[patch net-next v2 0/5] mlxsw: Offload bridge device mrouter

[patch net-next v2 0/5] mlxsw: Offload bridge device mrouter

[Jiri Pirko]

From: Jiri Pirko <jiri@mellanox.com>

Yotam says:

Similarly to a bridged port, the bridge device itself can be configured by
the user to be an mrouter port. In this case, all multicast traffic should
be forwarded to it. Make the mlxsw Spectrum driver offload these directives
to the Spectrum hardware.

Patches 1 and 2 add a new switchdev notification for bridge device mrouter
port status and make the bridge module notify about it.

Patches 3-5 change the mlxsw Spectrum driver to handle these notifications
by adding the Spectrum router port to the bridge MDB entries.

---
v1->v2:
 - patch1:
   - Don't add the MDB_RTR_TYPE_TEMP state and use the timer_pending to
     distinguish between learning-on and learning-off states

Yotam Gigi (5):
  net: bridge: Notify on bridge device mrouter state changes
  net: bridge: Export bridge multicast router state
  mlxsw: spectrum: router: Export the mlxsw_sp_router_port function
  mlxsw: spectrum_switchdev: Add support for router port in SMID entries
  mlxsw: spectrum_switchdev: Support bridge mrouter notifications

 .../net/ethernet/mellanox/mlxsw/spectrum_router.c  |  2 +-
 .../net/ethernet/mellanox/mlxsw/spectrum_router.h  |  1 +
 .../ethernet/mellanox/mlxsw/spectrum_switchdev.c   | 76 +++++++++++++++++++++-
 include/linux/if_bridge.h                          |  5 ++
 include/net/switchdev.h                            |  1 +
 net/bridge/br_multicast.c                          | 50 +++++++++++++-
 6 files changed, 128 insertions(+), 7 deletions(-)

-- 
2.9.5

[patch net-next v2 1/5] net: bridge: Notify on bridge device mrouter state changes

[Jiri Pirko]

From: Yotam Gigi <yotamg@mellanox.com>

Add the SWITCHDEV_ATTR_ID_BRIDGE_MROUTER switchdev notification type, used
to indicate whether the bridge is or isn't mrouter. Notify when the bridge
changes its state, similarly to the already existing bridged port mrouter
notifications.

The notification uses the switchdev_attr.u.mrouter boolean flag to indicate
the current bridge mrouter status. Thus, it only indicates whether the
bridge is currently used as an mrouter or not, and does not indicate the
exact mrouter state of the bridge (learning, permanent, etc.).

Signed-off-by: Yotam Gigi <yotamg@mellanox.com>
Signed-off-by: Jiri Pirko <jiri@mellanox.com>
---
v1->v2:
 - use the timer_pending to distinguish between learning-on and
   learning-off states
---
 include/net/switchdev.h   |  1 +
 net/bridge/br_multicast.c | 38 +++++++++++++++++++++++++++++++++++---
 2 files changed, 36 insertions(+), 3 deletions(-)

diff --git a/include/net/switchdev.h b/include/net/switchdev.h
index d767b79..d756fbe 100644
--- a/include/net/switchdev.h
+++ b/include/net/switchdev.h
@@ -51,6 +51,7 @@ enum switchdev_attr_id {
 	SWITCHDEV_ATTR_ID_BRIDGE_AGEING_TIME,
 	SWITCHDEV_ATTR_ID_BRIDGE_VLAN_FILTERING,
 	SWITCHDEV_ATTR_ID_BRIDGE_MC_DISABLED,
+	SWITCHDEV_ATTR_ID_BRIDGE_MROUTER,
 };
 
 struct switchdev_attr {
diff --git a/net/bridge/br_multicast.c b/net/bridge/br_multicast.c
index 8dc5c8d..bd50550 100644
--- a/net/bridge/br_multicast.c
+++ b/net/bridge/br_multicast.c
@@ -859,8 +859,32 @@ static void br_multicast_router_expired(unsigned long data)
 	spin_unlock(&br->multicast_lock);
 }
 
