B.A.T.M.A.N July 2011

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

15 participants
32 discussions

[B.A.T.M.A.N.] pull request: batman-adv 2011-07-08

by Marek Lindner

Hi, I have another small but important set of changes I'd like to get pulled into net-next-2.6/3.1. Despite our best efforts, recent tests and reviews of the protocol changes we have introduced a couple of weeks ago made it quite obvious that some bugs slipped through. The implementation of said protocol does not handle all corner cases in the way it should, thus provoking inconsistent states in the mesh. These 4 patches are meant to address the found issues. Thanks, Marek

11 years, 8 months

[B.A.T.M.A.N.] [PATCHv3 0/3] Implement the AP-Isolation mechanism

by Antonio Quartulli

This is the third version of the AP-Isolation patchset. The patches have been rebased on top of the master branch but they have been left mostly unmodified. The Ap-isolation mechanism will prevent (if enabled) WIFI clients to communicate with other WIFI clients over the mesh network. WIFI check on a client is done by default while the isolation is performed only if the mechanism is enabled either on the receiver and on the transmitter side. Ap-isolation can be enabled or disabled by means of a new sysfs attribute (a batctl patch has also been provided to toggle interact with it). Thank you all. Regards, Antonio Quartulli

11 years, 8 months

[B.A.T.M.A.N.] [PATCH] batman-adv: request the full table if tt_crc doesn't match

by Antonio Quartulli

In case of tt_crc mismatching for a certain orig_node after applying the changes, the node must request the full table immediately. Signed-off-by: Antonio Quartulli <ordex(a)autistici.org> --- routing.c | 5 +++++ 1 files changed, 5 insertions(+), 0 deletions(-) diff --git a/routing.c b/routing.c index 2cb98be..010ebb7 100644 --- a/routing.c +++ b/routing.c @@ -91,6 +91,11 @@ static void update_transtable(struct bat_priv *bat_priv, * to recompute it to spot any possible inconsistency * in the global table */ orig_node->tt_crc = tt_global_crc(bat_priv, orig_node); + + /* If something went wrong we must request the full table now */ + if (orig_node->tt_crc != tt_crc) + goto request_table; + /* Roaming phase is over: tables are in sync again. I can * unset the flag */ orig_node->tt_poss_change = false; -- 1.7.3.4

11 years, 8 months

[B.A.T.M.A.N.] [PATCHv2 0/3] Solve consistency issues in TT code

by Antonio Quartulli

Hello all, this is the second version of the TT consistency fix patchset. The beautification patches have been removed as they were probably introducing a few new problems (thanks Sven for pointing me to the C99 documentation) while the remaining 4 patches have been merged in a more logical way. Here is a little overview: - 1/3 solves a 'little' bug found today. - 2/3 fixes the way local tables are handled - 3/3 fixes the roaming phase in order to keep consistency For more details, please read the commit messages. Thansk you! Regards, Antonio Quartulli

11 years, 8 months

[B.A.T.M.A.N.] [PATCH 0/6] Solve consistency issues in TT code

by Antonio Quartulli

Hello everybody. In the current TT implementation there are some cases where transtable consistency is not kept. In particular, it is possible to observe this problem in case of a tt_response issued after a request for full-table: if the responding node is _the real destination of the request_ then it will send within the response message also those clients added after the last ttvn increment. This will cause an inconsistency because now we have different nodes with the same ttvn value but with different tables for the same originator. These patches are meant to solve this issue and keep consistency among nodes. Patches 1) and 2) introduce beautifications only. This change helped in writing patches 5) and 6) by reducing the code complexity. I would like to read some feedbacks even about the patch decomposition, which could probably be improved :-) Thank you all. Regards, Antonio Quartulli

