diff options
Diffstat (limited to 'source/Plugins/ObjectFile')
| -rw-r--r-- | source/Plugins/ObjectFile/ELF/ELFHeader.cpp | 607 | ||||
| -rw-r--r-- | source/Plugins/ObjectFile/ELF/ELFHeader.h | 645 | ||||
| -rw-r--r-- | source/Plugins/ObjectFile/ELF/ObjectFileELF.cpp | 5918 | ||||
| -rw-r--r-- | source/Plugins/ObjectFile/ELF/ObjectFileELF.h | 766 | ||||
| -rw-r--r-- | source/Plugins/ObjectFile/JIT/ObjectFileJIT.cpp | 429 | ||||
| -rw-r--r-- | source/Plugins/ObjectFile/JIT/ObjectFileJIT.h | 201 | ||||
| -rw-r--r-- | source/Plugins/ObjectFile/Mach-O/ObjectFileMachO.cpp | 11217 | ||||
| -rw-r--r-- | source/Plugins/ObjectFile/Mach-O/ObjectFileMachO.h | 396 | ||||
| -rw-r--r-- | source/Plugins/ObjectFile/PECOFF/ObjectFilePECOFF.cpp | 1748 | ||||
| -rw-r--r-- | source/Plugins/ObjectFile/PECOFF/ObjectFilePECOFF.h | 540 | ||||
| -rw-r--r-- | source/Plugins/ObjectFile/PECOFF/WindowsMiniDump.cpp | 63 | ||||
| -rw-r--r-- | source/Plugins/ObjectFile/PECOFF/WindowsMiniDump.h | 9 |
12 files changed, 10951 insertions, 11588 deletions
diff --git a/source/Plugins/ObjectFile/ELF/ELFHeader.cpp b/source/Plugins/ObjectFile/ELF/ELFHeader.cpp index 625cce3c10627..a3e82390d262e 100644 --- a/source/Plugins/ObjectFile/ELF/ELFHeader.cpp +++ b/source/Plugins/ObjectFile/ELF/ELFHeader.cpp @@ -24,240 +24,201 @@ using namespace llvm::ELF; // // GetMaxU64 and GetMaxS64 wrap the similarly named methods from DataExtractor // with error handling code and provide for parsing a sequence of values. -static bool -GetMaxU64(const lldb_private::DataExtractor &data, - lldb::offset_t *offset, - uint64_t *value, - uint32_t byte_size) -{ - const lldb::offset_t saved_offset = *offset; - *value = data.GetMaxU64(offset, byte_size); - return *offset != saved_offset; +static bool GetMaxU64(const lldb_private::DataExtractor &data, + lldb::offset_t *offset, uint64_t *value, + uint32_t byte_size) { + const lldb::offset_t saved_offset = *offset; + *value = data.GetMaxU64(offset, byte_size); + return *offset != saved_offset; } -static bool -GetMaxU64(const lldb_private::DataExtractor &data, - lldb::offset_t *offset, - uint64_t *value, - uint32_t byte_size, - uint32_t count) -{ - lldb::offset_t saved_offset = *offset; - - for (uint32_t i = 0; i < count; ++i, ++value) - { - if (GetMaxU64(data, offset, value, byte_size) == false) - { - *offset = saved_offset; - return false; - } +static bool GetMaxU64(const lldb_private::DataExtractor &data, + lldb::offset_t *offset, uint64_t *value, + uint32_t byte_size, uint32_t count) { + lldb::offset_t saved_offset = *offset; + + for (uint32_t i = 0; i < count; ++i, ++value) { + if (GetMaxU64(data, offset, value, byte_size) == false) { + *offset = saved_offset; + return false; } - return true; + } + return true; } -static bool -GetMaxS64(const lldb_private::DataExtractor &data, - lldb::offset_t *offset, - int64_t *value, - uint32_t byte_size) -{ - const lldb::offset_t saved_offset = *offset; - *value = data.GetMaxS64(offset, byte_size); - return *offset != saved_offset; +static bool GetMaxS64(const lldb_private::DataExtractor &data, + lldb::offset_t *offset, int64_t *value, + uint32_t byte_size) { + const lldb::offset_t saved_offset = *offset; + *value = data.GetMaxS64(offset, byte_size); + return *offset != saved_offset; } -static bool -GetMaxS64(const lldb_private::DataExtractor &data, - lldb::offset_t *offset, - int64_t *value, - uint32_t byte_size, - uint32_t count) -{ - lldb::offset_t saved_offset = *offset; - - for (uint32_t i = 0; i < count; ++i, ++value) - { - if (GetMaxS64(data, offset, value, byte_size) == false) - { - *offset = saved_offset; - return false; - } +static bool GetMaxS64(const lldb_private::DataExtractor &data, + lldb::offset_t *offset, int64_t *value, + uint32_t byte_size, uint32_t count) { + lldb::offset_t saved_offset = *offset; + + for (uint32_t i = 0; i < count; ++i, ++value) { + if (GetMaxS64(data, offset, value, byte_size) == false) { + *offset = saved_offset; + return false; } - return true; + } + return true; } //------------------------------------------------------------------------------ // ELFHeader -ELFHeader::ELFHeader() -{ - memset(this, 0, sizeof(ELFHeader)); -} +ELFHeader::ELFHeader() { memset(this, 0, sizeof(ELFHeader)); } -ByteOrder -ELFHeader::GetByteOrder() const -{ - if (e_ident[EI_DATA] == ELFDATA2MSB) - return eByteOrderBig; - if (e_ident[EI_DATA] == ELFDATA2LSB) - return eByteOrderLittle; - return eByteOrderInvalid; +ByteOrder ELFHeader::GetByteOrder() const { + if (e_ident[EI_DATA] == ELFDATA2MSB) + return eByteOrderBig; + if (e_ident[EI_DATA] == ELFDATA2LSB) + return eByteOrderLittle; + return eByteOrderInvalid; } -bool -ELFHeader::Parse(lldb_private::DataExtractor &data, lldb::offset_t *offset) -{ - // Read e_ident. This provides byte order and address size info. - if (data.GetU8(offset, &e_ident, EI_NIDENT) == NULL) - return false; +bool ELFHeader::Parse(lldb_private::DataExtractor &data, + lldb::offset_t *offset) { + // Read e_ident. This provides byte order and address size info. + if (data.GetU8(offset, &e_ident, EI_NIDENT) == NULL) + return false; - const unsigned byte_size = Is32Bit() ? 4 : 8; - data.SetByteOrder(GetByteOrder()); - data.SetAddressByteSize(byte_size); + const unsigned byte_size = Is32Bit() ? 4 : 8; + data.SetByteOrder(GetByteOrder()); + data.SetAddressByteSize(byte_size); - // Read e_type and e_machine. - if (data.GetU16(offset, &e_type, 2) == NULL) - return false; + // Read e_type and e_machine. + if (data.GetU16(offset, &e_type, 2) == NULL) + return false; - // Read e_version. - if (data.GetU32(offset, &e_version, 1) == NULL) - return false; + // Read e_version. + if (data.GetU32(offset, &e_version, 1) == NULL) + return false; - // Read e_entry, e_phoff and e_shoff. - if (GetMaxU64(data, offset, &e_entry, byte_size, 3) == false) - return false; + // Read e_entry, e_phoff and e_shoff. + if (GetMaxU64(data, offset, &e_entry, byte_size, 3) == false) + return false; - // Read e_flags. - if (data.GetU32(offset, &e_flags, 1) == NULL) - return false; + // Read e_flags. + if (data.GetU32(offset, &e_flags, 1) == NULL) + return false; - // Read e_ehsize, e_phentsize, e_phnum, e_shentsize, e_shnum and - // e_shstrndx. - if (data.GetU16(offset, &e_ehsize, 6) == NULL) - return false; + // Read e_ehsize, e_phentsize, e_phnum, e_shentsize, e_shnum and + // e_shstrndx. + if (data.GetU16(offset, &e_ehsize, 6) == NULL) + return false; - return true; + return true; } -bool -ELFHeader::MagicBytesMatch(const uint8_t *magic) -{ - return memcmp(magic, ElfMagic, strlen(ElfMagic)) == 0; +bool ELFHeader::MagicBytesMatch(const uint8_t *magic) { + return memcmp(magic, ElfMagic, strlen(ElfMagic)) == 0; } -unsigned -ELFHeader::AddressSizeInBytes(const uint8_t *magic) -{ - unsigned address_size = 0; - - switch (magic[EI_CLASS]) - { - case ELFCLASS32: - address_size = 4; - break; - - case ELFCLASS64: - address_size = 8; - break; - } - return address_size; -} +unsigned ELFHeader::AddressSizeInBytes(const uint8_t *magic) { + unsigned address_size = 0; -unsigned -ELFHeader::GetRelocationJumpSlotType() const -{ - unsigned slot = 0; - - switch (e_machine) - { - default: - assert(false && "architecture not supported"); - break; - case EM_PPC: - slot = R_PPC_JMP_SLOT; - break; - case EM_PPC64: - slot = R_PPC64_JMP_SLOT; - break; - case EM_386: - case EM_IAMCU: // FIXME: is this correct? - slot = R_386_JUMP_SLOT; - break; - case EM_X86_64: - slot = R_X86_64_JUMP_SLOT; - break; - case EM_ARM: - slot = R_ARM_JUMP_SLOT; - break; - case EM_HEXAGON: - slot = R_HEX_JMP_SLOT; - break; - case EM_AARCH64: - slot = R_AARCH64_JUMP_SLOT; - break; - case EM_MIPS: - slot = R_MIPS_JUMP_SLOT; - break; - case EM_S390: - slot = R_390_JMP_SLOT; - break; - } + switch (magic[EI_CLASS]) { + case ELFCLASS32: + address_size = 4; + break; - return slot; + case ELFCLASS64: + address_size = 8; + break; + } + return address_size; +} + +unsigned ELFHeader::GetRelocationJumpSlotType() const { + unsigned slot = 0; + + switch (e_machine) { + default: + assert(false && "architecture not supported"); + break; + case EM_PPC: + slot = R_PPC_JMP_SLOT; + break; + case EM_PPC64: + slot = R_PPC64_JMP_SLOT; + break; + case EM_386: + case EM_IAMCU: // FIXME: is this correct? + slot = R_386_JUMP_SLOT; + break; + case EM_X86_64: + slot = R_X86_64_JUMP_SLOT; + break; + case EM_ARM: + slot = R_ARM_JUMP_SLOT; + break; + case EM_HEXAGON: + slot = R_HEX_JMP_SLOT; + break; + case EM_AARCH64: + slot = R_AARCH64_JUMP_SLOT; + break; + case EM_MIPS: + slot = R_MIPS_JUMP_SLOT; + break; + case EM_S390: + slot = R_390_JMP_SLOT; + break; + } + + return slot; } //------------------------------------------------------------------------------ // ELFSectionHeader -ELFSectionHeader::ELFSectionHeader() -{ - memset(this, 0, sizeof(ELFSectionHeader)); +ELFSectionHeader::ELFSectionHeader() { + memset(this, 0, sizeof(ELFSectionHeader)); } -bool -ELFSectionHeader::Parse(const lldb_private::DataExtractor &data, - lldb::offset_t *offset) -{ - const unsigned byte_size = data.GetAddressByteSize(); +bool ELFSectionHeader::Parse(const lldb_private::DataExtractor &data, + lldb::offset_t *offset) { + const unsigned byte_size = data.GetAddressByteSize(); - // Read sh_name and sh_type. - if (data.GetU32(offset, &sh_name, 2) == NULL) - return false; + // Read sh_name and sh_type. + if (data.GetU32(offset, &sh_name, 2) == NULL) + return false; - // Read sh_flags. - if (GetMaxU64(data, offset, &sh_flags, byte_size) == false) - return false; + // Read sh_flags. + if (GetMaxU64(data, offset, &sh_flags, byte_size) == false) + return false; - // Read sh_addr, sh_off and sh_size. - if (GetMaxU64(data, offset, &sh_addr, byte_size, 3) == false) - return false; + // Read sh_addr, sh_off and sh_size. + if (GetMaxU64(data, offset, &sh_addr, byte_size, 3) == false) + return false; - // Read sh_link and sh_info. - if (data.GetU32(offset, &sh_link, 2) == NULL) - return false; + // Read sh_link and sh_info. + if (data.GetU32(offset, &sh_link, 2) == NULL) + return false; - // Read sh_addralign and sh_entsize. - if (GetMaxU64(data, offset, &sh_addralign, byte_size, 2) == false) - return false; + // Read sh_addralign and sh_entsize. + if (GetMaxU64(data, offset, &sh_addralign, byte_size, 2) == false) + return false; - return true; + return true; } //------------------------------------------------------------------------------ // ELFSymbol -ELFSymbol::ELFSymbol() -{ - memset(this, 0, sizeof(ELFSymbol)); -} +ELFSymbol::ELFSymbol() { memset(this, 0, sizeof(ELFSymbol)); } -#define ENUM_TO_CSTR(e) case e: return #e +#define ENUM_TO_CSTR(e) \ + case e: \ + return #e -const char * -ELFSymbol::bindingToCString(unsigned char binding) -{ - switch (binding) - { +const char *ELFSymbol::bindingToCString(unsigned char binding) { + switch (binding) { ENUM_TO_CSTR(STB_LOCAL); ENUM_TO_CSTR(STB_GLOBAL); ENUM_TO_CSTR(STB_WEAK); @@ -265,15 +226,12 @@ ELFSymbol::bindingToCString(unsigned char binding) ENUM_TO_CSTR(STB_HIOS); ENUM_TO_CSTR(STB_LOPROC); ENUM_TO_CSTR(STB_HIPROC); - } - return ""; + } + return ""; } -const char * -ELFSymbol::typeToCString(unsigned char type) -{ - switch (type) - { +const char *ELFSymbol::typeToCString(unsigned char type) { + switch (type) { ENUM_TO_CSTR(STT_NOTYPE); ENUM_TO_CSTR(STT_OBJECT); ENUM_TO_CSTR(STT_FUNC); @@ -285,16 +243,13 @@ ELFSymbol::typeToCString(unsigned char type) ENUM_TO_CSTR(STT_HIOS); ENUM_TO_CSTR(STT_LOPROC); ENUM_TO_CSTR(STT_HIPROC); - } - return ""; + } + return ""; } -const char * -ELFSymbol::sectionIndexToCString (elf_half shndx, - const lldb_private::SectionList *section_list) -{ - switch (shndx) - { +const char *ELFSymbol::sectionIndexToCString( + elf_half shndx, const lldb_private::SectionList *section_list) { + switch (shndx) { ENUM_TO_CSTR(SHN_UNDEF); ENUM_TO_CSTR(SHN_LOPROC); ENUM_TO_CSTR(SHN_HIPROC); @@ -303,180 +258,148 @@ ELFSymbol::sectionIndexToCString (elf_half shndx, ENUM_TO_CSTR(SHN_ABS); ENUM_TO_CSTR(SHN_COMMON); ENUM_TO_CSTR(SHN_XINDEX); - default: - { - const lldb_private::Section *section = section_list->GetSectionAtIndex(shndx).get(); - if (section) - return section->GetName().AsCString(""); - } - break; - } - return ""; + default: { + const lldb_private::Section *section = + section_list->GetSectionAtIndex(shndx).get(); + if (section) + return section->GetName().AsCString(""); + } break; + } + return ""; } -void -ELFSymbol::Dump (lldb_private::Stream *s, - uint32_t idx, - const lldb_private::DataExtractor *strtab_data, - const lldb_private::SectionList *section_list) -{ - s->Printf("[%3u] 0x%16.16" PRIx64 " 0x%16.16" PRIx64 " 0x%8.8x 0x%2.2x (%-10s %-13s) 0x%2.2x 0x%4.4x (%-10s) %s\n", - idx, - st_value, - st_size, - st_name, - st_info, - bindingToCString (getBinding()), - typeToCString (getType()), - st_other, - st_shndx, - sectionIndexToCString (st_shndx, section_list), - strtab_data ? strtab_data->PeekCStr(st_name) : ""); +void ELFSymbol::Dump(lldb_private::Stream *s, uint32_t idx, + const lldb_private::DataExtractor *strtab_data, + const lldb_private::SectionList *section_list) { + s->Printf("[%3u] 0x%16.16" PRIx64 " 0x%16.16" PRIx64 + " 0x%8.8x 0x%2.2x (%-10s %-13s) 0x%2.2x 0x%4.4x (%-10s) %s\n", + idx, st_value, st_size, st_name, st_info, + bindingToCString(getBinding()), typeToCString(getType()), st_other, + st_shndx, sectionIndexToCString(st_shndx, section_list), + strtab_data ? strtab_data->PeekCStr(st_name) : ""); } -bool -ELFSymbol::Parse(const lldb_private::DataExtractor &data, lldb::offset_t *offset) -{ - const unsigned byte_size = data.GetAddressByteSize(); - const bool parsing_32 = byte_size == 4; - - // Read st_name. - if (data.GetU32(offset, &st_name, 1) == NULL) - return false; - - if (parsing_32) - { - // Read st_value and st_size. - if (GetMaxU64(data, offset, &st_value, byte_size, 2) == false) - return false; - - // Read st_info and st_other. - if (data.GetU8(offset, &st_info, 2) == NULL) - return false; - - // Read st_shndx. - if (data.GetU16(offset, &st_shndx, 1) == NULL) - return false; - } - else - { - // Read st_info and st_other. - if (data.GetU8(offset, &st_info, 2) == NULL) - return false; - - // Read st_shndx. - if (data.GetU16(offset, &st_shndx, 1) == NULL) - return false; - - // Read st_value and st_size. - if (data.GetU64(offset, &st_value, 2) == NULL) - return false; - } - return true; +bool ELFSymbol::Parse(const lldb_private::DataExtractor &data, + lldb::offset_t *offset) { + const unsigned byte_size = data.GetAddressByteSize(); + const bool parsing_32 = byte_size == 4; + + // Read st_name. + if (data.GetU32(offset, &st_name, 1) == NULL) + return false; + + if (parsing_32) { + // Read st_value and st_size. + if (GetMaxU64(data, offset, &st_value, byte_size, 2) == false) + return false; + + // Read st_info and st_other. + if (data.GetU8(offset, &st_info, 2) == NULL) + return false; + + // Read st_shndx. + if (data.GetU16(offset, &st_shndx, 1) == NULL) + return false; + } else { + // Read st_info and st_other. + if (data.GetU8(offset, &st_info, 2) == NULL) + return false; + + // Read st_shndx. + if (data.GetU16(offset, &st_shndx, 1) == NULL) + return false; + + // Read st_value and st_size. + if (data.GetU64(offset, &st_value, 2) == NULL) + return false; + } + return true; } //------------------------------------------------------------------------------ // ELFProgramHeader -ELFProgramHeader::ELFProgramHeader() -{ - memset(this, 0, sizeof(ELFProgramHeader)); +ELFProgramHeader::ELFProgramHeader() { + memset(this, 0, sizeof(ELFProgramHeader)); } -bool -ELFProgramHeader::Parse(const lldb_private::DataExtractor &data, - lldb::offset_t *offset) -{ - const uint32_t byte_size = data.GetAddressByteSize(); - const bool parsing_32 = byte_size == 4; - - // Read p_type; - if (data.GetU32(offset, &p_type, 1) == NULL) - return false; - - if (parsing_32) { - // Read p_offset, p_vaddr, p_paddr, p_filesz and p_memsz. - if (GetMaxU64(data, offset, &p_offset, byte_size, 5) == false) - return false; - - // Read p_flags. - if (data.GetU32(offset, &p_flags, 1) == NULL) - return false; - - // Read p_align. - if (GetMaxU64(data, offset, &p_align, byte_size) == false) - return false; - } - else { - // Read p_flags. - if (data.GetU32(offset, &p_flags, 1) == NULL) - return false; - - // Read p_offset, p_vaddr, p_paddr, p_filesz, p_memsz and p_align. - if (GetMaxU64(data, offset, &p_offset, byte_size, 6) == false) - return false; - } - - return true; +bool ELFProgramHeader::Parse(const lldb_private::DataExtractor &data, + lldb::offset_t *offset) { + const uint32_t byte_size = data.GetAddressByteSize(); + const bool parsing_32 = byte_size == 4; + + // Read p_type; + if (data.GetU32(offset, &p_type, 1) == NULL) + return false; + + if (parsing_32) { + // Read p_offset, p_vaddr, p_paddr, p_filesz and p_memsz. + if (GetMaxU64(data, offset, &p_offset, byte_size, 5) == false) + return false; + + // Read p_flags. + if (data.GetU32(offset, &p_flags, 1) == NULL) + return false; + + // Read p_align. + if (GetMaxU64(data, offset, &p_align, byte_size) == false) + return false; + } else { + // Read p_flags. + if (data.GetU32(offset, &p_flags, 1) == NULL) + return false; + + // Read p_offset, p_vaddr, p_paddr, p_filesz, p_memsz and p_align. + if (GetMaxU64(data, offset, &p_offset, byte_size, 6) == false) + return false; + } + + return true; } //------------------------------------------------------------------------------ // ELFDynamic -ELFDynamic::ELFDynamic() -{ - memset(this, 0, sizeof(ELFDynamic)); -} +ELFDynamic::ELFDynamic() { memset(this, 0, sizeof(ELFDynamic)); } -bool -ELFDynamic::Parse(const lldb_private::DataExtractor &data, lldb::offset_t *offset) -{ - const unsigned byte_size = data.GetAddressByteSize(); - return GetMaxS64(data, offset, &d_tag, byte_size, 2); +bool ELFDynamic::Parse(const lldb_private::DataExtractor &data, + lldb::offset_t *offset) { + const unsigned byte_size = data.GetAddressByteSize(); + return GetMaxS64(data, offset, &d_tag, byte_size, 2); } //------------------------------------------------------------------------------ // ELFRel -ELFRel::ELFRel() -{ - memset(this, 0, sizeof(ELFRel)); -} +ELFRel::ELFRel() { memset(this, 0, sizeof(ELFRel)); } -bool -ELFRel::Parse(const lldb_private::DataExtractor &data, lldb::offset_t *offset) -{ - const unsigned byte_size = data.GetAddressByteSize(); +bool ELFRel::Parse(const lldb_private::DataExtractor &data, + lldb::offset_t *offset) { + const unsigned byte_size = data.GetAddressByteSize(); - // Read r_offset and r_info. - if (GetMaxU64(data, offset, &r_offset, byte_size, 2) == false) - return false; + // Read r_offset and r_info. + if (GetMaxU64(data, offset, &r_offset, byte_size, 2) == false) + return false; - return true; + return true; } //------------------------------------------------------------------------------ // ELFRela -ELFRela::ELFRela() -{ - memset(this, 0, sizeof(ELFRela)); -} +ELFRela::ELFRela() { memset(this, 0, sizeof(ELFRela)); } -bool -ELFRela::Parse(const lldb_private::DataExtractor &data, lldb::offset_t *offset) -{ - const unsigned byte_size = data.GetAddressByteSize(); +bool ELFRela::Parse(const lldb_private::DataExtractor &data, + lldb::offset_t *offset) { + const unsigned byte_size = data.GetAddressByteSize(); - // Read r_offset and r_info. - if (GetMaxU64(data, offset, &r_offset, byte_size, 2) == false) - return false; + // Read r_offset and r_info. + if (GetMaxU64(data, offset, &r_offset, byte_size, 2) == false) + return false; - // Read r_addend; - if (GetMaxS64(data, offset, &r_addend, byte_size) == false) - return false; + // Read r_addend; + if (GetMaxS64(data, offset, &r_addend, byte_size) == false) + return false; - return true; + return true; } - - diff --git a/source/Plugins/ObjectFile/ELF/ELFHeader.h b/source/Plugins/ObjectFile/ELF/ELFHeader.h index 4ea22b51baf70..71b200f1c16b6 100644 --- a/source/Plugins/ObjectFile/ELF/ELFHeader.h +++ b/source/Plugins/ObjectFile/ELF/ELFHeader.h @@ -25,13 +25,11 @@ #include "lldb/lldb-enumerations.h" -namespace lldb_private -{ +namespace lldb_private { class DataExtractor; } // End namespace lldb_private. -namespace elf -{ +namespace elf { //------------------------------------------------------------------------------ /// @name ELF type definitions. @@ -44,10 +42,10 @@ typedef uint64_t elf_addr; typedef uint64_t elf_off; typedef uint16_t elf_half; typedef uint32_t elf_word; -typedef int32_t elf_sword; +typedef int32_t elf_sword; typedef uint64_t elf_size; typedef uint64_t elf_xword; -typedef int64_t elf_sxword; +typedef int64_t elf_sxword; //@} //------------------------------------------------------------------------------ @@ -56,376 +54,331 @@ typedef int64_t elf_sxword; /// /// This object is used to identify the general attributes on an ELF file and to /// locate additional sections within the file. -struct ELFHeader -{ - unsigned char e_ident[llvm::ELF::EI_NIDENT]; ///< ELF file identification. - elf_addr e_entry; ///< Virtual address program entry point. - elf_off e_phoff; ///< File offset of program header table. - elf_off e_shoff; ///< File offset of section header table. - elf_word e_flags; ///< Processor specific flags. - elf_word e_version; ///< Version of object file (always 1). - elf_half e_type; ///< Object file type. - elf_half e_machine; ///< Target architecture. - elf_half e_ehsize; ///< Byte size of the ELF header. - elf_half e_phentsize; ///< Size of a program header table entry. - elf_half e_phnum; ///< Number of program header entries. - elf_half e_shentsize; ///< Size of a section header table entry. - elf_half e_shnum; ///< Number of section header entries. - elf_half e_shstrndx; ///< String table section index. - - ELFHeader(); - - //-------------------------------------------------------------------------- - /// Returns true if this is a 32 bit ELF file header. - /// - /// @return - /// True if this is a 32 bit ELF file header. - bool Is32Bit() const { - return e_ident[llvm::ELF::EI_CLASS] == llvm::ELF::ELFCLASS32; - } - - //-------------------------------------------------------------------------- - /// Returns true if this is a 64 bit ELF file header. - /// - /// @return - /// True if this is a 64 bit ELF file header. - bool Is64Bit() const { - return e_ident[llvm::ELF::EI_CLASS] == llvm::ELF::ELFCLASS64; - } - - //-------------------------------------------------------------------------- - /// The byte order of this ELF file header. - /// - /// @return - /// The byte order of this ELF file as described by the header. - lldb::ByteOrder - GetByteOrder() const; - - //-------------------------------------------------------------------------- - /// The jump slot relocation type of this ELF. - unsigned - GetRelocationJumpSlotType() const; - - //-------------------------------------------------------------------------- - /// Parse an ELFHeader entry starting at position \p offset and - /// update the data extractor with the address size and byte order - /// attributes as defined by the header. - /// - /// @param[in,out] data - /// The DataExtractor to read from. Updated with the address size and - /// byte order attributes appropriate to this header. - /// - /// @param[in,out] offset - /// Pointer to an offset in the data. On return the offset will be - /// advanced by the number of bytes read. - /// - /// @return - /// True if the ELFHeader was successfully read and false - /// otherwise. - bool - Parse(lldb_private::DataExtractor &data, lldb::offset_t *offset); - - //-------------------------------------------------------------------------- - /// Examines at most EI_NIDENT bytes starting from the given pointer and - /// determines if the magic ELF identification exists. - /// - /// @return - /// True if the given sequence of bytes identifies an ELF file. - static bool - MagicBytesMatch(const uint8_t *magic); - - //-------------------------------------------------------------------------- - /// Examines at most EI_NIDENT bytes starting from the given address and - /// determines the address size of the underlying ELF file. This function - /// should only be called on an pointer for which MagicBytesMatch returns - /// true. - /// - /// @return - /// The number of bytes forming an address in the ELF file (either 4 or - /// 8), else zero if the address size could not be determined. - static unsigned - AddressSizeInBytes(const uint8_t *magic); +struct ELFHeader { + unsigned char e_ident[llvm::ELF::EI_NIDENT]; ///< ELF file identification. + elf_addr e_entry; ///< Virtual address program entry point. + elf_off e_phoff; ///< File offset of program header table. + elf_off e_shoff; ///< File offset of section header table. + elf_word e_flags; ///< Processor specific flags. + elf_word e_version; ///< Version of object file (always 1). + elf_half e_type; ///< Object file type. + elf_half e_machine; ///< Target architecture. + elf_half e_ehsize; ///< Byte size of the ELF header. + elf_half e_phentsize; ///< Size of a program header table entry. + elf_half e_phnum; ///< Number of program header entries. + elf_half e_shentsize; ///< Size of a section header table entry. + elf_half e_shnum; ///< Number of section header entries. + elf_half e_shstrndx; ///< String table section index. + + ELFHeader(); + + //-------------------------------------------------------------------------- + /// Returns true if this is a 32 bit ELF file header. + /// + /// @return + /// True if this is a 32 bit ELF file header. + bool Is32Bit() const { + return e_ident[llvm::ELF::EI_CLASS] == llvm::ELF::ELFCLASS32; + } + + //-------------------------------------------------------------------------- + /// Returns true if this is a 64 bit ELF file header. + /// + /// @return + /// True if this is a 64 bit ELF file header. + bool Is64Bit() const { + return e_ident[llvm::ELF::EI_CLASS] == llvm::ELF::ELFCLASS64; + } + + //-------------------------------------------------------------------------- + /// The byte order of this ELF file header. + /// + /// @return + /// The byte order of this ELF file as described by the header. + lldb::ByteOrder GetByteOrder() const; + + //-------------------------------------------------------------------------- + /// The jump slot relocation type of this ELF. + unsigned GetRelocationJumpSlotType() const; + + //-------------------------------------------------------------------------- + /// Parse an ELFHeader entry starting at position \p offset and + /// update the data extractor with the address size and byte order + /// attributes as defined by the header. + /// + /// @param[in,out] data + /// The DataExtractor to read from. Updated with the address size and + /// byte order attributes appropriate to this header. + /// + /// @param[in,out] offset + /// Pointer to an offset in the data. On return the offset will be + /// advanced by the number of bytes read. + /// + /// @return + /// True if the ELFHeader was successfully read and false + /// otherwise. + bool Parse(lldb_private::DataExtractor &data, lldb::offset_t *offset); + + //-------------------------------------------------------------------------- + /// Examines at most EI_NIDENT bytes starting from the given pointer and + /// determines if the magic ELF identification exists. + /// + /// @return + /// True if the given sequence of bytes identifies an ELF file. + static bool MagicBytesMatch(const uint8_t *magic); + + //-------------------------------------------------------------------------- + /// Examines at most EI_NIDENT bytes starting from the given address and + /// determines the address size of the underlying ELF file. This function + /// should only be called on an pointer for which MagicBytesMatch returns + /// true. + /// + /// @return + /// The number of bytes forming an address in the ELF file (either 4 or + /// 8), else zero if the address size could not be determined. + static unsigned AddressSizeInBytes(const uint8_t *magic); }; //------------------------------------------------------------------------------ /// @class ELFSectionHeader /// @brief Generic representation of an ELF section header. -struct ELFSectionHeader -{ - elf_word sh_name; ///< Section name string index. - elf_word sh_type; ///< Section type. - elf_xword sh_flags; ///< Section attributes. - elf_addr sh_addr; ///< Virtual address of the section in memory. - elf_off sh_offset; ///< Start of section from beginning of file. - elf_xword sh_size; ///< Number of bytes occupied in the file. - elf_word sh_link; ///< Index of associated section. - elf_word sh_info; ///< Extra section info (overloaded). - elf_xword sh_addralign; ///< Power of two alignment constraint. - elf_xword sh_entsize; ///< Byte size of each section entry. - - ELFSectionHeader(); - - //-------------------------------------------------------------------------- - /// Parse an ELFSectionHeader entry from the given DataExtracter starting at - /// position \p offset. - /// - /// @param[in] data - /// The DataExtractor to read from. The address size of the extractor - /// determines if a 32 or 64 bit object should be read. - /// - /// @param[in,out] offset - /// Pointer to an offset in the data. On return the offset will be - /// advanced by the number of bytes read. - /// - /// @return - /// True if the ELFSectionHeader was successfully read and false - /// otherwise. - bool - Parse(const lldb_private::DataExtractor &data, lldb::offset_t *offset); +struct ELFSectionHeader { + elf_word sh_name; ///< Section name string index. + elf_word sh_type; ///< Section type. + elf_xword sh_flags; ///< Section attributes. + elf_addr sh_addr; ///< Virtual address of the section in memory. + elf_off sh_offset; ///< Start of section from beginning of file. + elf_xword sh_size; ///< Number of bytes occupied in the file. + elf_word sh_link; ///< Index of associated section. + elf_word sh_info; ///< Extra section info (overloaded). + elf_xword sh_addralign; ///< Power of two alignment constraint. + elf_xword sh_entsize; ///< Byte size of each section entry. + + ELFSectionHeader(); + + //-------------------------------------------------------------------------- + /// Parse an ELFSectionHeader entry from the given DataExtracter starting at + /// position \p offset. + /// + /// @param[in] data + /// The DataExtractor to read from. The address size of the extractor + /// determines if a 32 or 64 bit object should be read. + /// + /// @param[in,out] offset + /// Pointer to an offset in the data. On return the offset will be + /// advanced by the number of bytes read. + /// + /// @return + /// True if the ELFSectionHeader was successfully read and false + /// otherwise. + bool Parse(const lldb_private::DataExtractor &data, lldb::offset_t *offset); }; //------------------------------------------------------------------------------ /// @class ELFProgramHeader /// @brief Generic representation of an ELF program header. -struct ELFProgramHeader -{ - elf_word p_type; ///< Type of program segment. - elf_word p_flags; ///< Segment attributes. - elf_off p_offset; ///< Start of segment from beginning of file. - elf_addr p_vaddr; ///< Virtual address of segment in memory. - elf_addr p_paddr; ///< Physical address (for non-VM systems). - elf_xword p_filesz; ///< Byte size of the segment in file. - elf_xword p_memsz; ///< Byte size of the segment in memory. - elf_xword p_align; ///< Segment alignment constraint. - - ELFProgramHeader(); - - /// Parse an ELFProgramHeader entry from the given DataExtractor starting at - /// position \p offset. The address size of the DataExtractor determines if - /// a 32 or 64 bit object is to be parsed. - /// - /// @param[in] data - /// The DataExtractor to read from. The address size of the extractor - /// determines if a 32 or 64 bit object should be read. - /// - /// @param[in,out] offset - /// Pointer to an offset in the data. On return the offset will be - /// advanced by the number of bytes read. - /// - /// @return - /// True if the ELFProgramHeader was successfully read and false - /// otherwise. - bool - Parse(const lldb_private::DataExtractor &data, lldb::offset_t *offset); +struct ELFProgramHeader { + elf_word p_type; ///< Type of program segment. + elf_word p_flags; ///< Segment attributes. + elf_off p_offset; ///< Start of segment from beginning of file. + elf_addr p_vaddr; ///< Virtual address of segment in memory. + elf_addr p_paddr; ///< Physical address (for non-VM systems). + elf_xword p_filesz; ///< Byte size of the segment in file. + elf_xword p_memsz; ///< Byte size of the segment in memory. + elf_xword p_align; ///< Segment alignment constraint. + + ELFProgramHeader(); + + /// Parse an ELFProgramHeader entry from the given DataExtractor starting at + /// position \p offset. The address size of the DataExtractor determines if + /// a 32 or 64 bit object is to be parsed. + /// + /// @param[in] data + /// The DataExtractor to read from. The address size of the extractor + /// determines if a 32 or 64 bit object should be read. + /// + /// @param[in,out] offset + /// Pointer to an offset in the data. On return the offset will be + /// advanced by the number of bytes read. + /// + /// @return + /// True if the ELFProgramHeader was successfully read and false + /// otherwise. + bool Parse(const lldb_private::DataExtractor &data, lldb::offset_t *offset); }; //------------------------------------------------------------------------------ /// @class ELFSymbol /// @brief Represents a symbol within an ELF symbol table. -struct ELFSymbol -{ - elf_addr st_value; ///< Absolute or relocatable address. - elf_xword st_size; ///< Size of the symbol or zero. - elf_word st_name; ///< Symbol name string index. - unsigned char st_info; ///< Symbol type and binding attributes. - unsigned char st_other; ///< Reserved for future use. - elf_half st_shndx; ///< Section to which this symbol applies. - - ELFSymbol(); - - /// Returns the binding attribute of the st_info member. - unsigned char getBinding() const { return st_info >> 4; } - - /// Returns the type attribute of the st_info member. - unsigned char getType() const { return st_info & 0x0F; } - - /// Sets the binding and type of the st_info member. - void setBindingAndType(unsigned char binding, unsigned char type) { - st_info = (binding << 4) + (type & 0x0F); - } - - static const char * - bindingToCString(unsigned char binding); - - static const char * - typeToCString(unsigned char type); - - static const char * - sectionIndexToCString(elf_half shndx, - const lldb_private::SectionList *section_list); - - /// Parse an ELFSymbol entry from the given DataExtractor starting at - /// position \p offset. The address size of the DataExtractor determines if - /// a 32 or 64 bit object is to be parsed. - /// - /// @param[in] data - /// The DataExtractor to read from. The address size of the extractor - /// determines if a 32 or 64 bit object should be read. - /// - /// @param[in,out] offset - /// Pointer to an offset in the data. On return the offset will be - /// advanced by the number of bytes read. - /// - /// @return - /// True if the ELFSymbol was successfully read and false otherwise. - bool - Parse(const lldb_private::DataExtractor &data, lldb::offset_t *offset); - - void - Dump (lldb_private::Stream *s, - uint32_t idx, - const lldb_private::DataExtractor *strtab_data, - const lldb_private::SectionList *section_list); +struct ELFSymbol { + elf_addr st_value; ///< Absolute or relocatable address. + elf_xword st_size; ///< Size of the symbol or zero. + elf_word st_name; ///< Symbol name string index. + unsigned char st_info; ///< Symbol type and binding attributes. + unsigned char st_other; ///< Reserved for future use. + elf_half st_shndx; ///< Section to which this symbol applies. + + ELFSymbol(); + + /// Returns the binding attribute of the st_info member. + unsigned char getBinding() const { return st_info >> 4; } + + /// Returns the type attribute of the st_info member. + unsigned char getType() const { return st_info & 0x0F; } + + /// Sets the binding and type of the st_info member. + void setBindingAndType(unsigned char binding, unsigned char type) { + st_info = (binding << 4) + (type & 0x0F); + } + + static const char *bindingToCString(unsigned char binding); + + static const char *typeToCString(unsigned char type); + + static const char * + sectionIndexToCString(elf_half shndx, + const lldb_private::SectionList *section_list); + + /// Parse an ELFSymbol entry from the given DataExtractor starting at + /// position \p offset. The address size of the DataExtractor determines if + /// a 32 or 64 bit object is to be parsed. + /// + /// @param[in] data + /// The DataExtractor to read from. The address size of the extractor + /// determines if a 32 or 64 bit object should be read. + /// + /// @param[in,out] offset + /// Pointer to an offset in the data. On return the offset will be + /// advanced by the number of bytes read. + /// + /// @return + /// True if the ELFSymbol was successfully read and false otherwise. + bool Parse(const lldb_private::DataExtractor &data, lldb::offset_t *offset); + + void Dump(lldb_private::Stream *s, uint32_t idx, + const lldb_private::DataExtractor *strtab_data, + const lldb_private::SectionList *section_list); }; //------------------------------------------------------------------------------ /// @class ELFDynamic /// @brief Represents an entry in an ELF dynamic table. -struct ELFDynamic -{ - elf_sxword d_tag; ///< Type of dynamic table entry. - union - { - elf_xword d_val; ///< Integer value of the table entry. - elf_addr d_ptr; ///< Pointer value of the table entry. - }; - - ELFDynamic(); - - /// Parse an ELFDynamic entry from the given DataExtractor starting at - /// position \p offset. The address size of the DataExtractor determines if - /// a 32 or 64 bit object is to be parsed. - /// - /// @param[in] data - /// The DataExtractor to read from. The address size of the extractor - /// determines if a 32 or 64 bit object should be read. - /// - /// @param[in,out] offset - /// Pointer to an offset in the data. On return the offset will be - /// advanced by the number of bytes read. - /// - /// @return - /// True if the ELFDynamic entry was successfully read and false - /// otherwise. - bool - Parse(const lldb_private::DataExtractor &data, lldb::offset_t *offset); +struct ELFDynamic { + elf_sxword d_tag; ///< Type of dynamic table entry. + union { + elf_xword d_val; ///< Integer value of the table entry. + elf_addr d_ptr; ///< Pointer value of the table entry. + }; + + ELFDynamic(); + + /// Parse an ELFDynamic entry from the given DataExtractor starting at + /// position \p offset. The address size of the DataExtractor determines if + /// a 32 or 64 bit object is to be parsed. + /// + /// @param[in] data + /// The DataExtractor to read from. The address size of the extractor + /// determines if a 32 or 64 bit object should be read. + /// + /// @param[in,out] offset + /// Pointer to an offset in the data. On return the offset will be + /// advanced by the number of bytes read. + /// + /// @return + /// True if the ELFDynamic entry was successfully read and false + /// otherwise. + bool Parse(const lldb_private::DataExtractor &data, lldb::offset_t *offset); }; //------------------------------------------------------------------------------ /// @class ELFRel /// @brief Represents a relocation entry with an implicit addend. -struct ELFRel -{ - elf_addr r_offset; ///< Address of reference. - elf_xword r_info; ///< symbol index and type of relocation. - - ELFRel(); - - /// Parse an ELFRel entry from the given DataExtractor starting at position - /// \p offset. The address size of the DataExtractor determines if a 32 or - /// 64 bit object is to be parsed. - /// - /// @param[in] data - /// The DataExtractor to read from. The address size of the extractor - /// determines if a 32 or 64 bit object should be read. - /// - /// @param[in,out] offset - /// Pointer to an offset in the data. On return the offset will be - /// advanced by the number of bytes read. - /// - /// @return - /// True if the ELFRel entry was successfully read and false otherwise. - bool - Parse(const lldb_private::DataExtractor &data, lldb::offset_t *offset); - - /// Returns the type when the given entry represents a 32-bit relocation. - static unsigned - RelocType32(const ELFRel &rel) - { - return rel.r_info & 0x0ff; - } - - /// Returns the type when the given entry represents a 64-bit relocation. - static unsigned - RelocType64(const ELFRel &rel) - { - return rel.r_info & 0xffffffff; - } - - /// Returns the symbol index when the given entry represents a 32-bit - /// relocation. - static unsigned - RelocSymbol32(const ELFRel &rel) - { - return rel.r_info >> 8; - } - - /// Returns the symbol index when the given entry represents a 64-bit - /// relocation. - static unsigned - RelocSymbol64(const ELFRel &rel) - { - return rel.r_info >> 32; - } +struct ELFRel { + elf_addr r_offset; ///< Address of reference. + elf_xword r_info; ///< symbol index and type of relocation. + + ELFRel(); + + /// Parse an ELFRel entry from the given DataExtractor starting at position + /// \p offset. The address size of the DataExtractor determines if a 32 or + /// 64 bit object is to be parsed. + /// + /// @param[in] data + /// The DataExtractor to read from. The address size of the extractor + /// determines if a 32 or 64 bit object should be read. + /// + /// @param[in,out] offset + /// Pointer to an offset in the data. On return the offset will be + /// advanced by the number of bytes read. + /// + /// @return + /// True if the ELFRel entry was successfully read and false otherwise. + bool Parse(const lldb_private::DataExtractor &data, lldb::offset_t *offset); + + /// Returns the type when the given entry represents a 32-bit relocation. + static unsigned RelocType32(const ELFRel &rel) { return rel.r_info & 0x0ff; } + + /// Returns the type when the given entry represents a 64-bit relocation. + static unsigned RelocType64(const ELFRel &rel) { + return rel.r_info & 0xffffffff; + } + + /// Returns the symbol index when the given entry represents a 32-bit + /// relocation. + static unsigned RelocSymbol32(const ELFRel &rel) { return rel.r_info >> 8; } + + /// Returns the symbol index when the given entry represents a 64-bit + /// relocation. + static unsigned RelocSymbol64(const ELFRel &rel) { return rel.r_info >> 32; } }; //------------------------------------------------------------------------------ /// @class ELFRela /// @brief Represents a relocation entry with an explicit addend. -struct ELFRela -{ - elf_addr r_offset; ///< Address of reference. - elf_xword r_info; ///< Symbol index and type of relocation. - elf_sxword r_addend; ///< Constant part of expression. - - ELFRela(); - - /// Parse an ELFRela entry from the given DataExtractor starting at position - /// \p offset. The address size of the DataExtractor determines if a 32 or - /// 64 bit object is to be parsed. - /// - /// @param[in] data - /// The DataExtractor to read from. The address size of the extractor - /// determines if a 32 or 64 bit object should be read. - /// - /// @param[in,out] offset - /// Pointer to an offset in the data. On return the offset will be - /// advanced by the number of bytes read. - /// - /// @return - /// True if the ELFRela entry was successfully read and false otherwise. - bool - Parse(const lldb_private::DataExtractor &data, lldb::offset_t *offset); - - /// Returns the type when the given entry represents a 32-bit relocation. - static unsigned - RelocType32(const ELFRela &rela) - { - return rela.r_info & 0x0ff; - } - - /// Returns the type when the given entry represents a 64-bit relocation. - static unsigned - RelocType64(const ELFRela &rela) - { - return rela.r_info & 0xffffffff; - } - - /// Returns the symbol index when the given entry represents a 32-bit - /// relocation. - static unsigned - RelocSymbol32(const ELFRela &rela) - { - return rela.r_info >> 8; - } - - /// Returns the symbol index when the given entry represents a 64-bit - /// relocation. - static unsigned - RelocSymbol64(const ELFRela &rela) - { - return rela.r_info >> 32; - } +struct ELFRela { + elf_addr r_offset; ///< Address of reference. + elf_xword r_info; ///< Symbol index and type of relocation. + elf_sxword r_addend; ///< Constant part of expression. + + ELFRela(); + + /// Parse an ELFRela entry from the given DataExtractor starting at position + /// \p offset. The address size of the DataExtractor determines if a 32 or + /// 64 bit object is to be parsed. + /// + /// @param[in] data + /// The DataExtractor to read from. The address size of the extractor + /// determines if a 32 or 64 bit object should be read. + /// + /// @param[in,out] offset + /// Pointer to an offset in the data. On return the offset will be + /// advanced by the number of bytes read. + /// + /// @return + /// True if the ELFRela entry was successfully read and false otherwise. + bool Parse(const lldb_private::DataExtractor &data, lldb::offset_t *offset); + + /// Returns the type when the given entry represents a 32-bit relocation. + static unsigned RelocType32(const ELFRela &rela) { + return rela.r_info & 0x0ff; + } + + /// Returns the type when the given entry represents a 64-bit relocation. + static unsigned RelocType64(const ELFRela &rela) { + return rela.r_info & 0xffffffff; + } + + /// Returns the symbol index when the given entry represents a 32-bit + /// relocation. + static unsigned RelocSymbol32(const ELFRela &rela) { + return rela.r_info >> 8; + } + + /// Returns the symbol index when the given entry represents a 64-bit + /// relocation. + static unsigned RelocSymbol64(const ELFRela &rela) { + return rela.r_info >> 32; + } }; } // End namespace elf. diff --git a/source/Plugins/ObjectFile/ELF/ObjectFileELF.cpp b/source/Plugins/ObjectFile/ELF/ObjectFileELF.cpp index 68e4e50a96e55..eb983154618bb 100644 --- a/source/Plugins/ObjectFile/ELF/ObjectFileELF.cpp +++ b/source/Plugins/ObjectFile/ELF/ObjectFileELF.cpp @@ -9,8 +9,8 @@ #include "ObjectFileELF.h" -#include <cassert> #include <algorithm> +#include <cassert> #include <unordered_map> #include "lldb/Core/ArchSpec.h" @@ -35,8 +35,10 @@ #include "llvm/Support/MathExtras.h" #include "llvm/Support/MipsABIFlags.h" -#define CASE_AND_STREAM(s, def, width) \ - case def: s->Printf("%-*s", width, #def); break; +#define CASE_AND_STREAM(s, def, width) \ + case def: \ + s->Printf("%-*s", width, #def); \ + break; using namespace lldb; using namespace lldb_private; @@ -47,64 +49,64 @@ namespace { // ELF note owner definitions const char *const LLDB_NT_OWNER_FREEBSD = "FreeBSD"; -const char *const LLDB_NT_OWNER_GNU = "GNU"; -const char *const LLDB_NT_OWNER_NETBSD = "NetBSD"; -const char *const LLDB_NT_OWNER_CSR = "csr"; +const char *const LLDB_NT_OWNER_GNU = "GNU"; +const char *const LLDB_NT_OWNER_NETBSD = "NetBSD"; +const char *const LLDB_NT_OWNER_CSR = "csr"; const char *const LLDB_NT_OWNER_ANDROID = "Android"; -const char *const LLDB_NT_OWNER_CORE = "CORE"; -const char *const LLDB_NT_OWNER_LINUX = "LINUX"; +const char *const LLDB_NT_OWNER_CORE = "CORE"; +const char *const LLDB_NT_OWNER_LINUX = "LINUX"; // ELF note type definitions -const elf_word LLDB_NT_FREEBSD_ABI_TAG = 0x01; +const elf_word LLDB_NT_FREEBSD_ABI_TAG = 0x01; const elf_word LLDB_NT_FREEBSD_ABI_SIZE = 4; -const elf_word LLDB_NT_GNU_ABI_TAG = 0x01; -const elf_word LLDB_NT_GNU_ABI_SIZE = 16; +const elf_word LLDB_NT_GNU_ABI_TAG = 0x01; +const elf_word LLDB_NT_GNU_ABI_SIZE = 16; const elf_word LLDB_NT_GNU_BUILD_ID_TAG = 0x03; -const elf_word LLDB_NT_NETBSD_ABI_TAG = 0x01; -const elf_word LLDB_NT_NETBSD_ABI_SIZE = 4; +const elf_word LLDB_NT_NETBSD_ABI_TAG = 0x01; +const elf_word LLDB_NT_NETBSD_ABI_SIZE = 4; // GNU ABI note OS constants -const elf_word LLDB_NT_GNU_ABI_OS_LINUX = 0x00; -const elf_word LLDB_NT_GNU_ABI_OS_HURD = 0x01; +const elf_word LLDB_NT_GNU_ABI_OS_LINUX = 0x00; +const elf_word LLDB_NT_GNU_ABI_OS_HURD = 0x01; const elf_word LLDB_NT_GNU_ABI_OS_SOLARIS = 0x02; // LLDB_NT_OWNER_CORE and LLDB_NT_OWNER_LINUX note contants -#define NT_PRSTATUS 1 -#define NT_PRFPREG 2 -#define NT_PRPSINFO 3 -#define NT_TASKSTRUCT 4 -#define NT_AUXV 6 -#define NT_SIGINFO 0x53494749 -#define NT_FILE 0x46494c45 -#define NT_PRXFPREG 0x46e62b7f -#define NT_PPC_VMX 0x100 -#define NT_PPC_SPE 0x101 -#define NT_PPC_VSX 0x102 -#define NT_386_TLS 0x200 -#define NT_386_IOPERM 0x201 -#define NT_X86_XSTATE 0x202 -#define NT_S390_HIGH_GPRS 0x300 -#define NT_S390_TIMER 0x301 -#define NT_S390_TODCMP 0x302 -#define NT_S390_TODPREG 0x303 -#define NT_S390_CTRS 0x304 -#define NT_S390_PREFIX 0x305 -#define NT_S390_LAST_BREAK 0x306 -#define NT_S390_SYSTEM_CALL 0x307 -#define NT_S390_TDB 0x308 -#define NT_S390_VXRS_LOW 0x309 -#define NT_S390_VXRS_HIGH 0x30a -#define NT_ARM_VFP 0x400 -#define NT_ARM_TLS 0x401 -#define NT_ARM_HW_BREAK 0x402 -#define NT_ARM_HW_WATCH 0x403 -#define NT_ARM_SYSTEM_CALL 0x404 -#define NT_METAG_CBUF 0x500 -#define NT_METAG_RPIPE 0x501 -#define NT_METAG_TLS 0x502 +#define NT_PRSTATUS 1 +#define NT_PRFPREG 2 +#define NT_PRPSINFO 3 +#define NT_TASKSTRUCT 4 +#define NT_AUXV 6 +#define NT_SIGINFO 0x53494749 +#define NT_FILE 0x46494c45 +#define NT_PRXFPREG 0x46e62b7f +#define NT_PPC_VMX 0x100 +#define NT_PPC_SPE 0x101 +#define NT_PPC_VSX 0x102 +#define NT_386_TLS 0x200 +#define NT_386_IOPERM 0x201 +#define NT_X86_XSTATE 0x202 +#define NT_S390_HIGH_GPRS 0x300 +#define NT_S390_TIMER 0x301 +#define NT_S390_TODCMP 0x302 +#define NT_S390_TODPREG 0x303 +#define NT_S390_CTRS 0x304 +#define NT_S390_PREFIX 0x305 +#define NT_S390_LAST_BREAK 0x306 +#define NT_S390_SYSTEM_CALL 0x307 +#define NT_S390_TDB 0x308 +#define NT_S390_VXRS_LOW 0x309 +#define NT_S390_VXRS_HIGH 0x30a +#define NT_ARM_VFP 0x400 +#define NT_ARM_TLS 0x401 +#define NT_ARM_HW_BREAK 0x402 +#define NT_ARM_HW_WATCH 0x403 +#define NT_ARM_SYSTEM_CALL 0x404 +#define NT_METAG_CBUF 0x500 +#define NT_METAG_RPIPE 0x501 +#define NT_METAG_TLS 0x502 //===----------------------------------------------------------------------===// /// @class ELFRelocation @@ -112,401 +114,335 @@ const elf_word LLDB_NT_GNU_ABI_OS_SOLARIS = 0x02; /// /// This helper class allows us to parse both ELFRel and ELFRela relocation /// entries in a generic manner. -class ELFRelocation -{ +class ELFRelocation { public: + /// Constructs an ELFRelocation entry with a personality as given by @p + /// type. + /// + /// @param type Either DT_REL or DT_RELA. Any other value is invalid. + ELFRelocation(unsigned type); - /// Constructs an ELFRelocation entry with a personality as given by @p - /// type. - /// - /// @param type Either DT_REL or DT_RELA. Any other value is invalid. - ELFRelocation(unsigned type); - - ~ELFRelocation(); + ~ELFRelocation(); - bool - Parse(const lldb_private::DataExtractor &data, lldb::offset_t *offset); + bool Parse(const lldb_private::DataExtractor &data, lldb::offset_t *offset); - static unsigned - RelocType32(const ELFRelocation &rel); + static unsigned RelocType32(const ELFRelocation &rel); - static unsigned - RelocType64(const ELFRelocation &rel); + static unsigned RelocType64(const ELFRelocation &rel); - static unsigned - RelocSymbol32(const ELFRelocation &rel); + static unsigned RelocSymbol32(const ELFRelocation &rel); - static unsigned - RelocSymbol64(const ELFRelocation &rel); + static unsigned RelocSymbol64(const ELFRelocation &rel); - static unsigned - RelocOffset32(const ELFRelocation &rel); + static unsigned RelocOffset32(const ELFRelocation &rel); - static unsigned - RelocOffset64(const ELFRelocation &rel); + static unsigned RelocOffset64(const ELFRelocation &rel); - static unsigned - RelocAddend32(const ELFRelocation &rel); + static unsigned RelocAddend32(const ELFRelocation &rel); - static unsigned - RelocAddend64(const ELFRelocation &rel); + static unsigned RelocAddend64(const ELFRelocation &rel); private: - typedef llvm::PointerUnion<ELFRel*, ELFRela*> RelocUnion; + typedef llvm::PointerUnion<ELFRel *, ELFRela *> RelocUnion; - RelocUnion reloc; + RelocUnion reloc; }; -ELFRelocation::ELFRelocation(unsigned type) -{ - if (type == DT_REL || type == SHT_REL) - reloc = new ELFRel(); - else if (type == DT_RELA || type == SHT_RELA) - reloc = new ELFRela(); - else { - assert(false && "unexpected relocation type"); - reloc = static_cast<ELFRel*>(NULL); - } +ELFRelocation::ELFRelocation(unsigned type) { + if (type == DT_REL || type == SHT_REL) + reloc = new ELFRel(); + else if (type == DT_RELA || type == SHT_RELA) + reloc = new ELFRela(); + else { + assert(false && "unexpected relocation type"); + reloc = static_cast<ELFRel *>(NULL); + } } -ELFRelocation::~ELFRelocation() -{ - if (reloc.is<ELFRel*>()) - delete reloc.get<ELFRel*>(); - else - delete reloc.get<ELFRela*>(); +ELFRelocation::~ELFRelocation() { + if (reloc.is<ELFRel *>()) + delete reloc.get<ELFRel *>(); + else + delete reloc.get<ELFRela *>(); } -bool -ELFRelocation::Parse(const lldb_private::DataExtractor &data, lldb::offset_t *offset) -{ - if (reloc.is<ELFRel*>()) - return reloc.get<ELFRel*>()->Parse(data, offset); - else - return reloc.get<ELFRela*>()->Parse(data, offset); +bool ELFRelocation::Parse(const lldb_private::DataExtractor &data, + lldb::offset_t *offset) { + if (reloc.is<ELFRel *>()) + return reloc.get<ELFRel *>()->Parse(data, offset); + else + return reloc.get<ELFRela *>()->Parse(data, offset); } -unsigned -ELFRelocation::RelocType32(const ELFRelocation &rel) -{ - if (rel.reloc.is<ELFRel*>()) - return ELFRel::RelocType32(*rel.reloc.get<ELFRel*>()); - else - return ELFRela::RelocType32(*rel.reloc.get<ELFRela*>()); +unsigned ELFRelocation::RelocType32(const ELFRelocation &rel) { + if (rel.reloc.is<ELFRel *>()) + return ELFRel::RelocType32(*rel.reloc.get<ELFRel *>()); + else + return ELFRela::RelocType32(*rel.reloc.get<ELFRela *>()); } -unsigned -ELFRelocation::RelocType64(const ELFRelocation &rel) -{ - if (rel.reloc.is<ELFRel*>()) - return ELFRel::RelocType64(*rel.reloc.get<ELFRel*>()); - else - return ELFRela::RelocType64(*rel.reloc.get<ELFRela*>()); +unsigned ELFRelocation::RelocType64(const ELFRelocation &rel) { + if (rel.reloc.is<ELFRel *>()) + return ELFRel::RelocType64(*rel.reloc.get<ELFRel *>()); + else + return ELFRela::RelocType64(*rel.reloc.get<ELFRela *>()); } -unsigned -ELFRelocation::RelocSymbol32(const ELFRelocation &rel) -{ - if (rel.reloc.is<ELFRel*>()) - return ELFRel::RelocSymbol32(*rel.reloc.get<ELFRel*>()); - else - return ELFRela::RelocSymbol32(*rel.reloc.get<ELFRela*>()); +unsigned ELFRelocation::RelocSymbol32(const ELFRelocation &rel) { + if (rel.reloc.is<ELFRel *>()) + return ELFRel::RelocSymbol32(*rel.reloc.get<ELFRel *>()); + else + return ELFRela::RelocSymbol32(*rel.reloc.get<ELFRela *>()); } -unsigned -ELFRelocation::RelocSymbol64(const ELFRelocation &rel) -{ - if (rel.reloc.is<ELFRel*>()) - return ELFRel::RelocSymbol64(*rel.reloc.get<ELFRel*>()); - else - return ELFRela::RelocSymbol64(*rel.reloc.get<ELFRela*>()); +unsigned ELFRelocation::RelocSymbol64(const ELFRelocation &rel) { + if (rel.reloc.is<ELFRel *>()) + return ELFRel::RelocSymbol64(*rel.reloc.get<ELFRel *>()); + else + return ELFRela::RelocSymbol64(*rel.reloc.get<ELFRela *>()); } -unsigned -ELFRelocation::RelocOffset32(const ELFRelocation &rel) -{ - if (rel.reloc.is<ELFRel*>()) - return rel.reloc.get<ELFRel*>()->r_offset; - else - return rel.reloc.get<ELFRela*>()->r_offset; +unsigned ELFRelocation::RelocOffset32(const ELFRelocation &rel) { + if (rel.reloc.is<ELFRel *>()) + return rel.reloc.get<ELFRel *>()->r_offset; + else + return rel.reloc.get<ELFRela *>()->r_offset; } -unsigned -ELFRelocation::RelocOffset64(const ELFRelocation &rel) -{ - if (rel.reloc.is<ELFRel*>()) - return rel.reloc.get<ELFRel*>()->r_offset; - else - return rel.reloc.get<ELFRela*>()->r_offset; +unsigned ELFRelocation::RelocOffset64(const ELFRelocation &rel) { + if (rel.reloc.is<ELFRel *>()) + return rel.reloc.get<ELFRel *>()->r_offset; + else + return rel.reloc.get<ELFRela *>()->r_offset; } -unsigned -ELFRelocation::RelocAddend32(const ELFRelocation &rel) -{ - if (rel.reloc.is<ELFRel*>()) - return 0; - else - return rel.reloc.get<ELFRela*>()->r_addend; +unsigned ELFRelocation::RelocAddend32(const ELFRelocation &rel) { + if (rel.reloc.is<ELFRel *>()) + return 0; + else + return rel.reloc.get<ELFRela *>()->r_addend; } -unsigned -ELFRelocation::RelocAddend64(const ELFRelocation &rel) -{ - if (rel.reloc.is<ELFRel*>()) - return 0; - else - return rel.reloc.get<ELFRela*>()->r_addend; +unsigned ELFRelocation::RelocAddend64(const ELFRelocation &rel) { + if (rel.reloc.is<ELFRel *>()) + return 0; + else + return rel.reloc.get<ELFRela *>()->r_addend; } } // end anonymous namespace -bool -ELFNote::Parse(const DataExtractor &data, lldb::offset_t *offset) -{ - // Read all fields. - if (data.GetU32(offset, &n_namesz, 3) == NULL) - return false; - - // The name field is required to be nul-terminated, and n_namesz - // includes the terminating nul in observed implementations (contrary - // to the ELF-64 spec). A special case is needed for cores generated - // by some older Linux versions, which write a note named "CORE" - // without a nul terminator and n_namesz = 4. - if (n_namesz == 4) - { - char buf[4]; - if (data.ExtractBytes (*offset, 4, data.GetByteOrder(), buf) != 4) - return false; - if (strncmp (buf, "CORE", 4) == 0) - { - n_name = "CORE"; - *offset += 4; - return true; - } - } - - const char *cstr = data.GetCStr(offset, llvm::alignTo (n_namesz, 4)); - if (cstr == NULL) - { - Log *log(lldb_private::GetLogIfAllCategoriesSet (LIBLLDB_LOG_SYMBOLS)); - if (log) - log->Printf("Failed to parse note name lacking nul terminator"); - - return false; - } - n_name = cstr; - return true; -} - -static uint32_t -kalimbaVariantFromElfFlags(const elf::elf_word e_flags) -{ - const uint32_t dsp_rev = e_flags & 0xFF; - uint32_t kal_arch_variant = LLDB_INVALID_CPUTYPE; - switch(dsp_rev) - { - // TODO(mg11) Support more variants - case 10: - kal_arch_variant = llvm::Triple::KalimbaSubArch_v3; - break; - case 14: - kal_arch_variant = llvm::Triple::KalimbaSubArch_v4; - break; - case 17: - case 20: - kal_arch_variant = llvm::Triple::KalimbaSubArch_v5; - break; - default: - break; - } - return kal_arch_variant; -} - -static uint32_t -mipsVariantFromElfFlags(const elf::elf_word e_flags, uint32_t endian) -{ - const uint32_t mips_arch = e_flags & llvm::ELF::EF_MIPS_ARCH; - uint32_t arch_variant = ArchSpec::eMIPSSubType_unknown; - - switch (mips_arch) - { - case llvm::ELF::EF_MIPS_ARCH_1: - case llvm::ELF::EF_MIPS_ARCH_2: - case llvm::ELF::EF_MIPS_ARCH_32: - return (endian == ELFDATA2LSB) ? ArchSpec::eMIPSSubType_mips32el : ArchSpec::eMIPSSubType_mips32; - case llvm::ELF::EF_MIPS_ARCH_32R2: - return (endian == ELFDATA2LSB) ? ArchSpec::eMIPSSubType_mips32r2el : ArchSpec::eMIPSSubType_mips32r2; - case llvm::ELF::EF_MIPS_ARCH_32R6: - return (endian == ELFDATA2LSB) ? ArchSpec::eMIPSSubType_mips32r6el : ArchSpec::eMIPSSubType_mips32r6; - case llvm::ELF::EF_MIPS_ARCH_3: - case llvm::ELF::EF_MIPS_ARCH_4: - case llvm::ELF::EF_MIPS_ARCH_5: - case llvm::ELF::EF_MIPS_ARCH_64: - return (endian == ELFDATA2LSB) ? ArchSpec::eMIPSSubType_mips64el : ArchSpec::eMIPSSubType_mips64; - case llvm::ELF::EF_MIPS_ARCH_64R2: - return (endian == ELFDATA2LSB) ? ArchSpec::eMIPSSubType_mips64r2el : ArchSpec::eMIPSSubType_mips64r2; - case llvm::ELF::EF_MIPS_ARCH_64R6: - return (endian == ELFDATA2LSB) ? ArchSpec::eMIPSSubType_mips64r6el : ArchSpec::eMIPSSubType_mips64r6; - default: - break; - } - - return arch_variant; -} +bool ELFNote::Parse(const DataExtractor &data, lldb::offset_t *offset) { + // Read all fields. + if (data.GetU32(offset, &n_namesz, 3) == NULL) + return false; -static uint32_t -subTypeFromElfHeader(const elf::ELFHeader& header) -{ - if (header.e_machine == llvm::ELF::EM_MIPS) - return mipsVariantFromElfFlags (header.e_flags, - header.e_ident[EI_DATA]); + // The name field is required to be nul-terminated, and n_namesz + // includes the terminating nul in observed implementations (contrary + // to the ELF-64 spec). A special case is needed for cores generated + // by some older Linux versions, which write a note named "CORE" + // without a nul terminator and n_namesz = 4. + if (n_namesz == 4) { + char buf[4]; + if (data.ExtractBytes(*offset, 4, data.GetByteOrder(), buf) != 4) + return false; + if (strncmp(buf, "CORE", 4) == 0) { + n_name = "CORE"; + *offset += 4; + return true; + } + } + + const char *cstr = data.GetCStr(offset, llvm::alignTo(n_namesz, 4)); + if (cstr == NULL) { + Log *log(lldb_private::GetLogIfAllCategoriesSet(LIBLLDB_LOG_SYMBOLS)); + if (log) + log->Printf("Failed to parse note name lacking nul terminator"); - return - llvm::ELF::EM_CSR_KALIMBA == header.e_machine ? - kalimbaVariantFromElfFlags(header.e_flags) : - LLDB_INVALID_CPUTYPE; + return false; + } + n_name = cstr; + return true; +} + +static uint32_t kalimbaVariantFromElfFlags(const elf::elf_word e_flags) { + const uint32_t dsp_rev = e_flags & 0xFF; + uint32_t kal_arch_variant = LLDB_INVALID_CPUTYPE; + switch (dsp_rev) { + // TODO(mg11) Support more variants + case 10: + kal_arch_variant = llvm::Triple::KalimbaSubArch_v3; + break; + case 14: + kal_arch_variant = llvm::Triple::KalimbaSubArch_v4; + break; + case 17: + case 20: + kal_arch_variant = llvm::Triple::KalimbaSubArch_v5; + break; + default: + break; + } + return kal_arch_variant; +} + +static uint32_t mipsVariantFromElfFlags(const elf::elf_word e_flags, + uint32_t endian) { + const uint32_t mips_arch = e_flags & llvm::ELF::EF_MIPS_ARCH; + uint32_t arch_variant = ArchSpec::eMIPSSubType_unknown; + + switch (mips_arch) { + case llvm::ELF::EF_MIPS_ARCH_1: + case llvm::ELF::EF_MIPS_ARCH_2: + case llvm::ELF::EF_MIPS_ARCH_32: + return (endian == ELFDATA2LSB) ? ArchSpec::eMIPSSubType_mips32el + : ArchSpec::eMIPSSubType_mips32; + case llvm::ELF::EF_MIPS_ARCH_32R2: + return (endian == ELFDATA2LSB) ? ArchSpec::eMIPSSubType_mips32r2el + : ArchSpec::eMIPSSubType_mips32r2; + case llvm::ELF::EF_MIPS_ARCH_32R6: + return (endian == ELFDATA2LSB) ? ArchSpec::eMIPSSubType_mips32r6el + : ArchSpec::eMIPSSubType_mips32r6; + case llvm::ELF::EF_MIPS_ARCH_3: + case llvm::ELF::EF_MIPS_ARCH_4: + case llvm::ELF::EF_MIPS_ARCH_5: + case llvm::ELF::EF_MIPS_ARCH_64: + return (endian == ELFDATA2LSB) ? ArchSpec::eMIPSSubType_mips64el + : ArchSpec::eMIPSSubType_mips64; + case llvm::ELF::EF_MIPS_ARCH_64R2: + return (endian == ELFDATA2LSB) ? ArchSpec::eMIPSSubType_mips64r2el + : ArchSpec::eMIPSSubType_mips64r2; + case llvm::ELF::EF_MIPS_ARCH_64R6: + return (endian == ELFDATA2LSB) ? ArchSpec::eMIPSSubType_mips64r6el + : ArchSpec::eMIPSSubType_mips64r6; + default: + break; + } + + return arch_variant; +} + +static uint32_t subTypeFromElfHeader(const elf::ELFHeader &header) { + if (header.e_machine == llvm::ELF::EM_MIPS) + return mipsVariantFromElfFlags(header.e_flags, header.e_ident[EI_DATA]); + + return llvm::ELF::EM_CSR_KALIMBA == header.e_machine + ? kalimbaVariantFromElfFlags(header.e_flags) + : LLDB_INVALID_CPUTYPE; } //! The kalimba toolchain identifies a code section as being //! one with the SHT_PROGBITS set in the section sh_type and the top //! bit in the 32-bit address field set. static lldb::SectionType -kalimbaSectionType( - const elf::ELFHeader& header, - const elf::ELFSectionHeader& sect_hdr) -{ - if (llvm::ELF::EM_CSR_KALIMBA != header.e_machine) - { - return eSectionTypeOther; - } +kalimbaSectionType(const elf::ELFHeader &header, + const elf::ELFSectionHeader §_hdr) { + if (llvm::ELF::EM_CSR_KALIMBA != header.e_machine) { + return eSectionTypeOther; + } - if (llvm::ELF::SHT_NOBITS == sect_hdr.sh_type) - { - return eSectionTypeZeroFill; - } + if (llvm::ELF::SHT_NOBITS == sect_hdr.sh_type) { + return eSectionTypeZeroFill; + } - if (llvm::ELF::SHT_PROGBITS == sect_hdr.sh_type) - { - const lldb::addr_t KAL_CODE_BIT = 1 << 31; - return KAL_CODE_BIT & sect_hdr.sh_addr ? - eSectionTypeCode : eSectionTypeData; - } + if (llvm::ELF::SHT_PROGBITS == sect_hdr.sh_type) { + const lldb::addr_t KAL_CODE_BIT = 1 << 31; + return KAL_CODE_BIT & sect_hdr.sh_addr ? eSectionTypeCode + : eSectionTypeData; + } - return eSectionTypeOther; + return eSectionTypeOther; } // Arbitrary constant used as UUID prefix for core files. -const uint32_t -ObjectFileELF::g_core_uuid_magic(0xE210C); +const uint32_t ObjectFileELF::g_core_uuid_magic(0xE210C); //------------------------------------------------------------------ // Static methods. //------------------------------------------------------------------ -void -ObjectFileELF::Initialize() -{ - PluginManager::RegisterPlugin(GetPluginNameStatic(), - GetPluginDescriptionStatic(), - CreateInstance, - CreateMemoryInstance, - GetModuleSpecifications); -} - -void -ObjectFileELF::Terminate() -{ - PluginManager::UnregisterPlugin(CreateInstance); -} - -lldb_private::ConstString -ObjectFileELF::GetPluginNameStatic() -{ - static ConstString g_name("elf"); - return g_name; -} - -const char * -ObjectFileELF::GetPluginDescriptionStatic() -{ - return "ELF object file reader."; -} - -ObjectFile * -ObjectFileELF::CreateInstance (const lldb::ModuleSP &module_sp, - DataBufferSP &data_sp, - lldb::offset_t data_offset, - const lldb_private::FileSpec* file, - lldb::offset_t file_offset, - lldb::offset_t length) -{ - if (!data_sp) - { - data_sp = file->MemoryMapFileContentsIfLocal(file_offset, length); - data_offset = 0; - } +void ObjectFileELF::Initialize() { + PluginManager::RegisterPlugin(GetPluginNameStatic(), + GetPluginDescriptionStatic(), CreateInstance, + CreateMemoryInstance, GetModuleSpecifications); +} - if (data_sp && data_sp->GetByteSize() > (llvm::ELF::EI_NIDENT + data_offset)) - { - const uint8_t *magic = data_sp->GetBytes() + data_offset; - if (ELFHeader::MagicBytesMatch(magic)) - { - // Update the data to contain the entire file if it doesn't already - if (data_sp->GetByteSize() < length) { - data_sp = file->MemoryMapFileContentsIfLocal(file_offset, length); - data_offset = 0; - magic = data_sp->GetBytes(); - } - unsigned address_size = ELFHeader::AddressSizeInBytes(magic); - if (address_size == 4 || address_size == 8) - { - std::unique_ptr<ObjectFileELF> objfile_ap(new ObjectFileELF(module_sp, data_sp, data_offset, file, file_offset, length)); - ArchSpec spec; - if (objfile_ap->GetArchitecture(spec) && - objfile_ap->SetModulesArchitecture(spec)) - return objfile_ap.release(); - } - } - } - return NULL; +void ObjectFileELF::Terminate() { + PluginManager::UnregisterPlugin(CreateInstance); } +lldb_private::ConstString ObjectFileELF::GetPluginNameStatic() { + static ConstString g_name("elf"); + return g_name; +} -ObjectFile* -ObjectFileELF::CreateMemoryInstance (const lldb::ModuleSP &module_sp, - DataBufferSP& data_sp, - const lldb::ProcessSP &process_sp, - lldb::addr_t header_addr) -{ - if (data_sp && data_sp->GetByteSize() > (llvm::ELF::EI_NIDENT)) - { - const uint8_t *magic = data_sp->GetBytes(); - if (ELFHeader::MagicBytesMatch(magic)) - { - unsigned address_size = ELFHeader::AddressSizeInBytes(magic); - if (address_size == 4 || address_size == 8) - { - std::auto_ptr<ObjectFileELF> objfile_ap(new ObjectFileELF(module_sp, data_sp, process_sp, header_addr)); - ArchSpec spec; - if (objfile_ap->GetArchitecture(spec) && - objfile_ap->SetModulesArchitecture(spec)) - return objfile_ap.release(); - } - } - } - return NULL; +const char *ObjectFileELF::GetPluginDescriptionStatic() { + return "ELF object file reader."; } -bool -ObjectFileELF::MagicBytesMatch (DataBufferSP& data_sp, - lldb::addr_t data_offset, - lldb::addr_t data_length) -{ - if (data_sp && data_sp->GetByteSize() > (llvm::ELF::EI_NIDENT + data_offset)) - { - const uint8_t *magic = data_sp->GetBytes() + data_offset; - return ELFHeader::MagicBytesMatch(magic); - } - return false; +ObjectFile *ObjectFileELF::CreateInstance(const lldb::ModuleSP &module_sp, + DataBufferSP &data_sp, + lldb::offset_t data_offset, + const lldb_private::FileSpec *file, + lldb::offset_t file_offset, + lldb::offset_t length) { + if (!data_sp) { + data_sp = file->MemoryMapFileContentsIfLocal(file_offset, length); + data_offset = 0; + } + + if (data_sp && + data_sp->GetByteSize() > (llvm::ELF::EI_NIDENT + data_offset)) { + const uint8_t *magic = data_sp->GetBytes() + data_offset; + if (ELFHeader::MagicBytesMatch(magic)) { + // Update the data to contain the entire file if it doesn't already + if (data_sp->GetByteSize() < length) { + data_sp = file->MemoryMapFileContentsIfLocal(file_offset, length); + data_offset = 0; + magic = data_sp->GetBytes(); + } + unsigned address_size = ELFHeader::AddressSizeInBytes(magic); + if (address_size == 4 || address_size == 8) { + std::unique_ptr<ObjectFileELF> objfile_ap(new ObjectFileELF( + module_sp, data_sp, data_offset, file, file_offset, length)); + ArchSpec spec; + if (objfile_ap->GetArchitecture(spec) && + objfile_ap->SetModulesArchitecture(spec)) + return objfile_ap.release(); + } + } + } + return NULL; +} + +ObjectFile *ObjectFileELF::CreateMemoryInstance( + const lldb::ModuleSP &module_sp, DataBufferSP &data_sp, + const lldb::ProcessSP &process_sp, lldb::addr_t header_addr) { + if (data_sp && data_sp->GetByteSize() > (llvm::ELF::EI_NIDENT)) { + const uint8_t *magic = data_sp->GetBytes(); + if (ELFHeader::MagicBytesMatch(magic)) { + unsigned address_size = ELFHeader::AddressSizeInBytes(magic); + if (address_size == 4 || address_size == 8) { + std::auto_ptr<ObjectFileELF> objfile_ap( + new ObjectFileELF(module_sp, data_sp, process_sp, header_addr)); + ArchSpec spec; + if (objfile_ap->GetArchitecture(spec) && + objfile_ap->SetModulesArchitecture(spec)) + return objfile_ap.release(); + } + } + } + return NULL; +} + +bool ObjectFileELF::MagicBytesMatch(DataBufferSP &data_sp, + lldb::addr_t data_offset, + lldb::addr_t data_length) { + if (data_sp && + data_sp->GetByteSize() > (llvm::ELF::EI_NIDENT + data_offset)) { + const uint8_t *magic = data_sp->GetBytes() + data_offset; + return ELFHeader::MagicBytesMatch(magic); + } + return false; } /* @@ -515,130 +451,117 @@ ObjectFileELF::MagicBytesMatch (DataBufferSP& data_sp, * COPYRIGHT (C) 1986 Gary S. Brown. You may use this program, or * code or tables extracted from it, as desired without restriction. */ -static uint32_t -calc_crc32(uint32_t crc, const void *buf, size_t size) -{ - static const uint32_t g_crc32_tab[] = - { - 0x00000000, 0x77073096, 0xee0e612c, 0x990951ba, 0x076dc419, 0x706af48f, - 0xe963a535, 0x9e6495a3, 0x0edb8832, 0x79dcb8a4, 0xe0d5e91e, 0x97d2d988, - 0x09b64c2b, 0x7eb17cbd, 0xe7b82d07, 0x90bf1d91, 0x1db71064, 0x6ab020f2, - 0xf3b97148, 0x84be41de, 0x1adad47d, 0x6ddde4eb, 0xf4d4b551, 0x83d385c7, - 0x136c9856, 0x646ba8c0, 0xfd62f97a, 0x8a65c9ec, 0x14015c4f, 0x63066cd9, - 0xfa0f3d63, 0x8d080df5, 0x3b6e20c8, 0x4c69105e, 0xd56041e4, 0xa2677172, - 0x3c03e4d1, 0x4b04d447, 0xd20d85fd, 0xa50ab56b, 0x35b5a8fa, 0x42b2986c, - 0xdbbbc9d6, 0xacbcf940, 0x32d86ce3, 0x45df5c75, 0xdcd60dcf, 0xabd13d59, - 0x26d930ac, 0x51de003a, 0xc8d75180, 0xbfd06116, 0x21b4f4b5, 0x56b3c423, - 0xcfba9599, 0xb8bda50f, 0x2802b89e, 0x5f058808, 0xc60cd9b2, 0xb10be924, - 0x2f6f7c87, 0x58684c11, 0xc1611dab, 0xb6662d3d, 0x76dc4190, 0x01db7106, - 0x98d220bc, 0xefd5102a, 0x71b18589, 0x06b6b51f, 0x9fbfe4a5, 0xe8b8d433, - 0x7807c9a2, 0x0f00f934, 0x9609a88e, 0xe10e9818, 0x7f6a0dbb, 0x086d3d2d, - 0x91646c97, 0xe6635c01, 0x6b6b51f4, 0x1c6c6162, 0x856530d8, 0xf262004e, - 0x6c0695ed, 0x1b01a57b, 0x8208f4c1, 0xf50fc457, 0x65b0d9c6, 0x12b7e950, - 0x8bbeb8ea, 0xfcb9887c, 0x62dd1ddf, 0x15da2d49, 0x8cd37cf3, 0xfbd44c65, - 0x4db26158, 0x3ab551ce, 0xa3bc0074, 0xd4bb30e2, 0x4adfa541, 0x3dd895d7, - 0xa4d1c46d, 0xd3d6f4fb, 0x4369e96a, 0x346ed9fc, 0xad678846, 0xda60b8d0, - 0x44042d73, 0x33031de5, 0xaa0a4c5f, 0xdd0d7cc9, 0x5005713c, 0x270241aa, - 0xbe0b1010, 0xc90c2086, 0x5768b525, 0x206f85b3, 0xb966d409, 0xce61e49f, - 0x5edef90e, 0x29d9c998, 0xb0d09822, 0xc7d7a8b4, 0x59b33d17, 0x2eb40d81, - 0xb7bd5c3b, 0xc0ba6cad, 0xedb88320, 0x9abfb3b6, 0x03b6e20c, 0x74b1d29a, - 0xead54739, 0x9dd277af, 0x04db2615, 0x73dc1683, 0xe3630b12, 0x94643b84, - 0x0d6d6a3e, 0x7a6a5aa8, 0xe40ecf0b, 0x9309ff9d, 0x0a00ae27, 0x7d079eb1, - 0xf00f9344, 0x8708a3d2, 0x1e01f268, 0x6906c2fe, 0xf762575d, 0x806567cb, - 0x196c3671, 0x6e6b06e7, 0xfed41b76, 0x89d32be0, 0x10da7a5a, 0x67dd4acc, - 0xf9b9df6f, 0x8ebeeff9, 0x17b7be43, 0x60b08ed5, 0xd6d6a3e8, 0xa1d1937e, - 0x38d8c2c4, 0x4fdff252, 0xd1bb67f1, 0xa6bc5767, 0x3fb506dd, 0x48b2364b, - 0xd80d2bda, 0xaf0a1b4c, 0x36034af6, 0x41047a60, 0xdf60efc3, 0xa867df55, - 0x316e8eef, 0x4669be79, 0xcb61b38c, 0xbc66831a, 0x256fd2a0, 0x5268e236, - 0xcc0c7795, 0xbb0b4703, 0x220216b9, 0x5505262f, 0xc5ba3bbe, 0xb2bd0b28, - 0x2bb45a92, 0x5cb36a04, 0xc2d7ffa7, 0xb5d0cf31, 0x2cd99e8b, 0x5bdeae1d, - 0x9b64c2b0, 0xec63f226, 0x756aa39c, 0x026d930a, 0x9c0906a9, 0xeb0e363f, - 0x72076785, 0x05005713, 0x95bf4a82, 0xe2b87a14, 0x7bb12bae, 0x0cb61b38, - 0x92d28e9b, 0xe5d5be0d, 0x7cdcefb7, 0x0bdbdf21, 0x86d3d2d4, 0xf1d4e242, - 0x68ddb3f8, 0x1fda836e, 0x81be16cd, 0xf6b9265b, 0x6fb077e1, 0x18b74777, - 0x88085ae6, 0xff0f6a70, 0x66063bca, 0x11010b5c, 0x8f659eff, 0xf862ae69, - 0x616bffd3, 0x166ccf45, 0xa00ae278, 0xd70dd2ee, 0x4e048354, 0x3903b3c2, - 0xa7672661, 0xd06016f7, 0x4969474d, 0x3e6e77db, 0xaed16a4a, 0xd9d65adc, - 0x40df0b66, 0x37d83bf0, 0xa9bcae53, 0xdebb9ec5, 0x47b2cf7f, 0x30b5ffe9, - 0xbdbdf21c, 0xcabac28a, 0x53b39330, 0x24b4a3a6, 0xbad03605, 0xcdd70693, - 0x54de5729, 0x23d967bf, 0xb3667a2e, 0xc4614ab8, 0x5d681b02, 0x2a6f2b94, - 0xb40bbe37, 0xc30c8ea1, 0x5a05df1b, 0x2d02ef8d - }; - const uint8_t *p = (const uint8_t *)buf; - - crc = crc ^ ~0U; - while (size--) - crc = g_crc32_tab[(crc ^ *p++) & 0xFF] ^ (crc >> 8); - return crc ^ ~0U; -} - -static uint32_t -calc_gnu_debuglink_crc32(const void *buf, size_t size) -{ - return calc_crc32(0U, buf, size); -} - -uint32_t -ObjectFileELF::CalculateELFNotesSegmentsCRC32 (const ProgramHeaderColl& program_headers, - DataExtractor& object_data) -{ - typedef ProgramHeaderCollConstIter Iter; - - uint32_t core_notes_crc = 0; - - for (Iter I = program_headers.begin(); I != program_headers.end(); ++I) - { - if (I->p_type == llvm::ELF::PT_NOTE) - { - const elf_off ph_offset = I->p_offset; - const size_t ph_size = I->p_filesz; - - DataExtractor segment_data; - if (segment_data.SetData(object_data, ph_offset, ph_size) != ph_size) - { - // The ELF program header contained incorrect data, - // probably corefile is incomplete or corrupted. - break; - } - - core_notes_crc = calc_crc32(core_notes_crc, - segment_data.GetDataStart(), - segment_data.GetByteSize()); - } - } - - return core_notes_crc; -} - -static const char* -OSABIAsCString (unsigned char osabi_byte) -{ -#define _MAKE_OSABI_CASE(x) case x: return #x - switch (osabi_byte) - { - _MAKE_OSABI_CASE(ELFOSABI_NONE); - _MAKE_OSABI_CASE(ELFOSABI_HPUX); - _MAKE_OSABI_CASE(ELFOSABI_NETBSD); - _MAKE_OSABI_CASE(ELFOSABI_GNU); - _MAKE_OSABI_CASE(ELFOSABI_HURD); - _MAKE_OSABI_CASE(ELFOSABI_SOLARIS); - _MAKE_OSABI_CASE(ELFOSABI_AIX); - _MAKE_OSABI_CASE(ELFOSABI_IRIX); - _MAKE_OSABI_CASE(ELFOSABI_FREEBSD); - _MAKE_OSABI_CASE(ELFOSABI_TRU64); - _MAKE_OSABI_CASE(ELFOSABI_MODESTO); - _MAKE_OSABI_CASE(ELFOSABI_OPENBSD); - _MAKE_OSABI_CASE(ELFOSABI_OPENVMS); - _MAKE_OSABI_CASE(ELFOSABI_NSK); - _MAKE_OSABI_CASE(ELFOSABI_AROS); - _MAKE_OSABI_CASE(ELFOSABI_FENIXOS); - _MAKE_OSABI_CASE(ELFOSABI_C6000_ELFABI); - _MAKE_OSABI_CASE(ELFOSABI_C6000_LINUX); - _MAKE_OSABI_CASE(ELFOSABI_ARM); - _MAKE_OSABI_CASE(ELFOSABI_STANDALONE); - default: - return "<unknown-osabi>"; - } +static uint32_t calc_crc32(uint32_t crc, const void *buf, size_t size) { + static const uint32_t g_crc32_tab[] = { + 0x00000000, 0x77073096, 0xee0e612c, 0x990951ba, 0x076dc419, 0x706af48f, + 0xe963a535, 0x9e6495a3, 0x0edb8832, 0x79dcb8a4, 0xe0d5e91e, 0x97d2d988, + 0x09b64c2b, 0x7eb17cbd, 0xe7b82d07, 0x90bf1d91, 0x1db71064, 0x6ab020f2, + 0xf3b97148, 0x84be41de, 0x1adad47d, 0x6ddde4eb, 0xf4d4b551, 0x83d385c7, + 0x136c9856, 0x646ba8c0, 0xfd62f97a, 0x8a65c9ec, 0x14015c4f, 0x63066cd9, + 0xfa0f3d63, 0x8d080df5, 0x3b6e20c8, 0x4c69105e, 0xd56041e4, 0xa2677172, + 0x3c03e4d1, 0x4b04d447, 0xd20d85fd, 0xa50ab56b, 0x35b5a8fa, 0x42b2986c, + 0xdbbbc9d6, 0xacbcf940, 0x32d86ce3, 0x45df5c75, 0xdcd60dcf, 0xabd13d59, + 0x26d930ac, 0x51de003a, 0xc8d75180, 0xbfd06116, 0x21b4f4b5, 0x56b3c423, + 0xcfba9599, 0xb8bda50f, 0x2802b89e, 0x5f058808, 0xc60cd9b2, 0xb10be924, + 0x2f6f7c87, 0x58684c11, 0xc1611dab, 0xb6662d3d, 0x76dc4190, 0x01db7106, + 0x98d220bc, 0xefd5102a, 0x71b18589, 0x06b6b51f, 0x9fbfe4a5, 0xe8b8d433, + 0x7807c9a2, 0x0f00f934, 0x9609a88e, 0xe10e9818, 0x7f6a0dbb, 0x086d3d2d, + 0x91646c97, 0xe6635c01, 0x6b6b51f4, 0x1c6c6162, 0x856530d8, 0xf262004e, + 0x6c0695ed, 0x1b01a57b, 0x8208f4c1, 0xf50fc457, 0x65b0d9c6, 0x12b7e950, + 0x8bbeb8ea, 0xfcb9887c, 0x62dd1ddf, 0x15da2d49, 0x8cd37cf3, 0xfbd44c65, + 0x4db26158, 0x3ab551ce, 0xa3bc0074, 0xd4bb30e2, 0x4adfa541, 0x3dd895d7, + 0xa4d1c46d, 0xd3d6f4fb, 0x4369e96a, 0x346ed9fc, 0xad678846, 0xda60b8d0, + 0x44042d73, 0x33031de5, 0xaa0a4c5f, 0xdd0d7cc9, 0x5005713c, 0x270241aa, + 0xbe0b1010, 0xc90c2086, 0x5768b525, 0x206f85b3, 0xb966d409, 0xce61e49f, + 0x5edef90e, 0x29d9c998, 0xb0d09822, 0xc7d7a8b4, 0x59b33d17, 0x2eb40d81, + 0xb7bd5c3b, 0xc0ba6cad, 0xedb88320, 0x9abfb3b6, 0x03b6e20c, 0x74b1d29a, + 0xead54739, 0x9dd277af, 0x04db2615, 0x73dc1683, 0xe3630b12, 0x94643b84, + 0x0d6d6a3e, 0x7a6a5aa8, 0xe40ecf0b, 0x9309ff9d, 0x0a00ae27, 0x7d079eb1, + 0xf00f9344, 0x8708a3d2, 0x1e01f268, 0x6906c2fe, 0xf762575d, 0x806567cb, + 0x196c3671, 0x6e6b06e7, 0xfed41b76, 0x89d32be0, 0x10da7a5a, 0x67dd4acc, + 0xf9b9df6f, 0x8ebeeff9, 0x17b7be43, 0x60b08ed5, 0xd6d6a3e8, 0xa1d1937e, + 0x38d8c2c4, 0x4fdff252, 0xd1bb67f1, 0xa6bc5767, 0x3fb506dd, 0x48b2364b, + 0xd80d2bda, 0xaf0a1b4c, 0x36034af6, 0x41047a60, 0xdf60efc3, 0xa867df55, + 0x316e8eef, 0x4669be79, 0xcb61b38c, 0xbc66831a, 0x256fd2a0, 0x5268e236, + 0xcc0c7795, 0xbb0b4703, 0x220216b9, 0x5505262f, 0xc5ba3bbe, 0xb2bd0b28, + 0x2bb45a92, 0x5cb36a04, 0xc2d7ffa7, 0xb5d0cf31, 0x2cd99e8b, 0x5bdeae1d, + 0x9b64c2b0, 0xec63f226, 0x756aa39c, 0x026d930a, 0x9c0906a9, 0xeb0e363f, + 0x72076785, 0x05005713, 0x95bf4a82, 0xe2b87a14, 0x7bb12bae, 0x0cb61b38, + 0x92d28e9b, 0xe5d5be0d, 0x7cdcefb7, 0x0bdbdf21, 0x86d3d2d4, 0xf1d4e242, + 0x68ddb3f8, 0x1fda836e, 0x81be16cd, 0xf6b9265b, 0x6fb077e1, 0x18b74777, + 0x88085ae6, 0xff0f6a70, 0x66063bca, 0x11010b5c, 0x8f659eff, 0xf862ae69, + 0x616bffd3, 0x166ccf45, 0xa00ae278, 0xd70dd2ee, 0x4e048354, 0x3903b3c2, + 0xa7672661, 0xd06016f7, 0x4969474d, 0x3e6e77db, 0xaed16a4a, 0xd9d65adc, + 0x40df0b66, 0x37d83bf0, 0xa9bcae53, 0xdebb9ec5, 0x47b2cf7f, 0x30b5ffe9, + 0xbdbdf21c, 0xcabac28a, 0x53b39330, 0x24b4a3a6, 0xbad03605, 0xcdd70693, + 0x54de5729, 0x23d967bf, 0xb3667a2e, 0xc4614ab8, 0x5d681b02, 0x2a6f2b94, + 0xb40bbe37, 0xc30c8ea1, 0x5a05df1b, 0x2d02ef8d}; + const uint8_t *p = (const uint8_t *)buf; + + crc = crc ^ ~0U; + while (size--) + crc = g_crc32_tab[(crc ^ *p++) & 0xFF] ^ (crc >> 8); + return crc ^ ~0U; +} + +static uint32_t calc_gnu_debuglink_crc32(const void *buf, size_t size) { + return calc_crc32(0U, buf, size); +} + +uint32_t ObjectFileELF::CalculateELFNotesSegmentsCRC32( + const ProgramHeaderColl &program_headers, DataExtractor &object_data) { + typedef ProgramHeaderCollConstIter Iter; + + uint32_t core_notes_crc = 0; + + for (Iter I = program_headers.begin(); I != program_headers.end(); ++I) { + if (I->p_type == llvm::ELF::PT_NOTE) { + const elf_off ph_offset = I->p_offset; + const size_t ph_size = I->p_filesz; + + DataExtractor segment_data; + if (segment_data.SetData(object_data, ph_offset, ph_size) != ph_size) { + // The ELF program header contained incorrect data, + // probably corefile is incomplete or corrupted. + break; + } + + core_notes_crc = calc_crc32(core_notes_crc, segment_data.GetDataStart(), + segment_data.GetByteSize()); + } + } + + return core_notes_crc; +} + +static const char *OSABIAsCString(unsigned char osabi_byte) { +#define _MAKE_OSABI_CASE(x) \ + case x: \ + return #x + switch (osabi_byte) { + _MAKE_OSABI_CASE(ELFOSABI_NONE); + _MAKE_OSABI_CASE(ELFOSABI_HPUX); + _MAKE_OSABI_CASE(ELFOSABI_NETBSD); + _MAKE_OSABI_CASE(ELFOSABI_GNU); + _MAKE_OSABI_CASE(ELFOSABI_HURD); + _MAKE_OSABI_CASE(ELFOSABI_SOLARIS); + _MAKE_OSABI_CASE(ELFOSABI_AIX); + _MAKE_OSABI_CASE(ELFOSABI_IRIX); + _MAKE_OSABI_CASE(ELFOSABI_FREEBSD); + _MAKE_OSABI_CASE(ELFOSABI_TRU64); + _MAKE_OSABI_CASE(ELFOSABI_MODESTO); + _MAKE_OSABI_CASE(ELFOSABI_OPENBSD); + _MAKE_OSABI_CASE(ELFOSABI_OPENVMS); + _MAKE_OSABI_CASE(ELFOSABI_NSK); + _MAKE_OSABI_CASE(ELFOSABI_AROS); + _MAKE_OSABI_CASE(ELFOSABI_FENIXOS); + _MAKE_OSABI_CASE(ELFOSABI_C6000_ELFABI); + _MAKE_OSABI_CASE(ELFOSABI_C6000_LINUX); + _MAKE_OSABI_CASE(ELFOSABI_ARM); + _MAKE_OSABI_CASE(ELFOSABI_STANDALONE); + default: + return "<unknown-osabi>"; + } #undef _MAKE_OSABI_CASE } @@ -648,2480 +571,2427 @@ OSABIAsCString (unsigned char osabi_byte) // This function is only being kept to validate the move. // // TODO : Remove this function -static bool -GetOsFromOSABI (unsigned char osabi_byte, llvm::Triple::OSType &ostype) -{ - switch (osabi_byte) - { - case ELFOSABI_AIX: ostype = llvm::Triple::OSType::AIX; break; - case ELFOSABI_FREEBSD: ostype = llvm::Triple::OSType::FreeBSD; break; - case ELFOSABI_GNU: ostype = llvm::Triple::OSType::Linux; break; - case ELFOSABI_NETBSD: ostype = llvm::Triple::OSType::NetBSD; break; - case ELFOSABI_OPENBSD: ostype = llvm::Triple::OSType::OpenBSD; break; - case ELFOSABI_SOLARIS: ostype = llvm::Triple::OSType::Solaris; break; - default: - ostype = llvm::Triple::OSType::UnknownOS; - } - return ostype != llvm::Triple::OSType::UnknownOS; -} - -size_t -ObjectFileELF::GetModuleSpecifications (const lldb_private::FileSpec& file, - lldb::DataBufferSP& data_sp, - lldb::offset_t data_offset, - lldb::offset_t file_offset, - lldb::offset_t length, - lldb_private::ModuleSpecList &specs) -{ - Log *log(lldb_private::GetLogIfAllCategoriesSet (LIBLLDB_LOG_MODULES)); - - const size_t initial_count = specs.GetSize(); - - if (ObjectFileELF::MagicBytesMatch(data_sp, 0, data_sp->GetByteSize())) - { - DataExtractor data; - data.SetData(data_sp); - elf::ELFHeader header; - if (header.Parse(data, &data_offset)) - { - if (data_sp) - { - ModuleSpec spec (file); - - const uint32_t sub_type = subTypeFromElfHeader(header); - spec.GetArchitecture().SetArchitecture(eArchTypeELF, - header.e_machine, - sub_type, - header.e_ident[EI_OSABI]); - - if (spec.GetArchitecture().IsValid()) - { - llvm::Triple::OSType ostype; - llvm::Triple::VendorType vendor; - llvm::Triple::OSType spec_ostype = spec.GetArchitecture ().GetTriple ().getOS (); - - if (log) - log->Printf ("ObjectFileELF::%s file '%s' module OSABI: %s", __FUNCTION__, file.GetPath ().c_str (), OSABIAsCString (header.e_ident[EI_OSABI])); - - // SetArchitecture should have set the vendor to unknown - vendor = spec.GetArchitecture ().GetTriple ().getVendor (); - assert(vendor == llvm::Triple::UnknownVendor); - - // - // Validate it is ok to remove GetOsFromOSABI - GetOsFromOSABI (header.e_ident[EI_OSABI], ostype); - assert(spec_ostype == ostype); - if (spec_ostype != llvm::Triple::OSType::UnknownOS) - { - if (log) - log->Printf ("ObjectFileELF::%s file '%s' set ELF module OS type from ELF header OSABI.", __FUNCTION__, file.GetPath ().c_str ()); - } - - // Try to get the UUID from the section list. Usually that's at the end, so - // map the file in if we don't have it already. - size_t section_header_end = header.e_shoff + header.e_shnum * header.e_shentsize; - if (section_header_end > data_sp->GetByteSize()) - { - data_sp = file.MemoryMapFileContentsIfLocal (file_offset, section_header_end); - data.SetData(data_sp); - } - - uint32_t gnu_debuglink_crc = 0; - std::string gnu_debuglink_file; - SectionHeaderColl section_headers; - lldb_private::UUID &uuid = spec.GetUUID(); - - using namespace std::placeholders; - const SetDataFunction set_data = std::bind(&ObjectFileELF::SetData, std::cref(data), _1, _2, _3); - GetSectionHeaderInfo(section_headers, set_data, header, uuid, gnu_debuglink_file, gnu_debuglink_crc, spec.GetArchitecture ()); - - - llvm::Triple &spec_triple = spec.GetArchitecture ().GetTriple (); - - if (log) - log->Printf ("ObjectFileELF::%s file '%s' module set to triple: %s (architecture %s)", __FUNCTION__, file.GetPath ().c_str (), spec_triple.getTriple ().c_str (), spec.GetArchitecture ().GetArchitectureName ()); - - if (!uuid.IsValid()) - { - uint32_t core_notes_crc = 0; - - if (!gnu_debuglink_crc) - { - lldb_private::Timer scoped_timer (__PRETTY_FUNCTION__, - "Calculating module crc32 %s with size %" PRIu64 " KiB", - file.GetLastPathComponent().AsCString(), - (file.GetByteSize()-file_offset)/1024); - - // For core files - which usually don't happen to have a gnu_debuglink, - // and are pretty bulky - calculating whole contents crc32 would be too much of luxury. - // Thus we will need to fallback to something simpler. - if (header.e_type == llvm::ELF::ET_CORE) - { - size_t program_headers_end = header.e_phoff + header.e_phnum * header.e_phentsize; - if (program_headers_end > data_sp->GetByteSize()) - { - data_sp = file.MemoryMapFileContentsIfLocal(file_offset, program_headers_end); - data.SetData(data_sp); - } - ProgramHeaderColl program_headers; - GetProgramHeaderInfo(program_headers, set_data, header); - - size_t segment_data_end = 0; - for (ProgramHeaderCollConstIter I = program_headers.begin(); - I != program_headers.end(); ++I) - { - segment_data_end = std::max<unsigned long long> (I->p_offset + I->p_filesz, segment_data_end); - } - - if (segment_data_end > data_sp->GetByteSize()) - { - data_sp = file.MemoryMapFileContentsIfLocal(file_offset, segment_data_end); - data.SetData(data_sp); - } - - core_notes_crc = CalculateELFNotesSegmentsCRC32 (program_headers, data); - } - else - { - // Need to map entire file into memory to calculate the crc. - data_sp = file.MemoryMapFileContentsIfLocal (file_offset, SIZE_MAX); - data.SetData(data_sp); - gnu_debuglink_crc = calc_gnu_debuglink_crc32 (data.GetDataStart(), data.GetByteSize()); - } - } - if (gnu_debuglink_crc) - { - // Use 4 bytes of crc from the .gnu_debuglink section. - uint32_t uuidt[4] = { gnu_debuglink_crc, 0, 0, 0 }; - uuid.SetBytes (uuidt, sizeof(uuidt)); - } - else if (core_notes_crc) - { - // Use 8 bytes - first 4 bytes for *magic* prefix, mainly to make it look different form - // .gnu_debuglink crc followed by 4 bytes of note segments crc. - uint32_t uuidt[4] = { g_core_uuid_magic, core_notes_crc, 0, 0 }; - uuid.SetBytes (uuidt, sizeof(uuidt)); - } - } - - specs.Append(spec); +static bool GetOsFromOSABI(unsigned char osabi_byte, + llvm::Triple::OSType &ostype) { + switch (osabi_byte) { + case ELFOSABI_AIX: + ostype = llvm::Triple::OSType::AIX; + break; + case ELFOSABI_FREEBSD: + ostype = llvm::Triple::OSType::FreeBSD; + break; + case ELFOSABI_GNU: + ostype = llvm::Triple::OSType::Linux; + break; + case ELFOSABI_NETBSD: + ostype = llvm::Triple::OSType::NetBSD; + break; + case ELFOSABI_OPENBSD: + ostype = llvm::Triple::OSType::OpenBSD; + break; + case ELFOSABI_SOLARIS: + ostype = llvm::Triple::OSType::Solaris; + break; + default: + ostype = llvm::Triple::OSType::UnknownOS; + } + return ostype != llvm::Triple::OSType::UnknownOS; +} + +size_t ObjectFileELF::GetModuleSpecifications( + const lldb_private::FileSpec &file, lldb::DataBufferSP &data_sp, + lldb::offset_t data_offset, lldb::offset_t file_offset, + lldb::offset_t length, lldb_private::ModuleSpecList &specs) { + Log *log(lldb_private::GetLogIfAllCategoriesSet(LIBLLDB_LOG_MODULES)); + + const size_t initial_count = specs.GetSize(); + + if (ObjectFileELF::MagicBytesMatch(data_sp, 0, data_sp->GetByteSize())) { + DataExtractor data; + data.SetData(data_sp); + elf::ELFHeader header; + if (header.Parse(data, &data_offset)) { + if (data_sp) { + ModuleSpec spec(file); + + const uint32_t sub_type = subTypeFromElfHeader(header); + spec.GetArchitecture().SetArchitecture( + eArchTypeELF, header.e_machine, sub_type, header.e_ident[EI_OSABI]); + + if (spec.GetArchitecture().IsValid()) { + llvm::Triple::OSType ostype; + llvm::Triple::VendorType vendor; + llvm::Triple::OSType spec_ostype = + spec.GetArchitecture().GetTriple().getOS(); + + if (log) + log->Printf("ObjectFileELF::%s file '%s' module OSABI: %s", + __FUNCTION__, file.GetPath().c_str(), + OSABIAsCString(header.e_ident[EI_OSABI])); + + // SetArchitecture should have set the vendor to unknown + vendor = spec.GetArchitecture().GetTriple().getVendor(); + assert(vendor == llvm::Triple::UnknownVendor); + + // + // Validate it is ok to remove GetOsFromOSABI + GetOsFromOSABI(header.e_ident[EI_OSABI], ostype); + assert(spec_ostype == ostype); + if (spec_ostype != llvm::Triple::OSType::UnknownOS) { + if (log) + log->Printf("ObjectFileELF::%s file '%s' set ELF module OS type " + "from ELF header OSABI.", + __FUNCTION__, file.GetPath().c_str()); + } + + // Try to get the UUID from the section list. Usually that's at the + // end, so + // map the file in if we don't have it already. + size_t section_header_end = + header.e_shoff + header.e_shnum * header.e_shentsize; + if (section_header_end > data_sp->GetByteSize()) { + data_sp = file.MemoryMapFileContentsIfLocal(file_offset, + section_header_end); + data.SetData(data_sp); + } + + uint32_t gnu_debuglink_crc = 0; + std::string gnu_debuglink_file; + SectionHeaderColl section_headers; + lldb_private::UUID &uuid = spec.GetUUID(); + + using namespace std::placeholders; + const SetDataFunction set_data = + std::bind(&ObjectFileELF::SetData, std::cref(data), _1, _2, _3); + GetSectionHeaderInfo(section_headers, set_data, header, uuid, + gnu_debuglink_file, gnu_debuglink_crc, + spec.GetArchitecture()); + + llvm::Triple &spec_triple = spec.GetArchitecture().GetTriple(); + + if (log) + log->Printf("ObjectFileELF::%s file '%s' module set to triple: %s " + "(architecture %s)", + __FUNCTION__, file.GetPath().c_str(), + spec_triple.getTriple().c_str(), + spec.GetArchitecture().GetArchitectureName()); + + if (!uuid.IsValid()) { + uint32_t core_notes_crc = 0; + + if (!gnu_debuglink_crc) { + lldb_private::Timer scoped_timer( + LLVM_PRETTY_FUNCTION, + "Calculating module crc32 %s with size %" PRIu64 " KiB", + file.GetLastPathComponent().AsCString(), + (file.GetByteSize() - file_offset) / 1024); + + // For core files - which usually don't happen to have a + // gnu_debuglink, + // and are pretty bulky - calculating whole contents crc32 would + // be too much of luxury. + // Thus we will need to fallback to something simpler. + if (header.e_type == llvm::ELF::ET_CORE) { + size_t program_headers_end = + header.e_phoff + header.e_phnum * header.e_phentsize; + if (program_headers_end > data_sp->GetByteSize()) { + data_sp = file.MemoryMapFileContentsIfLocal( + file_offset, program_headers_end); + data.SetData(data_sp); + } + ProgramHeaderColl program_headers; + GetProgramHeaderInfo(program_headers, set_data, header); + + size_t segment_data_end = 0; + for (ProgramHeaderCollConstIter I = program_headers.begin(); + I != program_headers.end(); ++I) { + segment_data_end = std::max<unsigned long long>( + I->p_offset + I->p_filesz, segment_data_end); } + + if (segment_data_end > data_sp->GetByteSize()) { + data_sp = file.MemoryMapFileContentsIfLocal(file_offset, + segment_data_end); + data.SetData(data_sp); + } + + core_notes_crc = + CalculateELFNotesSegmentsCRC32(program_headers, data); + } else { + // Need to map entire file into memory to calculate the crc. + data_sp = + file.MemoryMapFileContentsIfLocal(file_offset, SIZE_MAX); + data.SetData(data_sp); + gnu_debuglink_crc = calc_gnu_debuglink_crc32( + data.GetDataStart(), data.GetByteSize()); + } } + if (gnu_debuglink_crc) { + // Use 4 bytes of crc from the .gnu_debuglink section. + uint32_t uuidt[4] = {gnu_debuglink_crc, 0, 0, 0}; + uuid.SetBytes(uuidt, sizeof(uuidt)); + } else if (core_notes_crc) { + // Use 8 bytes - first 4 bytes for *magic* prefix, mainly to make + // it look different form + // .gnu_debuglink crc followed by 4 bytes of note segments crc. + uint32_t uuidt[4] = {g_core_uuid_magic, core_notes_crc, 0, 0}; + uuid.SetBytes(uuidt, sizeof(uuidt)); + } + } + + specs.Append(spec); } + } } + } - return specs.GetSize() - initial_count; + return specs.GetSize() - initial_count; } //------------------------------------------------------------------ // PluginInterface protocol //------------------------------------------------------------------ -lldb_private::ConstString -ObjectFileELF::GetPluginName() -{ - return GetPluginNameStatic(); +lldb_private::ConstString ObjectFileELF::GetPluginName() { + return GetPluginNameStatic(); } -uint32_t -ObjectFileELF::GetPluginVersion() -{ - return m_plugin_version; -} +uint32_t ObjectFileELF::GetPluginVersion() { return m_plugin_version; } //------------------------------------------------------------------ // ObjectFile protocol //------------------------------------------------------------------ -ObjectFileELF::ObjectFileELF (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) : - ObjectFile(module_sp, file, file_offset, length, data_sp, data_offset), - m_header(), - m_uuid(), - m_gnu_debuglink_file(), - m_gnu_debuglink_crc(0), - m_program_headers(), - m_section_headers(), - m_dynamic_symbols(), - m_filespec_ap(), - m_entry_point_address(), - m_arch_spec() -{ - if (file) - m_file = *file; - ::memset(&m_header, 0, sizeof(m_header)); -} - -ObjectFileELF::ObjectFileELF (const lldb::ModuleSP &module_sp, - DataBufferSP& header_data_sp, - const lldb::ProcessSP &process_sp, - addr_t header_addr) : - ObjectFile(module_sp, process_sp, header_addr, header_data_sp), - m_header(), - m_uuid(), - m_gnu_debuglink_file(), - m_gnu_debuglink_crc(0), - m_program_headers(), - m_section_headers(), - m_dynamic_symbols(), - m_filespec_ap(), - m_entry_point_address(), - m_arch_spec() -{ - ::memset(&m_header, 0, sizeof(m_header)); -} - -ObjectFileELF::~ObjectFileELF() -{ -} - -bool -ObjectFileELF::IsExecutable() const -{ - return ((m_header.e_type & ET_EXEC) != 0) || (m_header.e_entry != 0); -} - -bool -ObjectFileELF::SetLoadAddress (Target &target, - lldb::addr_t value, - bool value_is_offset) -{ - ModuleSP module_sp = GetModule(); - if (module_sp) - { - size_t num_loaded_sections = 0; - SectionList *section_list = GetSectionList (); - if (section_list) - { - if (!value_is_offset) - { - bool found_offset = false; - for (size_t i = 0, count = GetProgramHeaderCount(); i < count; ++i) - { - const elf::ELFProgramHeader* header = GetProgramHeaderByIndex(i); - if (header == nullptr) - continue; - - if (header->p_type != PT_LOAD || header->p_offset != 0) - continue; - - value = value - header->p_vaddr; - found_offset = true; - break; - } - if (!found_offset) - return false; - } +ObjectFileELF::ObjectFileELF(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) + : ObjectFile(module_sp, file, file_offset, length, data_sp, data_offset), + m_header(), m_uuid(), m_gnu_debuglink_file(), m_gnu_debuglink_crc(0), + m_program_headers(), m_section_headers(), m_dynamic_symbols(), + m_filespec_ap(), m_entry_point_address(), m_arch_spec() { + if (file) + m_file = *file; + ::memset(&m_header, 0, sizeof(m_header)); +} + +ObjectFileELF::ObjectFileELF(const lldb::ModuleSP &module_sp, + DataBufferSP &header_data_sp, + const lldb::ProcessSP &process_sp, + addr_t header_addr) + : ObjectFile(module_sp, process_sp, header_addr, header_data_sp), + m_header(), m_uuid(), m_gnu_debuglink_file(), m_gnu_debuglink_crc(0), + m_program_headers(), m_section_headers(), m_dynamic_symbols(), + m_filespec_ap(), m_entry_point_address(), m_arch_spec() { + ::memset(&m_header, 0, sizeof(m_header)); +} + +ObjectFileELF::~ObjectFileELF() {} + +bool ObjectFileELF::IsExecutable() const { + return ((m_header.e_type & ET_EXEC) != 0) || (m_header.e_entry != 0); +} + +bool ObjectFileELF::SetLoadAddress(Target &target, lldb::addr_t value, + bool value_is_offset) { + ModuleSP module_sp = GetModule(); + if (module_sp) { + size_t num_loaded_sections = 0; + SectionList *section_list = GetSectionList(); + if (section_list) { + if (!value_is_offset) { + bool found_offset = false; + for (size_t i = 0, count = GetProgramHeaderCount(); i < count; ++i) { + const elf::ELFProgramHeader *header = GetProgramHeaderByIndex(i); + if (header == nullptr) + continue; - const size_t num_sections = section_list->GetSize(); - size_t sect_idx = 0; - - for (sect_idx = 0; sect_idx < num_sections; ++sect_idx) - { - // Iterate through the object file sections to find all - // of the sections that have SHF_ALLOC in their flag bits. - SectionSP section_sp (section_list->GetSectionAtIndex (sect_idx)); - if (section_sp && section_sp->Test(SHF_ALLOC)) - { - lldb::addr_t load_addr = section_sp->GetFileAddress(); - // We don't want to update the load address of a section with type - // eSectionTypeAbsoluteAddress as they already have the absolute load address - // already specified - if (section_sp->GetType() != eSectionTypeAbsoluteAddress) - load_addr += value; - - // On 32-bit systems the load address have to fit into 4 bytes. The rest of - // the bytes are the overflow from the addition. - if (GetAddressByteSize() == 4) - load_addr &= 0xFFFFFFFF; - - if (target.GetSectionLoadList().SetSectionLoadAddress (section_sp, load_addr)) - ++num_loaded_sections; - } - } - return num_loaded_sections > 0; + if (header->p_type != PT_LOAD || header->p_offset != 0) + continue; + + value = value - header->p_vaddr; + found_offset = true; + break; + } + if (!found_offset) + return false; + } + + const size_t num_sections = section_list->GetSize(); + size_t sect_idx = 0; + + for (sect_idx = 0; sect_idx < num_sections; ++sect_idx) { + // Iterate through the object file sections to find all + // of the sections that have SHF_ALLOC in their flag bits. + SectionSP section_sp(section_list->GetSectionAtIndex(sect_idx)); + if (section_sp && section_sp->Test(SHF_ALLOC)) { + lldb::addr_t load_addr = section_sp->GetFileAddress(); + // We don't want to update the load address of a section with type + // eSectionTypeAbsoluteAddress as they already have the absolute load + // address + // already specified + if (section_sp->GetType() != eSectionTypeAbsoluteAddress) + load_addr += value; + + // On 32-bit systems the load address have to fit into 4 bytes. The + // rest of + // the bytes are the overflow from the addition. + if (GetAddressByteSize() == 4) + load_addr &= 0xFFFFFFFF; + + if (target.GetSectionLoadList().SetSectionLoadAddress(section_sp, + load_addr)) + ++num_loaded_sections; } + } + return num_loaded_sections > 0; } - return false; + } + return false; } -ByteOrder -ObjectFileELF::GetByteOrder() const -{ - if (m_header.e_ident[EI_DATA] == ELFDATA2MSB) - return eByteOrderBig; - if (m_header.e_ident[EI_DATA] == ELFDATA2LSB) - return eByteOrderLittle; - return eByteOrderInvalid; -} - -uint32_t -ObjectFileELF::GetAddressByteSize() const -{ - return m_data.GetAddressByteSize(); -} - -AddressClass -ObjectFileELF::GetAddressClass (addr_t file_addr) -{ - Symtab* symtab = GetSymtab(); - if (!symtab) - return eAddressClassUnknown; - - // The address class is determined based on the symtab. Ask it from the object file what - // contains the symtab information. - ObjectFile* symtab_objfile = symtab->GetObjectFile(); - if (symtab_objfile != nullptr && symtab_objfile != this) - return symtab_objfile->GetAddressClass(file_addr); - - auto res = ObjectFile::GetAddressClass (file_addr); - if (res != eAddressClassCode) - return res; - - auto ub = m_address_class_map.upper_bound(file_addr); - if (ub == m_address_class_map.begin()) - { - // No entry in the address class map before the address. Return - // default address class for an address in a code section. - return eAddressClassCode; - } +ByteOrder ObjectFileELF::GetByteOrder() const { + if (m_header.e_ident[EI_DATA] == ELFDATA2MSB) + return