If journal superblock is written only in disk's caches and other transaction
starts reusing space of the transaction cleaned from the log, it can happen
blocks of a new transaction reach the disk before journal superblock. When
power failure happens in such case, subsequent journal replay would still try
to replay the old transaction but some of it's blocks may be already
overwritten by the new transaction. For this reason we must use WRITE_FUA when
updating log tail and we must first write new log tail to disk and update
in-memory information only after that.
Signed-off-by: Jan Kara <jack@suse.cz>
There are some log tail updates that are not protected by j_checkpoint_mutex.
Some of these are harmless because they happen during startup or shutdown but
updates in journal_commit_transaction() and journal_flush() can really race
with other log tail updates (e.g. someone doing journal_flush() with someone
running cleanup_journal_tail()). So protect all log tail updates with
j_checkpoint_mutex.
Signed-off-by: Jan Kara <jack@suse.cz>
There are three case of updating journal superblock. In the first case, we want
to mark journal as empty (setting s_sequence to 0), in the second case we want
to update log tail, in the third case we want to update s_errno. Split these
cases into separate functions. It makes the code slightly more straightforward
and later patches will make the distinction even more important.
Signed-off-by: Jan Kara <jack@suse.cz>
Currently we write out all journal buffers in WRITE_SYNC mode. This improves
performance for fsync heavy workloads but hinders performance when writes
are mostly asynchronous, most noticably it slows down readers and users
complain about slow desktop response etc.
So submit writes as asynchronous in the normal case and only submit writes as
WRITE_SYNC if we detect someone is waiting for current transaction commit.
I've gathered some numbers to back this change. The first is the read latency
test. It measures time to read 1 MB after several seconds of sleeping in
presence of streaming writes.
Top 10 times (out of 90) in us:
Before After
2131586 697473
1709932 557487
1564598 535642
1480462 347573
1478579 323153
1408496 222181
1388960 181273
1329565 181070
1252486 172832
1223265 172278
Average:
619377 82180
So the improvement in both maximum and average latency is massive.
I've measured fsync throughput by:
fs_mark -n 100 -t 1 -s 16384 -d /mnt/fsync/ -S 1 -L 4
in presence of streaming reader. The numbers (fsyncs/s) are:
Before After
9.9 6.3
6.8 6.0
6.3 6.2
5.8 6.1
So fsync performance seems unharmed by this change.
Signed-off-by: Jan Kara <jack@suse.cz>
Pull power management updates for 3.4 from Rafael Wysocki:
"Assorted extensions and fixes including:
* Introduction of early/late suspend/hibernation device callbacks.
* Generic PM domains extensions and fixes.
* devfreq updates from Axel Lin and MyungJoo Ham.
* Device PM QoS updates.
* Fixes of concurrency problems with wakeup sources.
* System suspend and hibernation fixes."
* tag 'pm-for-3.4' of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm: (43 commits)
PM / Domains: Check domain status during hibernation restore of devices
PM / devfreq: add relation of recommended frequency.
PM / shmobile: Make MTU2 driver use pm_genpd_dev_always_on()
PM / shmobile: Make CMT driver use pm_genpd_dev_always_on()
PM / shmobile: Make TMU driver use pm_genpd_dev_always_on()
PM / Domains: Introduce "always on" device flag
PM / Domains: Fix hibernation restore of devices, v2
PM / Domains: Fix handling of wakeup devices during system resume
sh_mmcif / PM: Use PM QoS latency constraint
tmio_mmc / PM: Use PM QoS latency constraint
PM / QoS: Make it possible to expose PM QoS latency constraints
PM / Sleep: JBD and JBD2 missing set_freezable()
PM / Domains: Fix include for PM_GENERIC_DOMAINS=n case
PM / Freezer: Remove references to TIF_FREEZE in comments
PM / Sleep: Add more wakeup source initialization routines
PM / Hibernate: Enable usermodehelpers in hibernate() error path
PM / Sleep: Make __pm_stay_awake() delete wakeup source timers
PM / Sleep: Fix race conditions related to wakeup source timer function
PM / Sleep: Fix possible infinite loop during wakeup source destruction
PM / Hibernate: print physical addresses consistently with other parts of kernel
...
