Discussion:
ASCBEJST
Add Reply
Charles Mills
2017-09-29 17:11:01 UTC
Reply
Permalink
Raw Message
ASCBEJST -- which *is* documented as a programming interface -- is
documented as "elapsed job step timing." Am I confused or is it actually
"job step TCB CPU time"?

Charles

----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to ***@listserv.ua.edu with the message: INFO IBM-MAIN
Greg Dyck
2017-09-29 17:58:42 UTC
Reply
Permalink
Raw Message
Post by Charles Mills
ASCBEJST -- which *is* documented as a programming interface -- is
documented as "elapsed job step timing." Am I confused or is it actually
"job step TCB CPU time"?
The documentation is correct, ASCBEJST is *elapsed* *time*.

CPU time is accumulated in a number of different buckets in the ASSB,
diced and sliced several ways.

Regards,
Greg
Jim Mulder
2017-09-29 19:32:16 UTC
Reply
Permalink
Raw Message
Yes, it is CPU time, not elapsed time.
The comment on the field which says "ELAPSED TIME" goes all the
way back to MVS 3.8.

Jim Mulder z/OS Diagnosis, Design, Development, Test IBM Corp.
Poughkeepsie NY
Date: 09/29/2017 03:29 PM
Subject: ASCBEJST
ASCBEJST -- which *is* documented as a programming interface -- is
documented as "elapsed job step timing." Am I confused or is it actually
"job step TCB CPU time"?
Charles
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to ***@listserv.ua.edu with the message: INFO IBM-MAIN
Charles Mills
2017-09-29 20:15:44 UTC
Reply
Permalink
Raw Message
Thanks, @Jim.

In STCK format? Microseconds * 4096?

It accumulates through the entire jobstep or started task, without ever
being reset to zero until the end of the jobstep or STC?

As you can perhaps infer I have a problem that I have not quite been able to
put my finger on.

Charles


-----Original Message-----
From: IBM Mainframe Discussion List [mailto:IBM-***@LISTSERV.UA.EDU] On
Behalf Of Jim Mulder
Sent: Friday, September 29, 2017 12:33 PM
To: IBM-***@LISTSERV.UA.EDU
Subject: Re: ASCBEJST

Yes, it is CPU time, not elapsed time.
The comment on the field which says "ELAPSED TIME" goes all the way back to
MVS 3.8.

Jim Mulder z/OS Diagnosis, Design, Development, Test IBM Corp.
Poughkeepsie NY
Date: 09/29/2017 03:29 PM
Subject: ASCBEJST
ASCBEJST -- which *is* documented as a programming interface -- is
documented as "elapsed job step timing." Am I confused or is it
actually "job step TCB CPU time"?
Charles
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions, send email
to ***@listserv.ua.edu with the message: INFO IBM-MAIN

----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to ***@listserv.ua.edu with the message: INFO IBM-MAIN
Martin Packer
2017-09-30 06:56:56 UTC
Reply
Permalink
Raw Message
I would expect STCK or STCKE. Try bit 51 as 1 microsecond.

Cheers, Martin

Sent from my iPad
Post by Charles Mills
In STCK format? Microseconds * 4096?
It accumulates through the entire jobstep or started task, without ever
being reset to zero until the end of the jobstep or STC?
As you can perhaps infer I have a problem that I have not quite been able to
put my finger on.
Charles
-----Original Message-----
Behalf Of Jim Mulder
Sent: Friday, September 29, 2017 12:33 PM
Subject: Re: ASCBEJST
Yes, it is CPU time, not elapsed time.
The comment on the field which says "ELAPSED TIME" goes all the way back to
MVS 3.8.
Jim Mulder z/OS Diagnosis, Design, Development, Test IBM Corp.
Poughkeepsie NY
Date: 09/29/2017 03:29 PM
Subject: ASCBEJST
ASCBEJST -- which *is* documented as a programming interface -- is
documented as "elapsed job step timing." Am I confused or is it
actually "job step TCB CPU time"?
Charles
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions, send email
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
Unless stated otherwise above:
IBM United Kingdom Limited - Registered in England and Wales with number 741598.
Registered office: PO Box 41, North Harbour, Portsmouth, Hampshire PO6 3AU


----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to ***@listserv.ua.edu with the message: INFO IBM-MAIN
Ed Jaffe
2017-09-30 15:00:52 UTC
Reply
Permalink
Raw Message
Post by Martin Packer
I would expect STCK or STCKE. Try bit 51 as 1 microsecond.
|    LM    R14,R15,ASCBEJST        Load CPU time
|    SRDL  R14,12                  Convert to microseconds
--
Phoenix Software International
Edward E. Jaffe
831 Parkview Drive North
El Segundo, CA 90245
http://www.phoenixsoftware.com/

