src/llvm/lib/Target/NVPTX/NVPTXAsmPrinter.cpp

   1 //===-- NVPTXAsmPrinter.cpp - NVPTX LLVM assembly writer ------------------===//
   2 //
   3 //                     The LLVM Compiler Infrastructure
   4 //
   5 // This file is distributed under the University of Illinois Open Source
   6 // License. See LICENSE.TXT for details.
   7 //
   8 //===----------------------------------------------------------------------===//
   9 //
  10 // This file contains a printer that converts from our internal representation
  11 // of machine-dependent LLVM code to NVPTX assembly language.
  12 //
  13 //===----------------------------------------------------------------------===//
  14
  15 #include "NVPTXAsmPrinter.h"
  16 #include "InstPrinter/NVPTXInstPrinter.h"
  17 #include "MCTargetDesc/NVPTXMCAsmInfo.h"
  18 #include "NVPTX.h"
  19 #include "NVPTXInstrInfo.h"
  20 #include "NVPTXMCExpr.h"
  21 #include "NVPTXMachineFunctionInfo.h"
  22 #include "NVPTXRegisterInfo.h"
  23 #include "NVPTXTargetMachine.h"
  24 #include "NVPTXUtilities.h"
  25 #include "cl_common_defines.h"
  26 #include "llvm/ADT/StringExtras.h"
  27 #include "llvm/Analysis/ConstantFolding.h"
  28 #include "llvm/CodeGen/Analysis.h"
  29 #include "llvm/CodeGen/MachineFrameInfo.h"
  30 #include "llvm/CodeGen/MachineModuleInfo.h"
  31 #include "llvm/CodeGen/MachineRegisterInfo.h"
  32 #include "llvm/IR/DebugInfo.h"
  33 #include "llvm/IR/DerivedTypes.h"
  34 #include "llvm/IR/Function.h"
  35 #include "llvm/IR/GlobalVariable.h"
  36 #include "llvm/IR/Mangler.h"
  37 #include "llvm/IR/Module.h"
  38 #include "llvm/IR/Operator.h"
  39 #include "llvm/MC/MCStreamer.h"
  40 #include "llvm/MC/MCSymbol.h"
  41 #include "llvm/Support/CommandLine.h"
  42 #include "llvm/Support/ErrorHandling.h"
  43 #include "llvm/Support/FormattedStream.h"
  44 #include "llvm/Support/Path.h"
  45 #include "llvm/Support/TargetRegistry.h"
  46 #include "llvm/Support/TimeValue.h"
  47 #include "llvm/Target/TargetLoweringObjectFile.h"
  48 #include <sstream>
  49 using namespace llvm;
  50
  51 #define DEPOTNAME "__local_depot"
  52
  53 static cl::opt<bool>
  54 EmitLineNumbers("nvptx-emit-line-numbers", cl::Hidden,
  55                 cl::desc("NVPTX Specific: Emit Line numbers even without -G"),
  56                 cl::init(true));
  57
  58 static cl::opt<bool>
  59 InterleaveSrc("nvptx-emit-src", cl::ZeroOrMore, cl::Hidden,
  60               cl::desc("NVPTX Specific: Emit source line in ptx file"),
  61               cl::init(false));
  62
  63 namespace {
  64 /// DiscoverDependentGlobals - Return a set of GlobalVariables on which \p V
  65 /// depends.
  66 void DiscoverDependentGlobals(const Value *V,
  67                               DenseSet<const GlobalVariable *> &Globals) {
  68   if (const GlobalVariable *GV = dyn_cast<GlobalVariable>(V))
  69     Globals.insert(GV);
  70   else {
  71     if (const User *U = dyn_cast<User>(V)) {
  72       for (unsigned i = 0, e = U->getNumOperands(); i != e; ++i) {
  73         DiscoverDependentGlobals(U->getOperand(i), Globals);
  74       }
  75     }
  76   }
  77 }
  78
  79 /// VisitGlobalVariableForEmission - Add \p GV to the list of GlobalVariable
  80 /// instances to be emitted, but only after any dependents have been added
  81 /// first.
  82 void VisitGlobalVariableForEmission(
  83     const GlobalVariable *GV, SmallVectorImpl<const GlobalVariable *> &Order,
  84     DenseSet<const GlobalVariable *> &Visited,
  85     DenseSet<const GlobalVariable *> &Visiting) {
  86   // Have we already visited this one?
  87   if (Visited.count(GV))
  88     return;
  89
  90   // Do we have a circular dependency?
  91   if (!Visiting.insert(GV).second)
  92     report_fatal_error("Circular dependency found in global variable set");
  93
  94   // Make sure we visit all dependents first
  95   DenseSet<const GlobalVariable *> Others;
  96   for (unsigned i = 0, e = GV->getNumOperands(); i != e; ++i)
  97     DiscoverDependentGlobals(GV->getOperand(i), Others);
  98
  99   for (DenseSet<const GlobalVariable *>::iterator I = Others.begin(),
 100                                                   E = Others.end();
 101        I != E; ++I)
 102     VisitGlobalVariableForEmission(*I, Order, Visited, Visiting);
 103
 104   // Now we can visit ourself
 105   Order.push_back(GV);
 106   Visited.insert(GV);
 107   Visiting.erase(GV);
 108 }
 109 }
 110
 111 void NVPTXAsmPrinter::emitLineNumberAsDotLoc(const MachineInstr &MI) {
 112   if (!EmitLineNumbers)
 113     return;
 114   if (ignoreLoc(MI))
 115     return;
 116
 117   DebugLoc curLoc = MI.getDebugLoc();
 118
 119   if (prevDebugLoc.isUnknown() && curLoc.isUnknown())
 120     return;
 121
 122   if (prevDebugLoc == curLoc)
 123     return;
 124
 125   prevDebugLoc = curLoc;
 126
 127   if (curLoc.isUnknown())
 128     return;
 129
 130   const MachineFunction *MF = MI.getParent()->getParent();
 131   //const TargetMachine &TM = MF->getTarget();
 132
 133   const LLVMContext &ctx = MF->getFunction()->getContext();
 134   DIScope Scope(curLoc.getScope(ctx));
 135
 136   assert((!Scope || Scope.isScope()) &&
 137     "Scope of a DebugLoc should be null or a DIScope.");
 138   if (!Scope)
 139      return;
 140
 141   StringRef fileName(Scope.getFilename());
 142   StringRef dirName(Scope.getDirectory());
 143   SmallString<128> FullPathName = dirName;
 144   if (!dirName.empty() && !sys::path::is_absolute(fileName)) {
 145     sys::path::append(FullPathName, fileName);
 146     fileName = FullPathName.str();
 147   }
 148
 149   if (filenameMap.find(fileName.str()) == filenameMap.end())
 150     return;
 151
 152   // Emit the line from the source file.
 153   if (InterleaveSrc)
 154     this->emitSrcInText(fileName.str(), curLoc.getLine());
 155
 156   std::stringstream temp;
 157   temp << "\t.loc " << filenameMap[fileName.str()] << " " << curLoc.getLine()
 158        << " " << curLoc.getCol();
 159   OutStreamer.EmitRawText(Twine(temp.str().c_str()));
 160 }
 161
 162 void NVPTXAsmPrinter::EmitInstruction(const MachineInstr *MI) {
 163   SmallString<128> Str;
 164   raw_svector_ostream OS(Str);
 165   if (nvptxSubtarget.getDrvInterface() == NVPTX::CUDA)
 166     emitLineNumberAsDotLoc(*MI);
 167
 168   MCInst Inst;
 169   lowerToMCInst(MI, Inst);
 170   EmitToStreamer(OutStreamer, Inst);
 171 }
 172
 173 // Handle symbol backtracking for targets that do not support image handles
 174 bool NVPTXAsmPrinter::lowerImageHandleOperand(const MachineInstr *MI,
 175                                            unsigned OpNo, MCOperand &MCOp) {
 176   const MachineOperand &MO = MI->getOperand(OpNo);
 177   const MCInstrDesc &MCID = MI->getDesc();
 178
 179   if (MCID.TSFlags & NVPTXII::IsTexFlag) {
 180     // This is a texture fetch, so operand 4 is a texref and operand 5 is
 181     // a samplerref
 182     if (OpNo == 4 && MO.isImm()) {
 183       lowerImageHandleSymbol(MO.getImm(), MCOp);
 184       return true;
 185     }
 186     if (OpNo == 5 && MO.isImm() && !(MCID.TSFlags & NVPTXII::IsTexModeUnifiedFlag)) {
 187       lowerImageHandleSymbol(MO.getImm(), MCOp);
 188       return true;
 189     }
 190
 191     return false;
 192   } else if (MCID.TSFlags & NVPTXII::IsSuldMask) {
 193     unsigned VecSize =
 194       1 << (((MCID.TSFlags & NVPTXII::IsSuldMask) >> NVPTXII::IsSuldShift) - 1);
 195
 196     // For a surface load of vector size N, the Nth operand will be the surfref
 197     if (OpNo == VecSize && MO.isImm()) {
 198       lowerImageHandleSymbol(MO.getImm(), MCOp);
 199       return true;
 200     }
 201
 202     return false;
 203   } else if (MCID.TSFlags & NVPTXII::IsSustFlag) {
 204     // This is a surface store, so operand 0 is a surfref
 205     if (OpNo == 0 && MO.isImm()) {
 206       lowerImageHandleSymbol(MO.getImm(), MCOp);
 207       return true;
 208     }
 209
 210     return false;
 211   } else if (MCID.TSFlags & NVPTXII::IsSurfTexQueryFlag) {
 212     // This is a query, so operand 1 is a surfref/texref
 213     if (OpNo == 1 && MO.isImm()) {
 214       lowerImageHandleSymbol(MO.getImm(), MCOp);
 215       return true;
 216     }
 217
 218     return false;
 219   }
 220
 221   return false;
 222 }
 223
 224 void NVPTXAsmPrinter::lowerImageHandleSymbol(unsigned Index, MCOperand &MCOp) {
 225   // Ewwww
 226   TargetMachine &TM = const_cast<TargetMachine&>(MF->getTarget());
 227   NVPTXTargetMachine &nvTM = static_cast<NVPTXTargetMachine&>(TM);
 228   const NVPTXMachineFunctionInfo *MFI = MF->getInfo<NVPTXMachineFunctionInfo>();
 229   const char *Sym = MFI->getImageHandleSymbol(Index);
 230   std::string *SymNamePtr =
 231     nvTM.getManagedStrPool()->getManagedString(Sym);
 232   MCOp = GetSymbolRef(OutContext.GetOrCreateSymbol(
 233     StringRef(SymNamePtr->c_str())));
 234 }
 235
 236 void NVPTXAsmPrinter::lowerToMCInst(const MachineInstr *MI, MCInst &OutMI) {
 237   OutMI.setOpcode(MI->getOpcode());
