bool DWARFUnit::extractImpl(DataExtractor debug_info, uint32_t *offset_ptr) {
Length = debug_info.getU32(offset_ptr);
Version = debug_info.getU16(offset_ptr);
uint64_t AbbrOffset = debug_info.getU32(offset_ptr);
if (IndexEntry) {
if (AbbrOffset)
return false;
auto *UnitContrib = IndexEntry->getOffset();
if (!UnitContrib || UnitContrib->Length != (Length + 4))
return false;
auto *AbbrEntry = IndexEntry->getOffset(DW_SECT_ABBREV);
if (!AbbrEntry)
return false;
AbbrOffset = AbbrEntry->Offset;
}
AddrSize = debug_info.getU8(offset_ptr);
bool LengthOK = debug_info.isValidOffset(getNextUnitOffset() - 1);
bool VersionOK = DWARFContext::isSupportedVersion(Version);
bool AddrSizeOK = AddrSize == 4 || AddrSize == 8;
if (!LengthOK || !VersionOK || !AddrSizeOK)
return false;
Abbrevs = Abbrev->getAbbreviationDeclarationSet(AbbrOffset);
return Abbrevs != nullptr;
}
bool DWARFCompileUnit::extract(DataExtractor debug_info, uint32_t *offset_ptr) {
clear();
Offset = *offset_ptr;
if (debug_info.isValidOffset(*offset_ptr)) {
uint64_t abbrOffset;
Length = debug_info.getU32(offset_ptr);
Version = debug_info.getU16(offset_ptr);
abbrOffset = debug_info.getU32(offset_ptr);
AddrSize = debug_info.getU8(offset_ptr);
bool lengthOK = debug_info.isValidOffset(getNextCompileUnitOffset()-1);
bool versionOK = DWARFContext::isSupportedVersion(Version);
bool abbrOffsetOK = AbbrevSection.size() > abbrOffset;
bool addrSizeOK = AddrSize == 4 || AddrSize == 8;
if (lengthOK && versionOK && addrSizeOK && abbrOffsetOK && Abbrev != NULL) {
Abbrevs = Abbrev->getAbbreviationDeclarationSet(abbrOffset);
return true;
}
// reset the offset to where we tried to parse from if anything went wrong
*offset_ptr = Offset;
}
return false;
}
bool DWARFUnitIndex::parseImpl(DataExtractor IndexData) {
uint32_t Offset = 0;
if (!Header.parse(IndexData, &Offset))
return false;
if (!IndexData.isValidOffsetForDataOfSize(
Offset, Header.NumBuckets * (8 + 4) +
(2 * Header.NumUnits + 1) * 4 * Header.NumColumns))
return false;
Rows = llvm::make_unique<Entry[]>(Header.NumBuckets);
auto Contribs =
llvm::make_unique<Entry::SectionContribution *[]>(Header.NumUnits);
ColumnKinds = llvm::make_unique<DWARFSectionKind[]>(Header.NumColumns);
// Read Hash Table of Signatures
for (unsigned i = 0; i != Header.NumBuckets; ++i)
Rows[i].Signature = IndexData.getU64(&Offset);
// Read Parallel Table of Indexes
for (unsigned i = 0; i != Header.NumBuckets; ++i) {
auto Index = IndexData.getU32(&Offset);
if (!Index)
continue;
Rows[i].Index = this;
Rows[i].Contributions =
llvm::make_unique<Entry::SectionContribution[]>(Header.NumColumns);
Contribs[Index - 1] = Rows[i].Contributions.get();
}
// Read the Column Headers
for (unsigned i = 0; i != Header.NumColumns; ++i) {
ColumnKinds[i] = static_cast<DWARFSectionKind>(IndexData.getU32(&Offset));
if (ColumnKinds[i] == InfoColumnKind) {
if (InfoColumn != -1)
return false;
InfoColumn = i;
}
}
if (InfoColumn == -1)
return false;
// Read Table of Section Offsets
for (unsigned i = 0; i != Header.NumUnits; ++i) {
auto *Contrib = Contribs[i];
for (unsigned i = 0; i != Header.NumColumns; ++i)
Contrib[i].Offset = IndexData.getU32(&Offset);
}
// Read Table of Section Sizes
for (unsigned i = 0; i != Header.NumUnits; ++i) {
auto *Contrib = Contribs[i];
for (unsigned i = 0; i != Header.NumColumns; ++i)
Contrib[i].Length = IndexData.getU32(&Offset);
}
return true;
}
bool DWARFDebugLine::Prologue::parse(DataExtractor debug_line_data,
uint32_t *offset_ptr) {
const uint32_t prologue_offset = *offset_ptr;
clear();
TotalLength = debug_line_data.getU32(offset_ptr);
Version = debug_line_data.getU16(offset_ptr);
if (Version < 2)
return false;
PrologueLength = debug_line_data.getU32(offset_ptr);
const uint32_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%8.8x should"
