On Thu, 7 May 2015 08:36:20 -0400, Shmuel Metz (Seymour J.) wrote:
>
>I've heard that there was an APAR because it didn't have an
>eye-catcher. If so, the eye catcher disappeared when standards
>changed.
> 
The same collection of lore avers that IEFBR14 was given a special
dispensation to deviate from standards by not saving registers on
entry and restoring them before return.  I suppose this might make
a small difference to someone reading a dump.

On Thu, 7 May 2015 08:39:53 -0400, Shmuel Metz (Seymour J.) wrote:
>
>>But it's inappropriate that the OS create a new data set when it
>>knows that it will be deleted immediately, never used.
>
>That's not clear. If you have a valid jobstep with IEFBR14 and change
>the program name, would it be appropriate to have an error message or
>is the appropriate behavior that you get the same DD processing
>regardless of program name?
>
One might make the same argument about deleting migrated data sets
without recall because PGM=IEFBR14.  IBM employees have stated here
that the decision is based solely on the name, never on the provenance.

In days of yore, when TSO DELETE recalled migrated data sets before
deleting them, I coded in scripts HDELETE; DELETE -- no need to check
status.  DELETE is smarter now.  Is the optimization in TSO or in
IDCAMS?

-- gil

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

Reply via email to