/usr/src/gnu/usr.bin/clang/liblldELF/../../../llvm/llvm/include/llvm/Object/IRSymtab.h

Bug Summary

File:	src/gnu/usr.bin/clang/liblldELF/../../../llvm/llvm/include/llvm/Object/IRSymtab.h
Warning:	line 328, column 20 Called C++ object pointer is null

Annotated Source Code

Press '?' to see keyboard shortcuts

Show analyzer invocation

clang -cc1 -cc1 -triple amd64-unknown-openbsd7.0 -analyze -disable-free -disable-llvm-verifier -discard-value-names -main-file-name InputFiles.cpp -analyzer-store=region -analyzer-opt-analyze-nested-blocks -analyzer-checker=core -analyzer-checker=apiModeling -analyzer-checker=unix -analyzer-checker=deadcode -analyzer-checker=cplusplus -analyzer-checker=security.insecureAPI.UncheckedReturn -analyzer-checker=security.insecureAPI.getpw -analyzer-checker=security.insecureAPI.gets -analyzer-checker=security.insecureAPI.mktemp -analyzer-checker=security.insecureAPI.mkstemp -analyzer-checker=security.insecureAPI.vfork -analyzer-checker=nullability.NullPassedToNonnull -analyzer-checker=nullability.NullReturnedFromNonnull -analyzer-output plist -w -setup-static-analyzer -mrelocation-model static -mframe-pointer=all -relaxed-aliasing -fno-rounding-math -mconstructor-aliases -munwind-tables -target-cpu x86-64 -tune-cpu generic -debugger-tuning=gdb -fcoverage-compilation-dir=/usr/src/gnu/usr.bin/clang/liblldELF/obj -resource-dir /usr/local/lib/clang/13.0.0 -I /usr/src/gnu/usr.bin/clang/liblldELF/obj/../include/lld/ELF -I /usr/src/gnu/usr.bin/clang/liblldELF/../../../llvm/lld/include -I /usr/src/gnu/usr.bin/clang/liblldELF/../../../llvm/lld/ELF -I /usr/src/gnu/usr.bin/clang/liblldELF/../../../llvm/llvm/include -I /usr/src/gnu/usr.bin/clang/liblldELF/../include -I /usr/src/gnu/usr.bin/clang/liblldELF/obj -I /usr/src/gnu/usr.bin/clang/liblldELF/obj/../include -D NDEBUG -D __STDC_LIMIT_MACROS -D __STDC_CONSTANT_MACROS -D __STDC_FORMAT_MACROS -D LLVM_PREFIX="/usr" -internal-isystem /usr/include/c++/v1 -internal-isystem /usr/local/lib/clang/13.0.0/include -internal-externc-isystem /usr/include -O2 -Wno-unused-parameter -Wwrite-strings -Wno-missing-field-initializers -Wno-long-long -Wno-comment -std=c++14 -fdeprecated-macro -fdebug-compilation-dir=/usr/src/gnu/usr.bin/clang/liblldELF/obj -ferror-limit 19 -fvisibility-inlines-hidden -fwrapv -stack-protector 2 -fno-rtti -fgnuc-version=4.2.1 -vectorize-loops -vectorize-slp -fno-builtin-malloc -fno-builtin-calloc -fno-builtin-realloc -fno-builtin-valloc -fno-builtin-free -fno-builtin-strdup -fno-builtin-strndup -analyzer-output=html -faddrsig -D__GCC_HAVE_DWARF2_CFI_ASM=1 -o /home/ben/Projects/vmm/scan-build/2022-01-12-194120-40624-1 -x c++ /usr/src/gnu/usr.bin/clang/liblldELF/../../../llvm/lld/ELF/InputFiles.cpp

/usr/src/gnu/usr.bin/clang/liblldELF/../../../llvm/lld/ELF/InputFiles.cpp

→

1//===- InputFiles.cpp -----------------------------------------------------===//
2//
3// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
4// See https://llvm.org/LICENSE.txt for license information.
5// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
6//
7//===----------------------------------------------------------------------===//

9#include "InputFiles.h"
10#include "Driver.h"
11#include "InputSection.h"
12#include "LinkerScript.h"
13#include "SymbolTable.h"
14#include "Symbols.h"
15#include "SyntheticSections.h"
16#include "lld/Common/DWARF.h"
17#include "lld/Common/ErrorHandler.h"
18#include "lld/Common/Memory.h"
19#include "llvm/ADT/STLExtras.h"
20#include "llvm/CodeGen/Analysis.h"
21#include "llvm/IR/LLVMContext.h"
22#include "llvm/IR/Module.h"
23#include "llvm/LTO/LTO.h"
24#include "llvm/MC/StringTableBuilder.h"
25#include "llvm/Object/ELFObjectFile.h"
26#include "llvm/Support/ARMAttributeParser.h"
27#include "llvm/Support/ARMBuildAttributes.h"
28#include "llvm/Support/Endian.h"
29#include "llvm/Support/Path.h"
30#include "llvm/Support/RISCVAttributeParser.h"
31#include "llvm/Support/TarWriter.h"
32#include "llvm/Support/raw_ostream.h"

34using namespace llvm;
35using namespace llvm::ELF;
36using namespace llvm::object;
37using namespace llvm::sys;
38using namespace llvm::sys::fs;
39using namespace llvm::support::endian;
40using namespace lld;
41using namespace lld::elf;

43bool InputFile::isInGroup;
44uint32_t InputFile::nextGroupId;

46std::vector<ArchiveFile *> elf::archiveFiles;
47std::vector<BinaryFile *> elf::binaryFiles;
48std::vector<BitcodeFile *> elf::bitcodeFiles;
49std::vector<LazyObjFile *> elf::lazyObjFiles;
50std::vector<InputFile *> elf::objectFiles;
51std::vector<SharedFile *> elf::sharedFiles;

53std::unique_ptr<TarWriter> elf::tar;

55// Returns "<internal>", "foo.a(bar.o)" or "baz.o".
56std::string lld::toString(const InputFile *f) {
if (!f)
  return "<internal>";

if (f->toStringCache.empty()) {
  if (f->archiveName.empty())
    f->toStringCache = std::string(f->getName());
  else
    f->toStringCache = (f->archiveName + "(" + f->getName() + ")").str();
}
return f->toStringCache;
67}

69static ELFKind getELFKind(MemoryBufferRef mb, StringRef archiveName) {
unsigned char size;
unsigned char endian;
std::tie(size, endian) = getElfArchType(mb.getBuffer());

auto report = [&](StringRef msg) {
  StringRef filename = mb.getBufferIdentifier();
  if (archiveName.empty())
    fatal(filename + ": " + msg);
  else
    fatal(archiveName + "(" + filename + "): " + msg);
};

if (!mb.getBuffer().startswith(ElfMagic))
  report("not an ELF file");
if (endian != ELFDATA2LSB && endian != ELFDATA2MSB)
  report("corrupted ELF file: invalid data encoding");
if (size != ELFCLASS32 && size != ELFCLASS64)
  report("corrupted ELF file: invalid file class");

size_t bufSize = mb.getBuffer().size();
if ((size == ELFCLASS32 && bufSize < sizeof(Elf32_Ehdr)) ||
    (size == ELFCLASS64 && bufSize < sizeof(Elf64_Ehdr)))
  report("corrupted ELF file: file is too short");

if (size == ELFCLASS32)
  return (endian == ELFDATA2LSB) ? ELF32LEKind : ELF32BEKind;
return (endian == ELFDATA2LSB) ? ELF64LEKind : ELF64BEKind;
97}

99InputFile::InputFile(Kind k, MemoryBufferRef m)
  : mb(m), groupId(nextGroupId), fileKind(k) {
// All files within the same --{start,end}-group get the same group ID.
// Otherwise, a new file will get a new group ID.
if (!isInGroup)
  ++nextGroupId;
105}

107Optional<MemoryBufferRef> elf::readFile(StringRef path) {
llvm::TimeTraceScope timeScope("Load input files", path);

// The --chroot option changes our virtual root directory.
// This is useful when you are dealing with files created by --reproduce.
if (!config->chroot.empty() && path.startswith("/"))
  path = saver.save(config->chroot + path);

log(path);
config->dependencyFiles.insert(llvm::CachedHashString(path));

auto mbOrErr = MemoryBuffer::getFile(path, /*IsText=*/false,
                                     /*RequiresNullTerminator=*/false);
if (auto ec = mbOrErr.getError()) {
  error("cannot open " + path + ": " + ec.message());
  return None;
}

std::unique_ptr<MemoryBuffer> &mb = *mbOrErr;
MemoryBufferRef mbref = mb->getMemBufferRef();
make<std::unique_ptr<MemoryBuffer>>(std::move(mb)); // take MB ownership

if (tar)
  tar->append(relativeToRoot(path), mbref.getBuffer());
return mbref;
132}

134// All input object files must be for the same architecture
135// (e.g. it does not make sense to link x86 object files with
136// MIPS object files.) This function checks for that error.
137static bool isCompatible(InputFile *file) {
if (!file->isElf() && !isa<BitcodeFile>(file))
  return true;

if (file->ekind == config->ekind && file->emachine == config->emachine) {
  if (config->emachine != EM_MIPS)
    return true;
  if (isMipsN32Abi(file) == config->mipsN32Abi)
    return true;
}

StringRef target =
    !config->bfdname.empty() ? config->bfdname : config->emulation;
if (!target.empty()) {
  error(toString(file) + " is incompatible with " + target);
  return false;
}

InputFile *existing;
if (!objectFiles.empty())
  existing = objectFiles[0];
else if (!sharedFiles.empty())
  existing = sharedFiles[0];
else if (!bitcodeFiles.empty())
  existing = bitcodeFiles[0];
else
  llvm_unreachable("Must have -m, OUTPUT_FORMAT or existing input file to "__builtin_unreachable()
                   "determine target emulation")__builtin_unreachable();

error(toString(file) + " is incompatible with " + toString(existing));
return false;
168}

170template <class ELFT> static void doParseFile(InputFile *file) {
if (!isCompatible(file))
  return;

// Binary file
if (auto *f = dyn_cast<BinaryFile>(file)) {
  binaryFiles.push_back(f);
  f->parse();
  return;
}

// .a file
if (auto *f = dyn_cast<ArchiveFile>(file)) {
  archiveFiles.push_back(f);
  f->parse();
  return;
}

// Lazy object file
if (auto *f = dyn_cast<LazyObjFile>(file)) {
  lazyObjFiles.push_back(f);
  f->parse<ELFT>();
  return;
}

if (config->trace)
  message(toString(file));

// .so file
if (auto *f = dyn_cast<SharedFile>(file)) {
  f->parse<ELFT>();
  return;
}

// LLVM bitcode file
if (auto *f = dyn_cast<BitcodeFile>(file)) {
  bitcodeFiles.push_back(f);
  f->parse<ELFT>();
  return;
}

// Regular object file
objectFiles.push_back(file);
cast<ObjFile<ELFT>>(file)->parse();
214}

216// Add symbols in File to the symbol table.
217void elf::parseFile(InputFile *file) {
switch (config->ekind) {
case ELF32LEKind:
  doParseFile<ELF32LE>(file);
  return;
case ELF32BEKind:
  doParseFile<ELF32BE>(file);
  return;
case ELF64LEKind:
  doParseFile<ELF64LE>(file);
  return;
case ELF64BEKind:
  doParseFile<ELF64BE>(file);
  return;
default:
  llvm_unreachable("unknown ELFT")__builtin_unreachable();
}
234}

236// Concatenates arguments to construct a string representing an error location.
237static std::string createFileLineMsg(StringRef path, unsigned line) {
std::string filename = std::string(path::filename(path));
std::string lineno = ":" + std::to_string(line);
if (filename == path)
  return filename + lineno;
return filename + lineno + " (" + path.str() + lineno + ")";
243}

245template <class ELFT>
246static std::string getSrcMsgAux(ObjFile<ELFT> &file, const Symbol &sym,
                              InputSectionBase &sec, uint64_t offset) {
// In DWARF, functions and variables are stored to different places.
// First, lookup a function for a given offset.
if (Optional<DILineInfo> info = file.getDILineInfo(&sec, offset))
  return createFileLineMsg(info->FileName, info->Line);

