B.A.T.M.A.N

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

15 participants
4308 discussions

[B.A.T.M.A.N.] [PATCH 0/5] Preparation for SPEEDY_JOIN/ROAM
by Antonio Quartulli 18 Apr '12

18 Apr '12

Hello everybody, in preparation to my next patchset, which will enable clients to be detected from nodes in the network _before_ they are announced by means of the TT mechanism (more details will come within the companion email email of the related patchset), I'm sending this 4 patches that fix/clean up something here and there plus a little behaviour change in the OGM preparation. In particular (for further details please refer to the commit message of each patch): patch 1) changes the behaviour of how an OGM is built. Instead of preparing/building it at the same time we schedule it, this patch delays such operation to the instant before sending the message. This patch also introduce a new entry in the protocol API called ".bat_fill_ogm". patch 2) in case of ADD and DEL events (or viceversa) of the same client during the same originator interval, such events can be both nullified instead of being sent. patch 3) future operations will require to look for a particular originator in the orig_list of a global entry, therefore tt_global_entry_has_orig() is modified accordingly so that it returns the found originator, if any. patch 4) updates the tt_global_entry ttvn in case of reannouncement of the same client. patch 5) changes tt_global_Add() sign. We use to pass several boolean flags to this function that are then converted to the TT_CLIENT_* ones. This patch modifies the tt_global_add() sign so that we can directly pass a combination of TT_CLIENT_* flags instead of several boolean parameters. Thank you, Antonio

2 15

[B.A.T.M.A.N.] pull request: batman-adv 2012-04-17
by Antonio Quartulli 18 Apr '12

18 Apr '12

Hello, here are a set of changes I would like to propose for next-next/3.5. Most of the patches are simple cleanups and small fixes. The only important change is patch 3 which changes the Originator Message sequence number starting value from 0 to a random generated value. Thank you, Antonio The following changes since commit ecffe75f934b4e3c5301fe5db278068e0efb0d6b: hippi: fix printk format in rrunner.c (2012-04-16 23:48:38 -0400) are available in the git repository at: git://git.open-mesh.org/linux-merge.git tags/batman-adv-for-davem for you to fetch changes up to 520ada6df73ab595f7d51378b9df937371f13460: batman-adv: skip the window protection test when the originator has no neighbours (2012-04-17 13:29:19 +0200) ---------------------------------------------------------------- Included changes: * several cleanups and fixes * OriGinator Message seqno initial value is now random ---------------------------------------------------------------- Antonio Quartulli (3): batman-adv: use ETH_HLEN instead of sizeof(struct ethhdr) batman-adv: print OGM seq numbers as unsigned long batman-adv: skip the window protection test when the originator has no neighbours Marek Lindner (8): batman-adv: move ogm initialization into the proper function batman-adv: refactoring API: find generalized name for bat_ogm_init callback batman-adv: randomize initial seqno to avoid collision batman-adv: add iface_disable() callback to routing API batman-adv: handle routing code initialization properly batman-adv: refactoring API: find generalized name for bat_ogm_init_primary callback batman-adv: rename BATMAN_OGM_LEN to BATMAN_OGM_HLEN batman-adv: mark existing ogm variables as batman iv net/batman-adv/bat_iv_ogm.c | 63 +++++++++++++++++++++----------- net/batman-adv/bridge_loop_avoidance.c | 10 ++--- net/batman-adv/hard-interface.c | 33 +++++++++-------- net/batman-adv/icmp_socket.c | 4 +- net/batman-adv/main.c | 5 ++- net/batman-adv/packet.h | 6 +-- net/batman-adv/routing.c | 10 ++--- net/batman-adv/send.c | 14 +++---- net/batman-adv/soft-interface.c | 2 +- net/batman-adv/types.h | 12 +++--- net/batman-adv/vis.c | 8 ++-- 11 files changed, 95 insertions(+), 72 deletions(-)

2 20

[B.A.T.M.A.N.] Batman as a kernel module
by Mitar 17 Apr '12

17 Apr '12

Hi! What are the reasons (except for performance) of Batman being a kernel module? Is there a way for running it in user-space? We have some servers where we get donated bandwidth, but as people are running also other stuff on those servers they do not really like adding kernel modules and probably even rebooting servers for upgrading to newer Batman versions. Would some compatibility layer be possible to have to run Batman in user-space? Are there any existing known? Mitar

3 3

[B.A.T.M.A.N.] Module crash when reading log and deleting interface
by Martin Hundebøll 17 Apr '12

