[PATCH] multicast: Extend ip address command to enable multicast group join/leave on IP level.

[PATCH] multicast: Extend ip address command to enable multicast group join/leave on IP level.

[Madhu Challa]

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 <challa@noironetworks.com>
---
 include/net/ip.h             |  1 +
 include/net/ipv6.h           |  2 ++
 include/net/multicast.h      | 16 +++++++++++++
 include/net/netns/ipv4.h     |  1 +
 include/net/netns/ipv6.h     |  1 +
 include/uapi/linux/if_addr.h |  1 +
 net/ipv4/devinet.c           | 11 +++++++++
 net/ipv4/igmp.c              | 57 ++++++++++++++++++++++++++++++++++++++++++++
 net/ipv6/addrconf.c          | 11 ++++++++-
 net/ipv6/mcast.c             | 39 ++++++++++++++++++++++++++++++
 10 files changed, 139 insertions(+), 1 deletion(-)
 create mode 100644 include/net/multicast.h

diff --git a/include/net/ip.h b/include/net/ip.h
index 025c61c..e759bf4 100644
--- a/include/net/ip.h
+++ b/include/net/ip.h
@@ -571,4 +571,5 @@ extern int sysctl_icmp_msgs_burst;
 int ip_misc_proc_init(void);
 #endif
 
+void ip_mc_config_async(struct sock *sk, bool join, __be32 saddr, int ifindex);
 #endif	/* _IP_H */
diff --git a/include/net/ipv6.h b/include/net/ipv6.h
index 4c9fe22..9da5537 100644
--- a/include/net/ipv6.h
+++ b/include/net/ipv6.h
@@ -940,4 +940,6 @@ int ipv6_sysctl_register(void);
 void ipv6_sysctl_unregister(void);
 #endif
 
+void ipv6_mc_config_async(struct sock *sk, bool join,
+			  const struct in6_addr *addr, int ifindex);
 #endif /* _NET_IPV6_H */
diff --git a/include/net/multicast.h b/include/net/multicast.h
new file mode 100644
index 0000000..eb0a70c
--- /dev/null
+++ b/include/net/multicast.h
@@ -0,0 +1,16 @@
+#ifndef _MULTICAST_H
+#define _MULTICAST_H
+
+struct mc_autojoin_request {
+	union {
+		struct sockaddr_in sin;
+		struct sockaddr_in6 sin6;
+	} addr;
+	int ifindex;
+	struct sock *sk;
+	struct work_struct ipv4_work;
+	struct work_struct ipv6_work;
+	bool join;
+};
+
+#endif
diff --git a/include/net/netns/ipv4.h b/include/net/netns/ipv4.h
index dbe2254..9c1f01e 100644
--- a/include/net/netns/ipv4.h
+++ b/include/net/netns/ipv4.h
@@ -47,6 +47,7 @@ struct netns_ipv4 {
 #endif
 	struct hlist_head	*fib_table_hash;
 	struct sock		*fibnl;
+	struct sock		*mc_autojoin_sock;
 
 	struct sock  * __percpu	*icmp_sk;
 
diff --git a/include/net/netns/ipv6.h b/include/net/netns/ipv6.h
index 69ae41f..fd2cef8 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_sock;
 #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 f0b4a31..86888b0 100644
--- a/net/ipv4/devinet.c
+++ b/net/ipv4/devinet.c
@@ -584,6 +584,11 @@ 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_async(net->ipv4.mc_autojoin_sock,
+					   false, ifa->ifa_address,
+					   ifa->ifa_dev->dev->ifindex);
+		}
 		__inet_del_ifa(in_dev, ifap, 1, nlh, NETLINK_CB(skb).portid);
 		return 0;
 	}
@@ -838,6 +843,12 @@ 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) {
+			WARN_ON(!ipv4_is_multicast(ifa->ifa_address));
+			ip_mc_config_async(net->ipv4.mc_autojoin_sock,
+					   true, ifa->ifa_address,
+					   ifa->ifa_dev->dev->ifindex);
+		}
 		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 666cf36..0b3a000 100644
--- a/net/ipv4/igmp.c
+++ b/net/ipv4/igmp.c
@@ -105,6 +105,7 @@
 #include <linux/proc_fs.h>
 #include <linux/seq_file.h>
 #endif
+#include <net/multicast.h>
 
