B.A.T.M.A.N April 2012

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

25 participants
63 discussions

[B.A.T.M.A.N.] [PATCH] batman-adv: Add get_ethtool_stats() support
by Martin Hundebøll 22 Apr '12

22 Apr '12

Added additional counters in a bat_stats structure, which are exported through the ethtool api. The counters are specific to batman-adv and includes: forwarded packets management packets (OGMs at this point) translation table packets distributed arp table packets I would like you all to check if the increments are added at the right locations and also if more counters would be relevant. (E.g. in bridge loop avoidance code?) This is the reworked approach from the previous stat counters patch I send, where ethtool stats was suggested. Signed-off-by: Martin Hundebøll <martin(a)hundeboll.net> --- bat_iv_ogm.c | 6 ++++- distributed-arp-table.c | 8 ++++++- routing.c | 9 +++++++ soft-interface.c | 61 +++++++++++++++++++++++++++++++++++++++++++++++ translation-table.c | 8 +++++++ types.h | 17 +++++++++++++ 6 files changed, 107 insertions(+), 2 deletions(-) diff --git a/bat_iv_ogm.c b/bat_iv_ogm.c index 994369d..288326a 100644 --- a/bat_iv_ogm.c +++ b/bat_iv_ogm.c @@ -196,8 +196,10 @@ static void bat_iv_ogm_send_to_if(struct forw_packet *forw_packet, /* create clone because function is called more than once */ skb = skb_clone(forw_packet->skb, GFP_ATOMIC); - if (skb) + if (skb) { + bat_priv->bat_stats.mgmt_tx++; send_skb_packet(skb, hard_iface, broadcast_addr); + } } /* send a batman ogm packet */ @@ -956,6 +958,8 @@ static void bat_iv_ogm_process(const struct ethhdr *ethhdr, if (batman_ogm_packet->header.packet_type != BAT_IV_OGM) return; + bat_priv->bat_stats.mgmt_rx++; + /* could be changed by schedule_own_packet() */ if_incoming_seqno = atomic_read(&if_incoming->seqno); diff --git a/distributed-arp-table.c b/distributed-arp-table.c index b43bece..cc3182e 100644 --- a/distributed-arp-table.c +++ b/distributed-arp-table.c @@ -395,9 +395,11 @@ bool dat_snoop_outgoing_arp_request(struct bat_priv *bat_priv, netif_rx(skb_new); bat_dbg(DBG_DAT, bat_priv, "ARP request replied locally\n"); - } else + } else { /* Send the request on the DHT */ + bat_priv->bat_stats.dat_request_tx++; ret = dht_send_data(bat_priv, skb, ip_dst, BAT_P_DAT_DHT_GET); + } out: if (n) neigh_release(n); @@ -450,6 +452,8 @@ bool dat_snoop_incoming_arp_request(struct bat_priv *bat_priv, if (!skb_new) goto out; + bat_priv->bat_stats.dat_reply_tx++; + unicast_4addr_send_skb(skb_new, bat_priv, BAT_P_DAT_CACHE_REPLY); ret = true; @@ -488,6 +492,8 @@ bool dat_snoop_outgoing_arp_reply(struct bat_priv *bat_priv, arp_neigh_update(bat_priv, ip_src, hw_src); arp_neigh_update(bat_priv, ip_dst, hw_dst); + bat_priv->bat_stats.dat_reply_tx++; + /* Send the ARP reply to the candidates for both the IP addresses we * fetched from the ARP reply */ dht_send_data(bat_priv, skb, ip_src, BAT_P_DAT_DHT_PUT); diff --git a/routing.c b/routing.c index 795d3af..28ce7b9 100644 --- a/routing.c +++ b/routing.c @@ -600,6 +600,8 @@ int recv_tt_query(struct sk_buff *skb, struct hard_iface *recv_if) switch (tt_query->flags & TT_QUERY_TYPE_MASK) { case TT_REQUEST: + bat_priv->bat_stats.tt_request_rx++; + /* If we cannot provide an answer the tt_request is * forwarded */ if (!send_tt_response(bat_priv, tt_query)) { @@ -612,6 +614,8 @@ int recv_tt_query(struct sk_buff *skb, struct hard_iface *recv_if) } break; case TT_RESPONSE: + bat_priv->bat_stats.tt_response_rx++; + if (is_my_mac(tt_query->dst)) { /* packet needs to be linearized to access the TT * changes */ @@ -663,6 +667,8 @@ int recv_roam_adv(struct sk_buff *skb, struct hard_iface *recv_if) if (is_broadcast_ether_addr(ethhdr->h_source)) goto out; + bat_priv->bat_stats.tt_roam_adv_rx++; + roam_adv_packet = (struct roam_adv_packet *)skb->data; if (!is_my_mac(roam_adv_packet->dst)) @@ -869,6 +875,9 @@ static int route_unicast_packet(struct sk_buff *skb, struct hard_iface *recv_if) /* decrement ttl */ unicast_packet->header.ttl--; + /* Update stats counter */ + bat_priv->bat_stats.forward++; + /* route it */ send_skb_packet(skb, neigh_node->if_incoming, neigh_node->addr); ret = NET_RX_SUCCESS; diff --git a/soft-interface.c b/soft-interface.c index 92137af..095ee83 100644 --- a/soft-interface.c +++ b/soft-interface.c @@ -46,6 +46,10 @@ static void bat_get_drvinfo(struct net_device *dev, static u32 bat_get_msglevel(struct net_device *dev); static void