C++ AllocaInst::getAlignment方法代码示例

SizeOffsetType ObjectSizeOffsetVisitor::visitAllocaInst(AllocaInst &I) {
  if (!I.getAllocatedType()->isSized())
    return unknown();

  APInt Size(IntTyBits, TD->getTypeAllocSize(I.getAllocatedType()));
  if (!I.isArrayAllocation())
    return std::make_pair(align(Size, I.getAlignment()), Zero);

  Value *ArraySize = I.getArraySize();
  if (const ConstantInt *C = dyn_cast<ConstantInt>(ArraySize)) {
    Size *= C->getValue().zextOrSelf(IntTyBits);
    return std::make_pair(align(Size, I.getAlignment()), Zero);
  }
  return unknown();
}

SizeOffsetType ObjectSizeOffsetVisitor::visitAllocaInst(AllocaInst &I) {
  if (!I.getAllocatedType()->isSized())
    return unknown();

  APInt Size(IntTyBits, DL.getTypeAllocSize(I.getAllocatedType()));
  if (!I.isArrayAllocation())
    return std::make_pair(align(Size, I.getAlignment()), Zero);

  Value *ArraySize = I.getArraySize();
  if (const ConstantInt *C = dyn_cast<ConstantInt>(ArraySize)) {
    APInt NumElems = C->getValue();
    if (!CheckedZextOrTrunc(NumElems))
      return unknown();

    bool Overflow;
    Size = Size.umul_ov(NumElems, Overflow);
    return Overflow ? unknown() : std::make_pair(align(Size, I.getAlignment()),
                                                 Zero);
  }
  return unknown();
}

void ConstantInsertExtractElementIndex::fixNonConstantVectorIndices(
    BasicBlock &BB, const Instructions &Instrs) const {
  for (Instructions::const_iterator IB = Instrs.begin(), IE = Instrs.end();
       IB != IE; ++IB) {
    Instruction *I = *IB;
    Value *Vec = I->getOperand(0);
    Value *Idx = getInsertExtractElementIdx(I);
    VectorType *VecTy = cast<VectorType>(Vec->getType());
    Type *ElemTy = VecTy->getElementType();
    unsigned ElemAlign = DL->getPrefTypeAlignment(ElemTy);
    unsigned VecAlign = std::max(ElemAlign, DL->getPrefTypeAlignment(VecTy));

    IRBuilder<> IRB(I);
    AllocaInst *Alloca = IRB.CreateAlloca(
        ElemTy, ConstantInt::get(Type::getInt32Ty(M->getContext()),
                                 vectorNumElements(I)));
    Alloca->setAlignment(VecAlign);
    Value *AllocaAsVec = IRB.CreateBitCast(Alloca, VecTy->getPointerTo());
    IRB.CreateAlignedStore(Vec, AllocaAsVec, Alloca->getAlignment());
    Value *GEP = IRB.CreateGEP(Alloca, Idx);

    Value *Res;
    switch (I->getOpcode()) {
    default:
      llvm_unreachable("expected InsertElement or ExtractElement");
    case Instruction::InsertElement:
      IRB.CreateAlignedStore(I->getOperand(1), GEP, ElemAlign);
      Res = IRB.CreateAlignedLoad(AllocaAsVec, Alloca->getAlignment());
      break;
    case Instruction::ExtractElement:
      Res = IRB.CreateAlignedLoad(GEP, ElemAlign);
      break;
    }

    I->replaceAllUsesWith(Res);
    I->eraseFromParent();
  }
}

bool MisalignStackPass::runOnBasicBlock (BasicBlock &BB)
{
    bool Changed = false;
    const unsigned alignLimit = sizeof(uint32_t);

    for (BasicBlock::iterator I = BB.begin(), E = BB.end(); I != E; ++I) {
        AllocaInst *AI = dyn_cast<AllocaInst>(I);

        if (AI && AI->getAlignment() > alignLimit) {
            AI->setAlignment(alignLimit);
            Changed = true;
        }
    }

    return Changed;
}

bool IRTranslator::translateStaticAlloca(const AllocaInst &AI) {
  assert(AI.isStaticAlloca() && "only handle static allocas now");
  MachineFunction &MF = MIRBuilder.getMF();
  unsigned ElementSize = DL->getTypeStoreSize(AI.getAllocatedType());
  unsigned Size =
      ElementSize * cast<ConstantInt>(AI.getArraySize())->getZExtValue();