+static void br_mc_router_state_change(struct net_bridge *p,
+				      bool is_mc_router)
+{
+	struct switchdev_attr attr = {
+		.orig_dev = p->dev,
+		.id = SWITCHDEV_ATTR_ID_BRIDGE_MROUTER,
+		.flags = SWITCHDEV_F_DEFER,
+		.u.mrouter = is_mc_router,
+	};
+
+	switchdev_port_attr_set(p->dev, &attr);
+}
+
 static void br_multicast_local_router_expired(unsigned long data)
 {
+	struct net_bridge *br = (struct net_bridge *)data;
+
+	spin_lock(&br->multicast_lock);
+	if (br->multicast_router == MDB_RTR_TYPE_DISABLED ||
+	    br->multicast_router == MDB_RTR_TYPE_PERM ||
+	    timer_pending(&br->multicast_router_timer))
+		goto out;
+
+	br_mc_router_state_change(br, false);
+out:
+	spin_unlock(&br->multicast_lock);
 }
 
 static void br_multicast_querier_expired(struct net_bridge *br,
@@ -1364,9 +1388,12 @@ static void br_multicast_mark_router(struct net_bridge *br,
 	unsigned long now = jiffies;
 
 	if (!port) {
-		if (br->multicast_router == MDB_RTR_TYPE_TEMP_QUERY)
+		if (br->multicast_router == MDB_RTR_TYPE_TEMP_QUERY) {
+			if (!timer_pending(&br->multicast_router_timer))
+				br_mc_router_state_change(br, true);
 			mod_timer(&br->multicast_router_timer,
 				  now + br->multicast_querier_interval);
+		}
 		return;
 	}
 
@@ -1952,7 +1979,7 @@ void br_multicast_init(struct net_bridge *br)
 
 	spin_lock_init(&br->multicast_lock);
 	setup_timer(&br->multicast_router_timer,
-		    br_multicast_local_router_expired, 0);
+		    br_multicast_local_router_expired, (unsigned long)br);
 	setup_timer(&br->ip4_other_query.timer,
 		    br_ip4_multicast_querier_expired, (unsigned long)br);
 	setup_timer(&br->ip4_own_query.timer, br_ip4_multicast_query_expired,
@@ -2042,9 +2069,14 @@ int br_multicast_set_router(struct net_bridge *br, unsigned long val)
 	switch (val) {
 	case MDB_RTR_TYPE_DISABLED:
 	case MDB_RTR_TYPE_PERM:
+		br_mc_router_state_change(br, val == MDB_RTR_TYPE_PERM);
 		del_timer(&br->multicast_router_timer);
-		/* fall through */
+		br->multicast_router = val;
+		err = 0;
+		break;
 	case MDB_RTR_TYPE_TEMP_QUERY:
+		if (br->multicast_router != MDB_RTR_TYPE_TEMP_QUERY)
+			br_mc_router_state_change(br, false);
 		br->multicast_router = val;
 		err = 0;
 		break;
-- 
2.9.5

[patch net-next v2 2/5] net: bridge: Export bridge multicast router state

[Jiri Pirko]

From: Yotam Gigi <yotamg@mellanox.com>

Add an access function that, given a bridge netdevice, returns whether the
bridge device is currently an mrouter or not. The function uses the already
existing br_multicast_is_router function to check that.

This function is needed in order to allow ports that join an already
existing bridge to know the current mrouter state of the bridge device.
Together with the bridge device mrouter ports switchdev notifications, it
is possible to have full offloading of the semantics of the bridge device
mcast router state.

Due to the fact that the bridge multicast router status can change in
packet RX path, take the multicast_router bridge spinlock to protect the
read.

Signed-off-by: Yotam Gigi <yotamg@mellanox.com>
Reviewed-by: Nogah Frankel <nogahf@mellanox.com>
Reviewed-by: Nikolay Aleksandrov <nikolay@cumulusnetworks.com>
Signed-off-by: Jiri Pirko <jiri@mellanox.com>
---
 include/linux/if_bridge.h |  5 +++++
 net/bridge/br_multicast.c | 12 ++++++++++++
 2 files changed, 17 insertions(+)

