From mboxrd@z Thu Jan 1 00:00:00 1970 From: YOSHIFUJI Hideaki Subject: Re: [PATCH] net IPv6: Remove unnecessary argument of ndisc_send_ns and ndisc_send_na Date: Mon, 14 Apr 2014 14:13:18 +0900 Message-ID: <534B6E6E.5020104@linux-ipv6.org> References: <534B58AA.80704@gmail.com> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit To: Balakumaran Kannan , davem@davemloft.net, kuznet@ms2.inr.ac.ru, jmorris@namei.org, kaber@trash.net, netdev@vger.kernel.org, YOSHIFUJI Hideaki Return-path: Received: from 94.43.138.210.xn.2iij.net ([210.138.43.94]:52751 "EHLO mail.st-paulia.net" rhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with ESMTP id S1750905AbaDNFNU (ORCPT ); Mon, 14 Apr 2014 01:13:20 -0400 In-Reply-To: <534B58AA.80704@gmail.com> Sender: netdev-owner@vger.kernel.org List-ID: Balakumaran Kannan wrote: > The 'neigh' argument of 'ndisc_send_ns' and 'ndisc_send_na' are not used in > these functions. So this could be removed. > > And a minor change of 'optlen' assignment in those functions. the compound > addition and assignment operator is not needed as 'optlen' value is always > zero. Do not mix changes. Well, it is for future extension so that we can prepend another option easier. Do you see any penalty here? > > Signed-off-by: Balakumaran Kannan > --- > I am not sure whether 'neigh' argument is intentional or planned for any > future use. If you find this is unnecessary, this patch could be considered. > > And ndisc_send_na is used as a function pointer in ipv6_stub. This may affect anybody > using this structure. But with default configuration built, I didn't get any error or > warning. Kindly consider this scenario also. drivers/net/usb/cdc_mbim.c uses it at least. --yoshfuji