7;
}
def: InstRW<[JWriteSHLDrrCL], (instrs SHLD16rrCL, SHLD32rrCL, SHLD64rrCL,
                                      SHRD16rrCL, SHRD32rrCL, SHRD64rrCL)>;

def JWriteSHLDm : SchedWriteRes<[JLAGU, JALU01]> {
  let Latency = 9;
  let ResourceCycles = [1, 22];
  let NumMicroOps = 8;
}
def: InstRW<[JWriteSHLDm],(instrs SHLD16mri8, SHLD32mri8, SHLD64mri8,
                                  SHLD16mrCL, SHLD32mrCL, SHLD64mrCL,
                                  SHRD16mri8, SHRD32mri8, SHRD64mri8,
                                  SHRD16mrCL, SHRD32mrCL, SHRD64mrCL)>;

////////////////////////////////////////////////////////////////////////////////
// Loads, stores, and moves, not folded with other operations.
////////////////////////////////////////////////////////////////////////////////

def : WriteRes<WriteLoad,  [JLAGU]> { let Latency = 5; }
def : WriteRes<WriteStore, [JSAGU]>;
def : WriteRes<WriteMove,  [JALU01]>;

// Load/store MXCSR.
// FIXME: These are copy and pasted from WriteLoad/Store.
def : WriteRes<WriteLDMXCSR, [JLAGU]> { let Latency = 5; }
def : WriteRes<WriteSTMXCSR, [JSAGU]>;

// Treat misc copies as a move.
def : InstRW<[WriteMove], (instrs COPY)>;

////////////////////////////////////////////////////////////////////////////////
// Idioms that clear a register, like xorps %xmm0, %xmm0.
// These can often bypass execution ports completely.
////////////////////////////////////////////////////////////////////////////////

def : WriteRes<WriteZero,  []>;

////////////////////////////////////////////////////////////////////////////////
// Branches don't produce values, so they have no latency, but they still
// consume resources. Indirect branches can fold loads.
////////////////////////////////////////////////////////////////////////////////

defm : JWriteResIntPair<WriteJump,  [JALU01], 1>;

////////////////////////////////////////////////////////////////////////////////
// Special case scheduling classes.
////////////////////////////////////////////////////////////////////////////////

def : WriteRes<WriteSystem,     [JALU01]> { let Latency = 100; }
def : WriteRes<WriteMicrocoded, [JALU01]> { let Latency = 100; }
def : WriteRes<WriteFence,  [JSAGU]>;

// Nops don't have dependencies, so there's no actual latency, but we set this
// to '1' to tell the scheduler that the nop uses an ALU slot for a cycle.
def : WriteRes<WriteNop, [JALU01]> { let Latency = 1; }

////////////////////////////////////////////////////////////////////////////////
// Floating point. This covers both scalar and vector operations.
////////////////////////////////////////////////////////////////////////////////

def  : WriteRes<WriteFLoad,        [JLAGU, JFPU01, JFPX]> { let Latency = 5; }
def  : WriteRes<WriteFStore,        [JSAGU, JFPU1, JSTC]>;
def  : WriteRes<WriteFMove,               [JFPU01, JFPX]>;