diff --git a/include/linux/if_bridge.h b/include/linux/if_bridge.h
index 316ee11..02639eb 100644
--- a/include/linux/if_bridge.h
+++ b/include/linux/if_bridge.h
@@ -64,6 +64,7 @@ int br_multicast_list_adjacent(struct net_device *dev,
 bool br_multicast_has_querier_anywhere(struct net_device *dev, int proto);
 bool br_multicast_has_querier_adjacent(struct net_device *dev, int proto);
 bool br_multicast_enabled(const struct net_device *dev);
+bool br_multicast_router(const struct net_device *dev);
 #else
 static inline int br_multicast_list_adjacent(struct net_device *dev,
 					     struct list_head *br_ip_list)
@@ -84,6 +85,10 @@ static inline bool br_multicast_enabled(const struct net_device *dev)
 {
 	return false;
 }
+static inline bool br_multicast_router(const struct net_device *dev)
+{
+	return false;
+}
 #endif
 
 #if IS_ENABLED(CONFIG_BRIDGE) && IS_ENABLED(CONFIG_BRIDGE_VLAN_FILTERING)
diff --git a/net/bridge/br_multicast.c b/net/bridge/br_multicast.c
index bd50550..7947e04 100644
--- a/net/bridge/br_multicast.c
+++ b/net/bridge/br_multicast.c
@@ -2216,6 +2216,18 @@ bool br_multicast_enabled(const struct net_device *dev)
 }
 EXPORT_SYMBOL_GPL(br_multicast_enabled);
 
+bool br_multicast_router(const struct net_device *dev)
+{
+	struct net_bridge *br = netdev_priv(dev);
+	bool is_router;
+
+	spin_lock_bh(&br->multicast_lock);
+	is_router = br_multicast_is_router(br);
+	spin_unlock_bh(&br->multicast_lock);
+	return is_router;
+}
+EXPORT_SYMBOL_GPL(br_multicast_router);
+
 int br_multicast_set_querier(struct net_bridge *br, unsigned long val)
 {
 	unsigned long max_delay;
-- 
2.9.5

[patch net-next v2 3/5] mlxsw: spectrum: router: Export the mlxsw_sp_router_port function

[Jiri Pirko]

From: Yotam Gigi <yotamg@mellanox.com>

In Spectrum hardware, the router port is a virtual port that is the gateway
to the routing mechanism. Hence, in order for a packet to be L3 forwarded,
it must first be L2 forwarded to the router port inside the hardware.

Further patches in this patchset are going to introduce support in bridge
device used as an mrouter port. In this case, the router port index will be
needed in order to update the MDB entries to include the router port. Thus,
export the mlxsw_sp_router_port function, which returns the index of the
Spectrum router port.

Signed-off-by: Yotam Gigi <yotamg@mellanox.com>
Reviewed-by: Nogah Frankel <nogahf@mellanox.com>
Signed-off-by: Jiri Pirko <jiri@mellanox.com>
---
 drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c | 2 +-
 drivers/net/ethernet/mellanox/mlxsw/spectrum_router.h | 1 +
 2 files changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c
index e0f8ea4..a072903 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c
@@ -5957,7 +5957,7 @@ static int mlxsw_sp_rif_vlan_fid_op(struct mlxsw_sp_rif *rif,
 	return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(ritr), ritr_pl);
 }
 
-static u8 mlxsw_sp_router_port(const struct mlxsw_sp *mlxsw_sp)
+u8 mlxsw_sp_router_port(const struct mlxsw_sp *mlxsw_sp)
 {
 	return mlxsw_core_max_ports(mlxsw_sp->core) + 1;
 }
diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.h b/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.h
index 3d44918..3f2d840 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.h
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.h
@@ -70,6 +70,7 @@ u16 mlxsw_sp_rif_index(const struct mlxsw_sp_rif *rif);
 u16 mlxsw_sp_ipip_lb_rif_index(const struct mlxsw_sp_rif_ipip_lb *rif);
 u16 mlxsw_sp_ipip_lb_ul_vr_id(const struct mlxsw_sp_rif_ipip_lb *rif);
 int mlxsw_sp_rif_dev_ifindex(const struct mlxsw_sp_rif *rif);
+u8 mlxsw_sp_router_port(const struct mlxsw_sp *mlxsw_sp);
 const struct net_device *mlxsw_sp_rif_dev(const struct mlxsw_sp_rif *rif);
 int mlxsw_sp_rif_counter_value_get(struct mlxsw_sp *mlxsw_sp,
 				   struct mlxsw_sp_rif *rif,
-- 
2.9.5

[patch net-next v2 4/5] mlxsw: spectrum_switchdev: Add support for router port in SMID entries

[Jiri Pirko]

From: Yotam Gigi <yotamg@mellanox.com>

In Spectrum, MDB entries point to MID entries, that indicate which ports a
packet should be forwarded to. Add the support in creating MID entries that
forward the packet to the Spectrum router port.

