Re: [PATCH] mm/shrinker: do not hold RCU lock in shrinker_debugfs_count_show()

[Shakeel Butt]

On Wed, Jun 10, 2026 at 05:26:47PM -0700, SeongJae Park wrote:
> On Wed, 10 Jun 2026 17:22:51 -0700 SeongJae Park <sj@xxxxxxxxxx> wrote:
> 
> > On Wed, 10 Jun 2026 16:20:48 -0700 Shakeel Butt <shakeel.butt@xxxxxxxxx> wrote:
> > 
> > > Reading the debugfs "count" file of a memcg-aware shrinker can sleep
> > > inside an RCU read-side critical section:
> > > 
> > >   BUG: sleeping function called from invalid context at kernel/cgroup/rstat.c:421
> > >   RCU nest depth: 1, expected: 0
> > >    css_rstat_flush
> > >    mem_cgroup_flush_stats
> > >    zswap_shrinker_count
> > >    shrinker_debugfs_count_show
> > > 
> > > shrinker_debugfs_count_show() invokes the ->count_objects() callback
> > > under rcu_read_lock(). The zswap callback flushes memcg stats via
> > > css_rstat_flush(), which may sleep, so it must not run under RCU.
> > > 
> > > The RCU lock is not needed here. mem_cgroup_iter() takes RCU internally
> > > and returns a memcg holding a css reference (dropped on the next
> > > iteration or by mem_cgroup_iter_break()), so the memcg stays alive
> > > without it. The shrinker is kept alive by the open debugfs file:
> > > shrinker_free() removes the debugfs entries via
> > > debugfs_remove_recursive(), which waits for in-flight readers to drain,
> > > before call_rcu(..., shrinker_free_rcu_cb). The sibling "scan" handler
> > > already invokes the sleeping ->scan_objects() callback with no RCU
> > > section.
> > > 
> > > Drop the rcu_read_lock()/rcu_read_unlock().
> > 
> > All make sense to me, thank you for the nice description and the fix!
> > 
> > > 
> > > Fixes: 5035ebc644ae ("mm: shrinkers: introduce debugfs interface for memory shrinkers")
> 
> Forgot asking this, sorry.  Are you intentionally not adding Cc: stable@ here?
> I think the user impact is arguably minor enough to not Cc-ing stable@, but
> just thought it would be good to make the intention clear.
> 

Haha I was just being lazy to think through if this should be CCed to stable or
not and letting others/reviewers do that for me :P

> > > Reported-by: Zenghui Yu <zenghui.yu@xxxxxxxxx>
> > > Closes: https://lore.kernel.org/all/c052a064-cddb-494f-a0d8-f8a10b4b1c4d@xxxxxxxxx/
> > > Suggested-by: Nhat Pham <nphamcs@xxxxxxxxx>
> > > Signed-off-by: Shakeel Butt <shakeel.butt@xxxxxxxxx>
> > 
> > Reviewed-by: SeongJae Park <sj@xxxxxxxxxx>

Thanks.