#include "llvm/Transforms/Instrumentation.h"
#include "llvm/ADT/ArrayRef.h"
#include "llvm/ADT/DenseMap.h"
+#include "llvm/ADT/DenseSet.h"
#include "llvm/ADT/DepthFirstIterator.h"
#include "llvm/ADT/SmallSet.h"
#include "llvm/ADT/SmallString.h"
#include "llvm/Support/DataTypes.h"
#include "llvm/Support/Debug.h"
#include "llvm/Support/Endian.h"
-#include "llvm/Support/system_error.h"
#include "llvm/Transforms/Utils/ASanStackFrameLayout.h"
#include "llvm/Transforms/Utils/BasicBlockUtils.h"
#include "llvm/Transforms/Utils/Cloning.h"
#include "llvm/Transforms/Utils/Local.h"
#include "llvm/Transforms/Utils/ModuleUtils.h"
-#include "llvm/Transforms/Utils/SpecialCaseList.h"
#include <algorithm>
#include <string>
+#include <system_error>
using namespace llvm;
"__asan_unregister_globals";
static const char *const kAsanPoisonGlobalsName = "__asan_before_dynamic_init";
static const char *const kAsanUnpoisonGlobalsName = "__asan_after_dynamic_init";
-static const char *const kAsanInitName = "__asan_init_v3";
+static const char *const kAsanInitName = "__asan_init_v4";
static const char *const kAsanCovModuleInitName = "__sanitizer_cov_module_init";
static const char *const kAsanCovName = "__sanitizer_cov";
static const char *const kAsanPtrCmp = "__sanitizer_ptr_cmp";
// This flag may need to be replaced with -f[no]asan-stack.
static cl::opt<bool> ClStack("asan-stack",
cl::desc("Handle stack memory"), cl::Hidden, cl::init(true));
-// This flag may need to be replaced with -f[no]asan-use-after-return.
static cl::opt<bool> ClUseAfterReturn("asan-use-after-return",
- cl::desc("Check return-after-free"), cl::Hidden, cl::init(false));
+ cl::desc("Check return-after-free"), cl::Hidden, cl::init(true));
// This flag may need to be replaced with -f[no]asan-globals.
static cl::opt<bool> ClGlobals("asan-globals",
cl::desc("Handle global objects"), cl::Hidden, cl::init(true));
"are more than this number of blocks."),
cl::Hidden, cl::init(1500));
static cl::opt<bool> ClInitializers("asan-initialization-order",
- cl::desc("Handle C++ initializer order"), cl::Hidden, cl::init(false));
+ cl::desc("Handle C++ initializer order"), cl::Hidden, cl::init(true));
static cl::opt<bool> ClInvalidPointerPairs("asan-detect-invalid-pointer-pair",
cl::desc("Instrument <, <=, >, >=, - with pointer operands"),
cl::Hidden, cl::init(false));
static cl::opt<unsigned> ClRealignStack("asan-realign-stack",
cl::desc("Realign stack to the value of this flag (power of two)"),
cl::Hidden, cl::init(32));
-static cl::opt<std::string> ClBlacklistFile("asan-blacklist",
- cl::desc("File containing the list of objects to ignore "
- "during instrumentation"), cl::Hidden);
static cl::opt<int> ClInstrumentationWithCallsThreshold(
"asan-instrumentation-with-call-threshold",
cl::desc("If the function being instrumented contains more than "
"Number of optimized accesses to global vars");
namespace {
-/// A set of dynamically initialized globals extracted from metadata.
-class SetOfDynamicallyInitializedGlobals {
+/// Frontend-provided metadata for source location.
+struct LocationMetadata {
+ StringRef Filename;
+ int LineNo;
+ int ColumnNo;
+
+ LocationMetadata() : Filename(), LineNo(0), ColumnNo(0) {}
+
+ bool empty() const { return Filename.empty(); }
+
+ void parse(MDNode *MDN) {
+ assert(MDN->getNumOperands() == 3);
+ MDString *MDFilename = cast<MDString>(MDN->getOperand(0));
+ Filename = MDFilename->getString();
+ LineNo = cast<ConstantInt>(MDN->getOperand(1))->getLimitedValue();
+ ColumnNo = cast<ConstantInt>(MDN->getOperand(2))->getLimitedValue();
+ }
+};
+
+/// Frontend-provided metadata for global variables.