// If it failed, lookup again as a variable.
if (Optional<std::pair<std::string, unsigned>> fileLine =
        file.getVariableLoc(sym.getName()))
  return createFileLineMsg(fileLine->first, fileLine->second);

// File.sourceFile contains STT_FILE symbol, and that is a last resort.
return std::string(file.sourceFile);
260}

262std::string InputFile::getSrcMsg(const Symbol &sym, InputSectionBase &sec,
                               uint64_t offset) {
if (kind() != ObjKind)
  return "";
switch (config->ekind) {
default:
  llvm_unreachable("Invalid kind")__builtin_unreachable();
case ELF32LEKind:
  return getSrcMsgAux(cast<ObjFile<ELF32LE>>(*this), sym, sec, offset);
case ELF32BEKind:
  return getSrcMsgAux(cast<ObjFile<ELF32BE>>(*this), sym, sec, offset);
case ELF64LEKind:
  return getSrcMsgAux(cast<ObjFile<ELF64LE>>(*this), sym, sec, offset);
case ELF64BEKind:
  return getSrcMsgAux(cast<ObjFile<ELF64BE>>(*this), sym, sec, offset);
}
278}

280StringRef InputFile::getNameForScript() const {
if (archiveName.empty())
  return getName();

if (nameForScriptCache.empty())
  nameForScriptCache = (archiveName + Twine(':') + getName()).str();

return nameForScriptCache;
288}

290template <class ELFT> DWARFCache *ObjFile<ELFT>::getDwarf() {
llvm::call_once(initDwarf, [this]() {
  dwarf = std::make_unique<DWARFCache>(std::make_unique<DWARFContext>(
      std::make_unique<LLDDwarfObj<ELFT>>(this), "",
      [&](Error err) { warn(getName() + ": " + toString(std::move(err))); },
      [&](Error warning) {
        warn(getName() + ": " + toString(std::move(warning)));
      }));
});

return dwarf.get();
301}

303// Returns the pair of file name and line number describing location of data
304// object (variable, array, etc) definition.
305template <class ELFT>
306Optional<std::pair<std::string, unsigned>>
307ObjFile<ELFT>::getVariableLoc(StringRef name) {
return getDwarf()->getVariableLoc(name);
309}

311// Returns source line information for a given offset
312// using DWARF debug info.
313template <class ELFT>
314Optional<DILineInfo> ObjFile<ELFT>::getDILineInfo(InputSectionBase *s,
                                                uint64_t offset) {
// Detect SectionIndex for specified section.
uint64_t sectionIndex = object::SectionedAddress::UndefSection;
ArrayRef<InputSectionBase *> sections = s->file->getSections();
for (uint64_t curIndex = 0; curIndex < sections.size(); ++curIndex) {
  if (s == sections[curIndex]) {
    sectionIndex = curIndex;
    break;
  }
}

return getDwarf()->getDILineInfo(offset, sectionIndex);
327}

329ELFFileBase::ELFFileBase(Kind k, MemoryBufferRef mb) : InputFile(k, mb) {
ekind = getELFKind(mb, "");

switch (ekind) {
case ELF32LEKind:
  init<ELF32LE>();
  break;
case ELF32BEKind:
  init<ELF32BE>();
  break;
case ELF64LEKind:
  init<ELF64LE>();
  break;
case ELF64BEKind:
  init<ELF64BE>();
  break;
default:
  llvm_unreachable("getELFKind")__builtin_unreachable();
}
348}

350template <typename Elf_Shdr>
351static const Elf_Shdr *findSection(ArrayRef<Elf_Shdr> sections, uint32_t type) {
for (const Elf_Shdr &sec : sections)
  if (sec.sh_type == type)
    return &sec;
return nullptr;
356}

358template <class ELFT> void ELFFileBase::init() {
using Elf_Shdr = typename ELFT::Shdr;
using Elf_Sym = typename ELFT::Sym;

// Initialize trivial attributes.
const ELFFile<ELFT> &obj = getObj<ELFT>();
emachine = obj.getHeader().e_machine;
osabi = obj.getHeader().e_ident[llvm::ELF::EI_OSABI];
abiVersion = obj.getHeader().e_ident[llvm::ELF::EI_ABIVERSION];

ArrayRef<Elf_Shdr> sections = CHECK(obj.sections(), this)check2((obj.sections()), [&] { return toString(this); });

// Find a symbol table.
bool isDSO =
    (identify_magic(mb.getBuffer()) == file_magic::elf_shared_object);
const Elf_Shdr *symtabSec =
    findSection(sections, isDSO ? SHT_DYNSYM : SHT_SYMTAB);

if (!symtabSec)
  return;

// Initialize members corresponding to a symbol table.
firstGlobal = symtabSec->sh_info;

ArrayRef<Elf_Sym> eSyms = CHECK(obj.symbols(symtabSec), this)check2((obj.symbols(symtabSec)), [&] { return toString(this
); });
if (firstGlobal == 0 || firstGlobal > eSyms.size())
  fatal(toString(this) + ": invalid sh_info in symbol table");

elfSyms = reinterpret_cast<const void *>(eSyms.data());
numELFSyms = eSyms.size();
stringTable = CHECK(obj.getStringTableForSymtab(*symtabSec, sections), this)check2((obj.getStringTableForSymtab(*symtabSec, sections)), [
&] { return toString(this); });
389}

391template <class ELFT>
392uint32_t ObjFile<ELFT>::getSectionIndex(const Elf_Sym &sym) const {
return CHECK(check2((this->getObj().getSectionIndex(sym, getELFSyms<
ELFT>(), shndxTable)), [&] { return toString(this); })
    this->getObj().getSectionIndex(sym, getELFSyms<ELFT>(), shndxTable),check2((this->getObj().getSectionIndex(sym, getELFSyms<
ELFT>(), shndxTable)), [&] { return toString(this); })
    this)check2((this->getObj().getSectionIndex(sym, getELFSyms<
ELFT>(), shndxTable)), [&] { return toString(this); });
396}

398template <class ELFT> ArrayRef<Symbol *> ObjFile<ELFT>::getLocalSymbols() {
if (this->symbols.empty())
  return {};
return makeArrayRef(this->symbols).slice(1, this->firstGlobal - 1);
402}

404template <class ELFT> ArrayRef<Symbol *> ObjFile<ELFT>::getGlobalSymbols() {
return makeArrayRef(this->symbols).slice(this->firstGlobal);
406}

408template <class ELFT> void ObjFile<ELFT>::parse(bool ignoreComdats) {
// Read a section table. justSymbols is usually false.
if (this->justSymbols)
  initializeJustSymbols();
else
  initializeSections(ignoreComdats);

// Read a symbol table.
initializeSymbols();
417}

419// Sections with SHT_GROUP and comdat bits define comdat section groups.
420// They are identified and deduplicated by group name. This function
421// returns a group name.
422template <class ELFT>
423StringRef ObjFile<ELFT>::getShtGroupSignature(ArrayRef<Elf_Shdr> sections,
                                            const Elf_Shdr &sec) {
typename ELFT::SymRange symbols = this->getELFSyms<ELFT>();
if (sec.sh_info >= symbols.size())
  fatal(toString(this) + ": invalid symbol index");
const typename ELFT::Sym &sym = symbols[sec.sh_info];
StringRef signature = CHECK(sym.getName(this->stringTable), this)check2((sym.getName(this->stringTable)), [&] { return toString
(this); });

// As a special case, if a symbol is a section symbol and has no name,
// we use a section name as a signature.
//
// Such SHT_GROUP sections are invalid from the perspective of the ELF
// standard, but GNU gold 1.14 (the newest version as of July 2017) or
// older produce such sections as outputs for the -r option, so we need
// a bug-compatibility.
if (signature.empty() && sym.getType() == STT_SECTION)
  return getSectionName(sec);
return signature;
441}

443template <class ELFT>
444bool ObjFile<ELFT>::shouldMerge(const Elf_Shdr &sec, StringRef name) {
if (!(sec.sh_flags & SHF_MERGE))
  return false;

// On a regular link we don't merge sections if -O0 (default is -O1). This
// sometimes makes the linker significantly faster, although the output will
// be bigger.
//
// Doing the same for -r would create a problem as it would combine sections
// with different sh_entsize. One option would be to just copy every SHF_MERGE
// section as is to the output. While this would produce a valid ELF file with
// usable SHF_MERGE sections, tools like (llvm-)?dwarfdump get confused when
// they see two .debug_str. We could have separate logic for combining
// SHF_MERGE sections based both on their name and sh_entsize, but that seems
// to be more trouble than it is worth. Instead, we just use the regular (-O1)
// logic for -r.
if (config->optimize == 0 && !config->relocatable)
  return false;

// A mergeable section with size 0 is useless because they don't have
// any data to merge. A mergeable string section with size 0 can be
// argued as invalid because it doesn't end with a null character.
// We'll avoid a mess by handling them as if they were non-mergeable.
if (sec.sh_size == 0)
  return false;

// Check for sh_entsize. The ELF spec is not clear about the zero
// sh_entsize. It says that "the member [sh_entsize] contains 0 if
// the section does not hold a table of fixed-size entries". We know
// that Rust 1.13 produces a string mergeable section with a zero
// sh_entsize. Here we just accept it rather than being picky about it.
uint64_t entSize = sec.sh_entsize;
if (entSize == 0)
  return false;
if (sec.sh_size % entSize)
  fatal(toString(this) + ":(" + name + "): SHF_MERGE section size (" +
        Twine(sec.sh_size) + ") must be a multiple of sh_entsize (" +
        Twine(entSize) + ")");

if (sec.sh_flags & SHF_WRITE)
  fatal(toString(this) + ":(" + name +
        "): writable SHF_MERGE section is not supported");

return true;
488}

490// This is for --just-symbols.
491//
492// --just-symbols is a very minor feature that allows you to link your
493// output against other existing program, so that if you load both your
494// program and the other program into memory, your output can refer the
495// other program's symbols.
496//
497// When the option is given, we link "just symbols". The section table is
498// initialized with null pointers.
499template <class ELFT> void ObjFile<ELFT>::initializeJustSymbols() {
ArrayRef<Elf_Shdr> sections = CHECK(this->getObj().sections(), this)check2((this->getObj().sections()), [&] { return toString
(this); });
this->sections.resize(sections.size());
502}

504// An ELF object file may contain a `.deplibs` section. If it exists, the
505// section contains a list of library specifiers such as `m` for libm. This
506// function resolves a given name by finding the first matching library checking
507// the various ways that a library can be specified to LLD. This ELF extension
508// is a form of autolinking and is called `dependent libraries`. It is currently
509// unique to LLVM and lld.
510static void addDependentLibrary(StringRef specifier, const InputFile *f) {
if (!config->dependentLibraries)
  return;
if (fs::exists(specifier))
  driver->addFile(specifier, /*withLOption=*/false);
else if (Optional<std::string> s = findFromSearchPaths(specifier))
  driver->addFile(*s, /*withLOption=*/true);
else if (Optional<std::string> s = searchLibraryBaseName(specifier))
  driver->addFile(*s, /*withLOption=*/true);
else
  error(toString(f) +
        ": unable to find library from dependent library specifier: " +
        specifier);
523}

525// Record the membership of a section group so that in the garbage collection
526// pass, section group members are kept or discarded as a unit.
527template <class ELFT>
528static void handleSectionGroup(ArrayRef<InputSectionBase *> sections,
                             ArrayRef<typename ELFT::Word> entries) {
bool hasAlloc = false;
for (uint32_t index : entries.slice(1)) {
  if (index >= sections.size())
    return;
  if (InputSectionBase *s = sections[index])
    if (s != &InputSection::discarded && s->flags & SHF_ALLOC)
      hasAlloc = true;
}

// If any member has the SHF_ALLOC flag, the whole group is subject to garbage
// collection. See the comment in markLive(). This rule retains .debug_types
// and .rela.debug_types.
if (!hasAlloc)
  return;

// Connect the members in a circular doubly-linked list via
// nextInSectionGroup.
InputSectionBase *head;
InputSectionBase *prev = nullptr;
for (uint32_t index : entries.slice(1)) {
  InputSectionBase *s = sections[index];
  if (!s || s == &InputSection::discarded)
    continue;
  if (prev)
    prev->nextInSectionGroup = s;
  else
    head = s;
  prev = s;
}
if (prev)
  prev->nextInSectionGroup = head;
561}

