[PATCH net] bonding: Fix stacked device detection in arp monitoring

[PATCH net] bonding: Fix stacked device detection in arp monitoring

[Vlad Yasevich]

Prior to commit fbd929f2dce460456807a51e18d623db3db9f077
	bonding: support QinQ for bond arp interval

the arp monitoring code allowed for proper detection of devices
stacked on top of vlans.  Since the above commit, the
code can still detect a device stacked on top of single
vlan, but not a device stacked on top of Q-in-Q configuration.
The search will only set the inner vlan tag if the route
device is the vlan device.  However, this is not always the
case, as it is possible to extend the stacked configuration.

With this patch it is possible to provision devices on
top Q-in-Q vlan configuration that should be used as
a source of ARP monitoring information.

For example:
ip link add link bond0 vlan10 type vlan proto 802.1q id 10
ip link add link vlan10 vlan100 type vlan proto 802.1q id 100
ip link add link vlan100 type macvlan

Note:  This patch limits the number of stacked VLANs to 2,
just like before.  The original, however had another issue
in that if we had more then 2 levels of VLANs, we would end
up generating incorrectly tagged traffic.  This is no longer
possible.

Fixes: fbd929f2dce460456807a51e18d623db3db9f077 (bonding: support QinQ for bond arp interval)
CC: Jay Vosburgh <fubar@us.ibm.com>
CC: Veaceslav Falico <vfalico@redhat.com>
CC: Andy Gospodarek <andy@greyhouse.net>
CC: Ding Tianhong <dingtianhong@huawei.com>
CC: Patric McHardy <kaber@trash.net>
Signed-off-by: Vlad Yasevich <vyasevic@redhat.com>
---
 drivers/net/bonding/bond_main.c | 90 +++++++++++++++++------------------------
 1 file changed, 37 insertions(+), 53 deletions(-)

diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c
index 9d08e00..4822728 100644
--- a/drivers/net/bonding/bond_main.c
+++ b/drivers/net/bonding/bond_main.c
@@ -2164,22 +2164,46 @@ static void bond_arp_send(struct net_device *slave_dev, int arp_op,
 	arp_xmit(skb);
 }
 
+static bool bond_collect_vlans(struct net_device *start, struct net_device *end,
+				struct bond_vlan_tag *tag, int idx)
+{
+	struct net_device *upper;
+	struct list_head  *iter;
+
+	/* We do not support more then 2 levels of VLAN nesting */
+	if (idx >= 2)
+		return false;
+
+	netdev_for_each_all_upper_dev_rcu(start, upper, iter) {
+		if (is_vlan_dev(upper)) {
+			tag[idx].vlan_proto = vlan_dev_vlan_proto(upper);
+			tag[idx].vlan_id = vlan_dev_vlan_id(upper);
+			idx++;
+		}
+		if (upper == end)
+			return true;
+
+		/* Look at upper devices list only if current device
+		 * is a vlan
+		 */
+		if (is_vlan_dev(upper) && bond_collect_vlans(upper, end, tag, idx))
+			return true;
+	}
+	return false;
+}
+
 
 static void bond_arp_send_all(struct bonding *bond, struct slave *slave)
 {
-	struct net_device *upper, *vlan_upper;
-	struct list_head *iter, *vlan_iter;
 	struct rtable *rt;
-	struct bond_vlan_tag inner, outer;
+	struct bond_vlan_tag tags[2];
 	__be32 *targets = bond->params.arp_targets, addr;
 	int i;
+	bool ret;
 
 	for (i = 0; i < BOND_MAX_ARP_TARGETS && targets[i]; i++) {
 		pr_debug("basa: target %pI4\n", &targets[i]);
-		inner.vlan_proto = 0;
-		inner.vlan_id = 0;
-		outer.vlan_proto = 0;
-		outer.vlan_id = 0;
+		memset(tags, 0, sizeof(tags));
 
 		/* Find out through which dev should the packet go */
 		rt = ip_route_output(dev_net(bond->dev), targets[i], 0,
@@ -2192,7 +2216,7 @@ static void bond_arp_send_all(struct bonding *bond, struct slave *slave)
 				net_warn_ratelimited("%s: no route to arp_ip_target %pI4 and arp_validate is set\n",
 						     bond->dev->name,
 						     &targets[i]);
-			bond_arp_send(slave->dev, ARPOP_REQUEST, targets[i], 0, &inner, &outer);
+			bond_arp_send(slave->dev, ARPOP_REQUEST, targets[i], 0, &tags[1], &tags[0]);
 			continue;
 		}
 
