void ProgramCFG::setFuncVariable(const Function *F,string func, CFG* cfg, bool initial){ for (Function::const_arg_iterator it = F->arg_begin(), E = F->arg_end();it != E; ++it) { Type *Ty = it->getType(); if(initial){ string varNum = it->getName(); string varName = func+"_"+varNum; if(Ty->isPointerTy()){ Type *ETy = Ty->getPointerElementType(); int ID = cfg->counter_variable++; Variable var(varName, ID, PTR); cfg->variableList.push_back(var); InstParser::setVariable(cfg, NULL, ETy, varName, true); } else{ VarType type; if(Ty->isIntegerTy()) type = INT; else if(Ty->isFloatingPointTy()) type = FP; else errs()<<"0:programCFG.type error\n"; int ID = cfg->counter_variable++; Variable var(varName, ID, type); cfg->variableList.push_back(var); cfg->mainInput.push_back(ID); } } else{ int ID = cfg->counter_variable++; string varNum = it->getName(); string varName = func+"_"+varNum; VarType type; if(Ty->isPointerTy()) type = PTR; else if(Ty->isIntegerTy()) type = INT; else if(Ty->isFloatingPointTy()) type = FP; else errs()<<"1:programCFG.type error\n"; if(!cfg->hasVariable(varName)){ Variable var(varName, ID, type); cfg->variableList.push_back(var); } else errs()<<"1:setFuncVariable error 10086!!\t"<<varName<<"\n"; } } }
/// CloneFunction - Return a copy of the specified function, but without /// embedding the function into another module. Also, any references specified /// in the VMap are changed to refer to their mapped value instead of the /// original one. If any of the arguments to the function are in the VMap, /// the arguments are deleted from the resultant function. The VMap is /// updated to include mappings from all of the instructions and basicblocks in /// the function from their old to new values. /// Function *llvm::CloneFunction(const Function *F, ValueToValueMapTy &VMap, bool ModuleLevelChanges, ClonedCodeInfo *CodeInfo) { std::vector<Type*> ArgTypes; // The user might be deleting arguments to the function by specifying them in // the VMap. If so, we need to not add the arguments to the arg ty vector // for (Function::const_arg_iterator I = F->arg_begin(), E = F->arg_end(); I != E; ++I) if (VMap.count(I) == 0) // Haven't mapped the argument to anything yet? ArgTypes.push_back(I->getType()); // Create a new function type... FunctionType *FTy = FunctionType::get(F->getFunctionType()->getReturnType(), ArgTypes, F->getFunctionType()->isVarArg()); // Create the new function... Function *NewF = Function::Create(FTy, F->getLinkage(), F->getName()); // Loop over the arguments, copying the names of the mapped arguments over... Function::arg_iterator DestI = NewF->arg_begin(); for (Function::const_arg_iterator I = F->arg_begin(), E = F->arg_end(); I != E; ++I) if (VMap.count(I) == 0) { // Is this argument preserved? DestI->setName(I->getName()); // Copy the name over... VMap[I] = DestI++; // Add mapping to VMap } SmallVector<ReturnInst*, 8> Returns; // Ignore returns cloned. CloneFunctionInto(NewF, F, VMap, ModuleLevelChanges, Returns, "", CodeInfo); return NewF; }
void llvm::copyFunctionBody(Function &New, const Function &Orig, ValueToValueMapTy &VMap) { if (!Orig.isDeclaration()) { Function::arg_iterator DestI = New.arg_begin(); for (Function::const_arg_iterator J = Orig.arg_begin(); J != Orig.arg_end(); ++J) { DestI->setName(J->getName()); VMap[J] = DestI++; } SmallVector<ReturnInst *, 8> Returns; // Ignore returns cloned. CloneFunctionInto(&New, &Orig, VMap, /*ModuleLevelChanges=*/true, Returns); } }
std::unique_ptr<Module> llvm::CloneModule( const Module *M, ValueToValueMapTy &VMap, std::function<bool(const GlobalValue *)> ShouldCloneDefinition) { // First off, we need to create the new module. std::unique_ptr<Module> New = llvm::make_unique<Module>(M->getModuleIdentifier(), M->getContext()); New->setDataLayout(M->getDataLayout()); New->setTargetTriple(M->getTargetTriple()); New->setModuleInlineAsm(M->getModuleInlineAsm()); // Loop over all of the global variables, making corresponding globals in the // new module. Here we add them to the VMap and to the new Module. We // don't worry about attributes or initializers, they will come later. // for (Module::const_global_iterator I = M->global_begin(), E = M->global_end(); I != E; ++I) { GlobalVariable *GV = new GlobalVariable(*New, I->getValueType(), I->isConstant(), I->getLinkage(), (Constant*) nullptr, I->getName(), (GlobalVariable*) nullptr, I->getThreadLocalMode(), I->getType()->getAddressSpace()); GV->copyAttributesFrom(&*I); VMap[&*I] = GV; } // Loop over the functions in the module, making external