IF YOU WOULD LIKE TO GET AN ACCOUNT, please write an
email to Administrator. User accounts are meant only to access repo
and report issues and/or generate pull requests.
This is a purpose-specific Git hosting for
BaseALT
projects. Thank you for your understanding!
Только зарегистрированные пользователи имеют доступ к сервису!
Для получения аккаунта, обратитесь к администратору.
[ Upstream commit f5ea16137a3fa2858620dc9084466491c128535f ]
There's a small window where a LOCK sent during a delegation return can
race with another OPEN on client, but the open stateid has not yet been
updated. In this case, the client doesn't handle the OLD_STATEID error
from the server and will lose this lock, emitting:
"NFS: nfs4_handle_delegation_recall_error: unhandled error -10024".
Fix this by sending the task through the nfs4 error handling in
nfs4_lock_done() when we may have to reconcile our stateid with what the
server believes it to be. For this case, the result is a retry of the
LOCK operation with the updated stateid.
Reported-by: Gonzalo Siero Humet <gsierohu@redhat.com>
Signed-off-by: Benjamin Coddington <bcodding@redhat.com>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
commit c8af247de385ce49afabc3bf1cf4fd455c94bfe8 upstream.
Syzbot reported a slab-out-of-bounds Write bug:
loop0: detected capacity change from 0 to 2048
==================================================================
BUG: KASAN: slab-out-of-bounds in udf_find_entry+0x8a5/0x14f0
fs/udf/namei.c:253
Write of size 105 at addr ffff8880123ff896 by task syz-executor323/3610
CPU: 0 PID: 3610 Comm: syz-executor323 Not tainted
6.1.0-rc2-syzkaller-00105-gb229b6ca5abb #0
Hardware name: Google Compute Engine/Google Compute Engine, BIOS
Google 10/11/2022
Call Trace:
<TASK>
__dump_stack lib/dump_stack.c:88 [inline]
dump_stack_lvl+0x1b1/0x28e lib/dump_stack.c:106
print_address_description+0x74/0x340 mm/kasan/report.c:284
print_report+0x107/0x1f0 mm/kasan/report.c:395
kasan_report+0xcd/0x100 mm/kasan/report.c:495
kasan_check_range+0x2a7/0x2e0 mm/kasan/generic.c:189
memcpy+0x3c/0x60 mm/kasan/shadow.c:66
udf_find_entry+0x8a5/0x14f0 fs/udf/namei.c:253
udf_lookup+0xef/0x340 fs/udf/namei.c:309
lookup_open fs/namei.c:3391 [inline]
open_last_lookups fs/namei.c:3481 [inline]
path_openat+0x10e6/0x2df0 fs/namei.c:3710
do_filp_open+0x264/0x4f0 fs/namei.c:3740
do_sys_openat2+0x124/0x4e0 fs/open.c:1310
do_sys_open fs/open.c:1326 [inline]
__do_sys_creat fs/open.c:1402 [inline]
__se_sys_creat fs/open.c:1396 [inline]
__x64_sys_creat+0x11f/0x160 fs/open.c:1396
do_syscall_x64 arch/x86/entry/common.c:50 [inline]
do_syscall_64+0x3d/0xb0 arch/x86/entry/common.c:80
entry_SYSCALL_64_after_hwframe+0x63/0xcd
RIP: 0033:0x7ffab0d164d9
Code: ff ff c3 66 2e 0f 1f 84 00 00 00 00 00 0f 1f 40 00 48 89 f8 48 89
f7 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 05 <48> 3d 01
f0 ff ff 73 01 c3 48 c7 c1 c0 ff ff ff f7 d8 64 89 01 48
RSP: 002b:00007ffe1a7e6bb8 EFLAGS: 00000246 ORIG_RAX: 0000000000000055
RAX: ffffffffffffffda RBX: 0000000000000000 RCX: 00007ffab0d164d9
RDX: 00007ffab0d164d9 RSI: 0000000000000000 RDI: 0000000020000180
RBP: 00007ffab0cd5a10 R08: 0000000000000000 R09: 0000000000000000
R10: 00005555573552c0 R11: 0000000000000246 R12: 00007ffab0cd5aa0
R13: 0000000000000000 R14: 0000000000000000 R15: 0000000000000000
</TASK>
Allocated by task 3610:
kasan_save_stack mm/kasan/common.c:45 [inline]
kasan_set_track+0x3d/0x60 mm/kasan/common.c:52
____kasan_kmalloc mm/kasan/common.c:371 [inline]
__kasan_kmalloc+0x97/0xb0 mm/kasan/common.c:380
kmalloc include/linux/slab.h:576 [inline]
udf_find_entry+0x7b6/0x14f0 fs/udf/namei.c:243
udf_lookup+0xef/0x340 fs/udf/namei.c:309
lookup_open fs/namei.c:3391 [inline]
open_last_lookups fs/namei.c:3481 [inline]
path_openat+0x10e6/0x2df0 fs/namei.c:3710
do_filp_open+0x264/0x4f0 fs/namei.c:3740
do_sys_openat2+0x124/0x4e0 fs/open.c:1310
do_sys_open fs/open.c:1326 [inline]
__do_sys_creat fs/open.c:1402 [inline]
__se_sys_creat fs/open.c:1396 [inline]
__x64_sys_creat+0x11f/0x160 fs/open.c:1396
do_syscall_x64 arch/x86/entry/common.c:50 [inline]
do_syscall_64+0x3d/0xb0 arch/x86/entry/common.c:80
entry_SYSCALL_64_after_hwframe+0x63/0xcd
The buggy address belongs to the object at ffff8880123ff800
which belongs to the cache kmalloc-256 of size 256
The buggy address is located 150 bytes inside of
256-byte region [ffff8880123ff800, ffff8880123ff900)
The buggy address belongs to the physical page:
page:ffffea000048ff80 refcount:1 mapcount:0 mapping:0000000000000000
index:0x0 pfn:0x123fe
head:ffffea000048ff80 order:1 compound_mapcount:0 compound_pincount:0
flags: 0xfff00000010200(slab|head|node=0|zone=1|lastcpupid=0x7ff)
raw: 00fff00000010200 ffffea00004b8500 dead000000000003 ffff888012041b40
raw: 0000000000000000 0000000080100010 00000001ffffffff 0000000000000000
page dumped because: kasan: bad access detected
page_owner tracks the page as allocated
page last allocated via order 0, migratetype Unmovable, gfp_mask 0x0(),
pid 1, tgid 1 (swapper/0), ts 1841222404, free_ts 0
create_dummy_stack mm/page_owner.c:67 [inline]
register_early_stack+0x77/0xd0 mm/page_owner.c:83
init_page_owner+0x3a/0x731 mm/page_owner.c:93
kernel_init_freeable+0x41c/0x5d5 init/main.c:1629
kernel_init+0x19/0x2b0 init/main.c:1519
page_owner free stack trace missing
Memory state around the buggy address:
ffff8880123ff780: fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc
ffff8880123ff800: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
>ffff8880123ff880: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 06
^
ffff8880123ff900: fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc
ffff8880123ff980: fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc
==================================================================
Fix this by changing the memory size allocated for copy_name from
UDF_NAME_LEN(254) to UDF_NAME_LEN_CS0(255), because the total length
(lfi) of subsequent memcpy can be up to 255.
CC: stable@vger.kernel.org
Reported-by: syzbot+69c9fdccc6dd08961d34@syzkaller.appspotmail.com
Fixes: 066b9cded00b ("udf: Use separate buffer for copying split names")
Signed-off-by: ZhangPeng <zhangpeng362@huawei.com>
Signed-off-by: Jan Kara <jack@suse.cz>
Link: https://lore.kernel.org/r/20221109013542.442790-1-zhangpeng362@huawei.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 8cccf05fe857a18ee26e20d11a8455a73ffd4efd upstream.