@@ -2201,52 +2225,12 @@ static void bond_arp_send_all(struct bonding *bond, struct slave *slave)
 			goto found;
 
 		rcu_read_lock();
-		/* first we search only for vlan devices. for every vlan
-		 * found we verify its upper dev list, searching for the
-		 * rt->dst.dev. If found we save the tag of the vlan and
-		 * proceed to send the packet.
-		 */
-		netdev_for_each_all_upper_dev_rcu(bond->dev, vlan_upper,
-						  vlan_iter) {
-			if (!is_vlan_dev(vlan_upper))
-				continue;
-
-			if (vlan_upper == rt->dst.dev) {
-				outer.vlan_proto = vlan_dev_vlan_proto(vlan_upper);
-				outer.vlan_id = vlan_dev_vlan_id(vlan_upper);
-				rcu_read_unlock();
-				goto found;
-			}
-			netdev_for_each_all_upper_dev_rcu(vlan_upper, upper,
-							  iter) {
-				if (upper == rt->dst.dev) {
-					/* If the upper dev is a vlan dev too,
-					 *  set the vlan tag to inner tag.
-					 */
-					if (is_vlan_dev(upper)) {
-						inner.vlan_proto = vlan_dev_vlan_proto(upper);
-						inner.vlan_id = vlan_dev_vlan_id(upper);
-					}
-					outer.vlan_proto = vlan_dev_vlan_proto(vlan_upper);
-					outer.vlan_id = vlan_dev_vlan_id(vlan_upper);
-					rcu_read_unlock();
-					goto found;
-				}
-			}
-		}
-
-		/* if the device we're looking for is not on top of any of
-		 * our upper vlans, then just search for any dev that
-		 * matches, and in case it's a vlan - save the id
-		 */
-		netdev_for_each_all_upper_dev_rcu(bond->dev, upper, iter) {
-			if (upper == rt->dst.dev) {
-				rcu_read_unlock();
-				goto found;
-			}
-		}
+		ret = bond_collect_vlans(bond->dev, rt->dst.dev, tags, 0);
 		rcu_read_unlock();
 
+		if (ret)
+			goto found;
+
 		/* Not our device - skip */
 		pr_debug("%s: no path to arp_ip_target %pI4 via rt.dev %s\n",
 			 bond->dev->name, &targets[i],
@@ -2259,7 +2243,7 @@ found:
 		addr = bond_confirm_addr(rt->dst.dev, targets[i], 0);
 		ip_rt_put(rt);
 		bond_arp_send(slave->dev, ARPOP_REQUEST, targets[i],
-			      addr, &inner, &outer);
+			      addr, &tags[1], &tags[0]);
 	}
 }
 
-- 
1.9.0

Re: [PATCH net] bonding: Fix stacked device detection in arp monitoring

[Jay Vosburgh]

Vlad Yasevich <vyasevic@redhat.com> wrote:

>Prior to commit fbd929f2dce460456807a51e18d623db3db9f077
>	bonding: support QinQ for bond arp interval
>
>the arp monitoring code allowed for proper detection of devices
>stacked on top of vlans.  Since the above commit, the
>code can still detect a device stacked on top of single
>vlan, but not a device stacked on top of Q-in-Q configuration.
>The search will only set the inner vlan tag if the route
>device is the vlan device.  However, this is not always the
>case, as it is possible to extend the stacked configuration.
>
>With this patch it is possible to provision devices on
>top Q-in-Q vlan configuration that should be used as
>a source of ARP monitoring information.
>
>For example:
>ip link add link bond0 vlan10 type vlan proto 802.1q id 10
>ip link add link vlan10 vlan100 type vlan proto 802.1q id 100
>ip link add link vlan100 type macvlan
>
>Note:  This patch limits the number of stacked VLANs to 2,
>just like before.  The original, however had another issue
>in that if we had more then 2 levels of VLANs, we would end
>up generating incorrectly tagged traffic.  This is no longer
>possible.

	I like this in general, with one minor exception: we lose the
"inner" and "outer" nomenclature, which I think made the code clearer.
I think the bond_collect_vlans function deserves at least a sentence or
two of commentary explaining that we're walking up the layers from
bottom to top, and collecting the outer tag and then the inner tag,
thus, outer is tags[0] and inner is tags[1].

	And, just as a random observation, it looks like it will do the
right thing (or at least not crash) if things are layered in odd ways,
e.g., if the macvlan is nested between the vlans.

