void DWARFDebugLoc::parse(DataExtractor data, unsigned AddressSize) {
uint32_t Offset = 0;
while (data.isValidOffset(Offset+AddressSize-1)) {
Locations.resize(Locations.size() + 1);
LocationList &Loc = Locations.back();
Loc.Offset = Offset;
// 2.6.2 Location Lists
// A location list entry consists of:
while (true) {
Entry E;
RelocAddrMap::const_iterator AI = RelocMap.find(Offset);
// 1. A beginning address offset. ...
E.Begin = data.getUnsigned(&Offset, AddressSize);
if (AI != RelocMap.end())
E.Begin += AI->second.second;
AI = RelocMap.find(Offset);
// 2. An ending address offset. ...
E.End = data.getUnsigned(&Offset, AddressSize);
if (AI != RelocMap.end())
E.End += AI->second.second;
// The end of any given location list is marked by an end of list entry,
// which consists of a 0 for the beginning address offset and a 0 for the
// ending address offset.
if (E.Begin == 0 && E.End == 0)
break;
unsigned Bytes = data.getU16(&Offset);
// A single location description describing the location of the object...
StringRef str = data.getData().substr(Offset, Bytes);
Offset += Bytes;
E.Loc.reserve(str.size());
std::copy(str.begin(), str.end(), std::back_inserter(E.Loc));
Loc.Entries.push_back(llvm_move(E));
}
}
if (data.isValidOffset(Offset))
llvm::errs() << "error: failed to consume entire .debug_loc section\n";
}
void DWARFDebugFrame::parse(DataExtractor Data) {
uint32_t Offset = 0;
DenseMap<uint32_t, CIE *> CIEs;
while (Data.isValidOffset(Offset)) {
uint32_t StartOffset = Offset;
bool IsDWARF64 = false;
uint64_t Length = Data.getU32(&Offset);
uint64_t Id;
if (Length == UINT32_MAX) {
// DWARF-64 is distinguished by the first 32 bits of the initial length
// field being 0xffffffff. Then, the next 64 bits are the actual entry
// length.
IsDWARF64 = true;
Length = Data.getU64(&Offset);
}
// At this point, Offset points to the next field after Length.
// Length is the structure size excluding itself. Compute an offset one
// past the end of the structure (needed to know how many instructions to
// read).
// TODO: For honest DWARF64 support, DataExtractor will have to treat
// offset_ptr as uint64_t*
uint32_t EndStructureOffset = Offset + static_cast<uint32_t>(Length);
// The Id field's size depends on the DWARF format
Id = Data.getUnsigned(&Offset, IsDWARF64 ? 8 : 4);
bool IsCIE = ((IsDWARF64 && Id == DW64_CIE_ID) || Id == DW_CIE_ID);
if (IsCIE) {
uint8_t Version = Data.getU8(&Offset);
const char *Augmentation = Data.getCStr(&Offset);
uint8_t AddressSize = Version < 4 ? Data.getAddressSize() : Data.getU8(&Offset);
Data.setAddressSize(AddressSize);
uint8_t SegmentDescriptorSize = Version < 4 ? 0 : Data.getU8(&Offset);
uint64_t CodeAlignmentFactor = Data.getULEB128(&Offset);
int64_t DataAlignmentFactor = Data.getSLEB128(&Offset);
uint64_t ReturnAddressRegister = Data.getULEB128(&Offset);
auto Cie = make_unique<CIE>(StartOffset, Length, Version,
StringRef(Augmentation), AddressSize,
SegmentDescriptorSize, CodeAlignmentFactor,
DataAlignmentFactor, ReturnAddressRegister);
CIEs[StartOffset] = Cie.get();
Entries.emplace_back(std::move(Cie));
} else {
// FDE
uint64_t CIEPointer = Id;
uint64_t InitialLocation = Data.getAddress(&Offset);
uint64_t AddressRange = Data.getAddress(&Offset);
Entries.emplace_back(new FDE(StartOffset, Length, CIEPointer,
InitialLocation, AddressRange,
CIEs[CIEPointer]));
}
Entries.back()->parseInstructions(Data, &Offset, EndStructureOffset);
if (Offset != EndStructureOffset) {
std::string Str;
raw_string_ostream OS(Str);
