Sir,

Forgive me ... but a question if I may ... 

DS has inherently been used for nothing more than to define an area
that was initialized during 'housekeeping' .. 

>From the last few threads, am I to take it that DS act's like a DC 
in some circumstances ??. 

Please acknowledge and advise. 


Kind Regards

Jim Thomas
617-233-4130             (mobile)
636-294-1014                (res)
[email protected] (Email)

-----Original Message-----
From: IBM Mainframe Discussion List [mailto:[email protected]] On Behalf
Of John Gilmore
Sent: Sunday, October 30, 2011 2:11 PM
To: [email protected]
Subject: Re: filling in the holes

Paul Gilmartin's question is a curious one.

As the HLASM LR makes clear repeatedly, DS nominal values are
optional, and nothing is ever assembled into the space they map or
reserve.

The binder has an option, FILL, that permits any character, x'00'
through x'ff', to be specified for use in filling uninitialized areas
in [most] program objects.  (It is ignored for load modules and for PM
format 1 program objects.)   I use nuls, x'00', to initialize any such
areas in [other] program objects; and I have found this choice
unproblematic; but I  do not write non-refreshable code even in jest;
and my use of such areas is exiguous for that reason.

For STORAGE OBTAIN macro instructions things are a little more complicated:

o no initializations are performed for fewer than 4 Kibytes, it being
assumed that
   they may have specialized requirements that are better addressed by
the requester
   or no such requirement at all;

o for 8 Kibytes or more in a pageable private subpool, storage is
cleared to nuls;

o for 4 Kibytes or more in a pageable private subpool with BNDRY=PAGE
specified,
   storage is cleared to nuls.

Specifying CHECKZERO=YES yields information about what in fgact
happens in these and, I think, all other cases too.


The last time I looked into buffer initialization I found that the
different access methods did things quite differently, but that was
more than 10 years ago, and I cannot recite on what they do currently.
 Where it is important, e.g.,  for locate-mode I/O, my practice  is to
initialize them myself.

As this not at all exhaustive catalogue I hope makes clear, there is
no simple answer to questions like this one.

John Gilmore, Ashland, MA 01721 - USA

----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [email protected] with the message: GET IBM-MAIN INFO
Search the archives at http://bama.ua.edu/archives/ibm-main.html



-----
No virus found in this message.
Checked by AVG - www.avg.com
Version: 2012.0.1834 / Virus Database: 2092/4584 - Release Date: 10/30/11

----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [email protected] with the message: GET IBM-MAIN INFO
Search the archives at http://bama.ua.edu/archives/ibm-main.html

Reply via email to