	-J

>Fixes: fbd929f2dce460456807a51e18d623db3db9f077 (bonding: support QinQ for bond arp interval)
>CC: Jay Vosburgh <fubar@us.ibm.com>
>CC: Veaceslav Falico <vfalico@redhat.com>
>CC: Andy Gospodarek <andy@greyhouse.net>
>CC: Ding Tianhong <dingtianhong@huawei.com>
>CC: Patric McHardy <kaber@trash.net>
>Signed-off-by: Vlad Yasevich <vyasevic@redhat.com>
>---
> drivers/net/bonding/bond_main.c | 90 +++++++++++++++++------------------------
> 1 file changed, 37 insertions(+), 53 deletions(-)
>
>diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c
>index 9d08e00..4822728 100644
>--- a/drivers/net/bonding/bond_main.c
>+++ b/drivers/net/bonding/bond_main.c
>@@ -2164,22 +2164,46 @@ static void bond_arp_send(struct net_device *slave_dev, int arp_op,
> 	arp_xmit(skb);
> }
> 
>+static bool bond_collect_vlans(struct net_device *start, struct net_device *end,
>+				struct bond_vlan_tag *tag, int idx)
>+{
>+	struct net_device *upper;
>+	struct list_head  *iter;
>+
>+	/* We do not support more then 2 levels of VLAN nesting */
>+	if (idx >= 2)
>+		return false;
>+
>+	netdev_for_each_all_upper_dev_rcu(start, upper, iter) {
>+		if (is_vlan_dev(upper)) {
>+			tag[idx].vlan_proto = vlan_dev_vlan_proto(upper);
>+			tag[idx].vlan_id = vlan_dev_vlan_id(upper);
>+			idx++;
>+		}
>+		if (upper == end)
>+			return true;
>+
>+		/* Look at upper devices list only if current device
>+		 * is a vlan
>+		 */
>+		if (is_vlan_dev(upper) && bond_collect_vlans(upper, end, tag, idx))
>+			return true;
>+	}
>+	return false;
>+}
>+
> 
> static void bond_arp_send_all(struct bonding *bond, struct slave *slave)
> {
>-	struct net_device *upper, *vlan_upper;
>-	struct list_head *iter, *vlan_iter;
> 	struct rtable *rt;
>-	struct bond_vlan_tag inner, outer;
>+	struct bond_vlan_tag tags[2];
> 	__be32 *targets = bond->params.arp_targets, addr;
> 	int i;
>+	bool ret;
> 
> 	for (i = 0; i < BOND_MAX_ARP_TARGETS && targets[i]; i++) {
> 		pr_debug("basa: target %pI4\n", &targets[i]);
>-		inner.vlan_proto = 0;
>-		inner.vlan_id = 0;
>-		outer.vlan_proto = 0;
>-		outer.vlan_id = 0;
>+		memset(tags, 0, sizeof(tags));
> 
> 		/* Find out through which dev should the packet go */
> 		rt = ip_route_output(dev_net(bond->dev), targets[i], 0,
>@@ -2192,7 +2216,7 @@ static void bond_arp_send_all(struct bonding *bond, struct slave *slave)
> 				net_warn_ratelimited("%s: no route to arp_ip_target %pI4 and arp_validate is set\n",
> 						     bond->dev->name,
> 						     &targets[i]);
>-			bond_arp_send(slave->dev, ARPOP_REQUEST, targets[i], 0, &inner, &outer);
>+			bond_arp_send(slave->dev, ARPOP_REQUEST, targets[i], 0, &tags[1], &tags[0]);
> 			continue;
> 		}
> 
>@@ -2201,52 +2225,12 @@ static void bond_arp_send_all(struct bonding *bond, struct slave *slave)
> 			goto found;
> 
> 		rcu_read_lock();
>-		/* first we search only for vlan devices. for every vlan
>-		 * found we verify its upper dev list, searching for the
>-		 * rt->dst.dev. If found we save the tag of the vlan and
>-		 * proceed to send the packet.
>-		 */
>-		netdev_for_each_all_upper_dev_rcu(bond->dev, vlan_upper,
>-						  vlan_iter) {
>-			if (!is_vlan_dev(vlan_upper))
>-				continue;
>-
>-			if (vlan_upper == rt->dst.dev) {
>-				outer.vlan_proto = vlan_dev_vlan_proto(vlan_upper);
>-				outer.vlan_id = vlan_dev_vlan_id(vlan_upper);
>-				rcu_read_unlock();
>-				goto found;
>-			}
>-			netdev_for_each_all_upper_dev_rcu(vlan_upper, upper,
>-							  iter) {
>-				if (upper == rt->dst.dev) {
>-					/* If the upper dev is a vlan dev too,
>-					 *  set the vlan tag to inner tag.
>-					 */
>-					if (is_vlan_dev(upper)) {
>-						inner.vlan_proto = vlan_dev_vlan_proto(upper);
>-						inner.vlan_id = vlan_dev_vlan_id(upper);
>-					}
>-					outer.vlan_proto = vlan_dev_vlan_proto(vlan_upper);
>-					outer.vlan_id = vlan_dev_vlan_id(vlan_upper);
>-					rcu_read_unlock();
>-					goto found;
>-				}
>-			}
>-		}
>-
>-		/* if the device we're looking for is not on top of any of
>-		 * our upper vlans, then just search for any dev that
>-		 * matches, and in case it's a vlan - save the id
>-		 */
>-		netdev_for_each_all_upper_dev_rcu(bond->dev, upper, iter) {
>-			if (upper == rt->dst.dev) {
>-				rcu_read_unlock();
>-				goto found;
>-			}
>-		}
>+		ret = bond_collect_vlans(bond->dev, rt->dst.dev, tags, 0);
> 		rcu_read_unlock();
> 
>+		if (ret)
>+			goto found;
>+
> 		/* Not our device - skip */
> 		pr_debug("%s: no path to arp_ip_target %pI4 via rt.dev %s\n",
> 			 bond->dev->name, &targets[i],
>@@ -2259,7 +2243,7 @@ found:
> 		addr = bond_confirm_addr(rt->dst.dev, targets[i], 0);
> 		ip_rt_put(rt);
> 		bond_arp_send(slave->dev, ARPOP_REQUEST, targets[i],
>-			      addr, &inner, &outer);
>+			      addr, &tags[1], &tags[0]);
> 	}
> }