+class GlobalsMetadata {
public:
- void Init(Module& M) {
- // Clang generates metadata identifying all dynamically initialized globals.
- NamedMDNode *DynamicGlobals =
- M.getNamedMetadata("llvm.asan.dynamically_initialized_globals");
- if (!DynamicGlobals)
+ struct Entry {
+ Entry()
+ : SourceLoc(), Name(), IsDynInit(false),
+ IsBlacklisted(false) {}
+ LocationMetadata SourceLoc;
+ StringRef Name;
+ bool IsDynInit;
+ bool IsBlacklisted;
+ };
+
+ GlobalsMetadata() : inited_(false) {}
+
+ void init(Module& M) {
+ assert(!inited_);
+ inited_ = true;
+ NamedMDNode *Globals = M.getNamedMetadata("llvm.asan.globals");
+ if (!Globals)
return;
- for (const auto MDN : DynamicGlobals->operands()) {
- assert(MDN->getNumOperands() == 1);
- Value *VG = MDN->getOperand(0);
- // The optimizer may optimize away a global entirely, in which case we
- // cannot instrument access to it.
- if (!VG)
+ for (auto MDN : Globals->operands()) {
+ // Metadata node contains the global and the fields of "Entry".
+ assert(MDN->getNumOperands() == 5);
+ Value *V = MDN->getOperand(0);
+ // The optimizer may optimize away a global entirely.
+ if (!V)
continue;
- DynInitGlobals.insert(cast<GlobalVariable>(VG));
+ GlobalVariable *GV = cast<GlobalVariable>(V);
+ // We can already have an entry for GV if it was merged with another
+ // global.
+ Entry &E = Entries[GV];
+ if (Value *Loc = MDN->getOperand(1))
+ E.SourceLoc.parse(cast<MDNode>(Loc));
+ if (Value *Name = MDN->getOperand(2)) {
+ MDString *MDName = cast<MDString>(Name);
+ E.Name = MDName->getString();
+ }
+ ConstantInt *IsDynInit = cast<ConstantInt>(MDN->getOperand(3));
+ E.IsDynInit |= IsDynInit->isOne();
+ ConstantInt *IsBlacklisted = cast<ConstantInt>(MDN->getOperand(4));
+ E.IsBlacklisted |= IsBlacklisted->isOne();
}
}
- bool Contains(GlobalVariable *G) { return DynInitGlobals.count(G) != 0; }
+
+ /// Returns metadata entry for a given global.
+ Entry get(GlobalVariable *G) const {
+ auto Pos = Entries.find(G);
+ return (Pos != Entries.end()) ? Pos->second : Entry();
+ }
+
private:
- SmallSet<GlobalValue*, 32> DynInitGlobals;
+ bool inited_;
+ DenseMap<GlobalVariable*, Entry> Entries;
};
/// This struct defines the shadow mapping using the rule:
/// AddressSanitizer: instrument the code in module to find memory bugs.
struct AddressSanitizer : public FunctionPass {
- AddressSanitizer(bool CheckInitOrder = true,
- bool CheckUseAfterReturn = false,
- bool CheckLifetime = false,
- StringRef BlacklistFile = StringRef())
- : FunctionPass(ID),
- CheckInitOrder(CheckInitOrder || ClInitializers),
- CheckUseAfterReturn(CheckUseAfterReturn || ClUseAfterReturn),
- CheckLifetime(CheckLifetime || ClCheckLifetime),
- BlacklistFile(BlacklistFile.empty() ? ClBlacklistFile
- : BlacklistFile) {}
+ AddressSanitizer() : FunctionPass(ID) {}
const char *getPassName() const override {
return "AddressSanitizerFunctionPass";
}
bool InjectCoverage(Function &F, const ArrayRef<BasicBlock*> AllBlocks);
void InjectCoverageAtBlock(Function &F, BasicBlock &BB);
- bool CheckInitOrder;
- bool CheckUseAfterReturn;
- bool CheckLifetime;
- SmallString<64> BlacklistFile;
-
LLVMContext *C;
const DataLayout *DL;
int LongSize;
Function *AsanHandleNoReturnFunc;
Function *AsanCovFunction;
Function *AsanPtrCmpFunction, *AsanPtrSubFunction;
- std::unique_ptr<SpecialCaseList> BL;
// This array is indexed by AccessIsWrite and log2(AccessSize).
