{"title":"API for Performance Monitoring in Embedded Multicore Systems","authors":"G. Gracioli, A. A. Fröhlich","doi":"10.1109/SBESC.2011.27","DOIUrl":null,"url":null,"abstract":"Hardware Performance Counters (HPCs) are special registers available in the most modern processors that can be used to monitor shared hardware resources in multicore processors. Specifically for embedded real-time applications running on a multicore processor, such shared resources can affect their performance and cause deadline misses. This paper presents a hardware performance counter interface designed to embedded systems. The use of the interface is demonstrated through a benchmark that shares data between two threads executing in different cores of a multicore processor. As a result, the operating system can obtain an accurate view of software's behavior.","PeriodicalId":147899,"journal":{"name":"2011 Brazilian Symposium on Computing System Engineering","volume":"24 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2011-11-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"1","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2011 Brazilian Symposium on Computing System Engineering","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/SBESC.2011.27","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 1
Abstract
Hardware Performance Counters (HPCs) are special registers available in the most modern processors that can be used to monitor shared hardware resources in multicore processors. Specifically for embedded real-time applications running on a multicore processor, such shared resources can affect their performance and cause deadline misses. This paper presents a hardware performance counter interface designed to embedded systems. The use of the interface is demonstrated through a benchmark that shares data between two threads executing in different cores of a multicore processor. As a result, the operating system can obtain an accurate view of software's behavior.