Bounding the execution time of real-time tasks on modern processors

Abstract

Finding a tight estimate on the execution time of tasks in embedded hard real time systems is gaining complexity and importance. Up-to-date processors are equipped with acceleration techniques to bridge the gap between the fast core frequency of the processors and the comparably slow main memory. On the other hand, the application field of real time systems is continuously widening to more complex systems. To keep up with the evolution of processors, it is proposed to lay more weight on measurement and less on modeling. By analyzing the control flow graph the compiler uses for optimization, systematic information on how the code has to be measured can be gained. Using this information, the code can be automatically instrumented and measured to bound the execution time of the real time tasks.