Re: [Kernel Bug] INFO: rcu detected stall in count_memcg_event_mm

[Longxing Li]

  Hi Shakeel,

  Thank you for the reply.

  This seems like a CPU starvation scenario. The reproduction triggers
extreme CPU contention where rcu_preempt kthread cannot get sufficient
CPU time to complete grace periods.
  1. The stall occurs in count_memcg_event_mm / get_obj_cgroup_from_current
  2. Both paths correctly hold RCU read-side locks
  3. The deadlock is: page fault → dentry allocation → memcg kmem
tracking → RCU lookup → need RCU grace period → rcu_preempt starved
  4. This may require extreme stress to reproduce, and can be random

 We will try to reproduce with the latest kernel and report back if we
find any specific deficiency.

  Best regards,
  Longxing Li

Shakeel Butt <shakeel.butt@xxxxxxxxx> 于2026年6月12日周五 01:21写道：
>
> Hi Longxing,
>
> Thanks for the report.
>
> On Tue, Jun 09, 2026 at 07:57:56PM +0800, Longxing Li wrote:
> > Dear Linux kernel developers and maintainers,
> >
> > We would like to report a new kernel bug found by our tool. INFO: rcu
> > detected stall in count_memcg_event_mm. Details are as follows.
> >
> > Kernel commit: v5.15.189
>
> This is an old kernel, can you reproduce with the latest kernel? Also at the
> high level this seems like CPU starvation. Can you also describe your system and
> the workload/test which is trigerring this issue?
>