| Age | Commit message (Collapse) | Author |
|---|
|
When the AIL tries to flush the CIL, it relies on the CIL push
ending up on stable storage without having to wait for and
manipulate iclog state directly. However, if there is already a
pending CIL push when the AIL tries to flush the CIL, it won't set
the cil->xc_push_commit_stable flag and so the CIL push will not
actively flush the commit record iclog.
generic/530 when run on a single CPU test VM can trigger this fairly
reliably. This test exercises unlinked inode recovery, and can
result in inodes being pinned in memory by ongoing modifications to
the inode cluster buffer to record unlinked list modifications. As a
result, the first inode unlinked in a buffer can pin the tail of the
log whilst the inode cluster buffer is pinned by the current
checkpoint that has been pushed but isn't on stable storage because
because the cil->xc_push_commit_stable was not set. This results in
the log/AIL effectively deadlocking until something triggers the
commit record iclog to be pushed to stable storage (i.e. the
periodic log worker calling xfs_log_force()).
The fix is two-fold - first we should always set the
cil->xc_push_commit_stable when xlog_cil_flush() is called,
regardless of whether there is already a pending push or not.
Second, if the CIL is empty, we should trigger an iclog flush to
ensure that the iclogs of the last checkpoint have actually been
submitted to disk as that checkpoint may not have been run under
stable completion constraints.
Reported-and-tested-by: Matthew Wilcox <willy@infradead.org>
Fixes: 0020a190cf3e ("xfs: AIL needs asynchronous CIL forcing")
Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
|
|
xfs_ail_push_all_sync() has a loop like this:
while max_ail_lsn {
prepare_to_wait(ail_empty)
target = max_ail_lsn
wake_up(ail_task);
schedule()
}
Which is designed to sleep until the AIL is emptied. When
xfs_ail_update_finish() moves the tail of the log, it does:
if (list_empty(&ailp->ail_head))
wake_up_all(&ailp->ail_empty);
So it will only wake up the sync push waiter when the AIL goes
empty. If, by the time the push waiter has woken, the AIL has more
in it, it will reset the target, wake the push task and go back to
sleep.
The problem here is that if the AIL is having items added to it
when xfs_ail_push_all_sync() is called, then they may get inserted
into the AIL at a LSN higher than the target LSN. At this point,
xfsaild_push() will see that the target is X, the item LSNs are
(X+N) and skip over them, hence never pushing the out.
The result of this the AIL will not get emptied by the AIL push
thread, hence xfs_ail_finish_update() will never see the AIL being
empty even if it moves the tail. Hence xfs_ail_push_all_sync() never
gets woken and hence cannot update the push target to capture the
items beyond the current target on the LSN.
This is a TOCTOU type of issue so the way to avoid it is to not
use the push target at all for sync pushes. We know that a sync push
is being requested by the fact the ail_empty wait queue is active,
hence the xfsaild can just set the target to max_ail_lsn on every
push that we see the wait queue active. Hence we no longer will
leave items on the AIL that are beyond the LSN sampled at the start
of a sync push.
Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Chandan Babu R <chandan.babu@oracle.com>
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
|
|
AIL flushing can get stuck here:
[316649.005769] INFO: task xfsaild/pmem1:324525 blocked for more than 123 seconds.
[316649.007807] Not tainted 5.17.0-rc6-dgc+ #975
[316649.009186] "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.
[316649.011720] task:xfsaild/pmem1 state:D stack:14544 pid:324525 ppid: 2 flags:0x00004000
[316649.014112] Call Trace:
[316649.014841] <TASK>
[316649.015492] __schedule+0x30d/0x9e0
[316649.017745] schedule+0x55/0xd0
[316649.018681] io_schedule+0x4b/0x80
[316649.019683] xfs_buf_wait_unpin+0x9e/0xf0
[316649.021850] __xfs_buf_submit+0x14a/0x230
[316649.023033] xfs_buf_delwri_submit_buffers+0x107/0x280
[316649.024511] xfs_buf_delwri_submit_nowait+0x10/0x20
[316649.025931] xfsaild+0x27e/0x9d0
[316649.028283] kthread+0xf6/0x120
[316649.030602] ret_from_fork+0x1f/0x30
in the situation where flushing gets preempted between the unpin
check and the buffer trylock under nowait conditions:
blk_start_plug(&plug);
list_for_each_entry_safe(bp, n, buffer_list, b_list) {
if (!wait_list) {
if (xfs_buf_ispinned(bp)) {
pinned++;
continue;
}
Here >>>>>>
if (!xfs_buf_trylock(bp))
continue;
This means submission is stuck until something else triggers a log
force to unpin the buffer.
To get onto the delwri list to begin with, the buffer pin state has
already been checked, and hence it's relatively rare we get a race
between flushing and encountering a pinned buffer in delwri
submission to begin with. Further, to increase the pin count the
buffer has to be locked, so the only way we can hit this race
without failing the trylock is to be preempted between the pincount
check seeing zero and the trylock being run.
Hence to avoid this problem, just invert the order of trylock vs
pin check. We shouldn't hit that many pinned buffers here, so
optimising away the trylock for pinned buffers should not matter for
performance at all.
Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Chandan Babu R <chandan.babu@oracle.com>
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
|
|
After 963 iterations of generic/530, it deadlocked during recovery
on a pinned inode cluster buffer like so:
XFS (pmem1): Starting recovery (logdev: internal)
INFO: task kworker/8:0:306037 blocked for more than 122 seconds.
Not tainted 5.17.0-rc6-dgc+ #975
"echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.
task:kworker/8:0 state:D stack:13024 pid:306037 ppid: 2 flags:0x00004000
Workqueue: xfs-inodegc/pmem1 xfs_inodegc_worker
Call Trace:
<TASK>
__schedule+0x30d/0x9e0
schedule+0x55/0xd0
schedule_timeout+0x114/0x160
__down+0x99/0xf0
down+0x5e/0x70
xfs_buf_lock+0x36/0xf0
xfs_buf_find+0x418/0x850
xfs_buf_get_map+0x47/0x380
xfs_buf_read_map+0x54/0x240
xfs_trans_read_buf_map+0x1bd/0x490
xfs_imap_to_bp+0x4f/0x70
xfs_iunlink_map_ino+0x66/0xd0
xfs_iunlink_map_prev.constprop.0+0x148/0x2f0
xfs_iunlink_remove_inode+0xf2/0x1d0
xfs_inactive_ifree+0x1a3/0x900
xfs_inode_unlink+0xcc/0x210
xfs_inodegc_worker+0x1ac/0x2f0
process_one_work+0x1ac/0x390
worker_thread+0x56/0x3c0
kthread+0xf6/0x120
ret_from_fork+0x1f/0x30
</TASK>
task:mount state:D stack:13248 pid:324509 ppid:324233 flags:0x00004000
Call Trace:
<TASK>
__schedule+0x30d/0x9e0
schedule+0x55/0xd0
schedule_timeout+0x114/0x160
__down+0x99/0xf0
down+0x5e/0x70
xfs_buf_lock+0x36/0xf0
xfs_buf_find+0x418/0x850
xfs_buf_get_map+0x47/0x380
xfs_buf_read_map+0x54/0x240
xfs_trans_read_buf_map+0x1bd/0x490
xfs_imap_to_bp+0x4f/0x70
xfs_iget+0x300/0xb40
xlog_recover_process_one_iunlink+0x4c/0x170
xlog_recover_process_iunlinks.isra.0+0xee/0x130
xlog_recover_finish+0x57/0x110
xfs_log_mount_finish+0xfc/0x1e0
xfs_mountfs+0x540/0x910
xfs_fs_fill_super+0x495/0x850
get_tree_bdev+0x171/0x270
xfs_fs_get_tree+0x15/0x20
vfs_get_tree+0x24/0xc0
path_mount+0x304/0xba0
__x64_sys_mount+0x108/0x140
do_syscall_64+0x35/0x80
entry_SYSCALL_64_after_hwframe+0x44/0xae
</TASK>
task:xfsaild/pmem1 state:D stack:14544 pid:324525 ppid: 2 flags:0x00004000
Call Trace:
<TASK>
__schedule+0x30d/0x9e0
schedule+0x55/0xd0
io_schedule+0x4b/0x80
xfs_buf_wait_unpin+0x9e/0xf0
__xfs_buf_submit+0x14a/0x230
xfs_buf_delwri_submit_buffers+0x107/0x280
xfs_buf_delwri_submit_nowait+0x10/0x20
xfsaild+0x27e/0x9d0
kthread+0xf6/0x120
ret_from_fork+0x1f/0x30
We have the mount process waiting on an inode cluster buffer read,
inodegc doing unlink waiting on the same inode cluster buffer, and
the AIL push thread blocked in writeback waiting for the inode
cluster buffer to become unpinned.
