pwndbg/DEVELOPING.md

31 KiB

Development Basics

Environment setup

After installing pwndbg by running setup.sh, you additionally need to run ./setup-dev.sh to install the necessary development dependencies.

If you would like to use Docker, you can create a Docker image with everything already installed for you. To build and run the container, run the following commands:

docker build -t pwndbg . 
docker run -it --cap-add=SYS_PTRACE --security-opt seccomp=unconfined -v `pwd`:/pwndbg pwndbg bash

If you'd like to use docker compose, you can run

docker compose run -i main

Development using Nix

There is a development shell defined in the flake that should install all of the development requirements. To enter the environment run nix develop or automatically enter the environment using direnv.

When testing changes run nix build .#pwndbg-dev and use the copy of the files in the results/ folder.

Testing

It's highly recommended you write a new test or update an existing test whenever adding new functionality to pwndbg.

We have four types of tests: gdb-tests,qemu-tests, unit-tests, and Linux kernel tests, which are all located in subdirectories of tests.

gdb-tests refers to our x86 tests, which are located tests/gdb-tests.

To run these tests, run ./tests.sh. You can filter the tests to run by providing an argument to the script, such as ./tests.sh heap, which will only run tests that contain "heap" in the name. You can also drop into the PDB debugger when a test fails with ./tests.sh --pdb.

To invoke cross-architecture tests, use ./qemu-tests.sh, and to run unit tests, use ./unit-tests.sh

Writing Tests

Each test is a Python function that runs inside of an isolated GDB session. Using a pytest fixture at the beginning of each test, GDB will attach to a binary or connect to a QEMU instance. Each test runs some commands and uses Python assert statements to verify correctness. We can access pwndbg library code like pwndbg.gdblib.regs.rsp as well as execute GDB commands with gdb.execute().

We can take a look at tests/gdb-tests/tests/test_symbol.py for an example of a simple test. Looking at a simplified version of the top-level code, we have this:

import gdb
import pwndbg
import tests

BINARY = tests.binaries.get("symbol_1600_and_752.out")

Since these tests run inside GDB, we can import the gdb Python library. We also import the tests module, which makes it easy to get the path to the test binaries located in tests/gdb-tests/tests/binaries. You should be able to reuse the binaries in this folder for most tests, but if not feel free to add a new one.

Here's a small snippet of the actual test:

def test_hexdump(start_binary):
    start_binary(BINARY)
    pwndbg.gdblib.config.hexdump_group_width = -1

    gdb.execute("set hexdump-byte-separator")
    stack_addr = pwndbg.gdblib.regs.rsp - 0x100

pytest will run any function that starts with test_ as a new test, so there is no need to register your new test anywhere. The start_binary argument is a function that will run the binary you give it, and it will set some common options before starting the binary. Using start_binary is recommended if you don't need any additional customization to GDB settings before starting the binary, but if you do it's fine to not use it.

QEMU Tests

Our gdb-tests run in x86. To debug other architectures, we use QEMU for emulation, and attach to its debug port. These tests are located in tests/qemu-tests/tests/user. Test creation is identical to our x86 tests - create a Python function with a Pytest fixture name as the parameter (it matches based on the name), and call the argument to start debugging a binary. The qemu_assembly_run fixture takes in a Python string of assembly code, compiles it in the appropriate architecture, and runs it - no need to create an external file or edit a Makefile.

Kernel Tests

We use qemu-system for full system level emulation for our Linux kernel tests. These are located in tests/qemu-tests/tests/system. The tests will run for a variety kernel configurations and architectures.

Testing Under Nix

You will need to build a nix-compatible gdbinit.py file, which you can do with nix build .#pwndbg-dev. Then simply run the test by adding the --nix flag:

./tests.sh --nix [filter]

Linting

The lint.sh script runs isort, black, ruff, shfmt, and vermin. isort and black are able to automatically fix any issues they detect, and you can enable this by running ./lint.sh -f. You can find the configuration files for these tools in setup.cfg and pyproject.toml.

