* [B.A.T.M.A.N.] [PATCHv14 0/6] Basic Multicast Optimizations
@ 2014-02-15 16:47 Linus Lüssing
2014-02-15 16:47 ` [B.A.T.M.A.N.] [PATCHv14 1/6] batman-adv: Multicast Listener Announcements via Translation Table Linus Lüssing
` (5 more replies)
0 siblings, 6 replies; 14+ messages in thread
From: Linus Lüssing @ 2014-02-15 16:47 UTC (permalink / raw)
To: b.a.t.m.a.n
This is the fourteenth revision of the basic multicast optimization patches.
Changes in v14 include:
* fixed compilation for CONFIG_BATMAN_ADV_MCAST=n (issue was introduced by v13)
* fixed potential null pointer exception (moved ethhdr assignment in
batadv_mcast_forw_mode() to after the check methods / pskb_may_pull()s)
* removed flag-list-"helper" functions
* removed methods which were performing both counting and orig_node retrieval
in batadv_mcast_forw_mode(), instead doing that in two logical, seperate
sections in batadv_mcast_forw_mode()
Cheers, Linus
^ permalink raw reply [flat|nested] 14+ messages in thread
* [B.A.T.M.A.N.] [PATCHv14 1/6] batman-adv: Multicast Listener Announcements via Translation Table
2014-02-15 16:47 [B.A.T.M.A.N.] [PATCHv14 0/6] Basic Multicast Optimizations Linus Lüssing
@ 2014-02-15 16:47 ` Linus Lüssing
2014-02-16 6:23 ` Marek Lindner
2014-02-15 16:47 ` [B.A.T.M.A.N.] [PATCHv14 2/6] batman-adv: introduce capability initialization bitfield Linus Lüssing
` (4 subsequent siblings)
5 siblings, 1 reply; 14+ messages in thread
From: Linus Lüssing @ 2014-02-15 16:47 UTC (permalink / raw)
To: b.a.t.m.a.n
With this patch a node which has no bridge interface on top of its soft
interface announces its local multicast listeners via the translation
table.
Signed-off-by: Linus Lüssing <linus.luessing@web.de>
---
Makefile | 2 +
Makefile.kbuild | 1 +
compat.h | 37 ++++++++
gen-compat-autoconf.sh | 1 +
main.c | 6 ++
main.h | 1 +
multicast.c | 242 ++++++++++++++++++++++++++++++++++++++++++++++++
multicast.h | 38 ++++++++
translation-table.c | 22 ++++-
types.h | 24 +++++
10 files changed, 370 insertions(+), 4 deletions(-)
create mode 100644 multicast.c
create mode 100644 multicast.h
diff --git a/Makefile b/Makefile
index 9a7cbae..4f6c30a 100644
--- a/Makefile
+++ b/Makefile
@@ -25,6 +25,8 @@ export CONFIG_BATMAN_ADV_BLA=y
export CONFIG_BATMAN_ADV_DAT=y
# B.A.T.M.A.N network coding (catwoman):
export CONFIG_BATMAN_ADV_NC=n
+# B.A.T.M.A.N. multicast optimizations:
+export CONFIG_BATMAN_ADV_MCAST=y
PWD:=$(shell pwd)
KERNELPATH ?= /lib/modules/$(shell uname -r)/build
diff --git a/Makefile.kbuild b/Makefile.kbuild
index 42df18f..eb7d8c0 100644
--- a/Makefile.kbuild
+++ b/Makefile.kbuild
@@ -36,3 +36,4 @@ batman-adv-y += send.o
batman-adv-y += soft-interface.o
batman-adv-y += sysfs.o
batman-adv-y += translation-table.o
+batman-adv-$(CONFIG_BATMAN_ADV_MCAST) += multicast.o
diff --git a/compat.h b/compat.h
index 12bc8d8..7a3d235 100644
--- a/compat.h
+++ b/compat.h
@@ -103,6 +103,24 @@ static inline struct sk_buff *netdev_alloc_skb_ip_align(struct net_device *dev,
#define pr_warn pr_warning
+#undef netdev_for_each_mc_addr
+#define netdev_for_each_mc_addr(mclist, dev) \
+ for (mclist = (struct batadv_dev_addr_list *)dev->mc_list; mclist; \
+ mclist = (struct batadv_dev_addr_list *)mclist->next)
+
+/* Note, that this breaks the usage of the normal 'struct netdev_hw_addr'
+ * for kernels < 2.6.35 in batman-adv!
+ */
+#define netdev_hw_addr batadv_dev_addr_list
+struct batadv_dev_addr_list {
+ struct dev_addr_list *next;
+ u8 addr[MAX_ADDR_LEN];
+ u8 da_addrlen;
+ u8 da_synced;
+ int da_users;
+ int da_gusers;
+};
+
#endif /* < KERNEL_VERSION(2, 6, 35) */
@@ -143,6 +161,14 @@ static inline int batadv_param_set_copystring(const char *val,
#define addr_assign_type ifindex
#define NET_ADDR_RANDOM 0
+#define netdev_master_upper_dev_get_rcu(dev) \
+ NULL; \
+ if (dev->br_port ? 1 : 0) { \
+ rcu_read_unlock(); \
+ dev_hold(dev); \
+ return dev; \
+ }
+
#endif /* < KERNEL_VERSION(2, 6, 36) */
#if LINUX_VERSION_CODE < KERNEL_VERSION(2, 6, 37)
@@ -343,6 +369,17 @@ static int __batadv_interface_tx(struct sk_buff *skb, \
pos && ({ n = pos->member.next; 1; }); \
pos = hlist_entry_safe(n, typeof(*pos), member))
+#ifndef netdev_master_upper_dev_get_rcu
+#define netdev_master_upper_dev_get_rcu(dev) \
+ NULL; \
+ if (dev->priv_flags & IFF_BRIDGE_PORT) { \
+ rcu_read_unlock(); \
+ dev_hold(dev); \
+ return dev; \
+ }
+
+#endif /* netdev_master_upper_dev_get_rcu */
+
#endif /* < KERNEL_VERSION(3, 9, 0) */
#if LINUX_VERSION_CODE < KERNEL_VERSION(3, 10, 0)
diff --git a/gen-compat-autoconf.sh b/gen-compat-autoconf.sh
index 78573e4..f625b6f 100755
--- a/gen-compat-autoconf.sh
+++ b/gen-compat-autoconf.sh
@@ -39,6 +39,7 @@ gen_config() {
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}"
+gen_config 'CONFIG_BATMAN_ADV_MCAST' ${CONFIG_BATMAN_ADV_MCAST:="y"} >> "${TMP}"
gen_config 'CONFIG_BATMAN_ADV_NC' ${CONFIG_BATMAN_ADV_NC:="n"} >> "${TMP}"
# only regenerate compat-autoconf.h when config was changed
diff --git a/main.c b/main.c
index fbeaebd..58e98c8 100644
--- a/main.c
+++ b/main.c
@@ -34,6 +34,7 @@
#include "gateway_client.h"
#include "bridge_loop_avoidance.h"
#include "distributed-arp-table.h"
+#include "multicast.h"
#include "gateway_common.h"
#include "hash.h"
#include "bat_algo.h"
@@ -120,6 +121,9 @@ int batadv_mesh_init(struct net_device *soft_iface)
INIT_LIST_HEAD(&bat_priv->tt.changes_list);
INIT_LIST_HEAD(&bat_priv->tt.req_list);
INIT_LIST_HEAD(&bat_priv->tt.roam_list);
+#ifdef CONFIG_BATMAN_ADV_MCAST
+ INIT_HLIST_HEAD(&bat_priv->mcast.mla_list);
+#endif
INIT_HLIST_HEAD(&bat_priv->tvlv.container_list);
INIT_HLIST_HEAD(&bat_priv->tvlv.handler_list);
INIT_HLIST_HEAD(&bat_priv->softif_vlan_list);
@@ -169,6 +173,8 @@ void batadv_mesh_free(struct net_device *soft_iface)
batadv_dat_free(bat_priv);
batadv_bla_free(bat_priv);
+ batadv_mcast_free(bat_priv);
+
/* Free the TT and the originator tables only after having terminated
* all the other depending components which may use these structures for
* their purposes.
diff --git a/main.h b/main.h
index 34e85a2..dc6b4a2 100644
--- a/main.h
+++ b/main.h
@@ -176,6 +176,7 @@ enum batadv_uev_type {
#include <linux/percpu.h>
#include <linux/slab.h>
#include <net/sock.h> /* struct sock */
+#include <net/addrconf.h> /* ipv6 address stuff */
#include <net/rtnetlink.h>
#include <linux/jiffies.h>
#include <linux/seq_file.h>
diff --git a/multicast.c b/multicast.c
new file mode 100644
index 0000000..66e270c
--- /dev/null
+++ b/multicast.c
@@ -0,0 +1,242 @@
+/* Copyright (C) 2014 B.A.T.M.A.N. contributors:
+ *
+ * Linus Lüssing
+ *
+ * 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.
+ */
+
+#include "main.h"
+#include "multicast.h"
+#include "originator.h"
+#include "hard-interface.h"
+#include "translation-table.h"
+
+/**
+ * batadv_mcast_mla_softif_get - get softif multicast listeners
+ * @dev: the device to collect multicast addresses from
+ * @mcast_list: a list to put found addresses into
+ *
+ * Collect multicast addresses of the local multicast listeners
+ * on the given soft interface, dev, in the given mcast_list.
+ *
+ * Return -ENOMEM on memory allocation error or the number of
+ * items added to the mcast_list otherwise.
+ */
+static int batadv_mcast_mla_softif_get(struct net_device *dev,
+ struct hlist_head *mcast_list)
+{
+ struct netdev_hw_addr *mc_list_entry;
+ struct batadv_hw_addr *new;
+ int ret = 0;
+
+ netif_addr_lock_bh(dev);
+ netdev_for_each_mc_addr(mc_list_entry, dev) {
+ new = kmalloc(sizeof(*new), GFP_ATOMIC);
+ if (!new) {
+ ret = -ENOMEM;
+ break;
+ }
+
+ memcpy(&new->addr, &mc_list_entry->addr, ETH_ALEN);
+ hlist_add_head(&new->list, mcast_list);
+ ret++;
+ }
+ netif_addr_unlock_bh(dev);
+
+ return ret;
+}
+
+/**
+ * batadv_mcast_mla_is_duplicate - check whether an address is in a list
+ * @mcast_addr: the multicast address to check
+ * @mcast_list: the list with multicast addresses to search in
+ *
+ * Return true if the given address is already in the given list.
+ * Otherwise returns false.
+ */
+static bool batadv_mcast_mla_is_duplicate(uint8_t *mcast_addr,
+ struct hlist_head *mcast_list)
+{
+ struct batadv_hw_addr *mcast_entry;
+
+ hlist_for_each_entry(mcast_entry, mcast_list, list)
+ if (batadv_compare_eth(mcast_entry->addr, mcast_addr))
+ return true;
+
+ return false;
+}
+
+/**
+ * batadv_mcast_mla_list_free - free a list of multicast addresses
+ * @mcast_list: the list to free
+ *
+ * Remove and free all items in the given mcast_list.
+ */
+static void batadv_mcast_mla_list_free(struct hlist_head *mcast_list)
+{
+ struct batadv_hw_addr *mcast_entry;
+ struct hlist_node *tmp;
+
+ hlist_for_each_entry_safe(mcast_entry, tmp, mcast_list, list) {
+ hlist_del(&mcast_entry->list);
+ kfree(mcast_entry);
+ }
+}
+
+/**
+ * batadv_mcast_mla_tt_retract - clean up multicast listener announcements
+ * @bat_priv: the bat priv with all the soft interface information
+ * @mcast_list: a list of addresses which should _not_ be removed
+ *
+ * Retract the announcement of any multicast listener from the
+ * translation table except the ones listed in the given mcast_list.
+ *
+ * If mcast_list is NULL then all are retracted.
+ */
+static void batadv_mcast_mla_tt_retract(struct batadv_priv *bat_priv,
+ struct hlist_head *mcast_list)
+{
+ struct batadv_hw_addr *mcast_entry;
+ struct hlist_node *tmp;
+
+ hlist_for_each_entry_safe(mcast_entry, tmp, &bat_priv->mcast.mla_list,
+ list) {
+ if (mcast_list &&
+ batadv_mcast_mla_is_duplicate(mcast_entry->addr,
+ mcast_list))
+ continue;
+
+ batadv_tt_local_remove(bat_priv, mcast_entry->addr,
+ BATADV_NO_FLAGS,
+ "mcast TT outdated", false);
+
+ hlist_del(&mcast_entry->list);
+ kfree(mcast_entry);
+ }
+}
+
+/**
+ * batadv_mcast_mla_tt_add - add multicast listener announcements
+ * @bat_priv: the bat priv with all the soft interface information
+ * @mcast_list: a list of addresses which are going to get added
+ *
+ * Add multicast listener announcements from the given mcast_list to the
+ * translation table if they have not been added yet.
+ */
+static void batadv_mcast_mla_tt_add(struct batadv_priv *bat_priv,
+ struct hlist_head *mcast_list)
+{
+ struct batadv_hw_addr *mcast_entry;
+ struct hlist_node *tmp;
+
+ if (!mcast_list)
+ return;
+
+ hlist_for_each_entry_safe(mcast_entry, tmp, mcast_list, list) {
+ if (batadv_mcast_mla_is_duplicate(mcast_entry->addr,
+ &bat_priv->mcast.mla_list))
+ continue;
+
+ if (!batadv_tt_local_add(bat_priv->soft_iface,
+ mcast_entry->addr, BATADV_NO_FLAGS,
+ BATADV_NULL_IFINDEX, BATADV_NO_MARK))
+ continue;
+
+ hlist_del(&mcast_entry->list);
+ hlist_add_head(&mcast_entry->list, &bat_priv->mcast.mla_list);
+ }
+}
+
+/**
+ * batadv_mcast_get_bridge - get the bridge interface on our soft interface
+ * @bat_priv: the bat priv with all the soft interface information
+ *
+ * Return the next bridge interface on top of our soft interface and increase
+ * its refcount. If no such bridge interface exists, then return NULL.
+ */
+static struct net_device *
+batadv_mcast_get_bridge(struct batadv_priv *bat_priv)
+{
+ struct net_device *upper = bat_priv->soft_iface;
+
+ rcu_read_lock();
+
+ do {
+ upper = netdev_master_upper_dev_get_rcu(upper);
+ } while (upper && !(upper->priv_flags & IFF_EBRIDGE));
+
+ if (upper)
+ dev_hold(upper);
+
+ rcu_read_unlock();
+
+ return upper;
+}
+
+/**
+ * batadv_mcast_has_bridge - check whether the soft-iface is bridged
+ * @bat_priv: the bat priv with all the soft interface information
+ *
+ * Check whether there is a bridge on top of our soft interface. Return
+ * true if so, false otherwise.
+ */
+static bool batadv_mcast_has_bridge(struct batadv_priv *bat_priv)
+{
+ struct net_device *bridge;
+
+ bridge = batadv_mcast_get_bridge(bat_priv);
+ if (!bridge)
+ goto out;
+
+ dev_put(bridge);
+ return true;
+out:
+ return false;
+}
+
+/**
+ * batadv_mcast_mla_update - update the own MLAs
+ * @bat_priv: the bat priv with all the soft interface information
+ *
+ * Update the own multicast listener announcements in the translation
+ * table.
