From owner-freebsd-net@FreeBSD.ORG Mon Feb 2 17:36:02 2015 Return-Path: Delivered-To: freebsd-net@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:1900:2254:206a::19:1]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by hub.freebsd.org (Postfix) with ESMTPS id 8ADE771E; Mon, 2 Feb 2015 17:36:02 +0000 (UTC) Received: from mga09.intel.com (mga09.intel.com [134.134.136.24]) by mx1.freebsd.org (Postfix) with ESMTP id 5A7D17B9; Mon, 2 Feb 2015 17:36:02 +0000 (UTC) Received: from orsmga003.jf.intel.com ([10.7.209.27]) by orsmga102.jf.intel.com with ESMTP; 02 Feb 2015 09:32:34 -0800 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="5.09,507,1418112000"; d="scan'208";a="521423158" Received: from orsmsx110.amr.corp.intel.com ([10.22.240.8]) by orsmga003.jf.intel.com with ESMTP; 02 Feb 2015 09:28:29 -0800 Received: from orsmsx157.amr.corp.intel.com (10.22.240.23) by ORSMSX110.amr.corp.intel.com (10.22.240.8) with Microsoft SMTP Server (TLS) id 14.3.195.1; Mon, 2 Feb 2015 09:35:59 -0800 Received: from orsmsx111.amr.corp.intel.com ([169.254.11.110]) by ORSMSX157.amr.corp.intel.com ([169.254.9.139]) with mapi id 14.03.0195.001; Mon, 2 Feb 2015 09:35:59 -0800 From: "Pieper, Jeffrey E" To: "sbruno@freebsd.org" , Jack Vogel , hiren panchasara Subject: RE: Intel 82574L (em) Thread-Topic: Intel 82574L (em) Thread-Index: AQHQPNFkP3MptScjxEq0dTQWShT445zZ7wMAgAA6nwCAA2nSEIAAlB4A//96LpA= Date: Mon, 2 Feb 2015 17:35:58 +0000 Message-ID: <2A35EA60C3C77D438915767F458D6568806C268D@ORSMSX111.amr.corp.intel.com> References: <54CBF396.3090903@ignoranthack.me> <20150131010014.GB19333@strugglingcoder.info> <2A35EA60C3C77D438915767F458D6568806C25DE@ORSMSX111.amr.corp.intel.com> <54CFB38E.1040408@ignoranthack.me> In-Reply-To: <54CFB38E.1040408@ignoranthack.me> Accept-Language: en-US Content-Language: en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: x-originating-ip: [10.22.254.138] Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: quoted-printable MIME-Version: 1.0 Cc: FreeBSD Net X-BeenThere: freebsd-net@freebsd.org X-Mailman-Version: 2.1.18-1 Precedence: list List-Id: Networking and TCP/IP with FreeBSD List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 02 Feb 2015 17:36:02 -0000 In the past we have, yes. This was a few years ago, but iirc the current im= plementation is supposed to be the official solution. Jeff -----Original Message----- From: Sean Bruno [mailto:sbruno@ignoranthack.me]=20 Sent: Monday, February 02, 2015 9:28 AM To: Pieper, Jeffrey E; Jack Vogel; hiren panchasara Cc: FreeBSD Net Subject: Re: Intel 82574L (em) -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA512 On 02/02/15 08:42, Pieper, Jeffrey E wrote: > Iirc, we experienced issues with 82574L, where the interface will > hang/die. This is resolved in both FreeBSD and Linux by forcing > ASPM off and disabling MSIX. >=20 > Jeff >=20 >=20 Are you running tests with the multi-queue implementation in the h/w and driver turned on? sean > ----Original Message----- From: owner-freebsd-net@freebsd.org > [mailto:owner-freebsd-net@freebsd.org] On Behalf Of Jack Vogel=20 > Sent: Friday, January 30, 2015 8:30 PM To: hiren panchasara Cc: > FreeBSD Net Subject: Re: Intel 82574L (em) >=20 > Yup, I wrote that :) >=20 > Sean, I will check around to see if anything may have changed in > that regard. >=20 > Jack >=20 >=20 > On Fri, Jan 30, 2015 at 5:00 PM, hiren panchasara <=20 > hiren@strugglingcoder.info> wrote: >=20 >> On Fri, Jan 30, 2015 at 01:11:50PM -0800, Sean Bruno wrote: >>>=20 >>> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA512 >>>=20 >>>=20 >> http://www.intel.com/content/dam/doc/datasheet/82574l-gbe-controller-dat= asheet.pdf >>> >>> >>=20 According to 7.1.11, this device does indeed have 2 queues for stuff and >>> or things. So, basic RSS would be possible in something like >>> an Atom >> box. >>>=20 >>> I note that the em(4) driver intentionally disables this on=20 >>> initialization. I'm up for some science on my new shiny, soon >>> to be router box. Any reason not to default to 1 queue and >>> allow loader.conf to raise it to 2? >>=20 >> Intel folks know better but it seems this is hartwell. >>=20 >> em_setup_msix() in very start says: >>=20 >> /* ** Setup MSI/X for Hartwell: tests have shown ** use of two >> queues to be unstable, and to ** provide no great gain anyway, so >> we simply ** seperate the interrupts and use a single queue. */ >>=20 >> Things may have changed now. I guess you can try enabling it and >> find out :-) >>=20 >> cheers, Hiren >>=20 > _______________________________________________=20 > freebsd-net@freebsd.org mailing list=20 > http://lists.freebsd.org/mailman/listinfo/freebsd-net To > unsubscribe, send any mail to > "freebsd-net-unsubscribe@freebsd.org"=20 > _______________________________________________=20 > freebsd-net@freebsd.org mailing list=20 > http://lists.freebsd.org/mailman/listinfo/freebsd-net To > unsubscribe, send any mail to > "freebsd-net-unsubscribe@freebsd.org" >=20 >=20 -----BEGIN PGP SIGNATURE----- Version: GnuPG v2 iQF8BAEBCgBmBQJUz7OLXxSAAAAAAC4AKGlzc3Vlci1mcHJAbm90YXRpb25zLm9w ZW5wZ3AuZmlmdGhob3JzZW1hbi5uZXRCQUFENDYzMkU3MTIxREU4RDIwOTk3REQx MjAxRUZDQTFFNzI3RTY0AAoJEBIB78oecn5kYn0IALVCtNNmWMWIRLlbcVGDg9wo KUbKpvN4UBxOuAvsav8Hussxvy+gh4UXZvgqZ2opTElRrPiUb/iGXa967LWsRaTB TrbvnFE7rJp2xGVlG+rsID+wSdsEAX/isTJWOvpWIqPEULaFvtFh/LUPUrux51Ca SPNzJ+LAh/vWk4sOXN+4PxICiaprlRDs0HF/Mqh5mh8W5TwE3OZy74js7izhNqZS NgowzDyYOq6uhxHZRAC5/GUdpX+ybyCRFZgqKBrNy8BYZV9f+wts2vh7IzwDBSE+ 0ikcM2QGunsr9HH2mnermE2lYF4QKL1Zm8m0mh+TNNIA1O/8Qi4mbE1uMdrzJv4=3D =3D9wHn -----END PGP SIGNATURE-----