B.A.T.M.A.N May 2012

b.a.t.m.a.n@lists.open-mesh.org

20 participants
51 discussions

[B.A.T.M.A.N.] pull request: batman-adv 2012-04-29
by Antonio Quartulli 24 May '12

24 May '12

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

5 25

[B.A.T.M.A.N.] [PATCH] batman-adv: Avoid gcc inline problems with vararg functions
by Sven Eckelmann 22 May '12

22 May '12

gcc doesn't support the inlining of functions using vararg. Sparse fails to build the source when having such functions marked using the keyword "inline". This problem was introduced in 3b896e321c49e3cbbfd540f537c007db883c3900 Signed-off-by: Sven Eckelmann <sven(a)narfation.org> --- main.h | 20 +++++++++++++------- 1 file changed, 13 insertions(+), 7 deletions(-) diff --git a/main.h b/main.h index 977de45..4721c86 100644 --- a/main.h +++ b/main.h @@ -196,6 +196,17 @@ static inline void batadv_vdbg(int type, struct bat_priv *bat_priv, batadv_debug_log(bat_priv, fmt, args); } +__printf(3, 4) +static inline void batadv_dbg(int type, struct bat_priv *bat_priv, + const char *fmt, ...) +{ + va_list args; + + va_start(args, fmt); + batadv_vdbg(type, bat_priv, fmt, args); + va_end(args); +} + #else /* !CONFIG_BATMAN_ADV_DEBUG */ static inline void batadv_vdbg(int type __always_unused, @@ -205,19 +216,14 @@ static inline void batadv_vdbg(int type __always_unused, { } -#endif - __printf(3, 4) static inline void batadv_dbg(int type, struct bat_priv *bat_priv, const char *fmt, ...) { - va_list args; - - va_start(args, fmt); - batadv_vdbg(type, bat_priv, fmt, args); - va_end(args); } +#endif + #define batadv_info(net_dev, fmt, arg...) \ do { \ struct net_device *_netdev = (net_dev); \ -- 1.7.10

2 2

[B.A.T.M.A.N.] [PATCH] batman-adv: Revert Distributed ARP Table
by Sven Eckelmann 19 May '12