eByteOrderBig; + if (m_header.e_ident[EI_DATA] == ELFDATA2LSB) + return eByteOrderLittle; + return eByteOrderInvalid; +} + +uint32_t ObjectFileELF::GetAddressByteSize() const { + return m_data.GetAddressByteSize(); +} + +AddressClass ObjectFileELF::GetAddressClass(addr_t file_addr) { + Symtab *symtab = GetSymtab(); + if (!symtab) + return eAddressClassUnknown; + + // The address class is determined based on the symtab. Ask it from the object + // file what + // contains the symtab information. + ObjectFile *symtab_objfile = symtab->GetObjectFile(); + if (symtab_objfile != nullptr && symtab_objfile != this) + return symtab_objfile->GetAddressClass(file_addr); + + auto res = ObjectFile::GetAddressClass(file_addr); + if (res != eAddressClassCode) + return res; + + auto ub = m_address_class_map.upper_bound(file_addr); + if (ub == m_address_class_map.begin()) { + // No entry in the address class map before the address. Return + // default address class for an address in a code section. + return eAddressClassCode; + } - // Move iterator to the address class entry preceding address - --ub; + // Move iterator to the address class entry preceding address + --ub; - return ub->second; + return ub->second; } -size_t -ObjectFileELF::SectionIndex(const SectionHeaderCollIter &I) -{ - return std::distance(m_section_headers.begin(), I) + 1u; +size_t ObjectFileELF::SectionIndex(const SectionHeaderCollIter &I) { + return std::distance(m_section_headers.begin(), I) + 1u; } -size_t -ObjectFileELF::SectionIndex(const SectionHeaderCollConstIter &I) const -{ - return std::distance(m_section_headers.begin(), I) + 1u; +size_t ObjectFileELF::SectionIndex(const SectionHeaderCollConstIter &I) const { + return std::distance(m_section_headers.begin(), I) + 1u; } -bool -ObjectFileELF::ParseHeader() -{ - lldb::offset_t offset = 0; - if (!m_header.Parse(m_data, &offset)) - return false; - - if (!IsInMemory()) - return true; - - // For in memory object files m_data might not contain the full object file. Try to load it - // until the end of the "Section header table" what is at the end of the ELF file. - addr_t file_size = m_header.e_shoff + m_header.e_shnum * m_header.e_shentsize; - if (m_data.GetByteSize() < file_size) - { - ProcessSP process_sp (m_process_wp.lock()); - if (!process_sp) - return false; - - DataBufferSP data_sp = ReadMemory(process_sp, m_memory_addr, file_size); - if (!data_sp) - return false; - m_data.SetData(data_sp, 0, file_size); - } +bool ObjectFileELF::ParseHeader() { + lldb::offset_t offset = 0; + if (!m_header.Parse(m_data, &offset)) + return false; + if (!IsInMemory()) return true; + + // For in memory object files m_data might not contain the full object file. + // Try to load it + // until the end of the "Section header table" what is at the end of the ELF + // file. + addr_t file_size = m_header.e_shoff + m_header.e_shnum * m_header.e_shentsize; + if (m_data.GetByteSize() < file_size) { + ProcessSP process_sp(m_process_wp.lock()); + if (!process_sp) + return false; + + DataBufferSP data_sp = ReadMemory(process_sp, m_memory_addr, file_size); + if (!data_sp) + return false; + m_data.SetData(data_sp, 0, file_size); + } + + return true; } -bool -ObjectFileELF::GetUUID(lldb_private::UUID* uuid) -{ - // Need to parse the section list to get the UUIDs, so make sure that's been done. - if (!ParseSectionHeaders() && GetType() != ObjectFile::eTypeCoreFile) - return false; +bool ObjectFileELF::GetUUID(lldb_private::UUID *uuid) { + // Need to parse the section list to get the UUIDs, so make sure that's been + // done. + if (!ParseSectionHeaders() && GetType() != ObjectFile::eTypeCoreFile) + return false; - if (m_uuid.IsValid()) - { - // We have the full build id uuid. - *uuid = m_uuid; - return true; - } - else if (GetType() == ObjectFile::eTypeCoreFile) - { - uint32_t core_notes_crc = 0; - - if (!ParseProgramHeaders()) - return false; - - core_notes_crc = CalculateELFNotesSegmentsCRC32(m_program_headers, m_data); - - if (core_notes_crc) - { - // Use 8 bytes - first 4 bytes for *magic* prefix, mainly to make it - // look different form .gnu_debuglink crc - followed by 4 bytes of note - // segments crc. - uint32_t uuidt[4] = { g_core_uuid_magic, core_notes_crc, 0, 0 }; - m_uuid.SetBytes (uuidt, sizeof(uuidt)); - } - } - else - { - if (!m_gnu_debuglink_crc) - m_gnu_debuglink_crc = calc_gnu_debuglink_crc32 (m_data.GetDataStart(), m_data.GetByteSize()); - if (m_gnu_debuglink_crc) - { - // Use 4 bytes of crc from the .gnu_debuglink section. - uint32_t uuidt[4] = { m_gnu_debuglink_crc, 0, 0, 0 }; - m_uuid.SetBytes (uuidt, sizeof(uuidt)); - } - } + if (m_uuid.IsValid()) { + // We have the full build id uuid. + *uuid = m_uuid; + return true; + } else if (GetType() == ObjectFile::eTypeCoreFile) { + uint32_t core_notes_crc = 0; - if (m_uuid.IsValid()) - { - *uuid = m_uuid; - return true; - } + if (!ParseProgramHeaders()) + return false; + + core_notes_crc = CalculateELFNotesSegmentsCRC32(m_program_headers, m_data); + + if (core_notes_crc) { + // Use 8 bytes - first 4 bytes for *magic* prefix, mainly to make it + // look different form .gnu_debuglink crc - followed by 4 bytes of note + // segments crc. + uint32_t uuidt[4] = {g_core_uuid_magic, core_notes_crc, 0, 0}; + m_uuid.SetBytes(uuidt, sizeof(uuidt)); + } + } else { + if (!m_gnu_debuglink_crc) + m_gnu_debuglink_crc = + calc_gnu_debuglink_crc32(m_data.GetDataStart(), m_data.GetByteSize()); + if (m_gnu_debuglink_crc) { + // Use 4 bytes of crc from the .gnu_debuglink section. + uint32_t uuidt[4] = {m_gnu_debuglink_crc, 0, 0, 0}; + m_uuid.SetBytes(uuidt, sizeof(uuidt)); + } + } + + if (m_uuid.IsValid()) { + *uuid = m_uuid; + return true; + } - return false; + return false; } -lldb_private::FileSpecList -ObjectFileELF::GetDebugSymbolFilePaths() -{ - FileSpecList file_spec_list; +lldb_private::FileSpecList ObjectFileELF::GetDebugSymbolFilePaths() { + FileSpecList file_spec_list; - if (!m_gnu_debuglink_file.empty()) - { - FileSpec file_spec (m_gnu_debuglink_file.c_str(), false); - file_spec_list.Append (file_spec); - } - return file_spec_list; + if (!m_gnu_debuglink_file.empty()) { + FileSpec file_spec(m_gnu_debuglink_file, false); + file_spec_list.Append(file_spec); + } + return file_spec_list; } -uint32_t -ObjectFileELF::GetDependentModules(FileSpecList &files) -{ - size_t num_modules = ParseDependentModules(); - uint32_t num_specs = 0; +uint32_t ObjectFileELF::GetDependentModules(FileSpecList &files) { + size_t num_modules = ParseDependentModules(); + uint32_t num_specs = 0; - for (unsigned i = 0; i < num_modules; ++i) - { - if (files.AppendIfUnique(m_filespec_ap->GetFileSpecAtIndex(i))) - num_specs++; - } + for (unsigned i = 0; i < num_modules; ++i) { + if (files.AppendIfUnique(m_filespec_ap->GetFileSpecAtIndex(i))) + num_specs++; + } - return num_specs; + return num_specs; } -Address -ObjectFileELF::GetImageInfoAddress(Target *target) -{ - if (!ParseDynamicSymbols()) - return Address(); +Address ObjectFileELF::GetImageInfoAddress(Target *target) { + if (!ParseDynamicSymbols()) + return Address(); - SectionList *section_list = GetSectionList(); - if (!section_list) - return Address(); + SectionList *section_list = GetSectionList(); + if (!section_list) + return Address(); - // Find the SHT_DYNAMIC (.dynamic) section. - SectionSP dynsym_section_sp (section_list->FindSectionByType (eSectionTypeELFDynamicLinkInfo, true)); - if (!dynsym_section_sp) - return Address(); - assert (dynsym_section_sp->GetObjectFile() == this); + // Find the SHT_DYNAMIC (.dynamic) section. + SectionSP dynsym_section_sp( + section_list->FindSectionByType(eSectionTypeELFDynamicLinkInfo, true)); + if (!dynsym_section_sp) + return Address(); + assert(dynsym_section_sp->GetObjectFile() == this); - user_id_t dynsym_id = dynsym_section_sp->GetID(); - const ELFSectionHeaderInfo *dynsym_hdr = GetSectionHeaderByIndex(dynsym_id); - if (!dynsym_hdr) - return Address(); + user_id_t dynsym_id = dynsym_section_sp->GetID(); + const ELFSectionHeaderInfo *dynsym_hdr = GetSectionHeaderByIndex(dynsym_id); + if (!dynsym_hdr) + return Address(); - for (size_t i = 0; i < m_dynamic_symbols.size(); ++i) - { - ELFDynamic &symbol = m_dynamic_symbols[i]; + for (size_t i = 0; i < m_dynamic_symbols.size(); ++i) { + ELFDynamic &symbol = m_dynamic_symbols[i]; + + if (symbol.d_tag == DT_DEBUG) { + // Compute the offset as the number of previous entries plus the + // size of d_tag. + addr_t offset = i * dynsym_hdr->sh_entsize + GetAddressByteSize(); + return Address(dynsym_section_sp, offset); + } + // MIPS executables uses DT_MIPS_RLD_MAP_REL to support PIE. DT_MIPS_RLD_MAP + // exists in non-PIE. + else if ((symbol.d_tag == DT_MIPS_RLD_MAP || + symbol.d_tag == DT_MIPS_RLD_MAP_REL) && + target) { + addr_t offset = i * dynsym_hdr->sh_entsize + GetAddressByteSize(); + addr_t dyn_base = dynsym_section_sp->GetLoadBaseAddress(target); + if (dyn_base == LLDB_INVALID_ADDRESS) + return Address(); - if (symbol.d_tag == DT_DEBUG) - { - // Compute the offset as the number of previous entries plus the - // size of d_tag. - addr_t offset = i * dynsym_hdr->sh_entsize + GetAddressByteSize(); - return Address(dynsym_section_sp, offset); - } - // MIPS executables uses DT_MIPS_RLD_MAP_REL to support PIE. DT_MIPS_RLD_MAP exists in non-PIE. - else if ((symbol.d_tag == DT_MIPS_RLD_MAP || symbol.d_tag == DT_MIPS_RLD_MAP_REL) && target) - { - addr_t offset = i * dynsym_hdr->sh_entsize + GetAddressByteSize(); - addr_t dyn_base = dynsym_section_sp->GetLoadBaseAddress(target); - if (dyn_base == LLDB_INVALID_ADDRESS) - return Address(); - - Error error; - if (symbol.d_tag == DT_MIPS_RLD_MAP) - { - // DT_MIPS_RLD_MAP tag stores an absolute address of the debug pointer. - Address addr; - if (target->ReadPointerFromMemory(dyn_base + offset, false, error, addr)) - return addr; - } - if (symbol.d_tag == DT_MIPS_RLD_MAP_REL) - { - // DT_MIPS_RLD_MAP_REL tag stores the offset to the debug pointer, relative to the address of the tag. - uint64_t rel_offset; - rel_offset = target->ReadUnsignedIntegerFromMemory(dyn_base + offset, false, GetAddressByteSize(), UINT64_MAX, error); - if (error.Success() && rel_offset != UINT64_MAX) - { - Address addr; - addr_t debug_ptr_address = dyn_base + (offset - GetAddressByteSize()) + rel_offset; - addr.SetOffset (debug_ptr_address); - return addr; - } - } + Error error; + if (symbol.d_tag == DT_MIPS_RLD_MAP) { + // DT_MIPS_RLD_MAP tag stores an absolute address of the debug pointer. + Address addr; + if (target->ReadPointerFromMemory(dyn_base + offset, false, error, + addr)) + return addr; + } + if (symbol.d_tag == DT_MIPS_RLD_MAP_REL) { + // DT_MIPS_RLD_MAP_REL tag stores the offset to the debug pointer, + // relative to the address of the tag. + uint64_t rel_offset; + rel_offset = target->ReadUnsignedIntegerFromMemory( + dyn_base + offset, false, GetAddressByteSize(), UINT64_MAX, error); + if (error.Success() && rel_offset != UINT64_MAX) { + Address addr; + addr_t debug_ptr_address = + dyn_base + (offset - GetAddressByteSize()) + rel_offset; + addr.SetOffset(debug_ptr_address); + return addr; } + } } + } - return Address(); + return Address(); } -lldb_private::Address -ObjectFileELF::GetEntryPointAddress () -{ - if (m_entry_point_address.IsValid()) - return m_entry_point_address; +lldb_private::Address ObjectFileELF::GetEntryPointAddress() { + if (m_entry_point_address.IsValid()) + return m_entry_point_address; - if (!ParseHeader() || !IsExecutable()) - return m_entry_point_address; + if (!ParseHeader() || !IsExecutable()) + return m_entry_point_address; - SectionList *section_list = GetSectionList(); - addr_t offset = m_header.e_entry; + SectionList *section_list = GetSectionList(); + addr_t offset = m_header.e_entry; - if (!section_list) - m_entry_point_address.SetOffset(offset); - else - m_entry_point_address.ResolveAddressUsingFileSections(offset, section_list); - return m_entry_point_address; + if (!section_list) + m_entry_point_address.SetOffset(offset); + else + m_entry_point_address.ResolveAddressUsingFileSections(offset, section_list); + return m_entry_point_address; } //---------------------------------------------------------------------- // ParseDependentModules //---------------------------------------------------------------------- -size_t -ObjectFileELF::ParseDependentModules() -{ - if (m_filespec_ap.get()) - return m_filespec_ap->GetSize(); +size_t ObjectFileELF::ParseDependentModules() { + if (m_filespec_ap.get()) + return m_filespec_ap->GetSize(); - m_filespec_ap.reset(new FileSpecList()); + m_filespec_ap.reset(new FileSpecList()); - if (!ParseSectionHeaders()) - return 0; + if (!ParseSectionHeaders()) + return 0; - SectionList *section_list = GetSectionList(); - if (!section_list) - return 0; - - // Find the SHT_DYNAMIC section. - Section *dynsym = section_list->FindSectionByType (eSectionTypeELFDynamicLinkInfo, true).get(); - if (!dynsym) - return 0; - assert (dynsym->GetObjectFile() == this); - - const ELFSectionHeaderInfo *header = GetSectionHeaderByIndex (dynsym->GetID()); - if (!header) - return 0; - // sh_link: section header index of string table used by entries in the section. - Section *dynstr = section_list->FindSectionByID (header->sh_link + 1).get(); - if (!dynstr) - return 0; - - DataExtractor dynsym_data; - DataExtractor dynstr_data; - if (ReadSectionData(dynsym, dynsym_data) && - ReadSectionData(dynstr, dynstr_data)) - { - ELFDynamic symbol; - const lldb::offset_t section_size = dynsym_data.GetByteSize(); - lldb::offset_t offset = 0; - - // The only type of entries we are concerned with are tagged DT_NEEDED, - // yielding the name of a required library. - while (offset < section_size) - { - if (!symbol.Parse(dynsym_data, &offset)) - break; + SectionList *section_list = GetSectionList(); + if (!section_list) + return 0; - if (symbol.d_tag != DT_NEEDED) - continue; + // Find the SHT_DYNAMIC section. + Section *dynsym = + section_list->FindSectionByType(eSectionTypeELFDynamicLinkInfo, true) + .get(); + if (!dynsym) + return 0; + assert(dynsym->GetObjectFile() == this); - uint32_t str_index = static_cast<uint32_t>(symbol.d_val); - const char *lib_name = dynstr_data.PeekCStr(str_index); - m_filespec_ap->Append(FileSpec(lib_name, true)); - } + const ELFSectionHeaderInfo *header = GetSectionHeaderByIndex(dynsym->GetID()); + if (!header) + return 0; + // sh_link: section header index of string table used by entries in the + // section. + Section *dynstr = section_list->FindSectionByID(header->sh_link + 1).get(); + if (!dynstr) + return 0; + + DataExtractor dynsym_data; + DataExtractor dynstr_data; + if (ReadSectionData(dynsym, dynsym_data) && + ReadSectionData(dynstr, dynstr_data)) { + ELFDynamic symbol; + const lldb::offset_t section_size = dynsym_data.GetByteSize(); + lldb::offset_t offset = 0; + + // The only type of entries we are concerned with are tagged DT_NEEDED, + // yielding the name of a required library. + while (offset < section_size) { + if (!symbol.Parse(dynsym_data, &offset)) + break; + + if (symbol.d_tag != DT_NEEDED) + continue; + + uint32_t str_index = static_cast<uint32_t>(symbol.d_val); + const char *lib_name = dynstr_data.PeekCStr(str_index); + m_filespec_ap->Append(FileSpec(lib_name, true)); } + } - return m_filespec_ap->GetSize(); + return m_filespec_ap->GetSize(); } //---------------------------------------------------------------------- // GetProgramHeaderInfo //---------------------------------------------------------------------- -size_t -ObjectFileELF::GetProgramHeaderInfo(ProgramHeaderColl &program_headers, - const SetDataFunction &set_data, - const ELFHeader &header) -{ - // We have already parsed the program headers - if (!program_headers.empty()) - return program_headers.size(); - - // If there are no program headers to read we are done. - if (header.e_phnum == 0) - return 0; - - program_headers.resize(header.e_phnum); - if (program_headers.size() != header.e_phnum) - return 0; - - const size_t ph_size = header.e_phnum * header.e_phentsize; - const elf_off ph_offset = header.e_phoff; - DataExtractor data; - if (set_data(data, ph_offset, ph_size) != ph_size) - return 0; - - uint32_t idx; - lldb::offset_t offset; - for (idx = 0, offset = 0; idx < header.e_phnum; ++idx) - { - if (program_headers[idx].Parse(data, &offset) == false) - break; - } +size_t ObjectFileELF::GetProgramHeaderInfo(ProgramHeaderColl &program_headers, + const SetDataFunction &set_data, + const ELFHeader &header) { + // We have already parsed the program headers + if (!program_headers.empty()) + return program_headers.size(); - if (idx < program_headers.size()) - program_headers.resize(idx); + // If there are no program headers to read we are done. + if (header.e_phnum == 0) + return 0; - return program_headers.size(); + program_headers.resize(header.e_phnum); + if (program_headers.size() != header.e_phnum) + return 0; + + const size_t ph_size = header.e_phnum * header.e_phentsize; + const elf_off ph_offset = header.e_phoff; + DataExtractor data; + if (set_data(data, ph_offset, ph_size) != ph_size) + return 0; + + uint32_t idx; + lldb::offset_t offset; + for (idx = 0, offset = 0; idx < header.e_phnum; ++idx) { + if (program_headers[idx].Parse(data, &offset) == false) + break; + } + if (idx < program_headers.size()) + program_headers.resize(idx); + + return program_headers.size(); } //---------------------------------------------------------------------- // ParseProgramHeaders //---------------------------------------------------------------------- -size_t -ObjectFileELF::ParseProgramHeaders() -{ - using namespace std::placeholders; - return GetProgramHeaderInfo(m_program_headers, - std::bind(&ObjectFileELF::SetDataWithReadMemoryFallback, this, _1, _2, _3), - m_header); +size_t ObjectFileELF::ParseProgramHeaders() { + using namespace std::placeholders; + return GetProgramHeaderInfo( + m_program_headers, + std::bind(&ObjectFileELF::SetDataWithReadMemoryFallback, this, _1, _2, + _3), + m_header); } lldb_private::Error -ObjectFileELF::RefineModuleDetailsFromNote (lldb_private::DataExtractor &data, lldb_private::ArchSpec &arch_spec, lldb_private::UUID &uuid) -{ - Log *log(lldb_private::GetLogIfAllCategoriesSet (LIBLLDB_LOG_MODULES)); - Error error; +ObjectFileELF::RefineModuleDetailsFromNote(lldb_private::DataExtractor &data, + lldb_private::ArchSpec &arch_spec, + lldb_private::UUID &uuid) { + Log *log(lldb_private::GetLogIfAllCategoriesSet(LIBLLDB_LOG_MODULES)); + Error error; + + lldb::offset_t offset = 0; + + while (true) { + // Parse the note header. If this fails, bail out. + const lldb::offset_t note_offset = offset; + ELFNote note = ELFNote(); + if (!note.Parse(data, &offset)) { + // We're done. + return error; + } + + if (log) + log->Printf("ObjectFileELF::%s parsing note name='%s', type=%" PRIu32, + __FUNCTION__, note.n_name.c_str(), note.n_type); + + // Process FreeBSD ELF notes. + if ((note.n_name == LLDB_NT_OWNER_FREEBSD) && + (note.n_type == LLDB_NT_FREEBSD_ABI_TAG) && + (note.n_descsz == LLDB_NT_FREEBSD_ABI_SIZE)) { + // Pull out the min version info. + uint32_t version_info; + if (data.GetU32(&offset, &version_info, 1) == nullptr) { + error.SetErrorString("failed to read FreeBSD ABI note payload"); + return error; + } + + // Convert the version info into a major/minor number. + const uint32_t version_major = version_info / 100000; + const uint32_t version_minor = (version_info / 1000) % 100; + + char os_name[32]; + snprintf(os_name, sizeof(os_name), "freebsd%" PRIu32 ".%" PRIu32, + version_major, version_minor); + + // Set the elf OS version to FreeBSD. Also clear the vendor. + arch_spec.GetTriple().setOSName(os_name); + arch_spec.GetTriple().setVendor(llvm::Triple::VendorType::UnknownVendor); + + if (log) + log->Printf("ObjectFileELF::%s detected FreeBSD %" PRIu32 ".%" PRIu32 + ".%" PRIu32, + __FUNCTION__, version_major, version_minor, + static_cast<uint32_t>(version_info % 1000)); + } + // Process GNU ELF notes. + else if (note.n_name == LLDB_NT_OWNER_GNU) { + switch (note.n_type) { + case LLDB_NT_GNU_ABI_TAG: + if (note.n_descsz == LLDB_NT_GNU_ABI_SIZE) { + // Pull out the min OS version supporting the ABI. + uint32_t version_info[4]; + if (data.GetU32(&offset, &version_info[0], note.n_descsz / 4) == + nullptr) { + error.SetErrorString("failed to read GNU ABI note payload"); + return error; + } - lldb::offset_t offset = 0; + // Set the OS per the OS field. + switch (version_info[0]) { + case LLDB_NT_GNU_ABI_OS_LINUX: + arch_spec.GetTriple().setOS(llvm::Triple::OSType::Linux); + arch_spec.GetTriple().setVendor( + llvm::Triple::VendorType::UnknownVendor); + if (log) + log->Printf( + "ObjectFileELF::%s detected Linux, min version %" PRIu32 + ".%" PRIu32 ".%" PRIu32, + __FUNCTION__, version_info[1], version_info[2], + version_info[3]); + // FIXME we have the minimal version number, we could be propagating + // that. version_info[1] = OS Major, version_info[2] = OS Minor, + // version_info[3] = Revision. + break; + case LLDB_NT_GNU_ABI_OS_HURD: + arch_spec.GetTriple().setOS(llvm::Triple::OSType::UnknownOS); + arch_spec.GetTriple().setVendor( + llvm::Triple::VendorType::UnknownVendor); + if (log) + log->Printf("ObjectFileELF::%s detected Hurd (unsupported), min " + "version %" PRIu32 ".%" PRIu32 ".%" PRIu32, + __FUNCTION__, version_info[1], version_info[2], + version_info[3]); + break; + case LLDB_NT_GNU_ABI_OS_SOLARIS: + arch_spec.GetTriple().setOS(llvm::Triple::OSType::Solaris); + arch_spec.GetTriple().setVendor( + llvm::Triple::VendorType::UnknownVendor); + if (log) + log->Printf( + "ObjectFileELF::%s detected Solaris, min version %" PRIu32 + ".%" PRIu32 ".%" PRIu32, + __FUNCTION__, version_info[1], version_info[2], + version_info[3]); + break; + default: + if (log) + log->Printf( + "ObjectFileELF::%s unrecognized OS in note, id %" PRIu32 + ", min version %" PRIu32 ".%" PRIu32 ".%" PRIu32, + __FUNCTION__, version_info[0], version_info[1], + version_info[2], version_info[3]); + break; + } + } + break; + + case LLDB_NT_GNU_BUILD_ID_TAG: + // Only bother processing this if we don't already have the uuid set. + if (!uuid.IsValid()) { + // 16 bytes is UUID|MD5, 20 bytes is SHA1. Other linkers may produce a + // build-id of a different + // length. Accept it as long as it's at least 4 bytes as it will be + // better than our own crc32. + if (note.n_descsz >= 4 && note.n_descsz <= 20) { + uint8_t uuidbuf[20]; + if (data.GetU8(&offset, &uuidbuf, note.n_descsz) == nullptr) { + error.SetErrorString("failed to read GNU_BUILD_ID note payload"); + return error; + } - while (true) - { - // Parse the note header. If this fails, bail out. - const lldb::offset_t note_offset = offset; - ELFNote note = ELFNote(); - if (!note.Parse(data, &offset)) - { - // We're done. + // Save the build id as the UUID for the module. + uuid.SetBytes(uuidbuf, note.n_descsz); + } + } + break; + } + } + // Process NetBSD ELF notes. + else if ((note.n_name == LLDB_NT_OWNER_NETBSD) && + (note.n_type == LLDB_NT_NETBSD_ABI_TAG) && + (note.n_descsz == LLDB_NT_NETBSD_ABI_SIZE)) { + // Pull out the min version info. + uint32_t version_info; + if (data.GetU32(&offset, &version_info, 1) == nullptr) { + error.SetErrorString("failed to read NetBSD ABI note payload"); + return error; + } + + // Set the elf OS version to NetBSD. Also clear the vendor. + arch_spec.GetTriple().setOS(llvm::Triple::OSType::NetBSD); + arch_spec.GetTriple().setVendor(llvm::Triple::VendorType::UnknownVendor); + + if (log) + log->Printf( + "ObjectFileELF::%s detected NetBSD, min version constant %" PRIu32, + __FUNCTION__, version_info); + } + // Process CSR kalimba notes + else if ((note.n_type == LLDB_NT_GNU_ABI_TAG) && + (note.n_name == LLDB_NT_OWNER_CSR)) { + arch_spec.GetTriple().setOS(llvm::Triple::OSType::UnknownOS); + arch_spec.GetTriple().setVendor(llvm::Triple::VendorType::CSR); + + // TODO At some point the description string could be processed. + // It could provide a steer towards the kalimba variant which + // this ELF targets. + if (note.n_descsz) { + const char *cstr = + data.GetCStr(&offset, llvm::alignTo(note.n_descsz, 4)); + (void)cstr; + } + } else if (note.n_name == LLDB_NT_OWNER_ANDROID) { + arch_spec.GetTriple().setOS(llvm::Triple::OSType::Linux); + arch_spec.GetTriple().setEnvironment( + llvm::Triple::EnvironmentType::Android); + } else if (note.n_name == LLDB_NT_OWNER_LINUX) { + // This is sometimes found in core files and usually contains extended + // register info + arch_spec.GetTriple().setOS(llvm::Triple::OSType::Linux); + } else if (note.n_name == LLDB_NT_OWNER_CORE) { + // Parse the NT_FILE to look for stuff in paths to shared libraries + // As the contents look like this in a 64 bit ELF core file: + // count = 0x000000000000000a (10) + // page_size = 0x0000000000001000 (4096) + // Index start end file_ofs path + // ===== ------------------ ------------------ ------------------ + // ------------------------------------- + // [ 0] 0x0000000000400000 0x0000000000401000 0x0000000000000000 + // /tmp/a.out + // [ 1] 0x0000000000600000 0x0000000000601000 0x0000000000000000 + // /tmp/a.out + // [ 2] 0x0000000000601000 0x0000000000602000 0x0000000000000001 + // /tmp/a.out + // [ 3] 0x00007fa79c9ed000 0x00007fa79cba8000 0x0000000000000000 + // /lib/x86_64-linux-gnu/libc-2.19.so + // [ 4] 0x00007fa79cba8000 0x00007fa79cda7000 0x00000000000001bb + // /lib/x86_64-linux-gnu/libc-2.19.so + // [ 5] 0x00007fa79cda7000 0x00007fa79cdab000 0x00000000000001ba + // /lib/x86_64-linux-gnu/libc-2.19.so + // [ 6] 0x00007fa79cdab000 0x00007fa79cdad000 0x00000000000001be + // /lib/x86_64-linux-gnu/libc-2.19.so + // [ 7] 0x00007fa79cdb2000 0x00007fa79cdd5000 0x0000000000000000 + // /lib/x86_64-linux-gnu/ld-2.19.so + // [ 8] 0x00007fa79cfd4000 0x00007fa79cfd5000 0x0000000000000022 + // /lib/x86_64-linux-gnu/ld-2.19.so + // [ 9] 0x00007fa79cfd5000 0x00007fa79cfd6000 0x0000000000000023 + // /lib/x86_64-linux-gnu/ld-2.19.so + // In the 32 bit ELFs the count, page_size, start, end, file_ofs are + // uint32_t + // For reference: see readelf source code (in binutils). + if (note.n_type == NT_FILE) { + uint64_t count = data.GetAddress(&offset); + const char *cstr; + data.GetAddress(&offset); // Skip page size + offset += count * 3 * + data.GetAddressByteSize(); // Skip all start/end/file_ofs + for (size_t i = 0; i < count; ++i) { + cstr = data.GetCStr(&offset); + if (cstr == nullptr) { + error.SetErrorStringWithFormat("ObjectFileELF::%s trying to read " + "at an offset after the end " + "(GetCStr returned nullptr)", + __FUNCTION__); return error; + } + llvm::StringRef path(cstr); + if (path.contains("/lib/x86_64-linux-gnu") || path.contains("/lib/i386-linux-gnu")) { + arch_spec.GetTriple().setOS(llvm::Triple::OSType::Linux); + break; + } } + } + } - if (log) - log->Printf ("ObjectFileELF::%s parsing note name='%s', type=%" PRIu32, __FUNCTION__, note.n_name.c_str (), note.n_type); - - // Process FreeBSD ELF notes. - if ((note.n_name == LLDB_NT_OWNER_FREEBSD) && - (note.n_type == LLDB_NT_FREEBSD_ABI_TAG) && - (note.n_descsz == LLDB_NT_FREEBSD_ABI_SIZE)) - { - // Pull out the min version info. - uint32_t version_info; - if (data.GetU32 (&offset, &version_info, 1) == nullptr) - { - error.SetErrorString ("failed to read FreeBSD ABI note payload"); - return error; - } + // Calculate the offset of the next note just in case "offset" has been used + // to poke at the contents of the note data + offset = note_offset + note.GetByteSize(); + } - // Convert the version info into a major/minor number. - const uint32_t version_major = version_info / 100000; - const uint32_t version_minor = (version_info / 1000) % 100; + return error; +} - char os_name[32]; - snprintf (os_name, sizeof (os_name), "freebsd%" PRIu32 ".%" PRIu32, version_major, version_minor); +void ObjectFileELF::ParseARMAttributes(DataExtractor &data, uint64_t length, + ArchSpec &arch_spec) { + lldb::offset_t Offset = 0; - // Set the elf OS version to FreeBSD. Also clear the vendor. - arch_spec.GetTriple ().setOSName (os_name); - arch_spec.GetTriple ().setVendor (llvm::Triple::VendorType::UnknownVendor); + uint8_t FormatVersion = data.GetU8(&Offset); + if (FormatVersion != llvm::ARMBuildAttrs::Format_Version) + return; - if (log) - log->Printf ("ObjectFileELF::%s detected FreeBSD %" PRIu32 ".%" PRIu32 ".%" PRIu32, __FUNCTION__, version_major, version_minor, static_cast<uint32_t> (version_info % 1000)); - } - // Process GNU ELF notes. - else if (note.n_name == LLDB_NT_OWNER_GNU) - { - switch (note.n_type) - { - case LLDB_NT_GNU_ABI_TAG: - if (note.n_descsz == LLDB_NT_GNU_ABI_SIZE) - { - // Pull out the min OS version supporting the ABI. - uint32_t version_info[4]; - if (data.GetU32 (&offset, &version_info[0], note.n_descsz / 4) == nullptr) - { - error.SetErrorString ("failed to read GNU ABI note payload"); - return error; - } - - // Set the OS per the OS field. - switch (version_info[0]) - { - case LLDB_NT_GNU_ABI_OS_LINUX: - arch_spec.GetTriple ().setOS (llvm::Triple::OSType::Linux); - arch_spec.GetTriple ().setVendor (llvm::Triple::VendorType::UnknownVendor); - if (log) - log->Printf ("ObjectFileELF::%s detected Linux, min version %" PRIu32 ".%" PRIu32 ".%" PRIu32, __FUNCTION__, version_info[1], version_info[2], version_info[3]); - // FIXME we have the minimal version number, we could be propagating that. version_info[1] = OS Major, version_info[2] = OS Minor, version_info[3] = Revision. - break; - case LLDB_NT_GNU_ABI_OS_HURD: - arch_spec.GetTriple ().setOS (llvm::Triple::OSType::UnknownOS); - arch_spec.GetTriple ().setVendor (llvm::Triple::VendorType::UnknownVendor); - if (log) - log->Printf ("ObjectFileELF::%s detected Hurd (unsupported), min version %" PRIu32 ".%" PRIu32 ".%" PRIu32, __FUNCTION__, version_info[1], version_info[2], version_info[3]); - break; - case LLDB_NT_GNU_ABI_OS_SOLARIS: - arch_spec.GetTriple ().setOS (llvm::Triple::OSType::Solaris); - arch_spec.GetTriple ().setVendor (llvm::Triple::VendorType::UnknownVendor); - if (log) - log->Printf ("ObjectFileELF::%s detected Solaris, min version %" PRIu32 ".%" PRIu32 ".%" PRIu32, __FUNCTION__, version_info[1], version_info[2], version_info[3]); - break; - default: - if (log) - log->Printf ("ObjectFileELF::%s unrecognized OS in note, id %" PRIu32 ", min version %" PRIu32 ".%" PRIu32 ".%" PRIu32, __FUNCTION__, version_info[0], version_info[1], version_info[2], version_info[3]); - break; - } - } - break; - - case LLDB_NT_GNU_BUILD_ID_TAG: - // Only bother processing this if we don't already have the uuid set. - if (!uuid.IsValid()) - { - // 16 bytes is UUID|MD5, 20 bytes is SHA1. Other linkers may produce a build-id of a different - // length. Accept it as long as it's at least 4 bytes as it will be better than our own crc32. - if (note.n_descsz >= 4 && note.n_descsz <= 20) - { - uint8_t uuidbuf[20]; - if (data.GetU8 (&offset, &uuidbuf, note.n_descsz) == nullptr) - { - error.SetErrorString ("failed to read GNU_BUILD_ID note payload"); - return error; - } - - // Save the build id as the UUID for the module. - uuid.SetBytes (uuidbuf, note.n_descsz); - } - } - break; - } - } - // Process NetBSD ELF notes. - else if ((note.n_name == LLDB_NT_OWNER_NETBSD) && - (note.n_type == LLDB_NT_NETBSD_ABI_TAG) && - (note.n_descsz == LLDB_NT_NETBSD_ABI_SIZE)) - { - // Pull out the min version info. - uint32_t version_info; - if (data.GetU32 (&offset, &version_info, 1) == nullptr) - { - error.SetErrorString ("failed to read NetBSD ABI note payload"); - return error; - } + Offset = Offset + sizeof(uint32_t); // Section Length + llvm::StringRef VendorName = data.GetCStr(&Offset); - // Set the elf OS version to NetBSD. Also clear the vendor. - arch_spec.GetTriple ().setOS (llvm::Triple::OSType::NetBSD); - arch_spec.GetTriple ().setVendor (llvm::Triple::VendorType::UnknownVendor); + if (VendorName != "aeabi") + return; - if (log) - log->Printf ("ObjectFileELF::%s detected NetBSD, min version constant %" PRIu32, __FUNCTION__, version_info); - } - // Process CSR kalimba notes - else if ((note.n_type == LLDB_NT_GNU_ABI_TAG) && - (note.n_name == LLDB_NT_OWNER_CSR)) - { - arch_spec.GetTriple().setOS(llvm::Triple::OSType::UnknownOS); - arch_spec.GetTriple().setVendor(llvm::Triple::VendorType::CSR); - - // TODO At some point the description string could be processed. - // It could provide a steer towards the kalimba variant which - // this ELF targets. - if(note.n_descsz) - { - const char *cstr = data.GetCStr(&offset, llvm::alignTo (note.n_descsz, 4)); - (void)cstr; - } - } - else if (note.n_name == LLDB_NT_OWNER_ANDROID) - { - arch_spec.GetTriple().setOS(llvm::Triple::OSType::Linux); - arch_spec.GetTriple().setEnvironment(llvm::Triple::EnvironmentType::Android); - } - else if (note.n_name == LLDB_NT_OWNER_LINUX) - { - // This is sometimes found in core files and usually contains extended register info - arch_spec.GetTriple().setOS(llvm::Triple::OSType::Linux); - } - else if (note.n_name == LLDB_NT_OWNER_CORE) - { - // Parse the NT_FILE to look for stuff in paths to shared libraries - // As the contents look like: - // count = 0x000000000000000a (10) - // page_size = 0x0000000000001000 (4096) - // Index start end file_ofs path - // ===== ------------------ ------------------ ------------------ ------------------------------------- - // [ 0] 0x0000000000400000 0x0000000000401000 0x0000000000000000 /tmp/a.out - // [ 1] 0x0000000000600000 0x0000000000601000 0x0000000000000000 /tmp/a.out - // [ 2] 0x0000000000601000 0x0000000000602000 0x0000000000000001 /tmp/a.out - // [ 3] 0x00007fa79c9ed000 0x00007fa79cba8000 0x0000000000000000 /lib/x86_64-linux-gnu/libc-2.19.so - // [ 4] 0x00007fa79cba8000 0x00007fa79cda7000 0x00000000000001bb /lib/x86_64-linux-gnu/libc-2.19.so - // [ 5] 0x00007fa79cda7000 0x00007fa79cdab000 0x00000000000001ba /lib/x86_64-linux-gnu/libc-2.19.so - // [ 6] 0x00007fa79cdab000 0x00007fa79cdad000 0x00000000000001be /lib/x86_64-linux-gnu/libc-2.19.so - // [ 7] 0x00007fa79cdb2000 0x00007fa79cdd5000 0x0000000000000000 /lib/x86_64-linux-gnu/ld-2.19.so - // [ 8] 0x00007fa79cfd4000 0x00007fa79cfd5000 0x0000000000000022 /lib/x86_64-linux-gnu/ld-2.19.so - // [ 9] 0x00007fa79cfd5000 0x00007fa79cfd6000 0x0000000000000023 /lib/x86_64-linux-gnu/ld-2.19.so - if (note.n_type == NT_FILE) - { - uint64_t count = data.GetU64(&offset); - offset += 8 + 3*8*count; // Skip page size and all start/end/file_ofs - for (size_t i=0; i<count; ++i) - { - llvm::StringRef path(data.GetCStr(&offset)); - if (path.startswith("/lib/x86_64-linux-gnu")) - { - arch_spec.GetTriple().setOS(llvm::Triple::OSType::Linux); - break; - } - } - } - } + if (arch_spec.GetTriple().getEnvironment() == + llvm::Triple::UnknownEnvironment) + arch_spec.GetTriple().setEnvironment(llvm::Triple::EABI); - // Calculate the offset of the next note just in case "offset" has been used - // to poke at the contents of the note data - offset = note_offset + note.GetByteSize(); - } + while (Offset < length) { + uint8_t Tag = data.GetU8(&Offset); + uint32_t Size = data.GetU32(&Offset); - return error; -} + if (Tag != llvm::ARMBuildAttrs::File || Size == 0) + continue; -void -ObjectFileELF::ParseARMAttributes(DataExtractor &data, uint64_t length, ArchSpec &arch_spec) -{ - lldb::offset_t Offset = 0; + while (Offset < length) { + uint64_t Tag = data.GetULEB128(&Offset); + switch (Tag) { + default: + if (Tag < 32) + data.GetULEB128(&Offset); + else if (Tag % 2 == 0) + data.GetULEB128(&Offset); + else + data.GetCStr(&Offset); - uint8_t FormatVersion = data.GetU8(&Offset); - if (FormatVersion != llvm::ARMBuildAttrs::Format_Version) - return; + break; - Offset = Offset + sizeof(uint32_t); // Section Length - llvm::StringRef VendorName = data.GetCStr(&Offset); + case llvm::ARMBuildAttrs::CPU_raw_name: + case llvm::ARMBuildAttrs::CPU_name: + data.GetCStr(&Offset); - if (VendorName != "aeabi") - return; + break; - if (arch_spec.GetTriple().getEnvironment() == llvm::Triple::UnknownEnvironment) - arch_spec.GetTriple().setEnvironment(llvm::Triple::EABI); + case llvm::ARMBuildAttrs::ABI_VFP_args: { + uint64_t VFPArgs = data.GetULEB128(&Offset); - while (Offset < length) - { - uint8_t Tag = data.GetU8(&Offset); - uint32_t Size = data.GetU32(&Offset); + if (VFPArgs == llvm::ARMBuildAttrs::BaseAAPCS) { + if (arch_spec.GetTriple().getEnvironment() == + llvm::Triple::UnknownEnvironment || + arch_spec.GetTriple().getEnvironment() == llvm::Triple::EABIHF) + arch_spec.GetTriple().setEnvironment(llvm::Triple::EABI); - if (Tag != llvm::ARMBuildAttrs::File || Size == 0) - continue; + arch_spec.SetFlags(ArchSpec::eARM_abi_soft_float); + } else if (VFPArgs == llvm::ARMBuildAttrs::HardFPAAPCS) { + if (arch_spec.GetTriple().getEnvironment() == + llvm::Triple::UnknownEnvironment || + arch_spec.GetTriple().getEnvironment() == llvm::Triple::EABI) + arch_spec.GetTriple().setEnvironment(llvm::Triple::EABIHF); - while (Offset < length) - { - uint64_t Tag = data.GetULEB128(&Offset); - switch (Tag) - { - default: - if (Tag < 32) - data.GetULEB128(&Offset); - else if (Tag % 2 == 0) - data.GetULEB128(&Offset); - else - data.GetCStr(&Offset); - - break; - - case llvm::ARMBuildAttrs::CPU_raw_name: - case llvm::ARMBuildAttrs::CPU_name: - data.GetCStr(&Offset); - - break; - - case llvm::ARMBuildAttrs::ABI_VFP_args: - { - uint64_t VFPArgs = data.GetULEB128(&Offset); - - if (VFPArgs == llvm::ARMBuildAttrs::BaseAAPCS) - { - if (arch_spec.GetTriple().getEnvironment() == llvm::Triple::UnknownEnvironment || - arch_spec.GetTriple().getEnvironment() == llvm::Triple::EABIHF) - arch_spec.GetTriple().setEnvironment(llvm::Triple::EABI); - - arch_spec.SetFlags(ArchSpec::eARM_abi_soft_float); - } - else if (VFPArgs == llvm::ARMBuildAttrs::HardFPAAPCS) - { - if (arch_spec.GetTriple().getEnvironment() == llvm::Triple::UnknownEnvironment || - arch_spec.GetTriple().getEnvironment() == llvm::Triple::EABI) - arch_spec.GetTriple().setEnvironment(llvm::Triple::EABIHF); - - arch_spec.SetFlags(ArchSpec::eARM_abi_hard_float); - } - - break; - } - } + arch_spec.SetFlags(ArchSpec::eARM_abi_hard_float); } + + break; + } + } } + } } //---------------------------------------------------------------------- // GetSectionHeaderInfo //---------------------------------------------------------------------- -size_t -ObjectFileELF::GetSectionHeaderInfo(SectionHeaderColl §ion_headers, - const SetDataFunction &set_data, - const elf::ELFHeader &header, - lldb_private::UUID &uuid, - std::string &gnu_debuglink_file, - uint32_t &gnu_debuglink_crc, - ArchSpec &arch_spec) -{ - // Don't reparse the section headers if we already did that. - if (!section_headers.empty()) - return section_headers.size(); - - // Only initialize the arch_spec to okay defaults if they're not already set. - // We'll refine this with note data as we parse the notes. - if (arch_spec.GetTriple ().getOS () == llvm::Triple::OSType::UnknownOS) - { - llvm::Triple::OSType ostype; - llvm::Triple::OSType spec_ostype; - const uint32_t sub_type = subTypeFromElfHeader(header); - arch_spec.SetArchitecture (eArchTypeELF, header.e_machine, sub_type, header.e_ident[EI_OSABI]); - // - // Validate if it is ok to remove GetOsFromOSABI - GetOsFromOSABI (header.e_ident[EI_OSABI], ostype); - spec_ostype = arch_spec.GetTriple ().getOS (); - assert(spec_ostype == ostype); - } - - if (arch_spec.GetMachine() == llvm::Triple::mips || arch_spec.GetMachine() == llvm::Triple::mipsel - || arch_spec.GetMachine() == llvm::Triple::mips64 || arch_spec.GetMachine() == llvm::Triple::mips64el) - { - switch (header.e_flags & llvm::ELF::EF_MIPS_ARCH_ASE) - { - case llvm::ELF::EF_MIPS_MICROMIPS: - arch_spec.SetFlags (ArchSpec::eMIPSAse_micromips); - break; - case llvm::ELF::EF_MIPS_ARCH_ASE_M16: - arch_spec.SetFlags (ArchSpec::eMIPSAse_mips16); - break; - case llvm::ELF::EF_MIPS_ARCH_ASE_MDMX: - arch_spec.SetFlags (ArchSpec::eMIPSAse_mdmx); - break; - default: - break; - } +size_t ObjectFileELF::GetSectionHeaderInfo(SectionHeaderColl §ion_headers, + const SetDataFunction &set_data, + const elf::ELFHeader &header, + lldb_private::UUID &uuid, + std::string &gnu_debuglink_file, + uint32_t &gnu_debuglink_crc, + ArchSpec &arch_spec) { + // Don't reparse the section headers if we already did that. + if (!section_headers.empty()) + return section_headers.size(); + + // Only initialize the arch_spec to okay defaults if they're not already set. + // We'll refine this with note data as we parse the notes. + if (arch_spec.GetTriple().getOS() == llvm::Triple::OSType::UnknownOS) { + llvm::Triple::OSType ostype; + llvm::Triple::OSType spec_ostype; + const uint32_t sub_type = subTypeFromElfHeader(header); + arch_spec.SetArchitecture(eArchTypeELF, header.e_machine, sub_type, + header.e_ident[EI_OSABI]); + + // Validate if it is ok to remove GetOsFromOSABI. + // Note, that now the OS is determined based on EI_OSABI flag and + // the info extracted from ELF notes (see RefineModuleDetailsFromNote). + // However in some cases that still might be not enough: for example + // a shared library might not have any notes at all + // and have EI_OSABI flag set to System V, + // as result the OS will be set to UnknownOS. + GetOsFromOSABI(header.e_ident[EI_OSABI], ostype); + spec_ostype = arch_spec.GetTriple().getOS(); + assert(spec_ostype == ostype); + } + + if (arch_spec.GetMachine() == llvm::Triple::mips || + arch_spec.GetMachine() == llvm::Triple::mipsel || + arch_spec.GetMachine() == llvm::Triple::mips64 || + arch_spec.GetMachine() == llvm::Triple::mips64el) { + switch (header.e_flags & llvm::ELF::EF_MIPS_ARCH_ASE) { + case llvm::ELF::EF_MIPS_MICROMIPS: + arch_spec.SetFlags(ArchSpec::eMIPSAse_micromips); + break; + case llvm::ELF::EF_MIPS_ARCH_ASE_M16: + arch_spec.SetFlags(ArchSpec::eMIPSAse_mips16); + break; + case llvm::ELF::EF_MIPS_ARCH_ASE_MDMX: + arch_spec.SetFlags(ArchSpec::eMIPSAse_mdmx); + break; + default: + break; } + } - if (arch_spec.GetMachine() == llvm::Triple::arm || - arch_spec.GetMachine() == llvm::Triple::thumb) - { - if (header.e_flags & llvm::ELF::EF_ARM_SOFT_FLOAT) - arch_spec.SetFlags (ArchSpec::eARM_abi_soft_float); - else if (header.e_flags & llvm::ELF::EF_ARM_VFP_FLOAT) - arch_spec.SetFlags (ArchSpec::eARM_abi_hard_float); - } + if (arch_spec.GetMachine() == llvm::Triple::arm || + arch_spec.GetMachine() == llvm::Triple::thumb) { + if (header.e_flags & llvm::ELF::EF_ARM_SOFT_FLOAT) + arch_spec.SetFlags(ArchSpec::eARM_abi_soft_float); + else if (header.e_flags & llvm::ELF::EF_ARM_VFP_FLOAT) + arch_spec.SetFlags(ArchSpec::eARM_abi_hard_float); + } - // If there are no section headers we are done. - if (header.e_shnum == 0) - return 0; + // If there are no section headers we are done. + if (header.e_shnum == 0) + return 0; - Log *log(lldb_private::GetLogIfAllCategoriesSet (LIBLLDB_LOG_MODULES)); + Log *log(lldb_private::GetLogIfAllCategoriesSet(LIBLLDB_LOG_MODULES)); - section_headers.resize(header.e_shnum); - if (section_headers.size() != header.e_shnum) - return 0; + section_headers.resize(header.e_shnum); + if (section_headers.size() != header.e_shnum) + return 0; - const size_t sh_size = header.e_shnum * header.e_shentsize; - const elf_off sh_offset = header.e_shoff; - DataExtractor sh_data; - if (set_data (sh_data, sh_offset, sh_size) != sh_size) - return 0; + const size_t sh_size = header.e_shnum * header.e_shentsize; + const elf_off sh_offset = header.e_shoff; + DataExtractor sh_data; + if (set_data(sh_data, sh_offset, sh_size) != sh_size) + return 0; - uint32_t idx; - lldb::offset_t offset; - for (idx = 0, offset = 0; idx < header.e_shnum; ++idx) - { - if (section_headers[idx].Parse(sh_data, &offset) == false) + uint32_t idx; + lldb::offset_t offset; + for (idx = 0, offset = 0; idx < header.e_shnum; ++idx) { + if (section_headers[idx].Parse(sh_data, &offset) == false) + break; + } + if (idx < section_headers.size()) + section_headers.resize(idx); + + const unsigned strtab_idx = header.e_shstrndx; + if (strtab_idx && strtab_idx < section_headers.size()) { + const ELFSectionHeaderInfo &sheader = section_headers[strtab_idx]; + const size_t byte_size = sheader.sh_size; + const Elf64_Off offset = sheader.sh_offset; + lldb_private::DataExtractor shstr_data; + + if (set_data(shstr_data, offset, byte_size) == byte_size) { + for (SectionHeaderCollIter I = section_headers.begin(); + I != section_headers.end(); ++I) { + static ConstString g_sect_name_gnu_debuglink(".gnu_debuglink"); + const ELFSectionHeaderInfo &sheader = *I; + const uint64_t section_size = + sheader.sh_type == SHT_NOBITS ? 0 : sheader.sh_size; + ConstString name(shstr_data.PeekCStr(I->sh_name)); + + I->section_name = name; + + if (arch_spec.IsMIPS()) { + uint32_t arch_flags = arch_spec.GetFlags(); + DataExtractor data; + if (sheader.sh_type == SHT_MIPS_ABIFLAGS) { + + if (section_size && (set_data(data, sheader.sh_offset, + section_size) == section_size)) { + // MIPS ASE Mask is at offset 12 in MIPS.abiflags section + lldb::offset_t offset = 12; // MIPS ABI Flags Version: 0 + arch_flags |= data.GetU32(&offset); + + // The floating point ABI is at offset 7 + offset = 7; + switch (data.GetU8(&offset)) { + case llvm::Mips::Val_GNU_MIPS_ABI_FP_ANY: + arch_flags |= lldb_private::ArchSpec::eMIPS_ABI_FP_ANY; + break; + case llvm::Mips::Val_GNU_MIPS_ABI_FP_DOUBLE: + arch_flags |= lldb_private::ArchSpec::eMIPS_ABI_FP_DOUBLE; + break; + case llvm::Mips::Val_GNU_MIPS_ABI_FP_SINGLE: + arch_flags |= lldb_private::ArchSpec::eMIPS_ABI_FP_SINGLE; + break; + case llvm::Mips::Val_GNU_MIPS_ABI_FP_SOFT: + arch_flags |= lldb_private::ArchSpec::eMIPS_ABI_FP_SOFT; + break; + case llvm::Mips::Val_GNU_MIPS_ABI_FP_OLD_64: + arch_flags |= lldb_private::ArchSpec::eMIPS_ABI_FP_OLD_64; + break; + case llvm::Mips::Val_GNU_MIPS_ABI_FP_XX: + arch_flags |= lldb_private::ArchSpec::eMIPS_ABI_FP_XX; + break; + case llvm::Mips::Val_GNU_MIPS_ABI_FP_64: + arch_flags |= lldb_private::ArchSpec::eMIPS_ABI_FP_64; + break; + case llvm::Mips::Val_GNU_MIPS_ABI_FP_64A: + arch_flags |= lldb_private::ArchSpec::eMIPS_ABI_FP_64A; + break; + } + } + } + // Settings appropriate ArchSpec ABI Flags + switch (header.e_flags & llvm::ELF::EF_MIPS_ABI) { + case llvm::ELF::EF_MIPS_ABI_O32: + arch_flags |= lldb_private::ArchSpec::eMIPSABI_O32; break; - } - if (idx < section_headers.size()) - section_headers.resize(idx); - - const unsigned strtab_idx = header.e_shstrndx; - if (strtab_idx && strtab_idx < section_headers.size()) - { - const ELFSectionHeaderInfo &sheader = section_headers[strtab_idx]; - const size_t byte_size = sheader.sh_size; - const Elf64_Off offset = sheader.sh_offset; - lldb_private::DataExtractor shstr_data; - - if (set_data (shstr_data, offset, byte_size) == byte_size) - { - for (SectionHeaderCollIter I = section_headers.begin(); - I != section_headers.end(); ++I) - { - static ConstString g_sect_name_gnu_debuglink (".gnu_debuglink"); - const ELFSectionHeaderInfo &sheader = *I; - const uint64_t section_size = sheader.sh_type == SHT_NOBITS ? 0 : sheader.sh_size; - ConstString name(shstr_data.PeekCStr(I->sh_name)); - - I->section_name = name; - - if (arch_spec.IsMIPS()) - { - uint32_t arch_flags = arch_spec.GetFlags (); - DataExtractor data; - if (sheader.sh_type == SHT_MIPS_ABIFLAGS) - { - - if (section_size && (set_data (data, sheader.sh_offset, section_size) == section_size)) - { - // MIPS ASE Mask is at offset 12 in MIPS.abiflags section - lldb::offset_t offset = 12; // MIPS ABI Flags Version: 0 - arch_flags |= data.GetU32 (&offset); - - // The floating point ABI is at offset 7 - offset = 7; - switch (data.GetU8 (&offset)) - { - case llvm::Mips::Val_GNU_MIPS_ABI_FP_ANY : - arch_flags |= lldb_private::ArchSpec::eMIPS_ABI_FP_ANY; - break; - case llvm::Mips::Val_GNU_MIPS_ABI_FP_DOUBLE : - arch_flags |= lldb_private::ArchSpec::eMIPS_ABI_FP_DOUBLE; - break; - case llvm::Mips::Val_GNU_MIPS_ABI_FP_SINGLE : - arch_flags |= lldb_private::ArchSpec::eMIPS_ABI_FP_SINGLE; - break; - case llvm::Mips::Val_GNU_MIPS_ABI_FP_SOFT : - arch_flags |= lldb_private::ArchSpec::eMIPS_ABI_FP_SOFT; - break; - case llvm::Mips::Val_GNU_MIPS_ABI_FP_OLD_64 : - arch_flags |= lldb_private::ArchSpec::eMIPS_ABI_FP_OLD_64; - break; - case llvm::Mips::Val_GNU_MIPS_ABI_FP_XX : - arch_flags |= lldb_private::ArchSpec::eMIPS_ABI_FP_XX; - break; - case llvm::Mips::Val_GNU_MIPS_ABI_FP_64 : - arch_flags |= lldb_private::ArchSpec::eMIPS_ABI_FP_64; - break; - case llvm::Mips::Val_GNU_MIPS_ABI_FP_64A : - arch_flags |= lldb_private::ArchSpec::eMIPS_ABI_FP_64A; - break; - } - } - } - // Settings appropriate ArchSpec ABI Flags - switch(header.e_flags & llvm::ELF::EF_MIPS_ABI) - { - case llvm::ELF::EF_MIPS_ABI_O32: - arch_flags |= lldb_private::ArchSpec::eMIPSABI_O32; - break; - case EF_MIPS_ABI_O64: - arch_flags |= lldb_private::ArchSpec::eMIPSABI_O64; - break; - case EF_MIPS_ABI_EABI32: - arch_flags |= lldb_private::ArchSpec::eMIPSABI_EABI32; - break; - case EF_MIPS_ABI_EABI64: - arch_flags |= lldb_private::ArchSpec::eMIPSABI_EABI64; - break; - default: - // ABI Mask doesn't cover N32 and N64 ABI. - if (header.e_ident[EI_CLASS] == llvm::ELF::ELFCLASS64) - arch_flags |= lldb_private::ArchSpec::eMIPSABI_N64; - else if (header.e_flags && llvm::ELF::EF_MIPS_ABI2) - arch_flags |= lldb_private::ArchSpec::eMIPSABI_N32; - break; - } - arch_spec.SetFlags (arch_flags); - } + case EF_MIPS_ABI_O64: + arch_flags |= lldb_private::ArchSpec::eMIPSABI_O64; + break; + case EF_MIPS_ABI_EABI32: + arch_flags |= lldb_private::ArchSpec::eMIPSABI_EABI32; + break; + case EF_MIPS_ABI_EABI64: + arch_flags |= lldb_private::ArchSpec::eMIPSABI_EABI64; + break; + default: + // ABI Mask doesn't cover N32 and N64 ABI. + if (header.e_ident[EI_CLASS] == llvm::ELF::ELFCLASS64) + arch_flags |= lldb_private::ArchSpec::eMIPSABI_N64; + else if (header.e_flags && llvm::ELF::EF_MIPS_ABI2) + arch_flags |= lldb_private::ArchSpec::eMIPSABI_N32; + break; + } + arch_spec.SetFlags(arch_flags); + } - if (arch_spec.GetMachine() == llvm::Triple::arm || arch_spec.GetMachine() == llvm::Triple::thumb) - { - DataExtractor data; + if (arch_spec.GetMachine() == llvm::Triple::arm || + arch_spec.GetMachine() == llvm::Triple::thumb) { + DataExtractor data; - if (sheader.sh_type == SHT_ARM_ATTRIBUTES && section_size != 0 && - set_data(data, sheader.sh_offset, section_size) == section_size) - ParseARMAttributes(data, section_size, arch_spec); - } + if (sheader.sh_type == SHT_ARM_ATTRIBUTES && section_size != 0 && + set_data(data, sheader.sh_offset, section_size) == section_size) + ParseARMAttributes(data, section_size, arch_spec); + } - if (name == g_sect_name_gnu_debuglink) - { - DataExtractor data; - if (section_size && (set_data (data, sheader.sh_offset, section_size) == section_size)) - { - lldb::offset_t gnu_debuglink_offset = 0; - gnu_debuglink_file = data.GetCStr (&gnu_debuglink_offset); - gnu_debuglink_offset = llvm::alignTo (gnu_debuglink_offset, 4); - data.GetU32 (&gnu_debuglink_offset, &gnu_debuglink_crc, 1); - } - } + if (name == g_sect_name_gnu_debuglink) { + DataExtractor data; + if (section_size && (set_data(data, sheader.sh_offset, + section_size) == section_size)) { + lldb::offset_t gnu_debuglink_offset = 0; + gnu_debuglink_file = data.GetCStr(&gnu_debuglink_offset); + gnu_debuglink_offset = llvm::alignTo(gnu_debuglink_offset, 4); + data.GetU32(&gnu_debuglink_offset, &gnu_debuglink_crc, 1); + } + } - // Process ELF note section entries. - bool is_note_header = (sheader.sh_type == SHT_NOTE); - - // The section header ".note.android.ident" is stored as a - // PROGBITS type header but it is actually a note header. - static ConstString g_sect_name_android_ident (".note.android.ident"); - if (!is_note_header && name == g_sect_name_android_ident) - is_note_header = true; - - if (is_note_header) - { - // Allow notes to refine module info. - DataExtractor data; - if (section_size && (set_data (data, sheader.sh_offset, section_size) == section_size)) - { - Error error = RefineModuleDetailsFromNote (data, arch_spec, uuid); - if (error.Fail ()) - { - if (log) - log->Printf ("ObjectFileELF::%s ELF note processing failed: %s", __FUNCTION__, error.AsCString ()); - } - } - } + // Process ELF note section entries. + bool is_note_header = (sheader.sh_type == SHT_NOTE); + + // The section header ".note.android.ident" is stored as a + // PROGBITS type header but it is actually a note header. + static ConstString g_sect_name_android_ident(".note.android.ident"); + if (!is_note_header && name == g_sect_name_android_ident) + is_note_header = true; + + if (is_note_header) { + // Allow notes to refine module info. + DataExtractor data; + if (section_size && (set_data(data, sheader.sh_offset, + section_size) == section_size)) { + Error error = RefineModuleDetailsFromNote(data, arch_spec, uuid); + if (error.Fail()) { + if (log) + log->Printf("ObjectFileELF::%s ELF note processing failed: %s", + __FUNCTION__, error.AsCString()); } + } + } + } - // Make any unknown triple components to be unspecified unknowns. - if (arch_spec.GetTriple().getVendor() == llvm::Triple::UnknownVendor) - arch_spec.GetTriple().setVendorName (llvm::StringRef()); - if (arch_spec.GetTriple().getOS() == llvm::Triple::UnknownOS) - arch_spec.GetTriple().setOSName (llvm::StringRef()); + // Make any unknown triple components to be unspecified unknowns. + if (arch_spec.GetTriple().getVendor() == llvm::Triple::UnknownVendor) + arch_spec.GetTriple().setVendorName(llvm::StringRef()); + if (arch_spec.GetTriple().getOS() == llvm::Triple::UnknownOS) + arch_spec.GetTriple().setOSName(llvm::StringRef()); - return section_headers.size(); - } + return section_headers.size(); } + } - section_headers.clear(); - return 0; + section_headers.clear(); + return 0; } -size_t -ObjectFileELF::GetProgramHeaderCount() -{ - return ParseProgramHeaders(); -} +size_t ObjectFileELF::GetProgramHeaderCount() { return ParseProgramHeaders(); } const elf::ELFProgramHeader * -ObjectFileELF::GetProgramHeaderByIndex(lldb::user_id_t id) -{ - if (!id || !ParseProgramHeaders()) - return NULL; +ObjectFileELF::GetProgramHeaderByIndex(lldb::user_id_t id) { + if (!id || !ParseProgramHeaders()) + return NULL; - if (--id < m_program_headers.size()) - return &m_program_headers[id]; + if (--id < m_program_headers.size()) + return &m_program_headers[id]; - return NULL; + return NULL; } -DataExtractor -ObjectFileELF::GetSegmentDataByIndex(lldb::user_id_t id) -{ - const elf::ELFProgramHeader *segment_header = GetProgramHeaderByIndex(id); - if (segment_header == NULL) - return DataExtractor(); - return DataExtractor(m_data, segment_header->p_offset, segment_header->p_filesz); +DataExtractor ObjectFileELF::GetSegmentDataByIndex(lldb::user_id_t id) { + const elf::ELFProgramHeader *segment_header = GetProgramHeaderByIndex(id); + if (segment_header == NULL) + return DataExtractor(); + return DataExtractor(m_data, segment_header->p_offset, + segment_header->p_filesz); } std::string -ObjectFileELF::StripLinkerSymbolAnnotations(llvm::StringRef symbol_name) const -{ - size_t pos = symbol_name.find('@'); - return symbol_name.substr(0, pos).str(); +ObjectFileELF::StripLinkerSymbolAnnotations(llvm::StringRef symbol_name) const { + size_t pos = symbol_name.find('@'); + return symbol_name.substr(0, pos).str(); } //---------------------------------------------------------------------- // ParseSectionHeaders //---------------------------------------------------------------------- -size_t -ObjectFileELF::ParseSectionHeaders() -{ - using namespace std::placeholders; +size_t ObjectFileELF::ParseSectionHeaders() { + using namespace std::placeholders; - return GetSectionHeaderInfo(m_section_headers, - std::bind(&ObjectFileELF::SetDataWithReadMemoryFallback, this, _1, _2, _3), - m_header, - m_uuid, - m_gnu_debuglink_file, - m_gnu_debuglink_crc, - m_arch_spec); + return GetSectionHeaderInfo( + m_section_headers, + std::bind(&ObjectFileELF::SetDataWithReadMemoryFallback, this, _1, _2, + _3), + m_header, m_uuid, m_gnu_debuglink_file, m_gnu_debuglink_crc, m_arch_spec); } -lldb::offset_t -ObjectFileELF::SetData(const lldb_private::DataExtractor &src, lldb_private::DataExtractor &dst, lldb::offset_t offset, lldb::offset_t length) -{ - return dst.SetData(src, offset, length); +lldb::offset_t ObjectFileELF::SetData(const lldb_private::DataExtractor &src, + lldb_private::DataExtractor &dst, + lldb::offset_t offset, + lldb::offset_t length) { + return dst.SetData(src, offset, length); } lldb::offset_t -ObjectFileELF::SetDataWithReadMemoryFallback(lldb_private::DataExtractor &dst, lldb::offset_t offset, lldb::offset_t length) -{ - if (offset + length <= m_data.GetByteSize()) - return dst.SetData(m_data, offset, length); - - const auto process_sp = m_process_wp.lock(); - if (process_sp != nullptr) - { - addr_t file_size = offset + length; - - DataBufferSP data_sp = ReadMemory(process_sp, m_memory_addr, file_size); - if (!data_sp) - return false; - m_data.SetData(data_sp, 0, file_size); - } - +ObjectFileELF::SetDataWithReadMemoryFallback(lldb_private::DataExtractor &dst, + lldb::offset_t offset, + lldb::offset_t length) { + if (offset + length <= m_data.GetByteSize()) return dst.SetData(m_data, offset, length); + + const auto process_sp = m_process_wp.lock(); + if (process_sp != nullptr) { + addr_t file_size = offset + length; + + DataBufferSP data_sp = ReadMemory(process_sp, m_memory_addr, file_size); + if (!data_sp) + return false; + m_data.SetData(data_sp, 0, file_size); + } + + return dst.SetData(m_data, offset, length); } const ObjectFileELF::ELFSectionHeaderInfo * -ObjectFileELF::GetSectionHeaderByIndex(lldb::user_id_t id) -{ - if (!id || !ParseSectionHeaders()) - return NULL; +ObjectFileELF::GetSectionHeaderByIndex(lldb::user_id_t id) { + if (!id || !ParseSectionHeaders()) + return NULL; - if (--id < m_section_headers.size()) - return &m_section_headers[id]; + if (--id < m_section_headers.size()) + return &m_section_headers[id]; - return NULL; + return NULL; } -lldb::user_id_t -ObjectFileELF::GetSectionIndexByName(const char* name) -{ - if (!name || !name[0] || !ParseSectionHeaders()) - return 0; - for (size_t i = 1; i < m_section_headers.size(); ++i) - if (m_section_headers[i].section_name == ConstString(name)) - return i; +lldb::user_id_t ObjectFileELF::GetSectionIndexByName(const char *name) { + if (!name || !name[0] || !ParseSectionHeaders()) return 0; + for (size_t i = 1; i < m_section_headers.size(); ++i) + if (m_section_headers[i].section_name == ConstString(name)) + return i; + return 0; } -void -ObjectFileELF::CreateSections(SectionList &unified_section_list) -{ - if (!m_sections_ap.get() && ParseSectionHeaders()) - { - m_sections_ap.reset(new SectionList()); - - for (SectionHeaderCollIter I = m_section_headers.begin(); - I != m_section_headers.end(); ++I) - { - const ELFSectionHeaderInfo &header = *I; - - ConstString& name = I->section_name; - const uint64_t file_size = header.sh_type == SHT_NOBITS ? 0 : header.sh_size; - const uint64_t vm_size = header.sh_flags & SHF_ALLOC ? header.sh_size : 0; - - static ConstString g_sect_name_text (".text"); - static ConstString g_sect_name_data (".data"); - static ConstString g_sect_name_bss (".bss"); - static ConstString g_sect_name_tdata (".tdata"); - static ConstString g_sect_name_tbss (".tbss"); - static ConstString g_sect_name_dwarf_debug_abbrev (".debug_abbrev"); - static ConstString g_sect_name_dwarf_debug_addr (".debug_addr"); - static ConstString g_sect_name_dwarf_debug_aranges (".debug_aranges"); - static ConstString g_sect_name_dwarf_debug_frame (".debug_frame"); - static ConstString g_sect_name_dwarf_debug_info (".debug_info"); - static ConstString g_sect_name_dwarf_debug_line (".debug_line"); - static ConstString g_sect_name_dwarf_debug_loc (".debug_loc"); - static ConstString g_sect_name_dwarf_debug_macinfo (".debug_macinfo"); - static ConstString g_sect_name_dwarf_debug_macro (".debug_macro"); - static ConstString g_sect_name_dwarf_debug_pubnames (".debug_pubnames"); - static ConstString g_sect_name_dwarf_debug_pubtypes (".debug_pubtypes"); - static ConstString g_sect_name_dwarf_debug_ranges (".debug_ranges"); - static ConstString g_sect_name_dwarf_debug_str (".debug_str"); - static ConstString g_sect_name_dwarf_debug_str_offsets (".debug_str_offsets"); - static ConstString g_sect_name_dwarf_debug_abbrev_dwo (".debug_abbrev.dwo"); - static ConstString g_sect_name_dwarf_debug_info_dwo (".debug_info.dwo"); - static ConstString g_sect_name_dwarf_debug_line_dwo (".debug_line.dwo"); - static ConstString g_sect_name_dwarf_debug_macro_dwo (".debug_macro.dwo"); - static ConstString g_sect_name_dwarf_debug_loc_dwo (".debug_loc.dwo"); - static ConstString g_sect_name_dwarf_debug_str_dwo (".debug_str.dwo"); - static ConstString g_sect_name_dwarf_debug_str_offsets_dwo (".debug_str_offsets.dwo"); - static ConstString g_sect_name_eh_frame (".eh_frame"); - static ConstString g_sect_name_arm_exidx (".ARM.exidx"); - static ConstString g_sect_name_arm_extab (".ARM.extab"); - static ConstString g_sect_name_go_symtab (".gosymtab"); - - SectionType sect_type = eSectionTypeOther; - - bool is_thread_specific = false; - - if (name == g_sect_name_text) sect_type = eSectionTypeCode; - else if (name == g_sect_name_data) sect_type = eSectionTypeData; - else if (name == g_sect_name_bss) sect_type = eSectionTypeZeroFill; - else if (name == g_sect_name_tdata) - { - sect_type = eSectionTypeData; - is_thread_specific = true; - } - else if (name == g_sect_name_tbss) - { - sect_type = eSectionTypeZeroFill; - is_thread_specific = true; - } - // .debug_abbrev – Abbreviations used in the .debug_info section - // .debug_aranges – Lookup table for mapping addresses to compilation units - // .debug_frame – Call frame information - // .debug_info – The core DWARF information section - // .debug_line – Line number information - // .debug_loc – Location lists used in DW_AT_location attributes - // .debug_macinfo – Macro information - // .debug_pubnames – Lookup table for mapping object and function names to compilation units - // .debug_pubtypes – Lookup table for mapping type names to compilation units - // .debug_ranges – Address ranges used in DW_AT_ranges attributes - // .debug_str – String table used in .debug_info - // MISSING? .gnu_debugdata - "mini debuginfo / MiniDebugInfo" section, http://sourceware.org/gdb/onlinedocs/gdb/MiniDebugInfo.html - // MISSING? .debug-index - http://src.chromium.org/viewvc/chrome/trunk/src/build/gdb-add-index?pathrev=144644 - // MISSING? .debug_types - Type descriptions from DWARF 4? See http://gcc.gnu.org/wiki/DwarfSeparateTypeInfo - else if (name == g_sect_name_dwarf_debug_abbrev) sect_type = eSectionTypeDWARFDebugAbbrev; - else if (name == g_sect_name_dwarf_debug_addr) sect_type = eSectionTypeDWARFDebugAddr; - else if (name == g_sect_name_dwarf_debug_aranges) sect_type = eSectionTypeDWARFDebugAranges; - else if (name == g_sect_name_dwarf_debug_frame) sect_type = eSectionTypeDWARFDebugFrame; - else if (name == g_sect_name_dwarf_debug_info) sect_type = eSectionTypeDWARFDebugInfo; - else if (name == g_sect_name_dwarf_debug_line) sect_type = eSectionTypeDWARFDebugLine; - else if (name == g_sect_name_dwarf_debug_loc) sect_type = eSectionTypeDWARFDebugLoc; - else if (name == g_sect_name_dwarf_debug_macinfo) sect_type = eSectionTypeDWARFDebugMacInfo; - else if (name == g_sect_name_dwarf_debug_macro) sect_type = eSectionTypeDWARFDebugMacro; - else if (name == g_sect_name_dwarf_debug_pubnames) sect_type = eSectionTypeDWARFDebugPubNames; - else if (name == g_sect_name_dwarf_debug_pubtypes) sect_type = eSectionTypeDWARFDebugPubTypes; - else if (name == g_sect_name_dwarf_debug_ranges) sect_type = eSectionTypeDWARFDebugRanges; - else if (name == g_sect_name_dwarf_debug_str) sect_type = eSectionTypeDWARFDebugStr; - else if (name == g_sect_name_dwarf_debug_str_offsets) sect_type = eSectionTypeDWARFDebugStrOffsets; - else if (name == g_sect_name_dwarf_debug_abbrev_dwo) sect_type = eSectionTypeDWARFDebugAbbrev; - else if (name == g_sect_name_dwarf_debug_info_dwo) sect_type = eSectionTypeDWARFDebugInfo; - else if (name == g_sect_name_dwarf_debug_line_dwo) sect_type = eSectionTypeDWARFDebugLine; - else if (name == g_sect_name_dwarf_debug_macro_dwo) sect_type = eSectionTypeDWARFDebugMacro; - else if (name == g_sect_name_dwarf_debug_loc_dwo) sect_type = eSectionTypeDWARFDebugLoc; - else if (name == g_sect_name_dwarf_debug_str_dwo) sect_type = eSectionTypeDWARFDebugStr; - else if (name == g_sect_name_dwarf_debug_str_offsets_dwo) sect_type = eSectionTypeDWARFDebugStrOffsets; - else if (name == g_sect_name_eh_frame) sect_type = eSectionTypeEHFrame; - else if (name == g_sect_name_arm_exidx) sect_type = eSectionTypeARMexidx; - else if (name == g_sect_name_arm_extab) sect_type = eSectionTypeARMextab; - else if (name == g_sect_name_go_symtab) sect_type = eSectionTypeGoSymtab; - - const uint32_t permissions = ((header.sh_flags & SHF_ALLOC) ? ePermissionsReadable : 0) | - ((header.sh_flags & SHF_WRITE) ? ePermissionsWritable : 0) | - ((header.sh_flags & SHF_EXECINSTR) ? ePermissionsExecutable : 0); - switch (header.sh_type) - { - case SHT_SYMTAB: - assert (sect_type == eSectionTypeOther); - sect_type = eSectionTypeELFSymbolTable; - break; - case SHT_DYNSYM: - assert (sect_type == eSectionTypeOther); - sect_type = eSectionTypeELFDynamicSymbols; - break; - case SHT_RELA: - case SHT_REL: - assert (sect_type == eSectionTypeOther); - sect_type = eSectionTypeELFRelocationEntries; - break; - case SHT_DYNAMIC: - assert (sect_type == eSectionTypeOther); - sect_type = eSectionTypeELFDynamicLinkInfo; - break; - } - - if (eSectionTypeOther == sect_type) - { - // the kalimba toolchain assumes that ELF section names are free-form. It does - // support linkscripts which (can) give rise to various arbitrarily named - // sections being "Code" or "Data". - sect_type = kalimbaSectionType(m_header, header); - } - - const uint32_t target_bytes_size = - (eSectionTypeData == sect_type || eSectionTypeZeroFill == sect_type) ? - m_arch_spec.GetDataByteSize() : - eSectionTypeCode == sect_type ? - m_arch_spec.GetCodeByteSize() : 1; - - elf::elf_xword log2align = (header.sh_addralign==0) - ? 0 - : llvm::Log2_64(header.sh_addralign); - SectionSP section_sp (new Section(GetModule(), // Module to which this section belongs. - this, // ObjectFile to which this section belongs and should read section data from. - SectionIndex(I), // Section ID. - name, // Section name. - sect_type, // Section type. - header.sh_addr, // VM address. - vm_size, // VM size in bytes of this section. - header.sh_offset, // Offset of this section in the file. - file_size, // Size of the section as found in the file. - log2align, // Alignment of the section - header.sh_flags, // Flags for this section. - target_bytes_size));// Number of host bytes per target byte - - section_sp->SetPermissions(permissions); - if (is_thread_specific) - section_sp->SetIsThreadSpecific (is_thread_specific); - m_sections_ap->AddSection(section_sp); - } - } +void ObjectFileELF::CreateSections(SectionList &unified_section_list) { + if (!m_sections_ap.get() && ParseSectionHeaders()) { + m_sections_ap.reset(new SectionList()); - if (m_sections_ap.get()) - { - if (GetType() == eTypeDebugInfo) - { - static const SectionType g_sections[] = - { - eSectionTypeDWARFDebugAbbrev, - eSectionTypeDWARFDebugAddr, - eSectionTypeDWARFDebugAranges, - eSectionTypeDWARFDebugFrame, - eSectionTypeDWARFDebugInfo, - eSectionTypeDWARFDebugLine, - eSectionTypeDWARFDebugLoc, - eSectionTypeDWARFDebugMacInfo, - eSectionTypeDWARFDebugPubNames, - eSectionTypeDWARFDebugPubTypes, - eSectionTypeDWARFDebugRanges, - eSectionTypeDWARFDebugStr, - eSectionTypeDWARFDebugStrOffsets, - eSectionTypeELFSymbolTable, - }; - SectionList *elf_section_list = m_sections_ap.get(); - for (size_t idx = 0; idx < sizeof(g_sections) / sizeof(g_sections[0]); ++idx) - { - SectionType section_type = g_sections[idx]; - SectionSP section_sp (elf_section_list->FindSectionByType (section_type, true)); - if (section_sp) - { - SectionSP module_section_sp (unified_section_list.FindSectionByType (section_type, true)); - if (module_section_sp) - unified_section_list.ReplaceSection (module_section_sp->GetID(), section_sp); - else - unified_section_list.AddSection (section_sp); - } - } - } - else - { - unified_section_list = *m_sections_ap; + for (SectionHeaderCollIter I = m_section_headers.begin(); + I != m_section_headers.end(); ++I) { + const ELFSectionHeaderInfo &header = *I; + + ConstString &name = I->section_name; + const uint64_t file_size = + header.sh_type == SHT_NOBITS ? 0 : header.sh_size; + const uint64_t vm_size = header.sh_flags & SHF_ALLOC ? header.sh_size : 0; + + static ConstString g_sect_name_text(".text"); + static ConstString g_sect_name_data(".data"); + static ConstString g_sect_name_bss(".bss"); + static ConstString g_sect_name_tdata(".tdata"); + static ConstString g_sect_name_tbss(".tbss"); + static ConstString g_sect_name_dwarf_debug_abbrev(".debug_abbrev"); + static ConstString g_sect_name_dwarf_debug_addr(".debug_addr"); + static ConstString g_sect_name_dwarf_debug_aranges(".debug_aranges"); + static ConstString g_sect_name_dwarf_debug_frame(".debug_frame"); + static ConstString g_sect_name_dwarf_debug_info(".debug_info"); + static ConstString g_sect_name_dwarf_debug_line(".debug_line"); + static ConstString g_sect_name_dwarf_debug_loc(".debug_loc"); + static ConstString g_sect_name_dwarf_debug_macinfo(".debug_macinfo"); + static ConstString g_sect_name_dwarf_debug_macro(".debug_macro"); + static ConstString g_sect_name_dwarf_debug_pubnames(".debug_pubnames"); + static ConstString g_sect_name_dwarf_debug_pubtypes(".debug_pubtypes"); + static ConstString g_sect_name_dwarf_debug_ranges(".debug_ranges"); + static ConstString g_sect_name_dwarf_debug_str(".debug_str"); + static ConstString g_sect_name_dwarf_debug_str_offsets( + ".debug_str_offsets"); + static ConstString g_sect_name_dwarf_debug_abbrev_dwo( + ".debug_abbrev.dwo"); + static ConstString g_sect_name_dwarf_debug_info_dwo(".debug_info.dwo"); + static ConstString g_sect_name_dwarf_debug_line_dwo(".debug_line.dwo"); + static ConstString g_sect_name_dwarf_debug_macro_dwo(".debug_macro.dwo"); + static ConstString g_sect_name_dwarf_debug_loc_dwo(".debug_loc.dwo"); + static ConstString g_sect_name_dwarf_debug_str_dwo(".debug_str.dwo"); + static ConstString g_sect_name_dwarf_debug_str_offsets_dwo( + ".debug_str_offsets.dwo"); + static ConstString g_sect_name_eh_frame(".eh_frame"); + static ConstString g_sect_name_arm_exidx(".ARM.exidx"); + static ConstString g_sect_name_arm_extab(".ARM.extab"); + static ConstString g_sect_name_go_symtab(".gosymtab"); + + SectionType sect_type = eSectionTypeOther; + + bool is_thread_specific = false; + + if (name == g_sect_name_text) + sect_type = eSectionTypeCode; + else if (name == g_sect_name_data) + sect_type = eSectionTypeData; + else if (name == g_sect_name_bss) + sect_type = eSectionTypeZeroFill; + else if (name == g_sect_name_tdata) { + sect_type = eSectionTypeData; + is_thread_specific = true; + } else if (name == g_sect_name_tbss) { + sect_type = eSectionTypeZeroFill; + is_thread_specific = true; + } + // .debug_abbrev – Abbreviations used in the .debug_info section + // .debug_aranges – Lookup table for mapping addresses to compilation + // units + // .debug_frame – Call frame information + // .debug_info – The core DWARF information section + // .debug_line – Line number information + // .debug_loc – Location lists used in DW_AT_location attributes + // .debug_macinfo – Macro information + // .debug_pubnames – Lookup table for mapping object and function names to + // compilation units + // .debug_pubtypes – Lookup table for mapping type names to compilation + // units + // .debug_ranges – Address ranges used in DW_AT_ranges attributes + // .debug_str – String table used in .debug_info + // MISSING? .gnu_debugdata - "mini debuginfo / MiniDebugInfo" section, + // http://sourceware.org/gdb/onlinedocs/gdb/MiniDebugInfo.html + // MISSING? .debug-index - + // http://src.chromium.org/viewvc/chrome/trunk/src/build/gdb-add-index?pathrev=144644 + // MISSING? .debug_types - Type descriptions from DWARF 4? See + // http://gcc.gnu.org/wiki/DwarfSeparateTypeInfo + else if (name == g_sect_name_dwarf_debug_abbrev) + sect_type = eSectionTypeDWARFDebugAbbrev; + else if (name == g_sect_name_dwarf_debug_addr) + sect_type = eSectionTypeDWARFDebugAddr; + else if (name == g_sect_name_dwarf_debug_aranges) + sect_type = eSectionTypeDWARFDebugAranges; + else if (name == g_sect_name_dwarf_debug_frame) + sect_type = eSectionTypeDWARFDebugFrame; + else if (name == g_sect_name_dwarf_debug_info) + sect_type = eSectionTypeDWARFDebugInfo; + else if (name == g_sect_name_dwarf_debug_line) + sect_type = eSectionTypeDWARFDebugLine; + else if (name == g_sect_name_dwarf_debug_loc) + sect_type = eSectionTypeDWARFDebugLoc; + else if (name == g_sect_name_dwarf_debug_macinfo) + sect_type = eSectionTypeDWARFDebugMacInfo; + else if (name == g_sect_name_dwarf_debug_macro) + sect_type = eSectionTypeDWARFDebugMacro; + else if (name == g_sect_name_dwarf_debug_pubnames) + sect_type = eSectionTypeDWARFDebugPubNames; + else if (name == g_sect_name_dwarf_debug_pubtypes) + sect_type = eSectionTypeDWARFDebugPubTypes; + else if (name == g_sect_name_dwarf_debug_ranges) + sect_type = eSectionTypeDWARFDebugRanges; + else if (name == g_sect_name_dwarf_debug_str) + sect_type = eSectionTypeDWARFDebugStr; + else if (name == g_sect_name_dwarf_debug_str_offsets) + sect_type = eSectionTypeDWARFDebugStrOffsets; + else if (name == g_sect_name_dwarf_debug_abbrev_dwo) + sect_type = eSectionTypeDWARFDebugAbbrev; + else if (name == g_sect_name_dwarf_debug_info_dwo) + sect_type = eSectionTypeDWARFDebugInfo; + else if (name == g_sect_name_dwarf_debug_line_dwo) + sect_type = eSectionTypeDWARFDebugLine; + else if (name == g_sect_name_dwarf_debug_macro_dwo) + sect_type = eSectionTypeDWARFDebugMacro; + else if (name == g_sect_name_dwarf_debug_loc_dwo) + sect_type = eSectionTypeDWARFDebugLoc; + else if (name == g_sect_name_dwarf_debug_str_dwo) + sect_type = eSectionTypeDWARFDebugStr; + else if (name == g_sect_name_dwarf_debug_str_offsets_dwo) + sect_type = eSectionTypeDWARFDebugStrOffsets; + else if (name == g_sect_name_eh_frame) + sect_type = eSectionTypeEHFrame; + else if (name == g_sect_name_arm_exidx) + sect_type = eSectionTypeARMexidx; + else if (name == g_sect_name_arm_extab) + sect_type = eSectionTypeARMextab; + else if (name == g_sect_name_go_symtab) + sect_type = eSectionTypeGoSymtab; + + const uint32_t permissions = + ((header.sh_flags & SHF_ALLOC) ? ePermissionsReadable : 0u) | + ((header.sh_flags & SHF_WRITE) ? ePermissionsWritable : 0u) | + ((header.sh_flags & SHF_EXECINSTR) ? ePermissionsExecutable : 0u); + switch (header.sh_type) { + case SHT_SYMTAB: + assert(sect_type == eSectionTypeOther); + sect_type = eSectionTypeELFSymbolTable; + break; + case SHT_DYNSYM: + assert(sect_type == eSectionTypeOther); + sect_type = eSectionTypeELFDynamicSymbols; + break; + case SHT_RELA: + case SHT_REL: + assert(sect_type == eSectionTypeOther); + sect_type = eSectionTypeELFRelocationEntries; + break; + case SHT_DYNAMIC: + assert(sect_type == eSectionTypeOther); + sect_type = eSectionTypeELFDynamicLinkInfo; + break; + } + + if (eSectionTypeOther == sect_type) { + // the kalimba toolchain assumes that ELF section names are free-form. + // It does + // support linkscripts which (can) give rise to various arbitrarily + // named + // sections being "Code" or "Data". + sect_type = kalimbaSectionType(m_header, header); + } + + const uint32_t target_bytes_size = + (eSectionTypeData == sect_type || eSectionTypeZeroFill == sect_type) + ? m_arch_spec.GetDataByteSize() + : eSectionTypeCode == sect_type ? m_arch_spec.GetCodeByteSize() + : 1; + + elf::elf_xword log2align = + (header.sh_addralign == 0) ? 0 : llvm::Log2_64(header.sh_addralign); + SectionSP section_sp(new Section( + GetModule(), // Module to which this section belongs. + this, // ObjectFile to which this section belongs and should read + // section data from. + SectionIndex(I), // Section ID. + name, // Section name. + sect_type, // Section type. + header.sh_addr, // VM address. + vm_size, // VM size in bytes of this section. + header.sh_offset, // Offset of this section in the file. + file_size, // Size of the section as found in the file. + log2align, // Alignment of the section + header.sh_flags, // Flags for this section. + target_bytes_size)); // Number of host bytes per target byte + + section_sp->SetPermissions(permissions); + if (is_thread_specific) + section_sp->SetIsThreadSpecific(is_thread_specific); + m_sections_ap->AddSection(section_sp); + } + } + + if (m_sections_ap.get()) { + if (GetType() == eTypeDebugInfo) { + static const SectionType g_sections[] = { + eSectionTypeDWARFDebugAbbrev, eSectionTypeDWARFDebugAddr, + eSectionTypeDWARFDebugAranges, eSectionTypeDWARFDebugFrame, + eSectionTypeDWARFDebugInfo, eSectionTypeDWARFDebugLine, + eSectionTypeDWARFDebugLoc, eSectionTypeDWARFDebugMacInfo, + eSectionTypeDWARFDebugPubNames, eSectionTypeDWARFDebugPubTypes, + eSectionTypeDWARFDebugRanges, eSectionTypeDWARFDebugStr, + eSectionTypeDWARFDebugStrOffsets, eSectionTypeELFSymbolTable, + }; + SectionList *elf_section_list = m_sections_ap.get(); + for (size_t idx = 0; idx < sizeof(g_sections) / sizeof(g_sections[0]); + ++idx) { + SectionType section_type = g_sections[idx]; + SectionSP section_sp( + elf_section_list->FindSectionByType(section_type, true)); + if (section_sp) { + SectionSP module_section_sp( + unified_section_list.FindSectionByType(section_type, true)); + if (module_section_sp) + unified_section_list.ReplaceSection(module_section_sp->GetID(), + section_sp); + else + unified_section_list.AddSection(section_sp); } + } + } else { + unified_section_list = *m_sections_ap; } + } } -// Find the arm/aarch64 mapping symbol character in the given symbol name. Mapping symbols have the -// form of "$<char>[.<any>]*". Additionally we recognize cases when the mapping symbol prefixed by -// an arbitrary string because if a symbol prefix added to each symbol in the object file with +// Find the arm/aarch64 mapping symbol character in the given symbol name. +// Mapping symbols have the +// form of "$<char>[.<any>]*". Additionally we recognize cases when the mapping +// symbol prefixed by +// an arbitrary string because if a symbol prefix added to each symbol in the +// object file with // objcopy then the mapping symbols are also prefixed. -static char -FindArmAarch64MappingSymbol(const char* symbol_name) -{ - if (!symbol_name) - return '\0'; - - const char* dollar_pos = ::strchr(symbol_name, '$'); - if (!dollar_pos || dollar_pos[1] == '\0') - return '\0'; +static char FindArmAarch64MappingSymbol(const char *symbol_name) { + if (!symbol_name) + return '\0'; - if (dollar_pos[2] == '\0' || dollar_pos[2] == '.') - return dollar_pos[1]; + const char *dollar_pos = ::strchr(symbol_name, '$'); + if (!dollar_pos || dollar_pos[1] == '\0') return '\0'; + + if (dollar_pos[2] == '\0' || dollar_pos[2] == '.') + return dollar_pos[1]; + return '\0'; } -#define STO_MIPS_ISA (3 << 6) -#define STO_MICROMIPS (2 << 6) -#define IS_MICROMIPS(ST_OTHER) (((ST_OTHER) & STO_MIPS_ISA) == STO_MICROMIPS) +#define STO_MIPS_ISA (3 << 6) +#define STO_MICROMIPS (2 << 6) +#define IS_MICROMIPS(ST_OTHER) (((ST_OTHER)&STO_MIPS_ISA) == STO_MICROMIPS) // private -unsigned -ObjectFileELF::ParseSymbols (Symtab *symtab, - user_id_t start_id, - SectionList *section_list, - const size_t num_symbols, - const DataExtractor &symtab_data, - const DataExtractor &strtab_data) -{ - ELFSymbol symbol; - lldb::offset_t offset = 0; +unsigned ObjectFileELF::ParseSymbols(Symtab *symtab, user_id_t start_id, + SectionList *section_list, + const size_t num_symbols, + const DataExtractor &symtab_data, + const DataExtractor &strtab_data) { + ELFSymbol symbol; + lldb::offset_t offset = 0; + + static ConstString text_section_name(".text"); + static ConstString init_section_name(".init"); + static ConstString fini_section_name(".fini"); + static ConstString ctors_section_name(".ctors"); + static ConstString dtors_section_name(".dtors"); + + static ConstString data_section_name(".data"); + static ConstString rodata_section_name(".rodata"); + static ConstString rodata1_section_name(".rodata1"); + static ConstString data2_section_name(".data1"); + static ConstString bss_section_name(".bss"); + static ConstString opd_section_name(".opd"); // For ppc64 + + // On Android the oatdata and the oatexec symbols in the oat and odex files + // covers the full + // .text section what causes issues with displaying unusable symbol name to + // the user and very + // slow unwinding speed because the instruction emulation based unwind plans + // try to emulate all + // instructions in these symbols. Don't add these symbols to the symbol list + // as they have no + // use for the debugger and they are causing a lot of trouble. + // Filtering can't be restricted to Android because this special object file + // don't contain the + // note section specifying the environment to Android but the custom extension + // and file name + // makes it highly unlikely that this will collide with anything else. + ConstString file_extension = m_file.GetFileNameExtension(); + bool skip_oatdata_oatexec = file_extension == ConstString("oat") || + file_extension == ConstString("odex"); + + ArchSpec arch; + GetArchitecture(arch); + ModuleSP module_sp(GetModule()); + SectionList *module_section_list = + module_sp ? module_sp->GetSectionList() : nullptr; + + // Local cache to avoid doing a FindSectionByName for each symbol. The "const + // char*" key must + // came from a ConstString object so they can be compared by pointer + std::unordered_map<const char *, lldb::SectionSP> section_name_to_section; + + unsigned i; + for (i = 0; i < num_symbols; ++i) { + if (symbol.Parse(symtab_data, &offset) == false) + break; + + const char *symbol_name = strtab_data.PeekCStr(symbol.st_name); + if (!symbol_name) + symbol_name = ""; + + // No need to add non-section symbols that have no names + if (symbol.getType() != STT_SECTION && + (symbol_name == nullptr || symbol_name[0] == '\0')) + continue; + + // Skipping oatdata and oatexec sections if it is requested. See details + // above the + // definition of skip_oatdata_oatexec for the reasons. + if (skip_oatdata_oatexec && (::strcmp(symbol_name, "oatdata") == 0 || + ::strcmp(symbol_name, "oatexec") == 0)) + continue; + + SectionSP symbol_section_sp; + SymbolType symbol_type = eSymbolTypeInvalid; + Elf64_Half section_idx = symbol.st_shndx; + + switch (section_idx) { + case SHN_ABS: + symbol_type = eSymbolTypeAbsolute; + break; + case SHN_UNDEF: + symbol_type = eSymbolTypeUndefined; + break; + default: + symbol_section_sp = section_list->GetSectionAtIndex(section_idx); + break; + } - static ConstString text_section_name(".text"); - static ConstString init_section_name(".init"); - static ConstString fini_section_name(".fini"); - static ConstString ctors_section_name(".ctors"); - static ConstString dtors_section_name(".dtors"); - - static ConstString data_section_name(".data"); - static ConstString rodata_section_name(".rodata"); - static ConstString rodata1_section_name(".rodata1"); - static ConstString data2_section_name(".data1"); - static ConstString bss_section_name(".bss"); - static ConstString opd_section_name(".opd"); // For ppc64 - - // On Android the oatdata and the oatexec symbols in the oat and odex files covers the full - // .text section what causes issues with displaying unusable symbol name to the user and very - // slow unwinding speed because the instruction emulation based unwind plans try to emulate all - // instructions in these symbols. Don't add these symbols to the symbol list as they have no - // use for the debugger and they are causing a lot of trouble. - // Filtering can't be restricted to Android because this special object file don't contain the - // note section specifying the environment to Android but the custom extension and file name - // makes it highly unlikely that this will collide with anything else. - ConstString file_extension = m_file.GetFileNameExtension(); - bool skip_oatdata_oatexec = file_extension == ConstString("oat") || file_extension == ConstString("odex"); - - ArchSpec arch; - GetArchitecture(arch); - ModuleSP module_sp(GetModule()); - SectionList* module_section_list = module_sp ? module_sp->GetSectionList() : nullptr; - - // Local cache to avoid doing a FindSectionByName for each symbol. The "const char*" key must - // came from a ConstString object so they can be compared by pointer - std::unordered_map<const char*, lldb::SectionSP> section_name_to_section; - - unsigned i; - for (i = 0; i < num_symbols; ++i) - { - if (symbol.Parse(symtab_data, &offset) == false) - break; + // If a symbol is undefined do not process it further even if it has a STT + // type + if (symbol_type != eSymbolTypeUndefined) { + switch (symbol.getType()) { + default: + case STT_NOTYPE: + // The symbol's type is not specified. + break; - const char *symbol_name = strtab_data.PeekCStr(symbol.st_name); + case STT_OBJECT: + // The symbol is associated with a data object, such as a variable, + // an array, etc. + symbol_type = eSymbolTypeData; + break; - // No need to add non-section symbols that have no names - if (symbol.getType() != STT_SECTION && - (symbol_name == NULL || symbol_name[0] == '\0')) - continue; + case STT_FUNC: + // The symbol is associated with a function or other executable code. + symbol_type = eSymbolTypeCode; + break; - // Skipping oatdata and oatexec sections if it is requested. See details above the - // definition of skip_oatdata_oatexec for the reasons. - if (skip_oatdata_oatexec && (::strcmp(symbol_name, "oatdata") == 0 || ::strcmp(symbol_name, "oatexec") == 0)) - continue; + case STT_SECTION: + // The symbol is associated with a section. Symbol table entries of + // this type exist primarily for relocation and normally have + // STB_LOCAL binding. + break; - SectionSP symbol_section_sp; - SymbolType symbol_type = eSymbolTypeInvalid; - Elf64_Half section_idx = symbol.st_shndx; + case STT_FILE: + // Conventionally, the symbol's name gives the name of the source + // file associated with the object file. A file symbol has STB_LOCAL + // binding, its section index is SHN_ABS, and it precedes the other + // STB_LOCAL symbols for the file, if it is present. + symbol_type = eSymbolTypeSourceFile; + break; - switch (section_idx) - { - case SHN_ABS: - symbol_type = eSymbolTypeAbsolute; - break; - case SHN_UNDEF: - symbol_type = eSymbolTypeUndefined; - break; - default: - symbol_section_sp = section_list->GetSectionAtIndex(section_idx); - break; + case STT_GNU_IFUNC: + // The symbol is associated with an indirect function. The actual + // function will be resolved if it is referenced. + symbol_type = eSymbolTypeResolver; + break; + } + } + + if (symbol_type == eSymbolTypeInvalid && symbol.getType() != STT_SECTION) { + if (symbol_section_sp) { + const ConstString §_name = symbol_section_sp->GetName(); + if (sect_name == text_section_name || sect_name == init_section_name || + sect_name == fini_section_name || sect_name == ctors_section_name || + sect_name == dtors_section_name) { + symbol_type = eSymbolTypeCode; + } else if (sect_name == data_section_name || + sect_name == data2_section_name || + sect_name == rodata_section_name || + sect_name == rodata1_section_name || + sect_name == bss_section_name) { + symbol_type = eSymbolTypeData; } - - // If a symbol is undefined do not process it further even if it has a STT type - if (symbol_type != eSymbolTypeUndefined) - { - switch (symbol.getType()) - { - default: - case STT_NOTYPE: - // The symbol's type is not specified. - break; - - case STT_OBJECT: - // The symbol is associated with a data object, such as a variable, - // an array, etc. - symbol_type = eSymbolTypeData; - break; - - case STT_FUNC: - // The symbol is associated with a function or other executable code. - symbol_type = eSymbolTypeCode; - break; - - case STT_SECTION: - // The symbol is associated with a section. Symbol table entries of - // this type exist primarily for relocation and normally have - // STB_LOCAL binding. - break; - - case STT_FILE: - // Conventionally, the symbol's name gives the name of the source - // file associated with the object file. A file symbol has STB_LOCAL - // binding, its section index is SHN_ABS, and it precedes the other - // STB_LOCAL symbols for the file, if it is present. - symbol_type = eSymbolTypeSourceFile; - break; - - case STT_GNU_IFUNC: - // The symbol is associated with an indirect function. The actual - // function will be resolved if it is referenced. - symbol_type = eSymbolTypeResolver; - break; - } - } - - if (symbol_type == eSymbolTypeInvalid && symbol.getType() != STT_SECTION) - { - if (symbol_section_sp) - { - const ConstString §_name = symbol_section_sp->GetName(); - if (sect_name == text_section_name || - sect_name == init_section_name || - sect_name == fini_section_name || - sect_name == ctors_section_name || - sect_name == dtors_section_name) - { - symbol_type = eSymbolTypeCode; - } - else if (sect_name == data_section_name || - sect_name == data2_section_name || - sect_name == rodata_section_name || - sect_name == rodata1_section_name || - sect_name == bss_section_name) - { - symbol_type = eSymbolTypeData; - } + } + } + + int64_t symbol_value_offset = 0; + uint32_t additional_flags = 0; + + if (arch.IsValid()) { + if (arch.GetMachine() == llvm::Triple::arm) { + if (symbol.getBinding() == STB_LOCAL) { + char mapping_symbol = FindArmAarch64MappingSymbol(symbol_name); + if (symbol_type == eSymbolTypeCode) { + switch (mapping_symbol) { + case 'a': + // $a[.<any>]* - marks an ARM instruction sequence + m_address_class_map[symbol.st_value] = eAddressClassCode; + break; + case 'b': + case 't': + // $b[.<any>]* - marks a THUMB BL instruction sequence + // $t[.<any>]* - marks a THUMB instruction sequence + m_address_class_map[symbol.st_value] = + eAddressClassCodeAlternateISA; + break; + case 'd': + // $d[.<any>]* - marks a data item sequence (e.g. lit pool) + m_address_class_map[symbol.st_value] = eAddressClassData; + break; } + } + if (mapping_symbol) + continue; } - - int64_t symbol_value_offset = 0; - uint32_t additional_flags = 0; - - if (arch.IsValid()) - { - if (arch.GetMachine() == llvm::Triple::arm) - { - if (symbol.getBinding() == STB_LOCAL) - { - char mapping_symbol = FindArmAarch64MappingSymbol(symbol_name); - if (symbol_type == eSymbolTypeCode) - { - switch (mapping_symbol) - { - case 'a': - // $a[.<any>]* - marks an ARM instruction sequence - m_address_class_map[symbol.st_value] = eAddressClassCode; - break; - case 'b': - case 't': - // $b[.<any>]* - marks a THUMB BL instruction sequence - // $t[.<any>]* - marks a THUMB instruction sequence - m_address_class_map[symbol.st_value] = eAddressClassCodeAlternateISA; - break; - case 'd': - // $d[.<any>]* - marks a data item sequence (e.g. lit pool) - m_address_class_map[symbol.st_value] = eAddressClassData; - break; - } - } - if (mapping_symbol) - continue; - } - } - else if (arch.GetMachine() == llvm::Triple::aarch64) - { - if (symbol.getBinding() == STB_LOCAL) - { - char mapping_symbol = FindArmAarch64MappingSymbol(symbol_name); - if (symbol_type == eSymbolTypeCode) - { - switch (mapping_symbol) - { - case 'x': - // $x[.<any>]* - marks an A64 instruction sequence - m_address_class_map[symbol.st_value] = eAddressClassCode; - break; - case 'd': - // $d[.<any>]* - marks a data item sequence (e.g. lit pool) - m_address_class_map[symbol.st_value] = eAddressClassData; - break; - } - } - if (mapping_symbol) - continue; - } - } - - if (arch.GetMachine() == llvm::Triple::arm) - { - if (symbol_type == eSymbolTypeCode) - { - if (symbol.st_value & 1) - { - // Subtracting 1 from the address effectively unsets - // the low order bit, which results in the address - // actually pointing to the beginning of the symbol. - // This delta will be used below in conjunction with - // symbol.st_value to produce the final symbol_value - // that we store in the symtab. - symbol_value_offset = -1; - m_address_class_map[symbol.st_value^1] = eAddressClassCodeAlternateISA; - } - else - { - // This address is ARM - m_address_class_map[symbol.st_value] = eAddressClassCode; - } - } - } - - /* - * MIPS: - * The bit #0 of an address is used for ISA mode (1 for microMIPS, 0 for MIPS). - * This allows processer to switch between microMIPS and MIPS without any need - * for special mode-control register. However, apart from .debug_line, none of - * the ELF/DWARF sections set the ISA bit (for symbol or section). Use st_other - * flag to check whether the symbol is microMIPS and then set the address class - * accordingly. - */ - const llvm::Triple::ArchType llvm_arch = arch.GetMachine(); - if (llvm_arch == llvm::Triple::mips || llvm_arch == llvm::Triple::mipsel - || llvm_arch == llvm::Triple::mips64 || llvm_arch == llvm::Triple::mips64el) - { - if (IS_MICROMIPS(symbol.st_other)) - m_address_class_map[symbol.st_value] = eAddressClassCodeAlternateISA; - else if ((symbol.st_value & 1) && (symbol_type == eSymbolTypeCode)) - { - symbol.st_value = symbol.st_value & (~1ull); - m_address_class_map[symbol.st_value] = eAddressClassCodeAlternateISA; - } - else - { - if (symbol_type == eSymbolTypeCode) - m_address_class_map[symbol.st_value] = eAddressClassCode; - else if (symbol_type == eSymbolTypeData) - m_address_class_map[symbol.st_value] = eAddressClassData; - else - m_address_class_map[symbol.st_value] = eAddressClassUnknown; - } + } else if (arch.GetMachine() == llvm::Triple::aarch64) { + if (symbol.getBinding() == STB_LOCAL) { + char mapping_symbol = FindArmAarch64MappingSymbol(symbol_name); + if (symbol_type == eSymbolTypeCode) { + switch (mapping_symbol) { + case 'x': + // $x[.<any>]* - marks an A64 instruction sequence + m_address_class_map[symbol.st_value] = eAddressClassCode; + break; + case 'd': + // $d[.