My IEFUSI still prevents REGION=0M for all but STCs. Sort of worthless these days on 95% of the systems, but I actually still have some small monoplex LPARs that only have 1.2G - 2G, so the old region=0M and variable length getmain issue is still a valid concern for those LPARs. Actually, "prevent" is a poor wording choice, you can code REGION=0M but you end up with 256M. But I also removed code a long time ago to let people code higher regions sizes for batch (other than 0M). So I could rip out all of that checking and just force a "max pvt above the line" minus some reserved ELSQA like what is done for below the line. Maybe as I roll out z/OS 2.1.
My IEFUSI also prevents any MEMLIMIT >5G - except for STCs that code REGION=0M, they will get MEMLIMIT=NOLIMIT. It worked this way by default (excluding IEFUSI processing) until OA14391 in 2007, which changed the way it worked. At that point I updated my IEFUSI to force MEMLIMIT=NOLIMIT for an STC with REGION=0M, basically making it work "the old way". If any job / STC specifies more than 5G for MEMLIMIT, they are forced to 5G anyway from my IEFUSI. So the bottom line is I have complete control via SMFPRMxx, (with the exception of STCs running with REGION=0M) which I can update dynamically if there was ever a need to resolve a problem. I have MEMLIMIT set to 10G in SMFPRMxx. It has been this way for something like 8 years and I haven't needed to change it yet (most likely due to any STCs that would need more than 10G having REGION=0M coded). I've posted this table before from my IEFUSI source, but maybe prior to the MEMLIMIT code I added (fixed font helps when viewing): *********************************************************************** * RESULTS OF IEFUSI FOR 24-BIT AND 31-BIT REGION SIZES: * * ----------------------------------------------------- * * JCL DEFINITION REG. LIMIT REG. SIZE REGION SIZE WTO * * BELOW BELOW & LIMIT ABOVE * * ----------------------------------------------------------------- * * NO DEFINITION 8 MB DEF. SIZE-64K 256 MB DEFAULT NO * * 0 MB (JOB) ALL-512K SIZE-64K 256 MB YES * * 0 MB (STC) ALL-512K SIZE-64K ALL AVAIL YES * * >0 MB AND <=8MB 8 MB SIZE-64K 256 MB NO * * >8 MB AND LIMITB AS REQ. SIZE-64K 256 MB NO * * LIMITB AND <=256MB ALL-512K SIZE-64K 256 MB YES * * >256 AND <=1 GB ALL-512K SIZE-64K AS REQUESTED YES * * >1GB AND < 2047MB ALL-512K SIZE-64K ALL AVAIL YES * * OMVS PROCESS ALL-512K SIZE-64K ALL AVAIL NO * * DFHSIP (CICS) ALL-512K SIZE-64K AS REQUESTED DEPENDS * * >2047M JCL ERROR (FROM CONVERTER) * *********************************************************************** * RESULTS OF IEFUSI FOR 64-BIT MEMLIMIT SIZE: * * ------------------------------------------- * * JCL DEFINITION MEMLIMIT WTO * * ----------------------------------------------------------------- * * NO MEMLIMIT DEFINITION TAKEN FROM SMFPRMXX MEMLIMIT NO * * REGION=0M (JOB) + NO MEMLIMIT 5GB NO * * REGION=0M (STC) + NO MEMLIMIT NOLIMIT (16383PB) NO * * MEMLIMIT= <5GB AS REQUESTED NO * * MEMLIMIT= >=5GB 5GB NO * *********************************************************************** -- Mark Zelden - Zelden Consulting Services - z/OS, OS/390 and MVS ITIL v3 Foundation Certified mailto:[email protected] Mark's MVS Utilities: http://www.mzelden.com/mvsutil.html Systems Programming expert at http://search390.techtarget.com/ateExperts/ ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [email protected] with the message: INFO IBM-MAIN
