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 and 802.xx
by Harms, Hannes 17 Feb '14

17 Feb '14

Dear list, I have a problems to understand whre are the borders of batman-adv. For exampel if I use enddvices that support 802.n, I will achive a higher throughput than with 802.g devices. I tested this in practice. But what happens if I use devices that support 802.ac ? Will the throughput also be higher in my network, and would the devices use beamforming? I assume that all devices in the network are the same. Best regrads and thanks for your time Hannes

2 1

[B.A.T.M.A.N.] [A.L.F.R.E.D] Can not get data from alfred client.
by Le Tran Dat 17 Feb '14

17 Feb '14

Hi list, I have trouble with Alfred server and client. I have a Batman-adv serve as a server and another as client. In the server node (batman-adv server) I start alfred as master. In the client node (batman-adv client) I start alfred as slave. Then I try to set data from server node by command: echo "data from server" | alfred -s 65 I can see the data in server side by command alfred -r 65 But I can not read the data from client by using the same command. How can I troubleshoot this issue since I have successfully set and get data via alfred before. I have no ideas why it is not working now. Thanks, Dat

2 1

[B.A.T.M.A.N.] Bridge loop avoidance required?
by tjhowse 16 Feb '14

16 Feb '14

Behold my shitty diagram: http://i.imgur.com/9tODwRu.png In this configuration, where the only route out of the batman-adv-supported layer 2 network is via a layer 3 router to the WAN port on the nodes, is there any benefit to enabling bridge loop avoidance? My understanding is that forming a loop would only be accidentally possible if the link to the non-mesh LAN doesn't go via a gateway. The gateway is configured to block all WAN->LAN traffic, except SSH to the node itself. The reason I ask is that the operator of the non-mesh network has demanded that we cease connecting any form of mesh routers to their network on the basis that it could mesh a route between two routers connected to their network and form a loop. The situation is quite political, and the facts aren't necessarily the most important factors. Thanks for your time, Travis.

1 0

[B.A.T.M.A.N.] [PATCH] batman-adv: silence smatch errors in multicast compat code
by Simon Wunderlich 16 Feb '14

16 Feb '14

smatch is reporting problems of unreachable code parts: multicast.c:170 batadv_mcast_get_bridge() info: ignoring unreachable code. tmp/multicast.c:172 batadv_mcast_get_bridge() info: ignoring unreachable code. These are caused by compat code which returns within a loop, never giving the rest of the function a chance to be processed. This can be changed by slightly adjusting the compat code, setting the iterator variable to NULL and let the original function handle the rest. This does not introduce any functional change. Introduced by: a3f58139efedb96222dffdda89896c243fdff49e ("batman-adv: Multicast Listener Announcements via Translation Table") Signed-off-by: Simon Wunderlich <sw(a)simonwunderlich.de> --- compat.h | 14 +++++--------- 1 file changed, 5 insertions(+), 9 deletions(-) diff --git a/compat.h b/compat.h index 186547b..a99c9d5 100644 --- a/compat.h +++ b/compat.h @@ -160,14 +160,12 @@ static inline int batadv_param_set_copystring(const char *val, #define NET_ADDR_RANDOM 0 #define netdev_master_upper_dev_get_rcu(dev) \ - upper; \ - rcu_read_unlock(); \ + NULL; \ if (dev->br_port ? 1 : 0) { \ + rcu_read_unlock(); \ dev_hold(dev); \ return dev; \ } \ -\ - return NULL\ #endif /* < KERNEL_VERSION(2, 6, 36) */ @@ -344,14 +342,12 @@ static int __batadv_interface_set_mac_addr(x, y) #ifndef netdev_master_upper_dev_get_rcu #define netdev_master_upper_dev_get_rcu(dev) \ - upper; \ - rcu_read_unlock(); \ + NULL; \ if (dev->priv_flags & IFF_BRIDGE_PORT) { \ + rcu_read_unlock(); \ dev_hold(dev); \ return dev; \ - } \ -\ - return NULL\ + } #endif /* netdev_master_upper_dev_get_rcu */ -- 1.7.10.4

