googleprojectzero / googleprojectzero/fuzzilli

Implement More Intelligent Mutation/Code Generator/Constant Selection Scheduler

Open
#172 7 comments 4 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

Dominant language
Swift
Stars
2.3k
Forks
367
Avg merge
23h 53m
Merged PRs (30d)
1

Description

Fuzzilli current uses hard-coded weights to select between the various mutators, code generators, and constants. These weights are just approximations, and a more intelligent mutator scheduler could help improve performance, by giving additional weight to those likely to find new coverage, or be successful.

One possible example would be something like MOpt, although it may be overkill at this time.

First thoughts on requirements for a new scheduler:

  • Applies more emphasis to mutators/code generators/constant selectors that are more likely to run successfully or produce new coverage
  • Have a "weight floor", intermittent re-normalization, etc, such that some no mutator/code generator/constant is ever completely ignored

Contributor guide

Open the contributing guide

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Research direction

Start by reading the scheduler setup in Sources/FuzzilliCli/main.swift, Sources/FuzzilliCli/CodeGeneratorWeights.swift, and Sources/Fuzzilli/Core/ProgramBuilder.swift, then review the linked MOpt paper. Done means a scheduler gives more weight to successful or coverage-producing choices while retaining a weight floor and periodically re-normalizing all mutators, generators, and constants.

Written by the indexing model from the issue text.

Assessment

Tech stack
swift
Domain
devtools, testing-qa
Issue type
Feature
Difficulty
5/5
Estimated time
Over a week
Activity status
Stale
Clarity
Needs clarification
Newbie friendliness
25/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.