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 1/2] alfred: Externalized synchronization interval
by Jonathan Haws 06 Sep '16

06 Sep '16

Signed-off-by: Jonathan Haws <jhaws(a)sdl.usu.edu> ALFRED_INTERVAL is now externalized via the -p option (synchronization period). If specified as option, user supplied interval is used, otherwise the default interval of ALFRED_INTERVAL is used. --- alfred.h | 1 + main.c | 14 +++++++++++++- man/alfred.8 | 5 +++++ server.c | 4 ++-- 4 files changed, 21 insertions(+), 3 deletions(-) diff --git a/alfred.h b/alfred.h index a9906c8..5b7e965 100644 --- a/alfred.h +++ b/alfred.h @@ -131,6 +131,7 @@ struct globals { uint16_t changed_data_type_count; /* maximum is 256 */ struct timespec if_check; + struct timespec sync_period; struct hashtable_t *data_hash; struct hashtable_t *transaction_hash; diff --git a/main.c b/main.c index 9cab705..3aa89fd 100644 --- a/main.c +++ b/main.c @@ -59,6 +59,8 @@ static void alfred_usage(void) printf(" -m, --master start up the daemon in master mode, which\n"); printf(" accepts data from slaves and syncs it with\n"); printf(" other masters\n"); + printf(" -p, --sync-period [period] set synchronization period, in seconds\n"); + printf(" fractional seconds are supported (i.e. 0.2 = 5 Hz)\n"); printf("\n"); printf(" -u, --unix-path [path] path to unix socket used for client-server\n"); printf(" communication (default: \""ALFRED_SOCK_PATH_DEFAULT"\")\n"); @@ -156,6 +158,7 @@ out: static struct globals *alfred_init(int argc, char *argv[]) { int opt, opt_ind, i, ret; + double sync_period = 0.0; struct globals *globals; struct option long_options[] = { {"set-data", required_argument, NULL, 's'}, @@ -170,6 +173,7 @@ static struct globals *alfred_init(int argc, char *argv[]) {"update-command", required_argument, NULL, 'c'}, {"version", no_argument, NULL, 'v'}, {"verbose", no_argument, NULL, 'd'}, + {"sync-period", required_argument, NULL, 'p'}, {NULL, 0, NULL, 0}, }; @@ -193,12 +197,14 @@ static struct globals *alfred_init(int argc, char *argv[]) globals->unix_path = ALFRED_SOCK_PATH_DEFAULT; globals->verbose = 0; globals->update_command = NULL; + globals->sync_period.tv_sec = ALFRED_INTERVAL; + globals->sync_period.tv_nsec = 0; INIT_LIST_HEAD(&globals->changed_data_types); globals->changed_data_type_count = 0; time_random_seed(); - while ((opt = getopt_long(argc, argv, "ms:r:hi:b:vV:M:I:u:dc:", long_options, + while ((opt = getopt_long(argc, argv, "ms:r:hi:b:vV:M:I:u:dc:p:", long_options, &opt_ind)) != -1) { switch (opt) { case 'r': @@ -265,6 +271,12 @@ static struct globals *alfred_init(int argc, char *argv[]) printf("%s %s\n", argv[0], SOURCE_VERSION); printf("A.L.F.R.E.D. - Almighty Lightweight Remote Fact Exchange Daemon\n"); return NULL; + case 'p': + sync_period = strtod(optarg, NULL); + globals->sync_period.tv_sec = (int) sync_period; + globals->sync_period.tv_nsec = (double)(sync_period - (int)sync_period) * 1e9; + printf(" ** Setting sync interval to: %.9f seconds (%ld.%09ld)\n", sync_period, globals->sync_period.tv_sec, globals->sync_period.tv_nsec); + break; case 'h': default: alfred_usage(); diff --git a/man/alfred.8 b/man/alfred.8 index 49d42bc..df37040 100644 --- a/man/alfred.8 +++ b/man/alfred.8 @@ -118,6 +118,11 @@ overhead). \fB\-c\fP, \fB\-\-update-command\fP \fIcommand\fP Specify command to execute on data change. It will be called with data-type list as arguments. +.TP +\fB\-p\fP, \fB\-\-sync-period\fP \fIperiod\fP +Specify alfred synchronization period, in seconds. If not specified, the default +ALFRED_INTERVAL setting of 10 seconds will be used. Fractional seconds are +supported. . .SH EXAMPLES Start an alfred server listening on bridge br0 (assuming that this bridge diff --git a/server.c b/server.c index 47aee4f..c5df945 100644 --- a/server.c +++ b/server.c @@ -372,7 +372,7 @@ int alfred_server(struct globals *globals) while (1) { clock_gettime(CLOCK_MONOTONIC, &now); - now.tv_sec -= ALFRED_INTERVAL; + time_diff(&now, &globals->sync_period, &now); if (!time_diff(&last_check, &now, &tv)) { tv.tv_sec = 0; tv.tv_nsec = 0; @@ -409,7 +409,7 @@ int alfred_server(struct globals *globals) if (globals->opmode == OPMODE_MASTER) { /* we are a master */ - printf("announce master ...\n"); + printf("[%ld.%09ld] announce master ...\n", last_check.tv_sec, last_check.tv_nsec); announce_master(globals); sync_data(globals); } else { -- 2.7.4

