For archival purposes, forwarding an incoming command email to
linux-kernel@vger.kernel.org, syzkaller-bugs@googlegroups.com.

***

Subject: [PATCH] ntfs3: initialize run_lock for $Extend inode records
Author: kartikey406@gmail.com

#syz test: git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git master

Inodes from the $Extend directory (NTFS system metadata files) were not
having their run_lock rwsem initialized. These inodes are assigned
ntfs_file_inode_operations but skip the normal S_ISREG initialization
path where run_lock is initialized.

When operations like truncate are called on these inodes, the code
attempts to acquire the uninitialized run_lock, triggering lockdep
warnings about using non-static keys.

Initialize run_lock for $Extend records to match the initialization done
for regular files.

Reported-by: syzbot+3e58a7dc1a8c00243999@syzkaller.appspotmail.com
Signed-off-by: Deepanshu Kartikey <kartikey406@gmail.com>
---
 fs/ntfs3/inode.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/fs/ntfs3/inode.c b/fs/ntfs3/inode.c
index 3959f23c487a..180cd984339b 100644
--- a/fs/ntfs3/inode.c
+++ b/fs/ntfs3/inode.c
@@ -472,6 +472,7 @@ static struct inode *ntfs_read_mft(struct inode *inode,
 		/* Records in $Extend are not a files or general directories. */
 		inode->i_op = &ntfs_file_inode_operations;
 		mode = S_IFREG;
+		init_rwsem(&ni->file.run_lock);
 	} else {
 		err = -EINVAL;
 		goto out;
-- 
2.43.0