| Message ID | 1439863433-27017-1-git-send-email-quwenruo@cn.fujitsu.com (mailing list archive) |
|---|---|
| State | Not Applicable |
| Headers | show |
On Tue, Aug 18, 2015 at 3:03 AM, Qu Wenruo <quwenruo@cn.fujitsu.com> wrote: > Btrfs qgroup reserve codes lacks check for rewrite dirty page, causing > every write, even rewriting a uncommitted dirty page, to reserve space. > > But only written data will free the reserved space, causing reserved > space leaking. > > The bug exists almost from the beginning of btrfs qgroup codes, but > nobody found it. > > For example: > > 1)Write [0, 12K) into file A > reserve 12K space > > File A: > 0 4K 8K 12K > |<--------dirty-------->| > reserved: 12K > > 2)Write [0,4K) into file A > 0 4K 8K 12K > |<--------dirty-------->| > reserved: 16K <<< Should be 12K > > 3) Commit transaction > Dirty pages [0,12) written to disk. > Free 12K reserved space. > reserved: 4K <<< Should be 0 > > This testcase will test such problem. > Kernel fix will need some huge change, so won't be soon. > > Signed-off-by: Qu Wenruo <quwenruo@cn.fujitsu.com> Reviewed-by: Filipe Manana <fdmanana@suse.com> Btw, you should update your local fstests repository. btrfs/089 already exists for quite some time: https://git.kernel.org/cgit/fs/xfs/xfstests-dev.git/commit/tests/btrfs/089?id=0e2a022a670b21769dc4cb7c9a0e24cb38c25b99 > --- > Changelog: > v2: > Use smaller write size inside loop, in case commit is trigger by dirty page > threshold, and ensure following write won't trigger EQUOT > v3: > Add more comments and fix some expression. > --- > tests/btrfs/089 | 87 +++++++++++++++++++++++++++++++++++++++++++++++++++++ > tests/btrfs/089.out | 13 ++++++++ > tests/btrfs/group | 1 + > 3 files changed, 101 insertions(+) > create mode 100755 tests/btrfs/089 > create mode 100644 tests/btrfs/089.out > > diff --git a/tests/btrfs/089 b/tests/btrfs/089 > new file mode 100755 > index 0000000..afbb1e4 > --- /dev/null > +++ b/tests/btrfs/089 > @@ -0,0 +1,87 @@ > +#! /bin/bash > +# FS QA Test 089 > +# > +# Check for qgroup reserved space leaks caused by re-writing dirty ranges > +# This bug has been present in btrfs qgroup for a long time > +# > +#----------------------------------------------------------------------- > +# Copyright (c) 2015 Fujitsu. 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 > + > +# real QA test starts here > + > +# Modify as appropriate. > +_supported_fs btrfs > +_supported_os Linux > +_require_scratch > +_need_to_be_root > + > +# Use big blocksize to ensure there is still enough space left for metadata > +# space reserve. > +BLOCKSIZE=$(( 2 * 1024 * 1024 )) # 2M block size > +FILESIZE=$(( 128 * 1024 * 1024 )) # 128M file size > + > +_scratch_mkfs >> $seqres.full 2>&1 > +_scratch_mount > +_require_fs_space $SCRATCH_MNT $(($FILESIZE * 2 / 1024)) > + > +_run_btrfs_util_prog quota enable $SCRATCH_MNT > +_run_btrfs_util_prog qgroup limit $FILESIZE 5 $SCRATCH_MNT > + > +# loop 5 times without sync to ensure reserved space leak will happen > +for i in `seq 1 5`; do > + # Use 1/4 of the file size, to ensure even commit is trigger by > + # dirty page threshold or commit interval, we should still be > + # able to continue write > + $XFS_IO_PROG -f -c "pwrite -b $BLOCKSIZE 0 $(($FILESIZE / 4))" \ > + $SCRATCH_MNT/foo | _filter_xfs_io > +done > + > +# Sync to make sure all the dirty pages are written to disk, which should > +# free all the reserved space > +sync > + > +# remove the file and sync, to ensure all quota space freed > +rm $SCRATCH_MNT/foo > +sync > + > +# We should be able to write $FILESIZE - $BLOCKSIZE data now > +$XFS_IO_PROG -f -c "pwrite -b $BLOCKSIZE 0 $(($FILESIZE - $BLOCKSIZE))" \ > + $SCRATCH_MNT/foo | _filter_xfs_io > + > +# success, all done > +status=0 > +exit > diff --git a/tests/btrfs/089.out b/tests/btrfs/089.out > new file mode 100644 > index 0000000..642bede > --- /dev/null > +++ b/tests/btrfs/089.out > @@ -0,0 +1,13 @@ > +QA output created by 089 > +wrote 33554432/33554432 bytes at offset 0 > +XXX Bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec) > +wrote 33554432/33554432 bytes at offset 0 > +XXX Bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec) > +wrote 33554432/33554432 bytes at offset 0 > +XXX Bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec) > +wrote 33554432/33554432 bytes at offset 0 > +XXX Bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec) > +wrote 33554432/33554432 bytes at offset 0 > +XXX Bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec) > +wrote 132120576/132120576 bytes at offset 0 > +XXX Bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec) > diff --git a/tests/btrfs/group b/tests/btrfs/group > index ffe18bf..da37e46 100644 > --- a/tests/btrfs/group > +++ b/tests/btrfs/group > @@ -91,6 +91,7 @@ > 086 auto quick clone > 087 auto quick send > 088 auto quick metadata > +089 auto qgroup > 090 auto quick metadata > 091 auto quick qgroup > 092 auto quick send > -- > 1.8.3.1 > > -- > 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
Filipe David Manana wrote on 2015/08/19 11:07 +0100: > On Tue, Aug 18, 2015 at 3:03 AM, Qu Wenruo <quwenruo@cn.fujitsu.com> wrote: >> Btrfs qgroup reserve codes lacks check for rewrite dirty page, causing >> every write, even rewriting a uncommitted dirty page, to reserve space. >> >> But only written data will free the reserved space, causing reserved >> space leaking. >> >> The bug exists almost from the beginning of btrfs qgroup codes, but >> nobody found it. >> >> For example: >> >> 1)Write [0, 12K) into file A >> reserve 12K space >> >> File A: >> 0 4K 8K 12K >> |<--------dirty-------->| >> reserved: 12K >> >> 2)Write [0,4K) into file A >> 0 4K 8K 12K >> |<--------dirty-------->| >> reserved: 16K <<< Should be 12K >> >> 3) Commit transaction >> Dirty pages [0,12) written to disk. >> Free 12K reserved space. >> reserved: 4K <<< Should be 0 >> >> This testcase will test such problem. >> Kernel fix will need some huge change, so won't be soon. >> >> Signed-off-by: Qu Wenruo <quwenruo@cn.fujitsu.com> > > Reviewed-by: Filipe Manana <fdmanana@suse.com> > > Btw, you should update your local fstests repository. btrfs/089 > already exists for quite some time: > https://git.kernel.org/cgit/fs/xfs/xfstests-dev.git/commit/tests/btrfs/089?id=0e2a022a670b21769dc4cb7c9a0e24cb38c25b99 > Thanks for the review and the info. I'll update it to latest version. Thanks, Qu >> --- >> Changelog: >> v2: >> Use smaller write size inside loop, in case commit is trigger by dirty page >> threshold, and ensure following write won't trigger EQUOT >> v3: >> Add more comments and fix some expression. >> --- >> tests/btrfs/089 | 87 +++++++++++++++++++++++++++++++++++++++++++++++++++++ >> tests/btrfs/089.out | 13 ++++++++ >> tests/btrfs/group | 1 + >> 3 files changed, 101 insertions(+) >> create mode 100755 tests/btrfs/089 >> create mode 100644 tests/btrfs/089.out >> >> diff --git a/tests/btrfs/089 b/tests/btrfs/089 >> new file mode 100755 >> index 0000000..afbb1e4 >> --- /dev/null >> +++ b/tests/btrfs/089 >> @@ -0,0 +1,87 @@ >> +#! /bin/bash >> +# FS QA Test 089 >> +# >> +# Check for qgroup reserved space leaks caused by re-writing dirty ranges >> +# This bug has been present in btrfs qgroup for a long time >> +# >> +#----------------------------------------------------------------------- >> +# Copyright (c) 2015 Fujitsu. 