  // Always allocate at least one byte.
  Size = std::max(Size, 1u);

  unsigned Alignment = AI.getAlignment();
  if (!Alignment)
    Alignment = DL->getABITypeAlignment(AI.getAllocatedType());

  unsigned Res = getOrCreateVReg(AI);
  int FI = MF.getFrameInfo().CreateStackObject(Size, Alignment, false, &AI);
  MIRBuilder.buildFrameIndex(LLT::pointer(0), Res, FI);
  return true;
}

int IRTranslator::getOrCreateFrameIndex(const AllocaInst &AI) {
  if (FrameIndices.find(&AI) != FrameIndices.end())
    return FrameIndices[&AI];

  MachineFunction &MF = MIRBuilder.getMF();
  unsigned ElementSize = DL->getTypeStoreSize(AI.getAllocatedType());
  unsigned Size =
      ElementSize * cast<ConstantInt>(AI.getArraySize())->getZExtValue();

  // Always allocate at least one byte.
  Size = std::max(Size, 1u);

  unsigned Alignment = AI.getAlignment();
  if (!Alignment)
    Alignment = DL->getABITypeAlignment(AI.getAllocatedType());

  int &FI = FrameIndices[&AI];
  FI = MF.getFrameInfo().CreateStackObject(Size, Alignment, false, &AI);
  return FI;
}

/// performCallSlotOptzn - takes a memcpy and a call that it depends on,
/// and checks for the possibility of a call slot optimization by having
/// the call write its result directly into the destination of the memcpy.
bool MemCpyOpt::performCallSlotOptzn(Instruction *cpy,
                                     Value *cpyDest, Value *cpySrc,
                                     uint64_t cpyLen, unsigned cpyAlign,
                                     CallInst *C) {
    // The general transformation to keep in mind is
    //
    //   call @func(..., src, ...)
    //   memcpy(dest, src, ...)
    //
    // ->
    //
    //   memcpy(dest, src, ...)
    //   call @func(..., dest, ...)
    //
    // Since moving the memcpy is technically awkward, we additionally check that
    // src only holds uninitialized values at the moment of the call, meaning that
    // the memcpy can be discarded rather than moved.

    // Deliberately get the source and destination with bitcasts stripped away,
    // because we'll need to do type comparisons based on the underlying type.
    CallSite CS(C);

    // Require that src be an alloca.  This simplifies the reasoning considerably.
    AllocaInst *srcAlloca = dyn_cast<AllocaInst>(cpySrc);
    if (!srcAlloca)
        return false;

    // Check that all of src is copied to dest.
    if (TD == 0) return false;

    ConstantInt *srcArraySize = dyn_cast<ConstantInt>(srcAlloca->getArraySize());
    if (!srcArraySize)
        return false;

    uint64_t srcSize = TD->getTypeAllocSize(srcAlloca->getAllocatedType()) *
                       srcArraySize->getZExtValue();

    if (cpyLen < srcSize)
        return false;

    // Check that dest points to memory that is at least as aligned as src.
    unsigned srcAlign = srcAlloca->getAlignment();
    if (!srcAlign)
        srcAlign = TD->getABITypeAlignment(srcAlloca->getAllocatedType());
    bool isDestSufficientlyAligned = srcAlign <= cpyAlign;
    // If dest is not aligned enough and we can't increase its alignment then
    // bail out.
    if (!isDestSufficientlyAligned && !isa<AllocaInst>(cpyDest))
        return false;

    // Check that accessing the first srcSize bytes of dest will not cause a
    // trap.  Otherwise the transform is invalid since it might cause a trap
    // to occur earlier than it otherwise would.
    if (AllocaInst *A = dyn_cast<AllocaInst>(cpyDest)) {
        // The destination is an alloca.  Check it is larger than srcSize.
        ConstantInt *destArraySize = dyn_cast<ConstantInt>(A->getArraySize());
        if (!destArraySize)
            return false;

        uint64_t destSize = TD->getTypeAllocSize(A->getAllocatedType()) *
                            destArraySize->getZExtValue();

        if (destSize < srcSize)
            return false;
    } else if (Argument *A = dyn_cast<Argument>(cpyDest)) {
        // If the destination is an sret parameter then only accesses that are
        // outside of the returned struct type can trap.
        if (!A->hasStructRetAttr())
            return false;

        Type *StructTy = cast<PointerType>(A->getType())->getElementType();
        uint64_t destSize = TD->getTypeAllocSize(StructTy);

        if (destSize < srcSize)
            return false;
    } else {
        return false;
    }