563template <class ELFT>
564void ObjFile<ELFT>::initializeSections(bool ignoreComdats) {
const ELFFile<ELFT> &obj = this->getObj();

ArrayRef<Elf_Shdr> objSections = CHECK(obj.sections(), this)check2((obj.sections()), [&] { return toString(this); });
uint64_t size = objSections.size();
this->sections.resize(size);
this->sectionStringTable =
    CHECK(obj.getSectionStringTable(objSections), this)check2((obj.getSectionStringTable(objSections)), [&] { return
 toString(this); });

std::vector<ArrayRef<Elf_Word>> selectedGroups;

for (size_t i = 0, e = objSections.size(); i < e; ++i) {
  if (this->sections[i] == &InputSection::discarded)
    continue;
  const Elf_Shdr &sec = objSections[i];

  if (sec.sh_type == ELF::SHT_LLVM_CALL_GRAPH_PROFILE)
    cgProfileSectionIndex = i;

  // SHF_EXCLUDE'ed sections are discarded by the linker. However,
  // if -r is given, we'll let the final link discard such sections.
  // This is compatible with GNU.
  if ((sec.sh_flags & SHF_EXCLUDE) && !config->relocatable) {
    if (sec.sh_type == SHT_LLVM_ADDRSIG) {
      // We ignore the address-significance table if we know that the object
      // file was created by objcopy or ld -r. This is because these tools
      // will reorder the symbols in the symbol table, invalidating the data
      // in the address-significance table, which refers to symbols by index.
      if (sec.sh_link != 0)
        this->addrsigSec = &sec;
      else if (config->icf == ICFLevel::Safe)
        warn(toString(this) +
             ": --icf=safe conservatively ignores "
             "SHT_LLVM_ADDRSIG [index " +
             Twine(i) +
             "] with sh_link=0 "
             "(likely created using objcopy or ld -r)");
    }
    this->sections[i] = &InputSection::discarded;
    continue;
  }

  switch (sec.sh_type) {
  case SHT_GROUP: {
    // De-duplicate section groups by their signatures.
    StringRef signature = getShtGroupSignature(objSections, sec);
    this->sections[i] = &InputSection::discarded;

    ArrayRef<Elf_Word> entries =
        CHECK(obj.template getSectionContentsAsArray<Elf_Word>(sec), this)check2((obj.template getSectionContentsAsArray<Elf_Word>
(sec)), [&] { return toString(this); });
    if (entries.empty())
      fatal(toString(this) + ": empty SHT_GROUP");

    Elf_Word flag = entries[0];
    if (flag && flag != GRP_COMDAT)
      fatal(toString(this) + ": unsupported SHT_GROUP format");

    bool keepGroup =
        (flag & GRP_COMDAT) == 0 || ignoreComdats ||
        symtab->comdatGroups.try_emplace(CachedHashStringRef(signature), this)
            .second;
    if (keepGroup) {
      if (config->relocatable)
        this->sections[i] = createInputSection(sec);
      selectedGroups.push_back(entries);
      continue;
    }

    // Otherwise, discard group members.
    for (uint32_t secIndex : entries.slice(1)) {
      if (secIndex >= size)
        fatal(toString(this) +
              ": invalid section index in group: " + Twine(secIndex));
      this->sections[secIndex] = &InputSection::discarded;
    }
    break;
  }
  case SHT_SYMTAB_SHNDX:
    shndxTable = CHECK(obj.getSHNDXTable(sec, objSections), this)check2((obj.getSHNDXTable(sec, objSections)), [&] { return
 toString(this); });
    break;
  case SHT_SYMTAB:
  case SHT_STRTAB:
  case SHT_REL:
  case SHT_RELA:
  case SHT_NULL:
    break;
  default:
    this->sections[i] = createInputSection(sec);
  }
}

// We have a second loop. It is used to:
// 1) handle SHF_LINK_ORDER sections.
// 2) create SHT_REL[A] sections. In some cases the section header index of a
//    relocation section may be smaller than that of the relocated section. In
//    such cases, the relocation section would attempt to reference a target
//    section that has not yet been created. For simplicity, delay creation of
//    relocation sections until now.
for (size_t i = 0, e = objSections.size(); i < e; ++i) {
  if (this->sections[i] == &InputSection::discarded)
    continue;
  const Elf_Shdr &sec = objSections[i];

  if (sec.sh_type == SHT_REL || sec.sh_type == SHT_RELA)
    this->sections[i] = createInputSection(sec);

  // A SHF_LINK_ORDER section with sh_link=0 is handled as if it did not have
  // the flag.
  if (!(sec.sh_flags & SHF_LINK_ORDER) || !sec.sh_link)
    continue;

  InputSectionBase *linkSec = nullptr;
  if (sec.sh_link < this->sections.size())
    linkSec = this->sections[sec.sh_link];
  if (!linkSec)
    fatal(toString(this) + ": invalid sh_link index: " + Twine(sec.sh_link));

  // A SHF_LINK_ORDER section is discarded if its linked-to section is
  // discarded.
  InputSection *isec = cast<InputSection>(this->sections[i]);
  linkSec->dependentSections.push_back(isec);
  if (!isa<InputSection>(linkSec))
    error("a section " + isec->name +
          " with SHF_LINK_ORDER should not refer a non-regular section: " +
          toString(linkSec));
}

for (ArrayRef<Elf_Word> entries : selectedGroups)
  handleSectionGroup<ELFT>(this->sections, entries);
693}

695// For ARM only, to set the EF_ARM_ABI_FLOAT_SOFT or EF_ARM_ABI_FLOAT_HARD
696// flag in the ELF Header we need to look at Tag_ABI_VFP_args to find out how
697// the input objects have been compiled.
698static void updateARMVFPArgs(const ARMAttributeParser &attributes,
                           const InputFile *f) {
Optional<unsigned> attr =
    attributes.getAttributeValue(ARMBuildAttrs::ABI_VFP_args);
if (!attr.hasValue())
  // If an ABI tag isn't present then it is implicitly given the value of 0
  // which maps to ARMBuildAttrs::BaseAAPCS. However many assembler files,
  // including some in glibc that don't use FP args (and should have value 3)
  // don't have the attribute so we do not consider an implicit value of 0
  // as a clash.
  return;

unsigned vfpArgs = attr.getValue();
ARMVFPArgKind arg;
switch (vfpArgs) {
case ARMBuildAttrs::BaseAAPCS:
  arg = ARMVFPArgKind::Base;
  break;
case ARMBuildAttrs::HardFPAAPCS:
  arg = ARMVFPArgKind::VFP;
  break;
case ARMBuildAttrs::ToolChainFPPCS:
  // Tool chain specific convention that conforms to neither AAPCS variant.
  arg = ARMVFPArgKind::ToolChain;
  break;
case ARMBuildAttrs::CompatibleFPAAPCS:
  // Object compatible with all conventions.
  return;
default:
  error(toString(f) + ": unknown Tag_ABI_VFP_args value: " + Twine(vfpArgs));
  return;
}
// Follow ld.bfd and error if there is a mix of calling conventions.
if (config->armVFPArgs != arg && config->armVFPArgs != ARMVFPArgKind::Default)
  error(toString(f) + ": incompatible Tag_ABI_VFP_args");
else
  config->armVFPArgs = arg;
735}

737// The ARM support in lld makes some use of instructions that are not available
738// on all ARM architectures. Namely:
739// - Use of BLX instruction for interworking between ARM and Thumb state.
740// - Use of the extended Thumb branch encoding in relocation.
741// - Use of the MOVT/MOVW instructions in Thumb Thunks.
742// The ARM Attributes section contains information about the architecture chosen
743// at compile time. We follow the convention that if at least one input object
744// is compiled with an architecture that supports these features then lld is
745// permitted to use them.
746static void updateSupportedARMFeatures(const ARMAttributeParser &attributes) {
Optional<unsigned> attr =
    attributes.getAttributeValue(ARMBuildAttrs::CPU_arch);
if (!attr.hasValue())
  return;
auto arch = attr.getValue();
switch (arch) {
case ARMBuildAttrs::Pre_v4:
case ARMBuildAttrs::v4:
case ARMBuildAttrs::v4T:
  // Architectures prior to v5 do not support BLX instruction
  break;
case ARMBuildAttrs::v5T:
case ARMBuildAttrs::v5TE:
case ARMBuildAttrs::v5TEJ:
case ARMBuildAttrs::v6:
case ARMBuildAttrs::v6KZ:
case ARMBuildAttrs::v6K:
  config->armHasBlx = true;
  // Architectures used in pre-Cortex processors do not support
  // The J1 = 1 J2 = 1 Thumb branch range extension, with the exception
  // of Architecture v6T2 (arm1156t2-s and arm1156t2f-s) that do.
  break;
default:
  // All other Architectures have BLX and extended branch encoding
  config->armHasBlx = true;
  config->armJ1J2BranchEncoding = true;
  if (arch != ARMBuildAttrs::v6_M && arch != ARMBuildAttrs::v6S_M)
    // All Architectures used in Cortex processors with the exception
    // of v6-M and v6S-M have the MOVT and MOVW instructions.
    config->armHasMovtMovw = true;
  break;
}
779}

781// If a source file is compiled with x86 hardware-assisted call flow control
782// enabled, the generated object file contains feature flags indicating that
783// fact. This function reads the feature flags and returns it.
784//
785// Essentially we want to read a single 32-bit value in this function, but this
786// function is rather complicated because the value is buried deep inside a
787// .note.gnu.property section.
788//
789// The section consists of one or more NOTE records. Each NOTE record consists
790// of zero or more type-length-value fields. We want to find a field of a
791// certain type. It seems a bit too much to just store a 32-bit value, perhaps
792// the ABI is unnecessarily complicated.
793template <class ELFT> static uint32_t readAndFeatures(const InputSection &sec) {
using Elf_Nhdr = typename ELFT::Nhdr;
using Elf_Note = typename ELFT::Note;

uint32_t featuresSet = 0;
ArrayRef<uint8_t> data = sec.data();
auto reportFatal = [&](const uint8_t *place, const char *msg) {
  fatal(toString(sec.file) + ":(" + sec.name + "+0x" +
        Twine::utohexstr(place - sec.data().data()) + "): " + msg);
};
while (!data.empty()) {
  // Read one NOTE record.
  auto *nhdr = reinterpret_cast<const Elf_Nhdr *>(data.data());
  if (data.size() < sizeof(Elf_Nhdr) || data.size() < nhdr->getSize())
    reportFatal(data.data(), "data is too short");

  Elf_Note note(*nhdr);
  if (nhdr->n_type != NT_GNU_PROPERTY_TYPE_0 || note.getName() != "GNU") {
    data = data.slice(nhdr->getSize());
    continue;
  }

  uint32_t featureAndType = config->emachine == EM_AARCH64
                                ? GNU_PROPERTY_AARCH64_FEATURE_1_AND
                                : GNU_PROPERTY_X86_FEATURE_1_AND;

  // Read a body of a NOTE record, which consists of type-length-value fields.
  ArrayRef<uint8_t> desc = note.getDesc();
  while (!desc.empty()) {
    const uint8_t *place = desc.data();
    if (desc.size() < 8)
      reportFatal(place, "program property is too short");
    uint32_t type = read32<ELFT::TargetEndianness>(desc.data());
    uint32_t size = read32<ELFT::TargetEndianness>(desc.data() + 4);
    desc = desc.slice(8);
    if (desc.size() < size)
      reportFatal(place, "program property is too short");

    if (type == featureAndType) {
      // We found a FEATURE_1_AND field. There may be more than one of these
      // in a .note.gnu.property section, for a relocatable object we
      // accumulate the bits set.
      if (size < 4)
        reportFatal(place, "FEATURE_1_AND entry is too short");
      featuresSet |= read32<ELFT::TargetEndianness>(desc.data());
    }

    // Padding is present in the note descriptor, if necessary.
    desc = desc.slice(alignTo<(ELFT::Is64Bits ? 8 : 4)>(size));
  }

