We have another case where after an fsync log replay we get an inode with
a wrong link count (smaller than it should be) and a number of directory
entries greater than its link count. This happens when we add a new link
hard link to our inode A and then we fsync some other inode B that has
the side effect of logging the parent directory inode too. In this case
at log replay time we add the new hard link to our inode (the item with
key BTRFS_INODE_REF_KEY) when processing the parent directory but we
never adjust the link count of our inode A. As a result we get stale dir
entries for our inode A that can never be deleted and therefore it makes
it impossible to remove the parent directory (as its i_size can never
decrease back to 0).
A simple reproducer for fstests that triggers this issue:
seq=`basename $0`
seqres=$RESULT_DIR/$seq
echo "QA output created by $seq"
tmp=/tmp/$$
status=1 # failure is the default!
trap "_cleanup; exit \$status" 0 1 2 3 15
_cleanup()
{
_cleanup_flakey
rm -f $tmp.*
}
# get standard environment, filters and checks
. ./common/rc
. ./common/filter
. ./common/dmflakey
# real QA test starts here
_need_to_be_root
_supported_fs generic
_supported_os Linux
_require_scratch
_require_dm_flakey
_require_metadata_journaling $SCRATCH_DEV
rm -f $seqres.full
_scratch_mkfs >>$seqres.full 2>&1
_init_flakey
_mount_flakey
# Create our test directory and files.
mkdir $SCRATCH_MNT/testdir
touch $SCRATCH_MNT/testdir/foo
touch $SCRATCH_MNT/testdir/bar
# Make sure everything done so far is durably persisted.
sync
# Create one hard link for file foo and another one for file bar. After
# that fsync only the file bar.
ln $SCRATCH_MNT/testdir/bar $SCRATCH_MNT/testdir/bar_link
ln $SCRATCH_MNT/testdir/foo $SCRATCH_MNT/testdir/foo_link
$XFS_IO_PROG -c "fsync" $SCRATCH_MNT/testdir/bar
# Silently drop all writes on scratch device to simulate power failure.
_load_flakey_table $FLAKEY_DROP_WRITES
_unmount_flakey
# Allow writes again and mount the fs to trigger log/journal replay.
_load_flakey_table $FLAKEY_ALLOW_WRITES
_mount_flakey
# Now verify both our files have a link count of 2.
echo "Link count for file foo: $(stat --format=%h $SCRATCH_MNT/testdir/foo)"
echo "Link count for file bar: $(stat --format=%h $SCRATCH_MNT/testdir/bar)"
# We should be able to remove all the links of our files in testdir, and
# after that the parent directory should become empty and therefore
# possible to remove it.
rm -f $SCRATCH_MNT/testdir/*
rmdir $SCRATCH_MNT/testdir
_unmount_flakey
# The fstests framework will call fsck against our filesystem which will verify
# that all metadata is in a consistent state.
status=0
exit
The test fails with:
-Link count for file foo: 2
+Link count for file foo: 1
Link count for file bar: 2
+rm: cannot remove '/home/fdmanana/btrfs-tests/scratch_1/testdir/foo_link': Stale file handle
+rmdir: failed to remove '/home/fdmanana/btrfs-tests/scratch_1/testdir': Directory not empty
(...)
_check_btrfs_filesystem: filesystem on /dev/sdc is inconsistent
And fsck's output:
(...)
checking fs roots
root 5 inode 258 errors 2001, no inode item, link count wrong
unresolved ref dir 257 index 5 namelen 8 name foo_link filetype 1 errors 4, no inode ref
Checking filesystem on /dev/sdc
(...)
So fix this by marking inodes for link count fixup at log replay time
whenever a directory entry is replayed if the entry was created in the
transaction where the fsync was made and if it points to a non-directory
inode.
This isn't a new problem/regression, the issue exists for a long time,
possibly since the log tree feature was added (2008).
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Signed-off-by: Chris Mason <clm@fb.com>
When we have the no_holes feature enabled, if a we truncate a file to a
smaller size, truncate it again but to a size greater than or equals to
its original size and fsync it, the log tree will not have any information
about the hole covering the range [truncate_1_offset, new_file_size[.
Which means if the fsync log is replayed, the file will remain with the
state it had before both truncate operations.
Without the no_holes feature this does not happen, since when the inode
is logged (full sync flag is set) it will find in the fs/subvol tree a
leaf with a generation matching the current transaction id that has an
explicit extent item representing the hole.
Fix this by adding an explicit extent item representing a hole between
the last extent and the inode's i_size if we are doing a full sync.
The issue is easy to reproduce with the following test case for fstests:
. ./common/rc
. ./common/filter
. ./common/dmflakey
_need_to_be_root
_supported_fs generic
_supported_os Linux
_require_scratch
_require_dm_flakey
# This test was motivated by an issue found in btrfs when the btrfs
# no-holes feature is enabled (introduced in kernel 3.14). So enable
# the feature if the fs being tested is btrfs.
if [ $FSTYP == "btrfs" ]; then
_require_btrfs_fs_feature "no_holes"
_require_btrfs_mkfs_feature "no-holes"
MKFS_OPTIONS="$MKFS_OPTIONS -O no-holes"
fi
rm -f $seqres.full
_scratch_mkfs >>$seqres.full 2>&1
_init_flakey
_mount_flakey
# Create our test files and make sure everything is durably persisted.
$XFS_IO_PROG -f -c "pwrite -S 0xaa 0 64K" \
-c "pwrite -S 0xbb 64K 61K" \
$SCRATCH_MNT/foo | _filter_xfs_io
$XFS_IO_PROG -f -c "pwrite -S 0xee 0 64K" \
-c "pwrite -S 0xff 64K 61K" \
$SCRATCH_MNT/bar | _filter_xfs_io
sync
# Now truncate our file foo to a smaller size (64Kb) and then truncate
# it to the size it had before the shrinking truncate (125Kb). Then
# fsync our file. If a power failure happens after the fsync, we expect
# our file to have a size of 125Kb, with the first 64Kb of data having
# the value 0xaa and the second 61Kb of data having the value 0x00.
$XFS_IO_PROG -c "truncate 64K" \
-c "truncate 125K" \
-c "fsync" \
$SCRATCH_MNT/foo
# Do something similar to our file bar, but the first truncation sets
# the file size to 0 and the second truncation expands the size to the
# double of what it was initially.
$XFS_IO_PROG -c "truncate 0" \
-c "truncate 253K" \
-c "fsync" \
$SCRATCH_MNT/bar
_load_flakey_table $FLAKEY_DROP_WRITES
_unmount_flakey
# Allow writes again, mount to trigger log replay and validate file
# contents.
_load_flakey_table $FLAKEY_ALLOW_WRITES
_mount_flakey
# We expect foo to have a size of 125Kb, the first 64Kb of data all
# having the value 0xaa and the remaining 61Kb to be a hole (all bytes
# with value 0x00).
echo "File foo content after log replay:"
od -t x1 $SCRATCH_MNT/foo
# We expect bar to have a size of 253Kb and no extents (any byte read
# from bar has the value 0x00).
echo "File bar content after log replay:"
od -t x1 $SCRATCH_MNT/bar
status=0
exit
The expected file contents in the golden output are:
File foo content after log replay:
0000000 aa aa aa aa aa aa aa aa aa aa aa aa aa aa aa aa
*
0200000 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
*
0372000
File bar content after log replay:
0000000 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
*
0772000
Without this fix, their contents are:
File foo content after log replay:
0000000 aa aa aa aa aa aa aa aa aa aa aa aa aa aa aa aa
*
0200000 bb bb bb bb bb bb bb bb bb bb bb bb bb bb bb bb
*
0372000
File bar content after log replay:
0000000 ee ee ee ee ee ee ee ee ee ee ee ee ee ee ee ee
*
0200000 ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff
*
0372000 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
*
0772000
A test case submission for fstests follows soon.
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Reviewed-by: Liu Bo <bo.li.liu@oracle.com>
Signed-off-by: Chris Mason <clm@fb.com>
After commit 4f764e5153 ("Btrfs: remove deleted xattrs on fsync log
replay"), we can end up in a situation where during log replay we end up
deleting xattrs that were never deleted when their file was last fsynced.
This happens in the fast fsync path (flag BTRFS_INODE_NEEDS_FULL_SYNC is
not set in the inode) if the inode has the flag BTRFS_INODE_COPY_EVERYTHING
set, the xattr was added in a past transaction and the leaf where the
xattr is located was not updated (COWed or created) in the current
transaction. In this scenario the xattr item never ends up in the log
tree and therefore at log replay time, which makes the replay code delete
the xattr from the fs/subvol tree as it thinks that xattr was deleted
prior to the last fsync.
Fix this by always logging all xattrs, which is the simplest and most
reliable way to detect deleted xattrs and replay the deletes at log replay
time.
This issue is reproducible with the following test case for fstests:
seq=`basename $0`
seqres=$RESULT_DIR/$seq
echo "QA output created by $seq"
here=`pwd`
tmp=/tmp/$$
status=1 # failure is the default!
_cleanup()
{
_cleanup_flakey
rm -f $tmp.*
}
trap "_cleanup; exit \$status" 0 1 2 3 15
# get standard environment, filters and checks
. ./common/rc
. ./common/filter
. ./common/dmflakey
. ./common/attr
# real QA test starts here
# We create a lot of xattrs for a single file. Only btrfs and xfs are currently
# able to store such a large mount of xattrs per file, other filesystems such
# as ext3/4 and f2fs for example, fail with ENOSPC even if we attempt to add
# less than 1000 xattrs with very small values.
_supported_fs btrfs xfs
_supported_os Linux
_need_to_be_root
_require_scratch
_require_dm_flakey
_require_attrs
_require_metadata_journaling $SCRATCH_DEV
rm -f $seqres.full
_scratch_mkfs >> $seqres.full 2>&1
_init_flakey
_mount_flakey
# Create the test file with some initial data and make sure everything is
# durably persisted.
$XFS_IO_PROG -f -c "pwrite -S 0xaa 0 32k" $SCRATCH_MNT/foo | _filter_xfs_io
sync
# Add many small xattrs to our file.
# We create such a large amount because it's needed to trigger the issue found
# in btrfs - we need to have an amount that causes the fs to have at least 3
# btree leafs with xattrs stored in them, and it must work on any leaf size
# (maximum leaf/node size is 64Kb).
num_xattrs=2000
for ((i = 1; i <= $num_xattrs; i++)); do
name="user.attr_$(printf "%04d" $i)"
$SETFATTR_PROG -n $name -v "val_$(printf "%04d" $i)" $SCRATCH_MNT/foo
done
# Sync the filesystem to force a commit of the current btrfs transaction, this
# is a necessary condition to trigger the bug on btrfs.
sync
# Now update our file's data and fsync the file.
# After a successful fsync, if the fsync log/journal is replayed we expect to
# see all the xattrs we added before with the same values (and the updated file
# data of course). Btrfs used to delete some of these xattrs when it replayed
# its fsync log/journal.
$XFS_IO_PROG -c "pwrite -S 0xbb 8K 16K" \
-c "fsync" \
$SCRATCH_MNT/foo | _filter_xfs_io
# Simulate a crash/power loss.
_load_flakey_table $FLAKEY_DROP_WRITES
_unmount_flakey
# Allow writes again and mount. This makes the fs replay its fsync log.
_load_flakey_table $FLAKEY_ALLOW_WRITES
_mount_flakey
echo "File content after crash and log replay:"
od -t x1 $SCRATCH_MNT/foo
echo "File xattrs after crash and log replay:"
for ((i = 1; i <= $num_xattrs; i++)); do
name="user.attr_$(printf "%04d" $i)"
echo -n "$name="
$GETFATTR_PROG --absolute-names -n $name --only-values $SCRATCH_MNT/foo
echo
done
status=0
exit
The golden output expects all xattrs to be available, and with the correct
values, after the fsync log is replayed.
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Signed-off-by: Chris Mason <clm@fb.com>
If we do an append write to a file (which increases its inode's i_size)
that does not have the flag BTRFS_INODE_NEEDS_FULL_SYNC set in its inode,
and the previous transaction added a new hard link to the file, which sets
the flag BTRFS_INODE_COPY_EVERYTHING in the file's inode, and then fsync
the file, the inode's new i_size isn't logged. This has the consequence
that after the fsync log is replayed, the file size remains what it was
before the append write operation, which means users/applications will
not be able to read the data that was successsfully fsync'ed before.
This happens because neither the inode item nor the delayed inode get
their i_size updated when the append write is made - doing so would
require starting a transaction in the buffered write path, something that
we do not do intentionally for performance reasons.
Fix this by making sure that when the flag BTRFS_INODE_COPY_EVERYTHING is
set the inode is logged with its current i_size (log the in-memory inode
into the log tree).
This issue is not a recent regression and is easy to reproduce with the
following test case for fstests:
seq=`basename $0`
seqres=$RESULT_DIR/$seq
echo "QA output created by $seq"
here=`pwd`
tmp=/tmp/$$
status=1 # failure is the default!
_cleanup()
{
_cleanup_flakey
rm -f $tmp.*
}
trap "_cleanup; exit \$status" 0 1 2 3 15
# get standard environment, filters and checks
. ./common/rc
. ./common/filter
. ./common/dmflakey
# real QA test starts here
_supported_fs generic
_supported_os Linux
_need_to_be_root
_require_scratch
_require_dm_flakey
_require_metadata_journaling $SCRATCH_DEV
_crash_and_mount()
{
# Simulate a crash/power loss.
_load_flakey_table $FLAKEY_DROP_WRITES
_unmount_flakey
# Allow writes again and mount. This makes the fs replay its fsync log.
_load_flakey_table $FLAKEY_ALLOW_WRITES
_mount_flakey
}
rm -f $seqres.full
_scratch_mkfs >> $seqres.full 2>&1
_init_flakey
_mount_flakey
# Create the test file with some initial data and then fsync it.
# The fsync here is only needed to trigger the issue in btrfs, as it causes the
# the flag BTRFS_INODE_NEEDS_FULL_SYNC to be removed from the btrfs inode.
$XFS_IO_PROG -f -c "pwrite -S 0xaa 0 32k" \
-c "fsync" \
$SCRATCH_MNT/foo | _filter_xfs_io
sync
# Add a hard link to our file.
# On btrfs this sets the flag BTRFS_INODE_COPY_EVERYTHING on the btrfs inode,
# which is a necessary condition to trigger the issue.
ln $SCRATCH_MNT/foo $SCRATCH_MNT/bar
# Sync the filesystem to force a commit of the current btrfs transaction, this
# is a necessary condition to trigger the bug on btrfs.
sync
# Now append more data to our file, increasing its size, and fsync the file.
# In btrfs because the inode flag BTRFS_INODE_COPY_EVERYTHING was set and the
# write path did not update the inode item in the btree nor the delayed inode
# item (in memory struture) in the current transaction (created by the fsync
# handler), the fsync did not record the inode's new i_size in the fsync
# log/journal. This made the data unavailable after the fsync log/journal is
# replayed.
$XFS_IO_PROG -c "pwrite -S 0xbb 32K 32K" \
-c "fsync" \
$SCRATCH_MNT/foo | _filter_xfs_io
echo "File content after fsync and before crash:"
od -t x1 $SCRATCH_MNT/foo
_crash_and_mount
echo "File content after crash and log replay:"
od -t x1 $SCRATCH_MNT/foo
status=0
exit
The expected file output before and after the crash/power failure expects the
appended data to be available, which is:
0000000 aa aa aa aa aa aa aa aa aa aa aa aa aa aa aa aa
*
0100000 bb bb bb bb bb bb bb bb bb bb bb bb bb bb bb bb
*
0200000
Cc: stable@vger.kernel.org
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Reviewed-by: Liu Bo <bo.li.liu@oracle.com>
Signed-off-by: Chris Mason <clm@fb.com>
After commit 8407f55326
("Btrfs: fix data corruption after fast fsync and writeback error"),
during wait_ordered_extents(), we wait for ordered extent setting
BTRFS_ORDERED_IO_DONE or BTRFS_ORDERED_IOERR, at which point we've
already got checksum information, so we don't need to check
(csum_bytes_left == 0) in the whole logging path.
Signed-off-by: Liu Bo <bo.li.liu@oracle.com>
Signed-off-by: Chris Mason <clm@fb.com>
Pull fourth vfs update from Al Viro:
"d_inode() annotations from David Howells (sat in for-next since before
the beginning of merge window) + four assorted fixes"
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs:
RCU pathwalk breakage when running into a symlink overmounting something
fix I_DIO_WAKEUP definition
direct-io: only inc/dec inode->i_dio_count for file systems
fs/9p: fix readdir()
VFS: assorted d_backing_inode() annotations
VFS: fs/inode.c helpers: d_inode() annotations
VFS: fs/cachefiles: d_backing_inode() annotations
VFS: fs library helpers: d_inode() annotations
VFS: assorted weird filesystems: d_inode() annotations
VFS: normal filesystems (and lustre): d_inode() annotations
VFS: security/: d_inode() annotations
VFS: security/: d_backing_inode() annotations
VFS: net/: d_inode() annotations
VFS: net/unix: d_backing_inode() annotations
VFS: kernel/: d_inode() annotations
VFS: audit: d_backing_inode() annotations
VFS: Fix up some ->d_inode accesses in the chelsio driver
VFS: Cachefiles should perform fs modifications on the top layer only
VFS: AF_UNIX sockets should call mknod on the top layer only
that's the bulk of filesystem drivers dealing with inodes of their own
Signed-off-by: David Howells <dhowells@redhat.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
When we are deleting large files with large extents, we are building up
a huge set of delayed refs for processing. Truncate isn't checking
often enough to see if we need to back off and process those, or let
a commit proceed.
The end result is long stalls after the rm, and very long commit times.
During the commits, other processes back up waiting to start new
transactions and we get into trouble.
Signed-off-by: Chris Mason <clm@fb.com>
We can get into inconsistency between inodes and directory entries
after fsyncing a directory. The issue is that while a directory gets
the new dentries persisted in the fsync log and replayed at mount time,
the link count of the inode that directory entries point to doesn't
get updated, staying with an incorrect link count (smaller then the
correct value). This later leads to stale file handle errors when
accessing (including attempt to delete) some of the links if all the
other ones are removed, which also implies impossibility to delete the
parent directories, since the dentries can not be removed.
Another issue is that (unlike ext3/4, xfs, f2fs, reiserfs, nilfs2),
when fsyncing a directory, new files aren't logged (their metadata and
dentries) nor any child directories. So this patch fixes this issue too,
since it has the same resolution as the incorrect inode link count issue
mentioned before.