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 $tmp.* >> +} >> + >> +# get standard environment, filters and checks >> +. ./common/rc >> +. ./common/filter >> + >> +# real QA test starts here >> + >> +# Modify as appropriate. >> +_supported_fs btrfs >> +_supported_os Linux >> +_require_scratch >> +_need_to_be_root >> + >> +# Use big blocksize to ensure there is still enough space left for metadata >> +# space reserve. >> +BLOCKSIZE=$(( 2 * 1024 * 1024 )) # 2M block size >> +FILESIZE=$(( 128 * 1024 * 1024 )) # 128M file size >> + >> +_scratch_mkfs >> $seqres.full 2>&1 >> +_scratch_mount >> +_require_fs_space $SCRATCH_MNT $(($FILESIZE * 2 / 1024)) >> + >> +_run_btrfs_util_prog quota enable $SCRATCH_MNT >> +_run_btrfs_util_prog qgroup limit $FILESIZE 5 $SCRATCH_MNT >> + >> +# loop 5 times without sync to ensure reserved space leak will happen >> +for i in `seq 1 5`; do >> + # Use 1/4 of the file size, to ensure even commit is trigger by >> + # dirty page threshold or commit interval, we should still be >> + # able to continue write >> + $XFS_IO_PROG -f -c "pwrite -b $BLOCKSIZE 0 $(($FILESIZE / 4))" \ >> + $SCRATCH_MNT/foo | _filter_xfs_io >> +done >> + >> +# Sync to make sure all the dirty pages are written to disk, which should >> +# free all the reserved space >> +sync >> + >> +# remove the file and sync, to ensure all quota space freed >> +rm $SCRATCH_MNT/foo >> +sync >> + >> +# We should be able to write $FILESIZE - $BLOCKSIZE data now >> +$XFS_IO_PROG -f -c "pwrite -b $BLOCKSIZE 0 $(($FILESIZE - $BLOCKSIZE))" \ >> + $SCRATCH_MNT/foo | _filter_xfs_io >> + >> +# success, all done >> +status=0 >> +exit >> diff --git a/tests/btrfs/089.out b/tests/btrfs/089.out >> new file mode 100644 >> index 0000000..642bede >> --- /dev/null >> +++ b/tests/btrfs/089.out >> @@ -0,0 +1,13 @@ >> +QA output created by 089 >> +wrote 33554432/33554432 bytes at offset 0 >> +XXX Bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec) >> +wrote 33554432/33554432 bytes at offset 0 >> +XXX Bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec) >> +wrote 33554432/33554432 bytes at offset 0 >> +XXX Bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec) >> +wrote 33554432/33554432 bytes at offset 0 >> +XXX Bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec) >> +wrote 33554432/33554432 bytes at offset 0 >> +XXX Bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec) >> +wrote 132120576/132120576 bytes at offset 0 >> +XXX Bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec) >> diff --git a/tests/btrfs/group b/tests/btrfs/group >> index ffe18bf..da37e46 100644 >> --- a/tests/btrfs/group >> +++ b/tests/btrfs/group >> @@ -91,6 +91,7 @@ >> 086 auto quick clone >> 087 auto quick send >> 088 auto quick metadata >> +089 auto qgroup >> 090 auto quick metadata >> 091 auto quick qgroup >> 092 auto quick send >> -- >> 1.8.3.1 >> >> -- >> 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 > > > -- To unsubscribe from this list: send the line "unsubscribe linux-btrfs" 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/btrfs/089 b/tests/btrfs/089 new file mode 100755 index 0000000..afbb1e4 --- /dev/null +++ b/tests/btrfs/089 @@ -0,0 +1,87 @@ +#! /bin/bash +# FS QA Test 089 +# +# Check for qgroup reserved space leaks caused by re-writing dirty ranges +# This bug has been present in btrfs qgroup for a long time +# +#----------------------------------------------------------------------- +# Copyright (c) 2015 Fujitsu. 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 $tmp.