17 Apr '12

Hi, I can make the module crash by deleting eth0 from bat0, while reading the log with batctl log: root@OpenWrt:/# batctl ll all batman_adv: bat0: log_level: Changing from: 0 to: 31 root@OpenWrt:/# batctl log > /dev/null & root@OpenWrt:/# batctl if del eth0 batman_adv: bat0: Interface deactivated: eth0 batman_adv: bat0: Removing interface: eth0 BUG: unable to handle kernel paging request at 00002000 IP: [<c4961424>] debug_log+0xe4/0x220 [batman_adv] *pde = 00000000 Oops: 0000 [#2] last sysfs file: /sys/devices/pci0000:00/0000:00:03.0/net/eth0/batman_adv/mesh_iface Modules linked in: e1000 batman_adv crc16 Pid: 826, comm: batctl Tainted: G D W (2.6.32.27 #4) Bochs EIP: 0060:[<c4961424>] EFLAGS: 00010246 CPU: 0 EIP is at debug_log+0xe4/0x220 [batman_adv] EAX: c3049000 EBX: 00001000 ECX: 00000000 EDX: 08eca070 ESI: fffffff2 EDI: 00000000 EBP: 08ec9070 ESP: c3049f40 DS: 007b ES: 007b FS: 0000 GS: 0000 SS: 0068 Process batctl (pid: 826, ti=c3049000 task=c3b70360 task.ti=c3049000) Stack: 010611a7 00000000 00000000 00000002 c34b8b90 c34b8b90 c3049f58 00000000 <0> c38cac00 08ec9070 c3049f9c 00001000 c106e4ba c3049f9c 00000000 00000000 <0> c49613b0 c38cac00 fffffff7 00000003 c3049000 c106e8b1 c3049f9c 00000000 Call Trace: [<c106e4ba>] ? vfs_read+0x9a/0x160 [<c49613b0>] ? debug_log+0x70/0x220 [batman_adv] [<c106e8b1>] ? sys_read+0x41/0x70 [<c1002ae5>] ? syscall_call+0x7/0xb Code: 00 31 f6 85 db 0f 84 3c 01 00 00 89 e0 89 ea 25 00 f0 ff ff be f2 ff ff ff 01 da 19 c9 39 50 18 83 d9 00 85 c9 0f 85 1c 01 00 00 <8b> 87 00 20 00 00 3b 87 04 20 00 00 0f 85 84 00 00 00 31 c0 31 EIP: [<c4961424>] debug_log+0xe4/0x220 [batman_adv] SS:ESP 0068:c3049f40 CR2: 0000000000002000 ---[ end trace 4eaa2a86a8e2da26 ]--- [1]+ Killed batctl log 1>/dev/null -- Kind Regards Martin Hundebøll Frederiks Allé 99A, 1.th 8000 Aarhus C Denmark +45 61 65 54 61 martin(a)hundeboll.net

1 0

[B.A.T.M.A.N.] batman-adv 2012.1.0 released
by Marek Lindner 17 Apr '12

17 Apr '12

The B.A.T.M.A.N. team is delighted to announce its newest release, 2012.1.0, bringing mainly bug fixes and code cleanups. Also part of the release is a newly developed routing algorithm framework which allows switching between different routing algorithms. As the kernel module always depends on the Linux kernel it was compiled against, it does not make sense to provide binaries on our website. As usual, you will find the signed tarballs in our download section: http://downloads.open-mesh.org/batman/releases/batman-adv-2012.1.0/ as well as prepackaged binaries in your distribution. Thanks ------ Thanks to all people sending in patches: * Alexey Fisher <bug-track(a)fisher-privat.net> * Antonio Quartulli <ordex(a)autistici.org> * Danny Kukawka <danny.kukawka(a)bisect.de> * Marek Lindner <lindner_marek(a)yahoo.de> * Martin Hundebøll <martin(a)hundeboll.net> * Simon Wunderlich <siwu(a)hrz.tu-chemnitz.de> * Sven Eckelmann <sven(a)narfation.org> batman-adv ---------- Most changes coming with this release happened under the hood to either improve stability or pave the way for new features, most notably the routing algorithm abstraction. Over the last months several ideas centering around the question how to improve the current routing algorithm have surfaced. To better experiment with these ideas while keeping a working system for our users the routing algorithm abstraction was added. It allows to easily develop and test new routing concepts while taking advantage of the existing batman-adv features and infrastructure. The API will be further refined with the next releases to enable alternative algorithms to be included as well. The translation table code received further attention which led to the discovery of endless loops in the tt-request mechanism (under certain circumstances). This was addressed along with a TT_CLIENT_NEW flag race condition when a new client was added to the internal hash. All batman-adv internal timeouts have been converted to milliseconds for consistency purposes. Even the batman-adv makefile was modernized to provide an install target and a selection system for conveniently compiling features into the module. batctl ------ Each batman-adv (layer 2) ICMP packet used to ping or traceroute a node comes with a sequence number to detect failures. Prior to this release batctl ping/traceroute did not compare the outgoing sequence number with the incoming sequence number to verify whether they indeed match. In WiFi environments exhibiting high packet loss and excessive packet retransmissions this missing check led to false results. Furthermore, the return code of batctl ping was improved to properly indicate a failure. Happy routing, The B.A.T.M.A.N. team