bat_set_msglevel(struct net_device *dev, u32 value); static u32 bat_get_link(struct net_device *dev); +static void bat_get_strings(struct net_device *dev, u32 stringset, u8 *data); +static void bat_get_ethtool_stats(struct net_device *dev, + struct ethtool_stats *stats, u64 *data); +static int bat_get_sset_count(struct net_device *dev, int stringset); static const struct ethtool_ops bat_ethtool_ops = { .get_settings = bat_get_settings, @@ -53,6 +57,9 @@ static const struct ethtool_ops bat_ethtool_ops = { .get_msglevel = bat_get_msglevel, .set_msglevel = bat_set_msglevel, .get_link = bat_get_link, + .get_strings = bat_get_strings, + .get_ethtool_stats = bat_get_ethtool_stats, + .get_sset_count = bat_get_sset_count, }; int my_skb_head_push(struct sk_buff *skb, unsigned int len) @@ -498,3 +505,57 @@ static u32 bat_get_link(struct net_device *dev) { return 1; } + +/* Inspired by drivers/net/ethernet/dlink/sundance.c:1702 */ +static const struct { + const char name[ETH_GSTRING_LEN]; +} bat_stats_strings[] = { + { "forward" }, + { "mgmt_tx" }, + { "mgmt_rx" }, + { "tt_request_tx" }, + { "tt_request_rx" }, + { "tt_response_tx" }, + { "tt_response_rx" }, + { "tt_roam_adv_tx" }, + { "tt_roam_adv_rx" }, + { "dat_request_tx" }, + { "dat_request_rx" }, + { "dat_reply_tx" }, + { "dat_reply_rx" }, +}; +#define BAT_STATS_NUM 13 + +static void bat_get_strings(struct net_device *dev, u32 stringset, u8 *data) +{ + if (stringset == ETH_SS_STATS) + memcpy(data, bat_stats_strings, sizeof(bat_stats_strings)); +} + +static void bat_get_ethtool_stats(struct net_device *dev, + struct ethtool_stats *stats, u64 *data) +{ + struct bat_priv *bat_priv = netdev_priv(dev); + + data[0] = bat_priv->bat_stats.forward; + data[1] = bat_priv->bat_stats.mgmt_tx; + data[2] = bat_priv->bat_stats.mgmt_rx; + data[3] = bat_priv->bat_stats.tt_request_tx; + data[4] = bat_priv->bat_stats.tt_request_rx; + data[5] = bat_priv->bat_stats.tt_response_tx; + data[6] = bat_priv->bat_stats.tt_response_rx; + data[7] = bat_priv->bat_stats.tt_roam_adv_tx; + data[8] = bat_priv->bat_stats.tt_roam_adv_rx; + data[9] = bat_priv->bat_stats.dat_request_tx; + data[10] = bat_priv->bat_stats.dat_request_rx; + data[11] = bat_priv->bat_stats.dat_reply_tx; + data[12] = bat_priv->bat_stats.dat_reply_rx; +} + +static int bat_get_sset_count(struct net_device *dev, int stringset) +{ + if (stringset == ETH_SS_STATS) + return BAT_STATS_NUM; + + return -EOPNOTSUPP; +} diff --git a/translation-table.c b/translation-table.c index b3e608a..aa0afb1 100644 --- a/translation-table.c +++ b/translation-table.c @@ -1352,6 +1352,8 @@ static int send_tt_request(struct bat_priv *bat_priv, dst_orig_node->orig, neigh_node->addr, (full_table ? 'F' : '.')); + bat_priv->bat_stats.tt_request_tx++; + send_skb_packet(skb, neigh_node->if_incoming, neigh_node->addr); ret = 0; @@ -1476,6 +1478,8 @@ static bool send_other_tt_response(struct bat_priv *bat_priv, res_dst_orig_node->orig, neigh_node->addr, req_dst_orig_node->orig, req_ttvn); + bat_priv->bat_stats.tt_response_tx++; + send_skb_packet(skb, neigh_node->if_incoming, neigh_node->addr); ret = true; goto out; @@ -1592,6 +1596,8 @@ static bool send_my_tt_response(struct bat_priv *bat_priv, orig_node->orig, neigh_node->addr, (tt_response->flags & TT_FULL_TABLE ? 'F' : '.')); + bat_priv->bat_stats.tt_response_tx++; + send_skb_packet(skb, neigh_node->if_incoming, neigh_node->addr); ret = true; goto out; @@ -1891,6 +1897,8 @@ static void send_roam_adv(struct bat_priv *bat_priv, uint8_t *client, "Sending ROAMING_ADV to %pM (client %pM) via %pM\n", orig_node->orig, client, neigh_node->addr); + bat_priv->bat_stats.tt_roam_adv_tx++; + send_skb_packet(skb, neigh_node->if_incoming, neigh_node->addr); ret = 0; diff --git a/types.h b/types.h index 15f538a..c4c1e13 100644 --- a/types.h +++ b/types.h @@ -162,9 +162,26 @@ struct bcast_duplist_entry { }; #endif +struct bat_stats { + uint64_t forward; + uint64_t mgmt_tx; + uint64_t mgmt_rx; + uint64_t tt_request_tx; + uint64_t tt_request_rx; + uint64_t tt_response_tx; + uint64_t tt_response_rx; + uint64_t tt_roam_adv_tx; + uint64_t tt_roam_adv_rx; + uint64_t dat_request_tx; + uint64_t dat_request_rx; + uint64_t dat_reply_tx; + uint64_t dat_reply_rx; +}; + struct bat_priv { atomic_t mesh_state; struct net_device_stats stats; + struct bat_stats bat_stats; atomic_t aggregated_ogms; /* boolean */ atomic_t bonding; /* boolean */ atomic_t fragmentation; /* boolean */ -- 1.7.10

