Re: [PATCH] btrfs: check truncate can update file size correctly when truncate fails

[Wang Xiaoguang <wangxg.fnst@cn.fujitsu.com>]

hello,

Thanks for your comments.
I will update the patch according to your comments and make it to 
generic tests.

Regards,
Xiaoguang Wang

On 06/23/2016 04:27 PM, Eryu Guan wrote:
> On Thu, Jun 23, 2016 at 03:36:40PM +0800, Wang Xiaoguang wrote:
>> In btrfs, when truncate operation fails for enospc reason, file may still
>> have some disk blocks, but it will fail to update filesize accordingly.
>>
>> Signed-off-by: Wang Xiaoguang <wangxg.fnst@cn.fujitsu.com>
>> ---
>>   tests/btrfs/124     | 86 +++++++++++++++++++++++++++++++++++++++++++++++++++++
>>   tests/btrfs/124.out |  2 ++
>>   tests/btrfs/group   |  1 +
>>   3 files changed, 89 insertions(+)
>>   create mode 100755 tests/btrfs/124
>>   create mode 100644 tests/btrfs/124.out
>>
>> diff --git a/tests/btrfs/124 b/tests/btrfs/124
>> new file mode 100755
>> index 0000000..1a13ddf
>> --- /dev/null
>> +++ b/tests/btrfs/124
>> @@ -0,0 +1,86 @@
>> +#! /bin/bash
>> +# FS QA Test 124
>> +#
>> +# Test whether truncate can update file size correctly when truncate fails
>> +#
>> +# When truncate operation fails for enospc reason, file will still
>> +# have some disk blocks, but it will fail to update filesize accordingly.
>> +#
>> +#-----------------------------------------------------------------------
>> +# Copyright (c) 2016 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
>> +. ./common/reflink
>> +
>> +# remove previous $seqres.full before test
>> +rm -f $seqres.full
>> +
>> +# Modify as appropriate.
>> +_supported_fs generic
>> +_supported_os Linux
>> +_supported_fs btrfs
> Two "_supported_fs" here.
>
> And I don't see any other btrfs-specific configs except MKFS_OPTIONS, is
> it possible to make it a generic test?
>
>> +_require_scratch_reflink
>> +
>> +MKFS_OPTIONS="-O ^mixed-bg"
> Need comments on why testing with this MKFS_OPTIONS.
>
>> +_scratch_mkfs_sized $((101 * 1024 * 1024)) >> $seqres.full 2>&1
> 101M is too small to make btrfs on ppc64 host for me.
>
>> +_scratch_mount
>> +
>> +blocksize=$((128 * 1024))
>> +file="$SCRATCH_MNT/testfile"
>> +filesize=$((12 * 1024 * 1024 * 1024))
> Comments on these numbers picked would be good, i.e. why 128k blocksize,
> why 12M filesize. Any special reason or just good enough for the test?
>
>> +nr=$((filesize / blocksize))
>> +
>> +# write the initial block for later reflink
>> +_pwrite_byte 0xcdcdcdcd 0 $blocksize $file > /dev/null
>> +
>> +# use reflink to create the rest of the file, whose all extents are all
>> +# pointing to the first extent
>> +for ((i = 1; i < $nr; i++)); do
>> +	_reflink_range $file 0 $file $(($i * $blocksize)) $blocksize > /dev/null
>> +done
> It's not clear to me in test description that why reflink is needed in
> this test. Perhaps test description and comments above can be updated to
> reflect the purpose of reflink?
>
>> +
>> +$XFS_IO_PROG -f -c "truncate 0" $file > /dev/null 2>&1
>> +disk_usage=`du $file | $AWK_PROG '{print $1}'`
>> +new_filesize=`ls -l $file | $AWK_PROG '{print $5}'`
> I think "stat -c %s $file" is much easier :)
>
>> +if [ $disk_usage -gt 0 ] && [ $new_filesize -eq 0 ]; then
>> +	echo "after truncate, file size is 0, but file still owns disk blocks"
>> +	status=1
>> +	exit
>> +fi
>> +
>> +echo "Silence is golden"
>> +# success, all done
>> +status=0
>> +exit
>> diff --git a/tests/btrfs/124.out b/tests/btrfs/124.out
>> new file mode 100644
>> index 0000000..d7339e6
>> --- /dev/null
>> +++ b/tests/btrfs/124.out
>> @@ -0,0 +1,2 @@
>> +QA output created by 124
>> +Silence is golden
>> diff --git a/tests/btrfs/group b/tests/btrfs/group
>> index 5a26ed7..8b5050e 100644
>> --- a/tests/btrfs/group
>> +++ b/tests/btrfs/group
>> @@ -126,3 +126,4 @@
>>   121 auto quick snapshot qgroup
>>   122 auto quick snapshot qgroup
>>   123 auto quick qgroup
>> +124 auto quick metadata
> Add it to 'clone' group as well?
>
> Thanks,
> Eryu
>
>