Merge branch 'ip-ingress-skb-reason'

Menglong Dong says:

====================
net: ip: add skb drop reasons to ip ingress

In the series "net: use kfree_skb_reason() for ip/udp packet receive",
skb drop reasons are added to the basic ingress path of IPv4. And in
the series "net: use kfree_skb_reason() for ip/neighbour", the egress
paths of IPv4 and IPv6 are handled. Related links:

https://lore.kernel.org/netdev/[e-mail]/
https://lore.kernel.org/netdev/[e-mail]/

Seems we still have a lot work to do with IP layer, including IPv6 basic
ingress path, IPv4/IPv6 forwarding, IPv6 exthdrs, fragment and defrag,
etc.

In this series, skb drop reasons are added to the basic ingress path of
IPv6 protocol and IPv4/IPv6 packet forwarding. Following functions, which
are used for IPv6 packet receiving are handled:

ip6_pkt_drop()
ip6_rcv_core()
ip6_protocol_deliver_rcu()

And following functions that used for IPv6 TLV parse are handled:

ip6_parse_tlv()
ipv6_hop_ra()
ipv6_hop_ioam()
ipv6_hop_jumbo()
ipv6_hop_calipso()
ipv6_dest_hao()

Besides, ip_forward() and ip6_forward(), which are used for IPv4/IPv6
forwarding, are also handled. And following new drop reasons are added:

/* host unreachable, corresponding to IPSTATS_MIB_INADDRERRORS */
SKB_DROP_REASON_IP_INADDRERRORS
/* network unreachable, corresponding to IPSTATS_MIB_INADDRERRORS */
SKB_DROP_REASON_IP_INNOROUTES
/* packet size is too big, corresponding to
* IPSTATS_MIB_INTOOBIGERRORS
*/
SKB_DROP_REASON_PKT_TOO_BIG

In order to simply the definition and assignment for
'enum skb_drop_reason', some helper functions are introduced in the 1th
patch. I'm not such if this is necessary, but it makes the code simpler.
For example, we can replace the code:

if (reason == SKB_DROP_REASON_NOT_SPECIFIED)
reason = SKB_DROP_REASON_IP_INHDR;

with:

SKB_DR_OR(reason, IP_INHDR);

In the 6th patch, the statistics for skb in ipv6_hop_jum() is removed,
as I think it is redundant. There are two call chains for
ipv6_hop_jumbo(). The first one is:

ipv6_destopt_rcv() -> ip6_parse_tlv() -> ipv6_hop_jumbo()

On this call chain, the drop statistics will be done in
ipv6_destopt_rcv() with 'IPSTATS_MIB_INHDRERRORS' if ipv6_hop_jumbo()
returns false.

The second call chain is:

ip6_rcv_core() -> ipv6_parse_hopopts() -> ip6_parse_tlv()

And the drop statistics will also be done in ip6_rcv_core() with
'IPSTATS_MIB_INHDRERRORS' if ipv6_hop_jumbo() returns false.

Therefore, the statistics in ipv6_hop_jumbo() is redundant, which
means the drop is counted twice. The statistics in ipv6_hop_jumbo()
is almost the same as the outside, except the
'IPSTATS_MIB_INTRUNCATEDPKTS', which seems that we have to ignore it.

======================================================================

Here is a basic test for IPv6 forwarding packet drop that monitored by
'dropwatch' tool:

drop at: ip6_forward+0x81a/0xb70 (0xffffffff86c73f8a)
origin: software
input port ifindex: 7
timestamp: Wed Apr 13 11:51:06 2022 130010176 nsec
protocol: 0x86dd
length: 94
original length: 94
drop reason: IP_INADDRERRORS

The origin cause of this case is that IPv6 doesn't allow to forward the
packet with LOCAL-LINK saddr, and results the 'IP_INADDRERRORS' drop
reason.
====================

