pull request: batman-adv 2012-04-25

pull request: batman-adv 2012-04-25

[Antonio Quartulli]

I'd like to propose the following changes for net-next/linux-3.5.

The first 8 patches introduce a new feature called D.A.T. (Distributed ARP
Table) which is the resulting implementation of Antonio Quartulli's GSOC 2011
Project[0]. This feature aims to improve reliability of ARP resolutions in
sparse wireless mesh networks by creating either a local and a
network-wide-distributed ARP cache. Major details can be found on its
wikipage[1].
It is an optional feature and it is possible to choose whether to compile it
or not by using the proper Kconfig option (=n by default). We are evaluating the
possibility of introducing a runtime switch to turn the behaviour on an off
without recompiling the module (like for the bridge loop avoidance).

Patch 1 also introduces a new packet type, but backward compatibility is
preserved.
Patch 9 fixes a bug in the "gw propagation" feature introduced so far.
Patch 11 and 12 are yet other improvements for the new
routing-protocol-API.

The others are minor fixes and cleanups.

Thank you very much,
	Antonio

[0] http://www.google-melange.com/gsoc/project/google/gsoc2011/ordex/4001
[1] http://www.open-mesh.org/wiki/batman-adv/DistributedArpTable

The following changes since commit 7a2a66a0ac1cf93d30869c4ecbfc71a2fda19397:

  Add linux-next specific files for 20120423 (2012-04-23 16:58:43 +1000)

are available in the git repository at:

  git://git.open-mesh.org/linux-merge.git tags/batman-adv-for-davem

for you to fetch changes up to cc01b7a1f16291cc48fa6d261aa952d45604f10c:

  batman-adv: split neigh_new function into generic and batman iv specific parts (2012-04-24 10:11:31 +0200)

----------------------------------------------------------------
Included changes:

* a new feature has been introduced: D.A.T. (Distributed ARP Table). It is a
  mechanism based on DHT theory that creates a distributed (mesh network wide)
  ARP cache in order to speed up ARP resolutions in sparse wireless mesh
  networks.
* in order to satisfy DAT requirements a new unicast packet type, namely
  UNICAST_4ADDR, has been introduced. Backward compatibility has been kept: not
  updated nodes will simply drop the packet and ignore DAT mechanism.
* minor fixes and cleanups
* minor routing protocol API cleanups

----------------------------------------------------------------
Antonio Quartulli (9):
      batman-adv: add UNICAST_4ADDR packet type
      batman-adv: add a new log level for DAT debugging
      batman-adv: add biggest_unsigned_int(x) macro
      batman-adv: Distributed ARP Table - create DHT helper functions
      batman-adv: Distributed ARP Table - add ARP parsing functions
      batman-adv: Distributed ARP Table - add snooping functions for ARP messages
      batman-adv: Distributed ARP Table - increase default soft_iface ARP table timeout
      batman-adv: Distributed ARP Table - add compile option
      batman-adv: fix wrong dhcp option list browsing

Marek Lindner (6):
      batman-adv: introduce is_single_hop_neigh variable to increase readability
      batman-adv: introduce packet type handler array for incoming packets
      batman-adv: register batman ogm receive function during protocol init
      batman-adv: rename last_valid to last_seen
      batman-adv: replace HZ calculations with jiffies_to_msecs()
      batman-adv: split neigh_new function into generic and batman iv specific parts

 Documentation/networking/batman-adv.txt |    3 +-
 net/batman-adv/Kconfig                  |   10 +
 net/batman-adv/Makefile                 |    1 +
 net/batman-adv/bat_debugfs.c            |    4 +-
 net/batman-adv/bat_iv_ogm.c             |   95 +++--
 net/batman-adv/bat_sysfs.c              |    2 +-
 net/batman-adv/distributed-arp-table.c  |  605 +++++++++++++++++++++++++++++++
 net/batman-adv/distributed-arp-table.h  |  140 +++++++
 net/batman-adv/gateway_client.c         |    6 +-
 net/batman-adv/hard-interface.c         |  116 +-----
 net/batman-adv/main.c                   |  124 ++++++-
 net/batman-adv/main.h                   |   24 +-
 net/batman-adv/originator.c             |   52 +--
 net/batman-adv/originator.h             |    6 +-
 net/batman-adv/packet.h                 |   30 +-
 net/batman-adv/routing.c                |   30 +-
 net/batman-adv/routing.h                |    4 +-
 net/batman-adv/send.c                   |    6 +-
 net/batman-adv/soft-interface.c         |   17 +-
 net/batman-adv/types.h                  |   33 +-
 net/batman-adv/unicast.c                |  102 +++++-
 net/batman-adv/unicast.h                |   21 +-
 22 files changed, 1205 insertions(+), 226 deletions(-)
 create mode 100644 net/batman-adv/distributed-arp-table.c
 create mode 100644 net/batman-adv/distributed-arp-table.h

[B.A.T.M.A.N.] pull request: batman-adv 2012-04-25

[Antonio Quartulli]

I'd like to propose the following changes for net-next/linux-3.5.

The first 8 patches introduce a new feature called D.A.T. (Distributed ARP
Table) which is the resulting implementation of Antonio Quartulli's GSOC 2011
Project[0]. This feature aims to improve reliability of ARP resolutions in
sparse wireless mesh networks by creating either a local and a
network-wide-distributed ARP cache. Major details can be found on its
wikipage[1].
It is an optional feature and it is possible to choose whether to compile it
or not by using the proper Kconfig option (=n by default). We are evaluating the
possibility of introducing a runtime switch to turn the behaviour on an off
without recompiling the module (like for the bridge loop avoidance).

Patch 1 also introduces a new packet type, but backward compatibility is
preserved.
Patch 9 fixes a bug in the "gw propagation" feature introduced so far.
Patch 11 and 12 are yet other improvements for the new
routing-protocol-API.

The others are minor fixes and cleanups.

Thank you very much,
	Antonio

[0] http://www.google-melange.com/gsoc/project/google/gsoc2011/ordex/4001
[1] http://www.open-mesh.org/wiki/batman-adv/DistributedArpTable

The following changes since commit 7a2a66a0ac1cf93d30869c4ecbfc71a2fda19397:

  Add linux-next specific files for 20120423 (2012-04-23 16:58:43 +1000)

are available in the git repository at:

  git://git.open-mesh.org/linux-merge.git tags/batman-adv-for-davem

for you to fetch changes up to cc01b7a1f16291cc48fa6d261aa952d45604f10c:

  batman-adv: split neigh_new function into generic and batman iv specific parts (2012-04-24 10:11:31 +0200)

----------------------------------------------------------------
Included changes:

* a new feature has been introduced: D.A.T. (Distributed ARP Table). It is a
  mechanism based on DHT theory that creates a distributed (mesh network wide)
  ARP cache in order to speed up ARP resolutions in sparse wireless mesh
  networks.
* in order to satisfy DAT requirements a new unicast packet type, namely
  UNICAST_4ADDR, has been introduced. Backward compatibility has been kept: not
  updated nodes will simply drop the packet and ignore DAT mechanism.
* minor fixes and cleanups
* minor routing protocol API cleanups

----------------------------------------------------------------
Antonio Quartulli (9):
      batman-adv: add UNICAST_4ADDR packet type
      batman-adv: add a new log level for DAT debugging
      batman-adv: add biggest_unsigned_int(x) macro
      batman-adv: Distributed ARP Table - create DHT helper functions
      batman-adv: Distributed ARP Table - add ARP parsing functions
      batman-adv: Distributed ARP Table - add snooping functions for ARP messages
      batman-adv: Distributed ARP Table - increase default soft_iface ARP table timeout
      batman-adv: Distributed ARP Table - add compile option
      batman-adv: fix wrong dhcp option list browsing

Marek Lindner (6):
      batman-adv: introduce is_single_hop_neigh variable to increase readability
      batman-adv: introduce packet type handler array for incoming packets
      batman-adv: register batman ogm receive function during protocol init
      batman-adv: rename last_valid to last_seen
      batman-adv: replace HZ calculations with jiffies_to_msecs()
      batman-adv: split neigh_new function into generic and batman iv specific parts

 Documentation/networking/batman-adv.txt |    3 +-
 net/batman-adv/Kconfig                  |   10 +
 net/batman-adv/Makefile                 |    1 +
 net/batman-adv/bat_debugfs.c            |    4 +-
 net/batman-adv/bat_iv_ogm.c             |   95 +++--
 net/batman-adv/bat_sysfs.c              |    2 +-
 net/batman-adv/distributed-arp-table.c  |  605 +++++++++++++++++++++++++++++++
 net/batman-adv/distributed-arp-table.h  |  140 +++++++
 net/batman-adv/gateway_client.c         |    6 +-
 net/batman-adv/hard-interface.c         |  116 +-----
 net/batman-adv/main.c                   |  124 ++++++-
 net/batman-adv/main.h                   |   24 +-
 net/batman-adv/originator.c             |   52 +--
 net/batman-adv/originator.h             |    6 +-
 net/batman-adv/packet.h                 |   30 +-
 net/batman-adv/routing.c                |   30 +-
 net/batman-adv/routing.h                |    4 +-
 net/batman-adv/send.c                   |    6 +-
 net/batman-adv/soft-interface.c         |   17 +-
 net/batman-adv/types.h                  |   33 +-
 net/batman-adv/unicast.c                |  102 +++++-
 net/batman-adv/unicast.h                |   21 +-
 22 files changed, 1205 insertions(+), 226 deletions(-)
 create mode 100644 net/batman-adv/distributed-arp-table.c
 create mode 100644 net/batman-adv/distributed-arp-table.h

[PATCH 01/15] batman-adv: add UNICAST_4ADDR packet type

[Antonio Quartulli]

The current unicast packet type does not contain the orig source address. This
patches add a new unicast packet (called UNICAST_4ADDR) which provides two new
fields: the originator source address and the subtype (the type of the data
contained in the packet payload). The former is useful to identify the node
which injected the packet into the network and the latter is useful to avoid
creating new unicast packet types in the future: a macro defining a new subtype
will be enough.

Signed-off-by: Antonio Quartulli <ordex-GaUfNO9RBHfsrOwW+9ziJQ@public.gmane.org>
---
 net/batman-adv/hard-interface.c |    1 +
 net/batman-adv/packet.h         |   27 ++++++++---
 net/batman-adv/routing.c        |    8 ++-
 net/batman-adv/unicast.c        |  102 +++++++++++++++++++++++++++++++--------
 net/batman-adv/unicast.h        |   17 ++++++-
 5 files changed, 125 insertions(+), 30 deletions(-)

diff --git a/net/batman-adv/hard-interface.c b/net/batman-adv/hard-interface.c
index 47c79d7..220cb02 100644
--- a/net/batman-adv/hard-interface.c
+++ b/net/batman-adv/hard-interface.c
@@ -614,6 +614,7 @@ static int batman_skb_recv(struct sk_buff *skb, struct net_device *dev,
 
 		/* unicast packet */
 	case BAT_UNICAST:
+	case BAT_UNICAST_4ADDR:
 		ret = recv_unicast_packet(skb, hard_iface);
 		break;
 
diff --git a/net/batman-adv/packet.h b/net/batman-adv/packet.h
index f54969c..e9f5184 100644
--- a/net/batman-adv/packet.h
+++ b/net/batman-adv/packet.h
@@ -25,14 +25,19 @@
 #define ETH_P_BATMAN  0x4305	/* unofficial/not registered Ethertype */
 
 enum bat_packettype {
-	BAT_IV_OGM	 = 0x01,
-	BAT_ICMP	 = 0x02,
-	BAT_UNICAST	 = 0x03,
-	BAT_BCAST	 = 0x04,
-	BAT_VIS		 = 0x05,
-	BAT_UNICAST_FRAG = 0x06,
-	BAT_TT_QUERY	 = 0x07,
-	BAT_ROAM_ADV	 = 0x08
+	BAT_IV_OGM		= 0x01,
+	BAT_ICMP		= 0x02,
+	BAT_UNICAST		= 0x03,
+	BAT_BCAST		= 0x04,
+	BAT_VIS			= 0x05,
+	BAT_UNICAST_FRAG	= 0x06,
+	BAT_TT_QUERY		= 0x07,
+	BAT_ROAM_ADV		= 0x08,
+	BAT_UNICAST_4ADDR	= 0x09
+};
+
+enum bat_subtype {
+	BAT_P_DATA		= 0x01
 };
 
 /* this file is included by batctl which needs these defines */
@@ -159,6 +164,12 @@ struct unicast_packet {
 	uint8_t  dest[ETH_ALEN];
 } __packed;
 
+struct unicast_4addr_packet {
+	struct unicast_packet u;
+	uint8_t src[ETH_ALEN];
+	uint8_t subtype;
+} __packed;
+
 struct unicast_frag_packet {
 	struct batman_header header;
 	uint8_t  ttvn; /* destination translation table version number */
diff --git a/net/batman-adv/routing.c b/net/batman-adv/routing.c
index ff56086..5f19bbb 100644
--- a/net/batman-adv/routing.c
+++ b/net/batman-adv/routing.c
@@ -959,14 +959,18 @@ int recv_unicast_packet(struct sk_buff *skb, struct hard_iface *recv_if)
 	struct unicast_packet *unicast_packet;
 	int hdr_size = sizeof(*unicast_packet);
 
+	unicast_packet = (struct unicast_packet *)skb->data;
+
+	/* the caller function should have already pulled 2 bytes */
+	if (unicast_packet->header.packet_type == BAT_UNICAST_4ADDR)
+		hdr_size = sizeof(struct unicast_4addr_packet);
+
 	if (check_unicast_packet(skb, hdr_size) < 0)
 		return NET_RX_DROP;
 
 	if (!check_unicast_ttvn(bat_priv, skb))
 		return NET_RX_DROP;
 
-	unicast_packet = (struct unicast_packet *)skb->data;
-
 	/* packet for me */
 	if (is_my_mac(unicast_packet->dest)) {
 		interface_rx(recv_if->soft_iface, skb, recv_if, hdr_size);
diff --git a/net/batman-adv/unicast.c b/net/batman-adv/unicast.c
index 676f6a6..96c7cec 100644
--- a/net/batman-adv/unicast.c
+++ b/net/batman-adv/unicast.c
@@ -283,13 +283,78 @@ out:
 	return ret;
 }
 
-int unicast_send_skb(struct sk_buff *skb, struct bat_priv *bat_priv)
+static bool pull_and_fill_unicast(struct sk_buff *skb, int hdr_size,
+				  struct orig_node *orig_node)
+{
+	struct unicast_packet *unicast_packet;
+
+	if (my_skb_head_push(skb, hdr_size) < 0)
+		return false;
+
+	unicast_packet = (struct unicast_packet *)skb->data;
+	unicast_packet->header.version = COMPAT_VERSION;
+	/* batman packet type: unicast */
+	unicast_packet->header.packet_type = BAT_UNICAST;
+	/* set unicast ttl */
+	unicast_packet->header.ttl = TTL;
+	/* copy the destination for faster routing */
+	memcpy(unicast_packet->dest, orig_node->orig, ETH_ALEN);
+	/* set the destination tt version number */
+	unicast_packet->ttvn =
+		(uint8_t)atomic_read(&orig_node->last_ttvn);
+
+	return true;
+}
+
+static bool prepare_unicast_packet(struct sk_buff *skb,
+				   struct orig_node *orig_node)
+{
+	return pull_and_fill_unicast(skb, sizeof(struct unicast_packet),
+				     orig_node);
+}
+
+static bool prepare_unicast_4addr_packet(struct bat_priv *bat_priv,
+					 struct sk_buff *skb,
+					 struct orig_node *orig_node,
+					 int packet_subtype)
+{
+	struct hard_iface *primary_if;
+	struct unicast_4addr_packet *unicast_4addr_packet;
+	bool ret = false;
+
+	primary_if = primary_if_get_selected(bat_priv);
+	if (!primary_if)
+		goto out;
+
+	/* pull the header space and fill the unicast_packet substructure.
+	 * We can do that because the first member of the unicast_4addr_packet
+	 * is of type struct unicast_packet
+	 */
+	if (!pull_and_fill_unicast(skb, sizeof(*unicast_4addr_packet),
+				   orig_node))
+		goto out;
+
+	unicast_4addr_packet = (struct unicast_4addr_packet *)skb->data;
+	unicast_4addr_packet->u.header.packet_type = BAT_UNICAST_4ADDR;
+	memcpy(unicast_4addr_packet->src, primary_if->net_dev->dev_addr,
+	       ETH_ALEN);
+	unicast_4addr_packet->subtype = packet_subtype;
+
+	ret = true;
+out:
+	if (primary_if)
+		hardif_free_ref(primary_if);
+	return ret;
+}
+
+int unicast_generic_send_skb(struct sk_buff *skb, struct bat_priv *bat_priv,
+			     int packet_type, int packet_subtype)
 {
 	struct ethhdr *ethhdr = (struct ethhdr *)skb->data;
-	struct unicast_packet *unicast_packet;
 	struct orig_node *orig_node;
 	struct neigh_node *neigh_node;
 	int data_len = skb->len;
+	struct unicast_packet *unicast_packet;
 	int ret = 1;
 
 	/* get routing information */
@@ -303,7 +368,6 @@ int unicast_send_skb(struct sk_buff *skb, struct bat_priv *bat_priv)
 	 * returns NULL in case of AP isolation */
 	orig_node = transtable_search(bat_priv, ethhdr->h_source,
 				      ethhdr->h_dest);
-
 find_router:
 	/**
 	 * find_router():
@@ -311,27 +375,28 @@ find_router:
 	 *  - increases neigh_nodes refcount if found.
 	 */
 	neigh_node = find_router(bat_priv, orig_node, NULL);
-
 	if (!neigh_node)
 		goto out;
 
-	if (my_skb_head_push(skb, sizeof(*unicast_packet)) < 0)
+	switch (packet_type) {
+	case BAT_UNICAST:
+		prepare_unicast_packet(skb, orig_node);
+		break;
+	case BAT_UNICAST_4ADDR:
+		prepare_unicast_4addr_packet(bat_priv, skb, orig_node,
+					     packet_subtype);
+		break;
+	default:
+		/* this function supports UNICAST and UNICAST_4ADDR only. It
+		 * should never be invoked with any other packet type
+		 */
 		goto out;
+	}
 
 	unicast_packet = (struct unicast_packet *)skb->data;
-
-	unicast_packet->header.version = COMPAT_VERSION;
-	/* batman packet type: unicast */
-	unicast_packet->header.packet_type = BAT_UNICAST;
-	/* set unicast ttl */
-	unicast_packet->header.ttl = TTL;
-	/* copy the destination for faster routing */
-	memcpy(unicast_packet->dest, orig_node->orig, ETH_ALEN);
-	/* set the destination tt version number */
-	unicast_packet->ttvn =
-		(uint8_t)atomic_read(&orig_node->last_ttvn);
-
-	if (atomic_read(&bat_priv->fragmentation) &&
+	/* fragmentation mechanism only works for UNICAST (now) */
+	if (packet_type == BAT_UNICAST &&
+	    atomic_read(&bat_priv->fragmentation) &&
 	    data_len + sizeof(*unicast_packet) >
 				neigh_node->if_incoming->net_dev->mtu) {
 		/* send frag skb decreases ttl */
@@ -343,7 +408,6 @@ find_router:
 
 	send_skb_packet(skb, neigh_node->if_incoming, neigh_node->addr);
 	ret = 0;
-	goto out;
 
 out:
 	if (neigh_node)
diff --git a/net/batman-adv/unicast.h b/net/batman-adv/unicast.h
index a9faf6b..ae9775b 100644
--- a/net/batman-adv/unicast.h
+++ b/net/batman-adv/unicast.h
@@ -30,9 +30,24 @@
 int frag_reassemble_skb(struct sk_buff *skb, struct bat_priv *bat_priv,
 			struct sk_buff **new_skb);
 void frag_list_free(struct list_head *head);
-int unicast_send_skb(struct sk_buff *skb, struct bat_priv *bat_priv);
 int frag_send_skb(struct sk_buff *skb, struct bat_priv *bat_priv,
 		  struct hard_iface *hard_iface, const uint8_t dstaddr[]);
+int unicast_generic_send_skb(struct sk_buff *skb, struct bat_priv *bat_priv,
+			     int packet_type, int packet_subtype);
+
+static inline int unicast_send_skb(struct sk_buff *skb,
+				   struct bat_priv *bat_priv)
+{
+	return unicast_generic_send_skb(skb, bat_priv, BAT_UNICAST, 0);
+}
+
+static inline int unicast_4addr_send_skb(struct sk_buff *skb,
+					 struct bat_priv *bat_priv,
+					 int packet_subtype)
+{
+	return unicast_generic_send_skb(skb, bat_priv, BAT_UNICAST_4ADDR,
+					packet_subtype);
+}
 
 static inline int frag_can_reassemble(const struct sk_buff *skb, int mtu)
 {
-- 
1.7.9.4

[B.A.T.M.A.N.] [PATCH 01/15] batman-adv: add UNICAST_4ADDR packet type

[Antonio Quartulli]

The current unicast packet type does not contain the orig source address. This
patches add a new unicast packet (called UNICAST_4ADDR) which provides two new
fields: the originator source address and the subtype (the type of the data
contained in the packet payload). The former is useful to identify the node
which injected the packet into the network and the latter is useful to avoid
creating new unicast packet types in the future: a macro defining a new subtype
will be enough.

Signed-off-by: Antonio Quartulli <ordex@autistici.org>
---
 net/batman-adv/hard-interface.c |    1 +
 net/batman-adv/packet.h         |   27 ++++++++---
 net/batman-adv/routing.c        |    8 ++-
 net/batman-adv/unicast.c        |  102 +++++++++++++++++++++++++++++++--------
 net/batman-adv/unicast.h        |   17 ++++++-
 5 files changed, 125 insertions(+), 30 deletions(-)

diff --git a/net/batman-adv/hard-interface.c b/net/batman-adv/hard-interface.c
index 47c79d7..220cb02 100644
--- a/net/batman-adv/hard-interface.c
+++ b/net/batman-adv/hard-interface.c
@@ -614,6 +614,7 @@ static int batman_skb_recv(struct sk_buff *skb, struct net_device *dev,
 
 		/* unicast packet */
 	case BAT_UNICAST:
+	case BAT_UNICAST_4ADDR:
 		ret = recv_unicast_packet(skb, hard_iface);
 		break;
 
diff --git a/net/batman-adv/packet.h b/net/batman-adv/packet.h
index f54969c..e9f5184 100644
--- a/net/batman-adv/packet.h
+++ b/net/batman-adv/packet.h
@@ -25,14 +25,19 @@
 #define ETH_P_BATMAN  0x4305	/* unofficial/not registered Ethertype */
 
 enum bat_packettype {
-	BAT_IV_OGM	 = 0x01,
-	BAT_ICMP	 = 0x02,
-	BAT_UNICAST	 = 0x03,
-	BAT_BCAST	 = 0x04,
-	BAT_VIS		 = 0x05,
-	BAT_UNICAST_FRAG = 0x06,
-	BAT_TT_QUERY	 = 0x07,
-	BAT_ROAM_ADV	 = 0x08
+	BAT_IV_OGM		= 0x01,
+	BAT_ICMP		= 0x02,
+	BAT_UNICAST		= 0x03,
+	BAT_BCAST		= 0x04,
+	BAT_VIS			= 0x05,
+	BAT_UNICAST_FRAG	= 0x06,
+	BAT_TT_QUERY		= 0x07,
+	BAT_ROAM_ADV		= 0x08,
+	BAT_UNICAST_4ADDR	= 0x09
+};
+
+enum bat_subtype {
+	BAT_P_DATA		= 0x01
 };
 
 /* this file is included by batctl which needs these defines */
@@ -159,6 +164,12 @@ struct unicast_packet {
 	uint8_t  dest[ETH_ALEN];
 } __packed;
 
+struct unicast_4addr_packet {
+	struct unicast_packet u;
+	uint8_t src[ETH_ALEN];
+	uint8_t subtype;
+} __packed;
+
 struct unicast_frag_packet {
 	struct batman_header header;
 	uint8_t  ttvn; /* destination translation table version number */
diff --git a/net/batman-adv/routing.c b/net/batman-adv/routing.c
index ff56086..5f19bbb 100644
--- a/net/batman-adv/routing.c
+++ b/net/batman-adv/routing.c
@@ -959,14 +959,18 @@ int recv_unicast_packet(struct sk_buff *skb, struct hard_iface *recv_if)
 	struct unicast_packet *unicast_packet;
 	int hdr_size = sizeof(*unicast_packet);
 
+	unicast_packet = (struct unicast_packet *)skb->data;
+
+	/* the caller function should have already pulled 2 bytes */
+	if (unicast_packet->header.packet_type == BAT_UNICAST_4ADDR)
+		hdr_size = sizeof(struct unicast_4addr_packet);
+
 	if (check_unicast_packet(skb, hdr_size) < 0)
 		return NET_RX_DROP;
 
 	if (!check_unicast_ttvn(bat_priv, skb))
 		return NET_RX_DROP;
 
-	unicast_packet = (struct unicast_packet *)skb->data;
-
 	/* packet for me */
 	if (is_my_mac(unicast_packet->dest)) {
 		interface_rx(recv_if->soft_iface, skb, recv_if, hdr_size);
diff --git a/net/batman-adv/unicast.c b/net/batman-adv/unicast.c
index 676f6a6..96c7cec 100644
--- a/net/batman-adv/unicast.c
+++ b/net/batman-adv/unicast.c
@@ -283,13 +283,78 @@ out:
 	return ret;
 }
 
-int unicast_send_skb(struct sk_buff *skb, struct bat_priv *bat_priv)
+static bool pull_and_fill_unicast(struct sk_buff *skb, int hdr_size,
+				  struct orig_node *orig_node)
+{
+	struct unicast_packet *unicast_packet;
+
+	if (my_skb_head_push(skb, hdr_size) < 0)
+		return false;
+
+	unicast_packet = (struct unicast_packet *)skb->data;
+	unicast_packet->header.version = COMPAT_VERSION;
+	/* batman packet type: unicast */
+	unicast_packet->header.packet_type = BAT_UNICAST;
+	/* set unicast ttl */
+	unicast_packet->header.ttl = TTL;
+	/* copy the destination for faster routing */
+	memcpy(unicast_packet->dest, orig_node->orig, ETH_ALEN);
+	/* set the destination tt version number */
+	unicast_packet->ttvn =
+		(uint8_t)atomic_read(&orig_node->last_ttvn);
+
+	return true;
+}
+
+static bool prepare_unicast_packet(struct sk_buff *skb,
+				   struct orig_node *orig_node)
+{
+	return pull_and_fill_unicast(skb, sizeof(struct unicast_packet),
+				     orig_node);
+}
+
+static bool prepare_unicast_4addr_packet(struct bat_priv *bat_priv,
+					 struct sk_buff *skb,
+					 struct orig_node *orig_node,
+					 int packet_subtype)
+{
+	struct hard_iface *primary_if;
+	struct unicast_4addr_packet *unicast_4addr_packet;
+	bool ret = false;
+
+	primary_if = primary_if_get_selected(bat_priv);
+	if (!primary_if)
+		goto out;
+
+	/* pull the header space and fill the unicast_packet substructure.
+	 * We can do that because the first member of the unicast_4addr_packet
+	 * is of type struct unicast_packet
+	 */
+	if (!pull_and_fill_unicast(skb, sizeof(*unicast_4addr_packet),
+				   orig_node))
+		goto out;
+
+	unicast_4addr_packet = (struct unicast_4addr_packet *)skb->data;
+	unicast_4addr_packet->u.header.packet_type = BAT_UNICAST_4ADDR;
+	memcpy(unicast_4addr_packet->src, primary_if->net_dev->dev_addr,
+	       ETH_ALEN);
+	unicast_4addr_packet->subtype = packet_subtype;
+
+	ret = true;
+out:
+	if (primary_if)
+		hardif_free_ref(primary_if);
+	return ret;
+}
+
+int unicast_generic_send_skb(struct sk_buff *skb, struct bat_priv *bat_priv,
+			     int packet_type, int packet_subtype)
 {
 	struct ethhdr *ethhdr = (struct ethhdr *)skb->data;
-	struct unicast_packet *unicast_packet;
 	struct orig_node *orig_node;
 	struct neigh_node *neigh_node;
 	int data_len = skb->len;
+	struct unicast_packet *unicast_packet;
 	int ret = 1;
 
 	/* get routing information */
@@ -303,7 +368,6 @@ int unicast_send_skb(struct sk_buff *skb, struct bat_priv *bat_priv)
 	 * returns NULL in case of AP isolation */
 	orig_node = transtable_search(bat_priv, ethhdr->h_source,
 				      ethhdr->h_dest);
-
 find_router:
 	/**
 	 * find_router():
@@ -311,27 +375,28 @@ find_router:
 	 *  - increases neigh_nodes refcount if found.
 	 */
 	neigh_node = find_router(bat_priv, orig_node, NULL);
-
 	if (!neigh_node)
 		goto out;
 
-	if (my_skb_head_push(skb, sizeof(*unicast_packet)) < 0)
+	switch (packet_type) {
+	case BAT_UNICAST:
+		prepare_unicast_packet(skb, orig_node);
+		break;
+	case BAT_UNICAST_4ADDR:
+		prepare_unicast_4addr_packet(bat_priv, skb, orig_node,
+					     packet_subtype);
+		break;
+	default:
+		/* this function supports UNICAST and UNICAST_4ADDR only. It
+		 * should never be invoked with any other packet type
+		 */
 		goto out;
+	}
 
 	unicast_packet = (struct unicast_packet *)skb->data;
-
-	unicast_packet->header.version = COMPAT_VERSION;
-	/* batman packet type: unicast */
-	unicast_packet->header.packet_type = BAT_UNICAST;
-	/* set unicast ttl */
-	unicast_packet->header.ttl = TTL;
-	/* copy the destination for faster routing */
-	memcpy(unicast_packet->dest, orig_node->orig, ETH_ALEN);
-	/* set the destination tt version number */
-	unicast_packet->ttvn =
-		(uint8_t)atomic_read(&orig_node->last_ttvn);
-
-	if (atomic_read(&bat_priv->fragmentation) &&
+	/* fragmentation mechanism only works for UNICAST (now) */
+	if (packet_type == BAT_UNICAST &&
+	    atomic_read(&bat_priv->fragmentation) &&
 	    data_len + sizeof(*unicast_packet) >
 				neigh_node->if_incoming->net_dev->mtu) {
 		/* send frag skb decreases ttl */
@@ -343,7 +408,6 @@ find_router:
 
 	send_skb_packet(skb, neigh_node->if_incoming, neigh_node->addr);
 	ret = 0;
-	goto out;
 
 out:
 	if (neigh_node)
diff --git a/net/batman-adv/unicast.h b/net/batman-adv/unicast.h
index a9faf6b..ae9775b 100644
--- a/net/batman-adv/unicast.h
+++ b/net/batman-adv/unicast.h
@@ -30,9 +30,24 @@
 int frag_reassemble_skb(struct sk_buff *skb, struct bat_priv *bat_priv,
 			struct sk_buff **new_skb);
 void frag_list_free(struct list_head *head);
-int unicast_send_skb(struct sk_buff *skb, struct bat_priv *bat_priv);
 int frag_send_skb(struct sk_buff *skb, struct bat_priv *bat_priv,
 		  struct hard_iface *hard_iface, const uint8_t dstaddr[]);
+int unicast_generic_send_skb(struct sk_buff *skb, struct bat_priv *bat_priv,
+			     int packet_type, int packet_subtype);
+
+static inline int unicast_send_skb(struct sk_buff *skb,
+				   struct bat_priv *bat_priv)
+{
+	return unicast_generic_send_skb(skb, bat_priv, BAT_UNICAST, 0);
+}
+
+static inline int unicast_4addr_send_skb(struct sk_buff *skb,
+					 struct bat_priv *bat_priv,
+					 int packet_subtype)
+{
+	return unicast_generic_send_skb(skb, bat_priv, BAT_UNICAST_4ADDR,
+					packet_subtype);
+}
 
 static inline int frag_can_reassemble(const struct sk_buff *skb, int mtu)
 {
-- 
1.7.9.4

[PATCH 02/15] batman-adv: add a new log level for DAT debugging

[Antonio Quartulli]

A new log level has been added to concentrate messages regarding DAT: ARP
snooping, requests, response and DHT related messages.
The new log level is named DBG_DAT