When submitting a PR, the CI job defined in .github/workflows/lint.yml will verify that running ./lint.sh succeeds, otherwise the job will fail and we won't be able to merge your PR.

You can optionally set the contents of .git/hooks/pre-push to the following if you would like lint.sh to automatically be run before every push:

#!/usr/bin/env bash

./lint.sh || exit 1

Minimum Supported Versions

Our goal is to fully support all Ubuntu LTS releases that have not reach end-of-life, with support for other platforms on a best-effort basis. Currently that means all code should work on Ubuntu 22.04, and 24.04 with GDB 12.1 and later. This means that the minimum supported Python version is 3.10, and we cannot use any newer Python features unless those features are backported to this minimum version.

Note that while all code should run without errors on these supported LTS versions, it's fine if older versions don't support all of the features of newer versions, as long as this is handled correctly and this information is shown to the user. For example, we may make use of some GDB APIs in newer versions that we aren't able to provide alternative implementations for in older versions, and so in these cases we should inform the user that the functionality can't be provided due to the version of GDB.

The lint.sh script described in the previous section runs vermin to ensure that our code does not use any features that aren't supported on Python 3.6.

Adding a Command

Create a new Python file in pwndbg/commands/my_command.py, replacing my_command with a reasonable name for the command. The most basic command looks like this:

import argparse

import pwndbg.commands

parser = argparse.ArgumentParser(description="Command description.")
parser.add_argument("arg", type=str, help="An example argument.")


@pwndbg.commands.ArgparsedCommand(parser)
def my_command(arg: str) -> None:
    """Print the argument"""
    print(f"Argument is {arg}")

In addition, you need to import this file in the load_commands function in pwndbg/commands/__init__.py. After this, running my_command foo in GDB will print out "Argument is foo".

Adding a Configuration Option

import pwndbg.gdblib.config

pwndbg.gdblib.config.add_param("config-name", False, "example configuration option")

pwndbg.gdblib.config.config_name will now refer to the value of the configuration option, and it will default to False if not set.

Configuration Docstrings

TODO: There are many places GDB shows docstrings, and they show up slightly differently in each place, we should give examples of this

  • When using pwndbg.gdblib.config.add_param to add a new config, there are a few things to keep in mind:
    • For the set_show_doc parameter, it is best to use a noun phrase like "the value of something" to ensure that the output is grammatically correct.
    • For the help_docstring parameter, you can use the output of help set follow-fork-mode as a guide for formatting the documentation string if the config is an enum type.
    • For the param_class parameter
      • See the documentation for more information.
      • If you use gdb.PARAM_ENUM as param_class, you must pass a list of strings to the enum_sequence parameter.

Triggers

TODO: If we want to do something when user changes config/theme - we can do it defining a function and decorating it with pwndbg.config.Trigger.

Porting public tools

If porting a public tool to pwndbg, please make a point of crediting the original author. This can be added to CREDITS.md noting the original author/inspiration, and linking to the original tool/article. Also please be sure that the license of the original tool is suitable to porting into pwndbg, such as MIT.

Random developer notes

Feel free to update the list below!

  • If you want to play with pwndbg functions under GDB, you can always use GDB's pi which launches python interpreter or just py <some python line>.

  • If there is possibility, don't use gdb.execute as this requires us to parse the string and so on; there are some cases in which there is no other choice. Most of the time we try to wrap GDB's API to our own/easier API.

  • We have our own pwndbg.config.Parameter (which extends gdb.Parameter) - all of our parameters can be seen using config or theme commands.

  • The dashboard/display/context we are displaying is done by pwndbg/commands/context.py which is invoked through GDB's prompt hook (which we defined in pwndbg/prompt.py as prompt_hook_on_stop).

  • We change a bit GDB settings - this can be seen in pwndbg/__init__.py - there are also imports for all pwndbg submodules

  • We have a wrapper for GDB's events in pwndbg/events.py - thx to that we can e.g. invoke something based upon some event

  • We have a caching mechanism ("memoization") which we use through Python's decorators - those are defined in pwndbg/lib/cache.py - just check its usages

  • To block a function before the first prompt was displayed use the pwndbg.decorators.only_after_first_prompt decorator.

