Experimental Physics and
Industrial Control System

Andrew Johnson <[email protected]> · Wed, 17 Dec 2014 15:12:11 -0600

Hi Mark,

On 12/17/2014 01:42 PM, Mark Rivers wrote:
> Questions:
> The way I have currently written it the info field is only read
> during init_record.  This means that this feature if set at iocInit,
> and cannot be turned on and off at runtime.  Is this OK?
> 
> It seems to me that it is because:
> 
> - I don't think one can access the info tags over Channel Access so
> one would need to type a "dbpf" command to the IOC to change the
> setting.
> 
> - The info tags are not Process Passive so one would need to force
> the record to process to have it read the info tag again?
> 
> Is the info tag approach acceptable, and if so is the name CALLBACK
> OK?

Info tags are not fields, they are just a way of loading metadata for a
record from a .db file. As you said above you can't get at them through
CA, or even through dbpf — there are deliberately no iocsh commands
provided for manipulating them at runtime. You could however provide
your own command that would find the record, set the info item and also
trigger the device support to adjust its settings to match. It should
lock the record while it's doing that though.

I might suggest using some kind of namespace (e.g. asyn:CALLBACK or
asyn.CALLBACK) to distinguish asyn-specific info tag names, although
it's probably a bit late for me to suggest that now.

I don't see CALLBACK as obviously describing the functionality that
you're providing, so I would encourage you to look for a better name;
how about asyn:DUPLEX or asyn:READBACK instead?

HTH,

- Andrew
-- 
People everywhere confuse what they read in newspapers with news.
-- A. J. Liebling

Experimental Physics and Industrial Control System

Experimental Physics and
Industrial Control System