    // Check that src is not accessed except via the call and the memcpy.  This
    // guarantees that it holds only undefined values when passed in (so the final
    // memcpy can be dropped), that it is not read or written between the call and
    // the memcpy, and that writing beyond the end of it is undefined.
    SmallVector<User*, 8> srcUseList(srcAlloca->use_begin(),
                                     srcAlloca->use_end());
    while (!srcUseList.empty()) {
        User *UI = srcUseList.pop_back_val();

        if (isa<BitCastInst>(UI)) {
            for (User::use_iterator I = UI->use_begin(), E = UI->use_end();
                    I != E; ++I)
                srcUseList.push_back(*I);
        } else if (GetElementPtrInst *G = dyn_cast<GetElementPtrInst>(UI)) {
            if (G->hasAllZeroIndices())
                for (User::use_iterator I = UI->use_begin(), E = UI->use_end();
                        I != E; ++I)
                    srcUseList.push_back(*I);
//.........这里部分代码省略.........

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.........

Instruction *InstCombiner::visitAllocaInst(AllocaInst &AI) {
  if (auto *I = simplifyAllocaArraySize(*this, AI))
    return I;

  if (AI.getAllocatedType()->isSized()) {
    // If the alignment is 0 (unspecified), assign it the preferred alignment.
    if (AI.getAlignment() == 0)
      AI.setAlignment(DL.getPrefTypeAlignment(AI.getAllocatedType()));

    // Move all alloca's of zero byte objects to the entry block and merge them
    // together.  Note that we only do this for alloca's, because malloc should
    // allocate and return a unique pointer, even for a zero byte allocation.
    if (DL.getTypeAllocSize(AI.getAllocatedType()) == 0) {
      // For a zero sized alloca there is no point in doing an array allocation.
      // This is helpful if the array size is a complicated expression not used
      // elsewhere.
      if (AI.isArrayAllocation()) {
        AI.setOperand(0, ConstantInt::get(AI.getArraySize()->getType(), 1));
        return &AI;
      }

      // Get the first instruction in the entry block.
      BasicBlock &EntryBlock = AI.getParent()->getParent()->getEntryBlock();
      Instruction *FirstInst = EntryBlock.getFirstNonPHIOrDbg();
      if (FirstInst != &AI) {
        // If the entry block doesn't start with a zero-size alloca then move
        // this one to the start of the entry block.  There is no problem with
        // dominance as the array size was forced to a constant earlier already.
        AllocaInst *EntryAI = dyn_cast<AllocaInst>(FirstInst);
        if (!EntryAI || !EntryAI->getAllocatedType()->isSized() ||
            DL.getTypeAllocSize(EntryAI->getAllocatedType()) != 0) {
          AI.moveBefore(FirstInst);
          return &AI;
        }

        // If the alignment of the entry block alloca is 0 (unspecified),
        // assign it the preferred alignment.
        if (EntryAI->getAlignment() == 0)
          EntryAI->setAlignment(
              DL.getPrefTypeAlignment(EntryAI->getAllocatedType()));
        // Replace this zero-sized alloca with the one at the start of the entry
        // block after ensuring that the address will be aligned enough for both
        // types.
        unsigned MaxAlign = std::max(EntryAI->getAlignment(),
                                     AI.getAlignment());
        EntryAI->setAlignment(MaxAlign);
        if (AI.getType() != EntryAI->getType())
          return new BitCastInst(EntryAI, AI.getType());
        return ReplaceInstUsesWith(AI, EntryAI);
      }
    }
  }

  if (AI.getAlignment()) {
    // Check to see if this allocation is only modified by a memcpy/memmove from
    // a constant global whose alignment is equal to or exceeds that of the
    // allocation.  If this is the case, we can change all users to use
    // the constant global instead.  This is commonly produced by the CFE by
    // constructs like "void foo() { int A[] = {1,2,3,4,5,6,7,8,9...}; }" if 'A'
    // is only subsequently read.
    SmallVector<Instruction *, 4> ToDelete;
    if (MemTransferInst *Copy = isOnlyCopiedFromConstantGlobal(&AI, ToDelete)) {
      unsigned SourceAlign = getOrEnforceKnownAlignment(
          Copy->getSource(), AI.getAlignment(), DL, &AI, AC, DT);
      if (AI.getAlignment() <= SourceAlign) {
        DEBUG(dbgs() << "Found alloca equal to global: " << AI << '\n');
        DEBUG(dbgs() << "  memcpy = " << *Copy << '\n');
        for (unsigned i = 0, e = ToDelete.size(); i != e; ++i)
          EraseInstFromFunction(*ToDelete[i]);
        Constant *TheSrc = cast<Constant>(Copy->getSource());
        Constant *Cast
          = ConstantExpr::getPointerBitCastOrAddrSpaceCast(TheSrc, AI.getType());
        Instruction *NewI = ReplaceInstUsesWith(AI, Cast);
        EraseInstFromFunction(*Copy);
        ++NumGlobalCopies;
        return NewI;
      }
    }
  }

