From patchwork Thu Nov 10 03:18:48 2016 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Qu Wenruo X-Patchwork-Id: 9420525 Return-Path: Received: from mail.wl.linuxfoundation.org (pdx-wl-mail.web.codeaurora.org [172.30.200.125]) by pdx-korg-patchwork.web.codeaurora.org (Postfix) with ESMTP id 06FAA6048E for ; Thu, 10 Nov 2016 03:19:19 +0000 (UTC) Received: from mail.wl.linuxfoundation.org (localhost [127.0.0.1]) by mail.wl.linuxfoundation.org (Postfix) with ESMTP id EB098292E2 for ; Thu, 10 Nov 2016 03:19:18 +0000 (UTC) Received: by mail.wl.linuxfoundation.org (Postfix, from userid 486) id DB96D2948C; Thu, 10 Nov 2016 03:19:18 +0000 (UTC) X-Spam-Checker-Version: SpamAssassin 3.3.1 (2010-03-16) on pdx-wl-mail.web.codeaurora.org X-Spam-Level: X-Spam-Status: No, score=-6.9 required=2.0 tests=BAYES_00,RCVD_IN_DNSWL_HI autolearn=ham version=3.3.1 Received: from vger.kernel.org (vger.kernel.org [209.132.180.67]) by mail.wl.linuxfoundation.org (Postfix) with ESMTP id C88FA292E2 for ; Thu, 10 Nov 2016 03:19:17 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1752831AbcKJDTR (ORCPT ); Wed, 9 Nov 2016 22:19:17 -0500 Received: from cn.fujitsu.com ([222.73.24.84]:23086 "EHLO song.cn.fujitsu.com" rhost-flags-OK-FAIL-OK-OK) by vger.kernel.org with ESMTP id S1751564AbcKJDTQ (ORCPT ); Wed, 9 Nov 2016 22:19:16 -0500 X-IronPort-AV: E=Sophos;i="5.20,367,1444665600"; d="scan'208";a="953630" Received: from unknown (HELO cn.fujitsu.com) ([10.167.250.3]) by song.cn.fujitsu.com with ESMTP; 10 Nov 2016 11:19:05 +0800 Received: from localhost.localdomain (unknown [10.167.226.34]) by cn.fujitsu.com (Postfix) with ESMTP id CC8FC41B4BC4; Thu, 10 Nov 2016 11:19:02 +0800 (CST) From: Qu Wenruo To: linux-btrfs@vger.kernel.org, fstests@vger.kernel.org Subject: [PATCH v3] fstests: btrfs: Check false ENOSPC bug caused by incorrect metadata reserve Date: Thu, 10 Nov 2016 11:18:48 +0800 Message-Id: <20161110031848.21127-1-quwenruo@cn.fujitsu.com> X-Mailer: git-send-email 2.10.2 MIME-Version: 1.0 X-yoursite-MailScanner-ID: CC8FC41B4BC4.A1A33 X-yoursite-MailScanner: Found to be clean X-yoursite-MailScanner-From: quwenruo@cn.fujitsu.com Sender: fstests-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: fstests@vger.kernel.org X-Virus-Scanned: ClamAV using ClamSMTP Due to the fact that btrfs uses different max extent size for compressed and non-compressed write, it calculates metadata reservation incorrectly. This can leads to false ENOSPC alert for compressed write. This test case will check it by using small fs and do parallel write to trigger it. It's highly recommened to use MOUNT_OPTIONS="-o compress" and MKFS_OPTIONS="-n 64k" to trigger it. Without compression, it won't be triggered. The fix is not merged and may change in the future, but the function is good enough: btrfs: improve inode's outstanding_extents computation btrfs: fix false enospc for compression Signed-off-by: Qu Wenruo --- v2: Use $XFS_IO_PROGS to replace hard coded xfs_io. Fix "kill 0" which will also kill the script itself. Add missing "Silence is golden" output. Add workaround for EBUSY while umount for compression v3: Remove mount and mkfs option override to increase coverity Increase test time to 30 * TIME_FACTOR to increase the possibility for other MKFS_OPTIONS --- tests/btrfs/132 | 105 ++++++++++++++++++++++++++++++++++++++++++++++++++++ tests/btrfs/132.out | 2 + tests/btrfs/group | 1 + 3 files changed, 108 insertions(+) create mode 100755 tests/btrfs/132 create mode 