Skip to content

Modular, structure-aware, and feedback-driven fuzzing engine for Rust functions

License

Notifications You must be signed in to change notification settings

loiclec/fuzzcheck-rs

Repository files navigation

Fuzzcheck

CI Docs MIT licensed crates.io

Fuzzcheck is a modular, structure-aware, and feedback-driven fuzzing engine for Rust functions.

Given a function test: (T) -> bool, you can use fuzzcheck to find a value of type T that fails the test or leads to a crash.

The tool fuzzcheck-view is available to visualise the code coverage of each/all test cases generated by fuzzcheck. It is still just a prototype though.

Follow the guide at fuzzcheck.neocities.org to get started or read the documentation on docs.rs.

Setup

Linux or macOS is required. Windows support is planned but I need help with it.

Rust nightly is also required. You can install it with:

rustup toolchain install nightly

While it is not strictly necessary, installing the cargo-fuzzcheck executable will make it easier to run fuzzcheck.

cargo install cargo-fuzzcheck

In your Cargo.toml file, add fuzzcheck as a dev-dependency:

[dev-dependencies]
fuzzcheck = "0.12"

Then, we need a way to serialize values. By default, fuzzcheck uses serde_json for that purpose (but it can be changed). That means our data types should implement serde's traits. In Cargo.toml, add:

[dependencies]
serde = { version = "1.0", features = ["derive"] }

Usage

Below is an example of how to use fuzz test. Note:

  1. every code related to fuzzcheck is conditional on #[cfg(test)] because we don't want to carry the fuzzcheck dependency in normal builds
  2. the #![cfg_attr(test, feature(no_coverage))] is required by fuzzcheck’s procedural macros
  3. the use of derive(fuzzcheck::DefaultMutator) makes a custom type fuzzable
#![cfg_attr(fuzzing, feature(no_coverage))]
use serde::{Deserialize, Serialize};

#[cfg_attr(fuzzing, derive(fuzzcheck::DefaultMutator))]
#[derive(Clone, Serialize, Deserialize)]
struct SampleStruct<T, U> {
    x: T,
    y: U,
}

#[cfg_attr(fuzzing, derive(fuzzcheck::DefaultMutator))]
#[derive(Clone, Serialize, Deserialize)]
enum SampleEnum {
    A(u16),
    B,
    C { x: bool, y: bool },
}

fn should_not_crash(xs: &[SampleStruct<u8, SampleEnum>]) {
    if xs.len() > 3
        && xs[0].x == 100
        && matches!(xs[0].y, SampleEnum::C { x: false, y: true })
        && xs[1].x == 55
        && matches!(xs[1].y, SampleEnum::C { x: true, y: false })
        && xs[2].x == 87
        && matches!(xs[2].y, SampleEnum::C { x: false, y: false })
        && xs[3].x == 24
        && matches!(xs[3].y, SampleEnum::C { x: true, y: true })
    {
        panic!()
    }
}

// fuzz tests reside along your other tests and have the #[test] attribute
#[cfg(all(fuzzing, test))]
mod tests {
    #[test]
    fn test_function_shouldn_t_crash() {
        let result = fuzzcheck::fuzz_test(super::should_not_crash) // the test function to fuzz
            .default_mutator() // the mutator to generate values of &[SampleStruct<u8, SampleEnum>]
            .serde_serializer() // save the test cases to the file system using serde
            .default_sensor_and_pool() // gather observations using the default sensor (i.e. recording code coverage)
            .arguments_from_cargo_fuzzcheck() // take arguments from the cargo-fuzzcheck command line tool
            .stop_after_first_test_failure(true) // stop the fuzzer as soon as a test failure is found
            .launch();
        assert!(!result.found_test_failure);
    }
}

We can now use cargo-fuzzcheck to launch the test, using Rust nightly:

rustup override set nightly
# the argument is the *exact* path to the test function
cargo fuzzcheck tests::test_function_shouldn_t_crash

This starts a loop that will stop when a failing test has been found. After about ~50ms of fuzz-testing on my machine, the following line is printed:

Failing test case found. Saving at "fuzz/tests::test_function_shouldn_t_crash/artifacts/59886edc1de2dcc1.json"

The file 59886edc1de2dcc1.json contains the JSON-encoded input that failed the test.

