RE: MBBI Value Type

[Mark Rivers <rivers@cars.uchicago.edu>]

The data type of the ZRVL field is ULONG:

 

This is from mbbiRecord.dbd for 3.14.12.4:

 

        field(ZRVL,DBF_ULONG) {

                prompt("Zero Value")

                promptgroup(GUI_BITS1)

                special(SPC_MOD)

                pp(TRUE)

                base(HEX)

                interest(1)

        }

 

However, because Channel Access does not support a ULONG datatype, it promotes it to double.

 

Mark

 

 

From: tech-talk-bounces@aps.anl.gov [mailto:tech-talk-bounces@aps.anl.gov] On Behalf Of Elmer Pensack
Sent: Tuesday, November 25, 2014 12:14 PM
To: tech-talk@aps.anl.gov
Subject: MBBI Value Type

 

In working with the value types of a MBBI, I noticed that the native type of the mbbi values is DBF_DOUBLE even though the documentation states ULONG. When a connection handler completes a connection with those attributes it returns the DOUBLE type.  Also, when a caget with the –D option is executed the following is displayed.

 

>  caget –d DBR_LONG <PV NAME>.ZRVL

 

<PV NAME>.ZRVL

     Native Data Type:     DBF_DOUBLE

     Request Type:         DBR_LONG

     Element Count:        1

     Value:                0

 

However a caput with a double value does get truncated during the put.  Which makes sense as these values should be integers. So why is the native type a double?  Currently I am basing display of data by the EPICS type.  So this data is shows as a decimal value because it comes back as a DOUBLE.

We can code for this configuration but I was just curious as why the native type for MBBI values was set to DOUBLE.

 

Thanks