---
	-Jay Vosburgh, jay.vosburgh@canonical.com

Re: [PATCH net] bonding: Fix stacked device detection in arp monitoring

[Vlad Yasevich]

On 05/06/2014 03:22 PM, Jay Vosburgh wrote:
> Vlad Yasevich <vyasevic@redhat.com> wrote:
> 
>> Prior to commit fbd929f2dce460456807a51e18d623db3db9f077
>> 	bonding: support QinQ for bond arp interval
>>
>> the arp monitoring code allowed for proper detection of devices
>> stacked on top of vlans.  Since the above commit, the
>> code can still detect a device stacked on top of single
>> vlan, but not a device stacked on top of Q-in-Q configuration.
>> The search will only set the inner vlan tag if the route
>> device is the vlan device.  However, this is not always the
>> case, as it is possible to extend the stacked configuration.
>>
>> With this patch it is possible to provision devices on
>> top Q-in-Q vlan configuration that should be used as
>> a source of ARP monitoring information.
>>
>> For example:
>> ip link add link bond0 vlan10 type vlan proto 802.1q id 10
>> ip link add link vlan10 vlan100 type vlan proto 802.1q id 100
>> ip link add link vlan100 type macvlan
>>
>> Note:  This patch limits the number of stacked VLANs to 2,
>> just like before.  The original, however had another issue
>> in that if we had more then 2 levels of VLANs, we would end
>> up generating incorrectly tagged traffic.  This is no longer
>> possible.
> 
> 	I like this in general, with one minor exception: we lose the
> "inner" and "outer" nomenclature, which I think made the code clearer.

I can completely remove the inner/outer nomenclature if you wish, or can
save it for follow-on patch.

> I think the bond_collect_vlans function deserves at least a sentence or
> two of commentary explaining that we're walking up the layers from
> bottom to top, and collecting the outer tag and then the inner tag,
> thus, outer is tags[0] and inner is tags[1].

Will do.

> 
> 	And, just as a random observation, it looks like it will do the
> right thing (or at least not crash) if things are layered in odd ways,
> e.g., if the macvlan is nested between the vlans.

Right.  There is a slight bug though that I found last night.  I'll fix
that in v2 as well.