  // Go to next NOTE record to look for more FEATURE_1_AND descriptions.
  data = data.slice(nhdr->getSize());
}

return featuresSet;
849}

851template <class ELFT>
852InputSectionBase *ObjFile<ELFT>::getRelocTarget(const Elf_Shdr &sec) {
uint32_t idx = sec.sh_info;
if (idx >= this->sections.size())
  fatal(toString(this) + ": invalid relocated section index: " + Twine(idx));
InputSectionBase *target = this->sections[idx];

// Strictly speaking, a relocation section must be included in the
// group of the section it relocates. However, LLVM 3.3 and earlier
// would fail to do so, so we gracefully handle that case.
if (target == &InputSection::discarded)
  return nullptr;

if (!target)
  fatal(toString(this) + ": unsupported relocation reference");
return target;
867}

869// Create a regular InputSection class that has the same contents
870// as a given section.
871static InputSection *toRegularSection(MergeInputSection *sec) {
return make<InputSection>(sec->file, sec->flags, sec->type, sec->alignment,
                          sec->data(), sec->name);
874}

876template <class ELFT>
877InputSectionBase *ObjFile<ELFT>::createInputSection(const Elf_Shdr &sec) {
StringRef name = getSectionName(sec);

if (config->emachine == EM_ARM && sec.sh_type == SHT_ARM_ATTRIBUTES) {
  ARMAttributeParser attributes;
  ArrayRef<uint8_t> contents = check(this->getObj().getSectionContents(sec));
  if (Error e = attributes.parse(contents, config->ekind == ELF32LEKind
                                               ? support::little
                                               : support::big)) {
    auto *isec = make<InputSection>(*this, sec, name);
    warn(toString(isec) + ": " + llvm::toString(std::move(e)));
  } else {
    updateSupportedARMFeatures(attributes);
    updateARMVFPArgs(attributes, this);

    // FIXME: Retain the first attribute section we see. The eglibc ARM
    // dynamic loaders require the presence of an attribute section for dlopen
    // to work. In a full implementation we would merge all attribute
    // sections.
    if (in.attributes == nullptr) {
      in.attributes = make<InputSection>(*this, sec, name);
      return in.attributes;
    }
    return &InputSection::discarded;
  }
}

if (config->emachine == EM_RISCV && sec.sh_type == SHT_RISCV_ATTRIBUTES) {
  RISCVAttributeParser attributes;
  ArrayRef<uint8_t> contents = check(this->getObj().getSectionContents(sec));
  if (Error e = attributes.parse(contents, support::little)) {
    auto *isec = make<InputSection>(*this, sec, name);
    warn(toString(isec) + ": " + llvm::toString(std::move(e)));
  } else {
    // FIXME: Validate arch tag contains C if and only if EF_RISCV_RVC is
    // present.

    // FIXME: Retain the first attribute section we see. Tools such as
    // llvm-objdump make use of the attribute section to determine which
    // standard extensions to enable. In a full implementation we would merge
    // all attribute sections.
    if (in.attributes == nullptr) {
      in.attributes = make<InputSection>(*this, sec, name);
      return in.attributes;
    }
    return &InputSection::discarded;
  }
}

switch (sec.sh_type) {
case SHT_LLVM_DEPENDENT_LIBRARIES: {
  if (config->relocatable)
    break;
  ArrayRef<char> data =
      CHECK(this->getObj().template getSectionContentsAsArray<char>(sec), this)check2((this->getObj().template getSectionContentsAsArray<
char>(sec)), [&] { return toString(this); });
  if (!data.empty() && data.back() != '\0') {
    error(toString(this) +
          ": corrupted dependent libraries section (unterminated string): " +
          name);
    return &InputSection::discarded;
  }
  for (const char *d = data.begin(), *e = data.end(); d < e;) {
    StringRef s(d);
    addDependentLibrary(s, this);
    d += s.size() + 1;
  }
  return &InputSection::discarded;
}
case SHT_RELA:
case SHT_REL: {
  // Find a relocation target section and associate this section with that.
  // Target may have been discarded if it is in a different section group
  // and the group is discarded, even though it's a violation of the
  // spec. We handle that situation gracefully by discarding dangling
  // relocation sections.
  InputSectionBase *target = getRelocTarget(sec);
  if (!target)
    return nullptr;

  // ELF spec allows mergeable sections with relocations, but they are
  // rare, and it is in practice hard to merge such sections by contents,
  // because applying relocations at end of linking changes section
  // contents. So, we simply handle such sections as non-mergeable ones.
  // Degrading like this is acceptable because section merging is optional.
  if (auto *ms = dyn_cast<MergeInputSection>(target)) {
    target = toRegularSection(ms);
    this->sections[sec.sh_info] = target;
  }

  if (target->firstRelocation)
    fatal(toString(this) +
          ": multiple relocation sections to one section are not supported");

  if (sec.sh_type == SHT_RELA) {
    ArrayRef<Elf_Rela> rels = CHECK(getObj().relas(sec), this)check2((getObj().relas(sec)), [&] { return toString(this)
; });
    target->firstRelocation = rels.begin();
    target->numRelocations = rels.size();
    target->areRelocsRela = true;
  } else {
    ArrayRef<Elf_Rel> rels = CHECK(getObj().rels(sec), this)check2((getObj().rels(sec)), [&] { return toString(this);
 });
    target->firstRelocation = rels.begin();
    target->numRelocations = rels.size();
    target->areRelocsRela = false;
  }
  assert(isUInt<31>(target->numRelocations))((void)0);

  // Relocation sections are usually removed from the output, so return
  // `nullptr` for the normal case. However, if -r or --emit-relocs is
  // specified, we need to copy them to the output. (Some post link analysis
  // tools specify --emit-relocs to obtain the information.)
  if (!config->relocatable && !config->emitRelocs)
    return nullptr;
  InputSection *relocSec = make<InputSection>(*this, sec, name);
  // If the relocated section is discarded (due to /DISCARD/ or
  // --gc-sections), the relocation section should be discarded as well.
  target->dependentSections.push_back(relocSec);
  return relocSec;
}
}

// The GNU linker uses .note.GNU-stack section as a marker indicating
// that the code in the object file does not expect that the stack is
// executable (in terms of NX bit). If all input files have the marker,
// the GNU linker adds a PT_GNU_STACK segment to tells the loader to
// make the stack non-executable. Most object files have this section as
// of 2017.
//
// But making the stack non-executable is a norm today for security
// reasons. Failure to do so may result in a serious security issue.
// Therefore, we make LLD always add PT_GNU_STACK unless it is
// explicitly told to do otherwise (by -z execstack). Because the stack
// executable-ness is controlled solely by command line options,
// .note.GNU-stack sections are simply ignored.
if (name == ".note.GNU-stack")
  return &InputSection::discarded;

// Object files that use processor features such as Intel Control-Flow
// Enforcement (CET) or AArch64 Branch Target Identification BTI, use a
// .note.gnu.property section containing a bitfield of feature bits like the
// GNU_PROPERTY_X86_FEATURE_1_IBT flag. Read a bitmap containing the flag.
//
// Since we merge bitmaps from multiple object files to create a new
// .note.gnu.property containing a single AND'ed bitmap, we discard an input
// file's .note.gnu.property section.
if (name == ".note.gnu.property") {
  this->andFeatures = readAndFeatures<ELFT>(InputSection(*this, sec, name));
  return &InputSection::discarded;
}

// Split stacks is a feature to support a discontiguous stack,
// commonly used in the programming language Go. For the details,
// see https://gcc.gnu.org/wiki/SplitStacks. An object file compiled
// for split stack will include a .note.GNU-split-stack section.
if (name == ".note.GNU-split-stack") {
  if (config->relocatable) {
    error("cannot mix split-stack and non-split-stack in a relocatable link");
    return &InputSection::discarded;
  }
  this->splitStack = true;
  return &InputSection::discarded;
}

// An object file cmpiled for split stack, but where some of the
// functions were compiled with the no_split_stack_attribute will
// include a .note.GNU-no-split-stack section.
if (name == ".note.GNU-no-split-stack") {
  this->someNoSplitStack = true;
  return &InputSection::discarded;
}

// The linkonce feature is a sort of proto-comdat. Some glibc i386 object
// files contain definitions of symbol "__x86.get_pc_thunk.bx" in linkonce
// sections. Drop those sections to avoid duplicate symbol errors.
// FIXME: This is glibc PR20543, we should remove this hack once that has been
// fixed for a while.
if (name == ".gnu.linkonce.t.__x86.get_pc_thunk.bx" ||
    name == ".gnu.linkonce.t.__i686.get_pc_thunk.bx")
  return &InputSection::discarded;

// If we are creating a new .build-id section, strip existing .build-id
// sections so that the output won't have more than one .build-id.
// This is not usually a problem because input object files normally don't
// have .build-id sections, but you can create such files by
// "ld.{bfd,gold,lld} -r --build-id", and we want to guard against it.
if (name == ".note.gnu.build-id" && config->buildId != BuildIdKind::None)
  return &InputSection::discarded;

// The linker merges EH (exception handling) frames and creates a
// .eh_frame_hdr section for runtime. So we handle them with a special
// class. For relocatable outputs, they are just passed through.
if (name == ".eh_frame" && !config->relocatable)
  return make<EhInputSection>(*this, sec, name);

if (shouldMerge(sec, name))
  return make<MergeInputSection>(*this, sec, name);
return make<InputSection>(*this, sec, name);
1073}

1075template <class ELFT>
1076StringRef ObjFile<ELFT>::getSectionName(const Elf_Shdr &sec) {
return CHECK(getObj().getSectionName(sec, sectionStringTable), this)check2((getObj().getSectionName(sec, sectionStringTable)), [&
] { return toString(this); });
1078}

1080// Initialize this->Symbols. this->Symbols is a parallel array as
1081// its corresponding ELF symbol table.
1082template <class ELFT> void ObjFile<ELFT>::initializeSymbols() {
ArrayRef<Elf_Sym> eSyms = this->getELFSyms<ELFT>();
this->symbols.resize(eSyms.size());

// Fill in InputFile::symbols. Some entries have been initialized
// because of LazyObjFile.
for (size_t i = 0, end = eSyms.size(); i != end; ++i) {
  if (this->symbols[i])
    continue;
  const Elf_Sym &eSym = eSyms[i];
  uint32_t secIdx = getSectionIndex(eSym);
  if (secIdx >= this->sections.size())
    fatal(toString(this) + ": invalid section index: " + Twine(secIdx));
  if (eSym.getBinding() != STB_LOCAL) {
    if (i < firstGlobal)
      error(toString(this) + ": non-local symbol (" + Twine(i) +
            ") found at index < .symtab's sh_info (" + Twine(firstGlobal) +
            ")");
    this->symbols[i] =
        symtab->insert(CHECK(eSyms[i].getName(this->stringTable), this)check2((eSyms[i].getName(this->stringTable)), [&] { return
 toString(this); }));
    continue;
  }

  // Handle local symbols. Local symbols are not added to the symbol
  // table because they are not visible from other object files. We
  // allocate symbol instances and add their pointers to symbols.
  if (i >= firstGlobal)
    errorOrWarn(toString(this) + ": STB_LOCAL symbol (" + Twine(i) +
                ") found at index >= .symtab's sh_info (" +
                Twine(firstGlobal) + ")");

  InputSectionBase *sec = this->sections[secIdx];
  uint8_t type = eSym.getType();
  if (type == STT_FILE)
    sourceFile = CHECK(eSym.getName(this->stringTable), this)check2((eSym.getName(this->stringTable)), [&] { return
 toString(this); });
  if (this->stringTable.size() <= eSym.st_name)
    fatal(toString(this) + ": invalid symbol name offset");
  StringRefZ name = this->stringTable.data() + eSym.st_name;

  if (eSym.st_shndx == SHN_UNDEF)
    this->symbols[i] =
        make<Undefined>(this, name, STB_LOCAL, eSym.st_other, type);
  else if (sec == &InputSection::discarded)
    this->symbols[i] =
        make<Undefined>(this, name, STB_LOCAL, eSym.st_other, type,
                        /*discardedSecIdx=*/secIdx);
  else
    this->symbols[i] = make<Defined>(this, name, STB_LOCAL, eSym.st_other,
                                     type, eSym.st_value, eSym.st_size, sec);
}

