Hello,
I'm quite pleased to announce that I got rtlinux running
smoothly here in really no time at all! (except that building
the kernel on my 486/33 does take some time.. )
Overall tho - very smooth install! I was so pleased that I told
my advisor... and he thinks that a small presentation on rtl
is in order for our weekly meeting.. Once again, going out and doing
something independantly has generated more work! :-)
So, I've read thru M. Barabanov's thesis and the white paper,
however still had a few questions, so, I hope this is the right
forum to ask 'em.
How does rtl gurantee that a process (rt or otherwise) will
give up the cpu w/in a bounded period of time? IOW - if there's an
interrupt that's to be handled by a real-time handler, how does rtl
get the currently executing prcess out of the cpu? Time slices, perhaps?
Do we just assume that the curently running process can
be removed at any time? or am I asking the wrong question?
Is the following a correct idea of how the scheduler interacts with
the kernel and various processes? I'm thinking that the
scheduler (the one I load as a module (rtl_sched)) directly controls
when the rt processes and the kernel get to run, while the linux
kernel, when running, gets to choose between running itself of some user
processes. (thus the question, how to get the kernel and/or user process
to give up the cpu...)
User Process 2
/
/
Kernel
/ \
/ \
scheduler User Process 1
|\
| \
| RealTime process 1
\
\
RealTime process 2
Thank You!! And again - great product!
Josh
--- [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/
