From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mga04.intel.com (mga04.intel.com [192.55.52.120]) by dpdk.org (Postfix) with ESMTP id 6B20B2A5E for ; Wed, 28 Feb 2018 11:15:21 +0100 (CET) X-Amp-Result: SKIPPED(no attachment in message) X-Amp-File-Uploaded: False Received: from orsmga004.jf.intel.com ([10.7.209.38]) by fmsmga104.fm.intel.com with ESMTP/TLS/DHE-RSA-AES256-GCM-SHA384; 28 Feb 2018 02:15:20 -0800 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="5.47,405,1515484800"; d="scan'208";a="178698067" Received: from aburakov-mobl.ger.corp.intel.com (HELO [10.237.220.78]) ([10.237.220.78]) by orsmga004.jf.intel.com with ESMTP; 28 Feb 2018 02:15:18 -0800 To: "Tan, Jianfeng" , "dev@dpdk.org" References: <31f6d9ef676fb1eb0a664c06d62d66f32876dcb6.1519672713.git.anatoly.burakov@intel.com> From: "Burakov, Anatoly" Message-ID: Date: Wed, 28 Feb 2018 10:15:18 +0000 User-Agent: Mozilla/5.0 (Windows NT 10.0; WOW64; rv:52.0) Gecko/20100101 Thunderbird/52.6.0 MIME-Version: 1.0 In-Reply-To: Content-Type: text/plain; charset=utf-8; format=flowed Content-Language: en-US Content-Transfer-Encoding: 7bit Subject: Re: [dpdk-dev] [PATCH v3 3/5] eal: use locks to determine if secondary process is active 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: Wed, 28 Feb 2018 10:15:21 -0000 On 28-Feb-18 1:26 AM, Tan, Jianfeng wrote: > > >> -----Original Message----- >> From: Burakov, Anatoly >> Sent: Tuesday, February 27, 2018 10:36 PM >> To: dev@dpdk.org >> Cc: Tan, Jianfeng >> Subject: [PATCH v3 3/5] eal: use locks to determine if secondary process is >> active >> >> Previously, IPC would remove sockets it considers to be "inactive" >> based on whether they have responded. > > To be more precise, it was not depending on if the other side responses or not; it was depending on sendmsg return error, ECONNREFUSED. > >> Change this to create lock >> files in addition to socket files, so that we can determine if >> secondary process is active before attempting to communicate with >> it. That way, we can distinguish secondaries that are alive but >> are not responding, from those that have already died. > > I think, by the old way, we can also "distinguish secondaries that are alive but are not responding, from those that have already died", can't we? > > Thanks, > Jianfeng > I rechecked, and you're right. For some reason i thought that nb_sent gets incremented even if there was ECONNREFUSED error. It doesn't, so the effect is the same. I'll drop this patch so (well, i'll keep the naming stuff, as it makes things a bit easier). -- Thanks, Anatoly