From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: <dev-bounces@dpdk.org> Received: from mails.dpdk.org (mails.dpdk.org [217.70.189.124]) by inbox.dpdk.org (Postfix) with ESMTP id 9B707A034F; Mon, 8 Nov 2021 03:44:28 +0100 (CET) Received: from [217.70.189.124] (localhost [127.0.0.1]) by mails.dpdk.org (Postfix) with ESMTP id 1A21140151; Mon, 8 Nov 2021 03:44:28 +0100 (CET) Received: from mga04.intel.com (mga04.intel.com [192.55.52.120]) by mails.dpdk.org (Postfix) with ESMTP id BA6D240040 for <dev@dpdk.org>; Mon, 8 Nov 2021 03:44:26 +0100 (CET) X-IronPort-AV: E=McAfee;i="6200,9189,10161"; a="230869604" X-IronPort-AV: E=Sophos;i="5.87,217,1631602800"; d="scan'208";a="230869604" Received: from orsmga004.jf.intel.com ([10.7.209.38]) by fmsmga104.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 07 Nov 2021 18:44:25 -0800 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="5.87,217,1631602800"; d="scan'208";a="601254652" Received: from fmsmsx602.amr.corp.intel.com ([10.18.126.82]) by orsmga004.jf.intel.com with ESMTP; 07 Nov 2021 18:44:25 -0800 Received: from shsmsx604.ccr.corp.intel.com (10.109.6.214) by fmsmsx602.amr.corp.intel.com (10.18.126.82) with Microsoft SMTP Server (version=TLS1_2, cipher=TLS_ECDHE_RSA_WITH_AES_128_GCM_SHA256) id 15.1.2242.12; Sun, 7 Nov 2021 18:44:24 -0800 Received: from shsmsx601.ccr.corp.intel.com (10.109.6.141) by SHSMSX604.ccr.corp.intel.com (10.109.6.214) with Microsoft SMTP Server (version=TLS1_2, cipher=TLS_ECDHE_RSA_WITH_AES_128_GCM_SHA256) id 15.1.2242.12; Mon, 8 Nov 2021 10:44:22 +0800 Received: from shsmsx601.ccr.corp.intel.com ([10.109.6.141]) by SHSMSX601.ccr.corp.intel.com ([10.109.6.141]) with mapi id 15.01.2242.012; Mon, 8 Nov 2021 10:44:22 +0800 From: "Zhang, Qi Z" <qi.z.zhang@intel.com> To: "Xu, Ting" <ting.xu@intel.com>, Thomas Monjalon <thomas@monjalon.net> CC: "dev@dpdk.org" <dev@dpdk.org>, "Yang, Qiming" <qiming.yang@intel.com>, "Guo, Junfeng" <junfeng.guo@intel.com>, "orika@nvidia.com" <orika@nvidia.com>, "viacheslavo@nvidia.com" <viacheslavo@nvidia.com>, "Yigit, Ferruh" <ferruh.yigit@intel.com>, "andrew.rybchenko@oktetlabs.ru" <andrew.rybchenko@oktetlabs.ru> Thread-Topic: [dpdk-dev] [PATCH v4 0/2] enable protocol agnostic flow offloading in RSS Thread-Index: AQHX0SKIxyMfN/h4bEWTGgAbptQM6qv0ZGqAgAACjACAAAEvAIAAGU4AgARg0OA= Date: Mon, 8 Nov 2021 02:44:22 +0000 Message-ID: <212e24c9848f4fd382f032d3558c099e@intel.com> References: <20211008070934.6956-1-ting.xu@intel.com> <2404603.ePvfOphbkP@thomas> <CO6PR11MB56345CEACEA35A35E23BE0A4F88E9@CO6PR11MB5634.namprd11.prod.outlook.com> <4597919.4CLRmV2U1y@thomas> <CO6PR11MB56340A59CB99AA6A1653CB24F88E9@CO6PR11MB5634.namprd11.prod.outlook.com> In-Reply-To: <CO6PR11MB56340A59CB99AA6A1653CB24F88E9@CO6PR11MB5634.namprd11.prod.outlook.com> Accept-Language: en-US Content-Language: en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: dlp-reaction: no-action dlp-version: 11.6.200.16 dlp-product: dlpe-windows x-originating-ip: [10.239.127.36] Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: quoted-printable MIME-Version: 1.0 Subject: Re: [dpdk-dev] [PATCH v4 0/2] enable protocol agnostic flow offloading in RSS X-BeenThere: dev@dpdk.