{"title":"Minimizing stack usage for AUTOSAR/OSEK's restricted fixed-priority preemption threshold support","authors":"L. Hatvani, R. J. Bril","doi":"10.1109/SIES.2016.7509417","DOIUrl":null,"url":null,"abstract":"Fixed priority preemption threshold scheduling (FPTS) may significantly improve the schedulability ratio of task sets compared to both fixed-priority pre-emptive scheduling (FPPS) and fixed-priority non-preemptive scheduling (FPNS). Moreover, FPTS reduces stack memory requirements compared to FPPS. Unfortunately, the scheduling policy defined by the standard automotive platform AUTOSAR/OSEK only supports a restricted version of FPTS. In earlier work, the consequences of these limitations have been investigated for the schedulability ratio of task sets on a uniprocessor platform. This paper considers the consequences for the stack memory requirements. To that end, it presents a preemption threshold assignment algorithm for minimizing stack usage under FPTS on an AUTOSAR/OSEK platform. The paper includes a comparative evaluation of the stack usage of FPTS without restrictions and FPTS as defined by AUTOSAR/OSEK.","PeriodicalId":185636,"journal":{"name":"2016 11th IEEE Symposium on Industrial Embedded Systems (SIES)","volume":"84 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2016-05-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"4","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2016 11th IEEE Symposium on Industrial Embedded Systems (SIES)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/SIES.2016.7509417","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 4
Abstract
Fixed priority preemption threshold scheduling (FPTS) may significantly improve the schedulability ratio of task sets compared to both fixed-priority pre-emptive scheduling (FPPS) and fixed-priority non-preemptive scheduling (FPNS). Moreover, FPTS reduces stack memory requirements compared to FPPS. Unfortunately, the scheduling policy defined by the standard automotive platform AUTOSAR/OSEK only supports a restricted version of FPTS. In earlier work, the consequences of these limitations have been investigated for the schedulability ratio of task sets on a uniprocessor platform. This paper considers the consequences for the stack memory requirements. To that end, it presents a preemption threshold assignment algorithm for minimizing stack usage under FPTS on an AUTOSAR/OSEK platform. The paper includes a comparative evaluation of the stack usage of FPTS without restrictions and FPTS as defined by AUTOSAR/OSEK.