3 11

[B.A.T.M.A.N.] [PATCH] batctl: fix man page originator entry
by Marek Lindner 22 Apr '12

22 Apr '12

Reported-by: Jo-Philipp Wich <jow(a)openwrt.org> Signed-off-by: Marek Lindner <lindner_marek(a)yahoo.de> --- man/batctl.8 | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-) diff --git a/man/batctl.8 b/man/batctl.8 index 036412b..4f2e1c6 100644 --- a/man/batctl.8 +++ b/man/batctl.8 @@ -59,7 +59,7 @@ protocol. If no parameter is given or the first parameter is neither "add" nor "del" the current interface settings are displayed. In order to add or delete interfaces specify "add" or "del" as first argument and append the interface names you wish to add or delete. .br .IP "\fBoriginators\fP|\fBo\fP [\fB\-w\fP [\fI\interval\fP]][\fB\-n\fP][\fB\-t\fP]" -Once started batctl will display the list of announced gateways in the network. Use the "\-w" option to let batctl refresh the list every second or add a number to let it refresh at a custom interval in seconds (with optional decimal places). If "\-n" is given batctl will not replace the MAC addresses with bat\-host names in the output. The "\-t" option filters all originators that have not been seen for the specified amount of seconds (with optional decimal places) from the output. +Once started batctl will display the list of announced originators in the network. Use the "\-w" option to let batctl refresh the list every second or add a number to let it refresh at a custom interval in seconds (with optional decimal places). If "\-n" is given batctl will not replace the MAC addresses with bat\-host names in the output. The "\-t" option filters all originators that have not been seen for the specified amount of seconds (with optional decimal places) from the output. .br .IP "\fBinterval\fP|\fBit\fP [\fBorig_interval\fP]" If no parameter is given the current originator interval setting is displayed otherwise the parameter is used to set the originator interval. The interval is in units of milliseconds. -- 1.7.9.1

1 1

[B.A.T.M.A.N.] [PATCH] batman-adv: fix types in README.external
by Marek Lindner 22 Apr '12

22 Apr '12

Signed-off-by: Marek Lindner <lindner_marek(a)yahoo.de> --- README.external | 4 ++-- 1 files changed, 2 insertions(+), 2 deletions(-) diff --git a/README.external b/README.external index 874ee7f..787172e 100644 --- a/README.external +++ b/README.external @@ -21,7 +21,7 @@ if you want to compile against some other kernel, use: # make KERNELPATH=/path/to/kernel -if you wont to install this module: +if you want to install this module: # sudo make install @@ -29,7 +29,7 @@ CONFIGURATION ------------- The in-kernel module can be configured through -menuconfig. When compiling outside of the kernel, tree it is +menuconfig. When compiling outside of the kernel tree, it is necessary to configure it using the make options. Each option can be set to to y (enabled), n (disabled) or m (build as module). Available options and their possible values are -- 1.7.9.1

