From 9f9818f8c1cf44055634297247620be4755e7af2 Mon Sep 17 00:00:00 2001 From: Jesse Gross Date: Sat, 19 Mar 2016 09:32:02 -0700 Subject: tunnels: Remove encapsulation offloads on decap. commit a09a4c8dd1ec7f830e1fb9e59eb72bddc965d168 upstream. If a packet is either locally encapsulated or processed through GRO it is marked with the offloads that it requires. However, when it is decapsulated these tunnel offload indications are not removed. This means that if we receive an encapsulated TCP packet, aggregate it with GRO, decapsulate, and retransmit the resulting frame on a NIC that does not support encapsulation, we won't be able to take advantage of hardware offloads even though it is just a simple TCP packet at this point. This fixes the problem by stripping off encapsulation offload indications when packets are decapsulated. The performance impacts of this bug are significant. In a test where a Geneve encapsulated TCP stream is sent to a hypervisor, GRO'ed, decapsulated, and bridged to a VM performance is improved by 60% (5Gbps->8Gbps) as a result of avoiding unnecessary segmentation at the VM tap interface. Reported-by: Ramu Ramamurthy Fixes: 68c33163 ("v4 GRE: Add TCP segmentation offload for GRE") Signed-off-by: Jesse Gross Signed-off-by: David S. Miller (backported from commit a09a4c8dd1ec7f830e1fb9e59eb72bddc965d168) [adapt iptunnel_pull_header arguments, avoid 7f290c9] Signed-off-by: Stefan Bader Signed-off-by: Juerg Haefliger Signed-off-by: Greg Kroah-Hartman --- include/net/ip_tunnels.h | 16 ++++++++++++++++ 1 file changed, 16 insertions(+) (limited to 'include/net') diff --git a/include/net/ip_tunnels.h b/include/net/ip_tunnels.h index af40bc586a1b..86a7bdd61d1a 100644 --- a/include/net/ip_tunnels.h +++ b/include/net/ip_tunnels.h @@ -283,6 +283,22 @@ struct metadata_dst *iptunnel_metadata_reply(struct metadata_dst *md, struct sk_buff *iptunnel_handle_offloads(struct sk_buff *skb, bool gre_csum, int gso_type_mask); +static inline int iptunnel_pull_offloads(struct sk_buff *skb) +{ + if (skb_is_gso(skb)) { + int err; + + err = skb_unclone(skb, GFP_ATOMIC); + if (unlikely(err)) + return err; + skb_shinfo(skb)->gso_type &= ~(NETIF_F_GSO_ENCAP_ALL >> + NETIF_F_GSO_SHIFT); + } + + skb->encapsulation = 0; + return 0; +} + static inline void iptunnel_xmit_stats(int err, struct net_device_stats *err_stats, struct pcpu_sw_netstats __percpu *stats) -- cgit v1.2.1