Experimental Physics and
Industrial Control System

Noboru Yamamoto <[email protected]> · Fri, 21 Sep 2001 08:18:33 +0900

Hi,

Frank Lenkszus wrote:
> 
> The 3.13 Record Reference manual pages for the SubArray record state that
> the field MALM is the "Maximum Number of Elements in the Sub-array". This is
> not true. If I wish to extract one element and only one element from a
> waveform record, and therefore the "Maximum Number of Elements in the Sub-array"
> is "1" the record support forces the INDX field to "0";i.e, I can never
> access any element greater than "0" From the record support code for  the
> Sub-Array record:
> 
>         if (psa->indx >= psa->malm)
>         {
>            psa->indx = psa->malm - 1;
>         }
> 
> The consequences of this is if one is required to blow out each and every
> element of an array into a separate process variable (Don't ask why - it's
> imposed on me by other considerations) MALM must be set to the input array size,
> not the "Maximum Sub-Array Size" which in this case is "1".  Since the record
> support allocates memory based on MALM, breaking up an N element array into N
> process variables means NxN memory locations will be allocated, even though only
> N elements will be used!  Not a problem for small arrays but when larger arrays
> are involved the memory consumption seems excessive.
> 
> If this has been covered before, please excuse the noise.
We have developed a compact Subarray record to solve this problem. This 
record requires a minor modification to the DB access library, though.

> 
> Frank
>

Noboru Yamamoto
KEKB control group
KEK, JAPAN

Experimental Physics and Industrial Control System

Experimental Physics and
Industrial Control System