kern/132222: The latest kernel causes my machine panic!

[Robert Watson]

On Sun, 8 Mar 2009, Ethan Hsiao wrote:

> I'm using PPP. This system is always crashed after PPP dail-up.

Hi Ethan--

What's going on here is that the current thread is trying to acquire a lock 
held by a second thread (100075) that is improperly sleeping while holding the 
lock.  We would like to know two things (a) what lock is it and (b) what is 
the other thread doing that caused it to sleep -- probably it should be 
dropping the lock before sleeping.  To extract this information, use kgdb on a 
vmcore matching the panic message you're looking at (doesn't matter which as 
long as you're using the right thread id (tid)).

(a) Do a backtrace of the default thread in the vmcore using the "backtrace"
     command.  It will be operating on a lock, which we'd like to print -- most
     likely it's a mutex, in which case you can just print *mutexpointer from
     the arguments in the trace.  If you don't know how to extract this
     information, just send me the backtrace output and I can tell you how to
     do that.

(b) Type in  "tid 100075" (or whatever the tid from the specific panic you're
     using is, if not the one listed in the original message), and then
     "backtrace".

Follow up to this e-mail with that information, and my hope is we can fix this 
in pretty short order.

Thanks,

Robert N M Watson
Computer Laboratory
University of Cambridge


>
> Thanks!
>
> Regards,
> Ethan Hsiao
>
> 2009/3/6 Gavin Atkinson <gavin at freebsd.org>:
>> On Tue, 2009-03-03 at 10:11 +0800, Ethan Hsiao wrote:
>>> Hi,
>>>
>>> kern/132222 should be the same as kern/132215 (threads/132215).
>>> It is happened after patched CTM src-7.0568.
>>
>> Hi,
>>
>> To be honest, I can't actually see how you have come to the conclusion
>> that this PR (132222) is related to 132215.
>>
>> Are you using IPv6 and/or PPP?  Can you please give some more details
>> about what this system is used for?
>>
>> Thanks,
>>
>> Gaivn
>>
>