  // At last, use the generic allocation site handler to aggressively remove
  // unused allocas.
  return visitAllocSite(AI);
}

/// If it is possible to inline the specified call site,
/// do so and update the CallGraph for this operation.
///
/// This function also does some basic book-keeping to update the IR.  The
/// InlinedArrayAllocas map keeps track of any allocas that are already
/// available from other functions inlined into the caller.  If we are able to
/// inline this call site we attempt to reuse already available allocas or add
/// any new allocas to the set if not possible.
static bool InlineCallIfPossible(CallSite CS, InlineFunctionInfo &IFI,
                                 InlinedArrayAllocasTy &InlinedArrayAllocas,
                                 int InlineHistory, bool InsertLifetime) {
  Function *Callee = CS.getCalledFunction();
  Function *Caller = CS.getCaller();

  // Try to inline the function.  Get the list of static allocas that were
  // inlined.
  if (!InlineFunction(CS, IFI, InsertLifetime))
    return false;

  AdjustCallerSSPLevel(Caller, Callee);

  // Look at all of the allocas that we inlined through this call site.  If we
  // have already inlined other allocas through other calls into this function,
  // then we know that they have disjoint lifetimes and that we can merge them.
  //
  // There are many heuristics possible for merging these allocas, and the
  // different options have different tradeoffs.  One thing that we *really*
  // don't want to hurt is SRoA: once inlining happens, often allocas are no
  // longer address taken and so they can be promoted.
  //
  // Our "solution" for that is to only merge allocas whose outermost type is an
  // array type.  These are usually not promoted because someone is using a
  // variable index into them.  These are also often the most important ones to
  // merge.
  //
  // A better solution would be to have real memory lifetime markers in the IR
  // and not have the inliner do any merging of allocas at all.  This would
  // allow the backend to do proper stack slot coloring of all allocas that
  // *actually make it to the backend*, which is really what we want.
  //
  // Because we don't have this information, we do this simple and useful hack.
  //
  SmallPtrSet<AllocaInst*, 16> UsedAllocas;
  
  // When processing our SCC, check to see if CS was inlined from some other
  // call site.  For example, if we're processing "A" in this code:
  //   A() { B() }
  //   B() { x = alloca ... C() }
  //   C() { y = alloca ... }
  // Assume that C was not inlined into B initially, and so we're processing A
  // and decide to inline B into A.  Doing this makes an alloca available for
  // reuse and makes a callsite (C) available for inlining.  When we process
  // the C call site we don't want to do any alloca merging between X and Y
  // because their scopes are not disjoint.  We could make this smarter by
  // keeping track of the inline history for each alloca in the
  // InlinedArrayAllocas but this isn't likely to be a significant win.
  if (InlineHistory != -1)  // Only do merging for top-level call sites in SCC.
    return true;
  
  // Loop over all the allocas we have so far and see if they can be merged with
  // a previously inlined alloca.  If not, remember that we had it.
  for (unsigned AllocaNo = 0, e = IFI.StaticAllocas.size();
       AllocaNo != e; ++AllocaNo) {
    AllocaInst *AI = IFI.StaticAllocas[AllocaNo];
    
    // Don't bother trying to merge array allocations (they will usually be
    // canonicalized to be an allocation *of* an array), or allocations whose
    // type is not itself an array (because we're afraid of pessimizing SRoA).
    ArrayType *ATy = dyn_cast<ArrayType>(AI->getAllocatedType());
    if (!ATy || AI->isArrayAllocation())
      continue;
    
    // Get the list of all available allocas for this array type.
    std::vector<AllocaInst*> &AllocasForType = InlinedArrayAllocas[ATy];
    
