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.] Batman-adv simple configuration example
by Pedro Nuno Costa Rodrigues 07 May '12

07 May '12

Hello! Sorry about taking your time, but i really need some help... First of all, sorry about my english, i will try to express the better i can. I send this need of help for you and for the others developers envolved in the batman-adv project, hoping some reply. My name is Pedro, and i'm writing from Portugal, Oporto. I'm a student of telecomunications and i'm doing the last work to finish my master thesis. It was proposed to me in my school doing a management aplication for a batman-adv mesh network... by the way my school is ISEP - Oporto Superior Institute of Engeneering. In this moment i'm trying to setup a basic 2 nodes using batman-adv... I'm using UBUNTU 10.10 - linux. I'm using OPENWRT in 2 routers. The routers are 2 linksys wrt54gs (v1.1 and v3). I compiled the backfire stable release from: svn co svn://svn.openwrt.org/openwrt/branches/backfire Here are my feeds: src-svn packages svn://svn.openwrt.org/openwrt/branches/packages_10.03.2 svn://svn.openwrt.org/openwrt/packages src-svn luci http://svn.luci.subsignal.org/luci/branches/luci-0.10/contrib/package I installed the kmod-batman-adv. I attached a image with a scheme from my test configuration. .................. When i do: root@OpenWrt:~# batctl if br-lan: active wlan0: active the two interfaces are active... it seems ok! root@OpenWrt:~# batctl o [B.A.T.M.A.N. adv 2012.0.0, MainIF/MAC: br-lan/00:13:10:7c:da:71 (bat0)] Originator last-seen (#/255) Nexthop [outgoingIF]: Potential nexthops ... 00:06:25:00:63:7b 0.412s (247) 00:06:25:00:63:7b [ wlan0]: 00:06:25:00:63:7b (247) 00:06:25:00:63:79 0.076s (246) 00:06:25:00:63:7b [ wlan0]: 00:06:25:00:63:7b (246) the node 1 see the node 2... it seems ok! Here is my /etc/config files for router 1, that is supossed to work as a batman-adv gateway: root@OpenWrt:/etc/config# vi batman-adv config 'mesh' 'bat0' option 'interfaces' 'wifi lan' option 'aggregated_ogms' '1' option 'ap_isolation' '0' option 'bonding' '0' option 'fragmentation' '1' option 'gw_bandwidth' '5mbit/1024kbit' option 'gw_mode' 'server' #option 'gw_sel_class' option 'log_level' 'all' option 'orig_interval' '1000' option 'vis_mode' 'server' after the boot when i do: root@OpenWrt:~# batctl gw off then i do: root@OpenWrt:~# batctl gw server 5mbit root@OpenWrt:~# batctl gw server (announced bw: 4MBit/1024KBit) should't be necessary do "batctl gw server 5mbit"... it already is determinated in the /etc/config/batman-adv... why this happens? also with the vis mode: root@OpenWrt:~# batctl vm client then i do: root@OpenWrt:~# batctl vm server root@OpenWrt:~# batctl vm server the same thing happens... why? ----- the gw definition and the vm definition is mandatory to router 2 access internet? or doesn't matter? here is is my /etc/config files for router 2, that is supossed to work as a batman-adv gateway client: root@OpenWrt:~# vi /etc/config/batman-adv config 'mesh' 'bat0' option 'interfaces' 'wifi lan' option 'aggregated_ogms' '1' option 'ap_isolation' '0' option 'bonding' '0' option 'fragmentation' '1' #option 'gw_bandwidth' option 'gw_mode' 'client' option 'gw_sel_class' '2' option 'log_level' 'all' option 'orig_interval' '1000' option 'vis_mode' 'client' is the file ok? when i do in node 2: root@OpenWrt:~# batctl if br-lan: active wlan0: active root@OpenWrt:~# batctl o [B.A.T.M.A.N. adv 2012.0.0, MainIF/MAC: br-lan/00:06:25:00:63:79 (bat0)] Originator last-seen (#/255) Nexthop [outgoingIF]: Potential nexthops ... 00:13:10:7c:da:71 0.916s (235) 00:13:10:7c:da:73 [ wlan0]: 00:13:10:7c:da:73 (235) 00:13:10:7c:da:73 0.080s (235) 00:13:10:7c:da:73 [ wlan0]: 00:13:10:7c:da:73 (235) it seems ok... root@OpenWrt:~# batctl gw off doesn't seem ok... then i do: root@OpenWrt:~# batctl gw client 2 root@OpenWrt:~# batctl gw client (selection class: 2) root@OpenWrt:~# batctl vm client root@OpenWrt:~# batctl vm client the gw and vm doesn't start/boot with the start configuration... it's the same problem... why this happens? it's crucial for the batman-adv works correctly? ____________________________________________________________ _____end________________________________________________________ ____________________________________________________________ Here is my network file for router 1 and 2 ( the only difference is "lan option ipaddr" and the "wifi option ipaddr"... Question: is necessary to wifi (wlan0) have static ipaddr or it shouldn't? The mtu size is ok? root@OpenWrt:/etc/config# vi network #### VLAN configuration 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" ######## for V3 is "4 5" #### Loopback configuration config interface loopback option ifname "lo" option proto static option ipaddr 127.0.0.1 option netmask 255.0.0.0 #### LAN configuration config interface lan option type bridge option ifname "eth0.0 bat0 wlan0" option proto static option ipaddr 10.3.1.1 #######for router 2 is 10.3.1.2 option netmask 255.255.255.0 option broadcast 10.3.1.255 option mtu 1500 #### WLAN configuration config interface wifi option proto static option ifname "wlan0" option ipaddr 10.1.1.1 ########for router 2 is 10.1.1.2 option netmask 255.255.255.0 option mtu 1528 #### WAN configuration config interface wan option ifname "eth0.1" option proto dhcp #### BAT configuration config interface bat option ifname "bat0" option proto none option mtu 1500 ___________________________________________________ __end_________________________________________________ _____________________________________________________ Here is the dhcp file: P.S. i can access the internet through the eth0.0 (ethernet cable) both in router 1 and router 2, but i can't