//===----------------------------------------------------------------------===//
#include "llvm-objdump.h"
-#include "MCFunction.h"
-#include "llvm/Support/MachO.h"
-#include "llvm/Object/MachO.h"
-#include "llvm/ADT/OwningPtr.h"
-#include "llvm/ADT/Triple.h"
#include "llvm/ADT/STLExtras.h"
+#include "llvm/ADT/StringExtras.h"
+#include "llvm/ADT/Triple.h"
#include "llvm/DebugInfo/DIContext.h"
#include "llvm/MC/MCAsmInfo.h"
+#include "llvm/MC/MCContext.h"
#include "llvm/MC/MCDisassembler.h"
#include "llvm/MC/MCInst.h"
#include "llvm/MC/MCInstPrinter.h"
#include "llvm/MC/MCInstrAnalysis.h"
#include "llvm/MC/MCInstrDesc.h"
#include "llvm/MC/MCInstrInfo.h"
+#include "llvm/MC/MCRegisterInfo.h"
#include "llvm/MC/MCSubtargetInfo.h"
+#include "llvm/Object/MachO.h"
+#include "llvm/Support/Casting.h"
#include "llvm/Support/CommandLine.h"
#include "llvm/Support/Debug.h"
+#include "llvm/Support/Endian.h"
#include "llvm/Support/Format.h"
#include "llvm/Support/GraphWriter.h"
+#include "llvm/Support/MachO.h"
#include "llvm/Support/MemoryBuffer.h"
#include "llvm/Support/TargetRegistry.h"
#include "llvm/Support/TargetSelect.h"
#include "llvm/Support/raw_ostream.h"
-#include "llvm/Support/system_error.h"
#include <algorithm>
#include <cstring>
+#include <system_error>
using namespace llvm;
using namespace object;
-static cl::opt<bool>
- CFG("cfg", cl::desc("Create a CFG for every symbol in the object file and"
- "write it to a graphviz file (MachO-only)"));
-
static cl::opt<bool>
UseDbg("g", cl::desc("Print line information from debug info if available"));
static cl::opt<std::string>
DSYMFile("dsym", cl::desc("Use .dSYM file for debug info"));
-static const Target *GetTarget(const MachOObject *MachOObj) {
+static const Target *GetTarget(const MachOObjectFile *MachOObj) {
// Figure out the target triple.
- llvm::Triple TT("unknown-unknown-unknown");
- switch (MachOObj->getHeader().CPUType) {
- case llvm::MachO::CPUTypeI386:
- TT.setArch(Triple::ArchType(Triple::x86));
- break;
- case llvm::MachO::CPUTypeX86_64:
- TT.setArch(Triple::ArchType(Triple::x86_64));
- break;
- case llvm::MachO::CPUTypeARM:
- TT.setArch(Triple::ArchType(Triple::arm));
- break;
- case llvm::MachO::CPUTypePowerPC:
- TT.setArch(Triple::ArchType(Triple::ppc));
- break;
- case llvm::MachO::CPUTypePowerPC64:
- TT.setArch(Triple::ArchType(Triple::ppc64));
- break;
+ if (TripleName.empty()) {
+ llvm::Triple TT("unknown-unknown-unknown");
+ TT.setArch(Triple::ArchType(MachOObj->getArch()));
+ TripleName = TT.str();
}
- TripleName = TT.str();
-
// Get the target specific parser.
std::string Error;
const Target *TheTarget = TargetRegistry::lookupTarget(TripleName, Error);
errs() << "llvm-objdump: error: unable to get target for '" << TripleName
<< "', see --version and --triple.\n";
- return 0;
+ return nullptr;
}
struct SymbolSorter {
}
};
-// Print additional information about an address, if available.
-static void DumpAddress(uint64_t Address, ArrayRef<SectionRef> Sections,
- MachOObject *MachOObj, raw_ostream &OS) {
- for (unsigned i = 0; i != Sections.size(); ++i) {
- uint64_t SectAddr = 0, SectSize = 0;
- Sections[i].getAddress(SectAddr);
- Sections[i].getSize(SectSize);
- uint64_t addr = SectAddr;
- if (SectAddr <= Address &&
- SectAddr + SectSize > Address) {
- StringRef bytes, name;
- Sections[i].getContents(bytes);
- Sections[i].getName(name);
- // Print constant strings.
- if (!name.compare("__cstring"))
- OS << '"' << bytes.substr(addr, bytes.find('\0', addr)) << '"';
- // Print constant CFStrings.
- if (!name.compare("__cfstring"))
- OS << "@\"" << bytes.substr(addr, bytes.find('\0', addr)) << '"';
- }
- }
-}
+// Types for the storted data in code table that is built before disassembly
+// and the predicate function to sort them.
+typedef std::pair<uint64_t, DiceRef> DiceTableEntry;
+typedef std::vector<DiceTableEntry> DiceTable;
+typedef DiceTable::iterator dice_table_iterator;
-typedef std::map<uint64_t, MCFunction*> FunctionMapTy;
-typedef SmallVector<MCFunction, 16> FunctionListTy;
-static void createMCFunctionAndSaveCalls(StringRef Name,
- const MCDisassembler *DisAsm,
- MemoryObject &Object, uint64_t Start,
- uint64_t End,
- MCInstrAnalysis *InstrAnalysis,
- uint64_t Address,
- raw_ostream &DebugOut,
- FunctionMapTy &FunctionMap,
- FunctionListTy &Functions) {
- SmallVector<uint64_t, 16> Calls;
- MCFunction f =
- MCFunction::createFunctionFromMC(Name, DisAsm, Object, Start, End,
- InstrAnalysis, DebugOut, Calls);
- Functions.push_back(f);
- FunctionMap[Address] = &Functions.back();
-
- // Add the gathered callees to the map.
- for (unsigned i = 0, e = Calls.size(); i != e; ++i)
- FunctionMap.insert(std::make_pair(Calls[i], (MCFunction*)0));
+static bool
+compareDiceTableEntries(const DiceTableEntry i,
+ const DiceTableEntry j) {
+ return i.first == j.first;
}
-// Write a graphviz file for the CFG inside an MCFunction.
-static void emitDOTFile(const char *FileName, const MCFunction &f,
- MCInstPrinter *IP) {
- // Start a new dot file.
- std::string Error;
- raw_fd_ostream Out(FileName, Error);
- if (!Error.empty()) {
- errs() << "llvm-objdump: warning: " << Error << '\n';
- return;
- }
-
- Out << "digraph " << f.getName() << " {\n";
- Out << "graph [ rankdir = \"LR\" ];\n";
- for (MCFunction::iterator i = f.begin(), e = f.end(); i != e; ++i) {
- bool hasPreds = false;
- // Only print blocks that have predecessors.
- // FIXME: Slow.
