On the complexity of flow-sensitive dataflow analyses

Robert Muth, Saumya K Debray

Research output: Chapter in Book/Report/Conference proceedingConference contribution

18 Citations (Scopus)

Abstract

This paper attempts to address the question of why certain dataflow analysis problems can be solved efficiently, but not others. We focus on flow-sensitive analyses, and give a simple and general result that shows that analyses that require the use of relational attributes for precision must be PSPACE-hard in general. We then show that if the language constructs are slightly strengthened to allow a computation to maintain a very limited summary of what happens along an execution path, inter-procedural analyses become EXPTIME-hard. We discuss applications of our results to a variety of analyses discussed in the literature. Our work elucidates the reasons behind the complexity results given by a number of authors, improves on a number of such complexity results, and exposes conceptual commonalities underlying such results that are not readily apparent otherwise.

Original languageEnglish (US)
Title of host publicationConference Record of the Annual ACM Symposium on Principles of Programming Languages
PublisherACM
Pages67-80
Number of pages14
StatePublished - 2000
EventPOPL'00 - The 27th ACM SIGPLAN-SIGACT Symposium on Principles og Programming Languages - Boston, MA, USA
Duration: Jan 19 2000Jan 21 2000

Other

OtherPOPL'00 - The 27th ACM SIGPLAN-SIGACT Symposium on Principles og Programming Languages
CityBoston, MA, USA
Period1/19/001/21/00

Fingerprint

Data flow analysis

ASJC Scopus subject areas

  • Software

Cite this

Muth, R., & Debray, S. K. (2000). On the complexity of flow-sensitive dataflow analyses. In Conference Record of the Annual ACM Symposium on Principles of Programming Languages (pp. 67-80). ACM.

On the complexity of flow-sensitive dataflow analyses. / Muth, Robert; Debray, Saumya K.

Conference Record of the Annual ACM Symposium on Principles of Programming Languages. ACM, 2000. p. 67-80.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Muth, R & Debray, SK 2000, On the complexity of flow-sensitive dataflow analyses. in Conference Record of the Annual ACM Symposium on Principles of Programming Languages. ACM, pp. 67-80, POPL'00 - The 27th ACM SIGPLAN-SIGACT Symposium on Principles og Programming Languages, Boston, MA, USA, 1/19/00.
Muth R, Debray SK. On the complexity of flow-sensitive dataflow analyses. In Conference Record of the Annual ACM Symposium on Principles of Programming Languages. ACM. 2000. p. 67-80
Muth, Robert ; Debray, Saumya K. / On the complexity of flow-sensitive dataflow analyses. Conference Record of the Annual ACM Symposium on Principles of Programming Languages. ACM, 2000. pp. 67-80
@inproceedings{cde5a960e4284a2a809b304cf5ca68cb,
title = "On the complexity of flow-sensitive dataflow analyses",
abstract = "This paper attempts to address the question of why certain dataflow analysis problems can be solved efficiently, but not others. We focus on flow-sensitive analyses, and give a simple and general result that shows that analyses that require the use of relational attributes for precision must be PSPACE-hard in general. We then show that if the language constructs are slightly strengthened to allow a computation to maintain a very limited summary of what happens along an execution path, inter-procedural analyses become EXPTIME-hard. We discuss applications of our results to a variety of analyses discussed in the literature. Our work elucidates the reasons behind the complexity results given by a number of authors, improves on a number of such complexity results, and exposes conceptual commonalities underlying such results that are not readily apparent otherwise.",
author = "Robert Muth and Debray, {Saumya K}",
year = "2000",
language = "English (US)",
pages = "67--80",
booktitle = "Conference Record of the Annual ACM Symposium on Principles of Programming Languages",
publisher = "ACM",

}

TY - GEN

T1 - On the complexity of flow-sensitive dataflow analyses

AU - Muth, Robert

AU - Debray, Saumya K

PY - 2000

Y1 - 2000

N2 - This paper attempts to address the question of why certain dataflow analysis problems can be solved efficiently, but not others. We focus on flow-sensitive analyses, and give a simple and general result that shows that analyses that require the use of relational attributes for precision must be PSPACE-hard in general. We then show that if the language constructs are slightly strengthened to allow a computation to maintain a very limited summary of what happens along an execution path, inter-procedural analyses become EXPTIME-hard. We discuss applications of our results to a variety of analyses discussed in the literature. Our work elucidates the reasons behind the complexity results given by a number of authors, improves on a number of such complexity results, and exposes conceptual commonalities underlying such results that are not readily apparent otherwise.

AB - This paper attempts to address the question of why certain dataflow analysis problems can be solved efficiently, but not others. We focus on flow-sensitive analyses, and give a simple and general result that shows that analyses that require the use of relational attributes for precision must be PSPACE-hard in general. We then show that if the language constructs are slightly strengthened to allow a computation to maintain a very limited summary of what happens along an execution path, inter-procedural analyses become EXPTIME-hard. We discuss applications of our results to a variety of analyses discussed in the literature. Our work elucidates the reasons behind the complexity results given by a number of authors, improves on a number of such complexity results, and exposes conceptual commonalities underlying such results that are not readily apparent otherwise.

UR - http://www.scopus.com/inward/record.url?scp=0033717006&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0033717006&partnerID=8YFLogxK

M3 - Conference contribution

SP - 67

EP - 80

BT - Conference Record of the Annual ACM Symposium on Principles of Programming Languages

PB - ACM

ER -