Practical Compile Time Analyses for Evolving C Systems

在大型和不断发展的软件系统中，编译器、测试器和调试器等工具需要维护和更新静态信息。数据流分析算法收集有关程序中数据的使用和定义的事实，是这些信息的主要来源。增量更新算法只计算受程序更改影响的信息，避免完全重新计算，有效地为大型、不断发展的软件系统提供一致的文档。在以前的研究中，C系统的增量分析的实现一直受到用于确定指针诱导的别名的已知技术的不精确的阻碍。这项研究的目的是在我们的指针混叠近似算法的条件分析方法的基础上，开发更精确的C系统的过程间分析。具体地说，计划：o优化过程间指针别名近似算法，O开发一种新的、针对类C语言的过程间修改副作用分析算法，并设计其他对流敏感的数据流分析(例如，过程间到达定义、静态程序切片)，O为C系统构建第二个原型增量MOD分析器，并分析其在实际C系统上的性能，和o研究扩大大规模编程所涉及的问题。

In large and evolving software systems, tools such as compilers, testers, and debuggers require that static information be maintained and updated. Data flow analysis algorithms, which gather facts about use and definition of data in programs, are a major source of this information. Incremental update algorithms, which only calculate information affected by the program changes, avoiding total recalculation, efficiently provide consistent documentation for a large, evolving software system. In previous research, implementation of incremental analysis for C systems has been hampered by the imprecision of the known techniques for determining pointer-induced aliases. This research aims at developing more precise interprocedural analyses of C systems, based on the conditional analysis approach of our pointer aliasing approximation algorithm. Specifically, it is planned to: o optimize the interprocedural pointer aliasing approximation algorithm, o develop a new, interprocedural modification side effect analysis algorithm for C-like languages and design other flow sensitive data flow analyses (e.g., interprocedural reaching definitions, static program slicing), o build a second prototype incremental MOD analyzer for C systems and profile its performance on actual C systems, and o study issues involved in scaling up for programming-in-the-large.