AwareCompiler: Agentic Context-Aware Compiler Optimization via a Synergistic Knowledge-Data Driven Framework
This work addresses compiler optimization for software developers, presenting an incremental improvement through a hybrid knowledge-data-driven approach.
The paper tackles automating compiler optimization by addressing semantic misalignment, inefficient agent interactions, and reward sparsity, resulting in a framework that significantly outperforms existing baselines in performance and efficiency on standard benchmarks.
Compiler optimization is crucial for enhancing program performance by transforming the sequence of optimization passes while maintaining correctness. Despite the promising potential of large language models (LLMs)-based agent for software optimization, automating compiler optimization remains challenging due to: (1) semantic misalignment between abstract program representations and concrete optimization passes, (2) inefficient interaction mechanisms between agents and compiler environments, and (3) reward sparsity from the extensive decision-making process within large optimization spaces. This paper introduces \textbf{AwareCompiler}, an agentic framework for compiler optimization that addresses these challenges through three key innovations: structured knowledge integration and dataset construction, knowledge-driven adaptive pass generation, and data-driven hybrid training pipeline. Experimental results on standard benchmarks demonstrate that AwareCompiler significantly outperforms existing baselines in both performance and efficiency, highlighting the effectiveness of our synergistic knowledge-data-driven approach. Our code is publicly available at https://github.com/LHY-24/AwareCompiler.