X-Git-Url: http://demsky.eecs.uci.edu/git/?a=blobdiff_plain;f=lib%2FCodeGen%2FShadowStackGC.cpp;h=6ab0cb03c065042be0ce2a56b5be1598bc7bbe16;hb=6b5252db2db5eeeadec4602329ac56beb6dea54a;hp=0576228bd23b52f5cddfdb0e7593247e4b3b332c;hpb=0598866c052147c31b808391f58434ce3dbfb838;p=oota-llvm.git diff --git a/lib/CodeGen/ShadowStackGC.cpp b/lib/CodeGen/ShadowStackGC.cpp index 0576228bd23..6ab0cb03c06 100644 --- a/lib/CodeGen/ShadowStackGC.cpp +++ b/lib/CodeGen/ShadowStackGC.cpp @@ -31,95 +31,98 @@ #include "llvm/CodeGen/GCStrategy.h" #include "llvm/IntrinsicInst.h" #include "llvm/Module.h" +#include "llvm/Support/CallSite.h" #include "llvm/Support/IRBuilder.h" using namespace llvm; namespace { - - class VISIBILITY_HIDDEN ShadowStackGC : public GCStrategy { + + class ShadowStackGC : public GCStrategy { /// RootChain - This is the global linked-list that contains the chain of GC /// roots. GlobalVariable *Head; - + /// StackEntryTy - Abstract type of a link in the shadow stack. - /// + /// const StructType *StackEntryTy; - + /// Roots - GC roots in the current function. Each is a pair of the /// intrinsic call and its corresponding alloca. std::vector > Roots; - + public: ShadowStackGC(); - + bool initializeCustomLowering(Module &M); bool performCustomLowering(Function &F); - + private: bool IsNullValue(Value *V); Constant *GetFrameMap(Function &F); const Type* GetConcreteStackEntryType(Function &F); void CollectRoots(Function &F); - static GetElementPtrInst *CreateGEP(IRBuilder<> &B, Value *BasePtr, + static GetElementPtrInst *CreateGEP(LLVMContext &Context, + IRBuilder<> &B, Value *BasePtr, int Idx1, const char *Name); - static GetElementPtrInst *CreateGEP(IRBuilder<> &B, Value *BasePtr, + static GetElementPtrInst *CreateGEP(LLVMContext &Context, + IRBuilder<> &B, Value *BasePtr, int Idx1, int Idx2, const char *Name); }; } - + static GCRegistry::Add X("shadow-stack", "Very portable GC for uncooperative code generators"); - + namespace { /// EscapeEnumerator - This is a little algorithm to find all escape points /// from a function so that "finally"-style code can be inserted. In addition /// to finding the existing return and unwind instructions, it also (if /// necessary) transforms any call instructions into invokes and sends them to /// a landing pad. - /// + /// /// It's wrapped up in a state machine using the same transform C# uses for /// 'yield return' enumerators, This transform allows it to be non-allocating. - class VISIBILITY_HIDDEN EscapeEnumerator { + class EscapeEnumerator { Function &F; const char *CleanupBBName; - + // State. int State; Function::iterator StateBB, StateE; IRBuilder<> Builder; - + public: EscapeEnumerator(Function &F, const char *N = "cleanup") - : F(F), CleanupBBName(N), State(0) {} - + : F(F), CleanupBBName(N), State(0), Builder(F.getContext()) {} + IRBuilder<> *Next() { switch (State) { default: return 0; - + case 0: StateBB = F.begin(); StateE = F.end(); State = 1; - + case 1: // Find all 'return' and 'unwind' instructions. while (StateBB != StateE) { BasicBlock *CurBB = StateBB++; - + // Branches and invokes do not escape, only unwind and return do. TerminatorInst *TI = CurBB->getTerminator(); if (!isa(TI) && !isa(TI)) continue; - + Builder.SetInsertPoint(TI->getParent(), TI); return &Builder; } - + State = 2; - + // Find all 'call' instructions. SmallVector Calls; for (Function::iterator BB = F.begin(), @@ -130,34 +133,36 @@ namespace { if (!CI->getCalledFunction() || !CI->getCalledFunction()->getIntrinsicID()) Calls.push_back(CI); - + if (Calls.empty()) return 0; - + // Create a cleanup block. - BasicBlock *CleanupBB = BasicBlock::Create(CleanupBBName, &F); - UnwindInst *UI = new UnwindInst(CleanupBB); - + BasicBlock *CleanupBB = BasicBlock::Create(F.getContext(), + CleanupBBName, &F); + UnwindInst *UI = new UnwindInst(F.getContext(), CleanupBB); + // Transform the 'call' instructions into 'invoke's branching to the // cleanup block. Go in reverse order to make prettier BB names. SmallVector Args; for (unsigned I = Calls.size(); I != 0; ) { CallInst *CI = cast(Calls[--I]); - + // Split the basic block containing the function call. BasicBlock *CallBB = CI->getParent(); BasicBlock *NewBB = CallBB->splitBasicBlock(CI, CallBB->getName() + ".cont"); - + // Remove the unconditional branch inserted at the end of CallBB. CallBB->getInstList().pop_back(); NewBB->getInstList().remove(CI); - + // Create a new invoke instruction. Args.clear(); - Args.append(CI->op_begin() + 1, CI->op_end()); - - InvokeInst *II = InvokeInst::Create(CI->getOperand(0), + CallSite CS(CI); + Args.append(CS.arg_begin(), CS.arg_end()); + + InvokeInst *II = InvokeInst::Create(CI->getCalledValue(), NewBB, CleanupBB, Args.begin(), Args.end(), CI->getName(), CallBB); @@ -166,7 +171,7 @@ namespace { CI->replaceAllUsesWith(II); delete CI; } - + Builder.SetInsertPoint(UI->getParent(), UI); return &Builder; } @@ -185,56 +190,57 @@ ShadowStackGC::ShadowStackGC() : Head(0), StackEntryTy(0) { Constant *ShadowStackGC::GetFrameMap(Function &F) { // doInitialization creates the abstract type of this value. - - Type *VoidPtr = PointerType::getUnqual(Type::Int8Ty); - + const Type *VoidPtr = Type::getInt8PtrTy(F.getContext()); + // Truncate the ShadowStackDescriptor if some metadata is null. unsigned NumMeta = 0; SmallVector Metadata; for (unsigned I = 0; I != Roots.size(); ++I) { - Constant *C = cast(Roots[I].first->getOperand(2)); + Constant *C = cast(Roots[I].first->getArgOperand(1)); if (!C->isNullValue()) NumMeta = I + 1; Metadata.push_back(ConstantExpr::getBitCast(C, VoidPtr)); } - + Constant *BaseElts[] = { - ConstantInt::get(Type::Int32Ty, Roots.size(), false), - ConstantInt::get(Type::Int32Ty, NumMeta, false), + ConstantInt::get(Type::getInt32Ty(F.getContext()), Roots.size(), false), + ConstantInt::get(Type::getInt32Ty(F.getContext()), NumMeta, false), }; - + Constant *DescriptorElts[] = { - ConstantStruct::get(BaseElts, 2), + ConstantStruct::get(F.getContext(), BaseElts, 2, false), ConstantArray::get(ArrayType::get(VoidPtr, NumMeta), Metadata.begin(), NumMeta) }; - - Constant *FrameMap = ConstantStruct::get(DescriptorElts, 2); - + + Constant *FrameMap = ConstantStruct::get(F.getContext(), DescriptorElts, 2, + false); + std::string TypeName("gc_map."); TypeName += utostr(NumMeta); F.getParent()->addTypeName(TypeName, FrameMap->getType()); - + // FIXME: Is this actually dangerous as WritingAnLLVMPass.html claims? Seems // that, short of multithreaded LLVM, it should be safe; all that