With the latest and greatest changes to the freezer, I started seeing
panics that were caused by jbd2 running post-process freezing and
hitting the canary BUG_ON for non-TuxOnIce I/O submission. I've traced
this back to a lack of set_freezable calls in both jbd and jbd2. Since
they're clearly meant to be frozen (there are tests for freezing()), I
submit the following patch to add the missing calls.
Signed-off-by: Nigel Cunningham <nigel@tuxonice.net>
Acked-by: Jan Kara <jack@suse.cz>
Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
When we reach cleanup_journal_tail(), there is no guarantee that
checkpointed buffers are on a stable storage - especially if buffers were
written out by log_do_checkpoint(), they are likely to be only in disk's
caches. Thus when we update journal superblock, effectively removing old
transaction from journal, this write of superblock can get to stable storage
before those checkpointed buffers which can result in filesystem corruption
after a crash.
A similar problem can happen if we replay the journal and wipe it before
flushing disk's caches.
Thus we must unconditionally issue a cache flush before we update journal
superblock in these cases. The fix is slightly complicated by the fact that we
have to get log tail before we issue cache flush but we can store it in the
journal superblock only after the cache flush. Otherwise we risk races where
new tail is written before appropriate cache flush is finished.
I managed to reproduce the corruption using somewhat tweaked Chris Mason's
barrier-test scheduler. Also this should fix occasional reports of 'Bit already
freed' filesystem errors which are totally unreproducible but inspection of
several fs images I've gathered over time points to a problem like this.
CC: stable@kernel.org
Signed-off-by: Jan Kara <jack@suse.cz>
* 'for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/jack/linux-fs:
ext2/3/4: delete unneeded includes of module.h
ext{3,4}: Fix potential race when setversion ioctl updates inode
udf: Mark LVID buffer as uptodate before marking it dirty
ext3: Don't warn from writepage when readonly inode is spotted after error
jbd: Remove j_barrier mutex
reiserfs: Force inode evictions before umount to avoid crash
reiserfs: Fix quota mount option parsing
udf: Treat symlink component of type 2 as /
udf: Fix deadlock when converting file from in-ICB one to normal one
udf: Cleanup calling convention of inode_getblk()
ext2: Fix error handling on inode bitmap corruption
ext3: Fix error handling on inode bitmap corruption
ext3: replace ll_rw_block with other functions
ext3: NULL dereference in ext3_evict_inode()
jbd: clear revoked flag on buffers before a new transaction started
ext3: call ext3_mark_recovery_complete() when recovery is really needed
j_barrier mutex is used for serializing different journal lock operations. The
problem with it is that e.g. FIFREEZE ioctl results in process leaving kernel
with j_barrier mutex held which makes lockdep freak out. Also hibernation code
wants to freeze filesystem but it cannot do so because it then cannot hibernate
the system because of mutex being locked.
So we remove j_barrier mutex and use direct wait on j_barrier_count instead.
Since locking journal is a rare operation we don't have to care about fairness
or such things.
CC: Andrew Morton <akpm@linux-foundation.org>
Acked-by: Joel Becker <jlbec@evilplan.org>
Signed-off-by: Jan Kara <jack@suse.cz>
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/jikos/trivial: (53 commits)
Kconfig: acpi: Fix typo in comment.
misc latin1 to utf8 conversions
devres: Fix a typo in devm_kfree comment
btrfs: free-space-cache.c: remove extra semicolon.
fat: Spelling s/obsolate/obsolete/g
SCSI, pmcraid: Fix spelling error in a pmcraid_err() call
tools/power turbostat: update fields in manpage
mac80211: drop spelling fix
types.h: fix comment spelling for 'architectures'
typo fixes: aera -> area, exntension -> extension
devices.txt: Fix typo of 'VMware'.
sis900: Fix enum typo 'sis900_rx_bufer_status'
decompress_bunzip2: remove invalid vi modeline
treewide: Fix comment and string typo 'bufer'
hyper-v: Update MAINTAINERS
treewide: Fix typos in various parts of the kernel, and fix some comments.
clockevents: drop unknown Kconfig symbol GENERIC_CLOCKEVENTS_MIGR
gpio: Kconfig: drop unknown symbol 'CS5535_GPIO'
leds: Kconfig: Fix typo 'D2NET_V2'
sound: Kconfig: drop unknown symbol ARCH_CLPS7500
...
