Re: [PATCH v4] ext4/030: Ext4 online resize with bigalloc tests.

[Date Prev] [Date Next] [Thread Prev] [Thread Next] [Date Index] [Thread Index]



On Wed, Jan 10, 2018 at 11:07 PM, Eryu Guan <eguan@xxxxxxxxxx> wrote:
> On Sun, Jan 07, 2018 at 08:18:15PM -0800, harshads wrote:
>> Add tests to verify Ext4 online resizing feature with bigalloc feature
>> enabled. We test various resizing scenarios with different cluster
>> sizes.
>>
>> Signed-off-by: Harshad Shirwadkar <harshads@xxxxxxxxxx>
>
> I have some more comments inline, sorry for not bringing up all comments
> in previous reviews.. And please cc ext4 list for new ext4 test.
>
> One minor issue regarding to the patch summary:
>
> ext4/030: Ext4 online resize with bigalloc tests
>
> there's no need to include a fixed test seq number in summary for new
> test, new tests usually will be re-numbered on commit.
Done.
>
>> ---
>>  tests/ext4/030     | 156 +++++++++++++++++++++++++++++++++++++++++++++++++++++
>>  tests/ext4/030.out | 148 ++++++++++++++++++++++++++++++++++++++++++++++++++
>>  tests/ext4/group   |   1 +
>>  3 files changed, 305 insertions(+)
>>  create mode 100755 tests/ext4/030
>>  create mode 100644 tests/ext4/030.out
>>
>> diff --git a/tests/ext4/030 b/tests/ext4/030
>> new file mode 100755
>> index 00000000..99ca9362
>> --- /dev/null
>> +++ b/tests/ext4/030
>> @@ -0,0 +1,156 @@
>> +#! /bin/bash
>> +# FS QA Test ext4/030
>> +#
>> +# Ext4 online resize tests of small and crucial resizes with bigalloc
>> +# feature.
>> +#
>> +#-----------------------------------------------------------------------
>> +# Copyright (c) 2017 Google, Inc.  All Rights Reserved.
>> +#
>> +# Author: Harshad Shirwadkar <harshads@xxxxxxxxxx>
>> +#
>> +# 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
>> +
>> +_bytes2blk()
>
> Usually the leading underscore is only used for common helper names, not
> a local function, "bytes2blk" is fine.
Done.
>
>> +{
>> +    local bytes=$1
>
> And please indent with tab not 4 spaces.
Done.
>
>> +    BLKSIZ=4096
>
> BLKSIZ is always 4096 and is only used in this function, move it out of
> bytes2blk? e.g.
>
> BLKSIZ=4096
> bytes2blk()
> {
> ...
> }
>
Done.
> This leads me to wonder that do we need to test other block sizes like
> 1k and 2k?
As of now, I have performed testing with 4k and 2k block sizes.
Everything looks good. With 1k block size, there are a few small
fixes. So I was we could get this version of the test in. When 1k
block fixes are in the kernel, we could enhance the test suite to test
that.
>
>> +    echo $((bytes / $BLKSIZ))
>> +}
>> +
>> +_ext4_online_resize()
>> +{
>> +    local image_file=$1
>> +    local image_mount=$2
>> +    local original_size=$3
>> +    local final_size=$4
>> +    local cluster_size=$5
>> +
>> +    echo "+ truncate image file to $(_bytes2blk $final_size)"
>
> Append such logs to $seqres.full too? I might be easier for debugging.
Done.
>
>> +    $XFS_IO_PROG -f -c "truncate ${final_size}" ${image_file}
>> +    LOOP_DEVICE=`_create_loop_device $image_file || _fail "losetup failed"`
>> +
>> +    echo "+ create fs on image file $(_bytes2blk $original_size)"
>> +    ${MKFS_PROG}.${FSTYP} -F -O bigalloc,resize_inode,metadata_csum \
>
> Why do we need "metadata_csum" here? Some comments would be fine if it's
> really needed. Better to have comments on the "resize_inode" feature
> too.