What has happened here is that the AIL push thread has raced with
the inodegc process modifying, committing and pinning the inode
cluster buffer here in xfs_buf_delwri_submit_buffers() here:
blk_start_plug(&plug);
list_for_each_entry_safe(bp, n, buffer_list, b_list) {
if (!wait_list) {
if (xfs_buf_ispinned(bp)) {
pinned++;
continue;
}
Here >>>>>>
if (!xfs_buf_trylock(bp))
continue;
Basically, the AIL has found the buffer wasn't pinned and got the
lock without blocking, but then the buffer was pinned. This implies
the processing here was pre-empted between the pin check and the
lock, because the pin count can only be increased while holding the
buffer locked. Hence when it has gone to submit the IO, it has
blocked waiting for the buffer to be unpinned.
With all executing threads now waiting on the buffer to be unpinned,
we normally get out of situations like this via the background log
worker issuing a log force which will unpinned stuck buffers like
this. But at this point in recovery, we haven't started the log
worker. In fact, the first thing we do after processing intents and
unlinked inodes is *start the log worker*. IOWs, we start it too
late to have it break deadlocks like this.
Avoid this and any other similar deadlock vectors in intent and
unlinked inode recovery by starting the log worker before we recover
intents and unlinked inodes. This part of recovery runs as though
the filesystem is fully active, so we really should have the same
infrastructure running as we normally do at runtime.
Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Chandan Babu R <chandan.babu@oracle.com>
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
|
|
The symbol xfs_name_dotdot is a global variable that the xfs codebase
uses here and there to look up directory dotdot entries. Currently it's
a non-const variable, which means that it's a mutable global variable.
So far nobody's abused this to cause problems, but let's use the
compiler to enforce that.
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
|
|
Various directory functions do not modify their @name parameter,
so mark it const to make that clear. This will enable us to mark
the global xfs_name_dotdot variable as const to prevent mischief.
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
|
|
XFS does not reserve quota for directory expansion when renaming
children into a directory. This means that we don't reject the
expansion with EDQUOT when we're at or near a hard limit, which means
that unprivileged userspace can use rename() to exceed quota.
Rename operations don't always expand the target directory, and we allow
a rename to proceed with no space reservation if we don't need to add a
block to the target directory to handle the addition. Moreover, the
unlink operation on the source directory generally does not expand the
directory (you'd have to free a block and then cause a btree split) and
it's probably of little consequence to leave the corner case that
renaming a file out of a directory can increase its size.
As with link and unlink, there is a further bug in that we do not
trigger the blockgc workers to try to clear space when we're out of
quota.
Because rename is its own special tricky animal, we'll patch xfs_rename
directly to reserve quota to the rename transaction. We'll leave
cleaning up the rest of xfs_rename for the metadata directory tree
patchset.
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
|
|
XFS does not reserve quota for directory expansion when linking or
unlinking children from a directory. This means that we don't reject
the expansion with EDQUOT when we're at or near a hard limit, which
means that unprivileged userspace can use link()/unlink() to exceed
quota.
The fix for this is nuanced -- link operations don't always expand the
directory, and we allow a link to proceed with no space reservation if
we don't need to add a block to the directory to handle the addition.
Unlink operations generally do not expand the directory (you'd have to
free a block and then cause a btree split) and we can defer the
directory block freeing if there is no space reservation.
Moreover, there is a further bug in that we do not trigger the blockgc
workers to try to clear space when we're out of quota.
To fix both cases, create a new xfs_trans_alloc_dir function that
allocates the transaction, locks and joins the inodes, and reserves
quota for the directory. If there isn't sufficient space or quota,
we'll switch the caller to reservationless mode. This should prevent
quota usage overruns with the least restriction in functionality.
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
|
|
Combine if tests to reduce the indentation levels of the quota chown
calls in xfs_setattr_nonsize.
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Christian Brauner <brauner@kernel.org>
|
|
Filipe Manana pointed out that XFS' behavior w.r.t. setuid/setgid
revocation isn't consistent with btrfs[1] or ext4. Those two
filesystems use the VFS function setattr_copy to convey certain
attributes from struct iattr into the VFS inode structure.
Andrey Zhadchenko reported[2] that XFS uses the wrong user namespace to
decide if it should clear setgid and setuid on a file attribute update.
This is a second symptom of the problem that Filipe noticed.
XFS, on the other hand, open-codes setattr_copy in xfs_setattr_mode,
xfs_setattr_nonsize, and xfs_setattr_time. Regrettably, setattr_copy is
/not/ a simple copy function; it contains additional logic to clear the
setgid bit when setting the mode, and XFS' version no longer matches.
The VFS implements its own setuid/setgid stripping logic, which
establishes consistent behavior. It's a tad unfortunate that it's
scattered across notify_change, should_remove_suid, and setattr_copy but
XFS should really follow the Linux VFS. Adapt XFS to use the VFS
functions and get rid of the old functions.
[1] https://lore.kernel.org/fstests/CAL3q7H47iNQ=Wmk83WcGB-KBJVOEtR9+qGczzCeXJ9Y2KCV25Q@mail.gmail.com/
[2] https://lore.kernel.org/linux-xfs/20220221182218.748084-1-andrey.zhadchenko@virtuozzo.com/
Fixes: 7fa294c8991c ("userns: Allow chown and setgid preservation")
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Christian Brauner <brauner@kernel.org>
|
|
There are a few places where we test the current process' capability set
to decide if we're going to be more or less generous with resource
acquisition for a system call. If the process doesn't have the
capability, we can continue the call, albeit in a degraded mode.
These are /not/ the actual security decisions, so it's not proper to use
capable(), which (in certain selinux setups) causes audit messages to
get logged. Switch them to has_capability_noaudit.
Fixes: 7317a03df703f ("xfs: refactor inode ownership change transaction/inode/quota allocation idiom")
Fixes: ea9a46e1c4925 ("xfs: only return detailed fsmap info if the caller has CAP_SYS_ADMIN")
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Cc: Dave Chinner <david@fromorbit.com>
Reviewed-by: Ondrej Mosnacek <omosnace@redhat.com>
Acked-by: Serge Hallyn <serge@hallyn.com>
Reviewed-by: Eric Sandeen <sandeen@redhat.com>
|
|
COW extents are already converted into written real extents after
xfs_reflink_convert_cow_locked(), therefore cmap->br_state should
reflect it.
Otherwise, there is another necessary unwritten convertion
triggered in xfs_dio_write_end_io() for direct I/O cases.
Signed-off-by: Gao Xiang <hsiangkao@linux.alibaba.com>
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
|
|
Pull xfs fixes from Darrick Wong:
"Nothing exciting, just more fixes for not returning sync_filesystem
error values (and eliding it when it's not necessary).