- for (MCFunction::iterator pi = f.begin(), pe = f.end(); pi != pe;
- ++pi)
- if (pi->second.contains(i->first)) {
- hasPreds = true;
- break;
- }
-
- if (!hasPreds && i != f.begin())
- continue;
-
- Out << '"' << i->first << "\" [ label=\"<a>";
- // Print instructions.
- for (unsigned ii = 0, ie = i->second.getInsts().size(); ii != ie;
- ++ii) {
- // Escape special chars and print the instruction in mnemonic form.
- std::string Str;
- raw_string_ostream OS(Str);
- IP->printInst(&i->second.getInsts()[ii].Inst, OS, "");
- Out << DOT::EscapeString(OS.str()) << '|';
+static void DumpDataInCode(const char *bytes, uint64_t Size,
+ unsigned short Kind) {
+ uint64_t Value;
+
+ switch (Kind) {
+ case MachO::DICE_KIND_DATA:
+ switch (Size) {
+ case 4:
+ Value = bytes[3] << 24 |
+ bytes[2] << 16 |
+ bytes[1] << 8 |
+ bytes[0];
+ outs() << "\t.long " << Value;
+ break;
+ case 2:
+ Value = bytes[1] << 8 |
+ bytes[0];
+ outs() << "\t.short " << Value;
+ break;
+ case 1:
+ Value = bytes[0];
+ outs() << "\t.byte " << Value;
+ break;
}
- Out << "<o>\" shape=\"record\" ];\n";
-
- // Add edges.
- for (MCBasicBlock::succ_iterator si = i->second.succ_begin(),
- se = i->second.succ_end(); si != se; ++si)
- Out << i->first << ":o -> " << *si <<":a\n";
+ outs() << "\t@ KIND_DATA\n";
+ break;
+ case MachO::DICE_KIND_JUMP_TABLE8:
+ Value = bytes[0];
+ outs() << "\t.byte " << Value << "\t@ KIND_JUMP_TABLE8";
+ break;
+ case MachO::DICE_KIND_JUMP_TABLE16:
+ Value = bytes[1] << 8 |
+ bytes[0];
+ outs() << "\t.short " << Value << "\t@ KIND_JUMP_TABLE16";
+ break;
+ case MachO::DICE_KIND_JUMP_TABLE32:
+ Value = bytes[3] << 24 |
+ bytes[2] << 16 |
+ bytes[1] << 8 |
+ bytes[0];
+ outs() << "\t.long " << Value << "\t@ KIND_JUMP_TABLE32";
+ break;
+ default:
+ outs() << "\t@ data in code kind = " << Kind << "\n";
+ break;
}
- Out << "}\n";
}
-static void getSectionsAndSymbols(const macho::Header &Header,
+static void getSectionsAndSymbols(const MachO::mach_header Header,
MachOObjectFile *MachOObj,
- InMemoryStruct<macho::SymtabLoadCommand> *SymtabLC,
std::vector<SectionRef> &Sections,
std::vector<SymbolRef> &Symbols,
- SmallVectorImpl<uint64_t> &FoundFns) {
- error_code ec;
- for (symbol_iterator SI = MachOObj->begin_symbols(),
- SE = MachOObj->end_symbols(); SI != SE; SI.increment(ec))
- Symbols.push_back(*SI);
-
- for (section_iterator SI = MachOObj->begin_sections(),
- SE = MachOObj->end_sections(); SI != SE; SI.increment(ec)) {
- SectionRef SR = *SI;
+ SmallVectorImpl<uint64_t> &FoundFns,
+ uint64_t &BaseSegmentAddress) {
+ for (const SymbolRef &Symbol : MachOObj->symbols())
+ Symbols.push_back(Symbol);
+
+ for (const SectionRef &Section : MachOObj->sections()) {
StringRef SectName;
- SR.getName(SectName);
- Sections.push_back(*SI);
+ Section.getName(SectName);
+ Sections.push_back(Section);
}
- for (unsigned i = 0; i != Header.NumLoadCommands; ++i) {
- const MachOObject::LoadCommandInfo &LCI =
- MachOObj->getObject()->getLoadCommandInfo(i);
- if (LCI.Command.Type == macho::LCT_FunctionStarts) {
+ MachOObjectFile::LoadCommandInfo Command =
+ MachOObj->getFirstLoadCommandInfo();
+ bool BaseSegmentAddressSet = false;
+ for (unsigned i = 0; ; ++i) {
+ if (Command.C.cmd == MachO::LC_FUNCTION_STARTS) {
// We found a function starts segment, parse the addresses for later
// consumption.
- InMemoryStruct<macho::LinkeditDataLoadCommand> LLC;
- MachOObj->getObject()->ReadLinkeditDataLoadCommand(LCI, LLC);
+ MachO::linkedit_data_command LLC =
+ MachOObj->getLinkeditDataLoadCommand(Command);
- MachOObj->getObject()->ReadULEB128s(LLC->DataOffset, FoundFns);
+ MachOObj->ReadULEB128s(LLC.dataoff, FoundFns);
+ }
+ else if (Command.C.cmd == MachO::LC_SEGMENT) {
+ MachO::segment_command SLC =
+ MachOObj->getSegmentLoadCommand(Command);
+ StringRef SegName = SLC.segname;
+ if(!BaseSegmentAddressSet && SegName != "__PAGEZERO") {
+ BaseSegmentAddressSet = true;
+ BaseSegmentAddress = SLC.vmaddr;
+ }
}
+
+ if (i == Header.ncmds - 1)
+ break;
+ else
+ Command = MachOObj->getNextLoadCommandInfo(Command);
}
}
-void llvm::DisassembleInputMachO(StringRef Filename) {
- OwningPtr<MemoryBuffer> Buff;
+static void DisassembleInputMachO2(StringRef Filename,
+ MachOObjectFile *MachOOF);
- if (error_code ec = MemoryBuffer::getFileOrSTDIN(Filename, Buff)) {
- errs() << "llvm-objdump: " << Filename << ": " << ec.message() << "\n";
+void llvm::DisassembleInputMachO(StringRef Filename) {
+ ErrorOr<std::unique_ptr<MemoryBuffer>> Buff =
+ MemoryBuffer::getFileOrSTDIN(Filename);
+ if (std::error_code EC = Buff.getError()) {
+ errs() << "llvm-objdump: " << Filename << ": " << EC.message() << "\n";
return;
}
- OwningPtr<MachOObjectFile> MachOOF(static_cast<MachOObjectFile*>(
- ObjectFile::createMachOObjectFile(Buff.take())));
- MachOObject *MachOObj = MachOOF->getObject();
+ std::unique_ptr<MachOObjectFile> MachOOF =
+ std::move(ObjectFile::createMachOObjectFile(Buff.get()).get());
- const Target *TheTarget = GetTarget(MachOObj);
+ DisassembleInputMachO2(Filename, MachOOF.get());
+}
+
+static void DisassembleInputMachO2(StringRef Filename,
+ MachOObjectFile *MachOOF) {
+ const Target *TheTarget = GetTarget(MachOOF);
if (!TheTarget) {
// GetTarget prints out stuff.