    // Loop over the allocas in AllocasForType to see if we can reuse one.  Note
    // that we have to be careful not to reuse the same "available" alloca for
    // multiple different allocas that we just inlined, we use the 'UsedAllocas'
    // set to keep track of which "available" allocas are being used by this
    // function.  Also, AllocasForType can be empty of course!
    bool MergedAwayAlloca = false;
    for (unsigned i = 0, e = AllocasForType.size(); i != e; ++i) {
      AllocaInst *AvailableAlloca = AllocasForType[i];

      unsigned Align1 = AI->getAlignment(),
               Align2 = AvailableAlloca->getAlignment();
      
      // The available alloca has to be in the right function, not in some other
      // function in this SCC.
      if (AvailableAlloca->getParent() != AI->getParent())
        continue;
      
      // If the inlined function already uses this alloca then we can't reuse
      // it.
      if (!UsedAllocas.insert(AvailableAlloca).second)
        continue;
      
      // Otherwise, we *can* reuse it, RAUW AI into AvailableAlloca and declare
      // success!
      DEBUG(dbgs() << "    ***MERGED ALLOCA: " << *AI << "\n\t\tINTO: "
//.........这里部分代码省略.........

/// performCallSlotOptzn - takes a memcpy and a call that it depends on,
/// and checks for the possibility of a call slot optimization by having
/// the call write its result directly into the destination of the memcpy.
bool MemCpyOpt::performCallSlotOptzn(Instruction *cpy,
                                     Value *cpyDest, Value *cpySrc,
                                     uint64_t cpyLen, unsigned cpyAlign,
                                     CallInst *C) {
  // The general transformation to keep in mind is
  //
  //   call @func(..., src, ...)
  //   memcpy(dest, src, ...)
  //
  // ->
  //
  //   memcpy(dest, src, ...)
  //   call @func(..., dest, ...)
  //
  // Since moving the memcpy is technically awkward, we additionally check that
  // src only holds uninitialized values at the moment of the call, meaning that
  // the memcpy can be discarded rather than moved.

  // Deliberately get the source and destination with bitcasts stripped away,
  // because we'll need to do type comparisons based on the underlying type.
  CallSite CS(C);

  // Require that src be an alloca.  This simplifies the reasoning considerably.
  AllocaInst *srcAlloca = dyn_cast<AllocaInst>(cpySrc);
  if (!srcAlloca)
    return false;

  ConstantInt *srcArraySize = dyn_cast<ConstantInt>(srcAlloca->getArraySize());
  if (!srcArraySize)
    return false;

  const DataLayout &DL = cpy->getModule()->getDataLayout();
  uint64_t srcSize = DL.getTypeAllocSize(srcAlloca->getAllocatedType()) *
                     srcArraySize->getZExtValue();

  if (cpyLen < srcSize)
    return false;

  // Check that accessing the first srcSize bytes of dest will not cause a
  // trap.  Otherwise the transform is invalid since it might cause a trap
  // to occur earlier than it otherwise would.
  if (AllocaInst *A = dyn_cast<AllocaInst>(cpyDest)) {
    // The destination is an alloca.  Check it is larger than srcSize.
    ConstantInt *destArraySize = dyn_cast<ConstantInt>(A->getArraySize());
    if (!destArraySize)
      return false;

    uint64_t destSize = DL.getTypeAllocSize(A->getAllocatedType()) *
                        destArraySize->getZExtValue();

    if (destSize < srcSize)
      return false;
  } else if (Argument *A = dyn_cast<Argument>(cpyDest)) {
    if (A->getDereferenceableBytes() < srcSize) {
      // If the destination is an sret parameter then only accesses that are
      // outside of the returned struct type can trap.
      if (!A->hasStructRetAttr())
        return false;

      Type *StructTy = cast<PointerType>(A->getType())->getElementType();
      if (!StructTy->isSized()) {
        // The call may never return and hence the copy-instruction may never
        // be executed, and therefore it's not safe to say "the destination
        // has at least <cpyLen> bytes, as implied by the copy-instruction",
        return false;
      }

      uint64_t destSize = DL.getTypeAllocSize(StructTy);
      if (destSize < srcSize)
        return false;
    }
  } else {
    return false;
  }

  // Check that dest points to memory that is at least as aligned as src.
  unsigned srcAlign = srcAlloca->getAlignment();
  if (!srcAlign)
    srcAlign = DL.getABITypeAlignment(srcAlloca->getAllocatedType());
  bool isDestSufficientlyAligned = srcAlign <= cpyAlign;
  // If dest is not aligned enough and we can't increase its alignment then
  // bail out.
  if (!isDestSufficientlyAligned && !isa<AllocaInst>(cpyDest))
    return false;

