diff mbox series

[net-next,v3,2/5] net: ethernet: stmicro: stmmac: first disable all queues in release

Message ID 20220716230802.20788-3-ansuelsmth@gmail.com (mailing list archive)
State New, archived
Headers show
Series Add MTU change with stmmac interface running | expand

Commit Message

Christian Marangi July 16, 2022, 11:07 p.m. UTC
Disable all queues before tx_disable in stmmac_release to prevent a
corner case where packet may be still queued at the same time tx_disable
is called resulting in kernel panic if some packet still has to be
processed.

Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
---
 drivers/net/ethernet/stmicro/stmmac/stmmac_main.c | 10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)

Comments

Russell King (Oracle) July 18, 2022, 9:19 a.m. UTC | #1
On Sun, Jul 17, 2022 at 01:07:59AM +0200, Christian Marangi wrote:
> Disable all queues before tx_disable in stmmac_release to prevent a
> corner case where packet may be still queued at the same time tx_disable
> is called resulting in kernel panic if some packet still has to be
> processed.
> 
> Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
> ---
>  drivers/net/ethernet/stmicro/stmmac/stmmac_main.c | 10 +++++-----
>  1 file changed, 5 insertions(+), 5 deletions(-)
> 
> diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
> index 5578abb14949..1854dcdd6095 100644
> --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
> +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
> @@ -3758,6 +3758,11 @@ static int stmmac_release(struct net_device *dev)
>  	struct stmmac_priv *priv = netdev_priv(dev);
>  	u32 chan;
>  
> +	stmmac_disable_all_queues(priv);
> +
> +	for (chan = 0; chan < priv->plat->tx_queues_to_use; chan++)
> +		hrtimer_cancel(&priv->tx_queue[chan].txtimer);
> +
>  	netif_tx_disable(dev);

Is there a reason not to call phylink_stop() as the very first thing in
this function? That will bring the link (and therefore carrier) down
before phylink_stop() returns which should also prevent packets being
queued into the driver for transmission.

Thanks.
diff mbox series

Patch

diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
index 5578abb14949..1854dcdd6095 100644
--- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
+++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
@@ -3758,6 +3758,11 @@  static int stmmac_release(struct net_device *dev)
 	struct stmmac_priv *priv = netdev_priv(dev);
 	u32 chan;
 
+	stmmac_disable_all_queues(priv);
+
+	for (chan = 0; chan < priv->plat->tx_queues_to_use; chan++)
+		hrtimer_cancel(&priv->tx_queue[chan].txtimer);
+
 	netif_tx_disable(dev);
 
 	if (device_may_wakeup(priv->device))
@@ -3766,11 +3771,6 @@  static int stmmac_release(struct net_device *dev)
 	phylink_stop(priv->phylink);
 	phylink_disconnect_phy(priv->phylink);
 
-	stmmac_disable_all_queues(priv);
-
-	for (chan = 0; chan < priv->plat->tx_queues_to_use; chan++)
-		hrtimer_cancel(&priv->tx_queue[chan].txtimer);
-
 	/* Free the IRQ lines */
 	stmmac_free_irq(dev, REQ_IRQ_ERR_ALL, 0);