Experimental Physics and
Industrial Control System

[email protected] · Fri, 4 Nov 1994 10:54:20 -0700 (MST)

Marty is right in that it is the device support layer and not the driver that
determines how you specify CAMAC addresses.  This is true of both the CEBAF and
the LANL drivers.

We (the PSR Upgrade Team) would cast our vote in favor of proliferating DTYPs
however.  The advantage of this approach is that it cuts down on data entry
errors when you are building your system.  It also has the added advantage of
documenting (in the Capfast diagrams, for example) exactly what kind of CAMAC
module your record is talking to.

The LANL CAMAC device-support layer is being developed to support both methods.
There is a set of "generic" device support routines for ai, ao, bi, bo, li, lo,
mbbi, and mbbo records which are globally accessible.  Device support for other
DTYP's can call these routines and pass them the parms that would normally have
been passed in the param field (note that this method does not eliminate the
param field, it only cuts down on the overloading).

For demonstration purposes, I would vote for overloading the param field, since
this makes it easier to understand what is going on (it also requires no
additional device-support coding).

					Eric Bjorklund

Experimental Physics and Industrial Control System

Experimental Physics and
Industrial Control System