KernelGPT: Enhanced Kernel Fuzzing via Large Language Models

Abstract

Bugs in operating system kernels can affect billions of devices and users all over the world. As a result, a large body of research has been focused on kernel fuzzing, i.e., automatically generating syscall (system call) sequences to detect potential kernel bugs or vulnerabilities. Kernel fuzzing aims to generate valid syscall sequences guided by syscall specifications that define both the syntax and semantics of syscalls. While there has been existing work trying to automate syscall specification generation, this remains largely manual work, and a large number of important syscalls are still uncovered.