C++ GlobalVariable::setUnnamedAddr方法代码示例

/// CreateGlobalString - Make a new global variable with an initializer that
/// has array of i8 type filled in with the nul terminated string value
/// specified.  If Name is specified, it is the name of the global variable
/// created.
Value *IRBuilderBase::CreateGlobalString(StringRef Str, const Twine &Name) {
  Constant *StrConstant = ConstantDataArray::getString(Context, Str);
  Module &M = *BB->getParent()->getParent();
  GlobalVariable *GV = new GlobalVariable(M, StrConstant->getType(),
                                          true, GlobalValue::PrivateLinkage,
                                          StrConstant);
  GV->setName(Name);
  GV->setUnnamedAddr(true);
  return GV;
}

GlobalVariable *GCOVProfiler::getEdgeStateValue() {
  GlobalVariable *GV = M->getGlobalVariable("__llvm_gcov_global_state_pred");
  if (!GV) {
    GV = new GlobalVariable(*M, Type::getInt32Ty(*Ctx), false,
                            GlobalValue::InternalLinkage,
                            ConstantInt::get(Type::getInt32Ty(*Ctx),
                                             0xffffffff),
                            "__llvm_gcov_global_state_pred");
    GV->setUnnamedAddr(true);
  }
  return GV;
}

/// CreateGlobalString - Make a new global variable with an initializer that
/// has array of i8 type filled in with the nul terminated string value
/// specified.  If Name is specified, it is the name of the global variable
/// created.
GlobalVariable *IRBuilderBase::CreateGlobalString(StringRef Str,
                                                  const Twine &Name,
                                                  unsigned AddressSpace) {
  Constant *StrConstant = ConstantDataArray::getString(Context, Str);
  Module &M = *BB->getParent()->getParent();
  GlobalVariable *GV = new GlobalVariable(M, StrConstant->getType(),
                                          true, GlobalValue::PrivateLinkage,
                                          StrConstant, Name, nullptr,
                                          GlobalVariable::NotThreadLocal,
                                          AddressSpace);
  GV->setUnnamedAddr(GlobalValue::UnnamedAddr::Global);
  return GV;
}

static Constant *getGlobalString(LLVMContext &C, Module &M,
                                 const StringRef &str) {
  Constant *strArray = ConstantDataArray::getString(C, str);
  GlobalVariable *strVar =
        new GlobalVariable(M, strArray->getType(), true,
                           GlobalValue::PrivateLinkage, strArray, "");
  strVar->setUnnamedAddr(true);
  strVar->setAlignment(1);

  std::vector<Value *> params;
  params.push_back(ConstantInt::get(TypeBuilder<types::i<32>, true>::get(C), 0));
  params.push_back(ConstantInt::get(TypeBuilder<types::i<32>, true>::get(C), 0));

  return ConstantExpr::getInBoundsGetElementPtr(strVar, params);
}

void AMDGPUPromoteAlloca::handleAlloca(AllocaInst &I) {
  // Array allocations are probably not worth handling, since an allocation of
  // the array type is the canonical form.
  if (!I.isStaticAlloca() || I.isArrayAllocation())
    return;

  IRBuilder<> Builder(&I);

  // First try to replace the alloca with a vector
  Type *AllocaTy = I.getAllocatedType();

  DEBUG(dbgs() << "Trying to promote " << I << '\n');

  if (tryPromoteAllocaToVector(&I))
    return;

  DEBUG(dbgs() << " alloca is not a candidate for vectorization.\n");

  const Function &ContainingFunction = *I.getParent()->getParent();

  // FIXME: We should also try to get this value from the reqd_work_group_size
  // function attribute if it is available.
  unsigned WorkGroupSize = AMDGPU::getMaximumWorkGroupSize(ContainingFunction);

  int AllocaSize =
      WorkGroupSize * Mod->getDataLayout().getTypeAllocSize(AllocaTy);

  if (AllocaSize > LocalMemAvailable) {
    DEBUG(dbgs() << " Not enough local memory to promote alloca.\n");
    return;
  }

  std::vector<Value*> WorkList;

  if (!collectUsesWithPtrTypes(&I, WorkList)) {
    DEBUG(dbgs() << " Do not know how to convert all uses\n");
    return;
  }

  DEBUG(dbgs() << "Promoting alloca to local memory\n");
  LocalMemAvailable -= AllocaSize;

  Function *F = I.getParent()->getParent();

  Type *GVTy = ArrayType::get(I.getAllocatedType(), WorkGroupSize);
  GlobalVariable *GV = new GlobalVariable(
      *Mod, GVTy, false, GlobalValue::InternalLinkage,
      UndefValue::get(GVTy),
      Twine(F->getName()) + Twine('.') + I.getName(),
      nullptr,
      GlobalVariable::NotThreadLocal,
      AMDGPUAS::LOCAL_ADDRESS);
  GV->setUnnamedAddr(true);
  GV->setAlignment(I.getAlignment());

