public inbox for linux-unionfs@vger.kernel.org
 help / color / mirror / Atom feed
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
> 

  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