摘要
智能汽车的座舱系统往往采用高性能SOC计算平台,实时任务与非实时任务共存,促使嵌入式虚拟化技术快速发展。针对1型嵌入式虚拟机管理器Xvisor的实时调度问题,采用组合调度分析法,对RTOS中VCPU的周期和时间片参数进行优化,在保证实时任务可调度的基础上,最大化非实时任务处理器资源。修改Xvisor-0.3.0的VCPU调度策略,在树莓派3b上进行实验,验证结果显示,所求最优参数可以保证实时任务可调度,旨在确保嵌入式虚拟机管理器在智能汽车领域应用中尽量节约计算资源,充分发挥处理器性能。
The high performance SOC computing platform is often used in the cockpit system of intelligent vehicles.The coexistence of real-time tasks and non real-time tasks promotes the rapid development of embedded virtualization technology.Aiming at the real-time scheduling problem of type-1 embedded hypervisor,Xvisor,the Compositional Schedulability Analysis method is adopted to optimize the periodicity and time slice parameters of VCPU in RTOS,and maximize the processor resources of non real-time tasks on the basis of ensuring the schedulability of real-time tasks.The VCPU scheduling policy of Xvisor-0.3.0 is modified and performed on Raspberry Pi 3 Model B.It is verified that the optimal parameters can ensure real-time tasks to be scheduled.It aims to ensure that the embedded virtual machine man-ager can save computing resources as much as possible in the application of intelligent vehicles and give full play to the processor performance.
作者
吕孟军
石春
吴刚
秦琳琳
LYU Mengjun;SHI Chun;WU Gang;QIN Linin(Department of Automation,University of Science and Technology of China,Hefei 230027,China)
出处
《仪表技术》
2023年第3期31-36,共6页
Instrumentation Technology
关键词
嵌入式
虚拟机管理器
组合调度分析
调度算法
实时性
embedded
hypervisor
compositional schedulability analysis
scheduling algorithm
real-time per-formance