functions as before for (Module::const_iterator I = M->begin(), E = M->end(); I != E; ++I) { Function *NF = Function::Create(cast<FunctionType>(I->getValueType()), I->getLinkage(), I->getName(), New.get()); NF->copyAttributesFrom(&*I); VMap[&*I] = NF; } // Loop over the aliases in the module for (Module::const_alias_iterator I = M->alias_begin(), E = M->alias_end(); I != E; ++I) { if (!ShouldCloneDefinition(&*I)) { // An alias cannot act as an external reference, so we need to create // either a function or a global variable depending on the value type. // FIXME: Once pointee types are gone we can probably pick one or the // other. GlobalValue *GV; if (I->getValueType()->isFunctionTy()) GV = Function::Create(cast<FunctionType>(I->getValueType()), GlobalValue::ExternalLinkage, I->getName(), New.get()); else GV = new GlobalVariable( *New, I->getValueType(), false, GlobalValue::ExternalLinkage, (Constant *)nullptr, I->getName(), (GlobalVariable *)nullptr, I->getThreadLocalMode(), I->getType()->getAddressSpace()); VMap[&*I] = GV; // We do not copy attributes (mainly because copying between different // kinds of globals is forbidden), but this is generally not required for // correctness. continue; } auto *GA = GlobalAlias::create(I->getValueType(), I->getType()->getPointerAddressSpace(), I->getLinkage(), I->getName(), New.get()); GA->copyAttributesFrom(&*I); VMap[&*I] = GA; } // Now that all of the things that global variable initializer can refer to // have been created, loop through and copy the global variable referrers // over... We also set the attributes on the global now. // for (Module::const_global_iterator I = M->global_begin(), E = M->global_end(); I != E; ++I) { GlobalVariable *GV = cast<GlobalVariable>(VMap[&*I]); if (!ShouldCloneDefinition(&*I)) { // Skip after setting the correct linkage for an external reference. GV->setLinkage(GlobalValue::ExternalLinkage); continue; } if (I->hasInitializer()) GV->setInitializer(MapValue(I->getInitializer(), VMap)); } // Similarly, copy over function bodies now... // for (Module::const_iterator I = M->begin(), E = M->end(); I != E; ++I) { Function *F = cast<Function>(VMap[&*I]); if (!ShouldCloneDefinition(&*I)) { // Skip after setting the correct linkage for an external reference. F->setLinkage(GlobalValue::ExternalLinkage); // Personality function is not valid on a declaration. F->setPersonalityFn(nullptr); continue; } if (!I->isDeclaration()) { Function::arg_iterator DestI = F->arg_begin(); for (Function::const_arg_iterator J = I->arg_begin(); J != I->arg_end(); ++J) { DestI->setName(J->getName()); VMap[&*J] = &*DestI++; } SmallVector<ReturnInst*, 8> Returns; // Ignore returns cloned. CloneFunctionInto(F, &*I, VMap, /*ModuleLevelChanges=*/true, Returns); } if (I->hasPersonalityFn()) F->setPersonalityFn(MapValue(I->getPersonalityFn(), VMap)); } // And aliases for (Module::const_alias_iterator I = M->alias_begin(), E = M->alias_end(); I != E; ++I) { // We already dealt with undefined aliases above. if (!ShouldCloneDefinition(&*I)) continue; GlobalAlias *GA = cast<GlobalAlias>(VMap[&*I]); if (const Constant *C = I->getAliasee()) GA->setAliasee(MapValue(C, VMap)); } // And named metadata.... for (Module::const_named_metadata_iterator I = M->named_metadata_begin(), E = M->named_metadata_end(); I != E; ++I) { const NamedMDNode &NMD = *I; NamedMDNode *NewNMD = New->getOrInsertNamedMetadata(NMD.getName()); for (unsigned i = 0, e = NMD.getNumOperands(); i != e; ++i) NewNMD->addOperand(MapMetadata(NMD.getOperand(i), VMap)); } return New; }
/* ************************************************************************** */ bool RangedAddressSanitizer::doInitialization(Module &M) { // Link FastAddressSanitizer functions into the target module LLVMContext & context = M.getContext(); const char * fasanPath = getenv("FASANMODULE"); if (! fasanPath) { return false; } std::stringstream ss; ss << fasanPath; SMDiagnostic diag; Module * fasanModule = ParseIRFile(ss.str(), diag, context); if (!fasanModule) { abort(); } #if 0 /* using LLVM linking facilities */ Linker linker(&M); std::string linkErr; if (linker.linkInModule(fasanModule, Linker::DestroySource, &linkErr)) { errs() << "[FASAN] Error while linking runtime module: " << fasanModule << "(!!)