11 years, 8 months

[B.A.T.M.A.N.] Disconnecting the nodes in batman-adv 2011.1.0

by Max Ip

Hi all, I have three nodes (connected through wifi) running batman-adv, node_1 ------ node_2 ------node_3 They have ip address ipv4 10.42.43.1 , 2 , 3 respectively. Now, all the nodes are connected to each other. To see the bandwidth and jitter patterns through relaying, I decided to disconnect node_1 and node_3. In node_1 , sudo iptables -A INPUT -s 10.42.43.3 -j DROP When, I now run batman-adv protocol, node_1 still finds node_3 without node_2 being involved. Though they cannot ping and avahi doesn't run, batman-adv shows them as 1 hop neighbors. How can I disconnect node_1 from node_3 so that I can see the 2 hop patterns. Thankyou in advance. Max

11 years, 8 months

[B.A.T.M.A.N.] make batctl 2011.1.0

by Max Ip

Hi all, I downloaded the batctl package 2011.1.0 from the batman website. I installed it using command make all. It generates the following two warnings: functions.c: In function ‘read_file’: functions.c:153:9: warning: ignoring return value of ‘system’, declared with attribute warn_unused_result and bisect.c: In function ‘bisect’: bisect.c:197:9: warning: ‘j’ may be used uninitialized in this function I just ignored the warnings. But, I wasn't able to run the batctl command neither could I see the manual page for batctl. I couldn't figure out what exactly could be the issue though I have been going through the codes and the makefile. Do anyone of you have an idea if my way of installation is correct? Thanking you. Max

11 years, 8 months

[B.A.T.M.A.N.] pull request: batman-adv 2011-07-05

by Marek Lindner

Hi, I'd like to get the following patches pulled into net-next-2.6/3.1. This patchset contains a mix of cleanup code (patch 1-3) for the protocol changes we sent you with our last pull request, 2 bug fixes and an improved version handling patch. Thanks, Marek PS: This is my first pull request - have mercy on my soul and let me know if does not meet your expectations. The following changes since commit 43676ab590c3f8686fd047d34c3e33803eef71f0: batman-adv: improved gateway tq-based selection (2011-06-20 11:37:41 +0200) are available in the git repository at: git://git.open-mesh.org/linux-merge.git batman-adv/next Antonio Quartulli (3): batman-adv: unify flags for tt_change/tt_local_entry/tt_global_entry batman-adv: add_bcast_packet_to_list() takes the sending delay as parameter batman-adv: pass a unique flag arg instead of a sequence of bool ones Marek Lindner (2): batman-adv: broadcast primary OGM on all active hard-interfaces batman-adv: aggregation checks should use the primary_if pointer Sven Eckelmann (1): batman-adv: Replace version info instead of appending them net/batman-adv/aggregation.c | 25 ++++++++++++++++++++----- net/batman-adv/gateway_client.c | 5 ++--- net/batman-adv/main.c | 9 ++------- net/batman-adv/main.h | 11 +++-------- net/batman-adv/originator.c | 5 ++--- net/batman-adv/packet.h | 11 +++++++---- net/batman-adv/routing.c | 2 +- net/batman-adv/send.c | 21 +++++++++++++++------ net/batman-adv/send.h | 2 +- net/batman-adv/soft-interface.c | 2 +- net/batman-adv/translation-table.c | 30 ++++++++++++++---------------- net/batman-adv/translation-table.h | 3 +-- net/batman-adv/types.h | 4 ++-- 13 files changed, 71 insertions(+), 59 deletions(-)

11 years, 8 months

[B.A.T.M.A.N.] [PATCH] batman-adv: Replace version info instead of appending them

