diff options
Diffstat (limited to 'tools/llvm-objdump')
-rw-r--r-- | tools/llvm-objdump/COFFDump.cpp | 82 | ||||
-rw-r--r-- | tools/llvm-objdump/ELFDump.cpp | 252 | ||||
-rw-r--r-- | tools/llvm-objdump/MachODump.cpp | 978 | ||||
-rw-r--r-- | tools/llvm-objdump/WasmDump.cpp | 40 | ||||
-rw-r--r-- | tools/llvm-objdump/llvm-objdump.cpp | 2215 | ||||
-rw-r--r-- | tools/llvm-objdump/llvm-objdump.h | 153 |
6 files changed, 1989 insertions, 1731 deletions
diff --git a/tools/llvm-objdump/COFFDump.cpp b/tools/llvm-objdump/COFFDump.cpp index 55607ec299be..1ba0a68902c9 100644 --- a/tools/llvm-objdump/COFFDump.cpp +++ b/tools/llvm-objdump/COFFDump.cpp @@ -1,9 +1,8 @@ //===-- COFFDump.cpp - COFF-specific dumper ---------------------*- C++ -*-===// // -// The LLVM Compiler Infrastructure -// -// This file is distributed under the University of Illinois Open Source -// License. See LICENSE.TXT for details. +// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions. +// See https://llvm.org/LICENSE.txt for license information. +// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception // //===----------------------------------------------------------------------===// /// @@ -25,10 +24,10 @@ #include "llvm/Support/WithColor.h" #include "llvm/Support/raw_ostream.h" -using namespace llvm; -using namespace object; +using namespace llvm::object; using namespace llvm::Win64EH; +namespace llvm { // Returns the name of the unwind code. static StringRef getUnwindCodeTypeName(uint8_t Code) { switch(Code) { @@ -156,70 +155,68 @@ static void printAllUnwindCodes(ArrayRef<UnwindCode> UCs) { } // Given a symbol sym this functions returns the address and section of it. -static std::error_code -resolveSectionAndAddress(const COFFObjectFile *Obj, const SymbolRef &Sym, - const coff_section *&ResolvedSection, - uint64_t &ResolvedAddr) { +static Error resolveSectionAndAddress(const COFFObjectFile *Obj, + const SymbolRef &Sym, + const coff_section *&ResolvedSection, + uint64_t &ResolvedAddr) { Expected<uint64_t> ResolvedAddrOrErr = Sym.getAddress(); if (!ResolvedAddrOrErr) - return errorToErrorCode(ResolvedAddrOrErr.takeError()); + return ResolvedAddrOrErr.takeError(); ResolvedAddr = *ResolvedAddrOrErr; Expected<section_iterator> Iter = Sym.getSection(); if (!Iter) - return errorToErrorCode(Iter.takeError()); + return Iter.takeError(); ResolvedSection = Obj->getCOFFSection(**Iter); - return std::error_code(); + return Error::success(); } // Given a vector of relocations for a section and an offset into this section // the function returns the symbol used for the relocation at the offset. -static std::error_code resolveSymbol(const std::vector<RelocationRef> &Rels, +static Error resolveSymbol(const std::vector<RelocationRef> &Rels, uint64_t Offset, SymbolRef &Sym) { for (auto &R : Rels) { uint64_t Ofs = R.getOffset(); if (Ofs == Offset) { Sym = *R.getSymbol(); - return std::error_code(); + return Error::success(); } } - return object_error::parse_failed; + return make_error<BinaryError>(); } // Given a vector of relocations for a section and an offset into this section // the function resolves the symbol used for the relocation at the offset and // returns the section content and the address inside the content pointed to // by the symbol. -static std::error_code +static Error getSectionContents(const COFFObjectFile *Obj, const std::vector<RelocationRef> &Rels, uint64_t Offset, ArrayRef<uint8_t> &Contents, uint64_t &Addr) { SymbolRef Sym; - if (std::error_code EC = resolveSymbol(Rels, Offset, Sym)) - return EC; + if (Error E = resolveSymbol(Rels, Offset, Sym)) + return E; const coff_section *Section; - if (std::error_code EC = resolveSectionAndAddress(Obj, Sym, Section, Addr)) - return EC; - if (std::error_code EC = Obj->getSectionContents(Section, Contents)) - return EC; - return std::error_code(); + if (Error E = resolveSectionAndAddress(Obj, Sym, Section, Addr)) + return E; + return Obj->getSectionContents(Section, Contents); } // Given a vector of relocations for a section and an offset into this section // the function returns the name of the symbol used for the relocation at the // offset. -static std::error_code resolveSymbolName(const std::vector<RelocationRef> &Rels, - uint64_t Offset, StringRef &Name) { +static Error resolveSymbolName(const std::vector<RelocationRef> &Rels, + uint64_t Offset, StringRef &Name) { SymbolRef Sym; - if (std::error_code EC = resolveSymbol(Rels, Offset, Sym)) + if (Error EC = resolveSymbol(Rels, Offset, Sym)) return EC; Expected<StringRef> NameOrErr = Sym.getName(); if (!NameOrErr) - return errorToErrorCode(NameOrErr.takeError()); + return NameOrErr.takeError(); Name = *NameOrErr; - return std::error_code(); + return Error::success(); } -static void printCOFFSymbolAddress(llvm::raw_ostream &Out, +static void printCOFFSymbolAddress(raw_ostream &Out, const std::vector<RelocationRef> &Rels, uint64_t Offset, uint32_t Disp) { StringRef Sym; @@ -469,6 +466,18 @@ static bool getPDataSection(const COFFObjectFile *Obj, return false; } +Error getCOFFRelocationValueString(const COFFObjectFile *Obj, + const RelocationRef &Rel, + SmallVectorImpl<char> &Result) { + symbol_iterator SymI = Rel.getSymbol(); + Expected<StringRef> SymNameOrErr = SymI->getName(); + if (!SymNameOrErr) + return SymNameOrErr.takeError(); + StringRef SymName = *SymNameOrErr; + Result.append(SymName.begin(), SymName.end()); + return Error::success(); +} + static void printWin64EHUnwindInfo(const Win64EH::UnwindInfo *UI) { // The casts to int are required in order to output the value as number. // Without the casts the value would be interpreted as char data (which @@ -578,7 +587,7 @@ static void printRuntimeFunctionRels(const COFFObjectFile *Obj, printWin64EHUnwindInfo(UI); } -void llvm::printCOFFUnwindInfo(const COFFObjectFile *Obj) { +void printCOFFUnwindInfo(const COFFObjectFile *Obj) { if (Obj->getMachine() != COFF::IMAGE_FILE_MACHINE_AMD64) { WithColor::error(errs(), "llvm-objdump") << "unsupported image machine type " @@ -607,7 +616,7 @@ void llvm::printCOFFUnwindInfo(const COFFObjectFile *Obj) { } } -void llvm::printCOFFFileHeader(const object::ObjectFile *Obj) { +void printCOFFFileHeader(const object::ObjectFile *Obj) { const COFFObjectFile *file = dyn_cast<const COFFObjectFile>(Obj); printTLSDirectory(file); printLoadConfiguration(file); @@ -615,7 +624,7 @@ void llvm::printCOFFFileHeader(const object::ObjectFile *Obj) { printExportTable(file); } -void llvm::printCOFFSymbolTable(const object::COFFImportFile *i) { +void printCOFFSymbolTable(const object::COFFImportFile *i) { unsigned Index = 0; bool IsCode = i->getCOFFImportHeader()->getType() == COFF::IMPORT_CODE; @@ -623,7 +632,7 @@ void llvm::printCOFFSymbolTable(const object::COFFImportFile *i) { std::string Name; raw_string_ostream NS(Name); - Sym.printName(NS); + cantFail(Sym.printName(NS)); NS.flush(); outs() << "[" << format("%2d", Index) << "]" @@ -638,11 +647,11 @@ void llvm::printCOFFSymbolTable(const object::COFFImportFile *i) { } } -void llvm::printCOFFSymbolTable(const COFFObjectFile *coff) { +void printCOFFSymbolTable(const COFFObjectFile *coff) { for (unsigned SI = 0, SE = coff->getNumberOfSymbols(); SI != SE; ++SI) { Expected<COFFSymbolRef> Symbol = coff->getSymbol(SI); StringRef Name; - error(errorToErrorCode(Symbol.takeError())); + error(Symbol.takeError()); error(coff->getSymbolName(*Symbol, Name)); outs() << "[" << format("%2d", SI) << "]" @@ -709,3 +718,4 @@ void llvm::printCOFFSymbolTable(const COFFObjectFile *coff) { } } } +} // namespace llvm diff --git a/tools/llvm-objdump/ELFDump.cpp b/tools/llvm-objdump/ELFDump.cpp index b17a15a0d8fc..9c4d67d0f1bd 100644 --- a/tools/llvm-objdump/ELFDump.cpp +++ b/tools/llvm-objdump/ELFDump.cpp @@ -1,9 +1,8 @@ //===-- ELFDump.cpp - ELF-specific dumper -----------------------*- C++ -*-===// // -// The LLVM Compiler Infrastructure -// -// This file is distributed under the University of Illinois Open Source -// License. See LICENSE.TXT for details. +// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions. +// See https://llvm.org/LICENSE.txt for license information. +// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception // //===----------------------------------------------------------------------===// /// @@ -13,23 +12,22 @@ //===----------------------------------------------------------------------===// #include "llvm-objdump.h" +#include "llvm/Demangle/Demangle.h" #include "llvm/Object/ELFObjectFile.h" #include "llvm/Support/Format.h" #include "llvm/Support/MathExtras.h" #include "llvm/Support/raw_ostream.h" -using namespace llvm; using namespace llvm::object; +namespace llvm { template <class ELFT> -Expected<StringRef> getDynamicStrTab(const ELFFile<ELFT> *Elf) { - typedef ELFFile<ELFT> ELFO; - +static Expected<StringRef> getDynamicStrTab(const ELFFile<ELFT> *Elf) { auto DynamicEntriesOrError = Elf->dynamicEntries(); if (!DynamicEntriesOrError) return DynamicEntriesOrError.takeError(); - for (const typename ELFO::Elf_Dyn &Dyn : *DynamicEntriesOrError) { + for (const typename ELFT::Dyn &Dyn : *DynamicEntriesOrError) { if (Dyn.d_tag == ELF::DT_STRTAB) { auto MappedAddrOrError = Elf->toMappedAddr(Dyn.getPtr()); if (!MappedAddrOrError) @@ -43,7 +41,7 @@ Expected<StringRef> getDynamicStrTab(const ELFFile<ELFT> *Elf) { if (!SectionsOrError) return SectionsOrError.takeError(); - for (const typename ELFO::Elf_Shdr &Sec : *SectionsOrError) { + for (const typename ELFT::Shdr &Sec : *SectionsOrError) { if (Sec.sh_type == ELF::SHT_DYNSYM) return Elf->getStringTableForSymtab(Sec); } @@ -52,40 +50,135 @@ Expected<StringRef> getDynamicStrTab(const ELFFile<ELFT> *Elf) { } template <class ELFT> -void printDynamicSection(const ELFFile<ELFT> *Elf, StringRef Filename) { - auto ProgramHeaderOrError = Elf->program_headers(); - if (!ProgramHeaderOrError) - report_error(Filename, ProgramHeaderOrError.takeError()); +static Error getRelocationValueString(const ELFObjectFile<ELFT> *Obj, + const RelocationRef &RelRef, + SmallVectorImpl<char> &Result) { + const ELFFile<ELFT> &EF = *Obj->getELFFile(); + DataRefImpl Rel = RelRef.getRawDataRefImpl(); + auto SecOrErr = EF.getSection(Rel.d.a); + if (!SecOrErr) + return SecOrErr.takeError(); - auto DynamicEntriesOrError = Elf->dynamicEntries(); - if (!DynamicEntriesOrError) - report_error(Filename, DynamicEntriesOrError.takeError()); + int64_t Addend = 0; + // If there is no Symbol associated with the relocation, we set the undef + // boolean value to 'true'. This will prevent us from calling functions that + // requires the relocation to be associated with a symbol. + // + // In SHT_REL case we would need to read the addend from section data. + // GNU objdump does not do that and we just follow for simplicity atm. + bool Undef = false; + if ((*SecOrErr)->sh_type == ELF::SHT_RELA) { + const typename ELFT::Rela *ERela = Obj->getRela(Rel); + Addend = ERela->r_addend; + Undef = ERela->getSymbol(false) == 0; + } else if ((*SecOrErr)->sh_type != ELF::SHT_REL) { + return make_error<BinaryError>(); + } + + // Default scheme is to print Target, as well as "+ <addend>" for nonzero + // addend. Should be acceptable for all normal purposes. + std::string FmtBuf; + raw_string_ostream Fmt(FmtBuf); + + if (!Undef) { + symbol_iterator SI = RelRef.getSymbol(); + const typename ELFT::Sym *Sym = Obj->getSymbol(SI->getRawDataRefImpl()); + if (Sym->getType() == ELF::STT_SECTION) { + Expected<section_iterator> SymSI = SI->getSection(); + if (!SymSI) + return SymSI.takeError(); + const typename ELFT::Shdr *SymSec = + Obj->getSection((*SymSI)->getRawDataRefImpl()); + auto SecName = EF.getSectionName(SymSec); + if (!SecName) + return SecName.takeError(); + Fmt << *SecName; + } else { + Expected<StringRef> SymName = SI->getName(); + if (!SymName) + return SymName.takeError(); + if (Demangle) + Fmt << demangle(*SymName); + else + Fmt << *SymName; + } + } else { + Fmt << "*ABS*"; + } + + if (Addend != 0) + Fmt << (Addend < 0 ? "" : "+") << Addend; + Fmt.flush(); + Result.append(FmtBuf.begin(), FmtBuf.end()); + return Error::success(); +} +Error getELFRelocationValueString(const ELFObjectFileBase *Obj, + const RelocationRef &Rel, + SmallVectorImpl<char> &Result) { + if (auto *ELF32LE = dyn_cast<ELF32LEObjectFile>(Obj)) + return getRelocationValueString(ELF32LE, Rel, Result); + if (auto *ELF64LE = dyn_cast<ELF64LEObjectFile>(Obj)) + return getRelocationValueString(ELF64LE, Rel, Result); + if (auto *ELF32BE = dyn_cast<ELF32BEObjectFile>(Obj)) + return getRelocationValueString(ELF32BE, Rel, Result); + auto *ELF64BE = cast<ELF64BEObjectFile>(Obj); + return getRelocationValueString(ELF64BE, Rel, Result); +} + +template <class ELFT> +static uint64_t getSectionLMA(const ELFFile<ELFT> *Obj, + const object::ELFSectionRef &Sec) { + auto PhdrRangeOrErr = Obj->program_headers(); + if (!PhdrRangeOrErr) + report_fatal_error(toString(PhdrRangeOrErr.takeError())); + + // Search for a PT_LOAD segment containing the requested section. Use this + // segment's p_addr to calculate the section's LMA. + for (const typename ELFT::Phdr &Phdr : *PhdrRangeOrErr) + if ((Phdr.p_type == ELF::PT_LOAD) && (Phdr.p_vaddr <= Sec.getAddress()) && + (Phdr.p_vaddr + Phdr.p_memsz > Sec.getAddress())) + return Sec.getAddress() - Phdr.p_vaddr + Phdr.p_paddr; + + // Return section's VMA if it isn't in a PT_LOAD segment. + return Sec.getAddress(); +} + +uint64_t getELFSectionLMA(const object::ELFSectionRef &Sec) { + if (const auto *ELFObj = dyn_cast<ELF32LEObjectFile>(Sec.getObject())) + return getSectionLMA(ELFObj->getELFFile(), Sec); + else if (const auto *ELFObj = dyn_cast<ELF32BEObjectFile>(Sec.getObject())) + return getSectionLMA(ELFObj->getELFFile(), Sec); + else if (const auto *ELFObj = dyn_cast<ELF64LEObjectFile>(Sec.getObject())) + return getSectionLMA(ELFObj->getELFFile(), Sec); + const auto *ELFObj = cast<ELF64BEObjectFile>(Sec.getObject()); + return getSectionLMA(ELFObj->getELFFile(), Sec); +} + +template <class ELFT> +void printDynamicSection(const ELFFile<ELFT> *Elf, StringRef Filename) { + ArrayRef<typename ELFT::Dyn> DynamicEntries = + unwrapOrError(Elf->dynamicEntries(), Filename); outs() << "Dynamic Section:\n"; - for (const auto &Dyn : *DynamicEntriesOrError) { + for (const typename ELFT::Dyn &Dyn : DynamicEntries) { if (Dyn.d_tag == ELF::DT_NULL) continue; - StringRef Str = StringRef(Elf->getDynamicTagAsString(Dyn.d_tag)); - - if (Str.empty()) { - std::string HexStr = utohexstr(static_cast<uint64_t>(Dyn.d_tag), true); - outs() << format(" 0x%-19s", HexStr.c_str()); - } else { - // We use "-21" in order to match GNU objdump's output. - outs() << format(" %-21s", Str.data()); - } + std::string Str = Elf->getDynamicTagAsString(Dyn.d_tag); + outs() << format(" %-21s", Str.c_str()); const char *Fmt = ELFT::Is64Bits ? "0x%016" PRIx64 "\n" : "0x%08" PRIx64 "\n"; - if (Dyn.d_tag == ELF::DT_NEEDED) { + if (Dyn.d_tag == ELF::DT_NEEDED || Dyn.d_tag == ELF::DT_RPATH || + Dyn.d_tag == ELF::DT_RUNPATH || Dyn.d_tag == ELF::DT_SONAME || + Dyn.d_tag == ELF::DT_AUXILIARY || Dyn.d_tag == ELF::DT_FILTER) { Expected<StringRef> StrTabOrErr = getDynamicStrTab(Elf); if (StrTabOrErr) { const char *Data = StrTabOrErr.get().data(); outs() << (Data + Dyn.d_un.d_val) << "\n"; continue; } - warn(errorToErrorCode(StrTabOrErr.takeError()).message()); + warn(toString(StrTabOrErr.takeError())); consumeError(StrTabOrErr.takeError()); } outs() << format(Fmt, (uint64_t)Dyn.d_un.d_val); @@ -93,13 +186,11 @@ void printDynamicSection(const ELFFile<ELFT> *Elf, StringRef Filename) { } template <class ELFT> void printProgramHeaders(const ELFFile<ELFT> *o) { - typedef ELFFile<ELFT> ELFO; outs() << "Program Header:\n"; auto ProgramHeaderOrError = o->program_headers(); if (!ProgramHeaderOrError) - report_fatal_error( - errorToErrorCode(ProgramHeaderOrError.takeError()).message()); - for (const typename ELFO::Elf_Phdr &Phdr : *ProgramHeaderOrError) { + report_fatal_error(toString(ProgramHeaderOrError.takeError())); + for (const typename ELFT::Phdr &Phdr : *ProgramHeaderOrError) { switch (Phdr.p_type) { case ELF::PT_DYNAMIC: outs() << " DYNAMIC "; @@ -157,7 +248,86 @@ template <class ELFT> void printProgramHeaders(const ELFFile<ELFT> *o) { outs() << "\n"; } -void llvm::printELFFileHeader(const object::ObjectFile *Obj) { +template <class ELFT> +void printSymbolVersionDependency(ArrayRef<uint8_t> Contents, + StringRef StrTab) { + outs() << "Version References:\n"; + + const uint8_t *Buf = Contents.data(); + while (Buf) { + auto *Verneed = reinterpret_cast<const typename ELFT::Verneed *>(Buf); + outs() << " required from " + << StringRef(StrTab.drop_front(Verneed->vn_file).data()) << ":\n"; + + const uint8_t *BufAux = Buf + Verneed->vn_aux; + while (BufAux) { + auto *Vernaux = reinterpret_cast<const typename ELFT::Vernaux *>(BufAux); + outs() << " " + << format("0x%08" PRIx32 " ", (uint32_t)Vernaux->vna_hash) + << format("0x%02" PRIx16 " ", (uint16_t)Vernaux->vna_flags) + << format("%02" PRIu16 " ", (uint16_t)Vernaux->vna_other) + << StringRef(StrTab.drop_front(Vernaux->vna_name).data()) << '\n'; + BufAux = Vernaux->vna_next ? BufAux + Vernaux->vna_next : nullptr; + } + Buf = Verneed->vn_next ? Buf + Verneed->vn_next : nullptr; + } +} + +template <class ELFT> +void printSymbolVersionDefinition(const typename ELFT::Shdr &Shdr, + ArrayRef<uint8_t> Contents, + StringRef StrTab) { + outs() << "Version definitions:\n"; + + const uint8_t *Buf = Contents.data(); + uint32_t VerdefIndex = 1; + // sh_info contains the number of entries in the SHT_GNU_verdef section. To + // make the index column have consistent width, we should insert blank spaces + // according to sh_info. + uint16_t VerdefIndexWidth = std::to_string(Shdr.sh_info).size(); + while (Buf) { + auto *Verdef = reinterpret_cast<const typename ELFT::Verdef *>(Buf); + outs() << format_decimal(VerdefIndex++, VerdefIndexWidth) << " " + << format("0x%02" PRIx16 " ", (uint16_t)Verdef->vd_flags) + << format("0x%08" PRIx32 " ", (uint32_t)Verdef->vd_hash); + + const uint8_t *BufAux = Buf + Verdef->vd_aux; + uint16_t VerdauxIndex = 0; + while (BufAux) { + auto *Verdaux = reinterpret_cast<const typename ELFT::Verdaux *>(BufAux); + if (VerdauxIndex) + outs() << std::string(VerdefIndexWidth + 17, ' '); + outs() << StringRef(StrTab.drop_front(Verdaux->vda_name).data()) << '\n'; + BufAux = Verdaux->vda_next ? BufAux + Verdaux->vda_next : nullptr; + ++VerdauxIndex; + } + Buf = Verdef->vd_next ? Buf + Verdef->vd_next : nullptr; + } +} + +template <class ELFT> +void printSymbolVersionInfo(const ELFFile<ELFT> *Elf, StringRef FileName) { + ArrayRef<typename ELFT::Shdr> Sections = + unwrapOrError(Elf->sections(), FileName); + for (const typename ELFT::Shdr &Shdr : Sections) { + if (Shdr.sh_type != ELF::SHT_GNU_verneed && + Shdr.sh_type != ELF::SHT_GNU_verdef) + continue; + + ArrayRef<uint8_t> Contents = + unwrapOrError(Elf->getSectionContents(&Shdr), FileName); + const typename ELFT::Shdr *StrTabSec = + unwrapOrError(Elf->getSection(Shdr.sh_link), FileName); + StringRef StrTab = unwrapOrError(Elf->getStringTable(StrTabSec), FileName); + + if (Shdr.sh_type == ELF::SHT_GNU_verneed) + printSymbolVersionDependency<ELFT>(Contents, StrTab); + else + printSymbolVersionDefinition<ELFT>(Shdr, Contents, StrTab); + } +} + +void printELFFileHeader(const object::ObjectFile *Obj) { if (const auto *ELFObj = dyn_cast<ELF32LEObjectFile>(Obj)) printProgramHeaders(ELFObj->getELFFile()); else if (const auto *ELFObj = dyn_cast<ELF32BEObjectFile>(Obj)) @@ -168,7 +338,7 @@ void llvm::printELFFileHeader(const object::ObjectFile *Obj) { printProgramHeaders(ELFObj->getELFFile()); } -void llvm::printELFDynamicSection(const object::ObjectFile *Obj) { +void printELFDynamicSection(const object::ObjectFile *Obj) { if (const auto *ELFObj = dyn_cast<ELF32LEObjectFile>(Obj)) printDynamicSection(ELFObj->getELFFile(), Obj->getFileName()); else if (const auto *ELFObj = dyn_cast<ELF32BEObjectFile>(Obj)) @@ -178,3 +348,15 @@ void llvm::printELFDynamicSection(const object::ObjectFile *Obj) { else if (const auto *ELFObj = dyn_cast<ELF64BEObjectFile>(Obj)) printDynamicSection(ELFObj->getELFFile(), Obj->getFileName()); } + +void printELFSymbolVersionInfo(const object::ObjectFile *Obj) { + if (const auto *ELFObj = dyn_cast<ELF32LEObjectFile>(Obj)) + printSymbolVersionInfo(ELFObj->getELFFile(), Obj->getFileName()); + else if (const auto *ELFObj = dyn_cast<ELF32BEObjectFile>(Obj)) + printSymbolVersionInfo(ELFObj->getELFFile(), Obj->getFileName()); + else if (const auto *ELFObj = dyn_cast<ELF64LEObjectFile>(Obj)) + printSymbolVersionInfo(ELFObj->getELFFile(), Obj->getFileName()); + else if (const auto *ELFObj = dyn_cast<ELF64BEObjectFile>(Obj)) + printSymbolVersionInfo(ELFObj->getELFFile(), Obj->getFileName()); +} +} // namespace llvm diff --git a/tools/llvm-objdump/MachODump.cpp b/tools/llvm-objdump/MachODump.cpp index 5ef7058ec9da..58ff7be4543c 100644 --- a/tools/llvm-objdump/MachODump.cpp +++ b/tools/llvm-objdump/MachODump.cpp @@ -1,9 +1,8 @@ //===-- MachODump.cpp - Object file dumping utility for llvm --------------===// // -// The LLVM Compiler Infrastructure -// -// This file is distributed under the University of Illinois Open Source -// License. See LICENSE.TXT for details. +// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions. +// See https://llvm.org/LICENSE.txt for license information. +// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception // //===----------------------------------------------------------------------===// // @@ -56,83 +55,140 @@ extern "C" { } #endif -using namespace llvm; -using namespace object; +using namespace llvm::object; + +namespace llvm { + +cl::OptionCategory MachOCat("llvm-objdump MachO Specific Options"); + +extern cl::opt<bool> ArchiveHeaders; +extern cl::opt<bool> Disassemble; +extern cl::opt<bool> DisassembleAll; +extern cl::opt<DIDumpType> DwarfDumpType; +extern cl::list<std::string> FilterSections; +extern cl::list<std::string> MAttrs; +extern cl::opt<std::string> MCPU; +extern cl::opt<bool> NoShowRawInsn; +extern cl::opt<bool> NoLeadingAddr; +extern cl::opt<bool> PrintImmHex; +extern cl::opt<bool> PrivateHeaders; +extern cl::opt<bool> Relocations; +extern cl::opt<bool> SectionHeaders; +extern cl::opt<bool> SectionContents; +extern cl::opt<bool> SymbolTable; +extern cl::opt<std::string> TripleName; +extern cl::opt<bool> UnwindInfo; + +cl::opt<bool> + FirstPrivateHeader("private-header", + cl::desc("Display only the first format specific file " + "header"), + cl::cat(MachOCat)); + +cl::opt<bool> ExportsTrie("exports-trie", + cl::desc("Display mach-o exported symbols"), + cl::cat(MachOCat)); + +cl::opt<bool> Rebase("rebase", cl::desc("Display mach-o rebasing info"), + cl::cat(MachOCat)); + +cl::opt<bool> Bind("bind", cl::desc("Display mach-o binding info"), + cl::cat(MachOCat)); + +cl::opt<bool> LazyBind("lazy-bind", + cl::desc("Display mach-o lazy binding info"), + cl::cat(MachOCat)); + +cl::opt<bool> WeakBind("weak-bind", + cl::desc("Display mach-o weak binding info"), + cl::cat(MachOCat)); static cl::opt<bool> - UseDbg("g", - cl::desc("Print line information from debug info if available")); + UseDbg("g", cl::Grouping, + cl::desc("Print line information from debug info if available"), + cl::cat(MachOCat)); static cl::opt<std::string> DSYMFile("dsym", - cl::desc("Use .dSYM file for debug info")); + cl::desc("Use .dSYM file for debug info"), + cl::cat(MachOCat)); static cl::opt<bool> FullLeadingAddr("full-leading-addr", - cl::desc("Print full leading address")); + cl::desc("Print full leading address"), + cl::cat(MachOCat)); static cl::opt<bool> NoLeadingHeaders("no-leading-headers", - cl::desc("Print