diff options
author | Linus Torvalds <torvalds@linux-foundation.org> | 2013-11-13 17:40:34 +0900 |
---|---|---|
committer | Linus Torvalds <torvalds@linux-foundation.org> | 2013-11-13 17:40:34 +0900 |
commit | 42a2d923cc349583ebf6fdd52a7d35e1c2f7e6bd (patch) | |
tree | 2b2b0c03b5389c1301800119333967efafd994ca /net/batman-adv/gateway_client.c | |
parent | 5cbb3d216e2041700231bcfc383ee5f8b7fc8b74 (diff) | |
parent | 75ecab1df14d90e86cebef9ec5c76befde46e65f (diff) |
Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-next
Pull networking updates from David Miller:
1) The addition of nftables. No longer will we need protocol aware
firewall filtering modules, it can all live in userspace.
At the core of nftables is a, for lack of a better term, virtual
machine that executes byte codes to inspect packet or metadata
(arriving interface index, etc.) and make verdict decisions.
Besides support for loading packet contents and comparing them, the
interpreter supports lookups in various datastructures as
fundamental operations. For example sets are supports, and
therefore one could create a set of whitelist IP address entries
which have ACCEPT verdicts attached to them, and use the appropriate
byte codes to do such lookups.
Since the interpreted code is composed in userspace, userspace can
do things like optimize things before giving it to the kernel.
Another major improvement is the capability of atomically updating
portions of the ruleset. In the existing netfilter implementation,
one has to update the entire rule set in order to make a change and
this is very expensive.
Userspace tools exist to create nftables rules using existing
netfilter rule sets, but both kernel implementations will need to
co-exist for quite some time as we transition from the old to the
new stuff.
Kudos to Patrick McHardy, Pablo Neira Ayuso, and others who have
worked so hard on this.
2) Daniel Borkmann and Hannes Frederic Sowa made several improvements
to our pseudo-random number generator, mostly used for things like
UDP port randomization and netfitler, amongst other things.
In particular the taus88 generater is updated to taus113, and test
cases are added.
3) Support 64-bit rates in HTB and TBF schedulers, from Eric Dumazet
and Yang Yingliang.
4) Add support for new 577xx tigon3 chips to tg3 driver, from Nithin
Sujir.
5) Fix two fatal flaws in TCP dynamic right sizing, from Eric Dumazet,
Neal Cardwell, and Yuchung Cheng.
6) Allow IP_TOS and IP_TTL to be specified in sendmsg() ancillary
control message data, much like other socket option attributes.
From Francesco Fusco.
7) Allow applications to specify a cap on the rate computed
automatically by the kernel for pacing flows, via a new
SO_MAX_PACING_RATE socket option. From Eric Dumazet.
8) Make the initial autotuned send buffer sizing in TCP more closely
reflect actual needs, from Eric Dumazet.
9) Currently early socket demux only happens for TCP sockets, but we
can do it for connected UDP sockets too. Implementation from Shawn
Bohrer.
10) Refactor inet socket demux with the goal of improving hash demux
performance for listening sockets. With the main goals being able
to use RCU lookups on even request sockets, and eliminating the
listening lock contention. From Eric Dumazet.
11) The bonding layer has many demuxes in it's fast path, and an RCU
conversion was started back in 3.11, several changes here extend the
RCU usage to even more locations. From Ding Tianhong and Wang
Yufen, based upon suggestions by Nikolay Aleksandrov and Veaceslav
Falico.
12) Allow stackability of segmentation offloads to, in particular, allow
segmentation offloading over tunnels. From Eric Dumazet.
13) Significantly improve the handling of secret keys we input into the
various hash functions in the inet hashtables, TCP fast open, as
well as syncookies. From Hannes Frederic Sowa. The key fundamental
operation is "net_get_random_once()" which uses static keys.
Hannes even extended this to ipv4/ipv6 fragmentation handling and
our generic flow dissector.
14) The generic driver layer takes care now to set the driver data to
NULL on device removal, so it's no longer necessary for drivers to
explicitly set it to NULL any more. Many drivers have been cleaned
up in this way, from Jingoo Han.
