On Fri, 15 Dec 1995, Andrew Gillham wrote:

> > 
> > I'm not sure what the problem with this approach is.  We did try this at
> > one time and ran into some problem...
> > 
> > -allen
> 
> Uhmm, I can think of one.. :)  In the process of using asm to talk to
> video ram, I sorta tromped all over the memory after the video ram.
> (couldn't happen right? :) )

> but on 1MB boundaries or similar might not run into the problems of
> runaway 'jmp loop' commands.. :-)

Just avoid the video ram.  It shouldn't be that hard to avoid.  Start at 
2 MB.  Why not?  It shouldn't be too hard to use STA $xxxxxxxx or 
whatever.  The video RAM, I believe is a special case and doesn't take 
too highly to random storage.  That's why we have toolbox calls :-).  
Seriously, though, there must be something.  How many pages until you 
*do* get a page fault -- or do you ever?  Is the PMMU  MC68851 or 
somebody else's?  If I read this 68020/68030 reference manual correctly, 
the processor will never issue a page fault because only the PMMU is 
privy to the amt. of memory you have.  Ah.  Got it.  Here's the problem.  
Page faults only occur at the chip level "in a virtual memory system."  
Are we taking full advantage of those PMMU's?

Just trying to comprehend these manuals and filter out the unrelated junk.

 /---------------------------------------------------------------------\
|David A. Gatwood             And Richard Cory, one calm summer night,  |
|davagatw@mars              Went home and put a bullet through his head.|
|dgatwood@nyx.cs.du.edu              --Edwin Arlington Robinson         |
|http://mars.utm.edu/~davagatw -or- http://nox.cs.du.edu:8001/~dgatwood |
 \---------------------------------------------------------------------/