This is very easy to reproduce, and the following excerpt from my test
case for xfstests shows how:
_scratch_mkfs >> $seqres.full 2>&1
_init_flakey
_mount_flakey
# Create our main test file and directory.
$XFS_IO_PROG -f -c "pwrite -S 0xaa 0 8K" $SCRATCH_MNT/foo | _filter_xfs_io
mkdir $SCRATCH_MNT/mydir
# Make sure all metadata and data are durably persisted.
sync
# Add a hard link to 'foo' inside our test directory and fsync only the
# directory. The btrfs fsync implementation had a bug that caused the new
# directory entry to be visible after the fsync log replay but, the inode
# of our file remained with a link count of 1.
ln $SCRATCH_MNT/foo $SCRATCH_MNT/mydir/foo_2
# Add a few more links and new files.
# This is just to verify nothing breaks or gives incorrect results after the
# fsync log is replayed.
ln $SCRATCH_MNT/foo $SCRATCH_MNT/mydir/foo_3
$XFS_IO_PROG -f -c "pwrite -S 0xff 0 64K" $SCRATCH_MNT/hello | _filter_xfs_io
ln $SCRATCH_MNT/hello $SCRATCH_MNT/mydir/hello_2
# Add some subdirectories and new files and links to them. This is to verify
# that after fsyncing our top level directory 'mydir', all the subdirectories
# and their files/links are registered in the fsync log and exist after the
# fsync log is replayed.
mkdir -p $SCRATCH_MNT/mydir/x/y/z
ln $SCRATCH_MNT/foo $SCRATCH_MNT/mydir/x/y/foo_y_link
ln $SCRATCH_MNT/foo $SCRATCH_MNT/mydir/x/y/z/foo_z_link
touch $SCRATCH_MNT/mydir/x/y/z/qwerty
# Now fsync only our top directory.
$XFS_IO_PROG -c "fsync" $SCRATCH_MNT/mydir
# And fsync now our new file named 'hello', just to verify later that it has
# the expected content and that the previous fsync on the directory 'mydir' had
# no bad influence on this fsync.
$XFS_IO_PROG -c "fsync" $SCRATCH_MNT/hello
# Simulate a crash/power loss.
_load_flakey_table $FLAKEY_DROP_WRITES
_unmount_flakey
_load_flakey_table $FLAKEY_ALLOW_WRITES
_mount_flakey
# Verify the content of our file 'foo' remains the same as before, 8192 bytes,
# all with the value 0xaa.
echo "File 'foo' content after log replay:"
od -t x1 $SCRATCH_MNT/foo
# Remove the first name of our inode. Because of the directory fsync bug, the
# inode's link count was 1 instead of 5, so removing the 'foo' name ended up
# deleting the inode and the other names became stale directory entries (still
# visible to applications). Attempting to remove or access the remaining
# dentries pointing to that inode resulted in stale file handle errors and
# made it impossible to remove the parent directories since it was impossible
# for them to become empty.
echo "file 'foo' link count after log replay: $(stat -c %h $SCRATCH_MNT/foo)"
rm -f $SCRATCH_MNT/foo
# Now verify that all files, links and directories created before fsyncing our
# directory exist after the fsync log was replayed.
[ -f $SCRATCH_MNT/mydir/foo_2 ] || echo "Link mydir/foo_2 is missing"
[ -f $SCRATCH_MNT/mydir/foo_3 ] || echo "Link mydir/foo_3 is missing"
[ -f $SCRATCH_MNT/hello ] || echo "File hello is missing"
[ -f $SCRATCH_MNT/mydir/hello_2 ] || echo "Link mydir/hello_2 is missing"
[ -f $SCRATCH_MNT/mydir/x/y/foo_y_link ] || \
echo "Link mydir/x/y/foo_y_link is missing"
[ -f $SCRATCH_MNT/mydir/x/y/z/foo_z_link ] || \
echo "Link mydir/x/y/z/foo_z_link is missing"
[ -f $SCRATCH_MNT/mydir/x/y/z/qwerty ] || \
echo "File mydir/x/y/z/qwerty is missing"
# We expect our file here to have a size of 64Kb and all the bytes having the
# value 0xff.
echo "file 'hello' content after log replay:"
od -t x1 $SCRATCH_MNT/hello
# Now remove all files/links, under our test directory 'mydir', and verify we
# can remove all the directories.
rm -f $SCRATCH_MNT/mydir/x/y/z/*
rmdir $SCRATCH_MNT/mydir/x/y/z
rm -f $SCRATCH_MNT/mydir/x/y/*
rmdir $SCRATCH_MNT/mydir/x/y
rmdir $SCRATCH_MNT/mydir/x
rm -f $SCRATCH_MNT/mydir/*
rmdir $SCRATCH_MNT/mydir
# An fsck, run by the fstests framework everytime a test finishes, also detected
# the inconsistency and printed the following error message:
#
# root 5 inode 257 errors 2001, no inode item, link count wrong
# unresolved ref dir 258 index 2 namelen 5 name foo_2 filetype 1 errors 4, no inode ref
# unresolved ref dir 258 index 3 namelen 5 name foo_3 filetype 1 errors 4, no inode ref
status=0
exit
The expected golden output for the test is:
wrote 8192/8192 bytes at offset 0
XXX Bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
wrote 65536/65536 bytes at offset 0
XXX Bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
File 'foo' content after log replay:
0000000 aa aa aa aa aa aa aa aa aa aa aa aa aa aa aa aa
*
0020000
file 'foo' link count after log replay: 5
file 'hello' content after log replay:
0000000 ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff
*
0200000
Which is the output after this patch and when running the test against
ext3/4, xfs, f2fs, reiserfs or nilfs2. Without this patch, the test's
output is:
wrote 8192/8192 bytes at offset 0
XXX Bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
wrote 65536/65536 bytes at offset 0
XXX Bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
File 'foo' content after log replay:
0000000 aa aa aa aa aa aa aa aa aa aa aa aa aa aa aa aa
*
0020000
file 'foo' link count after log replay: 1
Link mydir/foo_2 is missing
Link mydir/foo_3 is missing
Link mydir/x/y/foo_y_link is missing
Link mydir/x/y/z/foo_z_link is missing
File mydir/x/y/z/qwerty is missing
file 'hello' content after log replay:
0000000 ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff
*
0200000
rmdir: failed to remove '/home/fdmanana/btrfs-tests/scratch_1/mydir/x/y/z': No such file or directory
rmdir: failed to remove '/home/fdmanana/btrfs-tests/scratch_1/mydir/x/y': No such file or directory
rmdir: failed to remove '/home/fdmanana/btrfs-tests/scratch_1/mydir/x': No such file or directory
rm: cannot remove '/home/fdmanana/btrfs-tests/scratch_1/mydir/foo_2': Stale file handle
rm: cannot remove '/home/fdmanana/btrfs-tests/scratch_1/mydir/foo_3': Stale file handle
rmdir: failed to remove '/home/fdmanana/btrfs-tests/scratch_1/mydir': Directory not empty
Fsck, without this fix, also complains about the wrong link count:
root 5 inode 257 errors 2001, no inode item, link count wrong
unresolved ref dir 258 index 2 namelen 5 name foo_2 filetype 1 errors 4, no inode ref
unresolved ref dir 258 index 3 namelen 5 name foo_3 filetype 1 errors 4, no inode ref
So fix this by logging the inodes that the dentries point to when
fsyncing a directory.
A test case for xfstests follows.
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Signed-off-by: Chris Mason <clm@fb.com>
If we deleted xattrs from a file and fsynced the file, after a log replay
the xattrs would remain associated to the file. This was an unexpected
behaviour and differs from what other filesystems do, such as for example
xfs and ext3/4.
Fix this by, on fsync log replay, check if every xattr in the fs/subvol
tree (that belongs to a logged inode) has a matching xattr in the log,
and if it does not, delete it from the fs/subvol tree. This is a similar
approach to what we do for dentries when we replay a directory from the
fsync log.
This issue is trivial to reproduce, and the following excerpt from my
test for xfstests triggers the issue:
_crash_and_mount()
{
# Simulate a crash/power loss.
_load_flakey_table $FLAKEY_DROP_WRITES
_unmount_flakey
_load_flakey_table $FLAKEY_ALLOW_WRITES
_mount_flakey
}
rm -f $seqres.full
_scratch_mkfs >> $seqres.full 2>&1
_init_flakey
_mount_flakey
# Create out test file and add 3 xattrs to it.
touch $SCRATCH_MNT/foobar
$SETFATTR_PROG -n user.attr1 -v val1 $SCRATCH_MNT/foobar
$SETFATTR_PROG -n user.attr2 -v val2 $SCRATCH_MNT/foobar
$SETFATTR_PROG -n user.attr3 -v val3 $SCRATCH_MNT/foobar
# Make sure everything is durably persisted.
sync
# Now delete the second xattr and fsync the inode.
$SETFATTR_PROG -x user.attr2 $SCRATCH_MNT/foobar
$XFS_IO_PROG -c "fsync" $SCRATCH_MNT/foobar
_crash_and_mount
# After the fsync log is replayed, the file should have only 2 xattrs, the ones
# named user.attr1 and user.attr3. The btrfs fsync log replay bug left the file
# with the 3 xattrs that we had before deleting the second one and fsyncing the
# file.
echo "xattr names and values after first fsync log replay:"
$GETFATTR_PROG --absolute-names --dump $SCRATCH_MNT/foobar | _filter_scratch
# Now write some data to our file, fsync it, remove the first xattr, add a new
# hard link to our file and commit the fsync log by fsyncing some other new
# file. This is to verify that after log replay our first xattr does not exist
# anymore.
echo "hello world!" >> $SCRATCH_MNT/foobar
$XFS_IO_PROG -c "fsync" $SCRATCH_MNT/foobar
$SETFATTR_PROG -x user.attr1 $SCRATCH_MNT/foobar
ln $SCRATCH_MNT/foobar $SCRATCH_MNT/foobar_link
touch $SCRATCH_MNT/qwerty
$XFS_IO_PROG -c "fsync" $SCRATCH_MNT/qwerty
_crash_and_mount
# Now only the xattr with name user.attr3 should be set in our file.
echo "xattr names and values after second fsync log replay:"
$GETFATTR_PROG --absolute-names --dump $SCRATCH_MNT/foobar | _filter_scratch
status=0
exit
The expected golden output, which is produced with this patch applied or
when testing against xfs or ext3/4, is:
xattr names and values after first fsync log replay:
# file: SCRATCH_MNT/foobar
user.attr1="val1"
user.attr3="val3"
xattr names and values after second fsync log replay:
# file: SCRATCH_MNT/foobar
user.attr3="val3"
Without this patch applied, the output is:
xattr names and values after first fsync log replay:
# file: SCRATCH_MNT/foobar
user.attr1="val1"
user.attr2="val2"
user.attr3="val3"
xattr names and values after second fsync log replay:
# file: SCRATCH_MNT/foobar
user.attr1="val1"
user.attr2="val2"
user.attr3="val3"
A patch with a test case for xfstests follows soon.
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Signed-off-by: Chris Mason <clm@fb.com>
Pull btrfs fixes from Chris Mason:
"Outside of misc fixes, Filipe has a few fsync corners and we're
pulling in one more of Josef's fixes from production use here"
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mason/linux-btrfs:
Btrfs:__add_inode_ref: out of bounds memory read when looking for extended ref.
Btrfs: fix data loss in the fast fsync path
Btrfs: remove extra run_delayed_refs in update_cowonly_root
Btrfs: incremental send, don't rename a directory too soon
btrfs: fix lost return value due to variable shadowing
Btrfs: do not ignore errors from btrfs_lookup_xattr in do_setxattr
Btrfs: fix off-by-one logic error in btrfs_realloc_node
Btrfs: add missing inode update when punching hole
Btrfs: abort the transaction if we fail to update the free space cache inode
Btrfs: fix fsync race leading to ordered extent memory leaks
Improper arithmetics when calculting the address of the extended ref could
lead to an out of bounds memory read and kernel panic.
Signed-off-by: Quentin Casasnovas <quentin.casasnovas@oracle.com>
Reviewed-by: David Sterba <dsterba@suse.cz>
cc: stable@vger.kernel.org # v3.7+
Signed-off-by: Chris Mason <clm@fb.com>
Pull btrfs updates from Chris Mason:
"This pull is mostly cleanups and fixes:
- The raid5/6 cleanups from Zhao Lei fixup some long standing warts
in the code and add improvements on top of the scrubbing support
from 3.19.
- Josef has round one of our ENOSPC fixes coming from large btrfs
clusters here at FB.
- Dave Sterba continues a long series of cleanups (thanks Dave), and
Filipe continues hammering on corner cases in fsync and others
This all was held up a little trying to track down a use-after-free in
btrfs raid5/6. It's not clear yet if this is just made easier to
trigger with this pull or if its a new bug from the raid5/6 cleanups.
Dave Sterba is the only one to trigger it so far, but he has a
consistent way to reproduce, so we'll get it nailed shortly"
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mason/linux-btrfs: (68 commits)
Btrfs: don't remove extents and xattrs when logging new names
Btrfs: fix fsync data loss after adding hard link to inode
Btrfs: fix BUG_ON in btrfs_orphan_add() when delete unused block group
Btrfs: account for large extents with enospc
Btrfs: don't set and clear delalloc for O_DIRECT writes
Btrfs: only adjust outstanding_extents when we do a short write
btrfs: Fix out-of-space bug
Btrfs: scrub, fix sleep in atomic context
Btrfs: fix scheduler warning when syncing log
Btrfs: Remove unnecessary placeholder in btrfs_err_code
btrfs: cleanup init for list in free-space-cache
btrfs: delete chunk allocation attemp when setting block group ro
btrfs: clear bio reference after submit_one_bio()
Btrfs: fix scrub race leading to use-after-free
Btrfs: add missing cleanup on sysfs init failure
Btrfs: fix race between transaction commit and empty block group removal
btrfs: add more checks to btrfs_read_sys_array
btrfs: cleanup, rename a few variables in btrfs_read_sys_array
btrfs: add checks for sys_chunk_array sizes
btrfs: more superblock checks, lower bounds on devices and sectorsize/nodesize
...
This is the 3rd independent patch of a larger project to cleanup btrfs's
internal usage of btrfs_root. Many functions take btrfs_root only to
grab the fs_info struct.
By requiring a root these functions cause programmer overhead. That
these functions can accept any valid root is not obvious until
inspection.
This patch reduces the specificity of such functions to accept the
fs_info directly.
These patches can be applied independently and thus are not being
submitted as a patch series. There should be about 26 patches by the
project's completion. Each patch will cleanup between 1 and 34 functions
apiece. Each patch covers a single file's functions.
This patch affects the following function(s):
1) csum_tree_block
2) csum_dirty_buffer
3) check_tree_block_fsid
4) btrfs_find_tree_block
5) clean_tree_block
Signed-off-by: Daniel Dressler <danieru.dressler@gmail.com>
Signed-off-by: David Sterba <dsterba@suse.cz>
If we are recording in the tree log that an inode has new names (new hard
links were added), we would drop items, belonging to the inode, that we
shouldn't:
1) When the flag BTRFS_INODE_COPY_EVERYTHING is set in the inode's runtime
flags, we ended up dropping all the extent and xattr items that were
previously logged. This was done only in memory, since logging a new
name doesn't imply syncing the log;
2) When the flag BTRFS_INODE_COPY_EVERYTHING is set in the inode's runtime
flags, we ended up dropping all the xattr items that were previously
logged. Like the case before, this was done only in memory because
logging a new name doesn't imply syncing the log.
This led to some surprises in scenarios such as the following:
1) write some extents to an inode;
2) fsync the inode;
3) truncate the inode or delete/modify some of its xattrs
4) add a new hard link for that inode
5) fsync some other file, to force the log tree to be durably persisted
6) power failure happens
The next time the fs is mounted, the fsync log replay code is executed,
and the resulting file doesn't have the content it had when the last fsync
against it was performed, instead if has a content matching what it had
when the last transaction commit happened.
So change the behaviour such that when a new name is logged, only the inode
item and reference items are processed.
This is easy to reproduce with the test I just made for xfstests, whose
main body is:
_scratch_mkfs >> $seqres.full 2>&1
_init_flakey
_mount_flakey
# Create our test file with some data.
$XFS_IO_PROG -f -c "pwrite -S 0xaa -b 8K 0 8K" \
$SCRATCH_MNT/foo | _filter_xfs_io
# Make sure the file is durably persisted.
sync
# Append some data to our file, to increase its size.
$XFS_IO_PROG -f -c "pwrite -S 0xcc -b 4K 8K 4K" \
$SCRATCH_MNT/foo | _filter_xfs_io
# Fsync the file, so from this point on if a crash/power failure happens, our
# new data is guaranteed to be there next time the fs is mounted.
$XFS_IO_PROG -c "fsync" $SCRATCH_MNT/foo
# Now shrink our file to 5000 bytes.
$XFS_IO_PROG -c "truncate 5000" $SCRATCH_MNT/foo
# Now do an expanding truncate to a size larger than what we had when we last
# fsync'ed our file. This is just to verify that after power failure and
# replaying the fsync log, our file matches what it was when we last fsync'ed
# it - 12Kb size, first 8Kb of data had a value of 0xaa and the last 4Kb of
# data had a value of 0xcc.
$XFS_IO_PROG -c "truncate 32K" $SCRATCH_MNT/foo
# Add one hard link to our file. This made btrfs drop all of our file's
# metadata from the fsync log, including the metadata relative to the
# extent we just wrote and fsync'ed. This change was made only to the fsync
# log in memory, so adding the hard link alone doesn't change the persisted
# fsync log. This happened because the previous truncates set the runtime
# flag BTRFS_INODE_NEEDS_FULL_SYNC in the btrfs inode structure.
ln $SCRATCH_MNT/foo $SCRATCH_MNT/foo_link
# Now make sure the in memory fsync log is durably persisted.
# Creating and fsync'ing another file will do it.
# After this our persisted fsync log will no longer have metadata for our file
# foo that points to the extent we wrote and fsync'ed before.
touch $SCRATCH_MNT/bar
$XFS_IO_PROG -c "fsync" $SCRATCH_MNT/bar
# As expected, before the crash/power failure, we should be able to see a file
# with a size of 32Kb, with its first 5000 bytes having the value 0xaa and all
# the remaining bytes with value 0x00.
echo "File content before:"
od -t x1 $SCRATCH_MNT/foo
# Simulate a crash/power loss.
_load_flakey_table $FLAKEY_DROP_WRITES
_unmount_flakey
_load_flakey_table $FLAKEY_ALLOW_WRITES
_mount_flakey
# After mounting the fs again, the fsync log was replayed.
# The expected result is to see a file with a size of 12Kb, with its first 8Kb
# of data having the value 0xaa and its last 4Kb of data having a value of 0xcc.
# The btrfs bug used to leave the file as it used te be as of the last
# transaction commit - that is, with a size of 8Kb with all bytes having a
# value of 0xaa.
echo "File content after:"
od -t x1 $SCRATCH_MNT/foo
The test case for xfstests follows soon.
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Signed-off-by: Chris Mason <clm@fb.com>
We have a scenario where after the fsync log replay we can lose file data
that had been previously fsync'ed if we added an hard link for our inode
and after that we sync'ed the fsync log (for example by fsync'ing some
other file or directory).
This is because when adding an hard link we updated the inode item in the
log tree with an i_size value of 0. At that point the new inode item was
in memory only and a subsequent fsync log replay would not make us lose
the file data. However if after adding the hard link we sync the log tree
to disk, by fsync'ing some other file or directory for example, we ended
up losing the file data after log replay, because the inode item in the
persisted log tree had an an i_size of zero.
