Target/AMDGPU/SIRegisterInfo.td

//===-- SIRegisterInfo.td - SI Register defs ---------------*- 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
//
//===----------------------------------------------------------------------===//

//===----------------------------------------------------------------------===//
//  Subregister declarations
//===----------------------------------------------------------------------===//

class Indexes<int N> {
  list<int> all = [0,   1,  2,  3,  4,  5,  6 , 7,
                   8,   9, 10, 11, 12, 13, 14, 15,
                   16, 17, 18, 19, 20, 21, 22, 23,
                   24, 25, 26, 27, 28, 29, 30, 31];

  // Returns list of indexes [0..N)
  list<int> slice =
    !foldl([]<int>, all, acc, cur,
           !listconcat(acc, !if(!lt(cur, N), [cur], [])));
}

let Namespace = "AMDGPU" in {

def lo16 : SubRegIndex<16, 0>;
def hi16 : SubRegIndex<16, 16>;

foreach Index = 0-31 in {
  def sub#Index : SubRegIndex<32, !shl(Index, 5)>;
}

foreach Index = 1-31 in {
  def sub#Index#_lo16 : ComposedSubRegIndex<!cast<SubRegIndex>(sub#Index), lo16>;
  def sub#Index#_hi16 : ComposedSubRegIndex<!cast<SubRegIndex>(sub#Index), hi16>;
}

foreach Size = {2-6,8,16} in {
  foreach Index = Indexes<!add(33, !mul(Size, -1))>.slice in {
    def !foldl("", Indexes<Size>.slice, acc, cur,
               !strconcat(acc#!if(!eq(acc,""),"","_"), "sub"#!add(cur, Index))) :
      SubRegIndex<!mul(Size, 32), !shl(Index, 5)> {
      let CoveringSubRegIndices =
        !foldl([]<SubRegIndex>, Indexes<Size>.slice, acc, cur,
               !listconcat(acc, [!cast<SubRegIndex>(sub#!add(cur, Index))]));
    }
  }
}

}

//===----------------------------------------------------------------------===//
//  Helpers
//===----------------------------------------------------------------------===//

class getSubRegs<int size> {
  list<SubRegIndex> ret2 = [sub0, sub1];
  list<SubRegIndex> ret3 = [sub0, sub1, sub2];
  list<SubRegIndex> ret4 = [sub0, sub1, sub2, sub3];
  list<SubRegIndex> ret5 = [sub0, sub1, sub2, sub3, sub4];
  list<SubRegIndex> ret6 = [sub0, sub1, sub2, sub3, sub4, sub5];
  list<SubRegIndex> ret8 = [sub0, sub1, sub2, sub3, sub4, sub5, sub6, sub7];
  list<SubRegIndex> ret16 = [sub0, sub1, sub2, sub3,
                             sub4, sub5, sub6, sub7,
                             sub8, sub9, sub10, sub11,
                             sub12, sub13, sub14, sub15];
  list<SubRegIndex> ret32 = [sub0, sub1, sub2, sub3,
                             sub4, sub5, sub6, sub7,
                             sub8, sub9, sub10, sub11,
                             sub12, sub13, sub14, sub15,
                             sub16, sub17, sub18, sub19,
                             sub20, sub21, sub22, sub23,
                             sub24, sub25, sub26, sub27,
                             sub28, sub29, sub30, sub31];

  list<SubRegIndex> ret = !if(!eq(size, 2), ret2,
                              !if(!eq(size, 3), ret3,
                                  !if(!eq(size, 4), ret4,
                                      !if(!eq(size, 5), ret5,
                                          !if(!eq(size, 6), ret6,
                                              !if(!eq(size, 8), ret8,
                                                  !if(!eq(size, 16), ret16,
                                                      ret32)))))));
}

// Generates list of sequential register tuple names.
// E.g. RegSeq<3,2,2,"s">.ret -> [ "s[0:1]", "s[2:3]" ]
class RegSeqNames<int last_reg, int stride, int size, string prefix,
                  int start = 0> {
  int next = !add(start, stride);
  int end_reg = !add(!add(start, size), -1);
  list<string> ret =
    !if(!le(end_reg, last_reg),
        !listconcat([prefix # "[" # start # ":" # end_reg # "]"],
                    RegSeqNames<last_reg, stride, size, prefix, next>.ret),
                    []);
}

// Generates list of dags for register tupless.
class RegSeqDags<RegisterClass RC, int last_reg, int stride, int size,
                int start = 0> {
  dag trunc_rc = (trunc RC,
                  !if(!and(!eq(stride, 1), !eq(start, 0)),
                      !add(!add(last_reg, 2), !mul(size, -1)),
                      !add(last_reg, 1)));
  list<dag> ret =
    !if(!lt(start, size),
        !listconcat([(add (decimate (shl trunc_rc, start), stride))],
                    RegSeqDags<RC, last_reg, stride, size, !add(start, 1)>.ret),
        []);
}

class SIRegisterTuples<list<SubRegIndex> Indices, RegisterClass RC,
                       int last_reg, int stride, int size, string prefix> :
  RegisterTuples<Indices,
                 RegSeqDags<RC, last_reg, stride, size>.ret,
                 RegSeqNames<last_reg, stride, size, prefix>.ret>;

//===----------------------------------------------------------------------===//
//  Declarations that describe the SI registers
//===----------------------------------------------------------------------===//
class SIReg <string n, bits<16> regIdx = 0> :
  Register<n> {
  let Namespace = "AMDGPU";
  let HWEncoding = regIdx;
}

class SIRegWithSubRegs <string n, list<Register> subregs, bits<16> regIdx> :
  RegisterWithSubRegs<n, subregs> {
}

multiclass SIRegLoHi16 <string n, bits<16> regIdx, bit ArtificialHigh = 1,
                        bit HWEncodingHigh = 0> {
  // There is no special encoding for 16 bit subregs, these are not real
  // registers but rather operands for instructions preserving other 16 bits
  // of the result or reading just 16 bits of a 32 bit VGPR.
  // It is encoded as a corresponding 32 bit register.
  // Non-VGPR register classes use it as we need to have matching subregisters
  // to move instructions and data between ALUs.
  def _LO16 : SIReg<n#".l", regIdx> {
    let HWEncoding{8} = HWEncodingHigh;
  }
  def _HI16 : SIReg<!if(ArtificialHigh, "", n#".h"), regIdx> {
    let isArtificial = ArtificialHigh;
    let HWEncoding{8} = HWEncodingHigh;
  }
  def "" : RegisterWithSubRegs<n, [!cast<Register>(NAME#"_LO16"),
                                   !cast<Register>(NAME#"_HI16")]> {
    let Namespace = "AMDGPU";
    let SubRegIndices = [lo16, hi16];
    let CoveredBySubRegs = !if(ArtificialHigh,0,1);
    let HWEncoding = regIdx;
    let HWEncoding{8} = HWEncodingHigh;
  }
}

