Fuzz testing is based on generating a large amount of random inputs for a software component in order to trigger bugs and unexpected behavior. Basic fuzzing uses randomly generated arrays of bytes as inputs, which works great for some applications but is problematic for testing the logic that operates on highly structured data, as most random inputs are immediately rejected as invalid before any interesting parts of the code get a chance to run.
Structured fuzzing addresses this issue by ensuring the randomly generated data is itself structured, and as such has a high chance of presenting a valid input.
explorer_fuzzer is a structured fuzzer based on
libprotobuf-mutator, which is a
library to randomly mutate
protobuffers.
The input to the fuzzer is an instance of Carbon::Fuzzing::Carbon proto
randomly generated by the libprotobuf-mutator framework. explorer_fuzzer
converts the proto to a Carbon source code string, and tries to parse and
execute the code using explorer implementation.
libprotobuf-mutator supports fuzzer inputs in either text or binary protocol
buffer format. explorer_fuzzer uses text proto format with Carbon proto
message definition in common/fuzzing/carbon.proto.
Fuzzer AST representation in carbon.proto needs to be updated when changes are made to the AST, like adding a new AST node classes or changing relevant data members of existing nodes.
There are two unit tests which normally should not require direct changes, as both tests work off of Carbon test files in testdata.
ast_to_proto_test.cpp is a 'smoke' test which verifies that each field of Carbon proto is populated at least once after converting all of test Carbon files and merging the results into a single protocol buffer.
proto_to_carbon_test.cpp
uses a 'roundtrip' approach, by converting each parseable Carbon file to a
proto representation, then back to Carbon source, parsing this source into a
second instance of an AST, and comparing the second AST with the original
AST using AST::Dump() method. The goal of the test is to ensure that
carbon.proto is able to represent ASTs correctly without information loss.
To incorporate AST changes into fuzzing logic:
Add appropriate AST information to carbon.proto. Use existing similar cases as examples.
Add logic to populate the proto to ast_to_proto.cpp.
Make sure ast_to_proto_test passes with the new changes.
Modify proto_to_carbon.cpp which handles printing of a Carbon proto instance as a Carbon source string. For example, add code to print newly introduced proto fields.
Make sure proto_to_carbon_test passes after the changes.
The fuzzer can be run in 'unit test' mode, where the fuzzer executes on each
input file from the fuzzer_corpus/ folder, or in 'fuzzing' mode, where the
fuzzer will keep generating random inputs and executing the logic on them until
a crash is triggered, or forever in a bug-free program ;).
To run in 'unit test' mode:
bazel test --config=proto-fuzzer --test_output=all //explorer/fuzzing:explorer_fuzzer
To run in 'fuzzing' mode:
bazel build --config=proto-fuzzer //explorer/fuzzing:explorer_fuzzer
bazel-bin/explorer/fuzzing/explorer_fuzzer
It's also possible to run the fuzzer on a single input:
bazel-bin/explorer/fuzzing/explorer_fuzzer /tmp/crash.textproto
To reproduce a crash, run the fuzzer on the crashing input as described above.
A separate tool called fuzzverter can be used for things like converting a
crashing input to Carbon source code for running explorer on the code
directly.
To convert a Fuzzing::Carbon text proto to Carbon source:
bazel-bin/explorer/fuzzing/fuzzverter --mode proto_to_carbon --input /tmp/crash.textproto
The ability of the fuzzing framework to generate 'interesting' inputs can be
improved by providing 'seed' inputs known as the fuzzer corpus. The inputs need
to be a Fuzzing::Carbon text proto.
To generate a text proto from Carbon source:
bazel-bin/explorer/fuzzing/fuzzverter --mode carbon_to_proto --input /tmp/crash.carbon --output /tmp/crash.textproto