Signed-off-by: David S. Miller <[e-mail]>

include/linux/icmpv6.h

···
 extern int				icmpv6_err_convert(u8 type, u8 code,
 							   int *err);
 extern void				icmpv6_cleanup(void);
-extern void				icmpv6_param_prob(struct sk_buff *skb,
-							  u8 code, int pos);
+extern void				icmpv6_param_prob_reason(struct sk_buff *skb,
+								 u8 code, int pos,
+								 enum skb_drop_reason reason);
 
 struct flowi6;
 struct in6_addr;
···
 							 const struct in6_addr *saddr,
 							 const struct in6_addr *daddr,
 							 int oif);
+
+static inline void icmpv6_param_prob(struct sk_buff *skb, u8 code, int pos)
+{
+	icmpv6_param_prob_reason(skb, code, pos,
+				 SKB_DROP_REASON_NOT_SPECIFIED);
+}
 
 static inline bool icmpv6_is_err(int type)
 {

include/linux/skbuff.h

···
 					 * 2211, such as a broadcasts
 					 * ICMP_TIMESTAMP
 					 */
+	SKB_DROP_REASON_IP_INADDRERRORS,	/* host unreachable, corresponding
+						 * to IPSTATS_MIB_INADDRERRORS
+						 */
+	SKB_DROP_REASON_IP_INNOROUTES,	/* network unreachable, corresponding
+					 * to IPSTATS_MIB_INADDRERRORS
+					 */
+	SKB_DROP_REASON_PKT_TOO_BIG,	/* packet size is too big (maybe exceed
+					 * the MTU)
+					 */
 	SKB_DROP_REASON_MAX,
 };
+
+#define SKB_DR_INIT(name, reason)				\
+	enum skb_drop_reason name = SKB_DROP_REASON_##reason
+#define SKB_DR(name)						\
+	SKB_DR_INIT(name, NOT_SPECIFIED)
+#define SKB_DR_SET(name, reason)				\
+	(name = SKB_DROP_REASON_##reason)
+#define SKB_DR_OR(name, reason)					\
+	do {							\
+		if (name == SKB_DROP_REASON_NOT_SPECIFIED)	\
+			SKB_DR_SET(name, reason);		\
+	} while (0)
 
 /* To allow 64K frame to be packed as single skb without frag_list we
  * require 64K/PAGE_SIZE pages plus 1 additional page to allow for

include/trace/events/skb.h

···
 	EM(SKB_DROP_REASON_TAP_TXFILTER, TAP_TXFILTER)		\
 	EM(SKB_DROP_REASON_ICMP_CSUM, ICMP_CSUM)		\
 	EM(SKB_DROP_REASON_INVALID_PROTO, INVALID_PROTO)	\
+	EM(SKB_DROP_REASON_IP_INADDRERRORS, IP_INADDRERRORS)	\
+	EM(SKB_DROP_REASON_IP_INNOROUTES, IP_INNOROUTES)	\
+	EM(SKB_DROP_REASON_PKT_TOO_BIG, PKT_TOO_BIG)		\
 	EMe(SKB_DROP_REASON_MAX, MAX)
 
 #undef EM

net/ipv4/ip_forward.c

···
 	struct rtable *rt;	/* Route we use */
 	struct ip_options *opt	= &(IPCB(skb)->opt);
 	struct net *net;
+	SKB_DR(reason);
 
 	/* that should never happen */
 	if (skb->pkt_type != PACKET_HOST)
···
 	if (skb_warn_if_lro(skb))
 		goto drop;
 
-	if (!xfrm4_policy_check(NULL, XFRM_POLICY_FWD, skb))
+	if (!xfrm4_policy_check(NULL, XFRM_POLICY_FWD, skb)) {
+		SKB_DR_SET(reason, XFRM_POLICY);
 		goto drop;
+	}
 
 	if (IPCB(skb)->opt.router_alert && ip_call_ra_chain(skb))
 		return NET_RX_SUCCESS;
···
 	if (ip_hdr(skb)->ttl <= 1)
 		goto too_many_hops;
 
-	if (!xfrm4_route_forward(skb))
+	if (!xfrm4_route_forward(skb)) {
+		SKB_DR_SET(reason, XFRM_POLICY);
 		goto drop;
+	}
 
 	rt = skb_rtable(skb);
 
