[PATCH 0/3] Improve block device testing coverage

[PATCH 0/3] Improve block device testing coverage

[Dmitry Monakhov]

During LSFMM we have discussed how to test lower-backend of linux IO-stack.
Common opinion was that xfstests is the most obvious solution which cover
most of use cases filesystem care about.

I'm working on integration T10-DIF/DIF data integrity features to ext4,
for that reason we need to be shure that linux integrity framework is
in working state, which is currently broken in several places.

In fact, it is relatively simple to add basic coverage tests for basic
IO operations over virtual device with integrity support. All we need
is to add lio target support.

TOC:
add lio target helpers
add test: generic/420 check information lead for lio-fileio
add test: generic/421 basic blockdev T10-DIF-TYPE1 IO

[PATCH 1/3] add lio-target helpers

[Dmitry Monakhov]

Linux-IO Target is very good framework for testing block backend.
It is more flexible than scsi_debug.

http://linux-iscsi.org/wiki/LIO
Signed-off-by: Dmitry Monakhov <dmonakhov@openvz.org>
---
 common/config    |   2 +
 common/liotarget | 111 +++++++++++++++++++++++++++++++++++++++++++++++++++++++
 2 files changed, 113 insertions(+)
 create mode 100644 common/liotarget

diff --git a/common/config b/common/config
index 59041a3..cfe7913 100644
--- a/common/config
+++ b/common/config
@@ -212,6 +212,8 @@ export XZ_PROG="`set_prog_path xz`"
 export FLOCK_PROG="`set_prog_path flock`"
 export LDD_PROG="`set_prog_path ldd`"
 export TIMEOUT_PROG="`set_prog_path timeout`"
+export TARGETCLI_PROG="`set_prog_path targetcli`"
+export TARGETCTL_PROG="`set_prog_path targetctl`"
 
 # use 'udevadm settle' or 'udevsettle' to wait for lv to be settled.
 # newer systems have udevadm command but older systems like RHEL5 don't.
diff --git a/common/liotarget b/common/liotarget
new file mode 100644
index 0000000..f821692
--- /dev/null
+++ b/common/liotarget
@@ -0,0 +1,111 @@
+#!/bin/bash
+#
+# Copyright (c) 2017 Virtuozzo Inc
+# All Rights Reserved.
+#
+# Written by Dmitry Monakhov <dmonakhov@openvz.org>
+#
+# 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
+#
+#
+# Functions for Linux-IO Target manipulation
+#
+
+_require_liotarget()
+{
+	which $TARGETCLI_PROG >>$seqres.full 2>&1  || \
+	    _notrun "this test requires 'targetcli' tool"
+	which $TARGETCLI_PROG >>$seqres.full 2>&1  || \
+	    _notrun "this test requires 'targetcli' tool"
+
+	$TARGETCLI_PROG ls /backstores/ramdisk >>$seqres.full 2>&1 ||\
+	    _notrun "kernel compiled w/o CONFIG_TARGET_CORE"
+	$TARGETCLI_PROG ls /backstores/fileio >>$seqres.full 2>&1 ||\
+	    _notrun "kernel compiled w/o CONFIG_TCM_FILEIO"
+	$TARGETCLI_PROG ls /loopback >>$seqres.full 2>&1 ||\
+	    _notrun "kernel compiled w/o CONFIG_LOOPBACK_TARGET"
+}
+
+_liotgt_create_fileio()
+{
+	local name=$1
+	local img=$2
+	local sz=$3
+
+	$TARGETCLI_PROG /backstores/fileio create \
+			name=$name file_or_dev=$img  size=$sz >>$seqres.full ||\
+	    _fail "Can not create /backstores/fileio/$name"
+
+	local cfg_path=`ls -d /sys/kernel/config/target/core/fileio_*/$name`
+	[ -d "$cfg_path" ] || _fail "Bad config path"
+	echo $cfg_path
+}
+
+_liotgt_set_attribute()
+{
+	local path=$1
+	local attr_name=$2
+	local attr_val=$3
+
+	[ -f $path/attrib/$attr_name ] || _fail "Bad attribute $attr_name"
+	echo "echo $attr_val >  $path/attrib/$attr_name " >>$seqres.full
+	echo $attr_val >  $path/attrib/$attr_name
+}
+
+_liotgt_create_loopback()
+{
+	local out=`$TARGETCLI_PROG /loopback/ create 2>&1`
+	[ $? -eq 0 ] || _fail "Can not create loopback target"
+	echo $out >>$seqres.full
+
+	local naa=`echo $out | gawk '{ print $3 }'`
+	echo ${naa:0:20} >>$seqres.full
+
+	echo ${naa:0:20}
+}
+
+_liotgt_find_dev()
+{
+	local found=""
+	local name=$1
+	local drives=`find /sys/devices -type f -name model`
+	for d in $drives;do
+		local dir=`dirname $d`
+		local vendor=`cat $dir/vendor`
+		local model=`cat $dir/model`
+		if [ "${vendor//[[:space:]]/}" == "LIO-ORG" ] && \
+		       [ "${model//[[:space:]]/}" == "$name" ]; then
+			found=/dev/$(ls $dir/block)
+			break
+		fi
+	done
+	[ -z "$found" ] && _fail "Can not find device with backend $name"
+	echo "$found"
+}
+
+_liotgt_attach_target()
+{
+	local path=$1
+	local name=`basename $path`
+	local naa=$(_liotgt_create_loopback)
+
+	$TARGETCLI_PROG /loopback/$naa/luns create $path >>$seqres.full 2>&1 ||\
+	    _fail "Can not attach $path to tcm_loop"
+	_liotgt_find_dev $name
+}
+
+_liotgt_cleanup()
+{
+	$TARGETCTL_PROG clear
+}
-- 
2.9.3

