From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mails.dpdk.org (mails.dpdk.org [217.70.189.124]) by inbox.dpdk.org (Postfix) with ESMTP id 1B286A0C41; Wed, 17 Nov 2021 13:39:45 +0100 (CET) Received: from [217.70.189.124] (localhost [127.0.0.1]) by mails.dpdk.org (Postfix) with ESMTP id 99CF2426ED; Wed, 17 Nov 2021 13:39:44 +0100 (CET) Received: from NAM12-MW2-obe.outbound.protection.outlook.com (mail-mw2nam12on2049.outbound.protection.outlook.com [40.107.244.49]) by mails.dpdk.org (Postfix) with ESMTP id EADA4426EC for ; Wed, 17 Nov 2021 13:39:42 +0100 (CET) ARC-Seal: i=1; a=rsa-sha256; s=arcselector9901; d=microsoft.com; cv=none; b=Qn/8Y5mS3OECj6HY/x1TVg1rLiFq1GsQBqgLUy8IOePr28lGhoXyoBXke4RvQOEwDLDJOUgErW9QawjYpcBhl79ywjYUsABJ1xzRPTBlaJzbfGObercSYFbUG0yiBVrOWLLa1wLGhy8gvq9EJD5lfmxPLLLoji5FruaxokPN7fcTwxHB/BxhGLYXoRP+qZWa/Gdu6YiNjBIkJEfFQn5haL6SXrLSAEvU2Tw/ABzgboRMLI6wgdeNuWWkPSPQb9t6pFxGXpisT8exWUVykqYm2zvtalcGPdtf6+NJT8X2aqyC5IP26SsX20K2KtEc6xrh5JXA/vHMNLR8IhzSHOx0GQ== ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=microsoft.com; s=arcselector9901; h=From:Date:Subject:Message-ID:Content-Type:MIME-Version:X-MS-Exchange-AntiSpam-MessageData-ChunkCount:X-MS-Exchange-AntiSpam-MessageData-0:X-MS-Exchange-AntiSpam-MessageData-1; bh=kNm8mLNMJX3QWwe8WoY/QcWekEYfwDn2a4MvNPz1lQI=; b=JtJHuBU2E2UgxMC/6F4h5J9HuE+djD16Xu3LSe75eXUkf8WKVA/b/YMUQRqEKERwwpTS/B5+HkEgnuJeoRGYwq7HhEdWJbAXb9zl3fqgkjmbMARkKZzrMjEu8Pv6uMRM8uxFyUJar/wNeFxXtp/j/d2sKm04T/XDgfQGtCOQELjVvwjm23x6e1yFeZ6ea3P7KzztUIcaw0RIqhpmOQFwhrQ0rzldKXsdqU9MWH8bnISlsUtV5w0YK4rldN80toYByz2+E5dC4kjVP0l0ESLnWFDa59cMMkJBlJpA6EZBzB2scJgmYO2jaPBCX14r5Rk7gyiuPj51UsdUHqH4YDqHyQ== ARC-Authentication-Results: i=1; mx.microsoft.com 1; spf=pass smtp.mailfrom=nvidia.com; dmarc=pass action=none header.from=nvidia.com; dkim=pass header.d=nvidia.com; arc=none DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=Nvidia.com; s=selector2; h=From:Date:Subject:Message-ID:Content-Type:MIME-Version:X-MS-Exchange-SenderADCheck; bh=kNm8mLNMJX3QWwe8WoY/QcWekEYfwDn2a4MvNPz1lQI=; b=pqaWBVaoU1/uaKQ22BtW92ZBFxbfWpe6uj3unBjvIpVny5+qYj2SN8Q9uN4bBTi+JGPHM+gY5KEZ33/dY5XeerbY8oSVbwOVduxmbV+n6FD57ty6U8dP17wsxlh69rRyjuCW2b6nrvgtsitMqJKPKgs7mR9vz4ho/Pl4qV2aCYYApRgCri+nd2MW7luLTjew6Y07if2kwEpZ5z0JdsVrUMHIBhsGbSm3qfusdWIUre6GNu+thLQvbtjdpV+HjoZhrn1+3EUgmJD0LqUHrSnQMOO5n81OzLWTy8bQzkRLIWx9OVXYZKPMfSLRv2FBQk1rW2pEFyRbMWVbCEq8cj8E7g== Received: from DM6PR12MB4107.namprd12.prod.outlook.com (2603:10b6:5:218::7) by DM5PR12MB1484.namprd12.prod.outlook.com (2603:10b6:4:11::10) with Microsoft SMTP Server (version=TLS1_2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id 