Message ID | 20171113085350.42455-1-yuchao0@huawei.com (mailing list archive) |
---|---|
State | New, archived |
Headers | show |
On Mon, Nov 13, 2017 at 04:53:50PM +0800, Chao Yu wrote: > f2fs can skip isize updating in fsync(), since during mount, f2fs tries > to recovery isize according to valid block address or preallocated flag > in last fsynced dnode block. > > However, fallocate() breaks our rule with setting FALLOC_FL_KEEP_SIZE > flag, since it can preallocated block cross EOF, once the file is fsynced, > in POR, we will recover isize incorrectly based on these fallocated > blocks. > > This patch adds a new testcase to test fallocate, in order to verify > whether filesystem will do correct recovery on isize. > > Signed-off-by: Chao Yu <yuchao0@huawei.com> > --- > tests/generic/468 | 117 ++++++++++++++++++++++++++++++++++++++++++++++++++ > tests/generic/468.out | 13 ++++++ > tests/generic/group | 1 + > 3 files changed, 131 insertions(+) > create mode 100755 tests/generic/468 > create mode 100644 tests/generic/468.out > > diff --git a/tests/generic/468 b/tests/generic/468 > new file mode 100755 > index 000000000000..af3c5958750f > --- /dev/null > +++ b/tests/generic/468 > @@ -0,0 +1,117 @@ > +#! /bin/bash > +# FS QA Test 468 > +# > +# Do test with fallocate, f{data,}sync, power-cuts, check inode's metadata > +# after filesystem recovery. > +# In the case of fsync, filesystem should recover all the inode metadata, while > +# recovering i_blocks and i_size at least for fdatasync. Better to mention that this test is an fallocate variant of generic/392. > +# > +#----------------------------------------------------------------------- > +# Copyright (c) 2017 Chao Yu. All Rights Reserved. I didn't pay much attention to copyright statement before, but Dave pointed out similar issue in one of his reviews: You're signing off this patch a Huawei employee, but claiming personal copyright. Please clarify who owns the copyright - if it's your personal copyright then please sign off with a personal email address, not your employer's. > +# > +# 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=0 # failure is the default! status=1 means "failure is the default!" :) I noticed that you set status=1 explicitly on failure case, I don't think that's needed, just default to 'status=1' and set status=0 before exit in the end. The echos on failure case would break golden image anyway. Thanks, Eryu > +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 generic > +_supported_os Linux > +_require_test > + > +_require_scratch > +_require_scratch_shutdown > +_require_xfs_io_command "falloc" "-k" > + > +_scratch_mkfs >/dev/null 2>&1 > +_require_metadata_journaling $SCRATCH_DEV > +_scratch_mount > + > +testfile=$SCRATCH_MNT/testfile > + > +# check inode metadata after shutdown > +check_inode_metadata() > +{ > + sync_mode=$1 > + > + # fsync or fdatasync > + if [ $sync_mode = "fsync" ]; then > + stat_opt='-c "b: %b s: %s a: %x m: %y c: %z"' > + else > + stat_opt='-c "b: %b s: %s"' > + fi > + > + before=`stat "$stat_opt" $testfile` > + > + $XFS_IO_PROG -c "$sync_mode" $testfile | _filter_xfs_io > + src/godown $SCRATCH_MNT | tee -a $seqres.full > + _scratch_cycle_mount > + > + after=`stat "$stat_opt" $testfile` > + > + if [ "$before" != "$after" ]; then > + echo "Before: $before" > + echo "After : $after" > + status=1; # this is a failure! > + fi > + echo "Before: $before" >> $seqres.full > + echo "After : $after" >> $seqres.full > + rm $testfile > +} > + > +# fallocate XX KB with f{data}sync, followed by power-cut > +test_falloc() > +{ > + echo "==== falloc $2$3 test with $1 ====" | tee -a $seqres.full > + $XFS_IO_PROG -f -c "truncate 4202496" \ > + -c "pwrite 0 4202496" \ > + -c "fsync" \ > + -c "falloc $2 4202496 $3"\ > + $testfile >/dev/null > + check_inode_metadata $1 > +} > + > +for i in fsync fdatasync; do > + test_falloc $i "" 1024 > + test_falloc $i "" 4096 > + test_falloc $i "" 104857600 > + test_falloc $i "-k " 1024 > + test_falloc $i "-k " 4096 > + test_falloc $i "-k " 104857600 > +done > + > +exit > diff --git a/tests/generic/468.out b/tests/generic/468.out > new file mode 100644 > index 000000000000..b3a28d5ea07c > --- /dev/null > +++ b/tests/generic/468.out > @@ -0,0 +1,13 @@ > +QA output created by 468 > +==== falloc 1024 test with fsync ==== > +==== falloc 4096 test with fsync ==== > +==== falloc 104857600 test with fsync ==== > +==== falloc -k 1024 test with fsync ==== > +==== falloc -k 4096 test with fsync ==== > +==== falloc -k 104857600 test with fsync ==== > +==== falloc 1024 test with fdatasync ==== > +==== falloc 4096 test with fdatasync ==== > +==== falloc 104857600 test with fdatasync ==== > +==== falloc -k 1024 test with fdatasync ==== > +==== falloc -k 4096 test with fdatasync ==== > +==== falloc -k 104857600 test with fdatasync ==== > diff --git a/tests/generic/group b/tests/generic/group > index 9183950c8dfc..ea2dc04d931e 100644 > --- a/tests/generic/group > +++ b/tests/generic/group > @@ -470,3 +470,4 @@ > 465 auto rw quick aio > 466 auto quick rw > 467 auto quick exportfs > +468 shutdown auto quick metadata > -- > 2.15.0.55.gc2ece9dc4de6 > -- 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
On 2017/11/15 16:13, Eryu Guan wrote: > On Mon, Nov 13, 2017 at 04:53:50PM +0800, Chao Yu wrote: >> f2fs can skip isize updating in fsync(), since during mount, f2fs tries >> to recovery isize according to valid block address or preallocated flag >> in last fsynced dnode block. >> >> However, fallocate() breaks our rule with setting FALLOC_FL_KEEP_SIZE >> flag, since it can preallocated block cross EOF, once the file is fsynced, >> in POR, we will recover isize incorrectly based on these fallocated >> blocks. >> >> This patch adds a new testcase to test fallocate, in order to verify >> whether filesystem will do correct recovery on isize. >> >> Signed-off-by: Chao Yu <yuchao0@huawei.com> >> --- >> tests/generic/468 | 117 ++++++++++++++++++++++++++++++++++++++++++++++++++ >> tests/generic/468.out | 13 ++++++ >> tests/generic/group | 1 + >> 3 files changed, 131 insertions(+) >> create mode 100755 tests/generic/468 >> create mode 100644 tests/generic/468.out >> >> diff --git a/tests/generic/468 b/tests/generic/468 >> new file mode 100755 >> index 000000000000..af3c5958750f >> --- /dev/null >> +++ b/tests/generic/468 >> @@ -0,0 +1,117 @@ >> +#! /bin/bash >> +# FS QA Test 468 >> +# >> +# Do test with fallocate, f{data,}sync, power-cuts, check inode's metadata >> +# after filesystem recovery. >> +# In the case of fsync, filesystem should recover all the inode metadata, while >> +# recovering i_blocks and i_size at least for fdatasync. > > Better to mention that this test is an fallocate variant of generic/392. > >> +# >> +#----------------------------------------------------------------------- >> +# Copyright (c) 2017 Chao Yu. All Rights Reserved. > > I didn't pay much attention to copyright statement before, but Dave > pointed out similar issue in one of his reviews: You're signing off this > patch a Huawei employee, but claiming personal copyright. Please clarify > who owns the copyright - if it's your personal copyright then please > sign off with a personal email address, not your employer's. > >> +# >> +# 