defm : JWriteResFpuPair<WriteFAdd,         [JFPU0, JFPA],  3>;
defm : JWriteResYMMPair<WriteFAddY,        [JFPU0, JFPA],  3, [2,2], 2>;
defm : JWriteResFpuPair<WriteFCmp,         [JFPU0, JFPA],  2>;
defm : JWriteResYMMPair<WriteFCmpY,        [JFPU0, JFPA],  2, [2,2], 2>;
defm : JWriteResFpuPair<WriteFCom,  [JFPU0, JFPA, JALU0],  3>;
defm : JWriteResFpuPair<WriteFMul,         [JFPU1, JFPM],  2>;
defm : JWriteResYMMPair<WriteFMulY,        [JFPU1, JFPM],  2, [2,2], 2>;
defm : JWriteResFpuPair<WriteFMA,          [JFPU1, JFPM],  2>; // NOTE: Doesn't exist on Jaguar.
defm : JWriteResFpuPair<WriteFMAS,         [JFPU1, JFPM],  2>; // NOTE: Doesn't exist on Jaguar.
defm : JWriteResFpuPair<WriteFMAY,         [JFPU1, JFPM],  2>; // NOTE: Doesn't exist on Jaguar.
defm : JWriteResFpuPair<WriteDPPD,   [JFPU1, JFPM, JFPA],  9, [1, 3, 3],  3>;
defm : JWriteResFpuPair<WriteDPPS,   [JFPU1, JFPM, JFPA], 11, [1, 3, 3],  5>;
defm : JWriteResYMMPair<WriteDPPSY,  [JFPU1, JFPM, JFPA], 12, [2, 6, 6], 10>;
defm : JWriteResFpuPair<WriteFRcp,         [JFPU1, JFPM],  2>;
defm : JWriteResYMMPair<WriteFRcpY,        [JFPU1, JFPM],  2, [2,2], 2>;
defm : JWriteResFpuPair<WriteFRsqrt,       [JFPU1, JFPM],  2>;
defm : JWriteResYMMPair<WriteFRsqrtY,      [JFPU1, JFPM],  2, [2,2], 2>;
defm : JWriteResFpuPair<WriteFDiv,         [JFPU1, JFPM], 19, [1, 19]>;
defm : JWriteResYMMPair<WriteFDivY,        [JFPU1, JFPM], 38, [2, 38], 2>;
defm : JWriteResFpuPair<WriteFSqrt,        [JFPU1, JFPM], 21, [1, 21]>;
defm : JWriteResYMMPair<WriteFSqrtY,       [JFPU1, JFPM], 42, [2, 42], 2>;
defm : JWriteResFpuPair<WriteFSign,        [JFPU1, JFPM],  2>;
defm : JWriteResFpuPair<WriteFRnd,         [JFPU1, JSTC],  3>;
defm : JWriteResYMMPair<WriteFRndY,        [JFPU1, JSTC],  3, [2,2], 2>;
defm : JWriteResFpuPair<WriteFLogic,      [JFPU01, JFPX],  1>;
defm : JWriteResYMMPair<WriteFLogicY,     [JFPU01, JFPX],  1, [2, 2], 2>;
defm : JWriteResFpuPair<WriteFShuffle,    [JFPU01, JFPX],  1>;
defm : JWriteResYMMPair<WriteFShuffleY,   [JFPU01, JFPX],  1, [2, 2], 2>;
defm : JWriteResFpuPair<WriteFVarShuffle, [JFPU01, JFPX],  2, [1, 4], 3>;
defm : JWriteResYMMPair<WriteFVarShuffleY,[JFPU01, JFPX],  3, [2, 6], 6>;
defm : JWriteResFpuPair<WriteFBlend,      [JFPU01, JFPX],  1>;
defm : JWriteResYMMPair<WriteFBlendY,     [JFPU01, JFPX],  1, [2, 2], 2>;
defm : JWriteResFpuPair<WriteFVarBlend,   [JFPU01, JFPX],  2, [1, 4], 3>;
defm : JWriteResYMMPair<WriteFVarBlendY,  [JFPU01, JFPX],  3, [2, 6], 6>;
defm : JWriteResFpuPair<WriteFShuffle256, [JFPU01, JFPX],  1>;
defm : JWriteResFpuPair<WriteFVarShuffle256, [JFPU01, JFPX],  1>; // NOTE: Doesn't exist on Jaguar.

////////////////////////////////////////////////////////////////////////////////
// Conversions.
////////////////////////////////////////////////////////////////////////////////

defm : JWriteResFpuPair<WriteCvtF2I,       [JFPU1, JSTC], 3>; // Float -> Integer.
defm : JWriteResFpuPair<WriteCvtI2F,       [JFPU1, JSTC], 3>; // Integer -> Float.
defm : JWriteResFpuPair<WriteCvtF2F,       [JFPU1, JSTC], 3>; // Float -> Float size conversion.
def  : WriteRes<WriteCvtF2FSt, [JFPU1, JSTC, JSAGU]> { let Latency = 4; }