// Symbol resolution of non-local symbols.
SmallVector<unsigned, 32> undefineds;
for (size_t i = firstGlobal, end = eSyms.size(); i != end; ++i) {
  const Elf_Sym &eSym = eSyms[i];
  uint8_t binding = eSym.getBinding();
  if (binding == STB_LOCAL)
    continue; // Errored above.

  uint32_t secIdx = getSectionIndex(eSym);
  InputSectionBase *sec = this->sections[secIdx];
  uint8_t stOther = eSym.st_other;
  uint8_t type = eSym.getType();
  uint64_t value = eSym.st_value;
  uint64_t size = eSym.st_size;
  StringRefZ name = this->stringTable.data() + eSym.st_name;

  // Handle global undefined symbols.
  if (eSym.st_shndx == SHN_UNDEF) {
    undefineds.push_back(i);
    continue;
  }

  // Handle global common symbols.
  if (eSym.st_shndx == SHN_COMMON) {
    if (value == 0 || value >= UINT32_MAX0xffffffffU)
      fatal(toString(this) + ": common symbol '" + StringRef(name.data) +
            "' has invalid alignment: " + Twine(value));
    this->symbols[i]->resolve(
        CommonSymbol{this, name, binding, stOther, type, value, size});
    continue;
  }

  // If a defined symbol is in a discarded section, handle it as if it
  // were an undefined symbol. Such symbol doesn't comply with the
  // standard, but in practice, a .eh_frame often directly refer
  // COMDAT member sections, and if a comdat group is discarded, some
  // defined symbol in a .eh_frame becomes dangling symbols.
  if (sec == &InputSection::discarded) {
    Undefined und{this, name, binding, stOther, type, secIdx};
    Symbol *sym = this->symbols[i];
    // !ArchiveFile::parsed or LazyObjFile::fetched means that the file
    // containing this object has not finished processing, i.e. this symbol is
    // a result of a lazy symbol fetch. We should demote the lazy symbol to an
    // Undefined so that any relocations outside of the group to it will
    // trigger a discarded section error.
    if ((sym->symbolKind == Symbol::LazyArchiveKind &&
         !cast<ArchiveFile>(sym->file)->parsed) ||
        (sym->symbolKind == Symbol::LazyObjectKind &&
         cast<LazyObjFile>(sym->file)->fetched))
      sym->replace(und);
    else
      sym->resolve(und);
    continue;
  }

  // Handle global defined symbols.
  if (binding == STB_GLOBAL || binding == STB_WEAK ||
      binding == STB_GNU_UNIQUE) {
    this->symbols[i]->resolve(
        Defined{this, name, binding, stOther, type, value, size, sec});
    continue;
  }

  fatal(toString(this) + ": unexpected binding: " + Twine((int)binding));
}

// Undefined symbols (excluding those defined relative to non-prevailing
// sections) can trigger recursive fetch. Process defined symbols first so
// that the relative order between a defined symbol and an undefined symbol
// does not change the symbol resolution behavior. In addition, a set of
// interconnected symbols will all be resolved to the same file, instead of
// being resolved to different files.
for (unsigned i : undefineds) {
  const Elf_Sym &eSym = eSyms[i];
  StringRefZ name = this->stringTable.data() + eSym.st_name;
  this->symbols[i]->resolve(Undefined{this, name, eSym.getBinding(),
                                      eSym.st_other, eSym.getType()});
  this->symbols[i]->referenced = true;
}
1212}

1214ArchiveFile::ArchiveFile(std::unique_ptr<Archive> &&file)
  : InputFile(ArchiveKind, file->getMemoryBufferRef()),
    file(std::move(file)) {}

1218void ArchiveFile::parse() {
for (const Archive::Symbol &sym : file->symbols())
  symtab->addSymbol(LazyArchive{*this, sym});

// Inform a future invocation of ObjFile<ELFT>::initializeSymbols() that this
// archive has been processed.
parsed = true;
1225}

1227// Returns a buffer pointing to a member file containing a given symbol.
1228void ArchiveFile::fetch(const Archive::Symbol &sym) {
Archive::Child c =
    CHECK(sym.getMember(), toString(this) +check2((sym.getMember()), [&] { return toString(toString(
this) + ": could not get the member for symbol " + toELFString
(sym)); })
                               ": could not get the member for symbol " +check2((sym.getMember()), [&] { return toString(toString(
this) + ": could not get the member for symbol " + toELFString
(sym)); })
                               toELFString(sym))check2((sym.getMember()), [&] { return toString(toString(
this) + ": could not get the member for symbol " + toELFString
(sym)); });

if (!seen.insert(c.getChildOffset()).second)
  return;

MemoryBufferRef mb =
    CHECK(c.getMemoryBufferRef(),check2((c.getMemoryBufferRef()), [&] { return toString(toString
(this) + ": could not get the buffer for the member defining symbol "
 + toELFString(sym)); })
          toString(this) +check2((c.getMemoryBufferRef()), [&] { return toString(toString
(this) + ": could not get the buffer for the member defining symbol "
 + toELFString(sym)); })
              ": could not get the buffer for the member defining symbol " +check2((c.getMemoryBufferRef()), [&] { return toString(toString
(this) + ": could not get the buffer for the member defining symbol "
 + toELFString(sym)); })
              toELFString(sym))check2((c.getMemoryBufferRef()), [&] { return toString(toString
(this) + ": could not get the buffer for the member defining symbol "
 + toELFString(sym)); });

if (tar && c.getParent()->isThin())
  tar->append(relativeToRoot(CHECK(c.getFullName(), this)check2((c.getFullName()), [&] { return toString(this); })), mb.getBuffer());

InputFile *file = createObjectFile(mb, getName(), c.getChildOffset());
file->groupId = groupId;
parseFile(file);
1249}

1251// The handling of tentative definitions (COMMON symbols) in archives is murky.
1252// A tentative definition will be promoted to a global definition if there are
1253// no non-tentative definitions to dominate it. When we hold a tentative
1254// definition to a symbol and are inspecting archive members for inclusion
1255// there are 2 ways we can proceed:
1256//
1257// 1) Consider the tentative definition a 'real' definition (ie promotion from
1258//    tentative to real definition has already happened) and not inspect
1259//    archive members for Global/Weak definitions to replace the tentative
1260//    definition. An archive member would only be included if it satisfies some
1261//    other undefined symbol. This is the behavior Gold uses.
1262//
1263// 2) Consider the tentative definition as still undefined (ie the promotion to
1264//    a real definition happens only after all symbol resolution is done).
1265//    The linker searches archive members for STB_GLOBAL definitions to
1266//    replace the tentative definition with. This is the behavior used by
1267//    GNU ld.
1268//
1269//  The second behavior is inherited from SysVR4, which based it on the FORTRAN
1270//  COMMON BLOCK model. This behavior is needed for proper initialization in old
1271//  (pre F90) FORTRAN code that is packaged into an archive.
1272//
1273//  The following functions search archive members for definitions to replace
1274//  tentative definitions (implementing behavior 2).
1275static bool isBitcodeNonCommonDef(MemoryBufferRef mb, StringRef symName,
                                StringRef archiveName) {
IRSymtabFile symtabFile = check(readIRSymtab(mb));
for (const irsymtab::Reader::SymbolRef &sym :
     symtabFile.TheReader.symbols()) {
3
←
Calling 'Reader::symbols'→
  if (sym.isGlobal() && sym.getName() == symName)
    return !sym.isUndefined() && !sym.isWeak() && !sym.isCommon();
}
return false;
1284}

1286template <class ELFT>
1287static bool isNonCommonDef(MemoryBufferRef mb, StringRef symName,
                         StringRef archiveName) {
ObjFile<ELFT> *obj = make<ObjFile<ELFT>>(mb, archiveName);
StringRef stringtable = obj->getStringTable();

for (auto sym : obj->template getGlobalELFSyms<ELFT>()) {
  Expected<StringRef> name = sym.getName(stringtable);
  if (name && name.get() == symName)
    return sym.isDefined() && sym.getBinding() == STB_GLOBAL &&
           !sym.isCommon();
}
return false;
1299}

1301static bool isNonCommonDef(MemoryBufferRef mb, StringRef symName,
                         StringRef archiveName) {
switch (getELFKind(mb, archiveName)) {
case ELF32LEKind:
  return isNonCommonDef<ELF32LE>(mb, symName, archiveName);
case ELF32BEKind:
  return isNonCommonDef<ELF32BE>(mb, symName, archiveName);
case ELF64LEKind:
  return isNonCommonDef<ELF64LE>(mb, symName, archiveName);
case ELF64BEKind:
  return isNonCommonDef<ELF64BE>(mb, symName, archiveName);
default:
  llvm_unreachable("getELFKind")__builtin_unreachable();
}
1315}

1317bool ArchiveFile::shouldFetchForCommon(const Archive::Symbol &sym) {
Archive::Child c =
    CHECK(sym.getMember(), toString(this) +check2((sym.getMember()), [&] { return toString(toString(
this) + ": could not get the member for symbol " + toELFString
(sym)); })
                               ": could not get the member for symbol " +check2((sym.getMember()), [&] { return toString(toString(
this) + ": could not get the member for symbol " + toELFString
(sym)); })
                               toELFString(sym))check2((sym.getMember()), [&] { return toString(toString(
this) + ": could not get the member for symbol " + toELFString
(sym)); });
MemoryBufferRef mb =
    CHECK(c.getMemoryBufferRef(),check2((c.getMemoryBufferRef()), [&] { return toString(toString
(this) + ": could not get the buffer for the member defining symbol "
 + toELFString(sym)); })
          toString(this) +check2((c.getMemoryBufferRef()), [&] { return toString(toString
(this) + ": could not get the buffer for the member defining symbol "
 + toELFString(sym)); })
              ": could not get the buffer for the member defining symbol " +check2((c.getMemoryBufferRef()), [&] { return toString(toString
(this) + ": could not get the buffer for the member defining symbol "
 + toELFString(sym)); })
              toELFString(sym))check2((c.getMemoryBufferRef()), [&] { return toString(toString
(this) + ": could not get the buffer for the member defining symbol "
 + toELFString(sym)); });

if (isBitcode(mb))
  return isBitcodeNonCommonDef(mb, sym.getName(), getName());

return isNonCommonDef(mb, sym.getName(), getName());
1332}

1334size_t ArchiveFile::getMemberCount() const {
size_t count = 0;
Error err = Error::success();
for (const Archive::Child &c : file->children(err)) {
  (void)c;
  ++count;
}
// This function is used by --print-archive-stats=, where an error does not
// really matter.
consumeError(std::move(err));
return count;
1345}

1347unsigned SharedFile::vernauxNum;

1349// Parse the version definitions in the object file if present, and return a
1350// vector whose nth element contains a pointer to the Elf_Verdef for version
1351// identifier n. Version identifiers that are not definitions map to nullptr.
1352template <typename ELFT>
1353static std::vector<const void *> parseVerdefs(const uint8_t *base,
                                            const typename ELFT::Shdr *sec) {
if (!sec)
  return {};

// We cannot determine the largest verdef identifier without inspecting
// every Elf_Verdef, but both bfd and gold assign verdef identifiers
// sequentially starting from 1, so we predict that the largest identifier
// will be verdefCount.
unsigned verdefCount = sec->sh_info;
std::vector<const void *> verdefs(verdefCount + 1);

// Build the Verdefs array by following the chain of Elf_Verdef objects
// from the start of the .gnu.version_d section.
const uint8_t *verdef = base + sec->sh_offset;
for (unsigned i = 0; i != verdefCount; ++i) {
  auto *curVerdef = reinterpret_cast<const typename ELFT::Verdef *>(verdef);
  verdef += curVerdef->vd_next;
  unsigned verdefIndex = curVerdef->vd_ndx;
  verdefs.resize(verdefIndex + 1);
  verdefs[verdefIndex] = curVerdef;
}
return verdefs;
1376}

