Skip to contentSkip to conversation

virtio-net: synchronize operstate with admin state on up/down

PATCH Superseded Jason Wang 2 participants 10 messages May 30, 2024 – Jun 20, 2024

Subsystems (4)
lore
Share & tools
lore

Jason Wang

author
#0 lore raw

PATCH v2 1/1 virtio-net: synchronize operstate with admin state on up/down

This patch synchronize operstate with admin state per RFC2863.

This is done by trying to toggle the carrier upon open/close and synchronize with the config change work. This allows propagate status correctly to stacked devices like:

ip link add link enp0s3 macvlan0 type macvlan
ip link set link enp0s3 down
ip link show

Before this patch:

3: enp0s3: <BROADCAST,MULTICAST> mtu 1500 qdisc pfifo_fast state DOWN mode DEFAULT group default qlen 1000

    link/ether 00:00:05:00:00:09 brd ff:ff:ff:ff:ff:ff
......
5: macvlan0@enp0s3: <BROADCAST,MULTICAST,UP,LOWER_UP,M-DOWN> mtu 1500 qdisc noqueue state UP mode DEFAULT group default qlen 1000
    link/ether b2:a9:c5:04:da:53 brd ff:ff:ff:ff:ff:ff

After this patch:

3: enp0s3: <BROADCAST,MULTICAST> mtu 1500 qdisc pfifo_fast state DOWN mode DEFAULT group default qlen 1000

    link/ether 00:00:05:00:00:09 brd ff:ff:ff:ff:ff:ff
...
5: macvlan0@enp0s3: <NO-CARRIER,BROADCAST,MULTICAST,UP,M-DOWN> mtu 1500 qdisc noqueue state LOWERLAYERDOWN mode DEFAULT group default qlen 1000
    link/ether b2:a9:c5:04:da:53 brd ff:ff:ff:ff:ff:ff

---
Changes since V1:
- rebase
- add ack/review tags

---
 drivers/net/virtio_net.c | 94 +++++++++++++++++++++++++++-------------
 1 file changed, 63 insertions(+), 31 deletions(-)
drivers/net/virtio_net.c +63 -31
index 4a802c0ea2cb..69e4ae353c51 100644
@@ -433,6 +433,12 @@ struct virtnet_info {
 	/* The lock to synchronize the access to refill_enabled */
 	spinlock_t refill_lock;
 
+	/* Is config change enabled? */
+	bool config_change_enabled;
+
+	/* The lock to synchronize the access to config_change_enabled */
+	spinlock_t config_change_lock;
+
 	/* Work struct for config space updates */
 	struct work_struct config_work;
 
@@ -623,6 +629,20 @@ static void disable_delayed_refill(struct virtnet_info *vi)
 	spin_unlock_bh(&vi->refill_lock);
 }
 
+static void enable_config_change(struct virtnet_info *vi)
+{
+	spin_lock_irq(&vi->config_change_lock);
+	vi->config_change_enabled = true;
+	spin_unlock_irq(&vi->config_change_lock);
+}
+
+static void disable_config_change(struct virtnet_info *vi)
+{
+	spin_lock_irq(&vi->config_change_lock);
+	vi->config_change_enabled = false;
+	spin_unlock_irq(&vi->config_change_lock);
+}
+
 static void enable_rx_mode_work(struct virtnet_info *vi)
 {
 	rtnl_lock();
@@ -2421,6 +2441,25 @@ static int virtnet_enable_queue_pair(struct virtnet_info *vi, int qp_index)
 	return err;
 }
 
+static void virtnet_update_settings(struct virtnet_info *vi)
+{
+	u32 speed;
+	u8 duplex;
+
+	if (!virtio_has_feature(vi->vdev, VIRTIO_NET_F_SPEED_DUPLEX))
+		return;
+
+	virtio_cread_le(vi->vdev, struct virtio_net_config, speed, &speed);
+
+	if (ethtool_validate_speed(speed))
+		vi->speed = speed;
+
+	virtio_cread_le(vi->vdev, struct virtio_net_config, duplex, &duplex);
+
+	if (ethtool_validate_duplex(duplex))
+		vi->duplex = duplex;
+}
+
 static int virtnet_open(struct net_device *dev)
 {
 	struct virtnet_info *vi = netdev_priv(dev);
@@ -2439,6 +2478,18 @@ static int virtnet_open(struct net_device *dev)
 			goto err_enable_qp;
 	}
 
+	/* Assume link up if device can't report link status,
+	   otherwise get link status from config. */
+	netif_carrier_off(dev);
+	if (virtio_has_feature(vi->vdev, VIRTIO_NET_F_STATUS)) {
+		enable_config_change(vi);
+		schedule_work(&vi->config_work);
+	} else {
+		vi->status = VIRTIO_NET_S_LINK_UP;
+		virtnet_update_settings(vi);
+		netif_carrier_on(dev);
+	}
+
 	return 0;
 
 err_enable_qp:
@@ -2875,12 +2926,19 @@ static int virtnet_close(struct net_device *dev)
 	disable_delayed_refill(vi);
 	/* Make sure refill_work doesn't re-enable napi! */
 	cancel_delayed_work_sync(&vi->refill);
+	/* Make sure config notification doesn't schedule config work */
+	disable_config_change(vi);
+	/* Make sure status updating is cancelled */
+	cancel_work_sync(&vi->config_work);
 
 	for (i = 0; i < vi->max_queue_pairs; i++) {
 		virtnet_disable_queue_pair(vi, i);
 		cancel_work_sync(&vi->rq[i].dim.work);
 	}
 
+	vi->status &= ~VIRTIO_NET_S_LINK_UP;
+	netif_carrier_off(dev);
+
 	return 0;
 }
 
@@ -4583,25 +4641,6 @@ static void virtnet_init_settings(struct net_device *dev)
 	vi->duplex = DUPLEX_UNKNOWN;
 }
 
