[asan] Fix dead stripping of globals on Linux.
Use a combination of !associated, comdat, @llvm.compiler.used and
custom sections to allow dead stripping of globals and their asan
metadata. Sometimes.
Currently this works on LLD, which supports SHF_LINK_ORDER with
sh_link pointing to the associated section.
This also works on BFD, which seems to treat comdats as
all-or-nothing with respect to linker GC. There is a weird quirk
where the "first" global in each link is never GC-ed because of the
section symbols.
At this moment it does not work on Gold (as in the globals are never
stripped).
Differential Revision: https://reviews.llvm.org/D30121
llvm-svn: 298158
diff --git a/llvm/lib/Transforms/IPO/ThinLTOBitcodeWriter.cpp b/llvm/lib/Transforms/IPO/ThinLTOBitcodeWriter.cpp
index ac62496..c7e91cd 100644
--- a/llvm/lib/Transforms/IPO/ThinLTOBitcodeWriter.cpp
+++ b/llvm/lib/Transforms/IPO/ThinLTOBitcodeWriter.cpp
@@ -28,51 +28,11 @@
#include "llvm/Support/ScopedPrinter.h"
#include "llvm/Transforms/IPO/FunctionAttrs.h"
#include "llvm/Transforms/Utils/Cloning.h"
+#include "llvm/Transforms/Utils/ModuleUtils.h"
using namespace llvm;
namespace {
-// Produce a unique identifier for this module by taking the MD5 sum of the
-// names of the module's strong external symbols. This identifier is
-// normally guaranteed to be unique, or the program would fail to link due to
-// multiply defined symbols.
-//
-// If the module has no strong external symbols (such a module may still have a
-// semantic effect if it performs global initialization), we cannot produce a
-// unique identifier for this module, so we return the empty string, which
-// causes the entire module to be written as a regular LTO module.
-std::string getModuleId(Module *M) {
- MD5 Md5;
- bool ExportsSymbols = false;
- auto AddGlobal = [&](GlobalValue &GV) {
- if (GV.isDeclaration() || GV.getName().startswith("llvm.") ||
- !GV.hasExternalLinkage())
- return;
- ExportsSymbols = true;
- Md5.update(GV.getName());
- Md5.update(ArrayRef<uint8_t>{0});
- };
-
- for (auto &F : *M)
- AddGlobal(F);
- for (auto &GV : M->globals())
- AddGlobal(GV);
- for (auto &GA : M->aliases())
- AddGlobal(GA);
- for (auto &IF : M->ifuncs())
- AddGlobal(IF);
-
- if (!ExportsSymbols)
- return "";
-
- MD5::MD5Result R;
- Md5.final(R);
-
- SmallString<32> Str;
- MD5::stringifyResult(R, Str);
- return ("$" + Str).str();
-}
-
// Promote each local-linkage entity defined by ExportM and used by ImportM by
// changing visibility and appending the given ModuleId.
void promoteInternals(Module &ExportM, Module &ImportM, StringRef ModuleId) {
@@ -253,7 +213,7 @@
void splitAndWriteThinLTOBitcode(
raw_ostream &OS, function_ref<AAResults &(Function &)> AARGetter,
Module &M) {
- std::string ModuleId = getModuleId(&M);
+ std::string ModuleId = getUniqueModuleId(&M);
if (ModuleId.empty()) {
// We couldn't generate a module ID for this module, just write it out as a
// regular LTO module.
diff --git a/llvm/lib/Transforms/Instrumentation/AddressSanitizer.cpp b/llvm/lib/Transforms/Instrumentation/AddressSanitizer.cpp
index afcf865..c20ad6d 100644
--- a/llvm/lib/Transforms/Instrumentation/AddressSanitizer.cpp
+++ b/llvm/lib/Transforms/Instrumentation/AddressSanitizer.cpp
@@ -101,6 +101,10 @@
"__asan_register_image_globals";
static const char *const kAsanUnregisterImageGlobalsName =
"__asan_unregister_image_globals";
+static const char *const kAsanRegisterElfGlobalsName =
+ "__asan_register_elf_globals";
+static const char *const kAsanUnregisterElfGlobalsName =
+ "__asan_unregister_elf_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";
@@ -120,8 +124,11 @@
"__asan_poison_stack_memory";
static const char *const kAsanUnpoisonStackMemoryName =
"__asan_unpoison_stack_memory";
+
+// ASan version script has __asan_* wildcard. Triple underscore prevents a
+// linker (gold) warning about attempting to export a local symbol.