This is easy to reproduce, and the following excerpt from my test for
xfstests shows this:
_scratch_mkfs >> $seqres.full 2>&1
_init_flakey
_mount_flakey
# Create one file with data and fsync it.
# This made the btrfs fsync log persist the data and the inode metadata with
# a correct inode->i_size (4096 bytes).
$XFS_IO_PROG -f -c "pwrite -S 0xaa -b 4K 0 4K" -c "fsync" \
$SCRATCH_MNT/foo | _filter_xfs_io
# Now add one hard link to our file. This made the btrfs code update the fsync
# log, in memory only, with an inode metadata having a size of 0.
ln $SCRATCH_MNT/foo $SCRATCH_MNT/foo_link
# Now force persistence of the fsync log to disk, for example, by fsyncing some
# other file.
touch $SCRATCH_MNT/bar
$XFS_IO_PROG -c "fsync" $SCRATCH_MNT/bar
# Before a power loss or crash, we could read the 4Kb of data from our file as
# expected.
echo "File content before:"
od -t x1 $SCRATCH_MNT/foo
# Simulate a crash/power loss.
_load_flakey_table $FLAKEY_DROP_WRITES
_unmount_flakey
_load_flakey_table $FLAKEY_ALLOW_WRITES
_mount_flakey
# After the fsync log replay, because the fsync log had a value of 0 for our
# inode's i_size, we couldn't read anymore the 4Kb of data that we previously
# wrote and fsync'ed. The size of the file became 0 after the fsync log replay.
echo "File content after:"
od -t x1 $SCRATCH_MNT/foo
Another alternative test, that doesn't need to fsync an inode in the same
transaction it was created, is:
_scratch_mkfs >> $seqres.full 2>&1
_init_flakey
_mount_flakey
# Create our test file with some data.
$XFS_IO_PROG -f -c "pwrite -S 0xaa -b 8K 0 8K" \
$SCRATCH_MNT/foo | _filter_xfs_io
# Make sure the file is durably persisted.
sync
# Append some data to our file, to increase its size.
$XFS_IO_PROG -f -c "pwrite -S 0xcc -b 4K 8K 4K" \
$SCRATCH_MNT/foo | _filter_xfs_io
# Fsync the file, so from this point on if a crash/power failure happens, our
# new data is guaranteed to be there next time the fs is mounted.
$XFS_IO_PROG -c "fsync" $SCRATCH_MNT/foo
# Add one hard link to our file. This made btrfs write into the in memory fsync
# log a special inode with generation 0 and an i_size of 0 too. Note that this
# didn't update the inode in the fsync log on disk.
ln $SCRATCH_MNT/foo $SCRATCH_MNT/foo_link
# Now make sure the in memory fsync log is durably persisted.
# Creating and fsync'ing another file will do it.
touch $SCRATCH_MNT/bar
$XFS_IO_PROG -c "fsync" $SCRATCH_MNT/bar
# As expected, before the crash/power failure, we should be able to read the
# 12Kb of file data.
echo "File content before:"
od -t x1 $SCRATCH_MNT/foo
# Simulate a crash/power loss.
_load_flakey_table $FLAKEY_DROP_WRITES
_unmount_flakey
_load_flakey_table $FLAKEY_ALLOW_WRITES
_mount_flakey
# After mounting the fs again, the fsync log was replayed.
# The btrfs fsync log replay code didn't update the i_size of the persisted
# inode because the inode item in the log had a special generation with a
# value of 0 (and it couldn't know the correct i_size, since that inode item
# had a 0 i_size too). This made the last 4Kb of file data inaccessible and
# effectively lost.
echo "File content after:"
od -t x1 $SCRATCH_MNT/foo
This isn't a new issue/regression. This problem has been around since the
log tree code was added in 2008:
Btrfs: Add a write ahead tree log to optimize synchronous operations
(commit e02119d5a7)
Test cases for xfstests follow soon.
CC: <stable@vger.kernel.org>
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Signed-off-by: Chris Mason <clm@fb.com>
Add missing blk_finish_plug in btrfs_sync_log()
Signed-off-by: Forrest Liu <forrestl@synology.com>
Reviewed-by: David Sterba <dsterba@suse.cz>
Signed-off-by: Chris Mason <clm@fb.com>
If we have an inode with a large number of hard links, some of which may
be extrefs, turn a regular ref into an extref, fsync the inode and then
replay the fsync log (after a crash/reboot), we can endup with an fsync
log that makes the replay code always fail with -EOVERFLOW when processing
the inode's references.
This is easy to reproduce with the test case I made for xfstests. Its steps
are the following:
_scratch_mkfs "-O extref" >> $seqres.full 2>&1
_init_flakey
_mount_flakey
# Create a test file with 3001 hard links. This number is large enough to
# make btrfs start using extrefs at some point even if the fs has the maximum
# possible leaf/node size (64Kb).
echo "hello world" > $SCRATCH_MNT/foo
for i in `seq 1 3000`; do
ln $SCRATCH_MNT/foo $SCRATCH_MNT/foo_link_`printf "%04d" $i`
done
# Make sure all metadata and data are durably persisted.
sync
# Now remove one link, add a new one with a new name, add another new one with
# the same name as the one we just removed and fsync the inode.
rm -f $SCRATCH_MNT/foo_link_0001
ln $SCRATCH_MNT/foo $SCRATCH_MNT/foo_link_3001
ln $SCRATCH_MNT/foo $SCRATCH_MNT/foo_link_0001
rm -f $SCRATCH_MNT/foo_link_0002
ln $SCRATCH_MNT/foo $SCRATCH_MNT/foo_link_3002
ln $SCRATCH_MNT/foo $SCRATCH_MNT/foo_link_3003
$XFS_IO_PROG -c "fsync" $SCRATCH_MNT/foo
# Simulate a crash/power loss. This makes sure the next mount
# will see an fsync log and will replay that log.
_load_flakey_table $FLAKEY_DROP_WRITES
_unmount_flakey
_load_flakey_table $FLAKEY_ALLOW_WRITES
_mount_flakey
# Check that the number of hard links is correct, we are able to remove all
# the hard links and read the file's data. This is just to verify we don't
# get stale file handle errors (due to dangling directory index entries that
# point to inodes that no longer exist).
echo "Link count: $(stat --format=%h $SCRATCH_MNT/foo)"
[ -f $SCRATCH_MNT/foo ] || echo "Link foo is missing"
for ((i = 1; i <= 3003; i++)); do
name=foo_link_`printf "%04d" $i`
if [ $i -eq 2 ]; then
[ -f $SCRATCH_MNT/$name ] && echo "Link $name found"
else
[ -f $SCRATCH_MNT/$name ] || echo "Link $name is missing"
fi
done
rm -f $SCRATCH_MNT/foo_link_*
cat $SCRATCH_MNT/foo
rm -f $SCRATCH_MNT/foo
status=0
exit
The fix is simply to correct the overflow condition when overwriting a
reference item because it was wrong, trying to increase the item in the
fs/subvol tree by an impossible amount. Also ensure that we don't insert
one normal ref and one ext ref for the same dentry - this happened because
processing a dir index entry from the parent in the log happened when
the normal ref item was full, which made the logic insert an extref and
later when the normal ref had enough room, it would be inserted again
when processing the ref item from the child inode in the log.
This issue has been present since the introduction of the extrefs feature
(2012).
A test case for xfstests follows soon. This test only passes if the previous
patch titled "Btrfs: fix fsync when extend references are added to an inode"
is applied too.
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Signed-off-by: Chris Mason <clm@fb.com>
If we added an extended reference to an inode and fsync'ed it, the log
replay code would make our inode have an incorrect link count, which
was lower then the expected/correct count.
This resulted in stale directory index entries after deleting some of
the hard links, and any access to the dangling directory entries resulted
in -ESTALE errors because the entries pointed to inode items that don't
exist anymore.
This is easy to reproduce with the test case I made for xfstests, and
the bulk of that test is:
_scratch_mkfs "-O extref" >> $seqres.full 2>&1
_init_flakey
_mount_flakey
# Create a test file with 3001 hard links. This number is large enough to
# make btrfs start using extrefs at some point even if the fs has the maximum
# possible leaf/node size (64Kb).
echo "hello world" > $SCRATCH_MNT/foo
for i in `seq 1 3000`; do
ln $SCRATCH_MNT/foo $SCRATCH_MNT/foo_link_`printf "%04d" $i`
done
# Make sure all metadata and data are durably persisted.
sync
# Add one more link to the inode that ends up being a btrfs extref and fsync
# the inode.
ln $SCRATCH_MNT/foo $SCRATCH_MNT/foo_link_3001
$XFS_IO_PROG -c "fsync" $SCRATCH_MNT/foo
# Simulate a crash/power loss. This makes sure the next mount
# will see an fsync log and will replay that log.
_load_flakey_table $FLAKEY_DROP_WRITES
_unmount_flakey
_load_flakey_table $FLAKEY_ALLOW_WRITES
_mount_flakey
# Now after the fsync log replay btrfs left our inode with a wrong link count N,
# which was smaller than the correct link count M (N < M).
# So after removing N hard links, the remaining M - N directory entries were
# still visible to user space but it was impossible to do anything with them
# because they pointed to an inode that didn't exist anymore. This resulted in
# stale file handle errors (-ESTALE) when accessing those dentries for example.
#
# So remove all hard links except the first one and then attempt to read the
# file, to verify we don't get an -ESTALE error when accessing the inodel
#
# The btrfs fsck tool also detected the incorrect inode link count and it
# reported an error message like the following:
#
# root 5 inode 257 errors 2001, no inode item, link count wrong
# unresolved ref dir 256 index 2978 namelen 13 name foo_link_2976 filetype 1 errors 4, no inode ref
#
# The fstests framework automatically calls fsck after a test is run, so we
# don't need to call fsck explicitly here.
rm -f $SCRATCH_MNT/foo_link_*
cat $SCRATCH_MNT/foo
status=0
exit
So make sure an fsync always flushes the delayed inode item, so that the
fsync log contains it (needed in order to trigger the link count fixup
code) and fix the extref counting function, which always return -ENOENT
to its caller (and made it assume there were always 0 extrefs).
This issue has been present since the introduction of the extrefs feature
(2012).
A test case for xfstests follows soon.
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Signed-off-by: Chris Mason <clm@fb.com>
If we have an inode (file) with a link count greater than 1, remove
one of its hard links, fsync the inode, power fail/crash and then
replay the fsync log on the next mount, we end up getting the parent
directory's metadata inconsistent - its i_size still reflects the
deleted hard link and has dangling index entries (with no matching
inode reference entries). This prevents the directory from ever being
deletable, as its i_size can never decrease to BTRFS_EMPTY_DIR_SIZE
even if all of its children inodes are deleted, and the dangling index
entries can never be removed (as they point to an inode that does not
exist anymore).
This is easy to reproduce with the following excerpt from the test case
for xfstests that I just made:
_scratch_mkfs >> $seqres.full 2>&1
_init_flakey
_mount_flakey
# Create a test file with 2 hard links in the same directory.
mkdir -p $SCRATCH_MNT/a/b
echo "hello world" > $SCRATCH_MNT/a/b/foo
ln $SCRATCH_MNT/a/b/foo $SCRATCH_MNT/a/b/bar
# Make sure all metadata and data are durably persisted.
sync
# Now remove one of the hard links and fsync the inode.
rm -f $SCRATCH_MNT/a/b/bar
$XFS_IO_PROG -c "fsync" $SCRATCH_MNT/a/b/foo
# Simulate a crash/power loss. This makes sure the next mount
# will see an fsync log and will replay that log.
_load_flakey_table $FLAKEY_DROP_WRITES
_unmount_flakey
_load_flakey_table $FLAKEY_ALLOW_WRITES
_mount_flakey
# Remove the last hard link of the file and attempt to remove its parent
# directory - this failed in btrfs because the fsync log and replay code
# didn't decrement the parent directory's i_size and left dangling directory
# index entries - this made the btrfs rmdir implementation always fail with
# the error -ENOTEMPTY.
#
# The dangling directory index entries were visible to user space, but it was
# impossible to do anything on them (unlink, open, read, write, stat, etc)
# because the inode they pointed to did not exist anymore.
#
# The parent directory's metadata inconsistency (stale index entries) was
# also detected by btrfs' fsck tool, which is run automatically by the fstests
# framework when the test finishes. The error message reported by fsck was:
#
# root 5 inode 259 errors 2001, no inode item, link count wrong
# unresolved ref dir 258 index 3 namelen 3 name bar filetype 1 errors 4, no inode ref
#
rm -f $SCRATCH_MNT/a/b/*
rmdir $SCRATCH_MNT/a/b
rmdir $SCRATCH_MNT/a
To fix this just make sure that after an unlink, if the inode is fsync'ed,
he parent inode is fully logged in the fsync log.
A test case for xfstests follows soon.
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Signed-off-by: Chris Mason <clm@fb.com>
We can search and add the orphan item in one go,
btrfs_insert_orphan_item will find out if the item already exists.
Signed-off-by: David Sterba <dsterba@suse.cz>
If btrfs_find_item is called with NULL path it allocates one locally but
does not free it. Affected paths are inserting an orphan item for a file
and for a subvol root.
Move the path allocation to the callers.
CC: <stable@vger.kernel.org> # 3.14+
Fixes: 3f870c2899 ("btrfs: expand btrfs_find_item() to include find_orphan_item functionality")
Signed-off-by: David Sterba <dsterba@suse.cz>
Finally it's clear that the requested blocksize is always equal to
nodesize, with one exception, the superblock.
Superblock has fixed size regardless of the metadata block size, but
uses the same helpers to initialize sys array/chunk tree and to work
with the chunk items. So it pretends to be an extent_buffer for a
moment, btrfs_read_sys_array is full of special cases, we're adding one
more.
Signed-off-by: David Sterba <dsterba@suse.cz>
When doing a fsync with a fast path we have a time window where we can miss
the fact that writeback of some file data failed, and therefore we endup
returning success (0) from fsync when we should return an error.
The steps that lead to this are the following:
1) We start all ordered extents by calling filemap_fdatawrite_range();
2) We do some other work like locking the inode's i_mutex, start a transaction,
start a log transaction, etc;
3) We enter btrfs_log_inode(), acquire the inode's log_mutex and collect all the
ordered extents from inode's ordered tree into a list;
4) But by the time we do ordered extent collection, some ordered extents we started
at step 1) might have already completed with an error, and therefore we didn't
found them in the ordered tree and had no idea they finished with an error. This
makes our fsync return success (0) to userspace, but has no bad effects on the log
like for example insertion of file extent items into the log that point to unwritten
extents, because the invalid extent maps were removed before the ordered extent
completed (in inode.c:btrfs_finish_ordered_io).
So after collecting the ordered extents just check if the inode's i_mapping has any
error flags set (AS_EIO or AS_ENOSPC) and leave with an error if it does. Whenever
writeback fails for a page of an ordered extent, we call mapping_set_error (done in
extent_io.c:end_extent_writepage, called by extent_io.c:end_bio_extent_writepage)
that sets one of those error flags in the inode's i_mapping flags.
This change also has the side effect of fixing the issue where for fast fsyncs we
never checked/cleared the error flags from the inode's i_mapping flags, which means
that a full fsync performed after a fast fsync could get such errors that belonged
to the fast fsync - because the full fsync calls btrfs_wait_ordered_range() which
calls filemap_fdatawait_range(), and the later checks for and clears those flags,
while for fast fsyncs we never call filemap_fdatawait_range() or anything else
that checks for and clears the error flags from the inode's i_mapping.
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Signed-off-by: Chris Mason <clm@fb.com>
Instead of collecting all ordered extents from the inode's ordered tree
and then wait for all of them to complete, just collect the ones that
overlap the fsync range.
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Signed-off-by: Chris Mason <clm@fb.com>
If an error happens during writeback of log btree extents, make sure the
error is returned to the caller (fsync), so that it takes proper action
(commit current transaction) instead of writing a superblock that points
to log btrees with all or some nodes that weren't durably persisted.
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Signed-off-by: Chris Mason <clm@fb.com>
Liu Bo pointed out that my previous fix would lose the generation update in the
scenario I described. It is actually much worse than that, we could lose the
entire extent if we lose power right after the transaction commits. Consider
the following
write extent 0-4k
log extent in log tree
commit transaction
< power fail happens here
ordered extent completes
We would lose the 0-4k extent because it hasn't updated the actual fs tree, and
the transaction commit will reset the log so it isn't replayed. If we lose
power before the transaction commit we are save, otherwise we are not.
Fix this by keeping track of all extents we logged in this transaction. Then
when we go to commit the transaction make sure we wait for all of those ordered
extents to complete before proceeding. This will make sure that if we lose
power after the transaction commit we still have our data. This also fixes the
problem of the improperly updated extent generation. Thanks,
cc: stable@vger.kernel.org
Signed-off-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
If we have two fsync()'s race on different subvols one will do all of its work
to get into the log_tree, wait on it's outstanding IO, and then allow the
log_tree to finish it's commit. The problem is we were just free'ing that
subvols logged extents instead of waiting on them, so whoever lost the race
wouldn't really have their data on disk. Fix this by waiting properly instead
of freeing the logged extents. Thanks,
cc: stable@vger.kernel.org
Signed-off-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
Pull btrfs fixes from Chris Mason:
"Filipe is nailing down some problems with our skinny extent variation,
and Dave's patch fixes endian problems in the new super block checks"
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mason/linux-btrfs:
Btrfs: fix race that makes btrfs_lookup_extent_info miss skinny extent items
Btrfs: properly clean up btrfs_end_io_wq_cache
Btrfs: fix invalid leaf slot access in btrfs_lookup_extent()
btrfs: use macro accessors in superblock validation checks
If we couldn't find our extent item, we accessed the current slot
(path->slots[0]) to check if it corresponds to an equivalent skinny
metadata item. However this slot could be beyond our last item in the
leaf (i.e. path->slots[0] >= btrfs_header_nritems(leaf)), in which case
we shouldn't process it.
Since btrfs_lookup_extent() is only used to find extent items for data
extents, fix this by removing completely the logic that looks up for an
equivalent skinny metadata item, since it can not exist.
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Signed-off-by: Chris Mason <clm@fb.com>
Pull btrfs updates from Chris Mason:
"The largest set of changes here come from Miao Xie. He's cleaning up
and improving read recovery/repair for raid, and has a number of
related fixes.
I've merged another set of fsync fixes from Filipe, and he's also
improved the way we handle metadata write errors to make sure we force
the FS readonly if things go wrong.
Otherwise we have a collection of fixes and cleanups. Dave Sterba
gets a cookie for removing the most lines (thanks Dave)"
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mason/linux-btrfs: (139 commits)
btrfs: Fix compile error when CONFIG_SECURITY is not set.
Btrfs: fix compiles when CONFIG_BTRFS_FS_RUN_SANITY_TESTS is off
btrfs: Make btrfs handle security mount options internally to avoid losing security label.
Btrfs: send, don't delay dir move if there's a new parent inode
btrfs: add more superblock checks
Btrfs: fix race in WAIT_SYNC ioctl
Btrfs: be aware of btree inode write errors to avoid fs corruption
Btrfs: remove redundant btrfs_verify_qgroup_counts declaration.
btrfs: fix shadow warning on cmp
Btrfs: fix compilation errors under DEBUG
Btrfs: fix crash of btrfs_release_extent_buffer_page
Btrfs: add missing end_page_writeback on submit_extent_page failure
btrfs: Fix the wrong condition judgment about subset extent map
Btrfs: fix build_backref_tree issue with multiple shared blocks
Btrfs: cleanup error handling in build_backref_tree
btrfs: move checks for DUMMY_ROOT into a helper
btrfs: new define for the inline extent data start
btrfs: kill extent_buffer_page helper
btrfs: drop constant param from btrfs_release_extent_buffer_page
btrfs: hide typecast to definition of BTRFS_SEND_TRANS_STUB
...
When we do a fast fsync, we start all ordered operations and then while
they're running in parallel we visit the list of modified extent maps
and construct their matching file extent items and write them to the
log btree. After that, in btrfs_sync_log() we wait for all the ordered
operations to finish (via btrfs_wait_logged_extents).
The problem with this is that we were completely ignoring errors that
can happen in the extent write path, such as -ENOSPC, a temporary -ENOMEM
or -EIO errors for example. When such error happens, it means we have parts
of the on disk extent that weren't written to, and so we end up logging
file extent items that point to these extents that contain garbage/random
data - so after a crash/reboot plus log replay, we get our inode's metadata
pointing to those extents.
This worked in contrast with the full (non-fast) fsync path, where we
start all ordered operations, wait for them to finish and then write
to the log btree. In this path, after each ordered operation completes
we check if it's flagged with an error (BTRFS_ORDERED_IOERR) and return
-EIO if so (via btrfs_wait_ordered_range).
So if an error happens with any ordered operation, just return a -EIO
error to userspace, so that it knows that not all of its previous writes
were durably persisted and the application can take proper action (like
redo the writes for e.g.) - and definitely not leave any file extent items
in the log refer to non fully written extents.
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Signed-off-by: Chris Mason <clm@fb.com>
When replaying a directory from the fsync log, if a directory entry
exists both in the fs/subvol tree and in the log, the directory's inode
got its i_size updated incorrectly, accounting for the dentry's name
twice.
Reproducer, from a test for xfstests:
_scratch_mkfs >> $seqres.full 2>&1
_init_flakey
_mount_flakey
touch $SCRATCH_MNT/foo
sync
touch $SCRATCH_MNT/bar
xfs_io -c "fsync" $SCRATCH_MNT
xfs_io -c "fsync" $SCRATCH_MNT/bar
_load_flakey_table $FLAKEY_DROP_WRITES
_unmount_flakey
_load_flakey_table $FLAKEY_ALLOW_WRITES
_mount_flakey
[ -f $SCRATCH_MNT/foo ] || echo "file foo is missing"
[ -f $SCRATCH_MNT/bar ] || echo "file bar is missing"
_unmount_flakey
_check_scratch_fs $FLAKEY_DEV
The filesystem check at the end failed with the message:
"root 5 root dir 256 error".
