Google Git
Sign in
llvm / llvm-project / llvm / 49a8cd29ed54c64b3b1fd80612508262a4a7b519 / . / lib / Transforms / Instrumentation / MemorySanitizer.cpp
blob: 3e4fae586aae22ddcd176b5d1a9a85653c954baa [file] [log] [blame]
//===- MemorySanitizer.cpp - detector of uninitialized reads --------------===//
//
// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
// See https://llvm.org/LICENSE.txt for license information.
// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
//
//===----------------------------------------------------------------------===//
//
/// \file
/// This file is a part of MemorySanitizer, a detector of uninitialized
/// reads.
///
/// The algorithm of the tool is similar to Memcheck
/// (http://goo.gl/QKbem). We associate a few shadow bits with every
/// byte of the application memory, poison the shadow of the malloc-ed
/// or alloca-ed memory, load the shadow bits on every memory read,
/// propagate the shadow bits through some of the arithmetic
/// instruction (including MOV), store the shadow bits on every memory
/// write, report a bug on some other instructions (e.g. JMP) if the
/// associated shadow is poisoned.
///
/// But there are differences too. The first and the major one:
/// compiler instrumentation instead of binary instrumentation. This
/// gives us much better register allocation, possible compiler
/// optimizations and a fast start-up. But this brings the major issue
/// as well: msan needs to see all program events, including system
/// calls and reads/writes in system libraries, so we either need to
/// compile *everything* with msan or use a binary translation
/// component (e.g. DynamoRIO) to instrument pre-built libraries.
/// Another difference from Memcheck is that we use 8 shadow bits per
/// byte of application memory and use a direct shadow mapping. This
/// greatly simplifies the instrumentation code and avoids races on
/// shadow updates (Memcheck is single-threaded so races are not a
/// concern there. Memcheck uses 2 shadow bits per byte with a slow
/// path storage that uses 8 bits per byte).
///
/// The default value of shadow is 0, which means "clean" (not poisoned).
///
/// Every module initializer should call __msan_init to ensure that the
/// shadow memory is ready. On error, __msan_warning is called. Since
/// parameters and return values may be passed via registers, we have a
/// specialized thread-local shadow for return values
/// (__msan_retval_tls) and parameters (__msan_param_tls).
///
/// Origin tracking.
///
/// MemorySanitizer can track origins (allocation points) of all uninitialized
/// values. This behavior is controlled with a flag (msan-track-origins) and is
/// disabled by default.
///
/// Origins are 4-byte values created and interpreted by the runtime library.
/// They are stored in a second shadow mapping, one 4-byte value for 4 bytes
/// of application memory. Propagation of origins is basically a bunch of
/// "select" instructions that pick the origin of a dirty argument, if an
/// instruction has one.
///
/// Every 4 aligned, consecutive bytes of application memory have one origin
/// value associated with them. If these bytes contain uninitialized data
/// coming from 2 different allocations, the last store wins. Because of this,
/// MemorySanitizer reports can show unrelated origins, but this is unlikely in
/// practice.
///
/// Origins are meaningless for fully initialized values, so MemorySanitizer
/// avoids storing origin to memory when a fully initialized value is stored.
/// This way it avoids needless overwriting origin of the 4-byte region on
/// a short (i.e. 1 byte) clean store, and it is also good for performance.
///
/// Atomic handling.
///
/// Ideally, every atomic store of application value should update the
/// corresponding shadow location in an atomic way. Unfortunately, atomic store
/// of two disjoint locations can not be done without severe slowdown.
///
/// Therefore, we implement an approximation that may err on the safe side.
/// In this implementation, every atomically accessed location in the program
/// may only change from (partially) uninitialized to fully initialized, but
/// not the other way around. We load the shadow _after_ the application load,
/// and we store the shadow _before_ the app store. Also, we always store clean
/// shadow (if the application store is atomic). This way, if the store-load
/// pair constitutes a happens-before arc, shadow store and load are correctly
/// ordered such that the load will get either the value that was stored, or
/// some later value (which is always clean).
///
/// This does not work very well with Compare-And-Swap (CAS) and
/// Read-Modify-Write (RMW) operations. To follow the above logic, CAS and RMW
/// must store the new shadow before the app operation, and load the shadow
/// after the app operation. Computers don't work this way. Current
/// implementation ignores the load aspect of CAS/RMW, always returning a clean
/// value. It implements the store part as a simple atomic store by storing a
/// clean shadow.
///
/// Instrumenting inline assembly.
///
/// For inline assembly code LLVM has little idea about which memory locations
/// become initialized depending on the arguments. It can be possible to figure
/// out which arguments are meant to point to inputs and outputs, but the
/// actual semantics can be only visible at runtime. In the Linux kernel it's
/// also possible that the arguments only indicate the offset for a base taken
/// from a segment register, so it's dangerous to treat any asm() arguments as
/// pointers. We take a conservative approach generating calls to
/// __msan_instrument_asm_store(ptr, size)
/// , which defer the memory unpoisoning to the runtime library.
/// The latter can perform more complex address checks to figure out whether
/// it's safe to touch the shadow memory.
/// Like with atomic operations, we call __msan_instrument_asm_store() before
/// the assembly call, so that changes to the shadow memory will be seen by
/// other threads together with main memory initialization.
///
/// KernelMemorySanitizer (KMSAN) implementation.
///
/// The major differences between KMSAN and MSan instrumentation are:
/// - KMSAN always tracks the origins and implies msan-keep-going=true;
/// - KMSAN allocates shadow and origin memory for each page separately, so
/// there are no explicit accesses to shadow and origin in the
/// instrumentation.
/// Shadow and origin values for a particular X-byte memory location
/// (X=1,2,4,8) are accessed through pointers obtained via the
/// __msan_metadata_ptr_for_load_X(ptr)
/// __msan_metadata_ptr_for_store_X(ptr)
/// functions. The corresponding functions check that the X-byte accesses
/// are possible and returns the pointers to shadow and origin memory.
/// Arbitrary sized accesses are handled with:
/// __msan_metadata_ptr_for_load_n(ptr, size)
/// __msan_metadata_ptr_for_store_n(ptr, size);
/// - TLS variables are stored in a single per-task struct. A call to a
/// function __msan_get_context_state() returning a pointer to that struct
/// is inserted into every instrumented function before the entry block;
/// - __msan_warning() takes a 32-bit origin parameter;
/// - local variables are poisoned with __msan_poison_alloca() upon function
/// entry and unpoisoned with __msan_unpoison_alloca() before leaving the
/// function;
/// - the pass doesn't declare any global variables or add global constructors
/// to the translation unit.
///
/// Also, KMSAN currently ignores uninitialized memory passed into inline asm
/// calls, making sure we're on the safe side wrt. possible false positives.
///
/// KernelMemorySanitizer only supports X86_64 at the moment.
///
//
// FIXME: This sanitizer does not yet handle scalable vectors
//
//===----------------------------------------------------------------------===//
#include "llvm/Transforms/Instrumentation/MemorySanitizer.h"
#include "llvm/ADT/APInt.h"
#include "llvm/ADT/ArrayRef.h"
#include "llvm/ADT/DepthFirstIterator.h"
#include "llvm/ADT/SmallSet.h"
#include "llvm/ADT/SmallString.h"
#include "llvm/ADT/SmallVector.h"
#include "llvm/ADT/StringExtras.h"
#include "llvm/ADT/StringRef.h"
#include "llvm/ADT/Triple.h"
#include "llvm/Analysis/TargetLibraryInfo.h"
#include "llvm/Analysis/ValueTracking.h"
#include "llvm/IR/Argument.h"
#include "llvm/IR/Attributes.h"
#include "llvm/IR/BasicBlock.h"
#include "llvm/IR/CallingConv.h"
#include "llvm/IR/Constant.h"
#include "llvm/IR/Constants.h"
#include "llvm/IR/DataLayout.h"
#include "llvm/IR/DerivedTypes.h"
#include "llvm/IR/Function.h"
#include "llvm/IR/GlobalValue.h"
#include "llvm/IR/GlobalVariable.h"
#include "llvm/IR/IRBuilder.h"
#include "llvm/IR/InlineAsm.h"
#include "llvm/IR/InstVisitor.h"
#include "llvm/IR/InstrTypes.h"
#include "llvm/IR/Instruction.h"
#include "llvm/IR/Instructions.h"
#include "llvm/IR/IntrinsicInst.h"
#include "llvm/IR/Intrinsics.h"
#include "llvm/IR/IntrinsicsX86.h"
#include "llvm/IR/LLVMContext.h"
#include "llvm/IR/MDBuilder.h"
#include "llvm/IR/Module.h"
#include "llvm/IR/Type.h"
#include "llvm/IR/Value.h"
#include "llvm/IR/ValueMap.h"
#include "llvm/InitializePasses.h"
#include "llvm/Pass.h"
#include "llvm/Support/AtomicOrdering.h"
#include "llvm/Support/Casting.h"
#include "llvm/Support/CommandLine.h"
#include "llvm/Support/Compiler.h"
#include "llvm/Support/Debug.h"
#include "llvm/Support/ErrorHandling.h"
#include "llvm/Support/MathExtras.h"
#include "llvm/Support/raw_ostream.h"
#include "llvm/Transforms/Instrumentation.h"
#include "llvm/Transforms/Utils/BasicBlockUtils.h"
#include "llvm/Transforms/Utils/Local.h"
#include "llvm/Transforms/Utils/ModuleUtils.h"
#include <algorithm>
#include <cassert>
#include <cstddef>
#include <cstdint>
#include <memory>
#include <string>
#include <tuple>
using namespace llvm;
#define DEBUG_TYPE "msan"
static const unsigned kOriginSize = 4;
static const Align kMinOriginAlignment = Align(4);
static const Align kShadowTLSAlignment = Align(8);
// These constants must be kept in sync with the ones in msan.h.
static const unsigned kParamTLSSize = 800;
static const unsigned kRetvalTLSSize = 800;
// Accesses sizes are powers of two: 1, 2, 4, 8.
static const size_t kNumberOfAccessSizes = 4;
/// Track origins of uninitialized values.
///
/// Adds a section to MemorySanitizer report that points to the allocation
/// (stack or heap) the uninitialized bits came from originally.
static cl::opt<int> ClTrackOrigins("msan-track-origins",
cl::desc("Track origins (allocation sites) of poisoned memory"),
cl::Hidden, cl::init(0));
static cl::opt<bool> ClKeepGoing("msan-keep-going",
cl::desc("keep going after reporting a UMR"),
cl::Hidden, cl::init(false));
static cl::opt<bool> ClPoisonStack("msan-poison-stack",
cl::desc("poison uninitialized stack variables"),
cl::Hidden, cl::init(true));
static cl::opt<bool> ClPoisonStackWithCall("msan-poison-stack-with-call",
cl::desc("poison uninitialized stack variables with a call"),
cl::Hidden, cl::init(false));
static cl::opt<int> ClPoisonStackPattern("msan-poison-stack-pattern",
cl::desc("poison uninitialized stack variables with the given pattern"),
cl::Hidden, cl::init(0xff));
static cl::opt<bool> ClPoisonUndef("msan-poison-undef",
cl::desc("poison undef temps"),
cl::Hidden, cl::init(true));
static cl::opt<bool> ClHandleICmp("msan-handle-icmp",
cl::desc("propagate shadow through ICmpEQ and ICmpNE"),
cl::Hidden, cl::init(true));
static cl::opt<bool> ClHandleICmpExact("msan-handle-icmp-exact",
cl::desc("exact handling of relational integer ICmp"),
cl::Hidden, cl::init(false));
static cl::opt<bool> ClHandleLifetimeIntrinsics(
"msan-handle-lifetime-intrinsics",
cl::desc(
"when possible, poison scoped variables at the beginning of the scope "
"(slower, but more precise)"),
cl::Hidden, cl::init(true));
// When compiling the Linux kernel, we sometimes see false positives related to
// MSan being unable to understand that inline assembly calls may initialize
// local variables.
// This flag makes the compiler conservatively unpoison every memory location
// passed into an assembly call. Note that this may cause false positives.
// Because it's impossible to figure out the array sizes, we can only unpoison
// the first sizeof(type) bytes for each type* pointer.
// The instrumentation is only enabled in KMSAN builds, and only if
// -msan-handle-asm-conservative is on. This is done because we may want to
// quickly disable assembly instrumentation when it breaks.
static cl::opt<bool> ClHandleAsmConservative(
"msan-handle-asm-conservative",
cl::desc("conservative handling of inline assembly"), cl::Hidden,
cl::init(true));
// This flag controls whether we check the shadow of the address
// operand of load or store. Such bugs are very rare, since load from
// a garbage address typically results in SEGV, but still happen
// (e.g. only lower bits of address are garbage, or the access happens
// early at program startup where malloc-ed memory is more likely to
// be zeroed. As of 2012-08-28 this flag adds 20% slowdown.
static cl::opt<bool> ClCheckAccessAddress("msan-check-access-address",
cl::desc("report accesses through a pointer which has poisoned shadow"),
cl::Hidden, cl::init(true));
static cl::opt<bool> ClEagerChecks(
"msan-eager-checks",
cl::desc("check arguments and return values at function call boundaries"),
cl::Hidden, cl::init(false));
static cl::opt<bool> ClDumpStrictInstructions("msan-dump-strict-instructions",
cl::desc("print out instructions with default strict semantics"),
cl::Hidden, cl::init(false));
static cl::opt<int> ClInstrumentationWithCallThreshold(
"msan-instrumentation-with-call-threshold",
cl::desc(
"If the function being instrumented requires more than "
"this number of checks and origin stores, use callbacks instead of "
"inline checks (-1 means never use callbacks)."),
cl::Hidden, cl::init(3500));
static cl::opt<bool>
ClEnableKmsan("msan-kernel",
cl::desc("Enable KernelMemorySanitizer instrumentation"),
cl::Hidden, cl::init(false));
// This is an experiment to enable handling of cases where shadow is a non-zero
// compile-time constant. For some unexplainable reason they were silently
// ignored in the instrumentation.
static cl::opt<bool> ClCheckConstantShadow("msan-check-constant-shadow",
cl::desc("Insert checks for constant shadow values"),
cl::Hidden, cl::init(false));
// This is off by default because of a bug in gold:
// https://sourceware.org/bugzilla/show_bug.cgi?id=19002
static cl::opt<bool> ClWithComdat("msan-with-comdat",
cl::desc("Place MSan constructors in comdat sections"),
cl::Hidden, cl::init(false));
// These options allow to specify custom memory map parameters
// See MemoryMapParams for details.
static cl::opt<uint64_t> ClAndMask("msan-and-mask",
cl::desc("Define custom MSan AndMask"),
cl::Hidden, cl::init(0));
static cl::opt<uint64_t> ClXorMask("msan-xor-mask",
cl::desc("Define custom MSan XorMask"),
cl::Hidden, cl::init(0));
static cl::opt<uint64_t> ClShadowBase("msan-shadow-base",
cl::desc("Define custom MSan ShadowBase"),
cl::Hidden, cl::init(0));
static cl::opt<uint64_t> ClOriginBase("msan-origin-base",
cl::desc("Define custom MSan OriginBase"),
cl::Hidden, cl::init(0));
const char kMsanModuleCtorName[] = "msan.module_ctor";
const char kMsanInitName[] = "__msan_init";
namespace {
// Memory map parameters used in application-to-shadow address calculation.