  // Check that src is not accessed except via the call and the memcpy.  This
  // guarantees that it holds only undefined values when passed in (so the final
  // memcpy can be dropped), that it is not read or written between the call and
  // the memcpy, and that writing beyond the end of it is undefined.
  SmallVector<User*, 8> srcUseList(srcAlloca->user_begin(),
                                   srcAlloca->user_end());
  while (!srcUseList.empty()) {
    User *U = srcUseList.pop_back_val();

    if (isa<BitCastInst>(U) || isa<AddrSpaceCastInst>(U)) {
      for (User *UU : U->users())
        srcUseList.push_back(UU);
//.........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.........

Instruction *InstCombiner::visitAllocaInst(AllocaInst &AI) {
  // Ensure that the alloca array size argument has type intptr_t, so that
  // any casting is exposed early.
  if (DL) {
    Type *IntPtrTy = DL->getIntPtrType(AI.getType());
    if (AI.getArraySize()->getType() != IntPtrTy) {
      Value *V = Builder->CreateIntCast(AI.getArraySize(),
                                        IntPtrTy, false);
      AI.setOperand(0, V);
      return &AI;
    }
  }

  // Convert: alloca Ty, C - where C is a constant != 1 into: alloca [C x Ty], 1
  if (AI.isArrayAllocation()) {  // Check C != 1
    if (const ConstantInt *C = dyn_cast<ConstantInt>(AI.getArraySize())) {
      Type *NewTy =
        ArrayType::get(AI.getAllocatedType(), C->getZExtValue());
      AllocaInst *New = Builder->CreateAlloca(NewTy, nullptr, AI.getName());
      New->setAlignment(AI.getAlignment());

      // Scan to the end of the allocation instructions, to skip over a block of
      // allocas if possible...also skip interleaved debug info
      //
      BasicBlock::iterator It = New;
      while (isa<AllocaInst>(*It) || isa<DbgInfoIntrinsic>(*It)) ++It;

      // Now that I is pointing to the first non-allocation-inst in the block,
      // insert our getelementptr instruction...
      //
      Type *IdxTy = DL
                  ? DL->getIntPtrType(AI.getType())
                  : Type::getInt64Ty(AI.getContext());
      Value *NullIdx = Constant::getNullValue(IdxTy);
      Value *Idx[2] = { NullIdx, NullIdx };
      Instruction *GEP =
        GetElementPtrInst::CreateInBounds(New, Idx, New->getName() + ".sub");
      InsertNewInstBefore(GEP, *It);

      // Now make everything use the getelementptr instead of the original
      // allocation.
      return ReplaceInstUsesWith(AI, GEP);
    } else if (isa<UndefValue>(AI.getArraySize())) {
      return ReplaceInstUsesWith(AI, Constant::getNullValue(AI.getType()));
    }
  }

  if (DL && AI.getAllocatedType()->isSized()) {
    // If the alignment is 0 (unspecified), assign it the preferred alignment.
    if (AI.getAlignment() == 0)
      AI.setAlignment(DL->getPrefTypeAlignment(AI.getAllocatedType()));

    // Move all alloca's of zero byte objects to the entry block and merge them
    // together.  Note that we only do this for alloca's, because malloc should
    // allocate and return a unique pointer, even for a zero byte allocation.
    if (DL->getTypeAllocSize(AI.getAllocatedType()) == 0) {
      // For a zero sized alloca there is no point in doing an array allocation.
      // This is helpful if the array size is a complicated expression not used
      // elsewhere.
      if (AI.isArrayAllocation()) {
        AI.setOperand(0, ConstantInt::get(AI.getArraySize()->getType(), 1));
        return &AI;
      }

      // Get the first instruction in the entry block.
      BasicBlock &EntryBlock = AI.getParent()->getParent()->getEntryBlock();
      Instruction *FirstInst = EntryBlock.getFirstNonPHIOrDbg();
      if (FirstInst != &AI) {
        // If the entry block doesn't start with a zero-size alloca then move
        // this one to the start of the entry block.  There is no problem with
        // dominance as the array size was forced to a constant earlier already.
        AllocaInst *EntryAI = dyn_cast<AllocaInst>(FirstInst);
        if (!EntryAI || !EntryAI->getAllocatedType()->isSized() ||
            DL->getTypeAllocSize(EntryAI->getAllocatedType()) != 0) {
          AI.moveBefore(FirstInst);
          return &AI;
        }

