Extracted ObjCARCContract from ObjCARCOpts into its own file.

This also required adding 2x headers Dependency Analysis.h/Provenance Analysis.h
and a .cpp file DependencyAnalysis.cpp to unentangle the dependencies inbetween
ObjCARCContract and ObjCARCOpts.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@173760 91177308-0d34-0410-b5e6-96231b3b80d8

lib/Transforms/ObjCARC/CMakeLists.txt

@@ -5,6 +5,9 @@ add_llvm_library(LLVMObjCARCOpts
   ObjCARCAPElim.cpp
   ObjCARCAliasAnalysis.cpp
   ObjCARCUtil.cpp
+  ObjCARCContract.cpp
+  DependencyAnalysis.cpp
+  ProvenanceAnalysis.cpp
   )
 
 add_dependencies(LLVMObjCARCOpts intrinsics_gen)

lib/Transforms/ObjCARC/DependencyAnalysis.cpp

@@ -0,0 +1,262 @@
+//===- DependencyAnalysis.cpp - ObjC ARC Optimization ---------------------===//
+//
+//                     The LLVM Compiler Infrastructure
+//
+// This file is distributed under the University of Illinois Open Source
+// License. See LICENSE.TXT for details.
+//
+//===----------------------------------------------------------------------===//
+/// \file
+///
+/// This file defines special dependency analysis routines used in Objective C
+/// ARC Optimizations.
+///
+/// WARNING: This file knows about certain library functions. It recognizes them
+/// by name, and hardwires knowledge of their semantics.
+///
+/// WARNING: This file knows about how certain Objective-C library functions are
+/// used. Naive LLVM IR transformations which would otherwise be
+/// behavior-preserving may break these assumptions.
+///
+//===----------------------------------------------------------------------===//
+
+#define DEBUG_TYPE "objc-arc-dependency"
+#include "ObjCARC.h"
+#include "ProvenanceAnalysis.h"
+#include "DependencyAnalysis.h"
+
+#include "llvm/Support/CFG.h"
+
+using namespace llvm;
+using namespace llvm::objcarc;
+
+/// Test whether the given instruction can result in a reference count
+/// modification (positive or negative) for the pointer's object.
+bool
+llvm::objcarc::CanAlterRefCount(const Instruction *Inst, const Value *Ptr,
+                                ProvenanceAnalysis &PA,
+                                InstructionClass Class) {
+  switch (Class) {
+  case IC_Autorelease:
+  case IC_AutoreleaseRV:
+  case IC_User:
+    // These operations never directly modify a reference count.
+    return false;
+  default: break;
+  }
+
+  ImmutableCallSite CS = static_cast<const Value *>(Inst);
+  assert(CS && "Only calls can alter reference counts!");
+
+  // See if AliasAnalysis can help us with the call.
+  AliasAnalysis::ModRefBehavior MRB = PA.getAA()->getModRefBehavior(CS);
+  if (AliasAnalysis::onlyReadsMemory(MRB))
+    return false;
+  if (AliasAnalysis::onlyAccessesArgPointees(MRB)) {
+    for (ImmutableCallSite::arg_iterator I = CS.arg_begin(), E = CS.arg_end();
+         I != E; ++I) {
+      const Value *Op = *I;
+      if (IsPotentialRetainableObjPtr(Op, *PA.getAA()) && PA.related(Ptr, Op))
+        return true;
+    }
+    return false;
+  }
+
+  // Assume the worst.
+  return true;
+}
+
+/// Test whether the given instruction can "use" the given pointer's object in a
+/// way that requires the reference count to be positive.
+bool
+llvm::objcarc::CanUse(const Instruction *Inst, const Value *Ptr,
+                      ProvenanceAnalysis &PA, InstructionClass Class) {
+  // IC_Call operations (as opposed to IC_CallOrUser) never "use" objc pointers.
+  if (Class == IC_Call)
+    return false;
+
+  // Consider various instructions which may have pointer arguments which are
+  // not "uses".
+  if (const ICmpInst *ICI = dyn_cast<ICmpInst>(Inst)) {
+    // Comparing a pointer with null, or any other constant, isn't really a use,
+    // because we don't care what the pointer points to, or about the values
+    // of any other dynamic reference-counted pointers.
+    if (!IsPotentialRetainableObjPtr(ICI->getOperand(1), *PA.getAA()))
+      return false;
+  } else if (ImmutableCallSite CS = static_cast<const Value *>(Inst)) {
+    // For calls, just check the arguments (and not the callee operand).
+    for (ImmutableCallSite::arg_iterator OI = CS.arg_begin(),
+         OE = CS.arg_end(); OI != OE; ++OI) {
+      const Value *Op = *OI;
+      if (IsPotentialRetainableObjPtr(Op, *PA.getAA()) && PA.related(Ptr, Op))
+        return true;
+    }
+    return false;
+  } else if (const StoreInst *SI = dyn_cast<StoreInst>(Inst)) {
+    // Special-case stores, because we don't care about the stored value, just
+    // the store address.
+    const Value *Op = GetUnderlyingObjCPtr(SI->getPointerOperand());
+    // If we can't tell what the underlying object was, assume there is a
+    // dependence.
+    return IsPotentialRetainableObjPtr(Op, *PA.getAA()) && PA.related(Op, Ptr);
+  }
+
+  // Check each operand for a match.
+  for (User::const_op_iterator OI = Inst->op_begin(), OE = Inst->op_end();
+       OI != OE; ++OI) {
+    const Value *Op = *OI;
+    if (IsPotentialRetainableObjPtr(Op, *PA.getAA()) && PA.related(Ptr, Op))
+      return true;
+  }
+  return false;
+}
+
+/// Test if there can be dependencies on Inst through Arg. This function only
+/// tests dependencies relevant for removing pairs of calls.
+bool
+llvm::objcarc::Depends(DependenceKind Flavor, Instruction *Inst,
+                       const Value *Arg, ProvenanceAnalysis &PA) {
+  // If we've reached the definition of Arg, stop.
+  if (Inst == Arg)
+    return true;
+
+  switch (Flavor) {
+  case NeedsPositiveRetainCount: {
+    InstructionClass Class = GetInstructionClass(Inst);
+    switch (Class) {
+    case IC_AutoreleasepoolPop:
+    case IC_AutoreleasepoolPush:
+    case IC_None:
+      return false;
+    default:
+      return CanUse(Inst, Arg, PA, Class);
+    }
+  }
+
+  case AutoreleasePoolBoundary: {
+    InstructionClass Class = GetInstructionClass(Inst);
+    switch (Class) {
+    case IC_AutoreleasepoolPop:
+    case IC_AutoreleasepoolPush:
+      // These mark the end and begin of an autorelease pool scope.
+      return true;
+    default:
+      // Nothing else does this.
+      return false;
+    }
+  }
+
+  case CanChangeRetainCount: {
+    InstructionClass Class = GetInstructionClass(Inst);
+    switch (Class) {
+    case IC_AutoreleasepoolPop:
+      // Conservatively assume this can decrement any count.
+      return true;
+    case IC_AutoreleasepoolPush:
+    case IC_None:
+      return false;
+    default:
+      return CanAlterRefCount(Inst, Arg, PA, Class);
+    }
+  }
+
+  case RetainAutoreleaseDep:
+    switch (GetBasicInstructionClass(Inst)) {
+    case IC_AutoreleasepoolPop:
+    case IC_AutoreleasepoolPush:
+      // Don't merge an objc_autorelease with an objc_retain inside a different
+      // autoreleasepool scope.
+      return true;
+    case IC_Retain:
+    case IC_RetainRV:
+      // Check for a retain of the same pointer for merging.
+      return GetObjCArg(Inst) == Arg;
+    default:
+      // Nothing else matters for objc_retainAutorelease formation.
+      return false;
+    }
+
+  case RetainAutoreleaseRVDep: {
+    InstructionClass Class = GetBasicInstructionClass(Inst);
+    switch (Class) {
+    case IC_Retain:
+    case IC_RetainRV:
+      // Check for a retain of the same pointer for merging.
+      return GetObjCArg(Inst) == Arg;
+    default:
+      // Anything that can autorelease interrupts
+      // retainAutoreleaseReturnValue formation.
+      return CanInterruptRV(Class);
+    }
+  }
+
+  case RetainRVDep:
+    return CanInterruptRV(GetBasicInstructionClass(Inst));
+  }
+
+  llvm_unreachable("Invalid dependence flavor");
+}
+
+/// Walk up the CFG from StartPos (which is in StartBB) and find local and
+/// non-local dependencies on Arg.
+///
+/// TODO: Cache results?
+void
+llvm::objcarc::FindDependencies(DependenceKind Flavor,
+                                const Value *Arg,
+                                BasicBlock *StartBB, Instruction *StartInst,
+                                SmallPtrSet<Instruction *, 4> &DependingInsts,
+                                SmallPtrSet<const BasicBlock *, 4> &Visited,
+                                ProvenanceAnalysis &PA) {
+  BasicBlock::iterator StartPos = StartInst;
+
+  SmallVector<std::pair<BasicBlock *, BasicBlock::iterator>, 4> Worklist;
+  Worklist.push_back(std::make_pair(StartBB, StartPos));
+  do {
+    std::pair<BasicBlock *, BasicBlock::iterator> Pair =
+      Worklist.pop_back_val();
+    BasicBlock *LocalStartBB = Pair.first;
+    BasicBlock::iterator LocalStartPos = Pair.second;
+    BasicBlock::iterator StartBBBegin = LocalStartBB->begin();
+    for (;;) {
+      if (LocalStartPos == StartBBBegin) {
+        pred_iterator PI(LocalStartBB), PE(LocalStartBB, false);
+        if (PI == PE)
+          // If we've reached the function entry, produce a null dependence.
+          DependingInsts.insert(0);
+        else
+          // Add the predecessors to the worklist.
+          do {
+            BasicBlock *PredBB = *PI;
+            if (Visited.insert(PredBB))
+              Worklist.push_back(std::make_pair(PredBB, PredBB->end()));
+          } while (++PI != PE);
+        break;
+      }
+
+      Instruction *Inst = --LocalStartPos;
+      if (Depends(Flavor, Inst, Arg, PA)) {
+        DependingInsts.insert(Inst);
+        break;
+      }
+    }
+  } while (!Worklist.empty());
+
+  // Determine whether the original StartBB post-dominates all of the blocks we
+  // visited. If not, insert a sentinal indicating that most optimizations are
+  // not safe.
+  for (SmallPtrSet<const BasicBlock *, 4>::const_iterator I = Visited.begin(),
+       E = Visited.end(); I != E; ++I) {
+    const BasicBlock *BB = *I;
+    if (BB == StartBB)
+      continue;
+    const TerminatorInst *TI = cast<TerminatorInst>(&BB->back());
+    for (succ_const_iterator SI(TI), SE(TI, false); SI != SE; ++SI) {
+      const BasicBlock *Succ = *SI;
+      if (Succ != StartBB && !Visited.count(Succ)) {
+        DependingInsts.insert(reinterpret_cast<Instruction *>(-1));
+        return;
+      }
+    }
+  }
+}

