fsnotify_foreach_iter_mark_type() is used to reduce boilerplate code
of iterating all marks of a specific group interested in an event
by consulting the iterator report_mask.
Use an open coded version of that iterator in fsnotify_iter_next()
that collects all marks of the current iteration group without
consulting the iterator report_mask.
At the moment, the two iterator variants are the same, but this
decoupling will allow us to exclude some of the group's marks from
reporting the event, for example for event on child and inode marks
on parent did not request to watch events on children.
Fixes: 2f02fd3fa1 ("fanotify: fix ignore mask logic for events on child and on dir")
Reported-by: Jan Kara <jack@suse.com>
Signed-off-by: Amir Goldstein <amir73il@gmail.com>
Signed-off-by: Jan Kara <jack@suse.cz>
Link: https://lore.kernel.org/r/20220511190213.831646-2-amir73il@gmail.com
fsnotify_add_mark() and variants implicitly take a reference on inode
when attaching a mark to an inode.
Make that behavior opt-out with the mark flag FSNOTIFY_MARK_FLAG_NO_IREF.
Instead of taking the inode reference when attaching connector to inode
and dropping the inode reference when detaching connector from inode,
take the inode reference on attach of the first mark that wants to hold
an inode reference and drop the inode reference on detach of the last
mark that wants to hold an inode reference.
Backends can "upgrade" an existing mark to take an inode reference, but
cannot "downgrade" a mark with inode reference to release the refernce.
This leaves the choice to the backend whether or not to pin the inode
when adding an inode mark.
This is intended to be used when adding a mark with ignored mask that is
used for optimization in cases where group can afford getting unneeded
events and reinstate the mark with ignored mask when inode is accessed
again after being evicted.
Link: https://lore.kernel.org/r/20220422120327.3459282-12-amir73il@gmail.com
Signed-off-by: Amir Goldstein <amir73il@gmail.com>
Signed-off-by: Jan Kara <jack@suse.cz>
Create helpers to take and release the group mark_mutex lock.
Define a flag FSNOTIFY_GROUP_NOFS in fsnotify_group that determines
if the mark_mutex lock is fs reclaim safe or not. If not safe, the
lock helpers take the lock and disable direct fs reclaim.
In that case we annotate the mutex with a different lockdep class to
express to lockdep that an allocation of mark of an fs reclaim safe group
may take the group lock of another "NOFS" group to evict inodes.
For now, converted only the callers in common code and no backend
defines the NOFS flag. It is intended to be set by fanotify for
evictable marks support.
Link: https://lore.kernel.org/r/20220422120327.3459282-7-amir73il@gmail.com
Suggested-by: Jan Kara <jack@suse.cz>
Link: https://lore.kernel.org/r/20220321112310.vpr7oxro2xkz5llh@quack3.lan/
Signed-off-by: Amir Goldstein <amir73il@gmail.com>
Signed-off-by: Jan Kara <jack@suse.cz>
Instead of passing the allow_dups argument to fsnotify_add_mark()
as an argument, define the group flag FSNOTIFY_GROUP_DUPS to express
the allow_dups behavior and set this behavior at group creation time
for all calls of fsnotify_add_mark().
Rename the allow_dups argument to generic add_flags argument for future
use.
Link: https://lore.kernel.org/r/20220422120327.3459282-6-amir73il@gmail.com
Suggested-by: Jan Kara <jack@suse.cz>
Signed-off-by: Amir Goldstein <amir73il@gmail.com>
Signed-off-by: Jan Kara <jack@suse.cz>
Add flags argument to fsnotify_alloc_group(), define and use the flag
FSNOTIFY_GROUP_USER in inotify and fanotify instead of the helper
fsnotify_alloc_user_group() to indicate user allocation.
Although the flag FSNOTIFY_GROUP_USER is currently not used after group
allocation, we store the flags argument in the group struct for future
use of other group flags.
