pwnlib.elf.elf — ELF Files

Exposes functionality for manipulating ELF files

Stop hard-coding things! Look them up at runtime with pwnlib.elf.

Example Usage

>>> e = ELF('/bin/cat')
>>> print(hex(e.address)) 
0x400000
>>> print(hex(e.symbols['write'])) 
0x401680
>>> print(hex(e.got['write'])) 
0x60b070
>>> print(hex(e.plt['write'])) 
0x401680

You can even patch and save the files.

>>> e = ELF('/bin/cat')
>>> e.read(e.address+1, 3)
b'ELF'
>>> e.asm(e.address, 'ret')
>>> e.save('/tmp/quiet-cat')
>>> disasm(open('/tmp/quiet-cat','rb').read(1))
'   0:   c3                      ret'

Module Members

class pwnlib.elf.elf.ELF(path, checksec=True)

Bases: ELFFile

Encapsulates information about an ELF file.

Example

>>> bash = ELF(which('bash'))
>>> hex(bash.symbols['read'])
0x41dac0
>>> hex(bash.plt['read'])
0x41dac0
>>> u32(bash.read(bash.got['read'], 4))
0x41dac6
>>> print(bash.disasm(bash.plt.read, 16))
0:   ff 25 1a 18 2d 00       jmp    QWORD PTR [rip+0x2d181a]        # 0x2d1820
6:   68 59 00 00 00          push   0x59
b:   e9 50 fa ff ff          jmp    0xfffffffffffffa60

__getitem__(name)

Implement dict-like access to header entries

__init__(path, checksec=True)

__repr__()

Return repr(self).

static _decompress_dwarf_section(section)

Returns the uncompressed contents of the provided DWARF section.

_get_section_header(n)

Find the header of section #n, parse it and return the struct

_get_section_header_stringtable()

Get the string table section corresponding to the section header table.

_get_section_name(section_header)

Given a section header, find this section’s name in the file’s string table

_get_segment_header(n)

Find the header of segment #n, parse it and return the struct

_identify_file()

Verify the ELF file and identify its class and endianness.

_make_gnu_verdef_section(section_header, name)

Create a GNUVerDefSection

_make_gnu_verneed_section(section_header, name)

Create a GNUVerNeedSection

_make_gnu_versym_section(section_header, name)

Create a GNUVerSymSection

_make_section(section_header)

Create a section object of the appropriate type

_make_segment(segment_header)

Create a Segment object of the appropriate type

_make_sunwsyminfo_table_section(section_header, name)

Create a SUNWSyminfoTableSection

_make_symbol_table_index_section(section_header, name)

Create a SymbolTableIndexSection object

_make_symbol_table_section(section_header, name)

Create a SymbolTableSection

_parse_elf_header()

Parses the ELF file header and assigns the result to attributes of this object.

_patch_elf_and_read_maps()

patch_elf_and_read_maps(self) -> dict

Read /proc/self/maps as if the ELF were executing.

This is done by replacing the code at the entry point with shellcode which dumps /proc/self/maps and exits, and actually executing the binary.

Returns

A dict mapping file paths to the lowest address they appear at. Does not do any translation for e.g. QEMU emulation, the raw results are returned.

If there is not enough space to inject the shellcode in the segment which contains the entry point, returns {}.

Doctests:

These tests are just to ensure that our shellcode is correct.

>>> for arch in CAT_PROC_MAPS_EXIT:
...   context.clear()
...   with context.local(arch=arch):
...     sc = shellcraft.cat2("/proc/self/maps")
...     sc += shellcraft.exit()
...     sc = asm(sc)
...     sc = enhex(sc)
...     assert sc == CAT_PROC_MAPS_EXIT[arch], (arch, sc)

_populate_functions()

Builds a dict of ‘functions’ (i.e. symbols of type ‘STT_FUNC’) by function name that map to a tuple consisting of the func address and size in bytes.