1378// Parse SHT_GNU_verneed to properly set the name of a versioned undefined
1379// symbol. We detect fatal issues which would cause vulnerabilities, but do not
1380// implement sophisticated error checking like in llvm-readobj because the value
1381// of such diagnostics is low.
1382template <typename ELFT>
1383std::vector<uint32_t> SharedFile::parseVerneed(const ELFFile<ELFT> &obj,
                                             const typename ELFT::Shdr *sec) {
if (!sec)
  return {};
std::vector<uint32_t> verneeds;
ArrayRef<uint8_t> data = CHECK(obj.getSectionContents(*sec), this)check2((obj.getSectionContents(*sec)), [&] { return toString
(this); });
const uint8_t *verneedBuf = data.begin();
for (unsigned i = 0; i != sec->sh_info; ++i) {
  if (verneedBuf + sizeof(typename ELFT::Verneed) > data.end())
    fatal(toString(this) + " has an invalid Verneed");
  auto *vn = reinterpret_cast<const typename ELFT::Verneed *>(verneedBuf);
  const uint8_t *vernauxBuf = verneedBuf + vn->vn_aux;
  for (unsigned j = 0; j != vn->vn_cnt; ++j) {
    if (vernauxBuf + sizeof(typename ELFT::Vernaux) > data.end())
      fatal(toString(this) + " has an invalid Vernaux");
    auto *aux = reinterpret_cast<const typename ELFT::Vernaux *>(vernauxBuf);
    if (aux->vna_name >= this->stringTable.size())
      fatal(toString(this) + " has a Vernaux with an invalid vna_name");
    uint16_t version = aux->vna_other & VERSYM_VERSION;
    if (version >= verneeds.size())
      verneeds.resize(version + 1);
    verneeds[version] = aux->vna_name;
    vernauxBuf += aux->vna_next;
  }
  verneedBuf += vn->vn_next;
}
return verneeds;
1410}

1412// We do not usually care about alignments of data in shared object
1413// files because the loader takes care of it. However, if we promote a
1414// DSO symbol to point to .bss due to copy relocation, we need to keep
1415// the original alignment requirements. We infer it in this function.
1416template <typename ELFT>
1417static uint64_t getAlignment(ArrayRef<typename ELFT::Shdr> sections,
                           const typename ELFT::Sym &sym) {
uint64_t ret = UINT64_MAX0xffffffffffffffffULL;
if (sym.st_value)
  ret = 1ULL << countTrailingZeros((uint64_t)sym.st_value);
if (0 < sym.st_shndx && sym.st_shndx < sections.size())
  ret = std::min<uint64_t>(ret, sections[sym.st_shndx].sh_addralign);
return (ret > UINT32_MAX0xffffffffU) ? 0 : ret;
1425}

1427// Fully parse the shared object file.
1428//
1429// This function parses symbol versions. If a DSO has version information,
1430// the file has a ".gnu.version_d" section which contains symbol version
1431// definitions. Each symbol is associated to one version through a table in
1432// ".gnu.version" section. That table is a parallel array for the symbol
1433// table, and each table entry contains an index in ".gnu.version_d".
1434//
1435// The special index 0 is reserved for VERF_NDX_LOCAL and 1 is for
1436// VER_NDX_GLOBAL. There's no table entry for these special versions in
1437// ".gnu.version_d".
1438//
1439// The file format for symbol versioning is perhaps a bit more complicated
1440// than necessary, but you can easily understand the code if you wrap your
1441// head around the data structure described above.
1442template <class ELFT> void SharedFile::parse() {
using Elf_Dyn = typename ELFT::Dyn;
using Elf_Shdr = typename ELFT::Shdr;
using Elf_Sym = typename ELFT::Sym;
using Elf_Verdef = typename ELFT::Verdef;
using Elf_Versym = typename ELFT::Versym;

ArrayRef<Elf_Dyn> dynamicTags;
const ELFFile<ELFT> obj = this->getObj<ELFT>();
ArrayRef<Elf_Shdr> sections = CHECK(obj.sections(), this)check2((obj.sections()), [&] { return toString(this); });

const Elf_Shdr *versymSec = nullptr;
const Elf_Shdr *verdefSec = nullptr;
const Elf_Shdr *verneedSec = nullptr;

// Search for .dynsym, .dynamic, .symtab, .gnu.version and .gnu.version_d.
for (const Elf_Shdr &sec : sections) {
  switch (sec.sh_type) {
  default:
    continue;
  case SHT_DYNAMIC:
    dynamicTags =
        CHECK(obj.template getSectionContentsAsArray<Elf_Dyn>(sec), this)check2((obj.template getSectionContentsAsArray<Elf_Dyn>
(sec)), [&] { return toString(this); });
    break;
  case SHT_GNU_versym:
    versymSec = &sec;
    break;
  case SHT_GNU_verdef:
    verdefSec = &sec;
    break;
  case SHT_GNU_verneed:
    verneedSec = &sec;
    break;
  }
}

if (versymSec && numELFSyms == 0) {
  error("SHT_GNU_versym should be associated with symbol table");
  return;
}

// Search for a DT_SONAME tag to initialize this->soName.
for (const Elf_Dyn &dyn : dynamicTags) {
  if (dyn.d_tag == DT_NEEDED) {
    uint64_t val = dyn.getVal();
    if (val >= this->stringTable.size())
      fatal(toString(this) + ": invalid DT_NEEDED entry");
    dtNeeded.push_back(this->stringTable.data() + val);
  } else if (dyn.d_tag == DT_SONAME) {
    uint64_t val = dyn.getVal();
    if (val >= this->stringTable.size())
      fatal(toString(this) + ": invalid DT_SONAME entry");
    soName = this->stringTable.data() + val;
  }
}

// DSOs are uniquified not by filename but by soname.
DenseMap<StringRef, SharedFile *>::iterator it;
bool wasInserted;
std::tie(it, wasInserted) = symtab->soNames.try_emplace(soName, this);

// If a DSO appears more than once on the command line with and without
// --as-needed, --no-as-needed takes precedence over --as-needed because a
// user can add an extra DSO with --no-as-needed to force it to be added to
// the dependency list.
it->second->isNeeded |= isNeeded;
if (!wasInserted)
  return;

sharedFiles.push_back(this);

verdefs = parseVerdefs<ELFT>(obj.base(), verdefSec);
std::vector<uint32_t> verneeds = parseVerneed<ELFT>(obj, verneedSec);

// Parse ".gnu.version" section which is a parallel array for the symbol
// table. If a given file doesn't have a ".gnu.version" section, we use
// VER_NDX_GLOBAL.
size_t size = numELFSyms - firstGlobal;
std::vector<uint16_t> versyms(size, VER_NDX_GLOBAL);
if (versymSec) {
  ArrayRef<Elf_Versym> versym =
      CHECK(obj.template getSectionContentsAsArray<Elf_Versym>(*versymSec),check2((obj.template getSectionContentsAsArray<Elf_Versym>
(*versymSec)), [&] { return toString(this); })
            this)check2((obj.template getSectionContentsAsArray<Elf_Versym>
(*versymSec)), [&] { return toString(this); })
          .slice(firstGlobal);
  for (size_t i = 0; i < size; ++i)
    versyms[i] = versym[i].vs_index;
}

// System libraries can have a lot of symbols with versions. Using a
// fixed buffer for computing the versions name (foo@ver) can save a
// lot of allocations.
SmallString<0> versionedNameBuffer;

// Add symbols to the symbol table.
ArrayRef<Elf_Sym> syms = this->getGlobalELFSyms<ELFT>();
for (size_t i = 0; i < syms.size(); ++i) {
  const Elf_Sym &sym = syms[i];

  // ELF spec requires that all local symbols precede weak or global
  // symbols in each symbol table, and the index of first non-local symbol
  // is stored to sh_info. If a local symbol appears after some non-local
  // symbol, that's a violation of the spec.
  StringRef name = CHECK(sym.getName(this->stringTable), this)check2((sym.getName(this->stringTable)), [&] { return toString
(this); });
  if (sym.getBinding() == STB_LOCAL) {
    warn("found local symbol '" + name +
         "' in global part of symbol table in file " + toString(this));
    continue;
  }

  uint16_t idx = versyms[i] & ~VERSYM_HIDDEN;
  if (sym.isUndefined()) {
    // For unversioned undefined symbols, VER_NDX_GLOBAL makes more sense but
    // as of binutils 2.34, GNU ld produces VER_NDX_LOCAL.
    if (idx != VER_NDX_LOCAL && idx != VER_NDX_GLOBAL) {
      if (idx >= verneeds.size()) {
        error("corrupt input file: version need index " + Twine(idx) +
              " for symbol " + name + " is out of bounds\n>>> defined in " +
              toString(this));
        continue;
      }
      StringRef verName = this->stringTable.data() + verneeds[idx];
      versionedNameBuffer.clear();
      name =
          saver.save((name + "@" + verName).toStringRef(versionedNameBuffer));
    }
    Symbol *s = symtab->addSymbol(
        Undefined{this, name, sym.getBinding(), sym.st_other, sym.getType()});
    s->exportDynamic = true;
    if (s->isUndefined() && sym.getBinding() != STB_WEAK &&
        config->unresolvedSymbolsInShlib != UnresolvedPolicy::Ignore)
      requiredSymbols.push_back(s);
    continue;
  }

  // MIPS BFD linker puts _gp_disp symbol into DSO files and incorrectly
  // assigns VER_NDX_LOCAL to this section global symbol. Here is a
  // workaround for this bug.
  if (config->emachine == EM_MIPS && idx == VER_NDX_LOCAL &&
      name == "_gp_disp")
    continue;

  uint32_t alignment = getAlignment<ELFT>(sections, sym);
  if (!(versyms[i] & VERSYM_HIDDEN)) {
    symtab->addSymbol(SharedSymbol{*this, name, sym.getBinding(),
                                   sym.st_other, sym.getType(), sym.st_value,
                                   sym.st_size, alignment, idx});
  }

  // Also add the symbol with the versioned name to handle undefined symbols
  // with explicit versions.
  if (idx == VER_NDX_GLOBAL)
    continue;

  if (idx >= verdefs.size() || idx == VER_NDX_LOCAL) {
    error("corrupt input file: version definition index " + Twine(idx) +
          " for symbol " + name + " is out of bounds\n>>> defined in " +
          toString(this));
    continue;
  }

  StringRef verName =
      this->stringTable.data() +
      reinterpret_cast<const Elf_Verdef *>(verdefs[idx])->getAux()->vda_name;
  versionedNameBuffer.clear();
  name = (name + "@" + verName).toStringRef(versionedNameBuffer);
  symtab->addSymbol(SharedSymbol{*this, saver.save(name), sym.getBinding(),
                                 sym.st_other, sym.getType(), sym.st_value,
                                 sym.st_size, alignment, idx});
}
1611}

1613static ELFKind getBitcodeELFKind(const Triple &t) {
if (t.isLittleEndian())
  return t.isArch64Bit() ? ELF64LEKind : ELF32LEKind;
return t.isArch64Bit() ? ELF64BEKind : ELF32BEKind;
1617}

1619static uint16_t getBitcodeMachineKind(StringRef path, const Triple &t) {
switch (t.getArch()) {
case Triple::aarch64:
case Triple::aarch64_be:
  return EM_AARCH64;
case Triple::amdgcn:
case Triple::r600:
  return EM_AMDGPU;
case Triple::arm:
case Triple::thumb:
  return EM_ARM;
case Triple::avr:
  return EM_AVR;
case Triple::mips:
case Triple::mipsel:
case Triple::mips64:
case Triple::mips64el:
  return EM_MIPS;
case Triple::msp430:
  return EM_MSP430;
case Triple::ppc:
case Triple::ppcle:
  return EM_PPC;
case Triple::ppc64:
case Triple::ppc64le:
  return EM_PPC64;
case Triple::riscv32:
case Triple::riscv64:
  return EM_RISCV;
case Triple::x86:
  return t.isOSIAMCU() ? EM_IAMCU : EM_386;
case Triple::x86_64:
  return EM_X86_64;
default:
  error(path + ": could not infer e_machine from bitcode target triple " +
        t.str());
  return EM_NONE;
}
1657}

1659static uint8_t getOsAbi(const Triple &t) {
switch (t.getOS()) {
case Triple::AMDHSA:
  return ELF::ELFOSABI_AMDGPU_HSA;
case Triple::AMDPAL:
  return ELF::ELFOSABI_AMDGPU_PAL;
case Triple::Mesa3D:
  return ELF::ELFOSABI_AMDGPU_MESA3D;
default:
  return ELF::ELFOSABI_NONE;
}
1670}