This will be later used to handle the bridge mrouter port switchdev
notifications.

Signed-off-by: Yotam Gigi <yotamg@mellanox.com>
Reviewed-by: Nogah Frankel <nogahf@mellanox.com>
Signed-off-by: Jiri Pirko <jiri@mellanox.com>
---
 drivers/net/ethernet/mellanox/mlxsw/spectrum_switchdev.c | 13 +++++++++++--
 1 file changed, 11 insertions(+), 2 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_switchdev.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_switchdev.c
index 2cfdf22..4b4584f 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_switchdev.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_switchdev.c
@@ -49,6 +49,7 @@
 #include <linux/netlink.h>
 #include <net/switchdev.h>
 
+#include "spectrum_router.h"
 #include "spectrum.h"
 #include "core.h"
 #include "reg.h"
@@ -1243,7 +1244,8 @@ static int mlxsw_sp_port_mdb_op(struct mlxsw_sp *mlxsw_sp, const char *addr,
 }
 
 static int mlxsw_sp_port_smid_full_entry(struct mlxsw_sp *mlxsw_sp, u16 mid_idx,
-					 long *ports_bitmap)
+					 long *ports_bitmap,
+					 bool set_router_port)
 {
 	char *smid_pl;
 	int err, i;
@@ -1258,9 +1260,15 @@ static int mlxsw_sp_port_smid_full_entry(struct mlxsw_sp *mlxsw_sp, u16 mid_idx,
 			mlxsw_reg_smid_port_mask_set(smid_pl, i, 1);
 	}
 
+	mlxsw_reg_smid_port_mask_set(smid_pl,
+				     mlxsw_sp_router_port(mlxsw_sp), 1);
+
 	for_each_set_bit(i, ports_bitmap, mlxsw_core_max_ports(mlxsw_sp->core))
 		mlxsw_reg_smid_port_set(smid_pl, i, 1);
 
+	mlxsw_reg_smid_port_set(smid_pl, mlxsw_sp_router_port(mlxsw_sp),
+				set_router_port);
+
 	err = mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(smid), smid_pl);
 	kfree(smid_pl);
 	return err;
@@ -1364,7 +1372,8 @@ mlxsw_sp_mc_write_mdb_entry(struct mlxsw_sp *mlxsw_sp,
 	mlxsw_sp_mc_get_mrouters_bitmap(flood_bitmap, bridge_device, mlxsw_sp);
 
 	mid->mid = mid_idx;
-	err = mlxsw_sp_port_smid_full_entry(mlxsw_sp, mid_idx, flood_bitmap);
+	err = mlxsw_sp_port_smid_full_entry(mlxsw_sp, mid_idx, flood_bitmap,
+					    false);
 	kfree(flood_bitmap);
 	if (err)
 		return false;
-- 
2.9.5

[patch net-next v2 5/5] mlxsw: spectrum_switchdev: Support bridge mrouter notifications

[Jiri Pirko]

From: Yotam Gigi <yotamg@mellanox.com>

Support the SWITCHDEV_ATTR_ID_BRIDGE_MROUTER port attribute switchdev
notification.

To do that, add the mrouter flag to struct mlxsw_sp_bridge_device, which
indicates whether the bridge device was set to be mrouter port. This field
is set when:
 - A new bridge is created, where the value is taken from the kernel
   bridge value.
 - A switchdev SWITCHDEV_ATTR_ID_BRIDGE_MROUTER notification is sent.