If a nilfs2 filesystem is downgraded to read-only due to metadata
corruption on disk and is remounted read/write, or if emergency read-only
remount is performed, detaching a log writer and synchronizing the
filesystem can be done at the same time.
In these cases, use-after-free of the log writer (hereinafter
nilfs->ns_writer) can happen as shown in the scenario below:
Task1 Task2
-------------------------------- ------------------------------
nilfs_construct_segment
nilfs_segctor_sync
init_wait
init_waitqueue_entry
add_wait_queue
schedule
nilfs_remount (R/W remount case)
nilfs_attach_log_writer
nilfs_detach_log_writer
nilfs_segctor_destroy
kfree
finish_wait
_raw_spin_lock_irqsave
__raw_spin_lock_irqsave
do_raw_spin_lock
debug_spin_lock_before <-- use-after-free
While Task1 is sleeping, nilfs->ns_writer is freed by Task2. After Task1
waked up, Task1 accesses nilfs->ns_writer which is already freed. This
scenario diagram is based on the Shigeru Yoshida's post [1].
This patch fixes the issue by not detaching nilfs->ns_writer on remount so
that this UAF race doesn't happen. Along with this change, this patch
also inserts a few necessary read-only checks with superblock instance
where only the ns_writer pointer was used to check if the filesystem is
read-only.
Link: https://syzkaller.appspot.com/bug?id=79a4c002e960419ca173d55e863bd09e8112df8b
Link: https://lkml.kernel.org/r/20221103141759.1836312-1-syoshida@redhat.com [1]
Link: https://lkml.kernel.org/r/20221104142959.28296-1-konishi.ryusuke@gmail.com
Signed-off-by: Ryusuke Konishi <konishi.ryusuke@gmail.com>
Reported-by: syzbot+f816fa82f8783f7a02bb@syzkaller.appspotmail.com
Reported-by: Shigeru Yoshida <syoshida@redhat.com>
Tested-by: Ryusuke Konishi <konishi.ryusuke@gmail.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 8ac932a4921a96ca52f61935dbba64ea87bbd5dc upstream.
A semaphore deadlock can occur if nilfs_get_block() detects metadata
corruption while locating data blocks and a superblock writeback occurs at
the same time:
task 1 task 2
------ ------
* A file operation *
nilfs_truncate()
nilfs_get_block()
down_read(rwsem A) <--
nilfs_bmap_lookup_contig()
... generic_shutdown_super()
nilfs_put_super()
* Prepare to write superblock *
down_write(rwsem B) <--
nilfs_cleanup_super()
* Detect b-tree corruption * nilfs_set_log_cursor()
nilfs_bmap_convert_error() nilfs_count_free_blocks()
__nilfs_error() down_read(rwsem A) <--
nilfs_set_error()
down_write(rwsem B) <--
*** DEADLOCK ***
Here, nilfs_get_block() readlocks rwsem A (= NILFS_MDT(dat_inode)->mi_sem)
and then calls nilfs_bmap_lookup_contig(), but if it fails due to metadata
corruption, __nilfs_error() is called from nilfs_bmap_convert_error()
inside the lock section.
Since __nilfs_error() calls nilfs_set_error() unless the filesystem is
read-only and nilfs_set_error() attempts to writelock rwsem B (=
nilfs->ns_sem) to write back superblock exclusively, hierarchical lock
acquisition occurs in the order rwsem A -> rwsem B.
Now, if another task starts updating the superblock, it may writelock
rwsem B during the lock sequence above, and can deadlock trying to
readlock rwsem A in nilfs_count_free_blocks().
However, there is actually no need to take rwsem A in
nilfs_count_free_blocks() because it, within the lock section, only reads
a single integer data on a shared struct with
nilfs_sufile_get_ncleansegs(). This has been the case after commit
aa474a220180 ("nilfs2: add local variable to cache the number of clean
segments"), that is, even before this bug was introduced.
So, this resolves the deadlock problem by just not taking the semaphore in
nilfs_count_free_blocks().
Link: https://lkml.kernel.org/r/20221029044912.9139-1-konishi.ryusuke@gmail.com
Fixes: e828949e5b42 ("nilfs2: call nilfs_error inside bmap routines")
Signed-off-by: Ryusuke Konishi <konishi.ryusuke@gmail.com>
Reported-by: syzbot+45d6ce7b7ad7ef455d03@syzkaller.appspotmail.com
Tested-by: Ryusuke Konishi <konishi.ryusuke@gmail.com>
Cc: <stable@vger.kernel.org> [2.6.38+
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit f376b45e861d8b7b34bf0eceeecfdd00dbe65cde upstream.
xfsaild is racy with respect to transaction abort and shutdown in
that the task can idle or exit with an empty AIL but buffers still
on the delwri queue. This was partly addressed by cancelling the
delwri queue before the task exits to prevent memory leaks, but it's
also possible for xfsaild to empty and idle with buffers on the
delwri queue. For example, a transaction that pins a buffer that
also happens to sit on the AIL delwri queue will explicitly remove
the associated log item from the AIL if the transaction aborts. The
side effect of this is an unmount hang in xfs_wait_buftarg() as the
associated buffers remain held by the delwri queue indefinitely.
This is reproduced on repeated runs of generic/531 with an fs format
(-mrmapbt=1 -bsize=1k) that happens to also reproduce transaction
aborts.
Update xfsaild to not idle until both the AIL and associated delwri
queue are empty and update the push code to continue delwri queue
submission attempts even when the AIL is empty. This allows the AIL
to eventually release aborted buffers stranded on the delwri queue
when they are unlocked by the associated transaction. This should
have no significant effect on normal runtime behavior because the
xfsaild currently idles only when the AIL is empty and in practice
the AIL is rarely empty with a populated delwri queue. The items
must be AIL resident to land in the queue in the first place and
generally aren't removed until writeback completes.
Note that the pre-existing delwri queue cancel logic in the exit
path is retained because task stop is external, could technically
come at any point, and xfsaild is still responsible to release its
buffer references before it exits.
Signed-off-by: Brian Foster <bfoster@redhat.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
Acked-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Chandan Babu R <chandan.babu@oracle.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 4750a171c3290f9bbebca16c6372db723a4cfa3b upstream.
[ For 5.4.y, SB_I_VERSION should be set in xfs_fs_remount() ]
The MS_I_VERSION mount flag is exposed via the VFS, as documented
in the mount manpages etc; see the iversion and noiversion mount
options in mount(8).
As a result, mount -o remount looks for this option in /proc/mounts
and will only send the I_VERSION flag back in during remount it it
is present. Since it's not there, a remount will /remove/ the
I_VERSION flag at the vfs level, and iversion functionality is lost.
xfs v5 superblocks intend to always have i_version enabled; it is
set as a default at mount time, but is lost during remount for the
reasons above.
The generic fix would be to expose this documented option in
/proc/mounts, but since that was rejected, fix it up again in the
xfs remount path instead, so that at least xfs won't suffer from
this misbehavior.
Signed-off-by: Eric Sandeen <sandeen@redhat.com>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com>
Acked-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Chandan Babu R <chandan.babu@oracle.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit cd647d5651c0b0deaa26c1acb9e1789437ba9bc7 upstream.
The page faultround path ->map_pages is implemented in XFS via
filemap_map_pages(). This function checks that pages found in page
cache lookups have not raced with truncate based invalidation by
checking page->mapping is correct and page->index is within EOF.
