Re: [memcg BUG] unable to handle kernel NULL pointer derefence at 00000000

[Li Zefan <lizf@cn.fujitsu.com>]

KAMEZAWA Hiroyuki wrote:
> On Tue, 21 Oct 2008 14:20:27 +0800
> Li Zefan <lizf@cn.fujitsu.com> wrote:
> 
>>> BTW, "allocate all page_cgroup at boot" patch goes to Linus' git. Wow.
>>>
>> But seems this patch causes kernel panic at system boot ... (or maybe one of other
>> memcg patches?)
>>
>> I wrote down the panic manually:
>>
>> BUG: unable to handle kernel NULL pointer dereference at 00000000
>> IP: page_cgroup_zoneinfo + 0xa
>>
>> Call Trace:
>> ? mem_cgroup_charge_common + 0x17d
>> ? mem_cgroup_charge
>> ? add_to_page_cache_locked
>> ? add_to_page_cache_lru
>> ? find_or_create_page
>> ? __getblk
>> ? ext3_get_inode_loc
>> ? ext3_iget
>> ? ext3_lookup
>>
>> Tell me if you need extra information.
>>
> Hmm, it's curious but page_cgroup->page seems to be NULL...
> Could you show disassemble code of page_cgroup_zoneinfo() ?
> 

You mean this ?

000881c8 <page_cgroup_zoneinfo>:
   881c8:       55                      push   %ebp
   881c9:       8b 50 04                mov    0x4(%eax),%edx
   881cc:       8b 40 08                mov    0x8(%eax),%eax
   881cf:       89 e5                   mov    %esp,%ebp
   881d1:       5d                      pop    %ebp
   881d2:       8b 00                   mov    (%eax),%eax
   881d4:       c1 e8 1e                shr    $0x1e,%eax
   881d7:       6b c0 58                imul   $0x58,%eax,%eax
   881da:       03 42 48                add    0x48(%edx),%eax
   881dd:       c3                      ret

> BTW, this happens always ?
> 

Sometimes just freezed, and sometimes panic. It never boots successfully.

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>