Cooper Knows the Shortest Stave: Finding 134 Bugs in the Binding Code of Scripting Languages
summary
Presents cooperative mutation: a technique that mutates both scripting-language code and the program’s native inputs to expose bugs in binding code. Implements three concrete steps for popular scripting languages—clustering native objects into semantic groups to shrink the native-input search space, statistically inferring relationships between script operations and those object groups from many executions, and using those inferred links to pick objects and script edits for targeted two-dimensional mutations.
AI-written summary · about
chapters (10)
- 0:00Start
- 4:26Motivating Example
- 6:49Challenges
- 8:15Cooper: Overview
- 14:37Relationship Inference
- 16:11Relationship guided mutation
- 17:20Implementation
- 23:09Case Study #1
- 25:36Case Study #3 Demo
- 28:38Takeaways
related talks
34mHow To Find And Prevent Entire Classes Of Security Vulnerabilities
50mCracking the Code to Secure Software
52mStructure aware fuzzing
47mKeynote: Code-Reuse Attacks and Beyond
29mOne-Click Code Fix: Securing Code Using AI
24mSymbiotic: finding bugs in C programs
17mDiscussion - Protocols, Session 2
51mPreventing Security Bugs through Software Design