Summary:
- Only call sync_filesystem when we're remounting the filesystem
readonly readonly, and actually check its return value"
* tag 'xfs-5.17-fixes-2' of git://git.kernel.org/pub/scm/fs/xfs/xfs-linux:
xfs: only bother with sync_filesystem during readonly remount
|
|
In commit 02b9984d6408, we pushed a sync_filesystem() call from the VFS
into xfs_fs_remount. The only time that we ever need to push dirty file
data or metadata to disk for a remount is if we're remounting the
filesystem read only, so this really could be moved to xfs_remount_ro.
Once we've moved the call site, actually check the return value from
sync_filesystem.
Fixes: 02b9984d6408 ("fs: push sync_filesystem() down to the file system's remount_fs()")
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
|
|
Pull xfs fixes from Darrick Wong:
"I was auditing operations in XFS that clear file privileges, and
realized that XFS' fallocate implementation drops suid/sgid but
doesn't clear file capabilities the same way that file writes and
reflink do.
There are VFS helpers that do it correctly, so refactor XFS to use
them. I also noticed that we weren't flushing the log at the correct
point in the fallocate operation, so that's fixed too.
Summary:
- Fix fallocate so that it drops all file privileges when files are
modified instead of open-coding that incompletely.
- Fix fallocate to flush the log if the caller wanted synchronous
file updates"
* tag 'xfs-5.17-fixes-1' of git://git.kernel.org/pub/scm/fs/xfs/xfs-linux:
xfs: ensure log flush at the end of a synchronous fallocate call
xfs: move xfs_update_prealloc_flags() to xfs_pnfs.c
xfs: set prealloc flag in xfs_alloc_file_space()
xfs: fallocate() should call file_modified()
xfs: remove XFS_PREALLOC_SYNC
xfs: reject crazy array sizes being fed to XFS_IOC_GETBMAP*
|
|
Pull vfs fixes from Darrick Wong:
"I was auditing the sync_fs code paths recently and noticed that most
callers of ->sync_fs ignore its return value (and many implementations
never return nonzero even if the fs is broken!), which means that
internal fs errors and corruption are not passed up to userspace
callers of syncfs(2) or FIFREEZE. Hence fixing the common code and
XFS, and I'll start working on the ext4/btrfs folks if this is merged.
Summary:
- Fix a bug where callers of ->sync_fs (e.g. sync_filesystem and
syncfs(2)) ignore the return value.
- Fix a bug where callers of sync_filesystem (e.g. fs freeze) ignore
the return value.
- Fix a bug in XFS where xfs_fs_sync_fs never passed back error
returns"
* tag 'vfs-5.17-fixes-2' of git://git.kernel.org/pub/scm/fs/xfs/xfs-linux:
xfs: return errors in xfs_fs_sync_fs
quota: make dquot_quota_sync return errors from ->sync_fs
vfs: make sync_filesystem return errors from ->sync_fs
vfs: make freeze_super abort when sync_filesystem returns error
|
|
Since we've started treating fallocate more like a file write, we
should flush the log to disk if the user has asked for synchronous
writes either by setting it via fcntl flags, or inode flags, or with
the sync mount option. We've already got a helper for this, so use
it.
[The original patch by Darrick was massaged by Dave to fit this patchset]
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
|
|
The operations that xfs_update_prealloc_flags() perform are now
unique to xfs_fs_map_blocks(), so move xfs_update_prealloc_flags()
to be a static function in xfs_pnfs.c and cut out all the
other functionality that is doesn't use anymore.
Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
|
|
Now that we only call xfs_update_prealloc_flags() from
xfs_file_fallocate() in the case where we need to set the
preallocation flag, do this in xfs_alloc_file_space() where we
already have the inode joined into a transaction and get
rid of the call to xfs_update_prealloc_flags() from the fallocate
code.
This also means that we now correctly avoid setting the
XFS_DIFLAG_PREALLOC flag when xfs_is_always_cow_inode() is true, as
these inodes will never have preallocated extents.
Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
|
|
In XFS, we always update the inode change and modification time when
any fallocate() operation succeeds. Furthermore, as various
fallocate modes can change the file contents (extending EOF,
punching holes, zeroing things, shifting extents), we should drop
file privileges like suid just like we do for a regular write().
There's already a VFS helper that figures all this out for us, so
use that.
The net effect of this is that we no longer drop suid/sgid if the
caller is root, but we also now drop file capabilities.
We also move the xfs_update_prealloc_flags() function so that it now
is only called by the scope that needs to set the the prealloc flag.
Based on a patch from Darrick Wong.
Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
|
|
Callers can acheive the same thing by calling xfs_log_force_inode()
after making their modifications. There is no need for
xfs_update_prealloc_flags() to do this.
Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
|
|
Syzbot tripped over the following complaint from the kernel:
WARNING: CPU: 2 PID: 15402 at mm/util.c:597 kvmalloc_node+0x11e/0x125 mm/util.c:597
While trying to run XFS_IOC_GETBMAP against the following structure:
struct getbmap fubar = {
.bmv_count = 0x22dae649,
};
Obviously, this is a crazy huge value since the next thing that the
ioctl would do is allocate 37GB of memory. This is enough to make
kvmalloc mad, but isn't large enough to trip the validation functions.
In other words, I'm fussing with checks that were **already sufficient**
because that's easier than dealing with 644 internal bug reports. Yes,
that's right, six hundred and forty-four.
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Allison Henderson <allison.henderson@oracle.com>
Reviewed-by: Catherine Hoang <catherine.hoang@oracle.com>
|
|
Now that the VFS will do something with the return values from
->sync_fs, make ours pass on error codes.
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Jan Kara <jack@suse.cz>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Acked-by: Christian Brauner <brauner@kernel.org>
|
|
Trond Myklebust reported soft lockups in XFS IO completion such as
this:
watchdog: BUG: soft lockup - CPU#12 stuck for 23s! [kworker/12:1:3106]
CPU: 12 PID: 3106 Comm: kworker/12:1 Not tainted 4.18.0-305.10.2.el8_4.x86_64 #1
Workqueue: xfs-conv/md127 xfs_end_io [xfs]
RIP: 0010:_raw_spin_unlock_irqrestore+0x11/0x20
Call Trace:
wake_up_page_bit+0x8a/0x110
iomap_finish_ioend+0xd7/0x1c0
iomap_finish_ioends+0x7f/0xb0
xfs_end_ioend+0x6b/0x100 [xfs]
xfs_end_io+0xb9/0xe0 [xfs]
process_one_work+0x1a7/0x360
worker_thread+0x1fa/0x390
kthread+0x116/0x130
ret_from_fork+0x35/0x40
Ioends are processed as an atomic completion unit when all the
chained bios in the ioend have completed their IO. Logically
contiguous ioends can also be merged and completed as a single,
larger unit. Both of these things can be problematic as both the
bio chains per ioend and the size of the merged ioends processed as
a single completion are both unbound.
If we have a large sequential dirty region in the page cache,
write_cache_pages() will keep feeding us sequential pages and we
will keep mapping them into ioends and bios until we get a dirty
page at a non-sequential file offset. These large sequential runs
can will result in bio and ioend chaining to optimise the io
patterns. The pages iunder writeback are pinned within these chains
until the submission chaining is broken, allowing the entire chain
to be completed. This can result in huge chains being processed
in IO completion context.
We get deep bio chaining if we have large contiguous physical
extents. We will keep adding pages to the current bio until it is
full, then we'll chain a new bio to keep adding pages for writeback.
Hence we can build bio chains that map millions of pages and tens of
gigabytes of RAM if the page cache contains big enough contiguous
dirty file regions. This long bio chain pins those pages until the
final bio in the chain completes and the ioend can iterate all the
chained bios and complete them.
OTOH, if we have a physically fragmented file, we end up submitting
one ioend per physical fragment that each have a small bio or bio
chain attached to them. We do not chain these at IO submission time,
but instead we chain them at completion time based on file
offset via iomap_ioend_try_merge(). Hence we can end up with unbound
ioend chains being built via completion merging.
XFS can then do COW remapping or unwritten extent conversion on that
merged chain, which involves walking an extent fragment at a time
and running a transaction to modify the physical extent information.