Signed-off-by: Antonio Quartulli <ordex@autistici.org>
---
 Documentation/networking/batman-adv.txt |    3 ++-
 net/batman-adv/bat_sysfs.c              |    2 +-
 net/batman-adv/main.h                   |    3 ++-
 3 files changed, 5 insertions(+), 3 deletions(-)

diff --git a/Documentation/networking/batman-adv.txt b/Documentation/networking/batman-adv.txt
index 220a58c..f804635 100644
--- a/Documentation/networking/batman-adv.txt
+++ b/Documentation/networking/batman-adv.txt
@@ -203,7 +203,8 @@ abled  during run time. Following log_levels are defined:
 2 - Enable messages related to route added / changed / deleted
 4 - Enable messages related to translation table operations
 8 - Enable messages related to bridge loop avoidance
-15 - enable all messages
+16 - Enable messaged related to DAT, ARP snooping and parsing
+31 - Enable all messages
 
 The debug output can be changed at runtime  using  the  file
 /sys/class/net/bat0/mesh/log_level. e.g.
diff --git a/net/batman-adv/bat_sysfs.c b/net/batman-adv/bat_sysfs.c
index 2c81688..271c070 100644
--- a/net/batman-adv/bat_sysfs.c
+++ b/net/batman-adv/bat_sysfs.c
@@ -401,7 +401,7 @@ BAT_ATTR_UINT(gw_sel_class, S_IRUGO | S_IWUSR, 1, TQ_MAX_VALUE,
 static BAT_ATTR(gw_bandwidth, S_IRUGO | S_IWUSR, show_gw_bwidth,
 		store_gw_bwidth);
 #ifdef CONFIG_BATMAN_ADV_DEBUG
-BAT_ATTR_UINT(log_level, S_IRUGO | S_IWUSR, 0, 15, NULL);
+BAT_ATTR_UINT(log_level, S_IRUGO | S_IWUSR, 0, 31, NULL);
 #endif
 
 static struct bat_attribute *mesh_attrs[] = {
diff --git a/net/batman-adv/main.h b/net/batman-adv/main.h
index d9832ac..a2b18d0 100644
--- a/net/batman-adv/main.h
+++ b/net/batman-adv/main.h
@@ -124,7 +124,8 @@ enum dbg_level {
 	DBG_ROUTES = 1 << 1, /* route added / changed / deleted */
 	DBG_TT	   = 1 << 2, /* translation table operations */
 	DBG_BLA    = 1 << 3, /* bridge loop avoidance */
-	DBG_ALL    = 15
+	DBG_DAT    = 1 << 4, /* snooped arp messages / dat operations */
+	DBG_ALL    = 31
 };
 
 /* Kernel headers */
-- 
1.7.9.4

[B.A.T.M.A.N.] [PATCH 02/15] batman-adv: add a new log level for DAT debugging

[Antonio Quartulli]

A new log level has been added to concentrate messages regarding DAT: ARP
snooping, requests, response and DHT related messages.
The new log level is named DBG_DAT

Signed-off-by: Antonio Quartulli <ordex@autistici.org>
---
 Documentation/networking/batman-adv.txt |    3 ++-
 net/batman-adv/bat_sysfs.c              |    2 +-
 net/batman-adv/main.h                   |    3 ++-
 3 files changed, 5 insertions(+), 3 deletions(-)

diff --git a/Documentation/networking/batman-adv.txt b/Documentation/networking/batman-adv.txt
index 220a58c..f804635 100644
--- a/Documentation/networking/batman-adv.txt
+++ b/Documentation/networking/batman-adv.txt
@@ -203,7 +203,8 @@ abled  during run time. Following log_levels are defined:
 2 - Enable messages related to route added / changed / deleted
 4 - Enable messages related to translation table operations
 8 - Enable messages related to bridge loop avoidance
-15 - enable all messages
+16 - Enable messaged related to DAT, ARP snooping and parsing
+31 - Enable all messages
 
 The debug output can be changed at runtime  using  the  file
 /sys/class/net/bat0/mesh/log_level. e.g.
diff --git a/net/batman-adv/bat_sysfs.c b/net/batman-adv/bat_sysfs.c
index 2c81688..271c070 100644
--- a/net/batman-adv/bat_sysfs.c
+++ b/net/batman-adv/bat_sysfs.c
@@ -401,7 +401,7 @@ BAT_ATTR_UINT(gw_sel_class, S_IRUGO | S_IWUSR, 1, TQ_MAX_VALUE,
 static BAT_ATTR(gw_bandwidth, S_IRUGO | S_IWUSR, show_gw_bwidth,
 		store_gw_bwidth);
 #ifdef CONFIG_BATMAN_ADV_DEBUG
-BAT_ATTR_UINT(log_level, S_IRUGO | S_IWUSR, 0, 15, NULL);
+BAT_ATTR_UINT(log_level, S_IRUGO | S_IWUSR, 0, 31, NULL);
 #endif
 
 static struct bat_attribute *mesh_attrs[] = {
diff --git a/net/batman-adv/main.h b/net/batman-adv/main.h
index d9832ac..a2b18d0 100644
--- a/net/batman-adv/main.h
+++ b/net/batman-adv/main.h
@@ -124,7 +124,8 @@ enum dbg_level {
 	DBG_ROUTES = 1 << 1, /* route added / changed / deleted */
 	DBG_TT	   = 1 << 2, /* translation table operations */
 	DBG_BLA    = 1 << 3, /* bridge loop avoidance */
-	DBG_ALL    = 15
+	DBG_DAT    = 1 << 4, /* snooped arp messages / dat operations */
+	DBG_ALL    = 31
 };
 
 /* Kernel headers */
-- 
1.7.9.4

[PATCH 03/15] batman-adv: add biggest_unsigned_int(x) macro

[Antonio Quartulli]

in case of dynamic type variable, it could be needed to compute at compile time
its maximal value. This macro helps in doing that for unsigned integer types

Signed-off-by: Antonio Quartulli <ordex@autistici.org>
---
 net/batman-adv/main.h |    3 +++
 1 file changed, 3 insertions(+)

diff --git a/net/batman-adv/main.h b/net/batman-adv/main.h
index a2b18d0..d9ef4ca 100644
--- a/net/batman-adv/main.h
+++ b/net/batman-adv/main.h
@@ -221,6 +221,9 @@ static inline bool has_timed_out(unsigned long timestamp, unsigned int timeout)
 /* Returns the smallest signed integer in two's complement with the sizeof x */
 #define smallest_signed_int(x) (1u << (7u + 8u * (sizeof(x) - 1u)))
 
+/* Returns the biggest unsigned integer with the sizeof x */
+#define biggest_unsigned_int(x) (~(x)0)
+
 /* Checks if a sequence number x is a predecessor/successor of y.
  * they handle overflows/underflows and can correctly check for a
  * predecessor/successor unless the variable sequence number has grown by
-- 
1.7.9.4

[B.A.T.M.A.N.] [PATCH 03/15] batman-adv: add biggest_unsigned_int(x) macro

[Antonio Quartulli]

in case of dynamic type variable, it could be needed to compute at compile time
its maximal value. This macro helps in doing that for unsigned integer types

Signed-off-by: Antonio Quartulli <ordex@autistici.org>
---
 net/batman-adv/main.h |    3 +++
 1 file changed, 3 insertions(+)

diff --git a/net/batman-adv/main.h b/net/batman-adv/main.h
index a2b18d0..d9ef4ca 100644
--- a/net/batman-adv/main.h
+++ b/net/batman-adv/main.h
@@ -221,6 +221,9 @@ static inline bool has_timed_out(unsigned long timestamp, unsigned int timeout)
 /* Returns the smallest signed integer in two's complement with the sizeof x */
 #define smallest_signed_int(x) (1u << (7u + 8u * (sizeof(x) - 1u)))
 
+/* Returns the biggest unsigned integer with the sizeof x */
+#define biggest_unsigned_int(x) (~(x)0)
+
 /* Checks if a sequence number x is a predecessor/successor of y.
  * they handle overflows/underflows and can correctly check for a
  * predecessor/successor unless the variable sequence number has grown by
-- 
1.7.9.4

RE: [PATCH 03/15] batman-adv: add biggest_unsigned_int(x) macro

[David Laight]

 

> +/* Returns the biggest unsigned integer with the sizeof x */
> +#define biggest_unsigned_int(x) (~(x)0)

Nope - biggest_unsigned_int(unsigned char) is probably 0xffffffff

	David

Re: [B.A.T.M.A.N.] [PATCH 03/15] batman-adv: add biggest_unsigned_int(x) macro

[David Laight]

 

> +/* Returns the biggest unsigned integer with the sizeof x */
> +#define biggest_unsigned_int(x) (~(x)0)

Nope - biggest_unsigned_int(unsigned char) is probably 0xffffffff

	David

Re: [PATCH 03/15] batman-adv: add biggest_unsigned_int(x) macro

[David Laight]

 
> > +/* Returns the biggest unsigned integer with the sizeof x */
> > +#define biggest_unsigned_int(x) (~(x)0)
> 
> Nope - biggest_unsigned_int(unsigned char) is probably 0xffffffff

To correct myself, it is actually ((int)(-1)) (ie a signed value).

	David

Re: [B.A.T.M.A.N.] [PATCH 03/15] batman-adv: add biggest_unsigned_int(x) macro

[David Laight]

 
> > +/* Returns the biggest unsigned integer with the sizeof x */
> > +#define biggest_unsigned_int(x) (~(x)0)
> 
> Nope - biggest_unsigned_int(unsigned char) is probably 0xffffffff

To correct myself, it is actually ((int)(-1)) (ie a signed value).

	David

Re: [PATCH 03/15] batman-adv: add biggest_unsigned_int(x) macro

[David Miller]

From: "David Laight" <David.Laight-ZS65k/vG3HxXrIkS9f7CXA@public.gmane.org>
Date: Wed, 25 Apr 2012 15:57:39 +0100

>  
>> > +/* Returns the biggest unsigned integer with the sizeof x */
>> > +#define biggest_unsigned_int(x) (~(x)0)
>> 
>> Nope - biggest_unsigned_int(unsigned char) is probably 0xffffffff
> 
> To correct myself, it is actually ((int)(-1)) (ie a signed value).

In any event this is just a terrible macro, and too stupid to live.

Re: [B.A.T.M.A.N.] [PATCH 03/15] batman-adv: add biggest_unsigned_int(x) macro

[David Miller]

From: "David Laight" <David.Laight@ACULAB.COM>
Date: Wed, 25 Apr 2012 15:57:39 +0100

>  
>> > +/* Returns the biggest unsigned integer with the sizeof x */
>> > +#define biggest_unsigned_int(x) (~(x)0)
>> 
>> Nope - biggest_unsigned_int(unsigned char) is probably 0xffffffff
> 
> To correct myself, it is actually ((int)(-1)) (ie a signed value).

In any event this is just a terrible macro, and too stupid to live.

[B.A.T.M.A.N.] Traffic Control in batman-adv

[gtolon]

Hi,

We are doing some tests with Traffic Control (tc) in routers with 
Openwrt running batman-adv, want to be able to share the bandwidth with 
some degree of fairness between users, so we think that SFQ could help, 
differentiating the flows depending on ip addresses and ports.

The problem is that when adding the SFQ qdisc to the wireless interface 
managed by batman it doesn't work as expected. We tested the SFQ in 
other configurations and it worked fine, in full queues, the dropped 
packets affected the big flows, allowing the others to pass, but in the 
interface managed by batman-adv when packets are dropped, the result 
affects all flows almost equally. We thought this could be related with 
the bridge we use to connect the batman with the non-batman interfaces, 
but in a wireless Acces Point bridged to bat0 it also worked fine. Maybe 
it's related with the way bat0 works with it's managed interface? We'd 
appreciate any advices. Thanks in advance!

Gabriel

Re: [B.A.T.M.A.N.] Traffic Control in batman-adv

[Sven Eckelmann]

[-- Attachment #1: Type: text/plain, Size: 1554 bytes --]

On Wednesday 25 April 2012 17:27:03 gtolon@inti.gob.ar wrote:
> Hi,

First thing: Don't reply to random messages when you actually want to start a 
new topic.

> We are doing some tests with Traffic Control (tc) in routers with
> Openwrt running batman-adv, want to be able to share the bandwidth with
> some degree of fairness between users, so we think that SFQ could help,
> differentiating the flows depending on ip addresses and ports.
> 
> The problem is that when adding the SFQ qdisc to the wireless interface
> managed by batman it doesn't work as expected. We tested the SFQ in
> other configurations and it worked fine, in full queues, the dropped
> packets affected the big flows, allowing the others to pass, but in the
> interface managed by batman-adv when packets are dropped, the result
> affects all flows almost equally. We thought this could be related with
> the bridge we use to connect the batman with the non-batman interfaces,
> but in a wireless Acces Point bridged to bat0 it also worked fine. Maybe
> it's related with the way bat0 works with it's managed interface? We'd
> appreciate any advices. Thanks in advance!

Why do you add the sfq qdisc to the wireless device instead to the bat0 
device? It doesn't really makes sense to me.

Also your statement about "batman-adv bridged with non-batman-adv" doesn't 
work, but "batman-adv bridged with non-batman-adv interface" works statement 
is slightly irritating. Please explain it a little bit more verbose to help us 
understand what is the difference.

Kind regards,
	Sven

[-- Attachment #2: This is a digitally signed message part. --]
[-- Type: application/pgp-signature, Size: 836 bytes --]

Re: [B.A.T.M.A.N.] Traffic Control in batman-adv

[gtolon]

Hi Sven, thanks for replying.

El 25/04/2012 05:51 p.m., Sven Eckelmann escribió:
> On Wednesday 25 April 2012 17:27:03 gtolon@inti.gob.ar wrote:
>> Hi,
> First thing: Don't reply to random messages when you actually want to start a
> new topic.

Ok.

>> We are doing some tests with Traffic Control (tc) in routers with
>> Openwrt running batman-adv, want to be able to share the bandwidth with
>> some degree of fairness between users, so we think that SFQ could help,
>> differentiating the flows depending on ip addresses and ports.
>>
>> The problem is that when adding the SFQ qdisc to the wireless interface
>> managed by batman it doesn't work as expected. We tested the SFQ in
>> other configurations and it worked fine, in full queues, the dropped
>> packets affected the big flows, allowing the others to pass, but in the
>> interface managed by batman-adv when packets are dropped, the result
>> affects all flows almost equally. We thought this could be related with
>> the bridge we use to connect the batman with the non-batman interfaces,
>> but in a wireless Acces Point bridged to bat0 it also worked fine. Maybe
>> it's related with the way bat0 works with it's managed interface? We'd
>> appreciate any advices. Thanks in advance!
> Why do you add the sfq qdisc to the wireless device instead to the bat0
> device? It doesn't really makes sense to me.
I also tryed with bat0, but the packets continued to dropp on the 
wireless device, not in bat0, so i thought that the qdisc of bat0 was 
not being used. Anyway, i'll try again.


> Also your statement about "batman-adv bridged with non-batman-adv" doesn't
> work, but "batman-adv bridged with non-batman-adv interface" works statement
> is slightly irritating. Please explain it a little bit more verbose to help us
> understand what is the difference.
Sorry for the bad explanation,what i wanted to say is:

We have a bridge with bat0, eth0 and wlan0 attached. wlan0 is in ap 
mode. There's another wireless interface, wlan0-1 in ad-hoc mode, 
managed by bat0. When SFQ qdisc is applied to wlan0, it works as 
expected. When it's aplied to wlan0-1, it doesn't. In both cases there's 
a bridge involved, that´s what i was trying to point. I hope now it's 
more clear.

Best regards

Gabriel

Re: [B.A.T.M.A.N.] Traffic Control in batman-adv

[3zl Trizonelabs]

maybe pic this will help

http://www.bild.me/bild.php?file=9951400Batman-CPE-config.png

regards
3zl

2012/4/26  <gtolon@inti.gob.ar>:
> Hi Sven, thanks for replying.
>
> El 25/04/2012 05:51 p.m., Sven Eckelmann escribió:
>>
>> On Wednesday 25 April 2012 17:27:03 gtolon@inti.gob.ar wrote:
>>>
>>> Hi,
>>
>> First thing: Don't reply to random messages when you actually want to
>> start a
>> new topic.
>
>
> Ok.
>
>>> We are doing some tests with Traffic Control (tc) in routers with
>>> Openwrt running batman-adv, want to be able to share the bandwidth with
>>> some degree of fairness between users, so we think that SFQ could help,
>>> differentiating the flows depending on ip addresses and ports.
>>>
>>> The problem is that when adding the SFQ qdisc to the wireless interface
>>> managed by batman it doesn't work as expected. We tested the SFQ in
>>> other configurations and it worked fine, in full queues, the dropped
>>> packets affected the big flows, allowing the others to pass, but in the
>>> interface managed by batman-adv when packets are dropped, the result
>>> affects all flows almost equally. We thought this could be related with
>>> the bridge we use to connect the batman with the non-batman interfaces,
>>> but in a wireless Acces Point bridged to bat0 it also worked fine. Maybe
>>> it's related with the way bat0 works with it's managed interface? We'd
>>> appreciate any advices. Thanks in advance!
>>
>> Why do you add the sfq qdisc to the wireless device instead to the bat0
>> device? It doesn't really makes sense to me.
>
> I also tryed with bat0, but the packets continued to dropp on the wireless
> device, not in bat0, so i thought that the qdisc of bat0 was not being used.
> Anyway, i'll try again.
>
>
>> Also your statement about "batman-adv bridged with non-batman-adv" doesn't
>> work, but "batman-adv bridged with non-batman-adv interface" works
>> statement
>> is slightly irritating. Please explain it a little bit more verbose to
>> help us
>> understand what is the difference.
>
> Sorry for the bad explanation,what i wanted to say is:
>
> We have a bridge with bat0, eth0 and wlan0 attached. wlan0 is in ap mode.
> There's another wireless interface, wlan0-1 in ad-hoc mode, managed by bat0.
> When SFQ qdisc is applied to wlan0, it works as expected. When it's aplied
> to wlan0-1, it doesn't. In both cases there's a bridge involved, that´s what
> i was trying to point. I hope now it's more clear.
>
> Best regards
>
> Gabriel

Re: [B.A.T.M.A.N.] Traffic Control in batman-adv

[Marek Lindner]

On Thursday, April 26, 2012 22:49:24 3zl Trizonelabs wrote:
> maybe pic this will help
> 
> http://www.bild.me/bild.php?file=9951400Batman-CPE-config.png


Hey,

that is a very good graphical explanation of the setup! Do you happen to have 
more of those ? Maybe we should create a page talking about possible network 
setups and include your stuff there ?

Cheers,
Marek

Re: [B.A.T.M.A.N.] Traffic Control in batman-adv

[3zl Trizonelabs]

Hi Marek,
since WBM Athens, where i met you guys,i decided to prepare some stuff
and organize it on a collaborating site for batman-adv applications
Everybody is  welcome to join.
I'll add your email to the Mindmono Mindmap site to get some ideas how
to organize the site. ( teamlab.com would be nice)
Have a look at the mindmap
:http://www.mindomo.com/view.htm?m=bbde50c10b0b4864bde98b83f2a7cf3b
regards
3zl
Greece/Germany

2012/4/26 Marek Lindner <lindner_marek@yahoo.de>:
> On Thursday, April 26, 2012 22:49:24 3zl Trizonelabs wrote:
>> maybe pic this will help
>>
>> http://www.bild.me/bild.php?file=9951400Batman-CPE-config.png
>
>
> Hey,
>
> that is a very good graphical explanation of the setup! Do you happen to have
> more of those ? Maybe we should create a page talking about possible network
> setups and include your stuff there ?
>
> Cheers,
> Marek

Re: [B.A.T.M.A.N.] Traffic Control in batman-adv

[gtolon]

Hi,
That's a very good diagram!

Resuming the thread, Sven said that adding a qdisc to wlan0-1 makes no 
sense. I don't understand completely why, but i guess that once bat0 
manages wlan0-1, the latter can't work anymore at IP layer, right?

So i've done some tests with bat0. I realized that maybe when i had 
applied a SFQ qdisc to bat0, the bottleneck was still on wlan0-1, and 
for that reasons packets were dropped on wlan0-1. Then using a HTB qdisc 
i limited bat0 rate to a minor value than the wlan0-1 actual output 
rate, and added some classes inside with SFQ. This way i could see 
dropped packets at bat0 and SFQ working as expected (and no dropped 
packets at wlan0-1). However, for this configuration to work, the limit 
value rate for bat0 has to be smaller than the output rate of wlan0-1, 
but this value is not fixed in a wireless link. I don't know if it's 
possible to use another configuration in which it's not necessary to set 
a fixed limit value to allow bat0 to control the bottleneck.

Regards

Gabriel


El 26/04/2012 01:04 p.m., 3zl Trizonelabs escribió:
> Hi Marek,
> since WBM Athens, where i met you guys,i decided to prepare some stuff
> and organize it on a collaborating site for batman-adv applications
> Everybody is  welcome to join.
> I'll add your email to the Mindmono Mindmap site to get some ideas how
> to organize the site. ( teamlab.com would be nice)
> Have a look at the mindmap
> :http://www.mindomo.com/view.htm?m=bbde50c10b0b4864bde98b83f2a7cf3b
> regards
> 3zl
> Greece/Germany
>
> 2012/4/26 Marek Lindner<lindner_marek@yahoo.de>:
>> On Thursday, April 26, 2012 22:49:24 3zl Trizonelabs wrote:
>>> maybe pic this will help
>>>
>>> http://www.bild.me/bild.php?file=9951400Batman-CPE-config.png
>>
>> Hey,
>>
>> that is a very good graphical explanation of the setup! Do you happen to have
>> more of those ? Maybe we should create a page talking about possible network
>> setups and include your stuff there ?
>>
>> Cheers,
>> Marek

Re: [B.A.T.M.A.N.] Traffic Control in batman-adv

[Guido Iribarren]

Hi gabriel,
maybe rate limiting is not what you're looking for, but instead TOS
marks on packets, which allow you to reorder packets in the queue, so
as to give priority to some protocols, without limiting overall rate
(or maybe i got it all wrong about what TOS is intented for :) )

on the other hand, you might check out a feature of gargoyle firmware
as inspiration. it's based on openwrt backfire, and implements an
'active congestion check' mechanism, that measures ping times along
the link, to estimate the actual troughput and adjust the qdisc max
rate accordingly, so that QoS rules continue working properly.

Cheers!
Guido

On 4/27/12, gtolon@inti.gob.ar <gtolon@inti.gob.ar> wrote:
> Hi,
> That's a very good diagram!
>
> Resuming the thread, Sven said that adding a qdisc to wlan0-1 makes no
> sense. I don't understand completely why, but i guess that once bat0
> manages wlan0-1, the latter can't work anymore at IP layer, right?
>
> So i've done some tests with bat0. I realized that maybe when i had
> applied a SFQ qdisc to bat0, the bottleneck was still on wlan0-1, and
> for that reasons packets were dropped on wlan0-1. Then using a HTB qdisc
> i limited bat0 rate to a minor value than the wlan0-1 actual output
> rate, and added some classes inside with SFQ. This way i could see
> dropped packets at bat0 and SFQ working as expected (and no dropped
> packets at wlan0-1). However, for this configuration to work, the limit
> value rate for bat0 has to be smaller than the output rate of wlan0-1,
> but this value is not fixed in a wireless link. I don't know if it's
> possible to use another configuration in which it's not necessary to set
> a fixed limit value to allow bat0 to control the bottleneck.
>
> Regards
>
> Gabriel
>
>
> El 26/04/2012 01:04 p.m., 3zl Trizonelabs escribió:
>> Hi Marek,
>> since WBM Athens, where i met you guys,i decided to prepare some stuff
>> and organize it on a collaborating site for batman-adv applications
>> Everybody is  welcome to join.
>> I'll add your email to the Mindmono Mindmap site to get some ideas how
>> to organize the site. ( teamlab.com would be nice)
>> Have a look at the mindmap
>> :http://www.mindomo.com/view.htm?m=bbde50c10b0b4864bde98b83f2a7cf3b
>> regards
>> 3zl
>> Greece/Germany
>>
>> 2012/4/26 Marek Lindner<lindner_marek@yahoo.de>:
>>> On Thursday, April 26, 2012 22:49:24 3zl Trizonelabs wrote:
>>>> maybe pic this will help
>>>>
>>>> http://www.bild.me/bild.php?file=9951400Batman-CPE-config.png
>>>
>>> Hey,
>>>
>>> that is a very good graphical explanation of the setup! Do you happen to
>>> have
>>> more of those ? Maybe we should create a page talking about possible
>>> network
>>> setups and include your stuff there ?
>>>
>>> Cheers,
>>> Marek
>

Re: [B.A.T.M.A.N.] Traffic Control in batman-adv

[3zl Trizonelabs]

measuring bandwidth we test with "assolo" packet in afrimesh repos
(builds ok with trunk)

maybe also have a look at


http://wiki.openwrt.org/doc/howto/packet.scheduler/packet.scheduler.theory

  TBF with netem discussed here
                    https://forum.openwrt.org/viewtopic.php?pid=161108

regards
3zl

Re: [B.A.T.M.A.N.] Traffic Control in batman-adv

[gtolon]

Guido and 3zl, thanks for the replies!

Regarding packet marking, i think it's a good idea, but when i tryed it 
i had some issues with the interaction between iptables/ebtables and the 
bridge, i'll read more about that.

I've heard about gargoyle, didn't know it was based on backfire. That 
active congestion check sounds great!

In this link there's an abstract about that assolo method, interesting:

http://www.computer.org/portal/web/csdl/doi/10.1109/ICIMP.2009.28


Best regards

Gabriel




El 27/04/2012 07:34 p.m., 3zl Trizonelabs escribió:
> measuring bandwidth we test with "assolo" packet in afrimesh repos
> (builds ok with trunk)
>
> maybe also have a look at
>
>
> http://wiki.openwrt.org/doc/howto/packet.scheduler/packet.scheduler.theory
>
>    TBF with netem discussed here
>                      https://forum.openwrt.org/viewtopic.php?pid=161108
>
> regards
> 3zl
>

Re: [PATCH 03/15] batman-adv: add biggest_unsigned_int(x) macro

[Antonio Quartulli]

Il 25.04.2012 20:09 David Miller ha scritto:
> From: "David Laight" <David.Laight-ZS65k/vG3HxXrIkS9f7CXA@public.gmane.org>
> Date: Wed, 25 Apr 2012 15:57:39 +0100
>
>>
>>> > +/* Returns the biggest unsigned integer with the sizeof x */
>>> > +#define biggest_unsigned_int(x) (~(x)0)
>>>
>>> Nope - biggest_unsigned_int(unsigned char) is probably 0xffffffff
>>
>> To correct myself, it is actually ((int)(-1)) (ie a signed value).
>
> In any event this is just a terrible macro, and too stupid to live.

Hello David,

ok. So the suggestion is to directly use the ~ operator in the code 
without creating a macro.
If the rest of the patchset is ok, I will eliminate this macro and 
resend the pull request again.

Cheers,


-- 
Antonio Quartulli

Re: [B.A.T.M.A.N.] [PATCH 03/15] batman-adv: add biggest_unsigned_int(x) macro

[Antonio Quartulli]

Il 25.04.2012 20:09 David Miller ha scritto:
> From: "David Laight" <David.Laight@ACULAB.COM>
> Date: Wed, 25 Apr 2012 15:57:39 +0100
>
>>
>>> > +/* Returns the biggest unsigned integer with the sizeof x */
>>> > +#define biggest_unsigned_int(x) (~(x)0)
>>>
>>> Nope - biggest_unsigned_int(unsigned char) is probably 0xffffffff
>>
>> To correct myself, it is actually ((int)(-1)) (ie a signed value).
>
> In any event this is just a terrible macro, and too stupid to live.

Hello David,

ok. So the suggestion is to directly use the ~ operator in the code 
without creating a macro.
If the rest of the patchset is ok, I will eliminate this macro and 
resend the pull request again.

