#include "AMDGPUInstPrinter.h"
#include "MCTargetDesc/AMDGPUMCTargetDesc.h"
+#include "SIDefines.h"
#include "llvm/MC/MCExpr.h"
#include "llvm/MC/MCInst.h"
+#include "llvm/MC/MCInstrInfo.h"
+#include "llvm/MC/MCRegisterInfo.h"
+#include "llvm/Support/MathExtras.h"
using namespace llvm;
printAnnotation(OS, Annot);
}
+void AMDGPUInstPrinter::printU8ImmOperand(const MCInst *MI, unsigned OpNo,
+ raw_ostream &O) {
+ O << formatHex(MI->getOperand(OpNo).getImm() & 0xff);
+}
+
+void AMDGPUInstPrinter::printU16ImmOperand(const MCInst *MI, unsigned OpNo,
+ raw_ostream &O) {
+ O << formatHex(MI->getOperand(OpNo).getImm() & 0xffff);
+}
+
+void AMDGPUInstPrinter::printU32ImmOperand(const MCInst *MI, unsigned OpNo,
+ raw_ostream &O) {
+ O << formatHex(MI->getOperand(OpNo).getImm() & 0xffffffff);
+}
+
+void AMDGPUInstPrinter::printU8ImmDecOperand(const MCInst *MI, unsigned OpNo,
+ raw_ostream &O) {
+ O << formatDec(MI->getOperand(OpNo).getImm() & 0xff);
+}
+
+void AMDGPUInstPrinter::printU16ImmDecOperand(const MCInst *MI, unsigned OpNo,
+ raw_ostream &O) {
+ O << formatDec(MI->getOperand(OpNo).getImm() & 0xffff);
+}
+
+void AMDGPUInstPrinter::printOffen(const MCInst *MI, unsigned OpNo,
+ raw_ostream &O) {
+ if (MI->getOperand(OpNo).getImm())
+ O << " offen";
+}
+
+void AMDGPUInstPrinter::printIdxen(const MCInst *MI, unsigned OpNo,
+ raw_ostream &O) {
+ if (MI->getOperand(OpNo).getImm())
+ O << " idxen";
+}
+
+void AMDGPUInstPrinter::printAddr64(const MCInst *MI, unsigned OpNo,
+ raw_ostream &O) {
+ if (MI->getOperand(OpNo).getImm())
+ O << " addr64";
+}
+
+void AMDGPUInstPrinter::printMBUFOffset(const MCInst *MI, unsigned OpNo,
+ raw_ostream &O) {
+ if (MI->getOperand(OpNo).getImm()) {
+ O << " offset:";
+ printU16ImmDecOperand(MI, OpNo, O);
+ }
+}
+
+void AMDGPUInstPrinter::printDSOffset(const MCInst *MI, unsigned OpNo,
+ raw_ostream &O) {
+ uint16_t Imm = MI->getOperand(OpNo).getImm();
+ if (Imm != 0) {
+ O << " offset:";
+ printU16ImmDecOperand(MI, OpNo, O);
+ }
+}
+
+void AMDGPUInstPrinter::printDSOffset0(const MCInst *MI, unsigned OpNo,
+ raw_ostream &O) {
+ O << " offset0:";
+ printU8ImmDecOperand(MI, OpNo, O);
+}
+
+void AMDGPUInstPrinter::printDSOffset1(const MCInst *MI, unsigned OpNo,
+ raw_ostream &O) {
+ O << " offset1:";
+ printU8ImmDecOperand(MI, OpNo, O);
+}
+
+void AMDGPUInstPrinter::printGLC(const MCInst *MI, unsigned OpNo,
+ raw_ostream &O) {
+ if (MI->getOperand(OpNo).getImm())
+ O << " glc";
+}
+
+void AMDGPUInstPrinter::printSLC(const MCInst *MI, unsigned OpNo,
+ raw_ostream &O) {
+ if (MI->getOperand(OpNo).getImm())
+ O << " slc";
+}
+
+void AMDGPUInstPrinter::printTFE(const MCInst *MI, unsigned OpNo,
+ raw_ostream &O) {
+ if (MI->getOperand(OpNo).getImm())
+ O << " tfe";
+}
+
+void AMDGPUInstPrinter::printRegOperand(unsigned reg, raw_ostream &O) {
+ switch (reg) {
+ case AMDGPU::VCC:
+ O << "vcc";
+ return;
+ case AMDGPU::SCC:
+ O << "scc";
+ return;
+ case AMDGPU::EXEC:
+ O << "exec";
+ return;
+ case AMDGPU::M0:
+ O << "m0";
+ return;
+ case AMDGPU::FLAT_SCR:
+ O << "flat_scratch";
+ return;
+ case AMDGPU::VCC_LO:
+ O << "vcc_lo";
+ return;
+ case AMDGPU::VCC_HI:
+ O << "vcc_hi";
+ return;
+ case AMDGPU::EXEC_LO:
+ O << "exec_lo";
+ return;
+ case AMDGPU::EXEC_HI:
+ O << "exec_hi";
+ return;