\n"; abort(); } #else PointerType * voidPtrTy = PointerType::getInt8PtrTy(context, 0); IntegerType * boolTy = IntegerType::get(context, 1); Type * voidTy = Type::getVoidTy(context); FunctionType * touchFunType = FunctionType::get(voidTy, ArrayRef<Type*>(voidPtrTy), false); FunctionType * verifyFunType = FunctionType::get(boolTy, ArrayRef<Type*>(voidPtrTy), false); ValueToValueMapTy reMap; reMap[fasanModule->getFunction("__fasan_touch")] = M.getOrInsertFunction("__fasan_touch", touchFunType); reMap[fasanModule->getFunction("__fasan_verify")] = M.getOrInsertFunction("__fasan_verify", verifyFunType); // migrate check function { std::string errMsg; Function * checkFunc = fasanModule->getFunction("__fasan_check"); if (!checkFunc) { abort(); } #if 1 FunctionType * checkFuncType = checkFunc->getFunctionType(); Function * targetFunc = dyn_cast<Function>(M.getOrInsertFunction("__fasan_check", checkFuncType)); assert(targetFunc && "function cast to const by getOrInsertFunc..?"); // Loop over the arguments, copying the names of the mapped arguments over... Function::arg_iterator DestI = targetFunc->arg_begin(); for (Function::const_arg_iterator I = checkFunc->arg_begin(), E = checkFunc->arg_end(); I != E; ++I) if (reMap.count(I) == 0) { // Is this argument preserved? DestI->setName(I->getName()); // Copy the name over... reMap[I] = DestI++; // Add mapping to VMap } SmallVector<ReturnInst*, 8> Returns; // Ignore returns cloned. CloneFunctionInto(targetFunc, checkFunc, reMap, false, Returns, "", nullptr); targetFunc->addAttribute(0,Attribute::SanitizeAddress); #else Function * clonedCheckFunc = CloneFunction(checkFunc, reMap, false, 0); assert(!M.getFunction("__fasan_check") && "already exists in module"); M.getFunctionList().push_back(clonedCheckFunc); ReuseFn_ = clonedCheckFunc; clonedCheckFunc->setLinkage(GlobalValue::InternalLinkage); // avoid conflicts during linking // re-map fake use to local copy for (auto & BB : *clonedCheckFunc) { for (auto & Inst : BB) { RemapInstruction(&Inst, reMap, RF_IgnoreMissingEntries, 0, 0); } } #endif #endif } delete fasanModule; return true; }
Module *llvm::CloneModule(const Module *M, ValueToValueMapTy &VMap) { // First off, we need to create the new module. Module *New = new Module(M->getModuleIdentifier(), M->getContext()); New->setDataLayout(M->getDataLayout()); New->setTargetTriple(M->getTargetTriple()); New->setModuleInlineAsm(M->getModuleInlineAsm()); // Loop over all of the global variables, making corresponding globals in the // new module. Here we add them to the VMap and to the new Module. We // don't worry about attributes or initializers, they will come later. // for (Module::const_global_iterator I = M->global_begin(), E = M->global_end(); I != E; ++I) { GlobalVariable *GV = new GlobalVariable(*New, I->getType()->getElementType(), I->isConstant(), I->getLinkage(), (Constant*) nullptr, I->getName(), (GlobalVariable*) nullptr, I->getThreadLocalMode(), I->getType()->getAddressSpace()); GV->copyAttributesFrom(I); VMap[I] = GV; } // Loop over the functions in the module, making external functions as before for (Module::const_iterator I = M->begin(), E = M->end(); I != E; ++I) { Function *NF = Function::Create(cast<FunctionType>(I->getType()->getElementType()), I->getLinkage(), I->getName(), New); NF->copyAttributesFrom(I); VMap[I] = NF; } // Loop over the aliases in the module for (Module::const_alias_iterator I = M->alias_begin(), E = M->alias_end(); I != E; ++I) { auto *PTy = cast<PointerType>(I->getType()); auto *GA = GlobalAlias::create(PTy->getElementType(), PTy->getAddressSpace(), I->getLinkage(), I->getName(), New); GA->copyAttributesFrom(I); VMap[I] = GA; } // Now that all of the things that global variable initializer can refer to // have been created, loop through and copy the global variable referrers // over... We also set the attributes on the global now. // for (Module::const_global_iterator I = M->global_begin(), E = M->global_end(); I != E; ++I) { GlobalVariable *GV = cast<GlobalVariable>(VMap[I]); if (I->hasInitializer()) GV->setInitializer(MapValue(I->getInitializer(), VMap)); } // Similarly, copy over function bodies now... // for (Module::const_iterator I = M->begin(), E = M->end(); I != E; ++I) { Function *F = cast<Function>(VMap[I]); if (!I->isDeclaration()) { Function::arg_iterator DestI = F->arg_begin(); for (Function::const_arg_iterator J = I->arg_begin(); J != I->arg_end(); ++J) { DestI->setName(J->getName()); VMap[J] = DestI++; } SmallVector<ReturnInst*, 8> Returns; // Ignore returns cloned. CloneFunctionInto(F, I, VMap, /*ModuleLevelChanges=*/true, Returns); } } // And aliases for (Module::const_alias_iterator I = M->alias_begin(), E = M->alias_end(); I != E; ++I) { GlobalAlias *GA = cast<GlobalAlias>(VMap[I]); if (const Constant *C = I->getAliasee()) GA->setAliasee(cast<GlobalObject>(MapValue(C, VMap))); } // And named metadata.... for (Module::const_named_metadata_iterator I = M->named_metadata_begin(), E = M->named_metadata_end(); I != E; ++I) { const NamedMDNode &NMD = *I; NamedMDNode *NewNMD = New->getOrInsertNamedMetadata(NMD.getName()); for (unsigned i = 0, e = NMD.getNumOperands(); i != e; ++i) NewNMD->addOperand(MapValue(NMD.getOperand(i), VMap)); } return New; }