access through wireless (neither router 1 and router 2)... Both routers give IP through cable and trough wireless... the conections are always made, and it's possible to access using ssh both routers, doesn't matter if the connection is made by cable or by air... the difference is to access to internet... only works by cable in both routers with the scheme i attached... Is possible to exist conflict with the IP distribution, because the "option authoritative 1"? Is necessary to change some settings? It is necessary wifi dhcp to the wireless stations to connect? or should they manually set an static IP? root@OpenWrt:/etc/config# vi dhcp config dnsmasq option domainneeded 1 option boguspriv 1 option filterwin2k 0 # enable for dial on demand option localise_queries 1 option rebind_protection 1 # disable if upstream must serve RFC1918 addresses option rebind_localhost 1 # enable for RBL checking and similar services #list rebind_domain example.lan # whitelist RFC1918 responses for domains option local '/lan/' option domain 'lan' option expandhosts 1 option nonegcache 0 option authoritative 1 option readethers 1 option leasefile '/tmp/dhcp.leases' option resolvfile '/tmp/resolv.conf.auto' #list server '/mycompany.local/1.2.3.4' #option nonwildcard 1 #list interface br-lan #list notinterface lo #list bogusnxdomain '64.94.110.11' config dhcp lan option interface lan option start 51 #########for router 2 is start 101 option limit 70 #########for router 2 is limit 130 #option leasetime 12h config dhcp wan option interface wan option ignore 1 config dhcp wifi option interface wifi option start 51 #########for router 2 is start 101 option limit 70 #########for router 2 is limit 130 _______________________________________________ ______end_______________________________________________ __________________________________________________ Here is my /etc/config/firewall: what do you think? it's ok? root@OpenWrt:~# vi /etc/config/firewall config defaults option syn_flood 1 option input ACCEPT option output ACCEPT option forward ACCEPT # Uncomment this line to disable ipv6 rules # option disable_ipv6 1 config zone option name lan option network 'lan' option input ACCEPT option output ACCEPT option forward ACCEPT config zone option name wan option network 'wan' option input ACCEPT option output ACCEPT option forward ACCPET option masq 1 option mtu_fix 1 config forwarding option src lan option dest wan config forwarding option src wan option dest lan config forwarding option src wifi option dest wan config forwarding option src wan option dest wifi config forwarding option src lan option dest wifi config forwarding option src wifi option dest lan # We need to accept udp packets on port 68, # see https://dev.openwrt.org/ticket/4108 config rule option name Allow-DHCP-Renew option src wan option proto udp option dest_port 68 option target ACCEPT option family ipv4 # Allow IPv4 ping config rule option name Allow-Ping option src wan option proto icmp option icmp_type echo-request option family ipv4 option target ACCEPT # Allow DHCPv6 replies # see https://dev.openwrt.org/ticket/10381 config rule option name Allow-DHCPv6 option src wan option proto udp option src_ip fe80::/10 option src_port 547 option dest_ip fe80::/10 option dest_port 546 option family ipv6 option target ACCEPT # Allow essential incoming IPv6 ICMP traffic config rule option name Allow-ICMPv6-Input option src wan option proto icmp list icmp_type echo-request list icmp_type destination-unreachable list icmp_type packet-too-big list icmp_type time-exceeded list icmp_type bad-header list icmp_type unknown-header-type list icmp_type router-solicitation list icmp_type neighbour-solicitation option limit 1000/sec option family ipv6 option target ACCEPT # Allow essential forwarded IPv6 ICMP traffic config rule option name Allow-ICMPv6-Forward option src wan option dest * option proto icmp list icmp_type echo-request list icmp_type destination-unreachable list icmp_type packet-too-big list icmp_type time-exceeded list icmp_type bad-header list icmp_type unknown-header-type option limit 1000/sec option family ipv6 option target ACCEPT # include a file with users custom iptables rules config include option path /etc/firewall.user _____________________________________ _____end_____________________________________ __________________________________________ Finally here is my /etc/config/wireless file: I also have some questions here... Is necessary the both routers have the same BSSID? Is mandatory to have BSSID, or isn't mandatory to aggregate a batman-adv mesh network? "option network wifi" is correct or should be "option network lan"? root@OpenWrt:/etc/config# vi wireless config wifi-device radio0 option type mac80211 option channel 6 option phy phy0 option hwmode 11g option country PT option txpower 20 option disabled 0 config wifi-iface option device radio0 option network wifi option mode adhoc option ssid PR option bssid 00:13:10:7c:da:71 option encryption none ________________________________ ______end___________________________ ___________________________________ This is it... Perhaps you don't have the time to analyze all this... I'm about two months to get this thing done... but i am not capable... I already asked to my colleagues and to my teachers, but they don't know how to... It's my job... So i decided to ask for your help, otherwise i think i don't get there only by myself... i can't find nothing concrete about these configurations anywhere... Could you please give some light? I would be very appreciate, and of course i would refer your kindness and contribution in the greetings section of my thesis... Thank you very much. Best regards and good work. Pedro Rodrigues Portugal, Europe