IOWs, we merge all the discontiguous ioends together into a
contiguous file range, only to then process them individually as
discontiguous extents.
This extent manipulation is computationally expensive and can run in
a tight loop, so merging logically contiguous but physically
discontigous ioends gains us nothing except for hiding the fact the
fact we broke the ioends up into individual physical extents at
submission and then need to loop over those individual physical
extents at completion.
Hence we need to have mechanisms to limit ioend sizes and
to break up completion processing of large merged ioend chains:
1. bio chains per ioend need to be bound in length. Pure overwrites
go straight to iomap_finish_ioend() in softirq context with the
exact bio chain attached to the ioend by submission. Hence the only
way to prevent long holdoffs here is to bound ioend submission
sizes because we can't reschedule in softirq context.
2. iomap_finish_ioends() has to handle unbound merged ioend chains
correctly. This relies on any one call to iomap_finish_ioend() being
bound in runtime so that cond_resched() can be issued regularly as
the long ioend chain is processed. i.e. this relies on mechanism #1
to limit individual ioend sizes to work correctly.
3. filesystems have to loop over the merged ioends to process
physical extent manipulations. This means they can loop internally,
and so we break merging at physical extent boundaries so the
filesystem can easily insert reschedule points between individual
extent manipulations.
Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reported-and-tested-by: Trond Myklebust <trondmy@hammerspace.com>
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
|
|
Pull xfs fixes from Darrick Wong:
"One of the patches removes some dead code from xfs_ioctl32.h and the
other fixes broken workqueue flushing in the inode garbage collector.
- Minor cleanup of ioctl32 cruft
- Clean up open coded inodegc workqueue function calls"
* tag 'xfs-5.17-merge-7' of git://git.kernel.org/pub/scm/fs/xfs/xfs-linux:
xfs: flush inodegc workqueue tasks before cancel
xfs: remove unused xfs_ioctl32.h declarations
|
|
Pull more xfs irix ioctl housecleaning from Darrick Wong:
"Withdraw the XFS_IOC_ALLOCSP* and XFS_IOC_FREESP* ioctl definitions.
This is the third and final of a series of small pull requests that
perform some long overdue housecleaning of XFS ioctls. This time,
we're withdrawing all variants of the ALLOCSP and FREESP ioctls from
XFS' userspace API. This might be a little premature since we've only
just removed the functionality, but as I pointed out in the last pull
request, nobody (including fstests) noticed that it was broken for 20
years.
In response to the patch, we received a single comment from someone
who stated that they 'augment' the ioctl for their own purposes, but
otherwise acquiesced to the withdrawal. I still want to try to clobber
these old ioctl definitions in 5.17.
So remove the header definitions for these ioctls. The just-removed
implementation has allowed callers to read stale disk contents for
more than **21 years** and nobody noticed or complained, which implies
a lack of users aside from exploit programs"
* tag 'xfs-5.17-merge-6' of git://git.kernel.org/pub/scm/fs/xfs/xfs-linux:
xfs: remove the XFS_IOC_{ALLOC,FREE}SP* definitions
|
|
Pull xfs irix ioctl housecleaning from Darrick Wong:
"Remove the XFS_IOC_ALLOCSP* and XFS_IOC_FREESP* ioctl families.
This is the second of a series of small pull requests that perform
some long overdue housecleaning of XFS ioctls. This time, we're
vacating the implementation of all variants of the ALLOCSP and FREESP
ioctls, which are holdovers from EFS in Irix, circa 1993. Roughly
equivalent functionality have been available for both ioctls since
2.6.25 (April 2008):
- XFS_IOC_FREESP ftruncates a file.
- XFS_IOC_ALLOCSP is the equivalent of fallocate.
As noted in the fix patch for CVE 2021-4155, the ALLOCSP ioctl has
been serving up stale disk blocks since 2000, and in 21 years
**nobody** noticed. On those grounds I think it's safe to vacate the
implementation.
Note that we lose the ability to preallocate and truncate relative to
the current file position, but as nobody's ever implemented that for
the VFS, I conclude that it's not in high demand.
Linux has always used fallocate as the space management system call,
whereas these Irix legacy ioctls only ever worked on XFS, and have
been the cause of recent stale data disclosure vulnerabilities. As
equivalent functionality is available elsewhere, remove the code"
* tag 'xfs-5.17-merge-5' of git://git.kernel.org/pub/scm/fs/xfs/xfs-linux:
xfs: kill the XFS_IOC_{ALLOC,FREE}SP* ioctls
|
|
Pull xfs ioctl housecleaning from Darrick Wong:
"This is the first of a series of small pull requests that perform some
long overdue housecleaning of XFS ioctls. This first pull request
removes the FSSETDM ioctl, which was used to set DMAPI event
attributes on XFS files. The DMAPI support has never been merged
upstream and the implementation of FSSETDM itself was removed two
years ago, so let's withdraw it completely.
- Withdraw the ioctl definition for the FSSETDM ioctl"
* tag 'xfs-5.17-merge-4' of git://git.kernel.org/pub/scm/fs/xfs/xfs-linux:
xfs: remove the XFS_IOC_FSSETDM definitions
|
|
The xfs_inodegc_stop() helper performs a high level flush of pending
work on the percpu queues and then runs a cancel_work_sync() on each
of the percpu work tasks to ensure all work has completed before
returning. While cancel_work_sync() waits for wq tasks to complete,
it does not guarantee work tasks have started. This means that the
_stop() helper can queue and instantly cancel a wq task without
having completed the associated work. This can be observed by
tracepoint inspection of a simple "rm -f <file>; fsfreeze -f <mnt>"
test:
xfs_destroy_inode: ... ino 0x83 ...
xfs_inode_set_need_inactive: ... ino 0x83 ...
xfs_inodegc_stop: ...
...
xfs_inodegc_start: ...
xfs_inodegc_worker: ...
xfs_inode_inactivating: ... ino 0x83 ...
The first few lines show that the inode is removed and need inactive
state set, but the inactivation work has not completed before the
inodegc mechanism stops. The inactivation doesn't actually occur
until the fs is unfrozen and the gc mechanism starts back up. Note
that this test requires fsfreeze to reproduce because xfs_freeze
indirectly invokes xfs_fs_statfs(), which calls xfs_inodegc_flush().
When this occurs, the workqueue try_to_grab_pending() logic first
tries to steal the pending bit, which does not succeed because the
bit has been set by queue_work_on(). Subsequently, it checks for
association of a pool workqueue from the work item under the pool
lock. This association is set at the point a work item is queued and
cleared when dequeued for processing. If the association exists, the
work item is removed from the queue and cancel_work_sync() returns
true. If the pwq association is cleared, the remove attempt assumes
the task is busy and retries (eventually returning false to the
caller after waiting for the work task to complete).
To avoid this race, we can flush each work item explicitly before
cancel. However, since the _queue_all() already schedules each
underlying work item, the workqueue level helpers are sufficient to
achieve the same ordering effect. E.g., the inodegc enabled flag
prevents scheduling any further work in the _stop() case. Use the
drain_workqueue() helper in this particular case to make the intent
a bit more self explanatory.
Signed-off-by: Brian Foster <bfoster@redhat.com>
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
|
|
Remove these unused ia32 compat declarations; all the bits involved have
either been withdrawn or hoisted to the VFS.
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Eric Sandeen <sandeen@redhat.com>
|
|
Now that we've made these ioctls defunct, move them from xfs_fs.h to
xfs_ioctl.c, which effectively removes them from the publicly supported
ioctl interfaces for XFS.
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Eric Sandeen <sandeen@redhat.com>
|
|
According to the glibc compat header for Irix 4, these ioctls originated
in April 1991 as a (somewhat clunky) way to preallocate space at the end
of a file on an EFS filesystem. XFS, which was released in Irix 5.3 in
December 1993, picked up these ioctls to maintain compatibility and they
were ported to Linux in the early 2000s.
