Re: Odd CALLPRC behavior when using RPG indicator with RTNVAL

Mark S. Waterbury wrote:

One problem with this, and with the proposed "solution" (to add
"EXTPROC(*CL:name)" to all the RPG procedure prototypes and recompile
ALL callers), is that the previous behavior that David is talking about
was _not_ just some /word-of-mouth/ "_work around_" but it was
*documented* in the _IBM manuals_, as recently as 1999, vis:

Folks:

Got an answer from IBM ... not what I was hoping for:

"6.1 is working as designed. The pre-V5R1 recommendation in the RPG
Programmer's Reference to declare a 2-byte character variable to receive a
1-byte return value was mistaken. It relied on unarchitected, then-current,
implementation details.

The change in behavior can occur only under very limited circumstances.
Among the necessary conditions are interlanguage procedure calls where the
calling and called procedures disagree about the size of the returned
value. It is likely that only the very specific scenario reported will
affect any customers.

Because an IBM recommendation prompted this issue, we are working on
suitable means to help customers determine whether they might be affected.
We will also be updating the 6.1 Memo To Users and will describe this
scenario in the Memo To Users for the next release.

Thank you for finding and reporting the mistaken reference material."