19 May '12

David S. Miller did not like the idea of batman-adv modifying and reading the ARP table and refused to pull this feature. It has to be rewritten after an acceptable solution was found. This reverts following commits: * batman-adv: add UNICAST_4ADDR packet type * batman-adv: add a new log level for DAT debugging * 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 Signed-off-by: Sven Eckelmann <sven(a)narfation.org> --- This patch is for next... I repeat: this patch is for the branch next Makefile | 2 - Makefile.kbuild | 1 - README | 3 +- README.external | 1 - bat_sysfs.c | 2 +- distributed-arp-table.c | 605 ----------------------------------------------- distributed-arp-table.h | 140 ----------- gen-compat-autoconf.sh | 1 - hard-interface.c | 3 - main.c | 2 - main.h | 15 +- originator.c | 2 - packet.h | 30 +-- routing.c | 8 +- send.c | 4 - soft-interface.c | 17 +- types.h | 22 -- unicast.c | 101 ++------ unicast.h | 21 +- 19 files changed, 33 insertions(+), 947 deletions(-) delete mode 100644 distributed-arp-table.c delete mode 100644 distributed-arp-table.h diff --git a/Makefile b/Makefile index ac84fba..08f8c39 100644 --- a/Makefile +++ b/Makefile @@ -23,8 +23,6 @@ export CONFIG_BATMAN_ADV_DEBUG=n # B.A.T.M.A.N. bridge loop avoidance: export CONFIG_BATMAN_ADV_BLA=y -# B.A.T.M.A.N. distributed ARP table: -export CONFIG_BATMAN_ADV_DAT=y PWD:=$(shell pwd) KERNELPATH ?= /lib/modules/$(shell uname -r)/build diff --git a/Makefile.kbuild b/Makefile.kbuild index ad002cd..6d5c194 100644 --- a/Makefile.kbuild +++ b/Makefile.kbuild @@ -24,7 +24,6 @@ batman-adv-y += bat_iv_ogm.o batman-adv-y += bat_sysfs.o batman-adv-y += bitarray.o batman-adv-$(CONFIG_BATMAN_ADV_BLA) += bridge_loop_avoidance.o -batman-adv-$(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/README b/README index 82c075f..75a5923 100644 --- a/README +++ b/README @@ -202,8 +202,7 @@ 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 -16 - Enable messaged related to DAT, ARP snooping and parsing -31 - Enable all messages +15 - 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/README.external b/README.external index 874ee7f..da5afd4 100644 --- a/README.external +++ b/README.external @@ -37,7 +37,6 @@ module). Available options and their possible values are * CONFIG_BATMAN_ADV_DEBUG=[y|n*] (B.A.T.M.A.N. debugging) * CONFIG_BATMAN_ADV_BLA=[y*|n] (B.A.T.M.A.N. bridge loop avoidance) - * CONFIG_BATMAN_ADV_DAT=[y*|n] (B.A.T.M.A.N. Distributed ARP Table) e.g., debugging can be enabled by diff --git a/bat_sysfs.c b/bat_sysfs.c index acb2640..5bc7b66 100644 --- a/bat_sysfs.c +++ b/bat_sysfs.c @@ -445,7 +445,7 @@ BAT_ATTR_SIF_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_SIF_UINT(log_level, S_IRUGO | S_IWUSR, 0, 31, NULL); +BAT_ATTR_SIF_UINT(log_level, S_IRUGO | S_IWUSR, 0, 15, NULL); #endif static struct bat_attribute *mesh_attrs[] = { diff --git a/distributed-arp-table.c b/distributed-arp-table.c deleted file mode 100644 index 6935be9..0000000 --- a/distributed-arp-table.c +++ /dev/null @@ -1,605 +0,0 @@ -/* - * Copyright (C) 2011-2012 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> -/* needed to use arp_tbl */ -#include <net/arp.h> -#include <linux/inetdevice.h> - -#include "main.h" -#include "distributed-arp-table.h" -#include "hard-interface.h" -#include "originator.h" -#include "send.h" -#include "types.h" -#include "translation-table.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, - 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; -} - -/* 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 - */ -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; -} - -/* 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; -} - -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/distributed-arp-table.h b/distributed-arp-table.h deleted file mode 100644 index 6c0acde..0000000 --- a/distributed-arp-table.h +++ /dev/null @@ -1,140 +0,0 @@ -/* - * Copyright (C) 2011-2012 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_ - -#ifdef CONFIG_BATMAN_ADV_DAT - -#include "types.h" -#include "originator.h" - -#include <linux/if_arp.h> - -#define DAT_ADDR_MAX ((dat_addr_t)~(dat_addr_t)0) - -#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)) - -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); -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 - */ -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); -} - -#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/gen-compat-autoconf.sh b/gen-compat-autoconf.sh index 7ea42aa..7cf621b 100755 --- a/gen-compat-autoconf.sh +++ b/gen-compat-autoconf.sh @@ -38,7 +38,6 @@ gen_config() { # write config variables gen_config 'CONFIG_BATMAN_ADV_DEBUG' ${CONFIG_BATMAN_ADV_DEBUG:="n"} >> "${TMP}" gen_config 'CONFIG_BATMAN_ADV_BLA' ${CONFIG_BATMAN_ADV_BLA:="y"} >> "${TMP}" -gen_config 'CONFIG_BATMAN_ADV_DAT' ${CONFIG_BATMAN_ADV_DAT:="y"} >> "${TMP}" # only regenerate compat-autoconf.h when config was changed diff "${TMP}" "${TARGET}" > /dev/null 2>&1 || cp "${TMP}" "${TARGET}" diff --git a/hard-interface.c b/hard-interface.c index 843caa7..ce78c6d 100644 --- a/hard-interface.c +++ b/hard-interface.c @@ -20,7 +20,6 @@ */ #include "main.h" -#include "distributed-arp-table.h" #include "hard-interface.h" #include "soft-interface.h" #include "send.h" @@ -113,8 +112,6 @@ 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/main.c