>
>> +             -C ${cluster_size} -b 4096 \
>> +             ${LOOP_DEVICE} $(_bytes2blk $original_size) > /dev/null 2>&1 || \
>> +     _fail "mkfs failed"
>> +
>> +    echo "+ mount image file"
>> +    $MOUNT_PROG -t ${FSTYP} ${LOOP_DEVICE} ${image_mount} > \
>> +             /dev/null 2>&1 || _fail "mount failed"
>> +
>> +    echo "+ resize fs to $(_bytes2blk $final_size)"
>> +    resize2fs -f ${LOOP_DEVICE} $(_bytes2blk $final_size) >> $seqres.full 2>&1 || \
>> +     _fail "resize2fs failed"
>
> Hmm, online resize support for bigalloc ext4 is added in v4.15-rc1,
> tests would fail on old kernels. I think we need a new _require rule to
> make sure current kernel does support ext4 online resize with bigalloc
> feature.
>
>> +
>> +    echo "+ umount fs"
>> +    $UMOUNT_PROG ${image_mount}
>> +
>> +    echo "+ check fs"
>> +    _check_generic_filesystem $LOOP_DEVICE >> $seqres.full 2>&1 || \
>> +     _fail "fsck should not fail"
>> +    _destroy_loop_device $LOOP_DEVICE && LOOP_DEVICE=
>> +}
>> +
>> +_cleanup()
>> +{
>> +    cd /
>> +    [ -n "$LOOP_DEVICE" ] && _destroy_loop_device $LOOP_DEVICE > /dev/null 2>&1
>> +    rm -f $tmp.*
>> +    umount ${IMG_MNT} > /dev/null 2>&1
>
> $UMOUNT_PROG
Done.
>
>> +    rm -f ${IMG_FILE} > /dev/null 2>&1
>> +}
>> +
>> +# get standard environment and checks
>> +. ./common/rc
>> +
>> +# remove previous $seqres.full before test
>> +rm -f $seqres.full
>> +
>> +# real QA test starts here
>> +_supported_fs ext4
>> +_supported_os Linux
>> +
>> +_require_loop
>> +_require_scratch
>> +_require_scratch_ext4_feature "bigalloc,resize_inode,metadata_csum"
>> +
>> +IMG_FILE=$TEST_DIR/$seq.fs
>> +IMG_MNT=$TEST_DIR/$seq.mnt
>
> We required scratch device, then just create fs images on $SCRATCH_MNT?
> i.e.
>
> rm -f $seqres.full
> _scratch_mkfs >>$seqres.full 2>&1
> _scratch_mount
>
> IMG_FILE=$SCRATCH_MNT/$seq.fs
> IMG_MNT=$SCRATCH_MNT/$seq.mnt
Done.
>
>> +
>> +rm -f $seqres.full
>> +
>> +rmdir $IMG_MNT 2>/dev/null
>
> Then there's no need to do rmdir on newly created filesystem.
Done.
>
>> +mkdir -p $IMG_MNT || _fail "cannot create loopback mount point"
>> +
>> +## We perform resizing to various multiples of block group sizes to
>> +## ensure that we cover maximum edge cases in the kernel code.
>
> I found that your reply to Ted in the ext4 bigalloc online resize path
> thread provided more details of the test:
>
> "... ensure that all different code paths that resize IOCTL triggers get
> executed. At a high level, these different code paths are - 1) extending
> last block group, 2) adding new block groups 3) conversion of file
> system to meta-bg."
>
> I think it's better to add similar comments in the test too.
I added comments in the test.
>
> Thanks,
> Eryu
>
>> +for cluster_size in 4096 16384 65536; do
>> +    echo "+ set cluster size to ${cluster_size}"
>> +    ## Extending a 1/2 block group to a 2/3 block group.
>> +    _ext4_online_resize ${IMG_FILE} ${IMG_MNT} \
>> +                     $((16384 * $cluster_size)) \
>> +                     $((24576 * $cluster_size)) ${cluster_size}
>> +
>> +    ## Extending a 2/3rd block group to one cluster less than a
>> +    ## full block group.
