[ Upstream commit a0851ea9cd ]
When committing transaction in jbd2_journal_commit_transaction(), the
disk caches for the filesystem device should be flushed before updating
the journal tail sequence. However, this step is missed if the journal
is not located on the filesystem device. As a result, the filesystem may
become inconsistent following a power failure or system crash. Fix it by
ensuring that the filesystem device is flushed appropriately.
Fixes: 3339578f05 ("jbd2: cleanup journal tail after transaction commit")
Signed-off-by: Zhang Yi <yi.zhang@huawei.com>
Link: https://lore.kernel.org/r/20241203014407.805916-3-yi.zhang@huaweicloud.com
Reviewed-by: Jan Kara <jack@suse.cz>
Signed-off-by: Christian Brauner <brauner@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
commit f5cacdc6f2 upstream.
In __jbd2_log_wait_for_space(), we might call jbd2_cleanup_journal_tail()
to recover some journal space. But if an error occurs while executing
jbd2_cleanup_journal_tail() (e.g., an EIO), we don't stop waiting for free
space right away, we try other branches, and if j_committing_transaction
is NULL (i.e., the tid is 0), we will get the following complain:
============================================
JBD2: I/O error when updating journal superblock for sdd-8.
__jbd2_log_wait_for_space: needed 256 blocks and only had 217 space available
__jbd2_log_wait_for_space: no way to get more journal space in sdd-8
------------[ cut here ]------------
WARNING: CPU: 2 PID: 139804 at fs/jbd2/checkpoint.c:109 __jbd2_log_wait_for_space+0x251/0x2e0
Modules linked in:
CPU: 2 PID: 139804 Comm: kworker/u8:3 Not tainted 6.6.0+ #1
RIP: 0010:__jbd2_log_wait_for_space+0x251/0x2e0
Call Trace:
<TASK>
add_transaction_credits+0x5d1/0x5e0
start_this_handle+0x1ef/0x6a0
jbd2__journal_start+0x18b/0x340
ext4_dirty_inode+0x5d/0xb0
__mark_inode_dirty+0xe4/0x5d0
generic_update_time+0x60/0x70
[...]
============================================
So only if jbd2_cleanup_journal_tail() returns 1, i.e., there is nothing to
clean up at the moment, continue to try to reclaim free space in other ways.
Note that this fix relies on commit 6f6a6fda29 ("jbd2: fix ocfs2 corrupt
when updating journal superblock fails") to make jbd2_cleanup_journal_tail
return the correct error code.
Fixes: 8c3f25d895 ("jbd2: don't give up looking for space so easily in __jbd2_log_wait_for_space")
Cc: stable@kernel.org
Signed-off-by: Baokun Li <libaokun1@huawei.com>
Reviewed-by: Jan Kara <jack@suse.cz>
Link: https://patch.msgid.link/20240718115336.2554501-1-libaokun@huaweicloud.com
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 6db3c1575a upstream.
When a full journal commit is on-going, any fast commit has to be enqueued
into a different queue: FC_Q_STAGING instead of FC_Q_MAIN. This enqueueing
is done only once, i.e. if an inode is already queued in a previous fast
commit entry it won't be enqueued again. However, if a full commit starts
_after_ the inode is enqueued into FC_Q_MAIN, the next fast commit needs to
be done into FC_Q_STAGING. And this is not being done in function
ext4_fc_track_template().
This patch fixes the issue by re-enqueuing an inode into the STAGING queue
during the fast commit clean-up callback when doing a full commit. However,
to prevent a race with a fast-commit, the clean-up callback has to be called
with the journal locked.
This bug was found using fstest generic/047. This test creates several 32k
bytes files, sync'ing each of them after it's creation, and then shutting
down the filesystem. Some data may be loss in this operation; for example a
file may have it's size truncated to zero.
Suggested-by: Jan Kara <jack@suse.cz>
Signed-off-by: Luis Henriques (SUSE) <luis.henriques@linux.dev>
Reviewed-by: Jan Kara <jack@suse.cz>
Link: https://patch.msgid.link/20240717172220.14201-1-luis.henriques@linux.dev
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
Cc: stable@kernel.org
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
[ Upstream commit 0bab8db415 ]
We encountered a problem that the file system could not be mounted in
the power-off scenario. The analysis of the file system mirror shows that
only part of the data is written to the last commit block.
The valid data of the commit block is concentrated in the first sector.
However, the data of the entire block is involved in the checksum calculation.
For different hardware, the minimum atomic unit may be different.
