Experimental Physics and
Industrial Control System

Emmanuel Mayssat <[email protected]> · Tue, 18 Mar 2008 10:40:49 -0700

On Tue, 2008-03-18 at 13:09 +0100, Korhonen Timo wrote:
> set to the same number. The ai record could then look like this:
> 
> record(ai, "Test:RecordToProcessOnEvent1or2") {
>   field(SCAN, "Event")
>   field(EVNT, "1")   
> #EVNT is the soft event number
> 
> ...other ai fields here...
> }

The problem with this method is that you have to keep track of the event
numbers in use. I believe epics support only 256 events. If a device
driver sends the same event ~number~ than the one expected by another
driver, then you are toasted!

> 
> This way you do not have to FLNK to each record, and there are also 
> other possibilities

With FLNK you don't have the same issue

> to dispatch the events, like from a calcout record (using OEVT and 
> OOPT). Usually  this is
> much more flexible method for the applications.

Interesting. I never noticed that calcout records could be configured to
generate events.

> > record(event, "Test:Event1") {
> >   field(DTYP, "MRF Event Receiver")
> >   field(INP, "#C0 S1 @")
> >   field(SCAN, "I/O Intr")
> >   field(VAL, "1")
> >   field(FLNK, "Test:RecordToProcessOnEvent1or2")
> > }

When I red the above, I was a little confused.
event records are usually generating events, not receiving them.(source
epics record manual). The record above seems to be using a DTYP I am
unfamiliar with. Where can I find the source? Is that event driver
specific?

Cheers,

--
E

Experimental Physics and Industrial Control System

Experimental Physics and
Industrial Control System