surrealpatch/lib/fuzz

Fuzzing

Surrealdb maintains a set of fuzz testing harnesses that are managed by cargo-fuzz.

To build and run the fuzzer we will need to;

• Install a specific version of the nightly compiler
• Install cargo fuzz
• Build a fuzz friendly version of surrealdb with our harnesses

Installing nightly

One of the key requirements for high-performance fuzzing is the ability to collect code-coverage feedback at runtime. With the current stable version of rustc we can't instrument our fuzz-harnesses with coverage feedback. Because of this we need to use some of the more bleeding edge features available in the nightly release.

Unfortunately for us the nightly release is a little unstable and there was a bug in the latest version of the nightly compiler that prevents use from compiling some of surrealdb's depdendencies. To workaround this issue we've carefully picked a version of the nightly compiler that works with both cargo-fuzz and our dependencies. This version is nightly-2023-04-21. To install this version we simply need to run;

rustup install nightly-2023-04-21

Installing cargo-fuzz

Full details on the different install options are available, in the cargo-fuzz book. but for the sake of brevity you can just install the basics with the command below.

cargo +nightly-2023-04-21 install cargo-fuzz

Building the fuzzers

Now that we've install cargo-fuzz we can go ahead and build our fuzzers.

cd lib  
# -O: Optimised build
# --debug-assertions: Catch common bugs, e.g. integer overflow.
cargo +nightly-2023-04-21 fuzz build -O --debug-assertions

Running the fuzzer

Now that the fuzzer has successfully built we can actually run them. To list the available fuzz harnesses we can use the command.

cargo +nightly-2023-04-21 fuzz list

Once we know what fuzzer (in this case fuzz_executor) we want to run we can it using the command;

cargo +nightly-2023-04-21 fuzz run -O --debug-assertions fuzz_executor

The previous command will run the fuzzer in libfuzzer's default mode, which means as a single thread. If you would like to speed fuzzing up we can make use of all cores, and use a dictionary file. e.g.

# -fork: Run N seperate process fuzzing in parralell in this case we
#        use nproc to match the number of processors on our local
#        machine.
# -dict: Make use the fuzzer specific dictionary file.
cargo +nightly-2023-04-21 fuzz run -O --debug-assertions \
  fuzz_executor -- -fork=$(nproc) \
  -dict=fuzz/fuzz_targets/fuzz_executor.dict