Skip to main content

A coverage-guided fuzzer for Python and Python extensions.

Project description

Atheris: A Coverage-Guided, Native Python Fuzzer

Atheris is a coverage-guided Python fuzzing engine. It supports fuzzing of Python code, but also native extensions written for CPython. Atheris is based off of libFuzzer. When fuzzing native code, Atheris can be used in combination with Address Sanitizer or Undefined Behavior Sanitizer to catch extra bugs.

Installation Instructions

Atheris supports Linux (32- and 64-bit) and Mac OS X, Python versions 3.6-3.9.

You can install prebuilt versions of Atheris with pip:

pip3 install atheris

These wheels come with a built-in libFuzzer, which is fine for fuzzing Python code. If you plan to fuzz native extensions, you may need to build from source to ensure the libFuzzer version in Atheris matches your Clang version.

Building from Source

Atheris relies on libFuzzer, which is distributed with Clang. If you have a sufficiently new version of clang on your path, installation from source is as simple as:

# Build latest release from source
pip3 install --no-binary atheris atheris
# Build development code from source
git clone https://github.com/google/atheris.git
cd atheris
pip3 install .

If you don't have clang installed or it's too old, you'll need to download and build the latest version of LLVM. Follow the instructions in Installing Against New LLVM below.

Mac

Apple Clang doesn't come with libFuzzer, so you'll need to install a new version of LLVM from head. Follow the instructions in Installing Against New LLVM below.

Installing Against New LLVM

# Building LLVM
git clone https://github.com/llvm/llvm-project.git
cd llvm-project
mkdir build
cd build
cmake -DLLVM_ENABLE_PROJECTS='clang;compiler-rt' -G "Unix Makefiles" ../llvm
make -j 10  # This step is very slow

# Installing Atheris
CLANG_BIN="$(pwd)/bin/clang" pip3 install <whatever>

Using Atheris

Example

import atheris

with atheris.instrument_imports():
  import some_library
  import sys

def TestOneInput(data):
  some_library.parse(data)

atheris.Setup(sys.argv, TestOneInput)
atheris.Fuzz()

When fuzzing Python, Atheris will report a failure if the Python code under test throws an uncaught exception.

Python coverage

Atheris collects Python coverage information by instrumenting bytecode. There are 3 options for adding this instrumentation to the bytecode:

  • You can instrument the libraries you import:

    with atheris.instrument_imports():
      import foo
      from bar import baz
    

    This will cause instrumentation to be added to foo and bar, as well as any libraries they import.

  • Or, you can instrument individual functions:

    @atheris.instrument_func
    def my_function(foo, bar):
      print("instrumented")
    
  • Or finally, you can instrument everything:

    atheris.instrument_all()
    

    Put this right before atheris.Setup(). This will find every Python function currently loaded in the interpreter, and instrument it. This might take a while.

Atheris can additionally instrument regular expression checks, e.g. re.search. To enable this feature, you will need to add: atheris.enabled_hooks.add("RegEx") To your script before your code calls re.compile. Internally this will import the re module and instrument the necessary functions. This is currently an experimental feature.

Why am I getting "No interesting inputs were found"?

You might see this error:

ERROR: no interesting inputs were found. Is the code instrumented for coverage? Exiting.

You'll get this error if the first 2 calls to TestOneInput didn't produce any coverage events. Even if you have instrumented some Python code, this can happen if the instrumentation isn't reached in those first 2 calls. (For example, because you have a nontrivial TestOneInput). You can resolve this by adding an atheris.instrument_func decorator to TestOneInput, using atheris.instrument_all(), or moving your TestOneInput function into an instrumented module.

Visualizing Python code coverage

Examining which lines are executed is helpful for understanding the effectiveness of your fuzzer. Atheris is compatible with coverage.py: you can run your fuzzer using the coverage.py module as you would for any other Python program. Here's an example:

python3 -m coverage run your_fuzzer.py -atheris_runs=10000  # Times to run
python3 -m coverage html
(cd htmlcov && python3 -m http.server 8000)

Coverage reports are only generated when your fuzzer exits gracefully. This happens if:

  • you specify -atheris_runs=<number>, and that many runs have elapsed.
  • your fuzzer exits by Python exception.
  • your fuzzer exits by sys.exit().

