Re: [PATCH v4 5/5] cgroup: use subsystem-specific rstat locks to avoid contention

[Shakeel Butt <shakeel.butt@linux.dev>]

On Tue, Apr 22, 2025 at 07:01:15AM -0700, Yosry Ahmed wrote:
> >  	struct cgroup_subsys_state *updated_next;	/* NULL iff not on the list */
> > @@ -793,6 +796,9 @@ struct cgroup_subsys {
> >  	 * specifies the mask of subsystems that this one depends on.
> >  	 */
> >  	unsigned int depends_on;
> > +
> > +	spinlock_t rstat_ss_lock;
> > +	raw_spinlock_t __percpu *rstat_ss_cpu_lock;
> 
> Can we use local_lock_t here instead? I guess it would be annoying
> because we won't be able to have common code for locking/unlocking. It's
> annoying because the local lock is a spinlock under the hood for non-RT
> kernels anyway..
> 

local_lock_t are spinlocks for RT kernels and just preempt_disable() for
non-RT kernels. Orthogonally I am exploring making memcg stats nmi safe
for bpf stuff and may look into making cgroup_rstat_updated() nmi safe
as well (which would require exploring local_trylock_t here).