        // If the alignment of the entry block alloca is 0 (unspecified),
        // assign it the preferred alignment.
        if (EntryAI->getAlignment() == 0)
          EntryAI->setAlignment(
            DL->getPrefTypeAlignment(EntryAI->getAllocatedType()));
        // Replace this zero-sized alloca with the one at the start of the entry
        // block after ensuring that the address will be aligned enough for both
        // types.
        unsigned MaxAlign = std::max(EntryAI->getAlignment(),
                                     AI.getAlignment());
        EntryAI->setAlignment(MaxAlign);
        if (AI.getType() != EntryAI->getType())
          return new BitCastInst(EntryAI, AI.getType());
        return ReplaceInstUsesWith(AI, EntryAI);
      }
    }
  }

  if (AI.getAlignment()) {
    // Check to see if this allocation is only modified by a memcpy/memmove from
    // a constant global whose alignment is equal to or exceeds that of the
    // allocation.  If this is the case, we can change all users to use
//.........这里部分代码省略.........

/// Look at all of the allocas that we inlined through this call site.  If we
/// have already inlined other allocas through other calls into this function,
/// then we know that they have disjoint lifetimes and that we can merge them.
///
/// There are many heuristics possible for merging these allocas, and the
/// different options have different tradeoffs.  One thing that we *really*
/// don't want to hurt is SRoA: once inlining happens, often allocas are no
/// longer address taken and so they can be promoted.
///
/// Our "solution" for that is to only merge allocas whose outermost type is an
/// array type.  These are usually not promoted because someone is using a
/// variable index into them.  These are also often the most important ones to
/// merge.
///
/// A better solution would be to have real memory lifetime markers in the IR
/// and not have the inliner do any merging of allocas at all.  This would
/// allow the backend to do proper stack slot coloring of all allocas that
/// *actually make it to the backend*, which is really what we want.
///
/// Because we don't have this information, we do this simple and useful hack.
static void mergeInlinedArrayAllocas(
    Function *Caller, InlineFunctionInfo &IFI,
    InlinedArrayAllocasTy &InlinedArrayAllocas, int InlineHistory) {
  SmallPtrSet<AllocaInst *, 16> UsedAllocas;

  // When processing our SCC, check to see if CS was inlined from some other
  // call site.  For example, if we're processing "A" in this code:
  //   A() { B() }
  //   B() { x = alloca ... C() }
  //   C() { y = alloca ... }
  // Assume that C was not inlined into B initially, and so we're processing A
  // and decide to inline B into A.  Doing this makes an alloca available for
  // reuse and makes a callsite (C) available for inlining.  When we process
  // the C call site we don't want to do any alloca merging between X and Y
  // because their scopes are not disjoint.  We could make this smarter by
  // keeping track of the inline history for each alloca in the
  // InlinedArrayAllocas but this isn't likely to be a significant win.
  if (InlineHistory != -1) // Only do merging for top-level call sites in SCC.
    return;

  // Loop over all the allocas we have so far and see if they can be merged with
  // a previously inlined alloca.  If not, remember that we had it.
  for (unsigned AllocaNo = 0, e = IFI.StaticAllocas.size(); AllocaNo != e;
       ++AllocaNo) {
    AllocaInst *AI = IFI.StaticAllocas[AllocaNo];

    // Don't bother trying to merge array allocations (they will usually be
    // canonicalized to be an allocation *of* an array), or allocations whose
    // type is not itself an array (because we're afraid of pessimizing SRoA).
    ArrayType *ATy = dyn_cast<ArrayType>(AI->getAllocatedType());
    if (!ATy || AI->isArrayAllocation())
      continue;

    // Get the list of all available allocas for this array type.
    std::vector<AllocaInst *> &AllocasForType = InlinedArrayAllocas[ATy];

    // Loop over the allocas in AllocasForType to see if we can reuse one.  Note
    // that we have to be careful not to reuse the same "available" alloca for
    // multiple different allocas that we just inlined, we use the 'UsedAllocas'
    // set to keep track of which "available" allocas are being used by this
    // function.  Also, AllocasForType can be empty of course!
    bool MergedAwayAlloca = false;
    for (AllocaInst *AvailableAlloca : AllocasForType) {
      unsigned Align1 = AI->getAlignment(),
               Align2 = AvailableAlloca->getAlignment();

