Jaap Buurman <jaapbuur...@gmail.com> writes: > One more question if you don't mind :) With the "mac80211 debloated > intermediate queues" do you also mean fq_codel (phase 2), or just > phase 1?
That's "phase 2". If I'm understanding the distinction correctly, "phase 1" is everything not wifi-specific (i.e. ethernet drivers etc). There has been no debloating effort for the WiFi stack prior to the intermediate queues, apart from applying FQ-CoDel to the WiFi interfaces (which has limited effect). > Also, has cake ever been tried instead of fq_codel to try and debloat > wifi? The queueing structure in the WiFi stack is based on FQ-CoDel, but it's not just applying FQ-CoDel itself to WiFi interface. Some of the ideas from Cake has been incorporated into the WiFi stack implementation, others haven't. One of the main things Cake brings to the table is a better traffic shaper; but since we're operating on the actual queues in the WiFi stack, applying that does not make sense. -Toke _______________________________________________ Lede-dev mailing list Lede-dev@lists.infradead.org http://lists.infradead.org/mailman/listinfo/lede-dev
