From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from EUR03-AM5-obe.outbound.protection.outlook.com (mail-eopbgr30080.outbound.protection.outlook.com [40.107.3.80]) by dpdk.org (Postfix) with ESMTP id 21BB62BF5 for ; Mon, 8 Jan 2018 15:42:39 +0100 (CET) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=Mellanox.com; s=selector1; h=From:Date:Subject:Message-ID:Content-Type:MIME-Version; bh=GbX78ICbb04deQ4hoXxib1UNAhKzOlRr4ki4HumnlHo=; b=Tl9kBlPpsV024Lb54IZRSZAOq3fSzl8LA4R1YnsIy93z8pq8/EdOH6LYiodfonD4XllarLeZuo9GHnnU+BQah6oTwtvg2Qw8u+9jItiLfxCoOCY3m2lGVb9Bdqgzz0y8Xp4Q0zF1Ynmzu8Xo/neY9bnftOsVOX6xCgy84oopnzo= Received: from AM6PR0502MB3797.eurprd05.prod.outlook.com (52.133.21.26) by AM6PR0502MB3799.eurprd05.prod.outlook.com (52.133.21.28) with Microsoft SMTP Server (version=TLS1_2, cipher=TLS_ECDHE_RSA_WITH_AES_256_CBC_SHA384_P256) id 15.20.386.5; Mon, 8 Jan 2018 14:42:31 +0000 Received: from AM6PR0502MB3797.eurprd05.prod.outlook.com ([fe80::b4b4:7de8:cf70:aa3a]) by AM6PR0502MB3797.eurprd05.prod.outlook.com ([fe80::b4b4:7de8:cf70:aa3a%13]) with mapi id 15.20.0386.006; Mon, 8 Jan 2018 14:42:31 +0000 From: Matan Azrad To: =?iso-8859-1?Q?Ga=EBtan_Rivet?= CC: Thomas Monjalon , Jingjing Wu , "dev@dpdk.org" , Neil Horman , Bruce Richardson , Konstantin Ananyev Thread-Topic: [PATCH v2 6/6] app/testpmd: adjust ethdev port ownership Thread-Index: AQHTiHVt5VPixrMnKUK3vfdVuXmfmKNp4uPQgAAV1ICAAAEJkIAADViAgAACDGA= Date: Mon, 8 Jan 2018 14:42:31 +0000 Message-ID: References: <1511870281-15282-1-git-send-email-matan@mellanox.com> <1515318351-4756-1-git-send-email-matan@mellanox.com> <1515318351-4756-7-git-send-email-matan@mellanox.com> <20180108113946.hcgxvulamjsiepre@bidouze.vm.6wind.com> <20180108133015.an547uygefjz5gj4@bidouze.vm.6wind.com> <20180108142143.nk6c6l64nqnurfhz@bidouze.vm.6wind.com> In-Reply-To: <20180108142143.nk6c6l64nqnurfhz@bidouze.vm.6wind.com> Accept-Language: en-US, he-IL Content-Language: en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: authentication-results: spf=none (sender IP is ) smtp.mailfrom=matan@mellanox.com; x-originating-ip: [193.47.165.251] x-ms-publictraffictype: Email x-microsoft-exchange-diagnostics: 1; AM6PR0502MB3799; 7:kEgJM2VF7ORg9Db5Mu2DhiKrMXPj8YsmaJypCcZ3tItrLS9mVh5SzdoZfVDhsdh7YM5k67p+pbkzgnMbLXap1mpY4GmJWUTLdf9bdaIWKOci+fxFM+PZpyM8S4Ijtc9Mo50ZYtiAshopwLCoL5Vppi3Q1MdrpcoJpp2GD4pMnlRHTe++XissoilRd7CwFMDJqCQaWqt4zpLKVaCUxRalGFfev2eDJvWIomZUo3uCfF3T6sLurb64fxG8BcBcxoL4 x-ms-exchange-antispam-srfa-diagnostics: SSOS; x-ms-office365-filtering-ht: Tenant x-ms-office365-filtering-correlation-id: e6755002-0359-4563-c48d-08d556a60c2f x-microsoft-antispam: UriScan:; BCL:0; PCL:0; RULEID:(48565401081)(4534020)(4602075)(4627115)(201703031133081)(201702281549075)(5600026)(4604075)(3008032)(2017052603307)(7153060)(7193020); SRVR:AM6PR0502MB3799; x-ms-traffictypediagnostic: AM6PR0502MB3799: x-ld-processed: a652971c-7d2e-4d9b-a6a4-d149256f461b,ExtAddr x-microsoft-antispam-prvs: x-exchange-antispam-report-test: UriScan:(60795455431006)(278428928389397); x-exchange-antispam-report-cfa-test: BCL:0; PCL:0; RULEID:(6040470)(2401047)(8121501046)(5005006)(3231023)(944501075)(3002001)(10201501046)(93006095)(93001095)(6055026)(6041268)(20161123560045)(20161123558120)(20161123564045)(201703131423095)(201702281528075)(20161123555045)(201703061421075)(201703061406153)(20161123562045)(6072148)(201708071742011); SRVR:AM6PR0502MB3799; BCL:0; PCL:0; RULEID:(100000803101)(100110400095); SRVR:AM6PR0502MB3799; x-forefront-prvs: 054642504A x-forefront-antispam-report: SFV:NSPM; SFS:(10009020)(979002)(376002)(39380400002)(39860400002)(366004)(396003)(346002)(24454002)(189003)(199004)(5250100002)(316002)(8936002)(102836004)(54906003)(3846002)(86362001)(97736004)(55016002)(229853002)(6116002)(9686003)(6506007)(81166006)(81156014)(6436002)(93886005)(561944003)(8676002)(4326008)(76176011)(2950100002)(33656002)(53936002)(59450400001)(6246003)(6916009)(3280700002)(3660700001)(14454004)(7696005)(105586002)(25786009)(68736007)(5660300001)(305945005)(99286004)(7736002)(2900100001)(106356001)(478600001)(74316002)(2906002)(66066001)(969003)(989001)(999001)(1009001)(1019001); DIR:OUT; SFP:1101; SCL:1; SRVR:AM6PR0502MB3799; H:AM6PR0502MB3797.eurprd05.prod.outlook.com; FPR:; SPF:None; PTR:InfoNoRecords; MX:3; A:3; LANG:en; received-spf: None (protection.outlook.com: mellanox.com does not designate permitted sender hosts) x-microsoft-antispam-message-info: nfdQOfQNv/Ss+ZRD80+e14ZDdZyGlnMKpJBPpsnNdJJiFYOYxctQ2dTcvVbxE/JyZTeOda8EYRfFEF0rruJxxA== spamdiagnosticoutput: 1:99 spamdiagnosticmetadata: NSPM Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable MIME-Version: 1.0 X-OriginatorOrg: Mellanox.com X-MS-Exchange-CrossTenant-Network-Message-Id: e6755002-0359-4563-c48d-08d556a60c2f X-MS-Exchange-CrossTenant-originalarrivaltime: 08 Jan 2018 14:42:31.1028 (UTC) X-MS-Exchange-CrossTenant-fromentityheader: Hosted X-MS-Exchange-CrossTenant-id: a652971c-7d2e-4d9b-a6a4-d149256f461b X-MS-Exchange-Transport-CrossTenantHeadersStamped: AM6PR0502MB3799 Subject: Re: [dpdk-dev] [PATCH v2 6/6] app/testpmd: adjust ethdev port ownership X-BeenThere: dev@dpdk.org X-Mailman-Version: 2.1.15 Precedence: list List-Id: DPDK patches and discussions List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 08 Jan 2018 14:42:40 -0000 From: Ga=EBtan Rivet, Monday, January 8, 2018 4:22 PM > On Mon, Jan 08, 2018 at 01:55:52PM +0000, Matan Azrad wrote: > > Hi Gaetan > > > > From: Ga=EBtan Rivet, Monday, January 8, 2018 3:30 PM > > > On Mon, Jan 08, 2018 at 12:30:19PM +0000, Matan Azrad wrote: > > > > > > > > > > > > From: Ga=EBtan Rivet, Monday, January 8, 2018 1:40 PM > > > > > Hi Matan, > > > > > > > > > > On Sun, Jan 07, 2018 at 09:45:51AM +0000, Matan Azrad wrote: > > > > > > Testpmd should not use ethdev ports which are managed by other > > > > > > DPDK entities. > > > > > > > > > > > > Set Testpmd ownership to each port which is not used by other > > > > > > entity and prevent any usage of ethdev ports which are not > > > > > > owned by > > > Testpmd. > > > > > > > > > > > > > > > > This patch should not be necessary. > > > > > > > > > > Ideally, your API evolution should not impact the default case. > > > > > As such, the default iterator RTE_ETH_FOREACH_DEV should still > > > > > be used in > > > testpmd. > > > > > > > > > Why? We want to adjust testpmd to the port ownership. > > > > > > > > > > This adjustment should be seamless for existing application. > > > > > > > > RTE_ETH_FOREACH_DEV should call > > > RTE_ETH_FOREACH_DEV_OWNED_BY, with > > > > > the default owner (meaning that it would thus iterate on the > > > > > application-owned set of device). > > > > > > > > > > > > > It will break the API (we already talked about it). > > > > There is not any default owner: > > > > Any DPDK entity includes applications must to allocate an owner ID > > > > and use > > > it to own the ports they wants to use. > > > > The application can include more than 1 owner depends on the user > needs. > > > > Each DPDK entity which can synchronize all its port usage can be a > > > > valid > > > DPDK entity for the ownership mechanism. > > > > > > > > > > That's the point of my remark: you did not include a default owner. > > > I think there should be