From 80e73cc563c4359be809a03bcb8e7e28141a813a Mon Sep 17 00:00:00 2001 From: Nikolay Aleksandrov Date: Tue, 28 Jun 2016 16:57:05 +0200 Subject: net: rtnetlink: add support for the IFLA_STATS_LINK_XSTATS_SLAVE attribute This patch adds support for the IFLA_STATS_LINK_XSTATS_SLAVE attribute which allows to export per-slave statistics if the master device supports the linkxstats callback. The attribute is passed down to the linkxstats callback and it is up to the callback user to use it (an example has been added to the only current user - the bridge). This allows us to query only specific slaves of master devices like bridge ports and export only what we're interested in instead of having to dump all ports and searching only for a single one. This will be used to export per-port IGMP/MLD stats and also per-port vlan stats in the future, possibly other statistics as well. Signed-off-by: Nikolay Aleksandrov Signed-off-by: David S. Miller --- include/uapi/linux/if_link.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/uapi/linux/if_link.h') diff --git a/include/uapi/linux/if_link.h b/include/uapi/linux/if_link.h index bb36bd5675a7..db2458ade81c 100644 --- a/include/uapi/linux/if_link.h +++ b/include/uapi/linux/if_link.h @@ -822,6 +822,7 @@ enum { IFLA_STATS_UNSPEC, /* also used as 64bit pad attribute */ IFLA_STATS_LINK_64, IFLA_STATS_LINK_XSTATS, + IFLA_STATS_LINK_XSTATS_SLAVE, __IFLA_STATS_MAX, }; -- cgit v1.2.3-59-g8ed1b From 1080ab95e3c7bdd77870e209aff83c763fdcf439 Mon Sep 17 00:00:00 2001 From: Nikolay Aleksandrov Date: Tue, 28 Jun 2016 16:57:06 +0200 Subject: net: bridge: add support for IGMP/MLD stats and export them via netlink This patch adds stats support for the currently used IGMP/MLD types by the bridge. The stats are per-port (plus one stat per-bridge) and per-direction (RX/TX). The stats are exported via netlink via the new linkxstats API (RTM_GETSTATS). In order to minimize the performance impact, a new option is used to enable/disable the stats - multicast_stats_enabled, similar to the recent vlan stats. Also in order to avoid multiple IGMP/MLD type lookups and checks, we make use of the current "igmp" member of the bridge private skb->cb region to record the type on Rx (both host-generated and external packets pass by multicast_rcv()). We can do that since the igmp member was used as a boolean and all the valid IGMP/MLD types are positive values. The normal bridge fast-path is not affected at all, the only affected paths are the flooding ones and since we make use of the IGMP/MLD type, we can quickly determine if the packet should be counted using cache-hot data (cb's igmp member). We add counters for: * IGMP Queries * IGMP Leaves * IGMP v1/v2/v3 reports * MLD Queries * MLD Leaves * MLD v1/v2 reports These are invaluable when monitoring or debugging complex multicast setups with bridges. Signed-off-by: Nikolay Aleksandrov Signed-off-by: David S. Miller --- include/uapi/linux/if_bridge.h | 26 +++++ include/uapi/linux/if_link.h | 1 + net/bridge/br_device.c | 10 +- net/bridge/br_forward.c | 13 ++- net/bridge/br_if.c | 9 +- net/bridge/br_input.c | 3 + net/bridge/br_multicast.c | 217 ++++++++++++++++++++++++++++++++++++++--- net/bridge/br_netlink.c | 91 ++++++++++++----- net/bridge/br_private.h | 41 +++++++- net/bridge/br_sysfs_br.c | 25 +++++ 10 files changed, 390 insertions(+), 46 deletions(-) (limited to 'include/uapi/linux/if_link.h') diff --git a/include/uapi/linux/if_bridge.h b/include/uapi/linux/if_bridge.h index 397d503fdedb..8304fe6f0561 100644 --- a/include/uapi/linux/if_bridge.h +++ b/include/uapi/linux/if_bridge.h @@ -247,8 +247,34 @@ enum { enum { BRIDGE_XSTATS_UNSPEC, BRIDGE_XSTATS_VLAN, + BRIDGE_XSTATS_MCAST, + BRIDGE_XSTATS_PAD, __BRIDGE_XSTATS_MAX }; #define BRIDGE_XSTATS_MAX (__BRIDGE_XSTATS_MAX - 1) +enum { + BR_MCAST_DIR_RX, + BR_MCAST_DIR_TX, + BR_MCAST_DIR_SIZE +}; + +/* IGMP/MLD statistics */ +struct br_mcast_stats { + __u64 igmp_queries[BR_MCAST_DIR_SIZE]; + __u64 igmp_leaves[BR_MCAST_DIR_SIZE]; + __u64 igmp_v1reports[BR_MCAST_DIR_SIZE]; + __u64 igmp_v2reports[BR_MCAST_DIR_SIZE]; + __u64 igmp_v3reports[BR_MCAST_DIR_SIZE]; + __u64 