 238   const NVPTXSubtarget &ST = TM.getSubtarget<NVPTXSubtarget>();
 239
 240   // Special: Do not mangle symbol operand of CALL_PROTOTYPE
 241   if (MI->getOpcode() == NVPTX::CALL_PROTOTYPE) {
 242     const MachineOperand &MO = MI->getOperand(0);
 243     OutMI.addOperand(GetSymbolRef(
 244       OutContext.GetOrCreateSymbol(Twine(MO.getSymbolName()))));
 245     return;
 246   }
 247
 248   for (unsigned i = 0, e = MI->getNumOperands(); i != e; ++i) {
 249     const MachineOperand &MO = MI->getOperand(i);
 250
 251     MCOperand MCOp;
 252     if (!ST.hasImageHandles()) {
 253       if (lowerImageHandleOperand(MI, i, MCOp)) {
 254         OutMI.addOperand(MCOp);
 255         continue;
 256       }
 257     }
 258
 259     if (lowerOperand(MO, MCOp))
 260       OutMI.addOperand(MCOp);
 261   }
 262 }
 263
 264 bool NVPTXAsmPrinter::lowerOperand(const MachineOperand &MO,
 265                                    MCOperand &MCOp) {
 266   switch (MO.getType()) {
 267   default: llvm_unreachable("unknown operand type");
 268   case MachineOperand::MO_Register:
 269     MCOp = MCOperand::CreateReg(encodeVirtualRegister(MO.getReg()));
 270     break;
 271   case MachineOperand::MO_Immediate:
 272     MCOp = MCOperand::CreateImm(MO.getImm());
 273     break;
 274   case MachineOperand::MO_MachineBasicBlock:
 275     MCOp = MCOperand::CreateExpr(MCSymbolRefExpr::Create(
 276         MO.getMBB()->getSymbol(), OutContext));
 277     break;
 278   case MachineOperand::MO_ExternalSymbol:
 279     MCOp = GetSymbolRef(GetExternalSymbolSymbol(MO.getSymbolName()));
 280     break;
 281   case MachineOperand::MO_GlobalAddress:
 282     MCOp = GetSymbolRef(getSymbol(MO.getGlobal()));
 283     break;
 284   case MachineOperand::MO_FPImmediate: {
 285     const ConstantFP *Cnt = MO.getFPImm();
 286     APFloat Val = Cnt->getValueAPF();
 287
 288     switch (Cnt->getType()->getTypeID()) {
 289     default: report_fatal_error("Unsupported FP type"); break;
 290     case Type::FloatTyID:
 291       MCOp = MCOperand::CreateExpr(
 292         NVPTXFloatMCExpr::CreateConstantFPSingle(Val, OutContext));
 293       break;
 294     case Type::DoubleTyID:
 295       MCOp = MCOperand::CreateExpr(
 296         NVPTXFloatMCExpr::CreateConstantFPDouble(Val, OutContext));
 297       break;
 298     }
 299     break;
 300   }
 301   }
 302   return true;
 303 }
 304
 305 unsigned NVPTXAsmPrinter::encodeVirtualRegister(unsigned Reg) {
 306   if (TargetRegisterInfo::isVirtualRegister(Reg)) {
 307     const TargetRegisterClass *RC = MRI->getRegClass(Reg);
 308
 309     DenseMap<unsigned, unsigned> &RegMap = VRegMapping[RC];
 310     unsigned RegNum = RegMap[Reg];
 311
 312     // Encode the register class in the upper 4 bits
 313     // Must be kept in sync with NVPTXInstPrinter::printRegName
 314     unsigned Ret = 0;
 315     if (RC == &NVPTX::Int1RegsRegClass) {
 316       Ret = (1 << 28);
 317     } else if (RC == &NVPTX::Int16RegsRegClass) {
 318       Ret = (2 << 28);
 319     } else if (RC == &NVPTX::Int32RegsRegClass) {
 320       Ret = (3 << 28);
 321     } else if (RC == &NVPTX::Int64RegsRegClass) {
 322       Ret = (4 << 28);
 323     } else if (RC == &NVPTX::Float32RegsRegClass) {
 324       Ret = (5 << 28);
 325     } else if (RC == &NVPTX::Float64RegsRegClass) {
 326       Ret = (6 << 28);
 327     } else {
 328       report_fatal_error("Bad register class");
 329     }
 330
 331     // Insert the vreg number
 332     Ret |= (RegNum & 0x0FFFFFFF);
 333     return Ret;
 334   } else {
 335     // Some special-use registers are actually physical registers.
 336     // Encode this as the register class ID of 0 and the real register ID.
 337     return Reg & 0x0FFFFFFF;
 338   }
 339 }
 340
 341 MCOperand NVPTXAsmPrinter::GetSymbolRef(const MCSymbol *Symbol) {
 342   const MCExpr *Expr;
 343   Expr = MCSymbolRefExpr::Create(Symbol, MCSymbolRefExpr::VK_None,
 344                                  OutContext);
 345   return MCOperand::CreateExpr(Expr);
 346 }
 347
 348 void NVPTXAsmPrinter::printReturnValStr(const Function *F, raw_ostream &O) {
 349   const DataLayout *TD = TM.getSubtargetImpl()->getDataLayout();
 350   const TargetLowering *TLI = TM.getSubtargetImpl()->getTargetLowering();
 351
 352   Type *Ty = F->getReturnType();
 353
 354   bool isABI = (nvptxSubtarget.getSmVersion() >= 20);
 355
 356   if (Ty->getTypeID() == Type::VoidTyID)
 357     return;
 358
 359   O << " (";
 360
 361   if (isABI) {
 362     if (Ty->isFloatingPointTy() || Ty->isIntegerTy()) {
 363       unsigned size = 0;
 364       if (const IntegerType *ITy = dyn_cast<IntegerType>(Ty)) {
 365         size = ITy->getBitWidth();
 366         if (size < 32)
 367           size = 32;
 368       } else {
 369         assert(Ty->isFloatingPointTy() && "Floating point type expected here");
 370         size = Ty->getPrimitiveSizeInBits();
 371       }
 372
 373       O << ".param .b" << size << " func_retval0";
 374     } else if (isa<PointerType>(Ty)) {
 375       O << ".param .b" << TLI->getPointerTy().getSizeInBits()
 376         << " func_retval0";
 377     } else if ((Ty->getTypeID() == Type::StructTyID) || isa<VectorType>(Ty)) {
 378        unsigned totalsz = TD->getTypeAllocSize(Ty);
 379        unsigned retAlignment = 0;
 380        if (!llvm::getAlign(*F, 0, retAlignment))
 381          retAlignment = TD->getABITypeAlignment(Ty);
 382        O << ".param .align " << retAlignment << " .b8 func_retval0[" << totalsz
 383          << "]";
 384     } else
 385       llvm_unreachable("Unknown return type");
 386   } else {
 387     SmallVector<EVT, 16> vtparts;
 388     ComputeValueVTs(*TLI, Ty, vtparts);
 389     unsigned idx = 0;
 390     for (unsigned i = 0, e = vtparts.size(); i != e; ++i) {
 391       unsigned elems = 1;
 392       EVT elemtype = vtparts[i];
 393       if (vtparts[i].isVector()) {
 394         elems = vtparts[i].getVectorNumElements();
 395         elemtype = vtparts[i].getVectorElementType();
 396       }
 397
 398       for (unsigned j = 0, je = elems; j != je; ++j) {
 399         unsigned sz = elemtype.getSizeInBits();
 400         if (elemtype.isInteger() && (sz < 32))
 401           sz = 32;
 402         O << ".reg .b" << sz << " func_retval" << idx;
 403         if (j < je - 1)
 404           O << ", ";
 405         ++idx;
 406       }
 407       if (i < e - 1)
 408         O << ", ";
 409     }
 410   }
 411   O << ") ";
 412   return;
 413 }
 414
 415 void NVPTXAsmPrinter::printReturnValStr(const MachineFunction &MF,
 416                                         raw_ostream &O) {
 417   const Function *F = MF.getFunction();
 418   printReturnValStr(F, O);
 419 }
 420
 421 void NVPTXAsmPrinter::EmitFunctionEntryLabel() {
 422   SmallString<128> Str;
 423   raw_svector_ostream O(Str);
 424
 425   if (!GlobalsEmitted) {
 426     emitGlobals(*MF->getFunction()->getParent());
 427     GlobalsEmitted = true;
 428   }
 429
 430   // Set up
 431   MRI = &MF->getRegInfo();
 432   F = MF->getFunction();
 433   emitLinkageDirective(F, O);
 434   if (llvm::isKernelFunction(*F))
 435     O << ".entry ";
 436   else {
 437     O << ".func ";
 438     printReturnValStr(*MF, O);
 439   }
 440
 441   O << *CurrentFnSym;
 442
 443   emitFunctionParamList(*MF, O);
 444
 445   if (llvm::isKernelFunction(*F))
 446     emitKernelFunctionDirectives(*F, O);
 447
 448   OutStreamer.EmitRawText(O.str());
 449
 450   prevDebugLoc = DebugLoc();
 451 }
 452
 453 void NVPTXAsmPrinter::EmitFunctionBodyStart() {
 454   VRegMapping.clear();
 455   OutStreamer.EmitRawText(StringRef("{\n"));
 456   setAndEmitFunctionVirtualRegisters(*MF);
 457
 458   SmallString<128> Str;
 459   raw_svector_ostream O(Str);
 460   emitDemotedVars(MF->getFunction(), O);
 461   OutStreamer.EmitRawText(O.str());
 462 }
 463
 464 void NVPTXAsmPrinter::EmitFunctionBodyEnd() {
 465   OutStreamer.EmitRawText(StringRef("}\n"));
 466   VRegMapping.clear();
 467 }
 468
 469 void NVPTXAsmPrinter::emitImplicitDef(const MachineInstr *MI) const {
 470   unsigned RegNo = MI->getOperand(0).getReg();
 471   const TargetRegisterInfo *TRI = TM.getSubtargetImpl()->getRegisterInfo();
 472   if (TRI->isVirtualRegister(RegNo)) {
 473     OutStreamer.AddComment(Twine("implicit-def: ") +
 474                            getVirtualRegisterName(RegNo));
 475   } else {
 476     OutStreamer.AddComment(
 477         Twine("implicit-def: ") +
 478         TM.getSubtargetImpl()->getRegisterInfo()->getName(RegNo));
 479   }
 480   OutStreamer.AddBlankLine();
 481 }
 482
 483 void NVPTXAsmPrinter::emitKernelFunctionDirectives(const Function &F,
 484                                                    raw_ostream &O) const {
 485   // If the NVVM IR has some of reqntid* specified, then output
 486   // the reqntid directive, and set the unspecified ones to 1.
 487   // If none of reqntid* is specified, don't output reqntid directive.
