In a PR last month I changed the ObjectFile CreateInstance etc methods to accept an optional DataExtractorSP instead of a DataBufferSP, and retain the extractor in a shared pointer internally in all of the ObjectFile subclasses. This is laying the groundwork for using a VirtualDataExtractor for some Mach-O binaries on macOS, where the segments of the binary are out-of-order in actual memory, and we add a lookup table to make it appear that the TEXT segment is at offset 0 in the Extractor, etc. Working on the actual implementation, I realized we were still using DataBufferSP's in ModuleSpec and Module, as well as in ObjectFile::GetModuleSpecifications. I originally was making a much larger NFC change where I had all ObjectFile subclasses operating on DataExtractors throughout their implementation, as well as in the DWARF parser. It was a very large patchset. Many subclasses start with their DataExtractor, then create smaller DataExtractors for parts of the binary image - the string table, the symbol table, etc., for processing. After consideration and discussion with Jonas, we agreed that a segment/section of a binary will never require a lookup table to access the bytes within it, so I changed VirtualDataExtractor::GetSubsetExtractorSP to (1) require that the Subset be contained within a single lookup table entry, and (2) return a simple DataExtractor bounded on that byte range. By doing this, I was able to remove all of my very-invasive changes to the ObjectFile subclass internals; it's only when they are operating on the entire binary image that care is needed. One pattern that subclasses like ObjectFileBreakpad use is to take an ArrayRef of the DataBuffer for a binary, then create a StringRef of that, then look for strings in it. With a VirtualDataExtractor and out-of-order binary segments, with gaps between them, this allows us to search the entire buffer looking for a string, and segfault when it gets to an unmapped region of the buffer. I added a VirtualDataExtractor::GetSubsetExtractorSP(0) which gets the largest contiguous memory region starting at offset 0 for this use case, and I added a comment about what was being done there because I know it is not obvious, and people not working on macOS wouldn't be familiar with the requirement. (when we have a ModuleSpec with a DataExtractor, any of the ObjectFile subclasses get a shot at Creating, so they all have to be able to iterate on these) rdar://148939795
216 lines
7.9 KiB
C++
216 lines
7.9 KiB
C++
//===-- ObjectContainerUniversalMachO.cpp ---------------------------------===//
|
|
//
|
|
// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
|
|
// See https://llvm.org/LICENSE.txt for license information.
|
|
// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
|
|
//
|
|
//===----------------------------------------------------------------------===//
|
|
|
|
#include "ObjectContainerUniversalMachO.h"
|
|
#include "lldb/Core/Module.h"
|
|
#include "lldb/Core/ModuleSpec.h"
|
|
#include "lldb/Core/PluginManager.h"
|
|
#include "lldb/Symbol/ObjectFile.h"
|
|
#include "lldb/Target/Target.h"
|
|
#include "lldb/Utility/ArchSpec.h"
|
|
#include "lldb/Utility/DataBuffer.h"
|
|
#include "lldb/Utility/Stream.h"
|
|
|
|
using namespace lldb;
|
|
using namespace lldb_private;
|
|
using namespace llvm::MachO;