-static void virtnet_update_settings(struct virtnet_info *vi)
-{
-	u32 speed;
-	u8 duplex;
-
-	if (!virtio_has_feature(vi->vdev, VIRTIO_NET_F_SPEED_DUPLEX))
-		return;
-
-	virtio_cread_le(vi->vdev, struct virtio_net_config, speed, &speed);
-
-	if (ethtool_validate_speed(speed))
-		vi->speed = speed;
-
-	virtio_cread_le(vi->vdev, struct virtio_net_config, duplex, &duplex);
-
-	if (ethtool_validate_duplex(duplex))
-		vi->duplex = duplex;
-}
-
 static u32 virtnet_get_rxfh_key_size(struct net_device *dev)
 {
 	return ((struct virtnet_info *)netdev_priv(dev))->rss_key_size;
@@ -5163,7 +5202,10 @@ static void virtnet_config_changed(struct virtio_device *vdev)
 {
 	struct virtnet_info *vi = vdev->priv;
 
-	schedule_work(&vi->config_work);
+	spin_lock_irq(&vi->config_change_lock);
+	if (vi->config_change_enabled)
+		schedule_work(&vi->config_work);
+	spin_unlock_irq(&vi->config_change_lock);
 }
 
 static void virtnet_free_queues(struct virtnet_info *vi)
@@ -5706,6 +5748,7 @@ static int virtnet_probe(struct virtio_device *vdev)
 	INIT_WORK(&vi->config_work, virtnet_config_changed_work);
 	INIT_WORK(&vi->rx_mode_work, virtnet_rx_mode_work);
 	spin_lock_init(&vi->refill_lock);
+	spin_lock_init(&vi->config_change_lock);
 
 	if (virtio_has_feature(vdev, VIRTIO_NET_F_MRG_RXBUF)) {
 		vi->mergeable_rx_bufs = true;
@@ -5901,17 +5944,6 @@ static int virtnet_probe(struct virtio_device *vdev)
 		goto free_unregister_netdev;
 	}
 
-	/* Assume link up if device can't report link status,
-	   otherwise get link status from config. */
-	netif_carrier_off(dev);
-	if (virtio_has_feature(vi->vdev, VIRTIO_NET_F_STATUS)) {
-		schedule_work(&vi->config_work);
-	} else {
-		vi->status = VIRTIO_NET_S_LINK_UP;
-		virtnet_update_settings(vi);
-		netif_carrier_on(dev);
-	}
-
 	for (i = 0; i < ARRAY_SIZE(guest_offloads); i++)
 		if (virtio_has_feature(vi->vdev, guest_offloads[i]))
 			set_bit(guest_offloads[i], &vi->guest_offloads);
Signature · 1 line
2.42.0
#1 lore raw
Quoted 51 lines · Jason Wang

This patch synchronize operstate with admin state per RFC2863.

This is done by trying to toggle the carrier upon open/close and synchronize with the config change work. This allows propagate status correctly to stacked devices like:

ip link add link enp0s3 macvlan0 type macvlan
ip link set link enp0s3 down
ip link show

Before this patch:

3: enp0s3: <BROADCAST,MULTICAST> mtu 1500 qdisc pfifo_fast state DOWN mode DEFAULT group default qlen 1000

    link/ether 00:00:05:00:00:09 brd ff:ff:ff:ff:ff:ff
......
5: macvlan0@enp0s3: <BROADCAST,MULTICAST,UP,LOWER_UP,M-DOWN> mtu 1500 qdisc noqueue state UP mode DEFAULT group default qlen 1000
    link/ether b2:a9:c5:04:da:53 brd ff:ff:ff:ff:ff:ff

After this patch:

3: enp0s3: <BROADCAST,MULTICAST> mtu 1500 qdisc pfifo_fast state DOWN mode DEFAULT group default qlen 1000

    link/ether 00:00:05:00:00:09 brd ff:ff:ff:ff:ff:ff
...
5: macvlan0@enp0s3: <NO-CARRIER,BROADCAST,MULTICAST,UP,M-DOWN> mtu 1500 qdisc noqueue state LOWERLAYERDOWN mode DEFAULT group default qlen 1000
    link/ether b2:a9:c5:04:da:53 brd ff:ff:ff:ff:ff:ff

---
Changes since V1:
- rebase
- add ack/review tags

---
 drivers/net/virtio_net.c | 94 +++++++++++++++++++++++++++-------------
 1 file changed, 63 insertions(+), 31 deletions(-)
drivers/net/virtio_net.c +0 -0
index 4a802c0ea2cb..69e4ae353c51 100644
@@ -433,6 +433,12 @@ struct virtnet_info {
 	/* The lock to synchronize the access to refill_enabled */
 	spinlock_t refill_lock;

drivers/net/virtio_net.c struct virtnet_info { · line 444

Quoted 10 lines
drivers/net/virtio_net.c +6 -0
 
+	/* Is config change enabled? */
+	bool config_change_enabled;
+
+	/* The lock to synchronize the access to config_change_enabled */
+	spinlock_t config_change_lock;
+
 	/* Work struct for config space updates */
 	struct work_struct config_work;
 

But we already have dev->config_lock and dev->config_enabled.

And it actually works better - instead of discarding config change events it defers them until enabled.

