Rob Windsor wrote:
> 
> Now, it hangs upon bootup.  During bootup off of the drive, it (the kernel)
> detects vmel0 and vmes0 and stops.  Normally, it should be mounting root
> at this point.

Tada!  I am _not_ crazy after all..  My 4/110 does the exact same thing.

> I compiled in a kernel using DIAGNOSTIC, DDB, and DEBUG, and booting it up
> gives me:
> 
> [...standard kernel messages...]
> 
> vmel0 at mainbus0
> vmes0 at mainbus0
> Process (pid 1) got signal 11
> Process (pid 1) got signal 11
> Process (pid 1) got signal 11
> 
> [...repeat, 9600baud...]

Exactly what I am seeing.

> /usr/include/sys/signal.h shows 11 as:
> 
> #define SIGSEGV 11      /* segmentation violation */
> 
> Anyone know what's going on here, and what the fix is?

Don't have a fix.  I did submit a PR though. (a couple of weeks ago)
What I was able to determine is that then the kernel forks off a
new process inorder to exec /sbin/init, the forked process is getting
a SEGV _before_ ever trying to read from disk.
I've had this problem since the sun4m integration.  I don't know if the
recent fixes for compiling SUN4 only kernels might make a difference or
not. 

-Andrew