A test case for xfstests follows.
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Signed-off-by: Chris Mason <clm@fb.com>
None of the uses of btrfs_search_forward() need to have the path
nodes (level >= 1) read locked, only the leaf needs to be locked
while the caller processes it. Therefore make it return a path
with all nodes unlocked, except for the leaf.
This change is motivated by the observation that during a file
fsync we repeatdly call btrfs_search_forward() and process the
returned leaf while upper nodes of the returned path (level >= 1)
are read locked, which unnecessarily blocks other tasks that want
to write to the same fs/subvol btree.
Therefore instead of modifying the fsync code to unlock all nodes
with level >= 1 immediately after calling btrfs_search_forward(),
change btrfs_search_forward() to do it, so that it benefits all
callers.
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Signed-off-by: Chris Mason <clm@fb.com>
The nodesize and leafsize were never of different values. Unify the
usage and make nodesize the one. Cleanup the redundant checks and
helpers.
Shaves a few bytes from .text:
text data bss dec hex filename
852418 24560 23112 900090 dbbfa btrfs.ko.before
851074 24584 23112 898770 db6d2 btrfs.ko.after
Signed-off-by: David Sterba <dsterba@suse.cz>
Signed-off-by: Chris Mason <clm@fb.com>
btrfs_set_key_type and btrfs_key_type are used inconsistently along with
open coded variants. Other members of btrfs_key are accessed directly
without any helpers anyway.
Signed-off-by: David Sterba <dsterba@suse.cz>
Signed-off-by: Chris Mason <clm@fb.com>
When a ranged fsync finishes if there are still extent maps in the modified
list, still set the inode's logged_trans and last_log_commit. This is important
in case an inode is fsync'ed and unlinked in the same transaction, to ensure its
inode ref gets deleted from the log and the respective dentries in its parent
are deleted too from the log (if the parent directory was fsync'ed in the same
transaction).
Instead make btrfs_inode_in_log() return false if the list of modified extent
maps isn't empty.
This is an incremental on top of the v4 version of the patch:
"Btrfs: fix fsync data loss after a ranged fsync"
which was added to its v5, but didn't make it on time.
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Signed-off-by: Chris Mason <clm@fb.com>
While we're doing a full fsync (when the inode has the flag
BTRFS_INODE_NEEDS_FULL_SYNC set) that is ranged too (covers only a
portion of the file), we might have ordered operations that are started
before or while we're logging the inode and that fall outside the fsync
range.
Therefore when a full ranged fsync finishes don't remove every extent
map from the list of modified extent maps - as for some of them, that
fall outside our fsync range, their respective ordered operation hasn't
finished yet, meaning the corresponding file extent item wasn't inserted
into the fs/subvol tree yet and therefore we didn't log it, and we must
let the next fast fsync (one that checks only the modified list) see this
extent map and log a matching file extent item to the log btree and wait
for its ordered operation to finish (if it's still ongoing).
A test case for xfstests follows.
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Signed-off-by: Chris Mason <clm@fb.com>
The file hole detection logic during a file fsync wasn't correct,
because it didn't look back (in a previous leaf) for the last file
extent item that can be in a leaf to the left of our leaf and that
has a generation lower than the current transaction id. This made it
assume that a hole exists when it really doesn't exist in the file.
Such false positive hole detection happens in the following scenario:
* We have a file that has many file extent items, covering 3 or more
btree leafs (the first leaf must contain non file extent items too).
* Two ranges of the file are modified, with their extent items being
located at 2 different leafs and those leafs aren't consecutive.
* When processing the second modified leaf, we weren't checking if
some file extent item exists that is located in some leaf that is
between our 2 modified leafs, and therefore assumed the range defined
between the last file extent item in the first leaf and the first file
extent item in the second leaf matched a hole.
Fortunately this didn't result in overriding the log with wrong data,
instead it made the last loop in copy_items() attempt to insert a
duplicated key (for a hole file extent item), which makes the file
fsync code return with -EEXIST to file.c:btrfs_sync_file() which in
turn ends up doing a full transaction commit, which is much more
expensive then writing only to the log tree and wait for it to be
durably persisted (as well as the file's modified extents/pages).
Therefore fix the hole detection logic, so that we don't pay the
cost of doing full transaction commits.
I could trigger this issue with the following test for xfstests (which
never fails, either without or with this patch). The last fsync call
results in a full transaction commit, due to the -EEXIST error mentioned
above. I could also observe this behaviour happening frequently when
running xfstests/generic/075 in a loop.
Test:
_cleanup()
{
_cleanup_flakey
rm -fr $tmp
}
# get standard environment, filters and checks
. ./common/rc
. ./common/filter
. ./common/dmflakey
# real QA test starts here
_supported_fs btrfs
_supported_os Linux
_require_scratch
_require_dm_flakey
_need_to_be_root
rm -f $seqres.full
# Create a file with many file extent items, each representing a 4Kb extent.
# These items span 3 btree leaves, of 16Kb each (default mkfs.btrfs leaf size
# as of btrfs-progs 3.12).
_scratch_mkfs -l 16384 >/dev/null 2>&1
_init_flakey
SAVE_MOUNT_OPTIONS="$MOUNT_OPTIONS"
MOUNT_OPTIONS="$MOUNT_OPTIONS -o commit=999"
_mount_flakey
# First fsync, inode has BTRFS_INODE_NEEDS_FULL_SYNC flag set.
$XFS_IO_PROG -f -c "pwrite -S 0x01 -b 4096 0 4096" -c "fsync" \
$SCRATCH_MNT/foo | _filter_xfs_io
# For any of the following fsync calls, inode doesn't have the flag
# BTRFS_INODE_NEEDS_FULL_SYNC set.
for ((i = 1; i <= 500; i++)); do
OFFSET=$((4096 * i))
LEN=4096
$XFS_IO_PROG -c "pwrite -S 0x01 $OFFSET $LEN" -c "fsync" \
$SCRATCH_MNT/foo | _filter_xfs_io
done
# Commit transaction and bump next transaction's id (to 7).
sync
# Truncate will set the BTRFS_INODE_NEEDS_FULL_SYNC flag in the btrfs's
# inode runtime flags.
$XFS_IO_PROG -c "truncate 2048000" $SCRATCH_MNT/foo
# Commit transaction and bump next transaction's id (to 8).
sync
# Touch 1 extent item from the first leaf and 1 from the last leaf. The leaf
# in the middle, containing only file extent items, isn't touched. So the
# next fsync, when calling btrfs_search_forward(), won't visit that middle
# leaf. First and 3rd leaf have now a generation with value 8, while the
# middle leaf remains with a generation with value 6.
$XFS_IO_PROG \
-c "pwrite -S 0xee -b 4096 0 4096" \
-c "pwrite -S 0xff -b 4096 2043904 4096" \
-c "fsync" \
$SCRATCH_MNT/foo | _filter_xfs_io
_load_flakey_table $FLAKEY_DROP_WRITES
md5sum $SCRATCH_MNT/foo | _filter_scratch
_unmount_flakey
_load_flakey_table $FLAKEY_ALLOW_WRITES
# During mount, we'll replay the log created by the fsync above, and the file's
# md5 digest should be the same we got before the unmount.
_mount_flakey
md5sum $SCRATCH_MNT/foo | _filter_scratch
_unmount_flakey
MOUNT_OPTIONS="$SAVE_MOUNT_OPTIONS"
status=0
exit
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Signed-off-by: Chris Mason <clm@fb.com>
If the log sync fails, there is something wrong in the log tree, we
should not continue to join the log transaction and log the metadata.
What we should do is to do a full commit.
This patch fixes this problem by setting ->last_trans_log_full_commit
to the current transaction id, it will tell the tasks not to join
the log transaction, and do a full commit.
Signed-off-by: Miao Xie <miaox@cn.fujitsu.com>
Signed-off-by: Josef Bacik <jbacik@fb.com>
We might commit the log sub-transaction which didn't contain the metadata we
logged. It was because we didn't record the log transid and just select
the current log sub-transaction to commit, but the right one might be
committed by the other task already. Actually, we needn't do anything
and it is safe that we go back directly in this case.
This patch improves the log sync by the above idea. We record the transid
of the log sub-transaction in which we log the metadata, and the transid
of the log sub-transaction we have committed. If the committed transid
is >= the transid we record when logging the metadata, we just go back.
Signed-off-by: Miao Xie <miaox@cn.fujitsu.com>
Signed-off-by: Josef Bacik <jbacik@fb.com>
It is possible that many tasks sync the log tree at the same time, but
only one task can do the sync work, the others will wait for it. But those
wait tasks didn't get the result of the log sync, and returned 0 when they
ended the wait. It caused those tasks skipped the error handle, and the
serious problem was they told the users the file sync succeeded but in
fact they failed.
This patch fixes this problem by introducing a log context structure,
we insert it into the a global list. When the sync fails, we will set
the error number of every log context in the list, then the waiting tasks
get the error number of the log context and handle the error if need.
Signed-off-by: Miao Xie <miaox@cn.fujitsu.com>
Signed-off-by: Josef Bacik <jbacik@fb.com>
The log trans id is initialized to be 0 every time we create a log tree,
and the log tree need be re-created after a new transaction is started,
it means the log trans id is unlikely to be a huge number, so we can use
signed integer instead of unsigned long integer to save a bit space.
Signed-off-by: Miao Xie <miaox@cn.fujitsu.com>
Signed-off-by: Josef Bacik <jbacik@fb.com>
Mutex unlock implies certain memory barriers to make sure all the memory
operation completes before the unlock, and the next mutex lock implies memory
barriers to make sure the all the memory happens after the lock. So it is
a full memory barrier(smp_mb), we needn't add memory barriers. Remove them.
Signed-off-by: Miao Xie <miaox@cn.fujitsu.com>
Signed-off-by: Josef Bacik <jbacik@fb.com>
The old code would start the log transaction even the log tree init
failed, it was unnecessary. Fix it.
Signed-off-by: Miao Xie <miaox@cn.fujitsu.com>
Signed-off-by: Josef Bacik <jbacik@fb.com>
We may abort the wait earlier if ->last_trans_log_full_commit was set to
the current transaction id, at this case, we need commit the current
transaction instead of the log sub-transaction. But the current code
didn't tell the caller to do it (return 0, not -EAGAIN). Fix it.
Signed-off-by: Miao Xie <miaox@cn.fujitsu.com>
Signed-off-by: Josef Bacik <jbacik@fb.com>
->last_trans_log_full_commit may be changed by the other tasks without lock,
so we need prevent the compiler from the optimize access just like
tmp = fs_info->last_trans_log_full_commit
if (tmp == ...)
...
<do something>
if (tmp == ...)
...
In fact, we need get the new value of ->last_trans_log_full_commit during
the second access. Fix it by ACCESS_ONCE().
Signed-off-by: Miao Xie <miaox@cn.fujitsu.com>
Signed-off-by: Josef Bacik <jbacik@fb.com>
There was a problem in the old code:
If we failed to log the csum, we would free all the ordered extents in the log list
including those ordered extents that were logged successfully, it would make the
log committer not to wait for the completion of the ordered extents.
This patch doesn't insert the ordered extents that is about to be logged into
a global list, instead, we insert them into a local list. If we log the ordered
extents successfully, we splice them with the global list, or we will throw them
away, then do full sync. It can also reduce the lock contention and the traverse
time of list.
Signed-off-by: Miao Xie <miaox@cn.fujitsu.com>
Signed-off-by: Josef Bacik <jbacik@fb.com>
If we truncate an uncompressed inline item, ram_bytes isn't updated to reflect
the new size. The fixe uses the size directly from the item header when
reading uncompressed inlines, and also fixes truncate to update the
size as it goes.
Reported-by: Jens Axboe <axboe@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
CC: stable@vger.kernel.org
The performance of fsync dropped down suddenly sometimes, the main reason
of this problem was that we might only flush part dirty pages in a ordered
extent, then got that ordered extent, wait for the csum calcucation. But if
no task flushed the left part, we would wait until the flusher flushed them,
sometimes we need wait for several seconds, it made the performance drop
down suddenly. (On my box, it drop down from 56MB/s to 4-10MB/s)
This patch improves the above problem by flushing left dirty pages aggressively.
Test Environment:
CPU: 2CPU * 2Cores
Memory: 4GB
Partition: 20GB(HDD)
Test Command:
# sysbench --num-threads=8 --test=fileio --file-num=1 \
> --file-total-size=8G --file-block-size=32768 \
> --file-io-mode=sync --file-fsync-freq=100 \
> --file-fsync-end=no --max-requests=10000 \
> --file-test-mode=rndwr run
Signed-off-by: Miao Xie <miaox@cn.fujitsu.com>
Signed-off-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
When writing to a file we drop existing file extent items that cover the
write range and then add a new file extent item that represents that write
range.
Before this change we were doing a tree lookup to remove the file extent
items, and then after we did another tree lookup to insert the new file
extent item.
Most of the time all the file extent items we need to drop are located
within a single leaf - this is the leaf where our new file extent item ends
up at. Therefore, in this common case just combine these 2 operations into
a single one.
By avoiding the second btree navigation for insertion of the new file extent
item, we reduce btree node/leaf lock acquisitions/releases, btree block/leaf
COW operations, CPU time on btree node/leaf key binary searches, etc.
Besides for file writes, this is an operation that happens for file fsync's
as well. However log btrees are much less likely to big as big as regular
fs btrees, therefore the impact of this change is smaller.
The following benchmark was performed against an SSD drive and a
HDD drive, both for random and sequential writes:
sysbench --test=fileio --file-num=4096 --file-total-size=8G \
--file-test-mode=[rndwr|seqwr] --num-threads=512 \
--file-block-size=8192 \ --max-requests=1000000 \
--file-fsync-freq=0 --file-io-mode=sync [prepare|run]
All results below are averages of 10 runs of the respective test.
** SSD sequential writes
Before this change: 225.88 Mb/sec
After this change: 277.26 Mb/sec
** SSD random writes
Before this change: 49.91 Mb/sec
After this change: 56.39 Mb/sec
** HDD sequential writes
Before this change: 68.53 Mb/sec
After this change: 69.87 Mb/sec
** HDD random writes
Before this change: 13.04 Mb/sec
After this change: 14.39 Mb/sec
Signed-off-by: Filipe David Borba Manana <fdmanana@gmail.com>
Signed-off-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
This is the third step in bootstrapping the btrfs_find_item interface.
The function find_orphan_item(), in orphan.c, is similar to the two
functions already replaced by the new interface. It uses two parameters,
which are already present in the interface, and is nearly identical to
the function brought in in the previous patch.
Replace the two calls to find_orphan_item() with calls to
btrfs_find_item(), with the defined objectid and type that was used
internally by find_orphan_item(), a null path, and a null key. Add a
test for a null path to btrfs_find_item, and if it passes, allocate and
free the path. Finally, remove find_orphan_item().
Signed-off-by: Kelley Nielsen <kelleynnn@gmail.com>
Signed-off-by: Josef Bacik <jbacik@fusionio.com>
Signed-off-by: Chris Mason <clm@fb.com>
Btrfs has always had these filler extent data items for holes in inodes. This
has made somethings very easy, like logging hole punches and sending hole
punches. However for large holey files these extent data items are pure
overhead. So add an incompatible feature to no longer add hole extents to
reduce the amount of metadata used by these sort of files. This has a few
changes for logging and send obviously since they will need to detect holes and
log/send the holes if there are any. I've tested this thoroughly with xfstests
and it doesn't cause any issues with and without the incompat format set.
Thanks,
Signed-off-by: Josef Bacik <jbacik@fusionio.com>
Signed-off-by: Chris Mason <clm@fb.com>
If we fsync, seek and write, rename and then fsync again we will lose the
modified hole extent because the rename will drop all of the modified extents
since we didn't do the fast search. We need to only drop the modified extents
if we didn't do the fast search and we were logging the entire inode as we don't
need them anymore, otherwise this is being premature. Thanks,
Signed-off-by: Josef Bacik <jbacik@fusionio.com>
Signed-off-by: Chris Mason <chris.mason@fusionio.com>
If we rename a file that is already in the log and we fsync again we will lose
the new name. This is because we just log the inode update and not the new ref.
To fix this we just need to check if we are logging the new name of the inode
and copy all the metadata instead of just updating the inode itself. With this
patch my testcase now passes. Thanks,
Signed-off-by: Josef Bacik <jbacik@fusionio.com>
Signed-off-by: Chris Mason <chris.mason@fusionio.com>
Use WARN_ON()'s return value in place of WARN_ON(1) for cleaner source
code that outputs a more descriptive warnings. Also fix the styling
warning of redundant braces that came up as a result of this fix.
Signed-off-by: Dulshani Gunawardhana <dulshani.gunawardhana89@gmail.com>
Reviewed-by: Zach Brown <zab@redhat.com>
Signed-off-by: Josef Bacik <jbacik@fusionio.com>
Signed-off-by: Chris Mason <chris.mason@fusionio.com>
If we get any error while doing a dir index/item lookup in the
log tree, we were always unlinking the corresponding inode in
the subvolume. It makes sense to unlink only if the lookup failed
to find the dir index/item, which corresponds to NULL or -ENOENT,
and not when other errors happen (like a transient -ENOMEM or -EIO).
Signed-off-by: Filipe David Borba Manana <fdmanana@gmail.com>
Signed-off-by: Josef Bacik <jbacik@fusionio.com>
Signed-off-by: Chris Mason <chris.mason@fusionio.com>
We were setting the csums search offset and length to the right values if
the extent is compressed, but later on right before doing the csums lookup
we were overriding these two parameters regardless of compression being
set or not for the extent.
Signed-off-by: Filipe David Borba Manana <fdmanana@gmail.com>
Signed-off-by: Josef Bacik <jbacik@fusionio.com>
Signed-off-by: Chris Mason <chris.mason@fusionio.com>
Originally, we introduced scrub_super_lock to synchronize
tree log code with scrubbing super.
However we can replace scrub_super_lock with device_list_mutex,
because writing super will hold this mutex, this will reduce an extra
lock holding when writing supers in sync log code.
Signed-off-by: Wang Shilong <wangsl.fnst@cn.fujitsu.com>
Signed-off-by: Josef Bacik <jbacik@fusionio.com>
Signed-off-by: Chris Mason <chris.mason@fusionio.com>
fs/btrfs/compat.h only contained trivial macro wrappers of drop_nlink()
and inc_nlink(). This doesn't belong in mainline.
Signed-off-by: Zach Brown <zab@redhat.com>
Signed-off-by: Josef Bacik <jbacik@fusionio.com>
Signed-off-by: Chris Mason <chris.mason@fusionio.com>
Avoid repeated tree searches by processing all inode ref items in
a leaf at once instead of processing one at a time, followed by a
path release and a tree search for a key with a decremented offset.
Signed-off-by: Filipe David Borba Manana <fdmanana@gmail.com>
Signed-off-by: Josef Bacik <jbacik@fusionio.com>
Signed-off-by: Chris Mason <chris.mason@fusionio.com>
In add_inode_ref() function:
Initializes local pointers.
Reduces the logical condition with the __add_inode_ref() return
value by using only one 'goto out'.
Centralizes the exiting, ensuring the freeing of all used memory.
Signed-off-by: Geyslan G. Bem <geyslan@gmail.com>
Reviewed-by: Stefan Behrens <sbehrens@giantdisaster.de>
Signed-off-by: Josef Bacik <jbacik@fusionio.com>
Signed-off-by: Chris Mason <chris.mason@fusionio.com>
The btrfs_insert_empty_item() function doesn't modify its
key argument.
