On May 16, 2005, at 11:42 AM, Christos Zoulas wrote:

> In article <D406CA9D-2228-4838-B7F0-86889F4FBAD2@dtcc.edu>,
> Chris Tribo  <ctribo@dtcc.edu> wrote:
>
>> This is what shows up when I tail /kern/msgbuf on my devel box which
>> is running MPACPI on a single processor machine
>>
>> pool_cache_get(PR_WAITOK) with held simple_lock 0xd88853b4 CPU 0 / 
>> usr/
>> src/sys/uvm/uvm_fault.c:903
>> pool_cache_get_paddr(c087af60,2,0,d18a7938,1) at
>> netbsd:pool_cache_get_paddr+0x177
>> namei(d18a7e04,ffffffff,0,4,0) at netbsd:namei+0x42a
>> vn_open(d18a7e04,302,180,ce3c0360,989680) at netbsd:vn_open+0x6d
>> coredump(ce3cbe70,c08151c0,585,202,c0815074) at netbsd:coredump+0xe0
>> sigexit(ce3cbe70,b,4002,b,0) at netbsd:sigexit+0x75
>> postsig(b,d18a7f54,0,1,1) at netbsd:postsig+0x2a2
>> trap() at netbsd:trap+0x3e8
>> --- trap (number 6) ---
>> 0x8049439:
>> Stopped in pid 24825.1 (tail) at netbsd:cpu_Debugger+0x4: leave
>>
>> Doing this on a physically multiprocessor box hangs the machine.
>>
>
> Can you do it with a LOCKDEBUG kernel?
>

It is a debug,diagnostic,lockdebug kernel already.

Chris