no leading headers")); + cl::desc("Print no leading headers"), + cl::cat(MachOCat)); -cl::opt<bool> llvm::UniversalHeaders("universal-headers", - cl::desc("Print Mach-O universal headers " - "(requires -macho)")); +cl::opt<bool> UniversalHeaders("universal-headers", + cl::desc("Print Mach-O universal headers " + "(requires -macho)"), + cl::cat(MachOCat)); cl::opt<bool> ArchiveMemberOffsets("archive-member-offsets", cl::desc("Print the offset to each archive member for " "Mach-O archives (requires -macho and " - "-archive-headers)")); - -cl::opt<bool> - llvm::IndirectSymbols("indirect-symbols", - cl::desc("Print indirect symbol table for Mach-O " - "objects (requires -macho)")); + "-archive-headers)"), + cl::cat(MachOCat)); -cl::opt<bool> - llvm::DataInCode("data-in-code", - cl::desc("Print the data in code table for Mach-O objects " - "(requires -macho)")); - -cl::opt<bool> - llvm::LinkOptHints("link-opt-hints", - cl::desc("Print the linker optimization hints for " - "Mach-O objects (requires -macho)")); - -cl::opt<bool> - llvm::InfoPlist("info-plist", - cl::desc("Print the info plist section as strings for " - "Mach-O objects (requires -macho)")); +cl::opt<bool> IndirectSymbols("indirect-symbols", + cl::desc("Print indirect symbol table for Mach-O " + "objects (requires -macho)"), + cl::cat(MachOCat)); cl::opt<bool> - llvm::DylibsUsed("dylibs-used", - cl::desc("Print the shared libraries used for linked " - "Mach-O files (requires -macho)")); + DataInCode("data-in-code", + cl::desc("Print the data in code table for Mach-O objects " + "(requires -macho)"), + cl::cat(MachOCat)); + +cl::opt<bool> LinkOptHints("link-opt-hints", + cl::desc("Print the linker optimization hints for " + "Mach-O objects (requires -macho)"), + cl::cat(MachOCat)); + +cl::opt<bool> InfoPlist("info-plist", + cl::desc("Print the info plist section as strings for " + "Mach-O objects (requires -macho)"), + cl::cat(MachOCat)); + +cl::opt<bool> DylibsUsed("dylibs-used", + cl::desc("Print the shared libraries used for linked " + "Mach-O files (requires -macho)"), + cl::cat(MachOCat)); cl::opt<bool> - llvm::DylibId("dylib-id", - cl::desc("Print the shared library's id for the dylib Mach-O " - "file (requires -macho)")); + DylibId("dylib-id", + cl::desc("Print the shared library's id for the dylib Mach-O " + "file (requires -macho)"), + cl::cat(MachOCat)); cl::opt<bool> - llvm::NonVerbose("non-verbose", - cl::desc("Print the info for Mach-O objects in " - "non-verbose or numeric form (requires -macho)")); + NonVerbose("non-verbose", + cl::desc("Print the info for Mach-O objects in " + "non-verbose or numeric form (requires -macho)"), + cl::cat(MachOCat)); cl::opt<bool> - llvm::ObjcMetaData("objc-meta-data", - cl::desc("Print the Objective-C runtime meta data for " - "Mach-O files (requires -macho)")); + ObjcMetaData("objc-meta-data", + cl::desc("Print the Objective-C runtime meta data for " + "Mach-O files (requires -macho)"), + cl::cat(MachOCat)); -cl::opt<std::string> llvm::DisSymName( +cl::opt<std::string> DisSymName( "dis-symname", - cl::desc("disassemble just this symbol's instructions (requires -macho)")); + cl::desc("disassemble just this symbol's instructions (requires -macho)"), + cl::cat(MachOCat)); static cl::opt<bool> NoSymbolicOperands( "no-symbolic-operands", - cl::desc("do not symbolic operands when disassembling (requires -macho)")); + cl::desc("do not symbolic operands when disassembling (requires -macho)"), + cl::cat(MachOCat)); static cl::list<std::string> ArchFlags("arch", cl::desc("architecture(s) from a Mach-O file to dump"), - cl::ZeroOrMore); + cl::ZeroOrMore, cl::cat(MachOCat)); bool ArchAll = false; @@ -142,7 +198,7 @@ static const Target *GetTarget(const MachOObjectFile *MachOObj, const char **McpuDefault, const Target **ThumbTarget) { // Figure out the target triple. - llvm::Triple TT(TripleName); + Triple TT(TripleName); if (TripleName.empty()) { TT = MachOObj->getArchTriple(McpuDefault); TripleName = TT.str(); @@ -151,7 +207,7 @@ static const Target *GetTarget(const MachOObjectFile *MachOObj, if (TT.getArch() == Triple::arm) { // We've inferred a 32-bit ARM target from the object file. All MachO CPUs // that support ARM are also capable of Thumb mode. - llvm::Triple ThumbTriple = TT; + Triple ThumbTriple = TT; std::string ThumbName = (Twine("thumb") + TT.getArchName().substr(3)).str(); ThumbTriple.setArchName(ThumbName); ThumbTripleName = ThumbTriple.str(); @@ -180,11 +236,11 @@ struct SymbolSorter { bool operator()(const SymbolRef &A, const SymbolRef &B) { Expected<SymbolRef::Type> ATypeOrErr = A.getType(); if (!ATypeOrErr) - report_error(A.getObject()->getFileName(), ATypeOrErr.takeError()); + report_error(ATypeOrErr.takeError(), A.getObject()->getFileName()); SymbolRef::Type AType = *ATypeOrErr; Expected<SymbolRef::Type> BTypeOrErr = B.getType(); if (!BTypeOrErr) - report_error(B.getObject()->getFileName(), BTypeOrErr.takeError()); + report_error(BTypeOrErr.takeError(), B.getObject()->getFileName()); SymbolRef::Type BType = *BTypeOrErr; uint64_t AAddr = (AType != SymbolRef::ST_Function) ? 0 : A.getValue(); uint64_t BAddr = (BType != SymbolRef::ST_Function) ? 0 : B.getValue(); @@ -308,11 +364,10 @@ static void getSectionsAndSymbols(MachOObjectFile *MachOObj, std::vector<SymbolRef> &Symbols, SmallVectorImpl<uint64_t> &FoundFns, uint64_t &BaseSegmentAddress) { + const StringRef FileName = MachOObj->getFileName(); for (const SymbolRef &Symbol : MachOObj->symbols()) { - Expected<StringRef> SymName = Symbol.getName(); - if (!SymName) - report_error(MachOObj->getFileName(), SymName.takeError()); - if (!SymName->startswith("ltmp")) + StringRef SymName = unwrapOrError(Symbol.getName(), FileName); + if (!SymName.startswith("ltmp")) Symbols.push_back(Symbol); } @@ -342,6 +397,254 @@ static void getSectionsAndSymbols(MachOObjectFile *MachOObj, } } +static void printRelocationTargetName(const MachOObjectFile *O, + const MachO::any_relocation_info &RE, + raw_string_ostream &Fmt) { + // Target of a scattered relocation is an address. In the interest of + // generating pretty output, scan through the symbol table looking for a + // symbol that aligns with that address. If we find one, print it. + // Otherwise, we just print the hex address of the target. + const StringRef FileName = O->getFileName(); + if (O->isRelocationScattered(RE)) { + uint32_t Val = O->getPlainRelocationSymbolNum(RE); + + for (const SymbolRef &Symbol : O->symbols()) { + uint64_t Addr = unwrapOrError(Symbol.getAddress(), FileName); + if (Addr != Val) + continue; + Fmt << unwrapOrError(Symbol.getName(), FileName); + return; + } + + // If we couldn't find a symbol that this relocation refers to, try + // to find a section beginning instead. + for (const SectionRef &Section : ToolSectionFilter(*O)) { + StringRef Name; + uint64_t Addr = Section.getAddress(); + if (Addr != Val) + continue; + if (std::error_code EC = Section.getName(Name)) + report_error(errorCodeToError(EC), O->getFileName()); + Fmt << Name; + return; + } + + Fmt << format("0x%x", Val); + return; + } + + StringRef S; + bool isExtern = O->getPlainRelocationExternal(RE); + uint64_t Val = O->getPlainRelocationSymbolNum(RE); + + if (O->getAnyRelocationType(RE) == MachO::ARM64_RELOC_ADDEND) { + Fmt << format("0x%0" PRIx64, Val); + return; + } + + if (isExtern) { + symbol_iterator SI = O->symbol_begin(); + advance(SI, Val); + S = unwrapOrError(SI->getName(), FileName); + } else { + section_iterator SI = O->section_begin(); + // Adjust for the fact that sections are 1-indexed. + if (Val == 0) { + Fmt << "0 (?,?)"; + return; + } + uint32_t I = Val - 1; + while (I != 0 && SI != O->section_end()) { + --I; + advance(SI, 1); + } + if (SI == O->section_end()) + Fmt << Val << " (?,?)"; + else + SI->getName(S); + } + + Fmt << S; +} + +Error getMachORelocationValueString(const MachOObjectFile *Obj, + const RelocationRef &RelRef, + SmallVectorImpl<char> &Result) { + DataRefImpl Rel = RelRef.getRawDataRefImpl(); + MachO::any_relocation_info RE = Obj->getRelocation(Rel); + + unsigned Arch = Obj->getArch(); + + std::string FmtBuf; + raw_string_ostream Fmt(FmtBuf); + unsigned Type = Obj->getAnyRelocationType(RE); + bool IsPCRel = Obj->getAnyRelocationPCRel(RE); + + // Determine any addends that should be displayed with the relocation. + // These require decoding the relocation type, which is triple-specific. + + // X86_64 has entirely custom relocation types. + if (Arch == Triple::x86_64) { + switch (Type) { + case MachO::X86_64_RELOC_GOT_LOAD: + case MachO::X86_64_RELOC_GOT: { + printRelocationTargetName(Obj, RE, Fmt); + Fmt << "@GOT"; + if (IsPCRel) + Fmt << "PCREL"; + break; + } + case MachO::X86_64_RELOC_SUBTRACTOR: { + DataRefImpl RelNext = Rel; + Obj->moveRelocationNext(RelNext); + MachO::any_relocation_info RENext = Obj->getRelocation(RelNext); + + // X86_64_RELOC_SUBTRACTOR must be followed by a relocation of type + // X86_64_RELOC_UNSIGNED. + // NOTE: Scattered relocations don't exist on x86_64. + unsigned RType = Obj->getAnyRelocationType(RENext); + if (RType != MachO::X86_64_RELOC_UNSIGNED) + report_error(Obj->getFileName(), "Expected X86_64_RELOC_UNSIGNED after " + "X86_64_RELOC_SUBTRACTOR."); + + // The X86_64_RELOC_UNSIGNED contains the minuend symbol; + // X86_64_RELOC_SUBTRACTOR contains the subtrahend. + printRelocationTargetName(Obj, RENext, Fmt); + Fmt << "-"; + printRelocationTargetName(Obj, RE, Fmt); + break; + } + case MachO::X86_64_RELOC_TLV: + printRelocationTargetName(Obj, RE, Fmt); + Fmt << "@TLV"; + if (IsPCRel) + Fmt << "P"; + break; + case MachO::X86_64_RELOC_SIGNED_1: + printRelocationTargetName(Obj, RE, Fmt); + Fmt << "-1"; + break; + case MachO::X86_64_RELOC_SIGNED_2: + printRelocationTargetName(Obj, RE, Fmt); + Fmt << "-2"; + break; + case MachO::X86_64_RELOC_SIGNED_4: + printRelocationTargetName(Obj, RE, Fmt); + Fmt << "-4"; + break; + default: + printRelocationTargetName(Obj, RE, Fmt); + break; + } + // X86 and ARM share some relocation types in common. + } else if (Arch == Triple::x86 || Arch == Triple::arm || + Arch == Triple::ppc) { + // Generic relocation types... + switch (Type) { + case MachO::GENERIC_RELOC_PAIR: // prints no info + return Error::success(); + case MachO::GENERIC_RELOC_SECTDIFF: { + DataRefImpl RelNext = Rel; + Obj->moveRelocationNext(RelNext); + MachO::any_relocation_info RENext = Obj->getRelocation(RelNext); + + // X86 sect diff's must be followed by a relocation of type + // GENERIC_RELOC_PAIR. + unsigned RType = Obj->getAnyRelocationType(RENext); + + if (RType != MachO::GENERIC_RELOC_PAIR) + report_error(Obj->getFileName(), "Expected GENERIC_RELOC_PAIR after " + "GENERIC_RELOC_SECTDIFF."); + + printRelocationTargetName(Obj, RE, Fmt); + Fmt << "-"; + printRelocationTargetName(Obj, RENext, Fmt); + break; + } + } + + if (Arch == Triple::x86 || Arch == Triple::ppc) { + switch (Type) { + case MachO::GENERIC_RELOC_LOCAL_SECTDIFF: { + DataRefImpl RelNext = Rel; + Obj->moveRelocationNext(RelNext); + MachO::any_relocation_info RENext = Obj->getRelocation(RelNext); + + // X86 sect diff's must be followed by a relocation of type + // GENERIC_RELOC_PAIR. + unsigned RType = Obj->getAnyRelocationType(RENext); + if (RType != MachO::GENERIC_RELOC_PAIR) + report_error(Obj->getFileName(), "Expected GENERIC_RELOC_PAIR after " + "GENERIC_RELOC_LOCAL_SECTDIFF."); + + printRelocationTargetName(Obj, RE, Fmt); + Fmt << "-"; + printRelocationTargetName(Obj, RENext, Fmt); + break; + } + case MachO::GENERIC_RELOC_TLV: { + printRelocationTargetName(Obj, RE, Fmt); + Fmt << "@TLV"; + if (IsPCRel) + Fmt << "P"; + break; + } + default: + printRelocationTargetName(Obj, RE, Fmt); + } + } else { // ARM-specific relocations + switch (Type) { + case MachO::ARM_RELOC_HALF: + case MachO::ARM_RELOC_HALF_SECTDIFF: { + // Half relocations steal a bit from the length field to encode + // whether this is an upper16 or a lower16 relocation. + bool isUpper = (Obj->getAnyRelocationLength(RE) & 0x1) == 1; + + if (isUpper) + Fmt << ":upper16:("; + else + Fmt << ":lower16:("; + printRelocationTargetName(Obj, RE, Fmt); + + DataRefImpl RelNext = Rel; + Obj->moveRelocationNext(RelNext); + MachO::any_relocation_info RENext = Obj->getRelocation(RelNext); + + // ARM half relocs must be followed by a relocation of type + // ARM_RELOC_PAIR. + unsigned RType = Obj->getAnyRelocationType(RENext); + if (RType != MachO::ARM_RELOC_PAIR) + report_error(Obj->getFileName(), "Expected ARM_RELOC_PAIR after " + "ARM_RELOC_HALF"); + + // NOTE: The half of the target virtual address is stashed in the + // address field of the secondary relocation, but we can't reverse + // engineer the constant offset from it without decoding the movw/movt + // instruction to find the other half in its immediate field. + + // ARM_RELOC_HALF_SECTDIFF encodes the second section in the + // symbol/section pointer of the follow-on relocation. + if (Type == MachO::ARM_RELOC_HALF_SECTDIFF) { + Fmt << "-"; + printRelocationTargetName(Obj, RENext, Fmt); + } + + Fmt << ")"; + break; + } + default: { + printRelocationTargetName(Obj, RE, Fmt); + } + } + } + } else + printRelocationTargetName(Obj, RE, Fmt); + + Fmt.flush(); + Result.append(FmtBuf.begin(), FmtBuf.end()); + return Error::success(); +} + static void PrintIndirectSymbolTable(MachOObjectFile *O, bool verbose, uint32_t n, uint32_t count, uint32_t stride, uint64_t addr) { @@ -389,10 +692,7 @@ static void PrintIndirectSymbolTable(MachOObjectFile *O, bool verbose, if (indirect_symbol < Symtab.nsyms) { symbol_iterator Sym = O->getSymbolByIndex(indirect_symbol); SymbolRef Symbol = *Sym; - Expected<StringRef> SymName = Symbol.getName(); - if (!SymName) - report_error(O->getFileName(), SymName.takeError()); - outs() << *SymName; + outs() << unwrapOrError(Symbol.getName(), O->getFileName()); } else { outs() << "?"; } @@ -500,6 +800,7 @@ static void PrintRType(const uint64_t cputype, const unsigned r_type) { outs() << arm_r_types[r_type]; break; case MachO::CPU_TYPE_ARM64: + case MachO::CPU_TYPE_ARM64_32: outs() << arm64_r_types[r_type]; break; default: @@ -510,9 +811,8 @@ static void PrintRType(const uint64_t cputype, const unsigned r_type) { static void PrintRLength(const uint64_t cputype, const unsigned r_type, const unsigned r_length, const bool previous_arm_half){ if (cputype == MachO::CPU_TYPE_ARM && - (r_type == llvm::MachO::ARM_RELOC_HALF || - r_type == llvm::MachO::ARM_RELOC_HALF_SECTDIFF || - previous_arm_half == true)) { + (r_type == MachO::ARM_RELOC_HALF || + r_type == MachO::ARM_RELOC_HALF_SECTDIFF || previous_arm_half == true)) { if ((r_length & 0x1) == 0) outs() << "lo/"; else @@ -573,9 +873,8 @@ static void PrintRelocationEntries(const MachOObjectFile *O, if (verbose) { // scattered: address if ((cputype == MachO::CPU_TYPE_I386 && - r_type == llvm::MachO::GENERIC_RELOC_PAIR) || - (cputype == MachO::CPU_TYPE_ARM && - r_type == llvm::MachO::ARM_RELOC_PAIR)) + r_type == MachO::GENERIC_RELOC_PAIR) || + (cputype == MachO::CPU_TYPE_ARM && r_type == MachO::ARM_RELOC_PAIR)) outs() << " "; else outs() << format("%08x ", (unsigned int)r_address); @@ -597,29 +896,27 @@ static void PrintRelocationEntries(const MachOObjectFile *O, outs() << format("True 0x%08x", (unsigned int)r_value); if (previous_sectdiff == false) { if ((cputype == MachO::CPU_TYPE_ARM && - r_type == llvm::MachO::ARM_RELOC_PAIR)) + r_type == MachO::ARM_RELOC_PAIR)) outs() << format(" half = 0x%04x ", (unsigned int)r_address); - } - else if (cputype == MachO::CPU_TYPE_ARM && - sectdiff_r_type == llvm::MachO::ARM_RELOC_HALF_SECTDIFF) + } else if (cputype == MachO::CPU_TYPE_ARM && + sectdiff_r_type == MachO::ARM_RELOC_HALF_SECTDIFF) outs() << format(" other_half = 0x%04x ", (unsigned int)r_address); if ((cputype == MachO::CPU_TYPE_I386 && - (r_type == llvm::MachO::GENERIC_RELOC_SECTDIFF || - r_type == llvm::MachO::GENERIC_RELOC_LOCAL_SECTDIFF)) || + (r_type == MachO::GENERIC_RELOC_SECTDIFF || + r_type == MachO::GENERIC_RELOC_LOCAL_SECTDIFF)) || (cputype == MachO::CPU_TYPE_ARM && - (sectdiff_r_type == llvm::MachO::ARM_RELOC_SECTDIFF || - sectdiff_r_type == llvm::MachO::ARM_RELOC_LOCAL_SECTDIFF || - sectdiff_r_type == llvm::MachO::ARM_RELOC_HALF_SECTDIFF))) { - previous_sectdiff = true; - sectdiff_r_type = r_type; - } - else { + (sectdiff_r_type == MachO::ARM_RELOC_SECTDIFF || + sectdiff_r_type == MachO::ARM_RELOC_LOCAL_SECTDIFF || + sectdiff_r_type == MachO::ARM_RELOC_HALF_SECTDIFF))) { + previous_sectdiff = true; + sectdiff_r_type = r_type; + } else { previous_sectdiff = false; sectdiff_r_type = 0; } if (cputype == MachO::CPU_TYPE_ARM && - (r_type == llvm::MachO::ARM_RELOC_HALF || - r_type == llvm::MachO::ARM_RELOC_HALF_SECTDIFF)) + (r_type == MachO::ARM_RELOC_HALF || + r_type == MachO::ARM_RELOC_HALF_SECTDIFF)) previous_arm_half = true; else previous_arm_half = false; @@ -635,8 +932,7 @@ static void PrintRelocationEntries(const MachOObjectFile *O, else { if (verbose) { // plain: address - if (cputype == MachO::CPU_TYPE_ARM && - r_type == llvm::MachO::ARM_RELOC_PAIR) + if (cputype == MachO::CPU_TYPE_ARM && r_type == MachO::ARM_RELOC_PAIR) outs() << " "; else outs() << format("%08x ", (unsigned int)r_address); @@ -678,28 +974,27 @@ static void PrintRelocationEntries(const MachOObjectFile *O, outs() << "False "; // plain: symbolnum/value - if (cputype == MachO::CPU_TYPE_ARM && - r_type == llvm::MachO::ARM_RELOC_PAIR) + if (cputype == MachO::CPU_TYPE_ARM && r_type == MachO::ARM_RELOC_PAIR) outs() << format("other_half = 0x%04x\n", (unsigned int)r_address); - else if (cputype == MachO::CPU_TYPE_ARM64 && - r_type == llvm::MachO::ARM64_RELOC_ADDEND) + else if ((cputype == MachO::CPU_TYPE_ARM64 || + cputype == MachO::CPU_TYPE_ARM64_32) && + r_type == MachO::ARM64_RELOC_ADDEND) outs() << format("addend = 0x%06x\n", (unsigned int)r_symbolnum); else { outs() << format("%d ", r_symbolnum); - if (r_symbolnum == llvm::MachO::R_ABS) + if (r_symbolnum == MachO::R_ABS) outs() << "R_ABS\n"; else { // in this case, r_symbolnum is actually a 1-based section number uint32_t nsects = O->section_end()->getRawDataRefImpl().d.a; if (r_symbolnum > 0 && r_symbolnum <= nsects) { - llvm::object::DataRefImpl DRI; + object::DataRefImpl DRI; DRI.d.a = r_symbolnum-1; StringRef SegName = O->getSectionFinalSegmentName(DRI); - StringRef SectName; - if (O->getSectionName(DRI, SectName)) - outs() << "(?,?)\n"; + if (Expected<StringRef> NameOrErr = O->getSectionName(DRI)) + outs() << "(" << SegName << "," << *NameOrErr << ")\n"; else - outs() << "(" << SegName << "," << SectName << ")\n"; + outs() << "(?,?)\n"; } else { outs() << "(?,?)\n"; @@ -708,8 +1003,8 @@ static void PrintRelocationEntries(const MachOObjectFile *O, } } if (cputype == MachO::CPU_TYPE_ARM && - (r_type == llvm::MachO::ARM_RELOC_HALF || - r_type == llvm::MachO::ARM_RELOC_HALF_SECTDIFF)) + (r_type == MachO::ARM_RELOC_HALF || + r_type == MachO::ARM_RELOC_HALF_SECTDIFF)) previous_arm_half = true; else previous_arm_half = false; @@ -752,13 +1047,12 @@ static void PrintRelocations(const MachOObjectFile *O, const bool verbose) { DataRefImpl DRI; DRI.d.a = J; const StringRef SegName = O->getSectionFinalSegmentName(DRI); - StringRef SectName; - if (O->getSectionName(DRI, SectName)) + if (Expected<StringRef> NameOrErr = O->getSectionName(DRI)) + outs() << "Relocation information (" << SegName << "," << *NameOrErr + << format(") %u entries", Sec.nreloc); + else outs() << "Relocation information (" << SegName << ",?) " << format("%u entries", Sec.nreloc); - else - outs() << "Relocation information (" << SegName << "," - << SectName << format(") %u entries", Sec.nreloc); outs() << "\naddress pcrel length extern type scattered " "symbolnum/value\n"; PrintRelocationEntries(O, O->section_rel_begin(DRI), @@ -773,13 +1067,12 @@ static void PrintRelocations(const MachOObjectFile *O, const bool verbose) { DataRefImpl DRI; DRI.d.a = J; const StringRef SegName = O->getSectionFinalSegmentName(DRI); - StringRef SectName; - if (O->getSectionName(DRI, SectName)) + if (Expected<StringRef> NameOrErr = O->getSectionName(DRI)) + outs() << "Relocation information (" << SegName << "," << *NameOrErr + << format(") %u entries", Sec.nreloc); + else outs() << "Relocation information (" << SegName << ",?) " << format("%u entries", Sec.nreloc); - else - outs() << "Relocation information (" << SegName << "," - << SectName << format(") %u entries", Sec.nreloc); outs() << "\naddress pcrel length extern type scattered " "symbolnum/value\n"; PrintRelocationEntries(O, O->section_rel_begin(DRI), @@ -913,7 +1206,16 @@ static void PrintDylibs(MachOObjectFile *O, bool JustId) { outs() << " current version " << ((dl.dylib.current_version >> 16) & 0xffff) << "." << ((dl.dylib.current_version >> 8) & 0xff) << "." - << (dl.dylib.current_version & 0xff) << ")\n"; + << (dl.dylib.current_version & 0xff); + if (Load.C.cmd == MachO::LC_LOAD_WEAK_DYLIB) + outs() << ", weak"; + if (Load.C.cmd == MachO::LC_REEXPORT_DYLIB) + outs() << ", reexport"; + if (Load.C.cmd == MachO::LC_LOAD_UPWARD_DYLIB) + outs() << ", upward"; + if (Load.C.cmd == MachO::LC_LAZY_LOAD_DYLIB) + outs() << ", lazy"; + outs() << ")\n"; } } else { outs() << "\tBad offset (" << dl.dylib.name << ") for name of "; @@ -942,18 +1244,13 @@ typedef DenseMap<uint64_t, StringRef> SymbolAddressMap; static void CreateSymbolAddressMap(MachOObjectFile *O, SymbolAddressMap *AddrMap) { // Create a map of symbol addresses to symbol names. + const StringRef FileName = O->getFileName(); for (const SymbolRef &Symbol : O->symbols()) { - Expected<SymbolRef::Type> STOrErr = Symbol.getType(); - if (!STOrErr) - report_error(O->getFileName(), STOrErr.takeError()); - SymbolRef::Type ST = *STOrErr; + SymbolRef::Type ST = unwrapOrError(Symbol.getType(), FileName); if (ST == SymbolRef::ST_Function || ST == SymbolRef::ST_Data || ST == SymbolRef::ST_Other) { uint64_t Address = Symbol.getValue(); - Expected<StringRef> SymNameOrErr = Symbol.getName(); - if (!SymNameOrErr) - report_error(O->getFileName(), SymNameOrErr.takeError()); - StringRef SymName = *SymNameOrErr; + StringRef SymName = unwrapOrError(Symbol.getName(), FileName); if (!SymName.startswith(".objc")) (*AddrMap)[Address] = SymName; } @@ -1186,10 +1483,8 @@ static void DumpLiteralPointerSection(MachOObjectFile *O, }); if (Reloc != Relocs.end()) { symbol_iterator RelocSym = Reloc->second; - Expected<StringRef> SymName = RelocSym->getName(); - if (!SymName) - report_error(O->getFileName(), SymName.takeError()); - outs() << "external relocation entry for symbol:" << *SymName << "\n"; + StringRef SymName = unwrapOrError(RelocSym->getName(), O->getFileName()); + outs() << "external relocation entry for symbol:" << SymName << "\n"; continue; } @@ -1220,8 +1515,8 @@ static void DumpLiteralPointerSection(MachOObjectFile *O, section_type = Sec.flags & MachO::SECTION_TYPE; } - StringRef BytesStr; - Sect->getContents(BytesStr); + StringRef BytesStr = unwrapOrError(Sect->getContents(), O->getFileName()); + const char *Contents = reinterpret_cast<const char *>(BytesStr.data()); switch (section_type) { @@ -1333,10 +1628,7 @@ static void DumpInitTermPointerSection(MachOObjectFile *O, }); if (Reloc != Relocs.end()) { symbol_iterator RelocSym = Reloc->second; - Expected<StringRef> SymName = RelocSym->getName(); - if (!SymName) - report_error(O->getFileName(), SymName.takeError()); - outs() << " " << *SymName; + outs() << " " << unwrapOrError(RelocSym->getName(), O->getFileName()); } else { SymbolName = GuessSymbolName(p, AddrMap); if (SymbolName) @@ -1438,8 +1730,8 @@ static void DumpSectionContents(StringRef Filename, MachOObjectFile *O, } uint32_t section_type = section_flags & MachO::SECTION_TYPE; - StringRef BytesStr; - Section.getContents(BytesStr); + StringRef BytesStr = + unwrapOrError(Section.getContents(), O->getFileName()); const char *sect = reinterpret_cast<const char *>(BytesStr.data()); uint32_t sect_size = BytesStr.size(); uint64_t sect_addr = Section.getAddress(); @@ -1523,8 +1815,8 @@ static void DumpInfoPlistSectionContents(StringRef Filename, if (SegName == "__TEXT" && SectName == "__info_plist") { if (!NoLeadingHeaders) outs() << "Contents of (" << SegName << "," << SectName << ") section\n"; - StringRef BytesStr; - Section.getContents(BytesStr); + StringRef BytesStr = + unwrapOrError(Section.getContents(), O->getFileName()); const char *sect = reinterpret_cast<const char *>(BytesStr.data()); outs() << format("%.