If the checksum of a committed block is incorrect, clear the data except the
'commit_header' and then calculate the checksum. If the checkusm is correct,
it is considered that the block is partially committed, Then continue to replay
journal.
Signed-off-by: Ye Bin <yebin10@huawei.com>
Reviewed-by: Jan Kara <jack@suse.cz>
Link: https://patch.msgid.link/20240620072405.3533701-1-yebin@huaweicloud.com
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
Signed-off-by: Sasha Levin <sashal@kernel.org>
commit 27ba5b6731 upstream.
Commit 9f356e5a4f ("jbd2: Account descriptor blocks into
t_outstanding_credits") started to account descriptor blocks into
transactions outstanding credits. However it didn't appropriately
decrease the maximum amount of credits available to userspace. Thus if
the filesystem requests a transaction smaller than
j_max_transaction_buffers but large enough that when descriptor blocks
are added the size exceeds j_max_transaction_buffers, we confuse
add_transaction_credits() into thinking previous handles have grown the
transaction too much and enter infinite journal commit loop in
start_this_handle() -> add_transaction_credits() trying to create
transaction with enough credits available.
Fix the problem by properly accounting for transaction space reserved
for descriptor blocks when verifying requested transaction handle size.
CC: stable@vger.kernel.org
Fixes: 9f356e5a4f ("jbd2: Account descriptor blocks into t_outstanding_credits")
Reported-by: Alexander Coffin <alex.coffin@maticrobots.com>
Link: https://lore.kernel.org/all/CA+hUFcuGs04JHZ_WzA1zGN57+ehL2qmHOt5a7RMpo+rv6Vyxtw@mail.gmail.com
Signed-off-by: Jan Kara <jack@suse.cz>
Reviewed-by: Zhang Yi <yi.zhang@huawei.com>
Link: https://patch.msgid.link/20240624170127.3253-3-jack@suse.cz
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit e3a00a2378 upstream.
Instead of computing the number of descriptor blocks a transaction can
have each time we need it (which is currently when starting each
transaction but will become more frequent later) precompute the number
once during journal initialization together with maximum transaction
size. We perform the precomputation whenever journal feature set is
updated similarly as for computation of
journal->j_revoke_records_per_block.
CC: stable@vger.kernel.org
Signed-off-by: Jan Kara <jack@suse.cz>
Reviewed-by: Zhang Yi <yi.zhang@huawei.com>
Link: https://patch.msgid.link/20240624170127.3253-2-jack@suse.cz
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
[ Upstream commit 6c02757c93 ]
There's issue when do io test:
WARN: soft lockup - CPU#45 stuck for 11s! [jbd2/dm-2-8:4170]
CPU: 45 PID: 4170 Comm: jbd2/dm-2-8 Kdump: loaded Tainted: G OE
Call trace:
dump_backtrace+0x0/0x1a0
show_stack+0x24/0x30
dump_stack+0xb0/0x100
watchdog_timer_fn+0x254/0x3f8
__hrtimer_run_queues+0x11c/0x380
hrtimer_interrupt+0xfc/0x2f8
arch_timer_handler_phys+0x38/0x58
handle_percpu_devid_irq+0x90/0x248
generic_handle_irq+0x3c/0x58
__handle_domain_irq+0x68/0xc0
gic_handle_irq+0x90/0x320
el1_irq+0xcc/0x180
queued_spin_lock_slowpath+0x1d8/0x320
jbd2_journal_commit_transaction+0x10f4/0x1c78 [jbd2]
kjournald2+0xec/0x2f0 [jbd2]
kthread+0x134/0x138
ret_from_fork+0x10/0x18
Analyzed informations from vmcore as follows:
(1) There are about 5k+ jbd2_inode in 'commit_transaction->t_inode_list';
(2) Now is processing the 855th jbd2_inode;
(3) JBD2 task has TIF_NEED_RESCHED flag;
(4) There's no pags in address_space around the 855th jbd2_inode;
(5) There are some process is doing drop caches;
(6) Mounted with 'nodioread_nolock' option;
(7) 128 CPUs;
According to informations from vmcore we know 'journal->j_list_lock' spin lock
competition is fierce. So journal_finish_inode_data_buffers() maybe process
slowly. Theoretically, there is scheduling point in the filemap_fdatawait_range_keep_errors().
