Practical systems for overcoming processor imperfections

Processors are not perfect. Even the most modern, thoroughly verified processors contain imperfections. Processor imperfections, being embedded in the lowest layer of the system, pose a significant problem not only for software developers during design and debug, but also serve as weaknesses to the security mechanisms implemented in the upper layers of the system. With such a pervasive impact on computing systems, it is vital that processor vendors address these imperfections in a way that maintains the abstraction of a perfect processor promised to software developers.

In this talk, I will present my work on designing systems that help reinforce the abstraction of a perfect processor in the face of a range of processor imperfections. Specifically, we build three reference systems targeted at different design-time processor imperfections: malicious circuits inserted into the processor description, errata-like bugs, and a subset of errata-like bugs that threaten the security abstractions of the processor.
Matthew Hicks is a Postdoctoral Research Associate in the Department of Computer Science at the University of Illinois Urbana-Champaign. His research areas include embedded systems, low-level security, and architecture. Matthew earned his PhD in Computer Science (advised by Sam King) from the University of Illinois Urbana-Champaign in 2013, where he also earned his MS in 2008. Matthew earned his BS in Computer Science from the University of Central Florida in 2006.