b/main.c index 9137aa0..083a299 100644 --- a/main.c +++ b/main.c @@ -268,8 +268,6 @@ static void recv_handler_init(void) /* 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 */ diff --git a/main.h b/main.h index 426a0d5..19e5881 100644 --- a/main.h +++ b/main.h @@ -67,15 +67,6 @@ #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 -/* 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 */ #define VIS_INTERVAL 5000 /* 5 seconds */ @@ -120,9 +111,6 @@ 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 @@ -136,8 +124,7 @@ 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_DAT = 1 << 4, /* snooped arp messages / dat operations */ - DBG_ALL = 31 + DBG_ALL = 15 }; /* Kernel headers */ diff --git a/originator.c b/originator.c index c6a00b3..4114794 100644 --- a/originator.c +++ b/originator.c @@ -20,7 +20,6 @@ */ #include "main.h" -#include "distributed-arp-table.h" #include "originator.h" #include "hash.h" #include "translation-table.h" @@ -221,7 +220,6 @@ 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/packet.h b/packet.h index 307dbb3..0ee1af7 100644 --- a/packet.h +++ b/packet.h @@ -25,22 +25,14 @@ #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_UNICAST_4ADDR = 0x09 -}; - -enum bat_subtype { - BAT_P_DATA = 0x01, - BAT_P_DAT_DHT_GET = 0x02, - BAT_P_DAT_DHT_PUT = 0x03, - BAT_P_DAT_CACHE_REPLY = 0x04 + 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 }; /* this file is included by batctl which needs these defines */ @@ -168,12 +160,6 @@ 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/routing.c b/routing.c index 4f2059f..840e2c6 100644 --- a/routing.c +++ b/routing.c @@ -962,18 +962,14 @@ 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/send.c b/send.c index 815cc9c..f47299f 100644 --- a/send.c +++ b/send.c @@ -20,7 +20,6 @@ */ #include "main.h" -#include "distributed-arp-table.h" #include "send.h" #include "routing.h" #include "translation-table.h" @@ -275,9 +274,6 @@ 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/soft-interface.c b/soft-interface.c index b56dafd..6e2530b 100644 --- a/soft-interface.c +++ b/soft-interface.c @@ -22,7 +22,6 @@ #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" @@ -137,7 +136,6 @@ 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; @@ -199,9 +197,6 @@ 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; @@ -221,7 +216,7 @@ 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, brd_delay); + add_bcast_packet_to_list(bat_priv, skb, 1); /* a copy is stored in the bcast list, therefore removing * the original skb. */ @@ -235,8 +230,6 @@ 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; @@ -269,12 +262,6 @@ 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); @@ -381,8 +368,6 @@ 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); diff --git a/types.h b/types.h index 6474324..61308e8 100644 --- a/types.h +++ b/types.h @@ -27,17 +27,6 @@ #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. - * - * *Please be careful: dat_addr_t must be UNSIGNED* - */ -#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) : \ @@ -78,9 +67,6 @@ 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; @@ -235,9 +221,6 @@ 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; }; @@ -411,9 +394,4 @@ struct bat_algo_ops { void (*bat_ogm_emit)(struct forw_packet *forw_packet); }; -struct dht_candidate { - int type; - struct orig_node *orig_node; -}; - #endif /* _NET_BATMAN_ADV_TYPES_H_ */ diff --git a/unicast.c b/unicast.c index e5c7999..74175c2 100644 --- a/unicast.c +++ b/unicast.c @@ -283,78 +283,13 @@ out: return ret; } -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); -} - -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) +int unicast_send_skb(struct sk_buff *skb, struct bat_priv *bat_priv) { 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 */ @@ -368,6 +303,7 @@ int unicast_generic_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(): @@ -375,26 +311,26 @@ find_router: * - increases neigh_nodes refcount if found. */ neigh_node = find_router(bat_priv, orig_node, NULL); + if (!neigh_node) goto out; - 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 - */ + if (my_skb_head_push(skb, sizeof(*unicast_packet)) < 0) 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); + /* inform the destination node that we are still missing a correct route * for this client. The destination will receive this packet and will * try to reroute it because the ttvn contained in the header is less @@ -403,9 +339,7 @@ find_router: if (tt_global_client_is_roaming(bat_priv, ethhdr->h_dest)) unicast_packet->ttvn = unicast_packet->ttvn - 1; - /* fragmentation mechanism only works for UNICAST (now) */ - if (packet_type == BAT_UNICAST && - atomic_read(&bat_priv->fragmentation) && + if (atomic_read(&bat_priv->fragmentation) && data_len + sizeof(*unicast_packet) > neigh_node->if_incoming->net_dev->mtu) { /* send frag skb decreases ttl */ @@ -417,6 +351,7 @@ find_router: send_skb_packet(skb, neigh_node->if_incoming, neigh_node->addr); ret = 0; + goto out; out: if (neigh_node) diff --git a/unicast.h b/unicast.h index e15aa62..a9faf6b 100644 --- a/unicast.h +++ b/unicast.h @@ -30,28 +30,9 @@ 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[]); -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); - -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) { -- 1.7.10