Thanks
-vlad

> 
> 	-J
> 
>> Fixes: fbd929f2dce460456807a51e18d623db3db9f077 (bonding: support QinQ for bond arp interval)
>> CC: Jay Vosburgh <fubar@us.ibm.com>
>> CC: Veaceslav Falico <vfalico@redhat.com>
>> CC: Andy Gospodarek <andy@greyhouse.net>
>> CC: Ding Tianhong <dingtianhong@huawei.com>
>> CC: Patric McHardy <kaber@trash.net>
>> Signed-off-by: Vlad Yasevich <vyasevic@redhat.com>
>> ---
>> drivers/net/bonding/bond_main.c | 90 +++++++++++++++++------------------------
>> 1 file changed, 37 insertions(+), 53 deletions(-)
>>
>> diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c
>> index 9d08e00..4822728 100644
>> --- a/drivers/net/bonding/bond_main.c
>> +++ b/drivers/net/bonding/bond_main.c
>> @@ -2164,22 +2164,46 @@ static void bond_arp_send(struct net_device *slave_dev, int arp_op,
>> 	arp_xmit(skb);
>> }
>>
>> +static bool bond_collect_vlans(struct net_device *start, struct net_device *end,
>> +				struct bond_vlan_tag *tag, int idx)
>> +{
>> +	struct net_device *upper;
>> +	struct list_head  *iter;
>> +
>> +	/* We do not support more then 2 levels of VLAN nesting */
>> +	if (idx >= 2)
>> +		return false;
>> +
>> +	netdev_for_each_all_upper_dev_rcu(start, upper, iter) {
>> +		if (is_vlan_dev(upper)) {
>> +			tag[idx].vlan_proto = vlan_dev_vlan_proto(upper);
>> +			tag[idx].vlan_id = vlan_dev_vlan_id(upper);
>> +			idx++;
>> +		}
>> +		if (upper == end)
>> +			return true;
>> +
>> +		/* Look at upper devices list only if current device
>> +		 * is a vlan
>> +		 */
>> +		if (is_vlan_dev(upper) && bond_collect_vlans(upper, end, tag, idx))
>> +			return true;
>> +	}
>> +	return false;
>> +}
>> +
>>
>> static void bond_arp_send_all(struct bonding *bond, struct slave *slave)
>> {
>> -	struct net_device *upper, *vlan_upper;
>> -	struct list_head *iter, *vlan_iter;
>> 	struct rtable *rt;
>> -	struct bond_vlan_tag inner, outer;
>> +	struct bond_vlan_tag tags[2];
>> 	__be32 *targets = bond->params.arp_targets, addr;
>> 	int i;
>> +	bool ret;
>>
>> 	for (i = 0; i < BOND_MAX_ARP_TARGETS && targets[i]; i++) {
>> 		pr_debug("basa: target %pI4\n", &targets[i]);
>> -		inner.vlan_proto = 0;
>> -		inner.vlan_id = 0;
>> -		outer.vlan_proto = 0;
>> -		outer.vlan_id = 0;
>> +		memset(tags, 0, sizeof(tags));
>>
>> 		/* Find out through which dev should the packet go */
>> 		rt = ip_route_output(dev_net(bond->dev), targets[i], 0,
>> @@ -2192,7 +2216,7 @@ static void bond_arp_send_all(struct bonding *bond, struct slave *slave)
>> 				net_warn_ratelimited("%s: no route to arp_ip_target %pI4 and arp_validate is set\n",
>> 						     bond->dev->name,
>> 						     &targets[i]);
>> -			bond_arp_send(slave->dev, ARPOP_REQUEST, targets[i], 0, &inner, &outer);
>> +			bond_arp_send(slave->dev, ARPOP_REQUEST, targets[i], 0, &tags[1], &tags[0]);
>> 			continue;
>> 		}
>>
>> @@ -2201,52 +2225,12 @@ static void bond_arp_send_all(struct bonding *bond, struct slave *slave)
>> 			goto found;
>>
>> 		rcu_read_lock();
>> -		/* first we search only for vlan devices. for every vlan
>> -		 * found we verify its upper dev list, searching for the
>> -		 * rt->dst.dev. If found we save the tag of the vlan and
>> -		 * proceed to send the packet.
>> -		 */
>> -		netdev_for_each_all_upper_dev_rcu(bond->dev, vlan_upper,
>> -						  vlan_iter) {
>> -			if (!is_vlan_dev(vlan_upper))
>> -				continue;
>> -
>> -			if (vlan_upper == rt->dst.dev) {
>> -				outer.vlan_proto = vlan_dev_vlan_proto(vlan_upper);
>> -				outer.vlan_id = vlan_dev_vlan_id(vlan_upper);
>> -				rcu_read_unlock();
>> -				goto found;
>> -			}
>> -			netdev_for_each_all_upper_dev_rcu(vlan_upper, upper,
>> -							  iter) {
>> -				if (upper == rt->dst.dev) {
>> -					/* If the upper dev is a vlan dev too,
>> -					 *  set the vlan tag to inner tag.
>> -					 */
>> -					if (is_vlan_dev(upper)) {
>> -						inner.vlan_proto = vlan_dev_vlan_proto(upper);
>> -						inner.vlan_id = vlan_dev_vlan_id(upper);
>> -					}
>> -					outer.vlan_proto = vlan_dev_vlan_proto(vlan_upper);
>> -					outer.vlan_id = vlan_dev_vlan_id(vlan_upper);
>> -					rcu_read_unlock();
>> -					goto found;
>> -				}
>> -			}
>> -		}
>> -
>> -		/* if the device we're looking for is not on top of any of
>> -		 * our upper vlans, then just search for any dev that
>> -		 * matches, and in case it's a vlan - save the id
>> -		 */
>> -		netdev_for_each_all_upper_dev_rcu(bond->dev, upper, iter) {
>> -			if (upper == rt->dst.dev) {
>> -				rcu_read_unlock();
>> -				goto found;
>> -			}
>> -		}
>> +		ret = bond_collect_vlans(bond->dev, rt->dst.dev, tags, 0);
>> 		rcu_read_unlock();
>>
>> +		if (ret)
>> +			goto found;
>> +
>> 		/* Not our device - skip */
>> 		pr_debug("%s: no path to arp_ip_target %pI4 via rt.dev %s\n",
>> 			 bond->dev->name, &targets[i],
>> @@ -2259,7 +2243,7 @@ found:
>> 		addr = bond_confirm_addr(rt->dst.dev, targets[i], 0);
>> 		ip_rt_put(rt);
>> 		bond_arp_send(slave->dev, ARPOP_REQUEST, targets[i],
>> -			      addr, &inner, &outer);
>> +			      addr, &tags[1], &tags[0]);
>> 	}
>> }
> 
> ---
> 	-Jay Vosburgh, jay.vosburgh@canonical.com
> 

Re: [PATCH net] bonding: Fix stacked device detection in arp monitoring

[Veaceslav Falico]

