llvm-project/llvm/lib/LTO/LTOModule.cpp

661 lines
22 KiB
C++
Raw Normal View History

//===-- LTOModule.cpp - LLVM Link Time Optimizer --------------------------===//
//
// The LLVM Compiler Infrastructure
//
// This file is distributed under the University of Illinois Open Source
// License. See LICENSE.TXT for details.
//
//===----------------------------------------------------------------------===//
//
// This file implements the Link Time Optimization library. This library is
// intended to be used by linker to optimize code at link time.
//
//===----------------------------------------------------------------------===//
#include "llvm/LTO/LTOModule.h"
#include "llvm/ADT/Triple.h"
#include "llvm/Bitcode/ReaderWriter.h"
#include "llvm/CodeGen/Analysis.h"
#include "llvm/IR/Constants.h"
Use the DiagnosticHandler to print diagnostics when reading bitcode. The bitcode reading interface used std::error_code to report an error to the callers and it is the callers job to print diagnostics. This is not ideal for error handling or diagnostic reporting: * For error handling, all that the callers care about is 3 possibilities: * It worked * The bitcode file is corrupted/invalid. * The file is not bitcode at all. * For diagnostic, it is user friendly to include far more information about the invalid case so the user can find out what is wrong with the bitcode file. This comes up, for example, when a developer introduces a bug while extending the format. The compromise we had was to have a lot of error codes. With this patch we use the DiagnosticHandler to communicate with the human and std::error_code to communicate with the caller. This allows us to have far fewer error codes and adds the infrastructure to print better diagnostics. This is so because the diagnostics are printed when he issue is found. The code that detected the problem in alive in the stack and can pass down as much context as needed. As an example the patch updates test/Bitcode/invalid.ll. Using a DiagnosticHandler also moves the fatal/non-fatal error decision to the caller. A simple one like llvm-dis can just use fatal errors. The gold plugin needs a bit more complex treatment because of being passed non-bitcode files. An hypothetical interactive tool would make all bitcode errors non-fatal. llvm-svn: 225562
2015-01-10 08:07:30 +08:00
#include "llvm/IR/DiagnosticPrinter.h"
#include "llvm/IR/LLVMContext.h"
#include "llvm/IR/Mangler.h"
#include "llvm/IR/Metadata.h"
#include "llvm/IR/Module.h"
#include "llvm/MC/MCExpr.h"
#include "llvm/MC/MCInst.h"
#include "llvm/MC/MCInstrInfo.h"
#include "llvm/MC/MCParser/MCAsmParser.h"
#include "llvm/MC/MCParser/MCTargetAsmParser.h"
#include "llvm/MC/MCSection.h"
#include "llvm/MC/MCSubtargetInfo.h"
#include "llvm/MC/MCSymbol.h"
#include "llvm/MC/SubtargetFeature.h"
#include "llvm/Object/IRObjectFile.h"
#include "llvm/Object/ObjectFile.h"
#include "llvm/Support/FileSystem.h"
#include "llvm/Support/Host.h"
#include "llvm/Support/MemoryBuffer.h"
#include "llvm/Support/Path.h"
#include "llvm/Support/SourceMgr.h"
#include "llvm/Support/TargetRegistry.h"
#include "llvm/Support/TargetSelect.h"
#include "llvm/Target/TargetLowering.h"
#include "llvm/Target/TargetLoweringObjectFile.h"
#include "llvm/Target/TargetRegisterInfo.h"
#include "llvm/Target/TargetSubtargetInfo.h"
#include "llvm/Transforms/Utils/GlobalStatus.h"
#include <system_error>
using namespace llvm;
using namespace llvm::object;
LTOModule::LTOModule(std::unique_ptr<object::IRObjectFile> Obj,
llvm::TargetMachine *TM)
: IRFile(std::move(Obj)), _target(TM) {}
LTOModule::~LTOModule() {}
/// isBitcodeFile - Returns 'true' if the file (or memory contents) is LLVM
/// bitcode.
bool LTOModule::isBitcodeFile(const void *Mem, size_t Length) {
ErrorOr<MemoryBufferRef> BCData = IRObjectFile::findBitcodeInMemBuffer(
MemoryBufferRef(StringRef((const char *)Mem, Length), "<mem>"));
return bool(BCData);
}
bool LTOModule::isBitcodeFile(const char *Path) {
ErrorOr<std::unique_ptr<MemoryBuffer>> BufferOrErr =
MemoryBuffer::getFile(Path);
if (!BufferOrErr)
return false;
ErrorOr<MemoryBufferRef> BCData = IRObjectFile::findBitcodeInMemBuffer(
BufferOrErr.get()->getMemBufferRef());
return bool(BCData);
}
bool LTOModule::isThinLTO() {
// Right now the detection is only based on the summary presence. We may want
// to add a dedicated flag at some point.