_populate_got()

Loads the symbols for all relocations

_populate_libraries()

>>> from os.path import exists
>>> bash = ELF(which('bash'))
>>> all(map(exists, bash.libs.keys()))
True
>>> any(map(lambda x: 'libc' in x, bash.libs.keys()))
True

_populate_plt()

Loads the PLT symbols

>>> path = pwnlib.data.elf.path
>>> for test in glob(os.path.join(path, 'test-*')):
...     test = ELF(test)
...     assert '__stack_chk_fail' in test.got, test
...     if test.arch != 'ppc':
...         assert '__stack_chk_fail' in test.plt, test

_populate_symbols()

>>> bash = ELF(which('bash'))
>>> bash.symbols['_start'] == bash.entry
True

_populate_synthetic_symbols()

Adds symbols from the GOT and PLT to the symbols dictionary.

Does not overwrite any existing symbols, and prefers PLT symbols.

Synthetic plt.xxx and got.xxx symbols are added for each PLT and GOT entry, respectively.

Example:bash.

>>> bash = ELF(which('bash'))
>>> bash.symbols.wcscmp == bash.plt.wcscmp
True
>>> bash.symbols.wcscmp == bash.symbols.plt.wcscmp
True
>>> bash.symbols.stdin  == bash.got.stdin
True
>>> bash.symbols.stdin  == bash.symbols.got.stdin
True

_read_dwarf_section(section, relocate_dwarf_sections)

Read the contents of a DWARF section from the stream and return a DebugSectionDescriptor. Apply relocations if asked to.

_section_offset(n)

Compute the offset of section #n in the file

_segment_offset(n)

Compute the offset of segment #n in the file

asm(address, assembly)

Assembles the specified instructions and inserts them into the ELF at the specified address.

This modifies the ELF in-place. The resulting binary can be saved with ELF.save()

bss(offset=0) → int

Returns

Address of the .bss section, plus the specified offset.

checksec(banner=True, color=True)

Prints out information in the binary, similar to checksec.sh.

Parameters

• banner (bool) – Whether to print the path to the ELF binary.

• color (bool) – Whether to use colored output.

debug(argv=[], *a, **kw) → tube

Debug the ELF with gdb.debug().

Parameters

• argv (list) – List of arguments to the binary

• *args – Extra arguments to gdb.debug()

• **kwargs – Extra arguments to gdb.debug()

Returns

tube – See gdb.debug()

disable_nx()

Disables NX for the ELF.

Zeroes out the PT_GNU_STACK program header p_type field.

disasm(address, n_bytes) → str

Returns a string of disassembled instructions at the specified virtual memory address

dynamic_by_tag(tag) → tag

Parameters

tag (str) – Named DT_XXX tag (e.g. 'DT_STRTAB').

Returns

elftools.elf.dynamic.DynamicTag

dynamic_string(offset) → bytes

Fetches an enumerated string from the DT_STRTAB table.

Parameters

offset (int) – String index

Returns

str – String from the table as raw bytes.

dynamic_value_by_tag(tag) → int

Retrieve the value from a dynamic tag a la DT_XXX.

If the tag is missing, returns None.

fit(address, *a, **kw)

Writes fitted data into the specified address.

See: packing.fit()

flat(address, *a, **kw)

Writes a full array of values to the specified address.

See: packing.flat()

static from_assembly(assembly) → ELF

Given an assembly listing, return a fully loaded ELF object which contains that assembly at its entry point.

Parameters

• assembly (str) – Assembly language listing

• vma (int) – Address of the entry point and the module’s base address.

Example

>>> e = ELF.from_assembly('nop; foo: int 0x80', vma = 0x400000)
>>> e.symbols['foo'] = 0x400001
>>> e.disasm(e.entry, 1)
'  400000:       90                      nop'
>>> e.disasm(e.symbols['foo'], 2)
'  400001:       cd 80                   int    0x80'

static from_bytes(bytes) → ELF

Given a sequence of bytes, return a fully loaded ELF object which contains those bytes at its entry point.

Parameters

• bytes (str) – Shellcode byte string

• vma (int) – Desired base address for the ELF.