1 2

[B.A.T.M.A.N.] Batman configuration and MTU
by HeXiLeD 20 Apr '12

20 Apr '12

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 I have being doing some testing with several types of configurations with custom openwrt build and batman in oder to try to come with a basic default "plug & play" ready to deploy firmware evolving the least changes possible other than the router's ip and not much more. I have tried several configurations, more than one subnet and have being trying to see a way to make some nice HOW-Tos regarding this matter for several types of configurations. So far and from what i seen all working and half working examples do not work for everyone or at least no one or very few post complete configurations of all files in /etc/config/ that actually fully work. Then there is the question if bridging. Some say bridge this, others say do not bridge that. None supply a fully working example that can be replicated. (i am glad i access the routers over serial cable which i recommend to everyone if they are working on builds and configurations. I have now collected quite a few pdfs regarding all openwrt, batman and mesh netwroks and only a couple worth to use for configurations as they seem to work. The best one i found so far can be obtained here: http://wiki.villagetelco.org/images/0/07/SECN_UserGuideV1d4_compressed.pdf Mesh Potato: Small Enterprise Campus Network (user guide) SECN_UserGuideV1d4_compressed.pdf It shows basic fast working examples either using Luci and command line.It's not a perfect guide as it needs a few retouches but by far the best i have found until now. This guide leads me to quite a few questions as for example i was told not to bridge the lan: https://forum.openwrt.org/viewtopic.php?id=35517 This led me to more complex configurations that included 2 different subnets and more but still not 100% working. And following the guide i only got it working doing a bridge. It still find this bridge strange and something confuses me but seems to work. However there is still a problem that remains which is the MTU. I have tried several ways to enable and they all seem to fail. Bellow are my current configurations that match the described situation: - ----- # cat firewall config 'defaults' option 'syn_flood' '1' option 'input' 'ACCEPT' option 'output' 'ACCEPT' option 'forward' 'REJECT' option 'disable_ipv6' '1' config 'zone' option 'name' 'lan' option 'network' 'lan' option 'input' 'ACCEPT' option 'output' 'ACCEPT' option 'forward' 'REJECT' config 'zone' option 'name' 'wifi' option 'network' 'wifi' option 'input' 'ACCEPT' option 'output' 'ACCEPT' option 'forward' 'REJECT' config 'zone' option 'name' 'wan' option 'network' 'wan' option 'input' 'REJECT' option 'output' 'ACCEPT' option 'forward' 'REJECT' option 'masq' '1' config 'forwarding' option 'src' 'lan' option 'dest' 'wan' config 'rule' option 'name' 'Allow-DHCP-Renew' option 'src' 'wan' option 'proto' 'udp' option 'dest_port' '68' option 'target' 'ACCEPT' option 'family' 'ipv4' config 'rule' option 'name' 'Allow-Ping' option 'src' 'wan' option 'proto' 'icmp' option 'icmp_type' 'echo-request' option 'family' 'ipv4' option 'target' 'ACCEPT' config 'include' option 