| Age | Commit message (Collapse) | Author |
|---|
|
commit 827aa18e7b903c5ff3b3cd8fec328a99b1dbd411 upstream.
When initializing the security xattrs, we are holding a transaction handle
therefore we need to use a GFP_NOFS context in order to avoid a deadlock
with reclaim in case it's triggered.
Fixes: 39a27ec1004e8 ("btrfs: use GFP_KERNEL for xattr and acl allocations")
Reviewed-by: Nikolay Borisov <nborisov@suse.com>
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
commit 9a6f209e36500efac51528132a3e3083586eda5f upstream.
If the quota enable and snapshot creation ioctls are called concurrently
we can get into a deadlock where the task enabling quotas will deadlock
on the fs_info->qgroup_ioctl_lock mutex because it attempts to lock it
twice, or the task creating a snapshot tries to commit the transaction
while the task enabling quota waits for the former task to commit the
transaction while holding the mutex. The following time diagrams show how
both cases happen.
First scenario:
CPU 0 CPU 1
btrfs_ioctl()
btrfs_ioctl_quota_ctl()
btrfs_quota_enable()
mutex_lock(fs_info->qgroup_ioctl_lock)
btrfs_start_transaction()
btrfs_ioctl()
btrfs_ioctl_snap_create_v2
create_snapshot()
--> adds snapshot to the
list pending_snapshots
of the current
transaction
btrfs_commit_transaction()
create_pending_snapshots()
create_pending_snapshot()
qgroup_account_snapshot()
btrfs_qgroup_inherit()
mutex_lock(fs_info->qgroup_ioctl_lock)
--> deadlock, mutex already locked
by this task at
btrfs_quota_enable()
Second scenario:
CPU 0 CPU 1
btrfs_ioctl()
btrfs_ioctl_quota_ctl()
btrfs_quota_enable()
mutex_lock(fs_info->qgroup_ioctl_lock)
btrfs_start_transaction()
btrfs_ioctl()
btrfs_ioctl_snap_create_v2
create_snapshot()
--> adds snapshot to the
list pending_snapshots
of the current
transaction
btrfs_commit_transaction()
--> waits for task at
CPU 0 to release
its transaction
handle
btrfs_commit_transaction()
--> sees another task started
the transaction commit first
--> releases its transaction
handle
--> waits for the transaction
commit to be completed by
the task at CPU 1
create_pending_snapshot()
qgroup_account_snapshot()
btrfs_qgroup_inherit()
mutex_lock(fs_info->qgroup_ioctl_lock)
--> deadlock, task at CPU 0
has the mutex locked but
it is waiting for us to
finish the transaction
commit
So fix this by setting the quota enabled flag in fs_info after committing
the transaction at btrfs_quota_enable(). This ends up serializing quota
enable and snapshot creation as if the snapshot creation happened just
before the quota enable request. The quota rescan task, scheduled after
committing the transaction in btrfs_quote_enable(), will do the accounting.
Fixes: 6426c7ad697d ("btrfs: qgroup: Fix qgroup accounting when creating snapshot")
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
commit 5a8067c0d17feb7579db0476191417b441a8996e upstream.
The available allocation bits members from struct btrfs_fs_info are
protected by a sequence lock, and when starting balance we access them
incorrectly in two different ways:
1) In the read sequence lock loop at btrfs_balance() we use the values we
read from fs_info->avail_*_alloc_bits and we can immediately do actions
that have side effects and can not be undone (printing a message and
jumping to a label). This is wrong because a retry might be needed, so
our actions must not have side effects and must be repeatable as long
as read_seqretry() returns a non-zero value. In other words, we were
essentially ignoring the sequence lock;
2) Right below the read sequence lock loop, we were reading the values
from avail_metadata_alloc_bits and avail_data_alloc_bits without any
protection from concurrent writers, that is, reading them outside of
the read sequence lock critical section.
So fix this by making sure we only read the available allocation bits
while in a read sequence lock critical section and that what we do in the
critical section is repeatable (has nothing that can not be undone) so
that any eventual retry that is needed is handled properly.
Fixes: de98ced9e743 ("Btrfs: use seqlock to protect fs_info->avail_{data, metadata, system}_alloc_bits")
Fixes: 14506127979a ("btrfs: fix a bogus warning when converting only data or metadata")
Reviewed-by: Nikolay Borisov <nborisov@suse.com>
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
commit 191ce17876c9367819c4b0a25b503c0f6d9054d8 upstream.
The check for special (reserved) inode number checks in __ext4_iget()
was broken by commit 8a363970d1dc: ("ext4: avoid declaring fs
inconsistent due to invalid file handles"). This was caused by a
botched reversal of the sense of the flag now known as
EXT4_IGET_SPECIAL (when it was previously named EXT4_IGET_NORMAL).
Fix the logic appropriately.
Fixes: 8a363970d1dc ("ext4: avoid declaring fs inconsistent...")
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
Reported-by: Dan Carpenter <dan.carpenter@oracle.com>
Cc: stable@kernel.org
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
commit 95cb67138746451cc84cf8e516e14989746e93b0 upstream.
We already using mapping_set_error() in fs/ext4/page_io.c, so all we
need to do is to use file_check_and_advance_wb_err() when handling
fsync() requests in ext4_sync_file().
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
Cc: stable@kernel.org
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
commit ad211f3e94b314a910d4af03178a0b52a7d1ee0a upstream.
In no-journal mode, we previously used __generic_file_fsync() in
no-journal mode. This triggers a lockdep warning, and in addition,
it's not safe to depend on the inode writeback mechanism in the case
ext4. We can solve both problems by calling ext4_write_inode()
directly.
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
Cc: stable@kernel.org
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
commit e86807862e6880809f191c4cea7f88a489f0ed34 upstream.
The xfstests generic/475 test switches the underlying device with
dm-error while running a stress test. This results in a large number
of file system errors, and since we can't lock the buffer head when
marking the superblock dirty in the ext4_grp_locked_error() case, it's
possible the superblock to be !buffer_uptodate() without
buffer_write_io_error() being true.
We need to set buffer_uptodate() before we call mark_buffer_dirty() or
this will trigger a WARN_ON. It's safe to do this since the
superblock must have been properly read into memory or the mount would
have been successful. So if buffer_uptodate() is not set, we can
safely assume that this happened due to a failed attempt to write the
superblock.