igmp_parse_errors; + + __u64 mld_queries[BR_MCAST_DIR_SIZE]; + __u64 mld_leaves[BR_MCAST_DIR_SIZE]; + __u64 mld_v1reports[BR_MCAST_DIR_SIZE]; + __u64 mld_v2reports[BR_MCAST_DIR_SIZE]; + __u64 mld_parse_errors; + + __u64 mcast_bytes[BR_MCAST_DIR_SIZE]; + __u64 mcast_packets[BR_MCAST_DIR_SIZE]; +}; #endif /* _UAPI_LINUX_IF_BRIDGE_H */ diff --git a/include/uapi/linux/if_link.h b/include/uapi/linux/if_link.h index db2458ade81c..4285ac31e865 100644 --- a/include/uapi/linux/if_link.h +++ b/include/uapi/linux/if_link.h @@ -273,6 +273,7 @@ enum { IFLA_BR_VLAN_DEFAULT_PVID, IFLA_BR_PAD, IFLA_BR_VLAN_STATS_ENABLED, + IFLA_BR_MCAST_STATS_ENABLED, __IFLA_BR_MAX, }; diff --git a/net/bridge/br_device.c b/net/bridge/br_device.c index 2c8095a5d824..0c39e0f6da09 100644 --- a/net/bridge/br_device.c +++ b/net/bridge/br_device.c @@ -104,8 +104,16 @@ static int br_dev_init(struct net_device *dev) return -ENOMEM; err = br_vlan_init(br); - if (err) + if (err) { free_percpu(br->stats); + return err; + } + + err = br_multicast_init_stats(br); + if (err) { + free_percpu(br->stats); + br_vlan_flush(br); + } br_set_lockdep_class(dev); return err; diff --git a/net/bridge/br_forward.c b/net/bridge/br_forward.c index f47759f05b6d..6c196037d818 100644 --- a/net/bridge/br_forward.c +++ b/net/bridge/br_forward.c @@ -198,8 +198,10 @@ static void br_flood(struct net_bridge *br, struct sk_buff *skb, struct sk_buff *skb), bool unicast) { - struct net_bridge_port *p; + u8 igmp_type = br_multicast_igmp_type(skb); + __be16 proto = skb->protocol; struct net_bridge_port *prev; + struct net_bridge_port *p; prev = NULL; @@ -218,6 +220,9 @@ static void br_flood(struct net_bridge *br, struct sk_buff *skb, prev = maybe_deliver(prev, p, skb, __packet_hook); if (IS_ERR(prev)) goto out; + if (prev == p) + br_multicast_count(p->br, p, proto, igmp_type, + BR_MCAST_DIR_TX); } if (!prev) @@ -257,9 +262,12 @@ static void br_multicast_flood(struct net_bridge_mdb_entry *mdst, struct sk_buff *skb)) { struct net_device *dev = BR_INPUT_SKB_CB(skb)->brdev; + u8 igmp_type = br_multicast_igmp_type(skb); struct net_bridge *br = netdev_priv(dev); struct net_bridge_port *prev = NULL; struct net_bridge_port_group *p; + __be16 proto = skb->protocol; + struct hlist_node *rp; rp = rcu_dereference(hlist_first_rcu(&br->router_list)); @@ -277,6 +285,9 @@ static void br_multicast_flood(struct net_bridge_mdb_entry *mdst, prev = maybe_deliver(prev, port, skb, __packet_hook); if (IS_ERR(prev)) goto out; + if (prev == port) + br_multicast_count(port->br, port, proto, igmp_type, + BR_MCAST_DIR_TX); if ((unsigned long)lport >= (unsigned long)port) p = rcu_dereference(p->next); diff --git a/net/bridge/br_if.c b/net/bridge/br_if.c index 8217aecf025b..f2fede05d32c 100644 --- a/net/bridge/br_if.c +++ b/net/bridge/br_if.c @@ -345,8 +345,8 @@ static int find_portno(struct net_bridge *br) static struct net_bridge_port *new_nbp(struct net_bridge *br, struct net_device *dev) { - int index; struct net_bridge_port *p; + int index, err; index = find_portno(br); if (index < 0) @@ -366,7 +366,12 @@ static struct net_bridge_port *new_nbp(struct net_bridge *br, br_init_port(p); br_set_state(p, BR_STATE_DISABLED); br_stp_port_timer_init(p); - br_multicast_add_port(p); + err = br_multicast_add_port(p); + if (err) { + dev_put(dev); + kfree(p); + p = ERR_PTR(err); + } return p; } diff --git a/net/bridge/br_input.c b/net/bridge/br_input.c index 43d2cd862bc2..786602bc0567 100644 --- a/net/bridge/br_input.c +++ b/net/bridge/br_input.c @@ -60,6 +60,9 @@ static int br_pass_frame_up(struct sk_buff *skb) skb = br_handle_vlan(br, vg, skb); if (!skb) return NET_RX_DROP; + /* update the multicast stats if the packet is IGMP/MLD */ + br_multicast_count(br, NULL, skb->protocol, br_multicast_igmp_type(skb), + BR_MCAST_DIR_TX); return NF_HOOK(NFPROTO_BRIDGE, NF_BR_LOCAL_IN, dev_net(indev), NULL, skb, indev, NULL, diff --git a/net/bridge/br_multicast.c b/net/bridge/br_multicast.c index 43844144c9c4..e405eef0ae2e 100644 --- a/net/bridge/br_multicast.c +++ b/net/bridge/br_multicast.c @@ -361,7 +361,8 @@ out: } static struct sk_buff *br_ip4_multicast_alloc_query(struct