15.20.4690.16; Wed, 17 Nov 2021 12:39:41 +0000 Received: from DM6PR12MB4107.namprd12.prod.outlook.com ([fe80::98ea:e961:8212:62f4]) by DM6PR12MB4107.namprd12.prod.outlook.com ([fe80::98ea:e961:8212:62f4%7]) with mapi id 15.20.4713.021; Wed, 17 Nov 2021 12:39:41 +0000 From: Elena Agostini To: "Ananyev, Konstantin" , "Yigit, Ferruh" , Jerin Jacob CC: "Richardson, Bruce" , Stephen Hemminger , "dev@dpdk.org" Subject: Re: [PATCH v3 1/1] app/testpmd: add GPU memory option for mbuf pools Thread-Topic: [PATCH v3 1/1] app/testpmd: add GPU memory option for mbuf pools Thread-Index: AQHX2xtuLnccivIdaEyRk5cICPLtBqwGrVmAgADi4JqAAATLgIAAAH8ygAABeoCAAATgAIAADV+AgAAAvt4= Date: Wed, 17 Nov 2021 12:39:40 +0000 Message-ID: References: <20211029204909.21318-1-eagostini@nvidia.com> <20211117030459.8274-1-eagostini@nvidia.com> <20211117030459.8274-2-eagostini@nvidia.com> <20211116133449.7b7d21d1@hermes.local> <20243569-b7f0-53c5-02c5-ba29734e30c2@intel.com> In-Reply-To: Accept-Language: en-US Content-Language: en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: authentication-results: dkim=none (message not signed) header.d=none;dmarc=none action=none header.from=nvidia.com; x-ms-publictraffictype: Email x-ms-office365-filtering-correlation-id: 1bde1da8-333e-48bb-a7ea-08d9a9c75345 x-ms-traffictypediagnostic: DM5PR12MB1484: x-microsoft-antispam-prvs: x-ms-oob-tlc-oobclassifiers: OLM:9508; x-ms-exchange-senderadcheck: 1 x-ms-exchange-antispam-relay: 0 x-microsoft-antispam: BCL:0; x-microsoft-antispam-message-info: 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 x-forefront-antispam-report: CIP:255.255.255.255; CTRY:; LANG:en; SCL:1; SRV:; IPV:NLI; SFV:NSPM; H:DM6PR12MB4107.namprd12.prod.outlook.com; PTR:; CAT:NONE; SFS:(4636009)(366004)(66446008)(66946007)(66556008)(38070700005)(64756008)(86362001)(8676002)(66476007)(33656002)(2906002)(186003)(4326008)(38100700002)(5660300002)(122000001)(52536014)(26005)(83380400001)(55016002)(71200400001)(7696005)(9686003)(316002)(110136005)(54906003)(508600001)(8936002)(91956017)(6506007)(76116006); DIR:OUT; SFP:1101; x-ms-exchange-antispam-messagedata-chunkcount: 1 x-ms-exchange-antispam-messagedata-0: =?us-ascii?Q?8Qj/9p1n498Js+0Jd6mQ422TPuqwcX2i9TOedzpV/9WXueimN0Ai6I0sxJqP?= =?us-ascii?Q?9mhNdJ+Y08KptsUlQpFeF8Bo6Y+3AAhBvFtPGonpbORFczAKwIkCmbW7qt3P?= =?us-ascii?Q?9gqJREyciy0e4+Qpnk6vv90gBTwyLcG1ldzGD0JqhROLwWiNLEvPYXymLp7w?= =?us-ascii?Q?rqrvdPCQ245r7T+Vg+MVpNAT04y7CPHmhtRTtftXB709E/b+N6J6RhAFcUss?= =?us-ascii?Q?fhoSS9a9m3gIHvhRyIeLSxSlByljUkoih7L2x91OJaA8DZP8zkhJxZpVpebI?