OS << format("Parsing entry instructions at %lx failed", StartOffset);
report_fatal_error(Str);
}
}
}
bool DWARFFormValue::extractValue(DataExtractor data, uint32_t *offset_ptr,
const DWARFUnit *cu) {
bool indirect = false;
bool is_block = false;
Value.data = nullptr;
// Read the value for the form into value and follow and DW_FORM_indirect
// instances we run into
do {
indirect = false;
switch (Form) {
case DW_FORM_addr:
case DW_FORM_ref_addr: {
uint16_t AddrSize =
(Form == DW_FORM_addr)
? cu->getAddressByteSize()
: getRefAddrSize(cu->getAddressByteSize(), cu->getVersion());
RelocAddrMap::const_iterator AI = cu->getRelocMap()->find(*offset_ptr);
if (AI != cu->getRelocMap()->end()) {
const std::pair<uint8_t, int64_t> &R = AI->second;
Value.uval = data.getUnsigned(offset_ptr, AddrSize) + R.second;
} else
Value.uval = data.getUnsigned(offset_ptr, AddrSize);
break;
}
case DW_FORM_exprloc:
case DW_FORM_block:
Value.uval = data.getULEB128(offset_ptr);
is_block = true;
break;
case DW_FORM_block1:
Value.uval = data.getU8(offset_ptr);
is_block = true;
break;
case DW_FORM_block2:
Value.uval = data.getU16(offset_ptr);
is_block = true;
break;
case DW_FORM_block4:
Value.uval = data.getU32(offset_ptr);
is_block = true;
break;
case DW_FORM_data1:
case DW_FORM_ref1:
case DW_FORM_flag:
Value.uval = data.getU8(offset_ptr);
break;
case DW_FORM_data2:
case DW_FORM_ref2:
Value.uval = data.getU16(offset_ptr);
break;
case DW_FORM_data4:
case DW_FORM_ref4: {
RelocAddrMap::const_iterator AI = cu->getRelocMap()->find(*offset_ptr);
Value.uval = data.getU32(offset_ptr);
if (AI != cu->getRelocMap()->end())
Value.uval += AI->second.second;
break;
}
case DW_FORM_data8:
case DW_FORM_ref8:
Value.uval = data.getU64(offset_ptr);
break;
case DW_FORM_sdata:
Value.sval = data.getSLEB128(offset_ptr);
break;
case DW_FORM_strp: {
RelocAddrMap::const_iterator AI
= cu->getRelocMap()->find(*offset_ptr);
if (AI != cu->getRelocMap()->end()) {
const std::pair<uint8_t, int64_t> &R = AI->second;
Value.uval = data.getU32(offset_ptr) + R.second;
} else
Value.uval = data.getU32(offset_ptr);
break;
}
case DW_FORM_udata:
case DW_FORM_ref_udata:
Value.uval = data.getULEB128(offset_ptr);
break;
case DW_FORM_string:
Value.cstr = data.getCStr(offset_ptr);
break;
case DW_FORM_indirect:
Form = data.getULEB128(offset_ptr);
indirect = true;
break;
case DW_FORM_sec_offset: {
// FIXME: This is 64-bit for DWARF64.
RelocAddrMap::const_iterator AI
= cu->getRelocMap()->find(*offset_ptr);
if (AI != cu->getRelocMap()->end()) {
const std::pair<uint8_t, int64_t> &R = AI->second;
Value.uval = data.getU32(offset_ptr) + R.second;
} else
Value.uval = data.getU32(offset_ptr);
break;
}
case DW_FORM_flag_present:
Value.uval = 1;
break;
case DW_FORM_ref_sig8:
Value.uval = data.getU64(offset_ptr);
break;
case DW_FORM_GNU_addr_index:
case DW_FORM_GNU_str_index:
Value.uval = data.getULEB128(offset_ptr);
break;
default:
return false;
}
} while (indirect);
if (is_block) {
StringRef str = data.getData().substr(*offset_ptr, Value.uval);
Value.data = nullptr;
if (!str.empty()) {
Value.data = reinterpret_cast<const uint8_t *>(str.data());
*offset_ptr += Value.uval;
}
}
return true;
}
void DWARFDebugFrame::parse(DataExtractor Data) {
uint32_t Offset = 0;
while (Data.isValidOffset(Offset)) {
uint32_t StartOffset = Offset;
bool IsDWARF64 = false;
uint64_t Length = Data.getU32(&Offset);
uint64_t Id;
if (Length == UINT32_MAX) {
// DWARF-64 is distinguished by the first 32 bits of the initial length
// field being 0xffffffff. Then, the next 64 bits are the actual entry
// length.
