[PATCH 4.14 019/164] can: peak/pcie_fd: fix potential bug in restarting tx queue

[Greg Kroah-Hartman]

4.14-stable review patch.  If anyone has any objections, please let me know.

------------------

From: Stephane Grosjean <s.grosjean@xxxxxxxxxxxxxxx>

commit 91785de6f94b58c3fb6664609e3682f011bd28d2 upstream.

Don't rely on can_get_echo_skb() return value to wake the network tx
queue up: can_get_echo_skb() returns 0 if the echo array slot was not
occupied, but also when the DLC of the released echo frame was 0.

Signed-off-by: Stephane Grosjean <s.grosjean@xxxxxxxxxxxxxxx>
Signed-off-by: Marc Kleine-Budde <mkl@xxxxxxxxxxxxxx>
Signed-off-by: Greg Kroah-Hartman <gregkh@xxxxxxxxxxxxxxxxxxx>

---
 drivers/net/can/peak_canfd/peak_canfd.c |    9 +++------
 1 file changed, 3 insertions(+), 6 deletions(-)

--- a/drivers/net/can/peak_canfd/peak_canfd.c
+++ b/drivers/net/can/peak_canfd/peak_canfd.c
@@ -258,21 +258,18 @@ static int pucan_handle_can_rx(struct pe
 	/* if this frame is an echo, */
 	if ((rx_msg_flags & PUCAN_MSG_LOOPED_BACK) &&
 	    !(rx_msg_flags & PUCAN_MSG_SELF_RECEIVE)) {
-		int n;
 		unsigned long flags;
 
 		spin_lock_irqsave(&priv->echo_lock, flags);
-		n = can_get_echo_skb(priv->ndev, msg->client);
+		can_get_echo_skb(priv->ndev, msg->client);
 		spin_unlock_irqrestore(&priv->echo_lock, flags);
 
 		/* count bytes of the echo instead of skb */
 		stats->tx_bytes += cf_len;
 		stats->tx_packets++;
 
-		if (n) {
-			/* restart tx queue only if a slot is free */
-			netif_wake_queue(priv->ndev);
-		}
+		/* restart tx queue (a slot is free) */
+		netif_wake_queue(priv->ndev);
 
 		return 0;
 	}