----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to ***@listserv.ua.edu with the message: INFO IBM-MAIN
Steve Smith
2017-09-30 21:23:20 UTC
Reply
Permalink
Raw Message
Ed... this is z/Architecture, let's bring that into the 21st century :-)
Post by Martin Packer
I would expect STCK or STCKE. Try bit 51 as 1 microsecond.
| LM R14,R15,ASCBEJST Load CPU time
| SRDL R14,12 Convert to microseconds
sas

----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to ***@listserv.ua.edu with the message: INFO IBM-MAIN
Charles Mills
2017-09-30 22:47:48 UTC
Reply
Permalink
Raw Message
I hear that one register can hold all 64 bits now! <g>

Seriously, the vague problem I was chasing is solved. Problem had nothing to do with ASCBEJST. Customer was running an older version that used clock() to retrieve cumulative CPU time. clock(), in its wisdom, returns cumulative step CPU time in microseconds as a 32-bit integer. Good for the average batch job; not so great for a long-running STC.

Charles


-----Original Message-----
From: IBM Mainframe Discussion List [mailto:IBM-***@LISTSERV.UA.EDU] On Behalf Of Steve Smith
Sent: Saturday, September 30, 2017 2:25 PM
To: IBM-***@LISTSERV.UA.EDU
Subject: Re: ASCBEJST

Ed... this is z/Architecture, let's bring that into the 21st century :-)
Post by Martin Packer
I would expect STCK or STCKE. Try bit 51 as 1 microsecond.
| LM R14,R15,ASCBEJST Load CPU time
| SRDL R14,12 Convert to microseconds
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to ***@listserv.ua.edu with the message: INFO IBM-MAIN
Greg Dyck
2017-10-05 13:59:00 UTC
Reply
Permalink
Raw Message
Post by Jim Mulder
Yes, it is CPU time, not elapsed time.
The comment on the field which says "ELAPSED TIME" goes all the
way back to MVS 3.8.
Grrrr... how could I have already forgotten this, and said what I said
earlier. I spent too much time in the jobstep timing code to not know it.

I know they say memory is the second thing to go, and I seem to have
forgotten what the first is.

Regards,
Greg

p***@gmail.com
2017-09-30 14:13:16 UTC
Reply
Permalink
Raw Message
Post by Charles Mills
ASCBEJST -- which *is* documented as a programming interface -- is
documented as "elapsed job step timing." Am I confused or is it actually
"job step TCB CPU time"?
Charles
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
I think it's bit 51 as microsecs as stated. I have this snippet of code I used:
* - TCB Time
LG R3,ASCBEJST LOAD ASCB TCB CPU TIME
LGHI R0,1000
LGHI R2,0
SRLG R3,R3,12
DLGR R2,R0
CvD R3,wrkdbl
Martin Packer
2017-10-03 15:53:37 UTC
Reply
Permalink
Raw Message
It might help Charles (and anyone else) to review the QWAC* fields in
DSNDQWAC in SDSNMACS - where the DB2 analogues of these fields are. Their
provenance is, by and large, probably simiular. QWACEJST is a field, for
example.

Cheers, Martin

Martin Packer,
zChampion, Principal Systems Investigator,
Worldwide Cloud & Systems Performance, IBM

+44-7802-245-584

email: ***@uk.ibm.com

Twitter / Facebook IDs: MartinPacker

Blog:
https://www.ibm.com/developerworks/mydeveloperworks/blogs/MartinPacker

Podcast Series (With Marna Walle): https://developer.ibm.com/tv/mpt/ or

https://itunes.apple.com/gb/podcast/mainframe-performance-topics/id1127943573?mt=2



From: Jim Mulder <***@US.IBM.COM>
To: IBM-***@LISTSERV.UA.EDU
Date: 29/09/2017 20:33
Subject: Re: ASCBEJST
Sent by: IBM Mainframe Discussion List <IBM-***@LISTSERV.UA.EDU>



Yes, it is CPU time, not elapsed time.
The comment on the field which says "ELAPSED TIME" goes all the
way back to MVS 3.8.

Jim Mulder z/OS Diagnosis, Design, Development, Test IBM Corp.
Poughkeepsie NY
Date: 09/29/2017 03:29 PM
Subject: ASCBEJST
ASCBEJST -- which *is* documented as a programming interface -- is
documented as "elapsed job step timing." Am I confused or is it actually
"job step TCB CPU time"?
Charles
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to ***@listserv.ua.edu with the message: INFO IBM-MAIN




Unless stated otherwise above:
IBM United Kingdom Limited - Registered in England and Wales with number
741598.
Registered office: PO Box 41, North Harbour, Portsmouth, Hampshire PO6 3AU

----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to ***@listserv.ua.edu with the message: INFO IBM-MAIN
Loading...