···
 		IP_INC_STATS(net, IPSTATS_MIB_FRAGFAILS);
 		icmp_send(skb, ICMP_DEST_UNREACH, ICMP_FRAG_NEEDED,
 			  htonl(mtu));
+		SKB_DR_SET(reason, PKT_TOO_BIG);
 		goto drop;
 	}
 
···
 	/* Tell the sender its packet died... */
 	__IP_INC_STATS(net, IPSTATS_MIB_INHDRERRORS);
 	icmp_send(skb, ICMP_TIME_EXCEEDED, ICMP_EXC_TTL, 0);
+	SKB_DR_SET(reason, IP_INHDR);
 drop:
-	kfree_skb(skb);
+	kfree_skb_reason(skb, reason);
 	return NET_RX_DROP;
 }

net/ipv4/route.c

···
 	struct inet_peer *peer;
 	unsigned long now;
 	struct net *net;
+	SKB_DR(reason);
 	bool send;
 	int code;
 
···
 	if (!IN_DEV_FORWARD(in_dev)) {
 		switch (rt->dst.error) {
 		case EHOSTUNREACH:
+			SKB_DR_SET(reason, IP_INADDRERRORS);
 			__IP_INC_STATS(net, IPSTATS_MIB_INADDRERRORS);
 			break;
 
 		case ENETUNREACH:
+			SKB_DR_SET(reason, IP_INNOROUTES);
 			__IP_INC_STATS(net, IPSTATS_MIB_INNOROUTES);
 			break;
 		}
···
 		break;
 	case ENETUNREACH:
 		code = ICMP_NET_UNREACH;
+		SKB_DR_SET(reason, IP_INNOROUTES);
 		__IP_INC_STATS(net, IPSTATS_MIB_INNOROUTES);
 		break;
 	case EACCES:
···
 	if (send)
 		icmp_send(skb, ICMP_DEST_UNREACH, code, 0);
 
-out:	kfree_skb(skb);
+out:	kfree_skb_reason(skb, reason);
 	return 0;
 }
 

net/ipv6/exthdrs.c

···
 			break;
 		fallthrough;
 	case 2: /* send ICMP PARM PROB regardless and drop packet */
-		icmpv6_param_prob(skb, ICMPV6_UNK_OPTION, optoff);
+		icmpv6_param_prob_reason(skb, ICMPV6_UNK_OPTION, optoff,
+					 SKB_DROP_REASON_UNHANDLED_PROTO);
 		return false;
 	}
 
 drop:
-	kfree_skb(skb);
+	kfree_skb_reason(skb, SKB_DROP_REASON_UNHANDLED_PROTO);
 	return false;
 }
 
···
 	if (len == 0)
 		return true;
 bad:
-	kfree_skb(skb);
+	kfree_skb_reason(skb, SKB_DROP_REASON_IP_INHDR);
 	return false;
 }
 
···
 	struct ipv6_destopt_hao *hao;
 	struct inet6_skb_parm *opt = IP6CB(skb);
 	struct ipv6hdr *ipv6h = ipv6_hdr(skb);
+	SKB_DR(reason);
 	int ret;
 
 	if (opt->dsthao) {
···
 	if (hao->length != 16) {
 		net_dbg_ratelimited("hao invalid option length = %d\n",
 				    hao->length);
+		SKB_DR_SET(reason, IP_INHDR);
 		goto discard;
 	}
 
 	if (!(ipv6_addr_type(&hao->addr) & IPV6_ADDR_UNICAST)) {
 		net_dbg_ratelimited("hao is not an unicast addr: %pI6\n",
 				    &hao->addr);
+		SKB_DR_SET(reason, INVALID_PROTO);
 		goto discard;
 	}
 
 	ret = xfrm6_input_addr(skb, (xfrm_address_t *)&ipv6h->daddr,
 			       (xfrm_address_t *)&hao->addr, IPPROTO_DSTOPTS);
-	if (unlikely(ret < 0))
+	if (unlikely(ret < 0)) {
+		SKB_DR_SET(reason, XFRM_POLICY);
 		goto discard;
+	}
 
 	if (skb_cloned(skb)) {
 		if (pskb_expand_head(skb, 0, 0, GFP_ATOMIC))
···
 	return true;
 
  discard:
-	kfree_skb(skb);
+	kfree_skb_reason(skb, reason);
 	return false;
 }
 #endif
···
 	}
 	net_dbg_ratelimited("ipv6_hop_ra: wrong RA length %d\n",
 			    nh[optoff + 1]);
-	kfree_skb(skb);
+	kfree_skb_reason(skb, SKB_DROP_REASON_IP_INHDR);
 	return false;
 }
 
