Axis: Automatically fixing atomicity violations through solving control constraints

Abstract

Abstract-Atomicity, a general correctness criterion in concurrency programs, is often violated in real-world applications. The violations are difficult for developers to fix, making automatic bug fixing techniques attractive. The state of the art approach aims at automating the manual fixing process but cannot provide any theoretical reasoning and guarantees. We provide an automatic approach that applies well-studied discrete control theory to guarantee deadlocks are not introduced and maximal preservation of the concurrency of the original code. Under the hood, we reduce the problem of violation fixing to a constraint solving problem using the Petri net model. Our evaluation on 13 subjects shows that the slowdown incurred by our patches is only 40% of that of the state of the art. With the deadlock-free guarantee, our patches incur moderate overhead (around 10%), which is a worthwhile cost for safety.

Eliminating Concurrency Bugs in Multithreaded Software: An Approach Based on Control of Petri Nets

“…These include other types of deadlocks, such as reader-writer deadlock, condition wait/signal deadlock, inter-process deadlock, and other concurrency issues such as race, atomicity violation, and priority inversion. Results on the case of reader-writer deadlocks have recently appeared in [33], while certain types of atomicity violations have been addressed in [17,18].…”

“…The principal safety property of interest in the work to-date is deadlock avoidance in multithreaded programs that use locking primitives to control access to shared data [15,16]. Recent results address certain types of atomicity violations in multithreaded programs [17,18].…”

A Survey of Test Based Automatic Program Repair

“…Le Goues et al [71] recommended that test suite and specifications should be combined to enhance repair validity. Specifications like assertions stipulate the object behaviors or method innovations that can help reduce the searching space, and in certain situations specifications describe the program requirements better [72], [73], [74]. Our wild guess is that approaches like automatic programming and program sketching [75] may be helpful, and any innovations on the patch formulation deserve sustained attentions.…”

Griffin: grouping suspicious memory-access patterns to improve understanding of concurrency bugs

“…Other existing techniques attempt to semi-automatically fix concurrency bugs [10,12,14]. The techniques input concurrency bugs reported from bug detectors [9,27], select fix strategies, and produce patches that enforce correct orderings using synchronization operations.…”