Rex: Closing the language-verifier gap with safe and usable kernel extensions
summary
Rex claims to close the language‑verifier gap by running kernel extensions in safe Rust linked to a Rex kernel crate and a lightweight runtime that enforces stack safety, termination, and safe exception handling at load or runtime instead of relying on the in‑kernel verifier. Implementation targets Linux 6.15 with a rustc backend pass and kernel symbol fixups. A BMC case study shows 300 Rust lines vs 500 C, classroom assignments succeed, and performance is competitive: 5.4× over user‑space cache on 8 CPUs and slightly (<5%) faster than eBPF. no separate static verification
AI-written summary · about
chapters (7)
- 0:00Rex and kernel extensions
- 2:47Language-verifier gap problem
- 6:40Rex framework solution
- 8:05Extended tab safety in Rex
- 11:40Exception handling in Rex runtime
- 15:09Usability and performance of Rex
- 18:23Conclusion and future
related talks
10mKeynote: Rust in the Linux Kernel: A New Era for Cloud Native Performance and Security
46mRust for Linux
14mKeynote: Rust in the Linux Kernel, Why?
30mAnatomy of a run-time: your contract with the kernel
40mKeynote: Kernel Developer Roundtable
48mWriting Linux Kernel Modules in Safe Rust
51mRust in the Linux kernel
42mThe Kernel Report
