[PATCHv2 net-next 0/2] Add two new flags to bridge.

[PATCHv2 net-next 0/2] Add two new flags to bridge.

[Vlad Yasevich]

The following series adds 2 new flags to bridge.  One flag allows
the user to control whether mac learning is performed on the interface
or not.  By default mac learning is on.
The other flag allows the user to control whether unicast traffic
is flooded (send without an fdb) to a given unicast port.  Default is
on.

Changes since v1:
 - Integrated suggestion from MST to not impact RTM_NEWNEIGH and to
   skip lookups when learning is disabled.

Vlad Yasevich (2):
  bridge: Add flag to control mac learning.
  bridge: Add a flag to control unicast packet flood.

 include/uapi/linux/if_link.h |    2 ++
 net/bridge/br_device.c       |    8 ++++----
 net/bridge/br_fdb.c          |   17 ++++++++++++++---
 net/bridge/br_forward.c      |   14 +++++++++-----
 net/bridge/br_if.c           |    2 +-
 net/bridge/br_input.c        |    9 ++++++---
 net/bridge/br_netlink.c      |   10 +++++++++-
 net/bridge/br_private.h      |    7 +++++--
 net/bridge/br_sysfs_if.c     |    4 ++++
 9 files changed, 54 insertions(+), 19 deletions(-)

-- 
1.7.7.6

[PATCHv2 net-next 1/2] bridge: Add flag to control mac learning.

[Vlad Yasevich]

Allow user to control whether mac learning is enabled on the port.
By default, mac learning is enabled.  Disabling mac learning will
cause new dynamic FDB entries to not be created for a particular port.

Signed-off-by: Vlad Yasevich <vyasevic@redhat.com>
---
 include/uapi/linux/if_link.h |    1 +
 net/bridge/br_fdb.c          |   17 ++++++++++++++---
 net/bridge/br_if.c           |    2 +-
 net/bridge/br_netlink.c      |    6 +++++-
 net/bridge/br_private.h      |    1 +
 net/bridge/br_sysfs_if.c     |    2 ++
 6 files changed, 24 insertions(+), 5 deletions(-)

diff --git a/include/uapi/linux/if_link.h b/include/uapi/linux/if_link.h
index e316354..80fad7f 100644
--- a/include/uapi/linux/if_link.h
+++ b/include/uapi/linux/if_link.h
@@ -221,6 +221,7 @@ enum {
 	IFLA_BRPORT_GUARD,	/* bpdu guard              */
 	IFLA_BRPORT_PROTECT,	/* root port protection    */
 	IFLA_BRPORT_FAST_LEAVE,	/* multicast fast leave    */
+	IFLA_BRPORT_LEARNING,	/* mac learning */
 	__IFLA_BRPORT_MAX
 };
 #define IFLA_BRPORT_MAX (__IFLA_BRPORT_MAX - 1)
diff --git a/net/bridge/br_fdb.c b/net/bridge/br_fdb.c
index c581f12..f29eb0b 100644
--- a/net/bridge/br_fdb.c
+++ b/net/bridge/br_fdb.c
@@ -446,8 +446,9 @@ int br_fdb_insert(struct net_bridge *br, struct net_bridge_port *source,
 	return ret;
 }
 
-void br_fdb_update(struct net_bridge *br, struct net_bridge_port *source,
-		   const unsigned char *addr, u16 vid)
+static void __br_fdb_update(struct net_bridge *br,
+			    struct net_bridge_port *source,
+		     	    const unsigned char *addr, u16 vid)
 {
 	struct hlist_head *head = &br->hash[br_mac_hash(addr, vid)];
 	struct net_bridge_fdb_entry *fdb;
@@ -481,6 +482,7 @@ void br_fdb_update(struct net_bridge *br, struct net_bridge_port *source,
 			if (fdb)
 				fdb_notify(br, fdb, RTM_NEWNEIGH);
 		}
+unlock:
 		/* else  we lose race and someone else inserts
 		 * it first, don't bother updating
 		 */
@@ -488,6 +490,15 @@ void br_fdb_update(struct net_bridge *br, struct net_bridge_port *source,
 	}
 }
 
+void br_fdb_update(struct net_bridge *br, struct net_bridge_port *source,
+		     const unsigned char *addr, u16 vid)
+{
+	if (!(source->flags & BR_LEARNING))
+		return;
+
+	__br_fdb_update(br, source, addr, vid);
+}
+
 static int fdb_to_nud(const struct net_bridge_fdb_entry *fdb)
 {
 	if (fdb->is_local)
@@ -653,7 +664,7 @@ static int __br_fdb_add(struct ndmsg *ndm, struct net_bridge_port *p,
 
 	if (ndm->ndm_flags & NTF_USE) {
 		rcu_read_lock();
-		br_fdb_update(p->br, p, addr, vid);
+		__br_fdb_update(p->br, p, addr, vid);
 		rcu_read_unlock();
 	} else {
 		spin_lock_bh(&p->br->hash_lock);
diff --git a/net/bridge/br_if.c b/net/bridge/br_if.c
index f17fcb3..9751103 100644
--- a/net/bridge/br_if.c
+++ b/net/bridge/br_if.c
@@ -220,7 +220,7 @@ static struct net_bridge_port *new_nbp(struct net_bridge *br,
 	p->path_cost = port_cost(dev);
 	p->priority = 0x8000 >> BR_PORT_BITS;
 	p->port_no = index;
-	p->flags = 0;
+	p->flags = BR_LEARNING;
 	br_init_port(p);
 	p->state = BR_STATE_DISABLED;
 	br_stp_port_timer_init(p);
diff --git a/net/bridge/br_netlink.c b/net/bridge/br_netlink.c
index 8e3abf5..ce902bf 100644
--- a/net/bridge/br_netlink.c
+++ b/net/bridge/br_netlink.c
@@ -30,6 +30,7 @@ static inline size_t br_port_info_size(void)
 		+ nla_total_size(1)	/* IFLA_BRPORT_GUARD */
 		+ nla_total_size(1)	/* IFLA_BRPORT_PROTECT */
 		+ nla_total_size(1)	/* IFLA_BRPORT_FAST_LEAVE */
+		+ nla_total_size(1)	/* IFLA_BRPORT_LEARNING */
 		+ 0;
 }
 
