Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

IncreRTL: Traceability-Guided Incremental RTL Generation under Requirement Evolution

Luanrong Chen, Renzhi Chen, Xinyu Li, Shanshan Li, Rui Gong, Lei Wang

Abstract

Large language models (LLMs) have shown promise in generating RTL code from natural-language descriptions, but existing methods remain static and struggle to adapt to evolving design requirements, potentially causing structural drift and costly full regeneration. We propose IncreRTL, a LLM-driven framework for incremental RTL generation under requirement evolution. By constructing requirement-code traceability links to locate and regenerate affected code segments, IncreRTL achieves accurate and consistent updates. Evaluated on our newly constructed EvoRTL-Bench, IncreRTL demonstrates notable improvements in regeneration consistency and efficiency, advancing LLM-based RTL generation toward practical engineering deployment.

IncreRTL: Traceability-Guided Incremental RTL Generation under Requirement Evolution

Abstract

Large language models (LLMs) have shown promise in generating RTL code from natural-language descriptions, but existing methods remain static and struggle to adapt to evolving design requirements, potentially causing structural drift and costly full regeneration. We propose IncreRTL, a LLM-driven framework for incremental RTL generation under requirement evolution. By constructing requirement-code traceability links to locate and regenerate affected code segments, IncreRTL achieves accurate and consistent updates. Evaluated on our newly constructed EvoRTL-Bench, IncreRTL demonstrates notable improvements in regeneration consistency and efficiency, advancing LLM-based RTL generation toward practical engineering deployment.

Paper Structure

This paper contains 28 sections, 3 equations, 5 figures, 3 tables, 1 algorithm.

Table of Contents

  1. Introduction
  2. Related Work
  3. Large Language Models for RTL Code Generation
  4. Requirement Traceability Links
  5. Methodology
  6. Overview
  7. Traceability Link Construction
  8. Structured Representation Construction
  9. Syntax-Preserving Code Segmentation
  10. Semantic Requirement Representation Construction
  11. Trace-Guided Incremental Updates
  12. Requirement--Code Semantic Alignment
  13. Candidate Links Based on Hierarchical Relationships
  14. Link Scoring and Validation
  15. Experiment
  16. ...and 13 more sections

Figures (5)

  • Figure 1: Illustration of requirement-driven RTL updates. IncreRTL performs traceability-guided local regeneration, mitigating token and drift issues.
  • Figure 2: Overview of the proposed IncreRTL framework.
  • Figure 3: Task template for incremental generation, where the LLMs generate only the impacted code snippets, ensuring unchanged line ranges and interfaces.
  • Figure 4: Example of localized regeneration for AR1, where the requirement changes from full 32-bit output to exposing only the upper 24 bits; traceability links confine updates to Fragment 2, and IncreRTL regenerates only its output logic while keeping Fragment 1 unchanged.
  • Figure 5: Performance comparison between IncreRTL and Direct Generation across different large models.