net: geneve: support IPv4/IPv6 as inner protocol

[ Upstream commit 435fe1c0c1f74b682dba85641406abf4337aade6 ]

This patch adds support for encapsulating IPv4/IPv6 within GENEVE.

In order to use this, a new IFLA_GENEVE_INNER_PROTO_INHERIT flag needs
to be provided at device creation. This property cannot be changed for
the time being.

In case IP traffic is received on a non-tun device the drop count is
increased.

Signed-off-by: Eyal Birger <eyal.birger@gmail.com>
Link: https://lore.kernel.org/r/20220316061557.431872-1-eyal.birger@gmail.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Stable-dep-of: c471236b2359 ("bareudp: Pull inner IP header on xmit.")
Signed-off-by: Sasha Levin <sashal@kernel.org>

diff --git a/drivers/net/geneve.c b/drivers/net/geneve.c
index 08b479f04ed060ce84557dc36c40ede8af56aea4..88c3805978f2c6e4f7653b869d6c6f558c86cec3 100644 (file)
--- a/drivers/net/geneve.c
+++ b/drivers/net/geneve.c
@@ -54,6 +54,7 @@ struct geneve_config {
        bool                    use_udp6_rx_checksums;
        bool                    ttl_inherit;
        enum ifla_geneve_df     df;
+       bool                    inner_proto_inherit;
 };
 
 /* Pseudo network device */
@@ -249,17 +250,24 @@ static void geneve_rx(struct geneve_dev *geneve, struct geneve_sock *gs,
                }
        }
 
-       skb_reset_mac_header(skb);
-       skb->protocol = eth_type_trans(skb, geneve->dev);
-       skb_postpull_rcsum(skb, eth_hdr(skb), ETH_HLEN);
-
        if (tun_dst)
                skb_dst_set(skb, &tun_dst->dst);
 
-       /* Ignore packet loops (and multicast echo) */
-       if (ether_addr_equal(eth_hdr(skb)->h_source, geneve->dev->dev_addr)) {
-               geneve->dev->stats.rx_errors++;
-               goto drop;
+       if (gnvh->proto_type == htons(ETH_P_TEB)) {
+               skb_reset_mac_header(skb);
+               skb->protocol = eth_type_trans(skb, geneve->dev);
+               skb_postpull_rcsum(skb, eth_hdr(skb), ETH_HLEN);
+
+               /* Ignore packet loops (and multicast echo) */
+               if (ether_addr_equal(eth_hdr(skb)->h_source,
+                                    geneve->dev->dev_addr)) {
+                       geneve->dev->stats.rx_errors++;
+                       goto drop;
+               }
+       } else {
+               skb_reset_mac_header(skb);
+               skb->dev = geneve->dev;
+               skb->pkt_type = PACKET_HOST;
        }
 
        /* Save offset of outer header relative to skb->head,
@@ -357,6 +365,7 @@ static int geneve_udp_encap_recv(struct sock *sk, struct sk_buff *skb)
        struct genevehdr *geneveh;
        struct geneve_dev *geneve;
        struct geneve_sock *gs;
+       __be16 inner_proto;
        int opts_len;
 
        /* Need UDP and Geneve header to be present */
@@ -368,7 +377,11 @@ static int geneve_udp_encap_recv(struct sock *sk, struct sk_buff *skb)
        if (unlikely(geneveh->ver != GENEVE_VER))
                goto drop;
 
-       if (unlikely(geneveh->proto_type != htons(ETH_P_TEB)))
+       inner_proto = geneveh->proto_type;
+
+       if (unlikely((inner_proto != htons(ETH_P_TEB) &&
+                     inner_proto != htons(ETH_P_IP) &&
+                     inner_proto != htons(ETH_P_IPV6))))
                goto drop;
 
        gs = rcu_dereference_sk_user_data(sk);
@@ -379,9 +392,14 @@ static int geneve_udp_encap_recv(struct sock *sk, struct sk_buff *skb)
        if (!geneve)
                goto drop;
 
+       if (unlikely((!geneve->cfg.inner_proto_inherit &&
+                     inner_proto != htons(ETH_P_TEB)))) {
+               geneve->dev->stats.rx_dropped++;
+               goto drop;
+       }
+
        opts_len = geneveh->opt_len * 4;
-       if (iptunnel_pull_header(skb, GENEVE_BASE_HLEN + opts_len,
-                                htons(ETH_P_TEB),
+       if (iptunnel_pull_header(skb, GENEVE_BASE_HLEN + opts_len, inner_proto,
                                 !net_eq(geneve->net, dev_net(geneve->dev)))) {
                geneve->dev->stats.rx_dropped++;
                goto drop;
@@ -728,7 +746,8 @@ static int geneve_stop(struct net_device *dev)
 }
 
 static void geneve_build_header(struct genevehdr *geneveh,
-                               const struct ip_tunnel_info *info)
+                               const struct ip_tunnel_info *info,
+                               __be16 inner_proto)
 {
        geneveh->ver = GENEVE_VER;
        geneveh->opt_len = info->options_len / 4;
@@ -736,7 +755,7 @@ static void geneve_build_header(struct genevehdr *geneveh,
        geneveh->critical = !!(info->key.tun_flags & TUNNEL_CRIT_OPT);
        geneveh->rsvd1 = 0;
        tunnel_id_to_vni(info->key.tun_id, geneveh->vni);
-       geneveh->proto_type = htons(ETH_P_TEB);
+       geneveh->proto_type = inner_proto;
        geneveh->rsvd2 = 0;
 