3 5

[B.A.T.M.A.N.] Remaining cleanup/prefixing patches
by Sven Eckelmann 18 May '12

18 May '12

Hi, it seems that just too many changes were done and now all were committed in the order or the version that I send to the mailinglist. Therefore, I would ask to ignore all my remaining patches and wait for the the complete list of patches not yet committed. I will send them to this mailing list in some hours. Thanks, Sven

2 27

[B.A.T.M.A.N.] [PATCH 01/18] batman-adv: Prefix bat_debugfs local static functions with batadv_
by Sven Eckelmann 16 May '12

16 May '12

All non-static symbols of batman-adv were prefixed with batadv_ to avoid collisions with other symbols of the kernel. Other symbols of batman-adv should use the same prefix to keep the naming scheme consistent. Signed-off-by: Sven Eckelmann <sven(a)narfation.org> --- bat_debugfs.c | 143 +++++++++++++++++++++++++++++---------------------------- 1 file changed, 72 insertions(+), 71 deletions(-) diff --git a/bat_debugfs.c b/bat_debugfs.c index cd636db..004d94f 100644 --- a/bat_debugfs.c +++ b/bat_debugfs.c @@ -32,25 +32,25 @@ #include "icmp_socket.h" #include "bridge_loop_avoidance.h" -static struct dentry *bat_debugfs; +static struct dentry *batadv_debugfs; #ifdef CONFIG_BATMAN_ADV_DEBUG -#define LOG_BUFF_MASK (log_buff_len-1) +#define LOG_BUFF_MASK (batadv_log_buff_len - 1) #define LOG_BUFF(idx) (debug_log->log_buff[(idx) & LOG_BUFF_MASK]) -static int log_buff_len = LOG_BUF_LEN; +static int batadv_log_buff_len = LOG_BUF_LEN; -static void emit_log_char(struct debug_log *debug_log, char c) +static void batadv_emit_log_char(struct debug_log *debug_log, char c) { LOG_BUFF(debug_log->log_end) = c; debug_log->log_end++; - if (debug_log->log_end - debug_log->log_start > log_buff_len) - debug_log->log_start = debug_log->log_end - log_buff_len; + if (debug_log->log_end - debug_log->log_start > batadv_log_buff_len) + debug_log->log_start = debug_log->log_end - batadv_log_buff_len; } __printf(2, 3) -static int fdebug_log(struct debug_log *debug_log, const char *fmt, ...) +static int batadv_fdebug_log(struct debug_log *debug_log, const char *fmt, ...) { va_list args; static char debug_log_buf[256]; @@ -65,7 +65,7 @@ static int fdebug_log(struct debug_log *debug_log, const char *fmt, ...) va_end(args); for (p = debug_log_buf; *p != 0; p++) - emit_log_char(debug_log, *p); + batadv_emit_log_char(debug_log, *p); spin_unlock_bh(&debug_log->lock); @@ -81,14 +81,14 @@ int batadv_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, "[%10u] %s", - jiffies_to_msecs(jiffies), tmp_log_buf); + batadv_fdebug_log(bat_priv->debug_log, "[%10u] %s", + jiffies_to_msecs(jiffies), tmp_log_buf); va_end(args); return 0; } -static int log_open(struct inode *inode, struct file *file) +static int batadv_log_open(struct inode *inode, struct file *file) { nonseekable_open(inode, file); file->private_data = inode->i_private; @@ -96,14 +96,14 @@ static int log_open(struct inode *inode, struct file *file) return 0; } -static int log_release(struct inode *inode, struct file *file) +static int batadv_log_release(struct inode *inode, struct file *file) { batadv_dec_module_count(); return 0; } -static ssize_t log_read(struct file *file, char __user *buf, - size_t count, loff_t *ppos) +static ssize_t batadv_log_read(struct file *file, char __user *buf, + size_t count, loff_t *ppos) { struct bat_priv *bat_priv = file->private_data; struct debug_log *debug_log = bat_priv->debug_log; @@ -156,7 +156,7 @@ static ssize_t log_read(struct file *file, char __user *buf, return error; } -static unsigned int log_poll(struct file *file, poll_table *wait) +static unsigned int batadv_log_poll(struct file *file, poll_table *wait) { struct bat_priv *bat_priv = file->private_data; struct debug_log *debug_log = bat_priv->debug_log; @@ -169,15 +169,15 @@ static unsigned int log_poll(struct file *file, poll_table *wait) return 0; } -static const struct file_operations log_fops = { - .open = log_open, - .release = log_release, - .read = log_read, - .poll = log_poll, +static const struct file_operations batadv_log_fops = { + .open = batadv_log_open, + .release = batadv_log_release, + .read = batadv_log_read, + .poll = batadv_log_poll, .llseek = no_llseek, }; -static int debug_log_setup(struct bat_priv *bat_priv) +static int batadv_debug_log_setup(struct bat_priv *bat_priv) { struct dentry *d; @@ -192,7 +192,8 @@ static int debug_log_setup(struct bat_priv *bat_priv) init_waitqueue_head(&bat_priv->debug_log->queue_wait); d = debugfs_create_file("log", S_IFREG | S_IRUSR, - bat_priv->debug_dir, bat_priv, &log_fops); + bat_priv->debug_dir, bat_priv, + &batadv_log_fops); if (d) goto err; @@ -202,49 +203,49 @@ err: return 1; } -static void debug_log_cleanup(struct bat_priv *bat_priv) +static void batadv_debug_log_cleanup(struct bat_priv *bat_priv) { kfree(bat_priv->debug_log); bat_priv->debug_log = NULL; } #else /* CONFIG_BATMAN_ADV_DEBUG */ -static int debug_log_setup(struct bat_priv *bat_priv) +static int batadv_debug_log_setup(struct bat_priv *bat_priv) { bat_priv->debug_log = NULL; return 0; } -static void debug_log_cleanup(struct bat_priv *bat_priv) +static void batadv_debug_log_cleanup(struct bat_priv *bat_priv) { return; } #endif -static int bat_algorithms_open(struct inode *inode, struct file *file) +static int batadv_algorithms_open(struct inode *inode, struct file *file) { return single_open(file, batadv_algo_seq_print_text, NULL); } -static int originators_open(struct inode *inode, struct file *file) +static int batadv_originators_open(struct inode *inode, struct file *file) { struct net_device *net_dev = (struct net_device *)inode->i_private; return single_open(file, batadv_orig_seq_print_text, net_dev); } -static int gateways_open(struct inode *inode, struct