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
Cc: stable@vger.kernel.org
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
commit 2b08b1f12cd664dc7d5c84ead9ff25ae97ad5491 upstream.
The ext4_inline_data_fiemap() function calls fiemap_fill_next_extent()
while still holding the xattr semaphore. This is not necessary and it
triggers a circular lockdep warning. This is because
fiemap_fill_next_extent() could trigger a page fault when it writes
into page which triggers a page fault. If that page is mmaped from
the inline file in question, this could very well result in a
deadlock.
This problem can be reproduced using generic/519 with a file system
configuration which has the inline_data feature enabled.
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
Cc: stable@kernel.org
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
commit 812c0cab2c0dfad977605dbadf9148490ca5d93f upstream.
There are enough credits reserved for most dioread_nolock writes;
however, if the extent tree is sufficiently deep, and/or quota is
enabled, the code was not allowing for all eventualities when
reserving journal credits for the unwritten extent conversion.
This problem can be seen using xfstests ext4/034:
WARNING: CPU: 1 PID: 257 at fs/ext4/ext4_jbd2.c:271 __ext4_handle_dirty_metadata+0x10c/0x180
Workqueue: ext4-rsv-conversion ext4_end_io_rsv_work
RIP: 0010:__ext4_handle_dirty_metadata+0x10c/0x180
...
EXT4-fs: ext4_free_blocks:4938: aborting transaction: error 28 in __ext4_handle_dirty_metadata
EXT4: jbd2_journal_dirty_metadata failed: handle type 11 started at line 4921, credits 4/0, errcode -28
EXT4-fs error (device dm-1) in ext4_free_blocks:4950: error 28
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
Cc: stable@kernel.org
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
commit b9a74cde94957d82003fb9f7ab4777938ca851cd upstream.
If maxBuf is small but non-zero, it could result in a zero sized lock
element array which we would then try and access OOB.
Signed-off-by: Ross Lagerwall <ross.lagerwall@citrix.com>
Signed-off-by: Steve French <stfrench@microsoft.com>
CC: Stable <stable@vger.kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
commit 8544f4aa9dd19a04d1244dae10feecc813ccf175 upstream.
In SMB3 protocol every part of the compound chain consumes credits
individually, so we need to call wait_for_free_credits() for each
of the PDUs in the chain. If an operation is interrupted, we must
ensure we return all credits taken from the server structure back.
Without this patch server can sometimes disconnect the session
due to credit mismatches, especially when first operation(s)
are large writes.
Signed-off-by: Pavel Shilovsky <pshilov@microsoft.com>
Signed-off-by: Steve French <stfrench@microsoft.com>
CC: Stable <stable@vger.kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
commit ee13919c2e8d1f904e035ad4b4239029a8994131 upstream.
Currently we hide EINTR code returned from sock_sendmsg()
and return 0 instead. This makes a caller think that we
successfully completed the network operation which is not
true. Fix this by properly returning EINTR to callers.
Cc: <stable@vger.kernel.org>
Signed-off-by: Pavel Shilovsky <pshilov@microsoft.com>
Reviewed-by: Jeff Layton <jlayton@kernel.org>
Signed-off-by: Steve French <stfrench@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
commit 33fa5c8b8a7dbe6353a56eaa654b790348890d42 upstream.
Currently we reset the number of total credits granted by the server
to 1 if the server didn't grant us anything int the response. This
violates the SMB3 protocol - we need to trust the server and use
the credit values from the response. Fix this by removing the
corresponding code.
Signed-off-by: Pavel Shilovsky <pshilov@microsoft.com>
Signed-off-by: Steve French <stfrench@microsoft.com>
CC: Stable <stable@vger.kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
commit b983f7e92348d7e7d091db1b78b7915e9dd3d63a upstream.
Currently for MTU requests we allocate maximum possible credits
in advance and then adjust them according to the request size.
While we were adjusting the number of credits belonging to the
server, we were skipping adjustment of credits belonging to the
request. This patch fixes it by setting request credits to
CreditCharge field value of SMB2 packet header.
Also ask 1 credit more for async read and write operations to
increase parallelism and match the behavior of other operations.
Signed-off-by: Pavel Shilovsky <pshilov@microsoft.com>
Signed-off-by: Steve French <stfrench@microsoft.com>
CC: Stable <stable@vger.kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
commit a6d8654d885d7d79a3fb82da64eaa489ca332a82 upstream.
When modifying the free space tree we can end up COWing one of its extent
buffers which in turn might result in allocating a new chunk, which in
turn can result in flushing (finish creation) of pending block groups. If
that happens we can deadlock because creating a pending block group needs
to update the free space tree, and if any of the updates tries to modify
the same extent buffer that we are COWing, we end up in a deadlock since
we try to write lock twice the same extent buffer.
So fix this by skipping pending block group creation if we are COWing an
extent buffer from the free space tree. This is a case missed by commit
5ce555578e091 ("Btrfs: fix deadlock when writing out free space caches").
Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=202173
Fixes: 5ce555578e091 ("Btrfs: fix deadlock when writing out free space caches")
CC: stable@vger.kernel.org # 4.18+
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
commit 3c1392d4c49962a31874af14ae9ff289cb2b3851 upstream.
Updating mseq makes client think importer mds has accepted all prior
cap messages and importer mds knows what caps client wants. Actually
some cap messages may have been dropped because of mseq mismatch.
If mseq is left untouched, importing cap's mds_wanted later will get
reset by cap import message.
Cc: stable@vger.kernel.org
Signed-off-by: "Yan, Zheng" <zyan@redhat.com>
Signed-off-by: Ilya Dryomov <idryomov@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
commit fdec6114ee1f0f43b1ad081ad8d46b23ba126d70 upstream.
Zero-length writes are legal; from 5661 section 18.32.3: "If the count
is zero, the WRITE will succeed and return a count of zero subject to
permissions checking".
This check is unnecessary and is causing zero-length reads to return
EINVAL.
Cc: stable@vger.kernel.org
Fixes: 3fd9557aec91 "NFSD: Refactor the generic write vector fill helper"
Cc: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
commit b8eee0e90f9797b747113638bc75e739b192ad38 upstream.
Commit 9d5b86ac13c5 ("fs/locks: Remove fl_nspid and use fs-specific l_pid
for remote locks") specified that the l_pid returned for F_GETLK on a local
file that has a remote lock should be the pid of the lock manager process.