// Offset = (Addr & ~AndMask) ^ XorMask
// Shadow = ShadowBase + Offset
// Origin = OriginBase + Offset
struct MemoryMapParams {
uint64_t AndMask;
uint64_t XorMask;
uint64_t ShadowBase;
uint64_t OriginBase;
};
struct PlatformMemoryMapParams {
const MemoryMapParams *bits32;
const MemoryMapParams *bits64;
};
} // end anonymous namespace
// i386 Linux
static const MemoryMapParams Linux_I386_MemoryMapParams = {
0x000080000000, // AndMask
0, // XorMask (not used)
0, // ShadowBase (not used)
0x000040000000, // OriginBase
};
// x86_64 Linux
static const MemoryMapParams Linux_X86_64_MemoryMapParams = {
#ifdef MSAN_LINUX_X86_64_OLD_MAPPING
0x400000000000, // AndMask
0, // XorMask (not used)
0, // ShadowBase (not used)
0x200000000000, // OriginBase
#else
0, // AndMask (not used)
0x500000000000, // XorMask
0, // ShadowBase (not used)
0x100000000000, // OriginBase
#endif
};
// mips64 Linux
static const MemoryMapParams Linux_MIPS64_MemoryMapParams = {
0, // AndMask (not used)
0x008000000000, // XorMask
0, // ShadowBase (not used)
0x002000000000, // OriginBase
};
// ppc64 Linux
static const MemoryMapParams Linux_PowerPC64_MemoryMapParams = {
0xE00000000000, // AndMask
0x100000000000, // XorMask
0x080000000000, // ShadowBase
0x1C0000000000, // OriginBase
};
// s390x Linux
static const MemoryMapParams Linux_S390X_MemoryMapParams = {
0xC00000000000, // AndMask
0, // XorMask (not used)
0x080000000000, // ShadowBase
0x1C0000000000, // OriginBase
};
// aarch64 Linux
static const MemoryMapParams Linux_AArch64_MemoryMapParams = {
0, // AndMask (not used)
0x06000000000, // XorMask
0, // ShadowBase (not used)
0x01000000000, // OriginBase
};
// i386 FreeBSD
static const MemoryMapParams FreeBSD_I386_MemoryMapParams = {
0x000180000000, // AndMask
0x000040000000, // XorMask
0x000020000000, // ShadowBase
0x000700000000, // OriginBase
};
// x86_64 FreeBSD
static const MemoryMapParams FreeBSD_X86_64_MemoryMapParams = {
0xc00000000000, // AndMask
0x200000000000, // XorMask
0x100000000000, // ShadowBase
0x380000000000, // OriginBase
};
// x86_64 NetBSD
static const MemoryMapParams NetBSD_X86_64_MemoryMapParams = {
0, // AndMask
0x500000000000, // XorMask
0, // ShadowBase
0x100000000000, // OriginBase
};
static const PlatformMemoryMapParams Linux_X86_MemoryMapParams = {
&Linux_I386_MemoryMapParams,
&Linux_X86_64_MemoryMapParams,
};
static const PlatformMemoryMapParams Linux_MIPS_MemoryMapParams = {
nullptr,
&Linux_MIPS64_MemoryMapParams,
};
static const PlatformMemoryMapParams Linux_PowerPC_MemoryMapParams = {
nullptr,
&Linux_PowerPC64_MemoryMapParams,
};
static const PlatformMemoryMapParams Linux_S390_MemoryMapParams = {
nullptr,
&Linux_S390X_MemoryMapParams,
};
static const PlatformMemoryMapParams Linux_ARM_MemoryMapParams = {
nullptr,
&Linux_AArch64_MemoryMapParams,
};
static const PlatformMemoryMapParams FreeBSD_X86_MemoryMapParams = {
&FreeBSD_I386_MemoryMapParams,
&FreeBSD_X86_64_MemoryMapParams,
};
static const PlatformMemoryMapParams NetBSD_X86_MemoryMapParams = {
nullptr,
&NetBSD_X86_64_MemoryMapParams,
};
namespace {
/// Instrument functions of a module to detect uninitialized reads.
///
/// Instantiating MemorySanitizer inserts the msan runtime library API function
/// declarations into the module if they don't exist already. Instantiating
/// ensures the __msan_init function is in the list of global constructors for
/// the module.
class MemorySanitizer {
public:
MemorySanitizer(Module &M, MemorySanitizerOptions Options)
: CompileKernel(Options.Kernel), TrackOrigins(Options.TrackOrigins),
Recover(Options.Recover) {
initializeModule(M);
}
// MSan cannot be moved or copied because of MapParams.
MemorySanitizer(MemorySanitizer &&) = delete;
MemorySanitizer &operator=(MemorySanitizer &&) = delete;
MemorySanitizer(const MemorySanitizer &) = delete;
MemorySanitizer &operator=(const MemorySanitizer &) = delete;
bool sanitizeFunction(Function &F, TargetLibraryInfo &TLI);
private:
friend struct MemorySanitizerVisitor;
friend struct VarArgAMD64Helper;
friend struct VarArgMIPS64Helper;
friend struct VarArgAArch64Helper;
friend struct VarArgPowerPC64Helper;
friend struct VarArgSystemZHelper;
void initializeModule(Module &M);
void initializeCallbacks(Module &M);
void createKernelApi(Module &M);
void createUserspaceApi(Module &M);
/// True if we're compiling the Linux kernel.
bool CompileKernel;
/// Track origins (allocation points) of uninitialized values.
int TrackOrigins;
bool Recover;
LLVMContext *C;
Type *IntptrTy;
Type *OriginTy;
// XxxTLS variables represent the per-thread state in MSan and per-task state
// in KMSAN.
// For the userspace these point to thread-local globals. In the kernel land
// they point to the members of a per-task struct obtained via a call to
// __msan_get_context_state().
/// Thread-local shadow storage for function parameters.
Value *ParamTLS;
/// Thread-local origin storage for function parameters.
Value *ParamOriginTLS;
/// Thread-local shadow storage for function return value.
Value *RetvalTLS;
/// Thread-local origin storage for function return value.
Value *RetvalOriginTLS;
/// Thread-local shadow storage for in-register va_arg function
/// parameters (x86_64-specific).
Value *VAArgTLS;
/// Thread-local shadow storage for in-register va_arg function
/// parameters (x86_64-specific).
Value *VAArgOriginTLS;
/// Thread-local shadow storage for va_arg overflow area
/// (x86_64-specific).
Value *VAArgOverflowSizeTLS;
/// Are the instrumentation callbacks set up?
bool CallbacksInitialized = false;
/// The run-time callback to print a warning.
FunctionCallee WarningFn;
// These arrays are indexed by log2(AccessSize).
FunctionCallee MaybeWarningFn[kNumberOfAccessSizes];
FunctionCallee MaybeStoreOriginFn[kNumberOfAccessSizes];
/// Run-time helper that generates a new origin value for a stack
/// allocation.
FunctionCallee MsanSetAllocaOrigin4Fn;
/// Run-time helper that poisons stack on function entry.
FunctionCallee MsanPoisonStackFn;
/// Run-time helper that records a store (or any event) of an
/// uninitialized value and returns an updated origin id encoding this info.
FunctionCallee MsanChainOriginFn;
/// Run-time helper that paints an origin over a region.
FunctionCallee MsanSetOriginFn;
/// MSan runtime replacements for memmove, memcpy and memset.
FunctionCallee MemmoveFn, MemcpyFn, MemsetFn;
/// KMSAN callback for task-local function argument shadow.
StructType *MsanContextStateTy;
FunctionCallee MsanGetContextStateFn;
/// Functions for poisoning/unpoisoning local variables
FunctionCallee MsanPoisonAllocaFn, MsanUnpoisonAllocaFn;
/// Each of the MsanMetadataPtrXxx functions returns a pair of shadow/origin
/// pointers.
FunctionCallee MsanMetadataPtrForLoadN, MsanMetadataPtrForStoreN;
FunctionCallee MsanMetadataPtrForLoad_1_8[4];
FunctionCallee MsanMetadataPtrForStore_1_8[4];
FunctionCallee MsanInstrumentAsmStoreFn;
/// Helper to choose between different MsanMetadataPtrXxx().
FunctionCallee getKmsanShadowOriginAccessFn(bool isStore, int size);
/// Memory map parameters used in application-to-shadow calculation.
const MemoryMapParams *MapParams;
/// Custom memory map parameters used when -msan-shadow-base or
// -msan-origin-base is provided.
MemoryMapParams CustomMapParams;
MDNode *ColdCallWeights;
/// Branch weights for origin store.
MDNode *OriginStoreWeights;
};
void insertModuleCtor(Module &M) {
getOrCreateSanitizerCtorAndInitFunctions(
M, kMsanModuleCtorName, kMsanInitName,
/*InitArgTypes=*/{},
/*InitArgs=*/{},
// This callback is invoked when the functions are created the first
// time. Hook them into the global ctors list in that case:
[&](Function *Ctor, FunctionCallee) {
if (!ClWithComdat) {
appendToGlobalCtors(M, Ctor, 0);
return;
}
Comdat *MsanCtorComdat = M.getOrInsertComdat(kMsanModuleCtorName);
Ctor->setComdat(MsanCtorComdat);
appendToGlobalCtors(M, Ctor, 0, Ctor);
});
}
/// A legacy function pass for msan instrumentation.
///
/// Instruments functions to detect uninitialized reads.
struct MemorySanitizerLegacyPass : public FunctionPass {
// Pass identification, replacement for typeid.
static char ID;
MemorySanitizerLegacyPass(MemorySanitizerOptions Options = {})
: FunctionPass(ID), Options(Options) {
initializeMemorySanitizerLegacyPassPass(*PassRegistry::getPassRegistry());
}
StringRef getPassName() const override { return "MemorySanitizerLegacyPass"; }
void getAnalysisUsage(AnalysisUsage &AU) const override {
AU.addRequired<TargetLibraryInfoWrapperPass>();
}
bool runOnFunction(Function &F) override {
return MSan->sanitizeFunction(
F, getAnalysis<TargetLibraryInfoWrapperPass>().getTLI(F));
}
bool doInitialization(Module &M) override;
Optional<MemorySanitizer> MSan;
MemorySanitizerOptions Options;
};
template <class T> T getOptOrDefault(const cl::opt<T> &Opt, T Default) {
return (Opt.getNumOccurrences() > 0) ? Opt : Default;
}
} // end anonymous namespace
MemorySanitizerOptions::MemorySanitizerOptions(int TO, bool R, bool K)
: Kernel(getOptOrDefault(ClEnableKmsan, K)),
TrackOrigins(getOptOrDefault(ClTrackOrigins, Kernel ? 2 : TO)),
Recover(getOptOrDefault(ClKeepGoing, Kernel || R)) {}
PreservedAnalyses MemorySanitizerPass::run(Function &F,
FunctionAnalysisManager &FAM) {
MemorySanitizer Msan(*F.getParent(), Options);
if (Msan.sanitizeFunction(F, FAM.getResult<TargetLibraryAnalysis>(F)))
return PreservedAnalyses::none();
return PreservedAnalyses::all();
}
PreservedAnalyses MemorySanitizerPass::run(Module &M,
ModuleAnalysisManager &AM) {
if (Options.Kernel)
return PreservedAnalyses::all();
insertModuleCtor(M);
return PreservedAnalyses::none();
}
char MemorySanitizerLegacyPass::ID = 0;
INITIALIZE_PASS_BEGIN(MemorySanitizerLegacyPass, "msan",
"MemorySanitizer: detects uninitialized reads.", false,
false)
INITIALIZE_PASS_DEPENDENCY(TargetLibraryInfoWrapperPass)
INITIALIZE_PASS_END(MemorySanitizerLegacyPass, "msan",
"MemorySanitizer: detects uninitialized reads.", false,
false)
FunctionPass *
llvm::createMemorySanitizerLegacyPassPass(MemorySanitizerOptions Options) {
return new MemorySanitizerLegacyPass(Options);
}
/// Create a non-const global initialized with the given string.
///
/// Creates a writable global for Str so that we can pass it to the
/// run-time lib. Runtime uses first 4 bytes of the string to store the
/// frame ID, so the string needs to be mutable.
static GlobalVariable *createPrivateNonConstGlobalForString(Module &M,
StringRef Str) {
Constant *StrConst = ConstantDataArray::getString(M.getContext(), Str);
return new GlobalVariable(M, StrConst->getType(), /*isConstant=*/false,
GlobalValue::PrivateLinkage, StrConst, "");
}
/// Create KMSAN API callbacks.
void MemorySanitizer::createKernelApi(Module &M) {
IRBuilder<> IRB(*C);
// These will be initialized in insertKmsanPrologue().