IsDWARF64 = true;
Length = Data.getU64(&Offset);
}
// At this point, Offset points to the next field after Length.
// Length is the structure size excluding itself. Compute an offset one
// past the end of the structure (needed to know how many instructions to
// read).
// TODO: For honest DWARF64 support, DataExtractor will have to treat
// offset_ptr as uint64_t*
uint32_t EndStructureOffset = Offset + static_cast<uint32_t>(Length);
// The Id field's size depends on the DWARF format
Id = Data.getUnsigned(&Offset, IsDWARF64 ? 8 : 4);
bool IsCIE = ((IsDWARF64 && Id == DW64_CIE_ID) || Id == DW_CIE_ID);
if (IsCIE) {
// Note: this is specifically DWARFv3 CIE header structure. It was
// changed in DWARFv4. We currently don't support reading DWARFv4
// here because LLVM itself does not emit it (and LLDB doesn't
// support it either).
uint8_t Version = Data.getU8(&Offset);
const char *Augmentation = Data.getCStr(&Offset);
uint64_t CodeAlignmentFactor = Data.getULEB128(&Offset);
int64_t DataAlignmentFactor = Data.getSLEB128(&Offset);
uint64_t ReturnAddressRegister = Data.getULEB128(&Offset);
Entries.emplace_back(new CIE(StartOffset, Length, Version,
StringRef(Augmentation), CodeAlignmentFactor,
DataAlignmentFactor, ReturnAddressRegister));
} else {
// FDE
uint64_t CIEPointer = Id;
uint64_t InitialLocation = Data.getAddress(&Offset);
uint64_t AddressRange = Data.getAddress(&Offset);
Entries.emplace_back(new FDE(StartOffset, Length, CIEPointer,
InitialLocation, AddressRange));
}
Entries.back()->parseInstructions(Data, &Offset, EndStructureOffset);
if (Offset != EndStructureOffset) {
string_ostream Str;
Str << format("Parsing entry instructions at %lx failed", StartOffset);
report_fatal_error(Str.str());
}
}
}
bool
DWARFFormValue::extractValue(DataExtractor data, uint32_t *offset_ptr,
const DWARFCompileUnit *cu) {
bool indirect = false;
bool is_block = false;
Value.data = NULL;
// Read the value for the form into value and follow and DW_FORM_indirect
// instances we run into
do {
indirect = false;
switch (Form) {
case DW_FORM_addr:
case DW_FORM_ref_addr:
Value.uval = data.getUnsigned(offset_ptr, cu->getAddressByteSize());
break;
case DW_FORM_block:
Value.uval = data.getULEB128(offset_ptr);
is_block = true;
break;
case DW_FORM_block1:
Value.uval = data.getU8(offset_ptr);
is_block = true;
break;
case DW_FORM_block2:
Value.uval = data.getU16(offset_ptr);
is_block = true;
break;
case DW_FORM_block4:
Value.uval = data.getU32(offset_ptr);
is_block = true;
break;
case DW_FORM_data1:
case DW_FORM_ref1:
case DW_FORM_flag:
Value.uval = data.getU8(offset_ptr);
break;
case DW_FORM_data2:
case DW_FORM_ref2:
Value.uval = data.getU16(offset_ptr);
break;
case DW_FORM_data4:
case DW_FORM_ref4:
Value.uval = data.getU32(offset_ptr);
break;
case DW_FORM_data8:
case DW_FORM_ref8:
Value.uval = data.getU64(offset_ptr);
break;
case DW_FORM_sdata:
Value.sval = data.getSLEB128(offset_ptr);
break;
case DW_FORM_strp:
Value.uval = data.getU32(offset_ptr);
break;
case DW_FORM_udata:
case DW_FORM_ref_udata:
Value.uval = data.getULEB128(offset_ptr);
break;
case DW_FORM_string:
Value.cstr = data.getCStr(offset_ptr);
// Set the string value to also be the data for inlined cstr form
// values only so we can tell the differnence between DW_FORM_string
// and DW_FORM_strp form values