def JWriteCVTF2F : SchedWriteRes<[JFPU1, JSTC]> {
  let Latency = 7;
  let ResourceCycles = [1, 2];
  let NumMicroOps = 2;
}
def : InstRW<[JWriteCVTF2F], (instregex "(V)?CVTS(D|S)2S(D|S)rr")>;

def JWriteCVTF2FLd : SchedWriteRes<[JLAGU, JFPU1, JSTC]> {
  let Latency = 12;
  let ResourceCycles = [1, 1, 2];
  let NumMicroOps = 2;
}
def : InstRW<[JWriteCVTF2FLd], (instregex "(V)?CVTS(D|S)2S(D|S)rm")>;

def JWriteCVTF2SI : SchedWriteRes<[JFPU1, JSTC, JFPA, JALU0]> {
  let Latency = 7;
  let NumMicroOps = 2;
}
def : InstRW<[JWriteCVTF2SI], (instregex "(V)?CVT(T?)S(D|S)2SI(64)?rr")>;

def JWriteCVTF2SILd : SchedWriteRes<[JLAGU, JFPU1, JSTC, JFPA, JALU0]> {
  let Latency = 12;
  let NumMicroOps = 2;
}
def : InstRW<[JWriteCVTF2SILd], (instregex "(V)?CVT(T?)S(D|S)2SI(64)?rm")>;

// FIXME: f+3 ST, LD+STC latency
def JWriteCVTSI2F : SchedWriteRes<[JFPU1, JSTC]> {
  let Latency = 9;
  let NumMicroOps = 2;
}
def : InstRW<[JWriteCVTSI2F], (instregex "(V)?CVTSI(64)?2S(D|S)rr")>;

def JWriteCVTSI2FLd : SchedWriteRes<[JLAGU, JFPU1, JSTC]> {
  let Latency = 14;
  let NumMicroOps = 2;
}
def : InstRW<[JWriteCVTSI2FLd], (instregex "(V)?CVTSI(64)?2S(D|S)rm")>;

////////////////////////////////////////////////////////////////////////////////
// Vector integer operations.
////////////////////////////////////////////////////////////////////////////////

def  : WriteRes<WriteVecLoad,      [JLAGU, JFPU01, JVALU]> { let Latency = 5; }
def  : WriteRes<WriteVecStore,       [JSAGU, JFPU1, JSTC]>;
def  : WriteRes<WriteVecMove,             [JFPU01, JVALU]>;