···
 	return true;
 
 drop:
-	kfree_skb(skb);
+	kfree_skb_reason(skb, SKB_DROP_REASON_IP_INHDR);
 	return false;
 }
 
···
 static bool ipv6_hop_jumbo(struct sk_buff *skb, int optoff)
 {
 	const unsigned char *nh = skb_network_header(skb);
-	struct inet6_dev *idev = __in6_dev_get_safely(skb->dev);
-	struct net *net = ipv6_skb_net(skb);
+	SKB_DR(reason);
 	u32 pkt_len;
 
 	if (nh[optoff + 1] != 4 || (optoff & 3) != 2) {
 		net_dbg_ratelimited("ipv6_hop_jumbo: wrong jumbo opt length/alignment %d\n",
 				    nh[optoff+1]);
-		__IP6_INC_STATS(net, idev, IPSTATS_MIB_INHDRERRORS);
+		SKB_DR_SET(reason, IP_INHDR);
 		goto drop;
 	}
 
 	pkt_len = ntohl(*(__be32 *)(nh + optoff + 2));
 	if (pkt_len <= IPV6_MAXPLEN) {
-		__IP6_INC_STATS(net, idev, IPSTATS_MIB_INHDRERRORS);
-		icmpv6_param_prob(skb, ICMPV6_HDR_FIELD, optoff+2);
+		icmpv6_param_prob_reason(skb, ICMPV6_HDR_FIELD, optoff + 2,
+					 SKB_DROP_REASON_IP_INHDR);
 		return false;
 	}
 	if (ipv6_hdr(skb)->payload_len) {
-		__IP6_INC_STATS(net, idev, IPSTATS_MIB_INHDRERRORS);
-		icmpv6_param_prob(skb, ICMPV6_HDR_FIELD, optoff);
+		icmpv6_param_prob_reason(skb, ICMPV6_HDR_FIELD, optoff,
+					 SKB_DROP_REASON_IP_INHDR);
 		return false;
 	}
 
 	if (pkt_len > skb->len - sizeof(struct ipv6hdr)) {
-		__IP6_INC_STATS(net, idev, IPSTATS_MIB_INTRUNCATEDPKTS);
+		SKB_DR_SET(reason, PKT_TOO_SMALL);
 		goto drop;
 	}
 
···
 	return true;
 
 drop:
-	kfree_skb(skb);
+	kfree_skb_reason(skb, reason);
 	return false;
 }
 
···
 	return true;
 
 drop:
-	kfree_skb(skb);
+	kfree_skb_reason(skb, SKB_DROP_REASON_IP_INHDR);
 	return false;
 }
 

net/ipv6/icmp.c

···
 }
 EXPORT_SYMBOL(icmp6_send);
 
-/* Slightly more convenient version of icmp6_send.
+/* Slightly more convenient version of icmp6_send with drop reasons.
  */
-void icmpv6_param_prob(struct sk_buff *skb, u8 code, int pos)
+void icmpv6_param_prob_reason(struct sk_buff *skb, u8 code, int pos,
+			      enum skb_drop_reason reason)
 {
 	icmp6_send(skb, ICMPV6_PARAMPROB, code, pos, NULL, IP6CB(skb));
-	kfree_skb(skb);
+	kfree_skb_reason(skb, reason);
 }
 
 /* Generate icmpv6 with type/code ICMPV6_DEST_UNREACH/ICMPV6_ADDR_UNREACH

net/ipv6/ip6_input.c

···
 static struct sk_buff *ip6_rcv_core(struct sk_buff *skb, struct net_device *dev,
 				    struct net *net)
 {
+	enum skb_drop_reason reason;
 	const struct ipv6hdr *hdr;
 	u32 pkt_len;
 	struct inet6_dev *idev;
 
 	if (skb->pkt_type == PACKET_OTHERHOST) {
 		dev_core_stats_rx_otherhost_dropped_inc(skb->dev);
-		kfree_skb(skb);
+		kfree_skb_reason(skb, SKB_DROP_REASON_OTHERHOST);
 		return NULL;
 	}
 
