Given the success of ack filtering, and cake in the libreqos system (now scaling past 10Gbit) and pping and ebpf pping turning out to be pretty high overhead... I started to muse over here about the idea of even more detailed stats than what we get:
https://github.com/rchac/LibreQoS/issues/58 I don't know how the quic "spin bit" is supposed to work, but that too, perhaps. -- FQ World Domination pending: https://blog.cerowrt.org/post/state_of_fq_codel/ Dave Täht CEO, TekLibre, LLC _______________________________________________ Cake mailing list Cake@lists.bufferbloat.net https://lists.bufferbloat.net/listinfo/cake