Quoted 152 lines
drivers/net/virtio_net.c +57 -31
@@ -623,6 +629,20 @@ static void disable_delayed_refill(struct virtnet_info *vi)
 	spin_unlock_bh(&vi->refill_lock);
 }
 
+static void enable_config_change(struct virtnet_info *vi)
+{
+	spin_lock_irq(&vi->config_change_lock);
+	vi->config_change_enabled = true;
+	spin_unlock_irq(&vi->config_change_lock);
+}
+
+static void disable_config_change(struct virtnet_info *vi)
+{
+	spin_lock_irq(&vi->config_change_lock);
+	vi->config_change_enabled = false;
+	spin_unlock_irq(&vi->config_change_lock);
+}
+
 static void enable_rx_mode_work(struct virtnet_info *vi)
 {
 	rtnl_lock();
@@ -2421,6 +2441,25 @@ static int virtnet_enable_queue_pair(struct virtnet_info *vi, int qp_index)
 	return err;
 }
 
+static void virtnet_update_settings(struct virtnet_info *vi)
+{
+	u32 speed;
+	u8 duplex;
+
+	if (!virtio_has_feature(vi->vdev, VIRTIO_NET_F_SPEED_DUPLEX))
+		return;
+
+	virtio_cread_le(vi->vdev, struct virtio_net_config, speed, &speed);
+
+	if (ethtool_validate_speed(speed))
+		vi->speed = speed;
+
+	virtio_cread_le(vi->vdev, struct virtio_net_config, duplex, &duplex);
+
+	if (ethtool_validate_duplex(duplex))
+		vi->duplex = duplex;
+}
+
 static int virtnet_open(struct net_device *dev)
 {
 	struct virtnet_info *vi = netdev_priv(dev);
@@ -2439,6 +2478,18 @@ static int virtnet_open(struct net_device *dev)
 			goto err_enable_qp;
 	}
 
+	/* Assume link up if device can't report link status,
+	   otherwise get link status from config. */
+	netif_carrier_off(dev);
+	if (virtio_has_feature(vi->vdev, VIRTIO_NET_F_STATUS)) {
+		enable_config_change(vi);
+		schedule_work(&vi->config_work);
+	} else {
+		vi->status = VIRTIO_NET_S_LINK_UP;
+		virtnet_update_settings(vi);
+		netif_carrier_on(dev);
+	}
+
 	return 0;
 
 err_enable_qp:
@@ -2875,12 +2926,19 @@ static int virtnet_close(struct net_device *dev)
 	disable_delayed_refill(vi);
 	/* Make sure refill_work doesn't re-enable napi! */
 	cancel_delayed_work_sync(&vi->refill);
+	/* Make sure config notification doesn't schedule config work */
+	disable_config_change(vi);
+	/* Make sure status updating is cancelled */
+	cancel_work_sync(&vi->config_work);
 
 	for (i = 0; i < vi->max_queue_pairs; i++) {
 		virtnet_disable_queue_pair(vi, i);
 		cancel_work_sync(&vi->rq[i].dim.work);
 	}
 
+	vi->status &= ~VIRTIO_NET_S_LINK_UP;
+	netif_carrier_off(dev);
+
 	return 0;
 }
 
@@ -4583,25 +4641,6 @@ static void virtnet_init_settings(struct net_device *dev)
 	vi->duplex = DUPLEX_UNKNOWN;
 }
 
-static void virtnet_update_settings(struct virtnet_info *vi)
-{
-	u32 speed;
-	u8 duplex;
-
-	if (!virtio_has_feature(vi->vdev, VIRTIO_NET_F_SPEED_DUPLEX))
-		return;
-
-	virtio_cread_le(vi->vdev, struct virtio_net_config, speed, &speed);
-
-	if (ethtool_validate_speed(speed))
-		vi->speed = speed;
-
-	virtio_cread_le(vi->vdev, struct virtio_net_config, duplex, &duplex);
-
-	if (ethtool_validate_duplex(duplex))
-		vi->duplex = duplex;
-}
-
 static u32 virtnet_get_rxfh_key_size(struct net_device *dev)
 {
 	return ((struct virtnet_info *)netdev_priv(dev))->rss_key_size;
@@ -5163,7 +5202,10 @@ static void virtnet_config_changed(struct virtio_device *vdev)
 {
 	struct virtnet_info *vi = vdev->priv;
 
-	schedule_work(&vi->config_work);
+	spin_lock_irq(&vi->config_change_lock);
+	if (vi->config_change_enabled)
+		schedule_work(&vi->config_work);
+	spin_unlock_irq(&vi->config_change_lock);
 }
 
 static void virtnet_free_queues(struct virtnet_info *vi)
@@ -5706,6 +5748,7 @@ static int virtnet_probe(struct virtio_device *vdev)
 	INIT_WORK(&vi->config_work, virtnet_config_changed_work);
 	INIT_WORK(&vi->rx_mode_work, virtnet_rx_mode_work);
 	spin_lock_init(&vi->refill_lock);
+	spin_lock_init(&vi->config_change_lock);
 
 	if (virtio_has_feature(vdev, VIRTIO_NET_F_MRG_RXBUF)) {
 		vi->mergeable_rx_bufs = true;
@@ -5901,17 +5944,6 @@ static int virtnet_probe(struct virtio_device *vdev)
 		goto free_unregister_netdev;
 	}
 
-	/* Assume link up if device can't report link status,
-	   otherwise get link status from config. */
-	netif_carrier_off(dev);
-	if (virtio_has_feature(vi->vdev, VIRTIO_NET_F_STATUS)) {
-		schedule_work(&vi->config_work);
-	} else {
-		vi->status = VIRTIO_NET_S_LINK_UP;
-		virtnet_update_settings(vi);
-		netif_carrier_on(dev);
-	}
-
 	for (i = 0; i < ARRAY_SIZE(guest_offloads); i++)
 		if (virtio_has_feature(vi->vdev, guest_offloads[i]))
 			set_bit(guest_offloads[i], &vi->guest_offloads);
Signature · 1 line
2.42.0
author
#2 lore raw
Quoted 67 lines · Michael S. Tsirkin
Quoted 51 lines · Jason Wang

This patch synchronize operstate with admin state per RFC2863.

This is done by trying to toggle the carrier upon open/close and synchronize with the config change work. This allows propagate status correctly to stacked devices like:

ip link add link enp0s3 macvlan0 type macvlan
ip link set link enp0s3 down
ip link show

Before this patch:

3: enp0s3: <BROADCAST,MULTICAST> mtu 1500 qdisc pfifo_fast state DOWN mode DEFAULT group default qlen 1000

    link/ether 00:00:05:00:00:09 brd ff:ff:ff:ff:ff:ff
......
5: macvlan0@enp0s3: <BROADCAST,MULTICAST,UP,LOWER_UP,M-DOWN> mtu 1500 qdisc noqueue state UP mode DEFAULT group default qlen 1000
    link/ether b2:a9:c5:04:da:53 brd ff:ff:ff:ff:ff:ff

After this patch:

3: enp0s3: <BROADCAST,MULTICAST> mtu 1500 qdisc pfifo_fast state DOWN mode DEFAULT group default qlen 1000

    link/ether 00:00:05:00:00:09 brd ff:ff:ff:ff:ff:ff
...
5: macvlan0@enp0s3: <NO-CARRIER,BROADCAST,MULTICAST,UP,M-DOWN> mtu 1500 qdisc noqueue state LOWERLAYERDOWN mode DEFAULT group default qlen 1000
    link/ether b2:a9:c5:04:da:53 brd ff:ff:ff:ff:ff:ff

---
Changes since V1:
- rebase
- add ack/review tags

---
 drivers/net/virtio_net.c | 94 +++++++++++++++++++++++++++-------------
 1 file changed, 63 insertions(+), 31 deletions(-)
drivers/net/virtio_net.c +0 -0
index 4a802c0ea2cb..69e4ae353c51 100644
@@ -433,6 +433,12 @@ struct virtnet_info {
      /* The lock to synchronize the access to refill_enabled */
      spinlock_t refill_lock;

drivers/net/virtio_net.c struct virtnet_info { · line 444

Quoted 10 lines
drivers/net/virtio_net.c +6 -0
 
+     /* Is config change enabled? */
+     bool config_change_enabled;
+
+     /* The lock to synchronize the access to config_change_enabled */
+     spinlock_t config_change_lock;
+
      /* Work struct for config space updates */
      struct work_struct config_work;
 

But we already have dev->config_lock and dev->config_enabled.

Quoted 3 lines

And it actually works better - instead of discarding config change events it defers them until enabled.

Yes but then both virtio-net driver and virtio core can ask to enable and disable and then we need some kind of synchronization which is non-trivial.

And device enabling on the core is different from bringing the device up in the networking subsystem. Here we just delay to deal with the config change interrupt on ndo_open(). (E.g try to ack announce is meaningless when the device is down).

Thanks

author
#3 lore raw
Quoted 73 lines · Jason Wang
Quoted 67 lines · Michael S. Tsirkin
Quoted 51 lines · Jason Wang

This patch synchronize operstate with admin state per RFC2863.

This is done by trying to toggle the carrier upon open/close and synchronize with the config change work. This allows propagate status correctly to stacked devices like:

ip link add link enp0s3 macvlan0 type macvlan
ip link set link enp0s3 down
ip link show

Before this patch:

3: enp0s3: <BROADCAST,MULTICAST> mtu 1500 qdisc pfifo_fast state DOWN mode DEFAULT group default qlen 1000

    link/ether 00:00:05:00:00:09 brd ff:ff:ff:ff:ff:ff
......
5: macvlan0@enp0s3: <BROADCAST,MULTICAST,UP,LOWER_UP,M-DOWN> mtu 1500 qdisc noqueue state UP mode DEFAULT group default qlen 1000
    link/ether b2:a9:c5:04:da:53 brd ff:ff:ff:ff:ff:ff

After this patch:

3: enp0s3: <BROADCAST,MULTICAST> mtu 1500 qdisc pfifo_fast state DOWN mode DEFAULT group default qlen 1000

    link/ether 00:00:05:00:00:09 brd ff:ff:ff:ff:ff:ff
...
5: macvlan0@enp0s3: <NO-CARRIER,BROADCAST,MULTICAST,UP,M-DOWN> mtu 1500 qdisc noqueue state LOWERLAYERDOWN mode DEFAULT group default qlen 1000
    link/ether b2:a9:c5:04:da:53 brd ff:ff:ff:ff:ff:ff

---
Changes since V1:
- rebase
- add ack/review tags

---
 drivers/net/virtio_net.c | 94 +++++++++++++++++++++++++++-------------
 1 file changed, 63 insertions(+), 31 deletions(-)
drivers/net/virtio_net.c +0 -0
index 4a802c0ea2cb..69e4ae353c51 100644
@@ -433,6 +433,12 @@ struct virtnet_info {
      /* The lock to synchronize the access to refill_enabled */
      spinlock_t refill_lock;

drivers/net/virtio_net.c struct virtnet_info { · line 444

Quoted 10 lines
drivers/net/virtio_net.c +6 -0
 
+     /* Is config change enabled? */
+     bool config_change_enabled;
+
+     /* The lock to synchronize the access to config_change_enabled */
+     spinlock_t config_change_lock;
+
      /* Work struct for config space updates */
      struct work_struct config_work;
 

But we already have dev->config_lock and dev->config_enabled.

Quoted 3 lines

And it actually works better - instead of discarding config change events it defers them until enabled.

Quoted 8 lines

Yes but then both virtio-net driver and virtio core can ask to enable and disable and then we need some kind of synchronization which is non-trivial.

And device enabling on the core is different from bringing the device up in the networking subsystem. Here we just delay to deal with the config change interrupt on ndo_open(). (E.g try to ack announce is meaningless when the device is down).

Or maybe you meant to make config_enabled a nest counter?

Thanks

Quoted 2 lines

Thanks

#4 lore raw
In reply to Jason Wang
Quoted 68 lines · Jason Wang
Quoted 67 lines · Michael S. Tsirkin
Quoted 51 lines · Jason Wang

This patch synchronize operstate with admin state per RFC2863.

This is done by trying to toggle the carrier upon open/close and synchronize with the config change work. This allows propagate status correctly to stacked devices like:

ip link add link enp0s3 macvlan0 type macvlan
ip link set link enp0s3 down
ip link show

Before this patch:

3: enp0s3: <BROADCAST,MULTICAST> mtu 1500 qdisc pfifo_fast state DOWN mode DEFAULT group default qlen 1000

    link/ether 00:00:05:00:00:09 brd ff:ff:ff:ff:ff:ff
......
5: macvlan0@enp0s3: <BROADCAST,MULTICAST,UP,LOWER_UP,M-DOWN> mtu 1500 qdisc noqueue state UP mode DEFAULT group default qlen 1000
    link/ether b2:a9:c5:04:da:53 brd ff:ff:ff:ff:ff:ff

After this patch:

3: enp0s3: <BROADCAST,MULTICAST> mtu 1500 qdisc pfifo_fast state DOWN mode DEFAULT group default qlen 1000

    link/ether 00:00:05:00:00:09 brd ff:ff:ff:ff:ff:ff
...
5: macvlan0@enp0s3: <NO-CARRIER,BROADCAST,MULTICAST,UP,M-DOWN> mtu 1500 qdisc noqueue state LOWERLAYERDOWN mode DEFAULT group default qlen 1000
    link/ether b2:a9:c5:04:da:53 brd ff:ff:ff:ff:ff:ff

---
Changes since V1:
- rebase
- add ack/review tags

---
 drivers/net/virtio_net.c | 94 +++++++++++++++++++++++++++-------------
 1 file changed, 63 insertions(+), 31 deletions(-)
drivers/net/virtio_net.c +0 -0
index 4a802c0ea2cb..69e4ae353c51 100644
@@ -433,6 +433,12 @@ struct virtnet_info {
      /* The lock to synchronize the access to refill_enabled */
      spinlock_t refill_lock;

drivers/net/virtio_net.c struct virtnet_info { · line 444

Quoted 10 lines
drivers/net/virtio_net.c +6 -0
 
+     /* Is config change enabled? */
+     bool config_change_enabled;
+
+     /* The lock to synchronize the access to config_change_enabled */
+     spinlock_t config_change_lock;
+
      /* Work struct for config space updates */
      struct work_struct config_work;
 

But we already have dev->config_lock and dev->config_enabled.

Quoted 7 lines
Quoted 3 lines

And it actually works better - instead of discarding config change events it defers them until enabled.

Yes but then both virtio-net driver and virtio core can ask to enable and disable and then we need some kind of synchronization which is non-trivial.

Well for core it happens on bring up path before driver works and later on tear down after it is gone.
So I do not think they ever do it at the same time.

Quoted 6 lines

And device enabling on the core is different from bringing the device up in the networking subsystem. Here we just delay to deal with the config change interrupt on ndo_open(). (E.g try to ack announce is meaningless when the device is down).

Thanks

another thing is that it is better not to re-read all config on link up if there was no config interrupt - less vm exits.

Signature · 1 line
MST
author
#5 lore raw
Quoted 74 lines · Michael S. Tsirkin
Quoted 72 lines · Jason Wang
Quoted 70 lines · Michael S. Tsirkin
Quoted 51 lines · Jason Wang

This patch synchronize operstate with admin state per RFC2863.

This is done by trying to toggle the carrier upon open/close and synchronize with the config change work. This allows propagate status correctly to stacked devices like:

ip link add link enp0s3 macvlan0 type macvlan
ip link set link enp0s3 down
ip link show

Before this patch:

3: enp0s3: <BROADCAST,MULTICAST> mtu 1500 qdisc pfifo_fast state DOWN mode DEFAULT group default qlen 1000

    link/ether 00:00:05:00:00:09 brd ff:ff:ff:ff:ff:ff
......
5: macvlan0@enp0s3: <BROADCAST,MULTICAST,UP,LOWER_UP,M-DOWN> mtu 1500 qdisc noqueue state UP mode DEFAULT group default qlen 1000
    link/ether b2:a9:c5:04:da:53 brd ff:ff:ff:ff:ff:ff

After this patch:

3: enp0s3: <BROADCAST,MULTICAST> mtu 1500 qdisc pfifo_fast state DOWN mode DEFAULT group default qlen 1000

    link/ether 00:00:05:00:00:09 brd ff:ff:ff:ff:ff:ff
...
5: macvlan0@enp0s3: <NO-CARRIER,BROADCAST,MULTICAST,UP,M-DOWN> mtu 1500 qdisc noqueue state LOWERLAYERDOWN mode DEFAULT group default qlen 1000
    link/ether b2:a9:c5:04:da:53 brd ff:ff:ff:ff:ff:ff

---
Changes since V1:
- rebase
- add ack/review tags

---
 drivers/net/virtio_net.c | 94 +++++++++++++++++++++++++++-------------
 1 file changed, 63 insertions(+), 31 deletions(-)
drivers/net/virtio_net.c +0 -0
index 4a802c0ea2cb..69e4ae353c51 100644
@@ -433,6 +433,12 @@ struct virtnet_info {
      /* The lock to synchronize the access to refill_enabled */
      spinlock_t refill_lock;

drivers/net/virtio_net.c struct virtnet_info { · line 444

Quoted 10 lines
drivers/net/virtio_net.c +6 -0
 
+     /* Is config change enabled? */
+     bool config_change_enabled;
+
+     /* The lock to synchronize the access to config_change_enabled */
+     spinlock_t config_change_lock;
+
      /* Work struct for config space updates */
      struct work_struct config_work;
 

But we already have dev->config_lock and dev->config_enabled.

And it actually works better - instead of discarding config change events it defers them until enabled.

Quoted 7 lines
Quoted 3 lines

Yes but then both virtio-net driver and virtio core can ask to enable and disable and then we need some kind of synchronization which is non-trivial.

Well for core it happens on bring up path before driver works and later on tear down after it is gone.
So I do not think they ever do it at the same time.

For example, there could be a suspend/resume when the admin state is down.

Quoted 11 lines
Quoted 8 lines

And device enabling on the core is different from bringing the device up in the networking subsystem. Here we just delay to deal with the config change interrupt on ndo_open(). (E.g try to ack announce is meaningless when the device is down).

Thanks

another thing is that it is better not to re-read all config on link up if there was no config interrupt - less vm exits.

Yes, but it should not matter much as it's done in the ndo_open().

Thanks

Quoted 4 lines

--
MST

author
#6 lore raw
Quoted 93 lines · Jason Wang
Quoted 74 lines · Michael S. Tsirkin
Quoted 72 lines · Jason Wang
Quoted 70 lines · Michael S. Tsirkin
Quoted 51 lines · Jason Wang

This patch synchronize operstate with admin state per RFC2863.

This is done by trying to toggle the carrier upon open/close and synchronize with the config change work. This allows propagate status correctly to stacked devices like:

ip link add link enp0s3 macvlan0 type macvlan
ip link set link enp0s3 down
ip link show

Before this patch:

3: enp0s3: <BROADCAST,MULTICAST> mtu 1500 qdisc pfifo_fast state DOWN mode DEFAULT group default qlen 1000

    link/ether 00:00:05:00:00:09 brd ff:ff:ff:ff:ff:ff
......
5: macvlan0@enp0s3: <BROADCAST,MULTICAST,UP,LOWER_UP,M-DOWN> mtu 1500 qdisc noqueue state UP mode DEFAULT group default qlen 1000
    link/ether b2:a9:c5:04:da:53 brd ff:ff:ff:ff:ff:ff

After this patch:

3: enp0s3: <BROADCAST,MULTICAST> mtu 1500 qdisc pfifo_fast state DOWN mode DEFAULT group default qlen 1000

    link/ether 00:00:05:00:00:09 brd ff:ff:ff:ff:ff:ff
...
5: macvlan0@enp0s3: <NO-CARRIER,BROADCAST,MULTICAST,UP,M-DOWN> mtu 1500 qdisc noqueue state LOWERLAYERDOWN mode DEFAULT group default qlen 1000
    link/ether b2:a9:c5:04:da:53 brd ff:ff:ff:ff:ff:ff

---
Changes since V1:
- rebase
- add ack/review tags

---
 drivers/net/virtio_net.c | 94 +++++++++++++++++++++++++++-------------
 1 file changed, 63 insertions(+), 31 deletions(-)
drivers/net/virtio_net.c +0 -0
index 4a802c0ea2cb..69e4ae353c51 100644
@@ -433,6 +433,12 @@ struct virtnet_info {
      /* The lock to synchronize the access to refill_enabled */
      spinlock_t refill_lock;

drivers/net/virtio_net.c struct virtnet_info { · line 444

Quoted 10 lines
drivers/net/virtio_net.c +6 -0
 
+     /* Is config change enabled? */
+     bool config_change_enabled;
+
+     /* The lock to synchronize the access to config_change_enabled */
+     spinlock_t config_change_lock;
+
      /* Work struct for config space updates */
      struct work_struct config_work;
 

But we already have dev->config_lock and dev->config_enabled.

And it actually works better - instead of discarding config change events it defers them until enabled.

Quoted 7 lines
Quoted 3 lines

Yes but then both virtio-net driver and virtio core can ask to enable and disable and then we need some kind of synchronization which is non-trivial.

Well for core it happens on bring up path before driver works and later on tear down after it is gone.
So I do not think they ever do it at the same time.

For example, there could be a suspend/resume when the admin state is down.

Quoted 7 lines
Quoted 7 lines

And device enabling on the core is different from bringing the device up in the networking subsystem. Here we just delay to deal with the config change interrupt on ndo_open(). (E.g try to ack announce is meaningless when the device is down).

Quoted 6 lines
Quoted 4 lines
Quoted 1 line

Thanks

another thing is that it is better not to re-read all config on link up if there was no config interrupt - less vm exits.

Yes, but it should not matter much as it's done in the ndo_open().

Michael, any more comments on this?

Please confirm if this patch is ok or not. If you prefer to reuse the config_disable() I can change it from a boolean to a counter that allows to be nested.

Thanks

Quoted 7 lines

Thanks

Quoted 4 lines

--
MST

author
#7 lore raw
Quoted 95 lines · Jason Wang
Quoted 93 lines · Jason Wang
Quoted 74 lines · Michael S. Tsirkin
Quoted 72 lines · Jason Wang
Quoted 70 lines · Michael S. Tsirkin
Quoted 51 lines · Jason Wang

This patch synchronize operstate with admin state per RFC2863.

This is done by trying to toggle the carrier upon open/close and synchronize with the config change work. This allows propagate status correctly to stacked devices like:

ip link add link enp0s3 macvlan0 type macvlan
ip link set link enp0s3 down
ip link show

Before this patch:

3: enp0s3: <BROADCAST,MULTICAST> mtu 1500 qdisc pfifo_fast state DOWN mode DEFAULT group default qlen 1000

    link/ether 00:00:05:00:00:09 brd ff:ff:ff:ff:ff:ff
......
5: macvlan0@enp0s3: <BROADCAST,MULTICAST,UP,LOWER_UP,M-DOWN> mtu 1500 qdisc noqueue state UP mode DEFAULT group default qlen 1000
    link/ether b2:a9:c5:04:da:53 brd ff:ff:ff:ff:ff:ff

After this patch:

3: enp0s3: <BROADCAST,MULTICAST> mtu 1500 qdisc pfifo_fast state DOWN mode DEFAULT group default qlen 1000

    link/ether 00:00:05:00:00:09 brd ff:ff:ff:ff:ff:ff
...
5: macvlan0@enp0s3: <NO-CARRIER,BROADCAST,MULTICAST,UP,M-DOWN> mtu 1500 qdisc noqueue state LOWERLAYERDOWN mode DEFAULT group default qlen 1000
    link/ether b2:a9:c5:04:da:53 brd ff:ff:ff:ff:ff:ff

---
Changes since V1:
- rebase
- add ack/review tags

---
 drivers/net/virtio_net.c | 94 +++++++++++++++++++++++++++-------------
 1 file changed, 63 insertions(+), 31 deletions(-)
drivers/net/virtio_net.c +0 -0
index 4a802c0ea2cb..69e4ae353c51 100644
@@ -433,6 +433,12 @@ struct virtnet_info {
      /* The lock to synchronize the access to refill_enabled */
      spinlock_t refill_lock;

drivers/net/virtio_net.c struct virtnet_info { · line 444

Quoted 10 lines
drivers/net/virtio_net.c +6 -0
 
+     /* Is config change enabled? */
+     bool config_change_enabled;
+
+     /* The lock to synchronize the access to config_change_enabled */
+     spinlock_t config_change_lock;
+
      /* Work struct for config space updates */
      struct work_struct config_work;
 

But we already have dev->config_lock and dev->config_enabled.

And it actually works better - instead of discarding config change events it defers them until enabled.

Quoted 7 lines
Quoted 3 lines

Yes but then both virtio-net driver and virtio core can ask to enable and disable and then we need some kind of synchronization which is non-trivial.

Well for core it happens on bring up path before driver works and later on tear down after it is gone.
So I do not think they ever do it at the same time.

For example, there could be a suspend/resume when the admin state is down.

Quoted 7 lines
Quoted 7 lines

And device enabling on the core is different from bringing the device up in the networking subsystem. Here we just delay to deal with the config change interrupt on ndo_open(). (E.g try to ack announce is meaningless when the device is down).

Quoted 8 lines
Quoted 6 lines
Quoted 4 lines
Quoted 1 line

Thanks

another thing is that it is better not to re-read all config on link up if there was no config interrupt - less vm exits.

Yes, but it should not matter much as it's done in the ndo_open().

Michael, any more comments on this?

Gentle ping.

Thanks

Quoted 14 lines

Please confirm if this patch is ok or not. If you prefer to reuse the config_disable() I can change it from a boolean to a counter that allows to be nested.

Thanks

Quoted 7 lines

Thanks

Quoted 4 lines

--
MST

#8 lore raw
In reply to Jason Wang
Quoted 101 lines · Jason Wang
Quoted 93 lines · Jason Wang
Quoted 74 lines · Michael S. Tsirkin
Quoted 72 lines · Jason Wang
Quoted 70 lines · Michael S. Tsirkin
Quoted 51 lines · Jason Wang

This patch synchronize operstate with admin state per RFC2863.

This is done by trying to toggle the carrier upon open/close and synchronize with the config change work. This allows propagate status correctly to stacked devices like:

ip link add link enp0s3 macvlan0 type macvlan
ip link set link enp0s3 down
ip link show

Before this patch:

3: enp0s3: <BROADCAST,MULTICAST> mtu 1500 qdisc pfifo_fast state DOWN mode DEFAULT group default qlen 1000

    link/ether 00:00:05:00:00:09 brd ff:ff:ff:ff:ff:ff
......
5: macvlan0@enp0s3: <BROADCAST,MULTICAST,UP,LOWER_UP,M-DOWN> mtu 1500 qdisc noqueue state UP mode DEFAULT group default qlen 1000
    link/ether b2:a9:c5:04:da:53 brd ff:ff:ff:ff:ff:ff

After this patch:

3: enp0s3: <BROADCAST,MULTICAST> mtu 1500 qdisc pfifo_fast state DOWN mode DEFAULT group default qlen 1000

    link/ether 00:00:05:00:00:09 brd ff:ff:ff:ff:ff:ff
...
5: macvlan0@enp0s3: <NO-CARRIER,BROADCAST,MULTICAST,UP,M-DOWN> mtu 1500 qdisc noqueue state LOWERLAYERDOWN mode DEFAULT group default qlen 1000
    link/ether b2:a9:c5:04:da:53 brd ff:ff:ff:ff:ff:ff

---
Changes since V1:
- rebase
- add ack/review tags

---
 drivers/net/virtio_net.c | 94 +++++++++++++++++++++++++++-------------
 1 file changed, 63 insertions(+), 31 deletions(-)
drivers/net/virtio_net.c +0 -0
index 4a802c0ea2cb..69e4ae353c51 100644
@@ -433,6 +433,12 @@ struct virtnet_info {
      /* The lock to synchronize the access to refill_enabled */
      spinlock_t refill_lock;

drivers/net/virtio_net.c struct virtnet_info { · line 444

Quoted 10 lines
drivers/net/virtio_net.c +6 -0
 
+     /* Is config change enabled? */
+     bool config_change_enabled;
+
+     /* The lock to synchronize the access to config_change_enabled */
+     spinlock_t config_change_lock;
+
      /* Work struct for config space updates */
      struct work_struct config_work;
 

But we already have dev->config_lock and dev->config_enabled.

And it actually works better - instead of discarding config change events it defers them until enabled.

Quoted 7 lines
Quoted 3 lines

Yes but then both virtio-net driver and virtio core can ask to enable and disable and then we need some kind of synchronization which is non-trivial.

