Hello, Paolo,
> I tried to understand why our test with Michael's "measuraments" did
> had, contrary to what he said, a lot of jitter. It turned out that the
> cause was that somebody took away the "r_cli" that is there while
> setting it up for RTAI.
Ok. I've just downloaded rtlinux 1.1 from luz.cs.nmt.edu.
In the examples/measurements/rt_process.c file:
void fun(int t) {
RTIME diff,now;
int i;
struct sample samp;
if (minimize_jitter) {
r_cli(); /* Be careful with this! The task won't be preempted by
anything else. This is probably only appropriate for small
high-priority tasks. */
}
> So be warned that to compensate jitter you must, not only schedule in
> advance substantially, but keep interrupt disabled for all the critical
> part of your task, INPUT-OUTPUT TO THE EXTERNAL WORLD INCLUDED. In fact
> in the sw test if you do not keep the outb whitin interrupts disabled
> the jitter is unchanged even if you schedule precisely. That clear it is
> easily understood that such a techniques is, for me, not acceptable for
> computer intensive real time.
Thanks for the info,
Michael.
--- [rtl] ---
To unsubscribe:
echo "unsubscribe rtl" | mail [EMAIL PROTECTED] OR
echo "unsubscribe rtl <Your_email>" | mail [EMAIL PROTECTED]
----
For more information on Real-Time Linux see:
http://www.rtlinux.org/~rtlinux/