@@ -56,7 +57,8 @@ static int br_port_fill_attrs(struct sk_buff *skb,
 	    nla_put_u8(skb, IFLA_BRPORT_MODE, mode) ||
 	    nla_put_u8(skb, IFLA_BRPORT_GUARD, !!(p->flags & BR_BPDU_GUARD)) ||
 	    nla_put_u8(skb, IFLA_BRPORT_PROTECT, !!(p->flags & BR_ROOT_BLOCK)) ||
-	    nla_put_u8(skb, IFLA_BRPORT_FAST_LEAVE, !!(p->flags & BR_MULTICAST_FAST_LEAVE)))
+	    nla_put_u8(skb, IFLA_BRPORT_FAST_LEAVE, !!(p->flags & BR_MULTICAST_FAST_LEAVE)) ||
+	    nla_put_u8(skb, IFLA_BRPORT_LEARNING, !!(p->flags & BR_LEARNING)))
 		return -EMSGSIZE;
 
 	return 0;
@@ -281,6 +283,7 @@ static const struct nla_policy ifla_brport_policy[IFLA_BRPORT_MAX + 1] = {
 	[IFLA_BRPORT_MODE]	= { .type = NLA_U8 },
 	[IFLA_BRPORT_GUARD]	= { .type = NLA_U8 },
 	[IFLA_BRPORT_PROTECT]	= { .type = NLA_U8 },
+	[IFLA_BRPORT_LEARNING]	= { .type = NLA_U8 },
 };
 
 /* Change the state of the port and notify spanning tree */
@@ -328,6 +331,7 @@ static int br_setport(struct net_bridge_port *p, struct nlattr *tb[])
 	br_set_port_flag(p, tb, IFLA_BRPORT_GUARD, BR_BPDU_GUARD);
 	br_set_port_flag(p, tb, IFLA_BRPORT_FAST_LEAVE, BR_MULTICAST_FAST_LEAVE);
 	br_set_port_flag(p, tb, IFLA_BRPORT_PROTECT, BR_ROOT_BLOCK);
+	br_set_port_flag(p, tb, IFLA_BRPORT_LEARNING, BR_LEARNING);
 
 	if (tb[IFLA_BRPORT_COST]) {
 		err = br_stp_set_path_cost(p, nla_get_u32(tb[IFLA_BRPORT_COST]));
diff --git a/net/bridge/br_private.h b/net/bridge/br_private.h
index 3cbf5be..67842b9 100644
--- a/net/bridge/br_private.h
+++ b/net/bridge/br_private.h
@@ -156,6 +156,7 @@ struct net_bridge_port
 #define BR_BPDU_GUARD           0x00000002
 #define BR_ROOT_BLOCK		0x00000004
 #define BR_MULTICAST_FAST_LEAVE	0x00000008
+#define BR_LEARNING		0x00000010
 
 #ifdef CONFIG_BRIDGE_IGMP_SNOOPING
 	u32				multicast_startup_queries_sent;
diff --git a/net/bridge/br_sysfs_if.c b/net/bridge/br_sysfs_if.c
index a1ef1b6..707f362 100644
--- a/net/bridge/br_sysfs_if.c
+++ b/net/bridge/br_sysfs_if.c
@@ -158,6 +158,7 @@ static BRPORT_ATTR(flush, S_IWUSR, NULL, store_flush);
 BRPORT_ATTR_FLAG(hairpin_mode, BR_HAIRPIN_MODE);
 BRPORT_ATTR_FLAG(bpdu_guard, BR_BPDU_GUARD);
 BRPORT_ATTR_FLAG(root_block, BR_ROOT_BLOCK);
+BRPORT_ATTR_FLAG(learning, BR_LEARNING);
 
 #ifdef CONFIG_BRIDGE_IGMP_SNOOPING
 static ssize_t show_multicast_router(struct net_bridge_port *p, char *buf)
@@ -195,6 +196,7 @@ static const struct brport_attribute *brport_attrs[] = {
 	&brport_attr_hairpin_mode,
 	&brport_attr_bpdu_guard,
 	&brport_attr_root_block,
+	&brport_attr_learning,
 #ifdef CONFIG_BRIDGE_IGMP_SNOOPING
 	&brport_attr_multicast_router,
 	&brport_attr_multicast_fast_leave,
-- 
1.7.7.6

Re: [PATCHv2 net-next 1/2] bridge: Add flag to control mac learning.