defm : JWriteResFpuPair<WriteVecALU,      [JFPU01, JVALU], 1>;
defm : JWriteResFpuPair<WriteVecALUY,     [JFPU01, JVALU], 1>;
defm : JWriteResFpuPair<WriteVecShift,    [JFPU01, JVALU], 1>;
defm : JWriteResFpuPair<WriteVecShiftX,   [JFPU01, JVALU], 1>;
defm : JWriteResFpuPair<WriteVecShiftY,   [JFPU01, JVALU], 1>;
defm : JWriteResFpuPair<WriteVecShiftImmX,[JFPU01, JVALU], 1>;
defm : JWriteResFpuPair<WriteVecShiftImmY,[JFPU01, JVALU], 1>;
defm : JWriteResFpuPair<WriteVecIMul,     [JFPU0, JVIMUL], 2>;
defm : JWriteResFpuPair<WriteVecIMulY,    [JFPU0, JVIMUL], 2>;
defm : JWriteResFpuPair<WritePMULLD,      [JFPU0, JFPU01, JVIMUL, JVALU], 4, [2, 1, 2, 1], 3>;
defm : JWriteResFpuPair<WritePMULLDY,     [JFPU0, JFPU01, JVIMUL, JVALU], 4, [2, 1, 2, 1], 3>;
defm : JWriteResFpuPair<WriteMPSAD,       [JFPU0, JVIMUL], 3, [1, 2]>;
defm : JWriteResFpuPair<WriteMPSADY,      [JFPU0, JVIMUL], 3, [1, 2]>;
defm : JWriteResFpuPair<WritePSADBW,      [JFPU01, JVALU], 2>;
defm : JWriteResFpuPair<WritePSADBWY,     [JFPU01, JVALU], 2>;
defm : JWriteResFpuPair<WritePHMINPOS,    [JFPU0,  JVALU], 2>;
defm : JWriteResFpuPair<WriteShuffle,     [JFPU01, JVALU], 1>;
defm : JWriteResFpuPair<WriteShuffleY,    [JFPU01, JVALU], 1>;
defm : JWriteResFpuPair<WriteVarShuffle,  [JFPU01, JVALU], 2, [1, 4], 3>;
defm : JWriteResFpuPair<WriteVarShuffleY, [JFPU01, JVALU], 2, [1, 4], 3>;
defm : JWriteResFpuPair<WriteBlend,       [JFPU01, JVALU], 1>;
defm : JWriteResFpuPair<WriteBlendY,      [JFPU01, JVALU], 1>;
defm : JWriteResFpuPair<WriteVarBlend,    [JFPU01, JVALU], 2, [1, 4], 3>;
defm : JWriteResFpuPair<WriteVarBlendY,   [JFPU01, JVALU], 2, [1, 4], 3>;
defm : JWriteResFpuPair<WriteVecLogic,    [JFPU01, JVALU], 1>;
defm : JWriteResFpuPair<WriteVecLogicY,   [JFPU01, JVALU], 1>; // NOTE: Doesn't exist on Jaguar.
defm : JWriteResFpuPair<WriteShuffle256,  [JFPU01, JVALU], 1>;
defm : JWriteResFpuPair<WriteVarShuffle256, [JFPU01, JVALU], 1>; // NOTE: Doesn't exist on Jaguar.
defm : JWriteResFpuPair<WriteVarVecShift, [JFPU01, JVALU], 1>; // NOTE: Doesn't exist on Jaguar.
defm : JWriteResFpuPair<WriteVarVecShiftY,[JFPU01, JVALU], 1>; // NOTE: Doesn't exist on Jaguar.

////////////////////////////////////////////////////////////////////////////////
// Vector insert/extract operations.
////////////////////////////////////////////////////////////////////////////////

defm : JWriteResFpuPair<WriteVecInsert,   [JFPU01, JVALU], 1>;
def  : WriteRes<WriteVecExtract,     [JFPU0, JFPA, JALU0]> { let Latency = 3; }
def  : WriteRes<WriteVecExtractSt,   [JFPU1, JSTC, JSAGU]> { let Latency = 3; }

////////////////////////////////////////////////////////////////////////////////
// SSE42 String instructions.
////////////////////////////////////////////////////////////////////////////////

defm : JWriteResFpuPair<WritePCmpIStrI, [JFPU1, JVALU1, JFPA, JALU0], 7, [1, 2, 1, 1], 3>;
defm : JWriteResFpuPair<WritePCmpIStrM, [JFPU1, JVALU1, JFPA, JALU0], 8, [1, 2, 1, 1], 3>;
defm : JWriteResFpuPair<WritePCmpEStrI, [JFPU1, JSAGU, JLAGU, JVALU, JVALU1, JFPA, JALU0], 14, [1, 2, 2, 6, 4, 1, 1], 9>;
defm : JWriteResFpuPair<WritePCmpEStrM, [JFPU1, JSAGU, JLAGU, JVALU, JVALU1, JFPA, JALU0], 14, [1, 2, 2, 6, 4, 1, 1], 9>;

////////////////////////////////////////////////////////////////////////////////
// MOVMSK Instructions.
////////////////////////////////////////////////////////////////////////////////

def : WriteRes<WriteFMOVMSK,    [JFPU0, JFPA, JALU0]> { let Latency = 3; }
def : WriteRes<WriteVecMOVMSK,  [JFPU0, JFPA, JALU0]> { let Latency = 3; }
def : WriteRes<WriteVecMOVMSKY, [JFPU0, JFPA, JALU0]> { let Latency = 3; }
def : WriteRes<WriteMMXMOVMSK,  [JFPU0, JFPA, JALU0]> { let Latency = 3; }

