Fault-tolerant reinforcement method of dual-core-based operating system and operating system

Abstract

This invention relates to the field of computer technology and discloses a fault-tolerant and hardened method and operating system based on a dual-kernel operating system. The first kernel performs fault detection based on state data and sends a fault signal to a first process when a fault is detected. Upon receiving the fault signal, the first process controls the second kernel to manage the operating system. The second kernel parses and compiles the target task pointed to by the current fault, obtaining multiple redundant instructions. The second kernel executes these redundant instructions sequentially, and if the comparison results of the redundant instructions are inconsistent, it rolls back to the restoration point corresponding to the redundant instructions and re-executes the redundant instructions. When the first kernel detects a fault signal, the second kernel takes over the operating system, ensuring its real-time performance and reliability. Simultaneously, by rolling back to the restoration point of the faulty instruction and re-executing it when a fault is detected, it ensures correct task execution and improves the operating system's fault shielding and recovery capabilities under fault conditions.

Images