return hasGlobalValueSummary(IRFile->getMemoryBufferRef(),
[](const DiagnosticInfo &DI) {
DiagnosticPrinterRawOStream DP(errs());
DI.print(DP);
errs() << '\n';
return;
});
}
bool LTOModule::isBitcodeForTarget(MemoryBuffer *Buffer,
StringRef TriplePrefix) {
ErrorOr<MemoryBufferRef> BCOrErr =
IRObjectFile::findBitcodeInMemBuffer(Buffer->getMemBufferRef());
if (!BCOrErr)
return false;
LLVMContext Context;
std::string Triple = getBitcodeTargetTriple(*BCOrErr, Context);
return StringRef(Triple).startswith(TriplePrefix);
}
std::string LTOModule::getProducerString(MemoryBuffer *Buffer) {
ErrorOr<MemoryBufferRef> BCOrErr =
IRObjectFile::findBitcodeInMemBuffer(Buffer->getMemBufferRef());
if (!BCOrErr)
return "";
LLVMContext Context;
return getBitcodeProducerString(*BCOrErr, Context);
}
ErrorOr<std::unique_ptr<LTOModule>>
LTOModule::createFromFile(LLVMContext &Context, const char *path,
TargetOptions options) {
ErrorOr<std::unique_ptr<MemoryBuffer>> BufferOrErr =
MemoryBuffer::getFile(path);
if (std::error_code EC = BufferOrErr.getError()) {
Context.emitError(EC.message());
return EC;
}
std::unique_ptr<MemoryBuffer> Buffer = std::move(BufferOrErr.get());
return makeLTOModule(Buffer->getMemBufferRef(), options, Context,
/* ShouldBeLazy*/ false);
}
ErrorOr<std::unique_ptr<LTOModule>>
LTOModule::createFromOpenFile(LLVMContext &Context, int fd, const char *path,
size_t size, TargetOptions options) {
return createFromOpenFileSlice(Context, fd, path, size, 0, options);
}
ErrorOr<std::unique_ptr<LTOModule>>
LTOModule::createFromOpenFileSlice(LLVMContext &Context, int fd,
const char *path, size_t map_size,
off_t offset, TargetOptions options) {
ErrorOr<std::unique_ptr<MemoryBuffer>> BufferOrErr =
MemoryBuffer::getOpenFileSlice(fd, path, map_size, offset);
if (std::error_code EC = BufferOrErr.getError()) {
Context.emitError(EC.message());
return EC;
}
std::unique_ptr<MemoryBuffer> Buffer = std::move(BufferOrErr.get());
return makeLTOModule(Buffer->getMemBufferRef(), options, Context,
/* ShouldBeLazy */ false);
}
ErrorOr<std::unique_ptr<LTOModule>>
LTOModule::createFromBuffer(LLVMContext &Context, const void *mem,
size_t length, TargetOptions options,
StringRef path) {
StringRef Data((const char *)mem, length);
MemoryBufferRef Buffer(Data, path);
return makeLTOModule(Buffer, options, Context, /* ShouldBeLazy */ false);
}
ErrorOr<std::unique_ptr<LTOModule>>
LTOModule::createInLocalContext(std::unique_ptr<LLVMContext> Context,
const void *mem, size_t length,
TargetOptions options, StringRef path) {
StringRef Data((const char *)mem, length);
MemoryBufferRef Buffer(Data, path);
// If we own a context, we know this is being used only for symbol extraction,
// not linking. Be lazy in that case.