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: DPDK patches and discussions <dev.dpdk.org> List-Unsubscribe: <https://mails.dpdk.org/options/dev>, <mailto:dev-request@dpdk.org?subject=unsubscribe> List-Archive: <http://mails.dpdk.org/archives/dev/> List-Post: <mailto:dev@dpdk.org> List-Help: <mailto:dev-request@dpdk.org?subject=help> List-Subscribe: <https://mails.dpdk.org/listinfo/dev>, <mailto:dev-request@dpdk.org?subject=subscribe> Errors-To: dev-bounces@dpdk.org Sender: "dev" <dev-bounces@dpdk.org> > -----Original Message----- > From: Xu, Ting <ting.xu@intel.com> > Sent: Friday, November 5, 2021 10:54 PM > To: Thomas Monjalon <thomas@monjalon.net>; Zhang, Qi Z > <qi.z.zhang@intel.com> > Cc: dev@dpdk.org; Yang, Qiming <qiming.yang@intel.com>; Guo, Junfeng > <junfeng.guo@intel.com>; orika@nvidia.com; viacheslavo@nvidia.com; > Yigit, Ferruh <ferruh.yigit@intel.com>; andrew.rybchenko@oktetlabs.ru > Subject: RE: [dpdk-dev] [PATCH v4 0/2] enable protocol agnostic flow > offloading in RSS >=20 > > -----Original Message----- > > From: Thomas Monjalon <thomas@monjalon.net> > > Sent: Friday, November 5, 2021 9:23 PM > > To: Zhang, Qi Z <qi.z.zhang@intel.com>; Xu, Ting <ting.xu@intel.com> > > Cc: dev@dpdk.org; Yang, Qiming <qiming.yang@intel.com>; Guo, Junfeng > > <junfeng.guo@intel.com>; orika@nvidia.com; viacheslavo@nvidia.com; > > Yigit, Ferruh <ferruh.yigit@intel.com>; andrew.rybchenko@oktetlabs.ru > > Subject: Re: [dpdk-dev] [PATCH v4 0/2] enable protocol agnostic flow > > offloading in RSS > > > > 05/11/2021 14:18, Xu, Ting: > > > From: Thomas Monjalon <thomas@monjalon.net> > > > > 04/11/2021 03:22, Ting Xu: > > > > > Enable protocol agnostic flow offloading to support raw pattern > > > > > input for RSS hash flow rule creation. It is based on Parser > > > > > Library > > feature. > > > > > Current rte_flow raw API is utilized. > > > > > > > > I remember raw item was especially targetting encapsulation action. > > > > Which action do you achieve in ice PMD? Only RSS with raw item? > > > > > > > > For matching, raw item is quite limited. I think the new flex item > > > > is more appropriate for general custom protocol parsing. > > > > Did you look into flex item feature? > > > > > > > Hi, Thomas, we make use of the API for rte_raw to achieve our > > > protocol agnostic flow offloading method, including RSS and FDIR. > > > Behind it is the new Parser Library feature based on DDP, > > > > Where is this parser library code? >=20 > Parser Library is committed before as ice base code. >=20 > > > > > rather than the original logic I think. > > > > > Binary-format pattern spec and mask are passed to Parser Library and > > > the corresponding protocol ID and offset will be provided. > > > > Sorry you didn't reply about the new flex item added in this release. > > Do you know it was added? > > https://git.dpdk.org/dpdk/commit/?id=3Ddc4d860e8a89 > > >=20 > Protocol agnostic flow offloading is based on ice DDP feature, although > they all want to make flow creation more flexible, I think it is differen= t > with flex item. > @Zhang, Qi Z Hi, Qi, are you familiar with flex item? >=20 Yes, this is a very Intel DDP specific feature, the goal is to make our dri= ver to support any customized parser (DDP) without change code. Below steps are planned: 1. raw packet interface, this help on quick POC, as its simply the rte_flow= parser, but of cause it's not scale for an DPDK application to support dif= ferent vendor. 2. enable a build-in util to forge a packet from a sequence of rte_flow hea= ders, then the application will be scale, but the missing part is how to de= scribe a customized header, we have similar idea like json, but seems "Flex= Item" already help to fix this gap, and we will improve the solution base = on it. Thanks Qi