4 5

[B.A.T.M.A.N.] 802.1x on top of batman-adv
by dan 04 Sep '16

04 Sep '16

any consideration to making a batman-adv 802.1x interface? or anyone tested successfully using hostapd for wireless and wired clients and 802.1x? Combining batman-adv's layer2 mesh and some NAC would be very handy.

2 1

[B.A.T.M.A.N.] Rebroadcast Avoidance #2 / TVLV API generalization
by Linus Lüssing 04 Sep '16

04 Sep '16

Hi, Just wanted to let you know that, as I promised I would after the multicast part being merged, I'm now looking into an automatization for the no-rebroadcast flag patch. In Gluon, this patch is currently used in two different use-cases: A) Avoid rebroadcasts on VPN links over e.g. ADSL -> upstream is costly / low bandwidth B) Avoid rebroadcasts on wired interfaces -> link becomes "noisy" with a bunch of nodes quickly thanks to OGM and broadcast packet rebroadcasts => O(n^2) The former case is addressed with the simple rebroadcast avoidance patch currently pending in patchwork. For the latter, I hacked some code into this branch, which seems to do what it is supposed to: https://git.open-mesh.org/batman-adv.git/shortlog/refs/heads/linus/neighhash (top three commits - warning lot's of code cleanup still to do) As a goody, it's now so generic, that it works on wireless, too :-). The basic idea of using a simple hash to describe/summarize the neighborhood of an interface comes from Henning, when we were discussing rebroadcasts at the last Wireless Battle Mesh during dinner. Kudos to him! Anyways, before going into details about the neighborhood hash approach (I'll document and illustrate it in the wiki soon, anyway), I wanted to ask for some feedback for the second patch ("batman-adv: unify OGM+Unicast TVLV handler functions (WIP)"), a prerequisite to easily add TVLV support to ELP: I) What do you think about the proposal for the TVLV handler API? (e.g. generalizing by adding a packet_type attribute) ~~~ void batadv_tvlv_handler_register(struct batadv_priv *bat_priv, int (*handler)(struct batadv_priv *bat_priv, void *tvlv_value, u16 tvlv_value_len, void *ctx), u8 packet_type, u8 tvlv_type, u8 tvlv_version, u8 flags); void batadv_tvlv_handler_unregister(struct batadv_priv *bat_priv, u8 packet_type, u8 tvlv_type, u8 tvlv_version); int batadv_tvlv_containers_process(struct batadv_priv *bat_priv, const struct sk_buff *skb, u8 packet_type, unsigned int tvlv_offset, u16 tvlv_value_len, void *ctx) ~~~ II) If you like it, would you prefer having both APIs simultaneously for a limited time to be able to migrate TVLVs one by one once they were thoroughly tested (con: code size). Or dragging everything to the new API in one patchset? (con: more invasive, potentially more regressions?) Regards, Linus