is // necessary is that a simple Module::iterator loop not be invalidated. // Appending to the GlobalVariable list is safe in that sense. - // + // // All of the output passes emit globals last. The ExecutionEngine // explicitly supports adding globals to the module after // initialization. - // + // // Still, if it isn't deemed acceptable, then this transformation needs // to be a ModulePass (which means it cannot be in the 'llc' pipeline // (which uses a FunctionPassManager (which segfaults (not asserts) if // provided a ModulePass))). - Constant *GV = new GlobalVariable(FrameMap->getType(), true, + Constant *GV = new GlobalVariable(*F.getParent(), FrameMap->getType(), true, GlobalVariable::InternalLinkage, - FrameMap, "__gc_" + F.getName(), - F.getParent()); - - Constant *GEPIndices[2] = { ConstantInt::get(Type::Int32Ty, 0), - ConstantInt::get(Type::Int32Ty, 0) }; + FrameMap, "__gc_" + F.getName()); + + Constant *GEPIndices[2] = { + ConstantInt::get(Type::getInt32Ty(F.getContext()), 0), + ConstantInt::get(Type::getInt32Ty(F.getContext()), 0) + }; return ConstantExpr::getGetElementPtr(GV, GEPIndices, 2); } @@ -244,12 +250,12 @@ const Type* ShadowStackGC::GetConcreteStackEntryType(Function &F) { EltTys.push_back(StackEntryTy); for (size_t I = 0; I != Roots.size(); I++) EltTys.push_back(Roots[I].second->getAllocatedType()); - Type *Ty = StructType::get(EltTys); - + Type *Ty = StructType::get(F.getContext(), EltTys); + std::string TypeName("gc_stackentry."); TypeName += F.getName(); F.getParent()->addTypeName(TypeName, Ty); - + return Ty; } @@ -262,44 +268,46 @@ bool ShadowStackGC::initializeCustomLowering(Module &M) { // void *Meta[]; // May be absent for roots without metadata. // }; std::vector EltTys; - EltTys.push_back(Type::Int32Ty); // 32 bits is ok up to a 32GB stack frame. :) - EltTys.push_back(Type::Int32Ty); // Specifies length of variable length array. - StructType *FrameMapTy = StructType::get(EltTys); + // 32 bits is ok up to a 32GB stack frame. :) + EltTys.push_back(Type::getInt32Ty(M.getContext())); + // Specifies length of variable length array. + EltTys.push_back(Type::getInt32Ty(M.getContext())); + StructType *FrameMapTy = StructType::get(M.getContext(), EltTys); M.addTypeName("gc_map", FrameMapTy); PointerType *FrameMapPtrTy = PointerType::getUnqual(FrameMapTy); - + // struct StackEntry { // ShadowStackEntry *Next; // Caller's stack entry. // FrameMap *Map; // Pointer to constant FrameMap. // void *Roots[]; // Stack roots (in-place array, so we pretend). // }; - OpaqueType *RecursiveTy = OpaqueType::get(); - + OpaqueType *RecursiveTy = OpaqueType::get(M.getContext()); + EltTys.clear(); EltTys.push_back(PointerType::getUnqual(RecursiveTy)); EltTys.push_back(FrameMapPtrTy); - PATypeHolder LinkTyH = StructType::get(EltTys); - + PATypeHolder LinkTyH = StructType::get(M.getContext(), EltTys); + RecursiveTy->refineAbstractTypeTo(LinkTyH.get()); StackEntryTy = cast(LinkTyH.get()); const PointerType *StackEntryPtrTy = PointerType::getUnqual(StackEntryTy); M.addTypeName("gc_stackentry", LinkTyH.get()); // FIXME: Is this safe from // a FunctionPass? - + // Get the root chain if it already exists. Head = M.getGlobalVariable("llvm_gc_root_chain"); if (!Head) { // If the root