Return-path: Received: from mail-wg0-f45.google.com ([74.125.82.45]:44140 "EHLO mail-wg0-f45.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752568AbbAWCSV (ORCPT ); Thu, 22 Jan 2015 21:18:21 -0500 Received: by mail-wg0-f45.google.com with SMTP id x12so5046621wgg.4 for ; Thu, 22 Jan 2015 18:18:20 -0800 (PST) MIME-Version: 1.0 From: Avery Pennarun Date: Thu, 22 Jan 2015 21:17:57 -0500 Message-ID: (sfid-20150123_031824_948243_310025AD) Subject: Re: REMINDER: wireless workshop in Ottawa To: Arend van Spriel Cc: linux-wireless , Johannes Berg , Dave Taht Content-Type: text/plain; charset=UTF-8 Sender: linux-wireless-owner@vger.kernel.org List-ID: Arend van Spriel wrote: > I am afraid I will be "committing a netdev crime" [1] because I will > not attend. Still there is one topic that would be interesting for > this conf/workshop. Recent experiences with 11ac devices has shown us > that the congestion control mechanisms in the networking subsystem are > not optimal for wireless in general which becomes even more visible > with 11ac (or it is just our devices :-p ). If there are people in the > community working in this area it would be great to collaborate on it. > Let me know. I know Dave Taht is the usual suspect to talk to about > this so Cc'ed him explicitly. I'm planning to be there, and I work with Dave Taht regularly about such things. I would love to have a discussion about these issues with whoever is willing to listen :) Also, you have not experienced congestion control problems until you've tried to get TCP to do a gigabit per second with 30+ milliseconds of latency on a slightly lossy network. Have fun, Avery