DOVE: pinpointing firmware security vulnerabilities via symbolic control flow assertion mining (work-in-progress)

Abstract

In the past decade, the number of reported security attacks exploiting unchecked input firmware values has been on the rise. To address this concerning trend, this work proposes a novel detection framework, called DOVE, capable of identifying unlikely firmware execution flows, specifically those that may reveal a security vulnerability. The DOVE framework operates by leveraging a symbolic simulation of the firmware's execution, paired with a probability computation that can identify unlikely execution flows and provide to the user corresponding formal assertions.