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

[ Upstream commit d7bf2ebebc2bd61ab95e2a8e33541ef282f303d4 ]

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: d8b9605d2697 ("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>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

11 files changed, 22 insertions, 19 deletions

diff --git a/net/sched/act_connmark.c b/net/sched/act_connmark.c
index 538dedd84e21..6485f421c091 100644
--- a/net/sched/act_connmark.c
+++ b/net/sched/act_connmark.c
@@ -46,17 +46,20 @@ 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;
 	}
 
diff --git a/net/sched/act_csum.c b/net/sched/act_csum.c
index 9ecbf8edcf39..24ad4ceaca27 100644
--- a/net/sched/act_csum.c
+++ b/net/sched/act_csum.c
@@ -577,7 +577,7 @@ 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):
diff --git a/net/sched/act_skbedit.c b/net/sched/act_skbedit.c
index a80179c1075f..06648b79968e 100644
--- a/net/sched/act_skbedit.c
+++ b/net/sched/act_skbedit.c
@@ -51,7 +51,7 @@ 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))
diff --git a/net/sched/cls_api.c b/net/sched/cls_api.c
index e217ebc693f8..184c20b86393 100644
--- a/net/sched/cls_api.c
+++ b/net/sched/cls_api.c
@@ -969,7 +969,7 @@ int tcf_classify(struct sk_buff *skb, const struct tcf_proto *tp,
 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 &&
diff --git a/net/sched/cls_flow.c b/net/sched/cls_flow.c
index 2bb043cd436b..55bf75cb1f16 100644
--- a/net/sched/cls_flow.c
+++ b/net/sched/cls_flow.c
@@ -84,7 +84,7 @@ 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,
@@ -108,7 +108,7 @@ 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)
@@ -155,7 +155,7 @@ 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):
@@ -168,7 +168,7 @@ 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):
diff --git a/net/sched/cls_flower.c b/net/sched/cls_flower.c
index 44ca31f8538d..208436eb107c 100644
--- a/net/sched/cls_flower.c
+++ b/net/sched/cls_flower.c
@@ -203,7 +203,7 @@ 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(skb, &mask->dissector, &skb_key, 0);
 
diff --git a/net/sched/em_ipset.c b/net/sched/em_ipset.c
index c1b23e3060b8..ef3b6b66c26a 100644
--- a/net/sched/em_ipset.c
+++ b/net/sched/em_ipset.c
@@ -62,7 +62,7 @@ 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)))
diff --git a/net/sched/em_meta.c b/net/sched/em_meta.c
index d6e97115500b..e36fa9272259 100644
--- a/net/sched/em_meta.c
+++ b/net/sched/em_meta.c
@@ -199,7 +199,7 @@ 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)
diff --git a/net/sched/sch_cake.c b/net/sched/sch_cake.c
index d03f843647ae..32712e7dcbdc 100644
--- a/net/sched/sch_cake.c
+++ b/net/sched/sch_cake.c
@@ -589,7 +589,7 @@ static void cake_update_flowkeys(struct flow_keys *keys,
 	struct nf_conntrack_tuple tuple = {};
 	bool rev = !skb->_nfct;
 
-	if (tc_skb_protocol(skb) != htons(ETH_P_IP))
+	if (skb_protocol(skb, true) != htons(ETH_P_IP))
 		return;
 
 	if (!nf_ct_get_tuple_skb(&tuple, skb))
@@ -1514,7 +1514,7 @@ 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))
diff --git a/net/sched/sch_dsmark.c b/net/sched/sch_dsmark.c
index 84c948c91914..1c2fce8c45b2 100644
--- a/net/sched/sch_dsmark.c
+++ b/net/sched/sch_dsmark.c
@@ -207,7 +207,7 @@ 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) ||
@@ -300,7 +300,7 @@ 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);
@@ -317,7 +317,7 @@ 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;
 	}
 
diff --git a/net/sched/sch_teql.c b/net/sched/sch_teql.c
index 93f04cf5cac1..7ed53124069c 100644
--- a/net/sched/sch_teql.c
+++ b/net/sched/sch_teql.c
@@ -243,7 +243,7 @@ __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)