[Michael S. Tsirkin]

On Mon, Apr 29, 2013 at 01:35:44PM -0400, Vlad Yasevich wrote:
> Allow user to control whether mac learning is enabled on the port.
> By default, mac learning is enabled.  Disabling mac learning will
> cause new dynamic FDB entries to not be created for a particular port.
> 
> Signed-off-by: Vlad Yasevich <vyasevic@redhat.com>
> ---
>  include/uapi/linux/if_link.h |    1 +
>  net/bridge/br_fdb.c          |   17 ++++++++++++++---
>  net/bridge/br_if.c           |    2 +-
>  net/bridge/br_netlink.c      |    6 +++++-
>  net/bridge/br_private.h      |    1 +
>  net/bridge/br_sysfs_if.c     |    2 ++
>  6 files changed, 24 insertions(+), 5 deletions(-)
> 
> diff --git a/include/uapi/linux/if_link.h b/include/uapi/linux/if_link.h
> index e316354..80fad7f 100644
> --- a/include/uapi/linux/if_link.h
> +++ b/include/uapi/linux/if_link.h
> @@ -221,6 +221,7 @@ enum {
>  	IFLA_BRPORT_GUARD,	/* bpdu guard              */
>  	IFLA_BRPORT_PROTECT,	/* root port protection    */
>  	IFLA_BRPORT_FAST_LEAVE,	/* multicast fast leave    */
> +	IFLA_BRPORT_LEARNING,	/* mac learning */
>  	__IFLA_BRPORT_MAX
>  };
>  #define IFLA_BRPORT_MAX (__IFLA_BRPORT_MAX - 1)
> diff --git a/net/bridge/br_fdb.c b/net/bridge/br_fdb.c
> index c581f12..f29eb0b 100644
> --- a/net/bridge/br_fdb.c
> +++ b/net/bridge/br_fdb.c
> @@ -446,8 +446,9 @@ int br_fdb_insert(struct net_bridge *br, struct net_bridge_port *source,
>  	return ret;
>  }
>  
> -void br_fdb_update(struct net_bridge *br, struct net_bridge_port *source,
> -		   const unsigned char *addr, u16 vid)
> +static void __br_fdb_update(struct net_bridge *br,
> +			    struct net_bridge_port *source,
> +		     	    const unsigned char *addr, u16 vid)
>  {
>  	struct hlist_head *head = &br->hash[br_mac_hash(addr, vid)];
>  	struct net_bridge_fdb_entry *fdb;
> @@ -481,6 +482,7 @@ void br_fdb_update(struct net_bridge *br, struct net_bridge_port *source,
>  			if (fdb)
>  				fdb_notify(br, fdb, RTM_NEWNEIGH);
>  		}
> +unlock:
>  		/* else  we lose race and someone else inserts
>  		 * it first, don't bother updating
>  		 */


Unused now?

