From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mga11.intel.com (mga11.intel.com [192.55.52.93]) by dpdk.org (Postfix) with ESMTP id 854EA6AE0 for ; Tue, 4 Oct 2016 23:49:54 +0200 (CEST) Received: from fmsmga006.fm.intel.com ([10.253.24.20]) by fmsmga102.fm.intel.com with ESMTP; 04 Oct 2016 14:49:53 -0700 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="5.31,445,1473145200"; d="scan'208,217";a="16202047" Received: from fmsmsx107.amr.corp.intel.com ([10.18.124.205]) by fmsmga006.fm.intel.com with ESMTP; 04 Oct 2016 14:49:53 -0700 Received: from fmsmsx117.amr.corp.intel.com (10.18.116.17) by fmsmsx107.amr.corp.intel.com (10.18.124.205) with Microsoft SMTP Server (TLS) id 14.3.248.2; Tue, 4 Oct 2016 14:49:53 -0700 Received: from fmsmsx108.amr.corp.intel.com ([169.254.9.94]) by fmsmsx117.amr.corp.intel.com ([169.254.3.249]) with mapi id 14.03.0248.002; Tue, 4 Oct 2016 14:49:53 -0700 From: "Vangati, Narender" To: "dev@dpdk.org" Thread-Topic: [dpdk-dev] [RFC] libeventdev: event driven programming model framework for DPDK Thread-Index: AdIeiSak13I9OJseTpeT3ARREZ5dJA== Date: Tue, 4 Oct 2016 21:49:52 +0000 Message-ID: <3B5C8B5D2B969C4CBE78502867F1AD9A0134149F@FMSMSX108.amr.corp.intel.com> Accept-Language: en-US Content-Language: en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: x-titus-metadata-40: eyJDYXRlZ29yeUxhYmVscyI6IiIsIk1ldGFkYXRhIjp7Im5zIjoiaHR0cDpcL1wvd3d3LnRpdHVzLmNvbVwvbnNcL0ludGVsMyIsImlkIjoiMTE5MjYxOGQtMDQ4ZC00ODUwLTg2ZjMtMTNiOTA1NGE2MDM1IiwicHJvcHMiOlt7Im4iOiJDVFBDbGFzc2lmaWNhdGlvbiIsInZhbHMiOlt7InZhbHVlIjoiQ1RQX0lDIn1dfV19LCJTdWJqZWN0TGFiZWxzIjpbXSwiVE1DVmVyc2lvbiI6IjE1LjkuNi42IiwiVHJ1c3RlZExhYmVsSGFzaCI6Ill3Wmx6cE91WnFTd05Sb0orMFlGZzIzeXlyQzJDOHhOcXFSS2t6Z2hYaWM9In0= x-ctpclassification: CTP_IC x-originating-ip: [10.1.200.107] MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: quoted-printable X-Content-Filtered-By: Mailman/MimeDel 2.1.15 Subject: Re: [dpdk-dev] [RFC] libeventdev: event driven programming model framework for DPDK X-BeenThere: dev@dpdk.org X-Mailman-Version: 2.1.15 Precedence: list List-Id: patches and discussions about DPDK List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 04 Oct 2016 21:49:55 -0000 Hi Jerin, Here are some comments on the libeventdev RFC. These are collated thoughts after discussions with you & others to understa= nd the concepts and rationale for the current proposal. 1. Concept of flow queues. This is better abstracted as flow ids and not as= flow queues which implies there is a queueing structure per flow. A s/w im= plementation can do atomic load balancing on multiple flow ids more efficie= ntly than maintaining each event in a specific flow queue. 2. Scheduling group. A scheduling group is more a steam of events, so an ev= ent queue might be a better abstraction. 3. An event queue should support the concept of max active atomic flows (ma= ximum number of active flows this queue can track at any given time) and ma= x active ordered sequences (maximum number of outstanding events waiting to= be egress reordered by this queue). This allows a scheduler implementation= to dimension/partition its resources among event queues. 4. An event queue should support concept of a single consumer. In an applic= ation, a stream of events may need to be brought together to a single core = for some stages of processing, e.g. for TX at the end of the pipeline to av= oid NIC reordering of the packets. Having a 'single consumer' event queue f= or that stage allows the intensive scheduling logic to be short circuited a= nd can improve throughput for s/w implementations. 5. Instead of tying eventdev access to an lcore, a higher level of abstract= ion called event port is needed which is the application i/f to the eventde= v. Event ports are connected to event queues and is the object the applicat= ion uses to dequeue and enqueue events. There can be more than one event po= rt per lcore allowing multiple lightweight threads to have their own i/f in= to eventdev, if the implementation supports it. An event port abstraction a= lso encapsulates dequeue depth and enqueue depth for a scheduler implementa= tions which can schedule multiple events at a time and output events that c= an be buffered. 6. An event should support priority. Per event priority is useful for segre= gating high priority (control messages) traffic from low priority within th= e same flow. This needs to be part of the event definition for implementati= ons which support it. 7. Event port to event queue servicing priority. This allows two event port= s to connect to the same event queue with different priorities. For impleme= ntations which support it, this allows a worker core to participate in two = different workflows with different priorities (workflow 1 needing 3.5 cores= , workflow 2 needing 2.5 cores, and so on). 8. Define the workflow as schedule/dequeue/enqueue. An implementation is fr= ee to define schedule as NOOP. A distributed s/w scheduler can use this to = schedule events; also a centralized s/w scheduler can make this a NOOP on n= on-scheduler cores. 9. The schedule_from_group API does not fit the workflow. 10. The ctxt_update/ctxt_wait breaks the normal workflow. If the normal wor= kflow is a dequeue -> do work based on event type -> enqueue, a pin_event = argument to enqueue (where the pinned event is returned through the normal = dequeue) allows application workflow to remain the same whether or not an i= mplementation supports it. 11. Burst dequeue/enqueue needed. 12. Definition of a closed/open system - where open system is memory backed= and closed system eventdev has limited capacity. In such systems, it is al= so useful to denote per event port how many packets can be active in the sy= stem. This can serve as a threshold for ethdev like devices so they don't o= verwhelm core to core events. 13. There should be sort of device capabilities definition to address diffe= rent implementations. vnr ---