'path' '/etc/firewall.user' - ----- # cat dhcp config 'dnsmasq' option 'domainneeded' '1' option 'boguspriv' '1' option 'filterwin2k' '0' option 'localise_queries' '1' option 'rebind_protection' '1' option 'rebind_localhost' '1' option 'local' '/lan/' option 'domain' 'wireless.domain.com' option 'expandhosts' '1' option 'nonegcache' '0' option 'authoritative' '1' option 'readethers' '1' option 'leasefile' '/tmp/dhcp.leases' option 'resolvfile' '/tmp/resolv.conf.auto' config 'dhcp' 'lan' option 'interface' 'lan' option 'start' '51' option 'limit' '254' option 'leasetime' '24h' config 'dhcp' 'wan' option 'interface' 'wan' option 'ignore' '1' option 'domain' 'wireless.domain.com' - ----- g# cat network config 'switch' 'eth0' option 'enable' '1' config 'switch_vlan' 'eth0_0' option 'device' 'eth0' option 'vlan' '0' option 'ports' '1 2 3 4 5' config 'switch_vlan' 'eth0_1' option 'device' 'eth0' option 'vlan' '1' option 'ports' '0 5' config 'interface' 'loopback' option 'ifname' 'lo' option 'proto' 'static' option 'ipaddr' '127.0.0.1' option 'netmask' '255.0.0.0' config 'interface' 'lan' option 'type' 'bridge' option 'proto' 'static' option 'ipaddr' '192.168.1.1' option 'netmask' '255.255.255.0' option 'broadcast' '192.168.1.255' option 'ifname' 'eth0.0 bat0 wlan0' option 'mtu' '1528' #config 'interface' 'wifi' # option 'proto' 'none' # option 'ifname' 'wlan0' # option 'mtu' '1528' #config 'interface' 'lan' # option 'proto' 'none' # option 'ifname' 'lan' # option 'mtu' '1528' config 'interface' 'wan' option 'ifname' 'eth0.1' option 'proto' 'dhcp' - ----- # cat wireless config 'wifi-device' 'radio0' option 'type' 'mac80211' option 'channel' '11' option 'phy' 'phy0' option 'hwmode' '11g' option 'country' 'ca' option 'txpower' '20' option 'diversity' '1' option 'disabled' '0' option 'txantenna' '1' option 'rxantenna' '1' config 'wifi-iface' option 'device' 'radio0' option 'hidden' '0' option 'mode' 'adhoc' option 'ssid' 'wireless' option 'bssid' '02:16:b8:6b:0a:1a' option 'encryption' 'none' #option 'network' 'lan' #is optional or must match bridge name option 'mtu' '1528' #mtu seems to be ignored - needs fix - ----- # cat batman-adv config 'mesh' 'bat0' option 'ap_isolation' '0' option 'bonding' '0' option 'aggregation' '1' option 'fragmentation' '1' option 'gw_bandwidth' '3mbit/384kbit' option 'gw_mode' 'client' option 'gw_sel_class' '5' option 'orig_interval' '1000' option 'vis_mode' 'client' option 'interfaces' 'wlan0 eth0.0' # batctl if eth0.0: active wlan0: active # batctl o [B.A.T.M.A.N. adv 2012.0.0, MainIF/MAC: eth0.0/00:12:17:be:f9:46 (bat0)] Originator last-seen (#/255) Nexthop [outgoingIF]: Potential nexthops ... - ----- The mesh potato guide does not show configurations for batman-adv . I also noticed that replacing bat0 interfaces by "lan" seems to be working as the lan interface already bridges everything. Correct me if i am wrong. The last problem i now face (assuming that this configuration is valid) is the MTU settings which seem to be ignored no matter where i add them; including creating the following for the other interfaces: config 'interface' 'wifi' option 'proto' 'none' option 'ifname' 'wlan0' option 'mtu' '1528' I have lost quite a lot of sleep over these