> @@ -488,6 +490,15 @@ void br_fdb_update(struct net_bridge *br, struct net_bridge_port *source,
>  	}
>  }
>  
> +void br_fdb_update(struct net_bridge *br, struct net_bridge_port *source,
> +		     const unsigned char *addr, u16 vid)
> +{
> +	if (!(source->flags & BR_LEARNING))
> +		return;
> +
> +	__br_fdb_update(br, source, addr, vid);
> +}
> +
>  static int fdb_to_nud(const struct net_bridge_fdb_entry *fdb)
>  {
>  	if (fdb->is_local)
> @@ -653,7 +664,7 @@ static int __br_fdb_add(struct ndmsg *ndm, struct net_bridge_port *p,
>  
>  	if (ndm->ndm_flags & NTF_USE) {
>  		rcu_read_lock();
> -		br_fdb_update(p->br, p, addr, vid);
> +		__br_fdb_update(p->br, p, addr, vid);
>  		rcu_read_unlock();
>  	} else {
>  		spin_lock_bh(&p->br->hash_lock);
> diff --git a/net/bridge/br_if.c b/net/bridge/br_if.c
> index f17fcb3..9751103 100644
> --- a/net/bridge/br_if.c
> +++ b/net/bridge/br_if.c
> @@ -220,7 +220,7 @@ static struct net_bridge_port *new_nbp(struct net_bridge *br,
>  	p->path_cost = port_cost(dev);
>  	p->priority = 0x8000 >> BR_PORT_BITS;
>  	p->port_no = index;
> -	p->flags = 0;
> +	p->flags = BR_LEARNING;
>  	br_init_port(p);
>  	p->state = BR_STATE_DISABLED;
>  	br_stp_port_timer_init(p);
> diff --git a/net/bridge/br_netlink.c b/net/bridge/br_netlink.c
> index 8e3abf5..ce902bf 100644
> --- a/net/bridge/br_netlink.c
> +++ b/net/bridge/br_netlink.c
> @@ -30,6 +30,7 @@ static inline size_t br_port_info_size(void)
>  		+ nla_total_size(1)	/* IFLA_BRPORT_GUARD */
>  		+ nla_total_size(1)	/* IFLA_BRPORT_PROTECT */
>  		+ nla_total_size(1)	/* IFLA_BRPORT_FAST_LEAVE */
> +		+ nla_total_size(1)	/* IFLA_BRPORT_LEARNING */
>  		+ 0;
>  }
>  
> @@ -56,7 +57,8 @@ static int br_port_fill_attrs(struct sk_buff *skb,
>  	    nla_put_u8(skb, IFLA_BRPORT_MODE, mode) ||
>  	    nla_put_u8(skb, IFLA_BRPORT_GUARD, !!(p->flags & BR_BPDU_GUARD)) ||
>  	    nla_put_u8(skb, IFLA_BRPORT_PROTECT, !!(p->flags & BR_ROOT_BLOCK)) ||
> -	    nla_put_u8(skb, IFLA_BRPORT_FAST_LEAVE, !!(p->flags & BR_MULTICAST_FAST_LEAVE)))
> +	    nla_put_u8(skb, IFLA_BRPORT_FAST_LEAVE, !!(p->flags & BR_MULTICAST_FAST_LEAVE)) ||
> +	    nla_put_u8(skb, IFLA_BRPORT_LEARNING, !!(p->flags & BR_LEARNING)))
>  		return -EMSGSIZE;
>  
>  	return 0;
> @@ -281,6 +283,7 @@ static const struct nla_policy ifla_brport_policy[IFLA_BRPORT_MAX + 1] = {
>  	[IFLA_BRPORT_MODE]	= { .type = NLA_U8 },
>  	[IFLA_BRPORT_GUARD]	= { .type = NLA_U8 },
>  	[IFLA_BRPORT_PROTECT]	= { .type = NLA_U8 },
> +	[IFLA_BRPORT_LEARNING]	= { .type = NLA_U8 },
>  };
>  
>  /* Change the state of the port and notify spanning tree */
> @@ -328,6 +331,7 @@ static int br_setport(struct net_bridge_port *p, struct nlattr *tb[])
>  	br_set_port_flag(p, tb, IFLA_BRPORT_GUARD, BR_BPDU_GUARD);
>  	br_set_port_flag(p, tb, IFLA_BRPORT_FAST_LEAVE, BR_MULTICAST_FAST_LEAVE);
>  	br_set_port_flag(p, tb, IFLA_BRPORT_PROTECT, BR_ROOT_BLOCK);
> +	br_set_port_flag(p, tb, IFLA_BRPORT_LEARNING, BR_LEARNING);
>  
>  	if (tb[IFLA_BRPORT_COST]) {
>  		err = br_stp_set_path_cost(p, nla_get_u32(tb[IFLA_BRPORT_COST]));
> diff --git a/net/bridge/br_private.h b/net/bridge/br_private.h
> index 3cbf5be..67842b9 100644
> --- a/net/bridge/br_private.h
> +++ b/net/bridge/br_private.h
> @@ -156,6 +156,7 @@ struct net_bridge_port
>  #define BR_BPDU_GUARD           0x00000002
>  #define BR_ROOT_BLOCK		0x00000004
>  #define BR_MULTICAST_FAST_LEAVE	0x00000008
> +#define BR_LEARNING		0x00000010
>  
>  #ifdef CONFIG_BRIDGE_IGMP_SNOOPING
>  	u32				multicast_startup_queries_sent;
> diff --git a/net/bridge/br_sysfs_if.c b/net/bridge/br_sysfs_if.c
> index a1ef1b6..707f362 100644
> --- a/net/bridge/br_sysfs_if.c
> +++ b/net/bridge/br_sysfs_if.c
> @@ -158,6 +158,7 @@ static BRPORT_ATTR(flush, S_IWUSR, NULL, store_flush);
>  BRPORT_ATTR_FLAG(hairpin_mode, BR_HAIRPIN_MODE);
>  BRPORT_ATTR_FLAG(bpdu_guard, BR_BPDU_GUARD);
>  BRPORT_ATTR_FLAG(root_block, BR_ROOT_BLOCK);
> +BRPORT_ATTR_FLAG(learning, BR_LEARNING);
>  
>  #ifdef CONFIG_BRIDGE_IGMP_SNOOPING
>  static ssize_t show_multicast_router(struct net_bridge_port *p, char *buf)
> @@ -195,6 +196,7 @@ static const struct brport_attribute *brport_attrs[] = {
>  	&brport_attr_hairpin_mode,
>  	&brport_attr_bpdu_guard,
>  	&brport_attr_root_block,
> +	&brport_attr_learning,
>  #ifdef CONFIG_BRIDGE_IGMP_SNOOPING
>  	&brport_attr_multicast_router,
>  	&brport_attr_multicast_fast_leave,
> -- 
> 1.7.7.6

Re: [PATCHv2 net-next 1/2] bridge: Add flag to control mac learning.

[Stephen Hemminger]

On Mon, 29 Apr 2013 13:35:44 -0400
Vlad Yasevich <vyasevic@redhat.com> wrote:

> Allow user to control whether mac learning is enabled on the port.
> By default, mac learning is enabled.  Disabling mac learning will
> cause new dynamic FDB entries to not be created for a particular port.
> 
> Signed-off-by: Vlad Yasevich <vyasevic@redhat.com>

This looks like a good feature, thank you for doing it. It needed a couple
more changes before it is ready.

Your patch fails against current net-next, there is a new flag ADMIN_COST
that is not in your tree.

Also, it is clearer to just move the flag test back into the input side,
rather than adding wrapper functions.


Subject: [PATCHv3 net-next 1/2] bridge: Add flag to control mac learning.

Allow user to control whether mac learning is enabled on the port.
By default, mac learning is enabled.  Disabling mac learning will
cause new dynamic FDB entries to not be created for a particular port.

Signed-off-by: Vlad Yasevich <vyasevic@redhat.com>
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>

---
 include/uapi/linux/if_link.h |    1 +
 net/bridge/br_if.c           |    2 +-
 net/bridge/br_input.c        |    7 +++++--
 net/bridge/br_netlink.c      |    6 +++++-
 net/bridge/br_private.h      |    1 +
 net/bridge/br_sysfs_if.c     |    2 ++
 6 files changed, 15 insertions(+), 4 deletions(-)

