Efficient Demand Evaluation of Fixed-Point Attributes Using Static Analysis (Artifact)

This artifact supports the evaluation of a new demand-driven algorithm for efficient circular Reference Attribute Grammar evaluation, specifically designed to improve performance of higly circular applications, e.g., dataflow analyses for Java. The artifact includes all necessary tools, dependencies, benchmarks, and scripts to reproduce the experiments presented in the corresponding paper. It provides a Docker-based setup for easy deployment, as well as detailed instructions for manual installation. The artifact allows users to explore and validate the proposed algorithm's performance improvements through real-world case studies, demonstrating significant speedups in dead-assignment and null-pointer dereference analyses compared to existing algorithms. While the artifact does not require specific hardware, reproducing the experiments as described in the paper may take one to two days of computation time. The artifact has been tested on Linux and macOS systems.