 488   unsigned reqntidx, reqntidy, reqntidz;
 489   bool specified = false;
 490   if (llvm::getReqNTIDx(F, reqntidx) == false)
 491     reqntidx = 1;
 492   else
 493     specified = true;
 494   if (llvm::getReqNTIDy(F, reqntidy) == false)
 495     reqntidy = 1;
 496   else
 497     specified = true;
 498   if (llvm::getReqNTIDz(F, reqntidz) == false)
 499     reqntidz = 1;
 500   else
 501     specified = true;
 502
 503   if (specified)
 504     O << ".reqntid " << reqntidx << ", " << reqntidy << ", " << reqntidz
 505       << "\n";
 506
 507   // If the NVVM IR has some of maxntid* specified, then output
 508   // the maxntid directive, and set the unspecified ones to 1.
 509   // If none of maxntid* is specified, don't output maxntid directive.
 510   unsigned maxntidx, maxntidy, maxntidz;
 511   specified = false;
 512   if (llvm::getMaxNTIDx(F, maxntidx) == false)
 513     maxntidx = 1;
 514   else
 515     specified = true;
 516   if (llvm::getMaxNTIDy(F, maxntidy) == false)
 517     maxntidy = 1;
 518   else
 519     specified = true;
 520   if (llvm::getMaxNTIDz(F, maxntidz) == false)
 521     maxntidz = 1;
 522   else
 523     specified = true;
 524
 525   if (specified)
 526     O << ".maxntid " << maxntidx << ", " << maxntidy << ", " << maxntidz
 527       << "\n";
 528
 529   unsigned mincta;
 530   if (llvm::getMinCTASm(F, mincta))
 531     O << ".minnctapersm " << mincta << "\n";
 532 }
 533
 534 std::string
 535 NVPTXAsmPrinter::getVirtualRegisterName(unsigned Reg) const {
 536   const TargetRegisterClass *RC = MRI->getRegClass(Reg);
 537
 538   std::string Name;
 539   raw_string_ostream NameStr(Name);
 540
 541   VRegRCMap::const_iterator I = VRegMapping.find(RC);
 542   assert(I != VRegMapping.end() && "Bad register class");
 543   const DenseMap<unsigned, unsigned> &RegMap = I->second;
 544
 545   VRegMap::const_iterator VI = RegMap.find(Reg);
 546   assert(VI != RegMap.end() && "Bad virtual register");
 547   unsigned MappedVR = VI->second;
 548
 549   NameStr << getNVPTXRegClassStr(RC) << MappedVR;
 550
 551   NameStr.flush();
 552   return Name;
 553 }
 554
 555 void NVPTXAsmPrinter::emitVirtualRegister(unsigned int vr,
 556                                           raw_ostream &O) {
 557   O << getVirtualRegisterName(vr);
 558 }
 559
 560 void NVPTXAsmPrinter::printVecModifiedImmediate(
 561     const MachineOperand &MO, const char *Modifier, raw_ostream &O) {
 562   static const char vecelem[] = { '0', '1', '2', '3', '0', '1', '2', '3' };
 563   int Imm = (int) MO.getImm();
 564   if (0 == strcmp(Modifier, "vecelem"))
 565     O << "_" << vecelem[Imm];
 566   else if (0 == strcmp(Modifier, "vecv4comm1")) {
 567     if ((Imm < 0) || (Imm > 3))
 568       O << "//";
 569   } else if (0 == strcmp(Modifier, "vecv4comm2")) {
 570     if ((Imm < 4) || (Imm > 7))
 571       O << "//";
 572   } else if (0 == strcmp(Modifier, "vecv4pos")) {
 573     if (Imm < 0)
 574       Imm = 0;
 575     O << "_" << vecelem[Imm % 4];
 576   } else if (0 == strcmp(Modifier, "vecv2comm1")) {
 577     if ((Imm < 0) || (Imm > 1))
 578       O << "//";
 579   } else if (0 == strcmp(Modifier, "vecv2comm2")) {
 580     if ((Imm < 2) || (Imm > 3))
 581       O << "//";
 582   } else if (0 == strcmp(Modifier, "vecv2pos")) {
 583     if (Imm < 0)
 584       Imm = 0;
 585     O << "_" << vecelem[Imm % 2];
 586   } else
 587     llvm_unreachable("Unknown Modifier on immediate operand");
 588 }
 589
 590
 591
 592 void NVPTXAsmPrinter::emitDeclaration(const Function *F, raw_ostream &O) {
 593
 594   emitLinkageDirective(F, O);
 595   if (llvm::isKernelFunction(*F))
 596     O << ".entry ";
 597   else
 598     O << ".func ";
 599   printReturnValStr(F, O);
 600   O << *getSymbol(F) << "\n";
 601   emitFunctionParamList(F, O);
 602   O << ";\n";
 603 }
 604
 605 static bool usedInGlobalVarDef(const Constant *C) {
 606   if (!C)
 607     return false;
 608
 609   if (const GlobalVariable *GV = dyn_cast<GlobalVariable>(C)) {
 610     if (GV->getName().str() == "llvm.used")
 611       return false;
 612     return true;
 613   }
 614
 615   for (const User *U : C->users())
 616     if (const Constant *C = dyn_cast<Constant>(U))
 617       if (usedInGlobalVarDef(C))
 618         return true;
 619
 620   return false;
 621 }
 622
 623 static bool usedInOneFunc(const User *U, Function const *&oneFunc) {
 624   if (const GlobalVariable *othergv = dyn_cast<GlobalVariable>(U)) {
 625     if (othergv->getName().str() == "llvm.used")
 626       return true;
 627   }
 628
 629   if (const Instruction *instr = dyn_cast<Instruction>(U)) {
 630     if (instr->getParent() && instr->getParent()->getParent()) {
 631       const Function *curFunc = instr->getParent()->getParent();
 632       if (oneFunc && (curFunc != oneFunc))
 633         return false;
 634       oneFunc = curFunc;
 635       return true;
 636     } else
 637       return false;
 638   }
 639
 640   for (const User *UU : U->users())
 641     if (usedInOneFunc(UU, oneFunc) == false)
 642       return false;
 643
 644   return true;
 645 }
 646
 647 /* Find out if a global variable can be demoted to local scope.
 648  * Currently, this is valid for CUDA shared variables, which have local
 649  * scope and global lifetime. So the conditions to check are :
 650  * 1. Is the global variable in shared address space?
 651  * 2. Does it have internal linkage?
 652  * 3. Is the global variable referenced only in one function?
 653  */
 654 static bool canDemoteGlobalVar(const GlobalVariable *gv, Function const *&f) {
 655   if (gv->hasInternalLinkage() == false)
 656     return false;
 657   const PointerType *Pty = gv->getType();
 658   if (Pty->getAddressSpace() != llvm::ADDRESS_SPACE_SHARED)
 659     return false;
 660
 661   const Function *oneFunc = nullptr;
 662
 663   bool flag = usedInOneFunc(gv, oneFunc);
 664   if (flag == false)
 665     return false;
 666   if (!oneFunc)
 667     return false;
 668   f = oneFunc;
 669   return true;
 670 }
 671
 672 static bool useFuncSeen(const Constant *C,
 673                         llvm::DenseMap<const Function *, bool> &seenMap) {
 674   for (const User *U : C->users()) {
 675     if (const Constant *cu = dyn_cast<Constant>(U)) {
 676       if (useFuncSeen(cu, seenMap))
 677         return true;
 678     } else if (const Instruction *I = dyn_cast<Instruction>(U)) {
 679       const BasicBlock *bb = I->getParent();
 680       if (!bb)
 681         continue;
 682       const Function *caller = bb->getParent();
 683       if (!caller)
 684         continue;
 685       if (seenMap.find(caller) != seenMap.end())
 686         return true;
 687     }
 688   }
 689   return false;
 690 }
 691
 692 void NVPTXAsmPrinter::emitDeclarations(const Module &M, raw_ostream &O) {
 693   llvm::DenseMap<const Function *, bool> seenMap;
 694   for (Module::const_iterator FI = M.begin(), FE = M.end(); FI != FE; ++FI) {
 695     const Function *F = FI;
 696
 697     if (F->isDeclaration()) {
 698       if (F->use_empty())
 699         continue;
 700       if (F->getIntrinsicID())
 701         continue;
 702       emitDeclaration(F, O);
 703       continue;
 704     }
 705     for (const User *U : F->users()) {
 706       if (const Constant *C = dyn_cast<Constant>(U)) {
 707         if (usedInGlobalVarDef(C)) {
 708           // The use is in the initialization of a global variable
 709           // that is a function pointer, so print a declaration
 710           // for the original function
 711           emitDeclaration(F, O);
 712           break;
 713         }
 714         // Emit a declaration of this function if the function that
 715         // uses this constant expr has already been seen.
 716         if (useFuncSeen(C, seenMap)) {
 717           emitDeclaration(F, O);
 718           break;
 719         }
 720       }
 721
 722       if (!isa<Instruction>(U))
 723         continue;
 724       const Instruction *instr = cast<Instruction>(U);
 725       const BasicBlock *bb = instr->getParent();
 726       if (!bb)
 727         continue;
 728       const Function *caller = bb->getParent();
 729       if (!caller)
 730         continue;
 731
 732       // If a caller has already been seen, then the caller is
 733       // appearing in the module before the callee. so print out
 734       // a declaration for the callee.
 735       if (seenMap.find(caller) != seenMap.end()) {
 736         emitDeclaration(F, O);
 737         break;
 738       }
 739     }
 740     seenMap[F] = true;
 741   }
 742 }
 743
 744 void NVPTXAsmPrinter::recordAndEmitFilenames(Module &M) {
 745   DebugInfoFinder DbgFinder;
 746   DbgFinder.processModule(M);
 747
 748   unsigned i = 1;
 749   for (DICompileUnit DIUnit : DbgFinder.compile_units()) {
 750     StringRef Filename(DIUnit.getFilename());
 751     StringRef Dirname(DIUnit.getDirectory());
 752     SmallString<128> FullPathName = Dirname;
 753     if (!Dirname.empty() && !sys::path::is_absolute(Filename)) {
 754       sys::path::append(FullPathName, Filename);
 755       Filename = FullPathName.str();
 756     }
 757     if (filenameMap.find(Filename.str()) != filenameMap.end())
 758       continue;
 759     filenameMap[Filename.str()] = i;
 760     OutStreamer.EmitDwarfFileDirective(i, "", Filename.str());
 761     ++i;
 762   }
 763
 764   for (DISubprogram SP : DbgFinder.subprograms()) {
 765     StringRef Filename(SP.getFilename());
 766     StringRef Dirname(SP.getDirectory());
 767     SmallString<128> FullPathName = Dirname;
 768     if (!Dirname.empty() && !sys::path::is_absolute(Filename)) {
 769       sys::path::append(FullPathName, Filename);
 770       Filename = FullPathName.str();
 771     }
 772     if (filenameMap.find(Filename.str()) != filenameMap.end())
 773       continue;
 774     filenameMap[Filename.str()] = i;
 775     ++i;
 776   }
 777 }
 778
 779 bool NVPTXAsmPrinter::doInitialization(Module &M) {
 780
 781   SmallString<128> Str1;
 782   raw_svector_ostream OS1(Str1);
 783
 784   MMI = getAnalysisIfAvailable<MachineModuleInfo>();
 785   MMI->AnalyzeModule(M);
 786
 787   // We need to call the parent's one explicitly.
