Re: 2.6.23-rc6-mm1: BUG kmalloc-16: Object padding overwritten (sysfs?)

!MAILaRCHIVE_VOTE_RePLACE
Previous message: [thread] [date] [author]
Next message: [thread] [date] [author]
To: <ebiederm@...>
Cc: Alexey Dobriyan <adobriyan@...>, Andrew Morton <akpm@...>, <gregkh@...>, <linux-kernel@...>
Date: Thursday, September 20, 2007 - 1:25 pm

On Thu, 20 Sep 2007, Alexey Dobriyan wrote:

=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D
----
x5a
=3D3474
474
x8000000000000083

Hmmm.. A corrupted sysfs object at an offset of one word from the end of=20
the object that could never have been caught by SLAB since it does not=20
check more than 4 bytes. Does the value 0x5ea7 tell us anything? Maybe a=20
counter was incremented a couple of times from the initial value of 0x5a5a=
=20
that was put there by SLUB?

x5a
006
006
8000000000000083
a 5a =DB=F9=FB=FF....ZZZZZZZZ
b a5 kkkkkkkkkkkkkkk=A5
     =BB=BB=BB=BB=BB=BB=BB=BB       =20
     =D3^ZZZZZZ       =20

Ditto.... but here we have a freed object in the above case the object is=
=20
still in use. Done by different processes at different times.

Eric: Anything that comes to mind in sysfs?
Previous message: [thread] [date] [author]
Next message: [thread] [date] [author]

Messages in current thread:
Re: 2.6.23-rc6-mm1: BUG kmalloc-16: Object padding overwritt..., Christoph Lameter, (Wed Sep 19, 4:59 pm)
Re: 2.6.23-rc6-mm1: BUG kmalloc-16: Object padding overwritt..., Christoph Lameter, (Wed Sep 19, 5:36 pm)
Re: 2.6.23-rc6-mm1: BUG kmalloc-16: Object padding overwritt..., Christoph Lameter, (Thu Sep 20, 1:25 pm)
Re: 2.6.23-rc6-mm1: BUG kmalloc-16: Object padding overwritt..., Christoph Lameter, (Thu Sep 20, 1:37 pm)
Re: 2.6.23-rc6-mm1: BUG kmalloc-16: Object padding overwritt..., Eric W. Biederman, (Thu Sep 20, 2:54 pm)
Re: 2.6.23-rc6-mm1: BUG kmalloc-16: Object padding overwritt..., Christoph Lameter, (Thu Sep 20, 1:36 pm)