That commit, while updating other filesystems, failed to update lockd, such
that locks created by lockd had their fl_pid set to that of the remote
process holding the lock. Fix that here to be the pid of lockd.
Also, fix the client case so that the returned lock pid is negative, which
indicates a remote lock on a remote file.
Fixes: 9d5b86ac13c5 ("fs/locks: Remove fl_nspid and use fs-specific...")
Cc: stable@vger.kernel.org
Signed-off-by: Benjamin Coddington <bcodding@redhat.com>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
commit 2d29f6b96d8f80322ed2dd895bca590491c38d34 upstream.
Fix the resource group wrap-around logic in gfs2_rbm_find that commit
e579ed4f44 broke. The bug can lead to unnecessary repeated scanning of the
same bitmaps; there is a risk that future changes will turn this into an
endless loop.
Fixes: e579ed4f44 ("GFS2: Introduce rbm field bii")
Cc: stable@vger.kernel.org # v3.13+
Signed-off-by: Andreas Gruenbacher <agruenba@redhat.com>
Signed-off-by: Bob Peterson <rpeterso@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
commit 6ff9b09e00a441599f3aacdf577254455a048bc9 upstream.
In gfs2_create_inode, after setting and releasing the acl / default_acl, the
acl / default_acl pointers are not set to NULL as they should be. In that
state, when the function reaches label fail_free_acls, gfs2_create_inode will
try to release the same acls again.
Fix that by setting the pointers to NULL after releasing the acls. Slightly
simplify the logic. Also, posix_acl_release checks for NULL already, so
there is no need to duplicate those checks here.
Fixes: e01580bf9e4d ("gfs2: use generic posix ACL infrastructure")
Reported-by: Pan Bian <bianpan2016@163.com>
Cc: Christoph Hellwig <hch@lst.de>
Cc: stable@vger.kernel.org # v4.9+
Signed-off-by: Andreas Gruenbacher <agruenba@redhat.com>
Signed-off-by: Bob Peterson <rpeterso@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
commit d47b41aceeadc6b58abc9c7c6485bef7cfb75636 upstream.
According to comment in dlm_user_request() ua should be freed
in dlm_free_lkb() after successful attach to lkb.
However ua is attached to lkb not in set_lock_args() but later,
inside request_lock().
Fixes 597d0cae0f99 ("[DLM] dlm: user locks")
Cc: stable@kernel.org # 2.6.19
Signed-off-by: Vasily Averin <vvs@virtuozzo.com>
Signed-off-by: David Teigland <teigland@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
commit c0174726c3976e67da8649ac62cae43220ae173a upstream.
Fixes 6d40c4a708e0 ("dlm: improve error and debug messages")
Cc: stable@kernel.org # 3.5
Signed-off-by: Vasily Averin <vvs@virtuozzo.com>
Signed-off-by: David Teigland <teigland@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
commit 23851e978f31eda8b2d01bd410d3026659ca06c7 upstream.
Fixes 3d6aa675fff9 ("dlm: keep lkbs in idr")
Cc: stable@kernel.org # 3.1
Signed-off-by: Vasily Averin <vvs@virtuozzo.com>
Signed-off-by: David Teigland <teigland@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
commit b982896cdb6e6a6b89d86dfb39df489d9df51e14 upstream.
If allocation fails on last elements of array need to free already
allocated elements.
v2: just move existing out_rsbtbl label to right place
Fixes 789924ba635f ("dlm: fix race between remove and lookup")
Cc: stable@kernel.org # 3.6
Signed-off-by: Vasily Averin <vvs@virtuozzo.com>
Signed-off-by: David Teigland <teigland@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
commit 6d2f84eee098540ae857998fe32f29b9e2cd9613 upstream.
When passing a large read to receive_encrypted_read(), ensure that the
demultiplex_thread knows that a MID was processed. Without this, those
operations never complete.
This is a similar issue/fix to lease break handling:
commit 7af929d6d05ba5564139718e30d5bc96bdbc716a
("smb3: fix lease break problem introduced by compounding")
CC: Stable <stable@vger.kernel.org> # 4.19+
Fixes: b24df3e30cbf ("cifs: update receive_encrypted_standard to handle compounded responses")
Signed-off-by: Paul Aurich <paul@darkrain42.org>
Tested-by: Yves-Alexis Perez <corsac@corsac.net>
Signed-off-by: Steve French <stfrench@microsoft.com>
Reviewed-by: Ronnie Sahlberg <lsahlber@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
commit 9a596f5b39593414c0ec80f71b94a226286f084e upstream.
While resolving a bug with locks on samba shares found a strange behavior.
When a file locked by one node and we trying to lock it from another node
it fail with errno 5 (EIO) but in that case errno must be set to
(EACCES | EAGAIN).
This isn't happening when we try to lock file second time on same node.
In this case it returns EACCES as expected.
Also this issue not reproduces when we use SMB1 protocol (vers=1.0 in
mount options).
Further investigation showed that the mapping from status_to_posix_error
is different for SMB1 and SMB2+ implementations.
For SMB1 mapping is [NT_STATUS_LOCK_NOT_GRANTED to ERRlock]
(See fs/cifs/netmisc.c line 66)
but for SMB2+ mapping is [STATUS_LOCK_NOT_GRANTED to -EIO]
(see fs/cifs/smb2maperror.c line 383)
Quick changes in SMB2+ mapping from EIO to EACCES has fixed issue.
BUG: https://bugzilla.kernel.org/show_bug.cgi?id=201971
Signed-off-by: Georgy A Bystrenin <gkot@altlinux.org>
Reviewed-by: Pavel Shilovsky <pshilov@microsoft.com>
CC: Stable <stable@vger.kernel.org>
Signed-off-by: Steve French <stfrench@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
commit 64beba0558fce7b59e9a8a7afd77290e82a22163 upstream.
There is a security report where f2fs_getxattr() has a hole to expose wrong
memory region when the image is malformed like this.
f2fs_getxattr: entry->e_name_len: 4, size: 12288, buffer_size: 16384, len: 4
Cc: <stable@vger.kernel.org>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
commit 88960068f25fcc3759455d85460234dcc9d43fef upstream.
Treat "block_count" from struct f2fs_super_block as 64-bit little endian
value in sanity_check_raw_super() because struct f2fs_super_block
declares "block_count" as "__le64".
