Merge branch 'inet_fwmark_reflect'

Lorenzo Colitti says:

====================
Make mark-based routing work better with multiple separate networks.

Mark-based routing (ip rule fwmark 17 lookup 100) combined with
either iptables marking (iptables -j MARK --set-mark 17) or
application-based marking (the SO_MARK setsockopt) are a good
way to deal with connecting simultaneously to multiple networks.

Each network can be given a routing table, and ip rules can
be configured to make different fwmarks select different
networks. Applications can select networks them by setting
appropriate socket marks, and iptables rules can be used to
handle non-aware applications, enforce policy, etc.

This patch series improves functionality when mark-based routing
is used in this way. Current behaviour has the following
limitations:

1. Kernel-originated replies that are not associated with a
   socket always use a mark of zero. This means that, for
   example, when the kernel sends a ping reply or a TCP reset,
   it does not send it on the network from which it received the
   original packet.
2. Path MTU discovery, which is triggered by incoming packets,
   does not always work correctly, because the routing lookups it
   uses to clone routes do not take the fwmark into account and
   thus can happen in the wrong routing table.
3. Application-based marking works well for outbound connections,
   but does not work well for incoming connections. Marking a
   listening socket causes that socket to only accept
   connections from a given network, and sockets that are
   returned by accept() are not marked (and are thus not routed
   correctly).

sysctl. This causes route lookups for kernel-generated replies
and PMTUD to use the fwmark of the packet that caused them.

which causes TCP sockets returned by accept() to be marked with
the same mark that sent the intial SYN packet.
====================

Signed-off-by: David S. Miller <davem@davemloft.net>

18 files changed, 79 insertions, 9 deletions

diff --git a/include/net/inet_sock.h b/include/net/inet_sock.h
index 1833c3f389ee..b1edf17bec01 100644
--- a/include/net/inet_sock.h
+++ b/include/net/inet_sock.h
@@ -90,6 +90,7 @@ struct inet_request_sock {
         kmemcheck_bitfield_end(flags);
         struct ip_options_rcu   *opt;
         struct sk_buff          *pktopts;
+        u32                     ir_mark;
 };
 
 static inline struct inet_request_sock *inet_rsk(const struct request_sock *sk)
@@ -97,6 +98,15 @@ static inline struct inet_request_sock *inet_rsk(const struct request_sock *sk)
         return (struct inet_request_sock *)sk;
 }
 
+static inline u32 inet_request_mark(struct sock *sk, struct sk_buff *skb)
+{
+        if (!sk->sk_mark && sock_net(sk)->ipv4.sysctl_tcp_fwmark_accept) {
+                return skb->mark;
+        } else {
+                return sk->sk_mark;
+        }
+}
+
 struct inet_cork {
         unsigned int            flags;
         __be32                  addr;
diff --git a/include/net/ip.h b/include/net/ip.h
index 55752985c144..14c50a1650ef 100644
--- a/include/net/ip.h
+++ b/include/net/ip.h
@@ -231,6 +231,9 @@ void ipfrag_init(void);
 
 void ip_static_sysctl_init(void);
 
+#define IP4_REPLY_MARK(net, mark) \
+        ((net)->ipv4.sysctl_fwmark_reflect ? (mark) : 0)
+
 static inline bool ip_is_fragment(const struct iphdr *iph)
 {
         return (iph->frag_off & htons(IP_MF | IP_OFFSET)) != 0;
diff --git a/include/net/ipv6.h b/include/net/ipv6.h
index 5b40ad297b8c..ba810d0546bc 100644
--- a/include/net/ipv6.h
+++ b/include/net/ipv6.h
@@ -113,6 +113,9 @@ struct frag_hdr {
 #define IP6_MF          0x0001
 #define IP6_OFFSET      0xFFF8
 
+#define IP6_REPLY_MARK(net, mark) \
+        ((net)->ipv6.sysctl.fwmark_reflect ? (mark) : 0)
+
 #include <net/sock.h>
 