Link: https://lore.kernel.org/r/20220422120327.3459282-5-amir73il@gmail.com
Signed-off-by: Amir Goldstein <amir73il@gmail.com>
Signed-off-by: Jan Kara <jack@suse.cz>
The inotify control flags in the mark mask (e.g. FS_IN_ONE_SHOT) are not
relevant to object interest mask, so move them to the mark flags.
This frees up some bits in the object interest mask.
Link: https://lore.kernel.org/r/20220422120327.3459282-3-amir73il@gmail.com
Signed-off-by: Amir Goldstein <amir73il@gmail.com>
Signed-off-by: Jan Kara <jack@suse.cz>
fsnotify() treats FS_MODIFY events specially - it does not skip them
even if the FS_MODIFY event does not apear in the object's fsnotify
mask. This is because send_to_group() checks if FS_MODIFY needs to
clear ignored mask of marks.
The common case is that an object does not have any mark with ignored
mask and in particular, that it does not have a mark with ignored mask
and without the FSNOTIFY_MARK_FLAG_IGNORED_SURV_MODIFY flag.
Set FS_MODIFY in object's fsnotify mask during fsnotify_recalc_mask()
if object has a mark with an ignored mask and without the
FSNOTIFY_MARK_FLAG_IGNORED_SURV_MODIFY flag and remove the special
treatment of FS_MODIFY in fsnotify(), so that FS_MODIFY events could
be optimized in the common case.
Call fsnotify_recalc_mask() from fanotify after adding or removing an
ignored mask from a mark without FSNOTIFY_MARK_FLAG_IGNORED_SURV_MODIFY
or when adding the FSNOTIFY_MARK_FLAG_IGNORED_SURV_MODIFY flag to a mark
with ignored mask (the flag cannot be removed by fanotify uapi).
Performance results for doing 10000000 write(2)s to tmpfs:
vanilla patched
without notification mark 25.486+-1.054 24.965+-0.244
with notification mark 30.111+-0.139 26.891+-1.355
So we can see the overhead of notification subsystem has been
drastically reduced.
Link: https://lore.kernel.org/r/20220223151438.790268-3-amir73il@gmail.com
Suggested-by: Jan Kara <jack@suse.cz>
Signed-off-by: Amir Goldstein <amir73il@gmail.com>
Signed-off-by: Jan Kara <jack@suse.cz>
fsnotify_parent() does not consider the parent's mark at all unless
the parent inode shows interest in events on children and in the
specific event.
So unless parent added an event to both its mark mask and ignored mask,
the event will not be ignored.
Fix this by declaring the interest of an object in an event when the
event is in either a mark mask or ignored mask.
Link: https://lore.kernel.org/r/20220223151438.790268-2-amir73il@gmail.com
Signed-off-by: Amir Goldstein <amir73il@gmail.com>
Signed-off-by: Jan Kara <jack@suse.cz>
The dnotify FS_DN_RENAME event is used to request notification about
a move within the same parent directory and was always coupled with
the FS_MOVED_FROM event.
Rename the FS_DN_RENAME event flag to FS_RENAME, decouple it from
FS_MOVED_FROM and report it with the moved dentry instead of the moved
inode, so it has the information about both old and new parent and name.
Generate the FS_RENAME event regardless of same parent dir and apply
the "same parent" rule in the generic fsnotify_handle_event() helper
that is used to call backends with ->handle_inode_event() method
(i.e. dnotify). The ->handle_inode_event() method is not rich enough to
report both old and new parent and name anyway.
The enriched event is reported to fanotify over the ->handle_event()
method with the old and new dir inode marks in marks array slots for
ITER_TYPE_INODE and a new iter type slot ITER_TYPE_INODE2.
The enriched event will be used for reporting old and new parent+name to
fanotify groups with FAN_RENAME events.
Link: https://lore.kernel.org/r/20211129201537.1932819-5-amir73il@gmail.com
Signed-off-by: Amir Goldstein <amir73il@gmail.com>
Signed-off-by: Jan Kara <jack@suse.cz>
They are two different types that use the same enum, so this confusing.
Use the object type to indicate the type of object mark is attached to
and the iter type to indicate the type of watch.