+ */
+void batadv_mcast_mla_update(struct batadv_priv *bat_priv)
+{
+ struct net_device *soft_iface = bat_priv->soft_iface;
+ struct hlist_head mcast_list = HLIST_HEAD_INIT;
+ int ret;
+
+ /* Avoid attaching MLAs, if there is a bridge on top of our soft
+ * interface, we don't support that yet (TODO)
+ */
+ if (batadv_mcast_has_bridge(bat_priv))
+ goto update;
+
+ ret = batadv_mcast_mla_softif_get(soft_iface, &mcast_list);
+ if (ret < 0)
+ goto out;
+
+update:
+ batadv_mcast_mla_tt_retract(bat_priv, &mcast_list);
+ batadv_mcast_mla_tt_add(bat_priv, &mcast_list);
+
+out:
+ batadv_mcast_mla_list_free(&mcast_list);
+}
+
+/**
+ * batadv_mcast_free - free the multicast optimizations structures
+ * @bat_priv: the bat priv with all the soft interface information
+ */
+void batadv_mcast_free(struct batadv_priv *bat_priv)
+{
+ batadv_mcast_mla_tt_retract(bat_priv, NULL);
+}
diff --git a/multicast.h b/multicast.h
new file mode 100644
index 0000000..111aad7
--- /dev/null
+++ b/multicast.h
@@ -0,0 +1,38 @@
+/* Copyright (C) 2014 B.A.T.M.A.N. contributors:
+ *
+ * Linus Lüssing
+ *
+ * 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.
+ */
+
+#ifndef _NET_BATMAN_ADV_MULTICAST_H_
+#define _NET_BATMAN_ADV_MULTICAST_H_
+
+#ifdef CONFIG_BATMAN_ADV_MCAST
+
+void batadv_mcast_mla_update(struct batadv_priv *bat_priv);
+
+void batadv_mcast_free(struct batadv_priv *bat_priv);
+
+#else
+
+static inline void batadv_mcast_mla_update(struct batadv_priv *bat_priv)
+{
+ return;
+}
+
+static inline void batadv_mcast_free(struct batadv_priv *bat_priv)
+{
+ return;
+}
+
+#endif /* CONFIG_BATMAN_ADV_MCAST */
+
+#endif /* _NET_BATMAN_ADV_MULTICAST_H_ */
diff --git a/translation-table.c b/translation-table.c
index 6d0da58..4082d05 100644
--- a/translation-table.c
+++ b/translation-table.c
@@ -24,6 +24,7 @@
#include "originator.h"
#include "routing.h"
#include "bridge_loop_avoidance.h"
+#include "multicast.h"
#include <linux/crc32c.h>
@@ -484,7 +485,7 @@ bool batadv_tt_local_add(struct net_device *soft_iface, const uint8_t *addr,
{
struct batadv_priv *bat_priv = netdev_priv(soft_iface);
struct batadv_tt_local_entry *tt_local;
- struct batadv_tt_global_entry *tt_global;
+ struct batadv_tt_global_entry *tt_global = NULL;
struct net_device *in_dev = NULL;
struct hlist_head *head;
struct batadv_tt_orig_list_entry *orig_entry;
@@ -497,7 +498,9 @@ bool batadv_tt_local_add(struct net_device *soft_iface, const uint8_t *addr,
in_dev = dev_get_by_index(&init_net, ifindex);
tt_local = batadv_tt_local_hash_find(bat_priv, addr, vid);
- tt_global = batadv_tt_global_hash_find(bat_priv, addr, vid);
+
+ if (!is_multicast_ether_addr(addr))
+ tt_global = batadv_tt_global_hash_find(bat_priv, addr, vid);
if (tt_local) {
tt_local->last_seen = jiffies;
@@ -562,8 +565,11 @@ bool batadv_tt_local_add(struct net_device *soft_iface, const uint8_t *addr,
tt_local->last_seen = jiffies;
tt_local->common.added_at = tt_local->last_seen;
- /* the batman interface mac address should never be purged */
- if (batadv_compare_eth(addr, soft_iface->dev_addr))
+ /* the batman interface mac and multicast addresses should never be
+ * purged
+ */
+ if (batadv_compare_eth(addr, soft_iface->dev_addr) ||
+ is_multicast_ether_addr(addr))
tt_local->common.flags |= BATADV_TT_CLIENT_NOPURGE;
hash_added = batadv_hash_add(bat_priv->tt.local_hash, batadv_compare_tt,
@@ -1361,6 +1367,11 @@ add_orig_entry:
ret = true;
out_remove:
+ /* Do not remove multicast addresses from the local hash on
+ * global additions
+ */
+ if (is_multicast_ether_addr(tt_addr))
+ goto out;
/* remove address from local hash if present */
local_flags = batadv_tt_local_remove(bat_priv, tt_addr, vid,
@@ -3120,6 +3131,9 @@ static void batadv_tt_local_purge_pending_clients(struct batadv_priv *bat_priv)
*/
static void batadv_tt_local_commit_changes_nolock(struct batadv_priv *bat_priv)
{
+ /* Update multicast addresses in local translation table */
+ batadv_mcast_mla_update(bat_priv);
+
if (atomic_read(&bat_priv->tt.local_changes) < 1) {
if (!batadv_atomic_dec_not_zero(&bat_priv->tt.ogm_append_cnt))
batadv_tt_tvlv_container_update(bat_priv);
diff --git a/types.h b/types.h
index 9f52517..d553264 100644
--- a/types.h
+++ b/types.h
@@ -607,6 +607,16 @@ struct batadv_priv_dat {
};
#endif
+#ifdef CONFIG_BATMAN_ADV_MCAST
+/**
+ * struct batadv_priv_mcast - per mesh interface mcast data
+ * @mla_list: list of multicast addresses we are currently announcing via TT
+ */
+struct batadv_priv_mcast {
+ struct hlist_head mla_list;
+};
+#endif
+
/**
* struct batadv_priv_nc - per mesh interface network coding private data
* @work: work queue callback item for cleanup
@@ -702,6 +712,7 @@ struct batadv_softif_vlan {
* @tt: translation table data
* @tvlv: type-version-length-value data
* @dat: distributed arp table data
+ * @mcast: multicast data
* @network_coding: bool indicating whether network coding is enabled
* @batadv_priv_nc: network coding data
*/
@@ -759,6 +770,9 @@ struct batadv_priv {
#ifdef CONFIG_BATMAN_ADV_DAT
struct batadv_priv_dat dat;
#endif
+#ifdef CONFIG_BATMAN_ADV_MCAST
+ struct batadv_priv_mcast mcast;
+#endif
#ifdef CONFIG_BATMAN_ADV_NC
atomic_t network_coding;
struct batadv_priv_nc nc;
@@ -1116,6 +1130,16 @@ struct batadv_dat_entry {
};
/**
+ * struct batadv_hw_addr - a list entry for a MAC address
+ * @list: list node for the linking of entries
+ * @addr: the MAC address of this list entry
+ */
+struct batadv_hw_addr {
+ struct hlist_node list;
+ unsigned char addr[ETH_ALEN];
+};
+
+/**
* struct batadv_dat_candidate - candidate destination for DAT operations
* @type: the type of the selected candidate. It can one of the following:
* - BATADV_DAT_CANDIDATE_NOT_FOUND
--
1.7.10.4
^ permalink raw reply related [flat|nested] 14+ messages in thread
* [B.A.T.M.A.N.] [PATCHv14 2/6] batman-adv: introduce capability initialization bitfield
2014-02-15 16:47 [B.A.T.M.A.N.] [PATCHv14 0/6] Basic Multicast Optimizations Linus Lüssing
2014-02-15 16:47 ` [B.A.T.M.A.N.] [PATCHv14 1/6] batman-adv: Multicast Listener Announcements via Translation Table Linus Lüssing
@ 2014-02-15 16:47 ` Linus Lüssing
2014-02-16 6:24 ` Marek Lindner
2014-02-15 16:47 ` [B.A.T.M.A.N.] [PATCHv14 3/6] batman-adv: Announce new capability via multicast TVLV Linus Lüssing
` (3 subsequent siblings)
5 siblings, 1 reply; 14+ messages in thread
From: Linus Lüssing @ 2014-02-15 16:47 UTC (permalink / raw)
To: b.a.t.m.a.n
The new bitfield allows us to keep track whether capability subsets of
an originator have gone through their initialization phase yet.
The translation table is the only user right now, but a new one will be
added soon.
Signed-off-by: Linus Lüssing <linus.luessing@web.de>
---
originator.c | 1 -
translation-table.c | 12 +++++++-----
types.h | 7 ++++---
3 files changed, 11 insertions(+), 9 deletions(-)
diff --git a/originator.c b/originator.c
index 9f2da37..24a9300 100644
--- a/originator.c
+++ b/originator.c
@@ -664,7 +664,6 @@ struct batadv_orig_node *batadv_orig_node_new(struct batadv_priv *bat_priv,
/* extra reference for return */
atomic_set(&orig_node->refcount, 2);
- orig_node->tt_initialised = false;
orig_node->bat_priv = bat_priv;
ether_addr_copy(orig_node->orig, addr);
batadv_dat_init_orig_node_addr(orig_node);
diff --git a/translation-table.c b/translation-table.c
index 4082d05..f641dfe 100644
--- a/translation-table.c
+++ b/translation-table.c
@@ -1774,7 +1774,7 @@ void batadv_tt_global_del_orig(struct batadv_priv *bat_priv,
}
spin_unlock_bh(list_lock);
}
- orig_node->tt_initialised = false;
+ orig_node->capa_initialized &= ~BATADV_ORIG_CAPA_HAS_TT;
}
static bool batadv_tt_global_to_purge(struct batadv_tt_global_entry *tt_global,
@@ -2734,7 +2734,7 @@ static void _batadv_tt_update_changes(struct batadv_priv *bat_priv,
return;
}
}
- orig_node->tt_initialised = true;
+ orig_node->capa_initialized |= BATADV_ORIG_CAPA_HAS_TT;
}
static void batadv_tt_fill_gtable(struct batadv_priv *bat_priv,
@@ -3224,13 +3224,15 @@ static void batadv_tt_update_orig(struct batadv_priv *bat_priv,
uint8_t orig_ttvn = (uint8_t)atomic_read(&orig_node->last_ttvn);
struct batadv_tvlv_tt_vlan_data *tt_vlan;
bool full_table = true;
+ bool has_tt_init;
tt_vlan = (struct batadv_tvlv_tt_vlan_data *)tt_buff;
+ has_tt_init = orig_node->capa_initialized & BATADV_ORIG_CAPA_HAS_TT;
+
/* orig table not initialised AND first diff is in the OGM OR the ttvn
* increased by one -> we can apply the attached changes
*/
- if ((!orig_node->tt_initialised && ttvn == 1) ||
- ttvn - orig_ttvn == 1) {
+ if ((!has_tt_init && ttvn == 1) || ttvn - orig_ttvn == 1) {
/* the OGM could not contain the changes due to their size or
* because they have already been sent BATADV_TT_OGM_APPEND_MAX
* times.
@@ -3270,7 +3272,7 @@ static void batadv_tt_update_orig(struct batadv_priv *bat_priv,
/* if we missed more than one change or our tables are not
* in sync anymore -> request fresh tt data
*/
- if (!orig_node->tt_initialised || ttvn != orig_ttvn ||
+ if (!has_tt_init || ttvn != orig_ttvn ||
!batadv_tt_global_check_crc(orig_node, tt_vlan,
tt_num_vlan)) {
request_table:
diff --git a/types.h b/types.h
index d553264..b46117c 100644
--- a/types.h
+++ b/types.h
@@ -205,13 +205,12 @@ struct batadv_orig_bat_iv {
* @last_seen: time when last packet from this node was received
* @bcast_seqno_reset: time when the broadcast seqno window was reset
* @capabilities: announced capabilities of this originator
+ * @capa_initialized: bitfield to remember whether a capability was initialized
* @last_ttvn: last seen translation table version number
* @tt_buff: last tt changeset this node received from the orig node
* @tt_buff_len: length of the last tt changeset this node received from the
* orig node
* @tt_buff_lock: lock that protects tt_buff and tt_buff_len
- * @tt_initialised: bool keeping track of whether or not this node have received
- * any translation table information from the orig node yet
* @tt_lock: prevents from updating the table while reading it. Table update is
* made up by two operations (data structure update and metdata -CRC/TTVN-
* recalculation) and they have to be executed atomically in order to avoid
@@ -248,11 +247,11 @@ struct batadv_orig_node {
unsigned long last_seen;
unsigned long bcast_seqno_reset;
uint8_t capabilities;
+ uint8_t capa_initialized;
atomic_t last_ttvn;
unsigned char *tt_buff;
int16_t tt_buff_len;
spinlock_t tt_buff_lock; /* protects tt_buff & tt_buff_len */
- bool tt_initialised;
/* prevents from changing the table while reading it */
spinlock_t tt_lock;
DECLARE_BITMAP(bcast_bits, BATADV_TQ_LOCAL_WINDOW_SIZE);
@@ -282,10 +281,12 @@ struct batadv_orig_node {
* enum batadv_orig_capabilities - orig node capabilities
* @BATADV_ORIG_CAPA_HAS_DAT: orig node has distributed arp table enabled
* @BATADV_ORIG_CAPA_HAS_NC: orig node has network coding enabled
+ * @BATADV_ORIG_CAPA_HAS_TT: orig node has tt capability
*/
enum batadv_orig_capabilities {
BATADV_ORIG_CAPA_HAS_DAT = BIT(0),
BATADV_ORIG_CAPA_HAS_NC = BIT(1),
+ BATADV_ORIG_CAPA_HAS_TT = BIT(2),
};
/**
--
1.7.10.4
^ permalink raw reply related [flat|nested] 14+ messages in thread
* [B.A.T.M.A.N.] [PATCHv14 3/6] batman-adv: Announce new capability via multicast TVLV
2014-02-15 16:47 [B.A.T.M.A.N.] [PATCHv14 0/6] Basic Multicast Optimizations Linus Lüssing
2014-02-15 16:47 ` [B.A.T.M.A.N.] [PATCHv14 1/6] batman-adv: Multicast Listener Announcements via Translation Table Linus Lüssing
2014-02-15 16:47 ` [B.A.T.M.A.N.] [PATCHv14 2/6] batman-adv: introduce capability initialization bitfield Linus Lüssing
@ 2014-02-15 16:47 ` Linus Lüssing
2014-02-16 6:25 ` Marek Lindner
2014-02-15 16:47 ` [B.A.T.M.A.N.] [PATCHv14 4/6] batman-adv: Modified forwarding behaviour for multicast packets Linus Lüssing
` (2 subsequent siblings)
5 siblings, 1 reply; 14+ messages in thread
From: Linus Lüssing @ 2014-02-15 16:47 UTC (permalink / raw)
To: b.a.t.m.a.n
If the soft interface of a node is not part of a bridge then a node
announces a new multicast TVLV: The existence of this TVLV
signalizes that this node is announcing all of its multicast listeners
via the translation table infrastructure.
