writeback: Avoid softlockup when switching many inodes

[ Upstream commit 66c14dc ]

process_inode_switch_wbs_work() can be switching over 100 inodes to a
different cgroup. Since switching an inode requires counting all dirty &
under-writeback pages in the address space of each inode, this can take
a significant amount of time. Add a possibility to reschedule after
processing each inode to avoid softlockups.

Acked-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Jan Kara <jack@suse.cz>
Signed-off-by: Christian Brauner <brauner@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>

Author: jankara

fs/fs-writeback.c

@@ -503,6 +503,7 @@ static void inode_switch_wbs_work_fn(struct work_struct *work)
 	 */
 	down_read(&bdi->wb_switch_rwsem);
 
+	inodep = isw->inodes;
 	/*
 	 * By the time control reaches here, RCU grace period has passed
 	 * since I_WB_SWITCH assertion and all wb stat update transactions
@@ -513,6 +514,7 @@ static void inode_switch_wbs_work_fn(struct work_struct *work)
 	 * gives us exclusion against all wb related operations on @inode
 	 * including IO list manipulations and stat updates.
 	 */
+relock:
 	if (old_wb < new_wb) {
 		spin_lock(&old_wb->list_lock);
 		spin_lock_nested(&new_wb->list_lock, SINGLE_DEPTH_NESTING);
@@ -521,10 +523,17 @@ static void inode_switch_wbs_work_fn(struct work_struct *work)
 		spin_lock_nested(&old_wb->list_lock, SINGLE_DEPTH_NESTING);
 	}
 
-	for (inodep = isw->inodes; *inodep; inodep++) {
+	while (*inodep) {
 		WARN_ON_ONCE((*inodep)->i_wb != old_wb);
 		if (inode_do_switch_wbs(*inodep, old_wb, new_wb))
 			nr_switched++;
+		inodep++;
+		if (*inodep && need_resched()) {
+			spin_unlock(&new_wb->list_lock);
+			spin_unlock(&old_wb->list_lock);
+			cond_resched();
+			goto relock;
+		}
 	}
 
 	spin_unlock(&new_wb->list_lock);