一种基于多核心CPU提升Linux操作系统应用程序实时性的方法

By reallocating tasks and configuring interrupt priorities on a multi-core CPU, the problem of insufficient real-time performance of traditional Linux systems in industrial control scenarios is solved, achieving efficient and economical real-time performance improvement, which is applicable to fields such as distribution network automation and smart grids.

CN122132148BActive Publication Date: 2026-07-17ZHUHAI FEISEN POWER TECH CO LTD

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
ZHUHAI FEISEN POWER TECH CO LTD
Filing Date
2026-04-30
Publication Date
2026-07-17

AI Technical Summary

Technical Problem

Traditional Linux systems lack real-time performance in industrial control scenarios and cannot meet the needs of applications sensitive to deterministic latency, such as power distribution automation. They suffer from problems such as interrupt response delays, kernel non-preemption, spinlock blocking preemption, non-deterministic scheduler design, priority inversion, and memory allocation blocking.

Method used

By reallocating tasks across multi-core CPUs, utilizing CPU affinity and interrupt priority configuration, and modifying kernel boot parameters, real-time critical tasks are isolated to independent cores and interrupt preemptive mode is enabled. This ensures that high-priority tasks can interrupt low-priority processing, achieving physical isolation and priority protection for tasks.

Benefits of technology

It significantly reduces system costs, simplifies the development process, improves system real-time performance and stability, meets the application scenarios with extreme real-time requirements such as distribution network automation, and has good cross-platform compatibility and flexible configuration capabilities.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122132148B_ABST
    Figure CN122132148B_ABST
Patent Text Reader

Abstract

本发明提供一种基于多核心CPU提升Linux操作系统应用程序实时性的方法,包括在系统启动阶段,通过修改Linux内核的启动参数,对多核心CPU中的各个核心任务进行重新分配,将Linux系统的基础运行任务集中分配至第零核心上运行;目标应用程序在创建任务时,根据任务的实时性要求,通过调用系统函数设置任务亲和性;将非实时关键性任务分配至第一核心上运行;将实时关键性任务分配至第二核心上运行;将次实时关键性任务分配至第三核心上运行。本发明通过利用多核心CPU的不同核心分配不同任务,有效降低Linux系统中用于实时要求严格的特定中断的响应延迟时间,提高设备实时性,满足配网自动化等领域的实时性要求。
Need to check novelty before this filing date? Find Prior Art