RetvalTLS = nullptr;
RetvalOriginTLS = nullptr;
ParamTLS = nullptr;
ParamOriginTLS = nullptr;
VAArgTLS = nullptr;
VAArgOriginTLS = nullptr;
VAArgOverflowSizeTLS = nullptr;
WarningFn = M.getOrInsertFunction("__msan_warning", IRB.getVoidTy(),
IRB.getInt32Ty());
// Requests the per-task context state (kmsan_context_state*) from the
// runtime library.
MsanContextStateTy = StructType::get(
ArrayType::get(IRB.getInt64Ty(), kParamTLSSize / 8),
ArrayType::get(IRB.getInt64Ty(), kRetvalTLSSize / 8),
ArrayType::get(IRB.getInt64Ty(), kParamTLSSize / 8),
ArrayType::get(IRB.getInt64Ty(), kParamTLSSize / 8), /* va_arg_origin */
IRB.getInt64Ty(), ArrayType::get(OriginTy, kParamTLSSize / 4), OriginTy,
OriginTy);
MsanGetContextStateFn = M.getOrInsertFunction(
"__msan_get_context_state", PointerType::get(MsanContextStateTy, 0));
Type *RetTy = StructType::get(PointerType::get(IRB.getInt8Ty(), 0),
PointerType::get(IRB.getInt32Ty(), 0));
for (int ind = 0, size = 1; ind < 4; ind++, size <<= 1) {
std::string name_load =
"__msan_metadata_ptr_for_load_" + std::to_string(size);
std::string name_store =
"__msan_metadata_ptr_for_store_" + std::to_string(size);
MsanMetadataPtrForLoad_1_8[ind] = M.getOrInsertFunction(
name_load, RetTy, PointerType::get(IRB.getInt8Ty(), 0));
MsanMetadataPtrForStore_1_8[ind] = M.getOrInsertFunction(
name_store, RetTy, PointerType::get(IRB.getInt8Ty(), 0));
}
MsanMetadataPtrForLoadN = M.getOrInsertFunction(
"__msan_metadata_ptr_for_load_n", RetTy,
PointerType::get(IRB.getInt8Ty(), 0), IRB.getInt64Ty());
MsanMetadataPtrForStoreN = M.getOrInsertFunction(
"__msan_metadata_ptr_for_store_n", RetTy,
PointerType::get(IRB.getInt8Ty(), 0), IRB.getInt64Ty());
// Functions for poisoning and unpoisoning memory.
MsanPoisonAllocaFn =
M.getOrInsertFunction("__msan_poison_alloca", IRB.getVoidTy(),
IRB.getInt8PtrTy(), IntptrTy, IRB.getInt8PtrTy());
MsanUnpoisonAllocaFn = M.getOrInsertFunction(
"__msan_unpoison_alloca", IRB.getVoidTy(), IRB.getInt8PtrTy(), IntptrTy);
}
static Constant *getOrInsertGlobal(Module &M, StringRef Name, Type *Ty) {
return M.getOrInsertGlobal(Name, Ty, [&] {
return new GlobalVariable(M, Ty, false, GlobalVariable::ExternalLinkage,
nullptr, Name, nullptr,
GlobalVariable::InitialExecTLSModel);
});
}
/// Insert declarations for userspace-specific functions and globals.
void MemorySanitizer::createUserspaceApi(Module &M) {
IRBuilder<> IRB(*C);
// Create the callback.
// FIXME: this function should have "Cold" calling conv,
// which is not yet implemented.
StringRef WarningFnName = Recover ? "__msan_warning_with_origin"
: "__msan_warning_with_origin_noreturn";
WarningFn =
M.getOrInsertFunction(WarningFnName, IRB.getVoidTy(), IRB.getInt32Ty());
// Create the global TLS variables.
RetvalTLS =
getOrInsertGlobal(M, "__msan_retval_tls",
ArrayType::get(IRB.getInt64Ty(), kRetvalTLSSize / 8));
RetvalOriginTLS = getOrInsertGlobal(M, "__msan_retval_origin_tls", OriginTy);
ParamTLS =
getOrInsertGlobal(M, "__msan_param_tls",
ArrayType::get(IRB.getInt64Ty(), kParamTLSSize / 8));
ParamOriginTLS =
getOrInsertGlobal(M, "__msan_param_origin_tls",
ArrayType::get(OriginTy, kParamTLSSize / 4));
VAArgTLS =
getOrInsertGlobal(M, "__msan_va_arg_tls",
ArrayType::get(IRB.getInt64Ty(), kParamTLSSize / 8));
VAArgOriginTLS =
getOrInsertGlobal(M, "__msan_va_arg_origin_tls",
ArrayType::get(OriginTy, kParamTLSSize / 4));
VAArgOverflowSizeTLS =
getOrInsertGlobal(M, "__msan_va_arg_overflow_size_tls", IRB.getInt64Ty());
for (size_t AccessSizeIndex = 0; AccessSizeIndex < kNumberOfAccessSizes;
AccessSizeIndex++) {
unsigned AccessSize = 1 << AccessSizeIndex;
std::string FunctionName = "__msan_maybe_warning_" + itostr(AccessSize);
SmallVector<std::pair<unsigned, Attribute>, 2> MaybeWarningFnAttrs;
MaybeWarningFnAttrs.push_back(std::make_pair(
AttributeList::FirstArgIndex, Attribute::get(*C, Attribute::ZExt)));
MaybeWarningFnAttrs.push_back(std::make_pair(
AttributeList::FirstArgIndex + 1, Attribute::get(*C, Attribute::ZExt)));
MaybeWarningFn[AccessSizeIndex] = M.getOrInsertFunction(
FunctionName, AttributeList::get(*C, MaybeWarningFnAttrs),
IRB.getVoidTy(), IRB.getIntNTy(AccessSize * 8), IRB.getInt32Ty());
FunctionName = "__msan_maybe_store_origin_" + itostr(AccessSize);
SmallVector<std::pair<unsigned, Attribute>, 2> MaybeStoreOriginFnAttrs;
MaybeStoreOriginFnAttrs.push_back(std::make_pair(
AttributeList::FirstArgIndex, Attribute::get(*C, Attribute::ZExt)));
MaybeStoreOriginFnAttrs.push_back(std::make_pair(
AttributeList::FirstArgIndex + 2, Attribute::get(*C, Attribute::ZExt)));
MaybeStoreOriginFn[AccessSizeIndex] = M.getOrInsertFunction(
FunctionName, AttributeList::get(*C, MaybeStoreOriginFnAttrs),
IRB.getVoidTy(), IRB.getIntNTy(AccessSize * 8), IRB.getInt8PtrTy(),
IRB.getInt32Ty());
}
MsanSetAllocaOrigin4Fn = M.getOrInsertFunction(
"__msan_set_alloca_origin4", IRB.getVoidTy(), IRB.getInt8PtrTy(), IntptrTy,
IRB.getInt8PtrTy(), IntptrTy);
MsanPoisonStackFn =
M.getOrInsertFunction("__msan_poison_stack", IRB.getVoidTy(),
IRB.getInt8PtrTy(), IntptrTy);
}
/// Insert extern declaration of runtime-provided functions and globals.
void MemorySanitizer::initializeCallbacks(Module &M) {
// Only do this once.
if (CallbacksInitialized)
return;
IRBuilder<> IRB(*C);
// Initialize callbacks that are common for kernel and userspace
// instrumentation.
MsanChainOriginFn = M.getOrInsertFunction(
"__msan_chain_origin", IRB.getInt32Ty(), IRB.getInt32Ty());
MsanSetOriginFn =
M.getOrInsertFunction("__msan_set_origin", IRB.getVoidTy(),
IRB.getInt8PtrTy(), IntptrTy, IRB.getInt32Ty());
MemmoveFn = M.getOrInsertFunction(
"__msan_memmove", IRB.getInt8PtrTy(), IRB.getInt8PtrTy(),
IRB.getInt8PtrTy(), IntptrTy);
MemcpyFn = M.getOrInsertFunction(
"__msan_memcpy", IRB.getInt8PtrTy(), IRB.getInt8PtrTy(), IRB.getInt8PtrTy(),
IntptrTy);
MemsetFn = M.getOrInsertFunction(
"__msan_memset", IRB.getInt8PtrTy(), IRB.getInt8PtrTy(), IRB.getInt32Ty(),
IntptrTy);
MsanInstrumentAsmStoreFn =
M.getOrInsertFunction("__msan_instrument_asm_store", IRB.getVoidTy(),
PointerType::get(IRB.getInt8Ty(), 0), IntptrTy);
if (CompileKernel) {
createKernelApi(M);
} else {
createUserspaceApi(M);
}
CallbacksInitialized = true;
}
FunctionCallee MemorySanitizer::getKmsanShadowOriginAccessFn(bool isStore,
int size) {
FunctionCallee *Fns =
isStore ? MsanMetadataPtrForStore_1_8 : MsanMetadataPtrForLoad_1_8;
switch (size) {
case 1:
return Fns[0];
case 2:
return Fns[1];
case 4:
return Fns[2];
case 8:
return Fns[3];
default:
return nullptr;
}
}
/// Module-level initialization.
///
/// inserts a call to __msan_init to the module's constructor list.
void MemorySanitizer::initializeModule(Module &M) {
auto &DL = M.getDataLayout();
bool ShadowPassed = ClShadowBase.getNumOccurrences() > 0;
bool OriginPassed = ClOriginBase.getNumOccurrences() > 0;
// Check the overrides first
if (ShadowPassed || OriginPassed) {
CustomMapParams.AndMask = ClAndMask;
CustomMapParams.XorMask = ClXorMask;
CustomMapParams.ShadowBase = ClShadowBase;
CustomMapParams.OriginBase = ClOriginBase;
MapParams = &CustomMapParams;
} else {
Triple TargetTriple(M.getTargetTriple());
switch (TargetTriple.getOS()) {
case Triple::FreeBSD:
switch (TargetTriple.getArch()) {
case Triple::x86_64:
MapParams = FreeBSD_X86_MemoryMapParams.bits64;
break;
case Triple::x86:
MapParams = FreeBSD_X86_MemoryMapParams.bits32;
break;
default:
report_fatal_error("unsupported architecture");
}
break;
case Triple::NetBSD:
switch (TargetTriple.getArch()) {
case Triple::x86_64:
MapParams = NetBSD_X86_MemoryMapParams.bits64;
break;
default:
report_fatal_error("unsupported architecture");
}
break;
case Triple::Linux:
switch (TargetTriple.getArch()) {
case Triple::x86_64:
MapParams = Linux_X86_MemoryMapParams.bits64;
break;
case Triple::x86:
MapParams = Linux_X86_MemoryMapParams.bits32;
break;
case Triple::mips64:
case Triple::mips64el:
MapParams = Linux_MIPS_MemoryMapParams.bits64;
break;
case Triple::ppc64:
case Triple::ppc64le:
MapParams = Linux_PowerPC_MemoryMapParams.bits64;
break;
case Triple::systemz:
MapParams = Linux_S390_MemoryMapParams.bits64;
break;
case Triple::aarch64:
case Triple::aarch64_be:
MapParams = Linux_ARM_MemoryMapParams.bits64;
break;
default:
report_fatal_error("unsupported architecture");
}
break;
default:
report_fatal_error("unsupported operating system");
}
}
C = &(M.getContext());
IRBuilder<> IRB(*C);
IntptrTy = IRB.getIntPtrTy(DL);
OriginTy = IRB.getInt32Ty();
ColdCallWeights = MDBuilder(*C).createBranchWeights(1, 1000);
OriginStoreWeights = MDBuilder(*C).createBranchWeights(1, 1000);
if (!CompileKernel) {
if (TrackOrigins)
M.getOrInsertGlobal("__msan_track_origins", IRB.getInt32Ty(), [&] {
return new GlobalVariable(
M, IRB.getInt32Ty(), true, GlobalValue::WeakODRLinkage,
IRB.getInt32(TrackOrigins), "__msan_track_origins");
});
if (Recover)
M.getOrInsertGlobal("__msan_keep_going", IRB.getInt32Ty(), [&] {
return new GlobalVariable(M, IRB.getInt32Ty(), true,
GlobalValue::WeakODRLinkage,
IRB.getInt32(Recover), "__msan_keep_going");
});
}
}
bool MemorySanitizerLegacyPass::doInitialization(Module &M) {
if (!Options.Kernel)
insertModuleCtor(M);
MSan.emplace(M, Options);
return true;
}
namespace {
/// A helper class that handles instrumentation of VarArg
/// functions on a particular platform.
///
/// Implementations are expected to insert the instrumentation
/// necessary to propagate argument shadow through VarArg function
/// calls. Visit* methods are called during an InstVisitor pass over
/// the function, and should avoid creating new basic blocks. A new
/// instance of this class is created for each instrumented function.
struct VarArgHelper {
virtual ~VarArgHelper() = default;
/// Visit a CallBase.
virtual void visitCallBase(CallBase &CB, IRBuilder<> &IRB) = 0;
/// Visit a va_start call.
virtual void visitVAStartInst(VAStartInst &I) = 0;
/// Visit a va_copy call.
virtual void visitVACopyInst(VACopyInst &I) = 0;
/// Finalize function instrumentation.