file *file) +static int batadv_gateways_open(struct inode *inode, struct file *file) { struct net_device *net_dev = (struct net_device *)inode->i_private; return single_open(file, batadv_gw_client_seq_print_text, net_dev); } -static int transtable_global_open(struct inode *inode, struct file *file) +static int batadv_transtable_global_open(struct inode *inode, struct file *file) { struct net_device *net_dev = (struct net_device *)inode->i_private; return single_open(file, batadv_tt_global_seq_print_text, net_dev); } #ifdef CONFIG_BATMAN_ADV_BLA -static int bla_claim_table_open(struct inode *inode, struct file *file) +static int batadv_bla_claim_table_open(struct inode *inode, struct file *file) { struct net_device *net_dev = (struct net_device *)inode->i_private; return single_open(file, batadv_bla_claim_table_seq_print_text, @@ -252,13 +253,13 @@ static int bla_claim_table_open(struct inode *inode, struct file *file) } #endif -static int transtable_local_open(struct inode *inode, struct file *file) +static int batadv_transtable_local_open(struct inode *inode, struct file *file) { struct net_device *net_dev = (struct net_device *)inode->i_private; return single_open(file, batadv_tt_local_seq_print_text, net_dev); } -static int vis_data_open(struct inode *inode, struct file *file) +static int batadv_vis_data_open(struct inode *inode, struct file *file) { struct net_device *net_dev = (struct net_device *)inode->i_private; return single_open(file, batadv_vis_seq_print_text, net_dev); @@ -269,37 +270,37 @@ struct bat_debuginfo { const struct file_operations fops; }; -#define BAT_DEBUGINFO(_name, _mode, _open) \ -struct bat_debuginfo bat_debuginfo_##_name = { \ - .attr = { .name = __stringify(_name), \ - .mode = _mode, }, \ - .fops = { .owner = THIS_MODULE, \ - .open = _open, \ - .read = seq_read, \ - .llseek = seq_lseek, \ - .release = single_release, \ - } \ +#define BAT_DEBUGINFO(_name, _mode, _open) \ +struct bat_debuginfo batadv_debuginfo_##_name = { \ + .attr = { .name = __stringify(_name), \ + .mode = _mode, }, \ + .fops = { .owner = THIS_MODULE, \ + .open = _open, \ + .read = seq_read, \ + .llseek = seq_lseek, \ + .release = single_release, \ + } \ }; -static BAT_DEBUGINFO(routing_algos, S_IRUGO, bat_algorithms_open); -static BAT_DEBUGINFO(originators, S_IRUGO, originators_open); -static BAT_DEBUGINFO(gateways, S_IRUGO, gateways_open); -static BAT_DEBUGINFO(transtable_global, S_IRUGO, transtable_global_open); +static BAT_DEBUGINFO(routing_algos, S_IRUGO, batadv_algorithms_open); +static BAT_DEBUGINFO(originators, S_IRUGO, batadv_originators_open); +static BAT_DEBUGINFO(gateways, S_IRUGO, batadv_gateways_open); +static BAT_DEBUGINFO(transtable_global, S_IRUGO, batadv_transtable_global_open); #ifdef CONFIG_BATMAN_ADV_BLA -static BAT_DEBUGINFO(bla_claim_table, S_IRUGO, bla_claim_table_open); +static BAT_DEBUGINFO(bla_claim_table, S_IRUGO, batadv_bla_claim_table_open); #endif -static BAT_DEBUGINFO(transtable_local, S_IRUGO, transtable_local_open); -static BAT_DEBUGINFO(vis_data, S_IRUGO, vis_data_open); +static BAT_DEBUGINFO(transtable_local, S_IRUGO, batadv_transtable_local_open); +static BAT_DEBUGINFO(vis_data, S_IRUGO, batadv_vis_data_open); -static struct bat_debuginfo *mesh_debuginfos[] = { - &bat_debuginfo_originators, - &bat_debuginfo_gateways, - &bat_debuginfo_transtable_global, +static struct bat_debuginfo *batadv_mesh_debuginfos[] = { + &batadv_debuginfo_originators, + &batadv_debuginfo_gateways, + &batadv_debuginfo_transtable_global, #ifdef CONFIG_BATMAN_ADV_BLA - &bat_debuginfo_bla_claim_table, + &batadv_debuginfo_bla_claim_table, #endif - &bat_debuginfo_transtable_local, - &bat_debuginfo_vis_data, + &batadv_debuginfo_transtable_local, + &batadv_debuginfo_vis_data, NULL, }; @@ -308,17 +309,17 @@ void batadv_debugfs_init(void) struct bat_debuginfo *bat_debug; struct dentry *file; - bat_debugfs = debugfs_create_dir(DEBUGFS_BAT_SUBDIR, NULL); - if (bat_debugfs == ERR_PTR(-ENODEV)) - bat_debugfs = NULL; + batadv_debugfs = debugfs_create_dir(DEBUGFS_BAT_SUBDIR, NULL); + if (batadv_debugfs == ERR_PTR(-ENODEV)) + batadv_debugfs = NULL; - if (!bat_debugfs) + if (!batadv_debugfs) goto out; - bat_debug = &bat_debuginfo_routing_algos; + bat_debug = &batadv_debuginfo_routing_algos; file = debugfs_create_file(bat_debug->attr.name, S_IFREG | bat_debug->attr.mode, - bat_debugfs, NULL, &bat_debug->fops); + batadv_debugfs, NULL, &bat_debug->fops); if (!file) pr_err("Can't add debugfs file: %s\n", bat_debug->attr.name); @@ -328,9 +329,9 @@ out: void batadv_debugfs_destroy(void) { - if (bat_debugfs) { - debugfs_remove_recursive(bat_debugfs); - bat_debugfs = NULL; + if (batadv_debugfs) { + debugfs_remove_recursive(batadv_debugfs); + batadv_debugfs = NULL; } } @@ -340,17 +341,17 @@ int batadv_debugfs_add_meshif(struct net_device *dev) struct bat_debuginfo **bat_debug; struct dentry *file; - if (!bat_debugfs) + if (!batadv_debugfs) goto out; - bat_priv->debug_dir = debugfs_create_dir(dev->name, bat_debugfs); + bat_priv->debug_dir = debugfs_create_dir(dev->name, batadv_debugfs); if (!bat_priv->debug_dir) goto out; batadv_socket_setup(bat_priv); - debug_log_setup(bat_priv); + batadv_debug_log_setup(bat_priv); - for (bat_debug = mesh_debuginfos; *bat_debug; ++bat_debug) { + for (bat_debug = batadv_mesh_debuginfos; *bat_debug; ++bat_debug) { file = debugfs_create_file(((*bat_debug)->attr).name, S_IFREG | ((*bat_debug)->attr).mode, bat_priv->debug_dir, @@ -378,9 +379,9 @@ void batadv_debugfs_del_meshif(struct net_device *dev) { struct bat_priv *bat_priv = netdev_priv(dev); - debug_log_cleanup(bat_priv); + batadv_debug_log_cleanup(bat_priv); - if (bat_debugfs) { + if (batadv_debugfs) { debugfs_remove_recursive(bat_priv->debug_dir); bat_priv->debug_dir = NULL; } -- 1.7.10