A group can have two different watches of the same object type (parent
and child watches) that match the same event.
Link: https://lore.kernel.org/r/20211129201537.1932819-3-amir73il@gmail.com
Signed-off-by: Amir Goldstein <amir73il@gmail.com>
Signed-off-by: Jan Kara <jack@suse.cz>
In preparation for separating object type from iterator type, rename
some 'type' arguments in functions to 'obj_type' and remove the unused
interface to clear marks by object type mask.
Link: https://lore.kernel.org/r/20211129201537.1932819-2-amir73il@gmail.com
Signed-off-by: Amir Goldstein <amir73il@gmail.com>
Signed-off-by: Jan Kara <jack@suse.cz>
Pre-allocate slots for file system errors to have greater chances of
succeeding, since error events can happen in GFP_NOFS context. This
patch introduces a group-wide mempool of error events, shared by all
FAN_FS_ERROR marks in this group.
Link: https://lore.kernel.org/r/20211025192746.66445-20-krisman@collabora.com
Reviewed-by: Amir Goldstein <amir73il@gmail.com>
Reviewed-by: Jan Kara <jack@suse.cz>
Signed-off-by: Gabriel Krisman Bertazi <krisman@collabora.com>
Signed-off-by: Jan Kara <jack@suse.cz>
Expose a new type of fsnotify event for filesystems to report errors for
userspace monitoring tools. fanotify will send this type of
notification for FAN_FS_ERROR events. This also introduce a helper for
generating the new event.
Link: https://lore.kernel.org/r/20211025192746.66445-18-krisman@collabora.com
Reviewed-by: Amir Goldstein <amir73il@gmail.com>
Reviewed-by: Jan Kara <jack@suse.cz>
Signed-off-by: Gabriel Krisman Bertazi <krisman@collabora.com>
Signed-off-by: Jan Kara <jack@suse.cz>
For group-wide mempool backed events, like FS_ERROR, the free_event
callback will need to reference the group's mempool to free the memory.
Wire that argument into the current callers.
Link: https://lore.kernel.org/r/20211025192746.66445-13-krisman@collabora.com
Reviewed-by: Jan Kara <jack@suse.cz>
Reviewed-by: Amir Goldstein <amir73il@gmail.com>
Signed-off-by: Gabriel Krisman Bertazi <krisman@collabora.com>
Signed-off-by: Jan Kara <jack@suse.cz>
FAN_FS_ERROR allows events without inodes - i.e. for file system-wide
errors. Even though fsnotify_handle_inode_event is not currently used
by fanotify, this patch protects other backends from cases where neither
inode or dir are provided. Also document the constraints of the
interface (inode and dir cannot be both NULL).
Link: https://lore.kernel.org/r/20211025192746.66445-12-krisman@collabora.com
Suggested-by: Amir Goldstein <amir73il@gmail.com>
Signed-off-by: Gabriel Krisman Bertazi <krisman@collabora.com>
Reviewed-by: Amir Goldstein <amir73il@gmail.com>
Reviewed-by: Jan Kara <jack@suse.cz>
Signed-off-by: Jan Kara <jack@suse.cz>
Some file system events (i.e. FS_ERROR) might not be associated with an
inode or directory. For these, we can retrieve the super block from the
data field. But, since the super_block is available in the data field
on every event type, simplify the code to always retrieve it from there,
through a new helper.
Link: https://lore.kernel.org/r/20211025192746.66445-11-krisman@collabora.com
Suggested-by: Jan Kara <jack@suse.cz>
Reviewed-by: Amir Goldstein <amir73il@gmail.com>
Reviewed-by: Jan Kara <jack@suse.cz>
Signed-off-by: Gabriel Krisman Bertazi <krisman@collabora.com>
Signed-off-by: Jan Kara <jack@suse.cz>
fsnotify_add_event is growing in number of parameters, which in most
case are just passed a NULL pointer. So, split out a new
fsnotify_insert_event function to clean things up for users who don't
need an insert hook.
