USENIX Security2022Top-tier venue
MundoFuzz: Hypervisor Fuzzing with Statistical Coverage Testing and Grammar Inference
Cheolwoo Myung, Gwangmu Lee, Byoungyoung Lee
Abstract
A hypervisor is system software, managing and running virtual machines. Since the hypervisor is placed at the lowestlevel in the typical systems software stack, it has critical security implications. Once compromised, the entire software components running on top of the hypervisor (including all guest virtual machines and applications running within each guest virtual machine) are compromised as well, as the hypervisor has all the privileges to access those. This paper proposes MUNDOFUZZ, a hypervisor fuzzer to enable both coverage-guided and grammar-aware fuzzing. We find that the coverage measurement in hypervisors suffers from noises due to the hypervisor's asynchronous system event handling. In order to filter out such noises, MUNDOFUZZ develops a statistical differential coverage measurement methods, allowing MUNDOFUZZ to capture the clean coverage information for hypervisor inputs. Moreover, we observe that hypervisor inputs have complex input grammars because it supports many different devices and each device has its own input format. Thus, MUNDOFUZZ learns the input grammar through inspecting the coverage characteristics of the given hypervisor input, which is based on the idea that the hypervisor behaves in a different way if the grammatically correct (or incorrect) input is given. We evaluated MUNDOFUZZ with popular hypervisors, QEMU and Bhyve, and MUNDOFUZZ outperformed other state-of-the-art hypervisor fuzzers ranging from 4.91% to 6.60% in terms of coverage. More importantly, MUNDOFUZZ identified 40 previously unknown bugs (including 9 CVEs), demonstrating its strong practical effectiveness in finding real-world hypervisor vulnerabilities. * Corresponding author which enables a single host machine to run multiple guest VMs. This advantage becomes particularly important as the computing trends are rapidly shifting towards cloud computing. Using the hypervisor, cloud computing platforms can facilitate flexible resource managements on virtualized platforms, which can efficiently service for billions of users [18] . From the security point of view, hypervisors introduce crucial attack surface to be protected. Importantly, a hypervisor is running at the lowest layer in the typical systems software stack, so it is given the highest security privilege of the host machine. Consequently, any successful compromise exploiting hypervisor vulnerabilities would allow attackers to gain privilege of the entire host machine, including the hypervisor itself as well as all guest VMs running on top of them.
Ask about this paper
Your agent reads all of it.
Lune indexed this paper to the last equation, along with the top-tier papers that cite it. Ask a question and the answer quotes them.
Your agent calls
Luneget_paper_fulltext
Free to start. No credit card required.
Terminal
Install the CLIlune papers fulltext 66d78742-de84-41f3-8d7e-dae124d1c747Cited by top-tier papers9
- SoK: Prudent Evaluation Practices for FuzzingMoritz Schloegel, Nils Bars, Nico Schiller, Lukas Bernhard et al.S&P 2024 · 69 citations
- HYPERPILL: Fuzzing for Hypervisor-bugs by leveraging the Hardware Virtualization InterfaceAlexander Bulekov, Qiang Liu, Manuel Egele, Mathias PayerUSENIX Security 2024 · 15 citations
- ORANalyst: Systematic Testing Framework for Open RAN ImplementationsTianchang Yang, Syed Md. Mukit Rashid, Ali Ranjbar, Gang Tan et al.USENIX Security 2024 · 10 citations
- Breaking Isolation: A New Perspective on Hypervisor Exploitation via Cross-Domain AttacksGaoning Pan, Yiming Tao, Qinying Wang, Chunming Wu et al.NDSS 2026
- AidFuzzer: Adaptive Interrupt-Driven Firmware Fuzzing via Run-Time State RecognitionJianqiang Wang, Qinying Wang, Tobias Scharnowski, Li Shi et al.USENIX Security 2025
Builds on8
- Coverage-based Greybox Fuzzing as Markov ChainMarcel Böhme, Van-Thuan Pham, Abhik RoychoudhuryCCS 2016 · 1,026 citations
- Evaluating Fuzz TestingGeorge Klees, Andrew Ruef, Benji Cooper, Shiyi Wei et al.CCS 2018 · 753 citations
- MoonShine: Optimizing OS Fuzzer Seed Selection with Trace DistillationShankara Pailoor, Andrew Aday, Suman JanaUSENIX Security 2018 · 180 citations
- IMF: Inferred Model-based FuzzerHyungSeok Han, Sang Kil ChaCCS 2017 · 139 citations
- DifuzzRTL: Differential Fuzz Testing to Find CPU BugsJaewon Hur, Suhwan Song, Dongup Kwon, Eunjin Baek et al.S&P 2021 · 126 citations
Related papers
- HYPER-CUBE: High-Dimensional Hypervisor FuzzingSergej Schumilo, Cornelius Aschermann, Ali Abbasi, Simon Wörner et al.NDSS 2020
- Morphuzz: Bending (Input) Space to Fuzz Virtual DevicesAlexander Bulekov, Bandan Das, Stefan Hajnoczi, Manuel EgeleUSENIX Security 2022
- Nyx: Greybox Hypervisor Fuzzing using Fast Snapshots and Affine TypesSergej Schumilo, Cornelius Aschermann, Ali Abbasi, Simon Wörner et al.USENIX Security 2021 · 102 citations
- HyperMirage: Direct State Manipulation in Hybrid Virtual CPU FuzzingManuel Andreas, Fabian Specht, Marius MomeuNDSS 2026
- HyperFuzzer: An Efficient Hybrid Fuzzer for Virtual CPUsXinyang Ge, Ben Niu, Robert Brotzman, Yaohui Chen et al.CCS 2021 · 9 citations
