linux/net/batman-adv/aggregation.c

294 lines
8.7 KiB
C
Raw Normal View History

/*
* Copyright (C) 2007-2011 B.A.T.M.A.N. contributors:
*
* Marek Lindner, Simon Wunderlich
*
* 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 "main.h"
batman-adv: improved client announcement mechanism The client announcement mechanism informs every mesh node in the network of any connected non-mesh client, in order to find the path towards that client from any given point in the mesh. The old implementation was based on the simple idea of appending a data buffer to each OGM containing all the client MAC addresses the node is serving. All other nodes can populate their global translation tables (table which links client MAC addresses to node addresses) using this MAC address buffer and linking it to the node's address contained in the OGM. A node that wants to contact a client has to lookup the node the client is connected to and its address in the global translation table. It is easy to understand that this implementation suffers from several issues: - big overhead (each and every OGM contains the entire list of connected clients) - high latencies for client route updates due to long OGM trip time and OGM losses The new implementation addresses these issues by appending client changes (new client joined or a client left) to the OGM instead of filling it with all the client addresses each time. In this way nodes can modify their global tables by means of "updates", thus reducing the overhead within the OGMs. To keep the entire network in sync each node maintains a translation table version number (ttvn) and a translation table checksum. These values are spread with the OGM to allow all the network participants to determine whether or not they need to update their translation table information. When a translation table lookup is performed in order to send a packet to a client attached to another node, the destination's ttvn is added to the payload packet. Forwarding nodes can compare the packet's ttvn with their destination's ttvn (this node could have a fresher information than the source) and re-route the packet if necessary. This greatly reduces the packet loss of clients roaming from one AP to the next. Signed-off-by: Antonio Quartulli <ordex@autistici.org> Signed-off-by: Marek Lindner <lindner_marek@yahoo.de> Signed-off-by: Sven Eckelmann <sven@narfation.org>
2011-04-27 20:27:44 +08:00
#include "translation-table.h"
#include "aggregation.h"
#include "send.h"
#include "routing.h"
#include "hard-interface.h"
/* return true if new_packet can be aggregated with forw_packet */
static bool can_aggregate_with(const struct batman_packet *new_batman_packet,
struct bat_priv *bat_priv,
int packet_len,
unsigned long send_time,
bool directlink,
const struct hard_iface *if_incoming,
const struct forw_packet *forw_packet)
{
struct batman_packet *batman_packet =
(struct batman_packet *)forw_packet->skb->data;
int aggregated_bytes = forw_packet->packet_len + packet_len;
struct hard_iface *primary_if = NULL;
bool res = false;
/**
* we can aggregate the current packet to this aggregated packet
* if:
*
* - the send time is within our MAX_AGGREGATION_MS time
* - the resulting packet wont be bigger than
* MAX_AGGREGATION_BYTES
*/
if (time_before(send_time, forw_packet->send_time) &&
time_after_eq(send_time + msecs_to_jiffies(MAX_AGGREGATION_MS),
forw_packet->send_time) &&
(aggregated_bytes <= MAX_AGGREGATION_BYTES)) {
/**
* check aggregation compatibility
* -> direct link packets are broadcasted on
* their interface only
* -> aggregate packet if the current packet is
* a "global" packet as well as the base
* packet
*/
primary_if = primary_if_get_selected(bat_priv);
if (!primary_if)
goto out;
/* packets without direct link flag and high TTL
* are flooded through the net */
if ((!directlink) &&
(!(batman_packet->flags & DIRECTLINK)) &&
(batman_packet->ttl != 1) &&
/* own packets originating non-primary
* interfaces leave only that interface */
((!forw_packet->own) ||
(forw_packet->if_incoming == primary_if))) {
res = true;
goto out;
}
/* if the incoming packet is sent via this one
* interface only - we still can aggregate */
if ((directlink) &&
(new_batman_packet->ttl == 1) &&
(forw_packet->if_incoming == if_incoming) &&
/* packets from direct neighbors or
* own secondary interface packets
* (= secondary interface packets in general) */
(batman_packet->flags & DIRECTLINK ||
(forw_packet->own &&
forw_packet->if_incoming != primary_if))) {
res = true;
goto out;
}
}
out:
if (primary_if)
hardif_free_ref(primary_if);
return res;
}
/* create a new aggregated packet and add this packet to it */
