sched: consistently handle layer3 header accesses in the presence of VLANs

There are a couple of places in net/sched/ that check skb->protocol and act
on the value there. However, in the presence of VLAN tags, the value stored
in skb->protocol can be inconsistent based on whether VLAN acceleration is
enabled. The commit quoted in the Fixes tag below fixed the users of
skb->protocol to use a helper that will always see the VLAN ethertype.

However, most of the callers don't actually handle the VLAN ethertype, but
expect to find the IP header type in the protocol field. This means that
things like changing the ECN field, or parsing diffserv values, stops
working if there's a VLAN tag, or if there are multiple nested VLAN
tags (QinQ).

To fix this, change the helper to take an argument that indicates whether
the caller wants to skip the VLAN tags or not. When skipping VLAN tags, we
make sure to skip all of them, so behaviour is consistent even in QinQ
mode.

To make the helper usable from the ECN code, move it to if_vlan.h instead
of pkt_sched.h.

v3:
- Remove empty lines
- Move vlan variable definitions inside loop in skb_protocol()
- Also use skb_protocol() helper in IP{,6}_ECN_decapsulate() and
  bpf_skb_ecn_set_ce()

v2:
- Use eth_type_vlan() helper in skb_protocol()
- Also fix code that reads skb->protocol directly
- Change a couple of 'if/else if' statements to switch constructs to avoid
  calling the helper twice

Reported-by: Ilya Ponetayev <i.ponetaev@ndmsystems.com>
Fixes: d8b9605d26 ("net: sched: fix skb->protocol use in case of accelerated vlan path")
Signed-off-by: Toke Høiland-Jørgensen <toke@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

net/sched/act_connmark.c

@@ -43,17 +43,20 @@ static int tcf_connmark_act(struct sk_buff *skb, const struct tc_action *a,
 	tcf_lastuse_update(&ca->tcf_tm);
 	bstats_update(&ca->tcf_bstats, skb);
 
-	if (skb->protocol == htons(ETH_P_IP)) {
+	switch (skb_protocol(skb, true)) {
+	case htons(ETH_P_IP):
 		if (skb->len < sizeof(struct iphdr))
 			goto out;
 
 		proto = NFPROTO_IPV4;
-	} else if (skb->protocol == htons(ETH_P_IPV6)) {
+		break;
+	case htons(ETH_P_IPV6):
 		if (skb->len < sizeof(struct ipv6hdr))
 			goto out;
 
 		proto = NFPROTO_IPV6;
-	} else {
+		break;
+	default:
 		goto out;
 	}
 

net/sched/act_csum.c

@@ -587,7 +587,7 @@ static int tcf_csum_act(struct sk_buff *skb, const struct tc_action *a,
 		goto drop;
 
 	update_flags = params->update_flags;
-	protocol = tc_skb_protocol(skb);
+	protocol = skb_protocol(skb, false);
 again:
 	switch (protocol) {
 	case cpu_to_be16(ETH_P_IP):

net/sched/act_ct.c

@@ -624,7 +624,7 @@ static u8 tcf_ct_skb_nf_family(struct sk_buff *skb)
 {
 	u8 family = NFPROTO_UNSPEC;
 
-	switch (skb->protocol) {
+	switch (skb_protocol(skb, true)) {
 	case htons(ETH_P_IP):
 		family = NFPROTO_IPV4;
 		break;
@@ -748,6 +748,7 @@ static int ct_nat_execute(struct sk_buff *skb, struct nf_conn *ct,
 			  const struct nf_nat_range2 *range,
 			  enum nf_nat_manip_type maniptype)
 {
+	__be16 proto = skb_protocol(skb, true);
 	int hooknum, err = NF_ACCEPT;
 
 	/* See HOOK2MANIP(). */
@@ -759,14 +760,13 @@ static int ct_nat_execute(struct sk_buff *skb, struct nf_conn *ct,
 	switch (ctinfo) {
 	case IP_CT_RELATED:
 	case IP_CT_RELATED_REPLY:
-		if (skb->protocol == htons(ETH_P_IP) &&
+		if (proto == htons(ETH_P_IP) &&
 		    ip_hdr(skb)->protocol == IPPROTO_ICMP) {
 			if (!nf_nat_icmp_reply_translation(skb, ct, ctinfo,
 							   hooknum))
 				err = NF_DROP;
 			goto out;
-		} else if (IS_ENABLED(CONFIG_IPV6) &&
-			   skb->protocol == htons(ETH_P_IPV6)) {
+		} else if (IS_ENABLED(CONFIG_IPV6) && proto == htons(ETH_P_IPV6)) {
 			__be16 frag_off;
 			u8 nexthdr = ipv6_hdr(skb)->nexthdr;
 			int hdrlen = ipv6_skip_exthdr(skb,
@@ -1550,4 +1550,3 @@ MODULE_AUTHOR("Yossi Kuperman <yossiku@mellanox.com>");
 MODULE_AUTHOR("Marcelo Ricardo Leitner <marcelo.leitner@gmail.com>");
 MODULE_DESCRIPTION("Connection tracking action");
 MODULE_LICENSE("GPL v2");
-