This fixes a bug where the superblock validation fails on big endian
devices with the following error:
F2FS-fs (sda1): Wrong segment_count / block_count (61439 > 0)
F2FS-fs (sda1): Can't find valid F2FS filesystem in 1th superblock
F2FS-fs (sda1): Wrong segment_count / block_count (61439 > 0)
F2FS-fs (sda1): Can't find valid F2FS filesystem in 2th superblock
As result of this the partition cannot be mounted.
With this patch applied the superblock validation works fine and the
partition can be mounted again:
F2FS-fs (sda1): Mounted with checkpoint version = 7c84
My little endian x86-64 hardware was able to mount the partition without
this fix.
To confirm that mounting f2fs filesystems works on big endian machines
again I tested this on a 32-bit MIPS big endian (lantiq) device.
Fixes: 0cfe75c5b01199 ("f2fs: enhance sanity_check_raw_super() to avoid potential overflows")
Cc: stable@vger.kernel.org
Signed-off-by: Martin Blumenstingl <martin.blumenstingl@googlemail.com>
Reviewed-by: Chao Yu <yuchao0@huawei.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
commit 0ea295dd853e0879a9a30ab61f923c26be35b902 upstream.
The function truncate_node frees the page with f2fs_put_page. However,
the page index is read after that. So, the patch reads the index before
freeing the page.
Fixes: bf39c00a9a7f ("f2fs: drop obsolete node page when it is truncated")
Cc: <stable@vger.kernel.org>
Signed-off-by: Pan Bian <bianpan2016@163.com>
Reviewed-by: Chao Yu <yuchao0@huawei.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
commit d8a706414af4827fc0b4b1c0c631c607351938b9 upstream.
get_unlocked_entry() uses an exclusive wait because it is guaranteed to
eventually obtain the lock and follow on with an unlock+wakeup cycle.
The wait_entry_unlocked() path does not have the same guarantee. Rather
than open-code an extra wakeup, just switch to a non-exclusive wait.
Cc: Matthew Wilcox <willy@infradead.org>
Reported-by: Linus Torvalds <torvalds@linux-foundation.org>
Reviewed-by: Jan Kara <jack@suse.cz>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
|
|
commit 55e56f06ed71d9441f3abd5b1d3c1a870812b3fe upstream.
After we drop the i_pages lock, the inode can be freed at any time.
The get_unlocked_entry() code has no choice but to reacquire the lock,
so it can't be used here. Create a new wait_entry_unlocked() which takes
care not to acquire the lock or dereference the address_space in any way.
Fixes: c2a7d2a11552 ("filesystem-dax: Introduce dax_lock_mapping_entry()")
Cc: <stable@vger.kernel.org>
Signed-off-by: Matthew Wilcox <willy@infradead.org>
Reviewed-by: Jan Kara <jack@suse.cz>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
|
|
commit be6821f82c3cc36e026f5afd10249988852b35ea upstream.
If we create a snapshot of a snapshot currently being used by a send
operation, we can end up with send failing unexpectedly (returning
-ENOENT error to user space for example). The following diagram shows
how this happens.
CPU 1 CPU2 CPU3
btrfs_ioctl_send()
(...)
create_snapshot()
-> creates snapshot of a
root used by the send
task
btrfs_commit_transaction()
create_pending_snapshot()
__get_inode_info()
btrfs_search_slot()
btrfs_search_slot_get_root()
down_read commit_root_sem
get reference on eb of the
commit root
-> eb with bytenr == X
up_read commit_root_sem
btrfs_cow_block(root node)
btrfs_free_tree_block()
-> creates delayed ref to
free the extent
btrfs_run_delayed_refs()
-> runs the delayed ref,
adds extent to
fs_info->pinned_extents
btrfs_finish_extent_commit()
unpin_extent_range()
-> marks extent as free
in the free space cache
transaction commit finishes
btrfs_start_transaction()
(...)
btrfs_cow_block()
btrfs_alloc_tree_block()
btrfs_reserve_extent()
-> allocates extent at
bytenr == X
btrfs_init_new_buffer(bytenr X)
btrfs_find_create_tree_block()
alloc_extent_buffer(bytenr X)
find_extent_buffer(bytenr X)
-> returns existing eb,
which the send task got
(...)
-> modifies content of the
eb with bytenr == X
-> uses an eb that now
belongs to some other
tree and no more matches
the commit root of the
snapshot, resuts will be
unpredictable
The consequences of this race can be various, and can lead to searches in
the commit root performed by the send task failing unexpectedly (unable to
find inode items, returning -ENOENT to user space, for example) or not
failing because an inode item with the same number was added to the tree
that reused the metadata extent, in which case send can behave incorrectly
in the worst case or just fail later for some reason.
Fix this by performing a copy of the commit root's extent buffer when doing
a search in the context of a send operation.
CC: stable@vger.kernel.org # 4.4.x: 1fc28d8e2e9: Btrfs: move get root out of btrfs_search_slot to a helper
CC: stable@vger.kernel.org # 4.4.x: f9ddfd0592a: Btrfs: remove unused check of skip_locking
CC: stable@vger.kernel.org # 4.4.x
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
commit 0568e82dbe2510fc1fa664f58e5c997d3f1e649e upstream.
With my delayed refs patches in place we started seeing a large amount
of aborts in __btrfs_free_extent:
BTRFS error (device sdb1): unable to find ref byte nr 91947008 parent 0 root 35964 owner 1 offset 0
Call Trace:
? btrfs_merge_delayed_refs+0xaf/0x340
__btrfs_run_delayed_refs+0x6ea/0xfc0
? btrfs_set_path_blocking+0x31/0x60
btrfs_run_delayed_refs+0xeb/0x180
btrfs_commit_transaction+0x179/0x7f0
? btrfs_check_space_for_delayed_refs+0x30/0x50
? should_end_transaction.isra.19+0xe/0x40
btrfs_drop_snapshot+0x41c/0x7c0
btrfs_clean_one_deleted_snapshot+0xb5/0xd0
cleaner_kthread+0xf6/0x120
kthread+0xf8/0x130
? btree_invalidatepage+0x90/0x90
? kthread_bind+0x10/0x10
ret_from_fork+0x35/0x40
This was because btrfs_drop_snapshot depends on the root not being
modified while it's dropping the snapshot. It will unlock the root node
(and really every node) as it walks down the tree, only to re-lock it
when it needs to do something. This is a problem because if we modify
the tree we could cow a block in our path, which frees our reference to
that block. Then once we get back to that shared block we'll free our
reference to it again, and get ENOENT when trying to lookup our extent
reference to that block in __btrfs_free_extent.
