Time slicing is a natural product when an RTK supports
round robin multithreading and preemption. Time slicing typically
relies on software events that occur periodically (such as timer
interrupts) to provide a timing basis. When a simple logic detects
that a thread has been `hogging' the processor for more than a
certain amount of time, it invokes the preemption logic to
take control from the `hogging' thread and let the next thread
start/restart execution.
Time slicing ensures that even if threads do not voluntarily yield
to others, processor resources will still be fairly distributed
among threads of the same priority.
Copyright © 2008-10-25 by Tak Auyeung