3 2

[B.A.T.M.A.N.] [PATCH 1/3] batman-adv: Free local translation table on error in tt_init
by Sven Eckelmann 06 May '12

06 May '12

Signed-off-by: Sven Eckelmann <sven(a)narfation.org> --- translation-table.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/translation-table.c b/translation-table.c index 88c62f1..c3b7773 100644 --- a/translation-table.c +++ b/translation-table.c @@ -1776,8 +1776,10 @@ int tt_init(struct bat_priv *bat_priv) if (!tt_local_init(bat_priv)) return 0; - if (!tt_global_init(bat_priv)) + if (!tt_global_init(bat_priv)) { + tt_local_table_free(bat_priv); return 0; + } tt_start_timer(bat_priv); -- 1.7.10

3 9

[B.A.T.M.A.N.] [PATCH] batman-adv: Change "make install" path and run depmod
by Martin Hundebøll 06 May '12

06 May '12

Currently, when installing the out-of-tree module on distributions that use gzipped modules, batman-adv.ko will be installed next to batman-adv.ko.gz, which is selected by kmod. By changing the install path to /lib/modules/`uname -r`/updates/net/batman-adv/ and running "depmod -ae", we make the newly installed module used by kmod. Signed-off-by: Martin Hundebøll <martin(a)hundeboll.net> --- Makefile | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/Makefile b/Makefile index ac84fba..c68937d 100644 --- a/Makefile +++ b/Makefile @@ -55,7 +55,8 @@ clean: $(MAKE) -C $(KERNELPATH) M=$(PWD) PWD=$(PWD) clean install: config - $(MAKE) -C $(KERNELPATH) M=$(PWD) PWD=$(PWD) INSTALL_MOD_DIR=kernel/net/batman-adv/ modules_install + $(MAKE) -C $(KERNELPATH) M=$(PWD) PWD=$(PWD) INSTALL_MOD_DIR=updates/net/batman-adv/ modules_install + depmod -ae config: $(PWD)/gen-compat-autoconf.sh $(PWD)/compat-autoconf.h -- 1.7.10

4 7

[B.A.T.M.A.N.] [RFC 1/2] batman-adv: Free local translation table on error in tt_init
by Sven Eckelmann 04 May '12

04 May '12

3 6

[B.A.T.M.A.N.] pull request: batman-adv 2012-04-25
by Antonio Quartulli 03 May '12

