netfilter: nf_defrag_ipv4: Add sysctl to disable per interface

Add a sysctl nf_ipv4_defrag_skip to skip defragmentation per
interface. This is set 0 to preserve existing behavior (always
defrag per interface).

This is useful for pure ipv4 forwarding scenarios (without NAT)
in conjunction with xfrm. It appears that network stack defrags
the packets and then forwards them to xfrm which then encrypts
and then later fragments them on a different boundary compared
to the source.

CRs-Fixed: 2333588
Change-Id: I11956284a9692579274e8626f61cc6432232254c
Signed-off-by: Subash Abhinov Kasiviswanathan <subashab@codeaurora.org>

Documentation/networking/ip-sysctl.txt

@@ -1337,6 +1337,10 @@ igmp_link_local_mcast_reports - BOOLEAN
 	224.0.0.X range.
 	Default TRUE
 
+nf_ipv4_defrag_skip - BOOLEAN
+	Skip defragmentation per interface if set.
+	Default : 0 (always defrag)
+
 Alexey Kuznetsov.
 kuznet@ms2.inr.ac.ru
 

include/linux/inetdevice.h

@@ -131,6 +131,8 @@ static inline void ipv4_devconf_setall(struct in_device *in_dev)
 #define IN_DEV_ARP_ANNOUNCE(in_dev)	IN_DEV_MAXCONF((in_dev), ARP_ANNOUNCE)
 #define IN_DEV_ARP_IGNORE(in_dev)	IN_DEV_MAXCONF((in_dev), ARP_IGNORE)
 #define IN_DEV_ARP_NOTIFY(in_dev)	IN_DEV_MAXCONF((in_dev), ARP_NOTIFY)
+#define IN_DEV_NF_IPV4_DEFRAG_SKIP(in_dev) \
+	IN_DEV_ORCONF((in_dev), NF_IPV4_DEFRAG_SKIP)
 
 struct in_ifaddr {
 	struct hlist_node	hash;

include/uapi/linux/ip.h

@@ -169,6 +169,7 @@ enum
 	IPV4_DEVCONF_DROP_UNICAST_IN_L2_MULTICAST,
 	IPV4_DEVCONF_DROP_GRATUITOUS_ARP,
 	IPV4_DEVCONF_BC_FORWARDING,
+	IPV4_DEVCONF_NF_IPV4_DEFRAG_SKIP,
 	__IPV4_DEVCONF_MAX
 };
 

include/uapi/linux/sysctl.h

@@ -483,6 +483,7 @@ enum
 	NET_IPV4_CONF_PROMOTE_SECONDARIES=20,
 	NET_IPV4_CONF_ARP_ACCEPT=21,
 	NET_IPV4_CONF_ARP_NOTIFY=22,
+	NET_IPV4_CONF_NF_IPV4_DEFRAG_SKIP = 23,
 };
 
 /* /proc/sys/net/ipv4/netfilter */

kernel/sysctl_binary.c

@@ -257,6 +257,7 @@ static const struct bin_table bin_net_ipv4_conf_vars_table[] = {
 	{ CTL_INT,	NET_IPV4_CONF_NOPOLICY,			"disable_policy" },
 	{ CTL_INT,	NET_IPV4_CONF_FORCE_IGMP_VERSION,	"force_igmp_version" },
 	{ CTL_INT,	NET_IPV4_CONF_PROMOTE_SECONDARIES,	"promote_secondaries" },
+	{ CTL_INT, NET_IPV4_CONF_NF_IPV4_DEFRAG_SKIP, "nf_ipv4_defrag_skip" },
 	{}
 };
 

net/ipv4/devinet.c

@@ -2310,6 +2310,8 @@ static struct devinet_sysctl_table {
 					      "route_localnet"),
 		DEVINET_SYSCTL_FLUSHING_ENTRY(DROP_UNICAST_IN_L2_MULTICAST,
 					      "drop_unicast_in_l2_multicast"),
+		DEVINET_SYSCTL_RW_ENTRY(NF_IPV4_DEFRAG_SKIP,
+					"nf_ipv4_defrag_skip"),
 	},
 };
 

net/ipv4/netfilter/nf_defrag_ipv4.c

@@ -11,6 +11,7 @@
 #include <linux/netfilter.h>
 #include <linux/module.h>
 #include <linux/skbuff.h>
+#include <linux/inetdevice.h>
 #include <net/netns/generic.h>
 #include <net/route.h>
 #include <net/ip.h>
@@ -83,8 +84,13 @@ static unsigned int ipv4_conntrack_defrag(void *priv,
 #endif
 	/* Gather fragments. */
 	if (ip_is_fragment(ip_hdr(skb))) {
-		enum ip_defrag_users user =
-			nf_ct_defrag_user(state->hook, skb);
+		enum ip_defrag_users user;
+
+		if (skb->dev &&
+		    IN_DEV_NF_IPV4_DEFRAG_SKIP(__in_dev_get_rcu(skb->dev)))
+			return NF_ACCEPT;
+
+		user = nf_ct_defrag_user(state->hook, skb);
 
 		if (nf_ct_ipv4_gather_frags(state->net, skb, user))
 			return NF_STOLEN;