From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S264166AbTDOX5D (for ); Tue, 15 Apr 2003 19:57:03 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S264167AbTDOX5D (for ); Tue, 15 Apr 2003 19:57:03 -0400 Received: from air-2.osdl.org ([65.172.181.6]:15753 "EHLO mail.osdl.org") by vger.kernel.org with ESMTP id S264166AbTDOX5C (for ); Tue, 15 Apr 2003 19:57:02 -0400 Date: Tue, 15 Apr 2003 17:07:48 -0700 From: "Randy.Dunlap" To: walt Cc: linux-kernel@vger.kernel.org Subject: Re: [PATCH] Re: 2.5.67: ppa driver & preempt == oops Message-Id: <20030415170748.1c9f0e1c.rddunlap@osdl.org> In-Reply-To: <3E9C9B19.7000107@myrealbox.com> References: <3E9C9B19.7000107@myrealbox.com> Organization: OSDL X-Mailer: Sylpheed version 0.8.11 (GTK+ 1.2.10; i586-pc-linux-gnu) Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org On Tue, 15 Apr 2003 16:51:53 -0700 walt wrote: | walt wrote: | > Patrick Mansfield wrote: | > | >> On Sun, Apr 13, 2003 at 07:44:04PM +0200, Gert Vervoort wrote: | >> | >> Here is a patch against 2.5.67, can you try it out? | >> | >> I did not compile let alone run with this patch. | >> | >> We never hold the host_lock while calling the detect function (unlike the | >> io_request_lock, see the bizzare 2.4 code), so acquiring it inside | >> ppa_detect is very wrong. I don't know why your scsi scan did not hang. | >> ... | > | > | > | > Yes! Thank you. This patch fixes the segfault of modprobe that I've | > been seeing for ages. | | Hmm. An important addendum -- | | I can mount and unmount the Zip disk as many times as I want, as long as | I don't eject the disk from the drive. | | Once I eject the disk and insert it again, the mount commant just hangs | forever. No error messages anywhere -- it just hangs. I can't even | kill the mount command. The only way out is to reboot and then all is | well again until the next time I eject the disk from the drive. I don't know that it's _required_, but you can try doing umount /mnt/zip (or /dev/sdx) eject /mnt/zip (or dev/sdx) and then eject it, reinsert it, try to mount, etc.? -- ~Randy