Experimental Physics and
Industrial Control System

From: Johnson, Andrew N. <anj at anl.gov>
Sent: Friday, June 18, 2021 11:03 AM
To: Wang, Andrew <wang126 at llnl.gov>
Cc: EPICS tech-talk <tech-talk at aps.anl.gov>
Subject: Re: PINI field for lsi record doesn't seem to work?

On Jun 18, 2021, at 12:58 PM, Wang, Andrew <wang126 at llnl.gov> wrote:

 

Actually, I just realized I’m using EPICS 3. Let me try it on EPICS 7.

 

From: Johnson, Andrew N. <anj at anl.gov> 
Sent: Friday, June 18, 2021 10:44 AM
To: Wang, Andrew <wang126 at llnl.gov>
Cc: EPICS tech-talk <tech-talk at aps.anl.gov>
Subject: Re: PINI field for lsi record doesn't seem to work?

 

Hi Andy, 

 

On Jun 18, 2021, at 12:31 PM, Wang, Andrew via Tech-talk <tech-talk at aps.anl.gov> wrote:

 

I have a simple lsi record shown below.

 

record(lsi, “joke_record”) 

{

              field(SIZV,  128)

              field(VAL,  "If you read this, tell me a funny joke")

              field(PINI, “YES”)

}

 

When I start up the IOC and do a caget on “joke_record”, I do not get the initial value. Feel like there is something very simple I forgot…

 

The VAL field of an lsi record is an array, which gets allocated at record initialization time. That happens during iocInit, which is some time after the .db file has been loaded so there’s nowhere for the IOC to actually store your string. The PINI field is irrelevant to that.

 

The simplest way to initialize an lsi record’s VAL field is to set its INP field like this:

 

record(lsi, "lsi") {

  field(SIZV, 80)

  field(INP, ["This is a long string with more than 40 characters"])

}

 

The square brackets around the string literal ensure that the IOC doesn’t assume that what you’re giving it is a PV name. Without them the IOC would make the INP field a link to a record called “This”, either locally or over CA.

 

There is no need to set PINI, the initialization of the VAL field will happen during iocInit.

 

HTH,

 

- Andrew

 

-- 

Complexity comes for free, simplicity you have to work for.