Experimental Physics and
Industrial Control System

"Leo R. Dalesio" <[email protected]> · Fri, 4 Feb 2000 11:32:05 -0800

So there are several inadequacies with output records that need to be
handled:
1) We need to be able to say - read value at initialization - but not
process
2) We need a general purpose way to reread the value fromt he hardware in
the case where there are multiple maters to the hardware. This opens up to
several questions -
do we always read before write? If we write and find that someone else wrote
over that value - do we overwrite it? Do we provide some general idea in the
IOC that the IOC is either
    A) MAster and whatever it says is enforced - finding a different value
in the PLC is an error
    B) Slave - we do not do any writing in this state
    C) Multi-master - we write and then read to determine the value in the
remote.
        - what about passive records? Does this add a scan on remote change
to value?
3) We need to fix passive BO's that say UDF when DOL is not defined. (Did I
get this one right?)

This affects bo, mbbo, mbbo-direct, longout, and ao.
Does it relate to steppermotors?
Does it relate to algorithmic records with OUT fields?

I think it only relates to output records with device support. I also think
that we could omit the case for steppermotors (even though I said it would
be nice to be consistent.....)

What else does this need to turn into an action item?

    Bob

Experimental Physics and Industrial Control System

Experimental Physics and
Industrial Control System