net: ipv6: autoconf routes into per-device tables

Currently, IPv6 router discovery always puts routes into
RT6_TABLE_MAIN. This causes problems for connection managers
that want to support multiple simultaneous network connections
and want control over which one is used by default (e.g., wifi
and wired).

To work around this connection managers typically take the routes
they prefer and copy them to static routes with low metrics in
the main table. This puts the burden on the connection manager
to watch netlink to see if the routes have changed, delete the
routes when their lifetime expires, etc.

Instead, this patch adds a per-interface sysctl to have the
kernel put autoconf routes into different tables. This allows
each interface to have its own autoconf table, and choosing the
default interface (or using different interfaces at the same
time for different types of traffic) can be done using
appropriate ip rules.

The sysctl behaves as follows:

- = 0: default. Put routes into RT6_TABLE_MAIN as before.
- > 0: manual. Put routes into the specified table.
- < 0: automatic. Add the absolute value of the sysctl to the
       device's ifindex, and use that table.

The automatic mode is most useful in conjunction with
net.ipv6.conf.default.accept_ra_rt_table. A connection manager
or distribution could set it to, say, -100 on boot, and
thereafter just use IP rules.

Change-Id: I82d16e3737d9cdfa6489e649e247894d0d60cbb1
Signed-off-by: Lorenzo Colitti <lorenzo@google.com>

5 files changed, 71 insertions, 45 deletions

diff --git a/include/linux/ipv6.h b/include/linux/ipv6.h
index ff560537dd61..36549979c6b8 100644
--- a/include/linux/ipv6.h
+++ b/include/linux/ipv6.h
@@ -37,6 +37,7 @@ struct ipv6_devconf {
 	__s32		accept_ra_rt_info_max_plen;
 #endif
 #endif
+	__s32		accept_ra_rt_table;
 	__s32		proxy_ndp;
 	__s32		accept_source_route;
 	__s32		accept_ra_from_local;
diff --git a/include/net/addrconf.h b/include/net/addrconf.h
index d13573bb879e..9828539ae701 100644
--- a/include/net/addrconf.h
+++ b/include/net/addrconf.h
@@ -193,6 +193,8 @@ static inline bool ipv6_is_mld(struct sk_buff *skb, int nexthdr, int offset)
 void addrconf_prefix_rcv(struct net_device *dev,
 			 u8 *opt, int len, bool sllao);
 
+u32 addrconf_rt_table(const struct net_device *dev, u32 default_table);
+
 /*
  *	anycast prototypes (anycast.c)
  */
diff --git a/include/uapi/linux/ipv6.h b/include/uapi/linux/ipv6.h
index efa2666f4b8a..22bd2111d3a8 100644
--- a/include/uapi/linux/ipv6.h
+++ b/include/uapi/linux/ipv6.h
@@ -160,6 +160,7 @@ enum {
 	DEVCONF_ACCEPT_DAD,
 	DEVCONF_FORCE_TLLAO,
 	DEVCONF_NDISC_NOTIFY,
+	DEVCONF_ACCEPT_RA_RT_TABLE,
 	DEVCONF_MLDV1_UNSOLICITED_REPORT_INTERVAL,
 	DEVCONF_MLDV2_UNSOLICITED_REPORT_INTERVAL,
 	DEVCONF_SUPPRESS_FRAG_NDISC,
diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c
index 0169ccf5aa4f..3990a0e08e24 100644
--- a/net/ipv6/addrconf.c
+++ b/net/ipv6/addrconf.c
@@ -196,6 +196,7 @@ static struct ipv6_devconf ipv6_devconf __read_mostly = {
 	.accept_ra_rt_info_max_plen = 0,
 #endif
 #endif
+	.accept_ra_rt_table	= 0,
 	.proxy_ndp		= 0,
 	.accept_source_route	= 0,	/* we do not accept RH0 by default. */
 	.disable_ipv6		= 0,
@@ -233,6 +234,7 @@ static struct ipv6_devconf ipv6_devconf_dflt __read_mostly = {
 	.accept_ra_rt_info_max_plen = 0,
 #endif
 #endif
+	.accept_ra_rt_table	= 0,
 	.proxy_ndp		= 0,
 	.accept_source_route	= 0,	/* we do not accept RH0 by default. */
 	.disable_ipv6		= 0,
@@ -1966,6 +1968,31 @@ static void  __ipv6_try_regen_rndid(struct inet6_dev *idev, struct in6_addr *tmp
 		__ipv6_regen_rndid(idev);
 }
 