problems and it is natural that i might be missing something very simple. batman_adv: bat0: The MTU of interface eth0.0 is too small (1500) to handle the transport of batman-adv packets. Packets going over this interface will be fragmented on layer2 which could impact the performance. Setting the MTU to 1528 would solve the problem. batman_adv: bat0: The MTU of interface wlan0 is too small (1500) to handle the transport of batman-adv packets. Packets going over this interface will be fragmented on layer2 which could impact the performance. Setting the MTU to 1528 would solve the problem. Right now i am just trying the setup with 1 router and testing with a netbook. I am able to connect, obtain ip from the router in either wired or wireless. I am also able to connect to the wan but while doing so wirelessly the performance is lower and slower which i believe to be caused by the lack of proper MTU adjustment. Extra details: wrt54g v2.2 backfire 10.03.2 b43 drivers version 508.1084 batman 2012 v14 + batman-adv_load-module-when-needed.patch Loaded modules: nf_nat_tftp 432 0 nf_conntrack_tftp 2400 1 nf_nat_tftp nf_nat_irc 864 0 nf_conntrack_irc 2592 1 nf_nat_irc nf_nat_ftp 1360 0 nf_conntrack_ftp 4896 1 nf_nat_ftp ipt_MASQUERADE 992 1 iptable_nat 2304 1 nf_nat 11824 5 nf_nat_tftp,nf_nat_irc,nf_nat_ftp,ipt_MASQUERADE,iptable_nat xt_NOTRACK 672 0 iptable_raw 656 1 xt_state 800 3 nf_conntrack_ipv4 7808 6 iptable_nat,nf_nat nf_defrag_ipv4 608 1 nf_conntrack_ipv4 nf_conntrack 42560 12 nf_nat_tftp,nf_conntrack_tftp,nf_nat_irc,nf_conntrack_irc,nf_nat_ftp,nf_conntrack_ftp,ipt_MASQUERADE,iptable_nat,nf_nat,xt_NOTRACK,xt_state,nf_conntrack_ipv4 ipt_REJECT 1984 2 xt_TCPMSS 1952 0 ipt_LOG 4640 0 xt_comment 464 0 xt_multiport 1904 0 xt_mac 576 0 xt_limit 1152 1 iptable_mangle 992 0 iptable_filter 768 1 ip_tables 8864 4 iptable_nat,iptable_raw,iptable_mangle,iptable_filter xt_tcpudp 1856 4 x_tables 9840 13 ipt_MASQUERADE,iptable_nat,xt_NOTRACK,xt_state,ipt_REJECT,xt_TCPMSS,ipt_LOG,xt_comment,xt_multiport,xt_mac,xt_limit,ip_tables,xt_tcpudp b43 151056 0 mac80211 260560 1 b43 cfg80211 149136 2 b43,mac80211 compat 13312 3 b43,mac80211,cfg80211 arc4 816 2 aes_generic 31056 0 deflate 1328 0 ecb 1328 0 cbc 2016 0 batman_adv 105408 0 crc16 976 1 batman_adv diag 7504 0 switch_robo 4208 0 swicth_core 5216 1 switch_robo -----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.17 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/ iQIcBAEBAgAGBQJPjdVoAAoJEA1GFAMAMJs71XgP/jbQ4O2+OuddmP7zfKwvTJXz db2YoxmiXx3it0F2R/fV5689B5l0yrxPFtebjN27IDeq2zwJaSvkBuWg6LCFDO2/ W4YKX7P7AzqXIDlkuGO5qH72YKc6Bfw9u74dTusGY4OmmcoLdboN8ID3Qax2wSkY npFukKtqZJFlO4ehH/hneuXxEDmAlpAcDa6dD3+Zkttm61S9ktwslZdtC/tEqoYO g/dqW2r8GXekcd2E/AajG1J4BLSdU1+xQJGVjyWufhO9IAFTNdkkUHAKM6fo96Ea 7SBFvVdLMo6D64gImAUNi5IREgrTihOY41iC1/JnAXTGAfSrxRi8YBbu1aRkUIc3 oTQ8OOTrIE0MmqE0+z0y88Ie23QtADtwHgLUFBHaliqXnkyHB2HxykeyU1WJHSrK 3qsI04Ag4x9GEoMjvThdIhdyIiRFhkSmNg9dfDjaPfhXKOdlklcOOMHyGm1VYO8t acq3vF9TO2d1dsbLKi78QO9jP8pmaWCTLoHf0HE896S8ju1HfMuhSsacjQ+y1W/v XQyP1f3DT/XAIqIhdQ36ZHBSUv081UqdskhtUhPp3V3HEzP7YFYH9wV08P3vQgAM OLZhLrn8JEm1r1urWFzYhuZStdpgAxgNdr7WnC1aYkrx1rvT7WAmZKLx204Pqkjz tdg/kdoEBo6fo8wM3Kvd =TnUr -----END PGP SIGNATURE-----

