From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from DUB004-OMC2S35.hotmail.com (dub004-omc2s35.hotmail.com [157.55.1.174]) by dpdk.org (Postfix) with ESMTP id 13CCDB38C for ; Fri, 11 Jul 2014 15:15:35 +0200 (CEST) Received: from DUB131-W88 ([157.55.1.136]) by DUB004-OMC2S35.hotmail.com with Microsoft SMTPSVC(7.5.7601.22712); Fri, 11 Jul 2014 06:16:03 -0700 X-TMN: [5L39Zg/TzkXNCcUb+k3otgqlo2JPtD3E] X-Originating-Email: [hobywank@hotmail.com] Message-ID: From: Hobywan Kenoby To: "Sujith Sankar (ssujith)" , "dev@dpdk.org" Date: Fri, 11 Jul 2014 15:16:03 +0200 Importance: Normal In-Reply-To: References: <1404916536-1364-1-git-send-email-pablo.de.lara.guarch@intel.com>, <20140709152039.GB5250@localhost.localdomain>, , MIME-Version: 1.0 X-OriginalArrivalTime: 11 Jul 2014 13:16:03.0523 (UTC) FILETIME=[44203930:01CF9D0A] Content-Type: text/plain; charset="Windows-1252" Content-Transfer-Encoding: quoted-printable X-Content-Filtered-By: Mailman/MimeDel 2.1.15 Subject: Re: [dpdk-dev] PMD for Cisco VIC Ethernet NIC - Request for guidelines for submission 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: Fri, 11 Jul 2014 13:15:35 -0000 Hi Sujith=2C It makes sens=2C using VFIO makes a far cleaner implementation.=20 You worked on the performance=2C could you share some measurements ? Did yo= u introduce vectorization functions as Intel did a while ago? There are all= ways tradeoffs between pps and latency=2C do you include documentation to c= onfigure the card for one or the other? HK From: ssujith@cisco.com To: hobywank@hotmail.com=3B dev@dpdk.org Subject: Re: [dpdk-dev] PMD for Cisco VIC Ethernet NIC - Request for guidel= ines for submission Date: Fri=2C 11 Jul 2014 07:46:08 +0000 =0A= =0A= =0A= =0A= =0A= Hi Hobywan=2C=0A= Thanks for the email !=0A= =0A= =0A= We=92ve been working on performance benchmarking. Also=2C we felt that it = would be better to push the driver after Anatoly=92s patch (vfio-pci) got i= n so that we could make the necessary modification before submission. =0A= Now that 1.7.0 is out=2C we are hopeful of submitting the patch soon.=0A= =0A= =0A= Thanks=2C=0A= -Sujith=0A= =0A= =0A= =0A= =0A= From: Hobywan Kenoby =0A= Date: Thursday=2C 10 July 2014 1:37 am =0A= To: "Sujith Sankar (ssujith)" =2C "dev@dpdk.org" =0A= Subject: RE: [dpdk-dev] PMD for Cisco VIC Ethernet NIC - Request for guidel= ines for submission =0A= =0A= =0A= =0A= =0A= =0A= Hi Sujith=2C=0A= =0A= =0A= It was exciting to see open source code coming from Cisco ensuring a DPDK a= pplication can run on any platform and with any card....=0A= I haven't seen your patch yet. What happened?=0A= =0A= =0A= HK =0A= =0A= =0A= > Date : Wed=2C 28 May 2014 08:06=0A= > From : dev [mailto:dev-bounces@dpdk.org] =0A= > To :=0A= dev@dpdk.org=0A= > Subject : [dpdk-dev] PMD for Cisco VIC Ethernet NIC - Request for guideli= nes=0A= > for submission=0A= > =0A= > =0A= > Hi all=2C=0A= > =0A= > We have been working on development of poll-mode driver for Cisco VIC=0A= > Ethernet NIC and integration of it with DPDK. We would like to submit th= is=0A= > poll-mode driver (ENIC PMD) to the DPDK community so that it could be par= t=0A= > of the DPDK tree.=0A= > =0A= > Could someone please provide the guidelines and steps to do this? As of= =0A= > now=2C ENIC PMD is being tested with DPDK 1.6.0r2. Is it alright to subm= it a=0A= > patch for DPDK 1.6.0r2?=0A= > =0A= > One aspect of ENIC PMD is that it works with VFIO-PCI and not UIO. Hope= =0A= > this is acceptable. The following thread in dpdk-dev influenced this=0A= > decision.=0A= > =0A= http://dpdk.org/ml/archives/dev/2013-July/000373.html=0A= > =0A= > ENIC PMD uses one interrupt per interface and it is used by the NIC for= =0A= > signalling the driver in case of any error. Since this does not come in= =0A= > the fast path=2C it should be acceptable=2C isn=B9t it?=0A= > =0A= > Please give your suggestions and comments.=0A= > =0A= > Thanks=2C=0A= > -Sujith=0A= =0A= =0A= =0A= =0A= =0A= =0A= =