  • Memory accesses should be done through pwndbg/memory.py functions

  • Process properties can be retrieved thx to pwndbg/gdblib/proc.py - e.g. using pwndbg.gdblib.proc.pid will give us current process pid

  • We have a wrapper for handling exceptions that are thrown by commands - defined in pwndbg/exception.py - current approach seems to work fine - by using set exception-verbose on - we get a stacktrace. If we want to debug stuff we can always do set exception-debugger on.

  • Some of pwndbg's functionality - e.g. memory fetching - require us to have an instance of proper gdb.Type - the problem with that is that there is no way to define our own types - we have to ask gdb if it detected particular type in this particular binary (that sucks). We do it in pwndbg/typeinfo.py and it works most of the time. The known bug with that is that it might not work properly for Golang binaries compiled with debugging symbols.

  • If you want to use gdb.parse_and_eval("a_function_name()") or something similar that call a function, please remember this might cause another thread to continue execution without set scheduler-locking on. If you didn't expect that, you should use parse_and_eval_with_scheduler_lock from pwndbg.gdblib.scheduler instead.

Annotations

Alongside the disassembled instructions in the dashboard, Pwndbg also has the ability to display annotations - text that contains relevent information regarding the execution of the instruction. For example, on the x86 MOV instruction, we can display the concrete value that gets placed into the destination register. Likewise, we can indicate the results of mathematical operations and memory accesses. The annotation in question is always dependent on the exact instruction being annotated - we handle it in a case-by-case basis.

The main hurdle in providing annotations is determining what each instruction does, getting the relevent CPU registers and memory that are accessed, and then resolving concrete values of the operands. We call the process of determining this information "enhancement", as we enhance the information provided natively by GDB.

The Capstone Engine disassembly framework is used to statically determine information about instructions and their operands. Take the x86 instruction sub rax, rdx. Given the raw bytes of the machine instructions, Capstone creates an object that provides an API that, among many things, exposes the names of the operands and the fact that they are both 8-byte wide registers. It provides all the information necessary to describe each operand. It also tells the general 'group' that a instruction belongs to, like if its a JUMP-like instruction, a RET, or a CALL. These groups are architecture agnostic.

However, the Capstone Engine doesn't fill in concrete values that those registers take on. It has no way of knowing the value in rdx, nor can it actually read from memory.

To determine the actual values that the operands take on, and to determine the results of executing an instruction, we use the Unicorn Engine, a CPU emulator framework. The emulator has its own internal CPU register set and memory pages that mirror that of the host process, and it can execute instructions to mutate its internal state. Note that the Unicorn Engine cannot execute syscalls - it doesn't have knowledge of a kernel.

We have the ability to single-step the emulator - tell it to execute the instruction at the program counter inside the emulator. After doing so, we can inspect the state of the emulator - read from its registers and memory. The Unicorn Engine itself doesn't expose information regarding what each instruction is doing - what is the instruction (is it an add, mov, push?) and what registers/memory locations is it reading to and writing from? - which is why we use the Capstone engine to statically determine this information.

Using what we know about the instruction based on the Capstone engine - such as that it was a sub instruction and rax was written to - we query the emulator after stepping in to determine the results of the instruction.

We also read the program counter from the emulator to determine jumps and so we can display the instructions that will actually be executed, as opposed to displaying the instructions that follow consecutively in memory.

Enhancing

Everytime the inferior process stops (and when the disasm context section is displayed), we display the next handful of assembly instructions in the dashboard so the user can understand where the process is headed. The exact amount is determined by the context-disasm-lines setting.

We will be enhancing the instruction at the current program counter, as well as all the future instructions that are displayed. The end result of enhancement is that we get a list of PwndbgInstruction objects, each encapsulating relevent information regarding the instructions execution.

When the process stops, we instantiate the emulator from scratch. We copy all the registers from the host process into the emulator. For performance purposes, we register a handler to the Unicorn Engine to lazily map memory pages from the host to the emulator when they are accessed (a page fault from within the emulator), instead of immediately copying all the memory from the host to the emulator.