 #define IP_MAX_MEMBERSHIPS	20
 #define IP_MAX_MSF		10
@@ -1976,6 +1977,45 @@ out:
 }
 EXPORT_SYMBOL(ip_mc_leave_group);
 
+static void ip_mc_auto_join(struct work_struct *work)
+{
+	struct mc_autojoin_request *req =
+		container_of(work, struct mc_autojoin_request, ipv4_work);
+	struct ip_mreqn mreq = {
+		.imr_multiaddr.s_addr = req->addr.sin.sin_addr.s_addr,
+		.imr_ifindex = req->ifindex,
+	};
+
+	lock_sock(req->sk);
+	if (req->join)
+		ip_mc_join_group(req->sk, &mreq);
+	else
+		ip_mc_leave_group(req->sk, &mreq);
+	release_sock(req->sk);
+	sock_put(req->sk);
+	kfree(req);
+}
+
+void ip_mc_config_async(struct sock *sk, bool join, __be32 saddr,
+			int ifindex)
+{
+	struct mc_autojoin_request *req;
+	ASSERT_RTNL();
+
+	req = kzalloc(sizeof(*req), GFP_KERNEL);
+	if (!req)
+		return;
+
+	sock_hold(sk);
+	req->sk = sk;
+	req->addr.sin.sin_addr.s_addr = saddr;
+	req->ifindex = ifindex;
+	req->join = join;
+	INIT_WORK(&req->ipv4_work, ip_mc_auto_join);
+	schedule_work(&req->ipv4_work);
+}
+EXPORT_SYMBOL(ip_mc_config_async);
+
 int ip_mc_source(int add, int omode, struct sock *sk, struct
 	ip_mreq_source *mreqs, int ifindex)
 {
@@ -2724,6 +2764,8 @@ 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;
+	struct socket *sock = NULL;
 
 	pde = proc_create("igmp", S_IRUGO, net->proc_net, &igmp_mc_seq_fops);
 	if (!pde)
@@ -2732,8 +2774,16 @@ static int __net_init igmp_net_init(struct net *net)
 			  &igmp_mcf_seq_fops);
 	if (!pde)
 		goto out_mcfilter;
+	err = sock_create_kern(AF_INET, SOCK_DGRAM, 0, &sock);
+	if (err < 0)
+		goto out_sock;
+	sk_change_net(sock->sk, net);
+	net->ipv4.mc_autojoin_sock = sock->sk;
+
 	return 0;
 
+out_sock:
+	remove_proc_entry("mcfilter", net->proc_net);
 out_mcfilter:
 	remove_proc_entry("igmp", net->proc_net);
 out_igmp:
@@ -2742,8 +2792,15 @@ out_igmp:
 
 static void __net_exit igmp_net_exit(struct net *net)
 {
+	struct sock *sk = net->ipv4.mc_autojoin_sock;
+
 	remove_proc_entry("mcfilter", net->proc_net);
 	remove_proc_entry("igmp", net->proc_net);
+	if (sk) {
+		kernel_sock_shutdown(sk->sk_socket, SHUT_RDWR);
+		sk_release_kernel(sk);
+		net->ipv4.mc_autojoin_sock = NULL;
+	}
 }
 
 static struct pernet_operations igmp_net_ops = {
diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c
index 98e4a63..572598b 100644
--- a/net/ipv6/addrconf.c
+++ b/net/ipv6/addrconf.c
@@ -2540,6 +2540,11 @@ static int inet6_addr_add(struct net *net, int ifindex,
 			manage_tempaddrs(idev, ifp, valid_lft, prefered_lft,
 					 true, jiffies);
 		in6_ifa_put(ifp);
+		if (ifa_flags & IFA_F_MCAUTOJOIN) {
+			WARN_ON(!ipv6_addr_is_multicast(pfx));
+			ipv6_mc_config_async(net->ipv6.mc_autojoin_sock,
+					     true, pfx, ifindex);
+		}
 		addrconf_verify_rtnl();
 		return 0;
 	}
@@ -2578,6 +2583,10 @@ 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_async(net->ipv6.mc_autojoin_sock,
+						     false, pfx, dev->ifindex);
+			}
 			return 0;
 		}
 	}
@@ -3945,7 +3954,7 @@ 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 5ce107c..00fca26 100644
--- a/net/ipv6/mcast.c
+++ b/net/ipv6/mcast.c
@@ -63,6 +63,7 @@
 #include <net/inet_common.h>
 