return;
}
- OwningPtr<const MCInstrInfo> InstrInfo(TheTarget->createMCInstrInfo());
- OwningPtr<MCInstrAnalysis>
- InstrAnalysis(TheTarget->createMCInstrAnalysis(InstrInfo.get()));
+ std::unique_ptr<const MCInstrInfo> InstrInfo(TheTarget->createMCInstrInfo());
+ std::unique_ptr<MCInstrAnalysis> InstrAnalysis(
+ TheTarget->createMCInstrAnalysis(InstrInfo.get()));
+
+ // Package up features to be passed to target/subtarget
+ std::string FeaturesStr;
+ if (MAttrs.size()) {
+ SubtargetFeatures Features;
+ for (unsigned i = 0; i != MAttrs.size(); ++i)
+ Features.AddFeature(MAttrs[i]);
+ FeaturesStr = Features.getString();
+ }
// Set up disassembler.
- OwningPtr<const MCAsmInfo> AsmInfo(TheTarget->createMCAsmInfo(TripleName));
- OwningPtr<const MCSubtargetInfo>
- STI(TheTarget->createMCSubtargetInfo(TripleName, "", ""));
- OwningPtr<const MCDisassembler> DisAsm(TheTarget->createMCDisassembler(*STI));
+ std::unique_ptr<const MCRegisterInfo> MRI(
+ TheTarget->createMCRegInfo(TripleName));
+ std::unique_ptr<const MCAsmInfo> AsmInfo(
+ TheTarget->createMCAsmInfo(*MRI, TripleName));
+ std::unique_ptr<const MCSubtargetInfo> STI(
+ TheTarget->createMCSubtargetInfo(TripleName, MCPU, FeaturesStr));
+ MCContext Ctx(AsmInfo.get(), MRI.get(), nullptr);
+ std::unique_ptr<const MCDisassembler> DisAsm(
+ TheTarget->createMCDisassembler(*STI, Ctx));
int AsmPrinterVariant = AsmInfo->getAssemblerDialect();
- OwningPtr<MCInstPrinter> IP(TheTarget->createMCInstPrinter(
- AsmPrinterVariant, *AsmInfo, *STI));
+ std::unique_ptr<MCInstPrinter> IP(TheTarget->createMCInstPrinter(
+ AsmPrinterVariant, *AsmInfo, *InstrInfo, *MRI, *STI));
if (!InstrAnalysis || !AsmInfo || !STI || !DisAsm || !IP) {
errs() << "error: couldn't initialize disassembler for target "
outs() << '\n' << Filename << ":\n\n";
- const macho::Header &Header = MachOObj->getHeader();
-
- const MachOObject::LoadCommandInfo *SymtabLCI = 0;
- // First, find the symbol table segment.
- for (unsigned i = 0; i != Header.NumLoadCommands; ++i) {
- const MachOObject::LoadCommandInfo &LCI = MachOObj->getLoadCommandInfo(i);
- if (LCI.Command.Type == macho::LCT_Symtab) {
- SymtabLCI = &LCI;
- break;
- }
- }
-
- // Read and register the symbol table data.
- InMemoryStruct<macho::SymtabLoadCommand> SymtabLC;
- MachOObj->ReadSymtabLoadCommand(*SymtabLCI, SymtabLC);
- MachOObj->RegisterStringTable(*SymtabLC);
+ MachO::mach_header Header = MachOOF->getHeader();
+ // FIXME: FoundFns isn't used anymore. Using symbols/LC_FUNCTION_STARTS to
+ // determine function locations will eventually go in MCObjectDisassembler.
+ // FIXME: Using the -cfg command line option, this code used to be able to
+ // annotate relocations with the referenced symbol's name, and if this was
+ // inside a __[cf]string section, the data it points to. This is now replaced
+ // by the upcoming MCSymbolizer, which needs the appropriate setup done above.
std::vector<SectionRef> Sections;
std::vector<SymbolRef> Symbols;
SmallVector<uint64_t, 8> FoundFns;
+ uint64_t BaseSegmentAddress;
- getSectionsAndSymbols(Header, MachOOF.get(), &SymtabLC, Sections, Symbols,
- FoundFns);
+ getSectionsAndSymbols(Header, MachOOF, Sections, Symbols, FoundFns,
+ BaseSegmentAddress);
- // Make a copy of the unsorted symbol list. FIXME: duplication
- std::vector<SymbolRef> UnsortedSymbols(Symbols);
// Sort the symbols by address, just in case they didn't come in that way.
std::sort(Symbols.begin(), Symbols.end(), SymbolSorter());
+ // Build a data in code table that is sorted on by the address of each entry.
+ uint64_t BaseAddress = 0;
+ if (Header.filetype == MachO::MH_OBJECT)
+ Sections[0].getAddress(BaseAddress);
+ else
+ BaseAddress = BaseSegmentAddress;
+ DiceTable Dices;
+ for (dice_iterator DI = MachOOF->begin_dices(), DE = MachOOF->end_dices();
+ DI != DE; ++DI) {
+ uint32_t Offset;
+ DI->getOffset(Offset);
+ Dices.push_back(std::make_pair(BaseAddress + Offset, *DI));
+ }
+ array_pod_sort(Dices.begin(), Dices.end());
+
#ifndef NDEBUG
raw_ostream &DebugOut = DebugFlag ? dbgs() : nulls();
#else
raw_ostream &DebugOut = nulls();
#endif
- StringRef DebugAbbrevSection, DebugInfoSection, DebugArangesSection,
- DebugLineSection, DebugStrSection;
- OwningPtr<DIContext> diContext;
- OwningPtr<MachOObjectFile> DSYMObj;
- MachOObject *DbgInfoObj = MachOObj;
+ std::unique_ptr<DIContext> diContext;
+ ObjectFile *DbgObj = MachOOF;
// Try to find debug info and set up the DIContext for it.
if (UseDbg) {
- ArrayRef<SectionRef> DebugSections = Sections;
- std::vector<SectionRef> DSYMSections;
-
// 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()) {
- OwningPtr<MemoryBuffer> Buf;
- if (error_code ec = MemoryBuffer::getFileOrSTDIN(DSYMFile.c_str(), Buf)) {
- errs() << "llvm-objdump: " << Filename << ": " << ec.message() << '\n';
+ ErrorOr<std::unique_ptr<MemoryBuffer>> Buf =
+ MemoryBuffer::getFileOrSTDIN(DSYMFile);
+ if (std::error_code EC = Buf.getError()) {
+ errs() << "llvm-objdump: " << Filename << ": " << EC.message() << '\n';
return;
}
- DSYMObj.reset(static_cast<MachOObjectFile*>(
- ObjectFile::createMachOObjectFile(Buf.take())));
- const macho::Header &Header = DSYMObj->getObject()->getHeader();
-
- std::vector<SymbolRef> Symbols;
- SmallVector<uint64_t, 8> FoundFns;
- getSectionsAndSymbols(Header, DSYMObj.get(), 0, DSYMSections, Symbols,
- FoundFns);
- DebugSections = DSYMSections;
- DbgInfoObj = DSYMObj.get()->getObject();
+ DbgObj = ObjectFile::createMachOObjectFile(Buf.get()).get().release();
}
- // Find the named debug info sections.