···
 
 	__IP6_UPD_PO_STATS(net, idev, IPSTATS_MIB_IN, skb->len);
 
+	SKB_DR_SET(reason, NOT_SPECIFIED);
 	if ((skb = skb_share_check(skb, GFP_ATOMIC)) == NULL ||
 	    !idev || unlikely(idev->cnf.disable_ipv6)) {
 		__IP6_INC_STATS(net, idev, IPSTATS_MIB_INDISCARDS);
+		if (unlikely(idev->cnf.disable_ipv6))
+			SKB_DR_SET(reason, IPV6DISABLED);
 		goto drop;
 	}
 
···
 
 	hdr = ipv6_hdr(skb);
 
-	if (hdr->version != 6)
+	if (hdr->version != 6) {
+		SKB_DR_SET(reason, UNHANDLED_PROTO);
 		goto err;
+	}
 
 	__IP6_ADD_STATS(net, idev,
 			IPSTATS_MIB_NOECTPKTS +
···
 	if (!ipv6_addr_is_multicast(&hdr->daddr) &&
 	    (skb->pkt_type == PACKET_BROADCAST ||
 	     skb->pkt_type == PACKET_MULTICAST) &&
-	    idev->cnf.drop_unicast_in_l2_multicast)
+	    idev->cnf.drop_unicast_in_l2_multicast) {
+		SKB_DR_SET(reason, UNICAST_IN_L2_MULTICAST);
 		goto err;
+	}
 
 	/* RFC4291 2.7
 	 * Nodes must not originate a packet to a multicast address whose scope
···
 		if (pkt_len + sizeof(struct ipv6hdr) > skb->len) {
 			__IP6_INC_STATS(net,
 					idev, IPSTATS_MIB_INTRUNCATEDPKTS);
+			SKB_DR_SET(reason, PKT_TOO_SMALL);
 			goto drop;
 		}
-		if (pskb_trim_rcsum(skb, pkt_len + sizeof(struct ipv6hdr))) {
-			__IP6_INC_STATS(net, idev, IPSTATS_MIB_INHDRERRORS);
-			goto drop;
-		}
+		if (pskb_trim_rcsum(skb, pkt_len + sizeof(struct ipv6hdr)))
+			goto err;
 		hdr = ipv6_hdr(skb);
 	}
 
···
 	return skb;
 err:
 	__IP6_INC_STATS(net, idev, IPSTATS_MIB_INHDRERRORS);
+	SKB_DR_OR(reason, IP_INHDR);
 drop:
 	rcu_read_unlock();
-	kfree_skb(skb);
+	kfree_skb_reason(skb, reason);
 	return NULL;
 }
 
···
 	const struct inet6_protocol *ipprot;
 	struct inet6_dev *idev;
 	unsigned int nhoff;
+	SKB_DR(reason);
 	bool raw;
 
 	/*
···
 			if (ipv6_addr_is_multicast(&hdr->daddr) &&
 			    !ipv6_chk_mcast_addr(dev, &hdr->daddr,
 						 &hdr->saddr) &&
-			    !ipv6_is_mld(skb, nexthdr, skb_network_header_len(skb)))
+			    !ipv6_is_mld(skb, nexthdr, skb_network_header_len(skb))) {
+				SKB_DR_SET(reason, IP_INADDRERRORS);
 				goto discard;
+			}
 		}
 		if (!(ipprot->flags & INET6_PROTO_NOPOLICY) &&
-		    !xfrm6_policy_check(NULL, XFRM_POLICY_IN, skb))
+		    !xfrm6_policy_check(NULL, XFRM_POLICY_IN, skb)) {
+			SKB_DR_SET(reason, XFRM_POLICY);
 			goto discard;
+		}
 
