LACE：用于敏捷RISC-V指令扩展的大语言模型辅助多智能体框架

中文摘要 AI 辅助

特定领域指令集架构扩展（ISAX）被广泛应用于RISC-V生态系统以加速新兴工作负载，但在不同核心上实现和验证ISAX的过程仍缓慢且碎片化。现有框架仍需要针对每个核心进行接口适配，且一旦微架构或ISAX发生变化，差分测试通常会失效。我们提出LACE，这是一种大语言模型（LLM）辅助的多智能体工作流，可将自然语言形式的ISAX意图转换为紧凑的两级中间表示（操作级和硬件描述语言任务级），在大型代码仓库上执行检索引导的局部寄存器传输级（RTL）编辑，并通过与编译器无关的riscv-formal检查流程（假设存在RVFI可用性或检测工具）形成闭环。在我们的评估设置中，针对四个嵌入式RISC-V核心，LACE将pass@1生成准确率从接近零提升至72.8%，同时改进了代码定位并减少了集成返工。LACE的代码可在该https URL获取。

英文摘要

Domain-specific Instruction Set Architecture eXtensions (ISAX) are widely adopted in the RISC-V ecosystem to accelerate emerging workloads, but implementing and validating ISAXes across different cores remains slow and fragmented. Existing frameworks still require per-core interface adaptation, and differential testing often breaks once either the microarchitecture or the ISAX changes. We present LACE, an LLM-aided multi-agent workflow that translates natural-language ISAX intents into a compact two-level IR (operation-level and HDL task-level), performs retrieval-guided localized RTL edits over large repositories, and closes the loop with a compiler-agnostic riscv-formal checking flow (assuming RVFI availability or instrumentation). Across four embedded RISC-V cores, LACE raises pass@1 generation accuracy from near-zero to 72.8\% within our evaluation setup, while improving code localization and reducing integration rework. The code of LACE is available at https://github.com/UMN-ZhaoLab/LACE.

发表机构

• University of Minnesota, Twin Cities（明尼苏达大学双城分校）
• Columbia University（哥伦比亚大学）
• Northwestern University（西北大学）

机构由 AI 辅助整理，请以论文原文为准。