- for (unsigned SectIdx = 0; SectIdx != DebugSections.size(); SectIdx++) {
- StringRef SectName;
- if (!DebugSections[SectIdx].getName(SectName)) {
- if (SectName.equals("__DWARF,__debug_abbrev"))
- DebugSections[SectIdx].getContents(DebugAbbrevSection);
- else if (SectName.equals("__DWARF,__debug_info"))
- DebugSections[SectIdx].getContents(DebugInfoSection);
- else if (SectName.equals("__DWARF,__debug_aranges"))
- DebugSections[SectIdx].getContents(DebugArangesSection);
- else if (SectName.equals("__DWARF,__debug_line"))
- DebugSections[SectIdx].getContents(DebugLineSection);
- else if (SectName.equals("__DWARF,__debug_str"))
- DebugSections[SectIdx].getContents(DebugStrSection);
- }
- }
-
- // Setup the DIContext.
- diContext.reset(DIContext::getDWARFContext(DbgInfoObj->isLittleEndian(),
- DebugInfoSection,
- DebugAbbrevSection,
- DebugArangesSection,
- DebugLineSection,
- DebugStrSection));
+ // Setup the DIContext
+ diContext.reset(DIContext::getDWARFContext(*DbgObj));
}
- FunctionMapTy FunctionMap;
- FunctionListTy Functions;
-
for (unsigned SectIdx = 0; SectIdx != Sections.size(); SectIdx++) {
+
+ bool SectIsText = false;
+ Sections[SectIdx].isText(SectIsText);
+ if (SectIsText == false)
+ continue;
+
StringRef SectName;
if (Sections[SectIdx].getName(SectName) ||
- SectName.compare("__TEXT,__text"))
+ SectName != "__text")
continue; // Skip non-text sections
- // Insert the functions from the function starts segment into our map.
- uint64_t VMAddr;
- Sections[SectIdx].getAddress(VMAddr);
- for (unsigned i = 0, e = FoundFns.size(); i != e; ++i) {
- StringRef SectBegin;
- Sections[SectIdx].getContents(SectBegin);
- uint64_t Offset = (uint64_t)SectBegin.data();
- FunctionMap.insert(std::make_pair(VMAddr + FoundFns[i]-Offset,
- (MCFunction*)0));
- }
+ DataRefImpl DR = Sections[SectIdx].getRawDataRefImpl();
+
+ StringRef SegmentName = MachOOF->getSectionFinalSegmentName(DR);
+ if (SegmentName != "__TEXT")
+ continue;
StringRef Bytes;
Sections[SectIdx].getContents(Bytes);
bool symbolTableWorked = false;
// Parse relocations.
- std::vector<std::pair<uint64_t, SymbolRef> > Relocs;
- error_code ec;
- for (relocation_iterator RI = Sections[SectIdx].begin_relocations(),
- RE = Sections[SectIdx].end_relocations(); RI != RE; RI.increment(ec)) {
+ std::vector<std::pair<uint64_t, SymbolRef>> Relocs;
+ for (const RelocationRef &Reloc : Sections[SectIdx].relocations()) {
uint64_t RelocOffset, SectionAddress;
- RI->getAddress(RelocOffset);
+ Reloc.getOffset(RelocOffset);
Sections[SectIdx].getAddress(SectionAddress);
RelocOffset -= SectionAddress;
- SymbolRef RelocSym;
- RI->getSymbol(RelocSym);
+ symbol_iterator RelocSym = Reloc.getSymbol();
- Relocs.push_back(std::make_pair(RelocOffset, RelocSym));
+ Relocs.push_back(std::make_pair(RelocOffset, *RelocSym));
}
array_pod_sort(Relocs.begin(), Relocs.end());
continue;
// Start at the address of the symbol relative to the section's address.
+ uint64_t SectionAddress = 0;
uint64_t Start = 0;
- Symbols[SymIdx].getOffset(Start);
+ Sections[SectIdx].getAddress(SectionAddress);
+ Symbols[SymIdx].getAddress(Start);
+ Start -= SectionAddress;
// Stop disassembling either at the beginning of the next symbol or at
// the end of the section.
- bool containsNextSym = true;
+ bool containsNextSym = false;
uint64_t NextSym = 0;
uint64_t NextSymIdx = SymIdx+1;
while (Symbols.size() > NextSymIdx) {
if (NextSymType == SymbolRef::ST_Function) {
Sections[SectIdx].containsSymbol(Symbols[NextSymIdx],
containsNextSym);
- Symbols[NextSymIdx].getOffset(NextSym);
+ Symbols[NextSymIdx].getAddress(NextSym);
+ NextSym -= SectionAddress;
break;
}
++NextSymIdx;
symbolTableWorked = true;
- if (!CFG) {
- // Normal disassembly, print addresses, bytes and mnemonic form.
- StringRef SymName;
- Symbols[SymIdx].getName(SymName);
-
- outs() << SymName << ":\n";
- DILineInfo lastLine;
- for (uint64_t Index = Start; Index < End; Index += Size) {
- MCInst Inst;
-
- if (DisAsm->getInstruction(Inst, Size, memoryObject, Index,
- DebugOut, nulls())) {
- uint64_t SectAddress = 0;
- Sections[SectIdx].getAddress(SectAddress);
- outs() << format("%8" PRIx64 ":\t", SectAddress + Index);
-
- DumpBytes(StringRef(Bytes.data() + Index, Size));
- IP->printInst(&Inst, outs(), "");
-
- // Print debug info.
- if (diContext) {
- DILineInfo dli =
- diContext->getLineInfoForAddress(SectAddress + Index);
- // Print valid line info if it changed.
- if (dli != lastLine && dli.getLine() != 0)
- outs() << "\t## " << dli.getFileName() << ':'
- << dli.getLine() << ':' << dli.getColumn();
- lastLine = dli;
- }
- outs() << "\n";
- } else {
- errs() << "llvm-objdump: warning: invalid instruction encoding\n";
- if (Size == 0)
- Size = 1; // skip illegible bytes
+ outs() << SymName << ":\n";
+ DILineInfo lastLine;
+ for (uint64_t Index = Start; Index < End; Index += Size) {
+ MCInst Inst;
+
+ uint64_t SectAddress = 0;
+ Sections[SectIdx].getAddress(SectAddress);
+ outs() << format("%8" PRIx64 ":\t", SectAddress + Index);
+
+ // Check the data in code table here to see if this is data not an
+ // instruction to be disassembled.