" have ended at 0x%8.8x but it ended at 0x%8.8x\n",
prologue_offset, end_prologue_offset, *offset_ptr);
return false;
}
return true;
}
bool DWARFUnitIndex::Header::parse(DataExtractor IndexData,
uint32_t *OffsetPtr) {
if (!IndexData.isValidOffsetForDataOfSize(*OffsetPtr, 16))
return false;
Version = IndexData.getU32(OffsetPtr);
NumColumns = IndexData.getU32(OffsetPtr);
NumUnits = IndexData.getU32(OffsetPtr);
NumBuckets = IndexData.getU32(OffsetPtr);
return Version <= 2;
}
Optional<DWARFDebugLocDWO::LocationList>
DWARFDebugLocDWO::parseOneLocationList(DataExtractor Data, unsigned *Offset) {
LocationList LL;
LL.Offset = *Offset;
// dwarf::DW_LLE_end_of_list_entry is 0 and indicates the end of the list.
while (auto Kind =
static_cast<dwarf::LocationListEntry>(Data.getU8(Offset))) {
if (Kind != dwarf::DW_LLE_startx_length) {
llvm::errs() << "error: dumping support for LLE of kind " << (int)Kind
<< " not implemented\n";
return None;
}
Entry E;
E.Start = Data.getULEB128(Offset);
E.Length = Data.getU32(Offset);
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.resize(str.size());
std::copy(str.begin(), str.end(), E.Loc.begin());
LL.Entries.push_back(std::move(E));
}
return LL;
}
void DWARFDebugLocDWO::parse(DataExtractor data) {
uint32_t Offset = 0;
while (data.isValidOffset(Offset)) {
Locations.resize(Locations.size() + 1);
LocationList &Loc = Locations.back();
Loc.Offset = Offset;
dwarf::LocationListEntry Kind;
while ((Kind = static_cast<dwarf::LocationListEntry>(
data.getU8(&Offset))) != dwarf::DW_LLE_end_of_list) {
if (Kind != dwarf::DW_LLE_startx_length) {
llvm::errs() << "error: dumping support for LLE of kind " << (int)Kind
<< " not implemented\n";
return;
}
Entry E;
E.Start = data.getULEB128(&Offset);
E.Length = data.getU32(&Offset);
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.resize(str.size());
std::copy(str.begin(), str.end(), E.Loc.begin());
Loc.Entries.push_back(std::move(E));
}
}
}
bool DWARFUnit::extractImpl(DataExtractor debug_info, uint32_t *offset_ptr) {
Length = debug_info.getU32(offset_ptr);
Version = debug_info.getU16(offset_ptr);
uint64_t abbrOffset = debug_info.getU32(offset_ptr);
AddrSize = debug_info.getU8(offset_ptr);
bool lengthOK = debug_info.isValidOffset(getNextUnitOffset() - 1);
bool versionOK = DWARFContext::isSupportedVersion(Version);
bool abbrOffsetOK = AbbrevSection.size() > abbrOffset;
bool addrSizeOK = AddrSize == 4 || AddrSize == 8;
if (!lengthOK || !versionOK || !addrSizeOK || !abbrOffsetOK)
return false;
Abbrevs = Abbrev->getAbbreviationDeclarationSet(abbrOffset);
return true;
}
bool DWARFTypeUnit::extractImpl(DataExtractor debug_info,
uint32_t *offset_ptr) {
if (!DWARFUnit::extractImpl(debug_info, offset_ptr))
return false;
TypeHash = debug_info.getU64(offset_ptr);
TypeOffset = debug_info.getU32(offset_ptr);
return TypeOffset < getLength();
}
bool DWARFUnit::extractImpl(DataExtractor debug_info, uint32_t *offset_ptr) {
Length = debug_info.getU32(offset_ptr);
Version = debug_info.getU16(offset_ptr);
uint64_t AbbrOffset = debug_info.getU32(offset_ptr);
AddrSize = debug_info.getU8(offset_ptr);
bool LengthOK = debug_info.isValidOffset(getNextUnitOffset() - 1);
bool VersionOK = DWARFContext::isSupportedVersion(Version);
bool AddrSizeOK = AddrSize == 4 || AddrSize == 8;
if (!LengthOK || !VersionOK || !AddrSizeOK)
return false;
Abbrevs = Abbrev->getAbbreviationDeclarationSet(AbbrOffset);
if (Abbrevs == nullptr)
return false;
return true;
}
bool DWARFTypeUnit::extractImpl(DataExtractor debug_info,
uint32_t *offset_ptr) {
if (!DWARFUnit::extractImpl(debug_info, offset_ptr))
return false;
TypeHash = debug_info.getU64(offset_ptr);
TypeOffset = debug_info.getU32(offset_ptr);
// TypeOffset is relative to the beginning of the header,
// so we have to account for the leading length field.
