Message-Id:	<199704031226.OAA16009@math.amu.edu.pl>
Comments:	Authenticated sender is <grendel AT hoth DOT amu DOT edu DOT pl>
From:	"Mark Habersack" <grendel AT hoth DOT amu DOT edu DOT pl>
Organization:	PPP (Pesticide Powered Pumpkins)
To:	Eli Zaretskii <eliz AT is DOT elta DOT co DOT il>
Date:	Thu, 3 Apr 1997 14:27:05 +0100
MIME-Version:	1.0
Subject:	Re: Interrupt handlers and page locking
Reply-to:	grendel AT hoth DOT amu DOT edu DOT pl
CC:	djgpp AT delorie DOT com
References:	<Pine DOT SGI DOT 3 DOT 91 DOT 970401095017 DOT 24639D-100000 AT atmosp DOT physics DOT utoronto DOT ca>
In-reply-to:	<Pine.SUN.3.91.970402111824.23151R-100000@is>

Once upon a time (on  2 Apr 97 at 11:20) Eli Zaretskii said:

> > > 	Also, in my program, I -know- I don't have some parts of the interrupt
> > > handler code and data locked, yet I never get page fault errors. How do
> > > I force a page fault if it is possible to happen?
> > >
> > 
> > Try loading lots of TSR's, and run your program under a debugger. 
> 
> No amount of TSRs will cause your DJGPP program to page, because TSRs all
> load into conventional memory whereas the absolute majority of DJGPP
> installations use extended memory to load data and code of DJGPP programs.
With DPMI 1.0 it would be possible to allocate a single page, mark it as 
non-present or not-commited and then reference the memory area - this would 
generate a page fault. Probably (I haven't tried it) it is possible to do it 
with DPMI 0.9 as well: allocate some linear memory (__dpmi_allocate_memory), 
allocate LDT descriptor and set its base to match the linear address of the 
just allocated memory, deallocate memory (which *should* only mark the page 
as available!) and reference the segment. This **should** cause PF but may 
also generate a mere GPF (I don't think so, but it might). Again, I haven't 
tried it - it's taken off the top of my head.
==================================================
Stand straight, look me in the eye and say goodbye
Stand straight, we drifted past the point of
  reasons why.
Yesterday starts tommorow, tommorow starts today
And the problems seem to be we're picking up the
  pieces of a ricochet...

- Raw text -