[
  {
    "x": 100,
    "y": {
      "C": {
        "x": false,
        "y": true
      }
    }
  },
  {
    "x": 55,
    "y": {
      "C": {
        "x": true,
        "y": false
      }
    }
  },
  ..
]

Minifying failing test inputs

Fuzzcheck can also be used to minify a large input that fails a test. If the failure is recoverable (i.e. it is not a segfault/stack overflow), and the fuzzer is not instructed to stop after the first failure, then the failing test cases will be minified automatically. Otherwise, you can use the minify command.

Let's say you have a file crash.json containing an input that you would like to minify. Launch cargo fuzzcheck <exact name of fuzz test> with the minify command and an --input-file option.

cargo fuzzcheck "tests::test_function_shouldn_t_crash" --command minify --input-file "crash.json"

This will repeatedly launch the fuzzer in “minify” mode and save the artifacts in the folder artifacts/crash.minified. The name of each artifact will be prefixed with the complexity of its input. For example, crash.minified/800--fe958d4f003bd4f5.json has a complexity of 8.00.

You can stop the minifying fuzzer at any point and look for the least complex input in the crash.minified folder.

Alternatives

Other crates with the same goal are quickcheck and proptest. Fuzzcheck can be more powerful than these because it guides the generation of test cases based on feedback generated from running the test function. This feedback is most often code coverage, but can be different.

Another similar crate is cargo-fuzz, often paired with arbitrary. In this case, fuzzcheck has an advantage by being easier to use, more modular, and being more fundamentally structure-aware and thus potentially more efficient.

Previous work on fuzzing engines

As far as I know, evolutionary, coverage-guided fuzzing engines were popularized by American Fuzzy Lop (AFL).
Fuzzcheck is also evolutionary and coverage-guided.

Later on, LLVM released its own fuzzing engine, libFuzzer, which is based on the same ideas as AFL, but it uses Clang’s SanitizerCoverage and is in-process (it lives in the same process as the program being fuzz-tested.
Fuzzcheck is also in-process. It uses rustc’s -Z instrument-coverage option instead of SanitizerCoverage for code coverage instrumentation.

Both AFL and libFuzzer work by manipulating bitstrings (e.g. 1011101011). However, many programs work on structured data, and mutations at the bitstring level may not map to meaningful mutations at the level of the structured data. This problem can be partially addressed by using a compact binary encoding such as protobuf and providing custom mutation functions to libFuzzer that work on the structured data itself. This is a way to perform “structure-aware fuzzing” (talk, tutorial).

An alternative way to deal with structured data is to use generators just like QuickCheck’s Arbitrary trait. And then to “treat the raw byte buffer input provided by the coverage-guided fuzzer as a sequence of random values and implement a “random” number generator around it.” (cited blog post by @fitzgen). The tool cargo-fuzz has recently implemented that approach.

Fuzzcheck is also structure-aware, but unlike previous attempts at structure-aware fuzzing, it doesn't use an intermediary binary encoding such as protobuf nor does it use Quickcheck-like generators. Instead, it directly mutates the typed values in-process. This is better many ways. First, it is faster because there is no need to encode and decode inputs at each iteration. Second, the complexity of the input is given by a user-defined function, which will be more accurate than counting the bytes of the protobuf encoding. Finally, and most importantly, the mutations are faster and more meaningful than those done on protobuf or Arbitrary’s byte buffer-based RNG. A detail that I particularly like about fuzzcheck, and that is possible only because it mutates typed values, is that every mutation is done in-place and is reversable. That means that generating a new test case is super fast, and can often even be done with zero allocations.

As I was developing Fuzzcheck for Swift, a few researchers developed Fuzzchick for Coq (paper). It is a coverage-guided property-based testing tool implemented as an extension to Quickchick. As far as I know, it is the only other tool with the same philosophy as fuzzcheck. The similarity between the names fuzzcheck and Fuzzchick is a coincidence.

LibAFL is another modular fuzzer written in Rust. It was released relatively recently.

Sponsor this project

Packages

No packages published

Languages