// FIXME: The size of the length field is 12 in DWARF64.
unsigned SizeOfLength = 4;
return TypeOffset < getLength() + SizeOfLength;
}
static uint64_t readPointer(const DataExtractor &Data, uint32_t &Offset,
unsigned Encoding) {
switch (getSizeForEncoding(Data, Encoding)) {
case 2:
return Data.getU16(&Offset);
case 4:
return Data.getU32(&Offset);
case 8:
return Data.getU64(&Offset);
default:
llvm_unreachable("Illegal data size");
}
}
uint32_t
DWARFCompileUnit::extract(uint32_t offset, DataExtractor debug_info_data,
const DWARFAbbreviationDeclarationSet *abbrevs) {
clear();
Offset = offset;
if (debug_info_data.isValidOffset(offset)) {
Length = debug_info_data.getU32(&offset);
Version = debug_info_data.getU16(&offset);
bool abbrevsOK = debug_info_data.getU32(&offset) == abbrevs->getOffset();
Abbrevs = abbrevs;
AddrSize = debug_info_data.getU8(&offset);
bool versionOK = DWARFContext::isSupportedVersion(Version);
bool addrSizeOK = AddrSize == 4 || AddrSize == 8;
if (versionOK && addrSizeOK && abbrevsOK &&
debug_info_data.isValidOffset(offset))
return offset;
}
return 0;
}
bool DWARFUnit::extractImpl(DataExtractor debug_info, uint32_t *offset_ptr) {
Length = debug_info.getU32(offset_ptr);
// FIXME: Support DWARF64.
FormParams.Format = DWARF32;
FormParams.Version = debug_info.getU16(offset_ptr);
if (FormParams.Version >= 5) {
UnitType = debug_info.getU8(offset_ptr);
FormParams.AddrSize = debug_info.getU8(offset_ptr);
AbbrOffset = debug_info.getU32(offset_ptr);
} else {
AbbrOffset = debug_info.getU32(offset_ptr);
FormParams.AddrSize = debug_info.getU8(offset_ptr);
}
if (IndexEntry) {
if (AbbrOffset)
return false;
auto *UnitContrib = IndexEntry->getOffset();
if (!UnitContrib || UnitContrib->Length != (Length + 4))
return false;
auto *AbbrEntry = IndexEntry->getOffset(DW_SECT_ABBREV);
if (!AbbrEntry)
return false;
AbbrOffset = AbbrEntry->Offset;
}
bool LengthOK = debug_info.isValidOffset(getNextUnitOffset() - 1);
bool VersionOK = DWARFContext::isSupportedVersion(getVersion());
bool AddrSizeOK = getAddressByteSize() == 4 || getAddressByteSize() == 8;
if (!LengthOK || !VersionOK || !AddrSizeOK)
return false;
// Keep track of the highest DWARF version we encounter across all units.
Context.setMaxVersionIfGreater(getVersion());
return true;
}
/// State transition when a CustomEventMarker is encountered.
