Move inode inactivation to background work contexts so that it no
longer runs in the context that releases the final reference to an
inode. This will allow process work that ends up blocking on
inactivation to continue doing work while the filesytem processes
the inactivation in the background.
A typical demonstration of this is unlinking an inode with lots of
extents. The extents are removed during inactivation, so this blocks
the process that unlinked the inode from the directory structure. By
moving the inactivation to the background process, the userspace
applicaiton can keep working (e.g. unlinking the next inode in the
directory) while the inactivation work on the previous inode is
done by a different CPU.
The implementation of the queue is relatively simple. We use a
per-cpu lockless linked list (llist) to queue inodes for
inactivation without requiring serialisation mechanisms, and a work
item to allow the queue to be processed by a CPU bound worker
thread. We also keep a count of the queue depth so that we can
trigger work after a number of deferred inactivations have been
queued.
The use of a bound workqueue with a single work depth allows the
workqueue to run one work item per CPU. We queue the work item on
the CPU we are currently running on, and so this essentially gives
us affine per-cpu worker threads for the per-cpu queues. THis
maintains the effective CPU affinity that occurs within XFS at the
AG level due to all objects in a directory being local to an AG.
Hence inactivation work tends to run on the same CPU that last
accessed all the objects that inactivation accesses and this
maintains hot CPU caches for unlink workloads.
A depth of 32 inodes was chosen to match the number of inodes in an
inode cluster buffer. This hopefully allows sequential
allocation/unlink behaviours to defering inactivation of all the
inodes in a single cluster buffer at a time, further helping
maintain hot CPU and buffer cache accesses while running
inactivations.
A hard per-cpu queue throttle of 256 inode has been set to avoid
runaway queuing when inodes that take a long to time inactivate are
being processed. For example, when unlinking inodes with large
numbers of extents that can take a lot of processing to free.
Signed-off-by: Dave Chinner <dchinner@redhat.com>
[djwong: tweak comments and tracepoints, convert opflags to state bits]
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
If we don't need to inactivate an inode, we can detach the dquots and
move on to reclamation. This isn't strictly required here; it's a
preparation patch for deferred inactivation per reviewer request[1] to
move the creation of xfs_inode_needs_inactivation into a separate
change. Eventually this !need_inactive chunk will turn into the code
path for inodes that skip xfs_inactive and go straight to memory
reclaim.
[1] https://lore.kernel.org/linux-xfs/20210609012838.GW2945738@locust/T/#mca6d958521cb88bbc1bfe1a30767203328d410b5
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
This is just a simple wrapper around the per-ag inode allocation
that doesn't need to exist. The internal mechanism to select and
allocate within an AG does not need to be exposed outside
xfs_ialloc.c, and it being exposed simply makes it harder to follow
the code and simplify it.
This is simplified by internalising xf_dialloc_select_ag() and
xfs_dialloc_ag() into a single xfs_dialloc() function and then
xfs_dir_ialloc() can go away.
Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Brian Foster <bfoster@redhat.com>
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Move the crtime field from struct xfs_icdinode into stuct xfs_inode and
remove the now entirely unused struct xfs_icdinode.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
In preparation of removing the historic icinode struct, move the flags2
field into the containing xfs_inode structure.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
In preparation of removing the historic icinode struct, move the flags
field into the containing xfs_inode structure.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
In preparation of removing the historic icinode struct, move the
forkoff field into the containing xfs_inode structure.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
The i_cowextsize field is only used for v3 inodes, and the i_flushiter
field is only used for v1/v2 inodes. Use a union to pack the inode a
littler better after adding a few missing guards around their usage.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
In preparation of removing the historic icinode struct, move the
flushiter field into the containing xfs_inode structure.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
In preparation of removing the historic icinode struct, move the
cowextsize field into the containing xfs_inode structure. Also
switch to use the xfs_extlen_t instead of a uint32_t.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
In preparation of removing the historic icinode struct, move the extsize
field into the containing xfs_inode structure.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
In preparation of removing the historic icinode struct, move the nblocks
field into the containing xfs_inode structure.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
In preparation of removing the historic icinode struct, move the on-disk
size field into the containing xfs_inode structure.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
In preparation of removing the historic icinode struct, move the projid
field into the containing xfs_inode structure.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
The legacy DMAPI fields were never set by upstream Linux XFS, and have no
way to be read using the kernel APIs. So instead of bloating the in-core
inode for them just copy them from the on-disk inode into the log when
logging the inode. The only caveat is that we need to make sure to zero
the fields for newly read or deleted inodes, which is solved using a new
flag in the inode.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
When we allocate a new inode, we often need to add an attribute to
the inode as part of the create. This can happen as a result of
needing to add default ACLs or security labels before the inode is
made visible to userspace.
