Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!watmath!clyde!burl!ulysses!allegra!mit-eddie!genrad!decvax!ittatc!dcdwest!sdcsvax!ncr-sd!ncrcae!sauron!campbell
From: campbell@sauron.UUCP (Mark Campbell)
Newsgroups: net.unix-wizards
Subject: VAX Unix System V.2.2 Region Problem
Message-ID: <670@sauron.UUCP>
Date: Mon, 21-Apr-86 10:32:27 EST
Article-I.D.: sauron.670
Posted: Mon Apr 21 10:32:27 1986
Date-Received: Sat, 26-Apr-86 03:11:56 EST
Distribution: net
Organization: NCR Corp., Advanced System Development, Columbia, SC
Lines: 28

Note: This request concerns VAX Unix V.2.2 kernel source.  If you don't have
it, then you're probably not going to be interested.

I need some help understanding a small piece of VAX Unix V.2.2 kernel source.
Unfortunately, I don't have a VAX (any more) to help me test any theories I
might have.

In the kernel's "procdup" routine, called from "newproc" to duplicate the
process, the routine "dupreg" is asserted to duplicate the regions of
the process.  The problem I am having concerns the udot area.  When "dupreg"
is asserted for the udot, it seems like it will (amoung other things):

	1) Cause the parent and child udot PTE to be marked as C/W.
	2) Cause the paged-referenced count to be incremented for the
	   original udot page.

At a later point in this routine, a hand-crafted udot is placed in the
child.  My questions concerns the aforementioned two points -- what happens
to the marked PTE's and the page referenced count?  Even if the child's PTE
is fixed by the hand-crafting, what about the parents?

I would appreciate any replies, either E-mailed or posted.


Disclaimer: NCR is in no way responsible for this request.
-- 

Mark Campbell    Phone: (803)-791-6697     E-Mail: !ncsu!ncrcae!sauron!campbell