On 6/28/25 11:47 AM, Lorenzo Bianconi wrote:
>> On Fri, Jun 27, 2025 at 02:45:28PM +0200, Lorenzo Bianconi wrote:
>>> Introduce SW acceleration for IPIP tunnels in the netfilter flowtable
>>> infrastructure.
>>> IPIP SW acceleration can be tested running the following scenario where
>>> the traffic is forwarded between two NICs (eth0 and eth1) and an IPIP
>>> tunnel is used to access a remote site (using eth1 as the underlay device):
>>>
>>> ETH0 -- TUN0 <==> ETH1 -- [IP network] -- TUN1 (192.168.100.2)
>>>
>>> $ip addr show
>>> 6: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue state UP
>>> group default qlen 1000
>>> link/ether 00:00:22:33:11:55 brd ff:ff:ff:ff:ff:ff
>>> inet 192.168.0.2/24 scope global eth0
>>> valid_lft forever preferred_lft forever
>>> 7: eth1: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue state UP
>>> group default qlen 1000
>>> link/ether 00:11:22:33:11:55 brd ff:ff:ff:ff:ff:ff
>>> inet 192.168.1.1/24 scope global eth1
>>> valid_lft forever preferred_lft forever
>>> 8: tun0@NONE: <POINTOPOINT,NOARP,UP,LOWER_UP> mtu 1480 qdisc noqueue state
>>> UNKNOWN group default qlen 1000
>>> link/ipip 192.168.1.1 peer 192.168.1.2
>>> inet 192.168.100.1/24 scope global tun0
>>> valid_lft forever preferred_lft forever
>>>
>>> $ip route show
>>> default via 192.168.100.2 dev tun0
>>> 192.168.0.0/24 dev eth0 proto kernel scope link src 192.168.0.2
>>> 192.168.1.0/24 dev eth1 proto kernel scope link src 192.168.1.1
>>> 192.168.100.0/24 dev tun0 proto kernel scope link src 192.168.100.1
>>>
>>> $nft list ruleset
>>> table inet filter {
>>> flowtable ft {
>>> hook ingress priority filter
>>> devices = { eth0, eth1 }
>>> }
>>>
>>> chain forward {
>>> type filter hook forward priority filter; policy accept;
>>> meta l4proto { tcp, udp } flow add @ft
>>> }
>>> }
>>
>> Is there a proof that this accelerates forwarding?
>
> I reproduced the scenario described above using veths (something similar to
> what is done in nft_flowtable.sh) and I got the following results:
>
> - flowtable configured as above between the two router interfaces
> - TCP stream between client and server going via the IPIP tunnel
> - TCP stream transmitted into the IPIP tunnel:
> - net-next: ~41Gbps
> - net-next + IPIP flowtbale support: ~40Gbps
> - TCP stream received from the IPIP tunnel:
> - net-next: ~35Gbps
> - net-next + IPIP flowtbale support: ~49Gbps
>
>>
>>> Signed-off-by: Lorenzo Bianconi <[email protected]>
>>> ---
>>> net/ipv4/ipip.c | 21 +++++++++++++++++++++
>>> net/netfilter/nf_flow_table_ip.c | 28 ++++++++++++++++++++++++++--
>>> 2 files changed, 47 insertions(+), 2 deletions(-)
>>>
>
> [...]
>
>>> static bool nf_flow_skb_encap_protocol(struct sk_buff *skb, __be16 proto,
>>> u32 *offset)
>>> {
>>> struct vlan_ethhdr *veth;
>>> __be16 inner_proto;
>>> + u16 size;
>>>
>>> switch (skb->protocol) {
>>> + case htons(ETH_P_IP):
>>> + if (nf_flow_ip4_encap_proto(skb, &size))
>>> + *offset += size;
>>
>> This is blindly skipping the outer IP header.
>
> Do you mean we are supposed to validate the outer IP header performing the
> sanity checks done in nf_flow_tuple_ip()?
Yes.
Note that we could always obtain a possibly considerably tput
improvement stripping required validation ;)
I guess this should go via the netfilter tree, please adjust the patch
prefix accordingly.
Also why IP over IP specifically? I guess other kind of encapsulations
may benefit from similar path and are more ubiquitous.
/P
