> From: IBM Mainframe Discussion List [mailto:[email protected]] On > Behalf Of Edward Jaffe > Sent: Wednesday, December 16, 2009 11:08 AM > To: [email protected] > Subject: Re: PSATOLD = 0 when under "PRB" !?!? > > Thompson, Steve wrote: > > So the question that I have is, under what conditions can PSATOLD be > > zero? From days gone by, this would seem to be some kind of > dispatching > > bug if this is not set prior to dispatching a program that is PRB > based. > > > > Our code uses PSATOLD=0 as a way to check whether we're in SRB mode. > <SNIP> > > We have NO SRB code in this case. NONE, NADA, ZILCH. > > But the code path taken is only done with PSATOLD = zero. > > There are two of us here who say this is impossible (I'm one). But here > we are, and we have seen this at 1.7, 1.9, and 1.10 (that I know about). > And it is only here and there. > > So there must be some other way for PSATOLD to be zero. And that's all > I'm asking: SRB does it, what else does it? Do IRBs? Because we do use > IRBs from time to time (we do use VTAM).
Simply change your code so it Abends on this path where you think PSATOLD is zero, and get a dump of the Abend. Or set a SLIP IF in this path with A=SVCD. Whatever is really happening should be obvious when you look at the system trace in the dump. Jim Mulder z/OS System Test IBM Corp. Poughkeepsie, NY ---------------------------------------------------------------------- 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
