From mboxrd@z Thu Jan 1 00:00:00 1970 From: Thomas Graf Subject: Re: [PATCH 3/3] IPROUTE: generate compat messages that match nla_parse_nested_compat Date: Thu, 28 Aug 2008 11:51:31 +0200 Message-ID: <20080828095131.GX20815@postel.suug.ch> References: <20080828011340.7987.74684.stgit@jtkirshe-mobile.jf.intel.com> <20080828011347.7987.24759.stgit@jtkirshe-mobile.jf.intel.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: jeff@garzik.org, netdev@vger.kernel.org, davem@davemloft.net, Alexander Duyck To: Jeff Kirsher Return-path: Received: from postel.suug.ch ([194.88.212.233]:56804 "EHLO postel.suug.ch" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751675AbYH1JvL (ORCPT ); Thu, 28 Aug 2008 05:51:11 -0400 Content-Disposition: inline In-Reply-To: <20080828011347.7987.24759.stgit@jtkirshe-mobile.jf.intel.com> Sender: netdev-owner@vger.kernel.org List-ID: * Jeff Kirsher 2008-08-27 18:13 > From: Alexander Duyck > > This patch updates libnetlink.c to generate nested compat netlink > attributes in the same format that is parsed by the kernels 2.6.26 > and newer. > > This patch will fix the prio multiqueue option for 2.6.26 kernels, > but will break compatibility with all kernels prior to 2.6.26. I'm not sure what you mean by fix, I can't find any nested attributes being parsed in the proposed sch_multiq.c. Wouldn't it be simpler to have q_multiq.c use addattr_nest() instead of addattr_nest_compat() instead of changing addattr_nest_compat() to be equivalent to addattr_nest()?