Re: [PATCH] fstests: regression test for btrfs buffered read's repair

[Filipe Manana <fdmanana@gmail.com>]

On Wed, Apr 12, 2017 at 2:27 AM, Liu Bo <bo.li.liu@oracle.com> wrote:
> This case tests whether buffered read can repair the bad copy if we
> have a good copy.

Regardless of being a test we should have always had (thanks for
this!), it would be useful to mention we had a regression (as the test
description in the btrfs/141 file says) and which patch fixed it (and
possibly which kernel version or patch/commit introduced the
regression).

Just a couple comments/questions below.

>
> Signed-off-by: Liu Bo <bo.li.liu@oracle.com>
> ---
>  tests/btrfs/141     | 152 ++++++++++++++++++++++++++++++++++++++++++++++++++++
>  tests/btrfs/141.out |  39 ++++++++++++++
>  tests/btrfs/group   |   1 +
>  3 files changed, 192 insertions(+)
>  create mode 100755 tests/btrfs/141
>  create mode 100644 tests/btrfs/141.out
>
> diff --git a/tests/btrfs/141 b/tests/btrfs/141
> new file mode 100755
> index 0000000..53fd75c
> --- /dev/null
> +++ b/tests/btrfs/141
> @@ -0,0 +1,152 @@
> +#! /bin/bash
> +# FS QA Test 141
> +#
> +# Regression test for btrfs buffered read's repair during read.
> +#
> +#-----------------------------------------------------------------------
> +# Copyright (c) 2017 Liu Bo.  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
> +
> +# remove previous $seqres.full before test
> +rm -f $seqres.full
> +
> +# real QA test starts here
> +
> +# Modify as appropriate.
> +_supported_fs btrfs
> +_supported_os Linux
> +_require_scratch_dev_pool 2
> +_require_command "$BTRFS_MAP_LOGICAL_PROG" btrfs-map-logical
> +_require_command "$FILEFRAG_PROG" filefrag
> +
> +# helpe to convert 'file offset' to btrfs logical offset
> +FILEFRAG_FILTER='
> +       if (/blocks? of (\d+) bytes/) {
> +               $blocksize = $1;
> +               next
> +       }
> +       ($ext, $logical, $physical, $length) =
> +               (/^\s*(\d+):\s+(\d+)..\s+\d+:\s+(\d+)..\s+\d+:\s+(\d+):/)
> +       or next;
> +       ($flags) = /.*:\s*(\S*)$/;
> +       print $physical * $blocksize, "#",
> +             $length * $blocksize, "#",
> +             $logical * $blocksize, "#",
> +             $flags, " "'
> +
> +# this makes filefrag output script readable by using a perl helper.
> +# output is one extent per line, with three numbers separated by '#'
> +# the numbers are: physical, length, logical (all in bytes)
> +# sample output: "1234#10#5678" -> physical 1234, length 10, logical 5678
> +_filter_extents()
> +{
> +       tee -a $seqres.full | $PERL_PROG -ne "$FILEFRAG_FILTER"
> +}
> +
> +_check_file_extents()
> +{
> +       cmd="filefrag -v $1"
> +       echo "# $cmd" >> $seqres.full
> +       out=`$cmd | _filter_extents`
> +       if [ -z "$out" ]; then
> +               return 1
> +       fi
> +       echo "after filter: $out" >> $seqres.full
> +       echo $out
> +       return 0
> +}
> +
> +_check_repair()
> +{
> +       filter=${1:-cat}
> +       dmesg | tac | sed -ne "0,\#run fstests $seqnum at $date_time#p" | tac | $filter | grep -q -e "csum failed"
> +       if [ $? -eq 0 ]; then
> +               echo 1
> +       else
> +               echo 0
> +       fi
> +}
> +
> +_scratch_dev_pool_get 2
> +# step 1, create a raid1 btrfs which contains one 128k file.
> +echo "step 1......mkfs.btrfs" >>$seqres.full
> +
> +mkfs_opts="-d raid1"
> +_scratch_pool_mkfs $mkfs_opts >>$seqres.full 2>&1
> +
> +_scratch_mount -o nospace_cache

Same as the other test, why do we need to mount without space cache?
It isn't obvious if it's needed nor why - a comment in the test
explaining why would be useful for everyone.

> +
> +$XFS_IO_PROG -f -d -c "pwrite -S 0xaa -b 128K 0 128K" "$SCRATCH_MNT/foobar" | _filter_xfs_io
> +
> +sync
> +
> +# step 2, corrupt the first 64k of one copy (on SCRATCH_DEV which is the first
> +# one in $SCRATCH_DEV_POOL
> +echo "step 2......corrupt file extent" >>$seqres.full
> +
> +extents=`_check_file_extents $SCRATCH_MNT/foobar`
> +logical_in_btrfs=`echo ${extents} | cut -d '#' -f 1`
> +physical_on_scratch=`$BTRFS_MAP_LOGICAL_PROG -l ${logical_in_btrfs} $SCRATCH_DEV | grep "$SCRATCH_DEV" | awk '{print $6}'`
> +
> +_scratch_unmount
> +$XFS_IO_PROG -d -c "pwrite -S 0xbb -b 64K $physical_on_scratch 64K" $SCRATCH_DEV | _filter_xfs_io
> +
> +_scratch_mount -o nospace_cache
> +
> +# step 3, 128k dio read (this read can repair bad copy)

dio read -> buffered read

> +echo "step 3......repair the bad copy" >>$seqres.full
> +
> +# since raid1 consists of two copies, and the following read may read the good
> +# copy directly, so lets loop 10 times here and discard output that dio reads

dio reads -> buffered reads

thanks!

