| Message ID | 1415648713-110483-1-git-send-email-jesse@nicira.com |
|---|---|
| State | Accepted, archived |
| Delegated to: | David Miller |
| Headers | show |
From: Jesse Gross <jesse@nicira.com> Date: Mon, 10 Nov 2014 11:45:13 -0800 > When doing GRO processing for UDP tunnels, we never add > SKB_GSO_UDP_TUNNEL to gso_type - only the type of the inner protocol > is added (such as SKB_GSO_TCPV4). The result is that if the packet is > later resegmented we will do GSO but not treat it as a tunnel. This > results in UDP fragmentation of the outer header instead of (i.e.) TCP > segmentation of the inner header as was originally on the wire. > > Signed-off-by: Jesse Gross <jesse@nicira.com> > --- > v2: Update commit message to reflect result of problem. Applied, thanks. -- To unsubscribe from this list: send the line "unsubscribe netdev" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
diff --git a/drivers/net/vxlan.c b/drivers/net/vxlan.c index ca30982..cfb892b 100644 --- a/drivers/net/vxlan.c +++ b/drivers/net/vxlan.c @@ -621,6 +621,8 @@ static int vxlan_gro_complete(struct sk_buff *skb, int nhoff) int vxlan_len = sizeof(struct vxlanhdr) + sizeof(struct ethhdr); int err = -ENOSYS; + udp_tunnel_gro_complete(skb, nhoff); + eh = (struct ethhdr *)(skb->data + nhoff + sizeof(struct vxlanhdr)); type = eh->h_proto; diff --git a/include/net/udp_tunnel.h b/include/net/udp_tunnel.h index a47790b..2a50a70 100644 --- a/include/net/udp_tunnel.h +++ b/include/net/udp_tunnel.h @@ -100,6 +100,15 @@ static inline struct sk_buff *udp_tunnel_handle_offloads(struct sk_buff *skb, return iptunnel_handle_offloads(skb, udp_csum, type); } +static inline void udp_tunnel_gro_complete(struct sk_buff *skb, int nhoff) +{ + struct udphdr *uh; + + uh = (struct udphdr *)(skb->data + nhoff - sizeof(struct udphdr)); + skb_shinfo(skb)->gso_type |= uh->check ? + SKB_GSO_UDP_TUNNEL_CSUM : SKB_GSO_UDP_TUNNEL; +} + static inline void udp_tunnel_encap_enable(struct socket *sock) { #if IS_ENABLED(CONFIG_IPV6) diff --git a/net/ipv4/fou.c b/net/ipv4/fou.c index 32e7892..606c520 100644 --- a/net/ipv4/fou.c +++ b/net/ipv4/fou.c @@ -133,6 +133,8 @@ static int fou_gro_complete(struct sk_buff *skb, int nhoff) int err = -ENOSYS; const struct net_offload **offloads; + udp_tunnel_gro_complete(skb, nhoff); + rcu_read_lock(); offloads = NAPI_GRO_CB(skb)->is_ipv6 ? inet6_offloads : inet_offloads; ops = rcu_dereference(offloads[proto]);
When doing GRO processing for UDP tunnels, we never add SKB_GSO_UDP_TUNNEL to gso_type - only the type of the inner protocol is added (such as SKB_GSO_TCPV4). The result is that if the packet is later resegmented we will do GSO but not treat it as a tunnel. This results in UDP fragmentation of the outer header instead of (i.e.) TCP segmentation of the inner header as was originally on the wire. Signed-off-by: Jesse Gross <jesse@nicira.com> --- v2: Update commit message to reflect result of problem. --- drivers/net/vxlan.c | 2 ++ include/net/udp_tunnel.h | 9 +++++++++ net/ipv4/fou.c | 2 ++ 3 files changed, 13 insertions(+)