However, we've known for a long time that this is not sufficient to
protect against races with invalidations done by operations that do
not change EOF. e.g. hole punching and other fallocate() based
direct extent manipulations. The way we protect against these
races is we wrap the page fault operations in a XFS_MMAPLOCK_SHARED
lock so they serialise against fallocate and truncate before calling
into the filemap function that processes the fault.
Do the same for XFS's ->map_pages implementation to close this
potential data corruption issue.
Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Amir Goldstein <amir73il@gmail.com>
Reviewed-by: Brian Foster <bfoster@redhat.com>
Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
Acked-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Chandan Babu R <chandan.babu@oracle.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 00fd1d56dd08a8ceaa9e4ee1a41fefd9f6c6bc7d upstream.
The existing reflink remapping loop has some structural problems that
need addressing:
The biggest problem is that we create one transaction for each extent in
the source file without accounting for the number of mappings there are
for the same range in the destination file. In other words, we don't
know the number of remap operations that will be necessary and we
therefore cannot guess the block reservation required. On highly
fragmented filesystems (e.g. ones with active dedupe) we guess wrong,
run out of block reservation, and fail.
The second problem is that we don't actually use the bmap intents to
their full potential -- instead of calling bunmapi directly and having
to deal with its backwards operation, we could call the deferred ops
xfs_bmap_unmap_extent and xfs_refcount_decrease_extent instead. This
makes the frontend loop much simpler.
Solve all of these problems by refactoring the remapping loops so that
we only perform one remapping operation per transaction, and each
operation only tries to remap a single extent from source to dest.
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
Reviewed-by: Brian Foster <bfoster@redhat.com>
Reported-by: Edwin Török <edwin@etorok.net>
Tested-by: Edwin Török <edwin@etorok.net>
Acked-by: Darrick J. Wong <djwong@kernel.org>
[backported to 5.4.y - Tested-by above does not refer to the backport]
Signed-off-by: Chandan Babu R <chandan.babu@oracle.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 877f58f53684f14ca3202640f70592bf44890924 upstream.
[ Slightly modify fs/xfs/libxfs/xfs_rtbitmap.c & fs/xfs/xfs_reflink.c to
resolve merge conflict ]
The name of this predicate is a little misleading -- it decides if the
extent mapping is allocated and written. Change the name to be more
direct, as we're going to add a new predicate in the next patch.
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
Reviewed-by: Brian Foster <bfoster@redhat.com>
Acked-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Chandan Babu R <chandan.babu@oracle.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit f74681ba2006434be195402e0b15fc5763cddd7e upstream.
[Slightly modify xfs_trans_alloc() to fix a merge conflict due to missing
"atomic_inc(&mp->m_active_trans)" statement in v5.9 kernel]
The rmapbt extent swap algorithm remaps individual extents between
the source inode and the target to trigger reverse mapping metadata
updates. If either inode straddles a format or other bmap allocation
boundary, the individual unmap and map cycles can trigger repeated
bmap block allocations and frees as the extent count bounces back
and forth across the boundary. While net block usage is bound across
the swap operation, this behavior can prematurely exhaust the
transaction block reservation because it continuously drains as the
transaction rolls. Each allocation accounts against the reservation
and each free returns to global free space on transaction roll.
The previous workaround to this problem attempted to detect this
boundary condition and provide surplus block reservation to
acommodate it. This is insufficient because more remaps can occur
than implied by the extent counts; if start offset boundaries are
not aligned between the two inodes, for example.
To address this problem more generically and dynamically, add a
transaction accounting mode that returns freed blocks to the
transaction reservation instead of the superblock counters on
transaction roll and use it when the rmapbt based algorithm is
active. This allows the chain of remap transactions to preserve the
block reservation based own its own frees and prevent premature
exhaustion regardless of the remap pattern. Note that this is only
safe for superblocks with lazy sb accounting, but the latter is
required for v5 supers and the rmap feature depends on v5.
Fixes: b3fed434822d0 ("xfs: account format bouncing into rmapbt swapext tx reservation")
Root-caused-by: Darrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: Brian Foster <bfoster@redhat.com>
Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
Acked-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Chandan Babu R <chandan.babu@oracle.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 17a0bc9bd697f75cfdf9b378d5eb2d7409c91340 upstream.
The rec_len field in the directory entry has to be a multiple of 4. A
corrupted filesystem image can be used to hit a BUG() in
ext4_rec_len_to_disk(), called from make_indexed_dir().
------------[ cut here ]------------
kernel BUG at fs/ext4/ext4.h:2413!
...
RIP: 0010:make_indexed_dir+0x53f/0x5f0
...
Call Trace:
<TASK>
? add_dirent_to_buf+0x1b2/0x200
ext4_add_entry+0x36e/0x480
ext4_add_nondir+0x2b/0xc0
ext4_create+0x163/0x200
path_openat+0x635/0xe90
do_filp_open+0xb4/0x160
? __create_object.isra.0+0x1de/0x3b0
? _raw_spin_unlock+0x12/0x30
do_sys_openat2+0x91/0x150
__x64_sys_open+0x6c/0xa0
do_syscall_64+0x3c/0x80
entry_SYSCALL_64_after_hwframe+0x46/0xb0
The fix simply adds a call to ext4_check_dir_entry() to validate the
directory entry, returning -EFSCORRUPTED if the entry is invalid.
CC: stable@kernel.org
Link: https://bugzilla.kernel.org/show_bug.cgi?id=216540
Signed-off-by: Luís Henriques <lhenriques@suse.de>
Link: https://lore.kernel.org/r/20221012131330.32456-1-lhenriques@suse.de
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 2398091f9c2c8e0040f4f9928666787a3e8108a7 upstream.
The type of parameter generation has been u32 since the beginning,
however all callers pass a u64 generation, so unify the types to prevent
potential loss.
CC: stable@vger.kernel.org # 4.9+
Reviewed-by: Josef Bacik <josef@toxicpanda.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 8cc0072469723459dc6bd7beff81b2b3149f4cf4 upstream.
xfs_ifree_cluster() calls xfs_perag_get() at the beginning, but forgets to
call xfs_perag_put() in one failed path.
Add the missed function call to fix it.
Fixes: ce92464c180b ("xfs: make xfs_trans_get_buf return an error code")
Signed-off-by: Chuhong Yuan <hslester96@gmail.com>
Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
Acked-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Chandan Babu R <chandan.babu@oracle.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 1edd2c055dff9710b1e29d4df01902abb0a55f1f upstream.
During writeback, it's possible for the quota block reservation in
xfs_iomap_write_unwritten to fail with EDQUOT because we hit the quota
limit. This causes writeback errors for data that was already written
to disk, when it's not even guaranteed that the bmbt will expand to
exceed the quota limit. Irritatingly, this condition is reported to
userspace as EIO by fsync, which is confusing.
We wrote the data, so allow the reservation. That might put us slightly
above the hard limit, but it's better than losing data after a write.
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
Reviewed-by: Brian Foster <bfoster@redhat.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Acked-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Chandan Babu R <chandan.babu@oracle.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit c8d329f311c4d3d8f8e6dc5897ec235e37f48ae8 upstream.
Long ago, group & project quota were mutually exclusive, and so
when we turned on XFS_QMOPT_ENOSPC ("return ENOSPC if project quota
is exceeded") when project quota was enabled, we only needed to
disable it again for user quota.
When group & project quota got separated, this got missed, and as a
result if project quota is enabled and group quota is exceeded, the
error code returned is incorrectly returned as ENOSPC not EDQUOT.
Fix this by stripping XFS_QMOPT_ENOSPC out of flags for group
quota when we try to reserve the space.
Signed-off-by: Eric Sandeen <sandeen@redhat.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Brian Foster <bfoster@redhat.com>
Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
Acked-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Chandan Babu R <chandan.babu@oracle.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit dc3ffbb14060c943469d5e12900db3a60bc3fa64 upstream.