No coverage report will be generated if your fuzzer exits due to a crash in native code, or due to libFuzzer's -runs flag (use -atheris_runs). If your fuzzer exits via other methods, such as SIGINT (Ctrl+C), Atheris will attempt to generate a report but may be unable to (depending on your code). For consistent reports, we recommend always using -atheris_runs=<number>.

If you'd like to examine coverage when running with your corpus, you can do that with the following command:

python3 -m coverage run your_fuzzer.py corpus_dir/* -atheris_runs=$(ls corpus_dir | wc -l)

This will cause Atheris to run on each file in <corpus-dir>, then exit. Importantly, if you leave off the -atheris_runs=$(ls corpus_dir | wc -l), no coverage report will be generated.

Using coverage.py will significantly slow down your fuzzer, so only use it for visualizing coverage; don't use it all the time.

Fuzzing Native Extensions

In order for fuzzing native extensions to be effective, your native extensions must be instrumented. See Native Extension Fuzzing for instructions.

Integration with OSS-Fuzz

Atheris is fully supported by OSS-Fuzz, Google's continuous fuzzing service for open source projects. For integrating with OSS-Fuzz, please see https://google.github.io/oss-fuzz/getting-started/new-project-guide/python-lang.

API

The atheris module provides three key functions: instrument_imports(), Setup() and Fuzz().

In your source file, import all libraries you wish to fuzz inside a with atheris.instrument_imports():-block, like this:

# library_a will not get instrumented
import library_a

with atheris.instrument_imports():
    # library_b will get instrumented
    import library_b

Generally, it's best to import atheris first and then import all other libraries inside of a with atheris.instrument_imports() block.

Next, define a fuzzer entry point function and pass it to atheris.Setup() along with the fuzzer's arguments (typically sys.argv). Finally, call atheris.Fuzz() to start fuzzing. You must call atheris.Setup() before atheris.Fuzz().

instrument_imports(include=[], exclude=[])

  • include: A list of fully-qualified module names that shall be instrumented.
  • exclude: A list of fully-qualified module names that shall NOT be instrumented.

This should be used together with a with-statement. All modules imported in said statement will be instrumented. However, because Python imports all modules only once, this cannot be used to instrument any previously imported module, including modules required by Atheris. To add coverage to those modules, use instrument_all() instead.

A full list of unsupported modules can be retrieved as follows:

import sys
import atheris
print(sys.modules.keys())

instrument_func(func)

  • func: The function to instrument.

This will instrument the specified Python function and then return func. This is typically used as a decorator, but can be used to instrument individual functions too. Note that the func is instrumented in-place, so this will affect all call points of the function.

This cannot be called on a bound method - call it on the unbound version.

instrument_all()

This will scan over all objects in the interpreter and call instrument_func on every Python function. This works even on core Python interpreter functions, something which instrument_imports cannot do.

This function is experimental.

Setup(args, test_one_input, internal_libfuzzer=None)

  • args: A list of strings: the process arguments to pass to the fuzzer, typically sys.argv. This argument list may be modified in-place, to remove arguments consumed by the fuzzer. See the LibFuzzer docs for a list of such options.
  • test_one_input: your fuzzer's entry point. Must take a single bytes argument. This will be repeatedly invoked with a single bytes container.
  • internal_libfuzzer: Indicates whether libfuzzer will be provided by atheris or by an external library (see native_extension_fuzzing.md). If unspecified, Atheris will determine this automatically. If fuzzing pure Python, leave this as True.

Fuzz()

This starts the fuzzer. You must have called Setup() before calling this function. This function does not return.

In many cases Setup() and Fuzz() could be combined into a single function, but they are separated because you may want the fuzzer to consume the command-line arguments it handles before passing any remaining arguments to another setup function.

FuzzedDataProvider

Often, a bytes object is not convenient input to your code being fuzzed. Similar to libFuzzer, we provide a FuzzedDataProvider to translate these bytes into other input forms.