Example

>>> e = ELF.from_bytes(b'\x90\xcd\x80', vma=0xc000)
>>> print(e.disasm(e.entry, 3))
    c000:       90                      nop
    c001:       cd 80                   int    0x80

get_ehabi_infos()

Generally, shared library and executable contain 1 .ARM.exidx section. Object file contains many .ARM.exidx sections. So we must traverse every section and filter sections whose type is SHT_ARM_EXIDX.

get_machine_arch()

Return the machine architecture, as detected from the ELF header.

get_section_by_name(name)

Get a section from the file, by name. Return None if no such section exists.

get_section_index(section_name)

Gets the index of the section by name. Return None if no such section name exists.

get_segment_for_address(address, size=1) → Segment

Given a virtual address described by a PT_LOAD segment, return the first segment which describes the virtual address. An optional size may be provided to ensure the entire range falls into the same segment.

Parameters

• address (int) – Virtual address to find

• size (int) – Number of bytes which must be available after address in both the file-backed data for the segment, and the memory region which is reserved for the data.

Returns

Either returns a segments.Segment object, or None.

get_shstrndx()

Find the string table section index for the section header table

get_supplementary_dwarfinfo(dwarfinfo)

Read supplementary dwarfinfo, from either the standared .debug_sup section or the GNU proprietary .gnu_debugaltlink.

has_ehabi_info()

Check whether this file appears to have arm exception handler index table.

iter_segments_by_type(t)

Yields

Segments matching the specified type.

classmethod load_from_path(path)

Takes a path to a file on the local filesystem, and returns an ELFFile from it, setting up a correct stream_loader relative to the original file.

num_sections()

Number of sections in the file

num_segments()

Number of segments in the file

offset_to_vaddr(offset) → int

Translates the specified offset to a virtual address.

Parameters

offset (int) – Offset to translate

Returns

int – Virtual address which corresponds to the file offset, or None.

Examples

This example shows that regardless of changes to the virtual address layout by modifying ELF.address, the offset for any given address doesn’t change.

>>> bash = ELF('/bin/bash')
>>> bash.address == bash.offset_to_vaddr(0)
True
>>> bash.address += 0x123456
>>> bash.address == bash.offset_to_vaddr(0)
True

p16(address, data, *a, **kw)

Writes a 16-bit integer data to the specified address

p32(address, data, *a, **kw)

Writes a 32-bit integer data to the specified address

p64(address, data, *a, **kw)

Writes a 64-bit integer data to the specified address

p8(address, data, *a, **kw)

Writes a 8-bit integer data to the specified address

pack(address, data, *a, **kw)

Writes a packed integer data to the specified address

process(argv=[], *a, **kw) → process

Execute the binary with process. Note that argv is a list of arguments, and should not include argv[0].

Parameters

• argv (list) – List of arguments to the binary

• *args – Extra arguments to process

• **kwargs – Extra arguments to process

Returns

process

read(address, count) → bytes

Read data from the specified virtual address

Parameters

• address (int) – Virtual address to read

• count (int) – Number of bytes to read

Returns

A bytes object, or None.

Examples

The simplest example is just to read the ELF header.

>>> bash = ELF(which('bash'))
>>> bash.read(bash.address, 4)
b'\x7fELF'

ELF segments do not have to contain all of the data on-disk that gets loaded into memory.

First, let’s create an ELF file has some code in two sections.

>>> assembly = '''
... .section .A,"awx"
... .global A
... A: nop
... .section .B,"awx"
... .global B
... B: int3
... '''
>>> e = ELF.from_assembly(assembly, vma=False)

By default, these come right after eachother in memory.

>>> e.read(e.symbols.A, 2)
b'\x90\xcc'
>>> e.symbols.B - e.symbols.A
1

Let’s move the sections so that B is a little bit further away.

>>> objcopy = pwnlib.asm._objcopy()
>>> objcopy += [
...     '--change-section-vma', '.B+5',
...     '--change-section-lma', '.B+5',
...     e.path
... ]
>>> subprocess.check_call(objcopy)
0

Now let’s re-load the ELF, and check again

>>> e = ELF(e.path)
>>> e.symbols.B - e.symbols.A
6
>>> e.read(e.symbols.A, 2)
b'\x90\x00'
>>> e.read(e.symbols.A, 7)
b'\x90\x00\x00\x00\x00\x00\xcc'
>>> e.read(e.symbols.A, 10)
b'\x90\x00\x00\x00\x00\x00\xcc\x00\x00\x00'

Everything is relative to the user-selected base address, so moving things around keeps everything working.