ErrorOr<std::unique_ptr<LTOModule>> Ret =
makeLTOModule(Buffer, options, *Context, /* ShouldBeLazy */ true);
if (Ret)
(*Ret)->OwnedContext = std::move(Context);
return Ret;
}
static ErrorOr<std::unique_ptr<Module>>
parseBitcodeFileImpl(MemoryBufferRef Buffer, LLVMContext &Context,
bool ShouldBeLazy) {
Use the DiagnosticHandler to print diagnostics when reading bitcode. The bitcode reading interface used std::error_code to report an error to the callers and it is the callers job to print diagnostics. This is not ideal for error handling or diagnostic reporting: * For error handling, all that the callers care about is 3 possibilities: * It worked * The bitcode file is corrupted/invalid. * The file is not bitcode at all. * For diagnostic, it is user friendly to include far more information about the invalid case so the user can find out what is wrong with the bitcode file. This comes up, for example, when a developer introduces a bug while extending the format. The compromise we had was to have a lot of error codes. With this patch we use the DiagnosticHandler to communicate with the human and std::error_code to communicate with the caller. This allows us to have far fewer error codes and adds the infrastructure to print better diagnostics. This is so because the diagnostics are printed when he issue is found. The code that detected the problem in alive in the stack and can pass down as much context as needed. As an example the patch updates test/Bitcode/invalid.ll. Using a DiagnosticHandler also moves the fatal/non-fatal error decision to the caller. A simple one like llvm-dis can just use fatal errors. The gold plugin needs a bit more complex treatment because of being passed non-bitcode files. An hypothetical interactive tool would make all bitcode errors non-fatal. llvm-svn: 225562
2015-01-10 08:07:30 +08:00
// Find the buffer.
ErrorOr<MemoryBufferRef> MBOrErr =
IRObjectFile::findBitcodeInMemBuffer(Buffer);
if (std::error_code EC = MBOrErr.getError()) {
Context.emitError(EC.message());
return EC;
}
Use the DiagnosticHandler to print diagnostics when reading bitcode. The bitcode reading interface used std::error_code to report an error to the callers and it is the callers job to print diagnostics. This is not ideal for error handling or diagnostic reporting: * For error handling, all that the callers care about is 3 possibilities: * It worked * The bitcode file is corrupted/invalid. * The file is not bitcode at all. * For diagnostic, it is user friendly to include far more information about the invalid case so the user can find out what is wrong with the bitcode file. This comes up, for example, when a developer introduces a bug while extending the format. The compromise we had was to have a lot of error codes. With this patch we use the DiagnosticHandler to communicate with the human and std::error_code to communicate with the caller. This allows us to have far fewer error codes and adds the infrastructure to print better diagnostics. This is so because the diagnostics are printed when he issue is found. The code that detected the problem in alive in the stack and can pass down as much context as needed. As an example the patch updates test/Bitcode/invalid.ll. Using a DiagnosticHandler also moves the fatal/non-fatal error decision to the caller. A simple one like llvm-dis can just use fatal errors. The gold plugin needs a bit more complex treatment because of being passed non-bitcode files. An hypothetical interactive tool would make all bitcode errors non-fatal. llvm-svn: 225562
2015-01-10 08:07:30 +08:00
if (!ShouldBeLazy) {
// Parse the full file.
ErrorOr<std::unique_ptr<Module>> M = parseBitcodeFile(*MBOrErr, Context);
if (std::error_code EC = M.getError())
return EC;
return std::move(*M);
Use the DiagnosticHandler to print diagnostics when reading bitcode. The bitcode reading interface used std::error_code to report an error to the callers and it is the callers job to print diagnostics. This is not ideal for error handling or diagnostic reporting: * For error handling, all that the callers care about is 3 possibilities: * It worked * The bitcode file is corrupted/invalid. * The file is not bitcode at all. * For diagnostic, it is user friendly to include far more information about the invalid case so the user can find out what is wrong with the bitcode file. This comes up, for example, when a developer introduces a bug while extending the format. The compromise we had was to have a lot of error codes. With this patch we use the DiagnosticHandler to communicate with the human and std::error_code to communicate with the caller. This allows us to have far fewer error codes and adds the infrastructure to print better diagnostics. This is so because the diagnostics are printed when he issue is found. The code that detected the problem in alive in the stack and can pass down as much context as needed. As an example the patch updates test/Bitcode/invalid.ll. Using a DiagnosticHandler also moves the fatal/non-fatal error decision to the caller. A simple one like llvm-dis can just use fatal errors. The gold plugin needs a bit more complex treatment because of being passed non-bitcode files. An hypothetical interactive tool would make all bitcode errors non-fatal. llvm-svn: 225562
2015-01-10 08:07:30 +08:00
}
// Parse lazily.