*s", BytesStr.size(), sect) << "\n"; return; @@ -1609,8 +1901,8 @@ static void ProcessMachO(StringRef Name, MachOObjectFile *MachOOF, // the error message. if (Disassemble || IndirectSymbols || !FilterSections.empty() || UnwindInfo) if (Error Err = MachOOF->checkSymbolTable()) - report_error(ArchiveName, FileName, std::move(Err), ArchitectureName); - + report_error(std::move(Err), ArchiveName, FileName, ArchitectureName); + if (DisassembleAll) { for (const SectionRef &Section : MachOOF->sections()) { StringRef SectName; @@ -1774,6 +2066,21 @@ static void printCPUType(uint32_t cputype, uint32_t cpusubtype) { outs() << " cputype CPU_TYPE_ARM64\n"; outs() << " cpusubtype CPU_SUBTYPE_ARM64_ALL\n"; break; + case MachO::CPU_SUBTYPE_ARM64E: + outs() << " cputype CPU_TYPE_ARM64\n"; + outs() << " cpusubtype CPU_SUBTYPE_ARM64E\n"; + break; + default: + printUnknownCPUType(cputype, cpusubtype); + break; + } + break; + case MachO::CPU_TYPE_ARM64_32: + switch (cpusubtype & ~MachO::CPU_SUBTYPE_MASK) { + case MachO::CPU_SUBTYPE_ARM64_32_V8: + outs() << " cputype CPU_TYPE_ARM64_32\n"; + outs() << " cpusubtype CPU_SUBTYPE_ARM64_32_V8\n"; + break; default: printUnknownCPUType(cputype, cpusubtype); break; @@ -1862,10 +2169,8 @@ static void printArchiveChild(StringRef Filename, const Archive::Child &C, StringRef ArchitectureName = StringRef()) { if (print_offset) outs() << C.getChildOffset() << "\t"; - Expected<sys::fs::perms> ModeOrErr = C.getAccessMode(); - if (!ModeOrErr) - report_error(Filename, C, ModeOrErr.takeError(), ArchitectureName); - sys::fs::perms Mode = ModeOrErr.get(); + sys::fs::perms Mode = + unwrapOrError(C.getAccessMode(), Filename, C, ArchitectureName); if (verbose) { // FIXME: this first dash, "-", is for (Mode & S_IFMT) == S_IFREG. // But there is nothing in sys::fs::perms for S_IFMT or S_IFREG. @@ -1883,20 +2188,11 @@ static void printArchiveChild(StringRef Filename, const Archive::Child &C, outs() << format("0%o ", Mode); } - Expected<unsigned> UIDOrErr = C.getUID(); - if (!UIDOrErr) - report_error(Filename, C, UIDOrErr.takeError(), ArchitectureName); - unsigned UID = UIDOrErr.get(); - outs() << format("%3d/", UID); - Expected<unsigned> GIDOrErr = C.getGID(); - if (!GIDOrErr) - report_error(Filename, C, GIDOrErr.takeError(), ArchitectureName); - unsigned GID = GIDOrErr.get(); - outs() << format("%-3d ", GID); - Expected<uint64_t> Size = C.getRawSize(); - if (!Size) - report_error(Filename, C, Size.takeError(), ArchitectureName); - outs() << format("%5" PRId64, Size.get()) << " "; + outs() << format( + "%3d/%-3d %5" PRId64 " ", + unwrapOrError(C.getUID(), Filename, C, ArchitectureName), + unwrapOrError(C.getGID(), Filename, C, ArchitectureName), + unwrapOrError(C.getRawSize(), Filename, C, ArchitectureName)); StringRef RawLastModified = C.getRawLastModified(); if (verbose) { @@ -1919,21 +2215,15 @@ static void printArchiveChild(StringRef Filename, const Archive::Child &C, Expected<StringRef> NameOrErr = C.getName(); if (!NameOrErr) { consumeError(NameOrErr.takeError()); - Expected<StringRef> NameOrErr = C.getRawName(); - if (!NameOrErr) - report_error(Filename, C, NameOrErr.takeError(), ArchitectureName); - StringRef RawName = NameOrErr.get(); - outs() << RawName << "\n"; + outs() << unwrapOrError(C.getRawName(), Filename, C, ArchitectureName) + << "\n"; } else { StringRef Name = NameOrErr.get(); outs() << Name << "\n"; } } else { - Expected<StringRef> NameOrErr = C.getRawName(); - if (!NameOrErr) - report_error(Filename, C, NameOrErr.takeError(), ArchitectureName); - StringRef RawName = NameOrErr.get(); - outs() << RawName << "\n"; + outs() << unwrapOrError(C.getRawName(), Filename, C, ArchitectureName) + << "\n"; } } @@ -1941,12 +2231,11 @@ static void printArchiveHeaders(StringRef Filename, Archive *A, bool verbose, bool print_offset, StringRef ArchitectureName = StringRef()) { Error Err = Error::success(); - ; for (const auto &C : A->children(Err, false)) printArchiveChild(Filename, C, verbose, print_offset, ArchitectureName); if (Err) - report_error(StringRef(), Filename, std::move(Err), ArchitectureName); + report_error(std::move(Err), StringRef(), Filename, ArchitectureName); } static bool ValidateArchFlags() { @@ -1970,15 +2259,15 @@ static bool ValidateArchFlags() { // -arch flags selecting just those slices as specified by them and also parses // archive files. Then for each individual Mach-O file ProcessMachO() is // called to process the file based on the command line options. -void llvm::parseInputMachO(StringRef Filename) { +void parseInputMachO(StringRef Filename) { if (!ValidateArchFlags()) return; // Attempt to open the binary. Expected<OwningBinary<Binary>> BinaryOrErr = createBinary(Filename); if (!BinaryOrErr) { - if (auto E = isNotObjectErrorInvalidFileType(BinaryOrErr.takeError())) - report_error(Filename, std::move(E)); + if (Error E = isNotObjectErrorInvalidFileType(BinaryOrErr.takeError())) + report_error(std::move(E), Filename); else outs() << Filename << ": is not an object file\n"; return; @@ -1994,8 +2283,8 @@ void llvm::parseInputMachO(StringRef Filename) { for (auto &C : A->children(Err)) { Expected<std::unique_ptr<Binary>> ChildOrErr = C.getAsBinary(); if (!ChildOrErr) { - if (auto E = isNotObjectErrorInvalidFileType(ChildOrErr.takeError())) - report_error(Filename, C, std::move(E)); + if (Error E = isNotObjectErrorInvalidFileType(ChildOrErr.takeError())) + report_error(std::move(E), Filename, C); continue; } if (MachOObjectFile *O = dyn_cast<MachOObjectFile>(&*ChildOrErr.get())) { @@ -2005,7 +2294,7 @@ void llvm::parseInputMachO(StringRef Filename) { } } if (Err) - report_error(Filename, std::move(Err)); + report_error(std::move(Err), Filename); return; } if (MachOUniversalBinary *UB = dyn_cast<MachOUniversalBinary>(&Bin)) { @@ -2026,7 +2315,7 @@ void llvm::parseInputMachO(StringRef Filename) { llvm_unreachable("Input object can't be invalid at this point"); } -void llvm::parseInputMachO(MachOUniversalBinary *UB) { +void parseInputMachO(MachOUniversalBinary *UB) { if (!ValidateArchFlags()) return; @@ -2055,13 +2344,12 @@ void llvm::parseInputMachO(MachOUniversalBinary *UB) { ObjectFile &O = *ObjOrErr.get(); if (MachOObjectFile *MachOOF = dyn_cast<MachOObjectFile>(&O)) ProcessMachO(Filename, MachOOF, "", ArchitectureName); - } else if (auto E = isNotObjectErrorInvalidFileType( - ObjOrErr.takeError())) { - report_error(Filename, StringRef(), std::move(E), - ArchitectureName); + } else if (Error E = isNotObjectErrorInvalidFileType( + ObjOrErr.takeError())) { + report_error(std::move(E), Filename, StringRef(), ArchitectureName); continue; } else if (Expected<std::unique_ptr<Archive>> AOrErr = - I->getAsArchive()) { + I->getAsArchive()) { std::unique_ptr<Archive> &A = *AOrErr; outs() << "Archive : " << Filename; if (!ArchitectureName.empty()) @@ -2074,8 +2362,8 @@ void llvm::parseInputMachO(MachOUniversalBinary *UB) { for (auto &C : A->children(Err)) { Expected<std::unique_ptr<Binary>> ChildOrErr = C.getAsBinary(); if (!ChildOrErr) { - if (auto E = isNotObjectErrorInvalidFileType(ChildOrErr.takeError())) - report_error(Filename, C, std::move(E), ArchitectureName); + if (Error E = isNotObjectErrorInvalidFileType(ChildOrErr.takeError())) + report_error(std::move(E), Filename, C, ArchitectureName); continue; } if (MachOObjectFile *O = @@ -2083,7 +2371,7 @@ void llvm::parseInputMachO(MachOUniversalBinary *UB) { ProcessMachO(Filename, O, O->getFileName(), ArchitectureName); } if (Err) - report_error(Filename, std::move(Err)); + report_error(std::move(Err), Filename); } else { consumeError(AOrErr.takeError()); error("Mach-O universal file: " + Filename + " for " + @@ -2116,11 +2404,11 @@ void llvm::parseInputMachO(MachOUniversalBinary *UB) { ObjectFile &O = *ObjOrErr.get(); if (MachOObjectFile *MachOOF = dyn_cast<MachOObjectFile>(&O)) ProcessMachO(Filename, MachOOF); - } else if (auto E = isNotObjectErrorInvalidFileType( - ObjOrErr.takeError())) { - report_error(Filename, std::move(E)); + } else if (Error E = + isNotObjectErrorInvalidFileType(ObjOrErr.takeError())) { + report_error(std::move(E), Filename); } else if (Expected<std::unique_ptr<Archive>> AOrErr = - I->getAsArchive()) { + I->getAsArchive()) { std::unique_ptr<Archive> &A = *AOrErr; outs() << "Archive : " << Filename << "\n"; if (ArchiveHeaders) @@ -2130,8 +2418,9 @@ void llvm::parseInputMachO(MachOUniversalBinary *UB) { for (auto &C : A->children(Err)) { Expected<std::unique_ptr<Binary>> ChildOrErr = C.getAsBinary(); if (!ChildOrErr) { - if (auto E = isNotObjectErrorInvalidFileType(ChildOrErr.takeError())) - report_error(Filename, C, std::move(E)); + if (Error E = + isNotObjectErrorInvalidFileType(ChildOrErr.takeError())) + report_error(std::move(E), Filename, C); continue; } if (MachOObjectFile *O = @@ -2139,7 +2428,7 @@ void llvm::parseInputMachO(MachOUniversalBinary *UB) { ProcessMachO(Filename, O, O->getFileName()); } if (Err) - report_error(Filename, std::move(Err)); + report_error(std::move(Err), Filename); } else { consumeError(AOrErr.takeError()); error("Mach-O universal file: " + Filename + " for architecture " + @@ -2164,11 +2453,10 @@ void llvm::parseInputMachO(MachOUniversalBinary *UB) { ObjectFile &Obj = *ObjOrErr.get(); if (MachOObjectFile *MachOOF = dyn_cast<MachOObjectFile>(&Obj)) ProcessMachO(Filename, MachOOF, "", ArchitectureName); - } else if (auto E = isNotObjectErrorInvalidFileType( - ObjOrErr.takeError())) { - report_error(StringRef(), Filename, std::move(E), ArchitectureName); - } else if (Expected<std::unique_ptr<Archive>> AOrErr = - I->getAsArchive()) { + } else if (Error E = + isNotObjectErrorInvalidFileType(ObjOrErr.takeError())) { + report_error(std::move(E), StringRef(), Filename, ArchitectureName); + } else if (Expected<std::unique_ptr<Archive>> AOrErr = I->getAsArchive()) { std::unique_ptr<Archive> &A = *AOrErr; outs() << "Archive : " << Filename; if (!ArchitectureName.empty()) @@ -2181,8 +2469,8 @@ void llvm::parseInputMachO(MachOUniversalBinary *UB) { for (auto &C : A->children(Err)) { Expected<std::unique_ptr<Binary>> ChildOrErr = C.getAsBinary(); if (!ChildOrErr) { - if (auto E = isNotObjectErrorInvalidFileType(ChildOrErr.takeError())) - report_error(Filename, C, std::move(E), ArchitectureName); + if (Error E = isNotObjectErrorInvalidFileType(ChildOrErr.takeError())) + report_error(std::move(E), Filename, C, ArchitectureName); continue; } if (MachOObjectFile *O = @@ -2193,7 +2481,7 @@ void llvm::parseInputMachO(MachOUniversalBinary *UB) { } } if (Err) - report_error(Filename, std::move(Err)); + report_error(std::move(Err), Filename); } else { consumeError(AOrErr.takeError()); error("Mach-O universal file: " + Filename + " for architecture " + @@ -2308,12 +2596,9 @@ static int SymbolizerGetOpInfo(void *DisInfo, uint64_t Pc, uint64_t Offset, } } if (reloc_found && isExtern) { - Expected<StringRef> SymName = Symbol.getName(); - if (!SymName) - report_error(info->O->getFileName(), SymName.takeError()); - const char *name = SymName->data(); op_info->AddSymbol.Present = 1; - op_info->AddSymbol.Name = name; + op_info->AddSymbol.Name = + unwrapOrError(Symbol.getName(), info->O->getFileName()).data(); // For i386 extern relocation entries the value in the instruction is // the offset from the symbol, and value is already set in op_info->Value. return 1; @@ -2372,10 +2657,8 @@ static int SymbolizerGetOpInfo(void *DisInfo, uint64_t Pc, uint64_t Offset, // is the offset from the external symbol. if (info->O->getAnyRelocationPCRel(RE)) op_info->Value -= Pc + Offset + Size; - Expected<StringRef> SymName = Symbol.getName(); - if (!SymName) - report_error(info->O->getFileName(), SymName.takeError()); - const char *name = SymName->data(); + const char *name = + unwrapOrError(Symbol.getName(), info->O->getFileName()).data(); op_info->AddSymbol.Present = 1; op_info->AddSymbol.Name = name; return 1; @@ -2412,10 +2695,8 @@ static int SymbolizerGetOpInfo(void *DisInfo, uint64_t Pc, uint64_t Offset, // is the offset from the external symbol. if (info->O->getAnyRelocationPCRel(RE)) op_info->Value -= Pc + Offset + Size; - Expected<StringRef> SymName = Symbol.getName(); - if (!SymName) - report_error(info->O->getFileName(), SymName.takeError()); - const char *name = SymName->data(); + const char *name = + unwrapOrError(Symbol.getName(), info->O->getFileName()).data(); unsigned Type = info->O->getAnyRelocationType(RE); if (Type == MachO::X86_64_RELOC_SUBTRACTOR) { DataRefImpl RelNext = Rel; @@ -2429,10 +2710,7 @@ static int SymbolizerGetOpInfo(void *DisInfo, uint64_t Pc, uint64_t Offset, op_info->SubtractSymbol.Name = name; symbol_iterator RelocSymNext = info->O->getSymbolByIndex(SymbolNum); Symbol = *RelocSymNext; - Expected<StringRef> SymNameNext = Symbol.getName(); - if (!SymNameNext) - report_error(info->O->getFileName(), SymNameNext.takeError()); - name = SymNameNext->data(); + name = unwrapOrError(Symbol.getName(), info->O->getFileName()).data(); } } // TODO: add the VariantKinds to op_info->VariantKind for relocation types @@ -2501,10 +2779,8 @@ static int SymbolizerGetOpInfo(void *DisInfo, uint64_t Pc, uint64_t Offset, } if (isExtern) { - Expected<StringRef> SymName = Symbol.getName(); - if (!SymName) - report_error(info->O->getFileName(), SymName.takeError()); - const char *name = SymName->data(); + const char *name = + unwrapOrError(Symbol.getName(), info->O->getFileName()).data(); op_info->AddSymbol.Present = 1; op_info->AddSymbol.Name = name; switch (r_type) { @@ -2620,10 +2896,9 @@ static int SymbolizerGetOpInfo(void *DisInfo, uint64_t Pc, uint64_t Offset, // NOTE: Scattered relocations don't exist on arm64. if (!info->O->getPlainRelocationExternal(RE)) return 0; - Expected<StringRef> SymName = Reloc->getSymbol()->getName(); - if (!SymName) - report_error(info->O->getFileName(), SymName.takeError()); - const char *name = SymName->data(); + const char *name = + unwrapOrError(Reloc->getSymbol()->getName(), info->O->getFileName()) + .data(); op_info->AddSymbol.Present = 1; op_info->AddSymbol.Name = name; @@ -2749,12 +3024,8 @@ static const char *GuessIndirectSymbol(uint64_t ReferenceValue, info->O->getIndirectSymbolTableEntry(Dysymtab, index); if (indirect_symbol < Symtab.nsyms) { symbol_iterator Sym = info->O->getSymbolByIndex(indirect_symbol); - SymbolRef Symbol = *Sym; - Expected<StringRef> SymName = Symbol.getName(); - if (!SymName) - report_error(info->O->getFileName(), SymName.takeError()); - const char *name = SymName->data(); - return name; + return unwrapOrError(Sym->getName(), info->O->getFileName()) + .data(); } } } @@ -2784,12 +3055,8 @@ static const char *GuessIndirectSymbol(uint64_t ReferenceValue, info->O->getIndirectSymbolTableEntry(Dysymtab, index); if (indirect_symbol < Symtab.nsyms) { symbol_iterator Sym = info->O->getSymbolByIndex(indirect_symbol); - SymbolRef Symbol = *Sym; - Expected<StringRef> SymName = Symbol.getName(); - if (!SymName) - report_error(info->O->getFileName(), SymName.takeError()); - const char *name = SymName->data(); - return name; + return unwrapOrError(Sym->getName(), info->O->getFileName()) + .data(); } } } @@ -2960,8 +3227,8 @@ static const char *get_pointer_64(uint64_t Address, uint32_t &offset, S = (*(info->Sections))[SectIdx]; offset = Address - SectAddress; left = SectSize - offset; - StringRef SectContents; - ((*(info->Sections))[SectIdx]).getContents(SectContents); + StringRef SectContents = unwrapOrError( + ((*(info->Sections))[SectIdx]).getContents(), info->O->getFileName()); return SectContents.data() + offset; } } @@ -3015,10 +3282,7 @@ static const char *get_symbol_64(uint32_t sect_offset, SectionRef S, const char *SymbolName = nullptr; if (reloc_found && isExtern) { n_value = Symbol.getValue(); - Expected<StringRef> NameOrError = Symbol.getName(); - if (!NameOrError) - report_error(info->O->getFileName(), NameOrError.takeError()); - StringRef Name = *NameOrError; + StringRef Name = unwrapOrError(Symbol.getName(), info->O->getFileName()); if (!Name.empty()) { SymbolName = Name.data(); return SymbolName; @@ -3767,8 +4031,7 @@ walk_pointer_list_64(const char *listname, const SectionRef S, StringRef SegName = O->getSectionFinalSegmentName(Ref); outs() << "Contents of (" << SegName << "," << SectName << ") section\n"; - StringRef BytesStr; - S.getContents(BytesStr); + StringRef BytesStr = unwrapOrError(S.getContents(), O->getFileName()); const char *Contents = reinterpret_cast<const char *>(BytesStr.data()); for (uint32_t i = 0; i < S.getSize(); i += sizeof(uint64_t)) { @@ -3818,8 +4081,7 @@ walk_pointer_list_32(const char *listname, const SectionRef S, StringRef SegName = O->getSectionFinalSegmentName(Ref); outs() << "Contents of (" << SegName << "," << SectName << ") section\n"; - StringRef BytesStr; - S.getContents(BytesStr); + StringRef BytesStr = unwrapOrError(S.getContents(), O->getFileName()); const char *Contents = reinterpret_cast<const char *>(BytesStr.data()); for (uint32_t i = 0; i < S.getSize(); i += sizeof(uint32_t)) { @@ -6970,32 +7232,78 @@ static void DisassembleMachO(StringRef Filename, MachOObjectFile *MachOOF, raw_ostream &DebugOut = nulls(); #endif + // Try to find debug info and set up the DIContext for it. std::unique_ptr<DIContext> diContext; - ObjectFile *DbgObj = MachOOF; + std::unique_ptr<Binary> DSYMBinary; std::unique_ptr<MemoryBuffer> DSYMBuf; - // Try to find debug info and set up the DIContext for it. if (UseDbg) { + ObjectFile *DbgObj = MachOOF; + // A separate DSym file path was specified, parse it as a macho file, // get the sections and supply it to the section name parsing machinery. if (!DSYMFile.empty()) { ErrorOr<std::unique_ptr<MemoryBuffer>> BufOrErr = MemoryBuffer::getFileOrSTDIN(DSYMFile); if (std::error_code EC = BufOrErr.getError()) { - report_error(DSYMFile, errorCodeToError(EC)); + report_error(errorCodeToError(EC), DSYMFile); return; } - Expected<std::unique_ptr<MachOObjectFile>> DbgObjCheck = - ObjectFile::createMachOObjectFile(BufOrErr.get()->getMemBufferRef()); + // We need to keep the file alive, because we're replacing DbgObj with it. + DSYMBuf = std::move(BufOrErr.get()); - if (Error E = DbgObjCheck.takeError()) { - report_error(DSYMFile, std::move(E)); + Expected<std::unique_ptr<Binary>> BinaryOrErr = + createBinary(DSYMBuf.get()->getMemBufferRef()); + if (!BinaryOrErr) { + report_error(BinaryOrErr.takeError(), DSYMFile); return; } - DbgObj = DbgObjCheck.get().release(); - // We need to keep the file alive, because we're replacing DbgObj with it. - DSYMBuf = std::move(BufOrErr.get()); + // We need to keep the Binary elive with the buffer + DSYMBinary = std::move(BinaryOrErr.get()); + + if (ObjectFile *O = dyn_cast<ObjectFile>(DSYMBinary.get())) { + // this is a Mach-O object file, use it + if (MachOObjectFile *MachDSYM = dyn_cast<MachOObjectFile>(&*O)) { + DbgObj = MachDSYM; + } + else { + WithColor::error(errs(), "llvm-objdump") + << DSYMFile << " is not a Mach-O file type.\n"; + return; + } + } + else if (auto UB = dyn_cast<MachOUniversalBinary>(DSYMBinary.get())){ + // this is a Universal Binary, find a Mach-O for this architecture + uint32_t CPUType, CPUSubType; + const char *ArchFlag; + if (MachOOF->is64Bit()) { + const MachO::mach_header_64 H_64 = MachOOF->getHeader64(); + CPUType = H_64.cputype; + CPUSubType = H_64.cpusubtype; + } else { + const MachO::mach_header H = MachOOF->getHeader(); + CPUType = H.cputype; + CPUSubType = H.cpusubtype; + } + Triple T = MachOObjectFile::getArchTriple(CPUType, CPUSubType, nullptr, + &ArchFlag); + Expected<std::unique_ptr<MachOObjectFile>> MachDSYM = + UB->getObjectForArch(ArchFlag); + if (!MachDSYM) { + report_error(MachDSYM.takeError(), DSYMFile); + return; + } + + // We need to keep the Binary elive with the buffer + DbgObj = &*MachDSYM.get(); + DSYMBinary = std::move(*MachDSYM); + } + else { + WithColor::error(errs(), "llvm-objdump") + << DSYMFile << " is not a Mach-O or Universal file type.