This is highly inefficient right now. We do the create transaction
to allocate the inode, then we do an "add attr fork" transaction to
modify the just created empty inode to set the inode fork offset to
allow attributes to be stored, then we go and do the attribute
creation.
This means 3 transactions instead of 1 to allocate an inode, and
this greatly increases the load on the CIL commit code, resulting in
excessive contention on the CIL spin locks and performance
degradation:
18.99% [kernel] [k] __pv_queued_spin_lock_slowpath
3.57% [kernel] [k] do_raw_spin_lock
2.51% [kernel] [k] __raw_callee_save___pv_queued_spin_unlock
2.48% [kernel] [k] memcpy
2.34% [kernel] [k] xfs_log_commit_cil
The typical profile resulting from running fsmark on a selinux enabled
filesytem is adds this overhead to the create path:
- 15.30% xfs_init_security
- 15.23% security_inode_init_security
- 13.05% xfs_initxattrs
- 12.94% xfs_attr_set
- 6.75% xfs_bmap_add_attrfork
- 5.51% xfs_trans_commit
- 5.48% __xfs_trans_commit
- 5.35% xfs_log_commit_cil
- 3.86% _raw_spin_lock
- do_raw_spin_lock
__pv_queued_spin_lock_slowpath
- 0.70% xfs_trans_alloc
0.52% xfs_trans_reserve
- 5.41% xfs_attr_set_args
- 5.39% xfs_attr_set_shortform.constprop.0
- 4.46% xfs_trans_commit
- 4.46% __xfs_trans_commit
- 4.33% xfs_log_commit_cil
- 2.74% _raw_spin_lock
- do_raw_spin_lock
__pv_queued_spin_lock_slowpath
0.60% xfs_inode_item_format
0.90% xfs_attr_try_sf_addname
- 1.99% selinux_inode_init_security
- 1.02% security_sid_to_context_force
- 1.00% security_sid_to_context_core
- 0.92% sidtab_entry_to_string
- 0.90% sidtab_sid2str_get
0.59% sidtab_sid2str_put.part.0
- 0.82% selinux_determine_inode_label
- 0.77% security_transition_sid
0.70% security_compute_sid.part.0
And fsmark creation rate performance drops by ~25%. The key point to
note here is that half the additional overhead comes from adding the
attribute fork to the newly created inode. That's crazy, considering
we can do this same thing at inode create time with a couple of
lines of code and no extra overhead.
So, if we know we are going to add an attribute immediately after
creating the inode, let's just initialise the attribute fork inside
the create transaction and chop that whole chunk of code out of
the create fast path. This completely removes the performance
drop caused by enabling SELinux, and the profile looks like:
- 8.99% xfs_init_security
- 9.00% security_inode_init_security
- 6.43% xfs_initxattrs
- 6.37% xfs_attr_set
- 5.45% xfs_attr_set_args
- 5.42% xfs_attr_set_shortform.constprop.0
- 4.51% xfs_trans_commit
- 4.54% __xfs_trans_commit
- 4.59% xfs_log_commit_cil
- 2.67% _raw_spin_lock
- 3.28% do_raw_spin_lock
3.08% __pv_queued_spin_lock_slowpath
0.66% xfs_inode_item_format
- 0.90% xfs_attr_try_sf_addname
- 0.60% xfs_trans_alloc
- 2.35% selinux_inode_init_security
- 1.25% security_sid_to_context_force
- 1.21% security_sid_to_context_core
- 1.19% sidtab_entry_to_string
- 1.20% sidtab_sid2str_get
- 0.86% sidtab_sid2str_put.part.0
- 0.62% _raw_spin_lock_irqsave
- 0.77% do_raw_spin_lock
__pv_queued_spin_lock_slowpath
- 0.84% selinux_determine_inode_label
- 0.83% security_transition_sid
0.86% security_compute_sid.part.0
Which indicates the XFS overhead of creating the selinux xattr has
been halved. This doesn't fix the CIL lock contention problem, just
means it's not a limiting factor for this workload. Lock contention
in the security subsystems is going to be an issue soon, though...
Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
[djwong: fix compilation error when CONFIG_SECURITY=n]
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Gao Xiang <hsiangkao@redhat.com>
Files containing metadata (quota records, rt bitmap and summary info)
are fully managed by the filesystem, which means that all resource
cleanup must be explicit, not automatic. This means that they should
never be subjected automatic to post-eof truncation, nor should they be
freed automatically even if the link count drops to zero.
In other words, xfs_inactive() should leave these files alone. Add the
necessary predicate functions to make this happen. This adds a second
layer of prevention for the kinds of fs corruption that was fixed by
commit f4c32e87de. If we ever decide to support removing metadata
files, we should make all those metadata updates explicit.
Rearrange the order of #includes to fix compiler errors, since
xfs_mount.h is supposed to be included before xfs_inode.h
Followup-to: f4c32e87de ("xfs: fix realtime bitmap/summary file truncation when growing rt volume")
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Enable idmapped mounts for xfs. This basically just means passing down
the user_namespace argument from the VFS methods down to where it is
passed to the relevant helpers.
Note that full-filesystem bulkstat is not supported from inside idmapped
mounts as it is an administrative operation that acts on the whole file
system. The limitation is not applied to the bulkstat single operation
that just operates on a single inode.
Link: https://lore.kernel.org/r/20210121131959.646623-40-christian.brauner@ubuntu.com
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Christian Brauner <christian.brauner@ubuntu.com>
So xfs_ialloc() will only address in-core inode allocation then,
Also, rename xfs_ialloc() to xfs_dir_ialloc_init() in order to
keep everything in xfs_inode.c under the same namespace.
Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Dave Chinner <dchinner@redhat.com>
Signed-off-by: Gao Xiang <hsiangkao@redhat.com>
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
Redesign the ondisk inode timestamps to be a simple unsigned 64-bit
counter of nanoseconds since 14 Dec 1901 (i.e. the minimum time in the
32-bit unix time epoch). This enables us to handle dates up to 2486,
which solves the y2038 problem.
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Gao Xiang <hsiangkao@redhat.com>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
With the recent rework of the inode cluster flushing, we no longer
ever wait on the the inode flush "lock". It was never a lock in the
first place, just a completion to allow callers to wait for inode IO
to complete. We now never wait for flush completion as all inode
flushing is non-blocking. Hence we can get rid of all the iflock
infrastructure and instead just set and check a state flag.
Rename the XFS_IFLOCK flag to XFS_IFLUSHING, convert all the
xfs_iflock_nowait() test-and-set operations on that flag, and
replace all the xfs_ifunlock() calls to clear operations.
Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
Now that we have all the dirty inodes attached to the cluster
buffer, we don't actually have to do radix tree lookups to find
them. Sure, the radix tree is efficient, but walking a linked list
of just the dirty inodes attached to the buffer is much better.
We are also no longer dependent on having a locked inode passed into
the function to determine where to start the lookup. This means we
can drop it from the function call and treat all inodes the same.
We also make xfs_iflush_cluster skip inodes marked with
XFS_IRECLAIM. This we avoid races with inodes that reclaim is
actively referencing or are being re-initialised by inode lookup. If
they are actually dirty, they'll get written by a future cluster
flush....
We also add a shutdown check after obtaining the flush lock so that
we catch inodes that are dirty in memory and may have inconsistent
state due to the shutdown in progress. We abort these inodes
directly and so they remove themselves directly from the buffer list
and the AIL rather than having to wait for the buffer to be failed
and callbacks run to be processed correctly.
Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
Now we have a cached buffer on inode log items, we don't need
to do buffer lookups when flushing inodes anymore - all we need
to do is lock the buffer and we are ready to go.
This largely gets rid of the need for xfs_iflush(), which is
essentially just a mechanism to look up the buffer and flush the
inode to it. Instead, we can just call xfs_iflush_cluster() with a
few modifications to ensure it also flushes the inode we already
hold locked.
