Search papers, labs, and topics across Lattice.
AgRefactor is a novel LLM-based multi-agent workflow designed to refactor software into High-Level Synthesis (HLS)-compatible programs, addressing the challenges posed by restrictive language support and the disparity between software and hardware programming practices. By leveraging a self-evolving memory system, AgRefactor enhances the efficiency and robustness of refactoring tasks, achieving significant performance improvements on complex benchmarks. The system demonstrates a 6.51x speedup over existing state-of-the-art tools while maintaining resource efficiency, making it a compelling solution for real-world software synthesis challenges.
AgRefactor achieves a staggering 6.51x speedup over leading pragma tuning tools, revolutionizing the efficiency of software refactoring for HLS compatibility.
High-Level Synthesis (HLS) provides a fast path from concepts to silicon, but converting real-world software into synthesizable HLS code remains challenging due to restrictive language support and the gap between software and hardware programming practices. Existing automated and LLM-based refactoring approaches partially address this problem, yet they often lack flexibility, struggle to scale, and incur high computational costs. We introduce AgRefactor, an LLM-based multi-agent workflow for refactoring software into HLS-compatible programs. AgRefactor incorporates a self-evolving memory system that accumulates and retrieves factual and strategic knowledge across tasks, improving robustness and efficiency on unseen programs. To reduce cost and enhance scalability, it integrates automated refactoring tools, enabling agents to balance LLM-driven rewrites with efficient tool-based transformations. On 9 out of 11 challenging real-world benchmarks, which are 5-10x longer than the most complex cases studied in prior work, AgRefactor outperforms or matches the state-of-the-art automated refactoring tool and a strong LLM-based baseline built on the same framework backbone. Further agentic performance optimization yields a 6.51x geometric mean speedup over the SoTA pragma tuning tool and a 1.20x speedup over optimized open-source designs with less than 20% extra resources. AgRefactor is fully-automated and open-sourced.