+ DiceTable Dice;
+ Dice.push_back(std::make_pair(SectAddress + Index, DiceRef()));
+ dice_table_iterator DTI = std::search(Dices.begin(), Dices.end(),
+ Dice.begin(), Dice.end(),
+ compareDiceTableEntries);
+ if (DTI != Dices.end()){
+ uint16_t Length;
+ DTI->second.getLength(Length);
+ DumpBytes(StringRef(Bytes.data() + Index, Length));
+ uint16_t Kind;
+ DTI->second.getKind(Kind);
+ DumpDataInCode(Bytes.data() + Index, Length, Kind);
+ continue;
+ }
+
+ if (DisAsm->getInstruction(Inst, Size, memoryObject, Index,
+ DebugOut, nulls())) {
+ DumpBytes(StringRef(Bytes.data() + Index, Size));
+ IP->printInst(&Inst, outs(), "");
+
+ // Print debug info.
+ if (diContext) {
+ DILineInfo dli =
+ diContext->getLineInfoForAddress(SectAddress + Index);
+ // Print valid line info if it changed.
+ if (dli != lastLine && dli.Line != 0)
+ outs() << "\t## " << dli.FileName << ':' << dli.Line << ':'
+ << dli.Column;
+ lastLine = dli;
}
+ outs() << "\n";
+ } else {
+ errs() << "llvm-objdump: warning: invalid instruction encoding\n";
+ if (Size == 0)
+ Size = 1; // skip illegible bytes
}
- } else {
- // Create CFG and use it for disassembly.
- StringRef SymName;
- Symbols[SymIdx].getName(SymName);
- createMCFunctionAndSaveCalls(
- SymName, DisAsm.get(), memoryObject, Start, End,
- InstrAnalysis.get(), Start, DebugOut, FunctionMap, Functions);
}
}
-
- if (CFG) {
- if (!symbolTableWorked) {
- // Reading the symbol table didn't work, create a big __TEXT symbol.
- uint64_t SectSize = 0, SectAddress = 0;
- Sections[SectIdx].getSize(SectSize);
- Sections[SectIdx].getAddress(SectAddress);
- createMCFunctionAndSaveCalls("__TEXT", DisAsm.get(), memoryObject,
- 0, SectSize,
- InstrAnalysis.get(),
- SectAddress, DebugOut,
- FunctionMap, Functions);
- }
- for (std::map<uint64_t, MCFunction*>::iterator mi = FunctionMap.begin(),
- me = FunctionMap.end(); mi != me; ++mi)
- if (mi->second == 0) {
- // Create functions for the remaining callees we have gathered,
- // but we didn't find a name for them.
- uint64_t SectSize = 0;
- Sections[SectIdx].getSize(SectSize);
-
- SmallVector<uint64_t, 16> Calls;
- MCFunction f =
- MCFunction::createFunctionFromMC("unknown", DisAsm.get(),
- memoryObject, mi->first,
- SectSize,
- InstrAnalysis.get(), DebugOut,
- Calls);
- Functions.push_back(f);
- mi->second = &Functions.back();
- for (unsigned i = 0, e = Calls.size(); i != e; ++i) {
- std::pair<uint64_t, MCFunction*> p(Calls[i], (MCFunction*)0);
- if (FunctionMap.insert(p).second)
- mi = FunctionMap.begin();
- }
+ if (!symbolTableWorked) {
+ // Reading the symbol table didn't work, disassemble the whole section.
+ uint64_t SectAddress;
+ Sections[SectIdx].getAddress(SectAddress);
+ uint64_t SectSize;
+ Sections[SectIdx].getSize(SectSize);
+ uint64_t InstSize;
+ for (uint64_t Index = 0; Index < SectSize; Index += InstSize) {
+ MCInst Inst;
+
+ if (DisAsm->getInstruction(Inst, InstSize, memoryObject, Index,
+ DebugOut, nulls())) {
+ outs() << format("%8" PRIx64 ":\t", SectAddress + Index);
+ DumpBytes(StringRef(Bytes.data() + Index, InstSize));
+ IP->printInst(&Inst, outs(), "");
+ outs() << "\n";
+ } else {
+ errs() << "llvm-objdump: warning: invalid instruction encoding\n";
+ if (InstSize == 0)
+ InstSize = 1; // skip illegible bytes
}
+ }
+ }
+ }
+}
- DenseSet<uint64_t> PrintedBlocks;
- for (unsigned ffi = 0, ffe = Functions.size(); ffi != ffe; ++ffi) {
- MCFunction &f = Functions[ffi];
- for (MCFunction::iterator fi = f.begin(), fe = f.end(); fi != fe; ++fi){
- if (!PrintedBlocks.insert(fi->first).second)
- continue; // We already printed this block.
-
- // We assume a block has predecessors when it's the first block after
- // a symbol.
- bool hasPreds = FunctionMap.find(fi->first) != FunctionMap.end();
-
- // See if this block has predecessors.
- // FIXME: Slow.
- for (MCFunction::iterator pi = f.begin(), pe = f.end(); pi != pe;
- ++pi)
- if (pi->second.contains(fi->first)) {
- hasPreds = true;
- break;
- }
-
- uint64_t SectSize = 0, SectAddress;
- Sections[SectIdx].getSize(SectSize);
- Sections[SectIdx].getAddress(SectAddress);
-
- // No predecessors, this is a data block. Print as .byte directives.
- if (!hasPreds) {
- uint64_t End = llvm::next(fi) == fe ? SectSize :
- llvm::next(fi)->first;
- outs() << "# " << End-fi->first << " bytes of data:\n";
- for (unsigned pos = fi->first; pos != End; ++pos) {
- outs() << format("%8x:\t", SectAddress + pos);
- DumpBytes(StringRef(Bytes.data() + pos, 1));
- outs() << format("\t.byte 0x%02x\n", (uint8_t)Bytes[pos]);
- }
- continue;
- }
- if (fi->second.contains(fi->first)) // Print a header for simple loops
- outs() << "# Loop begin:\n";
-
- DILineInfo lastLine;
- // Walk over the instructions and print them.
- for (unsigned ii = 0, ie = fi->second.getInsts().size(); ii != ie;
- ++ii) {
- const MCDecodedInst &Inst = fi->second.getInsts()[ii];
-
- // If there's a symbol at this address, print its name.
- if (FunctionMap.find(SectAddress + Inst.Address) !=
- FunctionMap.end())
- outs() << FunctionMap[SectAddress + Inst.Address]-> getName()
- << ":\n";
-
- outs() << format("%8" PRIx64 ":\t", SectAddress + Inst.Address);
- DumpBytes(StringRef(Bytes.data() + Inst.Address, Inst.Size));
-
- if (fi->second.contains(fi->first)) // Indent simple loops.
