Hans Olminkhof wrote:
> My dumb question: What is a real time operating system?
Not a dumb question at all. In fact, AFAIK there is no single
universally accepted answer to this question.
Generally the implication is that an RTOS can guarantee that interrupt latency
and task scheduling will meet externally imposed hard real-time criteria. For
instance, it might be the case that the system has to respond to an input
contact closure (such as a soft limit switch [*]) within 100 microseconds to
turn off a motor and apply a brake. It is not acceptable for this system to
sometimes take 150 microseconds because the user interface is busy-waiting
for the mouse to be released.
Generally the way this is done is by use of priority-based preemptive
scheduling, and by prohibiting or strictly limiting the ability of tasks
to disable interrupts.
Eric
[*] In safety-critical systems, it is insufficient to have soft-limit
switches that require software intervention to operate. Hard-limit switches
directly wired to the controls are required. The usual criteria is that the
soft-limit switch is positioned to make sure that the system performs the
intended task, and the hard-limit switch is positioned such that it is
triggered before the system would cause injury to an operator.
A secondary objective of preventing damage to the machine might be handled
by either the soft or hard limit switch.
Received on Sat Apr 17 1999 - 19:42:37 BST
This archive was generated by hypermail 2.3.0
: Fri Oct 10 2014 - 23:31:43 BST