Error processCustomEventMarker(FDRState &State, uint8_t RecordFirstByte,
DataExtractor &RecordExtractor,
size_t &RecordSize) {
// We can encounter a CustomEventMarker anywhere in the log, so we can handle
// it regardless of the expectation. However, we do se the expectation to read
// a set number of fixed bytes, as described in the metadata.
uint32_t OffsetPtr = 1; // Read after the first byte.
uint32_t DataSize = RecordExtractor.getU32(&OffsetPtr);
uint64_t TSC = RecordExtractor.getU64(&OffsetPtr);
// FIXME: Actually represent the record through the API. For now we only skip
// through the data.
(void)TSC;
RecordSize = 16 + DataSize;
return Error::success();
}
void FrameEntry::parseInstructions(DataExtractor Data, uint32_t *Offset,
uint32_t EndOffset) {
while (*Offset < EndOffset) {
uint8_t Opcode = Data.getU8(Offset);
// Some instructions have a primary opcode encoded in the top bits.
uint8_t Primary = Opcode & DWARF_CFI_PRIMARY_OPCODE_MASK;
if (Primary) {
// If it's a primary opcode, the first operand is encoded in the bottom
// bits of the opcode itself.
uint64_t Op1 = Opcode & DWARF_CFI_PRIMARY_OPERAND_MASK;
switch (Primary) {
default: llvm_unreachable("Impossible primary CFI opcode");
case DW_CFA_advance_loc:
case DW_CFA_restore:
addInstruction(Primary, Op1);
break;
case DW_CFA_offset:
addInstruction(Primary, Op1, Data.getULEB128(Offset));
break;
}
} else {
// Extended opcode - its value is Opcode itself.
switch (Opcode) {
default: llvm_unreachable("Invalid extended CFI opcode");
case DW_CFA_nop:
case DW_CFA_remember_state:
case DW_CFA_restore_state:
case DW_CFA_GNU_window_save:
// No operands
addInstruction(Opcode);
break;
case DW_CFA_set_loc:
// Operands: Address
addInstruction(Opcode, Data.getAddress(Offset));
break;
case DW_CFA_advance_loc1:
// Operands: 1-byte delta
addInstruction(Opcode, Data.getU8(Offset));
break;
case DW_CFA_advance_loc2:
// Operands: 2-byte delta
addInstruction(Opcode, Data.getU16(Offset));
break;
case DW_CFA_advance_loc4:
// Operands: 4-byte delta
addInstruction(Opcode, Data.getU32(Offset));
break;
case DW_CFA_restore_extended:
case DW_CFA_undefined:
case DW_CFA_same_value:
case DW_CFA_def_cfa_register:
case DW_CFA_def_cfa_offset:
// Operands: ULEB128
addInstruction(Opcode, Data.getULEB128(Offset));
break;
case DW_CFA_def_cfa_offset_sf:
// Operands: SLEB128
addInstruction(Opcode, Data.getSLEB128(Offset));
break;
case DW_CFA_offset_extended:
case DW_CFA_register:
case DW_CFA_def_cfa:
case DW_CFA_val_offset:
// Operands: ULEB128, ULEB128
addInstruction(Opcode, Data.getULEB128(Offset),
Data.getULEB128(Offset));
break;
case DW_CFA_offset_extended_sf:
case DW_CFA_def_cfa_sf:
case DW_CFA_val_offset_sf:
// Operands: ULEB128, SLEB128
addInstruction(Opcode, Data.getULEB128(Offset),
Data.getSLEB128(Offset));
break;
case DW_CFA_def_cfa_expression:
case DW_CFA_expression:
case DW_CFA_val_expression:
// TODO: implement this
report_fatal_error("Values with expressions not implemented yet!");
}
}
}
}
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::skipValue(uint16_t form, DataExtractor debug_info_data,
uint32_t *offset_ptr, const DWARFUnit *cu) {
bool indirect = false;
do {
switch (form) {
// Blocks if inlined data that have a length field and the data bytes
// inlined in the .debug_info
case DW_FORM_exprloc:
case DW_FORM_block: {
uint64_t size = debug_info_data.getULEB128(offset_ptr);
*offset_ptr += size;
return true;
}
case DW_FORM_block1: {
uint8_t size = debug_info_data.getU8(offset_ptr);
*offset_ptr += size;
return true;
}
case DW_FORM_block2: {
uint16_t size = debug_info_data.getU16(offset_ptr);
*offset_ptr += size;
return true;
}
case DW_FORM_block4: {
uint32_t size = debug_info_data.getU32(offset_ptr);
*offset_ptr += size;
return true;
}
// Inlined NULL terminated C-strings
case DW_FORM_string:
debug_info_data.getCStr(offset_ptr);
return true;
// Compile unit address sized values
case DW_FORM_addr:
*offset_ptr += cu->getAddressByteSize();
return true;
case DW_FORM_ref_addr:
*offset_ptr += getRefAddrSize(cu->getAddressByteSize(), cu->getVersion());
return true;
// 0 byte values - implied from the form.