Value.data = (uint8_t*)Value.cstr;
break;
case DW_FORM_indirect:
Form = data.getULEB128(offset_ptr);
indirect = true;
break;
default:
return false;
}
} while (indirect);
if (is_block) {
StringRef str = data.getData().substr(*offset_ptr, Value.uval);
Value.data = NULL;
if (!str.empty()) {
Value.data = reinterpret_cast<const uint8_t *>(str.data());
*offset_ptr += Value.uval;
}
}
return true;
}
void DWARFDebugFrame::parse(DataExtractor Data) {
uint32_t Offset = 0;
DenseMap<uint32_t, CIE *> CIEs;
while (Data.isValidOffset(Offset)) {
uint32_t StartOffset = Offset;
auto ReportError = [StartOffset](const char *ErrorMsg) {
std::string Str;
raw_string_ostream OS(Str);
OS << format(ErrorMsg, StartOffset);
OS.flush();
report_fatal_error(Str);
};
bool IsDWARF64 = false;
uint64_t Length = Data.getU32(&Offset);
uint64_t Id;
if (Length == UINT32_MAX) {
// DWARF-64 is distinguished by the first 32 bits of the initial length
// field being 0xffffffff. Then, the next 64 bits are the actual entry
// length.
IsDWARF64 = true;
Length = Data.getU64(&Offset);
}
// At this point, Offset points to the next field after Length.
// Length is the structure size excluding itself. Compute an offset one
// past the end of the structure (needed to know how many instructions to
// read).
// TODO: For honest DWARF64 support, DataExtractor will have to treat
// offset_ptr as uint64_t*
uint32_t StartStructureOffset = Offset;
uint32_t EndStructureOffset = Offset + static_cast<uint32_t>(Length);
// The Id field's size depends on the DWARF format
Id = Data.getUnsigned(&Offset, (IsDWARF64 && !IsEH) ? 8 : 4);
bool IsCIE = ((IsDWARF64 && Id == DW64_CIE_ID) ||
Id == DW_CIE_ID ||
(IsEH && !Id));
if (IsCIE) {
uint8_t Version = Data.getU8(&Offset);
const char *Augmentation = Data.getCStr(&Offset);
StringRef AugmentationString(Augmentation ? Augmentation : "");
uint8_t AddressSize = Version < 4 ? Data.getAddressSize() :
Data.getU8(&Offset);
Data.setAddressSize(AddressSize);
uint8_t SegmentDescriptorSize = Version < 4 ? 0 : Data.getU8(&Offset);
uint64_t CodeAlignmentFactor = Data.getULEB128(&Offset);
int64_t DataAlignmentFactor = Data.getSLEB128(&Offset);
uint64_t ReturnAddressRegister = Data.getULEB128(&Offset);
// Parse the augmentation data for EH CIEs
StringRef AugmentationData("");
uint32_t FDEPointerEncoding = DW_EH_PE_omit;
uint32_t LSDAPointerEncoding = DW_EH_PE_omit;
if (IsEH) {
Optional<uint32_t> PersonalityEncoding;
Optional<uint64_t> Personality;
Optional<uint64_t> AugmentationLength;
uint32_t StartAugmentationOffset;
uint32_t EndAugmentationOffset;
// Walk the augmentation string to get all the augmentation data.
for (unsigned i = 0, e = AugmentationString.size(); i != e; ++i) {
switch (AugmentationString[i]) {
default:
ReportError("Unknown augmentation character in entry at %lx");
case 'L':
LSDAPointerEncoding = Data.getU8(&Offset);
break;
case 'P': {
if (Personality)
ReportError("Duplicate personality in entry at %lx");
PersonalityEncoding = Data.getU8(&Offset);
Personality = readPointer(Data, Offset, *PersonalityEncoding);
break;
}
case 'R':
FDEPointerEncoding = Data.getU8(&Offset);
break;
case 'z':
if (i)
ReportError("'z' must be the first character at %lx");
// Parse the augmentation length first. We only parse it if
// the string contains a 'z'.
