Test-driven repair of data races in structured parallel programs

Abstract

A common workflow for developing parallel software is as follows: 1) start with a sequential program, 2) identify subcomputations that should be converted to parallel tasks, 3) insert synchronization to achieve the same semantics as the sequential program, and repeat steps 2) and 3) as needed to improve performance. Though this is not the only approach to developing parallel software, it is sufficiently common to warrant special attention as parallel programming becomes ubiquitous. This paper focuses on automating step 3), which is usually the hardest step for developers who lack expertise in parallel programming. Past solutions to the problem of repairing parallel programs have used static-only or dynamic-only approaches, both of which incur significant limitations in practice. Static approaches can guarantee soundness in many cases but are limited in precision when analyzing medium or large-scale software with accesses to pointer-based data structures in multiple procedures. Dynamic approaches are more precise, but their proposed repairs are limited to a single input and are not reflected back in the original source program. In this paper, we introduce a hybrid static+dynamic test-driven approach to repairing data races in structured parallel programs. Our approach includes a novel coupling between static and dynamic analyses. First, we execute the program on a concrete test input and determine the set of data races for this input dynamically. Next, we compute a set of "finish" placements that prevent these races and also respects the static scoping rules of the program while maximizing parallelism. Empirical results on standard benchmarks and student homework submissions from a parallel computing course establish the effectiveness of our approach with respect to compile-time overhead, precision, and performance of the repaired code.

Synthesizing racy tests

“…Test-driven repair [27] combines static and dynamic analysis to identify modifications to code that will prevent races. We differ from these approaches in terms of the intended application and the underlying mechanism.…”

Pegasus: automatic barrier inference for stable multithreaded systems

“…As shown in our experiments, our approach gives better performance without manual intervention. Various techniques [31,44,40] have been proposed to maximize parallelism in programs. Burnim et al [6] propose an assertion framework for specifying that regions of a parallel program behave deterministically despite nondeterministic thread interleaving.…”

“…Cerny et al [7] propose a technique to transform and add synchronization constructs in the concurrent program so that the performance is optimized. A number of approaches have also been proposed for auto-repair of multithreaded programs [44,24]. While these approach work in a non-deterministic environment for structured parallel languages, we propose an approach for reducing execution times on stable multithreaded systems on unstructured parallel languages.…”

Automatic Detection, Validation, and Repair of Race Conditions in Interrupt-Driven Embedded Software

“…K. Wang is with Visa Research, USA (email: kewang@visa.com). concurrency fault repair [34], [38], [43], [44], [73]. These techniques, however, focus on thread-level races.…”