-----Original Message-----
From: IBM Mainframe Discussion List [mailto:[EMAIL PROTECTED] On
Behalf Of Jim Mulder
Sent: Thursday, November 08, 2007 6:07 PM
To: [email protected]
Subject: Re: CSA 'above the bar'

IBM Mainframe Discussion List <[email protected]> wrote on 11/08/2007
06:23:58 PM:
<SNIP>
 Actually, no.  We are talking about virtual addressability, and
x'000000' is certainly addressable as a 24-bit virtual address in every
address space, and addresses the PSA for the processor on which the code
is executing.

  Furthermore, absolute frame 0 is also addressable by using the SQA
virtual address of the PSA for the processor on which the code is
executing.  This will reverse prefix to absolute 0.
As of z/OS 1.5, if there was more than one CPU available at IPL time,
the SQA virtual address of the a PSA will be a 31-bit ESQA address.
Otherwise, it will be a 24-bit SQA address.

  As of MVS/XA, we always use a non-zero prefix for each online CPU, so
practically speaking, MVS does not use frame absolute 0 for anything
other than IPL processing and SADMP processing (SADMP only uses one CPU
and uses a prefix of zero). 
<SNIP>

You did say dedicated. And it certainly appears to me to be both
dedicated and reserved (by architectural definition).  

In the MVS world, if a Problem State program attempts to modify 00000xxx
(where x is 0-512 decimal and regardless of the content of the current
base register) and LAP is on... So it is not truly available (except to
the SCP). Otherwise, as I recall [MVS environments], that page is Key0
non-fetch protected.

[The above is only for non-Z architected machines. I honestly haven't
read the requisite chapters in the new PoOP.]

Regards,
Steve Thompson

-- All opinions expressed by me are my own and may not necessarily
reflect those of my employer. --

----------------------------------------------------------------------
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

Reply via email to