1672BitcodeFile::BitcodeFile(MemoryBufferRef mb, StringRef archiveName,
                       uint64_t offsetInArchive)
  : InputFile(BitcodeKind, mb) {
this->archiveName = std::string(archiveName);

std::string path = mb.getBufferIdentifier().str();
if (config->thinLTOIndexOnly)
  path = replaceThinLTOSuffix(mb.getBufferIdentifier());

// ThinLTO assumes that all MemoryBufferRefs given to it have a unique
// name. If two archives define two members with the same name, this
// causes a collision which result in only one of the objects being taken
// into consideration at LTO time (which very likely causes undefined
// symbols later in the link stage). So we append file offset to make
// filename unique.
StringRef name =
    archiveName.empty()
        ? saver.save(path)
        : saver.save(archiveName + "(" + path::filename(path) + " at " +
                     utostr(offsetInArchive) + ")");
MemoryBufferRef mbref(mb.getBuffer(), name);

obj = CHECK(lto::InputFile::create(mbref), this)check2((lto::InputFile::create(mbref)), [&] { return toString
(this); });

Triple t(obj->getTargetTriple());
ekind = getBitcodeELFKind(t);
emachine = getBitcodeMachineKind(mb.getBufferIdentifier(), t);
osabi = getOsAbi(t);
1700}

1702static uint8_t mapVisibility(GlobalValue::VisibilityTypes gvVisibility) {
switch (gvVisibility) {
case GlobalValue::DefaultVisibility:
  return STV_DEFAULT;
case GlobalValue::HiddenVisibility:
  return STV_HIDDEN;
case GlobalValue::ProtectedVisibility:
  return STV_PROTECTED;
}
llvm_unreachable("unknown visibility")__builtin_unreachable();
1712}

1714template <class ELFT>
1715static Symbol *createBitcodeSymbol(const std::vector<bool> &keptComdats,
                                 const lto::InputFile::Symbol &objSym,
                                 BitcodeFile &f) {
StringRef name = saver.save(objSym.getName());
uint8_t binding = objSym.isWeak() ? STB_WEAK : STB_GLOBAL;
uint8_t type = objSym.isTLS() ? STT_TLS : STT_NOTYPE;
uint8_t visibility = mapVisibility(objSym.getVisibility());
bool canOmitFromDynSym = objSym.canBeOmittedFromSymbolTable();

int c = objSym.getComdatIndex();
if (objSym.isUndefined() || (c != -1 && !keptComdats[c])) {
  Undefined newSym(&f, name, binding, visibility, type);
  if (canOmitFromDynSym)
    newSym.exportDynamic = false;
  Symbol *ret = symtab->addSymbol(newSym);
  ret->referenced = true;
  return ret;
}

if (objSym.isCommon())
  return symtab->addSymbol(
      CommonSymbol{&f, name, binding, visibility, STT_OBJECT,
                   objSym.getCommonAlignment(), objSym.getCommonSize()});

Defined newSym(&f, name, binding, visibility, type, 0, 0, nullptr);
if (canOmitFromDynSym)
  newSym.exportDynamic = false;
return symtab->addSymbol(newSym);
1743}

1745template <class ELFT> void BitcodeFile::parse() {
std::vector<bool> keptComdats;
for (std::pair<StringRef, Comdat::SelectionKind> s : obj->getComdatTable()) {
  keptComdats.push_back(
      s.second == Comdat::NoDeduplicate ||
      symtab->comdatGroups.try_emplace(CachedHashStringRef(s.first), this)
          .second);
}

for (const lto::InputFile::Symbol &objSym : obj->symbols())
  symbols.push_back(createBitcodeSymbol<ELFT>(keptComdats, objSym, *this));

for (auto l : obj->getDependentLibraries())
  addDependentLibrary(l, this);
1759}

1761void BinaryFile::parse() {
ArrayRef<uint8_t> data = arrayRefFromStringRef(mb.getBuffer());
auto *section = make<InputSection>(this, SHF_ALLOC | SHF_WRITE, SHT_PROGBITS,
                                   8, data, ".data");
sections.push_back(section);

// For each input file foo that is embedded to a result as a binary
// blob, we define _binary_foo_{start,end,size} symbols, so that
// user programs can access blobs by name. Non-alphanumeric
// characters in a filename are replaced with underscore.
std::string s = "_binary_" + mb.getBufferIdentifier().str();
for (size_t i = 0; i < s.size(); ++i)
  if (!isAlnum(s[i]))
    s[i] = '_';

symtab->addSymbol(Defined{nullptr, saver.save(s + "_start"), STB_GLOBAL,
                          STV_DEFAULT, STT_OBJECT, 0, 0, section});
symtab->addSymbol(Defined{nullptr, saver.save(s + "_end"), STB_GLOBAL,
                          STV_DEFAULT, STT_OBJECT, data.size(), 0, section});
symtab->addSymbol(Defined{nullptr, saver.save(s + "_size"), STB_GLOBAL,
                          STV_DEFAULT, STT_OBJECT, data.size(), 0, nullptr});
1782}

1784InputFile *elf::createObjectFile(MemoryBufferRef mb, StringRef archiveName,
                               uint64_t offsetInArchive) {
if (isBitcode(mb))
  return make<BitcodeFile>(mb, archiveName, offsetInArchive);

switch (getELFKind(mb, archiveName)) {
case ELF32LEKind:
  return make<ObjFile<ELF32LE>>(mb, archiveName);
case ELF32BEKind:
  return make<ObjFile<ELF32BE>>(mb, archiveName);
case ELF64LEKind:
  return make<ObjFile<ELF64LE>>(mb, archiveName);
case ELF64BEKind:
  return make<ObjFile<ELF64BE>>(mb, archiveName);
default:
  llvm_unreachable("getELFKind")__builtin_unreachable();
}
1801}

1803void LazyObjFile::fetch() {
if (fetched)
  return;
fetched = true;

InputFile *file = createObjectFile(mb, archiveName, offsetInArchive);
file->groupId = groupId;

// Copy symbol vector so that the new InputFile doesn't have to
// insert the same defined symbols to the symbol table again.
file->symbols = std::move(symbols);

parseFile(file);
1816}

1818template <class ELFT> void LazyObjFile::parse() {
using Elf_Sym = typename ELFT::Sym;

// A lazy object file wraps either a bitcode file or an ELF file.
if (isBitcode(this->mb)) {
  std::unique_ptr<lto::InputFile> obj =
      CHECK(lto::InputFile::create(this->mb), this)check2((lto::InputFile::create(this->mb)), [&] { return
 toString(this); });
  for (const lto::InputFile::Symbol &sym : obj->symbols()) {
    if (sym.isUndefined())
      continue;
    symtab->addSymbol(LazyObject{*this, saver.save(sym.getName())});
  }
  return;
}

if (getELFKind(this->mb, archiveName) != config->ekind) {
  error("incompatible file: " + this->mb.getBufferIdentifier());
  return;
}

// Find a symbol table.
ELFFile<ELFT> obj = check(ELFFile<ELFT>::create(mb.getBuffer()));
ArrayRef<typename ELFT::Shdr> sections = CHECK(obj.sections(), this)check2((obj.sections()), [&] { return toString(this); });

for (const typename ELFT::Shdr &sec : sections) {
  if (sec.sh_type != SHT_SYMTAB)
    continue;

  // A symbol table is found.
  ArrayRef<Elf_Sym> eSyms = CHECK(obj.symbols(&sec), this)check2((obj.symbols(&sec)), [&] { return toString(this
); });
  uint32_t firstGlobal = sec.sh_info;
  StringRef strtab = CHECK(obj.getStringTableForSymtab(sec, sections), this)check2((obj.getStringTableForSymtab(sec, sections)), [&] {
 return toString(this); });
  this->symbols.resize(eSyms.size());

  // Get existing symbols or insert placeholder symbols.
  for (size_t i = firstGlobal, end = eSyms.size(); i != end; ++i)
    if (eSyms[i].st_shndx != SHN_UNDEF)
      this->symbols[i] = symtab->insert(CHECK(eSyms[i].getName(strtab), this)check2((eSyms[i].getName(strtab)), [&] { return toString(
this); }));

  // Replace existing symbols with LazyObject symbols.
  //
  // resolve() may trigger this->fetch() if an existing symbol is an
  // undefined symbol. If that happens, this LazyObjFile has served
  // its purpose, and we can exit from the loop early.
  for (Symbol *sym : this->symbols) {
    if (!sym)
      continue;
    sym->resolve(LazyObject{*this, sym->getName()});

    // If fetched, stop iterating because this->symbols has been transferred
    // to the instantiated ObjFile.
    if (fetched)
      return;
  }
  return;
}
1874}

1876bool LazyObjFile::shouldFetchForCommon(const StringRef &name) {
if (isBitcode(mb))
1
Taking true branch→
  return isBitcodeNonCommonDef(mb, name, archiveName);
2
←
Calling 'isBitcodeNonCommonDef'→

return isNonCommonDef(mb, name, archiveName);
1881}

1883std::string elf::replaceThinLTOSuffix(StringRef path) {
StringRef suffix = config->thinLTOObjectSuffixReplace.first;
StringRef repl = config->thinLTOObjectSuffixReplace.second;

if (path.consume_back(suffix))
  return (path + repl).str();
return std::string(path);
1890}

1892template void BitcodeFile::parse<ELF32LE>();
1893template void BitcodeFile::parse<ELF32BE>();
1894template void BitcodeFile::parse<ELF64LE>();
1895template void BitcodeFile::parse<ELF64BE>();

1897template void LazyObjFile::parse<ELF32LE>();
1898template void LazyObjFile::parse<ELF32BE>();
1899template void LazyObjFile::parse<ELF64LE>();
1900template void LazyObjFile::parse<ELF64BE>();

1902template class elf::ObjFile<ELF32LE>;
1903template class elf::ObjFile<ELF32BE>;
1904template class elf::ObjFile<ELF64LE>;
1905template class elf::ObjFile<ELF64BE>;

1907template void SharedFile::parse<ELF32LE>();
1908template void SharedFile::parse<ELF32BE>();
1909template void SharedFile::parse<ELF64LE>();
1910template void SharedFile::parse<ELF64BE>();

←

/usr/src/gnu/usr.bin/clang/liblldELF/../../../llvm/llvm/include/llvm/Object/IRSymtab.h