static const char *const kAsanGlobalsRegisteredFlagName =
- "__asan_globals_registered";
+ "___asan_globals_registered";
static const char *const kAsanOptionDetectUseAfterReturn =
"__asan_option_detect_stack_use_after_return";
@@ -612,6 +619,10 @@
void InstrumentGlobalsCOFF(IRBuilder<> &IRB, Module &M,
ArrayRef<GlobalVariable *> ExtendedGlobals,
ArrayRef<Constant *> MetadataInitializers);
+ void InstrumentGlobalsELF(IRBuilder<> &IRB, Module &M,
+ ArrayRef<GlobalVariable *> ExtendedGlobals,
+ ArrayRef<Constant *> MetadataInitializers,
+ const std::string &UniqueModuleId);
void InstrumentGlobalsMachO(IRBuilder<> &IRB, Module &M,
ArrayRef<GlobalVariable *> ExtendedGlobals,
ArrayRef<Constant *> MetadataInitializers);
@@ -622,7 +633,8 @@
GlobalVariable *CreateMetadataGlobal(Module &M, Constant *Initializer,
StringRef OriginalName);
- void SetComdatForGlobalMetadata(GlobalVariable *G, GlobalVariable *Metadata);
+ void SetComdatForGlobalMetadata(GlobalVariable *G, GlobalVariable *Metadata,
+ StringRef InternalSuffix);
IRBuilder<> CreateAsanModuleDtor(Module &M);
bool ShouldInstrumentGlobal(GlobalVariable *G);
@@ -647,6 +659,8 @@
Function *AsanUnregisterGlobals;
Function *AsanRegisterImageGlobals;
Function *AsanUnregisterImageGlobals;
+ Function *AsanRegisterElfGlobals;
+ Function *AsanUnregisterElfGlobals;
};
// Stack poisoning does not play well with exception handling.
@@ -1596,12 +1610,22 @@
checkSanitizerInterfaceFunction(M.getOrInsertFunction(
kAsanUnregisterImageGlobalsName, IRB.getVoidTy(), IntptrTy, nullptr));
AsanUnregisterImageGlobals->setLinkage(Function::ExternalLinkage);
+
+ AsanRegisterElfGlobals = checkSanitizerInterfaceFunction(
+ M.getOrInsertFunction(kAsanRegisterElfGlobalsName, IRB.getVoidTy(),
+ IntptrTy, IntptrTy, IntptrTy, nullptr));
+ AsanRegisterElfGlobals->setLinkage(Function::ExternalLinkage);
+
+ AsanUnregisterElfGlobals = checkSanitizerInterfaceFunction(
+ M.getOrInsertFunction(kAsanUnregisterElfGlobalsName, IRB.getVoidTy(),
+ IntptrTy, IntptrTy, IntptrTy, nullptr));
+ AsanUnregisterElfGlobals->setLinkage(Function::ExternalLinkage);
}
// Put the metadata and the instrumented global in the same group. This ensures
// that the metadata is discarded if the instrumented global is discarded.
void AddressSanitizerModule::SetComdatForGlobalMetadata(
- GlobalVariable *G, GlobalVariable *Metadata) {
+ GlobalVariable *G, GlobalVariable *Metadata, StringRef InternalSuffix) {
Module &M = *G->getParent();
Comdat *C = G->getComdat();
if (!C) {
@@ -1611,7 +1635,15 @@
assert(G->hasLocalLinkage());
G->setName(Twine(kAsanGenPrefix) + "_anon_global");
}
- C = M.getOrInsertComdat(G->getName());
+
+ if (!InternalSuffix.empty() && G->hasLocalLinkage()) {
+ std::string Name = G->getName();
+ Name += InternalSuffix;
+ C = M.getOrInsertComdat(Name);
+ } else {
+ C = M.getOrInsertComdat(G->getName());
+ }
+
// Make this IMAGE_COMDAT_SELECT_NODUPLICATES on COFF.
if (TargetTriple.isOSBinFormatCOFF())
C->setSelectionKind(Comdat::NoDuplicates);
@@ -1666,10 +1698,69 @@
"global metadata will not be padded appropriately");
Metadata->setAlignment(SizeOfGlobalStruct);
- SetComdatForGlobalMetadata(G, Metadata);
+ SetComdatForGlobalMetadata(G, Metadata, "");
}
}
+void AddressSanitizerModule::InstrumentGlobalsELF(
+ IRBuilder<> &IRB, Module &M, ArrayRef<GlobalVariable *> ExtendedGlobals,
+ ArrayRef<Constant *> MetadataInitializers,
+ const std::string &UniqueModuleId) {
+ assert(ExtendedGlobals.size() == MetadataInitializers.size());
+
+ SmallVector<GlobalValue *, 16> MetadataGlobals(ExtendedGlobals.size());
+ for (size_t i = 0; i < ExtendedGlobals.size(); i++) {
+ GlobalVariable *G = ExtendedGlobals[i];
+ GlobalVariable *Metadata =
+ CreateMetadataGlobal(M, MetadataInitializers[i], G->getName());
+ MDNode *MD = MDNode::get(M.getContext(), ValueAsMetadata::get(G));
+ Metadata->setMetadata(LLVMContext::MD_associated, MD);
+ MetadataGlobals[i] = Metadata;
+
+ SetComdatForGlobalMetadata(G, Metadata, UniqueModuleId);
+ }
+
+ // Update llvm.compiler.used, adding the new metadata globals. This is
+ // needed so that during LTO these variables stay alive.