However, if inode's address_space has no pages which taged with PAGECACHE_TAG_WRITEBACK,
will not call cond_resched(). So may lead to soft lockup.
journal_finish_inode_data_buffers
filemap_fdatawait_range_keep_errors
__filemap_fdatawait_range
while (index <= end)
nr_pages = pagevec_lookup_range_tag(&pvec, mapping, &index, end, PAGECACHE_TAG_WRITEBACK);
if (!nr_pages)
break; --> If 'nr_pages' is equal zero will break, then will not call cond_resched()
for (i = 0; i < nr_pages; i++)
wait_on_page_writeback(page);
cond_resched();
To solve above issue, add scheduling point in the journal_finish_inode_data_buffers();
Signed-off-by: Ye Bin <yebin10@huawei.com>
Reviewed-by: Jan Kara <jack@suse.cz>
Link: https://lore.kernel.org/r/20231211112544.3879780-1-yebin10@huawei.com
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 6a3afb6ac6 ]
Current jbd2 only add REQ_SYNC for descriptor block, metadata log
buffer, commit buffer and superblock buffer, the submitted IO could be
throttled by writeback throttle in block layer, that could lead to
priority inversion in some cases. The log IO looks like a kind of high
priority metadata IO, so it should not be throttled by WBT like QOS
policies in block layer, let's add REQ_SYNC | REQ_IDLE to exempt from
writeback throttle, and also add REQ_META together indicates it's a
metadata IO.
Signed-off-by: Zhang Yi <yi.zhang@huawei.com>
Reviewed-by: Jan Kara <jack@suse.cz>
Link: https://lore.kernel.org/r/20231129114740.2686201-2-yi.zhang@huaweicloud.com
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
Signed-off-by: Sasha Levin <sashal@kernel.org>
commit 61187fce86 upstream.
JBD2 makes sure journal data is fallen on fs device by sync_blockdev(),
however, other process could intercept the EIO information from bdev's
mapping, which leads journal recovering successful even EIO occurs during
data written back to fs device.
We found this problem in our product, iscsi + multipath is chosen for block
device of ext4. Unstable network may trigger kpartx to rescan partitions in
device mapper layer. Detailed process is shown as following:
mount kpartx irq
jbd2_journal_recover
do_one_pass
memcpy(nbh->b_data, obh->b_data) // copy data to fs dev from journal
mark_buffer_dirty // mark bh dirty
vfs_read
generic_file_read_iter // dio
filemap_write_and_wait_range
__filemap_fdatawrite_range
do_writepages
block_write_full_folio
submit_bh_wbc
>> EIO occurs in disk <<
end_buffer_async_write
mark_buffer_write_io_error
mapping_set_error
set_bit(AS_EIO, &mapping->flags) // set!
filemap_check_errors
test_and_clear_bit(AS_EIO, &mapping->flags) // clear!
err2 = sync_blockdev
filemap_write_and_wait
filemap_check_errors
test_and_clear_bit(AS_EIO, &mapping->flags) // false
err2 = 0
Filesystem is mounted successfully even data from journal is failed written
into disk, and ext4/ocfs2 could become corrupted.
Fix it by comparing the wb_err state in fs block device before recovering
and after recovering.
A reproducer can be found in the kernel bugzilla referenced below.
Link: https://bugzilla.kernel.org/show_bug.cgi?id=217888
Cc: stable@vger.kernel.org
Signed-off-by: Zhihao Cheng <chengzhihao1@huawei.com>
Signed-off-by: Zhang Yi <yi.zhang@huawei.com>
Reviewed-by: Jan Kara <jack@suse.cz>
Link: https://lore.kernel.org/r/20230919012525.1783108-1-chengzhihao1@huawei.com
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Pull ext4 fixes from Ted Ts'o:
"Regression and bug fixes for ext4"
* tag 'ext4_for_linus-6.6-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/tytso/ext4:
ext4: fix rec_len verify error
ext4: do not let fstrim block system suspend
ext4: move setting of trimmed bit into ext4_try_to_trim_range()
jbd2: Fix memory leak in journal_init_common()
jbd2: Remove page size assumptions
buffer: Make bh_offset() work for compound pages
There is a memory leak reported by kmemleak:
unreferenced object 0xff11000105903b80 (size 64):
comm "mount", pid 3382, jiffies 4295032021 (age 27.826s)
hex dump (first 32 bytes):
04 00 00 00 00 00 00 00 01 00 00 00 00 00 00 00 ................
ff ff ff ff 00 00 00 00 00 00 00 00 00 00 00 00 ................
backtrace:
[<ffffffffae86ac40>] __kmalloc_node+0x50/0x160
[<ffffffffaf2486d8>] crypto_alloc_tfmmem.isra.0+0x38/0x110
[<ffffffffaf2498e5>] crypto_create_tfm_node+0x85/0x2f0
[<ffffffffaf24a92c>] crypto_alloc_tfm_node+0xfc/0x210
[<ffffffffaedde777>] journal_init_common+0x727/0x1ad0
[<ffffffffaede1715>] jbd2_journal_init_inode+0x2b5/0x500
[<ffffffffaed786b5>] ext4_load_and_init_journal+0x255/0x2440
[<ffffffffaed8b423>] ext4_fill_super+0x8823/0xa330
...