// Special Registers
defm VCC_LO : SIRegLoHi16<"vcc_lo", 106>;
defm VCC_HI : SIRegLoHi16<"vcc_hi", 107>;

// Pseudo-registers: Used as placeholders during isel and immediately
// replaced, never seeing the verifier.
def PRIVATE_RSRC_REG : SIReg<"private_rsrc", 0>;
def FP_REG : SIReg<"fp", 0>;
def SP_REG : SIReg<"sp", 0>;

// Pseudo-register to represent the program-counter DWARF register.
def PC_REG : SIReg<"pc", 0>, DwarfRegNum<[16, 16]> {
  // There is no physical register corresponding to a "program counter", but
  // we need to encode the concept in debug information in order to represent
  // things like the return value in unwind information.
  let isArtificial = 1;
}

// VCC for 64-bit instructions
def VCC : RegisterWithSubRegs<"vcc", [VCC_LO, VCC_HI]> {
  let Namespace = "AMDGPU";
  let SubRegIndices = [sub0, sub1];
  let HWEncoding = 106;
}

defm EXEC_LO : SIRegLoHi16<"exec_lo", 126>, DwarfRegNum<[1, 1]>;
defm EXEC_HI : SIRegLoHi16<"exec_hi", 127>;

def EXEC : RegisterWithSubRegs<"exec", [EXEC_LO, EXEC_HI]>, DwarfRegNum<[17, 1]> {
  let Namespace = "AMDGPU";
  let SubRegIndices = [sub0, sub1];
  let HWEncoding = 126;
}

// 32-bit real registers, for MC only.
// May be used with both 32-bit and 64-bit operands.
defm SRC_VCCZ : SIRegLoHi16<"src_vccz", 251>;
defm SRC_EXECZ : SIRegLoHi16<"src_execz", 252>;
defm SRC_SCC : SIRegLoHi16<"src_scc", 253>;

// 1-bit pseudo register, for codegen only.
// Should never be emitted.
def SCC : SIReg<"scc">;

defm M0 : SIRegLoHi16 <"m0", 124>;
defm SGPR_NULL : SIRegLoHi16 <"null", 125>;

defm SRC_SHARED_BASE : SIRegLoHi16<"src_shared_base", 235>;
defm SRC_SHARED_LIMIT : SIRegLoHi16<"src_shared_limit", 236>;
defm SRC_PRIVATE_BASE : SIRegLoHi16<"src_private_base", 237>;
defm SRC_PRIVATE_LIMIT : SIRegLoHi16<"src_private_limit", 238>;
defm SRC_POPS_EXITING_WAVE_ID : SIRegLoHi16<"src_pops_exiting_wave_id", 239>;

// Not addressable
def MODE : SIReg <"mode", 0>;

def LDS_DIRECT : SIReg <"src_lds_direct", 254> {
  // There is no physical register corresponding to this. This is an
  // encoding value in a source field, which will ultimately trigger a
  // read from m0.
  let isArtificial = 1;
}

defm XNACK_MASK_LO : SIRegLoHi16<"xnack_mask_lo", 104>;
defm XNACK_MASK_HI : SIRegLoHi16<"xnack_mask_hi", 105>;

def XNACK_MASK :
    RegisterWithSubRegs<"xnack_mask", [XNACK_MASK_LO, XNACK_MASK_HI]> {
  let Namespace = "AMDGPU";
  let SubRegIndices = [sub0, sub1];
  let HWEncoding = 104;
}

// Trap handler registers
defm TBA_LO : SIRegLoHi16<"tba_lo", 108>;
defm TBA_HI : SIRegLoHi16<"tba_hi", 109>;

def TBA : RegisterWithSubRegs<"tba", [TBA_LO, TBA_HI]> {
  let Namespace = "AMDGPU";
  let SubRegIndices = [sub0, sub1];
  let HWEncoding = 108;
}

defm TMA_LO : SIRegLoHi16<"tma_lo", 110>;
defm TMA_HI : SIRegLoHi16<"tma_hi", 111>;

def TMA : RegisterWithSubRegs<"tma", [TMA_LO, TMA_HI]> {
  let Namespace = "AMDGPU";
  let SubRegIndices = [sub0, sub1];
  let HWEncoding = 110;
}

foreach Index = 0-15 in {
  defm TTMP#Index#_vi         : SIRegLoHi16<"ttmp"#Index, !add(112, Index)>;
  defm TTMP#Index#_gfx9_gfx10 : SIRegLoHi16<"ttmp"#Index, !add(108, Index)>;
  defm TTMP#Index             : SIRegLoHi16<"ttmp"#Index, 0>;
}

multiclass FLAT_SCR_LOHI_m <string n, bits<16> ci_e, bits<16> vi_e> {
  defm _ci : SIRegLoHi16<n, ci_e>;
  defm _vi : SIRegLoHi16<n, vi_e>;
  defm "" : SIRegLoHi16<n, 0>;
}

class FlatReg <Register lo, Register hi, bits<16> encoding> :
    RegisterWithSubRegs<"flat_scratch", [lo, hi]> {
  let Namespace = "AMDGPU";
  let SubRegIndices = [sub0, sub1];
  let HWEncoding = encoding;
}

defm FLAT_SCR_LO : FLAT_SCR_LOHI_m<"flat_scratch_lo", 104, 102>; // Offset in units of 256-bytes.
defm FLAT_SCR_HI : FLAT_SCR_LOHI_m<"flat_scratch_hi", 105, 103>; // Size is the per-thread scratch size, in bytes.

