From: Dmitry Monakhov <dmonakhov@openvz.org>
To: fstests@vger.kernel.org
Cc: linux-fsdevel@vger.kernel.org, linux-block@vger.kernel.org,
Dmitry Monakhov <dmonakhov@openvz.org>
Subject: [PATCH 4/6] new: blockdev/003 basic blockdev T10-DIF-TYPE1 integrity checks
Date: Thu, 6 Apr 2017 17:19:08 +0400 [thread overview]
Message-ID: <1491484750-9164-5-git-send-email-dmonakhov@openvz.org> (raw)
In-Reply-To: <1491484750-9164-1-git-send-email-dmonakhov@openvz.org>
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/blockdev/003 | 164 +++++++++++++++++++++++++++++++++++++++++++++++++
tests/blockdev/003.out | 39 ++++++++++++
tests/blockdev/group | 1 +
3 files changed, 204 insertions(+)
create mode 100755 tests/blockdev/003
create mode 100644 tests/blockdev/003.out
diff --git a/tests/blockdev/003 b/tests/blockdev/003
new file mode 100755
index 0000000..8f31f0d
--- /dev/null
+++ b/tests/blockdev/003
@@ -0,0 +1,164 @@
+#! /bin/bash
+# FS QA Test 003
+#
+# Check basic T10-DIF integrity features for a block device
+#
+# DIF/DIX TYPE: T10-DIF-TYPE1-CRC
+# Kernel docs: Documentation/blockdev/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/$$ || _fail "Can not make test dir"
+
+# One full test suite
+_run_one()
+{
+ local img_name=$1
+
+ echo "Run: $img_name"
+ echo "T0: Test basic IO"
+ $XFS_IO_PROG -c "pwrite -S 0xa0 -b 4M 0 16M" -d $dev >>$seqres.full 2>&1 || \
+ _fail "pwrite failed"
+
+ $XFS_IO_PROG -c "pwrite -S 0xa3 -b 1k -V8 20M 32k" -d $dev >>$seqres.full 2>&1 || \
+ _fail "pwrite failed"
+
+ $XFS_IO_PROG -c "pwrite -S 0xa2 -b 1k -V8 2M 32k" -f $dev >>$seqres.full 2>&1 || \
+ _fail "pwrite failed"
+
+ $XFS_IO_PROG -c "pwrite -S 0xa1 -b 4k 1536000 8k" -f $dev >>$seqres.full 2>&1 || \
+ _fail "pwrite failed"
+
+ $XFS_IO_PROG -c "fsync" -d $dev >>$seqres.full 2>&1 || _fail "fsync failed"
+
+ echo "Check that buffered IO"
+ dd if=$dev bs=4k 2>>$seqres.full | md5sum
+ echo "Check that direct IO"
+ # Small requests will not being splitted
+ dd if=$dev bs=64k iflag=direct 2>>$seqres.full | md5sum
+ # Big requests may be splitted on flight, but result MUST be the same.
+ dd if=$dev bs=4M iflag=direct 2>>$seqres.full | md5sum
+
+ echo "Check csum corruption detection"
+ # LIO-fileio store t10 DIF data in separate file ${IMG}.protection
+ # struct t10_pi_tuple {
+ # __be16 guard_tag; /* Checksum */
+ # __be16 app_tag; /* Opaque storage */
+ # __be32 ref_tag; /* Target LBA or indirect LBA */
+ #}
+ # Play with 3000'th sector -> t10_pi_tuple offset == 3000 * 8 == 24000
+ #
+ echo "T1: Corrupt guard_tag, next read should fail"
+ $XFS_IO_PROG -c "pwrite -S 0xde -b2 24000 2 -w" \
+ -f $TEST_DIR/$$/$img_name.protection >>$seqres.full 2>&1
+ dd if=$dev of=/dev/null bs=1M count=2 iflag=direct >>$seqres.full 2>&1 &&
+ _fail "read should fail on 3000'th sector"
+
+ echo "T2: Check that unaffected blocks are still readable"
+ dd if=$dev of=/dev/null bs=1M count=1 iflag=direct >>$seqres.full 2>&1 || _fail "read failed"
+
+ echo "T3: Rewrite corrupted sector and check that read works"
+ $XFS_IO_PROG -c "pwrite -S 0xa1 -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/$$/$img_name.protection >>$seqres.full 2>&1
+ dd if=$dev bs=2M count=1 iflag=direct >>$seqres.full 2>&1 || _fail "read failed"
+
+ echo "T5: Corrupt ref_tag, next read should fail"
+ $XFS_IO_PROG -c "pwrite -S 0xde -b4 24004 4 -w" \
+ $TEST_DIR/$$/$img_name.protection >>$seqres.full 2>&1
+ dd if=$dev of=/dev/null bs=1M count=2 iflag=direct >>$seqres.full 2>&1 &&
+ _fail "read should fail on 3000'th sector"
+
+ echo "T6: Check that unaffected blocks are still readable"
+ dd if=$dev of=/dev/null bs=1M count=1 iflag=direct >>$seqres.full 2>&1 || _fail "read failed"
+
+ echo "T7: Rewrite corrupted sector and check that read works"
+ $XFS_IO_PROG -c "pwrite -S 0xa1 -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
+ # Small requests will not being splitted
+ dd if=$dev bs=64k iflag=direct 2>>$seqres.full | md5sum
+ # Big requests may be splitted on flight, but result MUST be the same.