3 3

[B.A.T.M.A.N.] pull request: batman-adv 2012-04-18
by Antonio Quartulli 19 Apr '12

19 Apr '12

Hello David, this is the fixed version of my previous pull request (issued on 2012-04-17). In this patchset the issues you reported have been fixed; moreover I'm also including two new patches (1/13 and 2/13) which are respectively: 1/13) a fix for Al Viro's report about the missing htons() 2/13) a fix for a wrongly duplicated line in a comment 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 1e5cc266dbc401d11aefb966ad35e651c2f67414: batman-adv: skip the window protection test when the originator has no neighbours (2012-04-18 09:54:02 +0200) ---------------------------------------------------------------- Included changes: * remove duplicated line in comment * add htons() invocation for tt_crc as suggested by Al Viro * OriGinator Message seqno initial value is now random * some cleanups and fixes ---------------------------------------------------------------- Antonio Quartulli (5): batman-adv: convert the tt_crc to network order batman-adv: remove duplicated line in comment batman-adv: use ETH_HLEN instead of sizeof(struct ethhdr) batman-adv: print OGM seq numbers as unsigned int 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 | 11 ++---- 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/translation-table.c | 2 +- net/batman-adv/types.h | 12 +++--- net/batman-adv/vis.c | 8 ++-- 12 files changed, 96 insertions(+), 74 deletions(-)

3 22

[B.A.T.M.A.N.] Determining on which node a client is connected to
by Mitar 18 Apr '12

18 Apr '12

Hi! Is it possible to determine from central location (for example gateway) to which node a client is currently connected to? So from DHCP you can get MAC-IP translation, but how to get from MAC on which node client currently is? Mitar

3 3

[B.A.T.M.A.N.] Problem with multi-radio, multi-channel in batman-adv
by puyou.lu 18 Apr '12

18 Apr '12

Hi guys. 　　I’m trying to make a for nodes WMN using four devices with two wireless cards each. I would like to use three different channels to reduce the interference, like this: 　　[Node A] <- channel 1 -> [Node B] <- channel 2 -> [Node C] <- channel 3 -> [Node D] 　　First I configure eight wireless cards to the specific channel in ad-hoc mode using “iw dev wlan0 ibss join …”. 　　Then, I think I should add wlan0 and wlan1 to the same batman-adv interface (that is bat0) in every device. So I need to configure four nodes with the same commands like: 　　 echo bat0 > /sys/class/net/wlan0/batman_adv/mesh_iface 　　 echo bat0 > /sys/class/net/wlan1/batman_adv/mesh_iface 　　Last configure an IP address to each bat0. (Node A-10.0.0.1, Node B-10.0.0.2, Node C-10.0.0.3, Node D-10.0.0.4). 　　But it seem Node A can reach Node B and Node C(using ping), but can’t for Node D. 　　　　Am I following the right procedure? 　　Sorry for my poor English. Any help would be appreciated. Puyou Lu.

