Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!rice!sun-spots-request
From: byrnes@ee.cornell.edu (Robert G. Byrnes)
Newsgroups: comp.sys.sun
Subject: watchdog reset problem on Sun-3/280 under SunOS 4.0.3
Keywords: SunOS
Message-ID: <5860@brazos.Rice.edu>
Date: 18 Mar 90 00:58:50 GMT
Sender: root@rice.edu
Organization: Sun-Spots
Lines: 33
Approved: Sun-Spots@rice.edu
X-Sun-Spots-Digest: Volume 9, Issue 85, message 5

I just upgraded one of our Sun-3/280 fileservers from SunOS 4.0.1 to
4.0.3, and have run into the following when I boot from the new kernel:

SunOS Release 4.0.3 (GENERIC) #1: Fri Mar 16 19:41:38 EST 1990
Copyright (c) 1989 by Sun Microsystems, Inc.
mem = 8192K (0x800000)
avail mem = 7168000
Ethernet address = 8:0:20:0:76:13
xyc0 at vme16d16 0xee40 vec 0x48 
xy0 at xyc0 slave 0
xy0: <Fujitsu-M2361 Eagle cyl 840 alt 2 hd 20 sec 67>
si0 at vme24d16 0x200000 vec 0x40 
st0 at si0 slave 32
zs0 at obio 0x20000 pri 3 
:
:	(a whole bunch of lines with just a single colon)
:
Watchdog Reset!

... after which the kernel halts.  The watchdog reset message occurs just
as autoconf is about to print

ie0 at obio 0xc0000 pri 3 

but it never gets this far (ie0 is the last device to be recognized).
This occurs for the generic kernel supplied on the distribution tape, a
generic kernel that I made myself, and a custom kernel ...  The old 4.0.1
kernel works fine (I am using it now until I can resolve this), and the
4.0.3 kernels work fine on diskless Sun-3/60's (these have le0 in place of
ie0).

Has anybody seen this before?  Is the ie0 driver broken?  If so, is there
a fix?