|
|
|
|
LLDB_PLUGIN_DEFINE_ADV(ObjectContainerUniversalMachO,
|
|
ObjectContainerMachOArchive)
|
|
|
|
void ObjectContainerUniversalMachO::Initialize() {
|
|
PluginManager::RegisterPlugin(GetPluginNameStatic(),
|
|
GetPluginDescriptionStatic(), CreateInstance,
|
|
GetModuleSpecifications);
|
|
}
|
|
|
|
void ObjectContainerUniversalMachO::Terminate() {
|
|
PluginManager::UnregisterPlugin(CreateInstance);
|
|
}
|
|
|
|
ObjectContainer *ObjectContainerUniversalMachO::CreateInstance(
|
|
const lldb::ModuleSP &module_sp, DataBufferSP &data_sp,
|
|
lldb::offset_t data_offset, const FileSpec *file,
|
|
lldb::offset_t file_offset, lldb::offset_t length) {
|
|
// We get data when we aren't trying to look for cached container
|
|
// information, so only try and look for an architecture slice if we get data
|
|
if (data_sp) {
|
|
DataExtractor data;
|
|
data.SetData(data_sp, data_offset, length);
|
|
if (ObjectContainerUniversalMachO::MagicBytesMatch(data)) {
|
|
std::unique_ptr<ObjectContainerUniversalMachO> container_up(
|
|
new ObjectContainerUniversalMachO(module_sp, data_sp, data_offset,
|
|
file, file_offset, length));
|
|
if (container_up->ParseHeader()) {
|
|
return container_up.release();
|
|
}
|
|
}
|
|
}
|
|
return nullptr;
|
|
}
|
|
|
|
bool ObjectContainerUniversalMachO::MagicBytesMatch(const DataExtractor &data) {
|
|
lldb::offset_t offset = 0;
|
|
uint32_t magic = data.GetU32(&offset);
|
|
return magic == FAT_MAGIC || magic == FAT_CIGAM || magic == FAT_MAGIC_64 ||
|
|
magic == FAT_CIGAM_64;
|
|
}
|
|
|
|
ObjectContainerUniversalMachO::ObjectContainerUniversalMachO(
|
|
const lldb::ModuleSP &module_sp, DataBufferSP &data_sp,
|
|
lldb::offset_t data_offset, const FileSpec *file,
|
|
lldb::offset_t file_offset, lldb::offset_t length)
|
|
: ObjectContainer(module_sp, file, file_offset, length, data_sp,
|
|
data_offset),
|
|
m_header(), m_fat_archs() {
|
|
memset(&m_header, 0, sizeof(m_header));
|
|
}
|
|
|
|
ObjectContainerUniversalMachO::~ObjectContainerUniversalMachO() = default;
|
|
|
|
bool ObjectContainerUniversalMachO::ParseHeader() {
|
|
bool success = ParseHeader(*m_extractor_sp.get(), m_header, m_fat_archs);
|
|
// We no longer need any data, we parsed all we needed to parse and cached it
|
|
// in m_header and m_fat_archs
|
|
m_extractor_sp = std::make_shared<DataExtractor>();
|
|
return success;
|
|
}
|
|
|
|
bool ObjectContainerUniversalMachO::ParseHeader(
|
|
lldb_private::DataExtractor &extractor, llvm::MachO::fat_header &header,
|
|
std::vector<FatArch> &fat_archs) {
|
|
// Store the file offset for this universal file as we could have a universal
|
|
// .o file in a BSD archive, or be contained in another kind of object.
|
|
lldb::offset_t offset = 0;
|
|
extractor.SetByteOrder(eByteOrderBig);
|
|
header.magic = extractor.GetU32(&offset);
|
|
fat_archs.clear();
|
|
|
|
// Universal mach-o files always have their headers in big endian.
|
|
if (header.magic == FAT_MAGIC || header.magic == FAT_MAGIC_64) {
|
|
const bool is_fat64 = header.magic == FAT_MAGIC_64;
|
|
extractor.SetAddressByteSize(is_fat64 ? 8 : 4);
|
|
|
|
header.nfat_arch = extractor.GetU32(&offset);
|
|
|
|
// Now we should have enough data for all of the fat headers, so lets index
|
|
// them so we know how many architectures that this universal binary
|
|
// contains.