+ dd if=$dev bs=4M iflag=direct 2>>$seqres.full | md5sum
+}
+
+# Create virtual scsi target with internal csum verification
+name=dif-type1-w
+cfg_path=$(_liotgt_create_fileio $name $TEST_DIR/$$/$name 32M)
+
+_liotgt_set_attribute $cfg_path pi_prot_type 1
+_liotgt_set_attribute $cfg_path pi_prot_format 1
+_liotgt_set_attribute $cfg_path pi_prot_verify 1
+dev=$(_liotgt_attach_target /backstores/fileio/$name)
+_run_one $name
+
+# Create virtual scsi target w/o internal csum verification,
+# check that core blk_integrity_profile->verify_fn works
+#
+name=dif-type1-wo
+cfg_path=$(_liotgt_create_fileio $name $TEST_DIR/$$/$name 32M)
+
+_liotgt_set_attribute $cfg_path pi_prot_type 1
+_liotgt_set_attribute $cfg_path pi_prot_format 1
+_liotgt_set_attribute $cfg_path pi_prot_verify 0
+dev=$(_liotgt_attach_target /backstores/fileio/$name)
+_run_one $name
+
+status=0
+exit
diff --git a/tests/blockdev/003.out b/tests/blockdev/003.out
new file mode 100644
index 0000000..be2efbb
--- /dev/null
+++ b/tests/blockdev/003.out
@@ -0,0 +1,39 @@
+QA output created by 003
+Run: dif-type1-w
+T0: Test basic IO
+Check that buffered IO
+f18a6c9b9d69ff8ed61daa62dfab4341 -
+Check that direct IO
+f18a6c9b9d69ff8ed61daa62dfab4341 -
+f18a6c9b9d69ff8ed61daa62dfab4341 -
+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
+f18a6c9b9d69ff8ed61daa62dfab4341 -
+f18a6c9b9d69ff8ed61daa62dfab4341 -
+f18a6c9b9d69ff8ed61daa62dfab4341 -
+Run: dif-type1-wo
+T0: Test basic IO
+Check that buffered IO
+f18a6c9b9d69ff8ed61daa62dfab4341 -
+Check that direct IO
+f18a6c9b9d69ff8ed61daa62dfab4341 -
+f18a6c9b9d69ff8ed61daa62dfab4341 -
+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
+f18a6c9b9d69ff8ed61daa62dfab4341 -
+f18a6c9b9d69ff8ed61daa62dfab4341 -
+f18a6c9b9d69ff8ed61daa62dfab4341 -
diff --git a/tests/blockdev/group b/tests/blockdev/group
index 0b511f3..e5ce864 100644
--- a/tests/blockdev/group
+++ b/tests/blockdev/group
@@ -5,3 +5,4 @@
#
001 rw blockdev
002 rw blockdev liotarget
+003 rw blockdev liotarget integrity dangerous
--
2.9.3
next prev parent reply other threads:[~2017-04-06 13:19 UTC|newest]
Thread overview: 16+ messages / expand[flat|nested] mbox.gz Atom feed top
2017-04-06 13:19 [PATCH 0/6] RFC add blkdev tests v2 Dmitry Monakhov
2017-04-06 13:19 ` [PATCH 1/6] add lio-target helpers Dmitry Monakhov
2017-04-06 13:19 ` [PATCH 2/6] add: blockdev/001 check page-cache coherency after BLKDISCARD Dmitry Monakhov
2017-04-06 13:19 ` [PATCH 3/6] new: blockdev/002 check information leak for lio-fileio Dmitry Monakhov
2017-04-06 13:19 ` Dmitry Monakhov [this message]
2017-04-06 13:19 ` [PATCH 5/6] new: blockdev/004 Check that live fs survives blkdev page truncation Dmitry Monakhov
2017-04-06 13:19 ` [PATCH 6/6] new: blockdev/005 Check that busy " Dmitry Monakhov
2017-04-06 13:55 ` [PATCH 0/6] RFC add blkdev tests v2 Christoph Hellwig
2017-04-06 14:33 ` Jens Axboe
2017-04-06 14:47 ` Christoph Hellwig
2017-04-06 14:51 ` Jens Axboe
2017-04-06 15:20 ` Johannes Thumshirn
2017-04-06 15:32 ` Eric Sandeen
2017-04-06 18:15 ` Johannes Thumshirn
2017-04-06 18:19 ` Jens Axboe
2017-04-06 18:22 ` Christoph Hellwig
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=1491484750-9164-5-git-send-email-dmonakhov@openvz.org \
--to=dmonakhov@openvz.org \
--cc=fstests@vger.kernel.org \
--cc=linux-block@vger.kernel.org \
--cc=linux-fsdevel@vger.kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).