15) Add a BPF based packet scheduler classifier, from Daniel Borkmann.
16) Improve CRC32 interfaces and generic SKB checksum iterators so that
SCTP's checksumming can more cleanly be handled. Also from Daniel
Borkmann.
17) Add a new PMTU discovery mode, IP_PMTUDISC_INTERFACE, which forces
using the interface MTU value. This helps avoid PMTU attacks,
particularly on DNS servers. From Hannes Frederic Sowa.
18) Use generic XPS for transmit queue steering rather than internal
(re-)implementation in virtio-net. From Jason Wang.
* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-next: (1622 commits)
random32: add test cases for taus113 implementation
random32: upgrade taus88 generator to taus113 from errata paper
random32: move rnd_state to linux/random.h
random32: add prandom_reseed_late() and call when nonblocking pool becomes initialized
random32: add periodic reseeding
random32: fix off-by-one in seeding requirement
PHY: Add RTL8201CP phy_driver to realtek
xtsonic: add missing platform_set_drvdata() in xtsonic_probe()
macmace: add missing platform_set_drvdata() in mace_probe()
ethernet/arc/arc_emac: add missing platform_set_drvdata() in arc_emac_probe()
ipv6: protect for_each_sk_fl_rcu in mem_check with rcu_read_lock_bh
vlan: Implement vlan_dev_get_egress_qos_mask as an inline.
ixgbe: add warning when max_vfs is out of range.
igb: Update link modes display in ethtool
netfilter: push reasm skb through instead of original frag skbs
ip6_output: fragment outgoing reassembled skb properly
MAINTAINERS: mv643xx_eth: take over maintainership from Lennart
net_sched: tbf: support of 64bit rates
ixgbe: deleting dfwd stations out of order can cause null ptr deref
ixgbe: fix build err, num_rx_queues is only available with CONFIG_RPS
...
Diffstat (limited to 'net/batman-adv/gateway_client.c')
-rw-r--r-- | net/batman-adv/gateway_client.c | 247 |
1 files changed, 159 insertions, 88 deletions
diff --git a/net/batman-adv/gateway_client.c b/net/batman-adv/gateway_client.c index 1ce4b8763ef..2449afaa763 100644 --- a/net/batman-adv/gateway_client.c +++ b/net/batman-adv/gateway_client.c @@ -118,7 +118,6 @@ batadv_gw_get_best_gw_node(struct batadv_priv *bat_priv) uint32_t max_gw_factor = 0, tmp_gw_factor = 0; uint32_t gw_divisor; uint8_t max_tq = 0; - int down, up; uint8_t tq_avg; struct batadv_orig_node *orig_node; @@ -138,14 +137,13 @@ batadv_gw_get_best_gw_node(struct batadv_priv *bat_priv) if (!atomic_inc_not_zero(&gw_node->refcount)) goto next; - tq_avg = router->tq_avg; + tq_avg = router->bat_iv.tq_avg; switch (atomic_read(&bat_priv->gw_sel_class)) { case 1: /* fast connection */ - batadv_gw_bandwidth_to_kbit(orig_node->gw_flags, - &down, &up); - - tmp_gw_factor = tq_avg * tq_avg * down * 100 * 100; + tmp_gw_factor = tq_avg * tq_avg; + tmp_gw_factor *= gw_node->bandwidth_down; + tmp_gw_factor *= 100 * 100; tmp_gw_factor /= gw_divisor; if ((tmp_gw_factor > max_gw_factor) || @@ -223,11 +221,6 @@ void batadv_gw_election(struct batadv_priv *bat_priv) struct batadv_neigh_node *router = NULL; char gw_addr[18] = { '\0' }; - /* The batman daemon checks here if we already passed a full originator - * cycle in order to make sure we don't choose the first gateway we - * hear about. This check is based on the daemon's uptime which we - * don't have. - */ if (atomic_read(&bat_priv->gw_mode) != BATADV_GW_MODE_CLIENT) goto out; @@ -258,16 +251,22 @@ void batadv_gw_election(struct batadv_priv *bat_priv) NULL); } else if ((!curr_gw) && (next_gw)) { batadv_dbg(BATADV_DBG_BATMAN, bat_priv, - "Adding route to gateway %pM (gw_flags: %i, tq: %i)\n", + "Adding route to gateway %pM (bandwidth: %u.