From patchwork Wed Feb 25 17:58:35 2015 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Madhu Challa X-Patchwork-Id: 443583 X-Patchwork-Delegate: davem@davemloft.net Return-Path: X-Original-To: patchwork-incoming@ozlabs.org Delivered-To: patchwork-incoming@ozlabs.org Received: from vger.kernel.org (vger.kernel.org [209.132.180.67]) by ozlabs.org (Postfix) with ESMTP id 2FFA414010F for ; Thu, 26 Feb 2015 05:15:23 +1100 (AEDT) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1753011AbbBYSPO (ORCPT ); Wed, 25 Feb 2015 13:15:14 -0500 Received: from mail-pd0-f180.google.com ([209.85.192.180]:33368 "EHLO mail-pd0-f180.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752959AbbBYSPF (ORCPT ); Wed, 25 Feb 2015 13:15:05 -0500 Received: by pdjz10 with SMTP id z10so6650898pdj.0 for ; Wed, 25 Feb 2015 10:15:04 -0800 (PST) X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20130820; h=x-gm-message-state:from:to:cc:subject:date:message-id:in-reply-to :references; bh=MFWkiju/e/SzJ+TCkkfpxmygbcXfRnxd8X0P2CtKmCc=; b=idzrQnZefogvkg86h35zr6EW0ZRv7RP14JeYrH5S5aIR+pEODw9Wo1CK3RVbbeeB3W x4ueBwKUgXyO1Hrt4uhfCfdxMIsv5QLrstOXdU+JgrJ2EQUFEnUm9KeiAYiX17NKcM6i D6ljrXnhAyFnFKiKSL/HBNYwVR7cwiOofRaCAAaVGQoL9R540VcViT5c5+WyES3TkNQD qfKar6UO1ZlGxMHfof8nOeMc7rLg4XV9oSyj917Dwg4upkBwkwDdD6yRHKempAjdF5fR OiRdxFea9Oz2C2FgDcrE0DOzVe1U7iCJzoeJgNqIfmtFo2P1xik4T+aI0Wen8nWXnWDu Z4rQ== X-Gm-Message-State: ALoCoQlDSCnHcpY4vu9Z/RD7Ueb9yfx0/M618MSLZd0gbbSODA8IdyuUcLSX0D6M+iF1L9MS39UR X-Received: by 10.68.224.101 with SMTP id rb5mr7586435pbc.67.1424888104031; Wed, 25 Feb 2015 10:15:04 -0800 (PST) Received: from localhost.localdomain ([173.36.240.158]) by mx.google.com with ESMTPSA id g3sm11598488pdg.39.2015.02.25.10.15.03 (version=TLSv1.2 cipher=ECDHE-RSA-AES128-SHA bits=128/128); Wed, 25 Feb 2015 10:15:03 -0800 (PST) From: Madhu Challa To: eric.dumazet@gmail.com, davem@davemloft.net, daniel@iogearbox.net, netdev@vger.kernel.org Cc: Madhu Challa Subject: [PATCH net-next v5 2/2] multicast: Extend ip address command to enable multicast group join/leave on Date: Wed, 25 Feb 2015 09:58:35 -0800 Message-Id: <1424887115-3483-3-git-send-email-challa@noironetworks.com> X-Mailer: git-send-email 1.9.1 In-Reply-To: <1424887115-3483-1-git-send-email-challa@noironetworks.com> References: <1424887115-3483-1-git-send-email-challa@noironetworks.com> Sender: netdev-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: netdev@vger.kernel.org Joining multicast group on ethernet level via "ip maddr" command would not work if we have an Ethernet switch that does igmp snooping since the switch would not replicate multicast packets on ports that did not have IGMP reports for the multicast addresses. Linux vxlan interfaces created via "ip link add vxlan" have the group option that enables then to do the required join. By extending ip address command with option "autojoin" we can get similar functionality for openvswitch vxlan interfaces as well as other tunneling mechanisms that need to receive multicast traffic. The kernel code is structured similar to how the vxlan driver does a group join / leave. example: ip address add 224.1.1.10/24 dev eth5 autojoin ip address del 224.1.1.10/24 dev eth5 Signed-off-by: Madhu Challa --- include/net/netns/ipv4.h | 1 + include/net/netns/ipv6.h | 1 + include/uapi/linux/if_addr.h | 1 + net/ipv4/devinet.c | 31 +++++++++++++++++++++++++++++++ net/ipv4/igmp.c | 13 +++++++++++++ net/ipv6/addrconf.c | 38 +++++++++++++++++++++++++++++++++++--- net/ipv6/mcast.c | 20 ++++++++++++++++---- 7 files changed, 98 insertions(+), 7 deletions(-) diff --git a/include/net/netns/ipv4.h b/include/net/netns/ipv4.h index dbe2254..1b26c6c 100644 --- a/include/net/netns/ipv4.h +++ b/include/net/netns/ipv4.h @@ -49,6 +49,7 @@ struct netns_ipv4 { struct sock *fibnl; struct sock * __percpu *icmp_sk; + struct sock *mc_autojoin_sk; struct inet_peer_base *peers; struct tcpm_hash_bucket *tcp_metrics_hash; diff --git a/include/net/netns/ipv6.h b/include/net/netns/ipv6.h index 69ae41f..ca0db12 100644 --- a/include/net/netns/ipv6.h +++ b/include/net/netns/ipv6.h @@ -67,6 +67,7 @@ struct netns_ipv6 { struct sock *ndisc_sk; struct sock *tcp_sk; struct sock *igmp_sk; + struct sock *mc_autojoin_sk; #ifdef CONFIG_IPV6_MROUTE #ifndef CONFIG_IPV6_MROUTE_MULTIPLE_TABLES struct mr6_table *mrt6; diff --git a/include/uapi/linux/if_addr.h b/include/uapi/linux/if_addr.h index dea10a8..40fdfea 100644 --- a/include/uapi/linux/if_addr.h +++ b/include/uapi/linux/if_addr.h @@ -50,6 +50,7 @@ enum { #define IFA_F_PERMANENT 0x80 #define IFA_F_MANAGETEMPADDR 0x100 #define IFA_F_NOPREFIXROUTE 0x200 +#define IFA_F_MCAUTOJOIN 0x400 struct ifa_cacheinfo { __u32 ifa_prefered; diff --git a/net/ipv4/devinet.c b/net/ipv4/devinet.c index 3a8985c..5105759 100644 --- a/net/ipv4/devinet.c +++ b/net/ipv4/devinet.c @@ -548,6 +548,26 @@ struct in_ifaddr *inet_ifa_byprefix(struct in_device *in_dev, __be32 prefix, return NULL; } +static int ip_mc_config(struct sock *sk, bool join, const struct in_ifaddr *ifa) +{ + struct ip_mreqn mreq = { + .imr_multiaddr.s_addr = ifa->ifa_address, + .imr_ifindex = ifa->ifa_dev->dev->ifindex, + }; + int ret; + + ASSERT_RTNL(); + + lock_sock(sk); + if (join) + ret = __ip_mc_join_group(sk, &mreq); + else + ret = __ip_mc_leave_group(sk, &mreq); + release_sock(sk); + + return ret; +} + static int inet_rtm_deladdr(struct sk_buff *skb, struct nlmsghdr *nlh) { struct net *net = sock_net(skb->sk); @@ -584,6 +604,8 @@ static int inet_rtm_deladdr(struct sk_buff *skb, struct nlmsghdr *nlh) !inet_ifa_match(nla_get_be32(tb[IFA_ADDRESS]), ifa))) continue; + if (ipv4_is_multicast(ifa->ifa_address)) + ip_mc_config(net->ipv4.mc_autojoin_sk, false, ifa); __inet_del_ifa(in_dev, ifap, 1, nlh, NETLINK_CB(skb).portid); return 0; } @@ -838,6 +860,15 @@ static int inet_rtm_newaddr(struct sk_buff *skb, struct nlmsghdr *nlh) * userspace already relies on not having to provide this. */ set_ifa_lifetime(ifa, valid_lft, prefered_lft); + if (ifa->ifa_flags & IFA_F_MCAUTOJOIN) { + int ret = ip_mc_config(net->ipv4.mc_autojoin_sk, + true, ifa); + + if (ret < 0) { + inet_free_ifa(ifa); + return ret; + } + } return __inet_insert_ifa(ifa, nlh, NETLINK_CB(skb).portid); } else { inet_free_ifa(ifa); diff --git a/net/ipv4/igmp.c b/net/ipv4/igmp.c index 4b1172d..5cb1ef4 100644 --- a/net/ipv4/igmp.c +++ b/net/ipv4/igmp.c @@ -97,6 +97,7 @@ #include #include #include +#include #include #ifdef CONFIG_IP_MROUTE #include @@ -2740,6 +2741,7 @@ static const struct file_operations igmp_mcf_seq_fops = { static int __net_init igmp_net_init(struct net *net) { struct proc_dir_entry *pde; + int err; pde = proc_create("igmp", S_IRUGO, net->proc_net, &igmp_mc_seq_fops); if (!pde) @@ -2748,8 +2750,18 @@ static int __net_init igmp_net_init(struct net *net) &igmp_mcf_seq_fops); if (!pde) goto out_mcfilter; + err = inet_ctl_sock_create(&net->ipv4.mc_autojoin_sk, AF_INET, + SOCK_DGRAM, 0, net); + if (err < 0) { + pr_err("Failed