// InlineFunction - This function inlines the called function into the basic // block of the caller. This returns false if it is not possible to inline this // call. The program is still in a well defined state if this occurs though. // // Note that this only does one level of inlining. For example, if the // instruction 'call B' is inlined, and 'B' calls 'C', then the call to 'C' now // exists in the instruction stream. Similiarly this will inline a recursive // function by one level. // bool llvm::InlineFunction(CallSite CS, CallGraph *CG, const TargetData *TD) { Instruction *TheCall = CS.getInstruction(); assert(TheCall->getParent() && TheCall->getParent()->getParent() && "Instruction not in function!"); const Function *CalledFunc = CS.getCalledFunction(); if (CalledFunc == 0 || // Can't inline external function or indirect CalledFunc->isDeclaration() || // call, or call to a vararg function! CalledFunc->getFunctionType()->isVarArg()) return false; // If the call to the callee is not a tail call, we must clear the 'tail' // flags on any calls that we inline. bool MustClearTailCallFlags = !(isa<CallInst>(TheCall) && cast<CallInst>(TheCall)->isTailCall()); // If the call to the callee cannot throw, set the 'nounwind' flag on any // calls that we inline. bool MarkNoUnwind = CS.doesNotThrow(); BasicBlock *OrigBB = TheCall->getParent(); Function *Caller = OrigBB->getParent(); // GC poses two hazards to inlining, which only occur when the callee has GC: // 1. If the caller has no GC, then the callee's GC must be propagated to the // caller. // 2. If the caller has a differing GC, it is invalid to inline. if (CalledFunc->hasGC()) { if (!Caller->hasGC()) Caller->setGC(CalledFunc->getGC()); else if (CalledFunc->getGC() != Caller->getGC()) return false; } // Get an iterator to the last basic block in the function, which will have // the new function inlined after it. // Function::iterator LastBlock = &Caller->back(); // Make sure to capture all of the return instructions from the cloned // function. std::vector<ReturnInst*> Returns; ClonedCodeInfo InlinedFunctionInfo; Function::iterator FirstNewBlock; { // Scope to destroy ValueMap after cloning. DenseMap<const Value*, Value*> ValueMap; assert(CalledFunc->arg_size() == CS.arg_size() && "No varargs calls can be inlined!"); // Calculate the vector of arguments to pass into the function cloner, which // matches up the formal to the actual argument values. CallSite::arg_iterator AI = CS.arg_begin(); unsigned ArgNo = 0; for (Function::const_arg_iterator I = CalledFunc->arg_begin(), E = CalledFunc->arg_end(); I != E; ++I, ++AI, ++ArgNo) { Value *ActualArg = *AI; // When byval arguments actually inlined, we need to make the copy implied // by them explicit. However, we don't do this if the callee is readonly // or readnone, because the copy would be unneeded: the callee doesn't // modify the struct. if (CalledFunc->paramHasAttr(ArgNo+1, Attribute::ByVal) && !CalledFunc->onlyReadsMemory()) { const Type *AggTy = cast<PointerType>(I->getType())->getElementType(); const Type *VoidPtrTy = PointerType::getUnqual(Type::Int8Ty); // Create the alloca. If we have TargetData, use nice alignment. unsigned Align = 1; if (TD) Align = TD->getPrefTypeAlignment(AggTy); Value *NewAlloca = new AllocaInst(AggTy, 0, Align, I->getName(), Caller->begin()->begin()); // Emit a memcpy. const Type *Tys[] = { Type::Int64Ty }; Function *MemCpyFn = Intrinsic::getDeclaration(Caller->getParent(), Intrinsic::memcpy, Tys, 1); Value *DestCast = new BitCastInst(NewAlloca, VoidPtrTy, "tmp", TheCall); Value *SrcCast = new BitCastInst(*AI, VoidPtrTy, "tmp", TheCall); Value *Size; if (TD == 0) Size = ConstantExpr::getSizeOf(AggTy); else Size = ConstantInt::get(Type::Int64Ty, TD->getTypeStoreSize(AggTy)); // Always generate a memcpy of alignment 1 here because we don't know // the alignment of the src pointer. Other optimizations can infer // better alignment. Value *CallArgs[] = { DestCast, SrcCast, Size, ConstantInt::get(Type::Int32Ty, 1) }; CallInst *TheMemCpy = CallInst::Create(MemCpyFn, CallArgs, CallArgs+4, "", TheCall); // If we