Experimental Physics and
Industrial Control System

Andrew Johnson <[email protected]> · Mon, 30 Apr 2018 14:32:17 -0500

Hi Mike,

On 04/30/2018 02:20 PM, Michael Westfall wrote:
> Thanks. We did a 'make distclean' in both the support module and the
> IOC, then recompile both, but error persists. I suspect a bug in the
> .dbd file for the custom record.

I have a vague memory that many years ago Gemini had some custom record
types (CAD/CAR/SIR, or maybe a predecessor of one of them) at least one
of which had the ability to vary the number of fields that it supported,
possibly to save memory in the IOC. I think the record type names used
had a number to indicate how many parameters the instance supported.

This isn't related to one of them is it? If so the check we added may
prevent the record type from being used, or require you to build
multiple record types with different sizes; if you can give a bit more
information about your problem we might be able to help...

- Andrew

> On Mon, Apr 30, 2018 at 11:06 AM, Mark Rivers <[email protected]
> <mailto:[email protected]>> wrote:
> 
>     Hi Mike,
> 
> 
>     My first guess would be that the record.dbd and record.h files are
>     inconsistent.  Try doing "make uninstall" in the module with the
>     record and the module with the IOC.
> 
> 
>     Mark
> 
> 
> 
>     ________________________________
>     From: [email protected]
>     <mailto:[email protected]>
>     <[email protected]
>     <mailto:[email protected]>> on behalf of Michael
>     Westfall <[email protected] <mailto:[email protected]>>
>     Sent: Monday, April 30, 2018 8:54 AM
>     To: [email protected] <mailto:[email protected]> Talk
>     Subject: Number of fields in record ssert error at runtime
> 
> 
>     What might be the cause of an assert error at runtime concerning the
>     number of fields in a record?
> 
>     This happens when the xxxRecordSizeOffset() function is called
>     during record initialization.
>     The above function is defined in the xxxRecord.h file that is
>     generated at compile time.
> 
>     The line where it fails says 'assert(pdbRecordType->no_fields==99);'
> 
>     Why would a record have a different number of fields at runtime than
>     were defined in the structure defining the record?
> 
>     This is a custom record that we are trying to use.
> 
>     Thanks for any help...
> 
>     --
> 
>     Mike Westfall
>     Control Systems Software Engineer
> 
>     [http://www.gemini.edu/images/GeminiLogo_4.5in_newSmall.png
>     <http://www.gemini.edu/images/GeminiLogo_4.5in_newSmall.png>]
> 
> 
> 
> 
> -- 
> Mike Westfall
> Control Systems Software Engineer
> 
> 

-- 
Arguing for surveillance because you have nothing to hide is no
different than making the claim, "I don't care about freedom of
speech because I have nothing to say." -- Edward Snowdon

Experimental Physics and Industrial Control System

Experimental Physics and
Industrial Control System