[PATCH net-next,v4] bonding: support balance-alb with openvswitch

[PATCH net-next,v4] bonding: support balance-alb with openvswitch

[Mat Kowalski]

Commit d5410ac7b0ba ("net:bonding:support balance-alb interface with
vlan to bridge") introduced a support for balance-alb mode for
interfaces connected to the linux bridge by fixing missing matching of
MAC entry in FDB. In our testing we discovered that it still does not
work when the bond is connected to the OVS bridge as show in diagram
below:

eth1(mac:eth1_mac)--bond0(balance-alb,mac:eth0_mac)--eth0(mac:eth0_mac)
                         |
                       bond0.150(mac:eth0_mac)
                         |
                       ovs_bridge(ip:bridge_ip,mac:eth0_mac)

This patch fixes it by checking not only if the device is a bridge but
also if it is an openvswitch.

Signed-off-by: Mateusz Kowalski <mko@redhat.com>
---
 drivers/net/bonding/bond_alb.c | 2 +-
 include/linux/netdevice.h      | 5 +++++
 2 files changed, 6 insertions(+), 1 deletion(-)

diff --git a/drivers/net/bonding/bond_alb.c b/drivers/net/bonding/bond_alb.c
index b9dbad3a8af8..cc5049eb25f8 100644
--- a/drivers/net/bonding/bond_alb.c
+++ b/drivers/net/bonding/bond_alb.c
@@ -668,7 +668,7 @@ static struct slave *rlb_arp_xmit(struct sk_buff *skb, struct bonding *bond)
 
 	dev = ip_dev_find(dev_net(bond->dev), arp->ip_src);
 	if (dev) {
-		if (netif_is_bridge_master(dev)) {
+		if (netif_is_any_bridge_master(dev)) {
 			dev_put(dev);
 			return NULL;
 		}
diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index 84c36a7f873f..27593c0d3c15 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -5103,6 +5103,11 @@ static inline bool netif_is_ovs_port(const struct net_device *dev)
 	return dev->priv_flags & IFF_OVS_DATAPATH;
 }
 
+static inline bool netif_is_any_bridge_master(const struct net_device *dev)
+{
+	return netif_is_bridge_master(dev) || netif_is_ovs_master(dev);
+}
+
 static inline bool netif_is_any_bridge_port(const struct net_device *dev)
 {
 	return netif_is_bridge_port(dev) || netif_is_ovs_port(dev);
-- 
2.41.0

Re: [PATCH net-next,v4] bonding: support balance-alb with openvswitch

[Jay Vosburgh]

Mat Kowalski <mko@redhat.com> wrote:

>Commit d5410ac7b0ba ("net:bonding:support balance-alb interface with
>vlan to bridge") introduced a support for balance-alb mode for
>interfaces connected to the linux bridge by fixing missing matching of
>MAC entry in FDB. In our testing we discovered that it still does not
>work when the bond is connected to the OVS bridge as show in diagram
>below:
>
>eth1(mac:eth1_mac)--bond0(balance-alb,mac:eth0_mac)--eth0(mac:eth0_mac)
>                         |
>                       bond0.150(mac:eth0_mac)
>                         |
>                       ovs_bridge(ip:bridge_ip,mac:eth0_mac)
>
>This patch fixes it by checking not only if the device is a bridge but
>also if it is an openvswitch.

	What changed between v3 and v4?

	-J

>Signed-off-by: Mateusz Kowalski <mko@redhat.com>
>---
> drivers/net/bonding/bond_alb.c | 2 +-
> include/linux/netdevice.h      | 5 +++++
> 2 files changed, 6 insertions(+), 1 deletion(-)
>
>diff --git a/drivers/net/bonding/bond_alb.c b/drivers/net/bonding/bond_alb.c
>index b9dbad3a8af8..cc5049eb25f8 100644
>--- a/drivers/net/bonding/bond_alb.c
>+++ b/drivers/net/bonding/bond_alb.c
>@@ -668,7 +668,7 @@ static struct slave *rlb_arp_xmit(struct sk_buff *skb, struct bonding *bond)
> 
> 	dev = ip_dev_find(dev_net(bond->dev), arp->ip_src);
> 	if (dev) {
>-		if (netif_is_bridge_master(dev)) {
>+		if (netif_is_any_bridge_master(dev)) {
> 			dev_put(dev);
> 			return NULL;
> 		}
>diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
>index 84c36a7f873f..27593c0d3c15 100644
>--- a/include/linux/netdevice.h
>+++ b/include/linux/netdevice.h
>@@ -5103,6 +5103,11 @@ static inline bool netif_is_ovs_port(const struct net_device *dev)
> 	return dev->priv_flags & IFF_OVS_DATAPATH;
> }
> 
>+static inline bool netif_is_any_bridge_master(const struct net_device *dev)
>+{
>+	return netif_is_bridge_master(dev) || netif_is_ovs_master(dev);
>+}
>+
> static inline bool netif_is_any_bridge_port(const struct net_device *dev)
> {
> 	return netif_is_bridge_port(dev) || netif_is_ovs_port(dev);
>-- 
>2.41.0