1 0

[B.A.T.M.A.N.] [PATCHv2] batman-adv: use static inline functions instead of foo(...)
by Antonio Quartulli 17 Apr '12

17 Apr '12

When the DAT component is not built, instead of having fake function like foo(...) we must use a static inline function in order to allow the compiler to check the number of the type of the passed arguments Signed-off-by: Antonio Quartulli <ordex(a)autistici.org> --- distributed-arp-table.h | 53 ++++++++++++++++++++++++++++++++++++++++------- 1 file changed, 45 insertions(+), 8 deletions(-) diff --git a/distributed-arp-table.h b/distributed-arp-table.h index 26b98c0..fcf6c15 100644 --- a/distributed-arp-table.h +++ b/distributed-arp-table.h @@ -88,14 +88,51 @@ static inline void dat_init_own_dht_addr(struct bat_priv *bat_priv, #else -#define dat_snoop_outgoing_arp_request(...) (0) -#define dat_snoop_incoming_arp_request(...) (0) -#define dat_snoop_outgoing_arp_reply(...) -#define dat_snoop_incoming_arp_reply(...) (0) -#define dat_drop_broadcast_packet(...) (0) -#define arp_change_timeout(...) -#define dat_init_orig_node_dht_addr(...) -#define dat_init_own_dht_addr(...) +static inline bool dat_snoop_outgoing_arp_request(struct bat_priv *bat_priv, + struct sk_buff *skb) +{ + return false; +} + +static inline bool dat_snoop_incoming_arp_request(struct bat_priv *bat_priv, + struct sk_buff *skb, + int hdr_size) +{ + return false; +} + +static inline bool dat_snoop_outgoing_arp_reply(struct bat_priv *bat_priv, + struct sk_buff *skb) +{ + return false; +} + +static inline bool dat_snoop_incoming_arp_reply(struct bat_priv *bat_priv, + struct sk_buff *skb, + int hdr_size) +{ + return false; +} + +static inline bool dat_drop_broadcast_packet(struct bat_priv *bat_priv, + struct forw_packet *forw_packet) +{ + return false; +} + +static inline void dat_init_orig_node_dht_addr(struct orig_node *orig_node) +{ +} + +static inline void dat_init_own_dht_addr(struct bat_priv *bat_priv, + struct hard_iface *primary_if) +{ +} + +static inline void arp_change_timeout(struct net_device *soft_iface, + const char *name) +{ +} #endif /* CONFIG_BATMAN_ADV_DAT */ -- 1.7.9.4

2 1

[B.A.T.M.A.N.] [PATCH 1/3] batman-adv: drop useless argument for arp_change_timeout()
by Antonio Quartulli 17 Apr '12