The error handling in xfs_trans_unreserve_and_mod_sb() is largely
incorrect - rolling back the changes in the transaction if only one
counter underruns makes all the other counters incorrect. We still
allow the change to proceed and committing the transaction, except
now we have multiple incorrect counters instead of a single
underflow.
Further, we don't actually report the error to the caller, so this
is completely silent except on debug kernels that will assert on
failure before we even get to the rollback code. Hence this error
handling is broken, untested, and largely unnecessary complexity.
Just remove it.
Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
Acked-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Chandan Babu R <chandan.babu@oracle.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 78bba5c812cc651cee51b64b786be926ab7fe2a9 upstream.
While QAing the new xfs_repair quotacheck code, I uncovered a quota
corruption bug resulting from a bad interaction between dquot buffer
initialization and quotacheck. The bug can be reproduced with the
following sequence:
# mkfs.xfs -f /dev/sdf
# mount /dev/sdf /opt -o usrquota
# su nobody -s /bin/bash -c 'touch /opt/barf'
# sync
# xfs_quota -x -c 'report -ahi' /opt
User quota on /opt (/dev/sdf)
Inodes
User ID Used Soft Hard Warn/Grace
---------- ---------------------------------
root 3 0 0 00 [------]
nobody 1 0 0 00 [------]
# xfs_io -x -c 'shutdown' /opt
# umount /opt
# mount /dev/sdf /opt -o usrquota
# touch /opt/man2
# xfs_quota -x -c 'report -ahi' /opt
User quota on /opt (/dev/sdf)
Inodes
User ID Used Soft Hard Warn/Grace
---------- ---------------------------------
root 1 0 0 00 [------]
nobody 1 0 0 00 [------]
# umount /opt
Notice how the initial quotacheck set the root dquot icount to 3
(rootino, rbmino, rsumino), but after shutdown -> remount -> recovery,
xfs_quota reports that the root dquot has only 1 icount. We haven't
deleted anything from the filesystem, which means that quota is now
under-counting. This behavior is not limited to icount or the root
dquot, but this is the shortest reproducer.
I traced the cause of this discrepancy to the way that we handle ondisk
dquot updates during quotacheck vs. regular fs activity. Normally, when
we allocate a disk block for a dquot, we log the buffer as a regular
(dquot) buffer. Subsequent updates to the dquots backed by that block
are done via separate dquot log item updates, which means that they
depend on the logged buffer update being written to disk before the
dquot items. Because individual dquots have their own LSN fields, that
initial dquot buffer must always be recovered.
However, the story changes for quotacheck, which can cause dquot block
allocations but persists the final dquot counter values via a delwri
list. Because recovery doesn't gate dquot buffer replay on an LSN, this
means that the initial dquot buffer can be replayed over the (newer)
contents that were delwritten at the end of quotacheck. In effect, this
re-initializes the dquot counters after they've been updated. If the
log does not contain any other dquot items to recover, the obsolete
dquot contents will not be corrected by log recovery.
Because quotacheck uses a transaction to log the setting of the CHKD
flags in the superblock, we skip quotacheck during the second mount
call, which allows the incorrect icount to remain.
Fix this by changing the ondisk dquot initialization function to use
ordered buffers to write out fresh dquot blocks if it detects that we're
running quotacheck. If the system goes down before quotacheck can
complete, the CHKD flags will not be set in the superblock and the next
mount will run quotacheck again, which can fix uninitialized dquot
buffers. This requires amending the defer code to maintaine ordered
buffer state across defer rolls for the sake of the dquot allocation
code.
For regular operations we preserve the current behavior since the dquot
items require properly initialized ondisk dquot records.
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
Reviewed-by: Brian Foster <bfoster@redhat.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Acked-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Chandan Babu R <chandan.babu@oracle.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit f28cef9e4daca11337cb9f144cdebedaab69d78c upstream.
The attr fork can transition from shortform to leaf format while
empty if the first xattr doesn't fit in shortform. While this empty
leaf block state is intended to be transient, it is technically not
due to the transactional implementation of the xattr set operation.
We historically have a couple of bandaids to work around this
problem. The first is to hold the buffer after the format conversion
to prevent premature writeback of the empty leaf buffer and the
second is to bypass the xattr count check in the verifier during
recovery. The latter assumes that the xattr set is also in the log
and will be recovered into the buffer soon after the empty leaf
buffer is reconstructed. This is not guaranteed, however.
If the filesystem crashes after the format conversion but before the
xattr set that induced it, only the format conversion may exist in
the log. When recovered, this creates a latent corrupted state on
the inode as any subsequent attempts to read the buffer fail due to
verifier failure. This includes further attempts to set xattrs on
the inode or attempts to destroy the attr fork, which prevents the
inode from ever being removed from the unlinked list.
To avoid this condition, accept that an empty attr leaf block is a
valid state and remove the count check from the verifier. This means
that on rare occasions an attr fork might exist in an unexpected
state, but is otherwise consistent and functional. Note that we
retain the logic to avoid racing with metadata writeback to reduce
the window where this can occur.
Signed-off-by: Brian Foster <bfoster@redhat.com>
Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Acked-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Chandan Babu R <chandan.babu@oracle.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
[ Upstream commit d37de92b38932d40e4a251e876cc388f9aee5f42 ]
In the test_no_shared_qgroup() and test_multiple_refs() qgroup self tests,
if we fail to add the tree ref, remove the extent item or remove the
extent ref, we are returning from the test function without freeing the
"old_roots" ulist that was allocated by the previous calls to
btrfs_find_all_roots(). Fix that by calling ulist_free() before returning.
Fixes: 442244c96332 ("btrfs: qgroup: Switch self test to extent-oriented qgroup mechanism.")
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 92876eec382a0f19f33d09d2c939e9ca49038ae5 ]
During backref walking, at find_parent_nodes(), if we are dealing with a
data extent and we get an error while resolving the indirect backrefs, at
resolve_indirect_refs(), or in the while loop that iterates over the refs
in the direct refs rbtree, we end up leaking the inode lists attached to
the direct refs we have in the direct refs rbtree that were not yet added
to the refs ulist passed as argument to find_parent_nodes(). Since they
were not yet added to the refs ulist and prelim_release() does not free
the lists, on error the caller can only free the lists attached to the
refs that were added to the refs ulist, all the remaining refs get their
inode lists never freed, therefore leaking their memory.
Fix this by having prelim_release() always free any attached inode list
to each ref found in the rbtree, and have find_parent_nodes() set the
ref's inode list to NULL once it transfers ownership of the inode list
to a ref added to the refs ulist passed to find_parent_nodes().
Fixes: 86d5f9944252 ("btrfs: convert prelimary reference tracking to use rbtrees")
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 5614dc3a47e3310fbc77ea3b67eaadd1c6417bf1 ]
During backref walking, at resolve_indirect_refs(), if we get an error
we jump to the 'out' label and call ulist_free() on the 'parents' ulist,
which frees all the elements in the ulist - however that does not free
any inode lists that may be attached to elements, through the 'aux' field
of a ulist node, so we end up leaking lists if we have any attached to
the unodes.
Fix this by calling free_leaf_list() instead of ulist_free() when we exit
from resolve_indirect_refs(). The static function free_leaf_list() is
moved up for this to be possible and it's slightly simplified by removing
unnecessary code.
Fixes: 3301958b7c1d ("Btrfs: add inodes before dropping the extent lock in find_all_leafs")
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit e59679f2b7e522ecad99974e5636291ffd47c184 ]
Currently, we are only guaranteed to send RECLAIM_COMPLETE if we have
open state to recover. Fix the client to always send RECLAIM_COMPLETE
after setting up the lease.
