From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S263016AbTDFPku (for ); Sun, 6 Apr 2003 11:40:50 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S263020AbTDFPku (for ); Sun, 6 Apr 2003 11:40:50 -0400 Received: from pc2-cwma1-4-cust86.swan.cable.ntl.com ([213.105.254.86]:27790 "EHLO lxorguk.ukuu.org.uk") by vger.kernel.org with ESMTP id S263016AbTDFPkt (for ); Sun, 6 Apr 2003 11:40:49 -0400 Subject: Re: [PATCH] e1000 close From: Alan Cox To: "J.A. Magallon" Cc: Marcelo Tosatti , lkml In-Reply-To: <20030405231220.GI12864@werewolf.able.es> References: <20030405224233.GA12746@werewolf.able.es> <20030405231220.GI12864@werewolf.able.es> Content-Type: text/plain Content-Transfer-Encoding: 7bit Organization: Message-Id: <1049640831.963.14.camel@dhcp22.swansea.linux.org.uk> Mime-Version: 1.0 X-Mailer: Ximian Evolution 1.2.2 (1.2.2-5) Date: 06 Apr 2003 15:53:52 +0100 Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org On Sul, 2003-04-06 at 00:12, J.A. Magallon wrote: > Supposed to cure a dev_close called without dev_open. > Is this still needed ? If ->close is called on a device that is not open then the fix is not the driver, the fix is the core code fix.