* +} + +# get standard environment, filters and checks +. ./common/rc +. ./common/filter + +# real QA test starts here + +# Modify as appropriate. +_supported_fs btrfs +_supported_os Linux +_require_scratch +_need_to_be_root + +# Use big blocksize to ensure there is still enough space left for metadata +# space reserve. +BLOCKSIZE=$(( 2 * 1024 * 1024 )) # 2M block size +FILESIZE=$(( 128 * 1024 * 1024 )) # 128M file size + +_scratch_mkfs >> $seqres.full 2>&1 +_scratch_mount +_require_fs_space $SCRATCH_MNT $(($FILESIZE * 2 / 1024)) + +_run_btrfs_util_prog quota enable $SCRATCH_MNT +_run_btrfs_util_prog qgroup limit $FILESIZE 5 $SCRATCH_MNT + +# loop 5 times without sync to ensure reserved space leak will happen +for i in `seq 1 5`; do + # Use 1/4 of the file size, to ensure even commit is trigger by + # dirty page threshold or commit interval, we should still be + # able to continue write + $XFS_IO_PROG -f -c "pwrite -b $BLOCKSIZE 0 $(($FILESIZE / 4))" \ + $SCRATCH_MNT/foo | _filter_xfs_io +done + +# Sync to make sure all the dirty pages are written to disk, which should +# free all the reserved space +sync + +# remove the file and sync, to ensure all quota space freed +rm $SCRATCH_MNT/foo +sync + +# We should be able to write $FILESIZE - $BLOCKSIZE data now +$XFS_IO_PROG -f -c "pwrite -b $BLOCKSIZE 0 $(($FILESIZE - $BLOCKSIZE))" \ + $SCRATCH_MNT/foo | _filter_xfs_io + +# success, all done +status=0 +exit diff --git a/tests/btrfs/089.out b/tests/btrfs/089.out new file mode 100644 index 0000000..642bede --- /dev/null +++ b/tests/btrfs/089.out @@ -0,0 +1,13 @@ +QA output created by 089 +wrote 33554432/33554432 bytes at offset 0 +XXX Bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec) +wrote 33554432/33554432 bytes at offset 0 +XXX Bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec) +wrote 33554432/33554432 bytes at offset 0 +XXX Bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec) +wrote 33554432/33554432 bytes at offset 0 +XXX Bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec) +wrote 33554432/33554432 bytes at offset 0 +XXX Bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec) +wrote 132120576/132120576 bytes at offset 0 +XXX Bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec) diff --git a/tests/btrfs/group b/tests/btrfs/group index ffe18bf..da37e46 100644 --- a/tests/btrfs/group +++ b/tests/btrfs/group @@ -91,6 +91,7 @@ 086 auto quick clone 087 auto quick send 088 auto quick metadata +089 auto qgroup 090 auto quick metadata 091 auto quick qgroup 092 auto quick send
Btrfs qgroup reserve codes lacks check for rewrite dirty page, causing every write, even rewriting a uncommitted dirty page, to reserve space. But only written data will free the reserved space, causing reserved space leaking. The bug exists almost from the beginning of btrfs qgroup codes, but nobody found it. For example: 1)Write [0, 12K) into file A reserve 12K space File A: 0 4K 8K 12K |<--------dirty-------->| reserved: 12K 2)Write [0,4K) into file A 0 4K 8K 12K |<--------dirty-------->| reserved: 16K <<< Should be 12K 3) Commit transaction Dirty pages [0,12) written to disk. Free 12K reserved space. reserved: 4K <<< Should be 0 This testcase will test such problem. Kernel fix will need some huge change, so won't be soon. Signed-off-by: Qu Wenruo <quwenruo@cn.fujitsu.com> --- Changelog: v2: Use smaller write size inside loop, in case commit is trigger by dirty page threshold, and ensure following write won't trigger EQUOT v3: Add more comments and fix some expression. --- tests/btrfs/089 | 87 +++++++++++++++++++++++++++++++++++++++++++++++++++++ tests/btrfs/089.out | 13 ++++++++ tests/btrfs/group | 1 + 3 files changed, 101 insertions(+) create mode 100755 tests/btrfs/089 create mode 100644 tests/btrfs/089.out