std::unique_ptr<MemoryBuffer> LightweightBuf =
MemoryBuffer::getMemBuffer(*MBOrErr, false);
ErrorOr<std::unique_ptr<Module>> M = getLazyBitcodeModule(
std::move(LightweightBuf), Context, true /*ShouldLazyLoadMetadata*/);
if (std::error_code EC = M.getError())
return EC;
return std::move(*M);
}
ErrorOr<std::unique_ptr<LTOModule>>
LTOModule::makeLTOModule(MemoryBufferRef Buffer, TargetOptions options,
LLVMContext &Context, bool ShouldBeLazy) {
ErrorOr<std::unique_ptr<Module>> MOrErr =
parseBitcodeFileImpl(Buffer, Context, ShouldBeLazy);
if (std::error_code EC = MOrErr.getError())
return EC;
std::unique_ptr<Module> &M = *MOrErr;
std::string TripleStr = M->getTargetTriple();
if (TripleStr.empty())
TripleStr = sys::getDefaultTargetTriple();
llvm::Triple Triple(TripleStr);
// find machine architecture for this module
std::string errMsg;
const Target *march = TargetRegistry::lookupTarget(TripleStr, errMsg);
if (!march)
return std::unique_ptr<LTOModule>(nullptr);
2011-04-21 09:54:08 +08:00
// construct LTOModule, hand over ownership of module and target
SubtargetFeatures Features;
Features.getDefaultSubtargetFeatures(Triple);
std::string FeatureStr = Features.getString();
// Set a default CPU for Darwin triples.
std::string CPU;
if (Triple.isOSDarwin()) {
if (Triple.getArch() == llvm::Triple::x86_64)
CPU = "core2";
else if (Triple.getArch() == llvm::Triple::x86)
CPU = "yonah";
else if (Triple.getArch() == llvm::Triple::aarch64)
CPU = "cyclone";
}
TargetMachine *target =
march->createTargetMachine(TripleStr, CPU, FeatureStr, options, None);
M->setDataLayout(target->createDataLayout());
std::unique_ptr<object::IRObjectFile> IRObj(
new object::IRObjectFile(Buffer, std::move(M)));
std::unique_ptr<LTOModule> Ret(new LTOModule(std::move(IRObj), target));
Ret->parseSymbols();
Ret->parseMetadata();
return std::move(Ret);
}
/// Create a MemoryBuffer from a memory range with an optional name.
std::unique_ptr<MemoryBuffer>
LTOModule::makeBuffer(const void *mem, size_t length, StringRef name) {
const char *startPtr = (const char*)mem;
return MemoryBuffer::getMemBuffer(StringRef(startPtr, length), name, false);
}
/// objcClassNameFromExpression - Get string that the data pointer points to.
bool
LTOModule::objcClassNameFromExpression(const Constant *c, std::string &name) {
if (const ConstantExpr *ce = dyn_cast<ConstantExpr>(c)) {
Constant *op = ce->getOperand(0);
if (GlobalVariable *gvn = dyn_cast<GlobalVariable>(op)) {
Constant *cn = gvn->getInitializer();
if (ConstantDataArray *ca = dyn_cast<ConstantDataArray>(cn)) {
if (ca->isCString()) {
name = (".objc_class_name_" + ca->getAsCString()).str();
return true;
}
}
}
}
return false;
}
/// addObjCClass - Parse i386/ppc ObjC class data structure.
void LTOModule::addObjCClass(const GlobalVariable *clgv) {
const ConstantStruct *c = dyn_cast<ConstantStruct>(clgv->getInitializer());
if (!c) return;
// second slot in __OBJC,__class is pointer to superclass name
std::string superclassName;
if (objcClassNameFromExpression(c->getOperand(1), superclassName)) {
auto IterBool =
_undefines.insert(std::make_pair(superclassName, NameAndAttributes()));
if (IterBool.second) {
NameAndAttributes &info = IterBool.first->second;
info.name = IterBool.first->first().data();
info.attributes = LTO_SYMBOL_DEFINITION_UNDEFINED;
info.isFunction = false;
info.symbol = clgv;
}
}
// third slot in __OBJC,__class is pointer to class name
std::string className;
if (objcClassNameFromExpression(c->getOperand(2), className)) {
auto Iter = _defines.insert(className).first;
NameAndAttributes info;
info.name = Iter->first().data();
info.attributes = LTO_SYMBOL_PERMISSIONS_DATA |
LTO_SYMBOL_DEFINITION_REGULAR | LTO_SYMBOL_SCOPE_DEFAULT;
info.isFunction = false;
info.symbol = clgv;
_symbols.push_back(info);
}
}
/// addObjCCategory - Parse i386/ppc ObjC category data structure.
void LTOModule::addObjCCategory(const GlobalVariable *clgv) {
const ConstantStruct *c = dyn_cast<ConstantStruct>(clgv->getInitializer());
if (!c) return;
// second slot in __OBJC,__category is pointer to target class name
std::string targetclassName;
if (!objcClassNameFromExpression(c->getOperand(1), targetclassName))
return;
auto IterBool =
_undefines.insert(std::make_pair(targetclassName, NameAndAttributes()));
if (!IterBool.second)
return;
NameAndAttributes &info = IterBool.first->second;
info.name = IterBool.first->first().data();
info.attributes = LTO_SYMBOL_DEFINITION_UNDEFINED;
info.isFunction = false;
info.symbol = clgv;
}
/// addObjCClassRef - Parse i386/ppc ObjC class list data structure.