 /* sysctls */
diff --git a/include/net/netns/ipv4.h b/include/net/netns/ipv4.h
index b2704fd0ec80..2f0cfad66666 100644
--- a/include/net/netns/ipv4.h
+++ b/include/net/netns/ipv4.h
@@ -77,6 +77,9 @@ struct netns_ipv4 {
         int sysctl_ip_no_pmtu_disc;
         int sysctl_ip_fwd_use_pmtu;
 
+        int sysctl_fwmark_reflect;
+        int sysctl_tcp_fwmark_accept;
+
         struct ping_group_range ping_group_range;
 
         atomic_t dev_addr_genid;
diff --git a/include/net/netns/ipv6.h b/include/net/netns/ipv6.h
index 21edaf1f7916..19d3446e59d2 100644
--- a/include/net/netns/ipv6.h
+++ b/include/net/netns/ipv6.h
@@ -30,6 +30,7 @@ struct netns_sysctl_ipv6 {
         int flowlabel_consistency;
         int icmpv6_time;
         int anycast_src_echo_reply;
+        int fwmark_reflect;
 };
 
 struct netns_ipv6 {
diff --git a/net/ipv4/icmp.c b/net/ipv4/icmp.c
index fe52666dc43c..79c3d947a481 100644
--- a/net/ipv4/icmp.c
+++ b/net/ipv4/icmp.c
@@ -337,6 +337,7 @@ static void icmp_reply(struct icmp_bxm *icmp_param, struct sk_buff *skb)
         struct sock *sk;
         struct inet_sock *inet;
         __be32 daddr, saddr;
+        u32 mark = IP4_REPLY_MARK(net, skb->mark);
 
         if (ip_options_echo(&icmp_param->replyopts.opt.opt, skb))
                 return;
@@ -349,6 +350,7 @@ static void icmp_reply(struct icmp_bxm *icmp_param, struct sk_buff *skb)
         icmp_param->data.icmph.checksum = 0;
 
         inet->tos = ip_hdr(skb)->tos;
+        sk->sk_mark = mark;
         daddr = ipc.addr = ip_hdr(skb)->saddr;
         saddr = fib_compute_spec_dst(skb);
         ipc.opt = NULL;
@@ -364,6 +366,7 @@ static void icmp_reply(struct icmp_bxm *icmp_param, struct sk_buff *skb)
         memset(&fl4, 0, sizeof(fl4));
         fl4.daddr = daddr;
         fl4.saddr = saddr;
+        fl4.flowi4_mark = mark;
         fl4.flowi4_tos = RT_TOS(ip_hdr(skb)->tos);
         fl4.flowi4_proto = IPPROTO_ICMP;
         security_skb_classify_flow(skb, flowi4_to_flowi(&fl4));
@@ -382,7 +385,7 @@ static struct rtable *icmp_route_lookup(struct net *net,
                                         struct flowi4 *fl4,
                                         struct sk_buff *skb_in,
                                         const struct iphdr *iph,
-                                        __be32 saddr, u8 tos,
+                                        __be32 saddr, u8 tos, u32 mark,
                                         int type, int code,
                                         struct icmp_bxm *param)
 {
@@ -394,6 +397,7 @@ static struct rtable *icmp_route_lookup(struct net *net,
         fl4->daddr = (param->replyopts.opt.opt.srr ?
                       param->replyopts.opt.opt.faddr : iph->saddr);
         fl4->saddr = saddr;
+        fl4->flowi4_mark = mark;
         fl4->flowi4_tos = RT_TOS(tos);
         fl4->flowi4_proto = IPPROTO_ICMP;
         fl4->fl4_icmp_type = type;
@@ -491,6 +495,7 @@ void icmp_send(struct sk_buff *skb_in, int type, int code, __be32 info)
         struct flowi4 fl4;
         __be32 saddr;
         u8  tos;
+        u32 mark;
         struct net *net;
         struct sock *sk;
 
