HP3000-L Archives

February 2010, Week 4

HP3000-L@RAVEN.UTC.EDU

Options: Use Monospaced Font
Show Text Part by Default
Show All Mail Headers

Message: [<< First] [< Prev] [Next >] [Last >>]
Topic: [<< First] [< Prev] [Next >] [Last >>]
Author: [<< First] [< Prev] [Next >] [Last >>]

Print Reply
Subject:
From:
Jeff Kell <[log in to unmask]>
Reply To:
Jeff Kell <[log in to unmask]>
Date:
Tue, 23 Feb 2010 14:38:00 -0500
Content-Type:
text/plain
Parts/Attachments:
text/plain (28 lines)
Some random thoughts (it's been ~10 years since writing any COBOL from
scratch, though I still maintain/patch/amend ours from time to time)...

On the "big" Cobol programs... yes, it used to take close to an hour to
compile some on the series ii/iii.  The cobol/ii compiler was even slower.

As for "scaling down" the bits -- yes the bounds/validate are costly in
compile time (and run time), useful for debugging, tradeoffs to consider.

Symdebug was another costly one in code size, but useful for debugging. 
Optimize is good for debugged code.

Alignment checking we always had to insert as our local code had some
legacy / switch-stubbed CM code in the background.

But one of the most useful parts for any "sizeable" COBOL program was
breaking it down into pieces and calling sections as separately-compiled
subroutines. 

Later additions in native mode (GLOBAL attributes, better program
linkage definitions and subprogram support) made this even more
manageable. 

Jeff

* To join/leave the list, search archives, change list settings, *
* etc., please visit http://raven.utc.edu/archives/hp3000-l.html *

ATOM RSS1 RSS2