[MiNT] Re[2]: 68030 MMU crash course?

["George (HOL, IMAP)" <ggn@hol.gr>]

Hi all,

  Thanks for the  overwhelming  amount  of replies! The article pointed by Mikro
seems to be easy to understand. I'll try some experiments when I have some time.
But  since  one  of  you  (Xavier) asked me to elaborate what I'm trying to do I
thought I might as well do that! (I have nothing to hide anyway :)

  A  few  months  back a very small group of people decided to start patching ST
games  so  that  they  can  be  ran  off  hard drives and furthermore add Falcon
compatibility.

  As you can  imagine  we  run into pretty weird problems at points, one of them
being  the  shadowing  of  the  YM2149  registers  on the ST (not present on the
Falcon). The course of action we're following now is to patch the offending code
and  get on with it, but this can prove tiresome and repetitive. Then we thought
"Wait  a minute, if we can make the MMU transparently map areas like that to the
Falcon registers, we won't have to patch anything at all!"

  So  I began researching on the MMU. First stop was a program called Backwards,
which,  as you most will know, makes a lot of games Falcon compatible. It messes
around  with  the  MMU  so I disassembled that in hopes of understanding what it
did.  When  it  became  apparent  that  I  wouldn't  understand  how  the MMU is
programmed from that example, I posted my question here :)

  Anyway,  hope  you  didn't  all  go  to  sleep!  You can check our progress at
http://www.tphf.karoo.net/dbug/falcon.htm. Any comments or criticism is welcome!

(BTW:  I didn't want to bore you with all this in the first place, that's why my
first post was rather Laconic!)

-- 
Best regards,
 George Nakos

[From the operation manual for the CI-300 Dot Matrix Line Printer, made
in Japan]:

The excellent output machine of MODEL CI-300 as extraordinary DOT MATRIX
LINE PRINTER, built in two MICRO-PROCESSORs as well as EAROM, is featured by
permitting wonderful co-existence such as; "high quality against low cost,"
"diversified functions with compact design," "flexibility in accessibleness
and durability of approx. 2000,000,00 Dot/Head," "being sophisticated in
mechanism but possibly agile operating under noises being extremely
suppressed" etc.

And as a matter of course, the final goal is just simply to help achieve
"super shuttle diplomacy" between cool data, perhaps earned by HOST
COMPUTER, and warm heart of human being.