Re: [Cake] Corrupted sit-tunnelled packets when using skb_gso_segment() on an IFB interface?

2018-09-10 Thread Eric Dumazet
On 09/10/2018 11:52 AM, Eric Dumazet wrote: > > > On 09/10/2018 09:04 AM, Toke Høiland-Jørgensen wrote: >> Hi everyone >> >> While investigating a bug report on CAKE[0], I've run into the following >> behaviour: >> >> When running CAKE as an ingress shaper on an IFB interface, if the GSO >>

Re: [Cake] Corrupted sit-tunnelled packets when using skb_gso_segment() on an IFB interface?

2018-09-10 Thread Eric Dumazet
On 09/10/2018 09:04 AM, Toke Høiland-Jørgensen wrote: > Hi everyone > > While investigating a bug report on CAKE[0], I've run into the following > behaviour: > > When running CAKE as an ingress shaper on an IFB interface, if the GSO > splitting feature is turned on, TCP throughput will drop

[Cake] Corrupted sit-tunnelled packets when using skb_gso_segment() on an IFB interface?

2018-09-10 Thread Toke Høiland-Jørgensen
Hi everyone While investigating a bug report on CAKE[0], I've run into the following behaviour: When running CAKE as an ingress shaper on an IFB interface, if the GSO splitting feature is turned on, TCP throughput will drop dramatically on 6in4 (sit) tunnels running over the interface in