Re: 15c8410c67 ("mm/slob.c: respect list_head abstraction layer"): WARNING: CPU: 0 PID: 1 at lib/list_debug.c:28 __list_add_valid

[Tobin C. Harding]

On Wed, Apr 03, 2019 at 03:54:17PM +1100, Tobin C. Harding wrote:
> On Wed, Apr 03, 2019 at 10:00:38AM +0800, kernel test robot wrote:
> > Greetings,
> > 
> > 0day kernel testing robot got the below dmesg and the first bad commit is
> > 
> > https://git.kernel.org/pub/scm/linux/kernel/git/next/linux-next.git master
> > 
> > commit 15c8410c67adefd26ea0df1f1b86e1836051784b
> > Author:     Tobin C. Harding <tobin@xxxxxxxxxx>
> > AuthorDate: Fri Mar 29 10:01:23 2019 +1100
> > Commit:     Stephen Rothwell <sfr@xxxxxxxxxxxxxxxx>
> > CommitDate: Sat Mar 30 16:09:41 2019 +1100
> > 
> >     mm/slob.c: respect list_head abstraction layer
> >     
> >     Currently we reach inside the list_head.  This is a violation of the layer
> >     of abstraction provided by the list_head.  It makes the code fragile.
> >     More importantly it makes the code wicked hard to understand.
> >     
> >     The code logic is based on the page in which an allocation was made, we
> >     want to modify the slob_list we are working on to have this page at the
> >     front.  We already have a function to check if an entry is at the front of
> >     the list.  Recently a function was added to list.h to do the list
> >     rotation.  We can use these two functions to reduce line count, reduce
> >     code fragility, and reduce cognitive load required to read the code.
> >     
> >     Use list_head functions to interact with lists thereby maintaining the
> >     abstraction provided by the list_head structure.
> >     
> >     Link: http://lkml.kernel.org/r/20190318000234.22049-3-tobin@xxxxxxxxxx
> >     Signed-off-by: Tobin C. Harding <tobin@xxxxxxxxxx>
> >     Cc: Christoph Lameter <cl@xxxxxxxxx>
> >     Cc: David Rientjes <rientjes@xxxxxxxxxx>
> >     Cc: Joonsoo Kim <iamjoonsoo.kim@xxxxxxx>
> >     Cc: Pekka Enberg <penberg@xxxxxxxxxx>
> >     Cc: Roman Gushchin <guro@xxxxxx>
> >     Signed-off-by: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx>
> >     Signed-off-by: Stephen Rothwell <sfr@xxxxxxxxxxxxxxxx>
> > 
> > 2e1f88301e  include/linux/list.h: add list_rotate_to_front()
> > 15c8410c67  mm/slob.c: respect list_head abstraction layer
> > 05d08e2995  Add linux-next specific files for 20190402
> > +-------------------------------------------------------+------------+------------+---------------+
> > |                                                       | 2e1f88301e | 15c8410c67 | next-20190402 |
> > +-------------------------------------------------------+------------+------------+---------------+
> > | boot_successes                                        | 1009       | 198        | 299           |
> > | boot_failures                                         | 0          | 2          | 44            |
> > | WARNING:at_lib/list_debug.c:#__list_add_valid         | 0          | 2          | 44            |
> > | RIP:__list_add_valid                                  | 0          | 2          | 44            |
> > | WARNING:at_lib/list_debug.c:#__list_del_entry_valid   | 0          | 2          | 25            |
> > | RIP:__list_del_entry_valid                            | 0          | 2          | 25            |
> > | WARNING:possible_circular_locking_dependency_detected | 0          | 2          | 44            |
> > | RIP:_raw_spin_unlock_irqrestore                       | 0          | 2          | 2             |
> > | BUG:kernel_hang_in_test_stage                         | 0          | 0          | 6             |
> > | BUG:unable_to_handle_kernel                           | 0          | 0          | 1             |
> > | Oops:#[##]                                            | 0          | 0          | 1             |
> > | RIP:slob_page_alloc                                   | 0          | 0          | 1             |
> > | Kernel_panic-not_syncing:Fatal_exception              | 0          | 0          | 1             |
> > | RIP:delay_tsc                                         | 0          | 0          | 2             |
> > +-------------------------------------------------------+------------+------------+---------------+
> > 
> > [    2.618737] db_root: cannot open: /etc/target
> > [    2.620114] mtdoops: mtd device (mtddev=name/number) must be supplied
> > [    2.620967] slram: not enough parameters.
> > [    2.621614] ------------[ cut here ]------------
> > [    2.622254] list_add corruption. prev->next should be next (ffffffffaeeb71b0), but was ffffcee1406d3f70. (prev=ffffcee140422508).
> 
> Is this perhaps a false positive because we hackishly move the list_head
> 'head' and insert it back into the list.  Perhaps this is confusing the
> validation functions?

This has got me stumped.  I cannot create a test case where manipulating
a list with list_rotate_to_front() causes the list validation functions
to emit an error.  Also I cannot come up with a way on paper that it can
happen either.

I don't really know how to go forwards from here.  I'll sleep on it and
see if something comes to me, any ideas to look into please?

thanks,
Tobin.