[PATCH 5.15.y] fs: rotate inodes whose page cache could not be dropped

[Sasha Levin <sashal@kernel.org>]

From: Nikolaos Karaolidis <nick@karaolidis.com>

[ Upstream commit 8ee09f597763acfd31c62a578ba88b792fc41b24 ]

When inode_lru_isolate() drops the page cache of an unused inode, it
has to unlock the list, so it returns LRU_RETRY and the walk restarts
at the head of the list, where the inode still is. If
invalidate_mapping_pages() could not empty the mapping, the restarted
walk strips the same inode again, and so on until nr_to_walk runs out.
Every walk of that list then ends on this inode, and nothing behind it
is reclaimed for as long as its cache stays.

Before commit 2a0629834cd8 ("vfs: Don't evict inode under the inode lru
traversing context") the walker pinned the inode with a reference, and
dropping it in iput_final() set I_REFERENCED, so the restarted walk
rotated the inode to the tail. Set I_REFERENCED again when the mapping
is still populated after the invalidation, in the same i_lock section
that clears I_LRU_ISOLATING. inode_unpin_lru_isolating() now expects
i_lock to be held, like inode_pin_lru_isolating().

On CONFIG_HIGHMEM kernels this can happen to any unused inode: a folio
is left behind if it is locked, under writeback, spliced into a pipe,
or still in another CPU's LRU-add batch after being read. Without
HIGHMEM it happens to DAX inodes whose only entry is at index 0, which
mapping_shrinkable() lets onto the LRU and invalidate_mapping_pages()
never removes.

Fixes: 2a0629834cd8 ("vfs: Don't evict inode under the inode lru traversing context")
Cc: stable@vger.kernel.org
Signed-off-by: Nikolaos Karaolidis <nick@karaolidis.com>
Link: https://patch.msgid.link/20261016102515.111285-1-nick@karaolidis.com
Reviewed-by: Jan Kara <jack@suse.cz>
Reviewed-by: Zhihao Cheng <chengzhihao1@huawei.com>
Signed-off-by: Christian Brauner (Amutable) <brauner@kernel.org>
[ sashal: replaced inode_state_set() with a direct inode->i_state bit update ]
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 fs/inode.c | 9 +++++++--
 1 file changed, 7 insertions(+), 2 deletions(-)

diff --git a/fs/inode.c b/fs/inode.c
index bfeec2830fb0d..2e03403ed50c0 100644
--- a/fs/inode.c
+++ b/fs/inode.c
@@ -490,12 +490,11 @@ static void inode_pin_lru_isolating(struct inode *inode)
 
 static void inode_unpin_lru_isolating(struct inode *inode)
 {
-	spin_lock(&inode->i_lock);
+	lockdep_assert_held(&inode->i_lock);
 	WARN_ON(!(inode->i_state & I_LRU_ISOLATING));
 	inode->i_state &= ~I_LRU_ISOLATING;
 	smp_mb();
 	wake_up_bit(&inode->i_state, __I_LRU_ISOLATING);
-	spin_unlock(&inode->i_lock);
 }
 
 static void inode_wait_for_lru_isolating(struct inode *inode)
@@ -846,7 +845,13 @@ static enum lru_status inode_lru_isolate(struct list_head *item,
 			if (current->reclaim_state)
 				current->reclaim_state->reclaimed_slab += reap;
 		}
+
+		spin_lock(&inode->i_lock);
+		/* The walk restarts here: don't retry leftovers right away */
+		if (!mapping_empty(&inode->i_data))
+			inode->i_state |= I_REFERENCED;
 		inode_unpin_lru_isolating(inode);
+		spin_unlock(&inode->i_lock);
 		spin_lock(lru_lock);
 		return LRU_RETRY;
 	}
-- 
2.53.0