On the interaction between sharing and linearity

nalysis of logic programs, abstract domains for detecting sharing and linearity ion are widely used. Devising abstract unification algorithms for such domains has proved to be rather hard. At the moment, the available algorithms are correct but not i.e., they cannot fully exploit the information conveyed by the abstract domains. In this paper, we define a new (infinite) domain ShLinω which can be thought of as a framework from which other domains can be easily derived by abstraction. ShLinω makes the interaction between sharing and linearity explicit. We provide a constructive rization of the optimal abstract unification operator on ShLinω , and we lift it to two wn abstractions of ShLinω , to the classical Sharing × Lin abstract domain and to the more precise ShLin2 domain by Andy King. In the case of single-binding tions, we obtain optimal abstract unification algorithms for such domains.

• Type: Journal
• Year: 2007
• Where: CORR
• Authors: Gianluca Amato, Francesca Scozzari
• Keywords: Abstract Domain, Abstract Unification Algorithms, CORR 2007, Education, Optimal Abstract Unification