This is ultimately happening because we have delayed items left to be
processed for our deleted snapshot _after_ all of the inodes are closed
for the snapshot. We only run the delayed inode item if we're deleting
the inode, and even then we do not run the delayed insertions or delayed
removals. These can be run at any point after our final inode does its
last iput, which is what triggers the snapshot deletion. We can end up
with the snapshot deletion happening and then have the delayed items run
on that file system, resulting in the above problem.
This problem has existed forever, however my patches made it much easier
to hit as I wake up the cleaner much more often to deal with delayed
iputs, which made us more likely to start the snapshot dropping work
before the transaction commits, which is when the delayed items would
generally be run. Before, generally speaking, we would run the delayed
items, commit the transaction, and wakeup the cleaner thread to start
deleting snapshots, which means we were less likely to hit this problem.
You could still hit it if you had multiple snapshots to be deleted and
ended up with lots of delayed items, but it was definitely harder.
Fix for now by simply running all the delayed items before starting to
drop the snapshot. We could make this smarter in the future by making
the delayed items per-root, and then simply drop any delayed items for
roots that we are going to delete. But for now just a quick and easy
solution is the safest.
CC: stable@vger.kernel.org # 4.4+
Reviewed-by: Filipe Manana <fdmanana@suse.com>
Signed-off-by: Josef Bacik <josef@toxicpanda.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
commit 41bd60676923822de1df2c50b3f9a10171f4338a upstream.
The log tree has a long standing problem that when a file is fsync'ed we
only check for new ancestors, created in the current transaction, by
following only the hard link for which the fsync was issued. We follow the
ancestors using the VFS' dget_parent() API. This means that if we create a
new link for a file in a directory that is new (or in an any other new
ancestor directory) and then fsync the file using an old hard link, we end
up not logging the new ancestor, and on log replay that new hard link and
ancestor do not exist. In some cases, involving renames, the file will not
exist at all.
Example:
mkfs.btrfs -f /dev/sdb
mount /dev/sdb /mnt
mkdir /mnt/A
touch /mnt/foo
ln /mnt/foo /mnt/A/bar
xfs_io -c fsync /mnt/foo
<power failure>
In this example after log replay only the hard link named 'foo' exists
and directory A does not exist, which is unexpected. In other major linux
filesystems, such as ext4, xfs and f2fs for example, both hard links exist
and so does directory A after mounting again the filesystem.
Checking if any new ancestors are new and need to be logged was added in
2009 by commit 12fcfd22fe5b ("Btrfs: tree logging unlink/rename fixes"),
however only for the ancestors of the hard link (dentry) for which the
fsync was issued, instead of checking for all ancestors for all of the
inode's hard links.
So fix this by tracking the id of the last transaction where a hard link
was created for an inode and then on fsync fallback to a full transaction
commit when an inode has more than one hard link and at least one new hard
link was created in the current transaction. This is the simplest solution
since this is not a common use case (adding frequently hard links for
which there's an ancestor created in the current transaction and then
fsync the file). In case it ever becomes a common use case, a solution
that consists of iterating the fs/subvol btree for each hard link and
check if any ancestor is new, could be implemented.
This solves many unexpected scenarios reported by Jayashree Mohan and
Vijay Chidambaram, and for which there is a new test case for fstests
under review.
Fixes: 12fcfd22fe5b ("Btrfs: tree logging unlink/rename fixes")
CC: stable@vger.kernel.org # 4.4+
Reported-by: Vijay Chidambaram <vvijay03@gmail.com>
Reported-by: Jayashree Mohan <jayashree2912@gmail.com>
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
run_delalloc_nocow
commit 27a7ff554e8d349627a90bda275c527b7348adae upstream.
The test case btrfs/001 with inode_cache mount option will encounter the
following warning:
WARNING: CPU: 1 PID: 23700 at fs/btrfs/inode.c:956 cow_file_range.isra.19+0x32b/0x430 [btrfs]
CPU: 1 PID: 23700 Comm: btrfs Kdump: loaded Tainted: G W O 4.20.0-rc4-custom+ #30
Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS 0.0.0 02/06/2015
RIP: 0010:cow_file_range.isra.19+0x32b/0x430 [btrfs]
Call Trace:
? free_extent_buffer+0x46/0x90 [btrfs]
run_delalloc_nocow+0x455/0x900 [btrfs]
btrfs_run_delalloc_range+0x1a7/0x360 [btrfs]
writepage_delalloc+0xf9/0x150 [btrfs]
__extent_writepage+0x125/0x3e0 [btrfs]
extent_write_cache_pages+0x1b6/0x3e0 [btrfs]
? __wake_up_common_lock+0x63/0xc0
extent_writepages+0x50/0x80 [btrfs]
do_writepages+0x41/0xd0
? __filemap_fdatawrite_range+0x9e/0xf0
__filemap_fdatawrite_range+0xbe/0xf0
btrfs_fdatawrite_range+0x1b/0x50 [btrfs]
__btrfs_write_out_cache+0x42c/0x480 [btrfs]
btrfs_write_out_ino_cache+0x84/0xd0 [btrfs]
btrfs_save_ino_cache+0x551/0x660 [btrfs]
commit_fs_roots+0xc5/0x190 [btrfs]
btrfs_commit_transaction+0x2bf/0x8d0 [btrfs]
btrfs_mksubvol+0x48d/0x4d0 [btrfs]
btrfs_ioctl_snap_create_transid+0x170/0x180 [btrfs]
btrfs_ioctl_snap_create_v2+0x124/0x180 [btrfs]
btrfs_ioctl+0x123f/0x3030 [btrfs]
The file extent generation of the free space inode is equal to the last
snapshot of the file root, so the inode will be passed to cow_file_rage.
But the inode was created and its extents were preallocated in
btrfs_save_ino_cache, there are no cow copies on disk.
The preallocated extent is not yet in the extent tree, and
btrfs_cross_ref_exist will ignore the -ENOENT returned by
check_committed_ref, so we can directly write the inode to the disk.
