Michael Joosten wrote:
> 
> Unfortunately, the beast panics on me. But, at least, the frodo chip is
> recognized:
...
> trap type 2, code = 0x0, v = 0x0
> kernel program counter = 0x1ffe
> kernel: Illegal instruction trap
> pid = 0, pc = 00001FFE, ps = 2704, sfc = 0001, dfc = 0001
> Registers:
>              0        1        2        3        4        5        6        7
> dreg: 00003911 0000000C 00000000 00000000 00000000 00000000 00000402 00000000
> areg: 000A3030 0010C000 0010C040 000B9B7C 0201FC00 FE002000 0010DE6C FFEFFFFC
> 
> Kernel stack (000A2F64):
> 0A2F64:
> panic: Illegal instruction
> Stopped at      _Debugger+0x6:  unlk    a6
> db> t
> _Debugger(2704,a2fa0,96cd8,9693e,0) + 6
> db> n
> 
> Surprising, isn't it? No stack to trace... I'll recompile a different kernel

No, I was seeing this regularly almost right up until Jason
released the latest snapshot.  Make sure you have pmap.c version
1.28 and locore.s 1.67.  There may be other files that are relevant,
but I'm fairly sure they're where it was at.  (No CVS access *mumble*)

The 'no stack' is fairly normal; something has eaten the kernel stack
and it's returned to neverland, hitting a bogus opcode on the way.

> I'm now recompiling, as I forgot to include a options HP375, as I've a 400t
> here. Anyways:
> frodo0 at intio0 addr 0x81c000
> apci0 at frodo0,1: 8250-like UART
> apci1: disabled due to conflict with dca0
> apci2 at frodo0,3: 8250-like UART
> apkbd0 at frodo0,0: Apollo keyboard
> 
> It was discovered, but I think the order of apcis is worng by one. Either there
> are four of them, incl. keyboard, or three plus keyboard. But then, the
> conflicting port is the first one, not the second one, so apci0 should be
> disabled. Of course only true if apkbd0 is the first SIO, apci0 the second and
> so on in the Frodo chip.

Oops, you're completely right.  I was comparing against the 
unit number, not the subdevice (legacy from a different arrangement).
Fixed in the next version 8)  I'll reorder stuff so that the keyboard
shows up first too.

I'm going to try to work out how to tell the difference between
the 3xx and 4xx machines; it's supposed to be possible.
 
> And the compiler grinds away... let's see what happens tomorrow (good morning,
> mike ! 8-)

G'day again 8)

--
Mike Smith  *BSD hack  Unix hardware collector
The question "why are the fundamental laws of nature mathematical"
invites the trivial response "because we define as fundamental those
laws which are mathematical".  Paul Davies, _The_Mind_of_God_