lib/Transforms/ObjCARC/DependencyAnalysis.h

@@ -0,0 +1,79 @@
+//===- DependencyAnalysis.h - ObjC ARC Optimization ---*- mode: c++ -*-----===//
+//
+//                     The LLVM Compiler Infrastructure
+//
+// This file is distributed under the University of Illinois Open Source
+// License. See LICENSE.TXT for details.
+//
+//===----------------------------------------------------------------------===//
+/// \file
+///
+/// This file declares special dependency analysis routines used in Objective C
+/// ARC Optimizations.
+///
+/// WARNING: This file knows about certain library functions. It recognizes them
+/// by name, and hardwires knowledge of their semantics.
+///
+/// WARNING: This file knows about how certain Objective-C library functions are
+/// used. Naive LLVM IR transformations which would otherwise be
+/// behavior-preserving may break these assumptions.
+///
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_TRANSFORMS_OBJCARC_DEPEDENCYANALYSIS_H
+#define LLVM_TRANSFORMS_OBJCARC_DEPEDENCYANALYSIS_H
+
+#include "llvm/ADT/SmallPtrSet.h"
+
+namespace llvm {
+  class BasicBlock;
+  class Instruction;
+  class Value;
+}
+
+namespace llvm {
+namespace objcarc {
+
+class ProvenanceAnalysis;
+
+/// \enum DependenceKind
+/// \brief Defines different dependence kinds among various ARC constructs.
+///
+/// There are several kinds of dependence-like concepts in use here.
+///
+enum DependenceKind {
+  NeedsPositiveRetainCount,
+  AutoreleasePoolBoundary,
+  CanChangeRetainCount,
+  RetainAutoreleaseDep,       ///< Blocks objc_retainAutorelease.
+  RetainAutoreleaseRVDep,     ///< Blocks objc_retainAutoreleaseReturnValue.
+  RetainRVDep                 ///< Blocks objc_retainAutoreleasedReturnValue.
+};
+
+void FindDependencies(DependenceKind Flavor,
+                      const Value *Arg,
+                      BasicBlock *StartBB, Instruction *StartInst,
+                      SmallPtrSet<Instruction *, 4> &DependingInstructions,
+                      SmallPtrSet<const BasicBlock *, 4> &Visited,
+                      ProvenanceAnalysis &PA);
+
+bool
+Depends(DependenceKind Flavor, Instruction *Inst, const Value *Arg,
+        ProvenanceAnalysis &PA);
+
+/// Test whether the given instruction can "use" the given pointer's object in a
+/// way that requires the reference count to be positive.
+bool
+CanUse(const Instruction *Inst, const Value *Ptr, ProvenanceAnalysis &PA,
+       InstructionClass Class);
+
+/// Test whether the given instruction can result in a reference count
+/// modification (positive or negative) for the pointer's object.
+bool
+CanAlterRefCount(const Instruction *Inst, const Value *Ptr,
+                 ProvenanceAnalysis &PA, InstructionClass Class);
+
+} // namespace objcarc
+} // namespace llvm
+
+#endif // LLVM_TRANSFORMS_OBJCARC_DEPEDENCYANALYSIS_H

lib/Transforms/ObjCARC/ObjCARC.h

@@ -29,10 +29,12 @@
 #include "llvm/Analysis/ValueTracking.h"
 #include "llvm/IR/Module.h"
 #include "llvm/Pass.h"
+#include "llvm/Support/CallSite.h"
 #include "llvm/Support/Debug.h"
 #include "llvm/Support/InstIterator.h"
 #include "llvm/Support/raw_ostream.h"
 #include "llvm/Transforms/ObjCARC.h"
+#include "llvm/Transforms/Utils/Local.h"
 
 namespace llvm {
 namespace objcarc {
@@ -163,6 +165,24 @@ static inline bool IsNoThrow(InstructionClass Class) {
          Class == IC_AutoreleasepoolPop;
 }
 