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=0 # failure is the default! > > status=1 means "failure is the default!" :) > > I noticed that you set status=1 explicitly on failure case, I don't > think that's needed, just default to 'status=1' and set status=0 before > exit in the end. The echos on failure case would break golden image > anyway. Have updated all of them, please help to check v2. Thanks, > > Thanks, > Eryu > >> +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 generic >> +_supported_os Linux >> +_require_test >> + >> +_require_scratch >> +_require_scratch_shutdown >> +_require_xfs_io_command "falloc" "-k" >> + >> +_scratch_mkfs >/dev/null 2>&1 >> +_require_metadata_journaling $SCRATCH_DEV >> +_scratch_mount >> + >> +testfile=$SCRATCH_MNT/testfile >> + >> +# check inode metadata after shutdown >> +check_inode_metadata() >> +{ >> + sync_mode=$1 >> + >> + # fsync or fdatasync >> + if [ $sync_mode = "fsync" ]; then >> + stat_opt='-c "b: %b s: %s a: %x m: %y c: %z"' >> + else >> + stat_opt='-c "b: %b s: %s"' >> + fi >> + >> + before=`stat "$stat_opt" $testfile` >> + >> + $XFS_IO_PROG -c "$sync_mode" $testfile | _filter_xfs_io >> + src/godown $SCRATCH_MNT | tee -a $seqres.full >> + _scratch_cycle_mount >> + >> + after=`stat "$stat_opt" $testfile` >> + >> + if [ "$before" != "$after" ]; then >> + echo "Before: $before" >> + echo "After : $after" >> + status=1; # this is a failure! >> + fi >> + echo "Before: $before" >> $seqres.full >> + echo "After : $after" >> $seqres.full >> + rm $testfile >> +} >> + >> +# fallocate XX KB with f{data}sync, followed by power-cut >> +test_falloc() >> +{ >> + echo "==== falloc $2$3 test with $1 ====" | tee -a $seqres.full >> + $XFS_IO_PROG -f -c "truncate 4202496" \ >> + -c "pwrite 0 4202496" \ >> + -c "fsync" \ >> + -c "falloc $2 4202496 $3"\ >> + $testfile >/dev/null >> + check_inode_metadata $1 >> +} >> + >> +for i in fsync fdatasync; do >> + test_falloc $i "" 1024 >> + test_falloc $i "" 4096 >> + test_falloc $i "" 104857600 >> + test_falloc $i "-k " 1024 >> + test_falloc $i "-k " 4096 >> + test_falloc $i "-k " 104857600 >> +done >> + >> +exit >> diff --git a/tests/generic/468.out b/tests/generic/468.out >> new file mode 100644 >> index 000000000000..b3a28d5ea07c >> --- /dev/null >> +++ b/tests/generic/468.out >> @@ -0,0 +1,13 @@ >> +QA output created by 468 >> +==== falloc 1024 test with fsync ==== >> +==== falloc 4096 test with fsync ==== >> +==== falloc 104857600 test with fsync ==== >> +==== falloc -k 1024 test with fsync ==== >> +==== falloc -k 4096 test with fsync ==== >> +==== falloc -k 104857600 test with fsync ==== >> +==== falloc 1024 test with fdatasync ==== >> +==== falloc 4096 test with fdatasync ==== >> +==== falloc 104857600 test with fdatasync ==== >> +==== falloc -k 1024 test with fdatasync ==== >> +==== falloc -k 4096 test with fdatasync ==== >> +==== falloc -k 104857600 test with fdatasync ==== >> diff --git a/tests/generic/group b/tests/generic/group >> index 9183950c8dfc..ea2dc04d931e 100644 >> --- a/tests/generic/group >> +++ b/tests/generic/group >> @@ -470,3 +470,4 @@ >> 465 auto rw quick aio >> 466 auto quick rw >> 467 auto quick exportfs >> +468 shutdown auto quick metadata >> -- >> 2.15.0.55.gc2ece9dc4de6 >> > > . > -- 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