///
/// This method is called after visiting all interesting (see above)
/// instructions in a function.
virtual void finalizeInstrumentation() = 0;
};
struct MemorySanitizerVisitor;
} // end anonymous namespace
static VarArgHelper *CreateVarArgHelper(Function &Func, MemorySanitizer &Msan,
MemorySanitizerVisitor &Visitor);
static unsigned TypeSizeToSizeIndex(unsigned TypeSize) {
if (TypeSize <= 8) return 0;
return Log2_32_Ceil((TypeSize + 7) / 8);
}
namespace {
/// This class does all the work for a given function. Store and Load
/// instructions store and load corresponding shadow and origin
/// values. Most instructions propagate shadow from arguments to their
/// return values. Certain instructions (most importantly, BranchInst)
/// test their argument shadow and print reports (with a runtime call) if it's
/// non-zero.
struct MemorySanitizerVisitor : public InstVisitor<MemorySanitizerVisitor> {
Function &F;
MemorySanitizer &MS;
SmallVector<PHINode *, 16> ShadowPHINodes, OriginPHINodes;
ValueMap<Value*, Value*> ShadowMap, OriginMap;
std::unique_ptr<VarArgHelper> VAHelper;
const TargetLibraryInfo *TLI;
Instruction *FnPrologueEnd;
// The following flags disable parts of MSan instrumentation based on
// exclusion list contents and command-line options.
bool InsertChecks;
bool PropagateShadow;
bool PoisonStack;
bool PoisonUndef;
struct ShadowOriginAndInsertPoint {
Value *Shadow;
Value *Origin;
Instruction *OrigIns;
ShadowOriginAndInsertPoint(Value *S, Value *O, Instruction *I)
: Shadow(S), Origin(O), OrigIns(I) {}
};
SmallVector<ShadowOriginAndInsertPoint, 16> InstrumentationList;
bool InstrumentLifetimeStart = ClHandleLifetimeIntrinsics;
SmallSet<AllocaInst *, 16> AllocaSet;
SmallVector<std::pair<IntrinsicInst *, AllocaInst *>, 16> LifetimeStartList;
SmallVector<StoreInst *, 16> StoreList;
MemorySanitizerVisitor(Function &F, MemorySanitizer &MS,
const TargetLibraryInfo &TLI)
: F(F), MS(MS), VAHelper(CreateVarArgHelper(F, MS, *this)), TLI(&TLI) {
bool SanitizeFunction = F.hasFnAttribute(Attribute::SanitizeMemory);
InsertChecks = SanitizeFunction;
PropagateShadow = SanitizeFunction;
PoisonStack = SanitizeFunction && ClPoisonStack;
PoisonUndef = SanitizeFunction && ClPoisonUndef;
// In the presence of unreachable blocks, we may see Phi nodes with
// incoming nodes from such blocks. Since InstVisitor skips unreachable
// blocks, such nodes will not have any shadow value associated with them.
// It's easier to remove unreachable blocks than deal with missing shadow.
removeUnreachableBlocks(F);
MS.initializeCallbacks(*F.getParent());
FnPrologueEnd = IRBuilder<>(F.getEntryBlock().getFirstNonPHI())
.CreateIntrinsic(Intrinsic::donothing, {}, {});
if (MS.CompileKernel) {
IRBuilder<> IRB(FnPrologueEnd);
insertKmsanPrologue(IRB);
}
LLVM_DEBUG(if (!InsertChecks) dbgs()
<< "MemorySanitizer is not inserting checks into '"
<< F.getName() << "'\n");
}
bool isInPrologue(Instruction &I) {
return I.getParent() == FnPrologueEnd->getParent() &&
(&I == FnPrologueEnd || I.comesBefore(FnPrologueEnd));
}
Value *updateOrigin(Value *V, IRBuilder<> &IRB) {
if (MS.TrackOrigins <= 1) return V;
return IRB.CreateCall(MS.MsanChainOriginFn, V);
}
Value *originToIntptr(IRBuilder<> &IRB, Value *Origin) {
const DataLayout &DL = F.getParent()->getDataLayout();
unsigned IntptrSize = DL.getTypeStoreSize(MS.IntptrTy);
if (IntptrSize == kOriginSize) return Origin;
assert(IntptrSize == kOriginSize * 2);
Origin = IRB.CreateIntCast(Origin, MS.IntptrTy, /* isSigned */ false);
return IRB.CreateOr(Origin, IRB.CreateShl(Origin, kOriginSize * 8));
}
/// Fill memory range with the given origin value.
void paintOrigin(IRBuilder<> &IRB, Value *Origin, Value *OriginPtr,
unsigned Size, Align Alignment) {
const DataLayout &DL = F.getParent()->getDataLayout();
const Align IntptrAlignment = DL.getABITypeAlign(MS.IntptrTy);
unsigned IntptrSize = DL.getTypeStoreSize(MS.IntptrTy);
assert(IntptrAlignment >= kMinOriginAlignment);
assert(IntptrSize >= kOriginSize);
unsigned Ofs = 0;
Align CurrentAlignment = Alignment;
if (Alignment >= IntptrAlignment && IntptrSize > kOriginSize) {
Value *IntptrOrigin = originToIntptr(IRB, Origin);
Value *IntptrOriginPtr =
IRB.CreatePointerCast(OriginPtr, PointerType::get(MS.IntptrTy, 0));
for (unsigned i = 0; i < Size / IntptrSize; ++i) {
Value *Ptr = i ? IRB.CreateConstGEP1_32(MS.IntptrTy, IntptrOriginPtr, i)
: IntptrOriginPtr;
IRB.CreateAlignedStore(IntptrOrigin, Ptr, CurrentAlignment);
Ofs += IntptrSize / kOriginSize;
CurrentAlignment = IntptrAlignment;
}
}
for (unsigned i = Ofs; i < (Size + kOriginSize - 1) / kOriginSize; ++i) {
Value *GEP =
i ? IRB.CreateConstGEP1_32(MS.OriginTy, OriginPtr, i) : OriginPtr;
IRB.CreateAlignedStore(Origin, GEP, CurrentAlignment);
CurrentAlignment = kMinOriginAlignment;
}
}
void storeOrigin(IRBuilder<> &IRB, Value *Addr, Value *Shadow, Value *Origin,
Value *OriginPtr, Align Alignment, bool AsCall) {
const DataLayout &DL = F.getParent()->getDataLayout();
const Align OriginAlignment = std::max(kMinOriginAlignment, Alignment);
unsigned StoreSize = DL.getTypeStoreSize(Shadow->getType());
Value *ConvertedShadow = convertShadowToScalar(Shadow, IRB);
if (auto *ConstantShadow = dyn_cast<Constant>(ConvertedShadow)) {
if (ClCheckConstantShadow && !ConstantShadow->isZeroValue())
paintOrigin(IRB, updateOrigin(Origin, IRB), OriginPtr, StoreSize,
OriginAlignment);
return;
}
unsigned TypeSizeInBits = DL.getTypeSizeInBits(ConvertedShadow->getType());
unsigned SizeIndex = TypeSizeToSizeIndex(TypeSizeInBits);
if (AsCall && SizeIndex < kNumberOfAccessSizes && !MS.CompileKernel) {
FunctionCallee Fn = MS.MaybeStoreOriginFn[SizeIndex];
Value *ConvertedShadow2 =
IRB.CreateZExt(ConvertedShadow, IRB.getIntNTy(8 * (1 << SizeIndex)));
IRB.CreateCall(Fn,
{ConvertedShadow2,
IRB.CreatePointerCast(Addr, IRB.getInt8PtrTy()), Origin});
} else {
Value *Cmp = convertToBool(ConvertedShadow, IRB, "_mscmp");
Instruction *CheckTerm = SplitBlockAndInsertIfThen(
Cmp, &*IRB.GetInsertPoint(), false, MS.OriginStoreWeights);
IRBuilder<> IRBNew(CheckTerm);
paintOrigin(IRBNew, updateOrigin(Origin, IRBNew), OriginPtr, StoreSize,
OriginAlignment);
}
}
void materializeStores(bool InstrumentWithCalls) {
for (StoreInst *SI : StoreList) {
IRBuilder<> IRB(SI);
Value *Val = SI->getValueOperand();
Value *Addr = SI->getPointerOperand();
Value *Shadow = SI->isAtomic() ? getCleanShadow(Val) : getShadow(Val);
Value *ShadowPtr, *OriginPtr;
Type *ShadowTy = Shadow->getType();
const Align Alignment = assumeAligned(SI->getAlignment());
const Align OriginAlignment = std::max(kMinOriginAlignment, Alignment);
std::tie(ShadowPtr, OriginPtr) =
getShadowOriginPtr(Addr, IRB, ShadowTy, Alignment, /*isStore*/ true);
StoreInst *NewSI = IRB.CreateAlignedStore(Shadow, ShadowPtr, Alignment);
LLVM_DEBUG(dbgs() << " STORE: " << *NewSI << "\n");
(void)NewSI;
if (SI->isAtomic())
SI->setOrdering(addReleaseOrdering(SI->getOrdering()));
if (MS.TrackOrigins && !SI->isAtomic())
storeOrigin(IRB, Addr, Shadow, getOrigin(Val), OriginPtr,
OriginAlignment, InstrumentWithCalls);
}
}
/// Helper function to insert a warning at IRB's current insert point.
void insertWarningFn(IRBuilder<> &IRB, Value *Origin) {
if (!Origin)
Origin = (Value *)IRB.getInt32(0);
assert(Origin->getType()->isIntegerTy());
IRB.CreateCall(MS.WarningFn, Origin)->setCannotMerge();
// FIXME: Insert UnreachableInst if !MS.Recover?
// This may invalidate some of the following checks and needs to be done
// at the very end.
}
void materializeOneCheck(Instruction *OrigIns, Value *Shadow, Value *Origin,
bool AsCall) {
IRBuilder<> IRB(OrigIns);
LLVM_DEBUG(dbgs() << " SHAD0 : " << *Shadow << "\n");
Value *ConvertedShadow = convertShadowToScalar(Shadow, IRB);
LLVM_DEBUG(dbgs() << " SHAD1 : " << *ConvertedShadow << "\n");
if (auto *ConstantShadow = dyn_cast<Constant>(ConvertedShadow)) {
if (ClCheckConstantShadow && !ConstantShadow->isZeroValue()) {
insertWarningFn(IRB, Origin);
}
return;
}
const DataLayout &DL = OrigIns->getModule()->getDataLayout();
unsigned TypeSizeInBits = DL.getTypeSizeInBits(ConvertedShadow->getType());
unsigned SizeIndex = TypeSizeToSizeIndex(TypeSizeInBits);
if (AsCall && SizeIndex < kNumberOfAccessSizes && !MS.CompileKernel) {
FunctionCallee Fn = MS.MaybeWarningFn[SizeIndex];
Value *ConvertedShadow2 =
IRB.CreateZExt(ConvertedShadow, IRB.getIntNTy(8 * (1 << SizeIndex)));
IRB.CreateCall(Fn, {ConvertedShadow2, MS.TrackOrigins && Origin
? Origin
: (Value *)IRB.getInt32(0)});
} else {
Value *Cmp = convertToBool(ConvertedShadow, IRB, "_mscmp");
Instruction *CheckTerm = SplitBlockAndInsertIfThen(
Cmp, OrigIns,
/* Unreachable */ !MS.Recover, MS.ColdCallWeights);
IRB.SetInsertPoint(CheckTerm);
insertWarningFn(IRB, Origin);
LLVM_DEBUG(dbgs() << " CHECK: " << *Cmp << "\n");
}
}
void materializeChecks(bool InstrumentWithCalls) {
for (const auto &ShadowData : InstrumentationList) {
Instruction *OrigIns = ShadowData.OrigIns;
Value *Shadow = ShadowData.Shadow;
Value *Origin = ShadowData.Origin;
materializeOneCheck(OrigIns, Shadow, Origin, InstrumentWithCalls);
}
LLVM_DEBUG(dbgs() << "DONE:\n" << F);
}
// Returns the last instruction in the new prologue
void insertKmsanPrologue(IRBuilder<> &IRB) {
Value *ContextState = IRB.CreateCall(MS.MsanGetContextStateFn, {});
Constant *Zero = IRB.getInt32(0);
MS.ParamTLS = IRB.CreateGEP(MS.MsanContextStateTy, ContextState,
{Zero, IRB.getInt32(0)}, "param_shadow");
MS.RetvalTLS = IRB.CreateGEP(MS.MsanContextStateTy, ContextState,
{Zero, IRB.getInt32(1)}, "retval_shadow");
MS.VAArgTLS = IRB.CreateGEP(MS.MsanContextStateTy, ContextState,
{Zero, IRB.getInt32(2)}, "va_arg_shadow");
MS.VAArgOriginTLS = IRB.CreateGEP(MS.MsanContextStateTy, ContextState,
{Zero, IRB.getInt32(3)}, "va_arg_origin");
MS.VAArgOverflowSizeTLS =
IRB.CreateGEP(MS.MsanContextStateTy, ContextState,
{Zero, IRB.getInt32(4)}, "va_arg_overflow_size");
MS.ParamOriginTLS = IRB.CreateGEP(MS.MsanContextStateTy, ContextState,
{Zero, IRB.getInt32(5)}, "param_origin");
MS.RetvalOriginTLS =
IRB.CreateGEP(MS.MsanContextStateTy, ContextState,
{Zero, IRB.getInt32(6)}, "retval_origin");
}
/// Add MemorySanitizer instrumentation to a function.
bool runOnFunction() {
// Iterate all BBs in depth-first order and create shadow instructions
// for all instructions (where applicable).
// For PHI nodes we create dummy shadow PHIs which will be finalized later.
for (BasicBlock *BB : depth_first(FnPrologueEnd->getParent()))
visit(*BB);
// Finalize PHI nodes.
for (PHINode *PN : ShadowPHINodes) {
PHINode *PNS = cast<PHINode>(getShadow(PN));
PHINode *PNO = MS.TrackOrigins ? cast<PHINode>(getOrigin(PN)) : nullptr;
size_t NumValues = PN->getNumIncomingValues();
for (size_t v = 0; v < NumValues; v++) {
PNS->addIncoming(getShadow(PN, v), PN->getIncomingBlock(v));
if (PNO) PNO->addIncoming(getOrigin(PN, v), PN->getIncomingBlock(v));
}
}
VAHelper->finalizeInstrumentation();
// Poison llvm.lifetime.start intrinsics, if we haven't fallen back to
// instrumenting only allocas.
if (InstrumentLifetimeStart) {
for (auto Item : LifetimeStartList) {
instrumentAlloca(*Item.second, Item.first);
AllocaSet.erase(Item.second);
}
}
// Poison the allocas for which we didn't instrument the corresponding
// lifetime intrinsics.
for (AllocaInst *AI : AllocaSet)
instrumentAlloca(*AI);
bool InstrumentWithCalls = ClInstrumentationWithCallThreshold >= 0 &&
InstrumentationList.size() + StoreList.size() >
(unsigned)ClInstrumentationWithCallThreshold;
// Insert shadow value checks.
materializeChecks(InstrumentWithCalls);
// Delayed instrumentation of StoreInst.
