apfstests/tests/btrfs/143

#! /bin/bash
# SPDX-License-Identifier: GPL-2.0
# Copyright (c) 2017 Liu Bo.  All Rights Reserved.
#
# FS QA Test 143
#
# Regression test for btrfs buffered read's repair during read without checksum.
#
# This is to test whether buffered read retry-repair code is able to work in
# raid1 case as expected.
#
# Please note that without checksum, btrfs doesn't know if the data used to
# repair is correct, so repair is more of resync which makes sure that both
# of the copy has the same content.
#
# Commit 20a7db8ab3f2 ("btrfs: add dummy callback for readpage_io_failed and drop
# checks") introduced the regression.
#
# The upstream fix is
#	commit 9d0d1c8b1c9d ("Btrfs: bring back repair during read")
#
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_fail_make_request
_require_scratch_dev_pool 2

_require_btrfs_command inspect-internal dump-tree
_require_command "$FILEFRAG_PROG" filefrag

get_physical()
{
	local logical=$1
	local stripe=$2
	$BTRFS_UTIL_PROG inspect-internal dump-tree -t 3 $SCRATCH_DEV | \
		grep $logical -A 6 | \
		$AWK_PROG "(\$1 ~ /stripe/ && \$3 ~ /devid/ && \$2 ~ /$stripe/) { print \$6 }"
}

get_devid()
{
	local logical=$1
	local stripe=$2
	$BTRFS_UTIL_PROG inspect-internal dump-tree -t 3 $SCRATCH_DEV | \
		grep $logical -A 6 | \
		$AWK_PROG "(\$1 ~ /stripe/ && \$3 ~ /devid/ && \$2 ~ /$stripe/) { print \$4 }"
}

get_device_path()
{
	local devid=$1
	echo "$SCRATCH_DEV_POOL" | $AWK_PROG "{print \$$devid}"
}

SYSFS_BDEV=`_sysfs_dev $SCRATCH_DEV`

start_fail()
{
	local sysfs_bdev="$1"
	echo 100 > $DEBUGFS_MNT/fail_make_request/probability
	# the 1st one fails the first bio which is reading 4k (or more due to
	# readahead), and the 2nd one fails the retry of validation so that it
	# triggers read-repair
	echo 2 > $DEBUGFS_MNT/fail_make_request/times
	echo 0 > $DEBUGFS_MNT/fail_make_request/verbose
	echo 1 > $sysfs_bdev/make-it-fail
}

stop_fail()
{
	local sysfs_bdev="$1"
	echo 0 > $DEBUGFS_MNT/fail_make_request/probability
	echo 0 > $DEBUGFS_MNT/fail_make_request/times
	echo 0 > $sysfs_bdev/make-it-fail
}

_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 -b 1G"
_scratch_pool_mkfs $mkfs_opts >>$seqres.full 2>&1

# -o nospace_cache makes sure data is written to the start position of the data
# chunk
_scratch_mount -o nospace_cache,nodatasum

$XFS_IO_PROG -f -d -c "pwrite -S 0xaa -b 128K 0 128K" "$SCRATCH_MNT/foobar" |\
	_filter_xfs_io_offset

# step 2, corrupt the first 64k of stripe #1
echo "step 2......corrupt file extent" >>$seqres.full

${FILEFRAG_PROG} -v $SCRATCH_MNT/foobar >> $seqres.full
logical_in_btrfs=`${FILEFRAG_PROG} -v $SCRATCH_MNT/foobar | _filter_filefrag | cut -d '#' -f 1`
physical=`get_physical ${logical_in_btrfs} 1`
devid=$(get_devid ${logical_in_btrfs} 1)
target_dev=$(get_device_path $devid)

SYSFS_BDEV=`_sysfs_dev $target_dev`
_scratch_unmount

echo "corrupt stripe 1 devid $devid devpath $target_dev physical $physical" \
	>> $seqres.full
$XFS_IO_PROG -d -c "pwrite -S 0xbb -b 64K $physical 64K" $target_dev > /dev/null

_scratch_mount -o nospace_cache

# step 3, 128k buffered read (this read can repair bad copy)
echo "step 3......repair the bad copy" >>$seqres.full

# since raid1 consists of two copies, and the bad copy was put on stripe #1
# while the good copy lies on stripe #0, the bad copy only gets access when the
# reader's pid % 2 == 1 is true
while [[ -z ${result} ]]; do
    # invalidate the page cache.
    _scratch_cycle_mount

    start_fail $SYSFS_BDEV
    result=$(bash -c "
        if [[ \$((\$\$ % 2)) -eq 1 ]]; then
                exec $XFS_IO_PROG -c \"pread 0 4K\" \"$SCRATCH_MNT/foobar\"
        fi");
    stop_fail $SYSFS_BDEV
done

_scratch_unmount

# check if the repair works
$XFS_IO_PROG -c "pread -v -b 512 $physical 512" $target_dev |\
	_filter_xfs_io_offset

_scratch_dev_pool_put
# success, all done
status=0
exit