Hi Johnny,

Johnny Billquist wrote:
...
>
> panic: Segv in kernel mode: pc %x addr %x
> Stack traceback :
> 0x8c01bd34: trap+0x4fc(0x8c01bdfc)
> 0x8c01bdfc: trap type=0x8 code=0x4 pc=0x801b00f6 psl=0x4
> 0x8c01bdc8: pmap_deactivate+0x1a(0x84fab100)
> ...
>
> I find it a bit interesting that it looks like it's more or less 
> directly after a fork that this happens, that it somehow gets into 
> uvm_pageout (I haven't figured out how it gets there), and that 
> process 0 is active.
> Is some kind of kernel thread created here? Or maybe activated for the 
> first time?
I don't have any disassembly of a kernel at hand, but simply looking at 
the first few
lines in pmap_deactivate() says that this routine may be called before 
everything is setup.

It was a long time since I was dealing with these routines, but they 
assumes that it is
a complete process that is deactivated.  If no user area exists this 
routine will fail.
Please do a disass of the beginning of pmap_deactivate() and see what is 
not there yet.
It may be necessary to put in a check for that.

-- Ragge