HDLxGraph: Bridging Large Language Models and HDL Repositories via HDL Graph Databases

TL;DR

HDLxGraph enhances large language model tasks on HDL code repositories by integrating graph structures like ASTs and DFGs, significantly improving search and debugging accuracy over existing methods.

Contribution

This paper introduces HDLxGraph, the first framework combining HDL graph structures with RAGs, and presents HDLSearch, a new HDL code benchmark dataset.

Findings

HDLxGraph improves search accuracy by over 12%.

HDLxGraph enhances debugging and completion tasks.

HDLSearch provides a new benchmark for HDL code retrieval.

Abstract

Retrieval Augmented Generation (RAG) is an essential agent for Large Language Model (LLM) aided Description Language (HDL) tasks, addressing the challenges of limited training data and prohibitively long prompts. However, its performance in handling ambiguous queries and real-world, repository-level HDL projects containing thousands or even tens of thousands of code lines remains limited. Our analysis demonstrates two fundamental mismatches, structural and vocabulary, between conventional semantic similarity-based RAGs and HDL codes. To this end, we propose HDLxGraph, the first framework that integrates the inherent graph characteristics of HDLs with RAGs for LLM-assisted tasks. Specifically, HDLxGraph incorporates Abstract Syntax Trees (ASTs) to capture HDLs' hierarchical structures and Data Flow Graphs (DFGs) to address the vocabulary mismatch. In addition, to overcome the lack of comprehensive HDL search benchmarks, we introduce HDLSearch, an LLM generated dataset derived from real-world, repository-level HDL projects. Evaluations show that HDLxGraph improves search, debugging, and completion accuracy by 12.04%/12.22%/5.04% and by 11.59%/8.18%/4.07% over state-of-the-art similarity-based RAG and software-code Graph RAG baselines, respectively. The code of HDLxGraph and HDLSearch benchmark are available at https://github.com/UMN-ZhaoLab/HDLxGraph.