2 1

[B.A.T.M.A.N.] [PATCHv2] batman-adv: return NULL instead of 0 in batadv_mcast_forw_ip_node_get()
by Linus Lüssing 16 Feb '14

16 Feb '14

batadv_mcast_forw_ip_node_get() returns a pointer, so returning NULL in error cases is always more appropriate than 0. Introduced by: c8fedbea81eac41a4971bb43353fd52a903d7c4c ("batman-adv: Send multicast packets to nodes with a WANT_ALL flag") Reported-by: Antonio Quartulli <antonio(a)meshcoding.com> Signed-off-by: Linus Lüssing <linus.luessing(a)web.de> --- * added Reported-by multicast.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/multicast.c b/multicast.c index b2abe00..00bb30e 100644 --- a/multicast.c +++ b/multicast.c @@ -498,7 +498,7 @@ batadv_mcast_forw_ip_node_get(struct batadv_priv *bat_priv, return batadv_mcast_forw_ipv6_node_get(bat_priv); default: /* we shouldn't be here... */ - return 0; + return NULL; } } -- 1.7.10.4

2 1

[B.A.T.M.A.N.] [PATCH] batman-adv: return NULL instead of 0 in batadv_mcast_forw_ip_node_get()
by Linus Lüssing 16 Feb '14

16 Feb '14

batadv_mcast_forw_ip_node_get() returns a pointer, so returning NULL in error cases is always more appropriate than 0. Introduced by: c8fedbea81eac41a4971bb43353fd52a903d7c4c ("batman-adv: Send multicast packets to nodes with a WANT_ALL flag") Signed-off-by: Linus Lüssing <linus.luessing(a)web.de> --- multicast.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/multicast.c b/multicast.c index b2abe00..00bb30e 100644 --- a/multicast.c +++ b/multicast.c @@ -498,7 +498,7 @@ batadv_mcast_forw_ip_node_get(struct batadv_priv *bat_priv, return batadv_mcast_forw_ipv6_node_get(bat_priv); default: /* we shouldn't be here... */ - return 0; + return NULL; } } -- 1.7.10.4

1 0

[B.A.T.M.A.N.] [PATCHv2] batman-adv: add kerneldoc for dst_hint argument
by Antonio Quartulli 16 Feb '14

16 Feb '14

Some helper functions used along the TX path have now a new "dst_hint" argument but the kerneldoc was missing. Signed-off-by: Antonio Quartulli <antonio(a)meshcoding.com> --- Changes from v1: - comment rephrased (thanks Marek!) send.c | 1 + send.h | 2 ++ 2 files changed, 3 insertions(+) diff --git a/send.c b/send.c index 31a8c19..e8b7124 100644 --- a/send.c +++ b/send.c @@ -311,6 +311,7 @@ out: * @packet_type: the batman unicast packet type to use * @packet_subtype: the unicast 4addr packet subtype (only relevant for unicast * 4addr packets) + * @dst_hint: can be used to override the destination contained in the skb * @vid: the vid to be used to search the translation table * * Look up the recipient node for the destination address in the ethernet diff --git a/send.h b/send.h index aaddaa9..31e8781 100644 --- a/send.h +++ b/send.h @@ -47,6 +47,7 @@ int batadv_send_skb_via_gw(struct batadv_priv *bat_priv, struct sk_buff *skb, * batadv_send_skb_via_tt - send an skb via TT lookup * @bat_priv: the bat priv with all the soft interface information * @skb: the payload to send + * @dst_hint: can be used to override the destination contained in the skb * @vid: the vid to be used to search the translation table * * Look up the recipient node for the destination address in the ethernet @@ -68,6 +69,7 @@ static inline int batadv_send_skb_via_tt(struct batadv_priv *bat_priv, * @bat_priv: the bat priv with all the soft interface information * @skb: the payload to send * @packet_subtype: the unicast 4addr packet subtype to use + * @dst_hint: can be used to override the destination contained in the skb * @vid: the vid to be used to search the translation table * * Look up the recipient node for the destination address in the ethernet -- 1.8.5.3