 #include <net/ip6_checksum.h>
+#include <net/multicast.h>
 
 /* Ensure that we have struct in6_addr aligned on 32bit word. */
 static void *__mld2_query_bugs[] __attribute__((__unused__)) = {
@@ -247,6 +248,44 @@ int ipv6_sock_mc_drop(struct sock *sk, int ifindex, const struct in6_addr *addr)
 	return -EADDRNOTAVAIL;
 }
 
+static void ipv6_mc_auto_join(struct work_struct *work)
+{
+	struct mc_autojoin_request *req =
+		container_of(work, struct mc_autojoin_request, ipv6_work);
+
+	lock_sock(req->sk);
+	if (req->join)
+		ipv6_sock_mc_join(req->sk, req->ifindex,
+				  &req->addr.sin6.sin6_addr);
+	else
+		ipv6_sock_mc_drop(req->sk, req->ifindex,
+				  &req->addr.sin6.sin6_addr);
+	release_sock(req->sk);
+	sock_put(req->sk);
+	kfree(req);
+}
+
+
+void ipv6_mc_config_async(struct sock *sk, bool join,
+			  const struct in6_addr *addr, int ifindex)
+{
+	struct mc_autojoin_request *req;
+	ASSERT_RTNL();
+
+	req = kzalloc(sizeof(*req), GFP_KERNEL);
+	if (!req)
+		return;
+
+	sock_hold(sk);
+	req->sk = sk;
+	memcpy(&req->addr.sin6.sin6_addr, addr, sizeof(*addr));
+	req->ifindex = ifindex;
+	req->join = join;
+	INIT_WORK(&req->ipv6_work, ipv6_mc_auto_join);
+	schedule_work(&req->ipv6_work);
+}
+EXPORT_SYMBOL(ipv6_mc_config_async);
+
 /* called with rcu_read_lock() */
 static struct inet6_dev *ip6_mc_find_dev_rcu(struct net *net,
 					     const struct in6_addr *group,
-- 
1.9.1

Re: [PATCH] multicast: Extend ip address command to enable multicast group join/leave on IP level.

[Eric Dumazet]

On Mon, 2015-02-16 at 14:20 -0800, Madhu Challa wrote:
> 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.
> 
...
> +struct mc_autojoin_request {
> +	union {
> +		struct sockaddr_in sin;
> +		struct sockaddr_in6 sin6;
> +	} addr;
> +	int ifindex;
> +	struct sock *sk;
> +	struct work_struct ipv4_work;
> +	struct work_struct ipv6_work;
> +	bool join;
> +};

This looks a net-next patch, right ?

You do not need 2 separate work_struct. A single one is enough, or even
no work queue at all as a matter of fact.

I'll send a preparation patch like this one, then you'll be able to call
__ip_mc_{join|leave}_group() directly while RTNL is held.

[PATCH net-next] igmp: add __ip_mc_{join|leave}_group()

There is a need to perform igmp join/leave operations while RTNL is
held. Make ip_mc_{join|leave}_group() wrappers around __ip_mc_{join|
leave}_group() to avoid the proliferation of work queues.

...

[PATCH net-next] igmp: add __ip_mc_{join|leave}_group()

[Eric Dumazet]

From: Eric Dumazet <edumazet@google.com>

There is a need to perform igmp join/leave operations while RTNL is
held.

Make ip_mc_{join|leave}_group() wrappers around 
__ip_mc_{join|leave}_group() to avoid the proliferation of work queues.