1//===- IRSymtab.h - data definitions for IR symbol tables -------*- C++ -*-===//
2//
3// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
4// See https://llvm.org/LICENSE.txt for license information.
5// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
6//
7//===----------------------------------------------------------------------===//
8//
9// This file contains data definitions and a reader and builder for a symbol
10// table for LLVM IR. Its purpose is to allow linkers and other consumers of
11// bitcode files to efficiently read the symbol table for symbol resolution
12// purposes without needing to construct a module in memory.
13//
14// As with most object files the symbol table has two parts: the symbol table
15// itself and a string table which is referenced by the symbol table.
16//
17// A symbol table corresponds to a single bitcode file, which may consist of
18// multiple modules, so symbol tables may likewise contain symbols for multiple
19// modules.
20//
21//===----------------------------------------------------------------------===//
22 
23#ifndef LLVM_OBJECT_IRSYMTAB_H
24#define LLVM_OBJECT_IRSYMTAB_H
25 
26#include "llvm/ADT/ArrayRef.h"
27#include "llvm/ADT/StringRef.h"
28#include "llvm/ADT/iterator_range.h"
29#include "llvm/IR/Comdat.h"
30#include "llvm/IR/GlobalValue.h"
31#include "llvm/Object/SymbolicFile.h"
32#include "llvm/Support/Allocator.h"
33#include "llvm/Support/Endian.h"
34#include "llvm/Support/Error.h"
35#include <cassert>
36#include <cstdint>
37#include <vector>
38 
39namespace llvm {
40 
41struct BitcodeFileContents;
42class StringTableBuilder;
43 
44namespace irsymtab {
45 
46namespace storage {
47 
48// The data structures in this namespace define the low-level serialization
49// format. Clients that just want to read a symbol table should use the
50// irsymtab::Reader class.
51 
52using Word = support::ulittle32_t;
53 
54/// A reference to a string in the string table.
55struct Str {
56  Word Offset, Size;
57 
58  StringRef get(StringRef Strtab) const {
59    return {Strtab.data() + Offset, Size};
60  }
61};
62 
63/// A reference to a range of objects in the symbol table.
64template <typename T> struct Range {
65  Word Offset, Size;
66 
67  ArrayRef<T> get(StringRef Symtab) const {
68    return {reinterpret_cast<const T *>(Symtab.data() + Offset), Size};
69  }
70};
71 
72/// Describes the range of a particular module's symbols within the symbol
73/// table.
74struct Module {
75  Word Begin, End;
76 
77  /// The index of the first Uncommon for this Module.
78  Word UncBegin;
79};
80 
81/// This is equivalent to an IR comdat.
82struct Comdat {
83  Str Name;
84 
85  // llvm::Comdat::SelectionKind
86  Word SelectionKind;
87};
88 
89/// Contains the information needed by linkers for symbol resolution, as well as
90/// by the LTO implementation itself.
91struct Symbol {
92  /// The mangled symbol name.
93  Str Name;
94 
95  /// The unmangled symbol name, or the empty string if this is not an IR
96  /// symbol.
97  Str IRName;
98 
99  /// The index into Header::Comdats, or -1 if not a comdat member.
100  Word ComdatIndex;
101 
102  Word Flags;
103  enum FlagBits {
104    FB_visibility, // 2 bits
105    FB_has_uncommon = FB_visibility + 2,
106    FB_undefined,
107    FB_weak,
108    FB_common,
109    FB_indirect,
110    FB_used,
111    FB_tls,
112    FB_may_omit,
113    FB_global,
114    FB_format_specific,
115    FB_unnamed_addr,
116    FB_executable,
117  };
118};
119 
120/// This data structure contains rarely used symbol fields and is optionally
121/// referenced by a Symbol.
122struct Uncommon {
123  Word CommonSize, CommonAlign;
124 
125  /// COFF-specific: the name of the symbol that a weak external resolves to
126  /// if not defined.
127  Str COFFWeakExternFallbackName;
128 
129  /// Specified section name, if any.
130  Str SectionName;
131};
132 
133 
134struct Header {
135  /// Version number of the symtab format. This number should be incremented
136  /// when the format changes, but it does not need to be incremented if a
137  /// change to LLVM would cause it to create a different symbol table.
138  Word Version;
139  enum { kCurrentVersion = 3 };
140 
141  /// The producer's version string (LLVM_VERSION_STRING " " LLVM_REVISION).
142  /// Consumers should rebuild the symbol table from IR if the producer's
143  /// version does not match the consumer's version due to potential differences
144  /// in symbol table format, symbol enumeration order and so on.
145  Str Producer;
146 
147  Range<Module> Modules;
148  Range<Comdat> Comdats;
149  Range<Symbol> Symbols;
150  Range<Uncommon> Uncommons;
151 
152  Str TargetTriple, SourceFileName;
153 
154  /// COFF-specific: linker directives.
155  Str COFFLinkerOpts;
156 
157  /// Dependent Library Specifiers
158  Range<Str> DependentLibraries;
159};
160 
161} // end namespace storage
162 
163/// Fills in Symtab and StrtabBuilder with a valid symbol and string table for
164/// Mods.
165Error build(ArrayRef<Module *> Mods, SmallVector<char, 0> &Symtab,
166            StringTableBuilder &StrtabBuilder, BumpPtrAllocator &Alloc);
167 
168/// This represents a symbol that has been read from a storage::Symbol and
169/// possibly a storage::Uncommon.
170struct Symbol {
171  // Copied from storage::Symbol.
172  StringRef Name, IRName;
173  int ComdatIndex;
174  uint32_t Flags;
175 
176  // Copied from storage::Uncommon.
177  uint32_t CommonSize, CommonAlign;
178  StringRef COFFWeakExternFallbackName;
179  StringRef SectionName;
180 
181  /// Returns the mangled symbol name.
182  StringRef getName() const { return Name; }
183 
184  /// Returns the unmangled symbol name, or the empty string if this is not an
185  /// IR symbol.
186  StringRef getIRName() const { return IRName; }
187 
188  /// Returns the index into the comdat table (see Reader::getComdatTable()), or
189  /// -1 if not a comdat member.
190  int getComdatIndex() const { return ComdatIndex; }
191 
192  using S = storage::Symbol;
193 
194  GlobalValue::VisibilityTypes getVisibility() const {
195    return GlobalValue::VisibilityTypes((Flags >> S::FB_visibility) & 3);
196  }
197 
198  bool isUndefined() const { return (Flags >> S::FB_undefined) & 1; }
199  bool isWeak() const { return (Flags >> S::FB_weak) & 1; }
200  bool isCommon() const { return (Flags >> S::FB_common) & 1; }
201  bool isIndirect() const { return (Flags >> S::FB_indirect) & 1; }
202  bool isUsed() const { return (Flags >> S::FB_used) & 1; }
203  bool isTLS() const { return (Flags >> S::FB_tls) & 1; }
204 
205  bool canBeOmittedFromSymbolTable() const {
206    return (Flags >> S::FB_may_omit) & 1;
207  }
208 
209  bool isGlobal() const { return (Flags >> S::FB_global) & 1; }
210  bool isFormatSpecific() const { return (Flags >> S::FB_format_specific) & 1; }
211  bool isUnnamedAddr() const { return (Flags >> S::FB_unnamed_addr) & 1; }
212  bool isExecutable() const { return (Flags >> S::FB_executable) & 1; }
213 
214  uint64_t getCommonSize() const {
215    assert(isCommon())((void)0);
216    return CommonSize;
217  }
218 
219  uint32_t getCommonAlignment() const {
220    assert(isCommon())((void)0);
221    return CommonAlign;
222  }
223 
224  /// COFF-specific: for weak externals, returns the name of the symbol that is
225  /// used as a fallback if the weak external remains undefined.
226  StringRef getCOFFWeakExternalFallback() const {
227    assert(isWeak() && isIndirect())((void)0);
228    return COFFWeakExternFallbackName;
229  }
230 
231  StringRef getSectionName() const { return SectionName; }
232};
233 
234/// This class can be used to read a Symtab and Strtab produced by
235/// irsymtab::build.
236class Reader {
237  StringRef Symtab, Strtab;
238 
239  ArrayRef<storage::Module> Modules;
240  ArrayRef<storage::Comdat> Comdats;
241  ArrayRef<storage::Symbol> Symbols;
242  ArrayRef<storage::Uncommon> Uncommons;
243  ArrayRef<storage::Str> DependentLibraries;
244 
245  StringRef str(storage::Str S) const { return S.get(Strtab); }
246 
247  template <typename T> ArrayRef<T> range(storage::Range<T> R) const {
248    return R.get(Symtab);
249  }
250 
251  const storage::Header &header() const {
252    return *reinterpret_cast<const storage::Header *>(Symtab.data());
253  }
254 
255public:
256  class SymbolRef;
257 
258  Reader() = default;
259  Reader(StringRef Symtab, StringRef Strtab) : Symtab(Symtab), Strtab(Strtab) {
260    Modules = range(header().Modules);
261    Comdats = range(header().Comdats);
262    Symbols = range(header().Symbols);
263    Uncommons = range(header().Uncommons);
264    DependentLibraries = range(header().DependentLibraries);
265  }
266 
267  using symbol_range = iterator_range<object::content_iterator<SymbolRef>>;
268 
269  /// Returns the symbol table for the entire bitcode file.
270  /// The symbols enumerated by this method are ephemeral, but they can be
271  /// copied into an irsymtab::Symbol object.
272  symbol_range symbols() const;
273 
274  size_t getNumModules() const { return Modules.size(); }
275 
276  /// Returns a slice of the symbol table for the I'th module in the file.
277  /// The symbols enumerated by this method are ephemeral, but they can be
278  /// copied into an irsymtab::Symbol object.
279  symbol_range module_symbols(unsigned I) const;
280 
281  StringRef getTargetTriple() const { return str(header().TargetTriple); }
282 
283  /// Returns the source file path specified at compile time.
284  StringRef getSourceFileName() const { return str(header().SourceFileName); }
285 
286  /// Returns a table with all the comdats used by this file.
287  std::vector<std::pair<StringRef, llvm::Comdat::SelectionKind>>
288  getComdatTable() const {
289    std::vector<std::pair<StringRef, llvm::Comdat::SelectionKind>> ComdatTable;
290    ComdatTable.reserve(Comdats.size());
291    for (auto C : Comdats)
292      ComdatTable.push_back({str(C.Name), llvm::Comdat::SelectionKind(
293                                              uint32_t(C.SelectionKind))});
294    return ComdatTable;
295  }
296 
297  /// COFF-specific: returns linker options specified in the input file.
298  StringRef getCOFFLinkerOpts() const { return str(header().COFFLinkerOpts); }
299 
300  /// Returns dependent library specifiers
301  std::vector<StringRef> getDependentLibraries() const {
302    std::vector<StringRef> Specifiers;
303    Specifiers.reserve(DependentLibraries.size());
304    for (auto S : DependentLibraries) {
305      Specifiers.push_back(str(S));
306    }
307    return Specifiers;
308  }
309};
310 
311/// Ephemeral symbols produced by Reader::symbols() and
312/// Reader::module_symbols().
313class Reader::SymbolRef : public Symbol {
314  const storage::Symbol *SymI, *SymE;
315  const storage::Uncommon *UncI;
316  const Reader *R;
317 
318  void read() {
319    if (SymI == SymE)
8
←
Assuming field 'SymI' is not equal to field 'SymE'→
9
←
Taking false branch→
320      return;
321 
322    Name = R->str(SymI->Name);
323    IRName = R->str(SymI->IRName);
324    ComdatIndex = SymI->ComdatIndex;
325    Flags = SymI->Flags;
326 
327    if (Flags & (1 << storage::Symbol::FB_has_uncommon)) {
10
←
Assuming the condition is true→
11
←
Taking true branch→
328      CommonSize = UncI->CommonSize;
12
←
Called C++ object pointer is null
329      CommonAlign = UncI->CommonAlign;
330      COFFWeakExternFallbackName = R->str(UncI->COFFWeakExternFallbackName);
331      SectionName = R->str(UncI->SectionName);
332    } else
333      // Reset this field so it can be queried unconditionally for all symbols.
334      SectionName = "";
335  }
336 
337public:
338  SymbolRef(const storage::Symbol *SymI, const storage::Symbol *SymE,
339            const storage::Uncommon *UncI, const Reader *R)
340      : SymI(SymI), SymE(SymE), UncI(UncI), R(R) {
6
←
Null pointer value stored to 'symb.UncI'→
341    read();
7
←
Calling 'SymbolRef::read'→
342  }
343 
344  void moveNext() {
345    ++SymI;
346    if (Flags & (1 << storage::Symbol::FB_has_uncommon))
347      ++UncI;
348    read();
349  }
350 
351  bool operator==(const SymbolRef &Other) const { return SymI == Other.SymI; }
352};
353 
354inline Reader::symbol_range Reader::symbols() const {
355  return {SymbolRef(Symbols.begin(), Symbols.end(), Uncommons.begin(), this),
356          SymbolRef(Symbols.end(), Symbols.end(), nullptr, this)};
4
←
Passing null pointer value via 3rd parameter 'UncI'→
5
←
Calling constructor for 'SymbolRef'→
357}
358 
359inline Reader::symbol_range Reader::module_symbols(unsigned I) const {
360  const storage::Module &M = Modules[I];
361  const storage::Symbol *MBegin = Symbols.begin() + M.Begin,
362                        *MEnd = Symbols.begin() + M.End;
363  return {SymbolRef(MBegin, MEnd, Uncommons.begin() + M.UncBegin, this),
364          SymbolRef(MEnd, MEnd, nullptr, this)};
365}
366 
367/// The contents of the irsymtab in a bitcode file. Any underlying data for the
368/// irsymtab are owned by Symtab and Strtab.
369struct FileContents {
370  SmallVector<char, 0> Symtab, Strtab;
371  Reader TheReader;
372};
373 
374/// Reads the contents of a bitcode file, creating its irsymtab if necessary.
375Expected<FileContents> readBitcode(const BitcodeFileContents &BFC);
376 
377} // end namespace irsymtab
378} // end namespace llvm
379 
380#endif // LLVM_OBJECT_IRSYMTAB_H