Signed-off-by: Linus Lüssing <linus.luessing@web.de>
---
main.c | 1 +
multicast.c | 121 +++++++++++++++++++++++++++++++++++++++++++++++++++---
multicast.h | 14 +++++++
originator.c | 6 +++
packet.h | 12 ++++++
soft-interface.c | 4 ++
types.h | 13 ++++++
7 files changed, 166 insertions(+), 5 deletions(-)
diff --git a/main.c b/main.c
index 58e98c8..8f11b67 100644
--- a/main.c
+++ b/main.c
@@ -149,6 +149,7 @@ int batadv_mesh_init(struct net_device *soft_iface)
goto err;
batadv_gw_init(bat_priv);
+ batadv_mcast_init(bat_priv);
atomic_set(&bat_priv->gw.reselect, 0);
atomic_set(&bat_priv->mesh_state, BATADV_MESH_ACTIVE);
diff --git a/multicast.c b/multicast.c
index 66e270c..fe9e1e3 100644
--- a/multicast.c
+++ b/multicast.c
@@ -202,11 +202,52 @@ out:
}
/**
+ * batadv_mcast_mla_tvlv_update - update multicast tvlv
+ * @bat_priv: the bat priv with all the soft interface information
+ *
+ * Update the own multicast tvlv with our current multicast related settings,
+ * capabilities and inabilities.
+ *
+ * Return true if the tvlv container is registered afterwards. Otherwise return
+ * false.
+ */
+static bool batadv_mcast_mla_tvlv_update(struct batadv_priv *bat_priv)
+{
+ struct batadv_tvlv_mcast_data mcast_data;
+
+ mcast_data.flags = BATADV_NO_FLAGS;
+ memset(mcast_data.reserved, 0, sizeof(mcast_data.reserved));
+
+ /* Avoid attaching MLAs, if there is a bridge on top of our soft
+ * interface, we don't support that yet (TODO)
+ */
+ if (batadv_mcast_has_bridge(bat_priv)) {
+ if (bat_priv->mcast.enabled) {
+ batadv_tvlv_container_unregister(bat_priv,
+ BATADV_TVLV_MCAST, 1);
+ bat_priv->mcast.enabled = false;
+ }
+
+ return false;
+ }
+
+ if (!bat_priv->mcast.enabled ||
+ mcast_data.flags != bat_priv->mcast.flags) {
+ batadv_tvlv_container_register(bat_priv, BATADV_TVLV_MCAST, 1,
+ &mcast_data, sizeof(mcast_data));
+ bat_priv->mcast.flags = mcast_data.flags;
+ bat_priv->mcast.enabled = true;
+ }
+
+ return true;
+}
+
+/**
* batadv_mcast_mla_update - update the own MLAs
* @bat_priv: the bat priv with all the soft interface information
*
* Update the own multicast listener announcements in the translation
- * table.
+ * table as well as the own, announced multicast tvlv container.
*/
void batadv_mcast_mla_update(struct batadv_priv *bat_priv)
{
@@ -214,10 +255,7 @@ void batadv_mcast_mla_update(struct batadv_priv *bat_priv)
struct hlist_head mcast_list = HLIST_HEAD_INIT;
int ret;
- /* Avoid attaching MLAs, if there is a bridge on top of our soft
- * interface, we don't support that yet (TODO)
- */
- if (batadv_mcast_has_bridge(bat_priv))
+ if (!batadv_mcast_mla_tvlv_update(bat_priv))
goto update;
ret = batadv_mcast_mla_softif_get(soft_iface, &mcast_list);
@@ -233,10 +271,83 @@ out:
}
/**
+ * batadv_mcast_tvlv_ogm_handler_v1 - process incoming multicast tvlv container
+ * @bat_priv: the bat priv with all the soft interface information
+ * @orig: the orig_node of the ogm
+ * @flags: flags indicating the tvlv state (see batadv_tvlv_handler_flags)
+ * @tvlv_value: tvlv buffer containing the multicast data
+ * @tvlv_value_len: tvlv buffer length
+ */
+static void batadv_mcast_tvlv_ogm_handler_v1(struct batadv_priv *bat_priv,
+ struct batadv_orig_node *orig,
+ uint8_t flags,
+ void *tvlv_value,
+ uint16_t tvlv_value_len)
+{
+ bool orig_mcast_enabled = !(flags & BATADV_TVLV_HANDLER_OGM_CIFNOTFND);
+ uint8_t mcast_flags = BATADV_NO_FLAGS;
+ bool orig_initialized;
+
+ orig_initialized = orig->capa_initialized & BATADV_ORIG_CAPA_HAS_MCAST;
+
+ /* If mcast support is turned on decrease the disabled mcast node
+ * counter only if we had increased it for this node before. If this
+ * is a completely new orig_node no need to decrease the counter.
+ */
+ if (orig_mcast_enabled &&
+ !(orig->capabilities & BATADV_ORIG_CAPA_HAS_MCAST)) {
+ if (orig_initialized)
+ atomic_dec(&bat_priv->mcast.num_disabled);
+ orig->capabilities |= BATADV_ORIG_CAPA_HAS_MCAST;
+ /* If mcast support is being switched off increase the disabled
+ * mcast node counter.
+ */
+ } else if (!orig_mcast_enabled &&
+ orig->capabilities & BATADV_ORIG_CAPA_HAS_MCAST) {
+ atomic_inc(&bat_priv->mcast.num_disabled);
+ orig->capabilities &= ~BATADV_ORIG_CAPA_HAS_MCAST;
+ }
+
+ orig->capa_initialized |= BATADV_ORIG_CAPA_HAS_MCAST;
+
+ if (orig_mcast_enabled && tvlv_value &&
+ (tvlv_value_len >= sizeof(mcast_flags)))
+ mcast_flags = *(uint8_t *)tvlv_value;
+
+ orig->mcast_flags = mcast_flags;
+}
+
+/**
+ * batadv_mcast_init - initialize the multicast optimizations structures
+ * @bat_priv: the bat priv with all the soft interface information
+ */
+void batadv_mcast_init(struct batadv_priv *bat_priv)
+{
+ batadv_tvlv_handler_register(bat_priv, batadv_mcast_tvlv_ogm_handler_v1,
+ NULL, BATADV_TVLV_MCAST, 1,
+ BATADV_TVLV_HANDLER_OGM_CIFNOTFND);
+}
+
+/**
* batadv_mcast_free - free the multicast optimizations structures
* @bat_priv: the bat priv with all the soft interface information
*/
void batadv_mcast_free(struct batadv_priv *bat_priv)
{
+ batadv_tvlv_container_unregister(bat_priv, BATADV_TVLV_MCAST, 1);
+ batadv_tvlv_handler_unregister(bat_priv, BATADV_TVLV_MCAST, 1);
+
batadv_mcast_mla_tt_retract(bat_priv, NULL);
}
+
+/**
+ * batadv_mcast_purge_orig - reset originator global mcast state modifications
+ * @orig: the originator which is going to get purged
+ */
+void batadv_mcast_purge_orig(struct batadv_orig_node *orig)
+{
+ struct batadv_priv *bat_priv = orig->bat_priv;
+
+ if (!(orig->capabilities & BATADV_ORIG_CAPA_HAS_MCAST))
+ atomic_dec(&bat_priv->mcast.num_disabled);
+}
diff --git a/multicast.h b/multicast.h
index 111aad7..d736335 100644
--- a/multicast.h
+++ b/multicast.h
@@ -19,8 +19,12 @@
void batadv_mcast_mla_update(struct batadv_priv *bat_priv);
+void batadv_mcast_init(struct batadv_priv *bat_priv);
+
void batadv_mcast_free(struct batadv_priv *bat_priv);
+void batadv_mcast_purge_orig(struct batadv_orig_node *orig_node);
+
#else
static inline void batadv_mcast_mla_update(struct batadv_priv *bat_priv)
@@ -28,11 +32,21 @@ static inline void batadv_mcast_mla_update(struct batadv_priv *bat_priv)
return;
}
+static inline int batadv_mcast_init(struct batadv_priv *bat_priv)
+{
+ return 0;
+}
+
static inline void batadv_mcast_free(struct batadv_priv *bat_priv)
{
return;
}
+static inline void batadv_mcast_purge_orig(struct batadv_orig_node *orig_node)
+{
+ return;
+}
+
#endif /* CONFIG_BATMAN_ADV_MCAST */
#endif /* _NET_BATMAN_ADV_MULTICAST_H_ */
diff --git a/originator.c b/originator.c
index 24a9300..ffd9dfb 100644
--- a/originator.c
+++ b/originator.c
@@ -27,6 +27,7 @@
#include "bridge_loop_avoidance.h"
#include "network-coding.h"
#include "fragmentation.h"
+#include "multicast.h"
/* hash class keys */
static struct lock_class_key batadv_orig_hash_lock_class_key;
@@ -557,6 +558,8 @@ static void batadv_orig_node_free_rcu(struct rcu_head *rcu)
}
spin_unlock_bh(&orig_node->neigh_list_lock);
+ batadv_mcast_purge_orig(orig_node);
+
/* Free nc_nodes */
batadv_nc_purge_orig(orig_node->bat_priv, orig_node, NULL);
@@ -672,6 +675,9 @@ struct batadv_orig_node *batadv_orig_node_new(struct batadv_priv *bat_priv,
orig_node->tt_buff_len = 0;
reset_time = jiffies - 1 - msecs_to_jiffies(BATADV_RESET_PROTECTION_MS);
orig_node->bcast_seqno_reset = reset_time;
+#ifdef CONFIG_BATMAN_ADV_MCAST
+ orig_node->mcast_flags = BATADV_NO_FLAGS;
+#endif
/* create a vlan object for the "untagged" LAN */
vlan = batadv_orig_node_vlan_new(orig_node, BATADV_NO_FLAGS);
diff --git a/packet.h b/packet.h
index 0a381d1..e8c483d 100644
--- a/packet.h
+++ b/packet.h
@@ -145,6 +145,7 @@ enum batadv_bla_claimframe {
* @BATADV_TVLV_NC: network coding tvlv
* @BATADV_TVLV_TT: translation table tvlv
* @BATADV_TVLV_ROAM: roaming advertisement tvlv
+ * @BATADV_TVLV_MCAST: multicast capability tvlv
*/
enum batadv_tvlv_type {
BATADV_TVLV_GW = 0x01,
@@ -152,6 +153,7 @@ enum batadv_tvlv_type {
BATADV_TVLV_NC = 0x03,
BATADV_TVLV_TT = 0x04,
BATADV_TVLV_ROAM = 0x05,
+ BATADV_TVLV_MCAST = 0x06,
};
#pragma pack(2)
@@ -504,4 +506,14 @@ struct batadv_tvlv_roam_adv {
__be16 vid;
};
+/**
+ * struct batadv_tvlv_mcast_data - payload of a multicast tvlv
+ * @flags: multicast flags announced by the orig node
+ * @reserved: reserved field
+ */
+struct batadv_tvlv_mcast_data {
+ uint8_t flags;
+ uint8_t reserved[3];
+};
+
#endif /* _NET_BATMAN_ADV_PACKET_H_ */
diff --git a/soft-interface.c b/soft-interface.c
index 633e9d6..8ff47b7 100644
--- a/soft-interface.c
+++ b/soft-interface.c
@@ -689,6 +689,10 @@ static int batadv_softif_init_late(struct net_device *dev)
#ifdef CONFIG_BATMAN_ADV_DAT
atomic_set(&bat_priv->distributed_arp_table, 1);
#endif
+#ifdef CONFIG_BATMAN_ADV_MCAST
+ bat_priv->mcast.flags = BATADV_NO_FLAGS;
+ atomic_set(&bat_priv->mcast.num_disabled, 0);
+#endif
atomic_set(&bat_priv->gw_mode, BATADV_GW_MODE_OFF);
atomic_set(&bat_priv->gw_sel_class, 20);
atomic_set(&bat_priv->gw.bandwidth_down, 100);
diff --git a/types.h b/types.h
index b46117c..96ee0d2 100644
--- a/types.h
+++ b/types.h
@@ -204,6 +204,7 @@ struct batadv_orig_bat_iv {
* @batadv_dat_addr_t: address of the orig node in the distributed hash
* @last_seen: time when last packet from this node was received
* @bcast_seqno_reset: time when the broadcast seqno window was reset
+ * @mcast_flags: multicast flags announced by the orig node
* @capabilities: announced capabilities of this originator
* @capa_initialized: bitfield to remember whether a capability was initialized
* @last_ttvn: last seen translation table version number
@@ -246,6 +247,9 @@ struct batadv_orig_node {
#endif
unsigned long last_seen;
unsigned long bcast_seqno_reset;
+#ifdef CONFIG_BATMAN_ADV_MCAST
+ uint8_t mcast_flags;
+#endif
uint8_t capabilities;
uint8_t capa_initialized;
atomic_t last_ttvn;
@@ -282,11 +286,14 @@ struct batadv_orig_node {
* @BATADV_ORIG_CAPA_HAS_DAT: orig node has distributed arp table enabled
* @BATADV_ORIG_CAPA_HAS_NC: orig node has network coding enabled
* @BATADV_ORIG_CAPA_HAS_TT: orig node has tt capability
+ * @BATADV_ORIG_CAPA_HAS_MCAST: orig node has some multicast capability
+ * (= orig node announces a tvlv of type BATADV_TVLV_MCAST)
*/
enum batadv_orig_capabilities {
BATADV_ORIG_CAPA_HAS_DAT = BIT(0),
BATADV_ORIG_CAPA_HAS_NC = BIT(1),
BATADV_ORIG_CAPA_HAS_TT = BIT(2),
+ BATADV_ORIG_CAPA_HAS_MCAST = BIT(3),
};
/**
@@ -612,9 +619,15 @@ struct batadv_priv_dat {
/**
* struct batadv_priv_mcast - per mesh interface mcast data
* @mla_list: list of multicast addresses we are currently announcing via TT
+ * @flags: the flags we have last sent in our mcast tvlv
+ * @enabled: whether the multicast tvlv is currently enabled
+ * @num_disabled: number of nodes that have no mcast tvlv
*/
struct batadv_priv_mcast {
struct hlist_head mla_list;
+ uint8_t flags;
+ bool enabled;
+ atomic_t num_disabled;
};
#endif
--
1.7.10.4
^ permalink raw reply related [flat|nested] 14+ messages in thread
* [B.A.T.M.A.N.] [PATCHv14 4/6] batman-adv: Modified forwarding behaviour for multicast packets
2014-02-15 16:47 [B.A.T.M.A.N.] [PATCHv14 0/6] Basic Multicast Optimizations Linus Lüssing
` (2 preceding siblings ...)
2014-02-15 16:47 ` [B.A.T.M.A.N.] [PATCHv14 3/6] batman-adv: Announce new capability via multicast TVLV Linus Lüssing
@ 2014-02-15 16:47 ` Linus Lüssing
2014-02-16 6:28 ` Marek Lindner
2014-02-15 16:47 ` [B.A.T.M.A.N.] [PATCHv14 5/6] batman-adv: Add IPv4 link-local/IPv6-ll-all-nodes multicast support Linus Lüssing
2014-02-15 16:47 ` [B.A.T.M.A.N.] [PATCHv14 6/6] batman-adv: Send multicast packets to nodes with a WANT_ALL flag Linus Lüssing
5 siblings, 1 reply; 14+ messages in thread
From: Linus Lüssing @ 2014-02-15 16:47 UTC (permalink / raw)
To: b.a.t.m.a.n
With this patch a multicast packet is not always simply flooded anymore,
the behaviour for the following cases is changed to reduce
unnecessary overhead:
If all nodes within the horizon of a certain node have signalized
multicast listener announcement capability then an IPv6 multicast packet
with a destination of IPv6 link-local scope (excluding ff02::1) coming
from the upstream of this node...
