Skip to content

Conversation

@kevin-valerio
Copy link
Contributor

Description

Adds a "git recency" testcase scoring/scheduler bias: still coverage-guided, but prefers inputs that execute recently changed lines (from git blame).

  • libafl: GitRecency* metadata + GitRecencyTestcaseScore, and GitAwareStdWeightedScheduler
  • libafl_cc (feature git-recency): LLVM pass + link-time mapping generation (LIBAFL_GIT_RECENCY_MAPPING_PATH) and libafl_git_recency_mapgen tool
  • Docs: mdBook page under Advanced Features

Checklist

  • I have run ./scripts/precommit.sh and addressed all comments

Copilot AI review requested due to automatic review settings January 21, 2026 10:51

This comment was marked as spam.

@kevin-valerio
Copy link
Contributor Author

^ sorry for the verbose Copilot comments, forgot I had that stuff activated...

@rmalmain
Copy link
Member

it seems we also have some options enabled about copilot auto review on the repo, i'm not really sure why

@kevin-valerio
Copy link
Contributor Author

it seems we also have some options enabled about copilot auto review on the repo, i'm not really sure why

Well at the end the comments left aren't that stupid, but it's still very verbose

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants