Linux Virtualization - Jan 2013 - [PATCH V3 1/2] virtio-net: fix the set affinity bug when CPU IDs are not consecutive

As Michael mentioned, set affinity and select queue will not work very
well when CPU IDs are not consecutive, this can happen with hot unplug.
Fix this bug by traversal the online CPUs, and create a per cpu variable
to find the mapping from CPU to the preferable virtual-queue.

Cc: Rusty Russell <rusty at rustcorp.com.au>
Cc: "Michael S. Tsirkin" <mst at redhat.com>
Cc: Jason Wang <jasowang at redhat.com>
Cc: Eric Dumazet <erdnetdev at gmail.com>
Cc: virtualization at lists.linux-foundation.org
Cc: netdev at vger.kernel.org
Signed-off-by: Wanlong Gao <gaowanlong at cn.fujitsu.com>
---
 drivers/net/virtio_net.c | 39 +++++++++++++++++++++++++++++----------
 1 file changed, 29 insertions(+), 10 deletions(-)

diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c
index a6fcf15..a77f86c 100644
--- a/drivers/net/virtio_net.c
+++ b/drivers/net/virtio_net.c
@@ -41,6 +41,8 @@ module_param(gso, bool, 0444);
 #define VIRTNET_SEND_COMMAND_SG_MAX    2
 #define VIRTNET_DRIVER_VERSION "1.0.0"
 
+DEFINE_PER_CPU(int, vq_index) = -1;
+
 struct virtnet_stats {
 	struct u64_stats_sync tx_syncp;
 	struct u64_stats_sync rx_syncp;
@@ -1016,6 +1018,7 @@ static int virtnet_vlan_rx_kill_vid(struct net_device
*dev, u16 vid)
 static void virtnet_set_affinity(struct virtnet_info *vi, bool set)
 {
 	int i;
+	int cpu;
 
 	/* In multiqueue mode, when the number of cpu is equal to the number of
 	 * queue pairs, we let the queue pairs to be private to one cpu by
@@ -1029,16 +1032,29 @@ static void virtnet_set_affinity(struct virtnet_info
*vi, bool set)
 			return;
 	}
 
-	for (i = 0; i < vi->max_queue_pairs; i++) {
-		int cpu = set ? i : -1;
-		virtqueue_set_affinity(vi->rq[i].vq, cpu);
-		virtqueue_set_affinity(vi->sq[i].vq, cpu);
-	}
+	if (set) {
+		i = 0;
+		for_each_online_cpu(cpu) {
+			virtqueue_set_affinity(vi->rq[i].vq, cpu);
+			virtqueue_set_affinity(vi->sq[i].vq, cpu);
+			per_cpu(vq_index, cpu) = i;
+			i++;
+			if (i >= vi->max_queue_pairs)
+				break;
+		}
 
-	if (set)
 		vi->affinity_hint_set = true;
-	else
+	} else {
+		for(i = 0; i < vi->max_queue_pairs; i++) {
+			virtqueue_set_affinity(vi->rq[i].vq, -1);
+			virtqueue_set_affinity(vi->sq[i].vq, -1);
+		}
+
+		for_each_online_cpu(cpu)
+			per_cpu(vq_index, cpu) = -1;
+
 		vi->affinity_hint_set = false;
+	}
 }
 
 static void virtnet_get_ringparam(struct net_device *dev,
@@ -1127,12 +1143,15 @@ static int virtnet_change_mtu(struct net_device *dev,
int new_mtu)
 
 /* To avoid contending a lock hold by a vcpu who would exit to host, select the
  * txq based on the processor id.
- * TODO: handle cpu hotplug.
  */
 static u16 virtnet_select_queue(struct net_device *dev, struct sk_buff *skb)
 {
-	int txq = skb_rx_queue_recorded(skb) ? skb_get_rx_queue(skb) :
-		  smp_processor_id();
+	int txq = 0;
+
+	if (skb_rx_queue_recorded(skb))
+		txq = skb_get_rx_queue(skb);
+	else if ((txq = per_cpu(vq_index, smp_processor_id())) == -1)
+		txq = 0;
 
 	while (unlikely(txq >= dev->real_num_tx_queues))
 		txq -= dev->real_num_tx_queues;
-- 
1.8.1

Add a cpu notifier to virtio-net, so that we can reset the
virtqueue affinity if the cpu hotplug happens. It improve
the performance through enabling or disabling the virtqueue
affinity after doing cpu hotplug.

Cc: Rusty Russell <rusty at rustcorp.com.au>
Cc: "Michael S. Tsirkin" <mst at redhat.com>
Cc: Jason Wang <jasowang at redhat.com>
Cc: Eric Dumazet <erdnetdev at gmail.com>
Cc: virtualization at lists.linux-foundation.org
Cc: netdev at vger.kernel.org
Signed-off-by: Wanlong Gao <gaowanlong at cn.fujitsu.com>
---
 drivers/net/virtio_net.c | 30 ++++++++++++++++++++++++++++++
 1 file changed, 30 insertions(+)

diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c
index a77f86c..31a25db 100644
--- a/drivers/net/virtio_net.c
+++ b/drivers/net/virtio_net.c
@@ -26,6 +26,7 @@
 #include <linux/scatterlist.h>
 #include <linux/if_vlan.h>
 #include <linux/slab.h>
+#include <linux/cpu.h>
 
 static int napi_weight = 128;
 module_param(napi_weight, int, 0444);
@@ -125,6 +126,9 @@ struct virtnet_info {
 
 	/* Does the affinity hint is set for virtqueues? */
 	bool affinity_hint_set;
+
+	/* CPU hot plug notifier */
+	struct notifier_block nb;
 };
 
 struct skb_vnet_hdr {
@@ -1057,6 +1061,23 @@ static void virtnet_set_affinity(struct virtnet_info *vi,
bool set)
 	}
 }
 
+static int virtnet_cpu_callback(struct notifier_block *nfb,
+			        unsigned long action, void *hcpu)
+{
+	struct virtnet_info *vi = container_of(nfb, struct virtnet_info, nb);
+	switch(action) {
+	case CPU_ONLINE:
+	case CPU_ONLINE_FROZEN:
+	case CPU_DEAD:
+	case CPU_DEAD_FROZEN:
+		virtnet_set_affinity(vi, true);
+		break;
+	default:
+		break;
+	}
+	return NOTIFY_OK;
+}
+
 static void virtnet_get_ringparam(struct net_device *dev,
 				struct ethtool_ringparam *ring)
 {
@@ -1518,6 +1539,13 @@ static int virtnet_probe(struct virtio_device *vdev)
 		}
 	}
 
+	vi->nb.notifier_call = &virtnet_cpu_callback;
+	err = register_hotcpu_notifier(&vi->nb);
+	if (err) {
+		pr_debug("virtio_net: registering cpu notifier failed\n");
+		goto free_recv_bufs;
+	}
+
 	/* Assume link up if device can't report link status,
 	   otherwise get link status from config. */
 	if (virtio_has_feature(vi->vdev, VIRTIO_NET_F_STATUS)) {
@@ -1562,6 +1590,8 @@ static void virtnet_remove(struct virtio_device *vdev)
 {
 	struct virtnet_info *vi = vdev->priv;
 
+	unregister_hotcpu_notifier(&vi->nb);
+
 	/* Prevent config work handler from accessing the device. */
 	mutex_lock(&vi->config_lock);
 	vi->config_enable = false;
-- 
1.8.1

On 01/08/2013 06:07 PM, Wanlong Gao wrote:
> As Michael mentioned, set affinity and select queue will not work very
> well when CPU IDs are not consecutive, this can happen with hot unplug.
> Fix this bug by traversal the online CPUs, and create a per cpu variable
> to find the mapping from CPU to the preferable virtual-queue.
>
> Cc: Rusty Russell <rusty at rustcorp.com.au>
> Cc: "Michael S. Tsirkin" <mst at redhat.com>
> Cc: Jason Wang <jasowang at redhat.com>
> Cc: Eric Dumazet <erdnetdev at gmail.com>
> Cc: virtualization at lists.linux-foundation.org
> Cc: netdev at vger.kernel.org
> Signed-off-by: Wanlong Gao <gaowanlong at cn.fujitsu.com>
> ---
>  drivers/net/virtio_net.c | 39 +++++++++++++++++++++++++++++----------
>  1 file changed, 29 insertions(+), 10 deletions(-)
>
> diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c
> index a6fcf15..a77f86c 100644
> --- a/drivers/net/virtio_net.c
> +++ b/drivers/net/virtio_net.c
> @@ -41,6 +41,8 @@ module_param(gso, bool, 0444);
>  #define VIRTNET_SEND_COMMAND_SG_MAX    2
>  #define VIRTNET_DRIVER_VERSION "1.0.0"
>  
> +DEFINE_PER_CPU(int, vq_index) = -1;
> +
I think this should not be a global one, consider we may have more than
one virtio-net cards with different max queues.
>  struct virtnet_stats {
>  	struct u64_stats_sync tx_syncp;
>  	struct u64_stats_sync rx_syncp;
> @@ -1016,6 +1018,7 @@ static int virtnet_vlan_rx_kill_vid(struct net_device
*dev, u16 vid)
>  static void virtnet_set_affinity(struct virtnet_info *vi, bool set)
>  {
>  	int i;
> +	int cpu;
>  
>  	/* In multiqueue mode, when the number of cpu is equal to the number of
>  	 * queue pairs, we let the queue pairs to be private to one cpu by
> @@ -1029,16 +1032,29 @@ static void virtnet_set_affinity(struct
virtnet_info *vi, bool set)
>  			return;
>  	}
>  
> -	for (i = 0; i < vi->max_queue_pairs; i++) {
> -		int cpu = set ? i : -1;
> -		virtqueue_set_affinity(vi->rq[i].vq, cpu);
> -		virtqueue_set_affinity(vi->sq[i].vq, cpu);
> -	}
> +	if (set) {
> +		i = 0;
> +		for_each_online_cpu(cpu) {
> +			virtqueue_set_affinity(vi->rq[i].vq, cpu);
> +			virtqueue_set_affinity(vi->sq[i].vq, cpu);
> +			per_cpu(vq_index, cpu) = i;
> +			i++;
> +			if (i >= vi->max_queue_pairs)
> +				break;
Can this happen? we check only set when the number are
equal.
> +		}
>  
> -	if (set)
>  		vi->affinity_hint_set = true;
> -	else
> +	} else {
> +		for(i = 0; i < vi->max_queue_pairs; i++) {
> +			virtqueue_set_affinity(vi->rq[i].vq, -1);
> +			virtqueue_set_affinity(vi->sq[i].vq, -1);
> +		}
> +
> +		for_each_online_cpu(cpu)
> +			per_cpu(vq_index, cpu) = -1;
> +
This looks suboptimal since it may leads only txq zero is
used.
>  		vi->affinity_hint_set = false;
> +	}
>  }
>  
>  static void virtnet_get_ringparam(struct net_device *dev,
> @@ -1127,12 +1143,15 @@ static int virtnet_change_mtu(struct net_device
*dev, int new_mtu)
>  
>  /* To avoid contending a lock hold by a vcpu who would exit to host,
select the
>   * txq based on the processor id.
> - * TODO: handle cpu hotplug.
>   */
>  static u16 virtnet_select_queue(struct net_device *dev, struct sk_buff
*skb)
>  {
> -	int txq = skb_rx_queue_recorded(skb) ? skb_get_rx_queue(skb) :
> -		  smp_processor_id();
> +	int txq = 0;
> +
> +	if (skb_rx_queue_recorded(skb))
> +		txq = skb_get_rx_queue(skb);
> +	else if ((txq = per_cpu(vq_index, smp_processor_id())) == -1)
> +		txq = 0;
>  
>  	while (unlikely(txq >= dev->real_num_tx_queues))
>  		txq -= dev->real_num_tx_queues;

Wanlong Gao <gaowanlong at cn.fujitsu.com> writes:
>   */
>  static u16 virtnet_select_queue(struct net_device *dev, struct sk_buff
*skb)
>  {
> -	int txq = skb_rx_queue_recorded(skb) ? skb_get_rx_queue(skb) :
> -		  smp_processor_id();
> +	int txq = 0;
> +
> +	if (skb_rx_queue_recorded(skb))
> +		txq = skb_get_rx_queue(skb);
> +	else if ((txq = per_cpu(vq_index, smp_processor_id())) == -1)
> +		txq = 0;
You should use __get_cpu_var() instead of smp_processor_id() here, ie:

        else if ((txq = __get_cpu_var(vq_index)) == -1)

And AFAICT, no reason to initialize txq to 0 to start with.

So:

        int txq;

        if (skb_rx_queue_recorded(skb))
		txq = skb_get_rx_queue(skb);
        else {
                txq = __get_cpu_var(vq_index);
                if (txq == -1)
                        txq = 0;
        }

Now, just to confirm, I assume this can happen even if we use vq_index,
right, because of races with virtnet_set_channels?

  	while (unlikely(txq >= dev->real_num_tx_queues))
  		txq -= dev->real_num_tx_queues;


Thanks,
Rusty.