 788   //bool Result = AsmPrinter::doInitialization(M);
 789
 790   // Initialize TargetLoweringObjectFile.
 791   const_cast<TargetLoweringObjectFile &>(getObjFileLowering())
 792       .Initialize(OutContext, TM);
 793
 794   Mang = new Mangler(TM.getSubtargetImpl()->getDataLayout());
 795
 796   // Emit header before any dwarf directives are emitted below.
 797   emitHeader(M, OS1);
 798   OutStreamer.EmitRawText(OS1.str());
 799
 800   // Already commented out
 801   //bool Result = AsmPrinter::doInitialization(M);
 802
 803   // Emit module-level inline asm if it exists.
 804   if (!M.getModuleInlineAsm().empty()) {
 805     OutStreamer.AddComment("Start of file scope inline assembly");
 806     OutStreamer.AddBlankLine();
 807     OutStreamer.EmitRawText(StringRef(M.getModuleInlineAsm()));
 808     OutStreamer.AddBlankLine();
 809     OutStreamer.AddComment("End of file scope inline assembly");
 810     OutStreamer.AddBlankLine();
 811   }
 812
 813   if (nvptxSubtarget.getDrvInterface() == NVPTX::CUDA)
 814     recordAndEmitFilenames(M);
 815
 816   GlobalsEmitted = false;
 817
 818   return false; // success
 819 }
 820
 821 void NVPTXAsmPrinter::emitGlobals(const Module &M) {
 822   SmallString<128> Str2;
 823   raw_svector_ostream OS2(Str2);
 824
 825   emitDeclarations(M, OS2);
 826
 827   // As ptxas does not support forward references of globals, we need to first
 828   // sort the list of module-level globals in def-use order. We visit each
 829   // global variable in order, and ensure that we emit it *after* its dependent
 830   // globals. We use a little extra memory maintaining both a set and a list to
 831   // have fast searches while maintaining a strict ordering.
 832   SmallVector<const GlobalVariable *, 8> Globals;
 833   DenseSet<const GlobalVariable *> GVVisited;
 834   DenseSet<const GlobalVariable *> GVVisiting;
 835
 836   // Visit each global variable, in order
 837   for (Module::const_global_iterator I = M.global_begin(), E = M.global_end();
 838        I != E; ++I)
 839     VisitGlobalVariableForEmission(I, Globals, GVVisited, GVVisiting);
 840
 841   assert(GVVisited.size() == M.getGlobalList().size() &&
 842          "Missed a global variable");
 843   assert(GVVisiting.size() == 0 && "Did not fully process a global variable");
 844
 845   // Print out module-level global variables in proper order
 846   for (unsigned i = 0, e = Globals.size(); i != e; ++i)
 847     printModuleLevelGV(Globals[i], OS2);
 848
 849   OS2 << '\n';
 850
 851   OutStreamer.EmitRawText(OS2.str());
 852 }
 853
 854 void NVPTXAsmPrinter::emitHeader(Module &M, raw_ostream &O) {
 855   O << "//\n";
 856   O << "// Generated by LLVM NVPTX Back-End\n";
 857   O << "//\n";
 858   O << "\n";
 859
 860   unsigned PTXVersion = nvptxSubtarget.getPTXVersion();
 861   O << ".version " << (PTXVersion / 10) << "." << (PTXVersion % 10) << "\n";
 862
 863   O << ".target ";
 864   O << nvptxSubtarget.getTargetName();
 865
 866   if (nvptxSubtarget.getDrvInterface() == NVPTX::NVCL)
 867     O << ", texmode_independent";
 868   if (nvptxSubtarget.getDrvInterface() == NVPTX::CUDA) {
 869     if (!nvptxSubtarget.hasDouble())
 870       O << ", map_f64_to_f32";
 871   }
 872
 873   if (MAI->doesSupportDebugInformation())
 874     O << ", debug";
 875
 876   O << "\n";
 877
 878   O << ".address_size ";
 879   if (nvptxSubtarget.is64Bit())
 880     O << "64";
 881   else
 882     O << "32";
 883   O << "\n";
 884
 885   O << "\n";
 886 }
 887
 888 bool NVPTXAsmPrinter::doFinalization(Module &M) {
 889
 890   // If we did not emit any functions, then the global declarations have not
 891   // yet been emitted.
 892   if (!GlobalsEmitted) {
 893     emitGlobals(M);
 894     GlobalsEmitted = true;
 895   }
 896
 897   // XXX Temproarily remove global variables so that doFinalization() will not
 898   // emit them again (global variables are emitted at beginning).
 899
 900   Module::GlobalListType &global_list = M.getGlobalList();
 901   int i, n = global_list.size();
 902   GlobalVariable **gv_array = new GlobalVariable *[n];
 903
 904   // first, back-up GlobalVariable in gv_array
 905   i = 0;
 906   for (Module::global_iterator I = global_list.begin(), E = global_list.end();
 907        I != E; ++I)
 908     gv_array[i++] = &*I;
 909
 910   // second, empty global_list
 911   while (!global_list.empty())
 912     global_list.remove(global_list.begin());
 913
 914   // call doFinalization
 915   bool ret = AsmPrinter::doFinalization(M);
 916
 917   // now we restore global variables
 918   for (i = 0; i < n; i++)
 919     global_list.insert(global_list.end(), gv_array[i]);
 920
 921   clearAnnotationCache(&M);
 922
 923   delete[] gv_array;
 924   return ret;
 925
 926   //bool Result = AsmPrinter::doFinalization(M);
 927   // Instead of calling the parents doFinalization, we may
 928   // clone parents doFinalization and customize here.
 929   // Currently, we if NVISA out the EmitGlobals() in
 930   // parent's doFinalization, which is too intrusive.
 931   //
 932   // Same for the doInitialization.
 933   //return Result;
 934 }
 935
 936 // This function emits appropriate linkage directives for
 937 // functions and global variables.
 938 //
 939 // extern function declaration            -> .extern
 940 // extern function definition             -> .visible
 941 // external global variable with init     -> .visible
 942 // external without init                  -> .extern
 943 // appending                              -> not allowed, assert.
 944 // for any linkage other than
 945 // internal, private, linker_private,
 946 // linker_private_weak, linker_private_weak_def_auto,
 947 // we emit                                -> .weak.
 948
 949 void NVPTXAsmPrinter::emitLinkageDirective(const GlobalValue *V,
 950                                            raw_ostream &O) {
 951   if (nvptxSubtarget.getDrvInterface() == NVPTX::CUDA) {
 952     if (V->hasExternalLinkage()) {
 953       if (isa<GlobalVariable>(V)) {
 954         const GlobalVariable *GVar = cast<GlobalVariable>(V);
 955         if (GVar) {
 956           if (GVar->hasInitializer())
 957             O << ".visible ";
 958           else
 959             O << ".extern ";
 960         }
 961       } else if (V->isDeclaration())
 962         O << ".extern ";
 963       else
 964         O << ".visible ";
 965     } else if (V->hasAppendingLinkage()) {
 966       std::string msg;
 967       msg.append("Error: ");
 968       msg.append("Symbol ");
 969       if (V->hasName())
 970         msg.append(V->getName().str());
 971       msg.append("has unsupported appending linkage type");
 972       llvm_unreachable(msg.c_str());
 973     } else if (!V->hasInternalLinkage() &&
 974                !V->hasPrivateLinkage()) {
 975       O << ".weak ";
 976     }
 977   }
 978 }
 979
 980 void NVPTXAsmPrinter::printModuleLevelGV(const GlobalVariable *GVar,
 981                                          raw_ostream &O,
 982                                          bool processDemoted) {
 983
 984   // Skip meta data
 985   if (GVar->hasSection()) {
 986     if (GVar->getSection() == StringRef("llvm.metadata"))
 987       return;
 988   }
 989
 990   // Skip LLVM intrinsic global variables
 991   if (GVar->getName().startswith("llvm.") ||
 992       GVar->getName().startswith("nvvm."))
 993     return;
 994
 995   const DataLayout *TD = TM.getSubtargetImpl()->getDataLayout();
 996
 997   // GlobalVariables are always constant pointers themselves.
 998   const PointerType *PTy = GVar->getType();
 999   Type *ETy = PTy->getElementType();
1000
1001   if (GVar->hasExternalLinkage()) {
1002     if (GVar->hasInitializer())
1003       O << ".visible ";
1004     else
1005       O << ".extern ";
1006   } else if (GVar->hasLinkOnceLinkage() || GVar->hasWeakLinkage() ||
1007              GVar->hasAvailableExternallyLinkage() ||
1008              GVar->hasCommonLinkage()) {
1009     O << ".weak ";
1010   }
1011
1012   if (llvm::isTexture(*GVar)) {
1013     O << ".global .texref " << llvm::getTextureName(*GVar) << ";\n";
1014     return;
1015   }
1016
1017   if (llvm::isSurface(*GVar)) {
1018     O << ".global .surfref " << llvm::getSurfaceName(*GVar) << ";\n";
1019     return;
1020   }
1021
1022   if (GVar->isDeclaration()) {
1023     // (extern) declarations, no definition or initializer
1024     // Currently the only known declaration is for an automatic __local
1025     // (.shared) promoted to global.