@@ -592,6 +597,7 @@ void icmp_send(struct sk_buff *skb_in, int type, int code, __be32 info)
         tos = icmp_pointers[type].error ? ((iph->tos & IPTOS_TOS_MASK) |
                                            IPTOS_PREC_INTERNETCONTROL) :
                                           iph->tos;
+        mark = IP4_REPLY_MARK(net, skb_in->mark);
 
         if (ip_options_echo(&icmp_param->replyopts.opt.opt, skb_in))
                 goto out_unlock;
@@ -608,13 +614,14 @@ void icmp_send(struct sk_buff *skb_in, int type, int code, __be32 info)
         icmp_param->skb   = skb_in;
         icmp_param->offset = skb_network_offset(skb_in);
         inet_sk(sk)->tos = tos;
+        sk->sk_mark = mark;
         ipc.addr = iph->saddr;
         ipc.opt = &icmp_param->replyopts.opt;
         ipc.tx_flags = 0;
         ipc.ttl = 0;
         ipc.tos = -1;
 
-        rt = icmp_route_lookup(net, &fl4, skb_in, iph, saddr, tos,
+        rt = icmp_route_lookup(net, &fl4, skb_in, iph, saddr, tos, mark,
                                type, code, icmp_param);
         if (IS_ERR(rt))
                 goto out_unlock;
diff --git a/net/ipv4/inet_connection_sock.c b/net/ipv4/inet_connection_sock.c
index a56b8e6e866a..12e502cbfdc7 100644
--- a/net/ipv4/inet_connection_sock.c
+++ b/net/ipv4/inet_connection_sock.c
@@ -408,7 +408,7 @@ struct dst_entry *inet_csk_route_req(struct sock *sk,
         struct net *net = sock_net(sk);
         int flags = inet_sk_flowi_flags(sk);
 
-        flowi4_init_output(fl4, sk->sk_bound_dev_if, sk->sk_mark,
+        flowi4_init_output(fl4, sk->sk_bound_dev_if, ireq->ir_mark,
                            RT_CONN_FLAGS(sk), RT_SCOPE_UNIVERSE,
                            sk->sk_protocol,
                            flags,
@@ -445,7 +445,7 @@ struct dst_entry *inet_csk_route_child_sock(struct sock *sk,
 
         rcu_read_lock();
         opt = rcu_dereference(newinet->inet_opt);
-        flowi4_init_output(fl4, sk->sk_bound_dev_if, sk->sk_mark,
+        flowi4_init_output(fl4, sk->sk_bound_dev_if, inet_rsk(req)->ir_mark,
                            RT_CONN_FLAGS(sk), RT_SCOPE_UNIVERSE,
                            sk->sk_protocol, inet_sk_flowi_flags(sk),
                            (opt && opt->opt.srr) ? opt->opt.faddr : ireq->ir_rmt_addr,
@@ -680,6 +680,8 @@ struct sock *inet_csk_clone_lock(const struct sock *sk,
                 inet_sk(newsk)->inet_sport = htons(inet_rsk(req)->ir_num);
                 newsk->sk_write_space = sk_stream_write_space;
 
+                newsk->sk_mark = inet_rsk(req)->ir_mark;
+
                 newicsk->icsk_retransmits = 0;
                 newicsk->icsk_backoff     = 0;
                 newicsk->icsk_probes_out  = 0;
diff --git a/net/ipv4/ip_output.c b/net/ipv4/ip_output.c
index 6aa4380fde1a..6e231ab58d65 100644
--- a/net/ipv4/ip_output.c
+++ b/net/ipv4/ip_output.c
@@ -1546,7 +1546,8 @@ void ip_send_unicast_reply(struct net *net, struct sk_buff *skb, __be32 daddr,
                         daddr = replyopts.opt.opt.faddr;
         }
 