* ...is dropped if there is no according multicast listener in the
translation table,
* ...is forwarded via unicast if there is a single node with interested
multicast listeners
* ...and otherwise still gets flooded.
Signed-off-by: Linus Lüssing <linus.luessing@web.de>
---
multicast.c | 127 ++++++++++++++++++++++++++++++++++++++++++++++++++
multicast.h | 25 ++++++++++
send.c | 10 ++--
send.h | 5 ++
soft-interface.c | 20 +++++++-
sysfs-class-net-mesh | 9 ++++
sysfs.c | 6 +++
translation-table.c | 92 ++++++++++++++++++++++++++++--------
translation-table.h | 2 +
types.h | 7 +++
10 files changed, 277 insertions(+), 26 deletions(-)
diff --git a/multicast.c b/multicast.c
index fe9e1e3..127ba50 100644
--- a/multicast.c
+++ b/multicast.c
@@ -17,6 +17,7 @@
#include "originator.h"
#include "hard-interface.h"
#include "translation-table.h"
+#include "multicast.h"
/**
* batadv_mcast_mla_softif_get - get softif multicast listeners
@@ -271,6 +272,132 @@ out:
}
/**
+ * batadv_mcast_forw_mode_check_ipv6 - check for optimized forwarding potential
+ * @bat_priv: the bat priv with all the soft interface information
+ * @skb: the IPv6 packet to check
+ *
+ * Check whether the given IPv6 packet has the potential to
+ * be forwarded with a mode more optimal than classic flooding.
+ *
+ * If so then return 0. Otherwise -EINVAL is returned or -ENOMEM if we are
+ * out of memory.
+ */
+static int batadv_mcast_forw_mode_check_ipv6(struct batadv_priv *bat_priv,
+ struct sk_buff *skb)
+{
+ struct ipv6hdr *ip6hdr;
+
+ /* We might fail due to out-of-memory -> drop it */
+ if (!pskb_may_pull(skb, sizeof(struct ethhdr) + sizeof(*ip6hdr)))
+ return -ENOMEM;
+
+ ip6hdr = ipv6_hdr(skb);
+
+ /* TODO: Implement Multicast Router Discovery (RFC4286),
+ * then allow scope > link local, too
+ */
+ if (IPV6_ADDR_MC_SCOPE(&ip6hdr->daddr) !=
+ IPV6_ADDR_SCOPE_LINKLOCAL)
+ return -EINVAL;
+
+ /* link-local-all-nodes multicast listeners behind a bridge are
+ * not snoopable (see RFC4541, section 3, paragraph 3)
+ */
+ if (ipv6_addr_is_ll_all_nodes(&ip6hdr->daddr))
+ return -EINVAL;
+
+ return 0;
+}
+
+/**
+ * batadv_mcast_forw_mode_check - check for optimized forwarding potential
+ * @bat_priv: the bat priv with all the soft interface information
+ * @skb: the multicast frame to check
+ *
+ * Check whether the given multicast ethernet frame has the potential to
+ * be forwarded with a mode more optimal than classic flooding.
+ *
+ * If so then return 0. Otherwise -EINVAL is returned or -ENOMEM if we are
+ * out of memory.
+ */
+static int batadv_mcast_forw_mode_check(struct batadv_priv *bat_priv,
+ struct sk_buff *skb)
+{
+ struct ethhdr *ethhdr = eth_hdr(skb);
+
+ if (!atomic_read(&bat_priv->multicast_mode))
+ return -EINVAL;
+
+ if (atomic_read(&bat_priv->mcast.num_disabled))
+ return -EINVAL;
+
+ switch (ntohs(ethhdr->h_proto)) {
+ case ETH_P_IPV6:
+ return batadv_mcast_forw_mode_check_ipv6(bat_priv, skb);
+ default:
+ return -EINVAL;
+ }
+}
+
+/**
+ * batadv_mcast_forw_tt_node_get - get a multicast tt node
+ * @bat_priv: the bat priv with all the soft interface information
+ * @ethhdr: the ether header containing the multicast destination
+ *
+ * Return an orig_node matching the multicast address provided by ethhdr
+ * via a translation table lookup. This increases the returned nodes refcount.
+ */
+static struct batadv_orig_node *
+batadv_mcast_forw_tt_node_get(struct batadv_priv *bat_priv,
+ struct ethhdr *ethhdr)
+{
+ return batadv_transtable_search(bat_priv, ethhdr->h_source,
+ ethhdr->h_dest, BATADV_NO_FLAGS);
+}
+
+/**
+ * batadv_mcast_forw_mode - check on how to forward a multicast packet
+ * @bat_priv: the bat priv with all the soft interface information
+ * @skb: The multicast packet to check
+ * @orig: an originator to be set to forward the skb to
+ *
+ * Return the forwarding mode as enum batadv_forw_mode and in case of
+ * BATADV_FORW_SINGLE set the orig to the single originator the skb
+ * should be forwarded to.
+ */
+enum batadv_forw_mode
+batadv_mcast_forw_mode(struct batadv_priv *bat_priv, struct sk_buff *skb,
+ struct batadv_orig_node **orig)
+{
+ struct ethhdr *ethhdr;
+ int ret, tt_count;
+
+ ret = batadv_mcast_forw_mode_check(bat_priv, skb);
+ if (ret == -ENOMEM)
+ return BATADV_FORW_NONE;
+ else if (ret < 0)
+ return BATADV_FORW_ALL;
+
+ ethhdr = eth_hdr(skb);
+
+ tt_count = batadv_tt_global_hash_count(bat_priv, ethhdr->h_dest,
+ BATADV_NO_FLAGS);
+
+ switch (tt_count) {
+ case 1:
+ *orig = batadv_mcast_forw_tt_node_get(bat_priv, ethhdr);
+ if (*orig)
+ return BATADV_FORW_SINGLE;
+
+ /* fall through */
+ case 0:
+ return BATADV_FORW_NONE;
+ default:
+ return BATADV_FORW_ALL;
+ }
+}
+
+/**
* batadv_mcast_tvlv_ogm_handler_v1 - process incoming multicast tvlv container
* @bat_priv: the bat priv with all the soft interface information
* @orig: the orig_node of the ogm
diff --git a/multicast.h b/multicast.h
index d736335..a76bc3a 100644
--- a/multicast.h
+++ b/multicast.h
@@ -15,10 +15,28 @@
#ifndef _NET_BATMAN_ADV_MULTICAST_H_
#define _NET_BATMAN_ADV_MULTICAST_H_
+/**
+ * batadv_forw_mode - the way a packet should be forwarded as
+ * @BATADV_FORW_ALL: forward the packet to all nodes
+ * (currently via classic flooding)
+ * @BATADV_FORW_SINGLE: forward the packet to a single node
+ * (currently via the BATMAN unicast routing protocol)
+ * @BATADV_FORW_NONE: don't forward, drop it
+ */
+enum batadv_forw_mode {
+ BATADV_FORW_ALL,
+ BATADV_FORW_SINGLE,
+ BATADV_FORW_NONE,
+};
+
#ifdef CONFIG_BATMAN_ADV_MCAST
void batadv_mcast_mla_update(struct batadv_priv *bat_priv);
+enum batadv_forw_mode
+batadv_mcast_forw_mode(struct batadv_priv *bat_priv, struct sk_buff *skb,
+ struct batadv_orig_node **mcast_single_orig);
+
void batadv_mcast_init(struct batadv_priv *bat_priv);
void batadv_mcast_free(struct batadv_priv *bat_priv);
@@ -32,6 +50,13 @@ static inline void batadv_mcast_mla_update(struct batadv_priv *bat_priv)
return;
}
+static inline enum batadv_forw_mode
+batadv_mcast_forw_mode(struct batadv_priv *bat_priv, struct sk_buff *skb,
+ struct batadv_orig_node **mcast_single_orig)
+{
+ return BATADV_FORW_ALL;
+}
+
static inline int batadv_mcast_init(struct batadv_priv *bat_priv)
{
return 0;
diff --git a/send.c b/send.c
index 31a8c19..ef4c93e 100644
--- a/send.c
+++ b/send.c
@@ -248,11 +248,11 @@ out:
*
* Returns NET_XMIT_DROP in case of error or NET_XMIT_SUCCESS otherwise.
*/
-static int batadv_send_skb_unicast(struct batadv_priv *bat_priv,
- struct sk_buff *skb, int packet_type,
- int packet_subtype,
- struct batadv_orig_node *orig_node,
- unsigned short vid)
+int batadv_send_skb_unicast(struct batadv_priv *bat_priv,
+ struct sk_buff *skb, int packet_type,
+ int packet_subtype,
+ struct batadv_orig_node *orig_node,
+ unsigned short vid)
{
struct batadv_unicast_packet *unicast_packet;
struct ethhdr *ethhdr;
diff --git a/send.h b/send.h
index aaddaa9..887b63f 100644
--- a/send.h
+++ b/send.h
@@ -36,6 +36,11 @@ bool batadv_send_skb_prepare_unicast_4addr(struct batadv_priv *bat_priv,
struct sk_buff *skb,
struct batadv_orig_node *orig_node,
int packet_subtype);
+int batadv_send_skb_unicast(struct batadv_priv *bat_priv,
+ struct sk_buff *skb, int packet_type,
+ int packet_subtype,
+ struct batadv_orig_node *orig_node,
+ unsigned short vid);
int batadv_send_skb_via_tt_generic(struct batadv_priv *bat_priv,
struct sk_buff *skb, int packet_type,
int packet_subtype, uint8_t *dst_hint,
diff --git a/soft-interface.c b/soft-interface.c
index 8ff47b7..1a643fe 100644
--- a/soft-interface.c
+++ b/soft-interface.c
@@ -32,6 +32,7 @@
#include <linux/ethtool.h>
#include <linux/etherdevice.h>
#include <linux/if_vlan.h>
+#include "multicast.h"
#include "bridge_loop_avoidance.h"
#include "network-coding.h"
@@ -170,6 +171,8 @@ static int batadv_interface_tx(struct sk_buff *skb,
unsigned short vid;
uint32_t seqno;
int gw_mode;
+ enum batadv_forw_mode forw_mode;
+ struct batadv_orig_node *mcast_single_orig = NULL;
if (atomic_read(&bat_priv->mesh_state) != BATADV_MESH_ACTIVE)
goto dropped;
@@ -247,9 +250,19 @@ static int batadv_interface_tx(struct sk_buff *skb,
* directed to a DHCP server
*/
goto dropped;
- }
send:
+ if (do_bcast && !is_broadcast_ether_addr(ethhdr->h_dest)) {
+ forw_mode = batadv_mcast_forw_mode(bat_priv, skb,
+ &mcast_single_orig);
+ if (forw_mode == BATADV_FORW_NONE)
+ goto dropped;
+
+ if (forw_mode == BATADV_FORW_SINGLE)
+ do_bcast = false;
+ }
+ }
+
batadv_skb_set_priority(skb, 0);
/* ethernet packet should be broadcasted */
@@ -301,6 +314,10 @@ send:
if (ret)
goto dropped;
ret = batadv_send_skb_via_gw(bat_priv, skb, vid);
+ } else if (mcast_single_orig) {
+ ret = batadv_send_skb_unicast(bat_priv, skb,
+ BATADV_UNICAST, 0,
+ mcast_single_orig, vid);
} else {
if (batadv_dat_snoop_outgoing_arp_request(bat_priv,
skb))
@@ -691,6 +708,7 @@ static int batadv_softif_init_late(struct net_device *dev)
#endif
#ifdef CONFIG_BATMAN_ADV_MCAST
bat_priv->mcast.flags = BATADV_NO_FLAGS;
+ atomic_set(&bat_priv->multicast_mode, 1);
atomic_set(&bat_priv->mcast.num_disabled, 0);
#endif
atomic_set(&bat_priv->gw_mode, BATADV_GW_MODE_OFF);
diff --git a/sysfs-class-net-mesh b/sysfs-class-net-mesh
index 4793d3d..a8e918c 100644
--- a/sysfs-class-net-mesh
+++ b/sysfs-class-net-mesh
@@ -76,6 +76,15 @@ Description:
is used to classify clients as "isolated" by the
Extended Isolation feature.
+What: /sys/class/net/<mesh_iface>/mesh/multicast_mode
+Date: June 2013
+Contact: Linus Lüssing <linus.luessing@web.de>
+Description:
+ Indicates whether multicast optimizations are enabled
+ or disabled. If set to zero then all nodes in the
+ mesh are going to use classic flooding for any
+ multicast packet with no optimizations.
+
What: /sys/class/net/<mesh_iface>/mesh/network_coding
Date: Nov 2012
Contact: Martin Hundeboll <martin@hundeboll.net>
diff --git a/sysfs.c b/sysfs.c
index e456bf6..1ebb0d9 100644
--- a/sysfs.c
+++ b/sysfs.c
@@ -539,6 +539,9 @@ BATADV_ATTR_SIF_UINT(gw_sel_class, S_IRUGO | S_IWUSR, 1, BATADV_TQ_MAX_VALUE,
batadv_post_gw_reselect);
static BATADV_ATTR(gw_bandwidth, S_IRUGO | S_IWUSR, batadv_show_gw_bwidth,
batadv_store_gw_bwidth);
+#ifdef CONFIG_BATMAN_ADV_MCAST
+BATADV_ATTR_SIF_BOOL(multicast_mode, S_IRUGO | S_IWUSR, NULL);
+#endif
#ifdef CONFIG_BATMAN_ADV_DEBUG
BATADV_ATTR_SIF_UINT(log_level, S_IRUGO | S_IWUSR, 0, BATADV_DBG_ALL, NULL);
#endif
@@ -558,6 +561,9 @@ static struct batadv_attribute *batadv_mesh_attrs[] = {
#ifdef CONFIG_BATMAN_ADV_DAT
&batadv_attr_distributed_arp_table,
#endif
+#ifdef CONFIG_BATMAN_ADV_MCAST
+ &batadv_attr_multicast_mode,
+#endif
&batadv_attr_fragmentation,
&batadv_attr_routing_algo,
&batadv_attr_gw_mode,
diff --git a/translation-table.c b/translation-table.c
index f641dfe..dab5c39 100644
--- a/translation-table.c
+++ b/translation-table.c
@@ -193,6 +193,32 @@ batadv_tt_global_entry_free_ref(struct batadv_tt_global_entry *tt_global_entry)
}
}
+/**
+ * batadv_tt_global_hash_count - count the number of orig entries
+ * @hash: hash table containing the tt entries
+ * @addr: the mac address of the client to count entries for
+ * @vid: VLAN identifier
+ *
+ * Return the number of originators advertising the given address/data
+ * (excluding ourself).