This allows the AIL inode item pushing to be almost entirely
non-blocking in XFS - we won't block unless memory allocation
for the cluster inode lookup blocks or the block device queues are
full.
Writeback during inode reclaim becomes a little more complex because
we now have to lock the buffer ourselves, but otherwise this change
is largely a functional no-op that removes a whole lot of code.
Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com>
Reviewed-by: Brian Foster <bfoster@redhat.com>
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
Move the double-inode locking helpers to xfs_inode.c since they're not
specific to reflink.
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
Reviewed-by: Brian Foster <bfoster@redhat.com>
- Introduce DONTCACHE flags for dentries and inodes. This hint will
cause the VFS to drop the associated objects immediately after the
last put, so that we can change the file access mode (DAX or page
cache) on the fly.
-----BEGIN PGP SIGNATURE-----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=ebl3
-----END PGP SIGNATURE-----
Merge tag 'vfs-5.8-merge-2' of git://git.kernel.org/pub/scm/fs/xfs/xfs-linux
Pull DAX updates part two from Darrick Wong:
"This time around, we're hoisting the DONTCACHE flag from XFS into the
VFS so that we can make the incore DAX mode changes become effective
sooner.
We can't change the file data access mode on a live inode because we
don't have a safe way to change the file ops pointers. The incore
state change becomes effective at inode loading time, which can happen
if the inode is evicted. Therefore, we're making it so that
filesystems can ask the VFS to evict the inode as soon as the last
holder drops.
The per-fs changes to make this call this will be in subsequent pull
requests from Ted and myself.
Summary:
- Introduce DONTCACHE flags for dentries and inodes. This hint will
cause the VFS to drop the associated objects immediately after the
last put, so that we can change the file access mode (DAX or page
cache) on the fly"
* tag 'vfs-5.8-merge-2' of git://git.kernel.org/pub/scm/fs/xfs/xfs-linux:
fs: Introduce DCACHE_DONTCACHE
fs: Lift XFS_IDONTCACHE to the VFS layer
Both the data and attr fork have a format that is stored in the legacy
idinode. Move it into the xfs_ifork structure instead, where it uses
up padding.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Brian Foster <bfoster@redhat.com>
Reviewed-by: Chandan Babu R <chandanrlinux@gmail.com>
Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
There are there are three extents counters per inode, one for each of
the forks. Two are in the legacy icdinode and one is directly in
struct xfs_inode. Switch to a single counter in the xfs_ifork structure
where it uses up padding at the end of the structure. This simplifies
various bits of code that just wants the number of extents counter and
can now directly dereference it.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Chandan Babu R <chandanrlinux@gmail.com>
Reviewed-by: Brian Foster <bfoster@redhat.com>
Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
Call the data/attr local fork verifiers as soon as we are ready for them.
This keeps them close to the code setting up the forks, and avoids a
few branches later on. Also open code xfs_inode_verify_forks in the
only remaining caller.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Brian Foster <bfoster@redhat.com>
Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
DAX effective mode (S_DAX) changes requires inode eviction.
XFS has an advisory flag (XFS_IDONTCACHE) to prevent caching of the
inode if no other additional references are taken. We lift this flag to
the VFS layer and change the behavior slightly by allowing the flag to
remain even if multiple references are taken.
This will expedite the eviction of inodes to change S_DAX.
Cc: Al Viro <viro@zeniv.linux.org.uk>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Jan Kara <jack@suse.cz>
Signed-off-by: Ira Weiny <ira.weiny@intel.com>
Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
The functionality in xfs_diflags_to_linux() and xfs_diflags_to_iflags() are
nearly identical. The only difference is that *_to_linux() is called after
inode setup and disallows changing the DAX flag.
Combining them can be done with a flag which indicates if this is the initial
setup to allow the DAX flag to be properly set only at init time.
So remove xfs_diflags_to_linux() and call the modified xfs_diflags_to_iflags()
directly.
While we are here simplify xfs_diflags_to_iflags() to take struct xfs_inode and
use xfs_ip2xflags() to ensure future diflags are included correctly.
Reviewed-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: Ira Weiny <ira.weiny@intel.com>
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
Create a new helper to force the log up to the last LSN touching an
inode.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Brian Foster <bfoster@redhat.com>
Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
There is no point in splitting the fields like this in an purely
in-memory structure.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
Add a new xfs_inode_buftarg helper that gets the data I/O buftarg for a
given inode. Replace the existing xfs_find_bdev_for_inode and
xfs_find_daxdev_for_inode helpers with this new general one and cleanup
some of the callers.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
Widen the incore inode's i_delayed_blks counter to be a 64-bit integer.
This is necessary to fix an integer overflow problem that can be
reproduced easily now that we use the counter to track blocks that are
assigned to the inode in memory but not on disk. This includes actual
delalloc reservations as well as real extents in the COW fork that
are waiting to be remapped into the data fork.
These 'delayed mapping' blocks can easily exceed 2^32 blocks if one
creates a very large sparse file of size approximately 2^33 bytes with
one byte written every 2^23 bytes, sets a very large COW extent size
hint of 2^23 blocks, reflinks the first file into a second file, and
then writes a single byte every 2^23 blocks in the original file.
When this happens, we'll try to create approximately 1024 2^23 extent
reservations in the COW fork, which will overflow the counter and cause
problems.
Note that on x64 we end up filling a 4-byte gap in the structure so this
doesn't increase the incore size.
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
Reviewed-by: Allison Collins <allison.henderson@oracle.com>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
When scheduling writeback of dirty file data in the page cache, XFS uses
IO completion workqueue items to ensure that filesystem metadata only
updates after the write completes successfully. This is essential for
converting unwritten extents to real extents at the right time and
performing COW remappings.
Unfortunately, XFS queues each IO completion work item to an unbounded
workqueue, which means that the kernel can spawn dozens of threads to
try to handle the items quickly. These threads need to take the ILOCK
to update file metadata, which results in heavy ILOCK contention if a
large number of the work items target a single file, which is
inefficient.
Worse yet, the writeback completion threads get stuck waiting for the
ILOCK while holding transaction reservations, which can use up all
available log reservation space. When that happens, metadata updates to
other parts of the filesystem grind to a halt, even if the filesystem
could otherwise have handled it.
Even worse, if one of the things grinding to a halt happens to be a
thread in the middle of a defer-ops finish holding the same ILOCK and
trying to obtain more log reservation having exhausted the permanent
reservation, we now have an ABBA deadlock - writeback completion has a
transaction reserved and wants the ILOCK, and someone else has the ILOCK
and wants a transaction reservation.
Therefore, we create a per-inode writeback io completion queue + work
item. When writeback finishes, it can add the ioend to the per-inode
queue and let the single worker item process that queue. This
dramatically cuts down on the number of kworkers and ILOCK contention in
the system, and seems to have eliminated an occasional deadlock I was
seeing while running generic/476.
Testing with a program that simulates a heavy random-write workload to a
single file demonstrates that the number of kworkers drops from
approximately 120 threads per file to 1, without dramatically changing
write bandwidth or pagecache access latency.
Note that we leave the xfs-conv workqueue's max_active alone because we
still want to be able to run ioend processing for as many inodes as the
system can handle.
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
Reviewed-by: Brian Foster <bfoster@redhat.com>
Add the necessary in-core metadata fields to keep track of which parts
of the filesystem have been observed and which parts were observed to be
unhealthy, and print a warning at unmount time if we have unfixed
problems.
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
Reviewed-by: Brian Foster <bfoster@redhat.com>
Use a rhashtable to cache the unlinked list incore. This should speed
up unlinked processing considerably when there are a lot of inodes on
the unlinked list because iunlink_remove no longer has to traverse an
entire bucket list to find which inode points to the one being removed.
The incore list structure records "X.next_unlinked = Y" relations, with
the rhashtable using Y to index the records. This makes finding the
inode X that points to a inode Y very quick. If our cache fails to find
anything we can always fall back on the old method.
FWIW this drastically reduces the amount of time it takes to remove
inodes from the unlinked list. I wrote a program to open a lot of
O_TMPFILE files and then close them in the same order, which takes
a very long time if we have to traverse the unlinked lists. With the
ptach, I see:
+ /d/t/tmpfile/tmpfile
Opened 193531 files in 6.33s.
Closed 193531 files in 5.86s
real 0m12.192s
user 0m0.064s
sys 0m11.619s
+ cd /
+ umount /mnt
real 0m0.050s
user 0m0.004s
sys 0m0.030s
And without the patch:
+ /d/t/tmpfile/tmpfile
Opened 193588 files in 6.35s.