|
|
for (uint32_t arch_idx = 0; arch_idx < header.nfat_arch; ++arch_idx) {
|
|
if (extractor.ValidOffsetForDataOfSize(offset, sizeof(fat_arch))) {
|
|
if (is_fat64) {
|
|
fat_arch_64 arch;
|
|
arch.cputype = extractor.GetU32(&offset);
|
|
arch.cpusubtype = extractor.GetU32(&offset);
|
|
arch.offset = extractor.GetU64(&offset);
|
|
arch.size = extractor.GetU64(&offset);
|
|
arch.align = extractor.GetU32(&offset);
|
|
arch.reserved = extractor.GetU32(&offset);
|
|
fat_archs.emplace_back(arch);
|
|
} else {
|
|
fat_arch arch;
|
|
arch.cputype = extractor.GetU32(&offset);
|
|
arch.cpusubtype = extractor.GetU32(&offset);
|
|
arch.offset = extractor.GetU32(&offset);
|
|
arch.size = extractor.GetU32(&offset);
|
|
arch.align = extractor.GetU32(&offset);
|
|
fat_archs.emplace_back(arch);
|
|
}
|
|
}
|
|
}
|
|
return true;
|
|
}
|
|
|
|
memset(&header, 0, sizeof(header));
|
|
return true;
|
|
}
|
|
|
|
size_t ObjectContainerUniversalMachO::GetNumArchitectures() const {
|
|
return m_header.nfat_arch;
|
|
}
|
|
|
|
bool ObjectContainerUniversalMachO::GetArchitectureAtIndex(
|
|
uint32_t idx, ArchSpec &arch) const {
|
|
if (idx < m_header.nfat_arch) {
|
|
arch.SetArchitecture(eArchTypeMachO, m_fat_archs[idx].GetCPUType(),
|
|
m_fat_archs[idx].GetCPUSubType());
|
|
return true;
|
|
}
|
|
return false;
|
|
}
|
|
|
|
ObjectFileSP
|
|
ObjectContainerUniversalMachO::GetObjectFile(const FileSpec *file) {
|
|
uint32_t arch_idx = 0;
|
|
ArchSpec arch;
|
|
// If the module hasn't specified an architecture yet, set it to the default
|
|
// architecture:
|
|
ModuleSP module_sp(GetModule());
|
|
if (module_sp) {
|
|
if (!module_sp->GetArchitecture().IsValid()) {
|
|
arch = Target::GetDefaultArchitecture();
|
|
if (!arch.IsValid())
|
|
arch.SetTriple(LLDB_ARCH_DEFAULT);
|
|
} else
|
|
arch = module_sp->GetArchitecture();
|
|
|
|
ArchSpec curr_arch;
|
|
// First, try to find an exact match for the Arch of the Target.
|
|
for (arch_idx = 0; arch_idx < m_header.nfat_arch; ++arch_idx) {
|
|
if (GetArchitectureAtIndex(arch_idx, curr_arch) &&
|
|
arch.IsExactMatch(curr_arch))
|
|
break;
|
|
}
|
|
|
|
// Failing an exact match, try to find a compatible Arch of the Target.
|
|
if (arch_idx >= m_header.nfat_arch) {
|
|
for (arch_idx = 0; arch_idx < m_header.nfat_arch; ++arch_idx) {
|
|
if (GetArchitectureAtIndex(arch_idx, curr_arch) &&
|
|
arch.IsCompatibleMatch(curr_arch))
|
|
break;
|
|
}
|
|
}
|
|
|
|
if (arch_idx < m_header.nfat_arch) {
|
|
DataExtractorSP extractor_sp;
|
|
lldb::offset_t data_offset = 0;
|
|
return ObjectFile::FindPlugin(
|
|
module_sp, file, m_offset + m_fat_archs[arch_idx].GetOffset(),
|
|
m_fat_archs[arch_idx].GetSize(), extractor_sp, data_offset);
|
|
}
|
|
}
|
|
return ObjectFileSP();
|
|
}
|
|
|
|
size_t ObjectContainerUniversalMachO::GetModuleSpecifications(
|
|
const lldb_private::FileSpec &file, lldb::DataExtractorSP &extractor_sp,
|
|
lldb::offset_t data_offset, lldb::offset_t file_offset,
|
|
lldb::offset_t file_size, lldb_private::ModuleSpecList &specs) {
|
|
const size_t initial_count = specs.GetSize();
|
|
if (!extractor_sp)
|
|
return initial_count;
|
|
|
|
DataExtractorSP data_extractor_sp = extractor_sp->GetSubsetExtractorSP(
|
|
data_offset, extractor_sp->GetByteSize());
|
|
if (ObjectContainerUniversalMachO::MagicBytesMatch(*data_extractor_sp)) {
|
|
llvm::MachO::fat_header header;
|
|
std::vector<FatArch> fat_archs;
|
|
if (ParseHeader(*data_extractor_sp, header, fat_archs)) {
|
|
for (const FatArch &fat_arch : fat_archs) {
|
|
const lldb::offset_t slice_file_offset =
|
|
fat_arch.GetOffset() + file_offset;
|
|
if (fat_arch.GetOffset() < file_size && file_size > slice_file_offset) {
|
|
ObjectFile::GetModuleSpecifications(
|
|
file, slice_file_offset, file_size - slice_file_offset, specs);
|
|
}
|
|
}
|
|
}
|
|
}
|
|
return specs.GetSize() - initial_count;
|
|
}
|