Filipe David Manana wrote on 2015/08/19 11:07 +0100:
On Tue, Aug 18, 2015 at 3:03 AM, Qu Wenruo <quwenruo@xxxxxxxxxxxxxx> 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@xxxxxxxxxxxxxx>
Reviewed-by: Filipe Manana <fdmanana@xxxxxxxx>
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@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
--
To unsubscribe from this list: send the line "unsubscribe fstests" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