-        flowi4_init_output(&fl4, arg->bound_dev_if, 0,
+        flowi4_init_output(&fl4, arg->bound_dev_if,
+                           IP4_REPLY_MARK(net, skb->mark),
                            RT_TOS(arg->tos),
                            RT_SCOPE_UNIVERSE, ip_hdr(skb)->protocol,
                            ip_reply_arg_flowi_flags(arg),
diff --git a/net/ipv4/route.c b/net/ipv4/route.c
index db1e0da871f4..50e1e0feddfc 100644
--- a/net/ipv4/route.c
+++ b/net/ipv4/route.c
@@ -993,6 +993,9 @@ void ipv4_update_pmtu(struct sk_buff *skb, struct net *net, u32 mtu,
         struct flowi4 fl4;
         struct rtable *rt;
 
+        if (!mark)
+                mark = IP4_REPLY_MARK(net, skb->mark);
+
         __build_flow_key(&fl4, NULL, iph, oif,
                          RT_TOS(iph->tos), protocol, mark, flow_flags);
         rt = __ip_route_output_key(net, &fl4);
@@ -1010,6 +1013,10 @@ static void __ipv4_sk_update_pmtu(struct sk_buff *skb, struct sock *sk, u32 mtu)
         struct rtable *rt;
 
         __build_flow_key(&fl4, sk, iph, 0, 0, 0, 0, 0);
+
+        if (!fl4.flowi4_mark)
+                fl4.flowi4_mark = IP4_REPLY_MARK(sock_net(sk), skb->mark);
+
         rt = __ip_route_output_key(sock_net(sk), &fl4);
         if (!IS_ERR(rt)) {
                 __ip_rt_update_pmtu(rt, &fl4, mtu);
diff --git a/net/ipv4/syncookies.c b/net/ipv4/syncookies.c
index f2ed13c2125f..c86624b36a62 100644
--- a/net/ipv4/syncookies.c
+++ b/net/ipv4/syncookies.c
@@ -303,6 +303,7 @@ struct sock *cookie_v4_check(struct sock *sk, struct sk_buff *skb,
         ireq->ir_rmt_port       = th->source;
         ireq->ir_loc_addr       = ip_hdr(skb)->daddr;
         ireq->ir_rmt_addr       = ip_hdr(skb)->saddr;
+        ireq->ir_mark           = inet_request_mark(sk, skb);
         ireq->ecn_ok            = ecn_ok;
         ireq->snd_wscale        = tcp_opt.snd_wscale;
         ireq->sack_ok           = tcp_opt.sack_ok;
@@ -339,7 +340,7 @@ struct sock *cookie_v4_check(struct sock *sk, struct sk_buff *skb,
          * hasn't changed since we received the original syn, but I see
          * no easy way to do this.
          */
-        flowi4_init_output(&fl4, sk->sk_bound_dev_if, sk->sk_mark,
+        flowi4_init_output(&fl4, sk->sk_bound_dev_if, ireq->ir_mark,
                            RT_CONN_FLAGS(sk), RT_SCOPE_UNIVERSE, IPPROTO_TCP,
                            inet_sk_flowi_flags(sk),
                            (opt && opt->srr) ? opt->faddr : ireq->ir_rmt_addr,
diff --git a/net/ipv4/sysctl_net_ipv4.c b/net/ipv4/sysctl_net_ipv4.c
index 5cde8f263d40..a33b9fbc1d80 100644
--- a/net/ipv4/sysctl_net_ipv4.c
+++ b/net/ipv4/sysctl_net_ipv4.c
@@ -838,6 +838,20 @@ static struct ctl_table ipv4_net_table[] = {
                 .mode           = 0644,
                 .proc_handler   = proc_dointvec,
         },
+        {
+                .procname       = "fwmark_reflect",
+                .data           = &init_net.ipv4.sysctl_fwmark_reflect,
+                .maxlen         = sizeof(int),
+                .mode           = 0644,
+                .proc_handler   = proc_dointvec,
+        },
+        {
+                .procname       = "tcp_fwmark_accept",
+                .data           = &init_net.ipv4.sysctl_tcp_fwmark_accept,
+                .maxlen         = sizeof(int),
+                .mode           = 0644,
+                .proc_handler   = proc_dointvec,
+        },
         { }
 };
 
diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c
index a2780e5334c9..77cccda1ad0c 100644
--- a/net/ipv4/tcp_ipv4.c
+++ b/net/ipv4/tcp_ipv4.c
@@ -1318,6 +1318,7 @@ int tcp_v4_conn_request(struct sock *sk, struct sk_buff *skb)
         ireq->ir_rmt_addr = saddr;
         ireq->no_srccheck = inet_sk(sk)->transparent;
         ireq->opt = tcp_v4_save_options(skb);
+        ireq->ir_mark = inet_request_mark(sk, skb);
 
         if (security_inet_conn_request(sk, skb, req))
                 goto drop_and_free;
diff --git a/net/ipv6/icmp.c b/net/ipv6/icmp.c
index 8d3952796d39..f6c84a6eb238 100644
--- a/net/ipv6/icmp.c
+++ b/net/ipv6/icmp.c
@@ -400,6 +400,7 @@ static void icmp6_send(struct sk_buff *skb, u8 type, u8 code, __u32 info)
         int len;
         int hlimit;
         int err = 0;
+        u32 mark = IP6_REPLY_MARK(net, skb->mark);
 
         if ((u8 *)hdr < skb->head ||
             (skb_network_header(skb) + sizeof(*hdr)) > skb_tail_pointer(skb))
@@ -466,6 +467,7 @@ static void icmp6_send(struct sk_buff *skb, u8 type, u8 code, __u32 info)
         fl6.daddr = hdr->saddr;
         if (saddr)
                 fl6.saddr = *saddr;
+        fl6.flowi6_mark = mark;
         fl6.flowi6_oif = iif;
         fl6.fl6_icmp_type = type;
         fl6.fl6_icmp_code = code;
@@ -474,6 +476,7 @@ static void icmp6_send(struct sk_buff *skb, u8 type, u8 code, __u32 info)
         sk = icmpv6_xmit_lock(net);
         if (sk == NULL)
                 return;
+        sk->sk_mark = mark;
         np = inet6_sk(sk);
 
         if (!icmpv6_xrlim_allow(sk, type, &fl6))
@@ -551,6 +554,7 @@ static void icmpv6_echo_reply(struct sk_buff *skb)
         int err = 0;
         int hlimit;
         u8 tclass;
+        u32 mark = IP6_REPLY_MARK(net, skb->mark);
 
         saddr = &ipv6_hdr(skb)->daddr;
 
@@ -569,11 +573,13 @@ static void icmpv6_echo_reply(struct sk_buff *skb)
                 fl6.saddr = *saddr;
         fl6.flowi6_oif = skb->dev->ifindex;
         fl6.fl6_icmp_type = ICMPV6_ECHO_REPLY;
+        fl6.flowi6_mark = mark;
         security_skb_classify_flow(skb, flowi6_to_flowi(&fl6));
 
         sk = icmpv6_xmit_lock(net);
         if (sk == NULL)
                 return;
+        sk->sk_mark = mark;
         np = inet6_sk(sk);
 
         if (!fl6.flowi6_oif && ipv6_addr_is_multicast(&fl6.daddr))
diff --git a/net/ipv6/inet6_connection_sock.c b/net/ipv6/inet6_connection_sock.c
index d4ade34ab375..a245e5ddffbd 100644
--- a/net/ipv6/inet6_connection_sock.c
+++ b/net/ipv6/inet6_connection_sock.c
@@ -81,7 +81,7 @@ struct dst_entry *inet6_csk_route_req(struct sock *sk,
         final_p = fl6_update_dst(fl6, np->opt, &final);
         fl6->saddr = ireq->ir_v6_loc_addr;
         fl6->flowi6_oif = ireq->ir_iif;
-        fl6->flowi6_mark = sk->sk_mark;
+        fl6->flowi6_mark = ireq->ir_mark;
         fl6->fl6_dport = ireq->ir_rmt_port;
         fl6->fl6_sport = htons(ireq->ir_num);
         security_req_classify_flow(req, flowi6_to_flowi(fl6));
diff --git a/net/ipv6/route.c b/net/ipv6/route.c
index 004fffb6c221..f0a8ff9ed891 100644
--- a/net/ipv6/route.c
+++ b/net/ipv6/route.c
@@ -1176,7 +1176,7 @@ void ip6_update_pmtu(struct sk_buff *skb, struct net *net, __be32 mtu,
 
