Re: nbd: fix locking in case of error

[Paul Clements]

Pavel Machek wrote:
> This cleans up error handling some more, and prevents return with lock
> held from nbd_handle_req when nbd_send_req() fails. It also switches

>  static struct request *nbd_find_request(struct nbd_device *lo,
> @@ -467,10 +467,9 @@ static void nbd_handle_req(struct nbd_de
>  
>  	mutex_lock(&lo->tx_lock);
>  	if (unlikely(!lo->sock)) {
> -		mutex_unlock(&lo->tx_lock);
>  		printk(KERN_ERR "%s: Attempted send on closed socket\n",
>  		       lo->disk->disk_name);
> -		goto error_out;
> +		goto error_out_unlock;
>  	}
>  
>  	lo->active_req = req;
> @@ -478,7 +477,7 @@ static void nbd_handle_req(struct nbd_de
>  	if (nbd_send_req(lo, req) != 0) {
>  		printk(KERN_ERR "%s: Request send failed\n",
>  				lo->disk->disk_name);
> -		goto error_out;
> +		goto error_out_unlock;
>  	} else {
>  		spin_lock(&lo->queue_lock);
>  		list_add(&req->queuelist, &lo->queue_head);
> @@ -491,6 +490,8 @@ static void nbd_handle_req(struct nbd_de
>  
>  	return;
>  
> +error_out_unlock:
> +	mutex_unlock(&lo->tx_lock);
>  error_out:
>  	req->errors++;
>  	nbd_end_request(req);

Not sure I see the problem. tx_lock is unlocked in every return path 
AFAICS. You introduced the tx_lock problem with the previous patch, 
where the second error_out was not correct.

--
Paul
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/