1026     emitPTXGlobalVariable(GVar, O);
1027     O << ";\n";
1028     return;
1029   }
1030
1031   if (llvm::isSampler(*GVar)) {
1032     O << ".global .samplerref " << llvm::getSamplerName(*GVar);
1033
1034     const Constant *Initializer = nullptr;
1035     if (GVar->hasInitializer())
1036       Initializer = GVar->getInitializer();
1037     const ConstantInt *CI = nullptr;
1038     if (Initializer)
1039       CI = dyn_cast<ConstantInt>(Initializer);
1040     if (CI) {
1041       unsigned sample = CI->getZExtValue();
1042
1043       O << " = { ";
1044
1045       for (int i = 0,
1046                addr = ((sample & __CLK_ADDRESS_MASK) >> __CLK_ADDRESS_BASE);
1047            i < 3; i++) {
1048         O << "addr_mode_" << i << " = ";
1049         switch (addr) {
1050         case 0:
1051           O << "wrap";
1052           break;
1053         case 1:
1054           O << "clamp_to_border";
1055           break;
1056         case 2:
1057           O << "clamp_to_edge";
1058           break;
1059         case 3:
1060           O << "wrap";
1061           break;
1062         case 4:
1063           O << "mirror";
1064           break;
1065         }
1066         O << ", ";
1067       }
1068       O << "filter_mode = ";
1069       switch ((sample & __CLK_FILTER_MASK) >> __CLK_FILTER_BASE) {
1070       case 0:
1071         O << "nearest";
1072         break;
1073       case 1:
1074         O << "linear";
1075         break;
1076       case 2:
1077         llvm_unreachable("Anisotropic filtering is not supported");
1078       default:
1079         O << "nearest";
1080         break;
1081       }
1082       if (!((sample & __CLK_NORMALIZED_MASK) >> __CLK_NORMALIZED_BASE)) {
1083         O << ", force_unnormalized_coords = 1";
1084       }
1085       O << " }";
1086     }
1087
1088     O << ";\n";
1089     return;
1090   }
1091
1092   if (GVar->hasPrivateLinkage()) {
1093
1094     if (!strncmp(GVar->getName().data(), "unrollpragma", 12))
1095       return;
1096
1097     // FIXME - need better way (e.g. Metadata) to avoid generating this global
1098     if (!strncmp(GVar->getName().data(), "filename", 8))
1099       return;
1100     if (GVar->use_empty())
1101       return;
1102   }
1103
1104   const Function *demotedFunc = nullptr;
1105   if (!processDemoted && canDemoteGlobalVar(GVar, demotedFunc)) {
1106     O << "// " << GVar->getName().str() << " has been demoted\n";
1107     if (localDecls.find(demotedFunc) != localDecls.end())
1108       localDecls[demotedFunc].push_back(GVar);
1109     else {
1110       std::vector<const GlobalVariable *> temp;
1111       temp.push_back(GVar);
1112       localDecls[demotedFunc] = temp;
1113     }
1114     return;
1115   }
1116
1117   O << ".";
1118   emitPTXAddressSpace(PTy->getAddressSpace(), O);
1119
1120   if (isManaged(*GVar)) {
1121     O << " .attribute(.managed)";
1122   }
1123
1124   if (GVar->getAlignment() == 0)
1125     O << " .align " << (int) TD->getPrefTypeAlignment(ETy);
1126   else
1127     O << " .align " << GVar->getAlignment();
1128
1129   if (ETy->isFloatingPointTy() || ETy->isIntegerTy() || ETy->isPointerTy()) {
1130     O << " .";
1131     // Special case: ABI requires that we use .u8 for predicates
1132     if (ETy->isIntegerTy(1))
1133       O << "u8";
1134     else
1135       O << getPTXFundamentalTypeStr(ETy, false);
1136     O << " ";
1137     O << *getSymbol(GVar);
1138
1139     // Ptx allows variable initilization only for constant and global state
1140     // spaces.
1141     if (GVar->hasInitializer()) {
1142       if ((PTy->getAddressSpace() == llvm::ADDRESS_SPACE_GLOBAL) ||
1143           (PTy->getAddressSpace() == llvm::ADDRESS_SPACE_CONST)) {
1144         const Constant *Initializer = GVar->getInitializer();
1145         // 'undef' is treated as there is no value spefied.
1146         if (!Initializer->isNullValue() && !isa<UndefValue>(Initializer)) {
1147           O << " = ";
1148           printScalarConstant(Initializer, O);
1149         }
1150       } else {
1151         // The frontend adds zero-initializer to variables that don't have an
1152         // initial value, so skip warning for this case.
1153         if (!GVar->getInitializer()->isNullValue()) {
1154           std::string warnMsg = "initial value of '" + GVar->getName().str() +
1155               "' is not allowed in addrspace(" +
1156               llvm::utostr_32(PTy->getAddressSpace()) + ")";
1157           report_fatal_error(warnMsg.c_str());
1158         }
1159       }
1160     }
1161   } else {
1162     unsigned int ElementSize = 0;
1163
1164     // Although PTX has direct support for struct type and array type and
1165     // LLVM IR is very similar to PTX, the LLVM CodeGen does not support for
1166     // targets that support these high level field accesses. Structs, arrays
1167     // and vectors are lowered into arrays of bytes.
1168     switch (ETy->getTypeID()) {
1169     case Type::StructTyID:
1170     case Type::ArrayTyID:
1171     case Type::VectorTyID:
1172       ElementSize = TD->getTypeStoreSize(ETy);
1173       // Ptx allows variable initilization only for constant and
1174       // global state spaces.
1175       if (((PTy->getAddressSpace() == llvm::ADDRESS_SPACE_GLOBAL) ||
1176            (PTy->getAddressSpace() == llvm::ADDRESS_SPACE_CONST)) &&
1177           GVar->hasInitializer()) {
1178         const Constant *Initializer = GVar->getInitializer();
1179         if (!isa<UndefValue>(Initializer) && !Initializer->isNullValue()) {
1180           AggBuffer aggBuffer(ElementSize, O, *this);
1181           bufferAggregateConstant(Initializer, &aggBuffer);
1182           if (aggBuffer.numSymbols) {
1183             if (nvptxSubtarget.is64Bit()) {
1184               O << " .u64 " << *getSymbol(GVar) << "[";
1185               O << ElementSize / 8;
1186             } else {
1187               O << " .u32 " << *getSymbol(GVar) << "[";
1188               O << ElementSize / 4;
1189             }
1190             O << "]";
1191           } else {
1192             O << " .b8 " << *getSymbol(GVar) << "[";
1193             O << ElementSize;
1194             O << "]";
1195           }
1196           O << " = {";
1197           aggBuffer.print();
1198           O << "}";
1199         } else {
1200           O << " .b8 " << *getSymbol(GVar);
1201           if (ElementSize) {
1202             O << "[";
1203             O << ElementSize;
1204             O << "]";
1205           }
1206         }
1207       } else {
1208         O << " .b8 " << *getSymbol(GVar);
1209         if (ElementSize) {
1210           O << "[";
1211           O << ElementSize;
1212           O << "]";
1213         }
1214       }
1215       break;
1216     default:
1217       llvm_unreachable("type not supported yet");
1218     }
1219
1220   }
1221   O << ";\n";
1222 }
1223
1224 void NVPTXAsmPrinter::emitDemotedVars(const Function *f, raw_ostream &O) {
1225   if (localDecls.find(f) == localDecls.end())
1226     return;
1227
1228   std::vector<const GlobalVariable *> &gvars = localDecls[f];
1229
1230   for (unsigned i = 0, e = gvars.size(); i != e; ++i) {
1231     O << "\t// demoted variable\n\t";
1232     printModuleLevelGV(gvars[i], O, true);
1233   }
1234 }
1235
1236 void NVPTXAsmPrinter::emitPTXAddressSpace(unsigned int AddressSpace,
1237                                           raw_ostream &O) const {
1238   switch (AddressSpace) {
1239   case llvm::ADDRESS_SPACE_LOCAL:
1240     O << "local";
1241     break;
1242   case llvm::ADDRESS_SPACE_GLOBAL:
1243     O << "global";
1244     break;
1245   case llvm::ADDRESS_SPACE_CONST:
1246     O << "const";
1247     break;
1248   case llvm::ADDRESS_SPACE_SHARED:
1249     O << "shared";
1250     break;
1251   default:
1252     report_fatal_error("Bad address space found while emitting PTX");
1253     break;
1254   }
1255 }
1256
1257 std::string
1258 NVPTXAsmPrinter::getPTXFundamentalTypeStr(const Type *Ty, bool useB4PTR) const {
1259   switch (Ty->getTypeID()) {
1260   default:
1261     llvm_unreachable("unexpected type");
1262     break;
1263   case Type::IntegerTyID: {
1264     unsigned NumBits = cast<IntegerType>(Ty)->getBitWidth();
1265     if (NumBits == 1)
1266       return "pred";
1267     else if (NumBits <= 64) {
1268       std::string name = "u";
1269       return name + utostr(NumBits);
1270     } else {
1271       llvm_unreachable("Integer too large");
1272       break;
1273     }
1274     break;
1275   }
1276   case Type::FloatTyID:
1277     return "f32";
1278   case Type::DoubleTyID:
1279     return "f64";
1280   case Type::PointerTyID:
1281     if (nvptxSubtarget.is64Bit())
1282       if (useB4PTR)
1283         return "b64";
1284       else
1285         return "u64";
1286     else if (useB4PTR)
1287       return "b32";
1288     else
1289       return "u32";
1290   }
1291   llvm_unreachable("unexpected type");
1292   return nullptr;
1293 }
1294
1295 void NVPTXAsmPrinter::emitPTXGlobalVariable(const GlobalVariable *GVar,
1296                                             raw_ostream &O) {
1297
1298   const DataLayout *TD = TM.getSubtargetImpl()->getDataLayout();
1299
1300   // GlobalVariables are always constant pointers themselves.
1301   const PointerType *PTy = GVar->getType();
1302   Type *ETy = PTy->getElementType();
1303
1304   O << ".";
1305   emitPTXAddressSpace(PTy->getAddressSpace(), O);
1306   if (GVar->getAlignment() == 0)
1307     O << " .align " << (int) TD->getPrefTypeAlignment(ETy);
1308   else
1309     O << " .align " << GVar->getAlignment();
1310
1311   if (ETy->isFloatingPointTy() || ETy->isIntegerTy() || ETy->isPointerTy()) {
1312     O << " .";
1313     O << getPTXFundamentalTypeStr(ETy);
1314     O << " ";
1315     O << *getSymbol(GVar);
1316     return;
1317   }
1318
1319   int64_t ElementSize = 0;
1320
1321   // Although PTX has direct support for struct type and array type and LLVM IR
1322   // is very similar to PTX, the LLVM CodeGen does not support for targets that
1323   // support these high level field accesses. Structs and arrays are lowered
1324   // into arrays of bytes.
1325   switch (ETy->getTypeID()) {
1326   case Type::StructTyID:
1327   case Type::ArrayTyID:
1328   case Type::VectorTyID:
1329     ElementSize = TD->getTypeStoreSize(ETy);
1330     O << " .b8 " << *getSymbol(GVar) << "[";
1331     if (ElementSize) {
1332       O << itostr(ElementSize);
1333     }
1334     O << "]";
1335     break;
1336   default:
1337     llvm_unreachable("type not supported yet");
1338   }
1339   return;
1340 }
1341
1342 static unsigned int getOpenCLAlignment(const DataLayout *TD, Type *Ty) {
1343   if (Ty->isSingleValueType())
1344     return TD->getPrefTypeAlignment(Ty);
1345
1346   const ArrayType *ATy = dyn_cast<ArrayType>(Ty);
1347   if (ATy)
1348     return getOpenCLAlignment(TD, ATy->getElementType());
1349
1350   const StructType *STy = dyn_cast<StructType>(Ty);
1351   if (STy) {
1352     unsigned int alignStruct = 1;
1353     // Go through each element of the struct and find the
1354     // largest alignment.