Recently it was pointed out to me they still lurk in the kernel, even
though the Linux fallocate syscall supplanted the functionality a long
time ago. fstests doesn't seem to include any real functional or stress
tests for these ioctls, which means that the code quality is ... very
questionable. Most notably, it was a stale disk block exposure vector
for 21 years and nobody noticed or complained. As mature programmers
say, "If you're not testing it, it's broken."
Given all that, let's withdraw these ioctls from the XFS userspace API.
Normally we'd set a long deprecation process, but I estimate that there
aren't any real users, so let's trigger a warning in dmesg and return
-ENOTTY.
See: CVE-2021-4155
Augments: 983d8e60f508 ("xfs: map unwritten blocks in XFS_IOC_{ALLOC,FREE}SP just like fallocate")
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Eric Sandeen <sandeen@redhat.com>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
|
|
Remove the definitions for these ioctls, since the functionality (and,
weirdly, the 32-bit compat ioctl definitions) were removed from the
kernel in November 2019.
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
|
|
Merge misc updates from Andrew Morton:
"146 patches.
Subsystems affected by this patch series: kthread, ia64, scripts,
ntfs, squashfs, ocfs2, vfs, and mm (slab-generic, slab, kmemleak,
dax, kasan, debug, pagecache, gup, shmem, frontswap, memremap,
memcg, selftests, pagemap, dma, vmalloc, memory-failure, hugetlb,
userfaultfd, vmscan, mempolicy, oom-kill, hugetlbfs, migration, thp,
ksm, page-poison, percpu, rmap, zswap, zram, cleanups, hmm, and
damon)"
* emailed patches from Andrew Morton <akpm@linux-foundation.org>: (146 commits)
mm/damon: hide kernel pointer from tracepoint event
mm/damon/vaddr: hide kernel pointer from damon_va_three_regions() failure log
mm/damon/vaddr: use pr_debug() for damon_va_three_regions() failure logging
mm/damon/dbgfs: remove an unnecessary variable
mm/damon: move the implementation of damon_insert_region to damon.h
mm/damon: add access checking for hugetlb pages
Docs/admin-guide/mm/damon/usage: update for schemes statistics
mm/damon/dbgfs: support all DAMOS stats
Docs/admin-guide/mm/damon/reclaim: document statistics parameters
mm/damon/reclaim: provide reclamation statistics
mm/damon/schemes: account how many times quota limit has exceeded
mm/damon/schemes: account scheme actions that successfully applied
mm/damon: remove a mistakenly added comment for a future feature
Docs/admin-guide/mm/damon/usage: update for kdamond_pid and (mk|rm)_contexts
Docs/admin-guide/mm/damon/usage: mention tracepoint at the beginning
Docs/admin-guide/mm/damon/usage: remove redundant information
Docs/admin-guide/mm/damon/usage: update for scheme quotas and watermarks
mm/damon: convert macro functions to static inline functions
mm/damon: modify damon_rand() macro to static inline function
mm/damon: move damon_rand() definition into damon.h
...
|
|
Various places in the kernel - largely in filesystems - respond to a
memory allocation failure by looping around and re-trying. Some of
these cannot conveniently use __GFP_NOFAIL, for reasons such as:
- a GFP_ATOMIC allocation, which __GFP_NOFAIL doesn't work on
- a need to check for the process being signalled between failures
- the possibility that other recovery actions could be performed
- the allocation is quite deep in support code, and passing down an
extra flag to say if __GFP_NOFAIL is wanted would be clumsy.
Many of these currently use congestion_wait() which (in almost all
cases) simply waits the given timeout - congestion isn't tracked for
most devices.
It isn't clear what the best delay is for loops, but it is clear that
the various filesystems shouldn't be responsible for choosing a timeout.
This patch introduces memalloc_retry_wait() with takes on that
responsibility. Code that wants to retry a memory allocation can call
this function passing the GFP flags that were used. It will wait
however is appropriate.
For now, it only considers __GFP_NORETRY and whatever
gfpflags_allow_blocking() tests. If blocking is allowed without
__GFP_NORETRY, then alloc_page either made some reclaim progress, or
waited for a while, before failing. So there is no need for much
further waiting. memalloc_retry_wait() will wait until the current
jiffie ends. If this condition is not met, then alloc_page() won't have
waited much if at all. In that case memalloc_retry_wait() waits about
200ms. This is the delay that most current loops uses.
linux/sched/mm.h needs to be included in some files now,
but linux/backing-dev.h does not.
Link: https://lkml.kernel.org/r/163754371968.13692.1277530886009912421@noble.neil.brown.name
Signed-off-by: NeilBrown <neilb@suse.de>
Cc: Dave Chinner <david@fromorbit.com>
Cc: Michal Hocko <mhocko@suse.com>
Cc: "Theodore Ts'o" <tytso@mit.edu>
Cc: Jaegeuk Kim <jaegeuk@kernel.org>
Cc: Chao Yu <chao@kernel.org>
Cc: Darrick J. Wong <djwong@kernel.org>
Cc: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
|
|
Pull xfs fixes from Darrick Wong:
"These are the last few obvious fixes that I found while stress testing
online fsck for XFS prior to initiating a design review of the whole
giant machinery.
- Fix a minor locking inconsistency in readdir
- Fix incorrect fs feature bit validation for secondary superblocks"
* tag 'xfs-5.17-merge-3' of git://git.kernel.org/pub/scm/fs/xfs/xfs-linux:
xfs: fix online fsck handling of v5 feature bits on secondary supers
xfs: take the ILOCK when readdir inspects directory mapping data
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/nvdimm/nvdimm
Pull dax and libnvdimm updates from Dan Williams:
"The bulk of this is a rework of the dax_operations API after
discovering the obstacles it posed to the work-in-progress DAX+reflink
support for XFS and other copy-on-write filesystem mechanics.
Primarily the need to plumb a block_device through the API to handle
partition offsets was a sticking point and Christoph untangled that
dependency in addition to other cleanups to make landing the
DAX+reflink support easier.
The DAX_PMEM_COMPAT option has been around for 4 years and not only
are distributions shipping userspace that understand the current
configuration API, but some are not even bothering to turn this option
on anymore, so it seems a good time to remove it per the deprecation
schedule. Recall that this was added after the device-dax subsystem
moved from /sys/class/dax to /sys/bus/dax for its sysfs organization.
All recent functionality depends on /sys/bus/dax.
Some other miscellaneous cleanups and reflink prep patches are
included as well.
Summary:
- Simplify the dax_operations API:
- Eliminate bdev_dax_pgoff() in favor of the filesystem
maintaining and applying a partition offset to all its DAX iomap
operations.
- Remove wrappers and device-mapper stacked callbacks for
->copy_from_iter() and ->copy_to_iter() in favor of moving
block_device relative offset responsibility to the
dax_direct_access() caller.
- Remove the need for an @bdev in filesystem-DAX infrastructure
- Remove unused uio helpers copy_from_iter_flushcache() and
copy_mc_to_iter() as only the non-check_copy_size() versions are
used for DAX.
- Prepare XFS for the pending (next merge window) DAX+reflink support
- Remove deprecated DEV_DAX_PMEM_COMPAT support
- Cleanup a straggling misuse of the GUID api"
* tag 'libnvdimm-for-5.17' of git://git.kernel.org/pub/scm/linux/kernel/git/nvdimm/nvdimm: (38 commits)
iomap: Fix error handling in iomap_zero_iter()
ACPI: NFIT: Import GUID before use
dax: remove the copy_from_iter and copy_to_iter methods
dax: remove the DAXDEV_F_SYNC flag
dax: simplify dax_synchronous and set_dax_synchronous
uio: remove copy_from_iter_flushcache() and copy_mc_to_iter()
iomap: turn the byte variable in iomap_zero_iter into a ssize_t
memremap: remove support for external pgmap refcounts
fsdax: don't require CONFIG_BLOCK
iomap: build the block based code conditionally
dax: fix up some of the block device related ifdefs
fsdax: shift partition offset handling into the file systems
dax: return the partition offset from fs_dax_get_by_bdev
iomap: add a IOMAP_DAX flag
xfs: pass the mapping flags to xfs_bmbt_to_iomap
xfs: use xfs_direct_write_iomap_ops for DAX zeroing
xfs: move dax device handling into xfs_{alloc,free}_buftarg
ext4: cleanup the dax handling in ext4_fill_super
ext2: cleanup the dax handling in ext2_fill_super
fsdax: decouple zeroing from the iomap buffered I/O code
...
|
|
Pull iomap updates from Matthew Wilcox:
"Convert xfs/iomap to use folios.