[PATCH 2/3] add test: generic/420 check information lead for lio-fileio

[Dmitry Monakhov]

Signed-off-by: Dmitry Monakhov <dmonakhov@openvz.org>
---
 tests/generic/420     | 77 +++++++++++++++++++++++++++++++++++++++++++++++++++
 tests/generic/420.out |  2 ++
 tests/generic/group   |  1 +
 3 files changed, 80 insertions(+)
 create mode 100755 tests/generic/420
 create mode 100644 tests/generic/420.out

diff --git a/tests/generic/420 b/tests/generic/420
new file mode 100755
index 0000000..592703d
--- /dev/null
+++ b/tests/generic/420
@@ -0,0 +1,77 @@
+#! /bin/bash
+# FS QA Test 420
+#
+# Regression test for  information leak for lio-fileio target
+# BUG: If image  file is less than virtual blockdev then read() may return
+#      unitilized data.
+#
+#-----------------------------------------------------------------------
+# Copyright (c) 2017 Dmitry Monakhov <dmonakhov@openvz.org>
+# 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/$$
+status=1	# failure is the default!
+trap "_cleanup; exit \$status" 0 1 2 3 15
+
+
+_cleanup()
+{
+	cd /
+	rm -f $tmp.*
+	_liotgt_cleanup
+	rm -rf $TEST_DIR/$$
+}
+
+# get standard environment, filters and checks
+. ./common/rc
+. ./common/filter
+. ./common/liotarget
+
+# remove previous $seqres.full before test
+rm -f $seqres.full
+
+# real QA test starts here
+
+# Modify as appropriate.
+_supported_fs generic
+_supported_os Linux
+_require_test
+_require_liotarget
+
+mkdir -p $TEST_DIR/$$ || _fail "Can not make test dir"
+
+cfg_path=$(_liotgt_create_fileio  420-test.img $TEST_DIR/$$/420-test.img 128M)
+dev=$(_liotgt_attach_target /backstores/fileio/420-test.img)
+
+$XFS_IO_PROG -f -c "truncate 1M" $TEST_DIR/$$/420-test.img >> $seqres.full
+
+$XFS_IO_PROG -c "pwrite -S 0xa0 -b 512 512 1k" -d $dev >>$seqres.full 2>&1 || \
+    _fail "pwrite failed"
+
+$XFS_IO_PROG -c "pwrite -S 0xab -b 512 2M 1k" -d $dev >>$seqres.full 2>&1 || \
+    _fail "pwrite failed"
+
+md5sum $dev | sed -e "s,$dev,loopdev,g"
+# success, all done
+status=0
+exit
diff --git a/tests/generic/420.out b/tests/generic/420.out
new file mode 100644
index 0000000..deed5da
--- /dev/null
+++ b/tests/generic/420.out
@@ -0,0 +1,2 @@
+QA output created by 420
+f5cfb0d8d7bfadbc2127f4f6ca4a0eba  loopdev
diff --git a/tests/generic/group b/tests/generic/group
index 0781f35..1261547 100644
--- a/tests/generic/group
+++ b/tests/generic/group
@@ -422,3 +422,4 @@
 417 auto quick shutdown log
 418 auto rw
 419 auto quick encrypt
+420 auto blockdev liotarget
-- 
2.9.3