by Sven Eckelmann

The version number of modules build outside of the tree can get revision numbers added. This is useful to give hints about the revision of a distribution package and the used patchset. The prepended source number or branch name doesn't add any additional information which would help to identify problems and can therefore be omitted. Signed-off-by: Sven Eckelmann <sven(a)narfation.org> --- Makefile | 2 +- Makefile.kbuild | 2 +- gateway_client.c | 5 ++--- main.c | 9 ++------- main.h | 11 +++-------- originator.c | 5 ++--- 6 files changed, 11 insertions(+), 23 deletions(-) diff --git a/Makefile b/Makefile index 84828ea..5762721 100644 --- a/Makefile +++ b/Makefile @@ -30,7 +30,7 @@ endif export KERNELPATH REVISION= $(shell if [ -d .git ]; then \ - echo $$(git describe --always --dirty 2> /dev/null || echo "[unknown]"); \ + echo $$(git describe --always --dirty --match "v*" |sed 's/^v//' 2> /dev/null || echo "[unknown]"); \ fi) NUM_CPUS = $(shell nproc 2> /dev/null || echo 1) diff --git a/Makefile.kbuild b/Makefile.kbuild index d469668..4dda02b 100644 --- a/Makefile.kbuild +++ b/Makefile.kbuild @@ -28,7 +28,7 @@ endif # EXTRA_CFLAGS += -DCONFIG_BATMAN_ADV_DEBUG ifneq ($(REVISION),) -EXTRA_CFLAGS += -DREVISION_VERSION=\"$(REVISION)\" +EXTRA_CFLAGS += -DSOURCE_VERSION=\"$(REVISION)\" endif obj-m += batman-adv.o diff --git a/gateway_client.c b/gateway_client.c index 8b25b52..056180e 100644 --- a/gateway_client.c +++ b/gateway_client.c @@ -487,10 +487,9 @@ int gw_client_seq_print_text(struct seq_file *seq, void *offset) } seq_printf(seq, " %-12s (%s/%i) %17s [%10s]: gw_class ... " - "[B.A.T.M.A.N. adv %s%s, MainIF/MAC: %s/%pM (%s)]\n", + "[B.A.T.M.A.N. adv %s, MainIF/MAC: %s/%pM (%s)]\n", "Gateway", "#", TQ_MAX_VALUE, "Nexthop", - "outgoingIF", SOURCE_VERSION, REVISION_VERSION_STR, - primary_if->net_dev->name, + "outgoingIF", SOURCE_VERSION, primary_if->net_dev->name, primary_if->net_dev->dev_addr, net_dev->name); rcu_read_lock(); diff --git a/main.c b/main.c index e367e69..b0f9068 100644 --- a/main.c +++ b/main.c @@ -58,9 +58,8 @@ static int __init batman_init(void) register_netdevice_notifier(&hard_if_notifier); - pr_info("B.A.T.M.A.N. advanced %s%s (compatibility version %i) " - "loaded\n", SOURCE_VERSION, REVISION_VERSION_STR, - COMPAT_VERSION); + pr_info("B.A.T.M.A.N. advanced %s (compatibility version %i) " + "loaded\n", SOURCE_VERSION, COMPAT_VERSION); return 0; } @@ -184,8 +183,4 @@ MODULE_LICENSE("GPL"); MODULE_AUTHOR(DRIVER_AUTHOR); MODULE_DESCRIPTION(DRIVER_DESC); MODULE_SUPPORTED_DEVICE(DRIVER_DEVICE); -#ifdef REVISION_VERSION -MODULE_VERSION(SOURCE_VERSION "-" REVISION_VERSION); -#else MODULE_VERSION(SOURCE_VERSION); -#endif diff --git a/main.h b/main.h index 175c9c9..16da679 100644 --- a/main.h +++ b/main.h @@ -27,8 +27,9 @@ #define DRIVER_DESC "B.A.T.M.A.N. advanced" #define DRIVER_DEVICE "batman-adv" -#define SOURCE_VERSION "next" - +#ifndef SOURCE_VERSION +#define SOURCE_VERSION "2011.2.0" +#endif /* B.A.T.M.A.N. parameters */ @@ -146,12 +147,6 @@ enum dbg_level { #include "types.h" -#ifndef REVISION_VERSION -#define REVISION_VERSION_STR "" -#else -#define REVISION_VERSION_STR " "REVISION_VERSION -#endif - extern struct list_head hardif_list; extern unsigned char broadcast_addr[]; diff --git a/originator.c b/originator.c index 338b3c5..4cc94d4 100644 --- a/originator.c +++ b/originator.c @@ -430,9 +430,8 @@ int orig_seq_print_text(struct seq_file *seq, void *offset) goto out; } - seq_printf(seq, "[B.A.T.M.A.N. adv %s%s, MainIF/MAC: %s/%pM (%s)]\n", - SOURCE_VERSION, REVISION_VERSION_STR, - primary_if->net_dev->name, + seq_printf(seq, "[B.A.T.M.A.N. adv %s, MainIF/MAC: %s/%pM (%s)]\n", + SOURCE_VERSION, primary_if->net_dev->name, primary_if->net_dev->dev_addr, net_dev->name); seq_printf(seq, " %-15s %s (%s/%i) %17s [%10s]: %20s ...\n", "Originator", "last-seen", "#", TQ_MAX_VALUE, "Nexthop", -- 1.7.5.4

