A method and system for detecting context-sensitive memory vulnerabilities in Rust programs
By performing context-sensitive alias analysis and data flow analysis at the MIR layer of the Rust compiler, the problems of false negatives and false negatives in memory error detection in Rust programs are solved, achieving high-precision memory error detection and supporting memory safety detection in the Rust development process.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- NANJING UNIV
- Filing Date
- 2026-05-05
- Publication Date
- 2026-06-30
AI Technical Summary
Existing Rust program memory error detection tools suffer from high false negative rates and high false positive rates, are unable to track memory errors across functions, and rely on runtime detection and context-insensitive alias analysis, making it difficult to detect complex memory safety issues.
By implementing context-sensitive alias analysis and data flow analysis in the MIR layer of the Rust compiler, and utilizing the precise control flow and data flow information of the MIR, static detection of memory errors such as use-after-free and double free is performed. Combined with the working list algorithm and abstract state domain, accurate memory error detection is achieved.
It significantly reduces the false negative and false positive rates, improves the accuracy of memory error detection, can detect complex cross-function memory errors, generates structured diagnostic reports, and supports memory safety testing in the Rust development process.
Smart Images
Figure CN122309332A_ABST