// This may not add new address checks.
materializeStores(InstrumentWithCalls);
return true;
}
/// Compute the shadow type that corresponds to a given Value.
Type *getShadowTy(Value *V) {
return getShadowTy(V->getType());
}
/// Compute the shadow type that corresponds to a given Type.
Type *getShadowTy(Type *OrigTy) {
if (!OrigTy->isSized()) {
return nullptr;
}
// For integer type, shadow is the same as the original type.
// This may return weird-sized types like i1.
if (IntegerType *IT = dyn_cast<IntegerType>(OrigTy))
return IT;
const DataLayout &DL = F.getParent()->getDataLayout();
if (VectorType *VT = dyn_cast<VectorType>(OrigTy)) {
uint32_t EltSize = DL.getTypeSizeInBits(VT->getElementType());
return FixedVectorType::get(IntegerType::get(*MS.C, EltSize),
cast<FixedVectorType>(VT)->getNumElements());
}
if (ArrayType *AT = dyn_cast<ArrayType>(OrigTy)) {
return ArrayType::get(getShadowTy(AT->getElementType()),
AT->getNumElements());
}
if (StructType *ST = dyn_cast<StructType>(OrigTy)) {
SmallVector<Type*, 4> Elements;
for (unsigned i = 0, n = ST->getNumElements(); i < n; i++)
Elements.push_back(getShadowTy(ST->getElementType(i)));
StructType *Res = StructType::get(*MS.C, Elements, ST->isPacked());
LLVM_DEBUG(dbgs() << "getShadowTy: " << *ST << " ===> " << *Res << "\n");
return Res;
}
uint32_t TypeSize = DL.getTypeSizeInBits(OrigTy);
return IntegerType::get(*MS.C, TypeSize);
}
/// Flatten a vector type.
Type *getShadowTyNoVec(Type *ty) {
if (VectorType *vt = dyn_cast<VectorType>(ty))
return IntegerType::get(*MS.C,
vt->getPrimitiveSizeInBits().getFixedSize());
return ty;
}
/// Extract combined shadow of struct elements as a bool
Value *collapseStructShadow(StructType *Struct, Value *Shadow,
IRBuilder<> &IRB) {
Value *FalseVal = IRB.getIntN(/* width */ 1, /* value */ 0);
Value *Aggregator = FalseVal;
for (unsigned Idx = 0; Idx < Struct->getNumElements(); Idx++) {
// Combine by ORing together each element's bool shadow
Value *ShadowItem = IRB.CreateExtractValue(Shadow, Idx);
Value *ShadowInner = convertShadowToScalar(ShadowItem, IRB);
Value *ShadowBool = convertToBool(ShadowInner, IRB);
if (Aggregator != FalseVal)
Aggregator = IRB.CreateOr(Aggregator, ShadowBool);
else
Aggregator = ShadowBool;
}
return Aggregator;
}
// Extract combined shadow of array elements
Value *collapseArrayShadow(ArrayType *Array, Value *Shadow,
IRBuilder<> &IRB) {
if (!Array->getNumElements())
return IRB.getIntN(/* width */ 1, /* value */ 0);
Value *FirstItem = IRB.CreateExtractValue(Shadow, 0);
Value *Aggregator = convertShadowToScalar(FirstItem, IRB);
for (unsigned Idx = 1; Idx < Array->getNumElements(); Idx++) {
Value *ShadowItem = IRB.CreateExtractValue(Shadow, Idx);
Value *ShadowInner = convertShadowToScalar(ShadowItem, IRB);
Aggregator = IRB.CreateOr(Aggregator, ShadowInner);
}
return Aggregator;
}
/// Convert a shadow value to it's flattened variant. The resulting
/// shadow may not necessarily have the same bit width as the input
/// value, but it will always be comparable to zero.
Value *convertShadowToScalar(Value *V, IRBuilder<> &IRB) {
if (StructType *Struct = dyn_cast<StructType>(V->getType()))
return collapseStructShadow(Struct, V, IRB);
if (ArrayType *Array = dyn_cast<ArrayType>(V->getType()))
return collapseArrayShadow(Array, V, IRB);
Type *Ty = V->getType();
Type *NoVecTy = getShadowTyNoVec(Ty);
if (Ty == NoVecTy) return V;
return IRB.CreateBitCast(V, NoVecTy);
}
// Convert a scalar value to an i1 by comparing with 0
Value *convertToBool(Value *V, IRBuilder<> &IRB, const Twine &name = "") {
Type *VTy = V->getType();
assert(VTy->isIntegerTy());
if (VTy->getIntegerBitWidth() == 1)
// Just converting a bool to a bool, so do nothing.
return V;
return IRB.CreateICmpNE(V, ConstantInt::get(VTy, 0), name);
}
/// Compute the integer shadow offset that corresponds to a given
/// application address.
///
/// Offset = (Addr & ~AndMask) ^ XorMask
Value *getShadowPtrOffset(Value *Addr, IRBuilder<> &IRB) {
Value *OffsetLong = IRB.CreatePointerCast(Addr, MS.IntptrTy);
uint64_t AndMask = MS.MapParams->AndMask;
if (AndMask)
OffsetLong =
IRB.CreateAnd(OffsetLong, ConstantInt::get(MS.IntptrTy, ~AndMask));
uint64_t XorMask = MS.MapParams->XorMask;
if (XorMask)
OffsetLong =
IRB.CreateXor(OffsetLong, ConstantInt::get(MS.IntptrTy, XorMask));
return OffsetLong;
}
/// Compute the shadow and origin addresses corresponding to a given
/// application address.
///
/// Shadow = ShadowBase + Offset
/// Origin = (OriginBase + Offset) & ~3ULL
std::pair<Value *, Value *>
getShadowOriginPtrUserspace(Value *Addr, IRBuilder<> &IRB, Type *ShadowTy,
MaybeAlign Alignment) {
Value *ShadowOffset = getShadowPtrOffset(Addr, IRB);
Value *ShadowLong = ShadowOffset;
uint64_t ShadowBase = MS.MapParams->ShadowBase;
if (ShadowBase != 0) {
ShadowLong =
IRB.CreateAdd(ShadowLong,
ConstantInt::get(MS.IntptrTy, ShadowBase));
}
Value *ShadowPtr =
IRB.CreateIntToPtr(ShadowLong, PointerType::get(ShadowTy, 0));
Value *OriginPtr = nullptr;
if (MS.TrackOrigins) {
Value *OriginLong = ShadowOffset;
uint64_t OriginBase = MS.MapParams->OriginBase;
if (OriginBase != 0)
OriginLong = IRB.CreateAdd(OriginLong,
ConstantInt::get(MS.IntptrTy, OriginBase));
if (!Alignment || *Alignment < kMinOriginAlignment) {
uint64_t Mask = kMinOriginAlignment.value() - 1;
OriginLong =
IRB.CreateAnd(OriginLong, ConstantInt::get(MS.IntptrTy, ~Mask));
}
OriginPtr =
IRB.CreateIntToPtr(OriginLong, PointerType::get(MS.OriginTy, 0));
}
return std::make_pair(ShadowPtr, OriginPtr);
}
std::pair<Value *, Value *> getShadowOriginPtrKernel(Value *Addr,
IRBuilder<> &IRB,
Type *ShadowTy,
bool isStore) {
Value *ShadowOriginPtrs;
const DataLayout &DL = F.getParent()->getDataLayout();
int Size = DL.getTypeStoreSize(ShadowTy);
FunctionCallee Getter = MS.getKmsanShadowOriginAccessFn(isStore, Size);
Value *AddrCast =
IRB.CreatePointerCast(Addr, PointerType::get(IRB.getInt8Ty(), 0));
if (Getter) {
ShadowOriginPtrs = IRB.CreateCall(Getter, AddrCast);
} else {
Value *SizeVal = ConstantInt::get(MS.IntptrTy, Size);
ShadowOriginPtrs = IRB.CreateCall(isStore ? MS.MsanMetadataPtrForStoreN
: MS.MsanMetadataPtrForLoadN,
{AddrCast, SizeVal});
}
Value *ShadowPtr = IRB.CreateExtractValue(ShadowOriginPtrs, 0);
ShadowPtr = IRB.CreatePointerCast(ShadowPtr, PointerType::get(ShadowTy, 0));
Value *OriginPtr = IRB.CreateExtractValue(ShadowOriginPtrs, 1);
return std::make_pair(ShadowPtr, OriginPtr);
}
std::pair<Value *, Value *> getShadowOriginPtr(Value *Addr, IRBuilder<> &IRB,
Type *ShadowTy,
MaybeAlign Alignment,
bool isStore) {
if (MS.CompileKernel)
return getShadowOriginPtrKernel(Addr, IRB, ShadowTy, isStore);
return getShadowOriginPtrUserspace(Addr, IRB, ShadowTy, Alignment);
}
/// Compute the shadow address for a given function argument.
///
/// Shadow = ParamTLS+ArgOffset.
Value *getShadowPtrForArgument(Value *A, IRBuilder<> &IRB,
int ArgOffset) {
Value *Base = IRB.CreatePointerCast(MS.ParamTLS, MS.IntptrTy);
if (ArgOffset)
Base = IRB.CreateAdd(Base, ConstantInt::get(MS.IntptrTy, ArgOffset));
return IRB.CreateIntToPtr(Base, PointerType::get(getShadowTy(A), 0),
"_msarg");
}
/// Compute the origin address for a given function argument.
Value *getOriginPtrForArgument(Value *A, IRBuilder<> &IRB,
int ArgOffset) {
if (!MS.TrackOrigins)
return nullptr;
Value *Base = IRB.CreatePointerCast(MS.ParamOriginTLS, MS.IntptrTy);
if (ArgOffset)
Base = IRB.CreateAdd(Base, ConstantInt::get(MS.IntptrTy, ArgOffset));
return IRB.CreateIntToPtr(Base, PointerType::get(MS.OriginTy, 0),
"_msarg_o");
}
/// Compute the shadow address for a retval.
Value *getShadowPtrForRetval(Value *A, IRBuilder<> &IRB) {
return IRB.CreatePointerCast(MS.RetvalTLS,
PointerType::get(getShadowTy(A), 0),
"_msret");
}
/// Compute the origin address for a retval.
Value *getOriginPtrForRetval(IRBuilder<> &IRB) {
// We keep a single origin for the entire retval. Might be too optimistic.
return MS.RetvalOriginTLS;
}
/// Set SV to be the shadow value for V.
void setShadow(Value *V, Value *SV) {
assert(!ShadowMap.count(V) && "Values may only have one shadow");
ShadowMap[V] = PropagateShadow ? SV : getCleanShadow(V);
}
/// Set Origin to be the origin value for V.
void setOrigin(Value *V, Value *Origin) {
if (!MS.TrackOrigins) return;
assert(!OriginMap.count(V) && "Values may only have one origin");
LLVM_DEBUG(dbgs() << "ORIGIN: " << *V << " ==> " << *Origin << "\n");
OriginMap[V] = Origin;
}
Constant *getCleanShadow(Type *OrigTy) {
Type *ShadowTy = getShadowTy(OrigTy);
if (!ShadowTy)
return nullptr;
return Constant::getNullValue(ShadowTy);
}
/// Create a clean shadow value for a given value.
///
/// Clean shadow (all zeroes) means all bits of the value are defined
/// (initialized).
Constant *getCleanShadow(Value *V) {
return getCleanShadow(V->getType());
}
/// Create a dirty shadow of a given shadow type.
Constant *getPoisonedShadow(Type *ShadowTy) {
assert(ShadowTy);
if (isa<IntegerType>(ShadowTy) || isa<VectorType>(ShadowTy))
return Constant::getAllOnesValue(ShadowTy);
if (ArrayType *AT = dyn_cast<ArrayType>(ShadowTy)) {
SmallVector<Constant *, 4> Vals(AT->getNumElements(),
getPoisonedShadow(AT->getElementType()));
return ConstantArray::get(AT, Vals);
}
if (StructType *ST = dyn_cast<StructType>(ShadowTy)) {
SmallVector<Constant *, 4> Vals;
for (unsigned i = 0, n = ST->getNumElements(); i < n; i++)
Vals.push_back(getPoisonedShadow(ST->getElementType(i)));
return ConstantStruct::get(ST, Vals);
}
llvm_unreachable("Unexpected shadow type");
}
/// Create a dirty shadow for a given value.
Constant *getPoisonedShadow(Value *V) {
Type *ShadowTy = getShadowTy(V);
if (!ShadowTy)
return nullptr;
return getPoisonedShadow(ShadowTy);
}
/// Create a clean (zero) origin.
Value *getCleanOrigin() {
return Constant::getNullValue(MS.OriginTy);
}
/// Get the shadow value for a given Value.
///
/// This function either returns the value set earlier with setShadow,
/// or extracts if from ParamTLS (for function arguments).
Value *getShadow(Value *V) {
if (!PropagateShadow) return getCleanShadow(V);
if (Instruction *I = dyn_cast<Instruction>(V)) {
if (I->getMetadata("nosanitize"))
return getCleanShadow(V);
// For instructions the shadow is already stored in the map.
