From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mga09.intel.com (mga09.intel.com [134.134.136.24]) by dpdk.org (Postfix) with ESMTP id 7F9E01B63E for ; Mon, 23 Oct 2017 20:27:54 +0200 (CEST) Received: from orsmga005.jf.intel.com ([10.7.209.41]) by orsmga102.jf.intel.com with ESMTP/TLS/DHE-RSA-AES256-GCM-SHA384; 23 Oct 2017 11:27:53 -0700 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="5.43,424,1503385200"; d="scan'208";a="163926871" Received: from fmsmsx104.amr.corp.intel.com ([10.18.124.202]) by orsmga005.jf.intel.com with ESMTP; 23 Oct 2017 11:27:52 -0700 Received: from fmsmsx124.amr.corp.intel.com (10.18.125.39) by fmsmsx104.amr.corp.intel.com (10.18.124.202) with Microsoft SMTP Server (TLS) id 14.3.319.2; Mon, 23 Oct 2017 11:27:52 -0700 Received: from fmsmsx108.amr.corp.intel.com ([169.254.9.115]) by fmsmsx124.amr.corp.intel.com ([169.254.8.94]) with mapi id 14.03.0319.002; Mon, 23 Oct 2017 11:27:52 -0700 From: "Eads, Gage" To: Jerin Jacob , "dev@dpdk.org" CC: "Richardson, Bruce" , "Van Haaren, Harry" , Hemant Agrawal , Nipun Gupta , "Rao, Nikhil" , Pavan Nikhilesh , Thomas Monjalon Thread-Topic: [dpdk-dev] [PATCH] eventdev: remove experimental label Thread-Index: AQHTRmpTRlK6K7CWvkG/Xm4GCmCXCqLxp6Pg Date: Mon, 23 Oct 2017 18:27:52 +0000 Message-ID: <9184057F7FC11744A2107296B6B8EB1E1400BE33@FMSMSX108.amr.corp.intel.com> References: <20171016103255.16322-1-jerin.jacob@caviumnetworks.com> In-Reply-To: <20171016103255.16322-1-jerin.jacob@caviumnetworks.com> Accept-Language: en-US Content-Language: en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: x-titus-metadata-40: eyJDYXRlZ29yeUxhYmVscyI6IiIsIk1ldGFkYXRhIjp7Im5zIjoiaHR0cDpcL1wvd3d3LnRpdHVzLmNvbVwvbnNcL0ludGVsMyIsImlkIjoiMWJmYjk4NzYtNmYxOS00YjBiLWFlZDEtODQzMzBmY2U0OWEzIiwicHJvcHMiOlt7Im4iOiJDVFBDbGFzc2lmaWNhdGlvbiIsInZhbHMiOlt7InZhbHVlIjoiQ1RQX0lDIn1dfV19LCJTdWJqZWN0TGFiZWxzIjpbXSwiVE1DVmVyc2lvbiI6IjE2LjUuOS4zIiwiVHJ1c3RlZExhYmVsSGFzaCI6Ik5PTzdOd0JLSGpyZEZhWjNrUmt0WUs0XC9pRmxDdzRyQU5ieUtzZmVoVG1RPSJ9 x-ctpclassification: CTP_IC dlp-product: dlpe-windows dlp-version: 11.0.0.116 dlp-reaction: no-action x-originating-ip: [10.1.200.106] Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: quoted-printable MIME-Version: 1.0 Subject: Re: [dpdk-dev] [PATCH] eventdev: remove experimental label 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, 23 Oct 2017 18:27:55 -0000 Hi Jerin, I have one concern with the API that may delay changing the label. The implicit release that in rte_event_dequeue_burst() is a problem when us= ing asynchronous/look-aside hardware, like a cryptodev. For instance, let's= say in pipeline stage N the worker takes the event's mbuf and places it in= a per-worker crypto request queue. When the worker next calls rte_event_de= queue_burst(), that function will release the previous event which could ca= use the flow to migrate to another worker, and this could result in packet = reordering. To prevent this, the worker can't call dequeue until the look-aside operati= on completes...in effect treating the asynchronous/look-aside hardware as s= ynchronous. Another option is to feed stage N's queue to a single port to a= void the flow migration, but that port may become a bottleneck. We could remove the implicit release functionality or add a port configurat= ion flag to disable it, so the default behavior is unchanged. Removing it w= ill completely will likely require changes in existing code, but it simplif= ies the usage model (all dequeued events must be either forwarded or releas= ed) and the PMD's dequeue code. This functionality could be removed from th= e software eventdev fairly easily, but I haven't looked into the hardware P= MDs. Thanks, Gage > -----Original Message----- > From: Jerin Jacob [mailto:jerin.jacob@caviumnetworks.com] > Sent: Monday, October 16, 2017 5:33 AM > To: dev@dpdk.org > Cc: Jerin Jacob ; Richardson, Bruce > ; Van Haaren, Harry > ; Eads, Gage ; Hemant > Agrawal ; Nipun Gupta ; > Rao, Nikhil ; Pavan Nikhilesh > ; Thomas Monjalon > > Subject: [dpdk-dev] [PATCH] eventdev: remove experimental label >=20 > The eventdev API was introduced in DPDK 17.05 release. > Since then it > - has been reviewed and iterated for 17.08, 17.11 releases > - three drivers were implemented using the API. > - introduced another subsystem like service core and ethdev-eventdev Rx > adapter APIs to abstract the difference between HW and SW eventdev > implementations in a transparent way. > - had extensive use by the app/test-eventdev/ and > examples/eventdev_pipeline_sw_pmd/ >=20 > I believe the API is now stable and the EXPERIMENTAL label should be remo= ved. >=20 > CC: Bruce Richardson > CC: Harry van Haaren > CC: Gage Eads > CC: Hemant Agrawal > CC: Nipun Gupta > CC: Nikhil Rao > CC: Pavan Nikhilesh > CC: Thomas Monjalon > Signed-off-by: Jerin Jacob > --- >=20 > There are two more outstanding eventdev API changes. Please find below. > Please express if you have any concern in changing those APIs. I would > like to fix this API issue and remove experimental tag in v17.11, > if we all agree. >=20 > - evendev: fix inconsistency in event queue config > http://dpdk.org/dev/patchwork/patch/30293/ > - remove rte_event_schedule() API and use service core infrastructure > http://dpdk.org/dev/patchwork/patch/30375/ >=20 > --- > MAINTAINERS | 2 +- > 1 file changed, 1 insertion(+), 1 deletion(-) >=20 > diff --git a/MAINTAINERS b/MAINTAINERS > index 2a58378b7..4a4be3a54 100644 > --- a/MAINTAINERS > +++ b/MAINTAINERS > @@ -262,7 +262,7 @@ F: lib/librte_cryptodev/ > F: test/test/test_cryptodev* > F: examples/l2fwd-crypto/ >=20 > -Eventdev API - EXPERIMENTAL > +Eventdev API > M: Jerin Jacob > T: git://dpdk.org/next/dpdk-next-eventdev > F: lib/librte_eventdev/ > -- > 2.14.2