--- a/include/uapi/linux/if_link.h	2013-04-29 15:45:49.000000000 -0700
+++ b/include/uapi/linux/if_link.h	2013-04-29 16:16:04.246906829 -0700
@@ -221,6 +221,7 @@ enum {
 	IFLA_BRPORT_GUARD,	/* bpdu guard              */
 	IFLA_BRPORT_PROTECT,	/* root port protection    */
 	IFLA_BRPORT_FAST_LEAVE,	/* multicast fast leave    */
+	IFLA_BRPORT_LEARNING,	/* mac learning */
 	__IFLA_BRPORT_MAX
 };
 #define IFLA_BRPORT_MAX (__IFLA_BRPORT_MAX - 1)
--- a/net/bridge/br_if.c	2013-04-26 16:27:59.000000000 -0700
+++ b/net/bridge/br_if.c	2013-04-29 16:16:04.250906777 -0700
@@ -221,7 +221,7 @@ static struct net_bridge_port *new_nbp(s
 	p->path_cost = port_cost(dev);
 	p->priority = 0x8000 >> BR_PORT_BITS;
 	p->port_no = index;
-	p->flags = 0;
+	p->flags = BR_LEARNING;
 	br_init_port(p);
 	p->state = BR_STATE_DISABLED;
 	br_stp_port_timer_init(p);
--- a/net/bridge/br_netlink.c	2013-03-28 14:26:20.000000000 -0700
+++ b/net/bridge/br_netlink.c	2013-04-29 16:16:04.250906777 -0700
@@ -30,6 +30,7 @@ static inline size_t br_port_info_size(v
 		+ nla_total_size(1)	/* IFLA_BRPORT_GUARD */
 		+ nla_total_size(1)	/* IFLA_BRPORT_PROTECT */
 		+ nla_total_size(1)	/* IFLA_BRPORT_FAST_LEAVE */
+		+ nla_total_size(1)	/* IFLA_BRPORT_LEARNING */
 		+ 0;
 }
 
@@ -56,7 +57,8 @@ static int br_port_fill_attrs(struct sk_
 	    nla_put_u8(skb, IFLA_BRPORT_MODE, mode) ||
 	    nla_put_u8(skb, IFLA_BRPORT_GUARD, !!(p->flags & BR_BPDU_GUARD)) ||
 	    nla_put_u8(skb, IFLA_BRPORT_PROTECT, !!(p->flags & BR_ROOT_BLOCK)) ||
-	    nla_put_u8(skb, IFLA_BRPORT_FAST_LEAVE, !!(p->flags & BR_MULTICAST_FAST_LEAVE)))
+	    nla_put_u8(skb, IFLA_BRPORT_FAST_LEAVE, !!(p->flags & BR_MULTICAST_FAST_LEAVE)) ||
+	    nla_put_u8(skb, IFLA_BRPORT_LEARNING, !!(p->flags & BR_LEARNING)))
 		return -EMSGSIZE;
 
 	return 0;
@@ -281,6 +283,7 @@ static const struct nla_policy ifla_brpo
 	[IFLA_BRPORT_MODE]	= { .type = NLA_U8 },
 	[IFLA_BRPORT_GUARD]	= { .type = NLA_U8 },
 	[IFLA_BRPORT_PROTECT]	= { .type = NLA_U8 },
+	[IFLA_BRPORT_LEARNING]	= { .type = NLA_U8 },
 };
 
 /* Change the state of the port and notify spanning tree */
@@ -328,6 +331,7 @@ static int br_setport(struct net_bridge_
 	br_set_port_flag(p, tb, IFLA_BRPORT_GUARD, BR_BPDU_GUARD);
 	br_set_port_flag(p, tb, IFLA_BRPORT_FAST_LEAVE, BR_MULTICAST_FAST_LEAVE);
 	br_set_port_flag(p, tb, IFLA_BRPORT_PROTECT, BR_ROOT_BLOCK);
+	br_set_port_flag(p, tb, IFLA_BRPORT_LEARNING, BR_LEARNING);
 
 	if (tb[IFLA_BRPORT_COST]) {
 		err = br_stp_set_path_cost(p, nla_get_u32(tb[IFLA_BRPORT_COST]));
--- a/net/bridge/br_private.h	2013-04-26 16:27:59.000000000 -0700
+++ b/net/bridge/br_private.h	2013-04-29 16:19:16.728417454 -0700
@@ -157,6 +157,7 @@ struct net_bridge_port
 #define BR_ROOT_BLOCK		0x00000004
 #define BR_MULTICAST_FAST_LEAVE	0x00000008
 #define BR_ADMIN_COST		0x00000010
+#define BR_LEARNING		0x00000020
 
 #ifdef CONFIG_BRIDGE_IGMP_SNOOPING
 	u32				multicast_startup_queries_sent;
--- a/net/bridge/br_sysfs_if.c	2013-03-07 18:12:53.000000000 -0800
+++ b/net/bridge/br_sysfs_if.c	2013-04-29 16:16:04.250906777 -0700
@@ -158,6 +158,7 @@ static BRPORT_ATTR(flush, S_IWUSR, NULL,
 BRPORT_ATTR_FLAG(hairpin_mode, BR_HAIRPIN_MODE);
 BRPORT_ATTR_FLAG(bpdu_guard, BR_BPDU_GUARD);
 BRPORT_ATTR_FLAG(root_block, BR_ROOT_BLOCK);
+BRPORT_ATTR_FLAG(learning, BR_LEARNING);
 
 #ifdef CONFIG_BRIDGE_IGMP_SNOOPING
 static ssize_t show_multicast_router(struct net_bridge_port *p, char *buf)
@@ -195,6 +196,7 @@ static const struct brport_attribute *br
 	&brport_attr_hairpin_mode,
 	&brport_attr_bpdu_guard,
 	&brport_attr_root_block,
+	&brport_attr_learning,
 #ifdef CONFIG_BRIDGE_IGMP_SNOOPING
 	&brport_attr_multicast_router,
 	&brport_attr_multicast_fast_leave,
--- a/net/bridge/br_input.c	2013-03-14 14:22:58.151958077 -0700
+++ b/net/bridge/br_input.c	2013-04-29 16:25:20.467713190 -0700
@@ -75,7 +75,9 @@ int br_handle_frame_finish(struct sk_buf
 