In addition, change the bridge MID entries to include the router port when
the bridge device is configured to be mrouter port. The MID entries are
updated in the following cases:
 - When a new MID entry is created, update the router port according to the
   bridge mrouter state.
 - When a SWITCHDEV_ATTR_ID_BRIDGE_MROUTER notification is sent, update all
   the bridge's MID entries.

This is aligned with the case where a bridge slave is configured to be
mrouter port.

Signed-off-by: Yotam Gigi <yotamg@mellanox.com>
Reviewed-by: Nogah Frankel <nogahf@mellanox.com>
Signed-off-by: Jiri Pirko <jiri@mellanox.com>
---
 .../ethernet/mellanox/mlxsw/spectrum_switchdev.c   | 65 +++++++++++++++++++++-
 1 file changed, 63 insertions(+), 2 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_switchdev.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_switchdev.c
index 4b4584f..7b8548e 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_switchdev.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_switchdev.c
@@ -80,7 +80,8 @@ struct mlxsw_sp_bridge_device {
 	struct list_head ports_list;
 	struct list_head mids_list;
 	u8 vlan_enabled:1,
-	   multicast_enabled:1;
+	   multicast_enabled:1,
+	   mrouter:1;
 	const struct mlxsw_sp_bridge_ops *ops;
 };
 
@@ -171,6 +172,7 @@ mlxsw_sp_bridge_device_create(struct mlxsw_sp_bridge *bridge,
 	bridge_device->dev = br_dev;
 	bridge_device->vlan_enabled = vlan_enabled;
 	bridge_device->multicast_enabled = br_multicast_enabled(br_dev);
+	bridge_device->mrouter = br_multicast_router(br_dev);
 	INIT_LIST_HEAD(&bridge_device->ports_list);
 	if (vlan_enabled) {
 		bridge->vlan_enabled_exists = true;
@@ -813,6 +815,60 @@ static int mlxsw_sp_port_mc_disabled_set(struct mlxsw_sp_port *mlxsw_sp_port,
 	return 0;
 }
 
+static int mlxsw_sp_smid_router_port_set(struct mlxsw_sp *mlxsw_sp,
+					 u16 mid_idx, bool add)
+{
+	char *smid_pl;
+	int err;
+
+	smid_pl = kmalloc(MLXSW_REG_SMID_LEN, GFP_KERNEL);
+	if (!smid_pl)
+		return -ENOMEM;
+
+	mlxsw_reg_smid_pack(smid_pl, mid_idx,
+			    mlxsw_sp_router_port(mlxsw_sp), add);
+	err = mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(smid), smid_pl);
+	kfree(smid_pl);
+	return err;
+}
+
+static void
+mlxsw_sp_bridge_mrouter_update_mdb(struct mlxsw_sp *mlxsw_sp,
+				   struct mlxsw_sp_bridge_device *bridge_device,
+				   bool add)
+{
+	struct mlxsw_sp_mid *mid;
+
+	list_for_each_entry(mid, &bridge_device->mids_list, list)
+		mlxsw_sp_smid_router_port_set(mlxsw_sp, mid->mid, add);
+}
+
+static int
+mlxsw_sp_port_attr_br_mrouter_set(struct mlxsw_sp_port *mlxsw_sp_port,
+				  struct switchdev_trans *trans,
+				  struct net_device *orig_dev,
+				  bool is_mrouter)
+{
+	struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp;
+	struct mlxsw_sp_bridge_device *bridge_device;
+
+	if (switchdev_trans_ph_prepare(trans))
+		return 0;
+
+	/* It's possible we failed to enslave the port, yet this
+	 * operation is executed due to it being deferred.
+	 */
+	bridge_device = mlxsw_sp_bridge_device_find(mlxsw_sp->bridge, orig_dev);
+	if (!bridge_device)
+		return 0;
+
+	if (bridge_device->mrouter != is_mrouter)
+		mlxsw_sp_bridge_mrouter_update_mdb(mlxsw_sp, bridge_device,
+						   is_mrouter);
+	bridge_device->mrouter = is_mrouter;
+	return 0;
+}
+
 static int mlxsw_sp_port_attr_set(struct net_device *dev,
 				  const struct switchdev_attr *attr,
 				  struct switchdev_trans *trans)
@@ -850,6 +906,11 @@ static int mlxsw_sp_port_attr_set(struct net_device *dev,
 						    attr->orig_dev,
 						    attr->u.mc_disabled);
 		break;
+	case SWITCHDEV_ATTR_ID_BRIDGE_MROUTER:
+		err = mlxsw_sp_port_attr_br_mrouter_set(mlxsw_sp_port, trans,
+							attr->orig_dev,
+							attr->u.mrouter);
+		break;
 	default:
 		err = -EOPNOTSUPP;
 		break;
@@ -1373,7 +1434,7 @@ mlxsw_sp_mc_write_mdb_entry(struct mlxsw_sp *mlxsw_sp,
 
