From patchwork Fri Dec 20 22:59:17 2013 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Abbas Raza X-Patchwork-Id: 304338 Return-Path: X-Original-To: patchwork-incoming@ozlabs.org Delivered-To: patchwork-incoming@ozlabs.org Received: from vger.kernel.org (vger.kernel.org [209.132.180.67]) by ozlabs.org (Postfix) with ESMTP id 7A2EB2C00AE for ; Sat, 21 Dec 2013 09:59:52 +1100 (EST) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1754123Ab3LTW7u (ORCPT ); Fri, 20 Dec 2013 17:59:50 -0500 Received: from relay1.mentorg.com ([192.94.38.131]:59761 "EHLO relay1.mentorg.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753063Ab3LTW7t (ORCPT ); Fri, 20 Dec 2013 17:59:49 -0500 Received: from svr-orw-fem-01.mgc.mentorg.com ([147.34.98.93]) by relay1.mentorg.com with esmtp id 1Vu92q-0006y7-TE from Abbas_Raza@mentor.com ; Fri, 20 Dec 2013 14:59:48 -0800 Received: from abbas.pkl.mentorg.com ([137.202.156.207]) by svr-orw-fem-01.mgc.mentorg.com with Microsoft SMTPSVC(6.0.3790.4675); Fri, 20 Dec 2013 14:59:48 -0800 From: Abbas Raza To: linux-fsdevel@vger.kernel.org, linux-ext4@vger.kernel.org Cc: Abbas Raza , Theodore Ts'o , Alexander Viro Subject: [PATCH 2/2] ext4: handle mount failures more gracefully Date: Sat, 21 Dec 2013 03:59:17 +0500 Message-Id: <1387580357-28440-2-git-send-email-abbas_raza@mentor.com> X-Mailer: git-send-email 1.8.3.2 In-Reply-To: <1387580357-28440-1-git-send-email-abbas_raza@mentor.com> References: <1387580357-28440-1-git-send-email-abbas_raza@mentor.com> X-OriginalArrivalTime: 20 Dec 2013 22:59:48.0754 (UTC) FILETIME=[2EEF1720:01CEFDD7] Sender: linux-ext4-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: linux-ext4@vger.kernel.org From: Abbas Raza ext4 code throws WARNINGs if a mounted card having ext4 filesystem is ejected without unmounting first and inserted again. This happens because ext4 procfs entries for this card doesn't get cleaned when it is ejected this way. Steps to reproduce this problem: 1) card having ext4 filesystem is already mounted: root@mx6q:~# mount /dev/mmcblk0p1 on /media/sd-mmcblk0p1 type ext4 (rw,relatime,data=ordered) 2) cd root@mx6q:~# cd /media/sd-mmcblk0p1/ root@mx6q:/media/sd-mmcblk0p1# root@mx6q:/media/sd-mmcblk0p1# ls /proc/fs/ext4/ mmcblk0p1 root@mx6q:/media/sd-mmcblk0p1# 3) Remove card: root@mx6q:/media/sd-mmcblk0p1# [ 165.167264] mmc1: card efa2 removed 4) Proc entries for the removed card are still present: root@mx6q:/media/sd-mmcblk0p1# ls /proc/fs/ext4/ mmcblk0p1 5) Re-insert card (mount will fail with a following warnings): root@mx6q:/media/sd-mmcblk0p1# [ 74.987270] mmc1: new SDHC card at address efa2 [ 74.992426] mmcblk0: mmc1:efa2 SD04G 3.69 GiB [ 75.003689] mmcblk0: p1 [ 75.590601] ------------[ cut here ]------------ [ 75.595296] WARNING: at fs/proc/generic.c:573 proc_register+0xe8/0x138() [ 75.602013] proc_dir_entry 'ext4/mmcblk0p1' already registered [ 75.607882] Modules linked in: ip_tables x_tables [ 75.662964] Backtrace: [ 75.665500] [<80011734>] (dump_backtrace+0x0/0x10c) from [<803a2eb8>] (dump_stack+0x18/0x1c) [ 75.674005] r6:80488400 r5:0000023d r4:ac637c60 r3:80508444 [ 75.679779] [<803a2ea0>] (dump_stack+0x0/0x1c) from [<80022908>] (warn_slowpath_common+0x54/0x6c) [ 75.688885] [<800228b4>] (warn_slowpath_common+0x0/0x6c) from [<800229c4>] (warn_slowpath_fmt+0x38/0x40) [ 75.698596] r8:a9968b59 r7:a816b200 r6:00000000 r5:ac1a1c80 r4:a9968b00 r3:00000009 [ 75.706615] [<8002298c>] (warn_slowpath_fmt+0x0/0x40) from [<8013b03c>] (proc_register+0xe8/0x138) [ 75.715783] r3:ac1a1cd9 r2:80488466 [ 75.719437] [<8013af54>] (proc_register+0x0/0x138) from [<8013b264>] (proc_mkdir_mode+0x40/0x60) [ 75.728411] r8:805bac20 r7:ac721800 