Well for core it happens on bring up path before driver works and later on tear down after it is gone.
So I do not think they ever do it at the same time.

For example, there could be a suspend/resume when the admin state is down.

Quoted 7 lines
Quoted 7 lines

And device enabling on the core is different from bringing the device up in the networking subsystem. Here we just delay to deal with the config change interrupt on ndo_open(). (E.g try to ack announce is meaningless when the device is down).

Quoted 6 lines
Quoted 4 lines
Quoted 1 line

Thanks

another thing is that it is better not to re-read all config on link up if there was no config interrupt - less vm exits.

Yes, but it should not matter much as it's done in the ndo_open().

Quoted 7 lines

Michael, any more comments on this?

Please confirm if this patch is ok or not. If you prefer to reuse the config_disable() I can change it from a boolean to a counter that allows to be nested.

Thanks

I think doing this in core and re-using config_lock is a cleaner approach for sure, yes.

For remove we do not need stacking. But yes we need it for freeze.

I am not sure how will a counter work. Let's see the patch.

It might be easier, besides config_enabled, to add config_disabled, document that config_enabled is controlled by core, config_disabled by driver.
And we'd have 2 APIs virtio_config_disable_core and virtio_config_disable_driver.

But I leave that decision in your capable hands.

Quoted 7 lines
Quoted 7 lines

Thanks

Quoted 4 lines

--
MST

author
#9 lore raw
Quoted 118 lines · Michael S. Tsirkin
Quoted 101 lines · Jason Wang
Quoted 93 lines · Jason Wang
Quoted 74 lines · Michael S. Tsirkin
Quoted 72 lines · Jason Wang
Quoted 70 lines · Michael S. Tsirkin
Quoted 51 lines · Jason Wang

This patch synchronize operstate with admin state per RFC2863.

This is done by trying to toggle the carrier upon open/close and synchronize with the config change work. This allows propagate status correctly to stacked devices like:

ip link add link enp0s3 macvlan0 type macvlan
ip link set link enp0s3 down
ip link show

Before this patch:

3: enp0s3: <BROADCAST,MULTICAST> mtu 1500 qdisc pfifo_fast state DOWN mode DEFAULT group default qlen 1000

    link/ether 00:00:05:00:00:09 brd ff:ff:ff:ff:ff:ff
......
5: macvlan0@enp0s3: <BROADCAST,MULTICAST,UP,LOWER_UP,M-DOWN> mtu 1500 qdisc noqueue state UP mode DEFAULT group default qlen 1000
    link/ether b2:a9:c5:04:da:53 brd ff:ff:ff:ff:ff:ff

After this patch:

3: enp0s3: <BROADCAST,MULTICAST> mtu 1500 qdisc pfifo_fast state DOWN mode DEFAULT group default qlen 1000

    link/ether 00:00:05:00:00:09 brd ff:ff:ff:ff:ff:ff
...
5: macvlan0@enp0s3: <NO-CARRIER,BROADCAST,MULTICAST,UP,M-DOWN> mtu 1500 qdisc noqueue state LOWERLAYERDOWN mode DEFAULT group default qlen 1000
    link/ether b2:a9:c5:04:da:53 brd ff:ff:ff:ff:ff:ff

---
Changes since V1:
- rebase
- add ack/review tags

---
 drivers/net/virtio_net.c | 94 +++++++++++++++++++++++++++-------------
 1 file changed, 63 insertions(+), 31 deletions(-)
drivers/net/virtio_net.c +0 -0
index 4a802c0ea2cb..69e4ae353c51 100644
@@ -433,6 +433,12 @@ struct virtnet_info {
      /* The lock to synchronize the access to refill_enabled */
      spinlock_t refill_lock;

drivers/net/virtio_net.c struct virtnet_info { · line 444

Quoted 10 lines
drivers/net/virtio_net.c +6 -0
 
+     /* Is config change enabled? */
+     bool config_change_enabled;
+
+     /* The lock to synchronize the access to config_change_enabled */
+     spinlock_t config_change_lock;
+
      /* Work struct for config space updates */
      struct work_struct config_work;
 

But we already have dev->config_lock and dev->config_enabled.

And it actually works better - instead of discarding config change events it defers them until enabled.

Quoted 7 lines
Quoted 3 lines

Yes but then both virtio-net driver and virtio core can ask to enable and disable and then we need some kind of synchronization which is non-trivial.

Well for core it happens on bring up path before driver works and later on tear down after it is gone.
So I do not think they ever do it at the same time.

For example, there could be a suspend/resume when the admin state is down.

Quoted 7 lines
Quoted 7 lines

And device enabling on the core is different from bringing the device up in the networking subsystem. Here we just delay to deal with the config change interrupt on ndo_open(). (E.g try to ack announce is meaningless when the device is down).

Quoted 6 lines
Quoted 4 lines
Quoted 1 line

Thanks

another thing is that it is better not to re-read all config on link up if there was no config interrupt - less vm exits.

Yes, but it should not matter much as it's done in the ndo_open().

Quoted 7 lines

Michael, any more comments on this?

Please confirm if this patch is ok or not. If you prefer to reuse the config_disable() I can change it from a boolean to a counter that allows to be nested.

Thanks

I think doing this in core and re-using config_lock is a cleaner approach for sure, yes.

For remove we do not need stacking. But yes we need it for freeze.

I am not sure how will a counter work. Let's see the patch.

Quoted 8 lines

It might be easier, besides config_enabled, to add config_disabled, document that config_enabled is controlled by core, config_disabled by driver.
And we'd have 2 APIs virtio_config_disable_core and virtio_config_disable_driver.

But I leave that decision in your capable hands.

Let me try to switch to using a counter for config_enabled first.

Thanks

Quoted 11 lines
Quoted 10 lines
Quoted 10 lines

Thanks

Quoted 4 lines

--
MST