+/// Test whether the given instruction can autorelease any pointer or cause an
+/// autoreleasepool pop.
+static inline bool
+CanInterruptRV(InstructionClass Class) {
+  switch (Class) {
+  case IC_AutoreleasepoolPop:
+  case IC_CallOrUser:
+  case IC_Call:
+  case IC_Autorelease:
+  case IC_AutoreleaseRV:
+  case IC_FusedRetainAutorelease:
+  case IC_FusedRetainAutoreleaseRV:
+    return true;
+  default:
+    return false;
+  }
+}
+
 /// \brief Determine if F is one of the special known Functions.  If it isn't,
 /// return IC_CallOrUser.
 InstructionClass GetFunctionClass(const Function *F);
@@ -184,6 +204,8 @@ static inline InstructionClass GetBasicInstructionClass(const Value *V) {
   return isa<InvokeInst>(V) ? IC_CallOrUser : IC_User;
 }
 
+/// \brief Determine what kind of construct V is.
+InstructionClass GetInstructionClass(const Value *V);
 
 /// \brief This is a wrapper around getUnderlyingObject which also knows how to
 /// look through objc_retain and objc_autorelease calls, which we know to return
@@ -225,6 +247,137 @@ static inline Value *StripPointerCastsAndObjCCalls(Value *V) {
   return V;
 }
 
+/// \brief Assuming the given instruction is one of the special calls such as
+/// objc_retain or objc_release, return the argument value, stripped of no-op
+/// casts and forwarding calls.
+static inline Value *GetObjCArg(Value *Inst) {
+  return StripPointerCastsAndObjCCalls(cast<CallInst>(Inst)->getArgOperand(0));
+}
+
+static inline bool isNullOrUndef(const Value *V) {
+  return isa<ConstantPointerNull>(V) || isa<UndefValue>(V);
+}
+
+static inline bool isNoopInstruction(const Instruction *I) {
+  return isa<BitCastInst>(I) ||
+    (isa<GetElementPtrInst>(I) &&
+     cast<GetElementPtrInst>(I)->hasAllZeroIndices());
+}
+
+
+/// \brief Erase the given instruction.
+///
+/// Many ObjC calls return their argument verbatim,
+/// so if it's such a call and the return value has users, replace them with the
+/// argument value.
+///
+static inline void EraseInstruction(Instruction *CI) {
+  Value *OldArg = cast<CallInst>(CI)->getArgOperand(0);
+
+  bool Unused = CI->use_empty();
+
+  if (!Unused) {
+    // Replace the return value with the argument.
+    assert(IsForwarding(GetBasicInstructionClass(CI)) &&
+           "Can't delete non-forwarding instruction with users!");
+    CI->replaceAllUsesWith(OldArg);
+  }
+
+  CI->eraseFromParent();
+
+  if (Unused)
+    RecursivelyDeleteTriviallyDeadInstructions(OldArg);
+}
+
+/// \brief Test whether the given value is possible a retainable object pointer.
+static inline bool IsPotentialRetainableObjPtr(const Value *Op) {
+  // Pointers to static or stack storage are not valid retainable object pointers.
+  if (isa<Constant>(Op) || isa<AllocaInst>(Op))
+    return false;
+  // Special arguments can not be a valid retainable object pointer.
+  if (const Argument *Arg = dyn_cast<Argument>(Op))
+    if (Arg->hasByValAttr() ||
+        Arg->hasNestAttr() ||
+        Arg->hasStructRetAttr())
+      return false;
+  // Only consider values with pointer types.
+  //
+  // It seemes intuitive to exclude function pointer types as well, since
+  // functions are never retainable object pointers, however clang occasionally
+  // bitcasts retainable object pointers to function-pointer type temporarily.
+  PointerType *Ty = dyn_cast<PointerType>(Op->getType());
+  if (!Ty)
+    return false;
+  // Conservatively assume anything else is a potential retainable object pointer.
+  return true;
+}
+
+static inline bool IsPotentialRetainableObjPtr(const Value *Op,
+                                               AliasAnalysis &AA) {
+  // First make the rudimentary check.
+  if (!IsPotentialRetainableObjPtr(Op))
+    return false;
+
+  // Objects in constant memory are not reference-counted.
+  if (AA.pointsToConstantMemory(Op))
+    return false;
+
+  // Pointers in constant memory are not pointing to reference-counted objects.
+  if (const LoadInst *LI = dyn_cast<LoadInst>(Op))
+    if (AA.pointsToConstantMemory(LI->getPointerOperand()))
+      return false;
+
+  // Otherwise assume the worst.
+  return true;
+}
+
+/// \brief Helper for GetInstructionClass. Determines what kind of construct CS
+/// is.
+static inline InstructionClass GetCallSiteClass(ImmutableCallSite CS) {
+  for (ImmutableCallSite::arg_iterator I = CS.arg_begin(), E = CS.arg_end();
+       I != E; ++I)
+    if (IsPotentialRetainableObjPtr(*I))
+      return CS.onlyReadsMemory() ? IC_User : IC_CallOrUser;
+
+  return CS.onlyReadsMemory() ? IC_None : IC_Call;
+}
+
+/// \brief Return true if this value refers to a distinct and identifiable
+/// object.
+///
+/// This is similar to AliasAnalysis's isIdentifiedObject, except that it uses
+/// special knowledge of ObjC conventions.
+static inline bool IsObjCIdentifiedObject(const Value *V) {
+  // Assume that call results and arguments have their own "provenance".
+  // Constants (including GlobalVariables) and Allocas are never
+  // reference-counted.
+  if (isa<CallInst>(V) || isa<InvokeInst>(V) ||
+      isa<Argument>(V) || isa<Constant>(V) ||
+      isa<AllocaInst>(V))
+    return true;
+
+  if (const LoadInst *LI = dyn_cast<LoadInst>(V)) {
+    const Value *Pointer =
+      StripPointerCastsAndObjCCalls(LI->getPointerOperand());
+    if (const GlobalVariable *GV = dyn_cast<GlobalVariable>(Pointer)) {
+      // A constant pointer can't be pointing to an object on the heap. It may
+      // be reference-counted, but it won't be deleted.
+      if (GV->isConstant())
+        return true;
+      StringRef Name = GV->getName();
+      // These special variables are known to hold values which are not
+      // reference-counted pointers.
+      if (Name.startswith("\01L_OBJC_SELECTOR_REFERENCES_") ||
+          Name.startswith("\01L_OBJC_CLASSLIST_REFERENCES_") ||
+          Name.startswith("\01L_OBJC_CLASSLIST_SUP_REFS_$_") ||
+          Name.startswith("\01L_OBJC_METH_VAR_NAME_") ||
+          Name.startswith("\01l_objc_msgSend_fixup_"))
+        return true;
+    }
+  }
+
+  return false;
+}
 