- outs() << '\t';
-
- IP->printInst(&Inst.Inst, outs(), "");
-
- // Look for relocations inside this instructions, if there is one
- // print its target and additional information if available.
- for (unsigned j = 0; j != Relocs.size(); ++j)
- if (Relocs[j].first >= SectAddress + Inst.Address &&
- Relocs[j].first < SectAddress + Inst.Address + Inst.Size) {
- StringRef SymName;
- uint64_t Addr;
- Relocs[j].second.getAddress(Addr);
- Relocs[j].second.getName(SymName);
-
- outs() << "\t# " << SymName << ' ';
- DumpAddress(Addr, Sections, MachOObj, outs());
- }
-
- // If this instructions contains an address, see if we can evaluate
- // it and print additional information.
- uint64_t targ = InstrAnalysis->evaluateBranch(Inst.Inst,
- Inst.Address,
- Inst.Size);
- if (targ != -1ULL)
- DumpAddress(targ, Sections, MachOObj, outs());
-
- // Print debug info.
- if (diContext) {
- DILineInfo dli =
- diContext->getLineInfoForAddress(SectAddress + Inst.Address);
- // Print valid line info if it changed.
- if (dli != lastLine && dli.getLine() != 0)
- outs() << "\t## " << dli.getFileName() << ':'
- << dli.getLine() << ':' << dli.getColumn();
- lastLine = dli;
- }
-
- outs() << '\n';
- }
- }
+//===----------------------------------------------------------------------===//
+// __compact_unwind section dumping
+//===----------------------------------------------------------------------===//
- emitDOTFile((f.getName().str() + ".dot").c_str(), f, IP.get());
- }
+namespace {
+
+template <typename T> static uint64_t readNext(const char *&Buf) {
+ using llvm::support::little;
+ using llvm::support::unaligned;
+
+ uint64_t Val = support::endian::read<T, little, unaligned>(Buf);
+ Buf += sizeof(T);
+ return Val;
+ }
+
+struct CompactUnwindEntry {
+ uint32_t OffsetInSection;
+
+ uint64_t FunctionAddr;
+ uint32_t Length;
+ uint32_t CompactEncoding;
+ uint64_t PersonalityAddr;
+ uint64_t LSDAAddr;
+
+ RelocationRef FunctionReloc;
+ RelocationRef PersonalityReloc;
+ RelocationRef LSDAReloc;
+
+ CompactUnwindEntry(StringRef Contents, unsigned Offset, bool Is64)
+ : OffsetInSection(Offset) {
+ if (Is64)
+ read<uint64_t>(Contents.data() + Offset);
+ else
+ read<uint32_t>(Contents.data() + Offset);
+ }
+
+private:
+ template<typename UIntPtr>
+ void read(const char *Buf) {
+ FunctionAddr = readNext<UIntPtr>(Buf);
+ Length = readNext<uint32_t>(Buf);
+ CompactEncoding = readNext<uint32_t>(Buf);
+ PersonalityAddr = readNext<UIntPtr>(Buf);
+ LSDAAddr = readNext<UIntPtr>(Buf);
+ }
+};
+}
+
+/// Given a relocation from __compact_unwind, consisting of the RelocationRef
+/// and data being relocated, determine the best base Name and Addend to use for
+/// display purposes.
+///
+/// 1. An Extern relocation will directly reference a symbol (and the data is
+/// then already an addend), so use that.
+/// 2. Otherwise the data is an offset in the object file's layout; try to find
+// a symbol before it in the same section, and use the offset from there.
+/// 3. Finally, if all that fails, fall back to an offset from the start of the
+/// referenced section.
+static void findUnwindRelocNameAddend(const MachOObjectFile *Obj,
+ std::map<uint64_t, SymbolRef> &Symbols,
+ const RelocationRef &Reloc,
+ uint64_t Addr,
+ StringRef &Name, uint64_t &Addend) {
+ if (Reloc.getSymbol() != Obj->symbol_end()) {
+ Reloc.getSymbol()->getName(Name);
+ Addend = Addr;
+ return;
+ }
+
+ auto RE = Obj->getRelocation(Reloc.getRawDataRefImpl());
+ SectionRef RelocSection = Obj->getRelocationSection(RE);
+
+ uint64_t SectionAddr;
+ RelocSection.getAddress(SectionAddr);
+
+ auto Sym = Symbols.upper_bound(Addr);
+ if (Sym == Symbols.begin()) {
+ // The first symbol in the object is after this reference, the best we can
+ // do is section-relative notation.
+ RelocSection.getName(Name);
+ Addend = Addr - SectionAddr;
+ return;
+ }
+
+ // Go back one so that SymbolAddress <= Addr.
+ --Sym;
+
+ section_iterator SymSection = Obj->section_end();
+ Sym->second.getSection(SymSection);
+ if (RelocSection == *SymSection) {
+ // There's a valid symbol in the same section before this reference.
+ Sym->second.getName(Name);
+ Addend = Addr - Sym->first;
+ return;
+ }
+
+ // There is a symbol before this reference, but it's in a different
+ // section. Probably not helpful to mention it, so use the section name.
+ RelocSection.getName(Name);
+ Addend = Addr - SectionAddr;
+}
+
+static void printUnwindRelocDest(const MachOObjectFile *Obj,
+ std::map<uint64_t, SymbolRef> &Symbols,
+ const RelocationRef &Reloc,
+ uint64_t Addr) {
+ StringRef Name;
+ uint64_t Addend;
+
+ findUnwindRelocNameAddend(Obj, Symbols, Reloc, Addr, Name, Addend);
+
+ outs() << Name;
+ if (Addend)
+ outs() << " + " << format("0x%" PRIx64, Addend);
+}
+
+static void
+printMachOCompactUnwindSection(const MachOObjectFile *Obj,
+ std::map<uint64_t, SymbolRef> &Symbols,
+ const SectionRef &CompactUnwind) {
+
+ assert(Obj->isLittleEndian() &&
+ "There should not be a big-endian .o with __compact_unwind");
+
+ bool Is64 = Obj->is64Bit();
+ uint32_t PointerSize = Is64 ? sizeof(uint64_t) : sizeof(uint32_t);
+ uint32_t EntrySize = 3 * PointerSize + 2 * sizeof(uint32_t);
+
+ StringRef Contents;
+ CompactUnwind.getContents(Contents);
+
+ SmallVector<CompactUnwindEntry, 4> CompactUnwinds;
+
+ // First populate the initial raw offsets, encodings and so on from the entry.
+ for (unsigned Offset = 0; Offset < Contents.size(); Offset += EntrySize) {
+ CompactUnwindEntry Entry(Contents.data(), Offset, Is64);
+ CompactUnwinds.push_back(Entry);
+ }
+
+ // Next we need to look at the relocations to find out what objects are
+ // actually being referred to.