This should be all that is needed for XFS to use large folios. There
is no code in this pull request to create large folios, but no
additional changes should be needed to XFS or iomap once they are
created.
Usually this would have come from Darrick, and we had intended that it
would come that route. Between the holidays and various things which
Darrick needed to work on, he asked if I could send things directly.
There weren't any other iomap patches pending for this release, which
probably also played a role"
* tag 'iomap-5.17' of git://git.infradead.org/users/willy/linux: (26 commits)
iomap: Inline __iomap_zero_iter into its caller
xfs: Support large folios
iomap: Support large folios in invalidatepage
iomap: Convert iomap_migrate_page() to use folios
iomap: Convert iomap_add_to_ioend() to take a folio
iomap: Simplify iomap_do_writepage()
iomap: Simplify iomap_writepage_map()
iomap,xfs: Convert ->discard_page to ->discard_folio
iomap: Convert iomap_write_end_inline to take a folio
iomap: Convert iomap_write_begin() and iomap_write_end() to folios
iomap: Convert __iomap_zero_iter to use a folio
iomap: Allow iomap_write_begin() to be called with the full length
iomap: Convert iomap_page_mkwrite to use a folio
iomap: Convert readahead and readpage to use a folio
iomap: Convert iomap_read_inline_data to take a folio
iomap: Use folio offsets instead of page offsets
iomap: Convert bio completions to use folios
iomap: Pass the iomap_page into iomap_set_range_uptodate
iomap: Add iomap_invalidate_folio
iomap: Convert iomap_releasepage to use a folio
...
|
|
While I was auditing the code in xfs_repair that adds feature bits to
existing V5 filesystems, I decided to have a look at how online fsck
handles feature bits, and I found a few problems:
1) ATTR2 is added to the primary super when an xattr is set to a file,
but that isn't consistently propagated to secondary supers. This isn't
a corruption, merely a discrepancy that repair will fix if it ever has
to restore the primary from a secondary. Hence, if we find a mismatch
on a secondary, this is a preen condition, not a corruption.
2) There are more compat and ro_compat features now than there used to
be, but we mask off the newer features from testing. This means we
ignore inconsistencies in the INOBTCOUNT and BIGTIME features, which is
wrong. Get rid of the masking and compare directly.
3) NEEDSREPAIR, when set on a secondary, is ignored by everyone. Hence
a mismatch here should also be flagged for preening, and online repair
should clear the flag. Right now we ignore it due to (2).
4) log_incompat features are ephemeral, since we can clear the feature
bit as soon as the log no longer contains live records for a particular
log feature. As such, the only copy we care about is the one in the
primary super. If we find any bits set in the secondary super, we
should flag that for preening, and clear the bits if the user elects to
repair it.
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
|
|
I was poking around in the directory code while diagnosing online fsck
bugs, and noticed that xfs_readdir doesn't actually take the directory
ILOCK when it calls xfs_dir2_isblock. xfs_dir_open most probably loaded
the data fork mappings and the VFS took i_rwsem (aka IOLOCK_SHARED) so
we're protected against writer threads, but we really need to follow the
locking model like we do in other places.
To avoid unnecessarily cycling the ILOCK for fairly small directories,
change the block/leaf _getdents functions to consume the ILOCK hold that
the parent readdir function took to decide on a _getdents implementation.
It is ok to cycle the ILOCK in readdir because the VFS takes the IOLOCK
in the appropriate mode during lookups and writes, and we don't want to
be holding the ILOCK when we copy directory entries to userspace in case
there's a page fault. We really only need it to protect against data
fork lookups, like we do for other files.
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
|
|
Pull xfs updates from Darrick Wong:
"The big new feature here is that the mount code now only bothers to
try to free stale COW staging extents if the fs unmounted uncleanly.
This should reduce mount times, particularly on filesystems supporting
reflink and containing a large number of allocation groups.
Everything else this cycle are bugfixes, as the iomap folios
conversion should be plenty enough excitement for anyone. That and I
ran out of brain bandwidth after Thanksgiving last year.
Summary:
- Fix log recovery with da btree buffers when metauuid is in use.
- Fix type coercion problems in xattr buffer size validation.
- Fix a bug in online scrub dir leaf bestcount checking.
- Only run COW recovery when recovering the log.
- Fix symlink target buffer UAF problems and symlink locking problems
by not exposing xfs innards to the VFS.
- Fix incorrect quotaoff lock usage.
- Don't let transactions cancel cleanly if they have deferred work
items attached.
- Fix a UAF when we're deciding if we need to relog an intent item.
- Reduce kvmalloc overhead for log shadow buffers.
- Clean up sysfs attr group usage.
- Fix a bug where scrub's bmap/rmap checking could race with a quota
file block allocation due to insufficient locking.
- Teach scrub to complain about invalid project ids"
* tag 'xfs-5.17-merge-2' of git://git.kernel.org/pub/scm/fs/xfs/xfs-linux:
xfs: warn about inodes with project id of -1
xfs: hold quota inode ILOCK_EXCL until the end of dqalloc
xfs: Remove redundant assignment of mp
xfs: reduce kvmalloc overhead for CIL shadow buffers
xfs: sysfs: use default_groups in kobj_type
xfs: prevent UAF in xfs_log_item_in_current_chkpt
xfs: prevent a WARN_ONCE() in xfs_ioc_attr_list()
xfs: Fix comments mentioning xfs_ialloc
xfs: check sb_meta_uuid for dabuf buffer recovery
xfs: fix a bug in the online fsck directory leaf1 bestcount check
xfs: only run COW extent recovery when there are no live extents
xfs: don't expose internal symlink metadata buffers to the vfs
xfs: fix quotaoff mutex usage now that we don't support disabling it
xfs: shut down filesystem if we xfs_trans_cancel with deferred work items
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/brauner/linux
Pull fs idmapping updates from Christian Brauner:
"This contains the work to enable the idmapping infrastructure to
support idmapped mounts of filesystems mounted with an idmapping.
In addition this contains various cleanups that avoid repeated
open-coding of the same functionality and simplify the code in quite a
few places.
We also finish the renaming of the mapping helpers we started a few
kernel releases back and move them to a dedicated header to not
continue polluting the fs header needlessly with low-level idmapping
helpers. With this series the fs header only contains idmapping
helpers that interact with fs objects.
Currently we only support idmapped mounts for filesystems mounted
without an idmapping themselves. This was a conscious decision
mentioned in multiple places (cf. [1]).
As explained at length in [3] it is perfectly fine to extend support
for idmapped mounts to filesystem's mounted with an idmapping should
the need arise. The need has been there for some time now (cf. [2]).
Before we can port any filesystem that is mountable with an idmapping
to support idmapped mounts in the coming cycles, we need to first
extend the mapping helpers to account for the filesystem's idmapping.
This again, is explained at length in our documentation at [3] and
also in the individual commit messages so here's an overview.
Currently, the low-level mapping helpers implement the remapping
algorithms described in [3] in a simplified manner as we could rely on
the fact that all filesystems supporting idmapped mounts are mounted
without an idmapping.
In contrast, filesystems mounted with an idmapping are very likely to
not use an identity mapping and will instead use a non-identity
mapping. So the translation step from or into the filesystem's
idmapping in the remapping algorithm cannot be skipped for such
filesystems.