1355     for (unsigned i = 0, e = STy->getNumElements(); i != e; i++) {
1356       Type *ETy = STy->getElementType(i);
1357       unsigned int align = getOpenCLAlignment(TD, ETy);
1358       if (align > alignStruct)
1359         alignStruct = align;
1360     }
1361     return alignStruct;
1362   }
1363
1364   const FunctionType *FTy = dyn_cast<FunctionType>(Ty);
1365   if (FTy)
1366     return TD->getPointerPrefAlignment();
1367   return TD->getPrefTypeAlignment(Ty);
1368 }
1369
1370 void NVPTXAsmPrinter::printParamName(Function::const_arg_iterator I,
1371                                      int paramIndex, raw_ostream &O) {
1372   if ((nvptxSubtarget.getDrvInterface() == NVPTX::NVCL) ||
1373       (nvptxSubtarget.getDrvInterface() == NVPTX::CUDA))
1374     O << *getSymbol(I->getParent()) << "_param_" << paramIndex;
1375   else {
1376     std::string argName = I->getName();
1377     const char *p = argName.c_str();
1378     while (*p) {
1379       if (*p == '.')
1380         O << "_";
1381       else
1382         O << *p;
1383       p++;
1384     }
1385   }
1386 }
1387
1388 void NVPTXAsmPrinter::printParamName(int paramIndex, raw_ostream &O) {
1389   Function::const_arg_iterator I, E;
1390   int i = 0;
1391
1392   if ((nvptxSubtarget.getDrvInterface() == NVPTX::NVCL) ||
1393       (nvptxSubtarget.getDrvInterface() == NVPTX::CUDA)) {
1394     O << *CurrentFnSym << "_param_" << paramIndex;
1395     return;
1396   }
1397
1398   for (I = F->arg_begin(), E = F->arg_end(); I != E; ++I, i++) {
1399     if (i == paramIndex) {
1400       printParamName(I, paramIndex, O);
1401       return;
1402     }
1403   }
1404   llvm_unreachable("paramIndex out of bound");
1405 }
1406
1407 void NVPTXAsmPrinter::emitFunctionParamList(const Function *F, raw_ostream &O) {
1408   const DataLayout *TD = TM.getSubtargetImpl()->getDataLayout();
1409   const AttributeSet &PAL = F->getAttributes();
1410   const TargetLowering *TLI = TM.getSubtargetImpl()->getTargetLowering();
1411   Function::const_arg_iterator I, E;
1412   unsigned paramIndex = 0;
1413   bool first = true;
1414   bool isKernelFunc = llvm::isKernelFunction(*F);
1415   bool isABI = (nvptxSubtarget.getSmVersion() >= 20);
1416   MVT thePointerTy = TLI->getPointerTy();
1417
1418   O << "(\n";
1419
1420   for (I = F->arg_begin(), E = F->arg_end(); I != E; ++I, paramIndex++) {
1421     Type *Ty = I->getType();
1422
1423     if (!first)
1424       O << ",\n";
1425
1426     first = false;
1427
1428     // Handle image/sampler parameters
1429     if (isKernelFunction(*F)) {
1430       if (isSampler(*I) || isImage(*I)) {
1431         if (isImage(*I)) {
1432           std::string sname = I->getName();
1433           if (isImageWriteOnly(*I) || isImageReadWrite(*I)) {
1434             if (nvptxSubtarget.hasImageHandles())
1435               O << "\t.param .u64 .ptr .surfref ";
1436             else
1437               O << "\t.param .surfref ";
1438             O << *CurrentFnSym << "_param_" << paramIndex;
1439           }
1440           else { // Default image is read_only
1441             if (nvptxSubtarget.hasImageHandles())
1442               O << "\t.param .u64 .ptr .texref ";
1443             else
1444               O << "\t.param .texref ";
1445             O << *CurrentFnSym << "_param_" << paramIndex;
1446           }
1447         } else {
1448           if (nvptxSubtarget.hasImageHandles())
1449             O << "\t.param .u64 .ptr .samplerref ";
1450           else
1451             O << "\t.param .samplerref ";
1452           O << *CurrentFnSym << "_param_" << paramIndex;
1453         }
1454         continue;
1455       }
1456     }
1457
1458     if (PAL.hasAttribute(paramIndex + 1, Attribute::ByVal) == false) {
1459       if (Ty->isAggregateType() || Ty->isVectorTy()) {
1460         // Just print .param .align <a> .b8 .param[size];
1461         // <a> = PAL.getparamalignment
1462         // size = typeallocsize of element type
1463         unsigned align = PAL.getParamAlignment(paramIndex + 1);
1464         if (align == 0)
1465           align = TD->getABITypeAlignment(Ty);
1466
1467         unsigned sz = TD->getTypeAllocSize(Ty);
1468         O << "\t.param .align " << align << " .b8 ";
1469         printParamName(I, paramIndex, O);
1470         O << "[" << sz << "]";
1471
1472         continue;
1473       }
1474       // Just a scalar
1475       const PointerType *PTy = dyn_cast<PointerType>(Ty);
1476       if (isKernelFunc) {
1477         if (PTy) {
1478           // Special handling for pointer arguments to kernel
1479           O << "\t.param .u" << thePointerTy.getSizeInBits() << " ";
1480
1481           if (nvptxSubtarget.getDrvInterface() != NVPTX::CUDA) {
1482             Type *ETy = PTy->getElementType();
1483             int addrSpace = PTy->getAddressSpace();
1484             switch (addrSpace) {
1485             default:
1486               O << ".ptr ";
1487               break;
1488             case llvm::ADDRESS_SPACE_CONST:
1489               O << ".ptr .const ";
1490               break;
1491             case llvm::ADDRESS_SPACE_SHARED:
1492               O << ".ptr .shared ";
1493               break;
1494             case llvm::ADDRESS_SPACE_GLOBAL:
1495               O << ".ptr .global ";
1496               break;
1497             }
1498             O << ".align " << (int) getOpenCLAlignment(TD, ETy) << " ";
1499           }
1500           printParamName(I, paramIndex, O);
1501           continue;
1502         }
1503
1504         // non-pointer scalar to kernel func
1505         O << "\t.param .";
1506         // Special case: predicate operands become .u8 types
1507         if (Ty->isIntegerTy(1))
1508           O << "u8";
1509         else
1510           O << getPTXFundamentalTypeStr(Ty);
1511         O << " ";
1512         printParamName(I, paramIndex, O);
1513         continue;
1514       }
1515       // Non-kernel function, just print .param .b<size> for ABI
1516       // and .reg .b<size> for non-ABI
1517       unsigned sz = 0;
1518       if (isa<IntegerType>(Ty)) {
1519         sz = cast<IntegerType>(Ty)->getBitWidth();
1520         if (sz < 32)
1521           sz = 32;
1522       } else if (isa<PointerType>(Ty))
1523         sz = thePointerTy.getSizeInBits();
1524       else
1525         sz = Ty->getPrimitiveSizeInBits();
1526       if (isABI)
1527         O << "\t.param .b" << sz << " ";
1528       else
1529         O << "\t.reg .b" << sz << " ";
1530       printParamName(I, paramIndex, O);
1531       continue;
1532     }
1533
1534     // param has byVal attribute. So should be a pointer
1535     const PointerType *PTy = dyn_cast<PointerType>(Ty);
1536     assert(PTy && "Param with byval attribute should be a pointer type");
1537     Type *ETy = PTy->getElementType();
1538
1539     if (isABI || isKernelFunc) {
1540       // Just print .param .align <a> .b8 .param[size];
1541       // <a> = PAL.getparamalignment
1542       // size = typeallocsize of element type
1543       unsigned align = PAL.getParamAlignment(paramIndex + 1);
1544       if (align == 0)
1545         align = TD->getABITypeAlignment(ETy);
1546
1547       unsigned sz = TD->getTypeAllocSize(ETy);
1548       O << "\t.param .align " << align << " .b8 ";
1549       printParamName(I, paramIndex, O);
1550       O << "[" << sz << "]";
1551       continue;
1552     } else {
1553       // Split the ETy into constituent parts and
1554       // print .param .b<size> <name> for each part.
1555       // Further, if a part is vector, print the above for
1556       // each vector element.
1557       SmallVector<EVT, 16> vtparts;
1558       ComputeValueVTs(*TLI, ETy, vtparts);
1559       for (unsigned i = 0, e = vtparts.size(); i != e; ++i) {
1560         unsigned elems = 1;
1561         EVT elemtype = vtparts[i];
1562         if (vtparts[i].isVector()) {
1563           elems = vtparts[i].getVectorNumElements();
1564           elemtype = vtparts[i].getVectorElementType();
1565         }
1566
1567         for (unsigned j = 0, je = elems; j != je; ++j) {
1568           unsigned sz = elemtype.getSizeInBits();
1569           if (elemtype.isInteger() && (sz < 32))
1570             sz = 32;
1571           O << "\t.reg .b" << sz << " ";
1572           printParamName(I, paramIndex, O);
1573           if (j < je - 1)
1574             O << ",\n";
1575           ++paramIndex;
1576         }
1577         if (i < e - 1)
1578           O << ",\n";
1579       }
1580       --paramIndex;
1581       continue;
1582     }
1583   }
1584
1585   O << "\n)\n";
1586 }
1587
1588 void NVPTXAsmPrinter::emitFunctionParamList(const MachineFunction &MF,
1589                                             raw_ostream &O) {
1590   const Function *F = MF.getFunction();
1591   emitFunctionParamList(F, O);
1592 }
1593
1594 void NVPTXAsmPrinter::setAndEmitFunctionVirtualRegisters(
1595     const MachineFunction &MF) {
1596   SmallString<128> Str;
1597   raw_svector_ostream O(Str);
1598
1599   // Map the global virtual register number to a register class specific
1600   // virtual register number starting from 1 with that class.
1601   const TargetRegisterInfo *TRI = MF.getSubtarget().getRegisterInfo();
1602   //unsigned numRegClasses = TRI->getNumRegClasses();
1603
1604   // Emit the Fake Stack Object
1605   const MachineFrameInfo *MFI = MF.getFrameInfo();
1606   int NumBytes = (int) MFI->getStackSize();
1607   if (NumBytes) {
1608     O << "\t.local .align " << MFI->getMaxAlignment() << " .b8 \t" << DEPOTNAME
1609       << getFunctionNumber() << "[" << NumBytes << "];\n";
1610     if (nvptxSubtarget.is64Bit()) {
1611       O << "\t.reg .b64 \t%SP;\n";
1612       O << "\t.reg .b64 \t%SPL;\n";
1613     } else {
1614       O << "\t.reg .b32 \t%SP;\n";
1615       O << "\t.reg .b32 \t%SPL;\n";
1616     }
1617   }
1618
1619   // Go through all virtual registers to establish the mapping between the
1620   // global virtual
1621   // register number and the per class virtual register number.
1622   // We use the per class virtual register number in the ptx output.
