A concurrent program logic with a future and history

Abstract

Verifying fine-grained optimistic concurrent programs remains an open problem. Modern program logics provide abstraction mechanisms and compositional reasoning principles to deal with the inherent complexity. However, their use is mostly confined to pencil-and-paper or mechanized proofs. We devise a new separation logic geared towards the lacking automation. While local reasoning is known to be crucial for automation, we are the first to show how to retain this locality for (i) reasoning about inductive properties without the need for ghost code, and (ii) reasoning about computation histories in hindsight. We implemented our new logic in a tool and used it to automatically verify challenging concurrent search structures that require inductive properties and hindsight reasoning, such as the Harris set.

Make Flows Small Again: Revisiting the Flow Framework

“…In practice, path-counting flows, keyset flows, reachability flows, shortest-path flows, and priority inheritance flows are relevant [22][23][24]27] and compatible with our theory.…”

“…1, the flow framework has been used to verify complex concurrent data structures. More specifically, it has been used for automated proof construction by the plankton tool [26,27]. plankton performs an exhaustive proof search over a separation logic with support for flows-and further advanced features for establishing linearizability that do not matter for the present evaluation.…”

“…This algorithm enables the automation of the frame rule for reasoning about programs manipulating flow graphs. We evaluate the presented algorithms on a benchmark suite of flow graph updates that are extracted from linearizability proofs for concurrent search structures constructed by the tool plankton [26,27]. The evaluation demonstrates that our algorithms help to automate key aspects of these proofs that have previously relied on user guidance or heuristics.…”

“…The benchmark suite for our evaluation is obtained from plankton [26,27], a tool for verifying concurrent search structures using keyset flows. When the program mutates the symbolic heap, plankton creates a flow graph for the mutated nodes plus all nodes with a distance of k or less from those nodes.…”

“…The flow framework [23,24] is an abstraction mechanism based on separation logic [6,32,40] that enables reasoning about global inductive invariants of general graphs in a local manner. The framework has proved useful to verify intricate algorithms that are difficult to handle by other techniques, such as the Priority Inheritance Protocol, object-oriented design patterns, and complex concurrent data structures [22,24,27,34]. However, these efforts have also exposed some rough corners in the underlying meta theory that either limit expressivity or automation.…”