gre: copy ToS/DiffServ bits to outer IP header

When tunneling IP traffic with GRE this patch makes it possible to export the ToS/DiffServ information to the outer IP header. This is particularly useful in a scenario with ESP/AH where the inner IP header is encrypted but the packet priority/DiffServ information should still be respected by the transporting routers (for example in an MPLS backbone network).

The feature is disabled by default and can be enabled on a per-interface basis (/proc/sys/net/ipv4/conf/ethX/gre_copy_tos).

Also does this bring Linux back in the game, as JunOS/IOS provide this for quite some time:
http://www.cisco.com/en/US/docs/ios/11_3/feature/guide/greqos.html
http://www.juniper.net/techpubs/software/junos/junos94/swconfig-services/configuring-a-gre-tunnel-to-copy-tos-bits-to-the-outer-ip-header.html

diff -urN vanilla-linux-2.6.29.4/include/linux/sysctl.h gre-copy-tos/include/linux/sysctl.h
@@ -490,6 +490,7 @@

	NET_IPV4_CONF_ARP_IGNORE=19,
	NET_IPV4_CONF_PROMOTE_SECONDARIES=20,
	NET_IPV4_CONF_ARP_ACCEPT=21,
+	NET_IPV4_CONF_GRE_COPY_TOS=22,
	__NET_IPV4_CONF_MAX
};

diff -urN vanilla-linux-2.6.29.4/kernel/sysctl_check.c gre-copy-tos/kernel/sysctl_check.c
@@ -219,6 +219,7 @@

	{ NET_IPV4_CONF_ARP_IGNORE,		"arp_ignore" },
	{ NET_IPV4_CONF_PROMOTE_SECONDARIES,	"promote_secondaries" },
	{ NET_IPV4_CONF_ARP_ACCEPT,		"arp_accept" },
+	{ NET_IPV4_CONF_GRE_COPY_TOS,		"gre_copy_tos" },
	{}
};

diff -urN vanilla-linux-2.6.29.4/net/ipv4/devinet.c gre-copy-tos/net/ipv4/devinet.c
@@ -1439,6 +1439,7 @@

DEVINET_SYSCTL_RW_ENTRY(ARP_ANNOUNCE, "arp_announce"), DEVINET_SYSCTL_RW_ENTRY(ARP_IGNORE, "arp_ignore"), DEVINET_SYSCTL_RW_ENTRY(ARP_ACCEPT, "arp_accept"),
+ DEVINET_SYSCTL_RW_ENTRY(GRE_COPY_TOS, "gre_copy_tos"),

DEVINET_SYSCTL_FLUSHING_ENTRY(NOXFRM, "disable_xfrm"), DEVINET_SYSCTL_FLUSHING_ENTRY(NOPOLICY, "disable_policy"),

diff -urN vanilla-linux-2.6.29.4/net/ipv4/ip_gre.c gre-copy-tos/net/ipv4/ip_gre.c
@@ -610,6 +610,7 @@

	struct net_device_stats *stats = &tunnel->dev->stats;
	struct iphdr  *old_iph = ip_hdr(skb);
	struct iphdr  *tiph;
+	struct in_device *in_dev;
	u8     tos;
	__be16 df;
	struct rtable *rt;     			/* Route to the other host */
@@ -677,11 +678,13 @@
	}

	tos = tiph->tos;
-	if (tos&1) {
+	in_dev = in_dev_get(dev);
+	if ((in_dev && IN_DEV_ORCONF(in_dev, GRE_COPY_TOS)) || tos&1) {
		if (skb->protocol == htons(ETH_P_IP))
			tos = old_iph->tos;
		tos &= ~1;
	}

+ in_dev_put(in_dev);

	{
		struct flowi fl = { .oif = tunnel->parms.link,

Now using the rtnl_link interface.

diff -urN vanilla-linux-2.6.29.4/include/linux/if_tunnel.h dev-gre/include/linux/if_tunnel.h
@@ -34,6 +34,7 @@

	__be32			i_key;
	__be32			o_key;
	struct iphdr		iph;

+ __u8 copy_tos;
};

/* SIT-mode i_flags */
@@ -63,6 +64,7 @@
IFLA_GRE_REMOTE,
IFLA_GRE_TTL,
IFLA_GRE_TOS,
+ IFLA_GRE_COPY_TOS,

	IFLA_GRE_PMTUDISC,
	__IFLA_GRE_MAX,
};

diff -urN vanilla-linux-2.6.29.4/net/ipv4/ip_gre.c dev-gre/net/ipv4/ip_gre.c
@@ -677,7 +677,7 @@

