From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail-we0-f171.google.com (mail-we0-f171.google.com [74.125.82.171]) (using TLSv1 with cipher RC4-SHA (128/128 bits)) (Client CN "smtp.gmail.com", Issuer "Google Internet Authority" (verified OK)) by huchra.bufferbloat.net (Postfix) with ESMTPS id 567BF2022BC for ; Sun, 24 Jun 2012 22:24:26 -0700 (PDT) Received: by wejx9 with SMTP id x9so4955234wej.16 for ; Sun, 24 Jun 2012 22:24:24 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=subject:from:to:cc:in-reply-to:references:content-type:date :message-id:mime-version:x-mailer:content-transfer-encoding; bh=mTGy/5RnBbTkWy9B+qv4evSfl3qb1MSO09sF+XtGuqk=; b=gXN8ADGpO2FLHRpi9NdGbqvQNKfTxcbJdIlC5Uf2it6xgnFkRskdezjJUFOPpXKop/ C98vL0W6Rw/nStfEAL3fCV0QO7SWU8gHIWzLiCoxwc0ny1nPahrpsT2ekz2xUvpgiInm fvif6iOdUva0HvAxqxJmAI/41LmCyAKV3P55GFvH/9gYwPZGF0Ur8lqcdztiv/WVTV4v aJ9inaid9ZXUpd+Tc3iA3goo0kmhl89/NDO/e+h+PZiNXw2ysqNv5ZQJsCuaooVjfeWL 519ybOo/YJMBisoA/K1qNdfxv0mq8FBj5CAQ+AM4kU/8oRWIU/J/I3QLVGnyg3O11V2d xhfA== Received: by 10.180.80.134 with SMTP id r6mr6383451wix.1.1340601864529; Sun, 24 Jun 2012 22:24:24 -0700 (PDT) Received: from [192.168.1.37] (122.237.66.86.rev.sfr.net. [86.66.237.122]) by mx.google.com with ESMTPS id m4sm1114963wie.1.2012.06.24.22.24.22 (version=SSLv3 cipher=OTHER); Sun, 24 Jun 2012 22:24:23 -0700 (PDT) From: Eric Dumazet To: Dave =?ISO-8859-1?Q?T=E4ht?= In-Reply-To: <1340600422-1806-1-git-send-email-dave.taht@bufferbloat.net> References: <1340600422-1806-1-git-send-email-dave.taht@bufferbloat.net> Content-Type: text/plain; charset="UTF-8" Date: Mon, 25 Jun 2012 07:24:22 +0200 Message-ID: <1340601862.23933.18.camel@edumazet-glaptop> Mime-Version: 1.0 X-Mailer: Evolution 2.28.3 Content-Transfer-Encoding: 8bit Cc: codel@lists.bufferbloat.net Subject: Re: [Codel] [RFCv2 PATCH] iproute2: Add ecn_target option to codel and fq_codel X-BeenThere: codel@lists.bufferbloat.net X-Mailman-Version: 2.1.13 Precedence: list List-Id: CoDel AQM discussions List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 25 Jun 2012 05:24:26 -0000 On Sun, 2012-06-24 at 22:00 -0700, Dave Täht wrote: > From: Dave Taht > > ECN was not part of the original codel design and adding support > for it revealed problems in two areas. 1) ECN can be gamed. > 2) Dropping packets under overload more rapidly frees up > bandwidth than marking packets. > > Two possible scenarios of use - on egress from a network, > ecn_target could be set low, to drop more often, to > ensure lowest latency for other packets. > > On ingress, it could be set high to mark packets more often, > to lower data loss while still signalling the end application > that bandwidth is a problem. > > ecn_target is not engaged until after codel enters a dropping > state overall. > --- > include/linux/pkt_sched.h | 2 ++ > man/man8/tc-codel.8 | 3 +++ > man/man8/tc-fq_codel.8 | 5 +++++ > tc/q_codel.c | 17 ++++++++++++++++- > tc/q_fq_codel.c | 17 ++++++++++++++++- > 5 files changed, 42 insertions(+), 2 deletions(-) > > diff --git a/include/linux/pkt_sched.h b/include/linux/pkt_sched.h > index 32aef0a..f0dcf8c 100644 > --- a/include/linux/pkt_sched.h > +++ b/include/linux/pkt_sched.h > @@ -663,6 +663,7 @@ enum { > TCA_CODEL_LIMIT, > TCA_CODEL_INTERVAL, > TCA_CODEL_ECN, > + TCA_CODEL_ECN_TARGET, > __TCA_CODEL_MAX > }; > > @@ -691,6 +692,7 @@ enum { > TCA_FQ_CODEL_ECN, > TCA_FQ_CODEL_FLOWS, > TCA_FQ_CODEL_QUANTUM, > + TCA_FQ_CODEL_ECN_TARGET, > __TCA_FQ_CODEL_MAX > }; > > diff --git a/man/man8/tc-codel.8 b/man/man8/tc-codel.8 > index 61f163f..0e97249 100644 > --- a/man/man8/tc-codel.8 > +++ b/man/man8/tc-codel.8 > @@ -64,6 +64,9 @@ is the acceptable minimum standing/persistent queue delay. This minimum delay > is identified by tracking the local minimum queue delay that packets experience. > Default and recommended value is 5ms. > > +.SS ecn_target > +is the overall delay at which ecn marked packets will be dropped rather than marked. > + > .SS interval > is used to ensure that the measured minimum delay does not become too stale. The > minimum delay must be experienced in the last epoch of length > diff --git a/man/man8/tc-fq_codel.8 b/man/man8/tc-fq_codel.8 > index 8b43c10..9cb5be8 100644 > --- a/man/man8/tc-fq_codel.8 > +++ b/man/man8/tc-fq_codel.8 > @@ -49,6 +49,11 @@ and is the acceptable minimum > standing/persistent queue delay. This minimum delay is identified by tracking > the local minimum queue delay that packets experience. Default value is 5ms. > > +.SS ecn_target > +has the same semantics as > +.B codel > +and is the delay beyond which ecn packets will be dropped rather than marked. > + What do you mean by "ecn_target has the same semantics as codel" ?