>>>>> "Matthias" == Matthias Scheler <tron@lyssa.owl.de> writes:

Matthias>     Hi Daniel,

Matthias> you wrote in <199409251559.QAA13460@albireo.tdb.uu.se>:
>> No, but some fast backwards engeneering tests have given me some
>> ideas I'd like to try! Just let me say: It's amazing what one can
>> acomplish when one doesn't known what one can't do! I.e. I WANT A
>> DRIVER...

Matthias> A Fastlane driver won't be easy, if you don't have
Matthias> documentation don't waste your time.

I must say I don't like such assertions, no matter how correct you
are.  It would be rather nice to know why you state such facts.  Does
the driver use a special undocumented SCSI controller chip?  Is it
impossible to drive the chip programmatically at the same time as the
DMAC is almost impossible to figure out?  If you explain your position
it'll be easier for the guy to understand why you consider it a waste
of time.

Of course, I don't say this would be an easy project.  I just don't
like the attitude when saying "Don't even bother..." to a positive
mind without explaining why.

I'd say, if he calls the SCSI chip manufacturer explaining what he aims
to do, he'll probably get databooks for free (I and Markus W have done
so in the past).  Then he just have to read the AmigaDOS driver to
understand where the chip is mapped.  If he is satisfied with programmed
I/O this is all he needs to know.  Now if he wants DMA rolling he has
to contact the manufacturers of the board and see if they like to share
info (which they often don't want by some obscure reason) or read the
driver some more.  I did this with the GVP without too big problems
although, I must admit, the DMAC is very simple and it really took some
time understanding the AmigaDOS driver.

Niklas

Niklas Hallqvist	Phone: +46-(0)31-40 75 00
Applitron Datasystem	Fax:   +46-(0)31-83 39 50
Molndalsvagen 95	Email: niklas@appli.se
S-412 63  GOTEBORG, Sweden