[v1] net: cadence: macb: Synchronize stats calculations

[PATCH net] net: cadence: macb: Synchronize stats calculations

Posted by Sean Anderson 10 months ago

Stats calculations involve a RMW to add the stat update to the existing
value. This is currently not protected by any synchronization mechanism,
so data races are possible. Add a spinlock to protect the update. The
reader side could be protected using u64_stats, but we would still need
a spinlock for the update side anyway. And we always do an update
immediately before reading the stats anyway.

Fixes: 89e5785fc8a6 ("[PATCH] Atmel MACB ethernet driver")
Signed-off-by: Sean Anderson <sean.anderson@linux.dev>
---

 drivers/net/ethernet/cadence/macb.h      |  2 ++
 drivers/net/ethernet/cadence/macb_main.c | 12 ++++++++++--
 2 files changed, 12 insertions(+), 2 deletions(-)

diff --git a/drivers/net/ethernet/cadence/macb.h b/drivers/net/ethernet/cadence/macb.h
index 5740c98d8c9f..2847278d9cd4 100644
--- a/drivers/net/ethernet/cadence/macb.h
+++ b/drivers/net/ethernet/cadence/macb.h
@@ -1279,6 +1279,8 @@ struct macb {
 	struct clk		*rx_clk;
 	struct clk		*tsu_clk;
 	struct net_device	*dev;
+	/* Protects hw_stats and ethtool_stats */
+	spinlock_t		stats_lock;
 	union {
 		struct macb_stats	macb;
 		struct gem_stats	gem;
diff --git a/drivers/net/ethernet/cadence/macb_main.c b/drivers/net/ethernet/cadence/macb_main.c
index 48496209fb16..990a3863c6e1 100644
--- a/drivers/net/ethernet/cadence/macb_main.c
+++ b/drivers/net/ethernet/cadence/macb_main.c
@@ -1978,10 +1978,12 @@ static irqreturn_t macb_interrupt(int irq, void *dev_id)
 
 		if (status & MACB_BIT(ISR_ROVR)) {
 			/* We missed at least one packet */
+			spin_lock(&bp->stats_lock);
 			if (macb_is_gem(bp))
 				bp->hw_stats.gem.rx_overruns++;
 			else
 				bp->hw_stats.macb.rx_overruns++;
+			spin_unlock(&bp->stats_lock);
 
 			if (bp->caps & MACB_CAPS_ISR_CLEAR_ON_WRITE)
 				queue_writel(queue, ISR, MACB_BIT(ISR_ROVR));
@@ -3102,6 +3104,7 @@ static struct net_device_stats *gem_get_stats(struct macb *bp)
 	if (!netif_running(bp->dev))
 		return nstat;
 
+	spin_lock(&bp->stats_lock);
 	gem_update_stats(bp);
 
 	nstat->rx_errors = (hwstat->rx_frame_check_sequence_errors +
@@ -3131,6 +3134,7 @@ static struct net_device_stats *gem_get_stats(struct macb *bp)
 	nstat->tx_aborted_errors = hwstat->tx_excessive_collisions;
 	nstat->tx_carrier_errors = hwstat->tx_carrier_sense_errors;
 	nstat->tx_fifo_errors = hwstat->tx_underrun;
+	spin_unlock(&bp->stats_lock);
 
 	return nstat;
 }
@@ -3138,12 +3142,13 @@ static struct net_device_stats *gem_get_stats(struct macb *bp)
 static void gem_get_ethtool_stats(struct net_device *dev,
 				  struct ethtool_stats *stats, u64 *data)
 {
-	struct macb *bp;
+	struct macb *bp = netdev_priv(dev);
 
-	bp = netdev_priv(dev);
+	spin_lock(&bp->stats_lock);
 	gem_update_stats(bp);
 	memcpy(data, &bp->ethtool_stats, sizeof(u64)
 			* (GEM_STATS_LEN + QUEUE_STATS_LEN * MACB_MAX_QUEUES));
+	spin_unlock(&bp->stats_lock);
 }
 
 static int gem_get_sset_count(struct net_device *dev, int sset)
@@ -3193,6 +3198,7 @@ static struct net_device_stats *macb_get_stats(struct net_device *dev)
 		return gem_get_stats(bp);
 
 	/* read stats from hardware */
+	spin_lock(&bp->stats_lock);
 	macb_update_stats(bp);
 
 	/* Convert HW stats into netdevice stats */
@@ -3226,6 +3232,7 @@ static struct net_device_stats *macb_get_stats(struct net_device *dev)
 	nstat->tx_carrier_errors = hwstat->tx_carrier_errors;
 	nstat->tx_fifo_errors = hwstat->tx_underruns;
 	/* Don't know about heartbeat or window errors... */
+	spin_unlock(&bp->stats_lock);
 
 	return nstat;
 }
@@ -5097,6 +5104,7 @@ static int macb_probe(struct platform_device *pdev)
 		}
 	}
 	spin_lock_init(&bp->lock);
+	spin_lock_init(&bp->stats_lock);
 
