Experimental Physics and
Industrial Control System

Marty Kraimer <[email protected]> · Thu, 20 Oct 2005 07:24:41 -0500

Andrew Johnson wrote:

Tim Mooney wrote:

I just wonder how an end user is going to specify that a link
should wait or block.  Right now, I have a custom seq record that
allows the user to specify wait/nowait for each output link, but
of course this requires an extra field associated with each link.

Tim's last point is IMHO an important one which I personally think

might be part of a better solution to this whole issue of sequencing

and synchronization of parallel links. Where we want to give users

control over the seq/sync of sets of links there needs to be

information supplied to configure the seq/sync process, but this

information is not usually meaningful where individual links are used.

It really only seems to make sense for this level of control to be

provided for the arrays of inputs to the sub and calc records and for

the sequence record, and there we might want to provide additional and

more complex sequencing between the input and output links. It should

become obvious from that particular example that this sequencing and

synchronization between links is really nothing to do with the link

type at all, so it shouldn't be specified through the link configuration.

- Andrew

Are you saying that we should not implement block, process, and wait

semantics as part of a link definition?

I am under the impression that the general concensus that it is a good

idea and we were deciding semantics.

Am I wrong?

Marty

Experimental Physics and Industrial Control System

Experimental Physics and
Industrial Control System