The enhancement is broken into a couple stops:

  1. First, we resolve the values of all the operands of the instruction before stepping the emulator. This means we read values from registers and dereference memory depending on the operand type. This gives us the values of operands before the instruction executes.
  2. Then, we step the emulator, executing a single instruction.
  3. We resolve the values of all operands again, giving us the after_value of each operand.
  4. Then, we enhance the "condition" field of PwndbgInstructions, where we determine if the instruction is conditional (conditional branch or conditional mov are common) and if the action is taken.
  5. We then determine the next and target fields of PwndbgInstructions. next is the address that the program counter will take on after using the GDB command nexti, and target indicates the target address of branch/jump/PC-changing instructions.
  6. With all this information determined, we now effectively have a big switch statement, matching on the instruction type, where we set the annotation string value, which is the text that will be printed alongside the instruction in question.

We go through the enhancement process for the instruction at the program counter and then ensuing handful of instructions that are shown in the dashboard.

When to use emulation / reasoning about process state

When possible, we code aims to use emulation as little as possible. If there is information that can be determined statically or without the emulator, then we try to avoid emulation. This is so we can display annotations even when the Unicorn Engine is disabled. For example, say we come to a stop, and are faced with enhancing the following three instructions in the dashboard:

1.     lea    rax, [rip + 0xd55]
2. >   mov    rsi, rax      # The host process program counter is here
3.     mov    rax, rsi

Instruction #1, the lea instruction, is already in the past - we pull our enhanced PwndbgInstruction for it from a cache.

Instruction #2, the first mov instruction, is where the host process program counter is at. If we did stepi in GDB, this instruction would be executed. In this case, there is two ways we can determine the value that gets written to rsi.

  1. After stepping the emulator, read from the emulators rsi register.
  2. Given the context of the instruction, we know the value in rsi will come from rax. We can just read the rax register from the host. This avoids emulation.

The decision on which option to take is implemented in the annotation handler for the specific instruction. When possible, we have a preference for the second option, because it makes the annotations work even when emulation is off.

The reason we could do the second option, in this case, is because we could reason about the process state at the time this instruction would execute. This instruction is about to be executed (Program PC == instruction.address). We can safely read from rax from the host, knowing that the value we get is the true value it takes on when the instruction will execute. It must - there are no instructions in-between that could have mutated rax.

However, this will not be the case while enhancing instruction #3 while we are paused at instruction #2. This instruction is in the future, and without emulation, we cannot safely reason about the operands in question. It is reading from rsi, which might be mutated from the current value that rsi has in the stopped process (and in this case, we happen to know that it will be mutated). We must use emulation to determine the before_value of rsi in this case, and can't just read from the host processes register set. This principle applies in general - future instructions must be emulated to be fully annotated. When emulation is disable, the annotations are not as detailed since we can't fully reason about process state for future instructions.

What if the emulator fails?

It is possible for the emulator to fail to execute an instruction - either due to a restrictions in the engine itself, or the instruction inside segfaults and cannot continue. If the Unicorn Engine fails, there is no real way we can recover. When this happens, we simply stop emulating for the current step, and we try again the next time the process stops when we instantiate the emulator from scratch again.

Caching annotations

When we are stepping through the emulator, we want to remember the annotations of the past couple instructions. We don't want to nexti, and suddenly have the annotation of the previously executed instruction deleted. At the same time, we also never want stale annotations that might result from coming back to point in the program to which we have stepped before, such as the middle of a loop via a breakpoint.

New annotations are only created when the process stops, and we create annotations for next handful of instructions to be executed. If we continue in GDB and stop at a breakpoint, we don't want annotations to appear behind the PC that are from a previous time we were near the location in question. To avoid stale annotations while still remembering them when stepping, we have a simple caching method:

While we are doing our enhancement, we create a list containing the addresses of the future instructions that are displayed.

