From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail-wr1-x42a.google.com (mail-wr1-x42a.google.com [IPv6:2a00:1450:4864:20::42a]) (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 2388C3B2A4 for ; Fri, 6 Jul 2018 07:58:51 -0400 (EDT) Received: by mail-wr1-x42a.google.com with SMTP id u7-v6so3935570wrn.12 for ; Fri, 06 Jul 2018 04:58:51 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=heistp.net; s=google; h=mime-version:subject:from:in-reply-to:date:cc :content-transfer-encoding:message-id:references:to; bh=SJIRVtgkftafxvZUTLCO+pitblTxu12UVfNXSwtjMh0=; b=FIHgr3siYYoCNOsEk2l4OFJLNDGNKyVpgf+Mxyqc1j6tPDO0tW1ibNRUlaDDegI4Tp GO7x7cVs3Qs/k6Nae72rFivsvUYoauC35fo1CMfzaSjhzRVPbpcH7GfH2D9uydmnsI6i q5bqgVTYIv+lRKugNZtEmVZZTCX6EFO69hzRqPlxkU5lg4xVxeTKKYsrvt+hVmJNKyq+ keBSO7euDH73xk6A9XpdbSXqgIqw7aj07BFP4FF62+MEf1XUzrcgXkn3dUL+Jd+fBtJt 4mvmtPMGkB1fb+EBBF9PySj5Ohth8GbQd6tBmALyijQ3ImDrQmNR4hqRful60BQwIJwG nflA== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20161025; h=x-gm-message-state:mime-version:subject:from:in-reply-to:date:cc :content-transfer-encoding:message-id:references:to; bh=SJIRVtgkftafxvZUTLCO+pitblTxu12UVfNXSwtjMh0=; b=IOoczxGJB3Oyjy6/T3MqwUN8gBCGGomq61XB8K7jLcjckIjP+dw2mX4YLo2ylqx3R4 sMzw4xd4hQaAYuqmLtGyX+x2sDUwVk/HD2ozkcs90UhsFejgV1Z9f2So1ZLyTfmPt/X+ tvg7a1IUOXIAHS1KCxQofvpQ0JQIw786YAAaoBpg/j5xoH+Iq8MvDV7xxOLkXjtWamTn XT31ZsI4NGfJqMFPX5rdPo062KqSQcvIfmThnTy7hLUe00uBVNccyuA0h+v92t2ijVek 58JxucyX6d1MEUwVXxlrhA+Mag9DCX/Ufl3rnSQxq0y8oygo/wlJBs5l5GjqljPQWULQ 6f1w== X-Gm-Message-State: APt69E3zvWVi0jCDgAnqy5oMWTji0uavu0AVB+ZVLP2uBhtZ6G0CRo44 efltj+AoE8TTqVnNLxFublxmFFz2ryw= X-Google-Smtp-Source: AAOMgpdU+cQx1WNqsGvztUnhGBr/p2qt2p6Q2c+I/zptT39dQ9KDJnGrZCkm1UN8Zmz/eJhxHn+28A== X-Received: by 2002:adf:8162:: with SMTP id 89-v6mr7098547wrm.192.1530878330331; Fri, 06 Jul 2018 04:58:50 -0700 (PDT) Received: from tron.luk.heistp.net (h-1169.lbcfree.net. [185.193.85.130]) by smtp.gmail.com with ESMTPSA id n10-v6sm7888963wrs.64.2018.07.06.04.58.49 (version=TLS1_2 cipher=ECDHE-RSA-AES128-GCM-SHA256 bits=128/128); Fri, 06 Jul 2018 04:58:49 -0700 (PDT) Content-Type: text/plain; charset=utf-8 Mime-Version: 1.0 (Mac OS X Mail 11.4 \(3445.8.2\)) From: Pete Heist In-Reply-To: <87in5sbnuo.fsf@toke.dk> Date: Fri, 6 Jul 2018 13:58:48 +0200 Cc: Jonathan Morton , Cake List Content-Transfer-Encoding: quoted-printable Message-Id: <8184CEEA-64C0-4CCD-A831-D90CFDC56F22@heistp.net> References: <87fu11ipir.fsf@toke.dk> <871scligay.fsf@toke.dk> <2AE036E5-BD3D-4176-9476-9EC824EC1D18@darbyshire-bryant.me.uk> <87r2klh1fz.fsf@toke.dk> <87lgath01v.fsf@toke.dk> <52B2B44D-4382-404C-8F6D-03F12A72B11F@heistp.net> <31667353-48F2-4FAB-AC05-163680451719@toke.dk> <48ECB6C8-5D22-4785-A6CE-696D87EC5496@toke.dk> <73DD74AD-C2E7-4A12-AE49-C06D4486660E@gmail.com> <87fu10haw7.fsf@toke.dk> <8736wxco28.fsf@toke.dk> <87o9fkbtky.fsf@toke.dk> <87lgaobq0m.fsf@toke.dk> <87in5sbnuo.fsf@toke.dk> To: =?utf-8?Q?Toke_H=C3=B8iland-J=C3=B8rgensen?= X-Mailer: Apple Mail (2.3445.8.2) Subject: Re: [Cake] cake at 60gbit 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, 06 Jul 2018 11:58:51 -0000 > On Jul 6, 2018, at 1:33 PM, Toke H=C3=B8iland-J=C3=B8rgensen = wrote: >=20 > AHA! Found the culprit! >=20 > The bulk dequeue mechanism in sch_generic.c will dequeue a bunch of > packets at once, then check if they belong on the same hardware txq. = If > they don't, they will be put back on a separate queue in the qdisc > structure (sch->skb_bad_txq), and the qlen will be increased, without > telling the qdisc about it. Solid, nice work! > This obviously only happens on hardware with multiple TXQs, which is = why > the bug doesn't happen on veth. It would be nice if veth were mq capable. For whatever reason, I didn=E2=80=99t see this on my i210at=E2=80=99s = (1gbit ethernet with 4 transmit and 4 receive queues). I=E2=80=99m now playing with netem, cake and veth for the first time = (two namespaces with netem as the parent qdisc to cake for each = namespace). I=E2=80=99ve gotten the setup not to lock up in an infinite = loop but to occasionally stop passing traffic sometimes after a netperf = test. This could easily be a problem specific to netns though, so I=E2=80=99= ll be playing with it some more and will post if I can narrow it down to = something specific.=