>>> e.address += 0x1000
>>> e.read(e.symbols.A, 10)
b'\x90\x00\x00\x00\x00\x00\xcc\x00\x00\x00'

save(path=None)

Save the ELF to a file

>>> bash = ELF(which('bash'))
>>> bash.save('/tmp/bash_copy')
>>> copy = open('/tmp/bash_copy', 'rb')
>>> bash = open(which('bash'), 'rb')
>>> bash.read() == copy.read()
True

search(needle, writable=False, executable=False) → generator

Search the ELF’s virtual address space for the specified string.

Notes

Does not search empty space between segments, or uninitialized data. This will only return data that actually exists in the ELF file. Searching for a long string of NULL bytes probably won’t work.

Parameters

• needle (bytes) – String to search for.

• writable (bool) – Search only writable sections.

• executable (bool) – Search only executable sections.

Yields

An iterator for each virtual address that matches.

Examples

An ELF header starts with the bytes \x7fELF, so we sould be able to find it easily.

>>> bash = ELF('/bin/bash')
>>> bash.address + 1 == next(bash.search(b'ELF'))
True

We can also search for string the binary.

>>> len(list(bash.search(b'GNU bash'))) > 0
True

It is also possible to search for instructions in executable sections.

>>> binary = ELF.from_assembly('nop; mov eax, 0; jmp esp; ret')
>>> jmp_addr = next(binary.search(asm('jmp esp'), executable = True))
>>> binary.read(jmp_addr, 2) == asm('jmp esp')
True

section(name) → bytes

Gets data for the named section

Parameters

name (str) – Name of the section

Returns

str – String containing the bytes for that section

string(address) → str

Reads a null-terminated string from the specified address

Returns

A str with the string contents (NUL terminator is omitted), or an empty string if no NUL terminator could be found.

u16(address, *a, **kw)

Unpacks an integer from the specified address.

u32(address, *a, **kw)

Unpacks an integer from the specified address.

u64(address, *a, **kw)

Unpacks an integer from the specified address.

u8(address, *a, **kw)

Unpacks an integer from the specified address.

unpack(address, *a, **kw)

Unpacks an integer from the specified address.

vaddr_to_offset(address) → int

Translates the specified virtual address to a file offset

Parameters

address (int) – Virtual address to translate

Returns

int – Offset within the ELF file which corresponds to the address, or None.

Examples

>>> bash = ELF(which('bash'))
>>> bash.vaddr_to_offset(bash.address)
0
>>> bash.address += 0x123456
>>> bash.vaddr_to_offset(bash.address)
0
>>> bash.vaddr_to_offset(0) is None
True

write(address, data)

Writes data to the specified virtual address

Parameters

• address (int) – Virtual address to write

• data (str) – Bytes to write

Note

This routine does not check the bounds on the write to ensure that it stays in the same segment.

Examples

>>> bash = ELF(which('bash'))
>>> bash.read(bash.address+1, 3)
b'ELF'
>>> bash.write(bash.address, b"HELO")
>>> bash.read(bash.address, 4)
b'HELO'

__weakref__

list of weak references to the object (if defined)

property address

Address of the lowest segment loaded in the ELF.

When updated, the addresses of the following fields are also updated:

• symbols

• got

• plt

• functions

However, the following fields are NOT updated:

• segments

• sections

Example

>>> bash = ELF('/bin/bash')
>>> read = bash.symbols['read']
>>> text = bash.get_section_by_name('.text').header.sh_addr
>>> bash.address += 0x1000
>>> read + 0x1000 == bash.symbols['read']
True
>>> text == bash.get_section_by_name('.text').header.sh_addr
True

Type

int

arch

Architecture of the file (e.g. 'i386', 'arm').

See: ContextType.arch

Type

str

property asan

Whether the current binary was built with Address Sanitizer (ASAN).

Type

bool

property aslr

Whether the current binary is position-independent.