Signed-off-by: Filipe David Borba Manana <fdmanana@gmail.com>
Reviewed-by: Zach Brown <zab@redhat.com>
Signed-off-by: Josef Bacik <jbacik@fusionio.com>
Signed-off-by: Chris Mason <chris.mason@fusionio.com>
I added an assert to make sure we were looking up aligned offsets for csums and
I tripped it when running xfstests. This is because log_one_extent was checking
if block_start == 0 for a hole instead of EXTENT_MAP_HOLE. This worked out fine
in practice it seems, but it adds a lot of extra work that is uneeded. With
this fix I'm no longer tripping my assert. Thanks,
Signed-off-by: Josef Bacik <jbacik@fusionio.com>
Signed-off-by: Chris Mason <chris.mason@fusionio.com>
On error we will wait and free the tree log at unmount without a transaction.
This means that the actual freeing of the blocks doesn't happen which means we
complain about space leaks on unmount. So to fix this just skip the transaction
specific cleanup part of the tree log free'ing if we don't have a transaction
and that way we can free up our reserved space and our counters stay happy.
Thanks,
Signed-off-by: Josef Bacik <jbacik@fusionio.com>
Signed-off-by: Chris Mason <chris.mason@fusionio.com>
In tree-log.c:btrfs_log_inode(), we keep calling btrfs_search_forward()
until it returns a key whose objectid is higher than our inode or until
the key's type is higher than our maximum allowed type.
At the end of the loop, we increment our mininum search key's objectid
and type regardless of our desired target objectid and maximum desired
type, which causes another loop iteration that will call again
btrfs_search_forward() just to figure out we've gone beyond our maximum
key and exit the loop. Therefore while incrementing our minimum key,
don't do it blindly and exit the loop immiediately if the next search
key's objectid or type is beyond what we seek.
Also after incrementing the type, set the key's offset to 0, which was
missing and could make us loose some of the inode's items.
Signed-off-by: Filipe David Borba Manana <fdmanana@gmail.com>
Signed-off-by: Josef Bacik <jbacik@fusionio.com>
Signed-off-by: Chris Mason <chris.mason@fusionio.com>
It is not used for anything.
Signed-off-by: Filipe David Borba Manana <fdmanana@gmail.com>
Signed-off-by: Josef Bacik <jbacik@fusionio.com>
Signed-off-by: Chris Mason <chris.mason@fusionio.com>
So if we have dir_index items in the log that means we also have the inode item
as well, which means that the inode's i_size is correct. However when we
process dir_index'es we call btrfs_add_link() which will increase the
directory's i_size for the new entry. To fix this we need to just set the dir
items i_size to 0, and then as we find dir_index items we adjust the i_size.
btrfs_add_link() will do it for new entries, and if the entry already exists we
can just add the name_len to the i_size ourselves. Thanks,
Signed-off-by: Josef Bacik <jbacik@fusionio.com>
Signed-off-by: Chris Mason <chris.mason@fusionio.com>
A user reported a bug where his log would not replay because he was getting
-EEXIST back. This was because he had a file moved into a directory that was
logged. What happens is the file had a lower inode number, and so it is
processed first when replaying the log, and so we add the inode ref in for the
directory it was moved to. But then we process the directories DIR_INDEX item
and try to add the inode ref for that inode and it fails because we already
added it when we replayed the inode. To solve this problem we need to just
process any DIR_INDEX items we have in the log first so this all is taken care
of, and then we can replay the rest of the items. With this patch my reproducer
can remount the file system properly instead of erroring out. Thanks,
Signed-off-by: Josef Bacik <jbacik@fusionio.com>
Signed-off-by: Chris Mason <chris.mason@fusionio.com>
If you just create a directory and then fsync that directory and then pull the
power plug you will come back up and the directory will not be there. That is
because we won't actually create directories if we've logged files inside of
them since they will be created on replay, but in this check we will set our
logged_trans of our current directory if it happens to be a directory, making us
think it doesn't need to be logged. Fix the logic to only do this to parent
directories. Thanks,
Signed-off-by: Josef Bacik <jbacik@fusionio.com>
Signed-off-by: Chris Mason <chris.mason@fusionio.com>
tree-log.c was ignoring the return value from btrfs_run_delayed_items()
in several places.
Signed-off-by: Filipe David Borba Manana <fdmanana@gmail.com>
Reviewed-by: Miao Xie <miaox@cn.fujitsu.com>
Signed-off-by: Josef Bacik <jbacik@fusionio.com>
Signed-off-by: Chris Mason <chris.mason@fusionio.com>
In tree-log.c:replay_one_name(), if memory allocation for
the name fails, ensure we iput the dir inode we got before
before we return.
Signed-off-by: Filipe David Borba Manana <fdmanana@gmail.com>
Signed-off-by: Josef Bacik <jbacik@fusionio.com>
Signed-off-by: Chris Mason <chris.mason@fusionio.com>
The ceph guys tripped over this bug where we were still holding onto the
original path that we used to copy the inode with when logging. This is based
on Chris's fix which was reported to fix the problem. We need to drop the paths
in two cases anyway so just move the drop up so that we don't have duplicate
code. Thanks,
Cc: stable@vger.kernel.org
Signed-off-by: Josef Bacik <jbacik@fusionio.com>
Signed-off-by: Chris Mason <chris.mason@fusionio.com>
With non-mixed block groups we replay the logs before we're allowed to do any
writes, so we get away with not pinning/removing the data extents until right
when we replay them. However with mixed block groups we allocate out of the
same pool, so we could easily allocate a metadata block that was logged in our
tree log. To deal with this we just need to notice that we have mixed block
groups and do the normal excluding/removal dance during the pin stage of the log
replay and that way we don't allocate metadata blocks from areas we have logged
data extents. With this patch we now pass xfstests generic/311 with mixed
block groups turned on. Thanks,
Signed-off-by: Josef Bacik <jbacik@fusionio.com>
Before applying this patch, we flushed the log tree of the fs/file
tree firstly, and then flushed the log root tree. It is ineffective,
especially on the hard disk. This patch improved this problem by wrapping
the above two flushes by the same blk_plug.
By test, the performance of the sync write went up ~60%(2.9MB/s -> 4.6MB/s)
on my scsi disk whose disk buffer was enabled.
Test step:
# mkfs.btrfs -f -m single <disk>
# mount <disk> <mnt>
# dd if=/dev/zero of=<mnt>/file0 bs=32K count=1024 oflag=sync
Signed-off-by: Miao Xie <miaox@cn.fujitsu.com>
Signed-off-by: Josef Bacik <jbacik@fusionio.com>
EXTREF is treated same as REF, so we can make the code tidy.
Signed-off-by: Liu Bo <bo.li.liu@oracle.com>
Signed-off-by: Josef Bacik <jbacik@fusionio.com>
There are several functions whose code is similar, such as
btrfs_find_last_root()
btrfs_read_fs_root_no_radix()
Besides that, some functions are invoked twice, it is unnecessary,
for example, we are sure that all roots which is found in
btrfs_find_orphan_roots()
have their orphan items, so it is unnecessary to check the orphan
item again.
So cleanup it.
Signed-off-by: Miao Xie <miaox@cn.fujitsu.com>
Signed-off-by: Josef Bacik <jbacik@fusionio.com>
Big patch, but all it does is add statics to functions which
are in fact static, then remove the associated dead-code fallout.
removed functions:
btrfs_iref_to_path()
__btrfs_lookup_delayed_deletion_item()
__btrfs_search_delayed_insertion_item()
__btrfs_search_delayed_deletion_item()
find_eb_for_page()
btrfs_find_block_group()
range_straddles_pages()
extent_range_uptodate()
btrfs_file_extent_length()
btrfs_scrub_cancel_devid()
btrfs_start_transaction_lflush()
btrfs_print_tree() is left because it is used for debugging.
btrfs_start_transaction_lflush() and btrfs_reada_detach() are
left for symmetry.
ulist.c functions are left, another patch will take care of those.
Signed-off-by: Eric Sandeen <sandeen@redhat.com>
Signed-off-by: Josef Bacik <jbacik@fusionio.com>
There were a whole bunch and I was doing it for other things. I haven't tested
these error paths but at the very least this is better than panicing. I've only
left 2 BUG_ON()'s since they are logic errors and I want to replace them with a
ASSERT framework that we can compile out for production users. Thanks,
Signed-off-by: Josef Bacik <jbacik@fusionio.com>
So everybody who got hit by my fsync bug will still continue to hit this
BUG_ON() in the free space cache, which is pretty heavy handed. So I took a
file system that had this bug and fixed up all the BUG_ON()'s and leaks that
popped up when I tried to mount a broken file system like this. With this patch
we just fail to mount instead of panicing. Thanks,
Signed-off-by: Josef Bacik <jbacik@fusionio.com>
We need to check the return value of the commit in case something goes wrong,
otherwise we could end up going down the line and doing more stuff (like orphan
cleanup) before we notice we should have errored out. We need to do this before
we free up the log_tree_root since the caller will handle all of that. Thanks,
Signed-off-by: Josef Bacik <jbacik@fusionio.com>
Argument 'trans' is not used in btrfs_extend_item().
Signed-off-by: Tsutomu Itoh <t-itoh@jp.fujitsu.com>
Signed-off-by: Josef Bacik <jbacik@fusionio.com>
If argument 'trans' is unnecessary in the function where
fixup_low_keys() is called, 'trans' is deleted.
Signed-off-by: Tsutomu Itoh <t-itoh@jp.fujitsu.com>
Signed-off-by: Josef Bacik <jbacik@fusionio.com>
A user sent me a btrfs-image of a file system that was panicing on mount during
the log recovery. I had originally thought these problems were from a bug in
the free space cache code, but that was just a symptom of the problem. The
problem is if your application does something like this
[prealloc][prealloc][prealloc]
the internal extent maps will merge those all together into one extent map, even
though on disk they are 3 separate extents. So if you go to write into one of
these ranges the extent map will be right since we use the physical extent when
doing the write, but when we log the extents they will use the wrong sizes for
the remainder prealloc space. If this doesn't happen to trip up the free space
cache (which it won't in a lot of cases) then you will get bogus entries in your
extent tree which will screw stuff up later. The data and such will still work,
but everything else is broken. This patch fixes this by not allowing extents
that are on the modified list to be merged. This has the side effect that we
are no longer adding everything to the modified list all the time, which means
we now have to call btrfs_drop_extents every time we log an extent into the
tree. So this allows me to drop all this speciality code I was using to get
around calling btrfs_drop_extents. With this patch the testcase I've created no
longer creates a bogus file system after replaying the log. Thanks,
Signed-off-by: Josef Bacik <jbacik@fusionio.com>
When logging changed extents I was logging ram_bytes as the current length,
which isn't correct, it's supposed to be the ram bytes of the original extent.
This is for compression where even if we split the extent we need to know the
ram bytes so when we uncompress the extent we know how big it will be. This was
still working out right with compression for some reason but I think we were
getting lucky. It was definitely off for prealloc which is why I noticed it,
btrfsck was complaining about it. With this patch btrfsck no longer complains
after a log replay. Thanks,
Signed-off-by: Josef Bacik <jbacik@fusionio.com>
While trying to track down a tree log replay bug I noticed that fsck was always
complaining about nbytes not being right for our fsynced file. That is because
the new fsync stuff doesn't wait for ordered extents to complete, so the inodes
nbytes are not necessarily updated properly when we log it. So to fix this we
need to set nbytes to whatever it is on the inode that is on disk, so when we
replay the extents we can just add the bytes that are being added as we replay
the extent. This makes it work for the case that we have the wrong nbytes or
the case that we logged everything and nbytes is actually correct. With this
I'm no longer getting nbytes errors out of btrfsck.
Cc: stable@vger.kernel.org
Signed-off-by: Josef Bacik <jbacik@fusionio.com>
Signed-off-by: Chris Mason <chris.mason@fusionio.com>
We need to inc the nlink of deleted entries when running replay so we can do the
unlink on the fs_root and get everything cleaned up and then have the orphan
cleanup do the right thing. The problem is inc_nlink complains about this, even
thought it still does the right thing. So use set_nlink() if our i_nlink is 0
to keep users from seeing the warnings during log replay. Thanks,
Signed-off-by: Josef Bacik <jbacik@fusionio.com>
Apparently when we do inline extents we allow the data to overlap the last chunk
of the btrfs_file_extent_item, which means that we can possibly have a
btrfs_file_extent_item that isn't actually as large as a btrfs_file_extent_item.
This messes with us when we try to overwrite the extent when logging new extents
since we expect for it to be the right size. To fix this just delete the item
and try to do the insert again which will give us the proper sized
btrfs_file_extent_item. This fixes a panic where map_private_extent_buffer
would blow up because we're trying to write past the end of the leaf. Thanks,
Cc: stable@vger.kernel.org
Signed-off-by: Josef Bacik <jbacik@fusionio.com>
When we abort a transaction while fsyncing, we'll skip freeing log roots
part of committing a transaction, which leads to memory leak.
This adds a 'free log roots' in putting super when no more users hold
references on log roots, so it's safe and clean.
Signed-off-by: Liu Bo <bo.li.liu@oracle.com>
Signed-off-by: Josef Bacik <jbacik@fusionio.com>
Though most of the btrfs codes are using ALIGN macro for page alignment,
there are still some codes using open-coded alignment like the
following:
------
u64 mask = ((u64)root->stripesize - 1);
u64 ret = (val + mask) & ~mask;
------
Or even hidden one:
------
num_bytes = (end - start + blocksize) & ~(blocksize - 1);
------
Sometimes these open-coded alignment is not so easy to understand for
newbie like me.
This commit changes the open-coded alignment to the ALIGN macro for a
better readability.
Also there is a previous patch from David Sterba with similar changes,
but the patch is for 3.2 kernel and seems not merged.
http://www.spinics.net/lists/linux-btrfs/msg12747.html
Cc: David Sterba <dave@jikos.cz>
Signed-off-by: Qu Wenruo <quwenruo@cn.fujitsu.com>
Signed-off-by: Josef Bacik <jbacik@fusionio.com>
The entry point at the defrag ioctl always sets "cache only" to 0;
the codepaths haven't run for a long time as far as I can
tell. Chris says they're dead code, so remove them.
Signed-off-by: Eric Sandeen <sandeen@redhat.com>
Signed-off-by: Josef Bacik <jbacik@fusionio.com>
Since we don't actually copy the extent information from the source tree in
the fast case we don't need to wait for ordered io to be completed in order
to fsync, we just need to wait for the io to be completed. So when we're
logging our file just attach all of the ordered extents to the log, and then
when the log syncs just wait for IO_DONE on the ordered extents and then
write the super. Thanks,
Signed-off-by: Josef Bacik <jbacik@fusionio.com>
For compressed extents, the range of checksum is covered by disk length,
and the disk length is different with ram length, so we need to use disk
length instead to get us the right checksum.
Signed-off-by: Liu Bo <bo.li.liu@oracle.com>
Signed-off-by: Josef Bacik <jbacik@fusionio.com>
We drop the extent map tree lock while we're logging extents, so somebody
could come in and merge another extent into this one and screw up our
logging, or they could even remove us from the list which would keep us from
logging the extent or freeing our ref on it, so we need to make sure to not
clear LOGGING until after the extent is logged, and then we can merge it to
adjacent extents. Thanks,
Signed-off-by: Josef Bacik <jbacik@fusionio.com>
If we are syncing over and over the overhead of doing all those maps in
fill_inode_item and log_changed_extents really starts to hurt, so use map
tokens so we can avoid all the extra mapping. Since the token maps from our
offset to the end of the page make sure to set the first thing in the item
first so we really only do one map. Thanks,
Signed-off-by: Josef Bacik <jbacik@fusionio.com>
Signed-off-by: Chris Mason <chris.mason@fusionio.com>
We don't really need to copy extents from the source tree since we have all
of the information already available to us in the extent_map tree. So
instead just write the extents straight to the log tree and don't bother to
copy the extent items from the source tree.
Signed-off-by: Josef Bacik <jbacik@fusionio.com>
Signed-off-by: Chris Mason <chris.mason@fusionio.com>
We don't copy inode items anwyay, we just copy them straight into the log
from the in memory inode. So if we know we're only logging the inode, don't
bother dropping anything, just try to insert it and either if it succeeds or
we get EEXIST we can update the inode item in the log and carry on. Thanks,
Signed-off-by: Josef Bacik <jbacik@fusionio.com>
Signed-off-by: Chris Mason <chris.mason@fusionio.com>
Currently we copy all the file information into the log, inode item, the
refs, xattrs etc. Except most of this doesn't change from fsync to fsync,
just the inode item changes. So set a flag if an xattr changes or a link is
added, and otherwise only log the inode item. Thanks,
Signed-off-by: Josef Bacik <jbacik@fusionio.com>
Signed-off-by: Chris Mason <chris.mason@fusionio.com>
If we set BTRFS_INODE_NEEDS_FULL_SYNC, we should log all the extent,
but now we forget to take it into account, and set a wrong max key,
if so, we will skip the file extent metadata when doing logging. Fix it.
Signed-off-by: Miao Xie <miaox@cn.fujitsu.com>
Signed-off-by: Chris Mason <chris.mason@fusionio.com>
We forget to protect the modified_extents list, fix it.
Signed-off-by: Miao Xie <miaox@cn.fujitsu.com>
Reviewed-by: Liu Bo <bo.li.liu@oracle.com>
Signed-off-by: Chris Mason <chris.mason@fusionio.com>
There are two types of the file extent - inline extent and regular extent,
When we log file extents, we didn't take inline extent into account, fix it.
Signed-off-by: Miao Xie <miaox@cn.fujitsu.com>
Reviewed-by: Liu Bo <bo.li.liu@oracle.com>
Signed-off-by: Chris Mason <chris.mason@fusionio.com>
When we log new names, we need to log just enough to recreate the inode
during log replay, and there is no need to log extents along with it.
This actually fixes a bug revealed by xfstests 241, where it shows
that we're logging some extents that have not updated metadata,
so we don't get proper EXTENT_DATA items to be copied to log tree.
Signed-off-by: Liu Bo <bo.li.liu@oracle.com>
Signed-off-by: Chris Mason <chris.mason@fusionio.com>
When compiling with user namespace support btrfs fails like:
fs/btrfs/tree-log.c: In function ‘fill_inode_item’:
fs/btrfs/tree-log.c:2955:2: error: incompatible type for argument 3 of ‘btrfs_set_inode_uid’
fs/btrfs/ctree.h:2026:1: note: expected ‘u32’ but argument is of type ‘kuid_t’
fs/btrfs/tree-log.c:2956:2: error: incompatible type for argument 3 of ‘btrfs_set_inode_gid’
fs/btrfs/ctree.h:2027:1: note: expected ‘u32’ but argument is of type ‘kgid_t’
Fix this by using i_uid_read and i_gid_read in
Cc: Chris Mason <chris.mason@fusionio.com>
Cc: Josef Bacik <jbacik@fusionio.com>
Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com>
So far the return code of barrier_all_devices() is ignored, which
means that errors are ignored. The result can be a corrupt
filesystem which is not consistent.
This commit adds code to evaluate the return code of
barrier_all_devices(). The normal btrfs_error() mechanism is used to
switch the filesystem into read-only mode when errors are detected.
In order to decide whether barrier_all_devices() should return
error or success, the number of disks that are allowed to fail the
barrier submission is calculated. This calculation accounts for the
worst RAID level of metadata, system and data. If single, dup or
RAID0 is in use, a single disk error is already considered to be
fatal. Otherwise a single disk error is tolerated.
The calculation of the number of disks that are tolerated to fail
the barrier operation is performed when the filesystem gets mounted,
when a balance operation is started and finished, and when devices
are added or removed.