net_bridge *br, - __be32 group) + __be32 group, + u8 *igmp_type) { struct sk_buff *skb; struct igmphdr *ih; @@ -411,6 +412,7 @@ static struct sk_buff *br_ip4_multicast_alloc_query(struct net_bridge *br, skb_set_transport_header(skb, skb->len); ih = igmp_hdr(skb); + *igmp_type = IGMP_HOST_MEMBERSHIP_QUERY; ih->type = IGMP_HOST_MEMBERSHIP_QUERY; ih->code = (group ? br->multicast_last_member_interval : br->multicast_query_response_interval) / @@ -428,7 +430,8 @@ out: #if IS_ENABLED(CONFIG_IPV6) static struct sk_buff *br_ip6_multicast_alloc_query(struct net_bridge *br, - const struct in6_addr *group) + const struct in6_addr *grp, + u8 *igmp_type) { struct sk_buff *skb; struct ipv6hdr *ip6h; @@ -487,16 +490,17 @@ static struct sk_buff *br_ip6_multicast_alloc_query(struct net_bridge *br, skb_set_transport_header(skb, skb->len); mldq = (struct mld_msg *) icmp6_hdr(skb); - interval = ipv6_addr_any(group) ? + interval = ipv6_addr_any(grp) ? br->multicast_query_response_interval : br->multicast_last_member_interval; + *igmp_type = ICMPV6_MGM_QUERY; mldq->mld_type = ICMPV6_MGM_QUERY; mldq->mld_code = 0; mldq->mld_cksum = 0; mldq->mld_maxdelay = htons((u16)jiffies_to_msecs(interval)); mldq->mld_reserved = 0; - mldq->mld_mca = *group; + mldq->mld_mca = *grp; /* checksum */ mldq->mld_cksum = csum_ipv6_magic(&ip6h->saddr, &ip6h->daddr, @@ -513,14 +517,16 @@ out: #endif static struct sk_buff *br_multicast_alloc_query(struct net_bridge *br, - struct br_ip *addr) + struct br_ip *addr, + u8 *igmp_type) { switch (addr->proto) { case htons(ETH_P_IP): - return br_ip4_multicast_alloc_query(br, addr->u.ip4); + return br_ip4_multicast_alloc_query(br, addr->u.ip4, igmp_type); #if IS_ENABLED(CONFIG_IPV6) case htons(ETH_P_IPV6): - return br_ip6_multicast_alloc_query(br, &addr->u.ip6); + return br_ip6_multicast_alloc_query(br, &addr->u.ip6, + igmp_type); #endif } return NULL; @@ -829,18 +835,23 @@ static void __br_multicast_send_query(struct net_bridge *br, struct br_ip *ip) { struct sk_buff *skb; + u8 igmp_type; - skb = br_multicast_alloc_query(br, ip); + skb = br_multicast_alloc_query(br, ip, &igmp_type); if (!skb) return; if (port) { skb->dev = port->dev; + br_multicast_count(br, port, skb->protocol, igmp_type, + BR_MCAST_DIR_TX); NF_HOOK(NFPROTO_BRIDGE, NF_BR_LOCAL_OUT, dev_net(port->dev), NULL, skb, NULL, skb->dev, br_dev_queue_push_xmit); } else { br_multicast_select_own_querier(br, ip, skb); + br_multicast_count(br, port, skb->protocol, igmp_type, + BR_MCAST_DIR_RX); netif_rx(skb); } } @@ -918,7 +929,7 @@ static void br_ip6_multicast_port_query_expired(unsigned long data) } #endif -void br_multicast_add_port(struct net_bridge_port *port) +int br_multicast_add_port(struct net_bridge_port *port) { port->multicast_router = MDB_RTR_TYPE_TEMP_QUERY; @@ -930,6 +941,11 @@ void br_multicast_add_port(struct net_bridge_port *port) setup_timer(&port->ip6_own_query.timer, br_ip6_multicast_port_query_expired, (unsigned long)port); #endif + port->mcast_stats = netdev_alloc_pcpu_stats(struct bridge_mcast_stats); + if (!port->mcast_stats) + return -ENOMEM; + + return 0; } void br_multicast_del_port(struct net_bridge_port *port) @@ -944,6 +960,7 @@ void br_multicast_del_port(struct net_bridge_port *port) br_multicast_del_pg(br, pg); spin_unlock_bh(&br->multicast_lock); del_timer_sync(&port->multicast_router_timer); + free_percpu(port->mcast_stats); } static void br_multicast_enable(struct bridge_mcast_own_query *query) @@ -1583,6 +1600,39 @@ static void br_ip6_multicast_leave_group(struct net_bridge *br, } #endif +static void br_multicast_err_count(const struct net_bridge *br, + const struct net_bridge_port *p, + __be16 proto) +{ + struct bridge_mcast_stats __percpu *stats; + struct bridge_mcast_stats *pstats; + + if (!br->multicast_stats_enabled) + return; + + if (p) + stats = p->mcast_stats; + else + stats = br->mcast_stats; + if (WARN_ON(!stats)) + return; + + pstats = this_cpu_ptr(stats); + + u64_stats_update_begin(&pstats->syncp); + switch (proto) { + case htons(ETH_P_IP): + pstats->mstats.igmp_parse_errors++; + break; +#if IS_ENABLED(CONFIG_IPV6) + case htons(ETH_P_IPV6): + pstats->mstats.mld_parse_errors++; + break; +#endif + } + u64_stats_update_end(&pstats->syncp); +} + static int br_multicast_ipv4_rcv(struct net_bridge *br, struct net_bridge_port *port, struct sk_buff *skb, @@ -1599,11 +1649,12 @@ static int br_multicast_ipv4_rcv(struct net_bridge *br, BR_INPUT_SKB_CB(skb)->mrouters_only = 1; return 0; } else if (err < 0) { + br_multicast_err_count(br, port, skb->protocol); return err; } - BR_INPUT_SKB_CB(skb)->igmp = 1; ih = igmp_hdr(skb); + BR_INPUT_SKB_CB(skb)->igmp = ih->type; switch (ih->type) { case IGMP_HOST_MEMBERSHIP_REPORT: @@ -1625,6 +1676,9 @@ static int br_multicast_ipv4_rcv(struct net_bridge *br, if (skb_trimmed && skb_trimmed != skb) kfree_skb(skb_trimmed); + br_multicast_count(br, port, skb->protocol, BR_INPUT_SKB_CB(skb)->igmp, + BR_MCAST_DIR_RX); + return err; } @@ -1645,11 +1699,12 @@ static int br_multicast_ipv6_rcv(struct net_bridge *br, BR_INPUT_SKB_CB(skb)->mrouters_only = 1; return 0; } else if (err < 0) { + br_multicast_err_count(br, port, skb->protocol); return err; } - BR_INPUT_SKB_CB(skb)->igmp = 1; mld = (struct mld_msg *)skb_transport_header(skb); + BR_INPUT_SKB_CB(skb)->igmp = mld->mld_type; switch (mld->mld_type) { case ICMPV6_MGM_REPORT: @@ -1670,6 +1725,9 @@ static int br_multicast_ipv6_rcv(struct net_bridge *br, if (skb_trimmed && skb_trimmed != skb) kfree_skb(skb_trimmed); + br_multicast_count(br, port, skb->protocol, BR_INPUT_SKB_CB(skb)->igmp, + BR_MCAST_DIR_RX); + return err; } #endif @@ -1677,6 +1735,8 @@ static int br_multicast_ipv6_rcv(struct net_bridge *br, int br_multicast_rcv(struct net_bridge *br, struct net_bridge_port *port, struct sk_buff *skb, u16 vid) { + int ret = 0; + BR_INPUT_SKB_CB(skb)->igmp = 0; BR_INPUT_SKB_CB(skb)->mrouters_only = 0; @@ -1685,14 +1745,16 @@ int br_multicast_rcv(struct net_bridge *br, struct net_bridge_port *port, switch (skb->protocol) { case htons(ETH_P_IP): - return br_multicast_ipv4_rcv(br, port, skb, vid); + ret = br_multicast_ipv4_rcv(br, port, skb, vid); + break; #if IS_ENABLED(CONFIG_IPV6) case htons(ETH_P_IPV6): - return br_multicast_ipv6_rcv(br, port, skb, vid); + ret = br_multicast_ipv6_rcv(br, port, skb, vid); + break; #endif } - return 0; + return ret; } static void br_multicast_query_expired(struct net_bridge *br, @@ -1831,6 +1893,8 @@ void br_multicast_dev_del(struct net_bridge *br) out: spin_unlock_bh(&br->multicast_lock); + + free_percpu(br->mcast_stats); } int br_multicast_set_router(struct net_bridge *br, unsigned long val) @@ -2185,3 +2249,128 @@ unlock: return ret; } EXPORT_SYMBOL_GPL(br_multicast_has_querier_adjacent); + +static void br_mcast_stats_add(struct bridge_mcast_stats __percpu *stats, + __be16 proto, u8 type, u8 dir) +{ + struct bridge_mcast_stats *pstats = this_cpu_ptr(stats); + + u64_stats_update_begin(&pstats->syncp); + switch (proto) { + case htons(ETH_P_IP): + switch (type) { + case IGMP_HOST_MEMBERSHIP_REPORT: + pstats->mstats.igmp_v1reports[dir]++; + break; + case IGMPV2_HOST_MEMBERSHIP_REPORT: + pstats->mstats.igmp_v2reports[dir]++; + break; + case IGMPV3_HOST_MEMBERSHIP_REPORT: + pstats->mstats.igmp_v3reports[dir]++; + break; + case IGMP_HOST_MEMBERSHIP_QUERY: + pstats->mstats.igmp_queries[dir]++; + break; + case IGMP_HOST_LEAVE_MESSAGE: + pstats->mstats.igmp_leaves[dir]++; + break; + } + break; +#if IS_ENABLED(CONFIG_IPV6) + case htons(ETH_P_IPV6): + switch (type) { + case ICMPV6_MGM_REPORT: + pstats->mstats.mld_v1reports[dir]++; + break; + case ICMPV6_MLD2_REPORT: + pstats->mstats.mld_v2reports[dir]++; + break; + case ICMPV6_MGM_QUERY: + pstats->mstats.mld_queries[dir]++; + break; + case ICMPV6_MGM_REDUCTION: + pstats->mstats.mld_leaves[dir]++; + break; + } + break; +#endif /* CONFIG_IPV6 */ + } + u64_stats_update_end(&pstats->syncp); +} + +void br_multicast_count(struct net_bridge *br, const struct net_bridge_port *p, + __be16 proto, u8 type, u8 dir) +{ + struct bridge_mcast_stats __percpu *stats; + + /* if multicast_disabled is true then igmp type can't be set */ + if (!type || !br->multicast_stats_enabled) + return; + + if (p) + stats = p->mcast_stats; + else + stats = br->mcast_stats; + if (WARN_ON(!stats)) + return; + + br_mcast_stats_add(stats, proto, type, dir); +} + +int br_multicast_init_stats(struct net_bridge *br) +{ + br->mcast_stats = netdev_alloc_pcpu_stats(struct bridge_mcast_stats); + if (!br->mcast_stats) + return -ENOMEM; + + return 0; +} + +static void mcast_stats_add_dir(u64 *dst, u64 *src) +{ + dst[BR_MCAST_DIR_RX] += src[BR_MCAST_DIR_RX]; + dst[BR_MCAST_DIR_TX] += src[BR_MCAST_DIR_TX]; +} + +void br_multicast_get_stats(const struct net_bridge *br, + const struct net_bridge_port *p, + struct br_mcast_stats *dest) +{ + struct bridge_mcast_stats __percpu *stats; + struct br_mcast_stats tdst; + int i; + + memset(dest, 0, sizeof(*dest)); + if (p) + stats = p->mcast_stats; + else + stats = br->mcast_stats; + if (WARN_ON(!stats)) + return; + + memset(&tdst, 0, sizeof(tdst)); + for_each_possible_cpu(i) { + struct bridge_mcast_stats *cpu_stats = per_cpu_ptr(stats, i); + struct br_mcast_stats temp; + unsigned int start; + + do { + start = u64_stats_fetch_begin_irq(&cpu_stats->syncp); + memcpy(&temp, &cpu_stats->mstats, sizeof(temp)); + } while (u64_stats_fetch_retry_irq(&cpu_stats->syncp, start)); + + mcast_stats_add_dir(tdst.igmp_queries, temp.igmp_queries); + mcast_stats_add_dir(tdst.igmp_leaves, temp.igmp_leaves); + mcast_stats_add_dir(tdst.igmp_v1reports, temp.igmp_v1reports); + mcast_stats_add_dir(tdst.igmp_v2reports, temp.igmp_v2reports); + mcast_stats_add_dir(tdst.igmp_v3reports, temp.igmp_v3reports); + tdst.igmp_parse_errors += temp.igmp_parse_errors; + + mcast_stats_add_dir(tdst.mld_queries, temp.mld_queries); + mcast_stats_add_dir(tdst.mld_leaves, temp.mld_leaves); + mcast_stats_add_dir(tdst.mld_v1reports, temp.mld_v1reports); + mcast_stats_add_dir(tdst.mld_v2reports, temp.mld_v2reports); + tdst.mld_parse_errors += temp.mld_parse_errors; + } + memcpy(dest, &tdst, sizeof(*dest)); +} diff --git a/net/bridge/br_netlink.c b/net/bridge/br_netlink.c index ed75ff9ff9e6..f2a29e467e78 100644 --- a/net/bridge/br_netlink.c +++ b/net/bridge/br_netlink.c @@ -851,6 +851,7 @@ static const struct nla_policy br_policy[IFLA_BR_MAX + 1] = { [IFLA_BR_NF_CALL_ARPTABLES] = { .type = NLA_U8 }, [IFLA_BR_VLAN_DEFAULT_PVID] = { .type = NLA_U16 }, [IFLA_BR_VLAN_STATS_ENABLED] = { .type = NLA_U8 }, + [IFLA_BR_MCAST_STATS_ENABLED] = { .type = NLA_U8 }, }; static int br_changelink(struct net_device *brdev, struct nlattr *tb[], @@ -1055,6 +1056,13 @@ static int br_changelink(struct net_device *brdev, struct nlattr *tb[], br->multicast_startup_query_interval = clock_t_to_jiffies(val); } + + if (data[IFLA_BR_MCAST_STATS_ENABLED]) { + __u8 mcast_stats; + + mcast_stats = nla_get_u8(data[IFLA_BR_MCAST_STATS_ENABLED]); + br->multicast_stats_enabled = !!mcast_stats; + } #endif #if IS_ENABLED(CONFIG_BRIDGE_NETFILTER) if (data[IFLA_BR_NF_CALL_IPTABLES]) { @@ -1110,6 +1118,7 @@ static size_t br_get_size(const struct net_device *brdev) nla_total_size(sizeof(u8)) + /* IFLA_BR_MCAST_SNOOPING */ nla_total_size(sizeof(u8)) + /* IFLA_BR_MCAST_QUERY_USE_IFADDR */ nla_total_size(sizeof(u8)) + /* IFLA_BR_MCAST_QUERIER */ + nla_total_size(sizeof(u8)) + /* IFLA_BR_MCAST_STATS_ENABLED */ nla_total_size(sizeof(u32)) + /* IFLA_BR_MCAST_HASH_ELASTICITY */ nla_total_size(sizeof(u32)) + /* IFLA_BR_MCAST_HASH_MAX */ nla_total_size(sizeof(u32)) + /* IFLA_BR_MCAST_LAST_MEMBER_CNT */ @@ -1187,6 +1196,8 @@ static int br_fill_info(struct sk_buff *skb, const struct net_device *brdev) nla_put_u8(skb, IFLA_BR_MCAST_QUERY_USE_IFADDR, br->multicast_query_use_ifaddr) || nla_put_u8(skb, IFLA_BR_MCAST_QUERIER, br->multicast_querier) || + nla_put_u8(skb, IFLA_BR_MCAST_STATS_ENABLED, + br->multicast_stats_enabled) || nla_put_u32(skb, IFLA_BR_MCAST_HASH_ELASTICITY, br->hash_elasticity) || nla_put_u32(skb, IFLA_BR_MCAST_HASH_MAX, br->hash_max) || @@ -1242,21 +1253,21 @@ static size_t bridge_get_linkxstats_size(const struct net_device *dev) int numvls = 0; vg = br_vlan_group(br); - if (!vg) - return 0; - - /* we need to count all, even placeholder entries */ - list_for_each_entry(v, &vg->vlan_list, vlist) - numvls++; + if (vg) { + /* we need to count all, even placeholder entries */ + list_for_each_entry(v, &vg->vlan_list, vlist) + numvls++; + } - /* account