%u/%u.%u MBit, tq: %i)\n", next_gw->orig_node->orig, - next_gw->orig_node->gw_flags, router->tq_avg); + next_gw->bandwidth_down / 10, + next_gw->bandwidth_down % 10, + next_gw->bandwidth_up / 10, + next_gw->bandwidth_up % 10, router->bat_iv.tq_avg); batadv_throw_uevent(bat_priv, BATADV_UEV_GW, BATADV_UEV_ADD, gw_addr); } else { batadv_dbg(BATADV_DBG_BATMAN, bat_priv, - "Changing route to gateway %pM (gw_flags: %i, tq: %i)\n", + "Changing route to gateway %pM (bandwidth: %u.%u/%u.%u MBit, tq: %i)\n", next_gw->orig_node->orig, - next_gw->orig_node->gw_flags, router->tq_avg); + next_gw->bandwidth_down / 10, + next_gw->bandwidth_down % 10, + next_gw->bandwidth_up / 10, + next_gw->bandwidth_up % 10, router->bat_iv.tq_avg); batadv_throw_uevent(bat_priv, BATADV_UEV_GW, BATADV_UEV_CHANGE, gw_addr); } @@ -306,8 +305,8 @@ void batadv_gw_check_election(struct batadv_priv *bat_priv, if (!router_orig) goto out; - gw_tq_avg = router_gw->tq_avg; - orig_tq_avg = router_orig->tq_avg; + gw_tq_avg = router_gw->bat_iv.tq_avg; + orig_tq_avg = router_orig->bat_iv.tq_avg; /* the TQ value has to be better */ if (orig_tq_avg < gw_tq_avg) @@ -337,12 +336,20 @@ out: return; } +/** + * batadv_gw_node_add - add gateway node to list of available gateways + * @bat_priv: the bat priv with all the soft interface information + * @orig_node: originator announcing gateway capabilities + * @gateway: announced bandwidth information + */ static void batadv_gw_node_add(struct batadv_priv *bat_priv, struct batadv_orig_node *orig_node, - uint8_t new_gwflags) + struct batadv_tvlv_gateway_data *gateway) { struct batadv_gw_node *gw_node; - int down, up; + + if (gateway->bandwidth_down == 0) + return; gw_node = kzalloc(sizeof(*gw_node), GFP_ATOMIC); if (!gw_node) @@ -356,73 +363,116 @@ static void batadv_gw_node_add(struct batadv_priv *bat_priv, hlist_add_head_rcu(&gw_node->list, &bat_priv->gw.list); spin_unlock_bh(&bat_priv->gw.list_lock); - batadv_gw_bandwidth_to_kbit(new_gwflags, &down, &up); batadv_dbg(BATADV_DBG_BATMAN, bat_priv, - "Found new gateway %pM -> gw_class: %i - %i%s/%i%s\n", - orig_node->orig, new_gwflags, - (down > 2048 ? down / 1024 : down), - (down > 2048 ? "MBit" : "KBit"), - (up > 2048 ? up / 1024 : up), - (up > 2048 ? "MBit" : "KBit")); + "Found new gateway %pM -> gw bandwidth: %u.%u/%u.