Cheers,


-- 
Antonio Quartulli

[PATCH 04/15] batman-adv: Distributed ARP Table - create DHT helper functions

[Antonio Quartulli]

Add all the relevant functions in order to manage a Distributed Hash Table over
the B.A.T.M.A.N.-adv network. It will later be used to store several ARP entries
and implement DAT (Distributed ARP Table)

Signed-off-by: Antonio Quartulli <ordex@autistici.org>
---
 net/batman-adv/Makefile                |    1 +
 net/batman-adv/distributed-arp-table.c |  207 ++++++++++++++++++++++++++++++++
 net/batman-adv/distributed-arp-table.h |   66 ++++++++++
 net/batman-adv/hard-interface.c        |    3 +
 net/batman-adv/main.h                  |    6 +
 net/batman-adv/originator.c            |    2 +
 net/batman-adv/types.h                 |   14 +++
 net/batman-adv/unicast.c               |    8 +-
 net/batman-adv/unicast.h               |    4 +
 9 files changed, 307 insertions(+), 4 deletions(-)
 create mode 100644 net/batman-adv/distributed-arp-table.c
 create mode 100644 net/batman-adv/distributed-arp-table.h

diff --git a/net/batman-adv/Makefile b/net/batman-adv/Makefile
index 6d5c194..84955b3 100644
--- a/net/batman-adv/Makefile
+++ b/net/batman-adv/Makefile
@@ -24,6 +24,7 @@ batman-adv-y += bat_iv_ogm.o
 batman-adv-y += bat_sysfs.o
 batman-adv-y += bitarray.o
 batman-adv-$(CONFIG_BATMAN_ADV_BLA) += bridge_loop_avoidance.o
+batman-adv-y += distributed-arp-table.o
 batman-adv-y += gateway_client.o
 batman-adv-y += gateway_common.o
 batman-adv-y += hard-interface.o
diff --git a/net/batman-adv/distributed-arp-table.c b/net/batman-adv/distributed-arp-table.c
new file mode 100644
index 0000000..0078478
--- /dev/null
+++ b/net/batman-adv/distributed-arp-table.c
@@ -0,0 +1,207 @@
+/*
+ * Copyright (C) 2011 B.A.T.M.A.N. contributors:
+ *
+ * Antonio Quartulli
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of version 2 of the GNU General Public
+ * License as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA
+ * 02110-1301, USA
+ *
+ */
+
+#include <linux/if_ether.h>
+#include <linux/if_arp.h>
+
+#include "main.h"
+#include "distributed-arp-table.h"
+#include "hard-interface.h"
+#include "originator.h"
+#include "send.h"
+#include "types.h"
+#include "unicast.h"
+
+static bool is_orig_node_eligible(struct dht_candidate *res, int select,
+				  dat_addr_t tmp_max, dat_addr_t max,
+				  dat_addr_t last_max,
+				  struct orig_node *candidate,
+				  struct orig_node *max_orig_node)
+{
+	bool ret = false;
+	int j;
+
+	/* Check if we have already selected this neighbour... */
+	for (j = 0; j < select; j++)
+		if (res[j].orig_node == candidate)
+			break;
+	/* ..and possibly skip it */
+	if (j < select)
+		goto out;
+	/* sanity check: has it already been selected? This should not happen */
+	if (tmp_max > last_max)
+		goto out;
+	/* check if during this iteration we have already found an originator
+	 * with a closer dht address
+	 */
+	if (tmp_max < max)
+		goto out;
+	/* this is an hash collision with the temporary selected node. Choose
+	 * the one with the lowest address
+	 */
+	if ((tmp_max == max) &&
+	    (compare_eth(candidate->orig, max_orig_node->orig) > 0))
+		goto out;
+
+	ret = true;
+out:
+	return ret;
+}
+
+/* selects the next candidate by populating cands[select] and modifies last_max
+ * accordingly
+ */
+static void choose_next_candidate(struct bat_priv *bat_priv,
+				  struct dht_candidate *cands, int select,
+				  dat_addr_t ip_key, dat_addr_t *last_max)
+{
+	dat_addr_t max = 0, tmp_max = 0;
+	struct orig_node *orig_node, *max_orig_node = NULL;
+	struct hashtable_t *hash = bat_priv->orig_hash;
+	struct hlist_node *node;
+	struct hlist_head *head;
+	int i;
+
+	/* if no node is eligible as candidate, we will leave the candidate as
+	 * NOT_FOUND
+	 */
+	cands[select].type = DHT_CANDIDATE_NOT_FOUND;
+
+	/* iterate over the originator list and find the node with closest
+	 * dht_address which has not been selected yet
+	 */
+	for (i = 0; i < hash->size; i++) {
+		head = &hash->table[i];
+
+		rcu_read_lock();
+		hlist_for_each_entry_rcu(orig_node, node, head, hash_entry) {
+			/* the dht space is a ring and addresses are unsigned */
+			tmp_max = DAT_ADDR_MAX - orig_node->dht_addr + ip_key;
+
+			if (!is_orig_node_eligible(cands, select, tmp_max, max,
+						   *last_max, orig_node,
+						   max_orig_node))
+				continue;
+
+			if (!atomic_inc_not_zero(&orig_node->refcount))
+				continue;
+
+			max = tmp_max;
+			if (max_orig_node)
+				orig_node_free_ref(max_orig_node);
+			max_orig_node = orig_node;
+		}
+		rcu_read_unlock();
+	}
+	if (max_orig_node) {
+		cands[select].type = DHT_CANDIDATE_ORIG;
+		cands[select].orig_node = max_orig_node;
+		bat_dbg(DBG_DAT, bat_priv,
+			"dht_select_candidates() %d: selected %pM addr=%u dist=%u\n",
+			select, max_orig_node->orig, max_orig_node->dht_addr,
+			max);
+	}
+	*last_max = max;
+}
+
+/* Given a key, selects the candidates which the DHT message has to be sent to.
+ * An originator O is selected if and only if its DHT_ID value is one of three
+ * closest values (from the LEFT, with wrap around if needed) then the hash
+ * value of the key. ip_dst is the key.
+ *
+ * return an array of size DHT_CANDIDATES_NUM
+ */
+static struct dht_candidate *dht_select_candidates(struct bat_priv *bat_priv,
+						   uint32_t ip_dst)
+{
+	int select;
+	dat_addr_t last_max = DAT_ADDR_MAX, ip_key;
+	struct dht_candidate *res;
+
+	if (!bat_priv->orig_hash)
+		return NULL;
+
+	res = kmalloc(DHT_CANDIDATES_NUM * sizeof(*res), GFP_ATOMIC);
+	if (!res)
+		return NULL;
+
+	ip_key = (dat_addr_t)hash_ipv4(&ip_dst, DAT_ADDR_MAX);
+
+	bat_dbg(DBG_DAT, bat_priv,
+		"dht_select_candidates(): IP=%pI4 hash(IP)=%u\n", &ip_dst,
+		ip_key);
+
+	for (select = 0; select < DHT_CANDIDATES_NUM; select++)
+		choose_next_candidate(bat_priv, res, select, ip_key, &last_max);
+
+	return res;
+}
+
+/* Sends the skb payload passed as argument to the candidates selected for
+ * the data represented by 'ip'. The skb is copied by means of pskb_copy()
+ * and is sent as unicast packet to each of the selected candidate.
+ *
+ * If the packet is successfully sent to at least one candidate, then this
+ * function returns true
+ */
+static bool dht_send_data(struct bat_priv *bat_priv, struct sk_buff *skb,
+			  uint32_t ip, int packet_subtype)
+{
+	int i;
+	bool ret = false;
+	struct neigh_node *neigh_node = NULL;
+	struct sk_buff *tmp_skb;
+	struct dht_candidate *cand = dht_select_candidates(bat_priv, ip);
+
+	if (!cand)
+		goto out;
+
+	bat_dbg(DBG_DAT, bat_priv, "DHT_SEND for %pI4\n", &ip);
+
+	for (i = 0; i < DHT_CANDIDATES_NUM; i++) {
+		if (cand[i].type == DHT_CANDIDATE_NOT_FOUND)
+			continue;
+
+		neigh_node = orig_node_get_router(cand[i].orig_node);
+		if (!neigh_node)
+			goto free_orig;
+
+		tmp_skb = pskb_copy(skb, GFP_ATOMIC);
+		if (!prepare_unicast_4addr_packet(bat_priv, tmp_skb,
+						  cand[i].orig_node,
+						  packet_subtype)) {
+			kfree_skb(tmp_skb);
+			goto free_neigh;
+		}
+		if (send_skb_packet(tmp_skb, neigh_node->if_incoming,
+				    neigh_node->addr) == NET_XMIT_SUCCESS)
+			/* packet sent to a candidate: we can return true */
+			ret = true;
+free_neigh:
+		neigh_node_free_ref(neigh_node);
+free_orig:
+		orig_node_free_ref(cand[i].orig_node);
+	}
+
+out:
+	kfree(cand);
+	return ret;
+}
diff --git a/net/batman-adv/distributed-arp-table.h b/net/batman-adv/distributed-arp-table.h
new file mode 100644
index 0000000..94ee0fd
--- /dev/null
+++ b/net/batman-adv/distributed-arp-table.h
@@ -0,0 +1,66 @@
+/*
+ * Copyright (C) 2011 B.A.T.M.A.N. contributors:
+ *
+ * Antonio Quartulli
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of version 2 of the GNU General Public
+ * License as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA
+ * 02110-1301, USA
+ *
+ */
+
+#ifndef _NET_BATMAN_ADV_ARP_H_
+#define _NET_BATMAN_ADV_ARP_H_
+
+#include "types.h"
+#include "originator.h"
+
+#define DAT_ADDR_MAX biggest_unsigned_int(dat_addr_t)
+
+/* hash function to choose an entry in a hash table of given size.
+ * hash algorithm from http://en.wikipedia.org/wiki/Hash_table
+ */
+static inline uint32_t hash_ipv4(const void *data, uint32_t size)
+{
+	const unsigned char *key = data;
+	uint32_t hash = 0;
+	size_t i;
+
+	for (i = 0; i < 4; i++) {
+		hash += key[i];
+		hash += (hash << 10);
+		hash ^= (hash >> 6);
+	}
+
+	hash += (hash << 3);
+	hash ^= (hash >> 11);
+	hash += (hash << 15);
+
+	return hash % size;
+}
+
+static inline void dat_init_orig_node_dht_addr(struct orig_node *orig_node)
+{
+	orig_node->dht_addr = (dat_addr_t)choose_orig(orig_node->orig,
+						      DAT_ADDR_MAX);
+}
+
+static inline void dat_init_own_dht_addr(struct bat_priv *bat_priv,
+					 struct hard_iface *primary_if)
+{
+	bat_priv->dht_addr = (dat_addr_t)
+				choose_orig(primary_if->net_dev->dev_addr,
+					    DAT_ADDR_MAX);
+}
+
+#endif /* _NET_BATMAN_ADV_ARP_H_ */
diff --git a/net/batman-adv/hard-interface.c b/net/batman-adv/hard-interface.c
index 220cb02..1393939 100644
--- a/net/batman-adv/hard-interface.c
+++ b/net/batman-adv/hard-interface.c
@@ -20,6 +20,7 @@
  */
 
 #include "main.h"
+#include "distributed-arp-table.h"
 #include "hard-interface.h"
 #include "soft-interface.h"
 #include "send.h"
@@ -118,6 +119,8 @@ static void primary_if_update_addr(struct bat_priv *bat_priv,
 	if (!primary_if)
 		goto out;
 
+	dat_init_own_dht_addr(bat_priv, primary_if);
+
 	vis_packet = (struct vis_packet *)
 				bat_priv->my_vis_info->skb_packet->data;
 	memcpy(vis_packet->vis_orig, primary_if->net_dev->dev_addr, ETH_ALEN);
diff --git a/net/batman-adv/main.h b/net/batman-adv/main.h
index d9ef4ca..395e59d 100644
--- a/net/batman-adv/main.h
+++ b/net/batman-adv/main.h
@@ -67,6 +67,9 @@
 
 #define NUM_WORDS BITS_TO_LONGS(TQ_LOCAL_WINDOW_SIZE)
 
+/* numbers of originator to contact for any PUT/GET DHT operation */
+#define DHT_CANDIDATES_NUM 3
+
 #define LOG_BUF_LEN 8192	  /* has to be a power of 2 */
 
 #define VIS_INTERVAL 5000	/* 5 seconds */
@@ -111,6 +114,9 @@ enum uev_type {
 
 #define GW_THRESHOLD	50
 
+#define DHT_CANDIDATE_NOT_FOUND	0
+#define DHT_CANDIDATE_ORIG	1
+
 /* Debug Messages */
 #ifdef pr_fmt
 #undef pr_fmt
diff --git a/net/batman-adv/originator.c b/net/batman-adv/originator.c
index ce49698..31d7b58 100644
--- a/net/batman-adv/originator.c
+++ b/net/batman-adv/originator.c
@@ -20,6 +20,7 @@
  */
 
 #include "main.h"
+#include "distributed-arp-table.h"
 #include "originator.h"
 #include "hash.h"
 #include "translation-table.h"
@@ -224,6 +225,7 @@ struct orig_node *get_orig_node(struct bat_priv *bat_priv, const uint8_t *addr)
 	orig_node->tt_poss_change = false;
 	orig_node->bat_priv = bat_priv;
 	memcpy(orig_node->orig, addr, ETH_ALEN);
+	dat_init_orig_node_dht_addr(orig_node);
 	orig_node->router = NULL;
 	orig_node->tt_crc = 0;
 	atomic_set(&orig_node->last_ttvn, 0);
diff --git a/net/batman-adv/types.h b/net/batman-adv/types.h
index 2f4848b..61e2d78 100644
--- a/net/batman-adv/types.h
+++ b/net/batman-adv/types.h
@@ -27,6 +27,13 @@
 #include "packet.h"
 #include "bitarray.h"
 
+/* dat_addr_t is the type used for all DHT addresses. If it is changed,
+ * DAT_ADDR_MAX is changed as well.
+ *
+ * *Please be careful: dat_addr_t must be UNSIGNED*
+ */
+#define dat_addr_t uint16_t
+
 #define BAT_HEADER_LEN (ETH_HLEN + \
 	((sizeof(struct unicast_packet) > sizeof(struct bcast_packet) ? \
 	 sizeof(struct unicast_packet) : \
@@ -67,6 +74,7 @@ struct hard_iface {
 struct orig_node {
 	uint8_t orig[ETH_ALEN];
 	uint8_t primary_addr[ETH_ALEN];
+	dat_addr_t dht_addr;
 	struct neigh_node __rcu *router; /* rcu protected pointer */
 	unsigned long *bcast_own;
 	uint8_t *bcast_own_sum;
@@ -221,6 +229,7 @@ struct bat_priv {
 	struct gw_node __rcu *curr_gw;  /* rcu protected pointer */
 	atomic_t gw_reselect;
 	struct hard_iface __rcu *primary_if;  /* rcu protected pointer */
+	dat_addr_t dht_addr;
 	struct vis_info *my_vis_info;
 	struct bat_algo_ops *bat_algo_ops;
 };
@@ -395,4 +404,9 @@ struct bat_algo_ops {
 				struct sk_buff *skb);
 };
 
+struct dht_candidate {
+	int type;
+	struct orig_node *orig_node;
+};
+
 #endif /* _NET_BATMAN_ADV_TYPES_H_ */
diff --git a/net/batman-adv/unicast.c b/net/batman-adv/unicast.c
index 96c7cec..b7152c6 100644
--- a/net/batman-adv/unicast.c
+++ b/net/batman-adv/unicast.c
@@ -313,10 +313,10 @@ static bool prepare_unicast_packet(struct sk_buff *skb,
 				     orig_node);
 }
 
-static bool prepare_unicast_4addr_packet(struct bat_priv *bat_priv,
-					 struct sk_buff *skb,
-					 struct orig_node *orig_node,
-					 int packet_subtype)
+bool prepare_unicast_4addr_packet(struct bat_priv *bat_priv,
+				  struct sk_buff *skb,
+				  struct orig_node *orig_node,
+				  int packet_subtype)
 {
 	struct hard_iface *primary_if;
 	struct unicast_4addr_packet *unicast_4addr_packet;
diff --git a/net/batman-adv/unicast.h b/net/batman-adv/unicast.h
index ae9775b..e15aa62 100644
--- a/net/batman-adv/unicast.h
+++ b/net/batman-adv/unicast.h
@@ -32,6 +32,10 @@ int frag_reassemble_skb(struct sk_buff *skb, struct bat_priv *bat_priv,
 void frag_list_free(struct list_head *head);
 int frag_send_skb(struct sk_buff *skb, struct bat_priv *bat_priv,
 		  struct hard_iface *hard_iface, const uint8_t dstaddr[]);
+bool prepare_unicast_4addr_packet(struct bat_priv *bat_priv,
+				  struct sk_buff *skb,
+				  struct orig_node *orig_node,
+				  int packet_subtype);
 int unicast_generic_send_skb(struct sk_buff *skb, struct bat_priv *bat_priv,
 			     int packet_type, int packet_subtype);
 
-- 
1.7.9.4

[B.A.T.M.A.N.] [PATCH 04/15] batman-adv: Distributed ARP Table - create DHT helper functions

[Antonio Quartulli]

Add all the relevant functions in order to manage a Distributed Hash Table over
the B.A.T.M.A.N.-adv network. It will later be used to store several ARP entries
and implement DAT (Distributed ARP Table)

Signed-off-by: Antonio Quartulli <ordex@autistici.org>
---
 net/batman-adv/Makefile                |    1 +
 net/batman-adv/distributed-arp-table.c |  207 ++++++++++++++++++++++++++++++++
 net/batman-adv/distributed-arp-table.h |   66 ++++++++++
 net/batman-adv/hard-interface.c        |    3 +
 net/batman-adv/main.h                  |    6 +
 net/batman-adv/originator.c            |    2 +
 net/batman-adv/types.h                 |   14 +++
 net/batman-adv/unicast.c               |    8 +-
 net/batman-adv/unicast.h               |    4 +
 9 files changed, 307 insertions(+), 4 deletions(-)
 create mode 100644 net/batman-adv/distributed-arp-table.c
 create mode 100644 net/batman-adv/distributed-arp-table.h

diff --git a/net/batman-adv/Makefile b/net/batman-adv/Makefile
index 6d5c194..84955b3 100644
--- a/net/batman-adv/Makefile
+++ b/net/batman-adv/Makefile
@@ -24,6 +24,7 @@ batman-adv-y += bat_iv_ogm.o
 batman-adv-y += bat_sysfs.o
 batman-adv-y += bitarray.o
 batman-adv-$(CONFIG_BATMAN_ADV_BLA) += bridge_loop_avoidance.o
+batman-adv-y += distributed-arp-table.o
 batman-adv-y += gateway_client.o
 batman-adv-y += gateway_common.o
 batman-adv-y += hard-interface.o
diff --git a/net/batman-adv/distributed-arp-table.c b/net/batman-adv/distributed-arp-table.c
new file mode 100644
index 0000000..0078478
--- /dev/null
+++ b/net/batman-adv/distributed-arp-table.c
@@ -0,0 +1,207 @@
+/*
+ * Copyright (C) 2011 B.A.T.M.A.N. contributors:
+ *
+ * Antonio Quartulli
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of version 2 of the GNU General Public
+ * License as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA
+ * 02110-1301, USA
+ *
+ */
+
+#include <linux/if_ether.h>
+#include <linux/if_arp.h>
+
+#include "main.h"
+#include "distributed-arp-table.h"
+#include "hard-interface.h"
+#include "originator.h"
+#include "send.h"
+#include "types.h"
+#include "unicast.h"
+
+static bool is_orig_node_eligible(struct dht_candidate *res, int select,
+				  dat_addr_t tmp_max, dat_addr_t max,
+				  dat_addr_t last_max,
+				  struct orig_node *candidate,
+				  struct orig_node *max_orig_node)
+{
+	bool ret = false;
+	int j;
+
+	/* Check if we have already selected this neighbour... */
+	for (j = 0; j < select; j++)
+		if (res[j].orig_node == candidate)
+			break;
+	/* ..and possibly skip it */
+	if (j < select)
+		goto out;
+	/* sanity check: has it already been selected? This should not happen */
+	if (tmp_max > last_max)
+		goto out;
+	/* check if during this iteration we have already found an originator
+	 * with a closer dht address
+	 */
+	if (tmp_max < max)
+		goto out;
+	/* this is an hash collision with the temporary selected node. Choose
+	 * the one with the lowest address
+	 */
+	if ((tmp_max == max) &&
+	    (compare_eth(candidate->orig, max_orig_node->orig) > 0))
+		goto out;
+
+	ret = true;
+out:
+	return ret;
+}
+
+/* selects the next candidate by populating cands[select] and modifies last_max
+ * accordingly
+ */
+static void choose_next_candidate(struct bat_priv *bat_priv,
+				  struct dht_candidate *cands, int select,
+				  dat_addr_t ip_key, dat_addr_t *last_max)
+{
+	dat_addr_t max = 0, tmp_max = 0;
+	struct orig_node *orig_node, *max_orig_node = NULL;
+	struct hashtable_t *hash = bat_priv->orig_hash;
+	struct hlist_node *node;
+	struct hlist_head *head;
+	int i;
+
+	/* if no node is eligible as candidate, we will leave the candidate as
+	 * NOT_FOUND
+	 */
+	cands[select].type = DHT_CANDIDATE_NOT_FOUND;
+
+	/* iterate over the originator list and find the node with closest
+	 * dht_address which has not been selected yet
+	 */
+	for (i = 0; i < hash->size; i++) {
+		head = &hash->table[i];
+
+		rcu_read_lock();
+		hlist_for_each_entry_rcu(orig_node, node, head, hash_entry) {
+			/* the dht space is a ring and addresses are unsigned */
+			tmp_max = DAT_ADDR_MAX - orig_node->dht_addr + ip_key;
+
+			if (!is_orig_node_eligible(cands, select, tmp_max, max,
+						   *last_max, orig_node,
+						   max_orig_node))
+				continue;
+
+			if (!atomic_inc_not_zero(&orig_node->refcount))
+				continue;
+
+			max = tmp_max;
+			if (max_orig_node)
+				orig_node_free_ref(max_orig_node);
+			max_orig_node = orig_node;
+		}
+		rcu_read_unlock();
+	}
+	if (max_orig_node) {
+		cands[select].type = DHT_CANDIDATE_ORIG;
+		cands[select].orig_node = max_orig_node;
+		bat_dbg(DBG_DAT, bat_priv,
+			"dht_select_candidates() %d: selected %pM addr=%u dist=%u\n",
+			select, max_orig_node->orig, max_orig_node->dht_addr,
+			max);
+	}
+	*last_max = max;
+}
+
+/* Given a key, selects the candidates which the DHT message has to be sent to.
+ * An originator O is selected if and only if its DHT_ID value is one of three
+ * closest values (from the LEFT, with wrap around if needed) then the hash
+ * value of the key. ip_dst is the key.
+ *
+ * return an array of size DHT_CANDIDATES_NUM
+ */
+static struct dht_candidate *dht_select_candidates(struct bat_priv *bat_priv,
+						   uint32_t ip_dst)
+{
+	int select;
+	dat_addr_t last_max = DAT_ADDR_MAX, ip_key;
+	struct dht_candidate *res;
+
+	if (!bat_priv->orig_hash)
+		return NULL;
+
+	res = kmalloc(DHT_CANDIDATES_NUM * sizeof(*res), GFP_ATOMIC);
+	if (!res)
+		return NULL;
+
+	ip_key = (dat_addr_t)hash_ipv4(&ip_dst, DAT_ADDR_MAX);
+
+	bat_dbg(DBG_DAT, bat_priv,
+		"dht_select_candidates(): IP=%pI4 hash(IP)=%u\n", &ip_dst,
+		ip_key);
+
+	for (select = 0; select < DHT_CANDIDATES_NUM; select++)
+		choose_next_candidate(bat_priv, res, select, ip_key, &last_max);
+
+	return res;
+}
+
+/* Sends the skb payload passed as argument to the candidates selected for
+ * the data represented by 'ip'. The skb is copied by means of pskb_copy()
+ * and is sent as unicast packet to each of the selected candidate.
+ *
+ * If the packet is successfully sent to at least one candidate, then this
+ * function returns true
+ */
+static bool dht_send_data(struct bat_priv *bat_priv, struct sk_buff *skb,
+			  uint32_t ip, int packet_subtype)
+{
+	int i;
+	bool ret = false;
+	struct neigh_node *neigh_node = NULL;
+	struct sk_buff *tmp_skb;
+	struct dht_candidate *cand = dht_select_candidates(bat_priv, ip);
+
+	if (!cand)
+		goto out;
+
+	bat_dbg(DBG_DAT, bat_priv, "DHT_SEND for %pI4\n", &ip);
+
+	for (i = 0; i < DHT_CANDIDATES_NUM; i++) {
+		if (cand[i].type == DHT_CANDIDATE_NOT_FOUND)
+			continue;
+
+		neigh_node = orig_node_get_router(cand[i].orig_node);
+		if (!neigh_node)
+			goto free_orig;
+
+		tmp_skb = pskb_copy(skb, GFP_ATOMIC);
+		if (!prepare_unicast_4addr_packet(bat_priv, tmp_skb,
+						  cand[i].orig_node,
+						  packet_subtype)) {
+			kfree_skb(tmp_skb);
+			goto free_neigh;
+		}
+		if (send_skb_packet(tmp_skb, neigh_node->if_incoming,
+				    neigh_node->addr) == NET_XMIT_SUCCESS)
+			/* packet sent to a candidate: we can return true */
+			ret = true;
+free_neigh:
+		neigh_node_free_ref(neigh_node);
+free_orig:
+		orig_node_free_ref(cand[i].orig_node);
+	}
+
+out:
+	kfree(cand);
+	return ret;
+}
diff --git a/net/batman-adv/distributed-arp-table.h b/net/batman-adv/distributed-arp-table.h
new file mode 100644
index 0000000..94ee0fd
--- /dev/null
+++ b/net/batman-adv/distributed-arp-table.h
@@ -0,0 +1,66 @@
+/*
+ * Copyright (C) 2011 B.A.T.M.A.N. contributors:
+ *
+ * Antonio Quartulli
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of version 2 of the GNU General Public
+ * License as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA
+ * 02110-1301, USA
+ *
+ */
+
+#ifndef _NET_BATMAN_ADV_ARP_H_
+#define _NET_BATMAN_ADV_ARP_H_
+
+#include "types.h"
+#include "originator.h"
+
+#define DAT_ADDR_MAX biggest_unsigned_int(dat_addr_t)
+
+/* hash function to choose an entry in a hash table of given size.
+ * hash algorithm from http://en.wikipedia.org/wiki/Hash_table
+ */
+static inline uint32_t hash_ipv4(const void *data, uint32_t size)
+{
+	const unsigned char *key = data;
+	uint32_t hash = 0;
+	size_t i;
+
+	for (i = 0; i < 4; i++) {
+		hash += key[i];
+		hash += (hash << 10);
+		hash ^= (hash >> 6);
+	}
+
+	hash += (hash << 3);
+	hash ^= (hash >> 11);
+	hash += (hash << 15);
+
+	return hash % size;
+}
+
+static inline void dat_init_orig_node_dht_addr(struct orig_node *orig_node)
+{
+	orig_node->dht_addr = (dat_addr_t)choose_orig(orig_node->orig,
+						      DAT_ADDR_MAX);
+}
+
+static inline void dat_init_own_dht_addr(struct bat_priv *bat_priv,
+					 struct hard_iface *primary_if)
+{
+	bat_priv->dht_addr = (dat_addr_t)
+				choose_orig(primary_if->net_dev->dev_addr,
+					    DAT_ADDR_MAX);
+}
+
+#endif /* _NET_BATMAN_ADV_ARP_H_ */
diff --git a/net/batman-adv/hard-interface.c b/net/batman-adv/hard-interface.c
index 220cb02..1393939 100644
--- a/net/batman-adv/hard-interface.c
+++ b/net/batman-adv/hard-interface.c
@@ -20,6 +20,7 @@
  */
 
 #include "main.h"
+#include "distributed-arp-table.h"
 #include "hard-interface.h"
 #include "soft-interface.h"
 #include "send.h"
@@ -118,6 +119,8 @@ static void primary_if_update_addr(struct bat_priv *bat_priv,
 	if (!primary_if)
 		goto out;
 
+	dat_init_own_dht_addr(bat_priv, primary_if);
+
 	vis_packet = (struct vis_packet *)
 				bat_priv->my_vis_info->skb_packet->data;
 	memcpy(vis_packet->vis_orig, primary_if->net_dev->dev_addr, ETH_ALEN);
diff --git a/net/batman-adv/main.h b/net/batman-adv/main.h
index d9ef4ca..395e59d 100644
--- a/net/batman-adv/main.h
+++ b/net/batman-adv/main.h
@@ -67,6 +67,9 @@
 
 #define NUM_WORDS BITS_TO_LONGS(TQ_LOCAL_WINDOW_SIZE)
 
+/* numbers of originator to contact for any PUT/GET DHT operation */
+#define DHT_CANDIDATES_NUM 3
+
 #define LOG_BUF_LEN 8192	  /* has to be a power of 2 */
 
 #define VIS_INTERVAL 5000	/* 5 seconds */
@@ -111,6 +114,9 @@ enum uev_type {
 
 #define GW_THRESHOLD	50
 
+#define DHT_CANDIDATE_NOT_FOUND	0
+#define DHT_CANDIDATE_ORIG	1
+
 /* Debug Messages */
 #ifdef pr_fmt
 #undef pr_fmt
diff --git a/net/batman-adv/originator.c b/net/batman-adv/originator.c
index ce49698..31d7b58 100644
--- a/net/batman-adv/originator.c
+++ b/net/batman-adv/originator.c
@@ -20,6 +20,7 @@
  */
 
 #include "main.h"
+#include "distributed-arp-table.h"
 #include "originator.h"
 #include "hash.h"
 #include "translation-table.h"
@@ -224,6 +225,7 @@ struct orig_node *get_orig_node(struct bat_priv *bat_priv, const uint8_t *addr)
 	orig_node->tt_poss_change = false;
 	orig_node->bat_priv = bat_priv;
 	memcpy(orig_node->orig, addr, ETH_ALEN);
+	dat_init_orig_node_dht_addr(orig_node);
 	orig_node->router = NULL;
 	orig_node->tt_crc = 0;
 	atomic_set(&orig_node->last_ttvn, 0);
diff --git a/net/batman-adv/types.h b/net/batman-adv/types.h
index 2f4848b..61e2d78 100644
--- a/net/batman-adv/types.h
+++ b/net/batman-adv/types.h
@@ -27,6 +27,13 @@
 #include "packet.h"
 #include "bitarray.h"
 
+/* dat_addr_t is the type used for all DHT addresses. If it is changed,
+ * DAT_ADDR_MAX is changed as well.
+ *
+ * *Please be careful: dat_addr_t must be UNSIGNED*
+ */
+#define dat_addr_t uint16_t
+
 #define BAT_HEADER_LEN (ETH_HLEN + \
 	((sizeof(struct unicast_packet) > sizeof(struct bcast_packet) ? \
 	 sizeof(struct unicast_packet) : \
@@ -67,6 +74,7 @@ struct hard_iface {
 struct orig_node {
 	uint8_t orig[ETH_ALEN];
 	uint8_t primary_addr[ETH_ALEN];
+	dat_addr_t dht_addr;
 	struct neigh_node __rcu *router; /* rcu protected pointer */
 	unsigned long *bcast_own;
 	uint8_t *bcast_own_sum;
@@ -221,6 +229,7 @@ struct bat_priv {
 	struct gw_node __rcu *curr_gw;  /* rcu protected pointer */
 	atomic_t gw_reselect;
 	struct hard_iface __rcu *primary_if;  /* rcu protected pointer */
+	dat_addr_t dht_addr;
 	struct vis_info *my_vis_info;
 	struct bat_algo_ops *bat_algo_ops;
 };
@@ -395,4 +404,9 @@ struct bat_algo_ops {
 				struct sk_buff *skb);
 };
 
+struct dht_candidate {
+	int type;
+	struct orig_node *orig_node;
+};
+
 #endif /* _NET_BATMAN_ADV_TYPES_H_ */
diff --git a/net/batman-adv/unicast.c b/net/batman-adv/unicast.c
index 96c7cec..b7152c6 100644
--- a/net/batman-adv/unicast.c
+++ b/net/batman-adv/unicast.c
@@ -313,10 +313,10 @@ static bool prepare_unicast_packet(struct sk_buff *skb,
 				     orig_node);
 }
 
-static bool prepare_unicast_4addr_packet(struct bat_priv *bat_priv,
-					 struct sk_buff *skb,
-					 struct orig_node *orig_node,
-					 int packet_subtype)
+bool prepare_unicast_4addr_packet(struct bat_priv *bat_priv,
+				  struct sk_buff *skb,
+				  struct orig_node *orig_node,
+				  int packet_subtype)
 {
 	struct hard_iface *primary_if;
 	struct unicast_4addr_packet *unicast_4addr_packet;
diff --git a/net/batman-adv/unicast.h b/net/batman-adv/unicast.h
index ae9775b..e15aa62 100644
--- a/net/batman-adv/unicast.h
+++ b/net/batman-adv/unicast.h
@@ -32,6 +32,10 @@ int frag_reassemble_skb(struct sk_buff *skb, struct bat_priv *bat_priv,
 void frag_list_free(struct list_head *head);
 int frag_send_skb(struct sk_buff *skb, struct bat_priv *bat_priv,
 		  struct hard_iface *hard_iface, const uint8_t dstaddr[]);
+bool prepare_unicast_4addr_packet(struct bat_priv *bat_priv,
+				  struct sk_buff *skb,
+				  struct orig_node *orig_node,
+				  int packet_subtype);
 int unicast_generic_send_skb(struct sk_buff *skb, struct bat_priv *bat_priv,
 			     int packet_type, int packet_subtype);
 
-- 
1.7.9.4

[PATCH 05/15] batman-adv: Distributed ARP Table - add ARP parsing functions

[Antonio Quartulli]

ARP messages are now parsed to make it possible to trigger special actions
depending on their types (snooping).

