FWIW, prior to z/OS 1.12, depending on how your exit routine is written, 
you could try adding an exit routine with a different name than IEFUTL to 
the applicable exit point(s). For example, rename IEFUTL to IEFUTL1 in 
your data set and do an exit add specifying MODNAME(IEFUTL1). Both IEFUTL 
and IEFUTL1 will run. 

I would guess that it's the last one that runs, setting data into the 
parameter area(s) and/or control blocks, that wins (that's my recollection 
of how SMF processes). 

The order of exit routines shown on DISPLAY PROG,EXIT,EXITNAME=xxx is the 
order in which they will be called, if you want to experiment.

As to the questions
>1. Is it still needed to reload SMFPRMxx after doing the SET PROG=xx?

>2. Is it still needed in PROGxx to use EXIT DELETE and LPA DELETE and 
>then later again with LPA ADD and EXIT ADD? If not, can I then just 
>replace all those EXIT commands with one which is using REPLACE? 

1. No. Nor was it ever. You only reload SMFPRMxx with respect to exits 
when you are changing the definitions of which exits are to be processed. 
You do not do so when you are changing the exit routine(s) associated with 
an exit.

2.You certainly do not need to do EXIT DELETE then ADD if you are using 
REPLACE. But if you were ever doing LPA DELETE you probably ought not to 
be doing that, unless you can guarantee that no exit processing is 
underway when you do the LPA DELETE.  LPA DELETE says that you have made 
sure (in whatever way you can do that; often there is no such way) that no 
processing is currently using the module and thus the storage can be 
freed. Adding to LPA is fine (I'd even say desirable) but not necessary if 
your EXIT ADD specifies the data set in which the system is to find the 
module.

Peter Relson
z/OS Core Technology Design

----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [email protected] with the message: INFO IBM-MAIN

Reply via email to