void LTOModule::addObjCClassRef(const GlobalVariable *clgv) {
std::string targetclassName;
if (!objcClassNameFromExpression(clgv->getInitializer(), targetclassName))
return;
auto IterBool =
_undefines.insert(std::make_pair(targetclassName, NameAndAttributes()));
if (!IterBool.second)
return;
NameAndAttributes &info = IterBool.first->second;
info.name = IterBool.first->first().data();
info.attributes = LTO_SYMBOL_DEFINITION_UNDEFINED;
info.isFunction = false;
info.symbol = clgv;
}
void LTOModule::addDefinedDataSymbol(const object::BasicSymbolRef &Sym) {
SmallString<64> Buffer;
{
raw_svector_ostream OS(Buffer);
Sym.printName(OS);
}
const GlobalValue *V = IRFile->getSymbolGV(Sym.getRawDataRefImpl());
addDefinedDataSymbol(Buffer.c_str(), V);
}
void LTOModule::addDefinedDataSymbol(const char *Name, const GlobalValue *v) {
// Add to list of defined symbols.
addDefinedSymbol(Name, v, false);
if (!v->hasSection() /* || !isTargetDarwin */)
return;
// Special case i386/ppc ObjC data structures in magic sections:
// The issue is that the old ObjC object format did some strange
// contortions to avoid real linker symbols. For instance, the
// ObjC class data structure is allocated statically in the executable
// that defines that class. That data structures contains a pointer to
// its superclass. But instead of just initializing that part of the
// struct to the address of its superclass, and letting the static and
// dynamic linkers do the rest, the runtime works by having that field
// instead point to a C-string that is the name of the superclass.
// At runtime the objc initialization updates that pointer and sets
// it to point to the actual super class. As far as the linker
// knows it is just a pointer to a string. But then someone wanted the
// linker to issue errors at build time if the superclass was not found.
// So they figured out a way in mach-o object format to use an absolute
// symbols (.objc_class_name_Foo = 0) and a floating reference
// (.reference .objc_class_name_Bar) to cause the linker into erroring when
// a class was missing.
// The following synthesizes the implicit .objc_* symbols for the linker
// from the ObjC data structures generated by the front end.
// special case if this data blob is an ObjC class definition
std::string Section = v->getSection();
if (Section.compare(0, 15, "__OBJC,__class,") == 0) {
if (const GlobalVariable *gv = dyn_cast<GlobalVariable>(v)) {
addObjCClass(gv);
}
}
// special case if this data blob is an ObjC category definition
else if (Section.compare(0, 18, "__OBJC,__category,") == 0) {
if (const GlobalVariable *gv = dyn_cast<GlobalVariable>(v)) {
addObjCCategory(gv);
}
}
// special case if this data blob is the list of referenced classes
else if (Section.compare(0, 18, "__OBJC,__cls_refs,") == 0) {
if (const GlobalVariable *gv = dyn_cast<GlobalVariable>(v)) {
addObjCClassRef(gv);
}
}
}
void LTOModule::addDefinedFunctionSymbol(const object::BasicSymbolRef &Sym) {
SmallString<64> Buffer;
{
raw_svector_ostream OS(Buffer);
Sym.printName(OS);
}
const Function *F =
cast<Function>(IRFile->getSymbolGV(Sym.getRawDataRefImpl()));
addDefinedFunctionSymbol(Buffer.c_str(), F);
}
void LTOModule::addDefinedFunctionSymbol(const char *Name, const Function *F) {
// add to list of defined symbols
addDefinedSymbol(Name, F, true);
}
void LTOModule::addDefinedSymbol(const char *Name, const GlobalValue *def,
bool isFunction) {
// set alignment part log2() can have rounding errors
uint32_t align = def->getAlignment();
uint32_t attr = align ? countTrailingZeros(align) : 0;
// set permissions part
if (isFunction) {
attr |= LTO_SYMBOL_PERMISSIONS_CODE;
} else {
const GlobalVariable *gv = dyn_cast<GlobalVariable>(def);
if (gv && gv->isConstant())
attr |= LTO_SYMBOL_PERMISSIONS_RODATA;
else
attr |= LTO_SYMBOL_PERMISSIONS_DATA;
}
// set definition part