case DW_FORM_flag_present:
return true;
// 1 byte values
case DW_FORM_data1:
case DW_FORM_flag:
case DW_FORM_ref1:
*offset_ptr += 1;
return true;
// 2 byte values
case DW_FORM_data2:
case DW_FORM_ref2:
*offset_ptr += 2;
return true;
// 4 byte values
case DW_FORM_strp:
case DW_FORM_data4:
case DW_FORM_ref4:
*offset_ptr += 4;
return true;
// 8 byte values
case DW_FORM_data8:
case DW_FORM_ref8:
case DW_FORM_ref_sig8:
*offset_ptr += 8;
return true;
// signed or unsigned LEB 128 values
// case DW_FORM_APPLE_db_str:
case DW_FORM_sdata:
case DW_FORM_udata:
case DW_FORM_ref_udata:
case DW_FORM_GNU_str_index:
case DW_FORM_GNU_addr_index:
debug_info_data.getULEB128(offset_ptr);
return true;
case DW_FORM_indirect:
indirect = true;
form = debug_info_data.getULEB128(offset_ptr);
break;
// FIXME: 4 for DWARF32, 8 for DWARF64.
case DW_FORM_sec_offset:
*offset_ptr += 4;
return true;
default:
return false;
}
} while (indirect);
return true;
}
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;
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");
}
}
void FrameEntry::parseInstructions(DataExtractor Data, uint32_t *Offset,
uint32_t EndOffset) {
while (*Offset < EndOffset) {
uint8_t Opcode = Data.getU8(Offset);
// Some instructions have a primary opcode encoded in the top bits.
uint8_t Primary = Opcode & DWARF_CFI_PRIMARY_OPCODE_MASK;
if (Primary) {
// If it's a primary opcode, the first operand is encoded in the bottom
// bits of the opcode itself.
uint64_t Op1 = Opcode & DWARF_CFI_PRIMARY_OPERAND_MASK;
switch (Primary) {
default: llvm_unreachable("Impossible primary CFI opcode");
case DW_CFA_advance_loc:
case DW_CFA_restore:
addInstruction(Primary, Op1);
break;
case DW_CFA_offset:
addInstruction(Primary, Op1, Data.getULEB128(Offset));
break;
}
} else {
// Extended opcode - its value is Opcode itself.
switch (Opcode) {
default: llvm_unreachable("Invalid extended CFI opcode");
case DW_CFA_nop:
case DW_CFA_remember_state:
case DW_CFA_restore_state:
case DW_CFA_GNU_window_save:
// No operands
addInstruction(Opcode);
break;
case DW_CFA_set_loc:
// Operands: Address
addInstruction(Opcode, Data.getAddress(Offset));
break;
case DW_CFA_advance_loc1:
// Operands: 1-byte delta
addInstruction(Opcode, Data.getU8(Offset));
break;
case DW_CFA_advance_loc2:
// Operands: 2-byte delta
addInstruction(Opcode, Data.getU16(Offset));
break;
case DW_CFA_advance_loc4:
// Operands: 4-byte delta
addInstruction(Opcode, Data.getU32(Offset));
break;
case DW_CFA_restore_extended:
case DW_CFA_undefined:
case DW_CFA_same_value:
case DW_CFA_def_cfa_register:
case DW_CFA_def_cfa_offset:
case DW_CFA_GNU_args_size:
// Operands: ULEB128
addInstruction(Opcode, Data.getULEB128(Offset));
break;
case DW_CFA_def_cfa_offset_sf:
// Operands: SLEB128
addInstruction(Opcode, Data.getSLEB128(Offset));
break;
case DW_CFA_offset_extended:
case DW_CFA_register:
case DW_CFA_def_cfa:
case DW_CFA_val_offset: {
// Operands: ULEB128, ULEB128
// Note: We can not embed getULEB128 directly into function
// argument list. getULEB128 changes Offset and order of evaluation
// for arguments is unspecified.
