Moneta: Ex-Vivo GPU Driver Fuzzing by Recalling In-Vivo Execution States

Joonkyo Jung

Network and Distributed System Security (NDSS) Symposium 2025 · Day 2 · Fuzzing 1

The talk "Moneta: Ex-Vivo GPU Driver Fuzzing by Recalling In-Vivo Execution States," presented by Joonkyo Jung of Jon University, introduces a novel fuzzing framework designed to uncover vulnerabilities in Graphics Processing Unit (GPU) drivers. Moneta tackles long-standing challenges in device driver fuzzing, particularly the complex inter-syscall and hardware dependencies inherent to GPU interactions. The core innovation lies in its hybrid approach, combining the benefits of snapshotting with deterministic record-and-replay mechanisms to efficiently explore deep states within GPU drivers.

AI review

Moneta is legitimate systems security research with a clear technical contribution: a hybrid snapshot + record-replay ex-vivo fuzzer that solves real, concrete problems in GPU driver fuzzing. Five CVEs across Nvidia, AMD, and Mali, plus an open-source release, is the receipts. PhD student presenting novel doctoral work — this is exactly what a research track should look like.

Watch on YouTube