 	/* insert into forwarding database after filtering to avoid spoofing */
 	br = p->br;
-	br_fdb_update(br, p, eth_hdr(skb)->h_source, vid);
+
+	if (p->flags & BR_LEARNING)
+		br_fdb_update(br, p, eth_hdr(skb)->h_source, vid);
 
 	if (!is_broadcast_ether_addr(dest) && is_multicast_ether_addr(dest) &&
 	    br_multicast_rcv(br, p, skb))
@@ -142,7 +144,8 @@ static int br_handle_local_finish(struct
 	u16 vid = 0;
 
 	br_vlan_get_tag(skb, &vid);
-	br_fdb_update(p->br, p, eth_hdr(skb)->h_source, vid);
+	if (p->flags & BR_LEARNING)
+		br_fdb_update(p->br, p, eth_hdr(skb)->h_source, vid);
 	return 0;	 /* process further */
 }
 

[PATCHv2 net-next 2/2] bridge: Add a flag to control unicast packet flood.

[Vlad Yasevich]

Add a flag to control flood of unicast traffic.  By default, flood is
on and the bridge will flood unicast traffic if it doesn't know
the destination.  When the flag is turned off, unicast traffic
without an FDB will not be forwarded to the specified port.

Signed-off-by: Vlad Yasevich <vyasevic@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>

---
 include/uapi/linux/if_link.h |    1 +
 net/bridge/br_device.c       |    8 ++++----
 net/bridge/br_forward.c      |   14 +++++++++-----
 net/bridge/br_if.c           |    2 +-
 net/bridge/br_input.c        |    9 ++++++---
 net/bridge/br_netlink.c      |    6 +++++-
 net/bridge/br_private.h      |    6 ++++--
 net/bridge/br_sysfs_if.c     |    2 ++
 8 files changed, 32 insertions(+), 16 deletions(-)

diff --git a/include/uapi/linux/if_link.h b/include/uapi/linux/if_link.h
index 80fad7f..8d32021 100644
--- a/include/uapi/linux/if_link.h
+++ b/include/uapi/linux/if_link.h
@@ -222,6 +222,7 @@ enum {
 	IFLA_BRPORT_PROTECT,	/* root port protection    */
 	IFLA_BRPORT_FAST_LEAVE,	/* multicast fast leave    */
 	IFLA_BRPORT_LEARNING,	/* mac learning */
+	IFLA_BRPORT_UNICAST_FLOOD,	/* control flood of unicast traffic */
 	__IFLA_BRPORT_MAX
 };
 #define IFLA_BRPORT_MAX (__IFLA_BRPORT_MAX - 1)
diff --git a/net/bridge/br_device.c b/net/bridge/br_device.c
index 9673128..014d690 100644
--- a/net/bridge/br_device.c
+++ b/net/bridge/br_device.c
@@ -55,10 +55,10 @@ netdev_tx_t br_dev_xmit(struct sk_buff *skb, struct net_device *dev)
 	skb_pull(skb, ETH_HLEN);
 
 	if (is_broadcast_ether_addr(dest))
-		br_flood_deliver(br, skb);
+		br_flood_deliver(br, skb, false);
 	else if (is_multicast_ether_addr(dest)) {
 		if (unlikely(netpoll_tx_running(dev))) {
-			br_flood_deliver(br, skb);
+			br_flood_deliver(br, skb, false);
 			goto out;
 		}
 		if (br_multicast_rcv(br, NULL, skb)) {
@@ -70,11 +70,11 @@ netdev_tx_t br_dev_xmit(struct sk_buff *skb, struct net_device *dev)
 		if (mdst || BR_INPUT_SKB_CB_MROUTERS_ONLY(skb))
 			br_multicast_deliver(mdst, skb);
 		else
-			br_flood_deliver(br, skb);
+			br_flood_deliver(br, skb, false);
 	} else if ((dst = __br_fdb_get(br, dest, vid)) != NULL)
 		br_deliver(dst->dst, skb);
 	else
-		br_flood_deliver(br, skb);
+		br_flood_deliver(br, skb, true);
 
 out:
 	rcu_read_unlock();
diff --git a/net/bridge/br_forward.c b/net/bridge/br_forward.c
index 092b20e..3ca17eb 100644
--- a/net/bridge/br_forward.c
+++ b/net/bridge/br_forward.c
@@ -174,7 +174,8 @@ out:
 static void br_flood(struct net_bridge *br, struct sk_buff *skb,
 		     struct sk_buff *skb0,
 		     void (*__packet_hook)(const struct net_bridge_port *p,
-					   struct sk_buff *skb))
+					   struct sk_buff *skb),
+		     bool unicast)
 {
 	struct net_bridge_port *p;
 	struct net_bridge_port *prev;
@@ -182,6 +183,9 @@ static void br_flood(struct net_bridge *br, struct sk_buff *skb,
 	prev = NULL;
 
 	list_for_each_entry_rcu(p, &br->port_list, list) {
+		/* Do not flood unicast traffic to ports that turn it off */
+		if (unicast && !(p->flags & BR_UNICAST_FLOOD))
+			continue;
 		prev = maybe_deliver(prev, p, skb, __packet_hook);
 		if (IS_ERR(prev))
 			goto out;
@@ -203,16 +207,16 @@ out:
 
 
 /* called with rcu_read_lock */
-void br_flood_deliver(struct net_bridge *br, struct sk_buff *skb)
+void br_flood_deliver(struct net_bridge *br, struct sk_buff *skb, bool unicast)
 {
-	br_flood(br, skb, NULL, __br_deliver);
+	br_flood(br, skb, NULL, __br_deliver, unicast);
 }
 
