Execution time measure method for real-time system

Discussions on more advanced topics such as monolithic vs micro-kernels, transactional memory models, and paging vs segmentation should go here. Use this forum to expand and improve the wiki!
Post Reply
yuko
Posts: 1
Joined: Mon Mar 15, 2010 3:42 pm

Execution time measure method for real-time system

Post by yuko »

Hello, everyone,

I am currently trying to implement real-time system on Xen architecture. I will support Rate monotonic or EDF scheduler on Xen kernel and execute Linux guest OS on real-time Xen kernel. To provide real-time guarantee, I will patch Preemptive RT to guest Linux OS.

Anyway, my problem is how can we accurately measure task execution time in practical environment such as Linux. There are many research paper related to worst case execution time, but almost all works are targeting Hard real-time and impractical.
Anyone knows about practical methods to measure execution time of task in Linux environment?

And I found there are very few RTOS to support RM and EDF. Many commercial RTOS product such as LynxOs, pSOS, Vxworks and QNX seems to not support RM and EDF, RT Linux support RM and EDF. My question is why they did not provide RM and EDF? The RM and EDF theory came from around 30 years ago and there are huge amount of paper related to RM and EDF. Anyone knows about this?


Thank you for your concerns,
Post Reply