for the vlans and the link xstats type nest attribute */ return numvls * nla_total_size(sizeof(struct bridge_vlan_xstats)) + + nla_total_size(sizeof(struct br_mcast_stats)) + nla_total_size(0); } static size_t brport_get_linkxstats_size(const struct net_device *dev) { - return nla_total_size(0); + return nla_total_size(sizeof(struct br_mcast_stats)) + + nla_total_size(0); } static size_t br_get_linkxstats_size(const struct net_device *dev, int attr) @@ -1280,37 +1291,50 @@ static int bridge_fill_linkxstats(struct sk_buff *skb, int *prividx) { struct net_bridge *br = netdev_priv(dev); + struct nlattr *nla __maybe_unused; struct net_bridge_vlan_group *vg; struct net_bridge_vlan *v; struct nlattr *nest; int vl_idx = 0; - vg = br_vlan_group(br); - if (!vg) - goto out; nest = nla_nest_start(skb, LINK_XSTATS_TYPE_BRIDGE); if (!nest) return -EMSGSIZE; - list_for_each_entry(v, &vg->vlan_list, vlist) { - struct bridge_vlan_xstats vxi; - struct br_vlan_stats stats; - if (++vl_idx < *prividx) - continue; - memset(&vxi, 0, sizeof(vxi)); - vxi.vid = v->vid; - br_vlan_get_stats(v, &stats); - vxi.rx_bytes = stats.rx_bytes; - vxi.rx_packets = stats.rx_packets; - vxi.tx_bytes = stats.tx_bytes; - vxi.tx_packets = stats.tx_packets; - - if (nla_put(skb, BRIDGE_XSTATS_VLAN, sizeof(vxi), &vxi)) + vg = br_vlan_group(br); + if (vg) { + list_for_each_entry(v, &vg->vlan_list, vlist) { + struct bridge_vlan_xstats vxi; + struct br_vlan_stats stats; + + if (++vl_idx < *prividx) + continue; + memset(&vxi, 0, sizeof(vxi)); + vxi.vid = v->vid; + br_vlan_get_stats(v, &stats); + vxi.rx_bytes = stats.rx_bytes; + vxi.rx_packets = stats.rx_packets; + vxi.tx_bytes = stats.tx_bytes; + vxi.tx_packets = stats.tx_packets; + + if (nla_put(skb, BRIDGE_XSTATS_VLAN, sizeof(vxi), &vxi)) + goto nla_put_failure; + } + } + +#ifdef CONFIG_BRIDGE_IGMP_SNOOPING + if (++vl_idx >= *prividx) { + nla = nla_reserve_64bit(skb, BRIDGE_XSTATS_MCAST, + sizeof(struct br_mcast_stats), + BRIDGE_XSTATS_PAD); + if (!nla) goto nla_put_failure; + br_multicast_get_stats(br, NULL, nla_data(nla)); } +#endif nla_nest_end(skb, nest); *prividx = 0; -out: + return 0; nla_put_failure: @@ -1324,11 +1348,26 @@ static int brport_fill_linkxstats(struct sk_buff *skb, const struct net_device *dev, int *prividx) { + struct net_bridge_port *p = br_port_get_rtnl(dev); + struct nlattr *nla __maybe_unused; struct nlattr *nest; + if (!p) + return 0; + nest = nla_nest_start(skb, LINK_XSTATS_TYPE_BRIDGE); if (!nest) return -EMSGSIZE; +#ifdef CONFIG_BRIDGE_IGMP_SNOOPING + nla = nla_reserve_64bit(skb, BRIDGE_XSTATS_MCAST, + sizeof(struct br_mcast_stats), + BRIDGE_XSTATS_PAD); + if (!nla) { + nla_nest_end(skb, nest); + return -EMSGSIZE; + } + br_multicast_get_stats(p->br, p, nla_data(nla)); +#endif nla_nest_end(skb, nest); return 0; diff --git a/net/bridge/br_private.h b/net/bridge/br_private.h index 52edecf3c294..4dc851166ad1 100644 --- a/net/bridge/br_private.h +++ b/net/bridge/br_private.h @@ -75,6 +75,12 @@ struct bridge_mcast_querier { struct br_ip addr; struct net_bridge_port __rcu *port; }; + +/* IGMP/MLD statistics */ +struct bridge_mcast_stats { + struct br_mcast_stats mstats; + struct u64_stats_sync syncp; +}; #endif struct br_vlan_stats { @@ -229,6 +235,7 @@ struct net_bridge_port struct bridge_mcast_own_query ip6_own_query; #endif /* IS_ENABLED(CONFIG_IPV6) */ unsigned char multicast_router; + struct bridge_mcast_stats __percpu *mcast_stats; struct timer_list multicast_router_timer; struct hlist_head mglist; struct hlist_node rlist; @@ -315,6 +322,7 @@ struct net_bridge u8 multicast_querier:1; u8 multicast_query_use_ifaddr:1; u8 has_ipv6_addr:1; + u8 multicast_stats_enabled:1; u32 hash_elasticity; u32 hash_max; @@ -337,6 +345,7 @@ struct net_bridge struct bridge_mcast_other_query ip4_other_query; struct bridge_mcast_own_query ip4_own_query; struct bridge_mcast_querier ip4_querier; + struct bridge_mcast_stats __percpu *mcast_stats; #if IS_ENABLED(CONFIG_IPV6) struct bridge_mcast_other_query ip6_other_query; struct bridge_mcast_own_query ip6_own_query; @@ -543,7 +552,7 @@ int br_multicast_rcv(struct net_bridge *br, struct