Function *AsanErrorCallback[2][kNumberOfAccessSizes];
Function *AsanMemoryAccessCallback[2][kNumberOfAccessSizes];
*AsanMemoryAccessCallbackSized[2];
Function *AsanMemmove, *AsanMemcpy, *AsanMemset;
InlineAsm *EmptyAsm;
- SetOfDynamicallyInitializedGlobals DynamicallyInitializedGlobals;
+ GlobalsMetadata GlobalsMD;
friend struct FunctionStackPoisoner;
};
class AddressSanitizerModule : public ModulePass {
public:
- AddressSanitizerModule(bool CheckInitOrder = true,
- StringRef BlacklistFile = StringRef())
- : ModulePass(ID),
- CheckInitOrder(CheckInitOrder || ClInitializers),
- BlacklistFile(BlacklistFile.empty() ? ClBlacklistFile
- : BlacklistFile) {}
+ AddressSanitizerModule() : ModulePass(ID) {}
bool runOnModule(Module &M) override;
static char ID; // Pass identification, replacement for typeid
const char *getPassName() const override {
private:
void initializeCallbacks(Module &M);
+ bool InstrumentGlobals(IRBuilder<> &IRB, Module &M);
bool ShouldInstrumentGlobal(GlobalVariable *G);
void poisonOneInitializer(Function &GlobalInit, GlobalValue *ModuleName);
void createInitializerPoisonCalls(Module &M, GlobalValue *ModuleName);
return RedzoneSizeForScale(Mapping.Scale);
}
- bool CheckInitOrder;
- SmallString<64> BlacklistFile;
-
- std::unique_ptr<SpecialCaseList> BL;
- SetOfDynamicallyInitializedGlobals DynamicallyInitializedGlobals;
+ GlobalsMetadata GlobalsMD;
Type *IntptrTy;
LLVMContext *C;
const DataLayout *DL;
/// \brief Collect lifetime intrinsic calls to check for use-after-scope
/// errors.
void visitIntrinsicInst(IntrinsicInst &II) {
- if (!ASan.CheckLifetime) return;
+ if (!ClCheckLifetime) return;
Intrinsic::ID ID = II.getIntrinsicID();
if (ID != Intrinsic::lifetime_start &&
ID != Intrinsic::lifetime_end)
INITIALIZE_PASS(AddressSanitizer, "asan",
"AddressSanitizer: detects use-after-free and out-of-bounds bugs.",
false, false)
-FunctionPass *llvm::createAddressSanitizerFunctionPass(
- bool CheckInitOrder, bool CheckUseAfterReturn, bool CheckLifetime,
- StringRef BlacklistFile) {
- return new AddressSanitizer(CheckInitOrder, CheckUseAfterReturn,
- CheckLifetime, BlacklistFile);
+FunctionPass *llvm::createAddressSanitizerFunctionPass() {
+ return new AddressSanitizer();
}
char AddressSanitizerModule::ID = 0;
INITIALIZE_PASS(AddressSanitizerModule, "asan-module",
"AddressSanitizer: detects use-after-free and out-of-bounds bugs."
"ModulePass", false, false)
-ModulePass *llvm::createAddressSanitizerModulePass(
- bool CheckInitOrder, StringRef BlacklistFile) {
- return new AddressSanitizerModule(CheckInitOrder, BlacklistFile);
+ModulePass *llvm::createAddressSanitizerModulePass() {
+ return new AddressSanitizerModule();
}
static size_t TypeSizeToSizeIndex(uint32_t TypeSize) {
return GV;
}
+/// \brief Create a global describing a source location.
