From mboxrd@z Thu Jan 1 00:00:00 1970 From: Eric Dumazet Subject: Re: [Bugme-new] [Bug 39372] New: Problems with HFSC Scheduler Date: Fri, 29 Jul 2011 16:00:52 +0200 Message-ID: <1311948052.2843.19.camel@edumazet-HP-Compaq-6005-Pro-SFF-PC> References: <20110714151425.844b7738.akpm@linux-foundation.org> <4E32A796.8060104@ziu.info> <1311946060.2843.15.camel@edumazet-HP-Compaq-6005-Pro-SFF-PC> Mime-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: QUOTED-PRINTABLE Cc: Andrew Morton , netdev@vger.kernel.org, bugme-daemon@bugzilla.kernel.org, Jamal Hadi Salim , lucas.bocchi@gmail.com, Patrick McHardy , 631945@bugs.debian.org, 00bormoj@gmail.com, fdelawarde@wirelessmundi.com To: Michal Soltys Return-path: Received: from mail-ww0-f42.google.com ([74.125.82.42]:35743 "EHLO mail-ww0-f42.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753774Ab1G2OCG (ORCPT ); Fri, 29 Jul 2011 10:02:06 -0400 Received: by wwg11 with SMTP id 11so4965679wwg.1 for ; Fri, 29 Jul 2011 07:02:04 -0700 (PDT) In-Reply-To: <1311946060.2843.15.camel@edumazet-HP-Compaq-6005-Pro-SFF-PC> Sender: netdev-owner@vger.kernel.org List-ID: Le vendredi 29 juillet 2011 =C3=A0 15:27 +0200, Eric Dumazet a =C3=A9cr= it : > Le vendredi 29 juillet 2011 =C3=A0 14:29 +0200, Michal Soltys a =C3=A9= crit : > > On 11-07-15 00:14, Andrew Morton wrote: > > >=20 > > > (switched to email. Please respond via emailed reply-to-all, not= via > > > the bugzilla web interface). > > >=20 > > >=20 > > > Here: WARN_ON(next_time =3D=3D 0); > > >=20 > >=20 > > From the other thread on netfilter-devel: > >=20 > > > On 11-07-22 11:58, Michal Pokrywka wrote: After bisecting 2.6.39.= 1 it > > > turned out that the bug is caused independently by two patches: > > >=20 > > > commit b262a5da755cc6ed0cb4fba230cd9bf4037e1096 sch_sfq: fix peek= () > > > implementation > > >=20 > > > and > > >=20 > > > commit 9df49f2bfe862573911a080c75a6d81113c5c81d sch_sfq: avoid gi= ving > > > spurious NET_XMIT_CN signals > > >=20 > > > Reverting these patches makes HFSC work again. > > >=20 > >=20 > > This one (upstream 8efa885406359af300d46910642b50ca82c0fe47) seems = to be > > the culprit (does reverting only that one cures the problem ?) > >=20 > > It allows SFQ to return success on enqueuing, when the packet reall= y > > replaced some other packet in some other flow. This confuses outer = qdisc > > (in this particular case HFSC) which thinks new packet was actually > > added each time such situation happes. > >=20 >=20 > Technically speaking, _this_ packet was successfuly enqueued. >=20 > Returning NET_XMIT_CN or NET_XMIT_SUCCESS should not trigger a bug in > caller. >=20 > > This in turn causes additional dequeues and ends with attempt > > to schedule non-existent packets, and triggers the warning. > >=20 >=20 > Then its probably a bug in HFSC : It doesnt understand SFQ lost a > packet. >=20 > I'll take a look, thanks for the report. >=20 >=20 Oh well, it seems one qdisc_tree_decrease_qlen(sch, 1) is missing Maybe following patch would help... diff --git a/net/sched/sch_sfq.c b/net/sched/sch_sfq.c index 4536ee6..2a2d287 100644 --- a/net/sched/sch_sfq.c +++ b/net/sched/sch_sfq.c @@ -410,7 +410,12 @@ sfq_enqueue(struct sk_buff *skb, struct Qdisc *sch= ) /* Return Congestion Notification only if we dropped a packet * from this flow. */ - return (qlen !=3D slot->qlen) ? NET_XMIT_CN : NET_XMIT_SUCCESS; + if (qlen !=3D slot->qlen) + return NET_XMIT_CN; + + /* as we dropped a packet, better let upper stack know this */ + qdisc_tree_decrease_qlen(sch, 1); + return NET_XMIT_SUCCESS; } =20 static struct sk_buff *