Re: [PATCH v5 9/9] mm: switch deferred split shrinker to list_lru

[Wei Yang <richard.weiyang@gmail.com>]

On Wed, May 27, 2026 at 04:45:16PM -0400, Johannes Weiner wrote:
[...]
>@@ -3890,34 +3804,43 @@ static int __folio_freeze_and_split_unmapped(struct folio *folio, unsigned int n
> 	struct folio *end_folio = folio_next(folio);
> 	struct folio *new_folio, *next;
> 	int old_order = folio_order(folio);
>+	struct list_lru_one *lru;
>+	bool dequeue_deferred;
> 	int ret = 0;
>-	struct deferred_split *ds_queue;
> 
> 	VM_WARN_ON_ONCE(!mapping && end);
>-	/* Prevent deferred_split_scan() touching ->_refcount */
>-	ds_queue = folio_split_queue_lock(folio);
>+	/*
>+	 * If this folio can be on the deferred split queue, lock out
>+	 * the shrinker before freezing the ref. If the shrinker sees
>+	 * a 0-ref folio, it assumes it beat folio_put() to the list
>+	 * lock and must clean up the LRU state - the same dequeue we
>+	 * will do below as part of the split.
>+	 */
>+	dequeue_deferred = folio_test_anon(folio) && old_order > 1;

Looking at __folio_remove_rmap(), we check !folio_is_device_private() before
deferred_split_folio(). __folio_freeze_and_split_unmapped() is used in
folio_split_unmapped(). According to its comment, it could take device-private
folio.

This means for device-private folio, we still lock lru_list and try to remove
it from deferred_split_lru. The good news is this doesn't harm the system, but
does some extra work.

Would it be better to add !folio_is_device_private() here?

The purpose to lock here is to prevent shrinker seeing ref-0 folio. Since
device-private folio is not on deferred_split_lru, shrink won't see it.

>+	if (dequeue_deferred) {
>+		struct mem_cgroup *memcg;
>+
>+		rcu_read_lock();
>+		memcg = folio_memcg(folio);
>+		lru = list_lru_lock(&deferred_split_lru,
>+				    folio_nid(folio), &memcg);
>+	}
> 	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);
>-			}
>+		if (dequeue_deferred) {
>+			__list_lru_del(&deferred_split_lru, lru,
>+				       &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);
> 			}
>+			list_lru_unlock(lru);
>+			rcu_read_unlock();
> 		}
>-		split_queue_unlock(ds_queue);
>+
> 		if (mapping) {
> 			int nr = folio_nr_pages(folio);
> 
>@@ -4017,7 +3940,10 @@ 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);
>+		if (dequeue_deferred) {
>+			list_lru_unlock(lru);
>+			rcu_read_unlock();
>+		}
> 		return -EAGAIN;
> 	}
> 

-- 
Wei Yang
Help you, Help me