Fixes: 78d4295b1eee ("btrfs: lift some btrfs_cross_ref_exist checks in nocow path")
CC: stable@vger.kernel.org # 4.18+
Reviewed-by: Filipe Manana <fdmanana@suse.com>
Signed-off-by: Lu Fengqi <lufq.fnst@cn.fujitsu.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
commit 05c49e6bc1e8866ecfd674ebeeb58cdbff9145c2 upstream.
In a secnario where balance and replace co-exists as below,
- start balance
- pause balance
- start replace
- reboot
and when system restarts, balance resumes first. Then the replace is
attempted to restart but will fail as the EXCL_OP lock is already held
by the balance. If so place the replace state back to
BTRFS_IOCTL_DEV_REPLACE_STATE_SUSPENDED state.
Fixes: 010a47bde9420 ("btrfs: add proper safety check before resuming dev-replace")
CC: stable@vger.kernel.org # 4.18+
Signed-off-by: Anand Jain <anand.jain@oracle.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
commit 0d228ece59a35a9b9e8ff0d40653234a6d90f61e upstream.
At the time of forced unmount we place the running replace to
BTRFS_IOCTL_DEV_REPLACE_STATE_SUSPENDED state, so when the system comes
back and expect the target device is missing.
Then let the replace state continue to be in
BTRFS_IOCTL_DEV_REPLACE_STATE_SUSPENDED state instead of
BTRFS_IOCTL_DEV_REPLACE_STATE_STARTED as there isn't any matching scrub
running as part of replace.
Fixes: e93c89c1aaaa ("Btrfs: add new sources for device replace code")
CC: stable@vger.kernel.org # 4.4+
Signed-off-by: Anand Jain <anand.jain@oracle.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
commit 18f2c4fcebf2582f96cbd5f2238f4f354a0e4847 upstream.
If the file system has been shut down or is read-only, then
ext4_write_inode() needs to bail out early.
Also use jbd2_complete_transaction() instead of ext4_force_commit() so
we only force a commit if it is needed.
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
Cc: stable@kernel.org
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
commit fde872682e175743e0c3ef939c89e3c6008a1529 upstream.
Some time back, nfsd switched from calling vfs_fsync() to using a new
commit_metadata() hook in export_operations(). If the file system did
not provide a commit_metadata() hook, it fell back to using
sync_inode_metadata(). Unfortunately doesn't work on all file
systems. In particular, it doesn't work on ext4 due to how the inode
gets journalled --- the VFS writeback code will not always call
ext4_write_inode().
So we need to provide our own ext4_nfs_commit_metdata() method which
calls ext4_write_inode() directly.
Google-Bug-Id: 121195940
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
Cc: stable@kernel.org
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
commit 8a363970d1dc38c4ec4ad575c862f776f468d057 upstream.
If we receive a file handle, either from NFS or open_by_handle_at(2),
and it points at an inode which has not been initialized, and the file
system has metadata checksums enabled, we shouldn't try to get the
inode, discover the checksum is invalid, and then declare the file
system as being inconsistent.
This can be reproduced by creating a test file system via "mke2fs -t
ext4 -O metadata_csum /tmp/foo.img 8M", mounting it, cd'ing into that
directory, and then running the following program.
#define _GNU_SOURCE
#include <fcntl.h>
struct handle {
struct file_handle fh;
unsigned char fid[MAX_HANDLE_SZ];
};
int main(int argc, char **argv)
{
struct handle h = {{8, 1 }, { 12, }};
open_by_handle_at(AT_FDCWD, &h.fh, O_RDONLY);
return 0;
}
Google-Bug-Id: 120690101
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
Cc: stable@kernel.org
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
commit a805622a757b6d7f65def4141d29317d8e37b8a1 upstream.
In ext4_expand_extra_isize_ea(), we calculate the total size of the
xattr header, plus the xattr entries so we know how much of the
beginning part of the xattrs to move when expanding the inode extra
size. We need to include the terminating u32 at the end of the xattr
entries, or else if there is uninitialized, non-zero bytes after the
xattr entries and before the xattr values, the list of xattr entries
won't be properly terminated.
Reported-by: Steve Graham <stgraham2000@gmail.com>
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
Cc: stable@kernel.org
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
commit e647e29196b7f802f8242c39ecb7cc937f5ef217 upstream.
Commit e2b911c53584 ("ext4: clean up feature test macros with
predicate functions") broke the EXT4_IOC_GROUP_ADD ioctl. This was
not noticed since only very old versions of resize2fs (before
e2fsprogs 1.42) use this ioctl. However, using a new kernel with an
enterprise Linux userspace will cause attempts to use online resize to
fail with "No reserved GDT blocks".
Fixes: e2b911c53584 ("ext4: clean up feature test macros with predicate...")
Cc: stable@kernel.org # v4.4
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
Signed-off-by: ruippan (潘睿) <ruippan@tencent.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
commit 132d00becb31e88469334e1e62751c81345280e0 upstream.
In case of error, ext4_try_to_write_inline_data() should unlock
and release the page it holds.
Fixes: f19d5870cbf7 ("ext4: add normal write support for inline data")
Cc: stable@kernel.org # 3.8
Signed-off-by: Maurizio Lombardi <mlombard@redhat.com>
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
commit 61157b24e60fb3cd1f85f2c76a7b1d628f970144 upstream.
The function frees qf_inode via iput but then pass qf_inode to
lockdep_set_quota_inode on the failure path. This may result in a
use-after-free bug. The patch frees df_inode only when it is never used.
Fixes: daf647d2dd5 ("ext4: add lockdep annotations for i_data_sem")
Cc: stable@kernel.org # 4.6
Reviewed-by: Jan Kara <jack@suse.cz>
Signed-off-by: Pan Bian <bianpan2016@163.com>
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
commit fb265c9cb49e2074ddcdd4de99728aefdd3b3592 upstream.
Today, when sb_bread() returns NULL, this can either be because of an
I/O error or because the system failed to allocate the buffer. Since
it's an old interface, changing would require changing many call
sites.
So instead we create our own ext4_sb_bread(), which also allows us to
set the REQ_META flag.
Also fixed a problem in the xattr code where a NULL return in a
function could also mean that the xattr was not found, which could
lead to the wrong error getting returned to userspace.
