Re: [PATCH v6 31/33] mm: memcontrol: convert objcg to be per-memcg per-node type

[Usama Arif]

On Thu,  5 Mar 2026 19:52:49 +0800 Qi Zheng <qi.zheng@xxxxxxxxx> wrote:

> From: Qi Zheng <zhengqi.arch@xxxxxxxxxxxxx>
> 
> Convert objcg to be per-memcg per-node type, so that when reparent LRU
> folios later, we can hold the lru lock at the node level, thus avoiding
> holding too many lru locks at once.
> 
> Signed-off-by: Qi Zheng <zhengqi.arch@xxxxxxxxxxxxx>
> Acked-by: Shakeel Butt <shakeel.butt@xxxxxxxxx>
> ---
>  include/linux/memcontrol.h | 23 +++++------
>  include/linux/sched.h      |  2 +-
>  mm/memcontrol.c            | 79 +++++++++++++++++++++++---------------
>  3 files changed, 62 insertions(+), 42 deletions(-)
> 

[...]

> @@ -4087,7 +4100,13 @@ static int mem_cgroup_css_online(struct cgroup_subsys_state *css)
>  	xa_store(&mem_cgroup_private_ids, memcg->id.id, memcg, GFP_KERNEL);
>  
>  	return 0;
> -free_shrinker:
> +free_objcg:
> +	for_each_node(nid) {
> +		struct mem_cgroup_per_node *pn = memcg->nodeinfo[nid];
> +
> +		if (pn && pn->orig_objcg)
> +			obj_cgroup_put(pn->orig_objcg);

Is it possible that you might call obj_cgroup_put twice on the same cgroup?

If css_create fails, css_free_rwork_fn is queued, which ends up calling
mem_cgroup_css_free which calls obj_cgroup_put again?

Maybe adding pn->orig_objcg = NULL overhere after obj_cgroup_put
is enough to prevent the double put from causing issues?

> +	}
>  	free_shrinker_info(memcg);
>  offline_kmem:
>  	memcg_offline_kmem(memcg);
> -- 
> 2.20.1
> 
>