to initialize the IGMP autojoin socket (err %d)\n", + err); + goto out_sock; + } + return 0; +out_sock: + remove_proc_entry("mcfilter", net->proc_net); out_mcfilter: remove_proc_entry("igmp", net->proc_net); out_igmp: @@ -2760,6 +2772,7 @@ static void __net_exit igmp_net_exit(struct net *net) { remove_proc_entry("mcfilter", net->proc_net); remove_proc_entry("igmp", net->proc_net); + inet_ctl_sock_destroy(net->ipv4.mc_autojoin_sk); } static struct pernet_operations igmp_net_ops = { diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c index 98e4a63..783bccf 100644 --- a/net/ipv6/addrconf.c +++ b/net/ipv6/addrconf.c @@ -2464,6 +2464,23 @@ err_exit: return err; } +static int ipv6_mc_config(struct sock *sk, bool join, + const struct in6_addr *addr, int ifindex) +{ + int ret; + + ASSERT_RTNL(); + + lock_sock(sk); + if (join) + ret = __ipv6_sock_mc_join(sk, ifindex, addr); + else + ret = __ipv6_sock_mc_drop(sk, ifindex, addr); + release_sock(sk); + + return ret; +} + /* * Manual configuration of address on an interface */ @@ -2476,10 +2493,10 @@ static int inet6_addr_add(struct net *net, int ifindex, struct inet6_ifaddr *ifp; struct inet6_dev *idev; struct net_device *dev; + unsigned long timeout; + clock_t expires; int scope; u32 flags; - clock_t expires; - unsigned long timeout; ASSERT_RTNL(); @@ -2501,6 +2518,14 @@ static int inet6_addr_add(struct net *net, int ifindex, if (IS_ERR(idev)) return PTR_ERR(idev); + if (ifa_flags & IFA_F_MCAUTOJOIN) { + int ret = ipv6_mc_config(net->ipv6.mc_autojoin_sk, + true, pfx, ifindex); + + if (ret < 0) + return ret; + } + scope = ipv6_addr_scope(pfx); timeout = addrconf_timeout_fixup(valid_lft, HZ); @@ -2542,6 +2567,9 @@ static int inet6_addr_add(struct net *net, int ifindex, in6_ifa_put(ifp); addrconf_verify_rtnl(); return 0; + } else if (ifa_flags & IFA_F_MCAUTOJOIN) { + ipv6_mc_config(net->ipv6.mc_autojoin_sk, + false, pfx, ifindex); } return PTR_ERR(ifp); @@ -2578,6 +2606,10 @@ static int inet6_addr_del(struct net *net, int ifindex, u32 ifa_flags, jiffies); ipv6_del_addr(ifp); addrconf_verify_rtnl(); + if (ipv6_addr_is_multicast(pfx)) { + ipv6_mc_config(net->ipv6.mc_autojoin_sk, + false, pfx, dev->ifindex); + } return 0; } } @@ -3945,7 +3977,7 @@ inet6_rtm_newaddr(struct sk_buff *skb, struct nlmsghdr *nlh) /* We ignore other flags so far. */ ifa_flags &= IFA_F_NODAD | IFA_F_HOMEADDRESS | IFA_F_MANAGETEMPADDR | - IFA_F_NOPREFIXROUTE; + IFA_F_NOPREFIXROUTE | IFA_F_MCAUTOJOIN; ifa = ipv6_get_ifaddr(net, pfx, dev, 1); if (ifa == NULL) { diff --git a/net/ipv6/mcast.c b/net/ipv6/mcast.c index e4955d0..1dd1fed 100644 --- a/net/ipv6/mcast.c +++ b/net/ipv6/mcast.c @@ -2929,20 +2929,32 @@ static int __net_init igmp6_net_init(struct net *net) inet6_sk(net->ipv6.igmp_sk)->hop_limit = 1; + err = inet_ctl_sock_create(&net->ipv6.mc_autojoin_sk, PF_INET6, + SOCK_RAW, IPPROTO_ICMPV6, net); + if (err < 0) { + pr_err("Failed to initialize the IGMP6 autojoin socket (err %d)\n", + err); + goto out_sock_create; + } + err = igmp6_proc_init(net); if (err) - goto out_sock_create; -out: - return err; + goto out_sock_create_autojoin; + + return 0; +out_sock_create_autojoin: + inet_ctl_sock_destroy(net->ipv6.mc_autojoin_sk); out_sock_create: inet_ctl_sock_destroy(net->ipv6.igmp_sk); - goto out; +out: + return err; } static void __net_exit igmp6_net_exit(struct net *net) { inet_ctl_sock_destroy(net->ipv6.igmp_sk); + inet_ctl_sock_destroy(net->ipv6.mc_autojoin_sk); igmp6_proc_exit(net); }