[PATCH 3/3] add test: generic/421 basic blockdev T10-DIF-TYPE1 integrity checks

[Dmitry Monakhov]

Test create virtual block device via lio-targed infastructure and
perform basic IO operations with data corruption detection.

Temprorally mark is as dangerous, because currently it trigger BUG_ON
inside blkdev_issue_flush

BTW: I use 'dd' to test read from corrupted image instead of xfs_io
because even if pread failed, xfs_io still exit with success, BUG?

Signed-off-by: Dmitry Monakhov <dmonakhov@openvz.org>
---
 tests/generic/421     | 136 ++++++++++++++++++++++++++++++++++++++++++++++++++
 tests/generic/421.out |  16 ++++++
 tests/generic/group   |   1 +
 3 files changed, 153 insertions(+)
 create mode 100644 tests/generic/421
 create mode 100644 tests/generic/421.out

diff --git a/tests/generic/421 b/tests/generic/421
new file mode 100644
index 0000000..de1ba73
--- /dev/null
+++ b/tests/generic/421
@@ -0,0 +1,136 @@
+#! /bin/bash
+# FS QA Test 421
+#
+# Check basic T10-DIF integrity features for a block device
+#
+# DIF/DIX TYPE: T10-DIF-TYPE1-CRC
+# Kernel docs: Documentation/block/data-integrity.txt
+#-----------------------------------------------------------------------
+# Copyright (c) 2017 Dmitry Monakhov <dmonakhov@openvz.org>
+# 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/$$
+status=1	# failure is the default!
+trap "_cleanup; exit \$status" 0 1 2 3 15
+
+
+_cleanup()
+{
+	cd /
+	rm -f $tmp.*
+	_liotgt_cleanup
+	rm -rf $TEST_DIR/$$
+}
+
+# get standard environment, filters and checks
+. ./common/rc
+. ./common/filter
+. ./common/liotarget
+
+# remove previous $seqres.full before test
+rm -f $seqres.full
+
+# real QA test starts here
+
+# Modify as appropriate.
+_supported_fs generic
+_supported_os Linux
+_require_test
+_require_liotarget
+
+mkdir -p $TEST_DIR/$$ || _fail "Can not make test dir"
+
+# Create virtual block device
+cfg_path=$(_liotgt_create_fileio  t10-dif-t1 $TEST_DIR/$$/t10-dif-t1 32M)
+
+_liotgt_set_attribute $cfg_path  pi_prot_type 1
+_liotgt_set_attribute $cfg_path  pi_prot_format 1
+
+dev=$(_liotgt_attach_target /backstores/fileio/t10-dif-t1)
+
+echo "T0: Test basic IO"
+$XFS_IO_PROG -c "pwrite -S 0xa0 -b 4M 0 16M" -d $dev >>$seqres.full 2>&1 || \
+    _fail "pwrite failed"
+
+$XFS_IO_PROG -c "pwrite -S 0xa1 -b 1k -V8  20M 32k" -d $dev >>$seqres.full 2>&1 || \
+    _fail "pwrite failed"
+
+$XFS_IO_PROG -c "pwrite -S 0xa2 -b 1k -V8  2M 32k" -f $dev >>$seqres.full 2>&1 || \
+    _fail "pwrite failed"
+
+$XFS_IO_PROG -c "pwrite -S 0xa3 -b 4k  1536000 8k" -f $dev >>$seqres.full 2>&1 || \
+    _fail "pwrite failed"
+
+$XFS_IO_PROG -c "fsync" -d $dev >>$seqres.full 2>&1 || _fail "fsync failed"
+
+echo "Check that buffered and direct read works"
+dd if=$dev bs=4k  2>$seqres.full | md5sum
+dd if=$dev bs=4M iflag=direct 2>$seqres.full | md5sum
+
+echo "Check csum corruption detection"
+# LIO-fileio store t10 DIF data in separate file ${IMG}.protection
+# struct t10_pi_tuple {
+#	__be16 guard_tag;       /* Checksum */