17 Apr '12

The net_device structure already has a "name" field therefore we don't need to pass the name by using another parameter. Signed-off-by: Antonio Quartulli <ordex(a)autistici.org> --- distributed-arp-table.c | 4 ++-- distributed-arp-table.h | 2 +- soft-interface.c | 2 +- 3 files changed, 4 insertions(+), 4 deletions(-) diff --git a/distributed-arp-table.c b/distributed-arp-table.c index b43bece..5386af0 100644 --- a/distributed-arp-table.c +++ b/distributed-arp-table.c @@ -562,12 +562,12 @@ bool dat_drop_broadcast_packet(struct bat_priv *bat_priv, return false; } -void arp_change_timeout(struct net_device *soft_iface, const char *name) +void arp_change_timeout(struct net_device *soft_iface) { struct in_device *in_dev = in_dev_get(soft_iface); if (!in_dev) { pr_err("Unable to set ARP parameters for the batman interface '%s'\n", - name); + soft_iface->name); return; } diff --git a/distributed-arp-table.h b/distributed-arp-table.h index 26b98c0..c16aa1b 100644 --- a/distributed-arp-table.h +++ b/distributed-arp-table.h @@ -49,7 +49,7 @@ bool dat_snoop_incoming_arp_reply(struct bat_priv *bat_priv, struct sk_buff *skb, int hdr_size); bool dat_drop_broadcast_packet(struct bat_priv *bat_priv, struct forw_packet *forw_packet); -void arp_change_timeout(struct net_device *soft_iface, const char *name); +void arp_change_timeout(struct net_device *soft_iface); /* hash function to choose an entry in a hash table of given size */ /* hash algorithm from http://en.wikipedia.org/wiki/Hash_table */ diff --git a/soft-interface.c b/soft-interface.c index b56dafd..de0ceb2 100644 --- a/soft-interface.c +++ b/soft-interface.c @@ -381,7 +381,7 @@ struct net_device *softif_create(const char *name) goto free_soft_iface; } - arp_change_timeout(soft_iface, name); + arp_change_timeout(soft_iface); bat_priv = netdev_priv(soft_iface); -- 1.7.9.4

2 4

[B.A.T.M.A.N.] [PATCH 0/2] Fixes reported by Al Viro
by Antonio Quartulli 17 Apr '12

17 Apr '12

The following two patches fix the two issues found by Al Viro and reported in his email. Patch 1) simply convert the tt_crc field in the bat_priv structure from atomic_t to uint16_t. Actually there is no reason to declare it as atomic_t. Patch 2) fixes a little bug we have when sending a tt_request message: now we don't convert the tt_crc that we are sending within the request to network order. OTOH we convert this field from NO to HO on the receiver side. Therefore, nodes on the path of the request, which have HO different from NO, will fail to reply to the request and will forward it towards the final destination. Cheers, Antonio

2 7

[B.A.T.M.A.N.] [PATCH] net: cleanup unsigned to unsigned int
by Sven Eckelmann 17 Apr '12

17 Apr '12

From: Eric Dumazet <eric.dumazet(a)gmail.com> Use of "unsigned int" is preferred to bare "unsigned" in net tree. Signed-off-by: Eric Dumazet <eric.dumazet(a)gmail.com> Signed-off-by: David S. Miller <davem(a)davemloft.net> Signed-off-by: Sven Eckelmann <sven(a)narfation.org> --- This one is already in net-next. Please add it to our own next branch. bat_sysfs.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/bat_sysfs.c b/bat_sysfs.c index c6efd68..2c81688 100644 --- a/bat_sysfs.c +++ b/bat_sysfs.c @@ -149,7 +149,7 @@ static int store_bool_attr(char *buff, size_t count, atomic_read(attr) == 1 ? "enabled" : "disabled", enabled == 1 ? "enabled" : "disabled"); - atomic_set(attr, (unsigned)enabled); + atomic_set(attr, (unsigned int)enabled); return count; } @@ -268,7 +268,7 @@ static ssize_t store_vis_mode(struct kobject *kobj, struct attribute *attr, "client" : "server", vis_mode_tmp == VIS_TYPE_CLIENT_UPDATE ? "client" : "server"); - atomic_set(&bat_priv->vis_mode, (unsigned)vis_mode_tmp); + atomic_set(&bat_priv->vis_mode, (unsigned int)vis_mode_tmp); return count; } @@ -354,7 +354,7 @@ static ssize_t store_gw_mode(struct kobject *kobj, struct attribute *attr, curr_gw_mode_str, buff); gw_deselect(bat_priv); - atomic_set(&bat_priv->gw_mode, (unsigned)gw_mode_tmp); + atomic_set(&bat_priv->gw_mode, (unsigned int)gw_mode_tmp); return count; } -- 1.7.9.5

2 1

Re: [B.A.T.M.A.N.] Migration to Batman
by Jernej Kos 16 Apr '12

16 Apr '12

Hello! On 04/12/2012 05:17 PM, Gioacchino Mazzurco wrote: > You need a routing protocol between the gateways to do that, Batman creates Wouldn't just need something like an announce mechanism so each node would add an L3 route to the gateway announcing the said prefix? Since from layer 3, the batman mesh looks just like a big LAN, any node could simply add a route saying "prefix A via gateway B dev bat0" where B is a mesh node known to batman via its L2 identifier, no? Jernej

7 22

Jump to page:

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

B.A.T.M.A.N