////////////////////////////////////////////////////////////////////////////////
// AES Instructions.
////////////////////////////////////////////////////////////////////////////////

defm : JWriteResFpuPair<WriteAESIMC,      [JFPU0, JVIMUL], 2>;
defm : JWriteResFpuPair<WriteAESKeyGen,   [JFPU0, JVIMUL], 2>;
defm : JWriteResFpuPair<WriteAESDecEnc,   [JFPU0, JVIMUL], 3, [1, 1], 2>;

////////////////////////////////////////////////////////////////////////////////
// Horizontal add/sub  instructions.
////////////////////////////////////////////////////////////////////////////////

defm : JWriteResFpuPair<WriteFHAdd,         [JFPU0, JFPA], 3>;
defm : JWriteResYMMPair<WriteFHAddY,        [JFPU0, JFPA], 3, [2,2], 2>;
defm : JWriteResFpuPair<WritePHAdd,       [JFPU01, JVALU], 1>;
defm : JWriteResFpuPair<WritePHAddY,      [JFPU01, JVALU], 1>;

////////////////////////////////////////////////////////////////////////////////
// Carry-less multiplication instructions.
////////////////////////////////////////////////////////////////////////////////

defm : JWriteResFpuPair<WriteCLMul,       [JFPU0, JVIMUL], 2>;

////////////////////////////////////////////////////////////////////////////////
// SSE4A instructions.
////////////////////////////////////////////////////////////////////////////////

def JWriteINSERTQ: SchedWriteRes<[JFPU01, JVALU]> {
  let Latency = 2;
  let ResourceCycles = [1, 4];
}
def : InstRW<[JWriteINSERTQ], (instrs INSERTQ, INSERTQI)>;

////////////////////////////////////////////////////////////////////////////////
// F16C instructions.
////////////////////////////////////////////////////////////////////////////////

def JWriteCVTPS2PHY: SchedWriteRes<[JFPU1, JSTC, JFPX]> {
  let Latency = 6;
  let ResourceCycles = [2, 2, 2];
  let NumMicroOps = 3;
}
def : InstRW<[JWriteCVTPS2PHY], (instrs VCVTPS2PHYrr)>;

def JWriteCVTPS2PHYSt: SchedWriteRes<[JFPU1, JSTC, JFPX, JSAGU]> {
  let Latency = 7;
  let ResourceCycles = [2, 2, 2, 1];
  let NumMicroOps = 3;
}
def : InstRW<[JWriteCVTPS2PHYSt], (instrs VCVTPS2PHYmr)>;

def JWriteCVTPH2PSY: SchedWriteRes<[JFPU1, JSTC]> {
  let Latency = 3;
  let ResourceCycles = [2, 2];
  let NumMicroOps = 2;
}
def : InstRW<[JWriteCVTPH2PSY], (instrs VCVTPH2PSYrr)>;

def JWriteCVTPH2PSYLd: SchedWriteRes<[JLAGU, JFPU1, JSTC]> {
  let Latency = 8;
  let ResourceCycles = [1, 2, 2];
  let NumMicroOps = 2;
}
def : InstRW<[JWriteCVTPH2PSYLd], (instrs VCVTPH2PSYrm)>;

////////////////////////////////////////////////////////////////////////////////
// AVX instructions.
////////////////////////////////////////////////////////////////////////////////

def JWriteVMULYPD: SchedWriteRes<[JFPU1, JFPM]> {
  let Latency = 4;
  let ResourceCycles = [2, 4];
  let NumMicroOps = 2;
}
def : InstRW<[JWriteVMULYPD], (instrs VMULPDYrr)>;

def JWriteVMULYPDLd: SchedWriteRes<[JLAGU, JFPU1, JFPM]> {
  let Latency = 9;
  let ResourceCycles = [2, 2, 4];
  let NumMicroOps = 2;
}
def : InstRW<[JWriteVMULYPDLd, ReadAfterLd], (instrs VMULPDYrm)>;

def JWriteVMULPD: SchedWriteRes<[JFPU1, JFPM]> {
  let Latency = 4;
  let ResourceCycles = [1, 2];
}
def : InstRW<[JWriteVMULPD], (instrs MULPDrr, MULSDrr, VMULPDrr, VMULSDrr)>;