03 May '12

I'd like to propose the following changes for net-next/linux-3.5. The first 8 patches introduce a new feature called D.A.T. (Distributed ARP Table) which is the resulting implementation of Antonio Quartulli's GSOC 2011 Project[0]. This feature aims to improve reliability of ARP resolutions in sparse wireless mesh networks by creating either a local and a network-wide-distributed ARP cache. Major details can be found on its wikipage[1]. It is an optional feature and it is possible to choose whether to compile it or not by using the proper Kconfig option (=n by default). We are evaluating the possibility of introducing a runtime switch to turn the behaviour on an off without recompiling the module (like for the bridge loop avoidance). Patch 1 also introduces a new packet type, but backward compatibility is preserved. Patch 9 fixes a bug in the "gw propagation" feature introduced so far. Patch 11 and 12 are yet other improvements for the new routing-protocol-API. The others are minor fixes and cleanups. Thank you very much, Antonio [0] http://www.google-melange.com/gsoc/project/google/gsoc2011/ordex/4001 [1] http://www.open-mesh.org/wiki/batman-adv/DistributedArpTable The following changes since commit 7a2a66a0ac1cf93d30869c4ecbfc71a2fda19397: Add linux-next specific files for 20120423 (2012-04-23 16:58:43 +1000) 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 cc01b7a1f16291cc48fa6d261aa952d45604f10c: batman-adv: split neigh_new function into generic and batman iv specific parts (2012-04-24 10:11:31 +0200) ---------------------------------------------------------------- Included changes: * a new feature has been introduced: D.A.T. (Distributed ARP Table). It is a mechanism based on DHT theory that creates a distributed (mesh network wide) ARP cache in order to speed up ARP resolutions in sparse wireless mesh networks. * in order to satisfy DAT requirements a new unicast packet type, namely UNICAST_4ADDR, has been introduced. Backward compatibility has been kept: not updated nodes will simply drop the packet and ignore DAT mechanism. * minor fixes and cleanups * minor routing protocol API cleanups ---------------------------------------------------------------- Antonio Quartulli (9): batman-adv: add UNICAST_4ADDR packet type batman-adv: add a new log level for DAT debugging batman-adv: add biggest_unsigned_int(x) macro batman-adv: Distributed ARP Table - create DHT helper functions batman-adv: Distributed ARP Table - add ARP parsing functions batman-adv: Distributed ARP Table - add snooping functions for ARP messages batman-adv: Distributed ARP Table - increase default soft_iface ARP table timeout batman-adv: Distributed ARP Table - add compile option batman-adv: fix wrong dhcp option list browsing Marek Lindner (6): batman-adv: introduce is_single_hop_neigh variable to increase readability batman-adv: introduce packet type handler array for incoming packets batman-adv: register batman ogm receive function during protocol init batman-adv: rename last_valid to last_seen batman-adv: replace HZ calculations with jiffies_to_msecs() batman-adv: split neigh_new function into generic and batman iv specific parts Documentation/networking/batman-adv.txt | 3 +- net/batman-adv/Kconfig | 10 + net/batman-adv/Makefile | 1 + net/batman-adv/bat_debugfs.c | 4 +- net/batman-adv/bat_iv_ogm.c | 95 +++-- net/batman-adv/bat_sysfs.c | 2 +- net/batman-adv/distributed-arp-table.c | 605 +++++++++++++++++++++++++++++++ net/batman-adv/distributed-arp-table.h | 140 +++++++ net/batman-adv/gateway_client.c | 6 +- net/batman-adv/hard-interface.c | 116 +----- net/batman-adv/main.c | 124 ++++++- net/batman-adv/main.h | 24 +- net/batman-adv/originator.c | 52 +-- net/batman-adv/originator.h | 6 +- net/batman-adv/packet.h | 30 +- net/batman-adv/routing.c | 30 +- net/batman-adv/routing.h | 4 +- net/batman-adv/send.c | 6 +- net/batman-adv/soft-interface.c | 17 +- net/batman-adv/types.h | 33 +- net/batman-adv/unicast.c | 102 +++++- net/batman-adv/unicast.h | 21 +- 22 files changed, 1205 insertions(+), 226 deletions(-) create mode 100644 net/batman-adv/distributed-arp-table.c create mode 100644 net/batman-adv/distributed-arp-table.h

8 29

[B.A.T.M.A.N.] scripting batman-adv - best practice
by 3zl Trizonelabs 02 May '12

02 May '12

i did start to script some basic helper functions and would like some opinion Like say "best practice" : #------------get_bat_if ----------- get_bat_if(){ #returns batman_interface_name 0= ok 1=false(bat interface not existing) #local _bif = DONT !! if contruction needs uninit var if _bif=$(basename $(dirname $(find /sys/class/net/*/mesh -type d) )) 2&>/dev/null; then echo "$_bif" return 0 else return 1 fi } #-----------get_bat_wif-------------------- get_bat_wif(){ #return 0 active( wif=up) 1:inactive (wif=down) * echoes $_bat_wif set -- $(batctl if | awk '{print $1,$2}') # $1 "wlan0:" $2 "active" | "inactive" [ -z $(batctl if ] && exit 1 local _bat_wif=$1 _bat0_status=$2 echo ${_bat_wif%":"} [ "active" = "$_bat0_status" ] && return 0 || return 1 } #--------------get_bat_bridge ----------------- get_bat_bridge(){ # $1 bat_ifname returns bridgename attached to 0= ok 1=false(bat interface not existing) local _bat_ifn=${1:-"bat0"} #ls -l /sys/devices/virtual/net/${_bat_ifn}/brport/bridge if _bat_br=$(basename $(ls -l /sys/devices/virtual/net/${_bat_ifn}/brport/bridge | awk '{print $11}')) then echo "$_bat_br" return 0 else return 1 fi } #------------bat_isup----------------------- bat_isup(){ # test ok [ "down" = $(cat "${_bsysfs}operstate") ] && return 1 || return 0 #better using carrier ?? } #----------get_bat_neigh_cnt----------- get_bat_neighb_cnt(){ # return 0 if neighbours - echo no. of neighbours local _bneighb=$(( $(batctl o | awk 'END{print NR}') -2)) echo $_bneighb [ "$_bneighb" -gt 0 ] && return || return 1 } any input would be highly appreciated regards 3zl joker.eu5.org