+ for (const RelocationRef &Reloc : CompactUnwind.relocations()) {
+ uint64_t RelocAddress;
+ Reloc.getOffset(RelocAddress);
+
+ uint32_t EntryIdx = RelocAddress / EntrySize;
+ uint32_t OffsetInEntry = RelocAddress - EntryIdx * EntrySize;
+ CompactUnwindEntry &Entry = CompactUnwinds[EntryIdx];
+
+ if (OffsetInEntry == 0)
+ Entry.FunctionReloc = Reloc;
+ else if (OffsetInEntry == PointerSize + 2 * sizeof(uint32_t))
+ Entry.PersonalityReloc = Reloc;
+ else if (OffsetInEntry == 2 * PointerSize + 2 * sizeof(uint32_t))
+ Entry.LSDAReloc = Reloc;
+ else
+ llvm_unreachable("Unexpected relocation in __compact_unwind section");
+ }
+
+ // Finally, we're ready to print the data we've gathered.
+ outs() << "Contents of __compact_unwind section:\n";
+ for (auto &Entry : CompactUnwinds) {
+ outs() << " Entry at offset "
+ << format("0x%" PRIx32, Entry.OffsetInSection) << ":\n";
+
+ // 1. Start of the region this entry applies to.
+ outs() << " start: "
+ << format("0x%" PRIx64, Entry.FunctionAddr) << ' ';
+ printUnwindRelocDest(Obj, Symbols, Entry.FunctionReloc,
+ Entry.FunctionAddr);
+ outs() << '\n';
+
+ // 2. Length of the region this entry applies to.
+ outs() << " length: "
+ << format("0x%" PRIx32, Entry.Length) << '\n';
+ // 3. The 32-bit compact encoding.
+ outs() << " compact encoding: "
+ << format("0x%08" PRIx32, Entry.CompactEncoding) << '\n';
+
+ // 4. The personality function, if present.
+ if (Entry.PersonalityReloc.getObjectFile()) {
+ outs() << " personality function: "
+ << format("0x%" PRIx64, Entry.PersonalityAddr) << ' ';
+ printUnwindRelocDest(Obj, Symbols, Entry.PersonalityReloc,
+ Entry.PersonalityAddr);
+ outs() << '\n';
+ }
+
+ // 5. This entry's language-specific data area.
+ if (Entry.LSDAReloc.getObjectFile()) {
+ outs() << " LSDA: "
+ << format("0x%" PRIx64, Entry.LSDAAddr) << ' ';
+ printUnwindRelocDest(Obj, Symbols, Entry.LSDAReloc, Entry.LSDAAddr);
+ outs() << '\n';
}
}
}
+
+//===----------------------------------------------------------------------===//
+// __unwind_info section dumping
+//===----------------------------------------------------------------------===//
+
+static void printRegularSecondLevelUnwindPage(const char *PageStart) {
+ const char *Pos = PageStart;
+ uint32_t Kind = readNext<uint32_t>(Pos);
+ (void)Kind;
+ assert(Kind == 2 && "kind for a regular 2nd level index should be 2");
+
+ uint16_t EntriesStart = readNext<uint16_t>(Pos);
+ uint16_t NumEntries = readNext<uint16_t>(Pos);
+
+ Pos = PageStart + EntriesStart;
+ for (unsigned i = 0; i < NumEntries; ++i) {
+ uint32_t FunctionOffset = readNext<uint32_t>(Pos);
+ uint32_t Encoding = readNext<uint32_t>(Pos);
+
+ outs() << " [" << i << "]: "
+ << "function offset="
+ << format("0x%08" PRIx32, FunctionOffset) << ", "
+ << "encoding="
+ << format("0x%08" PRIx32, Encoding)
+ << '\n';
+ }
+}
+
+static void printCompressedSecondLevelUnwindPage(
+ const char *PageStart, uint32_t FunctionBase,
+ const SmallVectorImpl<uint32_t> &CommonEncodings) {
+ const char *Pos = PageStart;
+ uint32_t Kind = readNext<uint32_t>(Pos);
+ (void)Kind;
+ assert(Kind == 3 && "kind for a compressed 2nd level index should be 3");
+
+ uint16_t EntriesStart = readNext<uint16_t>(Pos);
+ uint16_t NumEntries = readNext<uint16_t>(Pos);
+
+ uint16_t EncodingsStart = readNext<uint16_t>(Pos);
+ readNext<uint16_t>(Pos);
+ const auto *PageEncodings = reinterpret_cast<const support::ulittle32_t *>(
+ PageStart + EncodingsStart);
+
+ Pos = PageStart + EntriesStart;
+ for (unsigned i = 0; i < NumEntries; ++i) {
+ uint32_t Entry = readNext<uint32_t>(Pos);
+ uint32_t FunctionOffset = FunctionBase + (Entry & 0xffffff);
+ uint32_t EncodingIdx = Entry >> 24;
+
+ uint32_t Encoding;
+ if (EncodingIdx < CommonEncodings.size())
+ Encoding = CommonEncodings[EncodingIdx];
+ else
+ Encoding = PageEncodings[EncodingIdx - CommonEncodings.size()];
+
+ outs() << " [" << i << "]: "
+ << "function offset="
+ << format("0x%08" PRIx32, FunctionOffset) << ", "
+ << "encoding[" << EncodingIdx << "]="
+ << format("0x%08" PRIx32, Encoding)
+ << '\n';
+ }
+}
+
+static void
+printMachOUnwindInfoSection(const MachOObjectFile *Obj,
+ std::map<uint64_t, SymbolRef> &Symbols,
+ const SectionRef &UnwindInfo) {
+
+ assert(Obj->isLittleEndian() &&
+ "There should not be a big-endian .o with __unwind_info");
+
+ outs() << "Contents of __unwind_info section:\n";
+
+ StringRef Contents;
+ UnwindInfo.getContents(Contents);
+ const char *Pos = Contents.data();
+
+ //===----------------------------------
+ // Section header
+ //===----------------------------------
+
+ uint32_t Version = readNext<uint32_t>(Pos);
+ outs() << " Version: "
+ << format("0x%" PRIx32, Version) << '\n';
+ assert(Version == 1 && "only understand version 1");
+
+ uint32_t CommonEncodingsStart = readNext<uint32_t>(Pos);
+ outs() << " Common encodings array section offset: "
+ << format("0x%" PRIx32, CommonEncodingsStart) << '\n';
+ uint32_t NumCommonEncodings = readNext<uint32_t>(Pos);
+ outs() << " Number of common encodings in array: "
+ << format("0x%" PRIx32, NumCommonEncodings) << '\n';
+
+ uint32_t PersonalitiesStart = readNext<uint32_t>(Pos);
+ outs() << " Personality function array section offset: "
+ << format("0x%" PRIx32, PersonalitiesStart) << '\n';
+ uint32_t NumPersonalities = readNext<uint32_t>(Pos);
+ outs() << " Number of personality functions in array: "
+ << format("0x%" PRIx32, NumPersonalities) << '\n';
+
+ uint32_t IndicesStart = readNext<uint32_t>(Pos);
+ outs() << " Index array section offset: "
+ << format("0x%" PRIx32, IndicesStart) << '\n';
+ uint32_t NumIndices = readNext<uint32_t>(Pos);
+ outs() << " Number of indices in array: "
+ << format("0x%" PRIx32, NumIndices) << '\n';
+
+ //===----------------------------------
+ // A shared list of common encodings
+ //===----------------------------------
+
+ // These occupy indices in the range [0, N] whenever an encoding is referenced
+ // from a compressed 2nd level index table. In practice the linker only
+ // creates ~128 of these, so that indices are available to embed encodings in
+ // the 2nd level index.
