The location of region tables



IBM OS/370, MVS, OS/390, Linux, TPF, VM/CMS, VM/ESA, VSE/ESA, z/VM, z/VSE, z/OS, z/OS.e etc...

The location of region tables

Postby Atsuko » Sun Jan 09, 2011 3:38 pm

Hello, is there anyone who can help me?
In my old understanding, page tables of the private area and segment tables are located in ELSQA.

Q1. If an application in 31-bit addressing mode which uses the high memory to store the large data is running, are the region tables used?

Q2. If Q1 is yes, where are those tables located? ELSQA?

Thanks in advance.
Atsuko
 
Posts: 2
Joined: Sun Jan 09, 2011 12:38 pm
Has thanked: 0 time
Been thanked: 0 time

Re: The location of region tables

 

Re: The location of region tables

Postby steve-myers » Sun Jan 09, 2011 7:38 pm

What earthly reason would you gain by even looking at the segment and page tables, even if you could find them? Any addresses in them are real storage addresses, not virtual addresses. The segment and page tables must be used; if they were not used your program could not access any storage. z/OS is a virtual memory operating system, not a real memory operating system!
steve-myers
Global moderator
 
Posts: 2066
Joined: Thu Jun 03, 2010 6:21 pm
Has thanked: 4 times
Been thanked: 228 times

Re: The location of region tables

Postby dick scherrer » Sun Jan 09, 2011 10:31 pm

Hello and welcome to the forum,

If you explain what you want to accomplish (i.e. the requirement), someone may have a suggestion . . .

As posted, the question appears (to me) to be more suited for tech people having a curiosity session - which is not a bad thing, but doesn't always lead to getting "something" done :)
Hope this helps,
d.sch.
User avatar
dick scherrer
Global moderator
 
Posts: 6269
Joined: Sat Jun 09, 2007 8:58 am
Has thanked: 3 times
Been thanked: 93 times

Re: The location of region tables

Postby steve-myers » Mon Jan 10, 2011 12:32 am

Dick scherrer is absolutely right. I vaguely recall looking at them for tech training for OS/VS2 Release 1 in 1972 or so. They weren't terribly useful in 1972, and are even less useful now. If I were a support person or developer shooting a possible issue (say an S0C5 ABEND, which were very common back in the days of OS/360, but should not be possible unless you have a faulty segment or page table in z/OS), then go at it. The real storage address of the current segment table is in a control register, but translating a real storage address to a virtual storage address is non-trivial.
steve-myers
Global moderator
 
Posts: 2066
Joined: Thu Jun 03, 2010 6:21 pm
Has thanked: 4 times
Been thanked: 228 times

Re: The location of region tables

Postby Atsuko » Mon Jan 10, 2011 6:17 pm

Pardon my curiosity, and thank you all for your rapid and practical suggestion^^.
Atsuko
 
Posts: 2
Joined: Sun Jan 09, 2011 12:38 pm
Has thanked: 0 time
Been thanked: 0 time

Re: The location of region tables

Postby dick scherrer » Tue Jan 11, 2011 4:08 am

You're welcome.

We have no problem with curiosity - several of us have the same affliction :)

d
User avatar
dick scherrer
Global moderator
 
Posts: 6269
Joined: Sat Jun 09, 2007 8:58 am
Has thanked: 3 times
Been thanked: 93 times


Return to Operating Systems