<any>]* - marks a data item sequence (e.g. lit pool) + m_address_class_map[symbol.st_value] = eAddressClassData; + break; } + } + if (mapping_symbol) + continue; } - - // symbol_value_offset may contain 0 for ARM symbols or -1 for THUMB symbols. See above for - // more details. - uint64_t symbol_value = symbol.st_value + symbol_value_offset; - - if (symbol_section_sp == nullptr && section_idx == SHN_ABS && symbol.st_size != 0) - { - // We don't have a section for a symbol with non-zero size. Create a new section for it - // so the address range covered by the symbol is also covered by the module (represented - // through the section list). It is needed so module lookup for the addresses covered - // by this symbol will be successfull. This case happens for absolute symbols. - ConstString fake_section_name(std::string(".absolute.") + symbol_name); - symbol_section_sp = std::make_shared<Section>(module_sp, - this, - SHN_ABS, - fake_section_name, - eSectionTypeAbsoluteAddress, - symbol_value, - symbol.st_size, - 0, 0, 0, - SHF_ALLOC); - - module_section_list->AddSection(symbol_section_sp); - section_list->AddSection(symbol_section_sp); + } + + if (arch.GetMachine() == llvm::Triple::arm) { + if (symbol_type == eSymbolTypeCode) { + if (symbol.st_value & 1) { + // Subtracting 1 from the address effectively unsets + // the low order bit, which results in the address + // actually pointing to the beginning of the symbol. + // This delta will be used below in conjunction with + // symbol.st_value to produce the final symbol_value + // that we store in the symtab. + symbol_value_offset = -1; + m_address_class_map[symbol.st_value ^ 1] = + eAddressClassCodeAlternateISA; + } else { + // This address is ARM + m_address_class_map[symbol.st_value] = eAddressClassCode; + } } - - if (symbol_section_sp && CalculateType() != ObjectFile::Type::eTypeObjectFile) - symbol_value -= symbol_section_sp->GetFileAddress(); - - if (symbol_section_sp && module_section_list && module_section_list != section_list) - { - const ConstString §_name = symbol_section_sp->GetName(); - auto section_it = section_name_to_section.find(sect_name.GetCString()); - if (section_it == section_name_to_section.end()) - section_it = section_name_to_section.emplace( - sect_name.GetCString(), - module_section_list->FindSectionByName (sect_name)).first; - if (section_it->second && section_it->second->GetFileSize()) - symbol_section_sp = section_it->second; + } + + /* + * MIPS: + * The bit #0 of an address is used for ISA mode (1 for microMIPS, 0 for + * MIPS). + * This allows processor to switch between microMIPS and MIPS without any + * need + * for special mode-control register. However, apart from .debug_line, + * none of + * the ELF/DWARF sections set the ISA bit (for symbol or section). Use + * st_other + * flag to check whether the symbol is microMIPS and then set the address + * class + * accordingly. + */ + const llvm::Triple::ArchType llvm_arch = arch.GetMachine(); + if (llvm_arch == llvm::Triple::mips || + llvm_arch == llvm::Triple::mipsel || + llvm_arch == llvm::Triple::mips64 || + llvm_arch == llvm::Triple::mips64el) { + if (IS_MICROMIPS(symbol.st_other)) + m_address_class_map[symbol.st_value] = eAddressClassCodeAlternateISA; + else if ((symbol.st_value & 1) && (symbol_type == eSymbolTypeCode)) { + symbol.st_value = symbol.st_value & (~1ull); + m_address_class_map[symbol.st_value] = eAddressClassCodeAlternateISA; + } else { + if (symbol_type == eSymbolTypeCode) + m_address_class_map[symbol.st_value] = eAddressClassCode; + else if (symbol_type == eSymbolTypeData) + m_address_class_map[symbol.st_value] = eAddressClassData; + else + m_address_class_map[symbol.st_value] = eAddressClassUnknown; } + } + } + + // symbol_value_offset may contain 0 for ARM symbols or -1 for THUMB + // symbols. See above for + // more details. + uint64_t symbol_value = symbol.st_value + symbol_value_offset; + + if (symbol_section_sp == nullptr && section_idx == SHN_ABS && + symbol.st_size != 0) { + // We don't have a section for a symbol with non-zero size. Create a new + // section for it + // so the address range covered by the symbol is also covered by the + // module (represented + // through the section list). It is needed so module lookup for the + // addresses covered + // by this symbol will be successfull. This case happens for absolute + // symbols. + ConstString fake_section_name(std::string(".absolute.") + symbol_name); + symbol_section_sp = + std::make_shared<Section>(module_sp, this, SHN_ABS, fake_section_name, + eSectionTypeAbsoluteAddress, symbol_value, + symbol.st_size, 0, 0, 0, SHF_ALLOC); + + module_section_list->AddSection(symbol_section_sp); + section_list->AddSection(symbol_section_sp); + } + + if (symbol_section_sp && + CalculateType() != ObjectFile::Type::eTypeObjectFile) + symbol_value -= symbol_section_sp->GetFileAddress(); + + if (symbol_section_sp && module_section_list && + module_section_list != section_list) { + const ConstString §_name = symbol_section_sp->GetName(); + auto section_it = section_name_to_section.find(sect_name.GetCString()); + if (section_it == section_name_to_section.end()) + section_it = + section_name_to_section + .emplace(sect_name.GetCString(), + module_section_list->FindSectionByName(sect_name)) + .first; + if (section_it->second && section_it->second->GetFileSize()) + symbol_section_sp = section_it->second; + } + + bool is_global = symbol.getBinding() == STB_GLOBAL; + uint32_t flags = symbol.st_other << 8 | symbol.st_info | additional_flags; + bool is_mangled = (symbol_name[0] == '_' && symbol_name[1] == 'Z'); + + llvm::StringRef symbol_ref(symbol_name); + + // Symbol names may contain @VERSION suffixes. Find those and strip them + // temporarily. + size_t version_pos = symbol_ref.find('@'); + bool has_suffix = version_pos != llvm::StringRef::npos; + llvm::StringRef symbol_bare = symbol_ref.substr(0, version_pos); + Mangled mangled(ConstString(symbol_bare), is_mangled); + + // Now append the suffix back to mangled and unmangled names. Only do it if + // the + // demangling was successful (string is not empty). + if (has_suffix) { + llvm::StringRef suffix = symbol_ref.substr(version_pos); + + llvm::StringRef mangled_name = mangled.GetMangledName().GetStringRef(); + if (!mangled_name.empty()) + mangled.SetMangledName(ConstString((mangled_name + suffix).str())); + + ConstString demangled = + mangled.GetDemangledName(lldb::eLanguageTypeUnknown); + llvm::StringRef demangled_name = demangled.GetStringRef(); + if (!demangled_name.empty()) + mangled.SetDemangledName(ConstString((demangled_name + suffix).str())); + } + + // In ELF all symbol should have a valid size but it is not true for some + // function symbols + // coming from hand written assembly. As none of the function symbol should + // have 0 size we + // try to calculate the size for these symbols in the symtab with saying + // that their original + // size is not valid. + bool symbol_size_valid = + symbol.st_size != 0 || symbol.getType() != STT_FUNC; + + Symbol dc_symbol( + i + start_id, // ID is the original symbol table index. + mangled, + symbol_type, // Type of this symbol + is_global, // Is this globally visible? + false, // Is this symbol debug info? + false, // Is this symbol a trampoline? + false, // Is this symbol artificial? + AddressRange(symbol_section_sp, // Section in which this symbol is + // defined or null. + symbol_value, // Offset in section or symbol value. + symbol.st_size), // Size in bytes of this symbol. + symbol_size_valid, // Symbol size is valid + has_suffix, // Contains linker annotations? + flags); // Symbol flags. + symtab->AddSymbol(dc_symbol); + } + return i; +} + +unsigned ObjectFileELF::ParseSymbolTable(Symtab *symbol_table, + user_id_t start_id, + lldb_private::Section *symtab) { + if (symtab->GetObjectFile() != this) { + // If the symbol table section is owned by a different object file, have it + // do the + // parsing. + ObjectFileELF *obj_file_elf = + static_cast<ObjectFileELF *>(symtab->GetObjectFile()); + return obj_file_elf->ParseSymbolTable(symbol_table, start_id, symtab); + } + + // Get section list for this object file. + SectionList *section_list = m_sections_ap.get(); + if (!section_list) + return 0; - bool is_global = symbol.getBinding() == STB_GLOBAL; - uint32_t flags = symbol.st_other << 8 | symbol.st_info | additional_flags; - bool is_mangled = symbol_name ? (symbol_name[0] == '_' && symbol_name[1] == 'Z') : false; - - llvm::StringRef symbol_ref(symbol_name); - - // Symbol names may contain @VERSION suffixes. Find those and strip them temporarily. - size_t version_pos = symbol_ref.find('@'); - bool has_suffix = version_pos != llvm::StringRef::npos; - llvm::StringRef symbol_bare = symbol_ref.substr(0, version_pos); - Mangled mangled(ConstString(symbol_bare), is_mangled); + user_id_t symtab_id = symtab->GetID(); + const ELFSectionHeaderInfo *symtab_hdr = GetSectionHeaderByIndex(symtab_id); + assert(symtab_hdr->sh_type == SHT_SYMTAB || + symtab_hdr->sh_type == SHT_DYNSYM); - // Now append the suffix back to mangled and unmangled names. Only do it if the - // demangling was successful (string is not empty). - if (has_suffix) - { - llvm::StringRef suffix = symbol_ref.substr(version_pos); + // sh_link: section header index of associated string table. + // Section ID's are ones based. + user_id_t strtab_id = symtab_hdr->sh_link + 1; + Section *strtab = section_list->FindSectionByID(strtab_id).get(); - llvm::StringRef mangled_name = mangled.GetMangledName().GetStringRef(); - if (! mangled_name.empty()) - mangled.SetMangledName( ConstString((mangled_name + suffix).str()) ); + if (symtab && strtab) { + assert(symtab->GetObjectFile() == this); + assert(strtab->GetObjectFile() == this); - ConstString demangled = mangled.GetDemangledName(lldb::eLanguageTypeUnknown); - llvm::StringRef demangled_name = demangled.GetStringRef(); - if (!demangled_name.empty()) - mangled.SetDemangledName( ConstString((demangled_name + suffix).str()) ); - } + DataExtractor symtab_data; + DataExtractor strtab_data; + if (ReadSectionData(symtab, symtab_data) && + ReadSectionData(strtab, strtab_data)) { + size_t num_symbols = symtab_data.GetByteSize() / symtab_hdr->sh_entsize; - // In ELF all symbol should have a valid size but it is not true for some function symbols - // coming from hand written assembly. As none of the function symbol should have 0 size we - // try to calculate the size for these symbols in the symtab with saying that their original - // size is not valid. - bool symbol_size_valid = symbol.st_size != 0 || symbol.getType() != STT_FUNC; - - Symbol dc_symbol( - i + start_id, // ID is the original symbol table index. - mangled, - symbol_type, // Type of this symbol - is_global, // Is this globally visible? - false, // Is this symbol debug info? - false, // Is this symbol a trampoline? - false, // Is this symbol artificial? - AddressRange( - symbol_section_sp, // Section in which this symbol is defined or null. - symbol_value, // Offset in section or symbol value. - symbol.st_size), // Size in bytes of this symbol. - symbol_size_valid, // Symbol size is valid - has_suffix, // Contains linker annotations? - flags); // Symbol flags. - symtab->AddSymbol(dc_symbol); + return ParseSymbols(symbol_table, start_id, section_list, num_symbols, + symtab_data, strtab_data); } - return i; -} + } -unsigned -ObjectFileELF::ParseSymbolTable(Symtab *symbol_table, - user_id_t start_id, - lldb_private::Section *symtab) -{ - if (symtab->GetObjectFile() != this) - { - // If the symbol table section is owned by a different object file, have it do the - // parsing. - ObjectFileELF *obj_file_elf = static_cast<ObjectFileELF *>(symtab->GetObjectFile()); - return obj_file_elf->ParseSymbolTable (symbol_table, start_id, symtab); - } + return 0; +} - // Get section list for this object file. - SectionList *section_list = m_sections_ap.get(); - if (!section_list) - return 0; - - user_id_t symtab_id = symtab->GetID(); - const ELFSectionHeaderInfo *symtab_hdr = GetSectionHeaderByIndex(symtab_id); - assert(symtab_hdr->sh_type == SHT_SYMTAB || - symtab_hdr->sh_type == SHT_DYNSYM); - - // sh_link: section header index of associated string table. - // Section ID's are ones based. - user_id_t strtab_id = symtab_hdr->sh_link + 1; - Section *strtab = section_list->FindSectionByID(strtab_id).get(); - - if (symtab && strtab) - { - assert (symtab->GetObjectFile() == this); - assert (strtab->GetObjectFile() == this); - - DataExtractor symtab_data; - DataExtractor strtab_data; - if (ReadSectionData(symtab, symtab_data) && - ReadSectionData(strtab, strtab_data)) - { - size_t num_symbols = symtab_data.GetByteSize() / symtab_hdr->sh_entsize; - - return ParseSymbols(symbol_table, start_id, section_list, - num_symbols, symtab_data, strtab_data); - } - } +size_t ObjectFileELF::ParseDynamicSymbols() { + if (m_dynamic_symbols.size()) + return m_dynamic_symbols.size(); + SectionList *section_list = GetSectionList(); + if (!section_list) return 0; -} -size_t -ObjectFileELF::ParseDynamicSymbols() -{ - if (m_dynamic_symbols.size()) - return m_dynamic_symbols.size(); - - SectionList *section_list = GetSectionList(); - if (!section_list) - return 0; + // Find the SHT_DYNAMIC section. + Section *dynsym = + section_list->FindSectionByType(eSectionTypeELFDynamicLinkInfo, true) + .get(); + if (!dynsym) + return 0; + assert(dynsym->GetObjectFile() == this); - // Find the SHT_DYNAMIC section. - Section *dynsym = section_list->FindSectionByType (eSectionTypeELFDynamicLinkInfo, true).get(); - if (!dynsym) - return 0; - assert (dynsym->GetObjectFile() == this); + ELFDynamic symbol; + DataExtractor dynsym_data; + if (ReadSectionData(dynsym, dynsym_data)) { + const lldb::offset_t section_size = dynsym_data.GetByteSize(); + lldb::offset_t cursor = 0; - ELFDynamic symbol; - DataExtractor dynsym_data; - if (ReadSectionData(dynsym, dynsym_data)) - { - const lldb::offset_t section_size = dynsym_data.GetByteSize(); - lldb::offset_t cursor = 0; - - while (cursor < section_size) - { - if (!symbol.Parse(dynsym_data, &cursor)) - break; + while (cursor < section_size) { + if (!symbol.Parse(dynsym_data, &cursor)) + break; - m_dynamic_symbols.push_back(symbol); - } + m_dynamic_symbols.push_back(symbol); } + } - return m_dynamic_symbols.size(); + return m_dynamic_symbols.size(); } -const ELFDynamic * -ObjectFileELF::FindDynamicSymbol(unsigned tag) -{ - if (!ParseDynamicSymbols()) - return NULL; +const ELFDynamic *ObjectFileELF::FindDynamicSymbol(unsigned tag) { + if (!ParseDynamicSymbols()) + return NULL; - DynamicSymbolCollIter I = m_dynamic_symbols.begin(); - DynamicSymbolCollIter E = m_dynamic_symbols.end(); - for ( ; I != E; ++I) - { - ELFDynamic *symbol = &*I; + DynamicSymbolCollIter I = m_dynamic_symbols.begin(); + DynamicSymbolCollIter E = m_dynamic_symbols.end(); + for (; I != E; ++I) { + ELFDynamic *symbol = &*I; - if (symbol->d_tag == tag) - return symbol; - } + if (symbol->d_tag == tag) + return symbol; + } - return NULL; + return NULL; } -unsigned -ObjectFileELF::PLTRelocationType() -{ - // DT_PLTREL - // This member specifies the type of relocation entry to which the - // procedure linkage table refers. The d_val member holds DT_REL or - // DT_RELA, as appropriate. All relocations in a procedure linkage table - // must use the same relocation. - const ELFDynamic *symbol = FindDynamicSymbol(DT_PLTREL); - - if (symbol) - return symbol->d_val; +unsigned ObjectFileELF::PLTRelocationType() { + // DT_PLTREL + // This member specifies the type of relocation entry to which the + // procedure linkage table refers. The d_val member holds DT_REL or + // DT_RELA, as appropriate. All relocations in a procedure linkage table + // must use the same relocation. + const ELFDynamic *symbol = FindDynamicSymbol(DT_PLTREL); - return 0; + if (symbol) + return symbol->d_val; + + return 0; } -// Returns the size of the normal plt entries and the offset of the first normal plt entry. The -// 0th entry in the plt table is usually a resolution entry which have different size in some +// Returns the size of the normal plt entries and the offset of the first normal +// plt entry. The +// 0th entry in the plt table is usually a resolution entry which have different +// size in some // architectures then the rest of the plt entries. static std::pair<uint64_t, uint64_t> -GetPltEntrySizeAndOffset(const ELFSectionHeader* rel_hdr, const ELFSectionHeader* plt_hdr) -{ - const elf_xword num_relocations = rel_hdr->sh_size / rel_hdr->sh_entsize; - - // Clang 3.3 sets entsize to 4 for 32-bit binaries, but the plt entries are 16 bytes. - // So round the entsize up by the alignment if addralign is set. - elf_xword plt_entsize = plt_hdr->sh_addralign ? - llvm::alignTo (plt_hdr->sh_entsize, plt_hdr->sh_addralign) : plt_hdr->sh_entsize; - - // Some linkers e.g ld for arm, fill plt_hdr->sh_entsize field incorrectly. - // PLT entries relocation code in general requires multiple instruction and - // should be greater than 4 bytes in most cases. Try to guess correct size just in case. - if (plt_entsize <= 4) - { - // The linker haven't set the plt_hdr->sh_entsize field. Try to guess the size of the plt - // entries based on the number of entries and the size of the plt section with the - // assumption that the size of the 0th entry is at least as big as the size of the normal - // entries and it isn't much bigger then that. - if (plt_hdr->sh_addralign) - plt_entsize = plt_hdr->sh_size / plt_hdr->sh_addralign / (num_relocations + 1) * plt_hdr->sh_addralign; - else - plt_entsize = plt_hdr->sh_size / (num_relocations + 1); - } - - elf_xword plt_offset = plt_hdr->sh_size - num_relocations * plt_entsize; - - return std::make_pair(plt_entsize, plt_offset); -} - -static unsigned -ParsePLTRelocations(Symtab *symbol_table, - user_id_t start_id, - unsigned rel_type, - const ELFHeader *hdr, - const ELFSectionHeader *rel_hdr, - const ELFSectionHeader *plt_hdr, - const ELFSectionHeader *sym_hdr, - const lldb::SectionSP &plt_section_sp, - DataExtractor &rel_data, - DataExtractor &symtab_data, - DataExtractor &strtab_data) -{ - ELFRelocation rel(rel_type); - ELFSymbol symbol; - lldb::offset_t offset = 0; - - uint64_t plt_offset, plt_entsize; - std::tie(plt_entsize, plt_offset) = GetPltEntrySizeAndOffset(rel_hdr, plt_hdr); - const elf_xword num_relocations = rel_hdr->sh_size / rel_hdr->sh_entsize; - - typedef unsigned (*reloc_info_fn)(const ELFRelocation &rel); - reloc_info_fn reloc_type; - reloc_info_fn reloc_symbol; - - if (hdr->Is32Bit()) - { - reloc_type = ELFRelocation::RelocType32; - reloc_symbol = ELFRelocation::RelocSymbol32; - } +GetPltEntrySizeAndOffset(const ELFSectionHeader *rel_hdr, + const ELFSectionHeader *plt_hdr) { + const elf_xword num_relocations = rel_hdr->sh_size / rel_hdr->sh_entsize; + + // Clang 3.3 sets entsize to 4 for 32-bit binaries, but the plt entries are 16 + // bytes. + // So round the entsize up by the alignment if addralign is set. + elf_xword plt_entsize = + plt_hdr->sh_addralign + ? llvm::alignTo(plt_hdr->sh_entsize, plt_hdr->sh_addralign) + : plt_hdr->sh_entsize; + + // Some linkers e.g ld for arm, fill plt_hdr->sh_entsize field incorrectly. + // PLT entries relocation code in general requires multiple instruction and + // should be greater than 4 bytes in most cases. Try to guess correct size + // just in case. + if (plt_entsize <= 4) { + // The linker haven't set the plt_hdr->sh_entsize field. Try to guess the + // size of the plt + // entries based on the number of entries and the size of the plt section + // with the + // assumption that the size of the 0th entry is at least as big as the size + // of the normal + // entries and it isn't much bigger then that. + if (plt_hdr->sh_addralign) + plt_entsize = plt_hdr->sh_size / plt_hdr->sh_addralign / + (num_relocations + 1) * plt_hdr->sh_addralign; else - { - reloc_type = ELFRelocation::RelocType64; - reloc_symbol = ELFRelocation::RelocSymbol64; - } - - unsigned slot_type = hdr->GetRelocationJumpSlotType(); - unsigned i; - for (i = 0; i < num_relocations; ++i) - { - if (rel.Parse(rel_data, &offset) == false) - break; - - if (reloc_type(rel) != slot_type) - continue; - - lldb::offset_t symbol_offset = reloc_symbol(rel) * sym_hdr->sh_entsize; - if (!symbol.Parse(symtab_data, &symbol_offset)) - break; - - const char *symbol_name = strtab_data.PeekCStr(symbol.st_name); - bool is_mangled = symbol_name ? (symbol_name[0] == '_' && symbol_name[1] == 'Z') : false; - uint64_t plt_index = plt_offset + i * plt_entsize; - - Symbol jump_symbol( - i + start_id, // Symbol table index - symbol_name, // symbol name. - is_mangled, // is the symbol name mangled? - eSymbolTypeTrampoline, // Type of this symbol - false, // Is this globally visible? - false, // Is this symbol debug info? - true, // Is this symbol a trampoline? - true, // Is this symbol artificial? - plt_section_sp, // Section in which this symbol is defined or null. - plt_index, // Offset in section or symbol value. - plt_entsize, // Size in bytes of this symbol. - true, // Size is valid - false, // Contains linker annotations? - 0); // Symbol flags. - - symbol_table->AddSymbol(jump_symbol); - } - - return i; + plt_entsize = plt_hdr->sh_size / (num_relocations + 1); + } + + elf_xword plt_offset = plt_hdr->sh_size - num_relocations * plt_entsize; + + return std::make_pair(plt_entsize, plt_offset); +} + +static unsigned ParsePLTRelocations( + Symtab *symbol_table, user_id_t start_id, unsigned rel_type, + const ELFHeader *hdr, const ELFSectionHeader *rel_hdr, + const ELFSectionHeader *plt_hdr, const ELFSectionHeader *sym_hdr, + const lldb::SectionSP &plt_section_sp, DataExtractor &rel_data, + DataExtractor &symtab_data, DataExtractor &strtab_data) { + ELFRelocation rel(rel_type); + ELFSymbol symbol; + lldb::offset_t offset = 0; + + uint64_t plt_offset, plt_entsize; + std::tie(plt_entsize, plt_offset) = + GetPltEntrySizeAndOffset(rel_hdr, plt_hdr); + const elf_xword num_relocations = rel_hdr->sh_size / rel_hdr->sh_entsize; + + typedef unsigned (*reloc_info_fn)(const ELFRelocation &rel); + reloc_info_fn reloc_type; + reloc_info_fn reloc_symbol; + + if (hdr->Is32Bit()) { + reloc_type = ELFRelocation::RelocType32; + reloc_symbol = ELFRelocation::RelocSymbol32; + } else { + reloc_type = ELFRelocation::RelocType64; + reloc_symbol = ELFRelocation::RelocSymbol64; + } + + unsigned slot_type = hdr->GetRelocationJumpSlotType(); + unsigned i; + for (i = 0; i < num_relocations; ++i) { + if (rel.Parse(rel_data, &offset) == false) + break; + + if (reloc_type(rel) != slot_type) + continue; + + lldb::offset_t symbol_offset = reloc_symbol(rel) * sym_hdr->sh_entsize; + if (!symbol.Parse(symtab_data, &symbol_offset)) + break; + + const char *symbol_name = strtab_data.PeekCStr(symbol.st_name); + bool is_mangled = + symbol_name ? (symbol_name[0] == '_' && symbol_name[1] == 'Z') : false; + uint64_t plt_index = plt_offset + i * plt_entsize; + + Symbol jump_symbol( + i + start_id, // Symbol table index + symbol_name, // symbol name. + is_mangled, // is the symbol name mangled? + eSymbolTypeTrampoline, // Type of this symbol + false, // Is this globally visible? + false, // Is this symbol debug info? + true, // Is this symbol a trampoline? + true, // Is this symbol artificial? + plt_section_sp, // Section in which this symbol is defined or null. + plt_index, // Offset in section or symbol value. + plt_entsize, // Size in bytes of this symbol. + true, // Size is valid + false, // Contains linker annotations? + 0); // Symbol flags. + + symbol_table->AddSymbol(jump_symbol); + } + + return i; } unsigned -ObjectFileELF::ParseTrampolineSymbols(Symtab *symbol_table, - user_id_t start_id, +ObjectFileELF::ParseTrampolineSymbols(Symtab *symbol_table, user_id_t start_id, const ELFSectionHeaderInfo *rel_hdr, - user_id_t rel_id) -{ - assert(rel_hdr->sh_type == SHT_RELA || rel_hdr->sh_type == SHT_REL); - - // The link field points to the associated symbol table. - user_id_t symtab_id = rel_hdr->sh_link; - - // If the link field doesn't point to the appropriate symbol name table then - // try to find it by name as some compiler don't fill in the link fields. - if (!symtab_id) - symtab_id = GetSectionIndexByName(".dynsym"); + user_id_t rel_id) { + assert(rel_hdr->sh_type == SHT_RELA || rel_hdr->sh_type == SHT_REL); - // Get PLT section. We cannot use rel_hdr->sh_info, since current linkers - // point that to the .got.plt or .got section instead of .plt. - user_id_t plt_id = GetSectionIndexByName(".plt"); + // The link field points to the associated symbol table. + user_id_t symtab_id = rel_hdr->sh_link; - if (!symtab_id || !plt_id) - return 0; + // If the link field doesn't point to the appropriate symbol name table then + // try to find it by name as some compiler don't fill in the link fields. + if (!symtab_id) + symtab_id = GetSectionIndexByName(".dynsym"); - // Section ID's are ones based; - symtab_id++; - plt_id++; + // Get PLT section. We cannot use rel_hdr->sh_info, since current linkers + // point that to the .got.plt or .got section instead of .plt. + user_id_t plt_id = GetSectionIndexByName(".plt"); - const ELFSectionHeaderInfo *plt_hdr = GetSectionHeaderByIndex(plt_id); - if (!plt_hdr) - return 0; + if (!symtab_id || !plt_id) + return 0; - const ELFSectionHeaderInfo *sym_hdr = GetSectionHeaderByIndex(symtab_id); - if (!sym_hdr) - return 0; + // Section ID's are ones based; + symtab_id++; + plt_id++; - SectionList *section_list = m_sections_ap.get(); - if (!section_list) - return 0; + const ELFSectionHeaderInfo *plt_hdr = GetSectionHeaderByIndex(plt_id); + if (!plt_hdr) + return 0; - Section *rel_section = section_list->FindSectionByID(rel_id).get(); - if (!rel_section) - return 0; + const ELFSectionHeaderInfo *sym_hdr = GetSectionHeaderByIndex(symtab_id); + if (!sym_hdr) + return 0; - SectionSP plt_section_sp (section_list->FindSectionByID(plt_id)); - if (!plt_section_sp) - return 0; + SectionList *section_list = m_sections_ap.get(); + if (!section_list) + return 0; - Section *symtab = section_list->FindSectionByID(symtab_id).get(); - if (!symtab) - return 0; + Section *rel_section = section_list->FindSectionByID(rel_id).get(); + if (!rel_section) + return 0; - // sh_link points to associated string table. - Section *strtab = section_list->FindSectionByID(sym_hdr->sh_link + 1).get(); - if (!strtab) - return 0; + SectionSP plt_section_sp(section_list->FindSectionByID(plt_id)); + if (!plt_section_sp) + return 0; - DataExtractor rel_data; - if (!ReadSectionData(rel_section, rel_data)) - return 0; + Section *symtab = section_list->FindSectionByID(symtab_id).get(); + if (!symtab) + return 0; - DataExtractor symtab_data; - if (!ReadSectionData(symtab, symtab_data)) - return 0; + // sh_link points to associated string table. + Section *strtab = section_list->FindSectionByID(sym_hdr->sh_link + 1).get(); + if (!strtab) + return 0; - DataExtractor strtab_data; - if (!ReadSectionData(strtab, strtab_data)) - return 0; - - unsigned rel_type = PLTRelocationType(); - if (!rel_type) - return 0; - - return ParsePLTRelocations (symbol_table, - start_id, - rel_type, - &m_header, - rel_hdr, - plt_hdr, - sym_hdr, - plt_section_sp, - rel_data, - symtab_data, - strtab_data); -} + DataExtractor rel_data; + if (!ReadSectionData(rel_section, rel_data)) + return 0; -unsigned -ObjectFileELF::RelocateSection(Symtab* symtab, const ELFHeader *hdr, const ELFSectionHeader *rel_hdr, - const ELFSectionHeader *symtab_hdr, const ELFSectionHeader *debug_hdr, - DataExtractor &rel_data, DataExtractor &symtab_data, - DataExtractor &debug_data, Section* rel_section) -{ - ELFRelocation rel(rel_hdr->sh_type); - lldb::addr_t offset = 0; - const unsigned num_relocations = rel_hdr->sh_size / rel_hdr->sh_entsize; - typedef unsigned (*reloc_info_fn)(const ELFRelocation &rel); - reloc_info_fn reloc_type; - reloc_info_fn reloc_symbol; - - if (hdr->Is32Bit()) - { - reloc_type = ELFRelocation::RelocType32; - reloc_symbol = ELFRelocation::RelocSymbol32; - } - else - { - reloc_type = ELFRelocation::RelocType64; - reloc_symbol = ELFRelocation::RelocSymbol64; - } + DataExtractor symtab_data; + if (!ReadSectionData(symtab, symtab_data)) + return 0; - for (unsigned i = 0; i < num_relocations; ++i) - { - if (rel.Parse(rel_data, &offset) == false) - break; + DataExtractor strtab_data; + if (!ReadSectionData(strtab, strtab_data)) + return 0; - Symbol* symbol = NULL; + unsigned rel_type = PLTRelocationType(); + if (!rel_type) + return 0; - if (hdr->Is32Bit()) - { - switch (reloc_type(rel)) { - case R_386_32: - case R_386_PC32: - default: - assert(false && "unexpected relocation type"); - } - } else { - switch (reloc_type(rel)) { - case R_X86_64_64: - { - symbol = symtab->FindSymbolByID(reloc_symbol(rel)); - if (symbol) - { - addr_t value = symbol->GetAddressRef().GetFileAddress(); - DataBufferSP& data_buffer_sp = debug_data.GetSharedDataBuffer(); - uint64_t* dst = reinterpret_cast<uint64_t*>(data_buffer_sp->GetBytes() + rel_section->GetFileOffset() + ELFRelocation::RelocOffset64(rel)); - *dst = value + ELFRelocation::RelocAddend64(rel); - } - break; - } - case R_X86_64_32: - case R_X86_64_32S: - { - symbol = symtab->FindSymbolByID(reloc_symbol(rel)); - if (symbol) - { - addr_t value = symbol->GetAddressRef().GetFileAddress(); - value += ELFRelocation::RelocAddend32(rel); - assert((reloc_type(rel) == R_X86_64_32 && (value <= UINT32_MAX)) || - (reloc_type(rel) == R_X86_64_32S && - ((int64_t)value <= INT32_MAX && (int64_t)value >= INT32_MIN))); - uint32_t truncated_addr = (value & 0xFFFFFFFF); - DataBufferSP& data_buffer_sp = debug_data.GetSharedDataBuffer(); - uint32_t* dst = reinterpret_cast<uint32_t*>(data_buffer_sp->GetBytes() + rel_section->GetFileOffset() + ELFRelocation::RelocOffset32(rel)); - *dst = truncated_addr; - } - break; - } - case R_X86_64_PC32: - default: - assert(false && "unexpected relocation type"); - } + return ParsePLTRelocations(symbol_table, start_id, rel_type, &m_header, + rel_hdr, plt_hdr, sym_hdr, plt_section_sp, + rel_data, symtab_data, strtab_data); +} + +unsigned ObjectFileELF::RelocateSection( + Symtab *symtab, const ELFHeader *hdr, const ELFSectionHeader *rel_hdr, + const ELFSectionHeader *symtab_hdr, const ELFSectionHeader *debug_hdr, + DataExtractor &rel_data, DataExtractor &symtab_data, + DataExtractor &debug_data, Section *rel_section) { + ELFRelocation rel(rel_hdr->sh_type); + lldb::addr_t offset = 0; + const unsigned num_relocations = rel_hdr->sh_size / rel_hdr->sh_entsize; + typedef unsigned (*reloc_info_fn)(const ELFRelocation &rel); + reloc_info_fn reloc_type; + reloc_info_fn reloc_symbol; + + if (hdr->Is32Bit()) { + reloc_type = ELFRelocation::RelocType32; + reloc_symbol = ELFRelocation::RelocSymbol32; + } else { + reloc_type = ELFRelocation::RelocType64; + reloc_symbol = ELFRelocation::RelocSymbol64; + } + + for (unsigned i = 0; i < num_relocations; ++i) { + if (rel.Parse(rel_data, &offset) == false) + break; + + Symbol *symbol = NULL; + + if (hdr->Is32Bit()) { + switch (reloc_type(rel)) { + case R_386_32: + case R_386_PC32: + default: + assert(false && "unexpected relocation type"); + } + } else { + switch (reloc_type(rel)) { + case R_X86_64_64: { + symbol = symtab->FindSymbolByID(reloc_symbol(rel)); + if (symbol) { + addr_t value = symbol->GetAddressRef().GetFileAddress(); + DataBufferSP &data_buffer_sp = debug_data.GetSharedDataBuffer(); + uint64_t *dst = reinterpret_cast<uint64_t *>( + data_buffer_sp->GetBytes() + rel_section->GetFileOffset() + + ELFRelocation::RelocOffset64(rel)); + *dst = value + ELFRelocation::RelocAddend64(rel); } + break; + } + case R_X86_64_32: + case R_X86_64_32S: { + symbol = symtab->FindSymbolByID(reloc_symbol(rel)); + if (symbol) { + addr_t value = symbol->GetAddressRef().GetFileAddress(); + value += ELFRelocation::RelocAddend32(rel); + assert( + (reloc_type(rel) == R_X86_64_32 && (value <= UINT32_MAX)) || + (reloc_type(rel) == R_X86_64_32S && + ((int64_t)value <= INT32_MAX && (int64_t)value >= INT32_MIN))); + uint32_t truncated_addr = (value & 0xFFFFFFFF); + DataBufferSP &data_buffer_sp = debug_data.GetSharedDataBuffer(); + uint32_t *dst = reinterpret_cast<uint32_t *>( + data_buffer_sp->GetBytes() + rel_section->GetFileOffset() + + ELFRelocation::RelocOffset32(rel)); + *dst = truncated_addr; + } + break; + } + case R_X86_64_PC32: + default: + assert(false && "unexpected relocation type"); + } } + } - return 0; + return 0; } -unsigned -ObjectFileELF::RelocateDebugSections(const ELFSectionHeader *rel_hdr, user_id_t rel_id) -{ - assert(rel_hdr->sh_type == SHT_RELA || rel_hdr->sh_type == SHT_REL); +unsigned ObjectFileELF::RelocateDebugSections(const ELFSectionHeader *rel_hdr, + user_id_t rel_id) { + assert(rel_hdr->sh_type == SHT_RELA || rel_hdr->sh_type == SHT_REL); - // Parse in the section list if needed. - SectionList *section_list = GetSectionList(); - if (!section_list) - return 0; + // Parse in the section list if needed. + SectionList *section_list = GetSectionList(); + if (!section_list) + return 0; - // Section ID's are ones based. - user_id_t symtab_id = rel_hdr->sh_link + 1; - user_id_t debug_id = rel_hdr->sh_info + 1; + // Section ID's are ones based. + user_id_t symtab_id = rel_hdr->sh_link + 1; + user_id_t debug_id = rel_hdr->sh_info + 1; - const ELFSectionHeader *symtab_hdr = GetSectionHeaderByIndex(symtab_id); - if (!symtab_hdr) - return 0; + const ELFSectionHeader *symtab_hdr = GetSectionHeaderByIndex(symtab_id); + if (!symtab_hdr) + return 0; - const ELFSectionHeader *debug_hdr = GetSectionHeaderByIndex(debug_id); - if (!debug_hdr) - return 0; + const ELFSectionHeader *debug_hdr = GetSectionHeaderByIndex(debug_id); + if (!debug_hdr) + return 0; - Section *rel = section_list->FindSectionByID(rel_id).get(); - if (!rel) - return 0; + Section *rel = section_list->FindSectionByID(rel_id).get(); + if (!rel) + return 0; - Section *symtab = section_list->FindSectionByID(symtab_id).get(); - if (!symtab) - return 0; + Section *symtab = section_list->FindSectionByID(symtab_id).get(); + if (!symtab) + return 0; - Section *debug = section_list->FindSectionByID(debug_id).get(); - if (!debug) - return 0; + Section *debug = section_list->FindSectionByID(debug_id).get(); + if (!debug) + return 0; - DataExtractor rel_data; - DataExtractor symtab_data; - DataExtractor debug_data; - - if (ReadSectionData(rel, rel_data) && - ReadSectionData(symtab, symtab_data) && - ReadSectionData(debug, debug_data)) - { - RelocateSection(m_symtab_ap.get(), &m_header, rel_hdr, symtab_hdr, debug_hdr, - rel_data, symtab_data, debug_data, debug); - } + DataExtractor rel_data; + DataExtractor symtab_data; + DataExtractor debug_data; - return 0; + if (ReadSectionData(rel, rel_data) && ReadSectionData(symtab, symtab_data) && + ReadSectionData(debug, debug_data)) { + RelocateSection(m_symtab_ap.get(), &m_header, rel_hdr, symtab_hdr, + debug_hdr, rel_data, symtab_data, debug_data, debug); + } + + return 0; } -Symtab * -ObjectFileELF::GetSymtab() -{ - ModuleSP module_sp(GetModule()); - if (!module_sp) - return NULL; - - // We always want to use the main object file so we (hopefully) only have one cached copy - // of our symtab, dynamic sections, etc. - ObjectFile *module_obj_file = module_sp->GetObjectFile(); - if (module_obj_file && module_obj_file != this) - return module_obj_file->GetSymtab(); - - if (m_symtab_ap.get() == NULL) - { - SectionList *section_list = module_sp->GetSectionList(); - if (!section_list) - return NULL; - - uint64_t symbol_id = 0; - std::lock_guard<std::recursive_mutex> guard(module_sp->GetMutex()); - - // Sharable objects and dynamic executables usually have 2 distinct symbol - // tables, one named ".symtab", and the other ".dynsym". The dynsym is a smaller - // version of the symtab that only contains global symbols. The information found - // in the dynsym is therefore also found in the symtab, while the reverse is not - // necessarily true. - Section *symtab = section_list->FindSectionByType (eSectionTypeELFSymbolTable, true).get(); - if (!symtab) - { - // The symtab section is non-allocable and can be stripped, so if it doesn't exist - // then use the dynsym section which should always be there. - symtab = section_list->FindSectionByType (eSectionTypeELFDynamicSymbols, true).get(); - } - if (symtab) - { - m_symtab_ap.reset(new Symtab(symtab->GetObjectFile())); - symbol_id += ParseSymbolTable (m_symtab_ap.get(), symbol_id, symtab); - } +Symtab *ObjectFileELF::GetSymtab() { + ModuleSP module_sp(GetModule()); + if (!module_sp) + return NULL; - // DT_JMPREL - // If present, this entry's d_ptr member holds the address of relocation - // entries associated solely with the procedure linkage table. Separating - // these relocation entries lets the dynamic linker ignore them during - // process initialization, if lazy binding is enabled. If this entry is - // present, the related entries of types DT_PLTRELSZ and DT_PLTREL must - // also be present. - const ELFDynamic *symbol = FindDynamicSymbol(DT_JMPREL); - if (symbol) - { - // Synthesize trampoline symbols to help navigate the PLT. - addr_t addr = symbol->d_ptr; - Section *reloc_section = section_list->FindSectionContainingFileAddress(addr).get(); - if (reloc_section) - { - user_id_t reloc_id = reloc_section->GetID(); - const ELFSectionHeaderInfo *reloc_header = GetSectionHeaderByIndex(reloc_id); - assert(reloc_header); - - if (m_symtab_ap == nullptr) - m_symtab_ap.reset(new Symtab(reloc_section->GetObjectFile())); - - ParseTrampolineSymbols (m_symtab_ap.get(), symbol_id, reloc_header, reloc_id); - } - } + // We always want to use the main object file so we (hopefully) only have one + // cached copy + // of our symtab, dynamic sections, etc. + ObjectFile *module_obj_file = module_sp->GetObjectFile(); + if (module_obj_file && module_obj_file != this) + return module_obj_file->GetSymtab(); - DWARFCallFrameInfo* eh_frame = GetUnwindTable().GetEHFrameInfo(); - if (eh_frame) - { - if (m_symtab_ap == nullptr) - m_symtab_ap.reset(new Symtab(this)); - ParseUnwindSymbols (m_symtab_ap.get(), eh_frame); - } + if (m_symtab_ap.get() == NULL) { + SectionList *section_list = module_sp->GetSectionList(); + if (!section_list) + return NULL; - // If we still don't have any symtab then create an empty instance to avoid do the section - // lookup next time. - if (m_symtab_ap == nullptr) - m_symtab_ap.reset(new Symtab(this)); + uint64_t symbol_id = 0; + std::lock_guard<std::recursive_mutex> guard(module_sp->GetMutex()); - m_symtab_ap->CalculateSymbolSizes(); - } + // Sharable objects and dynamic executables usually have 2 distinct symbol + // tables, one named ".symtab", and the other ".dynsym". The dynsym is a + // smaller + // version of the symtab that only contains global symbols. The information + // found + // in the dynsym is therefore also found in the symtab, while the reverse is + // not + // necessarily true. + Section *symtab = + section_list->FindSectionByType(eSectionTypeELFSymbolTable, true).get(); + if (!symtab) { + // The symtab section is non-allocable and can be stripped, so if it + // doesn't exist + // then use the dynsym section which should always be there. + symtab = + section_list->FindSectionByType(eSectionTypeELFDynamicSymbols, true) + .get(); + } + if (symtab) { + m_symtab_ap.reset(new Symtab(symtab->GetObjectFile())); + symbol_id += ParseSymbolTable(m_symtab_ap.get(), symbol_id, symtab); + } + + // DT_JMPREL + // If present, this entry's d_ptr member holds the address of + // relocation + // entries associated solely with the procedure linkage table. + // Separating + // these relocation entries lets the dynamic linker ignore them during + // process initialization, if lazy binding is enabled. If this entry is + // present, the related entries of types DT_PLTRELSZ and DT_PLTREL must + // also be present. + const ELFDynamic *symbol = FindDynamicSymbol(DT_JMPREL); + if (symbol) { + // Synthesize trampoline symbols to help navigate the PLT. + addr_t addr = symbol->d_ptr; + Section *reloc_section = + section_list->FindSectionContainingFileAddress(addr).get(); + if (reloc_section) { + user_id_t reloc_id = reloc_section->GetID(); + const ELFSectionHeaderInfo *reloc_header = + GetSectionHeaderByIndex(reloc_id); + assert(reloc_header); - for (SectionHeaderCollIter I = m_section_headers.begin(); - I != m_section_headers.end(); ++I) - { - if (I->sh_type == SHT_RELA || I->sh_type == SHT_REL) - { - if (CalculateType() == eTypeObjectFile) - { - const char *section_name = I->section_name.AsCString(""); - if (strstr(section_name, ".rela.debug") || - strstr(section_name, ".rel.debug")) - { - const ELFSectionHeader &reloc_header = *I; - user_id_t reloc_id = SectionIndex(I); - RelocateDebugSections(&reloc_header, reloc_id); - } - } + if (m_symtab_ap == nullptr) + m_symtab_ap.reset(new Symtab(reloc_section->GetObjectFile())); + + ParseTrampolineSymbols(m_symtab_ap.get(), symbol_id, reloc_header, + reloc_id); + } + } + + DWARFCallFrameInfo *eh_frame = GetUnwindTable().GetEHFrameInfo(); + if (eh_frame) { + if (m_symtab_ap == nullptr) + m_symtab_ap.reset(new Symtab(this)); + ParseUnwindSymbols(m_symtab_ap.get(), eh_frame); + } + + // If we still don't have any symtab then create an empty instance to avoid + // do the section + // lookup next time. + if (m_symtab_ap == nullptr) + m_symtab_ap.reset(new Symtab(this)); + + m_symtab_ap->CalculateSymbolSizes(); + } + + for (SectionHeaderCollIter I = m_section_headers.begin(); + I != m_section_headers.end(); ++I) { + if (I->sh_type == SHT_RELA || I->sh_type == SHT_REL) { + if (CalculateType() == eTypeObjectFile) { + const char *section_name = I->section_name.AsCString(""); + if (strstr(section_name, ".rela.debug") || + strstr(section_name, ".rel.debug")) { + const ELFSectionHeader &reloc_header = *I; + user_id_t reloc_id = SectionIndex(I); + RelocateDebugSections(&reloc_header, reloc_id); } + } + } + } + return m_symtab_ap.get(); +} + +void ObjectFileELF::ParseUnwindSymbols(Symtab *symbol_table, + DWARFCallFrameInfo *eh_frame) { + SectionList *section_list = GetSectionList(); + if (!section_list) + return; + + // First we save the new symbols into a separate list and add them to the + // symbol table after + // we colleced all symbols we want to add. This is neccessary because adding a + // new symbol + // invalidates the internal index of the symtab what causing the next lookup + // to be slow because + // it have to recalculate the index first. + std::vector<Symbol> new_symbols; + + eh_frame->ForEachFDEEntries([this, symbol_table, section_list, &new_symbols]( + lldb::addr_t file_addr, uint32_t size, dw_offset_t) { + Symbol *symbol = symbol_table->FindSymbolAtFileAddress(file_addr); + if (symbol) { + if (!symbol->GetByteSizeIsValid()) { + symbol->SetByteSize(size); + symbol->SetSizeIsSynthesized(true); + } + } else { + SectionSP section_sp = + section_list->FindSectionContainingFileAddress(file_addr); + if (section_sp) { + addr_t offset = file_addr - section_sp->GetFileAddress(); + const char *symbol_name = GetNextSyntheticSymbolName().GetCString(); + uint64_t symbol_id = symbol_table->GetNumSymbols(); + Symbol eh_symbol( + symbol_id, // Symbol table index. + symbol_name, // Symbol name. + false, // Is the symbol name mangled? + eSymbolTypeCode, // Type of this symbol. + true, // Is this globally visible? + false, // Is this symbol debug info? + false, // Is this symbol a trampoline? + true, // Is this symbol artificial? + section_sp, // Section in which this symbol is defined or null. + offset, // Offset in section or symbol value. + 0, // Size: Don't specify the size as an FDE can + false, // Size is valid: cover multiple symbols. + false, // Contains linker annotations? + 0); // Symbol flags. + new_symbols.push_back(eh_symbol); + } } - return m_symtab_ap.get(); -} - -void -ObjectFileELF::ParseUnwindSymbols(Symtab *symbol_table, DWARFCallFrameInfo* eh_frame) -{ - SectionList* section_list = GetSectionList(); - if (!section_list) - return; - - // First we save the new symbols into a separate list and add them to the symbol table after - // we colleced all symbols we want to add. This is neccessary because adding a new symbol - // invalidates the internal index of the symtab what causing the next lookup to be slow because - // it have to recalculate the index first. - std::vector<Symbol> new_symbols; - - eh_frame->ForEachFDEEntries( - [this, symbol_table, section_list, &new_symbols](lldb::addr_t file_addr, - uint32_t size, - dw_offset_t) { - Symbol* symbol = symbol_table->FindSymbolAtFileAddress(file_addr); - if (symbol) - { - if (!symbol->GetByteSizeIsValid()) - { - symbol->SetByteSize(size); - symbol->SetSizeIsSynthesized(true); - } - } - else - { - SectionSP section_sp = section_list->FindSectionContainingFileAddress(file_addr); - if (section_sp) - { - addr_t offset = file_addr - section_sp->GetFileAddress(); - const char* symbol_name = GetNextSyntheticSymbolName().GetCString(); - uint64_t symbol_id = symbol_table->GetNumSymbols(); - Symbol eh_symbol( - symbol_id, // Symbol table index. - symbol_name, // Symbol name. - false, // Is the symbol name mangled? - eSymbolTypeCode, // Type of this symbol. - true, // Is this globally visible? - false, // Is this symbol debug info? - false, // Is this symbol a trampoline? - true, // Is this symbol artificial? - section_sp, // Section in which this symbol is defined or null. - offset, // Offset in section or symbol value. - 0, // Size: Don't specify the size as an FDE can - false, // Size is valid: cover multiple symbols. - false, // Contains linker annotations? - 0); // Symbol flags. - new_symbols.push_back(eh_symbol); - } - } - return true; - }); + return true; + }); - for (const Symbol& s : new_symbols) - symbol_table->AddSymbol(s); + for (const Symbol &s : new_symbols) + symbol_table->AddSymbol(s); } -bool -ObjectFileELF::IsStripped () -{ - // TODO: determine this for ELF - return false; +bool ObjectFileELF::IsStripped() { + // TODO: determine this for ELF + return false; } //===----------------------------------------------------------------------===// @@ -3130,40 +3000,38 @@ ObjectFileELF::IsStripped () // Dump the specifics of the runtime file container (such as any headers // segments, sections, etc). //---------------------------------------------------------------------- -void -ObjectFileELF::Dump(Stream *s) -{ - ModuleSP module_sp(GetModule()); - if (!module_sp) - { - return; - } - - std::lock_guard<std::recursive_mutex> guard(module_sp->GetMutex()); - s->Printf("%p: ", static_cast<void *>(this)); - s->Indent(); - s->PutCString("ObjectFileELF"); - - ArchSpec header_arch; - GetArchitecture(header_arch); - - *s << ", file = '" << m_file << "', arch = " << header_arch.GetArchitectureName() << "\n"; - - DumpELFHeader(s, m_header); - s->EOL(); - DumpELFProgramHeaders(s); - s->EOL(); - DumpELFSectionHeaders(s); - s->EOL(); - SectionList *section_list = GetSectionList(); - if (section_list) - section_list->Dump(s, NULL, true, UINT32_MAX); - Symtab *symtab = GetSymtab(); - if (symtab) - symtab->Dump(s, NULL, eSortOrderNone); - s->EOL(); - DumpDependentModules(s); - s->EOL(); +void ObjectFileELF::Dump(Stream *s) { + ModuleSP module_sp(GetModule()); + if (!module_sp) { + return; + } + + std::lock_guard<std::recursive_mutex> guard(module_sp->GetMutex()); + s->Printf("%p: ", static_cast<void *>(this)); + s->Indent(); + s->PutCString("ObjectFileELF"); + + ArchSpec header_arch; + GetArchitecture(header_arch); + + *s << ", file = '" << m_file + << "', arch = " << header_arch.GetArchitectureName() << "\n"; + + DumpELFHeader(s, m_header); + s->EOL(); + DumpELFProgramHeaders(s); + s->EOL(); + DumpELFSectionHeaders(s); + s->EOL(); + SectionList *section_list = GetSectionList(); + if (section_list) + section_list->Dump(s, NULL, true, UINT32_MAX); + Symtab *symtab = GetSymtab(); + if (symtab) + symtab->Dump(s, NULL, eSortOrderNone); + s->EOL(); + DumpDependentModules(s); + s->EOL(); } //---------------------------------------------------------------------- @@ -3171,38 +3039,36 @@ ObjectFileELF::Dump(Stream *s) // // Dump the ELF header to the specified output stream //---------------------------------------------------------------------- -void -ObjectFileELF::DumpELFHeader(Stream *s, const ELFHeader &header) -{ - s->PutCString("ELF Header\n"); - s->Printf("e_ident[EI_MAG0 ] = 0x%2.2x\n", header.e_ident[EI_MAG0]); - s->Printf("e_ident[EI_MAG1 ] = 0x%2.2x '%c'\n", - header.e_ident[EI_MAG1], header.e_ident[EI_MAG1]); - s->Printf("e_ident[EI_MAG2 ] = 0x%2.2x '%c'\n", - header.e_ident[EI_MAG2], header.e_ident[EI_MAG2]); - s->Printf("e_ident[EI_MAG3 ] = 0x%2.2x '%c'\n", - header.e_ident[EI_MAG3], header.e_ident[EI_MAG3]); - - s->Printf("e_ident[EI_CLASS ] = 0x%2.2x\n", header.e_ident[EI_CLASS]); - s->Printf("e_ident[EI_DATA ] = 0x%2.2x ", header.e_ident[EI_DATA]); - DumpELFHeader_e_ident_EI_DATA(s, header.e_ident[EI_DATA]); - s->Printf ("\ne_ident[EI_VERSION] = 0x%2.2x\n", header.e_ident[EI_VERSION]); - s->Printf ("e_ident[EI_PAD ] = 0x%2.2x\n", header.e_ident[EI_PAD]); - - s->Printf("e_type = 0x%4.4x ", header.e_type); - DumpELFHeader_e_type(s, header.e_type); - s->Printf("\ne_machine = 0x%4.4x\n", header.e_machine); - s->Printf("e_version = 0x%8.8x\n", header.e_version); - s->Printf("e_entry = 0x%8.8" PRIx64 "\n", header.e_entry); - s->Printf("e_phoff = 0x%8.8" PRIx64 "\n", header.e_phoff); - s->Printf("e_shoff = 0x%8.8" PRIx64 "\n", header.e_shoff); - s->Printf("e_flags = 0x%8.8x\n", header.e_flags); - s->Printf("e_ehsize = 0x%4.4x\n", header.e_ehsize); - s->Printf("e_phentsize = 0x%4.4x\n", header.e_phentsize); - s->Printf("e_phnum = 0x%4.4x\n", header.e_phnum); - s->Printf("e_shentsize = 0x%4.4x\n", header.e_shentsize); - s->Printf("e_shnum = 0x%4.4x\n", header.e_shnum); - s->Printf("e_shstrndx = 0x%4.4x\n", header.e_shstrndx); +void ObjectFileELF::DumpELFHeader(Stream *s, const ELFHeader &header) { + s->PutCString("ELF Header\n"); + s->Printf("e_ident[EI_MAG0 ] = 0x%2.2x\n", header.e_ident[EI_MAG0]); + s->Printf("e_ident[EI_MAG1 ] = 0x%2.2x '%c'\n", header.e_ident[EI_MAG1], + header.e_ident[EI_MAG1]); + s->Printf("e_ident[EI_MAG2 ] = 0x%2.2x '%c'\n", header.e_ident[EI_MAG2], + header.e_ident[EI_MAG2]); + s->Printf("e_ident[EI_MAG3 ] = 0x%2.2x '%c'\n", header.e_ident[EI_MAG3], + header.e_ident[EI_MAG3]); + + s->Printf("e_ident[EI_CLASS ] = 0x%2.2x\n", header.e_ident[EI_CLASS]); + s->Printf("e_ident[EI_DATA ] = 0x%2.2x ", header.e_ident[EI_DATA]); + DumpELFHeader_e_ident_EI_DATA(s, header.e_ident[EI_DATA]); + s->Printf("\ne_ident[EI_VERSION] = 0x%2.2x\n", header.e_ident[EI_VERSION]); + s->Printf("e_ident[EI_PAD ] = 0x%2.2x\n", header.e_ident[EI_PAD]); + + s->Printf("e_type = 0x%4.4x ", header.e_type); + DumpELFHeader_e_type(s, header.e_type); + s->Printf("\ne_machine = 0x%4.4x\n", header.e_machine); + s->Printf("e_version = 0x%8.8x\n", header.e_version); + s->Printf("e_entry = 0x%8.8" PRIx64 "\n", header.e_entry); + s->Printf("e_phoff = 0x%8.8" PRIx64 "\n", header.e_phoff); + s->Printf("e_shoff = 0x%8.8" PRIx64 "\n", header.e_shoff); + s->Printf("e_flags = 0x%8.8x\n", header.e_flags); + s->Printf("e_ehsize = 0x%4.4x\n", header.e_ehsize); + s->Printf("e_phentsize = 0x%4.4x\n", header.e_phentsize); + s->Printf("e_phnum = 0x%4.4x\n", header.e_phnum); + s->Printf("e_shentsize = 0x%4.4x\n", header.e_shentsize); + s->Printf("e_shnum = 0x%4.4x\n", header.e_shnum); + s->Printf("e_shstrndx = 0x%4.4x\n", header.e_shstrndx); } //---------------------------------------------------------------------- @@ -3210,19 +3076,26 @@ ObjectFileELF::DumpELFHeader(Stream *s, const ELFHeader &header) // // Dump an token value for the ELF header member e_type //---------------------------------------------------------------------- -void -ObjectFileELF::DumpELFHeader_e_type(Stream *s, elf_half e_type) -{ - switch (e_type) - { - case ET_NONE: *s << "ET_NONE"; break; - case ET_REL: *s << "ET_REL"; break; - case ET_EXEC: *s << "ET_EXEC"; break; - case ET_DYN: *s << "ET_DYN"; break; - case ET_CORE: *s << "ET_CORE"; break; - default: - break; - } +void ObjectFileELF::DumpELFHeader_e_type(Stream *s, elf_half e_type) { + switch (e_type) { + case ET_NONE: + *s << "ET_NONE"; + break; + case ET_REL: + *s << "ET_REL"; + break; + case ET_EXEC: + *s << "ET_EXEC"; + break; + case ET_DYN: + *s << "ET_DYN"; + break; + case ET_CORE: + *s << "ET_CORE"; + break; + default: + break; + } } //---------------------------------------------------------------------- @@ -3230,34 +3103,38 @@ ObjectFileELF::DumpELFHeader_e_type(Stream *s, elf_half e_type) // // Dump an token value for the ELF header member e_ident[EI_DATA] //---------------------------------------------------------------------- -void -ObjectFileELF::DumpELFHeader_e_ident_EI_DATA(Stream *s, unsigned char ei_data) -{ - switch (ei_data) - { - case ELFDATANONE: *s << "ELFDATANONE"; break; - case ELFDATA2LSB: *s << "ELFDATA2LSB - Little Endian"; break; - case ELFDATA2MSB: *s << "ELFDATA2MSB - Big Endian"; break; - default: - break; - } +void ObjectFileELF::DumpELFHeader_e_ident_EI_DATA(Stream *s, + unsigned char ei_data) { + switch (ei_data) { + case ELFDATANONE: + *s << "ELFDATANONE"; + break; + case ELFDATA2LSB: + *s << "ELFDATA2LSB - Little Endian"; + break; + case ELFDATA2MSB: + *s << "ELFDATA2MSB - Big Endian"; + break; + default: + break; + } } - //---------------------------------------------------------------------- // DumpELFProgramHeader // // Dump a single ELF program header to the specified output stream //---------------------------------------------------------------------- -void -ObjectFileELF::DumpELFProgramHeader(Stream *s, const ELFProgramHeader &ph) -{ - DumpELFProgramHeader_p_type(s, ph.p_type); - s->Printf(" %8.8" PRIx64 " %8.8" PRIx64 " %8.8" PRIx64, ph.p_offset, ph.p_vaddr, ph.p_paddr); - s->Printf(" %8.8" PRIx64 " %8.8" PRIx64 " %8.8x (", ph.p_filesz, ph.p_memsz, ph.p_flags); +void ObjectFileELF::DumpELFProgramHeader(Stream *s, + const ELFProgramHeader &ph) { + DumpELFProgramHeader_p_type(s, ph.p_type); + s->Printf(" %8.8" PRIx64 " %8.8" PRIx64 " %8.8" PRIx64, ph.p_offset, + ph.p_vaddr, ph.p_paddr); + s->Printf(" %8.8" PRIx64 " %8.8" PRIx64 " %8.8x (", ph.p_filesz, ph.p_memsz, + ph.p_flags); - DumpELFProgramHeader_p_flags(s, ph.p_flags); - s->Printf(") %8.8" PRIx64, ph.p_align); + DumpELFProgramHeader_p_flags(s, ph.p_flags); + s->Printf(") %8.8" PRIx64, ph.p_align); } //---------------------------------------------------------------------- @@ -3266,41 +3143,35 @@ ObjectFileELF::DumpELFProgramHeader(Stream *s, const ELFProgramHeader &ph) // Dump an token value for the ELF program header member p_type which // describes the type of the program header // ---------------------------------------------------------------------- -void -ObjectFileELF::DumpELFProgramHeader_p_type(Stream *s, elf_word p_type) -{ - const int kStrWidth = 15; - switch (p_type) - { - CASE_AND_STREAM(s, PT_NULL , kStrWidth); - CASE_AND_STREAM(s, PT_LOAD , kStrWidth); - CASE_AND_STREAM(s, PT_DYNAMIC , kStrWidth); - CASE_AND_STREAM(s, PT_INTERP , kStrWidth); - CASE_AND_STREAM(s, PT_NOTE , kStrWidth); - CASE_AND_STREAM(s, PT_SHLIB , kStrWidth); - CASE_AND_STREAM(s, PT_PHDR , kStrWidth); - CASE_AND_STREAM(s, PT_TLS , kStrWidth); +void ObjectFileELF::DumpELFProgramHeader_p_type(Stream *s, elf_word p_type) { + const int kStrWidth = 15; + switch (p_type) { + CASE_AND_STREAM(s, PT_NULL, kStrWidth); + CASE_AND_STREAM(s, PT_LOAD, kStrWidth); + CASE_AND_STREAM(s, PT_DYNAMIC, kStrWidth); + CASE_AND_STREAM(s, PT_INTERP, kStrWidth); + CASE_AND_STREAM(s, PT_NOTE, kStrWidth); + CASE_AND_STREAM(s, PT_SHLIB, kStrWidth); + CASE_AND_STREAM(s, PT_PHDR, kStrWidth); + CASE_AND_STREAM(s, PT_TLS, kStrWidth); CASE_AND_STREAM(s, PT_GNU_EH_FRAME, kStrWidth); - default: - s->Printf("0x%8.8x%*s", p_type, kStrWidth - 10, ""); - break; - } + default: + s->Printf("0x%8.8x%*s", p_type, kStrWidth - 10, ""); + break; + } } - //---------------------------------------------------------------------- // DumpELFProgramHeader_p_flags // // Dump an token value for the ELF program header member p_flags //---------------------------------------------------------------------- -void -ObjectFileELF::DumpELFProgramHeader_p_flags(Stream *s, elf_word p_flags) -{ - *s << ((p_flags & PF_X) ? "PF_X" : " ") - << (((p_flags & PF_X) && (p_flags & PF_W)) ? '+' : ' ') - << ((p_flags & PF_W) ? "PF_W" : " ") - << (((p_flags & PF_W) && (p_flags & PF_R)) ? '+' : ' ') - << ((p_flags & PF_R) ? "PF_R" : " "); +void ObjectFileELF::DumpELFProgramHeader_p_flags(Stream *s, elf_word p_flags) { + *s << ((p_flags & PF_X) ? "PF_X" : " ") + << (((p_flags & PF_X) && (p_flags & PF_W)) ? '+' : ' ') + << ((p_flags & PF_W) ? "PF_W" : " ") + << (((p_flags & PF_W) && (p_flags & PF_R)) ? '+' : ' ') + << ((p_flags & PF_R) ? "PF_R" : " "); } //---------------------------------------------------------------------- @@ -3308,26 +3179,23 @@ ObjectFileELF::DumpELFProgramHeader_p_flags(Stream *s, elf_word p_flags) // // Dump all of the ELF program header to the specified output stream //---------------------------------------------------------------------- -void -ObjectFileELF::DumpELFProgramHeaders(Stream *s) -{ - if (!ParseProgramHeaders()) - return; - - s->PutCString("Program Headers\n"); - s->PutCString("IDX p_type p_offset p_vaddr p_paddr " - "p_filesz p_memsz p_flags p_align\n"); - s->PutCString("==== --------------- -------- -------- -------- " - "-------- -------- ------------------------- --------\n"); - - uint32_t idx = 0; - for (ProgramHeaderCollConstIter I = m_program_headers.begin(); - I != m_program_headers.end(); ++I, ++idx) - { - s->Printf("[%2u] ", idx); - ObjectFileELF::DumpELFProgramHeader(s, *I); - s->EOL(); - } +void ObjectFileELF::DumpELFProgramHeaders(Stream *s) { + if (!ParseProgramHeaders()) + return; + + s->PutCString("Program Headers\n"); + s->PutCString("IDX p_type p_offset p_vaddr p_paddr " + "p_filesz p_memsz p_flags p_align\n"); + s->PutCString("==== --------------- -------- -------- -------- " + "-------- -------- ------------------------- --------\n"); + + uint32_t idx = 0; + for (ProgramHeaderCollConstIter I = m_program_headers.begin(); + I != m_program_headers.end(); ++I, ++idx) { + s->Printf("[%2u] ", idx); + ObjectFileELF::DumpELFProgramHeader(s, *I); + s->EOL(); + } } //---------------------------------------------------------------------- @@ -3335,16 +3203,16 @@ ObjectFileELF::DumpELFProgramHeaders(Stream *s) // // Dump a single ELF section header to the specified output stream //---------------------------------------------------------------------- -void -ObjectFileELF::DumpELFSectionHeader(Stream *s, const ELFSectionHeaderInfo &sh) -{ - s->Printf("%8.8x ", sh.sh_name); - DumpELFSectionHeader_sh_type(s, sh.sh_type); - s->Printf(" %8.8" PRIx64 " (", sh.sh_flags); - DumpELFSectionHeader_sh_flags(s, sh.sh_flags); - s->Printf(") %8.8" PRIx64 " %8.8" PRIx64 " %8.8" PRIx64, sh.sh_addr, sh.sh_offset, sh.sh_size); - s->Printf(" %8.8x %8.8x", sh.sh_link, sh.sh_info); - s->Printf(" %8.8" PRIx64 " %8.8" PRIx64, sh.sh_addralign, sh.sh_entsize); +void ObjectFileELF::DumpELFSectionHeader(Stream *s, + const ELFSectionHeaderInfo &sh) { + s->Printf("%8.8x ", sh.sh_name); + DumpELFSectionHeader_sh_type(s, sh.sh_type); + s->Printf(" %8.8" PRIx64 " (", sh.sh_flags); + DumpELFSectionHeader_sh_flags(s, sh.sh_flags); + s->Printf(") %8.8" PRIx64 " %8.8" PRIx64 " %8.8" PRIx64, sh.sh_addr, + sh.sh_offset, sh.sh_size); + s->Printf(" %8.8x %8.8x", sh.sh_link, sh.sh_info); + s->Printf(" %8.8" PRIx64 " %8.8" PRIx64, sh.sh_addralign, sh.sh_entsize); } //---------------------------------------------------------------------- @@ -3353,32 +3221,29 @@ ObjectFileELF::DumpELFSectionHeader(Stream *s, const ELFSectionHeaderInfo &sh) // Dump an token value for the ELF section header member sh_type which // describes the type of the section //---------------------------------------------------------------------- -void -ObjectFileELF::DumpELFSectionHeader_sh_type(Stream *s, elf_word sh_type) -{ - const int kStrWidth = 12; - switch (sh_type) - { - CASE_AND_STREAM(s, SHT_NULL , kStrWidth); - CASE_AND_STREAM(s, SHT_PROGBITS , kStrWidth); - CASE_AND_STREAM(s, SHT_SYMTAB , kStrWidth); - CASE_AND_STREAM(s, SHT_STRTAB , kStrWidth); - CASE_AND_STREAM(s, SHT_RELA , kStrWidth); - CASE_AND_STREAM(s, SHT_HASH , kStrWidth); - CASE_AND_STREAM(s, SHT_DYNAMIC , kStrWidth); - CASE_AND_STREAM(s, SHT_NOTE , kStrWidth); - CASE_AND_STREAM(s, SHT_NOBITS , kStrWidth); - CASE_AND_STREAM(s, SHT_REL , kStrWidth); - CASE_AND_STREAM(s, SHT_SHLIB , kStrWidth); - CASE_AND_STREAM(s, SHT_DYNSYM , kStrWidth); - CASE_AND_STREAM(s, SHT_LOPROC , kStrWidth); - CASE_AND_STREAM(s, SHT_HIPROC , kStrWidth); - CASE_AND_STREAM(s, SHT_LOUSER , kStrWidth); - CASE_AND_STREAM(s, SHT_HIUSER , kStrWidth); - default: - s->Printf("0x%8.8x%*s", sh_type, kStrWidth - 10, ""); - break; - } +void ObjectFileELF::DumpELFSectionHeader_sh_type(Stream *s, elf_word sh_type) { + const int kStrWidth = 12; + switch (sh_type) { + CASE_AND_STREAM(s, SHT_NULL, kStrWidth); + CASE_AND_STREAM(s, SHT_PROGBITS, kStrWidth); + CASE_AND_STREAM(s, SHT_SYMTAB, kStrWidth); + CASE_AND_STREAM(s, SHT_STRTAB, kStrWidth); + CASE_AND_STREAM(s, SHT_RELA, kStrWidth); + CASE_AND_STREAM(s, SHT_HASH, kStrWidth); + CASE_AND_STREAM(s, SHT_DYNAMIC, kStrWidth); + CASE_AND_STREAM(s, SHT_NOTE, kStrWidth); + CASE_AND_STREAM(s, SHT_NOBITS, kStrWidth); + CASE_AND_STREAM(s, SHT_REL, kStrWidth); + CASE_AND_STREAM(s, SHT_SHLIB, kStrWidth); + CASE_AND_STREAM(s, SHT_DYNSYM, kStrWidth); + CASE_AND_STREAM(s, SHT_LOPROC, kStrWidth); + CASE_AND_STREAM(s, SHT_HIPROC, kStrWidth); + CASE_AND_STREAM(s, SHT_LOUSER, kStrWidth); + CASE_AND_STREAM(s, SHT_HIUSER, kStrWidth); + default: + s->Printf("0x%8.8x%*s", sh_type, kStrWidth - 10, ""); + break; + } } //---------------------------------------------------------------------- @@ -3386,14 +3251,13 @@ ObjectFileELF::DumpELFSectionHeader_sh_type(Stream *s, elf_word sh_type) // // Dump an token value for the ELF section header member sh_flags //---------------------------------------------------------------------- -void -ObjectFileELF::DumpELFSectionHeader_sh_flags(Stream *s, elf_xword sh_flags) -{ - *s << ((sh_flags & SHF_WRITE) ? "WRITE" : " ") - << (((sh_flags & SHF_WRITE) && (sh_flags & SHF_ALLOC)) ? '+' : ' ') - << ((sh_flags & SHF_ALLOC) ? "ALLOC" : " ") - << (((sh_flags & SHF_ALLOC) && (sh_flags & SHF_EXECINSTR)) ? '+' : ' ') - << ((sh_flags & SHF_EXECINSTR) ? "EXECINSTR" : " "); +void ObjectFileELF::DumpELFSectionHeader_sh_flags(Stream *s, + elf_xword sh_flags) { + *s << ((sh_flags & SHF_WRITE) ? "WRITE" : " ") + << (((sh_flags & SHF_WRITE) && (sh_flags & SHF_ALLOC)) ? '+' : ' ') + << ((sh_flags & SHF_ALLOC) ? "ALLOC" : " ") + << (((sh_flags & SHF_ALLOC) && (sh_flags & SHF_EXECINSTR)) ? '+' : ' ') + << ((sh_flags & SHF_EXECINSTR) ? "EXECINSTR" : " "); } //---------------------------------------------------------------------- @@ -3401,153 +3265,135 @@ ObjectFileELF::DumpELFSectionHeader_sh_flags(Stream *s, elf_xword sh_flags) // // Dump all of the ELF section header to the specified output stream //---------------------------------------------------------------------- -void -ObjectFileELF::DumpELFSectionHeaders(Stream *s) -{ - if (!ParseSectionHeaders()) - return; - - s->PutCString("Section Headers\n"); - s->PutCString("IDX name type flags " - "addr offset size link info addralgn " - "entsize Name\n"); - s->PutCString("==== -------- ------------ -------------------------------- " - "-------- -------- -------- -------- -------- -------- " - "-------- ====================\n"); - - uint32_t idx = 0; - for (SectionHeaderCollConstIter I = m_section_headers.begin(); - I != m_section_headers.end(); ++I, ++idx) - { - s->Printf("[%2u] ", idx); - ObjectFileELF::DumpELFSectionHeader(s, *I); - const char* section_name = I->section_name.AsCString(""); - if (section_name) - *s << ' ' << section_name << "\n"; - } -} - -void -ObjectFileELF::DumpDependentModules(lldb_private::Stream *s) -{ - size_t num_modules = ParseDependentModules(); +void ObjectFileELF::DumpELFSectionHeaders(Stream *s) { + if (!ParseSectionHeaders()) + return; + + s->PutCString("Section Headers\n"); + s->PutCString("IDX name type flags " + "addr offset size link info addralgn " + "entsize Name\n"); + s->PutCString("==== -------- ------------ -------------------------------- " + "-------- -------- -------- -------- -------- -------- " + "-------- ====================\n"); + + uint32_t idx = 0; + for (SectionHeaderCollConstIter I = m_section_headers.begin(); + I != m_section_headers.end(); ++I, ++idx) { + s->Printf("[%2u] ", idx); + ObjectFileELF::DumpELFSectionHeader(s, *I); + const char *section_name = I->section_name.AsCString(""); + if (section_name) + *s << ' ' << section_name << "\n"; + } +} + +void ObjectFileELF::DumpDependentModules(lldb_private::Stream *s) { + size_t num_modules = ParseDependentModules(); + + if (num_modules > 0) { + s->PutCString("Dependent Modules:\n"); + for (unsigned i = 0; i < num_modules; ++i) { + const FileSpec &spec = m_filespec_ap->GetFileSpecAtIndex(i); + s->Printf(" %s\n", spec.GetFilename().GetCString()); + } + } +} + +bool ObjectFileELF::GetArchitecture(ArchSpec &arch) { + if (!ParseHeader()) + return false; - if (num_modules > 0) - { - s->PutCString("Dependent Modules:\n"); - for (unsigned i = 0; i < num_modules; ++i) - { - const FileSpec &spec = m_filespec_ap->GetFileSpecAtIndex(i); - s->Printf(" %s\n", spec.GetFilename().GetCString()); + if (m_section_headers.empty()) { + // Allow elf notes to be parsed which may affect the detected architecture. + ParseSectionHeaders(); + } + + if (CalculateType() == eTypeCoreFile && + m_arch_spec.TripleOSIsUnspecifiedUnknown()) { + // Core files don't have section headers yet they have PT_NOTE program + // headers + // that might shed more light on the architecture + if (ParseProgramHeaders()) { + for (size_t i = 0, count = GetProgramHeaderCount(); i < count; ++i) { + const elf::ELFProgramHeader *header = GetProgramHeaderByIndex(i); + if (header && header->p_type == PT_NOTE && header->p_offset != 0 && + header->p_filesz > 0) { + DataExtractor data; + if (data.SetData(m_data, header->p_offset, header->p_filesz) == + header->p_filesz) { + lldb_private::UUID uuid; + RefineModuleDetailsFromNote(data, m_arch_spec, uuid); + } } + } } + } + arch = m_arch_spec; + return true; } -bool -ObjectFileELF::GetArchitecture (ArchSpec &arch) -{ - if (!ParseHeader()) - return false; +ObjectFile::Type ObjectFileELF::CalculateType() { + switch (m_header.e_type) { + case llvm::ELF::ET_NONE: + // 0 - No file type + return eTypeUnknown; - if (m_section_headers.empty()) - { - // Allow elf notes to be parsed which may affect the detected architecture. - ParseSectionHeaders(); - } + case llvm::ELF::ET_REL: + // 1 - Relocatable file + return eTypeObjectFile; - if (CalculateType() == eTypeCoreFile && m_arch_spec.TripleOSIsUnspecifiedUnknown()) - { - // Core files don't have section headers yet they have PT_NOTE program headers - // that might shed more light on the architecture - if (ParseProgramHeaders()) - { - for (size_t i = 0, count = GetProgramHeaderCount(); i < count; ++i) - { - const elf::ELFProgramHeader* header = GetProgramHeaderByIndex(i); - if (header && header->p_type == PT_NOTE && header->p_offset != 0 && header->p_filesz > 0) - { - DataExtractor data; - if (data.SetData (m_data, header->p_offset, header->p_filesz) == header->p_filesz) - { - lldb_private::UUID uuid; - RefineModuleDetailsFromNote (data, m_arch_spec, uuid); - } - } - } - } - } - arch = m_arch_spec; - return true; -} + case llvm::ELF::ET_EXEC: + // 2 - Executable file + return eTypeExecutable; -ObjectFile::Type -ObjectFileELF::CalculateType() -{ - switch (m_header.e_type) - { - case llvm::ELF::ET_NONE: - // 0 - No file type - return eTypeUnknown; + case llvm::ELF::ET_DYN: + // 3 - Shared object file + return eTypeSharedLibrary; - case llvm::ELF::ET_REL: - // 1 - Relocatable file - return eTypeObjectFile; + case ET_CORE: + // 4 - Core file + return eTypeCoreFile; - case llvm::ELF::ET_EXEC: - // 2 - Executable file - return eTypeExecutable; + default: + break; + } + return eTypeUnknown; +} - case llvm::ELF::ET_DYN: - // 3 - Shared object file - return eTypeSharedLibrary; +ObjectFile::Strata ObjectFileELF::CalculateStrata() { + switch (m_header.e_type) { + case llvm::ELF::ET_NONE: + // 0 - No file type + return eStrataUnknown; - case ET_CORE: - // 4 - Core file - return eTypeCoreFile; + case llvm::ELF::ET_REL: + // 1 - Relocatable file + return eStrataUnknown; - default: - break; - } - return eTypeUnknown; -} + case llvm::ELF::ET_EXEC: + // 2 - Executable file + // TODO: is there any way to detect that an executable is a kernel + // related executable by inspecting the program headers, section + // headers, symbols, or any other flag bits??? + return eStrataUser; + + case llvm::ELF::ET_DYN: + // 3 - Shared object file + // TODO: is there any way to detect that an shared library is a kernel + // related executable by inspecting the program headers, section + // headers, symbols, or any other flag bits??? + return eStrataUnknown; -ObjectFile::Strata -ObjectFileELF::CalculateStrata() -{ - switch (m_header.e_type) - { - case llvm::ELF::ET_NONE: - // 0 - No file type - return eStrataUnknown; - - case llvm::ELF::ET_REL: - // 1 - Relocatable file - return eStrataUnknown; - - case llvm::ELF::ET_EXEC: - // 2 - Executable file - // TODO: is there any way to detect that an executable is a kernel - // related executable by inspecting the program headers, section - // headers, symbols, or any other flag bits??? - return eStrataUser; - - case llvm::ELF::ET_DYN: - // 3 - Shared object file - // TODO: is there any way to detect that an shared library is a kernel - // related executable by inspecting the program headers, section - // headers, symbols, or any other flag bits??? - return eStrataUnknown; - - case ET_CORE: - // 4 - Core file - // TODO: is there any way to detect that an core file is a kernel - // related executable by inspecting the program headers, section - // headers, symbols, or any other flag bits??? - return eStrataUnknown; - - default: - break; - } + case ET_CORE: + // 4 - Core file + // TODO: is there any way to detect that an core file is a kernel + // related executable by inspecting the program headers, section + // headers, symbols, or any other flag bits??? return eStrataUnknown; -} + default: + break; + } + return eStrataUnknown; +} diff --git a/source/Plugins/ObjectFile/ELF/ObjectFileELF.h b/source/Plugins/ObjectFile/ELF/ObjectFileELF.h index e2f73f53ec63e..4ce5648cfed55 100644 --- a/source/Plugins/ObjectFile/ELF/ObjectFileELF.h +++ b/source/Plugins/ObjectFile/ELF/ObjectFileELF.h @@ -19,46 +19,40 @@ // Other libraries and framework includes // Project includes -#include "lldb/lldb-private.h" +#include "lldb/Core/ArchSpec.h" +#include "lldb/Core/UUID.h" #include "lldb/Host/FileSpec.h" #include "lldb/Symbol/ObjectFile.h" -#include "lldb/Core/UUID.h" -#include "lldb/Core/ArchSpec.h" +#include "lldb/lldb-private.h" #include "ELFHeader.h" -struct ELFNote -{ - elf::elf_word n_namesz; - elf::elf_word n_descsz; - elf::elf_word n_type; - - std::string n_name; - - ELFNote() : n_namesz(0), n_descsz(0), n_type(0) - { - } - - /// Parse an ELFNote entry from the given DataExtractor starting at position - /// \p offset. - /// - /// @param[in] data - /// The DataExtractor to read from. - /// - /// @param[in,out] offset - /// Pointer to an offset in the data. On return the offset will be - /// advanced by the number of bytes read. - /// - /// @return - /// True if the ELFRel entry was successfully read and false otherwise. - bool - Parse(const lldb_private::DataExtractor &data, lldb::offset_t *offset); - - size_t - GetByteSize() const - { - return 12 + llvm::alignTo (n_namesz, 4) + llvm::alignTo (n_descsz, 4); - } +struct ELFNote { + elf::elf_word n_namesz; + elf::elf_word n_descsz; + elf::elf_word n_type; + + std::string n_name; + + ELFNote() : n_namesz(0), n_descsz(0), n_type(0) {} + + /// Parse an ELFNote entry from the given DataExtractor starting at position + /// \p offset. + /// + /// @param[in] data + /// The DataExtractor to read from. + /// + /// @param[in,out] offset + /// Pointer to an offset in the data. On return the offset will be + /// advanced by the number of bytes read. + /// + /// @return + /// True if the ELFRel entry was successfully read and false otherwise. + bool Parse(const lldb_private::DataExtractor &data, lldb::offset_t *offset); + + size_t GetByteSize() const { + return 12 + llvm::alignTo(n_namesz, 4) + llvm::alignTo(n_descsz, 4); + } }; //------------------------------------------------------------------------------ @@ -67,390 +61,332 @@ struct ELFNote /// /// This class provides a generic ELF (32/64 bit) reader plugin implementing the /// ObjectFile protocol. -class ObjectFileELF : - public lldb_private::ObjectFile -{ +class ObjectFileELF : public lldb_private::ObjectFile { public: - ~ObjectFileELF() override; + ~ObjectFileELF() override; + + //------------------------------------------------------------------ + // Static Functions + //------------------------------------------------------------------ + static void Initialize(); + + static void Terminate(); + + static lldb_private::ConstString GetPluginNameStatic(); + + static const char *GetPluginDescriptionStatic(); + + static lldb_private::ObjectFile * + CreateInstance(const lldb::ModuleSP &module_sp, lldb::DataBufferSP &data_sp, + lldb::offset_t data_offset, const lldb_private::FileSpec *file, + lldb::offset_t file_offset, lldb::offset_t length); + + static lldb_private::ObjectFile *CreateMemoryInstance( + const lldb::ModuleSP &module_sp, lldb::DataBufferSP &data_sp, + const lldb::ProcessSP &process_sp, lldb::addr_t header_addr); + + static size_t GetModuleSpecifications(const lldb_private::FileSpec &file, + lldb::DataBufferSP &data_sp, + lldb::offset_t data_offset, + lldb::offset_t file_offset, + lldb::offset_t length, + lldb_private::ModuleSpecList &specs); + + static bool MagicBytesMatch(lldb::DataBufferSP &data_sp, lldb::addr_t offset, + lldb::addr_t length); + + //------------------------------------------------------------------ + // PluginInterface protocol + //------------------------------------------------------------------ + lldb_private::ConstString GetPluginName() override; + + uint32_t GetPluginVersion() override; + + //------------------------------------------------------------------ + // ObjectFile Protocol. + //------------------------------------------------------------------ + bool ParseHeader() override; + + bool SetLoadAddress(lldb_private::Target &target, lldb::addr_t value, + bool value_is_offset) override; + + lldb::ByteOrder GetByteOrder() const override; + + bool IsExecutable() const override; + + uint32_t GetAddressByteSize() const override; + + lldb::AddressClass GetAddressClass(lldb::addr_t file_addr) override; + + lldb_private::Symtab *GetSymtab() override; + + bool IsStripped() override; + + void CreateSections(lldb_private::SectionList &unified_section_list) override; + + void Dump(lldb_private::Stream *s) override; + + bool GetArchitecture(lldb_private::ArchSpec &arch) override; + + bool GetUUID(lldb_private::UUID *uuid) override; + + lldb_private::FileSpecList GetDebugSymbolFilePaths() override; + + uint32_t GetDependentModules(lldb_private::FileSpecList &files) override; + + lldb_private::Address + GetImageInfoAddress(lldb_private::Target *target) override; + + lldb_private::Address GetEntryPointAddress() override; + + ObjectFile::Type CalculateType() override; + + ObjectFile::Strata CalculateStrata() override; + + // Returns number of program headers found in the ELF file. + size_t GetProgramHeaderCount(); + + // Returns the program header with the given index. + const elf::ELFProgramHeader *GetProgramHeaderByIndex(lldb::user_id_t id); + + // Returns segment data for the given index. + lldb_private::DataExtractor GetSegmentDataByIndex(lldb::user_id_t id); - //------------------------------------------------------------------ - // Static Functions - //------------------------------------------------------------------ - static void - Initialize(); - - static void - Terminate(); - - static lldb_private::ConstString - GetPluginNameStatic(); - - static const char * - GetPluginDescriptionStatic(); - - static lldb_private::ObjectFile * - CreateInstance(const lldb::ModuleSP &module_sp, - lldb::DataBufferSP& data_sp, - lldb::offset_t data_offset, - const lldb_private::FileSpec* file, - lldb::offset_t file_offset, - lldb::offset_t length); - - static lldb_private::ObjectFile * - CreateMemoryInstance (const lldb::ModuleSP &module_sp, - lldb::DataBufferSP& data_sp, - const lldb::ProcessSP &process_sp, - lldb::addr_t header_addr); - - static size_t - GetModuleSpecifications (const lldb_private::FileSpec& file, - lldb::DataBufferSP& data_sp, - lldb::offset_t data_offset, - lldb::offset_t file_offset, - lldb::offset_t length, - lldb_private::ModuleSpecList &specs); - - static bool - MagicBytesMatch (lldb::DataBufferSP& data_sp, - lldb::addr_t offset, - lldb::addr_t length); - - //------------------------------------------------------------------ - // PluginInterface protocol - //------------------------------------------------------------------ - lldb_private::ConstString - GetPluginName() override; - - uint32_t - GetPluginVersion() override; - - //------------------------------------------------------------------ - // ObjectFile Protocol. - //------------------------------------------------------------------ - bool - ParseHeader() override; - - bool - SetLoadAddress (lldb_private::Target &target, - lldb::addr_t value, - bool value_is_offset) override; - - lldb::ByteOrder - GetByteOrder() const override; - - bool - IsExecutable () const override; - - uint32_t - GetAddressByteSize() const override; - - lldb::AddressClass - GetAddressClass (lldb::addr_t file_addr) override; - - lldb_private::Symtab * - GetSymtab() override; - - bool - IsStripped () override; - - void - CreateSections (lldb_private::SectionList &unified_section_list) override; - - void - Dump(lldb_private::Stream *s) override; - - bool - GetArchitecture (lldb_private::ArchSpec &arch) override; - - bool - GetUUID(lldb_private::UUID* uuid) override; - - lldb_private::FileSpecList - GetDebugSymbolFilePaths() override; - - uint32_t - GetDependentModules(lldb_private::FileSpecList& files) override; - - lldb_private::Address - GetImageInfoAddress(lldb_private::Target *target) override; - - lldb_private::Address - GetEntryPointAddress () override; - - ObjectFile::Type - CalculateType() override; - - ObjectFile::Strata - CalculateStrata() override; - - // Returns number of program headers found in the ELF file. - size_t - GetProgramHeaderCount(); - - // Returns the program header with the given index. - const elf::ELFProgramHeader * - GetProgramHeaderByIndex(lldb::user_id_t id); - - // Returns segment data for the given index. - lldb_private::DataExtractor - GetSegmentDataByIndex(lldb::user_id_t id); - - std::string - StripLinkerSymbolAnnotations(llvm::StringRef symbol_name) const override; + std::string + StripLinkerSymbolAnnotations(llvm::StringRef symbol_name) const override; private: - ObjectFileELF(const lldb::ModuleSP &module_sp, - lldb::DataBufferSP& data_sp, - lldb::offset_t data_offset, - const lldb_private::FileSpec* file, - lldb::offset_t offset, - lldb::offset_t length); - - ObjectFileELF (const lldb::ModuleSP &module_sp, - lldb::DataBufferSP& header_data_sp, - const lldb::ProcessSP &process_sp, - lldb::addr_t header_addr); - - typedef std::vector<elf::ELFProgramHeader> ProgramHeaderColl; - typedef ProgramHeaderColl::iterator ProgramHeaderCollIter; - typedef ProgramHeaderColl::const_iterator ProgramHeaderCollConstIter; - - struct ELFSectionHeaderInfo : public elf::ELFSectionHeader - { - lldb_private::ConstString section_name; - }; - - typedef std::vector<ELFSectionHeaderInfo> SectionHeaderColl; - typedef SectionHeaderColl::iterator SectionHeaderCollIter; - typedef SectionHeaderColl::const_iterator SectionHeaderCollConstIter; - - typedef std::vector<elf::ELFDynamic> DynamicSymbolColl; - typedef DynamicSymbolColl::iterator DynamicSymbolCollIter; - typedef DynamicSymbolColl::const_iterator DynamicSymbolCollConstIter; - - typedef std::map<lldb::addr_t, lldb::AddressClass> FileAddressToAddressClassMap; - typedef std::function<lldb::offset_t (lldb_private::DataExtractor &, lldb::offset_t, lldb::offset_t)> SetDataFunction; - - /// Version of this reader common to all plugins based on this class. - static const uint32_t m_plugin_version = 1; - static const uint32_t g_core_uuid_magic; - - /// ELF file header. - elf::ELFHeader m_header; - - /// ELF build ID. - lldb_private::UUID m_uuid; - - /// ELF .gnu_debuglink file and crc data if available. - std::string m_gnu_debuglink_file; - uint32_t m_gnu_debuglink_crc; - - /// Collection of program headers. - ProgramHeaderColl m_program_headers; - - /// Collection of section headers. - SectionHeaderColl m_section_headers; - - /// Collection of symbols from the dynamic table. - DynamicSymbolColl m_dynamic_symbols; - - /// List of file specifications corresponding to the modules (shared - /// libraries) on which this object file depends. - mutable std::unique_ptr<lldb_private::FileSpecList> m_filespec_ap; - - /// Cached value of the entry point for this module. - lldb_private::Address m_entry_point_address; - - /// The architecture detected from parsing elf file contents. - lldb_private::ArchSpec m_arch_spec; - - /// The address class for each symbol in the elf file - FileAddressToAddressClassMap m_address_class_map; - - /// Returns a 1 based index of the given section header. - size_t - SectionIndex(const SectionHeaderCollIter &I); - - /// Returns a 1 based index of the given section header. - size_t - SectionIndex(const SectionHeaderCollConstIter &I) const; - - // Parses the ELF program headers. - static size_t - GetProgramHeaderInfo(ProgramHeaderColl &program_headers, - const SetDataFunction &set_data, - const elf::ELFHeader &header); - - // Finds PT_NOTE segments and calculates their crc sum. - static uint32_t - CalculateELFNotesSegmentsCRC32(const ProgramHeaderColl& program_headers, - lldb_private::DataExtractor &data); - - /// Parses all section headers present in this object file and populates - /// m_program_headers. This method will compute the header list only once. - /// Returns the number of headers parsed. - size_t - ParseProgramHeaders(); - - /// Parses all section headers present in this object file and populates - /// m_section_headers. This method will compute the header list only once. - /// Returns the number of headers parsed. - size_t - ParseSectionHeaders(); - - static void - ParseARMAttributes(lldb_private::DataExtractor &data, uint64_t length, - lldb_private::ArchSpec &arch_spec); - - /// Parses the elf section headers and returns the uuid, debug link name, crc, archspec. - static size_t - GetSectionHeaderInfo(SectionHeaderColl §ion_headers, - const SetDataFunction &set_data, - const elf::ELFHeader &header, - lldb_private::UUID &uuid, - std::string &gnu_debuglink_file, - uint32_t &gnu_debuglink_crc, - lldb_private::ArchSpec &arch_spec); - - /// Scans the dynamic section and locates all dependent modules (shared - /// libraries) populating m_filespec_ap. This method will compute the - /// dependent module list only once. Returns the number of dependent - /// modules parsed. - size_t - ParseDependentModules(); - - /// Parses the dynamic symbol table and populates m_dynamic_symbols. The - /// vector retains the order as found in the object file. Returns the - /// number of dynamic symbols parsed. - size_t - ParseDynamicSymbols(); - - /// Populates m_symtab_ap will all non-dynamic linker symbols. This method - /// will parse the symbols only once. Returns the number of symbols parsed. - unsigned - ParseSymbolTable(lldb_private::Symtab *symbol_table, - lldb::user_id_t start_id, - lldb_private::Section *symtab); - - /// Helper routine for ParseSymbolTable(). - unsigned - ParseSymbols(lldb_private::Symtab *symbol_table, - lldb::user_id_t start_id, - lldb_private::SectionList *section_list, - const size_t num_symbols, - const lldb_private::DataExtractor &symtab_data, - const lldb_private::DataExtractor &strtab_data); - - /// Scans the relocation entries and adds a set of artificial symbols to the - /// given symbol table for each PLT slot. Returns the number of symbols - /// added. - unsigned - ParseTrampolineSymbols(lldb_private::Symtab *symbol_table, - lldb::user_id_t start_id, - const ELFSectionHeaderInfo *rela_hdr, - lldb::user_id_t section_id); - - void - ParseUnwindSymbols(lldb_private::Symtab *symbol_table, - lldb_private::DWARFCallFrameInfo* eh_frame); - - /// Relocates debug sections - unsigned - RelocateDebugSections(const elf::ELFSectionHeader *rel_hdr, lldb::user_id_t rel_id); - - unsigned - RelocateSection(lldb_private::Symtab* symtab, const elf::ELFHeader *hdr, const elf::ELFSectionHeader *rel_hdr, - const elf::ELFSectionHeader *symtab_hdr, const elf::ELFSectionHeader *debug_hdr, - lldb_private::DataExtractor &rel_data, lldb_private::DataExtractor &symtab_data, - lldb_private::DataExtractor &debug_data, lldb_private::Section* rel_section); - - /// Loads the section name string table into m_shstr_data. Returns the - /// number of bytes constituting the table. - size_t - GetSectionHeaderStringTable(); - - /// Utility method for looking up a section given its name. Returns the - /// index of the corresponding section or zero if no section with the given - /// name can be found (note that section indices are always 1 based, and so - /// section index 0 is never valid). - lldb::user_id_t - GetSectionIndexByName(const char *name); - - // Returns the ID of the first section that has the given type. - lldb::user_id_t - GetSectionIndexByType(unsigned type); - - /// Returns the section header with the given id or NULL. - const ELFSectionHeaderInfo * - GetSectionHeaderByIndex(lldb::user_id_t id); - - /// @name ELF header dump routines - //@{ - static void - DumpELFHeader(lldb_private::Stream *s, const elf::ELFHeader& header); - - static void - DumpELFHeader_e_ident_EI_DATA(lldb_private::Stream *s, - unsigned char ei_data); - - static void - DumpELFHeader_e_type(lldb_private::Stream *s, elf::elf_half e_type); - //@} - - /// @name ELF program header dump routines - //@{ - void - DumpELFProgramHeaders(lldb_private::Stream *s); - - static void - DumpELFProgramHeader(lldb_private::Stream *s, - const elf::ELFProgramHeader &ph); - - static void - DumpELFProgramHeader_p_type(lldb_private::Stream *s, elf::elf_word p_type); - - static void - DumpELFProgramHeader_p_flags(lldb_private::Stream *s, - elf::elf_word p_flags); - //@} - - /// @name ELF section header dump routines - //@{ - void - DumpELFSectionHeaders(lldb_private::Stream *s); - - static void - DumpELFSectionHeader(lldb_private::Stream *s, - const ELFSectionHeaderInfo& sh); - - static void - DumpELFSectionHeader_sh_type(lldb_private::Stream *s, - elf::elf_word sh_type); - - static void - DumpELFSectionHeader_sh_flags(lldb_private::Stream *s, - elf::elf_xword sh_flags); - //@} - - /// ELF dependent module dump routine. - void - DumpDependentModules(lldb_private::Stream *s); - - const elf::ELFDynamic * - FindDynamicSymbol(unsigned tag); - - unsigned - PLTRelocationType(); - - static lldb_private::Error - RefineModuleDetailsFromNote (lldb_private::DataExtractor &data, lldb_private::ArchSpec &arch_spec, lldb_private::UUID &uuid); - - - static lldb::offset_t - SetData(const lldb_private::DataExtractor &src, lldb_private::DataExtractor &dst, lldb::offset_t offset, lldb::offset_t length); - - lldb::offset_t - SetDataWithReadMemoryFallback(lldb_private::DataExtractor &dst, lldb::offset_t offset, lldb::offset_t length); + ObjectFileELF(const lldb::ModuleSP &module_sp, lldb::DataBufferSP &data_sp, + lldb::offset_t data_offset, const lldb_private::FileSpec *file, + lldb::offset_t offset, lldb::offset_t length); + + ObjectFileELF(const lldb::ModuleSP &module_sp, + lldb::DataBufferSP &header_data_sp, + const lldb::ProcessSP &process_sp, lldb::addr_t header_addr); + + typedef std::vector<elf::ELFProgramHeader> ProgramHeaderColl; + typedef ProgramHeaderColl::iterator ProgramHeaderCollIter; + typedef ProgramHeaderColl::const_iterator ProgramHeaderCollConstIter; + + struct ELFSectionHeaderInfo : public elf::ELFSectionHeader { + lldb_private::ConstString section_name; + }; + + typedef std::vector<ELFSectionHeaderInfo> SectionHeaderColl; + typedef SectionHeaderColl::iterator SectionHeaderCollIter; + typedef SectionHeaderColl::const_iterator SectionHeaderCollConstIter; + + typedef std::vector<elf::ELFDynamic> DynamicSymbolColl; + typedef DynamicSymbolColl::iterator DynamicSymbolCollIter; + typedef DynamicSymbolColl::const_iterator DynamicSymbolCollConstIter; + + typedef std::map<lldb::addr_t, lldb::AddressClass> + FileAddressToAddressClassMap; + typedef std::function<lldb::offset_t(lldb_private::DataExtractor &, + lldb::offset_t, lldb::offset_t)> + SetDataFunction; + + /// Version of this reader common to all plugins based on this class. + static const uint32_t m_plugin_version = 1; + static const uint32_t g_core_uuid_magic; + + /// ELF file header. + elf::ELFHeader m_header; + + /// ELF build ID. + lldb_private::UUID m_uuid; + + /// ELF .gnu_debuglink file and crc data if available. + std::string m_gnu_debuglink_file; + uint32_t m_gnu_debuglink_crc; + + /// Collection of program headers. + ProgramHeaderColl m_program_headers; + + /// Collection of section headers. + SectionHeaderColl m_section_headers; + + /// Collection of symbols from the dynamic table. + DynamicSymbolColl m_dynamic_symbols; + + /// List of file specifications corresponding to the modules (shared + /// libraries) on which this object file depends. + mutable std::unique_ptr<lldb_private::FileSpecList> m_filespec_ap; + + /// Cached value of the entry point for this module. + lldb_private::Address m_entry_point_address; + + /// The architecture detected from parsing elf file contents. + lldb_private::ArchSpec m_arch_spec; + + /// The address class for each symbol in the elf file + FileAddressToAddressClassMap m_address_class_map; + + /// Returns a 1 based index of the given section header. + size_t SectionIndex(const SectionHeaderCollIter &I); + + /// Returns a 1 based index of the given section header. + size_t SectionIndex(const SectionHeaderCollConstIter &I) const; + + // Parses the ELF program headers. + static size_t GetProgramHeaderInfo(ProgramHeaderColl &program_headers, + const SetDataFunction &set_data, + const elf::ELFHeader &header); + + // Finds PT_NOTE segments and calculates their crc sum. + static uint32_t + CalculateELFNotesSegmentsCRC32(const ProgramHeaderColl &program_headers, + lldb_private::DataExtractor &data); + + /// Parses all section headers present in this object file and populates + /// m_program_headers. This method will compute the header list only once. + /// Returns the number of headers parsed. + size_t ParseProgramHeaders(); + + /// Parses all section headers present in this object file and populates + /// m_section_headers. This method will compute the header list only once. + /// Returns the number of headers parsed. + size_t ParseSectionHeaders(); + + static void ParseARMAttributes(lldb_private::DataExtractor &data, + uint64_t length, + lldb_private::ArchSpec &arch_spec); + + /// Parses the elf section headers and returns the uuid, debug link name, crc, + /// archspec. + static size_t GetSectionHeaderInfo(SectionHeaderColl §ion_headers, + const SetDataFunction &set_data, + const elf::ELFHeader &header, + lldb_private::UUID &uuid, + std::string &gnu_debuglink_file, + uint32_t &gnu_debuglink_crc, + lldb_private::ArchSpec &arch_spec); + + /// Scans the dynamic section and locates all dependent modules (shared + /// libraries) populating m_filespec_ap. This method will compute the + /// dependent module list only once. Returns the number of dependent + /// modules parsed. + size_t ParseDependentModules(); + + /// Parses the dynamic symbol table and populates m_dynamic_symbols. The + /// vector retains the order as found in the object file. Returns the + /// number of dynamic symbols parsed. + size_t ParseDynamicSymbols(); + + /// Populates m_symtab_ap will all non-dynamic linker symbols. This method + /// will parse the symbols only once. Returns the number of symbols parsed. + unsigned ParseSymbolTable(lldb_private::Symtab *symbol_table, + lldb::user_id_t start_id, + lldb_private::Section *symtab); + + /// Helper routine for ParseSymbolTable(). + unsigned ParseSymbols(lldb_private::Symtab *symbol_table, + lldb::user_id_t start_id, + lldb_private::SectionList *section_list, + const size_t num_symbols, + const lldb_private::DataExtractor &symtab_data, + const lldb_private::DataExtractor &strtab_data); + + /// Scans the relocation entries and adds a set of artificial symbols to the + /// given symbol table for each PLT slot. Returns the number of symbols + /// added. + unsigned ParseTrampolineSymbols(lldb_private::Symtab *symbol_table, + lldb::user_id_t start_id, + const ELFSectionHeaderInfo *rela_hdr, + lldb::user_id_t section_id); + + void ParseUnwindSymbols(lldb_private::Symtab *symbol_table, + lldb_private::DWARFCallFrameInfo *eh_frame); + + /// Relocates debug sections + unsigned RelocateDebugSections(const elf::ELFSectionHeader *rel_hdr, + lldb::user_id_t rel_id); + + unsigned RelocateSection(lldb_private::Symtab *symtab, + const elf::ELFHeader *hdr, + const elf::ELFSectionHeader *rel_hdr, + const elf::ELFSectionHeader *symtab_hdr, + const elf::ELFSectionHeader *debug_hdr, + lldb_private::DataExtractor &rel_data, + lldb_private::DataExtractor &symtab_data, + lldb_private::DataExtractor &debug_data, + lldb_private::Section *rel_section); + + /// Loads the section name string table into m_shstr_data. Returns the + /// number of bytes constituting the table. + size_t GetSectionHeaderStringTable(); + + /// Utility method for looking up a section given its name. Returns the + /// index of the corresponding section or zero if no section with the given + /// name can be found (note that section indices are always 1 based, and so + /// section index 0 is never valid). + lldb::user_id_t GetSectionIndexByName(const char *name); + + // Returns the ID of the first section that has the given type. + lldb::user_id_t GetSectionIndexByType(unsigned type); + + /// Returns the section header with the given id or NULL. + const ELFSectionHeaderInfo *GetSectionHeaderByIndex(lldb::user_id_t id); + + /// @name ELF header dump routines + //@{ + static void DumpELFHeader(lldb_private::Stream *s, + const elf::ELFHeader &header); + + static void DumpELFHeader_e_ident_EI_DATA(lldb_private::Stream *s, + unsigned char ei_data); + + static void DumpELFHeader_e_type(lldb_private::Stream *s, + elf::elf_half e_type); + //@} + + /// @name ELF program header dump routines + //@{ + void DumpELFProgramHeaders(lldb_private::Stream *s); + + static void DumpELFProgramHeader(lldb_private::Stream *s, + const elf::ELFProgramHeader &ph); + + static void DumpELFProgramHeader_p_type(lldb_private::Stream *s, + elf::elf_word p_type); + + static void DumpELFProgramHeader_p_flags(lldb_private::Stream *s, + elf::elf_word p_flags); + //@} + + /// @name ELF section header dump routines + //@{ + void DumpELFSectionHeaders(lldb_private::Stream *s); + + static void DumpELFSectionHeader(lldb_private::Stream *s, + const ELFSectionHeaderInfo &sh); + + static void DumpELFSectionHeader_sh_type(lldb_private::Stream *s, + elf::elf_word sh_type); + + static void DumpELFSectionHeader_sh_flags(lldb_private::Stream *s, + elf::elf_xword sh_flags); + //@} + + /// ELF dependent module dump routine. + void DumpDependentModules(lldb_private::Stream *s); + + const elf::ELFDynamic *FindDynamicSymbol(unsigned tag); + + unsigned PLTRelocationType(); + + static lldb_private::Error + RefineModuleDetailsFromNote(lldb_private::DataExtractor &data, + lldb_private::ArchSpec &arch_spec, + lldb_private::UUID &uuid); + + static lldb::offset_t SetData(const lldb_private::DataExtractor &src, + lldb_private::DataExtractor &dst, + lldb::offset_t offset, lldb::offset_t length); + + lldb::offset_t SetDataWithReadMemoryFallback(lldb_private::DataExtractor &dst, + lldb::offset_t offset, + lldb::offset_t length); }; #endif // liblldb_ObjectFileELF_h_ diff --git a/source/Plugins/ObjectFile/JIT/ObjectFileJIT.cpp b/source/Plugins/ObjectFile/JIT/ObjectFileJIT.cpp index e3d8383337119..b5b5e38e1f6f3 100644 --- a/source/Plugins/ObjectFile/JIT/ObjectFileJIT.cpp +++ b/source/Plugins/ObjectFile/JIT/ObjectFileJIT.cpp @@ -26,8 +26,8 @@ #include "lldb/Core/StreamString.h" #include "lldb/Core/Timer.h" #include "lldb/Core/UUID.h" -#include "lldb/Host/Host.h" #include "lldb/Host/FileSpec.h" +#include "lldb/Host/Host.h" #include "lldb/Symbol/ObjectFile.h" #include "lldb/Target/Platform.h" #include "lldb/Target/Process.h" @@ -41,322 +41,227 @@ using namespace lldb; using namespace lldb_private; - -void -ObjectFileJIT::Initialize() -{ - PluginManager::RegisterPlugin (GetPluginNameStatic(), - GetPluginDescriptionStatic(), - CreateInstance, - CreateMemoryInstance, - GetModuleSpecifications); +void ObjectFileJIT::Initialize() { + PluginManager::RegisterPlugin(GetPluginNameStatic(), + GetPluginDescriptionStatic(), CreateInstance, + CreateMemoryInstance, GetModuleSpecifications); } -void -ObjectFileJIT::Terminate() -{ - PluginManager::UnregisterPlugin (CreateInstance); -} - - -lldb_private::ConstString -ObjectFileJIT::GetPluginNameStatic() -{ - static ConstString g_name("jit"); - return g_name; +void ObjectFileJIT::Terminate() { + PluginManager::UnregisterPlugin(CreateInstance); } -const char * -ObjectFileJIT::GetPluginDescriptionStatic() -{ - return "JIT code object file"; +lldb_private::ConstString ObjectFileJIT::GetPluginNameStatic() { + static ConstString g_name("jit"); + return g_name; } -ObjectFile * -ObjectFileJIT::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) -{ - // JIT'ed object file is backed by the ObjectFileJITDelegate, never - // read from a file - return NULL; +const char *ObjectFileJIT::GetPluginDescriptionStatic() { + return "JIT code object file"; } -ObjectFile * -ObjectFileJIT::CreateMemoryInstance (const lldb::ModuleSP &module_sp, - DataBufferSP& data_sp, - const ProcessSP &process_sp, - lldb::addr_t header_addr) -{ - // JIT'ed object file is backed by the ObjectFileJITDelegate, never - // read from memory - return NULL; +ObjectFile *ObjectFileJIT::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) { + // JIT'ed object file is backed by the ObjectFileJITDelegate, never + // read from a file + return NULL; } -size_t -ObjectFileJIT::GetModuleSpecifications (const lldb_private::FileSpec& file, - lldb::DataBufferSP& data_sp, - lldb::offset_t data_offset, - lldb::offset_t file_offset, - lldb::offset_t length, - lldb_private::ModuleSpecList &specs) -{ - // JIT'ed object file can't be read from a file on disk - return 0; +ObjectFile *ObjectFileJIT::CreateMemoryInstance(const lldb::ModuleSP &module_sp, + DataBufferSP &data_sp, + const ProcessSP &process_sp, + lldb::addr_t header_addr) { + // JIT'ed object file is backed by the ObjectFileJITDelegate, never + // read from memory + return NULL; } -ObjectFileJIT::ObjectFileJIT (const lldb::ModuleSP &module_sp, - const ObjectFileJITDelegateSP &delegate_sp) : - ObjectFile(module_sp, NULL, 0, 0, DataBufferSP(), 0), - m_delegate_wp () -{ - if (delegate_sp) - { - m_delegate_wp = delegate_sp; - m_data.SetByteOrder(delegate_sp->GetByteOrder()); - m_data.SetAddressByteSize(delegate_sp->GetAddressByteSize()); - } +size_t ObjectFileJIT::GetModuleSpecifications( + const lldb_private::FileSpec &file, lldb::DataBufferSP &data_sp, + lldb::offset_t data_offset, lldb::offset_t file_offset, + lldb::offset_t length, lldb_private::ModuleSpecList &specs) { + // JIT'ed object file can't be read from a file on disk + return 0; } -ObjectFileJIT::~ObjectFileJIT() -{ +ObjectFileJIT::ObjectFileJIT(const lldb::ModuleSP &module_sp, + const ObjectFileJITDelegateSP &delegate_sp) + : ObjectFile(module_sp, NULL, 0, 0, DataBufferSP(), 0), m_delegate_wp() { + if (delegate_sp) { + m_delegate_wp = delegate_sp; + m_data.SetByteOrder(delegate_sp->GetByteOrder()); + m_data.SetAddressByteSize(delegate_sp->GetAddressByteSize()); + } } +ObjectFileJIT::~ObjectFileJIT() {} -bool -ObjectFileJIT::ParseHeader () -{ - // JIT code is never in a file, nor is it required to have any header - return false; +bool ObjectFileJIT::ParseHeader() { + // JIT code is never in a file, nor is it required to have any header + return false; } -ByteOrder -ObjectFileJIT::GetByteOrder () const -{ - return m_data.GetByteOrder(); -} +ByteOrder ObjectFileJIT::GetByteOrder() const { return m_data.GetByteOrder(); } -bool -ObjectFileJIT::IsExecutable() const -{ - return false; -} +bool ObjectFileJIT::IsExecutable() const { return false; } -uint32_t -ObjectFileJIT::GetAddressByteSize () const -{ - return m_data.GetAddressByteSize(); +uint32_t ObjectFileJIT::GetAddressByteSize() const { + return m_data.GetAddressByteSize(); } - -Symtab * -ObjectFileJIT::GetSymtab() -{ - ModuleSP module_sp(GetModule()); - if (module_sp) - { - std::lock_guard<std::recursive_mutex> guard(module_sp->GetMutex()); - if (m_symtab_ap.get() == NULL) - { - m_symtab_ap.reset(new Symtab(this)); - std::lock_guard<std::recursive_mutex> symtab_guard(m_symtab_ap->GetMutex()); - ObjectFileJITDelegateSP delegate_sp (m_delegate_wp.lock()); - if (delegate_sp) - delegate_sp->PopulateSymtab(this, *m_symtab_ap); - // TODO: get symbols from delegate - m_symtab_ap->Finalize (); - } +Symtab *ObjectFileJIT::GetSymtab() { + ModuleSP module_sp(GetModule()); + if (module_sp) { + std::lock_guard<std::recursive_mutex> guard(module_sp->GetMutex()); + if (m_symtab_ap.get() == NULL) { + m_symtab_ap.reset(new Symtab(this)); + std::lock_guard<std::recursive_mutex> symtab_guard( + m_symtab_ap->GetMutex()); + ObjectFileJITDelegateSP delegate_sp(m_delegate_wp.lock()); + if (delegate_sp) + delegate_sp->PopulateSymtab(this, *m_symtab_ap); + // TODO: get symbols from delegate + m_symtab_ap->Finalize(); } - return m_symtab_ap.get(); + } + return m_symtab_ap.get(); } -bool -ObjectFileJIT::IsStripped () -{ - return false; // JIT code that is in a module is never stripped +bool ObjectFileJIT::IsStripped() { + return false; // JIT code that is in a module is never stripped } -void -ObjectFileJIT::CreateSections (SectionList &unified_section_list) -{ - if (!m_sections_ap.get()) - { - m_sections_ap.reset(new SectionList()); - ObjectFileJITDelegateSP delegate_sp (m_delegate_wp.lock()); - if (delegate_sp) - { - delegate_sp->PopulateSectionList(this, *m_sections_ap); - unified_section_list = *m_sections_ap; - } +void ObjectFileJIT::CreateSections(SectionList &unified_section_list) { + if (!m_sections_ap.get()) { + m_sections_ap.reset(new SectionList()); + ObjectFileJITDelegateSP delegate_sp(m_delegate_wp.lock()); + if (delegate_sp) { + delegate_sp->PopulateSectionList(this, *m_sections_ap); + unified_section_list = *m_sections_ap; } + } } -void -ObjectFileJIT::Dump (Stream *s) -{ - ModuleSP module_sp(GetModule()); - if (module_sp) - { - std::lock_guard<std::recursive_mutex> guard(module_sp->GetMutex()); - s->Printf("%p: ", static_cast<void*>(this)); - s->Indent(); - s->PutCString("ObjectFileJIT"); - - ArchSpec arch; - if (GetArchitecture(arch)) - *s << ", arch = " << arch.GetArchitectureName(); - - s->EOL(); - - SectionList *sections = GetSectionList(); - if (sections) - sections->Dump(s, NULL, true, UINT32_MAX); - - if (m_symtab_ap.get()) - m_symtab_ap->Dump(s, NULL, eSortOrderNone); - } -} +void ObjectFileJIT::Dump(Stream *s) { + ModuleSP module_sp(GetModule()); + if (module_sp) { + std::lock_guard<std::recursive_mutex> guard(module_sp->GetMutex()); + s->Printf("%p: ", static_cast<void *>(this)); + s->Indent(); + s->PutCString("ObjectFileJIT"); -bool -ObjectFileJIT::GetUUID (lldb_private::UUID* uuid) -{ - // TODO: maybe get from delegate, not needed for first pass - return false; -} + ArchSpec arch; + if (GetArchitecture(arch)) + *s << ", arch = " << arch.GetArchitectureName(); + s->EOL(); -uint32_t -ObjectFileJIT::GetDependentModules (FileSpecList& files) -{ - // JIT modules don't have dependencies, but they could - // if external functions are called and we know where they are - files.Clear(); - return 0; -} + SectionList *sections = GetSectionList(); + if (sections) + sections->Dump(s, NULL, true, UINT32_MAX); -lldb_private::Address -ObjectFileJIT::GetEntryPointAddress () -{ - return Address(); + if (m_symtab_ap.get()) + m_symtab_ap->Dump(s, NULL, eSortOrderNone); + } } -lldb_private::Address -ObjectFileJIT::GetHeaderAddress () -{ - return Address(); +bool ObjectFileJIT::GetUUID(lldb_private::UUID *uuid) { + // TODO: maybe get from delegate, not needed for first pass + return false; } - - -ObjectFile::Type -ObjectFileJIT::CalculateType() -{ - return eTypeJIT; +uint32_t ObjectFileJIT::GetDependentModules(FileSpecList &files) { + // JIT modules don't have dependencies, but they could + // if external functions are called and we know where they are + files.Clear(); + return 0; } -ObjectFile::Strata -ObjectFileJIT::CalculateStrata() -{ - return eStrataJIT; +lldb_private::Address ObjectFileJIT::GetEntryPointAddress() { + return Address(); } +lldb_private::Address ObjectFileJIT::GetHeaderAddress() { return Address(); } + +ObjectFile::Type ObjectFileJIT::CalculateType() { return eTypeJIT; } + +ObjectFile::Strata ObjectFileJIT::CalculateStrata() { return eStrataJIT; } -bool -ObjectFileJIT::GetArchitecture (ArchSpec &arch) -{ - ObjectFileJITDelegateSP delegate_sp (m_delegate_wp.lock()); - if (delegate_sp) - return delegate_sp->GetArchitecture(arch); - return false; +bool ObjectFileJIT::GetArchitecture(ArchSpec &arch) { + ObjectFileJITDelegateSP delegate_sp(m_delegate_wp.lock()); + if (delegate_sp) + return delegate_sp->GetArchitecture(arch); + return false; } //------------------------------------------------------------------ // PluginInterface protocol //------------------------------------------------------------------ -lldb_private::ConstString -ObjectFileJIT::GetPluginName() -{ - return GetPluginNameStatic(); +lldb_private::ConstString ObjectFileJIT::GetPluginName() { + return GetPluginNameStatic(); } -uint32_t -ObjectFileJIT::GetPluginVersion() -{ - return 1; -} - - -bool -ObjectFileJIT::SetLoadAddress (Target &target, - lldb::addr_t value, - bool value_is_offset) -{ - size_t num_loaded_sections = 0; - SectionList *section_list = GetSectionList (); - if (section_list) - { - const size_t num_sections = section_list->GetSize(); - // "value" is an offset to apply to each top level segment - for (size_t sect_idx = 0; sect_idx < num_sections; ++sect_idx) - { - // Iterate through the object file sections to find all - // of the sections that size on disk (to avoid __PAGEZERO) - // and load them - SectionSP section_sp (section_list->GetSectionAtIndex (sect_idx)); - if (section_sp && - section_sp->GetFileSize() > 0 && - section_sp->IsThreadSpecific() == false) - { - if (target.GetSectionLoadList().SetSectionLoadAddress (section_sp, section_sp->GetFileAddress() + value)) - ++num_loaded_sections; - } - } +uint32_t ObjectFileJIT::GetPluginVersion() { return 1; } + +bool ObjectFileJIT::SetLoadAddress(Target &target, lldb::addr_t value, + bool value_is_offset) { + size_t num_loaded_sections = 0; + SectionList *section_list = GetSectionList(); + if (section_list) { + const size_t num_sections = section_list->GetSize(); + // "value" is an offset to apply to each top level segment + for (size_t sect_idx = 0; sect_idx < num_sections; ++sect_idx) { + // Iterate through the object file sections to find all + // of the sections that size on disk (to avoid __PAGEZERO) + // and load them + SectionSP section_sp(section_list->GetSectionAtIndex(sect_idx)); + if (section_sp && section_sp->GetFileSize() > 0 && + section_sp->IsThreadSpecific() == false) { + if (target.GetSectionLoadList().SetSectionLoadAddress( + section_sp, section_sp->GetFileAddress() + value)) + ++num_loaded_sections; + } } - return num_loaded_sections > 0; + } + return num_loaded_sections > 0; } - -size_t -ObjectFileJIT::ReadSectionData (const lldb_private::Section *section, - lldb::offset_t section_offset, - void *dst, - size_t dst_len) const -{ - lldb::offset_t file_size = section->GetFileSize(); - if (section_offset < file_size) - { - size_t src_len = file_size - section_offset; - if (src_len > dst_len) - src_len = dst_len; - const uint8_t *src = ((uint8_t *)(uintptr_t)section->GetFileOffset()) + section_offset; - - memcpy (dst, src, src_len); - return src_len; - } - return 0; +size_t ObjectFileJIT::ReadSectionData(const lldb_private::Section *section, + lldb::offset_t section_offset, void *dst, + size_t dst_len) const { + lldb::offset_t file_size = section->GetFileSize(); + if (section_offset < file_size) { + size_t src_len = file_size - section_offset; + if (src_len > dst_len) + src_len = dst_len; + const uint8_t *src = + ((uint8_t *)(uintptr_t)section->GetFileOffset()) + section_offset; + + memcpy(dst, src, src_len); + return src_len; + } + return 0; } -size_t -ObjectFileJIT::ReadSectionData (const lldb_private::Section *section, - lldb_private::DataExtractor& section_data) const -{ - if (section->GetFileSize()) - { - const void *src = (void *)(uintptr_t)section->GetFileOffset(); - - DataBufferSP data_sp (new lldb_private::DataBufferHeap(src, section->GetFileSize())); - if (data_sp) - { - section_data.SetData (data_sp, 0, data_sp->GetByteSize()); - section_data.SetByteOrder (GetByteOrder()); - section_data.SetAddressByteSize (GetAddressByteSize()); - return section_data.GetByteSize(); - } +size_t ObjectFileJIT::ReadSectionData( + const lldb_private::Section *section, + lldb_private::DataExtractor §ion_data) const { + if (section->GetFileSize()) { + const void *src = (void *)(uintptr_t)section->GetFileOffset(); + + DataBufferSP data_sp( + new lldb_private::DataBufferHeap(src, section->GetFileSize())); + if (data_sp) { + section_data.SetData(data_sp, 0, data_sp->GetByteSize()); + section_data.SetByteOrder(GetByteOrder()); + section_data.SetAddressByteSize(GetAddressByteSize()); + return section_data.GetByteSize(); } - section_data.Clear(); - return 0; + } + section_data.Clear(); + return 0; } - diff --git a/source/Plugins/ObjectFile/JIT/ObjectFileJIT.h b/source/Plugins/ObjectFile/JIT/ObjectFileJIT.h index 39dbb3fb047b4..a211645d5d8b3 100644 --- a/source/Plugins/ObjectFile/JIT/ObjectFileJIT.h +++ b/source/Plugins/ObjectFile/JIT/ObjectFileJIT.h @@ -21,126 +21,93 @@ // This class needs to be hidden as eventually belongs in a plugin that // will export the ObjectFile protocol //---------------------------------------------------------------------- -class ObjectFileJIT : - public lldb_private::ObjectFile -{ +class ObjectFileJIT : public lldb_private::ObjectFile { public: - ObjectFileJIT(const lldb::ModuleSP &module_sp, - const lldb::ObjectFileJITDelegateSP &delegate_sp); - - ~ObjectFileJIT() override; - - //------------------------------------------------------------------ - // Static Functions - //------------------------------------------------------------------ - static void - Initialize(); - - static void - Terminate(); - - static lldb_private::ConstString - GetPluginNameStatic(); - - static const char * - GetPluginDescriptionStatic(); - - static lldb_private::ObjectFile * - CreateInstance (const lldb::ModuleSP &module_sp, - lldb::DataBufferSP& data_sp, - lldb::offset_t data_offset, - const lldb_private::FileSpec* file, - lldb::offset_t file_offset, - lldb::offset_t length); - - static lldb_private::ObjectFile * - CreateMemoryInstance (const lldb::ModuleSP &module_sp, - lldb::DataBufferSP& data_sp, - const lldb::ProcessSP &process_sp, - lldb::addr_t header_addr); - - static size_t - GetModuleSpecifications (const lldb_private::FileSpec& file, - lldb::DataBufferSP& data_sp, - lldb::offset_t data_offset, - lldb::offset_t file_offset, - lldb::offset_t length, - lldb_private::ModuleSpecList &specs); - - //------------------------------------------------------------------ - // Member Functions - //------------------------------------------------------------------ - bool - ParseHeader() override; - - bool - SetLoadAddress(lldb_private::Target &target, - lldb::addr_t value, - bool value_is_offset) override; - - lldb::ByteOrder - GetByteOrder() const override; - - bool - IsExecutable() const override; - - uint32_t - GetAddressByteSize() const override; - - lldb_private::Symtab * - GetSymtab() override; - - bool - IsStripped() override; - - void - CreateSections(lldb_private::SectionList &unified_section_list) override; - - void - Dump(lldb_private::Stream *s) override; - - bool - GetArchitecture(lldb_private::ArchSpec &arch) override; - - bool - GetUUID(lldb_private::UUID* uuid) override; - - uint32_t - GetDependentModules(lldb_private::FileSpecList& files) override; - - size_t - ReadSectionData(const lldb_private::Section *section, - lldb::offset_t section_offset, - void *dst, - size_t dst_len) const override; - - size_t - ReadSectionData(const lldb_private::Section *section, - lldb_private::DataExtractor& section_data) const override; - - lldb_private::Address - GetEntryPointAddress() override; - - lldb_private::Address - GetHeaderAddress() override; - - ObjectFile::Type - CalculateType() override; - - ObjectFile::Strata - CalculateStrata() override; - - //------------------------------------------------------------------ - // PluginInterface protocol - //------------------------------------------------------------------ - lldb_private::ConstString - GetPluginName() override; - - uint32_t - GetPluginVersion() override; + ObjectFileJIT(const lldb::ModuleSP &module_sp, + const lldb::ObjectFileJITDelegateSP &delegate_sp); + + ~ObjectFileJIT() override; + + //------------------------------------------------------------------ + // Static Functions + //------------------------------------------------------------------ + static void Initialize(); + + static void Terminate(); + + static lldb_private::ConstString GetPluginNameStatic(); + + static const char *GetPluginDescriptionStatic(); + + static lldb_private::ObjectFile * + CreateInstance(const lldb::ModuleSP &module_sp, lldb::DataBufferSP &data_sp, + lldb::offset_t data_offset, const lldb_private::FileSpec *file, + lldb::offset_t file_offset, lldb::offset_t length); + + static lldb_private::ObjectFile *CreateMemoryInstance( + const lldb::ModuleSP &module_sp, lldb::DataBufferSP &data_sp, + const lldb::ProcessSP &process_sp, lldb::addr_t header_addr); + + static size_t GetModuleSpecifications(const lldb_private::FileSpec &file, + lldb::DataBufferSP &data_sp, + lldb::offset_t data_offset, + lldb::offset_t file_offset, + lldb::offset_t length, + lldb_private::ModuleSpecList &specs); + + //------------------------------------------------------------------ + // Member Functions + //------------------------------------------------------------------ + bool ParseHeader() override; + + bool SetLoadAddress(lldb_private::Target &target, lldb::addr_t value, + bool value_is_offset) override; + + lldb::ByteOrder GetByteOrder() const override; + + bool IsExecutable() const override; + + uint32_t GetAddressByteSize() const override; + + lldb_private::Symtab *GetSymtab() override; + + bool IsStripped() override; + + void CreateSections(lldb_private::SectionList &unified_section_list) override; + + void Dump(lldb_private::Stream *s) override; + + bool GetArchitecture(lldb_private::ArchSpec &arch) override; + + bool GetUUID(lldb_private::UUID *uuid) override; + + uint32_t GetDependentModules(lldb_private::FileSpecList &files) override; + + size_t ReadSectionData(const lldb_private::Section *section, + lldb::offset_t section_offset, void *dst, + size_t dst_len) const override; + + size_t + ReadSectionData(const lldb_private::Section *section, + lldb_private::DataExtractor §ion_data) const override; + + lldb_private::Address GetEntryPointAddress() override; + + lldb_private::Address GetHeaderAddress() override; + + ObjectFile::Type CalculateType() override; + + ObjectFile::Strata CalculateStrata() override; + + //------------------------------------------------------------------ + // PluginInterface protocol + //------------------------------------------------------------------ + lldb_private::ConstString GetPluginName() override; + + uint32_t GetPluginVersion() override; protected: - lldb::ObjectFileJITDelegateWP m_delegate_wp; + lldb::ObjectFileJITDelegateWP m_delegate_wp; }; #endif // liblldb_ObjectFileJIT_h_ diff --git a/source/Plugins/ObjectFile/Mach-O/ObjectFileMachO.cpp b/source/Plugins/ObjectFile/Mach-O/ObjectFileMachO.cpp index 5f8242211b7f6..9ef4e9c3c9b19 100644 --- a/source/Plugins/ObjectFile/Mach-O/ObjectFileMachO.cpp +++ b/source/Plugins/ObjectFile/Mach-O/ObjectFileMachO.cpp @@ -10,9 +10,13 @@ // C Includes // C++ Includes // Other libraries and framework includes -#include "llvm/ADT/StringRef.h" +#include "llvm/ADT/StringRef.h" // Project includes +#include "Plugins/Process/Utility/RegisterContextDarwin_arm.h" +#include "Plugins/Process/Utility/RegisterContextDarwin_arm64.h" +#include "Plugins/Process/Utility/RegisterContextDarwin_i386.h" +#include "Plugins/Process/Utility/RegisterContextDarwin_x86_64.h" #include "lldb/Core/ArchSpec.h" #include "lldb/Core/DataBuffer.h" #include "lldb/Core/Debugger.h" @@ -28,10 +32,11 @@ #include "lldb/Core/StreamString.h" #include "lldb/Core/Timer.h" #include "lldb/Core/UUID.h" -#include "lldb/Host/Host.h" #include "lldb/Host/FileSpec.h" +#include "lldb/Host/Host.h" #include "lldb/Symbol/DWARFCallFrameInfo.h" #include "lldb/Symbol/ObjectFile.h" +#include "lldb/Target/DynamicLoader.h" #include "lldb/Target/MemoryRegionInfo.h" #include "lldb/Target/Platform.h" #include "lldb/Target/Process.h" @@ -39,16 +44,13 @@ #include "lldb/Target/Target.h" #include "lldb/Target/Thread.h" #include "lldb/Target/ThreadList.h" -#include "Plugins/Process/Utility/RegisterContextDarwin_arm.h" -#include "Plugins/Process/Utility/RegisterContextDarwin_arm64.h" -#include "Plugins/Process/Utility/RegisterContextDarwin_i386.h" -#include "Plugins/Process/Utility/RegisterContextDarwin_x86_64.h" #include "lldb/Utility/SafeMachO.h" #include "ObjectFileMachO.h" -#if defined (__APPLE__) && (defined (__arm__) || defined (__arm64__) || defined (__aarch64__)) +#if defined(__APPLE__) && \ + (defined(__arm__) || defined(__arm64__) || defined(__aarch64__)) // GetLLDBSharedCacheUUID() needs to call dlsym() #include <dlfcn.h> #endif @@ -65,3870 +67,2966 @@ using namespace llvm::MachO; // Some structure definitions needed for parsing the dyld shared cache files // found on iOS devices. -struct lldb_copy_dyld_cache_header_v1 -{ - char magic[16]; // e.g. "dyld_v0 i386", "dyld_v1 armv7", etc. - uint32_t mappingOffset; // file offset to first dyld_cache_mapping_info - uint32_t mappingCount; // number of dyld_cache_mapping_info entries - uint32_t imagesOffset; - uint32_t imagesCount; - uint64_t dyldBaseAddress; - uint64_t codeSignatureOffset; - uint64_t codeSignatureSize; - uint64_t slideInfoOffset; - uint64_t slideInfoSize; - uint64_t localSymbolsOffset; - uint64_t localSymbolsSize; - uint8_t uuid[16]; // v1 and above, also recorded in dyld_all_image_infos v13 and later +struct lldb_copy_dyld_cache_header_v1 { + char magic[16]; // e.g. "dyld_v0 i386", "dyld_v1 armv7", etc. + uint32_t mappingOffset; // file offset to first dyld_cache_mapping_info + uint32_t mappingCount; // number of dyld_cache_mapping_info entries + uint32_t imagesOffset; + uint32_t imagesCount; + uint64_t dyldBaseAddress; + uint64_t codeSignatureOffset; + uint64_t codeSignatureSize; + uint64_t slideInfoOffset; + uint64_t slideInfoSize; + uint64_t localSymbolsOffset; + uint64_t localSymbolsSize; + uint8_t uuid[16]; // v1 and above, also recorded in dyld_all_image_infos v13 + // and later }; -struct lldb_copy_dyld_cache_mapping_info -{ - uint64_t address; - uint64_t size; - uint64_t fileOffset; - uint32_t maxProt; - uint32_t initProt; +struct lldb_copy_dyld_cache_mapping_info { + uint64_t address; + uint64_t size; + uint64_t fileOffset; + uint32_t maxProt; + uint32_t initProt; }; -struct lldb_copy_dyld_cache_local_symbols_info -{ - uint32_t nlistOffset; - uint32_t nlistCount; - uint32_t stringsOffset; - uint32_t stringsSize; - uint32_t entriesOffset; - uint32_t entriesCount; +struct lldb_copy_dyld_cache_local_symbols_info { + uint32_t nlistOffset; + uint32_t nlistCount; + uint32_t stringsOffset; + uint32_t stringsSize; + uint32_t entriesOffset; + uint32_t entriesCount; }; -struct lldb_copy_dyld_cache_local_symbols_entry -{ - uint32_t dylibOffset; - uint32_t nlistStartIndex; - uint32_t nlistCount; +struct lldb_copy_dyld_cache_local_symbols_entry { + uint32_t dylibOffset; + uint32_t nlistStartIndex; + uint32_t nlistCount; }; - -class RegisterContextDarwin_x86_64_Mach : public RegisterContextDarwin_x86_64 -{ +class RegisterContextDarwin_x86_64_Mach : public RegisterContextDarwin_x86_64 { public: - RegisterContextDarwin_x86_64_Mach (lldb_private::Thread &thread, const DataExtractor &data) : - RegisterContextDarwin_x86_64 (thread, 0) - { - SetRegisterDataFrom_LC_THREAD (data); - } + RegisterContextDarwin_x86_64_Mach(lldb_private::Thread &thread, + const DataExtractor &data) + : RegisterContextDarwin_x86_64(thread, 0) { + SetRegisterDataFrom_LC_THREAD(data); + } - void - InvalidateAllRegisters() override - { - // Do nothing... registers are always valid... - } - - void - SetRegisterDataFrom_LC_THREAD (const DataExtractor &data) - { - lldb::offset_t offset = 0; - SetError (GPRRegSet, Read, -1); - SetError (FPURegSet, Read, -1); - SetError (EXCRegSet, Read, -1); - bool done = false; - - while (!done) - { - int flavor = data.GetU32 (&offset); - if (flavor == 0) - done = true; - else - { - uint32_t i; - uint32_t count = data.GetU32 (&offset); - switch (flavor) - { - case GPRRegSet: - for (i=0; i<count; ++i) - (&gpr.rax)[i] = data.GetU64(&offset); - SetError (GPRRegSet, Read, 0); - done = true; + void InvalidateAllRegisters() override { + // Do nothing... registers are always valid... + } - break; - case FPURegSet: - // TODO: fill in FPU regs.... - //SetError (FPURegSet, Read, -1); - done = true; - - break; - case EXCRegSet: - exc.trapno = data.GetU32(&offset); - exc.err = data.GetU32(&offset); - exc.faultvaddr = data.GetU64(&offset); - SetError (EXCRegSet, Read, 0); - done = true; - break; - case 7: - case 8: - case 9: - // fancy flavors that encapsulate of the above - // flavors... - break; + void SetRegisterDataFrom_LC_THREAD(const DataExtractor &data) { + lldb::offset_t offset = 0; + SetError(GPRRegSet, Read, -1); + SetError(FPURegSet, Read, -1); + SetError(EXCRegSet, Read, -1); + bool done = false; + + while (!done) { + int flavor = data.GetU32(&offset); + if (flavor == 0) + done = true; + else { + uint32_t i; + uint32_t count = data.GetU32(&offset); + switch (flavor) { + case GPRRegSet: + for (i = 0; i < count; ++i) + (&gpr.rax)[i] = data.GetU64(&offset); + SetError(GPRRegSet, Read, 0); + done = true; + + break; + case FPURegSet: + // TODO: fill in FPU regs.... + // SetError (FPURegSet, Read, -1); + done = true; + + break; + case EXCRegSet: + exc.trapno = data.GetU32(&offset); + exc.err = data.GetU32(&offset); + exc.faultvaddr = data.GetU64(&offset); + SetError(EXCRegSet, Read, 0); + done = true; + break; + case 7: + case 8: + case 9: + // fancy flavors that encapsulate of the above + // flavors... + break; - default: - done = true; - break; - } - } + default: + done = true; + break; } + } } - - static size_t - WriteRegister (RegisterContext *reg_ctx, const char *name, const char *alt_name, size_t reg_byte_size, Stream &data) - { - const RegisterInfo *reg_info = reg_ctx->GetRegisterInfoByName(name); - if (reg_info == NULL) - reg_info = reg_ctx->GetRegisterInfoByName(alt_name); - if (reg_info) - { - lldb_private::RegisterValue reg_value; - if (reg_ctx->ReadRegister(reg_info, reg_value)) - { - if (reg_info->byte_size >= reg_byte_size) - data.Write(reg_value.GetBytes(), reg_byte_size); - else - { - data.Write(reg_value.GetBytes(), reg_info->byte_size); - for (size_t i=0, n = reg_byte_size - reg_info->byte_size; i<n; ++ i) - data.PutChar(0); - } - return reg_byte_size; - } - } - // Just write zeros if all else fails - for (size_t i=0; i<reg_byte_size; ++ i) + } + + static size_t WriteRegister(RegisterContext *reg_ctx, const char *name, + const char *alt_name, size_t reg_byte_size, + Stream &data) { + const RegisterInfo *reg_info = reg_ctx->GetRegisterInfoByName(name); + if (reg_info == NULL) + reg_info = reg_ctx->GetRegisterInfoByName(alt_name); + if (reg_info) { + lldb_private::RegisterValue reg_value; + if (reg_ctx->ReadRegister(reg_info, reg_value)) { + if (reg_info->byte_size >= reg_byte_size) + data.Write(reg_value.GetBytes(), reg_byte_size); + else { + data.Write(reg_value.GetBytes(), reg_info->byte_size); + for (size_t i = 0, n = reg_byte_size - reg_info->byte_size; i < n; + ++i) data.PutChar(0); + } return reg_byte_size; + } } - - static bool - Create_LC_THREAD (Thread *thread, Stream &data) - { - RegisterContextSP reg_ctx_sp (thread->GetRegisterContext()); - if (reg_ctx_sp) - { - RegisterContext *reg_ctx = reg_ctx_sp.get(); - - data.PutHex32 (GPRRegSet); // Flavor - data.PutHex32 (GPRWordCount); - WriteRegister (reg_ctx, "rax", NULL, 8, data); - WriteRegister (reg_ctx, "rbx", NULL, 8, data); - WriteRegister (reg_ctx, "rcx", NULL, 8, data); - WriteRegister (reg_ctx, "rdx", NULL, 8, data); - WriteRegister (reg_ctx, "rdi", NULL, 8, data); - WriteRegister (reg_ctx, "rsi", NULL, 8, data); - WriteRegister (reg_ctx, "rbp", NULL, 8, data); - WriteRegister (reg_ctx, "rsp", NULL, 8, data); - WriteRegister (reg_ctx, "r8", NULL, 8, data); - WriteRegister (reg_ctx, "r9", NULL, 8, data); - WriteRegister (reg_ctx, "r10", NULL, 8, data); - WriteRegister (reg_ctx, "r11", NULL, 8, data); - WriteRegister (reg_ctx, "r12", NULL, 8, data); - WriteRegister (reg_ctx, "r13", NULL, 8, data); - WriteRegister (reg_ctx, "r14", NULL, 8, data); - WriteRegister (reg_ctx, "r15", NULL, 8, data); - WriteRegister (reg_ctx, "rip", NULL, 8, data); - WriteRegister (reg_ctx, "rflags", NULL, 8, data); - WriteRegister (reg_ctx, "cs", NULL, 8, data); - WriteRegister (reg_ctx, "fs", NULL, 8, data); - WriteRegister (reg_ctx, "gs", NULL, 8, data); - -// // Write out the FPU registers -// const size_t fpu_byte_size = sizeof(FPU); -// size_t bytes_written = 0; -// data.PutHex32 (FPURegSet); -// data.PutHex32 (fpu_byte_size/sizeof(uint64_t)); -// bytes_written += data.PutHex32(0); // uint32_t pad[0] -// bytes_written += data.PutHex32(0); // uint32_t pad[1] -// bytes_written += WriteRegister (reg_ctx, "fcw", "fctrl", 2, data); // uint16_t fcw; // "fctrl" -// bytes_written += WriteRegister (reg_ctx, "fsw" , "fstat", 2, data); // uint16_t fsw; // "fstat" -// bytes_written += WriteRegister (reg_ctx, "ftw" , "ftag", 1, data); // uint8_t ftw; // "ftag" -// bytes_written += data.PutHex8 (0); // uint8_t pad1; -// bytes_written += WriteRegister (reg_ctx, "fop" , NULL, 2, data); // uint16_t fop; // "fop" -// bytes_written += WriteRegister (reg_ctx, "fioff", "ip", 4, data); // uint32_t ip; // "fioff" -// bytes_written += WriteRegister (reg_ctx, "fiseg", NULL, 2, data); // uint16_t cs; // "fiseg" -// bytes_written += data.PutHex16 (0); // uint16_t pad2; -// bytes_written += WriteRegister (reg_ctx, "dp", "fooff" , 4, data); // uint32_t dp; // "fooff" -// bytes_written += WriteRegister (reg_ctx, "foseg", NULL, 2, data); // uint16_t ds; // "foseg" -// bytes_written += data.PutHex16 (0); // uint16_t pad3; -// bytes_written += WriteRegister (reg_ctx, "mxcsr", NULL, 4, data); // uint32_t mxcsr; -// bytes_written += WriteRegister (reg_ctx, "mxcsrmask", NULL, 4, data);// uint32_t mxcsrmask; -// bytes_written += WriteRegister (reg_ctx, "stmm0", NULL, sizeof(MMSReg), data); -// bytes_written += WriteRegister (reg_ctx, "stmm1", NULL, sizeof(MMSReg), data); -// bytes_written += WriteRegister (reg_ctx, "stmm2", NULL, sizeof(MMSReg), data); -// bytes_written += WriteRegister (reg_ctx, "stmm3", NULL, sizeof(MMSReg), data); -// bytes_written += WriteRegister (reg_ctx, "stmm4", NULL, sizeof(MMSReg), data); -// bytes_written += WriteRegister (reg_ctx, "stmm5", NULL, sizeof(MMSReg), data); -// bytes_written += WriteRegister (reg_ctx, "stmm6", NULL, sizeof(MMSReg), data); -// bytes_written += WriteRegister (reg_ctx, "stmm7", NULL, sizeof(MMSReg), data); -// bytes_written += WriteRegister (reg_ctx, "xmm0" , NULL, sizeof(XMMReg), data); -// bytes_written += WriteRegister (reg_ctx, "xmm1" , NULL, sizeof(XMMReg), data); -// bytes_written += WriteRegister (reg_ctx, "xmm2" , NULL, sizeof(XMMReg), data); -// bytes_written += WriteRegister (reg_ctx, "xmm3" , NULL, sizeof(XMMReg), data); -// bytes_written += WriteRegister (reg_ctx, "xmm4" , NULL, sizeof(XMMReg), data); -// bytes_written += WriteRegister (reg_ctx, "xmm5" , NULL, sizeof(XMMReg), data); -// bytes_written += WriteRegister (reg_ctx, "xmm6" , NULL, sizeof(XMMReg), data); -// bytes_written += WriteRegister (reg_ctx, "xmm7" , NULL, sizeof(XMMReg), data); -// bytes_written += WriteRegister (reg_ctx, "xmm8" , NULL, sizeof(XMMReg), data); -// bytes_written += WriteRegister (reg_ctx, "xmm9" , NULL, sizeof(XMMReg), data); -// bytes_written += WriteRegister (reg_ctx, "xmm10", NULL, sizeof(XMMReg), data); -// bytes_written += WriteRegister (reg_ctx, "xmm11", NULL, sizeof(XMMReg), data); -// bytes_written += WriteRegister (reg_ctx, "xmm12", NULL, sizeof(XMMReg), data); -// bytes_written += WriteRegister (reg_ctx, "xmm13", NULL, sizeof(XMMReg), data); -// bytes_written += WriteRegister (reg_ctx, "xmm14", NULL, sizeof(XMMReg), data); -// bytes_written += WriteRegister (reg_ctx, "xmm15", NULL, sizeof(XMMReg), data); -// -// // Fill rest with zeros -// for (size_t i=0, n = fpu_byte_size - bytes_written; i<n; ++ i) -// data.PutChar(0); - - // Write out the EXC registers - data.PutHex32 (EXCRegSet); - data.PutHex32 (EXCWordCount); - WriteRegister (reg_ctx, "trapno", NULL, 4, data); - WriteRegister (reg_ctx, "err", NULL, 4, data); - WriteRegister (reg_ctx, "faultvaddr", NULL, 8, data); - return true; - } - return false; + // Just write zeros if all else fails + for (size_t i = 0; i < reg_byte_size; ++i) + data.PutChar(0); + return reg_byte_size; + } + + static bool Create_LC_THREAD(Thread *thread, Stream &data) { + RegisterContextSP reg_ctx_sp(thread->GetRegisterContext()); + if (reg_ctx_sp) { + RegisterContext *reg_ctx = reg_ctx_sp.get(); + + data.PutHex32(GPRRegSet); // Flavor + data.PutHex32(GPRWordCount); + WriteRegister(reg_ctx, "rax", NULL, 8, data); + WriteRegister(reg_ctx, "rbx", NULL, 8, data); + WriteRegister(reg_ctx, "rcx", NULL, 8, data); + WriteRegister(reg_ctx, "rdx", NULL, 8, data); + WriteRegister(reg_ctx, "rdi", NULL, 8, data); + WriteRegister(reg_ctx, "rsi", NULL, 8, data); + WriteRegister(reg_ctx, "rbp", NULL, 8, data); + WriteRegister(reg_ctx, "rsp", NULL, 8, data); + WriteRegister(reg_ctx, "r8", NULL, 8, data); + WriteRegister(reg_ctx, "r9", NULL, 8, data); + WriteRegister(reg_ctx, "r10", NULL, 8, data); + WriteRegister(reg_ctx, "r11", NULL, 8, data); + WriteRegister(reg_ctx, "r12", NULL, 8, data); + WriteRegister(reg_ctx, "r13", NULL, 8, data); + WriteRegister(reg_ctx, "r14", NULL, 8, data); + WriteRegister(reg_ctx, "r15", NULL, 8, data); + WriteRegister(reg_ctx, "rip", NULL, 8, data); + WriteRegister(reg_ctx, "rflags", NULL, 8, data); + WriteRegister(reg_ctx, "cs", NULL, 8, data); + WriteRegister(reg_ctx, "fs", NULL, 8, data); + WriteRegister(reg_ctx, "gs", NULL, 8, data); + + // // Write out the FPU registers + // const size_t fpu_byte_size = sizeof(FPU); + // size_t bytes_written = 0; + // data.PutHex32 (FPURegSet); + // data.PutHex32 (fpu_byte_size/sizeof(uint64_t)); + // bytes_written += data.PutHex32(0); // uint32_t pad[0] + // bytes_written += data.PutHex32(0); // uint32_t pad[1] + // bytes_written += WriteRegister (reg_ctx, "fcw", "fctrl", 2, + // data); // uint16_t fcw; // "fctrl" + // bytes_written += WriteRegister (reg_ctx, "fsw" , "fstat", 2, + // data); // uint16_t fsw; // "fstat" + // bytes_written += WriteRegister (reg_ctx, "ftw" , "ftag", 1, + // data); // uint8_t ftw; // "ftag" + // bytes_written += data.PutHex8 (0); // uint8_t pad1; + // bytes_written += WriteRegister (reg_ctx, "fop" , NULL, 2, + // data); // uint16_t fop; // "fop" + // bytes_written += WriteRegister (reg_ctx, "fioff", "ip", 4, + // data); // uint32_t ip; // "fioff" + // bytes_written += WriteRegister (reg_ctx, "fiseg", NULL, 2, + // data); // uint16_t cs; // "fiseg" + // bytes_written += data.PutHex16 (0); // uint16_t pad2; + // bytes_written += WriteRegister (reg_ctx, "dp", "fooff" , 4, + // data); // uint32_t dp; // "fooff" + // bytes_written += WriteRegister (reg_ctx, "foseg", NULL, 2, + // data); // uint16_t ds; // "foseg" + // bytes_written += data.PutHex16 (0); // uint16_t pad3; + // bytes_written += WriteRegister (reg_ctx, "mxcsr", NULL, 4, + // data); // uint32_t mxcsr; + // bytes_written += WriteRegister (reg_ctx, "mxcsrmask", NULL, + // 4, data);// uint32_t mxcsrmask; + // bytes_written += WriteRegister (reg_ctx, "stmm0", NULL, + // sizeof(MMSReg), data); + // bytes_written += WriteRegister (reg_ctx, "stmm1", NULL, + // sizeof(MMSReg), data); + // bytes_written += WriteRegister (reg_ctx, "stmm2", NULL, + // sizeof(MMSReg), data); + // bytes_written += WriteRegister (reg_ctx, "stmm3", NULL, + // sizeof(MMSReg), data); + // bytes_written += WriteRegister (reg_ctx, "stmm4", NULL, + // sizeof(MMSReg), data); + // bytes_written += WriteRegister (reg_ctx, "stmm5", NULL, + // sizeof(MMSReg), data); + // bytes_written += WriteRegister (reg_ctx, "stmm6", NULL, + // sizeof(MMSReg), data); + // bytes_written += WriteRegister (reg_ctx, "stmm7", NULL, + // sizeof(MMSReg), data); + // bytes_written += WriteRegister (reg_ctx, "xmm0" , NULL, + // sizeof(XMMReg), data); + // bytes_written += WriteRegister (reg_ctx, "xmm1" , NULL, + // sizeof(XMMReg), data); + // bytes_written += WriteRegister (reg_ctx, "xmm2" , NULL, + // sizeof(XMMReg), data); + // bytes_written += WriteRegister (reg_ctx, "xmm3" , NULL, + // sizeof(XMMReg), data); + // bytes_written += WriteRegister (reg_ctx, "xmm4" , NULL, + // sizeof(XMMReg), data); + // bytes_written += WriteRegister (reg_ctx, "xmm5" , NULL, + // sizeof(XMMReg), data); + // bytes_written += WriteRegister (reg_ctx, "xmm6" , NULL, + // sizeof(XMMReg), data); + // bytes_written += WriteRegister (reg_ctx, "xmm7" , NULL, + // sizeof(XMMReg), data); + // bytes_written += WriteRegister (reg_ctx, "xmm8" , NULL, + // sizeof(XMMReg), data); + // bytes_written += WriteRegister (reg_ctx, "xmm9" , NULL, + // sizeof(XMMReg), data); + // bytes_written += WriteRegister (reg_ctx, "xmm10", NULL, + // sizeof(XMMReg), data); + // bytes_written += WriteRegister (reg_ctx, "xmm11", NULL, + // sizeof(XMMReg), data); + // bytes_written += WriteRegister (reg_ctx, "xmm12", NULL, + // sizeof(XMMReg), data); + // bytes_written += WriteRegister (reg_ctx, "xmm13", NULL, + // sizeof(XMMReg), data); + // bytes_written += WriteRegister (reg_ctx, "xmm14", NULL, + // sizeof(XMMReg), data); + // bytes_written += WriteRegister (reg_ctx, "xmm15", NULL, + // sizeof(XMMReg), data); + // + // // Fill rest with zeros + // for (size_t i=0, n = fpu_byte_size - bytes_written; i<n; ++ + // i) + // data.PutChar(0); + + // Write out the EXC registers + data.PutHex32(EXCRegSet); + data.PutHex32(EXCWordCount); + WriteRegister(reg_ctx, "trapno", NULL, 4, data); + WriteRegister(reg_ctx, "err", NULL, 4, data); + WriteRegister(reg_ctx, "faultvaddr", NULL, 8, data); + return true; } + return false; + } protected: - int - DoReadGPR(lldb::tid_t tid, int flavor, GPR &gpr) override - { - return 0; - } + int DoReadGPR(lldb::tid_t tid, int flavor, GPR &gpr) override { return 0; } - int - DoReadFPU(lldb::tid_t tid, int flavor, FPU &fpu) override - { - return 0; - } + int DoReadFPU(lldb::tid_t tid, int flavor, FPU &fpu) override { return 0; } - int - DoReadEXC(lldb::tid_t tid, int flavor, EXC &exc) override - { - return 0; - } + int DoReadEXC(lldb::tid_t tid, int flavor, EXC &exc) override { return 0; } - int - DoWriteGPR(lldb::tid_t tid, int flavor, const GPR &gpr) override - { - return 0; - } + int DoWriteGPR(lldb::tid_t tid, int flavor, const GPR &gpr) override { + return 0; + } - int - DoWriteFPU(lldb::tid_t tid, int flavor, const FPU &fpu) override - { - return 0; - } + int DoWriteFPU(lldb::tid_t tid, int flavor, const FPU &fpu) override { + return 0; + } - int - DoWriteEXC(lldb::tid_t tid, int flavor, const EXC &exc) override - { - return 0; - } + int DoWriteEXC(lldb::tid_t tid, int flavor, const EXC &exc) override { + return 0; + } }; -class RegisterContextDarwin_i386_Mach : public RegisterContextDarwin_i386 -{ +class RegisterContextDarwin_i386_Mach : public RegisterContextDarwin_i386 { public: - RegisterContextDarwin_i386_Mach (lldb_private::Thread &thread, const DataExtractor &data) : - RegisterContextDarwin_i386 (thread, 0) - { - SetRegisterDataFrom_LC_THREAD (data); - } - - void - InvalidateAllRegisters() override - { - // Do nothing... registers are always valid... - } + RegisterContextDarwin_i386_Mach(lldb_private::Thread &thread, + const DataExtractor &data) + : RegisterContextDarwin_i386(thread, 0) { + SetRegisterDataFrom_LC_THREAD(data); + } - void - SetRegisterDataFrom_LC_THREAD (const DataExtractor &data) - { - lldb::offset_t offset = 0; - SetError (GPRRegSet, Read, -1); - SetError (FPURegSet, Read, -1); - SetError (EXCRegSet, Read, -1); - bool done = false; - - while (!done) - { - int flavor = data.GetU32 (&offset); - if (flavor == 0) - done = true; - else - { - uint32_t i; - uint32_t count = data.GetU32 (&offset); - switch (flavor) - { - case GPRRegSet: - for (i=0; i<count; ++i) - (&gpr.eax)[i] = data.GetU32(&offset); - SetError (GPRRegSet, Read, 0); - done = true; + void InvalidateAllRegisters() override { + // Do nothing... registers are always valid... + } - break; - case FPURegSet: - // TODO: fill in FPU regs.... - //SetError (FPURegSet, Read, -1); - done = true; - - break; - case EXCRegSet: - exc.trapno = data.GetU32(&offset); - exc.err = data.GetU32(&offset); - exc.faultvaddr = data.GetU32(&offset); - SetError (EXCRegSet, Read, 0); - done = true; - break; - case 7: - case 8: - case 9: - // fancy flavors that encapsulate of the above - // flavors... - break; + void SetRegisterDataFrom_LC_THREAD(const DataExtractor &data) { + lldb::offset_t offset = 0; + SetError(GPRRegSet, Read, -1); + SetError(FPURegSet, Read, -1); + SetError(EXCRegSet, Read, -1); + bool done = false; + + while (!done) { + int flavor = data.GetU32(&offset); + if (flavor == 0) + done = true; + else { + uint32_t i; + uint32_t count = data.GetU32(&offset); + switch (flavor) { + case GPRRegSet: + for (i = 0; i < count; ++i) + (&gpr.eax)[i] = data.GetU32(&offset); + SetError(GPRRegSet, Read, 0); + done = true; + + break; + case FPURegSet: + // TODO: fill in FPU regs.... + // SetError (FPURegSet, Read, -1); + done = true; + + break; + case EXCRegSet: + exc.trapno = data.GetU32(&offset); + exc.err = data.GetU32(&offset); + exc.faultvaddr = data.GetU32(&offset); + SetError(EXCRegSet, Read, 0); + done = true; + break; + case 7: + case 8: + case 9: + // fancy flavors that encapsulate of the above + // flavors... + break; - default: - done = true; - break; - } - } + default: + done = true; + break; } + } } - - static size_t - WriteRegister (RegisterContext *reg_ctx, const char *name, const char *alt_name, size_t reg_byte_size, Stream &data) - { - const RegisterInfo *reg_info = reg_ctx->GetRegisterInfoByName(name); - if (reg_info == NULL) - reg_info = reg_ctx->GetRegisterInfoByName(alt_name); - if (reg_info) - { - lldb_private::RegisterValue reg_value; - if (reg_ctx->ReadRegister(reg_info, reg_value)) - { - if (reg_info->byte_size >= reg_byte_size) - data.Write(reg_value.GetBytes(), reg_byte_size); - else - { - data.Write(reg_value.GetBytes(), reg_info->byte_size); - for (size_t i=0, n = reg_byte_size - reg_info->byte_size; i<n; ++ i) - data.PutChar(0); - } - return reg_byte_size; - } - } - // Just write zeros if all else fails - for (size_t i=0; i<reg_byte_size; ++ i) + } + + static size_t WriteRegister(RegisterContext *reg_ctx, const char *name, + const char *alt_name, size_t reg_byte_size, + Stream &data) { + const RegisterInfo *reg_info = reg_ctx->GetRegisterInfoByName(name); + if (reg_info == NULL) + reg_info = reg_ctx->GetRegisterInfoByName(alt_name); + if (reg_info) { + lldb_private::RegisterValue reg_value; + if (reg_ctx->ReadRegister(reg_info, reg_value)) { + if (reg_info->byte_size >= reg_byte_size) + data.Write(reg_value.GetBytes(), reg_byte_size); + else { + data.Write(reg_value.GetBytes(), reg_info->byte_size); + for (size_t i = 0, n = reg_byte_size - reg_info->byte_size; i < n; + ++i) data.PutChar(0); + } return reg_byte_size; + } } - - static bool - Create_LC_THREAD (Thread *thread, Stream &data) - { - RegisterContextSP reg_ctx_sp (thread->GetRegisterContext()); - if (reg_ctx_sp) - { - RegisterContext *reg_ctx = reg_ctx_sp.get(); - - data.PutHex32 (GPRRegSet); // Flavor - data.PutHex32 (GPRWordCount); - WriteRegister (reg_ctx, "eax", NULL, 4, data); - WriteRegister (reg_ctx, "ebx", NULL, 4, data); - WriteRegister (reg_ctx, "ecx", NULL, 4, data); - WriteRegister (reg_ctx, "edx", NULL, 4, data); - WriteRegister (reg_ctx, "edi", NULL, 4, data); - WriteRegister (reg_ctx, "esi", NULL, 4, data); - WriteRegister (reg_ctx, "ebp", NULL, 4, data); - WriteRegister (reg_ctx, "esp", NULL, 4, data); - WriteRegister (reg_ctx, "ss", NULL, 4, data); - WriteRegister (reg_ctx, "eflags", NULL, 4, data); - WriteRegister (reg_ctx, "eip", NULL, 4, data); - WriteRegister (reg_ctx, "cs", NULL, 4, data); - WriteRegister (reg_ctx, "ds", NULL, 4, data); - WriteRegister (reg_ctx, "es", NULL, 4, data); - WriteRegister (reg_ctx, "fs", NULL, 4, data); - WriteRegister (reg_ctx, "gs", NULL, 4, data); - - // Write out the EXC registers - data.PutHex32 (EXCRegSet); - data.PutHex32 (EXCWordCount); - WriteRegister (reg_ctx, "trapno", NULL, 4, data); - WriteRegister (reg_ctx, "err", NULL, 4, data); - WriteRegister (reg_ctx, "faultvaddr", NULL, 4, data); - return true; - } - return false; + // Just write zeros if all else fails + for (size_t i = 0; i < reg_byte_size; ++i) + data.PutChar(0); + return reg_byte_size; + } + + static bool Create_LC_THREAD(Thread *thread, Stream &data) { + RegisterContextSP reg_ctx_sp(thread->GetRegisterContext()); + if (reg_ctx_sp) { + RegisterContext *reg_ctx = reg_ctx_sp.get(); + + data.PutHex32(GPRRegSet); // Flavor + data.PutHex32(GPRWordCount); + WriteRegister(reg_ctx, "eax", NULL, 4, data); + WriteRegister(reg_ctx, "ebx", NULL, 4, data); + WriteRegister(reg_ctx, "ecx", NULL, 4, data); + WriteRegister(reg_ctx, "edx", NULL, 4, data); + WriteRegister(reg_ctx, "edi", NULL, 4, data); + WriteRegister(reg_ctx, "esi", NULL, 4, data); + WriteRegister(reg_ctx, "ebp", NULL, 4, data); + WriteRegister(reg_ctx, "esp", NULL, 4, data); + WriteRegister(reg_ctx, "ss", NULL, 4, data); + WriteRegister(reg_ctx, "eflags", NULL, 4, data); + WriteRegister(reg_ctx, "eip", NULL, 4, data); + WriteRegister(reg_ctx, "cs", NULL, 4, data); + WriteRegister(reg_ctx, "ds", NULL, 4, data); + WriteRegister(reg_ctx, "es", NULL, 4, data); + WriteRegister(reg_ctx, "fs", NULL, 4, data); + WriteRegister(reg_ctx, "gs", NULL, 4, data); + + // Write out the EXC registers + data.PutHex32(EXCRegSet); + data.PutHex32(EXCWordCount); + WriteRegister(reg_ctx, "trapno", NULL, 4, data); + WriteRegister(reg_ctx, "err", NULL, 4, data); + WriteRegister(reg_ctx, "faultvaddr", NULL, 4, data); + return true; } + return false; + } protected: - int - DoReadGPR(lldb::tid_t tid, int flavor, GPR &gpr) override - { - return 0; - } + int DoReadGPR(lldb::tid_t tid, int flavor, GPR &gpr) override { return 0; } - int - DoReadFPU(lldb::tid_t tid, int flavor, FPU &fpu) override - { - return 0; - } + int DoReadFPU(lldb::tid_t tid, int flavor, FPU &fpu) override { return 0; } - int - DoReadEXC(lldb::tid_t tid, int flavor, EXC &exc) override - { - return 0; - } + int DoReadEXC(lldb::tid_t tid, int flavor, EXC &exc) override { return 0; } - int - DoWriteGPR(lldb::tid_t tid, int flavor, const GPR &gpr) override - { - return 0; - } + int DoWriteGPR(lldb::tid_t tid, int flavor, const GPR &gpr) override { + return 0; + } - int - DoWriteFPU(lldb::tid_t tid, int flavor, const FPU &fpu) override - { - return 0; - } + int DoWriteFPU(lldb::tid_t tid, int flavor, const FPU &fpu) override { + return 0; + } - int - DoWriteEXC(lldb::tid_t tid, int flavor, const EXC &exc) override - { - return 0; - } + int DoWriteEXC(lldb::tid_t tid, int flavor, const EXC &exc) override { + return 0; + } }; -class RegisterContextDarwin_arm_Mach : public RegisterContextDarwin_arm -{ +class RegisterContextDarwin_arm_Mach : public RegisterContextDarwin_arm { public: - RegisterContextDarwin_arm_Mach (lldb_private::Thread &thread, const DataExtractor &data) : - RegisterContextDarwin_arm (thread, 0) - { - SetRegisterDataFrom_LC_THREAD (data); - } - - void - InvalidateAllRegisters() override - { - // Do nothing... registers are always valid... - } - - void - SetRegisterDataFrom_LC_THREAD (const DataExtractor &data) - { - lldb::offset_t offset = 0; - SetError (GPRRegSet, Read, -1); - SetError (FPURegSet, Read, -1); - SetError (EXCRegSet, Read, -1); - bool done = false; - - while (!done) - { - int flavor = data.GetU32 (&offset); - uint32_t count = data.GetU32 (&offset); - lldb::offset_t next_thread_state = offset + (count * 4); - switch (flavor) - { - case GPRAltRegSet: - case GPRRegSet: - for (uint32_t i=0; i<count; ++i) - { - gpr.r[i] = data.GetU32(&offset); - } - - // Note that gpr.cpsr is also copied by the above loop; this loop technically extends - // one element past the end of the gpr.r[] array. - - SetError (GPRRegSet, Read, 0); - offset = next_thread_state; - break; + RegisterContextDarwin_arm_Mach(lldb_private::Thread &thread, + const DataExtractor &data) + : RegisterContextDarwin_arm(thread, 0) { + SetRegisterDataFrom_LC_THREAD(data); + } - case FPURegSet: - { - uint8_t *fpu_reg_buf = (uint8_t*) &fpu.floats.s[0]; - const int fpu_reg_buf_size = sizeof (fpu.floats); - if (data.ExtractBytes (offset, fpu_reg_buf_size, eByteOrderLittle, fpu_reg_buf) == fpu_reg_buf_size) - { - offset += fpu_reg_buf_size; - fpu.fpscr = data.GetU32(&offset); - SetError (FPURegSet, Read, 0); - } - else - { - done = true; - } - } - offset = next_thread_state; - break; + void InvalidateAllRegisters() override { + // Do nothing... registers are always valid... + } - case EXCRegSet: - if (count == 3) - { - exc.exception = data.GetU32(&offset); - exc.fsr = data.GetU32(&offset); - exc.far = data.GetU32(&offset); - SetError (EXCRegSet, Read, 0); - } - done = true; - offset = next_thread_state; - break; - - // Unknown register set flavor, stop trying to parse. - default: - done = true; - } + void SetRegisterDataFrom_LC_THREAD(const DataExtractor &data) { + lldb::offset_t offset = 0; + SetError(GPRRegSet, Read, -1); + SetError(FPURegSet, Read, -1); + SetError(EXCRegSet, Read, -1); + bool done = false; + + while (!done) { + int flavor = data.GetU32(&offset); + uint32_t count = data.GetU32(&offset); + lldb::offset_t next_thread_state = offset + (count * 4); + switch (flavor) { + case GPRAltRegSet: + case GPRRegSet: + for (uint32_t i = 0; i < count; ++i) { + gpr.r[i] = data.GetU32(&offset); } - } - static size_t - WriteRegister (RegisterContext *reg_ctx, const char *name, const char *alt_name, size_t reg_byte_size, Stream &data) - { - const RegisterInfo *reg_info = reg_ctx->GetRegisterInfoByName(name); - if (reg_info == NULL) - reg_info = reg_ctx->GetRegisterInfoByName(alt_name); - if (reg_info) - { - lldb_private::RegisterValue reg_value; - if (reg_ctx->ReadRegister(reg_info, reg_value)) - { - if (reg_info->byte_size >= reg_byte_size) - data.Write(reg_value.GetBytes(), reg_byte_size); - else - { - data.Write(reg_value.GetBytes(), reg_info->byte_size); - for (size_t i=0, n = reg_byte_size - reg_info->byte_size; i<n; ++ i) - data.PutChar(0); - } - return reg_byte_size; - } + // Note that gpr.cpsr is also copied by the above loop; this loop + // technically extends + // one element past the end of the gpr.r[] array. + + SetError(GPRRegSet, Read, 0); + offset = next_thread_state; + break; + + case FPURegSet: { + uint8_t *fpu_reg_buf = (uint8_t *)&fpu.floats.s[0]; + const int fpu_reg_buf_size = sizeof(fpu.floats); + if (data.ExtractBytes(offset, fpu_reg_buf_size, eByteOrderLittle, + fpu_reg_buf) == fpu_reg_buf_size) { + offset += fpu_reg_buf_size; + fpu.fpscr = data.GetU32(&offset); + SetError(FPURegSet, Read, 0); + } else { + done = true; + } + } + offset = next_thread_state; + break; + + case EXCRegSet: + if (count == 3) { + exc.exception = data.GetU32(&offset); + exc.fsr = data.GetU32(&offset); + exc.far = data.GetU32(&offset); + SetError(EXCRegSet, Read, 0); } - // Just write zeros if all else fails - for (size_t i=0; i<reg_byte_size; ++ i) + done = true; + offset = next_thread_state; + break; + + // Unknown register set flavor, stop trying to parse. + default: + done = true; + } + } + } + + static size_t WriteRegister(RegisterContext *reg_ctx, const char *name, + const char *alt_name, size_t reg_byte_size, + Stream &data) { + const RegisterInfo *reg_info = reg_ctx->GetRegisterInfoByName(name); + if (reg_info == NULL) + reg_info = reg_ctx->GetRegisterInfoByName(alt_name); + if (reg_info) { + lldb_private::RegisterValue reg_value; + if (reg_ctx->ReadRegister(reg_info, reg_value)) { + if (reg_info->byte_size >= reg_byte_size) + data.Write(reg_value.GetBytes(), reg_byte_size); + else { + data.Write(reg_value.GetBytes(), reg_info->byte_size); + for (size_t i = 0, n = reg_byte_size - reg_info->byte_size; i < n; + ++i) data.PutChar(0); + } return reg_byte_size; + } } - - static bool - Create_LC_THREAD (Thread *thread, Stream &data) - { - RegisterContextSP reg_ctx_sp (thread->GetRegisterContext()); - if (reg_ctx_sp) - { - RegisterContext *reg_ctx = reg_ctx_sp.get(); - - data.PutHex32 (GPRRegSet); // Flavor - data.PutHex32 (GPRWordCount); - WriteRegister (reg_ctx, "r0", NULL, 4, data); - WriteRegister (reg_ctx, "r1", NULL, 4, data); - WriteRegister (reg_ctx, "r2", NULL, 4, data); - WriteRegister (reg_ctx, "r3", NULL, 4, data); - WriteRegister (reg_ctx, "r4", NULL, 4, data); - WriteRegister (reg_ctx, "r5", NULL, 4, data); - WriteRegister (reg_ctx, "r6", NULL, 4, data); - WriteRegister (reg_ctx, "r7", NULL, 4, data); - WriteRegister (reg_ctx, "r8", NULL, 4, data); - WriteRegister (reg_ctx, "r9", NULL, 4, data); - WriteRegister (reg_ctx, "r10", NULL, 4, data); - WriteRegister (reg_ctx, "r11", NULL, 4, data); - WriteRegister (reg_ctx, "r12", NULL, 4, data); - WriteRegister (reg_ctx, "sp", NULL, 4, data); - WriteRegister (reg_ctx, "lr", NULL, 4, data); - WriteRegister (reg_ctx, "pc", NULL, 4, data); - WriteRegister (reg_ctx, "cpsr", NULL, 4, data); - - // Write out the EXC registers -// data.PutHex32 (EXCRegSet); -// data.PutHex32 (EXCWordCount); -// WriteRegister (reg_ctx, "exception", NULL, 4, data); -// WriteRegister (reg_ctx, "fsr", NULL, 4, data); -// WriteRegister (reg_ctx, "far", NULL, 4, data); - return true; - } - return false; + // Just write zeros if all else fails + for (size_t i = 0; i < reg_byte_size; ++i) + data.PutChar(0); + return reg_byte_size; + } + + static bool Create_LC_THREAD(Thread *thread, Stream &data) { + RegisterContextSP reg_ctx_sp(thread->GetRegisterContext()); + if (reg_ctx_sp) { + RegisterContext *reg_ctx = reg_ctx_sp.get(); + + data.PutHex32(GPRRegSet); // Flavor + data.PutHex32(GPRWordCount); + WriteRegister(reg_ctx, "r0", NULL, 4, data); + WriteRegister(reg_ctx, "r1", NULL, 4, data); + WriteRegister(reg_ctx, "r2", NULL, 4, data); + WriteRegister(reg_ctx, "r3", NULL, 4, data); + WriteRegister(reg_ctx, "r4", NULL, 4, data); + WriteRegister(reg_ctx, "r5", NULL, 4, data); + WriteRegister(reg_ctx, "r6", NULL, 4, data); + WriteRegister(reg_ctx, "r7", NULL, 4, data); + WriteRegister(reg_ctx, "r8", NULL, 4, data); + WriteRegister(reg_ctx, "r9", NULL, 4, data); + WriteRegister(reg_ctx, "r10", NULL, 4, data); + WriteRegister(reg_ctx, "r11", NULL, 4, data); + WriteRegister(reg_ctx, "r12", NULL, 4, data); + WriteRegister(reg_ctx, "sp", NULL, 4, data); + WriteRegister(reg_ctx, "lr", NULL, 4, data); + WriteRegister(reg_ctx, "pc", NULL, 4, data); + WriteRegister(reg_ctx, "cpsr", NULL, 4, data); + + // Write out the EXC registers + // data.PutHex32 (EXCRegSet); + // data.PutHex32 (EXCWordCount); + // WriteRegister (reg_ctx, "exception", NULL, 4, data); + // WriteRegister (reg_ctx, "fsr", NULL, 4, data); + // WriteRegister (reg_ctx, "far", NULL, 4, data); + return true; } + return false; + } protected: - int - DoReadGPR(lldb::tid_t tid, int flavor, GPR &gpr) override - { - return -1; - } + int DoReadGPR(lldb::tid_t tid, int flavor, GPR &gpr) override { return -1; } - int - DoReadFPU(lldb::tid_t tid, int flavor, FPU &fpu) override - { - return -1; - } + int DoReadFPU(lldb::tid_t tid, int flavor, FPU &fpu) override { return -1; } - int - DoReadEXC(lldb::tid_t tid, int flavor, EXC &exc) override - { - return -1; - } + int DoReadEXC(lldb::tid_t tid, int flavor, EXC &exc) override { return -1; } - int - DoReadDBG(lldb::tid_t tid, int flavor, DBG &dbg) override - { - return -1; - } + int DoReadDBG(lldb::tid_t tid, int flavor, DBG &dbg) override { return -1; } - int - DoWriteGPR(lldb::tid_t tid, int flavor, const GPR &gpr) override - { - return 0; - } + int DoWriteGPR(lldb::tid_t tid, int flavor, const GPR &gpr) override { + return 0; + } - int - DoWriteFPU(lldb::tid_t tid, int flavor, const FPU &fpu) override - { - return 0; - } + int DoWriteFPU(lldb::tid_t tid, int flavor, const FPU &fpu) override { + return 0; + } - int - DoWriteEXC(lldb::tid_t tid, int flavor, const EXC &exc) override - { - return 0; - } + int DoWriteEXC(lldb::tid_t tid, int flavor, const EXC &exc) override { + return 0; + } - int - DoWriteDBG(lldb::tid_t tid, int flavor, const DBG &dbg) override - { - return -1; - } + int DoWriteDBG(lldb::tid_t tid, int flavor, const DBG &dbg) override { + return -1; + } }; -class RegisterContextDarwin_arm64_Mach : public RegisterContextDarwin_arm64 -{ +class RegisterContextDarwin_arm64_Mach : public RegisterContextDarwin_arm64 { public: - RegisterContextDarwin_arm64_Mach (lldb_private::Thread &thread, const DataExtractor &data) : - RegisterContextDarwin_arm64 (thread, 0) - { - SetRegisterDataFrom_LC_THREAD (data); - } - - void - InvalidateAllRegisters() override - { - // Do nothing... registers are always valid... - } - - void - SetRegisterDataFrom_LC_THREAD (const DataExtractor &data) - { - lldb::offset_t offset = 0; - SetError (GPRRegSet, Read, -1); - SetError (FPURegSet, Read, -1); - SetError (EXCRegSet, Read, -1); - bool done = false; - while (!done) - { - int flavor = data.GetU32 (&offset); - uint32_t count = data.GetU32 (&offset); - lldb::offset_t next_thread_state = offset + (count * 4); - switch (flavor) - { - case GPRRegSet: - // x0-x29 + fp + lr + sp + pc (== 33 64-bit registers) plus cpsr (1 32-bit register) - if (count >= (33 * 2) + 1) - { - for (uint32_t i=0; i<33; ++i) - gpr.x[i] = data.GetU64(&offset); - gpr.cpsr = data.GetU32(&offset); - SetError (GPRRegSet, Read, 0); - } - offset = next_thread_state; - break; - case FPURegSet: - { - uint8_t *fpu_reg_buf = (uint8_t*) &fpu.v[0]; - const int fpu_reg_buf_size = sizeof (fpu); - if (fpu_reg_buf_size == count - && data.ExtractBytes (offset, fpu_reg_buf_size, eByteOrderLittle, fpu_reg_buf) == fpu_reg_buf_size) - { - SetError (FPURegSet, Read, 0); - } - else - { - done = true; - } - } - offset = next_thread_state; - break; - case EXCRegSet: - if (count == 4) - { - exc.far = data.GetU64(&offset); - exc.esr = data.GetU32(&offset); - exc.exception = data.GetU32(&offset); - SetError (EXCRegSet, Read, 0); - } - offset = next_thread_state; - break; - default: - done = true; - break; - } - } - } + RegisterContextDarwin_arm64_Mach(lldb_private::Thread &thread, + const DataExtractor &data) + : RegisterContextDarwin_arm64(thread, 0) { + SetRegisterDataFrom_LC_THREAD(data); + } - static size_t - WriteRegister (RegisterContext *reg_ctx, const char *name, const char *alt_name, size_t reg_byte_size, Stream &data) - { - const RegisterInfo *reg_info = reg_ctx->GetRegisterInfoByName(name); - if (reg_info == NULL) - reg_info = reg_ctx->GetRegisterInfoByName(alt_name); - if (reg_info) - { - lldb_private::RegisterValue reg_value; - if (reg_ctx->ReadRegister(reg_info, reg_value)) - { - if (reg_info->byte_size >= reg_byte_size) - data.Write(reg_value.GetBytes(), reg_byte_size); - else - { - data.Write(reg_value.GetBytes(), reg_info->byte_size); - for (size_t i=0, n = reg_byte_size - reg_info->byte_size; i<n; ++ i) - data.PutChar(0); - } - return reg_byte_size; - } + void InvalidateAllRegisters() override { + // Do nothing... registers are always valid... + } + + void SetRegisterDataFrom_LC_THREAD(const DataExtractor &data) { + lldb::offset_t offset = 0; + SetError(GPRRegSet, Read, -1); + SetError(FPURegSet, Read, -1); + SetError(EXCRegSet, Read, -1); + bool done = false; + while (!done) { + int flavor = data.GetU32(&offset); + uint32_t count = data.GetU32(&offset); + lldb::offset_t next_thread_state = offset + (count * 4); + switch (flavor) { + case GPRRegSet: + // x0-x29 + fp + lr + sp + pc (== 33 64-bit registers) plus cpsr (1 + // 32-bit register) + if (count >= (33 * 2) + 1) { + for (uint32_t i = 0; i < 29; ++i) + gpr.x[i] = data.GetU64(&offset); + gpr.fp = data.GetU64(&offset); + gpr.lr = data.GetU64(&offset); + gpr.sp = data.GetU64(&offset); + gpr.pc = data.GetU64(&offset); + gpr.cpsr = data.GetU32(&offset); + SetError(GPRRegSet, Read, 0); + } + offset = next_thread_state; + break; + case FPURegSet: { + uint8_t *fpu_reg_buf = (uint8_t *)&fpu.v[0]; + const int fpu_reg_buf_size = sizeof(fpu); + if (fpu_reg_buf_size == count && + data.ExtractBytes(offset, fpu_reg_buf_size, eByteOrderLittle, + fpu_reg_buf) == fpu_reg_buf_size) { + SetError(FPURegSet, Read, 0); + } else { + done = true; + } + } + offset = next_thread_state; + break; + case EXCRegSet: + if (count == 4) { + exc.far = data.GetU64(&offset); + exc.esr = data.GetU32(&offset); + exc.exception = data.GetU32(&offset); + SetError(EXCRegSet, Read, 0); } - // Just write zeros if all else fails - for (size_t i=0; i<reg_byte_size; ++ i) + offset = next_thread_state; + break; + default: + done = true; + break; + } + } + } + + static size_t WriteRegister(RegisterContext *reg_ctx, const char *name, + const char *alt_name, size_t reg_byte_size, + Stream &data) { + const RegisterInfo *reg_info = reg_ctx->GetRegisterInfoByName(name); + if (reg_info == NULL) + reg_info = reg_ctx->GetRegisterInfoByName(alt_name); + if (reg_info) { + lldb_private::RegisterValue reg_value; + if (reg_ctx->ReadRegister(reg_info, reg_value)) { + if (reg_info->byte_size >= reg_byte_size) + data.Write(reg_value.GetBytes(), reg_byte_size); + else { + data.Write(reg_value.GetBytes(), reg_info->byte_size); + for (size_t i = 0, n = reg_byte_size - reg_info->byte_size; i < n; + ++i) data.PutChar(0); + } return reg_byte_size; + } } - - static bool - Create_LC_THREAD (Thread *thread, Stream &data) - { - RegisterContextSP reg_ctx_sp (thread->GetRegisterContext()); - if (reg_ctx_sp) - { - RegisterContext *reg_ctx = reg_ctx_sp.get(); - - data.PutHex32 (GPRRegSet); // Flavor - data.PutHex32 (GPRWordCount); - WriteRegister (reg_ctx, "x0", NULL, 8, data); - WriteRegister (reg_ctx, "x1", NULL, 8, data); - WriteRegister (reg_ctx, "x2", NULL, 8, data); - WriteRegister (reg_ctx, "x3", NULL, 8, data); - WriteRegister (reg_ctx, "x4", NULL, 8, data); - WriteRegister (reg_ctx, "x5", NULL, 8, data); - WriteRegister (reg_ctx, "x6", NULL, 8, data); - WriteRegister (reg_ctx, "x7", NULL, 8, data); - WriteRegister (reg_ctx, "x8", NULL, 8, data); - WriteRegister (reg_ctx, "x9", NULL, 8, data); - WriteRegister (reg_ctx, "x10", NULL, 8, data); - WriteRegister (reg_ctx, "x11", NULL, 8, data); - WriteRegister (reg_ctx, "x12", NULL, 8, data); - WriteRegister (reg_ctx, "x13", NULL, 8, data); - WriteRegister (reg_ctx, "x14", NULL, 8, data); - WriteRegister (reg_ctx, "x15", NULL, 8, data); - WriteRegister (reg_ctx, "x16", NULL, 8, data); - WriteRegister (reg_ctx, "x17", NULL, 8, data); - WriteRegister (reg_ctx, "x18", NULL, 8, data); - WriteRegister (reg_ctx, "x19", NULL, 8, data); - WriteRegister (reg_ctx, "x20", NULL, 8, data); - WriteRegister (reg_ctx, "x21", NULL, 8, data); - WriteRegister (reg_ctx, "x22", NULL, 8, data); - WriteRegister (reg_ctx, "x23", NULL, 8, data); - WriteRegister (reg_ctx, "x24", NULL, 8, data); - WriteRegister (reg_ctx, "x25", NULL, 8, data); - WriteRegister (reg_ctx, "x26", NULL, 8, data); - WriteRegister (reg_ctx, "x27", NULL, 8, data); - WriteRegister (reg_ctx, "x28", NULL, 8, data); - WriteRegister (reg_ctx, "fp", NULL, 8, data); - WriteRegister (reg_ctx, "lr", NULL, 8, data); - WriteRegister (reg_ctx, "sp", NULL, 8, data); - WriteRegister (reg_ctx, "pc", NULL, 8, data); - WriteRegister (reg_ctx, "cpsr", NULL, 4, data); - - // Write out the EXC registers -// data.PutHex32 (EXCRegSet); -// data.PutHex32 (EXCWordCount); -// WriteRegister (reg_ctx, "far", NULL, 8, data); -// WriteRegister (reg_ctx, "esr", NULL, 4, data); -// WriteRegister (reg_ctx, "exception", NULL, 4, data); - return true; - } - return false; + // Just write zeros if all else fails + for (size_t i = 0; i < reg_byte_size; ++i) + data.PutChar(0); + return reg_byte_size; + } + + static bool Create_LC_THREAD(Thread *thread, Stream &data) { + RegisterContextSP reg_ctx_sp(thread->GetRegisterContext()); + if (reg_ctx_sp) { + RegisterContext *reg_ctx = reg_ctx_sp.get(); + + data.PutHex32(GPRRegSet); // Flavor + data.PutHex32(GPRWordCount); + WriteRegister(reg_ctx, "x0", NULL, 8, data); + WriteRegister(reg_ctx, "x1", NULL, 8, data); + WriteRegister(reg_ctx, "x2", NULL, 8, data); + WriteRegister(reg_ctx, "x3", NULL, 8, data); + WriteRegister(reg_ctx, "x4", NULL, 8, data); + WriteRegister(reg_ctx, "x5", NULL, 8, data); + WriteRegister(reg_ctx, "x6", NULL, 8, data); + WriteRegister(reg_ctx, "x7", NULL, 8, data); + WriteRegister(reg_ctx, "x8", NULL, 8, data); + WriteRegister(reg_ctx, "x9", NULL, 8, data); + WriteRegister(reg_ctx, "x10", NULL, 8, data); + WriteRegister(reg_ctx, "x11", NULL, 8, data); + WriteRegister(reg_ctx, "x12", NULL, 8, data); + WriteRegister(reg_ctx, "x13", NULL, 8, data); + WriteRegister(reg_ctx, "x14", NULL, 8, data); + WriteRegister(reg_ctx, "x15", NULL, 8, data); + WriteRegister(reg_ctx, "x16", NULL, 8, data); + WriteRegister(reg_ctx, "x17", NULL, 8, data); + WriteRegister(reg_ctx, "x18", NULL, 8, data); + WriteRegister(reg_ctx, "x19", NULL, 8, data); + WriteRegister(reg_ctx, "x20", NULL, 8, data); + WriteRegister(reg_ctx, "x21", NULL, 8, data); + WriteRegister(reg_ctx, "x22", NULL, 8, data); + WriteRegister(reg_ctx, "x23", NULL, 8, data); + WriteRegister(reg_ctx, "x24", NULL, 8, data); + WriteRegister(reg_ctx, "x25", NULL, 8, data); + WriteRegister(reg_ctx, "x26", NULL, 8, data); + WriteRegister(reg_ctx, "x27", NULL, 8, data); + WriteRegister(reg_ctx, "x28", NULL, 8, data); + WriteRegister(reg_ctx, "fp", NULL, 8, data); + WriteRegister(reg_ctx, "lr", NULL, 8, data); + WriteRegister(reg_ctx, "sp", NULL, 8, data); + WriteRegister(reg_ctx, "pc", NULL, 8, data); + WriteRegister(reg_ctx, "cpsr", NULL, 4, data); + + // Write out the EXC registers + // data.PutHex32 (EXCRegSet); + // data.PutHex32 (EXCWordCount); + // WriteRegister (reg_ctx, "far", NULL, 8, data); + // WriteRegister (reg_ctx, "esr", NULL, 4, data); + // WriteRegister (reg_ctx, "exception", NULL, 4, data); + return true; } + return false; + } protected: - int - DoReadGPR(lldb::tid_t tid, int flavor, GPR &gpr) override - { - return -1; - } - - int - DoReadFPU(lldb::tid_t tid, int flavor, FPU &fpu) override - { - return -1; - } - - int - DoReadEXC(lldb::tid_t tid, int flavor, EXC &exc) override - { - return -1; - } + int DoReadGPR(lldb::tid_t tid, int flavor, GPR &gpr) override { return -1; } - int - DoReadDBG(lldb::tid_t tid, int flavor, DBG &dbg) override - { - return -1; - } - - int - DoWriteGPR(lldb::tid_t tid, int flavor, const GPR &gpr) override - { - return 0; - } - - int - DoWriteFPU(lldb::tid_t tid, int flavor, const FPU &fpu) override - { - return 0; - } - - int - DoWriteEXC(lldb::tid_t tid, int flavor, const EXC &exc) override - { - return 0; - } - - int - DoWriteDBG(lldb::tid_t tid, int flavor, const DBG &dbg) override - { - return -1; - } -}; + int DoReadFPU(lldb::tid_t tid, int flavor, FPU &fpu) override { return -1; } -static uint32_t -MachHeaderSizeFromMagic(uint32_t magic) -{ - switch (magic) - { - case MH_MAGIC: - case MH_CIGAM: - return sizeof(struct mach_header); - - case MH_MAGIC_64: - case MH_CIGAM_64: - return sizeof(struct mach_header_64); - break; - - default: - break; - } + int DoReadEXC(lldb::tid_t tid, int flavor, EXC &exc) override { return -1; } + + int DoReadDBG(lldb::tid_t tid, int flavor, DBG &dbg) override { return -1; } + + int DoWriteGPR(lldb::tid_t tid, int flavor, const GPR &gpr) override { + return 0; + } + + int DoWriteFPU(lldb::tid_t tid, int flavor, const FPU &fpu) override { + return 0; + } + + int DoWriteEXC(lldb::tid_t tid, int flavor, const EXC &exc) override { return 0; + } + + int DoWriteDBG(lldb::tid_t tid, int flavor, const DBG &dbg) override { + return -1; + } +}; + +static uint32_t MachHeaderSizeFromMagic(uint32_t magic) { + switch (magic) { + case MH_MAGIC: + case MH_CIGAM: + return sizeof(struct mach_header); + + case MH_MAGIC_64: + case MH_CIGAM_64: + return sizeof(struct mach_header_64); + break; + + default: + break; + } + return 0; } #define MACHO_NLIST_ARM_SYMBOL_IS_THUMB 0x0008 -void -ObjectFileMachO::Initialize() -{ - PluginManager::RegisterPlugin (GetPluginNameStatic(), - GetPluginDescriptionStatic(), - CreateInstance, - CreateMemoryInstance, - GetModuleSpecifications, - SaveCore); +void ObjectFileMachO::Initialize() { + PluginManager::RegisterPlugin( + GetPluginNameStatic(), GetPluginDescriptionStatic(), CreateInstance, + CreateMemoryInstance, GetModuleSpecifications, SaveCore); } -void -ObjectFileMachO::Terminate() -{ - PluginManager::UnregisterPlugin (CreateInstance); +void ObjectFileMachO::Terminate() { + PluginManager::UnregisterPlugin(CreateInstance); } -lldb_private::ConstString -ObjectFileMachO::GetPluginNameStatic() -{ - static ConstString g_name("mach-o"); - return g_name; +lldb_private::ConstString ObjectFileMachO::GetPluginNameStatic() { + static ConstString g_name("mach-o"); + return g_name; } -const char * -ObjectFileMachO::GetPluginDescriptionStatic() -{ - return "Mach-o object file reader (32 and 64 bit)"; +const char *ObjectFileMachO::GetPluginDescriptionStatic() { + return "Mach-o object file reader (32 and 64 bit)"; } -ObjectFile * -ObjectFileMachO::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) -{ - if (!data_sp) - { - data_sp = file->MemoryMapFileContentsIfLocal(file_offset, length); - data_offset = 0; - } - - if (ObjectFileMachO::MagicBytesMatch(data_sp, data_offset, length)) - { - // Update the data to contain the entire file if it doesn't already - if (data_sp->GetByteSize() < length) - { - data_sp = file->MemoryMapFileContentsIfLocal(file_offset, length); - data_offset = 0; - } - std::unique_ptr<ObjectFile> objfile_ap(new ObjectFileMachO (module_sp, data_sp, data_offset, file, file_offset, length)); - if (objfile_ap.get() && objfile_ap->ParseHeader()) - return objfile_ap.release(); +ObjectFile *ObjectFileMachO::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) { + if (!data_sp) { + data_sp = file->MemoryMapFileContentsIfLocal(file_offset, length); + data_offset = 0; + } + + if (ObjectFileMachO::MagicBytesMatch(data_sp, data_offset, length)) { + // Update the data to contain the entire file if it doesn't already + if (data_sp->GetByteSize() < length) { + data_sp = file->MemoryMapFileContentsIfLocal(file_offset, length); + data_offset = 0; } - return NULL; + std::unique_ptr<ObjectFile> objfile_ap(new ObjectFileMachO( + module_sp, data_sp, data_offset, file, file_offset, length)); + if (objfile_ap.get() && objfile_ap->ParseHeader()) + return objfile_ap.release(); + } + return NULL; } -ObjectFile * -ObjectFileMachO::CreateMemoryInstance (const lldb::ModuleSP &module_sp, - DataBufferSP& data_sp, - const ProcessSP &process_sp, - lldb::addr_t header_addr) -{ - if (ObjectFileMachO::MagicBytesMatch(data_sp, 0, data_sp->GetByteSize())) - { - std::unique_ptr<ObjectFile> objfile_ap(new ObjectFileMachO (module_sp, data_sp, process_sp, header_addr)); - if (objfile_ap.get() && objfile_ap->ParseHeader()) - return objfile_ap.release(); - } - return NULL; +ObjectFile *ObjectFileMachO::CreateMemoryInstance( + const lldb::ModuleSP &module_sp, DataBufferSP &data_sp, + const ProcessSP &process_sp, lldb::addr_t header_addr) { + if (ObjectFileMachO::MagicBytesMatch(data_sp, 0, data_sp->GetByteSize())) { + std::unique_ptr<ObjectFile> objfile_ap( + new ObjectFileMachO(module_sp, data_sp, process_sp, header_addr)); + if (objfile_ap.get() && objfile_ap->ParseHeader()) + return objfile_ap.release(); + } + return NULL; } -size_t -ObjectFileMachO::GetModuleSpecifications (const lldb_private::FileSpec& file, - lldb::DataBufferSP& data_sp, - lldb::offset_t data_offset, - lldb::offset_t file_offset, - lldb::offset_t length, - lldb_private::ModuleSpecList &specs) -{ - const size_t initial_count = specs.GetSize(); - - if (ObjectFileMachO::MagicBytesMatch(data_sp, 0, data_sp->GetByteSize())) - { - DataExtractor data; +size_t ObjectFileMachO::GetModuleSpecifications( + const lldb_private::FileSpec &file, lldb::DataBufferSP &data_sp, + lldb::offset_t data_offset, lldb::offset_t file_offset, + lldb::offset_t length, lldb_private::ModuleSpecList &specs) { + const size_t initial_count = specs.GetSize(); + + if (ObjectFileMachO::MagicBytesMatch(data_sp, 0, data_sp->GetByteSize())) { + DataExtractor data; + data.SetData(data_sp); + llvm::MachO::mach_header header; + if (ParseHeader(data, &data_offset, header)) { + size_t header_and_load_cmds = + header.sizeofcmds + MachHeaderSizeFromMagic(header.magic); + if (header_and_load_cmds >= data_sp->GetByteSize()) { + data_sp = file.ReadFileContents(file_offset, header_and_load_cmds); data.SetData(data_sp); - llvm::MachO::mach_header header; - if (ParseHeader (data, &data_offset, header)) - { - size_t header_and_load_cmds = header.sizeofcmds + MachHeaderSizeFromMagic(header.magic); - if (header_and_load_cmds >= data_sp->GetByteSize()) - { - data_sp = file.ReadFileContents(file_offset, header_and_load_cmds); - data.SetData(data_sp); - data_offset = MachHeaderSizeFromMagic(header.magic); - } - if (data_sp) - { - ModuleSpec spec; - spec.GetFileSpec() = file; - spec.SetObjectOffset(file_offset); - spec.SetObjectSize(length); - - if (GetArchitecture (header, data, data_offset, spec.GetArchitecture())) - { - if (spec.GetArchitecture().IsValid()) - { - GetUUID (header, data, data_offset, spec.GetUUID()); - specs.Append(spec); - } - } - } + data_offset = MachHeaderSizeFromMagic(header.magic); + } + if (data_sp) { + ModuleSpec spec; + spec.GetFileSpec() = file; + spec.SetObjectOffset(file_offset); + spec.SetObjectSize(length); + + if (GetArchitecture(header, data, data_offset, + spec.GetArchitecture())) { + if (spec.GetArchitecture().IsValid()) { + GetUUID(header, data, data_offset, spec.GetUUID()); + specs.Append(spec); + } } + } } - return specs.GetSize() - initial_count; + } + return specs.GetSize() - initial_count; } -const ConstString & -ObjectFileMachO::GetSegmentNameTEXT() -{ - static ConstString g_segment_name_TEXT ("__TEXT"); - return g_segment_name_TEXT; +const ConstString &ObjectFileMachO::GetSegmentNameTEXT() { + static ConstString g_segment_name_TEXT("__TEXT"); + return g_segment_name_TEXT; } -const ConstString & -ObjectFileMachO::GetSegmentNameDATA() -{ - static ConstString g_segment_name_DATA ("__DATA"); - return g_segment_name_DATA; +const ConstString &ObjectFileMachO::GetSegmentNameDATA() { + static ConstString g_segment_name_DATA("__DATA"); + return g_segment_name_DATA; } -const ConstString & -ObjectFileMachO::GetSegmentNameDATA_DIRTY() -{ - static ConstString g_segment_name ("__DATA_DIRTY"); - return g_segment_name; +const ConstString &ObjectFileMachO::GetSegmentNameDATA_DIRTY() { + static ConstString g_segment_name("__DATA_DIRTY"); + return g_segment_name; } -const ConstString & -ObjectFileMachO::GetSegmentNameDATA_CONST() -{ - static ConstString g_segment_name ("__DATA_CONST"); - return g_segment_name; +const ConstString &ObjectFileMachO::GetSegmentNameDATA_CONST() { + static ConstString g_segment_name("__DATA_CONST"); + return g_segment_name; } -const ConstString & -ObjectFileMachO::GetSegmentNameOBJC() -{ - static ConstString g_segment_name_OBJC ("__OBJC"); - return g_segment_name_OBJC; +const ConstString &ObjectFileMachO::GetSegmentNameOBJC() { + static ConstString g_segment_name_OBJC("__OBJC"); + return g_segment_name_OBJC; } -const ConstString & -ObjectFileMachO::GetSegmentNameLINKEDIT() -{ - static ConstString g_section_name_LINKEDIT ("__LINKEDIT"); - return g_section_name_LINKEDIT; +const ConstString &ObjectFileMachO::GetSegmentNameLINKEDIT() { + static ConstString g_section_name_LINKEDIT("__LINKEDIT"); + return g_section_name_LINKEDIT; } -const ConstString & -ObjectFileMachO::GetSectionNameEHFrame() -{ - static ConstString g_section_name_eh_frame ("__eh_frame"); - return g_section_name_eh_frame; +const ConstString &ObjectFileMachO::GetSectionNameEHFrame() { + static ConstString g_section_name_eh_frame("__eh_frame"); + return g_section_name_eh_frame; } -bool -ObjectFileMachO::MagicBytesMatch (DataBufferSP& data_sp, - lldb::addr_t data_offset, - lldb::addr_t data_length) -{ - DataExtractor data; - data.SetData (data_sp, data_offset, data_length); - lldb::offset_t offset = 0; - uint32_t magic = data.GetU32(&offset); - return MachHeaderSizeFromMagic(magic) != 0; +bool ObjectFileMachO::MagicBytesMatch(DataBufferSP &data_sp, + lldb::addr_t data_offset, + lldb::addr_t data_length) { + DataExtractor data; + data.SetData(data_sp, data_offset, data_length); + lldb::offset_t offset = 0; + uint32_t magic = data.GetU32(&offset); + return MachHeaderSizeFromMagic(magic) != 0; } ObjectFileMachO::ObjectFileMachO(const lldb::ModuleSP &module_sp, - DataBufferSP& data_sp, + DataBufferSP &data_sp, lldb::offset_t data_offset, - const FileSpec* file, + const FileSpec *file, lldb::offset_t file_offset, - lldb::offset_t length) : - ObjectFile(module_sp, file, file_offset, length, data_sp, data_offset), - m_mach_segments(), - m_mach_sections(), - m_entry_point_address(), - m_thread_context_offsets(), - m_thread_context_offsets_valid(false), - m_reexported_dylibs (), - m_allow_assembly_emulation_unwind_plans (true) -{ - ::memset (&m_header, 0, sizeof(m_header)); - ::memset (&m_dysymtab, 0, sizeof(m_dysymtab)); + lldb::offset_t length) + : ObjectFile(module_sp, file, file_offset, length, data_sp, data_offset), + m_mach_segments(), m_mach_sections(), m_entry_point_address(), + m_thread_context_offsets(), m_thread_context_offsets_valid(false), + m_reexported_dylibs(), m_allow_assembly_emulation_unwind_plans(true) { + ::memset(&m_header, 0, sizeof(m_header)); + ::memset(&m_dysymtab, 0, sizeof(m_dysymtab)); } -ObjectFileMachO::ObjectFileMachO (const lldb::ModuleSP &module_sp, - lldb::DataBufferSP& header_data_sp, - const lldb::ProcessSP &process_sp, - lldb::addr_t header_addr) : - ObjectFile(module_sp, process_sp, header_addr, header_data_sp), - m_mach_segments(), - m_mach_sections(), - m_entry_point_address(), - m_thread_context_offsets(), - m_thread_context_offsets_valid(false), - m_reexported_dylibs (), - m_allow_assembly_emulation_unwind_plans (true) -{ - ::memset (&m_header, 0, sizeof(m_header)); - ::memset (&m_dysymtab, 0, sizeof(m_dysymtab)); +ObjectFileMachO::ObjectFileMachO(const lldb::ModuleSP &module_sp, + lldb::DataBufferSP &header_data_sp, + const lldb::ProcessSP &process_sp, + lldb::addr_t header_addr) + : ObjectFile(module_sp, process_sp, header_addr, header_data_sp), + m_mach_segments(), m_mach_sections(), m_entry_point_address(), + m_thread_context_offsets(), m_thread_context_offsets_valid(false), + m_reexported_dylibs(), m_allow_assembly_emulation_unwind_plans(true) { + ::memset(&m_header, 0, sizeof(m_header)); + ::memset(&m_dysymtab, 0, sizeof(m_dysymtab)); } -bool -ObjectFileMachO::ParseHeader (DataExtractor &data, - lldb::offset_t *data_offset_ptr, - llvm::MachO::mach_header &header) -{ - data.SetByteOrder (endian::InlHostByteOrder()); - // Leave magic in the original byte order - header.magic = data.GetU32(data_offset_ptr); - bool can_parse = false; - bool is_64_bit = false; - switch (header.magic) - { - case MH_MAGIC: - data.SetByteOrder (endian::InlHostByteOrder()); - data.SetAddressByteSize(4); - can_parse = true; - break; - - case MH_MAGIC_64: - data.SetByteOrder (endian::InlHostByteOrder()); - data.SetAddressByteSize(8); - can_parse = true; - is_64_bit = true; - break; - - case MH_CIGAM: - data.SetByteOrder(endian::InlHostByteOrder() == eByteOrderBig ? eByteOrderLittle : eByteOrderBig); - data.SetAddressByteSize(4); - can_parse = true; - break; - - case MH_CIGAM_64: - data.SetByteOrder(endian::InlHostByteOrder() == eByteOrderBig ? eByteOrderLittle : eByteOrderBig); - data.SetAddressByteSize(8); - is_64_bit = true; - can_parse = true; - break; - - default: - break; - } - - if (can_parse) - { - data.GetU32(data_offset_ptr, &header.cputype, 6); - if (is_64_bit) - *data_offset_ptr += 4; - return true; - } - else - { - memset(&header, 0, sizeof(header)); - } - return false; +bool ObjectFileMachO::ParseHeader(DataExtractor &data, + lldb::offset_t *data_offset_ptr, + llvm::MachO::mach_header &header) { + data.SetByteOrder(endian::InlHostByteOrder()); + // Leave magic in the original byte order + header.magic = data.GetU32(data_offset_ptr); + bool can_parse = false; + bool is_64_bit = false; + switch (header.magic) { + case MH_MAGIC: + data.SetByteOrder(endian::InlHostByteOrder()); + data.SetAddressByteSize(4); + can_parse = true; + break; + + case MH_MAGIC_64: + data.SetByteOrder(endian::InlHostByteOrder()); + data.SetAddressByteSize(8); + can_parse = true; + is_64_bit = true; + break; + + case MH_CIGAM: + data.SetByteOrder(endian::InlHostByteOrder() == eByteOrderBig + ? eByteOrderLittle + : eByteOrderBig); + data.SetAddressByteSize(4); + can_parse = true; + break; + + case MH_CIGAM_64: + data.SetByteOrder(endian::InlHostByteOrder() == eByteOrderBig + ? eByteOrderLittle + : eByteOrderBig); + data.SetAddressByteSize(8); + is_64_bit = true; + can_parse = true; + break; + + default: + break; + } + + if (can_parse) { + data.GetU32(data_offset_ptr, &header.cputype, 6); + if (is_64_bit) + *data_offset_ptr += 4; + return true; + } else { + memset(&header, 0, sizeof(header)); + } + return false; } -bool -ObjectFileMachO::ParseHeader () -{ - ModuleSP module_sp(GetModule()); - if (module_sp) - { - std::lock_guard<std::recursive_mutex> guard(module_sp->GetMutex()); - bool can_parse = false; - lldb::offset_t offset = 0; - m_data.SetByteOrder (endian::InlHostByteOrder()); - // Leave magic in the original byte order - m_header.magic = m_data.GetU32(&offset); - switch (m_header.magic) - { - case MH_MAGIC: - m_data.SetByteOrder (endian::InlHostByteOrder()); - m_data.SetAddressByteSize(4); - can_parse = true; - break; - - case MH_MAGIC_64: - m_data.SetByteOrder (endian::InlHostByteOrder()); - m_data.SetAddressByteSize(8); - can_parse = true; - break; - - case MH_CIGAM: - m_data.SetByteOrder(endian::InlHostByteOrder() == eByteOrderBig ? eByteOrderLittle : eByteOrderBig); - m_data.SetAddressByteSize(4); - can_parse = true; - break; - - case MH_CIGAM_64: - m_data.SetByteOrder(endian::InlHostByteOrder() == eByteOrderBig ? eByteOrderLittle : eByteOrderBig); - m_data.SetAddressByteSize(8); - can_parse = true; - break; - - default: - break; - } - - if (can_parse) - { - m_data.GetU32(&offset, &m_header.cputype, 6); +bool ObjectFileMachO::ParseHeader() { + ModuleSP module_sp(GetModule()); + if (module_sp) { + std::lock_guard<std::recursive_mutex> guard(module_sp->GetMutex()); + bool can_parse = false; + lldb::offset_t offset = 0; + m_data.SetByteOrder(endian::InlHostByteOrder()); + // Leave magic in the original byte order + m_header.magic = m_data.GetU32(&offset); + switch (m_header.magic) { + case MH_MAGIC: + m_data.SetByteOrder(endian::InlHostByteOrder()); + m_data.SetAddressByteSize(4); + can_parse = true; + break; + + case MH_MAGIC_64: + m_data.SetByteOrder(endian::InlHostByteOrder()); + m_data.SetAddressByteSize(8); + can_parse = true; + break; + + case MH_CIGAM: + m_data.SetByteOrder(endian::InlHostByteOrder() == eByteOrderBig + ? eByteOrderLittle + : eByteOrderBig); + m_data.SetAddressByteSize(4); + can_parse = true; + break; + + case MH_CIGAM_64: + m_data.SetByteOrder(endian::InlHostByteOrder() == eByteOrderBig + ? eByteOrderLittle + : eByteOrderBig); + m_data.SetAddressByteSize(8); + can_parse = true; + break; + + default: + break; + } - - ArchSpec mach_arch; - - if (GetArchitecture (mach_arch)) - { - // Check if the module has a required architecture - const ArchSpec &module_arch = module_sp->GetArchitecture(); - if (module_arch.IsValid() && !module_arch.IsCompatibleMatch(mach_arch)) - return false; - - if (SetModulesArchitecture (mach_arch)) - { - const size_t header_and_lc_size = m_header.sizeofcmds + MachHeaderSizeFromMagic(m_header.magic); - if (m_data.GetByteSize() < header_and_lc_size) - { - DataBufferSP data_sp; - ProcessSP process_sp (m_process_wp.lock()); - if (process_sp) - { - data_sp = ReadMemory (process_sp, m_memory_addr, header_and_lc_size); - } - else - { - // Read in all only the load command data from the file on disk - data_sp = m_file.ReadFileContents(m_file_offset, header_and_lc_size); - if (data_sp->GetByteSize() != header_and_lc_size) - return false; - } - if (data_sp) - m_data.SetData (data_sp); - } - } - return true; + if (can_parse) { + m_data.GetU32(&offset, &m_header.cputype, 6); + + ArchSpec mach_arch; + + if (GetArchitecture(mach_arch)) { + // Check if the module has a required architecture + const ArchSpec &module_arch = module_sp->GetArchitecture(); + if (module_arch.IsValid() && !module_arch.IsCompatibleMatch(mach_arch)) + return false; + + if (SetModulesArchitecture(mach_arch)) { + const size_t header_and_lc_size = + m_header.sizeofcmds + MachHeaderSizeFromMagic(m_header.magic); + if (m_data.GetByteSize() < header_and_lc_size) { + DataBufferSP data_sp; + ProcessSP process_sp(m_process_wp.lock()); + if (process_sp) { + data_sp = + ReadMemory(process_sp, m_memory_addr, header_and_lc_size); + } else { + // Read in all only the load command data from the file on disk + data_sp = + m_file.ReadFileContents(m_file_offset, header_and_lc_size); + if (data_sp->GetByteSize() != header_and_lc_size) + return false; } + if (data_sp) + m_data.SetData(data_sp); + } } - else - { - memset(&m_header, 0, sizeof(struct mach_header)); - } + return true; + } + } else { + memset(&m_header, 0, sizeof(struct mach_header)); } - return false; + } + return false; } -ByteOrder -ObjectFileMachO::GetByteOrder () const -{ - return m_data.GetByteOrder (); +ByteOrder ObjectFileMachO::GetByteOrder() const { + return m_data.GetByteOrder(); } -bool -ObjectFileMachO::IsExecutable() const -{ - return m_header.filetype == MH_EXECUTE; +bool ObjectFileMachO::IsExecutable() const { + return m_header.filetype == MH_EXECUTE; } -uint32_t -ObjectFileMachO::GetAddressByteSize () const -{ - return m_data.GetAddressByteSize (); +uint32_t ObjectFileMachO::GetAddressByteSize() const { + return m_data.GetAddressByteSize(); } -AddressClass -ObjectFileMachO::GetAddressClass (lldb::addr_t file_addr) -{ - Symtab *symtab = GetSymtab(); - if (symtab) - { - Symbol *symbol = symtab->FindSymbolContainingFileAddress(file_addr); - if (symbol) - { - if (symbol->ValueIsAddress()) - { - SectionSP section_sp (symbol->GetAddressRef().GetSection()); - if (section_sp) - { - const lldb::SectionType section_type = section_sp->GetType(); - switch (section_type) - { - case eSectionTypeInvalid: - return eAddressClassUnknown; - - case eSectionTypeCode: - if (m_header.cputype == llvm::MachO::CPU_TYPE_ARM) - { - // For ARM we have a bit in the n_desc field of the symbol - // that tells us ARM/Thumb which is bit 0x0008. - if (symbol->GetFlags() & MACHO_NLIST_ARM_SYMBOL_IS_THUMB) - return eAddressClassCodeAlternateISA; - } - return eAddressClassCode; - - case eSectionTypeContainer: - return eAddressClassUnknown; - - case eSectionTypeData: - case eSectionTypeDataCString: - case eSectionTypeDataCStringPointers: - case eSectionTypeDataSymbolAddress: - case eSectionTypeData4: - case eSectionTypeData8: - case eSectionTypeData16: - case eSectionTypeDataPointers: - case eSectionTypeZeroFill: - case eSectionTypeDataObjCMessageRefs: - case eSectionTypeDataObjCCFStrings: - case eSectionTypeGoSymtab: - return eAddressClassData; - - case eSectionTypeDebug: - case eSectionTypeDWARFDebugAbbrev: - case eSectionTypeDWARFDebugAddr: - case eSectionTypeDWARFDebugAranges: - case eSectionTypeDWARFDebugFrame: - case eSectionTypeDWARFDebugInfo: - case eSectionTypeDWARFDebugLine: - case eSectionTypeDWARFDebugLoc: - case eSectionTypeDWARFDebugMacInfo: - case eSectionTypeDWARFDebugMacro: - case eSectionTypeDWARFDebugPubNames: - case eSectionTypeDWARFDebugPubTypes: - case eSectionTypeDWARFDebugRanges: - case eSectionTypeDWARFDebugStr: - case eSectionTypeDWARFDebugStrOffsets: - case eSectionTypeDWARFAppleNames: - case eSectionTypeDWARFAppleTypes: - case eSectionTypeDWARFAppleNamespaces: - case eSectionTypeDWARFAppleObjC: - return eAddressClassDebug; - - case eSectionTypeEHFrame: - case eSectionTypeARMexidx: - case eSectionTypeARMextab: - case eSectionTypeCompactUnwind: - return eAddressClassRuntime; - - case eSectionTypeAbsoluteAddress: - case eSectionTypeELFSymbolTable: - case eSectionTypeELFDynamicSymbols: - case eSectionTypeELFRelocationEntries: - case eSectionTypeELFDynamicLinkInfo: - case eSectionTypeOther: - return eAddressClassUnknown; - } - } - } - - const SymbolType symbol_type = symbol->GetType(); - switch (symbol_type) - { - case eSymbolTypeAny: return eAddressClassUnknown; - case eSymbolTypeAbsolute: return eAddressClassUnknown; - - case eSymbolTypeCode: - case eSymbolTypeTrampoline: - case eSymbolTypeResolver: - if (m_header.cputype == llvm::MachO::CPU_TYPE_ARM) - { - // For ARM we have a bit in the n_desc field of the symbol - // that tells us ARM/Thumb which is bit 0x0008. - if (symbol->GetFlags() & MACHO_NLIST_ARM_SYMBOL_IS_THUMB) - return eAddressClassCodeAlternateISA; - } - return eAddressClassCode; - - case eSymbolTypeData: return eAddressClassData; - case eSymbolTypeRuntime: return eAddressClassRuntime; - case eSymbolTypeException: return eAddressClassRuntime; - case eSymbolTypeSourceFile: return eAddressClassDebug; - case eSymbolTypeHeaderFile: return eAddressClassDebug; - case eSymbolTypeObjectFile: return eAddressClassDebug; - case eSymbolTypeCommonBlock: return eAddressClassDebug; - case eSymbolTypeBlock: return eAddressClassDebug; - case eSymbolTypeLocal: return eAddressClassData; - case eSymbolTypeParam: return eAddressClassData; - case eSymbolTypeVariable: return eAddressClassData; - case eSymbolTypeVariableType: return eAddressClassDebug; - case eSymbolTypeLineEntry: return eAddressClassDebug; - case eSymbolTypeLineHeader: return eAddressClassDebug; - case eSymbolTypeScopeBegin: return eAddressClassDebug; - case eSymbolTypeScopeEnd: return eAddressClassDebug; - case eSymbolTypeAdditional: return eAddressClassUnknown; - case eSymbolTypeCompiler: return eAddressClassDebug; - case eSymbolTypeInstrumentation:return eAddressClassDebug; - case eSymbolTypeUndefined: return eAddressClassUnknown; - case eSymbolTypeObjCClass: return eAddressClassRuntime; - case eSymbolTypeObjCMetaClass: return eAddressClassRuntime; - case eSymbolTypeObjCIVar: return eAddressClassRuntime; - case eSymbolTypeReExported: return eAddressClassRuntime; +AddressClass ObjectFileMachO::GetAddressClass(lldb::addr_t file_addr) { + Symtab *symtab = GetSymtab(); + if (symtab) { + Symbol *symbol = symtab->FindSymbolContainingFileAddress(file_addr); + if (symbol) { + if (symbol->ValueIsAddress()) { + SectionSP section_sp(symbol->GetAddressRef().GetSection()); + if (section_sp) { + const lldb::SectionType section_type = section_sp->GetType(); + switch (section_type) { + case eSectionTypeInvalid: + return eAddressClassUnknown; + + case eSectionTypeCode: + if (m_header.cputype == llvm::MachO::CPU_TYPE_ARM) { + // For ARM we have a bit in the n_desc field of the symbol + // that tells us ARM/Thumb which is bit 0x0008. + if (symbol->GetFlags() & MACHO_NLIST_ARM_SYMBOL_IS_THUMB) + return eAddressClassCodeAlternateISA; } + return eAddressClassCode; + + case eSectionTypeContainer: + return eAddressClassUnknown; + + case eSectionTypeData: + case eSectionTypeDataCString: + case eSectionTypeDataCStringPointers: + case eSectionTypeDataSymbolAddress: + case eSectionTypeData4: + case eSectionTypeData8: + case eSectionTypeData16: + case eSectionTypeDataPointers: + case eSectionTypeZeroFill: + case eSectionTypeDataObjCMessageRefs: + case eSectionTypeDataObjCCFStrings: + case eSectionTypeGoSymtab: + return eAddressClassData; + + case eSectionTypeDebug: + case eSectionTypeDWARFDebugAbbrev: + case eSectionTypeDWARFDebugAddr: + case eSectionTypeDWARFDebugAranges: + case eSectionTypeDWARFDebugFrame: + case eSectionTypeDWARFDebugInfo: + case eSectionTypeDWARFDebugLine: + case eSectionTypeDWARFDebugLoc: + case eSectionTypeDWARFDebugMacInfo: + case eSectionTypeDWARFDebugMacro: + case eSectionTypeDWARFDebugPubNames: + case eSectionTypeDWARFDebugPubTypes: + case eSectionTypeDWARFDebugRanges: + case eSectionTypeDWARFDebugStr: + case eSectionTypeDWARFDebugStrOffsets: + case eSectionTypeDWARFAppleNames: + case eSectionTypeDWARFAppleTypes: + case eSectionTypeDWARFAppleNamespaces: + case eSectionTypeDWARFAppleObjC: + return eAddressClassDebug; + + case eSectionTypeEHFrame: + case eSectionTypeARMexidx: + case eSectionTypeARMextab: + case eSectionTypeCompactUnwind: + return eAddressClassRuntime; + + case eSectionTypeAbsoluteAddress: + case eSectionTypeELFSymbolTable: + case eSectionTypeELFDynamicSymbols: + case eSectionTypeELFRelocationEntries: + case eSectionTypeELFDynamicLinkInfo: + case eSectionTypeOther: + return eAddressClassUnknown; + } } + } + + const SymbolType symbol_type = symbol->GetType(); + switch (symbol_type) { + case eSymbolTypeAny: + return eAddressClassUnknown; + case eSymbolTypeAbsolute: + return eAddressClassUnknown; + + case eSymbolTypeCode: + case eSymbolTypeTrampoline: + case eSymbolTypeResolver: + if (m_header.cputype == llvm::MachO::CPU_TYPE_ARM) { + // For ARM we have a bit in the n_desc field of the symbol + // that tells us ARM/Thumb which is bit 0x0008. + if (symbol->GetFlags() & MACHO_NLIST_ARM_SYMBOL_IS_THUMB) + return eAddressClassCodeAlternateISA; + } + return eAddressClassCode; + + case eSymbolTypeData: + return eAddressClassData; + case eSymbolTypeRuntime: + return eAddressClassRuntime; + case eSymbolTypeException: + return eAddressClassRuntime; + case eSymbolTypeSourceFile: + return eAddressClassDebug; + case eSymbolTypeHeaderFile: + return eAddressClassDebug; + case eSymbolTypeObjectFile: + return eAddressClassDebug; + case eSymbolTypeCommonBlock: + return eAddressClassDebug; + case eSymbolTypeBlock: + return eAddressClassDebug; + case eSymbolTypeLocal: + return eAddressClassData; + case eSymbolTypeParam: + return eAddressClassData; + case eSymbolTypeVariable: + return eAddressClassData; + case eSymbolTypeVariableType: + return eAddressClassDebug; + case eSymbolTypeLineEntry: + return eAddressClassDebug; + case eSymbolTypeLineHeader: + return eAddressClassDebug; + case eSymbolTypeScopeBegin: + return eAddressClassDebug; + case eSymbolTypeScopeEnd: + return eAddressClassDebug; + case eSymbolTypeAdditional: + return eAddressClassUnknown; + case eSymbolTypeCompiler: + return eAddressClassDebug; + case eSymbolTypeInstrumentation: + return eAddressClassDebug; + case eSymbolTypeUndefined: + return eAddressClassUnknown; + case eSymbolTypeObjCClass: + return eAddressClassRuntime; + case eSymbolTypeObjCMetaClass: + return eAddressClassRuntime; + case eSymbolTypeObjCIVar: + return eAddressClassRuntime; + case eSymbolTypeReExported: + return eAddressClassRuntime; + } + } + } + return eAddressClassUnknown; +} + +Symtab *ObjectFileMachO::GetSymtab() { + ModuleSP module_sp(GetModule()); + if (module_sp) { + std::lock_guard<std::recursive_mutex> guard(module_sp->GetMutex()); + if (m_symtab_ap.get() == NULL) { + m_symtab_ap.reset(new Symtab(this)); + std::lock_guard<std::recursive_mutex> symtab_guard( + m_symtab_ap->GetMutex()); + ParseSymtab(); + m_symtab_ap->Finalize(); } - return eAddressClassUnknown; + } + return m_symtab_ap.get(); } -Symtab * -ObjectFileMachO::GetSymtab() -{ +bool ObjectFileMachO::IsStripped() { + if (m_dysymtab.cmd == 0) { ModuleSP module_sp(GetModule()); - if (module_sp) - { - std::lock_guard<std::recursive_mutex> guard(module_sp->GetMutex()); - if (m_symtab_ap.get() == NULL) - { - m_symtab_ap.reset(new Symtab(this)); - std::lock_guard<std::recursive_mutex> symtab_guard(m_symtab_ap->GetMutex()); - ParseSymtab (); - m_symtab_ap->Finalize (); + if (module_sp) { + lldb::offset_t offset = MachHeaderSizeFromMagic(m_header.magic); + for (uint32_t i = 0; i < m_header.ncmds; ++i) { + const lldb::offset_t load_cmd_offset = offset; + + load_command lc; + if (m_data.GetU32(&offset, &lc.cmd, 2) == NULL) + break; + if (lc.cmd == LC_DYSYMTAB) { + m_dysymtab.cmd = lc.cmd; + m_dysymtab.cmdsize = lc.cmdsize; + if (m_data.GetU32(&offset, &m_dysymtab.ilocalsym, + (sizeof(m_dysymtab) / sizeof(uint32_t)) - 2) == + NULL) { + // Clear m_dysymtab if we were unable to read all items from the + // load command + ::memset(&m_dysymtab, 0, sizeof(m_dysymtab)); + } } + offset = load_cmd_offset + lc.cmdsize; + } } - return m_symtab_ap.get(); + } + if (m_dysymtab.cmd) + return m_dysymtab.nlocalsym <= 1; + return false; } -bool -ObjectFileMachO::IsStripped () -{ - if (m_dysymtab.cmd == 0) - { - ModuleSP module_sp(GetModule()); - if (module_sp) - { - lldb::offset_t offset = MachHeaderSizeFromMagic(m_header.magic); - for (uint32_t i=0; i<m_header.ncmds; ++i) - { - const lldb::offset_t load_cmd_offset = offset; - - load_command lc; - if (m_data.GetU32(&offset, &lc.cmd, 2) == NULL) - break; - if (lc.cmd == LC_DYSYMTAB) - { - m_dysymtab.cmd = lc.cmd; - m_dysymtab.cmdsize = lc.cmdsize; - if (m_data.GetU32 (&offset, &m_dysymtab.ilocalsym, (sizeof(m_dysymtab) / sizeof(uint32_t)) - 2) == NULL) - { - // Clear m_dysymtab if we were unable to read all items from the load command - ::memset (&m_dysymtab, 0, sizeof(m_dysymtab)); - } - } - offset = load_cmd_offset + lc.cmdsize; - } +void ObjectFileMachO::CreateSections(SectionList &unified_section_list) { + if (!m_sections_ap.get()) { + m_sections_ap.reset(new SectionList()); + + const bool is_dsym = (m_header.filetype == MH_DSYM); + lldb::user_id_t segID = 0; + lldb::user_id_t sectID = 0; + lldb::offset_t offset = MachHeaderSizeFromMagic(m_header.magic); + uint32_t i; + const bool is_core = GetType() == eTypeCoreFile; + // bool dump_sections = false; + ModuleSP module_sp(GetModule()); + // First look up any LC_ENCRYPTION_INFO load commands + typedef RangeArray<uint32_t, uint32_t, 8> EncryptedFileRanges; + EncryptedFileRanges encrypted_file_ranges; + encryption_info_command encryption_cmd; + for (i = 0; i < m_header.ncmds; ++i) { + const lldb::offset_t load_cmd_offset = offset; + if (m_data.GetU32(&offset, &encryption_cmd, 2) == NULL) + break; + + // LC_ENCRYPTION_INFO and LC_ENCRYPTION_INFO_64 have the same sizes for + // the 3 fields we care about, so treat them the same. + if (encryption_cmd.cmd == LC_ENCRYPTION_INFO || + encryption_cmd.cmd == LC_ENCRYPTION_INFO_64) { + if (m_data.GetU32(&offset, &encryption_cmd.cryptoff, 3)) { + if (encryption_cmd.cryptid != 0) { + EncryptedFileRanges::Entry entry; + entry.SetRangeBase(encryption_cmd.cryptoff); + entry.SetByteSize(encryption_cmd.cryptsize); + encrypted_file_ranges.Append(entry); + } } + } + offset = load_cmd_offset + encryption_cmd.cmdsize; } - if (m_dysymtab.cmd) - return m_dysymtab.nlocalsym <= 1; - return false; -} -void -ObjectFileMachO::CreateSections (SectionList &unified_section_list) -{ - if (!m_sections_ap.get()) - { - m_sections_ap.reset(new SectionList()); - - const bool is_dsym = (m_header.filetype == MH_DSYM); - lldb::user_id_t segID = 0; - lldb::user_id_t sectID = 0; - lldb::offset_t offset = MachHeaderSizeFromMagic(m_header.magic); - uint32_t i; - const bool is_core = GetType() == eTypeCoreFile; - //bool dump_sections = false; - ModuleSP module_sp (GetModule()); - // First look up any LC_ENCRYPTION_INFO load commands - typedef RangeArray<uint32_t, uint32_t, 8> EncryptedFileRanges; - EncryptedFileRanges encrypted_file_ranges; - encryption_info_command encryption_cmd; - for (i=0; i<m_header.ncmds; ++i) - { - const lldb::offset_t load_cmd_offset = offset; - if (m_data.GetU32(&offset, &encryption_cmd, 2) == NULL) - break; + bool section_file_addresses_changed = false; + + offset = MachHeaderSizeFromMagic(m_header.magic); + + struct segment_command_64 load_cmd; + for (i = 0; i < m_header.ncmds; ++i) { + const lldb::offset_t load_cmd_offset = offset; + if (m_data.GetU32(&offset, &load_cmd, 2) == NULL) + break; + + if (load_cmd.cmd == LC_SEGMENT || load_cmd.cmd == LC_SEGMENT_64) { + if (m_data.GetU8(&offset, (uint8_t *)load_cmd.segname, 16)) { + bool add_section = true; + bool add_to_unified = true; + ConstString const_segname(load_cmd.segname, + std::min<size_t>(strlen(load_cmd.segname), + sizeof(load_cmd.segname))); + + SectionSP unified_section_sp( + unified_section_list.FindSectionByName(const_segname)); + if (is_dsym && unified_section_sp) { + if (const_segname == GetSegmentNameLINKEDIT()) { + // We need to keep the __LINKEDIT segment private to this object + // file only + add_to_unified = false; + } else { + // This is the dSYM file and this section has already been created + // by + // the object file, no need to create it. + add_section = false; + } + } + load_cmd.vmaddr = m_data.GetAddress(&offset); + load_cmd.vmsize = m_data.GetAddress(&offset); + load_cmd.fileoff = m_data.GetAddress(&offset); + load_cmd.filesize = m_data.GetAddress(&offset); + if (m_length != 0 && load_cmd.filesize != 0) { + if (load_cmd.fileoff > m_length) { + // We have a load command that says it extends past the end of the + // file. This is likely + // a corrupt file. We don't have any way to return an error + // condition here (this method + // was likely invoked from something like + // ObjectFile::GetSectionList()) -- all we can do + // is null out the SectionList vector and if a process has been + // set up, dump a message + // to stdout. The most common case here is core file debugging + // with a truncated file. + const char *lc_segment_name = load_cmd.cmd == LC_SEGMENT_64 + ? "LC_SEGMENT_64" + : "LC_SEGMENT"; + module_sp->ReportWarning( + "load command %u %s has a fileoff (0x%" PRIx64 + ") that extends beyond the end of the file (0x%" PRIx64 + "), ignoring this section", + i, lc_segment_name, load_cmd.fileoff, m_length); + + load_cmd.fileoff = 0; + load_cmd.filesize = 0; + } - // LC_ENCRYPTION_INFO and LC_ENCRYPTION_INFO_64 have the same sizes for - // the 3 fields we care about, so treat them the same. - if (encryption_cmd.cmd == LC_ENCRYPTION_INFO || encryption_cmd.cmd == LC_ENCRYPTION_INFO_64) - { - if (m_data.GetU32(&offset, &encryption_cmd.cryptoff, 3)) - { - if (encryption_cmd.cryptid != 0) - { - EncryptedFileRanges::Entry entry; - entry.SetRangeBase(encryption_cmd.cryptoff); - entry.SetByteSize(encryption_cmd.cryptsize); - encrypted_file_ranges.Append(entry); - } + if (load_cmd.fileoff + load_cmd.filesize > m_length) { + // We have a load command that says it extends past the end of the + // file. This is likely + // a corrupt file. We don't have any way to return an error + // condition here (this method + // was likely invoked from something like + // ObjectFile::GetSectionList()) -- all we can do + // is null out the SectionList vector and if a process has been + // set up, dump a message + // to stdout. The most common case here is core file debugging + // with a truncated file. + const char *lc_segment_name = load_cmd.cmd == LC_SEGMENT_64 + ? "LC_SEGMENT_64" + : "LC_SEGMENT"; + GetModule()->ReportWarning( + "load command %u %s has a fileoff + filesize (0x%" PRIx64 + ") that extends beyond the end of the file (0x%" PRIx64 + "), the segment will be truncated to match", + i, lc_segment_name, load_cmd.fileoff + load_cmd.filesize, + m_length); + + // Tuncase the length + load_cmd.filesize = m_length - load_cmd.fileoff; + } + } + if (m_data.GetU32(&offset, &load_cmd.maxprot, 4)) { + uint32_t segment_permissions = 0; + if (load_cmd.initprot & VM_PROT_READ) + segment_permissions |= ePermissionsReadable; + if (load_cmd.initprot & VM_PROT_WRITE) + segment_permissions |= ePermissionsWritable; + if (load_cmd.initprot & VM_PROT_EXECUTE) + segment_permissions |= ePermissionsExecutable; + + const bool segment_is_encrypted = + (load_cmd.flags & SG_PROTECTED_VERSION_1) != 0; + + // Keep a list of mach segments around in case we need to + // get at data that isn't stored in the abstracted Sections. + m_mach_segments.push_back(load_cmd); + + // Use a segment ID of the segment index shifted left by 8 so they + // never conflict with any of the sections. + SectionSP segment_sp; + if (add_section && (const_segname || is_core)) { + segment_sp.reset(new Section( + module_sp, // Module to which this section belongs + this, // Object file to which this sections belongs + ++segID << 8, // Section ID is the 1 based segment index + // shifted right by 8 bits as not to collide + // with any of the 256 section IDs that are + // possible + const_segname, // Name of this section + eSectionTypeContainer, // This section is a container of other + // sections. + load_cmd.vmaddr, // File VM address == addresses as they are + // found in the object file + load_cmd.vmsize, // VM size in bytes of this section + load_cmd.fileoff, // Offset to the data for this section in + // the file + load_cmd.filesize, // Size in bytes of this section as found + // in the file + 0, // Segments have no alignment information + load_cmd.flags)); // Flags for this section + + segment_sp->SetIsEncrypted(segment_is_encrypted); + m_sections_ap->AddSection(segment_sp); + segment_sp->SetPermissions(segment_permissions); + if (add_to_unified) + unified_section_list.AddSection(segment_sp); + } else if (unified_section_sp) { + if (is_dsym && + unified_section_sp->GetFileAddress() != load_cmd.vmaddr) { + // Check to see if the module was read from memory? + if (module_sp->GetObjectFile()->GetHeaderAddress().IsValid()) { + // We have a module that is in memory and needs to have its + // file address adjusted. We need to do this because when we + // load a file from memory, its addresses will be slid + // already, + // yet the addresses in the new symbol file will still be + // unslid. + // Since everything is stored as section offset, this + // shouldn't + // cause any problems. + + // Make sure we've parsed the symbol table from the + // ObjectFile before we go around changing its Sections. + module_sp->GetObjectFile()->GetSymtab(); + // eh_frame would present the same problems but we parse that + // on + // a per-function basis as-needed so it's more difficult to + // remove its use of the Sections. Realistically, the + // environments + // where this code path will be taken will not have eh_frame + // sections. + + unified_section_sp->SetFileAddress(load_cmd.vmaddr); + + // Notify the module that the section addresses have been + // changed once + // we're done so any file-address caches can be updated. + section_file_addresses_changed = true; } + } + m_sections_ap->AddSection(unified_section_sp); } - offset = load_cmd_offset + encryption_cmd.cmdsize; - } - bool section_file_addresses_changed = false; - - offset = MachHeaderSizeFromMagic(m_header.magic); + struct section_64 sect64; + ::memset(§64, 0, sizeof(sect64)); + // Push a section into our mach sections for the section at + // index zero (NO_SECT) if we don't have any mach sections yet... + if (m_mach_sections.empty()) + m_mach_sections.push_back(sect64); + uint32_t segment_sect_idx; + const lldb::user_id_t first_segment_sectID = sectID + 1; + + const uint32_t num_u32s = load_cmd.cmd == LC_SEGMENT ? 