Closed 193588 files in 751.61s
real 12m38.853s
user 0m0.084s
sys 12m34.470s
+ cd /
+ umount /mnt
real 0m0.086s
user 0m0.000s
sys 0m0.060s
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
Reviewed-by: Brian Foster <bfoster@redhat.com>
struct xfs_defer_ops has now been reduced to a single list_head. The
external dfops mechanism is unused and thus everywhere a (permanent)
transaction is accessible the associated dfops structure is as well.
Remove the xfs_defer_ops structure and fold the list_head into the
transaction. Also remove the last remnant of external dfops in
xfs_trans_dup().
Signed-off-by: Brian Foster <bfoster@redhat.com>
Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
We have a few places that already check if an inode has actual data in
the COW fork to avoid work on reflink inodes that do not actually have
outstanding COW blocks. There are a few more places that can avoid
working if doing the same check, so add a documented helper for this
condition and use it in all places where it makes sense.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
We only have a few more callers left, so seize the opportunity and kill
it off.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
Replace the IRELE macro with a proper function so that we can do proper
typechecking and so that we can stop open-coding iput in scrub, which
means that we'll be able to ftrace inode lifetimes going through scrub
correctly.
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
Reviewed-by: Carlos Maiolino <cmaiolino@redhat.com>
Reviewed-by: Brian Foster <bfoster@redhat.com>
The inode free callchain starting in xfs_inactive_ifree() already
associates its dfops with the transaction. It still passes the dfops
on the stack down through xfs_difree_inobt(), however.
Clean up the call stack and reference dfops directly from the
transaction. This patch does not change behavior.
Signed-off-by: Brian Foster <bfoster@redhat.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
- Strengthen metadata checking to avoid ASSERTing on bad disk contents
- Validate btree records that are being retrieved for clients
- Strengthen root inode verification
- Convert license blurbs to SPDX tags
- Enable changing DAX flag on directories
- Fix some writeback deadlocks in reflink
- Refactor out some old xfs helpers
- Move type verifiers to a separate file
- Fix some fuzzer crashes
- Various other bug fixes
-----BEGIN PGP SIGNATURE-----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=SD3u
-----END PGP SIGNATURE-----
Merge tag 'xfs-4.18-merge-10' of git://git.kernel.org/pub/scm/fs/xfs/xfs-linux
Pull more xfs updates from Darrick Wong:
"Here's the second round of patches for XFS for 4.18. Most of the
commits are small cleanups, bug fixes, and continued strengthening of
metadata verifiers; the bulk of the diff is the conversion of the
fs/xfs/ tree to use SPDX tags.
This series has been run through a full xfstests run over the weekend
and through a quick xfstests run against this morning's master, with
no major failures reported.
Summary:
- Strengthen metadata checking to avoid ASSERTing on bad disk
contents
- Validate btree records that are being retrieved for clients
- Strengthen root inode verification
- Convert license blurbs to SPDX tags
- Enable changing DAX flag on directories
- Fix some writeback deadlocks in reflink
- Refactor out some old xfs helpers
- Move type verifiers to a separate file
- Fix some fuzzer crashes
- Various other bug fixes"
* tag 'xfs-4.18-merge-10' of git://git.kernel.org/pub/scm/fs/xfs/xfs-linux: (31 commits)
xfs: update incore per-AG inode count
xfs: replace do_mod with native operations
xfs: don't call xfs_da_shrink_inode with NULL bp
xfs: clean up MIN/MAX
xfs: move various type verifiers to common file
xfs: xfs_reflink_convert_cow() memory allocation deadlock
xfs: setup VFS i_rwsem lockdep state correctly
xfs: fix string handling in label get/set functions
xfs: convert to SPDX license tags
xfs: validate btree records on retrieval
xfs: push corruption -> ESTALE conversion to xfs_nfs_get_inode()
xfs: verify root inode more thoroughly
xfs: verify COW extent size hint is valid in inode verifier
xfs: verify extent size hint is valid in inode verifier
xfs: catch bad stripe alignment configurations
iomap: fsync swap files before iterating mappings
xfs: use xfs_trans_getsb in xfs_sync_sb_buf
xfs: don't assert on corrupted unlinked inode list
xfs: explicitly pass buffer size to xfs_corruption_error
xfs: don't assert when on-disk btree pointers are garbage
...