Value *Shadow = ShadowMap[V];
if (!Shadow) {
LLVM_DEBUG(dbgs() << "No shadow: " << *V << "\n" << *(I->getParent()));
(void)I;
assert(Shadow && "No shadow for a value");
}
return Shadow;
}
if (UndefValue *U = dyn_cast<UndefValue>(V)) {
Value *AllOnes = PoisonUndef ? getPoisonedShadow(V) : getCleanShadow(V);
LLVM_DEBUG(dbgs() << "Undef: " << *U << " ==> " << *AllOnes << "\n");
(void)U;
return AllOnes;
}
if (Argument *A = dyn_cast<Argument>(V)) {
// For arguments we compute the shadow on demand and store it in the map.
Value **ShadowPtr = &ShadowMap[V];
if (*ShadowPtr)
return *ShadowPtr;
Function *F = A->getParent();
IRBuilder<> EntryIRB(FnPrologueEnd);
unsigned ArgOffset = 0;
const DataLayout &DL = F->getParent()->getDataLayout();
for (auto &FArg : F->args()) {
if (!FArg.getType()->isSized()) {
LLVM_DEBUG(dbgs() << "Arg is not sized\n");
continue;
}
bool FArgByVal = FArg.hasByValAttr();
bool FArgNoUndef = FArg.hasAttribute(Attribute::NoUndef);
bool FArgEagerCheck = ClEagerChecks && !FArgByVal && FArgNoUndef;
unsigned Size =
FArg.hasByValAttr()
? DL.getTypeAllocSize(FArg.getParamByValType())
: DL.getTypeAllocSize(FArg.getType());
if (A == &FArg) {
bool Overflow = ArgOffset + Size > kParamTLSSize;
if (FArgEagerCheck) {
*ShadowPtr = getCleanShadow(V);
setOrigin(A, getCleanOrigin());
continue;
} else if (FArgByVal) {
Value *Base = getShadowPtrForArgument(&FArg, EntryIRB, ArgOffset);
// ByVal pointer itself has clean shadow. We copy the actual
// argument shadow to the underlying memory.
// Figure out maximal valid memcpy alignment.
const Align ArgAlign = DL.getValueOrABITypeAlignment(
MaybeAlign(FArg.getParamAlignment()), FArg.getParamByValType());
Value *CpShadowPtr =
getShadowOriginPtr(V, EntryIRB, EntryIRB.getInt8Ty(), ArgAlign,
/*isStore*/ true)
.first;
// TODO(glider): need to copy origins.
if (Overflow) {
// ParamTLS overflow.
EntryIRB.CreateMemSet(
CpShadowPtr, Constant::getNullValue(EntryIRB.getInt8Ty()),
Size, ArgAlign);
} else {
const Align CopyAlign = std::min(ArgAlign, kShadowTLSAlignment);
Value *Cpy = EntryIRB.CreateMemCpy(CpShadowPtr, CopyAlign, Base,
CopyAlign, Size);
LLVM_DEBUG(dbgs() << " ByValCpy: " << *Cpy << "\n");
(void)Cpy;
}
*ShadowPtr = getCleanShadow(V);
} else {
// Shadow over TLS
Value *Base = getShadowPtrForArgument(&FArg, EntryIRB, ArgOffset);
if (Overflow) {
// ParamTLS overflow.
*ShadowPtr = getCleanShadow(V);
} else {
*ShadowPtr = EntryIRB.CreateAlignedLoad(getShadowTy(&FArg), Base,
kShadowTLSAlignment);
}
}
LLVM_DEBUG(dbgs()
<< " ARG: " << FArg << " ==> " << **ShadowPtr << "\n");
if (MS.TrackOrigins && !Overflow) {
Value *OriginPtr =
getOriginPtrForArgument(&FArg, EntryIRB, ArgOffset);
setOrigin(A, EntryIRB.CreateLoad(MS.OriginTy, OriginPtr));
} else {
setOrigin(A, getCleanOrigin());
}
break;
}
if (!FArgEagerCheck)
ArgOffset += alignTo(Size, kShadowTLSAlignment);
}
assert(*ShadowPtr && "Could not find shadow for an argument");
return *ShadowPtr;
}
// For everything else the shadow is zero.
return getCleanShadow(V);
}
/// Get the shadow for i-th argument of the instruction I.
Value *getShadow(Instruction *I, int i) {
return getShadow(I->getOperand(i));
}
/// Get the origin for a value.
Value *getOrigin(Value *V) {
if (!MS.TrackOrigins) return nullptr;
if (!PropagateShadow) return getCleanOrigin();
if (isa<Constant>(V)) return getCleanOrigin();
assert((isa<Instruction>(V) || isa<Argument>(V)) &&
"Unexpected value type in getOrigin()");
if (Instruction *I = dyn_cast<Instruction>(V)) {
if (I->getMetadata("nosanitize"))
return getCleanOrigin();
}
Value *Origin = OriginMap[V];
assert(Origin && "Missing origin");
return Origin;
}
/// Get the origin for i-th argument of the instruction I.
Value *getOrigin(Instruction *I, int i) {
return getOrigin(I->getOperand(i));
}
/// Remember the place where a shadow check should be inserted.
///
/// This location will be later instrumented with a check that will print a
/// UMR warning in runtime if the shadow value is not 0.
void insertShadowCheck(Value *Shadow, Value *Origin, Instruction *OrigIns) {
assert(Shadow);
if (!InsertChecks) return;
#ifndef NDEBUG
Type *ShadowTy = Shadow->getType();
assert((isa<IntegerType>(ShadowTy) || isa<VectorType>(ShadowTy) ||
isa<StructType>(ShadowTy) || isa<ArrayType>(ShadowTy)) &&
"Can only insert checks for integer, vector, and aggregate shadow "
"types");
#endif
InstrumentationList.push_back(
ShadowOriginAndInsertPoint(Shadow, Origin, OrigIns));
}
/// Remember the place where a shadow check should be inserted.
///
/// This location will be later instrumented with a check that will print a
/// UMR warning in runtime if the value is not fully defined.
void insertShadowCheck(Value *Val, Instruction *OrigIns) {
assert(Val);
Value *Shadow, *Origin;
if (ClCheckConstantShadow) {
Shadow = getShadow(Val);
if (!Shadow) return;
Origin = getOrigin(Val);
} else {
Shadow = dyn_cast_or_null<Instruction>(getShadow(Val));
if (!Shadow) return;
Origin = dyn_cast_or_null<Instruction>(getOrigin(Val));
}
insertShadowCheck(Shadow, Origin, OrigIns);
}
AtomicOrdering addReleaseOrdering(AtomicOrdering a) {
switch (a) {
case AtomicOrdering::NotAtomic:
return AtomicOrdering::NotAtomic;
case AtomicOrdering::Unordered:
case AtomicOrdering::Monotonic:
case AtomicOrdering::Release:
return AtomicOrdering::Release;
case AtomicOrdering::Acquire:
case AtomicOrdering::AcquireRelease:
return AtomicOrdering::AcquireRelease;
case AtomicOrdering::SequentiallyConsistent:
return AtomicOrdering::SequentiallyConsistent;
}
llvm_unreachable("Unknown ordering");
}
Value *makeAddReleaseOrderingTable(IRBuilder<> &IRB) {
constexpr int NumOrderings = (int)AtomicOrderingCABI::seq_cst + 1;
uint32_t OrderingTable[NumOrderings] = {};
OrderingTable[(int)AtomicOrderingCABI::relaxed] =
OrderingTable[(int)AtomicOrderingCABI::release] =
(int)AtomicOrderingCABI::release;
OrderingTable[(int)AtomicOrderingCABI::consume] =
OrderingTable[(int)AtomicOrderingCABI::acquire] =
OrderingTable[(int)AtomicOrderingCABI::acq_rel] =
(int)AtomicOrderingCABI::acq_rel;
OrderingTable[(int)AtomicOrderingCABI::seq_cst] =
(int)AtomicOrderingCABI::seq_cst;
return ConstantDataVector::get(IRB.getContext(),
makeArrayRef(OrderingTable, NumOrderings));
}
AtomicOrdering addAcquireOrdering(AtomicOrdering a) {
switch (a) {
case AtomicOrdering::NotAtomic:
return AtomicOrdering::NotAtomic;
case AtomicOrdering::Unordered:
case AtomicOrdering::Monotonic:
case AtomicOrdering::Acquire:
return AtomicOrdering::Acquire;
case AtomicOrdering::Release:
case AtomicOrdering::AcquireRelease:
return AtomicOrdering::AcquireRelease;
case AtomicOrdering::SequentiallyConsistent:
return AtomicOrdering::SequentiallyConsistent;
}
llvm_unreachable("Unknown ordering");
}
Value *makeAddAcquireOrderingTable(IRBuilder<> &IRB) {
constexpr int NumOrderings = (int)AtomicOrderingCABI::seq_cst + 1;
uint32_t OrderingTable[NumOrderings] = {};
OrderingTable[(int)AtomicOrderingCABI::relaxed] =
OrderingTable[(int)AtomicOrderingCABI::acquire] =
OrderingTable[(int)AtomicOrderingCABI::consume] =
(int)AtomicOrderingCABI::acquire;
OrderingTable[(int)AtomicOrderingCABI::release] =
OrderingTable[(int)AtomicOrderingCABI::acq_rel] =
(int)AtomicOrderingCABI::acq_rel;
OrderingTable[(int)AtomicOrderingCABI::seq_cst] =
(int)AtomicOrderingCABI::seq_cst;
return ConstantDataVector::get(IRB.getContext(),
makeArrayRef(OrderingTable, NumOrderings));
}
// ------------------- Visitors.
using InstVisitor<MemorySanitizerVisitor>::visit;
void visit(Instruction &I) {
if (I.getMetadata("nosanitize"))
return;
// Don't want to visit if we're in the prologue
if (isInPrologue(I))
return;
InstVisitor<MemorySanitizerVisitor>::visit(I);
}
/// Instrument LoadInst
///
/// Loads the corresponding shadow and (optionally) origin.
/// Optionally, checks that the load address is fully defined.
void visitLoadInst(LoadInst &I) {
assert(I.getType()->isSized() && "Load type must have size");
assert(!I.getMetadata("nosanitize"));
IRBuilder<> IRB(I.getNextNode());
Type *ShadowTy = getShadowTy(&I);
Value *Addr = I.getPointerOperand();
Value *ShadowPtr = nullptr, *OriginPtr = nullptr;
const Align Alignment = assumeAligned(I.getAlignment());
if (PropagateShadow) {
std::tie(ShadowPtr, OriginPtr) =
getShadowOriginPtr(Addr, IRB, ShadowTy, Alignment, /*isStore*/ false);
setShadow(&I,
IRB.CreateAlignedLoad(ShadowTy, ShadowPtr, Alignment, "_msld"));
} else {
setShadow(&I, getCleanShadow(&I));
}
if (ClCheckAccessAddress)
insertShadowCheck(I.getPointerOperand(), &I);
if (I.isAtomic())
I.setOrdering(addAcquireOrdering(I.getOrdering()));
if (MS.TrackOrigins) {
if (PropagateShadow) {
const Align OriginAlignment = std::max(kMinOriginAlignment, Alignment);
setOrigin(
&I, IRB.CreateAlignedLoad(MS.OriginTy, OriginPtr, OriginAlignment));
} else {
setOrigin(&I, getCleanOrigin());
}
}
}
/// Instrument StoreInst
///
/// Stores the corresponding shadow and (optionally) origin.
/// Optionally, checks that the store address is fully defined.
void visitStoreInst(StoreInst &I) {
StoreList.push_back(&I);
if (ClCheckAccessAddress)
insertShadowCheck(I.getPointerOperand(), &I);
}
void handleCASOrRMW(Instruction &I) {
assert(isa<AtomicRMWInst>(I) || isa<AtomicCmpXchgInst>(I));
IRBuilder<> IRB(&I);
Value *Addr = I.getOperand(0);
Value *Val = I.getOperand(1);
Value *ShadowPtr = getShadowOriginPtr(Addr, IRB, Val->getType(), Align(1),
/*isStore*/ true)
.first;
if (ClCheckAccessAddress)
insertShadowCheck(Addr, &I);
// Only test the conditional argument of cmpxchg instruction.