Remove the linker_private and linker_private_weak linkages. These linkages were introduced some time ago, but it was never very clear what exactly their semantics were or what they should be used for. Some investigation found these uses: * utf-16 strings in clang. * non-unnamed_addr strings produced by the sanitizers. It turns out they were just working around a more fundamental problem. For some sections a MachO linker needs a symbol in order to split the section into atoms, and llvm had no idea that was the case. I fixed that in r201700 and it is now safe to use the private linkage. When the object ends up in a section that requires symbols, llvm will use a 'l' prefix instead of a 'L' prefix and things just work. With that, these linkages were already dead, but there was a potential future user in the objc metadata information. I am still looking at CGObjcMac.cpp, but at this point I am convinced that linker_private and linker_private_weak are not what they need. The objc uses are currently split in * Regular symbols (no '\01' prefix). LLVM already directly provides whatever semantics they need. * Uses of a private name (start with "\01L" or "\01l") and private linkage. We can drop the "\01L" and "\01l" prefixes as soon as llvm agrees with clang on L being ok or not for a given section. I have two patches in code review for this. * Uses of private name and weak linkage. The last case is the one that one could think would fit one of these linkages. That is not the case. The semantics are * the linker will merge these symbol by *name*. * the linker will hide them in the final DSO. Given that the merging is done by name, any of the private (or internal) linkages would be a bad match. They allow llvm to rename the symbols, and that is really not what we want. From the llvm point of view, these objects should really be (linkonce|weak)(_odr)?. For now, just keeping the "\01l" prefix is probably the best for these symbols. If we one day want to have a more direct support in llvm, IMHO what we should add is not a linkage, it is just a hidden_symbol attribute. It would be applicable to multiple linkages. For example, on weak it would produce the current behavior we have for objc metadata. On internal, it would be equivalent to private (and we should then remove private). llvm-svn: 203866
2014-03-14 07:18:37 +08:00
if (def->hasWeakLinkage() || def->hasLinkOnceLinkage())
attr |= LTO_SYMBOL_DEFINITION_WEAK;
else if (def->hasCommonLinkage())
attr |= LTO_SYMBOL_DEFINITION_TENTATIVE;
else
attr |= LTO_SYMBOL_DEFINITION_REGULAR;
// set scope part
if (def->hasLocalLinkage())
// Ignore visibility if linkage is local.
attr |= LTO_SYMBOL_SCOPE_INTERNAL;
else if (def->hasHiddenVisibility())
attr |= LTO_SYMBOL_SCOPE_HIDDEN;
else if (def->hasProtectedVisibility())
attr |= LTO_SYMBOL_SCOPE_PROTECTED;
else if (canBeOmittedFromSymbolTable(def))
attr |= LTO_SYMBOL_SCOPE_DEFAULT_CAN_BE_HIDDEN;
else
attr |= LTO_SYMBOL_SCOPE_DEFAULT;
if (def->hasComdat())
attr |= LTO_SYMBOL_COMDAT;
if (isa<GlobalAlias>(def))
attr |= LTO_SYMBOL_ALIAS;
auto Iter = _defines.insert(Name).first;
// fill information structure
NameAndAttributes info;
StringRef NameRef = Iter->first();
info.name = NameRef.data();
assert(info.name[NameRef.size()] == '\0');
info.attributes = attr;
info.isFunction = isFunction;
info.symbol = def;
// add to table of symbols
_symbols.push_back(info);
}
/// addAsmGlobalSymbol - Add a global symbol from module-level ASM to the
/// defined list.
void LTOModule::addAsmGlobalSymbol(const char *name,
lto_symbol_attributes scope) {
auto IterBool = _defines.insert(name);
// only add new define if not already defined
if (!IterBool.second)
return;
NameAndAttributes &info = _undefines[IterBool.first->first().data()];
if (info.symbol == nullptr) {
// FIXME: This is trying to take care of module ASM like this:
//
// module asm ".zerofill __FOO, __foo, _bar_baz_qux, 0"
//
// but is gross and its mother dresses it funny. Have the ASM parser give us
// more details for this type of situation so that we're not guessing so
// much.