Fixes: ac27a0ec112a ("ext4: initial copy of files from ext3")
Cc: stable@kernel.org # 2.6.19
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
commit ea5751ccd665a2fd1b24f9af81f6167f0718c5f6 upstream.
proc_sys_lookup can fail with ENOMEM instead of ENOENT when the
corresponding sysctl table is being unregistered. In our case we see
this upon opening /proc/sys/net/*/conf files while network interfaces
are being deleted, which confuses our configuration daemon.
The problem was successfully reproduced and this fix tested on v4.9.122
and v4.20-rc6.
v2: return ERR_PTRs in all cases when proc_sys_make_inode fails instead
of mixing them with NULL. Thanks Al Viro for the feedback.
Fixes: ace0c791e6c3 ("proc/sysctl: Don't grab i_lock under sysctl_lock.")
Cc: stable@vger.kernel.org
Signed-off-by: Ivan Delalande <colona@arista.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
commit e58725d51fa8da9133f3f1c54170aa2e43056b91 upstream.
UBIFS's recovery code strictly assumes that a deleted inode will never
come back, therefore it removes all data which belongs to that inode
as soon it faces an inode with link count 0 in the replay list.
Before O_TMPFILE this assumption was perfectly fine. With O_TMPFILE
it can lead to data loss upon a power-cut.
Consider a journal with entries like:
0: inode X (nlink = 0) /* O_TMPFILE was created */
1: data for inode X /* Someone writes to the temp file */
2: inode X (nlink = 0) /* inode was changed, xattr, chmod, … */
3: inode X (nlink = 1) /* inode was re-linked via linkat() */
Upon replay of entry #2 UBIFS will drop all data that belongs to inode X,
this will lead to an empty file after mounting.
As solution for this problem, scan the replay list for a re-link entry
before dropping data.
Fixes: 474b93704f32 ("ubifs: Implement O_TMPFILE")
Cc: stable@vger.kernel.org
Cc: Russell Senior <russell@personaltelco.net>
Cc: Rafał Miłecki <zajec5@gmail.com>
Reported-by: Russell Senior <russell@personaltelco.net>
Reported-by: Rafał Miłecki <zajec5@gmail.com>
Tested-by: Rafał Miłecki <rafal@milecki.pl>
Signed-off-by: Richard Weinberger <richard@nod.at>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
commit 94f82008ce30e2624537d240d64ce718255e0b80 upstream.
This reverts commit 55956b59df336f6738da916dbb520b6e37df9fbd.
commit 55956b59df33 ("vfs: Allow userns root to call mknod on owned filesystems.")
enabled mknod() in user namespaces for userns root if CAP_MKNOD is
available. However, these device nodes are useless since any filesystem
mounted from a non-initial user namespace will set the SB_I_NODEV flag on
the filesystem. Now, when a device node s created in a non-initial user
namespace a call to open() on said device node will fail due to:
bool may_open_dev(const struct path *path)
{
return !(path->mnt->mnt_flags & MNT_NODEV) &&
!(path->mnt->mnt_sb->s_iflags & SB_I_NODEV);
}
The problem with this is that as of the aforementioned commit mknod()
creates partially functional device nodes in non-initial user namespaces.
In particular, it has the consequence that as of the aforementioned commit
open() will be more privileged with respect to device nodes than mknod().
Before it was the other way around. Specifically, if mknod() succeeded
then it was transparent for any userspace application that a fatal error
must have occured when open() failed.
All of this breaks multiple userspace workloads and a widespread assumption
about how to handle mknod(). Basically, all container runtimes and systemd
live by the slogan "ask for forgiveness not permission" when running user
namespace workloads. For mknod() the assumption is that if the syscall
succeeds the device nodes are useable irrespective of whether it succeeds
in a non-initial user namespace or not. This logic was chosen explicitly
to allow for the glorious day when mknod() will actually be able to create
fully functional device nodes in user namespaces.
A specific problem people are already running into when running 4.18 rc
kernels are failing systemd services. For any distro that is run in a
container systemd services started with the PrivateDevices= property set
will fail to start since the device nodes in question cannot be
opened (cf. the arguments in [1]).
Full disclosure, Seth made the very sound argument that it is already
possible to end up with partially functional device nodes. Any filesystem
mounted with MS_NODEV set will allow mknod() to succeed but will not allow
open() to succeed. The difference to the case here is that the MS_NODEV
case is transparent to userspace since it is an explicitly set mount option
while the SB_I_NODEV case is an implicit property enforced by the kernel
and hence opaque to userspace.
[1]: https://github.com/systemd/systemd/pull/9483
Signed-off-by: Christian Brauner <christian@brauner.io>
Cc: "Eric W. Biederman" <ebiederm@xmission.com>
Cc: Seth Forshee <seth.forshee@canonical.com>
Cc: Serge Hallyn <serge@hallyn.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
[ Upstream commit a837eca2412051628c0529768c9bc4f3580b040e ]
This reverts commit 61c6de667263184125d5ca75e894fcad632b0dd3.
The reverted commit added page reference counting to iomap page
structures that are used to track block size < page size state. This
was supposed to align the code with page migration page accounting
assumptions, but what it has done instead is break XFS filesystems.
Every fstests run I've done on sub-page block size XFS filesystems
has since picking up this commit 2 days ago has failed with bad page
state errors such as:
# ./run_check.sh "-m rmapbt=1,reflink=1 -i sparse=1 -b size=1k" "generic/038"
....
SECTION -- xfs
FSTYP -- xfs (debug)
PLATFORM -- Linux/x86_64 test1 4.20.0-rc6-dgc+
MKFS_OPTIONS -- -f -m rmapbt=1,reflink=1 -i sparse=1 -b size=1k /dev/sdc
MOUNT_OPTIONS -- /dev/sdc /mnt/scratch
generic/038 454s ...
run fstests generic/038 at 2018-12-20 18:43:05
XFS (sdc): Unmounting Filesystem
XFS (sdc): Mounting V5 Filesystem
XFS (sdc): Ending clean mount
BUG: Bad page state in process kswapd0 pfn:3a7fa
page:ffffea0000ccbeb0 count:0 mapcount:0 mapping:ffff88800d9b6360 index:0x1
flags: 0xfffffc0000000()
raw: 000fffffc0000000 dead000000000100 dead000000000200 ffff88800d9b6360
raw: 0000000000000001 0000000000000000 00000000ffffffff
page dumped because: non-NULL mapping
CPU: 0 PID: 676 Comm: kswapd0 Not tainted 4.20.0-rc6-dgc+ #915
Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.11.1-1 04/01/2014
Call Trace:
dump_stack+0x67/0x90
bad_page.cold.116+0x8a/0xbd
free_pcppages_bulk+0x4bf/0x6a0
free_unref_page_list+0x10f/0x1f0
shrink_page_list+0x49d/0xf50
shrink_inactive_list+0x19d/0x3b0
shrink_node_memcg.constprop.77+0x398/0x690
? shrink_slab.constprop.81+0x278/0x3f0
shrink_node+0x7a/0x2f0
kswapd+0x34b/0x6d0
? node_reclaim+0x240/0x240
kthread+0x11f/0x140
? __kthread_bind_mask+0x60/0x60
ret_from_fork+0x24/0x30
Disabling lock debugging due to kernel taint
....
