Program Dependence Graphs and Dataflow Analysis

程序依赖图和数据流分析

• 批准号: 9625656
• 负责人: Susan Horwitz
• 金额: $ 16.08万
• 依托单位: University of Wisconsin-Madison
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 1996
• 资助国家: 美国
• 起止时间: 1996-08-15 至 1999-07-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=9625656&HistoricalAwards=false
• 关键词: Program; Dependence; Graphs; Dataflow; Analysis

The goals of this project are to advance the state-of-the- art in language-based tools for software engineering. There are three related aspects to the research: (1) extending dependence-graph-based tools, (2) carrying out theoretical and practical studies of pointer analysis, and (3) quantifying trade-offs in dataflow analysis. The work builds on the prior research in two areas: the design and implementation of software engineering tools, and the design, implementation and evaluation of algorithms for dataflow analysis. The first part of the work includes the continuation of previous NSF-supported research extending underlying theory to permit graph-based tools to be used for a wider range of languages than is currently possible, and improving existing graph based tools. It also includes new directions: the design and implementation of new tools using a hybrid approach that combines text or syntax-based tools with semantic checks. The research on pointer analysis and dataflow analysis tradeoffs is complementary to the work on dependence graphs, since the results are expected to improve the graph-based tools. They are also expected to be useful to other researchers and practitioners concerned with applications that rely on the results of static analysis. ***

该项目的目标是推进软件工程中基于语言的工具的最新发展。本研究主要包括三个方面：（1）扩展基于依赖图的工具，（2）进行指针分析的理论和实践研究，（3）量化指针分析中的权衡。 这项工作建立在两个领域的先前研究：软件工程工具的设计和实现，以及算法的设计，实现和评估的EQUILLOW分析。 工作的第一部分包括继续以前NSF支持的研究，扩展基础理论，使基于图的工具能够用于比目前更广泛的语言，并改进现有的基于图的工具。它还包括新的方向：使用混合方法设计和实现新工具，该方法将基于文本或语法的工具与语义检查相结合。 对指针分析和递归分析权衡的研究是对依赖图工作的补充，因为其结果有望改进基于图的工具。他们也预计将是有用的其他研究人员和从业人员关注的应用程序，依赖于静态分析的结果。 ***