On 23 September 2012 21:28, micheal butz <[email protected]> wrote:
> I am writting a TSO command processor which I invoke IKJPARS to pass control > to a VALIDCK exit rtn Surely that's not the main reason you are calling IKJPARS... > The code in the routine runs ok which routine - the validity check exit? You know for sure that it ran? > however the problem I have is the with the > return address R14 seems to point to CVT field > CVTBSM2 offset 1DE in the CVT Why is that a problem?. Did you restore all the non linkage registers (2-13) before you branched to R14? From wherever you saved them on entry to your validity check routine? > The following is the syntax of the call CALLTSSR EP=IKJPARS,MF=(E,PPL) the > code picks up the address of the IKJPARS from CVT > and balrs to IKJPARS OK. > setting R14 to CVTBSM2 I see nothing at all like this in the CALLTSSR macro on any of my systems (z/OS 1.10 through 1.13). There are no references to CVTBSM2. But it's IKJPARS that sets this value, not CALLTSSR. > Am I suppoused to return with a register other then r14 >From the validity check routine? No - you are to return to the address passed to you in R14. Did you save and restore all registers as you do for any other program? In particular, the BSM 0,R2 requires that the caller's R2 be restored. This is just returning to the caller's AMODE. Tony H. ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [email protected] with the message: INFO IBM-MAIN
