> > Should I post the sc driver somewhere?  I'd say that it's slightly
> > better than alpha quality, I still wouldn't trust it with critical
> > data.  It does DMA and I haven't had any problems with it yet tho.
> >
> 	I seem to remember a bunch of people with sc boards on VME sun3
> 	and sparc boxes who might be interested in that driver :)
> 
> > I modelled it after the si driver in that it's broken up into
> > sc_<bus>.c and sc.c under arch/sun3/dev, and an MI sunscpal.c under
> > dev/ic.  (And "modelled" is misleading - all of the 'sc' and
> > 'sunscpal' files are literally modified copies of the si and
> > ncr5380sbc files.  Much credit goes to those authors.)  I wrote an
> > sc_vme attachment but I have no idea if it works since I don't have
> > any VME hardware.
> >
> 	Would you like some - we may be able to find someone in your
> 	area with a sparc VME sun3 or sparc box...

Thanks for the offer, but I don't think I'd need a box - just willing
testers that can compile kernels :) If anyone wants to test, over the
weekend I'll tar up the sources with some instructions and put it up
somewhere.

Also note that it should be possible to write MD backends to use sc
boards in other VME systems (atari?).  Roughly, all the backend has to
do is handle DMA setup, the rest of the scsibus handling is MI.

> > A multiuser session transcript follows.  Note that the kernel booted
> > from /dev/sd0a using the NetBSD boot blocks.  I would have root on sd0
> > if the disk were only big enough! :)
> 
> 	Woah, I've never seen a NetBSD box boot and show disk size in
> 	KB before :)

Ya, I need to go surplus hunting for some more SCSI drives.  All of my
decent sized ones are busy in other NetBSD boxes.

Matt

--
Matt Fredette
http://mit.edu/fredette/www