Path: utzoo!utgpu!jarvis.csri.toronto.edu!rutgers!apple!sun-barr!decwrl!ucbvax!ICSA.RICE.EDU!CSYSMAS%UCLAMVS
From: CSYSMAS%UCLAMVS@ICSA.RICE.EDU (Michael Stein)
Newsgroups: comp.lang.asm370
Subject: Re: tcb's and page faults
Message-ID: <8908041520.AA19607@brazos.rice.edu>
Date: 4 Aug 89 15:17:00 GMT
Sender: daemon@ucbvax.BERKELEY.EDU
Reply-To: IBM 370 Assembly Programming Discussion List <@rice.edu:ASM370%UCF1VM@icsa.rice.edu>
Distribution: inet
Organization: The Internet
Lines: 17

> Suppose that you have an address space with several tcb's
> and that one of these experiences a page fault. Now, is the
> other tcb's inhibited also until this p.f. is resolved or may they
> continue to execute. It may be so that /SP and /XA, /ESA works
> differently ?

The other tasks continue to run.  I would assume that all MVS
systems are the same as this is an "advertised feature" of
how page faults are handled (synchronously with the task
which faulted but independent between tasks).

Of course if another task tries to access the same page before
it's been resolved the other task will page fault too.

Remember that MVS dispatches tcb's (tasks) independently and more
than one of your tasks may be actually running at the SAME time
if there are multiple processors in the configuration.