1 0

[B.A.T.M.A.N.] pull request: batman-adv 2012-05-01
by Antonio Quartulli 01 May '12

01 May '12

Hello David, this is the new version of my previous pull request (issued on 2012-04-29). I'd like to see this changesin net-next/linux-3.5. In this pull request I entirely removed the D.A.T. code, that as we were discussing in the previous thread, needs some rewriting to avoid directly dealing with the neigh table. This patchset only contains fixes and cleanups. Letme know if there is any problem! Thank you, Antonio The following changes since commit 7a2a66a0ac1cf93d30869c4ecbfc71a2fda19397: Add linux-next specific files for 20120423 (2012-04-23 16:58:43 +1000) 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 969e62ad00763fed9ea28700ef7a7d97fdbfddf3: batman-adv: split neigh_new function into generic and batman iv specific parts (2012-05-01 00:37:58 +0200) ---------------------------------------------------------------- Included changes: * minor fixes and cleanups * minor routing protocol API cleanups ---------------------------------------------------------------- Antonio Quartulli (1): batman-adv: fix wrong dhcp option list browsing Marek Lindner (6): batman-adv: introduce is_single_hop_neigh variable to increase readability batman-adv: introduce packet type handler array for incoming packets batman-adv: register batman ogm receive function during protocol init batman-adv: rename last_valid to last_seen batman-adv: replace HZ calculations with jiffies_to_msecs() batman-adv: split neigh_new function into generic and batman iv specific parts net/batman-adv/bat_debugfs.c | 4 +- net/batman-adv/bat_iv_ogm.c | 95 +++++++++++++++++++++++------- net/batman-adv/gateway_client.c | 6 +- net/batman-adv/hard-interface.c | 113 ------------------------------------ net/batman-adv/main.c | 122 ++++++++++++++++++++++++++++++++++++++- net/batman-adv/main.h | 6 ++ net/batman-adv/originator.c | 50 ++++++++-------- net/batman-adv/originator.h | 6 +- net/batman-adv/routing.c | 22 ++++--- net/batman-adv/routing.h | 4 +- net/batman-adv/send.c | 2 +- net/batman-adv/types.h | 11 ++-- 12 files changed, 249 insertions(+), 192 deletions(-)

2 10

[B.A.T.M.A.N.] [RFC] batman-adv: turn tt commit code into routing protocol agnostic API
by Marek Lindner 29 Apr '12