+ if (!MetadataGlobals.empty())
+ appendToCompilerUsed(M, MetadataGlobals);
+
+ // RegisteredFlag serves two purposes. First, we can pass it to dladdr()
+ // to look up the loaded image that contains it. Second, we can store in it
+ // whether registration has already occurred, to prevent duplicate
+ // registration.
+ //
+ // Common linkage ensures that there is only one global per shared library.
+ GlobalVariable *RegisteredFlag = new GlobalVariable(
+ M, IntptrTy, false, GlobalVariable::CommonLinkage,
+ ConstantInt::get(IntptrTy, 0), kAsanGlobalsRegisteredFlagName);
+ RegisteredFlag->setVisibility(GlobalVariable::HiddenVisibility);
+
+ // Create start and stop symbols.
+ GlobalVariable *StartELFMetadata = new GlobalVariable(
+ M, IntptrTy, false, GlobalVariable::ExternalWeakLinkage, nullptr,
+ "__start_" + getGlobalMetadataSection());
+ StartELFMetadata->setVisibility(GlobalVariable::HiddenVisibility);
+ GlobalVariable *StopELFMetadata = new GlobalVariable(
+ M, IntptrTy, false, GlobalVariable::ExternalWeakLinkage, nullptr,
+ "__stop_" + getGlobalMetadataSection());
+ StopELFMetadata->setVisibility(GlobalVariable::HiddenVisibility);
+
+ // Create a call to register the globals with the runtime.
+ IRB.CreateCall(AsanRegisterElfGlobals,
+ {IRB.CreatePointerCast(RegisteredFlag, IntptrTy),
+ IRB.CreatePointerCast(StartELFMetadata, IntptrTy),
+ IRB.CreatePointerCast(StopELFMetadata, IntptrTy)});
+
+ // We also need to unregister globals at the end, e.g., when a shared library
+ // gets closed.
+ IRBuilder<> IRB_Dtor = CreateAsanModuleDtor(M);
+ IRB_Dtor.CreateCall(AsanUnregisterElfGlobals,
+ {IRB.CreatePointerCast(RegisteredFlag, IntptrTy),
+ IRB.CreatePointerCast(StartELFMetadata, IntptrTy),
+ IRB.CreatePointerCast(StopELFMetadata, IntptrTy)});
+}
+
void AddressSanitizerModule::InstrumentGlobalsMachO(
IRBuilder<> &IRB, Module &M, ArrayRef<GlobalVariable *> ExtendedGlobals,
ArrayRef<Constant *> MetadataInitializers) {
@@ -1912,7 +2003,12 @@
Initializers[i] = Initializer;
}
- if (TargetTriple.isOSBinFormatCOFF()) {
+ std::string ELFUniqueModuleId =
+ TargetTriple.isOSBinFormatELF() ? getUniqueModuleId(&M) : "";
+
+ if (!ELFUniqueModuleId.empty()) {
+ InstrumentGlobalsELF(IRB, M, NewGlobals, Initializers, ELFUniqueModuleId);
+ } else if (TargetTriple.isOSBinFormatCOFF()) {
InstrumentGlobalsCOFF(IRB, M, NewGlobals, Initializers);
} else if (ShouldUseMachOGlobalsSection()) {
InstrumentGlobalsMachO(IRB, M, NewGlobals, Initializers);
diff --git a/llvm/lib/Transforms/Utils/ModuleUtils.cpp b/llvm/lib/Transforms/Utils/ModuleUtils.cpp
index a53da85..27b4753 100644
--- a/llvm/lib/Transforms/Utils/ModuleUtils.cpp
+++ b/llvm/lib/Transforms/Utils/ModuleUtils.cpp
@@ -229,3 +229,35 @@
ComdatEntriesCovered.end();
});
}
+
+std::string llvm::getUniqueModuleId(Module *M) {
+ MD5 Md5;
+ bool ExportsSymbols = false;
+ auto AddGlobal = [&](GlobalValue &GV) {
+ if (GV.isDeclaration() || GV.getName().startswith("llvm.") ||
+ !GV.hasExternalLinkage())
+ return;
+ ExportsSymbols = true;
+ Md5.update(GV.getName());
+ Md5.update(ArrayRef<uint8_t>{0});
+ };
+
+ for (auto &F : *M)
+ AddGlobal(F);
+ for (auto &GV : M->globals())
+ AddGlobal(GV);
+ for (auto &GA : M->aliases())
+ AddGlobal(GA);
+ for (auto &IF : M->ifuncs())
+ AddGlobal(IF);
+
+ if (!ExportsSymbols)
+ return "";
+
+ MD5::MD5Result R;
+ Md5.final(R);
+
+ SmallString<32> Str;
+ MD5::stringifyResult(R, Str);
+ return ("$" + Str).str();
+}