From owner-freebsd-net@freebsd.org Tue Jan 24 01:40:49 2017 Return-Path: Delivered-To: freebsd-net@mailman.ysv.freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:1900:2254:206a::19:1]) by mailman.ysv.freebsd.org (Postfix) with ESMTP id A23B7CBE4A1; Tue, 24 Jan 2017 01:40:49 +0000 (UTC) (envelope-from sbruno@freebsd.org) Received: from mail.ignoranthack.me (ignoranthack.me [199.102.79.106]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (Client did not present a certificate) by mx1.freebsd.org (Postfix) with ESMTPS id 6B3DEB87; Tue, 24 Jan 2017 01:40:48 +0000 (UTC) (envelope-from sbruno@freebsd.org) Received: from [192.168.0.6] (67-0-248-244.albq.qwest.net [67.0.248.244]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) (Authenticated sender: sbruno@ignoranthack.me) by mail.ignoranthack.me (Postfix) with ESMTPSA id EF9941928BA; Tue, 24 Jan 2017 01:40:46 +0000 (UTC) Subject: Re: HEADS-UP: IFLIB implementations of sys/dev/e1000 em, lem, igb pending To: =?UTF-8?Q?Olivier_Cochard-Labb=c3=a9?= , Matthew Macy References: <30f21c75-d3a2-edcd-1999-d5ed9f970c06@freebsd.org> <1598d97bf2a.c6bcb76838987.6501340920645175463@nextbsd.org> <574a7ac7-4842-9518-8286-a4d89a9f7a27@freebsd.org> <6c6cb534-73c7-464b-8af1-7445a9c0188c@freebsd.org> <1598f29d379.ea6360351471.8752933472741761813@nextbsd.org> <1598f3f8588.d20017893749.339651164872952258@nextbsd.org> <1598f42ad77.eeec05be4113.9201780237587761460@nextbsd.org> <159902b73ed.10775291e21533.7488368455500235608@nextbsd.org> Cc: "freebsd-net@freebsd.org" , "freebsd-current@freebsd.org" From: Sean Bruno Message-ID: <18abdd64-08a6-50ca-fb6b-9c01a3d7b60c@freebsd.org> Date: Mon, 23 Jan 2017 18:40:43 -0700 User-Agent: Mozilla/5.0 (X11; FreeBSD amd64; rv:45.0) Gecko/20100101 Thunderbird/45.6.0 MIME-Version: 1.0 In-Reply-To: Content-Type: multipart/signed; micalg=pgp-sha512; protocol="application/pgp-signature"; boundary="4kX657crq1w9S5xfNFjXBVMHRSJKhxwCR" X-BeenThere: freebsd-net@freebsd.org X-Mailman-Version: 2.1.23 Precedence: list List-Id: Networking and TCP/IP with FreeBSD List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 24 Jan 2017 01:40:49 -0000 This is an OpenPGP/MIME signed message (RFC 4880 and 3156) --4kX657crq1w9S5xfNFjXBVMHRSJKhxwCR Content-Type: multipart/mixed; boundary="Ix9TlJ5t959KvkbWH2LshuUEiAmGO6UVL"; protected-headers="v1" From: Sean Bruno To: =?UTF-8?Q?Olivier_Cochard-Labb=c3=a9?= , Matthew Macy Cc: "freebsd-net@freebsd.org" , "freebsd-current@freebsd.org" Message-ID: <18abdd64-08a6-50ca-fb6b-9c01a3d7b60c@freebsd.org> Subject: Re: HEADS-UP: IFLIB implementations of sys/dev/e1000 em, lem, igb pending References: <30f21c75-d3a2-edcd-1999-d5ed9f970c06@freebsd.org> <1598d97bf2a.c6bcb76838987.6501340920645175463@nextbsd.org> <574a7ac7-4842-9518-8286-a4d89a9f7a27@freebsd.org> <6c6cb534-73c7-464b-8af1-7445a9c0188c@freebsd.org> <1598f29d379.ea6360351471.8752933472741761813@nextbsd.org> <1598f3f8588.d20017893749.339651164872952258@nextbsd.org> <1598f42ad77.eeec05be4113.9201780237587761460@nextbsd.org> <159902b73ed.10775291e21533.7488368455500235608@nextbsd.org> In-Reply-To: --Ix9TlJ5t959KvkbWH2LshuUEiAmGO6UVL Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: quoted-printable On 01/23/17 08:39, Olivier Cochard-Labb=C3=A9 wrote: >=20 > On Thu, Jan 12, 2017 at 1:54 AM, Matthew Macy > wrote: >=20 > > A flame graph for the core cycle count and a flame graph with > cache miss stats from pmc would be a great start. > > > > > > =E2=80=8BI didn't know the exact event name to use for cache mis= s stats, > but here are the flame graphs for CPU_CLK_UNHALTED_CORE: > > http://dev.bsdrp.net/netgate.r311848.CPU_CLK_UNHALTED_CORE.svg > > > http://dev.bsdrp.net/netgate.r311849.CPU_CLK_UNHALTED_CORE.svg > >=20 > Thanks. Having twice as many txqs would definitely help. It's also > clear that there may be some sort of peformance issue in > iflib_txq_drain. Although it could just be non-stop cache misses on= > the packet headers. >=20 >=20 > =E2=80=8BAny news about the performance issue in iflib_txq_drain ? >=20 > On a different hardware (PC Engine APU2), I've got -20% performance dro= p: >=20 > x head r311848: packets per second > + head r311849: packets per second > +----------------------------------------------------------------------= ----+ > | ++ = x| > |+++ x = xx x| > | |= _A_|| > ||A| = | > +----------------------------------------------------------------------= ----+ > N Min Max Median Avg St= ddev > x 5 580021 588650 585676 585406.1 3550.= 8673 > + 5 463865 467599 465428 465638.6 1437.= 9347 > Difference at 95.0% confidence > -119768 +/- 3950.78 > -20.4589% +/- 0.558328% > (Student's t, pooled s =3D 2708.9) > =E2=80=8B > =20 > =E2=80=8BBecause it's an AMD processor I didn't found the pmc equivalen= t of > CPU_CLK_UNHALTED_CORE, then I've used BU_CPU_CLK_UNHALTED but I've no > idea if it's the good one. >=20 > http://dev.bsdrp.net/apu2.r311848.BU_CPU_CLK_UNHALTED.svg > http://dev.bsdrp.net/apu2.r311849.BU_CPU_CLK_UNHALTED.svg > =E2=80=8B > =E2=80=8BThanks=E2=80=8B >=20 >=20 Olivier: Which set of configs from your test suite are you using for this? Specifically, what packet size are you slamming across? https://github.com/ocochard/netbenches/tree/master/pktgen.configs sean --Ix9TlJ5t959KvkbWH2LshuUEiAmGO6UVL-- --4kX657crq1w9S5xfNFjXBVMHRSJKhxwCR Content-Type: application/pgp-signature; name="signature.asc" Content-Description: OpenPGP digital signature Content-Disposition: attachment; filename="signature.asc" -----BEGIN PGP SIGNATURE----- iQGTBAEBCgB9FiEEuq1GMucSHejSCZfdEgHvyh5yfmQFAliGsJtfFIAAAAAALgAo aXNzdWVyLWZwckBub3RhdGlvbnMub3BlbnBncC5maWZ0aGhvcnNlbWFuLm5ldEJB QUQ0NjMyRTcxMjFERThEMjA5OTdERDEyMDFFRkNBMUU3MjdFNjQACgkQEgHvyh5y fmT5fgf7B9ORzSHAo7OzfCsB+pYIvU2hEcQt1CIf3+OFE4/0WdkIbkhGqVNZlbRI q+qK3CYoGCo/82HLR8mo2CAk17auFN+/MtWcCi8u5so87RYyl8bbMpNvmKXA5V+f EMCxQOSF41Z2t3Jmk/JmK27hQ2VFHxntmiqHlbx4rhSpysRmqgVVDHpi3zIaqOT2 Upzqvd+rdpBl5pPQUO+FrjpFnGUPyMt1MzDJNU2uJwe5U1SDbZEzC8PP3huJOrwI 0G4o05bJ9herkro1+y+cSH5wlE6mnJ8FGLzP7VxHH6AapBe3WJ4l9lfyx+O6tWdQ fyaglB/fRbGG/eS984pbmhF07bMkhQ== =c8G7 -----END PGP SIGNATURE----- --4kX657crq1w9S5xfNFjXBVMHRSJKhxwCR--