kernfs: fix missing kernfs_iattr_rwsem locking

When the kernfs_iattr_rwsem was introduced a case was missed.

The update of the kernfs directory node child count was also protected
by the kernfs_rwsem and needs to be included in the change so that the
child count (and so the inode n_link attribute) does not change while
holding the rwsem for read.

Fixes: 9caf6961 ("kernfs: Introduce separate rwsem to protect inode attributes.")
Cc: stable <stable@kernel.org>
Signed-off-by: Ian Kent <raven@themaw.net>
Reviewed-By: Imran Khan <imran.f.khan@oracle.com>
Acked-by: Miklos Szeredi <mszeredi@redhat.com>
Cc: Anders Roxell <anders.roxell@linaro.org>
Cc: Arnd Bergmann <arnd@arndb.de>
Cc: Minchan Kim <minchan@kernel.org>
Cc: Eric Sandeen <sandeen@sandeen.net>
Link: https://lore.kernel.org/r/169128520941.68052.15749253469930138901.stgit@donald.themaw.netSigned-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

fs/kernfs/dir.c

@@ -383,9 +383,11 @@ static int kernfs_link_sibling(struct kernfs_node *kn)
 	rb_insert_color(&kn->rb, &kn->parent->dir.children);
 
 	/* successfully added, account subdir number */
+	down_write(&kernfs_root(kn)->kernfs_iattr_rwsem);
 	if (kernfs_type(kn) == KERNFS_DIR)
 		kn->parent->dir.subdirs++;
 	kernfs_inc_rev(kn->parent);
+	up_write(&kernfs_root(kn)->kernfs_iattr_rwsem);
 
 	return 0;
 }
@@ -408,9 +410,11 @@ static bool kernfs_unlink_sibling(struct kernfs_node *kn)
 	if (RB_EMPTY_NODE(&kn->rb))
 		return false;
 
+	down_write(&kernfs_root(kn)->kernfs_iattr_rwsem);
 	if (kernfs_type(kn) == KERNFS_DIR)
 		kn->parent->dir.subdirs--;
 	kernfs_inc_rev(kn->parent);
+	up_write(&kernfs_root(kn)->kernfs_iattr_rwsem);
 
 	rb_erase(&kn->rb, &kn->parent->dir.children);
 	RB_CLEAR_NODE(&kn->rb);