r6:ac722000 r5:00000000 r4:a9968b00 [ 75.735284] [<8013b224>] (proc_mkdir_mode+0x0/0x60) from [<8013b29c>] (proc_mkdir+0x18/0x1c) [ 75.743825] r4:0000007f [ 75.746399] [<8013b284>] (proc_mkdir+0x0/0x1c) from [<8016f038>] (ext4_fill_super+0xe1c/0x2788) [ 75.755193] [<8016e21c>] (ext4_fill_super+0x0/0x2788) from [<800eae6c>] (mount_bdev+0x124/0x1a0) [ 75.764082] [<800ead48>] (mount_bdev+0x0/0x1a0) from [<8015e740>] (ext4_mount+0x1c/0x28) [ 75.772263] [<8015e724>] (ext4_mount+0x0/0x28) from [<800eb728>] (mount_fs+0x1c/0xc0) [ 75.780160] [<800eb70c>] (mount_fs+0x0/0xc0) from [<80102498>] (vfs_kern_mount+0x54/0xc8) [ 75.788409] r6:00000000 r5:ac3886c0 r4:acbd4b40 [ 75.793102] [<80102444>] (vfs_kern_mount+0x0/0xc8) from [<801045e8>] (do_mount+0x71c/0x834) [ 75.801557] r8:ac636000 r7:8051f3dc r6:805322c0 r5:ac3886c0 r4:00000020 r3:00000000 [ 75.809528] [<80103ecc>] (do_mount+0x0/0x834) from [<8010478c>] (sys_mount+0x8c/0xc0) [ 75.817446] [<80104700>] (sys_mount+0x0/0xc0) from [<8000d860>] (ret_fast_syscall+0x0/0x30) [ 75.825883] r7:00000015 r6:005956c0 r5:00200000 r4:00000000 [ 75.831606] ---[ end trace 041836e4555827df ]--- [ 75.839421] ------------[ cut here ]------------ [ 75.844096] WARNING: at fs/proc/generic.c:573 proc_register+0xe8/0x138() [ 75.850810] proc_dir_entry 'jbd2/mmcblk0p1-8' already registered ... [ 76.118095] ------------[ cut here ]------------ [ 76.122776] WARNING: at fs/sysfs/dir.c:536 sysfs_add_one+0x9c/0xbc() [ 76.129140] sysfs: cannot create duplicate filename '/fs/ext4/mmcblk0p1' ... [ 76.410445] ------------[ cut here ]------------ [ 76.415124] WARNING: at lib/kobject.c:196 kobject_add_internal+0x15c/0x1cc() [ 76.422214] kobject_add_internal failed for mmcblk0p1 with -EEXIST, don't try to register things with the same name in the same directory. ... [ 76.661850] ---[ end trace 041836e4555827e2 ]--- [ 76.666572] EXT4-fs (mmcblk0p1): mount failed With this patch: 5) Re-insert card (mount will fail with following message): root@mx6q:/media/sd-mmcblk0p1# root@mx6q:/media/sd-mmcblk0p1# [ 171.392220] mmc1: new SDHC card at address efa2 [ 171.398608] mmcblk0: mmc1:efa2 SD04G 3.69 GiB [ 171.410501] mmcblk0: p1 [ 171.990337] EXT4-fs (mmcblk0p1): could'nt mount because mount point is busy and /proc/ext4/mmcblk0p1 is already present 6) Free mount point and proc entries for the card will vanish root@mx6q:/media/sd-mmcblk0p1# cd .. root@mx6q:/media# ls /proc/fs/ext4/ 7) Re-insert card and it will mount successfully: root@mx6q:/media# [ 194.317263] mmc1: card efa2 removed root@mx6q:/media# [ 199.512198] mmc1: new SDHC card at address efa2 [ 199.518254] mmcblk0: mmc1:efa2 SD04G 3.69 GiB [ 199.527939] mmcblk0: p1 [ 200.101131] EXT4-fs (mmcblk0p1): recovery complete [ 200.105985] EXT4-fs (mmcblk0p1): mounted filesystem with ordered data mode. Opts: (null) root@mx6q:/media# ls /proc/fs/ext4/ mmcblk0p1 Cc: Theodore Ts'o Cc: Alexander Viro Signed-off-by: Abbas Raza --- fs/ext4/super.c | 11 ++++++++++- 1 file changed, 10 insertions(+), 1 deletion(-) diff --git a/fs/ext4/super.c b/fs/ext4/super.c index b83f75d..edb5613 100644 --- a/fs/ext4/super.c +++ b/fs/ext4/super.c @@ -3708,8 +3708,17 @@ static int ext4_fill_super(struct super_block *sb, void *data, int silent) goto failed_mount; } - if (ext4_proc_root) + if (ext4_proc_root) { + if (proc_lookup_subdir_by_name(ext4_proc_root, + sb->s_id) == 0) { + ext4_msg(sb, KERN_ERR, "could'nt mount because" + " mount point is busy and /proc/%s/%s is" + " already present", + ext4_proc_root->name, sb->s_id); + goto failed_mount; + } sbi->s_proc = proc_mkdir(sb->s_id, ext4_proc_root); + } if (sbi->s_proc) proc_create_data("options", S_IRUGO, sbi->s_proc,