+ */
+int batadv_tt_global_hash_count(struct batadv_priv *bat_priv,
+ const uint8_t *addr, unsigned short vid)
+{
+ struct batadv_tt_global_entry *tt_global_entry;
+ int count = 0;
+
+ tt_global_entry = batadv_tt_global_hash_find(bat_priv, addr, vid);
+ if (!tt_global_entry)
+ goto out;
+
+ count = atomic_read(&tt_global_entry->orig_list_count);
+ batadv_tt_global_entry_free_ref(tt_global_entry);
+
+out:
+ return count;
+}
+
static void batadv_tt_orig_list_entry_free_rcu(struct rcu_head *rcu)
{
struct batadv_tt_orig_list_entry *orig_entry;
@@ -1225,6 +1251,8 @@ batadv_tt_global_orig_entry_add(struct batadv_tt_global_entry *tt_global,
hlist_add_head_rcu(&orig_entry->list,
&tt_global->orig_list);
spin_unlock_bh(&tt_global->list_lock);
+ atomic_inc(&tt_global->orig_list_count);
+
out:
if (orig_entry)
batadv_tt_orig_list_entry_free_ref(orig_entry);
@@ -1298,6 +1326,7 @@ static bool batadv_tt_global_add(struct batadv_priv *bat_priv,
common->added_at = jiffies;
INIT_HLIST_HEAD(&tt_global_entry->orig_list);
+ atomic_set(&tt_global_entry->orig_list_count, 0);
spin_lock_init(&tt_global_entry->list_lock);
hash_added = batadv_hash_add(bat_priv->tt.global_hash,
@@ -1563,6 +1592,25 @@ out:
return 0;
}
+/**
+ * batadv_tt_global_del_orig_entry - remove and free an orig_entry
+ * @tt_global_entry: the global entry to remove the orig_entry from
+ * @orig_entry: the orig entry to remove and free
+ *
+ * Remove an orig_entry from its list in the given tt_global_entry and
+ * free this orig_entry afterwards.
+ */
+static void
+batadv_tt_global_del_orig_entry(struct batadv_tt_global_entry *tt_global_entry,
+ struct batadv_tt_orig_list_entry *orig_entry)
+{
+ batadv_tt_global_size_dec(orig_entry->orig_node,
+ tt_global_entry->common.vid);
+ atomic_dec(&tt_global_entry->orig_list_count);
+ hlist_del_rcu(&orig_entry->list);
+ batadv_tt_orig_list_entry_free_ref(orig_entry);
+}
+
/* deletes the orig list of a tt_global_entry */
static void
batadv_tt_global_del_orig_list(struct batadv_tt_global_entry *tt_global_entry)
@@ -1573,20 +1621,26 @@ batadv_tt_global_del_orig_list(struct batadv_tt_global_entry *tt_global_entry)
spin_lock_bh(&tt_global_entry->list_lock);
head = &tt_global_entry->orig_list;
- hlist_for_each_entry_safe(orig_entry, safe, head, list) {
- hlist_del_rcu(&orig_entry->list);
- batadv_tt_global_size_dec(orig_entry->orig_node,
- tt_global_entry->common.vid);
- batadv_tt_orig_list_entry_free_ref(orig_entry);
- }
+ hlist_for_each_entry_safe(orig_entry, safe, head, list)
+ batadv_tt_global_del_orig_entry(tt_global_entry, orig_entry);
spin_unlock_bh(&tt_global_entry->list_lock);
}
+/**
+ * batadv_tt_global_del_orig_node - remove orig_node from a global tt entry
+ * @bat_priv: the bat priv with all the soft interface information
+ * @tt_global_entry: the global entry to remove the orig_node from
+ * @orig_node: the originator announcing the client
+ * @message: message to append to the log on deletion
+ *
+ * Remove the given orig_node and its according orig_entry from the given
+ * global tt entry.
+ */
static void
-batadv_tt_global_del_orig_entry(struct batadv_priv *bat_priv,
- struct batadv_tt_global_entry *tt_global_entry,
- struct batadv_orig_node *orig_node,
- const char *message)
+batadv_tt_global_del_orig_node(struct batadv_priv *bat_priv,
+ struct batadv_tt_global_entry *tt_global_entry,
+ struct batadv_orig_node *orig_node,
+ const char *message)
{
struct hlist_head *head;
struct hlist_node *safe;
@@ -1603,10 +1657,8 @@ batadv_tt_global_del_orig_entry(struct batadv_priv *bat_priv,
orig_node->orig,
tt_global_entry->common.addr,
BATADV_PRINT_VID(vid), message);
- hlist_del_rcu(&orig_entry->list);
- batadv_tt_global_size_dec(orig_node,
- tt_global_entry->common.vid);
- batadv_tt_orig_list_entry_free_ref(orig_entry);
+ batadv_tt_global_del_orig_entry(tt_global_entry,
+ orig_entry);
}
}
spin_unlock_bh(&tt_global_entry->list_lock);
@@ -1648,8 +1700,8 @@ batadv_tt_global_del_roaming(struct batadv_priv *bat_priv,
/* there is another entry, we can simply delete this
* one and can still use the other one.
*/
- batadv_tt_global_del_orig_entry(bat_priv, tt_global_entry,
- orig_node, message);
+ batadv_tt_global_del_orig_node(bat_priv, tt_global_entry,
+ orig_node, message);
}
/**
@@ -1675,8 +1727,8 @@ static void batadv_tt_global_del(struct batadv_priv *bat_priv,
goto out;
if (!roaming) {
- batadv_tt_global_del_orig_entry(bat_priv, tt_global_entry,
- orig_node, message);
+ batadv_tt_global_del_orig_node(bat_priv, tt_global_entry,
+ orig_node, message);
if (hlist_empty(&tt_global_entry->orig_list))
batadv_tt_global_free(bat_priv, tt_global_entry,
@@ -1759,8 +1811,8 @@ void batadv_tt_global_del_orig(struct batadv_priv *bat_priv,
struct batadv_tt_global_entry,
common);
- batadv_tt_global_del_orig_entry(bat_priv, tt_global,
- orig_node, message);
+ batadv_tt_global_del_orig_node(bat_priv, tt_global,
+ orig_node, message);
if (hlist_empty(&tt_global->orig_list)) {
vid = tt_global->common.vid;
diff --git a/translation-table.h b/translation-table.h
index 20a1d78..ad84d7b 100644
--- a/translation-table.h
+++ b/translation-table.h
@@ -29,6 +29,8 @@ int batadv_tt_global_seq_print_text(struct seq_file *seq, void *offset);
void batadv_tt_global_del_orig(struct batadv_priv *bat_priv,
struct batadv_orig_node *orig_node,
int32_t match_vid, const char *message);
+int batadv_tt_global_hash_count(struct batadv_priv *bat_priv,
+ const uint8_t *addr, unsigned short vid);
struct batadv_orig_node *batadv_transtable_search(struct batadv_priv *bat_priv,
const uint8_t *src,
const uint8_t *addr,
diff --git a/types.h b/types.h
index 96ee0d2..c28fc4a 100644
--- a/types.h
+++ b/types.h
@@ -696,6 +696,8 @@ struct batadv_softif_vlan {
* enabled
* @distributed_arp_table: bool indicating whether distributed ARP table is
* enabled
+ * @multicast_mode: Enable or disable multicast optimizations on this node's
+ * sender/originating side
* @gw_mode: gateway operation: off, client or server (see batadv_gw_modes)
* @gw_sel_class: gateway selection class (applies if gw_mode client)
* @orig_interval: OGM broadcast interval in milliseconds
@@ -746,6 +748,9 @@ struct batadv_priv {
#ifdef CONFIG_BATMAN_ADV_DAT
atomic_t distributed_arp_table;
#endif
+#ifdef CONFIG_BATMAN_ADV_MCAST
+ atomic_t multicast_mode;
+#endif
atomic_t gw_mode;
atomic_t gw_sel_class;
atomic_t orig_interval;
@@ -909,12 +914,14 @@ struct batadv_tt_local_entry {
* struct batadv_tt_global_entry - translation table global entry data
* @common: general translation table data
* @orig_list: list of orig nodes announcing this non-mesh client
+ * @orig_list_count: number of items in the orig_list
* @list_lock: lock protecting orig_list
* @roam_at: time at which TT_GLOBAL_ROAM was set
*/
struct batadv_tt_global_entry {
struct batadv_tt_common_entry common;
struct hlist_head orig_list;
+ atomic_t orig_list_count;
spinlock_t list_lock; /* protects orig_list */
unsigned long roam_at;
};
--
1.7.10.4
^ permalink raw reply related [flat|nested] 14+ messages in thread
* [B.A.T.M.A.N.] [PATCHv14 5/6] batman-adv: Add IPv4 link-local/IPv6-ll-all-nodes multicast support
2014-02-15 16:47 [B.A.T.M.A.N.] [PATCHv14 0/6] Basic Multicast Optimizations Linus Lüssing
` (3 preceding siblings ...)
2014-02-15 16:47 ` [B.A.T.M.A.N.] [PATCHv14 4/6] batman-adv: Modified forwarding behaviour for multicast packets Linus Lüssing
@ 2014-02-15 16:47 ` Linus Lüssing
2014-02-16 6:30 ` Marek Lindner
2015-06-10 8:52 ` Sven Eckelmann
2014-02-15 16:47 ` [B.A.T.M.A.N.] [PATCHv14 6/6] batman-adv: Send multicast packets to nodes with a WANT_ALL flag Linus Lüssing
5 siblings, 2 replies; 14+ messages in thread
From: Linus Lüssing @ 2014-02-15 16:47 UTC (permalink / raw)
To: b.a.t.m.a.n
With this patch a node may additionally perform the dropping or
unicasting behaviour for a link-local IPv4 and link-local-all-nodes
IPv6 multicast packet, too.
The extra counter and BATADV_MCAST_WANT_ALL_UNSNOOPABLES flag is needed
because with a future bridge snooping support integration a node with a
bridge on top of its soft interface is not able to reliably detect its
multicast listeners for IPv4 link-local and the IPv6
link-local-all-nodes addresses anymore (see RFC4541, section 2.1.2.2
and section 3).
Even though this new flag does make "no difference" now, it'll ensure
a seamless integration of multicast bridge support without needing to
break compatibility later.
Also note, that even with multicast bridge support it won't be possible
to optimize 224.0.0.x and ff02::1 towards nodes with bridges, they will
always receive these ranges.
Signed-off-by: Linus Lüssing <linus.luessing@web.de>
---
main.c | 6 +++
main.h | 1 +
multicast.c | 135 ++++++++++++++++++++++++++++++++++++++++++++++++++----
packet.h | 9 ++++
soft-interface.c | 1 +
types.h | 12 +++++
6 files changed, 156 insertions(+), 8 deletions(-)
diff --git a/main.c b/main.c
index 8f11b67..57b09fa 100644
--- a/main.c
+++ b/main.c
@@ -111,6 +111,9 @@ int batadv_mesh_init(struct net_device *soft_iface)
spin_lock_init(&bat_priv->tt.last_changeset_lock);
spin_lock_init(&bat_priv->tt.commit_lock);
spin_lock_init(&bat_priv->gw.list_lock);
+#ifdef CONFIG_BATMAN_ADV_MCAST
+ spin_lock_init(&bat_priv->mcast.want_lists_lock);
+#endif
spin_lock_init(&bat_priv->tvlv.container_list_lock);
spin_lock_init(&bat_priv->tvlv.handler_list_lock);
spin_lock_init(&bat_priv->softif_vlan_list_lock);
@@ -118,6 +121,9 @@ int batadv_mesh_init(struct net_device *soft_iface)
INIT_HLIST_HEAD(&bat_priv->forw_bat_list);
INIT_HLIST_HEAD(&bat_priv->forw_bcast_list);
INIT_HLIST_HEAD(&bat_priv->gw.list);
+#ifdef CONFIG_BATMAN_ADV_MCAST
+ INIT_HLIST_HEAD(&bat_priv->mcast.want_all_unsnoopables_list);
+#endif
INIT_LIST_HEAD(&bat_priv->tt.changes_list);
INIT_LIST_HEAD(&bat_priv->tt.req_list);
INIT_LIST_HEAD(&bat_priv->tt.roam_list);
diff --git a/main.h b/main.h
index dc6b4a2..a31da5a 100644
--- a/main.h
+++ b/main.h
@@ -177,6 +177,7 @@ enum batadv_uev_type {
#include <linux/slab.h>
#include <net/sock.h> /* struct sock */
#include <net/addrconf.h> /* ipv6 address stuff */
+#include <linux/ip.h>
#include <net/rtnetlink.h>
#include <linux/jiffies.h>
#include <linux/seq_file.h>
diff --git a/multicast.c b/multicast.c
index 127ba50..ebfbf6e 100644
--- a/multicast.c
+++ b/multicast.c
@@ -272,9 +272,48 @@ out:
}
/**
+ * batadv_mcast_forw_mode_check_ipv4 - check for optimized forwarding potential
+ * @bat_priv: the bat priv with all the soft interface information
+ * @skb: the IPv4 packet to check
+ * @is_unsnoopable: stores whether the destination is snoopable
+ *
+ * Check whether the given IPv4 packet has the potential to
+ * be forwarded with a mode more optimal than classic flooding.
+ *
+ * If so then return 0. Otherwise -EINVAL is returned or -ENOMEM if we are
+ * out of memory.
+ */
+static int batadv_mcast_forw_mode_check_ipv4(struct batadv_priv *bat_priv,
+ struct sk_buff *skb,
+ bool *is_unsnoopable)
+{
+ struct iphdr *iphdr;
+
+ /* We might fail due to out-of-memory -> drop it */
+ if (!pskb_may_pull(skb, sizeof(struct ethhdr) + sizeof(*iphdr)))
+ return -ENOMEM;
+
+ iphdr = ip_hdr(skb);
+
+ /* TODO: Implement Multicast Router Discovery (RFC4286),
+ * then allow scope > link local, too
+ */
+ if (!ipv4_is_local_multicast(iphdr->daddr))
+ return -EINVAL;
+
+ /* link-local multicast listeners behind a bridge are
+ * not snoopable (see RFC4541, section 2.1.2.2)
+ */
+ *is_unsnoopable = true;
+
+ return 0;
+}
+
+/**
* batadv_mcast_forw_mode_check_ipv6 - check for optimized forwarding potential
* @bat_priv: the bat priv with all the soft interface information
* @skb: the IPv6 packet to check
+ * @is_unsnoopable: stores whether the destination is snoopable
*
* Check whether the given IPv6 packet has the potential to
* be forwarded with a mode more optimal than classic flooding.
@@ -283,7 +322,8 @@ out:
* out of memory.
*/
static int batadv_mcast_forw_mode_check_ipv6(struct batadv_priv *bat_priv,
- struct sk_buff *skb)
+ struct sk_buff *skb,
+ bool *is_unsnoopable)
{
struct ipv6hdr *ip6hdr;
@@ -304,7 +344,7 @@ static int batadv_mcast_forw_mode_check_ipv6(struct batadv_priv *bat_priv,
* not snoopable (see RFC4541, section 3, paragraph 3)
*/
if (ipv6_addr_is_ll_all_nodes(&ip6hdr->daddr))
- return -EINVAL;
+ *is_unsnoopable = true;
return 0;
}
@@ -313,6 +353,7 @@ static int batadv_mcast_forw_mode_check_ipv6(struct batadv_priv *bat_priv,
* batadv_mcast_forw_mode_check - check for optimized forwarding potential
* @bat_priv: the bat priv with all the soft interface information
* @skb: the multicast frame to check
+ * @is_unsnoopable: stores whether the destination is snoopable
*
* Check whether the given multicast ethernet frame has the potential to
* be forwarded with a mode more optimal than classic flooding.
