Modular static analysis with zonotopes

Eric Goubault; Sylvie Putot; Franck Védrine

doi:10.1007/978-3-642-33125-1_5

Modular static analysis with zonotopes

Eric Goubault
, Sylvie Putot
, Franck Védrine

LIST-DTSI-SLA CEA

Research output: Contribution to journal › Conference article › peer-review

Abstract

Being able to analyze programs function by function, or module by module is a key ingredient to scalable static analyses. The main difficulty for modular static analysis is to be able to do so while not losing too much precision. In this paper, we present a new summary-based approach that builds on previous work of the authors, a zonotopic functional abstraction, that is economical both in space and time complexity. This approach has been implemented, and experiments on numerical programs, reported here, show that this approach is very efficient, and that we still obtain precise analyses in realistic cases.

Original language	English
Journal	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume	7460 LNCS
DOIs	https://doi.org/10.1007/978-3-642-33125-1_5
Publication status	Published - 1 Jan 2012
Externally published	Yes
Event	19th International Static Analysis Symposium, SAS 2012 - Deauville, France Duration: 11 Sept 2012 → 13 Sept 2012

Access to Document

10.1007/978-3-642-33125-1_5

Cite this

@article{a40a2d129d83408e96ea59c1c86c9272,

title = "Modular static analysis with zonotopes",

abstract = "Being able to analyze programs function by function, or module by module is a key ingredient to scalable static analyses. The main difficulty for modular static analysis is to be able to do so while not losing too much precision. In this paper, we present a new summary-based approach that builds on previous work of the authors, a zonotopic functional abstraction, that is economical both in space and time complexity. This approach has been implemented, and experiments on numerical programs, reported here, show that this approach is very efficient, and that we still obtain precise analyses in realistic cases.",

author = "Eric Goubault and Sylvie Putot and Franck V{\'e}drine",

year = "2012",

month = jan,

day = "1",

doi = "10.1007/978-3-642-33125-1\_5",

language = "English",

volume = "7460 LNCS",

journal = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

issn = "0302-9743",

note = "19th International Static Analysis Symposium, SAS 2012 ; Conference date: 11-09-2012 Through 13-09-2012",

}

TY - JOUR

T1 - Modular static analysis with zonotopes

AU - Goubault, Eric

AU - Putot, Sylvie

AU - Védrine, Franck

PY - 2012/1/1

Y1 - 2012/1/1

N2 - Being able to analyze programs function by function, or module by module is a key ingredient to scalable static analyses. The main difficulty for modular static analysis is to be able to do so while not losing too much precision. In this paper, we present a new summary-based approach that builds on previous work of the authors, a zonotopic functional abstraction, that is economical both in space and time complexity. This approach has been implemented, and experiments on numerical programs, reported here, show that this approach is very efficient, and that we still obtain precise analyses in realistic cases.

AB - Being able to analyze programs function by function, or module by module is a key ingredient to scalable static analyses. The main difficulty for modular static analysis is to be able to do so while not losing too much precision. In this paper, we present a new summary-based approach that builds on previous work of the authors, a zonotopic functional abstraction, that is economical both in space and time complexity. This approach has been implemented, and experiments on numerical programs, reported here, show that this approach is very efficient, and that we still obtain precise analyses in realistic cases.

U2 - 10.1007/978-3-642-33125-1_5

DO - 10.1007/978-3-642-33125-1_5

M3 - Conference article

AN - SCOPUS:85054848775

SN - 0302-9743

VL - 7460 LNCS

JO - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

JF - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

T2 - 19th International Static Analysis Symposium, SAS 2012

Y2 - 11 September 2012 through 13 September 2012

ER -

Modular static analysis with zonotopes

Abstract

Access to Document

Fingerprint

Cite this