1 files changed, 43 insertions, 24 deletions

diff --git a/fs/debugfs/inode.c b/fs/debugfs/inode.c
index 9c0444cccbe1..4a9f0e0c6644 100644
--- a/fs/debugfs/inode.c
+++ b/fs/debugfs/inode.c
@@ -245,10 +245,19 @@ static int debugfs_show_options(struct seq_file *m, struct dentry *root)
 	return 0;
 }
 
+static void debugfs_evict_inode(struct inode *inode)
+{
+	truncate_inode_pages(&inode->i_data, 0);
+	clear_inode(inode);
+	if (S_ISLNK(inode->i_mode))
+		kfree(inode->i_private);
+}
+
 static const struct super_operations debugfs_super_operations = {
 	.statfs		= simple_statfs,
 	.remount_fs	= debugfs_remount,
 	.show_options	= debugfs_show_options,
+	.evict_inode	= debugfs_evict_inode,
 };
 
 static int debug_fill_super(struct super_block *sb, void *data, int silent)
@@ -465,23 +474,14 @@ static int __debugfs_remove(struct dentry *dentry, struct dentry *parent)
 	int ret = 0;
 
 	if (debugfs_positive(dentry)) {
-		if (dentry->d_inode) {
-			dget(dentry);
-			switch (dentry->d_inode->i_mode & S_IFMT) {
-			case S_IFDIR:
-				ret = simple_rmdir(parent->d_inode, dentry);
-				break;
-			case S_IFLNK:
-				kfree(dentry->d_inode->i_private);
-				/* fall through */
-			default:
-				simple_unlink(parent->d_inode, dentry);
-				break;
-			}
-			if (!ret)
-				d_delete(dentry);
-			dput(dentry);
-		}
+		dget(dentry);
+		if (S_ISDIR(dentry->d_inode->i_mode))
+			ret = simple_rmdir(parent->d_inode, dentry);
+		else
+			simple_unlink(parent->d_inode, dentry);
+		if (!ret)
+			d_delete(dentry);
+		dput(dentry);
 	}
 	return ret;
 }
@@ -533,7 +533,7 @@ EXPORT_SYMBOL_GPL(debugfs_remove);
  */
 void debugfs_remove_recursive(struct dentry *dentry)
 {
-	struct dentry *child, *next, *parent;
+	struct dentry *child, *parent;
 
 	if (IS_ERR_OR_NULL(dentry))
 		return;
@@ -545,30 +545,49 @@ void debugfs_remove_recursive(struct dentry *dentry)
 	parent = dentry;
  down:
 	mutex_lock(&parent->d_inode->i_mutex);
-	list_for_each_entry_safe(child, next, &parent->d_subdirs, d_u.d_child) {
+ loop:
+	/*
+	 * The parent->d_subdirs is protected by the d_lock. Outside that
+	 * lock, the child can be unlinked and set to be freed which can
+	 * use the d_u.d_child as the rcu head and corrupt this list.
+	 */
+	spin_lock(&parent->d_lock);
+	list_for_each_entry(child, &parent->d_subdirs, d_child) {
 		if (!debugfs_positive(child))
 			continue;
 
 		/* perhaps simple_empty(child) makes more sense */
 		if (!list_empty(&child->d_subdirs)) {
+			spin_unlock(&parent->d_lock);
 			mutex_unlock(&parent->d_inode->i_mutex);
 			parent = child;
 			goto down;
 		}
- up:
+
+		spin_unlock(&parent->d_lock);
+
 		if (!__debugfs_remove(child, parent))
 			simple_release_fs(&debugfs_mount, &debugfs_mount_count);
+
+		/*
+		 * The parent->d_lock protects agaist child from unlinking
+		 * from d_subdirs. When releasing the parent->d_lock we can
+		 * no longer trust that the next pointer is valid.
+		 * Restart the loop. We'll skip this one with the
+		 * debugfs_positive() check.
+		 */
+		goto loop;
 	}
+	spin_unlock(&parent->d_lock);
 
 	mutex_unlock(&parent->d_inode->i_mutex);
 	child = parent;
 	parent = parent->d_parent;
 	mutex_lock(&parent->d_inode->i_mutex);
 
-	if (child != dentry) {
-		next = list_next_entry(child, d_u.d_child);
-		goto up;
-	}
+	if (child != dentry)
+		/* go up */
+		goto loop;
 
 	if (!__debugfs_remove(child, parent))
 		simple_release_fs(&debugfs_mount, &debugfs_mount_count);