Conference Paper

Improving Abstract Interpretations by Combining Domains.

Department of Computer Science, University of Leuven, Louvain, Flemish, Belgium
DOI: 10.1145/200994.200998
Source: DBLP


This article considers static analysis based on abstract interpretation of logic programs over combined domains. It is known that analyses over combined domains provide more information potentially than obtained by the independent analyses. However, the construction of a combined analysis often requires redefining the basic operations for the combined domain. A practical approach to maintain precision in combined analyses of logic programs which reuses the individual analyses and does not redefine the basic operations is illustrated. The advantages of the approach are that (1) proofs of correctness for the new domains are not required and (2) implementations can be reused. The approach it; demonstrated by showing that a combined sharing analysis constructed from ''old'' proposals - compares well with other ''new'' proposals suggested in recent Literature both from the point of view of efficiency and accuracy.

0 Reads
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: ing Unification: A Key Step in the Design of Logic Program Analyses Maurice Bruynooghe 1 and Michael Codish 2 and Anne Mulkers 1 1 K.U.Leuven, Department of Computer Science, Celestijnenlaan 200A, 3001 Heverlee, Belgium 2 Ben-Gurion University, Department of Mathematics and Computer Science, P.O.B. 653, 84105 Beer-Sheva, Israel Abstract. This paper focuses on one of the key steps in the design of semantic based analyses for logic programs --- the definition of an abstract unification algorithm for a given notion of data description. We survey some of the major notions of data descriptions proposed in the context of mode and sharing analyses. We demonstrate how a careful and systematic analysis of the underlying concrete unification algorithm contributes to the design of the abstract algorithm. Several relevant properties of concrete substitutions which influence the design of abstract domains and algorithms are described. We make use of a novel representation called abstract eq...
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: The availability of freeness and sharing information for logic pro- grams has proven useful in a wide variety of applications. However, deriving correct freeness and sharing information turns out to be a complex task. In a recent paper, Codish et al. introduce a domain of abstract equation systems and illustrate the systematic derivation of an abstract unification algorithm for sharing and freeness. This paper follows the same approach, and using a more detailed abstraction, develops a more powerful transition system. The resulting algorithm is more prone to efficient implementation as it computes a single abstract solved form which exhibits sharing, freeness, groundness and linearity information.
    Static Analysis, Third International Workshop, WSA'93, Padova, Italy, September 22-24, 1993, Proceedings; 01/1993
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: . Goal independent analysis of logic programs is commonly discussed in the context of the bottom-up approach. However, while the literature is rich in descriptions of top-down analysers and their application, practical experience with bottom-up analysis is still in a preliminary stage. Moreover, the practical use of existing top-down frameworks for goal independent analysis has not been addressed in a practical system. We illustrate the efficient use of existing goal dependent, top-down frameworks for abstract interpretation in performing goal independent analyses of logic programs much the same as those usually derived from bottom-up frameworks. We present several optimizations for this flavour of top-down analysis. The approach is fully implemented within an existing top-down framework. Several implementation tradeoffs are discussed as well as the influence of domain characteristics. An experimental evaluation including a comparison with a bottom-up analysis for the domain Prop is p...
Show more