fs/drop_caches.c: avoid softlockups in drop_pagecache_sb()

When superblock has lots of inodes without any pagecache (like is the
case for /proc), drop_pagecache_sb() will iterate through all of them
without dropping sb->s_inode_list_lock which can lead to softlockups
(one of our customers hit this).

Fix the problem by going to the slow path and doing cond_resched() in
case the process needs rescheduling.

Link: http://lkml.kernel.org/r/[email protected]
Signed-off-by: Jan Kara <[email protected]>
Acked-by: Michal Hocko <[email protected]>
Reviewed-by: Andrew Morton <[email protected]>
Cc: Al Viro <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>

Author: jankara

fs/drop_caches.c

@@ -21,15 +21,21 @@ static void drop_pagecache_sb(struct super_block *sb, void *unused)
 	spin_lock(&sb->s_inode_list_lock);
 	list_for_each_entry(inode, &sb->s_inodes, i_sb_list) {
 		spin_lock(&inode->i_lock);
+		/*
+		 * We must skip inodes in unusual state. We may also skip
+		 * inodes without pages but we deliberately won't in case
+		 * we need to reschedule to avoid softlockups.
+		 */
 		if ((inode->i_state & (I_FREEING|I_WILL_FREE|I_NEW)) ||
-		    (inode->i_mapping->nrpages == 0)) {
+		    (inode->i_mapping->nrpages == 0 && !need_resched())) {
 			spin_unlock(&inode->i_lock);
 			continue;
 		}
 		__iget(inode);
 		spin_unlock(&inode->i_lock);
 		spin_unlock(&sb->s_inode_list_lock);
 
+		cond_resched();
 		invalidate_mapping_pages(inode->i_mapping, 0, -1);
 		iput(toput_inode);
 		toput_inode = inode;