%u MBit\n", + orig_node->orig, + ntohl(gateway->bandwidth_down) / 10, + ntohl(gateway->bandwidth_down) % 10, + ntohl(gateway->bandwidth_up) / 10, + ntohl(gateway->bandwidth_up) % 10); } -void batadv_gw_node_update(struct batadv_priv *bat_priv, - struct batadv_orig_node *orig_node, - uint8_t new_gwflags) +/** + * batadv_gw_node_get - retrieve gateway node from list of available gateways + * @bat_priv: the bat priv with all the soft interface information + * @orig_node: originator announcing gateway capabilities + * + * Returns gateway node if found or NULL otherwise. + */ +static struct batadv_gw_node * +batadv_gw_node_get(struct batadv_priv *bat_priv, + struct batadv_orig_node *orig_node) { - struct batadv_gw_node *gw_node, *curr_gw; - - /* Note: We don't need a NULL check here, since curr_gw never gets - * dereferenced. If curr_gw is NULL we also should not exit as we may - * have this gateway in our list (duplication check!) even though we - * have no currently selected gateway. - */ - curr_gw = batadv_gw_get_selected_gw_node(bat_priv); + struct batadv_gw_node *gw_node_tmp, *gw_node = NULL; rcu_read_lock(); - hlist_for_each_entry_rcu(gw_node, &bat_priv->gw.list, list) { - if (gw_node->orig_node != orig_node) + hlist_for_each_entry_rcu(gw_node_tmp, &bat_priv->gw.list, list) { + if (gw_node_tmp->orig_node != orig_node) continue; - batadv_dbg(BATADV_DBG_BATMAN, bat_priv, - "Gateway class of originator %pM changed from %i to %i\n", - orig_node->orig, gw_node->orig_node->gw_flags, - new_gwflags); + if (gw_node_tmp->deleted) + continue; - gw_node->deleted = 0; + if (!atomic_inc_not_zero(&gw_node_tmp->refcount)) + continue; - if (new_gwflags == BATADV_NO_FLAGS) { - gw_node->deleted = jiffies; - batadv_dbg(BATADV_DBG_BATMAN, bat_priv, - "Gateway %pM removed from gateway list\n", - orig_node->orig); + gw_node = gw_node_tmp; + break; + } + rcu_read_unlock(); - if (gw_node == curr_gw) - goto deselect; - } + return gw_node; +} - goto unlock; +/** + * batadv_gw_node_update - update list of available gateways with changed + * bandwidth information + * @bat_priv: the bat priv with all the soft interface information + * @orig_node: originator announcing gateway capabilities + * @gateway: announced bandwidth information + */ +void batadv_gw_node_update(struct batadv_priv *bat_priv, + struct batadv_orig_node *orig_node, + struct batadv_tvlv_gateway_data *gateway) +{ + struct batadv_gw_node *gw_node, *curr_gw = NULL; + + gw_node = batadv_gw_node_get(bat_priv, orig_node); + if (!gw_node) { + batadv_gw_node_add(bat_priv, orig_node, gateway); + goto out; } - if (new_gwflags == BATADV_NO_FLAGS) - goto unlock; + if ((gw_node->bandwidth_down == ntohl(gateway->bandwidth_down)) && + (gw_node->bandwidth_up == ntohl(gateway->bandwidth_up))) + goto out; - batadv_gw_node_add(bat_priv, orig_node, new_gwflags); - goto unlock; + batadv_dbg(BATADV_DBG_BATMAN, bat_priv, + "Gateway bandwidth of originator %pM changed from %u.%u/%u.%u MBit to %u.%u/%u.%u MBit\n", + orig_node->orig, + gw_node->bandwidth_down / 10, + gw_node->bandwidth_down % 10, + gw_node->bandwidth_up / 10, + gw_node->bandwidth_up % 10, + ntohl(gateway->bandwidth_down) / 10, + ntohl(gateway->bandwidth_down) % 10, + ntohl(gateway->bandwidth_up) / 10, + ntohl(gateway->bandwidth_up) % 10); + + gw_node->bandwidth_down = ntohl(gateway->bandwidth_down); + gw_node->bandwidth_up = ntohl(gateway->bandwidth_up); + + gw_node->deleted = 0; + if (ntohl(gateway->bandwidth_down) == 0) { + gw_node->deleted = jiffies; + batadv_dbg(BATADV_DBG_BATMAN, bat_priv, + "Gateway %pM removed from gateway list\n", + orig_node->orig); -deselect: - batadv_gw_deselect(bat_priv); -unlock: - rcu_read_unlock(); + /* Note: We don't need a NULL check here, since curr_gw never + * gets dereferenced. + */ + curr_gw = batadv_gw_get_selected_gw_node(bat_priv); + if (gw_node == curr_gw) + batadv_gw_deselect(bat_priv); + } +out: if (curr_gw) batadv_gw_node_free_ref(curr_gw); + if (gw_node) + batadv_gw_node_free_ref(gw_node); } void batadv_gw_node_delete(struct batadv_priv *bat_priv, struct batadv_orig_node *orig_node) { - batadv_gw_node_update(bat_priv, orig_node, 0); + struct batadv_tvlv_gateway_data gateway; + + gateway.bandwidth_down = 0; + gateway.bandwidth_up = 0; + + batadv_gw_node_update(bat_priv, orig_node, &gateway); } void batadv_gw_node_purge(struct batadv_priv *bat_priv) @@ -467,9 +517,7 @@ static int batadv_write_buffer_text(struct batadv_priv *bat_priv, { struct batadv_gw_node *curr_gw; struct batadv_neigh_node *router; - int down, up, ret = -1; - - batadv_gw_bandwidth_to_kbit(gw_node->orig_node->gw_flags, &down, &up); + int ret = -1; router = batadv_orig_node_get_router(gw_node->orig_node); if (!router) @@ -477,16 +525,15 @@ static int batadv_write_buffer_text(struct batadv_priv *bat_priv, curr_gw = batadv_gw_get_selected_gw_node(bat_priv); - ret = seq_printf(seq, "%s %pM (%3i) %pM [%10s]: %3i - %i%s/%i%s\n", + ret = seq_printf(seq, "%s %pM (%3i) %pM [%10s]: %u.%u/%u.%u MBit\n", (curr_gw == gw_node ? "=>" : " "), gw_node->orig_node->orig, - router->tq_avg, router->addr, + router->bat_iv.tq_avg, router->addr, router->if_incoming->net_dev->name, - gw_node->orig_node->gw_flags, - (down > 2048 ? down / 1024 : down), - (down > 2048 ? "MBit" : "KBit"), - (up > 2048 ? up / 1024 : up), - (up > 2048 ? "MBit" : "KBit")); + gw_node->bandwidth_down / 10, + gw_node->bandwidth_down % 10, + gw_node->bandwidth_up / 10, + gw_node->bandwidth_up % 10); batadv_neigh_node_free_ref(router); if (curr_gw) @@ -508,7 +555,7 @@ int batadv_gw_client_seq_print_text(struct seq_file *seq, void *offset) goto out; seq_printf(seq, - " %-12s (%s/%i) %17s [%10s]: gw_class ... [B.A.T.M.A.N. adv %s, MainIF/MAC: %s/%pM (%s)]\n", + " %-12s (%s/%i) %17s [%10s]: advertised uplink bandwidth ... [B.A.T.M.A.N. adv %s, MainIF/MAC: %s/%pM (%s)]\n", "Gateway", "#", BATADV_TQ_MAX_VALUE, "Nexthop", "outgoingIF", BATADV_SOURCE_VERSION, primary_if->net_dev->name, primary_if->net_dev->dev_addr, net_dev->name); @@ -603,24 +650,29 @@ bool batadv_gw_is_dhcp_target(struct sk_buff *skb, unsigned int *header_len) struct iphdr *iphdr; struct ipv6hdr *ipv6hdr; struct udphdr *udphdr; + struct vlan_ethhdr *vhdr; + __be16 proto; /* check for ethernet header */ if (!pskb_may_pull(skb, *header_len + ETH_HLEN)) return false; ethhdr = (struct ethhdr *)skb->data; + proto = ethhdr->h_proto; *header_len += ETH_HLEN; /* check for initial vlan header */ - if (ntohs(ethhdr->h_proto) == ETH_P_8021Q) { + if (proto == htons(ETH_P_8021Q)) { if (!pskb_may_pull(skb, *header_len + VLAN_HLEN)) return false; - ethhdr = (struct ethhdr *)(skb->data + VLAN_HLEN); + + vhdr = (struct vlan_ethhdr *)skb->data; + proto = vhdr->h_vlan_encapsulated_proto; *header_len += VLAN_HLEN; } /* check for ip header */ - switch (ntohs(ethhdr->h_proto)) { - case ETH_P_IP: + switch (proto) { + case