def FLAT_SCR_ci : FlatReg<FLAT_SCR_LO_ci, FLAT_SCR_HI_ci, 104>;
def FLAT_SCR_vi : FlatReg<FLAT_SCR_LO_vi, FLAT_SCR_HI_vi, 102>;
def FLAT_SCR : FlatReg<FLAT_SCR_LO, FLAT_SCR_HI, 0>;

// SGPR registers
foreach Index = 0-105 in {
  defm SGPR#Index :
     SIRegLoHi16 <"s"#Index, Index>,
     DwarfRegNum<[!if(!le(Index, 63), !add(Index, 32), !add(Index, 1024)),
                  !if(!le(Index, 63), !add(Index, 32), !add(Index, 1024))]>;
}

// VGPR registers
foreach Index = 0-255 in {
  defm VGPR#Index :
    SIRegLoHi16 <"v"#Index, Index, 0, 1>,
    DwarfRegNum<[!add(Index, 2560), !add(Index, 1536)]>;
}

// AccVGPR registers
foreach Index = 0-255 in {
  defm AGPR#Index :
      SIRegLoHi16 <"a"#Index, Index, 1, 1>,
      DwarfRegNum<[!add(Index, 3072), !add(Index, 2048)]>;
}

//===----------------------------------------------------------------------===//
//  Groupings using register classes and tuples
//===----------------------------------------------------------------------===//

def SCC_CLASS : RegisterClass<"AMDGPU", [i1], 1, (add SCC)> {
  let CopyCost = -1;
  let isAllocatable = 0;
}

def M0_CLASS : RegisterClass<"AMDGPU", [i32], 32, (add M0)> {
  let CopyCost = 1;
  let isAllocatable = 0;
}

def M0_CLASS_LO16 : RegisterClass<"AMDGPU", [i16, f16], 16, (add M0_LO16)> {
  let CopyCost = 1;
  let Size = 16;
  let isAllocatable = 0;
}

// TODO: Do we need to set DwarfRegAlias on register tuples?

def SGPR_LO16 : RegisterClass<"AMDGPU", [i16, f16], 16,
                              (add (sequence "SGPR%u_LO16", 0, 105))> {
  let AllocationPriority = 9;
  let Size = 16;
  let GeneratePressureSet = 0;
}

def SGPR_HI16 : RegisterClass<"AMDGPU", [i16, f16], 16,
                              (add (sequence "SGPR%u_HI16", 0, 105))> {
  let isAllocatable = 0;
  let Size = 16;
  let GeneratePressureSet = 0;
}

// SGPR 32-bit registers
def SGPR_32 : RegisterClass<"AMDGPU", [i32, f32, i16, f16, v2i16, v2f16], 32,
                            (add (sequence "SGPR%u", 0, 105))> {
  // Give all SGPR classes higher priority than VGPR classes, because
  // we want to spill SGPRs to VGPRs.
  let AllocationPriority = 9;
  let GeneratePressureSet = 0;
}

// SGPR 64-bit registers
def SGPR_64Regs : SIRegisterTuples<getSubRegs<2>.ret, SGPR_32, 105, 2, 2, "s">;

// SGPR 96-bit registers. No operations use these, but for symmetry with 96-bit VGPRs.
def SGPR_96Regs : SIRegisterTuples<getSubRegs<3>.ret, SGPR_32, 105, 3, 3, "s">;

// SGPR 128-bit registers
def SGPR_128Regs : SIRegisterTuples<getSubRegs<4>.ret, SGPR_32, 105, 4, 4, "s">;

// SGPR 160-bit registers. No operations use these, but for symmetry with 160-bit VGPRs.
def SGPR_160Regs : SIRegisterTuples<getSubRegs<5>.ret, SGPR_32, 105, 4, 5, "s">;

// SGPR 192-bit registers
def SGPR_192Regs : SIRegisterTuples<getSubRegs<6>.ret, SGPR_32, 105, 4, 6, "s">;

// SGPR 256-bit registers
def SGPR_256Regs : SIRegisterTuples<getSubRegs<8>.ret, SGPR_32, 105, 4, 8, "s">;

// SGPR 512-bit registers
def SGPR_512Regs : SIRegisterTuples<getSubRegs<16>.ret, SGPR_32, 105, 4, 16, "s">;

// SGPR 1024-bit registers
def SGPR_1024Regs : SIRegisterTuples<getSubRegs<32>.ret, SGPR_32, 105, 4, 32, "s">;

// Trap handler TMP 32-bit registers
def TTMP_32 : RegisterClass<"AMDGPU", [i32, f32, v2i16, v2f16], 32,
                            (add (sequence "TTMP%u", 0, 15))> {
  let isAllocatable = 0;
}

def TTMP_LO16 : RegisterClass<"AMDGPU", [i16, f16], 16,
                              (add (sequence "TTMP%u_LO16", 0, 15))> {
  let Size = 16;
  let isAllocatable = 0;
}

// Trap handler TMP 64-bit registers
def TTMP_64Regs : SIRegisterTuples<getSubRegs<2>.ret, TTMP_32, 15, 2, 2, "ttmp">;

// Trap handler TMP 128-bit registers
def TTMP_128Regs : SIRegisterTuples<getSubRegs<4>.ret, TTMP_32, 15, 4, 4, "ttmp">;

def TTMP_256Regs : SIRegisterTuples<getSubRegs<8>.ret, TTMP_32, 15, 4, 8, "ttmp">;

def TTMP_512Regs : SIRegisterTuples<getSubRegs<16>.ret, TTMP_32, 15, 4, 16, "ttmp">;

class TmpRegTuplesBase<int index, int size,
                       list<Register> subRegs,
                       list<SubRegIndex> indices = getSubRegs<size>.ret,
                       int index1 = !add(index, !add(size, -1)),
                       string name = "ttmp["#index#":"#index1#"]"> :
  RegisterWithSubRegs<name, subRegs> {
  let HWEncoding = subRegs[0].HWEncoding;
  let SubRegIndices = indices;
}