2 29

[B.A.T.M.A.N.] Some problems with batman-adv
by Jochen Jägers 15 May '12

15 May '12

Hello, i try to set up a ad-hoc-mesh network with batman-adv. There are three test-nodes with following configuration: Nvidia Tegra 2 Dual Core ARM-Cortex A9 Processor with Linux 2.6.36.2 USB-Wifi Dongle with Ralink 5370 Chipset Ralink rt5370sta driver kernel module batman-adv 2012.1.0 (also tried with 2012.0.0) I've set up the ad-hoc network with WPA-Encryption. This is working fine. I can access all nodes if i configure the ra0 (ralink wifi) interface with static ip addresses. I've done this for testing purpose only. Normally only the bat0 interface has an ip address. After wireless connection is established I activate the batman-adv interface. Example from first node: modprobe batman-adv batctl if add ra0 ifconfig bat0 up ifconfig bat0 192.168.50.1 netmask 255.255.255.0 ifconfig bat0 mtu 1472 After booting up all nodes "batctl o" returns: # batctl o [B.A.T.M.A.N. adv 2012.1.0, MainIF/MAC: ra0/80:1f:02:38:7a:bf (bat0)] Originator last-seen (#/255) Nexthop [outgoingIF]: Potential nexthops ... 00:0f:c9:22:00:17 0.020s (248) 00:0f:c9:22:00:17 [ ra0]: 00:0f:c9:22:11:78 (245) 00:0f:c9:22:00:17 (248) 00:0f:c9:22:11:78 0.560s (255) 00:0f:c9:22:11:78 [ ra0]: 00:0f:c9:22:00:17 (230) 00:0f:c9:22:11:78 (255) So far everything looks fine. But if i try to connect to the other nodes i mostly get no answers. No ping reply, no ssh or http connection. But sometimes it works for a couple of minutes. I also tried to ping with batctl but got unstable results: first time everything looks good: # batctl ping 00:0f:c9:22:00:17 PING 00:0f:c9:22:00:17 (00:0f:c9:22:00:17) 20(48) bytes of data 20 bytes from 00:0f:c9:22:00:17 icmp_seq=1 ttl=50 time=6.37 ms 20 bytes from 00:0f:c9:22:00:17 icmp_seq=2 ttl=50 time=13.22 ms 20 bytes from 00:0f:c9:22:00:17 icmp_seq=3 ttl=50 time=9.08 ms ^C--- 00:0f:c9:22:00:17 ping statistics --- 3 packets transmitted, 3 received, 0% packet loss rtt min/avg/max/mdev = 6.373/9.560/13.225/2.818 ms another try a few moments later did not work: # batctl ping 00:0f:c9:22:00:17 PING 00:0f:c9:22:00:17 (00:0f:c9:22:00:17) 20(48) bytes of data Reply from host 00:0f:c9:22:00:17 timed out Reply from host 00:0f:c9:22:00:17 timed out Reply from host 00:0f:c9:22:00:17 timed out Reply from host 00:0f:c9:22:00:17 timed out ^C--- 00:0f:c9:22:00:17 ping statistics --- 4 packets transmitted, 0 received, 100% packet loss rtt min/avg/max/mdev = 0.000/0.000/0.000/0.000 ms i tried many different things (changing MTU, Wifi channel, encryption SSID, batman-adv version,...) but the result was the same everytime. Some further information: # iwconfig lo no wireless extensions. dummy0 no wireless extensions. usb0 no wireless extensions. eth0 no wireless extensions. ra0 Ralink STA ESSID:"ceR{)|n$v#pp|fJX" Nickname:"RT2870STA" Mode:Ad-Hoc Frequency=2.442 GHz Cell: DE:BB:8C:08:B8:EC Bit Rate=36 Mb/s RTS thr:off Fragment thr:off Encryption key:D83E-521A-8B9F-7C95-3ACE-D477-****-**** Security mode:open Link Quality=52/100 Signal level:-71 dBm Noise level:-71 dBm Rx invalid nwid:0 Rx invalid crypt:0 Rx invalid frag:0 Tx excessive retries:0 Invalid misc:0 Missed beacon:0 bat0 no wireless extensions. #ifconfig ra0 ra0 Link encap:Ethernet HWaddr 80:1F:02:38:7A:BF UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 RX packets:0 errors:11 dropped:0 overruns:0 frame:0 TX packets:0 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:1000 RX bytes:258771 (252.7 KiB) TX bytes:13544 (13.2 KiB) #ifconfig bat0 bat0 Link encap:Ethernet HWaddr 02:DA:94:22:E1:67 inet addr:192.168.50.60 Bcast:192.168.50.255 Mask:255.255.255.0 UP BROADCAST RUNNING MULTICAST MTU:1472 Metric:1 RX packets:5 errors:0 dropped:0 overruns:0 frame:0 TX packets:11 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:0 RX bytes:434 (434.0 B) TX bytes:686 (686.0 B) Do you have any ideas what I can try to solve my problem or where to search for a solution? Thanks for any response. Greetings, -- Jochen Jägers Dr. Riedel Automatisierungstechnik GmbH Greifswalder Str. 4 10405 Berlin Tel. 030/428431-48 Fax. 030/428431-99 http://www.riedel-at.de/ Geschäftsführer: Dr. Manfred Riedel Amtsgericht Berlin Charlottenburg, HRB 39094 Umsatzsteuer-Identifikationsnummer DE 137 196 925

