From: Eryu Guan <eguan@redhat.com>
To: Chengguang Xu <cgxu519@icloud.com>
Cc: amir73il@gmail.com, fstests@vger.kernel.org,
linux-unionfs@vger.kernel.org
Subject: Re: [PATCH v2 3/3] generic/470: add syncfs test
Date: Thu, 7 Dec 2017 13:44:18 +0800 [thread overview]
Message-ID: <20171207054418.GC2749@eguan.usersys.redhat.com> (raw)
In-Reply-To: <1512613327-218016-3-git-send-email-cgxu519@icloud.com>
On Thu, Dec 07, 2017 at 10:22:07AM +0800, Chengguang Xu wrote:
> Inspired by syncfs bug of overlayfs which does not sync dirtyinodes in
> underlying filesystem.
> Run syncfs and shutdown filesystem(or underlying filesystem of overlayfs)
> to check syncfs result.
>
> Signed-off-by: Chengguang Xu <cgxu519@icloud.com>
> ---
>
> Changes since v1:
> Use fs shutdown and fssum to check syncfs result instead of
> checking delalloc state of extents.
>
> tests/generic/470 | 88 +++++++++++++++++++++++++++++++++++++++++++++++++++
> tests/generic/470.out | 2 ++
> tests/generic/group | 1 +
> 3 files changed, 91 insertions(+)
> create mode 100755 tests/generic/470
> create mode 100644 tests/generic/470.out
>
> diff --git a/tests/generic/470 b/tests/generic/470
> new file mode 100755
> index 0000000..b488747
> --- /dev/null
> +++ b/tests/generic/470
> @@ -0,0 +1,88 @@
> +#! /bin/bash
> +# FS QA Test 470
> +#
> +# Inspired by syncfs bug of overlayfs which does not sync dirty inodes in
> +# underlying filesystem.
Trailing whitespace in above line.
> +#
> +# Run syncfs and shutdown filesystem(or underlying filesystem of overlayfs)
> +# to check syncfs result.
> +#
> +# Test will be skipped if filesystem(or underlying filesystem of overlayfs)
> +# does not support shutdown.
> +#
> +#-----------------------------------------------------------------------
> +# Copyright (c) 2017 Chengguang Xu <cgxu519@icloud.com>
> +# 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
> +trap "_cleanup; exit \$status" 0 1 2 3 15
> +
> +_cleanup()
> +{
> + cd /
> + rm -f $tmp.*
> +}
> +
> +# 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
> +
> +_supported_fs generic
> +_supported_os Linux
> +_require_test
> +_require_fssum
> +_require_scratch
> +_require_scratch_shutdown
> +_require_xfs_io_command "syncfs"
> +
> +
> +FCNT=1000
> +
> +_scratch_mkfs >/dev/null 2>&1
> +_scratch_mount
> +
> +# In order to mitigate interference of write-back,
> +# create many files for test.
Sorry, I still don't understand how writeback could interfere this test
from this comment, what happens if we don't create such files? Why
writing files starting from offset 1k?
> +for i in `seq 1 $FCNT`; do
> + $XFS_IO_PROG -f -c "pwrite 1K 1M" \
> + $SCRATCH_MNT/$i >/dev/null 2>&1
> +done
> +
> +$FSSUM_PROG -ugomAcdES -f -w $TEST_DIR/before_test $SCRATCH_MNT
$tmp.fssum looks a better choice of the fssum record file, then you
don't need _require_test anymore.
> +$XFS_IO_PROG -c "syncfs" $SCRATCH_MNT/${FCNT} >/dev/null 2>&1
> +_scratch_shutdown
> +_scratch_cycle_mount
> +$FSSUM_PROG -ugomAcdES -f -w $TEST_DIR/after_test $SCRATCH_MNT
> +diff $TEST_DIR/before_test $TEST_DIR/after_test
> +if [ $? -eq 0 ]; then
> + status=0
> +fi
No need to check return value of diff here, just set status=0
unconditionally, test will break the golden image matching anyway if
diff finds something different.
But why not use "-r" option of fssum to read and compare the checksums?
e.g.
$FSSUM_PROG -r $tmp.fssum $SCRATCH_MNT
Thanks,
Eryu
> +
> +echo "Silence is golden"
> +exit
> diff --git a/tests/generic/470.out b/tests/generic/470.out
> new file mode 100644
> index 0000000..79fb532
> --- /dev/null
> +++ b/tests/generic/470.out
> @@ -0,0 +1,2 @@
> +QA output created by 470
> +Silence is golden
> diff --git a/tests/generic/group b/tests/generic/group
> index 6c3bb03..493a44c 100644
> --- a/tests/generic/group
> +++ b/tests/generic/group
> @@ -472,3 +472,4 @@
> 467 auto quick exportfs
> 468 shutdown auto quick metadata
> 469 auto quick
> +470 auto quick shutdown sync
> --
> 1.8.3.1
>
next prev parent reply other threads:[~2017-12-07 5:44 UTC|newest]
Thread overview: 22+ messages / expand[flat|nested] mbox.gz Atom feed top
2017-12-07 2:22 [PATCH v2 1/3] common/rc: add scratch shutdown support for overlayfs Chengguang Xu
2017-12-07 2:22 ` [PATCH v2 2/3] common/rc: add a check case in _require_xfs_io_command() to support syncfs Chengguang Xu
2017-12-07 2:22 ` [PATCH v2 3/3] generic/470: add syncfs test Chengguang Xu
2017-12-07 3:04 ` Amir Goldstein
2017-12-07 3:31 ` Chengguang Xu
2017-12-07 3:43 ` Amir Goldstein
2017-12-07 5:44 ` Eryu Guan [this message]
2017-12-07 6:20 ` Chengguang Xu
2017-12-07 7:13 ` Eryu Guan
2017-12-07 7:42 ` Chengguang Xu
2017-12-07 8:17 ` Amir Goldstein
2017-12-11 10:03 ` Chengguang Xu
2017-12-11 10:46 ` Amir Goldstein
2017-12-11 12:33 ` Chengguang Xu
2017-12-11 12:44 ` Amir Goldstein
2017-12-11 13:20 ` Chengguang Xu
2017-12-11 14:31 ` Chengguang Xu
2017-12-11 14:47 ` Amir Goldstein
2017-12-12 0:18 ` Dave Chinner
2017-12-07 2:54 ` [PATCH v2 1/3] common/rc: add scratch shutdown support for overlayfs Amir Goldstein
2017-12-07 5:31 ` Eryu Guan
2017-12-08 0:05 ` Dave Chinner
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20171207054418.GC2749@eguan.usersys.redhat.com \
--to=eguan@redhat.com \
--cc=amir73il@gmail.com \
--cc=cgxu519@icloud.com \
--cc=fstests@vger.kernel.org \
--cc=linux-unionfs@vger.kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox