Models and Separation Logics for Resource Trees

Abstract

In this article, we propose a new data structure, called resource tree, that is a node-labelled tree in which nodes contain resources which belong to a partial monoid. We define the resource tree model and a new separation logic (BI-Loc) that extends the Bunched Implications logic (BI) with a modality for locations. In addition, we consider quantifications on locations and paths and then we study decidability by model-checking in these models and logics. Moreover, we define a language to deal with resource trees and also an assertion logic derived from BI-Loc. Then soundness and completeness issues are studied, and we show how the model and its associated language can be used to manage heap structures and also permission accounting.