From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mga01.intel.com (mga01.intel.com [192.55.52.88]) by dpdk.org (Postfix) with ESMTP id 0F8C96C9B for ; Tue, 21 Jun 2016 12:29:34 +0200 (CEST) Received: from orsmga001.jf.intel.com ([10.7.209.18]) by fmsmga101.fm.intel.com with ESMTP; 21 Jun 2016 03:29:34 -0700 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="5.26,503,1459839600"; d="scan'208";a="980290236" Received: from bricha3-mobl3.ger.corp.intel.com ([10.237.220.108]) by orsmga001.jf.intel.com with SMTP; 21 Jun 2016 03:29:32 -0700 Received: by (sSMTP sendmail emulation); Tue, 21 Jun 2016 11:29:30 +0025 Date: Tue, 21 Jun 2016 11:29:30 +0100 From: Bruce Richardson To: Beilei Xing Cc: jingjing.wu@intel.com, dev@dpdk.org, thomas.monjalon@6wind.com, nhorman@tuxdriver.com, pmatilai@redhat.com Message-ID: <20160621102930.GA21016@bricha3-MOBL3> References: <1464247695-4694-1-git-send-email-beilei.xing@intel.com> <1465805012-2907-1-git-send-email-beilei.xing@intel.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <1465805012-2907-1-git-send-email-beilei.xing@intel.com> Organization: Intel Research and =?iso-8859-1?Q?De=ACvel?= =?iso-8859-1?Q?opment?= Ireland Ltd. User-Agent: Mutt/1.5.23 (2014-03-12) Subject: Re: [dpdk-dev] [PATCH v2] i40e: modify the meaning of single VLAN type 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, 21 Jun 2016 10:29:35 -0000 On Mon, Jun 13, 2016 at 04:03:32PM +0800, Beilei Xing wrote: > In current i40e codebase, if single VLAN header is added in a packet, > it's treated as inner VLAN. Generally, a single VLAN header is > treated as the outer VLAN header. So change corresponding register > for single VLAN. > At the meanwhile, change the meanings of inner VLAN and outer VLAN. > > Signed-off-by: Beilei Xing This patch changes the ABI, since an app written to the original API as specified e.g. to set a single vlan header, would no longer work with this change. Therefore, even though the original behaviour was inconsistent with other drivers it may still need to be preserved. I'm thinking that we may need to provide appropriately versioned copies of the vlan_offload_set and vlan_tpid_set functions for backward compatibility with the old ABI. Any other comments or thoughts on this? Neil, Thomas, Panu - is this fix something that we need to provide backward version-compatibility for, or given that the functions are being called through a generic ethdev API mean that this can just go in as a straight bug-fix? /Bruce