Subject: Re: NetBSD 2.1
To: Lars Hecking <lhecking@users.sourceforge.net>
From: Michael L. Hitch <mhitch@gemini.msu.montana.edu>
List: port-amiga
Date: 01/03/2006 12:05:32
On Tue, 3 Jan 2006, Lars Hecking wrote:

>
>  As with 2.0.2, I have no luck installing 2.1.
>
>  I tried a miniroot install onto a new drive, but get dropped into the
>  debugger. See below for the trace; any typos are mine ...
>
> ...
> Kernelized RAIDframe activated
> uvm_fault(0x214580, 0x4000, 0, 0x2) -> 0xe
>   type 8, code [mmu,,ssw]: 4a5
> trap type 8, code = 4a5, v = 4879
> pid = 0, lid = 1, pc = 0003B3E8, ps = 2300, sfc = 1, dfc = 1
> Registers:
> ...
> ...
> Kernel stack (002F5DA4):
> ...
> ...
> panic: MMU fault
> Stopped in pid 0.1 (swapper) at netbsd:cpu_Debugger 0x6: unlk a6
> db> trace
> cpu_Debugger(8,1,e6a66,2f5dc4,1a876a) + 6
> panic(1ef841,2,2f5e54,1a8ad2,8) + 8
> panictrap(8,4a5,4879,2f5edc) + 6c
> trapmmufault(8,4a5,4879,2f5edc,2606d8) + 328
> trap(8,4a5,4879) + 398
> in6_domifattach(?)
> if_attachdomain(263d50) + e
> if_attachdomain(0,5,ad303,dfffffc,0) + 24
> main(2f5fb4) + 350
> start() + 298

  OK, looking at the 2.1 INSTALL kernel, it looks like the address
0003B3E8 is in in6_domifattach(), but is not on an instruction boundary.

  Could you get some more information from DDB with:

db> x/i in6_domifattach,10

Michael