diff --git a/tests/generic/468 b/tests/generic/468 new file mode 100755 index 000000000000..af3c5958750f --- /dev/null +++ b/tests/generic/468 @@ -0,0 +1,117 @@ +#! /bin/bash +# FS QA Test 468 +# +# Do test with fallocate, f{data,}sync, power-cuts, check inode's metadata +# after filesystem recovery. +# In the case of fsync, filesystem should recover all the inode metadata, while +# recovering i_blocks and i_size at least for fdatasync. +# +#----------------------------------------------------------------------- +# Copyright (c) 2017 Chao Yu. 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=0 # 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 generic +_supported_os Linux +_require_test + +_require_scratch +_require_scratch_shutdown +_require_xfs_io_command "falloc" "-k" + +_scratch_mkfs >/dev/null 2>&1 +_require_metadata_journaling $SCRATCH_DEV +_scratch_mount + +testfile=$SCRATCH_MNT/testfile + +# check inode metadata after shutdown +check_inode_metadata() +{ + sync_mode=$1 + + # fsync or fdatasync + if [ $sync_mode = "fsync" ]; then + stat_opt='-c "b: %b s: %s a: %x m: %y c: %z"' + else + stat_opt='-c "b: %b s: %s"' + fi + + before=`stat "$stat_opt" $testfile` + + $XFS_IO_PROG -c "$sync_mode" $testfile | _filter_xfs_io + src/godown $SCRATCH_MNT | tee -a $seqres.full + _scratch_cycle_mount + + after=`stat "$stat_opt" $testfile` + + if [ "$before" != "$after" ]; then + echo "Before: $before" + echo "After : $after" + status=1; # this is a failure! + fi + echo "Before: $before" >> $seqres.full + echo "After : $after" >> $seqres.full + rm $testfile +} + +# fallocate XX KB with f{data}sync, followed by power-cut +test_falloc() +{ + echo "==== falloc $2$3 test with $1 ====" | tee -a $seqres.full + $XFS_IO_PROG -f -c "truncate 4202496" \ + -c "pwrite 0 4202496" \ + -c "fsync" \ + -c "falloc $2 4202496 $3"\ + $testfile >/dev/null + check_inode_metadata $1 +} + +for i in fsync fdatasync; do + test_falloc $i "" 1024 + test_falloc $i "" 4096 + test_falloc $i "" 104857600 + test_falloc $i "-k " 1024 + test_falloc $i "-k " 4096 + test_falloc $i "-k " 104857600 +done + +exit diff --git a/tests/generic/468.out b/tests/generic/468.out new file mode 100644 index 000000000000..b3a28d5ea07c --- /dev/null +++ b/tests/generic/468.out @@ -0,0 +1,13 @@ +QA output created by 468 +==== falloc 1024 test with fsync ==== +==== falloc 4096 test with fsync ==== +==== falloc 104857600 test with fsync ==== +==== falloc -k 1024 test with fsync ==== +==== falloc -k 4096 test with fsync ==== +==== falloc -k 104857600 test with fsync ==== +==== falloc 1024 test with fdatasync ==== +==== falloc 4096 test with fdatasync ==== +==== falloc 104857600 test with fdatasync ==== +==== falloc -k 1024 test with fdatasync ==== +==== falloc -k 4096 test with fdatasync ==== +==== falloc -k 104857600 test with fdatasync ==== diff --git a/tests/generic/group b/tests/generic/group index 9183950c8dfc..ea2dc04d931e 100644 --- a/tests/generic/group +++ b/tests/generic/group @@ -470,3 +470,4 @@ 465 auto rw quick aio 466 auto quick rw 467 auto quick exportfs +468 shutdown auto quick metadata
f2fs can skip isize updating in fsync(), since during mount, f2fs tries to recovery isize according to valid block address or preallocated flag in last fsynced dnode block. However, fallocate() breaks our rule with setting FALLOC_FL_KEEP_SIZE flag, since it can preallocated block cross EOF, once the file is fsynced, in POR, we will recover isize incorrectly based on these fallocated blocks. This patch adds a new testcase to test fallocate, in order to verify whether filesystem will do correct recovery on isize. Signed-off-by: Chao Yu <yuchao0@huawei.com> --- tests/generic/468 | 117 ++++++++++++++++++++++++++++++++++++++++++++++++++ tests/generic/468.out | 13 ++++++ tests/generic/group | 1 + 3 files changed, 131 insertions(+) create mode 100755 tests/generic/468 create mode 100644 tests/generic/468.out