 } // end namespace objcarc
 } // end namespace llvm

lib/Transforms/ObjCARC/ObjCARCContract.cpp

@@ -0,0 +1,537 @@
+//===- ObjCARCOpts.cpp - ObjC ARC Optimization ----------------------------===//
+//
+//                     The LLVM Compiler Infrastructure
+//
+// This file is distributed under the University of Illinois Open Source
+// License. See LICENSE.TXT for details.
+//
+//===----------------------------------------------------------------------===//
+/// \file
+/// This file defines late ObjC ARC optimizations. ARC stands for Automatic
+/// Reference Counting and is a system for managing reference counts for objects
+/// in Objective C.
+///
+/// WARNING: This file knows about certain library functions. It recognizes them
+/// by name, and hardwires knowledge of their semantics.
+///
+/// WARNING: This file knows about how certain Objective-C library functions are
+/// used. Naive LLVM IR transformations which would otherwise be
+/// behavior-preserving may break these assumptions.
+///
+//===----------------------------------------------------------------------===//
+
+// TODO: ObjCARCContract could insert PHI nodes when uses aren't
+// dominated by single calls.
+
+#define DEBUG_TYPE "objc-arc-contract"
+#include "ObjCARC.h"
+#include "ProvenanceAnalysis.h"
+#include "DependencyAnalysis.h"
+
+#include "llvm/ADT/Statistic.h"
+#include "llvm/Analysis/Dominators.h"
+#include "llvm/IR/InlineAsm.h"
+#include "llvm/IR/Operator.h"
+
+using namespace llvm;
+using namespace llvm::objcarc;
+
+STATISTIC(NumPeeps,       "Number of calls peephole-optimized");
+STATISTIC(NumStoreStrongs, "Number objc_storeStrong calls formed");
+
+namespace {
+  /// \brief Late ARC optimizations
+  ///
+  /// These change the IR in a way that makes it difficult to be analyzed by
+  /// ObjCARCOpt, so it's run late.
+  class ObjCARCContract : public FunctionPass {
+    bool Changed;
+    AliasAnalysis *AA;
+    DominatorTree *DT;
+    ProvenanceAnalysis PA;
+
+    /// A flag indicating whether this optimization pass should run.
+    bool Run;
+
+    /// Declarations for ObjC runtime functions, for use in creating calls to
+    /// them. These are initialized lazily to avoid cluttering up the Module
+    /// with unused declarations.
+
+    /// Declaration for objc_storeStrong().
+    Constant *StoreStrongCallee;
+    /// Declaration for objc_retainAutorelease().
+    Constant *RetainAutoreleaseCallee;
+    /// Declaration for objc_retainAutoreleaseReturnValue().
+    Constant *RetainAutoreleaseRVCallee;
+
+    /// The inline asm string to insert between calls and RetainRV calls to make
+    /// the optimization work on targets which need it.
+    const MDString *RetainRVMarker;
+
+    /// The set of inserted objc_storeStrong calls. If at the end of walking the
+    /// function we have found no alloca instructions, these calls can be marked
+    /// "tail".
+    SmallPtrSet<CallInst *, 8> StoreStrongCalls;
+
+    Constant *getStoreStrongCallee(Module *M);
+    Constant *getRetainAutoreleaseCallee(Module *M);
+    Constant *getRetainAutoreleaseRVCallee(Module *M);
+
+    bool ContractAutorelease(Function &F, Instruction *Autorelease,
+                             InstructionClass Class,
+                             SmallPtrSet<Instruction *, 4>
+                               &DependingInstructions,
+                             SmallPtrSet<const BasicBlock *, 4>
+                               &Visited);
+
+    void ContractRelease(Instruction *Release,
+                         inst_iterator &Iter);
+
+    virtual void getAnalysisUsage(AnalysisUsage &AU) const;
+    virtual bool doInitialization(Module &M);
+    virtual bool runOnFunction(Function &F);
+
+  public:
+    static char ID;
+    ObjCARCContract() : FunctionPass(ID) {
+      initializeObjCARCContractPass(*PassRegistry::getPassRegistry());
+    }
+  };
+}
+
+char ObjCARCContract::ID = 0;
+INITIALIZE_PASS_BEGIN(ObjCARCContract,
+                      "objc-arc-contract", "ObjC ARC contraction", false, false)
+INITIALIZE_AG_DEPENDENCY(AliasAnalysis)
+INITIALIZE_PASS_DEPENDENCY(DominatorTree)
+INITIALIZE_PASS_END(ObjCARCContract,
+                    "objc-arc-contract", "ObjC ARC contraction", false, false)
+
+Pass *llvm::createObjCARCContractPass() {
+  return new ObjCARCContract();
+}
+
+void ObjCARCContract::getAnalysisUsage(AnalysisUsage &AU) const {
+  AU.addRequired<AliasAnalysis>();
+  AU.addRequired<DominatorTree>();
+  AU.setPreservesCFG();
+}
+
+Constant *ObjCARCContract::getStoreStrongCallee(Module *M) {
+  if (!StoreStrongCallee) {
+    LLVMContext &C = M->getContext();
+    Type *I8X = PointerType::getUnqual(Type::getInt8Ty(C));
+    Type *I8XX = PointerType::getUnqual(I8X);
+    Type *Params[] = { I8XX, I8X };
+
+    AttributeSet Attr = AttributeSet()
+      .addAttribute(M->getContext(), AttributeSet::FunctionIndex,
+                    Attribute::NoUnwind)
+      .addAttribute(M->getContext(), 1, Attribute::NoCapture);
+
+    StoreStrongCallee =
+      M->getOrInsertFunction(
+        "objc_storeStrong",
+        FunctionType::get(Type::getVoidTy(C), Params, /*isVarArg=*/false),
+        Attr);
+  }
+  return StoreStrongCallee;
+}
+
+Constant *ObjCARCContract::getRetainAutoreleaseCallee(Module *M) {
+  if (!RetainAutoreleaseCallee) {
+    LLVMContext &C = M->getContext();
+    Type *I8X = PointerType::getUnqual(Type::getInt8Ty(C));
+    Type *Params[] = { I8X };
+    FunctionType *FTy = FunctionType::get(I8X, Params, /*isVarArg=*/false);
+    AttributeSet Attribute =
+      AttributeSet().addAttribute(M->getContext(), AttributeSet::FunctionIndex,
+                                  Attribute::NoUnwind);
+    RetainAutoreleaseCallee =
+      M->getOrInsertFunction("objc_retainAutorelease", FTy, Attribute);
+  }
+  return RetainAutoreleaseCallee;
+}
+
+Constant *ObjCARCContract::getRetainAutoreleaseRVCallee(Module *M) {
+  if (!RetainAutoreleaseRVCallee) {
+    LLVMContext &C = M->getContext();
+    Type *I8X = PointerType::getUnqual(Type::getInt8Ty(C));
+    Type *Params[] = { I8X };
+    FunctionType *FTy = FunctionType::get(I8X, Params, /*isVarArg=*/false);
+    AttributeSet Attribute =
+      AttributeSet().addAttribute(M->getContext(), AttributeSet::FunctionIndex,
+                                  Attribute::NoUnwind);
+    RetainAutoreleaseRVCallee =
+      M->getOrInsertFunction("objc_retainAutoreleaseReturnValue", FTy,
+                             Attribute);
+  }
+  return RetainAutoreleaseRVCallee;
+}
+
+/// Merge an autorelease with a retain into a fused call.
+bool
+ObjCARCContract::ContractAutorelease(Function &F, Instruction *Autorelease,
+                                     InstructionClass Class,
+                                     SmallPtrSet<Instruction *, 4>
+                                       &DependingInstructions,