= =?us-ascii?Q?wGUgbdorS2i0JP2ZVUfsLd4fcbuumMi6zYt57UDj5fTBGQzt9yI7XzTiQ5Iq?= =?us-ascii?Q?SoxLX/GQo8u5mcTUvQf709gr/6eCiq/rUMNjaRHlaHdaIB32Ox3khP/9DgZS?= =?us-ascii?Q?/+55D3H9YPXYe/wzWMSNIrwqdE0fgERyQFdaUNfeyts7mQr0ehbPPCQ4IrMr?= =?us-ascii?Q?5xMz1hNaYAOsZrjS6pmygRvjeZYMYMkhoYdrC2Bd/Gc0o1R/p+pPEGcSthkm?= =?us-ascii?Q?8mZh4onEZkHe7jfveSoJUh7Xl2XlK+EKfiNi1eVzKnN9cso2hf4ay3b3oOpi?= =?us-ascii?Q?q0JA2PwbscM6VblTjgf0uCJnOmkydbnMXKycwFZRMge2Gg7PXmoTooEph4na?= =?us-ascii?Q?fv2c+R1g+v2p3Tq7rHGOPFg/6Bfg3T+bbe84Ut/QeUCuATzw/EeNi2G7UpVA?= =?us-ascii?Q?xmDB2e5KsUi38HI23xuN/JOI7wUuf6Kxu3o+NvrI/g5Cw4gY1TSDurIexQ38?= =?us-ascii?Q?g57GdrizVYqxhDSPSYCtI2nUl9S1M4ZHFtHS0zAiFvPHAa6HpcBOSf+q7SeX?= =?us-ascii?Q?I8ONTQpF14ZDQWBHkWukR/1/SHOw8zhKy7XQRaZ5nSe2Rjg8elQ5h9E700ZW?= =?us-ascii?Q?2YVxWX5YHJN8tVQtpsWDfzEML5JJr1YApTnDCLbeyEnzdHiHPZeEo3ZNN3XC?= =?us-ascii?Q?afoD6fXk88QmP5+acpGNxU8PuB5YImYKFp6QlUgflX0TBScU5DG/3gh882Hq?= =?us-ascii?Q?ttJU/D7k2fbpT3zhQbBJyR36g83qJTClELbAZD8hJTmGvbb8svQLQ2puw004?= =?us-ascii?Q?qOewurRG2bXUG3MZsyslqcSnVkCi86SiKPOjcSsb4cQKaAlIxSYHqDt2oQse?= =?us-ascii?Q?CXytja8rwKTcb5o49L04hydMRMlKMpJhr5TnwRmEiWONfvhqheQTr3jKlIoZ?= =?us-ascii?Q?ejn7SQ7P2a4snDIyiP4Ayy1fG7Su8YUu+vaclRt9bVNXDZ8fD0PX5X18mqKf?= =?us-ascii?Q?2FTuo3FgEjpqV06M0dTl2F8REfc3wwzqUMAKL53/HDRZ701JWQ/rV8cJ9GfE?= =?us-ascii?Q?WAzs0WSrkmmsqlm6KVFewp8xAXECJlK/jkcUpyHvdT2C31QczNGnz6Eim1zb?= =?us-ascii?Q?SpEbOcuhx/3AmKeLopJtzQY7mrhzznpA8pQa2wqdvr8saWsQ01oSvp67Uypj?= =?us-ascii?Q?IN8xo2vf5BgbIIT/VBeMjudAFN+tTver+QK7YXwMr45nnlCbM3zkKammyBoW?= =?us-ascii?Q?nhKsUYku9GXh8qQmr6BELcza1cJiW3Y6/SG6GFo8f7kaJc8IlCLIdqOJXVRF?= =?us-ascii?Q?NTVGJiX3wOkj3XwGmzQXjsV+kIZn+2Xk4A6rLlaJbMRfZMWBVx6mdhgIT2pK?= =?us-ascii?Q?gYWAg0Kd/4z9sO1VzmouN0rM0KnWi7J/WHe6yRiOlfaU?= Content-Type: multipart/alternative; boundary="_000_DM6PR12MB4107DB7352F18469B91288B0CD9A9DM6PR12MB4107namp_" MIME-Version: 1.0 X-OriginatorOrg: Nvidia.com X-MS-Exchange-CrossTenant-AuthAs: Internal X-MS-Exchange-CrossTenant-AuthSource: DM6PR12MB4107.namprd12.prod.outlook.com X-MS-Exchange-CrossTenant-Network-Message-Id: 1bde1da8-333e-48bb-a7ea-08d9a9c75345 X-MS-Exchange-CrossTenant-originalarrivaltime: 17 Nov 2021 12:39:40.9459 (UTC) X-MS-Exchange-CrossTenant-fromentityheader: Hosted X-MS-Exchange-CrossTenant-id: 43083d15-7273-40c1-b7db-39efd9ccc17a X-MS-Exchange-CrossTenant-mailboxtype: HOSTED X-MS-Exchange-CrossTenant-userprincipalname: 1TpachjR6WQ6XhhMVeZazpge8Mg2ffQ5BVN7iO3YSlU09y+kEb6/pSRMJ6bTxRrq42jbkiEDUE04+64TqrYAog== X-MS-Exchange-Transport-CrossTenantHeadersStamped: DM5PR12MB1484 X-BeenThere: dev@dpdk.