btrfs: regression test for reading compressed data

We had a bug in btrfs compression code which could end up with a
kernel panic.

This is adding a regression test for the bug and I've also sent a
kernel patch to fix the bug.

The patch is "Btrfs: fix kernel oops while reading compressed data".

Signed-off-by: Liu Bo <bo.li.liu@oracle.com>
Reviewed-by: Eryu Guan <eguan@redhat.com>
Signed-off-by: Eryu Guan <eguan@redhat.com>

tests/btrfs/150

@@ -0,0 +1,105 @@
+#! /bin/bash
+# FS QA Test btrfs/150
+#
+# This is a regression test which ends up with a kernel oops in btrfs.
+# It occurs when btrfs's read repair happens while reading a compressed
+# extent.
+# The patch to fix it is
+#	Btrfs: fix kernel oops while reading compressed data
+#
+#-----------------------------------------------------------------------
+# Copyright (c) 2017 Oracle.  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
+_require_fail_make_request
+_require_scratch_dev_pool 2
+
+SYSFS_BDEV=`_sysfs_dev $SCRATCH_DEV`
+enable_io_failure()
+{
+	echo 100 > $DEBUGFS_MNT/fail_make_request/probability
+	echo 1000 > $DEBUGFS_MNT/fail_make_request/times
+	echo 0 > $DEBUGFS_MNT/fail_make_request/verbose
+	echo 1 > $DEBUGFS_MNT/fail_make_request/task-filter
+	echo 1 > $SYSFS_BDEV/make-it-fail
+}
+
+disable_io_failure()
+{
+	echo 0 > $DEBUGFS_MNT/fail_make_request/probability
+	echo 0 > $DEBUGFS_MNT/fail_make_request/times
+	echo 0 > $DEBUGFS_MNT/fail_make_request/task-filter
+	echo 0 > $SYSFS_BDEV/make-it-fail
+}
+
+_scratch_pool_mkfs "-d raid1 -b 1G" >> $seqres.full 2>&1
+
+# It doesn't matter which compression algorithm we use.
+_scratch_mount -ocompress
+
+# Create a file with all data being compressed
+$XFS_IO_PROG -f -c "pwrite -W 0 8K" $SCRATCH_MNT/foobar | _filter_xfs_io
+
+# Raid1 consists of two copies and btrfs decides which copy to read by reader's
+# %pid.  Now we inject errors to copy #1 and copy #0 is good.  We want to read
+# the bad copy to trigger read-repair.
+while [[ -z $result ]]; do
+	# invalidate the page cache
+	$XFS_IO_PROG -f -c "fadvise -d 0 8K" $SCRATCH_MNT/foobar
+
+	enable_io_failure
+
+	result=$(bash -c "
+	if [ \$((\$\$ % 2)) == 1 ]; then
+		echo 1 > /proc/\$\$/make-it-fail
+		exec $XFS_IO_PROG -c \"pread 0 8K\" \$SCRATCH_MNT/foobar
+	fi")
+
+	disable_io_failure
+done
+
+# success, all done
+status=0
+exit

tests/btrfs/150.out

@@ -0,0 +1,3 @@
+QA output created by 150
+wrote 8192/8192 bytes at offset 0
+XXX Bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)

tests/btrfs/group

@@ -152,3 +152,4 @@
 147 auto quick send
 148 auto quick rw
 149 auto quick send compress
+150 auto quick dangerous