static void new_aggregated_packet(const unsigned char *packet_buff,
int packet_len, unsigned long send_time,
bool direct_link,
struct hard_iface *if_incoming,
int own_packet)
{
struct bat_priv *bat_priv = netdev_priv(if_incoming->soft_iface);
struct forw_packet *forw_packet_aggr;
unsigned char *skb_buff;
if (!atomic_inc_not_zero(&if_incoming->refcount))
return;
/* own packet should always be scheduled */
if (!own_packet) {
if (!atomic_dec_not_zero(&bat_priv->batman_queue_left)) {
bat_dbg(DBG_BATMAN, bat_priv,
"batman packet queue full\n");
goto out;
}
}
forw_packet_aggr = kmalloc(sizeof(*forw_packet_aggr), GFP_ATOMIC);
if (!forw_packet_aggr) {
if (!own_packet)
atomic_inc(&bat_priv->batman_queue_left);
goto out;
}
if ((atomic_read(&bat_priv->aggregated_ogms)) &&
(packet_len < MAX_AGGREGATION_BYTES))
forw_packet_aggr->skb = dev_alloc_skb(MAX_AGGREGATION_BYTES +
sizeof(struct ethhdr));
else
forw_packet_aggr->skb = dev_alloc_skb(packet_len +
sizeof(struct ethhdr));
if (!forw_packet_aggr->skb) {
if (!own_packet)
atomic_inc(&bat_priv->batman_queue_left);
kfree(forw_packet_aggr);
goto out;
}
skb_reserve(forw_packet_aggr->skb, sizeof(struct ethhdr));
INIT_HLIST_NODE(&forw_packet_aggr->list);
skb_buff = skb_put(forw_packet_aggr->skb, packet_len);
forw_packet_aggr->packet_len = packet_len;
memcpy(skb_buff, packet_buff, packet_len);
forw_packet_aggr->own = own_packet;
forw_packet_aggr->if_incoming = if_incoming;
forw_packet_aggr->num_packets = 0;
forw_packet_aggr->direct_link_flags = NO_FLAGS;
forw_packet_aggr->send_time = send_time;
/* save packet direct link flag status */
if (direct_link)
forw_packet_aggr->direct_link_flags |= 1;
/* add new packet to packet list */
spin_lock_bh(&bat_priv->forw_bat_list_lock);
hlist_add_head(&forw_packet_aggr->list, &bat_priv->forw_bat_list);
spin_unlock_bh(&bat_priv->forw_bat_list_lock);
/* start timer for this packet */
INIT_DELAYED_WORK(&forw_packet_aggr->delayed_work,
send_outstanding_bat_packet);
queue_delayed_work(bat_event_workqueue,
&forw_packet_aggr->delayed_work,
send_time - jiffies);
return;
out:
hardif_free_ref(if_incoming);
}
/* aggregate a new packet into the existing aggregation */
static void aggregate(struct forw_packet *forw_packet_aggr,
const unsigned char *packet_buff, int packet_len,
bool direct_link)
{
unsigned char *skb_buff;
skb_buff = skb_put(forw_packet_aggr->skb, packet_len);
memcpy(skb_buff, packet_buff, packet_len);
forw_packet_aggr->packet_len += packet_len;
forw_packet_aggr->num_packets++;
/* save packet direct link flag status */
if (direct_link)
forw_packet_aggr->direct_link_flags |=
(1 << forw_packet_aggr->num_packets);
}
void add_bat_packet_to_list(struct bat_priv *bat_priv,
unsigned char *packet_buff, int packet_len,
struct hard_iface *if_incoming, int own_packet,
unsigned long send_time)
{
/**
* _aggr -> pointer to the packet we want to aggregate with
* _pos -> pointer to the position in the queue
*/
struct forw_packet *forw_packet_aggr = NULL, *forw_packet_pos = NULL;
struct hlist_node *tmp_node;
struct batman_packet *batman_packet =
(struct batman_packet *)packet_buff;
bool direct_link = batman_packet->flags & DIRECTLINK ? 1 : 0;
/* find position for the packet in the forward queue */
spin_lock_bh(&bat_priv->forw_bat_list_lock);
/* own packets are not to be aggregated */
if ((atomic_read(&bat_priv->aggregated_ogms)) && (!own_packet)) {
hlist_for_each_entry(forw_packet_pos, tmp_node,
&bat_priv->forw_bat_list, list) {
if (can_aggregate_with(batman_packet,
bat_priv,
packet_len,
send_time,
direct_link,
if_incoming,
forw_packet_pos)) {
forw_packet_aggr = forw_packet_pos;
break;
}
}
}
/* nothing to aggregate with - either aggregation disabled or no
* suitable aggregation packet found */
if (!forw_packet_aggr) {
/* the following section can run without the lock */
spin_unlock_bh(&bat_priv->forw_bat_list_lock);
/**
* if we could not aggregate this packet with one of the others
* we hold it back for a while, so that it might be aggregated
* later on
*/
if ((!own_packet) &&
(atomic_read(&bat_priv->aggregated_ogms)))
send_time += msecs_to_jiffies(MAX_AGGREGATION_MS);
new_aggregated_packet(packet_buff, packet_len,
send_time, direct_link,
if_incoming, own_packet);
} else {
aggregate(forw_packet_aggr,
packet_buff, packet_len,
direct_link);
spin_unlock_bh(&bat_priv->forw_bat_list_lock);
}
}
/* unpack the aggregated packets and process them one by one */
void receive_aggr_bat_packet(const struct ethhdr *ethhdr,
unsigned char *packet_buff, int packet_len,
struct hard_iface *if_incoming)
{
struct batman_packet *batman_packet;
int buff_pos = 0;
unsigned char *tt_buff;
