From mboxrd@z Thu Jan 1 00:00:00 1970 From: Ding Tianhong Subject: Re: [PATCH net-next] bonding: move the netdev_add_tso_features() to bonding module Date: Sun, 19 Jan 2014 13:34:21 +0800 Message-ID: <52DB63DD.50203@huawei.com> References: <52DA3BE5.5020500@huawei.com> <20140118114801.GA30549@redhat.com> <1390064923.31367.531.camel@edumazet-glaptop2.roam.corp.google.com> Mime-Version: 1.0 Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: 7bit Cc: Jay Vosburgh , Eric Dumazet , "David S. Miller" , Netdev To: Eric Dumazet , Veaceslav Falico Return-path: Received: from szxga03-in.huawei.com ([119.145.14.66]:25204 "EHLO szxga03-in.huawei.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1750739AbaASFe7 (ORCPT ); Sun, 19 Jan 2014 00:34:59 -0500 In-Reply-To: <1390064923.31367.531.camel@edumazet-glaptop2.roam.corp.google.com> Sender: netdev-owner@vger.kernel.org List-ID: On 2014/1/19 1:08, Eric Dumazet wrote: > On Sat, 2014-01-18 at 12:48 +0100, Veaceslav Falico wrote: >> On Sat, Jan 18, 2014 at 04:31:33PM +0800, Ding Tianhong wrote: >>> The function netdev_add_tso_features() was only be used for bonding, >>> so no need to export it in netdevice.h, move it to bonding module. >> >> Eric added it for a reason - like, other drivers might use it. Do you know >> if team, bridge, vlan etc. might use it? > > A helper can be used once, this is fine. A car can have 4 seats, and can > even be used with no passenger. > > I am quite bored by patches that break clean layering for wrong reasons. > > > static inline netdev_features_t netdev_add_tso_features(netdev_features_t features, > netdev_features_t mask) > { > return netdev_increment_features(features, NETIF_F_ALL_TSO, mask); > } > > There is _nothing_ in this helper that implies it should be private to bonding. > > yep, it is look so fool that my patch said, thanks for reminding me. > > >