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: DPDK patches and discussions List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: dev-bounces@dpdk.org --_000_DM6PR12MB4107DB7352F18469B91288B0CD9A9DM6PR12MB4107namp_ Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: quoted-printable > > >> > > >>>> > > >> > > >>>>> On Wed, 17 Nov 2021 03:04:59 +0000 > > >> > > >>>> > > >>>>>> This patch introduces GPU memory in testpmd through the gpudev l= ibrary. > > >> > > >>>> > > >> > > >>>>>> Testpmd can be used for network benchmarks when using GPU memory > > >> > > >>>> > > >> > > >>>>>> instead of regular CPU memory to send and receive packets. > > >> > > >>>> > > >> > > >>>>>> This option is currently limited to iofwd engine to ensure > > >> > > >>>> > > >> > > >>>>>> no workload is applied on packets not accessible from the CPU. > > >> > > >>>> > > >> > > >>>>>> > > >> > > >>>> > > >> > > >>>>>> The options chose is --mbuf-size so buffer split feature across > > >> > > >>>> > > >> > > >>>>>> different mempools can be enabled. > > >> > > >>>> > > >> > > >>>>>> > > >> > > >>>> > > >> > > >>>>>> Signed-off-by: Elena Agostini > > >> > > >>>> > > >> > > >>>>> > > >> > > >>>> > > >> > > >>>>> Won't this create a hard dependency of test-pmd on gpudev? > > >> > > >>>> > > >> > > >>>>> I thought gpudev was supposed to be optional > > >> > > >>>> > > >> > > >>>> > > >> > > >>>> > > >> > > >>>> Sure, let me submit another patch to make it optional > > >> > > >>> > > >> > > >>> Why to add yet another compile time macro everywhere in testpmd and > > >> > > >>> make hard to maintain? > > >> > > >>> Adding iofwd kind of code is very simple to add test/test-gpudev an= d > > >> > > >>> all GPU specific options > > >> > > >>> can be added in test-gpudev. It also helps to review the patches as > > >> > > >>> test cases focus on > > >> > > >>> each device class. > > >> > > >> > > >> > > >> Test-gpudev is standalone unit test to ensure gpudev functions work = correctly. > > >> > > >> In testpmd instead, there is a connection between gpudev and the net= work. > > > > > > I understand that. We had the same case with eventdev, where it needs= to > > > work with network. Testpmd is already complicated, IMO, we should > > > focus only ethdev > > > test cases on testpmd, test-gpudev can use ethdev API to enable > > > networking requirements for gpudev. > > > > > > > +1 > > +1 Testpmd already manages different type of memories for mempools. gpudev is just another type of memory, there is nothing more than that. --_000_DM6PR12MB4107DB7352F18469B91288B0CD9A9DM6PR12MB4107namp_ Content-Type: text/html; charset="us-ascii" Content-Transfer-Encoding: quoted-printable

