This one touches my heart. As a brand new system programmer, I did the same thing. In my case I inadvertently used JCL that specified RENT,REFR even though I knew the program was neither. Testing went great. Then got my first midnight phone call. Baptism by pager.
. . JO.Skip Robinson SCE Infrastructure Technology Services Electric Dragon Team Paddler SHARE MVS Program Co-Manager 626-302-7535 Office 323-715-0595 Mobile [email protected] From: Gary DiPillo <[email protected]> To: [email protected] Date: 07/19/2011 01:03 PM Subject: Re: Re-entrant module stores into itself with no 0C4 Sent by: IBM Mainframe Discussion List <[email protected]> Skip, Thanks. I never new that. And it also explains why my customer did get the 0C4. His module, which calls my module, is APF authorized, but his calling module was not re-entrant. So the linked stub did not fail, but the dynamically called module did. I guess I have more reading to do re: re-entrant programs. I mostly work in CICS, and CICS honors the RENT declaration when it loads the program. Regards, Gary DiPillo On 07/19/2011 3:50 PM, Skip Robinson wrote: > RENT is honored--i.e. failed if violated--only if the load library is APF > authorized. > > . > . > JO.Skip Robinson > SCE Infrastructure Technology Services > Electric Dragon Team Paddler > SHARE MVS Program Co-Manager > 626-302-7535 Office > 323-715-0595 Mobile > [email protected] > > > > From: Gary DiPillo<[email protected]> > To: [email protected] > Date: 07/19/2011 12:43 PM > Subject: Re-entrant module stores into itself with no 0C4 > Sent by: IBM Mainframe Discussion List<[email protected]> > > > > I have run into a situation I have never seen before and did not think was > possible. I must be missing something, but I haven't got a clue as to > what. > > I have a batch program (A) that is re-entrant. However, statically linked > to it is a program (B) that is not re-entrant. When this program B is > called, it stores program (A)'s R13 in a save area within program (B)'s > own storage. No ABEND, no program check. I used to think this was worthy > of an 0C4. > > This program (B) issues a LOAD EP=C and stores the entry point address of > the loaded program (C) in a data area within program (B). Again, no 0C4. > > Program (B) then CALLs the loaded program (C), which is also link-edited > as re-entrant. Module (C) stores all kinds of data in data areas within > itself. Again, no 0C4. > > The three programs involved are are all Assembler and linked RN, RU, even > though two of them (B) and (C) are not re-entrant (my error). They are > all in the same load library which is the only library the the batch job > step's STEPLIB. None of the programs is in the LPA, nor is the library in > the LINKLIST, nor are the programs in any other LINKLISTed library. > > My z/OS V1R10 image is a guest running under z/VM. > > Can someone suggest how these programs that are marked as re-entrant can > be loaded and executed without error even though they are storing data > within themselves? > > TIA, > Gary DiPillo ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [email protected] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
