On Jun 13, 12:49pm, Simon Burge wrote:
} On a 3100, I get the following with a kernel I build this morning (the
} same kernel I'm using on mona, the 5k/240).  This it just copied from
} the screen, so hopefully there aren't any typos:
} 
} > configuring pm at afc00000 interrupt number 3
} > pm0 at mainbus0(1024x864x8) (console)configuring dc at bc000000 interrupt number 2
} > dc0 at mainbus0 (kn01: intr_establish hardcoded)
} > configuring lance at b8000000 interrupt number 1
} > configuring sii at ba000000 interrupt number 0
} > sii0 at mainbus0 (kn01: intr_establish hardcoded)
} > configuring mc146818 at bd000000 interrupt number 16
} > clock0 at mainbus0
} > configuring dc at b5000000 interrupt number 4
} > configuring dc at b5200000 interrupt number 5
} > configuring  at 0 interrupt number 0
} 
} and hang.  Notice the double space and the 0's on the last line.  Just
} guessing it could be hanging trying to configure a "nothing"...

  I don't think that's where it's hanging.  I noticed that last line
quite some time ago, and it was booting fine on my 3100.  That entry
is due to a "null" entry in kn01_devs[].  There's an entry for "nvram"
that's #ifdef'd out, so the configuration loop picks up the last entry
and tries to configure it.  I don't know if the empty entry was to replace
the one that's #ifdef'd out since the loop uses a fixed counter.  Since
it wasn't causing any problem when I noticed it, I didn't try to figure
out what the proper solution was and left it alone.

  I'll see if I can try booting a current kernel on the 3100 tomorrow
if I get a chance, but I'm not sure if I will be able to get to it until
Friday or Saturday.  I've been using the 5000/25 all the time because
I've been working on the ASC driver bugs, so I haven't even had the
3100 turned on for a while.

Michael