μSwitch: Fast Kernel Context Isolation with Implicit Context Switches
Dinglan Peng, Congyu Liu, Tapti Palit, Pedro Fonseca, Anjo Vahldiek-Oberwagner, Mona Vij
Abstract
Isolating application components is crucial to limit the exposure of sensitive data and code to vulnerabilities in the untrusted components. Process-based isolation is the de facto isolation used in practice, e.g., web browsers. However, it incurs significant performance overhead and is typically infeasible when frequent switches between isolation domains are expected. To address this problem, many intra-process memory isolation techniques have been proposed using novel kernel abstractions, recent CPU extensions (e.g., Intel ® MPK), and software-based fault isolation (e.g., WebAssembly). However, these techniques insufficiently isolate kernel resources, such as file descriptors, or do so by incurring high overheads when resources are accessed. Other work virtualizes the kernel context inside a privileged user space domain, but this is ad-hoc, error-prone, and provides only limited kernel functionalities.
We propose μSWITCH, an efficient kernel context isolation mechanism with memory protection that addresses these limitations. We use a protected structure, shared by the kernel and the user space, for context switching and propose implicit context switching to improve its performance by deferring the kernel resource switch to the next system call. We apply μSWITCH to isolate libraries in the Firefox web browser and an HTTP server, and reduce the overhead of isolation by 32.7% to 98.4% compared with other isolation techniques.
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.
Cited by top-tier papers5
- Endokernel: A Thread Safe Monitor for Lightweight Subprocess IsolationFangfei Yang, Bumjin Im, Weijie Huang, Kelly Kaoudis et al.USENIX Security 2024 · 8 citations
- Capacity: Cryptographically-Enforced In-Process Capabilities for Modern ARM ArchitecturesKha Dinh Duy, Kyuwon Cho, Taehyun Noh, Hojoon LeeCCS 2023 · 5 citations
- LatticeBox: A Hardware-Software Co-Designed Framework for Scalable and Low-Latency CompartmentalizationZhanpeng Liu, Chenyang Li, Wende Tan, Yuan Li et al.NDSS 2026 · 1 citation
- SoK: Software CompartmentalizationHugo Lefeuvre, Nathan Dautenhahn, David Chisnall, Pierre OlivierS&P 2025
- Secpoline: A Scalable Approach to Build Secure In-Process Syscall InterposersRuben Sturm, Anton Schelfhout, Merve Gülmez, Adriaan Jacobs et al.USENIX Security 2026
Builds on29
- ERIM: Secure, Efficient In-process Isolation with Protection Keys (MPK)Anjo Vahldiek-Oberwagner, Eslam Elnikety, Nuno O. Duarte, Michael Sammler et al.USENIX Security 2019 · 247 citations
- Faastlane: Accelerating Function-as-a-Service WorkflowsSwaroop Kotni, Ajay Nayak, Vinod Ganapathy, Arkaprava BasuUSENIX ATC 2021 · 142 citations
- Shreds: Fine-Grained Execution Units with Private MemoryYaohui Chen, Sebassujeen Reymondjohnson, Zhichuang Sun, Long LuS&P 2016 · 116 citations
- Site Isolation: Process Separation for Web Sites within the BrowserCharles Reis, Alexander Moshchuk, Nasko OskovUSENIX Security 2019 · 105 citations
- xMP: Selective Memory Protection for Kernel and User SpaceSergej Proskurin, Marius Momeu, Seyedhamed Ghavamnia, Vasileios P. Kemerlis et al.S&P 2020 · 89 citations
Related papers
- EPK: Scalable and Efficient Memory Protection KeysJinyu Gu, Hao Li, Wentai Li, Yubin Xia et al.USENIX ATC 2022
- PKU Pitfalls: Attacks on PKU-based Memory Isolation SystemsR. Joseph Connor, Tyler McDaniel, Jared M. Smith, Max SchuchardUSENIX Security 2020
- VDom: Fast and Unlimited Virtual Domains on Multiple ArchitecturesZiqi Yuan, Siyu Hong, Rui Chang, Yajin Zhou et al.ASPLOS 2023 · 19 citations
- Enforcing Least Privilege Memory Views for Multithreaded ApplicationsTerry Ching-Hsiang Hsu, Kevin J. Hoffman, Patrick Eugster, Mathias PayerCCS 2016 · 71 citations
- Limitations and Opportunities of Modern Hardware Isolation MechanismsXiangdong Chen, Zhaofeng Li, Tirth Jain, Vikram Narayanan et al.USENIX ATC 2024 · 7 citations
