{"title":"具有FIFO通信的RTOS硬件任务状态管理器","authors":"P. Zaykov, G. Kuzmanov, A. Molnos, K. Goossens","doi":"10.1109/ReConFig.2014.7032527","DOIUrl":null,"url":null,"abstract":"In this paper, we address the problem of improving the performance of real-time embedded Multiprocessor System-on-Chip (MPSoC). Such MPSoCs often execute data-flow applications composed of multiple tasks, which communicate through First-In-First-Out (FIFO) queues. The tasks on each processor in the MPSoC are scheduled for execution by an instance of a Real-Time Operating System (RTOS). To improve performance, we propose a Hardware Task-Status Manager (HWTSM) block that reduces the Worst Case Execution Time (WCET) of the RTOS. The HWTSM is a Molen-style Custom Computing Unit (CCU), a coprocessor that determines the execution eligibility of tasks from FIFO-filling information. Furthermore, we propose a new processor-coprocessor execution model, denoted as parallel non-blocking. In this model the HWTSM execution overlaps with the execution of RTOS and user applications. The HWTSM is integrated into the existing CompSoC platform and this entire system is prototyped on a Xilinx XC5VFX130T FPGA chip. We experiment with two types of applications running in software, i.e., synthetic and real. With the synthetic applications, the results indicate a WCET reduction of the RTOS between 1.1 and 3.0 times. For each one of the real applications - JPEG and H.264 decoders, the experimental results indicate a WCET reduction of the RTOS by 1.3 and 1.6 times, respectively. The overall system performance gain vary from from 0.9% to 13.3% for synthetic applications, from 2.3% to 4.6% for the JPEG decoder, and from 3.8% to 7.5% for the H.264 decoder.","PeriodicalId":137331,"journal":{"name":"2014 International Conference on ReConFigurable Computing and FPGAs (ReConFig14)","volume":"33 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2014-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"4","resultStr":"{\"title\":\"Hardware Task-Status Manager for an RTOS with FIFO communication\",\"authors\":\"P. Zaykov, G. Kuzmanov, A. Molnos, K. Goossens\",\"doi\":\"10.1109/ReConFig.2014.7032527\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"In this paper, we address the problem of improving the performance of real-time embedded Multiprocessor System-on-Chip (MPSoC). Such MPSoCs often execute data-flow applications composed of multiple tasks, which communicate through First-In-First-Out (FIFO) queues. The tasks on each processor in the MPSoC are scheduled for execution by an instance of a Real-Time Operating System (RTOS). To improve performance, we propose a Hardware Task-Status Manager (HWTSM) block that reduces the Worst Case Execution Time (WCET) of the RTOS. The HWTSM is a Molen-style Custom Computing Unit (CCU), a coprocessor that determines the execution eligibility of tasks from FIFO-filling information. Furthermore, we propose a new processor-coprocessor execution model, denoted as parallel non-blocking. In this model the HWTSM execution overlaps with the execution of RTOS and user applications. The HWTSM is integrated into the existing CompSoC platform and this entire system is prototyped on a Xilinx XC5VFX130T FPGA chip. We experiment with two types of applications running in software, i.e., synthetic and real. With the synthetic applications, the results indicate a WCET reduction of the RTOS between 1.1 and 3.0 times. For each one of the real applications - JPEG and H.264 decoders, the experimental results indicate a WCET reduction of the RTOS by 1.3 and 1.6 times, respectively. The overall system performance gain vary from from 0.9% to 13.3% for synthetic applications, from 2.3% to 4.6% for the JPEG decoder, and from 3.8% to 7.5% for the H.264 decoder.\",\"PeriodicalId\":137331,\"journal\":{\"name\":\"2014 International Conference on ReConFigurable Computing and FPGAs (ReConFig14)\",\"volume\":\"33 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2014-12-01\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"4\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"2014 International Conference on ReConFigurable Computing and FPGAs (ReConFig14)\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1109/ReConFig.2014.7032527\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"2014 International Conference on ReConFigurable Computing and FPGAs (ReConFig14)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/ReConFig.2014.7032527","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
Hardware Task-Status Manager for an RTOS with FIFO communication
In this paper, we address the problem of improving the performance of real-time embedded Multiprocessor System-on-Chip (MPSoC). Such MPSoCs often execute data-flow applications composed of multiple tasks, which communicate through First-In-First-Out (FIFO) queues. The tasks on each processor in the MPSoC are scheduled for execution by an instance of a Real-Time Operating System (RTOS). To improve performance, we propose a Hardware Task-Status Manager (HWTSM) block that reduces the Worst Case Execution Time (WCET) of the RTOS. The HWTSM is a Molen-style Custom Computing Unit (CCU), a coprocessor that determines the execution eligibility of tasks from FIFO-filling information. Furthermore, we propose a new processor-coprocessor execution model, denoted as parallel non-blocking. In this model the HWTSM execution overlaps with the execution of RTOS and user applications. The HWTSM is integrated into the existing CompSoC platform and this entire system is prototyped on a Xilinx XC5VFX130T FPGA chip. We experiment with two types of applications running in software, i.e., synthetic and real. With the synthetic applications, the results indicate a WCET reduction of the RTOS between 1.1 and 3.0 times. For each one of the real applications - JPEG and H.264 decoders, the experimental results indicate a WCET reduction of the RTOS by 1.3 and 1.6 times, respectively. The overall system performance gain vary from from 0.9% to 13.3% for synthetic applications, from 2.3% to 4.6% for the JPEG decoder, and from 3.8% to 7.5% for the H.264 decoder.