one, and that all ports should pertain to > > > this default owner by default when created. > > > > > > This would not prevent a user or application from adding new owners > > > specific to their use and specialize ports if need be. > > > > > > However, for other applications that do not care for this > > > specialization, they should run with the current API and avoid the > > > ports that are configured by other third parties. > > > > > > > RTE_ETH_FOREACH_DEV means iterate over all devices and should stay as > is in my opinion. > > I understand your concern about changes in current application, But > > your "default" suggestion will cause to "non-default" applications to r= eset > all the default owners and will complicate them and hurt semantics. >=20 > Why should an application be able to iterate over all ports? Leave this > capability to the EAL (or ethdev layer) alone, while other components sho= uld > be restricted to their specific set. >=20 Yes, you right. > And if a need for this general iterator appears, solutions could be found= very > easily. >=20 > RTE_ETH_FOREACH_DEV currently does not iterate over deferred ports, it > iterates over the base set of ports available. Changing this behavior is = not > necessary, you could introduce your API while keeping it. >=20 Right. > > > > > I'm thinking about applications already written that would be used > > > with fail- safe ports: they would use RTE_ETH_FOREACH_DEV, and would > > > thus iterate over every ports, including those owned by the > > > fail-safe, unless they start following the new API. > > > > > > > They should start, it is really not complicated. >=20 > The point is not whether developpers downstream would be able to grasp > such complexity, but whether a project like DPDK should foster an unstabl= e > environment for its currently still limited ecosystem. >=20 > > What's about application which use count=3Drte_eth_dev_count and iterat= e > over all ports from 0 to count-1? > > We cannot save all the wrong application options. > > > > > This is unnecessary: adding a default owner for all created ports > > > and redefining RTE_ETH_FOREACH_DEV as follow > > > > > > #define RTE_ETH_FOREACH_DEV(i) > > > RTE_ETH_FOREACH_DEV_OWNED_BY(i, RTE_ETH_DEFAULT_OWNER) > > > > > > Is simple enough and will simplify the work of DPDK users. Moreover, > > > it would make fail-safe compatible with all applications using > > > RTE_ETH_FOREACH_DEV without additional evolution. It would actually > > > make any code using your API supported by those same applications, > > > which I think would help its adoption. > > > > > > > Will break API, will hurt semantic of FOREACH , and will complicate > ownership care applications as I wrote above. >=20 > Well, breaking an API is best before such API is integrated anyway. >=20 > I disagree regarding the added complexity for applications that would use > ownership. With your proposal, most applications will only add a single u= ser > and register all their ports with this user, then be forced to iterate up= on their > registered user. >=20 > You can save all of them the hassle of adding this code, by taking care o= f the > most common case, avoiding redundant code downstream and simplifying > possible future update to this default case. >=20 > So if anything, this would greatly simplify ownership for the vast majori= ty of > applications. > OK, got you. I will just document the API with the new semantic and will use the NO_OWNE= R for the old API. But actually I think testpmd should use the ownership mechanism as a good e= xample for it. Thanks! =20 > -- > Ga=EBtan Rivet > 6WIND