11 years, 8 months

[B.A.T.M.A.N.] [PATCH] batman-adv: aggregation checks should use the primary_if pointer

by Marek Lindner

The packet aggregation needs to ensure that only compatible packets are aggregated. Some of the checks are based on the interface number while assuming that the first interface also is the primary interface which is not always the case. This patch addresses the issue by using the primary_if pointer. Reported-by: Antonio Quartulli <ordex(a)autistici.org> Signed-off-by: Marek Lindner <lindner_marek(a)yahoo.de> --- aggregation.c | 25 ++++++++++++++++++++----- 1 files changed, 20 insertions(+), 5 deletions(-) diff --git a/aggregation.c b/aggregation.c index c583e04..69467fe 100644 --- a/aggregation.c +++ b/aggregation.c @@ -28,6 +28,7 @@ /* return true if new_packet can be aggregated with forw_packet */ static bool can_aggregate_with(const struct batman_packet *new_batman_packet, + struct bat_priv *bat_priv, int packet_len, unsigned long send_time, bool directlink, @@ -37,6 +38,8 @@ static bool can_aggregate_with(const struct batman_packet *new_batman_packet, struct batman_packet *batman_packet = (struct batman_packet *)forw_packet->skb->data; int aggregated_bytes = forw_packet->packet_len + packet_len; + struct hard_iface *primary_if = NULL; + bool res = false; /** * we can aggregate the current packet to this aggregated packet @@ -61,6 +64,10 @@ static bool can_aggregate_with(const struct batman_packet *new_batman_packet, * packet */ + primary_if = primary_if_get_selected(bat_priv); + if (!primary_if) + goto out; + /* packets without direct link flag and high TTL * are flooded through the net */ if ((!directlink) && @@ -70,8 +77,10 @@ static bool can_aggregate_with(const struct batman_packet *new_batman_packet, /* own packets originating non-primary * interfaces leave only that interface */ ((!forw_packet->own) || - (forw_packet->if_incoming->if_num == 0))) - return true; + (forw_packet->if_incoming == primary_if))) { + res = true; + goto out; + } /* if the incoming packet is sent via this one * interface only - we still can aggregate */ @@ -84,11 +93,16 @@ static bool can_aggregate_with(const struct batman_packet *new_batman_packet, * (= secondary interface packets in general) */ (batman_packet->flags & DIRECTLINK || (forw_packet->own && - forw_packet->if_incoming->if_num != 0))) - return true; + forw_packet->if_incoming != primary_if))) { + res = true; + goto out; + } } - return false; +out: + if (primary_if) + hardif_free_ref(primary_if); + return res; } /* create a new aggregated packet and add this packet to it */ @@ -210,6 +224,7 @@ void add_bat_packet_to_list(struct bat_priv *bat_priv, hlist_for_each_entry(forw_packet_pos, tmp_node, &bat_priv->forw_bat_list, list) { if (can_aggregate_with(batman_packet, + bat_priv, packet_len, send_time, direct_link, -- 1.7.5.3

11 years, 8 months

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

B.A.T.M.A.N July 2011