@@ -321,7 +362,8 @@ static int batadv_mcast_forw_mode_check_ipv6(struct batadv_priv *bat_priv,
* out of memory.
*/
static int batadv_mcast_forw_mode_check(struct batadv_priv *bat_priv,
- struct sk_buff *skb)
+ struct sk_buff *skb,
+ bool *is_unsnoopable)
{
struct ethhdr *ethhdr = eth_hdr(skb);
@@ -332,8 +374,12 @@ static int batadv_mcast_forw_mode_check(struct batadv_priv *bat_priv,
return -EINVAL;
switch (ntohs(ethhdr->h_proto)) {
+ case ETH_P_IP:
+ return batadv_mcast_forw_mode_check_ipv4(bat_priv, skb,
+ is_unsnoopable);
case ETH_P_IPV6:
- return batadv_mcast_forw_mode_check_ipv6(bat_priv, skb);
+ return batadv_mcast_forw_mode_check_ipv6(bat_priv, skb,
+ is_unsnoopable);
default:
return -EINVAL;
}
@@ -356,6 +402,33 @@ batadv_mcast_forw_tt_node_get(struct batadv_priv *bat_priv,
}
/**
+ * batadv_mcast_want_forw_unsnoop_node_get - get a node with an unsnoopable flag
+ * @bat_priv: the bat priv with all the soft interface information
+ *
+ * Return an orig_node which has the BATADV_MCAST_WANT_ALL_UNSNOOPABLES flag set
+ * and increase its refcount.
+ */
+static struct batadv_orig_node *
+batadv_mcast_forw_unsnoop_node_get(struct batadv_priv *bat_priv)
+{
+ struct batadv_orig_node *orig_node;
+
+ rcu_read_lock();
+ hlist_for_each_entry_rcu(orig_node,
+ &bat_priv->mcast.want_all_unsnoopables_list,
+ mcast_want_all_unsnoopables_node) {
+ if (atomic_inc_not_zero(&orig_node->refcount))
+ goto unlock;
+ }
+
+ orig_node = NULL;
+
+unlock:
+ rcu_read_unlock();
+ return orig_node;
+}
+
+/**
* batadv_mcast_forw_mode - check on how to forward a multicast packet
* @bat_priv: the bat priv with all the soft interface information
* @skb: The multicast packet to check
@@ -370,9 +443,10 @@ batadv_mcast_forw_mode(struct batadv_priv *bat_priv, struct sk_buff *skb,
struct batadv_orig_node **orig)
{
struct ethhdr *ethhdr;
- int ret, tt_count;
+ bool is_unsnoopable = false;
+ int ret, tt_count, unsnoop_count, total_count;
- ret = batadv_mcast_forw_mode_check(bat_priv, skb);
+ ret = batadv_mcast_forw_mode_check(bat_priv, skb, &is_unsnoopable);
if (ret == -ENOMEM)
return BATADV_FORW_NONE;
else if (ret < 0)
@@ -382,10 +456,18 @@ batadv_mcast_forw_mode(struct batadv_priv *bat_priv, struct sk_buff *skb,
tt_count = batadv_tt_global_hash_count(bat_priv, ethhdr->h_dest,
BATADV_NO_FLAGS);
+ unsnoop_count = !is_unsnoopable ? 0 :
+ atomic_read(&bat_priv->mcast.num_want_all_unsnoopables);
- switch (tt_count) {
+ total_count = tt_count + unsnoop_count;
+
+ switch (total_count) {
case 1:
- *orig = batadv_mcast_forw_tt_node_get(bat_priv, ethhdr);
+ if (tt_count)
+ *orig = batadv_mcast_forw_tt_node_get(bat_priv, ethhdr);
+ else if (unsnoop_count)
+ *orig = batadv_mcast_forw_unsnoop_node_get(bat_priv);
+
if (*orig)
return BATADV_FORW_SINGLE;
@@ -398,6 +480,39 @@ batadv_mcast_forw_mode(struct batadv_priv *bat_priv, struct sk_buff *skb,
}
/**
+ * batadv_mcast_want_unsnoop_update - update unsnoop counter and list
+ * @bat_priv: the bat priv with all the soft interface information
+ * @orig: the orig_node which multicast state might have changed of
+ * @mcast_flags: flags indicating the new multicast state
+ *
+ * If the BATADV_MCAST_WANT_ALL_UNSNOOPABLES flag of this originator,
+ * orig, has toggled then this method updates counter and list accordingly.
+ */
+static void batadv_mcast_want_unsnoop_update(struct batadv_priv *bat_priv,
+ struct batadv_orig_node *orig,
+ uint8_t mcast_flags)
+{
+ /* switched from flag unset to set */
+ if (mcast_flags & BATADV_MCAST_WANT_ALL_UNSNOOPABLES &&
+ !(orig->mcast_flags & BATADV_MCAST_WANT_ALL_UNSNOOPABLES)) {
+ atomic_inc(&bat_priv->mcast.num_want_all_unsnoopables);
+
+ spin_lock_bh(&bat_priv->mcast.want_lists_lock);
+ hlist_add_head_rcu(&orig->mcast_want_all_unsnoopables_node,
+ &bat_priv->mcast.want_all_unsnoopables_list);
+ spin_unlock_bh(&bat_priv->mcast.want_lists_lock);
+ /* switched from flag set to unset */
+ } else if (!(mcast_flags & BATADV_MCAST_WANT_ALL_UNSNOOPABLES) &&
+ orig->mcast_flags & BATADV_MCAST_WANT_ALL_UNSNOOPABLES) {
+ atomic_dec(&bat_priv->mcast.num_want_all_unsnoopables);
+
+ spin_lock_bh(&bat_priv->mcast.want_lists_lock);
+ hlist_del_rcu(&orig->mcast_want_all_unsnoopables_node);
+ spin_unlock_bh(&bat_priv->mcast.want_lists_lock);
+ }
+}
+
+/**
* batadv_mcast_tvlv_ogm_handler_v1 - process incoming multicast tvlv container
* @bat_priv: the bat priv with all the soft interface information
* @orig: the orig_node of the ogm
@@ -441,6 +556,8 @@ static void batadv_mcast_tvlv_ogm_handler_v1(struct batadv_priv *bat_priv,
(tvlv_value_len >= sizeof(mcast_flags)))
mcast_flags = *(uint8_t *)tvlv_value;
+ batadv_mcast_want_unsnoop_update(bat_priv, orig, mcast_flags);
+
orig->mcast_flags = mcast_flags;
}
@@ -477,4 +594,6 @@ void batadv_mcast_purge_orig(struct batadv_orig_node *orig)
if (!(orig->capabilities & BATADV_ORIG_CAPA_HAS_MCAST))
atomic_dec(&bat_priv->mcast.num_disabled);
+
+ batadv_mcast_want_unsnoop_update(bat_priv, orig, BATADV_NO_FLAGS);
}
diff --git a/packet.h b/packet.h
index e8c483d..0aada24 100644
--- a/packet.h
+++ b/packet.h
@@ -89,6 +89,15 @@ enum batadv_icmp_packettype {
BATADV_PARAMETER_PROBLEM = 12,
};
+/**
+ * enum batadv_mcast_flags - flags for multicast capabilities and settings
+ * @BATADV_MCAST_WANT_ALL_UNSNOOPABLES: we want all packets destined for
+ * 224.0.0.0/24 or ff02::1
+ */
+enum batadv_mcast_flags {
+ BATADV_MCAST_WANT_ALL_UNSNOOPABLES = BIT(0),
+};
+
/* tt data subtypes */
#define BATADV_TT_DATA_TYPE_MASK 0x0F
diff --git a/soft-interface.c b/soft-interface.c
index 1a643fe..db6feca 100644
--- a/soft-interface.c
+++ b/soft-interface.c
@@ -710,6 +710,7 @@ static int batadv_softif_init_late(struct net_device *dev)
bat_priv->mcast.flags = BATADV_NO_FLAGS;
atomic_set(&bat_priv->multicast_mode, 1);
atomic_set(&bat_priv->mcast.num_disabled, 0);
+ atomic_set(&bat_priv->mcast.num_want_all_unsnoopables, 0);
#endif
atomic_set(&bat_priv->gw_mode, BATADV_GW_MODE_OFF);
atomic_set(&bat_priv->gw_sel_class, 20);
diff --git a/types.h b/types.h
index c28fc4a..1a674cb 100644
--- a/types.h
+++ b/types.h
@@ -205,6 +205,8 @@ struct batadv_orig_bat_iv {
* @last_seen: time when last packet from this node was received
* @bcast_seqno_reset: time when the broadcast seqno window was reset
* @mcast_flags: multicast flags announced by the orig node
+ * @mcast_want_all_unsnoop_node: a list node for the
+ * mcast.want_all_unsnoopables list
* @capabilities: announced capabilities of this originator
* @capa_initialized: bitfield to remember whether a capability was initialized
* @last_ttvn: last seen translation table version number
@@ -249,6 +251,7 @@ struct batadv_orig_node {
unsigned long bcast_seqno_reset;
#ifdef CONFIG_BATMAN_ADV_MCAST
uint8_t mcast_flags;
+ struct hlist_node mcast_want_all_unsnoopables_node;
#endif
uint8_t capabilities;
uint8_t capa_initialized;
@@ -619,15 +622,24 @@ struct batadv_priv_dat {
/**
* struct batadv_priv_mcast - per mesh interface mcast data
* @mla_list: list of multicast addresses we are currently announcing via TT
+ * @want_all_unsnoopables_list: a list of orig_nodes wanting all unsnoopable
+ * multicast traffic
* @flags: the flags we have last sent in our mcast tvlv
* @enabled: whether the multicast tvlv is currently enabled
* @num_disabled: number of nodes that have no mcast tvlv
+ * @num_want_all_unsnoopables: number of nodes wanting unsnoopable IP traffic
+ * @want_lists_lock: lock for protecting modifications to mcast want lists
+ * (traversals are rcu-locked)
*/
struct batadv_priv_mcast {
struct hlist_head mla_list;
+ struct hlist_head want_all_unsnoopables_list;
uint8_t flags;
bool enabled;
atomic_t num_disabled;
+ atomic_t num_want_all_unsnoopables;
+ /* protects want_all_unsnoopables_list */
+ spinlock_t want_lists_lock;
};
#endif
--
1.7.10.4
^ permalink raw reply related [flat|nested] 14+ messages in thread
* [B.A.T.M.A.N.] [PATCHv14 6/6] batman-adv: Send multicast packets to nodes with a WANT_ALL flag
2014-02-15 16:47 [B.A.T.M.A.N.] [PATCHv14 0/6] Basic Multicast Optimizations Linus Lüssing
` (4 preceding siblings ...)
2014-02-15 16:47 ` [B.A.T.M.A.N.] [PATCHv14 5/6] batman-adv: Add IPv4 link-local/IPv6-ll-all-nodes multicast support Linus Lüssing
@ 2014-02-15 16:47 ` Linus Lüssing
2014-02-16 6:32 ` Marek Lindner
5 siblings, 1 reply; 14+ messages in thread
From: Linus Lüssing @ 2014-02-15 16:47 UTC (permalink / raw)
To: b.a.t.m.a.n
With this patch a node sends IPv4 multicast packets to nodes which
have a BATADV_MCAST_WANT_ALL_IPV4 flag set and IPv6 multicast packets
to nodes which have a BATADV_MCAST_WANT_ALL_IPV6 flag set, too.
Why is this needed? There are scenarios involving bridges where
multicast report snooping and multicast TT announcements are not
sufficient, which would lead to packet loss for some nodes otherwise:
MLDv1 and IGMPv1/IGMPv2 have a suppression mechanism
for multicast listener reports. When we have an MLDv1/IGMPv1/IGMPv2
querier behind a bridge then our snooping bridge is potentially not
going to see any reports even though listeners exist because according
to RFC4541 such reports are only forwarded to multicast routers:
-----------------------------------------------------------
---------------
{Querier}---|Snoop. Switch|----{Listener}
---------------
\ ^
-------
| br0 | < ???
-------
\
_-~---~_
_-~/ ~-_
~ batman-adv \-----{Sender}
\~_ cloud ~/
-~~__-__-~_/
I) MLDv1 Query: {Querier} -> flooded
II) MLDv1 Report: {Listener} -> {Querier}
-> br0 cannot detect the {Listener}
=> Packets from {Sender} need to be forwarded to all
detected listeners and MLDv1/IGMPv1/IGMPv2 queriers.
-----------------------------------------------------------
Note that we do not need to explicitly forward to MLDv2/IGMPv3 queriers,
because these protocols have no report suppression: A bridge has no
trouble detecting MLDv2/IGMPv3 listeners.
Even though we do not support bridges yet we need to provide the
according infrastructure already to not break compatibility later.
Signed-off-by: Linus Lüssing <linus.luessing@web.de>
---
main.c | 2 +
multicast.c | 178 +++++++++++++++++++++++++++++++++++++++++++++++++++++-
packet.h | 4 ++
send.c | 1 +
soft-interface.c | 2 +
types.h | 14 ++++-
6 files changed, 198 insertions(+), 3 deletions(-)
diff --git a/main.c b/main.c
index 57b09fa..d1183e8 100644
--- a/main.c
+++ b/main.c
@@ -123,6 +123,8 @@ int batadv_mesh_init(struct net_device *soft_iface)
INIT_HLIST_HEAD(&bat_priv->gw.list);
#ifdef CONFIG_BATMAN_ADV_MCAST
INIT_HLIST_HEAD(&bat_priv->mcast.want_all_unsnoopables_list);
+ INIT_HLIST_HEAD(&bat_priv->mcast.want_all_ipv4_list);
+ INIT_HLIST_HEAD(&bat_priv->mcast.want_all_ipv6_list);
#endif
INIT_LIST_HEAD(&bat_priv->tt.changes_list);
INIT_LIST_HEAD(&bat_priv->tt.req_list);
diff --git a/multicast.c b/multicast.c
index ebfbf6e..b2abe00 100644
--- a/multicast.c
+++ b/multicast.c
@@ -386,6 +386,29 @@ static int batadv_mcast_forw_mode_check(struct batadv_priv *bat_priv,
}
/**
+ * batadv_mcast_want_all_ip_count - count nodes with unspecific mcast interest
+ * @bat_priv: the bat priv with all the soft interface information
+ * @ethhdr: ethernet header of a packet
+ *
+ * Return the number of nodes which want all IPv4 multicast traffic if
+ * the given ethhdr is from an IPv4 packet or the number of nodes which want
+ * all IPv6 traffic if it matches an IPv6 packet.
+ */
+static int batadv_mcast_forw_want_all_ip_count(struct batadv_priv *bat_priv,
+ struct ethhdr *ethhdr)
+{
+ switch (ntohs(ethhdr->h_proto)) {
+ case ETH_P_IP:
+ return atomic_read(&bat_priv->mcast.num_want_all_ipv4);
+ case ETH_P_IPV6:
+ return atomic_read(&bat_priv->mcast.num_want_all_ipv6);
+ default:
+ /* we shouldn't be here... */
+ return 0;
+ }
+}
+
+/**
* batadv_mcast_forw_tt_node_get - get a multicast tt node
* @bat_priv: the bat priv with all the soft interface information
* @ethhdr: the ether header containing the multicast destination
@@ -402,6 +425,84 @@ batadv_mcast_forw_tt_node_get(struct batadv_priv *bat_priv,
}
/**
+ * batadv_mcast_want_forw_ipv4_node_get - get a node with an ipv4 flag
+ * @bat_priv: the bat priv with all the soft interface information
+ *
+ * Return an orig_node which has the BATADV_MCAST_WANT_ALL_IPV4 flag set
+ * and increase its refcount.