Signed-off-by: Stefan Behrens <sbehrens@giantdisaster.de>
We can just copy the in memory inode into the tree log directly, no sense in
updating the fs tree so we can copy it into the tree log tree. Thanks,
Signed-off-by: Josef Bacik <jbacik@fusionio.com>
When we truncate existing items in the tree log we've been searching for
each individual item and removing them. This is unnecessary churn and
searching, just keep track of the slot we are on and how many items we need
to delete and delete them all at once. Thanks,
Signed-off-by: Josef Bacik <jbacik@fusionio.com>
The tree logging stuff was looking up csums to copy over for prealloc
extents which is just work we don't need to be doing. Thanks,
Signed-off-by: Josef Bacik <jbacik@fusionio.com>
Everytime we write out dirty pages we search for an offset in the tree,
convert the bits in the state, and then when we wait we search for the
offset again and clear the bits. So for every dirty range in the io tree we
are doing 4 rb searches, which is suboptimal. With this patch we are only
doing 2 searches for every cycle (modulo weird things happening). Thanks,
Signed-off-by: Josef Bacik <jbacik@fusionio.com>
This patch adds basic support for extended inode refs. This includes support
for link and unlink of the refs, which basically gets us support for rename
as well.
Inode creation does not need changing - extended refs are only added after
the ref array is full.
Signed-off-by: Mark Fasheh <mfasheh@suse.de>
Moved part of the code into a sub function and replaced most of the gotos
by ifs, hoping that it will be easier to read now.
Signed-off-by: Jan Schmidt <list.btrfs@jan-o-sch.net>
Signed-off-by: Mark Fasheh <mfasheh@suse.de>
I started hitting warnings when running xfstest 68 in a loop because there
were EM's that were not lined up properly with the physical extents. This
is ok, if we do something like punch a hole or write to a preallocated space
or something like that we can have an EM that doesn't cover the entire
physical extent. So fix the tree logging stuff to cope with this case so we
don't just commit the transaction. With this patch I no longer see the
warnings from the tree logging code. Thanks,
Signed-off-by: Josef Bacik <jbacik@fusionio.com>
Dave Sterba pointed out a sleeping while atomic bug while doing fsync. This
is because I'm an idiot and didn't realize that rwlock's were spin locks, so
we've been holding this thing while doing allocations and such which is not
good. This patch fixes this by dropping the write lock before we do
anything heavy and re-acquire it when it is done. We also need to take a
ref on the em's in case their corresponding pages are evicted and mark them
as being logged so that releasepage does not remove them and doesn't remove
them from our local list. Thanks,
Reported-by: Dave Sterba <dave@jikos.cz>
Signed-off-by: Josef Bacik <jbacik@fusionio.com>
We forget to protect ->log_batch when syncing a file, this patch fix
this problem by atomic operation. And ->log_batch is used to check
if there are parallel sync operations or not, so it is unnecessary to
reset it to 0 after the sync operation of the current log tree complete.
Signed-off-by: Miao Xie <miaox@cn.fujitsu.com>
I audited all users of btrfs_drop_extents and found that nobody actually uses
the hint_byte argument. I'm sure it was used for something at some point but
it's not used now, and the way the pinning works the disk bytenr would never be
immediately useful anyway so lets just remove it. Thanks,
Signed-off-by: Josef Bacik <jbacik@fusionio.com>
This is based on Josef's "Btrfs: turbo charge fsync".
If an inode is a BTRFS_INODE_NODATASUM one, we don't need to look for csum
items any more.
Signed-off-by: Liu Bo <bo.li.liu@oracle.com>
This is based on Josef's "Btrfs: turbo charge fsync".
The current btrfs checks if an inode is in log by comparing
root's last_log_commit to inode's last_sub_trans[2].
But the problem is that this root->last_log_commit is shared among
inodes.
Say we have N inodes to be logged, after the first inode,
root's last_log_commit is updated and the N-1 remained files will
be skipped.
This fixes the bug by keeping a local copy of root's last_log_commit
inside each inode and this local copy will be maintained itself.
[1]: we regard each log transaction as a subset of btrfs's transaction,
i.e. sub_trans
Signed-off-by: Liu Bo <bo.li.liu@oracle.com>
This is based on Josef's "Btrfs: turbo charge fsync".
The above Josef's patch performs very good in random sync write test,
because we won't have too much extents to merge.
However, it does not performs good on the test:
dd if=/dev/zero of=foobar bs=4k count=12500 oflag=sync
The reason is when we do sequencial sync write, we need to merge the
current extent just with the previous one, so that we can get accumulated
extents to log:
A(4k) --> AA(8k) --> AAA(12k) --> AAAA(16k) ...
So we'll have to flush more and more checksum into log tree, which is the
bottleneck according to my tests.
But we can avoid this by telling fsync the real extents that are needed
to be logged.
With this, I did the above dd sync write test (size=50m),
w/o (orig) w/ (josef's) w/ (this)
SATA 104KB/s 109KB/s 121KB/s
ramdisk 1.5MB/s 1.5MB/s 10.7MB/s (613%)
Signed-off-by: Liu Bo <bo.li.liu@oracle.com>
This is based on Josef's "Btrfs: turbo charge fsync".
We should cleanup those extents after we've finished logging inode,
otherwise we may do redundant work on them.
Signed-off-by: Liu Bo <bo.li.liu@oracle.com>
I hit this a couple times while working on my fsync patch (all my bugs, not
normal operation), but with my new stuff we could have new errors from cases
I have not encountered, so instead of BUG()'ing we should be WARN()'ing so
that we are notified there is a problem but the user doesn't lose their
data. We can easily commit the transaction in the case that the tree
logging fails and still be fine, so let's try and be as nice to the user as
possible. Thanks,
Signed-off-by: Josef Bacik <jbacik@fusionio.com>
At least for the vm workload. Currently on fsync we will
1) Truncate all items in the log tree for the given inode if they exist
and
2) Copy all items for a given inode into the log
The problem with this is that for things like VMs you can have lots of
extents from the fragmented writing behavior, and worst yet you may have
only modified a few extents, not the entire thing. This patch fixes this
problem by tracking which transid modified our extent, and then when we do
the tree logging we find all of the extents we've modified in our current
transaction, sort them and commit them. We also only truncate up to the
xattrs of the inode and copy that stuff in normally, and then just drop any
extents in the range we have that exist in the log already. Here are some
numbers of a 50 meg fio job that does random writes and fsync()s after every
write
Original Patched
SATA drive 82KB/s 140KB/s
Fusion drive 431KB/s 2532KB/s
So around 2-6 times faster depending on your hardware. There are a few
corner cases, for example if you truncate at all we have to do it the old
way since there is no way to be sure what is in the log is ok. This
probably could be done smarter, but if you write-fsync-truncate-write-fsync
you deserve what you get. All this work is in RAM of course so if your
inode gets evicted from cache and you read it in and fsync it we'll do it
the slow way if we are still in the same transaction that we last modified
the inode in.
The biggest cool part of this is that it requires no changes to the recovery
code, so if you fsync with this patch and crash and load an old kernel, it
will run the recovery and be a-ok. I have tested this pretty thoroughly
with an fsync tester and everything comes back fine, as well as xfstests.
Thanks,
Signed-off-by: Josef Bacik <jbacik@fusionio.com>
We didn't check error of btrfs_update_inode(), but that error looks
easy to bubble back up.
Reviewed-by: David Sterba <dsterba@suse.cz>
Signed-off-by: Tsutomu Itoh <t-itoh@jp.fujitsu.com>
Signed-off-by: Josef Bacik <jbacik@fusionio.com>
While we are resolving directory modifications in the
tree log, we are triggering delayed metadata updates to
the filesystem btrees.
This commit forces the delayed updates to run so the
replay code can find any modifications done. It stops
us from crashing because the directory deleltion replay
expects items to be removed immediately from the tree.
Signed-off-by: Chris Mason <chris.mason@fusionio.com>
cc: stable@kernel.org
So dpkg fsync()'s the file and the directory containing the file whenever it
writes to a file which is really slow in btrfs. This is partly because
fsync()'ing a directory _always_ committed the transaction instead of just
going to the tree log. This is because drop_objectid_items() would return 1
since it does a btrfs_search_slot() which returns 1. In tree-log jargon
this means that we have to commit the transaction to be safe. So just check
if ret is greater than 0 and set it to 0 if it does. With this patch we now
use the tree-log instead of committing the entire transaction, which is
twice as fast on my box. Thanks,
Signed-off-by: Josef Bacik <josef@redhat.com>
We have this check down in the actual logging code, but this is after we
start a transaction and all that good stuff. So move the helper
inode_in_log() out so we can call it in fsync() and avoid starting a
transaction altogether and just exit if we've already fsync()'ed this file
recently. You would notice this issue if you fsync()'ed a file over and
over again until the transaction committed. Thanks,
Signed-off-by: Josef Bacik <josef@redhat.com>
btrfs_read_buffer() has the possibility of returning the error.
Therefore, I add the code in which the return value of btrfs_read_buffer()
is checked.
Signed-off-by: Tsutomu Itoh <t-itoh@jp.fujitsu.com>
verify_parent_transid needs to lock the extent range to make
sure no IO is underway, and so it can safely clear the
uptodate bits if our checks fail.
But, a few callers are using it with spinlocks held. Most
of the time, the generation numbers are going to match, and
we don't want to switch to a blocking lock just for the error
case. This adds an atomic flag to verify_parent_transid,
and changes it to return EAGAIN if it needs to block to
properly verifiy things.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
btrfs currently handles most errors with BUG_ON. This patch is a work-in-
progress but aims to handle most errors other than internal logic
errors and ENOMEM more gracefully.
This iteration prevents most crashes but can run into lockups with
the page lock on occasion when the timing "works out."
Signed-off-by: Jeff Mahoney <jeffm@suse.com>
wait_log_commit() and wait_for_writer() were using slightly different
conditions for deciding whether they should call schedule() and whether they
should continue in the wait loop. Thus it could happen that we busylooped when
the first condition was not true while the second one was. That is burning CPU
cycles needlessly and is deadly on UP machines...
Signed-off-by: Jan Kara <jack@suse.cz>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
Add a for_cow parameter to add_delayed_*_ref and pass the appropriate value
from every call site. The for_cow parameter will later on be used to
determine if a ref will change anything with respect to qgroups.
Delayed refs coming from relocation are always counted as for_cow, as they
don't change subvol quota.
Also pass in the fs_info for later use.
btrfs_find_all_roots() will use this as an optimization, as changes that are
for_cow will not change anything with respect to which root points to a
certain leaf. Thus, we don't need to add the current sequence number to
those delayed refs.
Signed-off-by: Arne Jansen <sensille@gmx.net>
Signed-off-by: Jan Schmidt <list.btrfs@jan-o-sch.net>
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mason/linux-btrfs: (114 commits)
Btrfs: check for a null fs root when writing to the backup root log
Btrfs: fix race during transaction joins
Btrfs: fix a potential btrfs_bio leak on scrub fixups
Btrfs: rename btrfs_bio multi -> bbio for consistency
Btrfs: stop leaking btrfs_bios on readahead
Btrfs: stop the readahead threads on failed mount
Btrfs: fix extent_buffer leak in the metadata IO error handling
Btrfs: fix the new inspection ioctls for 32 bit compat
Btrfs: fix delayed insertion reservation
Btrfs: ClearPageError during writepage and clean_tree_block
Btrfs: be smarter about committing the transaction in reserve_metadata_bytes
Btrfs: make a delayed_block_rsv for the delayed item insertion
Btrfs: add a log of past tree roots
btrfs: separate superblock items out of fs_info
Btrfs: use the global reserve when truncating the free space cache inode
Btrfs: release metadata from global reserve if we have to fallback for unlink
Btrfs: make sure to flush queued bios if write_cache_pages waits
Btrfs: fix extent pinning bugs in the tree log
Btrfs: make sure btrfs_remove_free_space doesn't leak EAGAIN
Btrfs: don't wait as long for more batches during SSD log commit
...
fs_info has now ~9kb, more than fits into one page. This will cause
mount failure when memory is too fragmented. Top space consumers are
super block structures super_copy and super_for_commit, ~2.8kb each.
Allocate them dynamically. fs_info will be ~3.5kb. (measured on x86_64)
Add a wrapper for freeing fs_info and all of it's dynamically allocated
members.
Signed-off-by: David Sterba <dsterba@suse.cz>
The tree log had two important bugs that could cause corruptions after a
crash. Sometimes we were allowing tree log blocks to be reused after
the tree log was committed but before the transaction commit was done.
This allowed a future metadata write to overwrite the tree log data. It
is fixed by adding a new variant of freeing reserved extents that always
pins them. Credit goes to Stefan Behrens and Arne Jansen for many many
hours spent tracking this bug down.
During tree log replay, we do a pass through the tree log and pin all
the extents we find. This makes sure the replay code won't go in and
use any of those blocks for new allocations during replay. The problem
is the free space cache isn't honoring these pinned extents. So the
allocator can end up handing them out, leading to all kinds of problems
during replay.
The fix here is to force any free space cache to load while we pin the
extents, and then to make sure we remove the pinned extents from the
free space rbtree.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
Reported-by: Stefan Behrens <sbehrens@giantdisaster.de>
When we're doing log commits, we try to wait for more writers to come in
and make the commit bigger. This helps improve performance on rotating
disks, but on SSDs it adds latencies.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
Replace remaining direct i_nlink updates with a new set_nlink()
updater function.
Signed-off-by: Miklos Szeredi <mszeredi@suse.cz>
Tested-by: Toshiyuki Okajima <toshi.okajima@jp.fujitsu.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
When btrfs recovers from a crash, it may hit the oops below:
------------[ cut here ]------------
kernel BUG at fs/btrfs/inode.c:4580!
[...]
RIP: 0010:[<ffffffffa03df251>] [<ffffffffa03df251>] btrfs_add_link+0x161/0x1c0 [btrfs]
[...]
Call Trace:
[<ffffffffa03e7b31>] ? btrfs_inode_ref_index+0x31/0x80 [btrfs]
[<ffffffffa04054e9>] add_inode_ref+0x319/0x3f0 [btrfs]
[<ffffffffa0407087>] replay_one_buffer+0x2c7/0x390 [btrfs]
[<ffffffffa040444a>] walk_down_log_tree+0x32a/0x480 [btrfs]
[<ffffffffa0404695>] walk_log_tree+0xf5/0x240 [btrfs]
[<ffffffffa0406cc0>] btrfs_recover_log_trees+0x250/0x350 [btrfs]
[<ffffffffa0406dc0>] ? btrfs_recover_log_trees+0x350/0x350 [btrfs]
[<ffffffffa03d18b2>] open_ctree+0x1442/0x17d0 [btrfs]
[...]
This comes from that while replaying an inode ref item, we forget to
check those old conflicting DIR_ITEM and DIR_INDEX items in fs/file tree,
then we will come to conflict corners which lead to BUG_ON().
Signed-off-by: Liu Bo <liubo2009@cn.fujitsu.com>
Tested-by: Andy Lutomirski <luto@mit.edu>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
The btrfs metadata btree is the source of significant
lock contention, especially in the root node. This
commit changes our locking to use a reader/writer
lock.
The lock is built on top of rw spinlocks, and it
extends the lock tracking to remember if we have a
read lock or a write lock when we go to blocking. Atomics
count the number of blocking readers or writers at any
given time.
It removes all of the adaptive spinning from the old code
and uses only the spinning/blocking hints inside of btrfs
to decide when it should continue spinning.
In read heavy workloads this is dramatically faster. In write
heavy workloads we're still faster because of less contention
on the root node lock.
We suffer slightly in dbench because we schedule more often
during write locks, but all other benchmarks so far are improved.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
The two ->process_func call sites in tree-log.c which were ignoring a return
code have also been updated to gracefully exit as well.
Signed-off-by: Mark Fasheh <mfasheh@suse.com>
If return value of btrfs_inc_extent_ref() is not 0, BUG() is called.
Signed-off-by: Tsutomu Itoh <t-itoh@jp.fujitsu.com>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
When read_one_inode() fails, error code is returned to caller instead
of BUG_ON().
Signed-off-by: Tsutomu Itoh <t-itoh@jp.fujitsu.com>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
Currently, btrfs_truncate_item and btrfs_extend_item returns only 0.
So, the check by BUG_ON in the caller is unnecessary.
Signed-off-by: Tsutomu Itoh <t-itoh@jp.fujitsu.com>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
The error code is returned instead of calling BUG_ON when
btrfs_del_item returns the error.
Signed-off-by: Tsutomu Itoh <t-itoh@jp.fujitsu.com>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
The current code relogs the entire inode every time during fsync log,
and it is much better suited to small files rather than large ones.
During my performance test, the fsync performace of large files sucks,
and we can ascribe this to the tremendous amount of csum infos of the
large ones, cause we have to flush all of these csum infos into log trees
even when there are only _one_ change in the whole file data. Apparently,
to optimize fsync, we need to create a filter to skip the unnecessary csum
ones, that is, the corresponding file data remains unchanged before this fsync.
Here I have some test results to show, I use sysbench to do "random write + fsync".
===
sysbench --test=fileio --num-threads=1 --file-num=2 --file-block-size=4K --file-total-size=8G --file-test-mode=rndwr --file-io-mode=sync --file-extra-flags= [prepare, run]
===
Sysbench args:
- Number of threads: 1
- Extra file open flags: 0
- 2 files, 4Gb each
- Block size 4Kb
- Number of random requests for random IO: 10000
- Read/Write ratio for combined random IO test: 1.50
- Periodic FSYNC enabled, calling fsync() each 100 requests.
- Calling fsync() at the end of test, Enabled.
- Using synchronous I/O mode
- Doing random write test
Sysbench results:
===
Operations performed: 0 Read, 10000 Write, 200 Other = 10200 Total
Read 0b Written 39.062Mb Total transferred 39.062Mb
===
a) without patch: (*SPEED* : 451.01Kb/sec)
112.75 Requests/sec executed
b) with patch: (*SPEED* : 4.7533Mb/sec)
1216.84 Requests/sec executed
PS: I've made a _sub transid_ stuff patch, but it does not perform as effectively as this patch,
and I'm wanderring where the problem is and trying to improve it more.
Signed-off-by: Liu Bo <liubo2009@cn.fujitsu.com>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
Changelog V5 -> V6:
- Fix oom when the memory load is high, by storing the delayed nodes into the
root's radix tree, and letting btrfs inodes go.
Changelog V4 -> V5:
- Fix the race on adding the delayed node to the inode, which is spotted by
Chris Mason.
- Merge Chris Mason's incremental patch into this patch.
- Fix deadlock between readdir() and memory fault, which is reported by
Itaru Kitayama.
Changelog V3 -> V4:
- Fix nested lock, which is reported by Itaru Kitayama, by updating space cache
inode in time.
Changelog V2 -> V3:
- Fix the race between the delayed worker and the task which does delayed items
balance, which is reported by Tsutomu Itoh.
- Modify the patch address David Sterba's comment.
- Fix the bug of the cpu recursion spinlock, reported by Chris Mason
Changelog V1 -> V2:
- break up the global rb-tree, use a list to manage the delayed nodes,
which is created for every directory and file, and used to manage the
delayed directory name index items and the delayed inode item.
- introduce a worker to deal with the delayed nodes.
Compare with Ext3/4, the performance of file creation and deletion on btrfs
is very poor. the reason is that btrfs must do a lot of b+ tree insertions,
such as inode item, directory name item, directory name index and so on.
If we can do some delayed b+ tree insertion or deletion, we can improve the
performance, so we made this patch which implemented delayed directory name
index insertion/deletion and delayed inode update.
Implementation:
- introduce a delayed root object into the filesystem, that use two lists to
manage the delayed nodes which are created for every file/directory.
One is used to manage all the delayed nodes that have delayed items. And the
other is used to manage the delayed nodes which is waiting to be dealt with
by the work thread.
- Every delayed node has two rb-tree, one is used to manage the directory name
index which is going to be inserted into b+ tree, and the other is used to
manage the directory name index which is going to be deleted from b+ tree.
- introduce a worker to deal with the delayed operation. This worker is used
to deal with the works of the delayed directory name index items insertion
and deletion and the delayed inode update.
When the delayed items is beyond the lower limit, we create works for some
delayed nodes and insert them into the work queue of the worker, and then
go back.
When the delayed items is beyond the upper bound, we create works for all
the delayed nodes that haven't been dealt with, and insert them into the work
queue of the worker, and then wait for that the untreated items is below some
threshold value.