* DAX broke a fundamental assumption of truncate of file mapped pages.
The truncate path assumed that it is safe to disconnect a pinned page
from a file and let the filesystem reclaim the physical block. With DAX
the page is equivalent to the filesystem block. Introduce
dax_layout_busy_page() to enable filesystems to wait for pinned DAX
pages to be released. Without this wait a filesystem could allocate
blocks under active device-DMA to a new file.
* DAX arranges for the block layer to be bypassed and uses
dax_direct_access() + copy_to_iter() to satisfy read(2) calls.
However, the memcpy_mcsafe() facility is available through the pmem
block driver. In order to safely handle media errors, via the DAX
block-layer bypass, introduce copy_to_iter_mcsafe().
* Fix cache management policy relative to the ACPI NFIT Platform
Capabilities Structure to properly elide cache flushes when they are not
necessary. The table indicates whether CPU caches are power-fail
protected. Clarify that a deep flush is always performed on
REQ_{FUA,PREFLUSH} requests.
-----BEGIN PGP SIGNATURE-----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=4bfw
-----END PGP SIGNATURE-----
Merge tag 'libnvdimm-for-4.18' of git://git.kernel.org/pub/scm/linux/kernel/git/nvdimm/nvdimm
Pull libnvdimm updates from Dan Williams:
"This adds a user for the new 'bytes-remaining' updates to
memcpy_mcsafe() that you already received through Ingo via the
x86-dax- for-linus pull.
Not included here, but still targeting this cycle, is support for
handling memory media errors (poison) consumed via userspace dax
mappings.
Summary:
- DAX broke a fundamental assumption of truncate of file mapped
pages. The truncate path assumed that it is safe to disconnect a
pinned page from a file and let the filesystem reclaim the physical
block. With DAX the page is equivalent to the filesystem block.
Introduce dax_layout_busy_page() to enable filesystems to wait for
pinned DAX pages to be released. Without this wait a filesystem
could allocate blocks under active device-DMA to a new file.
- DAX arranges for the block layer to be bypassed and uses
dax_direct_access() + copy_to_iter() to satisfy read(2) calls.
However, the memcpy_mcsafe() facility is available through the pmem
block driver. In order to safely handle media errors, via the DAX
block-layer bypass, introduce copy_to_iter_mcsafe().
- Fix cache management policy relative to the ACPI NFIT Platform
Capabilities Structure to properly elide cache flushes when they
are not necessary. The table indicates whether CPU caches are
power-fail protected. Clarify that a deep flush is always performed
on REQ_{FUA,PREFLUSH} requests"
* tag 'libnvdimm-for-4.18' of git://git.kernel.org/pub/scm/linux/kernel/git/nvdimm/nvdimm: (21 commits)
dax: Use dax_write_cache* helpers
libnvdimm, pmem: Do not flush power-fail protected CPU caches
libnvdimm, pmem: Unconditionally deep flush on *sync
libnvdimm, pmem: Complete REQ_FLUSH => REQ_PREFLUSH
acpi, nfit: Remove ecc_unit_size
dax: dax_insert_mapping_entry always succeeds
libnvdimm, e820: Register all pmem resources
libnvdimm: Debug probe times
linvdimm, pmem: Preserve read-only setting for pmem devices
x86, nfit_test: Add unit test for memcpy_mcsafe()
pmem: Switch to copy_to_iter_mcsafe()
dax: Report bytes remaining in dax_iomap_actor()
dax: Introduce a ->copy_to_iter dax operation
uio, lib: Fix CONFIG_ARCH_HAS_UACCESS_MCSAFE compilation
xfs, dax: introduce xfs_break_dax_layouts()
xfs: prepare xfs_break_layouts() for another layout type
xfs: prepare xfs_break_layouts() to be called with XFS_MMAPLOCK_EXCL
mm, fs, dax: handle layout changes to pinned dax mappings
mm: fix __gup_device_huge vs unmap
mm: introduce MEMORY_DEVICE_FS_DAX and CONFIG_DEV_PAGEMAP_OPS
...