2 1

[B.A.T.M.A.N.] [PATCH net 10/10] batman-adv: fix potential kernel paging error for unicast transmissions
by Antonio Quartulli 15 Feb '14

15 Feb '14

batadv_send_skb_prepare_unicast(_4addr) might reallocate the skb's data. If it does then our ethhdr pointer is not valid anymore in batadv_send_skb_unicast(), resulting in a kernel paging error. Fixing this by refetching the ethhdr pointer after the potential reallocation. Introduced by 3275e7cc84fb0574e9662e8e74c3b1dab38f7143 ("batman-adv: improve unicast packet (re)routing") Reported-by: Linus Lüssing <linus.luessing(a)web.de> Signed-off-by: Antonio Quartulli <antonio(a)meshcoding.com> --- ***This patch is not supposed to be merged in the batman-adv repository*** This is a substitutive patch that I am going to send to net instead of the one that we have in maint (same subject). I am doing this because the original patch together with all the bugfixes that followed it cannot be considered a "simple bugfix" anymore and I'd rather prefer to send it to net-next with a later pull request. This patch instead fixes the same problem but in a much simpler way (does not use of eth_hdr()). I sent it over the ml for peer review. If nobody has any objection I will send it to net. @Linus: I'd like to add your signed-off to this patch because it is essentially your patch but just reworked a bit. Please let me know what you think. Cheers, net/batman-adv/send.c | 9 +++++++-- 1 file changed, 7 insertions(+), 2 deletions(-) diff --git a/net/batman-adv/send.c b/net/batman-adv/send.c index 579f5f0..2f378a1 100644 --- a/net/batman-adv/send.c +++ b/net/batman-adv/send.c @@ -254,9 +254,9 @@ static int batadv_send_skb_unicast(struct batadv_priv *bat_priv, struct batadv_orig_node *orig_node, unsigned short vid) { - struct ethhdr *ethhdr = (struct ethhdr *)skb->data; + struct ethhdr *ethhdr; struct batadv_unicast_packet *unicast_packet; - int ret = NET_XMIT_DROP; + int ret = NET_XMIT_DROP, hdr_size; if (!orig_node) goto out; @@ -265,12 +265,16 @@ static int batadv_send_skb_unicast(struct batadv_priv *bat_priv, case BATADV_UNICAST: if (!batadv_send_skb_prepare_unicast(skb, orig_node)) goto out; + + hdr_size = sizeof(*unicast_packet); break; case BATADV_UNICAST_4ADDR: if (!batadv_send_skb_prepare_unicast_4addr(bat_priv, skb, orig_node, packet_subtype)) goto out; + + hdr_size = sizeof(struct batadv_unicast_packet_4addr); break; default: /* this function supports UNICAST and UNICAST_4ADDR only. It @@ -279,6 +283,7 @@ static int batadv_send_skb_unicast(struct batadv_priv *bat_priv, goto out; } + eth_hdr = (struct eth_hdr *)(skb->data + hdr_size); unicast_packet = (struct batadv_unicast_packet *)skb->data; /* inform the destination node that we are still missing a correct route -- 1.8.5.3

1 0

[B.A.T.M.A.N.] [PATCHv2] batman-adv: avoid double free when orig_node initialization fails
by Antonio Quartulli 15 Feb '14

