On Sep  5,  5:31pm, Matthias Drochner wrote:
> I looked a bit deeper, and found the following:
> -With a "graphics" console, the machine panics early in
>   consinit(), shortly after find_kbd()->scc_kbd()->lk_divert().
>   It seems that the "scc_cons_addr" global does not get initialized.
>   I don't know if scc is supposed to be initialized this early at all,
>   so I gave up here and used a serial console.

  I haven't seen this on a 5000/260 (which is currently on a serial
console), and didn't see it on a 5000/50 with a grapics console.  I
don't know what the vintage of -current the 5000/50 was running at the
time.

> -sfbinit() panics. The order of initializations looks a bit wrong
>   (fi->fi_base is used before initialized). After sorting this,
>   autoconfiguration completed, but the box still believes in
>   a graphics terminal, so I didn't get printouts after "init" was
>   started.

  I am aware of the sfbinit() problem, but have not had the time to
verify a fix yet.

> This all looks that sad that I'm beginning to believe that
> I'm missing an important point.
> 
> Can somebody tell me what is supposed to work at all with
> -current?
> Is somebody heavily working on this stuff - what would
> explain some temporary brokeness?

  I've been trying to work on some of this (but not 'heavily'), but
things have been way too hectic here lately.  [The semester has just
started and all the students are back, and both our Unix support person
and network support person quit this summer, which means more work for
me.]

  I was wondering if anyone else had seen the sfbinit() problem;  I
am going to try to get that fixed this weekend.

Michael

-- 
Michael L. Hitch			mhitch@montana.edu
Computer Consultant
Information Technology Center
Montana State University	Bozeman, MT	USA