- When we want to insert a directory name index into b+ tree, we just add the
information into the delayed inserting rb-tree.
And then we check the number of the delayed items and do delayed items
balance. (The balance policy is above.)
- When we want to delete a directory name index from the b+ tree, we search it
in the inserting rb-tree at first. If we look it up, just drop it. If not,
add the key of it into the delayed deleting rb-tree.
Similar to the delayed inserting rb-tree, we also check the number of the
delayed items and do delayed items balance.
(The same to inserting manipulation)
- When we want to update the metadata of some inode, we cached the data of the
inode into the delayed node. the worker will flush it into the b+ tree after
dealing with the delayed insertion and deletion.
- We will move the delayed node to the tail of the list after we access the
delayed node, By this way, we can cache more delayed items and merge more
inode updates.
- If we want to commit transaction, we will deal with all the delayed node.
- the delayed node will be freed when we free the btrfs inode.
- Before we log the inode items, we commit all the directory name index items
and the delayed inode update.
I did a quick test by the benchmark tool[1] and found we can improve the
performance of file creation by ~15%, and file deletion by ~20%.
Before applying this patch:
Create files:
Total files: 50000
Total time: 1.096108
Average time: 0.000022
Delete files:
Total files: 50000
Total time: 1.510403
Average time: 0.000030
After applying this patch:
Create files:
Total files: 50000
Total time: 0.932899
Average time: 0.000019
Delete files:
Total files: 50000
Total time: 1.215732
Average time: 0.000024
[1] http://marc.info/?l=linux-btrfs&m=128212635122920&q=p3
Many thanks for Kitayama-san's help!
Signed-off-by: Miao Xie <miaox@cn.fujitsu.com>
Reviewed-by: David Sterba <dave@jikos.cz>
Tested-by: Tsutomu Itoh <t-itoh@jp.fujitsu.com>
Tested-by: Itaru Kitayama <kitayama@cl.bb4u.ne.jp>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
This adds an initial implementation for scrub. It works quite
straightforward. The usermode issues an ioctl for each device in the
fs. For each device, it enumerates the allocated device chunks. For
each chunk, the contained extents are enumerated and the data checksums
fetched. The extents are read sequentially and the checksums verified.
If an error occurs (checksum or EIO), a good copy is searched for. If
one is found, the bad copy will be rewritten.
All enumerations happen from the commit roots. During a transaction
commit, the scrubs get paused and afterwards continue from the new
roots.
This commit is based on the series originally posted to linux-btrfs
with some improvements that resulted from comments from David Sterba,
Ilya Dryomov and Jan Schmidt.
Signed-off-by: Arne Jansen <sensille@gmx.net>
parameter tree root it's not used since commit
5f39d397df ("Btrfs: Create extent_buffer
interface for large blocksizes")
Signed-off-by: David Sterba <dsterba@suse.cz>
It is necessary to unlock mutex_lock before it return an error when
btrfs_alloc_path() fails.
Signed-off-by: Tsutomu Itoh <t-itoh@jp.fujitsu.com>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
There's a potential problem in 32bit system when we exhaust 32bit inode
numbers and start to allocate big inode numbers, because btrfs uses
inode->i_ino in many places.
So here we always use BTRFS_I(inode)->location.objectid, which is an
u64 variable.
There are 2 exceptions that BTRFS_I(inode)->location.objectid !=
inode->i_ino: the btree inode (0 vs 1) and empty subvol dirs (256 vs 2),
and inode->i_ino will be used in those cases.
Another reason to make this change is I'm going to use a special inode
to save free ino cache, and the inode number must be > (u64)-256.
Signed-off-by: Li Zefan <lizf@cn.fujitsu.com>
When we recover from crash via write-ahead log tree and process
the inode refs, for each btrfs_inode_ref item, we will
1) check if we already have a perfect match in fs/file tree, if
we have, then we're done.
2) search the corresponding back reference in fs/file tree, and
check all the names in this back reference to see if they are
also in the log to avoid conflict corners.
3) recover the logged inode refs to fs/file tree.
In current btrfs, however,
- for 2)'s check, once is enough, since the checked back reference
will remain unchanged after processing all the inode refs belonged
to the key.
- it has no need to do another 1) between 2) and 3).
I've made a small test to show how it improves,
$dd if=/dev/zero of=foobar bs=4K count=1
$sync
$make 100 hard links continuously, like ln foobar link_i
$fsync foobar
$echo b > /proc/sysrq-trigger
after reboot
$time mount DEV PATH
without patch:
real 0m0.285s
user 0m0.001s
sys 0m0.009s
with patch:
real 0m0.123s
user 0m0.000s
sys 0m0.010s
Changelog v1->v2:
- fix double free - pointed by David Sterba
Changelog v2->v3:
- adjust free order
Signed-off-by: Liu Bo <liubo2009@cn.fujitsu.com>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
This patch changes some BUG_ON() to the error return.
(but, most callers still use BUG_ON())
Signed-off-by: Tsutomu Itoh <t-itoh@jp.fujitsu.com>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
We need to make sure the dir items we get are valid dir items. So any time we
try and read one check it with verify_dir_item, which will do various sanity
checks to make sure it looks sane. Thanks,
Signed-off-by: Josef Bacik <josef@redhat.com>
The error check of btrfs_start_transaction() is added, and the mistake
of the error check on several places is corrected.
Signed-off-by: Tsutomu Itoh <t-itoh@jp.fujitsu.com>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
Because NULL is returned when the memory allocation fails,
it is checked whether it is NULL.
Signed-off-by: Tsutomu Itoh <t-itoh@jp.fujitsu.com>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
btrfs_sync_log returns -EAGAIN when we need full transaction commits
instead of small log commits, but sometimes we were dropping the return
value.
In practice, we check for this a few different ways, but this is still a
bug that can leave off full log commits when we really need them.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
To make btrfs more stable, add several missing necessary memory allocation
checks, and when no memory, return proper errno.
We've checked that some of those -ENOMEM errors will be returned to
userspace, and some will be catched by BUG_ON() in the upper callers,
and none will be ignored silently.
Signed-off-by: Liu Bo <liubo2009@cn.fujitsu.com>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
There are lots of places where we do dentry->d_parent->d_inode without holding
the dentry->d_lock. This could cause problems with rename. So instead we need
to use dget_parent() and hold the reference to the parent as long as we are
going to use it's inode and then dput it at the end.
Signed-off-by: Josef Bacik <josef@redhat.com>
Cc: raven@themaw.net
Signed-off-by: Chris Mason <chris.mason@oracle.com>
These are all the cases where a variable is set, but not read which are
not bugs as far as I can see, but simply leftovers.
Still needs more review.
Found by gcc 4.6's new warnings
Signed-off-by: Andi Kleen <ak@linux.intel.com>
Cc: Chris Mason <chris.mason@oracle.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
These are all the cases where a variable is set, but not
read which are really bugs.
- Couple of incorrect error handling fixed.
- One incorrect use of a allocation policy
- Some other things
Still needs more review.
Found by gcc 4.6's new warnings.
[akpm@linux-foundation.org: fix build. Might have been bitrot]
Signed-off-by: Andi Kleen <ak@linux.intel.com>
Cc: Chris Mason <chris.mason@oracle.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
Previous patches make the allocater return -ENOSPC if there is no
unreserved free metadata space. This patch updates tree log code
and various other places to propagate/handle the ENOSPC error.
Signed-off-by: Yan Zheng <zheng.yan@oracle.com>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
percpu.h is included by sched.h and module.h and thus ends up being
included when building most .c files. percpu.h includes slab.h which
in turn includes gfp.h making everything defined by the two files
universally available and complicating inclusion dependencies.
percpu.h -> slab.h dependency is about to be removed. Prepare for
this change by updating users of gfp and slab facilities include those
headers directly instead of assuming availability. As this conversion
needs to touch large number of source files, the following script is
used as the basis of conversion.
http://userweb.kernel.org/~tj/misc/slabh-sweep.py
The script does the followings.
* Scan files for gfp and slab usages and update includes such that
only the necessary includes are there. ie. if only gfp is used,
gfp.h, if slab is used, slab.h.
* When the script inserts a new include, it looks at the include
blocks and try to put the new include such that its order conforms
to its surrounding. It's put in the include block which contains
core kernel includes, in the same order that the rest are ordered -
alphabetical, Christmas tree, rev-Xmas-tree or at the end if there
doesn't seem to be any matching order.
* If the script can't find a place to put a new include (mostly
because the file doesn't have fitting include block), it prints out
an error message indicating which .h file needs to be added to the
file.
The conversion was done in the following steps.
1. The initial automatic conversion of all .c files updated slightly
over 4000 files, deleting around 700 includes and adding ~480 gfp.h
and ~3000 slab.h inclusions. The script emitted errors for ~400
files.
2. Each error was manually checked. Some didn't need the inclusion,
some needed manual addition while adding it to implementation .h or
embedding .c file was more appropriate for others. This step added
inclusions to around 150 files.
3. The script was run again and the output was compared to the edits
from #2 to make sure no file was left behind.
4. Several build tests were done and a couple of problems were fixed.
e.g. lib/decompress_*.c used malloc/free() wrappers around slab
APIs requiring slab.h to be added manually.
5. The script was run on all .h files but without automatically
editing them as sprinkling gfp.h and slab.h inclusions around .h
files could easily lead to inclusion dependency hell. Most gfp.h
inclusion directives were ignored as stuff from gfp.h was usually
wildly available and often used in preprocessor macros. Each
slab.h inclusion directive was examined and added manually as
necessary.
6. percpu.h was updated not to include slab.h.
7. Build test were done on the following configurations and failures
were fixed. CONFIG_GCOV_KERNEL was turned off for all tests (as my
distributed build env didn't work with gcov compiles) and a few
more options had to be turned off depending on archs to make things
build (like ipr on powerpc/64 which failed due to missing writeq).
* x86 and x86_64 UP and SMP allmodconfig and a custom test config.
* powerpc and powerpc64 SMP allmodconfig
* sparc and sparc64 SMP allmodconfig
* ia64 SMP allmodconfig
* s390 SMP allmodconfig
* alpha SMP allmodconfig
* um on x86_64 SMP allmodconfig
8. percpu.h modifications were reverted so that it could be applied as
a separate patch and serve as bisection point.
Given the fact that I had only a couple of failures from tests on step
6, I'm fairly confident about the coverage of this conversion patch.
If there is a breakage, it's likely to be something in one of the arch
headers which should be easily discoverable easily on most builds of
the specific arch.
Signed-off-by: Tejun Heo <tj@kernel.org>
Guess-its-ok-by: Christoph Lameter <cl@linux-foundation.org>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Lee Schermerhorn <Lee.Schermerhorn@hp.com>
This work is in preperation for being able to set a different root as the
default mounting root.
There is currently a problem with how we mount subvolumes. We cannot currently
mount a subvolume of a subvolume, you can only mount subvolumes/snapshots of the
default subvolume. So say you take a snapshot of the default subvolume and call
it snap1, and then take a snapshot of snap1 and call it snap2, so now you have
/
/snap1
/snap1/snap2
as your available volumes. Currently you can only mount / and /snap1,
you cannot mount /snap1/snap2. To fix this problem instead of passing
subvolid=<name> you must pass in subvolid=<treeid>, where <treeid> is
the tree id that gets spit out via the subvolume listing you get from
the subvolume listing patches (btrfs filesystem list). This allows us
to mount /, /snap1 and /snap1/snap2 as the root volume.
In addition to the above, we also now read the default dir item in the
tree root to get the root key that it points to. For now this just
points at what has always been the default subvolme, but later on I plan
to change it to point at whatever root you want to be the new default
root, so you can just set the default mount and not have to mount with
-o subvolid=<treeid>. I tested this out with the above scenario and it
worked perfectly. Thanks,
mount -o subvol operates inside the selected subvolid. For example:
mount -o subvol=snap1,subvolid=256 /dev/xxx /mnt
/mnt will have the snap1 directory for the subvolume with id
256.
mount -o subvol=snap /dev/xxx /mnt
/mnt will be the snap directory of whatever the default subvolume
is.
Signed-off-by: Josef Bacik <josef@redhat.com>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
We do log replay in a single transaction, so it's not good to do unbound
operations. This patch cleans up orphan inodes cleanup after replaying
the log. It also avoids doing other unbound operations such as truncating
a file during replaying log. These unbound operations are postponed to
the orphan inode cleanup stage.
Signed-off-by: Yan Zheng <zheng.yan@oracle.com>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
Rewrite btrfs_drop_extents by using btrfs_duplicate_item, so we can
avoid calling lock_extent within transaction.
Signed-off-by: Yan Zheng <zheng.yan@oracle.com>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
We allow two log transactions at a time, but use same flag
to mark dirty tree-log btree blocks. So we may flush dirty
blocks belonging to newer log transaction when committing a
log transaction. This patch fixes the issue by using two
flags to mark dirty tree-log btree blocks.
Signed-off-by: Yan Zheng <zheng.yan@oracle.com>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
* 'master' of git://git.kernel.org/pub/scm/linux/kernel/git/mason/btrfs-unstable:
Btrfs: always pin metadata in discard mode
Btrfs: enable discard support
Btrfs: add -o discard option
Btrfs: properly wait log writers during log sync
Btrfs: fix possible ENOSPC problems with truncate
Btrfs: fix btrfs acl #ifdef checks
Btrfs: streamline tree-log btree block writeout
Btrfs: avoid tree log commit when there are no changes
Btrfs: only write one super copy during fsync
A recently fsync optimization make btrfs_sync_log skip calling
wait_for_writer in the single log writer case. This is incorrect
since the writer count can also be increased by btrfs_pin_log.
Signed-off-by: Yan Zheng <zheng.yan@oracle.com>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
Syncing the tree log is a 3 phase operation.
1) write and wait for all the tree log blocks for a given root.
2) write and wait for all the tree log blocks for the
tree of tree log roots.
3) write and wait for the super blocks (barriers here)
This isn't as efficient as it could be because there is
no requirement to wait for the blocks from step one to hit the disk
before we start writing the blocks from step two. This commit
changes the sequence so that we don't start waiting until
all the tree blocks from both steps one and two have been sent
to disk.
We do this by breaking up btrfs_write_wait_marked_extents into
two functions, which is trivial because it was already broken
up into two parts.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
rpm has a habit of running fdatasync when the file hasn't
changed. We already detect if a file hasn't been changed
in the current transaction but it might have been sent to
the tree-log in this transaction and not changed since
the last call to fsync.
In this case, we want to avoid a tree log sync, which includes
a number of synchronous writes and barriers. This commit
extends the existing tracking of the last transaction to change
a file to also track the last sub-transaction.
The end result is that rpm -ivh and -Uvh are roughly twice as fast,
and on par with ext3.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
During a tree-log commit for fsync, we've been writing at least
two copies of the super block and forcing them to disk.
The other filesystems write only one, and this change brings us on
par with them. A full transaction commit will write all the super
copies, so we still have redundant info written on a regular
basis.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
* git://git.kernel.org/pub/scm/linux/kernel/git/mason/btrfs-unstable:
Btrfs: fix file clone ioctl for bookend extents
Btrfs: fix uninit compiler warning in cow_file_range_nocow
Btrfs: constify dentry_operations
Btrfs: optimize back reference update during btrfs_drop_snapshot
Btrfs: remove negative dentry when deleting subvolumne
Btrfs: optimize fsync for the single writer case
Btrfs: async delalloc flushing under space pressure
Btrfs: release delalloc reservations on extent item insertion
Btrfs: delay clearing EXTENT_DELALLOC for compressed extents
Btrfs: cleanup extent_clear_unlock_delalloc flags
Btrfs: fix possible softlockup in the allocator
Btrfs: fix deadlock on async thread startup
This patch optimizes the tree logging stuff so it doesn't always wait 1 jiffie
for new people to join the logging transaction if there is only ever 1 writer.
This helps a little bit with latency where we have something like RPM where it
will fdatasync every file it writes, and so waiting the 1 jiffie for every
fdatasync really starts to add up.
Signed-off-by: Josef Bacik <jbacik@redhat.com>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
This patch adds snapshot/subvolume destroy ioctl. A subvolume that isn't being
used and doesn't contains links to other subvolumes can be destroyed.
Signed-off-by: Yan Zheng <zheng.yan@oracle.com>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
The new back reference format does not allow reusing objectid of
deleted snapshot/subvol. So we use ++highest_objectid to allocate
objectid for new snapshot/subvol.
Now we use ++highest_objectid to allocate objectid for both new inode
and new snapshot/subvolume, so this patch removes 'find hole' code in
btrfs_find_free_objectid.
Signed-off-by: Yan Zheng <zheng.yan@oracle.com>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
This patch gets rid of two limitations of async block group caching.
The old code delays handling pinned extents when block group is in
caching. To allocate logged file extents, the old code need wait
until block group is fully cached. To get rid of the limitations,
This patch introduces a data structure to track the progress of
caching. Base on the caching progress, we know which extents should
be added to the free space cache when handling the pinned extents.
The logged file extents are also handled in a similar way.
This patch also changes how pinned extents are tracked. The old
code uses one tree to track pinned extents, and copy the pinned
extents tree at transaction commit time. This patch makes it use
two trees to track pinned extents. One tree for extents that are
pinned in the running transaction, one tree for extents that can
be unpinned. At transaction commit time, we swap the two trees.
Signed-off-by: Yan Zheng <zheng.yan@oracle.com>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
Data COW means that whenever we write to a file, we replace any old
extent pointers with new ones. There was a window where a readpage
might find the old extent pointers on disk and cache them in the
extent_map tree in ram in the middle of a given write replacing them.
Even though both the readpage and the write had their respective bytes
in the file locked, the extent readpage inserts may cover more bytes than
it had locked down.
This commit closes the race by keeping the new extent pinned in the extent
map tree until after the on-disk btree is properly setup with the new
extent pointers.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
We are racy with async block caching and unpinning extents. This patch makes
things much less complicated by only unpinning the extent if the block group is
cached. We check the block_group->cached var under the block_group->lock spin
lock. If it is set to BTRFS_CACHE_FINISHED then we update the pinned counters,
and unpin the extent and add the free space back. If it is not set to this, we
start the caching of the block group so the next time we unpin extents we can
unpin the extent. This keeps us from racing with the async caching threads,
lets us kill the fs wide async thread counter, and keeps us from having to set
DELALLOC bits for every extent we hit if there are caching kthreads going.
One thing that needed to be changed was btrfs_free_super_mirror_extents. Now
instead of just looking for LOCKED extents, we also look for DIRTY extents,
since we could have left some extents pinned in the previous transaction that
will never get freed now that we are unmounting, which would cause us to leak
memory. So btrfs_free_super_mirror_extents has been changed to
btrfs_free_pinned_extents, and it will clear the extents locked for the super
mirror, and any remaining pinned extents that may be present. Thank you,
Signed-off-by: Josef Bacik <jbacik@redhat.com>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
dir has already been tested. It seems that this test should be on the
recently returned value inode.
A simplified version of the semantic match that finds this problem is as
follows: (http://www.emn.fr/x-info/coccinelle/)
Signed-off-by: Julia Lawall <julia@diku.dk>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
This patch moves the caching of the block group off to a kthread in order to
allow people to allocate sooner. Instead of blocking up behind the caching
mutex, we instead kick of the caching kthread, and then attempt to make an
allocation. If we cannot, we wait on the block groups caching waitqueue, which
the caching kthread will wake the waiting threads up everytime it finds 2 meg
worth of space, and then again when its finished caching. This is how I tested
the speedup from this
mkfs the disk
mount the disk
fill the disk up with fs_mark
unmount the disk
mount the disk
time touch /mnt/foo
Without my changes this took 11 seconds on my box, with these changes it now
takes 1 second.
Another change thats been put in place is we lock the super mirror's in the
pinned extent map in order to keep us from adding that stuff as free space when
caching the block group. This doesn't really change anything else as far as the
pinned extent map is concerned, since for actual pinned extents we use
EXTENT_DIRTY, but it does mean that when we unmount we have to go in and unlock
those extents to keep from leaking memory.