htons(ETH_P_IP): if (!pskb_may_pull(skb, *header_len + sizeof(*iphdr))) return false; iphdr = (struct iphdr *)(skb->data + *header_len); @@ -631,7 +683,7 @@ bool batadv_gw_is_dhcp_target(struct sk_buff *skb, unsigned int *header_len) return false; break; - case ETH_P_IPV6: + case htons(ETH_P_IPV6): if (!pskb_may_pull(skb, *header_len + sizeof(*ipv6hdr))) return false; ipv6hdr = (struct ipv6hdr *)(skb->data + *header_len); @@ -658,28 +710,44 @@ bool batadv_gw_is_dhcp_target(struct sk_buff *skb, unsigned int *header_len) *header_len += sizeof(*udphdr); /* check for bootp port */ - if ((ntohs(ethhdr->h_proto) == ETH_P_IP) && - (ntohs(udphdr->dest) != 67)) + if ((proto == htons(ETH_P_IP)) && + (udphdr->dest != htons(67))) return false; - if ((ntohs(ethhdr->h_proto) == ETH_P_IPV6) && - (ntohs(udphdr->dest) != 547)) + if ((proto == htons(ETH_P_IPV6)) && + (udphdr->dest != htons(547))) return false; return true; } -/* this call might reallocate skb data */ +/** + * batadv_gw_out_of_range - check if the dhcp request destination is the best gw + * @bat_priv: the bat priv with all the soft interface information + * @skb: the outgoing packet + * + * Check if the skb is a DHCP request and if it is sent to the current best GW + * server. Due to topology changes it may be the case that the GW server + * previously selected is not the best one anymore. + * + * Returns true if the packet destination is unicast and it is not the best gw, + * false otherwise. + * + * This call might reallocate skb data. + */ bool batadv_gw_out_of_range(struct batadv_priv *bat_priv, struct sk_buff *skb) { struct batadv_neigh_node *neigh_curr = NULL, *neigh_old = NULL; struct batadv_orig_node *orig_dst_node = NULL; - struct batadv_gw_node *curr_gw = NULL; + struct batadv_gw_node *gw_node = NULL, *curr_gw = NULL; struct ethhdr *ethhdr; bool ret, out_of_range = false; unsigned int header_len = 0; uint8_t curr_tq_avg; + unsigned short vid; + + vid = batadv_get_vid(skb, 0); ret = batadv_gw_is_dhcp_target(skb, &header_len); if (!ret) @@ -687,11 +755,12 @@ bool batadv_gw_out_of_range(struct batadv_priv *bat_priv, ethhdr = (struct ethhdr *)skb->data; orig_dst_node = batadv_transtable_search(bat_priv, ethhdr->h_source, - ethhdr->h_dest); + ethhdr->h_dest, vid); if (!orig_dst_node) goto out; - if (!orig_dst_node->gw_flags) + gw_node = batadv_gw_node_get(bat_priv, orig_dst_node); + if (!gw_node->bandwidth_down == 0) goto out; ret = batadv_is_type_dhcprequest(skb, header_len); @@ -723,7 +792,7 @@ bool batadv_gw_out_of_range(struct batadv_priv *bat_priv, if (!neigh_curr) goto out; - curr_tq_avg = neigh_curr->tq_avg; + curr_tq_avg = neigh_curr->bat_iv.tq_avg; break; case BATADV_GW_MODE_OFF: default: @@ -734,7 +803,7 @@ bool batadv_gw_out_of_range(struct batadv_priv *bat_priv, if (!neigh_old) goto out; - if (curr_tq_avg - neigh_old->tq_avg > BATADV_GW_THRESHOLD) + if (curr_tq_avg - neigh_old->bat_iv.tq_avg > BATADV_GW_THRESHOLD) out_of_range = true; out: @@ -742,6 +811,8 @@ out: batadv_orig_node_free_ref(orig_dst_node); if (curr_gw) batadv_gw_node_free_ref(curr_gw); + if (gw_node) + batadv_gw_node_free_ref(gw_node); if (neigh_old) batadv_neigh_node_free_ref(neigh_old); if (neigh_curr) |