Signed-off-by: Antonio Quartulli <ordex@autistici.org>
---
 net/batman-adv/distributed-arp-table.c |  123 ++++++++++++++++++++++++++++++++
 net/batman-adv/distributed-arp-table.h |   10 +++
 net/batman-adv/packet.h                |    5 +-
 3 files changed, 137 insertions(+), 1 deletion(-)

diff --git a/net/batman-adv/distributed-arp-table.c b/net/batman-adv/distributed-arp-table.c
index 0078478..682fb1e 100644
--- a/net/batman-adv/distributed-arp-table.c
+++ b/net/batman-adv/distributed-arp-table.c
@@ -30,6 +30,75 @@
 #include "types.h"
 #include "unicast.h"
 
+#ifdef CONFIG_BATMAN_ADV_DEBUG
+
+static void bat_dbg_arp(struct bat_priv *bat_priv, struct sk_buff *skb,
+			uint16_t type, int hdr_size, char *msg)
+{
+	struct unicast_4addr_packet *unicast_4addr_packet;
+
+	if (msg)
+		bat_dbg(DBG_DAT, bat_priv, "%s\n", msg);
+
+	bat_dbg(DBG_DAT, bat_priv, "ARP MSG = [src: %pM-%pI4 dst: %pM-%pI4]\n",
+		ARP_HW_SRC(skb, hdr_size), &ARP_IP_SRC(skb, hdr_size),
+		ARP_HW_DST(skb, hdr_size), &ARP_IP_DST(skb, hdr_size));
+
+	if (hdr_size == 0)
+		return;
+
+	/* if the AP packet is encapsulated in a batman packet, let's print some
+	 * debug messages
+	 */
+	unicast_4addr_packet = (struct unicast_4addr_packet *)skb->data;
+
+	switch (unicast_4addr_packet->u.header.packet_type) {
+	case BAT_UNICAST:
+		bat_dbg(DBG_DAT, bat_priv,
+			"* encapsulated within a UNICAST packet\n");
+		break;
+	case BAT_UNICAST_4ADDR:
+		bat_dbg(DBG_DAT, bat_priv,
+			"* encapsulated within a UNICAST_4ADDR packet (src: %pM)\n",
+			unicast_4addr_packet->src);
+		switch (unicast_4addr_packet->subtype) {
+		case BAT_P_DAT_DHT_PUT:
+			bat_dbg(DBG_DAT, bat_priv, "* type: DAT_DHT_PUT\n");
+			break;
+		case BAT_P_DAT_DHT_GET:
+			bat_dbg(DBG_DAT, bat_priv, "* type: DAT_DHT_GET\n");
+			break;
+		case BAT_P_DAT_CACHE_REPLY:
+			bat_dbg(DBG_DAT, bat_priv, "* type: DAT_CACHE_REPLY\n");
+			break;
+		case BAT_P_DATA:
+			bat_dbg(DBG_DAT, bat_priv, "* type: DATA\n");
+			break;
+		default:
+			bat_dbg(DBG_DAT, bat_priv, "* type: Unknown!\n");
+		}
+		break;
+	case BAT_BCAST:
+		bat_dbg(DBG_DAT, bat_priv,
+			"* encapsulated within a BCAST packet (src: %pM)\n",
+			((struct bcast_packet *)unicast_4addr_packet)->orig);
+		break;
+	default:
+		bat_dbg(DBG_DAT, bat_priv,
+			"* encapsulated within an unknown packet type (0x%x)\n",
+			unicast_4addr_packet->u.header.packet_type);
+	}
+}
+
+#else
+
+static void bat_dbg_arp(struct bat_priv *bat_priv, struct sk_buff *skb,
+			uint16_t type, int hdr_size, char *msg)
+{
+}
+
+#endif /* CONFIG_BATMAN_ADV_DEBUG */
+
 static bool is_orig_node_eligible(struct dht_candidate *res, int select,
 				  dat_addr_t tmp_max, dat_addr_t max,
 				  dat_addr_t last_max,
@@ -205,3 +274,57 @@ out:
 	kfree(cand);
 	return ret;
 }
+
+/* Returns arphdr->ar_op if the skb contains a valid ARP packet, otherwise
+ * returns 0
+ */
+static uint16_t arp_get_type(struct bat_priv *bat_priv, struct sk_buff *skb,
+			     int hdr_size)
+{
+	struct arphdr *arphdr;
+	struct ethhdr *ethhdr;
+	uint32_t ip_src, ip_dst;
+	uint16_t type = 0;
+
+	/* pull the ethernet header */
+	if (unlikely(!pskb_may_pull(skb, hdr_size + ETH_HLEN)))
+		goto out;
+
+	ethhdr = (struct ethhdr *)(skb->data + hdr_size);
+
+	if (ethhdr->h_proto != htons(ETH_P_ARP))
+		goto out;
+
+	/* pull the ARP payload */
+	if (unlikely(!pskb_may_pull(skb, hdr_size + ETH_HLEN +
+				    arp_hdr_len(skb->dev))))
+		goto out;
+
+	arphdr = (struct arphdr *)(skb->data + hdr_size + ETH_HLEN);
+
+	/* Check whether the ARP packet carries a valid
+	 * IP information */
+	if (arphdr->ar_hrd != htons(ARPHRD_ETHER))
+		goto out;
+
+	if (arphdr->ar_pro != htons(ETH_P_IP))
+		goto out;
+
+	if (arphdr->ar_hln != ETH_ALEN)
+		goto out;
+
+	if (arphdr->ar_pln != 4)
+		goto out;
+
+	/* Check for bad reply/request. If the ARP message is not sane, DAT
+	 * will simply ignore it */
+	ip_src = ARP_IP_SRC(skb, hdr_size);
+	ip_dst = ARP_IP_DST(skb, hdr_size);
+	if (ipv4_is_loopback(ip_src) || ipv4_is_multicast(ip_src) ||
+	    ipv4_is_loopback(ip_dst) || ipv4_is_multicast(ip_dst))
+		goto out;
+
+	type = ntohs(arphdr->ar_op);
+out:
+	return type;
+}
diff --git a/net/batman-adv/distributed-arp-table.h b/net/batman-adv/distributed-arp-table.h
index 94ee0fd..cdd0484 100644
--- a/net/batman-adv/distributed-arp-table.h
+++ b/net/batman-adv/distributed-arp-table.h
@@ -25,8 +25,18 @@
 #include "types.h"
 #include "originator.h"
 
+#include <linux/if_arp.h>
+
 #define DAT_ADDR_MAX biggest_unsigned_int(dat_addr_t)
 
+#define ARP_HW_SRC(skb, hdr_size) ((uint8_t *)(skb->data + hdr_size) + \
+				   ETH_HLEN + sizeof(struct arphdr))
+#define ARP_IP_SRC(skb, hdr_size) (*(uint32_t *)(ARP_HW_SRC(skb, hdr_size) + \
+				   ETH_ALEN))
+#define ARP_HW_DST(skb, hdr_size) (ARP_HW_SRC(skb, hdr_size) + ETH_ALEN + 4)
+#define ARP_IP_DST(skb, hdr_size) (*(uint32_t *)(ARP_HW_SRC(skb, hdr_size) + \
+				   ETH_ALEN * 2 + 4))
+
 /* hash function to choose an entry in a hash table of given size.
  * hash algorithm from http://en.wikipedia.org/wiki/Hash_table
  */
diff --git a/net/batman-adv/packet.h b/net/batman-adv/packet.h
index e9f5184..c4533c4 100644
--- a/net/batman-adv/packet.h
+++ b/net/batman-adv/packet.h
@@ -37,7 +37,10 @@ enum bat_packettype {
 };
 
 enum bat_subtype {
-	BAT_P_DATA		= 0x01
+	BAT_P_DATA		= 0x01,
+	BAT_P_DAT_DHT_GET	= 0x02,
+	BAT_P_DAT_DHT_PUT	= 0x03,
+	BAT_P_DAT_CACHE_REPLY	= 0x04
 };
 
 /* this file is included by batctl which needs these defines */
-- 
1.7.9.4

[B.A.T.M.A.N.] [PATCH 05/15] batman-adv: Distributed ARP Table - add ARP parsing functions

[Antonio Quartulli]

ARP messages are now parsed to make it possible to trigger special actions
depending on their types (snooping).

Signed-off-by: Antonio Quartulli <ordex@autistici.org>
---
 net/batman-adv/distributed-arp-table.c |  123 ++++++++++++++++++++++++++++++++
 net/batman-adv/distributed-arp-table.h |   10 +++
 net/batman-adv/packet.h                |    5 +-
 3 files changed, 137 insertions(+), 1 deletion(-)

diff --git a/net/batman-adv/distributed-arp-table.c b/net/batman-adv/distributed-arp-table.c
index 0078478..682fb1e 100644
--- a/net/batman-adv/distributed-arp-table.c
+++ b/net/batman-adv/distributed-arp-table.c
@@ -30,6 +30,75 @@
 #include "types.h"
 #include "unicast.h"
 
+#ifdef CONFIG_BATMAN_ADV_DEBUG
+
+static void bat_dbg_arp(struct bat_priv *bat_priv, struct sk_buff *skb,
+			uint16_t type, int hdr_size, char *msg)
+{
+	struct unicast_4addr_packet *unicast_4addr_packet;
+
+	if (msg)
+		bat_dbg(DBG_DAT, bat_priv, "%s\n", msg);
+
+	bat_dbg(DBG_DAT, bat_priv, "ARP MSG = [src: %pM-%pI4 dst: %pM-%pI4]\n",
+		ARP_HW_SRC(skb, hdr_size), &ARP_IP_SRC(skb, hdr_size),
+		ARP_HW_DST(skb, hdr_size), &ARP_IP_DST(skb, hdr_size));
+
+	if (hdr_size == 0)
+		return;
+
+	/* if the AP packet is encapsulated in a batman packet, let's print some
+	 * debug messages
+	 */
+	unicast_4addr_packet = (struct unicast_4addr_packet *)skb->data;
+
+	switch (unicast_4addr_packet->u.header.packet_type) {
+	case BAT_UNICAST:
+		bat_dbg(DBG_DAT, bat_priv,
+			"* encapsulated within a UNICAST packet\n");
+		break;
+	case BAT_UNICAST_4ADDR:
+		bat_dbg(DBG_DAT, bat_priv,
+			"* encapsulated within a UNICAST_4ADDR packet (src: %pM)\n",
+			unicast_4addr_packet->src);
+		switch (unicast_4addr_packet->subtype) {
+		case BAT_P_DAT_DHT_PUT:
+			bat_dbg(DBG_DAT, bat_priv, "* type: DAT_DHT_PUT\n");
+			break;
+		case BAT_P_DAT_DHT_GET:
+			bat_dbg(DBG_DAT, bat_priv, "* type: DAT_DHT_GET\n");
+			break;
+		case BAT_P_DAT_CACHE_REPLY:
+			bat_dbg(DBG_DAT, bat_priv, "* type: DAT_CACHE_REPLY\n");
+			break;
+		case BAT_P_DATA:
+			bat_dbg(DBG_DAT, bat_priv, "* type: DATA\n");
+			break;
+		default:
+			bat_dbg(DBG_DAT, bat_priv, "* type: Unknown!\n");
+		}
+		break;
+	case BAT_BCAST:
+		bat_dbg(DBG_DAT, bat_priv,
+			"* encapsulated within a BCAST packet (src: %pM)\n",
+			((struct bcast_packet *)unicast_4addr_packet)->orig);
+		break;
+	default:
+		bat_dbg(DBG_DAT, bat_priv,
+			"* encapsulated within an unknown packet type (0x%x)\n",
+			unicast_4addr_packet->u.header.packet_type);
+	}
+}
+
+#else
+
+static void bat_dbg_arp(struct bat_priv *bat_priv, struct sk_buff *skb,
+			uint16_t type, int hdr_size, char *msg)
+{
+}
+
+#endif /* CONFIG_BATMAN_ADV_DEBUG */
+
 static bool is_orig_node_eligible(struct dht_candidate *res, int select,
 				  dat_addr_t tmp_max, dat_addr_t max,
 				  dat_addr_t last_max,
@@ -205,3 +274,57 @@ out:
 	kfree(cand);
 	return ret;
 }
+
+/* Returns arphdr->ar_op if the skb contains a valid ARP packet, otherwise
+ * returns 0
+ */
+static uint16_t arp_get_type(struct bat_priv *bat_priv, struct sk_buff *skb,
+			     int hdr_size)
+{
+	struct arphdr *arphdr;
+	struct ethhdr *ethhdr;
+	uint32_t ip_src, ip_dst;
+	uint16_t type = 0;
+
+	/* pull the ethernet header */
+	if (unlikely(!pskb_may_pull(skb, hdr_size + ETH_HLEN)))
+		goto out;
+
+	ethhdr = (struct ethhdr *)(skb->data + hdr_size);
+
+	if (ethhdr->h_proto != htons(ETH_P_ARP))
+		goto out;
+
+	/* pull the ARP payload */
+	if (unlikely(!pskb_may_pull(skb, hdr_size + ETH_HLEN +
+				    arp_hdr_len(skb->dev))))
+		goto out;
+
+	arphdr = (struct arphdr *)(skb->data + hdr_size + ETH_HLEN);
+
+	/* Check whether the ARP packet carries a valid
+	 * IP information */
+	if (arphdr->ar_hrd != htons(ARPHRD_ETHER))
+		goto out;
+
+	if (arphdr->ar_pro != htons(ETH_P_IP))
+		goto out;
+
+	if (arphdr->ar_hln != ETH_ALEN)
+		goto out;
+
+	if (arphdr->ar_pln != 4)
+		goto out;
+
+	/* Check for bad reply/request. If the ARP message is not sane, DAT
+	 * will simply ignore it */
+	ip_src = ARP_IP_SRC(skb, hdr_size);
+	ip_dst = ARP_IP_DST(skb, hdr_size);
+	if (ipv4_is_loopback(ip_src) || ipv4_is_multicast(ip_src) ||
+	    ipv4_is_loopback(ip_dst) || ipv4_is_multicast(ip_dst))
+		goto out;
+
+	type = ntohs(arphdr->ar_op);
+out:
+	return type;
+}
diff --git a/net/batman-adv/distributed-arp-table.h b/net/batman-adv/distributed-arp-table.h
index 94ee0fd..cdd0484 100644
--- a/net/batman-adv/distributed-arp-table.h
+++ b/net/batman-adv/distributed-arp-table.h
@@ -25,8 +25,18 @@
 #include "types.h"
 #include "originator.h"
 
+#include <linux/if_arp.h>
+
 #define DAT_ADDR_MAX biggest_unsigned_int(dat_addr_t)
 
+#define ARP_HW_SRC(skb, hdr_size) ((uint8_t *)(skb->data + hdr_size) + \
+				   ETH_HLEN + sizeof(struct arphdr))
+#define ARP_IP_SRC(skb, hdr_size) (*(uint32_t *)(ARP_HW_SRC(skb, hdr_size) + \
+				   ETH_ALEN))
+#define ARP_HW_DST(skb, hdr_size) (ARP_HW_SRC(skb, hdr_size) + ETH_ALEN + 4)
+#define ARP_IP_DST(skb, hdr_size) (*(uint32_t *)(ARP_HW_SRC(skb, hdr_size) + \
+				   ETH_ALEN * 2 + 4))
+
 /* hash function to choose an entry in a hash table of given size.
  * hash algorithm from http://en.wikipedia.org/wiki/Hash_table
  */
diff --git a/net/batman-adv/packet.h b/net/batman-adv/packet.h
index e9f5184..c4533c4 100644
--- a/net/batman-adv/packet.h
+++ b/net/batman-adv/packet.h
@@ -37,7 +37,10 @@ enum bat_packettype {
 };
 
 enum bat_subtype {
-	BAT_P_DATA		= 0x01
+	BAT_P_DATA		= 0x01,
+	BAT_P_DAT_DHT_GET	= 0x02,
+	BAT_P_DAT_DHT_PUT	= 0x03,
+	BAT_P_DAT_CACHE_REPLY	= 0x04
 };
 
 /* this file is included by batctl which needs these defines */
-- 
1.7.9.4

[PATCH 06/15] batman-adv: Distributed ARP Table - add snooping functions for ARP messages

[Antonio Quartulli]

In case of an ARP message going in or out the soft_iface, it is intercepted and
a special action is performed. In particular the DHT helper functions previously
implemented are used to store all the ARP entries belonging to the network in
order to provide a fast and unicast lookup instead of the classic broadcast
flooding mechanism.
Each node stores the entries it is responsible for (following the DHT rules) in
its soft_iface ARP table. This makes it possible to reuse the kernel data
structures and functions for ARP management.

Signed-off-by: Antonio Quartulli <ordex@autistici.org>
---
 net/batman-adv/Kconfig                 |    2 +-
 net/batman-adv/distributed-arp-table.c |  255 ++++++++++++++++++++++++++++++++
 net/batman-adv/distributed-arp-table.h |   11 ++
 net/batman-adv/main.h                  |    2 +
 net/batman-adv/send.c                  |    4 +
 net/batman-adv/soft-interface.c        |   15 +-
 6 files changed, 287 insertions(+), 2 deletions(-)

diff --git a/net/batman-adv/Kconfig b/net/batman-adv/Kconfig
index 53f5244..b25e20f 100644
--- a/net/batman-adv/Kconfig
+++ b/net/batman-adv/Kconfig
@@ -4,7 +4,7 @@
 
 config BATMAN_ADV
 	tristate "B.A.T.M.A.N. Advanced Meshing Protocol"
-	depends on NET
+	depends on NET && INET
 	select CRC16
         default n
 	help
diff --git a/net/batman-adv/distributed-arp-table.c b/net/batman-adv/distributed-arp-table.c
index 682fb1e..f3b63ef 100644
--- a/net/batman-adv/distributed-arp-table.c
+++ b/net/batman-adv/distributed-arp-table.c
@@ -21,6 +21,8 @@
 
 #include <linux/if_ether.h>
 #include <linux/if_arp.h>
+/* needed to use arp_tbl */
+#include <net/arp.h>
 
 #include "main.h"
 #include "distributed-arp-table.h"
@@ -28,6 +30,7 @@
 #include "originator.h"
 #include "send.h"
 #include "types.h"
+#include "translation-table.h"
 #include "unicast.h"
 
 #ifdef CONFIG_BATMAN_ADV_DEBUG
@@ -275,6 +278,32 @@ out:
 	return ret;
 }
 
+/* Update the neighbour entry corresponding to the IP passed as parameter with
+ * the hw address hw. If the neighbour entry doesn't exists, then it will be
+ * created
+ */
+static void arp_neigh_update(struct bat_priv *bat_priv, uint32_t ip,
+			     uint8_t *hw)
+{
+	struct neighbour *n = NULL;
+	struct hard_iface *primary_if = primary_if_get_selected(bat_priv);
+	if (!primary_if)
+		goto out;
+
+	n = __neigh_lookup(&arp_tbl, &ip, primary_if->soft_iface, 1);
+	if (!n)
+		goto out;
+
+	bat_dbg(DBG_DAT, bat_priv, "Updating neighbour: %pI4 - %pM\n", &ip, hw);
+
+	neigh_update(n, hw, NUD_REACHABLE, NEIGH_UPDATE_F_OVERRIDE);
+out:
+	if (n && !IS_ERR(n))
+		neigh_release(n);
+	if (primary_if)
+		hardif_free_ref(primary_if);
+}
+
 /* Returns arphdr->ar_op if the skb contains a valid ARP packet, otherwise
  * returns 0
  */
@@ -328,3 +357,229 @@ static uint16_t arp_get_type(struct bat_priv *bat_priv, struct sk_buff *skb,
 out:
 	return type;
 }
+
+/* return true if the message has been sent to the dht candidates, false
+ * otherwise. In case of true the message has to be enqueued to permit the
+ * fallback
+ */
+bool dat_snoop_outgoing_arp_request(struct bat_priv *bat_priv,
+				    struct sk_buff *skb)
+{
+	uint16_t type = 0;
+	uint32_t ip_dst, ip_src;
+	uint8_t *hw_src;
+	bool ret = false;
+	struct neighbour *n = NULL;
+	struct hard_iface *primary_if = NULL;
+	struct sk_buff *skb_new;
+
+	type = arp_get_type(bat_priv, skb, 0);
+	/* If we get an ARP_REQUEST we have to send the unicast message to the
+	 * selected DHT candidates
+	 */
+	if (type != ARPOP_REQUEST)
+		goto out;
+
+	bat_dbg_arp(bat_priv, skb, type, 0, "Parsing outgoing ARP REQUEST");
+
+	ip_src = ARP_IP_SRC(skb, 0);
+	hw_src = ARP_HW_SRC(skb, 0);
+	ip_dst = ARP_IP_DST(skb, 0);
+
+	primary_if = primary_if_get_selected(bat_priv);
+	if (!primary_if)
+		goto out;
+
+	arp_neigh_update(bat_priv, ip_src, hw_src);
+
+	n = neigh_lookup(&arp_tbl, &ip_dst, primary_if->soft_iface);
+	/* check if it is a valid neigh entry */
+	if (n && (n->nud_state & NUD_CONNECTED)) {
+		skb_new = arp_create(ARPOP_REPLY, ETH_P_ARP, ip_src,
+				     primary_if->soft_iface, ip_dst, hw_src,
+				     n->ha, hw_src);
+		if (!skb_new)
+			goto out;
+
+		skb_reset_mac_header(skb_new);
+		skb_new->protocol = eth_type_trans(skb_new,
+						   primary_if->soft_iface);
+		bat_priv->stats.rx_packets++;
+		bat_priv->stats.rx_bytes += skb->len + ETH_HLEN;
+		primary_if->soft_iface->last_rx = jiffies;
+
+		netif_rx(skb_new);
+		bat_dbg(DBG_DAT, bat_priv, "ARP request replied locally\n");
+	} else
+		/* Send the request on the DHT */
+		ret = dht_send_data(bat_priv, skb, ip_dst, BAT_P_DAT_DHT_GET);
+out:
+	if (n)
+		neigh_release(n);
+	if (primary_if)
+		hardif_free_ref(primary_if);
+	return ret;
+}
+
+/* This function is meant to be invoked for an ARP request which is coming into
+ * the bat0 interfaces from the mesh network. It will check for the needed data
+ * into the local table. If found, an ARP reply is sent immediately, otherwise
+ * the caller has to deliver the ARP request to the upper layer
+ */
+bool dat_snoop_incoming_arp_request(struct bat_priv *bat_priv,
+				    struct sk_buff *skb, int hdr_size)
+{
+	uint16_t type;
+	uint32_t ip_src, ip_dst;
+	uint8_t *hw_src;
+	struct hard_iface *primary_if = NULL;
+	struct sk_buff *skb_new;
+	struct neighbour *n = NULL;
+	bool ret = false;
+
+	type = arp_get_type(bat_priv, skb, hdr_size);
+	if (type != ARPOP_REQUEST)
+		goto out;
+
+	hw_src = ARP_HW_SRC(skb, hdr_size);
+	ip_src = ARP_IP_SRC(skb, hdr_size);
+	ip_dst = ARP_IP_DST(skb, hdr_size);
+
+	bat_dbg_arp(bat_priv, skb, type, hdr_size,
+		    "Parsing incoming ARP REQUEST");
+
+	primary_if = primary_if_get_selected(bat_priv);
+	if (!primary_if)
+		goto out;
+
+	arp_neigh_update(bat_priv, ip_src, hw_src);
+
+	n = neigh_lookup(&arp_tbl, &ip_dst, primary_if->soft_iface);
+	/* check if it is a valid neigh entry */
+	if (!n || !(n->nud_state & NUD_CONNECTED))
+		goto out;
+
+	skb_new = arp_create(ARPOP_REPLY, ETH_P_ARP, ip_src,
+			     primary_if->soft_iface, ip_dst, hw_src, n->ha,
+			     hw_src);
+
+	if (!skb_new)
+		goto out;
+
+	unicast_4addr_send_skb(skb_new, bat_priv, BAT_P_DAT_CACHE_REPLY);
+
+	ret = true;
+out:
+	if (n)
+		neigh_release(n);
+	if (primary_if)
+		hardif_free_ref(primary_if);
+	if (ret)
+		kfree_skb(skb);
+	return ret;
+}
+
+/* This function is meant to be invoked on an ARP reply packet going into the
+ * soft interface. The related neighbour entry has to be updated and the DHT has
+ * to be populated as well
+ */
+bool dat_snoop_outgoing_arp_reply(struct bat_priv *bat_priv,
+				  struct sk_buff *skb)
+{
+	uint16_t type;
+	uint32_t ip_src, ip_dst;
+	uint8_t *hw_src, *hw_dst;
+	bool ret = false;
+
+	type = arp_get_type(bat_priv, skb, 0);
+	if (type != ARPOP_REPLY)
+		goto out;
+
+	bat_dbg_arp(bat_priv, skb, type, 0, "Parsing outgoing ARP REPLY");
+
+	hw_src = ARP_HW_SRC(skb, 0);
+	ip_src = ARP_IP_SRC(skb, 0);
+	hw_dst = ARP_HW_DST(skb, 0);
+	ip_dst = ARP_IP_DST(skb, 0);
+
+	arp_neigh_update(bat_priv, ip_src, hw_src);
+	arp_neigh_update(bat_priv, ip_dst, hw_dst);
+
+	/* Send the ARP reply to the candidates for both the IP addresses we
+	 * fetched from the ARP reply
+	 */
+	dht_send_data(bat_priv, skb, ip_src, BAT_P_DAT_DHT_PUT);
+	dht_send_data(bat_priv, skb, ip_dst, BAT_P_DAT_DHT_PUT);
+	ret = true;
+out:
+	return ret;
+}
+
+/* This function has to be invoked on an ARP reply coming into the soft
+ * interface from the mesh network. The local table has to be updated
+ */
+bool dat_snoop_incoming_arp_reply(struct bat_priv *bat_priv,
+				  struct sk_buff *skb, int hdr_size)
+{
+	uint16_t type;
+	uint32_t ip_src, ip_dst;
+	uint8_t *hw_src, *hw_dst;
+	bool ret = false;
+
+	type = arp_get_type(bat_priv, skb, hdr_size);
+	if (type != ARPOP_REPLY)
+		goto out;
+
+	bat_dbg_arp(bat_priv, skb, type, hdr_size,
+		    "Parsing incoming ARP REPLY");
+
+	hw_src = ARP_HW_SRC(skb, hdr_size);
+	ip_src = ARP_IP_SRC(skb, hdr_size);
+	hw_dst = ARP_HW_DST(skb, hdr_size);
+	ip_dst = ARP_IP_DST(skb, hdr_size);
+
+	/* Update our internal cache with both the IP addresses we fetched from
+	 * the ARP reply
+	 */
+	arp_neigh_update(bat_priv, ip_src, hw_src);
+	arp_neigh_update(bat_priv, ip_dst, hw_dst);
+
+	/* if this REPLY is directed to a client of mine, let's deliver the
+	 * packet to the interface
+	 */
+	ret = !is_my_client(bat_priv, hw_dst);
+out:
+	/* if ret == false packet has to be delivered to the interface */
+	return ret;
+}
+
+bool dat_drop_broadcast_packet(struct bat_priv *bat_priv,
+			       struct forw_packet *forw_packet)
+{
+	struct neighbour *n;
+
+	/* If this packet is an ARP_REQUEST and we already have the information
+	 * that it is going to ask, we can drop the packet
+	 */
+	if (!forw_packet->num_packets &&
+	    (ARPOP_REQUEST == arp_get_type(bat_priv, forw_packet->skb,
+					   sizeof(struct bcast_packet)))) {
+		n = neigh_lookup(&arp_tbl,
+				 &ARP_IP_DST(forw_packet->skb,
+					     sizeof(struct bcast_packet)),
+				 forw_packet->if_incoming->soft_iface);
+		/* check if we already know this neigh */
+		if (n && (n->nud_state & NUD_CONNECTED)) {
+			bat_dbg(DBG_DAT, bat_priv,
+				"ARP Request for %pI4: fallback prevented\n",
+				&ARP_IP_DST(forw_packet->skb,
+					    sizeof(struct bcast_packet)));
+			return true;
+		}
+
+		bat_dbg(DBG_DAT, bat_priv, "ARP Request for %pI4: fallback\n",
+			&ARP_IP_DST(forw_packet->skb,
+				    sizeof(struct bcast_packet)));
+	}
+	return false;
+}
diff --git a/net/batman-adv/distributed-arp-table.h b/net/batman-adv/distributed-arp-table.h
index cdd0484..8b7aa87 100644
--- a/net/batman-adv/distributed-arp-table.h
+++ b/net/batman-adv/distributed-arp-table.h
@@ -37,6 +37,17 @@
 #define ARP_IP_DST(skb, hdr_size) (*(uint32_t *)(ARP_HW_SRC(skb, hdr_size) + \
 				   ETH_ALEN * 2 + 4))
 
+bool dat_snoop_outgoing_arp_request(struct bat_priv *bat_priv,
+				    struct sk_buff *skb);
+bool dat_snoop_incoming_arp_request(struct bat_priv *bat_priv,
+				    struct sk_buff *skb, int hdr_size);
+bool dat_snoop_outgoing_arp_reply(struct bat_priv *bat_priv,
+				  struct sk_buff *skb);
+bool dat_snoop_incoming_arp_reply(struct bat_priv *bat_priv,
+				  struct sk_buff *skb, int hdr_size);
+bool dat_drop_broadcast_packet(struct bat_priv *bat_priv,
+			       struct forw_packet *forw_packet);
+
 /* hash function to choose an entry in a hash table of given size.
  * hash algorithm from http://en.wikipedia.org/wiki/Hash_table
  */
diff --git a/net/batman-adv/main.h b/net/batman-adv/main.h
index 395e59d..57023c3 100644
--- a/net/batman-adv/main.h
+++ b/net/batman-adv/main.h
@@ -67,6 +67,8 @@
 
 #define NUM_WORDS BITS_TO_LONGS(TQ_LOCAL_WINDOW_SIZE)
 
+/* msecs after which an ARP_REQUEST is sent in broadcast as fallback */
+#define ARP_REQ_DELAY 250
 /* numbers of originator to contact for any PUT/GET DHT operation */
 #define DHT_CANDIDATES_NUM 3
 
diff --git a/net/batman-adv/send.c b/net/batman-adv/send.c
index 7c66b61..91eaa45 100644
--- a/net/batman-adv/send.c
+++ b/net/batman-adv/send.c
@@ -20,6 +20,7 @@
  */
 
 #include "main.h"
+#include "distributed-arp-table.h"
 #include "send.h"
 #include "routing.h"
 #include "translation-table.h"
@@ -274,6 +275,9 @@ static void send_outstanding_bcast_packet(struct work_struct *work)
 	if (atomic_read(&bat_priv->mesh_state) == MESH_DEACTIVATING)
 		goto out;
 
+	if (dat_drop_broadcast_packet(bat_priv, forw_packet))
+		goto out;
+
 	/* rebroadcast packet */
 	rcu_read_lock();
 	list_for_each_entry_rcu(hard_iface, &hardif_list, list) {
diff --git a/net/batman-adv/soft-interface.c b/net/batman-adv/soft-interface.c
index 6e2530b..3a1483a 100644
--- a/net/batman-adv/soft-interface.c
+++ b/net/batman-adv/soft-interface.c
@@ -22,6 +22,7 @@
 #include "main.h"
 #include "soft-interface.h"
 #include "hard-interface.h"
+#include "distributed-arp-table.h"
 #include "routing.h"
 #include "send.h"
 #include "bat_debugfs.h"
@@ -136,6 +137,7 @@ static int interface_tx(struct sk_buff *skb, struct net_device *soft_iface)
 	int data_len = skb->len, ret;
 	short vid __maybe_unused = -1;
 	bool do_bcast = false;
+	unsigned long brd_delay = 1;
 
 	if (atomic_read(&bat_priv->mesh_state) != MESH_ACTIVE)
 		goto dropped;
@@ -197,6 +199,9 @@ static int interface_tx(struct sk_buff *skb, struct net_device *soft_iface)
 		if (!primary_if)
 			goto dropped;
 
+		if (dat_snoop_outgoing_arp_request(bat_priv, skb))
+			brd_delay = msecs_to_jiffies(ARP_REQ_DELAY);
+
 		if (my_skb_head_push(skb, sizeof(*bcast_packet)) < 0)
 			goto dropped;
 
@@ -216,7 +221,7 @@ static int interface_tx(struct sk_buff *skb, struct net_device *soft_iface)
 		bcast_packet->seqno =
 			htonl(atomic_inc_return(&bat_priv->bcast_seqno));
 
-		add_bcast_packet_to_list(bat_priv, skb, 1);
+		add_bcast_packet_to_list(bat_priv, skb, brd_delay);
 
 		/* a copy is stored in the bcast list, therefore removing
 		 * the original skb. */
@@ -230,6 +235,8 @@ static int interface_tx(struct sk_buff *skb, struct net_device *soft_iface)
 				goto dropped;
 		}
 
+		dat_snoop_outgoing_arp_reply(bat_priv, skb);
+
 		ret = unicast_send_skb(skb, bat_priv);
 		if (ret != 0)
 			goto dropped_freed;
@@ -262,6 +269,12 @@ void interface_rx(struct net_device *soft_iface,
 	if (!pskb_may_pull(skb, hdr_size))
 		goto dropped;
 
+	if (dat_snoop_incoming_arp_request(bat_priv, skb, hdr_size))
+		goto out;
+
+	if (dat_snoop_incoming_arp_reply(bat_priv, skb, hdr_size))
+		goto out;
+
 	skb_pull_rcsum(skb, hdr_size);
 	skb_reset_mac_header(skb);
 
-- 
1.7.9.4

[B.A.T.M.A.N.] [PATCH 06/15] batman-adv: Distributed ARP Table - add snooping functions for ARP messages

[Antonio Quartulli]

In case of an ARP message going in or out the soft_iface, it is intercepted and
a special action is performed. In particular the DHT helper functions previously
implemented are used to store all the ARP entries belonging to the network in
order to provide a fast and unicast lookup instead of the classic broadcast
flooding mechanism.
Each node stores the entries it is responsible for (following the DHT rules) in
its soft_iface ARP table. This makes it possible to reuse the kernel data
structures and functions for ARP management.