Link: https://lore.kernel.org/r/20211025192746.66445-10-krisman@collabora.com
Suggested-by: Amir Goldstein <amir73il@gmail.com>
Reviewed-by: Amir Goldstein <amir73il@gmail.com>
Reviewed-by: Jan Kara <jack@suse.cz>
Signed-off-by: Gabriel Krisman Bertazi <krisman@collabora.com>
Signed-off-by: Jan Kara <jack@suse.cz>
Similarly to fanotify_is_perm_event and friends, provide a helper
predicate to say whether a mask is of an overflow event.
Link: https://lore.kernel.org/r/20211025192746.66445-9-krisman@collabora.com
Suggested-by: Amir Goldstein <amir73il@gmail.com>
Reviewed-by: Amir Goldstein <amir73il@gmail.com>
Reviewed-by: Jan Kara <jack@suse.cz>
Signed-off-by: Gabriel Krisman Bertazi <krisman@collabora.com>
Signed-off-by: Jan Kara <jack@suse.cz>
Define a new data type to pass for event - FSNOTIFY_EVENT_DENTRY.
Use it to pass the dentry instead of it's ->d_inode where available.
This is needed in preparation to the refactor to retrieve the super
block from the data field. In some cases (i.e. mkdir in kernfs), the
data inode comes from a negative dentry, such that no super block
information would be available. By receiving the dentry itself, instead
of the inode, fsnotify can derive the super block even on these cases.
Link: https://lore.kernel.org/r/20211025192746.66445-3-krisman@collabora.com
Reviewed-by: Jan Kara <jack@suse.cz>
Signed-off-by: Amir Goldstein <amir73il@gmail.com>
[Expand explanation in commit message]
Signed-off-by: Gabriel Krisman Bertazi <krisman@collabora.com>
Signed-off-by: Jan Kara <jack@suse.cz>
fanotify has some hardcoded limits. The only APIs to escape those limits
are FAN_UNLIMITED_QUEUE and FAN_UNLIMITED_MARKS.
Allow finer grained tuning of the system limits via sysfs tunables under
/proc/sys/fs/fanotify, similar to tunables under /proc/sys/fs/inotify,
with some minor differences.
- max_queued_events - global system tunable for group queue size limit.
Like the inotify tunable with the same name, it defaults to 16384 and
applies on initialization of a new group.
- max_user_marks - user ns tunable for marks limit per user.
Like the inotify tunable named max_user_watches, on a machine with
sufficient RAM and it defaults to 1048576 in init userns and can be
further limited per containing user ns.
- max_user_groups - user ns tunable for number of groups per user.
Like the inotify tunable named max_user_instances, it defaults to 128
in init userns and can be further limited per containing user ns.
The slightly different tunable names used for fanotify are derived from
the "group" and "mark" terminology used in the fanotify man pages and
throughout the code.
Considering the fact that the default value for max_user_instances was
increased in kernel v5.10 from 8192 to 1048576, leaving the legacy
fanotify limit of 8192 marks per group in addition to the max_user_marks
limit makes little sense, so the per group marks limit has been removed.
Note that when a group is initialized with FAN_UNLIMITED_MARKS, its own
marks are not accounted in the per user marks account, so in effect the
limit of max_user_marks is only for the collection of groups that are
not initialized with FAN_UNLIMITED_MARKS.
Link: https://lore.kernel.org/r/20210304112921.3996419-2-amir73il@gmail.com
Suggested-by: Jan Kara <jack@suse.cz>
Signed-off-by: Amir Goldstein <amir73il@gmail.com>
Signed-off-by: Jan Kara <jack@suse.cz>
In order to improve event merge performance, hash events in a 128 size
hash table by the event merge key.
The fanotify_event size grows by two pointers, but we just reduced its
size by removing the objectid member, so overall its size is increased
by one pointer.
Permission events and overflow event are not merged so they are also
not hashed.
