An Operational Semantic Basis for OpenMP Race Analysis

TL;DR

This paper introduces a simple operational semantics for OpenMP that aligns with current runtime events, aiding in the development of reliable race checkers and improving tooling for detecting data races.

Contribution

It provides a rigorous, practical operational semantics for OpenMP that can be directly used for building race detection tools and supporting future language extensions.

Findings

A new data race checker based on the semantics is under development.

The semantics aligns with existing OpenMP runtimes and tracing tools like OMPT.

The semantics can guide future OpenMP extensions and tooling efforts.

Abstract

OpenMP is the de facto standard to exploit the on-node parallelism in new generation supercomputers.Despite its overall ease of use, even expert users are known to create OpenMP programs that harbor concurrency errors, of which one of the most insidious of errors are {\em data races}.OpenMP is also a rapidly evolving standard, which means that future data races may be introduced within unfamiliar contexts.A simple and rigorous operational semantics for OpenMP can help build reliable race checkers and ward off future errors through programmer education and better tooling.This paper's key contribution is a simple operational semantics for OpenMP, with primitive events matching those generated by today's popular OpenMP runtimes and tracing methods such as OMPT.This makes our operational semantics more than a theoretical document for intellectual edification; it can serve as a blueprint for OpenMP event capture and tool building.We back this statement by summarizing the workings of a new data race checker for OpenMP being built based on this semantics.The larger purpose served by our semantics is to serve the needs of the OpenMP community with regard to their contemplated extensions to OpenMP, as well as future tooling efforts.