Re: RPG ILE and *INLR

["Simon Coulter" <shc@xxxxxxxxxxxxxxxxx>]

Hello Joe,

You wrote:
>I found something very interesting.  With the following situation:

>1. MAIN1 calls SUBP1 (using CALLB)
>2. SUBP1 does not set on *INLR
>3. MAIN1 does not set on *INLR (does a RETURN)

>Call MAIN1 twice in a row, and SUBP1 *INZSR gets called both times.  In
>fact, the *INZSR for MAIN1 gets called each time.  It's as if the concept
>of *INLR is ignored for the main module, and when the main module
>executes a RETURN, *INLR is set on automatically.  This is very bizarre.

Not bizzare at all.  I'll bet MAIN1 is running in a *NEW activation group.
Program ends, activation group ends, activation gone, storage gone!  Call
it again and it is a new activation therefore initialization occurs.

I believe there is special support in the activation group management to
handle the case of named activation group or *CALLER and a program ending
with LR on.  LR processing doesn't fit very well with an activation
group's idea of storage persistence.   Calling a program that ended with
LR on in a named or *CALLER activation group requires that the
initialization be performed -- even though the activation group may still
hold all the storage, and its state, from the previous invocation.

Regards,
Simon Coulter.

--------------------------------------------------------------------
   FlyByNight Software         AS/400 Technical Specialists
   http://www.flybynight.com.au/

   Phone: +61 3 9419 0175   Mobile: +61 0411 091 400        /"\
   Fax:   +61 3 9419 0175   mailto: shc@flybynight.com.au   \ /
                                                             X
                 ASCII Ribbon campaign against HTML E-Mail  / \
--------------------------------------------------------------------