1623   unsigned int numVRs = MRI->getNumVirtRegs();
1624   for (unsigned i = 0; i < numVRs; i++) {
1625     unsigned int vr = TRI->index2VirtReg(i);
1626     const TargetRegisterClass *RC = MRI->getRegClass(vr);
1627     DenseMap<unsigned, unsigned> &regmap = VRegMapping[RC];
1628     int n = regmap.size();
1629     regmap.insert(std::make_pair(vr, n + 1));
1630   }
1631
1632   // Emit register declarations
1633   // @TODO: Extract out the real register usage
1634   // O << "\t.reg .pred %p<" << NVPTXNumRegisters << ">;\n";
1635   // O << "\t.reg .s16 %rc<" << NVPTXNumRegisters << ">;\n";
1636   // O << "\t.reg .s16 %rs<" << NVPTXNumRegisters << ">;\n";
1637   // O << "\t.reg .s32 %r<" << NVPTXNumRegisters << ">;\n";
1638   // O << "\t.reg .s64 %rd<" << NVPTXNumRegisters << ">;\n";
1639   // O << "\t.reg .f32 %f<" << NVPTXNumRegisters << ">;\n";
1640   // O << "\t.reg .f64 %fd<" << NVPTXNumRegisters << ">;\n";
1641
1642   // Emit declaration of the virtual registers or 'physical' registers for
1643   // each register class
1644   for (unsigned i=0; i< TRI->getNumRegClasses(); i++) {
1645     const TargetRegisterClass *RC = TRI->getRegClass(i);
1646     DenseMap<unsigned, unsigned> &regmap = VRegMapping[RC];
1647     std::string rcname = getNVPTXRegClassName(RC);
1648     std::string rcStr = getNVPTXRegClassStr(RC);
1649     int n = regmap.size();
1650
1651     // Only declare those registers that may be used.
1652     if (n) {
1653        O << "\t.reg " << rcname << " \t" << rcStr << "<" << (n+1)
1654          << ">;\n";
1655     }
1656   }
1657
1658   OutStreamer.EmitRawText(O.str());
1659 }
1660
1661 void NVPTXAsmPrinter::printFPConstant(const ConstantFP *Fp, raw_ostream &O) {
1662   APFloat APF = APFloat(Fp->getValueAPF()); // make a copy
1663   bool ignored;
1664   unsigned int numHex;
1665   const char *lead;
1666
1667   if (Fp->getType()->getTypeID() == Type::FloatTyID) {
1668     numHex = 8;
1669     lead = "0f";
1670     APF.convert(APFloat::IEEEsingle, APFloat::rmNearestTiesToEven, &ignored);
1671   } else if (Fp->getType()->getTypeID() == Type::DoubleTyID) {
1672     numHex = 16;
1673     lead = "0d";
1674     APF.convert(APFloat::IEEEdouble, APFloat::rmNearestTiesToEven, &ignored);
1675   } else
1676     llvm_unreachable("unsupported fp type");
1677
1678   APInt API = APF.bitcastToAPInt();
1679   std::string hexstr(utohexstr(API.getZExtValue()));
1680   O << lead;
1681   if (hexstr.length() < numHex)
1682     O << std::string(numHex - hexstr.length(), '0');
1683   O << utohexstr(API.getZExtValue());
1684 }
1685
1686 void NVPTXAsmPrinter::printScalarConstant(const Constant *CPV, raw_ostream &O) {
1687   if (const ConstantInt *CI = dyn_cast<ConstantInt>(CPV)) {
1688     O << CI->getValue();
1689     return;
1690   }
1691   if (const ConstantFP *CFP = dyn_cast<ConstantFP>(CPV)) {
1692     printFPConstant(CFP, O);
1693     return;
1694   }
1695   if (isa<ConstantPointerNull>(CPV)) {
1696     O << "0";
1697     return;
1698   }
1699   if (const GlobalValue *GVar = dyn_cast<GlobalValue>(CPV)) {
1700     PointerType *PTy = dyn_cast<PointerType>(GVar->getType());
1701     bool IsNonGenericPointer = false;
1702     if (PTy && PTy->getAddressSpace() != 0) {
1703       IsNonGenericPointer = true;
1704     }
1705     if (EmitGeneric && !isa<Function>(CPV) && !IsNonGenericPointer) {
1706       O << "generic(";
1707       O << *getSymbol(GVar);
1708       O << ")";
1709     } else {
1710       O << *getSymbol(GVar);
1711     }
1712     return;
1713   }
1714   if (const ConstantExpr *Cexpr = dyn_cast<ConstantExpr>(CPV)) {
1715     const Value *v = Cexpr->stripPointerCasts();
1716     PointerType *PTy = dyn_cast<PointerType>(Cexpr->getType());
1717     bool IsNonGenericPointer = false;
1718     if (PTy && PTy->getAddressSpace() != 0) {
1719       IsNonGenericPointer = true;
1720     }
1721     if (const GlobalValue *GVar = dyn_cast<GlobalValue>(v)) {
1722       if (EmitGeneric && !isa<Function>(v) && !IsNonGenericPointer) {
1723         O << "generic(";
1724         O << *getSymbol(GVar);
1725         O << ")";
1726       } else {
1727         O << *getSymbol(GVar);
1728       }
1729       return;
1730     } else {
1731       O << *lowerConstant(CPV);
1732       return;
1733     }
1734   }
1735   llvm_unreachable("Not scalar type found in printScalarConstant()");
1736 }
1737
1738 void NVPTXAsmPrinter::bufferLEByte(const Constant *CPV, int Bytes,
1739                                    AggBuffer *aggBuffer) {
1740
1741   const DataLayout *TD = TM.getSubtargetImpl()->getDataLayout();
1742
1743   if (isa<UndefValue>(CPV) || CPV->isNullValue()) {
1744     int s = TD->getTypeAllocSize(CPV->getType());
1745     if (s < Bytes)
1746       s = Bytes;
1747     aggBuffer->addZeros(s);
1748     return;
1749   }
1750
1751   unsigned char *ptr;
1752   switch (CPV->getType()->getTypeID()) {
1753
1754   case Type::IntegerTyID: {
1755     const Type *ETy = CPV->getType();
1756     if (ETy == Type::getInt8Ty(CPV->getContext())) {
1757       unsigned char c =
1758           (unsigned char)(dyn_cast<ConstantInt>(CPV))->getZExtValue();
1759       ptr = &c;
1760       aggBuffer->addBytes(ptr, 1, Bytes);
1761     } else if (ETy == Type::getInt16Ty(CPV->getContext())) {
1762       short int16 = (short)(dyn_cast<ConstantInt>(CPV))->getZExtValue();
1763       ptr = (unsigned char *)&int16;
1764       aggBuffer->addBytes(ptr, 2, Bytes);
1765     } else if (ETy == Type::getInt32Ty(CPV->getContext())) {
1766       if (const ConstantInt *constInt = dyn_cast<ConstantInt>(CPV)) {
1767         int int32 = (int)(constInt->getZExtValue());
1768         ptr = (unsigned char *)&int32;
1769         aggBuffer->addBytes(ptr, 4, Bytes);
1770         break;
1771       } else if (const ConstantExpr *Cexpr = dyn_cast<ConstantExpr>(CPV)) {
1772         if (const ConstantInt *constInt = dyn_cast<ConstantInt>(
1773                 ConstantFoldConstantExpression(Cexpr, TD))) {
1774           int int32 = (int)(constInt->getZExtValue());
1775           ptr = (unsigned char *)&int32;
1776           aggBuffer->addBytes(ptr, 4, Bytes);
1777           break;
1778         }
1779         if (Cexpr->getOpcode() == Instruction::PtrToInt) {
1780           Value *v = Cexpr->getOperand(0)->stripPointerCasts();
1781           aggBuffer->addSymbol(v);
1782           aggBuffer->addZeros(4);
1783           break;
1784         }
1785       }
1786       llvm_unreachable("unsupported integer const type");
1787     } else if (ETy == Type::getInt64Ty(CPV->getContext())) {
1788       if (const ConstantInt *constInt = dyn_cast<ConstantInt>(CPV)) {
1789         long long int64 = (long long)(constInt->getZExtValue());
1790         ptr = (unsigned char *)&int64;
1791         aggBuffer->addBytes(ptr, 8, Bytes);
1792         break;
1793       } else if (const ConstantExpr *Cexpr = dyn_cast<ConstantExpr>(CPV)) {
1794         if (const ConstantInt *constInt = dyn_cast<ConstantInt>(
1795                 ConstantFoldConstantExpression(Cexpr, TD))) {
1796           long long int64 = (long long)(constInt->getZExtValue());
1797           ptr = (unsigned char *)&int64;
1798           aggBuffer->addBytes(ptr, 8, Bytes);
1799           break;
1800         }
1801         if (Cexpr->getOpcode() == Instruction::PtrToInt) {
1802           Value *v = Cexpr->getOperand(0)->stripPointerCasts();
1803           aggBuffer->addSymbol(v);
1804           aggBuffer->addZeros(8);
1805           break;
1806         }
1807       }
1808       llvm_unreachable("unsupported integer const type");
1809     } else
1810       llvm_unreachable("unsupported integer const type");
1811     break;
1812   }
1813   case Type::FloatTyID:
1814   case Type::DoubleTyID: {
1815     const ConstantFP *CFP = dyn_cast<ConstantFP>(CPV);
1816     const Type *Ty = CFP->getType();
1817     if (Ty == Type::getFloatTy(CPV->getContext())) {
1818       float float32 = (float) CFP->getValueAPF().convertToFloat();
1819       ptr = (unsigned char *)&float32;
1820       aggBuffer->addBytes(ptr, 4, Bytes);
1821     } else if (Ty == Type::getDoubleTy(CPV->getContext())) {
1822       double float64 = CFP->getValueAPF().convertToDouble();
1823       ptr = (unsigned char *)&float64;
1824       aggBuffer->addBytes(ptr, 8, Bytes);
1825     } else {
1826       llvm_unreachable("unsupported fp const type");
1827     }
1828     break;
1829   }
1830   case Type::PointerTyID: {
1831     if (const GlobalValue *GVar = dyn_cast<GlobalValue>(CPV)) {
1832       aggBuffer->addSymbol(GVar);
1833     } else if (const ConstantExpr *Cexpr = dyn_cast<ConstantExpr>(CPV)) {
1834       const Value *v = Cexpr->stripPointerCasts();
1835       aggBuffer->addSymbol(v);
1836     }
1837     unsigned int s = TD->getTypeAllocSize(CPV->getType());
1838     aggBuffer->addZeros(s);
1839     break;
1840   }
1841
1842   case Type::ArrayTyID:
1843   case Type::VectorTyID:
1844   case Type::StructTyID: {
1845     if (isa<ConstantArray>(CPV) || isa<ConstantVector>(CPV) ||
1846         isa<ConstantStruct>(CPV) || isa<ConstantDataSequential>(CPV)) {
1847       int ElementSize = TD->getTypeAllocSize(CPV->getType());
1848       bufferAggregateConstant(CPV, aggBuffer);
1849       if (Bytes > ElementSize)
1850         aggBuffer->addZeros(Bytes - ElementSize);
1851     } else if (isa<ConstantAggregateZero>(CPV))
1852       aggBuffer->addZeros(Bytes);
1853     else
1854       llvm_unreachable("Unexpected Constant type");
1855     break;
1856   }
1857
1858   default:
1859     llvm_unreachable("unsupported type");
1860   }
1861 }
1862
1863 void NVPTXAsmPrinter::bufferAggregateConstant(const Constant *CPV,
1864                                               AggBuffer *aggBuffer) {
1865   const DataLayout *TD = TM.getSubtargetImpl()->getDataLayout();
1866   int Bytes;
1867
1868   // Old constants
1869   if (isa<ConstantArray>(CPV) || isa<ConstantVector>(CPV)) {
1870     if (CPV->getNumOperands())
1871       for (unsigned i = 0, e = CPV->getNumOperands(); i != e; ++i)
1872         bufferLEByte(cast<Constant>(CPV->getOperand(i)), 0, aggBuffer);
1873     return;
1874   }
1875
1876   if (const ConstantDataSequential *CDS =
1877           dyn_cast<ConstantDataSequential>(CPV)) {
1878     if (CDS->getNumElements())
1879       for (unsigned i = 0; i < CDS->getNumElements(); ++i)
1880         bufferLEByte(cast<Constant>(CDS->getElementAsConstant(i)), 0,
1881                      aggBuffer);
1882     return;
1883   }
1884
1885   if (isa<ConstantStruct>(CPV)) {
1886     if (CPV->getNumOperands()) {
1887       StructType *ST = cast<StructType>(CPV->getType());
1888       for (unsigned i = 0, e = CPV->getNumOperands(); i != e; ++i) {
1889         if (i == (e - 1))
1890           Bytes = TD->getStructLayout(ST)->getElementOffset(0) +
1891                   TD->getTypeAllocSize(ST) -
1892                   TD->getStructLayout(ST)->getElementOffset(i);
1893         else
1894           Bytes = TD->getStructLayout(ST)->getElementOffset(i + 1) -
1895                   TD->getStructLayout(ST)->getElementOffset(i);
1896         bufferLEByte(cast<Constant>(CPV->getOperand(i)), Bytes, aggBuffer);
1897       }
1898     }
1899     return;
1900   }
1901   llvm_unreachable("unsupported constant type in printAggregateConstant()");
1902 }
1903
1904 // buildTypeNameMap - Run through symbol table looking for type names.