I've also added a check where when we are reading block groups from disk, if the
amount of space used == the size of the block group, we go ahead and mark the
block group as cached. This drastically reduces the amount of time it takes to
cache the block groups. Using the same test as above, except doing a dd to a
file and then unmounting, it used to take 33 seconds to umount, now it takes 3
seconds.
This version uses the commit_root in the caching kthread, and then keeps track
of how many async caching threads are running at any given time so if one of the
async threads is still running as we cross transactions we can wait until its
finished before handling the pinned extents. Thank you,
Signed-off-by: Josef Bacik <jbacik@redhat.com>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
During tree log replay, we read in the tree log roots,
process them and then free them. A recent change
takes an extra reference on the root node of the tree
when the root is read in, and stores that reference
in root->commit_root.
This reference was not being freed, leaving us with
one buffer pinned in ram for each subvol with
a tree log root after a crash.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
This commit introduces a new kind of back reference for btrfs metadata.
Once a filesystem has been mounted with this commit, IT WILL NO LONGER
BE MOUNTABLE BY OLDER KERNELS.
When a tree block in subvolume tree is cow'd, the reference counts of all
extents it points to are increased by one. At transaction commit time,
the old root of the subvolume is recorded in a "dead root" data structure,
and the btree it points to is later walked, dropping reference counts
and freeing any blocks where the reference count goes to 0.
The increments done during cow and decrements done after commit cancel out,
and the walk is a very expensive way to go about freeing the blocks that
are no longer referenced by the new btree root. This commit reduces the
transaction overhead by avoiding the need for dead root records.
When a non-shared tree block is cow'd, we free the old block at once, and the
new block inherits old block's references. When a tree block with reference
count > 1 is cow'd, we increase the reference counts of all extents
the new block points to by one, and decrease the old block's reference count by
one.
This dead tree avoidance code removes the need to modify the reference
counts of lower level extents when a non-shared tree block is cow'd.
But we still need to update back ref for all pointers in the block.
This is because the location of the block is recorded in the back ref
item.
We can solve this by introducing a new type of back ref. The new
back ref provides information about pointer's key, level and in which
tree the pointer lives. This information allow us to find the pointer
by searching the tree. The shortcoming of the new back ref is that it
only works for pointers in tree blocks referenced by their owner trees.
This is mostly a problem for snapshots, where resolving one of these
fuzzy back references would be O(number_of_snapshots) and quite slow.
The solution used here is to use the fuzzy back references in the common
case where a given tree block is only referenced by one root,
and use the full back references when multiple roots have a reference
on a given block.
This commit adds per subvolume red-black tree to keep trace of cached
inodes. The red-black tree helps the balancing code to find cached
inodes whose inode numbers within a given range.
This commit improves the balancing code by introducing several data
structures to keep the state of balancing. The most important one
is the back ref cache. It caches how the upper level tree blocks are
referenced. This greatly reduce the overhead of checking back ref.
The improved balancing code scales significantly better with a large
number of snapshots.
This is a very large commit and was written in a number of
pieces. But, they depend heavily on the disk format change and were
squashed together to make sure git bisect didn't end up in a
bad state wrt space balancing or the format change.
Signed-off-by: Yan Zheng <zheng.yan@oracle.com>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
The btrfs fallocate call takes an extent lock on the entire range
being fallocated, and then runs through insert_reserved_extent on each
extent as they are allocated.
The problem with this is that btrfs_drop_extents may decide to try
and take the same extent lock fallocate was already holding. The solution
used here is to push down knowledge of the range that is already locked
going into btrfs_drop_extents.
It turns out that at least one other caller had the same bug.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
Add a 'notreelog' mount option to disable the tree log (used by fsync,
O_SYNC writes). This is much slower, but the tree logging produces
inconsistent views into the FS for ceph.
Signed-off-by: Sage Weil <sage@newdream.net>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
This patch removes the pinned_mutex. The extent io map has an internal tree
lock that protects the tree itself, and since we only copy the extent io map
when we are committing the transaction we don't need it there. We also don't
need it when caching the block group since searching through the tree is also
protected by the internal map spin lock.
Signed-off-by: Josef Bacik <jbacik@redhat.com>
The fsync log has code to make sure all of the parents of a file are in the
log along with the file. It uses a minimal log of the parent directory
inodes, just enough to get the parent directory on disk.
If the transaction that originally created a file is fully on disk,
and the file hasn't been renamed or linked into other directories, we
can safely skip the parent directory walk. We know the file is on disk
somewhere and we can go ahead and just log that single file.
This is more important now because unrelated unlinks in the parent directory
might make us force a commit if we try to log the parent.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
The tree logging code allows individual files or directories to be logged
without including operations on other files and directories in the FS.
It tries to commit the minimal set of changes to disk in order to
fsync the single file or directory that was sent to fsync or O_SYNC.
The tree logging code was allowing files and directories to be unlinked
if they were part of a rename operation where only one directory
in the rename was in the fsync log. This patch adds a few new rules
to the tree logging.
1) on rename or unlink, if the inode being unlinked isn't in the fsync
log, we must force a full commit before doing an fsync of the directory
where the unlink was done. The commit isn't done during the unlink,
but it is forced the next time we try to log the parent directory.
Solution: record transid of last unlink/rename per directory when the
directory wasn't already logged. For renames this is only done when
renaming to a different directory.
mkdir foo/some_dir
normal commit
rename foo/some_dir foo2/some_dir
mkdir foo/some_dir
fsync foo/some_dir/some_file
The fsync above will unlink the original some_dir without recording
it in its new location (foo2). After a crash, some_dir will be gone
unless the fsync of some_file forces a full commit
2) we must log any new names for any file or dir that is in the fsync
log. This way we make sure not to lose files that are unlinked during
the same transaction.
2a) we must log any new names for any file or dir during rename
when the directory they are being removed from was logged.
2a is actually the more important variant. Without the extra logging
a crash might unlink the old name without recreating the new one
3) after a crash, we must go through any directories with a link count
of zero and redo the rm -rf
mkdir f1/foo
normal commit
rm -rf f1/foo
fsync(f1)
The directory f1 was fully removed from the FS, but fsync was never
called on f1, only its parent dir. After a crash the rm -rf must
be replayed. This must be able to recurse down the entire
directory tree. The inode link count fixup code takes care of the
ugly details.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
During log replay, inodes are copied from the log to the main filesystem
btrees. Sometimes they have a zero link count in the log but they actually
gain links during the replay or have some in the main btree.
This patch updates the link count to be at least one after copying the
inode out of the log. This makes sure the inode is deleted during an
iput while the rest of the replay code is still working on it.
The log replay has fixup code to make sure that link counts are correct
at the end of the replay, so we could use any non-zero number here and
it would work fine.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
btrfs_mark_buffer dirty would set dirty bits in the extent_io tree
for the buffers it was dirtying. This may require a kmalloc and it
was not atomic. So, anyone who called btrfs_mark_buffer_dirty had to
set any btree locks they were holding to blocking first.
This commit changes dirty tracking for extent buffers to just use a flag
in the extent buffer. Now that we have one and only one extent buffer
per page, this can be safely done without losing dirty bits along the way.
This also introduces a path->leave_spinning flag that callers of
btrfs_search_slot can use to indicate they will properly deal with a
path returned where all the locks are spinning instead of blocking.
Many of the btree search callers now expect spinning paths,
resulting in better btree concurrency overall.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
btrfs_record_root_in_trans needs the trans_mutex held to make sure two
callers don't race to setup the root in a given transaction. This adds
it to all the places that were missing it.
Signed-off-by: Yan Zheng <zheng.yan@oracle.com>
Most of the btrfs metadata operations can be protected by a spinlock,
but some operations still need to schedule.
So far, btrfs has been using a mutex along with a trylock loop,
most of the time it is able to avoid going for the full mutex, so
the trylock loop is a big performance gain.
This commit is step one for getting rid of the blocking locks entirely.
btrfs_tree_lock takes a spinlock, and the code explicitly switches
to a blocking lock when it starts an operation that can schedule.
We'll be able get rid of the blocking locks in smaller pieces over time.
Tracing allows us to find the most common cause of blocking, so we
can start with the hot spots first.
The basic idea is:
btrfs_tree_lock() returns with the spin lock held
btrfs_set_lock_blocking() sets the EXTENT_BUFFER_BLOCKING bit in
the extent buffer flags, and then drops the spin lock. The buffer is
still considered locked by all of the btrfs code.
If btrfs_tree_lock gets the spinlock but finds the blocking bit set, it drops
the spin lock and waits on a wait queue for the blocking bit to go away.
Much of the code that needs to set the blocking bit finishes without actually
blocking a good percentage of the time. So, an adaptive spin is still
used against the blocking bit to avoid very high context switch rates.
btrfs_clear_lock_blocking() clears the blocking bit and returns
with the spinlock held again.
btrfs_tree_unlock() can be called on either blocking or spinning locks,
it does the right thing based on the blocking bit.
ctree.c has a helper function to set/clear all the locked buffers in a
path as blocking.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
To improve performance, btrfs_sync_log merges tree log sync
requests. But it wrongly merges sync requests for different
tree logs. If multiple tree logs are synced at the same time,
only one of them actually gets synced.
This patch has following changes to fix the bug:
Move most tree log related fields in btrfs_fs_info to
btrfs_root. This allows merging sync requests separately
for each tree log.
Don't insert root item into the log root tree immediately
after log tree is allocated. Root item for log tree is
inserted when log tree get synced for the first time. This
allows syncing the log root tree without first syncing all
log trees.
At tree-log sync, btrfs_sync_log first sync the log tree;
then updates corresponding root item in the log root tree;
sync the log root tree; then update the super block.
Signed-off-by: Yan Zheng <zheng.yan@oracle.com>
Each subvolume has an extent_state_tree used to mark metadata
that needs to be sent to disk while syncing the tree. This is
used in addition to the dirty bits on the pages themselves so that
a single subvolume can be sent to disk efficiently in disk order.
Normally this marking happens in btrfs_alloc_free_block, which also does
special recording of dirty tree blocks for the tree log roots.
Yan Zheng noticed that when the root of the log tree is allocated, it is added
to the wrong writeback list. The fix used here is to explicitly set
it dirty as part of tree log creation.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
This patch contains following things.
1) Limit the max size of btrfs_ordered_sum structure to PAGE_SIZE. This
struct is kmalloced so we want to keep it reasonable.
2) Replace copy_extent_csums by btrfs_lookup_csums_range. This was
duplicated code in tree-log.c
3) Remove replay_one_csum. csum items are replayed at the same time as
replaying file extents. This guarantees we only replay useful csums.
4) nbytes accounting fix.
Signed-off-by: Yan Zheng <zheng.yan@oracle.com>
drop_one_dir_item does not properly update inode's link count. It can be
reproduced by executing following commands:
#touch test
#sync
#rm -f test
#dd if=/dev/zero bs=4k count=1 of=test conv=fsync
#echo b > /proc/sysrq-trigger
This fixes it by adding an BTRFS_ORPHAN_ITEM_KEY for the inode
Signed-off-by: Yan Zheng <zheng.yan@oracle.com>
btrfs_insert_empty_items takes the space needed by the btrfs_item
structure into account when calculating the required free space.
So the tree balancing code shouldn't add sizeof(struct btrfs_item)
to the size when checking the free space. This patch removes these
superfluous additions.
Signed-off-by: Yan Zheng <zheng.yan@oracle.com>
The fsync logging code makes sure to onl copy the relevant checksum for each
extent based on the file extent pointers it finds.
But for compressed extents, it needs to copy the checksum for the
entire extent.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
This patch implements superblock duplication. Superblocks
are stored at offset 16K, 64M and 256G on every devices.
Spaces used by superblocks are preserved by the allocator,
which uses a reverse mapping function to find the logical
addresses that correspond to superblocks. Thank you,
Signed-off-by: Yan Zheng <zheng.yan@oracle.com>
Btrfs stores checksums for each data block. Until now, they have
been stored in the subvolume trees, indexed by the inode that is
referencing the data block. This means that when we read the inode,
we've probably read in at least some checksums as well.
But, this has a few problems:
* The checksums are indexed by logical offset in the file. When
compression is on, this means we have to do the expensive checksumming
on the uncompressed data. It would be faster if we could checksum
the compressed data instead.
* If we implement encryption, we'll be checksumming the plain text and
storing that on disk. This is significantly less secure.
* For either compression or encryption, we have to get the plain text
back before we can verify the checksum as correct. This makes the raid
layer balancing and extent moving much more expensive.
* It makes the front end caching code more complex, as we have touch
the subvolume and inodes as we cache extents.
* There is potentitally one copy of the checksum in each subvolume
referencing an extent.
The solution used here is to store the extent checksums in a dedicated
tree. This allows us to index the checksums by phyiscal extent
start and length. It means:
* The checksum is against the data stored on disk, after any compression
or encryption is done.
* The checksum is stored in a central location, and can be verified without
following back references, or reading inodes.
This makes compression significantly faster by reducing the amount of
data that needs to be checksummed. It will also allow much faster
raid management code in general.
The checksums are indexed by a key with a fixed objectid (a magic value
in ctree.h) and offset set to the starting byte of the extent. This
allows us to copy the checksum items into the fsync log tree directly (or
any other tree), without having to invent a second format for them.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
This patch gives us the space we will need in order to have different csum
algorithims at some point in the future. We save the csum algorithim type
in the superblock, and use those instead of define's.
Signed-off-by: Josef Bacik <jbacik@redhat.com>
Shut up various sparse warnings about symbols that should be either
static or have their declarations in scope.
Signed-off-by: Christoph Hellwig <hch@lst.de>
This patch updates btrfs-progs for fallocate support.
fallocate is a little different in Btrfs because we need to tell the
COW system that a given preallocated extent doesn't need to be
cow'd as long as there are no snapshots of it. This leverages the
-o nodatacow checks.
Signed-off-by: Yan Zheng <zheng.yan@oracle.com>
This patch adds transaction IDs to root tree pointers.
Transaction IDs in tree pointers are compared with the
generation numbers in block headers when reading root
blocks of trees. This can detect some types of IO errors.
Signed-off-by: Yan Zheng <zheng.yan@oracle.com>
This patch removes the giant fs_info->alloc_mutex and replaces it with a bunch
of little locks.
There is now a pinned_mutex, which is used when messing with the pinned_extents
extent io tree, and the extent_ins_mutex which is used with the pending_del and
extent_ins extent io trees.
The locking for the extent tree stuff was inspired by a patch that Yan Zheng
wrote to fix a race condition, I cleaned it up some and changed the locking
around a little bit, but the idea remains the same. Basically instead of
holding the extent_ins_mutex throughout the processing of an extent on the
extent_ins or pending_del trees, we just hold it while we're searching and when
we clear the bits on those trees, and lock the extent for the duration of the
operations on the extent.
Also to keep from getting hung up waiting to lock an extent, I've added a
try_lock_extent so if we cannot lock the extent, move on to the next one in the
tree and we'll come back to that one. I have tested this heavily and it does
not appear to break anything. This has to be applied on top of my
find_free_extent redo patch.
I tested this patch on top of Yan's space reblancing code and it worked fine.
The only thing that has changed since the last version is I pulled out all my
debugging stuff, apparently I forgot to run guilt refresh before I sent the
last patch out. Thank you,
Signed-off-by: Josef Bacik <jbacik@redhat.com>
This is a large change for adding compression on reading and writing,
both for inline and regular extents. It does some fairly large
surgery to the writeback paths.
Compression is off by default and enabled by mount -o compress. Even
when the -o compress mount option is not used, it is possible to read
compressed extents off the disk.
If compression for a given set of pages fails to make them smaller, the
file is flagged to avoid future compression attempts later.
* While finding delalloc extents, the pages are locked before being sent down
to the delalloc handler. This allows the delalloc handler to do complex things
such as cleaning the pages, marking them writeback and starting IO on their
behalf.
* Inline extents are inserted at delalloc time now. This allows us to compress
the data before inserting the inline extent, and it allows us to insert
an inline extent that spans multiple pages.
* All of the in-memory extent representations (extent_map.c, ordered-data.c etc)
are changed to record both an in-memory size and an on disk size, as well
as a flag for compression.
From a disk format point of view, the extent pointers in the file are changed
to record the on disk size of a given extent and some encoding flags.
Space in the disk format is allocated for compression encoding, as well
as encryption and a generic 'other' field. Neither the encryption or the
'other' field are currently used.
In order to limit the amount of data read for a single random read in the
file, the size of a compressed extent is limited to 128k. This is a
software only limit, the disk format supports u64 sized compressed extents.
In order to limit the ram consumed while processing extents, the uncompressed
size of a compressed extent is limited to 256k. This is a software only limit
and will be subject to tuning later.
Checksumming is still done on compressed extents, and it is done on the
uncompressed version of the data. This way additional encodings can be
layered on without having to figure out which encoding to checksum.
Compression happens at delalloc time, which is basically singled threaded because
it is usually done by a single pdflush thread. This makes it tricky to
spread the compression load across all the cpus on the box. We'll have to
look at parallel pdflush walks of dirty inodes at a later time.
Decompression is hooked into readpages and it does spread across CPUs nicely.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
The offset field in struct btrfs_extent_ref records the position
inside file that file extent is referenced by. In the new back
reference system, tree leaves holding references to file extent
are recorded explicitly. We can scan these tree leaves very quickly, so the
offset field is not required.
This patch also makes the back reference system check the objectid
when extents are in deleting.
Signed-off-by: Yan Zheng <zheng.yan@oracle.com>
This patch makes btrfs count space allocated to file in bytes instead
of 512 byte sectors.
Everything else in btrfs uses a byte count instead of sector sizes or
blocks sizes, so this fits better.
Signed-off-by: Yan Zheng <zheng.yan@oracle.com>
This patch makes the back reference system to explicit record the
location of parent node for all types of extents. The location of
parent node is placed into the offset field of backref key. Every
time a tree block is balanced, the back references for the affected
lower level extents are updated.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
This is the same way the transaction code makes sure that all the
other tree blocks are safely on disk. There's an extent_io tree
for each root, and any blocks allocated to the tree logs are
recorded in that tree.
At tree-log sync, the extent_io tree is walked to flush down the
dirty pages and wait for them.
The main benefit is less time spent walking the tree log and skipping
clean pages, and getting sequential IO down to the drive.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
This changes the log tree copy code to use btrfs_insert_items and
to work in larger batches where possible.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
Since tree log blocks get freed every transaction, they never really
need to be written to disk. This skips the step where we update
metadata to record they were allocated.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
Drop i_mutex during the commit
Don't bother doing the fsync at all unless the dir is marked as dirtied
and needing fsync in this transaction. For directories, this means
that someone has unlinked a file from the dir without fsyncing the
file.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
* Pin down data blocks to prevent them from being reallocated like so:
trans 1: allocate file extent
trans 2: free file extent
trans 3: free file extent during old snapshot deletion
trans 3: allocate file extent to new file
trans 3: fsync new file
Before the tree logging code, this was legal because the fsync
would commit the transation that did the final data extent free
and the transaction that allocated the extent to the new file
at the same time.
With the tree logging code, the tree log subtransaction can commit
before the transaction that freed the extent. If we crash,
we're left with two different files using the extent.
* Don't wait in start_transaction if log replay is going on. This
avoids deadlocks from iput while we're cleaning up link counts in the
replay code.
* Don't deadlock in replay_one_name by trying to read an inode off
the disk while holding paths for the directory
* Hold the buffer lock while we mark a buffer as written. This
closes a race where someone is changing a buffer while we write it.
They are supposed to mark it dirty again after they change it, but
this violates the cow rules.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
File syncs and directory syncs are optimized by copying their
items into a special (copy-on-write) log tree. There is one log tree per
subvolume and the btrfs super block points to a tree of log tree roots.
After a crash, items are copied out of the log tree and back into the
subvolume. See tree-log.c for all the details.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