+ case AMDGPU::FLAT_SCR_LO:
+ O << "flat_scratch_lo";
+ return;
+ case AMDGPU::FLAT_SCR_HI:
+ O << "flat_scratch_hi";
+ return;
+ default:
+ break;
+ }
+
+ char Type;
+ unsigned NumRegs;
+
+ if (MRI.getRegClass(AMDGPU::VGPR_32RegClassID).contains(reg)) {
+ Type = 'v';
+ NumRegs = 1;
+ } else if (MRI.getRegClass(AMDGPU::SGPR_32RegClassID).contains(reg)) {
+ Type = 's';
+ NumRegs = 1;
+ } else if (MRI.getRegClass(AMDGPU::VReg_64RegClassID).contains(reg)) {
+ Type = 'v';
+ NumRegs = 2;
+ } else if (MRI.getRegClass(AMDGPU::SReg_64RegClassID).contains(reg)) {
+ Type = 's';
+ NumRegs = 2;
+ } else if (MRI.getRegClass(AMDGPU::VReg_128RegClassID).contains(reg)) {
+ Type = 'v';
+ NumRegs = 4;
+ } else if (MRI.getRegClass(AMDGPU::SReg_128RegClassID).contains(reg)) {
+ Type = 's';
+ NumRegs = 4;
+ } else if (MRI.getRegClass(AMDGPU::VReg_96RegClassID).contains(reg)) {
+ Type = 'v';
+ NumRegs = 3;
+ } else if (MRI.getRegClass(AMDGPU::VReg_256RegClassID).contains(reg)) {
+ Type = 'v';
+ NumRegs = 8;
+ } else if (MRI.getRegClass(AMDGPU::SReg_256RegClassID).contains(reg)) {
+ Type = 's';
+ NumRegs = 8;
+ } else if (MRI.getRegClass(AMDGPU::VReg_512RegClassID).contains(reg)) {
+ Type = 'v';
+ NumRegs = 16;
+ } else if (MRI.getRegClass(AMDGPU::SReg_512RegClassID).contains(reg)) {
+ Type = 's';
+ NumRegs = 16;
+ } else {
+ O << getRegisterName(reg);
+ return;
+ }
+
+ // The low 8 bits of the encoding value is the register index, for both VGPRs
+ // and SGPRs.
+ unsigned RegIdx = MRI.getEncodingValue(reg) & ((1 << 8) - 1);
+ if (NumRegs == 1) {
+ O << Type << RegIdx;
+ return;
+ }
+
+ O << Type << '[' << RegIdx << ':' << (RegIdx + NumRegs - 1) << ']';
+}
+
+void AMDGPUInstPrinter::printImmediate32(uint32_t Imm, raw_ostream &O) {
+ int32_t SImm = static_cast<int32_t>(Imm);
+ if (SImm >= -16 && SImm <= 64) {
+ O << SImm;
+ return;
+ }
+
+ if (Imm == FloatToBits(0.0f))
+ O << "0.0";
+ else if (Imm == FloatToBits(1.0f))
+ O << "1.0";
+ else if (Imm == FloatToBits(-1.0f))
+ O << "-1.0";
+ else if (Imm == FloatToBits(0.5f))
+ O << "0.5";
+ else if (Imm == FloatToBits(-0.5f))
+ O << "-0.5";
+ else if (Imm == FloatToBits(2.0f))
+ O << "2.0";
+ else if (Imm == FloatToBits(-2.0f))
+ O << "-2.0";
+ else if (Imm == FloatToBits(4.0f))
+ O << "4.0";
+ else if (Imm == FloatToBits(-4.0f))
+ O << "-4.0";
+ else
+ O << formatHex(static_cast<uint64_t>(Imm));
+}
+
+void AMDGPUInstPrinter::printImmediate64(uint64_t Imm, raw_ostream &O) {
+ int64_t SImm = static_cast<int64_t>(Imm);
+ if (SImm >= -16 && SImm <= 64) {
+ O << SImm;
+ return;
+ }
+
+ if (Imm == DoubleToBits(0.0))
+ O << "0.0";
+ else if (Imm == DoubleToBits(1.0))
+ O << "1.0";
+ else if (Imm == DoubleToBits(-1.0))
+ O << "-1.0";
+ else if (Imm == DoubleToBits(0.5))
+ O << "0.5";
+ else if (Imm == DoubleToBits(-0.5))
+ O << "-0.5";
+ else if (Imm == DoubleToBits(2.0))
+ O << "2.0";
+ else if (Imm == DoubleToBits(-2.0))
+ O << "-2.0";
+ else if (Imm == DoubleToBits(4.0))
+ O << "4.0";
+ else if (Imm == DoubleToBits(-4.0))
+ O << "-4.0";
+ else
+ llvm_unreachable("64-bit literal constants not supported");
+}
+
void AMDGPUInstPrinter::printOperand(const MCInst *MI, unsigned OpNo,
raw_ostream &O) {
if (Op.isReg()) {
switch (Op.getReg()) {
// This is the default predicate state, so we don't need to print it.
