What is it about?
With QFuzz we present a side-channel quantification technique based on greybox fuzzing. We propose two partitioning algorithms to characterize side-channel observations: one based on dynamic programming and one based on greedy selection.
Featured Image
Photo by Glenn Carstens-Peters on Unsplash
Why is it important?
Proper detection of side-channel vulnerabilities requires their quantification so that the exploitability and severity can be investigated. Our evaluation of QFuzz shows its scalability and usefulness when compared to state-of-the-art techniques as well as in real-world applications, including the discovery of a previously unknown vulnerability in a security-critical library.
Perspectives
With QFuzz we provide a new perspective on the quantification of side-channel vulnerabilities as we propose a dynamic analysis technique. It can be easily applied and does not require lots of preparation. We hope it can inspire practitioners to apply such techniques in practice to make software more secure and reliable.
Yannic Noller
National University of Singapore
Read the Original
This page is a summary of: QFuzz: quantitative fuzzing for side channels, July 2021, ACM (Association for Computing Machinery),
DOI: 10.1145/3460319.3464817.
You can read the full text:
Resources
QFuzz at ISSTA'2021
15min video presentation of QFuzz at the ISSTA'2021
Replication Package
First official release for QFuzz. We added all parts of our tool and all evaluation subjects to support the reproduction of our results. This release is submitted to the ISSTA 2021 Artifact Evaluation.
QFuzz Presentation Slides
Our slides from ISSTA'2021.
QFuzz Pre-print
Our pre-print for QFuzz.
Contributors
The following have contributed to this page