The root cause was traced to an error handing path in journal_init_common()
when malloc memory failed in register_shrinker(). The checksum driver is
used to reference to checksum algorithm via cryptoapi and the user should
release the memory when the driver is no longer needed or the journal
initialization failed.
Fix it by calling crypto_free_shash() on the "err_cleanup" error handing
path in journal_init_common().
Fixes: c30713084b ("jbd2: move load_superblock() into journal_init_common()")
Signed-off-by: Li Zetao <lizetao1@huawei.com>
Reviewed-by: Jan Kara <jack@suse.cz>
Reviewed-by: Zhang Yi <yi.zhang@huawei.com>
Reviewed-by: Ritesh Harjani (IBM) <ritesh.list@gmail.com>
Link: https://lore.kernel.org/r/20230911025138.983101-1-lizetao1@huawei.com
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
jbd2_alloc() allocates a buffer from slab when the block size is smaller
than PAGE_SIZE, and slab may be using a compound page. Before commit
8147c4c454, we set b_page to the precise page containing the buffer
and this code worked well. Now we set b_page to the head page of the
allocation, so we can no longer use offset_in_page(). While we could
do a 1:1 replacement with offset_in_folio(), use the more idiomatic
bh_offset() and the folio APIs to map the buffer.
This isn't enough to support a b_size larger than PAGE_SIZE on HIGHMEM
machines, but this is good enough to fix the actual bug we're seeing.
Fixes: 8147c4c454 ("jbd2: use a folio in jbd2_journal_write_metadata_buffer()")
Reported-by: Zorro Lang <zlang@kernel.org>
Signed-off-by: Ritesh Harjani (IBM) <ritesh.list@gmail.com>
[converted to be more folio]
Signed-off-by: Matthew Wilcox (Oracle) <willy@infradead.org>
Pull ext4 updates from Ted Ts'o:
"Many ext4 and jbd2 cleanups and bug fixes:
- Cleanups in the ext4 remount code when going to and from read-only
- Cleanups in ext4's multiblock allocator
- Cleanups in the jbd2 setup/mounting code paths
- Performance improvements when appending to a delayed allocation file
- Miscellaneous syzbot and other bug fixes"
* tag 'ext4_for_linus-6.6-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/tytso/ext4: (60 commits)
ext4: fix slab-use-after-free in ext4_es_insert_extent()
libfs: remove redundant checks of s_encoding
ext4: remove redundant checks of s_encoding
ext4: reject casefold inode flag without casefold feature
ext4: use LIST_HEAD() to initialize the list_head in mballoc.c
ext4: do not mark inode dirty every time when appending using delalloc
ext4: rename s_error_work to s_sb_upd_work
ext4: add periodic superblock update check
ext4: drop dio overwrite only flag and associated warning
ext4: add correct group descriptors and reserved GDT blocks to system zone
ext4: remove unused function declaration
ext4: mballoc: avoid garbage value from err
ext4: use sbi instead of EXT4_SB(sb) in ext4_mb_new_blocks_simple()
ext4: change the type of blocksize in ext4_mb_init_cache()
ext4: fix unttached inode after power cut with orphan file feature enabled
jbd2: correct the end of the journal recovery scan range
ext4: ext4_get_{dev}_journal return proper error value
ext4: cleanup ext4_get_dev_journal() and ext4_get_journal()
jbd2: jbd2_journal_init_{dev,inode} return proper error return value
jbd2: drop useless error tag in jbd2_journal_wipe()
...
We got a filesystem inconsistency issue below while running generic/475
I/O failure pressure test with fast_commit feature enabled.
Symlink /p3/d3/d1c/d6c/dd6/dce/l101 (inode #132605) is invalid.
