Subject: Re: Floating point in the kernel
To: Curt Sampson <firstname.lastname@example.org>
From: Bill Sommerfeld <email@example.com>
Date: 09/20/1998 09:58:48
> Well, not exactly. I don't see why a hard real-time system can't
> say `nobody gets more than 80% of the CPU, ever.' It's the equivilant
> of putting a slower CPU in the machine.
JFW already answered this, but I'll throw in another two bits..
Hard real time is about guaranteed responses to events within a
specified time period... correctness depends not only on computing the
right answer, but also on computing it *before* (e.g.) the robot arm
punches through the wall and kills someone...
The real question is:
Ok, with that scheme, nobody gets more than 80% of the CPU over *what
timescale*? 1s? 5s? 10ms? 100us?
That will provide a bound on exactly *how* hard the hard real-time
properties of the system are, and if those don't match your
requirements, it's time to find a new OS..