100644 tests/btrfs/132.out diff --git a/tests/btrfs/132 b/tests/btrfs/132 new file mode 100755 index 0000000..239707b --- /dev/null +++ b/tests/btrfs/132 @@ -0,0 +1,105 @@ +#! /bin/bash +# FS QA Test 132 +# +# Check if false ENOSPC will happen when parallel buffer write happens +# The problem is caused by incorrect metadata reservation for any buffered +# write whose max extent size is not 128M (including compression and in-band +# dedupe). +# +#----------------------------------------------------------------------- +# Copyright (c) 2016 Fujitsu. All Rights Reserved. +# +# This program is free software; you can redistribute it and/or +# modify it under the terms of the GNU General Public License as +# published by the Free Software Foundation. +# +# This program is distributed in the hope that it would be useful, +# but WITHOUT ANY WARRANTY; without even the implied warranty of +# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the +# GNU General Public License for more details. +# +# You should have received a copy of the GNU General Public License +# along with this program; if not, write the Free Software Foundation, +# Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA +#----------------------------------------------------------------------- +# + +seq=`basename $0` +seqres=$RESULT_DIR/$seq +echo "QA output created by $seq" + +here=`pwd` +tmp=/tmp/$$ +status=1 # failure is the default! +trap "_cleanup; exit \$status" 0 1 2 3 15 + +_cleanup() +{ + cd / + rm -f $tmp.* + kill $pids &> /dev/null + wait +} + +# get standard environment, filters and checks +. ./common/rc +. ./common/filter + +# remove previous $seqres.full before test +rm -f $seqres.full + +# real QA test starts here + +# Modify as appropriate. +_supported_fs btrfs +_supported_os Linux +_require_scratch + +# Use small filesystem to trigger the bug more easily +# It's highly recommened to run this test case with MKFS_OPTIONS="-n 64k" +# to further increase the possibility +# Since the false ENOSPC happens due to incorrect metadata reservation, +# larger nodesize and small fs will make it much easier to reproduce +_scratch_mkfs_sized $((512 * 1024 * 1024)) >> $seqres.full 2>&1 + +# Recommended to use MOUNT_OPTIONS="-o compress" to trigger the bug +_scratch_mount + +sleep_time=$(($TIME_FACTOR * 30)) +loop_writer() +{ + offset=$1 + len=$2 + file=$3 + + while true; + do + # Only need stderr, and we need to specify small block size + # but still large enough trigger compression + $XFS_IO_PROG -c "pwrite -b 8K $offset $len" $file 2>&1 \ + > /dev/null + done +} + +touch $SCRATCH_MNT/testfile + +# Start 2 writter writting into the same file +# The file is 142M, which is smaller than 1/2 of the filesystem, +# so no other cause can lead to ENOSPC. +loop_writer 0 128M $SCRATCH_MNT/testfile & +pids=$! +loop_writer 128M 16M $SCRATCH_MNT/testfile & +pids="$pids $!" + +sleep $sleep_time + +kill $pids +wait + +# Sync the fs to avoid EBUSY while umount, which is quite common for btrfs +# compression +sync + +echo "Silence is golden" +status=0 +exit diff --git a/tests/btrfs/132.out b/tests/btrfs/132.out new file mode 100644 index 0000000..b096312 --- /dev/null +++ b/tests/btrfs/132.out @@ -0,0 +1,2 @@ +QA output created by 132 +Silence is golden diff --git a/tests/btrfs/group b/tests/btrfs/group index c090604..ec8ad80 100644 --- a/tests/btrfs/group +++ b/tests/btrfs/group @@ -134,3 +134,4 @@ 129 auto quick send 130 auto clone send 131 auto quick +132 auto compress