+                                     SmallPtrSet<const BasicBlock *, 4>
+                                       &Visited) {
+  const Value *Arg = GetObjCArg(Autorelease);
+
+  // Check that there are no instructions between the retain and the autorelease
+  // (such as an autorelease_pop) which may change the count.
+  CallInst *Retain = 0;
+  if (Class == IC_AutoreleaseRV)
+    FindDependencies(RetainAutoreleaseRVDep, Arg,
+                     Autorelease->getParent(), Autorelease,
+                     DependingInstructions, Visited, PA);
+  else
+    FindDependencies(RetainAutoreleaseDep, Arg,
+                     Autorelease->getParent(), Autorelease,
+                     DependingInstructions, Visited, PA);
+
+  Visited.clear();
+  if (DependingInstructions.size() != 1) {
+    DependingInstructions.clear();
+    return false;
+  }
+
+  Retain = dyn_cast_or_null<CallInst>(*DependingInstructions.begin());
+  DependingInstructions.clear();
+
+  if (!Retain ||
+      GetBasicInstructionClass(Retain) != IC_Retain ||
+      GetObjCArg(Retain) != Arg)
+    return false;
+
+  Changed = true;
+  ++NumPeeps;
+
+  DEBUG(dbgs() << "ObjCARCContract::ContractAutorelease: Fusing "
+                  "retain/autorelease. Erasing: " << *Autorelease << "\n"
+                  "                                      Old Retain: "
+               << *Retain << "\n");
+
+  if (Class == IC_AutoreleaseRV)
+    Retain->setCalledFunction(getRetainAutoreleaseRVCallee(F.getParent()));
+  else
+    Retain->setCalledFunction(getRetainAutoreleaseCallee(F.getParent()));
+
+  DEBUG(dbgs() << "                                      New Retain: "
+               << *Retain << "\n");
+
+  EraseInstruction(Autorelease);
+  return true;
+}
+
+/// Attempt to merge an objc_release with a store, load, and objc_retain to form
+/// an objc_storeStrong. This can be a little tricky because the instructions
+/// don't always appear in order, and there may be unrelated intervening
+/// instructions.
+void ObjCARCContract::ContractRelease(Instruction *Release,
+                                      inst_iterator &Iter) {
+  LoadInst *Load = dyn_cast<LoadInst>(GetObjCArg(Release));
+  if (!Load || !Load->isSimple()) return;
+
+  // For now, require everything to be in one basic block.
+  BasicBlock *BB = Release->getParent();
+  if (Load->getParent() != BB) return;
+
+  // Walk down to find the store and the release, which may be in either order.
+  BasicBlock::iterator I = Load, End = BB->end();
+  ++I;
+  AliasAnalysis::Location Loc = AA->getLocation(Load);
+  StoreInst *Store = 0;
+  bool SawRelease = false;
+  for (; !Store || !SawRelease; ++I) {
+    if (I == End)
+      return;
+
+    Instruction *Inst = I;
+    if (Inst == Release) {
+      SawRelease = true;
+      continue;
+    }
+
+    InstructionClass Class = GetBasicInstructionClass(Inst);
+
+    // Unrelated retains are harmless.
+    if (IsRetain(Class))
+      continue;
+
+    if (Store) {
+      // The store is the point where we're going to put the objc_storeStrong,
+      // so make sure there are no uses after it.
+      if (CanUse(Inst, Load, PA, Class))
+        return;
+    } else if (AA->getModRefInfo(Inst, Loc) & AliasAnalysis::Mod) {
+      // We are moving the load down to the store, so check for anything
+      // else which writes to the memory between the load and the store.
+      Store = dyn_cast<StoreInst>(Inst);
+      if (!Store || !Store->isSimple()) return;
+      if (Store->getPointerOperand() != Loc.Ptr) return;
+    }
+  }
+
+  Value *New = StripPointerCastsAndObjCCalls(Store->getValueOperand());
+
+  // Walk up to find the retain.
+  I = Store;
+  BasicBlock::iterator Begin = BB->begin();
+  while (I != Begin && GetBasicInstructionClass(I) != IC_Retain)
+    --I;
+  Instruction *Retain = I;
+  if (GetBasicInstructionClass(Retain) != IC_Retain) return;
+  if (GetObjCArg(Retain) != New) return;
+
+  Changed = true;
+  ++NumStoreStrongs;
+
+  LLVMContext &C = Release->getContext();
+  Type *I8X = PointerType::getUnqual(Type::getInt8Ty(C));
+  Type *I8XX = PointerType::getUnqual(I8X);
+
+  Value *Args[] = { Load->getPointerOperand(), New };
+  if (Args[0]->getType() != I8XX)
+    Args[0] = new BitCastInst(Args[0], I8XX, "", Store);
+  if (Args[1]->getType() != I8X)
+    Args[1] = new BitCastInst(Args[1], I8X, "", Store);
+  CallInst *StoreStrong =
+    CallInst::Create(getStoreStrongCallee(BB->getParent()->getParent()),
+                     Args, "", Store);
+  StoreStrong->setDoesNotThrow();
+  StoreStrong->setDebugLoc(Store->getDebugLoc());
+
+  // We can't set the tail flag yet, because we haven't yet determined
+  // whether there are any escaping allocas. Remember this call, so that
+  // we can set the tail flag once we know it's safe.
+  StoreStrongCalls.insert(StoreStrong);
+
+  if (&*Iter == Store) ++Iter;
+  Store->eraseFromParent();
+  Release->eraseFromParent();
+  EraseInstruction(Retain);
+  if (Load->use_empty())
+    Load->eraseFromParent();
+}
+
+bool ObjCARCContract::doInitialization(Module &M) {
+  // If nothing in the Module uses ARC, don't do anything.
+  Run = ModuleHasARC(M);
+  if (!Run)
+    return false;
+
+  // These are initialized lazily.
+  StoreStrongCallee = 0;
+  RetainAutoreleaseCallee = 0;
+  RetainAutoreleaseRVCallee = 0;
+
+  // Initialize RetainRVMarker.
+  RetainRVMarker = 0;
+  if (NamedMDNode *NMD =
+        M.getNamedMetadata("clang.arc.retainAutoreleasedReturnValueMarker"))
+    if (NMD->getNumOperands() == 1) {
+      const MDNode *N = NMD->getOperand(0);
+      if (N->getNumOperands() == 1)
+        if (const MDString *S = dyn_cast<MDString>(N->getOperand(0)))
+          RetainRVMarker = S;
+    }
+
+  return false;
+}
+
+bool ObjCARCContract::runOnFunction(Function &F) {
+  if (!EnableARCOpts)
+    return false;
+
+  // If nothing in the Module uses ARC, don't do anything.
+  if (!Run)
+    return false;
+
+  Changed = false;
+  AA = &getAnalysis<AliasAnalysis>();
+  DT = &getAnalysis<DominatorTree>();
+
+  PA.setAA(&getAnalysis<AliasAnalysis>());
+
+  // Track whether it's ok to mark objc_storeStrong calls with the "tail"
+  // keyword. Be conservative if the function has variadic arguments.
+  // It seems that functions which "return twice" are also unsafe for the
+  // "tail" argument, because they are setjmp, which could need to
+  // return to an earlier stack state.
+  bool TailOkForStoreStrongs = !F.isVarArg() &&
+                               !F.callsFunctionThatReturnsTwice();
+
+  // For ObjC library calls which return their argument, replace uses of the
+  // argument with uses of the call return value, if it dominates the use. This
+  // reduces register pressure.
+  SmallPtrSet<Instruction *, 4> DependingInstructions;
+  SmallPtrSet<const BasicBlock *, 4> Visited;