If fast_commit feature is enabled, a special fast_commit journal area is
appended to the end of the normal journal area. The journal->j_last
point to the first unused block behind the normal journal area instead
of the whole log area, and the journal->j_fc_last point to the first
unused block behind the fast_commit journal area. While doing journal
recovery, do_one_pass(PASS_SCAN) should first scan the normal journal
area and turn around to the first block once it meet journal->j_last,
but the wrap() macro misuse the journal->j_fc_last, so the recovering
could not read the next magic block (commit block perhaps) and would end
early mistakenly and missing tN and every transaction after it in the
following example. Finally, it could lead to filesystem inconsistency.
| normal journal area | fast commit area |
+-------------------------------------------------+------------------+
| tN(rere) | tN+1 |~| tN-x |...| tN-1 | tN(front) | .... |
+-------------------------------------------------+------------------+
/ / /
start journal->j_last journal->j_fc_last
This patch fix it by use the correct ending journal->j_last.
Fixes: 5b849b5f96 ("jbd2: fast commit recovery path")
Cc: stable@kernel.org
Reported-by: Theodore Ts'o <tytso@mit.edu>
Link: https://lore.kernel.org/linux-ext4/20230613043120.GB1584772@mit.edu/
Signed-off-by: Zhang Yi <yi.zhang@huawei.com>
Reviewed-by: Jan Kara <jack@suse.cz>
Link: https://lore.kernel.org/r/20230626073322.3956567-1-yi.zhang@huaweicloud.com
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
If JBD2_FEATURE_INCOMPAT_FAST_COMMIT bit is set, it means the journal
have fast commit records need to recover, so the fast commit size
should not be too large, and the leftover normal journal size should
never less than JBD2_MIN_JOURNAL_BLOCKS. If it happens, the
journal->j_last is likely to be wrong and will probably lead to
incorrect journal recovery. So add a check into the
journal_check_superblock(), and drop the pointless check when
initializing the fastcommit parameters.
Signed-off-by: Zhang Yi <yi.zhang@huawei.com>
Reviewed-by: Jan Kara <jack@suse.cz>
Link: https://lore.kernel.org/r/20230811063610.2980059-8-yi.zhang@huaweicloud.com
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
Rename load_superblock() to journal_load_superblock(), move getting and
reading superblock from journal_init_common() and
journal_get_superblock() to this function, and also rename
journal_get_superblock() to journal_check_superblock(), make it a pure
check helper to check superblock validity from disk.
Signed-off-by: Zhang Yi <yi.zhang@huawei.com>
Reviewed-by: Jan Kara <jack@suse.cz>
Link: https://lore.kernel.org/r/20230811063610.2980059-7-yi.zhang@huaweicloud.com
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
Since load_superblock() has been moved to journal_init_common(), the
in-memory superblock structure is initialized and contains valid data
once the file system has a journal_t object, so it's safe to access it,
let's drop the call to journal_get_superblock() from
jbd2_journal_check_used_features() and also drop the setting/clearing of
the veirfy bit of the superblock buffer.
Signed-off-by: Zhang Yi <yi.zhang@huawei.com>
Reviewed-by: Jan Kara <jack@suse.cz>
Link: https://lore.kernel.org/r/20230811063610.2980059-4-yi.zhang@huaweicloud.com
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
Move the call to load_superblock() from jbd2_journal_load() and
jbd2_journal_wipe() early into journal_init_common(), the journal
superblock gets read and the in-memory journal_t structure gets
initialised after calling jbd2_journal_init_{dev,inode}, it's safe to
do following initialization according to it.
Signed-off-by: Zhang Yi <yi.zhang@huawei.com>
Reviewed-by: Jan Kara <jack@suse.cz>
Link: https://lore.kernel.org/r/20230811063610.2980059-3-yi.zhang@huaweicloud.com
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
Following process will corrupt ext4 image:
Step 1:
jbd2_journal_commit_transaction
__jbd2_journal_insert_checkpoint(jh, commit_transaction)
// Put jh into trans1->t_checkpoint_list
journal->j_checkpoint_transactions = commit_transaction
// Put trans1 into journal->j_checkpoint_transactions
Step 2:
do_get_write_access
test_clear_buffer_dirty(bh) // clear buffer dirty，set jbd dirty
__jbd2_journal_file_buffer(jh, transaction) // jh belongs to trans2
Step 3:
drop_cache
journal_shrink_one_cp_list
jbd2_journal_try_remove_checkpoint
if (!trylock_buffer(bh)) // lock bh, true
if (buffer_dirty(bh)) // buffer is not dirty
__jbd2_journal_remove_checkpoint(jh)
// remove jh from trans1->t_checkpoint_list
Step 4:
jbd2_log_do_checkpoint
trans1 = journal->j_checkpoint_transactions
// jh is not in trans1->t_checkpoint_list
jbd2_cleanup_journal_tail(journal) // trans1 is done
Step 5: Power cut, trans2 is not committed, jh is lost in next mounting.