auto op1 = Data.getULEB128(Offset);
auto op2 = Data.getULEB128(Offset);
addInstruction(Opcode, op1, op2);
break;
}
case DW_CFA_offset_extended_sf:
case DW_CFA_def_cfa_sf:
case DW_CFA_val_offset_sf: {
// Operands: ULEB128, SLEB128
// Note: see comment for the previous case
auto op1 = Data.getULEB128(Offset);
auto op2 = (uint64_t)Data.getSLEB128(Offset);
addInstruction(Opcode, op1, op2);
break;
}
case DW_CFA_def_cfa_expression:
// FIXME: Parse the actual instruction.
*Offset += Data.getULEB128(Offset);
break;
case DW_CFA_expression:
case DW_CFA_val_expression: {
// FIXME: Parse the actual instruction.
Data.getULEB128(Offset);
*Offset += Data.getULEB128(Offset);
break;
}
}
}
}
}
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;
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: {
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;
}
bool DWARFFormValue::skipValue(dwarf::Form Form, DataExtractor DebugInfoData,
uint32_t *OffsetPtr,
const DWARFFormParams Params) {
bool Indirect = false;
do {
switch (Form) {
// Blocks of inlined data that have a length field and the data bytes
// inlined in the .debug_info.
case DW_FORM_exprloc:
case DW_FORM_block: {
uint64_t size = DebugInfoData.getULEB128(OffsetPtr);
*OffsetPtr += size;
return true;
}
case DW_FORM_block1: {
uint8_t size = DebugInfoData.getU8(OffsetPtr);
*OffsetPtr += size;
return true;
}
case DW_FORM_block2: {
uint16_t size = DebugInfoData.getU16(OffsetPtr);
*OffsetPtr += size;
return true;
}
case DW_FORM_block4: {
uint32_t size = DebugInfoData.getU32(OffsetPtr);
*OffsetPtr += size;
return true;
}
// Inlined NULL terminated C-strings.
case DW_FORM_string:
DebugInfoData.getCStr(OffsetPtr);
return true;
case DW_FORM_addr:
case DW_FORM_ref_addr:
case DW_FORM_flag_present:
case DW_FORM_data1:
case DW_FORM_data2:
case DW_FORM_data4:
case DW_FORM_data8:
case DW_FORM_data16:
case DW_FORM_flag:
case DW_FORM_ref1:
case DW_FORM_ref2:
case DW_FORM_ref4:
case DW_FORM_ref8:
case DW_FORM_ref_sig8:
case DW_FORM_ref_sup4:
case DW_FORM_ref_sup8:
case DW_FORM_strx1:
case DW_FORM_strx2:
case DW_FORM_strx4:
case DW_FORM_addrx1:
case DW_FORM_addrx2:
case DW_FORM_addrx4:
case DW_FORM_sec_offset:
case DW_FORM_strp:
case DW_FORM_strp_sup:
case DW_FORM_line_strp:
case DW_FORM_GNU_ref_alt:
case DW_FORM_GNU_strp_alt:
if (Optional<uint8_t> FixedSize =
DWARFFormValue::getFixedByteSize(Form, Params)) {
*OffsetPtr += *FixedSize;
return true;
}
return false;
// signed or unsigned LEB 128 values.
case DW_FORM_sdata:
DebugInfoData.getSLEB128(OffsetPtr);
return true;
case DW_FORM_udata:
case DW_FORM_ref_udata:
case DW_FORM_strx:
case DW_FORM_addrx:
case DW_FORM_loclistx:
case DW_FORM_rnglistx:
case DW_FORM_GNU_addr_index:
case DW_FORM_GNU_str_index:
DebugInfoData.getULEB128(OffsetPtr);
return true;
case DW_FORM_indirect:
Indirect = true;
Form = static_cast<dwarf::Form>(DebugInfoData.getULEB128(OffsetPtr));
break;
default:
return false;
}
} while (Indirect);
return true;
}