        if (info->key.tun_flags & TUNNEL_GENEVE_OPT)
@@ -745,10 +764,12 @@ static void geneve_build_header(struct genevehdr *geneveh,
 
 static int geneve_build_skb(struct dst_entry *dst, struct sk_buff *skb,
                            const struct ip_tunnel_info *info,
-                           bool xnet, int ip_hdr_len)
+                           bool xnet, int ip_hdr_len,
+                           bool inner_proto_inherit)
 {
        bool udp_sum = !!(info->key.tun_flags & TUNNEL_CSUM);
        struct genevehdr *gnvh;
+       __be16 inner_proto;
        int min_headroom;
        int err;
 
@@ -766,8 +787,9 @@ static int geneve_build_skb(struct dst_entry *dst, struct sk_buff *skb,
                goto free_dst;
 
        gnvh = __skb_push(skb, sizeof(*gnvh) + info->options_len);
-       geneve_build_header(gnvh, info);
-       skb_set_inner_protocol(skb, htons(ETH_P_TEB));
+       inner_proto = inner_proto_inherit ? skb->protocol : htons(ETH_P_TEB);
+       geneve_build_header(gnvh, info, inner_proto);
+       skb_set_inner_protocol(skb, inner_proto);
        return 0;
 
 free_dst:
@@ -973,7 +995,8 @@ static int geneve_xmit_skb(struct sk_buff *skb, struct net_device *dev,
                }
        }
 
-       err = geneve_build_skb(&rt->dst, skb, info, xnet, sizeof(struct iphdr));
+       err = geneve_build_skb(&rt->dst, skb, info, xnet, sizeof(struct iphdr),
+                              geneve->cfg.inner_proto_inherit);
        if (unlikely(err))
                return err;
 
@@ -1052,7 +1075,8 @@ static int geneve6_xmit_skb(struct sk_buff *skb, struct net_device *dev,
                        ttl = key->ttl;
                ttl = ttl ? : ip6_dst_hoplimit(dst);
        }
-       err = geneve_build_skb(dst, skb, info, xnet, sizeof(struct ipv6hdr));
+       err = geneve_build_skb(dst, skb, info, xnet, sizeof(struct ipv6hdr),
+                              geneve->cfg.inner_proto_inherit);
        if (unlikely(err))
                return err;
 
@@ -1401,6 +1425,14 @@ static int geneve_configure(struct net *net, struct net_device *dev,
        dst_cache_reset(&geneve->cfg.info.dst_cache);
        memcpy(&geneve->cfg, cfg, sizeof(*cfg));
 
+       if (geneve->cfg.inner_proto_inherit) {
+               dev->header_ops = NULL;
+               dev->type = ARPHRD_NONE;
+               dev->hard_header_len = 0;
+               dev->addr_len = 0;
+               dev->flags = IFF_NOARP;
+       }
+
        err = register_netdevice(dev);
        if (err)
                return err;
@@ -1574,10 +1606,18 @@ static int geneve_nl2info(struct nlattr *tb[], struct nlattr *data[],
 #endif
        }
 
+       if (data[IFLA_GENEVE_INNER_PROTO_INHERIT]) {
+               if (changelink) {
+                       attrtype = IFLA_GENEVE_INNER_PROTO_INHERIT;
+                       goto change_notsup;
+               }
+               cfg->inner_proto_inherit = true;
+       }
+
        return 0;
 change_notsup:
        NL_SET_ERR_MSG_ATTR(extack, data[attrtype],
-                           "Changing VNI, Port, endpoint IP address family, external, and UDP checksum attributes are not supported");
+                           "Changing VNI, Port, endpoint IP address family, external, inner_proto_inherit, and UDP checksum attributes are not supported");
        return -EOPNOTSUPP;
 }
 
@@ -1812,6 +1852,10 @@ static int geneve_fill_info(struct sk_buff *skb, const struct net_device *dev)
        if (nla_put_u8(skb, IFLA_GENEVE_TTL_INHERIT, ttl_inherit))
                goto nla_put_failure;
 
+       if (geneve->cfg.inner_proto_inherit &&
+           nla_put_flag(skb, IFLA_GENEVE_INNER_PROTO_INHERIT))
+               goto nla_put_failure;
+
        return 0;
 
 nla_put_failure:

diff --git a/include/uapi/linux/if_link.h b/include/uapi/linux/if_link.h
index c4b23f06f69e0f503585c99d422969086821ce2d..9334f2128bb2e412abbd4c6841fa9154e00e67ed 100644 (file)
--- a/include/uapi/linux/if_link.h
+++ b/include/uapi/linux/if_link.h
@@ -761,6 +761,7 @@ enum {
        IFLA_GENEVE_LABEL,
        IFLA_GENEVE_TTL_INHERIT,
        IFLA_GENEVE_DF,
+       IFLA_GENEVE_INNER_PROTO_INHERIT,
        __IFLA_GENEVE_MAX
 };
 #define IFLA_GENEVE_MAX        (__IFLA_GENEVE_MAX - 1)