	}

	tos = tiph->tos;
-	if (tos&1) {
+	if (tunnel->parms.copy_tos || tos&1) {
		if (skb->protocol == htons(ETH_P_IP))
			tos = old_iph->tos;
		tos &= ~1;
@@ -991,6 +991,7 @@
				t->parms.iph.ttl = p.iph.ttl;
				t->parms.iph.tos = p.iph.tos;
				t->parms.iph.frag_off = p.iph.frag_off;
+				t->parms.copy_tos = p.copy_tos;
				if (t->parms.link != p.link) {
					t->parms.link = p.link;
					dev->mtu = ipgre_tunnel_bind_dev(dev);
@@ -1383,6 +1384,9 @@
	if (data[IFLA_GRE_TOS])
		parms->iph.tos = nla_get_u8(data[IFLA_GRE_TOS]);

+ if (data[IFLA_GRE_COPY_TOS])
+ parms->copy_tos = nla_get_u8(data[IFLA_GRE_COPY_TOS]);
+

	if (!data[IFLA_GRE_PMTUDISC] || nla_get_u8(data[IFLA_GRE_PMTUDISC]))
		parms->iph.frag_off = htons(IP_DF);
}
@@ -1536,6 +1540,8 @@
		nla_total_size(1) +
		/* IFLA_GRE_TOS */
		nla_total_size(1) +
+		/* IFLA_GRE_COPY_TOS */
+		nla_total_size(1) +
		/* IFLA_GRE_PMTUDISC */
		nla_total_size(1) +
		0;
@@ -1555,6 +1561,7 @@
	NLA_PUT_BE32(skb, IFLA_GRE_REMOTE, p->iph.daddr);
	NLA_PUT_U8(skb, IFLA_GRE_TTL, p->iph.ttl);
	NLA_PUT_U8(skb, IFLA_GRE_TOS, p->iph.tos);
+	NLA_PUT_U8(skb, IFLA_GRE_COPY_TOS, p->copy_tos);
	NLA_PUT_U8(skb, IFLA_GRE_PMTUDISC, !!(p->iph.frag_off & htons(IP_DF)));

	return 0;
@@ -1573,6 +1580,7 @@
	[IFLA_GRE_REMOTE]	= { .len = FIELD_SIZEOF(struct iphdr, daddr) },
	[IFLA_GRE_TTL]		= { .type = NLA_U8 },
	[IFLA_GRE_TOS]		= { .type = NLA_U8 },
+	[IFLA_GRE_COPY_TOS]	= { .type = NLA_U8 },
	[IFLA_GRE_PMTUDISC]	= { .type = NLA_U8 },
};

Your patches are seriously whitespace-damaged. Please make sure to send them properly (test by sending to yourself first).

And the corresponding patch for iproute2.

diff -urN iproute2-2.6.29-1/include/linux/if_tunnel.h iproute2-2.6.29-1-gre-dev/include/linux/if_tunnel.h
@@ -31,6 +31,7 @@

	__be32			i_key;
	__be32			o_key;
	struct iphdr		iph;

+ __u8 copy_tos;
};

/* SIT-mode i_flags */
@@ -60,6 +61,7 @@
IFLA_GRE_REMOTE,
IFLA_GRE_TTL,
IFLA_GRE_TOS,
+ IFLA_GRE_COPY_TOS,

	IFLA_GRE_PMTUDISC,
	__IFLA_GRE_MAX,
};

diff -urN iproute2-2.6.29-1/ip/iptunnel.c iproute2-2.6.29-1-gre-dev/ip/iptunnel.c
@@ -41,7 +41,7 @@