Fix it by checking 'jh->b_transaction' before remove it from checkpoint.
Cc: stable@kernel.org
Fixes: 46f881b5b1 ("jbd2: fix a race when checking checkpoint buffer busy")
Signed-off-by: Zhihao Cheng <chengzhihao1@huawei.com>
Signed-off-by: Zhang Yi <yi.zhang@huawei.com>
Reviewed-by: Jan Kara <jack@suse.cz>
Link: https://lore.kernel.org/r/20230714025528.564988-3-yi.zhang@huaweicloud.com
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
journal_clean_one_cp_list() has been merged into
journal_shrink_one_cp_list(), but do chekpoint buffer cleanup from the
committing process is just a best effort, it should stop scan once it
meet a busy buffer, or else it will cause a lot of invalid buffer scan
and checks. We catch a performance regression when doing fs_mark tests
below.
Test cmd:
./fs_mark -d scratch -s 1024 -n 10000 -t 1 -D 100 -N 100
Before merging checkpoint buffer cleanup:
FSUse% Count Size Files/sec App Overhead
95 10000 1024 8304.9 49033
After merging checkpoint buffer cleanup:
FSUse% Count Size Files/sec App Overhead
95 10000 1024 7649.0 50012
FSUse% Count Size Files/sec App Overhead
95 10000 1024 2107.1 50871
After merging checkpoint buffer cleanup, the total loop count in
journal_shrink_one_cp_list() could be up to 6,261,600+ (50,000+ ~
100,000+ in general), most of them are invalid. This patch fix it
through passing 'shrink_type' into journal_shrink_one_cp_list() and add
a new 'SHRINK_BUSY_STOP' to indicate it should stop once meet a busy
buffer. After fix, the loop count descending back to 10,000+.
After this fix:
FSUse% Count Size Files/sec App Overhead
95 10000 1024 8558.4 49109
Cc: stable@kernel.org
Fixes: b98dba273a ("jbd2: remove journal_clean_one_cp_list()")
Signed-off-by: Zhang Yi <yi.zhang@huawei.com>
Reviewed-by: Jan Kara <jack@suse.cz>
Link: https://lore.kernel.org/r/20230714025528.564988-2-yi.zhang@huaweicloud.com
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
Before removing checkpoint buffer from the t_checkpoint_list, we have to
check both BH_Dirty and BH_Lock bits together to distinguish buffers
have not been or were being written back. But __cp_buffer_busy() checks
them separately, it first check lock state and then check dirty, the
window between these two checks could be raced by writing back
procedure, which locks buffer and clears buffer dirty before I/O
completes. So it cannot guarantee checkpointing buffers been written
back to disk if some error happens later. Finally, it may clean
checkpoint transactions and lead to inconsistent filesystem.
jbd2_journal_forget() and __journal_try_to_free_buffer() also have the
same problem (journal_unmap_buffer() escape from this issue since it's
running under the buffer lock), so fix them through introducing a new
helper to try holding the buffer lock and remove really clean buffer.
Link: https://bugzilla.kernel.org/show_bug.cgi?id=217490
Cc: stable@vger.kernel.org
Suggested-by: Jan Kara <jack@suse.cz>
Signed-off-by: Zhang Yi <yi.zhang@huawei.com>
Reviewed-by: Jan Kara <jack@suse.cz>
Link: https://lore.kernel.org/r/20230606135928.434610-6-yi.zhang@huaweicloud.com
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
Following process,
jbd2_journal_commit_transaction
// there are several dirty buffer heads in transaction->t_checkpoint_list
P1 wb_workfn
jbd2_log_do_checkpoint
if (buffer_locked(bh)) // false
__block_write_full_page
trylock_buffer(bh)
test_clear_buffer_dirty(bh)
if (!buffer_dirty(bh))
__jbd2_journal_remove_checkpoint(jh)
if (buffer_write_io_error(bh)) // false
>> bh IO error occurs <<
jbd2_cleanup_journal_tail
__jbd2_update_log_tail
jbd2_write_superblock
// The bh won't be replayed in next mount.
, which could corrupt the ext4 image, fetch a reproducer in [Link].