have a call graph, update it. if (CG) { CallGraphNode *MemCpyCGN = CG->getOrInsertFunction(MemCpyFn); CallGraphNode *CallerNode = (*CG)[Caller]; CallerNode->addCalledFunction(TheMemCpy, MemCpyCGN); } // Uses of the argument in the function should use our new alloca // instead. ActualArg = NewAlloca; } ValueMap[I] = ActualArg; } // We want the inliner to prune the code as it copies. We would LOVE to // have no dead or constant instructions leftover after inlining occurs // (which can happen, e.g., because an argument was constant), but we'll be // happy with whatever the cloner can do. CloneAndPruneFunctionInto(Caller, CalledFunc, ValueMap, Returns, ".i", &InlinedFunctionInfo, TD); // Remember the first block that is newly cloned over. FirstNewBlock = LastBlock; ++FirstNewBlock; // Update the callgraph if requested. if (CG) UpdateCallGraphAfterInlining(CS, FirstNewBlock, ValueMap, *CG); } // If there are any alloca instructions in the block that used to be the entry // block for the callee, move them to the entry block of the caller. First // calculate which instruction they should be inserted before. We insert the // instructions at the end of the current alloca list. // { BasicBlock::iterator InsertPoint = Caller->begin()->begin(); for (BasicBlock::iterator I = FirstNewBlock->begin(), E = FirstNewBlock->end(); I != E; ) if (AllocaInst *AI = dyn_cast<AllocaInst>(I++)) { // If the alloca is now dead, remove it. This often occurs due to code // specialization. if (AI->use_empty()) { AI->eraseFromParent(); continue; } if (isa<Constant>(AI->getArraySize())) { // Scan for the block of allocas that we can move over, and move them // all at once. while (isa<AllocaInst>(I) && isa<Constant>(cast<AllocaInst>(I)->getArraySize())) ++I; // Transfer all of the allocas over in a block. Using splice means // that the instructions aren't removed from the symbol table, then // reinserted. Caller->getEntryBlock().getInstList().splice( InsertPoint, FirstNewBlock->getInstList(), AI, I); } } } // If the inlined code contained dynamic alloca instructions, wrap the inlined // code with llvm.stacksave/llvm.stackrestore intrinsics. if (InlinedFunctionInfo.ContainsDynamicAllocas) { Module *M = Caller->getParent(); // Get the two intrinsics we care about. Constant *StackSave, *StackRestore; StackSave = Intrinsic::getDeclaration(M, Intrinsic::stacksave); StackRestore = Intrinsic::getDeclaration(M, Intrinsic::stackrestore); // If we are preserving the callgraph, add edges to the stacksave/restore // functions for the calls we insert. CallGraphNode *StackSaveCGN = 0, *StackRestoreCGN = 0, *CallerNode = 0; if (CG) { // We know that StackSave/StackRestore are Function*'s, because they are // intrinsics which must have the right types. StackSaveCGN = CG->getOrInsertFunction(cast<Function>(StackSave)); StackRestoreCGN = CG->getOrInsertFunction(cast<Function>(StackRestore)); CallerNode = (*CG)[Caller]; } // Insert the llvm.stacksave. CallInst *SavedPtr = CallInst::Create(StackSave, "savedstack", FirstNewBlock->begin()); if (CG) CallerNode->addCalledFunction(SavedPtr, StackSaveCGN); // Insert a call to llvm.stackrestore before any return instructions in the // inlined function. for (unsigned i = 0, e = Returns.size(); i != e; ++i) { CallInst *CI = CallInst::Create(StackRestore, SavedPtr, "", Returns[i]); if (CG) CallerNode->addCalledFunction(CI, StackRestoreCGN); } // Count the number of StackRestore calls we insert. unsigned NumStackRestores = Returns.size(); // If we are inlining an invoke instruction, insert restores before each // unwind. These unwinds will be rewritten into branches later. if (InlinedFunctionInfo.ContainsUnwinds && isa<InvokeInst>(TheCall)) { for (Function::iterator BB = FirstNewBlock, E = Caller->end(); BB != E; ++BB) if (UnwindInst *UI = dyn_cast<UnwindInst>(BB->getTerminator())) { CallInst::Create(StackRestore, SavedPtr, "", UI); ++NumStackRestores; } } } // If we are inlining tail call instruction through a call site that isn't // marked 'tail', we must remove the tail marker for any calls in the inlined // code. Also, calls inlined through a 'nounwind' call site should be marked // 'nounwind'. if (InlinedFunctionInfo.ContainsCalls && (MustClearTailCallFlags || MarkNoUnwind)) { for (Function::iterator BB = FirstNewBlock, E = Caller->end(); BB != E; ++BB) for (BasicBlock::iterator I = BB->begin(), E = BB->end(); I != E; ++I) if (CallInst *CI = dyn_cast<CallInst>(I)) { if (MustClearTailCallFlags) CI->setTailCall(false); if (MarkNoUnwind) CI->setDoesNotThrow(); } } // If