Re: [PATCH v2] fstests: generic: Check if a bull fallocate will change extent number

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

在 2015年09月29日 17:55, Eryu Guan 写道:
> On Tue, Sep 29, 2015 at 05:34:24PM +0800, Qu Wenruo wrote:
>> Normally, a bull fallocate call on a fully written and synced file
>> should not add an extent.
>>
>> But not all filesystem follows the correct behavior.
>>
>> Btrfs has a bug to always truncate the last page if the fallocate start
>> offset is smaller than inode size.
>>
>> So add this test case to detect such malfunction.
>>
>> Signed-off-by: Qu Wenruo <quwenruo@cn.fujitsu.com>
>> ---
>> v2:
>>    Add author info...
>>    Fix some comment typo
>
> (I was going to point these out then saw a v2 :)
>
>> ---
>>   tests/generic/110     | 78 +++++++++++++++++++++++++++++++++++++++++++++++++++
>>   tests/generic/110.out |  3 ++
>>   tests/generic/group   |  1 +
>>   3 files changed, 82 insertions(+)
>>   create mode 100755 tests/generic/110
>>   create mode 100644 tests/generic/110.out
>>
>> diff --git a/tests/generic/110 b/tests/generic/110
>> new file mode 100755
>> index 0000000..b2b140c
>> --- /dev/null
>> +++ b/tests/generic/110
>> @@ -0,0 +1,78 @@
>> +#! /bin/bash
>> +# FS QA Test 110
>> +#
>> +# Test if fallocate will create uneeded extra tailing extent
>> +#
>> +#-----------------------------------------------------------------------
>> +# 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
>> +. ./common/defrag
>> +
>> +# remove previous $seqres.full before test
>> +rm -f $seqres.full
>> +
>> +# real QA test starts here
>> +
>> +_supported_fs generic
>> +_supported_os IRIX Linux
>> +_require_scratch
>> +_need_to_be_root
>
> need a '_require_xfs_io_command "falloc"' here, so test _notrun
> correctly on ext2/3/NFS/CIFS that don't support fallocate.

Will add it.

>
>> +
>> +# Use 64K file size to match any sectorsize
>> +# And with a unaligned tailing range to ensure it will be at least 2 pages
>> +filesize=$(( 64 * 1024 + 1024 ))
>> +
>> +_scratch_mkfs > /dev/null 2>&1
>> +_scratch_mount
>> +$XFS_IO_PROG -f -c "pwrite 0 $filesize" $SCRATCH_MNT/foo | _filter_xfs_io
>> +sync
>> +orig_extent_nr=`_extent_count $SCRATCH_MNT/foo`
>> +
>> +# As all space are allocated and even written to disk, this falloc
>> +# should do nothing with extent modification.
>> +$XFS_IO_PROG -f -c "falloc 0 $filesize" $SCRATCH_MNT/foo
>> +sync
>> +new_extent_nr=`_extent_count $SCRATCH_MNT/foo`
>> +
>> +echo "orig: $orig_extent_nr, new: $new_extent_nr" >> $seqres.full
>> +
>> +if [ "x$orig_extent_nr" != "x$new_extent_nr" ]; then
>> +	echo "number of extents mis-match after bull fallocate"
>
> print out the $orig_extent_nr and $new_extent_nr in this failure case? I
> think it's useful to see the difference just from the output diff, don't
> have to check the full file.

The problem is, we can't ensure orig/new_extent_nr always be a constant 
value(1 for btrfs case).

Maybe there is some file system which can only allocate extent up to 
64K, then the orig_extent_nr will be 2 and fails on that filesystem.

So I follow the method used in btrfs/010 to do it.

Thanks,
Qu
>
> Thanks,
> Eryu
>
>> +fi
>> +
>> +# success, all done
>> +status=0
>> +exit
>> diff --git a/tests/generic/110.out b/tests/generic/110.out
>> new file mode 100644
>> index 0000000..64049da
>> --- /dev/null
>> +++ b/tests/generic/110.out
>> @@ -0,0 +1,3 @@
>> +QA output created by 110
>> +wrote 66560/66560 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 4ae256f..428f3e3 100644
>> --- a/tests/generic/group
>> +++ b/tests/generic/group
>> @@ -112,6 +112,7 @@
>>   107 auto quick metadata
>>   108 auto quick rw
>>   109 auto metadata dir
>> +110 auto quick prealloc
>>   112 rw aio auto quick
>>   113 rw aio auto quick
>>   117 attr auto quick
>> --
>> 1.8.3.1
>>
>> --
>> 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
> --
> 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
>