7 : 8; + for (segment_sect_idx = 0; segment_sect_idx < load_cmd.nsects; + ++segment_sect_idx) { + if (m_data.GetU8(&offset, (uint8_t *)sect64.sectname, + sizeof(sect64.sectname)) == NULL) + break; + if (m_data.GetU8(&offset, (uint8_t *)sect64.segname, + sizeof(sect64.segname)) == NULL) + break; + sect64.addr = m_data.GetAddress(&offset); + sect64.size = m_data.GetAddress(&offset); - struct segment_command_64 load_cmd; - for (i=0; i<m_header.ncmds; ++i) - { - const lldb::offset_t load_cmd_offset = offset; - if (m_data.GetU32(&offset, &load_cmd, 2) == NULL) + if (m_data.GetU32(&offset, §64.offset, num_u32s) == NULL) break; - if (load_cmd.cmd == LC_SEGMENT || load_cmd.cmd == LC_SEGMENT_64) - { - if (m_data.GetU8(&offset, (uint8_t*)load_cmd.segname, 16)) - { - bool add_section = true; - bool add_to_unified = true; - ConstString const_segname (load_cmd.segname, std::min<size_t>(strlen(load_cmd.segname), sizeof(load_cmd.segname))); - - SectionSP unified_section_sp(unified_section_list.FindSectionByName(const_segname)); - if (is_dsym && unified_section_sp) - { - if (const_segname == GetSegmentNameLINKEDIT()) - { - // We need to keep the __LINKEDIT segment private to this object file only - add_to_unified = false; - } - else - { - // This is the dSYM file and this section has already been created by - // the object file, no need to create it. - add_section = false; - } + // Keep a list of mach sections around in case we need to + // get at data that isn't stored in the abstracted Sections. + m_mach_sections.push_back(sect64); + + if (add_section) { + ConstString section_name( + sect64.sectname, std::min<size_t>(strlen(sect64.sectname), + sizeof(sect64.sectname))); + if (!const_segname) { + // We have a segment with no name so we need to conjure up + // segments that correspond to the section's segname if there + // isn't already such a section. If there is such a section, + // we resize the section so that it spans all sections. + // We also mark these sections as fake so address matches + // don't + // hit if they land in the gaps between the child sections. + const_segname.SetTrimmedCStringWithLength( + sect64.segname, sizeof(sect64.segname)); + segment_sp = + unified_section_list.FindSectionByName(const_segname); + if (segment_sp.get()) { + Section *segment = segment_sp.get(); + // Grow the section size as needed. + const lldb::addr_t sect64_min_addr = sect64.addr; + const lldb::addr_t sect64_max_addr = + sect64_min_addr + sect64.size; + const lldb::addr_t curr_seg_byte_size = + segment->GetByteSize(); + const lldb::addr_t curr_seg_min_addr = + segment->GetFileAddress(); + const lldb::addr_t curr_seg_max_addr = + curr_seg_min_addr + curr_seg_byte_size; + if (sect64_min_addr >= curr_seg_min_addr) { + const lldb::addr_t new_seg_byte_size = + sect64_max_addr - curr_seg_min_addr; + // Only grow the section size if needed + if (new_seg_byte_size > curr_seg_byte_size) + segment->SetByteSize(new_seg_byte_size); + } else { + // We need to change the base address of the segment and + // adjust the child section offsets for all existing + // children. + const lldb::addr_t slide_amount = + sect64_min_addr - curr_seg_min_addr; + segment->Slide(slide_amount, false); + segment->GetChildren().Slide(-slide_amount, false); + segment->SetByteSize(curr_seg_max_addr - sect64_min_addr); } - load_cmd.vmaddr = m_data.GetAddress(&offset); - load_cmd.vmsize = m_data.GetAddress(&offset); - load_cmd.fileoff = m_data.GetAddress(&offset); - load_cmd.filesize = m_data.GetAddress(&offset); - if (m_length != 0 && load_cmd.filesize != 0) - { - if (load_cmd.fileoff > m_length) - { - // We have a load command that says it extends past the end of the file. This is likely - // a corrupt file. We don't have any way to return an error condition here (this method - // was likely invoked from something like ObjectFile::GetSectionList()) -- all we can do - // is null out the SectionList vector and if a process has been set up, dump a message - // to stdout. The most common case here is core file debugging with a truncated file. - const char *lc_segment_name = load_cmd.cmd == LC_SEGMENT_64 ? "LC_SEGMENT_64" : "LC_SEGMENT"; - module_sp->ReportWarning("load command %u %s has a fileoff (0x%" PRIx64 ") that extends beyond the end of the file (0x%" PRIx64 "), ignoring this section", - i, - lc_segment_name, - load_cmd.fileoff, - m_length); - - load_cmd.fileoff = 0; - load_cmd.filesize = 0; - } - - if (load_cmd.fileoff + load_cmd.filesize > m_length) - { - // We have a load command that says it extends past the end of the file. This is likely - // a corrupt file. We don't have any way to return an error condition here (this method - // was likely invoked from something like ObjectFile::GetSectionList()) -- all we can do - // is null out the SectionList vector and if a process has been set up, dump a message - // to stdout. The most common case here is core file debugging with a truncated file. - const char *lc_segment_name = load_cmd.cmd == LC_SEGMENT_64 ? "LC_SEGMENT_64" : "LC_SEGMENT"; - GetModule()->ReportWarning("load command %u %s has a fileoff + filesize (0x%" PRIx64 ") that extends beyond the end of the file (0x%" PRIx64 "), the segment will be truncated to match", - i, - lc_segment_name, - load_cmd.fileoff + load_cmd.filesize, - m_length); - - // Tuncase the length - load_cmd.filesize = m_length - load_cmd.fileoff; - } - } - if (m_data.GetU32(&offset, &load_cmd.maxprot, 4)) - { - const uint32_t segment_permissions = - ((load_cmd.initprot & VM_PROT_READ) ? ePermissionsReadable : 0) | - ((load_cmd.initprot & VM_PROT_WRITE) ? ePermissionsWritable : 0) | - ((load_cmd.initprot & VM_PROT_EXECUTE) ? ePermissionsExecutable : 0); - - const bool segment_is_encrypted = (load_cmd.flags & SG_PROTECTED_VERSION_1) != 0; - - // Keep a list of mach segments around in case we need to - // get at data that isn't stored in the abstracted Sections. - m_mach_segments.push_back (load_cmd); - - // Use a segment ID of the segment index shifted left by 8 so they - // never conflict with any of the sections. - SectionSP segment_sp; - if (add_section && (const_segname || is_core)) - { - segment_sp.reset(new Section (module_sp, // Module to which this section belongs - this, // Object file to which this sections belongs - ++segID << 8, // Section ID is the 1 based segment index shifted right by 8 bits as not to collide with any of the 256 section IDs that are possible - const_segname, // Name of this section - eSectionTypeContainer, // This section is a container of other sections. - load_cmd.vmaddr, // File VM address == addresses as they are found in the object file - load_cmd.vmsize, // VM size in bytes of this section - load_cmd.fileoff, // Offset to the data for this section in the file - load_cmd.filesize, // Size in bytes of this section as found in the file - 0, // Segments have no alignment information - load_cmd.flags)); // Flags for this section - - segment_sp->SetIsEncrypted (segment_is_encrypted); - m_sections_ap->AddSection(segment_sp); - segment_sp->SetPermissions(segment_permissions); - if (add_to_unified) - unified_section_list.AddSection(segment_sp); - } - else if (unified_section_sp) - { - if (is_dsym && unified_section_sp->GetFileAddress() != load_cmd.vmaddr) - { - // Check to see if the module was read from memory? - if (module_sp->GetObjectFile()->GetHeaderAddress().IsValid()) - { - // We have a module that is in memory and needs to have its - // file address adjusted. We need to do this because when we - // load a file from memory, its addresses will be slid already, - // yet the addresses in the new symbol file will still be unslid. - // Since everything is stored as section offset, this shouldn't - // cause any problems. - - // Make sure we've parsed the symbol table from the - // ObjectFile before we go around changing its Sections. - module_sp->GetObjectFile()->GetSymtab(); - // eh_frame would present the same problems but we parse that on - // a per-function basis as-needed so it's more difficult to - // remove its use of the Sections. Realistically, the environments - // where this code path will be taken will not have eh_frame sections. - - unified_section_sp->SetFileAddress(load_cmd.vmaddr); - - // Notify the module that the section addresses have been changed once - // we're done so any file-address caches can be updated. - section_file_addresses_changed = true; - } - } - m_sections_ap->AddSection(unified_section_sp); - } - - struct section_64 sect64; - ::memset (§64, 0, sizeof(sect64)); - // Push a section into our mach sections for the section at - // index zero (NO_SECT) if we don't have any mach sections yet... - if (m_mach_sections.empty()) - m_mach_sections.push_back(sect64); - uint32_t segment_sect_idx; - const lldb::user_id_t first_segment_sectID = sectID + 1; - - - const uint32_t num_u32s = load_cmd.cmd == LC_SEGMENT ? 7 : 8; - for (segment_sect_idx=0; segment_sect_idx<load_cmd.nsects; ++segment_sect_idx) - { - if (m_data.GetU8(&offset, (uint8_t*)sect64.sectname, sizeof(sect64.sectname)) == NULL) - break; - if (m_data.GetU8(&offset, (uint8_t*)sect64.segname, sizeof(sect64.segname)) == NULL) - break; - sect64.addr = m_data.GetAddress(&offset); - sect64.size = m_data.GetAddress(&offset); - - if (m_data.GetU32(&offset, §64.offset, num_u32s) == NULL) - break; - - // Keep a list of mach sections around in case we need to - // get at data that isn't stored in the abstracted Sections. - m_mach_sections.push_back (sect64); - - if (add_section) - { - ConstString section_name (sect64.sectname, std::min<size_t>(strlen(sect64.sectname), sizeof(sect64.sectname))); - if (!const_segname) - { - // We have a segment with no name so we need to conjure up - // segments that correspond to the section's segname if there - // isn't already such a section. If there is such a section, - // we resize the section so that it spans all sections. - // We also mark these sections as fake so address matches don't - // hit if they land in the gaps between the child sections. - const_segname.SetTrimmedCStringWithLength(sect64.segname, sizeof(sect64.segname)); - segment_sp = unified_section_list.FindSectionByName (const_segname); - if (segment_sp.get()) - { - Section *segment = segment_sp.get(); - // Grow the section size as needed. - const lldb::addr_t sect64_min_addr = sect64.addr; - const lldb::addr_t sect64_max_addr = sect64_min_addr + sect64.size; - const lldb::addr_t curr_seg_byte_size = segment->GetByteSize(); - const lldb::addr_t curr_seg_min_addr = segment->GetFileAddress(); - const lldb::addr_t curr_seg_max_addr = curr_seg_min_addr + curr_seg_byte_size; - if (sect64_min_addr >= curr_seg_min_addr) - { - const lldb::addr_t new_seg_byte_size = sect64_max_addr - curr_seg_min_addr; - // Only grow the section size if needed - if (new_seg_byte_size > curr_seg_byte_size) - segment->SetByteSize (new_seg_byte_size); - } - else - { - // We need to change the base address of the segment and - // adjust the child section offsets for all existing children. - const lldb::addr_t slide_amount = sect64_min_addr - curr_seg_min_addr; - segment->Slide(slide_amount, false); - segment->GetChildren().Slide(-slide_amount, false); - segment->SetByteSize (curr_seg_max_addr - sect64_min_addr); - } - - // Grow the section size as needed. - if (sect64.offset) - { - const lldb::addr_t segment_min_file_offset = segment->GetFileOffset(); - const lldb::addr_t segment_max_file_offset = segment_min_file_offset + segment->GetFileSize(); - - const lldb::addr_t section_min_file_offset = sect64.offset; - const lldb::addr_t section_max_file_offset = section_min_file_offset + sect64.size; - const lldb::addr_t new_file_offset = std::min (section_min_file_offset, segment_min_file_offset); - const lldb::addr_t new_file_size = std::max (section_max_file_offset, segment_max_file_offset) - new_file_offset; - segment->SetFileOffset (new_file_offset); - segment->SetFileSize (new_file_size); - } - } - else - { - // Create a fake section for the section's named segment - segment_sp.reset(new Section (segment_sp, // Parent section - module_sp, // Module to which this section belongs - this, // Object file to which this section belongs - ++segID << 8, // Section ID is the 1 based segment index shifted right by 8 bits as not to collide with any of the 256 section IDs that are possible - const_segname, // Name of this section - eSectionTypeContainer, // This section is a container of other sections. - sect64.addr, // File VM address == addresses as they are found in the object file - sect64.size, // VM size in bytes of this section - sect64.offset, // Offset to the data for this section in the file - sect64.offset ? sect64.size : 0, // Size in bytes of this section as found in the file - sect64.align, - load_cmd.flags)); // Flags for this section - segment_sp->SetIsFake(true); - segment_sp->SetPermissions(segment_permissions); - m_sections_ap->AddSection(segment_sp); - if (add_to_unified) - unified_section_list.AddSection(segment_sp); - segment_sp->SetIsEncrypted (segment_is_encrypted); - } - } - assert (segment_sp.get()); - - lldb::SectionType sect_type = eSectionTypeOther; - - if (sect64.flags & (S_ATTR_PURE_INSTRUCTIONS | S_ATTR_SOME_INSTRUCTIONS)) - sect_type = eSectionTypeCode; - else - { - uint32_t mach_sect_type = sect64.flags & SECTION_TYPE; - static ConstString g_sect_name_objc_data ("__objc_data"); - static ConstString g_sect_name_objc_msgrefs ("__objc_msgrefs"); - static ConstString g_sect_name_objc_selrefs ("__objc_selrefs"); - static ConstString g_sect_name_objc_classrefs ("__objc_classrefs"); - static ConstString g_sect_name_objc_superrefs ("__objc_superrefs"); - static ConstString g_sect_name_objc_const ("__objc_const"); - static ConstString g_sect_name_objc_classlist ("__objc_classlist"); - static ConstString g_sect_name_cfstring ("__cfstring"); - - static ConstString g_sect_name_dwarf_debug_abbrev ("__debug_abbrev"); - static ConstString g_sect_name_dwarf_debug_aranges ("__debug_aranges"); - static ConstString g_sect_name_dwarf_debug_frame ("__debug_frame"); - static ConstString g_sect_name_dwarf_debug_info ("__debug_info"); - static ConstString g_sect_name_dwarf_debug_line ("__debug_line"); - static ConstString g_sect_name_dwarf_debug_loc ("__debug_loc"); - static ConstString g_sect_name_dwarf_debug_macinfo ("__debug_macinfo"); - static ConstString g_sect_name_dwarf_debug_pubnames ("__debug_pubnames"); - static ConstString g_sect_name_dwarf_debug_pubtypes ("__debug_pubtypes"); - static ConstString g_sect_name_dwarf_debug_ranges ("__debug_ranges"); - static ConstString g_sect_name_dwarf_debug_str ("__debug_str"); - static ConstString g_sect_name_dwarf_apple_names ("__apple_names"); - static ConstString g_sect_name_dwarf_apple_types ("__apple_types"); - static ConstString g_sect_name_dwarf_apple_namespaces ("__apple_namespac"); - static ConstString g_sect_name_dwarf_apple_objc ("__apple_objc"); - static ConstString g_sect_name_eh_frame ("__eh_frame"); - static ConstString g_sect_name_compact_unwind ("__unwind_info"); - static ConstString g_sect_name_text ("__text"); - static ConstString g_sect_name_data ("__data"); - static ConstString g_sect_name_go_symtab ("__gosymtab"); - - if (section_name == g_sect_name_dwarf_debug_abbrev) - sect_type = eSectionTypeDWARFDebugAbbrev; - else if (section_name == g_sect_name_dwarf_debug_aranges) - sect_type = eSectionTypeDWARFDebugAranges; - else if (section_name == g_sect_name_dwarf_debug_frame) - sect_type = eSectionTypeDWARFDebugFrame; - else if (section_name == g_sect_name_dwarf_debug_info) - sect_type = eSectionTypeDWARFDebugInfo; - else if (section_name == g_sect_name_dwarf_debug_line) - sect_type = eSectionTypeDWARFDebugLine; - else if (section_name == g_sect_name_dwarf_debug_loc) - sect_type = eSectionTypeDWARFDebugLoc; - else if (section_name == g_sect_name_dwarf_debug_macinfo) - sect_type = eSectionTypeDWARFDebugMacInfo; - else if (section_name == g_sect_name_dwarf_debug_pubnames) - sect_type = eSectionTypeDWARFDebugPubNames; - else if (section_name == g_sect_name_dwarf_debug_pubtypes) - sect_type = eSectionTypeDWARFDebugPubTypes; - else if (section_name == g_sect_name_dwarf_debug_ranges) - sect_type = eSectionTypeDWARFDebugRanges; - else if (section_name == g_sect_name_dwarf_debug_str) - sect_type = eSectionTypeDWARFDebugStr; - else if (section_name == g_sect_name_dwarf_apple_names) - sect_type = eSectionTypeDWARFAppleNames; - else if (section_name == g_sect_name_dwarf_apple_types) - sect_type = eSectionTypeDWARFAppleTypes; - else if (section_name == g_sect_name_dwarf_apple_namespaces) - sect_type = eSectionTypeDWARFAppleNamespaces; - else if (section_name == g_sect_name_dwarf_apple_objc) - sect_type = eSectionTypeDWARFAppleObjC; - else if (section_name == g_sect_name_objc_selrefs) - sect_type = eSectionTypeDataCStringPointers; - else if (section_name == g_sect_name_objc_msgrefs) - sect_type = eSectionTypeDataObjCMessageRefs; - else if (section_name == g_sect_name_eh_frame) - sect_type = eSectionTypeEHFrame; - else if (section_name == g_sect_name_compact_unwind) - sect_type = eSectionTypeCompactUnwind; - else if (section_name == g_sect_name_cfstring) - sect_type = eSectionTypeDataObjCCFStrings; - else if (section_name == g_sect_name_go_symtab) - sect_type = eSectionTypeGoSymtab; - else if (section_name == g_sect_name_objc_data || - section_name == g_sect_name_objc_classrefs || - section_name == g_sect_name_objc_superrefs || - section_name == g_sect_name_objc_const || - section_name == g_sect_name_objc_classlist) - { - sect_type = eSectionTypeDataPointers; - } - - if (sect_type == eSectionTypeOther) - { - switch (mach_sect_type) - { - // TODO: categorize sections by other flags for regular sections - case S_REGULAR: - if (section_name == g_sect_name_text) - sect_type = eSectionTypeCode; - else if (section_name == g_sect_name_data) - sect_type = eSectionTypeData; - else - sect_type = eSectionTypeOther; - break; - case S_ZEROFILL: sect_type = eSectionTypeZeroFill; break; - case S_CSTRING_LITERALS: sect_type = eSectionTypeDataCString; break; // section with only literal C strings - case S_4BYTE_LITERALS: sect_type = eSectionTypeData4; break; // section with only 4 byte literals - case S_8BYTE_LITERALS: sect_type = eSectionTypeData8; break; // section with only 8 byte literals - case S_LITERAL_POINTERS: sect_type = eSectionTypeDataPointers; break; // section with only pointers to literals - case S_NON_LAZY_SYMBOL_POINTERS: sect_type = eSectionTypeDataPointers; break; // section with only non-lazy symbol pointers - case S_LAZY_SYMBOL_POINTERS: sect_type = eSectionTypeDataPointers; break; // section with only lazy symbol pointers - case S_SYMBOL_STUBS: sect_type = eSectionTypeCode; break; // section with only symbol stubs, byte size of stub in the reserved2 field - case S_MOD_INIT_FUNC_POINTERS: sect_type = eSectionTypeDataPointers; break; // section with only function pointers for initialization - case S_MOD_TERM_FUNC_POINTERS: sect_type = eSectionTypeDataPointers; break; // section with only function pointers for termination - case S_COALESCED: sect_type = eSectionTypeOther; break; - case S_GB_ZEROFILL: sect_type = eSectionTypeZeroFill; break; - case S_INTERPOSING: sect_type = eSectionTypeCode; break; // section with only pairs of function pointers for interposing - case S_16BYTE_LITERALS: sect_type = eSectionTypeData16; break; // section with only 16 byte literals - case S_DTRACE_DOF: sect_type = eSectionTypeDebug; break; - case S_LAZY_DYLIB_SYMBOL_POINTERS: sect_type = eSectionTypeDataPointers; break; - default: break; - } - } - } - SectionSP section_sp(new Section (segment_sp, - module_sp, - this, - ++sectID, - section_name, - sect_type, - sect64.addr - segment_sp->GetFileAddress(), - sect64.size, - sect64.offset, - sect64.offset == 0 ? 0 : sect64.size, - sect64.align, - sect64.flags)); - // Set the section to be encrypted to match the segment - - bool section_is_encrypted = false; - if (!segment_is_encrypted && load_cmd.filesize != 0) - section_is_encrypted = encrypted_file_ranges.FindEntryThatContains(sect64.offset) != NULL; - - section_sp->SetIsEncrypted (segment_is_encrypted || section_is_encrypted); - section_sp->SetPermissions(segment_permissions); - segment_sp->GetChildren().AddSection(section_sp); - - if (segment_sp->IsFake()) - { - segment_sp.reset(); - const_segname.Clear(); - } - } - } - if (segment_sp && is_dsym) - { - if (first_segment_sectID <= sectID) - { - lldb::user_id_t sect_uid; - for (sect_uid = first_segment_sectID; sect_uid <= sectID; ++sect_uid) - { - SectionSP curr_section_sp(segment_sp->GetChildren().FindSectionByID (sect_uid)); - SectionSP next_section_sp; - if (sect_uid + 1 <= sectID) - next_section_sp = segment_sp->GetChildren().FindSectionByID (sect_uid+1); - - if (curr_section_sp.get()) - { - if (curr_section_sp->GetByteSize() == 0) - { - if (next_section_sp.get() != NULL) - curr_section_sp->SetByteSize ( next_section_sp->GetFileAddress() - curr_section_sp->GetFileAddress() ); - else - curr_section_sp->SetByteSize ( load_cmd.vmsize ); - } - } - } - } - } + // Grow the section size as needed. + if (sect64.offset) { + const lldb::addr_t segment_min_file_offset = + segment->GetFileOffset(); + const lldb::addr_t segment_max_file_offset = + segment_min_file_offset + segment->GetFileSize(); + + const lldb::addr_t section_min_file_offset = + sect64.offset; + const lldb::addr_t section_max_file_offset = + section_min_file_offset + sect64.size; + const lldb::addr_t new_file_offset = std::min( + section_min_file_offset, segment_min_file_offset); + const lldb::addr_t new_file_size = + std::max(section_max_file_offset, + segment_max_file_offset) - + new_file_offset; + segment->SetFileOffset(new_file_offset); + segment->SetFileSize(new_file_size); + } + } else { + // Create a fake section for the section's named segment + segment_sp.reset(new Section( + segment_sp, // Parent section + module_sp, // Module to which this section belongs + this, // Object file to which this section belongs + ++segID << 8, // Section ID is the 1 based segment index + // shifted right by 8 bits as not to + // collide with any of the 256 section IDs + // that are possible + const_segname, // Name of this section + eSectionTypeContainer, // This section is a container of + // other sections. + sect64.addr, // File VM address == addresses as they are + // found in the object file + sect64.size, // VM size in bytes of this section + sect64.offset, // Offset to the data for this section in + // the file + sect64.offset ? sect64.size : 0, // Size in bytes of + // this section as + // found in the file + sect64.align, + load_cmd.flags)); // Flags for this section + segment_sp->SetIsFake(true); + segment_sp->SetPermissions(segment_permissions); + m_sections_ap->AddSection(segment_sp); + if (add_to_unified) + unified_section_list.AddSection(segment_sp); + segment_sp->SetIsEncrypted(segment_is_encrypted); + } + } + assert(segment_sp.get()); + + lldb::SectionType sect_type = eSectionTypeOther; + + if (sect64.flags & + (S_ATTR_PURE_INSTRUCTIONS | S_ATTR_SOME_INSTRUCTIONS)) + sect_type = eSectionTypeCode; + else { + uint32_t mach_sect_type = sect64.flags & SECTION_TYPE; + static ConstString g_sect_name_objc_data("__objc_data"); + static ConstString g_sect_name_objc_msgrefs("__objc_msgrefs"); + static ConstString g_sect_name_objc_selrefs("__objc_selrefs"); + static ConstString g_sect_name_objc_classrefs( + "__objc_classrefs"); + static ConstString g_sect_name_objc_superrefs( + "__objc_superrefs"); + static ConstString g_sect_name_objc_const("__objc_const"); + static ConstString g_sect_name_objc_classlist( + "__objc_classlist"); + static ConstString g_sect_name_cfstring("__cfstring"); + + static ConstString g_sect_name_dwarf_debug_abbrev( + "__debug_abbrev"); + static ConstString g_sect_name_dwarf_debug_aranges( + "__debug_aranges"); + static ConstString g_sect_name_dwarf_debug_frame( + "__debug_frame"); + static ConstString g_sect_name_dwarf_debug_info( + "__debug_info"); + static ConstString g_sect_name_dwarf_debug_line( + "__debug_line"); + static ConstString g_sect_name_dwarf_debug_loc("__debug_loc"); + static ConstString g_sect_name_dwarf_debug_macinfo( + "__debug_macinfo"); + static ConstString g_sect_name_dwarf_debug_pubnames( + "__debug_pubnames"); + static ConstString g_sect_name_dwarf_debug_pubtypes( + "__debug_pubtypes"); + static ConstString g_sect_name_dwarf_debug_ranges( + "__debug_ranges"); + static ConstString g_sect_name_dwarf_debug_str("__debug_str"); + static ConstString g_sect_name_dwarf_apple_names( + "__apple_names"); + static ConstString g_sect_name_dwarf_apple_types( + "__apple_types"); + static ConstString g_sect_name_dwarf_apple_namespaces( + "__apple_namespac"); + static ConstString g_sect_name_dwarf_apple_objc( + "__apple_objc"); + static ConstString g_sect_name_eh_frame("__eh_frame"); + static ConstString g_sect_name_compact_unwind( + "__unwind_info"); + static ConstString g_sect_name_text("__text"); + static ConstString g_sect_name_data("__data"); + static ConstString g_sect_name_go_symtab("__gosymtab"); + + if (section_name == g_sect_name_dwarf_debug_abbrev) + sect_type = eSectionTypeDWARFDebugAbbrev; + else if (section_name == g_sect_name_dwarf_debug_aranges) + sect_type = eSectionTypeDWARFDebugAranges; + else if (section_name == g_sect_name_dwarf_debug_frame) + sect_type = eSectionTypeDWARFDebugFrame; + else if (section_name == g_sect_name_dwarf_debug_info) + sect_type = eSectionTypeDWARFDebugInfo; + else if (section_name == g_sect_name_dwarf_debug_line) + sect_type = eSectionTypeDWARFDebugLine; + else if (section_name == g_sect_name_dwarf_debug_loc) + sect_type = eSectionTypeDWARFDebugLoc; + else if (section_name == g_sect_name_dwarf_debug_macinfo) + sect_type = eSectionTypeDWARFDebugMacInfo; + else if (section_name == g_sect_name_dwarf_debug_pubnames) + sect_type = eSectionTypeDWARFDebugPubNames; + else if (section_name == g_sect_name_dwarf_debug_pubtypes) + sect_type = eSectionTypeDWARFDebugPubTypes; + else if (section_name == g_sect_name_dwarf_debug_ranges) + sect_type = eSectionTypeDWARFDebugRanges; + else if (section_name == g_sect_name_dwarf_debug_str) + sect_type = eSectionTypeDWARFDebugStr; + else if (section_name == g_sect_name_dwarf_apple_names) + sect_type = eSectionTypeDWARFAppleNames; + else if (section_name == g_sect_name_dwarf_apple_types) + sect_type = eSectionTypeDWARFAppleTypes; + else if (section_name == g_sect_name_dwarf_apple_namespaces) + sect_type = eSectionTypeDWARFAppleNamespaces; + else if (section_name == g_sect_name_dwarf_apple_objc) + sect_type = eSectionTypeDWARFAppleObjC; + else if (section_name == g_sect_name_objc_selrefs) + sect_type = eSectionTypeDataCStringPointers; + else if (section_name == g_sect_name_objc_msgrefs) + sect_type = eSectionTypeDataObjCMessageRefs; + else if (section_name == g_sect_name_eh_frame) + sect_type = eSectionTypeEHFrame; + else if (section_name == g_sect_name_compact_unwind) + sect_type = eSectionTypeCompactUnwind; + else if (section_name == g_sect_name_cfstring) + sect_type = eSectionTypeDataObjCCFStrings; + else if (section_name == g_sect_name_go_symtab) + sect_type = eSectionTypeGoSymtab; + else if (section_name == g_sect_name_objc_data || + section_name == g_sect_name_objc_classrefs || + section_name == g_sect_name_objc_superrefs || + section_name == g_sect_name_objc_const || + section_name == g_sect_name_objc_classlist) { + sect_type = eSectionTypeDataPointers; + } + + if (sect_type == eSectionTypeOther) { + switch (mach_sect_type) { + // TODO: categorize sections by other flags for regular + // sections + case S_REGULAR: + if (section_name == g_sect_name_text) + sect_type = eSectionTypeCode; + else if (section_name == g_sect_name_data) + sect_type = eSectionTypeData; + else + sect_type = eSectionTypeOther; + break; + case S_ZEROFILL: + sect_type = eSectionTypeZeroFill; + break; + case S_CSTRING_LITERALS: + sect_type = eSectionTypeDataCString; + break; // section with only literal C strings + case S_4BYTE_LITERALS: + sect_type = eSectionTypeData4; + break; // section with only 4 byte literals + case S_8BYTE_LITERALS: + sect_type = eSectionTypeData8; + break; // section with only 8 byte literals + case S_LITERAL_POINTERS: + sect_type = eSectionTypeDataPointers; + break; // section with only pointers to literals + case S_NON_LAZY_SYMBOL_POINTERS: + sect_type = eSectionTypeDataPointers; + break; // section with only non-lazy symbol pointers + case S_LAZY_SYMBOL_POINTERS: + sect_type = eSectionTypeDataPointers; + break; // section with only lazy symbol pointers + case S_SYMBOL_STUBS: + sect_type = eSectionTypeCode; + break; // section with only symbol stubs, byte size of + // stub in the reserved2 field + case S_MOD_INIT_FUNC_POINTERS: + sect_type = eSectionTypeDataPointers; + break; // section with only function pointers for + // initialization + case S_MOD_TERM_FUNC_POINTERS: + sect_type = eSectionTypeDataPointers; + break; // section with only function pointers for + // termination + case S_COALESCED: + sect_type = eSectionTypeOther; + break; + case S_GB_ZEROFILL: + sect_type = eSectionTypeZeroFill; + break; + case S_INTERPOSING: + sect_type = eSectionTypeCode; + break; // section with only pairs of function pointers for + // interposing + case S_16BYTE_LITERALS: + sect_type = eSectionTypeData16; + break; // section with only 16 byte literals + case S_DTRACE_DOF: + sect_type = eSectionTypeDebug; + break; + case S_LAZY_DYLIB_SYMBOL_POINTERS: + sect_type = eSectionTypeDataPointers; + break; + default: + break; } + } } + + SectionSP section_sp(new Section( + segment_sp, module_sp, this, ++sectID, section_name, + sect_type, sect64.addr - segment_sp->GetFileAddress(), + sect64.size, sect64.offset, + sect64.offset == 0 ? 0 : sect64.size, sect64.align, + sect64.flags)); + // Set the section to be encrypted to match the segment + + bool section_is_encrypted = false; + if (!segment_is_encrypted && load_cmd.filesize != 0) + section_is_encrypted = + encrypted_file_ranges.FindEntryThatContains( + sect64.offset) != NULL; + + section_sp->SetIsEncrypted(segment_is_encrypted || + section_is_encrypted); + section_sp->SetPermissions(segment_permissions); + segment_sp->GetChildren().AddSection(section_sp); + + if (segment_sp->IsFake()) { + segment_sp.reset(); + const_segname.Clear(); + } + } } - else if (load_cmd.cmd == LC_DYSYMTAB) - { - m_dysymtab.cmd = load_cmd.cmd; - m_dysymtab.cmdsize = load_cmd.cmdsize; - m_data.GetU32 (&offset, &m_dysymtab.ilocalsym, (sizeof(m_dysymtab) / sizeof(uint32_t)) - 2); + if (segment_sp && is_dsym) { + if (first_segment_sectID <= sectID) { + lldb::user_id_t sect_uid; + for (sect_uid = first_segment_sectID; sect_uid <= sectID; + ++sect_uid) { + SectionSP curr_section_sp( + segment_sp->GetChildren().FindSectionByID(sect_uid)); + SectionSP next_section_sp; + if (sect_uid + 1 <= sectID) + next_section_sp = + segment_sp->GetChildren().FindSectionByID(sect_uid + 1); + + if (curr_section_sp.get()) { + if (curr_section_sp->GetByteSize() == 0) { + if (next_section_sp.get() != NULL) + curr_section_sp->SetByteSize( + next_section_sp->GetFileAddress() - + curr_section_sp->GetFileAddress()); + else + curr_section_sp->SetByteSize(load_cmd.vmsize); + } + } + } + } } - - offset = load_cmd_offset + load_cmd.cmdsize; + } } + } else if (load_cmd.cmd == LC_DYSYMTAB) { + m_dysymtab.cmd = load_cmd.cmd; + m_dysymtab.cmdsize = load_cmd.cmdsize; + m_data.GetU32(&offset, &m_dysymtab.ilocalsym, + (sizeof(m_dysymtab) / sizeof(uint32_t)) - 2); + } + + offset = load_cmd_offset + load_cmd.cmdsize; + } - - if (section_file_addresses_changed && module_sp.get()) - { - module_sp->SectionFileAddressesChanged(); - } + if (section_file_addresses_changed && module_sp.get()) { + module_sp->SectionFileAddressesChanged(); } + } } -class MachSymtabSectionInfo -{ +class MachSymtabSectionInfo { public: - MachSymtabSectionInfo (SectionList *section_list) : - m_section_list (section_list), - m_section_infos() - { - // Get the number of sections down to a depth of 1 to include - // all segments and their sections, but no other sections that - // may be added for debug map or - m_section_infos.resize(section_list->GetNumSections(1)); - } - - SectionSP - GetSection (uint8_t n_sect, addr_t file_addr) - { - if (n_sect == 0) - return SectionSP(); - if (n_sect < m_section_infos.size()) - { - if (!m_section_infos[n_sect].section_sp) - { - SectionSP section_sp (m_section_list->FindSectionByID (n_sect)); - m_section_infos[n_sect].section_sp = section_sp; - if (section_sp) - { - m_section_infos[n_sect].vm_range.SetBaseAddress (section_sp->GetFileAddress()); - m_section_infos[n_sect].vm_range.SetByteSize (section_sp->GetByteSize()); - } - else - { - Host::SystemLog (Host::eSystemLogError, "error: unable to find section for section %u\n", n_sect); - } - } - if (m_section_infos[n_sect].vm_range.Contains(file_addr)) - { - // Symbol is in section. - return m_section_infos[n_sect].section_sp; - } - else if (m_section_infos[n_sect].vm_range.GetByteSize () == 0 && - m_section_infos[n_sect].vm_range.GetBaseAddress() == file_addr) - { - // Symbol is in section with zero size, but has the same start - // address as the section. This can happen with linker symbols - // (symbols that start with the letter 'l' or 'L'. - return m_section_infos[n_sect].section_sp; - } + MachSymtabSectionInfo(SectionList *section_list) + : m_section_list(section_list), m_section_infos() { + // Get the number of sections down to a depth of 1 to include + // all segments and their sections, but no other sections that + // may be added for debug map or + m_section_infos.resize(section_list->GetNumSections(1)); + } + + SectionSP GetSection(uint8_t n_sect, addr_t file_addr) { + if (n_sect == 0) + return SectionSP(); + if (n_sect < m_section_infos.size()) { + if (!m_section_infos[n_sect].section_sp) { + SectionSP section_sp(m_section_list->FindSectionByID(n_sect)); + m_section_infos[n_sect].section_sp = section_sp; + if (section_sp) { + m_section_infos[n_sect].vm_range.SetBaseAddress( + section_sp->GetFileAddress()); + m_section_infos[n_sect].vm_range.SetByteSize( + section_sp->GetByteSize()); + } else { + Host::SystemLog(Host::eSystemLogError, + "error: unable to find section for section %u\n", + n_sect); } - return m_section_list->FindSectionContainingFileAddress(file_addr); + } + if (m_section_infos[n_sect].vm_range.Contains(file_addr)) { + // Symbol is in section. + return m_section_infos[n_sect].section_sp; + } else if (m_section_infos[n_sect].vm_range.GetByteSize() == 0 && + m_section_infos[n_sect].vm_range.GetBaseAddress() == + file_addr) { + // Symbol is in section with zero size, but has the same start + // address as the section. This can happen with linker symbols + // (symbols that start with the letter 'l' or 'L'. + return m_section_infos[n_sect].section_sp; + } } + return m_section_list->FindSectionContainingFileAddress(file_addr); + } protected: - struct SectionInfo - { - SectionInfo () : - vm_range(), - section_sp () - { - } - - VMRange vm_range; - SectionSP section_sp; - }; - SectionList *m_section_list; - std::vector<SectionInfo> m_section_infos; + struct SectionInfo { + SectionInfo() : vm_range(), section_sp() {} + + VMRange vm_range; + SectionSP section_sp; + }; + SectionList *m_section_list; + std::vector<SectionInfo> m_section_infos; }; -struct TrieEntry -{ - TrieEntry () : - name(), - address(LLDB_INVALID_ADDRESS), - flags (0), - other(0), - import_name() - { - } - - void - Clear () - { - name.Clear(); - address = LLDB_INVALID_ADDRESS; - flags = 0; - other = 0; - import_name.Clear(); - } - - void - Dump () const - { - printf ("0x%16.16llx 0x%16.16llx 0x%16.16llx \"%s\"", - static_cast<unsigned long long>(address), - static_cast<unsigned long long>(flags), - static_cast<unsigned long long>(other), name.GetCString()); - if (import_name) - printf (" -> \"%s\"\n", import_name.GetCString()); - else - printf ("\n"); - } - ConstString name; - uint64_t address; - uint64_t flags; - uint64_t other; - ConstString import_name; +struct TrieEntry { + TrieEntry() + : name(), address(LLDB_INVALID_ADDRESS), flags(0), other(0), + import_name() {} + + void Clear() { + name.Clear(); + address = LLDB_INVALID_ADDRESS; + flags = 0; + other = 0; + import_name.Clear(); + } + + void Dump() const { + printf("0x%16.16llx 0x%16.16llx 0x%16.16llx \"%s\"", + static_cast<unsigned long long>(address), + static_cast<unsigned long long>(flags), + static_cast<unsigned long long>(other), name.GetCString()); + if (import_name) + printf(" -> \"%s\"\n", import_name.GetCString()); + else + printf("\n"); + } + ConstString name; + uint64_t address; + uint64_t flags; + uint64_t other; + ConstString import_name; }; -struct TrieEntryWithOffset -{ - lldb::offset_t nodeOffset; - TrieEntry entry; +struct TrieEntryWithOffset { + lldb::offset_t nodeOffset; + TrieEntry entry; - TrieEntryWithOffset (lldb::offset_t offset) : - nodeOffset (offset), - entry() - { - } + TrieEntryWithOffset(lldb::offset_t offset) : nodeOffset(offset), entry() {} - void - Dump (uint32_t idx) const - { - printf ("[%3u] 0x%16.16llx: ", idx, - static_cast<unsigned long long>(nodeOffset)); - entry.Dump(); - } + void Dump(uint32_t idx) const { + printf("[%3u] 0x%16.16llx: ", idx, + static_cast<unsigned long long>(nodeOffset)); + entry.Dump(); + } - bool - operator<(const TrieEntryWithOffset& other) const - { - return ( nodeOffset < other.nodeOffset ); - } + bool operator<(const TrieEntryWithOffset &other) const { + return (nodeOffset < other.nodeOffset); + } }; -static bool -ParseTrieEntries (DataExtractor &data, - lldb::offset_t offset, - const bool is_arm, - std::vector<llvm::StringRef> &nameSlices, - std::set<lldb::addr_t> &resolver_addresses, - std::vector<TrieEntryWithOffset>& output) -{ - if (!data.ValidOffset(offset)) - return true; - - const uint64_t terminalSize = data.GetULEB128(&offset); - lldb::offset_t children_offset = offset + terminalSize; - if ( terminalSize != 0 ) { - TrieEntryWithOffset e (offset); - e.entry.flags = data.GetULEB128(&offset); - const char *import_name = NULL; - if ( e.entry.flags & EXPORT_SYMBOL_FLAGS_REEXPORT ) { - e.entry.address = 0; - e.entry.other = data.GetULEB128(&offset); // dylib ordinal - import_name = data.GetCStr(&offset); - } - else { - e.entry.address = data.GetULEB128(&offset); - if ( e.entry.flags & EXPORT_SYMBOL_FLAGS_STUB_AND_RESOLVER ) - { - e.entry.other = data.GetULEB128(&offset); - uint64_t resolver_addr = e.entry.other; - if (is_arm) - resolver_addr &= THUMB_ADDRESS_BIT_MASK; - resolver_addresses.insert(resolver_addr); - } - else - e.entry.other = 0; - } - // Only add symbols that are reexport symbols with a valid import name - if (EXPORT_SYMBOL_FLAGS_REEXPORT & e.entry.flags && import_name && import_name[0]) - { - std::string name; - if (!nameSlices.empty()) - { - for (auto name_slice: nameSlices) - name.append(name_slice.data(), name_slice.size()); - } - if (name.size() > 1) - { - // Skip the leading '_' - e.entry.name.SetCStringWithLength(name.c_str() + 1,name.size() - 1); - } - if (import_name) - { - // Skip the leading '_' - e.entry.import_name.SetCString(import_name+1); - } - output.push_back(e); - } - } - - const uint8_t childrenCount = data.GetU8(&children_offset); - for (uint8_t i=0; i < childrenCount; ++i) { - const char *cstr = data.GetCStr(&children_offset); - if (cstr) - nameSlices.push_back(llvm::StringRef(cstr)); - else - return false; // Corrupt data - lldb::offset_t childNodeOffset = data.GetULEB128(&children_offset); - if (childNodeOffset) - { - if (!ParseTrieEntries(data, - childNodeOffset, - is_arm, - nameSlices, - resolver_addresses, - output)) - { - return false; - } - } - nameSlices.pop_back(); - } +static bool ParseTrieEntries(DataExtractor &data, lldb::offset_t offset, + const bool is_arm, + std::vector<llvm::StringRef> &nameSlices, + std::set<lldb::addr_t> &resolver_addresses, + std::vector<TrieEntryWithOffset> &output) { + if (!data.ValidOffset(offset)) return true; + + const uint64_t terminalSize = data.GetULEB128(&offset); + lldb::offset_t children_offset = offset + terminalSize; + if (terminalSize != 0) { + TrieEntryWithOffset e(offset); + e.entry.flags = data.GetULEB128(&offset); + const char *import_name = NULL; + if (e.entry.flags & EXPORT_SYMBOL_FLAGS_REEXPORT) { + e.entry.address = 0; + e.entry.other = data.GetULEB128(&offset); // dylib ordinal + import_name = data.GetCStr(&offset); + } else { + e.entry.address = data.GetULEB128(&offset); + if (e.entry.flags & EXPORT_SYMBOL_FLAGS_STUB_AND_RESOLVER) { + e.entry.other = data.GetULEB128(&offset); + uint64_t resolver_addr = e.entry.other; + if (is_arm) + resolver_addr &= THUMB_ADDRESS_BIT_MASK; + resolver_addresses.insert(resolver_addr); + } else + e.entry.other = 0; + } + // Only add symbols that are reexport symbols with a valid import name + if (EXPORT_SYMBOL_FLAGS_REEXPORT & e.entry.flags && import_name && + import_name[0]) { + std::string name; + if (!nameSlices.empty()) { + for (auto name_slice : nameSlices) + name.append(name_slice.data(), name_slice.size()); + } + if (name.size() > 1) { + // Skip the leading '_' + e.entry.name.SetCStringWithLength(name.c_str() + 1, name.size() - 1); + } + if (import_name) { + // Skip the leading '_' + e.entry.import_name.SetCString(import_name + 1); + } + output.push_back(e); + } + } + + const uint8_t childrenCount = data.GetU8(&children_offset); + for (uint8_t i = 0; i < childrenCount; ++i) { + const char *cstr = data.GetCStr(&children_offset); + if (cstr) + nameSlices.push_back(llvm::StringRef(cstr)); + else + return false; // Corrupt data + lldb::offset_t childNodeOffset = data.GetULEB128(&children_offset); + if (childNodeOffset) { + if (!ParseTrieEntries(data, childNodeOffset, is_arm, nameSlices, + resolver_addresses, output)) { + return false; + } + } + nameSlices.pop_back(); + } + return true; } // Read the UUID out of a dyld_shared_cache file on-disk. -UUID -ObjectFileMachO::GetSharedCacheUUID (FileSpec dyld_shared_cache, const ByteOrder byte_order, const uint32_t addr_byte_size) -{ - UUID dsc_uuid; - DataBufferSP dsc_data_sp = dyld_shared_cache.MemoryMapFileContentsIfLocal(0, sizeof(struct lldb_copy_dyld_cache_header_v1)); - if (dsc_data_sp) - { - DataExtractor dsc_header_data (dsc_data_sp, byte_order, addr_byte_size); - - char version_str[7]; - lldb::offset_t offset = 0; - memcpy (version_str, dsc_header_data.GetData (&offset, 6), 6); - version_str[6] = '\0'; - if (strcmp (version_str, "dyld_v") == 0) - { - offset = offsetof (struct lldb_copy_dyld_cache_header_v1, uuid); - uint8_t uuid_bytes[sizeof (uuid_t)]; - memcpy (uuid_bytes, dsc_header_data.GetData (&offset, sizeof (uuid_t)), sizeof (uuid_t)); - dsc_uuid.SetBytes (uuid_bytes); - } +UUID ObjectFileMachO::GetSharedCacheUUID(FileSpec dyld_shared_cache, + const ByteOrder byte_order, + const uint32_t addr_byte_size) { + UUID dsc_uuid; + DataBufferSP dsc_data_sp = dyld_shared_cache.MemoryMapFileContentsIfLocal( + 0, sizeof(struct lldb_copy_dyld_cache_header_v1)); + if (dsc_data_sp) { + DataExtractor dsc_header_data(dsc_data_sp, byte_order, addr_byte_size); + + char version_str[7]; + lldb::offset_t offset = 0; + memcpy(version_str, dsc_header_data.GetData(&offset, 6), 6); + version_str[6] = '\0'; + if (strcmp(version_str, "dyld_v") == 0) { + offset = offsetof(struct lldb_copy_dyld_cache_header_v1, uuid); + uint8_t uuid_bytes[sizeof(uuid_t)]; + memcpy(uuid_bytes, dsc_header_data.GetData(&offset, sizeof(uuid_t)), + sizeof(uuid_t)); + dsc_uuid.SetBytes(uuid_bytes); } - return dsc_uuid; + } + return dsc_uuid; } -size_t -ObjectFileMachO::ParseSymtab () -{ - Timer scoped_timer(__PRETTY_FUNCTION__, - "ObjectFileMachO::ParseSymtab () module = %s", - m_file.GetFilename().AsCString("")); - ModuleSP module_sp (GetModule()); - if (!module_sp) - return 0; - - struct symtab_command symtab_load_command = { 0, 0, 0, 0, 0, 0 }; - struct linkedit_data_command function_starts_load_command = { 0, 0, 0, 0 }; - struct dyld_info_command dyld_info = { 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 }; - typedef AddressDataArray<lldb::addr_t, bool, 100> FunctionStarts; - FunctionStarts function_starts; - lldb::offset_t offset = MachHeaderSizeFromMagic(m_header.magic); - uint32_t i; - FileSpecList dylib_files; - Log *log(lldb_private::GetLogIfAllCategoriesSet (LIBLLDB_LOG_SYMBOLS)); - static const llvm::StringRef g_objc_v2_prefix_class ("_OBJC_CLASS_$_"); - static const llvm::StringRef g_objc_v2_prefix_metaclass ("_OBJC_METACLASS_$_"); - static const llvm::StringRef g_objc_v2_prefix_ivar ("_OBJC_IVAR_$_"); - - for (i=0; i<m_header.ncmds; ++i) - { - const lldb::offset_t cmd_offset = offset; - // Read in the load command and load command size - struct load_command lc; - if (m_data.GetU32(&offset, &lc, 2) == NULL) - break; - // Watch for the symbol table load command - switch (lc.cmd) - { - case LC_SYMTAB: - symtab_load_command.cmd = lc.cmd; - symtab_load_command.cmdsize = lc.cmdsize; - // Read in the rest of the symtab load command - if (m_data.GetU32(&offset, &symtab_load_command.symoff, 4) == 0) // fill in symoff, nsyms, stroff, strsize fields - return 0; - if (symtab_load_command.symoff == 0) - { - if (log) - module_sp->LogMessage(log, "LC_SYMTAB.symoff == 0"); - return 0; - } - - if (symtab_load_command.stroff == 0) - { - if (log) - module_sp->LogMessage(log, "LC_SYMTAB.stroff == 0"); - return 0; - } - - if (symtab_load_command.nsyms == 0) - { - if (log) - module_sp->LogMessage(log, "LC_SYMTAB.nsyms == 0"); - return 0; - } - - if (symtab_load_command.strsize == 0) - { - if (log) - module_sp->LogMessage(log, "LC_SYMTAB.strsize == 0"); - return 0; - } - break; - - case LC_DYLD_INFO: - case LC_DYLD_INFO_ONLY: - if (m_data.GetU32(&offset, &dyld_info.rebase_off, 10)) - { - dyld_info.cmd = lc.cmd; - dyld_info.cmdsize = lc.cmdsize; - } - else - { - memset (&dyld_info, 0, sizeof(dyld_info)); - } - break; +size_t ObjectFileMachO::ParseSymtab() { + Timer scoped_timer(LLVM_PRETTY_FUNCTION, + "ObjectFileMachO::ParseSymtab () module = %s", + m_file.GetFilename().AsCString("")); + ModuleSP module_sp(GetModule()); + if (!module_sp) + return 0; - case LC_LOAD_DYLIB: - case LC_LOAD_WEAK_DYLIB: - case LC_REEXPORT_DYLIB: - case LC_LOADFVMLIB: - case LC_LOAD_UPWARD_DYLIB: - { - uint32_t name_offset = cmd_offset + m_data.GetU32(&offset); - const char *path = m_data.PeekCStr(name_offset); - if (path) - { - FileSpec file_spec(path, false); - // Strip the path if there is @rpath, @executable, etc so we just use the basename - if (path[0] == '@') - file_spec.GetDirectory().Clear(); - - if (lc.cmd == LC_REEXPORT_DYLIB) - { - m_reexported_dylibs.AppendIfUnique(file_spec); - } + struct symtab_command symtab_load_command = {0, 0, 0, 0, 0, 0}; + struct linkedit_data_command function_starts_load_command = {0, 0, 0, 0}; + struct dyld_info_command dyld_info = {0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0}; + typedef AddressDataArray<lldb::addr_t, bool, 100> FunctionStarts; + FunctionStarts function_starts; + lldb::offset_t offset = MachHeaderSizeFromMagic(m_header.magic); + uint32_t i; + FileSpecList dylib_files; + Log *log(lldb_private::GetLogIfAllCategoriesSet(LIBLLDB_LOG_SYMBOLS)); + static const llvm::StringRef g_objc_v2_prefix_class("_OBJC_CLASS_$_"); + static const llvm::StringRef g_objc_v2_prefix_metaclass("_OBJC_METACLASS_$_"); + static const llvm::StringRef g_objc_v2_prefix_ivar("_OBJC_IVAR_$_"); + + for (i = 0; i < m_header.ncmds; ++i) { + const lldb::offset_t cmd_offset = offset; + // Read in the load command and load command size + struct load_command lc; + if (m_data.GetU32(&offset, &lc, 2) == NULL) + break; + // Watch for the symbol table load command + switch (lc.cmd) { + case LC_SYMTAB: + symtab_load_command.cmd = lc.cmd; + symtab_load_command.cmdsize = lc.cmdsize; + // Read in the rest of the symtab load command + if (m_data.GetU32(&offset, &symtab_load_command.symoff, 4) == + 0) // fill in symoff, nsyms, stroff, strsize fields + return 0; + if (symtab_load_command.symoff == 0) { + if (log) + module_sp->LogMessage(log, "LC_SYMTAB.symoff == 0"); + return 0; + } - dylib_files.Append(file_spec); - } - } - break; + if (symtab_load_command.stroff == 0) { + if (log) + module_sp->LogMessage(log, "LC_SYMTAB.stroff == 0"); + return 0; + } - case LC_FUNCTION_STARTS: - function_starts_load_command.cmd = lc.cmd; - function_starts_load_command.cmdsize = lc.cmdsize; - if (m_data.GetU32(&offset, &function_starts_load_command.dataoff, 2) == NULL) // fill in symoff, nsyms, stroff, strsize fields - memset (&function_starts_load_command, 0, sizeof(function_starts_load_command)); - break; + if (symtab_load_command.nsyms == 0) { + if (log) + module_sp->LogMessage(log, "LC_SYMTAB.nsyms == 0"); + return 0; + } - default: - break; + if (symtab_load_command.strsize == 0) { + if (log) + module_sp->LogMessage(log, "LC_SYMTAB.strsize == 0"); + return 0; + } + break; + + case LC_DYLD_INFO: + case LC_DYLD_INFO_ONLY: + if (m_data.GetU32(&offset, &dyld_info.rebase_off, 10)) { + dyld_info.cmd = lc.cmd; + dyld_info.cmdsize = lc.cmdsize; + } else { + memset(&dyld_info, 0, sizeof(dyld_info)); + } + break; + + case LC_LOAD_DYLIB: + case LC_LOAD_WEAK_DYLIB: + case LC_REEXPORT_DYLIB: + case LC_LOADFVMLIB: + case LC_LOAD_UPWARD_DYLIB: { + uint32_t name_offset = cmd_offset + m_data.GetU32(&offset); + const char *path = m_data.PeekCStr(name_offset); + if (path) { + FileSpec file_spec(path, false); + // Strip the path if there is @rpath, @executable, etc so we just use + // the basename + if (path[0] == '@') + file_spec.GetDirectory().Clear(); + + if (lc.cmd == LC_REEXPORT_DYLIB) { + m_reexported_dylibs.AppendIfUnique(file_spec); } - offset = cmd_offset + lc.cmdsize; - } - - if (symtab_load_command.cmd) - { - Symtab *symtab = m_symtab_ap.get(); - SectionList *section_list = GetSectionList(); - if (section_list == NULL) - return 0; - - const uint32_t addr_byte_size = m_data.GetAddressByteSize(); - const ByteOrder byte_order = m_data.GetByteOrder(); - bool bit_width_32 = addr_byte_size == 4; - const size_t nlist_byte_size = bit_width_32 ? sizeof(struct nlist) : sizeof(struct nlist_64); - - DataExtractor nlist_data (NULL, 0, byte_order, addr_byte_size); - DataExtractor strtab_data (NULL, 0, byte_order, addr_byte_size); - DataExtractor function_starts_data (NULL, 0, byte_order, addr_byte_size); - DataExtractor indirect_symbol_index_data (NULL, 0, byte_order, addr_byte_size); - DataExtractor dyld_trie_data (NULL, 0, byte_order, addr_byte_size); - - const addr_t nlist_data_byte_size = symtab_load_command.nsyms * nlist_byte_size; - const addr_t strtab_data_byte_size = symtab_load_command.strsize; - addr_t strtab_addr = LLDB_INVALID_ADDRESS; - - ProcessSP process_sp (m_process_wp.lock()); - Process *process = process_sp.get(); - - uint32_t memory_module_load_level = eMemoryModuleLoadLevelComplete; - - if (process && m_header.filetype != llvm::MachO::MH_OBJECT) - { - Target &target = process->GetTarget(); - - memory_module_load_level = target.GetMemoryModuleLoadLevel(); - - SectionSP linkedit_section_sp(section_list->FindSectionByName(GetSegmentNameLINKEDIT())); - // Reading mach file from memory in a process or core file... - - if (linkedit_section_sp) - { - addr_t linkedit_load_addr = linkedit_section_sp->GetLoadBaseAddress(&target); - if (linkedit_load_addr == LLDB_INVALID_ADDRESS) - { - // We might be trying to access the symbol table before the __LINKEDIT's load - // address has been set in the target. We can't fail to read the symbol table, - // so calculate the right address manually - linkedit_load_addr = CalculateSectionLoadAddressForMemoryImage(m_memory_addr, GetMachHeaderSection(), linkedit_section_sp.get()); - } - - const addr_t linkedit_file_offset = linkedit_section_sp->GetFileOffset(); - const addr_t symoff_addr = linkedit_load_addr + symtab_load_command.symoff - linkedit_file_offset; - strtab_addr = linkedit_load_addr + symtab_load_command.stroff - linkedit_file_offset; - - bool data_was_read = false; - -#if defined (__APPLE__) && (defined (__arm__) || defined (__arm64__) || defined (__aarch64__)) - if (m_header.flags & 0x80000000u && process->GetAddressByteSize() == sizeof (void*)) - { - // This mach-o memory file is in the dyld shared cache. If this - // program is not remote and this is iOS, then this process will - // share the same shared cache as the process we are debugging and - // we can read the entire __LINKEDIT from the address space in this - // process. This is a needed optimization that is used for local iOS - // debugging only since all shared libraries in the shared cache do - // not have corresponding files that exist in the file system of the - // device. They have been combined into a single file. This means we - // always have to load these files from memory. All of the symbol and - // string tables from all of the __LINKEDIT sections from the shared - // libraries in the shared cache have been merged into a single large - // symbol and string table. Reading all of this symbol and string table - // data across can slow down debug launch times, so we optimize this by - // reading the memory for the __LINKEDIT section from this process. - - UUID lldb_shared_cache(GetLLDBSharedCacheUUID()); - UUID process_shared_cache(GetProcessSharedCacheUUID(process)); - bool use_lldb_cache = true; - if (lldb_shared_cache.IsValid() && process_shared_cache.IsValid() && lldb_shared_cache != process_shared_cache) - { - use_lldb_cache = false; - ModuleSP module_sp (GetModule()); - if (module_sp) - module_sp->ReportWarning ("shared cache in process does not match lldb's own shared cache, startup will be slow."); - } - - PlatformSP platform_sp (target.GetPlatform()); - if (platform_sp && platform_sp->IsHost() && use_lldb_cache) - { - data_was_read = true; - nlist_data.SetData((void *)symoff_addr, nlist_data_byte_size, eByteOrderLittle); - strtab_data.SetData((void *)strtab_addr, strtab_data_byte_size, eByteOrderLittle); - if (function_starts_load_command.cmd) - { - const addr_t func_start_addr = linkedit_load_addr + function_starts_load_command.dataoff - linkedit_file_offset; - function_starts_data.SetData ((void *)func_start_addr, function_starts_load_command.datasize, eByteOrderLittle); - } - } - } -#endif + dylib_files.Append(file_spec); + } + } break; + + case LC_FUNCTION_STARTS: + function_starts_load_command.cmd = lc.cmd; + function_starts_load_command.cmdsize = lc.cmdsize; + if (m_data.GetU32(&offset, &function_starts_load_command.dataoff, 2) == + NULL) // fill in symoff, nsyms, stroff, strsize fields + memset(&function_starts_load_command, 0, + sizeof(function_starts_load_command)); + break; + + default: + break; + } + offset = cmd_offset + lc.cmdsize; + } - if (!data_was_read) - { - if (memory_module_load_level == eMemoryModuleLoadLevelComplete) - { - DataBufferSP nlist_data_sp (ReadMemory (process_sp, symoff_addr, nlist_data_byte_size)); - if (nlist_data_sp) - nlist_data.SetData (nlist_data_sp, 0, nlist_data_sp->GetByteSize()); - // Load strings individually from memory when loading from memory since shared cache - // string tables contain strings for all symbols from all shared cached libraries - //DataBufferSP strtab_data_sp (ReadMemory (process_sp, strtab_addr, strtab_data_byte_size)); - //if (strtab_data_sp) - // strtab_data.SetData (strtab_data_sp, 0, strtab_data_sp->GetByteSize()); - if (m_dysymtab.nindirectsyms != 0) - { - const addr_t indirect_syms_addr = linkedit_load_addr + m_dysymtab.indirectsymoff - linkedit_file_offset; - DataBufferSP indirect_syms_data_sp (ReadMemory (process_sp, indirect_syms_addr, m_dysymtab.nindirectsyms * 4)); - if (indirect_syms_data_sp) - indirect_symbol_index_data.SetData (indirect_syms_data_sp, 0, indirect_syms_data_sp->GetByteSize()); - } - } - - if (memory_module_load_level >= eMemoryModuleLoadLevelPartial) - { - if (function_starts_load_command.cmd) - { - const addr_t func_start_addr = linkedit_load_addr + function_starts_load_command.dataoff - linkedit_file_offset; - DataBufferSP func_start_data_sp (ReadMemory (process_sp, func_start_addr, function_starts_load_command.datasize)); - if (func_start_data_sp) - function_starts_data.SetData (func_start_data_sp, 0, func_start_data_sp->GetByteSize()); - } - } - } - } + if (symtab_load_command.cmd) { + Symtab *symtab = m_symtab_ap.get(); + SectionList *section_list = GetSectionList(); + if (section_list == NULL) + return 0; + + const uint32_t addr_byte_size = m_data.GetAddressByteSize(); + const ByteOrder byte_order = m_data.GetByteOrder(); + bool bit_width_32 = addr_byte_size == 4; + const size_t nlist_byte_size = + bit_width_32 ? sizeof(struct nlist) : sizeof(struct nlist_64); + + DataExtractor nlist_data(NULL, 0, byte_order, addr_byte_size); + DataExtractor strtab_data(NULL, 0, byte_order, addr_byte_size); + DataExtractor function_starts_data(NULL, 0, byte_order, addr_byte_size); + DataExtractor indirect_symbol_index_data(NULL, 0, byte_order, + addr_byte_size); + DataExtractor dyld_trie_data(NULL, 0, byte_order, addr_byte_size); + + const addr_t nlist_data_byte_size = + symtab_load_command.nsyms * nlist_byte_size; + const addr_t strtab_data_byte_size = symtab_load_command.strsize; + addr_t strtab_addr = LLDB_INVALID_ADDRESS; + + ProcessSP process_sp(m_process_wp.lock()); + Process *process = process_sp.get(); + + uint32_t memory_module_load_level = eMemoryModuleLoadLevelComplete; + + if (process && m_header.filetype != llvm::MachO::MH_OBJECT) { + Target &target = process->GetTarget(); + + memory_module_load_level = target.GetMemoryModuleLoadLevel(); + + SectionSP linkedit_section_sp( + section_list->FindSectionByName(GetSegmentNameLINKEDIT())); + // Reading mach file from memory in a process or core file... + + if (linkedit_section_sp) { + addr_t linkedit_load_addr = + linkedit_section_sp->GetLoadBaseAddress(&target); + if (linkedit_load_addr == LLDB_INVALID_ADDRESS) { + // We might be trying to access the symbol table before the + // __LINKEDIT's load + // address has been set in the target. We can't fail to read the + // symbol table, + // so calculate the right address manually + linkedit_load_addr = CalculateSectionLoadAddressForMemoryImage( + m_memory_addr, GetMachHeaderSection(), linkedit_section_sp.get()); } - else - { - nlist_data.SetData (m_data, - symtab_load_command.symoff, - nlist_data_byte_size); - strtab_data.SetData (m_data, - symtab_load_command.stroff, - strtab_data_byte_size); - - if (dyld_info.export_size > 0) - { - dyld_trie_data.SetData (m_data, - dyld_info.export_off, - dyld_info.export_size); - } - if (m_dysymtab.nindirectsyms != 0) - { - indirect_symbol_index_data.SetData (m_data, - m_dysymtab.indirectsymoff, - m_dysymtab.nindirectsyms * 4); - } - if (function_starts_load_command.cmd) - { - function_starts_data.SetData (m_data, - function_starts_load_command.dataoff, - function_starts_load_command.datasize); + const addr_t linkedit_file_offset = + linkedit_section_sp->GetFileOffset(); + const addr_t symoff_addr = linkedit_load_addr + + symtab_load_command.symoff - + linkedit_file_offset; + strtab_addr = linkedit_load_addr + symtab_load_command.stroff - + linkedit_file_offset; + + bool data_was_read = false; + +#if defined(__APPLE__) && \ + (defined(__arm__) || defined(__arm64__) || defined(__aarch64__)) + if (m_header.flags & 0x80000000u && + process->GetAddressByteSize() == sizeof(void *)) { + // This mach-o memory file is in the dyld shared cache. If this + // program is not remote and this is iOS, then this process will + // share the same shared cache as the process we are debugging and + // we can read the entire __LINKEDIT from the address space in this + // process. This is a needed optimization that is used for local iOS + // debugging only since all shared libraries in the shared cache do + // not have corresponding files that exist in the file system of the + // device. They have been combined into a single file. This means we + // always have to load these files from memory. All of the symbol and + // string tables from all of the __LINKEDIT sections from the shared + // libraries in the shared cache have been merged into a single large + // symbol and string table. Reading all of this symbol and string + // table + // data across can slow down debug launch times, so we optimize this + // by + // reading the memory for the __LINKEDIT section from this process. + + UUID lldb_shared_cache(GetLLDBSharedCacheUUID()); + UUID process_shared_cache(GetProcessSharedCacheUUID(process)); + bool use_lldb_cache = true; + if (lldb_shared_cache.IsValid() && process_shared_cache.IsValid() && + lldb_shared_cache != process_shared_cache) { + use_lldb_cache = false; + ModuleSP module_sp(GetModule()); + if (module_sp) + module_sp->ReportWarning("shared cache in process does not match " + "lldb's own shared cache, startup will " + "be slow."); + } + + PlatformSP platform_sp(target.GetPlatform()); + if (platform_sp && platform_sp->IsHost() && use_lldb_cache) { + data_was_read = true; + nlist_data.SetData((void *)symoff_addr, nlist_data_byte_size, + eByteOrderLittle); + strtab_data.SetData((void *)strtab_addr, strtab_data_byte_size, + eByteOrderLittle); + if (function_starts_load_command.cmd) { + const addr_t func_start_addr = + linkedit_load_addr + function_starts_load_command.dataoff - + linkedit_file_offset; + function_starts_data.SetData( + (void *)func_start_addr, + function_starts_load_command.datasize, eByteOrderLittle); } + } } +#endif - if (nlist_data.GetByteSize() == 0 && memory_module_load_level == eMemoryModuleLoadLevelComplete) - { - if (log) - module_sp->LogMessage(log, "failed to read nlist data"); - return 0; - } - - const bool have_strtab_data = strtab_data.GetByteSize() > 0; - if (!have_strtab_data) - { - if (process) - { - if (strtab_addr == LLDB_INVALID_ADDRESS) - { - if (log) - module_sp->LogMessage(log, "failed to locate the strtab in memory"); - return 0; - } + if (!data_was_read) { + // Always load dyld - the dynamic linker - from memory if we didn't + // find a binary anywhere else. + // lldb will not register dylib/framework/bundle loads/unloads if we + // don't have the dyld symbols, + // we force dyld to load from memory despite the user's + // target.memory-module-load-level setting. + if (memory_module_load_level == eMemoryModuleLoadLevelComplete || + m_header.filetype == llvm::MachO::MH_DYLINKER) { + DataBufferSP nlist_data_sp( + ReadMemory(process_sp, symoff_addr, nlist_data_byte_size)); + if (nlist_data_sp) + nlist_data.SetData(nlist_data_sp, 0, + nlist_data_sp->GetByteSize()); + // Load strings individually from memory when loading from memory + // since shared cache + // string tables contain strings for all symbols from all shared + // cached libraries + // DataBufferSP strtab_data_sp (ReadMemory (process_sp, strtab_addr, + // strtab_data_byte_size)); + // if (strtab_data_sp) + // strtab_data.SetData (strtab_data_sp, 0, + // strtab_data_sp->GetByteSize()); + if (m_dysymtab.nindirectsyms != 0) { + const addr_t indirect_syms_addr = linkedit_load_addr + + m_dysymtab.indirectsymoff - + linkedit_file_offset; + DataBufferSP indirect_syms_data_sp( + ReadMemory(process_sp, indirect_syms_addr, + m_dysymtab.nindirectsyms * 4)); + if (indirect_syms_data_sp) + indirect_symbol_index_data.SetData( + indirect_syms_data_sp, 0, + indirect_syms_data_sp->GetByteSize()); } - else - { - if (log) - module_sp->LogMessage(log, "failed to read strtab data"); - return 0; + } else if (memory_module_load_level >= + eMemoryModuleLoadLevelPartial) { + if (function_starts_load_command.cmd) { + const addr_t func_start_addr = + linkedit_load_addr + function_starts_load_command.dataoff - + linkedit_file_offset; + DataBufferSP func_start_data_sp( + ReadMemory(process_sp, func_start_addr, + function_starts_load_command.datasize)); + if (func_start_data_sp) + function_starts_data.SetData(func_start_data_sp, 0, + func_start_data_sp->GetByteSize()); } + } } + } + } else { + nlist_data.SetData(m_data, symtab_load_command.symoff, + nlist_data_byte_size); + strtab_data.SetData(m_data, symtab_load_command.stroff, + strtab_data_byte_size); + + if (dyld_info.export_size > 0) { + dyld_trie_data.SetData(m_data, dyld_info.export_off, + dyld_info.export_size); + } + + if (m_dysymtab.nindirectsyms != 0) { + indirect_symbol_index_data.SetData(m_data, m_dysymtab.indirectsymoff, + m_dysymtab.nindirectsyms * 4); + } + if (function_starts_load_command.cmd) { + function_starts_data.SetData(m_data, + function_starts_load_command.dataoff, + function_starts_load_command.datasize); + } + } - const ConstString &g_segment_name_TEXT = GetSegmentNameTEXT(); - const ConstString &g_segment_name_DATA = GetSegmentNameDATA(); - const ConstString &g_segment_name_DATA_DIRTY = GetSegmentNameDATA_DIRTY(); - const ConstString &g_segment_name_DATA_CONST = GetSegmentNameDATA_CONST(); - const ConstString &g_segment_name_OBJC = GetSegmentNameOBJC(); - const ConstString &g_section_name_eh_frame = GetSectionNameEHFrame(); - SectionSP text_section_sp(section_list->FindSectionByName(g_segment_name_TEXT)); - SectionSP data_section_sp(section_list->FindSectionByName(g_segment_name_DATA)); - SectionSP data_dirty_section_sp(section_list->FindSectionByName(g_segment_name_DATA_DIRTY)); - SectionSP data_const_section_sp(section_list->FindSectionByName(g_segment_name_DATA_CONST)); - SectionSP objc_section_sp(section_list->FindSectionByName(g_segment_name_OBJC)); - SectionSP eh_frame_section_sp; - if (text_section_sp.get()) - eh_frame_section_sp = text_section_sp->GetChildren().FindSectionByName (g_section_name_eh_frame); - else - eh_frame_section_sp = section_list->FindSectionByName (g_section_name_eh_frame); - - const bool is_arm = (m_header.cputype == llvm::MachO::CPU_TYPE_ARM); - - // lldb works best if it knows the start address of all functions in a module. - // Linker symbols or debug info are normally the best source of information for start addr / size but - // they may be stripped in a released binary. - // Two additional sources of information exist in Mach-O binaries: - // LC_FUNCTION_STARTS - a list of ULEB128 encoded offsets of each function's start address in the - // binary, relative to the text section. - // eh_frame - the eh_frame FDEs have the start addr & size of each function - // LC_FUNCTION_STARTS is the fastest source to read in, and is present on all modern binaries. - // Binaries built to run on older releases may need to use eh_frame information. - - if (text_section_sp && function_starts_data.GetByteSize()) - { - FunctionStarts::Entry function_start_entry; - function_start_entry.data = false; - lldb::offset_t function_start_offset = 0; - function_start_entry.addr = text_section_sp->GetFileAddress(); - uint64_t delta; - while ((delta = function_starts_data.GetULEB128(&function_start_offset)) > 0) - { - // Now append the current entry - function_start_entry.addr += delta; - function_starts.Append(function_start_entry); - } - } - else - { - // If m_type is eTypeDebugInfo, then this is a dSYM - it will have the load command claiming an eh_frame - // but it doesn't actually have the eh_frame content. And if we have a dSYM, we don't need to do any - // of this fill-in-the-missing-symbols works anyway - the debug info should give us all the functions in - // the module. - if (text_section_sp.get() && eh_frame_section_sp.get() && m_type != eTypeDebugInfo) - { - DWARFCallFrameInfo eh_frame(*this, eh_frame_section_sp, eRegisterKindEHFrame, true); - DWARFCallFrameInfo::FunctionAddressAndSizeVector functions; - eh_frame.GetFunctionAddressAndSizeVector (functions); - addr_t text_base_addr = text_section_sp->GetFileAddress(); - size_t count = functions.GetSize(); - for (size_t i = 0; i < count; ++i) - { - const DWARFCallFrameInfo::FunctionAddressAndSizeVector::Entry *func = functions.GetEntryAtIndex (i); - if (func) - { - FunctionStarts::Entry function_start_entry; - function_start_entry.addr = func->base - text_base_addr; - function_starts.Append(function_start_entry); - } - } - } - } + if (nlist_data.GetByteSize() == 0 && + memory_module_load_level == eMemoryModuleLoadLevelComplete) { + if (log) + module_sp->LogMessage(log, "failed to read nlist data"); + return 0; + } - const size_t function_starts_count = function_starts.GetSize(); - - // For user process binaries (executables, dylibs, frameworks, bundles), if we don't have - // LC_FUNCTION_STARTS/eh_frame section in this binary, we're going to assume the binary - // has been stripped. Don't allow assembly language instruction emulation because we don't - // know proper function start boundaries. - // - // For all other types of binaries (kernels, stand-alone bare board binaries, kexts), they - // may not have LC_FUNCTION_STARTS / eh_frame sections - we should not make any assumptions - // about them based on that. - if (function_starts_count == 0 && CalculateStrata() == eStrataUser) - { - m_allow_assembly_emulation_unwind_plans = false; - Log *unwind_or_symbol_log (lldb_private::GetLogIfAnyCategoriesSet (LIBLLDB_LOG_SYMBOLS | LIBLLDB_LOG_UNWIND)); - - if (unwind_or_symbol_log) - module_sp->LogMessage(unwind_or_symbol_log, "no LC_FUNCTION_STARTS, will not allow assembly profiled unwinds"); + const bool have_strtab_data = strtab_data.GetByteSize() > 0; + if (!have_strtab_data) { + if (process) { + if (strtab_addr == LLDB_INVALID_ADDRESS) { + if (log) + module_sp->LogMessage(log, "failed to locate the strtab in memory"); + return 0; } + } else { + if (log) + module_sp->LogMessage(log, "failed to read strtab data"); + return 0; + } + } - const user_id_t TEXT_eh_frame_sectID = - eh_frame_section_sp.get() ? eh_frame_section_sp->GetID() - : static_cast<user_id_t>(NO_SECT); - - lldb::offset_t nlist_data_offset = 0; - - uint32_t N_SO_index = UINT32_MAX; - - MachSymtabSectionInfo section_info (section_list); - std::vector<uint32_t> N_FUN_indexes; - std::vector<uint32_t> N_NSYM_indexes; - std::vector<uint32_t> N_INCL_indexes; - std::vector<uint32_t> N_BRAC_indexes; - std::vector<uint32_t> N_COMM_indexes; - typedef std::multimap <uint64_t, uint32_t> ValueToSymbolIndexMap; - typedef std::map <uint32_t, uint32_t> NListIndexToSymbolIndexMap; - typedef std::map <const char *, uint32_t> ConstNameToSymbolIndexMap; - ValueToSymbolIndexMap N_FUN_addr_to_sym_idx; - ValueToSymbolIndexMap N_STSYM_addr_to_sym_idx; - ConstNameToSymbolIndexMap N_GSYM_name_to_sym_idx; - // Any symbols that get merged into another will get an entry - // in this map so we know - NListIndexToSymbolIndexMap m_nlist_idx_to_sym_idx; - uint32_t nlist_idx = 0; - Symbol *symbol_ptr = NULL; - - uint32_t sym_idx = 0; - Symbol *sym = NULL; - size_t num_syms = 0; - std::string memory_symbol_name; - uint32_t unmapped_local_symbols_found = 0; - - std::vector<TrieEntryWithOffset> trie_entries; - std::set<lldb::addr_t> resolver_addresses; - - if (dyld_trie_data.GetByteSize() > 0) - { - std::vector<llvm::StringRef> nameSlices; - ParseTrieEntries (dyld_trie_data, - 0, - is_arm, - nameSlices, - resolver_addresses, - trie_entries); - - ConstString text_segment_name ("__TEXT"); - SectionSP text_segment_sp = GetSectionList()->FindSectionByName(text_segment_name); - if (text_segment_sp) - { - const lldb::addr_t text_segment_file_addr = text_segment_sp->GetFileAddress(); - if (text_segment_file_addr != LLDB_INVALID_ADDRESS) - { - for (auto &e : trie_entries) - e.entry.address += text_segment_file_addr; - } - } + const ConstString &g_segment_name_TEXT = GetSegmentNameTEXT(); + const ConstString &g_segment_name_DATA = GetSegmentNameDATA(); + const ConstString &g_segment_name_DATA_DIRTY = GetSegmentNameDATA_DIRTY(); + const ConstString &g_segment_name_DATA_CONST = GetSegmentNameDATA_CONST(); + const ConstString &g_segment_name_OBJC = GetSegmentNameOBJC(); + const ConstString &g_section_name_eh_frame = GetSectionNameEHFrame(); + SectionSP text_section_sp( + section_list->FindSectionByName(g_segment_name_TEXT)); + SectionSP data_section_sp( + section_list->FindSectionByName(g_segment_name_DATA)); + SectionSP data_dirty_section_sp( + section_list->FindSectionByName(g_segment_name_DATA_DIRTY)); + SectionSP data_const_section_sp( + section_list->FindSectionByName(g_segment_name_DATA_CONST)); + SectionSP objc_section_sp( + section_list->FindSectionByName(g_segment_name_OBJC)); + SectionSP eh_frame_section_sp; + if (text_section_sp.get()) + eh_frame_section_sp = text_section_sp->GetChildren().FindSectionByName( + g_section_name_eh_frame); + else + eh_frame_section_sp = + section_list->FindSectionByName(g_section_name_eh_frame); + + const bool is_arm = (m_header.cputype == llvm::MachO::CPU_TYPE_ARM); + + // lldb works best if it knows the start address of all functions in a + // module. + // Linker symbols or debug info are normally the best source of information + // for start addr / size but + // they may be stripped in a released binary. + // Two additional sources of information exist in Mach-O binaries: + // LC_FUNCTION_STARTS - a list of ULEB128 encoded offsets of each + // function's start address in the + // binary, relative to the text section. + // eh_frame - the eh_frame FDEs have the start addr & size of + // each function + // LC_FUNCTION_STARTS is the fastest source to read in, and is present on + // all modern binaries. + // Binaries built to run on older releases may need to use eh_frame + // information. + + if (text_section_sp && function_starts_data.GetByteSize()) { + FunctionStarts::Entry function_start_entry; + function_start_entry.data = false; + lldb::offset_t function_start_offset = 0; + function_start_entry.addr = text_section_sp->GetFileAddress(); + uint64_t delta; + while ((delta = function_starts_data.GetULEB128(&function_start_offset)) > + 0) { + // Now append the current entry + function_start_entry.addr += delta; + function_starts.Append(function_start_entry); + } + } else { + // If m_type is eTypeDebugInfo, then this is a dSYM - it will have the + // load command claiming an eh_frame + // but it doesn't actually have the eh_frame content. And if we have a + // dSYM, we don't need to do any + // of this fill-in-the-missing-symbols works anyway - the debug info + // should give us all the functions in + // the module. + if (text_section_sp.get() && eh_frame_section_sp.get() && + m_type != eTypeDebugInfo) { + DWARFCallFrameInfo eh_frame(*this, eh_frame_section_sp, + eRegisterKindEHFrame, true); + DWARFCallFrameInfo::FunctionAddressAndSizeVector functions; + eh_frame.GetFunctionAddressAndSizeVector(functions); + addr_t text_base_addr = text_section_sp->GetFileAddress(); + size_t count = functions.GetSize(); + for (size_t i = 0; i < count; ++i) { + const DWARFCallFrameInfo::FunctionAddressAndSizeVector::Entry *func = + functions.GetEntryAtIndex(i); + if (func) { + FunctionStarts::Entry function_start_entry; + function_start_entry.addr = func->base - text_base_addr; + function_starts.Append(function_start_entry); + } } + } + } - typedef std::set<ConstString> IndirectSymbols; - IndirectSymbols indirect_symbol_names; - -#if defined (__APPLE__) && (defined (__arm__) || defined (__arm64__) || defined (__aarch64__)) - - // Some recent builds of the dyld_shared_cache (hereafter: DSC) have been optimized by moving LOCAL - // symbols out of the memory mapped portion of the DSC. The symbol information has all been retained, - // but it isn't available in the normal nlist data. However, there *are* duplicate entries of *some* - // LOCAL symbols in the normal nlist data. To handle this situation correctly, we must first attempt - // to parse any DSC unmapped symbol information. If we find any, we set a flag that tells the normal - // nlist parser to ignore all LOCAL symbols. - - if (m_header.flags & 0x80000000u) - { - // Before we can start mapping the DSC, we need to make certain the target process is actually - // using the cache we can find. - - // Next we need to determine the correct path for the dyld shared cache. - - ArchSpec header_arch; - GetArchitecture(header_arch); - char dsc_path[PATH_MAX]; - char dsc_path_development[PATH_MAX]; - - snprintf(dsc_path, sizeof(dsc_path), "%s%s%s", - "/System/Library/Caches/com.apple.dyld/", /* IPHONE_DYLD_SHARED_CACHE_DIR */ - "dyld_shared_cache_", /* DYLD_SHARED_CACHE_BASE_NAME */ - header_arch.GetArchitectureName()); - - snprintf(dsc_path_development, sizeof(dsc_path), "%s%s%s%s", - "/System/Library/Caches/com.apple.dyld/", /* IPHONE_DYLD_SHARED_CACHE_DIR */ - "dyld_shared_cache_", /* DYLD_SHARED_CACHE_BASE_NAME */ - header_arch.GetArchitectureName(), - ".development"); - - FileSpec dsc_nondevelopment_filespec(dsc_path, false); - FileSpec dsc_development_filespec(dsc_path_development, false); - FileSpec dsc_filespec; - - UUID dsc_uuid; - UUID process_shared_cache_uuid; - - if (process) - { - process_shared_cache_uuid = GetProcessSharedCacheUUID(process); - } - - // First see if we can find an exact match for the inferior process shared cache UUID in - // the development or non-development shared caches on disk. - if (process_shared_cache_uuid.IsValid()) - { - if (dsc_development_filespec.Exists()) - { - UUID dsc_development_uuid = GetSharedCacheUUID (dsc_development_filespec, byte_order, addr_byte_size); - if (dsc_development_uuid.IsValid() && dsc_development_uuid == process_shared_cache_uuid) - { - dsc_filespec = dsc_development_filespec; - dsc_uuid = dsc_development_uuid; - } - } - if (!dsc_uuid.IsValid() && dsc_nondevelopment_filespec.Exists()) - { - UUID dsc_nondevelopment_uuid = GetSharedCacheUUID (dsc_nondevelopment_filespec, byte_order, addr_byte_size); - if (dsc_nondevelopment_uuid.IsValid() && dsc_nondevelopment_uuid == process_shared_cache_uuid) - { - dsc_filespec = dsc_nondevelopment_filespec; - dsc_uuid = dsc_nondevelopment_uuid; - } - } - } - - // Failing a UUID match, prefer the development dyld_shared cache if both are present. - if (!dsc_filespec.Exists()) - { - if (dsc_development_filespec.Exists()) - { - dsc_filespec = dsc_development_filespec; - } - else - { - dsc_filespec = dsc_nondevelopment_filespec; - } - } - - /* The dyld_cache_header has a pointer to the dyld_cache_local_symbols_info structure (localSymbolsOffset). - The dyld_cache_local_symbols_info structure gives us three things: - 1. The start and count of the nlist records in the dyld_shared_cache file - 2. The start and size of the strings for these nlist records - 3. The start and count of dyld_cache_local_symbols_entry entries - - There is one dyld_cache_local_symbols_entry per dylib/framework in the dyld shared cache. - The "dylibOffset" field is the Mach-O header of this dylib/framework in the dyld shared cache. - The dyld_cache_local_symbols_entry also lists the start of this dylib/framework's nlist records - and the count of how many nlist records there are for this dylib/framework. - */ - - // Process the dyld shared cache header to find the unmapped symbols - - DataBufferSP dsc_data_sp = dsc_filespec.MemoryMapFileContentsIfLocal(0, sizeof(struct lldb_copy_dyld_cache_header_v1)); - if (!dsc_uuid.IsValid()) - { - dsc_uuid = GetSharedCacheUUID (dsc_filespec, byte_order, addr_byte_size); - } - if (dsc_data_sp) - { - DataExtractor dsc_header_data (dsc_data_sp, byte_order, addr_byte_size); - - bool uuid_match = true; - if (dsc_uuid.IsValid() && process) - { - if (process_shared_cache_uuid.IsValid() && dsc_uuid != process_shared_cache_uuid) - { - // The on-disk dyld_shared_cache file is not the same as the one in this - // process' memory, don't use it. - uuid_match = false; - ModuleSP module_sp (GetModule()); - if (module_sp) - module_sp->ReportWarning ("process shared cache does not match on-disk dyld_shared_cache file, some symbol names will be missing."); - } - } - - offset = offsetof (struct lldb_copy_dyld_cache_header_v1, mappingOffset); - - uint32_t mappingOffset = dsc_header_data.GetU32(&offset); - - // If the mappingOffset points to a location inside the header, we've - // opened an old dyld shared cache, and should not proceed further. - if (uuid_match && mappingOffset >= sizeof(struct lldb_copy_dyld_cache_header_v1)) - { - - DataBufferSP dsc_mapping_info_data_sp = dsc_filespec.MemoryMapFileContentsIfLocal(mappingOffset, sizeof (struct lldb_copy_dyld_cache_mapping_info)); - DataExtractor dsc_mapping_info_data(dsc_mapping_info_data_sp, byte_order, addr_byte_size); - offset = 0; - - // The File addresses (from the in-memory Mach-O load commands) for the shared libraries - // in the shared library cache need to be adjusted by an offset to match up with the - // dylibOffset identifying field in the dyld_cache_local_symbol_entry's. This offset is - // recorded in mapping_offset_value. - const uint64_t mapping_offset_value = dsc_mapping_info_data.GetU64(&offset); - - offset = offsetof (struct lldb_copy_dyld_cache_header_v1, localSymbolsOffset); - uint64_t localSymbolsOffset = dsc_header_data.GetU64(&offset); - uint64_t localSymbolsSize = dsc_header_data.GetU64(&offset); + const size_t function_starts_count = function_starts.GetSize(); - if (localSymbolsOffset && localSymbolsSize) - { - // Map the local symbols - if (DataBufferSP dsc_local_symbols_data_sp = dsc_filespec.MemoryMapFileContentsIfLocal(localSymbolsOffset, localSymbolsSize)) - { - DataExtractor dsc_local_symbols_data(dsc_local_symbols_data_sp, byte_order, addr_byte_size); - - offset = 0; - - typedef std::map<ConstString, uint16_t> UndefinedNameToDescMap; - typedef std::map<uint32_t, ConstString> SymbolIndexToName; - UndefinedNameToDescMap undefined_name_to_desc; - SymbolIndexToName reexport_shlib_needs_fixup; - - - // Read the local_symbols_infos struct in one shot - struct lldb_copy_dyld_cache_local_symbols_info local_symbols_info; - dsc_local_symbols_data.GetU32(&offset, &local_symbols_info.nlistOffset, 6); - - SectionSP text_section_sp(section_list->FindSectionByName(GetSegmentNameTEXT())); - - uint32_t header_file_offset = (text_section_sp->GetFileAddress() - mapping_offset_value); - - offset = local_symbols_info.entriesOffset; - for (uint32_t entry_index = 0; entry_index < local_symbols_info.entriesCount; entry_index++) - { - struct lldb_copy_dyld_cache_local_symbols_entry local_symbols_entry; - local_symbols_entry.dylibOffset = dsc_local_symbols_data.GetU32(&offset); - local_symbols_entry.nlistStartIndex = dsc_local_symbols_data.GetU32(&offset); - local_symbols_entry.nlistCount = dsc_local_symbols_data.GetU32(&offset); - - if (header_file_offset == local_symbols_entry.dylibOffset) - { - unmapped_local_symbols_found = local_symbols_entry.nlistCount; - - // The normal nlist code cannot correctly size the Symbols array, we need to allocate it here. - sym = symtab->Resize (symtab_load_command.nsyms + m_dysymtab.nindirectsyms + unmapped_local_symbols_found - m_dysymtab.nlocalsym); - num_syms = symtab->GetNumSymbols(); - - nlist_data_offset = local_symbols_info.nlistOffset + (nlist_byte_size * local_symbols_entry.nlistStartIndex); - uint32_t string_table_offset = local_symbols_info.stringsOffset; - - for (uint32_t nlist_index = 0; nlist_index < local_symbols_entry.nlistCount; nlist_index++) - { - ///////////////////////////// - { - struct nlist_64 nlist; - if (!dsc_local_symbols_data.ValidOffsetForDataOfSize(nlist_data_offset, nlist_byte_size)) - break; - - nlist.n_strx = dsc_local_symbols_data.GetU32_unchecked(&nlist_data_offset); - nlist.n_type = dsc_local_symbols_data.GetU8_unchecked (&nlist_data_offset); - nlist.n_sect = dsc_local_symbols_data.GetU8_unchecked (&nlist_data_offset); - nlist.n_desc = dsc_local_symbols_data.GetU16_unchecked (&nlist_data_offset); - nlist.n_value = dsc_local_symbols_data.GetAddress_unchecked (&nlist_data_offset); - - SymbolType type = eSymbolTypeInvalid; - const char *symbol_name = dsc_local_symbols_data.PeekCStr(string_table_offset + nlist.n_strx); - - if (symbol_name == NULL) - { - // No symbol should be NULL, even the symbols with no - // string values should have an offset zero which points - // to an empty C-string - Host::SystemLog (Host::eSystemLogError, - "error: DSC unmapped local symbol[%u] has invalid string table offset 0x%x in %s, ignoring symbol\n", - entry_index, - nlist.n_strx, - module_sp->GetFileSpec().GetPath().c_str()); - continue; - } - if (symbol_name[0] == '\0') - symbol_name = NULL; - - const char *symbol_name_non_abi_mangled = NULL; - - SectionSP symbol_section; - uint32_t symbol_byte_size = 0; - bool add_nlist = true; - bool is_debug = ((nlist.n_type & N_STAB) != 0); - bool demangled_is_synthesized = false; - bool is_gsym = false; - bool set_value = true; - - assert (sym_idx < num_syms); - - sym[sym_idx].SetDebug (is_debug); - - if (is_debug) - { - switch (nlist.n_type) - { - case N_GSYM: - // global symbol: name,,NO_SECT,type,0 - // Sometimes the N_GSYM value contains the address. - - // FIXME: In the .o files, we have a GSYM and a debug symbol for all the ObjC data. They - // have the same address, but we want to ensure that we always find only the real symbol, - // 'cause we don't currently correctly attribute the GSYM one to the ObjCClass/Ivar/MetaClass - // symbol type. This is a temporary hack to make sure the ObjectiveC symbols get treated - // correctly. To do this right, we should coalesce all the GSYM & global symbols that have the - // same address. - - is_gsym = true; - sym[sym_idx].SetExternal(true); - - if (symbol_name && symbol_name[0] == '_' && symbol_name[1] == 'O') - { - llvm::StringRef symbol_name_ref(symbol_name); - if (symbol_name_ref.startswith(g_objc_v2_prefix_class)) - { - symbol_name_non_abi_mangled = symbol_name + 1; - symbol_name = symbol_name + g_objc_v2_prefix_class.size(); - type = eSymbolTypeObjCClass; - demangled_is_synthesized = true; - - } - else if (symbol_name_ref.startswith(g_objc_v2_prefix_metaclass)) - { - symbol_name_non_abi_mangled = symbol_name + 1; - symbol_name = symbol_name + g_objc_v2_prefix_metaclass.size(); - type = eSymbolTypeObjCMetaClass; - demangled_is_synthesized = true; - } - else if (symbol_name_ref.startswith(g_objc_v2_prefix_ivar)) - { - symbol_name_non_abi_mangled = symbol_name + 1; - symbol_name = symbol_name + g_objc_v2_prefix_ivar.size(); - type = eSymbolTypeObjCIVar; - demangled_is_synthesized = true; - } - } - else - { - if (nlist.n_value != 0) - symbol_section = section_info.GetSection (nlist.n_sect, nlist.n_value); - type = eSymbolTypeData; - } - break; - - case N_FNAME: - // procedure name (f77 kludge): name,,NO_SECT,0,0 - type = eSymbolTypeCompiler; - break; - - case N_FUN: - // procedure: name,,n_sect,linenumber,address - if (symbol_name) - { - type = eSymbolTypeCode; - symbol_section = section_info.GetSection (nlist.n_sect, nlist.n_value); - - N_FUN_addr_to_sym_idx.insert(std::make_pair(nlist.n_value, sym_idx)); - // We use the current number of symbols in the symbol table in lieu of - // using nlist_idx in case we ever start trimming entries out - N_FUN_indexes.push_back(sym_idx); - } - else - { - type = eSymbolTypeCompiler; - - if ( !N_FUN_indexes.empty() ) - { - // Copy the size of the function into the original STAB entry so we don't have - // to hunt for it later - symtab->SymbolAtIndex(N_FUN_indexes.back())->SetByteSize(nlist.n_value); - N_FUN_indexes.pop_back(); - // We don't really need the end function STAB as it contains the size which - // we already placed with the original symbol, so don't add it if we want a - // minimal symbol table - add_nlist = false; - } - } - break; - - case N_STSYM: - // static symbol: name,,n_sect,type,address - N_STSYM_addr_to_sym_idx.insert(std::make_pair(nlist.n_value, sym_idx)); - symbol_section = section_info.GetSection (nlist.n_sect, nlist.n_value); - if (symbol_name && symbol_name[0]) - { - type = ObjectFile::GetSymbolTypeFromName(symbol_name+1, eSymbolTypeData); - } - break; - - case N_LCSYM: - // .lcomm symbol: name,,n_sect,type,address - symbol_section = section_info.GetSection (nlist.n_sect, nlist.n_value); - type = eSymbolTypeCommonBlock; - break; - - case N_BNSYM: - // We use the current number of symbols in the symbol table in lieu of - // using nlist_idx in case we ever start trimming entries out - // Skip these if we want minimal symbol tables - add_nlist = false; - break; - - case N_ENSYM: - // Set the size of the N_BNSYM to the terminating index of this N_ENSYM - // so that we can always skip the entire symbol if we need to navigate - // more quickly at the source level when parsing STABS - // Skip these if we want minimal symbol tables - add_nlist = false; - break; - - case N_OPT: - // emitted with gcc2_compiled and in gcc source - type = eSymbolTypeCompiler; - break; - - case N_RSYM: - // register sym: name,,NO_SECT,type,register - type = eSymbolTypeVariable; - break; - - case N_SLINE: - // src line: 0,,n_sect,linenumber,address - symbol_section = section_info.GetSection (nlist.n_sect, nlist.n_value); - type = eSymbolTypeLineEntry; - break; - - case N_SSYM: - // structure elt: name,,NO_SECT,type,struct_offset - type = eSymbolTypeVariableType; - break; - - case N_SO: - // source file name - type = eSymbolTypeSourceFile; - if (symbol_name == NULL) - { - add_nlist = false; - if (N_SO_index != UINT32_MAX) - { - // Set the size of the N_SO to the terminating index of this N_SO - // so that we can always skip the entire N_SO if we need to navigate - // more quickly at the source level when parsing STABS - symbol_ptr = symtab->SymbolAtIndex(N_SO_index); - symbol_ptr->SetByteSize(sym_idx); - symbol_ptr->SetSizeIsSibling(true); - } - N_NSYM_indexes.clear(); - N_INCL_indexes.clear(); - N_BRAC_indexes.clear(); - N_COMM_indexes.clear(); - N_FUN_indexes.clear(); - N_SO_index = UINT32_MAX; - } - else - { - // We use the current number of symbols in the symbol table in lieu of - // using nlist_idx in case we ever start trimming entries out - const bool N_SO_has_full_path = symbol_name[0] == '/'; - if (N_SO_has_full_path) - { - if ((N_SO_index == sym_idx - 1) && ((sym_idx - 1) < num_syms)) - { - // We have two consecutive N_SO entries where the first contains a directory - // and the second contains a full path. - sym[sym_idx - 1].GetMangled().SetValue(ConstString(symbol_name), false); - m_nlist_idx_to_sym_idx[nlist_idx] = sym_idx - 1; - add_nlist = false; - } - else - { - // This is the first entry in a N_SO that contains a directory or - // a full path to the source file - N_SO_index = sym_idx; - } - } - else if ((N_SO_index == sym_idx - 1) && ((sym_idx - 1) < num_syms)) - { - // This is usually the second N_SO entry that contains just the filename, - // so here we combine it with the first one if we are minimizing the symbol table - const char *so_path = sym[sym_idx - 1].GetMangled().GetDemangledName(lldb::eLanguageTypeUnknown).AsCString(); - if (so_path && so_path[0]) - { - std::string full_so_path (so_path); - const size_t double_slash_pos = full_so_path.find("//"); - if (double_slash_pos != std::string::npos) - { - // The linker has been generating bad N_SO entries with doubled up paths - // in the format "%s%s" where the first string in the DW_AT_comp_dir, - // and the second is the directory for the source file so you end up with - // a path that looks like "/tmp/src//tmp/src/" - FileSpec so_dir(so_path, false); - if (!so_dir.Exists()) - { - so_dir.SetFile(&full_so_path[double_slash_pos + 1], false); - if (so_dir.Exists()) - { - // Trim off the incorrect path - full_so_path.erase(0, double_slash_pos + 1); - } - } - } - if (*full_so_path.rbegin() != '/') - full_so_path += '/'; - full_so_path += symbol_name; - sym[sym_idx - 1].GetMangled().SetValue(ConstString(full_so_path.c_str()), false); - add_nlist = false; - m_nlist_idx_to_sym_idx[nlist_idx] = sym_idx - 1; - } - } - else - { - // This could be a relative path to a N_SO - N_SO_index = sym_idx; - } - } - break; - - case N_OSO: - // object file name: name,,0,0,st_mtime - type = eSymbolTypeObjectFile; - break; - - case N_LSYM: - // local sym: name,,NO_SECT,type,offset - type = eSymbolTypeLocal; - break; - - //---------------------------------------------------------------------- - // INCL scopes - //---------------------------------------------------------------------- - case N_BINCL: - // include file beginning: name,,NO_SECT,0,sum - // We use the current number of symbols in the symbol table in lieu of - // using nlist_idx in case we ever start trimming entries out - N_INCL_indexes.push_back(sym_idx); - type = eSymbolTypeScopeBegin; - break; - - case N_EINCL: - // include file end: name,,NO_SECT,0,0 - // Set the size of the N_BINCL to the terminating index of this N_EINCL - // so that we can always skip the entire symbol if we need to navigate - // more quickly at the source level when parsing STABS - if ( !N_INCL_indexes.empty() ) - { - symbol_ptr = symtab->SymbolAtIndex(N_INCL_indexes.back()); - symbol_ptr->SetByteSize(sym_idx + 1); - symbol_ptr->SetSizeIsSibling(true); - N_INCL_indexes.pop_back(); - } - type = eSymbolTypeScopeEnd; - break; - - case N_SOL: - // #included file name: name,,n_sect,0,address - type = eSymbolTypeHeaderFile; - - // We currently don't use the header files on darwin - add_nlist = false; - break; - - case N_PARAMS: - // compiler parameters: name,,NO_SECT,0,0 - type = eSymbolTypeCompiler; - break; - - case N_VERSION: - // compiler version: name,,NO_SECT,0,0 - type = eSymbolTypeCompiler; - break; - - case N_OLEVEL: - // compiler -O level: name,,NO_SECT,0,0 - type = eSymbolTypeCompiler; - break; - - case N_PSYM: - // parameter: name,,NO_SECT,type,offset - type = eSymbolTypeVariable; - break; - - case N_ENTRY: - // alternate entry: name,,n_sect,linenumber,address - symbol_section = section_info.GetSection (nlist.n_sect, nlist.n_value); - type = eSymbolTypeLineEntry; - break; - - //---------------------------------------------------------------------- - // Left and Right Braces - //---------------------------------------------------------------------- - case N_LBRAC: - // left bracket: 0,,NO_SECT,nesting level,address - // We use the current number of symbols in the symbol table in lieu of - // using nlist_idx in case we ever start trimming entries out - symbol_section = section_info.GetSection (nlist.n_sect, nlist.n_value); - N_BRAC_indexes.push_back(sym_idx); - type = eSymbolTypeScopeBegin; - break; - - case N_RBRAC: - // right bracket: 0,,NO_SECT,nesting level,address - // Set the size of the N_LBRAC to the terminating index of this N_RBRAC - // so that we can always skip the entire symbol if we need to navigate - // more quickly at the source level when parsing STABS - symbol_section = section_info.GetSection (nlist.n_sect, nlist.n_value); - if ( !N_BRAC_indexes.empty() ) - { - symbol_ptr = symtab->SymbolAtIndex(N_BRAC_indexes.back()); - symbol_ptr->SetByteSize(sym_idx + 1); - symbol_ptr->SetSizeIsSibling(true); - N_BRAC_indexes.pop_back(); - } - type = eSymbolTypeScopeEnd; - break; - - case N_EXCL: - // deleted include file: name,,NO_SECT,0,sum - type = eSymbolTypeHeaderFile; - break; - - //---------------------------------------------------------------------- - // COMM scopes - //---------------------------------------------------------------------- - case N_BCOMM: - // begin common: name,,NO_SECT,0,0 - // We use the current number of symbols in the symbol table in lieu of - // using nlist_idx in case we ever start trimming entries out - type = eSymbolTypeScopeBegin; - N_COMM_indexes.push_back(sym_idx); - break; - - case N_ECOML: - // end common (local name): 0,,n_sect,0,address - symbol_section = section_info.GetSection (nlist.n_sect, nlist.n_value); - // Fall through - - case N_ECOMM: - // end common: name,,n_sect,0,0 - // Set the size of the N_BCOMM to the terminating index of this N_ECOMM/N_ECOML - // so that we can always skip the entire symbol if we need to navigate - // more quickly at the source level when parsing STABS - if ( !N_COMM_indexes.empty() ) - { - symbol_ptr = symtab->SymbolAtIndex(N_COMM_indexes.back()); - symbol_ptr->SetByteSize(sym_idx + 1); - symbol_ptr->SetSizeIsSibling(true); - N_COMM_indexes.pop_back(); - } - type = eSymbolTypeScopeEnd; - break; - - case N_LENG: - // second stab entry with length information - type = eSymbolTypeAdditional; - break; - - default: break; - } - } - else - { - //uint8_t n_pext = N_PEXT & nlist.n_type; - uint8_t n_type = N_TYPE & nlist.n_type; - sym[sym_idx].SetExternal((N_EXT & nlist.n_type) != 0); - - switch (n_type) - { - case N_INDR: - { - const char *reexport_name_cstr = strtab_data.PeekCStr(nlist.n_value); - if (reexport_name_cstr && reexport_name_cstr[0]) - { - type = eSymbolTypeReExported; - ConstString reexport_name(reexport_name_cstr + ((reexport_name_cstr[0] == '_') ? 1 : 0)); - sym[sym_idx].SetReExportedSymbolName(reexport_name); - set_value = false; - reexport_shlib_needs_fixup[sym_idx] = reexport_name; - indirect_symbol_names.insert(ConstString(symbol_name + ((symbol_name[0] == '_') ? 1 : 0))); - } - else - type = eSymbolTypeUndefined; - } - break; - - case N_UNDF: - if (symbol_name && symbol_name[0]) - { - ConstString undefined_name(symbol_name + ((symbol_name[0] == '_') ? 