Fixes: fce5c838e133 ("nfs41: RECLAIM_COMPLETE functionality")
Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 5d917cba3201e5c25059df96c29252fd99c4f6a7 ]
If RECLAIM_COMPLETE sets the NFS4CLNT_BIND_CONN_TO_SESSION flag, then we
need to loop back in order to handle it.
Fixes: 0048fdd06614 ("NFSv4.1: RECLAIM_COMPLETE must handle NFS4ERR_CONN_NOT_BOUND_TO_SESSION")
Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
From: "Darrick J. Wong" <darrick.wong@oracle.com>
commit 5ffce3cc22a0e89813ed0c7162a68b639aef9ab6 upstream.
Commit 5833112df7e9 tried to make it so that a remap operation would
force the log out to disk if the filesystem is mounted with mandatory
synchronous writes. Unfortunately, that commit failed to handle the
case where the inode or the file descriptor require mandatory
synchronous writes.
Refactor the check into into a helper that will look for all three
conditions, and now we can treat reflink just like any other synchronous
write.
Fixes: 5833112df7e9 ("xfs: reflink should force the log out if mounted with wsync")
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
Reviewed-by: Brian Foster <bfoster@redhat.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Acked-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Chandan Babu R <chandan.babu@oracle.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
From: "Darrick J. Wong" <darrick.wong@oracle.com>
commit c97738a960a86081a147e7d436138e6481757445 upstream.
In commit 8d3d7e2b35ea, we changed xfs_qm_dqpurge to bail out if we
can't lock the dquot buf to flush the dquot. This prevents the AIL from
blocking on the dquot, but it also forgets to clear the FREEING flag on
its way out. A subsequent purge attempt will see the FREEING flag is
set and bail out, which leads to dqpurge_all failing to purge all the
dquots.
(copy-pasting from Dave Chinner's identical patch)
This was found by inspection after having xfs/305 hang 1 in ~50
iterations in a quotaoff operation:
[ 8872.301115] xfs_quota D13888 92262 91813 0x00004002
[ 8872.302538] Call Trace:
[ 8872.303193] __schedule+0x2d2/0x780
[ 8872.304108] ? do_raw_spin_unlock+0x57/0xd0
[ 8872.305198] schedule+0x6e/0xe0
[ 8872.306021] schedule_timeout+0x14d/0x300
[ 8872.307060] ? __next_timer_interrupt+0xe0/0xe0
[ 8872.308231] ? xfs_qm_dqusage_adjust+0x200/0x200
[ 8872.309422] schedule_timeout_uninterruptible+0x2a/0x30
[ 8872.310759] xfs_qm_dquot_walk.isra.0+0x15a/0x1b0
[ 8872.311971] xfs_qm_dqpurge_all+0x7f/0x90
[ 8872.313022] xfs_qm_scall_quotaoff+0x18d/0x2b0
[ 8872.314163] xfs_quota_disable+0x3a/0x60
[ 8872.315179] kernel_quotactl+0x7e2/0x8d0
[ 8872.316196] ? __do_sys_newstat+0x51/0x80
[ 8872.317238] __x64_sys_quotactl+0x1e/0x30
[ 8872.318266] do_syscall_64+0x46/0x90
[ 8872.319193] entry_SYSCALL_64_after_hwframe+0x44/0xa9
[ 8872.320490] RIP: 0033:0x7f46b5490f2a
[ 8872.321414] Code: Bad RIP value.
Returning -EAGAIN from xfs_qm_dqpurge() without clearing the
XFS_DQ_FREEING flag means the xfs_qm_dqpurge_all() code can never
free the dquot, and we loop forever waiting for the XFS_DQ_FREEING
flag to go away on the dquot that leaked it via -EAGAIN.
Fixes: 8d3d7e2b35ea ("xfs: trylock underlying buffer on dquot flush")
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
Reviewed-by: Allison Collins <allison.henderson@oracle.com>
Reviewed-by: Chandan Babu R <chandanrlinux@gmail.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Brian Foster <bfoster@redhat.com>
Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
Acked-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Chandan Babu R <chandan.babu@oracle.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
From: Brian Foster <bfoster@redhat.com>
commit 9c516e0e4554e8f26ab73d46cbc789d7d8db664d upstream.
The recent change to make insert range an atomic operation used the
incorrect transaction rolling mechanism. The explicit transaction
roll does not finish deferred operations. This means that intents
for rmapbt updates caused by extent shifts are not logged until the
final transaction commits. Thus if a crash occurs during an insert
range, log recovery might leave the rmapbt in an inconsistent state.
This was discovered by repeated runs of generic/455.
Update insert range to finish dfops on every shift iteration. This
is similar to collapse range and ensures that intents are logged
with the transactions that make associated changes.
Fixes: dd87f87d87fa ("xfs: rework insert range into an atomic operation")
Signed-off-by: Brian Foster <bfoster@redhat.com>
Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
Acked-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Chandan Babu R <chandan.babu@oracle.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Commit 258f669e7e88 ("mm: /proc/pid/smaps_rollup: convert to single value
seq_file") introduced a null-deref if there are no vma's in the task in
show_smaps_rollup.
Fixes: 258f669e7e88 ("mm: /proc/pid/smaps_rollup: convert to single value seq_file")
Signed-off-by: Seth Jenkins <sethjenkins@google.com>
Reviewed-by: Alexey Dobriyan <adobriyan@gmail.com>
Tested-by: Alexey Dobriyan <adobriyan@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
[ Upstream commit 943553ef9b51db303ab2b955c1025261abfdf6fb ]
During backref walking, when processing a delayed reference with a type of
BTRFS_TREE_BLOCK_REF_KEY, we have two bugs there:
1) We are accessing the delayed references extent_op, and its key, without
the protection of the delayed ref head's lock;
2) If there's no extent op for the delayed ref head, we end up with an
uninitialized key in the stack, variable 'tmp_op_key', and then pass
it to add_indirect_ref(), which adds the reference to the indirect
refs rb tree.
This is wrong, because indirect references should have a NULL key
when we don't have access to the key, and in that case they should be
added to the indirect_missing_keys rb tree and not to the indirect rb
tree.
This means that if have BTRFS_TREE_BLOCK_REF_KEY delayed ref resulting
from freeing an extent buffer, therefore with a count of -1, it will
not cancel out the corresponding reference we have in the extent tree
(with a count of 1), since both references end up in different rb
trees.
When using fiemap, where we often need to check if extents are shared
through shared subtrees resulting from snapshots, it means we can
incorrectly report an extent as shared when it's no longer shared.
However this is temporary because after the transaction is committed
the extent is no longer reported as shared, as running the delayed
reference results in deleting the tree block reference from the extent
tree.
Outside the fiemap context, the result is unpredictable, as the key was
not initialized but it's used when navigating the rb trees to insert
and search for references (prelim_ref_compare()), and we expect all
references in the indirect rb tree to have valid keys.