	fprintf(stderr, "Usage: ip tunnel { add | change | del | show } [ NAME ]\n");
	fprintf(stderr, "          [ mode { ipip | gre | sit | isatap } ] [ remote ADDR ] [ local ADDR ]\n");
	fprintf(stderr, "          [ [i|o]seq ] [ [i|o]key KEY ] [ [i|o]csum ]\n");
-	fprintf(stderr, "          [ ttl TTL ] [ tos TOS ] [ [no]pmtudisc ] [ dev PHYS_DEV ]\n");
+	fprintf(stderr, "          [ ttl TTL ] [ tos TOS ] [ [no]pmtudisc ] [ [no]copytos ] [ dev PHYS_DEV ]\n");
	fprintf(stderr, "\n");
	fprintf(stderr, "Where: NAME := STRING\n");
	fprintf(stderr, "       ADDR := { IP_ADDRESS | any }\n");
@@ -160,6 +160,10 @@
			p->iph.frag_off = 0;
		} else if (strcmp(*argv, "pmtudisc") == 0) {
			p->iph.frag_off = htons(IP_DF);
+		} else if (strcmp(*argv, "nocopytos") == 0) {
+			p->copy_tos = 0;
+		} else if (strcmp(*argv, "copytos") == 0) {
+			p->copy_tos = 1;
		} else if (strcmp(*argv, "remote") == 0) {
			NEXT_ARG();
			if (strcmp(*argv, "any"))
@@ -353,6 +357,9 @@
	if (!(p->iph.frag_off&htons(IP_DF)))
		printf(" nopmtudisc");

+ if (p->copy_tos)
+ printf(" copytos");
+

	if ((p->i_flags&GRE_KEY) && (p->o_flags&GRE_KEY) && p->o_key == p->i_key)
		printf(" key %s", s3);
	else if ((p->i_flags|p->o_flags)&GRE_KEY) {

diff -urN iproute2-2.6.29-1/ip/link_gre.c iproute2-2.6.29-1-gre-dev/ip/link_gre.c
@@ -29,7 +29,7 @@

	fprintf(stderr, "Usage: ip link { add | set | change | replace | del } NAME\n");
	fprintf(stderr, "          type { gre | gretap } [ remote ADDR ] [ local ADDR ]\n");
	fprintf(stderr, "          [ [i|o]seq ] [ [i|o]key KEY ] [ [i|o]csum ]\n");
-	fprintf(stderr, "          [ ttl TTL ] [ tos TOS ] [ [no]pmtudisc ] [ dev PHYS_DEV ]\n");
+	fprintf(stderr, "          [ ttl TTL ] [ tos TOS ] [ [no]pmtudisc ] [ [no]copytos ] [ dev PHYS_DEV ]\n");
	fprintf(stderr, "\n");
	fprintf(stderr, "Where: NAME := STRING\n");
	fprintf(stderr, "       ADDR := { IP_ADDRESS | any }\n");
@@ -61,6 +61,7 @@
	__u8 pmtudisc = 1;
	__u8 ttl = 0;
	__u8 tos = 0;
+	__u8 copy_tos = 0;
	int len;

	if (!(n->nlmsg_flags & NLM_F_CREATE)) {
@@ -125,6 +126,9 @@
		if (greinfo[IFLA_GRE_TOS])
			tos = *(__u8 *)RTA_DATA(greinfo[IFLA_GRE_TOS]);

+ if (greinfo[IFLA_GRE_COPY_TOS])
+ copy_tos = *(__u8 *)RTA_DATA(greinfo[IFLA_GRE_COPY_TOS]);
+

		if (greinfo[IFLA_GRE_LINK])
			link = *(__u8 *)RTA_DATA(greinfo[IFLA_GRE_LINK]);
	}
@@ -196,6 +200,10 @@
			pmtudisc = 0;
		} else if (!matches(*argv, "pmtudisc")) {
			pmtudisc = 1;
+		} else if (!matches(*argv, "nocopytos")) {
+			copy_tos = 0;
+		} else if (!matches(*argv, "copytos")) {
+			copy_tos = 1;
		} else if (!matches(*argv, "remote")) {
			NEXT_ARG();
			if (strcmp(*argv, "any"))
@@ -262,6 +270,7 @@
		addattr32(n, 1024, IFLA_GRE_LINK, link);
	addattr_l(n, 1024, IFLA_GRE_TTL, &ttl, 1);
	addattr_l(n, 1024, IFLA_GRE_TOS, &tos, 1);

+ addattr_l(n, 1024, IFLA_GRE_COPY_TOS, &copy_tos, 1);

	return 0;
}
@@ -325,6 +334,10 @@
	    !*(__u8 *)RTA_DATA(tb[IFLA_GRE_PMTUDISC]))
		fputs("nopmtudisc ", f);

+ if (tb[IFLA_GRE_COPY_TOS] &&
+ *(__u8 *)RTA_DATA(tb[IFLA_GRE_COPY_TOS]))
+ fputs("copytos ", f);
+

	if (tb[IFLA_GRE_IFLAGS])
		iflags = *(__u16 *)RTA_DATA(tb[IFLA_GRE_IFLAGS]);