// The other argument can potentially be uninitialized, but we can not
// detect this situation reliably without possible false positives.
if (isa<AtomicCmpXchgInst>(I))
insertShadowCheck(Val, &I);
IRB.CreateStore(getCleanShadow(Val), ShadowPtr);
setShadow(&I, getCleanShadow(&I));
setOrigin(&I, getCleanOrigin());
}
void visitAtomicRMWInst(AtomicRMWInst &I) {
handleCASOrRMW(I);
I.setOrdering(addReleaseOrdering(I.getOrdering()));
}
void visitAtomicCmpXchgInst(AtomicCmpXchgInst &I) {
handleCASOrRMW(I);
I.setSuccessOrdering(addReleaseOrdering(I.getSuccessOrdering()));
}
// Vector manipulation.
void visitExtractElementInst(ExtractElementInst &I) {
insertShadowCheck(I.getOperand(1), &I);
IRBuilder<> IRB(&I);
setShadow(&I, IRB.CreateExtractElement(getShadow(&I, 0), I.getOperand(1),
"_msprop"));
setOrigin(&I, getOrigin(&I, 0));
}
void visitInsertElementInst(InsertElementInst &I) {
insertShadowCheck(I.getOperand(2), &I);
IRBuilder<> IRB(&I);
setShadow(&I, IRB.CreateInsertElement(getShadow(&I, 0), getShadow(&I, 1),
I.getOperand(2), "_msprop"));
setOriginForNaryOp(I);
}
void visitShuffleVectorInst(ShuffleVectorInst &I) {
IRBuilder<> IRB(&I);
setShadow(&I, IRB.CreateShuffleVector(getShadow(&I, 0), getShadow(&I, 1),
I.getShuffleMask(), "_msprop"));
setOriginForNaryOp(I);
}
// Casts.
void visitSExtInst(SExtInst &I) {
IRBuilder<> IRB(&I);
setShadow(&I, IRB.CreateSExt(getShadow(&I, 0), I.getType(), "_msprop"));
setOrigin(&I, getOrigin(&I, 0));
}
void visitZExtInst(ZExtInst &I) {
IRBuilder<> IRB(&I);
setShadow(&I, IRB.CreateZExt(getShadow(&I, 0), I.getType(), "_msprop"));
setOrigin(&I, getOrigin(&I, 0));
}
void visitTruncInst(TruncInst &I) {
IRBuilder<> IRB(&I);
setShadow(&I, IRB.CreateTrunc(getShadow(&I, 0), I.getType(), "_msprop"));
setOrigin(&I, getOrigin(&I, 0));
}
void visitBitCastInst(BitCastInst &I) {
// Special case: if this is the bitcast (there is exactly 1 allowed) between
// a musttail call and a ret, don't instrument. New instructions are not
// allowed after a musttail call.
if (auto *CI = dyn_cast<CallInst>(I.getOperand(0)))
if (CI->isMustTailCall())
return;
IRBuilder<> IRB(&I);
setShadow(&I, IRB.CreateBitCast(getShadow(&I, 0), getShadowTy(&I)));
setOrigin(&I, getOrigin(&I, 0));
}
void visitPtrToIntInst(PtrToIntInst &I) {
IRBuilder<> IRB(&I);
setShadow(&I, IRB.CreateIntCast(getShadow(&I, 0), getShadowTy(&I), false,
"_msprop_ptrtoint"));
setOrigin(&I, getOrigin(&I, 0));
}
void visitIntToPtrInst(IntToPtrInst &I) {
IRBuilder<> IRB(&I);
setShadow(&I, IRB.CreateIntCast(getShadow(&I, 0), getShadowTy(&I), false,
"_msprop_inttoptr"));
setOrigin(&I, getOrigin(&I, 0));
}
void visitFPToSIInst(CastInst& I) { handleShadowOr(I); }
void visitFPToUIInst(CastInst& I) { handleShadowOr(I); }
void visitSIToFPInst(CastInst& I) { handleShadowOr(I); }
void visitUIToFPInst(CastInst& I) { handleShadowOr(I); }
void visitFPExtInst(CastInst& I) { handleShadowOr(I); }
void visitFPTruncInst(CastInst& I) { handleShadowOr(I); }
/// Propagate shadow for bitwise AND.
///
/// This code is exact, i.e. if, for example, a bit in the left argument
/// is defined and 0, then neither the value not definedness of the
/// corresponding bit in B don't affect the resulting shadow.
void visitAnd(BinaryOperator &I) {
IRBuilder<> IRB(&I);
// "And" of 0 and a poisoned value results in unpoisoned value.
// 1&1 => 1; 0&1 => 0; p&1 => p;
// 1&0 => 0; 0&0 => 0; p&0 => 0;
// 1&p => p; 0&p => 0; p&p => p;
// S = (S1 & S2) | (V1 & S2) | (S1 & V2)
Value *S1 = getShadow(&I, 0);
Value *S2 = getShadow(&I, 1);
Value *V1 = I.getOperand(0);
Value *V2 = I.getOperand(1);
if (V1->getType() != S1->getType()) {
V1 = IRB.CreateIntCast(V1, S1->getType(), false);
V2 = IRB.CreateIntCast(V2, S2->getType(), false);
}
Value *S1S2 = IRB.CreateAnd(S1, S2);
Value *V1S2 = IRB.CreateAnd(V1, S2);
Value *S1V2 = IRB.CreateAnd(S1, V2);
setShadow(&I, IRB.CreateOr({S1S2, V1S2, S1V2}));
setOriginForNaryOp(I);
}
void visitOr(BinaryOperator &I) {
IRBuilder<> IRB(&I);
// "Or" of 1 and a poisoned value results in unpoisoned value.
// 1|1 => 1; 0|1 => 1; p|1 => 1;
// 1|0 => 1; 0|0 => 0; p|0 => p;
// 1|p => 1; 0|p => p; p|p => p;
// S = (S1 & S2) | (~V1 & S2) | (S1 & ~V2)
Value *S1 = getShadow(&I, 0);
Value *S2 = getShadow(&I, 1);
Value *V1 = IRB.CreateNot(I.getOperand(0));
Value *V2 = IRB.CreateNot(I.getOperand(1));
if (V1->getType() != S1->getType()) {
V1 = IRB.CreateIntCast(V1, S1->getType(), false);
V2 = IRB.CreateIntCast(V2, S2->getType(), false);
}
Value *S1S2 = IRB.CreateAnd(S1, S2);
Value *V1S2 = IRB.CreateAnd(V1, S2);
Value *S1V2 = IRB.CreateAnd(S1, V2);
setShadow(&I, IRB.CreateOr({S1S2, V1S2, S1V2}));
setOriginForNaryOp(I);
}
/// Default propagation of shadow and/or origin.
///
/// This class implements the general case of shadow propagation, used in all
/// cases where we don't know and/or don't care about what the operation
/// actually does. It converts all input shadow values to a common type
/// (extending or truncating as necessary), and bitwise OR's them.
///
/// This is much cheaper than inserting checks (i.e. requiring inputs to be
/// fully initialized), and less prone to false positives.
///
/// This class also implements the general case of origin propagation. For a
/// Nary operation, result origin is set to the origin of an argument that is
/// not entirely initialized. If there is more than one such arguments, the
/// rightmost of them is picked. It does not matter which one is picked if all
/// arguments are initialized.
template <bool CombineShadow>
class Combiner {
Value *Shadow = nullptr;
Value *Origin = nullptr;
IRBuilder<> &IRB;
MemorySanitizerVisitor *MSV;
public:
Combiner(MemorySanitizerVisitor *MSV, IRBuilder<> &IRB)
: IRB(IRB), MSV(MSV) {}
/// Add a pair of shadow and origin values to the mix.
Combiner &Add(Value *OpShadow, Value *OpOrigin) {
if (CombineShadow) {
assert(OpShadow);
if (!Shadow)
Shadow = OpShadow;
else {
OpShadow = MSV->CreateShadowCast(IRB, OpShadow, Shadow->getType());
Shadow = IRB.CreateOr(Shadow, OpShadow, "_msprop");
}
}
if (MSV->MS.TrackOrigins) {
assert(OpOrigin);
if (!Origin) {
Origin = OpOrigin;
} else {
Constant *ConstOrigin = dyn_cast<Constant>(OpOrigin);
// No point in adding something that might result in 0 origin value.
if (!ConstOrigin || !ConstOrigin->isNullValue()) {
Value *FlatShadow = MSV->convertShadowToScalar(OpShadow, IRB);
Value *Cond =
IRB.CreateICmpNE(FlatShadow, MSV->getCleanShadow(FlatShadow));
Origin = IRB.CreateSelect(Cond, OpOrigin, Origin);
}
}
}
return *this;
}
/// Add an application value to the mix.
Combiner &Add(Value *V) {
Value *OpShadow = MSV->getShadow(V);
Value *OpOrigin = MSV->MS.TrackOrigins ? MSV->getOrigin(V) : nullptr;
return Add(OpShadow, OpOrigin);
}
/// Set the current combined values as the given instruction's shadow
/// and origin.
void Done(Instruction *I) {
if (CombineShadow) {
assert(Shadow);
Shadow = MSV->CreateShadowCast(IRB, Shadow, MSV->getShadowTy(I));
MSV->setShadow(I, Shadow);
}
if (MSV->MS.TrackOrigins) {
assert(Origin);
MSV->setOrigin(I, Origin);
}
}
};
using ShadowAndOriginCombiner = Combiner<true>;
using OriginCombiner = Combiner<false>;
/// Propagate origin for arbitrary operation.
void setOriginForNaryOp(Instruction &I) {
if (!MS.TrackOrigins) return;
IRBuilder<> IRB(&I);
OriginCombiner OC(this, IRB);
for (Use &Op : I.operands())
OC.Add(Op.get());
OC.Done(&I);
}
size_t VectorOrPrimitiveTypeSizeInBits(Type *Ty) {
assert(!(Ty->isVectorTy() && Ty->getScalarType()->isPointerTy()) &&
"Vector of pointers is not a valid shadow type");
return Ty->isVectorTy() ? cast<FixedVectorType>(Ty)->getNumElements() *
Ty->getScalarSizeInBits()
: Ty->getPrimitiveSizeInBits();
}
/// Cast between two shadow types, extending or truncating as
/// necessary.
Value *CreateShadowCast(IRBuilder<> &IRB, Value *V, Type *dstTy,
bool Signed = false) {
Type *srcTy = V->getType();
size_t srcSizeInBits = VectorOrPrimitiveTypeSizeInBits(srcTy);
size_t dstSizeInBits = VectorOrPrimitiveTypeSizeInBits(dstTy);
if (srcSizeInBits > 1 && dstSizeInBits == 1)
return IRB.CreateICmpNE(V, getCleanShadow(V));
if (dstTy->isIntegerTy() && srcTy->isIntegerTy())
return IRB.CreateIntCast(V, dstTy, Signed);
if (dstTy->isVectorTy() && srcTy->isVectorTy() &&
cast<FixedVectorType>(dstTy)->getNumElements() ==
cast<FixedVectorType>(srcTy)->getNumElements())
return IRB.CreateIntCast(V, dstTy, Signed);
Value *V1 = IRB.CreateBitCast(V, Type::getIntNTy(*MS.C, srcSizeInBits));
Value *V2 =
IRB.CreateIntCast(V1, Type::getIntNTy(*MS.C, dstSizeInBits), Signed);
return IRB.CreateBitCast(V2, dstTy);
// TODO: handle struct types.
}
/// Cast an application value to the type of its own shadow.
Value *CreateAppToShadowCast(IRBuilder<> &IRB, Value *V) {
Type *ShadowTy = getShadowTy(V);
if (V->getType() == ShadowTy)
return V;
if (V->getType()->isPtrOrPtrVectorTy())
return IRB.CreatePtrToInt(V, ShadowTy);
else
return IRB.CreateBitCast(V, ShadowTy);
}
/// Propagate shadow for arbitrary operation.
void handleShadowOr(Instruction &I) {
IRBuilder<> IRB(&I);
ShadowAndOriginCombiner SC(this, IRB);
for (Use &Op : I.operands())
SC.Add(Op.get());
SC.Done(&I);
}
void visitFNeg(UnaryOperator &I) { handleShadowOr(I); }
// Handle multiplication by constant.
//
// Handle a special case of multiplication by constant that may have one or
// more zeros in the lower bits. This makes corresponding number of lower bits
// of the result zero as well. We model it by shifting the other operand
// shadow left by the required number of bits. Effectively, we transform
// (X * (A * 2**B)) to ((X << B) * A) and instrument (X << B) as (Sx << B).
// We use multiplication by 2**N instead of shift to cover the case of
// multiplication by 0, which may occur in some elements of a vector operand.
void handleMulByConstant(BinaryOperator &I, Constant *ConstArg,
Value *OtherArg) {
Constant *ShadowMul;
Type *Ty = ConstArg->getType();
if (auto *VTy = dyn_cast<VectorType>(Ty)) {
unsigned NumElements = cast<FixedVectorType>(VTy)->getNumElements();
Type *EltTy = VTy->getElementType();
SmallVector<Constant *, 16> Elements;
for (unsigned Idx = 0; Idx < NumElements; ++Idx) {
if (ConstantInt *Elt =
dyn_cast<ConstantInt>(ConstArg->getAggregateElement(Idx))) {
const APInt &V = Elt->getValue();
APInt V2 = APInt(V.getBitWidth(), 1) << V.countTrailingZeros();
Elements.push_back(ConstantInt::get(EltTy, V2));
} else {
Elements.push_back(ConstantInt::get(EltTy, 1));
}
}
ShadowMul = ConstantVector::get(Elements);
} else {
if (ConstantInt *Elt = dyn_cast<ConstantInt>(ConstArg)) {
const APInt &V = Elt->getValue();
APInt V2 = APInt(V.getBitWidth(), 1) << V.countTrailingZeros();
ShadowMul = ConstantInt::get(Ty, V2);
} else {
ShadowMul = ConstantInt::get(Ty, 1);
}
}
IRBuilder<> IRB(&I);
setShadow(&I,
IRB.CreateMul(getShadow(OtherArg), ShadowMul, "msprop_mul_cst"));
setOrigin(&I, getOrigin(OtherArg));
}
void visitMul(BinaryOperator &I) {
Constant *constOp0 = dyn_cast<Constant>(I.getOperand(0));
Constant *constOp1 = dyn_cast<Constant>(I.getOperand(1));
if (constOp0 && !constOp1)
handleMulByConstant(I, constOp0, I.getOperand(1));
else if (constOp1 && !constOp0)
handleMulByConstant(I, constOp1, I.getOperand(0));
else
handleShadowOr(I);
}
void visitFAdd(BinaryOperator &I) { handleShadowOr(I); }
void visitFSub(BinaryOperator &I) { handleShadowOr(I); }
void visitFMul(BinaryOperator &I) { handleShadowOr(I); }
void visitAdd(BinaryOperator &I) { handleShadowOr(I); }
void visitSub(BinaryOperator &I) { handleShadowOr(I); }
void visitXor(BinaryOperator &I) { handleShadowOr(I); }
void handleIntegerDiv(Instruction &I) {
IRBuilder<> IRB(&I);
// Strict on the second argument.
insertShadowCheck(I.getOperand(1), &I);
setShadow(&I, getShadow(&I, 0));
setOrigin(&I, getOrigin(&I, 0));
}
void visitUDiv(BinaryOperator &I) { handleIntegerDiv(I); }
void visitSDiv(BinaryOperator &I) { handleIntegerDiv(I); }
void visitURem(BinaryOperator &I) { handleIntegerDiv(I); }
void visitSRem(BinaryOperator &I) { handleIntegerDiv(I); }
// Floating point division is side-effect free. We can not require that the
// divisor is fully initialized and must propagate shadow. See PR37523.
void visitFDiv(BinaryOperator &I) { handleShadowOr(I); }
void visitFRem(BinaryOperator &I) { handleShadowOr(I); }
/// Instrument == and != comparisons.
