On Wed, 04 Jan 2023 13:28:57 +0100 Toke Høiland-Jørgensen wrote: > >> Interesting! :) My mental model was closer to GRO by frags > >> so the linear part would have no data, just headers. > > > > That is assumption as well. > > Right, okay, so how many headers? Only Ethernet, or all the way up to > L4 (TCP/UDP)? If we're speaking about guarantees or hard requirements - I think that we can only require / guarantee the Ethernet header. Requiring more will be defeated by tunnels (i.e. adjust_head() + redirect to a veth). > I do seem to recall a discussion around the header/data split for TCP > specifically, but I think I mentally put that down as "something people > may way to do at some point in the future", which is why it hasn't made > it into my own mental model (yet?) :)