def JWriteVMULPDLd: SchedWriteRes<[JLAGU, JFPU1, JFPM]> {
  let Latency = 9;
  let ResourceCycles = [1, 1, 2];
}
def : InstRW<[JWriteVMULPDLd], (instrs MULPDrm, MULSDrm, VMULPDrm, VMULSDrm)>;

def JWriteVCVTY: SchedWriteRes<[JFPU1, JSTC]> {
  let Latency = 3;
  let ResourceCycles = [2, 2];
  let NumMicroOps = 2;
}
def : InstRW<[JWriteVCVTY], (instrs VCVTDQ2PDYrr, VCVTDQ2PSYrr,
                                    VCVTPS2DQYrr, VCVTTPS2DQYrr)>;

def JWriteVCVTYLd: SchedWriteRes<[JLAGU, JFPU1, JSTC]> {
  let Latency = 8;
  let ResourceCycles = [2, 2, 2];
  let NumMicroOps = 2;
}
def : InstRW<[JWriteVCVTYLd, ReadAfterLd], (instrs VCVTDQ2PDYrm, VCVTDQ2PSYrm,
                                                   VCVTPS2DQYrm, VCVTTPS2DQYrm)>;

def JWriteVMOVNTDQSt: SchedWriteRes<[JFPU1, JSTC, JSAGU]> {
  let Latency = 2;
}
def : InstRW<[JWriteVMOVNTDQSt], (instrs MOVNTDQmr, VMOVNTDQmr)>;

def JWriteMOVNTSt: SchedWriteRes<[JFPU1, JSTC, JSAGU]> {
  let Latency = 3;
}
def : InstRW<[JWriteMOVNTSt], (instrs MOVNTPDmr, MOVNTPSmr, MOVNTSD, MOVNTSS, VMOVNTPDmr, VMOVNTPSmr)>;

def JWriteVMOVNTPYSt: SchedWriteRes<[JFPU1, JSTC, JSAGU]> {
  let Latency = 3;
  let ResourceCycles = [2, 2, 2];
}
def : InstRW<[JWriteVMOVNTPYSt], (instrs VMOVNTDQYmr, VMOVNTPDYmr, VMOVNTPSYmr)>;

def JWriteVCVTPDY: SchedWriteRes<[JFPU1, JSTC, JFPX]> {
  let Latency = 6;
  let ResourceCycles = [2, 2, 4];
  let NumMicroOps = 3;
}
def : InstRW<[JWriteVCVTPDY], (instrs VCVTPD2DQYrr, VCVTTPD2DQYrr, VCVTPD2PSYrr)>;

def JWriteVCVTPDYLd: SchedWriteRes<[JLAGU, JFPU1, JSTC, JFPX]> {
  let Latency = 11;
  let ResourceCycles = [2, 2, 2, 4];
  let NumMicroOps = 3;
}
def : InstRW<[JWriteVCVTPDYLd, ReadAfterLd], (instrs VCVTPD2DQYrm, VCVTTPD2DQYrm, VCVTPD2PSYrm)>;

def JWriteVBROADCASTYLd: SchedWriteRes<[JLAGU, JFPU01, JFPX]> {
  let Latency = 6;
  let ResourceCycles = [1, 2, 4];
  let NumMicroOps = 2;
}
def : InstRW<[JWriteVBROADCASTYLd, ReadAfterLd], (instrs VBROADCASTSDYrm,
                                                         VBROADCASTSSYrm)>;

def JWriteVMaskMovLd: SchedWriteRes<[JLAGU, JFPU01, JFPX]> {
  let Latency = 6;
  let ResourceCycles = [1, 1, 2];
}
def : InstRW<[JWriteVMaskMovLd], (instrs VMASKMOVPDrm, VMASKMOVPSrm)>;

def JWriteVMaskMovYLd: SchedWriteRes<[JLAGU, JFPU01, JFPX]> {
  let Latency = 6;
  let ResourceCycles = [2, 2, 4];
  let NumMicroOps = 2;
}
def : InstRW<[JWriteVMaskMovYLd], (instrs VMASKMOVPDYrm, VMASKMOVPSYrm)>;

