This repository contains the results of the RMARaceBench classification quality test suite [RRB23].
The MPI RMA, OpenSHMEM, and GASPI test cases were executed without any tool first. The results are available in the folder plain.
For the MPI RMA test cases, the tools MUST-RMA and PARCOACH were executed (PARCOACH provides two different kinds of analyses, so we split the results in PARCOACH-static and PARCOACH-dynamic). The detailed results are available in the folder tools
The following software packages are used for the evaluation in the Docker environment:
- LLVM / Clang compiler version 15.0.6
- OpenMPI 4.1.4
- CMake 3.25.1
- Python 3
- PARCOACH 2.3.1 (webpage, download)
- MUST-RMA 1.9.0 (webpage, download)
[RRB23] Simon Schwitanski, Joachim Jenke, Sven Klotz, and Matthias S. Müller. 2023. RMARaceBench: A Microbenchmark Suite to Evaluate Race Detection Tools for RMA Programs. In Workshops of The International Conference on High Performance Computing, Network, Storage, and Analysis (SC-W 2023), November 12–17, 2023, Denver, CO, USA. ACM, New York, NY, USA. https://doi.org/10.1145/3624062.3624087