{"title":"不同编译器优化水平下的嵌入式软件性能评估","authors":"P. Ruberg, Keijo Lass, Elvar Liiv, P. Ellervee","doi":"10.1109/AIEEE.2017.8270530","DOIUrl":null,"url":null,"abstract":"Tools for performance estimation based on instruction set simulators (ISS) are mostly available and show good results. However when the need arises to choose a different platform or to estimate performance without having the ISS the developer needs all the different software suits and devices and also must be able to work with them. In this case we propose an estimation method based on physical measurement for generating performance models. This paper extends our previous work on source-code level performance estimations for microcontrollers. We compare two proposed estimation methods to find the most suitable for estimating embedded software performance for C source-code level on microcontrollers with higher compiler optimisation levels than −O0. As a result we show that both methods could be applied with some exceptions.","PeriodicalId":224275,"journal":{"name":"2017 5th IEEE Workshop on Advances in Information, Electronic and Electrical Engineering (AIEEE)","volume":"161 ","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2017-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"2","resultStr":"{\"title\":\"Embedded software performance estimations at different compiler optimisation levels\",\"authors\":\"P. Ruberg, Keijo Lass, Elvar Liiv, P. Ellervee\",\"doi\":\"10.1109/AIEEE.2017.8270530\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Tools for performance estimation based on instruction set simulators (ISS) are mostly available and show good results. However when the need arises to choose a different platform or to estimate performance without having the ISS the developer needs all the different software suits and devices and also must be able to work with them. In this case we propose an estimation method based on physical measurement for generating performance models. This paper extends our previous work on source-code level performance estimations for microcontrollers. We compare two proposed estimation methods to find the most suitable for estimating embedded software performance for C source-code level on microcontrollers with higher compiler optimisation levels than −O0. As a result we show that both methods could be applied with some exceptions.\",\"PeriodicalId\":224275,\"journal\":{\"name\":\"2017 5th IEEE Workshop on Advances in Information, Electronic and Electrical Engineering (AIEEE)\",\"volume\":\"161 \",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2017-11-01\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"2\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"2017 5th IEEE Workshop on Advances in Information, Electronic and Electrical Engineering (AIEEE)\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1109/AIEEE.2017.8270530\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"2017 5th IEEE Workshop on Advances in Information, Electronic and Electrical Engineering (AIEEE)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/AIEEE.2017.8270530","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
Embedded software performance estimations at different compiler optimisation levels
Tools for performance estimation based on instruction set simulators (ISS) are mostly available and show good results. However when the need arises to choose a different platform or to estimate performance without having the ISS the developer needs all the different software suits and devices and also must be able to work with them. In this case we propose an estimation method based on physical measurement for generating performance models. This paper extends our previous work on source-code level performance estimations for microcontrollers. We compare two proposed estimation methods to find the most suitable for estimating embedded software performance for C source-code level on microcontrollers with higher compiler optimisation levels than −O0. As a result we show that both methods could be applied with some exceptions.