All of lore.kernel.org
 help / color / mirror / Atom feed
From: Eryu Guan <eguan@redhat.com>
To: Goldwyn Rodrigues <rgoldwyn@suse.de>
Cc: fstests@vger.kernel.org, darrick.wong@oracle.com,
	Goldwyn Rodrigues <rgoldwyn@suse.com>
Subject: Re: [PATCH v2 3/3] generic/471: Partial direct write test
Date: Wed, 29 Nov 2017 18:46:28 +0800	[thread overview]
Message-ID: <20171129104628.GR2749@eguan.usersys.redhat.com> (raw)
In-Reply-To: <20171127184920.13864-3-rgoldwyn@suse.de>

On Mon, Nov 27, 2017 at 12:49:20PM -0600, Goldwyn Rodrigues wrote:
> From: Goldwyn Rodrigues <rgoldwyn@suse.com>
> 
> Getting an error midway through a direct write would return an error
> and the error-code is returned in the write() call. However, partial
> data is over-written during the call.
> 
> This tests simulates the ENOSPC error to check for partial direct
> write consistency.
> 
> Signed-off-by: Goldwyn Rodrigues <rgoldwyn@suse.com>
> 
> Changes since v1:
>  - Check for option pwrite -O
>  - Create a 200M file instead of multiple 40M files.
> 
> ---
>  tests/generic/471     | 90 +++++++++++++++++++++++++++++++++++++++++++++++++++
>  tests/generic/471.out |  9 ++++++
>  tests/generic/group   |  1 +
>  3 files changed, 100 insertions(+)
>  create mode 100755 tests/generic/471
>  create mode 100644 tests/generic/471.out
> 
> diff --git a/tests/generic/471 b/tests/generic/471
> new file mode 100755
> index 00000000..a7780af1
> --- /dev/null
> +++ b/tests/generic/471
> @@ -0,0 +1,90 @@
> +#! /bin/bash
> +# FS QA Test No. 471
> +#
> +# write() to a file opened with O_DIRECT with count > remaining
> +# bytes. In case of a bug, the write returns ENOSPC wheras the
> +# showing no data is written, but the file contents are updated.
> +# Result should be the write should return remaining bytes
> +# (to aligned bytes) instead of ENOSPC error.
> +#
> +#-----------------------------------------------------------------------
> +# Copyright (c) 2017, SUSE Linux Products.  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 -rf $tmp.* $testdir1
> +}
> +
> +# get standard environment, filters and checks
> +. ./common/rc
> +. ./common/populate
> +. ./common/filter
> +. ./common/attr
> +
> +# real QA test starts here
> +_supported_os Linux
> +_require_odirect

Missed _require_scratch

> +_require_xfs_io_command pwrite -O
> +
> +rm -f $seqres.full
> +
> +echo "Reformat with 320M size"
> +umount $SCRATCH_MNT

No need to umount $SCRATCH_MNT, _require_scratch will do it for you.

> +sz_bytes=$((320 * 1024 * 1024))
> +_scratch_mkfs_sized $sz_bytes >> $seqres.full 2>&1
> +_scratch_mount >> $seqres.full 2>&1
> +testdir=$SCRATCH_MNT/test-$seq
> +rm -rf $testdir

No need to remove $testdir here, scratch_dev is freshly created.

> +mkdir $testdir
> +sync
> +
> +#Almost fill the filesystem
   ^^ add a space after '#' for comments

> +echo "Almost fill the filesystem"
> +$XFS_IO_PROG -f -c "pwrite 0 200M" $testdir/f1 > /dev/null
> +
> +# Create a file using buffered I/O which succeeds only partially
> +$XFS_IO_PROG -f -c "pwrite -w -S 0xaa 0 128M" $testdir/partial
> +
> +# re-write using direct I/O with another pattern using one single buffer, once
> +write_size=`$XFS_IO_PROG -d -c "pwrite -O -S 0xcd -V 1 -b 128M 0 128M" $testdir/partial | awk '/^wrote/ {split($2, bytes, "/"); print bytes[1]}'`

Hmm, I can see that "-O" requires bigger block size in this case. Then
how about reserving some space first and fill the rest then remove the
reserved file? So we have a fixed amount of space to do the test. e.g.

# reserve 2m space first
$XFS_IO_PROG -fc "pwrite 0 2m" $testdir/reserve
# fill the rest of the fs
$XFS_IO_PROG -fc "pwrite 0 320m" $testdir/fill
# remove the reserve file to get space free
rm -f $testdir/reserve
...
$XFS_IO_PROG -dc "pwrite -O -S 0xcd -V 1 -b 4m 0 4m" $testdir/partial
...

Thanks,
Eryu

> +if [ -z $write_size ]; then
> +	write_size=0
> +fi
> +
> +# If bug is not fixed, pwrite will return error and but still write data
> +if [ $write_size -gt 0 ]; then
> +	echo "pwrite wrote more than zero bytes."
> +else
> +	echo "pwrite wrote zero bytes"
> +fi
> +
> +# read the partial file to check if data written is of last write
> +$XFS_IO_PROG -c "pread 0 2M -v" $testdir/partial | _filter_xfs_io_unique
> +
> +# success, all done
> +status=0
> +exit
> diff --git a/tests/generic/471.out b/tests/generic/471.out
> new file mode 100644
> index 00000000..e01687c8
> --- /dev/null
> +++ b/tests/generic/471.out
> @@ -0,0 +1,9 @@
> +QA output created by 471
> +Reformat with 320M size
> +Almost fill the filesystem
> +pwrite: No space left on device
> +pwrite wrote more than zero bytes.
> +00000000:  cd cd cd cd cd cd cd cd cd cd cd cd cd cd cd cd  ................
> +*
> +read 2097152/2097152 bytes at offset 0
> +XXX Bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
> diff --git a/tests/generic/group b/tests/generic/group
> index 11ccfb01..f4591620 100644
> --- a/tests/generic/group
> +++ b/tests/generic/group
> @@ -473,3 +473,4 @@
>  468 shutdown auto quick metadata
>  469 auto quick
>  470 auto quick
> +471 auto quick rw
> -- 
> 2.14.2
> 
> --
> To unsubscribe from this list: send the line "unsubscribe fstests" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html

  reply	other threads:[~2017-11-29 10:46 UTC|newest]

Thread overview: 6+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2017-11-27 18:49 [PATCH 1/3] Check pwrite parameters Goldwyn Rodrigues
2017-11-27 18:49 ` [PATCH v2 2/3] generic/470: Test RWF_NOWAIT Goldwyn Rodrigues
2017-11-29 10:25   ` Eryu Guan
2017-11-27 18:49 ` [PATCH v2 3/3] generic/471: Partial direct write test Goldwyn Rodrigues
2017-11-29 10:46   ` Eryu Guan [this message]
2017-11-29  9:49 ` [PATCH 1/3] Check pwrite parameters Eryu Guan

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=20171129104628.GR2749@eguan.usersys.redhat.com \
    --to=eguan@redhat.com \
    --cc=darrick.wong@oracle.com \
    --cc=fstests@vger.kernel.org \
    --cc=rgoldwyn@suse.com \
    --cc=rgoldwyn@suse.de \
    /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.