Re: Re (6): installing NetBSD on a Sparcstation 2.

[Dan Oglesby <danno%serverunderground.com@localhost>]

----- Original Message -----
> From: peasthope%shaw.ca@localhost
> To: port-sparc%NetBSD.org@localhost
> Cc: peasthope%shaw.ca@localhost
> Sent: Monday, August 16, 2010 10:54:17 AM
> Subject: Re (6): installing NetBSD on a Sparcstation 2.
> Dan O. and everyone,
> 
> > So, the SS2 is being finicky with booting ...
> 
> The earliest documentation of problems installing on
> sun4* that I've found is PR 41194 by Jon Buller.
> 
> My difficulty with hdd recognition is just a blunder of
> my own, which can be explained later.
> 
> > The SS2 I have is currently running version 2.4 of the PROM
> > firmware. I am attempting to upgrade this to 2.9, since that seems
> > to be the version best supported by NetBSD and Linux.
> 
> Will be interesting to learn the outcome.
> 
> Regards, ... Peter E.
> 

The SS2 is waiting either an EPROM programmer, or for someone local to 
volunteer to reflash the EPROM for me.

The IPC appears to now have a non-working power supply.  It DOES have the last 
version of PROM installed (2.9), verified by the markings and part number on 
the EPROM.

Since I was 0 for 2, I decided to give the SS20 a try.  It successfully boots 
from the NetBSD-5.0.2 CD-ROM using the internal SCSI CD-ROM (Toshiba XM-4101), 
and installed to the internal 4.2GB SCA SCSI hard drive using the GENERIC 
kernel.  It panics when I tried using the multi-processor kernel on the reboot 
after installation.  This SS20 has dual SM71 processors, 256MB of RAM, and the 
2.25 PROM firmware.

The video on the CG6 framebuffer (SBus) appears to work for a few minutes after 
boot, then the console freezes.  I see no errors on the console, in dmesg (via 
SSH), or in /var/log/messages to indicate an issue with software.  I can SSH in 
to the box, and it appears to be running along just fine, minus the console 
issue.  After a Stop-A and a "boot disk", the system will fsck the filesystems, 
and have a working console again for a few minutes.

Something else I noticed, and I don't know if this is because I'm using the 
GENERIC kernel, but the CPU identification doesn't appear correct in dmesg:

cpu0 at mainbus0: mid 8: TMS390Z50 v0 or TMS390Z55 @ 75 MHz, on-chip FPU
cpu0: physical 20K instruction (64 b/l), 16K data (32 b/l), 1024K external (32 
b/l): cache enabled
cpu at mainbus0 not configured

So, there's what I've come up with so far.  I have some SS5 workstations with 
at least two different kind of CPUs (70Mhz, 110Mhz?) to try next.  I'll have to 
move the floppy from the IPC into one of the SS5s or the SS20 to test the 
5.1.3_RC3 floppy images as well.  I wish my sun4c hardware was up to the task, 
but time has apparently caught up with some of my hardware.  :-(

--Dan