From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mout.gmx.net (mout.gmx.net [212.227.17.21]) (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 96DAE3B2A3 for ; Fri, 3 Mar 2017 04:22:31 -0500 (EST) Received: from [10.120.32.28] ([80.187.107.134]) by mail.gmx.com (mrgmx101 [212.227.17.168]) with ESMTPSA (Nemesis) id 0MK4fR-1ckYEK2cth-001NHr; Fri, 03 Mar 2017 10:22:29 +0100 Date: Fri, 03 Mar 2017 10:22:29 +0100 User-Agent: K-9 Mail for Android In-Reply-To: References: MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: quoted-printable To: cake@lists.bufferbloat.net,Dave Taht From: Sebastian Moeller Message-ID: <972E3CF5-22BC-4CE0-BD42-29C5382CE90E@gmx.de> X-Provags-ID: V03:K0:OJUJg8u0gNalqZ495EK6T1DjQijHDjtAJJ7xbYC2KTJRLYgk6Xj c+L9taCppUmMyXXwG5wPbZN7Geh7ami1hympoKUn7WAn7wPesLRxAeLv8sYbACX1IU26f/i k2r1uOFel84rjC0UGuJXn529KEp9SEv0TuUra4NYqZY5kHSzgLDALIImWSAJRhx322RhHrk Sum/XvSMPbOYjhJ17O+Cg== X-UI-Out-Filterresults: notjunk:1;V01:K0:8Zf9pbCb7D4=:uA7eePtU0zUy8ZPjEyqFIl NaI6+hYBaR6MJ/FOAyc1otLKjCBL13JRiok5/mYsR4uNiu1obNFb0TkIQrdNCT2rnt8TxZDv8 QFt2GU518enCSzT7XPcANrj4l1RWbTvk2m0utKRDYC5ZXG3H/qwZT6ywbtXyD3BEX/C1XMyt/ nb+LyK2TlGQY9U+sWpplygRvGXQEzTtGYeKnAtF9OB/IutBtltwa0qLpgovk9qO9r5J4bheii sztfjkFrrpGnvWJnzXm2SknLdpJ17tks42yiJddsDCBIXLt8p/1kiRAGOV/ySef2C6uhWDLRs vL4q1tAYQcIFEA7fOfMyK9ViMD6e0UYL+/Ijk87d4Qv/PFkdoBnqclXbA2pGVuu9MwXjFX+FL kDS6oDBnc6k62ezHZZxPg6pnTHumgKL/m1JTYhwoT2k1sQj47AAQjmtK75duGrfY6+scNTErn GY6ILvE5o/Vhc/anEvaEfBTqVA+EFheG8//4fWyDjsKYBIu2AQXf64+mJB+UVJ2kSBJfihSCP 9mF/PNx99I+GNrd3ycVoZFdnWurIBsVQjewM+YfnCwQupQIUOPrkT+DasEcocLwiWVbtWcCKx vPolj0l4Ml2UVkiK9wbGH/AlJcuHcbmRyiwcCv/Oq/tQ3dPs/GTgmUYBQ5CmFljYtSZwp2bgk qQ6aCQwUxgKe0CfODXBsmjlzSnmoxsHSnTEmQvMyAjM2FmPcfKId88JjUCguR2itxFEuRNNRw D5coLut0NfB3mlLfEa4NuNAuVXJDhI1/3B0KPVE78VxRLZy/7t39MORlyNx2JaTMC5L/BYXYd 88pfVVY Subject: Re: [Cake] improving inbound shaping X-BeenThere: cake@lists.bufferbloat.net X-Mailman-Version: 2.1.20 Precedence: list List-Id: Cake - FQ_codel the next generation List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 03 Mar 2017 09:22:32 -0000 Hi Dave, Last time I tested this I came to the conclusion, that going via ifb and m= irred cost something like 5% shaper sqm performance on ingress=2E While not= nothing I am not sure this as big a performance issue as you seem to argue= =2E=20 Your proposal would improve usability quite a lot though if we could avoid= the whole ifb-dance=2E=2E=2E To assess the cost of the mirred ifb ingress, I simply instantiated the sh= aper for internet download not on ingress of the wan interface, but on egre= ss of the LAN interface connecting the sole testing host, so I believe the = only difference in work for the system should be the ifb/mirred processing= =2E On March 3, 2017 7:54:23 AM GMT+01:00, Dave Taht = wrote: >As that's the highest cpu user there is, (and the biggest problem I >have, on comcast, there's 2 sec of latency at 100mbit without shaping: > > http://www=2Etaht=2Enet/~d/comcast2/asmiserabledlasever=2Epng >(more flent data there)=2E > >It's always been a daydream to somehow bypass the existing tc_mirred >facility we use and be able to express: > >tc qdisc add dev eth0 ingress cake bandwidth 990mbit > >and have that "just work"=2E My hope would be that that would halve >the packet copies needed (don't know if that's the case in the first >place)=2E=2E=2E > >When I last looked at it (2+ years ago), that portion of linux was a >hairball that extended back to the late 90s, and I gave up=2E > >There were a few commits there recently - adding hardware offload >support for the flower classifier and this one: > >commit d2788d34885d4ce5ba17a8996fd95d28942e574e >Author: Daniel Borkmann >Date: Sat May 9 22:51:32 2015 +0200 > > net: sched: further simplify handle_ing > > Ingress qdisc has no other purpose than calling into tc_classify() > that executes attached classifier(s) and action(s)=2E > > It has a 1:1 relationship to dev->ingress_queue=2E After having commit > 087c1a601ad7 ("net: sched: run ingress qdisc without locks") removed > the central ingress lock, one major contention point is gone=2E > > The extra indirection layers however, are not necessary for calling > into ingress qdisc=2E pktgen calling locally into netif_receive_skb() > with a dummy u32, single CPU result on a Supermicro X10SLM-F, Xeon > E3-1240: before ~21,1 Mpps, after patch ~22,9 Mpps=2E > > > > >--=20 >Dave T=C3=A4ht >Let's go make home routers and wifi faster! With better software! >http://blog=2Ecerowrt=2Eorg >_______________________________________________ >Cake mailing list >Cake@lists=2Ebufferbloat=2Enet >https://lists=2Ebufferbloat=2Enet/listinfo/cake