+ */
+static struct batadv_orig_node *
+batadv_mcast_forw_ipv4_node_get(struct batadv_priv *bat_priv)
+{
+ struct batadv_orig_node *orig_node;
+
+ rcu_read_lock();
+ hlist_for_each_entry_rcu(orig_node,
+ &bat_priv->mcast.want_all_ipv4_list,
+ mcast_want_all_ipv4_node) {
+ if (atomic_inc_not_zero(&orig_node->refcount))
+ goto unlock;
+ }
+
+ orig_node = NULL;
+
+unlock:
+ rcu_read_unlock();
+ return orig_node;
+}
+
+/**
+ * batadv_mcast_want_forw_ipv6_node_get - get a node with an ipv6 flag
+ * @bat_priv: the bat priv with all the soft interface information
+ *
+ * Return an orig_node which has the BATADV_MCAST_WANT_ALL_IPV6 flag set
+ * and increase its refcount.
+ */
+static struct batadv_orig_node *
+batadv_mcast_forw_ipv6_node_get(struct batadv_priv *bat_priv)
+{
+ struct batadv_orig_node *orig_node;
+
+ rcu_read_lock();
+ hlist_for_each_entry_rcu(orig_node,
+ &bat_priv->mcast.want_all_ipv6_list,
+ mcast_want_all_ipv6_node) {
+ if (atomic_inc_not_zero(&orig_node->refcount))
+ goto unlock;
+ }
+
+ orig_node = NULL;
+
+unlock:
+ rcu_read_unlock();
+ return orig_node;
+}
+
+/**
+ * batadv_mcast_want_forw_ip_node_get - get a node with an ipv4/ipv6 flag
+ * @bat_priv: the bat priv with all the soft interface information
+ * @ethhdr: an ethernet header to determine the protocol family from
+ *
+ * Return an orig_node which has the BATADV_MCAST_WANT_ALL_IPV4 or
+ * BATADV_MCAST_WANT_ALL_IPV6 flag, depending on the provided ethhdr, set
+ * and increase its refcount.
+ */
+static struct batadv_orig_node *
+batadv_mcast_forw_ip_node_get(struct batadv_priv *bat_priv,
+ struct ethhdr *ethhdr)
+{
+ switch (ntohs(ethhdr->h_proto)) {
+ case ETH_P_IP:
+ return batadv_mcast_forw_ipv4_node_get(bat_priv);
+ case ETH_P_IPV6:
+ return batadv_mcast_forw_ipv6_node_get(bat_priv);
+ default:
+ /* we shouldn't be here... */
+ return 0;
+ }
+}
+
+/**
* batadv_mcast_want_forw_unsnoop_node_get - get a node with an unsnoopable flag
* @bat_priv: the bat priv with all the soft interface information
*
@@ -444,7 +545,7 @@ batadv_mcast_forw_mode(struct batadv_priv *bat_priv, struct sk_buff *skb,
{
struct ethhdr *ethhdr;
bool is_unsnoopable = false;
- int ret, tt_count, unsnoop_count, total_count;
+ int ret, tt_count, ip_count, unsnoop_count, total_count;
ret = batadv_mcast_forw_mode_check(bat_priv, skb, &is_unsnoopable);
if (ret == -ENOMEM)
@@ -456,15 +557,18 @@ batadv_mcast_forw_mode(struct batadv_priv *bat_priv, struct sk_buff *skb,
tt_count = batadv_tt_global_hash_count(bat_priv, ethhdr->h_dest,
BATADV_NO_FLAGS);
+ ip_count = batadv_mcast_forw_want_all_ip_count(bat_priv, ethhdr);
unsnoop_count = !is_unsnoopable ? 0 :
atomic_read(&bat_priv->mcast.num_want_all_unsnoopables);
- total_count = tt_count + unsnoop_count;
+ total_count = tt_count + ip_count + unsnoop_count;
switch (total_count) {
case 1:
if (tt_count)
*orig = batadv_mcast_forw_tt_node_get(bat_priv, ethhdr);
+ else if (ip_count)
+ *orig = batadv_mcast_forw_ip_node_get(bat_priv, ethhdr);
else if (unsnoop_count)
*orig = batadv_mcast_forw_unsnoop_node_get(bat_priv);
@@ -513,6 +617,72 @@ static void batadv_mcast_want_unsnoop_update(struct batadv_priv *bat_priv,
}
/**
+ * batadv_mcast_want_ipv4_update - update want-all-ipv4 counter and list
+ * @bat_priv: the bat priv with all the soft interface information
+ * @orig: the orig_node which multicast state might have changed of
+ * @mcast_flags: flags indicating the new multicast state
+ *
+ * If the BATADV_MCAST_WANT_ALL_IPV4 flag of this originator, orig, has
+ * toggled then this method updates counter and list accordingly.
+ */
+static void batadv_mcast_want_ipv4_update(struct batadv_priv *bat_priv,
+ struct batadv_orig_node *orig,
+ uint8_t mcast_flags)
+{
+ /* switched from flag unset to set */
+ if (mcast_flags & BATADV_MCAST_WANT_ALL_IPV4 &&
+ !(orig->mcast_flags & BATADV_MCAST_WANT_ALL_IPV4)) {
+ atomic_inc(&bat_priv->mcast.num_want_all_ipv4);
+
+ spin_lock_bh(&bat_priv->mcast.want_lists_lock);
+ hlist_add_head_rcu(&orig->mcast_want_all_ipv4_node,
+ &bat_priv->mcast.want_all_ipv4_list);
+ spin_unlock_bh(&bat_priv->mcast.want_lists_lock);
+ /* switched from flag set to unset */
+ } else if (!(mcast_flags & BATADV_MCAST_WANT_ALL_IPV4) &&
+ orig->mcast_flags & BATADV_MCAST_WANT_ALL_IPV4) {
+ atomic_dec(&bat_priv->mcast.num_want_all_ipv4);
+
+ spin_lock_bh(&bat_priv->mcast.want_lists_lock);
+ hlist_del_rcu(&orig->mcast_want_all_ipv4_node);
+ spin_unlock_bh(&bat_priv->mcast.want_lists_lock);
+ }
+}
+
+/**
+ * batadv_mcast_want_ipv6_update - update want-all-ipv6 counter and list
+ * @bat_priv: the bat priv with all the soft interface information
+ * @orig: the orig_node which multicast state might have changed of
+ * @mcast_flags: flags indicating the new multicast state
+ *
+ * If the BATADV_MCAST_WANT_ALL_IPV6 flag of this originator, orig, has
+ * toggled then this method updates counter and list accordingly.
+ */
+static void batadv_mcast_want_ipv6_update(struct batadv_priv *bat_priv,
+ struct batadv_orig_node *orig,
+ uint8_t mcast_flags)
+{
+ /* switched from flag unset to set */
+ if (mcast_flags & BATADV_MCAST_WANT_ALL_IPV6 &&
+ !(orig->mcast_flags & BATADV_MCAST_WANT_ALL_IPV6)) {
+ atomic_inc(&bat_priv->mcast.num_want_all_ipv6);
+
+ spin_lock_bh(&bat_priv->mcast.want_lists_lock);
+ hlist_add_head_rcu(&orig->mcast_want_all_ipv6_node,
+ &bat_priv->mcast.want_all_ipv6_list);
+ spin_unlock_bh(&bat_priv->mcast.want_lists_lock);
+ /* switched from flag set to unset */
+ } else if (!(mcast_flags & BATADV_MCAST_WANT_ALL_IPV6) &&
+ orig->mcast_flags & BATADV_MCAST_WANT_ALL_IPV6) {
+ atomic_dec(&bat_priv->mcast.num_want_all_ipv6);
+
+ spin_lock_bh(&bat_priv->mcast.want_lists_lock);
+ hlist_del_rcu(&orig->mcast_want_all_ipv6_node);
+ spin_unlock_bh(&bat_priv->mcast.want_lists_lock);
+ }
+}
+
+/**
* batadv_mcast_tvlv_ogm_handler_v1 - process incoming multicast tvlv container
* @bat_priv: the bat priv with all the soft interface information
* @orig: the orig_node of the ogm
@@ -557,6 +727,8 @@ static void batadv_mcast_tvlv_ogm_handler_v1(struct batadv_priv *bat_priv,
mcast_flags = *(uint8_t *)tvlv_value;
batadv_mcast_want_unsnoop_update(bat_priv, orig, mcast_flags);
+ batadv_mcast_want_ipv4_update(bat_priv, orig, mcast_flags);
+ batadv_mcast_want_ipv6_update(bat_priv, orig, mcast_flags);
orig->mcast_flags = mcast_flags;
}
@@ -596,4 +768,6 @@ void batadv_mcast_purge_orig(struct batadv_orig_node *orig)
atomic_dec(&bat_priv->mcast.num_disabled);
batadv_mcast_want_unsnoop_update(bat_priv, orig, BATADV_NO_FLAGS);
+ batadv_mcast_want_ipv4_update(bat_priv, orig, BATADV_NO_FLAGS);
+ batadv_mcast_want_ipv6_update(bat_priv, orig, BATADV_NO_FLAGS);
}
diff --git a/packet.h b/packet.h
index 0aada24..feaa336 100644
--- a/packet.h
+++ b/packet.h
@@ -93,9 +93,13 @@ enum batadv_icmp_packettype {
* enum batadv_mcast_flags - flags for multicast capabilities and settings
* @BATADV_MCAST_WANT_ALL_UNSNOOPABLES: we want all packets destined for
* 224.0.0.0/24 or ff02::1
+ * @BATADV_MCAST_WANT_ALL_IPV4: we want all IPv4 multicast packets
+ * @BATADV_MCAST_WANT_ALL_IPV6: we want all IPv6 multicast packets
*/
enum batadv_mcast_flags {
BATADV_MCAST_WANT_ALL_UNSNOOPABLES = BIT(0),
+ BATADV_MCAST_WANT_ALL_IPV4 = BIT(1),
+ BATADV_MCAST_WANT_ALL_IPV6 = BIT(2),
};
/* tt data subtypes */
diff --git a/send.c b/send.c
index ef4c93e..aad6d4b 100644
--- a/send.c
+++ b/send.c
@@ -27,6 +27,7 @@
#include "originator.h"
#include "network-coding.h"
#include "fragmentation.h"
+#include "multicast.h"
static void batadv_send_outstanding_bcast_packet(struct work_struct *work);
diff --git a/soft-interface.c b/soft-interface.c
index db6feca..744a59b 100644
--- a/soft-interface.c
+++ b/soft-interface.c
@@ -711,6 +711,8 @@ static int batadv_softif_init_late(struct net_device *dev)
atomic_set(&bat_priv->multicast_mode, 1);
atomic_set(&bat_priv->mcast.num_disabled, 0);
atomic_set(&bat_priv->mcast.num_want_all_unsnoopables, 0);
+ atomic_set(&bat_priv->mcast.num_want_all_ipv4, 0);
+ atomic_set(&bat_priv->mcast.num_want_all_ipv6, 0);
#endif
atomic_set(&bat_priv->gw_mode, BATADV_GW_MODE_OFF);
atomic_set(&bat_priv->gw_sel_class, 20);
diff --git a/types.h b/types.h
index 1a674cb..d4b923c 100644
--- a/types.h
+++ b/types.h
@@ -207,6 +207,8 @@ struct batadv_orig_bat_iv {
* @mcast_flags: multicast flags announced by the orig node
* @mcast_want_all_unsnoop_node: a list node for the
* mcast.want_all_unsnoopables list
+ * @mcast_want_all_ipv4_node: a list node for the mcast.want_all_ipv4 list
+ * @mcast_want_all_ipv6_node: a list node for the mcast.want_all_ipv6 list
* @capabilities: announced capabilities of this originator
* @capa_initialized: bitfield to remember whether a capability was initialized
* @last_ttvn: last seen translation table version number
@@ -252,6 +254,8 @@ struct batadv_orig_node {
#ifdef CONFIG_BATMAN_ADV_MCAST
uint8_t mcast_flags;
struct hlist_node mcast_want_all_unsnoopables_node;
+ struct hlist_node mcast_want_all_ipv4_node;
+ struct hlist_node mcast_want_all_ipv6_node;
#endif
uint8_t capabilities;
uint8_t capa_initialized;
@@ -624,21 +628,29 @@ struct batadv_priv_dat {
* @mla_list: list of multicast addresses we are currently announcing via TT
* @want_all_unsnoopables_list: a list of orig_nodes wanting all unsnoopable
* multicast traffic
+ * @want_all_ipv4_list: a list of orig_nodes wanting all IPv4 multicast traffic
+ * @want_all_ipv6_list: a list of orig_nodes wanting all IPv6 multicast traffic
* @flags: the flags we have last sent in our mcast tvlv
* @enabled: whether the multicast tvlv is currently enabled
* @num_disabled: number of nodes that have no mcast tvlv
* @num_want_all_unsnoopables: number of nodes wanting unsnoopable IP traffic
+ * @num_want_all_ipv4: counter for items in want_all_ipv4_list
+ * @num_want_all_ipv6: counter for items in want_all_ipv6_list
* @want_lists_lock: lock for protecting modifications to mcast want lists
* (traversals are rcu-locked)
*/
struct batadv_priv_mcast {
struct hlist_head mla_list;
struct hlist_head want_all_unsnoopables_list;
+ struct hlist_head want_all_ipv4_list;
+ struct hlist_head want_all_ipv6_list;
uint8_t flags;
bool enabled;
atomic_t num_disabled;
atomic_t num_want_all_unsnoopables;
- /* protects want_all_unsnoopables_list */
+ atomic_t num_want_all_ipv4;
+ atomic_t num_want_all_ipv6;
+ /* protects want_all_{unsnoopables,ipv4,ipv6}_list */
spinlock_t want_lists_lock;
};
#endif
--
1.7.10.4
^ permalink raw reply related [flat|nested] 14+ messages in thread
* Re: [B.A.T.M.A.N.] [PATCHv14 1/6] batman-adv: Multicast Listener Announcements via Translation Table
2014-02-15 16:47 ` [B.A.T.M.A.N.] [PATCHv14 1/6] batman-adv: Multicast Listener Announcements via Translation Table Linus Lüssing
@ 2014-02-16 6:23 ` Marek Lindner
0 siblings, 0 replies; 14+ messages in thread
From: Marek Lindner @ 2014-02-16 6:23 UTC (permalink / raw)
To: b.a.t.m.a.n
[-- Attachment #1: Type: text/plain, Size: 823 bytes --]
On Saturday 15 February 2014 17:47:49 Linus Lüssing wrote:
> With this patch a node which has no bridge interface on top of its soft
> interface announces its local multicast listeners via the translation
> table.
>
> Signed-off-by: Linus Lüssing <linus.luessing@web.de>
> ---
> Makefile | 2 +
> Makefile.kbuild | 1 +
> compat.h | 37 ++++++++
> gen-compat-autoconf.sh | 1 +
> main.c | 6 ++
> main.h | 1 +
> multicast.c | 242
> ++++++++++++++++++++++++++++++++++++++++++++++++
> multicast.h | 38 ++++++++
> translation-table.c | 22 ++++-
> types.h | 24 +++++
> 10 files changed, 370 insertions(+), 4 deletions(-)
Applied in revision e368857.