+u32 addrconf_rt_table(const struct net_device *dev, u32 default_table) {
+	/* Determines into what table to put autoconf PIO/RIO/default routes
+	 * learned on this device.
+	 *
+	 * - If 0, use the same table for every device. This puts routes into
+	 *   one of RT_TABLE_{PREFIX,INFO,DFLT} depending on the type of route
+	 *   (but note that these three are currently all equal to
+	 *   RT6_TABLE_MAIN).
+	 * - If > 0, use the specified table.
+	 * - If < 0, put routes into table dev->ifindex + (-rt_table).
+	 */
+	struct inet6_dev *idev = in6_dev_get(dev);
+	u32 table;
+	int sysctl = idev->cnf.accept_ra_rt_table;
+	if (sysctl == 0) {
+		table = default_table;
+	} else if (sysctl > 0) {
+		table = (u32) sysctl;
+	} else {
+		table = (unsigned) dev->ifindex + (-sysctl);
+	}
+	in6_dev_put(idev);
+	return table;
+}
+
 /*
  *	Add prefix route.
  */
@@ -1975,7 +2002,7 @@ addrconf_prefix_route(struct in6_addr *pfx, int plen, struct net_device *dev,
 		      unsigned long expires, u32 flags)
 {
 	struct fib6_config cfg = {
-		.fc_table = RT6_TABLE_PREFIX,
+		.fc_table = addrconf_rt_table(dev, RT6_TABLE_PREFIX),
 		.fc_metric = IP6_RT_PRIO_ADDRCONF,
 		.fc_ifindex = dev->ifindex,
 		.fc_expires = expires,
@@ -2009,7 +2036,8 @@ static struct rt6_info *addrconf_get_prefix_route(const struct in6_addr *pfx,
 	struct rt6_info *rt = NULL;
 	struct fib6_table *table;
 
-	table = fib6_get_table(dev_net(dev), RT6_TABLE_PREFIX);
+	table = fib6_get_table(dev_net(dev),
+			       addrconf_rt_table(dev, RT6_TABLE_PREFIX));
 	if (table == NULL)
 		return NULL;
 
@@ -4326,6 +4354,7 @@ static inline void ipv6_store_devconf(struct ipv6_devconf *cnf,
 	array[DEVCONF_ACCEPT_RA_RT_INFO_MAX_PLEN] = cnf->accept_ra_rt_info_max_plen;
 #endif
 #endif
+	array[DEVCONF_ACCEPT_RA_RT_TABLE] = cnf->accept_ra_rt_table;
 	array[DEVCONF_PROXY_NDP] = cnf->proxy_ndp;
 	array[DEVCONF_ACCEPT_SOURCE_ROUTE] = cnf->accept_source_route;
 #ifdef CONFIG_IPV6_OPTIMISTIC_DAD
@@ -5134,6 +5163,13 @@ static struct addrconf_sysctl_table
 #endif
 #endif
 		{
+			.procname	= "accept_ra_rt_table",
+			.data		= &ipv6_devconf.accept_ra_rt_table,
+			.maxlen		= sizeof(int),
+			.mode		= 0644,
+			.proc_handler	= proc_dointvec,
+		},
+		{
 			.procname	= "proxy_ndp",
 			.data		= &ipv6_devconf.proxy_ndp,
 			.maxlen		= sizeof(int),
diff --git a/net/ipv6/route.c b/net/ipv6/route.c
index a318dd89b6d9..2e173c12fd18 100644
--- a/net/ipv6/route.c
+++ b/net/ipv6/route.c
@@ -95,13 +95,12 @@ static void		rt6_do_redirect(struct dst_entry *dst, struct sock *sk,
 static int rt6_score_route(struct rt6_info *rt, int oif, int strict);
 
