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 DA7BB958D for ; Wed, 6 Jan 2016 08:35:03 +0100 (CET) Received: from fmsmga004.fm.intel.com ([10.253.24.48]) by fmsmga102.fm.intel.com with ESMTP; 05 Jan 2016 23:35:03 -0800 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="5.20,528,1444719600"; d="scan'208";a="24387640" Received: from yliu-dev.sh.intel.com (HELO yliu-dev) ([10.239.66.49]) by fmsmga004.fm.intel.com with ESMTP; 05 Jan 2016 23:35:01 -0800 Date: Wed, 6 Jan 2016 15:38:52 +0800 From: Yuanhan Liu To: Tetsuya Mukawa Message-ID: <20160106073852.GH26062@yliu-dev.sh.intel.com> References: <20151218041536.GI29571@yliu-dev.sh.intel.com> <56738B5C.1030206@igel.co.jp> <20151222034158.GH18863@yliu-dev.sh.intel.com> <567B61D6.1090806@igel.co.jp> <567BA5B9.5090006@igel.co.jp> <568C908A.2070008@igel.co.jp> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <568C908A.2070008@igel.co.jp> User-Agent: Mutt/1.5.23 (2014-03-12) Cc: dev@dpdk.org, ann.zhuangyanying@huawei.com Subject: Re: [dpdk-dev] [PATCH v5 1/3] vhost: Add callback and private data for vhost PMD 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: Wed, 06 Jan 2016 07:35:04 -0000 On Wed, Jan 06, 2016 at 12:56:58PM +0900, Tetsuya Mukawa wrote: > On 2015/12/29 6:59, Rich Lane wrote: > > On Wed, Dec 23, 2015 at 11:58 PM, Tetsuya Mukawa wrote: > >> Hi Rich and Yuanhan, > >> > >> I guess we have 2 implementations here. > >> > >> 1. rte_eth_vhost_get_queue_event() returns each event. > >> 2. rte_eth_vhost_get_queue_status() returns current status of the queues. > >> > >> I guess option "2" is more generic manner to handle interrupts from > >> device driver. > >> In the case of option "1", if DPDK application doesn't call > >> rte_eth_vhost_get_queue_event(), the vhost PMD needs to keep all events. > >> This may exhaust memory. > >> > > Option 1 can be implemented in constant space by only tracking the latest > > state of each > > queue. I pushed a rough implementation to https://github.com/rlane/dpdk > > vhost-queue-callback. > > > > One more example is current link status interrupt handling. > > Hi Rich, > > I appreciate your implementation. > I can understand what's your idea, and agree with it. > > > Hi Yuanhan, > > What do you think his implementation? With a quick glimpse, it looks good to me. --yliu