支持RTL动态适配的NoC验证方法、装置及设备

By using a virtual module-based NoC verification architecture and IP-XACT file parsing, the integrity and standardization of SoC-level NoC verification are achieved, solving the problems of incomplete verification layers and low standardization in existing technologies, and improving verification efficiency and accuracy.

CN121979733BActive Publication Date: 2026-07-17SIENGINE TECH CO LTD

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
SIENGINE TECH CO LTD
Filing Date
2026-04-03
Publication Date
2026-07-17

AI Technical Summary

Technical Problem

Existing NoC verification methods lack completeness in SoC-level verification environments, do not adopt the IEEE 1685-2009 standard, resulting in incomplete verification layers, low standardization, and a lack of pre-verification mechanisms, which affects verification efficiency.

Method used

The NoC verification architecture, based on the Dummy Module, is adopted. By parsing the IP-XACT file and NoC design document, it automatically connects to the AMBA VIP, performs content consistency detection and dynamic adaptation, and generates a NoC verification platform to achieve efficient verification of SoC-level end-to-end communication and cross-NoC access.

Benefits of technology

It achieves complete SoC-level verification, improves the applicability and portability of the verification platform, detects design errors at an early stage, and improves verification efficiency and accuracy.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN121979733B_ABST
    Figure CN121979733B_ABST
Patent Text Reader

Abstract

本发明公开了一种支持RTL动态适配的NoC验证方法、装置及设备,涉及芯片技术领域,该方法包括将RTL新版本与IP‑XACT文件、NoC设计文档和Database文件一并发布,并通过预先创建的自动化工具解析所述IP‑XACT文件和NoC设计文档,并自动连接和配置AMBA VIP以及更新NoC验证平台;通过所述自动化工具解析所述Database文件,并进行IP‑XACT文件、NoC设计文档与Database文件的内容一致性检测;其中,所述自动化工具包括用于生成NoC验证平台的组件,以及进行内容一致性检测的组件。本申请能够实现在SoC级验证环境中对多NoC的端到端通信、跨NoC访问等复杂场景的高效验证。
Need to check novelty before this filing date? Find Prior Art