Symbolic-heap separation logic with inductive definitions is a popular formalism for reasoning about heap-manipulating programs. The fragment SLIDbtw introduced by Iosif, Rogalewicz and Simacek, is one of the most expressive fragments with a decidable entailment problem. In recent work, we improved on the original decidability proof by providing a direct model-theoretic construction, obtaining a 2-Exptime upper bound. In this paper, we investigate separation logics built on top of the inductive definitions from SLIDbtw, i.e., logics that feature the standard Boolean and separation-logic operators. We give an almost tight delineation between decidability and undecidabilty. We establish the decidability of the satisfiability problem (in 2-Exptime) of a separation logic with conjunction, disjunction, separating conjunction and guarded forms of negation, magic wand, and septraction. We show that any further generalization leads to undecidabilty (under mild assumptions).

Most automated verifiers for separation logic are based on the symbolic-heap fragment, which disallows both the magic-wand operator and the application of classical Boolean operators to spatial formulas. This is not surprising, as support for the magic wand quickly leads to undecidability, especially when combined with inductive predicates for reasoning about data structures. To circumvent these undecidability results, we propose assigning a more restrictive semantics to the separating conjunction. We argue that the resulting logic, strong-separation logic, can be used for symbolic execution and abductive reasoning just like “standard” separation logic, while remaining decidable even in the presence of both the magic wand and the list-segment predicate—a combination of features that leads to undecidability for the standard semantics.

Most automated verifiers for separation logic are based on the symbolic-heap fragment, which disallows both the magic-wand operator and the application of classical Boolean operators to spatial formulas. This is not surprising, as support for the magic wand quickly leads to undecidability, especially when combined with inductive predicates for reasoning about data structures. To circumvent these undecidability results, we propose assigning a more restrictive semantics to the separating conjunction. We argue that the resulting logic, strong-separation logic, can be used for symbolic execution and abductive reasoning just like “standard” separation logic, while remaining decidable even in the presence of both the magic wand and inductive predicates (we consider a list-segment predicate and a tree predicate)—a combination of features that leads to undecidability for the standard semantics.

We develop a doubly-exponential decision procedure for the satisfiability problem of guarded separation logic —a novel fragment of separation logic featuring user-supplied inductive predicates, Boolean connectives, and separating connectives, including restricted (guarded) versions of negation, magic wand, and septraction. Moreover, we show that dropping the guards for any of the above connectives leads to an undecidable fragment. We further apply our decision procedure to reason about entailments in the popular symbolic heap fragment of separation logic. In particular, we obtain a doubly-exponential decision procedure for entailments between (quantifier-free) symbolic heaps with inductive predicate definitions of bounded treewidth ( \({\mathbf {SL}}_{\mathsf {btw}} \) )—one of the most expressive decidable fragments of separation logic. Together with the recently shown 2ExpTime -hardness for entailments in said fragment, we conclude that the entailment problem for \({\mathbf {SL}}_{\mathsf {btw}} \) is 2ExpTime -complete—thereby closing a previously open complexity gap.

No abstract