A Process-Based Parallel Scheduling Method for Simulation Tasks

Abstract

The invention discloses a simulation task parallel scheduling method based on a progress. The simulation task parallel scheduling method comprises the following steps: for an architecture of multiple processor plug-ins of an embedded device, starting a simulation management progress in a PC to be used as a main progress, and creating a plurality of sub-progresses by virtue of the main progress to virtualize the real-time processor plug-ins; creating a scheduling and data share memory by virtue of the main progress, after the sub-progresses are initialized, starting a single period simulation starting mark by virtue of the main progress, reading the simulation starting mark of the scheduling share memory by virtue of each sub-progress, starting a task operation, setting a sub-progress task operation ending mark after the task is ended, and then entering a waiting state; and after the ending marks of all sub-progresses are collected by virtue of the main progress, issuing a next period starting mark, circularly processing until the set upper limit of the simulation operation times is reached, and exiting the progress.

Description

technical field [0001] The invention belongs to the field of virtual simulation of embedded devices, in particular to a process-based simulation task parallel scheduling method. Background technique [0002] The current embedded device R&D model is still based on serial R&D, that is, it is necessary to develop the hardware with corresponding functions and adjust the system driver before proceeding with application development. For a device with multiple functions, the serial development cycle is long. In order to improve the efficiency of R&D and testing, it is necessary to simulate the actual operating environment and functional interface of the embedded device on the PC to realize the off-line simulation function, save R&D equipment investment, and improve Development and engineering implementation efficiency. The simulation methods in the existing literature often only consider the situation of one CPU, and the method of multi-CPU is considered in a small amount, and the...

AI Technical Summary

Problems solved by technology

Since the programs of multiple CPUs are integrated in one process, it is necessary to avoid duplication of variable names and functions, that is, to modify the programs of each CPU plug-in and add a layer of data encapsulation, resulting in a large difference between the embedded application program and the simulation program, which cannot Achieve homologous maintenance

Images