4 5

[B.A.T.M.A.N.] [PATCH 1/8] batman-adv: Move hash_ipv4 to distributed-arp-table.c
by Sven Eckelmann 15 May '12

15 May '12

hash_ipv4 is only used in distributed-arp-table.c and therefore can be placed inside this file instead of the header. Signed-off-by: Sven Eckelmann <sven(a)narfation.org> --- distributed-arp-table.c | 24 +++++++++++++++++++++++- distributed-arp-table.h | 22 ---------------------- 2 files changed, 23 insertions(+), 23 deletions(-) diff --git a/distributed-arp-table.c b/distributed-arp-table.c index e678ec4..4894a85 100644 --- a/distributed-arp-table.c +++ b/distributed-arp-table.c @@ -32,6 +32,28 @@ #include "translation-table.h" #include "unicast.h" +/* hash function to choose an entry in a hash table of given size. + * hash algorithm from http://en.wikipedia.org/wiki/Hash_table + */ +static uint32_t batadv_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; +} + #ifdef CONFIG_BATMAN_ADV_DEBUG static void bat_dbg_arp(struct bat_priv *bat_priv, struct sk_buff *skb, @@ -214,7 +236,7 @@ static struct dht_candidate *dht_select_candidates(struct bat_priv *bat_priv, if (!res) return NULL; - ip_key = (dat_addr_t)hash_ipv4(&ip_dst, DAT_ADDR_MAX); + ip_key = (dat_addr_t)batadv_hash_ipv4(&ip_dst, DAT_ADDR_MAX); bat_dbg(DBG_DAT, bat_priv, "dht_select_candidates(): IP=%pI4 hash(IP)=%u\n", &ip_dst, diff --git a/distributed-arp-table.h b/distributed-arp-table.h index 1ba7f01..ee5200a 100644 --- a/distributed-arp-table.h +++ b/distributed-arp-table.h @@ -49,28 +49,6 @@ bool batadv_dat_drop_broadcast_packet(struct bat_priv *bat_priv, struct forw_packet *forw_packet); void batadv_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 - */ -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 batadv_dat_init_orig_node_dht_addr(struct orig_node *orig_node) { -- 1.7.10

3 18

[B.A.T.M.A.N.] pull request: batman-adv 2012-05-14
by Antonio Quartulli 14 May '12