1 0

[B.A.T.M.A.N.] batman-adv 2016.3 released
by Simon Wunderlich 02 Sep '16

02 Sep '16

Sep 1st, 2016. Today the B.A.T.M.A.N. team publishes the August 2016 update to batman-adv, batctl and alfred! This release contains the multicast optimization support for bridged setups and frame priority support for fragmented packets. The initial set of patches to implement netlink support for batman-adv were integrated. As the first step, it is used to interact with the new kernel throughput benchmark code. Future releases will utilize netlink to replace the current debugfs tables in a more memory efficient way. Also several bugfixes, code cleanups and code documentation updates are included in this version. As from this release, we will report the summaries from our Changelog for each component instead of a full text. As the kernel module always depends on the Linux kernel it is 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: https://downloads.open-mesh.org/batman/releases/batman-adv-2016.3/ as well as prepackaged binaries in your distribution. Thanks Thanks to all people sending in patches: * Andrew Lunn <andrew(a)lunn.ch> * Antonio Quartulli <a(a)unstable.cc> * Florian Westphal <fw(a)strlen.de> * Linus Lüssing <linus.luessing(a)c0d3.blue> * Marek Lindner <mareklindner(a)neomailbox.ch> * Markus Pargmann <mpa(a)pengutronix.de> * Matthias Schiffer <mschiffer(a)universe-factory.net> * Simon Wunderlich <sw(a)simonwunderlich.de> * Sven Eckelmann <sven(a)narfation.org> batman-adv ---------- * support latest kernels (3.2 - 4.8) * multicast optimization support for bridged setups * initial, limited support for batman-adv netlink family * throughput meter implementation * support for frame priority in fragment packets * kernel doc updates and coding style cleanups * bugs squashed: - fix (re-)initialization of ELP tweaking options - fix elp packet data reservation - fix nullptr dereference after using vlan_insert_tag - fix various reference counting bugs in bonding, bla, tt and originator code - fix speedy join in gateway client mode batctl ------ * synchronize common headers with batman-adv * support multicast logging and debug table * split tcpdump OGM packet filter in OGM and OGMv2 filter * add infrastructure to communicate with batadv netlink family * integrate command to control new kernel throughput meter alfred ------ * integrate support for batman-adv netlink Happy routing, The B.A.T.M.A.N. team

1 0

[B.A.T.M.A.N.] [PATCH] cfg80211: Add stub for cfg80211_get_station()
by Linus Lüssing 30 Aug '16

30 Aug '16

This allows modules using this function (currently: batman-adv) to compile even if cfg80211 is not built at all, thus relaxing dependencies. Signed-off-by: Linus Lüssing <linus.luessing(a)c0d3.blue> --- include/net/cfg80211.h | 9 +++++++++ 1 file changed, 9 insertions(+) diff --git a/include/net/cfg80211.h b/include/net/cfg80211.h index 9c23f4d3..beb7610 100644 --- a/include/net/cfg80211.h +++ b/include/net/cfg80211.h @@ -1102,6 +1102,7 @@ struct station_info { struct cfg80211_tid_stats pertid[IEEE80211_NUM_TIDS + 1]; }; +#if IS_ENABLED(CONFIG_CFG80211) /** * cfg80211_get_station - retrieve information about a given station * @dev: the device where the station is supposed to be connected to @@ -1114,6 +1115,14 @@ struct station_info { */ int cfg80211_get_station(struct net_device *dev, const u8 *mac_addr, struct station_info *sinfo); +#else +static inline int cfg80211_get_station(struct net_device *dev, + const u8 *mac_addr, + struct station_info *sinfo) +{ + return -ENOENT; +} +#endif /** * enum monitor_flags - monitor flags -- 2.1.4

2 2

[B.A.T.M.A.N.] [PATCH] batman-adv: Use octal permissions instead of octal
by Sven Eckelmann 28 Aug '16