> +# give
> +for i in `seq 1 10`; do
> +       echo 3 > /proc/sys/vm/drop_caches
> +       $XFS_IO_PROG -c "pread -b 128K 0 128K" "$SCRATCH_MNT/foobar" > /dev/null
> +       repair=`_check_repair`
> +       if [ $repair -eq 1 ]; then
> +               break
> +       fi
> +done
> +
> +_scratch_unmount
> +
> +# check if the repair works
> +$XFS_IO_PROG -c "pread -v -b 512 $physical_on_scratch 512" $SCRATCH_DEV | _filter_xfs_io
> +
> +_scratch_dev_pool_put
> +# success, all done
> +status=0
> +exit
> diff --git a/tests/btrfs/141.out b/tests/btrfs/141.out
> new file mode 100644
> index 0000000..6b780ac
> --- /dev/null
> +++ b/tests/btrfs/141.out
> @@ -0,0 +1,39 @@
> +QA output created by 141
> +wrote 131072/131072 bytes at offset 0
> +XXX Bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
> +wrote 65536/65536 bytes at offset 244056064
> +XXX Bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
> +0e8c0000:  aa aa aa aa aa aa aa aa aa aa aa aa aa aa aa aa  ................
> +0e8c0010:  aa aa aa aa aa aa aa aa aa aa aa aa aa aa aa aa  ................
> +0e8c0020:  aa aa aa aa aa aa aa aa aa aa aa aa aa aa aa aa  ................
> +0e8c0030:  aa aa aa aa aa aa aa aa aa aa aa aa aa aa aa aa  ................
> +0e8c0040:  aa aa aa aa aa aa aa aa aa aa aa aa aa aa aa aa  ................
> +0e8c0050:  aa aa aa aa aa aa aa aa aa aa aa aa aa aa aa aa  ................
> +0e8c0060:  aa aa aa aa aa aa aa aa aa aa aa aa aa aa aa aa  ................
> +0e8c0070:  aa aa aa aa aa aa aa aa aa aa aa aa aa aa aa aa  ................
> +0e8c0080:  aa aa aa aa aa aa aa aa aa aa aa aa aa aa aa aa  ................
> +0e8c0090:  aa aa aa aa aa aa aa aa aa aa aa aa aa aa aa aa  ................
> +0e8c00a0:  aa aa aa aa aa aa aa aa aa aa aa aa aa aa aa aa  ................
> +0e8c00b0:  aa aa aa aa aa aa aa aa aa aa aa aa aa aa aa aa  ................
> +0e8c00c0:  aa aa aa aa aa aa aa aa aa aa aa aa aa aa aa aa  ................
> +0e8c00d0:  aa aa aa aa aa aa aa aa aa aa aa aa aa aa aa aa  ................
> +0e8c00e0:  aa aa aa aa aa aa aa aa aa aa aa aa aa aa aa aa  ................
> +0e8c00f0:  aa aa aa aa aa aa aa aa aa aa aa aa aa aa aa aa  ................
> +0e8c0100:  aa aa aa aa aa aa aa aa aa aa aa aa aa aa aa aa  ................
> +0e8c0110:  aa aa aa aa aa aa aa aa aa aa aa aa aa aa aa aa  ................
> +0e8c0120:  aa aa aa aa aa aa aa aa aa aa aa aa aa aa aa aa  ................
> +0e8c0130:  aa aa aa aa aa aa aa aa aa aa aa aa aa aa aa aa  ................
> +0e8c0140:  aa aa aa aa aa aa aa aa aa aa aa aa aa aa aa aa  ................
> +0e8c0150:  aa aa aa aa aa aa aa aa aa aa aa aa aa aa aa aa  ................
> +0e8c0160:  aa aa aa aa aa aa aa aa aa aa aa aa aa aa aa aa  ................
> +0e8c0170:  aa aa aa aa aa aa aa aa aa aa aa aa aa aa aa aa  ................
> +0e8c0180:  aa aa aa aa aa aa aa aa aa aa aa aa aa aa aa aa  ................
> +0e8c0190:  aa aa aa aa aa aa aa aa aa aa aa aa aa aa aa aa  ................
> +0e8c01a0:  aa aa aa aa aa aa aa aa aa aa aa aa aa aa aa aa  ................
> +0e8c01b0:  aa aa aa aa aa aa aa aa aa aa aa aa aa aa aa aa  ................
> +0e8c01c0:  aa aa aa aa aa aa aa aa aa aa aa aa aa aa aa aa  ................
> +0e8c01d0:  aa aa aa aa aa aa aa aa aa aa aa aa aa aa aa aa  ................
> +0e8c01e0:  aa aa aa aa aa aa aa aa aa aa aa aa aa aa aa aa  ................
> +0e8c01f0:  aa aa aa aa aa aa aa aa aa aa aa aa aa aa aa aa  ................
> +read 512/512 bytes at offset 244056064
> +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 1cb9c98..7b380e5 100644
> --- a/tests/btrfs/group
> +++ b/tests/btrfs/group
> @@ -142,3 +142,4 @@
>  138 auto compress
>  139 auto qgroup
>  140 auto quick
> +141 auto quick
> --
> 2.5.0
>
> --
> 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



-- 
Filipe David Manana,

“Whether you think you can, or you think you can't — you're right.”