From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail-lf0-x233.google.com (mail-lf0-x233.google.com [IPv6:2a00:1450:4010:c07::233]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by lists.bufferbloat.net (Postfix) with ESMTPS id C287E3CC02; Wed, 6 Apr 2016 12:46:47 -0400 (EDT) Received: by mail-lf0-x233.google.com with SMTP id c126so38303680lfb.2; Wed, 06 Apr 2016 09:46:47 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=mime-version:subject:from:in-reply-to:date:cc :content-transfer-encoding:message-id:references:to; bh=/UmUF9bpwmXNsljDuFj96VOVTfe5Cw0J1gT+O1meBSg=; b=gtw2ZQSCrP0ZHHi6WIWCcabsRSgrqst4B3gajmr303l87IjI8A3nms3JKsM07D0Lmp kygwqvvONpR37tRAvZwCetd+NnHq9pWwlUUjSBCUPW8KmlK9xWGplmBviyQgrSMuLWIa HOHiGb5qXIx62ZEXXUrCIcgxk32r2SX94+MnIGcK+pi/YV6BzIknOGvNQQ/zyTfrCJOc KSSvGZ9zVwuwfOBDJVgAWE1BfG8XkXmUK+Zb/xmZonFJ7EhSwUANpTB2jEa8cWrGjQ7l ixD1tHdaBALGLy6mhRt4r8pnoE+cDT7QWpzEhZfUWcAq+ehOJU3tonyo1yzJuJMCqM8H RtXQ== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20130820; h=x-gm-message-state:mime-version:subject:from:in-reply-to:date:cc :content-transfer-encoding:message-id:references:to; bh=/UmUF9bpwmXNsljDuFj96VOVTfe5Cw0J1gT+O1meBSg=; b=Tt2C5Y76DyeWJsr3Mvo9v6djUc1Wg3H+ry4EImqd7q6GkRV6JTMGFQMr3KvWJWSkE8 zlaA7CiP8zBDQnxeysmLZ1DXh9w/WFA2hGO9RTSa+z9KYG7atGcaw4LcyrhK8QWFLQiy JrmGFS6Qo5stGMXhxxa8kY/eID8hM7pOdkA7QhIFII/XYKDJcuneaUoHcXSeBTMOMCGv b50QYkSix5v/Ebc0Op6p55Im36tYY+M+n1egUe+tKMnn9E6dClsnVUIgp7PizvGJe6/C an+CSCbFlEqQ3QFEB98EqInJZhcDto/PjgDO4YliAGSJxZYmHOpW9FOTNhVnIUyrzd0C Biyw== X-Gm-Message-State: AD7BkJJddNjwuaWd+6WwfuhiFoRzCWl8OoDYhpzbTe6MihU62xgFaXEOZJwes75FBc/qOA== X-Received: by 10.25.41.140 with SMTP id p134mr11587102lfp.15.1459961206413; Wed, 06 Apr 2016 09:46:46 -0700 (PDT) Received: from bass.home.chromatix.fi (37-33-67-252.bb.dnainternet.fi. [37.33.67.252]) by smtp.gmail.com with ESMTPSA id au9sm555235lbc.24.2016.04.06.09.46.27 (version=TLSv1/SSLv3 cipher=OTHER); Wed, 06 Apr 2016 09:46:45 -0700 (PDT) Content-Type: text/plain; charset=utf-8 Mime-Version: 1.0 (Mac OS X Mail 9.2 \(3112\)) From: Jonathan Morton In-Reply-To: Date: Wed, 6 Apr 2016 19:46:11 +0300 Cc: Johannes Berg , make-wifi-fast@lists.bufferbloat.net, linux-wireless , codel@lists.bufferbloat.net Content-Transfer-Encoding: quoted-printable Message-Id: References: <1458898052-20601-1-git-send-email-michal.kazior@tieto.com> <1459420104-31554-1-git-send-email-michal.kazior@tieto.com> <1459420104-31554-2-git-send-email-michal.kazior@tieto.com> <1459864656.18188.60.camel@sipsolutions.net> <2233FCEB-7412-4F22-B262-068AFBB2FDCB@gmail.com> To: Michal Kazior X-Mailer: Apple Mail (2.3112) Subject: Re: [Codel] [Make-wifi-fast] [PATCHv2 1/2] mac80211: implement fair queuing per txq X-BeenThere: codel@lists.bufferbloat.net X-Mailman-Version: 2.1.20 Precedence: list List-Id: CoDel AQM discussions List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 06 Apr 2016 16:46:48 -0000 > On 6 Apr, 2016, at 10:16, Michal Kazior = wrote: >=20 > When a driver asks mac80211 to dequeue given txq it implies a > destination station as well. This is important because 802.11 > aggregation can be performed only on groups of packets going to a > single station on a single tid. >=20 > Cake - as I understand it - doesn't really *guarantee* maintaining > this. Keep in mind you can run with hundreds of stations connected. >=20 > You don't really want to burden drivers with sorting this grouping up > themselves (and hence coerce them into introducing another level of > intermediate queues, bis). Well, no. Cake isn=E2=80=99t designed to maintain per-station queues = explicitly, though it does have support for stochastic fairness between = hosts. It is also blissfuly unaware of the requirements of wifi = aggregation, largely because the standard qdisc interface is likewise = ignorant. I=E2=80=99m therefore not suggesting that you use Cake as-is. What I=E2=80=99m pointing at instead is the set-associative hash, which = could easily be tweaked to put greater emphasis on avoiding putting = multiple stations=E2=80=99 traffic in one queue, while maintaining the = performance benefits of a fixed queue pool indexed by a hash table, and = an extended operating region in which flow isolation is maintained. You = can then have a linked-list of queues assigned to a particular station, = so that when a packet for a particular station is requested, you can = easily locate one. I hadn=E2=80=99t appreciated, though, that the TXQ struct was = station-specific. This wasn=E2=80=99t obvious from the code fragments = posted, so it looked like packets that incurred hash collisions would be = dumped into a single overflow queue. - Jonathan Morton