+#       __be16 app_tag;         /* Opaque storage */
+#       __be32 ref_tag;         /* Target LBA or indirect LBA */
+#}
+# Play with 3000'th sector -> t10_pi_tuple offset == 3000 * 8 == 24000
+#
+echo "T1: Corrupt guard_tag, next read should fail"
+$XFS_IO_PROG -c "pwrite -S 0xde -b2 24000 2 -w" \
+	     -f $TEST_DIR/$$/t10-dif-t1.protection >>$seqres.full 2>&1
+dd if=$dev bs=1M count=2 iflag=direct >>$seqres.full 2>&1 &&
+    _fail "read should fail on 3000'th sector"
+
+echo "T2: Check that unaffected blocks are still readable"
+dd if=$dev bs=1M count=1 iflag=direct >>$seqres.full 2>&1 || _fail "read failed"
+
+echo "T3: Rewrite corrupted sector and check that read works"
+$XFS_IO_PROG -c "pwrite -S 0xa3 -b 4k 1536000 4k" -d $dev >>$seqres.full 2>&1 || \
+    _fail "pwrite failed"
+dd if=$dev bs=2M count=1 iflag=direct >>$seqres.full 2>&1 || _fail "read failed"
+
+echo "T4: Corrupt app_tag, should not affect read"
+$XFS_IO_PROG -c "pwrite -S 0xde -b2 24002 2 -w" \
+	     -f $TEST_DIR/$$/t10-dif-t1.protection >>$seqres.full 2>&1
+dd if=$dev bs=2M count=1 iflag=direct >>$seqres.full 2>&1 || _fail "read failed"
+
+echo "T5: Corrupt ref_tag, next read should fail"
+$XFS_IO_PROG -c "pwrite -S 0xde -b4 24004 4 -w"  \
+	     $TEST_DIR/$$/t10-dif-t1.protection  >>$seqres.full 2>&1
+dd if=$dev bs=1M count=2 iflag=direct >>$seqres.full 2>&1 &&
+    _fail "read should fail on 3000'th sector"
+
+echo "T6: Check that unaffected blocks are still readable"
+dd if=$dev bs=1M count=1 iflag=direct >>$seqres.full 2>&1 || _fail "read failed"
+
+echo "T7: Rewrite corrupted sector and check that read works"
+$XFS_IO_PROG -c "pwrite -S 0xa3 -b 4k 1536000 4k" -d $dev >>$seqres.full 2>&1 || \
+    _fail "pwrite failed"
+
+echo "Check that buffered and direct read works"
+dd if=$dev bs=4k  2>$seqres.full | md5sum
+dd if=$dev bs=4M iflag=direct 2>$seqres.full | md5sum
+# success, all done
+status=0
+exit
diff --git a/tests/generic/421.out b/tests/generic/421.out
new file mode 100644
index 0000000..087d717
--- /dev/null
+++ b/tests/generic/421.out
@@ -0,0 +1,16 @@
+QA output created by 421
+T0: Test basic IO
+Check that buffered and direct read works
+d4dacb57332e9125d56a0bad5fbfb8ff  -
+d4dacb57332e9125d56a0bad5fbfb8ff  -
+Check csum corruption detection
+T1: Corrupt guard_tag, next read should fail
+T2: Check that unaffected blocks are still readable
+T3: Rewrite corrupted sector and check that read works
+T4: Corrupt app_tag, should not affect read
+T5: Corrupt ref_tag, next read should fail
+T6: Check that unaffected blocks are still readable
+T7: Rewrite corrupted sector and check that read works
+Check that buffered and direct read works
+d4dacb57332e9125d56a0bad5fbfb8ff  -
+d4dacb57332e9125d56a0bad5fbfb8ff  -
diff --git a/tests/generic/group b/tests/generic/group
index 1261547..c3c6042 100644
--- a/tests/generic/group
+++ b/tests/generic/group
@@ -423,3 +423,4 @@
 418 auto rw
 419 auto quick encrypt
 420 auto blockdev liotarget
+421 auto blockdev dangerous liotarget integrity
-- 
2.9.3

