Re: [PATCH] fstests: generic: Test SHARED flag about fiemap ioctl before and after sync

[Qu Wenruo <quwenruo.btrfs@gmx.com>]

On 05/13/2016 01:17 PM, Darrick J. Wong wrote:
> On Fri, May 13, 2016 at 09:52:52AM +0800, Qu Wenruo wrote:
>> The test case will check SHARED flag returned by fiemap ioctl on
>> reflinked files before and after sync.
>>
>> Normally SHARED flag won't change just due to a normal sync operation.
>>
>> But btrfs doesn't handle SHARED flag well, and this time it won't check
>> any delayed extent tree(reverse extent searching tree) modification, but
>> only metadata already committed to disk.
>>
>> So btrfs will not return correct SHARED flag on reflinked files if there
>> is no sync to commit all metadata.
>>
>> This testcase will just check it.
>>
>> Signed-off-by: Qu Wenruo <quwenruo@cn.fujitsu.com>
>> ------
>> And of course, xfs handles it quite well. Nice work Darrick.
>> Also the test case needs the new infrastructure introduced in previous
>> generic/352 test case.
>> ---
>>  tests/generic/353     | 86 +++++++++++++++++++++++++++++++++++++++++++++++++++
>>  tests/generic/353.out |  9 ++++++
>>  tests/generic/group   |  1 +
>>  3 files changed, 96 insertions(+)
>>  create mode 100755 tests/generic/353
>>  create mode 100644 tests/generic/353.out
>>
>> diff --git a/tests/generic/353 b/tests/generic/353
>> new file mode 100755
>> index 0000000..1e9117e
>> --- /dev/null
>> +++ b/tests/generic/353
>> @@ -0,0 +1,86 @@
>> +#! /bin/bash
>> +# FS QA Test 353
>> +#
>> +# Check if fiemap ioctl returns correct SHARED flag on reflinked file
>> +# before and after sync the fs
>> +#
>> +# Btrfs has a bug in checking shared extent, which can only handle metadata
>> +# already committed to disk, but not delayed extent tree modification.
>> +# This caused SHARED flag only occurs after sync.
>
> I noticed this a while ago, but figured it was just btrfs being btrfs.  Ho hum.
>
> Thanks for writing a test and getting the problem fixed.
>
>> +#
>> +#-----------------------------------------------------------------------
>> +# 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.*
>
> tmp isn't used for anything in this testcase.

That's from template, and does no harm.

>
>> +}
>> +
>> +# get standard environment, filters and checks
>> +. ./common/rc
>> +. ./common/filter
>> +. ./common/reflink
>> +. ./common/punch
>> +
>> +# remove previous $seqres.full before test
>> +rm -f $seqres.full
>> +
>> +# real QA test starts here
>> +
>> +# Modify as appropriate.
>> +_supported_fs generic
>> +_supported_os Linux
>> +_require_scratch_reflink
>> +_require_fiemap
>> +
>> +_scratch_mkfs > /dev/null 2>&1
>> +_scratch_mount
>> +
>> +blocksize=64k
>> +file1="$SCRATCH_MNT/file1"
>> +file2="$SCRATCH_MNT/file2"
>> +
>> +# write the initial file
>> +_pwrite_byte 0xcdcdcdcd 0 $blocksize $file1 | _filter_xfs_io
>> +
>> +# reflink initial file
>> +_reflink_range $file1 0 $file2 0 $blocksize | _filter_xfs_io
>> +
>> +# check their fiemap to make sure it's correct
>> +$XFS_IO_PROG -c "fiemap -v" $file1 | _filter_fiemap_flags
>> +$XFS_IO_PROG -c "fiemap -v" $file2 | _filter_fiemap_flags
>> +
>> +# sync and recheck, to make sure the fiemap doesn't change just
>> +# due to sync
>> +sync
>> +$XFS_IO_PROG -c "fiemap -v" $file1 | _filter_fiemap_flags
>> +$XFS_IO_PROG -c "fiemap -v" $file2 | _filter_fiemap_flags
>
> Nowadays, when I write a test that prints similar output one after the other I
> will also write a comment to the output to distinguish the two cases, e.g.:
>
> +XXX Bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
> fiemap before sync
> +0: [0..127]: 0x2001
> +0: [0..127]: 0x2001
> fiemap after sync
> +0: [0..127]: 0x2001
> +0: [0..127]: 0x2001
>
> This way when a bunch of tests regress some months later it's easier
> for me to relearn what's going on.

Yes, that's pretty nice, I'll also add filename to the output, in case
the reflinked or original file layout changes.

Thanks for the review.
Qu
>
> (I wasn't always good at doing that.)
>
> Otherwise, everything looks ok.
>
> --D
>
>> +
>> +# success, all done
>> +status=0
>> +exit
>> diff --git a/tests/generic/353.out b/tests/generic/353.out
>> new file mode 100644
>> index 0000000..0cd8981
>> --- /dev/null
>> +++ b/tests/generic/353.out
>> @@ -0,0 +1,9 @@
>> +QA output created by 353
>> +wrote 65536/65536 bytes at offset 0
>> +XXX Bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
>> +linked 65536/65536 bytes at offset 0
>> +XXX Bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
>> +0: [0..127]: 0x2001
>> +0: [0..127]: 0x2001
>> +0: [0..127]: 0x2001
>> +0: [0..127]: 0x2001
>> diff --git a/tests/generic/group b/tests/generic/group
>> index 3f00386..0392d4d 100644
>> --- a/tests/generic/group
>> +++ b/tests/generic/group
>> @@ -355,3 +355,4 @@
>>  350 blockdev quick rw
>>  351 blockdev quick rw
>>  352 auto clone
>> +353 auto quick clone
>> --
>> 2.5.5
>>
>>
>>
>> --
>> To unsubscribe from this list: send the line "unsubscribe linux-btrfs" in
>> the body of a message to majordomo@vger.kernel.org
>> More majordomo info at  http://vger.kernel.org/majordomo-info.html
> --
> To unsubscribe from this list: send the line "unsubscribe linux-btrfs" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
>