The following reproducer triggers the second bug:
$ cat test.sh
#!/bin/bash
DEV=/dev/sdj
MNT=/mnt/sdj
mkfs.btrfs -f $DEV
mount -o compress $DEV $MNT
# With a compressed 128M file we get a tree height of 2 (level 1 root).
xfs_io -f -c "pwrite -b 1M 0 128M" $MNT/foo
btrfs subvolume snapshot $MNT $MNT/snap
# Fiemap should output 0x2008 in the flags column.
# 0x2000 means shared extent
# 0x8 means encoded extent (because it's compressed)
echo
echo "fiemap after snapshot, range [120M, 120M + 128K):"
xfs_io -c "fiemap -v 120M 128K" $MNT/foo
echo
# Overwrite one extent and fsync to flush delalloc and COW a new path
# in the snapshot's tree.
#
# After this we have a BTRFS_DROP_DELAYED_REF delayed ref of type
# BTRFS_TREE_BLOCK_REF_KEY with a count of -1 for every COWed extent
# buffer in the path.
#
# In the extent tree we have inline references of type
# BTRFS_TREE_BLOCK_REF_KEY, with a count of 1, for the same extent
# buffers, so they should cancel each other, and the extent buffers in
# the fs tree should no longer be considered as shared.
#
echo "Overwriting file range [120M, 120M + 128K)..."
xfs_io -c "pwrite -b 128K 120M 128K" $MNT/snap/foo
xfs_io -c "fsync" $MNT/snap/foo
# Fiemap should output 0x8 in the flags column. The extent in the range
# [120M, 120M + 128K) is no longer shared, it's now exclusive to the fs
# tree.
echo
echo "fiemap after overwrite range [120M, 120M + 128K):"
xfs_io -c "fiemap -v 120M 128K" $MNT/foo
echo
umount $MNT
Running it before this patch:
$ ./test.sh
(...)
wrote 134217728/134217728 bytes at offset 0
128 MiB, 128 ops; 0.1152 sec (1.085 GiB/sec and 1110.5809 ops/sec)
Create a snapshot of '/mnt/sdj' in '/mnt/sdj/snap'
fiemap after snapshot, range [120M, 120M + 128K):
/mnt/sdj/foo:
EXT: FILE-OFFSET BLOCK-RANGE TOTAL FLAGS
0: [245760..246015]: 34304..34559 256 0x2008
Overwriting file range [120M, 120M + 128K)...
wrote 131072/131072 bytes at offset 125829120
128 KiB, 1 ops; 0.0001 sec (683.060 MiB/sec and 5464.4809 ops/sec)
fiemap after overwrite range [120M, 120M + 128K):
/mnt/sdj/foo:
EXT: FILE-OFFSET BLOCK-RANGE TOTAL FLAGS
0: [245760..246015]: 34304..34559 256 0x2008
The extent in the range [120M, 120M + 128K) is still reported as shared
(0x2000 bit set) after overwriting that range and flushing delalloc, which
is not correct - an entire path was COWed in the snapshot's tree and the
extent is now only referenced by the original fs tree.
Running it after this patch:
$ ./test.sh
(...)
wrote 134217728/134217728 bytes at offset 0
128 MiB, 128 ops; 0.1198 sec (1.043 GiB/sec and 1068.2067 ops/sec)
Create a snapshot of '/mnt/sdj' in '/mnt/sdj/snap'
fiemap after snapshot, range [120M, 120M + 128K):
/mnt/sdj/foo:
EXT: FILE-OFFSET BLOCK-RANGE TOTAL FLAGS
0: [245760..246015]: 34304..34559 256 0x2008
Overwriting file range [120M, 120M + 128K)...
wrote 131072/131072 bytes at offset 125829120
128 KiB, 1 ops; 0.0001 sec (694.444 MiB/sec and 5555.5556 ops/sec)
fiemap after overwrite range [120M, 120M + 128K):
/mnt/sdj/foo:
EXT: FILE-OFFSET BLOCK-RANGE TOTAL FLAGS
0: [245760..246015]: 34304..34559 256 0x8
Now the extent is not reported as shared anymore.
So fix this by passing a NULL key pointer to add_indirect_ref() when
processing a delayed reference for a tree block if there's no extent op
for our delayed ref head with a defined key. Also access the extent op
only after locking the delayed ref head's lock.
The reproducer will be converted later to a test case for fstests.
Fixes: 86d5f994425252 ("btrfs: convert prelimary reference tracking to use rbtrees")
Fixes: a6dbceafb915e8 ("btrfs: Remove unused op_key var from add_delayed_refs")
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 4fc7b57228243d09c0d878873bf24fa64a90fa01 ]
When processing delayed data references during backref walking and we are
using a share context (we are being called through fiemap), whenever we
find a delayed data reference for an inode different from the one we are
interested in, then we immediately exit and consider the data extent as
shared. This is wrong, because:
1) This might be a DROP reference that will cancel out a reference in the
extent tree;
2) Even if it's an ADD reference, it may be followed by a DROP reference
that cancels it out.
In either case we should not exit immediately.
Fix this by never exiting when we find a delayed data reference for
another inode - instead add the reference and if it does not cancel out
other delayed reference, we will exit early when we call
extent_is_shared() after processing all delayed references. If we find
a drop reference, then signal the code that processes references from
the extent tree (add_inline_refs() and add_keyed_refs()) to not exit
immediately if it finds there a reference for another inode, since we
have delayed drop references that may cancel it out. In this later case
we exit once we don't have references in the rb trees that cancel out
each other and have two references for different inodes.
Example reproducer for case 1):
$ cat test-1.sh
#!/bin/bash
DEV=/dev/sdj
MNT=/mnt/sdj
mkfs.btrfs -f $DEV
mount $DEV $MNT
xfs_io -f -c "pwrite 0 64K" $MNT/foo
cp --reflink=always $MNT/foo $MNT/bar
echo
echo "fiemap after cloning:"
xfs_io -c "fiemap -v" $MNT/foo
rm -f $MNT/bar
echo
echo "fiemap after removing file bar:"
xfs_io -c "fiemap -v" $MNT/foo
umount $MNT
Running it before this patch, the extent is still listed as shared, it has
the flag 0x2000 (FIEMAP_EXTENT_SHARED) set:
$ ./test-1.sh
fiemap after cloning:
/mnt/sdj/foo:
EXT: FILE-OFFSET BLOCK-RANGE TOTAL FLAGS
0: [0..127]: 26624..26751 128 0x2001
fiemap after removing file bar:
/mnt/sdj/foo:
EXT: FILE-OFFSET BLOCK-RANGE TOTAL FLAGS
0: [0..127]: 26624..26751 128 0x2001
Example reproducer for case 2):
$ cat test-2.sh
#!/bin/bash
DEV=/dev/sdj
MNT=/mnt/sdj
mkfs.btrfs -f $DEV
mount $DEV $MNT
xfs_io -f -c "pwrite 0 64K" $MNT/foo
cp --reflink=always $MNT/foo $MNT/bar
# Flush delayed references to the extent tree and commit current
# transaction.
sync
echo
echo "fiemap after cloning:"
xfs_io -c "fiemap -v" $MNT/foo
rm -f $MNT/bar
echo
echo "fiemap after removing file bar:"
xfs_io -c "fiemap -v" $MNT/foo
umount $MNT
Running it before this patch, the extent is still listed as shared, it has
the flag 0x2000 (FIEMAP_EXTENT_SHARED) set:
$ ./test-2.sh
fiemap after cloning:
/mnt/sdj/foo:
EXT: FILE-OFFSET BLOCK-RANGE TOTAL FLAGS
0: [0..127]: 26624..26751 128 0x2001
fiemap after removing file bar:
/mnt/sdj/foo:
EXT: FILE-OFFSET BLOCK-RANGE TOTAL FLAGS
0: [0..127]: 26624..26751 128 0x2001
After this patch, after deleting bar in both tests, the extent is not
reported with the 0x2000 flag anymore, it gets only the flag 0x1
(which is FIEMAP_EXTENT_LAST):
$ ./test-1.sh
fiemap after cloning:
/mnt/sdj/foo:
EXT: FILE-OFFSET BLOCK-RANGE TOTAL FLAGS
0: [0..127]: 26624..26751 128 0x2001
fiemap after removing file bar:
/mnt/sdj/foo:
EXT: FILE-OFFSET BLOCK-RANGE TOTAL FLAGS
0: [0..127]: 26624..26751 128 0x1
$ ./test-2.sh
fiemap after cloning:
/mnt/sdj/foo:
EXT: FILE-OFFSET BLOCK-RANGE TOTAL FLAGS
0: [0..127]: 26624..26751 128 0x2001
fiemap after removing file bar:
/mnt/sdj/foo:
EXT: FILE-OFFSET BLOCK-RANGE TOTAL FLAGS
0: [0..127]: 26624..26751 128 0x1
These tests will later be converted to a test case for fstests.