14 May '12

Hello David, here is our last set of changes intended for net-next/linux-3.5. In this patchset there are: 1) several minor cleanups and fixes 2) an optimisation that avoid to linearise the whole packet when not needed 3) an optimisation for client traffic rerouting after a roaming event. Patches that modify the whole naming schema will be sent later for merging with linux-3.6. Thank you, Antonio The following changes since commit f96a8a0b78548c0ec06b0b4b438db6ee895d67e9: igb: Add Support for new i210/i211 devices. (2012-05-12 13:17:13 -0700) 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 521251f2f5fa16747cc21e71580e404af855d140: batman-adv: unset the TT_CLIENT_PENDING flag if the new local entry already exists (2012-05-14 09:05:08 +0200) ---------------------------------------------------------------- Included changes: * an improvement to avoid to linearise the whole received packet when not needed * an improvement for client traffic rerouting after roaming * a fix for the local translation table state-machine * minor cleanups and fixes ---------------------------------------------------------------- Antonio Quartulli (3): batman-adv: avoid skb_linearise() if not needed batman-adv: improve unicast packet (re)routing batman-adv: unset the TT_CLIENT_PENDING flag if the new local entry already exists Marek Lindner (2): batman-adv: prepare lq_update_lock to be shared among different protocols batman-adv: refactor window_protected to avoid unnecessary return statement Sven Eckelmann (3): batman-adv: use shorter pr_warn instead of pr_warning batman-adv: Start new development cycle batman-adv: README cleanups Documentation/networking/batman-adv.txt | 11 +++++------ net/batman-adv/bat_iv_ogm.c | 9 ++++----- net/batman-adv/hard-interface.c | 6 +++--- net/batman-adv/main.h | 2 +- net/batman-adv/originator.c | 1 + net/batman-adv/routing.c | 27 ++++++++++++++++----------- net/batman-adv/send.c | 4 ++-- net/batman-adv/translation-table.c | 21 +++++++++++++++++++++ net/batman-adv/translation-table.h | 2 ++ net/batman-adv/types.h | 2 +- net/batman-adv/unicast.c | 8 ++++++++ 11 files changed, 64 insertions(+), 29 deletions(-)

2 9

[B.A.T.M.A.N.] Possible OGM forwarding misbehavior
by Daniele Furlan 14 May '12

14 May '12

Hi all, incidentally I noticed that with the modification that introduced the 'NOT_BEST_NEXT_HOP' flag only OGM coming from best next hop neighbour are forwarded in case of multi-hop originators. The problem is that in the routing code it is forwarded only the first OGM received with a fresh sequence number (that could be received from a sub-optimal router). This means that if the OGM from best next hop neighbour is not received as the first OGM, it is not forwarded and I think this can be a problem. Maybe it can be better to maintain also an index of the last FORWARDED sequence number instead of checking the last RECEIVED sequence number. Regards -- Daniele Furlan

2 1

[B.A.T.M.A.N.] [PATCHv3 01/23] batman-adv: Fix namespace for hash_set_lock_class
by Sven Eckelmann 14 May '12

14 May '12

This function was introduced in 2c9aa3b4c37bd55148f5f25212c89e7bf5f28497 without a proper namespace prefix. Signed-off-by: Sven Eckelmann <sven(a)narfation.org> --- bridge_loop_avoidance.c | 7 ++++--- hash.c | 3 ++- hash.h | 3 ++- 3 files changed, 8 insertions(+), 5 deletions(-) diff --git a/bridge_loop_avoidance.c b/bridge_loop_avoidance.c index 89e9721..314e37b 100644 --- a/bridge_loop_avoidance.c +++ b/bridge_loop_avoidance.c @@ -1172,9 +1172,10 @@ int bla_init(struct bat_priv *bat_priv) if (!bat_priv->claim_hash || !bat_priv->backbone_hash) return -ENOMEM; - hash_set_lock_class(bat_priv->claim_hash, &claim_hash_lock_class_key); - hash_set_lock_class(bat_priv->backbone_hash, - &backbone_hash_lock_class_key); + batadv_hash_set_lock_class(bat_priv->claim_hash, + &claim_hash_lock_class_key); + batadv_hash_set_lock_class(bat_priv->backbone_hash, + &backbone_hash_lock_class_key); bat_dbg(DBG_BLA, bat_priv, "bla hashes initialized\n"); diff --git a/hash.c b/hash.c index 4578c20..a7dcff0 100644 --- a/hash.c +++ b/hash.c @@ -70,7 +70,8 @@ free_hash: return NULL; } -void hash_set_lock_class(struct hashtable_t *hash, struct lock_class_key *key) +void batadv_hash_set_lock_class(struct hashtable_t *hash, + struct lock_class_key *key) { uint32_t i; diff --git a/hash.h b/hash.h index 2e0409a..3d67ce4 100644 --- a/hash.h +++ b/hash.h @@ -46,7 +46,8 @@ struct hashtable_t { struct hashtable_t *hash_new(uint32_t size); /* set class key for all locks */ -void hash_set_lock_class(struct hashtable_t *hash, struct lock_class_key *key); +void batadv_hash_set_lock_class(struct hashtable_t *hash, + struct lock_class_key *key); /* free only the hashtable and the hash itself. */ void hash_destroy(struct hashtable_t *hash); -- 1.7.10

2 47

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

B.A.T.M.A.N May 2012