> > > Are you planning to do any design work on this? You _have_ to be the best
> > > person to do it. And with all your spare time... 8-).
> > 
> > Yes, i am.  In fact, I've already done some, but a fair bit of it
> > needs to be rethought.
> 
> Do you know if the Amiga ISA bus fits your model?

It should.  "my model" for a perfect driver includes using
byte/word/short read/write macros or functions for all memory and I/O
accesses ... so the amiga's potential need to byte-swap values there
should be taken care of.

ISA DMA is pretty straightforward, in general, on all machines
(including the alpha), it's just that some of them (the alpha, maybe
the amiga) REALLY need to have special mapping functions, while others
don't.  I've yet to hear about an ISA DMA implementation that DMA's
to, say, only one byte in two in main memory, or something.  (If the
amiga's like that, i dunno about it, and hell if i know how to deal
with it!  8-)


> > the problem is, i can think of few things that i'm not "supposed to"
> > work on that have a lower priority...  So, most likely, i'll do most
> > of the work on it when i have something that i'm supposed to do but
> > don't want to...  (that's one of the reasons i did the latest rev of
> > the alpha TC code...  8-)
> 
> Huh? You have priorities for things you're not supposed to work on?

"not supposed to work on for my job," vs. things that i _am_ supposed
to work on for my job...

many NetBSD/Alpha things fall into that category (e.g. X for the TGA),
but many don't.  (also, there are other things that NetBSD/Alpha that
i'm supposed to be working on for my job, from time to time, where
those times are becoming more frequent...  8-( )


> Seriously, though, when do you think something might happen?

"sometime within the next few months" if i can swing it.  i really
can't say any better than that.  As of today, when i turned my
NetBSD/Alpha crash machine into my "real" workstation (old one had
hardware problems), i don't even have a "crash machine" any more...


chris