Optimization Strategy apply to an Attribute Algorithm for the Abstract Interpretation

Kaninda Musumbu (1)

1 LaBRI - Laboratoire Bordelais de Recherche en Informatique

Presented in this work is a novel approach to implement an efficient abstract interpretation algorithm for logic programs by means of attributed grammar. Successful implementation of this algorithm yielded memory storage saving and run-time reduced. We first execute the algorithm presented in \cite{IEEE}(slightly modified), which generates a tree with four types of nodes. We then store in computer memory only the pertinent subtrees. In other words, after each step, we prune some branches of tree and develop only the subtree which can improve the "oracle" (i.e. set of abstract tuple).

Mots clés: incremental evaluation; Logic programming; static analysis; abstract interpretation; attribute grammar; incremental evaluation.

Domaines: Informatique et langage [cs.CL]

Kaninda Musumbu. Optimization Strategy apply to an Attribute Algorithm for the Abstract Interpretation. The 5the Indian Conference on Artificial Intelligence, Dec 2011, Siddaganga Institute of Technology, Tumkur, India. pp.618-631. ⟨hal-00712076⟩