On Tue, May 06, 2014 at 01:41:15PM -0400, Vlad Yasevich wrote:
>Prior to commit fbd929f2dce460456807a51e18d623db3db9f077
>	bonding: support QinQ for bond arp interval
>
>the arp monitoring code allowed for proper detection of devices
>stacked on top of vlans.  Since the above commit, the
>code can still detect a device stacked on top of single
>vlan, but not a device stacked on top of Q-in-Q configuration.
>The search will only set the inner vlan tag if the route
>device is the vlan device.  However, this is not always the
>case, as it is possible to extend the stacked configuration.
>
>With this patch it is possible to provision devices on
>top Q-in-Q vlan configuration that should be used as
>a source of ARP monitoring information.
>
>For example:
>ip link add link bond0 vlan10 type vlan proto 802.1q id 10
>ip link add link vlan10 vlan100 type vlan proto 802.1q id 100
>ip link add link vlan100 type macvlan
>
>Note:  This patch limits the number of stacked VLANs to 2,
>just like before.  The original, however had another issue
>in that if we had more then 2 levels of VLANs, we would end
>up generating incorrectly tagged traffic.  This is no longer
>possible.
>
>Fixes: fbd929f2dce460456807a51e18d623db3db9f077 (bonding: support QinQ for bond arp interval)
>CC: Jay Vosburgh <fubar@us.ibm.com>
>CC: Veaceslav Falico <vfalico@redhat.com>
>CC: Andy Gospodarek <andy@greyhouse.net>
>CC: Ding Tianhong <dingtianhong@huawei.com>
>CC: Patric McHardy <kaber@trash.net>
>Signed-off-by: Vlad Yasevich <vyasevic@redhat.com>
>---
> drivers/net/bonding/bond_main.c | 90 +++++++++++++++++------------------------
> 1 file changed, 37 insertions(+), 53 deletions(-)
>
>diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c
>index 9d08e00..4822728 100644
>--- a/drivers/net/bonding/bond_main.c
>+++ b/drivers/net/bonding/bond_main.c
>@@ -2164,22 +2164,46 @@ static void bond_arp_send(struct net_device *slave_dev, int arp_op,
> 	arp_xmit(skb);
> }
>
>+static bool bond_collect_vlans(struct net_device *start, struct net_device *end,
>+				struct bond_vlan_tag *tag, int idx)
>+{
>+	struct net_device *upper;
>+	struct list_head  *iter;
>+
>+	/* We do not support more then 2 levels of VLAN nesting */
>+	if (idx >= 2)
>+		return false;
>+
>+	netdev_for_each_all_upper_dev_rcu(start, upper, iter) {
>+		if (is_vlan_dev(upper)) {
>+			tag[idx].vlan_proto = vlan_dev_vlan_proto(upper);
>+			tag[idx].vlan_id = vlan_dev_vlan_id(upper);
>+			idx++;

Won't idx here skyrocket if we have more than 1 vlan on top of bonding?

Otherwise I like this approach, from the first view.

>+		}
>+		if (upper == end)
>+			return true;
>+
>+		/* Look at upper devices list only if current device
>+		 * is a vlan
>+		 */
>+		if (is_vlan_dev(upper) && bond_collect_vlans(upper, end, tag, idx))
>+			return true;
>+	}
>+	return false;
>+}
>+
>
> static void bond_arp_send_all(struct bonding *bond, struct slave *slave)
> {
>-	struct net_device *upper, *vlan_upper;
>-	struct list_head *iter, *vlan_iter;
> 	struct rtable *rt;
>-	struct bond_vlan_tag inner, outer;
>+	struct bond_vlan_tag tags[2];
> 	__be32 *targets = bond->params.arp_targets, addr;
> 	int i;
>+	bool ret;
>
> 	for (i = 0; i < BOND_MAX_ARP_TARGETS && targets[i]; i++) {
> 		pr_debug("basa: target %pI4\n", &targets[i]);
>-		inner.vlan_proto = 0;
>-		inner.vlan_id = 0;
>-		outer.vlan_proto = 0;
>-		outer.vlan_id = 0;
>+		memset(tags, 0, sizeof(tags));
>
> 		/* Find out through which dev should the packet go */
> 		rt = ip_route_output(dev_net(bond->dev), targets[i], 0,
>@@ -2192,7 +2216,7 @@ static void bond_arp_send_all(struct bonding *bond, struct slave *slave)
> 				net_warn_ratelimited("%s: no route to arp_ip_target %pI4 and arp_validate is set\n",
> 						     bond->dev->name,
> 						     &targets[i]);
>-			bond_arp_send(slave->dev, ARPOP_REQUEST, targets[i], 0, &inner, &outer);
>+			bond_arp_send(slave->dev, ARPOP_REQUEST, targets[i], 0, &tags[1], &tags[0]);
> 			continue;
> 		}
>
>@@ -2201,52 +2225,12 @@ static void bond_arp_send_all(struct bonding *bond, struct slave *slave)
> 			goto found;
>
> 		rcu_read_lock();
>-		/* first we search only for vlan devices. for every vlan
>-		 * found we verify its upper dev list, searching for the
>-		 * rt->dst.dev. If found we save the tag of the vlan and
>-		 * proceed to send the packet.
>-		 */
>-		netdev_for_each_all_upper_dev_rcu(bond->dev, vlan_upper,
>-						  vlan_iter) {
>-			if (!is_vlan_dev(vlan_upper))
>-				continue;
>-
>-			if (vlan_upper == rt->dst.dev) {
>-				outer.vlan_proto = vlan_dev_vlan_proto(vlan_upper);
>-				outer.vlan_id = vlan_dev_vlan_id(vlan_upper);
>-				rcu_read_unlock();
>-				goto found;
>-			}
>-			netdev_for_each_all_upper_dev_rcu(vlan_upper, upper,
>-							  iter) {
>-				if (upper == rt->dst.dev) {
>-					/* If the upper dev is a vlan dev too,
>-					 *  set the vlan tag to inner tag.
>-					 */
>-					if (is_vlan_dev(upper)) {
>-						inner.vlan_proto = vlan_dev_vlan_proto(upper);
>-						inner.vlan_id = vlan_dev_vlan_id(upper);
>-					}
>-					outer.vlan_proto = vlan_dev_vlan_proto(vlan_upper);
>-					outer.vlan_id = vlan_dev_vlan_id(vlan_upper);
>-					rcu_read_unlock();
>-					goto found;
>-				}
>-			}
>-		}
>-
>-		/* if the device we're looking for is not on top of any of
>-		 * our upper vlans, then just search for any dev that
>-		 * matches, and in case it's a vlan - save the id
>-		 */
>-		netdev_for_each_all_upper_dev_rcu(bond->dev, upper, iter) {
>-			if (upper == rt->dst.dev) {
>-				rcu_read_unlock();
>-				goto found;
>-			}
>-		}
>+		ret = bond_collect_vlans(bond->dev, rt->dst.dev, tags, 0);
> 		rcu_read_unlock();
>
>+		if (ret)
>+			goto found;
>+
> 		/* Not our device - skip */
> 		pr_debug("%s: no path to arp_ip_target %pI4 via rt.dev %s\n",
> 			 bond->dev->name, &targets[i],
>@@ -2259,7 +2243,7 @@ found:
> 		addr = bond_confirm_addr(rt->dst.dev, targets[i], 0);
> 		ip_rt_put(rt);
> 		bond_arp_send(slave->dev, ARPOP_REQUEST, targets[i],
>-			      addr, &inner, &outer);
>+			      addr, &tags[1], &tags[0]);
> 	}
> }
>
>-- 
>1.9.0
>