+  for (inst_iterator I = inst_begin(&F), E = inst_end(&F); I != E; ) {
+    Instruction *Inst = &*I++;
+
+    DEBUG(dbgs() << "ObjCARCContract: Visiting: " << *Inst << "\n");
+
+    // Only these library routines return their argument. In particular,
+    // objc_retainBlock does not necessarily return its argument.
+    InstructionClass Class = GetBasicInstructionClass(Inst);
+    switch (Class) {
+    case IC_Retain:
+    case IC_FusedRetainAutorelease:
+    case IC_FusedRetainAutoreleaseRV:
+      break;
+    case IC_Autorelease:
+    case IC_AutoreleaseRV:
+      if (ContractAutorelease(F, Inst, Class, DependingInstructions, Visited))
+        continue;
+      break;
+    case IC_RetainRV: {
+      // If we're compiling for a target which needs a special inline-asm
+      // marker to do the retainAutoreleasedReturnValue optimization,
+      // insert it now.
+      if (!RetainRVMarker)
+        break;
+      BasicBlock::iterator BBI = Inst;
+      BasicBlock *InstParent = Inst->getParent();
+
+      // Step up to see if the call immediately precedes the RetainRV call.
+      // If it's an invoke, we have to cross a block boundary. And we have
+      // to carefully dodge no-op instructions.
+      do {
+        if (&*BBI == InstParent->begin()) {
+          BasicBlock *Pred = InstParent->getSinglePredecessor();
+          if (!Pred)
+            goto decline_rv_optimization;
+          BBI = Pred->getTerminator();
+          break;
+        }
+        --BBI;
+      } while (isNoopInstruction(BBI));
+
+      if (&*BBI == GetObjCArg(Inst)) {
+        DEBUG(dbgs() << "ObjCARCContract: Adding inline asm marker for "
+                        "retainAutoreleasedReturnValue optimization.\n");
+        Changed = true;
+        InlineAsm *IA =
+          InlineAsm::get(FunctionType::get(Type::getVoidTy(Inst->getContext()),
+                                           /*isVarArg=*/false),
+                         RetainRVMarker->getString(),
+                         /*Constraints=*/"", /*hasSideEffects=*/true);
+        CallInst::Create(IA, "", Inst);
+      }
+    decline_rv_optimization:
+      break;
+    }
+    case IC_InitWeak: {
+      // objc_initWeak(p, null) => *p = null
+      CallInst *CI = cast<CallInst>(Inst);
+      if (isNullOrUndef(CI->getArgOperand(1))) {
+        Value *Null =
+          ConstantPointerNull::get(cast<PointerType>(CI->getType()));
+        Changed = true;
+        new StoreInst(Null, CI->getArgOperand(0), CI);
+
+        DEBUG(dbgs() << "OBJCARCContract: Old = " << *CI << "\n"
+                     << "                 New = " << *Null << "\n");
+
+        CI->replaceAllUsesWith(Null);
+        CI->eraseFromParent();
+      }
+      continue;
+    }
+    case IC_Release:
+      ContractRelease(Inst, I);
+      continue;
+    case IC_User:
+      // Be conservative if the function has any alloca instructions.
+      // Technically we only care about escaping alloca instructions,
+      // but this is sufficient to handle some interesting cases.
+      if (isa<AllocaInst>(Inst))
+        TailOkForStoreStrongs = false;
+      continue;
+    default:
+      continue;
+    }
+
+    DEBUG(dbgs() << "ObjCARCContract: Finished List.\n\n");
+
+    // Don't use GetObjCArg because we don't want to look through bitcasts
+    // and such; to do the replacement, the argument must have type i8*.
+    const Value *Arg = cast<CallInst>(Inst)->getArgOperand(0);
+    for (;;) {
+      // If we're compiling bugpointed code, don't get in trouble.
+      if (!isa<Instruction>(Arg) && !isa<Argument>(Arg))
+        break;
+      // Look through the uses of the pointer.
+      for (Value::const_use_iterator UI = Arg->use_begin(), UE = Arg->use_end();
+           UI != UE; ) {
+        Use &U = UI.getUse();
+        unsigned OperandNo = UI.getOperandNo();
+        ++UI; // Increment UI now, because we may unlink its element.
+
+        // If the call's return value dominates a use of the call's argument
+        // value, rewrite the use to use the return value. We check for
+        // reachability here because an unreachable call is considered to
+        // trivially dominate itself, which would lead us to rewriting its
+        // argument in terms of its return value, which would lead to
+        // infinite loops in GetObjCArg.
+        if (DT->isReachableFromEntry(U) && DT->dominates(Inst, U)) {
+          Changed = true;
+          Instruction *Replacement = Inst;
+          Type *UseTy = U.get()->getType();
+          if (PHINode *PHI = dyn_cast<PHINode>(U.getUser())) {
+            // For PHI nodes, insert the bitcast in the predecessor block.
+            unsigned ValNo = PHINode::getIncomingValueNumForOperand(OperandNo);
+            BasicBlock *BB = PHI->getIncomingBlock(ValNo);
+            if (Replacement->getType() != UseTy)
+              Replacement = new BitCastInst(Replacement, UseTy, "",
+                                            &BB->back());
+            // While we're here, rewrite all edges for this PHI, rather
+            // than just one use at a time, to minimize the number of
+            // bitcasts we emit.
+            for (unsigned i = 0, e = PHI->getNumIncomingValues(); i != e; ++i)
+              if (PHI->getIncomingBlock(i) == BB) {
+                // Keep the UI iterator valid.
+                if (&PHI->getOperandUse(
+                      PHINode::getOperandNumForIncomingValue(i)) ==
+                    &UI.getUse())
+                  ++UI;
+                PHI->setIncomingValue(i, Replacement);
+              }
+          } else {
+            if (Replacement->getType() != UseTy)
+              Replacement = new BitCastInst(Replacement, UseTy, "",
+                                            cast<Instruction>(U.getUser()));
+            U.set(Replacement);
+          }
+        }
+      }
+
+      // If Arg is a no-op casted pointer, strip one level of casts and iterate.
+      if (const BitCastInst *BI = dyn_cast<BitCastInst>(Arg))
+        Arg = BI->getOperand(0);
+      else if (isa<GEPOperator>(Arg) &&
+               cast<GEPOperator>(Arg)->hasAllZeroIndices())
+        Arg = cast<GEPOperator>(Arg)->getPointerOperand();
+      else if (isa<GlobalAlias>(Arg) &&
+               !cast<GlobalAlias>(Arg)->mayBeOverridden())
+        Arg = cast<GlobalAlias>(Arg)->getAliasee();
+      else
+        break;
+    }
+  }
+
+  // If this function has no escaping allocas or suspicious vararg usage,
+  // objc_storeStrong calls can be marked with the "tail" keyword.
+  if (TailOkForStoreStrongs)
+    for (SmallPtrSet<CallInst *, 8>::iterator I = StoreStrongCalls.begin(),
+         E = StoreStrongCalls.end(); I != E; ++I)
+      (*I)->setTailCall();
+  StoreStrongCalls.clear();
+
+  return Changed;
+}
+
+/// @}
+///

