Re: [PATCH] mm: list_lru: hold nlru lock to avoid reading transient negative nr_items

[Yang Shi <shy828301@gmail.com>]

On Mon, Nov 30, 2020 at 2:53 PM Roman Gushchin <guro@fb.com> wrote:
>
> On Mon, Nov 30, 2020 at 12:57:47PM -0800, Yang Shi wrote:
> > On Mon, Nov 30, 2020 at 12:09 PM Roman Gushchin <guro@fb.com> wrote:
> > >
> > > On Mon, Nov 30, 2020 at 10:45:14AM -0800, Yang Shi wrote:
> > > > When investigating a slab cache bloat problem, significant amount of
> > > > negative dentry cache was seen, but confusingly they neither got shrunk
> > > > by reclaimer (the host has very tight memory) nor be shrunk by dropping
> > > > cache.  The vmcore shows there are over 14M negative dentry objects on lru,
> > > > but tracing result shows they were even not scanned at all.  The further
> > > > investigation shows the memcg's vfs shrinker_map bit is not set.  So the
> > > > reclaimer or dropping cache just skip calling vfs shrinker.  So we have
> > > > to reboot the hosts to get the memory back.
> > > >
> > > > I didn't manage to come up with a reproducer in test environment, and the
> > > > problem can't be reproduced after rebooting.  But it seems there is race
> > > > between shrinker map bit clear and reparenting by code inspection.  The
> > > > hypothesis is elaborated as below.
> > > >
> > > > The memcg hierarchy on our production environment looks like:
> > > >                 root
> > > >                /    \
> > > >           system   user
> > > >
> > > > The main workloads are running under user slice's children, and it creates
> > > > and removes memcg frequently.  So reparenting happens very often under user
> > > > slice, but no task is under user slice directly.
> > > >
> > > > So with the frequent reparenting and tight memory pressure, the below
> > > > hypothetical race condition may happen:
> > > >
> > > >     CPU A                            CPU B                         CPU C
> > > > reparent
> > > >     dst->nr_items == 0
> > > >                                  shrinker:
> > > >                                      total_objects == 0
> > > >     add src->nr_items to dst
> > > >     set_bit
> > > >                                      retrun SHRINK_EMPTY
> > > >                                      clear_bit
> > > >                                                                   list_lru_del()
> > > > reparent again
> > > >     dst->nr_items may go negative
> > > >     due to current list_lru_del()
> > > >     on CPU C
> > > >                                  The second run of shrinker:
> > > >                                      read nr_items without any
> > > >                                      synchronization, so it may
> > > >                                      see intermediate negative
> > > >                                      nr_items then total_objects
> > > >                                      may return 0 conincidently
> > > >
> > > >                                      keep the bit cleared
> > > >     dst->nr_items != 0
> > > >     skip set_bit
> > > >     add scr->nr_item to dst
>
> Btw, I think I have a simpler explanation:
>
> A (0 objects)
> |
> B (N objects)
>
> Let's say the reparenting races with the deletion of a single slab object.
> list_lru_del() can see parent's lru list and substract 1 from nr_items == 0,
> setting A's nr_items to -1 (the item is actually still in B's list).
>
> memcg_drain_list_lru_node() will check !dst->nr_items && src->nr_items
> !-1 && N => 0 and not set the bit. But now we have (N-1) objects in A's list
> and the shrinker bit not set.

Yes, this is the exact race I elaborated in the commit log.

>
> My proposed fix should resolve it. Alternatively, we maybe can check if
> dst->nr_items <= 0 and only then set the bit, but it seems to be an unnecessary
> optimization.

Yes, I think "src->nr_items != 0" is good enough.

>
> Thanks!