The Complexity of Model Checking Multi-Stack Systems - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2013

The Complexity of Model Checking Multi-Stack Systems

Résumé

We consider the linear-time model checking problem for boolean concurrent programs with recursive procedure calls. While sequential recursive programs are usually modeled as pushdown automata, concurrent recursive programs involve several processes and can be naturally abstracted as pushdown automata with multiple stacks. Their behavior can be understood as words with multiple nesting relations, each relation connecting a procedure call with its corresponding return. To reason about multiply nested words, we consider the class of all temporal logics as defined in the book by Gabbay, Hodkinson, and Reynolds (1994). The unifying feature of these temporal logics is that their modalities are defined in monadic second-order (MSO) logic. In particular, this captures numerous temporal logics over concurrent and/or recursive programs that have been defined so far. Since the general model checking problem is undecidable, we restrict attention to phase bounded executions as proposed by La Torre, Madhusudan, and Parlato (LICS 2007). While the MSO model checking problem in this case is non-elementary, our main result states that the model checking (and satisfiability) problem for all MSO-definable temporal logics is decidable in elementary time. More precisely, it is solvable in (n + 2)-EXPTIME where n is the maximal level of the MSO modalities in the monadic quantifier alternation hierarchy. We complement this result and provide, for each level n, a temporal logic whose model checking problem is n-EXPSPACE-hard.

Dates et versions

hal-00926182 , version 1 (09-01-2014)

Identifiants

Citer

Benedikt Bollig, Dietrich Kuske, Roy Mennicke. The Complexity of Model Checking Multi-Stack Systems. 28th Annual IEEE Symposium on Logic in Computer Science (LICS'13), 2013, United States. pp.163-170, ⟨10.1109/LICS.2013.22⟩. ⟨hal-00926182⟩
51 Consultations
0 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More