Signed-off-by: Antonio Quartulli <ordex@autistici.org>
---
 net/batman-adv/Kconfig                 |    2 +-
 net/batman-adv/distributed-arp-table.c |  255 ++++++++++++++++++++++++++++++++
 net/batman-adv/distributed-arp-table.h |   11 ++
 net/batman-adv/main.h                  |    2 +
 net/batman-adv/send.c                  |    4 +
 net/batman-adv/soft-interface.c        |   15 +-
 6 files changed, 287 insertions(+), 2 deletions(-)

diff --git a/net/batman-adv/Kconfig b/net/batman-adv/Kconfig
index 53f5244..b25e20f 100644
--- a/net/batman-adv/Kconfig
+++ b/net/batman-adv/Kconfig
@@ -4,7 +4,7 @@
 
 config BATMAN_ADV
 	tristate "B.A.T.M.A.N. Advanced Meshing Protocol"
-	depends on NET
+	depends on NET && INET
 	select CRC16
         default n
 	help
diff --git a/net/batman-adv/distributed-arp-table.c b/net/batman-adv/distributed-arp-table.c
index 682fb1e..f3b63ef 100644
--- a/net/batman-adv/distributed-arp-table.c
+++ b/net/batman-adv/distributed-arp-table.c
@@ -21,6 +21,8 @@
 
 #include <linux/if_ether.h>
 #include <linux/if_arp.h>
+/* needed to use arp_tbl */
+#include <net/arp.h>
 
 #include "main.h"
 #include "distributed-arp-table.h"
@@ -28,6 +30,7 @@
 #include "originator.h"
 #include "send.h"
 #include "types.h"
+#include "translation-table.h"
 #include "unicast.h"
 
 #ifdef CONFIG_BATMAN_ADV_DEBUG
@@ -275,6 +278,32 @@ out:
 	return ret;
 }
 
+/* Update the neighbour entry corresponding to the IP passed as parameter with
+ * the hw address hw. If the neighbour entry doesn't exists, then it will be
+ * created
+ */
+static void arp_neigh_update(struct bat_priv *bat_priv, uint32_t ip,
+			     uint8_t *hw)
+{
+	struct neighbour *n = NULL;
+	struct hard_iface *primary_if = primary_if_get_selected(bat_priv);
+	if (!primary_if)
+		goto out;
+
+	n = __neigh_lookup(&arp_tbl, &ip, primary_if->soft_iface, 1);
+	if (!n)
+		goto out;
+
+	bat_dbg(DBG_DAT, bat_priv, "Updating neighbour: %pI4 - %pM\n", &ip, hw);
+
+	neigh_update(n, hw, NUD_REACHABLE, NEIGH_UPDATE_F_OVERRIDE);
+out:
+	if (n && !IS_ERR(n))
+		neigh_release(n);
+	if (primary_if)
+		hardif_free_ref(primary_if);
+}
+
 /* Returns arphdr->ar_op if the skb contains a valid ARP packet, otherwise
  * returns 0
  */
@@ -328,3 +357,229 @@ static uint16_t arp_get_type(struct bat_priv *bat_priv, struct sk_buff *skb,
 out:
 	return type;
 }
+
+/* return true if the message has been sent to the dht candidates, false
+ * otherwise. In case of true the message has to be enqueued to permit the
+ * fallback
+ */
+bool dat_snoop_outgoing_arp_request(struct bat_priv *bat_priv,
+				    struct sk_buff *skb)
+{
+	uint16_t type = 0;
+	uint32_t ip_dst, ip_src;
+	uint8_t *hw_src;
+	bool ret = false;
+	struct neighbour *n = NULL;
+	struct hard_iface *primary_if = NULL;
+	struct sk_buff *skb_new;
+
+	type = arp_get_type(bat_priv, skb, 0);
+	/* If we get an ARP_REQUEST we have to send the unicast message to the
+	 * selected DHT candidates
+	 */
+	if (type != ARPOP_REQUEST)
+		goto out;
+
+	bat_dbg_arp(bat_priv, skb, type, 0, "Parsing outgoing ARP REQUEST");
+
+	ip_src = ARP_IP_SRC(skb, 0);
+	hw_src = ARP_HW_SRC(skb, 0);
+	ip_dst = ARP_IP_DST(skb, 0);
+
+	primary_if = primary_if_get_selected(bat_priv);
+	if (!primary_if)
+		goto out;
+
+	arp_neigh_update(bat_priv, ip_src, hw_src);
+
+	n = neigh_lookup(&arp_tbl, &ip_dst, primary_if->soft_iface);
+	/* check if it is a valid neigh entry */
+	if (n && (n->nud_state & NUD_CONNECTED)) {
+		skb_new = arp_create(ARPOP_REPLY, ETH_P_ARP, ip_src,
+				     primary_if->soft_iface, ip_dst, hw_src,
+				     n->ha, hw_src);
+		if (!skb_new)
+			goto out;
+
+		skb_reset_mac_header(skb_new);
+		skb_new->protocol = eth_type_trans(skb_new,
+						   primary_if->soft_iface);
+		bat_priv->stats.rx_packets++;
+		bat_priv->stats.rx_bytes += skb->len + ETH_HLEN;
+		primary_if->soft_iface->last_rx = jiffies;
+
+		netif_rx(skb_new);
+		bat_dbg(DBG_DAT, bat_priv, "ARP request replied locally\n");
+	} else
+		/* Send the request on the DHT */
+		ret = dht_send_data(bat_priv, skb, ip_dst, BAT_P_DAT_DHT_GET);
+out:
+	if (n)
+		neigh_release(n);
+	if (primary_if)
+		hardif_free_ref(primary_if);
+	return ret;
+}
+
+/* This function is meant to be invoked for an ARP request which is coming into
+ * the bat0 interfaces from the mesh network. It will check for the needed data
+ * into the local table. If found, an ARP reply is sent immediately, otherwise
+ * the caller has to deliver the ARP request to the upper layer
+ */
+bool dat_snoop_incoming_arp_request(struct bat_priv *bat_priv,
+				    struct sk_buff *skb, int hdr_size)
+{
+	uint16_t type;
+	uint32_t ip_src, ip_dst;
+	uint8_t *hw_src;
+	struct hard_iface *primary_if = NULL;
+	struct sk_buff *skb_new;
+	struct neighbour *n = NULL;
+	bool ret = false;
+
+	type = arp_get_type(bat_priv, skb, hdr_size);
+	if (type != ARPOP_REQUEST)
+		goto out;
+
+	hw_src = ARP_HW_SRC(skb, hdr_size);
+	ip_src = ARP_IP_SRC(skb, hdr_size);
+	ip_dst = ARP_IP_DST(skb, hdr_size);
+
+	bat_dbg_arp(bat_priv, skb, type, hdr_size,
+		    "Parsing incoming ARP REQUEST");
+
+	primary_if = primary_if_get_selected(bat_priv);
+	if (!primary_if)
+		goto out;
+
+	arp_neigh_update(bat_priv, ip_src, hw_src);
+
+	n = neigh_lookup(&arp_tbl, &ip_dst, primary_if->soft_iface);
+	/* check if it is a valid neigh entry */
+	if (!n || !(n->nud_state & NUD_CONNECTED))
+		goto out;
+
+	skb_new = arp_create(ARPOP_REPLY, ETH_P_ARP, ip_src,
+			     primary_if->soft_iface, ip_dst, hw_src, n->ha,
+			     hw_src);
+
+	if (!skb_new)
+		goto out;
+
+	unicast_4addr_send_skb(skb_new, bat_priv, BAT_P_DAT_CACHE_REPLY);
+
+	ret = true;
+out:
+	if (n)
+		neigh_release(n);
+	if (primary_if)
+		hardif_free_ref(primary_if);
+	if (ret)
+		kfree_skb(skb);
+	return ret;
+}
+
+/* This function is meant to be invoked on an ARP reply packet going into the
+ * soft interface. The related neighbour entry has to be updated and the DHT has
+ * to be populated as well
+ */
+bool dat_snoop_outgoing_arp_reply(struct bat_priv *bat_priv,
+				  struct sk_buff *skb)
+{
+	uint16_t type;
+	uint32_t ip_src, ip_dst;
+	uint8_t *hw_src, *hw_dst;
+	bool ret = false;
+
+	type = arp_get_type(bat_priv, skb, 0);
+	if (type != ARPOP_REPLY)
+		goto out;
+
+	bat_dbg_arp(bat_priv, skb, type, 0, "Parsing outgoing ARP REPLY");
+
+	hw_src = ARP_HW_SRC(skb, 0);
+	ip_src = ARP_IP_SRC(skb, 0);
+	hw_dst = ARP_HW_DST(skb, 0);
+	ip_dst = ARP_IP_DST(skb, 0);
+
+	arp_neigh_update(bat_priv, ip_src, hw_src);
+	arp_neigh_update(bat_priv, ip_dst, hw_dst);
+
+	/* Send the ARP reply to the candidates for both the IP addresses we
+	 * fetched from the ARP reply
+	 */
+	dht_send_data(bat_priv, skb, ip_src, BAT_P_DAT_DHT_PUT);
+	dht_send_data(bat_priv, skb, ip_dst, BAT_P_DAT_DHT_PUT);
+	ret = true;
+out:
+	return ret;
+}
+
+/* This function has to be invoked on an ARP reply coming into the soft
+ * interface from the mesh network. The local table has to be updated
+ */
+bool dat_snoop_incoming_arp_reply(struct bat_priv *bat_priv,
+				  struct sk_buff *skb, int hdr_size)
+{
+	uint16_t type;
+	uint32_t ip_src, ip_dst;
+	uint8_t *hw_src, *hw_dst;
+	bool ret = false;
+
+	type = arp_get_type(bat_priv, skb, hdr_size);
+	if (type != ARPOP_REPLY)
+		goto out;
+
+	bat_dbg_arp(bat_priv, skb, type, hdr_size,
+		    "Parsing incoming ARP REPLY");
+
+	hw_src = ARP_HW_SRC(skb, hdr_size);
+	ip_src = ARP_IP_SRC(skb, hdr_size);
+	hw_dst = ARP_HW_DST(skb, hdr_size);
+	ip_dst = ARP_IP_DST(skb, hdr_size);
+
+	/* Update our internal cache with both the IP addresses we fetched from
+	 * the ARP reply
+	 */
+	arp_neigh_update(bat_priv, ip_src, hw_src);
+	arp_neigh_update(bat_priv, ip_dst, hw_dst);
+
+	/* if this REPLY is directed to a client of mine, let's deliver the
+	 * packet to the interface
+	 */
+	ret = !is_my_client(bat_priv, hw_dst);
+out:
+	/* if ret == false packet has to be delivered to the interface */
+	return ret;
+}
+
+bool dat_drop_broadcast_packet(struct bat_priv *bat_priv,
+			       struct forw_packet *forw_packet)
+{
+	struct neighbour *n;
+
+	/* If this packet is an ARP_REQUEST and we already have the information
+	 * that it is going to ask, we can drop the packet
+	 */
+	if (!forw_packet->num_packets &&
+	    (ARPOP_REQUEST == arp_get_type(bat_priv, forw_packet->skb,
+					   sizeof(struct bcast_packet)))) {
+		n = neigh_lookup(&arp_tbl,
+				 &ARP_IP_DST(forw_packet->skb,
+					     sizeof(struct bcast_packet)),
+				 forw_packet->if_incoming->soft_iface);
+		/* check if we already know this neigh */
+		if (n && (n->nud_state & NUD_CONNECTED)) {
+			bat_dbg(DBG_DAT, bat_priv,
+				"ARP Request for %pI4: fallback prevented\n",
+				&ARP_IP_DST(forw_packet->skb,
+					    sizeof(struct bcast_packet)));
+			return true;
+		}
+
+		bat_dbg(DBG_DAT, bat_priv, "ARP Request for %pI4: fallback\n",
+			&ARP_IP_DST(forw_packet->skb,
+				    sizeof(struct bcast_packet)));
+	}
+	return false;
+}
diff --git a/net/batman-adv/distributed-arp-table.h b/net/batman-adv/distributed-arp-table.h
index cdd0484..8b7aa87 100644
--- a/net/batman-adv/distributed-arp-table.h
+++ b/net/batman-adv/distributed-arp-table.h
@@ -37,6 +37,17 @@
 #define ARP_IP_DST(skb, hdr_size) (*(uint32_t *)(ARP_HW_SRC(skb, hdr_size) + \
 				   ETH_ALEN * 2 + 4))
 
+bool dat_snoop_outgoing_arp_request(struct bat_priv *bat_priv,
+				    struct sk_buff *skb);
+bool dat_snoop_incoming_arp_request(struct bat_priv *bat_priv,
+				    struct sk_buff *skb, int hdr_size);
+bool dat_snoop_outgoing_arp_reply(struct bat_priv *bat_priv,
+				  struct sk_buff *skb);
+bool dat_snoop_incoming_arp_reply(struct bat_priv *bat_priv,
+				  struct sk_buff *skb, int hdr_size);
+bool dat_drop_broadcast_packet(struct bat_priv *bat_priv,
+			       struct forw_packet *forw_packet);
+
 /* hash function to choose an entry in a hash table of given size.
  * hash algorithm from http://en.wikipedia.org/wiki/Hash_table
  */
diff --git a/net/batman-adv/main.h b/net/batman-adv/main.h
index 395e59d..57023c3 100644
--- a/net/batman-adv/main.h
+++ b/net/batman-adv/main.h
@@ -67,6 +67,8 @@
 
 #define NUM_WORDS BITS_TO_LONGS(TQ_LOCAL_WINDOW_SIZE)
 
+/* msecs after which an ARP_REQUEST is sent in broadcast as fallback */
+#define ARP_REQ_DELAY 250
 /* numbers of originator to contact for any PUT/GET DHT operation */
 #define DHT_CANDIDATES_NUM 3
 
diff --git a/net/batman-adv/send.c b/net/batman-adv/send.c
index 7c66b61..91eaa45 100644
--- a/net/batman-adv/send.c
+++ b/net/batman-adv/send.c
@@ -20,6 +20,7 @@
  */
 
 #include "main.h"
+#include "distributed-arp-table.h"
 #include "send.h"
 #include "routing.h"
 #include "translation-table.h"
@@ -274,6 +275,9 @@ static void send_outstanding_bcast_packet(struct work_struct *work)
 	if (atomic_read(&bat_priv->mesh_state) == MESH_DEACTIVATING)
 		goto out;
 
+	if (dat_drop_broadcast_packet(bat_priv, forw_packet))
+		goto out;
+
 	/* rebroadcast packet */
 	rcu_read_lock();
 	list_for_each_entry_rcu(hard_iface, &hardif_list, list) {
diff --git a/net/batman-adv/soft-interface.c b/net/batman-adv/soft-interface.c
index 6e2530b..3a1483a 100644
--- a/net/batman-adv/soft-interface.c
+++ b/net/batman-adv/soft-interface.c
@@ -22,6 +22,7 @@
 #include "main.h"
 #include "soft-interface.h"
 #include "hard-interface.h"
+#include "distributed-arp-table.h"
 #include "routing.h"
 #include "send.h"
 #include "bat_debugfs.h"
@@ -136,6 +137,7 @@ static int interface_tx(struct sk_buff *skb, struct net_device *soft_iface)
 	int data_len = skb->len, ret;
 	short vid __maybe_unused = -1;
 	bool do_bcast = false;
+	unsigned long brd_delay = 1;
 
 	if (atomic_read(&bat_priv->mesh_state) != MESH_ACTIVE)
 		goto dropped;
@@ -197,6 +199,9 @@ static int interface_tx(struct sk_buff *skb, struct net_device *soft_iface)
 		if (!primary_if)
 			goto dropped;
 
+		if (dat_snoop_outgoing_arp_request(bat_priv, skb))
+			brd_delay = msecs_to_jiffies(ARP_REQ_DELAY);
+
 		if (my_skb_head_push(skb, sizeof(*bcast_packet)) < 0)
 			goto dropped;
 
@@ -216,7 +221,7 @@ static int interface_tx(struct sk_buff *skb, struct net_device *soft_iface)
 		bcast_packet->seqno =
 			htonl(atomic_inc_return(&bat_priv->bcast_seqno));
 
-		add_bcast_packet_to_list(bat_priv, skb, 1);
+		add_bcast_packet_to_list(bat_priv, skb, brd_delay);
 
 		/* a copy is stored in the bcast list, therefore removing
 		 * the original skb. */
@@ -230,6 +235,8 @@ static int interface_tx(struct sk_buff *skb, struct net_device *soft_iface)
 				goto dropped;
 		}
 
+		dat_snoop_outgoing_arp_reply(bat_priv, skb);
+
 		ret = unicast_send_skb(skb, bat_priv);
 		if (ret != 0)
 			goto dropped_freed;
@@ -262,6 +269,12 @@ void interface_rx(struct net_device *soft_iface,
 	if (!pskb_may_pull(skb, hdr_size))
 		goto dropped;
 
+	if (dat_snoop_incoming_arp_request(bat_priv, skb, hdr_size))
+		goto out;
+
+	if (dat_snoop_incoming_arp_reply(bat_priv, skb, hdr_size))
+		goto out;
+
 	skb_pull_rcsum(skb, hdr_size);
 	skb_reset_mac_header(skb);
 
-- 
1.7.9.4

[PATCH 07/15] batman-adv: Distributed ARP Table - increase default soft_iface ARP table timeout

[Antonio Quartulli]

The default timeout value for ARP entries belonging to any soft_iface
ARP table has been incremented by a factor 4. This is necessary because the DHT
will store several network entries in the soft_iface ARP table.

Signed-off-by: Antonio Quartulli <ordex@autistici.org>
---
 net/batman-adv/distributed-arp-table.c |   20 ++++++++++++++++++++
 net/batman-adv/distributed-arp-table.h |    1 +
 net/batman-adv/main.h                  |    4 ++++
 net/batman-adv/soft-interface.c        |    2 ++
 4 files changed, 27 insertions(+)

diff --git a/net/batman-adv/distributed-arp-table.c b/net/batman-adv/distributed-arp-table.c
index f3b63ef..24b17131 100644
--- a/net/batman-adv/distributed-arp-table.c
+++ b/net/batman-adv/distributed-arp-table.c
@@ -23,6 +23,7 @@
 #include <linux/if_arp.h>
 /* needed to use arp_tbl */
 #include <net/arp.h>
+#include <linux/inetdevice.h>
 
 #include "main.h"
 #include "distributed-arp-table.h"
@@ -583,3 +584,22 @@ bool dat_drop_broadcast_packet(struct bat_priv *bat_priv,
 	}
 	return false;
 }
+
+void arp_change_timeout(struct net_device *soft_iface, const char *name)
+{
+	struct in_device *in_dev = in_dev_get(soft_iface);
+	if (!in_dev) {
+		pr_err("Unable to set ARP parameters for the batman interface '%s'\n",
+		       name);
+		return;
+	}
+
+	/* Introduce a delay in the ARP state-machine transactions. Entries
+	 * will be kept in the ARP table for the default time multiplied by 4
+	 */
+	in_dev->arp_parms->base_reachable_time *= ARP_TIMEOUT_FACTOR;
+	in_dev->arp_parms->gc_staletime *= ARP_TIMEOUT_FACTOR;
+	in_dev->arp_parms->reachable_time *= ARP_TIMEOUT_FACTOR;
+
+	in_dev_put(in_dev);
+}
diff --git a/net/batman-adv/distributed-arp-table.h b/net/batman-adv/distributed-arp-table.h
index 8b7aa87..3a4f710 100644
--- a/net/batman-adv/distributed-arp-table.h
+++ b/net/batman-adv/distributed-arp-table.h
@@ -47,6 +47,7 @@ bool dat_snoop_incoming_arp_reply(struct bat_priv *bat_priv,
 				  struct sk_buff *skb, int hdr_size);
 bool dat_drop_broadcast_packet(struct bat_priv *bat_priv,
 			       struct forw_packet *forw_packet);
+void arp_change_timeout(struct net_device *soft_iface, const char *name);
 
 /* hash function to choose an entry in a hash table of given size.
  * hash algorithm from http://en.wikipedia.org/wiki/Hash_table
diff --git a/net/batman-adv/main.h b/net/batman-adv/main.h
index 57023c3..c11c8df 100644
--- a/net/batman-adv/main.h
+++ b/net/batman-adv/main.h
@@ -71,6 +71,10 @@
 #define ARP_REQ_DELAY 250
 /* numbers of originator to contact for any PUT/GET DHT operation */
 #define DHT_CANDIDATES_NUM 3
+/* Factor which default ARP timeout values of the soft_iface table are
+ * multiplied by
+ */
+#define ARP_TIMEOUT_FACTOR 4
 
 #define LOG_BUF_LEN 8192	  /* has to be a power of 2 */
 
diff --git a/net/batman-adv/soft-interface.c b/net/batman-adv/soft-interface.c
index 3a1483a..b56dafd 100644
--- a/net/batman-adv/soft-interface.c
+++ b/net/batman-adv/soft-interface.c
@@ -381,6 +381,8 @@ struct net_device *softif_create(const char *name)
 		goto free_soft_iface;
 	}
 
+	arp_change_timeout(soft_iface, name);
+
 	bat_priv = netdev_priv(soft_iface);
 
 	atomic_set(&bat_priv->aggregated_ogms, 1);
-- 
1.7.9.4

[B.A.T.M.A.N.] [PATCH 07/15] batman-adv: Distributed ARP Table - increase default soft_iface ARP table timeout

[Antonio Quartulli]

The default timeout value for ARP entries belonging to any soft_iface
ARP table has been incremented by a factor 4. This is necessary because the DHT
will store several network entries in the soft_iface ARP table.

Signed-off-by: Antonio Quartulli <ordex@autistici.org>
---
 net/batman-adv/distributed-arp-table.c |   20 ++++++++++++++++++++
 net/batman-adv/distributed-arp-table.h |    1 +
 net/batman-adv/main.h                  |    4 ++++
 net/batman-adv/soft-interface.c        |    2 ++
 4 files changed, 27 insertions(+)

diff --git a/net/batman-adv/distributed-arp-table.c b/net/batman-adv/distributed-arp-table.c
index f3b63ef..24b17131 100644
--- a/net/batman-adv/distributed-arp-table.c
+++ b/net/batman-adv/distributed-arp-table.c
@@ -23,6 +23,7 @@
 #include <linux/if_arp.h>
 /* needed to use arp_tbl */
 #include <net/arp.h>
+#include <linux/inetdevice.h>
 
 #include "main.h"
 #include "distributed-arp-table.h"
@@ -583,3 +584,22 @@ bool dat_drop_broadcast_packet(struct bat_priv *bat_priv,
 	}
 	return false;
 }
+
+void arp_change_timeout(struct net_device *soft_iface, const char *name)
+{
+	struct in_device *in_dev = in_dev_get(soft_iface);
+	if (!in_dev) {
+		pr_err("Unable to set ARP parameters for the batman interface '%s'\n",
+		       name);
+		return;
+	}
+
+	/* Introduce a delay in the ARP state-machine transactions. Entries
+	 * will be kept in the ARP table for the default time multiplied by 4
+	 */
+	in_dev->arp_parms->base_reachable_time *= ARP_TIMEOUT_FACTOR;
+	in_dev->arp_parms->gc_staletime *= ARP_TIMEOUT_FACTOR;
+	in_dev->arp_parms->reachable_time *= ARP_TIMEOUT_FACTOR;
+
+	in_dev_put(in_dev);
+}
diff --git a/net/batman-adv/distributed-arp-table.h b/net/batman-adv/distributed-arp-table.h
index 8b7aa87..3a4f710 100644
--- a/net/batman-adv/distributed-arp-table.h
+++ b/net/batman-adv/distributed-arp-table.h
@@ -47,6 +47,7 @@ bool dat_snoop_incoming_arp_reply(struct bat_priv *bat_priv,
 				  struct sk_buff *skb, int hdr_size);
 bool dat_drop_broadcast_packet(struct bat_priv *bat_priv,
 			       struct forw_packet *forw_packet);
+void arp_change_timeout(struct net_device *soft_iface, const char *name);
 
 /* hash function to choose an entry in a hash table of given size.
  * hash algorithm from http://en.wikipedia.org/wiki/Hash_table
diff --git a/net/batman-adv/main.h b/net/batman-adv/main.h
index 57023c3..c11c8df 100644
--- a/net/batman-adv/main.h
+++ b/net/batman-adv/main.h
@@ -71,6 +71,10 @@
 #define ARP_REQ_DELAY 250
 /* numbers of originator to contact for any PUT/GET DHT operation */
 #define DHT_CANDIDATES_NUM 3
+/* Factor which default ARP timeout values of the soft_iface table are
+ * multiplied by
+ */
+#define ARP_TIMEOUT_FACTOR 4
 
 #define LOG_BUF_LEN 8192	  /* has to be a power of 2 */
 
diff --git a/net/batman-adv/soft-interface.c b/net/batman-adv/soft-interface.c
index 3a1483a..b56dafd 100644
--- a/net/batman-adv/soft-interface.c
+++ b/net/batman-adv/soft-interface.c
@@ -381,6 +381,8 @@ struct net_device *softif_create(const char *name)
 		goto free_soft_iface;
 	}
 
+	arp_change_timeout(soft_iface, name);
+
 	bat_priv = netdev_priv(soft_iface);
 
 	atomic_set(&bat_priv->aggregated_ogms, 1);
-- 
1.7.9.4

[PATCH 08/15] batman-adv: Distributed ARP Table - add compile option

[Antonio Quartulli]

This patch makes it possible to decide whether to include DAT within the
batman-adv binary or not.
It is extremely useful when the user wants to reduce the size of the resulting
module by cutting off any not needed feature.

Signed-off-by: Antonio Quartulli <ordex@autistici.org>
---
 net/batman-adv/Kconfig                 |   12 +++++++-
 net/batman-adv/Makefile                |    2 +-
 net/batman-adv/distributed-arp-table.h |   52 ++++++++++++++++++++++++++++++++
 net/batman-adv/types.h                 |    8 +++++
 4 files changed, 72 insertions(+), 2 deletions(-)

diff --git a/net/batman-adv/Kconfig b/net/batman-adv/Kconfig
index b25e20f..250e0b5 100644
--- a/net/batman-adv/Kconfig
+++ b/net/batman-adv/Kconfig
@@ -4,7 +4,7 @@
 
 config BATMAN_ADV
 	tristate "B.A.T.M.A.N. Advanced Meshing Protocol"
-	depends on NET && INET
+	depends on NET
 	select CRC16
         default n
 	help
@@ -25,6 +25,16 @@ config BATMAN_ADV_BLA
 	  more than one mesh node in the same LAN, you can safely remove
 	  this feature and save some space.
 
+config BATMAN_ADV_DAT
+	bool "Distributed ARP Table"
+	depends on BATMAN_ADV && INET
+	default n
+	help
+	  This option enables DAT (Distributed ARP Table), a DHT based
+	  mechanism that increases ARP reliability on sparse wireless
+	  mesh networks. If you think that your network does not need
+	  this option you can safely remove it and save some space.
+
 config BATMAN_ADV_DEBUG
 	bool "B.A.T.M.A.N. debugging"
 	depends on BATMAN_ADV
diff --git a/net/batman-adv/Makefile b/net/batman-adv/Makefile
index 84955b3..ad002cd 100644
--- a/net/batman-adv/Makefile
+++ b/net/batman-adv/Makefile
@@ -24,7 +24,7 @@ batman-adv-y += bat_iv_ogm.o
 batman-adv-y += bat_sysfs.o
 batman-adv-y += bitarray.o
 batman-adv-$(CONFIG_BATMAN_ADV_BLA) += bridge_loop_avoidance.o
-batman-adv-y += distributed-arp-table.o
+batman-adv-$(CONFIG_BATMAN_ADV_DAT) += distributed-arp-table.o
 batman-adv-y += gateway_client.o
 batman-adv-y += gateway_common.o
 batman-adv-y += hard-interface.o
diff --git a/net/batman-adv/distributed-arp-table.h b/net/batman-adv/distributed-arp-table.h
index 3a4f710..e55067b 100644
--- a/net/batman-adv/distributed-arp-table.h
+++ b/net/batman-adv/distributed-arp-table.h
@@ -22,6 +22,8 @@
 #ifndef _NET_BATMAN_ADV_ARP_H_
 #define _NET_BATMAN_ADV_ARP_H_
 
+#ifdef CONFIG_BATMAN_ADV_DAT
+
 #include "types.h"
 #include "originator.h"
 
@@ -85,4 +87,54 @@ static inline void dat_init_own_dht_addr(struct bat_priv *bat_priv,
 					    DAT_ADDR_MAX);
 }
 
+#else
+
+static inline bool dat_snoop_outgoing_arp_request(struct bat_priv *bat_priv,
+						  struct sk_buff *skb)
+{
+	return false;
+}
+
+static inline bool dat_snoop_incoming_arp_request(struct bat_priv *bat_priv,
+						  struct sk_buff *skb,
+						  int hdr_size)
+{
+	return false;
+}
+
+static inline bool dat_snoop_outgoing_arp_reply(struct bat_priv *bat_priv,
+						struct sk_buff *skb)
+{
+	return false;
+}
+
+static inline bool dat_snoop_incoming_arp_reply(struct bat_priv *bat_priv,
+						struct sk_buff *skb,
+						int hdr_size)
+{
+	return false;
+}
+
+static inline bool dat_drop_broadcast_packet(struct bat_priv *bat_priv,
+					     struct forw_packet *forw_packet)
+{
+	return false;
+}
+
+static inline void dat_init_orig_node_dht_addr(struct orig_node *orig_node)
+{
+}
+
+static inline void dat_init_own_dht_addr(struct bat_priv *bat_priv,
+					 struct hard_iface *primary_if)
+{
+}
+
+static inline void arp_change_timeout(struct net_device *soft_iface,
+				      const char *name)
+{
+}
+
+#endif /* CONFIG_BATMAN_ADV_DAT */
+
 #endif /* _NET_BATMAN_ADV_ARP_H_ */
diff --git a/net/batman-adv/types.h b/net/batman-adv/types.h
index 61e2d78..2b8ffcb 100644
--- a/net/batman-adv/types.h
+++ b/net/batman-adv/types.h
@@ -27,6 +27,8 @@
 #include "packet.h"
 #include "bitarray.h"
 
+#ifdef CONFIG_BATMAN_ADV_DAT
+
 /* dat_addr_t is the type used for all DHT addresses. If it is changed,
  * DAT_ADDR_MAX is changed as well.
  *
@@ -34,6 +36,8 @@
  */
 #define dat_addr_t uint16_t
 
+#endif /* CONFIG_BATMAN_ADV_DAT */
+
 #define BAT_HEADER_LEN (ETH_HLEN + \
 	((sizeof(struct unicast_packet) > sizeof(struct bcast_packet) ? \
 	 sizeof(struct unicast_packet) : \
@@ -74,7 +78,9 @@ struct hard_iface {
 struct orig_node {
 	uint8_t orig[ETH_ALEN];
 	uint8_t primary_addr[ETH_ALEN];
+#ifdef CONFIG_BATMAN_ADV_DAT
 	dat_addr_t dht_addr;
+#endif
 	struct neigh_node __rcu *router; /* rcu protected pointer */
 	unsigned long *bcast_own;
 	uint8_t *bcast_own_sum;
@@ -229,7 +235,9 @@ struct bat_priv {
 	struct gw_node __rcu *curr_gw;  /* rcu protected pointer */
 	atomic_t gw_reselect;
 	struct hard_iface __rcu *primary_if;  /* rcu protected pointer */
+#ifdef CONFIG_BATMAN_ADV_DAT
 	dat_addr_t dht_addr;
+#endif
 	struct vis_info *my_vis_info;
 	struct bat_algo_ops *bat_algo_ops;
 };
-- 
1.7.9.4

[B.A.T.M.A.N.] [PATCH 08/15] batman-adv: Distributed ARP Table - add compile option

[Antonio Quartulli]

This patch makes it possible to decide whether to include DAT within the
batman-adv binary or not.
It is extremely useful when the user wants to reduce the size of the resulting
module by cutting off any not needed feature.

Signed-off-by: Antonio Quartulli <ordex@autistici.org>
---
 net/batman-adv/Kconfig                 |   12 +++++++-
 net/batman-adv/Makefile                |    2 +-
 net/batman-adv/distributed-arp-table.h |   52 ++++++++++++++++++++++++++++++++
 net/batman-adv/types.h                 |    8 +++++
 4 files changed, 72 insertions(+), 2 deletions(-)

diff --git a/net/batman-adv/Kconfig b/net/batman-adv/Kconfig
index b25e20f..250e0b5 100644
--- a/net/batman-adv/Kconfig
+++ b/net/batman-adv/Kconfig
@@ -4,7 +4,7 @@
 
 config BATMAN_ADV
 	tristate "B.A.T.M.A.N. Advanced Meshing Protocol"
-	depends on NET && INET
+	depends on NET
 	select CRC16
         default n
 	help
@@ -25,6 +25,16 @@ config BATMAN_ADV_BLA
 	  more than one mesh node in the same LAN, you can safely remove
 	  this feature and save some space.
 