>> +    _ext4_online_resize ${IMG_FILE} ${IMG_MNT} \
>> +                     $((24576 * $cluster_size)) \
>> +                     $((32767 * $cluster_size)) ${cluster_size}
>> +
>> +    ## Extending a 2/3rd block group to a full block group.
>> +    _ext4_online_resize ${IMG_FILE} ${IMG_MNT} \
>> +                     $((24576 * $cluster_size)) \
>> +                     $((32768 * $cluster_size)) ${cluster_size}
>> +
>> +    ## Extending a 2/3rd block group to 2 block groups.
>> +    _ext4_online_resize ${IMG_FILE} ${IMG_MNT} \
>> +                     $((24576 * $cluster_size)) \
>> +                     $((65536 * $cluster_size)) ${cluster_size}
>> +
>> +    ## Extending a 2/3rd block group to 15 block groups and one
>> +    ## cluster.
>> +    _ext4_online_resize ${IMG_FILE} ${IMG_MNT} \
>> +                     $((24576 * $cluster_size)) \
>> +                     $((491521 * $cluster_size)) ${cluster_size}
>> +
>> +    ## Extending a 2/3rd block group to 15 and a half block groups.
>> +    _ext4_online_resize ${IMG_FILE} ${IMG_MNT} \
>> +                     $((24576 * $cluster_size)) \
>> +                     $((507904 * $cluster_size)) ${cluster_size}
>> +
>> +    ## Extending a 2/3rd block group to 16 block groups.
>> +    _ext4_online_resize ${IMG_FILE} ${IMG_MNT} \
>> +                     $((24576 * $cluster_size)) \
>> +                     $((524288 * $cluster_size)) ${cluster_size}
>> +
>> +    ## Extending a 2/3rd block group to 160 block groups.
>> +    _ext4_online_resize ${IMG_FILE} ${IMG_MNT} \
>> +                     $((24576 * $cluster_size)) \
>> +                     $((5242880 * $cluster_size)) ${cluster_size}
>> +done
>> +
>> +status=0
>> +exit
>> diff --git a/tests/ext4/030.out b/tests/ext4/030.out
>> new file mode 100644
>> index 00000000..17cc4d99
>> --- /dev/null
>> +++ b/tests/ext4/030.out
>> @@ -0,0 +1,148 @@
>> +QA output created by 030
>> ++ set cluster size to 4096
>> ++ truncate image file to 24576
>> ++ create fs on image file 16384
>> ++ mount image file
>> ++ resize fs to 24576
>> ++ umount fs
>> ++ check fs
>> ++ truncate image file to 32767
>> ++ create fs on image file 24576
>> ++ mount image file
>> ++ resize fs to 32767
>> ++ umount fs
>> ++ check fs
>> ++ truncate image file to 32768
>> ++ create fs on image file 24576
>> ++ mount image file
>> ++ resize fs to 32768
>> ++ umount fs
>> ++ check fs
>> ++ truncate image file to 65536
>> ++ create fs on image file 24576
>> ++ mount image file
>> ++ resize fs to 65536
>> ++ umount fs
>> ++ check fs
>> ++ truncate image file to 491521
>> ++ create fs on image file 24576
>> ++ mount image file
>> ++ resize fs to 491521
>> ++ umount fs
>> ++ check fs
>> ++ truncate image file to 507904
>> ++ create fs on image file 24576
>> ++ mount image file
>> ++ resize fs to 507904
>> ++ umount fs
>> ++ check fs
>> ++ truncate image file to 524288
>> ++ create fs on image file 24576
>> ++ mount image file
>> ++ resize fs to 524288
>> ++ umount fs
>> ++ check fs
>> ++ truncate image file to 5242880
>> ++ create fs on image file 24576
>> ++ mount image file
>> ++ resize fs to 5242880
>> ++ umount fs
>> ++ check fs
>> ++ set cluster size to 16384
>> ++ truncate image file to 98304