Re: [PATCH net] bonding: Fix stacked device detection in arp monitoring

[Vlad Yasevich]

On 05/07/2014 09:32 AM, Veaceslav Falico wrote:
> On Tue, May 06, 2014 at 01:41:15PM -0400, Vlad Yasevich wrote:
>> Prior to commit fbd929f2dce460456807a51e18d623db3db9f077
>>     bonding: support QinQ for bond arp interval
>>
>> the arp monitoring code allowed for proper detection of devices
>> stacked on top of vlans.  Since the above commit, the
>> code can still detect a device stacked on top of single
>> vlan, but not a device stacked on top of Q-in-Q configuration.
>> The search will only set the inner vlan tag if the route
>> device is the vlan device.  However, this is not always the
>> case, as it is possible to extend the stacked configuration.
>>
>> With this patch it is possible to provision devices on
>> top Q-in-Q vlan configuration that should be used as
>> a source of ARP monitoring information.
>>
>> For example:
>> ip link add link bond0 vlan10 type vlan proto 802.1q id 10
>> ip link add link vlan10 vlan100 type vlan proto 802.1q id 100
>> ip link add link vlan100 type macvlan
>>
>> Note:  This patch limits the number of stacked VLANs to 2,
>> just like before.  The original, however had another issue
>> in that if we had more then 2 levels of VLANs, we would end
>> up generating incorrectly tagged traffic.  This is no longer
>> possible.
>>
>> Fixes: fbd929f2dce460456807a51e18d623db3db9f077 (bonding: support QinQ
>> for bond arp interval)
>> CC: Jay Vosburgh <fubar@us.ibm.com>
>> CC: Veaceslav Falico <vfalico@redhat.com>
>> CC: Andy Gospodarek <andy@greyhouse.net>
>> CC: Ding Tianhong <dingtianhong@huawei.com>
>> CC: Patric McHardy <kaber@trash.net>
>> Signed-off-by: Vlad Yasevich <vyasevic@redhat.com>
>> ---
>> drivers/net/bonding/bond_main.c | 90
>> +++++++++++++++++------------------------
>> 1 file changed, 37 insertions(+), 53 deletions(-)
>>
>> diff --git a/drivers/net/bonding/bond_main.c
>> b/drivers/net/bonding/bond_main.c
>> index 9d08e00..4822728 100644
>> --- a/drivers/net/bonding/bond_main.c
>> +++ b/drivers/net/bonding/bond_main.c
>> @@ -2164,22 +2164,46 @@ static void bond_arp_send(struct net_device
>> *slave_dev, int arp_op,
>>     arp_xmit(skb);
>> }
>>
>> +static bool bond_collect_vlans(struct net_device *start, struct
>> net_device *end,
>> +                struct bond_vlan_tag *tag, int idx)
>> +{
>> +    struct net_device *upper;
>> +    struct list_head  *iter;
>> +
>> +    /* We do not support more then 2 levels of VLAN nesting */
>> +    if (idx >= 2)
>> +        return false;
>> +
>> +    netdev_for_each_all_upper_dev_rcu(start, upper, iter) {
>> +        if (is_vlan_dev(upper)) {
>> +            tag[idx].vlan_proto = vlan_dev_vlan_proto(upper);
>> +            tag[idx].vlan_id = vlan_dev_vlan_id(upper);
>> +            idx++;
> 
> Won't idx here skyrocket if we have more than 1 vlan on top of bonding?