 /* called under bridge lock */
 void br_flood_forward(struct net_bridge *br, struct sk_buff *skb,
-		      struct sk_buff *skb2)
+		      struct sk_buff *skb2, bool unicast)
 {
-	br_flood(br, skb, skb2, __br_forward);
+	br_flood(br, skb, skb2, __br_forward, unicast);
 }
 
 #ifdef CONFIG_BRIDGE_IGMP_SNOOPING
diff --git a/net/bridge/br_if.c b/net/bridge/br_if.c
index 9751103..3b15668 100644
--- a/net/bridge/br_if.c
+++ b/net/bridge/br_if.c
@@ -220,7 +220,7 @@ static struct net_bridge_port *new_nbp(struct net_bridge *br,
 	p->path_cost = port_cost(dev);
 	p->priority = 0x8000 >> BR_PORT_BITS;
 	p->port_no = index;
-	p->flags = BR_LEARNING;
+	p->flags = BR_LEARNING | BR_UNICAST_FLOOD;
 	br_init_port(p);
 	p->state = BR_STATE_DISABLED;
 	br_stp_port_timer_init(p);
diff --git a/net/bridge/br_input.c b/net/bridge/br_input.c
index 828e2bc..6b6a49e 100644
--- a/net/bridge/br_input.c
+++ b/net/bridge/br_input.c
@@ -65,6 +65,7 @@ int br_handle_frame_finish(struct sk_buff *skb)
 	struct net_bridge_fdb_entry *dst;
 	struct net_bridge_mdb_entry *mdst;
 	struct sk_buff *skb2;
+	bool unicast = true;
 	u16 vid = 0;
 
 	if (!p || p->state == BR_STATE_DISABLED)
@@ -94,9 +95,10 @@ int br_handle_frame_finish(struct sk_buff *skb)
 
 	dst = NULL;
 
-	if (is_broadcast_ether_addr(dest))
+	if (is_broadcast_ether_addr(dest)) {
 		skb2 = skb;
-	else if (is_multicast_ether_addr(dest)) {
+		unicast = false;
+	} else if (is_multicast_ether_addr(dest)) {
 		mdst = br_mdb_get(br, skb, vid);
 		if (mdst || BR_INPUT_SKB_CB_MROUTERS_ONLY(skb)) {
 			if ((mdst && mdst->mglist) ||
@@ -109,6 +111,7 @@ int br_handle_frame_finish(struct sk_buff *skb)
 		} else
 			skb2 = skb;
 
+		unicast = false;
 		br->dev->stats.multicast++;
 	} else if ((dst = __br_fdb_get(br, dest, vid)) &&
 			dst->is_local) {
@@ -122,7 +125,7 @@ int br_handle_frame_finish(struct sk_buff *skb)
 			dst->used = jiffies;
 			br_forward(dst->dst, skb, skb2);
 		} else
-			br_flood_forward(br, skb, skb2);
+			br_flood_forward(br, skb, skb2, unicast);
 	}
 
 	if (skb2)
diff --git a/net/bridge/br_netlink.c b/net/bridge/br_netlink.c
index ce902bf..d27599c 100644
--- a/net/bridge/br_netlink.c
+++ b/net/bridge/br_netlink.c
@@ -31,6 +31,7 @@ static inline size_t br_port_info_size(void)
 		+ nla_total_size(1)	/* IFLA_BRPORT_PROTECT */
 		+ nla_total_size(1)	/* IFLA_BRPORT_FAST_LEAVE */
 		+ nla_total_size(1)	/* IFLA_BRPORT_LEARNING */
+		+ nla_total_size(1)	/* IFLA_BRPORT_UNICAST_FLOOD */
 		+ 0;
 }
 
@@ -58,7 +59,8 @@ static int br_port_fill_attrs(struct sk_buff *skb,
 	    nla_put_u8(skb, IFLA_BRPORT_GUARD, !!(p->flags & BR_BPDU_GUARD)) ||
 	    nla_put_u8(skb, IFLA_BRPORT_PROTECT, !!(p->flags & BR_ROOT_BLOCK)) ||
 	    nla_put_u8(skb, IFLA_BRPORT_FAST_LEAVE, !!(p->flags & BR_MULTICAST_FAST_LEAVE)) ||
-	    nla_put_u8(skb, IFLA_BRPORT_LEARNING, !!(p->flags & BR_LEARNING)))
+	    nla_put_u8(skb, IFLA_BRPORT_LEARNING, !!(p->flags & BR_LEARNING)) ||
+	    nla_put_u8(skb, IFLA_BRPORT_UNICAST_FLOOD, !!(p->flags & BR_UNICAST_FLOOD)))
 		return -EMSGSIZE;
 
 	return 0;
@@ -284,6 +286,7 @@ static const struct nla_policy ifla_brport_policy[IFLA_BRPORT_MAX + 1] = {
 	[IFLA_BRPORT_GUARD]	= { .type = NLA_U8 },
 	[IFLA_BRPORT_PROTECT]	= { .type = NLA_U8 },
 	[IFLA_BRPORT_LEARNING]	= { .type = NLA_U8 },
+	[IFLA_BRPORT_UNICAST_FLOOD] = { .type = NLA_U8 },
 };
 
