Crafting a promela front-end with abstract data types to mitigate the sensitivity of (compositional) analysis to implementation choices

  • Authors:

  • Yung-Pin Cheng

  • Affiliations:

  • Department of Information and Computer Education, National Taiwan Normal University, Taipei, Taiwan

  • Venue:
  • SPIN'05 Proceedings of the 12th international conference on Model Checking Software
  • Year:
  • 2005

Quantified Score

Hi-index 0.00

Visualization

Abstract

Recently, an active research topic in software verification is applying model checkers to programs, such as multi-threaded Java code. However, a program typically consists of more behaviors, such as operations on complicated data structures or implementation details which are typically made for some criteria like performance. A brute-force model extraction may produce a poor model for analysis engine. In this paper, we give examples to show how subtle changes in implementation may result in considerable differences in analysis, particularly to compositional analysis. Unfortunately, these implementation choices are made by programmers – people who typically do not possess the knowledge of verification. To mitigate such sensitivity, we advocate that verification tools should recognize and support abstract data types and, in the meantime, prohibit or suppress the use of array. Programming process behaviors with abstract data types can hide and converge the implementation choices. More importantly, abstract data types are informative. They provide essential information for tool automation to select a best implementation for analysis. In this paper, we describe the design and implementation of such a prototype tool which can parse systems written in Promela syntax.