      // The available alloca has to be in the right function, not in some other
      // function in this SCC.
      if (AvailableAlloca->getParent() != AI->getParent())
        continue;

      // If the inlined function already uses this alloca then we can't reuse
      // it.
      if (!UsedAllocas.insert(AvailableAlloca).second)
        continue;

      // Otherwise, we *can* reuse it, RAUW AI into AvailableAlloca and declare
      // success!
      LLVM_DEBUG(dbgs() << "    ***MERGED ALLOCA: " << *AI
                        << "\n\t\tINTO: " << *AvailableAlloca << '\n');

      // Move affected dbg.declare calls immediately after the new alloca to
      // avoid the situation when a dbg.declare precedes its alloca.
      if (auto *L = LocalAsMetadata::getIfExists(AI))
        if (auto *MDV = MetadataAsValue::getIfExists(AI->getContext(), L))
          for (User *U : MDV->users())
            if (DbgDeclareInst *DDI = dyn_cast<DbgDeclareInst>(U))
              DDI->moveBefore(AvailableAlloca->getNextNode());

      AI->replaceAllUsesWith(AvailableAlloca);

      if (Align1 != Align2) {
        if (!Align1 || !Align2) {
          const DataLayout &DL = Caller->getParent()->getDataLayout();
          unsigned TypeAlign = DL.getABITypeAlignment(AI->getAllocatedType());

          Align1 = Align1 ? Align1 : TypeAlign;
          Align2 = Align2 ? Align2 : TypeAlign;
        }

//.........这里部分代码省略.........

//
// Method: insertBadAllocationSizes()
//
// Description:
//  This method will look for allocations and change their size to be
//  incorrect.  It does the following:
//    o) Changes the number of array elements allocated by alloca and malloc.
//
// Return value:
//  true  - The module was modified.
//  false - The module was left unmodified.
//
bool
FaultInjector::insertBadAllocationSizes  (Function & F) {
  // Worklist of allocation sites to rewrite
  std::vector<AllocaInst * > WorkList;

  for (Function::iterator fI = F.begin(), fE = F.end(); fI != fE; ++fI) {
    BasicBlock & BB = *fI;
    for (BasicBlock::iterator I = BB.begin(), bE = BB.end(); I != bE; ++I) {
      if (AllocaInst * AI = dyn_cast<AllocaInst>(I)) {
        if (AI->isArrayAllocation()) {
          // Skip if we should not insert a fault.
          if (!doFault()) continue;

          WorkList.push_back(AI);
        }
      }
    }
  }

  while (WorkList.size()) {
    AllocaInst * AI = WorkList.back();
    WorkList.pop_back();

    //
    // Print information about where the fault is being inserted.
    //
    printSourceInfo ("Bad allocation size", AI);

    Instruction * NewAlloc = 0;
    NewAlloc =  new AllocaInst (AI->getAllocatedType(),
                                ConstantInt::get(Int32Type,0),
                                AI->getAlignment(),
                                AI->getName(),
                                AI);
    AI->replaceAllUsesWith (NewAlloc);
    AI->eraseFromParent();
    ++BadSizes;
  }

  //
  // Try harder to make bad allocation sizes.
  //
  WorkList.clear();
  for (Function::iterator fI = F.begin(), fE = F.end(); fI != fE; ++fI) {
    BasicBlock & BB = *fI;
    for (BasicBlock::iterator I = BB.begin(), bE = BB.end(); I != bE; ++I) {
      if (AllocaInst * AI = dyn_cast<AllocaInst>(I)) {
        //
        // Determine if this is a data type that we can make smaller.
        //
        if (((TD->getTypeAllocSize(AI->getAllocatedType())) > 4) && doFault()) {
          WorkList.push_back(AI);
        }
      }
    }
  }

  //
  // Replace these allocations with an allocation of an integer and cast the
  // result back into the appropriate type.
  //
  while (WorkList.size()) {
    AllocaInst * AI = WorkList.back();
    WorkList.pop_back();

    Instruction * NewAlloc = 0;
    NewAlloc =  new AllocaInst (Int32Type,
                                AI->getArraySize(),
                                AI->getAlignment(),
                                AI->getName(),
                                AI);
    NewAlloc = castTo (NewAlloc, AI->getType(), "", AI);
    AI->replaceAllUsesWith (NewAlloc);
    AI->eraseFromParent();
    ++BadSizes;
  }

  return (BadSizes > 0);
}