From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id ; Wed, 11 Dec 2002 17:54:41 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id ; Wed, 11 Dec 2002 17:54:40 -0500 Received: from nat-pool-rdu.redhat.com ([66.187.233.200]:57885 "EHLO flossy.devel.redhat.com") by vger.kernel.org with ESMTP id ; Wed, 11 Dec 2002 17:54:38 -0500 Date: Wed, 11 Dec 2002 18:03:17 -0500 From: Doug Ledford To: Oliver Jehle Cc: linux-kernel@vger.kernel.org Subject: Re: Problems using /proc/scsi/gdth/ with 2.4.20aa1 Message-ID: <20021211230317.GD6513@redhat.com> Mail-Followup-To: Oliver Jehle , linux-kernel@vger.kernel.org References: <1039502939.1054.12.camel@vorab.monex.li> Mime-Version: 1.0 Content-Type: multipart/mixed; boundary="9amGYk9869ThD9tj" Content-Disposition: inline In-Reply-To: <1039502939.1054.12.camel@vorab.monex.li> User-Agent: Mutt/1.4i Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org --9amGYk9869ThD9tj Content-Type: text/plain; charset=us-ascii Content-Disposition: inline On Tue, Dec 10, 2002 at 07:49:00AM +0100, Oliver Jehle wrote: > Running linux 2.4.20-aa1 and the gdth driver works ,but accessing > /proc/scsi/gdth/0 for example with cat or the supplied icpcon utility > don't work... > > these messages are in system log when accessing /proc/scsi/gdth/0 with > cat for example (works with 2.4.18) > ... > > Dec 10 06:37:47 arena1 kernel: Unable to handle kernel NULL pointer > dereference at virtual address 00000000 > Dec 10 06:37:47 arena1 kernel: printing eip: > Dec 10 06:37:47 arena1 kernel: c024c879 > Dec 10 06:37:47 arena1 kernel: *pde = 00000000 > Dec 10 06:37:47 arena1 kernel: Oops: 0002 2.4.20aa1 #2 SMP Mon Dec 9 > 16:55:18 CET 2002 > Dec 10 06:37:47 arena1 kernel: CPU: 0 > Dec 10 06:37:47 arena1 kernel: EIP: > 0010:[scsi_release_commandblocks+17/92] Not tainted /me chuckles That looks *amazingly* familiar...a patch I wrote introduced that bug, sorry. It's fix is attached (assuming that Andrea picked up the same version of the 2.4.x iorl contention patch that we are using now, which is what introduced this problem for us here). -- Doug Ledford 919-754-3700 x44233 Red Hat, Inc. 1801 Varsity Dr. Raleigh, NC 27606 --9amGYk9869ThD9tj Content-Type: text/plain; charset=us-ascii Content-Disposition: attachment; filename="linux-2.4.9-gdthoops.patch" --- linux/drivers/scsi/scsi.c-stock Fri Oct 25 15:41:03 2002 +++ linux/drivers/scsi/scsi.c Fri Oct 25 15:47:04 2002 @@ -2672,10 +2672,10 @@ SDpnt->type = -1; SDpnt->queue_depth = 1; - scsi_build_commandblocks(SDpnt); - scsi_initialize_queue(SDpnt, SHpnt); + scsi_build_commandblocks(SDpnt); + SDpnt->online = TRUE; /* @@ -2705,13 +2705,12 @@ panic("Attempt to delete wrong device\n"); } - blk_cleanup_queue(&SDpnt->request_queue); - /* * We only have a single SCpnt attached to this device. Free * it now. */ scsi_release_commandblocks(SDpnt); + blk_cleanup_queue(&SDpnt->request_queue); kfree(SDpnt); } --9amGYk9869ThD9tj--