From mboxrd@z Thu Jan 1 00:00:00 1970 From: David Miller Subject: Re: the future of ethtool Date: Mon, 15 Nov 2010 18:02:25 -0800 (PST) Message-ID: <20101115.180225.189678628.davem@davemloft.net> References: <1289857936.2586.51.camel@bwh-desktop> <4CE1B8FD.3000007@garzik.org> <20101115233335.GB24292@canuck.infradead.org> Mime-Version: 1.0 Content-Type: Text/Plain; charset=us-ascii Content-Transfer-Encoding: 7bit Cc: jeff@garzik.org, bhutchings@solarflare.com, shemminger@vyatta.com, netdev@vger.kernel.org To: tgraf@infradead.org Return-path: Received: from 74-93-104-97-Washington.hfc.comcastbusiness.net ([74.93.104.97]:33525 "EHLO sunset.davemloft.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1754847Ab0KPCCA (ORCPT ); Mon, 15 Nov 2010 21:02:00 -0500 In-Reply-To: <20101115233335.GB24292@canuck.infradead.org> Sender: netdev-owner@vger.kernel.org List-ID: From: Thomas Graf Date: Mon, 15 Nov 2010 18:33:35 -0500 > I tried to solve this by splitting the validate/change operation and > thus be able to validate all requests before starting to commit > them. This would mean changing all drivers though which I wasn't > willing to do. It isn't sufficient. You can still get into unwindable failures. Earlier operations can consume fixed resources like TCAM filter slots or rx/tx queues, making a subsequent operation in the sequence fail. A validate/commit scheme cannot detect this effectively.