 	mid->mid = mid_idx;
 	err = mlxsw_sp_port_smid_full_entry(mlxsw_sp, mid_idx, flood_bitmap,
-					    false);
+					    bridge_device->mrouter);
 	kfree(flood_bitmap);
 	if (err)
 		return false;
-- 
2.9.5

Re: [patch net-next v2 1/5] net: bridge: Notify on bridge device mrouter state changes

[Ivan Vecera]

On 9.10.2017 11:15, Jiri Pirko wrote:
> From: Yotam Gigi <yotamg@mellanox.com>
> 
> Add the SWITCHDEV_ATTR_ID_BRIDGE_MROUTER switchdev notification type, used
> to indicate whether the bridge is or isn't mrouter. Notify when the bridge
> changes its state, similarly to the already existing bridged port mrouter
> notifications.
> 
> The notification uses the switchdev_attr.u.mrouter boolean flag to indicate
> the current bridge mrouter status. Thus, it only indicates whether the
> bridge is currently used as an mrouter or not, and does not indicate the
> exact mrouter state of the bridge (learning, permanent, etc.).
> 
> Signed-off-by: Yotam Gigi <yotamg@mellanox.com>
> Signed-off-by: Jiri Pirko <jiri@mellanox.com>
> ---
> v1->v2:
>  - use the timer_pending to distinguish between learning-on and
>    learning-off states
> ---
>  include/net/switchdev.h   |  1 +
>  net/bridge/br_multicast.c | 38 +++++++++++++++++++++++++++++++++++---
>  2 files changed, 36 insertions(+), 3 deletions(-)
> 
> diff --git a/include/net/switchdev.h b/include/net/switchdev.h
> index d767b79..d756fbe 100644
> --- a/include/net/switchdev.h
> +++ b/include/net/switchdev.h
> @@ -51,6 +51,7 @@ enum switchdev_attr_id {
>  	SWITCHDEV_ATTR_ID_BRIDGE_AGEING_TIME,
>  	SWITCHDEV_ATTR_ID_BRIDGE_VLAN_FILTERING,
>  	SWITCHDEV_ATTR_ID_BRIDGE_MC_DISABLED,
> +	SWITCHDEV_ATTR_ID_BRIDGE_MROUTER,
>  };
>  
>  struct switchdev_attr {
> diff --git a/net/bridge/br_multicast.c b/net/bridge/br_multicast.c
> index 8dc5c8d..bd50550 100644
> --- a/net/bridge/br_multicast.c
> +++ b/net/bridge/br_multicast.c
> @@ -859,8 +859,32 @@ static void br_multicast_router_expired(unsigned long data)
>  	spin_unlock(&br->multicast_lock);
>  }
>  
> +static void br_mc_router_state_change(struct net_bridge *p,
> +				      bool is_mc_router)
> +{
> +	struct switchdev_attr attr = {
> +		.orig_dev = p->dev,
> +		.id = SWITCHDEV_ATTR_ID_BRIDGE_MROUTER,
> +		.flags = SWITCHDEV_F_DEFER,
> +		.u.mrouter = is_mc_router,
> +	};
> +
> +	switchdev_port_attr_set(p->dev, &attr);
> +}
> +
>  static void br_multicast_local_router_expired(unsigned long data)
>  {
> +	struct net_bridge *br = (struct net_bridge *)data;
> +
> +	spin_lock(&br->multicast_lock);
> +	if (br->multicast_router == MDB_RTR_TYPE_DISABLED ||
> +	    br->multicast_router == MDB_RTR_TYPE_PERM ||
> +	    timer_pending(&br->multicast_router_timer))
> +		goto out;
> +
> +	br_mc_router_state_change(br, false);
> +out:
> +	spin_unlock(&br->multicast_lock);
>  }
>  
>  static void br_multicast_querier_expired(struct net_bridge *br,
> @@ -1364,9 +1388,12 @@ static void br_multicast_mark_router(struct net_bridge *br,
>  	unsigned long now = jiffies;
>  
>  	if (!port) {
> -		if (br->multicast_router == MDB_RTR_TYPE_TEMP_QUERY)
> +		if (br->multicast_router == MDB_RTR_TYPE_TEMP_QUERY) {
> +			if (!timer_pending(&br->multicast_router_timer))
> +				br_mc_router_state_change(br, true);
>  			mod_timer(&br->multicast_router_timer,
>  				  now + br->multicast_querier_interval);
> +		}
>  		return;
>  	}
>  
> @@ -1952,7 +1979,7 @@ void br_multicast_init(struct net_bridge *br)
>  
>  	spin_lock_init(&br->multicast_lock);
>  	setup_timer(&br->multicast_router_timer,
> -		    br_multicast_local_router_expired, 0);
> +		    br_multicast_local_router_expired, (unsigned long)br);
>  	setup_timer(&br->ip4_other_query.timer,
>  		    br_ip4_multicast_querier_expired, (unsigned long)br);
>  	setup_timer(&br->ip4_own_query.timer, br_ip4_multicast_query_expired,
> @@ -2042,9 +2069,14 @@ int br_multicast_set_router(struct net_bridge *br, unsigned long val)
>  	switch (val) {
>  	case MDB_RTR_TYPE_DISABLED:
>  	case MDB_RTR_TYPE_PERM:
> +		br_mc_router_state_change(br, val == MDB_RTR_TYPE_PERM);
>  		del_timer(&br->multicast_router_timer);
> -		/* fall through */
> +		br->multicast_router = val;
> +		err = 0;
> +		break;
>  	case MDB_RTR_TYPE_TEMP_QUERY:
> +		if (br->multicast_router != MDB_RTR_TYPE_TEMP_QUERY)
> +			br_mc_router_state_change(br, false);
>  		br->multicast_router = val;
>  		err = 0;
>  		break;
> 

