On Sep 24, 2006, at 1:48 PM, Nino Dehne wrote:

> On Sun, Sep 24, 2006 at 12:27:00PM +0100, Rui Paulo wrote:
>>
>> On Sep 24, 2006, at 9:18 AM, Nino Dehne wrote:
>>
>>> Hi,
>>>
>>> on a recent -current using a GENERIC kernel, ifconfig panics when
>>> doing
>>>
>>> # ifconfig ral0 up
>>>
>>> with (typing by hand):
>>>
>>> uvm_fault(0xcad4a700, 0, 1) -> 0xe
>>> kernel: supervisor trap page fault, code=0
>>> Stopped in pid 150.1 (ifconfig) at      netbsd:kauth_cred_ngroups
>>> +0x6:  movl 0x20(%eax),%eax
>>>
>>
>> What is the backtrace?
>
> It can't dump because I'm running via NFS root (which seems to be  
> exactly
> the problem):
>
> db> bt
> kauth_cred_ngroups( )[1] [...]+0xd [2]
> nfs_request( ) [...]+0xf7
> nfs_getattr( ) [...]+0x110
> VOP_GETATTR( ) [...]+0x2f
> firmware_open( ) [...]+0x111
> rt2661_init( ) [...]+0x571
> ieee80211_ioctl( ) [...]+0x47b
> rt2661_ioctl( ) [...]+0x4f
> in6_update_ifa( ) [...]+0x0x5ab
> in6_if_up( ) [...]+0x5b
> ifioctl( ) [...]+0x1e7
> sys_ioctl( ) [...]+0x15a
> syscall_plain() [...]+0xae
> --- syscall (number 54) ---
> 0xbbbb89f7:
> db>
>
> If you need all the hex, I can type it from the screen. I hope it's  
> useful
> anyway.
>
> Best regards,
>
> ND
>
>
> [1] I naively assume that a lot of hex codes are not much use in  
> order to save
> myself some typing. I hope that's OK.
> [2] This is always "at netbsd:" plus the function name from the  
> beginning of
> the line. Again, saving myself some typing.

Yes, the problem is NFS root.
I'm not sure of to fix this...

--
Rui Paulo