 /* Change the state of the port and notify spanning tree */
@@ -332,6 +335,7 @@ static int br_setport(struct net_bridge_port *p, struct nlattr *tb[])
 	br_set_port_flag(p, tb, IFLA_BRPORT_FAST_LEAVE, BR_MULTICAST_FAST_LEAVE);
 	br_set_port_flag(p, tb, IFLA_BRPORT_PROTECT, BR_ROOT_BLOCK);
 	br_set_port_flag(p, tb, IFLA_BRPORT_LEARNING, BR_LEARNING);
+	br_set_port_flag(p, tb, IFLA_BRPORT_UNICAST_FLOOD, BR_UNICAST_FLOOD);
 
 	if (tb[IFLA_BRPORT_COST]) {
 		err = br_stp_set_path_cost(p, nla_get_u32(tb[IFLA_BRPORT_COST]));
diff --git a/net/bridge/br_private.h b/net/bridge/br_private.h
index 67842b9..50b6f55 100644
--- a/net/bridge/br_private.h
+++ b/net/bridge/br_private.h
@@ -157,6 +157,7 @@ struct net_bridge_port
 #define BR_ROOT_BLOCK		0x00000004
 #define BR_MULTICAST_FAST_LEAVE	0x00000008
 #define BR_LEARNING		0x00000010
+#define BR_UNICAST_FLOOD	0x00000020
 
 #ifdef CONFIG_BRIDGE_IGMP_SNOOPING
 	u32				multicast_startup_queries_sent;
@@ -411,9 +412,10 @@ extern int br_dev_queue_push_xmit(struct sk_buff *skb);
 extern void br_forward(const struct net_bridge_port *to,
 		struct sk_buff *skb, struct sk_buff *skb0);
 extern int br_forward_finish(struct sk_buff *skb);
-extern void br_flood_deliver(struct net_bridge *br, struct sk_buff *skb);
+extern void br_flood_deliver(struct net_bridge *br, struct sk_buff *skb,
+			     bool unicast);
 extern void br_flood_forward(struct net_bridge *br, struct sk_buff *skb,
-			     struct sk_buff *skb2);
+			     struct sk_buff *skb2, bool unicast);
 
 /* br_if.c */
 extern void br_port_carrier_check(struct net_bridge_port *p);
diff --git a/net/bridge/br_sysfs_if.c b/net/bridge/br_sysfs_if.c
index 707f362..ea6990f 100644
--- a/net/bridge/br_sysfs_if.c
+++ b/net/bridge/br_sysfs_if.c
@@ -159,6 +159,7 @@ BRPORT_ATTR_FLAG(hairpin_mode, BR_HAIRPIN_MODE);
 BRPORT_ATTR_FLAG(bpdu_guard, BR_BPDU_GUARD);
 BRPORT_ATTR_FLAG(root_block, BR_ROOT_BLOCK);
 BRPORT_ATTR_FLAG(learning, BR_LEARNING);
+BRPORT_ATTR_FLAG(unicast_flood, BR_UNICAST_FLOOD);
 
 #ifdef CONFIG_BRIDGE_IGMP_SNOOPING
 static ssize_t show_multicast_router(struct net_bridge_port *p, char *buf)
@@ -197,6 +198,7 @@ static const struct brport_attribute *brport_attrs[] = {
 	&brport_attr_bpdu_guard,
 	&brport_attr_root_block,
 	&brport_attr_learning,
+	&brport_attr_unicast_flood,
 #ifdef CONFIG_BRIDGE_IGMP_SNOOPING
 	&brport_attr_multicast_router,
 	&brport_attr_multicast_fast_leave,
-- 
1.7.7.6

Re: [PATCHv2 net-next 2/2] bridge: Add a flag to control unicast packet flood.

[Stephen Hemminger]

On Mon, 29 Apr 2013 13:35:45 -0400
Vlad Yasevich <vyasevic@redhat.com> wrote:

> Add a flag to control flood of unicast traffic.  By default, flood is
> on and the bridge will flood unicast traffic if it doesn't know
> the destination.  When the flag is turned off, unicast traffic
> without an FDB will not be forwarded to the specified port.
> 
> Signed-off-by: Vlad Yasevich <vyasevic@redhat.com>
> Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
> 

This a good idea and thank you for doing it. I like that the flag is expressed
as a positive value (rather than an inverse value like no-flood).

The name BR_UNICAST_FLOOD is too long, just use BR_FLOOD and keep the code
shorter.

This doesn't apply against current net-next (same problem as first patch).

Also, I am not a fan of having lots of boolean flag variables in normal
code paths. It ends up reading like PASCAL code. But probably unavoidable in this case.

Re: [PATCHv2 net-next 2/2] bridge: Add a flag to control unicast packet flood.

[Vlad Yasevich]

On 04/29/2013 07:43 PM, Stephen Hemminger wrote:
> On Mon, 29 Apr 2013 13:35:45 -0400
> Vlad Yasevich <vyasevic@redhat.com> wrote:
>
>> Add a flag to control flood of unicast traffic.  By default, flood is
>> on and the bridge will flood unicast traffic if it doesn't know
>> the destination.  When the flag is turned off, unicast traffic
>> without an FDB will not be forwarded to the specified port.
>>
>> Signed-off-by: Vlad Yasevich <vyasevic@redhat.com>
>> Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
>>
>
> This a good idea and thank you for doing it. I like that the flag is expressed
> as a positive value (rather than an inverse value like no-flood).
>
> The name BR_UNICAST_FLOOD is too long, just use BR_FLOOD and keep the code
> shorter.
>
> This doesn't apply against current net-next (same problem as first patch).
>
> Also, I am not a fan of having lots of boolean flag variables in normal
> code paths. It ends up reading like PASCAL code. But probably unavoidable in this case.
>

Sorry, I'll rebase and resubmit.  The alternative there is to possibly
flag skb->cb.  Then we wouldn't need to pass the argument around.
If you'd rather I do that, it's easy enough.

Thanks
-vlad