 #ifdef CONFIG_IPV6_ROUTE_INFO
-static struct rt6_info *rt6_add_route_info(struct net *net,
+static struct rt6_info *rt6_add_route_info(struct net_device *dev,
 					   const struct in6_addr *prefix, int prefixlen,
-					   const struct in6_addr *gwaddr, int ifindex,
-					   unsigned int pref);
-static struct rt6_info *rt6_get_route_info(struct net *net,
+					   const struct in6_addr *gwaddr, unsigned int pref);
+static struct rt6_info *rt6_get_route_info(struct net_device *dev,
 					   const struct in6_addr *prefix, int prefixlen,
-					   const struct in6_addr *gwaddr, int ifindex);
+					   const struct in6_addr *gwaddr);
 #endif
 
 static void rt6_bind_peer(struct rt6_info *rt, int create)
@@ -700,7 +699,6 @@ static struct rt6_info *rt6_select(struct fib6_node *fn, int oif, int strict)
 int rt6_route_rcv(struct net_device *dev, u8 *opt, int len,
 		  const struct in6_addr *gwaddr)
 {
-	struct net *net = dev_net(dev);
 	struct route_info *rinfo = (struct route_info *) opt;
 	struct in6_addr prefix_buf, *prefix;
 	unsigned int pref;
@@ -745,8 +743,7 @@ int rt6_route_rcv(struct net_device *dev, u8 *opt, int len,
 	if (rinfo->prefix_len == 0)
 		rt = rt6_get_dflt_router(gwaddr, dev);
 	else
-		rt = rt6_get_route_info(net, prefix, rinfo->prefix_len,
-					gwaddr, dev->ifindex);
+		rt = rt6_get_route_info(net, prefix, rinfo->prefix_len,	gwaddr);
 
 	if (rt && !lifetime) {
 		ip6_del_rt(rt);
@@ -754,8 +751,7 @@ int rt6_route_rcv(struct net_device *dev, u8 *opt, int len,
 	}
 
 	if (!rt && lifetime)
-		rt = rt6_add_route_info(net, prefix, rinfo->prefix_len, gwaddr, dev->ifindex,
-					pref);
+		rt = rt6_add_route_info(dev, prefix, rinfo->prefix_len, gwaddr, pref);
 	else if (rt)
 		rt->rt6i_flags = RTF_ROUTEINFO |
 				 (rt->rt6i_flags & ~RTF_PREF_MASK) | RTF_PREF(pref);
@@ -1905,15 +1901,16 @@ static struct rt6_info *ip6_rt_copy(struct rt6_info *ort,
 }
 
 #ifdef CONFIG_IPV6_ROUTE_INFO
-static struct rt6_info *rt6_get_route_info(struct net *net,
+static struct rt6_info *rt6_get_route_info(struct net_device *dev,
 					   const struct in6_addr *prefix, int prefixlen,
-					   const struct in6_addr *gwaddr, int ifindex)
+					   const struct in6_addr *gwaddr)
 {
 	struct fib6_node *fn;
 	struct rt6_info *rt = NULL;
 	struct fib6_table *table;
 
-	table = fib6_get_table(net, RT6_TABLE_INFO);
+	table = fib6_get_table(dev_net(dev),
+			       addrconf_rt_table(dev, RT6_TABLE_INFO));
 	if (!table)
 		return NULL;
 
@@ -1923,7 +1920,7 @@ static struct rt6_info *rt6_get_route_info(struct net *net,
 		goto out;
 
 	for (rt = fn->leaf; rt; rt = rt->dst.rt6_next) {
-		if (rt->dst.dev->ifindex != ifindex)
+		if (rt->dst.dev->ifindex != dev->ifindex)
 			continue;
 		if ((rt->rt6i_flags & (RTF_ROUTEINFO|RTF_GATEWAY)) != (RTF_ROUTEINFO|RTF_GATEWAY))
 			continue;
@@ -1937,21 +1934,20 @@ out:
 	return rt;
 }
 
