On Thu, 20 Aug 2009 13:57:43 -0500, McKown, John <jmck...@healthmarkets.com> wrote:
<snip> >But there must be some way for the dispatcher to somehow know that CPU#n is a zIIP or a zAAP as opposed to a general CP. Otherwise how would the dispatcher know not to dispatch general work onto that CPU? I would guess that there is something in the PCCA(?) that says "this is a zIIP" or "this is a zAAP". Which I guess would be determined via some instruction at NIP time? Just a wild guess on my part. Hum. If that is true along with what you said, then the easy way to dispatch general work onto a zIIP or zAAP would be to "zap" that flag so that the dispatcher would think the zIIP or zAAP was a general CP. > >-- Well, now. there is something. The PCCAATTR field. x'04' is a zIIP. x'01' seems to be a zAAP (name PCCAIFA where IFA was he original name for a zAAP). Anybody want to try zapping that on a sandbox z/OS to see what happens? I don't have a zIIP or zAAP. -- John ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html