// fill information structure
info.name = IterBool.first->first().data();
info.attributes =
LTO_SYMBOL_PERMISSIONS_DATA | LTO_SYMBOL_DEFINITION_REGULAR | scope;
info.isFunction = false;
info.symbol = nullptr;
// add to table of symbols
_symbols.push_back(info);
return;
}
if (info.isFunction)
addDefinedFunctionSymbol(info.name, cast<Function>(info.symbol));
else
addDefinedDataSymbol(info.name, info.symbol);
_symbols.back().attributes &= ~LTO_SYMBOL_SCOPE_MASK;
_symbols.back().attributes |= scope;
}
/// addAsmGlobalSymbolUndef - Add a global symbol from module-level ASM to the
/// undefined list.
void LTOModule::addAsmGlobalSymbolUndef(const char *name) {
auto IterBool = _undefines.insert(std::make_pair(name, NameAndAttributes()));
_asm_undefines.push_back(IterBool.first->first().data());
// we already have the symbol
if (!IterBool.second)
return;
uint32_t attr = LTO_SYMBOL_DEFINITION_UNDEFINED;
attr |= LTO_SYMBOL_SCOPE_DEFAULT;
NameAndAttributes &info = IterBool.first->second;
info.name = IterBool.first->first().data();
info.attributes = attr;
info.isFunction = false;
info.symbol = nullptr;
}
/// Add a symbol which isn't defined just yet to a list to be resolved later.
void LTOModule::addPotentialUndefinedSymbol(const object::BasicSymbolRef &Sym,
bool isFunc) {
SmallString<64> name;
{
raw_svector_ostream OS(name);
Sym.printName(OS);
}
auto IterBool = _undefines.insert(std::make_pair(name, NameAndAttributes()));
// we already have the symbol
if (!IterBool.second)
return;
NameAndAttributes &info = IterBool.first->second;
info.name = IterBool.first->first().data();
const GlobalValue *decl = IRFile->getSymbolGV(Sym.getRawDataRefImpl());
if (decl->hasExternalWeakLinkage())
info.attributes = LTO_SYMBOL_DEFINITION_WEAKUNDEF;
else
info.attributes = LTO_SYMBOL_DEFINITION_UNDEFINED;
info.isFunction = isFunc;
info.symbol = decl;
}
void LTOModule::parseSymbols() {
for (auto &Sym : IRFile->symbols()) {
const GlobalValue *GV = IRFile->getSymbolGV(Sym.getRawDataRefImpl());
uint32_t Flags = Sym.getFlags();
if (Flags & object::BasicSymbolRef::SF_FormatSpecific)
continue;
bool IsUndefined = Flags & object::BasicSymbolRef::SF_Undefined;
if (!GV) {
SmallString<64> Buffer;
{
raw_svector_ostream OS(Buffer);
Sym.printName(OS);
}
const char *Name = Buffer.c_str();
if (IsUndefined)
addAsmGlobalSymbolUndef(Name);
else if (Flags & object::BasicSymbolRef::SF_Global)
addAsmGlobalSymbol(Name, LTO_SYMBOL_SCOPE_DEFAULT);
else
addAsmGlobalSymbol(Name, LTO_SYMBOL_SCOPE_INTERNAL);
continue;
}
2010-08-11 07:46:46 +08:00
auto *F = dyn_cast<Function>(GV);
if (IsUndefined) {
addPotentialUndefinedSymbol(Sym, F != nullptr);
continue;
}
2010-08-11 07:46:46 +08:00
if (F) {
addDefinedFunctionSymbol(Sym);
continue;
}
2010-08-11 07:46:46 +08:00
if (isa<GlobalVariable>(GV)) {
addDefinedDataSymbol(Sym);
continue;
}
assert(isa<GlobalAlias>(GV));
addDefinedDataSymbol(Sym);
}
2010-08-11 07:46:46 +08:00
// make symbols for all undefines
for (StringMap<NameAndAttributes>::iterator u =_undefines.begin(),
e = _undefines.end(); u != e; ++u) {
// If this symbol also has a definition, then don't make an undefine because
// it is a tentative definition.