Fix up trivial conflicts in arch/powerpc/platforms/40x/Kconfig (some new
kconfig additions, close to removed commented-out old ones)
Currently, we clear revoked flag only when a block is reused. However,
this can tigger a false journal error. Consider a situation when a block
is used as a meta block and is deleted(revoked) in ordered mode, then the
block is allocated as a data block to a file. At this moment, user changes
the file's journal mode from ordered to journaled and truncates the file.
The block will be considered re-revoked by journal because it has revoked
flag still pending from the last transaction and an assertion triggers.
We fix the problem by keeping the revoked status more uptodate - we clear
revoked flag when switching revoke tables to reflect there is no revoked
buffers in current transaction any more.
Signed-off-by: Yongqiang Yang <xiaoqiangnk@gmail.com>
Signed-off-by: Jan Kara <jack@suse.cz>
There is no reason to export two functions for entering the
refrigerator. Calling refrigerator() instead of try_to_freeze()
doesn't save anything noticeable or removes any race condition.
* Rename refrigerator() to __refrigerator() and make it return bool
indicating whether it scheduled out for freezing.
* Update try_to_freeze() to return bool and relay the return value of
__refrigerator() if freezing().
* Convert all refrigerator() users to try_to_freeze().
* Update documentation accordingly.
* While at it, add might_sleep() to try_to_freeze().
Signed-off-by: Tejun Heo <tj@kernel.org>
Cc: Samuel Ortiz <samuel@sortiz.org>
Cc: Chris Mason <chris.mason@oracle.com>
Cc: "Theodore Ts'o" <tytso@mit.edu>
Cc: Steven Whitehouse <swhiteho@redhat.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Jan Kara <jack@suse.cz>
Cc: KONISHI Ryusuke <konishi.ryusuke@lab.ntt.co.jp>
Cc: Christoph Hellwig <hch@infradead.org>
I hit a J_ASSERT(blocknr != 0) failure in cleanup_journal_tail() when
mounting a fsfuzzed ext3 image. It turns out that the corrupted ext3
image has s_first = 0 in journal superblock, and the 0 is passed to
journal->j_head in journal_reset(), then to blocknr in
cleanup_journal_tail(), in the end the J_ASSERT failed.
So validate s_first after reading journal superblock from disk in
journal_get_superblock() to ensure s_first is valid.
The following script could reproduce it:
fstype=ext3
blocksize=1024
img=$fstype.img
offset=0
found=0
magic="c0 3b 39 98"
dd if=/dev/zero of=$img bs=1M count=8
mkfs -t $fstype -b $blocksize -F $img
filesize=`stat -c %s $img`
while [ $offset -lt $filesize ]
do
if od -j $offset -N 4 -t x1 $img | grep -i "$magic";then
echo "Found journal: $offset"
found=1
break
fi
offset=`echo "$offset+$blocksize" | bc`
done
if [ $found -ne 1 ];then
echo "Magic \"$magic\" not found"
exit 1
fi
dd if=/dev/zero of=$img seek=$(($offset+23)) conv=notrunc bs=1 count=1
mkdir -p ./mnt
mount -o loop $img ./mnt
Cc: Jan Kara <jack@suse.cz>
Signed-off-by: Eryu Guan <guaneryu@gmail.com>
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
In journal checkpoint, we write the buffer and wait for its finish.
But in cfq, the async queue has a very low priority, and in our test,
if there are too many sync queues and every queue is filled up with
requests, and the process will hang waiting for the log space.
So this patch tries to use WRITE_SYNC in __flush_batch so that the request will
be moved into sync queue and handled by cfq timely. We also use the new plug,
sot that all the WRITE_SYNC requests can be given as a whole when we unplug it.
Reported-by: Robin Dong <sanbai@taobao.com>
Signed-off-by: Tao Ma <boyu.mt@taobao.com>
Signed-off-by: Jan Kara <jack@suse.cz>
journal_remove_journal_head() can oops when trying to access journal_head
returned by bh2jh(). This is caused for example by the following race:
TASK1 TASK2
journal_commit_transaction()
...
processing t_forget list
__journal_refile_buffer(jh);
if (!jh->b_transaction) {
jbd_unlock_bh_state(bh);
journal_try_to_free_buffers()
journal_grab_journal_head(bh)
jbd_lock_bh_state(bh)
__journal_try_to_free_buffer()
journal_put_journal_head(jh)
journal_remove_journal_head(bh);
journal_put_journal_head() in TASK2 sees that b_jcount == 0 and buffer is not
part of any transaction and thus frees journal_head before TASK1 gets to doing
so. Note that even buffer_head can be released by try_to_free_buffers() after
journal_put_journal_head() which adds even larger opportunity for oops (but I
didn't see this happen in reality).