class TmpRegTuples<string tgt,
                   int size,
                   int index0,
                   int index1 = !add(index0, 1),
                   int index2 = !add(index0, !if(!eq(size, 2), 1, 2)),
                   int index3 = !add(index0, !if(!eq(size, 2), 1, 3)),
                   int index4 = !add(index0, !if(!eq(size, 8), 4, 1)),
                   int index5 = !add(index0, !if(!eq(size, 8), 5, 1)),
                   int index6 = !add(index0, !if(!eq(size, 8), 6, 1)),
                   int index7 = !add(index0, !if(!eq(size, 8), 7, 1)),
                   Register r0 = !cast<Register>("TTMP"#index0#tgt),
                   Register r1 = !cast<Register>("TTMP"#index1#tgt),
                   Register r2 = !cast<Register>("TTMP"#index2#tgt),
                   Register r3 = !cast<Register>("TTMP"#index3#tgt),
                   Register r4 = !cast<Register>("TTMP"#index4#tgt),
                   Register r5 = !cast<Register>("TTMP"#index5#tgt),
                   Register r6 = !cast<Register>("TTMP"#index6#tgt),
                   Register r7 = !cast<Register>("TTMP"#index7#tgt)> :
  TmpRegTuplesBase<index0, size,
                   !if(!eq(size, 2), [r0, r1],
                       !if(!eq(size, 4), [r0, r1, r2, r3],
                                         [r0, r1, r2, r3, r4, r5, r6, r7])),
                   getSubRegs<size>.ret>;

foreach Index = {0, 2, 4, 6, 8, 10, 12, 14} in {
  def TTMP#Index#_TTMP#!add(Index,1)#_vi         : TmpRegTuples<"_vi",   2, Index>;
  def TTMP#Index#_TTMP#!add(Index,1)#_gfx9_gfx10 : TmpRegTuples<"_gfx9_gfx10", 2, Index>;
}

foreach Index = {0, 4, 8, 12} in {
  def TTMP#Index#_TTMP#!add(Index,1)#
                 _TTMP#!add(Index,2)#
                 _TTMP#!add(Index,3)#_vi : TmpRegTuples<"_vi",   4, Index>;
  def TTMP#Index#_TTMP#!add(Index,1)#
                 _TTMP#!add(Index,2)#
                 _TTMP#!add(Index,3)#_gfx9_gfx10 : TmpRegTuples<"_gfx9_gfx10", 4, Index>;
}

foreach Index = {0, 4, 8} in {
  def TTMP#Index#_TTMP#!add(Index,1)#
                 _TTMP#!add(Index,2)#
                 _TTMP#!add(Index,3)#
                 _TTMP#!add(Index,4)#
                 _TTMP#!add(Index,5)#
                 _TTMP#!add(Index,6)#
                 _TTMP#!add(Index,7)#_vi : TmpRegTuples<"_vi",   8, Index>;
  def TTMP#Index#_TTMP#!add(Index,1)#
                 _TTMP#!add(Index,2)#
                 _TTMP#!add(Index,3)#
                 _TTMP#!add(Index,4)#
                 _TTMP#!add(Index,5)#
                 _TTMP#!add(Index,6)#
                 _TTMP#!add(Index,7)#_gfx9_gfx10 : TmpRegTuples<"_gfx9_gfx10", 8, Index>;
}

def TTMP0_TTMP1_TTMP2_TTMP3_TTMP4_TTMP5_TTMP6_TTMP7_TTMP8_TTMP9_TTMP10_TTMP11_TTMP12_TTMP13_TTMP14_TTMP15_vi :
  TmpRegTuplesBase<0, 16,
                   [TTMP0_vi, TTMP1_vi, TTMP2_vi, TTMP3_vi,
                    TTMP4_vi, TTMP5_vi, TTMP6_vi, TTMP7_vi,
                    TTMP8_vi, TTMP9_vi, TTMP10_vi, TTMP11_vi,
                    TTMP12_vi, TTMP13_vi, TTMP14_vi, TTMP15_vi]>;

def TTMP0_TTMP1_TTMP2_TTMP3_TTMP4_TTMP5_TTMP6_TTMP7_TTMP8_TTMP9_TTMP10_TTMP11_TTMP12_TTMP13_TTMP14_TTMP15_gfx9_gfx10 :
  TmpRegTuplesBase<0, 16,
                   [TTMP0_gfx9_gfx10, TTMP1_gfx9_gfx10, TTMP2_gfx9_gfx10, TTMP3_gfx9_gfx10,
                    TTMP4_gfx9_gfx10, TTMP5_gfx9_gfx10, TTMP6_gfx9_gfx10, TTMP7_gfx9_gfx10,
                    TTMP8_gfx9_gfx10, TTMP9_gfx9_gfx10, TTMP10_gfx9_gfx10, TTMP11_gfx9_gfx10,
                    TTMP12_gfx9_gfx10, TTMP13_gfx9_gfx10, TTMP14_gfx9_gfx10, TTMP15_gfx9_gfx10]>;

class RegisterTypes<list<ValueType> reg_types> {
  list<ValueType> types = reg_types;
}

def Reg16Types : RegisterTypes<[i16, f16]>;
def Reg32Types : RegisterTypes<[i32, f32, v2i16, v2f16, p2, p3, p5, p6]>;

def VGPR_LO16 : RegisterClass<"AMDGPU", Reg16Types.types, 16,
                              (add (sequence "VGPR%u_LO16", 0, 255))> {
  let AllocationPriority = 1;
  let Size = 16;
  let GeneratePressureSet = 0;
}

def VGPR_HI16 : RegisterClass<"AMDGPU", Reg16Types.types, 16,
                              (add (sequence "VGPR%u_HI16", 0, 255))> {
  let AllocationPriority = 1;
  let Size = 16;
  let GeneratePressureSet = 0;
}

// VGPR 32-bit registers
// i16/f16 only on VI+
def VGPR_32 : RegisterClass<"AMDGPU", !listconcat(Reg32Types.types, Reg16Types.types), 32,
                            (add (sequence "VGPR%u", 0, 255))> {
  let AllocationPriority = 1;
  let Size = 32;
  let Weight = 1;
}

// VGPR 64-bit registers
def VGPR_64 : SIRegisterTuples<getSubRegs<2>.ret, VGPR_32, 255, 1, 2, "v">;

// VGPR 96-bit registers
def VGPR_96 : SIRegisterTuples<getSubRegs<3>.ret, VGPR_32, 255, 1, 3, "v">;

// VGPR 128-bit registers
def VGPR_128 : SIRegisterTuples<getSubRegs<4>.ret, VGPR_32, 255, 1, 4, "v">;