batman_packet = (struct batman_packet *)packet_buff;
do {
batman-adv: improved client announcement mechanism The client announcement mechanism informs every mesh node in the network of any connected non-mesh client, in order to find the path towards that client from any given point in the mesh. The old implementation was based on the simple idea of appending a data buffer to each OGM containing all the client MAC addresses the node is serving. All other nodes can populate their global translation tables (table which links client MAC addresses to node addresses) using this MAC address buffer and linking it to the node's address contained in the OGM. A node that wants to contact a client has to lookup the node the client is connected to and its address in the global translation table. It is easy to understand that this implementation suffers from several issues: - big overhead (each and every OGM contains the entire list of connected clients) - high latencies for client route updates due to long OGM trip time and OGM losses The new implementation addresses these issues by appending client changes (new client joined or a client left) to the OGM instead of filling it with all the client addresses each time. In this way nodes can modify their global tables by means of "updates", thus reducing the overhead within the OGMs. To keep the entire network in sync each node maintains a translation table version number (ttvn) and a translation table checksum. These values are spread with the OGM to allow all the network participants to determine whether or not they need to update their translation table information. When a translation table lookup is performed in order to send a packet to a client attached to another node, the destination's ttvn is added to the payload packet. Forwarding nodes can compare the packet's ttvn with their destination's ttvn (this node could have a fresher information than the source) and re-route the packet if necessary. This greatly reduces the packet loss of clients roaming from one AP to the next. Signed-off-by: Antonio Quartulli <ordex@autistici.org> Signed-off-by: Marek Lindner <lindner_marek@yahoo.de> Signed-off-by: Sven Eckelmann <sven@narfation.org>
2011-04-27 20:27:44 +08:00
/* network to host order for our 32bit seqno and the
orig_interval */
batman_packet->seqno = ntohl(batman_packet->seqno);
batman-adv: improved client announcement mechanism The client announcement mechanism informs every mesh node in the network of any connected non-mesh client, in order to find the path towards that client from any given point in the mesh. The old implementation was based on the simple idea of appending a data buffer to each OGM containing all the client MAC addresses the node is serving. All other nodes can populate their global translation tables (table which links client MAC addresses to node addresses) using this MAC address buffer and linking it to the node's address contained in the OGM. A node that wants to contact a client has to lookup the node the client is connected to and its address in the global translation table. It is easy to understand that this implementation suffers from several issues: - big overhead (each and every OGM contains the entire list of connected clients) - high latencies for client route updates due to long OGM trip time and OGM losses The new implementation addresses these issues by appending client changes (new client joined or a client left) to the OGM instead of filling it with all the client addresses each time. In this way nodes can modify their global tables by means of "updates", thus reducing the overhead within the OGMs. To keep the entire network in sync each node maintains a translation table version number (ttvn) and a translation table checksum. These values are spread with the OGM to allow all the network participants to determine whether or not they need to update their translation table information. When a translation table lookup is performed in order to send a packet to a client attached to another node, the destination's ttvn is added to the payload packet. Forwarding nodes can compare the packet's ttvn with their destination's ttvn (this node could have a fresher information than the source) and re-route the packet if necessary. This greatly reduces the packet loss of clients roaming from one AP to the next. Signed-off-by: Antonio Quartulli <ordex@autistici.org> Signed-off-by: Marek Lindner <lindner_marek@yahoo.de> Signed-off-by: Sven Eckelmann <sven@narfation.org>
2011-04-27 20:27:44 +08:00
batman_packet->tt_crc = ntohs(batman_packet->tt_crc);
tt_buff = packet_buff + buff_pos + BAT_PACKET_LEN;