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

Re: [PATCH net-next,v4] bonding: support balance-alb with openvswitch

[Mat Kowalski]

On 31/07/2023 00:51, Jay Vosburgh wrote:
> Mat Kowalski <mko@redhat.com> wrote:
> 
>> Commit d5410ac7b0ba ("net:bonding:support balance-alb interface with
>> vlan to bridge") introduced a support for balance-alb mode for
>> interfaces connected to the linux bridge by fixing missing matching of
>> MAC entry in FDB. In our testing we discovered that it still does not
>> work when the bond is connected to the OVS bridge as show in diagram
>> below:
>>
>> eth1(mac:eth1_mac)--bond0(balance-alb,mac:eth0_mac)--eth0(mac:eth0_mac)
>>                         |
>>                       bond0.150(mac:eth0_mac)
>>                         |
>>                       ovs_bridge(ip:bridge_ip,mac:eth0_mac)
>>
>> This patch fixes it by checking not only if the device is a bridge but
>> also if it is an openvswitch.
> 
> 	What changed between v3 and v4?
> 
> 	-J

All the changes were only about whitespaces and formatting of the diff itself. The code itself is the same in v1-v4

> 
>> Signed-off-by: Mateusz Kowalski <mko@redhat.com>
>> ---
>> drivers/net/bonding/bond_alb.c | 2 +-
>> include/linux/netdevice.h      | 5 +++++
>> 2 files changed, 6 insertions(+), 1 deletion(-)
>>
>> diff --git a/drivers/net/bonding/bond_alb.c b/drivers/net/bonding/bond_alb.c
>> index b9dbad3a8af8..cc5049eb25f8 100644
>> --- a/drivers/net/bonding/bond_alb.c
>> +++ b/drivers/net/bonding/bond_alb.c
>> @@ -668,7 +668,7 @@ static struct slave *rlb_arp_xmit(struct sk_buff *skb, struct bonding *bond)
>>
>> 	dev = ip_dev_find(dev_net(bond->dev), arp->ip_src);
>> 	if (dev) {
>> -		if (netif_is_bridge_master(dev)) {
>> +		if (netif_is_any_bridge_master(dev)) {
>> 			dev_put(dev);
>> 			return NULL;
>> 		}
>> diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
>> index 84c36a7f873f..27593c0d3c15 100644
>> --- a/include/linux/netdevice.h
>> +++ b/include/linux/netdevice.h
>> @@ -5103,6 +5103,11 @@ static inline bool netif_is_ovs_port(const struct net_device *dev)
>> 	return dev->priv_flags & IFF_OVS_DATAPATH;
>> }
>>
>> +static inline bool netif_is_any_bridge_master(const struct net_device *dev)
>> +{
>> +	return netif_is_bridge_master(dev) || netif_is_ovs_master(dev);
>> +}
>> +
>> static inline bool netif_is_any_bridge_port(const struct net_device *dev)
>> {
>> 	return netif_is_bridge_port(dev) || netif_is_ovs_port(dev);
>> -- 
>> 2.41.0
> 
> ---
> 	-Jay Vosburgh, jay.vosburgh@canonical.com
> 

Re: [PATCH net-next,v4] bonding: support balance-alb with openvswitch

[Mat Kowalski]

