Detection of asynchronous message passing errors using static analysis

Christakis, M; Sagonas, K

dc.contributor.author	Christakis, M	en
dc.contributor.author	Sagonas, K	en
dc.date.accessioned	2014-03-01T02:47:20Z
dc.date.available	2014-03-01T02:47:20Z
dc.date.issued	2011	en
dc.identifier.issn	03029743	en
dc.identifier.uri	https://dspace.lib.ntua.gr/xmlui/handle/123456789/33080
dc.subject	Concurrent Programs	en
dc.subject	Dynamical Processes	en
dc.subject	interprocess communication	en
dc.subject	Message Passing	en
dc.subject	Static Analysis	en
dc.subject	Long Period	en
dc.subject.other	Close approximation	en
dc.subject.other	Concurrent programming	en
dc.subject.other	Dynamic process	en
dc.subject.other	Erlang programs	en
dc.subject.other	False alarms	en
dc.subject.other	Interprocess communication	en
dc.subject.other	Software defects	en
dc.subject.other	Soundness and completeness	en
dc.subject.other	Errors	en
dc.subject.other	Message passing	en
dc.subject.other	Static analysis	en
dc.title	Detection of asynchronous message passing errors using static analysis	en
heal.type	conferenceItem	en
heal.identifier.primary	10.1007/978-3-642-18378-2_3	en
heal.identifier.secondary	http://dx.doi.org/10.1007/978-3-642-18378-2_3	en
heal.publicationDate	2011	en
heal.abstract	Concurrent programming is hard and prone to subtle errors. In this paper we present a static analysis that is able to detect some commonly occurring kinds of message passing errors in languages with dynamic process creation and communication based on asynchronous message passing. Our analysis is completely automatic, fast, and strikes a proper balance between soundness and completeness: it is effective in detecting errors and avoids false alarms by computing a close approximation of the interprocess communication topology of programs. We have integrated our analysis in dialyzer, a widely used tool for detecting software defects in Erlang programs, and demonstrate its effectiveness on libraries and applications of considerable size. Despite the fact that these applications have been developed over a long period of time and are reasonably well-tested, our analysis has managed to detect a significant number of previously unknown message passing errors in their code. © 2011 Springer-Verlag.	en
heal.journalName	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)	en
dc.identifier.doi	10.1007/978-3-642-18378-2_3	en
dc.identifier.volume	6539 LNCS	en
dc.identifier.spage	5	en
dc.identifier.epage	18	en