  Value *TCntY, *TCntZ;

  std::tie(TCntY, TCntZ) = getLocalSizeYZ(Builder);
  Value *TIdX = getWorkitemID(Builder, 0);
  Value *TIdY = getWorkitemID(Builder, 1);
  Value *TIdZ = getWorkitemID(Builder, 2);

  Value *Tmp0 = Builder.CreateMul(TCntY, TCntZ, "", true, true);
  Tmp0 = Builder.CreateMul(Tmp0, TIdX);
  Value *Tmp1 = Builder.CreateMul(TIdY, TCntZ, "", true, true);
  Value *TID = Builder.CreateAdd(Tmp0, Tmp1);
  TID = Builder.CreateAdd(TID, TIdZ);

  Value *Indices[] = {
    Constant::getNullValue(Type::getInt32Ty(Mod->getContext())),
    TID
  };

  Value *Offset = Builder.CreateInBoundsGEP(GVTy, GV, Indices);
  I.mutateType(Offset->getType());
  I.replaceAllUsesWith(Offset);
  I.eraseFromParent();

  for (Value *V : WorkList) {
    CallInst *Call = dyn_cast<CallInst>(V);
    if (!Call) {
      Type *EltTy = V->getType()->getPointerElementType();
      PointerType *NewTy = PointerType::get(EltTy, AMDGPUAS::LOCAL_ADDRESS);

      // The operand's value should be corrected on its own.
      if (isa<AddrSpaceCastInst>(V))
        continue;

      // FIXME: It doesn't really make sense to try to do this for all
      // instructions.
      V->mutateType(NewTy);
      continue;
    }

    IntrinsicInst *Intr = dyn_cast<IntrinsicInst>(Call);
    if (!Intr) {
      // FIXME: What is this for? It doesn't make sense to promote arbitrary
      // function calls. If the call is to a defined function that can also be
      // promoted, we should be able to do this once that function is also
      // rewritten.
//.........这里部分代码省略.........

// FIXME: Should try to pick the most likely to be profitable allocas first.
bool AMDGPUPromoteAlloca::handleAlloca(AllocaInst &I, bool SufficientLDS) {
  // Array allocations are probably not worth handling, since an allocation of
  // the array type is the canonical form.
  if (!I.isStaticAlloca() || I.isArrayAllocation())
    return false;

  IRBuilder<> Builder(&I);

  // First try to replace the alloca with a vector
  Type *AllocaTy = I.getAllocatedType();

  DEBUG(dbgs() << "Trying to promote " << I << '\n');

  if (tryPromoteAllocaToVector(&I, AS))
    return true; // Promoted to vector.

  const Function &ContainingFunction = *I.getParent()->getParent();
  CallingConv::ID CC = ContainingFunction.getCallingConv();

  // Don't promote the alloca to LDS for shader calling conventions as the work
  // item ID intrinsics are not supported for these calling conventions.
  // Furthermore not all LDS is available for some of the stages.
  switch (CC) {
  case CallingConv::AMDGPU_KERNEL:
  case CallingConv::SPIR_KERNEL:
    break;
  default:
    DEBUG(dbgs() << " promote alloca to LDS not supported with calling convention.\n");
    return false;
  }

  // Not likely to have sufficient local memory for promotion.
  if (!SufficientLDS)
    return false;

  const AMDGPUSubtarget &ST =
    TM->getSubtarget<AMDGPUSubtarget>(ContainingFunction);
  unsigned WorkGroupSize = ST.getFlatWorkGroupSizes(ContainingFunction).second;

  const DataLayout &DL = Mod->getDataLayout();

  unsigned Align = I.getAlignment();
  if (Align == 0)
    Align = DL.getABITypeAlignment(I.getAllocatedType());

  // FIXME: This computed padding is likely wrong since it depends on inverse
  // usage order.
  //
  // FIXME: It is also possible that if we're allowed to use all of the memory
  // could could end up using more than the maximum due to alignment padding.

  uint32_t NewSize = alignTo(CurrentLocalMemUsage, Align);
  uint32_t AllocSize = WorkGroupSize * DL.getTypeAllocSize(AllocaTy);
  NewSize += AllocSize;

  if (NewSize > LocalMemLimit) {
    DEBUG(dbgs() << "  " << AllocSize
          << " bytes of local memory not available to promote\n");
    return false;
  }

  CurrentLocalMemUsage = NewSize;

  std::vector<Value*> WorkList;

  if (!collectUsesWithPtrTypes(&I, &I, WorkList)) {
    DEBUG(dbgs() << " Do not know how to convert all uses\n");
    return false;
  }

  DEBUG(dbgs() << "Promoting alloca to local memory\n");

  Function *F = I.getParent()->getParent();

  Type *GVTy = ArrayType::get(I.getAllocatedType(), WorkGroupSize);
  GlobalVariable *GV = new GlobalVariable(
      *Mod, GVTy, false, GlobalValue::InternalLinkage,
      UndefValue::get(GVTy),
      Twine(F->getName()) + Twine('.') + I.getName(),
      nullptr,
      GlobalVariable::NotThreadLocal,
      AS.LOCAL_ADDRESS);
  GV->setUnnamedAddr(GlobalValue::UnnamedAddr::Global);
  GV->setAlignment(I.getAlignment());

