[A83] Re: MirageOS remembering program location


[Prev][Next][Index][Thread]

[A83] Re: MirageOS remembering program location




> Van: Henk Poley <HPoley@DDS.nl>
> 
> > Van: Dan Englender <dan@calc.org>
> > 
> > > > ...Actually, it needs to be a bit more complicated because it's
> > > > actually dependant on what page the code is actually running from,
> > > > not the mapped address.  IE. if you load RAM page 0 into 4000h,
> > > > and run some code there, it'll still crash.  The same applies for
the
> > > > Flash ROM execution protection.
> > >
> > > That means that the adress-bus part changes. But would cutting the
> > > RFSH-pin help? That pin isn't used all (might even not be connected,
> > > but then my theory isn't correct anyway), since Ti's don't have
dynamic
> > > RAM.
> > >
> > > And what is this Flash ROM execution protection?
> > 
> > ...I have no clue on the RFSH pin.
> 
> ???
> 
> The RFSH-pin seems to me like "the" way of handling this. It's high when
> the Z80 has fetched an OP-code. That way you can see if the Z80 is only
> reading data (which is approved) or executing stuff (which you don't want
> it to do).
> 
> That the hardware also needs to check if the adress-bus points to an
> 'illegal' RAM-page is only making it slightly difficult (but also
[almost]
> proves that it's external to the CPU).
> 
> Basicaly it means that you need to check A15 and A14 against the 4 banks
> with possible illegal pages. (XX?? ???? ???? ???? => A15, A14,..., A0)
> 
> If so, then "boom!"... (plainly set the RESET-pin)
> 
> So something like this:
> 
> A15-pin  AND  ProtPageBit1 -\
> 				> AND  -->  RESET-pin
> A14-pin  AND  ProtPageBit2 -/

A good page about the Z80 pin-out is:
http://www.bamafolks.com/~randy/students/embedded/Z80_intro.html

That also told me that it would be more logical to use the M1-pin (nr. 27),
since that pin should tell if the CPU is fetching an OP-code. (other
references decribe it as MI, for "maskable interrupt", but I think that's
not true).

	Henk Poley <><