\n"; + return; + } } // Setup the DIContext @@ -7016,10 +7324,9 @@ static void DisassembleMachO(StringRef Filename, MachOObjectFile *MachOOF, if (SegmentName != DisSegName) continue; - StringRef BytesStr; - Sections[SectIdx].getContents(BytesStr); - ArrayRef<uint8_t> Bytes(reinterpret_cast<const uint8_t *>(BytesStr.data()), - BytesStr.size()); + StringRef BytesStr = + unwrapOrError(Sections[SectIdx].getContents(), Filename); + ArrayRef<uint8_t> Bytes = arrayRefFromStringRef(BytesStr); uint64_t SectAddress = Sections[SectIdx].getAddress(); bool symbolTableWorked = false; @@ -7029,17 +7336,13 @@ static void DisassembleMachO(StringRef Filename, MachOObjectFile *MachOOF, SymbolAddressMap AddrMap; bool DisSymNameFound = false; for (const SymbolRef &Symbol : MachOOF->symbols()) { - Expected<SymbolRef::Type> STOrErr = Symbol.getType(); - if (!STOrErr) - report_error(MachOOF->getFileName(), STOrErr.takeError()); - SymbolRef::Type ST = *STOrErr; + SymbolRef::Type ST = + unwrapOrError(Symbol.getType(), MachOOF->getFileName()); if (ST == SymbolRef::ST_Function || ST == SymbolRef::ST_Data || ST == SymbolRef::ST_Other) { uint64_t Address = Symbol.getValue(); - Expected<StringRef> SymNameOrErr = Symbol.getName(); - if (!SymNameOrErr) - report_error(MachOOF->getFileName(), SymNameOrErr.takeError()); - StringRef SymName = *SymNameOrErr; + StringRef SymName = + unwrapOrError(Symbol.getName(), MachOOF->getFileName()); AddrMap[Address] = SymName; if (!DisSymName.empty() && DisSymName == SymName) DisSymNameFound = true; @@ -7076,15 +7379,10 @@ static void DisassembleMachO(StringRef Filename, MachOObjectFile *MachOOF, // Disassemble symbol by symbol. for (unsigned SymIdx = 0; SymIdx != Symbols.size(); SymIdx++) { - Expected<StringRef> SymNameOrErr = Symbols[SymIdx].getName(); - if (!SymNameOrErr) - report_error(MachOOF->getFileName(), SymNameOrErr.takeError()); - StringRef SymName = *SymNameOrErr; - - Expected<SymbolRef::Type> STOrErr = Symbols[SymIdx].getType(); - if (!STOrErr) - report_error(MachOOF->getFileName(), STOrErr.takeError()); - SymbolRef::Type ST = *STOrErr; + StringRef SymName = + unwrapOrError(Symbols[SymIdx].getName(), MachOOF->getFileName()); + SymbolRef::Type ST = + unwrapOrError(Symbols[SymIdx].getType(), MachOOF->getFileName()); if (ST != SymbolRef::ST_Function && ST != SymbolRef::ST_Data) continue; @@ -7137,10 +7435,8 @@ static void DisassembleMachO(StringRef Filename, MachOObjectFile *MachOOF, uint64_t NextSym = 0; uint64_t NextSymIdx = SymIdx + 1; while (Symbols.size() > NextSymIdx) { - Expected<SymbolRef::Type> STOrErr = Symbols[NextSymIdx].getType(); - if (!STOrErr) - report_error(MachOOF->getFileName(), STOrErr.takeError()); - SymbolRef::Type NextSymType = *STOrErr; + SymbolRef::Type NextSymType = unwrapOrError( + Symbols[NextSymIdx].getType(), MachOOF->getFileName()); if (NextSymType == SymbolRef::ST_Function) { containsNextSym = Sections[SectIdx].containsSymbol(Symbols[NextSymIdx]); @@ -7243,7 +7539,7 @@ static void DisassembleMachO(StringRef Filename, MachOObjectFile *MachOOF, // Print debug info. if (diContext) { - DILineInfo dli = diContext->getLineInfoForAddress(PC); + DILineInfo dli = diContext->getLineInfoForAddress({PC, SectIdx}); // Print valid line info if it changed. if (dli != lastLine && dli.Line != 0) outs() << "\t## " << dli.FileName << ':' << dli.Line << ':' @@ -7415,10 +7711,7 @@ static void findUnwindRelocNameAddend(const MachOObjectFile *Obj, const RelocationRef &Reloc, uint64_t Addr, StringRef &Name, uint64_t &Addend) { if (Reloc.getSymbol() != Obj->symbol_end()) { - Expected<StringRef> NameOrErr = Reloc.getSymbol()->getName(); - if (!NameOrErr) - report_error(Obj->getFileName(), NameOrErr.takeError()); - Name = *NameOrErr; + Name = unwrapOrError(Reloc.getSymbol()->getName(), Obj->getFileName()); Addend = Addr; return; } @@ -7440,16 +7733,11 @@ static void findUnwindRelocNameAddend(const MachOObjectFile *Obj, // Go back one so that SymbolAddress <= Addr. --Sym; - auto SectOrErr = Sym->second.getSection(); - if (!SectOrErr) - report_error(Obj->getFileName(), SectOrErr.takeError()); - section_iterator SymSection = *SectOrErr; + section_iterator SymSection = + unwrapOrError(Sym->second.getSection(), Obj->getFileName()); if (RelocSection == *SymSection) { // There's a valid symbol in the same section before this reference. - Expected<StringRef> NameOrErr = Sym->second.getName(); - if (!NameOrErr) - report_error(Obj->getFileName(), NameOrErr.takeError()); - Name = *NameOrErr; + Name = unwrapOrError(Sym->second.getName(), Obj->getFileName()); Addend = Addr - Sym->first; return; } @@ -7490,9 +7778,8 @@ printMachOCompactUnwindSection(const MachOObjectFile *Obj, uint32_t PointerSize = Is64 ? sizeof(uint64_t) : sizeof(uint32_t); uint32_t EntrySize = 3 * PointerSize + 2 * sizeof(uint32_t); - StringRef Contents; - CompactUnwind.getContents(Contents); - + StringRef Contents = + unwrapOrError(CompactUnwind.getContents(), Obj->getFileName()); SmallVector<CompactUnwindEntry, 4> CompactUnwinds; // First populate the initial raw offsets, encodings and so on from the entry. @@ -7633,8 +7920,8 @@ static void printMachOUnwindInfoSection(const MachOObjectFile *Obj, outs() << "Contents of __unwind_info section:\n"; - StringRef Contents; - UnwindInfo.getContents(Contents); + StringRef Contents = + unwrapOrError(UnwindInfo.getContents(), Obj->getFileName()); ptrdiff_t Pos = 0; //===---------------------------------- @@ -7801,7 +8088,7 @@ static void printMachOUnwindInfoSection(const MachOObjectFile *Obj, } } -void llvm::printMachOUnwindInfo(const MachOObjectFile *Obj) { +void printMachOUnwindInfo(const MachOObjectFile *Obj) { std::map<uint64_t, SymbolRef> Symbols; for (const SymbolRef &SymRef : Obj->symbols()) { // Discard any undefined or absolute symbols. They're not going to take part @@ -7917,6 +8204,20 @@ static void PrintMachHeader(uint32_t magic, uint32_t cputype, case MachO::CPU_SUBTYPE_ARM64_ALL: outs() << " ALL"; break; + case MachO::CPU_SUBTYPE_ARM64E: + outs() << " E"; + break; + default: + outs() << format(" %10d", cpusubtype & ~MachO::CPU_SUBTYPE_MASK); + break; + } + break; + case MachO::CPU_TYPE_ARM64_32: + outs() << " ARM64_32"; + switch (cpusubtype & ~MachO::CPU_SUBTYPE_MASK) { + case MachO::CPU_SUBTYPE_ARM64_32_V8: + outs() << " V8"; + break; default: outs() << format(" %10d", cpusubtype & ~MachO::CPU_SUBTYPE_MASK); break; @@ -9485,7 +9786,8 @@ static void PrintThreadCommand(MachO::thread_command t, const char *Ptr, begin += count * sizeof(uint32_t); } } - } else if (cputype == MachO::CPU_TYPE_ARM64) { + } else if (cputype == MachO::CPU_TYPE_ARM64 || + cputype == MachO::CPU_TYPE_ARM64_32) { while (begin < end) { if (end - begin > (ptrdiff_t)sizeof(uint32_t)) { memcpy((char *)&flavor, begin, sizeof(uint32_t)); @@ -9790,12 +10092,12 @@ static void PrintMachHeader(const MachOObjectFile *Obj, bool verbose) { } } -void llvm::printMachOFileHeader(const object::ObjectFile *Obj) { +void printMachOFileHeader(const object::ObjectFile *Obj) { const MachOObjectFile *file = dyn_cast<const MachOObjectFile>(Obj); PrintMachHeader(file, !NonVerbose); } -void llvm::printMachOLoadCommands(const object::ObjectFile *Obj) { +void printMachOLoadCommands(const object::ObjectFile *Obj) { const MachOObjectFile *file = dyn_cast<const MachOObjectFile>(Obj); uint32_t filetype = 0; uint32_t cputype = 0; @@ -9817,7 +10119,7 @@ void llvm::printMachOLoadCommands(const object::ObjectFile *Obj) { // export trie dumping //===----------------------------------------------------------------------===// -void llvm::printMachOExportsTrie(const object::MachOObjectFile *Obj) { +void printMachOExportsTrie(const object::MachOObjectFile *Obj) { uint64_t BaseSegmentAddress = 0; for (const auto &Command : Obj->load_commands()) { if (Command.C.cmd == MachO::LC_SEGMENT) { @@ -9835,7 +10137,7 @@ void llvm::printMachOExportsTrie(const object::MachOObjectFile *Obj) { } } Error Err = Error::success(); - for (const llvm::object::ExportEntry &Entry : Obj->exports(Err)) { + for (const object::ExportEntry &Entry : Obj->exports(Err)) { uint64_t Flags = Entry.flags(); bool ReExport = (Flags & MachO::EXPORT_SYMBOL_FLAGS_REEXPORT); bool WeakDef = (Flags & MachO::EXPORT_SYMBOL_FLAGS_WEAK_DEFINITION); @@ -9889,17 +10191,17 @@ void llvm::printMachOExportsTrie(const object::MachOObjectFile *Obj) { outs() << "\n"; } if (Err) - report_error(Obj->getFileName(), std::move(Err)); + report_error(std::move(Err), Obj->getFileName()); } //===----------------------------------------------------------------------===// // rebase table dumping //===----------------------------------------------------------------------===// -void llvm::printMachORebaseTable(object::MachOObjectFile *Obj) { +void printMachORebaseTable(object::MachOObjectFile *Obj) { outs() << "segment section address type\n"; Error Err = Error::success(); - for (const llvm::object::MachORebaseEntry &Entry : Obj->rebaseTable(Err)) { + for (const object::MachORebaseEntry &Entry : Obj->rebaseTable(Err)) { StringRef SegmentName = Entry.segmentName(); StringRef SectionName = Entry.sectionName(); uint64_t Address = Entry.address(); @@ -9910,7 +10212,7 @@ void llvm::printMachORebaseTable(object::MachOObjectFile *Obj) { Address, Entry.typeName().str().c_str()); } if (Err) - report_error(Obj->getFileName(), std::move(Err)); + report_error(std::move(Err), Obj->getFileName()); } static StringRef ordinalName(const object::MachOObjectFile *Obj, int Ordinal) { @@ -9938,12 +10240,12 @@ static StringRef ordinalName(const object::MachOObjectFile *Obj, int Ordinal) { // bind table dumping //===----------------------------------------------------------------------===// -void llvm::printMachOBindTable(object::MachOObjectFile *Obj) { +void printMachOBindTable(object::MachOObjectFile *Obj) { // Build table of sections so names can used in final output. outs() << "segment section address type " "addend dylib symbol\n"; Error Err = Error::success(); - for (const llvm::object::MachOBindEntry &Entry : Obj->bindTable(Err)) { + for (const object::MachOBindEntry &Entry : Obj->bindTable(Err)) { StringRef SegmentName = Entry.segmentName(); StringRef SectionName = Entry.sectionName(); uint64_t Address = Entry.address(); @@ -9962,18 +10264,18 @@ void llvm::printMachOBindTable(object::MachOObjectFile *Obj) { << Entry.symbolName() << Attr << "\n"; } if (Err) - report_error(Obj->getFileName(), std::move(Err)); + report_error(std::move(Err), Obj->getFileName()); } //===----------------------------------------------------------------------===// // lazy bind table dumping //===----------------------------------------------------------------------===// -void llvm::printMachOLazyBindTable(object::MachOObjectFile *Obj) { +void printMachOLazyBindTable(object::MachOObjectFile *Obj) { outs() << "segment section address " "dylib symbol\n"; Error Err = Error::success(); - for (const llvm::object::MachOBindEntry &Entry : Obj->lazyBindTable(Err)) { + for (const object::MachOBindEntry &Entry : Obj->lazyBindTable(Err)) { StringRef SegmentName = Entry.segmentName(); StringRef SectionName = Entry.sectionName(); uint64_t Address = Entry.address(); @@ -9987,18 +10289,18 @@ void llvm::printMachOLazyBindTable(object::MachOObjectFile *Obj) { << Entry.symbolName() << "\n"; } if (Err) - report_error(Obj->getFileName(), std::move(Err)); + report_error(std::move(Err), Obj->getFileName()); } //===----------------------------------------------------------------------===// // weak bind table dumping //===----------------------------------------------------------------------===// -void llvm::printMachOWeakBindTable(object::MachOObjectFile *Obj) { +void printMachOWeakBindTable(object::MachOObjectFile *Obj) { outs() << "segment section address " "type addend symbol\n"; Error Err = Error::success(); - for (const llvm::object::MachOBindEntry &Entry : Obj->weakBindTable(Err)) { + for (const object::MachOBindEntry &Entry : Obj->weakBindTable(Err)) { // Strong symbols don't have a location to update. if (Entry.flags() & MachO::BIND_SYMBOL_FLAGS_NON_WEAK_DEFINITION) { outs() << " strong " @@ -10019,7 +10321,7 @@ void llvm::printMachOWeakBindTable(object::MachOObjectFile *Obj) { << "\n"; } if (Err) - report_error(Obj->getFileName(), std::move(Err)); + report_error(std::move(Err), Obj->getFileName()); } // get_dyld_bind_info_symbolname() is used for disassembly and passed an @@ -10031,16 +10333,66 @@ static const char *get_dyld_bind_info_symbolname(uint64_t ReferenceValue, if (info->bindtable == nullptr) { info->bindtable = llvm::make_unique<SymbolAddressMap>(); Error Err = Error::success(); - for (const llvm::object::MachOBindEntry &Entry : info->O->bindTable(Err)) { + for (const object::MachOBindEntry &Entry : info->O->bindTable(Err)) { uint64_t Address = Entry.address(); StringRef name = Entry.symbolName(); if (!name.empty()) (*info->bindtable)[Address] = name; } if (Err) - report_error(info->O->getFileName(), std::move(Err)); + report_error(std::move(Err), info->O->getFileName()); } auto name = info->bindtable->lookup(ReferenceValue); return !name.empty() ? name.data() : nullptr; } +void printLazyBindTable(ObjectFile *o) { + outs() << "Lazy bind table:\n"; + if (MachOObjectFile *MachO = dyn_cast<MachOObjectFile>(o)) + printMachOLazyBindTable(MachO); + else + WithColor::error() + << "This operation is only currently supported " + "for Mach-O executable files.\n"; +} + +void printWeakBindTable(ObjectFile *o) { + outs() << "Weak bind table:\n"; + if (MachOObjectFile *MachO = dyn_cast<MachOObjectFile>(o)) + printMachOWeakBindTable(MachO); + else + WithColor::error() + << "This operation is only currently supported " + "for Mach-O executable files.\n"; +} + +void printExportsTrie(const ObjectFile *o) { + outs() << "Exports trie:\n"; + if (const MachOObjectFile *MachO = dyn_cast<MachOObjectFile>(o)) + printMachOExportsTrie(MachO); + else + WithColor::error() + << "This operation is only currently supported " + "for Mach-O executable files.\n"; +} + +void printRebaseTable(ObjectFile *o) { + outs() << "Rebase table:\n"; + if (MachOObjectFile *MachO = dyn_cast<MachOObjectFile>(o)) + printMachORebaseTable(MachO); + else + WithColor::error() + << "This operation is only currently supported " + "for Mach-O executable files.\n"; +} + +void printBindTable(ObjectFile *o) { + outs() << "Bind table:\n"; + if (MachOObjectFile *MachO = dyn_cast<MachOObjectFile>(o)) + printMachOBindTable(MachO); + else + WithColor::error() + << "This operation is only currently supported " + "for Mach-O executable files.\n"; +} +} // namespace llvm diff --git a/tools/llvm-objdump/WasmDump.cpp b/tools/llvm-objdump/WasmDump.cpp index 045002cd4b34..da27a4acbb5f 100644 --- a/tools/llvm-objdump/WasmDump.cpp +++ b/tools/llvm-objdump/WasmDump.cpp @@ -1,9 +1,8 @@ //===-- WasmDump.cpp - wasm-specific dumper ---------------------*- C++ -*-===// // -// The LLVM Compiler Infrastructure -// -// This file is distributed under the University of Illinois Open Source -// License. See LICENSE.TXT for details. +// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions. +// See https://llvm.org/LICENSE.txt for license information. +// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception // //===----------------------------------------------------------------------===// /// @@ -15,14 +14,39 @@ #include "llvm-objdump.h" #include "llvm/Object/Wasm.h" -using namespace llvm; -using namespace object; +using namespace llvm::object; -void llvm::printWasmFileHeader(const object::ObjectFile *Obj) { - const WasmObjectFile *File = dyn_cast<const WasmObjectFile>(Obj); +namespace llvm { +void printWasmFileHeader(const object::ObjectFile *Obj) { + const auto *File = dyn_cast<const WasmObjectFile>(Obj); outs() << "Program Header:\n"; outs() << "Version: 0x"; outs().write_hex(File->getHeader().Version); outs() << "\n"; } + +Error getWasmRelocationValueString(const WasmObjectFile *Obj, + const RelocationRef &RelRef, + SmallVectorImpl<char> &Result) { + const wasm::WasmRelocation &Rel = Obj->getWasmRelocation(RelRef); + symbol_iterator SI = RelRef.getSymbol(); + std::string FmtBuf; + raw_string_ostream Fmt(FmtBuf); + if (SI == Obj->symbol_end()) { + // Not all wasm relocations have symbols associated with them. + // In particular R_WASM_TYPE_INDEX_LEB. + Fmt << Rel.Index; + } else { + Expected<StringRef> SymNameOrErr = SI->getName(); + if (!SymNameOrErr) + return SymNameOrErr.takeError(); + StringRef SymName = *SymNameOrErr; + Result.append(SymName.begin(), SymName.end()); + } + Fmt << (Rel.Addend < 0 ? "" : "+") << Rel.Addend; + Fmt.flush(); + Result.append(FmtBuf.begin(), FmtBuf.end()); + return Error::success(); +} +} // namespace llvm diff --git a/tools/llvm-objdump/llvm-objdump.cpp b/tools/llvm-objdump/llvm-objdump.cpp index ba8d3c5b8d5c..58981203c59e 100644 --- a/tools/llvm-objdump/llvm-objdump.cpp +++ b/tools/llvm-objdump/llvm-objdump.cpp @@ -1,9 +1,8 @@ //===-- llvm-objdump.cpp - Object file dumping utility for llvm -----------===// // -// The LLVM Compiler Infrastructure -// -// This file is distributed under the University of Illinois Open Source -// License. See LICENSE.TXT for details. +// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions. +// See https://llvm.org/LICENSE.txt for license information. +// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception // //===----------------------------------------------------------------------===// // @@ -19,6 +18,7 @@ #include "llvm-objdump.h" #include "llvm/ADT/Optional.h" #include "llvm/ADT/STLExtras.h" +#include "llvm/ADT/SetOperations.h" #include "llvm/ADT/StringExtras.h" #include "llvm/ADT/StringSet.h" #include "llvm/ADT/Triple.h" @@ -68,283 +68,298 @@ #include <unordered_map> #include <utility> -using namespace llvm; -using namespace object; +using namespace llvm::object; + +namespace llvm { + +cl::OptionCategory ObjdumpCat("llvm-objdump Options"); + +// MachO specific +extern cl::OptionCategory MachOCat; +extern cl::opt<bool> Bind; +extern cl::opt<bool> DataInCode; +extern cl::opt<bool> DylibsUsed; +extern cl::opt<bool> DylibId; +extern cl::opt<bool> ExportsTrie; +extern cl::opt<bool> FirstPrivateHeader; +extern cl::opt<bool> IndirectSymbols; +extern cl::opt<bool> InfoPlist; +extern cl::opt<bool> LazyBind; +extern cl::opt<bool> LinkOptHints; +extern cl::opt<bool> ObjcMetaData; +extern cl::opt<bool> Rebase; +extern cl::opt<bool> UniversalHeaders; +extern cl::opt<bool> WeakBind; + +static cl::opt<uint64_t> AdjustVMA( + "adjust-vma", + cl::desc("Increase the displayed address by the specified offset"), + cl::value_desc("offset"), cl::init(0), cl::cat(ObjdumpCat)); -cl::opt<bool> - llvm::AllHeaders("all-headers", - cl::desc("Display all available header information")); +static cl::opt<bool> + AllHeaders("all-headers", + cl::desc("Display all available header information"), + cl::cat(ObjdumpCat)); static cl::alias AllHeadersShort("x", cl::desc("Alias for --all-headers"), + cl::NotHidden, cl::Grouping, cl::aliasopt(AllHeaders)); -static cl::list<std::string> -InputFilenames(cl::Positional, cl::desc("<input object files>"),cl::ZeroOrMore); +static cl::opt<std::string> + ArchName("arch-name", + cl::desc("Target arch to disassemble for, " + "see -version for available targets"), + cl::cat(ObjdumpCat)); + +cl::opt<bool> ArchiveHeaders("archive-headers", + cl::desc("Display archive header information"), + cl::cat(ObjdumpCat)); +static cl::alias ArchiveHeadersShort("a", + cl::desc("Alias for --archive-headers"), + cl::NotHidden, cl::Grouping, + cl::aliasopt(ArchiveHeaders)); + +cl::opt<bool> Demangle("demangle", cl::desc("Demangle symbols names"), + cl::init(false), cl::cat(ObjdumpCat)); +static cl::alias DemangleShort("C", cl::desc("Alias for --demangle"), + cl::NotHidden, cl::Grouping, + cl::aliasopt(Demangle)); + +cl::opt<bool> Disassemble( + "disassemble", + cl::desc("Display assembler mnemonics for the machine instructions"), + cl::cat(ObjdumpCat)); +static cl::alias DisassembleShort("d", cl::desc("Alias for --disassemble"), + cl::NotHidden, cl::Grouping, + cl::aliasopt(Disassemble)); + +cl::opt<bool> DisassembleAll( + "disassemble-all", + cl::desc("Display assembler mnemonics for the machine instructions"), + cl::cat(ObjdumpCat)); +static cl::alias DisassembleAllShort("D", + cl::desc("Alias for --disassemble-all"), + cl::NotHidden, cl::Grouping, + cl::aliasopt(DisassembleAll)); -cl::opt<bool> -llvm::Disassemble("disassemble", - cl::desc("Display assembler mnemonics for the machine instructions")); +static cl::list<std::string> + DisassembleFunctions("disassemble-functions", cl::CommaSeparated, + cl::desc("List of functions to disassemble. " + "Accept demangled names when --demangle is " + "specified, otherwise accept mangled names"), + cl::cat(ObjdumpCat)); + +static cl::opt<bool> DisassembleZeroes( + "disassemble-zeroes", + cl::desc("Do not skip blocks of zeroes when disassembling"), + cl::cat(ObjdumpCat)); static cl::alias -Disassembled("d", cl::desc("Alias for --disassemble"), - cl::aliasopt(Disassemble)); + DisassembleZeroesShort("z", cl::desc("Alias for --disassemble-zeroes"), + cl::NotHidden, cl::Grouping, + cl::aliasopt(DisassembleZeroes)); -cl::opt<bool> -llvm::DisassembleAll("disassemble-all", - cl::desc("Display assembler mnemonics for the machine instructions")); +static cl::list<std::string> + DisassemblerOptions("disassembler-options", + cl::desc("Pass target specific disassembler options"), + cl::value_desc("options"), cl::CommaSeparated, + cl::cat(ObjdumpCat)); static cl::alias -DisassembleAlld("D", cl::desc("Alias for --disassemble-all"), - cl::aliasopt(DisassembleAll)); - -cl::opt<bool> llvm::Demangle("demangle", cl::desc("Demangle symbols names"), - cl::init(false)); - -static cl::alias DemangleShort("C", cl::desc("Alias for --demangle"), - cl::aliasopt(llvm::Demangle)); + DisassemblerOptionsShort("M", cl::desc("Alias for --disassembler-options"), + cl::NotHidden, cl::Grouping, cl::Prefix, + cl::CommaSeparated, + cl::aliasopt(DisassemblerOptions)); -static cl::list<std::string> -DisassembleFunctions("df", - cl::CommaSeparated, - cl::desc("List of functions to disassemble")); -static StringSet<> DisasmFuncsSet; +cl::opt<DIDumpType> DwarfDumpType( + "dwarf", cl::init(DIDT_Null), cl::desc("Dump of dwarf debug sections:"), + cl::values(clEnumValN(DIDT_DebugFrame, "frames", ".debug_frame")), + cl::cat(ObjdumpCat)); + +static cl::opt<bool> DynamicRelocations( + "dynamic-reloc", + cl::desc("Display the dynamic relocation entries in the file"), + cl::cat(ObjdumpCat)); +static cl::alias DynamicRelocationShort("R", + cl::desc("Alias for --dynamic-reloc"), + cl::NotHidden, cl::Grouping, + cl::aliasopt(DynamicRelocations)); -cl::opt<bool> -llvm::Relocations("reloc", - cl::desc("Display the relocation entries in the file")); -static cl::alias RelocationsShort("r", cl::desc("Alias for --reloc"), - cl::NotHidden, - cl::aliasopt(llvm::Relocations)); +static cl::opt<bool> + FaultMapSection("fault-map-section", + cl::desc("Display contents of faultmap section"), + cl::cat(ObjdumpCat)); -cl::opt<bool> -llvm::DynamicRelocations("dynamic-reloc", - cl::desc("Display the dynamic relocation entries in the file")); -static cl::alias -DynamicRelocationsd("R", cl::desc("Alias for --dynamic-reloc"), - cl::aliasopt(DynamicRelocations)); +static cl::opt<bool> + FileHeaders("file-headers", + cl::desc("Display the contents of the overall file header"), + cl::cat(ObjdumpCat)); +static cl::alias FileHeadersShort("f", cl::desc("Alias for --file-headers"), + cl::NotHidden, cl::Grouping, + cl::aliasopt(FileHeaders)); -cl::opt<bool> - llvm::SectionContents("full-contents", - cl::desc("Display the content of each section")); +cl::opt<bool> SectionContents("full-contents", + cl::desc("Display the content of each section"), + cl::cat(ObjdumpCat)); static cl::alias SectionContentsShort("s", cl::desc("Alias for --full-contents"), + cl::NotHidden, cl::Grouping, cl::aliasopt(SectionContents)); -cl::opt<bool> llvm::SymbolTable("syms", cl::desc("Display the symbol table")); -static cl::alias SymbolTableShort("t", cl::desc("Alias for --syms"), - cl::NotHidden, - cl::aliasopt(llvm::SymbolTable)); - -cl::opt<bool> -llvm::ExportsTrie("exports-trie", cl::desc("Display mach-o exported symbols")); - -cl::opt<bool> -llvm::Rebase("rebase", cl::desc("Display mach-o rebasing info")); - -cl::opt<bool> -llvm::Bind("bind", cl::desc("Display mach-o binding info")); - -cl::opt<bool> -llvm::LazyBind("lazy-bind", cl::desc("Display mach-o lazy binding info")); - -cl::opt<bool> -llvm::WeakBind("weak-bind", cl::desc("Display mach-o weak binding info")); - -cl::opt<bool> -llvm::RawClangAST("raw-clang-ast", - cl::desc("Dump the raw binary contents of the clang AST section")); +static cl::list<std::string> InputFilenames(cl::Positional, + cl::desc("<input object files>"), + cl::ZeroOrMore, + cl::cat(ObjdumpCat)); static cl::opt<bool> -MachOOpt("macho", cl::desc("Use MachO specific object file parser")); -static cl::alias -MachOm("m", cl::desc("Alias for --macho"), cl::aliasopt(MachOOpt)); - -cl::opt<std::string> -llvm::TripleName("triple", cl::desc("Target triple to disassemble for, " - "see -version for available targets")); + PrintLines("line-numbers", + cl::desc("Display source line numbers with " + "disassembly. Implies disassemble object"), + cl::cat(ObjdumpCat)); +static cl::alias PrintLinesShort("l", cl::desc("Alias for --line-numbers"), + cl::NotHidden, cl::Grouping, + cl::aliasopt(PrintLines)); + +static cl::opt<bool> MachOOpt("macho", + cl::desc("Use MachO specific object file parser"), + cl::cat(ObjdumpCat)); +static cl::alias MachOm("m", cl::desc("Alias for --macho"), cl::NotHidden, + cl::Grouping, cl::aliasopt(MachOOpt)); cl::opt<std::string> -llvm::MCPU("mcpu", - cl::desc("Target a specific cpu type (-mcpu=help for details)"), - cl::value_desc("cpu-name"), - cl::init("")); - -cl::opt<std::string> -llvm::ArchName("arch-name", cl::desc("Target arch to disassemble for, " - "see -version for available targets")); + MCPU("mcpu", + cl::desc("Target a specific cpu type (-mcpu=help for details)"), + cl::value_desc("cpu-name"), cl::init(""), cl::cat(ObjdumpCat)); + +cl::list<std::string> MAttrs("mattr", cl::CommaSeparated, + cl::desc("Target specific attributes"), + cl::value_desc("a1,+a2,-a3,..."), + cl::cat(ObjdumpCat)); + +cl::opt<bool> NoShowRawInsn("no-show-raw-insn", + cl::desc("When disassembling " + "instructions, do not print " + "the instruction bytes."), + cl::cat(ObjdumpCat)); +cl::opt<bool> NoLeadingAddr("no-leading-addr", + cl::desc("Print no leading address"), + cl::cat(ObjdumpCat)); + +static cl::opt<bool> RawClangAST( + "raw-clang-ast", + cl::desc("Dump the raw