All of lore.kernel.org
 help / color / mirror / Atom feed
From: Tsutomu Itoh <t-itoh@jp.fujitsu.com>
To: linux-btrfs@vger.kernel.org, fstests@vger.kernel.org
Subject: Re: [PATCH v2] fstests: btrfs: Add regression test for reserved space leak.
Date: Wed, 5 Aug 2015 10:26:09 +0900	[thread overview]
Message-ID: <55C16631.1060505@jp.fujitsu.com> (raw)
In-Reply-To: <1438736914-28147-1-git-send-email-quwenruo@cn.fujitsu.com>

On 2015/08/05 10:08, Qu Wenruo wrote:
> The regression is introduced in v4.2-rc1, with the big btrfs qgroup
> change.
> The problem is, qgroup reserved space is never freed, causing even we
> increase the limit, we can still hit the EDQUOT much faster than it
> should.
> 
> Reported-by: Tsutomu Itoh <t-itoh@jp.fujitsu.com>
> Signed-off-by: Qu Wenruo <quwenruo@cn.fujitsu.com>
> ---
>   tests/btrfs/089     | 88 +++++++++++++++++++++++++++++++++++++++++++++++++++++
>   tests/btrfs/089.out |  5 +++
>   tests/btrfs/group   |  1 +
>   3 files changed, 94 insertions(+)
>   create mode 100755 tests/btrfs/089
>   create mode 100644 tests/btrfs/089.out
> 
> diff --git a/tests/btrfs/089 b/tests/btrfs/089
> new file mode 100755
> index 0000000..82db96c
> --- /dev/null
> +++ b/tests/btrfs/089
> @@ -0,0 +1,88 @@
> +#! /bin/bash
> +# FS QA Test 089
> +#
> +# Regression test for btrfs qgroup reserved space leak.
> +#
> +# Due to qgroup reserved space leak, EDQUOT can be trigged even it's not
> +# over limit after previous write.
> +#
> +#-----------------------------------------------------------------------
> +# Copyright (c) 2015 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 $tmp.*
> +}
> +
> +# get standard environment, filters and checks
> +. ./common/rc
> +. ./common/filter
> +
> +# real QA test starts here
> +
> +# Modify as appropriate.
> +_supported_fs btrfs
> +_supported_os Linux
> +_require_scratch
> +_need_to_be_root
> +
> +# Use big blocksize to ensure there is still enough space left
> +# for metadata reserve after hitting EDQUOT
> +BLOCKSIZE=$(( 2 * 1024 * 1024 ))
> +FILESIZE=$(( 128 * 1024 * 1024 )) # 128Mbytes
> +
> +# The last block won't be able to finish write, as metadata takes
> +# $NODESIZE space, causing the last block triggering EDQUOT
> +LENGTH=$(( $FILESIZE - $BLOCKSIZE ))
> +
> +_scratch_mkfs >>$seqres.full 2>&1
> +_scratch_mount
> +_require_fs_space $SCRATCH_MNT $(($FILESIZE * 2 / 1024))
> +
> +_run_btrfs_util_prog quota enable $SCRATCH_MNT
> +_run_btrfs_util_prog qgroup limit $FILESIZE 5 $SCRATCH_MNT
> +
> +$XFS_IO_PROG -f -c "pwrite -b $BLOCKSIZE 0 $LENGTH" \
> +	$SCRATCH_MNT/foo | _filter_xfs_io
> +
> +# A sync is needed to trigger a commit_transaction.
> +# As the reserved space freeing happens at commit_transaction time,
> +# without a transaction commit, no reserved space needs freeing and
> +# won't trigger the bug.
> +sync

Isn't '$BTRFS_UTIL_PROG filesystem sync' better instead of 'sync'?

Thanks,
Tsutomu

> +
> +# Double the limit to allow further write
> +_run_btrfs_util_prog qgroup limit $(($FILESIZE * 2)) 5 $SCRATCH_MNT
> +
> +# Test whether further write can succeed
> +$XFS_IO_PROG -f -c "pwrite -b $BLOCKSIZE $LENGTH $LENGTH" \
> +	$SCRATCH_MNT/foo | _filter_xfs_io
> +
> +# success, all done
> +status=0
> +exit
> diff --git a/tests/btrfs/089.out b/tests/btrfs/089.out
> new file mode 100644
> index 0000000..396888f
> --- /dev/null
> +++ b/tests/btrfs/089.out
> @@ -0,0 +1,5 @@
> +QA output created by 089
> +wrote 132120576/132120576 bytes at offset 0
> +XXX Bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
> +wrote 132120576/132120576 bytes at offset 132120576
> +XXX Bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
> diff --git a/tests/btrfs/group b/tests/btrfs/group
> index ffe18bf..225b532 100644
> --- a/tests/btrfs/group
> +++ b/tests/btrfs/group
> @@ -91,6 +91,7 @@
>   086 auto quick clone
>   087 auto quick send
>   088 auto quick metadata
> +089 auto quick qgroup
>   090 auto quick metadata
>   091 auto quick qgroup
>   092 auto quick send
> 



  reply	other threads:[~2015-08-05  1:36 UTC|newest]

Thread overview: 12+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-08-05  1:08 [PATCH v2] fstests: btrfs: Add regression test for reserved space leak Qu Wenruo
2015-08-05  1:08 ` Qu Wenruo
2015-08-05  1:26 ` Tsutomu Itoh [this message]
2015-08-05  1:39   ` Qu Wenruo
2015-08-05  1:39     ` Qu Wenruo
2015-08-05  1:57     ` Dave Chinner
2015-08-05  2:52       ` Tsutomu Itoh
2015-08-05  2:59         ` Qu Wenruo
2015-08-05  2:59           ` Qu Wenruo
2015-08-05  3:10         ` Dave Chinner
2015-08-05  4:04           ` Tsutomu Itoh
2015-08-05  7:46 ` Filipe David Manana

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=55C16631.1060505@jp.fujitsu.com \
    --to=t-itoh@jp.fujitsu.com \
    --cc=fstests@vger.kernel.org \
    --cc=linux-btrfs@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.