For example, say we have the following instructions with the first number being the memory address:

   0x555555556259 <main+553>    lea    rax, [rsp + 0x90]
   0x555555556261 <main+561>    mov    edi, 1                          EDI => 1
   0x555555556266 <main+566>    mov    rsi, rax
   0x555555556269 <main+569>    mov    qword ptr [rsp + 0x78], rax
   0x55555555626e <main+574>    call   qword ptr [rip + 0x6d6c]    <fstat64>

 ► 0x555555556274 <main+580>    mov    edx, 5                  EDX => 5
   0x555555556279 <main+585>    lea    rsi, [rip + 0x3f30]     RSI => 0x55555555a1b0 ◂— 'standard output'
   0x555555556280 <main+592>    test   eax, eax
   0x555555556282 <main+594>    js     main+3784                   <main+3784>

   0x555555556288 <main+600>    mov    rsi, qword ptr [rsp + 0xc8]
   0x555555556290 <main+608>    mov    edi, dword ptr [rsp + 0xa8]

In this case, our next_addresses_cache would be [0x555555556279, 0x555555556280, 0x555555556282, 0x555555556288, 0x555555556290].

Then, the next time our program comes to a stop (after using si, n, or any GDB command that continues the process), we immediately check if the current program counter is in this list. If it is, then we can infer that the annotations are still valid, as the program has only executed a couple instructions. In all other cases, we delete our cache of annotated instructions.

We might think "why not just check if it's the next address - 0x555555556279 in this case? Why a list of the next couple addresses?". This is because when source code is available, step and next often skip a couple instructions. It would be jarring to remove the annotations in this case. Likewise, this method has the added benefit that if we stop somewhere, and there happens to be a breakpoint only a couple instructions in front of us that we continue to, then previous couple annotations won't be wiped.

Other random annotation details

  • We don't emulate through CALL instructions. This is because the function might be very long.
  • We resolve symbols during the enhancement stage for operand values.
  • The folder pwndbg/gdblib/disasm contains the code for enhancement. It follows an object-oriented model, with arch.py implementing the parent class with shared functionality, and the per-architecture implementations are implemented as subclasses in their own files.
  • pwndbg/gdblib/nearpc.py is responsible for getting the list of enhanced PwndbgInstruction objects and converting them to the output seen in the 'disasm' view of the dashboard.

Adding or fixing annotations

We annotate on an instruction-by-instruction basis. Effectively, imagine a giant switch statement that selects the correct handler to create an annotation based on the specific instruction. Many instruction types can be grouped and annotated using the same logic, such as load, store, and arithmetic instructions.

See pwndbg/gdblib/disasm/aarch64.py as an example. We define sets that group instructions using the unique Capstone ID for each instruction, and inside the constructor of DisassemblyAssistant we have a mapping of instructions to a specific handler. The _set_annotation_string function will match the instruction to the correct handler, which set the instruction.annotation field.

If there is a bug in an annotation, the first order of business is finding its annotation handler. To track down where we are handling the instruction, you can search for its Capstone constant. For example, the RISC-V store byte instruction, sb, is represented as the Capstone constant RISCV_INS_SB. Or, if you are looking for the handler for the AArch64 instruction SUB, you can search the disasm code for _INS_SUB to find where we reference the appropriate Capstone constant for the instruction and following the code to the function that ultimately sets the annotation.

If an annotation is causing a crash, is it most likely due to a handler making an incorrect assumption on the number of operands, leading to a list index out of range error. One possible source of this is that a given instruction has multiple different disassembly representations. Take the RISC-V JALR instruction. It can be represented in 3 ways:

jalr rs1        # return register is implied as ra, and imm is implied as 0
jalr rs1, imm   # return register is implied as ra
jalr rd, rs1, imm

Capstone will expose the most "simplified" one possible, and the underlying list of register operands will change. If the handler doesn't take these different options into account, and rather assumes that jalr always has 3 operands, then an index error can occur if the handler accesses instruction.operands[2].

Bug root cause

When encountering an instruction that is behaving strangely (incorrect annotation, or there is a jump target when one shouldn't exist, or the target is incorrect), there are a couple routine things to check.

