A buffer overflow benchmark for software model checkers

  • Authors:
  • Kelvin Ku;Thomas E. Hart;Marsha Chechik;David Lie

  • Affiliations:
  • University of Toronto, Toronto, ON, Canada;University of Toronto, Toronto, ON, Canada;University of Toronto, Toronto, ON, Canada;University of Toronto, Toronto, ON, Canada

  • Venue:
  • Proceedings of the twenty-second IEEE/ACM international conference on Automated software engineering
  • Year:
  • 2007

Quantified Score

Hi-index 0.00

Visualization

Abstract

Software model checking based on abstraction-refinement has recently achieved widespread success in verifying API conformance in device drivers, and we believe this success can be replicated for the problem of buffer overflow detection. This paper presents a publicly-available benchmark suite to help guide and evaluate this research. The benchmark consists of 298 code fragments of varying complexity capturing 22 buffer overflow vulnerabilities in 12 open source applications. We give a preliminary evaluation of the benchmark using the SatAbs model checker