Re: [PATCH 0/3] Improve block device testing coverage

[Christoph Hellwig]

On Thu, Mar 30, 2017 at 05:19:01PM +0400, Dmitry Monakhov wrote:
> During LSFMM we have discussed how to test lower-backend of linux IO-stack.
> Common opinion was that xfstests is the most obvious solution which cover
> most of use cases filesystem care about.
> 
> I'm working on integration T10-DIF/DIF data integrity features to ext4,
> for that reason we need to be shure that linux integrity framework is
> in working state, which is currently broken in several places.
> 
> In fact, it is relatively simple to add basic coverage tests for basic
> IO operations over virtual device with integrity support. All we need
> is to add lio target support.

First:  Thanks for adding block layer testing!

Second: even more so than Darrick's blockdev fallocate test this is
the wrong place.  If I run xfstests I want to test my file system,
not random block device features.  Please start a proper block device
testsuite instead, possibly by copy and pasting code from xfstests.

That's how I started the test suite for qemu's block layer for example.

Re: [PATCH 0/3] Improve block device testing coverage

[Dmitry Monakhov]

Christoph Hellwig <hch@infradead.org> writes:

> On Thu, Mar 30, 2017 at 05:19:01PM +0400, Dmitry Monakhov wrote:
>> During LSFMM we have discussed how to test lower-backend of linux IO-stack.
>> Common opinion was that xfstests is the most obvious solution which cover
>> most of use cases filesystem care about.
>> 
>> I'm working on integration T10-DIF/DIF data integrity features to ext4,
>> for that reason we need to be shure that linux integrity framework is
>> in working state, which is currently broken in several places.
>> 
>> In fact, it is relatively simple to add basic coverage tests for basic
>> IO operations over virtual device with integrity support. All we need
>> is to add lio target support.
>
> First:  Thanks for adding block layer testing!
>
> Second: even more so than Darrick's blockdev fallocate test this is
> the wrong place.  If I run xfstests I want to test my file system,
> not random block device features.  Please start a proper block device
> testsuite instead, possibly by copy and pasting code from xfstests.
Fair enough. I also not happy to place blkdev feature  to tests/generic
namespace. But altearnative to fork xfstests infrastructure to dedicated
test-framework only for blkdevice seems not very good. Because fork is
always pain. I already maintain one internal fork of xfstests which
tests our Vituozzo's speciffic features.

May be it would be reasonable idea to add didicated namespace
'tests/blockdev' in xfstests, and move all blkdev related tests here?
IMHO this is good idea. Because filesystem relay on some basic
features from blkdev which should be tested explicitly, because
implicit testing is too hard to debug/investigation.