+config BATMAN_ADV_DAT
+	bool "Distributed ARP Table"
+	depends on BATMAN_ADV && INET
+	default n
+	help
+	  This option enables DAT (Distributed ARP Table), a DHT based
+	  mechanism that increases ARP reliability on sparse wireless
+	  mesh networks. If you think that your network does not need
+	  this option you can safely remove it and save some space.
+
 config BATMAN_ADV_DEBUG
 	bool "B.A.T.M.A.N. debugging"
 	depends on BATMAN_ADV
diff --git a/net/batman-adv/Makefile b/net/batman-adv/Makefile
index 84955b3..ad002cd 100644
--- a/net/batman-adv/Makefile
+++ b/net/batman-adv/Makefile
@@ -24,7 +24,7 @@ batman-adv-y += bat_iv_ogm.o
 batman-adv-y += bat_sysfs.o
 batman-adv-y += bitarray.o
 batman-adv-$(CONFIG_BATMAN_ADV_BLA) += bridge_loop_avoidance.o
-batman-adv-y += distributed-arp-table.o
+batman-adv-$(CONFIG_BATMAN_ADV_DAT) += distributed-arp-table.o
 batman-adv-y += gateway_client.o
 batman-adv-y += gateway_common.o
 batman-adv-y += hard-interface.o
diff --git a/net/batman-adv/distributed-arp-table.h b/net/batman-adv/distributed-arp-table.h
index 3a4f710..e55067b 100644
--- a/net/batman-adv/distributed-arp-table.h
+++ b/net/batman-adv/distributed-arp-table.h
@@ -22,6 +22,8 @@
 #ifndef _NET_BATMAN_ADV_ARP_H_
 #define _NET_BATMAN_ADV_ARP_H_
 
+#ifdef CONFIG_BATMAN_ADV_DAT
+
 #include "types.h"
 #include "originator.h"
 
@@ -85,4 +87,54 @@ static inline void dat_init_own_dht_addr(struct bat_priv *bat_priv,
 					    DAT_ADDR_MAX);
 }
 
+#else
+
+static inline bool dat_snoop_outgoing_arp_request(struct bat_priv *bat_priv,
+						  struct sk_buff *skb)
+{
+	return false;
+}
+
+static inline bool dat_snoop_incoming_arp_request(struct bat_priv *bat_priv,
+						  struct sk_buff *skb,
+						  int hdr_size)
+{
+	return false;
+}
+
+static inline bool dat_snoop_outgoing_arp_reply(struct bat_priv *bat_priv,
+						struct sk_buff *skb)
+{
+	return false;
+}
+
+static inline bool dat_snoop_incoming_arp_reply(struct bat_priv *bat_priv,
+						struct sk_buff *skb,
+						int hdr_size)
+{
+	return false;
+}
+
+static inline bool dat_drop_broadcast_packet(struct bat_priv *bat_priv,
+					     struct forw_packet *forw_packet)
+{
+	return false;
+}
+
+static inline void dat_init_orig_node_dht_addr(struct orig_node *orig_node)
+{
+}
+
+static inline void dat_init_own_dht_addr(struct bat_priv *bat_priv,
+					 struct hard_iface *primary_if)
+{
+}
+
+static inline void arp_change_timeout(struct net_device *soft_iface,
+				      const char *name)
+{
+}
+
+#endif /* CONFIG_BATMAN_ADV_DAT */
+
 #endif /* _NET_BATMAN_ADV_ARP_H_ */
diff --git a/net/batman-adv/types.h b/net/batman-adv/types.h
index 61e2d78..2b8ffcb 100644
--- a/net/batman-adv/types.h
+++ b/net/batman-adv/types.h
@@ -27,6 +27,8 @@
 #include "packet.h"
 #include "bitarray.h"
 
+#ifdef CONFIG_BATMAN_ADV_DAT
+
 /* dat_addr_t is the type used for all DHT addresses. If it is changed,
  * DAT_ADDR_MAX is changed as well.
  *
@@ -34,6 +36,8 @@
  */
 #define dat_addr_t uint16_t
 
+#endif /* CONFIG_BATMAN_ADV_DAT */
+
 #define BAT_HEADER_LEN (ETH_HLEN + \
 	((sizeof(struct unicast_packet) > sizeof(struct bcast_packet) ? \
 	 sizeof(struct unicast_packet) : \
@@ -74,7 +78,9 @@ struct hard_iface {
 struct orig_node {
 	uint8_t orig[ETH_ALEN];
 	uint8_t primary_addr[ETH_ALEN];
+#ifdef CONFIG_BATMAN_ADV_DAT
 	dat_addr_t dht_addr;
+#endif
 	struct neigh_node __rcu *router; /* rcu protected pointer */
 	unsigned long *bcast_own;
 	uint8_t *bcast_own_sum;
@@ -229,7 +235,9 @@ struct bat_priv {
 	struct gw_node __rcu *curr_gw;  /* rcu protected pointer */
 	atomic_t gw_reselect;
 	struct hard_iface __rcu *primary_if;  /* rcu protected pointer */
+#ifdef CONFIG_BATMAN_ADV_DAT
 	dat_addr_t dht_addr;
+#endif
 	struct vis_info *my_vis_info;
 	struct bat_algo_ops *bat_algo_ops;
 };
-- 
1.7.9.4

[PATCH 09/15] batman-adv: fix wrong dhcp option list browsing

[Antonio Quartulli]

In is_type_dhcprequest(), while parsing a DHCP message, if the entry we found in
the option list is neither a padding nor the dhcp-type, we have to ignore it and
jump as many bytes as its length + 1. The "+ 1" byte is given by the subtype
field itself that has to be jumped too.

Reported-by: Marek Lindner <lindner_marek@yahoo.de>
Signed-off-by: Antonio Quartulli <ordex@autistici.org>
---
 net/batman-adv/gateway_client.c |    6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/net/batman-adv/gateway_client.c b/net/batman-adv/gateway_client.c
index 6f9b9b7..47f7186 100644
--- a/net/batman-adv/gateway_client.c
+++ b/net/batman-adv/gateway_client.c
@@ -558,10 +558,10 @@ static bool is_type_dhcprequest(struct sk_buff *skb, int header_len)
 			p++;
 
 			/* ...and then we jump over the data */
-			if (pkt_len < *p)
+			if (pkt_len < 1 + (*p))
 				goto out;
-			pkt_len -= *p;
-			p += (*p);
+			pkt_len -= 1 + (*p);
+			p += 1 + (*p);
 		}
 	}
 out:
-- 
1.7.9.4

[B.A.T.M.A.N.] [PATCH 09/15] batman-adv: fix wrong dhcp option list browsing

[Antonio Quartulli]

In is_type_dhcprequest(), while parsing a DHCP message, if the entry we found in
the option list is neither a padding nor the dhcp-type, we have to ignore it and
jump as many bytes as its length + 1. The "+ 1" byte is given by the subtype
field itself that has to be jumped too.

Reported-by: Marek Lindner <lindner_marek@yahoo.de>
Signed-off-by: Antonio Quartulli <ordex@autistici.org>
---
 net/batman-adv/gateway_client.c |    6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/net/batman-adv/gateway_client.c b/net/batman-adv/gateway_client.c
index 6f9b9b7..47f7186 100644
--- a/net/batman-adv/gateway_client.c
+++ b/net/batman-adv/gateway_client.c
@@ -558,10 +558,10 @@ static bool is_type_dhcprequest(struct sk_buff *skb, int header_len)
 			p++;
 
 			/* ...and then we jump over the data */
-			if (pkt_len < *p)
+			if (pkt_len < 1 + (*p))
 				goto out;
-			pkt_len -= *p;
-			p += (*p);
+			pkt_len -= 1 + (*p);
+			p += 1 + (*p);
 		}
 	}
 out:
-- 
1.7.9.4

[PATCH 10/15] batman-adv: introduce is_single_hop_neigh variable to increase readability

[Antonio Quartulli]

From: Marek Lindner <lindner_marek@yahoo.de>

Signed-off-by: Marek Lindner <lindner_marek@yahoo.de>
Acked-by: Simon Wunderlich <siwu@hrz.tu-chemnitz.de>
Signed-off-by: Antonio Quartulli <ordex@autistici.org>
---
 net/batman-adv/bat_iv_ogm.c |   16 +++++++++-------
 1 file changed, 9 insertions(+), 7 deletions(-)

diff --git a/net/batman-adv/bat_iv_ogm.c b/net/batman-adv/bat_iv_ogm.c
index 8b2db2e..cd8f473 100644
--- a/net/batman-adv/bat_iv_ogm.c
+++ b/net/batman-adv/bat_iv_ogm.c
@@ -480,7 +480,8 @@ static void bat_iv_ogm_queue_add(struct bat_priv *bat_priv,
 static void bat_iv_ogm_forward(struct orig_node *orig_node,
 			       const struct ethhdr *ethhdr,
 			       struct batman_ogm_packet *batman_ogm_packet,
-			       int directlink, struct hard_iface *if_incoming)
+			       bool is_single_hop_neigh,
+			       struct hard_iface *if_incoming)
 {
 	struct bat_priv *bat_priv = netdev_priv(if_incoming->soft_iface);
 	struct neigh_node *router;
@@ -533,7 +534,7 @@ static void bat_iv_ogm_forward(struct orig_node *orig_node,
 
 	/* switch of primaries first hop flag when forwarding */
 	batman_ogm_packet->flags &= ~PRIMARIES_FIRST_HOP;
-	if (directlink)
+	if (is_single_hop_neigh)
 		batman_ogm_packet->flags |= DIRECTLINK;
 	else
 		batman_ogm_packet->flags &= ~DIRECTLINK;
@@ -918,7 +919,8 @@ static void bat_iv_ogm_process(const struct ethhdr *ethhdr,
 	struct neigh_node *orig_neigh_router = NULL;
 	int has_directlink_flag;
 	int is_my_addr = 0, is_my_orig = 0, is_my_oldorig = 0;
-	int is_broadcast = 0, is_bidirectional, is_single_hop_neigh;
+	int is_broadcast = 0, is_bidirectional;
+	bool is_single_hop_neigh = false;
 	int is_duplicate;
 	uint32_t if_incoming_seqno;
 
@@ -942,8 +944,8 @@ static void bat_iv_ogm_process(const struct ethhdr *ethhdr,
 
 	has_directlink_flag = (batman_ogm_packet->flags & DIRECTLINK ? 1 : 0);
 
-	is_single_hop_neigh = (compare_eth(ethhdr->h_source,
-					   batman_ogm_packet->orig) ? 1 : 0);
+	if (compare_eth(ethhdr->h_source, batman_ogm_packet->orig))
+		is_single_hop_neigh = true;
 
 	bat_dbg(DBG_BATMAN, bat_priv,
 		"Received BATMAN packet via NB: %pM, IF: %s [%pM] (from OG: %pM, via prev OG: %pM, seqno %u, ttvn %u, crc %u, changes %u, td %d, TTL %d, V %d, IDF %d)\n",
@@ -1114,7 +1116,7 @@ static void bat_iv_ogm_process(const struct ethhdr *ethhdr,
 
 		/* mark direct link on incoming interface */
 		bat_iv_ogm_forward(orig_node, ethhdr, batman_ogm_packet,
-				   1, if_incoming);
+				   is_single_hop_neigh, if_incoming);
 
 		bat_dbg(DBG_BATMAN, bat_priv,
 			"Forwarding packet: rebroadcast neighbor packet with direct link flag\n");
@@ -1137,7 +1139,7 @@ static void bat_iv_ogm_process(const struct ethhdr *ethhdr,
 	bat_dbg(DBG_BATMAN, bat_priv,
 		"Forwarding packet: rebroadcast originator packet\n");
 	bat_iv_ogm_forward(orig_node, ethhdr, batman_ogm_packet,
-			   0, if_incoming);
+			   is_single_hop_neigh, if_incoming);
 
 out_neigh:
 	if ((orig_neigh_node) && (!is_single_hop_neigh))
-- 
1.7.9.4

[B.A.T.M.A.N.] [PATCH 10/15] batman-adv: introduce is_single_hop_neigh variable to increase readability

[Antonio Quartulli]

From: Marek Lindner <lindner_marek@yahoo.de>

Signed-off-by: Marek Lindner <lindner_marek@yahoo.de>
Acked-by: Simon Wunderlich <siwu@hrz.tu-chemnitz.de>
Signed-off-by: Antonio Quartulli <ordex@autistici.org>
---
 net/batman-adv/bat_iv_ogm.c |   16 +++++++++-------
 1 file changed, 9 insertions(+), 7 deletions(-)

diff --git a/net/batman-adv/bat_iv_ogm.c b/net/batman-adv/bat_iv_ogm.c
index 8b2db2e..cd8f473 100644
--- a/net/batman-adv/bat_iv_ogm.c
+++ b/net/batman-adv/bat_iv_ogm.c
@@ -480,7 +480,8 @@ static void bat_iv_ogm_queue_add(struct bat_priv *bat_priv,
 static void bat_iv_ogm_forward(struct orig_node *orig_node,
 			       const struct ethhdr *ethhdr,
 			       struct batman_ogm_packet *batman_ogm_packet,
-			       int directlink, struct hard_iface *if_incoming)
+			       bool is_single_hop_neigh,
+			       struct hard_iface *if_incoming)
 {
 	struct bat_priv *bat_priv = netdev_priv(if_incoming->soft_iface);
 	struct neigh_node *router;
@@ -533,7 +534,7 @@ static void bat_iv_ogm_forward(struct orig_node *orig_node,
 
 	/* switch of primaries first hop flag when forwarding */
 	batman_ogm_packet->flags &= ~PRIMARIES_FIRST_HOP;
-	if (directlink)
+	if (is_single_hop_neigh)
 		batman_ogm_packet->flags |= DIRECTLINK;
 	else
 		batman_ogm_packet->flags &= ~DIRECTLINK;
@@ -918,7 +919,8 @@ static void bat_iv_ogm_process(const struct ethhdr *ethhdr,
 	struct neigh_node *orig_neigh_router = NULL;
 	int has_directlink_flag;
 	int is_my_addr = 0, is_my_orig = 0, is_my_oldorig = 0;
-	int is_broadcast = 0, is_bidirectional, is_single_hop_neigh;
+	int is_broadcast = 0, is_bidirectional;
+	bool is_single_hop_neigh = false;
 	int is_duplicate;
 	uint32_t if_incoming_seqno;
 
@@ -942,8 +944,8 @@ static void bat_iv_ogm_process(const struct ethhdr *ethhdr,
 
 	has_directlink_flag = (batman_ogm_packet->flags & DIRECTLINK ? 1 : 0);
 
-	is_single_hop_neigh = (compare_eth(ethhdr->h_source,
-					   batman_ogm_packet->orig) ? 1 : 0);
+	if (compare_eth(ethhdr->h_source, batman_ogm_packet->orig))
+		is_single_hop_neigh = true;
 
 	bat_dbg(DBG_BATMAN, bat_priv,
 		"Received BATMAN packet via NB: %pM, IF: %s [%pM] (from OG: %pM, via prev OG: %pM, seqno %u, ttvn %u, crc %u, changes %u, td %d, TTL %d, V %d, IDF %d)\n",
@@ -1114,7 +1116,7 @@ static void bat_iv_ogm_process(const struct ethhdr *ethhdr,
 
 		/* mark direct link on incoming interface */
 		bat_iv_ogm_forward(orig_node, ethhdr, batman_ogm_packet,
-				   1, if_incoming);
+				   is_single_hop_neigh, if_incoming);
 
 		bat_dbg(DBG_BATMAN, bat_priv,
 			"Forwarding packet: rebroadcast neighbor packet with direct link flag\n");
@@ -1137,7 +1139,7 @@ static void bat_iv_ogm_process(const struct ethhdr *ethhdr,
 	bat_dbg(DBG_BATMAN, bat_priv,
 		"Forwarding packet: rebroadcast originator packet\n");
 	bat_iv_ogm_forward(orig_node, ethhdr, batman_ogm_packet,
-			   0, if_incoming);
+			   is_single_hop_neigh, if_incoming);
 
 out_neigh:
 	if ((orig_neigh_node) && (!is_single_hop_neigh))
-- 
1.7.9.4

[PATCH 11/15] batman-adv: introduce packet type handler array for incoming packets

[Antonio Quartulli]

From: Marek Lindner <lindner_marek@yahoo.de>

The packet handler array replaces the growing switch statement, thus
dealing with incoming packets in a more efficient way. It also adds
to possibility to register packet handlers on the fly.

Signed-off-by: Marek Lindner <lindner_marek@yahoo.de>
Acked-by: Simon Wunderlich <siwu@hrz.tu-chemnitz.de>
Signed-off-by: Antonio Quartulli <ordex@autistici.org>
---
 net/batman-adv/hard-interface.c |  114 ------------------------------------
 net/batman-adv/main.c           |  123 +++++++++++++++++++++++++++++++++++++++
 net/batman-adv/main.h           |    6 ++
 3 files changed, 129 insertions(+), 114 deletions(-)

diff --git a/net/batman-adv/hard-interface.c b/net/batman-adv/hard-interface.c
index 1393939..b49bf0d 100644
--- a/net/batman-adv/hard-interface.c
+++ b/net/batman-adv/hard-interface.c
@@ -33,12 +33,6 @@
 
 #include <linux/if_arp.h>
 
-
-static int batman_skb_recv(struct sk_buff *skb,
-			   struct net_device *dev,
-			   struct packet_type *ptype,
-			   struct net_device *orig_dev);
-
 void hardif_free_rcu(struct rcu_head *rcu)
 {
 	struct hard_iface *hard_iface;
@@ -554,114 +548,6 @@ out:
 	return NOTIFY_DONE;
 }
 
-/* incoming packets with the batman ethertype received on any active hard
- * interface */
-static int batman_skb_recv(struct sk_buff *skb, struct net_device *dev,
-			   struct packet_type *ptype,
-			   struct net_device *orig_dev)
-{
-	struct bat_priv *bat_priv;
-	struct batman_ogm_packet *batman_ogm_packet;
-	struct hard_iface *hard_iface;
-	int ret;
-
-	hard_iface = container_of(ptype, struct hard_iface, batman_adv_ptype);
-	skb = skb_share_check(skb, GFP_ATOMIC);
-
-	/* skb was released by skb_share_check() */
-	if (!skb)
-		goto err_out;
-
-	/* packet should hold at least type and version */
-	if (unlikely(!pskb_may_pull(skb, 2)))
-		goto err_free;
-
-	/* expect a valid ethernet header here. */
-	if (unlikely(skb->mac_len != ETH_HLEN || !skb_mac_header(skb)))
-		goto err_free;
-
-	if (!hard_iface->soft_iface)
-		goto err_free;
-
-	bat_priv = netdev_priv(hard_iface->soft_iface);
-
-	if (atomic_read(&bat_priv->mesh_state) != MESH_ACTIVE)
-		goto err_free;
-
-	/* discard frames on not active interfaces */
-	if (hard_iface->if_status != IF_ACTIVE)
-		goto err_free;
-
-	batman_ogm_packet = (struct batman_ogm_packet *)skb->data;
-
-	if (batman_ogm_packet->header.version != COMPAT_VERSION) {
-		bat_dbg(DBG_BATMAN, bat_priv,
-			"Drop packet: incompatible batman version (%i)\n",
-			batman_ogm_packet->header.version);
-		goto err_free;
-	}
-
-	/* all receive handlers return whether they received or reused
-	 * the supplied skb. if not, we have to free the skb. */
-
-	switch (batman_ogm_packet->header.packet_type) {
-		/* batman originator packet */
-	case BAT_IV_OGM:
-		ret = recv_bat_ogm_packet(skb, hard_iface);
-		break;
-
-		/* batman icmp packet */
-	case BAT_ICMP:
-		ret = recv_icmp_packet(skb, hard_iface);
-		break;
-
-		/* unicast packet */
-	case BAT_UNICAST:
-	case BAT_UNICAST_4ADDR:
-		ret = recv_unicast_packet(skb, hard_iface);
-		break;
-
-		/* fragmented unicast packet */
-	case BAT_UNICAST_FRAG:
-		ret = recv_ucast_frag_packet(skb, hard_iface);
-		break;
-
-		/* broadcast packet */
-	case BAT_BCAST:
-		ret = recv_bcast_packet(skb, hard_iface);
-		break;
-
-		/* vis packet */
-	case BAT_VIS:
-		ret = recv_vis_packet(skb, hard_iface);
-		break;
-		/* Translation table query (request or response) */
-	case BAT_TT_QUERY:
-		ret = recv_tt_query(skb, hard_iface);
-		break;
-		/* Roaming advertisement */
-	case BAT_ROAM_ADV:
-		ret = recv_roam_adv(skb, hard_iface);
-		break;
-	default:
-		ret = NET_RX_DROP;
-	}
-
-	if (ret == NET_RX_DROP)
-		kfree_skb(skb);
-
-	/* return NET_RX_SUCCESS in any case as we
-	 * most probably dropped the packet for
-	 * routing-logical reasons. */
-
-	return NET_RX_SUCCESS;
-
-err_free:
-	kfree_skb(skb);
-err_out:
-	return NET_RX_DROP;
-}
-
 /* This function returns true if the interface represented by ifindex is a
  * 802.11 wireless device */
 bool is_wifi_iface(int ifindex)
diff --git a/net/batman-adv/main.c b/net/batman-adv/main.c
index 7913272..87af1a3 100644
--- a/net/batman-adv/main.c
+++ b/net/batman-adv/main.c
@@ -39,6 +39,7 @@
 /* List manipulations on hardif_list have to be rtnl_lock()'ed,
  * list traversals just rcu-locked */
 struct list_head hardif_list;
+static int (*recv_packet_handler[256])(struct sk_buff *, struct hard_iface *);
 char bat_routing_algo[20] = "BATMAN IV";
 static struct hlist_head bat_algo_list;
 
@@ -46,11 +47,15 @@ unsigned char broadcast_addr[] = {0xff, 0xff, 0xff, 0xff, 0xff, 0xff};
 
 struct workqueue_struct *bat_event_workqueue;
 
+static void recv_handler_init(void);
+
 static int __init batman_init(void)
 {
 	INIT_LIST_HEAD(&hardif_list);
 	INIT_HLIST_HEAD(&bat_algo_list);
 
+	recv_handler_init();
+
 	bat_iv_init();
 
 	/* the name should not be longer than 10 chars - see
@@ -179,6 +184,124 @@ int is_my_mac(const uint8_t *addr)
 	return 0;
 }
 
+static int recv_unhandled_packet(struct sk_buff *skb,
+				 struct hard_iface *recv_if)
+{
+	return NET_RX_DROP;
+}
+
+/* incoming packets with the batman ethertype received on any active hard
+ * interface
+ */
+int batman_skb_recv(struct sk_buff *skb, struct net_device *dev,
+		    struct packet_type *ptype, struct net_device *orig_dev)
+{
+	struct bat_priv *bat_priv;
+	struct batman_ogm_packet *batman_ogm_packet;
+	struct hard_iface *hard_iface;
+	uint8_t idx;
+	int ret;
+
+	hard_iface = container_of(ptype, struct hard_iface, batman_adv_ptype);
+	skb = skb_share_check(skb, GFP_ATOMIC);
+
+	/* skb was released by skb_share_check() */
+	if (!skb)
+		goto err_out;
+
+	/* packet should hold at least type and version */
+	if (unlikely(!pskb_may_pull(skb, 2)))
+		goto err_free;
+
+	/* expect a valid ethernet header here. */
+	if (unlikely(skb->mac_len != ETH_HLEN || !skb_mac_header(skb)))
+		goto err_free;
+
+	if (!hard_iface->soft_iface)
+		goto err_free;
+
+	bat_priv = netdev_priv(hard_iface->soft_iface);
+
+	if (atomic_read(&bat_priv->mesh_state) != MESH_ACTIVE)
+		goto err_free;
+
+	/* discard frames on not active interfaces */
+	if (hard_iface->if_status != IF_ACTIVE)
+		goto err_free;
+
+	batman_ogm_packet = (struct batman_ogm_packet *)skb->data;
+
+	if (batman_ogm_packet->header.version != COMPAT_VERSION) {
+		bat_dbg(DBG_BATMAN, bat_priv,
+			"Drop packet: incompatible batman version (%i)\n",
+			batman_ogm_packet->header.version);
+		goto err_free;
+	}
+
+	/* all receive handlers return whether they received or reused
+	 * the supplied skb. if not, we have to free the skb.
+	 */
+	idx = batman_ogm_packet->header.packet_type;
+	ret = (*recv_packet_handler[idx])(skb, hard_iface);
+
+	if (ret == NET_RX_DROP)
+		kfree_skb(skb);
+
+	/* return NET_RX_SUCCESS in any case as we
+	 * most probably dropped the packet for
+	 * routing-logical reasons.
+	 */
+	return NET_RX_SUCCESS;
+
+err_free:
+	kfree_skb(skb);
+err_out:
+	return NET_RX_DROP;
+}
+
+static void recv_handler_init(void)
+{
+	int i;
+
+	for (i = 0; i < ARRAY_SIZE(recv_packet_handler); i++)
+		recv_packet_handler[i] = recv_unhandled_packet;
+
+	/* batman originator packet */
+	recv_packet_handler[BAT_IV_OGM] = recv_bat_ogm_packet;
+	/* batman icmp packet */
+	recv_packet_handler[BAT_ICMP] = recv_icmp_packet;
+	/* unicast with 4 addresses packet */
+	recv_packet_handler[BAT_UNICAST_4ADDR] = recv_unicast_packet;
+	/* unicast packet */
+	recv_packet_handler[BAT_UNICAST] = recv_unicast_packet;
+	/* fragmented unicast packet */
+	recv_packet_handler[BAT_UNICAST_FRAG] = recv_ucast_frag_packet;
+	/* broadcast packet */
+	recv_packet_handler[BAT_BCAST] = recv_bcast_packet;
+	/* vis packet */
+	recv_packet_handler[BAT_VIS] = recv_vis_packet;
+	/* Translation table query (request or response) */
+	recv_packet_handler[BAT_TT_QUERY] = recv_tt_query;
+	/* Roaming advertisement */
+	recv_packet_handler[BAT_ROAM_ADV] = recv_roam_adv;
+}
+
+int recv_handler_register(uint8_t packet_type,
+			  int (*recv_handler)(struct sk_buff *,
+					      struct hard_iface *))
+{
+	if (recv_packet_handler[packet_type] != &recv_unhandled_packet)
+		return -EBUSY;
+
+	recv_packet_handler[packet_type] = recv_handler;
+	return 0;
+}
+
+void recv_handler_unregister(uint8_t packet_type)
+{
+	recv_packet_handler[packet_type] = recv_unhandled_packet;
+}
+
 static struct bat_algo_ops *bat_algo_get(char *name)
 {
 	struct bat_algo_ops *bat_algo_ops = NULL, *bat_algo_ops_tmp;
diff --git a/net/batman-adv/main.h b/net/batman-adv/main.h
index c11c8df..69b5173 100644
--- a/net/batman-adv/main.h
+++ b/net/batman-adv/main.h
@@ -168,6 +168,12 @@ void mesh_free(struct net_device *soft_iface);
 void inc_module_count(void);
 void dec_module_count(void);
 int is_my_mac(const uint8_t *addr);
+int batman_skb_recv(struct sk_buff *skb, struct net_device *dev,
+		    struct packet_type *ptype, struct net_device *orig_dev);
+int recv_handler_register(uint8_t packet_type,
+			  int (*recv_handler)(struct sk_buff *,
+					      struct hard_iface *));
+void recv_handler_unregister(uint8_t packet_type);
 int bat_algo_register(struct bat_algo_ops *bat_algo_ops);
 int bat_algo_select(struct bat_priv *bat_priv, char *name);
 int bat_algo_seq_print_text(struct seq_file *seq, void *offset);
-- 
1.7.9.4

[B.A.T.M.A.N.] [PATCH 11/15] batman-adv: introduce packet type handler array for incoming packets

[Antonio Quartulli]

From: Marek Lindner <lindner_marek@yahoo.de>

The packet handler array replaces the growing switch statement, thus
dealing with incoming packets in a more efficient way. It also adds
to possibility to register packet handlers on the fly.

Signed-off-by: Marek Lindner <lindner_marek@yahoo.de>
Acked-by: Simon Wunderlich <siwu@hrz.tu-chemnitz.de>
Signed-off-by: Antonio Quartulli <ordex@autistici.org>
---
 net/batman-adv/hard-interface.c |  114 ------------------------------------
 net/batman-adv/main.c           |  123 +++++++++++++++++++++++++++++++++++++++
 net/batman-adv/main.h           |    6 ++
 3 files changed, 129 insertions(+), 114 deletions(-)

diff --git a/net/batman-adv/hard-interface.c b/net/batman-adv/hard-interface.c
index 1393939..b49bf0d 100644
--- a/net/batman-adv/hard-interface.c
+++ b/net/batman-adv/hard-interface.c
@@ -33,12 +33,6 @@
 
 #include <linux/if_arp.h>
 
-
-static int batman_skb_recv(struct sk_buff *skb,
-			   struct net_device *dev,
-			   struct packet_type *ptype,
-			   struct net_device *orig_dev);
-
 void hardif_free_rcu(struct rcu_head *rcu)
 {
 	struct hard_iface *hard_iface;
@@ -554,114 +548,6 @@ out:
 	return NOTIFY_DONE;
 }
 
-/* incoming packets with the batman ethertype received on any active hard
- * interface */
-static int batman_skb_recv(struct sk_buff *skb, struct net_device *dev,
-			   struct packet_type *ptype,
-			   struct net_device *orig_dev)
-{
-	struct bat_priv *bat_priv;
-	struct batman_ogm_packet *batman_ogm_packet;
-	struct hard_iface *hard_iface;
-	int ret;
-
-	hard_iface = container_of(ptype, struct hard_iface, batman_adv_ptype);
-	skb = skb_share_check(skb, GFP_ATOMIC);
-
-	/* skb was released by skb_share_check() */
-	if (!skb)
-		goto err_out;
-
-	/* packet should hold at least type and version */
-	if (unlikely(!pskb_may_pull(skb, 2)))
-		goto err_free;
-
-	/* expect a valid ethernet header here. */
-	if (unlikely(skb->mac_len != ETH_HLEN || !skb_mac_header(skb)))
-		goto err_free;
-
-	if (!hard_iface->soft_iface)
-		goto err_free;
-
-	bat_priv = netdev_priv(hard_iface->soft_iface);
-
-	if (atomic_read(&bat_priv->mesh_state) != MESH_ACTIVE)
-		goto err_free;
-
-	/* discard frames on not active interfaces */
-	if (hard_iface->if_status != IF_ACTIVE)
-		goto err_free;
-
-	batman_ogm_packet = (struct batman_ogm_packet *)skb->data;
-
-	if (batman_ogm_packet->header.version != COMPAT_VERSION) {
-		bat_dbg(DBG_BATMAN, bat_priv,
-			"Drop packet: incompatible batman version (%i)\n",
-			batman_ogm_packet->header.version);
-		goto err_free;
-	}
-
-	/* all receive handlers return whether they received or reused
-	 * the supplied skb. if not, we have to free the skb. */
-
-	switch (batman_ogm_packet->header.packet_type) {
-		/* batman originator packet */
-	case BAT_IV_OGM:
-		ret = recv_bat_ogm_packet(skb, hard_iface);
-		break;
-
-		/* batman icmp packet */
-	case BAT_ICMP:
-		ret = recv_icmp_packet(skb, hard_iface);
-		break;
-
-		/* unicast packet */
-	case BAT_UNICAST:
-	case BAT_UNICAST_4ADDR:
-		ret = recv_unicast_packet(skb, hard_iface);
-		break;
-
-		/* fragmented unicast packet */
-	case BAT_UNICAST_FRAG:
-		ret = recv_ucast_frag_packet(skb, hard_iface);
-		break;
-
-		/* broadcast packet */
-	case BAT_BCAST:
-		ret = recv_bcast_packet(skb, hard_iface);
-		break;
-
-		/* vis packet */
-	case BAT_VIS:
-		ret = recv_vis_packet(skb, hard_iface);
-		break;
-		/* Translation table query (request or response) */
-	case BAT_TT_QUERY:
-		ret = recv_tt_query(skb, hard_iface);
-		break;
-		/* Roaming advertisement */
-	case BAT_ROAM_ADV:
-		ret = recv_roam_adv(skb, hard_iface);
-		break;
-	default:
-		ret = NET_RX_DROP;
-	}
-
-	if (ret == NET_RX_DROP)
-		kfree_skb(skb);
-
-	/* return NET_RX_SUCCESS in any case as we
-	 * most probably dropped the packet for
-	 * routing-logical reasons. */
-
-	return NET_RX_SUCCESS;
-
-err_free:
-	kfree_skb(skb);
-err_out:
-	return NET_RX_DROP;
-}
-
 /* This function returns true if the interface represented by ifindex is a
  * 802.11 wireless device */
 bool is_wifi_iface(int ifindex)
diff --git a/net/batman-adv/main.c b/net/batman-adv/main.c
index 7913272..87af1a3 100644
--- a/net/batman-adv/main.c
+++ b/net/batman-adv/main.c
@@ -39,6 +39,7 @@
 /* List manipulations on hardif_list have to be rtnl_lock()'ed,
  * list traversals just rcu-locked */
 struct list_head hardif_list;
+static int (*recv_packet_handler[256])(struct sk_buff *, struct hard_iface *);
 char bat_routing_algo[20] = "BATMAN IV";
 static struct hlist_head bat_algo_list;
 
