Philippe Gerum wrote:
On Tue, 2004-07-27 at 15:03, Marc Kleine-Budde wrote:
On Tue, Jul 27, 2004 at 02:38:01PM +0200, Gilles Chanteperdrix wrote:
Philippe Gerum wrote:
> _periodic_ nucleus watchdog that expresses time in ticks (this happens
> even if the nucleus is asked to use the timer hardware in one-shot mode;
> this is not related here), so your delay is always rounded to the next
The skin posix should operate in aperiodic mode by default (but not in
vesuvio, I agree).
I used the fusion branch.
What you need to use is fusion/testsuite/latency in order to compare
with vesuvio/rtai-testsuite/user/latency.
The numbers of testsuite/latency are actually a bit lower but still
around 70 us (compared to 30 us or so of old lxrt).
And why is there a difference between waking up a task which waits on a
periodic event compared to a aperiodic event as set up by the nanosleep
call in the other example? Please point us to the code passages, we
don't see it yet.
Jan
