From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail2.tohojo.dk (mail2.tohojo.dk [77.235.48.147]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by lists.bufferbloat.net (Postfix) with ESMTPS id 9F78B3B25E; Wed, 10 Aug 2016 11:04:11 -0400 (EDT) X-Virus-Scanned: amavisd-new at mail2.tohojo.dk DKIM-Filter: OpenDKIM Filter v2.10.3 mail2.tohojo.dk 44E3240D5E DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=toke.dk; s=201310; t=1470841449; bh=iKBkl7UU/vMWWSFkiAq3lM5S4Ij7kzRR2FGErNysL1g=; h=From:To:Cc:Subject:References:Date:In-Reply-To:From; b=l4zC9V99mffGN+2rF7oQMZqu7N3s7j3yxhYLzkH2q9MNLEVMEfHHqLyMNseJiGpw7 BkJGjA7ZSVtrx4qVa4AHcLR1o9i5lFJDJNPtqkG2bSlQosPMu79FAkwq81Xo3ym+KS 2KmWab5kUI01sAMZjx7gWzKsvFayqBipWD1ddMJA= Sender: toke@toke.dk Received: by alrua-x1.borgediget.toke.dk (Postfix, from userid 1000) id 30F425618; Wed, 10 Aug 2016 17:04:08 +0200 (CEST) From: =?utf-8?Q?Toke_H=C3=B8iland-J=C3=B8rgensen?= To: Noah Causin Cc: Dave Taht , make-wifi-fast@lists.bufferbloat.net, "cerowrt-devel\@lists.bufferbloat.net" , Felix Fietkau , Simon Wunderlich References: <76eb6a6a-2f39-3b36-d4c1-04198083c0f6@gmail.com> Date: Wed, 10 Aug 2016 17:04:08 +0200 In-Reply-To: <76eb6a6a-2f39-3b36-d4c1-04198083c0f6@gmail.com> (Noah Causin's message of "Wed, 10 Aug 2016 09:06:42 -0400") X-Clacks-Overhead: GNU Terry Pratchett Message-ID: <87oa50acnr.fsf@toke.dk> MIME-Version: 1.0 Content-Type: text/plain Subject: Re: [Make-wifi-fast] wifi airtime fairness patches could use eyeballs and testing X-BeenThere: make-wifi-fast@lists.bufferbloat.net X-Mailman-Version: 2.1.20 Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 10 Aug 2016 15:04:11 -0000 Noah Causin writes: > Hi, > > I am aware that a performance regression with using fq in mac80211 with multiple tcp streams has been reported, and this patch in lede disables it. > > https://github.com/lede-project/source/commit/4952469ff9278288d766b28247a17694b1c4faaa > > Has that been resolved? Just did a test of this and can reproduce with a wndr3800 as the access point. I see about a 25% reduction in aggregate throughput with two streams as opposed to one. Packet captures and time sequence graphs are here: https://kau.toke.dk/experiments/softq-debug/ - the onestream* and twostream* files. -Toke