@@ -46,11 +47,15 @@ unsigned char broadcast_addr[] = {0xff, 0xff, 0xff, 0xff, 0xff, 0xff};
 
 struct workqueue_struct *bat_event_workqueue;
 
+static void recv_handler_init(void);
+
 static int __init batman_init(void)
 {
 	INIT_LIST_HEAD(&hardif_list);
 	INIT_HLIST_HEAD(&bat_algo_list);
 
+	recv_handler_init();
+
 	bat_iv_init();
 
 	/* the name should not be longer than 10 chars - see
@@ -179,6 +184,124 @@ int is_my_mac(const uint8_t *addr)
 	return 0;
 }
 
+static int recv_unhandled_packet(struct sk_buff *skb,
+				 struct hard_iface *recv_if)
+{
+	return NET_RX_DROP;
+}
+
+/* incoming packets with the batman ethertype received on any active hard
+ * interface
+ */
+int batman_skb_recv(struct sk_buff *skb, struct net_device *dev,
+		    struct packet_type *ptype, struct net_device *orig_dev)
+{
+	struct bat_priv *bat_priv;
+	struct batman_ogm_packet *batman_ogm_packet;
+	struct hard_iface *hard_iface;
+	uint8_t idx;
+	int ret;
+
+	hard_iface = container_of(ptype, struct hard_iface, batman_adv_ptype);
+	skb = skb_share_check(skb, GFP_ATOMIC);
+
+	/* skb was released by skb_share_check() */
+	if (!skb)
+		goto err_out;
+
+	/* packet should hold at least type and version */
+	if (unlikely(!pskb_may_pull(skb, 2)))
+		goto err_free;
+
+	/* expect a valid ethernet header here. */
+	if (unlikely(skb->mac_len != ETH_HLEN || !skb_mac_header(skb)))
+		goto err_free;
+
+	if (!hard_iface->soft_iface)
+		goto err_free;
+
+	bat_priv = netdev_priv(hard_iface->soft_iface);
+
+	if (atomic_read(&bat_priv->mesh_state) != MESH_ACTIVE)
+		goto err_free;
+
+	/* discard frames on not active interfaces */
+	if (hard_iface->if_status != IF_ACTIVE)
+		goto err_free;
+
+	batman_ogm_packet = (struct batman_ogm_packet *)skb->data;
+
+	if (batman_ogm_packet->header.version != COMPAT_VERSION) {
+		bat_dbg(DBG_BATMAN, bat_priv,
+			"Drop packet: incompatible batman version (%i)\n",
+			batman_ogm_packet->header.version);
+		goto err_free;
+	}
+
+	/* all receive handlers return whether they received or reused
+	 * the supplied skb. if not, we have to free the skb.
+	 */
+	idx = batman_ogm_packet->header.packet_type;
+	ret = (*recv_packet_handler[idx])(skb, hard_iface);
+
+	if (ret == NET_RX_DROP)
+		kfree_skb(skb);
+
+	/* return NET_RX_SUCCESS in any case as we
+	 * most probably dropped the packet for
+	 * routing-logical reasons.
+	 */
+	return NET_RX_SUCCESS;
+
+err_free:
+	kfree_skb(skb);
+err_out:
+	return NET_RX_DROP;
+}
+
+static void recv_handler_init(void)
+{
+	int i;
+
+	for (i = 0; i < ARRAY_SIZE(recv_packet_handler); i++)
+		recv_packet_handler[i] = recv_unhandled_packet;
+
+	/* batman originator packet */
+	recv_packet_handler[BAT_IV_OGM] = recv_bat_ogm_packet;
+	/* batman icmp packet */
+	recv_packet_handler[BAT_ICMP] = recv_icmp_packet;
+	/* unicast with 4 addresses packet */
+	recv_packet_handler[BAT_UNICAST_4ADDR] = recv_unicast_packet;
+	/* unicast packet */
+	recv_packet_handler[BAT_UNICAST] = recv_unicast_packet;
+	/* fragmented unicast packet */
+	recv_packet_handler[BAT_UNICAST_FRAG] = recv_ucast_frag_packet;
+	/* broadcast packet */
+	recv_packet_handler[BAT_BCAST] = recv_bcast_packet;
+	/* vis packet */
+	recv_packet_handler[BAT_VIS] = recv_vis_packet;
+	/* Translation table query (request or response) */
+	recv_packet_handler[BAT_TT_QUERY] = recv_tt_query;
+	/* Roaming advertisement */
+	recv_packet_handler[BAT_ROAM_ADV] = recv_roam_adv;
+}
+
+int recv_handler_register(uint8_t packet_type,
+			  int (*recv_handler)(struct sk_buff *,
+					      struct hard_iface *))
+{
+	if (recv_packet_handler[packet_type] != &recv_unhandled_packet)
+		return -EBUSY;
+
+	recv_packet_handler[packet_type] = recv_handler;
+	return 0;
+}
+
+void recv_handler_unregister(uint8_t packet_type)
+{
+	recv_packet_handler[packet_type] = recv_unhandled_packet;
+}
+
 static struct bat_algo_ops *bat_algo_get(char *name)
 {
 	struct bat_algo_ops *bat_algo_ops = NULL, *bat_algo_ops_tmp;
diff --git a/net/batman-adv/main.h b/net/batman-adv/main.h
index c11c8df..69b5173 100644
--- a/net/batman-adv/main.h
+++ b/net/batman-adv/main.h
@@ -168,6 +168,12 @@ void mesh_free(struct net_device *soft_iface);
 void inc_module_count(void);
 void dec_module_count(void);
 int is_my_mac(const uint8_t *addr);
+int batman_skb_recv(struct sk_buff *skb, struct net_device *dev,
+		    struct packet_type *ptype, struct net_device *orig_dev);
+int recv_handler_register(uint8_t packet_type,
+			  int (*recv_handler)(struct sk_buff *,
+					      struct hard_iface *));
+void recv_handler_unregister(uint8_t packet_type);
 int bat_algo_register(struct bat_algo_ops *bat_algo_ops);
 int bat_algo_select(struct bat_priv *bat_priv, char *name);
 int bat_algo_seq_print_text(struct seq_file *seq, void *offset);
-- 
1.7.9.4

[PATCH 12/15] batman-adv: register batman ogm receive function during protocol init

[Antonio Quartulli]

From: Marek Lindner <lindner_marek@yahoo.de>

The B.A.T.M.A.N. IV OGM receive function still was hard-coded although
it is a routing protocol specific function. This patch takes advantage
of the dynamic packet handler registration to remove the hard-coded
function calls.

Signed-off-by: Marek Lindner <lindner_marek@yahoo.de>
Acked-by: Simon Wunderlich <siwu@hrz.tu-chemnitz.de>
Signed-off-by: Antonio Quartulli <ordex@autistici.org>
---
 net/batman-adv/bat_iv_ogm.c |   31 +++++++++++++++++++++++++++----
 net/batman-adv/main.c       |    5 +----
 net/batman-adv/routing.c    |   22 ++++++++++------------
 net/batman-adv/routing.h    |    4 +++-
 net/batman-adv/types.h      |    3 ---
 5 files changed, 41 insertions(+), 24 deletions(-)

diff --git a/net/batman-adv/bat_iv_ogm.c b/net/batman-adv/bat_iv_ogm.c
index cd8f473..e0aaf8c 100644
--- a/net/batman-adv/bat_iv_ogm.c
+++ b/net/batman-adv/bat_iv_ogm.c
@@ -1155,13 +1155,18 @@ out:
 	orig_node_free_ref(orig_node);
 }
 
-static void bat_iv_ogm_receive(struct hard_iface *if_incoming,
-			       struct sk_buff *skb)
+static int bat_iv_ogm_receive(struct sk_buff *skb,
+			      struct hard_iface *if_incoming)
 {
 	struct batman_ogm_packet *batman_ogm_packet;
 	struct ethhdr *ethhdr;
 	int buff_pos = 0, packet_len;
 	unsigned char *tt_buff, *packet_buff;
+	bool ret;
+
+	ret = check_management_packet(skb, if_incoming, BATMAN_OGM_HLEN);
+	if (!ret)
+		return NET_RX_DROP;
 
 	packet_len = skb_headlen(skb);
 	ethhdr = (struct ethhdr *)skb_mac_header(skb);
@@ -1187,6 +1192,9 @@ static void bat_iv_ogm_receive(struct hard_iface *if_incoming,
 						(packet_buff + buff_pos);
 	} while (bat_iv_ogm_aggr_packet(buff_pos, packet_len,
 					batman_ogm_packet->tt_num_changes));
+
+	kfree_skb(skb);
+	return NET_RX_SUCCESS;
 }
 
 static struct bat_algo_ops batman_iv __read_mostly = {
@@ -1197,10 +1205,25 @@ static struct bat_algo_ops batman_iv __read_mostly = {
 	.bat_ogm_update_mac = bat_iv_ogm_update_mac,
 	.bat_ogm_schedule = bat_iv_ogm_schedule,
 	.bat_ogm_emit = bat_iv_ogm_emit,
-	.bat_ogm_receive = bat_iv_ogm_receive,
 };
 
 int __init bat_iv_init(void)
 {
-	return bat_algo_register(&batman_iv);
+	int ret;
+
+	/* batman originator packet */
+	ret = recv_handler_register(BAT_IV_OGM, bat_iv_ogm_receive);
+	if (ret < 0)
+		goto out;
+
+	ret = bat_algo_register(&batman_iv);
+	if (ret < 0)
+		goto handler_unregister;
+
+	goto out;
+
+handler_unregister:
+	recv_handler_unregister(BAT_IV_OGM);
+out:
+	return ret;
 }
diff --git a/net/batman-adv/main.c b/net/batman-adv/main.c
index 87af1a3..7a3bfae 100644
--- a/net/batman-adv/main.c
+++ b/net/batman-adv/main.c
@@ -266,8 +266,6 @@ static void recv_handler_init(void)
 	for (i = 0; i < ARRAY_SIZE(recv_packet_handler); i++)
 		recv_packet_handler[i] = recv_unhandled_packet;
 
-	/* batman originator packet */
-	recv_packet_handler[BAT_IV_OGM] = recv_bat_ogm_packet;
 	/* batman icmp packet */
 	recv_packet_handler[BAT_ICMP] = recv_icmp_packet;
 	/* unicast with 4 addresses packet */
@@ -336,8 +334,7 @@ int bat_algo_register(struct bat_algo_ops *bat_algo_ops)
 	    !bat_algo_ops->bat_primary_iface_set ||
 	    !bat_algo_ops->bat_ogm_update_mac ||
 	    !bat_algo_ops->bat_ogm_schedule ||
-	    !bat_algo_ops->bat_ogm_emit ||
-	    !bat_algo_ops->bat_ogm_receive) {
+	    !bat_algo_ops->bat_ogm_emit) {
 		pr_info("Routing algo '%s' does not implement required ops\n",
 			bat_algo_ops->name);
 		goto out;
diff --git a/net/batman-adv/routing.c b/net/batman-adv/routing.c
index 5f19bbb..83f8115 100644
--- a/net/batman-adv/routing.c
+++ b/net/batman-adv/routing.c
@@ -248,37 +248,35 @@ int window_protected(struct bat_priv *bat_priv, int32_t seq_num_diff,
 	return 0;
 }
 
-int recv_bat_ogm_packet(struct sk_buff *skb, struct hard_iface *hard_iface)
+bool check_management_packet(struct sk_buff *skb,
+			     struct hard_iface *hard_iface,
+			     int header_len)
 {
-	struct bat_priv *bat_priv = netdev_priv(hard_iface->soft_iface);
 	struct ethhdr *ethhdr;
 
 	/* drop packet if it has not necessary minimum size */
-	if (unlikely(!pskb_may_pull(skb, BATMAN_OGM_HLEN)))
-		return NET_RX_DROP;
+	if (unlikely(!pskb_may_pull(skb, header_len)))
+		return false;
 
 	ethhdr = (struct ethhdr *)skb_mac_header(skb);
 
 	/* packet with broadcast indication but unicast recipient */
 	if (!is_broadcast_ether_addr(ethhdr->h_dest))
-		return NET_RX_DROP;
+		return false;
 
 	/* packet with broadcast sender address */
 	if (is_broadcast_ether_addr(ethhdr->h_source))
-		return NET_RX_DROP;
+		return false;
 
 	/* create a copy of the skb, if needed, to modify it. */
 	if (skb_cow(skb, 0) < 0)
-		return NET_RX_DROP;
+		return false;
 
 	/* keep skb linear */
 	if (skb_linearize(skb) < 0)
-		return NET_RX_DROP;
+		return false;
 
-	bat_priv->bat_algo_ops->bat_ogm_receive(hard_iface, skb);
-
-	kfree_skb(skb);
-	return NET_RX_SUCCESS;
+	return true;
 }
 
 static int recv_my_icmp_packet(struct bat_priv *bat_priv,
diff --git a/net/batman-adv/routing.h b/net/batman-adv/routing.h
index 3d729cb..d6bbbeb 100644
--- a/net/batman-adv/routing.h
+++ b/net/batman-adv/routing.h
@@ -23,6 +23,9 @@
 #define _NET_BATMAN_ADV_ROUTING_H_
 
 void slide_own_bcast_window(struct hard_iface *hard_iface);
+bool check_management_packet(struct sk_buff *skb,
+			     struct hard_iface *hard_iface,
+			     int header_len);
 void update_route(struct bat_priv *bat_priv, struct orig_node *orig_node,
 		  struct neigh_node *neigh_node);
 int recv_icmp_packet(struct sk_buff *skb, struct hard_iface *recv_if);
@@ -30,7 +33,6 @@ int recv_unicast_packet(struct sk_buff *skb, struct hard_iface *recv_if);
 int recv_ucast_frag_packet(struct sk_buff *skb, struct hard_iface *recv_if);
 int recv_bcast_packet(struct sk_buff *skb, struct hard_iface *recv_if);
 int recv_vis_packet(struct sk_buff *skb, struct hard_iface *recv_if);
-int recv_bat_ogm_packet(struct sk_buff *skb, struct hard_iface *recv_if);
 int recv_tt_query(struct sk_buff *skb, struct hard_iface *recv_if);
 int recv_roam_adv(struct sk_buff *skb, struct hard_iface *recv_if);
 struct neigh_node *find_router(struct bat_priv *bat_priv,
diff --git a/net/batman-adv/types.h b/net/batman-adv/types.h
index 2b8ffcb..ed6db5c 100644
--- a/net/batman-adv/types.h
+++ b/net/batman-adv/types.h
@@ -407,9 +407,6 @@ struct bat_algo_ops {
 				 int tt_num_changes);
 	/* send scheduled OGM */
 	void (*bat_ogm_emit)(struct forw_packet *forw_packet);
-	/* receive incoming OGM */
-	void (*bat_ogm_receive)(struct hard_iface *if_incoming,
-				struct sk_buff *skb);
 };
 
 struct dht_candidate {
-- 
1.7.9.4

[B.A.T.M.A.N.] [PATCH 12/15] batman-adv: register batman ogm receive function during protocol init

[Antonio Quartulli]

From: Marek Lindner <lindner_marek@yahoo.de>

The B.A.T.M.A.N. IV OGM receive function still was hard-coded although
it is a routing protocol specific function. This patch takes advantage
of the dynamic packet handler registration to remove the hard-coded
function calls.

Signed-off-by: Marek Lindner <lindner_marek@yahoo.de>
Acked-by: Simon Wunderlich <siwu@hrz.tu-chemnitz.de>
Signed-off-by: Antonio Quartulli <ordex@autistici.org>
---
 net/batman-adv/bat_iv_ogm.c |   31 +++++++++++++++++++++++++++----
 net/batman-adv/main.c       |    5 +----
 net/batman-adv/routing.c    |   22 ++++++++++------------
 net/batman-adv/routing.h    |    4 +++-
 net/batman-adv/types.h      |    3 ---
 5 files changed, 41 insertions(+), 24 deletions(-)

diff --git a/net/batman-adv/bat_iv_ogm.c b/net/batman-adv/bat_iv_ogm.c
index cd8f473..e0aaf8c 100644
--- a/net/batman-adv/bat_iv_ogm.c
+++ b/net/batman-adv/bat_iv_ogm.c
@@ -1155,13 +1155,18 @@ out:
 	orig_node_free_ref(orig_node);
 }
 
-static void bat_iv_ogm_receive(struct hard_iface *if_incoming,
-			       struct sk_buff *skb)
+static int bat_iv_ogm_receive(struct sk_buff *skb,
+			      struct hard_iface *if_incoming)
 {
 	struct batman_ogm_packet *batman_ogm_packet;
 	struct ethhdr *ethhdr;
 	int buff_pos = 0, packet_len;
 	unsigned char *tt_buff, *packet_buff;
+	bool ret;
+
+	ret = check_management_packet(skb, if_incoming, BATMAN_OGM_HLEN);
+	if (!ret)
+		return NET_RX_DROP;
 
 	packet_len = skb_headlen(skb);
 	ethhdr = (struct ethhdr *)skb_mac_header(skb);
@@ -1187,6 +1192,9 @@ static void bat_iv_ogm_receive(struct hard_iface *if_incoming,
 						(packet_buff + buff_pos);
 	} while (bat_iv_ogm_aggr_packet(buff_pos, packet_len,
 					batman_ogm_packet->tt_num_changes));
+
+	kfree_skb(skb);
+	return NET_RX_SUCCESS;
 }
 
 static struct bat_algo_ops batman_iv __read_mostly = {
@@ -1197,10 +1205,25 @@ static struct bat_algo_ops batman_iv __read_mostly = {
 	.bat_ogm_update_mac = bat_iv_ogm_update_mac,
 	.bat_ogm_schedule = bat_iv_ogm_schedule,
 	.bat_ogm_emit = bat_iv_ogm_emit,
-	.bat_ogm_receive = bat_iv_ogm_receive,
 };
 
 int __init bat_iv_init(void)
 {
-	return bat_algo_register(&batman_iv);
+	int ret;
+
+	/* batman originator packet */
+	ret = recv_handler_register(BAT_IV_OGM, bat_iv_ogm_receive);
+	if (ret < 0)
+		goto out;
+
+	ret = bat_algo_register(&batman_iv);
+	if (ret < 0)
+		goto handler_unregister;
+
+	goto out;
+
+handler_unregister:
+	recv_handler_unregister(BAT_IV_OGM);
+out:
+	return ret;
 }
diff --git a/net/batman-adv/main.c b/net/batman-adv/main.c
index 87af1a3..7a3bfae 100644
--- a/net/batman-adv/main.c
+++ b/net/batman-adv/main.c
@@ -266,8 +266,6 @@ static void recv_handler_init(void)
 	for (i = 0; i < ARRAY_SIZE(recv_packet_handler); i++)
 		recv_packet_handler[i] = recv_unhandled_packet;
 
-	/* batman originator packet */
-	recv_packet_handler[BAT_IV_OGM] = recv_bat_ogm_packet;
 	/* batman icmp packet */
 	recv_packet_handler[BAT_ICMP] = recv_icmp_packet;
 	/* unicast with 4 addresses packet */
@@ -336,8 +334,7 @@ int bat_algo_register(struct bat_algo_ops *bat_algo_ops)
 	    !bat_algo_ops->bat_primary_iface_set ||
 	    !bat_algo_ops->bat_ogm_update_mac ||
 	    !bat_algo_ops->bat_ogm_schedule ||
-	    !bat_algo_ops->bat_ogm_emit ||
-	    !bat_algo_ops->bat_ogm_receive) {
+	    !bat_algo_ops->bat_ogm_emit) {
 		pr_info("Routing algo '%s' does not implement required ops\n",
 			bat_algo_ops->name);
 		goto out;
diff --git a/net/batman-adv/routing.c b/net/batman-adv/routing.c
index 5f19bbb..83f8115 100644
--- a/net/batman-adv/routing.c
+++ b/net/batman-adv/routing.c
@@ -248,37 +248,35 @@ int window_protected(struct bat_priv *bat_priv, int32_t seq_num_diff,
 	return 0;
 }
 
-int recv_bat_ogm_packet(struct sk_buff *skb, struct hard_iface *hard_iface)
+bool check_management_packet(struct sk_buff *skb,
+			     struct hard_iface *hard_iface,
+			     int header_len)
 {
-	struct bat_priv *bat_priv = netdev_priv(hard_iface->soft_iface);
 	struct ethhdr *ethhdr;
 
 	/* drop packet if it has not necessary minimum size */
-	if (unlikely(!pskb_may_pull(skb, BATMAN_OGM_HLEN)))
-		return NET_RX_DROP;
+	if (unlikely(!pskb_may_pull(skb, header_len)))
+		return false;
 
 	ethhdr = (struct ethhdr *)skb_mac_header(skb);
 
 	/* packet with broadcast indication but unicast recipient */
 	if (!is_broadcast_ether_addr(ethhdr->h_dest))
-		return NET_RX_DROP;
+		return false;
 
 	/* packet with broadcast sender address */
 	if (is_broadcast_ether_addr(ethhdr->h_source))
-		return NET_RX_DROP;
+		return false;
 
 	/* create a copy of the skb, if needed, to modify it. */
 	if (skb_cow(skb, 0) < 0)
-		return NET_RX_DROP;
+		return false;
 
 	/* keep skb linear */
 	if (skb_linearize(skb) < 0)
-		return NET_RX_DROP;
+		return false;
 
-	bat_priv->bat_algo_ops->bat_ogm_receive(hard_iface, skb);
-
-	kfree_skb(skb);
-	return NET_RX_SUCCESS;
+	return true;
 }
 
 static int recv_my_icmp_packet(struct bat_priv *bat_priv,
diff --git a/net/batman-adv/routing.h b/net/batman-adv/routing.h
index 3d729cb..d6bbbeb 100644
--- a/net/batman-adv/routing.h
+++ b/net/batman-adv/routing.h
@@ -23,6 +23,9 @@
 #define _NET_BATMAN_ADV_ROUTING_H_
 
 void slide_own_bcast_window(struct hard_iface *hard_iface);
+bool check_management_packet(struct sk_buff *skb,
+			     struct hard_iface *hard_iface,
+			     int header_len);
 void update_route(struct bat_priv *bat_priv, struct orig_node *orig_node,
 		  struct neigh_node *neigh_node);
 int recv_icmp_packet(struct sk_buff *skb, struct hard_iface *recv_if);
@@ -30,7 +33,6 @@ int recv_unicast_packet(struct sk_buff *skb, struct hard_iface *recv_if);
 int recv_ucast_frag_packet(struct sk_buff *skb, struct hard_iface *recv_if);
 int recv_bcast_packet(struct sk_buff *skb, struct hard_iface *recv_if);
 int recv_vis_packet(struct sk_buff *skb, struct hard_iface *recv_if);
-int recv_bat_ogm_packet(struct sk_buff *skb, struct hard_iface *recv_if);
 int recv_tt_query(struct sk_buff *skb, struct hard_iface *recv_if);
 int recv_roam_adv(struct sk_buff *skb, struct hard_iface *recv_if);
 struct neigh_node *find_router(struct bat_priv *bat_priv,
diff --git a/net/batman-adv/types.h b/net/batman-adv/types.h
index 2b8ffcb..ed6db5c 100644
--- a/net/batman-adv/types.h
+++ b/net/batman-adv/types.h
@@ -407,9 +407,6 @@ struct bat_algo_ops {
 				 int tt_num_changes);
 	/* send scheduled OGM */
 	void (*bat_ogm_emit)(struct forw_packet *forw_packet);
-	/* receive incoming OGM */
-	void (*bat_ogm_receive)(struct hard_iface *if_incoming,
-				struct sk_buff *skb);
 };
 
 struct dht_candidate {
-- 
1.7.9.4

[PATCH 13/15] batman-adv: rename last_valid to last_seen

[Antonio Quartulli]

From: Marek Lindner <lindner_marek@yahoo.de>

Signed-off-by: Marek Lindner <lindner_marek@yahoo.de>
Acked-by: Simon Wunderlich <siwu@hrz.tu-chemnitz.de>
Signed-off-by: Antonio Quartulli <ordex@autistici.org>
---
 net/batman-adv/bat_iv_ogm.c |    8 ++++----
 net/batman-adv/originator.c |   16 ++++++++--------
 net/batman-adv/types.h      |    8 ++++----
 3 files changed, 16 insertions(+), 16 deletions(-)

diff --git a/net/batman-adv/bat_iv_ogm.c b/net/batman-adv/bat_iv_ogm.c
index e0aaf8c..8652a75 100644
--- a/net/batman-adv/bat_iv_ogm.c
+++ b/net/batman-adv/bat_iv_ogm.c
@@ -651,7 +651,7 @@ static void bat_iv_ogm_orig_update(struct bat_priv *bat_priv,
 	rcu_read_unlock();
 
 	orig_node->flags = batman_ogm_packet->flags;
-	neigh_node->last_valid = jiffies;
+	neigh_node->last_seen = jiffies;
 
 	spin_lock_bh(&neigh_node->tq_lock);
 	ring_buffer_set(neigh_node->tq_recv,
@@ -772,11 +772,11 @@ static int bat_iv_ogm_calc_tq(struct orig_node *orig_node,
 	if (!neigh_node)
 		goto out;
 
-	/* if orig_node is direct neighbor update neigh_node last_valid */
+	/* if orig_node is direct neighbor update neigh_node last_seen */
 	if (orig_node == orig_neigh_node)
-		neigh_node->last_valid = jiffies;
+		neigh_node->last_seen = jiffies;
 
-	orig_node->last_valid = jiffies;
+	orig_node->last_seen = jiffies;
 
 	/* find packet count of corresponding one hop neighbor */
 	spin_lock_bh(&orig_node->ogm_cnt_lock);
diff --git a/net/batman-adv/originator.c b/net/batman-adv/originator.c
index 31d7b58..936b2e6 100644
--- a/net/batman-adv/originator.c
+++ b/net/batman-adv/originator.c
@@ -285,7 +285,7 @@ static bool purge_orig_neighbors(struct bat_priv *bat_priv,
 	hlist_for_each_entry_safe(neigh_node, node, node_tmp,
 				  &orig_node->neigh_list, list) {
 
-		if ((has_timed_out(neigh_node->last_valid, PURGE_TIMEOUT)) ||
+		if ((has_timed_out(neigh_node->last_seen, PURGE_TIMEOUT)) ||
 		    (neigh_node->if_incoming->if_status == IF_INACTIVE) ||
 		    (neigh_node->if_incoming->if_status == IF_NOT_IN_USE) ||
 		    (neigh_node->if_incoming->if_status == IF_TO_BE_REMOVED)) {
@@ -302,9 +302,9 @@ static bool purge_orig_neighbors(struct bat_priv *bat_priv,
 					neigh_node->if_incoming->net_dev->name);
 			else
 				bat_dbg(DBG_BATMAN, bat_priv,
-					"neighbor timeout: originator %pM, neighbor: %pM, last_valid: %lu\n",
+					"neighbor timeout: originator %pM, neighbor: %pM, last_seen: %lu\n",
 					orig_node->orig, neigh_node->addr,
-					(neigh_node->last_valid / HZ));
+					(neigh_node->last_seen / HZ));
 
 			neigh_purged = true;
 
@@ -327,10 +327,10 @@ static bool purge_orig_node(struct bat_priv *bat_priv,
 {
 	struct neigh_node *best_neigh_node;
 
-	if (has_timed_out(orig_node->last_valid, 2 * PURGE_TIMEOUT)) {
+	if (has_timed_out(orig_node->last_seen, 2 * PURGE_TIMEOUT)) {
 		bat_dbg(DBG_BATMAN, bat_priv,
-			"Originator timeout: originator %pM, last_valid %lu\n",
-			orig_node->orig, (orig_node->last_valid / HZ));
+			"Originator timeout: originator %pM, last_seen %lu\n",
+			orig_node->orig, (orig_node->last_seen / HZ));
 		return true;
 	} else {
 		if (purge_orig_neighbors(bat_priv, orig_node,
@@ -448,9 +448,9 @@ int orig_seq_print_text(struct seq_file *seq, void *offset)
 				goto next;
 
 			last_seen_secs = jiffies_to_msecs(jiffies -
-						orig_node->last_valid) / 1000;
+						orig_node->last_seen) / 1000;
 			last_seen_msecs = jiffies_to_msecs(jiffies -
-						orig_node->last_valid) % 1000;
+						orig_node->last_seen) % 1000;
 
 			seq_printf(seq, "%pM %4i.%03is   (%3i) %pM [%10s]:",
 				   orig_node->orig, last_seen_secs,
diff --git a/net/batman-adv/types.h b/net/batman-adv/types.h
index ed6db5c..5ca9b58 100644
--- a/net/batman-adv/types.h
+++ b/net/batman-adv/types.h
@@ -63,7 +63,7 @@ struct hard_iface {
 /**
  *	orig_node - structure for orig_list maintaining nodes of mesh
  *	@primary_addr: hosts primary interface address
- *	@last_valid: when last packet from this node was received
+ *	@last_seen: when last packet from this node was received
  *	@bcast_seqno_reset: time when the broadcast seqno window was reset
  *	@batman_seqno_reset: time when the batman seqno window was reset
  *	@gw_flags: flags related to gateway class
@@ -84,7 +84,7 @@ struct orig_node {
 	struct neigh_node __rcu *router; /* rcu protected pointer */
 	unsigned long *bcast_own;
 	uint8_t *bcast_own_sum;
-	unsigned long last_valid;
+	unsigned long last_seen;
 	unsigned long bcast_seqno_reset;
 	unsigned long batman_seqno_reset;
 	uint8_t gw_flags;
@@ -134,7 +134,7 @@ struct gw_node {
 
 /**
  *	neigh_node
- *	@last_valid: when last packet via this neighbor was received
+ *	@last_seen: when last packet via this neighbor was received
  */
 struct neigh_node {
 	struct hlist_node list;
@@ -145,7 +145,7 @@ struct neigh_node {
 	uint8_t tq_avg;
 	uint8_t last_ttl;
 	struct list_head bonding_list;
-	unsigned long last_valid;
+	unsigned long last_seen;
 	DECLARE_BITMAP(real_bits, TQ_LOCAL_WINDOW_SIZE);
 	atomic_t refcount;
 	struct rcu_head rcu;
-- 
1.7.9.4

[B.A.T.M.A.N.] [PATCH 13/15] batman-adv: rename last_valid to last_seen

[Antonio Quartulli]

From: Marek Lindner <lindner_marek@yahoo.de>

Signed-off-by: Marek Lindner <lindner_marek@yahoo.de>
Acked-by: Simon Wunderlich <siwu@hrz.tu-chemnitz.de>
Signed-off-by: Antonio Quartulli <ordex@autistici.org>
---
 net/batman-adv/bat_iv_ogm.c |    8 ++++----
 net/batman-adv/originator.c |   16 ++++++++--------
 net/batman-adv/types.h      |    8 ++++----
 3 files changed, 16 insertions(+), 16 deletions(-)

diff --git a/net/batman-adv/bat_iv_ogm.c b/net/batman-adv/bat_iv_ogm.c
index e0aaf8c..8652a75 100644
--- a/net/batman-adv/bat_iv_ogm.c
+++ b/net/batman-adv/bat_iv_ogm.c
@@ -651,7 +651,7 @@ static void bat_iv_ogm_orig_update(struct bat_priv *bat_priv,
 	rcu_read_unlock();
 
 	orig_node->flags = batman_ogm_packet->flags;
-	neigh_node->last_valid = jiffies;
+	neigh_node->last_seen = jiffies;
 
 	spin_lock_bh(&neigh_node->tq_lock);
 	ring_buffer_set(neigh_node->tq_recv,
@@ -772,11 +772,11 @@ static int bat_iv_ogm_calc_tq(struct orig_node *orig_node,
 	if (!neigh_node)
 		goto out;
 
-	/* if orig_node is direct neighbor update neigh_node last_valid */
+	/* if orig_node is direct neighbor update neigh_node last_seen */
 	if (orig_node == orig_neigh_node)
-		neigh_node->last_valid = jiffies;
+		neigh_node->last_seen = jiffies;
 
-	orig_node->last_valid = jiffies;
+	orig_node->last_seen = jiffies;
 