28 Aug '16

Linus prefers to have octal permission numbers instead of combinations of macro names ("random line noise"). Also old existing "bad symbolic permission bit macro use" should be converted to octal numbers. (http://lkml.kernel.org/r/CA+55aFw5v23T-zvDZp-MmD_EYxF8WbafwwB59934FV7g21uMG…) Also remove the S_IFREG bit from the octal representation because it is filtered out by debugfs_create. Signed-off-by: Sven Eckelmann <sven(a)narfation.org> --- net/batman-adv/icmp_socket.c | 4 ++-- net/batman-adv/log.c | 3 +-- net/batman-adv/network-coding.c | 7 +++---- 3 files changed, 6 insertions(+), 8 deletions(-) diff --git a/net/batman-adv/icmp_socket.c b/net/batman-adv/icmp_socket.c index 378cc11..aed8125 100644 --- a/net/batman-adv/icmp_socket.c +++ b/net/batman-adv/icmp_socket.c @@ -322,8 +322,8 @@ int batadv_socket_setup(struct batadv_priv *bat_priv) if (!bat_priv->debug_dir) goto err; - d = debugfs_create_file(BATADV_ICMP_SOCKET, S_IFREG | S_IWUSR | S_IRUSR, - bat_priv->debug_dir, bat_priv, &batadv_fops); + d = debugfs_create_file(BATADV_ICMP_SOCKET, 0600, bat_priv->debug_dir, + bat_priv, &batadv_fops); if (!d) goto err; diff --git a/net/batman-adv/log.c b/net/batman-adv/log.c index 56dc532..46b0725 100644 --- a/net/batman-adv/log.c +++ b/net/batman-adv/log.c @@ -212,8 +212,7 @@ int batadv_debug_log_setup(struct batadv_priv *bat_priv) spin_lock_init(&bat_priv->debug_log->lock); init_waitqueue_head(&bat_priv->debug_log->queue_wait); - d = debugfs_create_file("log", S_IFREG | S_IRUSR, - bat_priv->debug_dir, bat_priv, + d = debugfs_create_file("log", 0400, bat_priv->debug_dir, bat_priv, &batadv_log_fops); if (!d) goto err; diff --git a/net/batman-adv/network-coding.c b/net/batman-adv/network-coding.c index e3baf69..0fc3b5f 100644 --- a/net/batman-adv/network-coding.c +++ b/net/batman-adv/network-coding.c @@ -1961,17 +1961,16 @@ int batadv_nc_init_debugfs(struct batadv_priv *bat_priv) if (!nc_dir) goto out; - file = debugfs_create_u8("min_tq", S_IRUGO | S_IWUSR, nc_dir, - &bat_priv->nc.min_tq); + file = debugfs_create_u8("min_tq", 0644, nc_dir, &bat_priv->nc.min_tq); if (!file) goto out; - file = debugfs_create_u32("max_fwd_delay", S_IRUGO | S_IWUSR, nc_dir, + file = debugfs_create_u32("max_fwd_delay", 0644, nc_dir, &bat_priv->nc.max_fwd_delay); if (!file) goto out; - file = debugfs_create_u32("max_buffer_time", S_IRUGO | S_IWUSR, nc_dir, + file = debugfs_create_u32("max_buffer_time", 0644, nc_dir, &bat_priv->nc.max_buffer_time); if (!file) goto out; -- 2.9.3

1 0

[B.A.T.M.A.N.] [PATCH maint v2] batman-adv: fix elp packet data reservation
by Linus Lüssing 26 Aug '16

26 Aug '16

The skb_reserve() call only reserved headroom for the mac header, but not the elp packet header itself. Fixing this by using skb_put()'ing towards the skb tail instead of skb_push()'ing towards the skb head. Fixes: a4b88af77e28 ("batman-adv: ELP - adding basic infrastructure") Signed-off-by: Linus Lüssing <linus.luessing(a)c0d3.blue> --- Changes in v2: * just switch from skb_push() to skb_put() => skb_put() actually returns the right point in the skb and is equivalent to a second skb->data assignment after skb_put() in this case * batamn-adv -> batman-adv * explicitly directed to maint net/batman-adv/bat_v_elp.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/batman-adv/bat_v_elp.c b/net/batman-adv/bat_v_elp.c index df42eb1..63762b3 100644 --- a/net/batman-adv/bat_v_elp.c +++ b/net/batman-adv/bat_v_elp.c @@ -334,7 +334,7 @@ int batadv_v_elp_iface_enable(struct batadv_hard_iface *hard_iface) goto out; skb_reserve(hard_iface->bat_v.elp_skb, ETH_HLEN + NET_IP_ALIGN); - elp_buff = skb_push(hard_iface->bat_v.elp_skb, BATADV_ELP_HLEN); + elp_buff = skb_put(hard_iface->bat_v.elp_skb, BATADV_ELP_HLEN); elp_packet = (struct batadv_elp_packet *)elp_buff; memset(elp_packet, 0, BATADV_ELP_HLEN); -- 2.1.4