// VGPR 160-bit registers
def VGPR_160 : SIRegisterTuples<getSubRegs<5>.ret, VGPR_32, 255, 1, 5, "v">;

// VGPR 192-bit registers
def VGPR_192 : SIRegisterTuples<getSubRegs<6>.ret, VGPR_32, 255, 1, 6, "v">;

// VGPR 256-bit registers
def VGPR_256 : SIRegisterTuples<getSubRegs<8>.ret, VGPR_32, 255, 1, 8, "v">;

// VGPR 512-bit registers
def VGPR_512 : SIRegisterTuples<getSubRegs<16>.ret, VGPR_32, 255, 1, 16, "v">;

// VGPR 1024-bit registers
def VGPR_1024 : SIRegisterTuples<getSubRegs<32>.ret, VGPR_32, 255, 1, 32, "v">;

def AGPR_LO16 : RegisterClass<"AMDGPU", Reg16Types.types, 16,
                              (add (sequence "AGPR%u_LO16", 0, 255))> {
  let isAllocatable = 0;
  let Size = 16;
  let GeneratePressureSet = 0;
}

// AccVGPR 32-bit registers
def AGPR_32 : RegisterClass<"AMDGPU", [i32, f32, i16, f16, v2i16, v2f16], 32,
                            (add (sequence "AGPR%u", 0, 255))> {
  let AllocationPriority = 1;
  let Size = 32;
  let Weight = 1;
}

// AGPR 64-bit registers
def AGPR_64 : SIRegisterTuples<getSubRegs<2>.ret, AGPR_32, 255, 1, 2, "a">;

// AGPR 96-bit registers
def AGPR_96 : SIRegisterTuples<getSubRegs<3>.ret, AGPR_32, 255, 1, 3, "a">;

// AGPR 128-bit registers
def AGPR_128 : SIRegisterTuples<getSubRegs<4>.ret, AGPR_32, 255, 1, 4, "a">;

// AGPR 160-bit registers
def AGPR_160 : SIRegisterTuples<getSubRegs<5>.ret, AGPR_32, 255, 1, 5, "a">;

// AGPR 192-bit registers
def AGPR_192 : SIRegisterTuples<getSubRegs<6>.ret, AGPR_32, 255, 1, 6, "a">;

// AGPR 256-bit registers
def AGPR_256 : SIRegisterTuples<getSubRegs<8>.ret, AGPR_32, 255, 1, 8, "a">;

// AGPR 512-bit registers
def AGPR_512 : SIRegisterTuples<getSubRegs<16>.ret, AGPR_32, 255, 1, 16, "a">;

// AGPR 1024-bit registers
def AGPR_1024 : SIRegisterTuples<getSubRegs<32>.ret, AGPR_32, 255, 1, 32, "a">;

//===----------------------------------------------------------------------===//
//  Register classes used as source and destination
//===----------------------------------------------------------------------===//

def Pseudo_SReg_32 : RegisterClass<"AMDGPU", [i32, f32, i16, f16, v2i16, v2f16], 32,
  (add FP_REG, SP_REG)> {
  let isAllocatable = 0;
  let CopyCost = -1;
}

def Pseudo_SReg_128 : RegisterClass<"AMDGPU", [v4i32, v2i64, v2f64], 32,
  (add PRIVATE_RSRC_REG)> {
  let isAllocatable = 0;
  let CopyCost = -1;
}

def LDS_DIRECT_CLASS : RegisterClass<"AMDGPU", [i32], 32,
  (add LDS_DIRECT)> {
  let isAllocatable = 0;
  let CopyCost = -1;
}

let GeneratePressureSet = 0 in {
// Subset of SReg_32 without M0 for SMRD instructions and alike.
// See comments in SIInstructions.td for more info.
def SReg_32_XM0_XEXEC : RegisterClass<"AMDGPU", [i32, f32, i16, f16, v2i16, v2f16, i1], 32,
  (add SGPR_32, VCC_LO, VCC_HI, FLAT_SCR_LO, FLAT_SCR_HI, XNACK_MASK_LO, XNACK_MASK_HI,
   SGPR_NULL, TTMP_32, TMA_LO, TMA_HI, TBA_LO, TBA_HI, SRC_SHARED_BASE, SRC_SHARED_LIMIT,
   SRC_PRIVATE_BASE, SRC_PRIVATE_LIMIT, SRC_POPS_EXITING_WAVE_ID,
   SRC_VCCZ, SRC_EXECZ, SRC_SCC)> {
  let AllocationPriority = 10;
}

def SReg_LO16_XM0_XEXEC : RegisterClass<"AMDGPU", [i16, f16], 16,
  (add SGPR_LO16, VCC_LO_LO16, VCC_HI_LO16, FLAT_SCR_LO_LO16, FLAT_SCR_HI_LO16,
   XNACK_MASK_LO_LO16, XNACK_MASK_HI_LO16, SGPR_NULL_LO16, TTMP_LO16, TMA_LO_LO16,
   TMA_HI_LO16, TBA_LO_LO16, TBA_HI_LO16, SRC_SHARED_BASE_LO16,
   SRC_SHARED_LIMIT_LO16, SRC_PRIVATE_BASE_LO16, SRC_PRIVATE_LIMIT_LO16,
   SRC_POPS_EXITING_WAVE_ID_LO16, SRC_VCCZ_LO16, SRC_EXECZ_LO16, SRC_SCC_LO16)> {
  let Size = 16;
  let AllocationPriority = 10;
}

def SReg_32_XEXEC_HI : RegisterClass<"AMDGPU", [i32, f32, i16, f16, v2i16, v2f16, i1], 32,
  (add SReg_32_XM0_XEXEC, EXEC_LO, M0_CLASS)> {
  let AllocationPriority = 10;
}

def SReg_LO16_XEXEC_HI : RegisterClass<"AMDGPU", [i16, f16], 16,
  (add SReg_LO16_XM0_XEXEC, EXEC_LO_LO16, M0_CLASS_LO16)> {
  let Size = 16;
  let AllocationPriority = 10;
}

def SReg_32_XM0 : RegisterClass<"AMDGPU", [i32, f32, i16, f16, v2i16, v2f16, i1], 32,
  (add SReg_32_XM0_XEXEC, EXEC_LO, EXEC_HI)> {
  let AllocationPriority = 10;
}

def SReg_LO16_XM0 : RegisterClass<"AMDGPU", [i16, f16], 16,
  (add SReg_LO16_XM0_XEXEC, EXEC_LO_LO16, EXEC_HI_LO16)> {
  let Size = 16;
  let AllocationPriority = 10;
}

def SReg_LO16 : RegisterClass<"AMDGPU", [i16, f16], 16,
  (add SGPR_LO16, SReg_LO16_XM0, M0_CLASS_LO16, EXEC_LO_LO16, EXEC_HI_LO16, SReg_LO16_XEXEC_HI)> {
  let Size = 16;
  let AllocationPriority = 10;
}
} // End GeneratePressureSet = 0