The failures are from anyway that frees pages and empties the
per-cpu page magazines, so it's not a predictable failure or an easy
to debug failure.
generic/038 is a reliable reproducer of this problem - it has a 9 in
10 failure rate on one of my test machines. Failure on other
machines have been at random points in fstests runs but every run
has ended up tripping this problem. Hence generic/038 was used to
bisect the failure because it was the most reliable failure.
It is too close to the 4.20 release (not to mention holidays) to
try to diagnose, fix and test the underlying cause of the problem,
so reverting the commit is the only option we have right now. The
revert has been tested against a current tot 4.20-rc7+ kernel across
multiple machines running sub-page block size XFs filesystems and
none of the bad page state failures have been seen.
Signed-off-by: Dave Chinner <dchinner@redhat.com>
Cc: Piotr Jaroszynski <pjaroszynski@nvidia.com>
Cc: Christoph Hellwig <hch@lst.de>
Cc: William Kucharski <william.kucharski@oracle.com>
Cc: Darrick J. Wong <darrick.wong@oracle.com>
Cc: Brian Foster <bfoster@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
|
|
[ Upstream commit d6fd0ae25c6495674dc5a41a8d16bc8e0073276d ]
There's a race between close_ctree() and cleaner_kthread().
close_ctree() sets btrfs_fs_closing(), and the cleaner stops when it
sees it set, but this is racy; the cleaner might have already checked
the bit and could be cleaning stuff. In particular, if it deletes unused
block groups, it will create delayed iputs for the free space cache
inodes. As of "btrfs: don't run delayed_iputs in commit", we're no
longer running delayed iputs after a commit. Therefore, if the cleaner
creates more delayed iputs after delayed iputs are run in
btrfs_commit_super(), we will leak inodes on unmount and get a busy
inode crash from the VFS.
Fix it by parking the cleaner before we actually close anything. Then,
any remaining delayed iputs will always be handled in
btrfs_commit_super(). This also ensures that the commit in close_ctree()
is really the last commit, so we can get rid of the commit in
cleaner_kthread().
The fstest/generic/475 followed by 476 can trigger a crash that
manifests as a slab corruption caused by accessing the freed kthread
structure by a wake up function. Sample trace:
[ 5657.077612] BUG: unable to handle kernel NULL pointer dereference at 00000000000000cc
[ 5657.079432] PGD 1c57a067 P4D 1c57a067 PUD da10067 PMD 0
[ 5657.080661] Oops: 0000 [#1] PREEMPT SMP
[ 5657.081592] CPU: 1 PID: 5157 Comm: fsstress Tainted: G W 4.19.0-rc8-default+ #323
[ 5657.083703] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.11.2-0-gf9626cc-prebuilt.qemu-project.org 04/01/2014
[ 5657.086577] RIP: 0010:shrink_page_list+0x2f9/0xe90
[ 5657.091937] RSP: 0018:ffffb5c745c8f728 EFLAGS: 00010287
[ 5657.092953] RAX: 0000000000000074 RBX: ffffb5c745c8f830 RCX: 0000000000000000
[ 5657.094590] RDX: 0000000000000000 RSI: 0000000000000001 RDI: ffff9a8747fdf3d0
[ 5657.095987] RBP: ffffb5c745c8f9e0 R08: 0000000000000000 R09: 0000000000000000
[ 5657.097159] R10: ffff9a8747fdf5e8 R11: 0000000000000000 R12: ffffb5c745c8f788
[ 5657.098513] R13: ffff9a877f6ff2c0 R14: ffff9a877f6ff2c8 R15: dead000000000200
[ 5657.099689] FS: 00007f948d853b80(0000) GS:ffff9a877d600000(0000) knlGS:0000000000000000
[ 5657.101032] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 5657.101953] CR2: 00000000000000cc CR3: 00000000684bd000 CR4: 00000000000006e0
[ 5657.103159] Call Trace:
[ 5657.103776] shrink_inactive_list+0x194/0x410
[ 5657.104671] shrink_node_memcg.constprop.84+0x39a/0x6a0
[ 5657.105750] shrink_node+0x62/0x1c0
[ 5657.106529] try_to_free_pages+0x1a4/0x500
[ 5657.107408] __alloc_pages_slowpath+0x2c9/0xb20
[ 5657.108418] __alloc_pages_nodemask+0x268/0x2b0
[ 5657.109348] kmalloc_large_node+0x37/0x90
[ 5657.110205] __kmalloc_node+0x236/0x310
[ 5657.111014] kvmalloc_node+0x3e/0x70
Fixes: 30928e9baac2 ("btrfs: don't run delayed_iputs in commit")
Signed-off-by: Omar Sandoval <osandov@fb.com>
Reviewed-by: David Sterba <dsterba@suse.com>
[ add trace ]
Signed-off-by: David Sterba <dsterba@suse.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
|
Filipe Manana
Theodore Ts'o
Ross Lagerwall
Pavel Shilovsky
Yan, Zheng
J. Bruce Fields
Benjamin Coddington
Andreas Gruenbacher
Vasily Averin
Paul Aurich
Georgy A Bystrenin
Jaegeuk Kim
Martin Blumenstingl
Pan Bian
Dan Williams
Matthew Wilcox
Josef Bacik
Lu Fengqi
Anand Jain
ruippan (潘睿)
Maurizio Lombardi
Ivan Delalande
Richard Weinberger
Christian Brauner
Dave Chinner
Omar Sandoval