 		ret = INDIRECT_CALL_2(ipprot->handler, tcp_v6_rcv, udpv6_rcv,
 				      skb);
···
 						IPSTATS_MIB_INUNKNOWNPROTOS);
 				icmpv6_send(skb, ICMPV6_PARAMPROB,
 					    ICMPV6_UNK_NEXTHDR, nhoff);
+				SKB_DR_SET(reason, IP_NOPROTO);
+			} else {
+				SKB_DR_SET(reason, XFRM_POLICY);
 			}
-			kfree_skb(skb);
+			kfree_skb_reason(skb, reason);
 		} else {
 			__IP6_INC_STATS(net, idev, IPSTATS_MIB_INDELIVERS);
 			consume_skb(skb);
···
 
 discard:
 	__IP6_INC_STATS(net, idev, IPSTATS_MIB_INDISCARDS);
-	kfree_skb(skb);
+	kfree_skb_reason(skb, reason);
 }
 
 static int ip6_input_finish(struct net *net, struct sock *sk, struct sk_buff *skb)

net/ipv6/ip6_output.c

···
 	struct inet6_skb_parm *opt = IP6CB(skb);
 	struct net *net = dev_net(dst->dev);
 	struct inet6_dev *idev;
+	SKB_DR(reason);
 	u32 mtu;
 
 	idev = __in6_dev_get_safely(dev_get_by_index_rcu(net, IP6CB(skb)->iif));
···
 		icmpv6_send(skb, ICMPV6_TIME_EXCEED, ICMPV6_EXC_HOPLIMIT, 0);
 		__IP6_INC_STATS(net, idev, IPSTATS_MIB_INHDRERRORS);
 
-		kfree_skb(skb);
+		kfree_skb_reason(skb, SKB_DROP_REASON_IP_INHDR);
 		return -ETIMEDOUT;
 	}
 
···
 
 	if (!xfrm6_route_forward(skb)) {
 		__IP6_INC_STATS(net, idev, IPSTATS_MIB_INDISCARDS);
+		SKB_DR_SET(reason, XFRM_POLICY);
 		goto drop;
 	}
 	dst = skb_dst(skb);
···
 		__IP6_INC_STATS(net, idev, IPSTATS_MIB_INTOOBIGERRORS);
 		__IP6_INC_STATS(net, ip6_dst_idev(dst),
 				IPSTATS_MIB_FRAGFAILS);
-		kfree_skb(skb);
+		kfree_skb_reason(skb, SKB_DROP_REASON_PKT_TOO_BIG);
 		return -EMSGSIZE;
 	}
 
···
 
 error:
 	__IP6_INC_STATS(net, idev, IPSTATS_MIB_INADDRERRORS);
+	SKB_DR_SET(reason, IP_INADDRERRORS);
 drop:
-	kfree_skb(skb);
+	kfree_skb_reason(skb, reason);
 	return -EINVAL;
 }
 

net/ipv6/route.c

···
 	struct dst_entry *dst = skb_dst(skb);
 	struct net *net = dev_net(dst->dev);
 	struct inet6_dev *idev;
+	SKB_DR(reason);
 	int type;
 
 	if (netif_is_l3_master(skb->dev) ||
···
 	case IPSTATS_MIB_INNOROUTES:
 		type = ipv6_addr_type(&ipv6_hdr(skb)->daddr);
 		if (type == IPV6_ADDR_ANY) {
+			SKB_DR_SET(reason, IP_INADDRERRORS);
 			IP6_INC_STATS(net, idev, IPSTATS_MIB_INADDRERRORS);
 			break;
 		}
+		SKB_DR_SET(reason, IP_INNOROUTES);
 		fallthrough;
 	case IPSTATS_MIB_OUTNOROUTES:
+		SKB_DR_OR(reason, IP_OUTNOROUTES);
 		IP6_INC_STATS(net, idev, ipstats_mib_noroutes);
 		break;
 	}
···
 		skb_dst_drop(skb);
 
 	icmpv6_send(skb, ICMPV6_DEST_UNREACH, code, 0);
-	kfree_skb(skb);
+	kfree_skb_reason(skb, reason);
 	return 0;
 }