On Sat, Feb 6, 2016 at 2:55 PM, Tom Herbert <t...@herbertland.com> wrote:
> On Fri, Feb 5, 2016 at 3:28 PM, Alexander Duyck <adu...@mirantis.com> wrote:
>> Instead of parsing headers to determine the inner protocol we can just pull
>> the value from inner_proto.
>>
>> Signed-off-by: Alexander Duyck <adu...@mirantis.com>
>> ---
>> net/ipv4/gre_offload.c | 6 ++----
>> 1 file changed, 2 insertions(+), 4 deletions(-)
>>
>> diff --git a/net/ipv4/gre_offload.c b/net/ipv4/gre_offload.c
>> index c15441b5ff61..003b0ebbcfdd 100644
>> --- a/net/ipv4/gre_offload.c
>> +++ b/net/ipv4/gre_offload.c
>> @@ -20,7 +20,6 @@ static struct sk_buff *gre_gso_segment(struct sk_buff *skb,
>> {
>> int tnl_hlen = skb_inner_mac_header(skb) - skb_transport_header(skb);
>> struct sk_buff *segs = ERR_PTR(-EINVAL);
>> - struct gre_base_hdr *greh;
>> u16 mac_offset = skb->mac_header;
>> __be16 protocol = skb->protocol;
>> u16 mac_len = skb->mac_len;
>> @@ -48,15 +47,13 @@ static struct sk_buff *gre_gso_segment(struct sk_buff
>> *skb,
>> if (unlikely(!pskb_may_pull(skb, tnl_hlen)))
>> goto out;
>>
>> - greh = (struct gre_base_hdr *)skb_transport_header(skb);
>> -
>> /* setup inner skb. */
>> - skb->protocol = greh->protocol;
>> skb->encapsulation = 0;
>> __skb_pull(skb, tnl_hlen);
>> skb_reset_mac_header(skb);
>> skb_set_network_header(skb, skb_inner_network_offset(skb));
>> skb->mac_len = skb_inner_network_offset(skb);
>> + skb->protocol = skb->inner_protocol;
>
> Is this going to be a problem if we try to perform GSO with nested
> encapsulations? (I suppose that's something not supported).
Right tunnels don't offload tunnels so there is no need to worry about
a tunnel nested in a tunnel. We might get there someday, but I don't
see it happening any time soon because we would end up likely needing
yet another level of header pointers for middle mac, network, and
transport in such a case.
- Alex