For example, vxlan_igmp_join() could possibly be removed.

Signed-off-by: Eric Dumazet <edumazet@google.com>
---
 include/linux/igmp.h |    2 +
 net/ipv4/igmp.c      |   52 ++++++++++++++++++++++++++---------------
 2 files changed, 36 insertions(+), 18 deletions(-)

diff --git a/include/linux/igmp.h b/include/linux/igmp.h
index 2c677afeea4782c96b79d0d8ede4846d87783b99..b5a6470e686c436f619ee086cbdfc3ece389db35 100644
--- a/include/linux/igmp.h
+++ b/include/linux/igmp.h
@@ -111,7 +111,9 @@ struct ip_mc_list {
 
 extern int ip_check_mc_rcu(struct in_device *dev, __be32 mc_addr, __be32 src_addr, u16 proto);
 extern int igmp_rcv(struct sk_buff *);
+extern int __ip_mc_join_group(struct sock *sk, struct ip_mreqn *imr);
 extern int ip_mc_join_group(struct sock *sk, struct ip_mreqn *imr);
+extern int __ip_mc_leave_group(struct sock *sk, struct ip_mreqn *imr);
 extern int ip_mc_leave_group(struct sock *sk, struct ip_mreqn *imr);
 extern void ip_mc_drop_socket(struct sock *sk);
 extern int ip_mc_source(int add, int omode, struct sock *sk,
diff --git a/net/ipv4/igmp.c b/net/ipv4/igmp.c
index 666cf364df86dffbab3cd2b7b1ea66e72d3ac7a8..4b1172d73e0309766d6d1a70473dd4c14d2c6fc0 100644
--- a/net/ipv4/igmp.c
+++ b/net/ipv4/igmp.c
@@ -1849,30 +1849,25 @@ static void ip_mc_clear_src(struct ip_mc_list *pmc)
 	pmc->sfcount[MCAST_EXCLUDE] = 1;
 }
 
-
-/*
- * Join a multicast group
- */
-int ip_mc_join_group(struct sock *sk , struct ip_mreqn *imr)
+int __ip_mc_join_group(struct sock *sk, struct ip_mreqn *imr)
 {
-	int err;
 	__be32 addr = imr->imr_multiaddr.s_addr;
-	struct ip_mc_socklist *iml = NULL, *i;
+	struct ip_mc_socklist *iml, *i;
 	struct in_device *in_dev;
 	struct inet_sock *inet = inet_sk(sk);
 	struct net *net = sock_net(sk);
 	int ifindex;
 	int count = 0;
+	int err;
+
+	ASSERT_RTNL();
 
 	if (!ipv4_is_multicast(addr))
 		return -EINVAL;
 
-	rtnl_lock();
-
 	in_dev = ip_mc_find_dev(net, imr);
 
 	if (!in_dev) {
-		iml = NULL;
 		err = -ENODEV;
 		goto done;
 	}
@@ -1900,9 +1895,22 @@ int ip_mc_join_group(struct sock *sk , struct ip_mreqn *imr)
 	ip_mc_inc_group(in_dev, addr);
 	err = 0;
 done:
-	rtnl_unlock();
 	return err;
 }
+EXPORT_SYMBOL(__ip_mc_join_group);
+
+/* Join a multicast group
+ */
+int ip_mc_join_group(struct sock *sk, struct ip_mreqn *imr)
+{
+	int ret;
+
+	rtnl_lock();
+	ret = __ip_mc_join_group(sk, imr);
+	rtnl_unlock();
+
+	return ret;
+}
 EXPORT_SYMBOL(ip_mc_join_group);
 
 static int ip_mc_leave_src(struct sock *sk, struct ip_mc_socklist *iml,
@@ -1925,11 +1933,7 @@ static int ip_mc_leave_src(struct sock *sk, struct ip_mc_socklist *iml,
 	return err;
 }
 