-static struct rt6_info *rt6_add_route_info(struct net *net,
+static struct rt6_info *rt6_add_route_info(struct net_device *dev,
 					   const struct in6_addr *prefix, int prefixlen,
-					   const struct in6_addr *gwaddr, int ifindex,
-					   unsigned int pref)
+					   const struct in6_addr *gwaddr, unsigned int pref)
 {
 	struct fib6_config cfg = {
-		.fc_table	= RT6_TABLE_INFO,
+		.fc_table	= addrconf_rt_table(dev, RT6_TABLE_INFO),
 		.fc_metric	= IP6_RT_PRIO_USER,
-		.fc_ifindex	= ifindex,
+		.fc_ifindex	= dev->ifindex,
 		.fc_dst_len	= prefixlen,
 		.fc_flags	= RTF_GATEWAY | RTF_ADDRCONF | RTF_ROUTEINFO |
 				  RTF_UP | RTF_PREF(pref),
 		.fc_nlinfo.portid = 0,
 		.fc_nlinfo.nlh = NULL,
-		.fc_nlinfo.nl_net = net,
+		.fc_nlinfo.nl_net = dev_net(dev),
 	};
 
 	cfg.fc_dst = *prefix;
@@ -1963,7 +1959,7 @@ static struct rt6_info *rt6_add_route_info(struct net *net,
 
 	ip6_route_add(&cfg);
 
-	return rt6_get_route_info(net, prefix, prefixlen, gwaddr, ifindex);
+	return rt6_get_route_info(dev, prefix, prefixlen, gwaddr);
 }
 #endif
 
@@ -1972,7 +1968,8 @@ struct rt6_info *rt6_get_dflt_router(const struct in6_addr *addr, struct net_dev
 	struct rt6_info *rt;
 	struct fib6_table *table;
 
-	table = fib6_get_table(dev_net(dev), RT6_TABLE_DFLT);
+	table = fib6_get_table(dev_net(dev),
+			       addrconf_rt_table(dev, RT6_TABLE_MAIN));
 	if (!table)
 		return NULL;
 
@@ -1994,7 +1991,7 @@ struct rt6_info *rt6_add_dflt_router(const struct in6_addr *gwaddr,
 				     unsigned int pref)
 {
 	struct fib6_config cfg = {
-		.fc_table	= RT6_TABLE_DFLT,
+		.fc_table	= addrconf_rt_table(dev, RT6_TABLE_DFLT),
 		.fc_metric	= IP6_RT_PRIO_USER,
 		.fc_ifindex	= dev->ifindex,
 		.fc_flags	= RTF_GATEWAY | RTF_ADDRCONF | RTF_DEFAULT |
@@ -2011,28 +2008,17 @@ struct rt6_info *rt6_add_dflt_router(const struct in6_addr *gwaddr,
 	return rt6_get_dflt_router(gwaddr, dev);
 }
 
-void rt6_purge_dflt_routers(struct net *net)
-{
-	struct rt6_info *rt;
-	struct fib6_table *table;
 
-	/* NOTE: Keep consistent with rt6_get_dflt_router */
-	table = fib6_get_table(net, RT6_TABLE_DFLT);
-	if (!table)
-		return;
+int rt6_addrconf_purge(struct rt6_info *rt, void *arg) {
+	if (rt->rt6i_flags & (RTF_DEFAULT | RTF_ADDRCONF) &&
+	    (!rt->rt6i_idev || rt->rt6i_idev->cnf.accept_ra != 2))
+		return -1;
+	return 0;
+}
 
-restart:
-	read_lock_bh(&table->tb6_lock);
-	for (rt = table->tb6_root.leaf; rt; rt = rt->dst.rt6_next) {
-		if (rt->rt6i_flags & (RTF_DEFAULT | RTF_ADDRCONF) &&
-		    (!rt->rt6i_idev || rt->rt6i_idev->cnf.accept_ra != 2)) {
-			dst_hold(&rt->dst);
-			read_unlock_bh(&table->tb6_lock);
-			ip6_del_rt(rt);
-			goto restart;
-		}
-	}
-	read_unlock_bh(&table->tb6_lock);
+void rt6_purge_dflt_routers(struct net *net)
+{
+	fib6_clean_all(net, rt6_addrconf_purge, NULL);
 }
 
 static void rtmsg_to_fib6_config(struct net *net,