Partial Order Reduction for Event-Driven Multi-threaded Programs

Abstract

Event-driven multi-threaded programming is fast becoming a preferred style of developing efficient and responsive applications. In this concurrency model, multiple threads execute concurrently, communicating through shared objects as well as by posting asynchronous events that are executed in their order of arrival. In this work, we consider partial order reduction (POR) for event-driven multi-threaded programs. The existing POR techniques treat event queues associated with threads as shared objects and thereby, reorder every pair of events handled on the same thread even if reordering them does not lead to different states. We do not treat event queues as shared objects and propose a new POR technique based on a novel backtracking set called the dependence-covering set. Events handled by the same thread are reordered by our POR technique only if necessary. We prove that exploring dependencecovering sets suffices to detect all deadlock cycles and assertion violations defined over local variables. To evaluate effectiveness of our POR scheme, we have implemented a dynamic algorithm to compute dependence-covering sets. On execution traces obtained from a few Android applications, we demonstrate that our technique explores many fewer transitions -often orders of magnitude fewer-compared to exploration based on persistent sets, wherein, event queues are considered as shared objects.

Stateful Dynamic Partial Order Reduction for Model Checking Event-Driven Applications that Do Not Terminate

“…We did not compare our implementation against other systems, e.g., eventdriven systems [22,30]. Not only that these systems do not perform stateful model checking and handle cyclic state spaces, but also they implemented their algorithms in different domains: web [22] and Android applications [30]-it will not be straightforward to adapt and compare these with our implementation on smart home apps.…”

“…Recent work on dynamic partial order reduction for event-driven programs has developed dynamic partial order reduction algorithms for stateless model checking of event-driven applications [22,30]. Jensen et al [22] consider a model similar to ours in which an event is treated as a single transition, while Maiya et al [30] consider a model in which event execution interleaves concurrently with threads. Neither of these approaches handle cyclic state spaces nor consider challenges that arise from stateful model checking.…”

“…Model-checking tools can be helpful for finding subtle concurrency bugs or understanding complex interactions between different applications [49]. In recent years, significant work has been expended on developing model checkers for multithreaded concurrency [2,19,22,25,63,61,56,59], but event-driven systems have received much less attention [22,30].…”

Lifestate: Event-Driven Protocols and Callback Control Flow (Extended Version)

“…The protocol verification problem for event-driven applications is related to typestate verification [34,26,17], but it is more complex since it requires reasoning about the asynchronous interaction of both callbacks and callins. Dynamic protocol verification is similar in spirit to dynamic event-race detection [32,23,7,31], which predicts if there is an event data-race from execution traces. However, a lifestate violation differs from, and is not directly comparable to, an event data-race.…”

Systematic testing of actor systems

“…Persistent sets (backtrack sets) are computed by considering transitions occurred in the past and not considering the effects of future transitions as done by the approach we follow, described in the paper by Flanagan and Godefroid . In the paper by Maiya et al , an effective POR strategy is proposed for a combined model of threads and events, what they call event‐driven multithreaded programs. The standard notion of dependency is not suitable in this framework, and thus, it is required to redefine the dependence relation to capture not only single‐threaded but also multithreaded dependencies.…”