Sam,
Thanks for your hint. However, I would expect the final fix to this error to
look different,
because a subpool number is not meaningful in the context of DESERV FUNC=GET_G .
(please see:
http://www-01.ibm.com/support/knowledgecenter/SSLTBW_2.1.0/com.ibm.zos.v2r1.idad500/deservgetg.htm
)
So this AIF statement should probably just be restored to the HDZ1D10 level and
again just read:
AIF ('&FUNC' EQ 'GET').CHKSUB 00398300
Anyway, I still have a PMR open regarding DESERV (see APAR OA45505) and I can
try to report this issue in that context if you like.
Regards,
Klaus Stanislawiak
On Thu, 26 Feb 2015 21:58:23 -0500, Sam Golob <sbgo...@cbttape.org> wrote:
Hi Folks,
There was an error in the PDS 8.6 program which wasn't an error in PDS.
It showed up when you wanted to do a HISTORY subcommand against a
load module in a PDSE. Instead of the display, you got a S378 abend.
The real error was caused by an error in IBM's DESERV macro in SYS1.MACLIB.
John Kalinich created a corrected version of the DESERV macro, which,
when concatenated ahead of the one in SYS1.MACLIB, fixed the problem.
The error in DESERV was detected by Greg Price, in summary:
DESERV does not set DESP_SUBPOOL_FLG, which causes default subpool 0 to be
used. Subsequently the FREEMAIN fails with ABEND S378-14 because storage
is not allocated in subpool 21. The cause is a logic error in
SYS1.MACLIB(DESERV) at the z/OS 2.1 level -
The fix (from Greg Price): change to the DESERV macro
The AIF instruction that checks *(('&FUNC' EQ 'GET') **AND** ('&FUNC' EQ
'GET_G'))* should be *(('&FUNC' EQ 'GET') **OR** ('&FUNC' EQ 'GET_G'))*.
The second condition (GET_G) was added in z/OS 2.1 (FMID HDZ2210, RMID
UA71462 on SYS9).
Code change in DESERV (our own copy of it):
D AIF (('&FUNC' EQ 'GET') AND X00398320 DEL= 1 02251 02251
I .* Change AND logic to OR on AIF below DRK FEB15 00398200 INS= 3 02251 02251
I .* 00398200 02252
I AIF (('&FUNC' EQ 'GET') OR X00398320 02253
For the CBT Tape, we had to create new load modules for PDS 8.6, which were
assembled against the revised version of the DESERV macro, concatenated
ahead of SYS1.MACLIB. I have just re-posted on the Updates page of
www.cbttape.org the following files:
File 035, File 135 (load modules), and File 182, member Z035XMIT changed.
I tested the revised version of the PDS 8.6 program and it correctly
displayed a HISTORY of a load module in a PDSE.
We hope that this clears up this problem with PDS. Now we have to enter
a PMR with IBM. (Who's gonna do it?)
All the best of everything to all of you.....
Sincerely, Sam
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