if (_defines.count(u->getKey())) continue;
NameAndAttributes info = u->getValue();
_symbols.push_back(info);
}
}
/// parseMetadata - Parse metadata from the module
void LTOModule::parseMetadata() {
raw_string_ostream OS(LinkerOpts);
// Linker Options
IR: Split Metadata from Value Split `Metadata` away from the `Value` class hierarchy, as part of PR21532. Assembly and bitcode changes are in the wings, but this is the bulk of the change for the IR C++ API. I have a follow-up patch prepared for `clang`. If this breaks other sub-projects, I apologize in advance :(. Help me compile it on Darwin I'll try to fix it. FWIW, the errors should be easy to fix, so it may be simpler to just fix it yourself. This breaks the build for all metadata-related code that's out-of-tree. Rest assured the transition is mechanical and the compiler should catch almost all of the problems. Here's a quick guide for updating your code: - `Metadata` is the root of a class hierarchy with three main classes: `MDNode`, `MDString`, and `ValueAsMetadata`. It is distinct from the `Value` class hierarchy. It is typeless -- i.e., instances do *not* have a `Type`. - `MDNode`'s operands are all `Metadata *` (instead of `Value *`). - `TrackingVH<MDNode>` and `WeakVH` referring to metadata can be replaced with `TrackingMDNodeRef` and `TrackingMDRef`, respectively. If you're referring solely to resolved `MDNode`s -- post graph construction -- just use `MDNode*`. - `MDNode` (and the rest of `Metadata`) have only limited support for `replaceAllUsesWith()`. As long as an `MDNode` is pointing at a forward declaration -- the result of `MDNode::getTemporary()` -- it maintains a side map of its uses and can RAUW itself. Once the forward declarations are fully resolved RAUW support is dropped on the ground. This means that uniquing collisions on changing operands cause nodes to become "distinct". (This already happened fairly commonly, whenever an operand went to null.) If you're constructing complex (non self-reference) `MDNode` cycles, you need to call `MDNode::resolveCycles()` on each node (or on a top-level node that somehow references all of the nodes). Also, don't do that. Metadata cycles (and the RAUW machinery needed to construct them) are expensive. - An `MDNode` can only refer to a `Constant` through a bridge called `ConstantAsMetadata` (one of the subclasses of `ValueAsMetadata`). As a side effect, accessing an operand of an `MDNode` that is known to be, e.g., `ConstantInt`, takes three steps: first, cast from `Metadata` to `ConstantAsMetadata`; second, extract the `Constant`; third, cast down to `ConstantInt`. The eventual goal is to introduce `MDInt`/`MDFloat`/etc. and have metadata schema owners transition away from using `Constant`s when the type isn't important (and they don't care about referring to `GlobalValue`s). In the meantime, I've added transitional API to the `mdconst` namespace that matches semantics with the old code, in order to avoid adding the error-prone three-step equivalent to every call site. If your old code was: MDNode *N = foo(); bar(isa <ConstantInt>(N->getOperand(0))); baz(cast <ConstantInt>(N->getOperand(1))); bak(cast_or_null <ConstantInt>(N->getOperand(2))); bat(dyn_cast <ConstantInt>(N->getOperand(3))); bay(dyn_cast_or_null<ConstantInt>(N->getOperand(4))); you can trivially match its semantics with: MDNode *N = foo(); bar(mdconst::hasa <ConstantInt>(N->getOperand(0))); baz(mdconst::extract <ConstantInt>(N->getOperand(1))); bak(mdconst::extract_or_null <ConstantInt>(N->getOperand(2))); bat(mdconst::dyn_extract <ConstantInt>(N->getOperand(3))); bay(mdconst::dyn_extract_or_null<ConstantInt>(N->getOperand(4))); and when you transition your metadata schema to `MDInt`: MDNode *N = foo(); bar(isa <MDInt>(N->getOperand(0))); baz(cast <MDInt>(N->getOperand(1))); bak(cast_or_null <MDInt>(N->getOperand(2))); bat(dyn_cast <MDInt>(N->getOperand(3))); bay(dyn_cast_or_null<MDInt>(N->getOperand(4))); - A `CallInst` -- specifically, intrinsic instructions -- can refer to metadata through a bridge called `MetadataAsValue`. This is a subclass of `Value` where `getType()->isMetadataTy()`. `MetadataAsValue` is the *only* class that can legally refer to a `LocalAsMetadata`, which is a bridged form of non-`Constant` values like `Argument` and `Instruction`. It can also refer to any other `Metadata` subclass. (I'll break all your testcases in a follow-up commit, when I propagate this change to assembly.) llvm-svn: 223802
2014-12-10 02:38:53 +08:00
if (Metadata *Val = getModule().getModuleFlag("Linker Options")) {
MDNode *LinkerOptions = cast<MDNode>(Val);
for (unsigned i = 0, e = LinkerOptions->getNumOperands(); i != e; ++i) {
MDNode *MDOptions = cast<MDNode>(LinkerOptions->getOperand(i));
for (unsigned ii = 0, ie = MDOptions->getNumOperands(); ii != ie; ++ii) {
MDString *MDOption = cast<MDString>(MDOptions->getOperand(ii));
OS << " " << MDOption->getString();
}
}
}
// Globals
Mangler Mang;
for (const NameAndAttributes &Sym : _symbols) {
if (!Sym.symbol)
continue;
_target->getObjFileLowering()->emitLinkerFlagsForGlobal(OS, Sym.symbol,
Mang);
}
// Add other interesting metadata here.
}