 	/* setup capabilities */
 	macb_configure_caps(bp, macb_config);
-- 
2.35.1.1320.gc452695387.dirty

Re: [PATCH net] net: cadence: macb: Synchronize stats calculations

Posted by Maxime Chevallier 10 months ago

Hello Sean,

On Tue, 18 Feb 2025 14:50:36 -0500
Sean Anderson <sean.anderson@linux.dev> wrote:

> Stats calculations involve a RMW to add the stat update to the existing
> value. This is currently not protected by any synchronization mechanism,
> so data races are possible. Add a spinlock to protect the update. The
> reader side could be protected using u64_stats, but we would still need
> a spinlock for the update side anyway. And we always do an update
> immediately before reading the stats anyway.
> 
> Fixes: 89e5785fc8a6 ("[PATCH] Atmel MACB ethernet driver")
> Signed-off-by: Sean Anderson <sean.anderson@linux.dev>
> ---
> 
>  drivers/net/ethernet/cadence/macb.h      |  2 ++
>  drivers/net/ethernet/cadence/macb_main.c | 12 ++++++++++--
>  2 files changed, 12 insertions(+), 2 deletions(-)
> 
> diff --git a/drivers/net/ethernet/cadence/macb.h b/drivers/net/ethernet/cadence/macb.h
> index 5740c98d8c9f..2847278d9cd4 100644
> --- a/drivers/net/ethernet/cadence/macb.h
> +++ b/drivers/net/ethernet/cadence/macb.h
> @@ -1279,6 +1279,8 @@ struct macb {
>  	struct clk		*rx_clk;
>  	struct clk		*tsu_clk;
>  	struct net_device	*dev;
> +	/* Protects hw_stats and ethtool_stats */
> +	spinlock_t		stats_lock;
>  	union {
>  		struct macb_stats	macb;
>  		struct gem_stats	gem;
> diff --git a/drivers/net/ethernet/cadence/macb_main.c b/drivers/net/ethernet/cadence/macb_main.c
> index 48496209fb16..990a3863c6e1 100644
> --- a/drivers/net/ethernet/cadence/macb_main.c
> +++ b/drivers/net/ethernet/cadence/macb_main.c
> @@ -1978,10 +1978,12 @@ static irqreturn_t macb_interrupt(int irq, void *dev_id)
>  
>  		if (status & MACB_BIT(ISR_ROVR)) {
>  			/* We missed at least one packet */
> +			spin_lock(&bp->stats_lock);
>  			if (macb_is_gem(bp))
>  				bp->hw_stats.gem.rx_overruns++;
>  			else
>  				bp->hw_stats.macb.rx_overruns++;
> +			spin_unlock(&bp->stats_lock);
>  
>  			if (bp->caps & MACB_CAPS_ISR_CLEAR_ON_WRITE)
>  				queue_writel(queue, ISR, MACB_BIT(ISR_ROVR));
> @@ -3102,6 +3104,7 @@ static struct net_device_stats *gem_get_stats(struct macb *bp)
>  	if (!netif_running(bp->dev))
>  		return nstat;
>  
> +	spin_lock(&bp->stats_lock);
>  	gem_update_stats(bp);
>  
>  	nstat->rx_errors = (hwstat->rx_frame_check_sequence_errors +
> @@ -3131,6 +3134,7 @@ static struct net_device_stats *gem_get_stats(struct macb *bp)
>  	nstat->tx_aborted_errors = hwstat->tx_excessive_collisions;
>  	nstat->tx_carrier_errors = hwstat->tx_carrier_sense_errors;
>  	nstat->tx_fifo_errors = hwstat->tx_underrun;
> +	spin_unlock(&bp->stats_lock);
>  
>  	return nstat;
>  }
> @@ -3138,12 +3142,13 @@ static struct net_device_stats *gem_get_stats(struct macb *bp)
>  static void gem_get_ethtool_stats(struct net_device *dev,
>  				  struct ethtool_stats *stats, u64 *data)
>  {
> -	struct macb *bp;
> +	struct macb *bp = netdev_priv(dev);
>  
> -	bp = netdev_priv(dev);
> +	spin_lock(&bp->stats_lock);

Sorry if I missed something, but as you're using that lock within the
macb_interrupt(), shouldn't it be a spin_lock_irqsave() for all the
callsites that aren't in irq context ?

You would risk a deadlock otherwise.

