RE: 16 Mb limit - possibly expanded in future?

Check out these:

D TS_malloc PR * ExtProc('_C_TS_malloc64')
D size 20U 0 value

D TS_realloc PR * ExtProc('_C_TS_realloc')
D ptr * value
D size 10U 0 value

D TS_free PR ExtProc('_C_TS_free')
D ptr * value

Just remember if you allocate it, you must clean it up afterwards



-----Original Message-----
From: midrange-l-bounces@xxxxxxxxxxxx
[mailto:midrange-l-bounces@xxxxxxxxxxxx] On Behalf Of Dennis Lovelady
Sent: 17 December 2009 11:02
To: 'RPG programming on the IBM i / System i'; 'Midrange Systems Technical
Discussion'
Subject: RE: 16 Mb limit - possibly expanded in future?

With the new version of Mq the maximum message size is 100 mb.

However - even with the highly appreciated extension of maximum
variable size in RPG - the maximum size in RPGLE is 16mb. Even the
%ALLOC statement signals 16 meg as the maximum size it can allocate to
a pointer.

Is it likely that IBM will extend the options beyond the 16mb limit in
a coming release?

Is there any work-around available?

I cannot speak for IBM's plans, but I can say that this is not an RPG-only
limitation. It is a limitation of 24-bit addressing that's implemented in
many (most?) objects throughout the i architecture. For example, this is
the maximum size of a user space also. Without knowing the construction of
a pointer, I'd be willing to bet small dollars that the pointer data type
also is comprised of a 24-bit address, limiting the value of such an
increase from a practicability standpoint.

Because of this, I have copied your message to midrange-l for further
discussion. (David may want to slap my hands for this; let's find out.)

Dennis Lovelady
http://www.linkedin.com/in/dennislovelady
--
"If I were two-faced, would I be wearing this one?"
-- Abraham Lincoln