// Register class for all scalar registers (SGPRs + Special Registers)
def SReg_32 : RegisterClass<"AMDGPU", [i32, f32, i16, f16, v2i16, v2f16, i1], 32,
  (add SReg_32_XM0, M0_CLASS, EXEC_LO, EXEC_HI, SReg_32_XEXEC_HI)> {
  let AllocationPriority = 10;
}

let GeneratePressureSet = 0 in {
def SRegOrLds_32 : RegisterClass<"AMDGPU", [i32, f32, i16, f16, v2i16, v2f16], 32,
  (add SReg_32, LDS_DIRECT_CLASS)> {
  let isAllocatable = 0;
}

def SGPR_64 : RegisterClass<"AMDGPU", [v2i32, i64, v2f32, f64, v4i16, v4f16], 32,
                            (add SGPR_64Regs)> {
  let CopyCost = 1;
  let AllocationPriority = 11;
}

// CCR (call clobbered registers) SGPR 64-bit registers
def CCR_SGPR_64 : RegisterClass<"AMDGPU", SGPR_64.RegTypes, 32,
                                (add (trunc SGPR_64, 16))> {
  let CopyCost = SGPR_64.CopyCost;
  let AllocationPriority = SGPR_64.AllocationPriority;
}

def TTMP_64 : RegisterClass<"AMDGPU", [v2i32, i64, f64, v4i16, v4f16], 32,
                            (add TTMP_64Regs)> {
  let isAllocatable = 0;
}

def SReg_64_XEXEC : RegisterClass<"AMDGPU", [v2i32, i64, v2f32, f64, i1, v4i16, v4f16], 32,
  (add SGPR_64, VCC, FLAT_SCR, XNACK_MASK, TTMP_64, TBA, TMA)> {
  let CopyCost = 1;
  let AllocationPriority = 13;
}

def SReg_64 : RegisterClass<"AMDGPU", [v2i32, i64, v2f32, f64, i1, v4i16, v4f16], 32,
  (add SReg_64_XEXEC, EXEC)> {
  let CopyCost = 1;
  let AllocationPriority = 13;
}

def SReg_1_XEXEC : RegisterClass<"AMDGPU", [i1], 32,
  (add SReg_64_XEXEC, SReg_32_XM0_XEXEC)> {
  let CopyCost = 1;
  let isAllocatable = 0;
}

def SReg_1 : RegisterClass<"AMDGPU", [i1], 32,
  (add SReg_1_XEXEC, EXEC, EXEC_LO)> {
  let CopyCost = 1;
  let isAllocatable = 0;
}

// Requires 2 s_mov_b64 to copy
let CopyCost = 2 in {

// There are no 3-component scalar instructions, but this is needed
// for symmetry with VGPRs.
def SGPR_96 : RegisterClass<"AMDGPU", [v3i32, v3f32], 32,
  (add SGPR_96Regs)> {
  let AllocationPriority = 14;
}

def SReg_96 : RegisterClass<"AMDGPU", [v3i32, v3f32], 32,
  (add SGPR_96)> {
  let AllocationPriority = 14;
}

def SGPR_128 : RegisterClass<"AMDGPU", [v4i32, v4f32, v2i64], 32,
                             (add SGPR_128Regs)> {
  let AllocationPriority = 15;
}

def TTMP_128 : RegisterClass<"AMDGPU", [v4i32, v4f32, v2i64], 32,
                             (add TTMP_128Regs)> {
  let isAllocatable = 0;
}

def SReg_128 : RegisterClass<"AMDGPU", [v4i32, v4f32, v2i64, v2f64], 32,
                             (add SGPR_128, TTMP_128)> {
  let isAllocatable = 0;
}

} // End CopyCost = 2

// There are no 5-component scalar instructions, but this is needed
// for symmetry with VGPRs.
def SGPR_160 : RegisterClass<"AMDGPU", [v5i32, v5f32], 32,
                             (add SGPR_160Regs)> {
  let AllocationPriority = 16;
}

def SReg_160 : RegisterClass<"AMDGPU", [v5i32, v5f32], 32,
                             (add SGPR_160)> {
  // FIXME: Should be isAllocatable = 0, but that causes all TableGen-generated
  // subclasses of SGPR_160 to be marked unallocatable too.
}

def SGPR_192 : RegisterClass<"AMDGPU", [untyped], 32, (add SGPR_192Regs)> {
  let Size = 192;
  let AllocationPriority = 17;
}

def SReg_192 : RegisterClass<"AMDGPU", [untyped], 32, (add SGPR_192)> {
  let Size = 192;
  let isAllocatable = 0;
}

def SGPR_256 : RegisterClass<"AMDGPU", [v8i32, v8f32, v4i64, v4f64], 32, (add SGPR_256Regs)> {
  let AllocationPriority = 18;
}

def TTMP_256 : RegisterClass<"AMDGPU", [v8i32, v8f32, v4i64, v4f64], 32, (add TTMP_256Regs)> {
  let isAllocatable = 0;
}

def SReg_256 : RegisterClass<"AMDGPU", [v8i32, v8f32, v4i64, v4f64], 32,
                             (add SGPR_256, TTMP_256)> {
  // Requires 4 s_mov_b64 to copy
  let CopyCost = 4;
  let isAllocatable = 0;
}

def SGPR_512 : RegisterClass<"AMDGPU", [v16i32, v16f32, v8i64, v8f64], 32,
                             (add SGPR_512Regs)> {
  let AllocationPriority = 19;
}