net_bridge_port *port, struct sk_buff *skb, u16 vid); struct net_bridge_mdb_entry *br_mdb_get(struct net_bridge *br, struct sk_buff *skb, u16 vid); -void br_multicast_add_port(struct net_bridge_port *port); +int br_multicast_add_port(struct net_bridge_port *port); void br_multicast_del_port(struct net_bridge_port *port); void br_multicast_enable_port(struct net_bridge_port *port); void br_multicast_disable_port(struct net_bridge_port *port); @@ -576,6 +585,12 @@ void br_mdb_notify(struct net_device *dev, struct net_bridge_port *port, struct br_ip *group, int type, u8 flags); void br_rtr_notify(struct net_device *dev, struct net_bridge_port *port, int type); +void br_multicast_count(struct net_bridge *br, const struct net_bridge_port *p, + __be16 proto, u8 type, u8 dir); +int br_multicast_init_stats(struct net_bridge *br); +void br_multicast_get_stats(const struct net_bridge *br, + const struct net_bridge_port *p, + struct br_mcast_stats *dest); #define mlock_dereference(X, br) \ rcu_dereference_protected(X, lockdep_is_held(&br->multicast_lock)) @@ -623,6 +638,11 @@ static inline bool br_multicast_querier_exists(struct net_bridge *br, return false; } } + +static inline int br_multicast_igmp_type(const struct sk_buff *skb) +{ + return BR_INPUT_SKB_CB(skb)->igmp; +} #else static inline int br_multicast_rcv(struct net_bridge *br, struct net_bridge_port *port, @@ -638,8 +658,9 @@ static inline struct net_bridge_mdb_entry *br_mdb_get(struct net_bridge *br, return NULL; } -static inline void br_multicast_add_port(struct net_bridge_port *port) +static inline int br_multicast_add_port(struct net_bridge_port *port) { + return 0; } static inline void br_multicast_del_port(struct net_bridge_port *port) @@ -695,6 +716,22 @@ static inline void br_mdb_init(void) static inline void br_mdb_uninit(void) { } + +static inline void br_multicast_count(struct net_bridge *br, + const struct net_bridge_port *p, + __be16 proto, u8 type, u8 dir) +{ +} + +static inline int br_multicast_init_stats(struct net_bridge *br) +{ + return 0; +} + +static inline int br_multicast_igmp_type(const struct sk_buff *skb) +{ + return 0; +} #endif /* br_vlan.c */ diff --git a/net/bridge/br_sysfs_br.c b/net/bridge/br_sysfs_br.c index beb47071e38d..e120307c6e36 100644 --- a/net/bridge/br_sysfs_br.c +++ b/net/bridge/br_sysfs_br.c @@ -618,6 +618,30 @@ static ssize_t multicast_startup_query_interval_store( return store_bridge_parm(d, buf, len, set_startup_query_interval); } static DEVICE_ATTR_RW(multicast_startup_query_interval); + +static ssize_t multicast_stats_enabled_show(struct device *d, + struct device_attribute *attr, + char *buf) +{ + struct net_bridge *br = to_bridge(d); + + return sprintf(buf, "%u\n", br->multicast_stats_enabled); +} + +static int set_stats_enabled(struct net_bridge *br, unsigned long val) +{ + br->multicast_stats_enabled = !!val; + return 0; +} + +static ssize_t multicast_stats_enabled_store(struct device *d, + struct device_attribute *attr, + const char *buf, + size_t len) +{ + return store_bridge_parm(d, buf, len, set_stats_enabled); +} +static DEVICE_ATTR_RW(multicast_stats_enabled); #endif #if IS_ENABLED(CONFIG_BRIDGE_NETFILTER) static ssize_t nf_call_iptables_show( @@ -784,6 +808,7 @@ static struct attribute *bridge_attrs[] = { &dev_attr_multicast_query_interval.attr, &dev_attr_multicast_query_response_interval.attr, &dev_attr_multicast_startup_query_interval.attr, + &dev_attr_multicast_stats_enabled.attr, #endif #if IS_ENABLED(CONFIG_BRIDGE_NETFILTER) &dev_attr_nf_call_iptables.attr, -- cgit v1.2.3-59-g8ed1b From d1fdd9138682e0f272beee0cb08b6328c5478b26 Mon Sep 17 00:00:00 2001 From: Brenden Blanco Date: Tue, 19 Jul 2016 12:16:49 -0700 Subject: rtnl: add option for setting link xdp prog Sets the bpf program represented by fd as an early filter in the rx path of the netdev. The fd must have been created as BPF_PROG_TYPE_XDP. Providing a negative value as fd clears the program. Getting the fd back via rtnl is not possible, therefore reading of this value merely provides a bool whether the program is valid on the link or not. Signed-off-by: Brenden Blanco Signed-off-by: David S. Miller --- include/uapi/linux/if_link.h | 12 +++++++++ net/core/rtnetlink.c | 64 ++++++++++++++++++++++++++++++++++++++++++++ 2 files changed, 76 insertions(+) (limited to 'include/uapi/linux/if_link.h') diff --git a/include/uapi/linux/if_link.h b/include/uapi/linux/if_link.h index 4285ac31e865..a1b5202c5f6b 100644 --- a/include/uapi/linux/if_link.h +++ b/include/uapi/linux/if_link.h @@ -156,6 +156,7 @@ enum { IFLA_GSO_MAX_SEGS, IFLA_GSO_MAX_SIZE, IFLA_PAD, + IFLA_XDP, __IFLA_MAX }; @@ -843,4 +844,15 @@ enum { }; #define LINK_XSTATS_TYPE_MAX (__LINK_XSTATS_TYPE_MAX - 1) +/* XDP section */ + +enum { + IFLA_XDP_UNSPEC, + IFLA_XDP_FD, + IFLA_XDP_ATTACHED, + __IFLA_XDP_MAX, +}; + +#define IFLA_XDP_MAX (__IFLA_XDP_MAX - 1) + #endif /* _UAPI_LINUX_IF_LINK_H */ diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c index a9e3805af739..eba2b8260dbd 100644 --- a/net/core/rtnetlink.c +++ b/net/core/rtnetlink.c @@ -891,6 +891,16 @@ static size_t rtnl_port_size(const struct net_device *dev, return port_self_size; } +static size_t rtnl_xdp_size(const struct net_device *dev) +{ + size_t xdp_size = nla_total_size(1); /* XDP_ATTACHED */ + + if (!dev->netdev_ops->ndo_xdp) + return 0; + else + return xdp_size; +} + static noinline size_t if_nlmsg_size(const struct net_device *dev, u32 ext_filter_mask) { @@ -927,6 +937,7 @@ static noinline size_t if_nlmsg_size(const struct net_device *dev, + nla_total_size(MAX_PHYS_ITEM_ID_LEN) /* IFLA_PHYS_PORT_ID */ + nla_total_size(MAX_PHYS_ITEM_ID_LEN) /* IFLA_PHYS_SWITCH_ID */ + nla_total_size(IFNAMSIZ) /* IFLA_PHYS_PORT_NAME */ + + rtnl_xdp_size(dev) /* IFLA_XDP */ + nla_total_size(1); /* IFLA_PROTO_DOWN */ } @@ -1211,6 +1222,33 @@ static int rtnl_fill_link_ifmap(struct sk_buff *skb, struct net_device *dev) return 0; } +static int rtnl_xdp_fill(struct sk_buff *skb, struct net_device *dev) +{ + struct netdev_xdp xdp_op = {}; + struct nlattr *xdp; + int err; + + if (!dev->netdev_ops->ndo_xdp) + return 0; + xdp = nla_nest_start(skb, IFLA_XDP); + if (!xdp) + return -EMSGSIZE; + xdp_op.command = XDP_QUERY_PROG; + err = dev->netdev_ops->ndo_xdp(dev, &xdp_op); + if (err) + goto err_cancel; + err = nla_put_u8(skb, IFLA_XDP_ATTACHED, xdp_op.prog_attached); + if (err) + goto err_cancel; + + nla_nest_end(skb, xdp); + return 0; + +err_cancel: + nla_nest_cancel(skb, xdp); + return err; +} + static int rtnl_fill_ifinfo(struct sk_buff *skb, struct net_device *dev, int type, u32 pid, u32 seq, u32 change, unsigned int flags, u32 ext_filter_mask) @@ -1307,6 +1345,9 @@ static int rtnl_fill_ifinfo(struct sk_buff *skb, struct net_device *dev, if (rtnl_port_fill(skb, dev, ext_filter_mask)) goto nla_put_failure; + if (rtnl_xdp_fill(skb, dev)) + goto nla_put_failure; + if (dev->rtnl_link_ops || rtnl_have_link_slave_info(dev)) { if (rtnl_link_fill(skb, dev) < 0) goto nla_put_failure; @@ -1392,6 +1433,7 @@ static const struct nla_policy ifla_policy[IFLA_MAX+1] = { [IFLA_PHYS_SWITCH_ID] = { .type = NLA_BINARY, .len = MAX_PHYS_ITEM_ID_LEN }, [IFLA_LINK_NETNSID] = { .type = NLA_S32 }, [IFLA_PROTO_DOWN] = { .type = NLA_U8 }, + [IFLA_XDP] = { .type = NLA_NESTED }, }; static const struct nla_policy ifla_info_policy[IFLA_INFO_MAX+1] = { @@ -1429,6 +1471,11 @@ static const struct nla_policy ifla_port_policy[IFLA_PORT_MAX+1] = { [IFLA_PORT_RESPONSE] = { .type = NLA_U16, }, }; +static const struct nla_policy ifla_xdp_policy[IFLA_XDP_MAX + 1] = { + [IFLA_XDP_FD] = { .type = NLA_S32 }, + [IFLA_XDP_ATTACHED] = { .type = NLA_U8 }, +}; + static const struct rtnl_link_ops *linkinfo_to_kind_ops(const struct nlattr *nla) { const struct rtnl_link_ops *ops = NULL; @@ -2054,6 +2101,23 @@ static int do_setlink(const struct sk_buff *skb, status |= DO_SETLINK_NOTIFY; } + if (tb[IFLA_XDP]) { + struct nlattr *xdp[IFLA_XDP_MAX + 1]; + + err = nla_parse_nested(xdp, IFLA_XDP_MAX, tb[IFLA_XDP], + ifla_xdp_policy); + if (err < 0) + goto errout; + + if (xdp[IFLA_XDP_FD]) { + err = dev_change_xdp_fd(dev, + nla_get_s32(xdp[IFLA_XDP_FD])); + if (err) + goto errout; + status |= DO_SETLINK_NOTIFY; + } + } + errout: if (status & DO_SETLINK_MODIFIED) { if (status & DO_SETLINK_NOTIFY) -- cgit v1.2.3-59-g8ed1b