  Value *TCntY, *TCntZ;

  std::tie(TCntY, TCntZ) = getLocalSizeYZ(Builder);
  Value *TIdX = getWorkitemID(Builder, 0);
  Value *TIdY = getWorkitemID(Builder, 1);
  Value *TIdZ = getWorkitemID(Builder, 2);

  Value *Tmp0 = Builder.CreateMul(TCntY, TCntZ, "", true, true);
  Tmp0 = Builder.CreateMul(Tmp0, TIdX);
  Value *Tmp1 = Builder.CreateMul(TIdY, TCntZ, "", true, true);
  Value *TID = Builder.CreateAdd(Tmp0, Tmp1);
  TID = Builder.CreateAdd(TID, TIdZ);

  Value *Indices[] = {
//.........这里部分代码省略.........

bool ConstantMerge::runOnModule(Module &M) {
  TD = getAnalysisIfAvailable<TargetData>();

  // Find all the globals that are marked "used".  These cannot be merged.
  SmallPtrSet<const GlobalValue*, 8> UsedGlobals;
  FindUsedValues(M.getGlobalVariable("llvm.used"), UsedGlobals);
  FindUsedValues(M.getGlobalVariable("llvm.compiler.used"), UsedGlobals);
  
  // Map unique <constants, has-unknown-alignment> pairs to globals.  We don't
  // want to merge globals of unknown alignment with those of explicit
  // alignment.  If we have TargetData, we always know the alignment.
  DenseMap<PointerIntPair<Constant*, 1, bool>, GlobalVariable*> CMap;

  // Replacements - This vector contains a list of replacements to perform.
  SmallVector<std::pair<GlobalVariable*, GlobalVariable*>, 32> Replacements;

  bool MadeChange = false;

  // Iterate constant merging while we are still making progress.  Merging two
  // constants together may allow us to merge other constants together if the
  // second level constants have initializers which point to the globals that
  // were just merged.
  while (1) {

    // First: Find the canonical constants others will be merged with.
    for (Module::global_iterator GVI = M.global_begin(), E = M.global_end();
         GVI != E; ) {
      GlobalVariable *GV = GVI++;

      // If this GV is dead, remove it.
      GV->removeDeadConstantUsers();
      if (GV->use_empty() && GV->hasLocalLinkage()) {
        GV->eraseFromParent();
        continue;
      }

      // Only process constants with initializers in the default address space.
      if (!GV->isConstant() || !GV->hasDefinitiveInitializer() ||
          GV->getType()->getAddressSpace() != 0 || GV->hasSection() ||
          // Don't touch values marked with attribute(used).
          UsedGlobals.count(GV))
        continue;

      // This transformation is legal for weak ODR globals in the sense it
      // doesn't change semantics, but we really don't want to perform it
      // anyway; it's likely to pessimize code generation, and some tools
      // (like the Darwin linker in cases involving CFString) don't expect it.
      if (GV->isWeakForLinker())
        continue;

      Constant *Init = GV->getInitializer();

      // Check to see if the initializer is already known.
      PointerIntPair<Constant*, 1, bool> Pair(Init, hasKnownAlignment(GV));
      GlobalVariable *&Slot = CMap[Pair];

      // If this is the first constant we find or if the old one is local,
      // replace with the current one. If the current is externally visible
      // it cannot be replace, but can be the canonical constant we merge with.
      if (Slot == 0 || IsBetterCannonical(*GV, *Slot))
        Slot = GV;
    }

    // Second: identify all globals that can be merged together, filling in
    // the Replacements vector.  We cannot do the replacement in this pass
    // because doing so may cause initializers of other globals to be rewritten,
    // invalidating the Constant* pointers in CMap.
    for (Module::global_iterator GVI = M.global_begin(), E = M.global_end();
         GVI != E; ) {
      GlobalVariable *GV = GVI++;

      // Only process constants with initializers in the default address space.
      if (!GV->isConstant() || !GV->hasDefinitiveInitializer() ||
          GV->getType()->getAddressSpace() != 0 || GV->hasSection() ||
          // Don't touch values marked with attribute(used).
          UsedGlobals.count(GV))
        continue;

      // We can only replace constant with local linkage.
      if (!GV->hasLocalLinkage())
        continue;

      Constant *Init = GV->getInitializer();

      // Check to see if the initializer is already known.
      PointerIntPair<Constant*, 1, bool> Pair(Init, hasKnownAlignment(GV));
      GlobalVariable *Slot = CMap[Pair];

      if (!Slot || Slot == GV)
        continue;

      if (!Slot->hasUnnamedAddr() && !GV->hasUnnamedAddr())
        continue;

      if (!GV->hasUnnamedAddr())
        Slot->setUnnamedAddr(false);

      // Make all uses of the duplicate constant use the canonical version.
      Replacements.push_back(std::make_pair(GV, Slot));
    }
//.........这里部分代码省略.........