lib/Transforms/ObjCARC/ObjCARCUtil.cpp

@@ -21,6 +21,8 @@
 
 #include "ObjCARC.h"
 
+#include "llvm/IR/Intrinsics.h"
+
 using namespace llvm;
 using namespace llvm::objcarc;
 
@@ -147,3 +149,94 @@ InstructionClass llvm::objcarc::GetFunctionClass(const Function *F) {
   // Anything else.
   return IC_CallOrUser;
 }
+
+/// \brief Determine what kind of construct V is.
+InstructionClass
+llvm::objcarc::GetInstructionClass(const Value *V) {
+  if (const Instruction *I = dyn_cast<Instruction>(V)) {
+    // Any instruction other than bitcast and gep with a pointer operand have a
+    // use of an objc pointer. Bitcasts, GEPs, Selects, PHIs transfer a pointer
+    // to a subsequent use, rather than using it themselves, in this sense.
+    // As a short cut, several other opcodes are known to have no pointer
+    // operands of interest. And ret is never followed by a release, so it's
+    // not interesting to examine.
+    switch (I->getOpcode()) {
+    case Instruction::Call: {
+      const CallInst *CI = cast<CallInst>(I);
+      // Check for calls to special functions.
+      if (const Function *F = CI->getCalledFunction()) {
+        InstructionClass Class = GetFunctionClass(F);
+        if (Class != IC_CallOrUser)
+          return Class;
+
+        // None of the intrinsic functions do objc_release. For intrinsics, the
+        // only question is whether or not they may be users.
+        switch (F->getIntrinsicID()) {
+        case Intrinsic::returnaddress: case Intrinsic::frameaddress:
+        case Intrinsic::stacksave: case Intrinsic::stackrestore:
+        case Intrinsic::vastart: case Intrinsic::vacopy: case Intrinsic::vaend:
+        case Intrinsic::objectsize: case Intrinsic::prefetch:
+        case Intrinsic::stackprotector:
+        case Intrinsic::eh_return_i32: case Intrinsic::eh_return_i64:
+        case Intrinsic::eh_typeid_for: case Intrinsic::eh_dwarf_cfa:
+        case Intrinsic::eh_sjlj_lsda: case Intrinsic::eh_sjlj_functioncontext:
+        case Intrinsic::init_trampoline: case Intrinsic::adjust_trampoline:
+        case Intrinsic::lifetime_start: case Intrinsic::lifetime_end:
+        case Intrinsic::invariant_start: case Intrinsic::invariant_end:
+        // Don't let dbg info affect our results.
+        case Intrinsic::dbg_declare: case Intrinsic::dbg_value:
+          // Short cut: Some intrinsics obviously don't use ObjC pointers.
+          return IC_None;
+        default:
+          break;
+        }
+      }
+      return GetCallSiteClass(CI);
+    }
+    case Instruction::Invoke:
+      return GetCallSiteClass(cast<InvokeInst>(I));
+    case Instruction::BitCast:
+    case Instruction::GetElementPtr:
+    case Instruction::Select: case Instruction::PHI:
+    case Instruction::Ret: case Instruction::Br:
+    case Instruction::Switch: case Instruction::IndirectBr:
+    case Instruction::Alloca: case Instruction::VAArg:
+    case Instruction::Add: case Instruction::FAdd:
+    case Instruction::Sub: case Instruction::FSub:
+    case Instruction::Mul: case Instruction::FMul:
+    case Instruction::SDiv: case Instruction::UDiv: case Instruction::FDiv:
+    case Instruction::SRem: case Instruction::URem: case Instruction::FRem:
+    case Instruction::Shl: case Instruction::LShr: case Instruction::AShr:
+    case Instruction::And: case Instruction::Or: case Instruction::Xor:
+    case Instruction::SExt: case Instruction::ZExt: case Instruction::Trunc:
+    case Instruction::IntToPtr: case Instruction::FCmp:
+    case Instruction::FPTrunc: case Instruction::FPExt:
+    case Instruction::FPToUI: case Instruction::FPToSI:
+    case Instruction::UIToFP: case Instruction::SIToFP:
+    case Instruction::InsertElement: case Instruction::ExtractElement:
+    case Instruction::ShuffleVector:
+    case Instruction::ExtractValue:
+      break;
+    case Instruction::ICmp:
+      // Comparing a pointer with null, or any other constant, isn't an
+      // interesting use, because we don't care what the pointer points to, or
+      // about the values of any other dynamic reference-counted pointers.
+      if (IsPotentialRetainableObjPtr(I->getOperand(1)))
+        return IC_User;
+      break;
+    default:
+      // For anything else, check all the operands.
+      // Note that this includes both operands of a Store: while the first
+      // operand isn't actually being dereferenced, it is being stored to
+      // memory where we can no longer track who might read it and dereference
+      // it, so we have to consider it potentially used.
+      for (User::const_op_iterator OI = I->op_begin(), OE = I->op_end();
+           OI != OE; ++OI)
+        if (IsPotentialRetainableObjPtr(*OI))
+          return IC_User;
+    }
+  }
+
+  // Otherwise, it's totally inert for ARC purposes.
+  return IC_None;
+}

