NDSS 2025 – Statically Discover Cross-Entry Use-After-Free Vulnerabilities In The Linux Kernel

The session discusses system-level security, focusing on a research paper about Use-After-Free (UAF) vulnerabilities. UAF flaws are common memory safety issues that are difficult to detect effectively. Existing methods are either dynamic, providing high precision but limited code coverage, or static, struggling with complex UAF scenarios. This paper introduces UAFX, a static analyzer designed to discover cross-entry UAF vulnerabilities in the Linux kernel. UAFX uses escape-fetch-based cross-entry alias analysis to accurately track memory relationships. A validation framework based on partial-order constraints helps eliminate false positives. Evaluations reveal that UAFX can find new UAF vulnerabilities with reasonable precision and performance. The tool has identified 80 true positive warnings in the kernel and a user-space program. This research aims to provide a more reliable method for identifying complex UAF errors. The presented content originates from the NDSS Symposium 2025 Conference, published on their YouTube channel.