net/sched/act_ctinfo.c

@@ -96,19 +96,22 @@ static int tcf_ctinfo_act(struct sk_buff *skb, const struct tc_action *a,
 	action = READ_ONCE(ca->tcf_action);
 
 	wlen = skb_network_offset(skb);
-	if (tc_skb_protocol(skb) == htons(ETH_P_IP)) {
+	switch (skb_protocol(skb, true)) {
+	case htons(ETH_P_IP):
 		wlen += sizeof(struct iphdr);
 		if (!pskb_may_pull(skb, wlen))
 			goto out;
 
 		proto = NFPROTO_IPV4;
-	} else if (tc_skb_protocol(skb) == htons(ETH_P_IPV6)) {
+		break;
+	case htons(ETH_P_IPV6):
 		wlen += sizeof(struct ipv6hdr);
 		if (!pskb_may_pull(skb, wlen))
 			goto out;
 
 		proto = NFPROTO_IPV6;
-	} else {
+		break;
+	default:
 		goto out;
 	}
 

net/sched/act_mpls.c

@@ -82,7 +82,7 @@ static int tcf_mpls_act(struct sk_buff *skb, const struct tc_action *a,
 			goto drop;
 		break;
 	case TCA_MPLS_ACT_PUSH:
-		new_lse = tcf_mpls_get_lse(NULL, p, !eth_p_mpls(skb->protocol));
+		new_lse = tcf_mpls_get_lse(NULL, p, !eth_p_mpls(skb_protocol(skb, true)));
 		if (skb_mpls_push(skb, new_lse, p->tcfm_proto, mac_len,
 				  skb->dev && skb->dev->type == ARPHRD_ETHER))
 			goto drop;

net/sched/act_skbedit.c

@@ -41,7 +41,7 @@ static int tcf_skbedit_act(struct sk_buff *skb, const struct tc_action *a,
 	if (params->flags & SKBEDIT_F_INHERITDSFIELD) {
 		int wlen = skb_network_offset(skb);
 
-		switch (tc_skb_protocol(skb)) {
+		switch (skb_protocol(skb, true)) {
 		case htons(ETH_P_IP):
 			wlen += sizeof(struct iphdr);
 			if (!pskb_may_pull(skb, wlen))

net/sched/cls_api.c

@@ -1538,7 +1538,7 @@ static inline int __tcf_classify(struct sk_buff *skb,
 reclassify:
 #endif
 	for (; tp; tp = rcu_dereference_bh(tp->next)) {
-		__be16 protocol = tc_skb_protocol(skb);
+		__be16 protocol = skb_protocol(skb, false);
 		int err;
 
 		if (tp->protocol != protocol &&

net/sched/cls_flow.c

@@ -80,7 +80,7 @@ static u32 flow_get_dst(const struct sk_buff *skb, const struct flow_keys *flow)
 	if (dst)
 		return ntohl(dst);
 
-	return addr_fold(skb_dst(skb)) ^ (__force u16) tc_skb_protocol(skb);
+	return addr_fold(skb_dst(skb)) ^ (__force u16)skb_protocol(skb, true);
 }
 
 static u32 flow_get_proto(const struct sk_buff *skb,
@@ -104,7 +104,7 @@ static u32 flow_get_proto_dst(const struct sk_buff *skb,
 	if (flow->ports.ports)
 		return ntohs(flow->ports.dst);
 
-	return addr_fold(skb_dst(skb)) ^ (__force u16) tc_skb_protocol(skb);
+	return addr_fold(skb_dst(skb)) ^ (__force u16)skb_protocol(skb, true);
 }
 
 static u32 flow_get_iif(const struct sk_buff *skb)
@@ -151,7 +151,7 @@ static u32 flow_get_nfct(const struct sk_buff *skb)
 static u32 flow_get_nfct_src(const struct sk_buff *skb,
 			     const struct flow_keys *flow)
 {
-	switch (tc_skb_protocol(skb)) {
+	switch (skb_protocol(skb, true)) {
 	case htons(ETH_P_IP):
 		return ntohl(CTTUPLE(skb, src.u3.ip));
 	case htons(ETH_P_IPV6):
@@ -164,7 +164,7 @@ fallback:
 static u32 flow_get_nfct_dst(const struct sk_buff *skb,
 			     const struct flow_keys *flow)
 {
-	switch (tc_skb_protocol(skb)) {
+	switch (skb_protocol(skb, true)) {
 	case htons(ETH_P_IP):
 		return ntohl(CTTUPLE(skb, dst.u3.ip));
 	case htons(ETH_P_IPV6):

