From mboxrd@z Thu Jan 1 00:00:00 1970 From: Yuanhan Liu Subject: Re: [PATCH] ethdev: fix wrong memset Date: Fri, 20 Jan 2017 18:34:28 +0800 Message-ID: <20170120103428.GY10293@yliu-dev.sh.intel.com> References: <1484899493-11051-1-git-send-email-yuanhan.liu@linux.intel.com> <2005966.sVKUI0LeEs@xps13> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: dev@dpdk.org, Ferruh Yigit To: Thomas Monjalon Return-path: Received: from mga06.intel.com (mga06.intel.com [134.134.136.31]) by dpdk.org (Postfix) with ESMTP id 77E922BA7 for ; Fri, 20 Jan 2017 11:32:16 +0100 (CET) Content-Disposition: inline In-Reply-To: <2005966.sVKUI0LeEs@xps13> List-Id: DPDK patches and discussions List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: dev-bounces@dpdk.org Sender: "dev" On Fri, Jan 20, 2017 at 11:20:06AM +0100, Thomas Monjalon wrote: > 2017-01-20 16:04, Yuanhan Liu: > > Fix an silly error by auto-complete while managing the merge conflicts. > > It's the eth_dev_data (but not eth_dev) entry should be memset. > > > > Fixes: d948f596fee2 ("ethdev: fix port data mismatched in multiple process model") > > You should describe the impact on this bug. Honestly, I don't know. I didn't met any issue while testing vhost. Maybe Ferruh knows what might be wrong, since it's him spotted this bug? > It will be helpful for those testing the RC1. > > > - memset(&rte_eth_devices[port_id], 0, sizeof(*eth_dev->data)); > > + memset(&rte_eth_dev_data[port_id], 0, sizeof(struct rte_eth_dev_data)); > > The title should be contain the scope of the bug. > I suggest "fix data reset when allocating port". Yeah, that's better. Thanks. --yliu