Right.  See v2 of the patch. :)

Thanks
-vlad

> 
> Otherwise I like this approach, from the first view.
> 
>> +        }
>> +        if (upper == end)
>> +            return true;
>> +
>> +        /* Look at upper devices list only if current device
>> +         * is a vlan
>> +         */
>> +        if (is_vlan_dev(upper) && bond_collect_vlans(upper, end, tag,
>> idx))
>> +            return true;
>> +    }
>> +    return false;
>> +}
>> +
>>
>> static void bond_arp_send_all(struct bonding *bond, struct slave *slave)
>> {
>> -    struct net_device *upper, *vlan_upper;
>> -    struct list_head *iter, *vlan_iter;
>>     struct rtable *rt;
>> -    struct bond_vlan_tag inner, outer;
>> +    struct bond_vlan_tag tags[2];
>>     __be32 *targets = bond->params.arp_targets, addr;
>>     int i;
>> +    bool ret;
>>
>>     for (i = 0; i < BOND_MAX_ARP_TARGETS && targets[i]; i++) {
>>         pr_debug("basa: target %pI4\n", &targets[i]);
>> -        inner.vlan_proto = 0;
>> -        inner.vlan_id = 0;
>> -        outer.vlan_proto = 0;
>> -        outer.vlan_id = 0;
>> +        memset(tags, 0, sizeof(tags));
>>
>>         /* Find out through which dev should the packet go */
>>         rt = ip_route_output(dev_net(bond->dev), targets[i], 0,
>> @@ -2192,7 +2216,7 @@ static void bond_arp_send_all(struct bonding
>> *bond, struct slave *slave)
>>                 net_warn_ratelimited("%s: no route to arp_ip_target
>> %pI4 and arp_validate is set\n",
>>                              bond->dev->name,
>>                              &targets[i]);
>> -            bond_arp_send(slave->dev, ARPOP_REQUEST, targets[i], 0,
>> &inner, &outer);
>> +            bond_arp_send(slave->dev, ARPOP_REQUEST, targets[i], 0,
>> &tags[1], &tags[0]);
>>             continue;
>>         }
>>
>> @@ -2201,52 +2225,12 @@ static void bond_arp_send_all(struct bonding
>> *bond, struct slave *slave)
>>             goto found;
>>
>>         rcu_read_lock();
>> -        /* first we search only for vlan devices. for every vlan
>> -         * found we verify its upper dev list, searching for the
>> -         * rt->dst.dev. If found we save the tag of the vlan and
>> -         * proceed to send the packet.
>> -         */
>> -        netdev_for_each_all_upper_dev_rcu(bond->dev, vlan_upper,
>> -                          vlan_iter) {
>> -            if (!is_vlan_dev(vlan_upper))
>> -                continue;
>> -
>> -            if (vlan_upper == rt->dst.dev) {
>> -                outer.vlan_proto = vlan_dev_vlan_proto(vlan_upper);
>> -                outer.vlan_id = vlan_dev_vlan_id(vlan_upper);
>> -                rcu_read_unlock();
>> -                goto found;
>> -            }
>> -            netdev_for_each_all_upper_dev_rcu(vlan_upper, upper,
>> -                              iter) {
>> -                if (upper == rt->dst.dev) {
>> -                    /* If the upper dev is a vlan dev too,
>> -                     *  set the vlan tag to inner tag.
>> -                     */
>> -                    if (is_vlan_dev(upper)) {
>> -                        inner.vlan_proto = vlan_dev_vlan_proto(upper);
>> -                        inner.vlan_id = vlan_dev_vlan_id(upper);
>> -                    }
>> -                    outer.vlan_proto = vlan_dev_vlan_proto(vlan_upper);
>> -                    outer.vlan_id = vlan_dev_vlan_id(vlan_upper);
>> -                    rcu_read_unlock();
>> -                    goto found;
>> -                }
>> -            }
>> -        }
>> -
>> -        /* if the device we're looking for is not on top of any of
>> -         * our upper vlans, then just search for any dev that
>> -         * matches, and in case it's a vlan - save the id
>> -         */
>> -        netdev_for_each_all_upper_dev_rcu(bond->dev, upper, iter) {
>> -            if (upper == rt->dst.dev) {
>> -                rcu_read_unlock();
>> -                goto found;
>> -            }
>> -        }
>> +        ret = bond_collect_vlans(bond->dev, rt->dst.dev, tags, 0);
>>         rcu_read_unlock();
>>
>> +        if (ret)
>> +            goto found;
>> +
>>         /* Not our device - skip */
>>         pr_debug("%s: no path to arp_ip_target %pI4 via rt.dev %s\n",
>>              bond->dev->name, &targets[i],
>> @@ -2259,7 +2243,7 @@ found:
>>         addr = bond_confirm_addr(rt->dst.dev, targets[i], 0);
>>         ip_rt_put(rt);
>>         bond_arp_send(slave->dev, ARPOP_REQUEST, targets[i],
>> -                  addr, &inner, &outer);
>> +                  addr, &tags[1], &tags[0]);
>>     }
>> }
>>
>> -- 
>> 1.9.0
>>