batman-adv: improved client announcement mechanism The client announcement mechanism informs every mesh node in the network of any connected non-mesh client, in order to find the path towards that client from any given point in the mesh. The old implementation was based on the simple idea of appending a data buffer to each OGM containing all the client MAC addresses the node is serving. All other nodes can populate their global translation tables (table which links client MAC addresses to node addresses) using this MAC address buffer and linking it to the node's address contained in the OGM. A node that wants to contact a client has to lookup the node the client is connected to and its address in the global translation table. It is easy to understand that this implementation suffers from several issues: - big overhead (each and every OGM contains the entire list of connected clients) - high latencies for client route updates due to long OGM trip time and OGM losses The new implementation addresses these issues by appending client changes (new client joined or a client left) to the OGM instead of filling it with all the client addresses each time. In this way nodes can modify their global tables by means of "updates", thus reducing the overhead within the OGMs. To keep the entire network in sync each node maintains a translation table version number (ttvn) and a translation table checksum. These values are spread with the OGM to allow all the network participants to determine whether or not they need to update their translation table information. When a translation table lookup is performed in order to send a packet to a client attached to another node, the destination's ttvn is added to the payload packet. Forwarding nodes can compare the packet's ttvn with their destination's ttvn (this node could have a fresher information than the source) and re-route the packet if necessary. This greatly reduces the packet loss of clients roaming from one AP to the next. Signed-off-by: Antonio Quartulli <ordex@autistici.org> Signed-off-by: Marek Lindner <lindner_marek@yahoo.de> Signed-off-by: Sven Eckelmann <sven@narfation.org>
2011-04-27 20:27:44 +08:00
receive_bat_packet(ethhdr, batman_packet, tt_buff, if_incoming);
buff_pos += BAT_PACKET_LEN +
tt_len(batman_packet->tt_num_changes);
batman_packet = (struct batman_packet *)
(packet_buff + buff_pos);
} while (aggregated_packet(buff_pos, packet_len,
batman-adv: improved client announcement mechanism The client announcement mechanism informs every mesh node in the network of any connected non-mesh client, in order to find the path towards that client from any given point in the mesh. The old implementation was based on the simple idea of appending a data buffer to each OGM containing all the client MAC addresses the node is serving. All other nodes can populate their global translation tables (table which links client MAC addresses to node addresses) using this MAC address buffer and linking it to the node's address contained in the OGM. A node that wants to contact a client has to lookup the node the client is connected to and its address in the global translation table. It is easy to understand that this implementation suffers from several issues: - big overhead (each and every OGM contains the entire list of connected clients) - high latencies for client route updates due to long OGM trip time and OGM losses The new implementation addresses these issues by appending client changes (new client joined or a client left) to the OGM instead of filling it with all the client addresses each time. In this way nodes can modify their global tables by means of "updates", thus reducing the overhead within the OGMs. To keep the entire network in sync each node maintains a translation table version number (ttvn) and a translation table checksum. These values are spread with the OGM to allow all the network participants to determine whether or not they need to update their translation table information. When a translation table lookup is performed in order to send a packet to a client attached to another node, the destination's ttvn is added to the payload packet. Forwarding nodes can compare the packet's ttvn with their destination's ttvn (this node could have a fresher information than the source) and re-route the packet if necessary. This greatly reduces the packet loss of clients roaming from one AP to the next. Signed-off-by: Antonio Quartulli <ordex@autistici.org> Signed-off-by: Marek Lindner <lindner_marek@yahoo.de> Signed-off-by: Sven Eckelmann <sven@narfation.org>
2011-04-27 20:27:44 +08:00
batman_packet->tt_num_changes));
}