On Thursday, 07/16/2009 at 01:33 EDT, David Boyes <dbo...@sinenomine.net> 
wrote:
> > -----Original Message-----
> > From: The IBM z/VM Operating System [mailto:ib...@listserv.uark.edu] 
On
> > Behalf Of Alan Altmark
> >    : <you get the idea>
> 
> ??? Actually, I didn't understand that at all. What were you trying to 
show -- 
> the steps in the IBM SYSPROF?

I was my attempt to
(a) Define the SYSPROF's domain:
    - handling auto-reIPL diagnostics
    - loading DCSSes and CSL routines
    - activating COMDIR
    - accessing the A and D disks
    - handling ACC(NOPROF
    - setting the default SFS filepool
    - initializing GLOBALV variables
    - trigger VMLINK autolinks
    - start batch processing

(b) Define the places where an exit would be called

(c) Define an interface that gives the exit a view of conditions within 
SYSPROF and the ability to alter what SYSPROF will do.

For example, SEGMENT LOAD and RTNLOAD would be the responsibility of 
SYSPROF, but your exit can alter the list of segments or CSL libraries 
that SYSPROF will load.  If you don't want SYSPROF to perform a GLOBALV 
INIT, set IBM._FINAL_GLOBALV_INIT to 0.

The analysis needs to be done to see if things that a significant 
percentage of sysprogs do in their current exits or mods should be moved 
into the IBM code.

Alan Altmark
z/VM Development
IBM Endicott

Reply via email to