Since writeback process clears buffer dirty after locking buffer head,
we can fix it by try locking buffer and check dirtiness while buffer is
locked, the buffer head can be removed if it is neither dirty nor locked.
Link: https://bugzilla.kernel.org/show_bug.cgi?id=217490
Fixes: 470decc613 ("[PATCH] jbd2: initial copy of files from jbd")
Signed-off-by: Zhihao Cheng <chengzhihao1@huawei.com>
Signed-off-by: Zhang Yi <yi.zhang@huawei.com>
Reviewed-by: Jan Kara <jack@suse.cz>
Link: https://lore.kernel.org/r/20230606135928.434610-5-yi.zhang@huaweicloud.com
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
There is a long-standing metadata corruption issue that happens from
time to time, but it's very difficult to reproduce and analyse, benefit
from the JBD2_CYCLE_RECORD option, we found out that the problem is the
checkpointing process miss to write out some buffers which are raced by
another do_get_write_access(). Looks below for detail.
jbd2_log_do_checkpoint() //transaction X
//buffer A is dirty and not belones to any transaction
__buffer_relink_io() //move it to the IO list
__flush_batch()
write_dirty_buffer()
do_get_write_access()
clear_buffer_dirty
__jbd2_journal_file_buffer()
//add buffer A to a new transaction Y
lock_buffer(bh)
//doesn't write out
__jbd2_journal_remove_checkpoint()
//finish checkpoint except buffer A
//filesystem corrupt if the new transaction Y isn't fully write out.
Due to the t_checkpoint_list walking loop in jbd2_log_do_checkpoint()
have already handles waiting for buffers under IO and re-added new
transaction to complete commit, and it also removing cleaned buffers,
this makes sure the list will eventually get empty. So it's fine to
leave buffers on the t_checkpoint_list while flushing out and completely
stop using the t_checkpoint_io_list.
Cc: stable@vger.kernel.org
Suggested-by: Jan Kara <jack@suse.cz>
Signed-off-by: Zhang Yi <yi.zhang@huawei.com>
Tested-by: Zhihao Cheng <chengzhihao1@huawei.com>
Reviewed-by: Jan Kara <jack@suse.cz>
Link: https://lore.kernel.org/r/20230606135928.434610-2-yi.zhang@huaweicloud.com
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
Pull ext4 updates from Ted Ts'o:
"Various cleanups and bug fixes in ext4's extent status tree,
journalling, and block allocator subsystems.
Also improve performance for parallel DIO overwrites"
* tag 'ext4_for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tytso/ext4: (55 commits)
ext4: avoid updating the superblock on a r/o mount if not needed
jbd2: skip reading super block if it has been verified
ext4: fix to check return value of freeze_bdev() in ext4_shutdown()
ext4: refactoring to use the unified helper ext4_quotas_off()
ext4: turn quotas off if mount failed after enabling quotas
ext4: update doc about journal superblock description
ext4: add journal cycled recording support
jbd2: continue to record log between each mount
jbd2: remove j_format_version
jbd2: factor out journal initialization from journal_get_superblock()
jbd2: switch to check format version in superblock directly
jbd2: remove unused feature macros
ext4: ext4_put_super: Remove redundant checking for 'sbi->s_journal_bdev'
ext4: Fix reusing stale buffer heads from last failed mounting
ext4: allow concurrent unaligned dio overwrites
ext4: clean up mballoc criteria comments
ext4: make ext4_zeroout_es() return void
ext4: make ext4_es_insert_extent() return void
ext4: make ext4_es_insert_delayed_block() return void
ext4: make ext4_es_remove_extent() return void
...
We got a NULL pointer dereference issue below while running generic/475
I/O failure pressure test.
BUG: kernel NULL pointer dereference, address: 0000000000000000
#PF: supervisor write access in kernel mode
#PF: error_code(0x0002) - not-present page
PGD 0 P4D 0
Oops: 0002 [#1] PREEMPT SMP PTI
CPU: 1 PID: 15600 Comm: fsstress Not tainted 6.4.0-rc5-xfstests-00055-gd3ab1bca26b4 #190
RIP: 0010:jbd2_journal_set_features+0x13d/0x430
...