AugmentationLength = Data.getULEB128(&Offset);
StartAugmentationOffset = Offset;
EndAugmentationOffset = Offset +
static_cast<uint32_t>(*AugmentationLength);
}
}
if (AugmentationLength.hasValue()) {
if (Offset != EndAugmentationOffset)
ReportError("Parsing augmentation data at %lx failed");
AugmentationData = Data.getData().slice(StartAugmentationOffset,
EndAugmentationOffset);
}
}
auto Cie = make_unique<CIE>(StartOffset, Length, Version,
AugmentationString, AddressSize,
SegmentDescriptorSize, CodeAlignmentFactor,
DataAlignmentFactor, ReturnAddressRegister,
AugmentationData, FDEPointerEncoding,
LSDAPointerEncoding);
CIEs[StartOffset] = Cie.get();
Entries.emplace_back(std::move(Cie));
} else {
// FDE
uint64_t CIEPointer = Id;
uint64_t InitialLocation = 0;
uint64_t AddressRange = 0;
CIE *Cie = CIEs[IsEH ? (StartStructureOffset - CIEPointer) : CIEPointer];
if (IsEH) {
// The address size is encoded in the CIE we reference.
if (!Cie)
ReportError("Parsing FDE data at %lx failed due to missing CIE");
InitialLocation = readPointer(Data, Offset,
Cie->getFDEPointerEncoding());
AddressRange = readPointer(Data, Offset,
Cie->getFDEPointerEncoding());
StringRef AugmentationString = Cie->getAugmentationString();
if (!AugmentationString.empty()) {
// Parse the augmentation length and data for this FDE.
uint64_t AugmentationLength = Data.getULEB128(&Offset);
uint32_t EndAugmentationOffset =
Offset + static_cast<uint32_t>(AugmentationLength);
// Decode the LSDA if the CIE augmentation string said we should.
if (Cie->getLSDAPointerEncoding() != DW_EH_PE_omit)
readPointer(Data, Offset, Cie->getLSDAPointerEncoding());
if (Offset != EndAugmentationOffset)
ReportError("Parsing augmentation data at %lx failed");
}
} else {
InitialLocation = Data.getAddress(&Offset);
AddressRange = Data.getAddress(&Offset);
}
Entries.emplace_back(new FDE(StartOffset, Length, CIEPointer,
InitialLocation, AddressRange,
Cie));
}
Entries.back()->parseInstructions(Data, &Offset, EndStructureOffset);
if (Offset != EndStructureOffset)
ReportError("Parsing entry instructions at %lx failed");
}
}
bool DWARFDebugLine::Prologue::parse(DataExtractor debug_line_data,
uint32_t *offset_ptr) {
const uint64_t prologue_offset = *offset_ptr;
clear();
TotalLength = debug_line_data.getU32(offset_ptr);
if (TotalLength == UINT32_MAX) {
IsDWARF64 = true;
TotalLength = debug_line_data.getU64(offset_ptr);
} else if (TotalLength > 0xffffff00) {
return false;
}
Version = debug_line_data.getU16(offset_ptr);
if (Version < 2)
return false;
PrologueLength =
debug_line_data.getUnsigned(offset_ptr, sizeofPrologueLength());
const uint64_t end_prologue_offset = PrologueLength + *offset_ptr;
MinInstLength = debug_line_data.getU8(offset_ptr);
if (Version >= 4)
MaxOpsPerInst = debug_line_data.getU8(offset_ptr);
DefaultIsStmt = debug_line_data.getU8(offset_ptr);
LineBase = debug_line_data.getU8(offset_ptr);
LineRange = debug_line_data.getU8(offset_ptr);
OpcodeBase = debug_line_data.getU8(offset_ptr);
StandardOpcodeLengths.reserve(OpcodeBase - 1);
for (uint32_t i = 1; i < OpcodeBase; ++i) {
uint8_t op_len = debug_line_data.getU8(offset_ptr);
StandardOpcodeLengths.push_back(op_len);
}
while (*offset_ptr < end_prologue_offset) {
const char *s = debug_line_data.getCStr(offset_ptr);
if (s && s[0])
IncludeDirectories.push_back(s);
else
break;
}
while (*offset_ptr < end_prologue_offset) {
const char *name = debug_line_data.getCStr(offset_ptr);