///
/// Sometimes the comparison result is known even if some of the bits of the
/// arguments are not.
void handleEqualityComparison(ICmpInst &I) {
IRBuilder<> IRB(&I);
Value *A = I.getOperand(0);
Value *B = I.getOperand(1);
Value *Sa = getShadow(A);
Value *Sb = getShadow(B);
// Get rid of pointers and vectors of pointers.
// For ints (and vectors of ints), types of A and Sa match,
// and this is a no-op.
A = IRB.CreatePointerCast(A, Sa->getType());
B = IRB.CreatePointerCast(B, Sb->getType());
// A == B <==> (C = A^B) == 0
// A != B <==> (C = A^B) != 0
// Sc = Sa | Sb
Value *C = IRB.CreateXor(A, B);
Value *Sc = IRB.CreateOr(Sa, Sb);
// Now dealing with i = (C == 0) comparison (or C != 0, does not matter now)
// Result is defined if one of the following is true
// * there is a defined 1 bit in C
// * C is fully defined
// Si = !(C & ~Sc) && Sc
Value *Zero = Constant::getNullValue(Sc->getType());
Value *MinusOne = Constant::getAllOnesValue(Sc->getType());
Value *Si =
IRB.CreateAnd(IRB.CreateICmpNE(Sc, Zero),
IRB.CreateICmpEQ(
IRB.CreateAnd(IRB.CreateXor(Sc, MinusOne), C), Zero));
Si->setName("_msprop_icmp");
setShadow(&I, Si);
setOriginForNaryOp(I);
}
/// Build the lowest possible value of V, taking into account V's
/// uninitialized bits.
Value *getLowestPossibleValue(IRBuilder<> &IRB, Value *A, Value *Sa,
bool isSigned) {
if (isSigned) {
// Split shadow into sign bit and other bits.
Value *SaOtherBits = IRB.CreateLShr(IRB.CreateShl(Sa, 1), 1);
Value *SaSignBit = IRB.CreateXor(Sa, SaOtherBits);
// Maximise the undefined shadow bit, minimize other undefined bits.
return
IRB.CreateOr(IRB.CreateAnd(A, IRB.CreateNot(SaOtherBits)), SaSignBit);
} else {
// Minimize undefined bits.
return IRB.CreateAnd(A, IRB.CreateNot(Sa));
}
}
/// Build the highest possible value of V, taking into account V's
/// uninitialized bits.
Value *getHighestPossibleValue(IRBuilder<> &IRB, Value *A, Value *Sa,
bool isSigned) {
if (isSigned) {
// Split shadow into sign bit and other bits.
Value *SaOtherBits = IRB.CreateLShr(IRB.CreateShl(Sa, 1), 1);
Value *SaSignBit = IRB.CreateXor(Sa, SaOtherBits);
// Minimise the undefined shadow bit, maximise other undefined bits.
return
IRB.CreateOr(IRB.CreateAnd(A, IRB.CreateNot(SaSignBit)), SaOtherBits);
} else {
// Maximize undefined bits.
return IRB.CreateOr(A, Sa);
}
}
/// Instrument relational comparisons.
///
/// This function does exact shadow propagation for all relational
/// comparisons of integers, pointers and vectors of those.
/// FIXME: output seems suboptimal when one of the operands is a constant
void handleRelationalComparisonExact(ICmpInst &I) {
IRBuilder<> IRB(&I);
Value *A = I.getOperand(0);
Value *B = I.getOperand(1);
Value *Sa = getShadow(A);
Value *Sb = getShadow(B);
// Get rid of pointers and vectors of pointers.
// For ints (and vectors of ints), types of A and Sa match,
// and this is a no-op.
A = IRB.CreatePointerCast(A, Sa->getType());
B = IRB.CreatePointerCast(B, Sb->getType());
// Let [a0, a1] be the interval of possible values of A, taking into account
// its undefined bits. Let [b0, b1] be the interval of possible values of B.
// Then (A cmp B) is defined iff (a0 cmp b1) == (a1 cmp b0).
bool IsSigned = I.isSigned();
Value *S1 = IRB.CreateICmp(I.getPredicate(),
getLowestPossibleValue(IRB, A, Sa, IsSigned),
getHighestPossibleValue(IRB, B, Sb, IsSigned));
Value *S2 = IRB.CreateICmp(I.getPredicate(),
getHighestPossibleValue(IRB, A, Sa, IsSigned),
getLowestPossibleValue(IRB, B, Sb, IsSigned));
Value *Si = IRB.CreateXor(S1, S2);
setShadow(&I, Si);
setOriginForNaryOp(I);
}
/// Instrument signed relational comparisons.
///
/// Handle sign bit tests: x<0, x>=0, x<=-1, x>-1 by propagating the highest
/// bit of the shadow. Everything else is delegated to handleShadowOr().
void handleSignedRelationalComparison(ICmpInst &I) {
Constant *constOp;
Value *op = nullptr;
CmpInst::Predicate pre;
if ((constOp = dyn_cast<Constant>(I.getOperand(1)))) {
op = I.getOperand(0);
pre = I.getPredicate();
} else if ((constOp = dyn_cast<Constant>(I.getOperand(0)))) {
op = I.getOperand(1);
pre = I.getSwappedPredicate();
} else {
handleShadowOr(I);
return;
}
if ((constOp->isNullValue() &&
(pre == CmpInst::ICMP_SLT || pre == CmpInst::ICMP_SGE)) ||
(constOp->isAllOnesValue() &&
(pre == CmpInst::ICMP_SGT || pre == CmpInst::ICMP_SLE))) {
IRBuilder<> IRB(&I);
Value *Shadow = IRB.CreateICmpSLT(getShadow(op), getCleanShadow(op),
"_msprop_icmp_s");
setShadow(&I, Shadow);
setOrigin(&I, getOrigin(op));
} else {
handleShadowOr(I);
}
}
void visitICmpInst(ICmpInst &I) {
if (!ClHandleICmp) {
handleShadowOr(I);
return;
}
if (I.isEquality()) {
handleEqualityComparison(I);
return;
}
assert(I.isRelational());
if (ClHandleICmpExact) {
handleRelationalComparisonExact(I);
return;
}
if (I.isSigned()) {
handleSignedRelationalComparison(I);
return;
}
assert(I.isUnsigned());
if ((isa<Constant>(I.getOperand(0)) || isa<Constant>(I.getOperand(1)))) {
handleRelationalComparisonExact(I);
return;
}
handleShadowOr(I);
}
void visitFCmpInst(FCmpInst &I) {
handleShadowOr(I);
}
void handleShift(BinaryOperator &I) {
IRBuilder<> IRB(&I);
// If any of the S2 bits are poisoned, the whole thing is poisoned.
// Otherwise perform the same shift on S1.
Value *S1 = getShadow(&I, 0);
Value *S2 = getShadow(&I, 1);
Value *S2Conv = IRB.CreateSExt(IRB.CreateICmpNE(S2, getCleanShadow(S2)),
S2->getType());
Value *V2 = I.getOperand(1);
Value *Shift = IRB.CreateBinOp(I.getOpcode(), S1, V2);
setShadow(&I, IRB.CreateOr(Shift, S2Conv));
setOriginForNaryOp(I);
}
void visitShl(BinaryOperator &I) { handleShift(I); }
void visitAShr(BinaryOperator &I) { handleShift(I); }
void visitLShr(BinaryOperator &I) { handleShift(I); }
/// Instrument llvm.memmove
///
/// At this point we don't know if llvm.memmove will be inlined or not.
/// If we don't instrument it and it gets inlined,
/// our interceptor will not kick in and we will lose the memmove.
/// If we instrument the call here, but it does not get inlined,
/// we will memove the shadow twice: which is bad in case
/// of overlapping regions. So, we simply lower the intrinsic to a call.
///
/// Similar situation exists for memcpy and memset.
void visitMemMoveInst(MemMoveInst &I) {
IRBuilder<> IRB(&I);
IRB.CreateCall(
MS.MemmoveFn,
{IRB.CreatePointerCast(I.getArgOperand(0), IRB.getInt8PtrTy()),
IRB.CreatePointerCast(I.getArgOperand(1), IRB.getInt8PtrTy()),
IRB.CreateIntCast(I.getArgOperand(2), MS.IntptrTy, false)});
I.eraseFromParent();
}
// Similar to memmove: avoid copying shadow twice.
// This is somewhat unfortunate as it may slowdown small constant memcpys.
// FIXME: consider doing manual inline for small constant sizes and proper
// alignment.
void visitMemCpyInst(MemCpyInst &I) {
IRBuilder<> IRB(&I);
IRB.CreateCall(
MS.MemcpyFn,
{IRB.CreatePointerCast(I.getArgOperand(0), IRB.getInt8PtrTy()),
IRB.CreatePointerCast(I.getArgOperand(1), IRB.getInt8PtrTy()),
IRB.CreateIntCast(I.getArgOperand(2), MS.IntptrTy, false)});
I.eraseFromParent();
}
// Same as memcpy.
void visitMemSetInst(MemSetInst &I) {
IRBuilder<> IRB(&I);
IRB.CreateCall(
MS.MemsetFn,
{IRB.CreatePointerCast(I.getArgOperand(0), IRB.getInt8PtrTy()),
IRB.CreateIntCast(I.getArgOperand(1), IRB.getInt32Ty(), false),
IRB.CreateIntCast(I.getArgOperand(2), MS.IntptrTy, false)});
I.eraseFromParent();
}
void visitVAStartInst(VAStartInst &I) {
VAHelper->visitVAStartInst(I);
}
void visitVACopyInst(VACopyInst &I) {
VAHelper->visitVACopyInst(I);
}
/// Handle vector store-like intrinsics.
///
/// Instrument intrinsics that look like a simple SIMD store: writes memory,
/// has 1 pointer argument and 1 vector argument, returns void.
bool handleVectorStoreIntrinsic(IntrinsicInst &I) {
IRBuilder<> IRB(&I);
Value* Addr = I.getArgOperand(0);
Value *Shadow = getShadow(&I, 1);
Value *ShadowPtr, *OriginPtr;
// We don't know the pointer alignment (could be unaligned SSE store!).
// Have to assume to worst case.
std::tie(ShadowPtr, OriginPtr) = getShadowOriginPtr(
Addr, IRB, Shadow->getType(), Align(1), /*isStore*/ true);
IRB.CreateAlignedStore(Shadow, ShadowPtr, Align(1));
if (ClCheckAccessAddress)
insertShadowCheck(Addr, &I);
// FIXME: factor out common code from materializeStores
if (MS.TrackOrigins) IRB.CreateStore(getOrigin(&I, 1), OriginPtr);
return true;
}
/// Handle vector load-like intrinsics.
///
/// Instrument intrinsics that look like a simple SIMD load: reads memory,
/// has 1 pointer argument, returns a vector.
bool handleVectorLoadIntrinsic(IntrinsicInst &I) {
IRBuilder<> IRB(&I);
Value *Addr = I.getArgOperand(0);
Type *ShadowTy = getShadowTy(&I);
Value *ShadowPtr = nullptr, *OriginPtr = nullptr;
if (PropagateShadow) {
// We don't know the pointer alignment (could be unaligned SSE load!).
// Have to assume to worst case.
const Align Alignment = Align(1);
std::tie(ShadowPtr, OriginPtr) =
getShadowOriginPtr(Addr, IRB, ShadowTy, Alignment, /*isStore*/ false);
setShadow(&I,
IRB.CreateAlignedLoad(ShadowTy, ShadowPtr, Alignment, "_msld"));
} else {
setShadow(&I, getCleanShadow(&I));
}
if (ClCheckAccessAddress)
insertShadowCheck(Addr, &I);
if (MS.TrackOrigins) {
if (PropagateShadow)
setOrigin(&I, IRB.CreateLoad(MS.OriginTy, OriginPtr));
else
setOrigin(&I, getCleanOrigin());
}
return true;
}
/// Handle (SIMD arithmetic)-like intrinsics.
///
/// Instrument intrinsics with any number of arguments of the same type,
/// equal to the return type. The type should be simple (no aggregates or
/// pointers; vectors are fine).
/// Caller guarantees that this intrinsic does not access memory.
bool maybeHandleSimpleNomemIntrinsic(IntrinsicInst &I) {
Type *RetTy = I.getType();
if (!(RetTy->isIntOrIntVectorTy() ||
RetTy->isFPOrFPVectorTy() ||
RetTy->isX86_MMXTy()))
return false;
unsigned NumArgOperands = I.getNumArgOperands();
for (unsigned i = 0; i < NumArgOperands; ++i) {
Type *Ty = I.getArgOperand(i)->getType();
if (Ty != RetTy)
return false;
}
IRBuilder<> IRB(&I);
ShadowAndOriginCombiner SC(this, IRB);
for (unsigned i = 0; i < NumArgOperands; ++i)
SC.Add(I.getArgOperand(i));
SC.Done(&I);
return true;
}
/// Heuristically instrument unknown intrinsics.
///
/// The main purpose of this code is to do something reasonable with all
/// random intrinsics we might encounter, most importantly - SIMD intrinsics.
/// We recognize several classes of intrinsics by their argument types and
/// ModRefBehaviour and apply special instrumentation when we are reasonably
/// sure that we know what the intrinsic does.
///
/// We special-case intrinsics where this approach fails. See llvm.bswap
/// handling as an example of that.
bool handleUnknownIntrinsic(IntrinsicInst &I) {
unsigned NumArgOperands = I.getNumArgOperands();
if (NumArgOperands == 0)
return false;
if (NumArgOperands == 2 &&
I.getArgOperand(0)->getType()->isPointerTy() &&
I.getArgOperand(1)->getType()->isVectorTy() &&
I.getType()->isVoidTy() &&
!I.onlyReadsMemory()) {