Ron, > 1) I'm not seeing any CLCK/EXT/DSP entries for the logical processor within > the the trace. > 2) The Physical CPU changes on every trace entry for the 15 second period, > There are only 11 trace entries. > > So it looks like the LCP is jumping all over the place onto different PCPs > without any trace indication, other than the CP number. > > From what Tom says z/OS doesn't know about the interrupt so the LCP info is > stored somewhere off z/OS and anything on that LCP would remain undispatched > until that LCP is put back on a PCP, which may be the same one or a different > one. I would like to add that I believe there is interaction between z/OS and PR/SM in state saving. Somehow somewhere. I believe I read somewhere that with Hiperdispatch the MVS dispatcher works with PR/SM and has one WUQ (work unit q) per physical (or was that logical?) cp. That change was made generally, regardless if Hiperdispatch is used or not.
> I'm beginning to suspect this LCP was just idle. and it's not my problem. Okay, the time gaps that you mentioned: I have seen some of those ever since we got a ZIIP (one physical for about 5 different lpars). There wasn't all that much work on the ZIIP, and all systrace entries with time gap in it are due to that ZIIP being mostly idle. You may want to check your cp setup, just to make sure you're not talking about a similar situation. What I mean by cp setup is to do a D M=CPU, note the number of logical cps and compare them in a *systrace all* output with the amount of logical cps (left column) you see in systrace. You might be amazed what goes on on the logical CPs when they're not running your STC/job (which is the default for systrace unless you explicitly request all address spaces). > Is there any trace put in the MVS trace table when a slip fires? Not that I remember. > Or is there any other way to see which LCP we were running on at the time? I would think that the firing of the slip trap is visible in the system trace. *If* it was a PER trap. For the life of me, I cannot remember what the first stages after the trap hit look like. But I would know it when I see it. Do you want to provide the 11 systrace entries? > I had a look in the PSA and the control registers but didn't spot anything. Did you dump ALLPSA? Or did you take the default (which would be just the PSA you happened to be running with). Keep in mind that unless you look at an sadump, data in PSA are very volatile and don't describe the state at the time of the slip anymore. Barbara ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [email protected] with the message: INFO IBM-MAIN