Non-idmapped filesystems and filesystems not supporting idmapped
mounts are unaffected by this change as the remapping algorithms can
take the same shortcut as before. If the low-level helpers detect that
they are dealing with an idmapped mount but the underlying filesystem
is mounted without an idmapping we can rely on the previous shortcut
and can continue to skip the translation step from or into the
filesystem's idmapping. And of course, if the low-level helpers detect
that they are not dealing with an idmapped mount they can simply
return the relevant id unchanged; no remapping needs to be performed
at all.
These checks guarantee that only the minimal amount of work is
performed. As before, if idmapped mounts aren't used the low-level
helpers are idempotent and no work is performed at all"
Link: 2ca4dcc4909d ("fs/mount_setattr: tighten permission checks") [1]
Link: https://github.com/containers/podman/issues/10374 [2]
Link: Documentations/filesystems/idmappings.rst [3]
Link: a65e58e791a1 ("fs: document and rename fsid helpers") [4]
* tag 'fs.idmapped.v5.17' of git://git.kernel.org/pub/scm/linux/kernel/git/brauner/linux:
fs: support mapped mounts of mapped filesystems
fs: add i_user_ns() helper
fs: port higher-level mapping helpers
fs: remove unused low-level mapping helpers
fs: use low-level mapping helpers
docs: update mapping documentation
fs: account for filesystem mappings
fs: tweak fsuidgid_has_mapping()
fs: move mapping helpers
fs: add is_idmapped_mnt() helper
|
|
Inodes aren't supposed to have a project id of -1U (aka 4294967295) but
the kernel hasn't always validated FSSETXATTR correctly. Flag this as
something for the sysadmin to check out.
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
|
|
Online fsck depends on callers holding ILOCK_EXCL from the time they
decide to update a block mapping until after they've updated the reverse
mapping records to guarantee the stability of both mapping records.
Unfortunately, the quota code drops ILOCK_EXCL at the first transaction
roll in the dquot allocation process, which breaks that assertion. This
leads to sporadic failures in the online rmap repair code if the repair
code grabs the AGF after bmapi_write maps a new block into the quota
file's data fork but before it can finish the deferred rmap update.
Fix this by rewriting the function to hold the ILOCK until after the
transaction commit like all other bmap updates do, and get rid of the
dqread wrapper that does nothing but complicate the codebase.
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
|
|
mp is being initialized to log->l_mp but this is never read
as record is overwritten later on. Remove the redundant
assignment.
Cleans up the following clang-analyzer warning:
fs/xfs/xfs_log_recover.c:3543:20: warning: Value stored to 'mp' during
its initialization is never read [clang-analyzer-deadcode.DeadStores].
Reported-by: Abaci Robot <abaci@linux.alibaba.com>
Signed-off-by: Jiapeng Chong <jiapeng.chong@linux.alibaba.com>
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
|
|
Oh, let me count the ways that the kvmalloc API sucks dog eggs.
The problem is when we are logging lots of large objects, we hit
kvmalloc really damn hard with costly order allocations, and
behaviour utterly sucks:
- 49.73% xlog_cil_commit
- 31.62% kvmalloc_node
- 29.96% __kmalloc_node
- 29.38% kmalloc_large_node
- 29.33% __alloc_pages
- 24.33% __alloc_pages_slowpath.constprop.0
- 18.35% __alloc_pages_direct_compact
- 17.39% try_to_compact_pages
- compact_zone_order
- 15.26% compact_zone
5.29% __pageblock_pfn_to_page
3.71% PageHuge
- 1.44% isolate_migratepages_block
0.71% set_pfnblock_flags_mask
1.11% get_pfnblock_flags_mask
- 0.81% get_page_from_freelist
- 0.59% _raw_spin_lock_irqsave
- do_raw_spin_lock
__pv_queued_spin_lock_slowpath
- 3.24% try_to_free_pages
- 3.14% shrink_node
- 2.94% shrink_slab.constprop.0
- 0.89% super_cache_count
- 0.66% xfs_fs_nr_cached_objects
- 0.65% xfs_reclaim_inodes_count
0.55% xfs_perag_get_tag
0.58% kfree_rcu_shrink_count
- 2.09% get_page_from_freelist
- 1.03% _raw_spin_lock_irqsave
- do_raw_spin_lock
__pv_queued_spin_lock_slowpath
- 4.88% get_page_from_freelist
- 3.66% _raw_spin_lock_irqsave
- do_raw_spin_lock
__pv_queued_spin_lock_slowpath
- 1.63% __vmalloc_node
- __vmalloc_node_range
- 1.10% __alloc_pages_bulk
- 0.93% __alloc_pages
- 0.92% get_page_from_freelist
- 0.89% rmqueue_bulk
- 0.69% _raw_spin_lock
- do_raw_spin_lock
__pv_queued_spin_lock_slowpath
13.73% memcpy_erms
- 2.22% kvfree
On this workload, that's almost a dozen CPUs all trying to compact
and reclaim memory inside kvmalloc_node at the same time. Yet it is
regularly falling back to vmalloc despite all that compaction, page
and shrinker reclaim that direct reclaim is doing. Copying all the
metadata is taking far less CPU time than allocating the storage!
Direct reclaim should be considered extremely harmful.
This is a high frequency, high throughput, CPU usage and latency
sensitive allocation. We've got memory there, and we're using
kvmalloc to allow memory allocation to avoid doing lots of work to
try to do contiguous allocations.
Except it still does *lots of costly work* that is unnecessary.
Worse: the only way to avoid the slowpath page allocation trying to
do compaction on costly allocations is to turn off direct reclaim
(i.e. remove __GFP_RECLAIM_DIRECT from the gfp flags).
Unfortunately, the stupid kvmalloc API then says "oh, this isn't a
GFP_KERNEL allocation context, so you only get kmalloc!". This
cuts off the vmalloc fallback, and this leads to almost instant OOM
problems which ends up in filesystems deadlocks, shutdowns and/or
kernel crashes.
I want some basic kvmalloc behaviour:
- kmalloc for a contiguous range with fail fast semantics - no
compaction direct reclaim if the allocation enters the slow path.
- run normal vmalloc (i.e. GFP_KERNEL) if kmalloc fails
The really, really stupid part about this is these kvmalloc() calls
are run under memalloc_nofs task context, so all the allocations are
always reduced to GFP_NOFS regardless of the fact that kvmalloc
requires GFP_KERNEL to be passed in. IOWs, we're already telling
kvmalloc to behave differently to the gfp flags we pass in, but it
still won't allow vmalloc to be run with anything other than
GFP_KERNEL.
So, this patch open codes the kvmalloc() in the commit path to have
the above described behaviour. The result is we more than halve the
CPU time spend doing kvmalloc() in this path and transaction commits
with 64kB objects in them more than doubles. i.e. we get ~5x
reduction in CPU usage per costly-sized kvmalloc() invocation and
the profile looks like this:
- 37.60% xlog_cil_commit
16.01% memcpy_erms
- 8.45% __kmalloc
- 8.04% kmalloc_order_trace
- 8.03% kmalloc_order
- 7.93% alloc_pages
- 7.90% __alloc_pages
- 4.05% __alloc_pages_slowpath.constprop.0
- 2.18% get_page_from_freelist
- 1.77% wake_all_kswapds
....