>
> That's how I started the test suite for qemu's block layer for example.
Do you mean qemu/tests/qemu-iotests ?

Re: [PATCH 0/3] Improve block device testing coverage

[Eryu Guan]

On Fri, Mar 31, 2017 at 10:43:19AM +0300, Dmitry Monakhov wrote:
> Christoph Hellwig <hch@infradead.org> writes:
> 
> > On Thu, Mar 30, 2017 at 05:19:01PM +0400, Dmitry Monakhov wrote:
> >> During LSFMM we have discussed how to test lower-backend of linux IO-stack.
> >> Common opinion was that xfstests is the most obvious solution which cover
> >> most of use cases filesystem care about.
> >> 
> >> I'm working on integration T10-DIF/DIF data integrity features to ext4,
> >> for that reason we need to be shure that linux integrity framework is
> >> in working state, which is currently broken in several places.
> >> 
> >> In fact, it is relatively simple to add basic coverage tests for basic
> >> IO operations over virtual device with integrity support. All we need
> >> is to add lio target support.
> >
> > First:  Thanks for adding block layer testing!
> >
> > Second: even more so than Darrick's blockdev fallocate test this is
> > the wrong place.  If I run xfstests I want to test my file system,
> > not random block device features.  Please start a proper block device
> > testsuite instead, possibly by copy and pasting code from xfstests.
> Fair enough. I also not happy to place blkdev feature  to tests/generic
> namespace. But altearnative to fork xfstests infrastructure to dedicated
> test-framework only for blkdevice seems not very good. Because fork is
> always pain. I already maintain one internal fork of xfstests which
> tests our Vituozzo's speciffic features.
> 
> May be it would be reasonable idea to add didicated namespace
> 'tests/blockdev' in xfstests, and move all blkdev related tests here?
> IMHO this is good idea. Because filesystem relay on some basic
> features from blkdev which should be tested explicitly, because
> implicit testing is too hard to debug/investigation.

I'm not sure if xfstests is the right place for blockdev tests,
especially for the pure blockdev level features (at least Darrick's
blockdev tests are testing fallocate(2) interface).

But yeah, a new tests/blockdev dir would be good if we eventually decide
adding blockdev tests to xfstests, so they're not run by default when
people want to test filesystems.

Thanks,
Eryu

Re: [PATCH 0/3] Improve block device testing coverage

[Dmitry Monakhov]

Eryu Guan <eguan@redhat.com> writes:

> On Fri, Mar 31, 2017 at 10:43:19AM +0300, Dmitry Monakhov wrote:
>> Christoph Hellwig <hch@infradead.org> writes:
>> 
>> > On Thu, Mar 30, 2017 at 05:19:01PM +0400, Dmitry Monakhov wrote:
>> >> During LSFMM we have discussed how to test lower-backend of linux IO-stack.
>> >> Common opinion was that xfstests is the most obvious solution which cover
>> >> most of use cases filesystem care about.
>> >> 
>> >> I'm working on integration T10-DIF/DIF data integrity features to ext4,
>> >> for that reason we need to be shure that linux integrity framework is
>> >> in working state, which is currently broken in several places.
>> >> 
>> >> In fact, it is relatively simple to add basic coverage tests for basic
>> >> IO operations over virtual device with integrity support. All we need
>> >> is to add lio target support.
>> >
>> > First:  Thanks for adding block layer testing!
>> >
>> > Second: even more so than Darrick's blockdev fallocate test this is
>> > the wrong place.  If I run xfstests I want to test my file system,
>> > not random block device features.  Please start a proper block device
>> > testsuite instead, possibly by copy and pasting code from xfstests.
>> Fair enough. I also not happy to place blkdev feature  to tests/generic
>> namespace. But altearnative to fork xfstests infrastructure to dedicated
>> test-framework only for blkdevice seems not very good. Because fork is
>> always pain. I already maintain one internal fork of xfstests which
>> tests our Vituozzo's speciffic features.
>> 
>> May be it would be reasonable idea to add didicated namespace
>> 'tests/blockdev' in xfstests, and move all blkdev related tests here?
>> IMHO this is good idea. Because filesystem relay on some basic
>> features from blkdev which should be tested explicitly, because
>> implicit testing is too hard to debug/investigation.
>
> I'm not sure if xfstests is the right place for blockdev tests,
> especially for the pure blockdev level features (at least Darrick's
> blockdev tests are testing fallocate(2) interface).
Another good example may be a bug with dirty page cache after blkdiscard
https://lkml.org/lkml/2017/3/22/789 . This simple bug  result in crappy
fsimage if mkfs relay on discard_zeroes_data behaviour.
So IMHO basic blkdev test coverage is important filesystem testing. i.e.
important for xfstests.
>
> But yeah, a new tests/blockdev dir would be good if we eventually decide
> adding blockdev tests to xfstests, so they're not run by default when
> people want to test filesystems.
>
> Thanks,
> Eryu