we are inlining through a 'nounwind' call site then any inlined 'unwind' // instructions are unreachable. if (InlinedFunctionInfo.ContainsUnwinds && MarkNoUnwind) for (Function::iterator BB = FirstNewBlock, E = Caller->end(); BB != E; ++BB) { TerminatorInst *Term = BB->getTerminator(); if (isa<UnwindInst>(Term)) { new UnreachableInst(Term); BB->getInstList().erase(Term); } } // If we are inlining for an invoke instruction, we must make sure to rewrite // any inlined 'unwind' instructions into branches to the invoke exception // destination, and call instructions into invoke instructions. if (InvokeInst *II = dyn_cast<InvokeInst>(TheCall)) HandleInlinedInvoke(II, FirstNewBlock, InlinedFunctionInfo); // If we cloned in _exactly one_ basic block, and if that block ends in a // return instruction, we splice the body of the inlined callee directly into // the calling basic block. if (Returns.size() == 1 && std::distance(FirstNewBlock, Caller->end()) == 1) { // Move all of the instructions right before the call. OrigBB->getInstList().splice(TheCall, FirstNewBlock->getInstList(), FirstNewBlock->begin(), FirstNewBlock->end()); // Remove the cloned basic block. Caller->getBasicBlockList().pop_back(); // If the call site was an invoke instruction, add a branch to the normal // destination. if (InvokeInst *II = dyn_cast<InvokeInst>(TheCall)) BranchInst::Create(II->getNormalDest(), TheCall); // If the return instruction returned a value, replace uses of the call with // uses of the returned value. if (!TheCall->use_empty()) { ReturnInst *R = Returns[0]; TheCall->replaceAllUsesWith(R->getReturnValue()); } // Since we are now done with the Call/Invoke, we can delete it. TheCall->eraseFromParent(); // Since we are now done with the return instruction, delete it also. Returns[0]->eraseFromParent(); // We are now done with the inlining. return true; } // Otherwise, we have the normal case, of more than one block to inline or // multiple return sites. // We want to clone the entire callee function into the hole between the // "starter" and "ender" blocks. How we accomplish this depends on whether // this is an invoke instruction or a call instruction. BasicBlock *AfterCallBB; if (InvokeInst *II = dyn_cast<InvokeInst>(TheCall)) { // Add an unconditional branch to make this look like the CallInst case... BranchInst *NewBr = BranchInst::Create(II->getNormalDest(), TheCall); // Split the basic block. This guarantees that no PHI nodes will have to be // updated due to new incoming edges, and make the invoke case more // symmetric to the call case. AfterCallBB = OrigBB->splitBasicBlock(NewBr, CalledFunc->getName()+".exit"); } else { // It's a call // If this is a call instruction, we need to split the basic block that // the call lives in. // AfterCallBB = OrigBB->splitBasicBlock(TheCall, CalledFunc->getName()+".exit"); } // Change the branch that used to go to AfterCallBB to branch to the first // basic block of the inlined function. // TerminatorInst *Br = OrigBB->getTerminator(); assert(Br && Br->getOpcode() == Instruction::Br && "splitBasicBlock broken!"); Br->setOperand(0, FirstNewBlock); // Now that the function is correct, make it a little bit nicer. In // particular, move the basic blocks inserted from the end of the function // into the space made by splitting the source basic block. Caller->getBasicBlockList().splice(AfterCallBB, Caller->getBasicBlockList(), FirstNewBlock, Caller->end()); // Handle all of the return instructions that we just cloned in, and eliminate // any users of the original call/invoke instruction. const Type *RTy = CalledFunc->getReturnType(); if (Returns.size() > 1) { // The PHI node should go at the front of the new basic block to merge all // possible incoming values. PHINode *PHI = 0; if (!TheCall->use_empty()) { PHI = PHINode::Create(RTy, TheCall->getName(), AfterCallBB->begin()); // Anything that used the result of the function call should now use the // PHI node as their operand. TheCall->replaceAllUsesWith(PHI); } // Loop over all of the return instructions adding entries to the PHI node // as appropriate. if (PHI) { for (unsigned i = 0, e = Returns.size(); i != e; ++i) { ReturnInst *RI = Returns[i]; assert(RI->getReturnValue()->getType() == PHI->getType() && "Ret value not consistent in function!"); PHI->addIncoming(RI->getReturnValue(), RI->getParent()); } } // Add a branch to the merge points and remove return instructions. for (unsigned i = 0, e = Returns.size(); i != e; ++i) { ReturnInst *RI = Returns[i]; BranchInst::Create(AfterCallBB, RI); RI->eraseFromParent(); } } else if (!Returns.empty()) { // Otherwise, if there is exactly one return value, just replace anything // using the return value of the call with the computed value. if (!TheCall->use_empty()) TheCall->replaceAllUsesWith(Returns[0]->getReturnValue()); // Splice the code from the return block into the block that it will return // to, which contains the code that was after the call. BasicBlock *ReturnBB = Returns[0]->getParent(); AfterCallBB->getInstList().splice(AfterCallBB->begin(), ReturnBB->getInstList()); // Update PHI nodes that use the ReturnBB to use the AfterCallBB. ReturnBB->replaceAllUsesWith(AfterCallBB); // Delete the return instruction now and empty ReturnBB now. Returns[0]->eraseFromParent(); ReturnBB->eraseFromParent(); } else if (!TheCall->use_empty()) { // No returns, but something is using the return value of the call. Just // nuke the result. TheCall->replaceAllUsesWith(UndefValue::get(TheCall->getType())); } // Since we are now done with the Call/Invoke, we can delete it. TheCall->eraseFromParent(); // We should always be able to fold the entry block of the function into the // single predecessor of the block... assert(cast<BranchInst>(Br)->isUnconditional() && "splitBasicBlock broken!"); BasicBlock *CalleeEntry = cast<BranchInst>(Br)->getSuccessor(0); // Splice the code entry block into calling block, right before the // unconditional branch. OrigBB->getInstList().splice(Br, CalleeEntry->getInstList()); CalleeEntry->replaceAllUsesWith(OrigBB); // Update PHI nodes // Remove the unconditional branch. OrigBB->getInstList().erase(Br); // Now we can remove the CalleeEntry block, which is now empty. Caller->getBasicBlockList().erase(CalleeEntry); return true; }
Module *llvm::CloneModule(const Module *M, DenseMap<const Value*, Value*> &ValueMap) { // First off, we need to create the new module... Module *New = new Module(M->getModuleIdentifier()); New->setDataLayout(M->getDataLayout()); New->setTargetTriple(M->getTargetTriple()); New->setModuleInlineAsm(M->getModuleInlineAsm()); // Copy all of the type symbol table entries over. const TypeSymbolTable &TST = M->getTypeSymbolTable(); for (TypeSymbolTable::const_iterator TI = TST.begin(), TE = TST.end(); TI != TE; ++TI) New->addTypeName(TI->first, TI->second); // Copy all of the dependent libraries over. for (Module::lib_iterator I = M->lib_begin(), E = M->lib_end(); I != E; ++I) New->addLibrary(*I); // Loop over all of the global variables, making corresponding globals in the // new module. Here we add them to the ValueMap and to the new Module. We // don't worry about attributes or initializers, they will come later. // for (Module::const_global_iterator I = M->global_begin(), E = M->global_end(); I != E; ++I) { GlobalVariable *GV = new GlobalVariable(I->getType()->getElementType(), false, GlobalValue::ExternalLinkage, 0, I->getName(), New); GV->setAlignment(I->getAlignment()); ValueMap[I] = GV; } // Loop over the functions in the module, making external functions as before for (Module::const_iterator I = M->begin(), E = M->end(); I != E; ++I) { Function *NF = Function::Create(cast<FunctionType>(I->getType()->getElementType()), GlobalValue::ExternalLinkage, I->getName(), New); NF->copyAttributesFrom(I); ValueMap[I] = NF; } // Loop over the aliases in the module for (Module::const_alias_iterator I = M->alias_begin(), E = M->alias_end(); I != E; ++I) ValueMap[I] = new GlobalAlias(I->getType(), GlobalAlias::ExternalLinkage, I->getName(), NULL, New); // Now that all of the things that global variable initializer can refer to // have been created, loop through and copy the global variable referrers // over... We also set the attributes on the global now. // for (Module::const_global_iterator I = M->global_begin(), E = M->global_end(); I != E; ++I) { GlobalVariable *GV = cast<GlobalVariable>(ValueMap[I]); if (I->hasInitializer()) GV->setInitializer(cast<Constant>(MapValue(I->getInitializer(), ValueMap))); GV->setLinkage(I->getLinkage()); GV->setThreadLocal(I->isThreadLocal()); GV->setConstant(I->isConstant()); } // Similarly, copy over function bodies now... // for (Module::const_iterator I = M->begin(), E = M->end(); I != E; ++I) { Function *F = cast<Function>(ValueMap[I]); if (!I->isDeclaration()) { Function::arg_iterator DestI = F->arg_begin(); for (Function::const_arg_iterator J = I->arg_begin(); J != I->arg_end(); ++J) { DestI->setName(J->getName()); ValueMap[J] = DestI++; } std::vector<ReturnInst*> Returns; // Ignore returns cloned... CloneFunctionInto(F, I, ValueMap, Returns); } F->setLinkage(I->getLinkage()); } // And aliases for (Module::const_alias_iterator I = M->alias_begin(), E = M->alias_end(); I != E; ++I) { GlobalAlias *GA = cast<GlobalAlias>(ValueMap[I]); GA->setLinkage(I->getLinkage()); if (const Constant* C = I->getAliasee()) GA->setAliasee(cast<Constant>(MapValue(C, ValueMap))); } return New; }