def JWriteVMaskMovSt: SchedWriteRes<[JFPU01, JFPX, JSAGU]> {
  let Latency = 6;
  let ResourceCycles = [1, 4, 1];
}
def : InstRW<[JWriteVMaskMovSt], (instrs VMASKMOVPDmr, VMASKMOVPSmr)>;

def JWriteVMaskMovYSt: SchedWriteRes<[JFPU01, JFPX, JSAGU]> {
  let Latency = 6;
  let ResourceCycles = [2, 4, 2];
  let NumMicroOps = 2;
}
def : InstRW<[JWriteVMaskMovYSt], (instrs VMASKMOVPDYmr, VMASKMOVPSYmr)>;

def JWriteVTESTY: SchedWriteRes<[JFPU01, JFPX, JFPA, JALU0]> {
  let Latency = 4;
  let ResourceCycles = [2, 2, 2, 1];
  let NumMicroOps = 3;
}
def : InstRW<[JWriteVTESTY], (instrs VPTESTYrr, VTESTPDYrr, VTESTPSYrr)>;

def JWriteVTESTYLd: SchedWriteRes<[JLAGU, JFPU01, JFPX, JFPA, JALU0]> {
  let Latency = 9;
  let ResourceCycles = [2, 2, 2, 2, 1];
  let NumMicroOps = 3;
}
def : InstRW<[JWriteVTESTYLd], (instrs VPTESTYrm, VTESTPDYrm, VTESTPSYrm)>;

def JWriteVTEST: SchedWriteRes<[JFPU0, JFPA, JALU0]> {
  let Latency = 3;
}
def : InstRW<[JWriteVTEST], (instrs PTESTrr, VPTESTrr, VTESTPDrr, VTESTPSrr)>;

def JWriteVTESTLd: SchedWriteRes<[JLAGU, JFPU0, JFPA, JALU0]> {
  let Latency = 8;
}
def : InstRW<[JWriteVTESTLd], (instrs PTESTrm, VPTESTrm, VTESTPDrm, VTESTPSrm)>;

def JWriteVSQRTPD: SchedWriteRes<[JFPU1, JFPM]> {
  let Latency = 27;
  let ResourceCycles = [1, 27];
}
def : InstRW<[JWriteVSQRTPD], (instrs SQRTPDr, VSQRTPDr,
                                      SQRTSDr, VSQRTSDr,
                                      SQRTSDr_Int, VSQRTSDr_Int)>;

def JWriteVSQRTPDLd: SchedWriteRes<[JLAGU, JFPU1, JFPM]> {
  let Latency = 32;
  let ResourceCycles = [1, 1, 27];
}
def : InstRW<[JWriteVSQRTPDLd], (instrs SQRTPDm, VSQRTPDm,
                                        SQRTSDm, VSQRTSDm,
                                        SQRTSDm_Int, VSQRTSDm_Int)>;

def JWriteVSQRTYPD: SchedWriteRes<[JFPU1, JFPM]> {
  let Latency = 54; // each uOp is 27cy.
  let ResourceCycles = [2, 54];
  let NumMicroOps = 2;
}
def : InstRW<[JWriteVSQRTYPD], (instrs VSQRTPDYr)>;

def JWriteVSQRTYPDLd: SchedWriteRes<[JLAGU, JFPU1, JFPM]> {
  let Latency = 59; // each uOp is 27cy (+5cy of memory load).
  let ResourceCycles = [2, 2, 54];
  let NumMicroOps = 2;
}
def : InstRW<[JWriteVSQRTYPDLd], (instrs VSQRTPDYm)>;

def JWriteJVZEROALL: SchedWriteRes<[]> {
  let Latency = 90;
  let NumMicroOps = 73;
}
def : InstRW<[JWriteJVZEROALL], (instrs VZEROALL)>;

def JWriteJVZEROUPPER: SchedWriteRes<[]> {
  let Latency = 46;
  let NumMicroOps = 37;
}
def : InstRW<[JWriteJVZEROUPPER], (instrs VZEROUPPER)>;
} // SchedModel