一种新的LTL性质的无状态模型检验方法
A New Approach to Stateless Model Checking of LTL Properties
摘要:
大型复杂并发程序的验证是软件界的一个重要课题。无状态模型检验是一种适合于大型程序系统化、自动化测试的方法,它在大型程序代码验证中的作用已得到证实。这方面的另一个众所周知的方法是运行时验证。无状态模型检验和运行时验证在某些方面是相似的。运行时验证中的一种常见方法是为线性时态逻辑中表示的属性构建运行时监视器。目前,在运行时验证领域提出了一些对有限路径上的线性时态逻辑公式进行检验的语义,这些语义也可以应用于无状态模型检验。然而,现有的无状态模型检查器不支持LTL公式。在某些设置中,使用无状态模型检验而不是运行时验证更有利。本文提出了一种新的编码的最近LTL语义有限路径到一个演员为基础的系统。我们采取了一个真正的并行方法,而不保存任何程序的状态或痕迹,这不仅解决了重要的问题,在运行时验证,但也可以应用于无状态的模型检验。
英文摘要:
Verification of large and complicated concurrent programs is an important issue in the software world. Stateless model checking is an appropriate method for systematically and automatically testing of large programs, which has proved its power in verifying code of large programs. Another well-known method in this area is runtime verification. Both stateless model checking and runtime verification are similar in some ways. One common approach in runtime verification is to construct runtime monitors for properties expressed in linear temporal logic. Currently, there are some semantics to check linear temporal logic formulae on finite paths proposed in the field of runtime verification, which can also be applied to stateless model checking. However, existing stateless model checkers do not support LTL formulae. In some settings, it is more advantageous to make use of stateless model checking instead of runtime verification. This paper proposes a novel encoding of one of the recent LTL semantics on finite paths into an actor-based system. We take a truly parallel approach without saving any program states or traces, which not only addresses important problems in runtime verification, but can also be applied to stateless model checking.
机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。