From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from nm27-vm0.access.bullet.mail.sp2.yahoo.com (nm27-vm0.access.bullet.mail.sp2.yahoo.com [98.139.44.188]) by huchra.bufferbloat.net (Postfix) with SMTP id 9397321F095 for ; Wed, 11 Jul 2012 11:09:29 -0700 (PDT) Received: from [98.139.44.102] by nm27.access.bullet.mail.sp2.yahoo.com with NNFMP; 11 Jul 2012 18:09:26 -0000 Received: from [98.139.44.74] by tm7.access.bullet.mail.sp2.yahoo.com with NNFMP; 11 Jul 2012 18:09:26 -0000 Received: from [127.0.0.1] by omp1011.access.mail.sp2.yahoo.com with NNFMP; 11 Jul 2012 18:09:26 -0000 X-Yahoo-Newman-Id: 853622.24234.bm@omp1011.access.mail.sp2.yahoo.com Received: (qmail 42731 invoked from network); 11 Jul 2012 18:09:26 -0000 DomainKey-Signature: a=rsa-sha1; q=dns; c=nofws; s=s1024; d=rogers.com; h=DKIM-Signature:X-Yahoo-Newman-Property:X-YMail-OSG:X-Yahoo-SMTP:Received:Message-ID:Date:From:Reply-To:User-Agent:MIME-Version:To:Subject:References:In-Reply-To:X-Enigmail-Version:Content-Type:Content-Transfer-Encoding; b=VwSlY4re1+wJ2clnzJ12eIIv0lJGwZPmOHaj15hL8B+HRZdcBQokxoxXFclyYcLTpHsWdN/AlCmW3US6Ti6d9q4fdfqXQp9WNGpk/I2P6zfor/FojjmmqJSlUbhnZDwsw3PmQdOhonpPN3JBkyxMGwr2wKlDi34bMNx3PKed3Jo= ; DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=rogers.com; s=s1024; t=1342030166; bh=bb1TVCDYdFcDp4Sg7sB6+f7AA6cHb9f2RDHvBwjz9bk=; h=X-Yahoo-Newman-Property:X-YMail-OSG:X-Yahoo-SMTP:Received:Message-ID:Date:From:Reply-To:User-Agent:MIME-Version:To:Subject:References:In-Reply-To:X-Enigmail-Version:Content-Type:Content-Transfer-Encoding; b=IaeNXcCQHziXymNPx0/plpUvOj3P9wad20btgumRHj/RSjsQoYugTdAgnxWfBGqMy35y9RBbUDYtJiNO/IwWBUxrEOGHgWqaFXXCpNRurCUM7paxZdK1BHb6x8QFb1ky/KkBTp4p5SthW8NGXTT/Mlx0PQORAqo9na2fNSmk5sA= X-Yahoo-Newman-Property: ymail-3 X-YMail-OSG: Rfe0Jq4VM1kQvjwOqnle8Xi2i2mMg1D14pnLLJMmEIgHgPJ GR0ouplT.FRhtfAeREikOQ9s4KqCmuM3wzWCzkcfIpwfUYrIyv3Z22zKjqiz NLj9sdT_n4AQW7rgX6hTfnfLPr7UEQfK7Crw49aybs_8S8QDd9aMbSgHY10m aKHpq6zxW2_RVeZdKADvV_W.peVMvtyeTXtqsdqJ_visx7rWWrzEn0kEb.XX s_osK9.1bNhpyAxv5UYxa4ZZLNyhf4mH_61AqznbLxMp1xXJ5DcVPuCdDrDj WrCnuH6MUCD3bNomtxLBCzZx.r0H4vMl92SYmU7bjrlWKACHnyCxE5RHr678 1OkNCPRmD0zssTMqG8qlXT_6lp1d8jd8WVcdGhEJlIlMG5kCgyntSt1i0N2s xV9xsGYiX0A9ropfhXtV0O0oZyoG1ami.7pWN9SjA.6w9dAin3BoL9wnvbKP 0kbJdVes- X-Yahoo-SMTP: sltvjZWswBCRD.ElTuB1l9j6s9wRYPpuyTNWOE5oEg-- Received: from [192.168.1.101] (davec-b@99.225.204.110 with plain) by smtp107.rog.mail.gq1.yahoo.com with SMTP; 11 Jul 2012 11:09:26 -0700 PDT Message-ID: <4FFDC1BC.9030909@rogers.com> Date: Wed, 11 Jul 2012 14:11:08 -0400 From: David Collier-Brown User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:12.0) Gecko/20120430 Thunderbird/12.0.1 MIME-Version: 1.0 To: codel@lists.bufferbloat.net References: In-Reply-To: X-Enigmail-Version: 1.4.2 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Subject: [Codel] Sidebar to TCP Small Queues - cooperative bandwidth shaping X-BeenThere: codel@lists.bufferbloat.net X-Mailman-Version: 2.1.13 Precedence: list Reply-To: davecb@spamcop.net List-Id: CoDel AQM discussions List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 11 Jul 2012 18:09:29 -0000 On 07/11/2012 12:17 PM, codel-request@lists.bufferbloat.net -> Ben Greear wrote > On 07/11/2012 08:11 AM, Eric Dumazet wrote: >> On Tue, 2012-07-10 at 17:13 +0200, Eric Dumazet wrote: >>> This introduce TSQ (TCP Small Queues) ... > > I haven't read your patch in detail, but I was wondering if this feature > would cause trouble for applications that are servicing many sockets at once > and so might take several ms between handling each individual socket. > > Or, applications that for other reasons cannot service sockets quite > as fast. Without this feature, they could poke more data into the > xmit queues to be handled by the kernel while the app goes about it's > other user-space work? > > Maybe this feature could be enabled/tuned on a per-socket basis? > I speculate[1] that sharing of resources between applications, or sockets within applications, can better be done as a separate action, by using either a purely local mechanism like cgroups, or by a mechanism that can use a global service-information daemon like the oldish "trickle" bandwidth manager (still available on Fedora via "yum install trickle") Strictly speaking, only the trickled component would be needed, and could conveniently reside on a router, while the the trickle shared library could be used initially, and retired in favour of the proper local control mechanism. --dave [1. Polite Canadian for "I'm pretty sure"] -- David Collier-Brown, | Always do right. This will gratify System Programmer and Author | some people and astonish the rest davecb@spamcop.net | -- Mark Twain (416) 223-8968