Call Trace:
<TASK>
? __die+0x23/0x60
? page_fault_oops+0xa4/0x170
? exc_page_fault+0x67/0x170
? asm_exc_page_fault+0x26/0x30
? jbd2_journal_set_features+0x13d/0x430
jbd2_journal_revoke+0x47/0x1e0
__ext4_forget+0xc3/0x1b0
ext4_free_blocks+0x214/0x2f0
ext4_free_branches+0xeb/0x270
ext4_ind_truncate+0x2bf/0x320
ext4_truncate+0x1e4/0x490
ext4_handle_inode_extension+0x1bd/0x2a0
? iomap_dio_complete+0xaf/0x1d0
The root cause is the journal super block had been failed to write out
due to I/O fault injection, it's uptodate bit was cleared by
end_buffer_write_sync() and didn't reset yet in jbd2_write_superblock().
And it raced by journal_get_superblock()->bh_read(), unfortunately, the
read IO is also failed, so the error handling in
journal_fail_superblock() unexpectedly clear the journal->j_sb_buffer,
finally lead to above NULL pointer dereference issue.
If the journal super block had been read and verified, there is no need
to call bh_read() read it again even if it has been failed to written
out. So the fix could be simply move buffer_verified(bh) in front of
bh_read(). Also remove a stale comment left in
jbd2_journal_check_used_features().
Fixes: 51bacdba23d8 ("jbd2: factor out journal initialization from journal_get_superblock()")
Reported-by: Theodore Ts'o <tytso@mit.edu>
Signed-off-by: Zhang Yi <yi.zhang@huawei.com>
Reviewed-by: Jan Kara <jack@suse.cz>
Link: https://lore.kernel.org/r/20230616015547.3155195-1-yi.zhang@huaweicloud.com
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
For a newly mounted file system, the journal committing thread always
record new transactions from the start of the journal area, no matter
whether the journal was clean or just has been recovered. So the logdump
code in debugfs cannot dump continuous logs between each mount, it is
disadvantageous to analysis corrupted file system image and locate the
file system inconsistency bugs.
If we get a corrupted file system in the running products and want to
find out what has happened, besides lookup the system log, one effective
way is to backtrack the journal log. But we may not always run e2fsck
before each mount and the default fsck -a mode also cannot always
checkout all inconsistencies, so it could left over some inconsistencies
into the next mount until we detect it. Finally, transactions in the
journal may probably discontinuous and some relatively new transactions
has been covered, it becomes hard to analyse. If we could record
transactions continuously between each mount, we could acquire more
useful info from the journal. Like this:
|Previous mount checkpointed/recovered logs|Current mount logs |
|{------}{---}{--------} ... {------}| ... |{======}{========}...000000|
And yes the journal area is limited and cannot record everything, the
problematic transaction may also be covered even if we do this, but
this is still useful for fuzzy tests and short-running products.
This patch save the head blocknr in the superblock after flushing the
journal or unmounting the file system, let the next mount could continue
to record new transaction behind it. This change is backward compatible
because the old kernel does not care about the head blocknr of the
journal. It is also fine if we mount a clean old image without valid
head blocknr, we fail back to set it to s_first just like before.
Finally, for the case of mount an unclean file system, we could also get
the journal head easily after scanning/replaying the journal, it will
continue to record new transaction after the recovered transactions.
Signed-off-by: Zhang Yi <yi.zhang@huawei.com>
Reviewed-by: Jan Kara <jack@suse.cz>
Link: https://lore.kernel.org/r/20230322013353.1843306-2-yi.zhang@huaweicloud.com
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
Theoretically possible that "%pg" will take all room for the j_devname
and hence the "-%lu" will go outside the boundary due to unconditional
sprintf() in use. To make this code more robust, replace two sequential
s*printf():s by a single call and then replace forbidden character.
It's possible to do this way, because '/' won't ever be in the result
of "-%lu".
Reviewed-by: Jan Kara <jack@suse.cz>
Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Signed-off-by: Kees Cook <keescook@chromium.org>
Link: https://lore.kernel.org/r/20230605170553.7835-2-andriy.shevchenko@linux.intel.com
When invalidating buffers under the partial tail page,
jbd2_journal_invalidate_folio() returns -EBUSY if the buffer is part of
the committing transaction as we cannot safely modify buffer state.
However if the buffer is already invalidated (due to previous
invalidation attempts from ext4_wait_for_tail_page_commit()), there's
nothing to do and there's no point in returning -EBUSY. This fixes
occasional warnings from ext4_journalled_invalidate_folio() triggered by
generic/051 fstest when blocksize < pagesize.
Fixes: 53e872681f ("ext4: fix deadlock in journal_unmap_buffer()")
Signed-off-by: Jan Kara <jack@suse.cz>
Link: https://lore.kernel.org/r/20230329154950.19720-1-jack@suse.cz
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
