Re: Is the Program Status Data Structure Externally Defined?

Jon,

I've always been interested in this 'performance penalty' of which IBM
speaks (OK, I've not *always* been interested in it - that would make me a
little *too* geeky!).

I typically code new RPG programs from a template which includes the above
copybook, and I've never noticed any performance hit. I assume the hit would
simply be at startup, where it has to call e.g. MATPGMNM or its equivalent
to get the program library etc. The only ones I could see that might have to
be re-resolved at runtime would be ExcpData, ExcpID and FileErr, and they'd
only be updated when an error occurs (at which point there would be all
sorts of other related performance hits due to exception handling, wouldn't
there).

I admit that I haven't run any tests on this, but my experience has been
that there isn't a noticeable problem with defining a 'full' PSDS...

Rory

On Mon, Dec 6, 2010 at 4:57 PM, Jon Paris <jon.paris@xxxxxxxxxxxxxx> wrote:

I suspect that the reason the system doesn't supply one Nathan is that
there is a performance penalty if the PSDS is longer than 80 bytes.
The first 80 can be "filled in" by RPG from its own data. To complete
many of the other fields requires calls to system functions and so the
recommendation is to only define as many as you need.


Jon Paris