With dm_2.2 there seems to be a requirement that did not formerly exist; that
the "Press Message" and "Release Message" properties match the contents of
the .ONAM and .ZNAM fields in the target record. In addition, it seems like
the only values that will work correctly are the values 1 and 0.
As I read the documentation (which I _think_ is current) there is no
indication of this restriction. That is, the OPI manual sez (again, as I
understand it) that sending a [Press/Release] Message of 1 to a binary record
should set the .VAL field record to one; sending a 0 chould set the record .VAL
field to 0.
The record reference manual doesn't say anything that I can find about using
the contents of .ONAM and .ZNAM to calculate the record value. And it certainly
doesn't say anything restricting the values (of .ONAM or .ZNAM) to 1 and 0.
I have a message button for a record with "Reset" and "Normal" in the .ONAM
and .ZNAM fields. The Press Message and Release message properties are 1 and
0 respectively. The Message Button seems to connect to the record OK, but
I cannot change the record .VAL, and attempting to do so causes dm to issue
the error message "Message button : value out of range" for each press or
release of the button. Changing the Press and Release Messages to "Reset"
and "Normal" does not change the results.
The only way I can get this to work is to make the Press and Release Message
properties 1 and 0 and change the .ONAM and .ZNAM fields to 1 and 0 also.
This will have a _major_ impact on moving to epics3.12.2/dm2.2, especially in
the case of converting older screens and databases. We (perhaps foolishly)
used .ZNAM and .ONAM to indicate the action that the "1" and "0" state would
cause so that attached monitors would display something meaningful, or in
many cases allowed them to default which I believe entered a "NULL" string
in the field in R3.12.1 and earlier. In any case, running the older .adl
files thru the 1.6 to 2.2 conversion script does not do the correct thing
when used with PVs defined in .db files generated for older versions
of epics.
I hope that this is all yet another dumb mistake on my part, and that someone
can tell me a magic answer that will make it "all better."
Still, clueless in Berkeley,
-bill
- Navigate by Date:
- Prev:
SCO/Open Desktop Port of EPICS Stefan Hippler
- Next:
next EPICS meeting in Europe Matthias Clausen DESY -MKV2/KRYK-
- Index:
1994
1995
<1996>
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
2021
2022
2023
2024
- Navigate by Thread:
- Prev:
[Q] Dynamic link object. MURANAKA Masaki
- Next:
next EPICS meeting in Europe Matthias Clausen DESY -MKV2/KRYK-
- Index:
1994
1995
<1996>
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
2021
2022
2023
2024
|