Fixes: dc046b10c8b7d4 ("Btrfs: make fiemap not blow when you have lots of snapshots")
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
commit 759a7c6126eef5635506453e9b9d55a6a3ac2084 upstream.
Commit b1529a41f777 "ocfs2: should reclaim the inode if
'__ocfs2_mknod_locked' returns an error" tried to reclaim the claimed
inode if __ocfs2_mknod_locked() fails later. But this introduce a race,
the freed bit may be reused immediately by another thread, which will
update dinode, e.g. i_generation. Then iput this inode will lead to BUG:
inode->i_generation != le32_to_cpu(fe->i_generation)
We could make this inode as bad, but we did want to do operations like
wipe in some cases. Since the claimed inode bit can only affect that an
dinode is missing and will return back after fsck, it seems not a big
problem. So just leave it as is by revert the reclaim logic.
Link: https://lkml.kernel.org/r/20221017130227.234480-1-joseph.qi@linux.alibaba.com
Fixes: b1529a41f777 ("ocfs2: should reclaim the inode if '__ocfs2_mknod_locked' returns an error")
Signed-off-by: Joseph Qi <joseph.qi@linux.alibaba.com>
Reported-by: Yan Wang <wangyan122@huawei.com>
Cc: Mark Fasheh <mark@fasheh.com>
Cc: Joel Becker <jlbec@evilplan.org>
Cc: Junxiao Bi <junxiao.bi@oracle.com>
Cc: Changwei Ge <gechangwei@live.cn>
Cc: Gang He <ghe@suse.com>
Cc: Jun Piao <piaojun@huawei.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 28f4821b1b53e0649706912e810c6c232fc506f9 upstream.
In ocfs2_mknod(), if error occurs after dinode successfully allocated,
ocfs2 i_links_count will not be 0.
So even though we clear inode i_nlink before iput in error handling, it
still won't wipe inode since we'll refresh inode from dinode during inode
lock. So just like clear inode i_nlink, we clear ocfs2 i_links_count as
well. Also do the same change for ocfs2_symlink().
Link: https://lkml.kernel.org/r/20221017130227.234480-2-joseph.qi@linux.alibaba.com
Signed-off-by: Joseph Qi <joseph.qi@linux.alibaba.com>
Reported-by: Yan Wang <wangyan122@huawei.com>
Cc: Mark Fasheh <mark@fasheh.com>
Cc: Joel Becker <jlbec@evilplan.org>
Cc: Junxiao Bi <junxiao.bi@oracle.com>
Cc: Changwei Ge <gechangwei@live.cn>
Cc: Gang He <ghe@suse.com>
Cc: Jun Piao <piaojun@huawei.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit c7f87f3984cfa1e6d32806a715f35c5947ad9c09 upstream.
xlog_wait() on the CIL context can reference a freed context if the
waiter doesn't get scheduled before the CIL context is freed. This
can happen when a task is on the hard throttle and the CIL push
aborts due to a shutdown. This was detected by generic/019:
thread 1 thread 2
__xfs_trans_commit
xfs_log_commit_cil
<CIL size over hard throttle limit>
xlog_wait
schedule
xlog_cil_push_work
wake_up_all
<shutdown aborts commit>
xlog_cil_committed
kmem_free
remove_wait_queue
spin_lock_irqsave --> UAF
Fix it by moving the wait queue to the CIL rather than keeping it in
in the CIL context that gets freed on push completion. Because the
wait queue is now independent of the CIL context and we might have
multiple contexts in flight at once, only wake the waiters on the
push throttle when the context we are pushing is over the hard
throttle size threshold.
Fixes: 0e7ab7efe7745 ("xfs: Throttle commits on delayed background CIL push")
Reported-by: Yu Kuai <yukuai3@huawei.com>
Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Acked-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Chandan Babu R <chandan.babu@oracle.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit f0f7a674d4df1510d8ca050a669e1420cf7d7fab upstream.
[ Modify fs/xfs/xfs_super.c to include the changes at locations suitable for
5.4-lts kernel ]
Move the inode dirty data flushing to a workqueue so that multiple
threads can take advantage of a single thread's flushing work. The
ratelimiting technique used in bdd4ee4 was not successful, because
threads that skipped the inode flush scan due to ratelimiting would
ENOSPC early, which caused occasional (but noticeable) changes in
behavior and sporadic fstest regressions.
Therefore, make all the writer threads wait on a single inode flush,
which eliminates both the stampeding hordes of flushers and the small
window in which a write could fail with ENOSPC because it lost the
ratelimit race after even another thread freed space.
Fixes: c6425702f21e ("xfs: ratelimit inode flush on buffered write ENOSPC")
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
Reviewed-by: Brian Foster <bfoster@redhat.com>
Acked-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Chandan Babu R <chandan.babu@oracle.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 5833112df7e9a306af9af09c60127b92ed723962 upstream.
Reflink should force the log out to disk if the filesystem was mounted
with wsync, the same as most other operations in xfs.
[Note: XFS_MOUNT_WSYNC is set when the admin mounts the filesystem
with either the 'wsync' or 'sync' mount options, which effectively means
that we're classifying reflink/dedupe as IO operations and making them
synchronous when required.]
Fixes: 3fc9f5e409319 ("xfs: remove xfs_reflink_remap_range")
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Brian Foster <bfoster@redhat.com>
[darrick: add more to the changelog]
Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
Acked-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Chandan Babu R <chandan.babu@oracle.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 54fbdd1035e3a4e4f4082c335b095426cdefd092 upstream.
Create a new helper to force the log up to the last LSN touching an
inode.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Brian Foster <bfoster@redhat.com>
Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
Acked-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Chandan Babu R <chandan.babu@oracle.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 8d3d7e2b35ea7d91d6e085c93b5efecfb0fba307 upstream.
A dquot flush currently blocks on the buffer lock for the underlying
dquot buffer. In turn, this causes xfsaild to block rather than
continue processing other items in the meantime. Update
xfs_qm_dqflush() to trylock the buffer, similar to how inode buffers
are handled, and return -EAGAIN if the lock fails. Fix up any
callers that don't currently handle the error properly.
Signed-off-by: Brian Foster <bfoster@redhat.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
Acked-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Chandan Babu R <chandan.babu@oracle.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 5cc3c006eb45524860c4d1dd4dd7ad4a506bf3f5 upstream.
[ Modify fs/xfs/xfs_log.c to include the changes at locations suitable for
5.4-lts kernel ]
In commit f467cad95f5e3, I added the ability to force a recalculation of
the filesystem summary counters if they seemed incorrect. This was done
(not entirely correctly) by tweaking the log code to write an unmount
record without the UMOUNT_TRANS flag set. At next mount, the log
recovery code will fail to find the unmount record and go into recovery,
which triggers the recalculation.
What actually gets written to the log is what ought to be an unmount
record, but without any flags set to indicate what kind of record it
actually is. This worked to trigger the recalculation, but we shouldn't
write bogus log records when we could simply write nothing.
Fixes: f467cad95f5e3 ("xfs: force summary counter recalc at next mount")
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Brian Foster <bfoster@redhat.com>
Acked-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Chandan Babu R <chandan.babu@oracle.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 8eb807bd839938b45bf7a97f0568d2a845ba6929 upstream.