def TTMP_512 : RegisterClass<"AMDGPU", [v16i32, v16f32, v8i64, v8f64], 32,
                             (add TTMP_512Regs)> {
  let isAllocatable = 0;
}

def SReg_512 : RegisterClass<"AMDGPU", [v16i32, v16f32, v8i64, v8f64], 32,
                             (add SGPR_512, TTMP_512)> {
  // Requires 8 s_mov_b64 to copy
  let CopyCost = 8;
  let isAllocatable = 0;
}

def VRegOrLds_32 : RegisterClass<"AMDGPU", [i32, f32, i16, f16, v2i16, v2f16], 32,
                                 (add VGPR_32, LDS_DIRECT_CLASS)> {
  let isAllocatable = 0;
}

def SGPR_1024 : RegisterClass<"AMDGPU", [v32i32, v32f32, v16i64, v16f64], 32,
                              (add SGPR_1024Regs)> {
  let AllocationPriority = 20;
}

def SReg_1024 : RegisterClass<"AMDGPU", [v32i32, v32f32, v16i64, v16f64], 32,
                              (add SGPR_1024)> {
  let CopyCost = 16;
  let isAllocatable = 0;
}

// Register class for all vector registers (VGPRs + Interpolation Registers)
class VRegClass<int numRegs, list<ValueType> regTypes, dag regList> :
    RegisterClass<"AMDGPU", regTypes, 32, regList> {
  let Size = !mul(numRegs, 32);

  // Requires n v_mov_b32 to copy
  let CopyCost = numRegs;
  let AllocationPriority = numRegs;
  let Weight = numRegs;
}

def VReg_64 : VRegClass<2, [i64, f64, v2i32, v2f32, v4f16, v4i16, p0, p1, p4],
                        (add VGPR_64)>;
def VReg_96 : VRegClass<3, [v3i32, v3f32], (add VGPR_96)>;
def VReg_128 : VRegClass<4, [v4i32, v4f32, v2i64, v2f64, i128], (add VGPR_128)>;
def VReg_160 : VRegClass<5, [v5i32, v5f32], (add VGPR_160)>;
def VReg_192 : VRegClass<6, [untyped], (add VGPR_192)>;
def VReg_256 : VRegClass<8, [v8i32, v8f32, v4i64, v4f64], (add VGPR_256)>;
def VReg_512 : VRegClass<16, [v16i32, v16f32, v8i64, v8f64], (add VGPR_512)>;
def VReg_1024 : VRegClass<32, [v32i32, v32f32, v16i64, v16f64], (add VGPR_1024)>;

class ARegClass<int numRegs, list<ValueType> regTypes, dag regList> :
    VRegClass<numRegs, regTypes, regList> {
  // Requires n v_accvgpr_write and n v_accvgpr_read to copy + burn 1 vgpr
  let CopyCost = !add(numRegs, numRegs, 1);
}

def AReg_64 : ARegClass<2, [i64, f64, v2i32, v2f32, v4f16, v4i16],
                        (add AGPR_64)>;
def AReg_96 : ARegClass<3, [v3i32, v3f32], (add AGPR_96)>;
def AReg_128 : ARegClass<4, [v4i32, v4f32, v2i64, v2f64], (add AGPR_128)>;
def AReg_160 : ARegClass<5, [v5i32, v5f32], (add AGPR_160)>;
def AReg_192 : ARegClass<6, [untyped], (add AGPR_192)>;
def AReg_256 : ARegClass<8, [v8i32, v8f32, v4i64, v4f64], (add AGPR_256)>;
def AReg_512 : ARegClass<16, [v16i32, v16f32, v8i64, v8f64], (add AGPR_512)>;
def AReg_1024 : ARegClass<32, [v32i32, v32f32, v16i64, v16f64], (add AGPR_1024)>;

} // End GeneratePressureSet = 0

// This is not a real register. This is just to have a register to add
// to VReg_1 that does not alias any real register that would
// introduce inferred register classess.
def ARTIFICIAL_VGPR : SIReg <"invalid vgpr", 0> {
  let isArtificial = 1;
}

let GeneratePressureSet = 0 in {
// FIXME: Should specify an empty set for this. No register should
// ever be allocated using VReg_1. This is a hack for SelectionDAG
// that should always be lowered by SILowerI1Copies. TableGen crashes
// on an empty register set, but also sorts register classes based on
// the number of registerss in them. Add only one register so this is
// sorted to the end and not preferred over VGPR_32.
def VReg_1 : RegisterClass<"AMDGPU", [i1], 32, (add ARTIFICIAL_VGPR)> {
  let Size = 1;
}

def VS_32 : RegisterClass<"AMDGPU", [i32, f32, i16, f16, v2i16, v2f16], 32,
                          (add VGPR_32, SReg_32, LDS_DIRECT_CLASS)> {
  let isAllocatable = 0;
}

def VS_64 : RegisterClass<"AMDGPU", [i64, f64], 32, (add VReg_64, SReg_64)> {
  let isAllocatable = 0;
}

def AV_32 : RegisterClass<"AMDGPU", [i32, f32, i16, f16, v2i16, v2f16], 32,
                          (add AGPR_32, VGPR_32)> {
  let isAllocatable = 0;
}

def AV_64 : RegisterClass<"AMDGPU", [i64, f64, v4f16], 32,
                          (add AReg_64, VReg_64)> {
  let isAllocatable = 0;
}
} // End GeneratePressureSet = 0

