fs/ntfs3: don't hold ni_lock when calling truncate_setsize()

syzbot is reporting hung task at do_user_addr_fault() [1], for there is
a silent deadlock between PG_locked bit and ni_lock lock.

Since filemap_update_page() calls filemap_read_folio() after calling
folio_trylock() which will set PG_locked bit, ntfs_truncate() must not
call truncate_setsize() which will wait for PG_locked bit to be cleared
when holding ni_lock lock.

Link: https://lore.kernel.org/all/00000000000060d41f05f139aa44@google.com/
Link: https://syzkaller.appspot.com/bug?extid=bed15dbf10294aa4f2ae
Reported-by: syzbot <syzbot+bed15dbf10294aa4f2ae@syzkaller.appspotmail.com>
Debugged-by: Linus Torvalds <torvalds@linux-foundation.org>
Co-developed-by: Hillf Danton <hdanton@sina.com>
Signed-off-by: Hillf Danton <hdanton@sina.com>
Signed-off-by: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp>
Fixes: 4342306f0f0d ("fs/ntfs3: Add file operations and implementation")
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

diff --git a/fs/ntfs3/file.c b/fs/ntfs3/file.c
index e5399ebc3a2b2d856bf5b81b392f97af2cd79814..d294cd9756887e13358e4fca461e48d06447f46e 100644 (file)
--- a/fs/ntfs3/file.c
+++ b/fs/ntfs3/file.c
@@ -390,10 +390,10 @@ static int ntfs_truncate(struct inode *inode, loff_t new_size)
 
        new_valid = ntfs_up_block(sb, min_t(u64, ni->i_valid, new_size));
 
-       ni_lock(ni);
-
        truncate_setsize(inode, new_size);
 
+       ni_lock(ni);
+
        down_write(&ni->file.run_lock);
        err = attr_set_size(ni, ATTR_DATA, NULL, 0, &ni->file.run, new_size,
                            &new_valid, ni->mi.sbi->options->prealloc, NULL);