Thanks,

Maxime

Re: [PATCH net] net: cadence: macb: Synchronize stats calculations

Posted by Sean Anderson 10 months ago

On 2/19/25 03:48, Maxime Chevallier wrote:
> Hello Sean,
> 
> On Tue, 18 Feb 2025 14:50:36 -0500
> Sean Anderson <sean.anderson@linux.dev> wrote:
> 
>> Stats calculations involve a RMW to add the stat update to the existing
>> value. This is currently not protected by any synchronization mechanism,
>> so data races are possible. Add a spinlock to protect the update. The
>> reader side could be protected using u64_stats, but we would still need
>> a spinlock for the update side anyway. And we always do an update
>> immediately before reading the stats anyway.
>> 
>> Fixes: 89e5785fc8a6 ("[PATCH] Atmel MACB ethernet driver")
>> Signed-off-by: Sean Anderson <sean.anderson@linux.dev>
>> ---
>> 
>>  drivers/net/ethernet/cadence/macb.h      |  2 ++
>>  drivers/net/ethernet/cadence/macb_main.c | 12 ++++++++++--
>>  2 files changed, 12 insertions(+), 2 deletions(-)
>> 
>> diff --git a/drivers/net/ethernet/cadence/macb.h b/drivers/net/ethernet/cadence/macb.h
>> index 5740c98d8c9f..2847278d9cd4 100644
>> --- a/drivers/net/ethernet/cadence/macb.h
>> +++ b/drivers/net/ethernet/cadence/macb.h
>> @@ -1279,6 +1279,8 @@ struct macb {
>>  	struct clk		*rx_clk;
>>  	struct clk		*tsu_clk;
>>  	struct net_device	*dev;
>> +	/* Protects hw_stats and ethtool_stats */
>> +	spinlock_t		stats_lock;
>>  	union {
>>  		struct macb_stats	macb;
>>  		struct gem_stats	gem;
>> diff --git a/drivers/net/ethernet/cadence/macb_main.c b/drivers/net/ethernet/cadence/macb_main.c
>> index 48496209fb16..990a3863c6e1 100644
>> --- a/drivers/net/ethernet/cadence/macb_main.c
>> +++ b/drivers/net/ethernet/cadence/macb_main.c
>> @@ -1978,10 +1978,12 @@ static irqreturn_t macb_interrupt(int irq, void *dev_id)
>>  
>>  		if (status & MACB_BIT(ISR_ROVR)) {
>>  			/* We missed at least one packet */
>> +			spin_lock(&bp->stats_lock);
>>  			if (macb_is_gem(bp))
>>  				bp->hw_stats.gem.rx_overruns++;
>>  			else
>>  				bp->hw_stats.macb.rx_overruns++;
>> +			spin_unlock(&bp->stats_lock);
>>  
>>  			if (bp->caps & MACB_CAPS_ISR_CLEAR_ON_WRITE)
>>  				queue_writel(queue, ISR, MACB_BIT(ISR_ROVR));
>> @@ -3102,6 +3104,7 @@ static struct net_device_stats *gem_get_stats(struct macb *bp)
>>  	if (!netif_running(bp->dev))
>>  		return nstat;
>>  
>> +	spin_lock(&bp->stats_lock);
>>  	gem_update_stats(bp);
>>  
>>  	nstat->rx_errors = (hwstat->rx_frame_check_sequence_errors +
>> @@ -3131,6 +3134,7 @@ static struct net_device_stats *gem_get_stats(struct macb *bp)
>>  	nstat->tx_aborted_errors = hwstat->tx_excessive_collisions;
>>  	nstat->tx_carrier_errors = hwstat->tx_carrier_sense_errors;
>>  	nstat->tx_fifo_errors = hwstat->tx_underrun;
>> +	spin_unlock(&bp->stats_lock);
>>  
>>  	return nstat;
>>  }
>> @@ -3138,12 +3142,13 @@ static struct net_device_stats *gem_get_stats(struct macb *bp)
>>  static void gem_get_ethtool_stats(struct net_device *dev,
>>  				  struct ethtool_stats *stats, u64 *data)
>>  {
>> -	struct macb *bp;
>> +	struct macb *bp = netdev_priv(dev);
>>  
>> -	bp = netdev_priv(dev);
>> +	spin_lock(&bp->stats_lock);
> 
> Sorry if I missed something, but as you're using that lock within the
> macb_interrupt(), shouldn't it be a spin_lock_irqsave() for all the
> callsites that aren't in irq context ?
> 
> You would risk a deadlock otherwise.

Yeah, looks like I missed this. Although I tested on a kernel with
lockdep enabled, and I thought that was supposed to catch this sort of
thing. Will send a v2.

--Sean

Re: [PATCH net] net: cadence: macb: Synchronize stats calculations

Posted by Andrew Lunn 10 months ago

> Yeah, looks like I missed this. Although I tested on a kernel with
> lockdep enabled, and I thought that was supposed to catch this sort of
> thing. Will send a v2.

It might be you need the sleep in atomic option, not lockdep. Better
still, just enable them all :-)

	Andrew

Re: [PATCH net] net: cadence: macb: Synchronize stats calculations

Posted by Sean Anderson 10 months ago

On 2/20/25 14:48, Andrew Lunn wrote:
>> Yeah, looks like I missed this. Although I tested on a kernel with
>> lockdep enabled, and I thought that was supposed to catch this sort of
>> thing. Will send a v2.
> 
> It might be you need the sleep in atomic option, not lockdep. Better
> still, just enable them all :-)
> 
> 	Andrew

I think the issue was that the lock is only taken in an error path and I
didn't trigger it during my testing. So lockdep never saw that this lock
could be taken in an interrupt.

--Sean