From mboxrd@z Thu Jan 1 00:00:00 1970 From: Brian Haley Subject: Re: [PATCH 2/2] ipvs: load balance ipv6 connections from a local process Date: Fri, 05 Sep 2008 11:55:44 -0400 Message-ID: <48C15680.7010004@hp.com> References: <20080905013609.GD14128@verge.net.au> <20080905013755.GE14128@verge.net.au> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Cc: Simon Horman , netdev@vger.kernel.org, lvs-devel@vger.kernel.org, Malcolm Turnbull , =?ISO-8859-1?Q?Siim_P=F5der?= , Vince Busam To: Julius Volz Return-path: Received: from g1t0027.austin.hp.com ([15.216.28.34]:9636 "EHLO g1t0027.austin.hp.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752336AbYIEPzu (ORCPT ); Fri, 5 Sep 2008 11:55:50 -0400 In-Reply-To: Sender: netdev-owner@vger.kernel.org List-ID: Julius Volz wrote: >> -out: >> - __ip_vs_conn_put(cp); >> - >> - return verdict; >> + snet.in6 = iph->saddr; > > I've always been told to use ipv6_addr_copy() for this. I'm not sure > what the problem with the direct struct assignment is though... would > be nice if someone could explain. Because an in6_addr is a union of 4 u32's, which won't all be copied in a struct assignment. That's the way I've always understood it. -Brian