Thompson, Steve (SCI TW) wrote:
<snip>
If this pervasive and well-established technique becomes unreliable, I
predict *lots* of software will break!
<snip>
Shall I give two for instances?
JES2 and JES3 exits and interfacing code?
The values of EQUates and the existence of field names are used to
determine which level of code to generate (PARTICULARLY WHEN a field has
moved from one C/B to another).
The JES developers have always done a pretty good job in this area.
Control block fields appear only when the function that needs them (or
toleration code on back releases) is present. If this wasn't true, I can
assure you that (E)JES would have encountered some *major* difficulties
over the past 16 years. And, I suspect numerous other JES interface
would have had problems as well ... including SDSF.
Not every component has an equivalent to SYSSTATE_OSREL. And, even if
such interfaces were pervasive, the release/modification boundary
granularity is insufficient to allow code to adapt to behavioral changes
that might occur when PTFs are installed. There's just no reliable way
at run time or assembly time to query the SMP/E data base!
Prior to this discussion, I would have stated unequivocally that the
_most_ reliable way to know if a particular behavior (implemented via
release boundary or APAR/PTF) is present is to test for the presence of
new control block fields that it defines (or old ones it removes). I
think it's still mostly true. But, obviously there are exceptions.
--
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