binary contents of the clang AST section"), + cl::cat(ObjdumpCat)); cl::opt<bool> -llvm::SectionHeaders("section-headers", cl::desc("Display summaries of the " - "headers for each section.")); -static cl::alias -SectionHeadersShort("headers", cl::desc("Alias for --section-headers"), - cl::aliasopt(SectionHeaders)); -static cl::alias -SectionHeadersShorter("h", cl::desc("Alias for --section-headers"), - cl::aliasopt(SectionHeaders)); - -cl::list<std::string> -llvm::FilterSections("section", cl::desc("Operate on the specified sections only. " - "With -macho dump segment,section")); -cl::alias -static FilterSectionsj("j", cl::desc("Alias for --section"), - cl::aliasopt(llvm::FilterSections)); - -cl::list<std::string> -llvm::MAttrs("mattr", - cl::CommaSeparated, - cl::desc("Target specific attributes"), - cl::value_desc("a1,+a2,-a3,...")); - -cl::opt<bool> -llvm::NoShowRawInsn("no-show-raw-insn", cl::desc("When disassembling " - "instructions, do not print " - "the instruction bytes.")); -cl::opt<bool> -llvm::NoLeadingAddr("no-leading-addr", cl::desc("Print no leading address")); - -cl::opt<bool> -llvm::UnwindInfo("unwind-info", cl::desc("Display unwind information")); - -static cl::alias -UnwindInfoShort("u", cl::desc("Alias for --unwind-info"), - cl::aliasopt(UnwindInfo)); - -cl::opt<bool> -llvm::PrivateHeaders("private-headers", - cl::desc("Display format specific file headers")); - -cl::opt<bool> -llvm::FirstPrivateHeader("private-header", - cl::desc("Display only the first format specific file " - "header")); - -static cl::alias -PrivateHeadersShort("p", cl::desc("Alias for --private-headers"), - cl::aliasopt(PrivateHeaders)); - -cl::opt<bool> llvm::FileHeaders( - "file-headers", - cl::desc("Display the contents of the overall file header")); - -static cl::alias FileHeadersShort("f", cl::desc("Alias for --file-headers"), - cl::aliasopt(FileHeaders)); - -cl::opt<bool> - llvm::ArchiveHeaders("archive-headers", - cl::desc("Display archive header information")); + Relocations("reloc", cl::desc("Display the relocation entries in the file"), + cl::cat(ObjdumpCat)); +static cl::alias RelocationsShort("r", cl::desc("Alias for --reloc"), + cl::NotHidden, cl::Grouping, + cl::aliasopt(Relocations)); -cl::alias -ArchiveHeadersShort("a", cl::desc("Alias for --archive-headers"), - cl::aliasopt(ArchiveHeaders)); +cl::opt<bool> PrintImmHex("print-imm-hex", + cl::desc("Use hex format for immediate values"), + cl::cat(ObjdumpCat)); -cl::opt<bool> - llvm::PrintImmHex("print-imm-hex", - cl::desc("Use hex format for immediate values")); +cl::opt<bool> PrivateHeaders("private-headers", + cl::desc("Display format specific file headers"), + cl::cat(ObjdumpCat)); +static cl::alias PrivateHeadersShort("p", + cl::desc("Alias for --private-headers"), + cl::NotHidden, cl::Grouping, + cl::aliasopt(PrivateHeaders)); -cl::opt<bool> PrintFaultMaps("fault-map-section", - cl::desc("Display contents of faultmap section")); +cl::list<std::string> + FilterSections("section", + cl::desc("Operate on the specified sections only. " + "With -macho dump segment,section"), + cl::cat(ObjdumpCat)); +static cl::alias FilterSectionsj("j", cl::desc("Alias for --section"), + cl::NotHidden, cl::Grouping, cl::Prefix, + cl::aliasopt(FilterSections)); + +cl::opt<bool> SectionHeaders("section-headers", + cl::desc("Display summaries of the " + "headers for each section."), + cl::cat(ObjdumpCat)); +static cl::alias SectionHeadersShort("headers", + cl::desc("Alias for --section-headers"), + cl::NotHidden, + cl::aliasopt(SectionHeaders)); +static cl::alias SectionHeadersShorter("h", + cl::desc("Alias for --section-headers"), + cl::NotHidden, cl::Grouping, + cl::aliasopt(SectionHeaders)); -cl::opt<DIDumpType> llvm::DwarfDumpType( - "dwarf", cl::init(DIDT_Null), cl::desc("Dump of dwarf debug sections:"), - cl::values(clEnumValN(DIDT_DebugFrame, "frames", ".debug_frame"))); +static cl::opt<bool> + ShowLMA("show-lma", + cl::desc("Display LMA column when dumping ELF section headers"), + cl::cat(ObjdumpCat)); -cl::opt<bool> PrintSource( +static cl::opt<bool> PrintSource( "source", cl::desc( - "Display source inlined with disassembly. Implies disassemble object")); - -cl::alias PrintSourceShort("S", cl::desc("Alias for -source"), - cl::aliasopt(PrintSource)); + "Display source inlined with disassembly. Implies disassemble object"), + cl::cat(ObjdumpCat)); +static cl::alias PrintSourceShort("S", cl::desc("Alias for -source"), + cl::NotHidden, cl::Grouping, + cl::aliasopt(PrintSource)); -cl::opt<bool> PrintLines("line-numbers", - cl::desc("Display source line numbers with " - "disassembly. Implies disassemble object")); - -cl::alias PrintLinesShort("l", cl::desc("Alias for -line-numbers"), - cl::aliasopt(PrintLines)); - -cl::opt<unsigned long long> +static cl::opt<uint64_t> StartAddress("start-address", cl::desc("Disassemble beginning at address"), - cl::value_desc("address"), cl::init(0)); -cl::opt<unsigned long long> - StopAddress("stop-address", - cl::desc("Stop disassembly at address"), - cl::value_desc("address"), cl::init(UINT64_MAX)); - -cl::opt<bool> DisassembleZeroes( - "disassemble-zeroes", - cl::desc("Do not skip blocks of zeroes when disassembling")); -cl::alias DisassembleZeroesShort("z", - cl::desc("Alias for --disassemble-zeroes"), - cl::aliasopt(DisassembleZeroes)); + cl::value_desc("address"), cl::init(0), cl::cat(ObjdumpCat)); +static cl::opt<uint64_t> StopAddress("stop-address", + cl::desc("Stop disassembly at address"), + cl::value_desc("address"), + cl::init(UINT64_MAX), cl::cat(ObjdumpCat)); + +cl::opt<bool> SymbolTable("syms", cl::desc("Display the symbol table"), + cl::cat(ObjdumpCat)); +static cl::alias SymbolTableShort("t", cl::desc("Alias for --syms"), + cl::NotHidden, cl::Grouping, + cl::aliasopt(SymbolTable)); -static StringRef ToolName; +cl::opt<std::string> TripleName("triple", + cl::desc("Target triple to disassemble for, " + "see -version for available targets"), + cl::cat(ObjdumpCat)); -typedef std::vector<std::tuple<uint64_t, StringRef, uint8_t>> SectionSymbolsTy; +cl::opt<bool> UnwindInfo("unwind-info", cl::desc("Display unwind information"), + cl::cat(ObjdumpCat)); +static cl::alias UnwindInfoShort("u", cl::desc("Alias for --unwind-info"), + cl::NotHidden, cl::Grouping, + cl::aliasopt(UnwindInfo)); -namespace { -typedef std::function<bool(llvm::object::SectionRef const &)> FilterPredicate; +static cl::opt<bool> + Wide("wide", cl::desc("Ignored for compatibility with GNU objdump"), + cl::cat(ObjdumpCat)); +static cl::alias WideShort("w", cl::Grouping, cl::aliasopt(Wide)); -class SectionFilterIterator { -public: - SectionFilterIterator(FilterPredicate P, - llvm::object::section_iterator const &I, - llvm::object::section_iterator const &E) - : Predicate(std::move(P)), Iterator(I), End(E) { - ScanPredicate(); - } - const llvm::object::SectionRef &operator*() const { return *Iterator; } - SectionFilterIterator &operator++() { - ++Iterator; - ScanPredicate(); - return *this; - } - bool operator!=(SectionFilterIterator const &Other) const { - return Iterator != Other.Iterator; - } +static cl::extrahelp + HelpResponse("\nPass @FILE as argument to read options from FILE.\n"); -private: - void ScanPredicate() { - while (Iterator != End && !Predicate(*Iterator)) { - ++Iterator; - } - } - FilterPredicate Predicate; - llvm::object::section_iterator Iterator; - llvm::object::section_iterator End; -}; +static StringSet<> DisasmFuncsSet; +static StringSet<> FoundSectionSet; +static StringRef ToolName; -class SectionFilter { -public: - SectionFilter(FilterPredicate P, llvm::object::ObjectFile const &O) - : Predicate(std::move(P)), Object(O) {} - SectionFilterIterator begin() { - return SectionFilterIterator(Predicate, Object.section_begin(), - Object.section_end()); - } - SectionFilterIterator end() { - return SectionFilterIterator(Predicate, Object.section_end(), - Object.section_end()); - } +typedef std::vector<std::tuple<uint64_t, StringRef, uint8_t>> SectionSymbolsTy; -private: - FilterPredicate Predicate; - llvm::object::ObjectFile const &Object; -}; -SectionFilter ToolSectionFilter(llvm::object::ObjectFile const &O) { - return SectionFilter( - [](llvm::object::SectionRef const &S) { - if (FilterSections.empty()) - return true; - llvm::StringRef String; - std::error_code error = S.getName(String); - if (error) - return false; - return is_contained(FilterSections, String); - }, - O); +static bool shouldKeep(object::SectionRef S) { + if (FilterSections.empty()) + return true; + StringRef SecName; + std::error_code error = S.getName(SecName); + if (error) + return false; + // StringSet does not allow empty key so avoid adding sections with + // no name (such as the section with index 0) here. + if (!SecName.empty()) + FoundSectionSet.insert(SecName); + return is_contained(FilterSections, SecName); } + +SectionFilter ToolSectionFilter(object::ObjectFile const &O) { + return SectionFilter([](object::SectionRef S) { return shouldKeep(S); }, O); } -void llvm::error(std::error_code EC) { +void error(std::error_code EC) { if (!EC) return; WithColor::error(errs(), ToolName) @@ -353,34 +368,39 @@ void llvm::error(std::error_code EC) { exit(1); } -LLVM_ATTRIBUTE_NORETURN void llvm::error(Twine Message) { +void error(Error E) { + if (!E) + return; + WithColor::error(errs(), ToolName) << toString(std::move(E)); + exit(1); +} + +LLVM_ATTRIBUTE_NORETURN void error(Twine Message) { WithColor::error(errs(), ToolName) << Message << ".\n"; errs().flush(); exit(1); } -void llvm::warn(StringRef Message) { +void warn(StringRef Message) { WithColor::warning(errs(), ToolName) << Message << ".\n"; errs().flush(); } -LLVM_ATTRIBUTE_NORETURN void llvm::report_error(StringRef File, - Twine Message) { - WithColor::error(errs(), ToolName) - << "'" << File << "': " << Message << ".\n"; - exit(1); +static void warn(Twine Message) { + // Output order between errs() and outs() matters especially for archive + // files where the output is per member object. + outs().flush(); + WithColor::warning(errs(), ToolName) << Message << "\n"; + errs().flush(); } -LLVM_ATTRIBUTE_NORETURN void llvm::report_error(StringRef File, - std::error_code EC) { - assert(EC); +LLVM_ATTRIBUTE_NORETURN void report_error(StringRef File, Twine Message) { WithColor::error(errs(), ToolName) - << "'" << File << "': " << EC.message() << ".\n"; + << "'" << File << "': " << Message << ".\n"; exit(1); } -LLVM_ATTRIBUTE_NORETURN void llvm::report_error(StringRef File, - llvm::Error E) { +LLVM_ATTRIBUTE_NORETURN void report_error(Error E, StringRef File) { assert(E); std::string Buf; raw_string_ostream OS(Buf); @@ -390,10 +410,9 @@ LLVM_ATTRIBUTE_NORETURN void llvm::report_error(StringRef File, exit(1); } -LLVM_ATTRIBUTE_NORETURN void llvm::report_error(StringRef ArchiveName, - StringRef FileName, - llvm::Error E, - StringRef ArchitectureName) { +LLVM_ATTRIBUTE_NORETURN void report_error(Error E, StringRef ArchiveName, + StringRef FileName, + StringRef ArchitectureName) { assert(E); WithColor::error(errs(), ToolName); if (ArchiveName != "") @@ -410,25 +429,39 @@ LLVM_ATTRIBUTE_NORETURN void llvm::report_error(StringRef ArchiveName, exit(1); } -LLVM_ATTRIBUTE_NORETURN void llvm::report_error(StringRef ArchiveName, - const object::Archive::Child &C, - llvm::Error E, - StringRef ArchitectureName) { +LLVM_ATTRIBUTE_NORETURN void report_error(Error E, StringRef ArchiveName, + const object::Archive::Child &C, + StringRef ArchitectureName) { Expected<StringRef> NameOrErr = C.getName(); // TODO: if we have a error getting the name then it would be nice to print // the index of which archive member this is and or its offset in the // archive instead of "???" as the name. if (!NameOrErr) { consumeError(NameOrErr.takeError()); - llvm::report_error(ArchiveName, "???", std::move(E), ArchitectureName); + report_error(std::move(E), ArchiveName, "???", ArchitectureName); } else - llvm::report_error(ArchiveName, NameOrErr.get(), std::move(E), - ArchitectureName); + report_error(std::move(E), ArchiveName, NameOrErr.get(), ArchitectureName); +} + +static void warnOnNoMatchForSections() { + SetVector<StringRef> MissingSections; + for (StringRef S : FilterSections) { + if (FoundSectionSet.count(S)) + return; + // User may specify a unnamed section. Don't warn for it. + if (!S.empty()) + MissingSections.insert(S); + } + + // Warn only if no section in FilterSections is matched. + for (StringRef S : MissingSections) + warn("section '" + S + "' mentioned in a -j/--section option, but not " + "found in any input file"); } static const Target *getTarget(const ObjectFile *Obj = nullptr) { // Figure out the target triple. - llvm::Triple TheTriple("unknown-unknown-unknown"); + Triple TheTriple("unknown-unknown-unknown"); if (TripleName.empty()) { if (Obj) TheTriple = Obj->makeTriple(); @@ -459,423 +492,21 @@ static const Target *getTarget(const ObjectFile *Obj = nullptr) { return TheTarget; } -bool llvm::isRelocAddressLess(RelocationRef A, RelocationRef B) { +bool isRelocAddressLess(RelocationRef A, RelocationRef B) { return A.getOffset() < B.getOffset(); } -static std::string demangle(StringRef Name) { - char *Demangled = nullptr; - if (Name.startswith("_Z")) - Demangled = itaniumDemangle(Name.data(), Demangled, nullptr, nullptr); - else if (Name.startswith("?")) - Demangled = microsoftDemangle(Name.data(), Demangled, nullptr, nullptr); - - if (!Demangled) - return Name; - - std::string Ret = Demangled; - free(Demangled); - return Ret; -} - -template <class ELFT> -static std::error_code getRelocationValueString(const ELFObjectFile<ELFT> *Obj, - const RelocationRef &RelRef, - SmallVectorImpl<char> &Result) { - typedef typename ELFObjectFile<ELFT>::Elf_Sym Elf_Sym; - typedef typename ELFObjectFile<ELFT>::Elf_Shdr Elf_Shdr; - typedef typename ELFObjectFile<ELFT>::Elf_Rela Elf_Rela; - - const ELFFile<ELFT> &EF = *Obj->getELFFile(); - DataRefImpl Rel = RelRef.getRawDataRefImpl(); - auto SecOrErr = EF.getSection(Rel.d.a); - if (!SecOrErr) - return errorToErrorCode(SecOrErr.takeError()); - const Elf_Shdr *Sec = *SecOrErr; - auto SymTabOrErr = EF.getSection(Sec->sh_link); - if (!SymTabOrErr) - return errorToErrorCode(SymTabOrErr.takeError()); - const Elf_Shdr *SymTab = *SymTabOrErr; - assert(SymTab->sh_type == ELF::SHT_SYMTAB || - SymTab->sh_type == ELF::SHT_DYNSYM); - auto StrTabSec = EF.getSection(SymTab->sh_link); - if (!StrTabSec) - return errorToErrorCode(StrTabSec.takeError()); - auto StrTabOrErr = EF.getStringTable(*StrTabSec); - if (!StrTabOrErr) - return errorToErrorCode(StrTabOrErr.takeError()); - StringRef StrTab = *StrTabOrErr; - int64_t Addend = 0; - // If there is no Symbol associated with the relocation, we set the undef - // boolean value to 'true'. This will prevent us from calling functions that - // requires the relocation to be associated with a symbol. - bool Undef = false; - switch (Sec->sh_type) { - default: - return object_error::parse_failed; - case ELF::SHT_REL: { - // TODO: Read implicit addend from section data. - break; - } - case ELF::SHT_RELA: { - const Elf_Rela *ERela = Obj->getRela(Rel); - Addend = ERela->r_addend; - Undef = ERela->getSymbol(false) == 0; - break; - } - } - std::string Target; - if (!Undef) { - symbol_iterator SI = RelRef.getSymbol(); - const Elf_Sym *symb = Obj->getSymbol(SI->getRawDataRefImpl()); - if (symb->getType() == ELF::STT_SECTION) { - Expected<section_iterator> SymSI = SI->getSection(); - if (!SymSI) - return errorToErrorCode(SymSI.takeError()); - const Elf_Shdr *SymSec = Obj->getSection((*SymSI)->getRawDataRefImpl()); - auto SecName = EF.getSectionName(SymSec); - if (!SecName) - return errorToErrorCode(SecName.takeError()); - Target = *SecName; - } else { - Expected<StringRef> SymName = symb->getName(StrTab); - if (!SymName) - return errorToErrorCode(SymName.takeError()); - if (Demangle) - Target = demangle(*SymName); - else - Target = *SymName; - } - } else - Target = "*ABS*"; - - // Default scheme is to print Target, as well as "+ <addend>" for nonzero - // addend. Should be acceptable for all normal purposes. - std::string FmtBuf; - raw_string_ostream Fmt(FmtBuf); - Fmt << Target; - if (Addend != 0) - Fmt << (Addend < 0 ? "" : "+") << Addend; - Fmt.flush(); - Result.append(FmtBuf.begin(), FmtBuf.end()); - return std::error_code(); -} - -static std::error_code getRelocationValueString(const ELFObjectFileBase *Obj, - const RelocationRef &Rel, - SmallVectorImpl<char> &Result) { - if (auto *ELF32LE = dyn_cast<ELF32LEObjectFile>(Obj)) - return getRelocationValueString(ELF32LE, Rel, Result); - if (auto *ELF64LE = dyn_cast<ELF64LEObjectFile>(Obj)) - return getRelocationValueString(ELF64LE, Rel, Result); - if (auto *ELF32BE = dyn_cast<ELF32BEObjectFile>(Obj)) - return getRelocationValueString(ELF32BE, Rel, Result); - auto *ELF64BE = cast<ELF64BEObjectFile>(Obj); - return getRelocationValueString(ELF64BE, Rel, Result); -} - -static std::error_code getRelocationValueString(const COFFObjectFile *Obj, - const RelocationRef &Rel, - SmallVectorImpl<char> &Result) { - symbol_iterator SymI = Rel.getSymbol(); - Expected<StringRef> SymNameOrErr = SymI->getName(); - if (!SymNameOrErr) - return errorToErrorCode(SymNameOrErr.takeError()); - StringRef SymName = *SymNameOrErr; - Result.append(SymName.begin(), SymName.end()); - return std::error_code(); -} - -static void printRelocationTargetName(const MachOObjectFile *O, - const MachO::any_relocation_info &RE, - raw_string_ostream &Fmt) { - // Target of a scattered relocation is an address. In the interest of - // generating pretty output, scan through the symbol table looking for a - // symbol that aligns with that address. If we find one, print it. - // Otherwise, we just print the hex address of the target. - if (O->isRelocationScattered(RE)) { - uint32_t Val = O->getPlainRelocationSymbolNum(RE); - - for (const SymbolRef &Symbol : O->symbols()) { - Expected<uint64_t> Addr = Symbol.getAddress(); - if (!Addr) - report_error(O->getFileName(), Addr.takeError()); - if (*Addr != Val) - continue; - Expected<StringRef> Name = Symbol.getName(); - if (!Name) - report_error(O->getFileName(), Name.takeError()); - Fmt << *Name; - return; - } - - // If we couldn't find a symbol that this relocation refers to, try - // to find a section beginning instead. - for (const SectionRef &Section : ToolSectionFilter(*O)) { - std::error_code ec; - - StringRef Name; - uint64_t Addr = Section.getAddress(); - if (Addr != Val) - continue; - if ((ec = Section.getName(Name))) - report_error(O->getFileName(), ec); - Fmt << Name; - return; - } - - Fmt << format("0x%x", Val); - return; - } - - StringRef S; - bool isExtern = O->getPlainRelocationExternal(RE); - uint64_t Val = O->getPlainRelocationSymbolNum(RE); - - if (O->getAnyRelocationType(RE) == MachO::ARM64_RELOC_ADDEND) { - Fmt << format("0x%0" PRIx64, Val); - return; - } - - if (isExtern) { - symbol_iterator SI = O->symbol_begin(); - advance(SI, Val); - Expected<StringRef> SOrErr = SI->getName(); - if (!SOrErr) - report_error(O->getFileName(), SOrErr.takeError()); - S = *SOrErr; - } else { - section_iterator SI = O->section_begin(); - // Adjust for the fact that sections are 1-indexed. - if (Val == 0) { - Fmt << "0 (?,?)"; - return; - } - uint32_t I = Val - 1; - while (I != 0 && SI != O->section_end()) { - --I; - advance(SI, 1); - } - if (SI == O->section_end()) - Fmt << Val << " (?,?)"; - else - SI->getName(S); - } - - Fmt << S; -} - -static std::error_code getRelocationValueString(const WasmObjectFile *Obj, - const RelocationRef &RelRef, - SmallVectorImpl<char> &Result) { - const wasm::WasmRelocation& Rel = Obj->getWasmRelocation(RelRef); - symbol_iterator SI = RelRef.getSymbol(); - std::string FmtBuf; - raw_string_ostream Fmt(FmtBuf); - if (SI == Obj->symbol_end()) { - // Not all wasm relocations have symbols associated with them. - // In particular R_WEBASSEMBLY_TYPE_INDEX_LEB. - Fmt << Rel.Index; - } else { - Expected<StringRef> SymNameOrErr = SI->getName(); - if (!SymNameOrErr) - return errorToErrorCode(SymNameOrErr.takeError()); - StringRef SymName = *SymNameOrErr; - Result.append(SymName.begin(), SymName.end()); - } - Fmt << (Rel.Addend < 0 ? "" : "+") << Rel.Addend; - Fmt.flush(); - Result.append(FmtBuf.begin(), FmtBuf.end()); - return std::error_code(); -} - -static std::error_code getRelocationValueString(const MachOObjectFile *Obj, - const RelocationRef &RelRef, - SmallVectorImpl<char> &Result) { - DataRefImpl Rel = RelRef.getRawDataRefImpl(); - MachO::any_relocation_info RE = Obj->getRelocation(Rel); - - unsigned Arch = Obj->getArch(); - - std::string FmtBuf; - raw_string_ostream Fmt(FmtBuf); - unsigned Type = Obj->getAnyRelocationType(RE); - bool IsPCRel = Obj->getAnyRelocationPCRel(RE); - - // Determine any addends that should be displayed with the relocation. - // These require decoding the relocation type, which is triple-specific. - - // X86_64 has entirely custom relocation types. - if (Arch == Triple::x86_64) { - switch (Type) { - case MachO::X86_64_RELOC_GOT_LOAD: - case MachO::X86_64_RELOC_GOT: { - printRelocationTargetName(Obj, RE, Fmt); - Fmt << "@GOT"; - if (IsPCRel) - Fmt << "PCREL"; - break; - } - case MachO::X86_64_RELOC_SUBTRACTOR: { - DataRefImpl RelNext = Rel; - Obj->moveRelocationNext(RelNext); - MachO::any_relocation_info RENext = Obj->getRelocation(RelNext); - - // X86_64_RELOC_SUBTRACTOR must be followed by a relocation of type - // X86_64_RELOC_UNSIGNED. - // NOTE: Scattered relocations don't exist on x86_64. - unsigned RType = Obj->getAnyRelocationType(RENext); - if (RType != MachO::X86_64_RELOC_UNSIGNED) - report_error(Obj->getFileName(), "Expected X86_64_RELOC_UNSIGNED after " - "X86_64_RELOC_SUBTRACTOR."); - - // The X86_64_RELOC_UNSIGNED contains the minuend symbol; - // X86_64_RELOC_SUBTRACTOR contains the subtrahend. - printRelocationTargetName(Obj, RENext, Fmt); - Fmt << "-"; - printRelocationTargetName(Obj, RE, Fmt); - break; - } - case MachO::X86_64_RELOC_TLV: - printRelocationTargetName(Obj, RE, Fmt); - Fmt << "@TLV"; - if (IsPCRel) - Fmt << "P"; - break; - case MachO::X86_64_RELOC_SIGNED_1: - printRelocationTargetName(Obj, RE, Fmt); - Fmt << "-1"; - break; - case MachO::X86_64_RELOC_SIGNED_2: - printRelocationTargetName(Obj, RE, Fmt); - Fmt << "-2"; - break; - case MachO::X86_64_RELOC_SIGNED_4: - printRelocationTargetName(Obj, RE, Fmt); - Fmt << "-4"; - break; - default: - printRelocationTargetName(Obj, RE, Fmt); - break; - } - // X86 and ARM share some relocation types in common. - } else if (Arch == Triple::x86 || Arch == Triple::arm || - Arch == Triple::ppc) { - // Generic relocation types... - switch (Type) { - case MachO::GENERIC_RELOC_PAIR: // prints no info - return std::error_code(); - case MachO::GENERIC_RELOC_SECTDIFF: { - DataRefImpl RelNext = Rel; - Obj->moveRelocationNext(RelNext); - MachO::any_relocation_info RENext = Obj->getRelocation(RelNext); - - // X86 sect diff's must be followed by a relocation of type - // GENERIC_RELOC_PAIR. - unsigned RType = Obj->getAnyRelocationType(RENext); - - if (RType != MachO::GENERIC_RELOC_PAIR) - report_error(Obj->getFileName(), "Expected GENERIC_RELOC_PAIR after " - "GENERIC_RELOC_SECTDIFF."); - - printRelocationTargetName(Obj, RE, Fmt); - Fmt << "-"; - printRelocationTargetName(Obj, RENext, Fmt); - break; - } - } - - if (Arch == Triple::x86 || Arch == Triple::ppc) { - switch (Type) { - case MachO::GENERIC_RELOC_LOCAL_SECTDIFF: { - DataRefImpl RelNext = Rel; - Obj->moveRelocationNext(RelNext); - MachO::any_relocation_info RENext = Obj->getRelocation(RelNext); - - // X86 sect diff's must be followed by a relocation of type - // GENERIC_RELOC_PAIR. - unsigned RType = Obj->getAnyRelocationType(RENext); - if (RType != MachO::GENERIC_RELOC_PAIR) - report_error(Obj->getFileName(), "Expected GENERIC_RELOC_PAIR after " - "GENERIC_RELOC_LOCAL_SECTDIFF."); - - printRelocationTargetName(Obj, RE, Fmt); - Fmt << "-"; - printRelocationTargetName(Obj, RENext, Fmt); - break; - } - case MachO::GENERIC_RELOC_TLV: { - printRelocationTargetName(Obj, RE, Fmt); - Fmt << "@TLV"; - if (IsPCRel) - Fmt << "P"; - break; - } - default: - printRelocationTargetName(Obj, RE, Fmt); - } - } else { // ARM-specific relocations - switch (Type) { - case MachO::ARM_RELOC_HALF: - case MachO::ARM_RELOC_HALF_SECTDIFF: { - // Half relocations steal a bit from the length field to encode - // whether this is an upper16 or a lower16 relocation. - bool isUpper = (Obj->getAnyRelocationLength(RE) & 0x1) == 1; - - if (isUpper) - Fmt << ":upper16:("; - else - Fmt << ":lower16:("; - printRelocationTargetName(Obj, RE, Fmt); - - DataRefImpl RelNext = Rel; - Obj->moveRelocationNext(RelNext); - MachO::any_relocation_info RENext = Obj->getRelocation(RelNext); - - // ARM half relocs must be followed by a relocation of type - // ARM_RELOC_PAIR. - unsigned RType = Obj->getAnyRelocationType(RENext); - if (RType != MachO::ARM_RELOC_PAIR) - report_error(Obj->getFileName(), "Expected ARM_RELOC_PAIR after " - "ARM_RELOC_HALF"); - - // NOTE: The half of the target virtual address is stashed in the - // address field of the secondary relocation, but we can't reverse - // engineer the constant offset from it without decoding the movw/movt - // instruction to find the other half in its immediate field. - - // ARM_RELOC_HALF_SECTDIFF encodes the second section in the - // symbol/section pointer of the follow-on relocation. - if (Type == MachO::ARM_RELOC_HALF_SECTDIFF) { - Fmt << "-"; - printRelocationTargetName(Obj, RENext, Fmt); - } - - Fmt << ")"; - break; - } - default: { printRelocationTargetName(Obj, RE, Fmt); } - } - } - } else - printRelocationTargetName(Obj, RE, Fmt); - - Fmt.flush(); - Result.append(FmtBuf.begin(), FmtBuf.end()); - return std::error_code(); -} - -static std::error_code getRelocationValueString(const RelocationRef &Rel, - SmallVectorImpl<char> &Result) { +static Error getRelocationValueString(const RelocationRef &Rel, + SmallVectorImpl<char> &Result) { const ObjectFile *Obj = Rel.getObject(); if (auto *ELF = dyn_cast<ELFObjectFileBase>(Obj)) - return getRelocationValueString(ELF, Rel, Result); + return getELFRelocationValueString(ELF, Rel, Result); if (auto *COFF = dyn_cast<COFFObjectFile>(Obj)) - return getRelocationValueString(COFF, Rel, Result); + return getCOFFRelocationValueString(COFF, Rel, Result); if (auto *Wasm = dyn_cast<WasmObjectFile>(Obj)) - return getRelocationValueString(Wasm, Rel, Result); + return getWasmRelocationValueString(Wasm, Rel, Result); if (auto *MachO = dyn_cast<MachOObjectFile>(Obj)) - return getRelocationValueString(MachO, Rel, Result); + return getMachORelocationValueString(MachO, Rel, Result); llvm_unreachable("unknown object file format"); } @@ -928,13 +559,15 @@ private: public: SourcePrinter() = default; SourcePrinter(const ObjectFile *Obj, StringRef DefaultArch) : Obj(Obj) { - symbolize::LLVMSymbolizer::Options SymbolizerOpts( - DILineInfoSpecifier::FunctionNameKind::None, true, false, false, - DefaultArch); + symbolize::LLVMSymbolizer::Options SymbolizerOpts; + SymbolizerOpts.PrintFunctions = DILineInfoSpecifier::FunctionNameKind::None; + SymbolizerOpts.Demangle = false; + SymbolizerOpts.DefaultArch = DefaultArch; Symbolizer.reset(new symbolize::LLVMSymbolizer(SymbolizerOpts)); } virtual ~SourcePrinter() = default; - virtual void printSourceLine(raw_ostream &OS, uint64_t Address, + virtual void printSourceLine(raw_ostream &OS, + object::SectionedAddress Address, StringRef Delimiter = "; "); }; @@ -949,35 +582,37 @@ bool SourcePrinter::cacheSource(const DILineInfo &LineInfo) { Buffer = std::move(*BufferOrError); } // Chomp the file to get lines - size_t BufferSize = Buffer->getBufferSize(); - const char *BufferStart = Buffer->getBufferStart(); - for (const char *Start = BufferStart, *End = BufferStart; - End < BufferStart + BufferSize; End++) - if (*End == '\n' || End == BufferStart + BufferSize - 1 || - (*End == '\r' && *(End + 1) == '\n')) { - LineCache[LineInfo.FileName].push_back(StringRef(Start, End - Start)); - if (*End == '\r') - End++; - Start = End + 1; + const char *BufferStart = Buffer->getBufferStart(), + *BufferEnd = Buffer->getBufferEnd(); + std::vector<StringRef> &Lines = LineCache[LineInfo.FileName]; + const char *Start = BufferStart; + for (const char *I = BufferStart; I != BufferEnd; ++I) + if (*I == '\n') { + Lines.emplace_back(Start, I - Start - (BufferStart < I && I[-1] == '\r')); + Start = I + 1; } + if (Start < BufferEnd) + Lines.emplace_back(Start, BufferEnd - Start); SourceCache[LineInfo.FileName] = std::move(Buffer); return true; } -void SourcePrinter::printSourceLine(raw_ostream &OS, uint64_t Address, +void SourcePrinter::printSourceLine(raw_ostream &OS, + object::SectionedAddress Address, StringRef Delimiter) { if (!Symbolizer) return; + DILineInfo LineInfo = DILineInfo(); - auto ExpectecLineInfo = - Symbolizer->symbolizeCode(Obj->getFileName(), Address); - if (!ExpectecLineInfo) - consumeError(ExpectecLineInfo.takeError()); + auto ExpectedLineInfo = Symbolizer->symbolizeCode(*Obj, Address); + if (!ExpectedLineInfo) + consumeError(ExpectedLineInfo.takeError()); else - LineInfo = *ExpectecLineInfo; + LineInfo = *ExpectedLineInfo; - if ((LineInfo.FileName == "<invalid>") || OldLineInfo.Line == LineInfo.Line || - LineInfo.Line == 0) + if ((LineInfo.FileName == "<invalid>") || LineInfo.Line == 0 || + ((OldLineInfo.Line == LineInfo.Line) && + (OldLineInfo.FileName == LineInfo.FileName))) return; if (PrintLines) @@ -986,53 +621,79 @@ void SourcePrinter::printSourceLine(raw_ostream &OS, uint64_t Address, if (SourceCache.find(LineInfo.FileName) == SourceCache.end()) if (!cacheSource(LineInfo)) return; - auto FileBuffer = SourceCache.find(LineInfo.FileName); - if (FileBuffer != SourceCache.end()) { - auto LineBuffer = LineCache.find(LineInfo.FileName); - if (LineBuffer != LineCache.end()) { - if (LineInfo.Line > LineBuffer->second.size()) - return; - // Vector begins at 0, line numbers are non-zero - OS << Delimiter << LineBuffer->second[LineInfo.Line - 1].ltrim() - << "\n"; - } + auto LineBuffer = LineCache.find(LineInfo.FileName); + if (LineBuffer != LineCache.end()) { + if (LineInfo.Line > LineBuffer->second.size()) + return; + // Vector begins at 0, line numbers are non-zero + OS << Delimiter << LineBuffer->second[LineInfo.Line - 1] << '\n'; } } OldLineInfo = LineInfo; } +static bool isAArch64Elf(const ObjectFile *Obj) { + const auto *Elf = dyn_cast<ELFObjectFileBase>(Obj); + return Elf && Elf->getEMachine() == ELF::EM_AARCH64; +} + static bool isArmElf(const ObjectFile *Obj) { - return (Obj->isELF() && - (Obj->getArch() == Triple::aarch64 || - Obj->getArch() == Triple::aarch64_be || - Obj->getArch() == Triple::arm || Obj->getArch() == Triple::armeb || - Obj->getArch() == Triple::thumb || - Obj->getArch() == Triple::thumbeb)); + const auto *Elf = dyn_cast<ELFObjectFileBase>(Obj); + return Elf && Elf->getEMachine() == ELF::EM_ARM; +} + +static bool hasMappingSymbols(const ObjectFile *Obj) { + return isArmElf(Obj) || isAArch64Elf(Obj); +} + +static void printRelocation(const RelocationRef &Rel, uint64_t Address, + bool Is64Bits) { + StringRef Fmt = Is64Bits ? "\t\t%016" PRIx64 ": " : "\t\t\t%08" PRIx64 ": "; + SmallString<16> Name; + SmallString<32> Val; + Rel.getTypeName(Name); + error(getRelocationValueString(Rel, Val)); + outs() << format(Fmt.data(), Address) << Name << "\t" << Val << "\n"; } class PrettyPrinter { public: virtual ~PrettyPrinter() = default; virtual void printInst(MCInstPrinter &IP, const MCInst *MI, - ArrayRef<uint8_t> Bytes, uint64_t Address, - raw_ostream &OS, StringRef Annot, - MCSubtargetInfo const &STI, SourcePrinter *SP, + ArrayRef<uint8_t> Bytes, + object::SectionedAddress Address, raw_ostream &OS, + StringRef Annot, MCSubtargetInfo const &STI, + SourcePrinter *SP, std::vector<RelocationRef> *Rels = nullptr) { if (SP && (PrintSource || PrintLines)) SP->printSourceLine(OS, Address); - if (!NoLeadingAddr) - OS << format("%8" PRIx64 ":", Address); - if (!NoShowRawInsn) { - OS << "\t"; - dumpBytes(Bytes, OS); + + { + formatted_raw_ostream FOS(OS); + if (!NoLeadingAddr) + FOS << format("%8" PRIx64 ":", Address.Address); + if (!NoShowRawInsn) { + FOS << ' '; + dumpBytes(Bytes, FOS); + } + FOS.flush(); + // The output of printInst starts with a tab. Print some spaces so that + // the tab has 1 column and advances to the target tab stop. + unsigned TabStop = NoShowRawInsn ? 16 : 40; + unsigned Column = FOS.getColumn(); + FOS.indent(Column < TabStop - 1 ? TabStop - 1 - Column : 7 - Column % 8); + + // The dtor calls flush() to ensure the indent comes before printInst(). } + if (MI) IP.printInst(MI, OS, "", STI); else - OS << " <unknown>"; + OS << "\t<unknown>"; } }; PrettyPrinter PrettyPrinterInst; + class HexagonPrettyPrinter : public PrettyPrinter { public: void printLead(ArrayRef<uint8_t> Bytes, uint64_t Address, @@ -1044,17 +705,17 @@ public: if (!NoShowRawInsn) { OS << "\t"; dumpBytes(Bytes.slice(0, 4), OS); - OS << format("%08" PRIx32, opcode); + OS << format("\t%08" PRIx32, opcode); } } void printInst(MCInstPrinter &IP, const MCInst *MI, ArrayRef<uint8_t> Bytes, - uint64_t Address, raw_ostream &OS, StringRef Annot, - MCSubtargetInfo const &STI, SourcePrinter *SP, + object::SectionedAddress Address, raw_ostream &OS, + StringRef Annot, MCSubtargetInfo const &STI, SourcePrinter *SP, std::vector<RelocationRef> *Rels) override { if (SP && (PrintSource || PrintLines)) SP->printSourceLine(OS, Address, ""); if (!MI) { - printLead(Bytes, Address, OS); + printLead(Bytes, Address.Address, OS); OS << " <unknown>"; return; } @@ -1070,21 +731,15 @@ public: auto HeadTail = PacketBundle.first.split('\n'); auto Preamble = " { "; auto Separator = ""; - StringRef Fmt = "\t\t\t%08" PRIx64 ": "; - std::vector<RelocationRef>::const_iterator RelCur = Rels->begin(); - std::vector<RelocationRef>::const_iterator RelEnd = Rels->end(); // Hexagon's packets require relocations to be inline rather than // clustered at the end of the packet. + std::vector<RelocationRef>::const_iterator RelCur = Rels->begin(); + std::vector<RelocationRef>::const_iterator RelEnd = Rels->end(); auto PrintReloc = [&]() -> void { - while ((RelCur != RelEnd) && (RelCur->getOffset() <= Address)) { - if (RelCur->getOffset() == Address) { - SmallString<16> Name; - SmallString<32> Val; - RelCur->getTypeName(Name); - error(getRelocationValueString(*RelCur, Val)); - OS << Separator << format(Fmt.data(), Address) << Name << "\t" << Val - << "\n"; + while ((RelCur != RelEnd) && (RelCur->getOffset() <= Address.Address)) { + if (RelCur->getOffset() == Address.Address) { + printRelocation(*RelCur, Address.Address, false); return; } ++RelCur; @@ -1096,7 +751,7 @@ public: Separator = "\n"; if (SP && (PrintSource || PrintLines)) SP->printSourceLine(OS, Address, ""); - printLead(Bytes, Address, OS); + printLead(Bytes, Address.Address, OS); OS << Preamble; Preamble = " "; StringRef Inst; @@ -1114,7 +769,7 @@ public: OS << " } " << PacketBundle.second; PrintReloc(); Bytes = Bytes.slice(4); - Address += 4; + Address.Address += 4; } } }; @@ -1123,14 +778,12 @@ HexagonPrettyPrinter HexagonPrettyPrinterInst; class AMDGCNPrettyPrinter : public PrettyPrinter { public: void printInst(MCInstPrinter &IP, const MCInst *MI, ArrayRef<uint8_t> Bytes, - uint64_t Address, raw_ostream &OS, StringRef Annot, - MCSubtargetInfo const &STI, SourcePrinter *SP, + object::SectionedAddress Address, raw_ostream &OS, + StringRef Annot, MCSubtargetInfo const &STI, SourcePrinter *SP, std::vector<RelocationRef> *Rels) override { if (SP && (PrintSource || PrintLines)) SP->printSourceLine(OS, Address); - typedef support::ulittle32_t U32; - if (MI) { SmallString<40> InstStr; raw_svector_ostream IS(InstStr); @@ -1144,7 +797,7 @@ public: // remaining if (Bytes.size() >= 4) { OS << format("\t.long 0x%08" PRIx32 " ", - static_cast<uint32_t>(*reinterpret_cast<const U32*>(Bytes.data()))); + support::endian::read32<support::little>(Bytes.data())); OS.indent(42); } else { OS << format("\t.byte 0x%02" PRIx8, Bytes[0]); @@ -1154,20 +807,21 @@ public: } } - OS << format("// %012" PRIX64 ": ", Address); - if (Bytes.size() >=4) { - for (auto D : makeArrayRef(reinterpret_cast<const U32*>(Bytes.data()), - Bytes.size() / sizeof(U32))) - // D should be explicitly casted to uint32_t here as it is passed - // by format to snprintf as vararg. - OS << format("%08" PRIX32 " ", static_cast<uint32_t>(D)); + OS << format("// %012" PRIX64 ":", Address.Address); + if (Bytes.size() >= 4) { + // D should be casted to uint32_t here as it is passed by format to + // snprintf as vararg. + for (uint32_t D : makeArrayRef( + reinterpret_cast<const support::little32_t *>(Bytes.data()), + Bytes.size() / 4)) + OS << format(" %08" PRIX32, D); } else { - for (unsigned int i = 0; i < Bytes.size(); i++) - OS << format("%02" PRIX8 " ", Bytes[i]); + for (unsigned char B : Bytes) + OS << format(" %02" PRIX8, B); } if (!Annot.empty()) - OS << "// " << Annot; + OS << " // " << Annot; } }; AMDGCNPrettyPrinter AMDGCNPrettyPrinterInst; @@ -1175,13 +829,13 @@ AMDGCNPrettyPrinter AMDGCNPrettyPrinterInst; class BPFPrettyPrinter : public PrettyPrinter { public: void printInst(MCInstPrinter &IP, const MCInst *MI, ArrayRef<uint8_t> Bytes, - uint64_t Address, raw_ostream &OS, StringRef Annot, - MCSubtargetInfo const &STI, SourcePrinter *SP, + object::SectionedAddress Address, raw_ostream &OS, + StringRef Annot, MCSubtargetInfo const &STI, SourcePrinter *SP, std::vector<RelocationRef> *Rels) override { if (SP && (PrintSource || PrintLines)) SP->printSourceLine(OS, Address); if (!NoLeadingAddr) - OS << format("%8" PRId64 ":", Address / 8); + OS << format("%8" PRId64 ":", Address.Address / 8); if (!NoShowRawInsn) { OS << "\t"; dumpBytes(Bytes, OS); @@ -1189,7 +843,7 @@ public: if (MI) IP.printInst(MI, OS, "", STI); else - OS << " <unknown>"; + OS << "\t<unknown>"; } }; BPFPrettyPrinter BPFPrettyPrinterInst; @@ -1227,27 +881,25 @@ addDynamicElfSymbols(const ELFObjectFile<ELFT> *Obj, std::map<SectionRef, SectionSymbolsTy> &AllSymbols) { for (auto Symbol : Obj->getDynamicSymbolIterators()) { uint8_t SymbolType = Symbol.getELFType(); - if (SymbolType != ELF::STT_FUNC || Symbol.getSize() == 0) + if (SymbolType == ELF::STT_SECTION) continue; - Expected<uint64_t> AddressOrErr = Symbol.getAddress(); - if (!AddressOrErr) - report_error(Obj->getFileName(), AddressOrErr.takeError()); + uint64_t Address = unwrapOrError(Symbol.getAddress(), Obj->getFileName()); + // ELFSymbolRef::getAddress() returns size instead of value for common + // symbols which is not desirable for disassembly output. Overriding. + if (SymbolType == ELF::STT_COMMON) + Address = Obj->getSymbol(Symbol.getRawDataRefImpl())->st_value; - Expected<StringRef> Name = Symbol.getName(); - if (!Name) - report_error(Obj->getFileName(), Name.takeError()); - if (Name->empty()) + StringRef Name = unwrapOrError(Symbol.getName(), Obj->getFileName()); + if (Name.empty()) continue; - Expected<section_iterator> SectionOrErr = Symbol.getSection(); - if (!SectionOrErr) - report_error(Obj->getFileName(), SectionOrErr.takeError()); - section_iterator SecI = *SectionOrErr; + section_iterator SecI = + unwrapOrError(Symbol.getSection(), Obj->getFileName()); if (SecI == Obj->section_end()) continue; - AllSymbols[*SecI].emplace_back(*AddressOrErr, *Name, SymbolType); + AllSymbols[*SecI].emplace_back(Address, Name, SymbolType); } } @@ -1285,14 +937,10 @@ static void addPltEntries(const ObjectFile *Obj, SymbolRef Symbol(PltEntry.first, ElfObj); uint8_t SymbolType = getElfSymbolType(Obj, Symbol); - Expected<StringRef> NameOrErr = Symbol.getName(); - if (!NameOrErr) - report_error(Obj->getFileName(), NameOrErr.takeError()); - if (NameOrErr->empty()) - continue; - StringRef Name = Saver.save((*NameOrErr + "@plt").str()); - - AllSymbols[*Plt].emplace_back(PltEntry.second, Name, SymbolType); + StringRef Name = unwrapOrError(Symbol.getName(), Obj->getFileName()); + if (!Name.empty()) + AllSymbols[*Plt].emplace_back( + PltEntry.second, Saver.save((Name + "@plt").str()), SymbolType); } } } @@ -1301,10 +949,6 @@ static void addPltEntries(const ObjectFile *Obj, // returns the number of zero bytes that can be skipped when dumping the // disassembly of the instructions in Buf. static size_t countSkippableZeroBytes(ArrayRef<uint8_t> Buf) { - // When -z or --disassemble-zeroes are given we always dissasemble them. - if (DisassembleZeroes) - return 0; - // Find the number of leading zeroes. size_t N = 0; while (N < Buf.size() && !Buf[N]) @@ -1320,108 +964,160 @@ static size_t countSkippableZeroBytes(ArrayRef<uint8_t> Buf) { return N & ~0x3; } -static void disassembleObject(const ObjectFile *Obj, bool InlineRelocs) { - if (StartAddress > StopAddress) - error("Start address should be less than stop address"); - - const Target *TheTarget = getTarget(Obj); - - // Package up features to be passed to target/subtarget - SubtargetFeatures Features = Obj->getFeatures(); - if (!MAttrs.empty()) - for (unsigned I = 0; I != MAttrs.size(); ++I) - Features.AddFeature(MAttrs[I]); - - std::unique_ptr<const MCRegisterInfo> MRI( - TheTarget->createMCRegInfo(TripleName)); - if (!MRI) - report_error(Obj->getFileName(), "no register info for target " + - TripleName); - - // Set up disassembler. - std::unique_ptr<const MCAsmInfo> AsmInfo( - TheTarget->createMCAsmInfo(*MRI, TripleName)); - if (!AsmInfo) - report_error(Obj->getFileName(), "no assembly info for target " + - TripleName); - std::unique_ptr<const MCSubtargetInfo> STI( - TheTarget->createMCSubtargetInfo(TripleName, MCPU, Features.getString())); - if (!STI) - report_error(Obj->getFileName(), "no subtarget info for target " + - TripleName); - std::unique_ptr<const MCInstrInfo> MII(TheTarget->createMCInstrInfo()); - if (!MII) - report_error(Obj->getFileName(), "no instruction info for target " + - TripleName); - MCObjectFileInfo MOFI; - MCContext Ctx(AsmInfo.get(), MRI.get(), &MOFI); - // FIXME: for now initialize MCObjectFileInfo with default values - MOFI.InitMCObjectFileInfo(Triple(TripleName), false, Ctx); - - std::unique_ptr<MCDisassembler> DisAsm( - TheTarget->createMCDisassembler(*STI, Ctx)); - if (!DisAsm) - report_error(Obj->getFileName(), "no disassembler for target " + - TripleName); +// Returns a map from sections to their relocations. +static std::map<SectionRef, std::vector<RelocationRef>> +getRelocsMap(object::ObjectFile const &Obj) { + std::map<SectionRef, std::vector<RelocationRef>> Ret; + for (SectionRef Sec : Obj.sections()) { + section_iterator Relocated = Sec.getRelocatedSection(); + if (Relocated == Obj.section_end() || !shouldKeep(*Relocated)) + continue; + std::vector<RelocationRef> &V = Ret[*Relocated]; + for (const RelocationRef &R : Sec.relocations()) + V.push_back(R); + // Sort relocations by address. + llvm::stable_sort(V, isRelocAddressLess); + } + return Ret; +} - std::unique_ptr<const MCInstrAnalysis> MIA( - TheTarget->createMCInstrAnalysis(MII.get())); +// Used for --adjust-vma to check if address should be adjusted by the +// specified value for a given section. +// For ELF we do not adjust non-allocatable sections like debug ones, +// because they are not loadable. +// TODO: implement for other file formats. +static bool shouldAdjustVA(const SectionRef &Section) { + const ObjectFile *Obj = Section.getObject(); + if (isa<object::ELFObjectFileBase>(Obj)) + return ELFSectionRef(Section).getFlags() & ELF::SHF_ALLOC; + return false; +} - int AsmPrinterVariant = AsmInfo->getAssemblerDialect(); - std::unique_ptr<MCInstPrinter> IP(TheTarget->createMCInstPrinter( - Triple(TripleName), AsmPrinterVariant, *AsmInfo, *MII, *MRI)); - if (!IP) - report_error(Obj->getFileName(), "no instruction printer for target " + - TripleName); - IP->setPrintImmHex(PrintImmHex); - PrettyPrinter &PIP = selectPrettyPrinter(Triple(TripleName)); - StringRef Fmt = Obj->getBytesInAddress() > 4 ? "\t\t%016" PRIx64 ": " : - "\t\t\t%08" PRIx64 ": "; +typedef std::pair<uint64_t, char> MappingSymbolPair; +static char getMappingSymbolKind(ArrayRef<MappingSymbolPair> MappingSymbols, + uint64_t Address) { + auto It = + partition_point(MappingSymbols, [Address](const MappingSymbolPair &Val) { + return Val.first <= Address; + }); + // Return zero for any address before the first mapping symbol; this means + // we should use the default disassembly mode, depending on the target. + if (It == MappingSymbols.begin()) + return '\x00'; + return (It - 1)->second; +} - SourcePrinter SP(Obj, TheTarget->getName()); +static uint64_t +dumpARMELFData(uint64_t SectionAddr, uint64_t Index, uint64_t End, + const ObjectFile *Obj, ArrayRef<uint8_t> Bytes, + ArrayRef<MappingSymbolPair> MappingSymbols) { + support::endianness Endian = + Obj->isLittleEndian() ? support::little : support::big; + while (Index < End) { + outs() << format("%8" PRIx64 ":", SectionAddr + Index); + outs() << "\t"; + if (Index + 4 <= End) { + dumpBytes(Bytes.slice(Index, 4), outs()); + outs() << "\t.word\t" + << format_hex( + support::endian::read32(Bytes.data() + Index, Endian), 10); + Index += 4; + } else if (Index + 2 <= End) { + dumpBytes(Bytes.slice(Index, 2), outs()); + outs() << "\t\t.short\t" + << format_hex( + support::endian::read16(Bytes.data() + Index, Endian), 6); + Index += 2; + } else { + dumpBytes(Bytes.slice(Index, 1), outs()); + outs() << "\t\t.byte\t" << format_hex(Bytes[0], 4); + ++Index; + } + outs() << "\n"; + if (getMappingSymbolKind(MappingSymbols, Index) != 'd') + break; + } + return Index; +} - // Create a mapping, RelocSecs = SectionRelocMap[S], where sections - // in RelocSecs contain the relocations for section S. - std::error_code EC; - std::map<SectionRef, SmallVector<SectionRef, 1>> SectionRelocMap; - for (const SectionRef &Section : ToolSectionFilter(*Obj)) { - section_iterator Sec2 = Section.getRelocatedSection(); - if (Sec2 != Obj->section_end()) - SectionRelocMap[*Sec2].push_back(Section); +static void dumpELFData(uint64_t SectionAddr, uint64_t Index, uint64_t End, + ArrayRef<uint8_t> Bytes) { + // print out data up to 8 bytes at a time in hex and ascii + uint8_t AsciiData[9] = {'\0'}; + uint8_t Byte; + int NumBytes = 0; + + for (; Index < End; ++Index) { + if (NumBytes == 0) + outs() << format("%8" PRIx64 ":", SectionAddr + Index); + Byte = Bytes.slice(Index)[0]; + outs() << format(" %02x", Byte); + AsciiData[NumBytes] = isPrint(Byte) ? Byte : '.'; + + uint8_t IndentOffset = 0; + NumBytes++; + if (Index == End - 1 || NumBytes > 8) { + // Indent the space for less than 8 bytes data. + // 2 spaces for byte and one for space between bytes + IndentOffset = 3 * (8 - NumBytes); + for (int Excess = NumBytes; Excess < 8; Excess++) + AsciiData[Excess] = '\0'; + NumBytes = 8; + } + if (NumBytes == 8) { + AsciiData[8] = '\0'; + outs() << std::string(IndentOffset, ' ') << " "; + outs() << reinterpret_cast<char *>(AsciiData); + outs() << '\n'; + NumBytes = 0; + } } +} + +static void disassembleObject(const Target *TheTarget, const ObjectFile *Obj, + MCContext &Ctx, MCDisassembler *PrimaryDisAsm, + MCDisassembler *SecondaryDisAsm, + const MCInstrAnalysis *MIA, MCInstPrinter *IP, + const MCSubtargetInfo *PrimarySTI, + const MCSubtargetInfo *SecondarySTI, + PrettyPrinter &PIP, + SourcePrinter &SP, bool InlineRelocs) { + const MCSubtargetInfo *STI = PrimarySTI; + MCDisassembler *DisAsm = PrimaryDisAsm; + bool