Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > linux.kernel > #1323066 > unrolled thread

[PATCH v2 net 6/6] net: mvneta: Fix race condition during stopping

Back to article view | Back to linux.kernel

This discussion starts older than the indexed window; earlier articles aren't shown. The article labeled Started by below is the oldest one visible, not the original post.

Contents

  [PATCH v2 net 6/6] net: mvneta: Fix race condition during stopping Gregory CLEMENT <gregory.clement@free-electrons.com> - 2016-02-01 14:10 +0100
    Re: [PATCH v2 net 6/6] net: mvneta: Fix race condition during  stopping Jisheng Zhang <jszhang@marvell.com> - 2016-02-03 09:10 +0100

#1323066 — [PATCH v2 net 6/6] net: mvneta: Fix race condition during stopping

When stopping the port, the CPU notifier are still there whereas the
mvneta_stop_dev function calls mvneta_percpu_disable() on each CPUs.
It was possible to have a new CPU coming at this point which could be
racy.

This patch adds a flag preventing executing the code notifier for a new CPU
when the port is stopping.

Signed-off-by: Gregory CLEMENT <gregory.clement@free-electrons.com>
---
 drivers/net/ethernet/marvell/mvneta.c | 14 ++++++++++++++
 1 file changed, 14 insertions(+)

diff --git a/drivers/net/ethernet/marvell/mvneta.c b/drivers/net/ethernet/marvell/mvneta.c
index 4d40d2fde7ca..2f53975aa6ec 100644
--- a/drivers/net/ethernet/marvell/mvneta.c
+++ b/drivers/net/ethernet/marvell/mvneta.c
@@ -374,6 +374,7 @@ struct mvneta_port {
 	 * ensuring that the configuration remains coherent.
 	 */
 	spinlock_t lock;
+	bool is_stopping;
 
 	/* Core clock */
 	struct clk *clk;
@@ -2916,6 +2917,11 @@ static int mvneta_percpu_notifier(struct notifier_block *nfb,
 	switch (action) {
 	case CPU_ONLINE:
 	case CPU_ONLINE_FROZEN:
+		/* Configuring the driver for a new CPU while the
+		 * driver is stopping is racy, so just avoid it.
+		 */
+		if (pp->is_stopping)
+			break;
 		netif_tx_stop_all_queues(pp->dev);
 
 		/* We have to synchronise on tha napi of each CPU
@@ -3054,9 +3060,17 @@ static int mvneta_stop(struct net_device *dev)
 {
 	struct mvneta_port *pp = netdev_priv(dev);
 
+	/* Inform that we are stopping so we don't want to setup the
+	 * driver for new CPUs in the notifiers
+	 */
+	pp->is_stopping = true;
 	mvneta_stop_dev(pp);
 	mvneta_mdio_remove(pp);
 	unregister_cpu_notifier(&pp->cpu_notifier);
+	/* Now that the notifier are unregistered, we can clear the
+	 * flag
+	 */
+	pp->is_stopping = false;
 	on_each_cpu(mvneta_percpu_disable, pp, true);
 	free_percpu_irq(dev->irq, pp->ports);
 	mvneta_cleanup_rxqs(pp);
-- 
2.5.0

[toc] | [next] | [standalone]

#1325016 — Re: [PATCH v2 net 6/6] net: mvneta: Fix race condition during stopping

On Mon, 1 Feb 2016 14:07:47 +0100 Gregory CLEMENT wrote:

> When stopping the port, the CPU notifier are still there whereas the
> mvneta_stop_dev function calls mvneta_percpu_disable() on each CPUs.
> It was possible to have a new CPU coming at this point which could be
> racy.
> 
> This patch adds a flag preventing executing the code notifier for a new CPU
> when the port is stopping.
> 
> Signed-off-by: Gregory CLEMENT <gregory.clement@free-electrons.com>
> ---
>  drivers/net/ethernet/marvell/mvneta.c | 14 ++++++++++++++
>  1 file changed, 14 insertions(+)
> 
> diff --git a/drivers/net/ethernet/marvell/mvneta.c b/drivers/net/ethernet/marvell/mvneta.c
> index 4d40d2fde7ca..2f53975aa6ec 100644
> --- a/drivers/net/ethernet/marvell/mvneta.c
> +++ b/drivers/net/ethernet/marvell/mvneta.c
> @@ -374,6 +374,7 @@ struct mvneta_port {
>  	 * ensuring that the configuration remains coherent.
>  	 */
>  	spinlock_t lock;
> +	bool is_stopping;
>  
>  	/* Core clock */
>  	struct clk *clk;
> @@ -2916,6 +2917,11 @@ static int mvneta_percpu_notifier(struct notifier_block *nfb,
>  	switch (action) {
>  	case CPU_ONLINE:
>  	case CPU_ONLINE_FROZEN:
> +		/* Configuring the driver for a new CPU while the
> +		 * driver is stopping is racy, so just avoid it.
> +		 */
> +		if (pp->is_stopping)
> +			break;

I still see race. What about another cpu set is_stopping at this point?

Thanks


>  		netif_tx_stop_all_queues(pp->dev);
>  
>  		/* We have to synchronise on tha napi of each CPU
> @@ -3054,9 +3060,17 @@ static int mvneta_stop(struct net_device *dev)
>  {
>  	struct mvneta_port *pp = netdev_priv(dev);
>  
> +	/* Inform that we are stopping so we don't want to setup the
> +	 * driver for new CPUs in the notifiers
> +	 */
> +	pp->is_stopping = true;
>  	mvneta_stop_dev(pp);
>  	mvneta_mdio_remove(pp);
>  	unregister_cpu_notifier(&pp->cpu_notifier);
> +	/* Now that the notifier are unregistered, we can clear the
> +	 * flag
> +	 */
> +	pp->is_stopping = false;
>  	on_each_cpu(mvneta_percpu_disable, pp, true);
>  	free_percpu_irq(dev->irq, pp->ports);
>  	mvneta_cleanup_rxqs(pp);

[toc] | [prev] | [standalone]

Back to top | Article view | linux.kernel

csiph-web