Type

bool

bits = 32

Bit-ness of the file

Type

int

build

Linux kernel build commit, if this is a Linux kernel image

Type

str

property buildid

GNU Build ID embedded into the binary

Type

bytes

bytes = 4

Pointer width, in bytes

Type

int

property canary

Whether the current binary uses stack canaries.

Type

bool

config

Linux kernel configuration, if this is a Linux kernel image

Type

dict

property data

Raw data of the ELF file.

See:

get_data()

Type

bytes

property dwarf

DWARF info for the elf

property elftype

ELF type (EXEC, DYN, etc)

Type

str

endian = 'little'

Endianness of the file (e.g. 'big', 'little')

Type

str

property entry

Address of the entry point for the ELF

Type

int

property entrypoint

Address of the entry point for the ELF

Type

int

property execstack

Whether the current binary uses an executable stack.

This is based on the presence of a program header PT_GNU_STACK being present, and its setting.

PT_GNU_STACK

The p_flags member specifies the permissions on the segment containing the stack and is used to indicate whether the stack should be executable. The absense of this header indicates that the stack will be executable.

In particular, if the header is missing the stack is executable. If the header is present, it may explicitly mark that the stack is executable.

This is only somewhat accurate. When using the GNU Linker, it usees DEFAULT_STACK_PERMS to decide whether a lack of PT_GNU_STACK should mark the stack as executable:

/* On most platforms presume that PT_GNU_STACK is absent and the stack is
 * executable.  Other platforms default to a nonexecutable stack and don't
 * need PT_GNU_STACK to do so.  */
uint_fast16_t stack_flags = DEFAULT_STACK_PERMS;

By searching the source for DEFAULT_STACK_PERMS, we can see which architectures have which settings.

$ git grep '#define DEFAULT_STACK_PERMS' | grep -v PF_X
sysdeps/aarch64/stackinfo.h:31:#define DEFAULT_STACK_PERMS (PF_R|PF_W)
sysdeps/nios2/stackinfo.h:31:#define DEFAULT_STACK_PERMS (PF_R|PF_W)
sysdeps/tile/stackinfo.h:31:#define DEFAULT_STACK_PERMS (PF_R|PF_W)

Type

bool

executable

True if the ELF is an executable

property executable_segments

List of all segments which are executable.

See:

ELF.segments

Type

list

file

Open handle to the ELF file on disk

Type

file

property fortify

Whether the current binary was built with Fortify Source (-DFORTIFY).

Type

bool

functions = {}

dotdict of name to Function for each function in the ELF

got = {}

dotdict of name to address for all Global Offset Table (GOT) entries

property libc

If this ELF imports any libraries which contain 'libc[.-], and we can determine the appropriate path to it on the local system, returns a new ELF object pertaining to that library.

If not found, the value will be None.

Type

ELF

property libc_start_main_return

Address of the return address into __libc_start_main from main.

>>> bash = ELF(which('bash'))
>>> libc = bash.libc
>>> libc.libc_start_main_return > 0
True

Try to find the return address from main into __libc_start_main. The heuristic to find the call to the function pointer of main is to list all calls inside __libc_start_main, find the call to exit after the call to main and select the previous call.

Type

int

library

True if the ELF is a shared library

property libs

address} for every library loaded for this ELF.

Type

Dictionary of {path

linker = None

Path to the linker for the ELF

property maps

address} for every mapping in this ELF’s address space.

Type

Dictionary of {name

memory

IntervalTree which maps all of the loaded memory segments

mmap

Memory-mapped copy of the ELF file on disk

Type

mmap.mmap

property msan

Whether the current binary was built with Memory Sanitizer (MSAN).

Type

bool

native

Whether this ELF should be able to run natively

property non_writable_segments

List of all segments which are NOT writeable.

See:

ELF.segments

Type

list

property nx

Whether the current binary uses NX protections.

Specifically, we are checking for READ_IMPLIES_EXEC being set by the kernel, as a result of honoring PT_GNU_STACK in the kernel.