lib/Transforms/ObjCARC/ProvenanceAnalysis.cpp

@@ -0,0 +1,177 @@
+//===- ProvenanceAnalysis.cpp - ObjC ARC Optimization ---------------------===//
+//
+//                     The LLVM Compiler Infrastructure
+//
+// This file is distributed under the University of Illinois Open Source
+// License. See LICENSE.TXT for details.
+//
+//===----------------------------------------------------------------------===//
+/// \file
+///
+/// This file defines a special form of Alias Analysis called ``Provenance
+/// Analysis''. The word ``provenance'' refers to the history of the ownership
+/// of an object. Thus ``Provenance Analysis'' is an analysis which attempts to
+/// use various techniques to determine if locally
+///
+/// WARNING: This file knows about certain library functions. It recognizes them
+/// by name, and hardwires knowledge of their semantics.
+///
+/// WARNING: This file knows about how certain Objective-C library functions are
+/// used. Naive LLVM IR transformations which would otherwise be
+/// behavior-preserving may break these assumptions.
+///
+//===----------------------------------------------------------------------===//
+
+#include "ObjCARC.h"
+#include "ProvenanceAnalysis.h"
+#include "llvm/ADT/SmallPtrSet.h"
+#include "llvm/ADT/STLExtras.h"
+
+using namespace llvm;
+using namespace llvm::objcarc;
+
+bool ProvenanceAnalysis::relatedSelect(const SelectInst *A,
+                                       const Value *B) {
+  // If the values are Selects with the same condition, we can do a more precise
+  // check: just check for relations between the values on corresponding arms.
+  if (const SelectInst *SB = dyn_cast<SelectInst>(B))
+    if (A->getCondition() == SB->getCondition())
+      return related(A->getTrueValue(), SB->getTrueValue()) ||
+             related(A->getFalseValue(), SB->getFalseValue());
+
+  // Check both arms of the Select node individually.
+  return related(A->getTrueValue(), B) ||
+         related(A->getFalseValue(), B);
+}
+
+bool ProvenanceAnalysis::relatedPHI(const PHINode *A,
+                                    const Value *B) {
+  // If the values are PHIs in the same block, we can do a more precise as well
+  // as efficient check: just check for relations between the values on
+  // corresponding edges.
+  if (const PHINode *PNB = dyn_cast<PHINode>(B))
+    if (PNB->getParent() == A->getParent()) {
+      for (unsigned i = 0, e = A->getNumIncomingValues(); i != e; ++i)
+        if (related(A->getIncomingValue(i),
+                    PNB->getIncomingValueForBlock(A->getIncomingBlock(i))))
+          return true;
+      return false;
+    }
+
+  // Check each unique source of the PHI node against B.
+  SmallPtrSet<const Value *, 4> UniqueSrc;
+  for (unsigned i = 0, e = A->getNumIncomingValues(); i != e; ++i) {
+    const Value *PV1 = A->getIncomingValue(i);
+    if (UniqueSrc.insert(PV1) && related(PV1, B))
+      return true;
+  }
+
+  // All of the arms checked out.
+  return false;
+}
+
+/// Test if the value of P, or any value covered by its provenance, is ever
+/// stored within the function (not counting callees).
+static bool isStoredObjCPointer(const Value *P) {
+  SmallPtrSet<const Value *, 8> Visited;
+  SmallVector<const Value *, 8> Worklist;
+  Worklist.push_back(P);
+  Visited.insert(P);
+  do {
+    P = Worklist.pop_back_val();
+    for (Value::const_use_iterator UI = P->use_begin(), UE = P->use_end();
+         UI != UE; ++UI) {
+      const User *Ur = *UI;
+      if (isa<StoreInst>(Ur)) {
+        if (UI.getOperandNo() == 0)
+          // The pointer is stored.
+          return true;
+        // The pointed is stored through.
+        continue;
+      }
+      if (isa<CallInst>(Ur))
+        // The pointer is passed as an argument, ignore this.
+        continue;
+      if (isa<PtrToIntInst>(P))
+        // Assume the worst.
+        return true;
+      if (Visited.insert(Ur))
+        Worklist.push_back(Ur);
+    }
+  } while (!Worklist.empty());
+
+  // Everything checked out.
+  return false;
+}
+
+bool ProvenanceAnalysis::relatedCheck(const Value *A,
+                                      const Value *B) {
+  // Skip past provenance pass-throughs.
+  A = GetUnderlyingObjCPtr(A);
+  B = GetUnderlyingObjCPtr(B);
+
+  // Quick check.
+  if (A == B)
+    return true;
+
+  // Ask regular AliasAnalysis, for a first approximation.
+  switch (AA->alias(A, B)) {
+  case AliasAnalysis::NoAlias:
+    return false;
+  case AliasAnalysis::MustAlias:
+  case AliasAnalysis::PartialAlias:
+    return true;
+  case AliasAnalysis::MayAlias:
+    break;
+  }
+
+  bool AIsIdentified = IsObjCIdentifiedObject(A);
+  bool BIsIdentified = IsObjCIdentifiedObject(B);
+
+  // An ObjC-Identified object can't alias a load if it is never locally stored.
+  if (AIsIdentified) {
+    // Check for an obvious escape.
+    if (isa<LoadInst>(B))
+      return isStoredObjCPointer(A);
+    if (BIsIdentified) {
+      // Check for an obvious escape.
+      if (isa<LoadInst>(A))
+        return isStoredObjCPointer(B);
+      // Both pointers are identified and escapes aren't an evident problem.
+      return false;
+    }
+  } else if (BIsIdentified) {
+    // Check for an obvious escape.
+    if (isa<LoadInst>(A))
+      return isStoredObjCPointer(B);
+  }
+
+   // Special handling for PHI and Select.
+  if (const PHINode *PN = dyn_cast<PHINode>(A))
+    return relatedPHI(PN, B);
+  if (const PHINode *PN = dyn_cast<PHINode>(B))
+    return relatedPHI(PN, A);
+  if (const SelectInst *S = dyn_cast<SelectInst>(A))
+    return relatedSelect(S, B);
+  if (const SelectInst *S = dyn_cast<SelectInst>(B))
+    return relatedSelect(S, A);
+
+  // Conservative.
+  return true;
+}
+
+bool ProvenanceAnalysis::related(const Value *A,
+                                 const Value *B) {
+  // Begin by inserting a conservative value into the map. If the insertion
+  // fails, we have the answer already. If it succeeds, leave it there until we
+  // compute the real answer to guard against recursive queries.
+  if (A > B) std::swap(A, B);
+  std::pair<CachedResultsTy::iterator, bool> Pair =
+    CachedResults.insert(std::make_pair(ValuePairTy(A, B), true));
+  if (!Pair.second)
+    return Pair.first->second;
+
+  bool Result = relatedCheck(A, B);
+  CachedResults[ValuePairTy(A, B)] = Result;
+  return Result;
+}

lib/Transforms/ObjCARC/ProvenanceAnalysis.h

@@ -0,0 +1,79 @@
+//===- ProvenanceAnalysis.h - ObjC ARC Optimization ---*- mode: c++ -*-----===//
+//
+//                     The LLVM Compiler Infrastructure
+//
+// This file is distributed under the University of Illinois Open Source
+// License. See LICENSE.TXT for details.
+//
+//===----------------------------------------------------------------------===//
+/// \file
+///
+/// This file declares a special form of Alias Analysis called ``Provenance
+/// Analysis''. The word ``provenance'' refers to the history of the ownership
+/// of an object. Thus ``Provenance Analysis'' is an analysis which attempts to
+/// use various techniques to determine if locally
+///
+/// WARNING: This file knows about certain library functions. It recognizes them
+/// by name, and hardwires knowledge of their semantics.
+///
+/// WARNING: This file knows about how certain Objective-C library functions are
+/// used. Naive LLVM IR transformations which would otherwise be
+/// behavior-preserving may break these assumptions.
+///
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_TRANSFORMS_OBJCARC_PROVENANCEANALYSIS_H
+#define LLVM_TRANSFORMS_OBJCARC_PROVENANCEANALYSIS_H
+
+#include "llvm/ADT/DenseMap.h"
+
+namespace llvm {
+  class Value;
+  class AliasAnalysis;
+  class PHINode;
+  class SelectInst;
+}
+
+namespace llvm {
+namespace objcarc {
+/// \brief This is similar to BasicAliasAnalysis, and it uses many of the same
+/// techniques, except it uses special ObjC-specific reasoning about pointer
+/// relationships.
+///
+/// In this context ``Provenance'' is defined as the history of an object's
+/// ownership. Thus ``Provenance Analysis'' is defined by using the notion of
+/// an ``independent provenance source'' of a pointer to determine whether or
+/// not two pointers have the same provenance source and thus could
+/// potentially be related.
+class ProvenanceAnalysis {
+  AliasAnalysis *AA;
+
+  typedef std::pair<const Value *, const Value *> ValuePairTy;
+  typedef DenseMap<ValuePairTy, bool> CachedResultsTy;
+  CachedResultsTy CachedResults;
+
+  bool relatedCheck(const Value *A, const Value *B);
+  bool relatedSelect(const SelectInst *A, const Value *B);
+  bool relatedPHI(const PHINode *A, const Value *B);
+
+  void operator=(const ProvenanceAnalysis &) LLVM_DELETED_FUNCTION;
+  ProvenanceAnalysis(const ProvenanceAnalysis &) LLVM_DELETED_FUNCTION;
+
+public:
+  ProvenanceAnalysis() {}
+
+  void setAA(AliasAnalysis *aa) { AA = aa; }
+
+  AliasAnalysis *getAA() const { return AA; }
+
+  bool related(const Value *A, const Value *B);
+
+  void clear() {
+    CachedResults.clear();
+  }
+};
+
+} // end namespace objcarc
+} // end namespace llvm
+
+#endif // LLVM_TRANSFORMS_OBJCARC_PROVENANCEANALYSIS_H