1905 //
1906
1907 bool NVPTXAsmPrinter::isImageType(const Type *Ty) {
1908
1909   std::map<const Type *, std::string>::iterator PI = TypeNameMap.find(Ty);
1910
1911   if (PI != TypeNameMap.end() && (!PI->second.compare("struct._image1d_t") ||
1912                                   !PI->second.compare("struct._image2d_t") ||
1913                                   !PI->second.compare("struct._image3d_t")))
1914     return true;
1915
1916   return false;
1917 }
1918
1919
1920 bool NVPTXAsmPrinter::ignoreLoc(const MachineInstr &MI) {
1921   switch (MI.getOpcode()) {
1922   default:
1923     return false;
1924   case NVPTX::CallArgBeginInst:
1925   case NVPTX::CallArgEndInst0:
1926   case NVPTX::CallArgEndInst1:
1927   case NVPTX::CallArgF32:
1928   case NVPTX::CallArgF64:
1929   case NVPTX::CallArgI16:
1930   case NVPTX::CallArgI32:
1931   case NVPTX::CallArgI32imm:
1932   case NVPTX::CallArgI64:
1933   case NVPTX::CallArgParam:
1934   case NVPTX::CallVoidInst:
1935   case NVPTX::CallVoidInstReg:
1936   case NVPTX::Callseq_End:
1937   case NVPTX::CallVoidInstReg64:
1938   case NVPTX::DeclareParamInst:
1939   case NVPTX::DeclareRetMemInst:
1940   case NVPTX::DeclareRetRegInst:
1941   case NVPTX::DeclareRetScalarInst:
1942   case NVPTX::DeclareScalarParamInst:
1943   case NVPTX::DeclareScalarRegInst:
1944   case NVPTX::StoreParamF32:
1945   case NVPTX::StoreParamF64:
1946   case NVPTX::StoreParamI16:
1947   case NVPTX::StoreParamI32:
1948   case NVPTX::StoreParamI64:
1949   case NVPTX::StoreParamI8:
1950   case NVPTX::StoreRetvalF32:
1951   case NVPTX::StoreRetvalF64:
1952   case NVPTX::StoreRetvalI16:
1953   case NVPTX::StoreRetvalI32:
1954   case NVPTX::StoreRetvalI64:
1955   case NVPTX::StoreRetvalI8:
1956   case NVPTX::LastCallArgF32:
1957   case NVPTX::LastCallArgF64:
1958   case NVPTX::LastCallArgI16:
1959   case NVPTX::LastCallArgI32:
1960   case NVPTX::LastCallArgI32imm:
1961   case NVPTX::LastCallArgI64:
1962   case NVPTX::LastCallArgParam:
1963   case NVPTX::LoadParamMemF32:
1964   case NVPTX::LoadParamMemF64:
1965   case NVPTX::LoadParamMemI16:
1966   case NVPTX::LoadParamMemI32:
1967   case NVPTX::LoadParamMemI64:
1968   case NVPTX::LoadParamMemI8:
1969   case NVPTX::PrototypeInst:
1970   case NVPTX::DBG_VALUE:
1971     return true;
1972   }
1973   return false;
1974 }
1975
1976 /// PrintAsmOperand - Print out an operand for an inline asm expression.
1977 ///
1978 bool NVPTXAsmPrinter::PrintAsmOperand(const MachineInstr *MI, unsigned OpNo,
1979                                       unsigned AsmVariant,
1980                                       const char *ExtraCode, raw_ostream &O) {
1981   if (ExtraCode && ExtraCode[0]) {
1982     if (ExtraCode[1] != 0)
1983       return true; // Unknown modifier.
1984
1985     switch (ExtraCode[0]) {
1986     default:
1987       // See if this is a generic print operand
1988       return AsmPrinter::PrintAsmOperand(MI, OpNo, AsmVariant, ExtraCode, O);
1989     case 'r':
1990       break;
1991     }
1992   }
1993
1994   printOperand(MI, OpNo, O);
1995
1996   return false;
1997 }
1998
1999 bool NVPTXAsmPrinter::PrintAsmMemoryOperand(
2000     const MachineInstr *MI, unsigned OpNo, unsigned AsmVariant,
2001     const char *ExtraCode, raw_ostream &O) {
2002   if (ExtraCode && ExtraCode[0])
2003     return true; // Unknown modifier
2004
2005   O << '[';
2006   printMemOperand(MI, OpNo, O);
2007   O << ']';
2008
2009   return false;
2010 }
2011
2012 void NVPTXAsmPrinter::printOperand(const MachineInstr *MI, int opNum,
2013                                    raw_ostream &O, const char *Modifier) {
2014   const MachineOperand &MO = MI->getOperand(opNum);
2015   switch (MO.getType()) {
2016   case MachineOperand::MO_Register:
2017     if (TargetRegisterInfo::isPhysicalRegister(MO.getReg())) {
2018       if (MO.getReg() == NVPTX::VRDepot)
2019         O << DEPOTNAME << getFunctionNumber();
2020       else
2021         O << NVPTXInstPrinter::getRegisterName(MO.getReg());
2022     } else {
2023       emitVirtualRegister(MO.getReg(), O);
2024     }
2025     return;
2026
2027   case MachineOperand::MO_Immediate:
2028     if (!Modifier)
2029       O << MO.getImm();
2030     else if (strstr(Modifier, "vec") == Modifier)
2031       printVecModifiedImmediate(MO, Modifier, O);
2032     else
2033       llvm_unreachable(
2034           "Don't know how to handle modifier on immediate operand");
2035     return;
2036
2037   case MachineOperand::MO_FPImmediate:
2038     printFPConstant(MO.getFPImm(), O);
2039     break;
2040
2041   case MachineOperand::MO_GlobalAddress:
2042     O << *getSymbol(MO.getGlobal());
2043     break;
2044
2045   case MachineOperand::MO_MachineBasicBlock:
2046     O << *MO.getMBB()->getSymbol();
2047     return;
2048
2049   default:
2050     llvm_unreachable("Operand type not supported.");
2051   }
2052 }
2053
2054 void NVPTXAsmPrinter::printMemOperand(const MachineInstr *MI, int opNum,
2055                                       raw_ostream &O, const char *Modifier) {
2056   printOperand(MI, opNum, O);
2057
2058   if (Modifier && !strcmp(Modifier, "add")) {
2059     O << ", ";
2060     printOperand(MI, opNum + 1, O);
2061   } else {
2062     if (MI->getOperand(opNum + 1).isImm() &&
2063         MI->getOperand(opNum + 1).getImm() == 0)
2064       return; // don't print ',0' or '+0'
2065     O << "+";
2066     printOperand(MI, opNum + 1, O);
2067   }
2068 }
2069
2070
2071 // Force static initialization.
2072 extern "C" void LLVMInitializeNVPTXBackendAsmPrinter() {
2073   RegisterAsmPrinter<NVPTXAsmPrinter> X(TheNVPTXTarget32);
2074   RegisterAsmPrinter<NVPTXAsmPrinter> Y(TheNVPTXTarget64);
2075 }
2076
2077 void NVPTXAsmPrinter::emitSrcInText(StringRef filename, unsigned line) {
2078   std::stringstream temp;
2079   LineReader *reader = this->getReader(filename.str());
2080   temp << "\n//";
2081   temp << filename.str();
2082   temp << ":";
2083   temp << line;
2084   temp << " ";
2085   temp << reader->readLine(line);
2086   temp << "\n";
2087   this->OutStreamer.EmitRawText(Twine(temp.str()));
2088 }
2089
2090 LineReader *NVPTXAsmPrinter::getReader(std::string filename) {
2091   if (!reader) {
2092     reader = new LineReader(filename);
2093   }
2094
2095   if (reader->fileName() != filename) {
2096     delete reader;
2097     reader = new LineReader(filename);
2098   }
2099
2100   return reader;
2101 }
2102
2103 std::string LineReader::readLine(unsigned lineNum) {
2104   if (lineNum < theCurLine) {
2105     theCurLine = 0;
2106     fstr.seekg(0, std::ios::beg);
2107   }
2108   while (theCurLine < lineNum) {
2109     fstr.getline(buff, 500);
2110     theCurLine++;
2111   }
2112   return buff;
2113 }
2114
2115 // Force static initialization.
2116 extern "C" void LLVMInitializeNVPTXAsmPrinter() {
2117   RegisterAsmPrinter<NVPTXAsmPrinter> X(TheNVPTXTarget32);
2118   RegisterAsmPrinter<NVPTXAsmPrinter> Y(TheNVPTXTarget64);
2119 }