Dynamic information-flow analysis for multi-threaded applications

  • Authors:
  • Laurent Mounier;Emmanuel Sifakis

  • Affiliations:
  • VERIMAG Laboratory, University of Grenoble, Gieres, France;VERIMAG Laboratory, University of Grenoble, Gieres, France

  • Venue:
  • ISoLA'12 Proceedings of the 5th international conference on Leveraging Applications of Formal Methods, Verification and Validation: technologies for mastering change - Volume Part I
  • Year:
  • 2012

Quantified Score

Hi-index 0.00

Visualization

Abstract

Information-flow analysis is one of the promising techniques to leverage the detection of software vulnerabilities and confidentiality breaches. However, in the context of multi-threaded applications running on multicore platforms, this analysis becomes highly challenging due to data races and inter-processor dependences. In this paper we first review some of the existing information-flow analysis techniques and we discuss their limits in this particular context. Then, we propose a dedicated runtime predictive approach. It consists in extending information-flow properties computed from a single parallel execution trace to a set of valid serialisations with respect to the execution platform. This approach can be applied for instance in runtime monitoring or security testing of multi-threaded applications.