chain does not exist, insert a new one with linkonce // linkage! - Head = new GlobalVariable(StackEntryPtrTy, false, - GlobalValue::LinkOnceLinkage, + Head = new GlobalVariable(M, StackEntryPtrTy, false, + GlobalValue::LinkOnceAnyLinkage, Constant::getNullValue(StackEntryPtrTy), - "llvm_gc_root_chain", &M); + "llvm_gc_root_chain"); } else if (Head->hasExternalLinkage() && Head->isDeclaration()) { Head->setInitializer(Constant::getNullValue(StackEntryPtrTy)); - Head->setLinkage(GlobalValue::LinkOnceLinkage); + Head->setLinkage(GlobalValue::LinkOnceAnyLinkage); } - + return true; } @@ -313,49 +321,49 @@ void ShadowStackGC::CollectRoots(Function &F) { // FIXME: Account for original alignment. Could fragment the root array. // Approach 1: Null initialize empty slots at runtime. Yuck. // Approach 2: Emit a map of the array instead of just a count. - + assert(Roots.empty() && "Not cleaned up?"); - - SmallVector,16> MetaRoots; - + + SmallVector, 16> MetaRoots; + for (Function::iterator BB = F.begin(), E = F.end(); BB != E; ++BB) for (BasicBlock::iterator II = BB->begin(), E = BB->end(); II != E;) if (IntrinsicInst *CI = dyn_cast(II++)) if (Function *F = CI->getCalledFunction()) if (F->getIntrinsicID() == Intrinsic::gcroot) { - std::pair Pair = std::make_pair( - CI, cast(CI->getOperand(1)->stripPointerCasts())); - if (IsNullValue(CI->getOperand(2))) + std::pair Pair = std::make_pair( + CI, cast(CI->getArgOperand(0)->stripPointerCasts())); + if (IsNullValue(CI->getArgOperand(1))) Roots.push_back(Pair); else MetaRoots.push_back(Pair); } - + // Number roots with metadata (usually empty) at the beginning, so that the // FrameMap::Meta array can be elided. Roots.insert(Roots.begin(), MetaRoots.begin(), MetaRoots.end()); } GetElementPtrInst * -ShadowStackGC::CreateGEP(IRBuilder<> &B, Value *BasePtr, +ShadowStackGC::CreateGEP(LLVMContext &Context, IRBuilder<> &B, Value *BasePtr, int Idx, int Idx2, const char *Name) { - Value *Indices[] = { ConstantInt::get(Type::Int32Ty, 0), - ConstantInt::get(Type::Int32Ty, Idx), - ConstantInt::get(Type::Int32Ty, Idx2) }; + Value *Indices[] = { ConstantInt::get(Type::getInt32Ty(Context), 0), + ConstantInt::get(Type::getInt32Ty(Context), Idx), + ConstantInt::get(Type::getInt32Ty(Context), Idx2) }; Value* Val = B.CreateGEP(BasePtr, Indices, Indices + 3, Name); - + assert(isa(Val) && "Unexpected folded constant"); - + return dyn_cast(Val); } GetElementPtrInst * -ShadowStackGC::CreateGEP(IRBuilder<> &B, Value *BasePtr, +ShadowStackGC::CreateGEP(LLVMContext &Context, IRBuilder<> &B, Value *BasePtr, int Idx, const char *Name) { - Value *Indices[] = { ConstantInt::get(Type::Int32Ty, 0), - ConstantInt::get(Type::Int32Ty, Idx) }; + Value *Indices[] = { ConstantInt::get(Type::getInt32Ty(Context), 0), + ConstantInt::get(Type::getInt32Ty(Context), Idx) }; Value *Val = B.CreateGEP(BasePtr, Indices, Indices + 2, Name); - + assert(isa(Val) && "Unexpected folded constant"); return dyn_cast(Val); @@ -363,68 +371,73 @@ ShadowStackGC::CreateGEP(IRBuilder<> &B, Value *BasePtr, /// runOnFunction - Insert code to maintain the shadow stack. bool ShadowStackGC::performCustomLowering(Function &F) { + LLVMContext &Context = F.getContext(); + // Find calls to llvm.gcroot. CollectRoots(F); - + // If