Hello,
I will try to clarify a bit:
We are developping this RT acquisition protocol in C. I thought it might
be a
good idea to separate the peaces of code with a distinct functionality
into different modules. For example, we have a module that contains the
buffers and the functions that manage the buffers, and that's it : this
module does not have any RT thread. There are other modules that have to
perform some task, and these do create a RT thread. A user application
consists of a Linux part that communicates with a RT module via shared
memory and FIFO. This user module just has a FIFO handler, and does not
contain any RT thread.
The user module can call functions from the protocol modules. The call of
certain functions should have the effect that the calling environnement
blocks on them. When for example a message arrives later on, the protocol
should be
able to send some signal to the blocked function in the user module, so
that the execution of it proceeds.
Since I am not at all a programming expert and I am pretty new to RTL, it
is very well possible that I do not handle the overall problem in an
efficient way. Any comments are very welcome.
Thanks for the interest,
Alain.
On Mon, 10 Apr 2000 [EMAIL PROTECTED] wrote:
> On Mon, Apr 10, 2000 at 04:49:07PM +0200, Alain Rolle wrote:
> > Hi again,
> >
> > I am implementing a RT protocol stack that consists of several kernel
> > modules. One of these modules is in fact the interface to the application
> > layer. An application is for example an other kernel module. (No task,
> > just a module that communicates with Linux trough the FIFO device)(RTL
> > 2.2)
> > Now comes the trouble : If this application module calls a certain
> > function from the interface module, this function should be able to
> > suspend the exection of the calling module. At some further moment in
> > time, the execution of the application module should be continued, and
> > this as a consequence of an event that occurred somewhere in the protocol
> > layer.
> >
> > I know this sounds complicated, but it is possible in e.g. the RTEMS RTOS.
> > Further more, I know that an IPC like module exists, but I understand that
> > this only applies to e.g. two RT tasks defined in one module that should
> > synchronise.
> > Is there a possibility of invoking software interrupts in RTLinux ?
> >
> > I know these are a lot of questions and I appreciate it already when you
> > kept on reading up to this point.
>
> Can you clarify the problem a little bit? If the modules don't create
> RTLinux threads, then do you have interrupt handlers? Fifo handlers?
> It's quite easy to produce software interrupts in RTLinux, but I'd like
> to get a better idea of your program before offering advice.
>
>
> >
> >
> > Thanks in advance,
> >
> > Alain Rolle
> >
> >
> >
> > -- [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/
>
> --
> ---------------------------------------------------------
> Victor Yodaiken
> FSMLabs: www.fsmlabs.com www.rtlinux.com
> FSMLabs is a servicemark and a service of
> VJY Associates L.L.C, New Mexico.
>
> -- [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/
>
>
-- [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/
