RTLinux_sigaction gives you user space access to RT within RTLinux.
As for passing pointers -- have you looked at mbuff?
Sharing complex data structures between RT and user space is most efficiently
done by using shared memory.
On Fri, Jan 12, 2001 at 07:36:41PM -0500, Dingrong Yi wrote:
> Hi Erwin and all,
>
> Got lots of responses, thanks all for making this list so helpful and
> informative.
>
> Since both you and Tomek adviced me with LXRT under RTAI, especially your
> case is very similar to mine: have a large crosslinked object model, need
> to perform real time actions on the model (actually, touch it) and display
> those actions,... I shall spend some time to investigate LXRT. Thanks.
> Well, a few more questions:
>
> What do you mean to promote a userspace process to a Realtime task?
> Besides, do you know any function of linux that provides better resolution
> than time(0) of #include<time.h> ? and what is the header file to
> include?
>
>
> If it turn out that LXRT fails for my purpose, one alternate is to decople
> the graphics display of the model and location of the action from the
> real time action itself (through FIFO) . However, this leads to another
> issue, is it possible to use list of lists (i.e., pointer of pointers)
> with real time linux? As I can use the simplest way to model the objects:
> their center-lines, with each line contains a list of points that it
> travels through.
>
>
> Thanks in advance.
>
> Dingrong Yi
>
>
>
>
> On Sun, 14 Jan 2001, Erwin Rol wrote:
>
> >
> > You could take a look at LXRT, which is part of RTAI, that makes
> > it possible to promote a userspace thread (or process) to a Realtime
> > task. Due to the internal setup you will loose a few usec in latency,
> > but
> > i haven't seen more than 30usec latency (appart with my S3 card and
> > XFree
> > 4.0, but thats an other story). Also it will not be posible to make anny
> > system calls from the promoted threads/processes, but there is also
> > being
> > worked on.
> >
> > I use it because i have a large crosslinked object model , and need to
> > preform
> > certain actions on the model in real time ( 40 times a second ), and
> > display
> > those actions at the same time. Because the displaying and the realtime
> > actions
> > not change the model there is no need for locking, because when you lock
> > it from
> > non realtime you of courrse loose realtime behavior in the realtime
> > task.
> >
> > So becarefull when you start to need locks on the model , you will
> > probably
> > loose realtime, when your not carefull.
> >
> > RTAI can be found at www.rtai.org
> >
> > - Erwin
> >
> >
> >
> > Dingrong Yi wrote:
> > >
> > > Hello all RTLinux users and developers,
> > >
> > > I need to schdule a procedure by a timer at every (around) 1 ms. I
> > > set the time interval of the XtAppAddTimeOut to be 1ms. But I am
> > > uncertain is the resolution of linux timer is such fine or not? Tijs
> > > suggested to modify parameter HZ of
> > > /usr/src/linux/include/asm-i386/param.h to 1024 for a 2ms timing
> > > (attached at the end). Does it mean if modify HZ to 2048 may get a 1ms
> > > resolution timing? If so, what should I do after this modification so
> > > that this modification may be effective?
> > >
> > > I am booting with RTLinux-2.0. The reason why I do this high frequency
> > > (1000 Hz) task in linux instead of RTLinux is because this task need to
> > > use lots of graphics, big memory (60M) for object modeling, etc which are
> > > cubersome to do in RTLinux.
> > >
> > > Thanks in advance.
> > >
> > > Dingrong Yi
> > >
> > > > > Hi!
> > > > >
> > > > > In RT-Linux Manifesto author writes:
> > > > >
> > > > > "Standard Linux takes up to 600 microseconds to start a handler and
> > > can
> > > > > easily be more than 20 milliseconds (20,000 microseconds) late for
> > > a
> > > > > periodic task."
> > > > >
> > > > > Is it possible to have 2 miliseconds ? I want to control very fast
> > > robot
> > > > > from rt-linux box.
> > > >
> > > > This works for x86 only:
> > > > Try using #define HZ 1024 or some other high number in
> > > > /usr/src/linux/include/asm-i386/param.h
> > > >
> > > > Usually it's 100 and that accounts for the (what I thought were 10) 20
> > > > ms.
> > > >
> > > > Note that this breaks some programs like 'top' and 'ps'
> > > >
> > > > --
> > > > Tijs van Bakel, <[EMAIL PROTECTED]>
> > >
> > > -- [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/
> >
> >
> >
>
> -- [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
Finite State Machine Labs: The RTLinux Company.
www.fsmlabs.com www.rtlinux.com
-- [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/