Reviewed by: Ivan Vecera <ivecera@redhat.com>

Re: [patch net-next v2 2/5] net: bridge: Export bridge multicast router state

[Ivan Vecera]

On 9.10.2017 11:15, Jiri Pirko wrote:
> From: Yotam Gigi <yotamg@mellanox.com>
> 
> Add an access function that, given a bridge netdevice, returns whether the
> bridge device is currently an mrouter or not. The function uses the already
> existing br_multicast_is_router function to check that.
> 
> This function is needed in order to allow ports that join an already
> existing bridge to know the current mrouter state of the bridge device.
> Together with the bridge device mrouter ports switchdev notifications, it
> is possible to have full offloading of the semantics of the bridge device
> mcast router state.
> 
> Due to the fact that the bridge multicast router status can change in
> packet RX path, take the multicast_router bridge spinlock to protect the
> read.
> 
> Signed-off-by: Yotam Gigi <yotamg@mellanox.com>
> Reviewed-by: Nogah Frankel <nogahf@mellanox.com>
> Reviewed-by: Nikolay Aleksandrov <nikolay@cumulusnetworks.com>
> Signed-off-by: Jiri Pirko <jiri@mellanox.com>
> ---
>  include/linux/if_bridge.h |  5 +++++
>  net/bridge/br_multicast.c | 12 ++++++++++++
>  2 files changed, 17 insertions(+)
> 
> diff --git a/include/linux/if_bridge.h b/include/linux/if_bridge.h
> index 316ee11..02639eb 100644
> --- a/include/linux/if_bridge.h
> +++ b/include/linux/if_bridge.h
> @@ -64,6 +64,7 @@ int br_multicast_list_adjacent(struct net_device *dev,
>  bool br_multicast_has_querier_anywhere(struct net_device *dev, int proto);
>  bool br_multicast_has_querier_adjacent(struct net_device *dev, int proto);
>  bool br_multicast_enabled(const struct net_device *dev);
> +bool br_multicast_router(const struct net_device *dev);
>  #else
>  static inline int br_multicast_list_adjacent(struct net_device *dev,
>  					     struct list_head *br_ip_list)
> @@ -84,6 +85,10 @@ static inline bool br_multicast_enabled(const struct net_device *dev)
>  {
>  	return false;
>  }
> +static inline bool br_multicast_router(const struct net_device *dev)
> +{
> +	return false;
> +}
>  #endif
>  
>  #if IS_ENABLED(CONFIG_BRIDGE) && IS_ENABLED(CONFIG_BRIDGE_VLAN_FILTERING)
> diff --git a/net/bridge/br_multicast.c b/net/bridge/br_multicast.c
> index bd50550..7947e04 100644
> --- a/net/bridge/br_multicast.c
> +++ b/net/bridge/br_multicast.c
> @@ -2216,6 +2216,18 @@ bool br_multicast_enabled(const struct net_device *dev)
>  }
>  EXPORT_SYMBOL_GPL(br_multicast_enabled);
>  
> +bool br_multicast_router(const struct net_device *dev)
> +{
> +	struct net_bridge *br = netdev_priv(dev);
> +	bool is_router;
> +
> +	spin_lock_bh(&br->multicast_lock);
> +	is_router = br_multicast_is_router(br);
> +	spin_unlock_bh(&br->multicast_lock);
> +	return is_router;
> +}
> +EXPORT_SYMBOL_GPL(br_multicast_router);
> +
>  int br_multicast_set_querier(struct net_bridge *br, unsigned long val)
>  {
>  	unsigned long max_delay;
> 