  1. Use the dev_dump_instruction command to print all the enhancement information. With no arguments, it will dump the info from the instruction at the current address. If given an address, it will pull from the instruction cache at the corresponding location.

If the issue is not related to branches, check the operands and the resolved values for registers and memory accesses. Verify that the values are correct - are the resolved memory locations correct? Step past the instruction and use instructions like telescope and regs to read memory and verify if the claim that the annotation is making is correct. For things like memory operands, you can try to look around the resolved memory location in memory to see the actual value that the instruction dereferenced, and see if the resolved memory location is simply off by a couple bytes.

Example output of dumping a mov instruction:

mov qword ptr [rsp], rsi at 0x55555555706c (size=4) (arch: x86)
        ID: 460, mov
        Raw asm: mov    qword ptr [rsp], rsi
        New asm: mov    qword ptr [rsp], rsi
        Next: 0x555555557070
        Target: 0x555555557070, Target string=, const=None
        Condition: UNDETERMINED
        Groups: []
        Annotation: [0x7fffffffe000] => 0x7fffffffe248 —▸ 0x7fffffffe618 ◂— '/usr/bin/ls'
        Operands: [['[0x7fffffffe000]': Symbol: None, Before: 0x7fffffffe000, After: 0x7fffffffe000, type=CS_OP_MEM, size=8, access=CS_AC_WRITE]] ['RSI': Symbol: None, Before: 0x7fffffffe248, After: 0x7fffffffe248, type=CS_OP_REG, size=8, access=CS_AC_READ]]]
        Conditional jump: False. Taken: False
        Unconditional jump: False
        Declare unconditional: None
        Can change PC: False
        Syscall:  N/A
        Causes Delay slot: False
        Split: NO_SPLIT
        Call-like: False
  1. Use the Capstone disassembler to verify the number of operands the instruction groups.

Taken the raw instruction bytes and pass them to cstool to see the information that we are working with:

cstool -d mips 0x0400000c

The number of operands may not match the visual appearance. You might also check the instruction groups, and verify that an instruction that we might consider a call has the Capstone call group. Capstone is not 100% correct in every single case in all architectures, so it's good to verify. Report a bug to Capstone if there appears to be an error, and in the meanwhile we can create a fix in Pwndbg to work around the current behavior.

  1. Check the state of the emulator.

Go to pwndbg/emu/emulator.py and uncomment the DEBUG = -1 line. This will enable verbose debug printing. The emulator will print it's current pc at every step, and indicate important events, like memory mappings. Likewise, in pwndbg/gdblib/disasm/arch.py you can set DEBUG_ENHANCEMENT = True to print register accesses to verify they are sane values.

Potential bugs:

  • A register is 0 (may also be the source of a Unicorn segfault if used as a memory operand) - often means we are not copying the host processes register into the emulator. By default, we map register by name - if in pwndbg, it's called rax, then we find the UC constant named U.x86_const.UC_X86_REG_RAX. Sometimes, this default mapping doesn't work, sometimes do to differences in underscores (FSBASE vs FS_BASE). In these cases, we have to manually add the mapping.
  • Unexpected crash - the instruction at hand might require a 'coprocessor', or some information that is unavailable to Unicorn (it's QEMU under the hood).
  • Instructions are just no executing - we've seen this in the case of Arm Thumb instructions. There might be some specific API/way to invoke the emulator that is required for a certain processor state.

Creating small cross-architecture programs

If you are encountering a strange behavior with a certain instruction or scenario in a non-native-architecture program, you can use some great functions from pwntools to handle the compilation and debugging. This is a great way to create a small reproducible example to isolate an issue.

The following Python program, when run from inside a tmux session, will take some AArch64 assembly, compile it, and run it with GDB attached in a new tmux pane. It will search your system for the appropriate cross compiler for the architecture at hand, and run the compiled binary with QEMU.

from pwn import *

context.arch = "aarch64"

AARCH64_GRACEFUL_EXIT = """
mov x0, 0
mov x8, 93
svc 0
"""

out = make_elf_from_assembly(STORE)
# Debug info
print(out)
gdb.debug(out)

pause()