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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.