net: add a sysctl to reflect the fwmark on replies

Kernel-originated IP packets that have no user socket associated
with them (e.g., ICMP errors and echo replies, TCP RSTs, etc.)
are emitted with a mark of zero. Add a sysctl to make them have
the same mark as the packet they are replying to.

This allows an administrator that wishes to do so to use
mark-based routing, firewalling, etc. for these replies by
marking the original packets inbound.

Tested using user-mode linux:
 - ICMP/ICMPv6 echo replies and errors.
 - TCP RST packets (IPv4 and IPv6).

Change-Id: I6873d973196797bcf32e2e91976df647c7e8b85a
Signed-off-by: Lorenzo Colitti <lorenzo@google.com>

11 files changed, 55 insertions, 3 deletions

diff --git a/Documentation/networking/ip-sysctl.txt b/Documentation/networking/ip-sysctl.txt
index 3458d6343e0..77731bba5c6 100644
--- a/Documentation/networking/ip-sysctl.txt
+++ b/Documentation/networking/ip-sysctl.txt
@@ -22,6 +22,13 @@ ip_no_pmtu_disc - BOOLEAN
 min_pmtu - INTEGER
 	default 552 - minimum discovered Path MTU
 
+fwmark_reflect - BOOLEAN
+	Controls the fwmark of kernel-generated IPv4 reply packets that are not
+	associated with a socket for example, TCP RSTs or ICMP echo replies).
+	If unset, these packets have a fwmark of zero. If set, they have the
+	fwmark of the packet they are replying to.
+	Default: 0
+
 route/max_size - INTEGER
 	Maximum number of routes allowed in the kernel.  Increase
 	this when using large numbers of interfaces and/or routes.
@@ -1087,6 +1094,13 @@ conf/all/forwarding - BOOLEAN
 proxy_ndp - BOOLEAN
 	Do proxy ndp.
 
+fwmark_reflect - BOOLEAN
+	Controls the fwmark of kernel-generated IPv6 reply packets that are not
+	associated with a socket for example, TCP RSTs or ICMPv6 echo replies).
+	If unset, these packets have a fwmark of zero. If set, they have the
+	fwmark of the packet they are replying to.
+	Default: 0
+
 conf/interface/*:
 	Change special settings per interface.
 
diff --git a/include/net/ip.h b/include/net/ip.h
index a68f838a132..509b8807927 100644
--- a/include/net/ip.h
+++ b/include/net/ip.h
@@ -225,6 +225,9 @@ extern void ipfrag_init(void);
 
 extern 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 ab47582f6c0..cc344ca9d0a 100644
--- a/include/net/ipv6.h
+++ b/include/net/ipv6.h
@@ -111,6 +111,9 @@ struct frag_hdr {
 
 #define	IP6_MF	0x0001
 
+#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 2ba9de89e8e..222461a7cc5 100644
--- a/include/net/netns/ipv4.h
+++ b/include/net/netns/ipv4.h
@@ -64,6 +64,8 @@ struct netns_ipv4 {
 
 	int sysctl_tcp_ecn;
 
+	int sysctl_fwmark_reflect;
+
 	kgid_t sysctl_ping_group_range[2];
 	long sysctl_tcp_mem[3];
 
diff --git a/include/net/netns/ipv6.h b/include/net/netns/ipv6.h
index 005e2c2e39a..4b9f99e3a91 100644
--- a/include/net/netns/ipv6.h
+++ b/include/net/netns/ipv6.h
@@ -28,6 +28,7 @@ struct netns_sysctl_ipv6 {
 	int ip6_rt_mtu_expires;
 	int ip6_rt_min_advmss;
 	int icmpv6_time;
+	int fwmark_reflect;
 };
 
 struct netns_ipv6 {
diff --git a/net/ipv4/icmp.c b/net/ipv4/icmp.c
index 562efd91f45..cc38f44306e 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;
@@ -361,6 +363,7 @@ 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));
@@ -379,7 +382,7 @@ 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)
 {
@@ -391,6 +394,7 @@ 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;
@@ -488,6 +492,7 @@ 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;
 
@@ -584,6 +589,7 @@ 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;
@@ -600,11 +606,12 @@ 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;
 
-	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/ip_output.c b/net/ipv4/ip_output.c
index 4bcabf3ab4c..c2ee385cecf 100644
--- a/net/ipv4/ip_output.c
+++ b/net/ipv4/ip_output.c
@@ -1497,7 +1497,8 @@ 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/sysctl_net_ipv4.c b/net/ipv4/sysctl_net_ipv4.c
index f9bb5d7488e..e6cdcb32b33 100644
--- a/net/ipv4/sysctl_net_ipv4.c
+++ b/net/ipv4/sysctl_net_ipv4.c
@@ -859,6 +859,13 @@ static struct ctl_table ipv4_net_table[] = {
 		.mode		= 0644,
 		.proc_handler	= ipv4_tcp_mem,
 	},
+	{
+		.procname	= "fwmark_reflect",
+		.data		= &init_net.ipv4.sysctl_fwmark_reflect,
+		.maxlen		= sizeof(int),
+		.mode		= 0644,
+		.proc_handler	= proc_dointvec,
+	},
 	{ }
 };
 
diff --git a/net/ipv6/icmp.c b/net/ipv6/icmp.c
index 1d2902e6178..28da4003e84 100644
--- a/net/ipv6/icmp.c
+++ b/net/ipv6/icmp.c
@@ -397,6 +397,7 @@ 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 + sizeof(*hdr)) > skb->tail)
@@ -462,6 +463,7 @@ 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;
@@ -470,6 +472,7 @@ 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)
 	struct dst_entry *dst;
 	int err = 0;
 	int hlimit;
+	u32 mark = IP6_REPLY_MARK(net, skb->mark);
 
 	saddr = &ipv6_hdr(skb)->daddr;
 
@@ -567,11 +571,13 @@ 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/sysctl_net_ipv6.c b/net/ipv6/sysctl_net_ipv6.c
index e85c48bd404..53a9f5a6453 100644
--- a/net/ipv6/sysctl_net_ipv6.c
+++ b/net/ipv6/sysctl_net_ipv6.c
@@ -24,6 +24,13 @@ static 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 0a17ed9eaf3..71545cb17ab 100644
--- a/net/ipv6/tcp_ipv6.c
+++ b/net/ipv6/tcp_ipv6.c
@@ -791,6 +791,7 @@ static void tcp_v6_send_response(struct sk_buff *skb, u32 seq, u32 ack, u32 win,
 	fl6.flowi6_proto = IPPROTO_TCP;
 	if (ipv6_addr_type(&fl6.daddr) & IPV6_ADDR_LINKLOCAL)
 		fl6.flowi6_oif = inet6_iif(skb);
+	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));