Re: [syzbot] [kernel?] KCSAN: data-race in __fput / __tty_hangup (4)

From: Tetsuo Handa
Date: Fri Apr 21 2023 - 11:12:38 EST


On 2023/04/21 17:21, Dmitry Vyukov wrote:
> If I am reading this correctly, this race can lead to NULL derefs
> among other things.
> hung_up_tty_fops does not have splice_read, while other fops have.
>
> So the following code in splice can execute NULL callback:
>
> if (unlikely(!in->f_op->splice_read))
> return warn_unsupported(in, "read");
> return in->f_op->splice_read(in, ppos, pipe, len, flags);
>

__fput(file) is called when the last reference to file is released.
Since __tty_hangup() traverses tty->tty_files under tty->files_lock,
tty_add_file() needs to hold a ref before adding to tty->tty_files
in order to defer concurrent __fput() by other threads?

diff --git a/drivers/tty/tty_io.c b/drivers/tty/tty_io.c
index 36fb945fdad4..2838703d48cf 100644
--- a/drivers/tty/tty_io.c
+++ b/drivers/tty/tty_io.c
@@ -197,7 +197,7 @@ void tty_add_file(struct tty_struct *tty, struct file *file)
struct tty_file_private *priv = file->private_data;

priv->tty = tty;
- priv->file = file;
+ priv->file = get_file(file);

spin_lock(&tty->files_lock);
list_add(&priv->list, &tty->tty_files);
@@ -228,6 +228,7 @@ static void tty_del_file(struct file *file)
spin_lock(&tty->files_lock);
list_del(&priv->list);
spin_unlock(&tty->files_lock);
+ fput(file);
tty_free_file(file);
}