We currently wake anything waiting on the log tail to move whenever
the log item at the tail of the log is removed. Historically this
was fine behaviour because there were very few items at any given
LSN. But with delayed logging, there may be thousands of items at
any given LSN, and we can't move the tail until they are all gone.
Hence if we are removing them in near tail-first order, we might be
waking up processes waiting on the tail LSN to change (e.g. log
space waiters) repeatedly without them being able to make progress.
This also occurs with the new sync push waiters, and can result in
thousands of spurious wakeups every second when under heavy direct
reclaim pressure.
To fix this, check that the tail LSN has actually changed on the
AIL before triggering wakeups. This will reduce the number of
spurious wakeups when doing bulk AIL removal and make this code much
more efficient.
Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Brian Foster <bfoster@redhat.com>
Reviewed-by: Allison Collins <allison.henderson@oracle.com>
Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
Acked-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Chandan Babu R <chandan.babu@oracle.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 4165994ac9672d91134675caa6de3645a9ace6c8 upstream.
Factor the common AIL deletion code that does all the wakeups into a
helper so we only have one copy of this somewhat tricky code to
interface with all the wakeups necessary when the LSN of the log
tail changes.
Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Allison Collins <allison.henderson@oracle.com>
Reviewed-by: Brian Foster <bfoster@redhat.com>
Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
Acked-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Chandan Babu R <chandan.babu@oracle.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 0e7ab7efe77451cba4cbecb6c9f5ef83cf32b36b upstream.
In certain situations the background CIL push can be indefinitely
delayed. While we have workarounds from the obvious cases now, it
doesn't solve the underlying issue. This issue is that there is no
upper limit on the CIL where we will either force or wait for
a background push to start, hence allowing the CIL to grow without
bound until it consumes all log space.
To fix this, add a new wait queue to the CIL which allows background
pushes to wait for the CIL context to be switched out. This happens
when the push starts, so it will allow us to block incoming
transaction commit completion until the push has started. This will
only affect processes that are running modifications, and only when
the CIL threshold has been significantly overrun.
This has no apparent impact on performance, and doesn't even trigger
until over 45 million inodes had been created in a 16-way fsmark
test on a 2GB log. That was limiting at 64MB of log space used, so
the active CIL size is only about 3% of the total log in that case.
The concurrent removal of those files did not trigger the background
sleep at all.
Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Allison Collins <allison.henderson@oracle.com>
Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
Acked-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Chandan Babu R <chandan.babu@oracle.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 108a42358a05312b2128533c6462a3fdeb410bdf upstream.
The current CIL size aggregation limit is 1/8th the log size. This
means for large logs we might be aggregating at least 250MB of dirty objects
in memory before the CIL is flushed to the journal. With CIL shadow
buffers sitting around, this means the CIL is often consuming >500MB
of temporary memory that is all allocated under GFP_NOFS conditions.
Flushing the CIL can take some time to do if there is other IO
ongoing, and can introduce substantial log force latency by itself.
It also pins the memory until the objects are in the AIL and can be
written back and reclaimed by shrinkers. Hence this threshold also
tends to determine the minimum amount of memory XFS can operate in
under heavy modification without triggering the OOM killer.
Modify the CIL space limit to prevent such huge amounts of pinned
metadata from aggregating. We can have 2MB of log IO in flight at
once, so limit aggregation to 16x this size. This threshold was
chosen as it little impact on performance (on 16-way fsmark) or log
traffic but pins a lot less memory on large logs especially under
heavy memory pressure. An aggregation limit of 8x had 5-10%
performance degradation and a 50% increase in log throughput for
the same workload, so clearly that was too small for highly
concurrent workloads on large logs.
This was found via trace analysis of AIL behaviour. e.g. insertion
from a single CIL flush:
xfs_ail_insert: old lsn 0/0 new lsn 1/3033090 type XFS_LI_INODE flags IN_AIL
$ grep xfs_ail_insert /mnt/scratch/s.t |grep "new lsn 1/3033090" |wc -l
1721823
$
So there were 1.7 million objects inserted into the AIL from this
CIL checkpoint, the first at 2323.392108, the last at 2325.667566 which
was the end of the trace (i.e. it hadn't finished). Clearly a major
problem.
Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Brian Foster <bfoster@redhat.com>
Reviewed-by: Allison Collins <allison.henderson@oracle.com>
Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
Acked-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Chandan Babu R <chandan.babu@oracle.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 5885539f0af371024d07afd14974bfdc3fff84c5 upstream.
When quotacheck runs, it zeroes all the timer fields in every dquot.
Unfortunately, it also does this to the root dquot, which erases any
preconfigured grace intervals and warning limits that the administrator
may have set. Worse yet, the incore copies of those variables remain
set. This cache coherence problem manifests itself as the grace
interval mysteriously being reset back to the defaults at the /next/
mount.
Fix it by not resetting the root disk dquot's timer and warning fields.
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Acked-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Chandan Babu R <chandan.babu@oracle.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 8a62714313391b9b2297d67c341b35edbf46c279 upstream.
AIL removal of the quotaoff start intent and free of both quotaoff
intents is currently limited to the ->iop_committed() handler of the
end intent. This executes when the end intent is committed to the
on-disk log and marks the completion of the operation. The problem
with this is it assumes the success of the operation. If a shutdown
or other error occurs during the quotaoff, it's possible for the
quotaoff task to exit without removing the start intent from the
AIL. This results in an unmount hang as the AIL cannot be emptied.
Further, no other codepath frees the intents and so this is also a
memory leak vector.
First, update the high level quotaoff error path to directly remove
and free the quotaoff start intent if it still exists in the AIL at
the time of the error. Next, update both of the start and end
quotaoff intents with an ->iop_release() callback to properly handle
transaction abort.
This means that If the quotaoff start transaction aborts, it frees
the start intent in the transaction commit path. If the filesystem
shuts down before the end transaction allocates, the quotaoff
sequence removes and frees the start intent. If the end transaction
aborts, it removes the start intent and frees both. This ensures
that a shutdown does not result in a hung unmount and that memory is
not leaked regardless of when a quotaoff error occurs.
Signed-off-by: Brian Foster <bfoster@redhat.com>
Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Acked-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Chandan Babu R <chandan.babu@oracle.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 854f82b1f6039a418b7d1407513f8640e05fd73f upstream.
AIL removal of the quotaoff start intent and free of both intents is
hardcoded to the ->iop_committed() handler of the end intent. Factor
out the start intent handling code so it can be used in a future
patch to properly handle quotaoff errors. Use xfs_trans_ail_remove()
instead of the _delete() variant to acquire the AIL lock and also
handle cases where an intent might not reside in the AIL at the
time of a failure.
Signed-off-by: Brian Foster <bfoster@redhat.com>
Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Acked-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Chandan Babu R <chandan.babu@oracle.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 1cc95e6f0d7cfd61c9d3c5cdd4e7345b173f764f upstream.
Signed-off-by: Pavel Reichl <preichl@redhat.com>
Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com>
[darrick: fix typo in subject line]
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
Acked-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Chandan Babu R <chandan.babu@oracle.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Benjamin Coddington
ZhangPeng
Zhang Xiaoxu
Ryusuke Konishi
Brian Foster
Eric Sandeen
Dave Chinner
Darrick J. Wong
Luís Henriques
Ye Bin
Miklos Szeredi
David Sterba
Chuhong Yuan
Dave Chinner
Filipe Manana
Trond Myklebust
Chandan Babu R
Christian A. Ehrhardt
Seth Jenkins
Joseph Qi
Christoph Hellwig
Pavel Reichl