Jason Thorpe wrote:
> 
>  > Well, I just blew the evening playing with this as an idea.  I'm not
>  > quite so sure I like it, unfortunately.
> 
> ...Oh, I think you looked at the wrong one... see below :-)

Drat.  Here I was thinking the sparc stuff would be more
conceptually up-to-date.  And I even spent time working out
a means for it to create extra instances of the keyboard
driver on the fly.  Pootle.
 
> The drivers for the sun3 keyboard and mice, which are children of the
> "zsc" (zs chip) driver are in sys/dev/sun.  The mvme68k port uses
> the MI "zs" driver, but attaches 4 ttys.  See sys/arch/mvme68k/dev.

Hmm, studying as we speak.  Curses on this stupid Wren-III, my
stocks of them are dropping like flies.  Any idea where I should
go to find a disk tray to mount internal disks in a 425t, or 
am I SOL stuck with my DECmate III box?
 
> ...well, I want to avoid the "pseudo-device" approach, because with
> that approach, you can't specifiy the location of the dnkbd when it's
> a pseudo-device.

Point taken.
 
> Well, CD-ROM drives work for data (no audio yet) with the current
> hp300 SCSI code:
> 
> oscsi0 at dio0 scode 14 ipl 3: 32 bit dma, async, scsi id 7
> sd3: CD-ROM, 249596 blocks, 2048 bytes/block
>      ^^^^^^
>      I use this all the time :-)

Aha, how do you mount it?  I tried 

# mount -t cd9660 /dev/sd1c /mnt

and variants therof with no luck (various responses).  This is
with an old Matsushita (CR-503B) that I keep because I can boot
old Suns off it.  Mechanically it's very tired, but logic-wise
it seems A-OK.

>  > If you wish, I'll clean up and rename the apkbd driver
>  > (de-cruft it in particular) and post the diffs in a day or two.
> 
> Sure... That'd be fine... I don't anticipate being able to do any
> serious integration work until this weekend (at which time I want
> to rework the HIL a little bit), and probably not until Sunday
> in any event.

OK.  Is there anything I can do to help with the SCSI stuff? 
I've been off looking for data on the MB87030, but apart from
the MipsCo driver for their old machines and the (seems similar)
driver in the x68k port (does it use the MI code?) I haven't 
turned much up.  Fuji have lots of docco for their (totally
different) newer parts on their website, but nothing at all
on their old parts, and nobody around here will own up to
carrying data for them 8(
 
> Jason R. Thorpe                                       thorpej@nas.nasa.gov

--
Mike Smith  *BSD hack  Unix hardware collector
The question "why are the fundamental laws of nature mathematical"
invites the trivial response "because we define as fundamental those
laws which are mathematical".  Paul Davies, _The_Mind_of_God_