net/sched/cls_flower.c

@@ -313,7 +313,7 @@ static int fl_classify(struct sk_buff *skb, const struct tcf_proto *tp,
 		/* skb_flow_dissect() does not set n_proto in case an unknown
 		 * protocol, so do it rather here.
 		 */
-		skb_key.basic.n_proto = skb->protocol;
+		skb_key.basic.n_proto = skb_protocol(skb, false);
 		skb_flow_dissect_tunnel_info(skb, &mask->dissector, &skb_key);
 		skb_flow_dissect_ct(skb, &mask->dissector, &skb_key,
 				    fl_ct_info_to_flower_map,

net/sched/em_ipset.c

@@ -59,7 +59,7 @@ static int em_ipset_match(struct sk_buff *skb, struct tcf_ematch *em,
 	};
 	int ret, network_offset;
 
-	switch (tc_skb_protocol(skb)) {
+	switch (skb_protocol(skb, true)) {
 	case htons(ETH_P_IP):
 		state.pf = NFPROTO_IPV4;
 		if (!pskb_network_may_pull(skb, sizeof(struct iphdr)))

net/sched/em_ipt.c

@@ -212,7 +212,7 @@ static int em_ipt_match(struct sk_buff *skb, struct tcf_ematch *em,
 	struct nf_hook_state state;
 	int ret;
 
-	switch (tc_skb_protocol(skb)) {
+	switch (skb_protocol(skb, true)) {
 	case htons(ETH_P_IP):
 		if (!pskb_network_may_pull(skb, sizeof(struct iphdr)))
 			return 0;

net/sched/em_meta.c

@@ -195,7 +195,7 @@ META_COLLECTOR(int_priority)
 META_COLLECTOR(int_protocol)
 {
 	/* Let userspace take care of the byte ordering */
-	dst->value = tc_skb_protocol(skb);
+	dst->value = skb_protocol(skb, false);
 }
 
 META_COLLECTOR(int_pkttype)

net/sched/sch_cake.c

@@ -592,7 +592,7 @@ static bool cake_update_flowkeys(struct flow_keys *keys,
 	bool rev = !skb->_nfct, upd = false;
 	__be32 ip;
 
-	if (tc_skb_protocol(skb) != htons(ETH_P_IP))
+	if (skb_protocol(skb, true) != htons(ETH_P_IP))
 		return false;
 
 	if (!nf_ct_get_tuple_skb(&tuple, skb))
@@ -1557,7 +1557,7 @@ static u8 cake_handle_diffserv(struct sk_buff *skb, bool wash)
 	u16 *buf, buf_;
 	u8 dscp;
 
-	switch (tc_skb_protocol(skb)) {
+	switch (skb_protocol(skb, true)) {
 	case htons(ETH_P_IP):
 		buf = skb_header_pointer(skb, offset, sizeof(buf_), &buf_);
 		if (unlikely(!buf))

net/sched/sch_dsmark.c

@@ -210,7 +210,7 @@ static int dsmark_enqueue(struct sk_buff *skb, struct Qdisc *sch,
 	if (p->set_tc_index) {
 		int wlen = skb_network_offset(skb);
 
-		switch (tc_skb_protocol(skb)) {
+		switch (skb_protocol(skb, true)) {
 		case htons(ETH_P_IP):
 			wlen += sizeof(struct iphdr);
 			if (!pskb_may_pull(skb, wlen) ||
@@ -303,7 +303,7 @@ static struct sk_buff *dsmark_dequeue(struct Qdisc *sch)
 	index = skb->tc_index & (p->indices - 1);
 	pr_debug("index %d->%d\n", skb->tc_index, index);
 
-	switch (tc_skb_protocol(skb)) {
+	switch (skb_protocol(skb, true)) {
 	case htons(ETH_P_IP):
 		ipv4_change_dsfield(ip_hdr(skb), p->mv[index].mask,
 				    p->mv[index].value);
@@ -320,7 +320,7 @@ static struct sk_buff *dsmark_dequeue(struct Qdisc *sch)
 		 */
 		if (p->mv[index].mask != 0xff || p->mv[index].value)
 			pr_warn("%s: unsupported protocol %d\n",
-				__func__, ntohs(tc_skb_protocol(skb)));
+				__func__, ntohs(skb_protocol(skb, true)));
 		break;
 	}
 

net/sched/sch_teql.c

@@ -239,7 +239,7 @@ __teql_resolve(struct sk_buff *skb, struct sk_buff *skb_res,
 		char haddr[MAX_ADDR_LEN];
 
 		neigh_ha_snapshot(haddr, n, dev);
-		err = dev_hard_header(skb, dev, ntohs(tc_skb_protocol(skb)),
+		err = dev_hard_header(skb, dev, ntohs(skb_protocol(skb, false)),
 				      haddr, NULL, skb->len);
 
 		if (err < 0)