You can construct the FuzzedDataProvider with:

fdp = atheris.FuzzedDataProvider(input_bytes)

The FuzzedDataProvider then supports the following functions:

def ConsumeBytes(count: int)

Consume count bytes.

def ConsumeUnicode(count: int)

Consume unicode characters. Might contain surrogate pair characters, which according to the specification are invalid in this situation. However, many core software tools (e.g. Windows file paths) support them, so other software often needs to too.

def ConsumeUnicodeNoSurrogates(count: int)

Consume unicode characters, but never generate surrogate pair characters.

def ConsumeString(count: int)

Alias for ConsumeBytes in Python 2, or ConsumeUnicode in Python 3.

def ConsumeInt(int: bytes)

Consume a signed integer of the specified size (when written in two's complement notation).

def ConsumeUInt(int: bytes)

Consume an unsigned integer of the specified size.

def ConsumeIntInRange(min: int, max: int)

Consume an integer in the range [min, max].

def ConsumeIntList(count: int, bytes: int)

Consume a list of count integers of size bytes.

def ConsumeIntListInRange(count: int, min: int, max: int)

Consume a list of count integers in the range [min, max].

def ConsumeFloat()

Consume an arbitrary floating-point value. Might produce weird values like NaN and Inf.

def ConsumeRegularFloat()

Consume an arbitrary numeric floating-point value; never produces a special type like NaN or Inf.

def ConsumeProbability()

Consume a floating-point value in the range [0, 1].

def ConsumeFloatInRange(min: float, max: float)

Consume a floating-point value in the range [min, max].

def ConsumeFloatList(count: int)

Consume a list of count arbitrary floating-point values. Might produce weird values like NaN and Inf.

def ConsumeRegularFloatList(count: int)

Consume a list of count arbitrary numeric floating-point values; never produces special types like NaN or Inf.

def ConsumeProbabilityList(count: int)

Consume a list of count floats in the range [0, 1].

def ConsumeFloatListInRange(count: int, min: float, max: float)

Consume a list of count floats in the range [min, max]

def PickValueInList(l: list)

Given a list, pick a random value

def ConsumeBool()

Consume either True or False.

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Files for atheris, version 2.0.7
Filename, size File type Python version Upload date Hashes
Filename, size atheris-2.0.7.tar.gz (65.4 kB) File type Source Python version None Upload date Hashes View
Filename, size atheris-2.0.7-cp39-cp39-manylinux2014_x86_64.whl (26.9 MB) File type Wheel Python version cp39 Upload date Hashes View
Filename, size atheris-2.0.7-cp39-cp39-macosx_11_0_x86_64.whl (2.6 MB) File type Wheel Python version cp39 Upload date Hashes View
Filename, size atheris-2.0.7-cp38-cp38-manylinux2014_x86_64.whl (26.9 MB) File type Wheel Python version cp38 Upload date Hashes View
Filename, size atheris-2.0.7-cp38-cp38-macosx_11_0_x86_64.whl (2.6 MB) File type Wheel Python version cp38 Upload date Hashes View
Filename, size atheris-2.0.7-cp37-cp37m-manylinux2014_x86_64.whl (27.1 MB) File type Wheel Python version cp37 Upload date Hashes View
Filename, size atheris-2.0.7-cp37-cp37m-macosx_11_0_x86_64.whl (2.6 MB) File type Wheel Python version cp37 Upload date Hashes View
Filename, size atheris-2.0.7-cp36-cp36m-manylinux2014_x86_64.whl (27.1 MB) File type Wheel Python version cp36 Upload date Hashes View
Filename, size atheris-2.0.7-cp36-cp36m-macosx_10_9_x86_64.whl (2.6 MB) File type Wheel Python version cp36 Upload date Hashes View

Supported by

AWS AWS Cloud computing Datadog Datadog Monitoring Facebook / Instagram Facebook / Instagram PSF Sponsor Fastly Fastly CDN Google Google Object Storage and Download Analytics Huawei Huawei PSF Sponsor Microsoft Microsoft PSF Sponsor NVIDIA NVIDIA PSF Sponsor Pingdom Pingdom Monitoring Salesforce Salesforce PSF Sponsor Sentry Sentry Error logging StatusPage StatusPage Status page