         memset(&fl6, 0, sizeof(fl6));
         fl6.flowi6_oif = oif;
-        fl6.flowi6_mark = mark;
+        fl6.flowi6_mark = mark ? mark : IP6_REPLY_MARK(net, skb->mark);
         fl6.daddr = iph->daddr;
         fl6.saddr = iph->saddr;
         fl6.flowlabel = ip6_flowinfo(iph);
diff --git a/net/ipv6/syncookies.c b/net/ipv6/syncookies.c
index bb53a5e73c1a..a822b880689b 100644
--- a/net/ipv6/syncookies.c
+++ b/net/ipv6/syncookies.c
@@ -216,6 +216,8 @@ struct sock *cookie_v6_check(struct sock *sk, struct sk_buff *skb)
             ipv6_addr_type(&ireq->ir_v6_rmt_addr) & IPV6_ADDR_LINKLOCAL)
                 ireq->ir_iif = inet6_iif(skb);
 
+        ireq->ir_mark = inet_request_mark(sk, skb);
+
         req->expires = 0UL;
         req->num_retrans = 0;
         ireq->ecn_ok            = ecn_ok;
@@ -242,7 +244,7 @@ struct sock *cookie_v6_check(struct sock *sk, struct sk_buff *skb)
                 final_p = fl6_update_dst(&fl6, np->opt, &final);
                 fl6.saddr = ireq->ir_v6_loc_addr;
                 fl6.flowi6_oif = sk->sk_bound_dev_if;
-                fl6.flowi6_mark = sk->sk_mark;
+                fl6.flowi6_mark = ireq->ir_mark;
                 fl6.fl6_dport = ireq->ir_rmt_port;
                 fl6.fl6_sport = inet_sk(sk)->inet_sport;
                 security_req_classify_flow(req, flowi6_to_flowi(&fl6));
diff --git a/net/ipv6/sysctl_net_ipv6.c b/net/ipv6/sysctl_net_ipv6.c
index 7f405a168822..058f3eca2e53 100644
--- a/net/ipv6/sysctl_net_ipv6.c
+++ b/net/ipv6/sysctl_net_ipv6.c
@@ -38,6 +38,13 @@ static struct ctl_table ipv6_table_template[] = {
                 .mode           = 0644,
                 .proc_handler   = proc_dointvec
         },
+        {
+                .procname       = "fwmark_reflect",
+                .data           = &init_net.ipv6.sysctl.fwmark_reflect,
+                .maxlen         = sizeof(int),
+                .mode           = 0644,
+                .proc_handler   = proc_dointvec
+        },
         { }
 };
 
diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c
index 3a267bf14f2f..f07b2abba359 100644
--- a/net/ipv6/tcp_ipv6.c
+++ b/net/ipv6/tcp_ipv6.c
@@ -812,6 +812,7 @@ static void tcp_v6_send_response(struct sk_buff *skb, u32 seq, u32 ack, u32 win,
                 fl6.flowi6_oif = inet6_iif(skb);
         else
                 fl6.flowi6_oif = oif;
+        fl6.flowi6_mark = IP6_REPLY_MARK(net, skb->mark);
         fl6.fl6_dport = t1->dest;
         fl6.fl6_sport = t1->source;
         security_skb_classify_flow(skb, flowi6_to_flowi(&fl6));
@@ -1033,6 +1034,7 @@ static int tcp_v6_conn_request(struct sock *sk, struct sk_buff *skb)
                 TCP_ECN_create_request(req, skb, sock_net(sk));
 
         ireq->ir_iif = sk->sk_bound_dev_if;
+        ireq->ir_mark = inet_request_mark(sk, skb);
 
         /* So that link locals have meaning */
         if (!sk->sk_bound_dev_if &&