3 2

[B.A.T.M.A.N.] [PATCH maint] batman-adv: Add missing refcnt for last_candidate
by Sven Eckelmann 26 Aug '16

26 Aug '16

batadv_find_router dereferences last_bonding_candidate from orig_node without making sure that it has a valid reference. This reference has to be retrieved by increasing the reference counter while holding neigh_list_lock. The lock is required to avoid that batadv_last_bonding_replace removes the current last_bonding_candidate, reduces the reference counter and maybe destroys the object in this process. Fixes: 797edd9e87ac ("batman-adv: add bonding again") Signed-off-by: Sven Eckelmann <sven(a)narfation.org> --- This patch was only build tested. --- net/batman-adv/routing.c | 28 +++++++++++++++++++++++++++- 1 file changed, 27 insertions(+), 1 deletion(-) diff --git a/net/batman-adv/routing.c b/net/batman-adv/routing.c index bfac086..69d425d 100644 --- a/net/batman-adv/routing.c +++ b/net/batman-adv/routing.c @@ -456,6 +456,29 @@ static int batadv_check_unicast_packet(struct batadv_priv *bat_priv, } /** + * batadv_last_bonding_get - Get last_bonding_candidate of orig_node + * @orig_node: originator node whose last bonding candidate should be retrieved + * + * Return: last bonding candidate of router or NULL if not found + * + * The object is returned with refcounter increased by 1. + */ +static struct batadv_orig_ifinfo * +batadv_last_bonding_get(struct batadv_orig_node *orig_node) +{ + struct batadv_orig_ifinfo *last_bonding_candidate; + + spin_lock_bh(&orig_node->neigh_list_lock); + last_bonding_candidate = orig_node->last_bonding_candidate; + + if (last_bonding_candidate) + kref_get(&last_bonding_candidate->refcount); + spin_unlock_bh(&orig_node->neigh_list_lock); + + return last_bonding_candidate; +} + +/** * batadv_last_bonding_replace - Replace last_bonding_candidate of orig_node * @orig_node: originator node whose bonding candidates should be replaced * @new_candidate: new bonding candidate or NULL @@ -525,7 +548,7 @@ batadv_find_router(struct batadv_priv *bat_priv, * router - obviously there are no other candidates. */ rcu_read_lock(); - last_candidate = orig_node->last_bonding_candidate; + last_candidate = batadv_last_bonding_get(orig_node); if (last_candidate) last_cand_router = rcu_dereference(last_candidate->router); @@ -617,6 +640,9 @@ next: batadv_orig_ifinfo_put(next_candidate); } + if (last_candidate) + batadv_orig_ifinfo_put(last_candidate); + return router; } -- 2.8.1

2 1

[B.A.T.M.A.N.] Consistent bat0 hardware address?
by Jonathan Haws 26 Aug '16

26 Aug '16

Is there a way to ensure that the assigned hardware address for bat0 is consistent? What I would like to have is keep the hardware address the same across reboots. Can this be done? Would it be done through /etc/udev/rules.d? Or some other mechanism? Thanks! Jon -- Jonathan R. Haws Embedded Engineer Space Dynamics Laboratory jhaws(a)sdl.usu.edu

2 1

[B.A.T.M.A.N.] [PATCH 1/2] alfred: Externalized synchronization interval
by Jonathan Haws 25 Aug '16

25 Aug '16

2 2

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