> One outstanding issue is recovery when a realtime task goes out of control...
In embedded systems, it is common to use a watchdog for this situation. It would seem to be straightforward to construct one using a highest priority task (doing a timed wait on a semaphore), and a low priority task which periodically posts to that semaphore. If the semwait times out, the watchdog task takes appropriate action against the offending process.
Any reason not to foist it off onto user space like that?
Copyright © 2017, Eklektix, Inc.
Comments and public postings are copyrighted by their creators.
Linux is a registered trademark of Linus Torvalds