SyzSpec: Specification Generation for Linux Kernel Fuzzing via Under-Constrained Symbolic Execution
Yu Hao, Juefei Pu, Xingyu Li, Zhiyun Qian, Ardalan Amiri Sani
摘要
Fuzzing has become one of the most effective and widely used techniques for discovering bugs and vulnerabilities, particularly in large-scale and complex programs like operating system kernels. A notable example is the kernel fuzzer syzkaller, which has identified over 6,800 bugs in the Linux kernel, with more than 5,500 already fixed. A crucial reason behind the success of the syzkaller is its collection of syscall descriptions, which are typically provided by human experts. Although some methods exist for automatically generating these syscall descriptions for device drivers, they often fall short when dealing with complex user inputs. These existing methods either lack precision or have a limited analysis scope, resulting in incomplete syscall descriptions.
问问这篇 Paper
智能体会读完全文。
Lune 把这篇 Paper 索引到了最后一个公式,引用它的顶会 Paper 也一样。你提问,回答直接引用原文。
引用它的顶会 Paper3
- Cottontail: Large Language Model-Driven Concolic Execution for Highly Structured Test Input GenerationHaoxin Tu, Seongmin Lee, Yuxian Li, Peng Chen 等S&P 2026 · 被引用 22 次
- A Compilation-Based Under-Constrained Execution EngineMingjun Yin, Zhaorui Li, Ju Chen, Haochen Zeng 等OSDI 2026
- HyperMirage: Direct State Manipulation in Hybrid Virtual CPU FuzzingManuel Andreas, Fabian Specht, Marius MomeuNDSS 2026
它引用的顶会 Paper19
- SOK: (State of) The Art of War: Offensive Techniques in Binary AnalysisYan Shoshitaishvili, Ruoyu Wang, Christopher Salls, Nick Stephens 等S&P 2016 · 被引用 1,085 次
- DIFUZE: Interface Aware Fuzzing for Kernel DriversJake Corina, Aravind Machiry, Christopher Salls, Yan Shoshitaishvili 等CCS 2017 · 被引用 195 次
- Where Does It Go?: Refining Indirect-Call Targets with Multi-Layer Type AnalysisKangjie Lu, Hong HuCCS 2019 · 被引用 142 次
- Charm: Facilitating Dynamic Analysis of Device Drivers of Mobile SystemsSeyed Mohammadjavad Seyed Talebi, Hamid Tavakoli, Hang Zhang, Zheng Zhang 等USENIX Security 2018 · 被引用 81 次
- HEALER: Relation Learning Guided Kernel FuzzingHao Sun, Yuheng Shen, Cong Wang, Jianzhong Liu 等SOSP 2021 · 被引用 59 次
相关 Paper
- SyzGen++: Dependency Inference for Augmenting Kernel Driver FuzzingWeiteng Chen, Yu Hao, Zheng Zhang, Xiaochen Zou 等S&P 2024 · 被引用 12 次
- SyzDescribe: Principled, Automated, Static Generation of Syscall Descriptions for Kernel DriversYu Hao, Guoren Li, Xiaochen Zou, Weiteng Chen 等S&P 2023
- KSG: Augmenting Kernel Fuzzing with System Call Specification GenerationHao Sun, Yuheng Shen, Jianzhong Liu, Yiru Xu 等USENIX ATC 2022 · 被引用 45 次
- KernelGPT: Enhanced Kernel Fuzzing via Large Language ModelsChenyuan Yang, Zijie Zhao, Lingming ZhangASPLOS 2025 · 被引用 45 次
- SyzGen: Automated Generation of Syscall Specification of Closed-Source macOS DriversWeiteng Chen, Yu Wang, Zheng Zhang, Zhiyun QianCCS 2021 · 被引用 25 次