+static GlobalVariable *createPrivateGlobalForSourceLoc(Module &M,
+ LocationMetadata MD) {
+ Constant *LocData[] = {
+ createPrivateGlobalForString(M, MD.Filename, true),
+ ConstantInt::get(Type::getInt32Ty(M.getContext()), MD.LineNo),
+ ConstantInt::get(Type::getInt32Ty(M.getContext()), MD.ColumnNo),
+ };
+ auto LocStruct = ConstantStruct::getAnon(LocData);
+ auto GV = new GlobalVariable(M, LocStruct->getType(), true,
+ GlobalValue::PrivateLinkage, LocStruct,
+ kAsanGenPrefix);
+ GV->setUnnamedAddr(true);
+ return GV;
+}
+
static bool GlobalWasGeneratedByAsan(GlobalVariable *G) {
return G->getName().find(kAsanGenPrefix) == 0;
}
// and set IsWrite/Alignment. Otherwise return NULL.
static Value *isInterestingMemoryAccess(Instruction *I, bool *IsWrite,
unsigned *Alignment) {
+ // Skip memory accesses inserted by another instrumentation.
+ if (I->getMetadata("nosanitize"))
+ return nullptr;
if (LoadInst *LI = dyn_cast<LoadInst>(I)) {
if (!ClInstrumentReads) return nullptr;
*IsWrite = false;
// If a global variable does not have dynamic initialization we don't
// have to instrument it. However, if a global does not have initializer
// at all, we assume it has dynamic initializer (in other TU).
- return G->hasInitializer() && !DynamicallyInitializedGlobals.Contains(G);
+ return G->hasInitializer() && !GlobalsMD.get(G).IsDynInit;
}
void
if (GlobalVariable *G = dyn_cast<GlobalVariable>(Addr)) {
// If initialization order checking is disabled, a simple access to a
// dynamically initialized global is always valid.
- if (!CheckInitOrder || GlobalIsLinkerInitialized(G)) {
+ if (!ClInitializers || GlobalIsLinkerInitialized(G)) {
NumOptimizedAccessesToGlobalVar++;
return;
}
Type *Ty = cast<PointerType>(G->getType())->getElementType();
DEBUG(dbgs() << "GLOBAL: " << *G << "\n");
- if (BL->isIn(*G)) return false;
+ if (GlobalsMD.get(G).IsBlacklisted) return false;
if (!Ty->isSized()) return false;
if (!G->hasInitializer()) return false;
if (GlobalWasGeneratedByAsan(G)) return false; // Our own global.
// Touch only those globals that will not be defined in other modules.
- // Don't handle ODR type linkages since other modules may be built w/o asan.
+ // Don't handle ODR linkage types and COMDATs since other modules may be built
+ // without ASan.
if (G->getLinkage() != GlobalVariable::ExternalLinkage &&
G->getLinkage() != GlobalVariable::PrivateLinkage &&
G->getLinkage() != GlobalVariable::InternalLinkage)
return false;
+ if (G->hasComdat())
+ return false;
// Two problems with thread-locals:
// - The address of the main thread's copy can't be computed at link-time.
// - Need to poison all copies, not just the main thread's one.
// This function replaces all global variables with new variables that have
// trailing redzones. It also creates a function that poisons
// redzones and inserts this function into llvm.global_ctors.
-bool AddressSanitizerModule::runOnModule(Module &M) {
- if (!ClGlobals) return false;
-
- DataLayoutPass *DLP = getAnalysisIfAvailable<DataLayoutPass>();
- if (!DLP)
- return false;
- DL = &DLP->getDataLayout();
-
- BL.reset(SpecialCaseList::createOrDie(BlacklistFile));
- if (BL->isIn(M)) return false;
- C = &(M.getContext());
- int LongSize = DL->getPointerSizeInBits();
- IntptrTy = Type::getIntNTy(*C, LongSize);
- Mapping = getShadowMapping(M, LongSize);
- initializeCallbacks(M);
- DynamicallyInitializedGlobals.Init(M);
+bool AddressSanitizerModule::InstrumentGlobals(IRBuilder<> &IRB, Module &M) {
+ GlobalsMD.init(M);
SmallVector<GlobalVariable *, 16> GlobalsToChange;
GlobalsToChange.push_back(&G);
}
- Function *CtorFunc = M.getFunction(kAsanModuleCtorName);
- assert(CtorFunc);
- IRBuilder<> IRB(CtorFunc->getEntryBlock().getTerminator());
-
- if (ClCoverage > 0) {
- Function *CovFunc = M.getFunction(kAsanCovName);
- int nCov = CovFunc ? CovFunc->getNumUses() : 0;
- IRB.CreateCall(AsanCovModuleInit, ConstantInt::get(IntptrTy, nCov));
- }
-
size_t n = GlobalsToChange.size();
if (n == 0) return false;
// const char *name;
// const char *module_name;
// size_t has_dynamic_init;
+ // void *source_location;
// We initialize an array of such structures and pass it to a run-time call.
