Real-time systems are characterized by the presence of timing constraints that a task must be completed within a given deadline. In this paper, we present a complete environment for determining best-case and worst-case execution time of a program when running on a given hardware. Our analysis technique is unique in that it allows user to annotate complex program path information and at the same time, models cache memory and pipeline accurately. This results in tight estimations even for complicated programs running on modern hardware. The technique has been implemented on a timing analysis tool - Cinderella 3, which provides retargetable back-ends for analyzing programs written in different languages and executed on different hardware. We present some experimental results of using this tool.
CITATION STYLE
Li, Y. T. S., Malik, S., & Wolfe, A. (1997). Cinderella: A retargetable environment for performance analysis of real-time software. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 1300 LNCS, pp. 1308–1315). Springer Verlag. https://doi.org/10.1007/bfb0002887
Mendeley helps you to discover research relevant for your work.