Inferring data polymorphism in systems code

  • Authors:
  • Brian Hackett;Alex Aiken

  • Affiliations:
  • Stanford University, Stanford, CA, USA;Stanford University, Stanford, CA, USA

  • Venue:
  • Proceedings of the 19th ACM SIGSOFT symposium and the 13th European conference on Foundations of software engineering
  • Year:
  • 2011

Quantified Score

Hi-index 0.00

Visualization

Abstract

We describe techniques for analyzing data polymorphism in C, and show that understanding data polymorphism is important for statically verifying type casts in the Linux kernel, where our techniques prove the safety of 75% of downcasts to structure types, out of a population of 28767. We also discuss prevalent patterns of data polymorphism in Linux, including code patterns we can handle and those we cannot.