- case AMDGPU::PRED_SEL_OFF: break;
- default: O << getRegisterName(Op.getReg()); break;
+ case AMDGPU::PRED_SEL_OFF:
+ break;
+
+ default:
+ printRegOperand(Op.getReg(), O);
+ break;
}
} else if (Op.isImm()) {
- O << Op.getImm();
+ const MCInstrDesc &Desc = MII.get(MI->getOpcode());
+ int RCID = Desc.OpInfo[OpNo].RegClass;
+ if (RCID != -1) {
+ const MCRegisterClass &ImmRC = MRI.getRegClass(RCID);
+ if (ImmRC.getSize() == 4)
+ printImmediate32(Op.getImm(), O);
+ else if (ImmRC.getSize() == 8)
+ printImmediate64(Op.getImm(), O);
+ else
+ llvm_unreachable("Invalid register class size");
+ } else {
+ // We hit this for the immediate instruction bits that don't yet have a
+ // custom printer.
+ // TODO: Eventually this should be unnecessary.
+ O << formatDec(Op.getImm());
+ }
} else if (Op.isFPImm()) {
- O << Op.getFPImm();
+ // We special case 0.0 because otherwise it will be printed as an integer.
+ if (Op.getFPImm() == 0.0)
+ O << "0.0";
+ else {
+ const MCInstrDesc &Desc = MII.get(MI->getOpcode());
+ const MCRegisterClass &ImmRC = MRI.getRegClass(Desc.OpInfo[OpNo].RegClass);
+
+ if (ImmRC.getSize() == 4)
+ printImmediate32(FloatToBits(Op.getFPImm()), O);
+ else if (ImmRC.getSize() == 8)
+ printImmediate64(DoubleToBits(Op.getFPImm()), O);
+ else
+ llvm_unreachable("Invalid register class size");
+ }
} else if (Op.isExpr()) {
const MCExpr *Exp = Op.getExpr();
Exp->print(O);
} else {
- assert(!"unknown operand type in printOperand");
+ llvm_unreachable("unknown operand type in printOperand");
}
}
+void AMDGPUInstPrinter::printOperandAndMods(const MCInst *MI, unsigned OpNo,
+ raw_ostream &O) {
+ unsigned InputModifiers = MI->getOperand(OpNo).getImm();
+ if (InputModifiers & SISrcMods::NEG)
+ O << '-';
+ if (InputModifiers & SISrcMods::ABS)
+ O << '|';
+ printOperand(MI, OpNo + 1, O);
+ if (InputModifiers & SISrcMods::ABS)
+ O << '|';
+}
+
void AMDGPUInstPrinter::printInterpSlot(const MCInst *MI, unsigned OpNum,
raw_ostream &O) {
unsigned Imm = MI->getOperand(OpNum).getImm();
} else if (Imm == 0) {
O << "P10";
} else {
- assert(!"Invalid interpolation parameter slot");
+ llvm_unreachable("Invalid interpolation parameter slot");
}
}
printIfSet(MI, OpNo, O, "_SAT");
}
-void AMDGPUInstPrinter::printLiteral(const MCInst *MI, unsigned OpNo,
+void AMDGPUInstPrinter::printClampSI(const MCInst *MI, unsigned OpNo,
raw_ostream &O) {
- union Literal {
- float f;
- int32_t i;
- } L;
+ if (MI->getOperand(OpNo).getImm())
+ O << " clamp";
+}
- L.i = MI->getOperand(OpNo).getImm();
- O << L.i << "(" << L.f << ")";
+void AMDGPUInstPrinter::printOModSI(const MCInst *MI, unsigned OpNo,
+ raw_ostream &O) {
+ int Imm = MI->getOperand(OpNo).getImm();
+ if (Imm == SIOutMods::MUL2)
+ O << " mul:2";
+ else if (Imm == SIOutMods::MUL4)
+ O << " mul:4";
+ else if (Imm == SIOutMods::DIV2)
+ O << " div:2";
+}
+
+void AMDGPUInstPrinter::printLiteral(const MCInst *MI, unsigned OpNo,
+ raw_ostream &O) {
+ int32_t Imm = MI->getOperand(OpNo).getImm();
+ O << Imm << '(' << BitsToFloat(Imm) << ')';
}
void AMDGPUInstPrinter::printLast(const MCInst *MI, unsigned OpNo,