The Linux kernel directly honors PT_GNU_STACK to mark the stack as executable.

case PT_GNU_STACK:
    if (elf_ppnt->p_flags & PF_X)
        executable_stack = EXSTACK_ENABLE_X;
    else
        executable_stack = EXSTACK_DISABLE_X;
    break;

Additionally, it then sets read_implies_exec, so that all readable pages are executable.

if (elf_read_implies_exec(loc->elf_ex, executable_stack))
    current->personality |= READ_IMPLIES_EXEC;

Type

bool

os

Operating system of the ELF

property packed

Whether the current binary is packed with UPX.

Type

bool

path = '/path/to/the/file'

Path to the file

Type

str

property pie

Whether the current binary is position-independent.

Type

bool

plt = {}

dotdict of name to address for all Procedure Linkate Table (PLT) entries

property relro

Whether the current binary uses RELRO protections.

This requires both presence of the dynamic tag DT_BIND_NOW, and a GNU_RELRO program header.

The ELF Specification describes how the linker should resolve symbols immediately, as soon as a binary is loaded. This can be emulated with the LD_BIND_NOW=1 environment variable.

DT_BIND_NOW

If present in a shared object or executable, this entry instructs the dynamic linker to process all relocations for the object containing this entry before transferring control to the program. The presence of this entry takes precedence over a directive to use lazy binding for this object when specified through the environment or via dlopen(BA_LIB).

(page 81)

Separately, an extension to the GNU linker allows a binary to specify a PT_GNU_RELRO program header, which describes the region of memory which is to be made read-only after relocations are complete.

Finally, a new-ish extension which doesn’t seem to have a canonical source of documentation is DF_BIND_NOW, which has supposedly superceded DT_BIND_NOW.

DF_BIND_NOW

If set in a shared object or executable, this flag instructs the dynamic linker to process all relocations for the object containing this entry before transferring control to the program. The presence of this entry takes precedence over a directive to use lazy binding for this object when specified through the environment or via dlopen(BA_LIB).

>>> path = pwnlib.data.elf.relro.path
>>> for test in glob(os.path.join(path, 'test-*')):
...     e = ELF(test)
...     expected = os.path.basename(test).split('-')[2]
...     actual = str(e.relro).lower()
...     assert actual == expected

Type

bool

property rpath

Whether the current binary has an RPATH.

Type

bool

property runpath

Whether the current binary has a RUNPATH.

Type

bool

property rwx_segments

List of all segments which are writeable and executable.

See:

ELF.segments

Type

list

property sections

A list of elftools.elf.sections.Section objects for the segments in the ELF.

Type

list

property segments

A list of elftools.elf.segments.Segment objects for the segments in the ELF.

Type

list

property start

Address of the entry point for the ELF

Type

int

statically_linked

True if the ELF is statically linked

property sym

Alias for ELF.symbols

Type

dotdict

symbols = {}

dotdict of name to address for all symbols in the ELF

property ubsan

Whether the current binary was built with Undefined Behavior Sanitizer (UBSAN).

Type

bool

version

Linux kernel version, if this is a Linux kernel image

Type

tuple

property writable_segments

List of all segments which are writeable.

See:

ELF.segments

Type

list

class pwnlib.elf.elf.Function(name, address, size, elf=None)

Encapsulates information about a function in an ELF binary.

Parameters

• name (str) – Name of the function

• address (int) – Address of the function

• size (int) – Size of the function, in bytes

• elf (ELF) – Encapsulating ELF object

__init__(name, address, size, elf=None)

__repr__()

Return repr(self).

__weakref__

list of weak references to the object (if defined)

address

Address of the function in the encapsulating ELF

elf

Encapsulating ELF object

name

Name of the function

size

Size of the function, in bytes

class pwnlib.elf.elf.dotdict

Wrapper to allow dotted access to dictionary elements.

Is a real dict object, but also serves up keys as attributes when reading attributes.

Supports recursive instantiation for keys which contain dots.

Example

>>> x = pwnlib.elf.elf.dotdict()
>>> isinstance(x, dict)
True
>>> x['foo'] = 3
>>> x.foo
3
>>> x['bar.baz'] = 4
>>> x.bar.baz
4

__weakref__

list of weak references to the object (if defined)