A Synchronous IPC Protocol for Predictable Access to Shared Resources in Mixed-Criticality Systems

Björn B. Brandenburg
{"title":"A Synchronous IPC Protocol for Predictable Access to Shared Resources in Mixed-Criticality Systems","authors":"Björn B. Brandenburg","doi":"10.1109/RTSS.2014.37","DOIUrl":null,"url":null,"abstract":"In mixed-criticality systems, highly critical tasks must be temporally and logically isolated from faults in lower-criticality tasks. Such strict isolation, however, is difficult to ensure even for independent tasks, and has not yet been attained if low- and high-criticality tasks share resources subject to mutual exclusion constraints (e.g., Shared data structures, peripheral I/O devices, or OS services), as it is often the case in practical systems. Taking a pragmatic, systems-oriented point of view, this paper argues that traditional real-time locking approaches are unsuitable in a mixed-criticality context: locking is a cooperative activity and requires trust, which is inherently in conflict with the paramount isolation requirements. Instead, a solution based on resource servers (in the microkernel sense) is proposed, and MC-IPC, a novel synchronous multiprocessor IPC protocol for invoking such servers, is presented. The MC-IPC protocol enables strict temporal and logical isolation among mutually untrusted tasks and thus can be used to share resources among tasks of different criticalities. It is shown to be practically viable with a prototype implementation in LITMUSRT and validated with a case study involving several antagonistic failure modes. Finally, MC-IPC is shown to offer analytical benefits in the context of Vestal's mixed-criticality task model.","PeriodicalId":353167,"journal":{"name":"2014 IEEE Real-Time Systems Symposium","volume":"15 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2014-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"27","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2014 IEEE Real-Time Systems Symposium","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/RTSS.2014.37","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 27

Abstract

In mixed-criticality systems, highly critical tasks must be temporally and logically isolated from faults in lower-criticality tasks. Such strict isolation, however, is difficult to ensure even for independent tasks, and has not yet been attained if low- and high-criticality tasks share resources subject to mutual exclusion constraints (e.g., Shared data structures, peripheral I/O devices, or OS services), as it is often the case in practical systems. Taking a pragmatic, systems-oriented point of view, this paper argues that traditional real-time locking approaches are unsuitable in a mixed-criticality context: locking is a cooperative activity and requires trust, which is inherently in conflict with the paramount isolation requirements. Instead, a solution based on resource servers (in the microkernel sense) is proposed, and MC-IPC, a novel synchronous multiprocessor IPC protocol for invoking such servers, is presented. The MC-IPC protocol enables strict temporal and logical isolation among mutually untrusted tasks and thus can be used to share resources among tasks of different criticalities. It is shown to be practically viable with a prototype implementation in LITMUSRT and validated with a case study involving several antagonistic failure modes. Finally, MC-IPC is shown to offer analytical benefits in the context of Vestal's mixed-criticality task model.
混合临界系统中可预测访问共享资源的同步IPC协议
在混合临界系统中,高临界任务必须在时间上和逻辑上与低临界任务中的故障隔离。然而,即使对于独立的任务,也很难确保这种严格的隔离,如果低临界和高临界的任务在互斥约束下共享资源(例如,共享数据结构、外设I/O设备或操作系统服务),就像在实际系统中经常出现的情况一样,这种隔离还没有实现。从实用的、面向系统的角度出发,本文认为传统的实时锁定方法不适用于混合临界环境:锁定是一种合作活动,需要信任,这本质上与最重要的隔离要求相冲突。本文提出了一种基于资源服务器(微内核意义上的)的解决方案,并提出了一种新的用于调用资源服务器的同步多处理器IPC协议MC-IPC。MC-IPC协议在相互不信任的任务之间实现严格的时间和逻辑隔离,从而可以用于在不同关键度的任务之间共享资源。通过LITMUSRT的原型实现证明了它实际上是可行的,并通过涉及几种拮抗失效模式的案例研究进行了验证。最后,MC-IPC被证明在Vestal的混合临界任务模型的背景下提供分析优势。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
求助全文
约1分钟内获得全文 求助全文
来源期刊
自引率
0.00%
发文量
0
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术官方微信