there are no roots in this function, then there is no need to add a // stack map entry for it. if (Roots.empty()) return false; - + // Build the constant map and figure the type of the shadow stack entry. Value *FrameMap = GetFrameMap(F); const Type *ConcreteStackEntryTy = GetConcreteStackEntryType(F); - + // Build the shadow stack entry at the very start of the function. BasicBlock::iterator IP = F.getEntryBlock().begin(); IRBuilder<> AtEntry(IP->getParent(), IP); - + Instruction *StackEntry = AtEntry.CreateAlloca(ConcreteStackEntryTy, 0, "gc_frame"); - + while (isa(IP)) ++IP; AtEntry.SetInsertPoint(IP->getParent(), IP); - + // Initialize the map pointer and load the current head of the shadow stack. Instruction *CurrentHead = AtEntry.CreateLoad(Head, "gc_currhead"); - Instruction *EntryMapPtr = CreateGEP(AtEntry, StackEntry,0,1,"gc_frame.map"); + Instruction *EntryMapPtr = CreateGEP(Context, AtEntry, StackEntry, + 0,1,"gc_frame.map"); AtEntry.CreateStore(FrameMap, EntryMapPtr); - + // After all the allocas... for (unsigned I = 0, E = Roots.size(); I != E; ++I) { // For each root, find the corresponding slot in the aggregate... - Value *SlotPtr = CreateGEP(AtEntry, StackEntry, 1 + I, "gc_root"); - + Value *SlotPtr = CreateGEP(Context, AtEntry, StackEntry, 1 + I, "gc_root"); + // And use it in lieu of the alloca. AllocaInst *OriginalAlloca = Roots[I].second; SlotPtr->takeName(OriginalAlloca); OriginalAlloca->replaceAllUsesWith(SlotPtr); } - + // Move past the original stores inserted by GCStrategy::InitRoots. This isn't // really necessary (the collector would never see the intermediate state at // runtime), but it's nicer not to push the half-initialized entry onto the // shadow stack. while (isa(IP)) ++IP; AtEntry.SetInsertPoint(IP->getParent(), IP); - + // Push the entry onto the shadow stack. - Instruction *EntryNextPtr = CreateGEP(AtEntry,StackEntry,0,0,"gc_frame.next"); - Instruction *NewHeadVal = CreateGEP(AtEntry,StackEntry, 0, "gc_newhead"); - AtEntry.CreateStore(CurrentHead, EntryNextPtr); - AtEntry.CreateStore(NewHeadVal, Head); - + Instruction *EntryNextPtr = CreateGEP(Context, AtEntry, + StackEntry,0,0,"gc_frame.next"); + Instruction *NewHeadVal = CreateGEP(Context, AtEntry, + StackEntry, 0, "gc_newhead"); + AtEntry.CreateStore(CurrentHead, EntryNextPtr); + AtEntry.CreateStore(NewHeadVal, Head); + // For each instruction that escapes... EscapeEnumerator EE(F, "gc_cleanup"); while (IRBuilder<> *AtExit = EE.Next()) { // Pop the entry from the shadow stack. Don't reuse CurrentHead from // AtEntry, since that would make the value live for the entire function. - Instruction *EntryNextPtr2 = CreateGEP(*AtExit, StackEntry, 0, 0, + Instruction *EntryNextPtr2 = CreateGEP(Context, *AtExit, StackEntry, 0, 0, "gc_frame.next"); Value *SavedHead = AtExit->CreateLoad(EntryNextPtr2, "gc_savedhead"); AtExit->CreateStore(SavedHead, Head); } - + // Delete the original allocas (which are no longer used) and the intrinsic // calls (which are no longer valid). Doing this last avoids invalidating // iterators. @@ -432,7 +445,7 @@ bool ShadowStackGC::performCustomLowering(Function &F) { Roots[I].first->eraseFromParent(); Roots[I].second->eraseFromParent(); } - + Roots.clear(); return true; }