if (name && name[0]) {
FileNameEntry fileEntry;
fileEntry.Name = name;
fileEntry.DirIdx = debug_line_data.getULEB128(offset_ptr);
fileEntry.ModTime = debug_line_data.getULEB128(offset_ptr);
fileEntry.Length = debug_line_data.getULEB128(offset_ptr);
FileNames.push_back(fileEntry);
} else {
break;
}
}
if (*offset_ptr != end_prologue_offset) {
fprintf(stderr, "warning: parsing line table prologue at 0x%8.8" PRIx64
" should have ended at 0x%8.8" PRIx64
" but it ended at 0x%8.8" PRIx64 "\n",
prologue_offset, end_prologue_offset, (uint64_t)*offset_ptr);
return false;
}
return true;
}
bool
DWARFFormValue::extractValue(DataExtractor data, uint32_t *offset_ptr,
const DWARFCompileUnit *cu) {
bool indirect = false;
bool is_block = false;
Value.data = NULL;
// Read the value for the form into value and follow and DW_FORM_indirect
// instances we run into
do {
indirect = false;
switch (Form) {
case DW_FORM_addr:
case DW_FORM_ref_addr: {
RelocAddrMap::const_iterator AI
= cu->getRelocMap()->find(*offset_ptr);
if (AI != cu->getRelocMap()->end()) {
const std::pair<uint8_t, int64_t> &R = AI->second;
Value.uval = data.getUnsigned(offset_ptr, cu->getAddressByteSize()) +
R.second;
} else
Value.uval = data.getUnsigned(offset_ptr, cu->getAddressByteSize());
break;
}
case DW_FORM_exprloc:
case DW_FORM_block:
Value.uval = data.getULEB128(offset_ptr);
is_block = true;
break;
case DW_FORM_block1:
Value.uval = data.getU8(offset_ptr);
is_block = true;
break;
case DW_FORM_block2:
Value.uval = data.getU16(offset_ptr);
is_block = true;
break;
case DW_FORM_block4:
Value.uval = data.getU32(offset_ptr);
is_block = true;
break;
case DW_FORM_data1:
case DW_FORM_ref1:
case DW_FORM_flag:
Value.uval = data.getU8(offset_ptr);
break;
case DW_FORM_data2:
case DW_FORM_ref2:
Value.uval = data.getU16(offset_ptr);
break;
case DW_FORM_data4:
case DW_FORM_ref4:
Value.uval = data.getU32(offset_ptr);
break;
case DW_FORM_data8:
case DW_FORM_ref8:
Value.uval = data.getU64(offset_ptr);
break;
case DW_FORM_sdata:
Value.sval = data.getSLEB128(offset_ptr);
break;
case DW_FORM_strp: {
RelocAddrMap::const_iterator AI
= cu->getRelocMap()->find(*offset_ptr);
if (AI != cu->getRelocMap()->end()) {
const std::pair<uint8_t, int64_t> &R = AI->second;
Value.uval = data.getU32(offset_ptr) + R.second;
} else
Value.uval = data.getU32(offset_ptr);
break;
}
case DW_FORM_udata:
case DW_FORM_ref_udata:
Value.uval = data.getULEB128(offset_ptr);
break;
case DW_FORM_string:
Value.cstr = data.getCStr(offset_ptr);
// Set the string value to also be the data for inlined cstr form
// values only so we can tell the differnence between DW_FORM_string
// and DW_FORM_strp form values
Value.data = (const uint8_t*)Value.cstr;
break;
case DW_FORM_indirect:
Form = data.getULEB128(offset_ptr);
indirect = true;
break;
case DW_FORM_sec_offset:
// FIXME: This is 64-bit for DWARF64.
Value.uval = data.getU32(offset_ptr);
break;
case DW_FORM_flag_present:
Value.uval = 1;
break;
case DW_FORM_ref_sig8:
Value.uval = data.getU64(offset_ptr);
break;
case DW_FORM_GNU_addr_index:
Value.uval = data.getULEB128(offset_ptr);
break;
case DW_FORM_GNU_str_index:
Value.uval = data.getULEB128(offset_ptr);
break;
default:
return false;
}
} while (indirect);
if (is_block) {
StringRef str = data.getData().substr(*offset_ptr, Value.uval);
Value.data = NULL;
if (!str.empty()) {
Value.data = reinterpret_cast<const uint8_t *>(str.data());
*offset_ptr += Value.uval;
}
}
return true;
}