sel -= 512;
int cb = sel >> 12;
sel &= 4095;
- O << cb << "[" << sel << "]";
+ O << cb << '[' << sel << ']';
} else if (sel >= 448) {
sel -= 448;
O << sel;
}
if (sel >= 0)
- O << "." << chans[chan];
+ O << '.' << chans[chan];
}
void AMDGPUInstPrinter::printBankSwizzle(const MCInst *MI, unsigned OpNo,
int BankSwizzle = MI->getOperand(OpNo).getImm();
switch (BankSwizzle) {
case 1:
- O << "BS:VEC_021";
+ O << "BS:VEC_021/SCL_122";
break;
case 2:
- O << "BS:VEC_120";
+ O << "BS:VEC_120/SCL_212";
break;
case 3:
- O << "BS:VEC_102";
+ O << "BS:VEC_102/SCL_221";
break;
case 4:
O << "BS:VEC_201";
unsigned Sel = MI->getOperand(OpNo).getImm();
switch (Sel) {
case 0:
- O << "X";
+ O << 'X';
break;
case 1:
- O << "Y";
+ O << 'Y';
break;
case 2:
- O << "Z";
+ O << 'Z';
break;
case 3:
- O << "W";
+ O << 'W';
break;
case 4:
- O << "0";
+ O << '0';
break;
case 5:
- O << "1";
+ O << '1';
break;
case 7:
- O << "_";
+ O << '_';
break;
default:
break;
unsigned CT = MI->getOperand(OpNo).getImm();
switch (CT) {
case 0:
- O << "U";
+ O << 'U';
break;
case 1:
- O << "N";
+ O << 'N';
break;
default:
break;
int KCacheMode = MI->getOperand(OpNo).getImm();
if (KCacheMode > 0) {
int KCacheBank = MI->getOperand(OpNo - 2).getImm();
- O << "CB" << KCacheBank <<":";
+ O << "CB" << KCacheBank << ':';
int KCacheAddr = MI->getOperand(OpNo + 2).getImm();
- int LineSize = (KCacheMode == 1)?16:32;
- O << KCacheAddr * 16 << "-" << KCacheAddr * 16 + LineSize;
+ int LineSize = (KCacheMode == 1) ? 16 : 32;
+ O << KCacheAddr * 16 << '-' << KCacheAddr * 16 + LineSize;
+ }
+}
+
+void AMDGPUInstPrinter::printSendMsg(const MCInst *MI, unsigned OpNo,
+ raw_ostream &O) {
+ unsigned SImm16 = MI->getOperand(OpNo).getImm();
+ unsigned Msg = SImm16 & 0xF;
+ if (Msg == 2 || Msg == 3) {
+ unsigned Op = (SImm16 >> 4) & 0xF;
+ if (Msg == 3)
+ O << "Gs_done(";
+ else
+ O << "Gs(";
+ if (Op == 0) {
+ O << "nop";
+ } else {
+ unsigned Stream = (SImm16 >> 8) & 0x3;
+ if (Op == 1)
+ O << "cut";
+ else if (Op == 2)
+ O << "emit";
+ else if (Op == 3)
+ O << "emit-cut";
+ O << " stream " << Stream;
+ }
+ O << "), [m0] ";
+ } else if (Msg == 1)
+ O << "interrupt ";
+ else if (Msg == 15)
+ O << "system ";
+ else
+ O << "unknown(" << Msg << ") ";
+}
+
+void AMDGPUInstPrinter::printWaitFlag(const MCInst *MI, unsigned OpNo,
+ raw_ostream &O) {
+ // Note: Mask values are taken from SIInsertWaits.cpp and not from ISA docs
+ // SIInsertWaits.cpp bits usage does not match ISA docs description but it
+ // works so it might be a misprint in docs.
+ unsigned SImm16 = MI->getOperand(OpNo).getImm();
+ unsigned Vmcnt = SImm16 & 0xF;
+ unsigned Expcnt = (SImm16 >> 4) & 0xF;
+ unsigned Lgkmcnt = (SImm16 >> 8) & 0xF;
+
+ bool NeedSpace = false;
+
+ if (Vmcnt != 0xF) {
+ O << "vmcnt(" << Vmcnt << ')';
+ NeedSpace = true;
+ }
+
+ if (Expcnt != 0x7) {
+ if (NeedSpace)
+ O << ' ';
+ O << "expcnt(" << Expcnt << ')';
+ NeedSpace = true;
+ }
+
+ if (Lgkmcnt != 0x7) {
+ if (NeedSpace)
+ O << ' ';
+ O << "lgkmcnt(" << Lgkmcnt << ')';
}
}
projects / oota-llvm.git / blobdiff