Re: ProDOS 32MB limit?

[GUDATH@EZINFO.VMSMAIL.ETHZ.CH (Henrik 'Ratte' Gudat)]

In <E60AB1.EB@serval.net.wsu.edu> mkelsey@eecs.wsu.edu writes:

> Would a rewrite of the the ProDOS block routines help overcome this 32MB
> limitation on an Apple //e?

Nope, I don't think so. The problem is that block numbers are part of the
catalog structure, and there's no space to fill the additional byte(s) for a
larger addressing space. Ok, maybe you could find an unused bit, but then all
routines that compute block addresses or otherwise deal with blocks have to be
modified accordingly. In the case of proDOS 8, this change would also make it
thru to the application software as they assume a 16 bit block range.

What really bugs me is that the extended Smartport interface support a mucho
wider address space, namely 24 bits if my memory serves me right. Fact is, the
hardware is present for supporting larger volumes, but the extended interface
is not used as far as I know. At least not in a way that would make it more
useful than the standard (16 bit) interface.
 
> Could a device driver be implemented to map more than two hard drives
> to an unused slot? (i.e. slot 5 or slot 4?)  Would ProDOS detect a 
> device driver conflict between the mouse I/O and the disk I/O if a
> physical drive was mapped logically to slot 4?

I'm not very much into device drivers (question: why am I posting this...good
point), but I'd say: under Prodos 8, no, because the disk -> slot allocation
scheme is too hard-wired in ProDOS. Under GS/OS, well, you can do about
everything.
 
- henrik

F.E.Systems
emutech@magnet.ch