//===----------------------------------------------------------------------===//
//  Register operands
//===----------------------------------------------------------------------===//

class RegImmMatcher<string name> : AsmOperandClass {
  let Name = name;
  let RenderMethod = "addRegOrImmOperands";
}

multiclass SIRegOperand32 <string rc, string MatchName, string opType,
                           string rc_suffix = "_32"> {
  let OperandNamespace = "AMDGPU" in {
    def _b16 : RegisterOperand<!cast<RegisterClass>(rc#rc_suffix)> {
      let OperandType = opType#"_INT16";
      let ParserMatchClass = RegImmMatcher<MatchName#"B16">;
      let DecoderMethod = "decodeOperand_VSrc16";
    }

    def _f16 : RegisterOperand<!cast<RegisterClass>(rc#rc_suffix)> {
      let OperandType = opType#"_FP16";
      let ParserMatchClass = RegImmMatcher<MatchName#"F16">;
      let DecoderMethod = "decodeOperand_" # rc # "_16";
    }

    def _b32 : RegisterOperand<!cast<RegisterClass>(rc#rc_suffix)> {
      let OperandType = opType#"_INT32";
      let ParserMatchClass = RegImmMatcher<MatchName#"B32">;
      let DecoderMethod = "decodeOperand_" # rc # rc_suffix;
    }

    def _f32 : RegisterOperand<!cast<RegisterClass>(rc#rc_suffix)> {
      let OperandType = opType#"_FP32";
      let ParserMatchClass = RegImmMatcher<MatchName#"F32">;
      let DecoderMethod = "decodeOperand_" # rc # rc_suffix;
    }

    def _v2b16 : RegisterOperand<!cast<RegisterClass>(rc#rc_suffix)> {
      let OperandType = opType#"_V2INT16";
      let ParserMatchClass = RegImmMatcher<MatchName#"V2B16">;
      let DecoderMethod = "decodeOperand_VSrcV216";
    }

    def _v2f16 : RegisterOperand<!cast<RegisterClass>(rc#rc_suffix)> {
      let OperandType = opType#"_V2FP16";
      let ParserMatchClass = RegImmMatcher<MatchName#"V2F16">;
      let DecoderMethod = "decodeOperand_VSrcV216";
    }
  }
}

multiclass SIRegOperand <string rc, string MatchName, string opType> :
  SIRegOperand32<rc, MatchName, opType> {
  let OperandNamespace = "AMDGPU" in {
    def _b64 : RegisterOperand<!cast<RegisterClass>(rc#"_64")> {
      let OperandType = opType#"_INT64";
      let ParserMatchClass = RegImmMatcher<MatchName#"B64">;
    }

    def _f64 : RegisterOperand<!cast<RegisterClass>(rc#"_64")> {
      let OperandType = opType#"_FP64";
      let ParserMatchClass = RegImmMatcher<MatchName#"F64">;
    }
  }
}

// FIXME: 64-bit sources can sometimes use 32-bit constants.
multiclass RegImmOperand <string rc, string MatchName>
  : SIRegOperand<rc, MatchName, "OPERAND_REG_IMM">;

multiclass RegInlineOperand <string rc, string MatchName>
  : SIRegOperand<rc, MatchName, "OPERAND_REG_INLINE_C">;

multiclass RegInlineOperand32 <string rc, string MatchName,
                               string rc_suffix = "_32">
  : SIRegOperand32<rc, MatchName, "OPERAND_REG_INLINE_C", rc_suffix>;

multiclass RegInlineOperandAC <string rc, string MatchName,
                               string rc_suffix = "_32">
  : SIRegOperand32<rc, MatchName, "OPERAND_REG_INLINE_AC", rc_suffix>;

//===----------------------------------------------------------------------===//
//  SSrc_* Operands with an SGPR or a 32-bit immediate
//===----------------------------------------------------------------------===//

defm SSrc : RegImmOperand<"SReg", "SSrc">;

def SSrcOrLds_b32 : RegisterOperand<SRegOrLds_32> {
  let OperandNamespace = "AMDGPU";
  let OperandType = "OPERAND_REG_IMM_INT32";
  let ParserMatchClass = RegImmMatcher<"SSrcOrLdsB32">;
}

//===----------------------------------------------------------------------===//
//  SCSrc_* Operands with an SGPR or a inline constant
//===----------------------------------------------------------------------===//

defm SCSrc : RegInlineOperand<"SReg", "SCSrc"> ;

//===----------------------------------------------------------------------===//
//  VSrc_* Operands with an SGPR, VGPR or a 32-bit immediate
//===----------------------------------------------------------------------===//

defm VSrc : RegImmOperand<"VS", "VSrc">;

def VSrc_128 : RegisterOperand<VReg_128> {
  let DecoderMethod = "DecodeVS_128RegisterClass";
}

//===----------------------------------------------------------------------===//
//  VSrc_* Operands with an VGPR
//===----------------------------------------------------------------------===//

// This is for operands with the enum(9), VSrc encoding restriction,
// but only allows VGPRs.
def VRegSrc_32 : RegisterOperand<VGPR_32> {
  //let ParserMatchClass = RegImmMatcher<"VRegSrc32">;
  let DecoderMethod = "DecodeVS_32RegisterClass";
}

//===----------------------------------------------------------------------===//
//  ASrc_* Operands with an AccVGPR
//===----------------------------------------------------------------------===//

def ARegSrc_32 : RegisterOperand<AGPR_32> {
  let DecoderMethod = "DecodeAGPR_32RegisterClass";
  let EncoderMethod = "getAVOperandEncoding";
}

//===----------------------------------------------------------------------===//
//  VCSrc_* Operands with an SGPR, VGPR or an inline constant
//===----------------------------------------------------------------------===//

defm VCSrc : RegInlineOperand<"VS", "VCSrc">;

//===----------------------------------------------------------------------===//
//  VISrc_* Operands with a VGPR or an inline constant
//===----------------------------------------------------------------------===//

defm VISrc : RegInlineOperand32<"VGPR", "VISrc">;

//===----------------------------------------------------------------------===//
//  AVSrc_* Operands with an AGPR or VGPR
//===----------------------------------------------------------------------===//

def AVSrc_32 : RegisterOperand<AV_32> {
  let DecoderMethod = "DecodeAV_32RegisterClass";
  let EncoderMethod = "getAVOperandEncoding";
}

def AVSrc_64 : RegisterOperand<AV_64> {
  let DecoderMethod = "DecodeAV_64RegisterClass";
  let EncoderMethod = "getAVOperandEncoding";
}

//===----------------------------------------------------------------------===//
//  ACSrc_* Operands with an AGPR or an inline constant
//===----------------------------------------------------------------------===//

defm AISrc      : RegInlineOperandAC<"AGPR", "AISrc">;
defm AISrc_128  : RegInlineOperandAC<"AReg", "AISrc_128",  "_128">;
defm AISrc_512  : RegInlineOperandAC<"AReg", "AISrc_512",  "_512">;
defm AISrc_1024 : RegInlineOperandAC<"AReg", "AISrc_1024", "_1024">;