On 31/07/2023 00:51, Jay Vosburgh wrote:
> Mat Kowalski <mko@redhat.com> wrote:
> 
>> Commit d5410ac7b0ba ("net:bonding:support balance-alb interface with
>> vlan to bridge") introduced a support for balance-alb mode for
>> interfaces connected to the linux bridge by fixing missing matching of
>> MAC entry in FDB. In our testing we discovered that it still does not
>> work when the bond is connected to the OVS bridge as show in diagram
>> below:
>>
>> eth1(mac:eth1_mac)--bond0(balance-alb,mac:eth0_mac)--eth0(mac:eth0_mac)
>>                         |
>>                       bond0.150(mac:eth0_mac)
>>                         |
>>                       ovs_bridge(ip:bridge_ip,mac:eth0_mac)
>>
>> This patch fixes it by checking not only if the device is a bridge but
>> also if it is an openvswitch.
> 
> 	What changed between v3 and v4?
> 
> 	-J

v4 changes:
- Fix additional space at the beginning of the line

v3 changes:
- Fix tab chars converted to spaces

v2 changes:
- Fix line wrapping

> 
>> Signed-off-by: Mateusz Kowalski <mko@redhat.com>
>> ---
>> drivers/net/bonding/bond_alb.c | 2 +-
>> include/linux/netdevice.h      | 5 +++++
>> 2 files changed, 6 insertions(+), 1 deletion(-)
>>
>> diff --git a/drivers/net/bonding/bond_alb.c b/drivers/net/bonding/bond_alb.c
>> index b9dbad3a8af8..cc5049eb25f8 100644
>> --- a/drivers/net/bonding/bond_alb.c
>> +++ b/drivers/net/bonding/bond_alb.c
>> @@ -668,7 +668,7 @@ static struct slave *rlb_arp_xmit(struct sk_buff *skb, struct bonding *bond)
>>
>> 	dev = ip_dev_find(dev_net(bond->dev), arp->ip_src);
>> 	if (dev) {
>> -		if (netif_is_bridge_master(dev)) {
>> +		if (netif_is_any_bridge_master(dev)) {
>> 			dev_put(dev);
>> 			return NULL;
>> 		}
>> diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
>> index 84c36a7f873f..27593c0d3c15 100644
>> --- a/include/linux/netdevice.h
>> +++ b/include/linux/netdevice.h
>> @@ -5103,6 +5103,11 @@ static inline bool netif_is_ovs_port(const struct net_device *dev)
>> 	return dev->priv_flags & IFF_OVS_DATAPATH;
>> }
>>
>> +static inline bool netif_is_any_bridge_master(const struct net_device *dev)
>> +{
>> +	return netif_is_bridge_master(dev) || netif_is_ovs_master(dev);
>> +}
>> +
>> static inline bool netif_is_any_bridge_port(const struct net_device *dev)
>> {
>> 	return netif_is_bridge_port(dev) || netif_is_ovs_port(dev);
>> -- 
>> 2.41.0
> 
> ---
> 	-Jay Vosburgh, jay.vosburgh@canonical.com
> 

Re: [PATCH net-next,v4] bonding: support balance-alb with openvswitch

[Paolo Abeni]

On Mon, 2023-07-31 at 22:21 +0200, Mat Kowalski wrote:
> 
> On 31/07/2023 00:51, Jay Vosburgh wrote:
> > Mat Kowalski <mko@redhat.com> wrote:
> > 
> > > Commit d5410ac7b0ba ("net:bonding:support balance-alb interface with
> > > vlan to bridge") introduced a support for balance-alb mode for
> > > interfaces connected to the linux bridge by fixing missing matching of
> > > MAC entry in FDB. In our testing we discovered that it still does not
> > > work when the bond is connected to the OVS bridge as show in diagram
> > > below:
> > > 
> > > eth1(mac:eth1_mac)--bond0(balance-alb,mac:eth0_mac)--eth0(mac:eth0_mac)
> > >                         |
> > >                       bond0.150(mac:eth0_mac)
> > >                         |
> > >                       ovs_bridge(ip:bridge_ip,mac:eth0_mac)
> > > 
> > > This patch fixes it by checking not only if the device is a bridge but
> > > also if it is an openvswitch.
> > 
> > 	What changed between v3 and v4?
> > 
> > 	-J
> 
> v4 changes:
> - Fix additional space at the beginning of the line
> 
> v3 changes:
> - Fix tab chars converted to spaces
> 
> v2 changes:
> - Fix line wrapping

Note that you should wait at least 24h before submitting a new version:

https://elixir.bootlin.com/linux/v6.5-rc4/source/Documentation/process/maintainer-netdev.rst#L350

3 new revisions in a raw is really not a good thing.

The used email address and the SoB mismatch. You should fix that either
changing the SoB or adding a suitable From: tag.

Please include the changelog in the next revision, after the SoB tag
and a '---' separator.

Cheers,

Paolo