Fix the problem by making transactions hold their own journal_head reference
(in b_jcount). That way we don't have to remove journal_head explicitely via
journal_remove_journal_head() and instead just remove journal_head when
b_jcount drops to zero. The result of this is that [__]journal_refile_buffer(),
[__]journal_unfile_buffer(), and __journal_remove_checkpoint() can free
journal_head which needs modification of a few callers. Also we have to be
careful because once journal_head is removed, buffer_head might be freed as
well. So we have to get our own buffer_head reference where it matters.
Signed-off-by: Jan Kara <jack@suse.cz>
journal_get_create_access should drop jh->b_jcount in error handling path
Signed-off-by: Ding Dinghua <dingdinghua@nrchpc.ac.cn>
Signed-off-by: Jan Kara <jack@suse.cz>
The callers of start_this_handle() (or better ext3_journal_start()) are not
really prepared to handle allocation failures. Such failures can for example
result in silent data loss when it happens in ext3_..._writepage(). OTOH
__GFP_NOFAIL is going away so we just retry allocation in start_this_handle().
This loop is potentially dangerous because the oom killer cannot be invoked
for GFP_NOFS allocation, so there is a potential for infinitely looping.
But still this is better than silent data loss.
Signed-off-by: Jan Kara <jack@suse.cz>
This commit adds fixed tracepoint for jbd. It has been based on fixed
tracepoints for jbd2, however there are missing those for collecting
statistics, since I think that it will require more intrusive patch so I
should have its own commit, if someone decide that it is needed. Also
there are new tracepoints in __journal_drop_transaction() and
journal_update_superblock().
The list of jbd tracepoints:
jbd_checkpoint
jbd_start_commit
jbd_commit_locking
jbd_commit_flushing
jbd_commit_logging
jbd_drop_transaction
jbd_end_commit
jbd_do_submit_data
jbd_cleanup_journal_tail
jbd_update_superblock_end
Signed-off-by: Lukas Czerner <lczerner@redhat.com>
Cc: Jan Kara <jack@suse.cz>
Signed-off-by: Jan Kara <jack@suse.cz>
journal_start returns an ERR_PTR() value rather than NULL on failure.
Cc: Jan Kara <jack@suse.cz>
Signed-off-by: Eryu Guan <guaneryu@gmail.com>
Signed-off-by: Jan Kara <jack@suse.cz>
summarise_journal_usage seems to be obsolete for a long time,
so remove it.
Cc: Jan Kara <jack@suse.cz>
Signed-off-by: Tao Ma <boyu.mt@taobao.com>
Signed-off-by: Jan Kara <jack@suse.cz>
In do_get_write_access() we wait on BH_Unshadow bit for buffer to get
from shadow state. The waking code in journal_commit_transaction() has
a bug because it does not issue a memory barrier after the buffer is moved
from the shadow state and before wake_up_bit() is called. Thus a waitqueue
check can happen before the buffer is actually moved from the shadow state
and waiting process may never be woken. Fix the problem by issuing proper
barrier.
CC: stable@kernel.org
Reported-by: Tao Ma <boyu.mt@taobao.com>
Signed-off-by: Jan Kara <jack@suse.cz>
If an application program does not make any changes to the indirect
blocks or extent tree, i_datasync_tid will not get updated. If there
are enough commits (i.e., 2**31) such that tid_geq()'s calculations
wrap, and there isn't a currently active transaction at the time of
the fdatasync() call, this can end up triggering a BUG_ON in
fs/jbd/commit.c:
J_ASSERT(journal->j_running_transaction != NULL);
It's pretty rare that this can happen, since it requires the use of
fdatasync() plus *very* frequent and excessive use of fsync(). But
with the right workload, it can.
We fix this by replacing the use of tid_geq() with an equality test,
since there's only one valid transaction id that is valid for us to
start: namely, the currently running transaction (if it exists).
CC: stable@kernel.org
Reported-by: Martin_Zielinski@McAfee.com
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
Signed-off-by: Jan Kara <jack@suse.cz>