Module *llvm::CloneModule(const Module *M, ValueToValueMapTy &VMap) { // First off, we need to create the new module... Module *New = new Module(M->getModuleIdentifier(), M->getContext()); New->setDataLayout(M->getDataLayout()); New->setTargetTriple(M->getTargetTriple()); New->setModuleInlineAsm(M->getModuleInlineAsm()); // Copy all of the type symbol table entries over. const TypeSymbolTable &TST = M->getTypeSymbolTable(); for (TypeSymbolTable::const_iterator TI = TST.begin(), TE = TST.end(); TI != TE; ++TI) New->addTypeName(TI->first, TI->second); // Copy all of the dependent libraries over. for (Module::lib_iterator I = M->lib_begin(), E = M->lib_end(); I != E; ++I) New->addLibrary(*I); // Loop over all of the global variables, making corresponding globals in the // new module. Here we add them to the VMap and to the new Module. We // don't worry about attributes or initializers, they will come later. // for (Module::const_global_iterator I = M->global_begin(), E = M->global_end(); I != E; ++I) { GlobalVariable *GV = new GlobalVariable(*New, I->getType()->getElementType(), false, GlobalValue::ExternalLinkage, 0, I->getName()); GV->setAlignment(I->getAlignment()); VMap[I] = GV; } // Loop over the functions in the module, making external functions as before for (Module::const_iterator I = M->begin(), E = M->end(); I != E; ++I) { Function *NF = Function::Create(cast<FunctionType>(I->getType()->getElementType()), GlobalValue::ExternalLinkage, I->getName(), New); NF->copyAttributesFrom(I); VMap[I] = NF; } // Loop over the aliases in the module for (Module::const_alias_iterator I = M->alias_begin(), E = M->alias_end(); I != E; ++I) VMap[I] = new GlobalAlias(I->getType(), GlobalAlias::ExternalLinkage, I->getName(), NULL, New); // Now that all of the things that global variable initializer can refer to // have been created, loop through and copy the global variable referrers // over... We also set the attributes on the global now. // for (Module::const_global_iterator I = M->global_begin(), E = M->global_end(); I != E; ++I) { GlobalVariable *GV = cast<GlobalVariable>(VMap[I]); if (I->hasInitializer()) GV->setInitializer(cast<Constant>(MapValue(I->getInitializer(), VMap))); GV->setLinkage(I->getLinkage()); GV->setThreadLocal(I->isThreadLocal()); GV->setConstant(I->isConstant()); } // Similarly, copy over function bodies now... // for (Module::const_iterator I = M->begin(), E = M->end(); I != E; ++I) { Function *F = cast<Function>(VMap[I]); if (!I->isDeclaration()) { Function::arg_iterator DestI = F->arg_begin(); for (Function::const_arg_iterator J = I->arg_begin(); J != I->arg_end(); ++J) { DestI->setName(J->getName()); VMap[J] = DestI++; } SmallVector<ReturnInst*, 8> Returns; // Ignore returns cloned. CloneFunctionInto(F, I, VMap, Returns); } F->setLinkage(I->getLinkage()); } // And aliases for (Module::const_alias_iterator I = M->alias_begin(), E = M->alias_end(); I != E; ++I) { GlobalAlias *GA = cast<GlobalAlias>(VMap[I]); GA->setLinkage(I->getLinkage()); if (const Constant* C = I->getAliasee()) GA->setAliasee(cast<Constant>(MapValue(C, VMap))); } // And named metadata.... for (Module::const_named_metadata_iterator I = M->named_metadata_begin(), E = M->named_metadata_end(); I != E; ++I) { const NamedMDNode &NMD = *I; SmallVector<MDNode*, 4> MDs; for (unsigned i = 0, e = NMD.getNumOperands(); i != e; ++i) MDs.push_back(cast<MDNode>(MapValue(NMD.getOperand(i), VMap))); NamedMDNode::Create(New->getContext(), NMD.getName(), MDs.data(), MDs.size(), New); } // Update metadata attach with instructions. for (Module::iterator MI = New->begin(), ME = New->end(); MI != ME; ++MI) for (Function::iterator FI = MI->begin(), FE = MI->end(); FI != FE; ++FI) for (BasicBlock::iterator BI = FI->begin(), BE = FI->end(); BI != BE; ++BI) { SmallVector<std::pair<unsigned, MDNode *>, 4 > MDs; BI->getAllMetadata(MDs); for (SmallVector<std::pair<unsigned, MDNode *>, 4>::iterator MDI = MDs.begin(), MDE = MDs.end(); MDI != MDE; ++MDI) { Value *MappedValue = MapValue(MDI->second, VMap); if (MDI->second != MappedValue && MappedValue) BI->setMetadata(MDI->first, cast<MDNode>(MappedValue)); } } return New; }