Integrated Timing Analysis of Application and Operating Systems Code

Autor: Abhik Roychoudhury, Lee Kee Chong, Van-Thuan Pham, Sudipta Chattopadhyay, Clément Ballabriga
Přispěvatelé: National University of Singapore (NUS), Centre de Recherche en Informatique, Signal et Automatique de Lille - UMR 9189 (CRIStAL), Centrale Lille-Université de Lille-Centre National de la Recherche Scientifique (CNRS)
Jazyk: angličtina
Rok vydání: 2013
Předmět:
Zdroj: 2013 IEEE 34th Real-Time Systems Symposium (RTSS)
2013 IEEE 34th Real-Time Systems Symposium (RTSS), Dec 2013, Vancouver, France. ⟨10.1109/RTSS.2013.21⟩
RTSS
DOI: 10.1109/RTSS.2013.21⟩
Popis: Real-time embedded software often runs on a supervisory operating system software layer on top of a modern processor. Thus, to give timing guarantees on the execution time and response time of such applications, one needs to consider the timing effects of the operating system, such as system calls and interrupts - over and above modeling the timing effects of micro-architectural features such as pipeline and cache. Previous works on Worst-case Execution Time (WCET) analysis have focused on micro-architectural modeling while ignoring the operating system's timing effects. As a result, WCET analyzers only estimate the maximum un-interrupted execution time of a program. In this work, we present a framework for RTOS aware WCET analysis - where the timing effects of system calls and interrupts can be accounted for. The key observation behind our analysis is to capture the timing effects of system calls and/or interrupts, as well as their effect on the micro-architectural states, compositionally via a damage function. This damage function is then composed in a controlled fashion to result in a RTOS-aware, micro-architecture-aware timing analysis of an application. We show the use of our analysis to compute the worst-case response time for a real-life robot controller software which runs several tasks such as balancing and/or navigation on top of a real-time operating system running on a modern processor.
Databáze: OpenAIRE