1 : 0)); - undefined_name_to_desc[undefined_name] = nlist.n_desc; - } - // Fall through - case N_PBUD: - type = eSymbolTypeUndefined; - break; - - case N_ABS: - type = eSymbolTypeAbsolute; - break; - - case N_SECT: - { - symbol_section = section_info.GetSection (nlist.n_sect, nlist.n_value); - - if (symbol_section == NULL) - { - // TODO: warn about this? - add_nlist = false; - break; - } - - if (TEXT_eh_frame_sectID == nlist.n_sect) - { - type = eSymbolTypeException; - } - else - { - uint32_t section_type = symbol_section->Get() & SECTION_TYPE; - - switch (section_type) - { - case S_CSTRING_LITERALS: type = eSymbolTypeData; break; // section with only literal C strings - case S_4BYTE_LITERALS: type = eSymbolTypeData; break; // section with only 4 byte literals - case S_8BYTE_LITERALS: type = eSymbolTypeData; break; // section with only 8 byte literals - case S_LITERAL_POINTERS: type = eSymbolTypeTrampoline; break; // section with only pointers to literals - case S_NON_LAZY_SYMBOL_POINTERS: type = eSymbolTypeTrampoline; break; // section with only non-lazy symbol pointers - case S_LAZY_SYMBOL_POINTERS: type = eSymbolTypeTrampoline; break; // section with only lazy symbol pointers - case S_SYMBOL_STUBS: type = eSymbolTypeTrampoline; break; // section with only symbol stubs, byte size of stub in the reserved2 field - case S_MOD_INIT_FUNC_POINTERS: type = eSymbolTypeCode; break; // section with only function pointers for initialization - case S_MOD_TERM_FUNC_POINTERS: type = eSymbolTypeCode; break; // section with only function pointers for termination - case S_INTERPOSING: type = eSymbolTypeTrampoline; break; // section with only pairs of function pointers for interposing - case S_16BYTE_LITERALS: type = eSymbolTypeData; break; // section with only 16 byte literals - case S_DTRACE_DOF: type = eSymbolTypeInstrumentation; break; - case S_LAZY_DYLIB_SYMBOL_POINTERS: type = eSymbolTypeTrampoline; break; - default: - switch (symbol_section->GetType()) - { - case lldb::eSectionTypeCode: - type = eSymbolTypeCode; - break; - case eSectionTypeData: - case eSectionTypeDataCString: // Inlined C string data - case eSectionTypeDataCStringPointers: // Pointers to C string data - case eSectionTypeDataSymbolAddress: // Address of a symbol in the symbol table - case eSectionTypeData4: - case eSectionTypeData8: - case eSectionTypeData16: - type = eSymbolTypeData; - break; - default: - break; - } - break; - } - - if (type == eSymbolTypeInvalid) - { - const char *symbol_sect_name = symbol_section->GetName().AsCString(); - if (symbol_section->IsDescendant (text_section_sp.get())) - { - if (symbol_section->IsClear(S_ATTR_PURE_INSTRUCTIONS | - S_ATTR_SELF_MODIFYING_CODE | - S_ATTR_SOME_INSTRUCTIONS)) - type = eSymbolTypeData; - else - type = eSymbolTypeCode; - } - else if (symbol_section->IsDescendant(data_section_sp.get()) || - symbol_section->IsDescendant(data_dirty_section_sp.get()) || - symbol_section->IsDescendant(data_const_section_sp.get())) - { - if (symbol_sect_name && ::strstr (symbol_sect_name, "__objc") == symbol_sect_name) - { - type = eSymbolTypeRuntime; - - if (symbol_name) - { - llvm::StringRef symbol_name_ref(symbol_name); - if (symbol_name_ref.startswith("_OBJC_")) - { - static const llvm::StringRef g_objc_v2_prefix_class ("_OBJC_CLASS_$_"); - static const llvm::StringRef g_objc_v2_prefix_metaclass ("_OBJC_METACLASS_$_"); - static const llvm::StringRef g_objc_v2_prefix_ivar ("_OBJC_IVAR_$_"); - if (symbol_name_ref.startswith(g_objc_v2_prefix_class)) - { - symbol_name_non_abi_mangled = symbol_name + 1; - symbol_name = symbol_name + g_objc_v2_prefix_class.size(); - type = eSymbolTypeObjCClass; - demangled_is_synthesized = true; - } - else if (symbol_name_ref.startswith(g_objc_v2_prefix_metaclass)) - { - symbol_name_non_abi_mangled = symbol_name + 1; - symbol_name = symbol_name + g_objc_v2_prefix_metaclass.size(); - type = eSymbolTypeObjCMetaClass; - demangled_is_synthesized = true; - } - else if (symbol_name_ref.startswith(g_objc_v2_prefix_ivar)) - { - symbol_name_non_abi_mangled = symbol_name + 1; - symbol_name = symbol_name + g_objc_v2_prefix_ivar.size(); - type = eSymbolTypeObjCIVar; - demangled_is_synthesized = true; - } - } - } - } - else if (symbol_sect_name && ::strstr (symbol_sect_name, "__gcc_except_tab") == symbol_sect_name) - { - type = eSymbolTypeException; - } - else - { - type = eSymbolTypeData; - } - } - else if (symbol_sect_name && ::strstr (symbol_sect_name, "__IMPORT") == symbol_sect_name) - { - type = eSymbolTypeTrampoline; - } - else if (symbol_section->IsDescendant(objc_section_sp.get())) - { - type = eSymbolTypeRuntime; - if (symbol_name && symbol_name[0] == '.') - { - llvm::StringRef symbol_name_ref(symbol_name); - static const llvm::StringRef g_objc_v1_prefix_class (".objc_class_name_"); - if (symbol_name_ref.startswith(g_objc_v1_prefix_class)) - { - symbol_name_non_abi_mangled = symbol_name; - symbol_name = symbol_name + g_objc_v1_prefix_class.size(); - type = eSymbolTypeObjCClass; - demangled_is_synthesized = true; - } - } - } - } - } - } - break; - } - } - - if (add_nlist) - { - uint64_t symbol_value = nlist.n_value; - if (symbol_name_non_abi_mangled) - { - sym[sym_idx].GetMangled().SetMangledName (ConstString(symbol_name_non_abi_mangled)); - sym[sym_idx].GetMangled().SetDemangledName (ConstString(symbol_name)); - } - else - { - bool symbol_name_is_mangled = false; - - if (symbol_name && symbol_name[0] == '_') - { - symbol_name_is_mangled = symbol_name[1] == '_'; - symbol_name++; // Skip the leading underscore - } - - if (symbol_name) - { - ConstString const_symbol_name(symbol_name); - sym[sym_idx].GetMangled().SetValue(const_symbol_name, symbol_name_is_mangled); - if (is_gsym && is_debug) - { - const char *gsym_name = sym[sym_idx].GetMangled().GetName(lldb::eLanguageTypeUnknown, Mangled::ePreferMangled).GetCString(); - if (gsym_name) - N_GSYM_name_to_sym_idx[gsym_name] = sym_idx; - } - } - } - if (symbol_section) - { - const addr_t section_file_addr = symbol_section->GetFileAddress(); - if (symbol_byte_size == 0 && function_starts_count > 0) - { - addr_t symbol_lookup_file_addr = nlist.n_value; - // Do an exact address match for non-ARM addresses, else get the closest since - // the symbol might be a thumb symbol which has an address with bit zero set - FunctionStarts::Entry *func_start_entry = function_starts.FindEntry (symbol_lookup_file_addr, !is_arm); - if (is_arm && func_start_entry) - { - // Verify that the function start address is the symbol address (ARM) - // or the symbol address + 1 (thumb) - if (func_start_entry->addr != symbol_lookup_file_addr && - func_start_entry->addr != (symbol_lookup_file_addr + 1)) - { - // Not the right entry, NULL it out... - func_start_entry = NULL; - } - } - if (func_start_entry) - { - func_start_entry->data = true; - - addr_t symbol_file_addr = func_start_entry->addr; - uint32_t symbol_flags = 0; - if (is_arm) - { - if (symbol_file_addr & 1) - symbol_flags = MACHO_NLIST_ARM_SYMBOL_IS_THUMB; - symbol_file_addr &= THUMB_ADDRESS_BIT_MASK; - } - - const FunctionStarts::Entry *next_func_start_entry = function_starts.FindNextEntry (func_start_entry); - const addr_t section_end_file_addr = section_file_addr + symbol_section->GetByteSize(); - if (next_func_start_entry) - { - addr_t next_symbol_file_addr = next_func_start_entry->addr; - // Be sure the clear the Thumb address bit when we calculate the size - // from the current and next address - if (is_arm) - next_symbol_file_addr &= THUMB_ADDRESS_BIT_MASK; - symbol_byte_size = std::min<lldb::addr_t>(next_symbol_file_addr - symbol_file_addr, section_end_file_addr - symbol_file_addr); - } - else - { - symbol_byte_size = section_end_file_addr - symbol_file_addr; - } - } - } - symbol_value -= section_file_addr; - } - - if (is_debug == false) - { - if (type == eSymbolTypeCode) - { - // See if we can find a N_FUN entry for any code symbols. - // If we do find a match, and the name matches, then we - // can merge the two into just the function symbol to avoid - // duplicate entries in the symbol table - std::pair<ValueToSymbolIndexMap::const_iterator, ValueToSymbolIndexMap::const_iterator> range; - range = N_FUN_addr_to_sym_idx.equal_range(nlist.n_value); - if (range.first != range.second) - { - bool found_it = false; - for (ValueToSymbolIndexMap::const_iterator pos = range.first; pos != range.second; ++pos) - { - if (sym[sym_idx].GetMangled().GetName(lldb::eLanguageTypeUnknown, Mangled::ePreferMangled) == sym[pos->second].GetMangled().GetName(lldb::eLanguageTypeUnknown, Mangled::ePreferMangled)) - { - m_nlist_idx_to_sym_idx[nlist_idx] = pos->second; - // We just need the flags from the linker symbol, so put these flags - // into the N_FUN flags to avoid duplicate symbols in the symbol table - sym[pos->second].SetExternal(sym[sym_idx].IsExternal()); - sym[pos->second].SetFlags (nlist.n_type << 16 | nlist.n_desc); - if (resolver_addresses.find(nlist.n_value) != resolver_addresses.end()) - sym[pos->second].SetType (eSymbolTypeResolver); - sym[sym_idx].Clear(); - found_it = true; - break; - } - } - if (found_it) - continue; - } - else - { - if (resolver_addresses.find(nlist.n_value) != resolver_addresses.end()) - type = eSymbolTypeResolver; - } - } - else if (type == eSymbolTypeData || - type == eSymbolTypeObjCClass || - type == eSymbolTypeObjCMetaClass || - type == eSymbolTypeObjCIVar ) - { - // See if we can find a N_STSYM entry for any data symbols. - // If we do find a match, and the name matches, then we - // can merge the two into just the Static symbol to avoid - // duplicate entries in the symbol table - std::pair<ValueToSymbolIndexMap::const_iterator, ValueToSymbolIndexMap::const_iterator> range; - range = N_STSYM_addr_to_sym_idx.equal_range(nlist.n_value); - if (range.first != range.second) - { - bool found_it = false; - for (ValueToSymbolIndexMap::const_iterator pos = range.first; pos != range.second; ++pos) - { - if (sym[sym_idx].GetMangled().GetName(lldb::eLanguageTypeUnknown, Mangled::ePreferMangled) == sym[pos->second].GetMangled().GetName(lldb::eLanguageTypeUnknown, Mangled::ePreferMangled)) - { - m_nlist_idx_to_sym_idx[nlist_idx] = pos->second; - // We just need the flags from the linker symbol, so put these flags - // into the N_STSYM flags to avoid duplicate symbols in the symbol table - sym[pos->second].SetExternal(sym[sym_idx].IsExternal()); - sym[pos->second].SetFlags (nlist.n_type << 16 | nlist.n_desc); - sym[sym_idx].Clear(); - found_it = true; - break; - } - } - if (found_it) - continue; - } - else - { - const char *gsym_name = sym[sym_idx].GetMangled().GetName(lldb::eLanguageTypeUnknown, Mangled::ePreferMangled).GetCString(); - if (gsym_name) - { - // Combine N_GSYM stab entries with the non stab symbol - ConstNameToSymbolIndexMap::const_iterator pos = N_GSYM_name_to_sym_idx.find(gsym_name); - if (pos != N_GSYM_name_to_sym_idx.end()) - { - const uint32_t GSYM_sym_idx = pos->second; - m_nlist_idx_to_sym_idx[nlist_idx] = GSYM_sym_idx; - // Copy the address, because often the N_GSYM address has an invalid address of zero - // when the global is a common symbol - sym[GSYM_sym_idx].GetAddressRef().SetSection (symbol_section); - sym[GSYM_sym_idx].GetAddressRef().SetOffset (symbol_value); - // We just need the flags from the linker symbol, so put these flags - // into the N_GSYM flags to avoid duplicate symbols in the symbol table - sym[GSYM_sym_idx].SetFlags (nlist.n_type << 16 | nlist.n_desc); - sym[sym_idx].Clear(); - continue; - } - } - } - } - } - - sym[sym_idx].SetID (nlist_idx); - sym[sym_idx].SetType (type); - if (set_value) - { - sym[sym_idx].GetAddressRef().SetSection (symbol_section); - sym[sym_idx].GetAddressRef().SetOffset (symbol_value); - } - sym[sym_idx].SetFlags (nlist.n_type << 16 | nlist.n_desc); - - if (symbol_byte_size > 0) - sym[sym_idx].SetByteSize(symbol_byte_size); - - if (demangled_is_synthesized) - sym[sym_idx].SetDemangledNameIsSynthesized(true); - ++sym_idx; - } - else - { - sym[sym_idx].Clear(); - } - - } - ///////////////////////////// - } - break; // No more entries to consider - } - } + // For user process binaries (executables, dylibs, frameworks, bundles), if + // we don't have + // LC_FUNCTION_STARTS/eh_frame section in this binary, we're going to assume + // the binary + // has been stripped. Don't allow assembly language instruction emulation + // because we don't + // know proper function start boundaries. + // + // For all other types of binaries (kernels, stand-alone bare board + // binaries, kexts), they + // may not have LC_FUNCTION_STARTS / eh_frame sections - we should not make + // any assumptions + // about them based on that. + if (function_starts_count == 0 && CalculateStrata() == eStrataUser) { + m_allow_assembly_emulation_unwind_plans = false; + Log *unwind_or_symbol_log(lldb_private::GetLogIfAnyCategoriesSet( + LIBLLDB_LOG_SYMBOLS | LIBLLDB_LOG_UNWIND)); + + if (unwind_or_symbol_log) + module_sp->LogMessage( + unwind_or_symbol_log, + "no LC_FUNCTION_STARTS, will not allow assembly profiled unwinds"); + } - for (const auto &pos :reexport_shlib_needs_fixup) - { - const auto undef_pos = undefined_name_to_desc.find(pos.second); - if (undef_pos != undefined_name_to_desc.end()) - { - const uint8_t dylib_ordinal = llvm::MachO::GET_LIBRARY_ORDINAL(undef_pos->second); - if (dylib_ordinal > 0 && dylib_ordinal < dylib_files.GetSize()) - sym[pos.first].SetReExportedSymbolSharedLibrary(dylib_files.GetFileSpecAtIndex(dylib_ordinal-1)); - } - } - } - } - } - } + const user_id_t TEXT_eh_frame_sectID = + eh_frame_section_sp.get() ? eh_frame_section_sp->GetID() + : static_cast<user_id_t>(NO_SECT); + + lldb::offset_t nlist_data_offset = 0; + + uint32_t N_SO_index = UINT32_MAX; + + MachSymtabSectionInfo section_info(section_list); + std::vector<uint32_t> N_FUN_indexes; + std::vector<uint32_t> N_NSYM_indexes; + std::vector<uint32_t> N_INCL_indexes; + std::vector<uint32_t> N_BRAC_indexes; + std::vector<uint32_t> N_COMM_indexes; + typedef std::multimap<uint64_t, uint32_t> ValueToSymbolIndexMap; + typedef std::map<uint32_t, uint32_t> NListIndexToSymbolIndexMap; + typedef std::map<const char *, uint32_t> ConstNameToSymbolIndexMap; + ValueToSymbolIndexMap N_FUN_addr_to_sym_idx; + ValueToSymbolIndexMap N_STSYM_addr_to_sym_idx; + ConstNameToSymbolIndexMap N_GSYM_name_to_sym_idx; + // Any symbols that get merged into another will get an entry + // in this map so we know + NListIndexToSymbolIndexMap m_nlist_idx_to_sym_idx; + uint32_t nlist_idx = 0; + Symbol *symbol_ptr = NULL; + + uint32_t sym_idx = 0; + Symbol *sym = NULL; + size_t num_syms = 0; + std::string memory_symbol_name; + uint32_t unmapped_local_symbols_found = 0; + + std::vector<TrieEntryWithOffset> trie_entries; + std::set<lldb::addr_t> resolver_addresses; + + if (dyld_trie_data.GetByteSize() > 0) { + std::vector<llvm::StringRef> nameSlices; + ParseTrieEntries(dyld_trie_data, 0, is_arm, nameSlices, + resolver_addresses, trie_entries); + + ConstString text_segment_name("__TEXT"); + SectionSP text_segment_sp = + GetSectionList()->FindSectionByName(text_segment_name); + if (text_segment_sp) { + const lldb::addr_t text_segment_file_addr = + text_segment_sp->GetFileAddress(); + if (text_segment_file_addr != LLDB_INVALID_ADDRESS) { + for (auto &e : trie_entries) + e.entry.address += text_segment_file_addr; } + } + } - // Must reset this in case it was mutated above! - nlist_data_offset = 0; -#endif - - if (nlist_data.GetByteSize() > 0) - { - - // If the sym array was not created while parsing the DSC unmapped - // symbols, create it now. - if (sym == NULL) - { - sym = symtab->Resize (symtab_load_command.nsyms + m_dysymtab.nindirectsyms); - num_syms = symtab->GetNumSymbols(); - } - - if (unmapped_local_symbols_found) - { - assert(m_dysymtab.ilocalsym == 0); - nlist_data_offset += (m_dysymtab.nlocalsym * nlist_byte_size); - nlist_idx = m_dysymtab.nlocalsym; - } - else - { - nlist_idx = 0; - } - - typedef std::map<ConstString, uint16_t> UndefinedNameToDescMap; - typedef std::map<uint32_t, ConstString> SymbolIndexToName; - UndefinedNameToDescMap undefined_name_to_desc; - SymbolIndexToName reexport_shlib_needs_fixup; - for (; nlist_idx < symtab_load_command.nsyms; ++nlist_idx) - { - struct nlist_64 nlist; - if (!nlist_data.ValidOffsetForDataOfSize(nlist_data_offset, nlist_byte_size)) - break; - - nlist.n_strx = nlist_data.GetU32_unchecked(&nlist_data_offset); - nlist.n_type = nlist_data.GetU8_unchecked (&nlist_data_offset); - nlist.n_sect = nlist_data.GetU8_unchecked (&nlist_data_offset); - nlist.n_desc = nlist_data.GetU16_unchecked (&nlist_data_offset); - nlist.n_value = nlist_data.GetAddress_unchecked (&nlist_data_offset); - - SymbolType type = eSymbolTypeInvalid; - const char *symbol_name = NULL; - - if (have_strtab_data) - { - symbol_name = strtab_data.PeekCStr(nlist.n_strx); + typedef std::set<ConstString> IndirectSymbols; + IndirectSymbols indirect_symbol_names; + +#if defined(__APPLE__) && \ + (defined(__arm__) || defined(__arm64__) || defined(__aarch64__)) + + // Some recent builds of the dyld_shared_cache (hereafter: DSC) have been + // optimized by moving LOCAL + // symbols out of the memory mapped portion of the DSC. The symbol + // information has all been retained, + // but it isn't available in the normal nlist data. However, there *are* + // duplicate entries of *some* + // LOCAL symbols in the normal nlist data. To handle this situation + // correctly, we must first attempt + // to parse any DSC unmapped symbol information. If we find any, we set a + // flag that tells the normal + // nlist parser to ignore all LOCAL symbols. + + if (m_header.flags & 0x80000000u) { + // Before we can start mapping the DSC, we need to make certain the target + // process is actually + // using the cache we can find. + + // Next we need to determine the correct path for the dyld shared cache. + + ArchSpec header_arch; + GetArchitecture(header_arch); + char dsc_path[PATH_MAX]; + char dsc_path_development[PATH_MAX]; + + snprintf( + dsc_path, sizeof(dsc_path), "%s%s%s", + "/System/Library/Caches/com.apple.dyld/", /* IPHONE_DYLD_SHARED_CACHE_DIR + */ + "dyld_shared_cache_", /* DYLD_SHARED_CACHE_BASE_NAME */ + header_arch.GetArchitectureName()); + + snprintf( + dsc_path_development, sizeof(dsc_path), "%s%s%s%s", + "/System/Library/Caches/com.apple.dyld/", /* IPHONE_DYLD_SHARED_CACHE_DIR + */ + "dyld_shared_cache_", /* DYLD_SHARED_CACHE_BASE_NAME */ + header_arch.GetArchitectureName(), ".development"); + + FileSpec dsc_nondevelopment_filespec(dsc_path, false); + FileSpec dsc_development_filespec(dsc_path_development, false); + FileSpec dsc_filespec; + + UUID dsc_uuid; + UUID process_shared_cache_uuid; + + if (process) { + process_shared_cache_uuid = GetProcessSharedCacheUUID(process); + } + + // First see if we can find an exact match for the inferior process shared + // cache UUID in + // the development or non-development shared caches on disk. + if (process_shared_cache_uuid.IsValid()) { + if (dsc_development_filespec.Exists()) { + UUID dsc_development_uuid = GetSharedCacheUUID( + dsc_development_filespec, byte_order, addr_byte_size); + if (dsc_development_uuid.IsValid() && + dsc_development_uuid == process_shared_cache_uuid) { + dsc_filespec = dsc_development_filespec; + dsc_uuid = dsc_development_uuid; + } + } + if (!dsc_uuid.IsValid() && dsc_nondevelopment_filespec.Exists()) { + UUID dsc_nondevelopment_uuid = GetSharedCacheUUID( + dsc_nondevelopment_filespec, byte_order, addr_byte_size); + if (dsc_nondevelopment_uuid.IsValid() && + dsc_nondevelopment_uuid == process_shared_cache_uuid) { + dsc_filespec = dsc_nondevelopment_filespec; + dsc_uuid = dsc_nondevelopment_uuid; + } + } + } + + // Failing a UUID match, prefer the development dyld_shared cache if both + // are present. + if (!dsc_filespec.Exists()) { + if (dsc_development_filespec.Exists()) { + dsc_filespec = dsc_development_filespec; + } else { + dsc_filespec = dsc_nondevelopment_filespec; + } + } + + /* The dyld_cache_header has a pointer to the + dyld_cache_local_symbols_info structure (localSymbolsOffset). + The dyld_cache_local_symbols_info structure gives us three things: + 1. The start and count of the nlist records in the dyld_shared_cache + file + 2. The start and size of the strings for these nlist records + 3. The start and count of dyld_cache_local_symbols_entry entries + + There is one dyld_cache_local_symbols_entry per dylib/framework in the + dyld shared cache. + The "dylibOffset" field is the Mach-O header of this dylib/framework in + the dyld shared cache. + The dyld_cache_local_symbols_entry also lists the start of this + dylib/framework's nlist records + and the count of how many nlist records there are for this + dylib/framework. + */ + + // Process the dyld shared cache header to find the unmapped symbols + + DataBufferSP dsc_data_sp = dsc_filespec.MemoryMapFileContentsIfLocal( + 0, sizeof(struct lldb_copy_dyld_cache_header_v1)); + if (!dsc_uuid.IsValid()) { + dsc_uuid = GetSharedCacheUUID(dsc_filespec, byte_order, addr_byte_size); + } + if (dsc_data_sp) { + DataExtractor dsc_header_data(dsc_data_sp, byte_order, addr_byte_size); + + bool uuid_match = true; + if (dsc_uuid.IsValid() && process) { + if (process_shared_cache_uuid.IsValid() && + dsc_uuid != process_shared_cache_uuid) { + // The on-disk dyld_shared_cache file is not the same as the one in + // this + // process' memory, don't use it. + uuid_match = false; + ModuleSP module_sp(GetModule()); + if (module_sp) + module_sp->ReportWarning("process shared cache does not match " + "on-disk dyld_shared_cache file, some " + "symbol names will be missing."); + } + } - if (symbol_name == NULL) + offset = offsetof(struct lldb_copy_dyld_cache_header_v1, mappingOffset); + + uint32_t mappingOffset = dsc_header_data.GetU32(&offset); + + // If the mappingOffset points to a location inside the header, we've + // opened an old dyld shared cache, and should not proceed further. + if (uuid_match && + mappingOffset >= sizeof(struct lldb_copy_dyld_cache_header_v1)) { + + DataBufferSP dsc_mapping_info_data_sp = + dsc_filespec.MemoryMapFileContentsIfLocal( + mappingOffset, + sizeof(struct lldb_copy_dyld_cache_mapping_info)); + DataExtractor dsc_mapping_info_data(dsc_mapping_info_data_sp, + byte_order, addr_byte_size); + offset = 0; + + // The File addresses (from the in-memory Mach-O load commands) for + // the shared libraries + // in the shared library cache need to be adjusted by an offset to + // match up with the + // dylibOffset identifying field in the + // dyld_cache_local_symbol_entry's. This offset is + // recorded in mapping_offset_value. + const uint64_t mapping_offset_value = + dsc_mapping_info_data.GetU64(&offset); + + offset = offsetof(struct lldb_copy_dyld_cache_header_v1, + localSymbolsOffset); + uint64_t localSymbolsOffset = dsc_header_data.GetU64(&offset); + uint64_t localSymbolsSize = dsc_header_data.GetU64(&offset); + + if (localSymbolsOffset && localSymbolsSize) { + // Map the local symbols + if (DataBufferSP dsc_local_symbols_data_sp = + dsc_filespec.MemoryMapFileContentsIfLocal( + localSymbolsOffset, localSymbolsSize)) { + DataExtractor dsc_local_symbols_data(dsc_local_symbols_data_sp, + byte_order, addr_byte_size); + + offset = 0; + + typedef std::map<ConstString, uint16_t> UndefinedNameToDescMap; + typedef std::map<uint32_t, ConstString> SymbolIndexToName; + UndefinedNameToDescMap undefined_name_to_desc; + SymbolIndexToName reexport_shlib_needs_fixup; + + // Read the local_symbols_infos struct in one shot + struct lldb_copy_dyld_cache_local_symbols_info local_symbols_info; + dsc_local_symbols_data.GetU32(&offset, + &local_symbols_info.nlistOffset, 6); + + SectionSP text_section_sp( + section_list->FindSectionByName(GetSegmentNameTEXT())); + + uint32_t header_file_offset = + (text_section_sp->GetFileAddress() - mapping_offset_value); + + offset = local_symbols_info.entriesOffset; + for (uint32_t entry_index = 0; + entry_index < local_symbols_info.entriesCount; + entry_index++) { + struct lldb_copy_dyld_cache_local_symbols_entry + local_symbols_entry; + local_symbols_entry.dylibOffset = + dsc_local_symbols_data.GetU32(&offset); + local_symbols_entry.nlistStartIndex = + dsc_local_symbols_data.GetU32(&offset); + local_symbols_entry.nlistCount = + dsc_local_symbols_data.GetU32(&offset); + + if (header_file_offset == local_symbols_entry.dylibOffset) { + unmapped_local_symbols_found = local_symbols_entry.nlistCount; + + // The normal nlist code cannot correctly size the Symbols + // array, we need to allocate it here. + sym = symtab->Resize( + symtab_load_command.nsyms + m_dysymtab.nindirectsyms + + unmapped_local_symbols_found - m_dysymtab.nlocalsym); + num_syms = symtab->GetNumSymbols(); + + nlist_data_offset = + local_symbols_info.nlistOffset + + (nlist_byte_size * local_symbols_entry.nlistStartIndex); + uint32_t string_table_offset = + local_symbols_info.stringsOffset; + + for (uint32_t nlist_index = 0; + nlist_index < local_symbols_entry.nlistCount; + nlist_index++) { + ///////////////////////////// { + struct nlist_64 nlist; + if (!dsc_local_symbols_data.ValidOffsetForDataOfSize( + nlist_data_offset, nlist_byte_size)) + break; + + nlist.n_strx = dsc_local_symbols_data.GetU32_unchecked( + &nlist_data_offset); + nlist.n_type = dsc_local_symbols_data.GetU8_unchecked( + &nlist_data_offset); + nlist.n_sect = dsc_local_symbols_data.GetU8_unchecked( + &nlist_data_offset); + nlist.n_desc = dsc_local_symbols_data.GetU16_unchecked( + &nlist_data_offset); + nlist.n_value = + dsc_local_symbols_data.GetAddress_unchecked( + &nlist_data_offset); + + SymbolType type = eSymbolTypeInvalid; + const char *symbol_name = dsc_local_symbols_data.PeekCStr( + string_table_offset + nlist.n_strx); + + if (symbol_name == NULL) { // No symbol should be NULL, even the symbols with no // string values should have an offset zero which points // to an empty C-string - Host::SystemLog (Host::eSystemLogError, - "error: symbol[%u] has invalid string table offset 0x%x in %s, ignoring symbol\n", - nlist_idx, - nlist.n_strx, - module_sp->GetFileSpec().GetPath().c_str()); + Host::SystemLog( + Host::eSystemLogError, + "error: DSC unmapped local symbol[%u] has invalid " + "string table offset 0x%x in %s, ignoring symbol\n", + entry_index, nlist.n_strx, + module_sp->GetFileSpec().GetPath().c_str()); continue; - } - if (symbol_name[0] == '\0') + } + if (symbol_name[0] == '\0') symbol_name = NULL; - } - else - { - const addr_t str_addr = strtab_addr + nlist.n_strx; - Error str_error; - if (process->ReadCStringFromMemory(str_addr, memory_symbol_name, str_error)) - symbol_name = memory_symbol_name.c_str(); - } - const char *symbol_name_non_abi_mangled = NULL; - - SectionSP symbol_section; - lldb::addr_t symbol_byte_size = 0; - bool add_nlist = true; - bool is_gsym = false; - bool is_debug = ((nlist.n_type & N_STAB) != 0); - bool demangled_is_synthesized = false; - bool set_value = true; - assert (sym_idx < num_syms); - - sym[sym_idx].SetDebug (is_debug); - - if (is_debug) - { - switch (nlist.n_type) - { - case N_GSYM: - // global symbol: name,,NO_SECT,type,0 - // Sometimes the N_GSYM value contains the address. - - // FIXME: In the .o files, we have a GSYM and a debug symbol for all the ObjC data. They - // have the same address, but we want to ensure that we always find only the real symbol, - // 'cause we don't currently correctly attribute the GSYM one to the ObjCClass/Ivar/MetaClass - // symbol type. This is a temporary hack to make sure the ObjectiveC symbols get treated - // correctly. To do this right, we should coalesce all the GSYM & global symbols that have the - // same address. - is_gsym = true; - sym[sym_idx].SetExternal(true); - - if (symbol_name && symbol_name[0] == '_' && symbol_name[1] == 'O') - { + + const char *symbol_name_non_abi_mangled = NULL; + + SectionSP symbol_section; + uint32_t symbol_byte_size = 0; + bool add_nlist = true; + bool is_debug = ((nlist.n_type & N_STAB) != 0); + bool demangled_is_synthesized = false; + bool is_gsym = false; + bool set_value = true; + + assert(sym_idx < num_syms); + + sym[sym_idx].SetDebug(is_debug); + + if (is_debug) { + switch (nlist.n_type) { + case N_GSYM: + // global symbol: name,,NO_SECT,type,0 + // Sometimes the N_GSYM value contains the address. + + // FIXME: In the .o files, we have a GSYM and a debug + // symbol for all the ObjC data. They + // have the same address, but we want to ensure that + // we always find only the real symbol, + // 'cause we don't currently correctly attribute the + // GSYM one to the ObjCClass/Ivar/MetaClass + // symbol type. This is a temporary hack to make sure + // the ObjectiveC symbols get treated + // correctly. To do this right, we should coalesce + // all the GSYM & global symbols that have the + // same address. + + is_gsym = true; + sym[sym_idx].SetExternal(true); + + if (symbol_name && symbol_name[0] == '_' && + symbol_name[1] == 'O') { llvm::StringRef symbol_name_ref(symbol_name); - if (symbol_name_ref.startswith(g_objc_v2_prefix_class)) - { - symbol_name_non_abi_mangled = symbol_name + 1; - symbol_name = symbol_name + g_objc_v2_prefix_class.size(); - type = eSymbolTypeObjCClass; - demangled_is_synthesized = true; - - } - else if (symbol_name_ref.startswith(g_objc_v2_prefix_metaclass)) - { - symbol_name_non_abi_mangled = symbol_name + 1; - symbol_name = symbol_name + g_objc_v2_prefix_metaclass.size(); - type = eSymbolTypeObjCMetaClass; - demangled_is_synthesized = true; + if (symbol_name_ref.startswith( + g_objc_v2_prefix_class)) { + symbol_name_non_abi_mangled = symbol_name + 1; + symbol_name = + symbol_name + g_objc_v2_prefix_class.size(); + type = eSymbolTypeObjCClass; + demangled_is_synthesized = true; + + } else if (symbol_name_ref.startswith( + g_objc_v2_prefix_metaclass)) { + symbol_name_non_abi_mangled = symbol_name + 1; + symbol_name = symbol_name + + g_objc_v2_prefix_metaclass.size(); + type = eSymbolTypeObjCMetaClass; + demangled_is_synthesized = true; + } else if (symbol_name_ref.startswith( + g_objc_v2_prefix_ivar)) { + symbol_name_non_abi_mangled = symbol_name + 1; + symbol_name = + symbol_name + g_objc_v2_prefix_ivar.size(); + type = eSymbolTypeObjCIVar; + demangled_is_synthesized = true; } - else if (symbol_name_ref.startswith(g_objc_v2_prefix_ivar)) - { - symbol_name_non_abi_mangled = symbol_name + 1; - symbol_name = symbol_name + g_objc_v2_prefix_ivar.size(); - type = eSymbolTypeObjCIVar; - demangled_is_synthesized = true; - } - } - else - { + } else { if (nlist.n_value != 0) - symbol_section = section_info.GetSection (nlist.n_sect, nlist.n_value); + symbol_section = section_info.GetSection( + nlist.n_sect, nlist.n_value); type = eSymbolTypeData; - } - break; + } + break; - case N_FNAME: - // procedure name (f77 kludge): name,,NO_SECT,0,0 - type = eSymbolTypeCompiler; - break; + case N_FNAME: + // procedure name (f77 kludge): name,,NO_SECT,0,0 + type = eSymbolTypeCompiler; + break; - case N_FUN: - // procedure: name,,n_sect,linenumber,address - if (symbol_name) - { + case N_FUN: + // procedure: name,,n_sect,linenumber,address + if (symbol_name) { type = eSymbolTypeCode; - symbol_section = section_info.GetSection (nlist.n_sect, nlist.n_value); - - N_FUN_addr_to_sym_idx.insert(std::make_pair(nlist.n_value, sym_idx)); - // We use the current number of symbols in the symbol table in lieu of - // using nlist_idx in case we ever start trimming entries out + symbol_section = section_info.GetSection( + nlist.n_sect, nlist.n_value); + + N_FUN_addr_to_sym_idx.insert( + std::make_pair(nlist.n_value, sym_idx)); + // We use the current number of symbols in the + // symbol table in lieu of + // using nlist_idx in case we ever start trimming + // entries out N_FUN_indexes.push_back(sym_idx); - } - else - { + } else { type = eSymbolTypeCompiler; - if ( !N_FUN_indexes.empty() ) - { - // Copy the size of the function into the original STAB entry so we don't have - // to hunt for it later - symtab->SymbolAtIndex(N_FUN_indexes.back())->SetByteSize(nlist.n_value); - N_FUN_indexes.pop_back(); - // We don't really need the end function STAB as it contains the size which - // we already placed with the original symbol, so don't add it if we want a - // minimal symbol table - add_nlist = false; + if (!N_FUN_indexes.empty()) { + // Copy the size of the function into the original + // STAB entry so we don't have + // to hunt for it later + symtab->SymbolAtIndex(N_FUN_indexes.back()) + ->SetByteSize(nlist.n_value); + N_FUN_indexes.pop_back(); + // We don't really need the end function STAB as + // it contains the size which + // we already placed with the original symbol, so + // don't add it if we want a + // minimal symbol table + add_nlist = false; } - } - break; - - case N_STSYM: - // static symbol: name,,n_sect,type,address - N_STSYM_addr_to_sym_idx.insert(std::make_pair(nlist.n_value, sym_idx)); - symbol_section = section_info.GetSection (nlist.n_sect, nlist.n_value); - if (symbol_name && symbol_name[0]) - { - type = ObjectFile::GetSymbolTypeFromName(symbol_name+1, eSymbolTypeData); - } - break; - - case N_LCSYM: - // .lcomm symbol: name,,n_sect,type,address - symbol_section = section_info.GetSection (nlist.n_sect, nlist.n_value); - type = eSymbolTypeCommonBlock; - break; - - case N_BNSYM: - // We use the current number of symbols in the symbol table in lieu of - // using nlist_idx in case we ever start trimming entries out - // Skip these if we want minimal symbol tables - add_nlist = false; - break; - - case N_ENSYM: - // Set the size of the N_BNSYM to the terminating index of this N_ENSYM - // so that we can always skip the entire symbol if we need to navigate - // more quickly at the source level when parsing STABS - // Skip these if we want minimal symbol tables - add_nlist = false; - break; - - - case N_OPT: - // emitted with gcc2_compiled and in gcc source - type = eSymbolTypeCompiler; - break; - - case N_RSYM: - // register sym: name,,NO_SECT,type,register - type = eSymbolTypeVariable; - break; - - case N_SLINE: - // src line: 0,,n_sect,linenumber,address - symbol_section = section_info.GetSection (nlist.n_sect, nlist.n_value); - type = eSymbolTypeLineEntry; - break; - - case N_SSYM: - // structure elt: name,,NO_SECT,type,struct_offset - type = eSymbolTypeVariableType; - break; - - case N_SO: - // source file name - type = eSymbolTypeSourceFile; - if (symbol_name == NULL) - { + } + break; + + case N_STSYM: + // static symbol: name,,n_sect,type,address + N_STSYM_addr_to_sym_idx.insert( + std::make_pair(nlist.n_value, sym_idx)); + symbol_section = section_info.GetSection( + nlist.n_sect, nlist.n_value); + if (symbol_name && symbol_name[0]) { + type = ObjectFile::GetSymbolTypeFromName( + symbol_name + 1, eSymbolTypeData); + } + break; + + case N_LCSYM: + // .lcomm symbol: name,,n_sect,type,address + symbol_section = section_info.GetSection( + nlist.n_sect, nlist.n_value); + type = eSymbolTypeCommonBlock; + break; + + case N_BNSYM: + // We use the current number of symbols in the symbol + // table in lieu of + // using nlist_idx in case we ever start trimming + // entries out + // Skip these if we want minimal symbol tables + add_nlist = false; + break; + + case N_ENSYM: + // Set the size of the N_BNSYM to the terminating + // index of this N_ENSYM + // so that we can always skip the entire symbol if we + // need to navigate + // more quickly at the source level when parsing STABS + // Skip these if we want minimal symbol tables + add_nlist = false; + break; + + case N_OPT: + // emitted with gcc2_compiled and in gcc source + type = eSymbolTypeCompiler; + break; + + case N_RSYM: + // register sym: name,,NO_SECT,type,register + type = eSymbolTypeVariable; + break; + + case N_SLINE: + // src line: 0,,n_sect,linenumber,address + symbol_section = section_info.GetSection( + nlist.n_sect, nlist.n_value); + type = eSymbolTypeLineEntry; + break; + + case N_SSYM: + // structure elt: name,,NO_SECT,type,struct_offset + type = eSymbolTypeVariableType; + break; + + case N_SO: + // source file name + type = eSymbolTypeSourceFile; + if (symbol_name == NULL) { add_nlist = false; - if (N_SO_index != UINT32_MAX) - { - // Set the size of the N_SO to the terminating index of this N_SO - // so that we can always skip the entire N_SO if we need to navigate - // more quickly at the source level when parsing STABS - symbol_ptr = symtab->SymbolAtIndex(N_SO_index); - symbol_ptr->SetByteSize(sym_idx); - symbol_ptr->SetSizeIsSibling(true); + if (N_SO_index != UINT32_MAX) { + // Set the size of the N_SO to the terminating + // index of this N_SO + // so that we can always skip the entire N_SO if + // we need to navigate + // more quickly at the source level when parsing + // STABS + symbol_ptr = symtab->SymbolAtIndex(N_SO_index); + symbol_ptr->SetByteSize(sym_idx); + symbol_ptr->SetSizeIsSibling(true); } N_NSYM_indexes.clear(); N_INCL_indexes.clear(); @@ -3936,2204 +3034,3061 @@ ObjectFileMachO::ParseSymtab () N_COMM_indexes.clear(); N_FUN_indexes.clear(); N_SO_index = UINT32_MAX; - } - else - { - // We use the current number of symbols in the symbol table in lieu of - // using nlist_idx in case we ever start trimming entries out - const bool N_SO_has_full_path = symbol_name[0] == '/'; - if (N_SO_has_full_path) - { - if ((N_SO_index == sym_idx - 1) && ((sym_idx - 1) < num_syms)) - { - // We have two consecutive N_SO entries where the first contains a directory - // and the second contains a full path. - sym[sym_idx - 1].GetMangled().SetValue(ConstString(symbol_name), false); - m_nlist_idx_to_sym_idx[nlist_idx] = sym_idx - 1; - add_nlist = false; - } - else - { - // This is the first entry in a N_SO that contains a directory or - // a full path to the source file - N_SO_index = sym_idx; - } - } - else if ((N_SO_index == sym_idx - 1) && ((sym_idx - 1) < num_syms)) - { - // This is usually the second N_SO entry that contains just the filename, - // so here we combine it with the first one if we are minimizing the symbol table - const char *so_path = sym[sym_idx - 1].GetMangled().GetDemangledName(lldb::eLanguageTypeUnknown).AsCString(); - if (so_path && so_path[0]) - { - std::string full_so_path (so_path); - const size_t double_slash_pos = full_so_path.find("//"); - if (double_slash_pos != std::string::npos) - { - // The linker has been generating bad N_SO entries with doubled up paths - // in the format "%s%s" where the first string in the DW_AT_comp_dir, - // and the second is the directory for the source file so you end up with - // a path that looks like "/tmp/src//tmp/src/" - FileSpec so_dir(so_path, false); - if (!so_dir.Exists()) - { - so_dir.SetFile(&full_so_path[double_slash_pos + 1], false); - if (so_dir.Exists()) - { - // Trim off the incorrect path - full_so_path.erase(0, double_slash_pos + 1); - } - } + } else { + // We use the current number of symbols in the + // symbol table in lieu of + // using nlist_idx in case we ever start trimming + // entries out + const bool N_SO_has_full_path = + symbol_name[0] == '/'; + if (N_SO_has_full_path) { + if ((N_SO_index == sym_idx - 1) && + ((sym_idx - 1) < num_syms)) { + // We have two consecutive N_SO entries where + // the first contains a directory + // and the second contains a full path. + sym[sym_idx - 1].GetMangled().SetValue( + ConstString(symbol_name), false); + m_nlist_idx_to_sym_idx[nlist_idx] = sym_idx - 1; + add_nlist = false; + } else { + // This is the first entry in a N_SO that + // contains a directory or + // a full path to the source file + N_SO_index = sym_idx; + } + } else if ((N_SO_index == sym_idx - 1) && + ((sym_idx - 1) < num_syms)) { + // This is usually the second N_SO entry that + // contains just the filename, + // so here we combine it with the first one if we + // are minimizing the symbol table + const char *so_path = + sym[sym_idx - 1] + .GetMangled() + .GetDemangledName( + lldb::eLanguageTypeUnknown) + .AsCString(); + if (so_path && so_path[0]) { + std::string full_so_path(so_path); + const size_t double_slash_pos = + full_so_path.find("//"); + if (double_slash_pos != std::string::npos) { + // The linker has been generating bad N_SO + // entries with doubled up paths + // in the format "%s%s" where the first string + // in the DW_AT_comp_dir, + // and the second is the directory for the + // source file so you end up with + // a path that looks like "/tmp/src//tmp/src/" + FileSpec so_dir(so_path, false); + if (!so_dir.Exists()) { + so_dir.SetFile( + &full_so_path[double_slash_pos + 1], + false); + if (so_dir.Exists()) { + // Trim off the incorrect path + full_so_path.erase(0, + double_slash_pos + 1); } - if (*full_so_path.rbegin() != '/') - full_so_path += '/'; - full_so_path += symbol_name; - sym[sym_idx - 1].GetMangled().SetValue(ConstString(full_so_path.c_str()), false); - add_nlist = false; - m_nlist_idx_to_sym_idx[nlist_idx] = sym_idx - 1; + } } + if (*full_so_path.rbegin() != '/') + full_so_path += '/'; + full_so_path += symbol_name; + sym[sym_idx - 1].GetMangled().SetValue( + ConstString(full_so_path.c_str()), false); + add_nlist = false; + m_nlist_idx_to_sym_idx[nlist_idx] = sym_idx - 1; + } + } else { + // This could be a relative path to a N_SO + N_SO_index = sym_idx; } - else - { - // This could be a relative path to a N_SO - N_SO_index = sym_idx; - } - } - break; - - case N_OSO: - // object file name: name,,0,0,st_mtime - type = eSymbolTypeObjectFile; - break; - - case N_LSYM: - // local sym: name,,NO_SECT,type,offset - type = eSymbolTypeLocal; - break; - - //---------------------------------------------------------------------- - // INCL scopes - //---------------------------------------------------------------------- - case N_BINCL: - // include file beginning: name,,NO_SECT,0,sum - // We use the current number of symbols in the symbol table in lieu of - // using nlist_idx in case we ever start trimming entries out - N_INCL_indexes.push_back(sym_idx); - type = eSymbolTypeScopeBegin; - break; - - case N_EINCL: - // include file end: name,,NO_SECT,0,0 - // Set the size of the N_BINCL to the terminating index of this N_EINCL - // so that we can always skip the entire symbol if we need to navigate - // more quickly at the source level when parsing STABS - if ( !N_INCL_indexes.empty() ) - { - symbol_ptr = symtab->SymbolAtIndex(N_INCL_indexes.back()); + } + break; + + case N_OSO: + // object file name: name,,0,0,st_mtime + type = eSymbolTypeObjectFile; + break; + + case N_LSYM: + // local sym: name,,NO_SECT,type,offset + type = eSymbolTypeLocal; + break; + + //---------------------------------------------------------------------- + // INCL scopes + //---------------------------------------------------------------------- + case N_BINCL: + // include file beginning: name,,NO_SECT,0,sum + // We use the current number of symbols in the symbol + // table in lieu of + // using nlist_idx in case we ever start trimming + // entries out + N_INCL_indexes.push_back(sym_idx); + type = eSymbolTypeScopeBegin; + break; + + case N_EINCL: + // include file end: name,,NO_SECT,0,0 + // Set the size of the N_BINCL to the terminating + // index of this N_EINCL + // so that we can always skip the entire symbol if we + // need to navigate + // more quickly at the source level when parsing STABS + if (!N_INCL_indexes.empty()) { + symbol_ptr = + symtab->SymbolAtIndex(N_INCL_indexes.back()); symbol_ptr->SetByteSize(sym_idx + 1); symbol_ptr->SetSizeIsSibling(true); N_INCL_indexes.pop_back(); - } - type = eSymbolTypeScopeEnd; - break; - - case N_SOL: - // #included file name: name,,n_sect,0,address - type = eSymbolTypeHeaderFile; - - // We currently don't use the header files on darwin - add_nlist = false; - break; - - case N_PARAMS: - // compiler parameters: name,,NO_SECT,0,0 - type = eSymbolTypeCompiler; - break; - - case N_VERSION: - // compiler version: name,,NO_SECT,0,0 - type = eSymbolTypeCompiler; - break; - - case N_OLEVEL: - // compiler -O level: name,,NO_SECT,0,0 - type = eSymbolTypeCompiler; - break; - - case N_PSYM: - // parameter: name,,NO_SECT,type,offset - type = eSymbolTypeVariable; - break; - - case N_ENTRY: - // alternate entry: name,,n_sect,linenumber,address - symbol_section = section_info.GetSection (nlist.n_sect, nlist.n_value); - type = eSymbolTypeLineEntry; - break; - - //---------------------------------------------------------------------- - // Left and Right Braces - //---------------------------------------------------------------------- - case N_LBRAC: - // left bracket: 0,,NO_SECT,nesting level,address - // We use the current number of symbols in the symbol table in lieu of - // using nlist_idx in case we ever start trimming entries out - symbol_section = section_info.GetSection (nlist.n_sect, nlist.n_value); - N_BRAC_indexes.push_back(sym_idx); - type = eSymbolTypeScopeBegin; - break; - - case N_RBRAC: - // right bracket: 0,,NO_SECT,nesting level,address - // Set the size of the N_LBRAC to the terminating index of this N_RBRAC - // so that we can always skip the entire symbol if we need to navigate - // more quickly at the source level when parsing STABS - symbol_section = section_info.GetSection (nlist.n_sect, nlist.n_value); - if ( !N_BRAC_indexes.empty() ) - { - symbol_ptr = symtab->SymbolAtIndex(N_BRAC_indexes.back()); + } + type = eSymbolTypeScopeEnd; + break; + + case N_SOL: + // #included file name: name,,n_sect,0,address + type = eSymbolTypeHeaderFile; + + // We currently don't use the header files on darwin + add_nlist = false; + break; + + case N_PARAMS: + // compiler parameters: name,,NO_SECT,0,0 + type = eSymbolTypeCompiler; + break; + + case N_VERSION: + // compiler version: name,,NO_SECT,0,0 + type = eSymbolTypeCompiler; + break; + + case N_OLEVEL: + // compiler -O level: name,,NO_SECT,0,0 + type = eSymbolTypeCompiler; + break; + + case N_PSYM: + // parameter: name,,NO_SECT,type,offset + type = eSymbolTypeVariable; + break; + + case N_ENTRY: + // alternate entry: name,,n_sect,linenumber,address + symbol_section = section_info.GetSection( + nlist.n_sect, nlist.n_value); + type = eSymbolTypeLineEntry; + break; + + //---------------------------------------------------------------------- + // Left and Right Braces + //---------------------------------------------------------------------- + case N_LBRAC: + // left bracket: 0,,NO_SECT,nesting level,address + // We use the current number of symbols in the symbol + // table in lieu of + // using nlist_idx in case we ever start trimming + // entries out + symbol_section = section_info.GetSection( + nlist.n_sect, nlist.n_value); + N_BRAC_indexes.push_back(sym_idx); + type = eSymbolTypeScopeBegin; + break; + + case N_RBRAC: + // right bracket: 0,,NO_SECT,nesting level,address + // Set the size of the N_LBRAC to the terminating + // index of this N_RBRAC + // so that we can always skip the entire symbol if we + // need to navigate + // more quickly at the source level when parsing STABS + symbol_section = section_info.GetSection( + nlist.n_sect, nlist.n_value); + if (!N_BRAC_indexes.empty()) { + symbol_ptr = + symtab->SymbolAtIndex(N_BRAC_indexes.back()); symbol_ptr->SetByteSize(sym_idx + 1); symbol_ptr->SetSizeIsSibling(true); N_BRAC_indexes.pop_back(); - } - type = eSymbolTypeScopeEnd; - break; - - case N_EXCL: - // deleted include file: name,,NO_SECT,0,sum - type = eSymbolTypeHeaderFile; - break; - - //---------------------------------------------------------------------- - // COMM scopes - //---------------------------------------------------------------------- - case N_BCOMM: - // begin common: name,,NO_SECT,0,0 - // We use the current number of symbols in the symbol table in lieu of - // using nlist_idx in case we ever start trimming entries out - type = eSymbolTypeScopeBegin; - N_COMM_indexes.push_back(sym_idx); - break; - - case N_ECOML: - // end common (local name): 0,,n_sect,0,address - symbol_section = section_info.GetSection (nlist.n_sect, nlist.n_value); - LLVM_FALLTHROUGH; - - case N_ECOMM: - // end common: name,,n_sect,0,0 - // Set the size of the N_BCOMM to the terminating index of this N_ECOMM/N_ECOML - // so that we can always skip the entire symbol if we need to navigate - // more quickly at the source level when parsing STABS - if ( !N_COMM_indexes.empty() ) - { - symbol_ptr = symtab->SymbolAtIndex(N_COMM_indexes.back()); + } + type = eSymbolTypeScopeEnd; + break; + + case N_EXCL: + // deleted include file: name,,NO_SECT,0,sum + type = eSymbolTypeHeaderFile; + break; + + //---------------------------------------------------------------------- + // COMM scopes + //---------------------------------------------------------------------- + case N_BCOMM: + // begin common: name,,NO_SECT,0,0 + // We use the current number of symbols in the symbol + // table in lieu of + // using nlist_idx in case we ever start trimming + // entries out + type = eSymbolTypeScopeBegin; + N_COMM_indexes.push_back(sym_idx); + break; + + case N_ECOML: + // end common (local name): 0,,n_sect,0,address + symbol_section = section_info.GetSection( + nlist.n_sect, nlist.n_value); + // Fall through + + case N_ECOMM: + // end common: name,,n_sect,0,0 + // Set the size of the N_BCOMM to the terminating + // index of this N_ECOMM/N_ECOML + // so that we can always skip the entire symbol if we + // need to navigate + // more quickly at the source level when parsing STABS + if (!N_COMM_indexes.empty()) { + symbol_ptr = + symtab->SymbolAtIndex(N_COMM_indexes.back()); symbol_ptr->SetByteSize(sym_idx + 1); symbol_ptr->SetSizeIsSibling(true); N_COMM_indexes.pop_back(); - } - type = eSymbolTypeScopeEnd; - break; - - case N_LENG: - // second stab entry with length information - type = eSymbolTypeAdditional; - break; + } + type = eSymbolTypeScopeEnd; + break; - default: break; - } - } - else - { - //uint8_t n_pext = N_PEXT & nlist.n_type; - uint8_t n_type = N_TYPE & nlist.n_type; - sym[sym_idx].SetExternal((N_EXT & nlist.n_type) != 0); + case N_LENG: + // second stab entry with length information + type = eSymbolTypeAdditional; + break; - switch (n_type) - { - case N_INDR: - { - const char *reexport_name_cstr = strtab_data.PeekCStr(nlist.n_value); - if (reexport_name_cstr && reexport_name_cstr[0]) - { - type = eSymbolTypeReExported; - ConstString reexport_name(reexport_name_cstr + ((reexport_name_cstr[0] == '_') ? 1 : 0)); - sym[sym_idx].SetReExportedSymbolName(reexport_name); - set_value = false; - reexport_shlib_needs_fixup[sym_idx] = reexport_name; - indirect_symbol_names.insert(ConstString(symbol_name + ((symbol_name[0] == '_') ? 1 : 0))); - } - else - type = eSymbolTypeUndefined; - } - break; - - case N_UNDF: - if (symbol_name && symbol_name[0]) - { - ConstString undefined_name(symbol_name + ((symbol_name[0] == '_') ? 1 : 0)); - undefined_name_to_desc[undefined_name] = nlist.n_desc; + default: + break; } - LLVM_FALLTHROUGH; - - case N_PBUD: - type = eSymbolTypeUndefined; - break; - - case N_ABS: - type = eSymbolTypeAbsolute; - break; - - case N_SECT: - { - symbol_section = section_info.GetSection (nlist.n_sect, nlist.n_value); - - if (!symbol_section) - { - // TODO: warn about this? - add_nlist = false; + } else { + // uint8_t n_pext = N_PEXT & nlist.n_type; + uint8_t n_type = N_TYPE & nlist.n_type; + sym[sym_idx].SetExternal((N_EXT & nlist.n_type) != 0); + + switch (n_type) { + case N_INDR: { + const char *reexport_name_cstr = + strtab_data.PeekCStr(nlist.n_value); + if (reexport_name_cstr && reexport_name_cstr[0]) { + type = eSymbolTypeReExported; + ConstString reexport_name( + reexport_name_cstr + + ((reexport_name_cstr[0] == '_') ? 1 : 0)); + sym[sym_idx].SetReExportedSymbolName(reexport_name); + set_value = false; + reexport_shlib_needs_fixup[sym_idx] = reexport_name; + indirect_symbol_names.insert( + ConstString(symbol_name + + ((symbol_name[0] == '_') ? 1 : 0))); + } else + type = eSymbolTypeUndefined; + } break; + + case N_UNDF: + if (symbol_name && symbol_name[0]) { + ConstString undefined_name( + symbol_name + + ((symbol_name[0] == '_') ? 1 : 0)); + undefined_name_to_desc[undefined_name] = + nlist.n_desc; + } + // Fall through + case N_PBUD: + type = eSymbolTypeUndefined; + break; + + case N_ABS: + type = eSymbolTypeAbsolute; + break; + + case N_SECT: { + symbol_section = section_info.GetSection( + nlist.n_sect, nlist.n_value); + + if (symbol_section == NULL) { + // TODO: warn about this? + add_nlist = false; + break; + } + + if (TEXT_eh_frame_sectID == nlist.n_sect) { + type = eSymbolTypeException; + } else { + uint32_t section_type = + symbol_section->Get() & SECTION_TYPE; + + switch (section_type) { + case S_CSTRING_LITERALS: + type = eSymbolTypeData; + break; // section with only literal C strings + case S_4BYTE_LITERALS: + type = eSymbolTypeData; + break; // section with only 4 byte literals + case S_8BYTE_LITERALS: + type = eSymbolTypeData; + break; // section with only 8 byte literals + case S_LITERAL_POINTERS: + type = eSymbolTypeTrampoline; + break; // section with only pointers to literals + case S_NON_LAZY_SYMBOL_POINTERS: + type = eSymbolTypeTrampoline; + break; // section with only non-lazy symbol + // pointers + case S_LAZY_SYMBOL_POINTERS: + type = eSymbolTypeTrampoline; + break; // section with only lazy symbol pointers + case S_SYMBOL_STUBS: + type = eSymbolTypeTrampoline; + break; // section with only symbol stubs, byte + // size of stub in the reserved2 field + case S_MOD_INIT_FUNC_POINTERS: + type = eSymbolTypeCode; + break; // section with only function pointers for + // initialization + case S_MOD_TERM_FUNC_POINTERS: + type = eSymbolTypeCode; + break; // section with only function pointers for + // termination + case S_INTERPOSING: + type = eSymbolTypeTrampoline; + break; // section with only pairs of function + // pointers for interposing + case S_16BYTE_LITERALS: + type = eSymbolTypeData; + break; // section with only 16 byte literals + case S_DTRACE_DOF: + type = eSymbolTypeInstrumentation; + break; + case S_LAZY_DYLIB_SYMBOL_POINTERS: + type = eSymbolTypeTrampoline; + break; + default: + switch (symbol_section->GetType()) { + case lldb::eSectionTypeCode: + type = eSymbolTypeCode; + break; + case eSectionTypeData: + case eSectionTypeDataCString: // Inlined C string + // data + case eSectionTypeDataCStringPointers: // Pointers + // to C + // string + // data + case eSectionTypeDataSymbolAddress: // Address of + // a symbol in + // the symbol + // table + case eSectionTypeData4: + case eSectionTypeData8: + case eSectionTypeData16: + type = eSymbolTypeData; + break; + default: break; + } + break; } - if (TEXT_eh_frame_sectID == nlist.n_sect) - { - type = eSymbolTypeException; - } - else - { - uint32_t section_type = symbol_section->Get() & SECTION_TYPE; - - switch (section_type) - { - case S_CSTRING_LITERALS: type = eSymbolTypeData; break; // section with only literal C strings - case S_4BYTE_LITERALS: type = eSymbolTypeData; break; // section with only 4 byte literals - case S_8BYTE_LITERALS: type = eSymbolTypeData; break; // section with only 8 byte literals - case S_LITERAL_POINTERS: type = eSymbolTypeTrampoline; break; // section with only pointers to literals - case S_NON_LAZY_SYMBOL_POINTERS: type = eSymbolTypeTrampoline; break; // section with only non-lazy symbol pointers - case S_LAZY_SYMBOL_POINTERS: type = eSymbolTypeTrampoline; break; // section with only lazy symbol pointers - case S_SYMBOL_STUBS: type = eSymbolTypeTrampoline; break; // section with only symbol stubs, byte size of stub in the reserved2 field - case S_MOD_INIT_FUNC_POINTERS: type = eSymbolTypeCode; break; // section with only function pointers for initialization - case S_MOD_TERM_FUNC_POINTERS: type = eSymbolTypeCode; break; // section with only function pointers for termination - case S_INTERPOSING: type = eSymbolTypeTrampoline; break; // section with only pairs of function pointers for interposing - case S_16BYTE_LITERALS: type = eSymbolTypeData; break; // section with only 16 byte literals - case S_DTRACE_DOF: type = eSymbolTypeInstrumentation; break; - case S_LAZY_DYLIB_SYMBOL_POINTERS: type = eSymbolTypeTrampoline; break; - default: - switch (symbol_section->GetType()) - { - case lldb::eSectionTypeCode: - type = eSymbolTypeCode; - break; - case eSectionTypeData: - case eSectionTypeDataCString: // Inlined C string data - case eSectionTypeDataCStringPointers: // Pointers to C string data - case eSectionTypeDataSymbolAddress: // Address of a symbol in the symbol table - case eSectionTypeData4: - case eSectionTypeData8: - case eSectionTypeData16: - type = eSymbolTypeData; - break; - default: - break; + if (type == eSymbolTypeInvalid) { + const char *symbol_sect_name = + symbol_section->GetName().AsCString(); + if (symbol_section->IsDescendant( + text_section_sp.get())) { + if (symbol_section->IsClear( + S_ATTR_PURE_INSTRUCTIONS | + S_ATTR_SELF_MODIFYING_CODE | + S_ATTR_SOME_INSTRUCTIONS)) + type = eSymbolTypeData; + else + type = eSymbolTypeCode; + } else if (symbol_section->IsDescendant( + data_section_sp.get()) || + symbol_section->IsDescendant( + data_dirty_section_sp.get()) || + symbol_section->IsDescendant( + data_const_section_sp.get())) { + if (symbol_sect_name && + ::strstr(symbol_sect_name, "__objc") == + symbol_sect_name) { + type = eSymbolTypeRuntime; + + if (symbol_name) { + llvm::StringRef symbol_name_ref( + symbol_name); + if (symbol_name_ref.startswith("_OBJC_")) { + static const llvm::StringRef + g_objc_v2_prefix_class( + "_OBJC_CLASS_$_"); + static const llvm::StringRef + g_objc_v2_prefix_metaclass( + "_OBJC_METACLASS_$_"); + static const llvm::StringRef + g_objc_v2_prefix_ivar( + "_OBJC_IVAR_$_"); + if (symbol_name_ref.startswith( + g_objc_v2_prefix_class)) { + symbol_name_non_abi_mangled = + symbol_name + 1; + symbol_name = + symbol_name + + g_objc_v2_prefix_class.size(); + type = eSymbolTypeObjCClass; + demangled_is_synthesized = true; + } else if ( + symbol_name_ref.startswith( + g_objc_v2_prefix_metaclass)) { + symbol_name_non_abi_mangled = + symbol_name + 1; + symbol_name = + symbol_name + + g_objc_v2_prefix_metaclass.size(); + type = eSymbolTypeObjCMetaClass; + demangled_is_synthesized = true; + } else if (symbol_name_ref.startswith( + g_objc_v2_prefix_ivar)) { + symbol_name_non_abi_mangled = + symbol_name + 1; + symbol_name = + symbol_name + + g_objc_v2_prefix_ivar.size(); + type = eSymbolTypeObjCIVar; + demangled_is_synthesized = true; + } } - break; + } + } else if (symbol_sect_name && + ::strstr(symbol_sect_name, + "__gcc_except_tab") == + symbol_sect_name) { + type = eSymbolTypeException; + } else { + type = eSymbolTypeData; } - - if (type == eSymbolTypeInvalid) - { - const char *symbol_sect_name = symbol_section->GetName().AsCString(); - if (symbol_section->IsDescendant (text_section_sp.get())) - { - if (symbol_section->IsClear(S_ATTR_PURE_INSTRUCTIONS | - S_ATTR_SELF_MODIFYING_CODE | - S_ATTR_SOME_INSTRUCTIONS)) - type = eSymbolTypeData; - else - type = eSymbolTypeCode; - } - else - if (symbol_section->IsDescendant(data_section_sp.get()) || - symbol_section->IsDescendant(data_dirty_section_sp.get()) || - symbol_section->IsDescendant(data_const_section_sp.get())) - { - if (symbol_sect_name && ::strstr (symbol_sect_name, "__objc") == symbol_sect_name) - { - type = eSymbolTypeRuntime; - - if (symbol_name) - { - llvm::StringRef symbol_name_ref(symbol_name); - if (symbol_name_ref.startswith("_OBJC_")) - { - static const llvm::StringRef g_objc_v2_prefix_class ("_OBJC_CLASS_$_"); - static const llvm::StringRef g_objc_v2_prefix_metaclass ("_OBJC_METACLASS_$_"); - static const llvm::StringRef g_objc_v2_prefix_ivar ("_OBJC_IVAR_$_"); - if (symbol_name_ref.startswith(g_objc_v2_prefix_class)) - { - symbol_name_non_abi_mangled = symbol_name + 1; - symbol_name = symbol_name + g_objc_v2_prefix_class.size(); - type = eSymbolTypeObjCClass; - demangled_is_synthesized = true; - } - else if (symbol_name_ref.startswith(g_objc_v2_prefix_metaclass)) - { - symbol_name_non_abi_mangled = symbol_name + 1; - symbol_name = symbol_name + g_objc_v2_prefix_metaclass.size(); - type = eSymbolTypeObjCMetaClass; - demangled_is_synthesized = true; - } - else if (symbol_name_ref.startswith(g_objc_v2_prefix_ivar)) - { - symbol_name_non_abi_mangled = symbol_name + 1; - symbol_name = symbol_name + g_objc_v2_prefix_ivar.size(); - type = eSymbolTypeObjCIVar; - demangled_is_synthesized = true; - } - } - } - } - else - if (symbol_sect_name && ::strstr (symbol_sect_name, "__gcc_except_tab") == symbol_sect_name) - { - type = eSymbolTypeException; - } - else - { - type = eSymbolTypeData; - } - } - else - if (symbol_sect_name && ::strstr (symbol_sect_name, "__IMPORT") == symbol_sect_name) - { - type = eSymbolTypeTrampoline; - } - else - if (symbol_section->IsDescendant(objc_section_sp.get())) - { - type = eSymbolTypeRuntime; - if (symbol_name && symbol_name[0] == '.') - { - llvm::StringRef symbol_name_ref(symbol_name); - static const llvm::StringRef g_objc_v1_prefix_class (".objc_class_name_"); - if (symbol_name_ref.startswith(g_objc_v1_prefix_class)) - { - symbol_name_non_abi_mangled = symbol_name; - symbol_name = symbol_name + g_objc_v1_prefix_class.size(); - type = eSymbolTypeObjCClass; - demangled_is_synthesized = true; - } - } - } + } else if (symbol_sect_name && + ::strstr(symbol_sect_name, + "__IMPORT") == + symbol_sect_name) { + type = eSymbolTypeTrampoline; + } else if (symbol_section->IsDescendant( + objc_section_sp.get())) { + type = eSymbolTypeRuntime; + if (symbol_name && symbol_name[0] == '.') { + llvm::StringRef symbol_name_ref(symbol_name); + static const llvm::StringRef + g_objc_v1_prefix_class( + ".objc_class_name_"); + if (symbol_name_ref.startswith( + g_objc_v1_prefix_class)) { + symbol_name_non_abi_mangled = symbol_name; + symbol_name = symbol_name + + g_objc_v1_prefix_class.size(); + type = eSymbolTypeObjCClass; + demangled_is_synthesized = true; + } } + } } + } + } break; } - break; - } - } - - if (add_nlist) - { - uint64_t symbol_value = nlist.n_value; - - if (symbol_name_non_abi_mangled) - { - sym[sym_idx].GetMangled().SetMangledName (ConstString(symbol_name_non_abi_mangled)); - sym[sym_idx].GetMangled().SetDemangledName (ConstString(symbol_name)); - } - else - { - bool symbol_name_is_mangled = false; - - if (symbol_name && symbol_name[0] == '_') - { + } + + if (add_nlist) { + uint64_t symbol_value = nlist.n_value; + if (symbol_name_non_abi_mangled) { + sym[sym_idx].GetMangled().SetMangledName( + ConstString(symbol_name_non_abi_mangled)); + sym[sym_idx].GetMangled().SetDemangledName( + ConstString(symbol_name)); + } else { + bool symbol_name_is_mangled = false; + + if (symbol_name && symbol_name[0] == '_') { symbol_name_is_mangled = symbol_name[1] == '_'; - symbol_name++; // Skip the leading underscore - } + symbol_name++; // Skip the leading underscore + } - if (symbol_name) - { + if (symbol_name) { ConstString const_symbol_name(symbol_name); - sym[sym_idx].GetMangled().SetValue(const_symbol_name, symbol_name_is_mangled); + sym[sym_idx].GetMangled().SetValue( + const_symbol_name, symbol_name_is_mangled); + if (is_gsym && is_debug) { + const char *gsym_name = + sym[sym_idx] + .GetMangled() + .GetName(lldb::eLanguageTypeUnknown, + Mangled::ePreferMangled) + .GetCString(); + if (gsym_name) + N_GSYM_name_to_sym_idx[gsym_name] = sym_idx; + } + } } - } - - if (is_gsym) - { - const char *gsym_name = sym[sym_idx].GetMangled().GetName(lldb::eLanguageTypeUnknown, Mangled::ePreferMangled).GetCString(); - if (gsym_name) - N_GSYM_name_to_sym_idx[gsym_name] = sym_idx; - } - - if (symbol_section) - { - const addr_t section_file_addr = symbol_section->GetFileAddress(); - if (symbol_byte_size == 0 && function_starts_count > 0) - { + if (symbol_section) { + const addr_t section_file_addr = + symbol_section->GetFileAddress(); + if (symbol_byte_size == 0 && + function_starts_count > 0) { addr_t symbol_lookup_file_addr = nlist.n_value; - // Do an exact address match for non-ARM addresses, else get the closest since - // the symbol might be a thumb symbol which has an address with bit zero set - FunctionStarts::Entry *func_start_entry = function_starts.FindEntry (symbol_lookup_file_addr, !is_arm); - if (is_arm && func_start_entry) - { - // Verify that the function start address is the symbol address (ARM) - // or the symbol address + 1 (thumb) - if (func_start_entry->addr != symbol_lookup_file_addr && - func_start_entry->addr != (symbol_lookup_file_addr + 1)) - { - // Not the right entry, NULL it out... - func_start_entry = NULL; - } + // Do an exact address match for non-ARM addresses, + // else get the closest since + // the symbol might be a thumb symbol which has an + // address with bit zero set + FunctionStarts::Entry *func_start_entry = + function_starts.FindEntry( + symbol_lookup_file_addr, !is_arm); + if (is_arm && func_start_entry) { + // Verify that the function start address is the + // symbol address (ARM) + // or the symbol address + 1 (thumb) + if (func_start_entry->addr != + symbol_lookup_file_addr && + func_start_entry->addr != + (symbol_lookup_file_addr + 1)) { + // Not the right entry, NULL it out... + func_start_entry = NULL; + } } - if (func_start_entry) - { - func_start_entry->data = true; - - addr_t symbol_file_addr = func_start_entry->addr; + if (func_start_entry) { + func_start_entry->data = true; + + addr_t symbol_file_addr = func_start_entry->addr; + uint32_t symbol_flags = 0; + if (is_arm) { + if (symbol_file_addr & 1) + symbol_flags = + MACHO_NLIST_ARM_SYMBOL_IS_THUMB; + symbol_file_addr &= THUMB_ADDRESS_BIT_MASK; + } + + const FunctionStarts::Entry + *next_func_start_entry = + function_starts.FindNextEntry( + func_start_entry); + const addr_t section_end_file_addr = + section_file_addr + + symbol_section->GetByteSize(); + if (next_func_start_entry) { + addr_t next_symbol_file_addr = + next_func_start_entry->addr; + // Be sure the clear the Thumb address bit when + // we calculate the size + // from the current and next address if (is_arm) - symbol_file_addr &= THUMB_ADDRESS_BIT_MASK; - - const FunctionStarts::Entry *next_func_start_entry = function_starts.FindNextEntry (func_start_entry); - const addr_t section_end_file_addr = section_file_addr + symbol_section->GetByteSize(); - if (next_func_start_entry) - { - addr_t next_symbol_file_addr = next_func_start_entry->addr; - // Be sure the clear the Thumb address bit when we calculate the size - // from the current and next address - if (is_arm) - next_symbol_file_addr &= THUMB_ADDRESS_BIT_MASK; - symbol_byte_size = std::min<lldb::addr_t>(next_symbol_file_addr - symbol_file_addr, section_end_file_addr - symbol_file_addr); - } - else - { - symbol_byte_size = section_end_file_addr - symbol_file_addr; - } + next_symbol_file_addr &= + THUMB_ADDRESS_BIT_MASK; + symbol_byte_size = std::min<lldb::addr_t>( + next_symbol_file_addr - symbol_file_addr, + section_end_file_addr - symbol_file_addr); + } else { + symbol_byte_size = + section_end_file_addr - symbol_file_addr; + } } + } + symbol_value -= section_file_addr; } - symbol_value -= section_file_addr; - } - if (is_debug == false) - { - if (type == eSymbolTypeCode) - { - // See if we can find a N_FUN entry for any code symbols. - // If we do find a match, and the name matches, then we - // can merge the two into just the function symbol to avoid + if (is_debug == false) { + if (type == eSymbolTypeCode) { + // See if we can find a N_FUN entry for any code + // symbols. + // If we do find a match, and the name matches, then + // we + // can merge the two into just the function symbol + // to avoid // duplicate entries in the symbol table - std::pair<ValueToSymbolIndexMap::const_iterator, ValueToSymbolIndexMap::const_iterator> range; - range = N_FUN_addr_to_sym_idx.equal_range(nlist.n_value); - if (range.first != range.second) - { - bool found_it = false; - for (ValueToSymbolIndexMap::const_iterator pos = range.first; pos != range.second; ++pos) - { - if (sym[sym_idx].GetMangled().GetName(lldb::eLanguageTypeUnknown, Mangled::ePreferMangled) == sym[pos->second].GetMangled().GetName(lldb::eLanguageTypeUnknown, Mangled::ePreferMangled)) - { - m_nlist_idx_to_sym_idx[nlist_idx] = pos->second; - // We just need the flags from the linker symbol, so put these flags - // into the N_FUN flags to avoid duplicate symbols in the symbol table - sym[pos->second].SetExternal(sym[sym_idx].IsExternal()); - sym[pos->second].SetFlags (nlist.n_type << 16 | nlist.n_desc); - if (resolver_addresses.find(nlist.n_value) != resolver_addresses.end()) - sym[pos->second].SetType (eSymbolTypeResolver); - sym[sym_idx].Clear(); - found_it = true; - break; - } + std::pair<ValueToSymbolIndexMap::const_iterator, + ValueToSymbolIndexMap::const_iterator> + range; + range = N_FUN_addr_to_sym_idx.equal_range( + nlist.n_value); + if (range.first != range.second) { + bool found_it = false; + for (ValueToSymbolIndexMap::const_iterator pos = + range.first; + pos != range.second; ++pos) { + if (sym[sym_idx].GetMangled().GetName( + lldb::eLanguageTypeUnknown, + Mangled::ePreferMangled) == + sym[pos->second].GetMangled().GetName( + lldb::eLanguageTypeUnknown, + Mangled::ePreferMangled)) { + m_nlist_idx_to_sym_idx[nlist_idx] = + pos->second; + // We just need the flags from the linker + // symbol, so put these flags + // into the N_FUN flags to avoid duplicate + // symbols in the symbol table + sym[pos->second].SetExternal( + sym[sym_idx].IsExternal()); + sym[pos->second].SetFlags(nlist.n_type << 16 | + nlist.n_desc); + if (resolver_addresses.find(nlist.n_value) != + resolver_addresses.end()) + sym[pos->second].SetType( + eSymbolTypeResolver); + sym[sym_idx].Clear(); + found_it = true; + break; } - if (found_it) - continue; - } - else - { - if (resolver_addresses.find(nlist.n_value) != resolver_addresses.end()) - type = eSymbolTypeResolver; + } + if (found_it) + continue; + } else { + if (resolver_addresses.find(nlist.n_value) != + resolver_addresses.end()) + type = eSymbolTypeResolver; } - } - else if (type == eSymbolTypeData || - type == eSymbolTypeObjCClass || - type == eSymbolTypeObjCMetaClass || - type == eSymbolTypeObjCIVar ) - { - // See if we can find a N_STSYM entry for any data symbols. - // If we do find a match, and the name matches, then we - // can merge the two into just the Static symbol to avoid + } else if (type == eSymbolTypeData || + type == eSymbolTypeObjCClass || + type == eSymbolTypeObjCMetaClass || + type == eSymbolTypeObjCIVar) { + // See if we can find a N_STSYM entry for any data + // symbols. + // If we do find a match, and the name matches, then + // we + // can merge the two into just the Static symbol to + // avoid // duplicate entries in the symbol table - std::pair<ValueToSymbolIndexMap::const_iterator, ValueToSymbolIndexMap::const_iterator> range; - range = N_STSYM_addr_to_sym_idx.equal_range(nlist.n_value); - if (range.first != range.second) - { - bool found_it = false; - for (ValueToSymbolIndexMap::const_iterator pos = range.first; pos != range.second; ++pos) - { - if (sym[sym_idx].GetMangled().GetName(lldb::eLanguageTypeUnknown, Mangled::ePreferMangled) == sym[pos->second].GetMangled().GetName(lldb::eLanguageTypeUnknown, Mangled::ePreferMangled)) - { - m_nlist_idx_to_sym_idx[nlist_idx] = pos->second; - // We just need the flags from the linker symbol, so put these flags - // into the N_STSYM flags to avoid duplicate symbols in the symbol table - sym[pos->second].SetExternal(sym[sym_idx].IsExternal()); - sym[pos->second].SetFlags (nlist.n_type << 16 | nlist.n_desc); - sym[sym_idx].Clear(); - found_it = true; - break; - } + std::pair<ValueToSymbolIndexMap::const_iterator, + ValueToSymbolIndexMap::const_iterator> + range; + range = N_STSYM_addr_to_sym_idx.equal_range( + nlist.n_value); + if (range.first != range.second) { + bool found_it = false; + for (ValueToSymbolIndexMap::const_iterator pos = + range.first; + pos != range.second; ++pos) { + if (sym[sym_idx].GetMangled().GetName( + lldb::eLanguageTypeUnknown, + Mangled::ePreferMangled) == + sym[pos->second].GetMangled().GetName( + lldb::eLanguageTypeUnknown, + Mangled::ePreferMangled)) { + m_nlist_idx_to_sym_idx[nlist_idx] = + pos->second; + // We just need the flags from the linker + // symbol, so put these flags + // into the N_STSYM flags to avoid duplicate + // symbols in the symbol table + sym[pos->second].SetExternal( + sym[sym_idx].IsExternal()); + sym[pos->second].SetFlags(nlist.n_type << 16 | + nlist.n_desc); + sym[sym_idx].Clear(); + found_it = true; + break; } - if (found_it) - continue; - } - else - { - // Combine N_GSYM stab entries with the non stab symbol - const char *gsym_name = sym[sym_idx].GetMangled().GetName(lldb::eLanguageTypeUnknown, Mangled::ePreferMangled).GetCString(); - if (gsym_name) - { - ConstNameToSymbolIndexMap::const_iterator pos = N_GSYM_name_to_sym_idx.find(gsym_name); - if (pos != N_GSYM_name_to_sym_idx.end()) - { - const uint32_t GSYM_sym_idx = pos->second; - m_nlist_idx_to_sym_idx[nlist_idx] = GSYM_sym_idx; - // Copy the address, because often the N_GSYM address has an invalid address of zero - // when the global is a common symbol - sym[GSYM_sym_idx].GetAddressRef().SetSection (symbol_section); - sym[GSYM_sym_idx].GetAddressRef().SetOffset (symbol_value); - // We just need the flags from the linker symbol, so put these flags - // into the N_GSYM flags to avoid duplicate symbols in the symbol table - sym[GSYM_sym_idx].SetFlags (nlist.n_type << 16 | nlist.n_desc); - sym[sym_idx].Clear(); - continue; - } + } + if (found_it) + continue; + } else { + const char *gsym_name = + sym[sym_idx] + .GetMangled() + .GetName(lldb::eLanguageTypeUnknown, + Mangled::ePreferMangled) + .GetCString(); + if (gsym_name) { + // Combine N_GSYM stab entries with the non stab + // symbol + ConstNameToSymbolIndexMap::const_iterator pos = + N_GSYM_name_to_sym_idx.find(gsym_name); + if (pos != N_GSYM_name_to_sym_idx.end()) { + const uint32_t GSYM_sym_idx = pos->second; + m_nlist_idx_to_sym_idx[nlist_idx] = + GSYM_sym_idx; + // Copy the address, because often the N_GSYM + // address has an invalid address of zero + // when the global is a common symbol + sym[GSYM_sym_idx].GetAddressRef().SetSection( + symbol_section); + sym[GSYM_sym_idx].GetAddressRef().SetOffset( + symbol_value); + // We just need the flags from the linker + // symbol, so put these flags + // into the N_GSYM flags to avoid duplicate + // symbols in the symbol table + sym[GSYM_sym_idx].SetFlags( + nlist.n_type << 16 | nlist.n_desc); + sym[sym_idx].Clear(); + continue; } + } } + } } - } - - sym[sym_idx].SetID (nlist_idx); - sym[sym_idx].SetType (type); - if (set_value) - { - sym[sym_idx].GetAddressRef().SetSection (symbol_section); - sym[sym_idx].GetAddressRef().SetOffset (symbol_value); - } - sym[sym_idx].SetFlags (nlist.n_type << 16 | nlist.n_desc); - if (symbol_byte_size > 0) - sym[sym_idx].SetByteSize(symbol_byte_size); + sym[sym_idx].SetID(nlist_idx); + sym[sym_idx].SetType(type); + if (set_value) { + sym[sym_idx].GetAddressRef().SetSection( + symbol_section); + sym[sym_idx].GetAddressRef().SetOffset(symbol_value); + } + sym[sym_idx].SetFlags(nlist.n_type << 16 | + nlist.n_desc); - if (demangled_is_synthesized) - sym[sym_idx].SetDemangledNameIsSynthesized(true); + if (symbol_byte_size > 0) + sym[sym_idx].SetByteSize(symbol_byte_size); - ++sym_idx; - } - else - { - sym[sym_idx].Clear(); + if (demangled_is_synthesized) + sym[sym_idx].SetDemangledNameIsSynthesized(true); + ++sym_idx; + } else { + sym[sym_idx].Clear(); + } + } + ///////////////////////////// + } + break; // No more entries to consider } - } + } - for (const auto &pos :reexport_shlib_needs_fixup) - { + for (const auto &pos : reexport_shlib_needs_fixup) { const auto undef_pos = undefined_name_to_desc.find(pos.second); - if (undef_pos != undefined_name_to_desc.end()) - { - const uint8_t dylib_ordinal = llvm::MachO::GET_LIBRARY_ORDINAL(undef_pos->second); - if (dylib_ordinal > 0 && dylib_ordinal < dylib_files.GetSize()) - sym[pos.first].SetReExportedSymbolSharedLibrary(dylib_files.GetFileSpecAtIndex(dylib_ordinal-1)); + if (undef_pos != undefined_name_to_desc.end()) { + const uint8_t dylib_ordinal = + llvm::MachO::GET_LIBRARY_ORDINAL(undef_pos->second); + if (dylib_ordinal > 0 && + dylib_ordinal < dylib_files.GetSize()) + sym[pos.first].SetReExportedSymbolSharedLibrary( + dylib_files.GetFileSpecAtIndex(dylib_ordinal - 1)); } + } } + } } + } + } - uint32_t synthetic_sym_id = symtab_load_command.nsyms; + // Must reset this in case it was mutated above! + nlist_data_offset = 0; +#endif - if (function_starts_count > 0) - { - uint32_t num_synthetic_function_symbols = 0; - for (i=0; i<function_starts_count; ++i) - { - if (function_starts.GetEntryRef (i).data == false) - ++num_synthetic_function_symbols; + if (nlist_data.GetByteSize() > 0) { + + // If the sym array was not created while parsing the DSC unmapped + // symbols, create it now. + if (sym == NULL) { + sym = symtab->Resize(symtab_load_command.nsyms + + m_dysymtab.nindirectsyms); + num_syms = symtab->GetNumSymbols(); + } + + if (unmapped_local_symbols_found) { + assert(m_dysymtab.ilocalsym == 0); + nlist_data_offset += (m_dysymtab.nlocalsym * nlist_byte_size); + nlist_idx = m_dysymtab.nlocalsym; + } else { + nlist_idx = 0; + } + + typedef std::map<ConstString, uint16_t> UndefinedNameToDescMap; + typedef std::map<uint32_t, ConstString> SymbolIndexToName; + UndefinedNameToDescMap undefined_name_to_desc; + SymbolIndexToName reexport_shlib_needs_fixup; + for (; nlist_idx < symtab_load_command.nsyms; ++nlist_idx) { + struct nlist_64 nlist; + if (!nlist_data.ValidOffsetForDataOfSize(nlist_data_offset, + nlist_byte_size)) + break; + + nlist.n_strx = nlist_data.GetU32_unchecked(&nlist_data_offset); + nlist.n_type = nlist_data.GetU8_unchecked(&nlist_data_offset); + nlist.n_sect = nlist_data.GetU8_unchecked(&nlist_data_offset); + nlist.n_desc = nlist_data.GetU16_unchecked(&nlist_data_offset); + nlist.n_value = nlist_data.GetAddress_unchecked(&nlist_data_offset); + + SymbolType type = eSymbolTypeInvalid; + const char *symbol_name = NULL; + + if (have_strtab_data) { + symbol_name = strtab_data.PeekCStr(nlist.n_strx); + + if (symbol_name == NULL) { + // No symbol should be NULL, even the symbols with no + // string values should have an offset zero which points + // to an empty C-string + Host::SystemLog(Host::eSystemLogError, + "error: symbol[%u] has invalid string table offset " + "0x%x in %s, ignoring symbol\n", + nlist_idx, nlist.n_strx, + module_sp->GetFileSpec().GetPath().c_str()); + continue; + } + if (symbol_name[0] == '\0') + symbol_name = NULL; + } else { + const addr_t str_addr = strtab_addr + nlist.n_strx; + Error str_error; + if (process->ReadCStringFromMemory(str_addr, memory_symbol_name, + str_error)) + symbol_name = memory_symbol_name.c_str(); + } + const char *symbol_name_non_abi_mangled = NULL; + + SectionSP symbol_section; + lldb::addr_t symbol_byte_size = 0; + bool add_nlist = true; + bool is_gsym = false; + bool is_debug = ((nlist.n_type & N_STAB) != 0); + bool demangled_is_synthesized = false; + bool set_value = true; + assert(sym_idx < num_syms); + + sym[sym_idx].SetDebug(is_debug); + + if (is_debug) { + switch (nlist.n_type) { + case N_GSYM: + // global symbol: name,,NO_SECT,type,0 + // Sometimes the N_GSYM value contains the address. + + // FIXME: In the .o files, we have a GSYM and a debug symbol for all + // the ObjC data. They + // have the same address, but we want to ensure that we always find + // only the real symbol, + // 'cause we don't currently correctly attribute the GSYM one to the + // ObjCClass/Ivar/MetaClass + // symbol type. This is a temporary hack to make sure the + // ObjectiveC symbols get treated + // correctly. To do this right, we should coalesce all the GSYM & + // global symbols that have the + // same address. + is_gsym = true; + sym[sym_idx].SetExternal(true); + + if (symbol_name && symbol_name[0] == '_' && symbol_name[1] == 'O') { + llvm::StringRef symbol_name_ref(symbol_name); + if (symbol_name_ref.startswith(g_objc_v2_prefix_class)) { + symbol_name_non_abi_mangled = symbol_name + 1; + symbol_name = symbol_name + g_objc_v2_prefix_class.size(); + type = eSymbolTypeObjCClass; + demangled_is_synthesized = true; + + } else if (symbol_name_ref.startswith( + g_objc_v2_prefix_metaclass)) { + symbol_name_non_abi_mangled = symbol_name + 1; + symbol_name = symbol_name + g_objc_v2_prefix_metaclass.size(); + type = eSymbolTypeObjCMetaClass; + demangled_is_synthesized = true; + } else if (symbol_name_ref.startswith(g_objc_v2_prefix_ivar)) { + symbol_name_non_abi_mangled = symbol_name + 1; + symbol_name = symbol_name + g_objc_v2_prefix_ivar.size(); + type = eSymbolTypeObjCIVar; + demangled_is_synthesized = true; + } + } else { + if (nlist.n_value != 0) + symbol_section = + section_info.GetSection(nlist.n_sect, nlist.n_value); + type = eSymbolTypeData; } + break; - if (num_synthetic_function_symbols > 0) - { - if (num_syms < sym_idx + num_synthetic_function_symbols) - { - num_syms = sym_idx + num_synthetic_function_symbols; - sym = symtab->Resize (num_syms); - } - for (i=0; i<function_starts_count; ++i) - { - const FunctionStarts::Entry *func_start_entry = function_starts.GetEntryAtIndex (i); - if (func_start_entry->data == false) - { - addr_t symbol_file_addr = func_start_entry->addr; - uint32_t symbol_flags = 0; - if (is_arm) - { - if (symbol_file_addr & 1) - symbol_flags = MACHO_NLIST_ARM_SYMBOL_IS_THUMB; - symbol_file_addr &= THUMB_ADDRESS_BIT_MASK; - } - Address symbol_addr; - if (module_sp->ResolveFileAddress (symbol_file_addr, symbol_addr)) - { - SectionSP symbol_section (symbol_addr.GetSection()); - uint32_t symbol_byte_size = 0; - if (symbol_section) - { - const addr_t section_file_addr = symbol_section->GetFileAddress(); - const FunctionStarts::Entry *next_func_start_entry = function_starts.FindNextEntry (func_start_entry); - const addr_t section_end_file_addr = section_file_addr + symbol_section->GetByteSize(); - if (next_func_start_entry) - { - addr_t next_symbol_file_addr = next_func_start_entry->addr; - if (is_arm) - next_symbol_file_addr &= THUMB_ADDRESS_BIT_MASK; - symbol_byte_size = std::min<lldb::addr_t>(next_symbol_file_addr - symbol_file_addr, section_end_file_addr - symbol_file_addr); - } - else - { - symbol_byte_size = section_end_file_addr - symbol_file_addr; - } - sym[sym_idx].SetID (synthetic_sym_id++); - sym[sym_idx].GetMangled().SetDemangledName(GetNextSyntheticSymbolName()); - sym[sym_idx].SetType (eSymbolTypeCode); - sym[sym_idx].SetIsSynthetic (true); - sym[sym_idx].GetAddressRef() = symbol_addr; - if (symbol_flags) - sym[sym_idx].SetFlags (symbol_flags); - if (symbol_byte_size) - sym[sym_idx].SetByteSize (symbol_byte_size); - ++sym_idx; - } - } - } - } + case N_FNAME: + // procedure name (f77 kludge): name,,NO_SECT,0,0 + type = eSymbolTypeCompiler; + break; + + case N_FUN: + // procedure: name,,n_sect,linenumber,address + if (symbol_name) { + type = eSymbolTypeCode; + symbol_section = + section_info.GetSection(nlist.n_sect, nlist.n_value); + + N_FUN_addr_to_sym_idx.insert( + std::make_pair(nlist.n_value, sym_idx)); + // We use the current number of symbols in the symbol table in + // lieu of + // using nlist_idx in case we ever start trimming entries out + N_FUN_indexes.push_back(sym_idx); + } else { + type = eSymbolTypeCompiler; + + if (!N_FUN_indexes.empty()) { + // Copy the size of the function into the original STAB entry so + // we don't have + // to hunt for it later + symtab->SymbolAtIndex(N_FUN_indexes.back()) + ->SetByteSize(nlist.n_value); + N_FUN_indexes.pop_back(); + // We don't really need the end function STAB as it contains the + // size which + // we already placed with the original symbol, so don't add it + // if we want a + // minimal symbol table + add_nlist = false; + } } - } + break; - // Trim our symbols down to just what we ended up with after - // removing any symbols. - if (sym_idx < num_syms) - { - num_syms = sym_idx; - sym = symtab->Resize (num_syms); - } + case N_STSYM: + // static symbol: name,,n_sect,type,address + N_STSYM_addr_to_sym_idx.insert( + std::make_pair(nlist.n_value, sym_idx)); + symbol_section = + section_info.GetSection(nlist.n_sect, nlist.n_value); + if (symbol_name && symbol_name[0]) { + type = ObjectFile::GetSymbolTypeFromName(symbol_name + 1, + eSymbolTypeData); + } + break; - // Now synthesize indirect symbols - if (m_dysymtab.nindirectsyms != 0) - { - if (indirect_symbol_index_data.GetByteSize()) - { - NListIndexToSymbolIndexMap::const_iterator end_index_pos = m_nlist_idx_to_sym_idx.end(); + case N_LCSYM: + // .lcomm symbol: name,,n_sect,type,address + symbol_section = + section_info.GetSection(nlist.n_sect, nlist.n_value); + type = eSymbolTypeCommonBlock; + break; - for (uint32_t sect_idx = 1; sect_idx < m_mach_sections.size(); ++sect_idx) - { - if ((m_mach_sections[sect_idx].flags & SECTION_TYPE) == S_SYMBOL_STUBS) - { - uint32_t symbol_stub_byte_size = m_mach_sections[sect_idx].reserved2; - if (symbol_stub_byte_size == 0) - continue; - - const uint32_t num_symbol_stubs = m_mach_sections[sect_idx].size / symbol_stub_byte_size; - - if (num_symbol_stubs == 0) - continue; - - const uint32_t symbol_stub_index_offset = m_mach_sections[sect_idx].reserved1; - for (uint32_t stub_idx = 0; stub_idx < num_symbol_stubs; ++stub_idx) - { - const uint32_t symbol_stub_index = symbol_stub_index_offset + stub_idx; - const lldb::addr_t symbol_stub_addr = m_mach_sections[sect_idx].addr + (stub_idx * symbol_stub_byte_size); - lldb::offset_t symbol_stub_offset = symbol_stub_index * 4; - if (indirect_symbol_index_data.ValidOffsetForDataOfSize(symbol_stub_offset, 4)) - { - const uint32_t stub_sym_id = indirect_symbol_index_data.GetU32 (&symbol_stub_offset); - if (stub_sym_id & (INDIRECT_SYMBOL_ABS | INDIRECT_SYMBOL_LOCAL)) - continue; - - NListIndexToSymbolIndexMap::const_iterator index_pos = m_nlist_idx_to_sym_idx.find (stub_sym_id); - Symbol *stub_symbol = NULL; - if (index_pos != end_index_pos) - { - // We have a remapping from the original nlist index to - // a current symbol index, so just look this up by index - stub_symbol = symtab->SymbolAtIndex (index_pos->second); - } - else - { - // We need to lookup a symbol using the original nlist - // symbol index since this index is coming from the - // S_SYMBOL_STUBS - stub_symbol = symtab->FindSymbolByID (stub_sym_id); - } + case N_BNSYM: + // We use the current number of symbols in the symbol table in lieu + // of + // using nlist_idx in case we ever start trimming entries out + // Skip these if we want minimal symbol tables + add_nlist = false; + break; - if (stub_symbol) - { - Address so_addr(symbol_stub_addr, section_list); - - if (stub_symbol->GetType() == eSymbolTypeUndefined) - { - // Change the external symbol into a trampoline that makes sense - // These symbols were N_UNDF N_EXT, and are useless to us, so we - // can re-use them so we don't have to make up a synthetic symbol - // for no good reason. - if (resolver_addresses.find(symbol_stub_addr) == resolver_addresses.end()) - stub_symbol->SetType (eSymbolTypeTrampoline); - else - stub_symbol->SetType (eSymbolTypeResolver); - stub_symbol->SetExternal (false); - stub_symbol->GetAddressRef() = so_addr; - stub_symbol->SetByteSize (symbol_stub_byte_size); - } - else - { - // Make a synthetic symbol to describe the trampoline stub - Mangled stub_symbol_mangled_name(stub_symbol->GetMangled()); - if (sym_idx >= num_syms) - { - sym = symtab->Resize (++num_syms); - stub_symbol = NULL; // this pointer no longer valid - } - sym[sym_idx].SetID (synthetic_sym_id++); - sym[sym_idx].GetMangled() = stub_symbol_mangled_name; - if (resolver_addresses.find(symbol_stub_addr) == resolver_addresses.end()) - sym[sym_idx].SetType (eSymbolTypeTrampoline); - else - sym[sym_idx].SetType (eSymbolTypeResolver); - sym[sym_idx].SetIsSynthetic (true); - sym[sym_idx].GetAddressRef() = so_addr; - sym[sym_idx].SetByteSize (symbol_stub_byte_size); - ++sym_idx; - } - } - else - { - if (log) - log->Warning ("symbol stub referencing symbol table symbol %u that isn't in our minimal symbol table, fix this!!!", stub_sym_id); - } - } - } - } - } - } - } + case N_ENSYM: + // Set the size of the N_BNSYM to the terminating index of this + // N_ENSYM + // so that we can always skip the entire symbol if we need to + // navigate + // more quickly at the source level when parsing STABS + // Skip these if we want minimal symbol tables + add_nlist = false; + break; - if (!trie_entries.empty()) - { - for (const auto &e : trie_entries) - { - if (e.entry.import_name) - { - // Only add indirect symbols from the Trie entries if we - // didn't have a N_INDR nlist entry for this already - if (indirect_symbol_names.find(e.entry.name) == indirect_symbol_names.end()) - { - // Make a synthetic symbol to describe re-exported symbol. - if (sym_idx >= num_syms) - sym = symtab->Resize (++num_syms); - sym[sym_idx].SetID (synthetic_sym_id++); - sym[sym_idx].GetMangled() = Mangled(e.entry.name); - sym[sym_idx].SetType (eSymbolTypeReExported); - sym[sym_idx].SetIsSynthetic (true); - sym[sym_idx].SetReExportedSymbolName(e.entry.import_name); - if (e.entry.other > 0 && e.entry.other <= dylib_files.GetSize()) - { - sym[sym_idx].SetReExportedSymbolSharedLibrary(dylib_files.GetFileSpecAtIndex(e.entry.other-1)); - } - ++sym_idx; + case N_OPT: + // emitted with gcc2_compiled and in gcc source + type = eSymbolTypeCompiler; + break; + + case N_RSYM: + // register sym: name,,NO_SECT,type,register + type = eSymbolTypeVariable; + break; + + case N_SLINE: + // src line: 0,,n_sect,linenumber,address + symbol_section = + section_info.GetSection(nlist.n_sect, nlist.n_value); + type = eSymbolTypeLineEntry; + break; + + case N_SSYM: + // structure elt: name,,NO_SECT,type,struct_offset + type = eSymbolTypeVariableType; + break; + + case N_SO: + // source file name + type = eSymbolTypeSourceFile; + if (symbol_name == NULL) { + add_nlist = false; + if (N_SO_index != UINT32_MAX) { + // Set the size of the N_SO to the terminating index of this + // N_SO + // so that we can always skip the entire N_SO if we need to + // navigate + // more quickly at the source level when parsing STABS + symbol_ptr = symtab->SymbolAtIndex(N_SO_index); + symbol_ptr->SetByteSize(sym_idx); + symbol_ptr->SetSizeIsSibling(true); + } + N_NSYM_indexes.clear(); + N_INCL_indexes.clear(); + N_BRAC_indexes.clear(); + N_COMM_indexes.clear(); + N_FUN_indexes.clear(); + N_SO_index = UINT32_MAX; + } else { + // We use the current number of symbols in the symbol table in + // lieu of + // using nlist_idx in case we ever start trimming entries out + const bool N_SO_has_full_path = symbol_name[0] == '/'; + if (N_SO_has_full_path) { + if ((N_SO_index == sym_idx - 1) && ((sym_idx - 1) < num_syms)) { + // We have two consecutive N_SO entries where the first + // contains a directory + // and the second contains a full path. + sym[sym_idx - 1].GetMangled().SetValue( + ConstString(symbol_name), false); + m_nlist_idx_to_sym_idx[nlist_idx] = sym_idx - 1; + add_nlist = false; + } else { + // This is the first entry in a N_SO that contains a directory + // or + // a full path to the source file + N_SO_index = sym_idx; + } + } else if ((N_SO_index == sym_idx - 1) && + ((sym_idx - 1) < num_syms)) { + // This is usually the second N_SO entry that contains just the + // filename, + // so here we combine it with the first one if we are minimizing + // the symbol table + const char *so_path = + sym[sym_idx - 1] + .GetMangled() + .GetDemangledName(lldb::eLanguageTypeUnknown) + .AsCString(); + if (so_path && so_path[0]) { + std::string full_so_path(so_path); + const size_t double_slash_pos = full_so_path.find("//"); + if (double_slash_pos != std::string::npos) { + // The linker has been generating bad N_SO entries with + // doubled up paths + // in the format "%s%s" where the first string in the + // DW_AT_comp_dir, + // and the second is the directory for the source file so + // you end up with + // a path that looks like "/tmp/src//tmp/src/" + FileSpec so_dir(so_path, false); + if (!so_dir.Exists()) { + so_dir.SetFile(&full_so_path[double_slash_pos + 1], + false); + if (so_dir.Exists()) { + // Trim off the incorrect path + full_so_path.erase(0, double_slash_pos + 1); + } } + } + if (*full_so_path.rbegin() != '/') + full_so_path += '/'; + full_so_path += symbol_name; + sym[sym_idx - 1].GetMangled().SetValue( + ConstString(full_so_path.c_str()), false); + add_nlist = false; + m_nlist_idx_to_sym_idx[nlist_idx] = sym_idx - 1; } + } else { + // This could be a relative path to a N_SO + N_SO_index = sym_idx; + } } - } - -// StreamFile s(stdout, false); -// s.Printf ("Symbol table before CalculateSymbolSizes():\n"); -// symtab->Dump(&s, NULL, eSortOrderNone); - // Set symbol byte sizes correctly since mach-o nlist entries don't have sizes - symtab->CalculateSymbolSizes(); + break; -// s.Printf ("Symbol table after CalculateSymbolSizes():\n"); -// symtab->Dump(&s, NULL, eSortOrderNone); + case N_OSO: + // object file name: name,,0,0,st_mtime + type = eSymbolTypeObjectFile; + break; - return symtab->GetNumSymbols(); - } - return 0; -} + case N_LSYM: + // local sym: name,,NO_SECT,type,offset + type = eSymbolTypeLocal; + break; -void -ObjectFileMachO::Dump (Stream *s) -{ - ModuleSP module_sp(GetModule()); - if (module_sp) - { - std::lock_guard<std::recursive_mutex> guard(module_sp->GetMutex()); - s->Printf("%p: ", static_cast<void*>(this)); - s->Indent(); - if (m_header.magic == MH_MAGIC_64 || m_header.magic == MH_CIGAM_64) - s->PutCString("ObjectFileMachO64"); - else - s->PutCString("ObjectFileMachO32"); + //---------------------------------------------------------------------- + // INCL scopes + //---------------------------------------------------------------------- + case N_BINCL: + // include file beginning: name,,NO_SECT,0,sum + // We use the current number of symbols in the symbol table in lieu + // of + // using nlist_idx in case we ever start trimming entries out + N_INCL_indexes.push_back(sym_idx); + type = eSymbolTypeScopeBegin; + break; - ArchSpec header_arch; - GetArchitecture(header_arch); + case N_EINCL: + // include file end: name,,NO_SECT,0,0 + // Set the size of the N_BINCL to the terminating index of this + // N_EINCL + // so that we can always skip the entire symbol if we need to + // navigate + // more quickly at the source level when parsing STABS + if (!N_INCL_indexes.empty()) { + symbol_ptr = symtab->SymbolAtIndex(N_INCL_indexes.back()); + symbol_ptr->SetByteSize(sym_idx + 1); + symbol_ptr->SetSizeIsSibling(true); + N_INCL_indexes.pop_back(); + } + type = eSymbolTypeScopeEnd; + break; - *s << ", file = '" << m_file << "', arch = " << header_arch.GetArchitectureName() << "\n"; + case N_SOL: + // #included file name: name,,n_sect,0,address + type = eSymbolTypeHeaderFile; - SectionList *sections = GetSectionList(); - if (sections) - sections->Dump(s, NULL, true, UINT32_MAX); + // We currently don't use the header files on darwin + add_nlist = false; + break; - if (m_symtab_ap.get()) - m_symtab_ap->Dump(s, NULL, eSortOrderNone); - } -} + case N_PARAMS: + // compiler parameters: name,,NO_SECT,0,0 + type = eSymbolTypeCompiler; + break; -bool -ObjectFileMachO::GetUUID (const llvm::MachO::mach_header &header, - const lldb_private::DataExtractor &data, - lldb::offset_t lc_offset, - lldb_private::UUID& uuid) -{ - uint32_t i; - struct uuid_command load_cmd; + case N_VERSION: + // compiler version: name,,NO_SECT,0,0 + type = eSymbolTypeCompiler; + break; - lldb::offset_t offset = lc_offset; - for (i=0; i<header.ncmds; ++i) - { - const lldb::offset_t cmd_offset = offset; - if (data.GetU32(&offset, &load_cmd, 2) == NULL) + case N_OLEVEL: + // compiler -O level: name,,NO_SECT,0,0 + type = eSymbolTypeCompiler; break; - - if (load_cmd.cmd == LC_UUID) - { - const uint8_t *uuid_bytes = data.PeekData(offset, 16); - - if (uuid_bytes) - { - // OpenCL on Mac OS X uses the same UUID for each of its object files. - // We pretend these object files have no UUID to prevent crashing. - - const uint8_t opencl_uuid[] = { 0x8c, 0x8e, 0xb3, 0x9b, - 0x3b, 0xa8, - 0x4b, 0x16, - 0xb6, 0xa4, - 0x27, 0x63, 0xbb, 0x14, 0xf0, 0x0d }; - - if (!memcmp(uuid_bytes, opencl_uuid, 16)) - return false; - - uuid.SetBytes (uuid_bytes); - return true; - } - return false; - } - offset = cmd_offset + load_cmd.cmdsize; - } - return false; -} -bool -ObjectFileMachO::GetArchitecture (const llvm::MachO::mach_header &header, - const lldb_private::DataExtractor &data, - lldb::offset_t lc_offset, - ArchSpec &arch) -{ - arch.SetArchitecture (eArchTypeMachO, header.cputype, header.cpusubtype); - - if (arch.IsValid()) - { - llvm::Triple &triple = arch.GetTriple(); - - // Set OS to an unspecified unknown or a "*" so it can match any OS - triple.setOS(llvm::Triple::UnknownOS); - triple.setOSName(llvm::StringRef()); - - if (header.filetype == MH_PRELOAD) - { - if (header.cputype == CPU_TYPE_ARM) - { - // If this is a 32-bit arm binary, and it's a standalone binary, - // force the Vendor to Apple so we don't accidentally pick up - // the generic armv7 ABI at runtime. Apple's armv7 ABI always uses - // r7 for the frame pointer register; most other armv7 ABIs use a - // combination of r7 and r11. - triple.setVendor(llvm::Triple::Apple); - } - else - { - // Set vendor to an unspecified unknown or a "*" so it can match any vendor - // This is required for correct behavior of EFI debugging on x86_64 - triple.setVendor(llvm::Triple::UnknownVendor); - triple.setVendorName(llvm::StringRef()); + case N_PSYM: + // parameter: name,,NO_SECT,type,offset + type = eSymbolTypeVariable; + break; + + case N_ENTRY: + // alternate entry: name,,n_sect,linenumber,address + symbol_section = + section_info.GetSection(nlist.n_sect, nlist.n_value); + type = eSymbolTypeLineEntry; + break; + + //---------------------------------------------------------------------- + // Left and Right Braces + //---------------------------------------------------------------------- + case N_LBRAC: + // left bracket: 0,,NO_SECT,nesting level,address + // We use the current number of symbols in the symbol table in lieu + // of + // using nlist_idx in case we ever start trimming entries out + symbol_section = + section_info.GetSection(nlist.n_sect, nlist.n_value); + N_BRAC_indexes.push_back(sym_idx); + type = eSymbolTypeScopeBegin; + break; + + case N_RBRAC: + // right bracket: 0,,NO_SECT,nesting level,address + // Set the size of the N_LBRAC to the terminating index of this + // N_RBRAC + // so that we can always skip the entire symbol if we need to + // navigate + // more quickly at the source level when parsing STABS + symbol_section = + section_info.GetSection(nlist.n_sect, nlist.n_value); + if (!N_BRAC_indexes.empty()) { + symbol_ptr = symtab->SymbolAtIndex(N_BRAC_indexes.back()); + symbol_ptr->SetByteSize(sym_idx + 1); + symbol_ptr->SetSizeIsSibling(true); + N_BRAC_indexes.pop_back(); } - return true; - } - else - { - struct load_command load_cmd; + type = eSymbolTypeScopeEnd; + break; - lldb::offset_t offset = lc_offset; - for (uint32_t i=0; i<header.ncmds; ++i) - { - const lldb::offset_t cmd_offset = offset; - if (data.GetU32(&offset, &load_cmd, 2) == NULL) - break; - - switch (load_cmd.cmd) - { - case llvm::MachO::LC_VERSION_MIN_IPHONEOS: - triple.setOS (llvm::Triple::IOS); - return true; - - case llvm::MachO::LC_VERSION_MIN_MACOSX: - triple.setOS (llvm::Triple::MacOSX); - return true; - - case llvm::MachO::LC_VERSION_MIN_TVOS: - triple.setOS (llvm::Triple::TvOS); - return true; - - case llvm::MachO::LC_VERSION_MIN_WATCHOS: - triple.setOS (llvm::Triple::WatchOS); - return true; + case N_EXCL: + // deleted include file: name,,NO_SECT,0,sum + type = eSymbolTypeHeaderFile; + break; - default: - break; - } + //---------------------------------------------------------------------- + // COMM scopes + //---------------------------------------------------------------------- + case N_BCOMM: + // begin common: name,,NO_SECT,0,0 + // We use the current number of symbols in the symbol table in lieu + // of + // using nlist_idx in case we ever start trimming entries out + type = eSymbolTypeScopeBegin; + N_COMM_indexes.push_back(sym_idx); + break; - offset = cmd_offset + load_cmd.cmdsize; + case N_ECOML: + // end common (local name): 0,,n_sect,0,address + symbol_section = + section_info.GetSection(nlist.n_sect, nlist.n_value); + LLVM_FALLTHROUGH; + + case N_ECOMM: + // end common: name,,n_sect,0,0 + // Set the size of the N_BCOMM to the terminating index of this + // N_ECOMM/N_ECOML + // so that we can always skip the entire symbol if we need to + // navigate + // more quickly at the source level when parsing STABS + if (!N_COMM_indexes.empty()) { + symbol_ptr = symtab->SymbolAtIndex(N_COMM_indexes.back()); + symbol_ptr->SetByteSize(sym_idx + 1); + symbol_ptr->SetSizeIsSibling(true); + N_COMM_indexes.pop_back(); } - - if (header.filetype != MH_KEXT_BUNDLE) - { - // We didn't find a LC_VERSION_MIN load command and this isn't a KEXT - // so lets not say our Vendor is Apple, leave it as an unspecified unknown - triple.setVendor(llvm::Triple::UnknownVendor); - triple.setVendorName(llvm::StringRef()); + type = eSymbolTypeScopeEnd; + break; + + case N_LENG: + // second stab entry with length information + type = eSymbolTypeAdditional; + break; + + default: + break; + } + } else { + // uint8_t n_pext = N_PEXT & nlist.n_type; + uint8_t n_type = N_TYPE & nlist.n_type; + sym[sym_idx].SetExternal((N_EXT & nlist.n_type) != 0); + + switch (n_type) { + case N_INDR: { + const char *reexport_name_cstr = + strtab_data.PeekCStr(nlist.n_value); + if (reexport_name_cstr && reexport_name_cstr[0]) { + type = eSymbolTypeReExported; + ConstString reexport_name( + reexport_name_cstr + + ((reexport_name_cstr[0] == '_') ? 1 : 0)); + sym[sym_idx].SetReExportedSymbolName(reexport_name); + set_value = false; + reexport_shlib_needs_fixup[sym_idx] = reexport_name; + indirect_symbol_names.insert( + ConstString(symbol_name + ((symbol_name[0] == '_') ? 1 : 0))); + } else + type = eSymbolTypeUndefined; + } break; + + case N_UNDF: + if (symbol_name && symbol_name[0]) { + ConstString undefined_name(symbol_name + + ((symbol_name[0] == '_') ? 