 	/* find packet count of corresponding one hop neighbor */
 	spin_lock_bh(&orig_node->ogm_cnt_lock);
diff --git a/net/batman-adv/originator.c b/net/batman-adv/originator.c
index 31d7b58..936b2e6 100644
--- a/net/batman-adv/originator.c
+++ b/net/batman-adv/originator.c
@@ -285,7 +285,7 @@ static bool purge_orig_neighbors(struct bat_priv *bat_priv,
 	hlist_for_each_entry_safe(neigh_node, node, node_tmp,
 				  &orig_node->neigh_list, list) {
 
-		if ((has_timed_out(neigh_node->last_valid, PURGE_TIMEOUT)) ||
+		if ((has_timed_out(neigh_node->last_seen, PURGE_TIMEOUT)) ||
 		    (neigh_node->if_incoming->if_status == IF_INACTIVE) ||
 		    (neigh_node->if_incoming->if_status == IF_NOT_IN_USE) ||
 		    (neigh_node->if_incoming->if_status == IF_TO_BE_REMOVED)) {
@@ -302,9 +302,9 @@ static bool purge_orig_neighbors(struct bat_priv *bat_priv,
 					neigh_node->if_incoming->net_dev->name);
 			else
 				bat_dbg(DBG_BATMAN, bat_priv,
-					"neighbor timeout: originator %pM, neighbor: %pM, last_valid: %lu\n",
+					"neighbor timeout: originator %pM, neighbor: %pM, last_seen: %lu\n",
 					orig_node->orig, neigh_node->addr,
-					(neigh_node->last_valid / HZ));
+					(neigh_node->last_seen / HZ));
 
 			neigh_purged = true;
 
@@ -327,10 +327,10 @@ static bool purge_orig_node(struct bat_priv *bat_priv,
 {
 	struct neigh_node *best_neigh_node;
 
-	if (has_timed_out(orig_node->last_valid, 2 * PURGE_TIMEOUT)) {
+	if (has_timed_out(orig_node->last_seen, 2 * PURGE_TIMEOUT)) {
 		bat_dbg(DBG_BATMAN, bat_priv,
-			"Originator timeout: originator %pM, last_valid %lu\n",
-			orig_node->orig, (orig_node->last_valid / HZ));
+			"Originator timeout: originator %pM, last_seen %lu\n",
+			orig_node->orig, (orig_node->last_seen / HZ));
 		return true;
 	} else {
 		if (purge_orig_neighbors(bat_priv, orig_node,
@@ -448,9 +448,9 @@ int orig_seq_print_text(struct seq_file *seq, void *offset)
 				goto next;
 
 			last_seen_secs = jiffies_to_msecs(jiffies -
-						orig_node->last_valid) / 1000;
+						orig_node->last_seen) / 1000;
 			last_seen_msecs = jiffies_to_msecs(jiffies -
-						orig_node->last_valid) % 1000;
+						orig_node->last_seen) % 1000;
 
 			seq_printf(seq, "%pM %4i.%03is   (%3i) %pM [%10s]:",
 				   orig_node->orig, last_seen_secs,
diff --git a/net/batman-adv/types.h b/net/batman-adv/types.h
index ed6db5c..5ca9b58 100644
--- a/net/batman-adv/types.h
+++ b/net/batman-adv/types.h
@@ -63,7 +63,7 @@ struct hard_iface {
 /**
  *	orig_node - structure for orig_list maintaining nodes of mesh
  *	@primary_addr: hosts primary interface address
- *	@last_valid: when last packet from this node was received
+ *	@last_seen: when last packet from this node was received
  *	@bcast_seqno_reset: time when the broadcast seqno window was reset
  *	@batman_seqno_reset: time when the batman seqno window was reset
  *	@gw_flags: flags related to gateway class
@@ -84,7 +84,7 @@ struct orig_node {
 	struct neigh_node __rcu *router; /* rcu protected pointer */
 	unsigned long *bcast_own;
 	uint8_t *bcast_own_sum;
-	unsigned long last_valid;
+	unsigned long last_seen;
 	unsigned long bcast_seqno_reset;
 	unsigned long batman_seqno_reset;
 	uint8_t gw_flags;
@@ -134,7 +134,7 @@ struct gw_node {
 
 /**
  *	neigh_node
- *	@last_valid: when last packet via this neighbor was received
+ *	@last_seen: when last packet via this neighbor was received
  */
 struct neigh_node {
 	struct hlist_node list;
@@ -145,7 +145,7 @@ struct neigh_node {
 	uint8_t tq_avg;
 	uint8_t last_ttl;
 	struct list_head bonding_list;
-	unsigned long last_valid;
+	unsigned long last_seen;
 	DECLARE_BITMAP(real_bits, TQ_LOCAL_WINDOW_SIZE);
 	atomic_t refcount;
 	struct rcu_head rcu;
-- 
1.7.9.4

[PATCH 14/15] batman-adv: replace HZ calculations with jiffies_to_msecs()

[Antonio Quartulli]

From: Marek Lindner <lindner_marek@yahoo.de>

Signed-off-by: Marek Lindner <lindner_marek@yahoo.de>
Acked-by: Simon Wunderlich <siwu@hrz.tu-chemnitz.de>
Signed-off-by: Antonio Quartulli <ordex@autistici.org>
---
 net/batman-adv/bat_debugfs.c |    4 ++--
 net/batman-adv/originator.c  |   15 ++++++++++-----
 net/batman-adv/send.c        |    2 +-
 3 files changed, 13 insertions(+), 8 deletions(-)

diff --git a/net/batman-adv/bat_debugfs.c b/net/batman-adv/bat_debugfs.c
index 916380c..3b588f8 100644
--- a/net/batman-adv/bat_debugfs.c
+++ b/net/batman-adv/bat_debugfs.c
@@ -83,8 +83,8 @@ int debug_log(struct bat_priv *bat_priv, const char *fmt, ...)
 
 	va_start(args, fmt);
 	vscnprintf(tmp_log_buf, sizeof(tmp_log_buf), fmt, args);
-	fdebug_log(bat_priv->debug_log, "[%10lu] %s",
-		   (jiffies / HZ), tmp_log_buf);
+	fdebug_log(bat_priv->debug_log, "[%10u] %s",
+		   jiffies_to_msecs(jiffies), tmp_log_buf);
 	va_end(args);
 
 	return 0;
diff --git a/net/batman-adv/originator.c b/net/batman-adv/originator.c
index 936b2e6..a64536c 100644
--- a/net/batman-adv/originator.c
+++ b/net/batman-adv/originator.c
@@ -36,7 +36,8 @@ static void purge_orig(struct work_struct *work);
 static void start_purge_timer(struct bat_priv *bat_priv)
 {
 	INIT_DELAYED_WORK(&bat_priv->orig_work, purge_orig);
-	queue_delayed_work(bat_event_workqueue, &bat_priv->orig_work, 1 * HZ);
+	queue_delayed_work(bat_event_workqueue,
+			   &bat_priv->orig_work, msecs_to_jiffies(1000));
 }
 
 /* returns 1 if they are the same originator */
@@ -276,6 +277,7 @@ static bool purge_orig_neighbors(struct bat_priv *bat_priv,
 	struct hlist_node *node, *node_tmp;
 	struct neigh_node *neigh_node;
 	bool neigh_purged = false;
+	unsigned long last_seen;
 
 	*best_neigh_node = NULL;
 
@@ -290,6 +292,8 @@ static bool purge_orig_neighbors(struct bat_priv *bat_priv,
 		    (neigh_node->if_incoming->if_status == IF_NOT_IN_USE) ||
 		    (neigh_node->if_incoming->if_status == IF_TO_BE_REMOVED)) {
 
+			last_seen = neigh_node->last_seen;
+
 			if ((neigh_node->if_incoming->if_status ==
 								IF_INACTIVE) ||
 			    (neigh_node->if_incoming->if_status ==
@@ -302,9 +306,9 @@ static bool purge_orig_neighbors(struct bat_priv *bat_priv,
 					neigh_node->if_incoming->net_dev->name);
 			else
 				bat_dbg(DBG_BATMAN, bat_priv,
-					"neighbor timeout: originator %pM, neighbor: %pM, last_seen: %lu\n",
+					"neighbor timeout: originator %pM, neighbor: %pM, last_seen: %u\n",
 					orig_node->orig, neigh_node->addr,
-					(neigh_node->last_seen / HZ));
+					jiffies_to_msecs(last_seen));
 
 			neigh_purged = true;
 
@@ -329,8 +333,9 @@ static bool purge_orig_node(struct bat_priv *bat_priv,
 
 	if (has_timed_out(orig_node->last_seen, 2 * PURGE_TIMEOUT)) {
 		bat_dbg(DBG_BATMAN, bat_priv,
-			"Originator timeout: originator %pM, last_seen %lu\n",
-			orig_node->orig, (orig_node->last_seen / HZ));
+			"Originator timeout: originator %pM, last_seen %u\n",
+			orig_node->orig,
+			jiffies_to_msecs(orig_node->last_seen));
 		return true;
 	} else {
 		if (purge_orig_neighbors(bat_priv, orig_node,
diff --git a/net/batman-adv/send.c b/net/batman-adv/send.c
index 91eaa45..20f6e89 100644
--- a/net/batman-adv/send.c
+++ b/net/batman-adv/send.c
@@ -296,7 +296,7 @@ static void send_outstanding_bcast_packet(struct work_struct *work)
 	/* if we still have some more bcasts to send */
 	if (forw_packet->num_packets < 3) {
 		_add_bcast_packet_to_list(bat_priv, forw_packet,
-					  ((5 * HZ) / 1000));
+					  msecs_to_jiffies(5));
 		return;
 	}
 
-- 
1.7.9.4

[B.A.T.M.A.N.] [PATCH 14/15] batman-adv: replace HZ calculations with jiffies_to_msecs()

[Antonio Quartulli]

From: Marek Lindner <lindner_marek@yahoo.de>

Signed-off-by: Marek Lindner <lindner_marek@yahoo.de>
Acked-by: Simon Wunderlich <siwu@hrz.tu-chemnitz.de>
Signed-off-by: Antonio Quartulli <ordex@autistici.org>
---
 net/batman-adv/bat_debugfs.c |    4 ++--
 net/batman-adv/originator.c  |   15 ++++++++++-----
 net/batman-adv/send.c        |    2 +-
 3 files changed, 13 insertions(+), 8 deletions(-)

diff --git a/net/batman-adv/bat_debugfs.c b/net/batman-adv/bat_debugfs.c
index 916380c..3b588f8 100644
--- a/net/batman-adv/bat_debugfs.c
+++ b/net/batman-adv/bat_debugfs.c
@@ -83,8 +83,8 @@ int debug_log(struct bat_priv *bat_priv, const char *fmt, ...)
 
 	va_start(args, fmt);
 	vscnprintf(tmp_log_buf, sizeof(tmp_log_buf), fmt, args);
-	fdebug_log(bat_priv->debug_log, "[%10lu] %s",
-		   (jiffies / HZ), tmp_log_buf);
+	fdebug_log(bat_priv->debug_log, "[%10u] %s",
+		   jiffies_to_msecs(jiffies), tmp_log_buf);
 	va_end(args);
 
 	return 0;
diff --git a/net/batman-adv/originator.c b/net/batman-adv/originator.c
index 936b2e6..a64536c 100644
--- a/net/batman-adv/originator.c
+++ b/net/batman-adv/originator.c
@@ -36,7 +36,8 @@ static void purge_orig(struct work_struct *work);
 static void start_purge_timer(struct bat_priv *bat_priv)
 {
 	INIT_DELAYED_WORK(&bat_priv->orig_work, purge_orig);
-	queue_delayed_work(bat_event_workqueue, &bat_priv->orig_work, 1 * HZ);
+	queue_delayed_work(bat_event_workqueue,
+			   &bat_priv->orig_work, msecs_to_jiffies(1000));
 }
 
 /* returns 1 if they are the same originator */
@@ -276,6 +277,7 @@ static bool purge_orig_neighbors(struct bat_priv *bat_priv,
 	struct hlist_node *node, *node_tmp;
 	struct neigh_node *neigh_node;
 	bool neigh_purged = false;
+	unsigned long last_seen;
 
 	*best_neigh_node = NULL;
 
@@ -290,6 +292,8 @@ static bool purge_orig_neighbors(struct bat_priv *bat_priv,
 		    (neigh_node->if_incoming->if_status == IF_NOT_IN_USE) ||
 		    (neigh_node->if_incoming->if_status == IF_TO_BE_REMOVED)) {
 
+			last_seen = neigh_node->last_seen;
+
 			if ((neigh_node->if_incoming->if_status ==
 								IF_INACTIVE) ||
 			    (neigh_node->if_incoming->if_status ==
@@ -302,9 +306,9 @@ static bool purge_orig_neighbors(struct bat_priv *bat_priv,
 					neigh_node->if_incoming->net_dev->name);
 			else
 				bat_dbg(DBG_BATMAN, bat_priv,
-					"neighbor timeout: originator %pM, neighbor: %pM, last_seen: %lu\n",
+					"neighbor timeout: originator %pM, neighbor: %pM, last_seen: %u\n",
 					orig_node->orig, neigh_node->addr,
-					(neigh_node->last_seen / HZ));
+					jiffies_to_msecs(last_seen));
 
 			neigh_purged = true;
 
@@ -329,8 +333,9 @@ static bool purge_orig_node(struct bat_priv *bat_priv,
 
 	if (has_timed_out(orig_node->last_seen, 2 * PURGE_TIMEOUT)) {
 		bat_dbg(DBG_BATMAN, bat_priv,
-			"Originator timeout: originator %pM, last_seen %lu\n",
-			orig_node->orig, (orig_node->last_seen / HZ));
+			"Originator timeout: originator %pM, last_seen %u\n",
+			orig_node->orig,
+			jiffies_to_msecs(orig_node->last_seen));
 		return true;
 	} else {
 		if (purge_orig_neighbors(bat_priv, orig_node,
diff --git a/net/batman-adv/send.c b/net/batman-adv/send.c
index 91eaa45..20f6e89 100644
--- a/net/batman-adv/send.c
+++ b/net/batman-adv/send.c
@@ -296,7 +296,7 @@ static void send_outstanding_bcast_packet(struct work_struct *work)
 	/* if we still have some more bcasts to send */
 	if (forw_packet->num_packets < 3) {
 		_add_bcast_packet_to_list(bat_priv, forw_packet,
-					  ((5 * HZ) / 1000));
+					  msecs_to_jiffies(5));
 		return;
 	}
 
-- 
1.7.9.4

[PATCH 15/15] batman-adv: split neigh_new function into generic and batman iv specific parts

[Antonio Quartulli]

From: Marek Lindner <lindner_marek@yahoo.de>

Signed-off-by: Marek Lindner <lindner_marek@yahoo.de>
Acked-by: Simon Wunderlich <siwu@hrz.tu-chemnitz.de>
Signed-off-by: Antonio Quartulli <ordex@autistici.org>
---
 net/batman-adv/bat_iv_ogm.c |   40 ++++++++++++++++++++++++++++++++++------
 net/batman-adv/originator.c |   27 ++++++++++-----------------
 net/batman-adv/originator.h |    6 ++----
 3 files changed, 46 insertions(+), 27 deletions(-)

diff --git a/net/batman-adv/bat_iv_ogm.c b/net/batman-adv/bat_iv_ogm.c
index 8652a75..4baabf9 100644
--- a/net/batman-adv/bat_iv_ogm.c
+++ b/net/batman-adv/bat_iv_ogm.c
@@ -30,6 +30,32 @@
 #include "send.h"
 #include "bat_algo.h"
 
+static struct neigh_node *bat_iv_ogm_neigh_new(struct hard_iface *hard_iface,
+					       const uint8_t *neigh_addr,
+					       struct orig_node *orig_node,
+					       struct orig_node *orig_neigh,
+					       uint32_t seqno)
+{
+	struct neigh_node *neigh_node;
+
+	neigh_node = neigh_node_new(hard_iface, neigh_addr, seqno);
+	if (!neigh_node)
+		goto out;
+
+	INIT_LIST_HEAD(&neigh_node->bonding_list);
+	spin_lock_init(&neigh_node->tq_lock);
+
+	neigh_node->orig_node = orig_neigh;
+	neigh_node->if_incoming = hard_iface;
+
+	spin_lock_bh(&orig_node->neigh_list_lock);
+	hlist_add_head_rcu(&neigh_node->list, &orig_node->neigh_list);
+	spin_unlock_bh(&orig_node->neigh_list_lock);
+
+out:
+	return neigh_node;
+}
+
 static int bat_iv_ogm_iface_enable(struct hard_iface *hard_iface)
 {
 	struct batman_ogm_packet *batman_ogm_packet;
@@ -638,8 +664,9 @@ static void bat_iv_ogm_orig_update(struct bat_priv *bat_priv,
 		if (!orig_tmp)
 			goto unlock;
 
-		neigh_node = create_neighbor(orig_node, orig_tmp,
-					     ethhdr->h_source, if_incoming);
+		neigh_node = bat_iv_ogm_neigh_new(if_incoming, ethhdr->h_source,
+						  orig_node, orig_tmp,
+						  batman_ogm_packet->seqno);
 
 		orig_node_free_ref(orig_tmp);
 		if (!neigh_node)
@@ -764,10 +791,11 @@ static int bat_iv_ogm_calc_tq(struct orig_node *orig_node,
 	rcu_read_unlock();
 
 	if (!neigh_node)
-		neigh_node = create_neighbor(orig_neigh_node,
-					     orig_neigh_node,
-					     orig_neigh_node->orig,
-					     if_incoming);
+		neigh_node = bat_iv_ogm_neigh_new(if_incoming,
+						  orig_neigh_node->orig,
+						  orig_neigh_node,
+						  orig_neigh_node,
+						  batman_ogm_packet->seqno);
 
 	if (!neigh_node)
 		goto out;
diff --git a/net/batman-adv/originator.c b/net/batman-adv/originator.c
index a64536c..4432d64 100644
--- a/net/batman-adv/originator.c
+++ b/net/batman-adv/originator.c
@@ -86,35 +86,28 @@ struct neigh_node *orig_node_get_router(struct orig_node *orig_node)
 	return router;
 }
 
-struct neigh_node *create_neighbor(struct orig_node *orig_node,
-				   struct orig_node *orig_neigh_node,
-				   const uint8_t *neigh,
-				   struct hard_iface *if_incoming)
+struct neigh_node *neigh_node_new(struct hard_iface *hard_iface,
+				  const uint8_t *neigh_addr, uint32_t seqno)
 {
-	struct bat_priv *bat_priv = netdev_priv(if_incoming->soft_iface);
+	struct bat_priv *bat_priv = netdev_priv(hard_iface->soft_iface);
 	struct neigh_node *neigh_node;
 
-	bat_dbg(DBG_BATMAN, bat_priv,
-		"Creating new last-hop neighbor of originator\n");
-
 	neigh_node = kzalloc(sizeof(*neigh_node), GFP_ATOMIC);
 	if (!neigh_node)
-		return NULL;
+		goto out;
 
 	INIT_HLIST_NODE(&neigh_node->list);
-	INIT_LIST_HEAD(&neigh_node->bonding_list);
-	spin_lock_init(&neigh_node->tq_lock);
 
-	memcpy(neigh_node->addr, neigh, ETH_ALEN);
-	neigh_node->orig_node = orig_neigh_node;
-	neigh_node->if_incoming = if_incoming;
+	memcpy(neigh_node->addr, neigh_addr, ETH_ALEN);
 
 	/* extra reference for return */
 	atomic_set(&neigh_node->refcount, 2);
 
-	spin_lock_bh(&orig_node->neigh_list_lock);
-	hlist_add_head_rcu(&neigh_node->list, &orig_node->neigh_list);
-	spin_unlock_bh(&orig_node->neigh_list_lock);
+	bat_dbg(DBG_BATMAN, bat_priv,
+		"Creating new neighbor %pM, initial seqno %d\n",
+		neigh_addr, seqno);
+
+out:
 	return neigh_node;
 }
 
diff --git a/net/batman-adv/originator.h b/net/batman-adv/originator.h
index 3fe2eda..64c5d94 100644
--- a/net/batman-adv/originator.h
+++ b/net/batman-adv/originator.h
@@ -29,10 +29,8 @@ void originator_free(struct bat_priv *bat_priv);
 void purge_orig_ref(struct bat_priv *bat_priv);
 void orig_node_free_ref(struct orig_node *orig_node);
 struct orig_node *get_orig_node(struct bat_priv *bat_priv, const uint8_t *addr);
-struct neigh_node *create_neighbor(struct orig_node *orig_node,
-				   struct orig_node *orig_neigh_node,
-				   const uint8_t *neigh,
-				   struct hard_iface *if_incoming);
+struct neigh_node *neigh_node_new(struct hard_iface *hard_iface,
+				  const uint8_t *neigh_addr, uint32_t seqno);
 void neigh_node_free_ref(struct neigh_node *neigh_node);
 struct neigh_node *orig_node_get_router(struct orig_node *orig_node);
 int orig_seq_print_text(struct seq_file *seq, void *offset);
-- 
1.7.9.4

[B.A.T.M.A.N.] [PATCH 15/15] batman-adv: split neigh_new function into generic and batman iv specific parts

[Antonio Quartulli]

From: Marek Lindner <lindner_marek@yahoo.de>

Signed-off-by: Marek Lindner <lindner_marek@yahoo.de>
Acked-by: Simon Wunderlich <siwu@hrz.tu-chemnitz.de>
Signed-off-by: Antonio Quartulli <ordex@autistici.org>
---
 net/batman-adv/bat_iv_ogm.c |   40 ++++++++++++++++++++++++++++++++++------
 net/batman-adv/originator.c |   27 ++++++++++-----------------
 net/batman-adv/originator.h |    6 ++----
 3 files changed, 46 insertions(+), 27 deletions(-)

diff --git a/net/batman-adv/bat_iv_ogm.c b/net/batman-adv/bat_iv_ogm.c
index 8652a75..4baabf9 100644
--- a/net/batman-adv/bat_iv_ogm.c
+++ b/net/batman-adv/bat_iv_ogm.c
@@ -30,6 +30,32 @@
 #include "send.h"
 #include "bat_algo.h"
 
+static struct neigh_node *bat_iv_ogm_neigh_new(struct hard_iface *hard_iface,
+					       const uint8_t *neigh_addr,
+					       struct orig_node *orig_node,
+					       struct orig_node *orig_neigh,
+					       uint32_t seqno)
+{
+	struct neigh_node *neigh_node;
+
+	neigh_node = neigh_node_new(hard_iface, neigh_addr, seqno);
+	if (!neigh_node)
+		goto out;
+
+	INIT_LIST_HEAD(&neigh_node->bonding_list);
+	spin_lock_init(&neigh_node->tq_lock);
+
+	neigh_node->orig_node = orig_neigh;
+	neigh_node->if_incoming = hard_iface;
+
+	spin_lock_bh(&orig_node->neigh_list_lock);
+	hlist_add_head_rcu(&neigh_node->list, &orig_node->neigh_list);
+	spin_unlock_bh(&orig_node->neigh_list_lock);
+
+out:
+	return neigh_node;
+}
+
 static int bat_iv_ogm_iface_enable(struct hard_iface *hard_iface)
 {
 	struct batman_ogm_packet *batman_ogm_packet;
@@ -638,8 +664,9 @@ static void bat_iv_ogm_orig_update(struct bat_priv *bat_priv,
 		if (!orig_tmp)
 			goto unlock;
 
-		neigh_node = create_neighbor(orig_node, orig_tmp,
-					     ethhdr->h_source, if_incoming);
+		neigh_node = bat_iv_ogm_neigh_new(if_incoming, ethhdr->h_source,
+						  orig_node, orig_tmp,
+						  batman_ogm_packet->seqno);
 
 		orig_node_free_ref(orig_tmp);
 		if (!neigh_node)
@@ -764,10 +791,11 @@ static int bat_iv_ogm_calc_tq(struct orig_node *orig_node,
 	rcu_read_unlock();
 
 	if (!neigh_node)
-		neigh_node = create_neighbor(orig_neigh_node,
-					     orig_neigh_node,
-					     orig_neigh_node->orig,
-					     if_incoming);
+		neigh_node = bat_iv_ogm_neigh_new(if_incoming,
+						  orig_neigh_node->orig,
+						  orig_neigh_node,
+						  orig_neigh_node,
+						  batman_ogm_packet->seqno);
 
 	if (!neigh_node)
 		goto out;
diff --git a/net/batman-adv/originator.c b/net/batman-adv/originator.c
index a64536c..4432d64 100644
--- a/net/batman-adv/originator.c
+++ b/net/batman-adv/originator.c
@@ -86,35 +86,28 @@ struct neigh_node *orig_node_get_router(struct orig_node *orig_node)
 	return router;
 }
 
-struct neigh_node *create_neighbor(struct orig_node *orig_node,
-				   struct orig_node *orig_neigh_node,
-				   const uint8_t *neigh,
-				   struct hard_iface *if_incoming)
+struct neigh_node *neigh_node_new(struct hard_iface *hard_iface,
+				  const uint8_t *neigh_addr, uint32_t seqno)
 {
-	struct bat_priv *bat_priv = netdev_priv(if_incoming->soft_iface);
+	struct bat_priv *bat_priv = netdev_priv(hard_iface->soft_iface);
 	struct neigh_node *neigh_node;
 
-	bat_dbg(DBG_BATMAN, bat_priv,
-		"Creating new last-hop neighbor of originator\n");
-
 	neigh_node = kzalloc(sizeof(*neigh_node), GFP_ATOMIC);
 	if (!neigh_node)
-		return NULL;
+		goto out;
 
 	INIT_HLIST_NODE(&neigh_node->list);
-	INIT_LIST_HEAD(&neigh_node->bonding_list);
-	spin_lock_init(&neigh_node->tq_lock);
 
-	memcpy(neigh_node->addr, neigh, ETH_ALEN);
-	neigh_node->orig_node = orig_neigh_node;
-	neigh_node->if_incoming = if_incoming;
+	memcpy(neigh_node->addr, neigh_addr, ETH_ALEN);
 
 	/* extra reference for return */
 	atomic_set(&neigh_node->refcount, 2);
 
-	spin_lock_bh(&orig_node->neigh_list_lock);
-	hlist_add_head_rcu(&neigh_node->list, &orig_node->neigh_list);
-	spin_unlock_bh(&orig_node->neigh_list_lock);
+	bat_dbg(DBG_BATMAN, bat_priv,
+		"Creating new neighbor %pM, initial seqno %d\n",
+		neigh_addr, seqno);
+
+out:
 	return neigh_node;
 }
 
diff --git a/net/batman-adv/originator.h b/net/batman-adv/originator.h
index 3fe2eda..64c5d94 100644
--- a/net/batman-adv/originator.h
+++ b/net/batman-adv/originator.h
@@ -29,10 +29,8 @@ void originator_free(struct bat_priv *bat_priv);
 void purge_orig_ref(struct bat_priv *bat_priv);
 void orig_node_free_ref(struct orig_node *orig_node);
 struct orig_node *get_orig_node(struct bat_priv *bat_priv, const uint8_t *addr);
-struct neigh_node *create_neighbor(struct orig_node *orig_node,
-				   struct orig_node *orig_neigh_node,
-				   const uint8_t *neigh,
-				   struct hard_iface *if_incoming);
+struct neigh_node *neigh_node_new(struct hard_iface *hard_iface,
+				  const uint8_t *neigh_addr, uint32_t seqno);
 void neigh_node_free_ref(struct neigh_node *neigh_node);
 struct neigh_node *orig_node_get_router(struct orig_node *orig_node);
 int orig_seq_print_text(struct seq_file *seq, void *offset);
-- 
1.7.9.4

pull request: batman-adv 2012-04-29

[Antonio Quartulli]

Hello David,

this is the new version of my last pull request (issued on 2012-04-25).
This patchset is intended for net-next/linux-3.5.

Since last series I only removed the biggest_unsigned_int() macro.

Let me know if there is any other problem.

Thank you,
	Antonio

===============================================================

The following changes since commit 7a2a66a0ac1cf93d30869c4ecbfc71a2fda19397:

  Add linux-next specific files for 20120423 (2012-04-23 16:58:43 +1000)

are available in the git repository at:

  git://git.open-mesh.org/linux-merge.git tags/batman-adv-for-davem

for you to fetch changes up to 3ad345e2521ef41500cc0890a717656b975c4321:

  batman-adv: split neigh_new function into generic and batman iv specific parts (2012-04-29 10:50:10 +0200)

----------------------------------------------------------------
Included changes:

* a new feature has been introduced: D.A.T. (Distributed ARP Table). It is a
  mechanism based on DHT theory that creates a distributed (mesh network wide)
  ARP cache in order to speed up ARP resolutions in sparse wireless mesh
  networks.
* in order to satisfy DAT requirements a new unicast packet type, namely
  UNICAST_4ADDR, has been introduced. Backward compatibility has been kept: not
  updated nodes will simply drop the packet and ignore DAT mechanism.
* minor fixes and cleanups
* minor routing protocol API cleanups

----------------------------------------------------------------
Antonio Quartulli (9):
      batman-adv: add UNICAST_4ADDR packet type
      batman-adv: add a new log level for DAT debugging
      batman-adv: add biggest_unsigned_int(x) macro
      batman-adv: Distributed ARP Table - create DHT helper functions
      batman-adv: Distributed ARP Table - add ARP parsing functions
      batman-adv: Distributed ARP Table - add snooping functions for ARP messages
      batman-adv: Distributed ARP Table - increase default soft_iface ARP table timeout
      batman-adv: Distributed ARP Table - add compile option
      batman-adv: fix wrong dhcp option list browsing

Marek Lindner (6):
      batman-adv: introduce is_single_hop_neigh variable to increase readability
      batman-adv: introduce packet type handler array for incoming packets
      batman-adv: register batman ogm receive function during protocol init
      batman-adv: rename last_valid to last_seen
      batman-adv: replace HZ calculations with jiffies_to_msecs()
      batman-adv: split neigh_new function into generic and batman iv specific parts

 Documentation/networking/batman-adv.txt |    3 +-
 net/batman-adv/Kconfig                  |   10 +
 net/batman-adv/Makefile                 |    1 +
 net/batman-adv/bat_debugfs.c            |    4 +-
 net/batman-adv/bat_iv_ogm.c             |   95 +++--
 net/batman-adv/bat_sysfs.c              |    2 +-
 net/batman-adv/distributed-arp-table.c  |  605 +++++++++++++++++++++++++++++++
 net/batman-adv/distributed-arp-table.h  |  140 +++++++
 net/batman-adv/gateway_client.c         |    6 +-
 net/batman-adv/hard-interface.c         |  116 +-----
 net/batman-adv/main.c                   |  124 ++++++-
 net/batman-adv/main.h                   |   21 +-
 net/batman-adv/originator.c             |   52 +--
 net/batman-adv/originator.h             |    6 +-
 net/batman-adv/packet.h                 |   30 +-
 net/batman-adv/routing.c                |   30 +-
 net/batman-adv/routing.h                |    4 +-
 net/batman-adv/send.c                   |    6 +-
 net/batman-adv/soft-interface.c         |   17 +-
 net/batman-adv/types.h                  |   33 +-
 net/batman-adv/unicast.c                |  102 +++++-
 net/batman-adv/unicast.h                |   21 +-
 22 files changed, 1202 insertions(+), 226 deletions(-)
 create mode 100644 net/batman-adv/distributed-arp-table.c
 create mode 100644 net/batman-adv/distributed-arp-table.h

[PATCH 03/15] batman-adv: add biggest_unsigned_int(x) macro

[Antonio Quartulli]

in case of dynamic type variable, it could be needed to compute at compile time
its maximal value. This macro helps in doing that for unsigned integer types

Signed-off-by: Antonio Quartulli <ordex-GaUfNO9RBHfsrOwW+9ziJQ@public.gmane.org>
---
 net/batman-adv/main.h |    3 +++
 1 file changed, 3 insertions(+)

diff --git a/net/batman-adv/main.h b/net/batman-adv/main.h
index a2b18d0..d9ef4ca 100644
--- a/net/batman-adv/main.h
+++ b/net/batman-adv/main.h
@@ -221,6 +221,9 @@ static inline bool has_timed_out(unsigned long timestamp, unsigned int timeout)
 /* Returns the smallest signed integer in two's complement with the sizeof x */
 #define smallest_signed_int(x) (1u << (7u + 8u * (sizeof(x) - 1u)))
 
+/* Returns the biggest unsigned integer with the sizeof x */
+#define biggest_unsigned_int(x) (~(x)0)
+
 /* Checks if a sequence number x is a predecessor/successor of y.
  * they handle overflows/underflows and can correctly check for a
  * predecessor/successor unless the variable sequence number has grown by
-- 
1.7.9.4