29 Apr '12

Prior to this patch the translation table code made assumptions about how the routing protocol works and where its buffers are stored (to directly modify them). Each protocol now calls the tt code with the relevant pointers, thereby abstracting the code. Signed-off-by: Marek Lindner <lindner_marek(a)yahoo.de> --- bat_iv_ogm.c | 21 +++++++--- send.c | 74 +---------------------------------- translation-table.c | 110 +++++++++++++++++++++++++++++++++++++++++++++------ translation-table.h | 6 +- types.h | 3 +- 5 files changed, 117 insertions(+), 97 deletions(-) diff --git a/bat_iv_ogm.c b/bat_iv_ogm.c index 2b48b82..a833640 100644 --- a/bat_iv_ogm.c +++ b/bat_iv_ogm.c @@ -562,13 +562,12 @@ static void bat_iv_ogm_forward(struct orig_node *orig_node, if_incoming, 0, bat_iv_ogm_fwd_send_time()); } -static void bat_iv_ogm_schedule(struct hard_iface *hard_iface, - int tt_num_changes) +static void bat_iv_ogm_schedule(struct hard_iface *hard_iface) { struct bat_priv *bat_priv = netdev_priv(hard_iface->soft_iface); struct batman_ogm_packet *batman_ogm_packet; struct hard_iface *primary_if; - int vis_server; + int vis_server, tt_num_changes = 0; vis_server = atomic_read(&bat_priv->vis_mode); primary_if = primary_if_get_selected(bat_priv); @@ -579,10 +578,7 @@ static void bat_iv_ogm_schedule(struct hard_iface *hard_iface, batman_ogm_packet->seqno = htonl((uint32_t)atomic_read(&hard_iface->seqno)); - batman_ogm_packet->ttvn = atomic_read(&bat_priv->ttvn); batman_ogm_packet->tt_crc = htons(bat_priv->tt_crc); - if (tt_num_changes >= 0) - batman_ogm_packet->tt_num_changes = tt_num_changes; if (vis_server == VIS_TYPE_SERVER_SYNC) batman_ogm_packet->flags |= VIS_SERVER; @@ -598,6 +594,19 @@ static void bat_iv_ogm_schedule(struct hard_iface *hard_iface, atomic_inc(&hard_iface->seqno); + if (hard_iface == primary_if) + tt_num_changes = tt_append_changes(bat_priv, + &hard_iface->packet_buff, + &hard_iface->packet_len, + BATMAN_OGM_HLEN); + + if (tt_num_changes > 0) + batman_ogm_packet->tt_num_changes = tt_num_changes; + else + batman_ogm_packet->tt_num_changes = 0; + + batman_ogm_packet->ttvn = atomic_read(&bat_priv->ttvn); + slide_own_bcast_window(hard_iface); bat_iv_ogm_queue_add(bat_priv, hard_iface->packet_buff, hard_iface->packet_len, hard_iface, 1, diff --git a/send.c b/send.c index cebc14a..836b70d 100644 --- a/send.c +++ b/send.c @@ -78,62 +78,9 @@ send_skb_err: return NET_XMIT_DROP; } -static void realloc_packet_buffer(struct hard_iface *hard_iface, - int new_len) -{ - unsigned char *new_buff; - - new_buff = kmalloc(new_len, GFP_ATOMIC); - - /* keep old buffer if kmalloc should fail */ - if (new_buff) { - memcpy(new_buff, hard_iface->packet_buff, - BATMAN_OGM_HLEN); - - kfree(hard_iface->packet_buff); - hard_iface->packet_buff = new_buff; - hard_iface->packet_len = new_len; - } -} - -/* when calling this function (hard_iface == primary_if) has to be true */ -static int prepare_packet_buffer(struct bat_priv *bat_priv, - struct hard_iface *hard_iface) -{ - int new_len; - - new_len = BATMAN_OGM_HLEN + - tt_len((uint8_t)atomic_read(&bat_priv->tt_local_changes)); - - /* if we have too many changes for one packet don't send any - * and wait for the tt table request which will be fragmented */ - if (new_len > hard_iface->soft_iface->mtu) - new_len = BATMAN_OGM_HLEN; - - realloc_packet_buffer(hard_iface, new_len); - - bat_priv->tt_crc = tt_local_crc(bat_priv); - - /* reset the sending counter */ - atomic_set(&bat_priv->tt_ogm_append_cnt, TT_OGM_APPEND_MAX); - - return tt_changes_fill_buffer(bat_priv, - hard_iface->packet_buff + BATMAN_OGM_HLEN, - hard_iface->packet_len - BATMAN_OGM_HLEN); -} - -static int reset_packet_buffer(struct bat_priv *bat_priv, - struct hard_iface *hard_iface) -{ - realloc_packet_buffer(hard_iface, BATMAN_OGM_HLEN); - return 0; -} - void schedule_bat_ogm(struct hard_iface *hard_iface) { struct bat_priv *bat_priv = netdev_priv(hard_iface->soft_iface); - struct hard_iface *primary_if; - int tt_num_changes = -1; if ((hard_iface->if_status == IF_NOT_IN_USE) || (hard_iface->if_status == IF_TO_BE_REMOVED)) @@ -149,26 +96,7 @@ void schedule_bat_ogm(struct hard_iface *hard_iface) if (hard_iface->if_status == IF_TO_BE_ACTIVATED) hard_iface->if_status = IF_ACTIVE; - primary_if = primary_if_get_selected(bat_priv); - - if (hard_iface == primary_if) { - /* if at least one change happened */ - if (atomic_read(&bat_priv->tt_local_changes) > 0) { - tt_commit_changes(bat_priv); - tt_num_changes = prepare_packet_buffer(bat_priv, - hard_iface); - } - - /* if the changes have been sent often enough */ - if (!atomic_dec_not_zero(&bat_priv->tt_ogm_append_cnt)) - tt_num_changes = reset_packet_buffer(bat_priv, - hard_iface); - } - - if (primary_if) - hardif_free_ref(primary_if); - - bat_priv->bat_algo_ops->bat_ogm_schedule(hard_iface, tt_num_changes); + bat_priv->bat_algo_ops->bat_ogm_schedule(hard_iface); } static void forw_packet_free(struct forw_packet *forw_packet) diff --git a/translation-table.c b/translation-table.c index 88c62f1..b032087 100644 --- a/translation-table.c +++ b/translation-table.c @@ -275,14 +275,63 @@ out: tt_global_entry_free_ref(tt_global_entry); } -int tt_changes_fill_buffer(struct bat_priv *bat_priv, - unsigned char *buff, int buff_len) +static void tt_realloc_packet_buff(unsigned char **packet_buff, + int *packet_buff_len, int min_packet_len, + int new_packet_len) +{ + unsigned char *new_buff; + + new_buff = kmalloc(new_packet_len, GFP_ATOMIC); + + /* keep old buffer if kmalloc should fail */ + if (new_buff) { + memcpy(new_buff, *packet_buff, min_packet_len); + kfree(*packet_buff); + *packet_buff = new_buff; + *packet_buff_len = new_packet_len; + } +} + +static void tt_prepare_packet_buff(struct bat_priv *bat_priv, + unsigned char **packet_buff, + int *packet_buff_len, int