1 : 0)); + undefined_name_to_desc[undefined_name] = nlist.n_desc; } - } - } - return arch.IsValid(); -} + LLVM_FALLTHROUGH; -bool -ObjectFileMachO::GetUUID (lldb_private::UUID* uuid) -{ - ModuleSP module_sp(GetModule()); - if (module_sp) - { - std::lock_guard<std::recursive_mutex> guard(module_sp->GetMutex()); - lldb::offset_t offset = MachHeaderSizeFromMagic(m_header.magic); - return GetUUID (m_header, m_data, offset, *uuid); - } - return false; -} + case N_PBUD: + type = eSymbolTypeUndefined; + break; -uint32_t -ObjectFileMachO::GetDependentModules (FileSpecList& files) -{ - uint32_t count = 0; - ModuleSP module_sp(GetModule()); - if (module_sp) - { - std::lock_guard<std::recursive_mutex> guard(module_sp->GetMutex()); - struct load_command load_cmd; - lldb::offset_t offset = MachHeaderSizeFromMagic(m_header.magic); - std::vector<std::string> rpath_paths; - std::vector<std::string> rpath_relative_paths; - const bool resolve_path = false; // Don't resolve the dependent file paths since they may not reside on this system - uint32_t i; - for (i=0; i<m_header.ncmds; ++i) - { - const uint32_t cmd_offset = offset; - if (m_data.GetU32(&offset, &load_cmd, 2) == NULL) - break; + case N_ABS: + type = eSymbolTypeAbsolute; + break; - switch (load_cmd.cmd) - { - case LC_RPATH: - case LC_LOAD_DYLIB: - case LC_LOAD_WEAK_DYLIB: - case LC_REEXPORT_DYLIB: - case LC_LOAD_DYLINKER: - case LC_LOADFVMLIB: - case LC_LOAD_UPWARD_DYLIB: - { - uint32_t name_offset = cmd_offset + m_data.GetU32(&offset); - const char *path = m_data.PeekCStr(name_offset); - if (path) - { - if (load_cmd.cmd == LC_RPATH) - rpath_paths.push_back(path); - else - { - if (path[0] == '@') - { - if (strncmp(path, "@rpath", strlen("@rpath")) == 0) - rpath_relative_paths.push_back(path + strlen("@rpath")); - } - else - { - FileSpec file_spec(path, resolve_path); - if (files.AppendIfUnique(file_spec)) - count++; - } + case N_SECT: { + symbol_section = + section_info.GetSection(nlist.n_sect, nlist.n_value); + + if (!symbol_section) { + // TODO: warn about this? + add_nlist = false; + break; + } + + if (TEXT_eh_frame_sectID == nlist.n_sect) { + type = eSymbolTypeException; + } else { + uint32_t section_type = symbol_section->Get() & SECTION_TYPE; + + switch (section_type) { + case S_CSTRING_LITERALS: + type = eSymbolTypeData; + break; // section with only literal C strings + case S_4BYTE_LITERALS: + type = eSymbolTypeData; + break; // section with only 4 byte literals + case S_8BYTE_LITERALS: + type = eSymbolTypeData; + break; // section with only 8 byte literals + case S_LITERAL_POINTERS: + type = eSymbolTypeTrampoline; + break; // section with only pointers to literals + case S_NON_LAZY_SYMBOL_POINTERS: + type = eSymbolTypeTrampoline; + break; // section with only non-lazy symbol pointers + case S_LAZY_SYMBOL_POINTERS: + type = eSymbolTypeTrampoline; + break; // section with only lazy symbol pointers + case S_SYMBOL_STUBS: + type = eSymbolTypeTrampoline; + break; // section with only symbol stubs, byte size of stub in + // the reserved2 field + case S_MOD_INIT_FUNC_POINTERS: + type = eSymbolTypeCode; + break; // section with only function pointers for initialization + case S_MOD_TERM_FUNC_POINTERS: + type = eSymbolTypeCode; + break; // section with only function pointers for termination + case S_INTERPOSING: + type = eSymbolTypeTrampoline; + break; // section with only pairs of function pointers for + // interposing + case S_16BYTE_LITERALS: + type = eSymbolTypeData; + break; // section with only 16 byte literals + case S_DTRACE_DOF: + type = eSymbolTypeInstrumentation; + break; + case S_LAZY_DYLIB_SYMBOL_POINTERS: + type = eSymbolTypeTrampoline; + break; + default: + switch (symbol_section->GetType()) { + case lldb::eSectionTypeCode: + type = eSymbolTypeCode; + break; + case eSectionTypeData: + case eSectionTypeDataCString: // Inlined C string data + case eSectionTypeDataCStringPointers: // Pointers to C string + // data + case eSectionTypeDataSymbolAddress: // Address of a symbol in + // the symbol table + case eSectionTypeData4: + case eSectionTypeData8: + case eSectionTypeData16: + type = eSymbolTypeData; + break; + default: + break; + } + break; + } + + if (type == eSymbolTypeInvalid) { + const char *symbol_sect_name = + symbol_section->GetName().AsCString(); + if (symbol_section->IsDescendant(text_section_sp.get())) { + if (symbol_section->IsClear(S_ATTR_PURE_INSTRUCTIONS | + S_ATTR_SELF_MODIFYING_CODE | + S_ATTR_SOME_INSTRUCTIONS)) + type = eSymbolTypeData; + else + type = eSymbolTypeCode; + } else if (symbol_section->IsDescendant( + data_section_sp.get()) || + symbol_section->IsDescendant( + data_dirty_section_sp.get()) || + symbol_section->IsDescendant( + data_const_section_sp.get())) { + if (symbol_sect_name && + ::strstr(symbol_sect_name, "__objc") == + symbol_sect_name) { + type = eSymbolTypeRuntime; + + if (symbol_name) { + llvm::StringRef symbol_name_ref(symbol_name); + if (symbol_name_ref.startswith("_OBJC_")) { + static const llvm::StringRef g_objc_v2_prefix_class( + "_OBJC_CLASS_$_"); + static const llvm::StringRef g_objc_v2_prefix_metaclass( + "_OBJC_METACLASS_$_"); + static const llvm::StringRef g_objc_v2_prefix_ivar( + "_OBJC_IVAR_$_"); + if (symbol_name_ref.startswith( + g_objc_v2_prefix_class)) { + symbol_name_non_abi_mangled = symbol_name + 1; + symbol_name = + symbol_name + g_objc_v2_prefix_class.size(); + type = eSymbolTypeObjCClass; + demangled_is_synthesized = true; + } else if (symbol_name_ref.startswith( + g_objc_v2_prefix_metaclass)) { + symbol_name_non_abi_mangled = symbol_name + 1; + symbol_name = + symbol_name + g_objc_v2_prefix_metaclass.size(); + type = eSymbolTypeObjCMetaClass; + demangled_is_synthesized = true; + } else if (symbol_name_ref.startswith( + g_objc_v2_prefix_ivar)) { + symbol_name_non_abi_mangled = symbol_name + 1; + symbol_name = + symbol_name + g_objc_v2_prefix_ivar.size(); + type = eSymbolTypeObjCIVar; + demangled_is_synthesized = true; } + } + } + } else if (symbol_sect_name && + ::strstr(symbol_sect_name, "__gcc_except_tab") == + symbol_sect_name) { + type = eSymbolTypeException; + } else { + type = eSymbolTypeData; + } + } else if (symbol_sect_name && + ::strstr(symbol_sect_name, "__IMPORT") == + symbol_sect_name) { + type = eSymbolTypeTrampoline; + } else if (symbol_section->IsDescendant( + objc_section_sp.get())) { + type = eSymbolTypeRuntime; + if (symbol_name && symbol_name[0] == '.') { + llvm::StringRef symbol_name_ref(symbol_name); + static const llvm::StringRef g_objc_v1_prefix_class( + ".objc_class_name_"); + if (symbol_name_ref.startswith(g_objc_v1_prefix_class)) { + symbol_name_non_abi_mangled = symbol_name; + symbol_name = symbol_name + g_objc_v1_prefix_class.size(); + type = eSymbolTypeObjCClass; + demangled_is_synthesized = true; } + } } - break; - - default: - break; + } } - offset = cmd_offset + load_cmd.cmdsize; + } break; + } } - if (!rpath_paths.empty()) - { - // Fixup all LC_RPATH values to be absolute paths - FileSpec this_file_spec(m_file); - this_file_spec.ResolvePath(); - std::string loader_path("@loader_path"); - std::string executable_path("@executable_path"); - for (auto &rpath : rpath_paths) - { - if (rpath.find(loader_path) == 0) - { - rpath.erase(0, loader_path.size()); - rpath.insert(0, this_file_spec.GetDirectory().GetCString()); + if (add_nlist) { + uint64_t symbol_value = nlist.n_value; + + if (symbol_name_non_abi_mangled) { + sym[sym_idx].GetMangled().SetMangledName( + ConstString(symbol_name_non_abi_mangled)); + sym[sym_idx].GetMangled().SetDemangledName( + ConstString(symbol_name)); + } else { + bool symbol_name_is_mangled = false; + + if (symbol_name && symbol_name[0] == '_') { + symbol_name_is_mangled = symbol_name[1] == '_'; + symbol_name++; // Skip the leading underscore + } + + if (symbol_name) { + ConstString const_symbol_name(symbol_name); + sym[sym_idx].GetMangled().SetValue(const_symbol_name, + symbol_name_is_mangled); + } + } + + if (is_gsym) { + const char *gsym_name = sym[sym_idx] + .GetMangled() + .GetName(lldb::eLanguageTypeUnknown, + Mangled::ePreferMangled) + .GetCString(); + if (gsym_name) + N_GSYM_name_to_sym_idx[gsym_name] = sym_idx; + } + + if (symbol_section) { + const addr_t section_file_addr = symbol_section->GetFileAddress(); + if (symbol_byte_size == 0 && function_starts_count > 0) { + addr_t symbol_lookup_file_addr = nlist.n_value; + // Do an exact address match for non-ARM addresses, else get the + // closest since + // the symbol might be a thumb symbol which has an address with + // bit zero set + FunctionStarts::Entry *func_start_entry = + function_starts.FindEntry(symbol_lookup_file_addr, !is_arm); + if (is_arm && func_start_entry) { + // Verify that the function start address is the symbol address + // (ARM) + // or the symbol address + 1 (thumb) + if (func_start_entry->addr != symbol_lookup_file_addr && + func_start_entry->addr != (symbol_lookup_file_addr + 1)) { + // Not the right entry, NULL it out... + func_start_entry = NULL; } - else if (rpath.find(executable_path) == 0) - { - rpath.erase(0, executable_path.size()); - rpath.insert(0, this_file_spec.GetDirectory().GetCString()); + } + if (func_start_entry) { + func_start_entry->data = true; + + addr_t symbol_file_addr = func_start_entry->addr; + if (is_arm) + symbol_file_addr &= THUMB_ADDRESS_BIT_MASK; + + const FunctionStarts::Entry *next_func_start_entry = + function_starts.FindNextEntry(func_start_entry); + const addr_t section_end_file_addr = + section_file_addr + symbol_section->GetByteSize(); + if (next_func_start_entry) { + addr_t next_symbol_file_addr = next_func_start_entry->addr; + // Be sure the clear the Thumb address bit when we calculate + // the size + // from the current and next address + if (is_arm) + next_symbol_file_addr &= THUMB_ADDRESS_BIT_MASK; + symbol_byte_size = std::min<lldb::addr_t>( + next_symbol_file_addr - symbol_file_addr, + section_end_file_addr - symbol_file_addr); + } else { + symbol_byte_size = section_end_file_addr - symbol_file_addr; } + } } - - for (const auto &rpath_relative_path : rpath_relative_paths) - { - for (const auto &rpath : rpath_paths) - { - std::string path = rpath; - path += rpath_relative_path; - // It is OK to resolve this path because we must find a file on - // disk for us to accept it anyway if it is rpath relative. - FileSpec file_spec(path, true); - // Remove any redundant parts of the path (like "../foo") since - // LC_RPATH values often contain "..". - file_spec.NormalizePath (); - if (file_spec.Exists() && files.AppendIfUnique(file_spec)) - { - count++; - break; - } + symbol_value -= section_file_addr; + } + + if (is_debug == false) { + if (type == eSymbolTypeCode) { + // See if we can find a N_FUN entry for any code symbols. + // If we do find a match, and the name matches, then we + // can merge the two into just the function symbol to avoid + // duplicate entries in the symbol table + std::pair<ValueToSymbolIndexMap::const_iterator, + ValueToSymbolIndexMap::const_iterator> + range; + range = N_FUN_addr_to_sym_idx.equal_range(nlist.n_value); + if (range.first != range.second) { + bool found_it = false; + for (ValueToSymbolIndexMap::const_iterator pos = range.first; + pos != range.second; ++pos) { + if (sym[sym_idx].GetMangled().GetName( + lldb::eLanguageTypeUnknown, + Mangled::ePreferMangled) == + sym[pos->second].GetMangled().GetName( + lldb::eLanguageTypeUnknown, + Mangled::ePreferMangled)) { + m_nlist_idx_to_sym_idx[nlist_idx] = pos->second; + // We just need the flags from the linker symbol, so put + // these flags + // into the N_FUN flags to avoid duplicate symbols in the + // symbol table + sym[pos->second].SetExternal(sym[sym_idx].IsExternal()); + sym[pos->second].SetFlags(nlist.n_type << 16 | + nlist.n_desc); + if (resolver_addresses.find(nlist.n_value) != + resolver_addresses.end()) + sym[pos->second].SetType(eSymbolTypeResolver); + sym[sym_idx].Clear(); + found_it = true; + break; + } + } + if (found_it) + continue; + } else { + if (resolver_addresses.find(nlist.n_value) != + resolver_addresses.end()) + type = eSymbolTypeResolver; + } + } else if (type == eSymbolTypeData || + type == eSymbolTypeObjCClass || + type == eSymbolTypeObjCMetaClass || + type == eSymbolTypeObjCIVar) { + // See if we can find a N_STSYM entry for any data symbols. + // If we do find a match, and the name matches, then we + // can merge the two into just the Static symbol to avoid + // duplicate entries in the symbol table + std::pair<ValueToSymbolIndexMap::const_iterator, + ValueToSymbolIndexMap::const_iterator> + range; + range = N_STSYM_addr_to_sym_idx.equal_range(nlist.n_value); + if (range.first != range.second) { + bool found_it = false; + for (ValueToSymbolIndexMap::const_iterator pos = range.first; + pos != range.second; ++pos) { + if (sym[sym_idx].GetMangled().GetName( + lldb::eLanguageTypeUnknown, + Mangled::ePreferMangled) == + sym[pos->second].GetMangled().GetName( + lldb::eLanguageTypeUnknown, + Mangled::ePreferMangled)) { + m_nlist_idx_to_sym_idx[nlist_idx] = pos->second; + // We just need the flags from the linker symbol, so put + // these flags + // into the N_STSYM flags to avoid duplicate symbols in the + // symbol table + sym[pos->second].SetExternal(sym[sym_idx].IsExternal()); + sym[pos->second].SetFlags(nlist.n_type << 16 | + nlist.n_desc); + sym[sym_idx].Clear(); + found_it = true; + break; + } + } + if (found_it) + continue; + } else { + // Combine N_GSYM stab entries with the non stab symbol + const char *gsym_name = sym[sym_idx] + .GetMangled() + .GetName(lldb::eLanguageTypeUnknown, + Mangled::ePreferMangled) + .GetCString(); + if (gsym_name) { + ConstNameToSymbolIndexMap::const_iterator pos = + N_GSYM_name_to_sym_idx.find(gsym_name); + if (pos != N_GSYM_name_to_sym_idx.end()) { + const uint32_t GSYM_sym_idx = pos->second; + m_nlist_idx_to_sym_idx[nlist_idx] = GSYM_sym_idx; + // Copy the address, because often the N_GSYM address has an + // invalid address of zero + // when the global is a common symbol + sym[GSYM_sym_idx].GetAddressRef().SetSection( + symbol_section); + sym[GSYM_sym_idx].GetAddressRef().SetOffset(symbol_value); + // We just need the flags from the linker symbol, so put + // these flags + // into the N_GSYM flags to avoid duplicate symbols in the + // symbol table + sym[GSYM_sym_idx].SetFlags(nlist.n_type << 16 | + nlist.n_desc); + sym[sym_idx].Clear(); + continue; + } } + } } - } - } - return count; -} + } -lldb_private::Address -ObjectFileMachO::GetEntryPointAddress () -{ - // If the object file is not an executable it can't hold the entry point. m_entry_point_address - // is initialized to an invalid address, so we can just return that. - // If m_entry_point_address is valid it means we've found it already, so return the cached value. + sym[sym_idx].SetID(nlist_idx); + sym[sym_idx].SetType(type); + if (set_value) { + sym[sym_idx].GetAddressRef().SetSection(symbol_section); + sym[sym_idx].GetAddressRef().SetOffset(symbol_value); + } + sym[sym_idx].SetFlags(nlist.n_type << 16 | nlist.n_desc); - if (!IsExecutable() || m_entry_point_address.IsValid()) - return m_entry_point_address; + if (symbol_byte_size > 0) + sym[sym_idx].SetByteSize(symbol_byte_size); - // Otherwise, look for the UnixThread or Thread command. The data for the Thread command is given in - // /usr/include/mach-o.h, but it is basically: - // - // uint32_t flavor - this is the flavor argument you would pass to thread_get_state - // uint32_t count - this is the count of longs in the thread state data - // struct XXX_thread_state state - this is the structure from <machine/thread_status.h> corresponding to the flavor. - // <repeat this trio> - // - // So we just keep reading the various register flavors till we find the GPR one, then read the PC out of there. - // FIXME: We will need to have a "RegisterContext data provider" class at some point that can get all the registers - // out of data in this form & attach them to a given thread. That should underlie the MacOS X User process plugin, - // and we'll also need it for the MacOS X Core File process plugin. When we have that we can also use it here. - // - // For now we hard-code the offsets and flavors we need: - // - // + if (demangled_is_synthesized) + sym[sym_idx].SetDemangledNameIsSynthesized(true); - ModuleSP module_sp(GetModule()); - if (module_sp) - { - std::lock_guard<std::recursive_mutex> guard(module_sp->GetMutex()); - struct load_command load_cmd; - lldb::offset_t offset = MachHeaderSizeFromMagic(m_header.magic); - uint32_t i; - lldb::addr_t start_address = LLDB_INVALID_ADDRESS; - bool done = false; + ++sym_idx; + } else { + sym[sym_idx].Clear(); + } + } + + for (const auto &pos : reexport_shlib_needs_fixup) { + const auto undef_pos = undefined_name_to_desc.find(pos.second); + if (undef_pos != undefined_name_to_desc.end()) { + const uint8_t dylib_ordinal = + llvm::MachO::GET_LIBRARY_ORDINAL(undef_pos->second); + if (dylib_ordinal > 0 && dylib_ordinal < dylib_files.GetSize()) + sym[pos.first].SetReExportedSymbolSharedLibrary( + dylib_files.GetFileSpecAtIndex(dylib_ordinal - 1)); + } + } + } - for (i=0; i<m_header.ncmds; ++i) - { - const lldb::offset_t cmd_offset = offset; - if (m_data.GetU32(&offset, &load_cmd, 2) == NULL) - break; + uint32_t synthetic_sym_id = symtab_load_command.nsyms; - switch (load_cmd.cmd) - { - case LC_UNIXTHREAD: - case LC_THREAD: - { - while (offset < cmd_offset + load_cmd.cmdsize) - { - uint32_t flavor = m_data.GetU32(&offset); - uint32_t count = m_data.GetU32(&offset); - if (count == 0) - { - // We've gotten off somehow, log and exit; - return m_entry_point_address; - } + if (function_starts_count > 0) { + uint32_t num_synthetic_function_symbols = 0; + for (i = 0; i < function_starts_count; ++i) { + if (function_starts.GetEntryRef(i).data == false) + ++num_synthetic_function_symbols; + } - switch (m_header.cputype) - { - case llvm::MachO::CPU_TYPE_ARM: - if (flavor == 1 || flavor == 9) // ARM_THREAD_STATE/ARM_THREAD_STATE32 from mach/arm/thread_status.h - { - offset += 60; // This is the offset of pc in the GPR thread state data structure. - start_address = m_data.GetU32(&offset); - done = true; - } - break; - case llvm::MachO::CPU_TYPE_ARM64: - if (flavor == 6) // ARM_THREAD_STATE64 from mach/arm/thread_status.h - { - offset += 256; // This is the offset of pc in the GPR thread state data structure. - start_address = m_data.GetU64(&offset); - done = true; - } - break; - case llvm::MachO::CPU_TYPE_I386: - if (flavor == 1) // x86_THREAD_STATE32 from mach/i386/thread_status.h - { - offset += 40; // This is the offset of eip in the GPR thread state data structure. - start_address = m_data.GetU32(&offset); - done = true; - } - break; - case llvm::MachO::CPU_TYPE_X86_64: - if (flavor == 4) // x86_THREAD_STATE64 from mach/i386/thread_status.h - { - offset += 16 * 8; // This is the offset of rip in the GPR thread state data structure. - start_address = m_data.GetU64(&offset); - done = true; - } - break; - default: - return m_entry_point_address; - } - // Haven't found the GPR flavor yet, skip over the data for this flavor: - if (done) - break; - offset += count * 4; - } - } - break; - case LC_MAIN: - { - ConstString text_segment_name ("__TEXT"); - uint64_t entryoffset = m_data.GetU64(&offset); - SectionSP text_segment_sp = GetSectionList()->FindSectionByName(text_segment_name); - if (text_segment_sp) - { - done = true; - start_address = text_segment_sp->GetFileAddress() + entryoffset; - } + if (num_synthetic_function_symbols > 0) { + if (num_syms < sym_idx + num_synthetic_function_symbols) { + num_syms = sym_idx + num_synthetic_function_symbols; + sym = symtab->Resize(num_syms); + } + for (i = 0; i < function_starts_count; ++i) { + const FunctionStarts::Entry *func_start_entry = + function_starts.GetEntryAtIndex(i); + if (func_start_entry->data == false) { + addr_t symbol_file_addr = func_start_entry->addr; + uint32_t symbol_flags = 0; + if (is_arm) { + if (symbol_file_addr & 1) + symbol_flags = MACHO_NLIST_ARM_SYMBOL_IS_THUMB; + symbol_file_addr &= THUMB_ADDRESS_BIT_MASK; + } + Address symbol_addr; + if (module_sp->ResolveFileAddress(symbol_file_addr, symbol_addr)) { + SectionSP symbol_section(symbol_addr.GetSection()); + uint32_t symbol_byte_size = 0; + if (symbol_section) { + const addr_t section_file_addr = + symbol_section->GetFileAddress(); + const FunctionStarts::Entry *next_func_start_entry = + function_starts.FindNextEntry(func_start_entry); + const addr_t section_end_file_addr = + section_file_addr + symbol_section->GetByteSize(); + if (next_func_start_entry) { + addr_t next_symbol_file_addr = next_func_start_entry->addr; + if (is_arm) + next_symbol_file_addr &= THUMB_ADDRESS_BIT_MASK; + symbol_byte_size = std::min<lldb::addr_t>( + next_symbol_file_addr - symbol_file_addr, + section_end_file_addr - symbol_file_addr); + } else { + symbol_byte_size = section_end_file_addr - symbol_file_addr; } - break; - - default: - break; + sym[sym_idx].SetID(synthetic_sym_id++); + sym[sym_idx].GetMangled().SetDemangledName( + GetNextSyntheticSymbolName()); + sym[sym_idx].SetType(eSymbolTypeCode); + sym[sym_idx].SetIsSynthetic(true); + sym[sym_idx].GetAddressRef() = symbol_addr; + if (symbol_flags) + sym[sym_idx].SetFlags(symbol_flags); + if (symbol_byte_size) + sym[sym_idx].SetByteSize(symbol_byte_size); + ++sym_idx; + } } - if (done) - break; - - // Go to the next load command: - offset = cmd_offset + load_cmd.cmdsize; + } } + } + } - if (start_address != LLDB_INVALID_ADDRESS) - { - // We got the start address from the load commands, so now resolve that address in the sections - // of this ObjectFile: - if (!m_entry_point_address.ResolveAddressUsingFileSections (start_address, GetSectionList())) - { - m_entry_point_address.Clear(); - } - } - else - { - // We couldn't read the UnixThread load command - maybe it wasn't there. As a fallback look for the - // "start" symbol in the main executable. + // Trim our symbols down to just what we ended up with after + // removing any symbols. + if (sym_idx < num_syms) { + num_syms = sym_idx; + sym = symtab->Resize(num_syms); + } - ModuleSP module_sp (GetModule()); + // Now synthesize indirect symbols + if (m_dysymtab.nindirectsyms != 0) { + if (indirect_symbol_index_data.GetByteSize()) { + NListIndexToSymbolIndexMap::const_iterator end_index_pos = + m_nlist_idx_to_sym_idx.end(); + + for (uint32_t sect_idx = 1; sect_idx < m_mach_sections.size(); + ++sect_idx) { + if ((m_mach_sections[sect_idx].flags & SECTION_TYPE) == + S_SYMBOL_STUBS) { + uint32_t symbol_stub_byte_size = + m_mach_sections[sect_idx].reserved2; + if (symbol_stub_byte_size == 0) + continue; + + const uint32_t num_symbol_stubs = + m_mach_sections[sect_idx].size / symbol_stub_byte_size; + + if (num_symbol_stubs == 0) + continue; + + const uint32_t symbol_stub_index_offset = + m_mach_sections[sect_idx].reserved1; + for (uint32_t stub_idx = 0; stub_idx < num_symbol_stubs; + ++stub_idx) { + const uint32_t symbol_stub_index = + symbol_stub_index_offset + stub_idx; + const lldb::addr_t symbol_stub_addr = + m_mach_sections[sect_idx].addr + + (stub_idx * symbol_stub_byte_size); + lldb::offset_t symbol_stub_offset = symbol_stub_index * 4; + if (indirect_symbol_index_data.ValidOffsetForDataOfSize( + symbol_stub_offset, 4)) { + const uint32_t stub_sym_id = + indirect_symbol_index_data.GetU32(&symbol_stub_offset); + if (stub_sym_id & (INDIRECT_SYMBOL_ABS | INDIRECT_SYMBOL_LOCAL)) + continue; + + NListIndexToSymbolIndexMap::const_iterator index_pos = + m_nlist_idx_to_sym_idx.find(stub_sym_id); + Symbol *stub_symbol = NULL; + if (index_pos != end_index_pos) { + // We have a remapping from the original nlist index to + // a current symbol index, so just look this up by index + stub_symbol = symtab->SymbolAtIndex(index_pos->second); + } else { + // We need to lookup a symbol using the original nlist + // symbol index since this index is coming from the + // S_SYMBOL_STUBS + stub_symbol = symtab->FindSymbolByID(stub_sym_id); + } - if (module_sp) - { - SymbolContextList contexts; - SymbolContext context; - if (module_sp->FindSymbolsWithNameAndType(ConstString ("start"), eSymbolTypeCode, contexts)) - { - if (contexts.GetContextAtIndex(0, context)) - m_entry_point_address = context.symbol->GetAddress(); + if (stub_symbol) { + Address so_addr(symbol_stub_addr, section_list); + + if (stub_symbol->GetType() == eSymbolTypeUndefined) { + // Change the external symbol into a trampoline that makes + // sense + // These symbols were N_UNDF N_EXT, and are useless to us, + // so we + // can re-use them so we don't have to make up a synthetic + // symbol + // for no good reason. + if (resolver_addresses.find(symbol_stub_addr) == + resolver_addresses.end()) + stub_symbol->SetType(eSymbolTypeTrampoline); + else + stub_symbol->SetType(eSymbolTypeResolver); + stub_symbol->SetExternal(false); + stub_symbol->GetAddressRef() = so_addr; + stub_symbol->SetByteSize(symbol_stub_byte_size); + } else { + // Make a synthetic symbol to describe the trampoline stub + Mangled stub_symbol_mangled_name(stub_symbol->GetMangled()); + if (sym_idx >= num_syms) { + sym = symtab->Resize(++num_syms); + stub_symbol = NULL; // this pointer no longer valid + } + sym[sym_idx].SetID(synthetic_sym_id++); + sym[sym_idx].GetMangled() = stub_symbol_mangled_name; + if (resolver_addresses.find(symbol_stub_addr) == + resolver_addresses.end()) + sym[sym_idx].SetType(eSymbolTypeTrampoline); + else + sym[sym_idx].SetType(eSymbolTypeResolver); + sym[sym_idx].SetIsSynthetic(true); + sym[sym_idx].GetAddressRef() = so_addr; + sym[sym_idx].SetByteSize(symbol_stub_byte_size); + ++sym_idx; + } + } else { + if (log) + log->Warning("symbol stub referencing symbol table symbol " + "%u that isn't in our minimal symbol table, " + "fix this!!!", + stub_sym_id); } + } } + } } + } } - return m_entry_point_address; -} - -lldb_private::Address -ObjectFileMachO::GetHeaderAddress () -{ - lldb_private::Address header_addr; - SectionList *section_list = GetSectionList(); - if (section_list) - { - SectionSP text_segment_sp (section_list->FindSectionByName (GetSegmentNameTEXT())); - if (text_segment_sp) - { - header_addr.SetSection (text_segment_sp); - header_addr.SetOffset (0); + if (!trie_entries.empty()) { + for (const auto &e : trie_entries) { + if (e.entry.import_name) { + // Only add indirect symbols from the Trie entries if we + // didn't have a N_INDR nlist entry for this already + if (indirect_symbol_names.find(e.entry.name) == + indirect_symbol_names.end()) { + // Make a synthetic symbol to describe re-exported symbol. + if (sym_idx >= num_syms) + sym = symtab->Resize(++num_syms); + sym[sym_idx].SetID(synthetic_sym_id++); + sym[sym_idx].GetMangled() = Mangled(e.entry.name); + sym[sym_idx].SetType(eSymbolTypeReExported); + sym[sym_idx].SetIsSynthetic(true); + sym[sym_idx].SetReExportedSymbolName(e.entry.import_name); + if (e.entry.other > 0 && e.entry.other <= dylib_files.GetSize()) { + sym[sym_idx].SetReExportedSymbolSharedLibrary( + dylib_files.GetFileSpecAtIndex(e.entry.other - 1)); + } + ++sym_idx; + } } + } } - return header_addr; + + // StreamFile s(stdout, false); + // s.Printf ("Symbol table before CalculateSymbolSizes():\n"); + // symtab->Dump(&s, NULL, eSortOrderNone); + // Set symbol byte sizes correctly since mach-o nlist entries don't have + // sizes + symtab->CalculateSymbolSizes(); + + // s.Printf ("Symbol table after CalculateSymbolSizes():\n"); + // symtab->Dump(&s, NULL, eSortOrderNone); + + return symtab->GetNumSymbols(); + } + return 0; } -uint32_t -ObjectFileMachO::GetNumThreadContexts () -{ - ModuleSP module_sp(GetModule()); - if (module_sp) - { - std::lock_guard<std::recursive_mutex> guard(module_sp->GetMutex()); - if (!m_thread_context_offsets_valid) - { - m_thread_context_offsets_valid = true; - lldb::offset_t offset = MachHeaderSizeFromMagic(m_header.magic); - FileRangeArray::Entry file_range; - thread_command thread_cmd; - for (uint32_t i=0; i<m_header.ncmds; ++i) - { - const uint32_t cmd_offset = offset; - if (m_data.GetU32(&offset, &thread_cmd, 2) == NULL) - break; +void ObjectFileMachO::Dump(Stream *s) { + ModuleSP module_sp(GetModule()); + if (module_sp) { + std::lock_guard<std::recursive_mutex> guard(module_sp->GetMutex()); + s->Printf("%p: ", static_cast<void *>(this)); + s->Indent(); + if (m_header.magic == MH_MAGIC_64 || m_header.magic == MH_CIGAM_64) + s->PutCString("ObjectFileMachO64"); + else + s->PutCString("ObjectFileMachO32"); - if (thread_cmd.cmd == LC_THREAD) - { - file_range.SetRangeBase (offset); - file_range.SetByteSize (thread_cmd.cmdsize - 8); - m_thread_context_offsets.Append (file_range); - } - offset = cmd_offset + thread_cmd.cmdsize; - } - } - } - return m_thread_context_offsets.GetSize(); + ArchSpec header_arch; + GetArchitecture(header_arch); + + *s << ", file = '" << m_file + << "', arch = " << header_arch.GetArchitectureName() << "\n"; + + SectionList *sections = GetSectionList(); + if (sections) + sections->Dump(s, NULL, true, UINT32_MAX); + + if (m_symtab_ap.get()) + m_symtab_ap->Dump(s, NULL, eSortOrderNone); + } } -lldb::RegisterContextSP -ObjectFileMachO::GetThreadContextAtIndex (uint32_t idx, lldb_private::Thread &thread) -{ - lldb::RegisterContextSP reg_ctx_sp; +bool ObjectFileMachO::GetUUID(const llvm::MachO::mach_header &header, + const lldb_private::DataExtractor &data, + lldb::offset_t lc_offset, + lldb_private::UUID &uuid) { + uint32_t i; + struct uuid_command load_cmd; - ModuleSP module_sp(GetModule()); - if (module_sp) - { - std::lock_guard<std::recursive_mutex> guard(module_sp->GetMutex()); - if (!m_thread_context_offsets_valid) - GetNumThreadContexts (); + lldb::offset_t offset = lc_offset; + for (i = 0; i < header.ncmds; ++i) { + const lldb::offset_t cmd_offset = offset; + if (data.GetU32(&offset, &load_cmd, 2) == NULL) + break; - const FileRangeArray::Entry *thread_context_file_range = m_thread_context_offsets.GetEntryAtIndex (idx); - if (thread_context_file_range) - { + if (load_cmd.cmd == LC_UUID) { + const uint8_t *uuid_bytes = data.PeekData(offset, 16); - DataExtractor data (m_data, - thread_context_file_range->GetRangeBase(), - thread_context_file_range->GetByteSize()); + if (uuid_bytes) { + // OpenCL on Mac OS X uses the same UUID for each of its object files. + // We pretend these object files have no UUID to prevent crashing. - switch (m_header.cputype) - { - case llvm::MachO::CPU_TYPE_ARM64: - reg_ctx_sp.reset (new RegisterContextDarwin_arm64_Mach (thread, data)); - break; - - case llvm::MachO::CPU_TYPE_ARM: - reg_ctx_sp.reset (new RegisterContextDarwin_arm_Mach (thread, data)); - break; + const uint8_t opencl_uuid[] = {0x8c, 0x8e, 0xb3, 0x9b, 0x3b, 0xa8, + 0x4b, 0x16, 0xb6, 0xa4, 0x27, 0x63, + 0xbb, 0x14, 0xf0, 0x0d}; - case llvm::MachO::CPU_TYPE_I386: - reg_ctx_sp.reset (new RegisterContextDarwin_i386_Mach (thread, data)); - break; + if (!memcmp(uuid_bytes, opencl_uuid, 16)) + return false; - case llvm::MachO::CPU_TYPE_X86_64: - reg_ctx_sp.reset (new RegisterContextDarwin_x86_64_Mach (thread, data)); - break; - } + uuid.SetBytes(uuid_bytes); + return true; + } + return false; + } + offset = cmd_offset + load_cmd.cmdsize; + } + return false; +} + +bool ObjectFileMachO::GetArchitecture(const llvm::MachO::mach_header &header, + const lldb_private::DataExtractor &data, + lldb::offset_t lc_offset, + ArchSpec &arch) { + arch.SetArchitecture(eArchTypeMachO, header.cputype, header.cpusubtype); + + if (arch.IsValid()) { + llvm::Triple &triple = arch.GetTriple(); + + // Set OS to an unspecified unknown or a "*" so it can match any OS + triple.setOS(llvm::Triple::UnknownOS); + triple.setOSName(llvm::StringRef()); + + if (header.filetype == MH_PRELOAD) { + if (header.cputype == CPU_TYPE_ARM) { + // If this is a 32-bit arm binary, and it's a standalone binary, + // force the Vendor to Apple so we don't accidentally pick up + // the generic armv7 ABI at runtime. Apple's armv7 ABI always uses + // r7 for the frame pointer register; most other armv7 ABIs use a + // combination of r7 and r11. + triple.setVendor(llvm::Triple::Apple); + } else { + // Set vendor to an unspecified unknown or a "*" so it can match any + // vendor + // This is required for correct behavior of EFI debugging on x86_64 + triple.setVendor(llvm::Triple::UnknownVendor); + triple.setVendorName(llvm::StringRef()); + } + return true; + } else { + struct load_command load_cmd; + + lldb::offset_t offset = lc_offset; + for (uint32_t i = 0; i < header.ncmds; ++i) { + const lldb::offset_t cmd_offset = offset; + if (data.GetU32(&offset, &load_cmd, 2) == NULL) + break; + + switch (load_cmd.cmd) { + case llvm::MachO::LC_VERSION_MIN_IPHONEOS: + triple.setOS(llvm::Triple::IOS); + return true; + + case llvm::MachO::LC_VERSION_MIN_MACOSX: + triple.setOS(llvm::Triple::MacOSX); + return true; + + case llvm::MachO::LC_VERSION_MIN_TVOS: + triple.setOS(llvm::Triple::TvOS); + return true; + + case llvm::MachO::LC_VERSION_MIN_WATCHOS: + triple.setOS(llvm::Triple::WatchOS); + return true; + + default: + break; } + + offset = cmd_offset + load_cmd.cmdsize; + } + + if (header.filetype != MH_KEXT_BUNDLE) { + // We didn't find a LC_VERSION_MIN load command and this isn't a KEXT + // so lets not say our Vendor is Apple, leave it as an unspecified + // unknown + triple.setVendor(llvm::Triple::UnknownVendor); + triple.setVendorName(llvm::StringRef()); + } } - return reg_ctx_sp; + } + return arch.IsValid(); } -ObjectFile::Type -ObjectFileMachO::CalculateType() -{ - switch (m_header.filetype) - { - case MH_OBJECT: // 0x1u - if (GetAddressByteSize () == 4) - { - // 32 bit kexts are just object files, but they do have a valid - // UUID load command. - UUID uuid; - if (GetUUID(&uuid)) - { - // this checking for the UUID load command is not enough - // we could eventually look for the symbol named - // "OSKextGetCurrentIdentifier" as this is required of kexts - if (m_strata == eStrataInvalid) - m_strata = eStrataKernel; - return eTypeSharedLibrary; - } +bool ObjectFileMachO::GetUUID(lldb_private::UUID *uuid) { + ModuleSP module_sp(GetModule()); + if (module_sp) { + std::lock_guard<std::recursive_mutex> guard(module_sp->GetMutex()); + lldb::offset_t offset = MachHeaderSizeFromMagic(m_header.magic); + return GetUUID(m_header, m_data, offset, *uuid); + } + return false; +} + +uint32_t ObjectFileMachO::GetDependentModules(FileSpecList &files) { + uint32_t count = 0; + ModuleSP module_sp(GetModule()); + if (module_sp) { + std::lock_guard<std::recursive_mutex> guard(module_sp->GetMutex()); + struct load_command load_cmd; + lldb::offset_t offset = MachHeaderSizeFromMagic(m_header.magic); + std::vector<std::string> rpath_paths; + std::vector<std::string> rpath_relative_paths; + const bool resolve_path = false; // Don't resolve the dependent file paths + // since they may not reside on this system + uint32_t i; + for (i = 0; i < m_header.ncmds; ++i) { + const uint32_t cmd_offset = offset; + if (m_data.GetU32(&offset, &load_cmd, 2) == NULL) + break; + + switch (load_cmd.cmd) { + case LC_RPATH: + case LC_LOAD_DYLIB: + case LC_LOAD_WEAK_DYLIB: + case LC_REEXPORT_DYLIB: + case LC_LOAD_DYLINKER: + case LC_LOADFVMLIB: + case LC_LOAD_UPWARD_DYLIB: { + uint32_t name_offset = cmd_offset + m_data.GetU32(&offset); + const char *path = m_data.PeekCStr(name_offset); + if (path) { + if (load_cmd.cmd == LC_RPATH) + rpath_paths.push_back(path); + else { + if (path[0] == '@') { + if (strncmp(path, "@rpath", strlen("@rpath")) == 0) + rpath_relative_paths.push_back(path + strlen("@rpath")); + } else { + FileSpec file_spec(path, resolve_path); + if (files.AppendIfUnique(file_spec)) + count++; } - return eTypeObjectFile; - - case MH_EXECUTE: return eTypeExecutable; // 0x2u - case MH_FVMLIB: return eTypeSharedLibrary; // 0x3u - case MH_CORE: return eTypeCoreFile; // 0x4u - case MH_PRELOAD: return eTypeSharedLibrary; // 0x5u - case MH_DYLIB: return eTypeSharedLibrary; // 0x6u - case MH_DYLINKER: return eTypeDynamicLinker; // 0x7u - case MH_BUNDLE: return eTypeSharedLibrary; // 0x8u - case MH_DYLIB_STUB: return eTypeStubLibrary; // 0x9u - case MH_DSYM: return eTypeDebugInfo; // 0xAu - case MH_KEXT_BUNDLE: return eTypeSharedLibrary; // 0xBu - default: + } + } + } break; + + default: + break; + } + offset = cmd_offset + load_cmd.cmdsize; + } + + if (!rpath_paths.empty()) { + // Fixup all LC_RPATH values to be absolute paths + FileSpec this_file_spec(m_file); + this_file_spec.ResolvePath(); + std::string loader_path("@loader_path"); + std::string executable_path("@executable_path"); + for (auto &rpath : rpath_paths) { + if (rpath.find(loader_path) == 0) { + rpath.erase(0, loader_path.size()); + rpath.insert(0, this_file_spec.GetDirectory().GetCString()); + } else if (rpath.find(executable_path) == 0) { + rpath.erase(0, executable_path.size()); + rpath.insert(0, this_file_spec.GetDirectory().GetCString()); + } + } + + for (const auto &rpath_relative_path : rpath_relative_paths) { + for (const auto &rpath : rpath_paths) { + std::string path = rpath; + path += rpath_relative_path; + // It is OK to resolve this path because we must find a file on + // disk for us to accept it anyway if it is rpath relative. + FileSpec file_spec(path, true); + // Remove any redundant parts of the path (like "../foo") since + // LC_RPATH values often contain "..". + file_spec = file_spec.GetNormalizedPath(); + if (file_spec.Exists() && files.AppendIfUnique(file_spec)) { + count++; break; + } + } + } } - return eTypeUnknown; + } + return count; } -ObjectFile::Strata -ObjectFileMachO::CalculateStrata() -{ - switch (m_header.filetype) - { - case MH_OBJECT: // 0x1u +lldb_private::Address ObjectFileMachO::GetEntryPointAddress() { + // If the object file is not an executable it can't hold the entry point. + // m_entry_point_address + // is initialized to an invalid address, so we can just return that. + // If m_entry_point_address is valid it means we've found it already, so + // return the cached value. + + if (!IsExecutable() || m_entry_point_address.IsValid()) + return m_entry_point_address; + + // Otherwise, look for the UnixThread or Thread command. The data for the + // Thread command is given in + // /usr/include/mach-o.h, but it is basically: + // + // uint32_t flavor - this is the flavor argument you would pass to + // thread_get_state + // uint32_t count - this is the count of longs in the thread state data + // struct XXX_thread_state state - this is the structure from + // <machine/thread_status.h> corresponding to the flavor. + // <repeat this trio> + // + // So we just keep reading the various register flavors till we find the GPR + // one, then read the PC out of there. + // FIXME: We will need to have a "RegisterContext data provider" class at some + // point that can get all the registers + // out of data in this form & attach them to a given thread. That should + // underlie the MacOS X User process plugin, + // and we'll also need it for the MacOS X Core File process plugin. When we + // have that we can also use it here. + // + // For now we hard-code the offsets and flavors we need: + // + // + + ModuleSP module_sp(GetModule()); + if (module_sp) { + std::lock_guard<std::recursive_mutex> guard(module_sp->GetMutex()); + struct load_command load_cmd; + lldb::offset_t offset = MachHeaderSizeFromMagic(m_header.magic); + uint32_t i; + lldb::addr_t start_address = LLDB_INVALID_ADDRESS; + bool done = false; + + for (i = 0; i < m_header.ncmds; ++i) { + const lldb::offset_t cmd_offset = offset; + if (m_data.GetU32(&offset, &load_cmd, 2) == NULL) + break; + + switch (load_cmd.cmd) { + case LC_UNIXTHREAD: + case LC_THREAD: { + while (offset < cmd_offset + load_cmd.cmdsize) { + uint32_t flavor = m_data.GetU32(&offset); + uint32_t count = m_data.GetU32(&offset); + if (count == 0) { + // We've gotten off somehow, log and exit; + return m_entry_point_address; + } + + switch (m_header.cputype) { + case llvm::MachO::CPU_TYPE_ARM: + if (flavor == 1 || + flavor == 9) // ARM_THREAD_STATE/ARM_THREAD_STATE32 from + // mach/arm/thread_status.h { - // 32 bit kexts are just object files, but they do have a valid - // UUID load command. - UUID uuid; - if (GetUUID(&uuid)) - { - // this checking for the UUID load command is not enough - // we could eventually look for the symbol named - // "OSKextGetCurrentIdentifier" as this is required of kexts - if (m_type == eTypeInvalid) - m_type = eTypeSharedLibrary; - - return eStrataKernel; - } + offset += 60; // This is the offset of pc in the GPR thread state + // data structure. + start_address = m_data.GetU32(&offset); + done = true; } - return eStrataUnknown; - - case MH_EXECUTE: // 0x2u - // Check for the MH_DYLDLINK bit in the flags - if (m_header.flags & MH_DYLDLINK) + break; + case llvm::MachO::CPU_TYPE_ARM64: + if (flavor == 6) // ARM_THREAD_STATE64 from mach/arm/thread_status.h { - return eStrataUser; + offset += 256; // This is the offset of pc in the GPR thread state + // data structure. + start_address = m_data.GetU64(&offset); + done = true; } - else + break; + case llvm::MachO::CPU_TYPE_I386: + if (flavor == + 1) // x86_THREAD_STATE32 from mach/i386/thread_status.h { - SectionList *section_list = GetSectionList(); - if (section_list) - { - static ConstString g_kld_section_name ("__KLD"); - if (section_list->FindSectionByName(g_kld_section_name)) - return eStrataKernel; - } + offset += 40; // This is the offset of eip in the GPR thread state + // data structure. + start_address = m_data.GetU32(&offset); + done = true; } - return eStrataRawImage; - - case MH_FVMLIB: return eStrataUser; // 0x3u - case MH_CORE: return eStrataUnknown; // 0x4u - case MH_PRELOAD: return eStrataRawImage; // 0x5u - case MH_DYLIB: return eStrataUser; // 0x6u - case MH_DYLINKER: return eStrataUser; // 0x7u - case MH_BUNDLE: return eStrataUser; // 0x8u - case MH_DYLIB_STUB: return eStrataUser; // 0x9u - case MH_DSYM: return eStrataUnknown; // 0xAu - case MH_KEXT_BUNDLE: return eStrataKernel; // 0xBu - default: break; + case llvm::MachO::CPU_TYPE_X86_64: + if (flavor == + 4) // x86_THREAD_STATE64 from mach/i386/thread_status.h + { + offset += 16 * 8; // This is the offset of rip in the GPR thread + // state data structure. + start_address = m_data.GetU64(&offset); + done = true; + } + break; + default: + return m_entry_point_address; + } + // Haven't found the GPR flavor yet, skip over the data for this + // flavor: + if (done) + break; + offset += count * 4; + } + } break; + case LC_MAIN: { + ConstString text_segment_name("__TEXT"); + uint64_t entryoffset = m_data.GetU64(&offset); + SectionSP text_segment_sp = + GetSectionList()->FindSectionByName(text_segment_name); + if (text_segment_sp) { + done = true; + start_address = text_segment_sp->GetFileAddress() + entryoffset; + } + } break; + + default: + break; + } + if (done) + break; + + // Go to the next load command: + offset = cmd_offset + load_cmd.cmdsize; } - return eStrataUnknown; + + if (start_address != LLDB_INVALID_ADDRESS) { + // We got the start address from the load commands, so now resolve that + // address in the sections + // of this ObjectFile: + if (!m_entry_point_address.ResolveAddressUsingFileSections( + start_address, GetSectionList())) { + m_entry_point_address.Clear(); + } + } else { + // We couldn't read the UnixThread load command - maybe it wasn't there. + // As a fallback look for the + // "start" symbol in the main executable. + + ModuleSP module_sp(GetModule()); + + if (module_sp) { + SymbolContextList contexts; + SymbolContext context; + if (module_sp->FindSymbolsWithNameAndType(ConstString("start"), + eSymbolTypeCode, contexts)) { + if (contexts.GetContextAtIndex(0, context)) + m_entry_point_address = context.symbol->GetAddress(); + } + } + } + } + + return m_entry_point_address; } -uint32_t -ObjectFileMachO::GetVersion (uint32_t *versions, uint32_t num_versions) -{ - ModuleSP module_sp(GetModule()); - if (module_sp) - { - std::lock_guard<std::recursive_mutex> guard(module_sp->GetMutex()); - struct dylib_command load_cmd; - lldb::offset_t offset = MachHeaderSizeFromMagic(m_header.magic); - uint32_t version_cmd = 0; - uint64_t version = 0; - uint32_t i; - for (i=0; i<m_header.ncmds; ++i) - { - const lldb::offset_t cmd_offset = offset; - if (m_data.GetU32(&offset, &load_cmd, 2) == NULL) - break; +lldb_private::Address ObjectFileMachO::GetHeaderAddress() { + lldb_private::Address header_addr; + SectionList *section_list = GetSectionList(); + if (section_list) { + SectionSP text_segment_sp( + section_list->FindSectionByName(GetSegmentNameTEXT())); + if (text_segment_sp) { + header_addr.SetSection(text_segment_sp); + header_addr.SetOffset(0); + } + } + return header_addr; +} - if (load_cmd.cmd == LC_ID_DYLIB) - { - if (version_cmd == 0) - { - version_cmd = load_cmd.cmd; - if (m_data.GetU32(&offset, &load_cmd.dylib, 4) == NULL) - break; - version = load_cmd.dylib.current_version; - } - break; // Break for now unless there is another more complete version - // number load command in the future. - } - offset = cmd_offset + load_cmd.cmdsize; +uint32_t ObjectFileMachO::GetNumThreadContexts() { + ModuleSP module_sp(GetModule()); + if (module_sp) { + std::lock_guard<std::recursive_mutex> guard(module_sp->GetMutex()); + if (!m_thread_context_offsets_valid) { + m_thread_context_offsets_valid = true; + lldb::offset_t offset = MachHeaderSizeFromMagic(m_header.magic); + FileRangeArray::Entry file_range; + thread_command thread_cmd; + for (uint32_t i = 0; i < m_header.ncmds; ++i) { + const uint32_t cmd_offset = offset; + if (m_data.GetU32(&offset, &thread_cmd, 2) == NULL) + break; + + if (thread_cmd.cmd == LC_THREAD) { + file_range.SetRangeBase(offset); + file_range.SetByteSize(thread_cmd.cmdsize - 8); + m_thread_context_offsets.Append(file_range); } + offset = cmd_offset + thread_cmd.cmdsize; + } + } + } + return m_thread_context_offsets.GetSize(); +} - if (version_cmd == LC_ID_DYLIB) - { - if (versions != NULL && num_versions > 0) - { - if (num_versions > 0) - versions[0] = (version & 0xFFFF0000ull) >> 16; - if (num_versions > 1) - versions[1] = (version & 0x0000FF00ull) >> 8; - if (num_versions > 2) - versions[2] = (version & 0x000000FFull); - // Fill in an remaining version numbers with invalid values - for (i=3; i<num_versions; ++i) - versions[i] = UINT32_MAX; - } - // The LC_ID_DYLIB load command has a version with 3 version numbers - // in it, so always return 3 - return 3; - } +lldb::RegisterContextSP +ObjectFileMachO::GetThreadContextAtIndex(uint32_t idx, + lldb_private::Thread &thread) { + lldb::RegisterContextSP reg_ctx_sp; + + ModuleSP module_sp(GetModule()); + if (module_sp) { + std::lock_guard<std::recursive_mutex> guard(module_sp->GetMutex()); + if (!m_thread_context_offsets_valid) + GetNumThreadContexts(); + + const FileRangeArray::Entry *thread_context_file_range = + m_thread_context_offsets.GetEntryAtIndex(idx); + if (thread_context_file_range) { + + DataExtractor data(m_data, thread_context_file_range->GetRangeBase(), + thread_context_file_range->GetByteSize()); + + switch (m_header.cputype) { + case llvm::MachO::CPU_TYPE_ARM64: + reg_ctx_sp.reset(new RegisterContextDarwin_arm64_Mach(thread, data)); + break; + + case llvm::MachO::CPU_TYPE_ARM: + reg_ctx_sp.reset(new RegisterContextDarwin_arm_Mach(thread, data)); + break; + + case llvm::MachO::CPU_TYPE_I386: + reg_ctx_sp.reset(new RegisterContextDarwin_i386_Mach(thread, data)); + break; + + case llvm::MachO::CPU_TYPE_X86_64: + reg_ctx_sp.reset(new RegisterContextDarwin_x86_64_Mach(thread, data)); + break; + } } - return false; + } + return reg_ctx_sp; } -bool -ObjectFileMachO::GetArchitecture (ArchSpec &arch) -{ - ModuleSP module_sp(GetModule()); - if (module_sp) - { - std::lock_guard<std::recursive_mutex> guard(module_sp->GetMutex()); - return GetArchitecture (m_header, m_data, MachHeaderSizeFromMagic(m_header.magic), arch); +ObjectFile::Type ObjectFileMachO::CalculateType() { + switch (m_header.filetype) { + case MH_OBJECT: // 0x1u + if (GetAddressByteSize() == 4) { + // 32 bit kexts are just object files, but they do have a valid + // UUID load command. + UUID uuid; + if (GetUUID(&uuid)) { + // this checking for the UUID load command is not enough + // we could eventually look for the symbol named + // "OSKextGetCurrentIdentifier" as this is required of kexts + if (m_strata == eStrataInvalid) + m_strata = eStrataKernel; + return eTypeSharedLibrary; + } } - return false; + return eTypeObjectFile; + + case MH_EXECUTE: + return eTypeExecutable; // 0x2u + case MH_FVMLIB: + return eTypeSharedLibrary; // 0x3u + case MH_CORE: + return eTypeCoreFile; // 0x4u + case MH_PRELOAD: + return eTypeSharedLibrary; // 0x5u + case MH_DYLIB: + return eTypeSharedLibrary; // 0x6u + case MH_DYLINKER: + return eTypeDynamicLinker; // 0x7u + case MH_BUNDLE: + return eTypeSharedLibrary; // 0x8u + case MH_DYLIB_STUB: + return eTypeStubLibrary; // 0x9u + case MH_DSYM: + return eTypeDebugInfo; // 0xAu + case MH_KEXT_BUNDLE: + return eTypeSharedLibrary; // 0xBu + default: + break; + } + return eTypeUnknown; } -UUID -ObjectFileMachO::GetProcessSharedCacheUUID (Process *process) -{ +ObjectFile::Strata ObjectFileMachO::CalculateStrata() { + switch (m_header.filetype) { + case MH_OBJECT: // 0x1u + { + // 32 bit kexts are just object files, but they do have a valid + // UUID load command. UUID uuid; - if (process) - { - addr_t all_image_infos = process->GetImageInfoAddress(); - - // The address returned by GetImageInfoAddress may be the address of dyld (don't want) - // or it may be the address of the dyld_all_image_infos structure (want). The first four - // bytes will be either the version field (all_image_infos) or a Mach-O file magic constant. - // Version 13 and higher of dyld_all_image_infos is required to get the sharedCacheUUID field. - - Error err; - uint32_t version_or_magic = process->ReadUnsignedIntegerFromMemory (all_image_infos, 4, -1, err); - if (version_or_magic != static_cast<uint32_t>(-1) - && version_or_magic != MH_MAGIC - && version_or_magic != MH_CIGAM - && version_or_magic != MH_MAGIC_64 - && version_or_magic != MH_CIGAM_64 - && version_or_magic >= 13) - { - addr_t sharedCacheUUID_address = LLDB_INVALID_ADDRESS; - int wordsize = process->GetAddressByteSize(); - if (wordsize == 8) - { - sharedCacheUUID_address = all_image_infos + 160; // sharedCacheUUID <mach-o/dyld_images.h> - } - if (wordsize == 4) - { - sharedCacheUUID_address = all_image_infos + 84; // sharedCacheUUID <mach-o/dyld_images.h> - } - if (sharedCacheUUID_address != LLDB_INVALID_ADDRESS) - { - uuid_t shared_cache_uuid; - if (process->ReadMemory (sharedCacheUUID_address, shared_cache_uuid, sizeof (uuid_t), err) == sizeof (uuid_t)) - { - uuid.SetBytes (shared_cache_uuid); - } - } + if (GetUUID(&uuid)) { + // this checking for the UUID load command is not enough + // we could eventually look for the symbol named + // "OSKextGetCurrentIdentifier" as this is required of kexts + if (m_type == eTypeInvalid) + m_type = eTypeSharedLibrary; + + return eStrataKernel; + } + } + return eStrataUnknown; + + case MH_EXECUTE: // 0x2u + // Check for the MH_DYLDLINK bit in the flags + if (m_header.flags & MH_DYLDLINK) { + return eStrataUser; + } else { + SectionList *section_list = GetSectionList(); + if (section_list) { + static ConstString g_kld_section_name("__KLD"); + if (section_list->FindSectionByName(g_kld_section_name)) + return eStrataKernel; + } + } + return eStrataRawImage; + + case MH_FVMLIB: + return eStrataUser; // 0x3u + case MH_CORE: + return eStrataUnknown; // 0x4u + case MH_PRELOAD: + return eStrataRawImage; // 0x5u + case MH_DYLIB: + return eStrataUser; // 0x6u + case MH_DYLINKER: + return eStrataUser; // 0x7u + case MH_BUNDLE: + return eStrataUser; // 0x8u + case MH_DYLIB_STUB: + return eStrataUser; // 0x9u + case MH_DSYM: + return eStrataUnknown; // 0xAu + case MH_KEXT_BUNDLE: + return eStrataKernel; // 0xBu + default: + break; + } + return eStrataUnknown; +} + +uint32_t ObjectFileMachO::GetVersion(uint32_t *versions, + uint32_t num_versions) { + ModuleSP module_sp(GetModule()); + if (module_sp) { + std::lock_guard<std::recursive_mutex> guard(module_sp->GetMutex()); + struct dylib_command load_cmd; + lldb::offset_t offset = MachHeaderSizeFromMagic(m_header.magic); + uint32_t version_cmd = 0; + uint64_t version = 0; + uint32_t i; + for (i = 0; i < m_header.ncmds; ++i) { + const lldb::offset_t cmd_offset = offset; + if (m_data.GetU32(&offset, &load_cmd, 2) == NULL) + break; + + if (load_cmd.cmd == LC_ID_DYLIB) { + if (version_cmd == 0) { + version_cmd = load_cmd.cmd; + if (m_data.GetU32(&offset, &load_cmd.dylib, 4) == NULL) + break; + version = load_cmd.dylib.current_version; } + break; // Break for now unless there is another more complete version + // number load command in the future. + } + offset = cmd_offset + load_cmd.cmdsize; + } + + if (version_cmd == LC_ID_DYLIB) { + if (versions != NULL && num_versions > 0) { + if (num_versions > 0) + versions[0] = (version & 0xFFFF0000ull) >> 16; + if (num_versions > 1) + versions[1] = (version & 0x0000FF00ull) >> 8; + if (num_versions > 2) + versions[2] = (version & 0x000000FFull); + // Fill in an remaining version numbers with invalid values + for (i = 3; i < num_versions; ++i) + versions[i] = UINT32_MAX; + } + // The LC_ID_DYLIB load command has a version with 3 version numbers + // in it, so always return 3 + return 3; } - return uuid; + } + return false; } -UUID -ObjectFileMachO::GetLLDBSharedCacheUUID () -{ - UUID uuid; -#if defined (__APPLE__) && (defined (__arm__) || defined (__arm64__) || defined (__aarch64__)) - uint8_t *(*dyld_get_all_image_infos)(void); - dyld_get_all_image_infos = (uint8_t*(*)()) dlsym (RTLD_DEFAULT, "_dyld_get_all_image_infos"); - if (dyld_get_all_image_infos) - { - uint8_t *dyld_all_image_infos_address = dyld_get_all_image_infos(); - if (dyld_all_image_infos_address) - { - uint32_t *version = (uint32_t*) dyld_all_image_infos_address; // version <mach-o/dyld_images.h> - if (*version >= 13) - { - uuid_t *sharedCacheUUID_address = 0; - int wordsize = sizeof (uint8_t *); - if (wordsize == 8) - { - sharedCacheUUID_address = (uuid_t*) ((uint8_t*) dyld_all_image_infos_address + 160); // sharedCacheUUID <mach-o/dyld_images.h> - } - else - { - sharedCacheUUID_address = (uuid_t*) ((uint8_t*) dyld_all_image_infos_address + 84); // sharedCacheUUID <mach-o/dyld_images.h> - } - uuid.SetBytes (sharedCacheUUID_address); - } +bool ObjectFileMachO::GetArchitecture(ArchSpec &arch) { + ModuleSP module_sp(GetModule()); + if (module_sp) { + std::lock_guard<std::recursive_mutex> guard(module_sp->GetMutex()); + return GetArchitecture(m_header, m_data, + MachHeaderSizeFromMagic(m_header.magic), arch); + } + return false; +} + +UUID ObjectFileMachO::GetProcessSharedCacheUUID(Process *process) { + UUID uuid; + if (process && process->GetDynamicLoader()) { + DynamicLoader *dl = process->GetDynamicLoader(); + addr_t load_address; + LazyBool using_shared_cache; + LazyBool private_shared_cache; + dl->GetSharedCacheInformation(load_address, uuid, using_shared_cache, + private_shared_cache); + } + return uuid; +} + +UUID ObjectFileMachO::GetLLDBSharedCacheUUID() { + UUID uuid; +#if defined(__APPLE__) && \ + (defined(__arm__) || defined(__arm64__) || defined(__aarch64__)) + uint8_t *(*dyld_get_all_image_infos)(void); + dyld_get_all_image_infos = + (uint8_t * (*)())dlsym(RTLD_DEFAULT, "_dyld_get_all_image_infos"); + if (dyld_get_all_image_infos) { + uint8_t *dyld_all_image_infos_address = dyld_get_all_image_infos(); + if (dyld_all_image_infos_address) { + uint32_t *version = (uint32_t *) + dyld_all_image_infos_address; // version <mach-o/dyld_images.h> + if (*version >= 13) { + uuid_t *sharedCacheUUID_address = 0; + int wordsize = sizeof(uint8_t *); + if (wordsize == 8) { + sharedCacheUUID_address = + (uuid_t *)((uint8_t *)dyld_all_image_infos_address + + 160); // sharedCacheUUID <mach-o/dyld_images.h> + } else { + sharedCacheUUID_address = + (uuid_t *)((uint8_t *)dyld_all_image_infos_address + + 84); // sharedCacheUUID <mach-o/dyld_images.h> } + uuid.SetBytes(sharedCacheUUID_address); + } } + } #endif - return uuid; + return uuid; } -uint32_t -ObjectFileMachO::GetMinimumOSVersion (uint32_t *versions, uint32_t num_versions) -{ - if (m_min_os_versions.empty()) - { - lldb::offset_t offset = MachHeaderSizeFromMagic(m_header.magic); - bool success = false; - for (uint32_t i=0; success == false && i < m_header.ncmds; ++i) - { - const lldb::offset_t load_cmd_offset = offset; - - version_min_command lc; - if (m_data.GetU32(&offset, &lc.cmd, 2) == NULL) - break; - if (lc.cmd == llvm::MachO::LC_VERSION_MIN_MACOSX - || lc.cmd == llvm::MachO::LC_VERSION_MIN_IPHONEOS - || lc.cmd == llvm::MachO::LC_VERSION_MIN_TVOS - || lc.cmd == llvm::MachO::LC_VERSION_MIN_WATCHOS) - { - if (m_data.GetU32 (&offset, &lc.version, (sizeof(lc) / sizeof(uint32_t)) - 2)) - { - const uint32_t xxxx = lc.version >> 16; - const uint32_t yy = (lc.version >> 8) & 0xffu; - const uint32_t zz = lc.version & 0xffu; - if (xxxx) - { - m_min_os_versions.push_back(xxxx); - m_min_os_versions.push_back(yy); - m_min_os_versions.push_back(zz); - } - success = true; - } - } - offset = load_cmd_offset + lc.cmdsize; - } - - if (success == false) - { - // Push an invalid value so we don't keep trying to - m_min_os_versions.push_back(UINT32_MAX); +uint32_t ObjectFileMachO::GetMinimumOSVersion(uint32_t *versions, + uint32_t num_versions) { + if (m_min_os_versions.empty()) { + lldb::offset_t offset = MachHeaderSizeFromMagic(m_header.magic); + bool success = false; + for (uint32_t i = 0; success == false && i < m_header.ncmds; ++i) { + const lldb::offset_t load_cmd_offset = offset; + + version_min_command lc; + if (m_data.GetU32(&offset, &lc.cmd, 2) == NULL) + break; + if (lc.cmd == llvm::MachO::LC_VERSION_MIN_MACOSX || + lc.cmd == llvm::MachO::LC_VERSION_MIN_IPHONEOS || + lc.cmd == llvm::MachO::LC_VERSION_MIN_TVOS || + lc.cmd == llvm::MachO::LC_VERSION_MIN_WATCHOS) { + if (m_data.GetU32(&offset, &lc.version, + (sizeof(lc) / sizeof(uint32_t)) - 2)) { + const uint32_t xxxx = lc.version >> 16; + const uint32_t yy = (lc.version >> 8) & 0xffu; + const uint32_t zz = lc.version & 0xffu; + if (xxxx) { + m_min_os_versions.push_back(xxxx); + m_min_os_versions.push_back(yy); + m_min_os_versions.push_back(zz); + } + success = true; } + } + offset = load_cmd_offset + lc.cmdsize; } - - if (m_min_os_versions.size() > 1 || m_min_os_versions[0] != UINT32_MAX) - { - if (versions != NULL && num_versions > 0) - { - for (size_t i=0; i<num_versions; ++i) - { - if (i < m_min_os_versions.size()) - versions[i] = m_min_os_versions[i]; - else - versions[i] = 0; - } - } - return m_min_os_versions.size(); + + if (success == false) { + // Push an invalid value so we don't keep trying to + m_min_os_versions.push_back(UINT32_MAX); + } + } + + if (m_min_os_versions.size() > 1 || m_min_os_versions[0] != UINT32_MAX) { + if (versions != NULL && num_versions > 0) { + for (size_t i = 0; i < num_versions; ++i) { + if (i < m_min_os_versions.size()) + versions[i] = m_min_os_versions[i]; + else + versions[i] = 0; + } } - // Call the superclasses version that will empty out the data - return ObjectFile::GetMinimumOSVersion (versions, num_versions); + return m_min_os_versions.size(); + } + // Call the superclasses version that will empty out the data + return ObjectFile::GetMinimumOSVersion(versions, num_versions); } -uint32_t -ObjectFileMachO::GetSDKVersion(uint32_t *versions, uint32_t num_versions) -{ - if (m_sdk_versions.empty()) - { - lldb::offset_t offset = MachHeaderSizeFromMagic(m_header.magic); - bool success = false; - for (uint32_t i=0; success == false && i < m_header.ncmds; ++i) - { - const lldb::offset_t load_cmd_offset = offset; - - version_min_command lc; - if (m_data.GetU32(&offset, &lc.cmd, 2) == NULL) - break; - if (lc.cmd == llvm::MachO::LC_VERSION_MIN_MACOSX - || lc.cmd == llvm::MachO::LC_VERSION_MIN_IPHONEOS - || lc.cmd == llvm::MachO::LC_VERSION_MIN_TVOS - || lc.cmd == llvm::MachO::LC_VERSION_MIN_WATCHOS) - { - if (m_data.GetU32 (&offset, &lc.version, (sizeof(lc) / sizeof(uint32_t)) - 2)) - { - const uint32_t xxxx = lc.sdk >> 16; - const uint32_t yy = (lc.sdk >> 8) & 0xffu; - const uint32_t zz = lc.sdk & 0xffu; - if (xxxx) - { - m_sdk_versions.push_back(xxxx); - m_sdk_versions.push_back(yy); - m_sdk_versions.push_back(zz); - } - success = true; - } - } - offset = load_cmd_offset + lc.cmdsize; - } - - if (success == false) - { - // Push an invalid value so we don't keep trying to - m_sdk_versions.push_back(UINT32_MAX); +uint32_t ObjectFileMachO::GetSDKVersion(uint32_t *versions, + uint32_t num_versions) { + if (m_sdk_versions.empty()) { + lldb::offset_t offset = MachHeaderSizeFromMagic(m_header.magic); + bool success = false; + for (uint32_t i = 0; success == false && i < m_header.ncmds; ++i) { + const lldb::offset_t load_cmd_offset = offset; + + version_min_command lc; + if (m_data.GetU32(&offset, &lc.cmd, 2) == NULL) + break; + if (lc.cmd == llvm::MachO::LC_VERSION_MIN_MACOSX || + lc.cmd == llvm::MachO::LC_VERSION_MIN_IPHONEOS || + lc.cmd == llvm::MachO::LC_VERSION_MIN_TVOS || + lc.cmd == llvm::MachO::LC_VERSION_MIN_WATCHOS) { + if (m_data.GetU32(&offset, &lc.version, + (sizeof(lc) / sizeof(uint32_t)) - 2)) { + const uint32_t xxxx = lc.sdk >> 16; + const uint32_t yy = (lc.sdk >> 8) & 0xffu; + const uint32_t zz = lc.sdk & 0xffu; + if (xxxx) { + m_sdk_versions.push_back(xxxx); + m_sdk_versions.push_back(yy); + m_sdk_versions.push_back(zz); + } + success = true; } + } + offset = load_cmd_offset + lc.cmdsize; } - - if (m_sdk_versions.size() > 1 || m_sdk_versions[0] != UINT32_MAX) - { - if (versions != NULL && num_versions > 0) - { - for (size_t i=0; i<num_versions; ++i) - { - if (i < m_sdk_versions.size()) - versions[i] = m_sdk_versions[i]; - else - versions[i] = 0; - } - } - return m_sdk_versions.size(); + + if (success == false) { + // Push an invalid value so we don't keep trying to + m_sdk_versions.push_back(UINT32_MAX); } - // Call the superclasses version that will empty out the data - return ObjectFile::GetSDKVersion (versions, num_versions); + } + + if (m_sdk_versions.size() > 1 || m_sdk_versions[0] != UINT32_MAX) { + if (versions != NULL && num_versions > 0) { + for (size_t i = 0; i < num_versions; ++i) { + if (i < m_sdk_versions.size()) + versions[i] = m_sdk_versions[i]; + else + versions[i] = 0; + } + } + return m_sdk_versions.size(); + } + // Call the superclasses version that will empty out the data + return ObjectFile::GetSDKVersion(versions, num_versions); } -bool -ObjectFileMachO::GetIsDynamicLinkEditor() -{ - return m_header.filetype == llvm::MachO::MH_DYLINKER; +bool ObjectFileMachO::GetIsDynamicLinkEditor() { + return m_header.filetype == llvm::MachO::MH_DYLINKER; } -bool -ObjectFileMachO::AllowAssemblyEmulationUnwindPlans () -{ - return m_allow_assembly_emulation_unwind_plans; +bool ObjectFileMachO::AllowAssemblyEmulationUnwindPlans() { + return m_allow_assembly_emulation_unwind_plans; } //------------------------------------------------------------------ // PluginInterface protocol //------------------------------------------------------------------ -lldb_private::ConstString -ObjectFileMachO::GetPluginName() -{ - return GetPluginNameStatic(); +lldb_private::ConstString ObjectFileMachO::GetPluginName() { + return GetPluginNameStatic(); } -uint32_t -ObjectFileMachO::GetPluginVersion() -{ - return 1; -} +uint32_t ObjectFileMachO::GetPluginVersion() { return 1; } -Section * -ObjectFileMachO::GetMachHeaderSection() -{ - // Find the first address of the mach header which is the first non-zero - // file sized section whose file offset is zero. This is the base file address - // of the mach-o file which can be subtracted from the vmaddr of the other - // segments found in memory and added to the load address - ModuleSP module_sp = GetModule(); - if (module_sp) - { - SectionList *section_list = GetSectionList (); - if (section_list) - { - lldb::addr_t mach_base_file_addr = LLDB_INVALID_ADDRESS; - const size_t num_sections = section_list->GetSize(); - - for (size_t sect_idx = 0; - sect_idx < num_sections && mach_base_file_addr == LLDB_INVALID_ADDRESS; - ++sect_idx) - { - Section *section = section_list->GetSectionAtIndex (sect_idx).get(); - if (section && - section->GetFileSize() > 0 && - section->GetFileOffset() == 0 && - section->IsThreadSpecific() == false && - module_sp.get() == section->GetModule().get()) - { - return section; - } - } +Section *ObjectFileMachO::GetMachHeaderSection() { + // Find the first address of the mach header which is the first non-zero + // file sized section whose file offset is zero. This is the base file address + // of the mach-o file which can be subtracted from the vmaddr of the other + // segments found in memory and added to the load address + ModuleSP module_sp = GetModule(); + if (module_sp) { + SectionList *section_list = GetSectionList(); + if (section_list) { + lldb::addr_t mach_base_file_addr = LLDB_INVALID_ADDRESS; + const size_t num_sections = section_list->GetSize(); + + for (size_t sect_idx = 0; sect_idx < num_sections && + mach_base_file_addr == LLDB_INVALID_ADDRESS; + ++sect_idx) { + Section *section = section_list->GetSectionAtIndex(sect_idx).get(); + if (section && section->GetFileSize() > 0 && + section->GetFileOffset() == 0 && + section->IsThreadSpecific() == false && + module_sp.get() == section->GetModule().get()) { + return section; } + } } - return nullptr; + } + return nullptr; } -lldb::addr_t -ObjectFileMachO::CalculateSectionLoadAddressForMemoryImage(lldb::addr_t mach_header_load_address, const Section *mach_header_section, const Section *section) -{ - ModuleSP module_sp = GetModule(); - if (module_sp && mach_header_section && section && mach_header_load_address != LLDB_INVALID_ADDRESS) - { - lldb::addr_t mach_header_file_addr = mach_header_section->GetFileAddress(); - if (mach_header_file_addr != LLDB_INVALID_ADDRESS) - { - if (section && - section->GetFileSize() > 0 && - section->IsThreadSpecific() == false && - module_sp.get() == section->GetModule().get()) - { - // Ignore __LINKEDIT and __DWARF segments - if (section->GetName() == GetSegmentNameLINKEDIT()) - { - // Only map __LINKEDIT if we have an in memory image and this isn't - // a kernel binary like a kext or mach_kernel. - const bool is_memory_image = (bool)m_process_wp.lock(); - const Strata strata = GetStrata(); - if (is_memory_image == false || strata == eStrataKernel) - return LLDB_INVALID_ADDRESS; - } - return section->GetFileAddress() - mach_header_file_addr + mach_header_load_address; - } +lldb::addr_t ObjectFileMachO::CalculateSectionLoadAddressForMemoryImage( + lldb::addr_t mach_header_load_address, const Section *mach_header_section, + const Section *section) { + ModuleSP module_sp = GetModule(); + if (module_sp && mach_header_section && section && + mach_header_load_address != LLDB_INVALID_ADDRESS) { + lldb::addr_t mach_header_file_addr = mach_header_section->GetFileAddress(); + if (mach_header_file_addr != LLDB_INVALID_ADDRESS) { + if (section && section->GetFileSize() > 0 && + section->IsThreadSpecific() == false && + module_sp.get() == section->GetModule().get()) { + // Ignore __LINKEDIT and __DWARF segments + if (section->GetName() == GetSegmentNameLINKEDIT()) { + // Only map __LINKEDIT if we have an in memory image and this isn't + // a kernel binary like a kext or mach_kernel. + const bool is_memory_image = (bool)m_process_wp.lock(); + const Strata strata = GetStrata(); + if (is_memory_image == false || strata == eStrataKernel) + return LLDB_INVALID_ADDRESS; } + return section->GetFileAddress() - mach_header_file_addr + + mach_header_load_address; + } } - return LLDB_INVALID_ADDRESS; + } + return LLDB_INVALID_ADDRESS; } -bool -ObjectFileMachO::SetLoadAddress (Target &target, - lldb::addr_t value, - bool value_is_offset) -{ - ModuleSP module_sp = GetModule(); - if (module_sp) - { - size_t num_loaded_sections = 0; - SectionList *section_list = GetSectionList (); - if (section_list) - { - const size_t num_sections = section_list->GetSize(); - - if (value_is_offset) - { - // "value" is an offset to apply to each top level segment - for (size_t sect_idx = 0; sect_idx < num_sections; ++sect_idx) - { - // Iterate through the object file sections to find all - // of the sections that size on disk (to avoid __PAGEZERO) - // and load them - SectionSP section_sp (section_list->GetSectionAtIndex (sect_idx)); - if (section_sp && - section_sp->GetFileSize() > 0 && - section_sp->IsThreadSpecific() == false && - module_sp.get() == section_sp->GetModule().get()) - { - // Ignore __LINKEDIT and __DWARF segments - if (section_sp->GetName() == GetSegmentNameLINKEDIT()) - { - // Only map __LINKEDIT if we have an in memory image and this isn't - // a kernel binary like a kext or mach_kernel. - const bool is_memory_image = (bool)m_process_wp.lock(); - const Strata strata = GetStrata(); - if (is_memory_image == false || strata == eStrataKernel) - continue; - } - if (target.GetSectionLoadList().SetSectionLoadAddress (section_sp, section_sp->GetFileAddress() + value)) - ++num_loaded_sections; - } - } +bool ObjectFileMachO::SetLoadAddress(Target &target, lldb::addr_t value, + bool value_is_offset) { + ModuleSP module_sp = GetModule(); + if (module_sp) { + size_t num_loaded_sections = 0; + SectionList *section_list = GetSectionList(); + if (section_list) { + const size_t num_sections = section_list->GetSize(); + + if (value_is_offset) { + // "value" is an offset to apply to each top level segment + for (size_t sect_idx = 0; sect_idx < num_sections; ++sect_idx) { + // Iterate through the object file sections to find all + // of the sections that size on disk (to avoid __PAGEZERO) + // and load them + SectionSP section_sp(section_list->GetSectionAtIndex(sect_idx)); + if (section_sp && section_sp->GetFileSize() > 0 && + section_sp->IsThreadSpecific() == false && + module_sp.get() == section_sp->GetModule().get()) { + // Ignore __LINKEDIT and __DWARF segments + if (section_sp->GetName() == GetSegmentNameLINKEDIT()) { + // Only map __LINKEDIT if we have an in memory image and this + // isn't + // a kernel binary like a kext or mach_kernel. + const bool is_memory_image = (bool)m_process_wp.lock(); + const Strata strata = GetStrata(); + if (is_memory_image == false || strata == eStrataKernel) + continue; } - else - { - // "value" is the new base address of the mach_header, adjust each - // section accordingly - - Section *mach_header_section = GetMachHeaderSection(); - if (mach_header_section) - { - for (size_t sect_idx = 0; sect_idx < num_sections; ++sect_idx) - { - SectionSP section_sp (section_list->GetSectionAtIndex (sect_idx)); - - lldb::addr_t section_load_addr = CalculateSectionLoadAddressForMemoryImage(value, mach_header_section, section_sp.get()); - if (section_load_addr != LLDB_INVALID_ADDRESS) - { - if (target.GetSectionLoadList().SetSectionLoadAddress (section_sp, section_load_addr)) - ++num_loaded_sections; - } - } - } + if (target.GetSectionLoadList().SetSectionLoadAddress( + section_sp, section_sp->GetFileAddress() + value)) + ++num_loaded_sections; + } + } + } else { + // "value" is the new base address of the mach_header, adjust each + // section accordingly + + Section *mach_header_section = GetMachHeaderSection(); + if (mach_header_section) { + for (size_t sect_idx = 0; sect_idx < num_sections; ++sect_idx) { + SectionSP section_sp(section_list->GetSectionAtIndex(sect_idx)); + + lldb::addr_t section_load_addr = + CalculateSectionLoadAddressForMemoryImage( + value, mach_header_section, section_sp.get()); + if (section_load_addr != LLDB_INVALID_ADDRESS) { + if (target.GetSectionLoadList().SetSectionLoadAddress( + section_sp, section_load_addr)) + ++num_loaded_sections; } + } } - return num_loaded_sections > 0; + } } - return false; + return num_loaded_sections > 0; + } + return false; } -bool -ObjectFileMachO::SaveCore (const lldb::ProcessSP &process_sp, - const FileSpec &outfile, - Error &error) -{ - if (process_sp) - { - Target &target = process_sp->GetTarget(); - const ArchSpec target_arch = target.GetArchitecture(); - const llvm::Triple &target_triple = target_arch.GetTriple(); - if (target_triple.getVendor() == llvm::Triple::Apple && - (target_triple.getOS() == llvm::Triple::MacOSX - || target_triple.getOS() == llvm::Triple::IOS - || target_triple.getOS() == llvm::Triple::WatchOS - || target_triple.getOS() == llvm::Triple::TvOS)) - { - bool make_core = false; - switch (target_arch.GetMachine()) - { - case llvm::Triple::aarch64: - case llvm::Triple::arm: - case llvm::Triple::thumb: - case llvm::Triple::x86: - case llvm::Triple::x86_64: - make_core = true; - break; - default: - error.SetErrorStringWithFormat ("unsupported core architecture: %s", target_triple.str().c_str()); - break; +bool ObjectFileMachO::SaveCore(const lldb::ProcessSP &process_sp, + const FileSpec &outfile, Error &error) { + if (process_sp) { + Target &target = process_sp->GetTarget(); + const ArchSpec target_arch = target.GetArchitecture(); + const llvm::Triple &target_triple = target_arch.GetTriple(); + if (target_triple.getVendor() == llvm::Triple::Apple && + (target_triple.getOS() == llvm::Triple::MacOSX || + target_triple.getOS() == llvm::Triple::IOS || + target_triple.getOS() == llvm::Triple::WatchOS || + target_triple.getOS() == llvm::Triple::TvOS)) { + bool make_core = false; + switch (target_arch.GetMachine()) { + case llvm::Triple::aarch64: + case llvm::Triple::arm: + case llvm::Triple::thumb: + case llvm::Triple::x86: + case llvm::Triple::x86_64: + make_core = true; + break; + default: + error.SetErrorStringWithFormat("unsupported core architecture: %s", + target_triple.str().c_str()); + break; + } + + if (make_core) { + std::vector<segment_command_64> segment_load_commands; + // uint32_t range_info_idx = 0; + MemoryRegionInfo range_info; + Error range_error = process_sp->GetMemoryRegionInfo(0, range_info); + const uint32_t addr_byte_size = target_arch.GetAddressByteSize(); + const ByteOrder byte_order = target_arch.GetByteOrder(); + if (range_error.Success()) { + while (range_info.GetRange().GetRangeBase() != LLDB_INVALID_ADDRESS) { + const addr_t addr = range_info.GetRange().GetRangeBase(); + const addr_t size = range_info.GetRange().GetByteSize(); + + if (size == 0) + break; + + // Calculate correct protections + uint32_t prot = 0; + if (range_info.GetReadable() == MemoryRegionInfo::eYes) + prot |= VM_PROT_READ; + if (range_info.GetWritable() == MemoryRegionInfo::eYes) + prot |= VM_PROT_WRITE; + if (range_info.GetExecutable() == MemoryRegionInfo::eYes) + prot |= VM_PROT_EXECUTE; + + // printf ("[%3u] [0x%16.16" PRIx64 " - + // 0x%16.16" PRIx64 ") %c%c%c\n", + // range_info_idx, + // addr, + // size, + // (prot & VM_PROT_READ ) ? 'r' : + // '-', + // (prot & VM_PROT_WRITE ) ? 'w' : + // '-', + // (prot & VM_PROT_EXECUTE) ? 'x' : + // '-'); + + if (prot != 0) { + uint32_t cmd_type = LC_SEGMENT_64; + uint32_t segment_size = sizeof(segment_command_64); + if (addr_byte_size == 4) { + cmd_type = LC_SEGMENT; + segment_size = sizeof(segment_command); + } + segment_command_64 segment = { + cmd_type, // uint32_t cmd; + segment_size, // uint32_t cmdsize; + {0}, // char segname[16]; + addr, // uint64_t vmaddr; // uint32_t for 32-bit Mach-O + size, // uint64_t vmsize; // uint32_t for 32-bit Mach-O + 0, // uint64_t fileoff; // uint32_t for 32-bit Mach-O + size, // uint64_t filesize; // uint32_t for 32-bit Mach-O + prot, // uint32_t maxprot; + prot, // uint32_t initprot; + 0, // uint32_t nsects; + 0}; // uint32_t flags; + segment_load_commands.push_back(segment); + } else { + // No protections and a size of 1 used to be returned from old + // debugservers when we asked about a region that was past the + // last memory region and it indicates the end... + if (size == 1) + break; } - - if (make_core) - { - std::vector<segment_command_64> segment_load_commands; -// uint32_t range_info_idx = 0; - MemoryRegionInfo range_info; - Error range_error = process_sp->GetMemoryRegionInfo(0, range_info); - const uint32_t addr_byte_size = target_arch.GetAddressByteSize(); - const ByteOrder byte_order = target_arch.GetByteOrder(); - if (range_error.Success()) - { - while (range_info.GetRange().GetRangeBase() != LLDB_INVALID_ADDRESS) - { - const addr_t addr = range_info.GetRange().GetRangeBase(); - const addr_t size = range_info.GetRange().GetByteSize(); - if (size == 0) - break; - - // Calculate correct protections - uint32_t prot = 0; - if (range_info.GetReadable() == MemoryRegionInfo::eYes) - prot |= VM_PROT_READ; - if (range_info.GetWritable() == MemoryRegionInfo::eYes) - prot |= VM_PROT_WRITE; - if (range_info.GetExecutable() == MemoryRegionInfo::eYes) - prot |= VM_PROT_EXECUTE; - -// printf ("[%3u] [0x%16.16" PRIx64 " - 0x%16.16" PRIx64 ") %c%c%c\n", -// range_info_idx, -// addr, -// size, -// (prot & VM_PROT_READ ) ? 'r' : '-', -// (prot & VM_PROT_WRITE ) ? 'w' : '-', -// (prot & VM_PROT_EXECUTE) ? 'x' : '-'); - - if (prot != 0) - { - uint32_t cmd_type = LC_SEGMENT_64; - uint32_t segment_size = sizeof (segment_command_64); - if (addr_byte_size == 4) - { - cmd_type = LC_SEGMENT; - segment_size = sizeof (segment_command); - } - segment_command_64 segment = { - cmd_type, // uint32_t cmd; - segment_size, // uint32_t cmdsize; - {0}, // char segname[16]; - addr, // uint64_t vmaddr; // uint32_t for 32-bit Mach-O - size, // uint64_t vmsize; // uint32_t for 32-bit Mach-O - 0, // uint64_t fileoff; // uint32_t for 32-bit Mach-O - size, // uint64_t filesize; // uint32_t for 32-bit Mach-O - prot, // uint32_t maxprot; - prot, // uint32_t initprot; - 0, // uint32_t nsects; - 0 }; // uint32_t flags; - segment_load_commands.push_back(segment); - } - else - { - // No protections and a size of 1 used to be returned from old - // debugservers when we asked about a region that was past the - // last memory region and it indicates the end... - if (size == 1) - break; - } - - range_error = process_sp->GetMemoryRegionInfo(range_info.GetRange().GetRangeEnd(), range_info); - if (range_error.Fail()) - break; - } - - StreamString buffer (Stream::eBinary, - addr_byte_size, - byte_order); + range_error = process_sp->GetMemoryRegionInfo( + range_info.GetRange().GetRangeEnd(), range_info); + if (range_error.Fail()) + break; + } + + StreamString buffer(Stream::eBinary, addr_byte_size, byte_order); + + mach_header_64 mach_header; + if (addr_byte_size == 8) { + mach_header.magic = MH_MAGIC_64; + } else { + mach_header.magic = MH_MAGIC; + } + mach_header.cputype = target_arch.GetMachOCPUType(); + mach_header.cpusubtype = target_arch.GetMachOCPUSubType(); + mach_header.filetype = MH_CORE; + mach_header.ncmds = segment_load_commands.size(); + mach_header.flags = 0; + mach_header.reserved = 0; + ThreadList &thread_list = process_sp->GetThreadList(); + const uint32_t num_threads = thread_list.GetSize(); + + // Make an array of LC_THREAD data items. Each one contains + // the contents of the LC_THREAD load command. The data doesn't + // contain the load command + load command size, we will + // add the load command and load command size as we emit the data. + std::vector<StreamString> LC_THREAD_datas(num_threads); + for (auto &LC_THREAD_data : LC_THREAD_datas) { + LC_THREAD_data.GetFlags().Set(Stream::eBinary); + LC_THREAD_data.SetAddressByteSize(addr_byte_size); + LC_THREAD_data.SetByteOrder(byte_order); + } + for (uint32_t thread_idx = 0; thread_idx < num_threads; + ++thread_idx) { + ThreadSP thread_sp(thread_list.GetThreadAtIndex(thread_idx)); + if (thread_sp) { + switch (mach_header.cputype) { + case llvm::MachO::CPU_TYPE_ARM64: + RegisterContextDarwin_arm64_Mach::Create_LC_THREAD( + thread_sp.get(), LC_THREAD_datas[thread_idx]); + break; - mach_header_64 mach_header; - if (addr_byte_size == 8) - { - mach_header.magic = MH_MAGIC_64; - } - else - { - mach_header.magic = MH_MAGIC; - } - mach_header.cputype = target_arch.GetMachOCPUType(); - mach_header.cpusubtype = target_arch.GetMachOCPUSubType(); - mach_header.filetype = MH_CORE; - mach_header.ncmds = segment_load_commands.size(); - mach_header.flags = 0; - mach_header.reserved = 0; - ThreadList &thread_list = process_sp->GetThreadList(); - const uint32_t num_threads = thread_list.GetSize(); - - // Make an array of LC_THREAD data items. Each one contains - // the contents of the LC_THREAD load command. The data doesn't - // contain the load command + load command size, we will - // add the load command and load command size as we emit the data. - std::vector<StreamString> LC_THREAD_datas(num_threads); - for (auto &LC_THREAD_data : LC_THREAD_datas) - { - LC_THREAD_data.GetFlags().Set(Stream::eBinary); - LC_THREAD_data.SetAddressByteSize(addr_byte_size); - LC_THREAD_data.SetByteOrder(byte_order); - } - for (uint32_t thread_idx = 0; thread_idx < num_threads; ++thread_idx) - { - ThreadSP thread_sp (thread_list.GetThreadAtIndex(thread_idx)); - if (thread_sp) - { - switch (mach_header.cputype) - { - case llvm::MachO::CPU_TYPE_ARM64: - RegisterContextDarwin_arm64_Mach::Create_LC_THREAD (thread_sp.get(), LC_THREAD_datas[thread_idx]); - break; - - case llvm::MachO::CPU_TYPE_ARM: - RegisterContextDarwin_arm_Mach::Create_LC_THREAD (thread_sp.get(), LC_THREAD_datas[thread_idx]); - break; - - case llvm::MachO::CPU_TYPE_I386: - RegisterContextDarwin_i386_Mach::Create_LC_THREAD (thread_sp.get(), LC_THREAD_datas[thread_idx]); - break; - - case llvm::MachO::CPU_TYPE_X86_64: - RegisterContextDarwin_x86_64_Mach::Create_LC_THREAD (thread_sp.get(), LC_THREAD_datas[thread_idx]); - break; - } - - } - } - - // The size of the load command is the size of the segments... - if (addr_byte_size == 8) - { - mach_header.sizeofcmds = segment_load_commands.size() * sizeof (struct segment_command_64); - } - else - { - mach_header.sizeofcmds = segment_load_commands.size() * sizeof (struct segment_command); - } - - // and the size of all LC_THREAD load command - for (const auto &LC_THREAD_data : LC_THREAD_datas) - { - ++mach_header.ncmds; - mach_header.sizeofcmds += 8 + LC_THREAD_data.GetSize(); - } + case llvm::MachO::CPU_TYPE_ARM: + RegisterContextDarwin_arm_Mach::Create_LC_THREAD( + thread_sp.get(), LC_THREAD_datas[thread_idx]); + break; - printf ("mach_header: 0x%8.8x 0x%8.8x 0x%8.8x 0x%8.8x 0x%8.8x 0x%8.8x 0x%8.8x 0x%8.8x\n", - mach_header.magic, - mach_header.cputype, - mach_header.cpusubtype, - mach_header.filetype, - mach_header.ncmds, - mach_header.sizeofcmds, - mach_header.flags, - mach_header.reserved); - - // Write the mach header - buffer.PutHex32(mach_header.magic); - buffer.PutHex32(mach_header.cputype); - buffer.PutHex32(mach_header.cpusubtype); - buffer.PutHex32(mach_header.filetype); - buffer.PutHex32(mach_header.ncmds); - buffer.PutHex32(mach_header.sizeofcmds); - buffer.PutHex32(mach_header.flags); - if (addr_byte_size == 8) - { - buffer.PutHex32(mach_header.reserved); - } - - // Skip the mach header and all load commands and align to the next - // 0x1000 byte boundary - addr_t file_offset = buffer.GetSize() + mach_header.sizeofcmds; - if (file_offset & 0x00000fff) - { - file_offset += 0x00001000ull; - file_offset &= (~0x00001000ull + 1); - } - - for (auto &segment : segment_load_commands) - { - segment.fileoff = file_offset; - file_offset += segment.filesize; - } - - // Write out all of the LC_THREAD load commands - for (const auto &LC_THREAD_data : LC_THREAD_datas) - { - const size_t LC_THREAD_data_size = LC_THREAD_data.GetSize(); - buffer.PutHex32(LC_THREAD); - buffer.PutHex32(8 + LC_THREAD_data_size); // cmd + cmdsize + data - buffer.Write(LC_THREAD_data.GetData(), LC_THREAD_data_size); - } + case llvm::MachO::CPU_TYPE_I386: + RegisterContextDarwin_i386_Mach::Create_LC_THREAD( + thread_sp.get(), LC_THREAD_datas[thread_idx]); + break; - // Write out all of the segment load commands - for (const auto &segment : segment_load_commands) - { - printf ("0x%8.8x 0x%8.8x [0x%16.16" PRIx64 " - 0x%16.16" PRIx64 ") [0x%16.16" PRIx64 " 0x%16.16" PRIx64 ") 0x%8.8x 0x%8.8x 0x%8.8x 0x%8.8x]\n", - segment.cmd, - segment.cmdsize, - segment.vmaddr, - segment.vmaddr + segment.vmsize, - segment.fileoff, - segment.filesize, - segment.maxprot, - segment.initprot, - segment.nsects, - segment.flags); - - buffer.PutHex32(segment.cmd); - buffer.PutHex32(segment.cmdsize); - buffer.PutRawBytes(segment.segname, sizeof(segment.segname)); - if (addr_byte_size == 8) - { - buffer.PutHex64(segment.vmaddr); - buffer.PutHex64(segment.vmsize); - buffer.PutHex64(segment.fileoff); - buffer.PutHex64(segment.filesize); - } - else - { - buffer.PutHex32(static_cast<uint32_t>(segment.vmaddr)); - buffer.PutHex32(static_cast<uint32_t>(segment.vmsize)); - buffer.PutHex32(static_cast<uint32_t>(segment.fileoff)); - buffer.PutHex32(static_cast<uint32_t>(segment.filesize)); - } - buffer.PutHex32(segment.maxprot); - buffer.PutHex32(segment.initprot); - buffer.PutHex32(segment.nsects); - buffer.PutHex32(segment.flags); - } - - File core_file; - std::string core_file_path(outfile.GetPath()); - error = core_file.Open(core_file_path.c_str(), - File::eOpenOptionWrite | - File::eOpenOptionTruncate | - File::eOpenOptionCanCreate); - if (error.Success()) - { - // Read 1 page at a time - uint8_t bytes[0x1000]; - // Write the mach header and load commands out to the core file - size_t bytes_written = buffer.GetString().size(); - error = core_file.Write(buffer.GetString().data(), bytes_written); - if (error.Success()) - { - // Now write the file data for all memory segments in the process - for (const auto &segment : segment_load_commands) - { - if (core_file.SeekFromStart(segment.fileoff) == -1) - { - error.SetErrorStringWithFormat("unable to seek to offset 0x%" PRIx64 " in '%s'", segment.fileoff, core_file_path.c_str()); - break; - } - - printf ("Saving %" PRId64 " bytes of data for memory region at 0x%" PRIx64 "\n", segment.vmsize, segment.vmaddr); - addr_t bytes_left = segment.vmsize; - addr_t addr = segment.vmaddr; - Error memory_read_error; - while (bytes_left > 0 && error.Success()) - { - const size_t bytes_to_read = bytes_left > sizeof(bytes) ? sizeof(bytes) : bytes_left; - const size_t bytes_read = process_sp->ReadMemory(addr, bytes, bytes_to_read, memory_read_error); - if (bytes_read == bytes_to_read) - { - size_t bytes_written = bytes_read; - error = core_file.Write(bytes, bytes_written); - bytes_left -= bytes_read; - addr += bytes_read; - } - else - { - // Some pages within regions are not readable, those - // should be zero filled - memset (bytes, 0, bytes_to_read); - size_t bytes_written = bytes_to_read; - error = core_file.Write(bytes, bytes_written); - bytes_left -= bytes_to_read; - addr += bytes_to_read; - } - } - } - } - } + case llvm::MachO::CPU_TYPE_X86_64: + RegisterContextDarwin_x86_64_Mach::Create_LC_THREAD( + thread_sp.get(), LC_THREAD_datas[thread_idx]); + break; + } + } + } + + // The size of the load command is the size of the segments... + if (addr_byte_size == 8) { + mach_header.sizeofcmds = segment_load_commands.size() * + sizeof(struct segment_command_64); + } else { + mach_header.sizeofcmds = + segment_load_commands.size() * sizeof(struct segment_command); + } + + // and the size of all LC_THREAD load command + for (const auto &LC_THREAD_data : LC_THREAD_datas) { + ++mach_header.ncmds; + mach_header.sizeofcmds += 8 + LC_THREAD_data.GetSize(); + } + + printf("mach_header: 0x%8.8x 0x%8.8x 0x%8.8x 0x%8.8x 0x%8.8x 0x%8.8x " + "0x%8.8x 0x%8.8x\n", + mach_header.magic, mach_header.cputype, mach_header.cpusubtype, + mach_header.filetype, mach_header.ncmds, + mach_header.sizeofcmds, mach_header.flags, + mach_header.reserved); + + // Write the mach header + buffer.PutHex32(mach_header.magic); + buffer.PutHex32(mach_header.cputype); + buffer.PutHex32(mach_header.cpusubtype); + buffer.PutHex32(mach_header.filetype); + buffer.PutHex32(mach_header.ncmds); + buffer.PutHex32(mach_header.sizeofcmds); + buffer.PutHex32(mach_header.flags); + if (addr_byte_size == 8) { + buffer.PutHex32(mach_header.reserved); + } + + // Skip the mach header and all load commands and align to the next + // 0x1000 byte boundary + addr_t file_offset = buffer.GetSize() + mach_header.sizeofcmds; + if (file_offset & 0x00000fff) { + file_offset += 0x00001000ull; + file_offset &= (~0x00001000ull + 1); + } + + for (auto &segment : segment_load_commands) { + segment.fileoff = file_offset; + file_offset += segment.filesize; + } + + // Write out all of the LC_THREAD load commands + for (const auto &LC_THREAD_data : LC_THREAD_datas) { + const size_t LC_THREAD_data_size = LC_THREAD_data.GetSize(); + buffer.PutHex32(LC_THREAD); + buffer.PutHex32(8 + LC_THREAD_data_size); // cmd + cmdsize + data + buffer.Write(LC_THREAD_data.GetString().data(), + LC_THREAD_data_size); + } + + // Write out all of the segment load commands + for (const auto &segment : segment_load_commands) { + printf("0x%8.8x 0x%8.8x [0x%16.16" PRIx64 " - 0x%16.16" PRIx64 + ") [0x%16.16" PRIx64 " 0x%16.16" PRIx64 + ") 0x%8.8x 0x%8.8x 0x%8.8x 0x%8.8x]\n", + segment.cmd, segment.cmdsize, segment.vmaddr, + segment.vmaddr + segment.vmsize, segment.fileoff, + segment.filesize, segment.maxprot, segment.initprot, + segment.nsects, segment.flags); + + buffer.PutHex32(segment.cmd); + buffer.PutHex32(segment.cmdsize); + buffer.PutRawBytes(segment.segname, sizeof(segment.segname)); + if (addr_byte_size == 8) { + buffer.PutHex64(segment.vmaddr); + buffer.PutHex64(segment.vmsize); + buffer.PutHex64(segment.fileoff); + buffer.PutHex64(segment.filesize); + } else { + buffer.PutHex32(static_cast<uint32_t>(segment.vmaddr)); + buffer.PutHex32(static_cast<uint32_t>(segment.vmsize)); + buffer.PutHex32(static_cast<uint32_t>(segment.fileoff)); + buffer.PutHex32(static_cast<uint32_t>(segment.filesize)); + } + buffer.PutHex32(segment.maxprot); + buffer.PutHex32(segment.initprot); + buffer.PutHex32(segment.nsects); + buffer.PutHex32(segment.flags); + } + + File core_file; + std::string core_file_path(outfile.GetPath()); + error = core_file.Open(core_file_path.c_str(), + File::eOpenOptionWrite | + File::eOpenOptionTruncate | + File::eOpenOptionCanCreate); + if (error.Success()) { + // Read 1 page at a time + uint8_t bytes[0x1000]; + // Write the mach header and load commands out to the core file + size_t bytes_written = buffer.GetString().size(); + error = core_file.Write(buffer.GetString().data(), bytes_written); + if (error.Success()) { + // Now write the file data for all memory segments in the process + for (const auto &segment : segment_load_commands) { + if (core_file.SeekFromStart(segment.fileoff) == -1) { + error.SetErrorStringWithFormat( + "unable to seek to offset 0x%" PRIx64 " in '%s'", + segment.fileoff, core_file_path.c_str()); + break; } - else - { - error.SetErrorString("process doesn't support getting memory region info"); + + printf("Saving %" PRId64 + " bytes of data for memory region at 0x%" PRIx64 "\n", + segment.vmsize, segment.vmaddr); + addr_t bytes_left = segment.vmsize; + addr_t addr = segment.vmaddr; + Error memory_read_error; + while (bytes_left > 0 && error.Success()) { + const size_t bytes_to_read = + bytes_left > sizeof(bytes) ? sizeof(bytes) : bytes_left; + const size_t bytes_read = process_sp->ReadMemory( + addr, bytes, bytes_to_read, memory_read_error); + if (bytes_read == bytes_to_read) { + size_t bytes_written = bytes_read; + error = core_file.Write(bytes, bytes_written); + bytes_left -= bytes_read; + addr += bytes_read; + } else { + // Some pages within regions are not readable, those + // should be zero filled + memset(bytes, 0, bytes_to_read); + size_t bytes_written = bytes_to_read; + error = core_file.Write(bytes, bytes_written); + bytes_left -= bytes_to_read; + addr += bytes_to_read; + } } + } } - return true; // This is the right plug to handle saving core files for this process + } + } else { + error.SetErrorString( + "process doesn't support getting memory region info"); } + } + return true; // This is the right plug to handle saving core files for + // this process } - return false; + } + return false; } diff --git a/source/Plugins/ObjectFile/Mach-O/ObjectFileMachO.h b/source/Plugins/ObjectFile/Mach-O/ObjectFileMachO.h index 251a0865e7823..cac176fe2ca67 100644 --- a/source/Plugins/ObjectFile/Mach-O/ObjectFileMachO.h +++ b/source/Plugins/ObjectFile/Mach-O/ObjectFileMachO.h @@ -14,242 +14,192 @@ // C++ Includes // Other libraries and framework includes // Project includes -#include "lldb/Utility/SafeMachO.h" #include "lldb/Core/Address.h" #include "lldb/Core/FileSpecList.h" #include "lldb/Core/RangeMap.h" #include "lldb/Host/FileSpec.h" #include "lldb/Symbol/ObjectFile.h" +#include "lldb/Utility/SafeMachO.h" //---------------------------------------------------------------------- // This class needs to be hidden as eventually belongs in a plugin that // will export the ObjectFile protocol //---------------------------------------------------------------------- -class ObjectFileMachO : - public lldb_private::ObjectFile -{ +class ObjectFileMachO : public lldb_private::ObjectFile { public: - ObjectFileMachO(const lldb::ModuleSP &module_sp, - lldb::DataBufferSP& data_sp, - lldb::offset_t data_offset, - const lldb_private::FileSpec* file, - lldb::offset_t offset, - lldb::offset_t length); - - ObjectFileMachO(const lldb::ModuleSP &module_sp, - lldb::DataBufferSP& data_sp, - const lldb::ProcessSP &process_sp, - lldb::addr_t header_addr); - - ~ObjectFileMachO() override = default; - - //------------------------------------------------------------------ - // Static Functions - //------------------------------------------------------------------ - static void - Initialize(); - - static void - Terminate(); - - static lldb_private::ConstString - GetPluginNameStatic(); - - static const char * - GetPluginDescriptionStatic(); - - static lldb_private::ObjectFile * - CreateInstance (const lldb::ModuleSP &module_sp, - lldb::DataBufferSP& data_sp, - lldb::offset_t data_offset, - const lldb_private::FileSpec* file, - lldb::offset_t file_offset, - lldb::offset_t length); - - static lldb_private::ObjectFile * - CreateMemoryInstance (const lldb::ModuleSP &module_sp, - lldb::DataBufferSP& data_sp, - const lldb::ProcessSP &process_sp, - lldb::addr_t header_addr); - - static size_t - GetModuleSpecifications (const lldb_private::FileSpec& file, - lldb::DataBufferSP& data_sp, - lldb::offset_t data_offset, - lldb::offset_t file_offset, - lldb::offset_t length, - lldb_private::ModuleSpecList &specs); - - static bool - SaveCore (const lldb::ProcessSP &process_sp, - const lldb_private::FileSpec &outfile, - lldb_private::Error &error); - - static bool - MagicBytesMatch (lldb::DataBufferSP& data_sp, - lldb::addr_t offset, - lldb::addr_t length); - - //------------------------------------------------------------------ - // Member Functions - //------------------------------------------------------------------ - bool - ParseHeader() override; - - bool - SetLoadAddress(lldb_private::Target &target, - lldb::addr_t value, - bool value_is_offset) override; - - lldb::ByteOrder - GetByteOrder() const override; - - bool - IsExecutable() const override; - - uint32_t - GetAddressByteSize() const override; - - lldb::AddressClass - GetAddressClass(lldb::addr_t file_addr) override; - - lldb_private::Symtab * - GetSymtab() override; - - bool - IsStripped() override; - - void - CreateSections(lldb_private::SectionList &unified_section_list) override; - - void - Dump(lldb_private::Stream *s) override; - - bool - GetArchitecture(lldb_private::ArchSpec &arch) override; - - bool - GetUUID(lldb_private::UUID* uuid) override; - - uint32_t - GetDependentModules(lldb_private::FileSpecList& files) override; - - lldb_private::FileSpecList - GetReExportedLibraries() override - { - return m_reexported_dylibs; - } - - lldb_private::Address - GetEntryPointAddress() override; - - lldb_private::Address - GetHeaderAddress() override; - - uint32_t - GetNumThreadContexts() override; - - lldb::RegisterContextSP - GetThreadContextAtIndex(uint32_t idx, lldb_private::Thread &thread) override; - - ObjectFile::Type - CalculateType() override; - - ObjectFile::Strata - CalculateStrata() override; - - uint32_t - GetVersion(uint32_t *versions, uint32_t num_versions) override; - - uint32_t - GetMinimumOSVersion(uint32_t *versions, uint32_t num_versions) override; - - uint32_t - GetSDKVersion(uint32_t *versions, uint32_t num_versions) override; - - bool - GetIsDynamicLinkEditor() override; - - static bool - ParseHeader (lldb_private::DataExtractor &data, - lldb::offset_t *data_offset_ptr, - llvm::MachO::mach_header &header); - - bool - AllowAssemblyEmulationUnwindPlans () override; - - //------------------------------------------------------------------ - // PluginInterface protocol - //------------------------------------------------------------------ - lldb_private::ConstString - GetPluginName() override; - - uint32_t - GetPluginVersion() override; + ObjectFileMachO(const lldb::ModuleSP &module_sp, lldb::DataBufferSP &data_sp, + lldb::offset_t data_offset, + const lldb_private::FileSpec *file, lldb::offset_t offset, + lldb::offset_t length); + + ObjectFileMachO(const lldb::ModuleSP &module_sp, lldb::DataBufferSP &data_sp, + const lldb::ProcessSP &process_sp, lldb::addr_t header_addr); + + ~ObjectFileMachO() override = default; + + //------------------------------------------------------------------ + // Static Functions + //------------------------------------------------------------------ + static void Initialize(); + + static void Terminate(); + + static lldb_private::ConstString GetPluginNameStatic(); + + static const char *GetPluginDescriptionStatic(); + + static lldb_private::ObjectFile * + CreateInstance(const lldb::ModuleSP &module_sp, lldb::DataBufferSP &data_sp, + lldb::offset_t data_offset, const lldb_private::FileSpec *file, + lldb::offset_t file_offset, lldb::offset_t length); + + static lldb_private::ObjectFile *CreateMemoryInstance( + const lldb::ModuleSP &module_sp, lldb::DataBufferSP &data_sp, + const lldb::ProcessSP &process_sp, lldb::addr_t header_addr); + + static size_t GetModuleSpecifications(const lldb_private::FileSpec &file, + lldb::DataBufferSP &data_sp, + lldb::offset_t data_offset, + lldb::offset_t file_offset, + lldb::offset_t length, + lldb_private::ModuleSpecList &specs); + + static bool SaveCore(const lldb::ProcessSP &process_sp, + const lldb_private::FileSpec &outfile, + lldb_private::Error &error); + + static bool MagicBytesMatch(lldb::DataBufferSP &data_sp, lldb::addr_t offset, + lldb::addr_t length); + + //------------------------------------------------------------------ + // Member Functions + //------------------------------------------------------------------ + bool ParseHeader() override; + + bool SetLoadAddress(lldb_private::Target &target, lldb::addr_t value, + bool value_is_offset) override; + + lldb::ByteOrder GetByteOrder() const override; + + bool IsExecutable() const override; + + uint32_t GetAddressByteSize() const override; + + lldb::AddressClass GetAddressClass(lldb::addr_t file_addr) override; + + lldb_private::Symtab *GetSymtab() override; + + bool IsStripped() override; + + void CreateSections(lldb_private::SectionList &unified_section_list) override; + + void Dump(lldb_private::Stream *s) override; + + bool GetArchitecture(lldb_private::ArchSpec &arch) override; + + bool GetUUID(lldb_private::UUID *uuid) override; + + uint32_t GetDependentModules(lldb_private::FileSpecList &files) override; + + lldb_private::FileSpecList GetReExportedLibraries() override { + return m_reexported_dylibs; + } + + lldb_private::Address GetEntryPointAddress() override; + + lldb_private::Address GetHeaderAddress() override; + + uint32_t GetNumThreadContexts() override; + + lldb::RegisterContextSP + GetThreadContextAtIndex(uint32_t idx, lldb_private::Thread &thread) override; + + ObjectFile::Type CalculateType() override; + + ObjectFile::Strata CalculateStrata() override; + + uint32_t GetVersion(uint32_t *versions, uint32_t num_versions) override; + + uint32_t GetMinimumOSVersion(uint32_t *versions, + uint32_t num_versions) override; + + uint32_t GetSDKVersion(uint32_t *versions, uint32_t num_versions) override; + + bool GetIsDynamicLinkEditor() override; + + static bool ParseHeader(lldb_private::DataExtractor &data, + lldb::offset_t *data_offset_ptr, + llvm::MachO::mach_header &header); + + bool AllowAssemblyEmulationUnwindPlans() override; + + //------------------------------------------------------------------ + // PluginInterface protocol + //------------------------------------------------------------------ + lldb_private::ConstString GetPluginName() override; + + uint32_t GetPluginVersion() override; protected: - static bool - GetUUID (const llvm::MachO::mach_header &header, - const lldb_private::DataExtractor &data, - lldb::offset_t lc_offset, // Offset to the first load command - lldb_private::UUID& uuid); - - static bool - GetArchitecture (const llvm::MachO::mach_header &header, - const lldb_private::DataExtractor &data, - lldb::offset_t lc_offset, - lldb_private::ArchSpec &arch); - - // Intended for same-host arm device debugging where lldb needs to - // detect libraries in the shared cache and augment the nlist entries - // with an on-disk dyld_shared_cache file. The process will record - // the shared cache UUID so the on-disk cache can be matched or rejected - // correctly. - lldb_private::UUID - GetProcessSharedCacheUUID (lldb_private::Process *); - - // Intended for same-host arm device debugging where lldb will read - // shared cache libraries out of its own memory instead of the remote - // process' memory as an optimization. If lldb's shared cache UUID - // does not match the process' shared cache UUID, this optimization - // should not be used. - lldb_private::UUID - GetLLDBSharedCacheUUID (); - - lldb_private::Section * - GetMachHeaderSection(); - - lldb::addr_t - CalculateSectionLoadAddressForMemoryImage(lldb::addr_t mach_header_load_address, - const lldb_private::Section *mach_header_section, - const lldb_private::Section *section); - - lldb_private::UUID - GetSharedCacheUUID (lldb_private::FileSpec dyld_shared_cache, const lldb::ByteOrder byte_order, const uint32_t addr_byte_size); - - size_t - ParseSymtab(); - - llvm::MachO::mach_header m_header; - static const lldb_private::ConstString &GetSegmentNameTEXT(); - static const lldb_private::ConstString &GetSegmentNameDATA(); - static const lldb_private::ConstString &GetSegmentNameDATA_DIRTY(); - static const lldb_private::ConstString &GetSegmentNameDATA_CONST(); - static const lldb_private::ConstString &GetSegmentNameOBJC(); - static const lldb_private::ConstString &GetSegmentNameLINKEDIT(); - static const lldb_private::ConstString &GetSectionNameEHFrame(); - - llvm::MachO::dysymtab_command m_dysymtab; - std::vector<llvm::MachO::segment_command_64> m_mach_segments; - std::vector<llvm::MachO::section_64> m_mach_sections; - std::vector<uint32_t> m_min_os_versions; - std::vector<uint32_t> m_sdk_versions; - typedef lldb_private::RangeVector<uint32_t, uint32_t> FileRangeArray; - lldb_private::Address m_entry_point_address; - FileRangeArray m_thread_context_offsets; - bool m_thread_context_offsets_valid; - lldb_private::FileSpecList m_reexported_dylibs; - bool m_allow_assembly_emulation_unwind_plans; + static bool + GetUUID(const llvm::MachO::mach_header &header, + const lldb_private::DataExtractor &data, + lldb::offset_t lc_offset, // Offset to the first load command + lldb_private::UUID &uuid); + + static bool GetArchitecture(const llvm::MachO::mach_header &header, + const lldb_private::DataExtractor &data, + lldb::offset_t lc_offset, + lldb_private::ArchSpec &arch); + + // Intended for same-host arm device debugging where lldb needs to + // detect libraries in the shared cache and augment the nlist entries + // with an on-disk dyld_shared_cache file. The process will record + // the shared cache UUID so the on-disk cache can be matched or rejected + // correctly. + lldb_private::UUID GetProcessSharedCacheUUID(lldb_private::Process *); + + // Intended for same-host arm device debugging where lldb will read + // shared cache libraries out of its own memory instead of the remote + // process' memory as an optimization. If lldb's shared cache UUID + // does not match the process' shared cache UUID, this optimization + // should not be used. + lldb_private::UUID GetLLDBSharedCacheUUID(); + + lldb_private::Section *GetMachHeaderSection(); + + lldb::addr_t CalculateSectionLoadAddressForMemoryImage( + lldb::addr_t mach_header_load_address, + const lldb_private::Section *mach_header_section, + const lldb_private::Section *section); + + lldb_private::UUID + GetSharedCacheUUID(lldb_private::FileSpec dyld_shared_cache, + const lldb::ByteOrder byte_order, + const uint32_t addr_byte_size); + + size_t ParseSymtab(); + + llvm::MachO::mach_header m_header; + static const lldb_private::ConstString &GetSegmentNameTEXT(); + static const lldb_private::ConstString &GetSegmentNameDATA(); + static const lldb_private::ConstString &GetSegmentNameDATA_DIRTY(); + static const lldb_private::ConstString &GetSegmentNameDATA_CONST(); + static const lldb_private::ConstString &GetSegmentNameOBJC(); + static const lldb_private::ConstString &GetSegmentNameLINKEDIT(); + static const lldb_private::ConstString &GetSectionNameEHFrame(); + + llvm::MachO::dysymtab_command m_dysymtab; + std::vector<llvm::MachO::segment_command_64> m_mach_segments; + std::vector<llvm::MachO::section_64> m_mach_sections; + std::vector<uint32_t> m_min_os_versions; + std::vector<uint32_t> m_sdk_versions; + typedef lldb_private::RangeVector<uint32_t, uint32_t> FileRangeArray; + lldb_private::Address m_entry_point_address; + FileRangeArray m_thread_context_offsets; + bool m_thread_context_offsets_valid; + lldb_private::FileSpecList m_reexported_dylibs; + bool m_allow_assembly_emulation_unwind_plans; }; #endif // liblldb_ObjectFileMachO_h_ diff --git a/source/Plugins/ObjectFile/PECOFF/ObjectFilePECOFF.cpp b/source/Plugins/ObjectFile/PECOFF/ObjectFilePECOFF.cpp index 5f5a0ab07ad69..f0647e02158be 100644 --- a/source/Plugins/ObjectFile/PECOFF/ObjectFilePECOFF.cpp +++ b/source/Plugins/ObjectFile/PECOFF/ObjectFilePECOFF.cpp @@ -14,7 +14,7 @@ #include "lldb/Core/ArchSpec.h" #include "lldb/Core/DataBuffer.h" -#include "lldb/Host/FileSpec.h" +#include "lldb/Core/DataBufferHeap.h" #include "lldb/Core/FileSpecList.h" #include "lldb/Core/Module.h" #include "lldb/Core/ModuleSpec.h" @@ -24,858 +24,818 @@ #include "lldb/Core/StreamString.h" #include "lldb/Core/Timer.h" #include "lldb/Core/UUID.h" +#include "lldb/Host/FileSpec.h" #include "lldb/Symbol/ObjectFile.h" #include "lldb/Target/Process.h" #include "lldb/Target/SectionLoadList.h" #include "lldb/Target/Target.h" -#define IMAGE_DOS_SIGNATURE 0x5A4D // MZ -#define IMAGE_NT_SIGNATURE 0x00004550 // PE00 -#define OPT_HEADER_MAGIC_PE32 0x010b -#define OPT_HEADER_MAGIC_PE32_PLUS 0x020b +#define IMAGE_DOS_SIGNATURE 0x5A4D // MZ +#define IMAGE_NT_SIGNATURE 0x00004550 // PE00 +#define OPT_HEADER_MAGIC_PE32 0x010b +#define OPT_HEADER_MAGIC_PE32_PLUS 0x020b using namespace lldb; using namespace lldb_private; -void -ObjectFilePECOFF::Initialize() -{ - PluginManager::RegisterPlugin (GetPluginNameStatic(), - GetPluginDescriptionStatic(), - CreateInstance, - CreateMemoryInstance, - GetModuleSpecifications, - SaveCore); +void ObjectFilePECOFF::Initialize() { + PluginManager::RegisterPlugin( + GetPluginNameStatic(), GetPluginDescriptionStatic(), CreateInstance, + CreateMemoryInstance, GetModuleSpecifications, SaveCore); } -void -ObjectFilePECOFF::Terminate() -{ - PluginManager::UnregisterPlugin (CreateInstance); +void ObjectFilePECOFF::Terminate() { + PluginManager::UnregisterPlugin(CreateInstance); } +lldb_private::ConstString ObjectFilePECOFF::GetPluginNameStatic() { + static ConstString g_name("pe-coff"); + return g_name; +} -lldb_private::ConstString -ObjectFilePECOFF::GetPluginNameStatic() -{ - static ConstString g_name("pe-coff"); - return g_name; +const char *ObjectFilePECOFF::GetPluginDescriptionStatic() { + return "Portable Executable and Common Object File Format object file reader " + "(32 and 64 bit)"; } -const char * -ObjectFilePECOFF::GetPluginDescriptionStatic() -{ - return "Portable Executable and Common Object File Format object file reader (32 and 64 bit)"; +ObjectFile *ObjectFilePECOFF::CreateInstance(const lldb::ModuleSP &module_sp, + DataBufferSP &data_sp, + lldb::offset_t data_offset, + const lldb_private::FileSpec *file, + lldb::offset_t file_offset, + lldb::offset_t length) { + if (!data_sp) { + data_sp = file->MemoryMapFileContentsIfLocal(file_offset, length); + data_offset = 0; + } + + if (ObjectFilePECOFF::MagicBytesMatch(data_sp)) { + // Update the data to contain the entire file if it doesn't already + if (data_sp->GetByteSize() < length) + data_sp = file->MemoryMapFileContentsIfLocal(file_offset, length); + std::unique_ptr<ObjectFile> objfile_ap(new ObjectFilePECOFF( + module_sp, data_sp, data_offset, file, file_offset, length)); + if (objfile_ap.get() && objfile_ap->ParseHeader()) + return objfile_ap.release(); + } + return NULL; } +ObjectFile *ObjectFilePECOFF::CreateMemoryInstance( + const lldb::ModuleSP &module_sp, lldb::DataBufferSP &data_sp, + const lldb::ProcessSP &process_sp, lldb::addr_t header_addr) { + if (!data_sp || !ObjectFilePECOFF::MagicBytesMatch(data_sp)) + return nullptr; + auto objfile_ap = llvm::make_unique<ObjectFilePECOFF>( + module_sp, data_sp, process_sp, header_addr); + if (objfile_ap.get() && objfile_ap->ParseHeader()) { + return objfile_ap.release(); + } + return nullptr; +} -ObjectFile * -ObjectFilePECOFF::CreateInstance (const lldb::ModuleSP &module_sp, - DataBufferSP& data_sp, - lldb::offset_t data_offset, - const lldb_private::FileSpec* file, - lldb::offset_t file_offset, - lldb::offset_t length) -{ - if (!data_sp) - { - data_sp = file->MemoryMapFileContentsIfLocal(file_offset, length); - data_offset = 0; - } +size_t ObjectFilePECOFF::GetModuleSpecifications( + const lldb_private::FileSpec &file, lldb::DataBufferSP &data_sp, + lldb::offset_t data_offset, lldb::offset_t file_offset, + lldb::offset_t length, lldb_private::ModuleSpecList &specs) { + const size_t initial_count = specs.GetSize(); - if (ObjectFilePECOFF::MagicBytesMatch(data_sp)) - { - // Update the data to contain the entire file if it doesn't already - if (data_sp->GetByteSize() < length) - data_sp = file->MemoryMapFileContentsIfLocal(file_offset, length); - std::unique_ptr<ObjectFile> objfile_ap(new ObjectFilePECOFF (module_sp, data_sp, data_offset, file, file_offset, length)); - if (objfile_ap.get() && objfile_ap->ParseHeader()) - return objfile_ap.release(); - } - return NULL; -} + if (ObjectFilePECOFF::MagicBytesMatch(data_sp)) { + DataExtractor data; + data.SetData(data_sp, data_offset, length); + data.SetByteOrder(eByteOrderLittle); -ObjectFile * -ObjectFilePECOFF::CreateMemoryInstance (const lldb::ModuleSP &module_sp, - lldb::DataBufferSP& data_sp, - const lldb::ProcessSP &process_sp, - lldb::addr_t header_addr) -{ - return NULL; -} + dos_header_t dos_header; + coff_header_t coff_header; -size_t -ObjectFilePECOFF::GetModuleSpecifications (const lldb_private::FileSpec& file, - lldb::DataBufferSP& data_sp, - lldb::offset_t data_offset, - lldb::offset_t file_offset, - lldb::offset_t length, - lldb_private::ModuleSpecList &specs) -{ - const size_t initial_count = specs.GetSize(); - - if (ObjectFilePECOFF::MagicBytesMatch(data_sp)) - { - DataExtractor data; - data.SetData(data_sp, data_offset, length); - data.SetByteOrder(eByteOrderLittle); - - dos_header_t dos_header; - coff_header_t coff_header; - - if (ParseDOSHeader(data, dos_header)) - { - lldb::offset_t offset = dos_header.e_lfanew; - uint32_t pe_signature = data.GetU32(&offset); - if (pe_signature != IMAGE_NT_SIGNATURE) - return false; - if (ParseCOFFHeader(data, &offset, coff_header)) - { - ArchSpec spec; - if (coff_header.machine == MachineAmd64) - { - spec.SetTriple("x86_64-pc-windows"); - specs.Append(ModuleSpec(file, spec)); - } - else if (coff_header.machine == MachineX86) - { - spec.SetTriple("i386-pc-windows"); - specs.Append(ModuleSpec(file, spec)); - spec.SetTriple("i686-pc-windows"); - specs.Append(ModuleSpec(file, spec)); - } - } + if (ParseDOSHeader(data, dos_header)) { + lldb::offset_t offset = dos_header.e_lfanew; + uint32_t pe_signature = data.GetU32(&offset); + if (pe_signature != IMAGE_NT_SIGNATURE) + return false; + if (ParseCOFFHeader(data, &offset, coff_header)) { + ArchSpec spec; + if (coff_header.machine == MachineAmd64) { + spec.SetTriple("x86_64-pc-windows"); + specs.Append(ModuleSpec(file, spec)); + } else if (coff_header.machine == MachineX86) { + spec.SetTriple("i386-pc-windows"); + specs.Append(ModuleSpec(file, spec)); + spec.SetTriple("i686-pc-windows"); + specs.Append(ModuleSpec(file, spec)); } + } } + } - return specs.GetSize() - initial_count; + return specs.GetSize() - initial_count; } -bool -ObjectFilePECOFF::SaveCore(const lldb::ProcessSP &process_sp, - const lldb_private::FileSpec &outfile, - lldb_private::Error &error) -{ - return SaveMiniDump(process_sp, outfile, error); +bool ObjectFilePECOFF::SaveCore(const lldb::ProcessSP &process_sp, + const lldb_private::FileSpec &outfile, + lldb_private::Error &error) { + return SaveMiniDump(process_sp, outfile, error); } - -bool -ObjectFilePECOFF::MagicBytesMatch (DataBufferSP& data_sp) -{ - DataExtractor data(data_sp, eByteOrderLittle, 4); - lldb::offset_t offset = 0; - uint16_t magic = data.GetU16 (&offset); - return magic == IMAGE_DOS_SIGNATURE; +bool ObjectFilePECOFF::MagicBytesMatch(DataBufferSP &data_sp) { + DataExtractor data(data_sp, eByteOrderLittle, 4); + lldb::offset_t offset = 0; + uint16_t magic = data.GetU16(&offset); + return magic == IMAGE_DOS_SIGNATURE; } -lldb::SymbolType -ObjectFilePECOFF::MapSymbolType(uint16_t coff_symbol_type) -{ - // TODO: We need to complete this mapping of COFF symbol types to LLDB ones. - // For now, here's a hack to make sure our function have types. - const auto complex_type = coff_symbol_type >> llvm::COFF::SCT_COMPLEX_TYPE_SHIFT; - if (complex_type == llvm::COFF::IMAGE_SYM_DTYPE_FUNCTION) - { - return lldb::eSymbolTypeCode; - } - return lldb::eSymbolTypeInvalid; +lldb::SymbolType ObjectFilePECOFF::MapSymbolType(uint16_t coff_symbol_type) { + // TODO: We need to complete this mapping of COFF symbol types to LLDB ones. + // For now, here's a hack to make sure our function have types. + const auto complex_type = + coff_symbol_type >> llvm::COFF::SCT_COMPLEX_TYPE_SHIFT; + if (complex_type == llvm::COFF::IMAGE_SYM_DTYPE_FUNCTION) { + return lldb::eSymbolTypeCode; + } + return lldb::eSymbolTypeInvalid; } -ObjectFilePECOFF::ObjectFilePECOFF (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) : - ObjectFile (module_sp, file, file_offset, length, data_sp, data_offset), - m_dos_header (), - m_coff_header (), - m_coff_header_opt (), - m_sect_headers (), - m_entry_point_address () -{ - ::memset (&m_dos_header, 0, sizeof(m_dos_header)); - ::memset (&m_coff_header, 0, sizeof(m_coff_header)); - ::memset (&m_coff_header_opt, 0, sizeof(m_coff_header_opt)); +ObjectFilePECOFF::ObjectFilePECOFF(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) + : ObjectFile(module_sp, file, file_offset, length, data_sp, data_offset), + m_dos_header(), m_coff_header(), m_coff_header_opt(), m_sect_headers(), + m_entry_point_address() { + ::memset(&m_dos_header, 0, sizeof(m_dos_header)); + ::memset(&m_coff_header, 0, sizeof(m_coff_header)); + ::memset(&m_coff_header_opt, 0, sizeof(m_coff_header_opt)); } - -ObjectFilePECOFF::~ObjectFilePECOFF() -{ +ObjectFilePECOFF::ObjectFilePECOFF(const lldb::ModuleSP &module_sp, + DataBufferSP &header_data_sp, + const lldb::ProcessSP &process_sp, + addr_t header_addr) + : ObjectFile(module_sp, process_sp, header_addr, header_data_sp), + m_dos_header(), m_coff_header(), m_coff_header_opt(), m_sect_headers(), + m_entry_point_address() { + ::memset(&m_dos_header, 0, sizeof(m_dos_header)); + ::memset(&m_coff_header, 0, sizeof(m_coff_header)); + ::memset(&m_coff_header_opt, 0, sizeof(m_coff_header_opt)); } +ObjectFilePECOFF::~ObjectFilePECOFF() {} -bool -ObjectFilePECOFF::ParseHeader () -{ - ModuleSP module_sp(GetModule()); - if (module_sp) - { - std::lock_guard<std::recursive_mutex> guard(module_sp->GetMutex()); - m_sect_headers.clear(); - m_data.SetByteOrder (eByteOrderLittle); - lldb::offset_t offset = 0; - - if (ParseDOSHeader(m_data, m_dos_header)) - { - offset = m_dos_header.e_lfanew; - uint32_t pe_signature = m_data.GetU32 (&offset); - if (pe_signature != IMAGE_NT_SIGNATURE) - return false; - if (ParseCOFFHeader(m_data, &offset, m_coff_header)) - { - if (m_coff_header.hdrsize > 0) - ParseCOFFOptionalHeader(&offset); - ParseSectionHeaders (offset); - } - return true; - } +bool ObjectFilePECOFF::ParseHeader() { + ModuleSP module_sp(GetModule()); + if (module_sp) { + std::lock_guard<std::recursive_mutex> guard(module_sp->GetMutex()); + m_sect_headers.clear(); + m_data.SetByteOrder(eByteOrderLittle); + lldb::offset_t offset = 0; + + if (ParseDOSHeader(m_data, m_dos_header)) { + offset = m_dos_header.e_lfanew; + uint32_t pe_signature = m_data.GetU32(&offset); + if (pe_signature != IMAGE_NT_SIGNATURE) + return false; + if (ParseCOFFHeader(m_data, &offset, m_coff_header)) { + if (m_coff_header.hdrsize > 0) + ParseCOFFOptionalHeader(&offset); + ParseSectionHeaders(offset); + } + return true; } - return false; + } + return false; } -bool -ObjectFilePECOFF::SetLoadAddress(Target &target, addr_t value, bool value_is_offset) -{ - bool changed = false; - ModuleSP module_sp = GetModule(); - if (module_sp) - { - size_t num_loaded_sections = 0; - SectionList *section_list = GetSectionList (); - if (section_list) - { - if (!value_is_offset) - { - value -= m_image_base; - } - - const size_t num_sections = section_list->GetSize(); - size_t sect_idx = 0; - - for (sect_idx = 0; sect_idx < num_sections; ++sect_idx) - { - // Iterate through the object file sections to find all - // of the sections that have SHF_ALLOC in their flag bits. - SectionSP section_sp (section_list->GetSectionAtIndex (sect_idx)); - if (section_sp && !section_sp->IsThreadSpecific()) - { - if (target.GetSectionLoadList().SetSectionLoadAddress (section_sp, section_sp->GetFileAddress() + value)) - ++num_loaded_sections; - } - } - changed = num_loaded_sections > 0; +bool ObjectFilePECOFF::SetLoadAddress(Target &target, addr_t value, + bool value_is_offset) { + bool changed = false; + ModuleSP module_sp = GetModule(); + if (module_sp) { + size_t num_loaded_sections = 0; + SectionList *section_list = GetSectionList(); + if (section_list) { + if (!value_is_offset) { + value -= m_image_base; + } + + const size_t num_sections = section_list->GetSize(); + size_t sect_idx = 0; + + for (sect_idx = 0; sect_idx < num_sections; ++sect_idx) { + // Iterate through the object file sections to find all + // of the sections that have SHF_ALLOC in their flag bits. + SectionSP section_sp(section_list->GetSectionAtIndex(sect_idx)); + if (section_sp && !section_sp->IsThreadSpecific()) { + if (target.GetSectionLoadList().SetSectionLoadAddress( + section_sp, section_sp->GetFileAddress() + value)) + ++num_loaded_sections; } + } + changed = num_loaded_sections > 0; } - return changed; + } + return changed; } +ByteOrder ObjectFilePECOFF::GetByteOrder() const { return eByteOrderLittle; } -ByteOrder -ObjectFilePECOFF::GetByteOrder () const -{ - return eByteOrderLittle; -} - -bool -ObjectFilePECOFF::IsExecutable() const -{ - return (m_coff_header.flags & llvm::COFF::IMAGE_FILE_DLL) == 0; +bool ObjectFilePECOFF::IsExecutable() const { + return (m_coff_header.flags & llvm::COFF::IMAGE_FILE_DLL) == 0; } -uint32_t -ObjectFilePECOFF::GetAddressByteSize () const -{ - if (m_coff_header_opt.magic == OPT_HEADER_MAGIC_PE32_PLUS) - return 8; - else if (m_coff_header_opt.magic == OPT_HEADER_MAGIC_PE32) - return 4; +uint32_t ObjectFilePECOFF::GetAddressByteSize() const { + if (m_coff_header_opt.magic == OPT_HEADER_MAGIC_PE32_PLUS) + return 8; + else if (m_coff_header_opt.magic == OPT_HEADER_MAGIC_PE32) return 4; + return 4; } //---------------------------------------------------------------------- // NeedsEndianSwap // -// Return true if an endian swap needs to occur when extracting data +// Return true if an endian swap needs to occur when extracting data // from this file. //---------------------------------------------------------------------- -bool -ObjectFilePECOFF::NeedsEndianSwap() const -{ +bool ObjectFilePECOFF::NeedsEndianSwap() const { #if defined(__LITTLE_ENDIAN__) - return false; + return false; #else - return true; + return true; #endif } //---------------------------------------------------------------------- // ParseDOSHeader //---------------------------------------------------------------------- -bool -ObjectFilePECOFF::ParseDOSHeader (DataExtractor &data, dos_header_t &dos_header) -{ - bool success = false; - lldb::offset_t offset = 0; - success = data.ValidOffsetForDataOfSize(0, sizeof(dos_header)); - - if (success) - { - dos_header.e_magic = data.GetU16(&offset); // Magic number - success = dos_header.e_magic == IMAGE_DOS_SIGNATURE; - - if (success) - { - dos_header.e_cblp = data.GetU16(&offset); // Bytes on last page of file - dos_header.e_cp = data.GetU16(&offset); // Pages in file - dos_header.e_crlc = data.GetU16(&offset); // Relocations - dos_header.e_cparhdr = data.GetU16(&offset); // Size of header in paragraphs - dos_header.e_minalloc = data.GetU16(&offset); // Minimum extra paragraphs needed - dos_header.e_maxalloc = data.GetU16(&offset); // Maximum extra paragraphs needed - dos_header.e_ss = data.GetU16(&offset); // Initial (relative) SS value - dos_header.e_sp = data.GetU16(&offset); // Initial SP value - dos_header.e_csum = data.GetU16(&offset); // Checksum - dos_header.e_ip = data.GetU16(&offset); // Initial IP value - dos_header.e_cs = data.GetU16(&offset); // Initial (relative) CS value - dos_header.e_lfarlc = data.GetU16(&offset); // File address of relocation table - dos_header.e_ovno = data.GetU16(&offset); // Overlay number - - dos_header.e_res[0] = data.GetU16(&offset); // Reserved words - dos_header.e_res[1] = data.GetU16(&offset); // Reserved words - dos_header.e_res[2] = data.GetU16(&offset); // Reserved words - dos_header.e_res[3] = data.GetU16(&offset); // Reserved words - - dos_header.e_oemid = data.GetU16(&offset); // OEM identifier (for e_oeminfo) - dos_header.e_oeminfo = data.GetU16(&offset); // OEM information; e_oemid specific - dos_header.e_res2[0] = data.GetU16(&offset); // Reserved words - dos_header.e_res2[1] = data.GetU16(&offset); // Reserved words - dos_header.e_res2[2] = data.GetU16(&offset); // Reserved words - dos_header.e_res2[3] = data.GetU16(&offset); // Reserved words - dos_header.e_res2[4] = data.GetU16(&offset); // Reserved words - dos_header.e_res2[5] = data.GetU16(&offset); // Reserved words - dos_header.e_res2[6] = data.GetU16(&offset); // Reserved words - dos_header.e_res2[7] = data.GetU16(&offset); // Reserved words - dos_header.e_res2[8] = data.GetU16(&offset); // Reserved words - dos_header.e_res2[9] = data.GetU16(&offset); // Reserved words - - dos_header.e_lfanew = data.GetU32(&offset); // File address of new exe header - } +bool ObjectFilePECOFF::ParseDOSHeader(DataExtractor &data, + dos_header_t &dos_header) { + bool success = false; + lldb::offset_t offset = 0; + success = data.ValidOffsetForDataOfSize(0, sizeof(dos_header)); + + if (success) { + dos_header.e_magic = data.GetU16(&offset); // Magic number + success = dos_header.e_magic == IMAGE_DOS_SIGNATURE; + + if (success) { + dos_header.e_cblp = data.GetU16(&offset); // Bytes on last page of file + dos_header.e_cp = data.GetU16(&offset); // Pages in file + dos_header.e_crlc = data.GetU16(&offset); // Relocations + dos_header.e_cparhdr = + data.GetU16(&offset); // Size of header in paragraphs + dos_header.e_minalloc = + data.GetU16(&offset); // Minimum extra paragraphs needed + dos_header.e_maxalloc = + data.GetU16(&offset); // Maximum extra paragraphs needed + dos_header.e_ss = data.GetU16(&offset); // Initial (relative) SS value + dos_header.e_sp = data.GetU16(&offset); // Initial SP value + dos_header.e_csum = data.GetU16(&offset); // Checksum + dos_header.e_ip = data.GetU16(&offset); // Initial IP value + dos_header.e_cs = data.GetU16(&offset); // Initial (relative) CS value + dos_header.e_lfarlc = + data.GetU16(&offset); // File address of relocation table + dos_header.e_ovno = data.GetU16(&offset); // Overlay number + + dos_header.e_res[0] = data.GetU16(&offset); // Reserved words + dos_header.e_res[1] = data.GetU16(&offset); // Reserved words + dos_header.e_res[2] = data.GetU16(&offset); // Reserved words + dos_header.e_res[3] = data.GetU16(&offset); // Reserved words + + dos_header.e_oemid = + data.GetU16(&offset); // OEM identifier (for e_oeminfo) + dos_header.e_oeminfo = + data.GetU16(&offset); // OEM information; e_oemid specific + dos_header.e_res2[0] = data.GetU16(&offset); // Reserved words + dos_header.e_res2[1] = data.GetU16(&offset); // Reserved words + dos_header.e_res2[2] = data.GetU16(&offset); // Reserved words + dos_header.e_res2[3] = data.GetU16(&offset); // Reserved words + dos_header.e_res2[4] = data.GetU16(&offset); // Reserved words + dos_header.e_res2[5] = data.GetU16(&offset); // Reserved words + dos_header.e_res2[6] = data.GetU16(&offset); // Reserved words + dos_header.e_res2[7] = data.GetU16(&offset); // Reserved words + dos_header.e_res2[8] = data.GetU16(&offset); // Reserved words + dos_header.e_res2[9] = data.GetU16(&offset); // Reserved words + + dos_header.e_lfanew = + data.GetU32(&offset); // File address of new exe header } - if (!success) - memset(&dos_header, 0, sizeof(dos_header)); - return success; + } + if (!success) + memset(&dos_header, 0, sizeof(dos_header)); + return success; } - //---------------------------------------------------------------------- // ParserCOFFHeader //---------------------------------------------------------------------- -bool -ObjectFilePECOFF::ParseCOFFHeader(DataExtractor &data, lldb::offset_t *offset_ptr, coff_header_t &coff_header) -{ - bool success = data.ValidOffsetForDataOfSize (*offset_ptr, sizeof(coff_header)); - if (success) - { - coff_header.machine = data.GetU16(offset_ptr); - coff_header.nsects = data.GetU16(offset_ptr); - coff_header.modtime = data.GetU32(offset_ptr); - coff_header.symoff = data.GetU32(offset_ptr); - coff_header.nsyms = data.GetU32(offset_ptr); - coff_header.hdrsize = data.GetU16(offset_ptr); - coff_header.flags = data.GetU16(offset_ptr); - } - if (!success) - memset(&coff_header, 0, sizeof(coff_header)); - return success; +bool ObjectFilePECOFF::ParseCOFFHeader(DataExtractor &data, + lldb::offset_t *offset_ptr, + coff_header_t &coff_header) { + bool success = + data.ValidOffsetForDataOfSize(*offset_ptr, sizeof(coff_header)); + if (success) { + coff_header.machine = data.GetU16(offset_ptr); + coff_header.nsects = data.GetU16(offset_ptr); + coff_header.modtime = data.GetU32(offset_ptr); + coff_header.symoff = data.GetU32(offset_ptr); + coff_header.nsyms = data.GetU32(offset_ptr); + coff_header.hdrsize = data.GetU16(offset_ptr); + coff_header.flags = data.GetU16(offset_ptr); + } + if (!success) + memset(&coff_header, 0, sizeof(coff_header)); + return success; } -bool -ObjectFilePECOFF::ParseCOFFOptionalHeader(lldb::offset_t *offset_ptr) -{ - bool success = false; - const lldb::offset_t end_offset = *offset_ptr + m_coff_header.hdrsize; - if (*offset_ptr < end_offset) - { - success = true; - m_coff_header_opt.magic = m_data.GetU16(offset_ptr); - m_coff_header_opt.major_linker_version = m_data.GetU8 (offset_ptr); - m_coff_header_opt.minor_linker_version = m_data.GetU8 (offset_ptr); - m_coff_header_opt.code_size = m_data.GetU32(offset_ptr); - m_coff_header_opt.data_size = m_data.GetU32(offset_ptr); - m_coff_header_opt.bss_size = m_data.GetU32(offset_ptr); - m_coff_header_opt.entry = m_data.GetU32(offset_ptr); - m_coff_header_opt.code_offset = m_data.GetU32(offset_ptr); - - const uint32_t addr_byte_size = GetAddressByteSize (); - - if (*offset_ptr < end_offset) - { - if (m_coff_header_opt.magic == OPT_HEADER_MAGIC_PE32) - { - // PE32 only - m_coff_header_opt.data_offset = m_data.GetU32(offset_ptr); - } - else - m_coff_header_opt.data_offset = 0; - - if (*offset_ptr < end_offset) - { - m_coff_header_opt.image_base = m_data.GetMaxU64 (offset_ptr, addr_byte_size); - m_coff_header_opt.sect_alignment = m_data.GetU32(offset_ptr); - m_coff_header_opt.file_alignment = m_data.GetU32(offset_ptr); - m_coff_header_opt.major_os_system_version = m_data.GetU16(offset_ptr); - m_coff_header_opt.minor_os_system_version = m_data.GetU16(offset_ptr); - m_coff_header_opt.major_image_version = m_data.GetU16(offset_ptr); - m_coff_header_opt.minor_image_version = m_data.GetU16(offset_ptr); - m_coff_header_opt.major_subsystem_version = m_data.GetU16(offset_ptr); - m_coff_header_opt.minor_subsystem_version = m_data.GetU16(offset_ptr); - m_coff_header_opt.reserved1 = m_data.GetU32(offset_ptr); - m_coff_header_opt.image_size = m_data.GetU32(offset_ptr); - m_coff_header_opt.header_size = m_data.GetU32(offset_ptr); - m_coff_header_opt.checksum = m_data.GetU32(offset_ptr); - m_coff_header_opt.subsystem = m_data.GetU16(offset_ptr); - m_coff_header_opt.dll_flags = m_data.GetU16(offset_ptr); - m_coff_header_opt.stack_reserve_size = m_data.GetMaxU64 (offset_ptr, addr_byte_size); - m_coff_header_opt.stack_commit_size = m_data.GetMaxU64 (offset_ptr, addr_byte_size); - m_coff_header_opt.heap_reserve_size = m_data.GetMaxU64 (offset_ptr, addr_byte_size); - m_coff_header_opt.heap_commit_size = m_data.GetMaxU64 (offset_ptr, addr_byte_size); - m_coff_header_opt.loader_flags = m_data.GetU32(offset_ptr); - uint32_t num_data_dir_entries = m_data.GetU32(offset_ptr); - m_coff_header_opt.data_dirs.clear(); - m_coff_header_opt.data_dirs.resize(num_data_dir_entries); - uint32_t i; - for (i=0; i<num_data_dir_entries; i++) - { - m_coff_header_opt.data_dirs[i].vmaddr = m_data.GetU32(offset_ptr); - m_coff_header_opt.data_dirs[i].vmsize = m_data.GetU32(offset_ptr); - } - - m_file_offset = m_coff_header_opt.image_base; - m_image_base = m_coff_header_opt.image_base; - } +bool ObjectFilePECOFF::ParseCOFFOptionalHeader(lldb::offset_t *offset_ptr) { + bool success = false; + const lldb::offset_t end_offset = *offset_ptr + m_coff_header.hdrsize; + if (*offset_ptr < end_offset) { + success = true; + m_coff_header_opt.magic = m_data.GetU16(offset_ptr); + m_coff_header_opt.major_linker_version = m_data.GetU8(offset_ptr); + m_coff_header_opt.minor_linker_version = m_data.GetU8(offset_ptr); + m_coff_header_opt.code_size = m_data.GetU32(offset_ptr); + m_coff_header_opt.data_size = m_data.GetU32(offset_ptr); + m_coff_header_opt.bss_size = m_data.GetU32(offset_ptr); + m_coff_header_opt.entry = m_data.GetU32(offset_ptr); + m_coff_header_opt.code_offset = m_data.GetU32(offset_ptr); + + const uint32_t addr_byte_size = GetAddressByteSize(); + + if (*offset_ptr < end_offset) { + if (m_coff_header_opt.magic == OPT_HEADER_MAGIC_PE32) { + // PE32 only + m_coff_header_opt.data_offset = m_data.GetU32(offset_ptr); + } else + m_coff_header_opt.data_offset = 0; + + if (*offset_ptr < end_offset) { + m_coff_header_opt.image_base = + m_data.GetMaxU64(offset_ptr, addr_byte_size); + m_coff_header_opt.sect_alignment = m_data.GetU32(offset_ptr); + m_coff_header_opt.file_alignment = m_data.GetU32(offset_ptr); + m_coff_header_opt.major_os_system_version = m_data.GetU16(offset_ptr); + m_coff_header_opt.minor_os_system_version = m_data.GetU16(offset_ptr); + m_coff_header_opt.major_image_version = m_data.GetU16(offset_ptr); + m_coff_header_opt.minor_image_version = m_data.GetU16(offset_ptr); + m_coff_header_opt.major_subsystem_version = m_data.GetU16(offset_ptr); + m_coff_header_opt.minor_subsystem_version = m_data.GetU16(offset_ptr); + m_coff_header_opt.reserved1 = m_data.GetU32(offset_ptr); + m_coff_header_opt.image_size = m_data.GetU32(offset_ptr); + m_coff_header_opt.header_size = m_data.GetU32(offset_ptr); + m_coff_header_opt.checksum = m_data.GetU32(offset_ptr); + m_coff_header_opt.subsystem = m_data.GetU16(offset_ptr); + m_coff_header_opt.dll_flags = m_data.GetU16(offset_ptr); + m_coff_header_opt.stack_reserve_size = + m_data.GetMaxU64(offset_ptr, addr_byte_size); + m_coff_header_opt.stack_commit_size = + m_data.GetMaxU64(offset_ptr, addr_byte_size); + m_coff_header_opt.heap_reserve_size = + m_data.GetMaxU64(offset_ptr, addr_byte_size); + m_coff_header_opt.heap_commit_size = + m_data.GetMaxU64(offset_ptr, addr_byte_size); + m_coff_header_opt.loader_flags = m_data.GetU32(offset_ptr); + uint32_t num_data_dir_entries = m_data.GetU32(offset_ptr); + m_coff_header_opt.data_dirs.clear(); + m_coff_header_opt.data_dirs.resize(num_data_dir_entries); + uint32_t i; + for (i = 0; i < num_data_dir_entries; i++) { + m_coff_header_opt.data_dirs[i].vmaddr = m_data.GetU32(offset_ptr); + m_coff_header_opt.data_dirs[i].vmsize = m_data.GetU32(offset_ptr); } + + m_file_offset = m_coff_header_opt.image_base; + m_image_base = m_coff_header_opt.image_base; + } } - // Make sure we are on track for section data which follows - *offset_ptr = end_offset; - return success; + } + // Make sure we are on track for section data which follows + *offset_ptr = end_offset; + return success; } +DataExtractor ObjectFilePECOFF::ReadImageData(uint32_t offset, size_t size) { + if (m_file) { + DataBufferSP buffer_sp(m_file.ReadFileContents(offset, size)); + return DataExtractor(buffer_sp, GetByteOrder(), GetAddressByteSize()); + } + ProcessSP process_sp(m_process_wp.lock()); + DataExtractor data; + if (process_sp) { + auto data_ap = llvm::make_unique<DataBufferHeap>(size, 0); + Error readmem_error; + size_t bytes_read = + process_sp->ReadMemory(m_image_base + offset, data_ap->GetBytes(), + data_ap->GetByteSize(), readmem_error); + if (bytes_read == size) { + DataBufferSP buffer_sp(data_ap.release()); + data.SetData(buffer_sp, 0, buffer_sp->GetByteSize()); + } + } + return data; +} //---------------------------------------------------------------------- // ParseSectionHeaders //---------------------------------------------------------------------- -bool -ObjectFilePECOFF::ParseSectionHeaders (uint32_t section_header_data_offset) -{ - const uint32_t nsects = m_coff_header.nsects; - m_sect_headers.clear(); - - if (nsects > 0) - { - const uint32_t addr_byte_size = GetAddressByteSize (); - const size_t section_header_byte_size = nsects * sizeof(section_header_t); - DataBufferSP section_header_data_sp(m_file.ReadFileContents (section_header_data_offset, section_header_byte_size)); - DataExtractor section_header_data (section_header_data_sp, GetByteOrder(), addr_byte_size); +bool ObjectFilePECOFF::ParseSectionHeaders( + uint32_t section_header_data_offset) { + const uint32_t nsects = m_coff_header.nsects; + m_sect_headers.clear(); - lldb::offset_t offset = 0; - if (section_header_data.ValidOffsetForDataOfSize (offset, section_header_byte_size)) - { - m_sect_headers.resize(nsects); - - for (uint32_t idx = 0; idx<nsects; ++idx) - { - const void *name_data = section_header_data.GetData(&offset, 8); - if (name_data) - { - memcpy(m_sect_headers[idx].name, name_data, 8); - m_sect_headers[idx].vmsize = section_header_data.GetU32(&offset); - m_sect_headers[idx].vmaddr = section_header_data.GetU32(&offset); - m_sect_headers[idx].size = section_header_data.GetU32(&offset); - m_sect_headers[idx].offset = section_header_data.GetU32(&offset); - m_sect_headers[idx].reloff = section_header_data.GetU32(&offset); - m_sect_headers[idx].lineoff = section_header_data.GetU32(&offset); - m_sect_headers[idx].nreloc = section_header_data.GetU16(&offset); - m_sect_headers[idx].nline = section_header_data.GetU16(&offset); - m_sect_headers[idx].flags = section_header_data.GetU32(&offset); - } - } + if (nsects > 0) { + const size_t section_header_byte_size = nsects * sizeof(section_header_t); + DataExtractor section_header_data = + ReadImageData(section_header_data_offset, section_header_byte_size); + + lldb::offset_t offset = 0; + if (section_header_data.ValidOffsetForDataOfSize( + offset, section_header_byte_size)) { + m_sect_headers.resize(nsects); + + for (uint32_t idx = 0; idx < nsects; ++idx) { + const void *name_data = section_header_data.GetData(&offset, 8); + if (name_data) { + memcpy(m_sect_headers[idx].name, name_data, 8); + m_sect_headers[idx].vmsize = section_header_data.GetU32(&offset); + m_sect_headers[idx].vmaddr = section_header_data.GetU32(&offset); + m_sect_headers[idx].size = section_header_data.GetU32(&offset); + m_sect_headers[idx].offset = section_header_data.GetU32(&offset); + m_sect_headers[idx].reloff = section_header_data.GetU32(&offset); + m_sect_headers[idx].lineoff = section_header_data.GetU32(&offset); + m_sect_headers[idx].nreloc = section_header_data.GetU16(&offset); + m_sect_headers[idx].nline = section_header_data.GetU16(&offset); + m_sect_headers[idx].flags = section_header_data.GetU32(&offset); } + } } - - return m_sect_headers.empty() == false; + } + + return m_sect_headers.empty() == false; } -bool -ObjectFilePECOFF::GetSectionName(std::string& sect_name, const section_header_t& sect) -{ - if (sect.name[0] == '/') - { - lldb::offset_t stroff = strtoul(§.name[1], NULL, 10); - lldb::offset_t string_file_offset = m_coff_header.symoff + (m_coff_header.nsyms * 18) + stroff; - const char *name = m_data.GetCStr (&string_file_offset); - if (name) - { - sect_name = name; - return true; - } - - return false; +bool ObjectFilePECOFF::GetSectionName(std::string §_name, + const section_header_t §) { + if (sect.name[0] == '/') { + lldb::offset_t stroff = strtoul(§.name[1], NULL, 10); + lldb::offset_t string_file_offset = + m_coff_header.symoff + (m_coff_header.nsyms * 18) + stroff; + const char *name = m_data.GetCStr(&string_file_offset); + if (name) { + sect_name = name; + return true; } - sect_name = sect.name; - return true; -} + + return false; + } + sect_name = sect.name; + return true; +} //---------------------------------------------------------------------- // GetNListSymtab //---------------------------------------------------------------------- -Symtab * -ObjectFilePECOFF::GetSymtab() -{ - ModuleSP module_sp(GetModule()); - if (module_sp) - { - std::lock_guard<std::recursive_mutex> guard(module_sp->GetMutex()); - if (m_symtab_ap.get() == NULL) - { - SectionList *sect_list = GetSectionList(); - m_symtab_ap.reset(new Symtab(this)); - std::lock_guard<std::recursive_mutex> guard(m_symtab_ap->GetMutex()); - - const uint32_t num_syms = m_coff_header.nsyms; - - if (num_syms > 0 && m_coff_header.symoff > 0) - { - const uint32_t symbol_size = 18; - const uint32_t addr_byte_size = GetAddressByteSize (); - const size_t symbol_data_size = num_syms * symbol_size; - // Include the 4-byte string table size at the end of the symbols - DataBufferSP symtab_data_sp(m_file.ReadFileContents (m_coff_header.symoff, symbol_data_size + 4)); - DataExtractor symtab_data (symtab_data_sp, GetByteOrder(), addr_byte_size); - lldb::offset_t offset = symbol_data_size; - const uint32_t strtab_size = symtab_data.GetU32 (&offset); - DataBufferSP strtab_data_sp(m_file.ReadFileContents (m_coff_header.symoff + symbol_data_size, strtab_size)); - DataExtractor strtab_data (strtab_data_sp, GetByteOrder(), addr_byte_size); - - // First 4 bytes should be zeroed after strtab_size has been read, - // because it is used as offset 0 to encode a NULL string. - uint32_t* strtab_data_start = (uint32_t*)strtab_data_sp->GetBytes(); - strtab_data_start[0] = 0; - - offset = 0; - std::string symbol_name; - Symbol *symbols = m_symtab_ap->Resize (num_syms); - for (uint32_t i=0; i<num_syms; ++i) - { - coff_symbol_t symbol; - const uint32_t symbol_offset = offset; - const char *symbol_name_cstr = NULL; - // If the first 4 bytes of the symbol string are zero, then they - // are followed by a 4-byte string table offset. Else these - // 8 bytes contain the symbol name - if (symtab_data.GetU32 (&offset) == 0) - { - // Long string that doesn't fit into the symbol table name, - // so now we must read the 4 byte string table offset - uint32_t strtab_offset = symtab_data.GetU32 (&offset); - symbol_name_cstr = strtab_data.PeekCStr (strtab_offset); - symbol_name.assign (symbol_name_cstr); - } - else - { - // Short string that fits into the symbol table name which is 8 bytes - offset += sizeof(symbol.name) - 4; // Skip remaining - symbol_name_cstr = symtab_data.PeekCStr (symbol_offset); - if (symbol_name_cstr == NULL) - break; - symbol_name.assign (symbol_name_cstr, sizeof(symbol.name)); - } - symbol.value = symtab_data.GetU32 (&offset); - symbol.sect = symtab_data.GetU16 (&offset); - symbol.type = symtab_data.GetU16 (&offset); - symbol.storage = symtab_data.GetU8 (&offset); - symbol.naux = symtab_data.GetU8 (&offset); - symbols[i].GetMangled ().SetValue (ConstString(symbol_name.c_str())); - if ((int16_t)symbol.sect >= 1) - { - Address symbol_addr(sect_list->GetSectionAtIndex(symbol.sect-1), symbol.value); - symbols[i].GetAddressRef() = symbol_addr; - symbols[i].SetType(MapSymbolType(symbol.type)); - } - - if (symbol.naux > 0) - { - i += symbol.naux; - offset += symbol_size; - } - } - +Symtab *ObjectFilePECOFF::GetSymtab() { + ModuleSP module_sp(GetModule()); + if (module_sp) { + std::lock_guard<std::recursive_mutex> guard(module_sp->GetMutex()); + if (m_symtab_ap.get() == NULL) { + SectionList *sect_list = GetSectionList(); + m_symtab_ap.reset(new Symtab(this)); + std::lock_guard<std::recursive_mutex> guard(m_symtab_ap->GetMutex()); + + const uint32_t num_syms = m_coff_header.nsyms; + + if (m_file && num_syms > 0 && m_coff_header.symoff > 0) { + const uint32_t symbol_size = 18; + const size_t symbol_data_size = num_syms * symbol_size; + // Include the 4-byte string table size at the end of the symbols + DataExtractor symtab_data = + ReadImageData(m_coff_header.symoff, symbol_data_size + 4); + lldb::offset_t offset = symbol_data_size; + const uint32_t strtab_size = symtab_data.GetU32(&offset); + if (strtab_size > 0) { + DataExtractor strtab_data = ReadImageData( + m_coff_header.symoff + symbol_data_size, strtab_size); + + // First 4 bytes should be zeroed after strtab_size has been read, + // because it is used as offset 0 to encode a NULL string. + uint32_t *strtab_data_start = (uint32_t *)strtab_data.GetDataStart(); + strtab_data_start[0] = 0; + + offset = 0; + std::string symbol_name; + Symbol *symbols = m_symtab_ap->Resize(num_syms); + for (uint32_t i = 0; i < num_syms; ++i) { + coff_symbol_t symbol; + const uint32_t symbol_offset = offset; + const char *symbol_name_cstr = NULL; + // If the first 4 bytes of the symbol string are zero, then they + // are followed by a 4-byte string table offset. Else these + // 8 bytes contain the symbol name + if (symtab_data.GetU32(&offset) == 0) { + // Long string that doesn't fit into the symbol table name, + // so now we must read the 4 byte string table offset + uint32_t strtab_offset = symtab_data.GetU32(&offset); + symbol_name_cstr = strtab_data.PeekCStr(strtab_offset); + symbol_name.assign(symbol_name_cstr); + } else { + // Short string that fits into the symbol table name which is 8 + // bytes + offset += sizeof(symbol.name) - 4; // Skip remaining + symbol_name_cstr = symtab_data.PeekCStr(symbol_offset); + if (symbol_name_cstr == NULL) + break; + symbol_name.assign(symbol_name_cstr, sizeof(symbol.name)); + } + symbol.value = symtab_data.GetU32(&offset); + symbol.sect = symtab_data.GetU16(&offset); + symbol.type = symtab_data.GetU16(&offset); + symbol.storage = symtab_data.GetU8(&offset); + symbol.naux = symtab_data.GetU8(&offset); + symbols[i].GetMangled().SetValue(ConstString(symbol_name.c_str())); + if ((int16_t)symbol.sect >= 1) { + Address symbol_addr(sect_list->GetSectionAtIndex(symbol.sect - 1), + symbol.value); + symbols[i].GetAddressRef() = symbol_addr; + symbols[i].SetType(MapSymbolType(symbol.type)); } - // Read export header - if (coff_data_dir_export_table < m_coff_header_opt.data_dirs.size() - && m_coff_header_opt.data_dirs[coff_data_dir_export_table].vmsize > 0 && m_coff_header_opt.data_dirs[coff_data_dir_export_table].vmaddr > 0) - { - export_directory_entry export_table; - uint32_t data_start = m_coff_header_opt.data_dirs[coff_data_dir_export_table].vmaddr; - Address address(m_coff_header_opt.image_base + data_start, sect_list); - DataBufferSP symtab_data_sp(m_file.ReadFileContents(address.GetSection()->GetFileOffset() + address.GetOffset(), m_coff_header_opt.data_dirs[0].vmsize)); - DataExtractor symtab_data (symtab_data_sp, GetByteOrder(), GetAddressByteSize()); - lldb::offset_t offset = 0; - - // Read export_table header - export_table.characteristics = symtab_data.GetU32(&offset); - export_table.time_date_stamp = symtab_data.GetU32(&offset); - export_table.major_version = symtab_data.GetU16(&offset); - export_table.minor_version = symtab_data.GetU16(&offset); - export_table.name = symtab_data.GetU32(&offset); - export_table.base = symtab_data.GetU32(&offset); - export_table.number_of_functions = symtab_data.GetU32(&offset); - export_table.number_of_names = symtab_data.GetU32(&offset); - export_table.address_of_functions = symtab_data.GetU32(&offset); - export_table.address_of_names = symtab_data.GetU32(&offset); - export_table.address_of_name_ordinals = symtab_data.GetU32(&offset); - - bool has_ordinal = export_table.address_of_name_ordinals != 0; - - lldb::offset_t name_offset = export_table.address_of_names - data_start; - lldb::offset_t name_ordinal_offset = export_table.address_of_name_ordinals - data_start; - - Symbol *symbols = m_symtab_ap->Resize(export_table.number_of_names); - - std::string symbol_name; - - // Read each export table entry - for (size_t i = 0; i < export_table.number_of_names; ++i) - { - uint32_t name_ordinal = has_ordinal ? symtab_data.GetU16(&name_ordinal_offset) : i; - uint32_t name_address = symtab_data.GetU32(&name_offset); - - const char* symbol_name_cstr = symtab_data.PeekCStr(name_address - data_start); - symbol_name.assign(symbol_name_cstr); - - lldb::offset_t function_offset = export_table.address_of_functions - data_start + sizeof(uint32_t) * name_ordinal; - uint32_t function_rva = symtab_data.GetU32(&function_offset); - - Address symbol_addr(m_coff_header_opt.image_base + function_rva, sect_list); - symbols[i].GetMangled().SetValue(ConstString(symbol_name.c_str())); - symbols[i].GetAddressRef() = symbol_addr; - symbols[i].SetType(lldb::eSymbolTypeCode); - symbols[i].SetDebug(true); - } + if (symbol.naux > 0) { + i += symbol.naux; + offset += symbol_size; } - m_symtab_ap->CalculateSymbolSizes(); + } } - } - return m_symtab_ap.get(); + } + + // Read export header + if (coff_data_dir_export_table < m_coff_header_opt.data_dirs.size() && + m_coff_header_opt.data_dirs[coff_data_dir_export_table].vmsize > 0 && + m_coff_header_opt.data_dirs[coff_data_dir_export_table].vmaddr > 0) { + export_directory_entry export_table; + uint32_t data_start = + m_coff_header_opt.data_dirs[coff_data_dir_export_table].vmaddr; + + uint32_t address_rva = data_start; + if (m_file) { + Address address(m_coff_header_opt.image_base + data_start, sect_list); + address_rva = + address.GetSection()->GetFileOffset() + address.GetOffset(); + } + DataExtractor symtab_data = + ReadImageData(address_rva, m_coff_header_opt.data_dirs[0].vmsize); + lldb::offset_t offset = 0; + // Read export_table header + export_table.characteristics = symtab_data.GetU32(&offset); + export_table.time_date_stamp = symtab_data.GetU32(&offset); + export_table.major_version = symtab_data.GetU16(&offset); + export_table.minor_version = symtab_data.GetU16(&offset); + export_table.name = symtab_data.GetU32(&offset); + export_table.base = symtab_data.GetU32(&offset); + export_table.number_of_functions = symtab_data.GetU32(&offset); + export_table.number_of_names = symtab_data.GetU32(&offset); + export_table.address_of_functions = symtab_data.GetU32(&offset); + export_table.address_of_names = symtab_data.GetU32(&offset); + export_table.address_of_name_ordinals = symtab_data.GetU32(&offset); + + bool has_ordinal = export_table.address_of_name_ordinals != 0; + + lldb::offset_t name_offset = export_table.address_of_names - data_start; + lldb::offset_t name_ordinal_offset = + export_table.address_of_name_ordinals - data_start; + + Symbol *symbols = m_symtab_ap->Resize(export_table.number_of_names); + + std::string symbol_name; + + // Read each export table entry + for (size_t i = 0; i < export_table.number_of_names; ++i) { + uint32_t name_ordinal = + has_ordinal ? symtab_data.GetU16(&name_ordinal_offset) : i; + uint32_t name_address = symtab_data.GetU32(&name_offset); + + const char *symbol_name_cstr = + symtab_data.PeekCStr(name_address - data_start); + symbol_name.assign(symbol_name_cstr); + + lldb::offset_t function_offset = export_table.address_of_functions - + data_start + + sizeof(uint32_t) * name_ordinal; + uint32_t function_rva = symtab_data.GetU32(&function_offset); + + Address symbol_addr(m_coff_header_opt.image_base + function_rva, + sect_list); + symbols[i].GetMangled().SetValue(ConstString(symbol_name.c_str())); + symbols[i].GetAddressRef() = symbol_addr; + symbols[i].SetType(lldb::eSymbolTypeCode); + symbols[i].SetDebug(true); + } + } + m_symtab_ap->CalculateSymbolSizes(); + } + } + return m_symtab_ap.get(); } -bool -ObjectFilePECOFF::IsStripped () -{ - // TODO: determine this for COFF - return false; +bool ObjectFilePECOFF::IsStripped() { + // TODO: determine this for COFF + return false; } +void ObjectFilePECOFF::CreateSections(SectionList &unified_section_list) { + if (!m_sections_ap.get()) { + m_sections_ap.reset(new SectionList()); - -void -ObjectFilePECOFF::CreateSections (SectionList &unified_section_list) -{ - if (!m_sections_ap.get()) - { - m_sections_ap.reset(new SectionList()); - - ModuleSP module_sp(GetModule()); - if (module_sp) - { - std::lock_guard<std::recursive_mutex> guard(module_sp->GetMutex()); - const uint32_t nsects = m_sect_headers.size(); - ModuleSP module_sp (GetModule()); - for (uint32_t idx = 0; idx<nsects; ++idx) - { - std::string sect_name; - GetSectionName (sect_name, m_sect_headers[idx]); - ConstString const_sect_name (sect_name.c_str()); - static ConstString g_code_sect_name (".code"); - static ConstString g_CODE_sect_name ("CODE"); - static ConstString g_data_sect_name (".data"); - static ConstString g_DATA_sect_name ("DATA"); - static ConstString g_bss_sect_name (".bss"); - static ConstString g_BSS_sect_name ("BSS"); - static ConstString g_debug_sect_name (".debug"); - static ConstString g_reloc_sect_name (".reloc"); - static ConstString g_stab_sect_name (".stab"); - static ConstString g_stabstr_sect_name (".stabstr"); - static ConstString g_sect_name_dwarf_debug_abbrev (".debug_abbrev"); - static ConstString g_sect_name_dwarf_debug_aranges (".debug_aranges"); - static ConstString g_sect_name_dwarf_debug_frame (".debug_frame"); - static ConstString g_sect_name_dwarf_debug_info (".debug_info"); - static ConstString g_sect_name_dwarf_debug_line (".debug_line"); - static ConstString g_sect_name_dwarf_debug_loc (".debug_loc"); - static ConstString g_sect_name_dwarf_debug_macinfo (".debug_macinfo"); - static ConstString g_sect_name_dwarf_debug_pubnames (".debug_pubnames"); - static ConstString g_sect_name_dwarf_debug_pubtypes (".debug_pubtypes"); - static ConstString g_sect_name_dwarf_debug_ranges (".debug_ranges"); - static ConstString g_sect_name_dwarf_debug_str (".debug_str"); - static ConstString g_sect_name_eh_frame (".eh_frame"); - static ConstString g_sect_name_go_symtab (".gosymtab"); - SectionType section_type = eSectionTypeOther; - if (m_sect_headers[idx].flags & llvm::COFF::IMAGE_SCN_CNT_CODE && - ((const_sect_name == g_code_sect_name) || (const_sect_name == g_CODE_sect_name))) - { - section_type = eSectionTypeCode; - } - else if (m_sect_headers[idx].flags & llvm::COFF::IMAGE_SCN_CNT_INITIALIZED_DATA && - ((const_sect_name == g_data_sect_name) || (const_sect_name == g_DATA_sect_name))) - { - section_type = eSectionTypeData; - } - else if (m_sect_headers[idx].flags & llvm::COFF::IMAGE_SCN_CNT_UNINITIALIZED_DATA && - ((const_sect_name == g_bss_sect_name) || (const_sect_name == g_BSS_sect_name))) - { - if (m_sect_headers[idx].size == 0) - section_type = eSectionTypeZeroFill; - else - section_type = eSectionTypeData; - } - else if (const_sect_name == g_debug_sect_name) - { - section_type = eSectionTypeDebug; - } - else if (const_sect_name == g_stabstr_sect_name) - { - section_type = eSectionTypeDataCString; - } - else if (const_sect_name == g_reloc_sect_name) - { - section_type = eSectionTypeOther; - } - else if (const_sect_name == g_sect_name_dwarf_debug_abbrev) section_type = eSectionTypeDWARFDebugAbbrev; - else if (const_sect_name == g_sect_name_dwarf_debug_aranges) section_type = eSectionTypeDWARFDebugAranges; - else if (const_sect_name == g_sect_name_dwarf_debug_frame) section_type = eSectionTypeDWARFDebugFrame; - else if (const_sect_name == g_sect_name_dwarf_debug_info) section_type = eSectionTypeDWARFDebugInfo; - else if (const_sect_name == g_sect_name_dwarf_debug_line) section_type = eSectionTypeDWARFDebugLine; - else if (const_sect_name == g_sect_name_dwarf_debug_loc) section_type = eSectionTypeDWARFDebugLoc; - else if (const_sect_name == g_sect_name_dwarf_debug_macinfo) section_type = eSectionTypeDWARFDebugMacInfo; - else if (const_sect_name == g_sect_name_dwarf_debug_pubnames) section_type = eSectionTypeDWARFDebugPubNames; - else if (const_sect_name == g_sect_name_dwarf_debug_pubtypes) section_type = eSectionTypeDWARFDebugPubTypes; - else if (const_sect_name == g_sect_name_dwarf_debug_ranges) section_type = eSectionTypeDWARFDebugRanges; - else if (const_sect_name == g_sect_name_dwarf_debug_str) section_type = eSectionTypeDWARFDebugStr; - else if (const_sect_name == g_sect_name_eh_frame) section_type = eSectionTypeEHFrame; - else if (const_sect_name == g_sect_name_go_symtab) section_type = eSectionTypeGoSymtab; - else if (m_sect_headers[idx].flags & llvm::COFF::IMAGE_SCN_CNT_CODE) - { - section_type = eSectionTypeCode; - } - else if (m_sect_headers[idx].flags & llvm::COFF::IMAGE_SCN_CNT_INITIALIZED_DATA) - { - section_type = eSectionTypeData; - } - else if (m_sect_headers[idx].flags & llvm::COFF::IMAGE_SCN_CNT_UNINITIALIZED_DATA) - { - if (m_sect_headers[idx].size == 0) - section_type = eSectionTypeZeroFill; - else - section_type = eSectionTypeData; - } - - // Use a segment ID of the segment index shifted left by 8 so they - // never conflict with any of the sections. - SectionSP section_sp (new Section (module_sp, // Module to which this section belongs - this, // Object file to which this section belongs - idx + 1, // Section ID is the 1 based segment index shifted right by 8 bits as not to collide with any of the 256 section IDs that are possible - const_sect_name, // Name of this section - section_type, // This section is a container of other sections. - m_coff_header_opt.image_base + m_sect_headers[idx].vmaddr, // File VM address == addresses as they are found in the object file - m_sect_headers[idx].vmsize, // VM size in bytes of this section - m_sect_headers[idx].offset, // Offset to the data for this section in the file - m_sect_headers[idx].size, // Size in bytes of this section as found in the file - m_coff_header_opt.sect_alignment, // Section alignment - m_sect_headers[idx].flags)); // Flags for this section - - //section_sp->SetIsEncrypted (segment_is_encrypted); - - unified_section_list.AddSection(section_sp); - m_sections_ap->AddSection (section_sp); - } + ModuleSP module_sp(GetModule()); + if (module_sp) { + std::lock_guard<std::recursive_mutex> guard(module_sp->GetMutex()); + const uint32_t nsects = m_sect_headers.size(); + ModuleSP module_sp(GetModule()); + for (uint32_t idx = 0; idx < nsects; ++idx) { + std::string sect_name; + GetSectionName(sect_name, m_sect_headers[idx]); + ConstString const_sect_name(sect_name.c_str()); + static ConstString g_code_sect_name(".code"); + static ConstString g_CODE_sect_name("CODE"); + static ConstString g_data_sect_name(".data"); + static ConstString g_DATA_sect_name("DATA"); + static ConstString g_bss_sect_name(".bss"); + static ConstString g_BSS_sect_name("BSS"); + static ConstString g_debug_sect_name(".debug"); + static ConstString g_reloc_sect_name(".reloc"); + static ConstString g_stab_sect_name(".stab"); + static ConstString g_stabstr_sect_name(".stabstr"); + static ConstString g_sect_name_dwarf_debug_abbrev(".debug_abbrev"); + static ConstString g_sect_name_dwarf_debug_aranges(".debug_aranges"); + static ConstString g_sect_name_dwarf_debug_frame(".debug_frame"); + static ConstString g_sect_name_dwarf_debug_info(".debug_info"); + static ConstString g_sect_name_dwarf_debug_line(".debug_line"); + static ConstString g_sect_name_dwarf_debug_loc(".debug_loc"); + static ConstString g_sect_name_dwarf_debug_macinfo(".debug_macinfo"); + static ConstString g_sect_name_dwarf_debug_pubnames(".debug_pubnames"); + static ConstString g_sect_name_dwarf_debug_pubtypes(".debug_pubtypes"); + static ConstString g_sect_name_dwarf_debug_ranges(".debug_ranges"); + static ConstString g_sect_name_dwarf_debug_str(".debug_str"); + static ConstString g_sect_name_eh_frame(".eh_frame"); + static ConstString g_sect_name_go_symtab(".gosymtab"); + SectionType section_type = eSectionTypeOther; + if (m_sect_headers[idx].flags & llvm::COFF::IMAGE_SCN_CNT_CODE && + ((const_sect_name == g_code_sect_name) || + (const_sect_name == g_CODE_sect_name))) { + section_type = eSectionTypeCode; + } else if (m_sect_headers[idx].flags & + llvm::COFF::IMAGE_SCN_CNT_INITIALIZED_DATA && + ((const_sect_name == g_data_sect_name) || + (const_sect_name == g_DATA_sect_name))) { + section_type = eSectionTypeData; + } else if (m_sect_headers[idx].flags & + llvm::COFF::IMAGE_SCN_CNT_UNINITIALIZED_DATA && + ((const_sect_name == g_bss_sect_name) || + (const_sect_name == g_BSS_sect_name))) { + if (m_sect_headers[idx].size == 0) + section_type = eSectionTypeZeroFill; + else + section_type = eSectionTypeData; + } else if (const_sect_name == g_debug_sect_name) { + section_type = eSectionTypeDebug; + } else if (const_sect_name == g_stabstr_sect_name) { + section_type = eSectionTypeDataCString; + } else if (const_sect_name == g_reloc_sect_name) { + section_type = eSectionTypeOther; + } else if (const_sect_name == g_sect_name_dwarf_debug_abbrev) + section_type = eSectionTypeDWARFDebugAbbrev; + else if (const_sect_name == g_sect_name_dwarf_debug_aranges) + section_type = eSectionTypeDWARFDebugAranges; + else if (const_sect_name == g_sect_name_dwarf_debug_frame) + section_type = eSectionTypeDWARFDebugFrame; + else if (const_sect_name == g_sect_name_dwarf_debug_info) + section_type = eSectionTypeDWARFDebugInfo; + else if (const_sect_name == g_sect_name_dwarf_debug_line) + section_type = eSectionTypeDWARFDebugLine; + else if (const_sect_name == g_sect_name_dwarf_debug_loc) + section_type = eSectionTypeDWARFDebugLoc; + else if (const_sect_name == g_sect_name_dwarf_debug_macinfo) + section_type = eSectionTypeDWARFDebugMacInfo; + else if (const_sect_name == g_sect_name_dwarf_debug_pubnames) + section_type = eSectionTypeDWARFDebugPubNames; + else if (const_sect_name == g_sect_name_dwarf_debug_pubtypes) + section_type = eSectionTypeDWARFDebugPubTypes; + else if (const_sect_name == g_sect_name_dwarf_debug_ranges) + section_type = eSectionTypeDWARFDebugRanges; + else if (const_sect_name == g_sect_name_dwarf_debug_str) + section_type = eSectionTypeDWARFDebugStr; + else if (const_sect_name == g_sect_name_eh_frame) + section_type = eSectionTypeEHFrame; + else if (const_sect_name == g_sect_name_go_symtab) + section_type = eSectionTypeGoSymtab; + else if (m_sect_headers[idx].flags & llvm::COFF::IMAGE_SCN_CNT_CODE) { + section_type = eSectionTypeCode; + } else if (m_sect_headers[idx].flags & + llvm::COFF::IMAGE_SCN_CNT_INITIALIZED_DATA) { + section_type = eSectionTypeData; + } else if (m_sect_headers[idx].flags & + llvm::COFF::IMAGE_SCN_CNT_UNINITIALIZED_DATA) { + if (m_sect_headers[idx].size == 0) + section_type = eSectionTypeZeroFill; + else + section_type = eSectionTypeData; } + + // Use a segment ID of the segment index shifted left by 8 so they + // never conflict with any of the sections. + SectionSP section_sp(new Section( + module_sp, // Module to which this section belongs + this, // Object file to which this section belongs + idx + 1, // Section ID is the 1 based segment index shifted right by + // 8 bits as not to collide with any of the 256 section IDs + // that are possible + const_sect_name, // Name of this section + section_type, // This section is a container of other sections. + m_coff_header_opt.image_base + + m_sect_headers[idx].vmaddr, // File VM address == addresses as + // they are found in the object file + m_sect_headers[idx].vmsize, // VM size in bytes of this section + m_sect_headers[idx] + .offset, // Offset to the data for this section in the file + m_sect_headers[idx] + .size, // Size in bytes of this section as found in the file + m_coff_header_opt.sect_alignment, // Section alignment + m_sect_headers[idx].flags)); // Flags for this section + + // section_sp->SetIsEncrypted (segment_is_encrypted); + + unified_section_list.AddSection(section_sp); + m_sections_ap->AddSection(section_sp); + } } + } } -bool -ObjectFilePECOFF::GetUUID (UUID* uuid) -{ - return false; -} +bool ObjectFilePECOFF::GetUUID(UUID *uuid) { return false; } -uint32_t -ObjectFilePECOFF::GetDependentModules (FileSpecList& files) -{ - return 0; +uint32_t ObjectFilePECOFF::GetDependentModules(FileSpecList &files) { + return 0; } -lldb_private::Address -ObjectFilePECOFF::GetEntryPointAddress () -{ - if (m_entry_point_address.IsValid()) - return m_entry_point_address; +lldb_private::Address ObjectFilePECOFF::GetEntryPointAddress() { + if (m_entry_point_address.IsValid()) + return m_entry_point_address; - if (!ParseHeader() || !IsExecutable()) - return m_entry_point_address; + if (!ParseHeader() || !IsExecutable()) + return m_entry_point_address; - SectionList *section_list = GetSectionList(); - addr_t offset = m_coff_header_opt.entry; + SectionList *section_list = GetSectionList(); + addr_t offset = m_coff_header_opt.entry; - if (!section_list) - m_entry_point_address.SetOffset(offset); - else - m_entry_point_address.ResolveAddressUsingFileSections(offset, section_list); - return m_entry_point_address; + if (!section_list) + m_entry_point_address.SetOffset(offset); + else + m_entry_point_address.ResolveAddressUsingFileSections(offset, section_list); + return m_entry_point_address; } - //---------------------------------------------------------------------- // Dump // // Dump the specifics of the runtime file container (such as any headers // segments, sections, etc). //---------------------------------------------------------------------- -void -ObjectFilePECOFF::Dump(Stream *s) -{ - ModuleSP module_sp(GetModule()); - if (module_sp) - { - std::lock_guard<std::recursive_mutex> guard(module_sp->GetMutex()); - s->Printf("%p: ", static_cast<void*>(this)); - s->Indent(); - s->PutCString("ObjectFilePECOFF"); - - ArchSpec header_arch; - GetArchitecture (header_arch); - - *s << ", file = '" << m_file << "', arch = " << header_arch.GetArchitectureName() << "\n"; - - SectionList *sections = GetSectionList(); - if (sections) - sections->Dump(s, NULL, true, UINT32_MAX); - - if (m_symtab_ap.get()) - m_symtab_ap->Dump(s, NULL, eSortOrderNone); - - if (m_dos_header.e_magic) - DumpDOSHeader (s, m_dos_header); - if (m_coff_header.machine) - { - DumpCOFFHeader (s, m_coff_header); - if (m_coff_header.hdrsize) - DumpOptCOFFHeader (s, m_coff_header_opt); - } - s->EOL(); - DumpSectionHeaders(s); - s->EOL(); +void ObjectFilePECOFF::Dump(Stream *s) { + ModuleSP module_sp(GetModule()); + if (module_sp) { + std::lock_guard<std::recursive_mutex> guard(module_sp->GetMutex()); + s->Printf("%p: ", static_cast<void *>(this)); + s->Indent(); + s->PutCString("ObjectFilePECOFF"); + + ArchSpec header_arch; + GetArchitecture(header_arch); + + *s << ", file = '" << m_file + << "', arch = " << header_arch.GetArchitectureName() << "\n"; + + SectionList *sections = GetSectionList(); + if (sections) + sections->Dump(s, NULL, true, UINT32_MAX); + + if (m_symtab_ap.get()) + m_symtab_ap->Dump(s, NULL, eSortOrderNone); + + if (m_dos_header.e_magic) + DumpDOSHeader(s, m_dos_header); + if (m_coff_header.machine) { + DumpCOFFHeader(s, m_coff_header); + if (m_coff_header.hdrsize) + DumpOptCOFFHeader(s, m_coff_header_opt); } + s->EOL(); + DumpSectionHeaders(s); + s->EOL(); + } } //---------------------------------------------------------------------- @@ -883,43 +843,33 @@ ObjectFilePECOFF::Dump(Stream *s) // // Dump the MS-DOS header to the specified output stream //---------------------------------------------------------------------- -void -ObjectFilePECOFF::DumpDOSHeader(Stream *s, const dos_header_t& header) -{ - s->PutCString ("MSDOS Header\n"); - s->Printf (" e_magic = 0x%4.4x\n", header.e_magic); - s->Printf (" e_cblp = 0x%4.4x\n", header.e_cblp); - s->Printf (" e_cp = 0x%4.4x\n", header.e_cp); - s->Printf (" e_crlc = 0x%4.4x\n", header.e_crlc); - s->Printf (" e_cparhdr = 0x%4.4x\n", header.e_cparhdr); - s->Printf (" e_minalloc = 0x%4.4x\n", header.e_minalloc); - s->Printf (" e_maxalloc = 0x%4.4x\n", header.e_maxalloc); - s->Printf (" e_ss = 0x%4.4x\n", header.e_ss); - s->Printf (" e_sp = 0x%4.4x\n", header.e_sp); - s->Printf (" e_csum = 0x%4.4x\n", header.e_csum); - s->Printf (" e_ip = 0x%4.4x\n", header.e_ip); - s->Printf (" e_cs = 0x%4.4x\n", header.e_cs); - s->Printf (" e_lfarlc = 0x%4.4x\n", header.e_lfarlc); - s->Printf (" e_ovno = 0x%4.4x\n", header.e_ovno); - s->Printf (" e_res[4] = { 0x%4.4x, 0x%4.4x, 0x%4.4x, 0x%4.4x }\n", - header.e_res[0], - header.e_res[1], - header.e_res[2], - header.e_res[3]); - s->Printf (" e_oemid = 0x%4.4x\n", header.e_oemid); - s->Printf (" e_oeminfo = 0x%4.4x\n", header.e_oeminfo); - s->Printf (" e_res2[10] = { 0x%4.4x, 0x%4.4x, 0x%4.4x, 0x%4.4x, 0x%4.4x, 0x%4.4x, 0x%4.4x, 0x%4.4x, 0x%4.4x, 0x%4.4x }\n", - header.e_res2[0], - header.e_res2[1], - header.e_res2[2], - header.e_res2[3], - header.e_res2[4], - header.e_res2[5], - header.e_res2[6], - header.e_res2[7], - header.e_res2[8], - header.e_res2[9]); - s->Printf (" e_lfanew = 0x%8.8x\n", header.e_lfanew); +void ObjectFilePECOFF::DumpDOSHeader(Stream *s, const dos_header_t &header) { + s->PutCString("MSDOS Header\n"); + s->Printf(" e_magic = 0x%4.4x\n", header.e_magic); + s->Printf(" e_cblp = 0x%4.4x\n", header.e_cblp); + s->Printf(" e_cp = 0x%4.4x\n", header.e_cp); + s->Printf(" e_crlc = 0x%4.4x\n", header.e_crlc); + s->Printf(" e_cparhdr = 0x%4.4x\n", header.e_cparhdr); + s->Printf(" e_minalloc = 0x%4.4x\n", header.e_minalloc); + s->Printf(" e_maxalloc = 0x%4.4x\n", header.e_maxalloc); + s->Printf(" e_ss = 0x%4.4x\n", header.e_ss); + s->Printf(" e_sp = 0x%4.4x\n", header.e_sp); + s->Printf(" e_csum = 0x%4.4x\n", header.e_csum); + s->Printf(" e_ip = 0x%4.4x\n", header.e_ip); + s->Printf(" e_cs = 0x%4.4x\n", header.e_cs); + s->Printf(" e_lfarlc = 0x%4.4x\n", header.e_lfarlc); + s->Printf(" e_ovno = 0x%4.4x\n", header.e_ovno); + s->Printf(" e_res[4] = { 0x%4.4x, 0x%4.4x, 0x%4.4x, 0x%4.4x }\n", + header.e_res[0], header.e_res[1], header.e_res[2], header.e_res[3]); + s->Printf(" e_oemid = 0x%4.4x\n", header.e_oemid); + s->Printf(" e_oeminfo = 0x%4.4x\n", header.e_oeminfo); + s->Printf(" e_res2[10] = { 0x%4.4x, 0x%4.4x, 0x%4.4x, 0x%4.4x, 0x%4.4x, " + "0x%4.4x, 0x%4.4x, 0x%4.4x, 0x%4.4x, 0x%4.4x }\n", + header.e_res2[0], header.e_res2[1], header.e_res2[2], + header.e_res2[3], header.e_res2[4], header.e_res2[5], + header.e_res2[6], header.e_res2[7], header.e_res2[8], + header.e_res2[9]); + s->Printf(" e_lfanew = 0x%8.8x\n", header.e_lfanew); } //---------------------------------------------------------------------- @@ -927,16 +877,14 @@ ObjectFilePECOFF::DumpDOSHeader(Stream *s, const dos_header_t& header) // // Dump the COFF header to the specified output stream //---------------------------------------------------------------------- -void -ObjectFilePECOFF::DumpCOFFHeader(Stream *s, const coff_header_t& header) -{ - s->PutCString ("COFF Header\n"); - s->Printf (" machine = 0x%4.4x\n", header.machine); - s->Printf (" nsects = 0x%4.4x\n", header.nsects); - s->Printf (" modtime = 0x%8.8x\n", header.modtime); - s->Printf (" symoff = 0x%8.8x\n", header.symoff); - s->Printf (" nsyms = 0x%8.8x\n", header.nsyms); - s->Printf (" hdrsize = 0x%4.4x\n", header.hdrsize); +void ObjectFilePECOFF::DumpCOFFHeader(Stream *s, const coff_header_t &header) { + s->PutCString("COFF Header\n"); + s->Printf(" machine = 0x%4.4x\n", header.machine); + s->Printf(" nsects = 0x%4.4x\n", header.nsects); + s->Printf(" modtime = 0x%8.8x\n", header.modtime); + s->Printf(" symoff = 0x%8.8x\n", header.symoff); + s->Printf(" nsyms = 0x%8.8x\n", header.nsyms); + s->Printf(" hdrsize = 0x%4.4x\n", header.hdrsize); } //---------------------------------------------------------------------- @@ -944,147 +892,145 @@ ObjectFilePECOFF::DumpCOFFHeader(Stream *s, const coff_header_t& header) // // Dump the optional COFF header to the specified output stream //---------------------------------------------------------------------- -void -ObjectFilePECOFF::DumpOptCOFFHeader(Stream *s, const coff_opt_header_t& header) -{ - s->PutCString ("Optional COFF Header\n"); - s->Printf (" magic = 0x%4.4x\n", header.magic); - s->Printf (" major_linker_version = 0x%2.2x\n", header.major_linker_version); - s->Printf (" minor_linker_version = 0x%2.2x\n", header.minor_linker_version); - s->Printf (" code_size = 0x%8.8x\n", header.code_size); - s->Printf (" data_size = 0x%8.8x\n", header.data_size); - s->Printf (" bss_size = 0x%8.8x\n", header.bss_size); - s->Printf (" entry = 0x%8.8x\n", header.entry); - s->Printf (" code_offset = 0x%8.8x\n", header.code_offset); - s->Printf (" data_offset = 0x%8.8x\n", header.data_offset); - s->Printf (" image_base = 0x%16.16" PRIx64 "\n", header.image_base); - s->Printf (" sect_alignment = 0x%8.8x\n", header.sect_alignment); - s->Printf (" file_alignment = 0x%8.8x\n", header.file_alignment); - s->Printf (" major_os_system_version = 0x%4.4x\n", header.major_os_system_version); - s->Printf (" minor_os_system_version = 0x%4.4x\n", header.minor_os_system_version); - s->Printf (" major_image_version = 0x%4.4x\n", header.major_image_version); - s->Printf (" minor_image_version = 0x%4.4x\n", header.minor_image_version); - s->Printf (" major_subsystem_version = 0x%4.4x\n", header.major_subsystem_version); - s->Printf (" minor_subsystem_version = 0x%4.4x\n", header.minor_subsystem_version); - s->Printf (" reserved1 = 0x%8.8x\n", header.reserved1); - s->Printf (" image_size = 0x%8.8x\n", header.image_size); - s->Printf (" header_size = 0x%8.8x\n", header.header_size); - s->Printf (" checksum = 0x%8.8x\n", header.checksum); - s->Printf (" subsystem = 0x%4.4x\n", header.subsystem); - s->Printf (" dll_flags = 0x%4.4x\n", header.dll_flags); - s->Printf (" stack_reserve_size = 0x%16.16" PRIx64 "\n", header.stack_reserve_size); - s->Printf (" stack_commit_size = 0x%16.16" PRIx64 "\n", header.stack_commit_size); - s->Printf (" heap_reserve_size = 0x%16.16" PRIx64 "\n", header.heap_reserve_size); - s->Printf (" heap_commit_size = 0x%16.16" PRIx64 "\n", header.heap_commit_size); - s->Printf (" loader_flags = 0x%8.8x\n", header.loader_flags); - s->Printf (" num_data_dir_entries = 0x%8.8x\n", (uint32_t)header.data_dirs.size()); - uint32_t i; - for (i=0; i<header.data_dirs.size(); i++) - { - s->Printf (" data_dirs[%2u] vmaddr = 0x%8.8x, vmsize = 0x%8.8x\n", - i, - header.data_dirs[i].vmaddr, - header.data_dirs[i].vmsize); - } +void ObjectFilePECOFF::DumpOptCOFFHeader(Stream *s, + const coff_opt_header_t &header) { + s->PutCString("Optional COFF Header\n"); + s->Printf(" magic = 0x%4.4x\n", header.magic); + s->Printf(" major_linker_version = 0x%2.2x\n", + header.major_linker_version); + s->Printf(" minor_linker_version = 0x%2.2x\n", + header.minor_linker_version); + s->Printf(" code_size = 0x%8.8x\n", header.code_size); + s->Printf(" data_size = 0x%8.8x\n", header.data_size); + s->Printf(" bss_size = 0x%8.8x\n", header.bss_size); + s->Printf(" entry = 0x%8.8x\n", header.entry); + s->Printf(" code_offset = 0x%8.8x\n", header.code_offset); + s->Printf(" data_offset = 0x%8.8x\n", header.data_offset); + s->Printf(" image_base = 0x%16.16" PRIx64 "\n", + header.image_base); + s->Printf(" sect_alignment = 0x%8.8x\n", header.sect_alignment); + s->Printf(" file_alignment = 0x%8.8x\n", header.file_alignment); + s->Printf(" major_os_system_version = 0x%4.4x\n", + header.major_os_system_version); + s->Printf(" minor_os_system_version = 0x%4.4x\n", + header.minor_os_system_version); + s->Printf(" major_image_version = 0x%4.4x\n", + header.major_image_version); + s->Printf(" minor_image_version = 0x%4.4x\n", + header.minor_image_version); + s->Printf(" major_subsystem_version = 0x%4.4x\n", + header.major_subsystem_version); + s->Printf(" minor_subsystem_version = 0x%4.4x\n", + header.minor_subsystem_version); + s->Printf(" reserved1 = 0x%8.8x\n", header.reserved1); + s->Printf(" image_size = 0x%8.8x\n", header.image_size); + s->Printf(" header_size = 0x%8.8x\n", header.header_size); + s->Printf(" checksum = 0x%8.8x\n", header.checksum); + s->Printf(" subsystem = 0x%4.4x\n", header.subsystem); + s->Printf(" dll_flags = 0x%4.4x\n", header.dll_flags); + s->Printf(" stack_reserve_size = 0x%16.16" PRIx64 "\n", + header.stack_reserve_size); + s->Printf(" stack_commit_size = 0x%16.16" PRIx64 "\n", + header.stack_commit_size); + s->Printf(" heap_reserve_size = 0x%16.16" PRIx64 "\n", + header.heap_reserve_size); + s->Printf(" heap_commit_size = 0x%16.16" PRIx64 "\n", + header.heap_commit_size); + s->Printf(" loader_flags = 0x%8.8x\n", header.loader_flags); + s->Printf(" num_data_dir_entries = 0x%8.8x\n", + (uint32_t)header.data_dirs.size()); + uint32_t i; + for (i = 0; i < header.data_dirs.size(); i++) { + s->Printf(" data_dirs[%2u] vmaddr = 0x%8.8x, vmsize = 0x%8.8x\n", i, + header.data_dirs[i].vmaddr, header.data_dirs[i].vmsize); + } } //---------------------------------------------------------------------- // DumpSectionHeader // // Dump a single ELF section header to the specified output stream //---------------------------------------------------------------------- -void -ObjectFilePECOFF::DumpSectionHeader(Stream *s, const section_header_t& sh) -{ - std::string name; - GetSectionName(name, sh); - s->Printf ("%-16s 0x%8.8x 0x%8.8x 0x%8.8x 0x%8.8x 0x%8.8x 0x%8.8x 0x%4.4x 0x%4.4x 0x%8.8x\n", - name.c_str(), - sh.vmaddr, - sh.vmsize, - sh.offset, - sh.size, - sh.reloff, - sh.lineoff, - sh.nreloc, - sh.nline, - sh.flags); +void ObjectFilePECOFF::DumpSectionHeader(Stream *s, + const section_header_t &sh) { + std::string name; + GetSectionName(name, sh); + s->Printf("%-16s 0x%8.8x 0x%8.8x 0x%8.8x 0x%8.8x 0x%8.8x 0x%8.8x 0x%4.4x " + "0x%4.4x 0x%8.8x\n", + name.c_str(), sh.vmaddr, sh.vmsize, sh.offset, sh.size, sh.reloff, + sh.lineoff, sh.nreloc, sh.nline, sh.flags); } - //---------------------------------------------------------------------- // DumpSectionHeaders // // Dump all of the ELF section header to the specified output stream //---------------------------------------------------------------------- -void -ObjectFilePECOFF::DumpSectionHeaders(Stream *s) -{ - - s->PutCString ("Section Headers\n"); - s->PutCString ("IDX name vm addr vm size file off file size reloc off line off nreloc nline flags\n"); - s->PutCString ("==== ---------------- ---------- ---------- ---------- ---------- ---------- ---------- ------ ------ ----------\n"); - - uint32_t idx = 0; - SectionHeaderCollIter pos, end = m_sect_headers.end(); - - for (pos = m_sect_headers.begin(); pos != end; ++pos, ++idx) - { - s->Printf ("[%2u] ", idx); - ObjectFilePECOFF::DumpSectionHeader(s, *pos); - } +void ObjectFilePECOFF::DumpSectionHeaders(Stream *s) { + + s->PutCString("Section Headers\n"); + s->PutCString("IDX name vm addr vm size file off file " + "size reloc off line off nreloc nline flags\n"); + s->PutCString("==== ---------------- ---------- ---------- ---------- " + "---------- ---------- ---------- ------ ------ ----------\n"); + + uint32_t idx = 0; + SectionHeaderCollIter pos, end = m_sect_headers.end(); + + for (pos = m_sect_headers.begin(); pos != end; ++pos, ++idx) { + s->Printf("[%2u] ", idx); + ObjectFilePECOFF::DumpSectionHeader(s, *pos); + } } -bool -ObjectFilePECOFF::GetArchitecture (ArchSpec &arch) -{ - uint16_t machine = m_coff_header.machine; - switch (machine) - { - case llvm::COFF::IMAGE_FILE_MACHINE_AMD64: - case llvm::COFF::IMAGE_FILE_MACHINE_I386: - case llvm::COFF::IMAGE_FILE_MACHINE_POWERPC: - case llvm::COFF::IMAGE_FILE_MACHINE_POWERPCFP: - case llvm::COFF::IMAGE_FILE_MACHINE_ARM: - case llvm::COFF::IMAGE_FILE_MACHINE_ARMNT: - case llvm::COFF::IMAGE_FILE_MACHINE_THUMB: - arch.SetArchitecture (eArchTypeCOFF, machine, LLDB_INVALID_CPUTYPE); - return true; - default: - break; - } +bool ObjectFilePECOFF::IsWindowsSubsystem() { + switch (m_coff_header_opt.subsystem) { + case llvm::COFF::IMAGE_SUBSYSTEM_NATIVE: + case llvm::COFF::IMAGE_SUBSYSTEM_WINDOWS_GUI: + case llvm::COFF::IMAGE_SUBSYSTEM_WINDOWS_CUI: + case llvm::COFF::IMAGE_SUBSYSTEM_NATIVE_WINDOWS: + case llvm::COFF::IMAGE_SUBSYSTEM_WINDOWS_CE_GUI: + case llvm::COFF::IMAGE_SUBSYSTEM_XBOX: + case llvm::COFF::IMAGE_SUBSYSTEM_WINDOWS_BOOT_APPLICATION: + return true; + default: return false; + } } -ObjectFile::Type -ObjectFilePECOFF::CalculateType() -{ - if (m_coff_header.machine != 0) - { - if ((m_coff_header.flags & llvm::COFF::IMAGE_FILE_DLL) == 0) - return eTypeExecutable; - else - return eTypeSharedLibrary; - } - return eTypeExecutable; +bool ObjectFilePECOFF::GetArchitecture(ArchSpec &arch) { + uint16_t machine = m_coff_header.machine; + switch (machine) { + case llvm::COFF::IMAGE_FILE_MACHINE_AMD64: + case llvm::COFF::IMAGE_FILE_MACHINE_I386: + case llvm::COFF::IMAGE_FILE_MACHINE_POWERPC: + case llvm::COFF::IMAGE_FILE_MACHINE_POWERPCFP: + case llvm::COFF::IMAGE_FILE_MACHINE_ARM: + case llvm::COFF::IMAGE_FILE_MACHINE_ARMNT: + case llvm::COFF::IMAGE_FILE_MACHINE_THUMB: + arch.SetArchitecture(eArchTypeCOFF, machine, LLDB_INVALID_CPUTYPE, + IsWindowsSubsystem() ? llvm::Triple::Win32 + : llvm::Triple::UnknownOS); + return true; + default: + break; + } + return false; } -ObjectFile::Strata -ObjectFilePECOFF::CalculateStrata() -{ - return eStrataUser; +ObjectFile::Type ObjectFilePECOFF::CalculateType() { + if (m_coff_header.machine != 0) { + if ((m_coff_header.flags & llvm::COFF::IMAGE_FILE_DLL) == 0) + return eTypeExecutable; + else + return eTypeSharedLibrary; + } + return eTypeExecutable; } + +ObjectFile::Strata ObjectFilePECOFF::CalculateStrata() { return eStrataUser; } //------------------------------------------------------------------ // PluginInterface protocol //------------------------------------------------------------------ -ConstString -ObjectFilePECOFF::GetPluginName() -{ - return GetPluginNameStatic(); -} - -uint32_t -ObjectFilePECOFF::GetPluginVersion() -{ - return 1; -} +ConstString ObjectFilePECOFF::GetPluginName() { return GetPluginNameStatic(); } +uint32_t ObjectFilePECOFF::GetPluginVersion() { return 1; } diff --git a/source/Plugins/ObjectFile/PECOFF/ObjectFilePECOFF.h b/source/Plugins/ObjectFile/PECOFF/ObjectFilePECOFF.h index 6c1cdbf56c6b8..2313047735e0c 100644 --- a/source/Plugins/ObjectFile/PECOFF/ObjectFilePECOFF.h +++ b/source/Plugins/ObjectFile/PECOFF/ObjectFilePECOFF.h @@ -18,290 +18,270 @@ // Project includes #include "lldb/Symbol/ObjectFile.h" -class ObjectFilePECOFF : - public lldb_private::ObjectFile -{ +class ObjectFilePECOFF : public lldb_private::ObjectFile { public: - typedef enum MachineType - { - MachineUnknown = 0x0, - MachineAm33 = 0x1d3, - MachineAmd64 = 0x8664, - MachineArm = 0x1c0, - MachineArmNt = 0x1c4, - MachineArm64 = 0xaa64, - MachineEbc = 0xebc, - MachineX86 = 0x14c, - MachineIA64 = 0x200, - MachineM32R = 0x9041, - MachineMips16 = 0x266, - MachineMipsFpu = 0x366, - MachineMipsFpu16 = 0x466, - MachinePowerPc = 0x1f0, - MachinePowerPcfp = 0x1f1, - MachineR4000 = 0x166, - MachineSh3 = 0x1a2, - MachineSh3dsp = 0x1a3, - MachineSh4 = 0x1a6, - MachineSh5 = 0x1a8, - MachineThumb = 0x1c2, - MachineWcemIpsv2 = 0x169 - } MachineType; - - ObjectFilePECOFF(const lldb::ModuleSP &module_sp, - lldb::DataBufferSP& data_sp, - lldb::offset_t data_offset, - const lldb_private::FileSpec* file, - lldb::offset_t file_offset, - lldb::offset_t length); - - ~ObjectFilePECOFF() override; - - //------------------------------------------------------------------ - // Static Functions - //------------------------------------------------------------------ - static void - Initialize(); - - static void - Terminate(); - - static lldb_private::ConstString - GetPluginNameStatic(); - - static const char * - GetPluginDescriptionStatic(); - - static ObjectFile * - CreateInstance (const lldb::ModuleSP &module_sp, - lldb::DataBufferSP& data_sp, - lldb::offset_t data_offset, - const lldb_private::FileSpec* file, - lldb::offset_t offset, - lldb::offset_t length); - - static lldb_private::ObjectFile * - CreateMemoryInstance (const lldb::ModuleSP &module_sp, - lldb::DataBufferSP& data_sp, - const lldb::ProcessSP &process_sp, - lldb::addr_t header_addr); - - static size_t - GetModuleSpecifications (const lldb_private::FileSpec& file, - lldb::DataBufferSP& data_sp, - lldb::offset_t data_offset, - lldb::offset_t file_offset, - lldb::offset_t length, - lldb_private::ModuleSpecList &specs); - - static bool - SaveCore (const lldb::ProcessSP &process_sp, - const lldb_private::FileSpec &outfile, - lldb_private::Error &error); - - static bool - MagicBytesMatch (lldb::DataBufferSP& data_sp); - - static lldb::SymbolType - MapSymbolType(uint16_t coff_symbol_type); - - bool - ParseHeader() override; - - bool - SetLoadAddress(lldb_private::Target &target, lldb::addr_t value, bool value_is_offset) override; - - lldb::ByteOrder - GetByteOrder() const override; - - bool - IsExecutable() const override; - - uint32_t - GetAddressByteSize() const override; - -// virtual lldb_private::AddressClass -// GetAddressClass (lldb::addr_t file_addr); - - lldb_private::Symtab * - GetSymtab() override; - - bool - IsStripped() override; - - void - CreateSections(lldb_private::SectionList &unified_section_list) override; - - void - Dump(lldb_private::Stream *s) override; - - bool - GetArchitecture(lldb_private::ArchSpec &arch) override; - - bool - GetUUID(lldb_private::UUID* uuid) override; - - uint32_t - GetDependentModules(lldb_private::FileSpecList& files) override; - - virtual lldb_private::Address - GetEntryPointAddress () override; - - ObjectFile::Type - CalculateType() override; - - ObjectFile::Strata - CalculateStrata() override; - - //------------------------------------------------------------------ - // PluginInterface protocol - //------------------------------------------------------------------ - lldb_private::ConstString - GetPluginName() override; - - uint32_t - GetPluginVersion() override; + typedef enum MachineType { + MachineUnknown = 0x0, + MachineAm33 = 0x1d3, + MachineAmd64 = 0x8664, + MachineArm = 0x1c0, + MachineArmNt = 0x1c4, + MachineArm64 = 0xaa64, + MachineEbc = 0xebc, + MachineX86 = 0x14c, + MachineIA64 = 0x200, + MachineM32R = 0x9041, + MachineMips16 = 0x266, + MachineMipsFpu = 0x366, + MachineMipsFpu16 = 0x466, + MachinePowerPc = 0x1f0, + MachinePowerPcfp = 0x1f1, + MachineR4000 = 0x166, + MachineSh3 = 0x1a2, + MachineSh3dsp = 0x1a3, + MachineSh4 = 0x1a6, + MachineSh5 = 0x1a8, + MachineThumb = 0x1c2, + MachineWcemIpsv2 = 0x169 + } MachineType; + + ObjectFilePECOFF(const lldb::ModuleSP &module_sp, lldb::DataBufferSP &data_sp, + lldb::offset_t data_offset, + const lldb_private::FileSpec *file, + lldb::offset_t file_offset, lldb::offset_t length); + + ObjectFilePECOFF(const lldb::ModuleSP &module_sp, + lldb::DataBufferSP &header_data_sp, + const lldb::ProcessSP &process_sp, lldb::addr_t header_addr); + + ~ObjectFilePECOFF() override; + + //------------------------------------------------------------------ + // Static Functions + //------------------------------------------------------------------ + static void Initialize(); + + static void Terminate(); + + static lldb_private::ConstString GetPluginNameStatic(); + + static const char *GetPluginDescriptionStatic(); + + static ObjectFile * + CreateInstance(const lldb::ModuleSP &module_sp, lldb::DataBufferSP &data_sp, + lldb::offset_t data_offset, const lldb_private::FileSpec *file, + lldb::offset_t offset, lldb::offset_t length); + + static lldb_private::ObjectFile *CreateMemoryInstance( + const lldb::ModuleSP &module_sp, lldb::DataBufferSP &data_sp, + const lldb::ProcessSP &process_sp, lldb::addr_t header_addr); + + static size_t GetModuleSpecifications(const lldb_private::FileSpec &file, + lldb::DataBufferSP &data_sp, + lldb::offset_t data_offset, + lldb::offset_t file_offset, + lldb::offset_t length, + lldb_private::ModuleSpecList &specs); + + static bool SaveCore(const lldb::ProcessSP &process_sp, + const lldb_private::FileSpec &outfile, + lldb_private::Error &error); + + static bool MagicBytesMatch(lldb::DataBufferSP &data_sp); + + static lldb::SymbolType MapSymbolType(uint16_t coff_symbol_type); + + bool ParseHeader() override; + + bool SetLoadAddress(lldb_private::Target &target, lldb::addr_t value, + bool value_is_offset) override; + + lldb::ByteOrder GetByteOrder() const override; + + bool IsExecutable() const override; + + uint32_t GetAddressByteSize() const override; + + // virtual lldb_private::AddressClass + // GetAddressClass (lldb::addr_t file_addr); + + lldb_private::Symtab *GetSymtab() override; + + bool IsStripped() override; + + void CreateSections(lldb_private::SectionList &unified_section_list) override; + + void Dump(lldb_private::Stream *s) override; + + bool GetArchitecture(lldb_private::ArchSpec &arch) override; + + bool GetUUID(lldb_private::UUID *uuid) override; + + uint32_t GetDependentModules(lldb_private::FileSpecList &files) override; + + virtual lldb_private::Address GetEntryPointAddress() override; + + ObjectFile::Type CalculateType() override; + + ObjectFile::Strata CalculateStrata() override; + + //------------------------------------------------------------------ + // PluginInterface protocol + //------------------------------------------------------------------ + lldb_private::ConstString GetPluginName() override; + + uint32_t GetPluginVersion() override; + + bool IsWindowsSubsystem(); + + lldb_private::DataExtractor ReadImageData(uint32_t offset, size_t size); protected: - bool NeedsEndianSwap() const; - - typedef struct dos_header { // DOS .EXE header - uint16_t e_magic; // Magic number - uint16_t e_cblp; // Bytes on last page of file - uint16_t e_cp; // Pages in file - uint16_t e_crlc; // Relocations - uint16_t e_cparhdr; // Size of header in paragraphs - uint16_t e_minalloc; // Minimum extra paragraphs needed - uint16_t e_maxalloc; // Maximum extra paragraphs needed - uint16_t e_ss; // Initial (relative) SS value - uint16_t e_sp; // Initial SP value - uint16_t e_csum; // Checksum - uint16_t e_ip; // Initial IP value - uint16_t e_cs; // Initial (relative) CS value - uint16_t e_lfarlc; // File address of relocation table - uint16_t e_ovno; // Overlay number - uint16_t e_res[4]; // Reserved words - uint16_t e_oemid; // OEM identifier (for e_oeminfo) - uint16_t e_oeminfo; // OEM information; e_oemid specific - uint16_t e_res2[10]; // Reserved words - uint32_t e_lfanew; // File address of new exe header - } dos_header_t; - - typedef struct coff_header { - uint16_t machine; - uint16_t nsects; - uint32_t modtime; - uint32_t symoff; - uint32_t nsyms; - uint16_t hdrsize; - uint16_t flags; - } coff_header_t; - - typedef struct data_directory { - uint32_t vmaddr; - uint32_t vmsize; - } data_directory_t; - - typedef struct coff_opt_header - { - uint16_t magic; - uint8_t major_linker_version; - uint8_t minor_linker_version; - uint32_t code_size; - uint32_t data_size; - uint32_t bss_size; - uint32_t entry; - uint32_t code_offset; - uint32_t data_offset; - - uint64_t image_base; - uint32_t sect_alignment; - uint32_t file_alignment; - uint16_t major_os_system_version; - uint16_t minor_os_system_version; - uint16_t major_image_version; - uint16_t minor_image_version; - uint16_t major_subsystem_version; - uint16_t minor_subsystem_version; - uint32_t reserved1; - uint32_t image_size; - uint32_t header_size; - uint32_t checksum; - uint16_t subsystem; - uint16_t dll_flags; - uint64_t stack_reserve_size; - uint64_t stack_commit_size; - uint64_t heap_reserve_size; - uint64_t heap_commit_size; - uint32_t loader_flags; - // uint32_t num_data_dir_entries; - std::vector<data_directory> data_dirs; // will contain num_data_dir_entries entries - } coff_opt_header_t; - - typedef enum coff_data_dir_type - { - coff_data_dir_export_table = 0, - coff_data_dir_import_table = 1, - } coff_data_dir_type; - - typedef struct section_header { - char name[8]; - uint32_t vmsize; // Virtual Size - uint32_t vmaddr; // Virtual Addr - uint32_t size; // File size - uint32_t offset; // File offset - uint32_t reloff; // Offset to relocations - uint32_t lineoff;// Offset to line table entries - uint16_t nreloc; // Number of relocation entries - uint16_t nline; // Number of line table entries - uint32_t flags; - } section_header_t; - - typedef struct coff_symbol { - char name[8]; - uint32_t value; - uint16_t sect; - uint16_t type; - uint8_t storage; - uint8_t naux; - } coff_symbol_t; - - typedef struct export_directory_entry { - uint32_t characteristics; - uint32_t time_date_stamp; - uint16_t major_version; - uint16_t minor_version; - uint32_t name; - uint32_t base; - uint32_t number_of_functions; - uint32_t number_of_names; - uint32_t address_of_functions; - uint32_t address_of_names; - uint32_t address_of_name_ordinals; - } export_directory_entry; - - static bool ParseDOSHeader (lldb_private::DataExtractor &data, dos_header_t &dos_header); - static bool ParseCOFFHeader (lldb_private::DataExtractor &data, lldb::offset_t *offset_ptr, coff_header_t &coff_header); - bool ParseCOFFOptionalHeader (lldb::offset_t *offset_ptr); - bool ParseSectionHeaders (uint32_t offset); - - static void DumpDOSHeader(lldb_private::Stream *s, const dos_header_t& header); - static void DumpCOFFHeader(lldb_private::Stream *s, const coff_header_t& header); - static void DumpOptCOFFHeader(lldb_private::Stream *s, const coff_opt_header_t& header); - void DumpSectionHeaders(lldb_private::Stream *s); - void DumpSectionHeader(lldb_private::Stream *s, const section_header_t& sh); - bool GetSectionName(std::string& sect_name, const section_header_t& sect); - - typedef std::vector<section_header_t> SectionHeaderColl; - typedef SectionHeaderColl::iterator SectionHeaderCollIter; - typedef SectionHeaderColl::const_iterator SectionHeaderCollConstIter; + bool NeedsEndianSwap() const; + + typedef struct dos_header { // DOS .EXE header + uint16_t e_magic; // Magic number + uint16_t e_cblp; // Bytes on last page of file + uint16_t e_cp; // Pages in file + uint16_t e_crlc; // Relocations + uint16_t e_cparhdr; // Size of header in paragraphs + uint16_t e_minalloc; // Minimum extra paragraphs needed + uint16_t e_maxalloc; // Maximum extra paragraphs needed + uint16_t e_ss; // Initial (relative) SS value + uint16_t e_sp; // Initial SP value + uint16_t e_csum; // Checksum + uint16_t e_ip; // Initial IP value + uint16_t e_cs; // Initial (relative) CS value + uint16_t e_lfarlc; // File address of relocation table + uint16_t e_ovno; // Overlay number + uint16_t e_res[4]; // Reserved words + uint16_t e_oemid; // OEM identifier (for e_oeminfo) + uint16_t e_oeminfo; // OEM information; e_oemid specific + uint16_t e_res2[10]; // Reserved words + uint32_t e_lfanew; // File address of new exe header + } dos_header_t; + + typedef struct coff_header { + uint16_t machine; + uint16_t nsects; + uint32_t modtime; + uint32_t symoff; + uint32_t nsyms; + uint16_t hdrsize; + uint16_t flags; + } coff_header_t; + + typedef struct data_directory { + uint32_t vmaddr; + uint32_t vmsize; + } data_directory_t; + + typedef struct coff_opt_header { + uint16_t magic; + uint8_t major_linker_version; + uint8_t minor_linker_version; + uint32_t code_size; + uint32_t data_size; + uint32_t bss_size; + uint32_t entry; + uint32_t code_offset; + uint32_t data_offset; + + uint64_t image_base; + uint32_t sect_alignment; + uint32_t file_alignment; + uint16_t major_os_system_version; + uint16_t minor_os_system_version; + uint16_t major_image_version; + uint16_t minor_image_version; + uint16_t major_subsystem_version; + uint16_t minor_subsystem_version; + uint32_t reserved1; + uint32_t image_size; + uint32_t header_size; + uint32_t checksum; + uint16_t subsystem; + uint16_t dll_flags; + uint64_t stack_reserve_size; + uint64_t stack_commit_size; + uint64_t heap_reserve_size; + uint64_t heap_commit_size; + uint32_t loader_flags; + // uint32_t num_data_dir_entries; + std::vector<data_directory> + data_dirs; // will contain num_data_dir_entries entries + } coff_opt_header_t; + + typedef enum coff_data_dir_type { + coff_data_dir_export_table = 0, + coff_data_dir_import_table = 1, + } coff_data_dir_type; + + typedef struct section_header { + char name[8]; + uint32_t vmsize; // Virtual Size + uint32_t vmaddr; // Virtual Addr + uint32_t size; // File size + uint32_t offset; // File offset + uint32_t reloff; // Offset to relocations + uint32_t lineoff; // Offset to line table entries + uint16_t nreloc; // Number of relocation entries + uint16_t nline; // Number of line table entries + uint32_t flags; + } section_header_t; + + typedef struct coff_symbol { + char name[8]; + uint32_t value; + uint16_t sect; + uint16_t type; + uint8_t storage; + uint8_t naux; + } coff_symbol_t; + + typedef struct export_directory_entry { + uint32_t characteristics; + uint32_t time_date_stamp; + uint16_t major_version; + uint16_t minor_version; + uint32_t name; + uint32_t base; + uint32_t number_of_functions; + uint32_t number_of_names; + uint32_t address_of_functions; + uint32_t address_of_names; + uint32_t address_of_name_ordinals; + } export_directory_entry; + + static bool ParseDOSHeader(lldb_private::DataExtractor &data, + dos_header_t &dos_header); + static bool ParseCOFFHeader(lldb_private::DataExtractor &data, + lldb::offset_t *offset_ptr, + coff_header_t &coff_header); + bool ParseCOFFOptionalHeader(lldb::offset_t *offset_ptr); + bool ParseSectionHeaders(uint32_t offset); + + static void DumpDOSHeader(lldb_private::Stream *s, + const dos_header_t &header); + static void DumpCOFFHeader(lldb_private::Stream *s, + const coff_header_t &header); + static void DumpOptCOFFHeader(lldb_private::Stream *s, + const coff_opt_header_t &header); + void DumpSectionHeaders(lldb_private::Stream *s); + void DumpSectionHeader(lldb_private::Stream *s, const section_header_t &sh); + bool GetSectionName(std::string §_name, const section_header_t §); + + typedef std::vector<section_header_t> SectionHeaderColl; + typedef SectionHeaderColl::iterator SectionHeaderCollIter; + typedef SectionHeaderColl::const_iterator SectionHeaderCollConstIter; + private: - dos_header_t m_dos_header; - coff_header_t m_coff_header; - coff_opt_header_t m_coff_header_opt; - SectionHeaderColl m_sect_headers; - lldb::addr_t m_image_base; - lldb_private::Address m_entry_point_address; + dos_header_t m_dos_header; + coff_header_t m_coff_header; + coff_opt_header_t m_coff_header_opt; + SectionHeaderColl m_sect_headers; + lldb::addr_t m_image_base; + lldb_private::Address m_entry_point_address; }; #endif // liblldb_ObjectFilePECOFF_h_ diff --git a/source/Plugins/ObjectFile/PECOFF/WindowsMiniDump.cpp b/source/Plugins/ObjectFile/PECOFF/WindowsMiniDump.cpp index 14c73eff2388a..3fd6cdc5c9a76 100644 --- a/source/Plugins/ObjectFile/PECOFF/WindowsMiniDump.cpp +++ b/source/Plugins/ObjectFile/PECOFF/WindowsMiniDump.cpp @@ -16,41 +16,44 @@ #ifdef _WIN32 #include "lldb/Host/windows/windows.h" -#include <DbgHelp.h> // for MiniDumpWriteDump +#include <dbghelp.h> // for MiniDumpWriteDump #endif namespace lldb_private { -bool -SaveMiniDump(const lldb::ProcessSP &process_sp, - const lldb_private::FileSpec &outfile, - lldb_private::Error &error) -{ - if (!process_sp) return false; +bool SaveMiniDump(const lldb::ProcessSP &process_sp, + const lldb_private::FileSpec &outfile, + lldb_private::Error &error) { + if (!process_sp) + return false; #ifdef _WIN32 - HANDLE process_handle = ::OpenProcess(PROCESS_QUERY_INFORMATION | PROCESS_VM_READ, FALSE, process_sp->GetID()); - const std::string file_name = outfile.GetCString(); - std::wstring wide_name; - wide_name.resize(file_name.size() + 1); - char * result_ptr = reinterpret_cast<char *>(&wide_name[0]); - const UTF8 *error_ptr = nullptr; - if (!llvm::ConvertUTF8toWide(sizeof(wchar_t), file_name, result_ptr, error_ptr)) { - error.SetErrorString("cannot convert file name"); - return false; - } - HANDLE file_handle = ::CreateFileW(wide_name.c_str(), GENERIC_WRITE, FILE_SHARE_READ, NULL, CREATE_ALWAYS, FILE_ATTRIBUTE_NORMAL, NULL); - const auto result = ::MiniDumpWriteDump(process_handle, process_sp->GetID(), file_handle, - MiniDumpWithFullMemoryInfo, NULL, NULL, NULL); - ::CloseHandle(file_handle); - ::CloseHandle(process_handle); - if (!result) - { - error.SetError(::GetLastError(), lldb::eErrorTypeWin32); - return false; - } - return true; -#endif + HANDLE process_handle = ::OpenProcess( + PROCESS_QUERY_INFORMATION | PROCESS_VM_READ, FALSE, process_sp->GetID()); + const std::string file_name = outfile.GetCString(); + std::wstring wide_name; + wide_name.resize(file_name.size() + 1); + char *result_ptr = reinterpret_cast<char *>(&wide_name[0]); + const llvm::UTF8 *error_ptr = nullptr; + if (!llvm::ConvertUTF8toWide(sizeof(wchar_t), file_name, result_ptr, + error_ptr)) { + error.SetErrorString("cannot convert file name"); return false; + } + HANDLE file_handle = + ::CreateFileW(wide_name.c_str(), GENERIC_WRITE, FILE_SHARE_READ, NULL, + CREATE_ALWAYS, FILE_ATTRIBUTE_NORMAL, NULL); + const auto result = + ::MiniDumpWriteDump(process_handle, process_sp->GetID(), file_handle, + MiniDumpWithFullMemoryInfo, NULL, NULL, NULL); + ::CloseHandle(file_handle); + ::CloseHandle(process_handle); + if (!result) { + error.SetError(::GetLastError(), lldb::eErrorTypeWin32); + return false; + } + return true; +#endif + return false; } -} // namesapce lldb_private +} // namesapce lldb_private diff --git a/source/Plugins/ObjectFile/PECOFF/WindowsMiniDump.h b/source/Plugins/ObjectFile/PECOFF/WindowsMiniDump.h index cbea88af1fb0a..3f741dc04f514 100644 --- a/source/Plugins/ObjectFile/PECOFF/WindowsMiniDump.h +++ b/source/Plugins/ObjectFile/PECOFF/WindowsMiniDump.h @@ -14,11 +14,10 @@ namespace lldb_private { -bool -SaveMiniDump(const lldb::ProcessSP &process_sp, - const lldb_private::FileSpec &outfile, - lldb_private::Error &error); +bool SaveMiniDump(const lldb::ProcessSP &process_sp, + const lldb_private::FileSpec &outfile, + lldb_private::Error &error); -} // namespace lldb_private +} // namespace lldb_private #endif |