Re: [PATCH] mm: switch deferred split shrinker to list_lru

[Usama Arif <usama.arif@linux.dev>]

On 11/03/2026 18:43, Johannes Weiner wrote:
> The deferred split queue handles cgroups in a suboptimal fashion. The
> queue is per-NUMA node or per-cgroup, not the intersection. That means
> on a cgrouped system, a node-restricted allocation entering reclaim
> can end up splitting large pages on other nodes:
> 
> 	alloc/unmap
> 	  deferred_split_folio()
> 	    list_add_tail(memcg->split_queue)
> 	    set_shrinker_bit(memcg, node, deferred_shrinker_id)
> 
> 	for_each_zone_zonelist_nodemask(restricted_nodes)
> 	  mem_cgroup_iter()
> 	    shrink_slab(node, memcg)
> 	      shrink_slab_memcg(node, memcg)
> 	        if test_shrinker_bit(memcg, node, deferred_shrinker_id)
> 	          deferred_split_scan()
> 	            walks memcg->split_queue
> 
> The shrinker bit adds an imperfect guard rail. As soon as the cgroup
> has a single large page on the node of interest, all large pages owned
> by that memcg, including those on other nodes, will be split.
> 
> list_lru properly sets up per-node, per-cgroup lists. As a bonus, it
> streamlines a lot of the list operations and reclaim walks. It's used
> widely by other major shrinkers already. Convert the deferred split
> queue as well.
> 
> The list_lru per-memcg heads are instantiated on demand when the first
> object of interest is allocated for a cgroup, by calling
> memcg_list_lru_alloc(). Add calls to where splittable pages are
> created: anon faults, swapin faults, khugepaged collapse.
> 
> These calls create all possible node heads for the cgroup at once, so
> the migration code (between nodes) doesn't need any special care.
> 
> The folio_test_partially_mapped() state is currently protected and
> serialized wrt LRU state by the deferred split queue lock. To
> facilitate the transition, add helpers to the list_lru API to allow
> caller-side locking.
> 
> Signed-off-by: Johannes Weiner <hannes@cmpxchg.org>
> ---
>  include/linux/huge_mm.h    |   6 +-
>  include/linux/list_lru.h   |  48 ++++++
>  include/linux/memcontrol.h |   4 -
>  include/linux/mmzone.h     |  12 --
>  mm/huge_memory.c           | 326 +++++++++++--------------------------
>  mm/internal.h              |   2 +-
>  mm/khugepaged.c            |   7 +
>  mm/list_lru.c              | 197 ++++++++++++++--------
>  mm/memcontrol.c            |  12 +-
>  mm/memory.c                |  52 +++---
>  mm/mm_init.c               |  14 --
>  11 files changed, 310 insertions(+), 370 deletions(-)
> 

[..]

> @@ -3802,33 +3706,25 @@ static int __folio_freeze_and_split_unmapped(struct folio *folio, unsigned int n
>  	struct folio *new_folio, *next;
>  	int old_order = folio_order(folio);
>  	int ret = 0;
> -	struct deferred_split *ds_queue;
> +	struct list_lru_one *l;
>  
>  	VM_WARN_ON_ONCE(!mapping && end);
>  	/* Prevent deferred_split_scan() touching ->_refcount */
> -	ds_queue = folio_split_queue_lock(folio);
> +	l = list_lru_lock(&deferred_split_lru, folio_nid(folio), folio_memcg(folio));

Hello Johannes!

I think we need folio_memcg() to be under rcu_read_lock()? 
folio_memcg() calls obj_cgroup_memcg() which has lockdep_assert_once(rcu_read_lock_held()).

folio_split_queue_lock() wraps split_queue_lock() under rcu_read_lock() so wasnt an issue.
 

>  	if (folio_ref_freeze(folio, folio_cache_ref_count(folio) + 1)) {
>  		struct swap_cluster_info *ci = NULL;
>  		struct lruvec *lruvec;
>  
>  		if (old_order > 1) {
> -			if (!list_empty(&folio->_deferred_list)) {
> -				ds_queue->split_queue_len--;
> -				/*
> -				 * Reinitialize page_deferred_list after removing the
> -				 * page from the split_queue, otherwise a subsequent
> -				 * split will see list corruption when checking the
> -				 * page_deferred_list.
> -				 */
> -				list_del_init(&folio->_deferred_list);
> -			}
> +			__list_lru_del(&deferred_split_lru, l,
> +				       &folio->_deferred_list, folio_nid(folio));
>  			if (folio_test_partially_mapped(folio)) {
>  				folio_clear_partially_mapped(folio);
>  				mod_mthp_stat(old_order,
>  					MTHP_STAT_NR_ANON_PARTIALLY_MAPPED, -1);
>  			}
>  		}
> -		split_queue_unlock(ds_queue);
> +		list_lru_unlock(l);
>  		if (mapping) {
>  			int nr = folio_nr_pages(folio);
>  
> @@ -3929,7 +3825,7 @@ static int __folio_freeze_and_split_unmapped(struct folio *folio, unsigned int n
>  		if (ci)
>  			swap_cluster_unlock(ci);
>  	} else {
> -		split_queue_unlock(ds_queue);
> +		list_lru_unlock(l);
>  		return -EAGAIN;
>  	}
>  
[..]