min_packet_len) +{ + struct hard_iface *primary_if; + int req_len; + + primary_if = primary_if_get_selected(bat_priv); + + req_len = min_packet_len; + req_len += tt_len((uint8_t)atomic_read(&bat_priv->tt_local_changes)); + + /* if we have too many changes for one packet don't send any + * and wait for the tt table request which will be fragmented */ + if ((!primary_if) || (req_len > primary_if->soft_iface->mtu)) + req_len = min_packet_len; + + tt_realloc_packet_buff(packet_buff, packet_buff_len, + min_packet_len, req_len); + + if (primary_if) + hardif_free_ref(primary_if); +} + +static int tt_changes_fill_buff(struct bat_priv *bat_priv, + unsigned char **packet_buff, + int *packet_buff_len, int min_packet_len) { - int count = 0, tot_changes = 0; struct tt_change_node *entry, *safe; + int count = 0, tot_changes = 0, new_len; + unsigned char *tt_buff; + + tt_prepare_packet_buff(bat_priv, packet_buff, + packet_buff_len, min_packet_len); - if (buff_len > 0) - tot_changes = buff_len / tt_len(1); + new_len = *packet_buff_len - min_packet_len; + tt_buff = *packet_buff + min_packet_len; + + if (new_len > 0) + tot_changes = new_len / tt_len(1); spin_lock_bh(&bat_priv->tt_changes_list_lock); atomic_set(&bat_priv->tt_local_changes, 0); @@ -290,7 +339,7 @@ int tt_changes_fill_buffer(struct bat_priv *bat_priv, list_for_each_entry_safe(entry, safe, &bat_priv->tt_changes_list, list) { if (count < tot_changes) { - memcpy(buff + tt_len(count), + memcpy(tt_buff + tt_len(count), &entry->change, sizeof(struct tt_change)); count++; } @@ -306,15 +355,15 @@ int tt_changes_fill_buffer(struct bat_priv *bat_priv, bat_priv->tt_buff = NULL; /* We check whether this new OGM has no changes due to size * problems */ - if (buff_len > 0) { + if (new_len > 0) { /** * if kmalloc() fails we will reply with the full table * instead of providing the diff */ - bat_priv->tt_buff = kmalloc(buff_len, GFP_ATOMIC); + bat_priv->tt_buff = kmalloc(new_len, GFP_ATOMIC); if (bat_priv->tt_buff) { - memcpy(bat_priv->tt_buff, buff, buff_len); - bat_priv->tt_buff_len = buff_len; + memcpy(bat_priv->tt_buff, tt_buff, new_len); + bat_priv->tt_buff_len = new_len; } } spin_unlock_bh(&bat_priv->tt_buff_lock); @@ -2019,20 +2068,55 @@ static void tt_local_purge_pending_clients(struct bat_priv *bat_priv) } -void tt_commit_changes(struct bat_priv *bat_priv) +static int tt_commit_changes(struct bat_priv *bat_priv, + unsigned char **packet_buff, int *packet_buff_len, + int packet_min_len) { - uint16_t changed_num = tt_set_flags(bat_priv->tt_local_hash, - TT_CLIENT_NEW, false); + uint16_t changed_num = 0; + + if (atomic_read(&bat_priv->tt_local_changes) < 1) + return 0; + + changed_num = tt_set_flags(bat_priv->tt_local_hash, + TT_CLIENT_NEW, false); + /* all the reset entries have now to be effectively counted as local * entries */ atomic_add(changed_num, &bat_priv->num_local_tt); tt_local_purge_pending_clients(bat_priv); + bat_priv->tt_crc = tt_local_crc(bat_priv); /* Increment the TTVN only once per OGM interval */ atomic_inc(&bat_priv->ttvn); bat_dbg(DBG_TT, bat_priv, "Local changes committed, updating to ttvn %u\n", (uint8_t)atomic_read(&bat_priv->ttvn)); bat_priv->tt_poss_change = false; + + /* reset the sending counter */ + atomic_set(&bat_priv->tt_ogm_append_cnt, TT_OGM_APPEND_MAX); + + return tt_changes_fill_buff(bat_priv, packet_buff, + packet_buff_len, packet_min_len); +} + +/* when calling this function (hard_iface == primary_if) has to be true */ +int tt_append_changes(struct bat_priv *bat_priv, + unsigned char **packet_buff, int *packet_buff_len, + int packet_min_len) +{ + int tt_num_changes; + + /* if at least one change happened */ + tt_num_changes = tt_commit_changes(bat_priv, packet_buff, + packet_buff_len, packet_min_len); + + /* if the changes have been sent often enough */ + if ((tt_num_changes == 0) && + (!atomic_dec_not_zero(&bat_priv->tt_ogm_append_cnt))) + tt_realloc_packet_buff(packet_buff, packet_buff_len, + packet_min_len, packet_min_len); + + return tt_num_changes; } bool is_ap_isolated(struct bat_priv *bat_priv, uint8_t *src, uint8_t *dst) diff --git a/translation-table.h b/translation-table.h index c43374d..58d9aae 100644 --- a/translation-table.h +++ b/translation-table.h @@ -23,8 +23,6 @@ #define _NET_BATMAN_ADV_TRANSLATION_TABLE_H_ int tt_len(int changes_num); -int tt_changes_fill_buffer(struct bat_priv *bat_priv, - unsigned char *buff, int buff_len); int tt_init(struct bat_priv *bat_priv); void tt_local_add(struct net_device *soft_iface, const uint8_t *addr, int ifindex); @@ -48,11 +46,13 @@ bool send_tt_response(struct bat_priv *bat_priv, bool is_my_client(struct bat_priv *bat_priv, const uint8_t *addr); void handle_tt_response(struct bat_priv *bat_priv, struct tt_query_packet *tt_response); -void tt_commit_changes(struct bat_priv *bat_priv); bool is_ap_isolated(struct bat_priv *bat_priv, uint8_t *src, uint8_t *dst); void tt_update_orig(struct bat_priv *bat_priv, struct orig_node *orig_node, const unsigned char *tt_buff, uint8_t tt_num_changes, uint8_t ttvn, uint16_t tt_crc); +int tt_append_changes(struct bat_priv *bat_priv, + unsigned char **packet_buff, int *packet_buff_len, + int packet_min_len); bool tt_global_client_is_roaming(struct bat_priv *bat_priv, uint8_t *addr); diff --git a/types.h b/types.h index 2944f77..9bde13d 100644 --- a/types.h +++ b/types.h @@ -427,8 +427,7 @@ struct bat_algo_ops { /* called when primary interface is selected / changed */ void (*bat_primary_iface_set)(struct hard_iface *hard_iface); /* prepare a new outgoing OGM for the send queue */ - void (*bat_ogm_schedule)(struct hard_iface *hard_iface, - int tt_num_changes); + void (*bat_ogm_schedule)(struct hard_iface *hard_iface); /* send scheduled OGM */ void (*bat_ogm_emit)(struct forw_packet *forw_packet); }; -- 1.7.9.1