Thanks,
Marek
[-- Attachment #2: This is a digitally signed message part. --]
[-- Type: application/pgp-signature, Size: 490 bytes --]
^ permalink raw reply [flat|nested] 14+ messages in thread
* Re: [B.A.T.M.A.N.] [PATCHv14 2/6] batman-adv: introduce capability initialization bitfield
2014-02-15 16:47 ` [B.A.T.M.A.N.] [PATCHv14 2/6] batman-adv: introduce capability initialization bitfield Linus Lüssing
@ 2014-02-16 6:24 ` Marek Lindner
0 siblings, 0 replies; 14+ messages in thread
From: Marek Lindner @ 2014-02-16 6:24 UTC (permalink / raw)
To: b.a.t.m.a.n
[-- Attachment #1: Type: text/plain, Size: 582 bytes --]
On Saturday 15 February 2014 17:47:50 Linus Lüssing wrote:
> The new bitfield allows us to keep track whether capability subsets of
> an originator have gone through their initialization phase yet.
>
> The translation table is the only user right now, but a new one will be
> added soon.
>
> Signed-off-by: Linus Lüssing <linus.luessing@web.de>
> ---
> originator.c | 1 -
> translation-table.c | 12 +++++++-----
> types.h | 7 ++++---
> 3 files changed, 11 insertions(+), 9 deletions(-)
Applied in revision 5d2121a.
Thanks,
Marek
[-- Attachment #2: This is a digitally signed message part. --]
[-- Type: application/pgp-signature, Size: 490 bytes --]
^ permalink raw reply [flat|nested] 14+ messages in thread
* Re: [B.A.T.M.A.N.] [PATCHv14 3/6] batman-adv: Announce new capability via multicast TVLV
2014-02-15 16:47 ` [B.A.T.M.A.N.] [PATCHv14 3/6] batman-adv: Announce new capability via multicast TVLV Linus Lüssing
@ 2014-02-16 6:25 ` Marek Lindner
0 siblings, 0 replies; 14+ messages in thread
From: Marek Lindner @ 2014-02-16 6:25 UTC (permalink / raw)
To: b.a.t.m.a.n
[-- Attachment #1: Type: text/plain, Size: 768 bytes --]
On Saturday 15 February 2014 17:47:51 Linus Lüssing wrote:
> If the soft interface of a node is not part of a bridge then a node
> announces a new multicast TVLV: The existence of this TVLV
> signalizes that this node is announcing all of its multicast listeners
> via the translation table infrastructure.
>
> Signed-off-by: Linus Lüssing <linus.luessing@web.de>
> ---
> main.c | 1 +
> multicast.c | 121
> +++++++++++++++++++++++++++++++++++++++++++++++++++---
> multicast.h | 14 +++++++
> originator.c | 6 +++
> packet.h | 12 ++++++
> soft-interface.c | 4 ++
> types.h | 13 ++++++
> 7 files changed, 166 insertions(+), 5 deletions(-)
Applied in revision 77ec494.
Thanks,
Marek
[-- Attachment #2: This is a digitally signed message part. --]
[-- Type: application/pgp-signature, Size: 490 bytes --]
^ permalink raw reply [flat|nested] 14+ messages in thread
* Re: [B.A.T.M.A.N.] [PATCHv14 4/6] batman-adv: Modified forwarding behaviour for multicast packets
2014-02-15 16:47 ` [B.A.T.M.A.N.] [PATCHv14 4/6] batman-adv: Modified forwarding behaviour for multicast packets Linus Lüssing
@ 2014-02-16 6:28 ` Marek Lindner
0 siblings, 0 replies; 14+ messages in thread
From: Marek Lindner @ 2014-02-16 6:28 UTC (permalink / raw)
To: b.a.t.m.a.n
[-- Attachment #1: Type: text/plain, Size: 1350 bytes --]
On Saturday 15 February 2014 17:47:52 Linus Lüssing wrote:
> With this patch a multicast packet is not always simply flooded anymore,
> the behaviour for the following cases is changed to reduce
> unnecessary overhead:
>
> If all nodes within the horizon of a certain node have signalized
> multicast listener announcement capability then an IPv6 multicast packet
> with a destination of IPv6 link-local scope (excluding ff02::1) coming
> from the upstream of this node...
>
> * ...is dropped if there is no according multicast listener in the
> translation table,
> * ...is forwarded via unicast if there is a single node with interested
> multicast listeners
> * ...and otherwise still gets flooded.
>
> Signed-off-by: Linus Lüssing <linus.luessing@web.de>
> ---
> multicast.c | 127
> ++++++++++++++++++++++++++++++++++++++++++++++++++
> multicast.h | 25 ++++++++++
> send.c | 10 ++--
> send.h | 5 ++
> soft-interface.c | 20 +++++++-
> sysfs-class-net-mesh | 9 ++++
> sysfs.c | 6 +++
> translation-table.c | 92 ++++++++++++++++++++++++++++--------
> translation-table.h | 2 +
> types.h | 7 +++
> 10 files changed, 277 insertions(+), 26 deletions(-)
Applied in revision 405cc1e.
Thanks,
Marek
[-- Attachment #2: This is a digitally signed message part. --]
[-- Type: application/pgp-signature, Size: 490 bytes --]
^ permalink raw reply [flat|nested] 14+ messages in thread
* Re: [B.A.T.M.A.N.] [PATCHv14 5/6] batman-adv: Add IPv4 link-local/IPv6-ll-all-nodes multicast support
2014-02-15 16:47 ` [B.A.T.M.A.N.] [PATCHv14 5/6] batman-adv: Add IPv4 link-local/IPv6-ll-all-nodes multicast support Linus Lüssing
@ 2014-02-16 6:30 ` Marek Lindner
2015-06-10 8:52 ` Sven Eckelmann
1 sibling, 0 replies; 14+ messages in thread
From: Marek Lindner @ 2014-02-16 6:30 UTC (permalink / raw)
To: b.a.t.m.a.n
[-- Attachment #1: Type: text/plain, Size: 1386 bytes --]
On Saturday 15 February 2014 17:47:53 Linus Lüssing wrote:
> With this patch a node may additionally perform the dropping or
> unicasting behaviour for a link-local IPv4 and link-local-all-nodes
> IPv6 multicast packet, too.
>
> The extra counter and BATADV_MCAST_WANT_ALL_UNSNOOPABLES flag is needed
> because with a future bridge snooping support integration a node with a
> bridge on top of its soft interface is not able to reliably detect its
> multicast listeners for IPv4 link-local and the IPv6
> link-local-all-nodes addresses anymore (see RFC4541, section 2.1.2.2
> and section 3).
>
> Even though this new flag does make "no difference" now, it'll ensure
> a seamless integration of multicast bridge support without needing to
> break compatibility later.
>
> Also note, that even with multicast bridge support it won't be possible
> to optimize 224.0.0.x and ff02::1 towards nodes with bridges, they will
> always receive these ranges.
>
> Signed-off-by: Linus Lüssing <linus.luessing@web.de>
> ---
> main.c | 6 +++
> main.h | 1 +
> multicast.c | 135
> ++++++++++++++++++++++++++++++++++++++++++++++++++----
> packet.h | 9 ++++
> soft-interface.c | 1 +
> types.h | 12 +++++
> 6 files changed, 156 insertions(+), 8 deletions(-)
Applied in revision 1c09034.
Thanks,
Marek
[-- Attachment #2: This is a digitally signed message part. --]
[-- Type: application/pgp-signature, Size: 490 bytes --]
^ permalink raw reply [flat|nested] 14+ messages in thread
* Re: [B.A.T.M.A.N.] [PATCHv14 6/6] batman-adv: Send multicast packets to nodes with a WANT_ALL flag
2014-02-15 16:47 ` [B.A.T.M.A.N.] [PATCHv14 6/6] batman-adv: Send multicast packets to nodes with a WANT_ALL flag Linus Lüssing
@ 2014-02-16 6:32 ` Marek Lindner
0 siblings, 0 replies; 14+ messages in thread
From: Marek Lindner @ 2014-02-16 6:32 UTC (permalink / raw)
To: b.a.t.m.a.n
[-- Attachment #1: Type: text/plain, Size: 2436 bytes --]
On Saturday 15 February 2014 17:47:54 Linus Lüssing wrote:
> With this patch a node sends IPv4 multicast packets to nodes which
> have a BATADV_MCAST_WANT_ALL_IPV4 flag set and IPv6 multicast packets
> to nodes which have a BATADV_MCAST_WANT_ALL_IPV6 flag set, too.
>
> Why is this needed? There are scenarios involving bridges where
> multicast report snooping and multicast TT announcements are not
> sufficient, which would lead to packet loss for some nodes otherwise:
>
> MLDv1 and IGMPv1/IGMPv2 have a suppression mechanism
> for multicast listener reports. When we have an MLDv1/IGMPv1/IGMPv2
> querier behind a bridge then our snooping bridge is potentially not
> going to see any reports even though listeners exist because according
> to RFC4541 such reports are only forwarded to multicast routers:
>
> -----------------------------------------------------------
> ---------------
> {Querier}---|Snoop. Switch|----{Listener}
> ---------------
> \ ^
> -------
> | br0 | < ???
> -------
> \
> _-~---~_
> _-~/ ~-_
> ~ batman-adv \-----{Sender}
> \~_ cloud ~/
> -~~__-__-~_/
>
> I) MLDv1 Query: {Querier} -> flooded
> II) MLDv1 Report: {Listener} -> {Querier}
>
> -> br0 cannot detect the {Listener}
> => Packets from {Sender} need to be forwarded to all
> detected listeners and MLDv1/IGMPv1/IGMPv2 queriers.
>
> -----------------------------------------------------------
>
> Note that we do not need to explicitly forward to MLDv2/IGMPv3 queriers,
> because these protocols have no report suppression: A bridge has no
> trouble detecting MLDv2/IGMPv3 listeners.
>
> Even though we do not support bridges yet we need to provide the
> according infrastructure already to not break compatibility later.
>
> Signed-off-by: Linus Lüssing <linus.luessing@web.de>
> ---
> main.c | 2 +
> multicast.c | 178
> +++++++++++++++++++++++++++++++++++++++++++++++++++++-
> packet.h | 4 ++
> send.c | 1 +
> soft-interface.c | 2 +
> types.h | 14 ++++-
> 6 files changed, 198 insertions(+), 3 deletions(-)
Applied in revision 02705e4.
Thanks,
Marek
[-- Attachment #2: This is a digitally signed message part. --]
[-- Type: application/pgp-signature, Size: 490 bytes --]
^ permalink raw reply [flat|nested] 14+ messages in thread
* Re: [B.A.T.M.A.N.] [PATCHv14 5/6] batman-adv: Add IPv4 link-local/IPv6-ll-all-nodes multicast support
2014-02-15 16:47 ` [B.A.T.M.A.N.] [PATCHv14 5/6] batman-adv: Add IPv4 link-local/IPv6-ll-all-nodes multicast support Linus Lüssing
2014-02-16 6:30 ` Marek Lindner
@ 2015-06-10 8:52 ` Sven Eckelmann
1 sibling, 0 replies; 14+ messages in thread
From: Sven Eckelmann @ 2015-06-10 8:52 UTC (permalink / raw)
To: b.a.t.m.a.n
[-- Attachment #1: Type: text/plain, Size: 1657 bytes --]
On Saturday 15 February 2014 17:47:53 Linus Lüssing wrote:
> +static void batadv_mcast_want_unsnoop_update(struct batadv_priv *bat_priv,
> + struct batadv_orig_node *orig,
> + uint8_t mcast_flags)
> +{
> + /* switched from flag unset to set */
> + if (mcast_flags & BATADV_MCAST_WANT_ALL_UNSNOOPABLES &&
> + !(orig->mcast_flags & BATADV_MCAST_WANT_ALL_UNSNOOPABLES)) {
> + atomic_inc(&bat_priv->mcast.num_want_all_unsnoopables);
> +
> + spin_lock_bh(&bat_priv->mcast.want_lists_lock);
> + hlist_add_head_rcu(&orig->mcast_want_all_unsnoopables_node,
> + &bat_priv->mcast.want_all_unsnoopables_list);
> + spin_unlock_bh(&bat_priv->mcast.want_lists_lock);
> + /* switched from flag set to unset */
> + } else if (!(mcast_flags & BATADV_MCAST_WANT_ALL_UNSNOOPABLES) &&
> + orig->mcast_flags & BATADV_MCAST_WANT_ALL_UNSNOOPABLES) {
> + atomic_dec(&bat_priv->mcast.num_want_all_unsnoopables);
> +
> + spin_lock_bh(&bat_priv->mcast.want_lists_lock);
> + hlist_del_rcu(&orig->mcast_want_all_unsnoopables_node);
> + spin_unlock_bh(&bat_priv->mcast.want_lists_lock);
> + }
Looks unsafe. How do you make sure that it is still in the list and not
already already removed in the meantime? hlist_del_rcu does not re-initialize
the pointers (only POISON them when CONFIG_DEBUG_LIST is activated). Thus
calling hlist_del_rcu again on an object not in the list either kills other
data structures or causes an Oops.
There are more *list_del* related problems of that type in your code. This is
just one example I've picked.
See http://www.open-mesh.org/issues/217#note-5
Kind regards,
Sven
[-- Attachment #2: This is a digitally signed message part. --]
[-- Type: application/pgp-signature, Size: 819 bytes --]
^ permalink raw reply [flat|nested] 14+ messages in thread
end of thread, other threads:[~2015-06-10 8:52 UTC | newest]
Thread overview: 14+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2014-02-15 16:47 [B.A.T.M.A.N.] [PATCHv14 0/6] Basic Multicast Optimizations Linus Lüssing
2014-02-15 16:47 ` [B.A.T.M.A.N.] [PATCHv14 1/6] batman-adv: Multicast Listener Announcements via Translation Table Linus Lüssing
2014-02-16 6:23 ` Marek Lindner
2014-02-15 16:47 ` [B.A.T.M.A.N.] [PATCHv14 2/6] batman-adv: introduce capability initialization bitfield Linus Lüssing
2014-02-16 6:24 ` Marek Lindner
2014-02-15 16:47 ` [B.A.T.M.A.N.] [PATCHv14 3/6] batman-adv: Announce new capability via multicast TVLV Linus Lüssing
2014-02-16 6:25 ` Marek Lindner
2014-02-15 16:47 ` [B.A.T.M.A.N.] [PATCHv14 4/6] batman-adv: Modified forwarding behaviour for multicast packets Linus Lüssing
2014-02-16 6:28 ` Marek Lindner
2014-02-15 16:47 ` [B.A.T.M.A.N.] [PATCHv14 5/6] batman-adv: Add IPv4 link-local/IPv6-ll-all-nodes multicast support Linus Lüssing
2014-02-16 6:30 ` Marek Lindner
2015-06-10 8:52 ` Sven Eckelmann
2014-02-15 16:47 ` [B.A.T.M.A.N.] [PATCHv14 6/6] batman-adv: Send multicast packets to nodes with a WANT_ALL flag Linus Lüssing
2014-02-16 6:32 ` Marek Lindner
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).