-/*
- *	Ask a socket to leave a group.
- */
-
-int ip_mc_leave_group(struct sock *sk, struct ip_mreqn *imr)
+int __ip_mc_leave_group(struct sock *sk, struct ip_mreqn *imr)
 {
 	struct inet_sock *inet = inet_sk(sk);
 	struct ip_mc_socklist *iml;
@@ -1940,7 +1944,8 @@ int ip_mc_leave_group(struct sock *sk, struct ip_mreqn *imr)
 	u32 ifindex;
 	int ret = -EADDRNOTAVAIL;
 
-	rtnl_lock();
+	ASSERT_RTNL();
+
 	in_dev = ip_mc_find_dev(net, imr);
 	if (!in_dev) {
 		ret = -ENODEV;
@@ -1964,14 +1969,25 @@ int ip_mc_leave_group(struct sock *sk, struct ip_mreqn *imr)
 		*imlp = iml->next_rcu;
 
 		ip_mc_dec_group(in_dev, group);
-		rtnl_unlock();
+
 		/* decrease mem now to avoid the memleak warning */
 		atomic_sub(sizeof(*iml), &sk->sk_omem_alloc);
 		kfree_rcu(iml, rcu);
 		return 0;
 	}
 out:
+	return ret;
+}
+EXPORT_SYMBOL(__ip_mc_leave_group);
+
+int ip_mc_leave_group(struct sock *sk, struct ip_mreqn *imr)
+{
+	int ret;
+
+	rtnl_lock();
+	ret = __ip_mc_leave_group(sk, imr);
 	rtnl_unlock();
+
 	return ret;
 }
 EXPORT_SYMBOL(ip_mc_leave_group);

Re: [PATCH] multicast: Extend ip address command to enable multicast group join/leave on IP level.

[Madhu Challa]

I realized we will have to do the same refactoring in ipv6 code as
well. I was also missing the mc_autojoin_sock initialization in ipv6
code that I just added. I will be happy to do the refactoring if you
want me to.

Thanks.

On Tue, Feb 17, 2015 at 8:08 AM, Madhu Challa <challa@noironetworks.com> wrote:
>
>
> On Tue, Feb 17, 2015 at 3:10 AM, Eric Dumazet <eric.dumazet@gmail.com>
> wrote:
>>
>> On Mon, 2015-02-16 at 14:20 -0800, Madhu Challa wrote:
>> > 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.
>> >
>> ...
>> > +struct mc_autojoin_request {
>> > +     union {
>> > +             struct sockaddr_in sin;
>> > +             struct sockaddr_in6 sin6;
>> > +     } addr;
>> > +     int ifindex;
>> > +     struct sock *sk;
>> > +     struct work_struct ipv4_work;
>> > +     struct work_struct ipv6_work;
>> > +     bool join;
>> > +};
>>
>> This looks a net-next patch, right ?
>
>
> yes.
>>
>>
>> You do not need 2 separate work_struct. A single one is enough, or even
>> no work queue at all as a matter of fact.
>>
>> I'll send a preparation patch like this one, then you'll be able to call
>> __ip_mc_{join|leave}_group() directly while RTNL is held.
>>
>> [PATCH net-next] igmp: add __ip_mc_{join|leave}_group()
>>
>> There is a need to perform igmp join/leave operations while RTNL is
>> held. Make ip_mc_{join|leave}_group() wrappers around __ip_mc_{join|
>> leave}_group() to avoid the proliferation of work queues.
>
>
> Thanks. I will update my patch to use the new apis.
>
> Thanks.
>>
>>
>>
>>
>> ...
>>
>>
>

Re: [PATCH net-next] igmp: add __ip_mc_{join|leave}_group()

[David Miller]

From: Eric Dumazet <eric.dumazet@gmail.com>
Date: Tue, 17 Feb 2015 03:19:24 -0800

> From: Eric Dumazet <edumazet@google.com>
> 
> There is a need to perform igmp join/leave operations while RTNL is
> held.
> 
> Make ip_mc_{join|leave}_group() wrappers around 
> __ip_mc_{join|leave}_group() to avoid the proliferation of work queues.
> 
> For example, vxlan_igmp_join() could possibly be removed.
> 
> Signed-off-by: Eric Dumazet <edumazet@google.com>

Applied, thanks Eric.