2 3

[B.A.T.M.A.N.] [PATCH] batman-adv: remove biggest_unsigned_int() macro
by Antonio Quartulli 29 Apr '12

29 Apr '12

As suggested by David S. Miller, this macro is not very useful. So we can remove it and use its expanded version instead. Signed-off-by: Antonio Quartulli <ordex(a)autistici.org> --- This patch should be committed into next because it has to be sent to David S. Miller within the next pull request. Thank you distributed-arp-table.h | 2 +- main.h | 3 --- 2 files changed, 1 insertion(+), 4 deletions(-) diff --git a/distributed-arp-table.h b/distributed-arp-table.h index 5e23577..6c0acde 100644 --- a/distributed-arp-table.h +++ b/distributed-arp-table.h @@ -29,7 +29,7 @@ #include <linux/if_arp.h> -#define DAT_ADDR_MAX biggest_unsigned_int(dat_addr_t) +#define DAT_ADDR_MAX ((dat_addr_t)~(dat_addr_t)0) #define ARP_HW_SRC(skb, hdr_size) ((uint8_t *)(skb->data + hdr_size) + \ ETH_HLEN + sizeof(struct arphdr)) diff --git a/main.h b/main.h index 97d0258..426a0d5 100644 --- a/main.h +++ b/main.h @@ -241,9 +241,6 @@ static inline bool has_timed_out(unsigned long timestamp, unsigned int timeout) /* Returns the smallest signed integer in two's complement with the sizeof x */ #define smallest_signed_int(x) (1u << (7u + 8u * (sizeof(x) - 1u))) -/* Returns the biggest unsigned integer with the sizeof x */ -#define biggest_unsigned_int(x) (~(x)0) - /* Checks if a sequence number x is a predecessor/successor of y. * they handle overflows/underflows and can correctly check for a * predecessor/successor unless the variable sequence number has grown by -- 1.7.9.4

2 1

[B.A.T.M.A.N.] [PATCH] batman-adv: Define attribute __percpu for < 2.6.33
by Martin Hundebøll 26 Apr '12

26 Apr '12

Signed-off-by: Martin Hundebøll <martin(a)hundeboll.net> --- compat.h | 1 + 1 file changed, 1 insertion(+) diff --git a/compat.h b/compat.h index 2dc41b1..33c52d3 100644 --- a/compat.h +++ b/compat.h @@ -59,6 +59,7 @@ #if LINUX_VERSION_CODE < KERNEL_VERSION(2, 6, 33) #define __always_unused __attribute__((unused)) +#define __percpu #define skb_iif iif -- 1.7.10

2 1

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