Stuart Hughes <[EMAIL PROTECTED]> writes:
> It is possible that you are getting an integer overflow in the multiply,
> RT_TICKS_PER_SEC is a big number and could oveflow so care is needed.
Maybe, although longlong_t should be long enough... Anyway, I need to
work with periods of up to about 500ms, so at least for now this is
not a big problem.
> If you can post a cut down sample that shows the same problem, it may be
> possible to spot the problem
Now it seems that I got something that looks like working. However, I
could not get the rate-monotonic scheduler to work correctly without
the debugging enabled. insmod would install it without any complaints,
but then one Linux program (that I use to send commands to another
Linux program via shared memory, which in turn activates some RT tasks
by sending commands via RT FIFOs, like in those examples) exhibits
very strange behavior. Namely, bash first complained that it could not
execute the binary. When I recompiled and ran it again, the whole
system froze. I did this a couple of times with the same results
(freezes). Later, when I uncommented the DEBUG line in the Makefile of
the rate-monotonic scheduler, and recompiled/reinstalled it,
everything got back to normal (I changed the RT FIFO number to 63 and
ignored it in my code).
>
> Regards
>
> Stuart.
Thanks,
Aleks
--- [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/
