bcore-mutation
A mutation testing tool for Bitcoin Core
"Mutation testing involves modifying a program in small ways. Each mutated version is called a mutant, and tests detect and reject mutants by causing the behaviour of the original version to differ from the mutant. This is called killing the mutant. Test suites are measured by the percentage of mutants that they kill." — Wikipedia
Features
- Generate mutants only for code touched in a specific PR or branch
- Security-based mutation operators for testing fuzzing scenarios
- Skip useless mutants (comments,
LogPrintfstatements, etc.) - One-mutant-per-line mode for faster analysis
- Coverage-guided mutation (only mutate covered lines)
- AST-based arid node filtering to reduce noise
- Persist results and resume analysis with a SQLite database
Installation
From source
From crates.io
Workflow
- Mutate — generate mutants and store them in a SQLite database.
- Analyze — run your test command against each mutant and report survivors.
mutate command
Generates mutants for the target code and optionally persists them to a SQLite database.
Flags
| Flag | Short | Default | Description |
|---|---|---|---|
--project NAME |
bitcoin-core |
Project to mutate. Accepts bitcoin-core or secp256k1. When --pr is used, the PR is fetched from this project's repository. |
|
--sqlite [PATH] |
mutation.db |
Persist mutants to a SQLite database. Accepts an optional custom path. | |
--file PATH |
-f |
File to mutate. Mutually exclusive with --pr. |
|
--pr NUMBER |
-p |
0 (current branch) |
PR number to mutate (fetched from the --project repository). Mutually exclusive with --file. |
--range START END |
-r |
Restrict mutation to a line range within the target file. Cannot be combined with --cov. |
|
--cov PATH |
-c |
Path to a coverage file (*.info generated with cmake -P build/Coverage.cmake). Only lines covered by tests will be mutated. Cannot be combined with --range. |
|
--skip-lines PATH |
Path to a JSON file listing lines to skip per file (see format below). | ||
--one-mutant |
Create only one mutant per line (prioritises harder-to-kill operators). Useful for large files. | ||
--test-only |
-t |
Only create mutants inside unit and functional test files. | |
--only-security-mutations |
-s |
Apply only security-focused mutation operators. Useful when evaluating fuzzing coverage. | |
--disable-ast-filtering |
Disable AST-based arid node detection. Generates more mutants, including potentially redundant ones. | ||
--add-expert-rule PATTERN |
Add a custom pattern for arid node detection (see AST filtering below). |
Examples
Mutate a specific file:
Mutate all files changed in a PR:
Mutate a secp256k1 PR (fetched from bitcoin-core/secp256k1):
Restrict to a line range:
Use a coverage file (only mutate covered lines):
One mutant per line (faster analysis):
Mutate only test files:
Security-only mutations (for fuzzing):
Skip specific lines:
Skip lines file format
Create a JSON file that maps file paths to line numbers to skip:
analyze command
Applies each mutant to the source tree, runs the test command, and reports whether the mutant was killed or survived.
When --sqlite is used, the mutate command prints a run_id that you pass to analyze with --run-id.
Flags
| Flag | Short | Default | Description |
|---|---|---|---|
--project NAME |
bitcoin-core |
Project being analyzed. Accepts bitcoin-core or secp256k1. |
|
--sqlite [PATH] |
mutation.db |
SQLite database to read mutants from. Requires --run-id. Accepts an optional custom path. |
|
--run-id ID |
Run ID returned by the mutate command. Requires --sqlite. |
||
--command CMD |
-c |
Shell command used to test each mutant (e.g. a build + test invocation). Required when using --run-id. |
|
--file-path PATH |
Only analyze mutants that belong to this file. Requires --run-id. |
||
--folder PATH |
-f |
Folder containing mutants (alternative to --sqlite / --run-id). |
|
--timeout SECONDS |
-t |
300 |
Timeout in seconds for each mutant's test run. |
--jobs N |
-j |
0 |
Number of parallel jobs passed to the compiler (e.g. make -j N). 0 uses the system default. |
--survival-threshold RATE |
0.75 |
Maximum acceptable mutant survival rate (e.g. 0.3 = 30%). The run exits with an error if the threshold is exceeded. |
|
--min-score RATE |
CI gate: fail with a non-zero exit code if the final mutation score (killed / total) is below this value (e.g. 0.8 = 80%). Aggregated across all analyzed folders. When unset, the score is not enforced. |
||
--surviving |
Only analyze mutants that survived a previous run. Requires --run-id. |
Examples
Basic analysis:
Per-file commands (useful when a PR spans multiple modules):
Retry only survivors from a previous run:
Fail CI when the mutation score drops below 80% (folder mode, no database):
# 1. Generate mutants for the PR — writes muts-* folders to disk
# 2. Analyze them and fail the job if the score is under 80%.
# With no --command, the built-in secp256k1 build/test commands are used.
Set a custom timeout and job count:
Set a survival rate threshold:
Testing
Contributing
- Fork the repository.
- Create a feature branch.
- Add tests for your changes.
- Ensure all tests pass:
cargo test - Run the linter:
cargo clippy -- -D warnings - Format code:
cargo fmt - Submit a pull request.
License
MIT License