Remove the verbose license text from XFS files and replace them
with SPDX tags. This does not change the license of any of the code,
merely refers to the common, up-to-date license files in LICENSES/
This change was mostly scripted. fs/xfs/Makefile and
fs/xfs/libxfs/xfs_fs.h were modified by hand, the rest were detected
and modified by the following command:
for f in `git grep -l "GNU General" fs/xfs/` ; do
echo $f
cat $f | awk -f hdr.awk > $f.new
mv -f $f.new $f
done
And the hdr.awk script that did the modification (including
detecting the difference between GPL-2.0 and GPL-2.0+ licenses)
is as follows:
$ cat hdr.awk
BEGIN {
hdr = 1.0
tag = "GPL-2.0"
str = ""
}
/^ \* This program is free software/ {
hdr = 2.0;
next
}
/any later version./ {
tag = "GPL-2.0+"
next
}
/^ \*\// {
if (hdr > 0.0) {
print "// SPDX-License-Identifier: " tag
print str
print $0
str=""
hdr = 0.0
next
}
print $0
next
}
/^ \* / {
if (hdr > 1.0)
next
if (hdr > 0.0) {
if (str != "")
str = str "\n"
str = str $0
next
}
print $0
next
}
/^ \*/ {
if (hdr > 0.0)
next
print $0
next
}
// {
if (hdr > 0.0) {
if (str != "")
str = str "\n"
str = str $0
next
}
print $0
}
END { }
$
Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
When xfs is operating as the back-end of a pNFS block server, it
prevents collisions between local and remote operations by requiring a
lease to be held for remotely accessed blocks. Local filesystem
operations break those leases before writing or mutating the extent map
of the file.
A similar mechanism is needed to prevent operations on pinned dax
mappings, like device-DMA, from colliding with extent unmap operations.
BREAK_WRITE and BREAK_UNMAP are introduced as two distinct levels of
layout breaking.
Layouts are broken in the BREAK_WRITE case to ensure that layout-holders
do not collide with local writes. Additionally, layouts are broken in
the BREAK_UNMAP case to make sure the layout-holder has a consistent
view of the file's extent map. While BREAK_WRITE breaks can be satisfied
be recalling FL_LAYOUT leases, BREAK_UNMAP breaks additionally require
waiting for busy dax-pages to go idle while holding XFS_MMAPLOCK_EXCL.
After this refactoring xfs_break_layouts() becomes the entry point for
coordinating both types of breaks. Finally, xfs_break_leased_layouts()
becomes just the BREAK_WRITE handler.
Note that the unlock tracking is needed in a follow on change. That will
coordinate retrying either break handler until both successfully test
for a lease break while maintaining the lock state.
Cc: Ross Zwisler <ross.zwisler@linux.intel.com>
Cc: "Darrick J. Wong" <darrick.wong@oracle.com>
Reported-by: Dave Chinner <david@fromorbit.com>
Reported-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
The changes to skip discards of speculative preallocation and
unwritten extents introduced several new wrapper functions through
the bunmapi -> extent free codepath to reduce churn in all of the
associated callers. In several cases, these wrappers simply toggle a
single flag to skip or not skip discards for the resulting blocks.
The explicit _nodiscard() wrappers for such an isolated set of
callers is a bit overkill. Kill off these wrappers and replace with
the calls to the underlying functions in the contexts that need to
control discard behavior. Retain the wrappers that preserve the
original calling conventions to serve the original purpose of
reducing code churn.
This is a refactoring patch and does not change behavior.
Signed-off-by: Brian Foster <bfoster@redhat.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
We've had reports of online discard operations being sent from XFS
on write-only workloads. These discards occur as a result of
eofblocks trims that can occur after a large file copy completes.
These discards are slightly confusing for users who might be paying
close attention to online discards (i.e., vdo) due to performance
sensitivity. They also happen to be spurious because freed post-eof
blocks by definition have not been written to during the current
allocation cycle.
Update xfs_free_eofblocks() to skip discards that are purely
attributed to eofblocks trims. This cuts down the number of spurious
discards that may occur on write-only workloads due to normal
preallocation activity.
Note that discards of post-eof extents can still occur from other
codepaths that do not isolate handling of post-eof blocks from those
within eof. For example, file unlinks and truncates may still cause
discards for any file blocks affected by the operation.
Signed-off-by: Brian Foster <bfoster@redhat.com>
Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