- StructType *GlobalStructTy = StructType::get(IntptrTy, IntptrTy,
- IntptrTy, IntptrTy,
- IntptrTy, IntptrTy, NULL);
+ StructType *GlobalStructTy =
+ StructType::get(IntptrTy, IntptrTy, IntptrTy, IntptrTy, IntptrTy,
+ IntptrTy, IntptrTy, NULL);
SmallVector<Constant *, 16> Initializers(n);
bool HasDynamicallyInitializedGlobals = false;
for (size_t i = 0; i < n; i++) {
static const uint64_t kMaxGlobalRedzone = 1 << 18;
GlobalVariable *G = GlobalsToChange[i];
+
+ auto MD = GlobalsMD.get(G);
+ // Create string holding the global name (use global name from metadata
+ // if it's available, otherwise just write the name of global variable).
+ GlobalVariable *Name = createPrivateGlobalForString(
+ M, MD.Name.empty() ? G->getName() : MD.Name,
+ /*AllowMerging*/ true);
+
PointerType *PtrTy = cast<PointerType>(G->getType());
Type *Ty = PtrTy->getElementType();
uint64_t SizeInBytes = DL->getTypeAllocSize(Ty);
RightRedzoneSize += MinRZ - (SizeInBytes % MinRZ);
assert(((RightRedzoneSize + SizeInBytes) % MinRZ) == 0);
Type *RightRedZoneTy = ArrayType::get(IRB.getInt8Ty(), RightRedzoneSize);
- // Determine whether this global should be poisoned in initialization.
- bool GlobalHasDynamicInitializer =
- DynamicallyInitializedGlobals.Contains(G);
StructType *NewTy = StructType::get(Ty, RightRedZoneTy, NULL);
Constant *NewInitializer = ConstantStruct::get(
NewTy, G->getInitializer(),
Constant::getNullValue(RightRedZoneTy), NULL);
- GlobalVariable *Name =
- createPrivateGlobalForString(M, G->getName(), /*AllowMerging*/true);
-
// Create a new global variable with enough space for a redzone.
GlobalValue::LinkageTypes Linkage = G->getLinkage();
if (G->isConstant() && Linkage == GlobalValue::PrivateLinkage)
NewGlobal->takeName(G);
G->eraseFromParent();
+ Constant *SourceLoc;
+ if (!MD.SourceLoc.empty()) {
+ auto SourceLocGlobal = createPrivateGlobalForSourceLoc(M, MD.SourceLoc);
+ SourceLoc = ConstantExpr::getPointerCast(SourceLocGlobal, IntptrTy);
+ } else {
+ SourceLoc = ConstantInt::get(IntptrTy, 0);
+ }
+
Initializers[i] = ConstantStruct::get(
- GlobalStructTy,
- ConstantExpr::getPointerCast(NewGlobal, IntptrTy),
+ GlobalStructTy, ConstantExpr::getPointerCast(NewGlobal, IntptrTy),
ConstantInt::get(IntptrTy, SizeInBytes),
ConstantInt::get(IntptrTy, SizeInBytes + RightRedzoneSize),
ConstantExpr::getPointerCast(Name, IntptrTy),
ConstantExpr::getPointerCast(ModuleName, IntptrTy),
- ConstantInt::get(IntptrTy, GlobalHasDynamicInitializer),
- NULL);
+ ConstantInt::get(IntptrTy, MD.IsDynInit), SourceLoc, NULL);
- // Populate the first and last globals declared in this TU.
- if (CheckInitOrder && GlobalHasDynamicInitializer)
+ if (ClInitializers && MD.IsDynInit)
HasDynamicallyInitializedGlobals = true;
DEBUG(dbgs() << "NEW GLOBAL: " << *NewGlobal << "\n");
ConstantArray::get(ArrayOfGlobalStructTy, Initializers), "");
// Create calls for poisoning before initializers run and unpoisoning after.