- __wake_up_common_lock
- 0.94% _raw_spin_lock_irqsave
- 3.72% get_page_from_freelist
- 2.43% _raw_spin_lock_irqsave
- 5.72% vmalloc
- 5.72% __vmalloc_node_range
- 4.81% __get_vm_area_node.constprop.0
- 3.26% alloc_vmap_area
- 2.52% _raw_spin_lock
- 1.46% _raw_spin_lock
0.56% __alloc_pages_bulk
- 4.66% kvfree
- 3.25% vfree
- __vfree
- 3.23% __vunmap
- 1.95% remove_vm_area
- 1.06% free_vmap_area_noflush
- 0.82% _raw_spin_lock
- 0.68% _raw_spin_lock
- 0.92% _raw_spin_lock
- 1.40% kfree
- 1.36% __free_pages
- 1.35% __free_pages_ok
- 1.02% _raw_spin_lock_irqsave
It's worth noting that over 50% of the CPU time spent allocating
these shadow buffers is now spent on spinlocks. So the shadow buffer
allocation overhead is greatly reduced by getting rid of direct
reclaim from kmalloc, and could probably be made even less costly if
vmalloc() didn't use global spinlocks to protect it's structures.
Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Allison Henderson <allison.henderson@oracle.com>
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
|
|
There are currently 2 ways to create a set of sysfs files for a
kobj_type, through the default_attrs field, and the default_groups
field. Move the xfs sysfs code to use default_groups field which has
been the preferred way since aa30f47cf666 ("kobject: Add support for
default attribute groups to kobj_type") so that we can soon get rid of
the obsolete default_attrs field.
Cc: "Darrick J. Wong" <djwong@kernel.org>
Cc: linux-xfs@vger.kernel.org
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
|
|
The old ALLOCSP/FREESP ioctls in XFS can be used to preallocate space at
the end of files, just like fallocate and RESVSP. Make the behavior
consistent with the other ioctls.
Reported-by: Kirill Tkhai <ktkhai@virtuozzo.com>
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Eric Sandeen <sandeen@redhat.com>
|
|
While I was running with KASAN and lockdep enabled, I stumbled upon an
KASAN report about a UAF to a freed CIL checkpoint. Looking at the
comment for xfs_log_item_in_current_chkpt, it seems pretty obvious to me
that the original patch to xfs_defer_finish_noroll should have done
something to lock the CIL to prevent it from switching the CIL contexts
while the predicate runs.
For upper level code that needs to know if a given log item is new
enough not to need relogging, add a new wrapper that takes the CIL
context lock long enough to sample the current CIL context. This is
kind of racy in that the CIL can switch the contexts immediately after
sampling, but that's ok because the consequence is that the defer ops
code is a little slow to relog items.
==================================================================
BUG: KASAN: use-after-free in xfs_log_item_in_current_chkpt+0x139/0x160 [xfs]
Read of size 8 at addr ffff88804ea5f608 by task fsstress/527999
CPU: 1 PID: 527999 Comm: fsstress Tainted: G D 5.16.0-rc4-xfsx #rc4
Call Trace:
<TASK>
dump_stack_lvl+0x45/0x59
print_address_description.constprop.0+0x1f/0x140
kasan_report.cold+0x83/0xdf
xfs_log_item_in_current_chkpt+0x139/0x160
xfs_defer_finish_noroll+0x3bb/0x1e30
__xfs_trans_commit+0x6c8/0xcf0
xfs_reflink_remap_extent+0x66f/0x10e0
xfs_reflink_remap_blocks+0x2dd/0xa90
xfs_file_remap_range+0x27b/0xc30
vfs_dedupe_file_range_one+0x368/0x420
vfs_dedupe_file_range+0x37c/0x5d0
do_vfs_ioctl+0x308/0x1260
__x64_sys_ioctl+0xa1/0x170
do_syscall_64+0x35/0x80
entry_SYSCALL_64_after_hwframe+0x44/0xae
RIP: 0033:0x7f2c71a2950b
Code: 0f 1e fa 48 8b 05 85 39 0d 00 64 c7 00 26 00 00 00 48 c7 c0 ff ff
ff ff c3 66 0f 1f 44 00 00 f3 0f 1e fa b8 10 00 00 00 0f 05 <48> 3d 01
f0 ff ff 73 01 c3 48 8b 0d 55 39 0d 00 f7 d8 64 89 01 48
RSP: 002b:00007ffe8c0e03c8 EFLAGS: 00000246 ORIG_RAX: 0000000000000010
RAX: ffffffffffffffda RBX: 00005600862a8740 RCX: 00007f2c71a2950b
RDX: 00005600862a7be0 RSI: 00000000c0189436 RDI: 0000000000000004
RBP: 000000000000000b R08: 0000000000000027 R09: 0000000000000003
R10: 0000000000000000 R11: 0000000000000246 R12: 000000000000005a
R13: 00005600862804a8 R14: 0000000000016000 R15: 00005600862a8a20
</TASK>
Allocated by task 464064:
kasan_save_stack+0x1e/0x50
__kasan_kmalloc+0x81/0xa0
kmem_alloc+0xcd/0x2c0 [xfs]
xlog_cil_ctx_alloc+0x17/0x1e0 [xfs]
xlog_cil_push_work+0x141/0x13d0 [xfs]
process_one_work+0x7f6/0x1380
worker_thread+0x59d/0x1040
kthread+0x3b0/0x490
ret_from_fork+0x1f/0x30
Freed by task 51:
kasan_save_stack+0x1e/0x50
kasan_set_track+0x21/0x30
kasan_set_free_info+0x20/0x30
__kasan_slab_free+0xed/0x130
slab_free_freelist_hook+0x7f/0x160
kfree+0xde/0x340
xlog_cil_committed+0xbfd/0xfe0 [xfs]
xlog_cil_process_committed+0x103/0x1c0 [xfs]
xlog_state_do_callback+0x45d/0xbd0 [xfs]
xlog_ioend_work+0x116/0x1c0 [xfs]
process_one_work+0x7f6/0x1380
worker_thread+0x59d/0x1040
kthread+0x3b0/0x490
ret_from_fork+0x1f/0x30
Last potentially related work creation:
kasan_save_stack+0x1e/0x50
__kasan_record_aux_stack+0xb7/0xc0
insert_work+0x48/0x2e0
__queue_work+0x4e7/0xda0
queue_work_on+0x69/0x80
xlog_cil_push_now.isra.0+0x16b/0x210 [xfs]
xlog_cil_force_seq+0x1b7/0x850 [xfs]
xfs_log_force_seq+0x1c7/0x670 [xfs]
xfs_file_fsync+0x7c1/0xa60 [xfs]
__x64_sys_fsync+0x52/0x80
do_syscall_64+0x35/0x80
entry_SYSCALL_64_after_hwframe+0x44/0xae
The buggy address belongs to the object at ffff88804ea5f600
which belongs to the cache kmalloc-256 of size 256
The buggy address is located 8 bytes inside of
256-byte region [ffff88804ea5f600, ffff88804ea5f700)
The buggy address belongs to the page:
page:ffffea00013a9780 refcount:1 mapcount:0 mapping:0000000000000000 index:0xffff88804ea5ea00 pfn:0x4ea5e
head:ffffea00013a9780 order:1 compound_mapcount:0
flags: 0x4fff80000010200(slab|head|node=1|zone=1|lastcpupid=0xfff)
raw: 04fff80000010200 ffffea0001245908 ffffea00011bd388 ffff888004c42b40
raw: ffff88804ea5ea00 0000000000100009 00000001ffffffff 0000000000000000
page dumped because: kasan: bad access detected
Memory state around the buggy address:
ffff88804ea5f500: fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc
ffff88804ea5f580: fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc
>ffff88804ea5f600: fa fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb
^
ffff88804ea5f680: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb
ffff88804ea5f700: fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc
==================================================================
Fixes: 4e919af7827a ("xfs: periodically relog deferred intent items")
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
|
|
The "bufsize" comes from the root user. If "bufsize" is negative then,
because of type promotion, neither of the validation checks at the start
of the function are able to catch it:
if (bufsize < sizeof(struct xfs_attrlist) ||
bufsize > XFS_XATTR_LIST_MAX)
return -EINVAL;
This means "bufsize" will trigger (WARN_ON_ONCE(size > INT_MAX)) in
kvmalloc_node(). Fix this by changing the type from int to size_t.
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
|