> > >><= o:p>

> > >>&= gt;>

> > >><= o:p>

> > >>&= gt;>> On Wed, 17 Nov 2021 03:04:59 +0000

> > >><= o:p>

> > >>&= gt;>

> > >>&= gt;>>> This patch introduces GPU memory in testpmd through the gpu= dev library.

> > >><= o:p>

> > >>&= gt;>

> > >><= o:p>

> > >>&= gt;>>> Testpmd can be used for network benchmarks when using GPU m= emory

> > >><= o:p>

> > >>&= gt;>

> > >><= o:p>

> > >>&= gt;>>> instead of regular CPU memory to send and receive packets.<= o:p>

> > >><= o:p>

> > >>&= gt;>

> > >><= o:p>

> > >>&= gt;>>> This option is currently limited to iofwd engine to ensure<= o:p>

> > >><= o:p>

> > >>&= gt;>

> > >><= o:p>

> > >>&= gt;>>> no workload is applied on packets not accessible from the C= PU.

> > >><= o:p>

> > >>&= gt;>

> > >><= o:p>

> > >>&= gt;>>>

> > >><= o:p>

> > >>&= gt;>

> > >><= o:p>

> > >>&= gt;>>> The options chose is --mbuf-size so buffer split feature ac= ross

> > >><= o:p>

> > >>&= gt;>

> > >><= o:p>

> > >>&= gt;>>> different mempools can be enabled.

> > >><= o:p>

> > >>&= gt;>

> > >><= o:p>

> > >>&= gt;>>>

> > >><= o:p>

> > >>&= gt;>

> > >><= o:p>

> > >>&= gt;>>> Signed-off-by: Elena Agostini <eagostini@nvidia.com><= o:p>

> > >><= o:p>

> > >>&= gt;>

> > >><= o:p>

> > >>&= gt;>>

> > >><= o:p>

> > >>&= gt;>

> > >><= o:p>

> > >>&= gt;>> Won't this create a hard dependency of test-pmd on gpudev?=

> > >><= o:p>

> > >>&= gt;>

> > >><= o:p>

> > >>&= gt;>> I thought gpudev was supposed to be optional<= /p>

> > >><= o:p>

> > >>&= gt;>

> > >><= o:p>

> > >>&= gt;>

> > >><= o:p>

> > >>&= gt;>

> > >><= o:p>

> > >>&= gt;> Sure, let me submit another patch to make it optional

> > >><= o:p>

> > >>&= gt;

> > >><= o:p>

> > >>&= gt; Why to add yet another compile time macro everywhere in testpmd and

> > >><= o:p>

> > >>&= gt; make hard to maintain?

> > >><= o:p>

> > >>&= gt; Adding iofwd kind of code is very simple to add test/test-gpudev and

> > >><= o:p>

> > >>&= gt; all GPU specific options

> > >><= o:p>

> > >>&= gt; can be added in test-gpudev. It also helps to review the patches as

> > >><= o:p>

> > >>&= gt; test cases focus on

> > >><= o:p>

> > >>&= gt; each device class.

> > >><= o:p>

> > >><= o:p>

> > >><= o:p>

> > >> = Test-gpudev is standalone unit test to ensure gpudev functions work correct= ly.

> > >><= o:p>

> > >> = In testpmd instead, there is a connection between gpudev and the network.

> > >=

> > > I un= derstand that. We had the same case with eventdev, where it needs to

> > > work= with network. Testpmd is already complicated, IMO, we should

> > > focu= s only ethdev

> > > test= cases on testpmd, test-gpudev can use ethdev API to enable

> > > netw= orking requirements for gpudev.

> > >=

> >

> > +1

>

> +1<= /span>

 

Test= pmd already manages different type of memories for mempools.

gpud= ev is just another type of memory, there is nothing more than that.

--_000_DM6PR12MB4107DB7352F18469B91288B0CD9A9DM6PR12MB4107namp_--