Re: [PATCH 0/3] Improve block device testing coverage

[Bart Van Assche]

On Fri, 2017-03-31 at 13:02 +0300, Dmitry Monakhov wrote:
> Another good example may be a bug with dirty page cache after blkdiscard
> https://lkml.org/lkml/2017/3/22/789 . This simple bug  result in crappy
> fsimage if mkfs relay on discard_zeroes_data behaviour.
> So IMHO basic blkdev test coverage is important filesystem testing. i.e.
> important for xfstests.

Mixing up filesystem tests and block layer / block driver tests in the same
directory is completely wrong. Block driver developers will be primarily
interested in the block tests and may want to skip the filesystem tests.
Filesystem developers will probably run the block tests only once and will
likely run the filesystem tests repeatedly. Mixing up different kinds of
tests in the same directory makes it unnecessarily hard to run block and
filesystem tests separately.

Bart.

Re: [PATCH 0/3] Improve block device testing coverage

[Darrick J. Wong]

On Fri, Mar 31, 2017 at 03:11:28PM +0000, Bart Van Assche wrote:
> On Fri, 2017-03-31 at 13:02 +0300, Dmitry Monakhov wrote:
> > Another good example may be a bug with dirty page cache after blkdiscard
> > https://lkml.org/lkml/2017/3/22/789 . This simple bug  result in crappy
> > fsimage if mkfs relay on discard_zeroes_data behaviour.
> > So IMHO basic blkdev test coverage is important filesystem testing. i.e.
> > important for xfstests.
> 
> Mixing up filesystem tests and block layer / block driver tests in the same
> directory is completely wrong. Block driver developers will be primarily
> interested in the block tests and may want to skip the filesystem tests.
> Filesystem developers will probably run the block tests only once and will
> likely run the filesystem tests repeatedly. Mixing up different kinds of
> tests in the same directory makes it unnecessarily hard to run block and
> filesystem tests separately.

During LSF I had started to wonder if we should just create a new
FSTYP=blockdev fs type with a no-op mkfs & mount.  "_require_fs generic"
could be taught to ignore FSTYP=blockdev; blockdev tests that should
work on all block devices can stay in tests/generic, and blockdev tests
that require specific features or complicated setup can go in
tests/blockdev.

The benefit (for the fs developers, anyway) of having complex block
device setup code helper functions in common/ is that then we can also
start writing tests to see how the fs reacts with more complex storage
setups.  We already have some of that for dm_{thin,flakey,delay,error}.

That way we keep the tests together and make it easy to run them (when
applicable) as part of regular fs testing, and avoid the situation where
bdevtests and xfstests slowly drift apart in terms of behaviors and
command line switches.

The downside ofc is the potential for bloat. :)

(The blockdev fallocate tests fit the fs/block split awkwardly --
they call what is nominally a fs feature on something that isn't itself
a filesystem...)

<shrug> Just my 5c.

--D

> 
> Bart.--
> 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