15 Feb '14

In the failure path of the orig_node initialization routine the orig_node->bat_iv.bcast_own field is free'd twice: first in batadv_iv_ogm_orig_get() and then later in batadv_orig_node_free_rcu(). Fix it by removing the kfree in batadv_iv_ogm_orig_get(). Signed-off-by: Antonio Quartulli <antonio(a)meshcoding.com> --- Change form v1: - remove kfree from batadv_iv_ogm_orig_get() instead of setting orig_node->bat_iv.bcast_own to NULL bat_iv_ogm.c | 6 ++---- 1 file changed, 2 insertions(+), 4 deletions(-) diff --git a/bat_iv_ogm.c b/bat_iv_ogm.c index c07e59f..fdf4322 100644 --- a/bat_iv_ogm.c +++ b/bat_iv_ogm.c @@ -243,18 +243,16 @@ batadv_iv_ogm_orig_get(struct batadv_priv *bat_priv, const uint8_t *addr) size = bat_priv->num_ifaces * sizeof(uint8_t); orig_node->bat_iv.bcast_own_sum = kzalloc(size, GFP_ATOMIC); if (!orig_node->bat_iv.bcast_own_sum) - goto free_bcast_own; + goto free_orig_node; hash_added = batadv_hash_add(bat_priv->orig_hash, batadv_compare_orig, batadv_choose_orig, orig_node, &orig_node->hash_entry); if (hash_added != 0) - goto free_bcast_own; + goto free_orig_node; return orig_node; -free_bcast_own: - kfree(orig_node->bat_iv.bcast_own); free_orig_node: /* free twice, as batadv_orig_node_new sets refcount to 2 */ batadv_orig_node_free_ref(orig_node); -- 1.8.5.3

2 1

[B.A.T.M.A.N.] [PATCH maint 1/2] batman-adv: fix TT CRC computation by ensuring byte order
by Antonio Quartulli 15 Feb '14

15 Feb '14

From: Antonio Quartulli <antonio(a)open-mesh.com> When computing the CRC on a 2byte variable the order of the bytes obviously alters the final result. This means that computing the CRC over the same value on two archs having different endianess leads to different numbers. The global and local translation table CRC computation routine makes this mistake while processing the clients VIDs. The result is a continuous CRC mismatching between nodes having different endianess. Fix this by converting the VID to Network Order before processing it. This guarantees that every node uses the same byte order. Introduced by 21a57f6e7a3b4455dfe68ee07a7b901d9e7f200b ("batman-adv: make the TT CRC logic VLAN specific") Signed-off-by: Antonio Quartulli <antonio(a)open-mesh.com> --- translation-table.c | 16 ++++++++++++---- 1 file changed, 12 insertions(+), 4 deletions(-) diff --git a/translation-table.c b/translation-table.c index 05c2a9b..24e3267 100644 --- a/translation-table.c +++ b/translation-table.c @@ -1961,6 +1961,7 @@ static uint32_t batadv_tt_global_crc(struct batadv_priv *bat_priv, struct hlist_head *head; uint32_t i, crc_tmp, crc = 0; uint8_t flags; + __be16 tmp_vid; for (i = 0; i < hash->size; i++) { head = &hash->table[i]; @@ -1997,8 +1998,11 @@ static uint32_t batadv_tt_global_crc(struct batadv_priv *bat_priv, orig_node)) continue; - crc_tmp = crc32c(0, &tt_common->vid, - sizeof(tt_common->vid)); + /* use network order to read the VID: this ensures that + * every node reads the bytes in the same order. + */ + tmp_vid = htons(tt_common->vid); + crc_tmp = crc32c(0, &tmp_vid, sizeof(tmp_vid)); /* compute the CRC on flags that have to be kept in sync * among nodes @@ -2032,6 +2036,7 @@ static uint32_t batadv_tt_local_crc(struct batadv_priv *bat_priv, struct hlist_head *head; uint32_t i, crc_tmp, crc = 0; uint8_t flags; + __be16 tmp_vid; for (i = 0; i < hash->size; i++) { head = &hash->table[i]; @@ -2050,8 +2055,11 @@ static uint32_t batadv_tt_local_crc(struct batadv_priv *bat_priv, if (tt_common->flags & BATADV_TT_CLIENT_NEW) continue; - crc_tmp = crc32c(0, &tt_common->vid, - sizeof(tt_common->vid)); + /* use network order to read the VID: this ensures that + * every node reads the bytes in the same order. + */ + tmp_vid = htons(tt_common->vid); + crc_tmp = crc32c(0, &tmp_vid, sizeof(tmp_vid)); /* compute the CRC on flags that have to be kept in sync * among nodes -- 1.8.5.3

3 6

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