Gilbert Saint-Flour wrote:
Edward Jaffe wrote:
However, this post serves to further underscore the validity of my
suggestion re: the REFR attribute. It is an old option that is not
currently being exploited by the operating system. It could be "added"
in an upward compatible way. (See Jim Mulder's post in this thread.) I'm
sure there are a few existing programs, erroneously marked REFR, that
deliberately modify themselves. Those programs will need to be relinked.
For everything else, it's business as usual. ...
You have my support: I also suggest that REFR be used to mean "loaded in
SP=252" under the control of a system-wide option.
I currently have to link-edit some of my REFR programs with a special utility
which rounds up the length of the load-modules to a multiple of 4K because
these programs issue PGSER PROTECT to set themselves read-only in order to
prevent accidental overlays. If these REFR programs were loaded into SP=252,
I wouldn't have to issue PGSER PROTECT or use a special link-edit procedure.
If REFR programs are automatically protected by a future operating
system release, why would you care into which subpool they are loaded? I
prefer SP 252 because it isn't fetch protected, thus allowing the
modules' storage to be "read" in any key. But, that's a different story
altogether.
--
Edward E Jaffe
Phoenix Software International, Inc
5200 W Century Blvd, Suite 800
Los Angeles, CA 90045
310-338-0400 x318
[EMAIL PROTECTED]
http://www.phoenixsoftware.com/
----------------------------------------------------------------------
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
