On Wed, Nov 6, 2013, at 06:10 AM, Michael Haberler wrote:
> 
> Am 06.11.2013 um 12:01 schrieb EBo <[email protected]>:
> 
> > This assumes that pins are either input, or output.  Have you ever 
> > played with tristate logic?  Is it even appropriate to think of tristate 
> > logic in HAL when considering a distributed system?
> 
> not played, but it is certainly possible to include IO pins in change
> detection, as well as setting them remotely.
> 
> if that makes a lot of sense in a remote UI scenario is a different matter.
> I cant even remember seeing a HAL_IO widget so far.

If you are limiting the use case for this stuff to only UIs, then tri-state
pins may or may not be important.  If it is a more general way to 
communicate between HAL instances on two computers, then we
will want tri-state to work.  It is typically used as a hand-shake mechanism,
and doesn't need to be hard real time to be useful.

For example, consider encoder index-enable (our software encoder
and all of our hardware encoder drivers do this).  The application
(typically but not neccessarily motion) sets the index enable pin.
On the next index pulse, the driver/hardware captures the position
and resets the enable pin, so the application knows that it is done.
In principle this could be done by using two separate pins at each end
and two separate signals, but using a single signal is cleaner.

-- 
  John Kasunich
  [email protected]

------------------------------------------------------------------------------
November Webinars for C, C++, Fortran Developers
Accelerate application performance with scalable programming models. Explore
techniques for threading, error checking, porting, and tuning. Get the most 
from the latest Intel processors and coprocessors. See abstracts and register
http://pubads.g.doubleclick.net/gampad/clk?id=60136231&iu=/4140/ostg.clktrk
_______________________________________________
Emc-developers mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/emc-developers

Reply via email to