- if (CheckInitOrder && HasDynamicallyInitializedGlobals)
+ if (HasDynamicallyInitializedGlobals)
createInitializerPoisonCalls(M, ModuleName);
IRB.CreateCall2(AsanRegisterGlobals,
IRB.CreatePointerCast(AllGlobals, IntptrTy),
return true;
}
+bool AddressSanitizerModule::runOnModule(Module &M) {
+ DataLayoutPass *DLP = getAnalysisIfAvailable<DataLayoutPass>();
+ if (!DLP)
+ return false;
+ DL = &DLP->getDataLayout();
+ C = &(M.getContext());
+ int LongSize = DL->getPointerSizeInBits();
+ IntptrTy = Type::getIntNTy(*C, LongSize);
+ Mapping = getShadowMapping(M, LongSize);
+ initializeCallbacks(M);
+
+ bool Changed = false;
+
+ Function *CtorFunc = M.getFunction(kAsanModuleCtorName);
+ assert(CtorFunc);
+ IRBuilder<> IRB(CtorFunc->getEntryBlock().getTerminator());
+
+ if (ClCoverage > 0) {
+ Function *CovFunc = M.getFunction(kAsanCovName);
+ int nCov = CovFunc ? CovFunc->getNumUses() : 0;
+ IRB.CreateCall(AsanCovModuleInit, ConstantInt::get(IntptrTy, nCov));
+ Changed = true;
+ }
+
+ if (ClGlobals)
+ Changed |= InstrumentGlobals(IRB, M);
+
+ return Changed;
+}
+
void AddressSanitizer::initializeCallbacks(Module &M) {
IRBuilder<> IRB(*C);
// Create __asan_report* callbacks.
report_fatal_error("data layout missing");
DL = &DLP->getDataLayout();
- BL.reset(SpecialCaseList::createOrDie(BlacklistFile));
- DynamicallyInitializedGlobals.Init(M);
+ GlobalsMD.init(M);
C = &(M.getContext());
LongSize = DL->getPointerSizeInBits();
break;
}
+ DebugLoc EntryLoc = IP->getDebugLoc().getFnDebugLoc(*C);
IRBuilder<> IRB(IP);
+ IRB.SetCurrentDebugLocation(EntryLoc);
Type *Int8Ty = IRB.getInt8Ty();
GlobalVariable *Guard = new GlobalVariable(
*F.getParent(), Int8Ty, false, GlobalValue::PrivateLinkage,
Instruction *Ins = SplitBlockAndInsertIfThen(
Cmp, IP, false, MDBuilder(*C).createBranchWeights(1, 100000));
IRB.SetInsertPoint(Ins);
- // We pass &F to __sanitizer_cov. We could avoid this and rely on
- // GET_CALLER_PC, but having the PC of the first instruction is just nice.
- Instruction *Call = IRB.CreateCall(AsanCovFunction);
- Call->setDebugLoc(IP->getDebugLoc());
+ IRB.SetCurrentDebugLocation(EntryLoc);
+ // __sanitizer_cov gets the PC of the instruction using GET_CALLER_PC.
+ IRB.CreateCall(AsanCovFunction);
StoreInst *Store = IRB.CreateStore(ConstantInt::get(Int8Ty, 1), Guard);
Store->setAtomic(Monotonic);
Store->setAlignment(1);
// as the function and inject this code into the entry block (-asan-coverage=1)
// or all blocks (-asan-coverage=2):
// if (*Guard) {
-// __sanitizer_cov(&F);
+// __sanitizer_cov();
// *Guard = 1;
// }
// The accesses to Guard are atomic. The rest of the logic is
}
bool AddressSanitizer::runOnFunction(Function &F) {
- if (BL->isIn(F)) return false;
if (&F == AsanCtorFunction) return false;
if (F.getLinkage() == GlobalValue::AvailableExternallyLinkage) return false;
DEBUG(dbgs() << "ASAN instrumenting:\n" << F << "\n");
DEBUG(dbgs() << L.DescriptionString << " --- " << L.FrameSize << "\n");
uint64_t LocalStackSize = L.FrameSize;
bool DoStackMalloc =
- ASan.CheckUseAfterReturn && LocalStackSize <= kMaxStackMallocSize;
+ ClUseAfterReturn && LocalStackSize <= kMaxStackMallocSize;
Type *ByteArrayTy = ArrayType::get(IRB.getInt8Ty(), LocalStackSize);
AllocaInst *MyAlloca =