On Sat, 2016-04-02 at 10:19 +0800, Herbert Xu wrote: > On Fri, Apr 01, 2016 at 07:15:33PM -0700, Eric Dumazet wrote: > > On Sat, 2016-04-02 at 09:57 +0800, Herbert Xu wrote: > > > > > > We could easily fix that by adding a feature bit to control this, > > > something like SKB_GSO_TCP_FIXEDID. > > > > I understood the patch allowed to aggregate 4 segments having > > > > ID=12 ID=40 ID=80 ID=1000 > > Right. But I haven't seen any justification that we should aggregate > such packets at all. The only valid case that I have seen is for > the case of unchanging IDs, no?
Presumably repeats of "DF=1 ID=0" should be what we really need to handle. On my wish list, having some reordering logic in GRO would be far more interesting than these minor details ;)
