At 12:02 -0400 on 04/13/2014, you wrote about Re: Testing (was
'Tesing') A SubSystem Initialization Routi:%0D
I'm not quite sure what Robert Rosenberg means by a 'static location',
but a sequence of directed loads to tghe same reused storage obtained
once dynamically would meet your needs.
John Gilmore, Ashland, MA 01721 - USA
What I was attempting to say was that when you initially set up the
SubSystem the Initialization routine is in memory at location X. This
address is remembered and stored into the control block for that SS.
There is no way to replace the Init routine with another - Even
loading another routine into the LPA will not update the address in
the control block (even though a search of the LPA will return the
address of the new copy).
My idea was that the Init Routines that is referenced by the SS
control block does NOT contain the actual Init Code but only a LOAD
of it followed by a Branch to the LOAD'ed copy followed by a DELETE.
I assume that the ABEND will reduce the load count from 1 to 0 so the
next LOAD will locate the replacement copy of the corrected routine
(which was placed into the LPA via a CSVDLYPA routine call). This
will overcome the inability to associate a new Init Routine with the
SS Init Interface since it will always be the same routine at the
same location. That code is a Top-Hat/Glue to the latest version of
the routine.
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