Reviewed by: Ivan Vecera <ivecera@redhat.com>

Re: [patch net-next v2 1/5] net: bridge: Notify on bridge device mrouter state changes

[Nikolay Aleksandrov]

On 09/10/17 12:15, Jiri Pirko wrote:
> From: Yotam Gigi <yotamg@mellanox.com>
> 
> Add the SWITCHDEV_ATTR_ID_BRIDGE_MROUTER switchdev notification type, used
> to indicate whether the bridge is or isn't mrouter. Notify when the bridge
> changes its state, similarly to the already existing bridged port mrouter
> notifications.
> 
> The notification uses the switchdev_attr.u.mrouter boolean flag to indicate
> the current bridge mrouter status. Thus, it only indicates whether the
> bridge is currently used as an mrouter or not, and does not indicate the
> exact mrouter state of the bridge (learning, permanent, etc.).
> 
> Signed-off-by: Yotam Gigi <yotamg@mellanox.com>
> Signed-off-by: Jiri Pirko <jiri@mellanox.com>
> ---
> v1->v2:
>  - use the timer_pending to distinguish between learning-on and
>    learning-off states
> ---
>  include/net/switchdev.h   |  1 +
>  net/bridge/br_multicast.c | 38 +++++++++++++++++++++++++++++++++++---
>  2 files changed, 36 insertions(+), 3 deletions(-)
> 

Much simpler, I like it. Thanks!

Signed-off-by: Nikolay Aleksandrov <nikolay@cumulusnetworks.com>

Re: [patch net-next v2 0/5] mlxsw: Offload bridge device mrouter

[David Miller]

From: Jiri Pirko <jiri@resnulli.us>
Date: Mon,  9 Oct 2017 11:15:30 +0200

> From: Jiri Pirko <jiri@mellanox.com>
> 
> Yotam says:
> 
> Similarly to a bridged port, the bridge device itself can be configured by
> the user to be an mrouter port. In this case, all multicast traffic should
> be forwarded to it. Make the mlxsw Spectrum driver offload these directives
> to the Spectrum hardware.
> 
> Patches 1 and 2 add a new switchdev notification for bridge device mrouter
> port status and make the bridge module notify about it.
> 
> Patches 3-5 change the mlxsw Spectrum driver to handle these notifications
> by adding the Spectrum router port to the bridge MDB entries.
> 
> ---
> v1->v2:
>  - patch1:
>    - Don't add the MDB_RTR_TYPE_TEMP state and use the timer_pending to
>      distinguish between learning-on and learning-off states

Series applied, thank you.