>> ++ create fs on image file 65536
>> ++ mount image file
>> ++ resize fs to 98304
>> ++ umount fs
>> ++ check fs
>> ++ truncate image file to 131068
>> ++ create fs on image file 98304
>> ++ mount image file
>> ++ resize fs to 131068
>> ++ umount fs
>> ++ check fs
>> ++ truncate image file to 131072
>> ++ create fs on image file 98304
>> ++ mount image file
>> ++ resize fs to 131072
>> ++ umount fs
>> ++ check fs
>> ++ truncate image file to 262144
>> ++ create fs on image file 98304
>> ++ mount image file
>> ++ resize fs to 262144
>> ++ umount fs
>> ++ check fs
>> ++ truncate image file to 1966084
>> ++ create fs on image file 98304
>> ++ mount image file
>> ++ resize fs to 1966084
>> ++ umount fs
>> ++ check fs
>> ++ truncate image file to 2031616
>> ++ create fs on image file 98304
>> ++ mount image file
>> ++ resize fs to 2031616
>> ++ umount fs
>> ++ check fs
>> ++ truncate image file to 2097152
>> ++ create fs on image file 98304
>> ++ mount image file
>> ++ resize fs to 2097152
>> ++ umount fs
>> ++ check fs
>> ++ truncate image file to 20971520
>> ++ create fs on image file 98304
>> ++ mount image file
>> ++ resize fs to 20971520
>> ++ umount fs
>> ++ check fs
>> ++ set cluster size to 65536
>> ++ truncate image file to 393216
>> ++ create fs on image file 262144
>> ++ mount image file
>> ++ resize fs to 393216
>> ++ umount fs
>> ++ check fs
>> ++ truncate image file to 524272
>> ++ create fs on image file 393216
>> ++ mount image file
>> ++ resize fs to 524272
>> ++ umount fs
>> ++ check fs
>> ++ truncate image file to 524288
>> ++ create fs on image file 393216
>> ++ mount image file
>> ++ resize fs to 524288
>> ++ umount fs
>> ++ check fs
>> ++ truncate image file to 1048576
>> ++ create fs on image file 393216
>> ++ mount image file
>> ++ resize fs to 1048576
>> ++ umount fs
>> ++ check fs
>> ++ truncate image file to 7864336
>> ++ create fs on image file 393216
>> ++ mount image file
>> ++ resize fs to 7864336
>> ++ umount fs
>> ++ check fs
>> ++ truncate image file to 8126464
>> ++ create fs on image file 393216
>> ++ mount image file
>> ++ resize fs to 8126464
>> ++ umount fs
>> ++ check fs
>> ++ truncate image file to 8388608
>> ++ create fs on image file 393216
>> ++ mount image file
>> ++ resize fs to 8388608
>> ++ umount fs
>> ++ check fs
>> ++ truncate image file to 83886080
>> ++ create fs on image file 393216
>> ++ mount image file
>> ++ resize fs to 83886080
>> ++ umount fs
>> ++ check fs
>> diff --git a/tests/ext4/group b/tests/ext4/group
>> index 257bb646..f29d3de6 100644
>> --- a/tests/ext4/group
>> +++ b/tests/ext4/group
>> @@ -32,6 +32,7 @@
>>  027 auto quick fsmap
>>  028 auto quick fsmap
>>  029 auto quick fsmap
>> +030 auto quick ioctl resize
>>  271 auto rw quick
>>  301 aio auto ioctl rw stress defrag
>>  302 aio auto ioctl rw stress defrag
>> --
>> 2.16.0.rc0.223.g4a4ac83678-goog
>>
--
To unsubscribe from this list: send the line "unsubscribe fstests" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html



[Index of Archives]     [Linux Filesystems Development]     [Linux NFS]     [Linux NILFS]     [Linux USB Devel]     [Linux Audio Users]     [Yosemite News]     [Linux Kernel]     [Linux SCSI]

  Powered by Linux