3 4

[B.A.T.M.A.N.] [PATCH] batman-adv: use real error codes instead of meaningless -1
by Antonio Quartulli 18 Apr '12

18 Apr '12

Reported-by: David Miller <davem(a)davemloft.net> Signed-off-by: Antonio Quartulli <ordex(a)autistici.org> --- This patch is for *next*. This patch fixes a problem pointed out by David S. Miller. Actually the idea is to use real error codes instead of meaningless -1, even if this codes are only used internally to the batman-adv code. IMHO it is a good idea. Most of the times we convert -1 to real error codes later, but I think it is better to user real code directly. This patch will be squashed with a patch which is going to be sent in the next pull request. Another patch which fix the other error codes should follow. Cheers, bat_iv_ogm.c | 2 +- hard-interface.c | 4 +--- 2 files changed, 2 insertions(+), 4 deletions(-) diff --git a/bat_iv_ogm.c b/bat_iv_ogm.c index ab2085c..2a1845e 100644 --- a/bat_iv_ogm.c +++ b/bat_iv_ogm.c @@ -59,7 +59,7 @@ static int bat_iv_ogm_iface_enable(struct hard_iface *hard_iface) { struct batman_ogm_packet *batman_ogm_packet; uint32_t random_seqno; - int res = -1; + int res = -ENOMEM; /* randomize initial seqno to avoid collision */ get_random_bytes(&random_seqno, sizeof(random_seqno)); diff --git a/hard-interface.c b/hard-interface.c index e647a43..843caa7 100644 --- a/hard-interface.c +++ b/hard-interface.c @@ -309,10 +309,8 @@ int hardif_enable_interface(struct hard_iface *hard_iface, bat_priv = netdev_priv(hard_iface->soft_iface); ret = bat_priv->bat_algo_ops->bat_iface_enable(hard_iface); - if (ret < 0) { - ret = -ENOMEM; + if (ret < 0) goto err_dev; - } hard_iface->if_num = bat_priv->num_ifaces; bat_priv->num_ifaces++; -- 1.7.9.4

2 1

[B.A.T.M.A.N.] [PATCH] batman-adv: use uint32_t instead of unsigned long to avoid cast
by Antonio Quartulli 18 Apr '12

18 Apr '12

Reported-by: David Miller <davem(a)davemloft.net> Signed-off-by: Antonio Quartulli <ordex(a)autistici.org> --- This patch is for *next*. It solves a problem pointed out by David S. Miller. This patch will be squashed with the one which introduced the issue and will be sent out in the next pull request. Cheers, bat_iv_ogm.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/bat_iv_ogm.c b/bat_iv_ogm.c index 994369d..ab2085c 100644 --- a/bat_iv_ogm.c +++ b/bat_iv_ogm.c @@ -58,12 +58,12 @@ out: static int bat_iv_ogm_iface_enable(struct hard_iface *hard_iface) { struct batman_ogm_packet *batman_ogm_packet; - unsigned long random_seqno; + uint32_t random_seqno; int res = -1; /* randomize initial seqno to avoid collision */ - get_random_bytes(&random_seqno, sizeof(unsigned long)); - atomic_set(&hard_iface->seqno, (uint32_t)random_seqno); + get_random_bytes(&random_seqno, sizeof(random_seqno)); + atomic_set(&hard_iface->seqno, random_seqno); hard_iface->packet_len = BATMAN_OGM_HLEN; hard_iface->packet_buff = kmalloc(hard_iface->packet_len, GFP_ATOMIC); -- 1.7.9.4

2 1

[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

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

B.A.T.M.A.N April 2012