PrimaryIsThumb = false; + if (isArmElf(Obj)) + PrimaryIsThumb = STI->checkFeatures("+thumb-mode"); + + std::map<SectionRef, std::vector<RelocationRef>> RelocMap; + if (InlineRelocs) + RelocMap = getRelocsMap(*Obj); + bool Is64Bits = Obj->getBytesInAddress() > 4; // Create a mapping from virtual address to symbol name. This is used to // pretty print the symbols while disassembling. std::map<SectionRef, SectionSymbolsTy> AllSymbols; SectionSymbolsTy AbsoluteSymbols; + const StringRef FileName = Obj->getFileName(); for (const SymbolRef &Symbol : Obj->symbols()) { - Expected<uint64_t> AddressOrErr = Symbol.getAddress(); - if (!AddressOrErr) - report_error(Obj->getFileName(), AddressOrErr.takeError()); - uint64_t Address = *AddressOrErr; - - Expected<StringRef> Name = Symbol.getName(); - if (!Name) - report_error(Obj->getFileName(), Name.takeError()); - if (Name->empty()) - continue; + uint64_t Address = unwrapOrError(Symbol.getAddress(), FileName); - Expected<section_iterator> SectionOrErr = Symbol.getSection(); - if (!SectionOrErr) - report_error(Obj->getFileName(), SectionOrErr.takeError()); + StringRef Name = unwrapOrError(Symbol.getName(), FileName); + if (Name.empty()) + continue; uint8_t SymbolType = ELF::STT_NOTYPE; - if (Obj->isELF()) + if (Obj->isELF()) { SymbolType = getElfSymbolType(Obj, Symbol); + if (SymbolType == ELF::STT_SECTION) + continue; + } - section_iterator SecI = *SectionOrErr; + section_iterator SecI = unwrapOrError(Symbol.getSection(), FileName); if (SecI != Obj->section_end()) - AllSymbols[*SecI].emplace_back(Address, *Name, SymbolType); + AllSymbols[*SecI].emplace_back(Address, Name, SymbolType); else - AbsoluteSymbols.emplace_back(Address, *Name, SymbolType); - - + AbsoluteSymbols.emplace_back(Address, Name, SymbolType); } if (AllSymbols.empty() && Obj->isELF()) addDynamicElfSymbols(Obj, AllSymbols); @@ -1448,31 +1144,28 @@ static void disassembleObject(const ObjectFile *Obj, bool InlineRelocs) { error(ExportEntry.getExportRVA(RVA)); uint64_t VA = COFFObj->getImageBase() + RVA; - auto Sec = std::upper_bound( - SectionAddresses.begin(), SectionAddresses.end(), VA, - [](uint64_t LHS, const std::pair<uint64_t, SectionRef> &RHS) { - return LHS < RHS.first; + auto Sec = partition_point( + SectionAddresses, [VA](const std::pair<uint64_t, SectionRef> &O) { + return O.first <= VA; }); - if (Sec != SectionAddresses.begin()) + if (Sec != SectionAddresses.begin()) { --Sec; - else - Sec = SectionAddresses.end(); - - if (Sec != SectionAddresses.end()) AllSymbols[Sec->second].emplace_back(VA, Name, ELF::STT_NOTYPE); - else + } else AbsoluteSymbols.emplace_back(VA, Name, ELF::STT_NOTYPE); } } // Sort all the symbols, this allows us to use a simple binary search to find // a symbol near an address. + StringSet<> FoundDisasmFuncsSet; for (std::pair<const SectionRef, SectionSymbolsTy> &SecSyms : AllSymbols) array_pod_sort(SecSyms.second.begin(), SecSyms.second.end()); array_pod_sort(AbsoluteSymbols.begin(), AbsoluteSymbols.end()); for (const SectionRef &Section : ToolSectionFilter(*Obj)) { - if (!DisassembleAll && (!Section.isText() || Section.isVirtual())) + if (FilterSections.empty() && !DisassembleAll && + (!Section.isText() || Section.isVirtual())) continue; uint64_t SectionAddr = Section.getAddress(); @@ -1482,25 +1175,23 @@ static void disassembleObject(const ObjectFile *Obj, bool InlineRelocs) { // Get the list of all the symbols in this section. SectionSymbolsTy &Symbols = AllSymbols[Section]; - std::vector<uint64_t> DataMappingSymsAddr; - std::vector<uint64_t> TextMappingSymsAddr; - if (isArmElf(Obj)) { + std::vector<MappingSymbolPair> MappingSymbols; + if (hasMappingSymbols(Obj)) { for (const auto &Symb : Symbols) { uint64_t Address = std::get<0>(Symb); StringRef Name = std::get<1>(Symb); if (Name.startswith("$d")) - DataMappingSymsAddr.push_back(Address - SectionAddr); + MappingSymbols.emplace_back(Address - SectionAddr, 'd'); if (Name.startswith("$x")) - TextMappingSymsAddr.push_back(Address - SectionAddr); + MappingSymbols.emplace_back(Address - SectionAddr, 'x'); if (Name.startswith("$a")) - TextMappingSymsAddr.push_back(Address - SectionAddr); + MappingSymbols.emplace_back(Address - SectionAddr, 'a'); if (Name.startswith("$t")) - TextMappingSymsAddr.push_back(Address - SectionAddr); + MappingSymbols.emplace_back(Address - SectionAddr, 't'); } } - llvm::sort(DataMappingSymsAddr); - llvm::sort(TextMappingSymsAddr); + llvm::sort(MappingSymbols); if (Obj->isELF() && Obj->getArch() == Triple::amdgcn) { // AMDGPU disassembler uses symbolizer for printing labels @@ -1514,19 +1205,6 @@ static void disassembleObject(const ObjectFile *Obj, bool InlineRelocs) { } } - // Make a list of all the relocations for this section. - std::vector<RelocationRef> Rels; - if (InlineRelocs) { - for (const SectionRef &RelocSec : SectionRelocMap[Section]) { - for (const RelocationRef &Reloc : RelocSec.relocations()) { - Rels.push_back(Reloc); - } - } - } - - // Sort relocations by address. - llvm::sort(Rels, isRelocAddressLess); - StringRef SegmentName = ""; if (const MachOObjectFile *MachO = dyn_cast<const MachOObjectFile>(Obj)) { DataRefImpl DR = Section.getRawDataRefImpl(); @@ -1546,56 +1224,54 @@ static void disassembleObject(const ObjectFile *Obj, bool InlineRelocs) { SmallString<40> Comments; raw_svector_ostream CommentStream(Comments); - StringRef BytesStr; - error(Section.getContents(BytesStr)); - ArrayRef<uint8_t> Bytes(reinterpret_cast<const uint8_t *>(BytesStr.data()), - BytesStr.size()); + ArrayRef<uint8_t> Bytes = arrayRefFromStringRef( + unwrapOrError(Section.getContents(), Obj->getFileName())); + + uint64_t VMAAdjustment = 0; + if (shouldAdjustVA(Section)) + VMAAdjustment = AdjustVMA; uint64_t Size; uint64_t Index; bool PrintedSection = false; - + std::vector<RelocationRef> Rels = RelocMap[Section]; std::vector<RelocationRef>::const_iterator RelCur = Rels.begin(); std::vector<RelocationRef>::const_iterator RelEnd = Rels.end(); // Disassemble symbol by symbol. for (unsigned SI = 0, SE = Symbols.size(); SI != SE; ++SI) { - uint64_t Start = std::get<0>(Symbols[SI]) - SectionAddr; - // The end is either the section end or the beginning of the next - // symbol. - uint64_t End = (SI == SE - 1) - ? SectSize - : std::get<0>(Symbols[SI + 1]) - SectionAddr; - // Don't try to disassemble beyond the end of section contents. - if (End > SectSize) - End = SectSize; - // If this symbol has the same address as the next symbol, then skip it. - if (Start >= End) - continue; + std::string SymbolName = std::get<1>(Symbols[SI]).str(); + if (Demangle) + SymbolName = demangle(SymbolName); - // Check if we need to skip symbol - // Skip if the symbol's data is not between StartAddress and StopAddress - if (End + SectionAddr < StartAddress || - Start + SectionAddr > StopAddress) { + // Skip if --disassemble-functions is not empty and the symbol is not in + // the list. + if (!DisasmFuncsSet.empty() && !DisasmFuncsSet.count(SymbolName)) continue; - } - /// Skip if user requested specific symbols and this is not in the list - if (!DisasmFuncsSet.empty() && - !DisasmFuncsSet.count(std::get<1>(Symbols[SI]))) + uint64_t Start = std::get<0>(Symbols[SI]); + if (Start < SectionAddr || StopAddress <= Start) continue; + else + FoundDisasmFuncsSet.insert(SymbolName); + + // The end is the section end, the beginning of the next symbol, or + // --stop-address. + uint64_t End = std::min<uint64_t>(SectionAddr + SectSize, StopAddress); + if (SI + 1 < SE) + End = std::min(End, std::get<0>(Symbols[SI + 1])); + if (Start >= End || End <= StartAddress) + continue; + Start -= SectionAddr; + End -= SectionAddr; if (!PrintedSection) { PrintedSection = true; - outs() << "Disassembly of section "; + outs() << "\nDisassembly of section "; if (!SegmentName.empty()) outs() << SegmentName << ","; - outs() << SectionName << ':'; + outs() << SectionName << ":\n"; } - // Stop disassembly at the stop address specified - if (End + SectionAddr > StopAddress) - End = StopAddress - SectionAddr; - if (Obj->isELF() && Obj->getArch() == Triple::amdgcn) { if (std::get<2>(Symbols[SI]) == ELF::STT_AMDGPU_HSA_KERNEL) { // skip amd_kernel_code_t at the begining of kernel symbol (256 bytes) @@ -1615,13 +1291,10 @@ static void disassembleObject(const ObjectFile *Obj, bool InlineRelocs) { outs() << '\n'; if (!NoLeadingAddr) - outs() << format("%016" PRIx64 " ", SectionAddr + Start); + outs() << format(Is64Bits ? "%016" PRIx64 " " : "%08" PRIx64 " ", + SectionAddr + Start + VMAAdjustment); - StringRef SymbolName = std::get<1>(Symbols[SI]); - if (Demangle) - outs() << demangle(SymbolName) << ":\n"; - else - outs() << SymbolName << ":\n"; + outs() << SymbolName << ":\n"; // Don't print raw contents of a virtual section. A virtual section // doesn't have any contents in the file. @@ -1636,143 +1309,82 @@ static void disassembleObject(const ObjectFile *Obj, bool InlineRelocs) { raw_ostream &DebugOut = nulls(); #endif - for (Index = Start; Index < End; Index += Size) { - MCInst Inst; + // Some targets (like WebAssembly) have a special prelude at the start + // of each symbol. + DisAsm->onSymbolStart(SymbolName, Size, Bytes.slice(Start, End - Start), + SectionAddr + Start, DebugOut, CommentStream); + Start += Size; + + Index = Start; + if (SectionAddr < StartAddress) + Index = std::max<uint64_t>(Index, StartAddress - SectionAddr); + + // If there is a data/common symbol inside an ELF text section and we are + // only disassembling text (applicable all architectures), we are in a + // situation where we must print the data and not disassemble it. + if (Obj->isELF() && !DisassembleAll && Section.isText()) { + uint8_t SymTy = std::get<2>(Symbols[SI]); + if (SymTy == ELF::STT_OBJECT || SymTy == ELF::STT_COMMON) { + dumpELFData(SectionAddr, Index, End, Bytes); + Index = End; + } + } - if (Index + SectionAddr < StartAddress || - Index + SectionAddr > StopAddress) { - // skip byte by byte till StartAddress is reached - Size = 1; + bool CheckARMELFData = hasMappingSymbols(Obj) && + std::get<2>(Symbols[SI]) != ELF::STT_OBJECT && + !DisassembleAll; + while (Index < End) { + // ARM and AArch64 ELF binaries can interleave data and text in the + // same section. We rely on the markers introduced to understand what + // we need to dump. If the data marker is within a function, it is + // denoted as a word/short etc. + if (CheckARMELFData && + getMappingSymbolKind(MappingSymbols, Index) == 'd') { + Index = dumpARMELFData(SectionAddr, Index, End, Obj, Bytes, + MappingSymbols); continue; } - // AArch64 ELF binaries can interleave data and text in the - // same section. We rely on the markers introduced to - // understand what we need to dump. If the data marker is within a - // function, it is denoted as a word/short etc - if (isArmElf(Obj) && std::get<2>(Symbols[SI]) != ELF::STT_OBJECT && - !DisassembleAll) { - uint64_t Stride = 0; - - auto DAI = std::lower_bound(DataMappingSymsAddr.begin(), - DataMappingSymsAddr.end(), Index); - if (DAI != DataMappingSymsAddr.end() && *DAI == Index) { - // Switch to data. - while (Index < End) { - outs() << format("%8" PRIx64 ":", SectionAddr + Index); - outs() << "\t"; - if (Index + 4 <= End) { - Stride = 4; - dumpBytes(Bytes.slice(Index, 4), outs()); - outs() << "\t.word\t"; - uint32_t Data = 0; - if (Obj->isLittleEndian()) { - const auto Word = - reinterpret_cast<const support::ulittle32_t *>( - Bytes.data() + Index); - Data = *Word; - } else { - const auto Word = reinterpret_cast<const support::ubig32_t *>( - Bytes.data() + Index); - Data = *Word; - } - outs() << "0x" << format("%08" PRIx32, Data); - } else if (Index + 2 <= End) { - Stride = 2; - dumpBytes(Bytes.slice(Index, 2), outs()); - outs() << "\t\t.short\t"; - uint16_t Data = 0; - if (Obj->isLittleEndian()) { - const auto Short = - reinterpret_cast<const support::ulittle16_t *>( - Bytes.data() + Index); - Data = *Short; - } else { - const auto Short = - reinterpret_cast<const support::ubig16_t *>(Bytes.data() + - Index); - Data = *Short; - } - outs() << "0x" << format("%04" PRIx16, Data); - } else { - Stride = 1; - dumpBytes(Bytes.slice(Index, 1), outs()); - outs() << "\t\t.byte\t"; - outs() << "0x" << format("%02" PRIx8, Bytes.slice(Index, 1)[0]); - } - Index += Stride; - outs() << "\n"; - auto TAI = std::lower_bound(TextMappingSymsAddr.begin(), - TextMappingSymsAddr.end(), Index); - if (TAI != TextMappingSymsAddr.end() && *TAI == Index) - break; - } + + // When -z or --disassemble-zeroes are given we always dissasemble + // them. Otherwise we might want to skip zero bytes we see. + if (!DisassembleZeroes) { + uint64_t MaxOffset = End - Index; + // For -reloc: print zero blocks patched by relocations, so that + // relocations can be shown in the dump. + if (RelCur != RelEnd) + MaxOffset = RelCur->getOffset() - Index; + + if (size_t N = + countSkippableZeroBytes(Bytes.slice(Index, MaxOffset))) { + outs() << "\t\t..." << '\n'; + Index += N; + continue; } } - // If there is a data symbol inside an ELF text section and we are only - // disassembling text (applicable all architectures), - // we are in a situation where we must print the data and not - // disassemble it. - if (Obj->isELF() && std::get<2>(Symbols[SI]) == ELF::STT_OBJECT && - !DisassembleAll && Section.isText()) { - // print out data up to 8 bytes at a time in hex and ascii - uint8_t AsciiData[9] = {'\0'}; - uint8_t Byte; - int NumBytes = 0; - - for (Index = Start; Index < End; Index += 1) { - if (((SectionAddr + Index) < StartAddress) || - ((SectionAddr + Index) > StopAddress)) - continue; - if (NumBytes == 0) { - outs() << format("%8" PRIx64 ":", SectionAddr + Index); - outs() << "\t"; - } - Byte = Bytes.slice(Index)[0]; - outs() << format(" %02x", Byte); - AsciiData[NumBytes] = isPrint(Byte) ? Byte : '.'; - - uint8_t IndentOffset = 0; - NumBytes++; - if (Index == End - 1 || NumBytes > 8) { - // Indent the space for less than 8 bytes data. - // 2 spaces for byte and one for space between bytes - IndentOffset = 3 * (8 - NumBytes); - for (int Excess = 8 - NumBytes; Excess < 8; Excess++) - AsciiData[Excess] = '\0'; - NumBytes = 8; - } - if (NumBytes == 8) { - AsciiData[8] = '\0'; - outs() << std::string(IndentOffset, ' ') << " "; - outs() << reinterpret_cast<char *>(AsciiData); - outs() << '\n'; - NumBytes = 0; - } + if (SecondarySTI) { + if (getMappingSymbolKind(MappingSymbols, Index) == 'a') { + STI = PrimaryIsThumb ? SecondarySTI : PrimarySTI; + DisAsm = PrimaryIsThumb ? SecondaryDisAsm : PrimaryDisAsm; + } else if (getMappingSymbolKind(MappingSymbols, Index) == 't') { + STI = PrimaryIsThumb ? PrimarySTI : SecondarySTI; + DisAsm = PrimaryIsThumb ? PrimaryDisAsm : SecondaryDisAsm; } } - if (Index >= End) - break; - - if (size_t N = - countSkippableZeroBytes(Bytes.slice(Index, End - Index))) { - outs() << "\t\t..." << '\n'; - Index += N; - if (Index >= End) - break; - } // Disassemble a real instruction or a data when disassemble all is // provided - bool Disassembled = DisAsm->getInstruction(Inst, Size, Bytes.slice(Index), - SectionAddr + Index, DebugOut, - CommentStream); + MCInst Inst; + bool Disassembled = DisAsm->getInstruction( + Inst, Size, Bytes.slice(Index), SectionAddr + Index, DebugOut, + CommentStream); if (Size == 0) Size = 1; - PIP.printInst(*IP, Disassembled ? &Inst : nullptr, - Bytes.slice(Index, Size), SectionAddr + Index, outs(), "", - *STI, &SP, &Rels); + PIP.printInst( + *IP, Disassembled ? &Inst : nullptr, Bytes.slice(Index, Size), + {SectionAddr + Index + VMAAdjustment, Section.getIndex()}, outs(), + "", *STI, &SP, &Rels); outs() << CommentStream.str(); Comments.clear(); @@ -1791,37 +1403,34 @@ static void disassembleObject(const ObjectFile *Obj, bool InlineRelocs) { // N.B. We don't walk the relocations in the relocatable case yet. auto *TargetSectionSymbols = &Symbols; if (!Obj->isRelocatableObject()) { - auto SectionAddress = std::upper_bound( - SectionAddresses.begin(), SectionAddresses.end(), Target, - [](uint64_t LHS, - const std::pair<uint64_t, SectionRef> &RHS) { - return LHS < RHS.first; + auto It = partition_point( + SectionAddresses, + [=](const std::pair<uint64_t, SectionRef> &O) { + return O.first <= Target; }); - if (SectionAddress != SectionAddresses.begin()) { - --SectionAddress; - TargetSectionSymbols = &AllSymbols[SectionAddress->second]; + if (It != SectionAddresses.begin()) { + --It; + TargetSectionSymbols = &AllSymbols[It->second]; } else { TargetSectionSymbols = &AbsoluteSymbols; } } - // Find the first symbol in the section whose offset is less than + // Find the last symbol in the section whose offset is less than // or equal to the target. If there isn't a section that contains // the target, find the nearest preceding absolute symbol. - auto TargetSym = std::upper_bound( - TargetSectionSymbols->begin(), TargetSectionSymbols->end(), - Target, [](uint64_t LHS, - const std::tuple<uint64_t, StringRef, uint8_t> &RHS) { - return LHS < std::get<0>(RHS); + auto TargetSym = partition_point( + *TargetSectionSymbols, + [=](const std::tuple<uint64_t, StringRef, uint8_t> &O) { + return std::get<0>(O) <= Target; }); if (TargetSym == TargetSectionSymbols->begin()) { TargetSectionSymbols = &AbsoluteSymbols; - TargetSym = std::upper_bound( - AbsoluteSymbols.begin(), AbsoluteSymbols.end(), - Target, [](uint64_t LHS, - const std::tuple<uint64_t, StringRef, uint8_t> &RHS) { - return LHS < std::get<0>(RHS); - }); + TargetSym = partition_point( + AbsoluteSymbols, + [=](const std::tuple<uint64_t, StringRef, uint8_t> &O) { + return std::get<0>(O) <= Target; + }); } if (TargetSym != TargetSectionSymbols->begin()) { --TargetSym; @@ -1838,34 +1447,125 @@ static void disassembleObject(const ObjectFile *Obj, bool InlineRelocs) { outs() << "\n"; // Hexagon does this in pretty printer - if (Obj->getArch() != Triple::hexagon) + if (Obj->getArch() != Triple::hexagon) { // Print relocation for instruction. while (RelCur != RelEnd) { - uint64_t Addr = RelCur->getOffset(); - SmallString<16> Name; - SmallString<32> Val; - + uint64_t Offset = RelCur->getOffset(); // If this relocation is hidden, skip it. - if (getHidden(*RelCur) || ((SectionAddr + Addr) < StartAddress)) { + if (getHidden(*RelCur) || SectionAddr + Offset < StartAddress) { ++RelCur; continue; } - // Stop when rel_cur's address is past the current instruction. - if (Addr >= Index + Size) + // Stop when RelCur's offset is past the current instruction. + if (Offset >= Index + Size) break; - RelCur->getTypeName(Name); - error(getRelocationValueString(*RelCur, Val)); - outs() << format(Fmt.data(), SectionAddr + Addr) << Name << "\t" - << Val << "\n"; + + // When --adjust-vma is used, update the address printed. + if (RelCur->getSymbol() != Obj->symbol_end()) { + Expected<section_iterator> SymSI = + RelCur->getSymbol()->getSection(); + if (SymSI && *SymSI != Obj->section_end() && + shouldAdjustVA(**SymSI)) + Offset += AdjustVMA; + } + + printRelocation(*RelCur, SectionAddr + Offset, Is64Bits); ++RelCur; } + } + + Index += Size; } } } + StringSet<> MissingDisasmFuncsSet = + set_difference(DisasmFuncsSet, FoundDisasmFuncsSet); + for (StringRef MissingDisasmFunc : MissingDisasmFuncsSet.keys()) + warn("failed to disassemble missing function " + MissingDisasmFunc); +} + +static void disassembleObject(const ObjectFile *Obj, bool InlineRelocs) { + const Target *TheTarget = getTarget(Obj); + + // Package up features to be passed to target/subtarget + SubtargetFeatures Features = Obj->getFeatures(); + if (!MAttrs.empty()) + for (unsigned I = 0; I != MAttrs.size(); ++I) + Features.AddFeature(MAttrs[I]); + + std::unique_ptr<const MCRegisterInfo> MRI( + TheTarget->createMCRegInfo(TripleName)); + if (!MRI) + report_error(Obj->getFileName(), + "no register info for target " + TripleName); + + // Set up disassembler. + std::unique_ptr<const MCAsmInfo> AsmInfo( + TheTarget->createMCAsmInfo(*MRI, TripleName)); + if (!AsmInfo) + report_error(Obj->getFileName(), + "no assembly info for target " + TripleName); + std::unique_ptr<const MCSubtargetInfo> STI( + TheTarget->createMCSubtargetInfo(TripleName, MCPU, Features.getString())); + if (!STI) + report_error(Obj->getFileName(), + "no subtarget info for target " + TripleName); + std::unique_ptr<const MCInstrInfo> MII(TheTarget->createMCInstrInfo()); + if (!MII) + report_error(Obj->getFileName(), + "no instruction info for target " + TripleName); + MCObjectFileInfo MOFI; + MCContext Ctx(AsmInfo.get(), MRI.get(), &MOFI); + // FIXME: for now initialize MCObjectFileInfo with default values + MOFI.InitMCObjectFileInfo(Triple(TripleName), false, Ctx); + + std::unique_ptr<MCDisassembler> DisAsm( + TheTarget->createMCDisassembler(*STI, Ctx)); + if (!DisAsm) + report_error(Obj->getFileName(), + "no disassembler for target " + TripleName); + + // If we have an ARM object file, we need a second disassembler, because + // ARM CPUs have two different instruction sets: ARM mode, and Thumb mode. + // We use mapping symbols to switch between the two assemblers, where + // appropriate. + std::unique_ptr<MCDisassembler> SecondaryDisAsm; + std::unique_ptr<const MCSubtargetInfo> SecondarySTI; + if (isArmElf(Obj) && !STI->checkFeatures("+mclass")) { + if (STI->checkFeatures("+thumb-mode")) + Features.AddFeature("-thumb-mode"); + else + Features.AddFeature("+thumb-mode"); + SecondarySTI.reset(TheTarget->createMCSubtargetInfo(TripleName, MCPU, + Features.getString())); + SecondaryDisAsm.reset(TheTarget->createMCDisassembler(*SecondarySTI, Ctx)); + } + + std::unique_ptr<const MCInstrAnalysis> MIA( + TheTarget->createMCInstrAnalysis(MII.get())); + + int AsmPrinterVariant = AsmInfo->getAssemblerDialect(); + std::unique_ptr<MCInstPrinter> IP(TheTarget->createMCInstPrinter( + Triple(TripleName), AsmPrinterVariant, *AsmInfo, *MII, *MRI)); + if (!IP) + report_error(Obj->getFileName(), + "no instruction printer for target " + TripleName); + IP->setPrintImmHex(PrintImmHex); + + PrettyPrinter &PIP = selectPrettyPrinter(Triple(TripleName)); + SourcePrinter SP(Obj, TheTarget->getName()); + + for (StringRef Opt : DisassemblerOptions) + if (!IP->applyTargetSpecificCLOption(Opt)) + error("Unrecognized disassembler option: " + Opt); + + disassembleObject(TheTarget, Obj, Ctx, DisAsm.get(), SecondaryDisAsm.get(), + MIA.get(), IP.get(), STI.get(), SecondarySTI.get(), PIP, + SP, InlineRelocs); } -void llvm::printRelocations(const ObjectFile *Obj) { +void printRelocations(const ObjectFile *Obj) { StringRef Fmt = Obj->getBytesInAddress() > 4 ? "%016" PRIx64 : "%08" PRIx64; // Regular objdump doesn't print relocations in non-relocatable object @@ -1873,28 +1573,40 @@ void llvm::printRelocations(const ObjectFile *Obj) { if (!Obj->isRelocatableObject()) return; + // Build a mapping from relocation target to a vector of relocation + // sections. Usually, there is an only one relocation section for + // each relocated section. + MapVector<SectionRef, std::vector<SectionRef>> SecToRelSec; for (const SectionRef &Section : ToolSectionFilter(*Obj)) { if (Section.relocation_begin() == Section.relocation_end()) continue; + const SectionRef TargetSec = *Section.getRelocatedSection(); + SecToRelSec[TargetSec].push_back(Section); + } + + for (std::pair<SectionRef, std::vector<SectionRef>> &P : SecToRelSec) { StringRef SecName; - error(Section.getName(SecName)); + error(P.first.getName(SecName)); outs() << "RELOCATION RECORDS FOR [" << SecName << "]:\n"; - for (const RelocationRef &Reloc : Section.relocations()) { - uint64_t Address = Reloc.getOffset(); - SmallString<32> RelocName; - SmallString<32> ValueStr; - if (Address < StartAddress || Address > StopAddress || getHidden(Reloc)) - continue; - Reloc.getTypeName(RelocName); - error(getRelocationValueString(Reloc, ValueStr)); - outs() << format(Fmt.data(), Address) << " " << RelocName << " " - << ValueStr << "\n"; + + for (SectionRef Section : P.second) { + for (const RelocationRef &Reloc : Section.relocations()) { + uint64_t Address = Reloc.getOffset(); + SmallString<32> RelocName; + SmallString<32> ValueStr; + if (Address < StartAddress || Address > StopAddress || getHidden(Reloc)) + continue; + Reloc.getTypeName(RelocName); + error(getRelocationValueString(Reloc, ValueStr)); + outs() << format(Fmt.data(), Address) << " " << RelocName << " " + << ValueStr << "\n"; + } } outs() << "\n"; } } -void llvm::printDynamicRelocations(const ObjectFile *Obj) { +void printDynamicRelocations(const ObjectFile *Obj) { // For the moment, this option is for ELF only if (!Obj->isELF()) return; @@ -1911,9 +1623,7 @@ void llvm::printDynamicRelocations(const ObjectFile *Obj) { outs() << "DYNAMIC RELOCATION RECORDS\n"; StringRef Fmt = Obj->getBytesInAddress() > 4 ? "%016" PRIx64 : "%08" PRIx64; - for (const SectionRef &Section : DynRelSec) { - if (Section.relocation_begin() == Section.relocation_end()) - continue; + for (const SectionRef &Section : DynRelSec) for (const RelocationRef &Reloc : Section.relocations()) { uint64_t Address = Reloc.getOffset(); SmallString<32> RelocName; @@ -1923,34 +1633,60 @@ void llvm::printDynamicRelocations(const ObjectFile *Obj) { outs() << format(Fmt.data(), Address) << " " << RelocName << " " << ValueStr << "\n"; } - } } -void llvm::printSectionHeaders(const ObjectFile *Obj) { - outs() << "Sections:\n" - "Idx Name Size Address Type\n"; +// Returns true if we need to show LMA column when dumping section headers. We +// show it only when the platform is ELF and either we have at least one section +// whose VMA and LMA are different and/or when --show-lma flag is used. +static bool shouldDisplayLMA(const ObjectFile *Obj) { + if (!Obj->isELF()) + return false; + for (const SectionRef &S : ToolSectionFilter(*Obj)) + if (S.getAddress() != getELFSectionLMA(S)) + return true; + return ShowLMA; +} + +void printSectionHeaders(const ObjectFile *Obj) { + bool HasLMAColumn = shouldDisplayLMA(Obj); + if (HasLMAColumn) + outs() << "Sections:\n" + "Idx Name Size VMA LMA " + "Type\n"; + else + outs() << "Sections:\n" + "Idx Name Size VMA Type\n"; + for (const SectionRef &Section : ToolSectionFilter(*Obj)) { StringRef Name; error(Section.getName(Name)); - uint64_t Address = Section.getAddress(); + uint64_t VMA = Section.getAddress(); + if (shouldAdjustVA(Section)) + VMA += AdjustVMA; + uint64_t Size = Section.getSize(); bool Text = Section.isText(); bool Data = Section.isData(); bool BSS = Section.isBSS(); std::string Type = (std::string(Text ? "TEXT " : "") + (Data ? "DATA " : "") + (BSS ? "BSS" : "")); - outs() << format("%3d %-13s %08" PRIx64 " %016" PRIx64 " %s\n", - (unsigned)Section.getIndex(), Name.str().c_str(), Size, - Address, Type.c_str()); + + if (HasLMAColumn) + outs() << format("%3d %-13s %08" PRIx64 " %016" PRIx64 " %016" PRIx64 + " %s\n", + (unsigned)Section.getIndex(), Name.str().c_str(), Size, + VMA, getELFSectionLMA(Section), Type.c_str()); + else + outs() << format("%3d %-13s %08" PRIx64 " %016" PRIx64 " %s\n", + (unsigned)Section.getIndex(), Name.str().c_str(), Size, + VMA, Type.c_str()); } outs() << "\n"; } -void llvm::printSectionContents(const ObjectFile *Obj) { - std::error_code EC; +void printSectionContents(const ObjectFile *Obj) { for (const SectionRef &Section : ToolSectionFilter(*Obj)) { StringRef Name; - StringRef Contents; error(Section.getName(Name)); uint64_t BaseAddr = Section.getAddress(); uint64_t Size = Section.getSize(); @@ -1965,7 +1701,7 @@ void llvm::printSectionContents(const ObjectFile *Obj) { continue; } - error(Section.getContents(Contents)); + StringRef Contents = unwrapOrError(Section.getContents(), Obj->getFileName()); // Dump out the content as hex and printable ascii characters. for (std::size_t Addr = 0, End = Contents.size(); Addr < End; Addr += 16) { @@ -1993,8 +1729,8 @@ void llvm::printSectionContents(const ObjectFile *Obj) { } } -void llvm::printSymbolTable(const ObjectFile *O, StringRef ArchiveName, - StringRef ArchitectureName) { +void printSymbolTable(const ObjectFile *O, StringRef ArchiveName, + StringRef ArchitectureName) { outs() << "SYMBOL TABLE:\n"; if (const COFFObjectFile *Coff = dyn_cast<const COFFObjectFile>(O)) { @@ -2002,41 +1738,24 @@ void llvm::printSymbolTable(const ObjectFile *O, StringRef ArchiveName, return; } + const StringRef FileName = O->getFileName(); for (auto I = O->symbol_begin(), E = O->symbol_end(); I != E; ++I) { - // Skip printing the special zero symbol when dumping an ELF file. - // This makes the output consistent with the GNU objdump. - if (I == O->symbol_begin() && isa<ELFObjectFileBase>(O)) - continue; - const SymbolRef &Symbol = *I; - Expected<uint64_t> AddressOrError = Symbol.getAddress(); - if (!AddressOrError) - report_error(ArchiveName, O->getFileName(), AddressOrError.takeError(), - ArchitectureName); - uint64_t Address = *AddressOrError; + uint64_t Address = unwrapOrError(Symbol.getAddress(), ArchiveName, FileName, + ArchitectureName); if ((Address < StartAddress) || (Address > StopAddress)) continue; - Expected<SymbolRef::Type> TypeOrError = Symbol.getType(); - if (!TypeOrError) - report_error(ArchiveName, O->getFileName(), TypeOrError.takeError(), - ArchitectureName); - SymbolRef::Type Type = *TypeOrError; + SymbolRef::Type Type = unwrapOrError(Symbol.getType(), ArchiveName, + FileName, ArchitectureName); uint32_t Flags = Symbol.getFlags(); - Expected<section_iterator> SectionOrErr = Symbol.getSection(); - if (!SectionOrErr) - report_error(ArchiveName, O->getFileName(), SectionOrErr.takeError(), - ArchitectureName); - section_iterator Section = *SectionOrErr; + section_iterator Section = unwrapOrError(Symbol.getSection(), ArchiveName, + FileName, ArchitectureName); StringRef Name; - if (Type == SymbolRef::ST_Debug && Section != O->section_end()) { + if (Type == SymbolRef::ST_Debug && Section != O->section_end()) Section->getName(Name); - } else { - Expected<StringRef> NameOrErr = Symbol.getName(); - if (!NameOrErr) - report_error(ArchiveName, O->getFileName(), NameOrErr.takeError(), - ArchitectureName); - Name = *NameOrErr; - } + else + Name = unwrapOrError(Symbol.getName(), ArchiveName, FileName, + ArchitectureName); bool Global = Flags & SymbolRef::SF_Global; bool Weak = Flags & SymbolRef::SF_Weak; @@ -2087,20 +1806,38 @@ void llvm::printSymbolTable(const ObjectFile *O, StringRef ArchiveName, outs() << SectionName; } - outs() << '\t'; if (Common || isa<ELFObjectFileBase>(O)) { uint64_t Val = Common ? Symbol.getAlignment() : ELFSymbolRef(Symbol).getSize(); - outs() << format("\t %08" PRIx64 " ", Val); + outs() << format("\t%08" PRIx64, Val); } - if (Hidden) - outs() << ".hidden "; + if (isa<ELFObjectFileBase>(O)) { + uint8_t Other = ELFSymbolRef(Symbol).getOther(); + switch (Other) { + case ELF::STV_DEFAULT: + break; + case ELF::STV_INTERNAL: + outs() << " .internal"; + break; + case ELF::STV_HIDDEN: + outs() << " .hidden"; + break; + case ELF::STV_PROTECTED: + outs() << " .protected"; + break; + default: + outs() << format(" 0x%02x", Other); + break; + } + } else if (Hidden) { + outs() << " .hidden"; + } if (Demangle) - outs() << demangle(Name) << '\n'; + outs() << ' ' << demangle(Name) << '\n'; else - outs() << Name << '\n'; + outs() << ' ' << Name << '\n'; } } @@ -2118,59 +1855,9 @@ static void printUnwindInfo(const ObjectFile *O) { "for COFF and MachO object files.\n"; } -void llvm::printExportsTrie(const ObjectFile *o) { - outs() << "Exports trie:\n"; - if (const MachOObjectFile *MachO = dyn_cast<MachOObjectFile>(o)) - printMachOExportsTrie(MachO); - else - WithColor::error(errs(), ToolName) - << "This operation is only currently supported " - "for Mach-O executable files.\n"; -} - -void llvm::printRebaseTable(ObjectFile *o) { - outs() << "Rebase table:\n"; - if (MachOObjectFile *MachO = dyn_cast<MachOObjectFile>(o)) - printMachORebaseTable(MachO); - else - WithColor::error(errs(), ToolName) - << "This operation is only currently supported " - "for Mach-O executable files.\n"; -} - -void llvm::printBindTable(ObjectFile *o) { - outs() << "Bind table:\n"; - if (MachOObjectFile *MachO = dyn_cast<MachOObjectFile>(o)) - printMachOBindTable(MachO); - else - WithColor::error(errs(), ToolName) - << "This operation is only currently supported " - "for Mach-O executable files.\n"; -} - -void llvm::printLazyBindTable(ObjectFile *o) { - outs() << "Lazy bind table:\n"; - if (MachOObjectFile *MachO = dyn_cast<MachOObjectFile>(o)) - printMachOLazyBindTable(MachO); - else - WithColor::error(errs(), ToolName) - << "This operation is only currently supported " - "for Mach-O executable files.\n"; -} - -void llvm::printWeakBindTable(ObjectFile *o) { - outs() << "Weak bind table:\n"; - if (MachOObjectFile *MachO = dyn_cast<MachOObjectFile>(o)) - printMachOWeakBindTable(MachO); - else - WithColor::error(errs(), ToolName) - << "This operation is only currently supported " - "for Mach-O executable files.\n"; -} - /// Dump the raw contents of the __clangast section so the output can be piped /// into llvm-bcanalyzer. -void llvm::printRawClangAST(const ObjectFile *Obj) { +void printRawClangAST(const ObjectFile *Obj) { if (outs().is_displayed()) { WithColor::error(errs(), ToolName) << "The -raw-clang-ast option will dump the raw binary contents of " @@ -2197,8 +1884,8 @@ void llvm::printRawClangAST(const ObjectFile *Obj) { if (!ClangASTSection) return; - StringRef ClangASTContents; - error(ClangASTSection.getValue().getContents(ClangASTContents)); + StringRef ClangASTContents = unwrapOrError( + ClangASTSection.getValue().getContents(), Obj->getFileName()); outs().write(ClangASTContents.data(), ClangASTContents.size()); } @@ -2234,9 +1921,8 @@ static void printFaultMaps(const ObjectFile *Obj) { return; } - StringRef FaultMapContents; - error(FaultMapSection.getValue().getContents(FaultMapContents)); - + StringRef FaultMapContents = + unwrapOrError(FaultMapSection.getValue().getContents(), Obj->getFileName()); FaultMapParser FMP(FaultMapContents.bytes_begin(), FaultMapContents.bytes_end()); @@ -2246,7 +1932,9 @@ static void printFaultMaps(const ObjectFile *Obj) { static void printPrivateFileHeaders(const ObjectFile *O, bool OnlyFirst) { if (O->isELF()) { printELFFileHeader(O); - return printELFDynamicSection(O); + printELFDynamicSection(O); + printELFSymbolVersionInfo(O); + return; } if (O->isCOFF()) return printCOFFFileHeader(O); @@ -2267,12 +1955,9 @@ static void printFileHeaders(const ObjectFile *O) { Triple::ArchType AT = O->getArch(); outs() << "architecture: " << Triple::getArchTypeName(AT) << "\n"; - Expected<uint64_t> StartAddrOrErr = O->getStartAddress(); - if (!StartAddrOrErr) - report_error(O->getFileName(), StartAddrOrErr.takeError()); + uint64_t Address = unwrapOrError(O->getStartAddress(), O->getFileName()); StringRef Fmt = O->getBytesInAddress() > 4 ? "%016" PRIx64 : "%08" PRIx64; - uint64_t Address = StartAddrOrErr.get(); outs() << "start address: " << "0x" << format(Fmt.data(), Address) << "\n\n"; } @@ -2297,22 +1982,9 @@ static void printArchiveChild(StringRef Filename, const Archive::Child &C) { outs() << " "; - Expected<unsigned> UIDOrErr = C.getUID(); - if (!UIDOrErr) - report_error(Filename, UIDOrErr.takeError()); - unsigned UID = UIDOrErr.get(); - outs() << format("%d/", UID); - - Expected<unsigned> GIDOrErr = C.getGID(); - if (!GIDOrErr) - report_error(Filename, GIDOrErr.takeError()); - unsigned GID = GIDOrErr.get(); - outs() << format("%-d ", GID); - - Expected<uint64_t> Size = C.getRawSize(); - if (!Size) - report_error(Filename, Size.takeError()); - outs() << format("%6" PRId64, Size.get()) << " "; + outs() << format("%d/%d %6" PRId64 " ", unwrapOrError(C.getUID(), Filename), + unwrapOrError(C.getGID(), Filename), + unwrapOrError(C.getRawSize(), Filename)); StringRef RawLastModified = C.getRawLastModified(); unsigned Seconds; @@ -2331,10 +2003,7 @@ static void printArchiveChild(StringRef Filename, const Archive::Child &C) { Expected<StringRef> NameOrErr = C.getName(); if (!NameOrErr) { consumeError(NameOrErr.takeError()); - Expected<StringRef> RawNameOrErr = C.getRawName(); - if (!RawNameOrErr) - report_error(Filename, NameOrErr.takeError()); - Name = RawNameOrErr.get(); + Name = unwrapOrError(C.getRawName(), Filename); } else { Name = NameOrErr.get(); } @@ -2386,7 +2055,7 @@ static void dumpObject(ObjectFile *O, const Archive *A = nullptr, printWeakBindTable(O); if (RawClangAST) printRawClangAST(O); - if (PrintFaultMaps) + if (FaultMapSection) printFaultMaps(O); if (DwarfDumpType != DIDT_Null) { std::unique_ptr<DIContext> DICtx = DWARFContext::create(*O); @@ -2421,7 +2090,7 @@ static void dumpArchive(const Archive *A) { Expected<std::unique_ptr<Binary>> ChildOrErr = C.getAsBinary(); if (!ChildOrErr) { if (auto E = isNotObjectErrorInvalidFileType(ChildOrErr.takeError())) - report_error(A->getFileName(), C, std::move(E)); + report_error(std::move(E), A->getFileName(), C); continue; } if (ObjectFile *O = dyn_cast<ObjectFile>(&*ChildOrErr.get())) @@ -2429,10 +2098,11 @@ static void dumpArchive(const Archive *A) { else if (COFFImportFile *I = dyn_cast<COFFImportFile>(&*ChildOrErr.get())) dumpObject(I, A, &C); else - report_error(A->getFileName(), object_error::invalid_file_type); + report_error(errorCodeToError(object_error::invalid_file_type), + A->getFileName()); } if (Err) - report_error(A->getFileName(), std::move(Err)); + report_error(std::move(Err), A->getFileName()); } /// Open file and figure out how to dump it. @@ -2446,10 +2116,8 @@ static void dumpInput(StringRef file) { } // Attempt to open the binary. - Expected<OwningBinary<Binary>> BinaryOrErr = createBinary(file); - if (!BinaryOrErr) - report_error(file, BinaryOrErr.takeError()); - Binary &Binary = *BinaryOrErr.get().getBinary(); + OwningBinary<Binary> OBinary = unwrapOrError(createBinary(file), file); + Binary &Binary = *OBinary.getBinary(); if (Archive *A = dyn_cast<Archive>(&Binary)) dumpArchive(A); @@ -2458,22 +2126,29 @@ static void dumpInput(StringRef file) { else if (MachOUniversalBinary *UB = dyn_cast<MachOUniversalBinary>(&Binary)) parseInputMachO(UB); else - report_error(file, object_error::invalid_file_type); + report_error(errorCodeToError(object_error::invalid_file_type), file); } +} // namespace llvm int main(int argc, char **argv) { + using namespace llvm; InitLLVM X(argc, argv); + const cl::OptionCategory *OptionFilters[] = {&ObjdumpCat, &MachOCat}; + cl::HideUnrelatedOptions(OptionFilters); // Initialize targets and assembly printers/parsers. - llvm::InitializeAllTargetInfos(); - llvm::InitializeAllTargetMCs(); - llvm::InitializeAllDisassemblers(); + InitializeAllTargetInfos(); + InitializeAllTargetMCs(); + InitializeAllDisassemblers(); // Register the target printer for --version. cl::AddExtraVersionPrinter(TargetRegistry::printRegisteredTargetsForVersion); cl::ParseCommandLineOptions(argc, argv, "llvm object file dumper\n"); + if (StartAddress >= StopAddress) + error("start address should be less than stop address"); + ToolName = argv[0]; // Defaults to a.out if no filenames specified. @@ -2481,40 +2156,22 @@ int main(int argc, char **argv) { InputFilenames.push_back("a.out"); if (AllHeaders) - FileHeaders = PrivateHeaders = Relocations = SectionHeaders = SymbolTable = - true; + ArchiveHeaders = FileHeaders = PrivateHeaders = Relocations = + SectionHeaders = SymbolTable = true; - if (DisassembleAll || PrintSource || PrintLines) + if (DisassembleAll || PrintSource || PrintLines || + (!DisassembleFunctions.empty())) Disassemble = true; - if (!Disassemble - && !Relocations - && !DynamicRelocations - && !SectionHeaders - && !SectionContents - && !SymbolTable - && !UnwindInfo - && !PrivateHeaders - && !FileHeaders - && !FirstPrivateHeader - && !ExportsTrie - && !Rebase - && !Bind - && !LazyBind - && !WeakBind - && !RawClangAST - && !(UniversalHeaders && MachOOpt) - && !ArchiveHeaders - && !(IndirectSymbols && MachOOpt) - && !(DataInCode && MachOOpt) - && !(LinkOptHints && MachOOpt) - && !(InfoPlist && MachOOpt) - && !(DylibsUsed && MachOOpt) - && !(DylibId && MachOOpt) - && !(ObjcMetaData && MachOOpt) - && !(!FilterSections.empty() && MachOOpt) - && !PrintFaultMaps - && DwarfDumpType == DIDT_Null) { + if (!ArchiveHeaders && !Disassemble && DwarfDumpType == DIDT_Null && + !DynamicRelocations && !FileHeaders && !PrivateHeaders && !RawClangAST && + !Relocations && !SectionHeaders && !SectionContents && !SymbolTable && + !UnwindInfo && !FaultMapSection && + !(MachOOpt && + (Bind || DataInCode || DylibId || DylibsUsed || ExportsTrie || + FirstPrivateHeader || IndirectSymbols || InfoPlist || LazyBind || + LinkOptHints || ObjcMetaData || Rebase || UniversalHeaders || + WeakBind || !FilterSections.empty()))) { cl::PrintHelpMessage(); return 2; } @@ -2524,5 +2181,7 @@ int main(int argc, char **argv) { llvm::for_each(InputFilenames, dumpInput); + warnOnNoMatchForSections(); + return EXIT_SUCCESS; } diff --git a/tools/llvm-objdump/llvm-objdump.h b/tools/llvm-objdump/llvm-objdump.h index fe2cb05fe227..e58d4a05c2e6 100644 --- a/tools/llvm-objdump/llvm-objdump.h +++ b/tools/llvm-objdump/llvm-objdump.h @@ -1,8 +1,7 @@ // -// The LLVM Compiler Infrastructure -// -// This file is distributed under the University of Illinois Open Source -// License. See LICENSE.TXT for details. +// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions. +// See https://llvm.org/LICENSE.txt for license information. +// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception // //===----------------------------------------------------------------------===// @@ -19,57 +18,86 @@ namespace llvm { class StringRef; namespace object { - class COFFObjectFile; - class COFFImportFile; - class MachOObjectFile; - class MachOUniversalBinary; - class ObjectFile; - class Archive; - class RelocationRef; +class COFFObjectFile; +class COFFImportFile; +class ELFObjectFileBase; +class ELFSectionRef; +class MachOObjectFile; +class MachOUniversalBinary; +class RelocationRef; } -extern cl::opt<std::string> TripleName; -extern cl::opt<std::string> ArchName; -extern cl::opt<std::string> MCPU; -extern cl::list<std::string> MAttrs; -extern cl::list<std::string> FilterSections; -extern cl::opt<bool> AllHeaders; extern cl::opt<bool> Demangle; -extern cl::opt<bool> Disassemble; -extern cl::opt<bool> DisassembleAll; -extern cl::opt<bool> NoShowRawInsn; -extern cl::opt<bool> NoLeadingAddr; -extern cl::opt<bool> PrivateHeaders; -extern cl::opt<bool> FileHeaders; -extern cl::opt<bool> FirstPrivateHeader; -extern cl::opt<bool> ExportsTrie; -extern cl::opt<bool> Rebase; -extern cl::opt<bool> Bind; -extern cl::opt<bool> LazyBind; -extern cl::opt<bool> WeakBind; -extern cl::opt<bool> RawClangAST; -extern cl::opt<bool> UniversalHeaders; -extern cl::opt<bool> ArchiveHeaders; -extern cl::opt<bool> IndirectSymbols; -extern cl::opt<bool> DataInCode; -extern cl::opt<bool> LinkOptHints; -extern cl::opt<bool> InfoPlist; -extern cl::opt<bool> DylibsUsed; -extern cl::opt<bool> DylibId; -extern cl::opt<bool> ObjcMetaData; -extern cl::opt<std::string> DisSymName; -extern cl::opt<bool> NonVerbose; -extern cl::opt<bool> Relocations; -extern cl::opt<bool> DynamicRelocations; -extern cl::opt<bool> SectionHeaders; -extern cl::opt<bool> SectionContents; -extern cl::opt<bool> SymbolTable; -extern cl::opt<bool> UnwindInfo; -extern cl::opt<bool> PrintImmHex; -extern cl::opt<DIDumpType> DwarfDumpType; + +typedef std::function<bool(llvm::object::SectionRef const &)> FilterPredicate; + +class SectionFilterIterator { +public: + SectionFilterIterator(FilterPredicate P, + llvm::object::section_iterator const &I, + llvm::object::section_iterator const &E) + : Predicate(std::move(P)), Iterator(I), End(E) { + ScanPredicate(); + } + const llvm::object::SectionRef &operator*() const { return *Iterator; } + SectionFilterIterator &operator++() { + ++Iterator; + ScanPredicate(); + return *this; + } + bool operator!=(SectionFilterIterator const &Other) const { + return Iterator != Other.Iterator; + } + +private: + void ScanPredicate() { + while (Iterator != End && !Predicate(*Iterator)) { + ++Iterator; + } + } + FilterPredicate Predicate; + llvm::object::section_iterator Iterator; + llvm::object::section_iterator End; +}; + +class SectionFilter { +public: + SectionFilter(FilterPredicate P, llvm::object::ObjectFile const &O) + : Predicate(std::move(P)), Object(O) {} + SectionFilterIterator begin() { + return SectionFilterIterator(Predicate, Object.section_begin(), + Object.section_end()); + } + SectionFilterIterator end() { + return SectionFilterIterator(Predicate, Object.section_end(), + Object.section_end()); + } + +private: + FilterPredicate Predicate; + llvm::object::ObjectFile const &Object; +}; // Various helper functions. +SectionFilter ToolSectionFilter(llvm::object::ObjectFile const &O); + +Error getELFRelocationValueString(const object::ELFObjectFileBase *Obj, + const object::RelocationRef &Rel, + llvm::SmallVectorImpl<char> &Result); +Error getCOFFRelocationValueString(const object::COFFObjectFile *Obj, + const object::RelocationRef &Rel, + llvm::SmallVectorImpl<char> &Result); +Error getWasmRelocationValueString(const object::WasmObjectFile *Obj, + const object::RelocationRef &RelRef, + llvm::SmallVectorImpl<char> &Result); +Error getMachORelocationValueString(const object::MachOObjectFile *Obj, + const object::RelocationRef &RelRef, + llvm::SmallVectorImpl<char> &Result); + +uint64_t getELFSectionLMA(const object::ELFSectionRef& Sec); + void error(std::error_code ec); +void error(Error E); bool isRelocAddressLess(object::RelocationRef A, object::RelocationRef B); void parseInputMachO(StringRef Filename); void parseInputMachO(object::MachOUniversalBinary *UB); @@ -82,6 +110,7 @@ void printMachOLazyBindTable(object::MachOObjectFile *O); void printMachOWeakBindTable(object::MachOObjectFile *O); void printELFFileHeader(const object::ObjectFile *O); void printELFDynamicSection(const object::ObjectFile *Obj); +void printELFSymbolVersionInfo(const object::ObjectFile *Obj); void printCOFFFileHeader(const object::ObjectFile *O); void printCOFFSymbolTable(const object::COFFImportFile *I); void printCOFFSymbolTable(const object::COFFObjectFile *O); @@ -103,18 +132,20 @@ void printSymbolTable(const object::ObjectFile *O, StringRef ArchiveName, void warn(StringRef Message); LLVM_ATTRIBUTE_NORETURN void error(Twine Message); LLVM_ATTRIBUTE_NORETURN void report_error(StringRef File, Twine Message); -LLVM_ATTRIBUTE_NORETURN void report_error(StringRef File, std::error_code EC); -LLVM_ATTRIBUTE_NORETURN void report_error(StringRef File, llvm::Error E); -LLVM_ATTRIBUTE_NORETURN void report_error(StringRef FileName, - StringRef ArchiveName, - llvm::Error E, - StringRef ArchitectureName - = StringRef()); -LLVM_ATTRIBUTE_NORETURN void report_error(StringRef ArchiveName, - const object::Archive::Child &C, - llvm::Error E, - StringRef ArchitectureName - = StringRef()); +LLVM_ATTRIBUTE_NORETURN void report_error(Error E, StringRef File); +LLVM_ATTRIBUTE_NORETURN void +report_error(Error E, StringRef FileName, StringRef ArchiveName, + StringRef ArchitectureName = StringRef()); +LLVM_ATTRIBUTE_NORETURN void +report_error(Error E, StringRef ArchiveName, const object::Archive::Child &C, + StringRef ArchitectureName = StringRef()); + +template <typename T, typename... Ts> +T unwrapOrError(Expected<T> EO, Ts &&... Args) { + if (EO) + return std::move(*EO); + report_error(EO.takeError(), std::forward<Ts>(Args)...); +} } // end namespace llvm |