+
+ SmallVector<uint32_t, 64> CommonEncodings;
+ outs() << " Common encodings: (count = " << NumCommonEncodings << ")\n";
+ Pos = Contents.data() + CommonEncodingsStart;
+ for (unsigned i = 0; i < NumCommonEncodings; ++i) {
+ uint32_t Encoding = readNext<uint32_t>(Pos);
+ CommonEncodings.push_back(Encoding);
+
+ outs() << " encoding[" << i << "]: " << format("0x%08" PRIx32, Encoding)
+ << '\n';
+ }
+
+
+ //===----------------------------------
+ // Personality functions used in this executable
+ //===----------------------------------
+
+ // There should be only a handful of these (one per source language,
+ // roughly). Particularly since they only get 2 bits in the compact encoding.
+
+ outs() << " Personality functions: (count = " << NumPersonalities << ")\n";
+ Pos = Contents.data() + PersonalitiesStart;
+ for (unsigned i = 0; i < NumPersonalities; ++i) {
+ uint32_t PersonalityFn = readNext<uint32_t>(Pos);
+ outs() << " personality[" << i + 1
+ << "]: " << format("0x%08" PRIx32, PersonalityFn) << '\n';
+ }
+
+ //===----------------------------------
+ // The level 1 index entries
+ //===----------------------------------
+
+ // These specify an approximate place to start searching for the more detailed
+ // information, sorted by PC.
+
+ struct IndexEntry {
+ uint32_t FunctionOffset;
+ uint32_t SecondLevelPageStart;
+ uint32_t LSDAStart;
+ };
+
+ SmallVector<IndexEntry, 4> IndexEntries;
+
+ outs() << " Top level indices: (count = " << NumIndices << ")\n";
+ Pos = Contents.data() + IndicesStart;
+ for (unsigned i = 0; i < NumIndices; ++i) {
+ IndexEntry Entry;
+
+ Entry.FunctionOffset = readNext<uint32_t>(Pos);
+ Entry.SecondLevelPageStart = readNext<uint32_t>(Pos);
+ Entry.LSDAStart = readNext<uint32_t>(Pos);
+ IndexEntries.push_back(Entry);
+
+ outs() << " [" << i << "]: "
+ << "function offset="
+ << format("0x%08" PRIx32, Entry.FunctionOffset) << ", "
+ << "2nd level page offset="
+ << format("0x%08" PRIx32, Entry.SecondLevelPageStart) << ", "
+ << "LSDA offset="
+ << format("0x%08" PRIx32, Entry.LSDAStart) << '\n';
+ }
+
+
+ //===----------------------------------
+ // Next come the LSDA tables
+ //===----------------------------------
+
+ // The LSDA layout is rather implicit: it's a contiguous array of entries from
+ // the first top-level index's LSDAOffset to the last (sentinel).
+
+ outs() << " LSDA descriptors:\n";
+ Pos = Contents.data() + IndexEntries[0].LSDAStart;
+ int NumLSDAs = (IndexEntries.back().LSDAStart - IndexEntries[0].LSDAStart) /
+ (2 * sizeof(uint32_t));
+ for (int i = 0; i < NumLSDAs; ++i) {
+ uint32_t FunctionOffset = readNext<uint32_t>(Pos);
+ uint32_t LSDAOffset = readNext<uint32_t>(Pos);
+ outs() << " [" << i << "]: "
+ << "function offset="
+ << format("0x%08" PRIx32, FunctionOffset) << ", "
+ << "LSDA offset="
+ << format("0x%08" PRIx32, LSDAOffset) << '\n';
+ }
+
+ //===----------------------------------
+ // Finally, the 2nd level indices
+ //===----------------------------------
+
+ // Generally these are 4K in size, and have 2 possible forms:
+ // + Regular stores up to 511 entries with disparate encodings
+ // + Compressed stores up to 1021 entries if few enough compact encoding
+ // values are used.
+ outs() << " Second level indices:\n";
+ for (unsigned i = 0; i < IndexEntries.size() - 1; ++i) {
+ // The final sentinel top-level index has no associated 2nd level page
+ if (IndexEntries[i].SecondLevelPageStart == 0)
+ break;
+
+ outs() << " Second level index[" << i << "]: "
+ << "offset in section="
+ << format("0x%08" PRIx32, IndexEntries[i].SecondLevelPageStart)
+ << ", "
+ << "base function offset="
+ << format("0x%08" PRIx32, IndexEntries[i].FunctionOffset) << '\n';
+
+ Pos = Contents.data() + IndexEntries[i].SecondLevelPageStart;
+ uint32_t Kind = *reinterpret_cast<const support::ulittle32_t *>(Pos);
+ if (Kind == 2)
+ printRegularSecondLevelUnwindPage(Pos);
+ else if (Kind == 3)
+ printCompressedSecondLevelUnwindPage(Pos, IndexEntries[i].FunctionOffset,
+ CommonEncodings);
+ else
+ llvm_unreachable("Do not know how to print this kind of 2nd level page");
+
+ }
+}
+
+void llvm::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
+ // in the convenience lookup for unwind info and just take up resources.
+ section_iterator Section = Obj->section_end();
+ SymRef.getSection(Section);
+ if (Section == Obj->section_end())
+ continue;
+
+ uint64_t Addr;
+ SymRef.getAddress(Addr);
+ Symbols.insert(std::make_pair(Addr, SymRef));
+ }
+
+ for (const SectionRef &Section : Obj->sections()) {
+ StringRef SectName;
+ Section.getName(SectName);
+ if (SectName == "__compact_unwind")
+ printMachOCompactUnwindSection(Obj, Symbols, Section);
+ else if (SectName == "__unwind_info")
+ printMachOUnwindInfoSection(Obj, Symbols, Section);
+ else if (SectName == "__eh_frame")
+ outs() << "llvm-objdump: warning: unhandled __eh_frame section\n";
+
+ }
+}