Link: https://lore.kernel.org/r/20210304104826.3993892-5-amir73il@gmail.com
Signed-off-by: Amir Goldstein <amir73il@gmail.com>
Signed-off-by: Jan Kara <jack@suse.cz>
objectid is only used by fanotify backend and it is just an optimization
for event merge before comparing all fields in event.
Move the objectid member from common struct fsnotify_event into struct
fanotify_event and reduce it to 29-bit hash to cram it together with the
3-bit event type.
Events of different types are never merged, so the combination of event
type and hash form a 32-bit key for fast compare of events.
This reduces the size of events by one pointer and paves the way for
adding hashed queue support for fanotify.
Link: https://lore.kernel.org/r/20210304104826.3993892-3-amir73il@gmail.com
Signed-off-by: Amir Goldstein <amir73il@gmail.com>
Signed-off-by: Jan Kara <jack@suse.cz>
Current code has an assumtion that fsnotify_notify_queue_is_empty() is
called to verify that queue is not empty before trying to peek or remove
an event from queue.
Remove this assumption by moving the fsnotify_notify_queue_is_empty()
into the functions, allow them to return NULL value and check return
value by all callers.
This is a prep patch for multi event queues.
Link: https://lore.kernel.org/r/20210304104826.3993892-2-amir73il@gmail.com
Signed-off-by: Amir Goldstein <amir73il@gmail.com>
Signed-off-by: Jan Kara <jack@suse.cz>
Currently the fs sysctl inotify/max_user_instances is used to limit the
number of inotify instances on the system. For systems running multiple
workloads, the per-user namespace sysctl max_inotify_instances can be
used to further partition inotify instances. However there is no easy
way to set a sensible system level max limit on inotify instances and
further partition it between the workloads. It is much easier to charge
the underlying resource (i.e. memory) behind the inotify instances to
the memcg of the workload and let their memory limits limit the number
of inotify instances they can create.
With inotify instances charged to memcg, the admin can simply set
max_user_instances to INT_MAX and let the memcg limits of the jobs limit
their inotify instances.
Link: https://lore.kernel.org/r/20201220044608.1258123-1-shakeelb@google.com
Reviewed-by: Amir Goldstein <amir73il@gmail.com>
Signed-off-by: Shakeel Butt <shakeelb@google.com>
Signed-off-by: Jan Kara <jack@suse.cz>
fsnotify_parent() used to send two separate events to backends when a
parent inode is watching children and the child inode is also watching.
In an attempt to avoid duplicate events in fanotify, we unified the two
backend callbacks to a single callback and handled the reporting of the
two separate events for the relevant backends (inotify and dnotify).
However the handling is buggy and can result in inotify and dnotify
listeners receiving events of the type they never asked for or spurious
events.
The problem is the unified event callback with two inode marks (parent and
child) is called when any of the parent and child inodes are watched and
interested in the event, but the parent inode's mark that is interested
in the event on the child is not necessarily the one we are currently
reporting to (it could belong to a different group).
So before reporting the parent or child event flavor to backend we need
to check that the mark is really interested in that event flavor.
The semantics of INODE and CHILD marks were hard to follow and made the
logic more complicated than it should have been. Replace it with INODE
and PARENT marks semantics to hopefully make the logic more clear.
Thanks to Hugh Dickins for spotting a bug in the earlier version of this
patch.
Fixes: 497b0c5a7c ("fsnotify: send event to parent and child with single callback")
CC: stable@vger.kernel.org
Link: https://lore.kernel.org/r/20201202120713.702387-4-amir73il@gmail.com
Reported-by: Hugh Dickins <hughd@google.com>
Signed-off-by: Amir Goldstein <amir73il@gmail.com>
Signed-off-by: Jan Kara <jack@suse.cz>
The handle_inode_event() interface was added as (quoting comment):
"a simple variant of handle_event() for groups that only have inode
marks and don't have ignore mask".
In other words, all backends except fanotify. The inotify backend
also falls under this category, but because it required extra arguments
it was left out of the initial pass of backends conversion to the
simple interface.
This results in code duplication between the generic helper
fsnotify_handle_event() and the inotify_handle_event() callback
which also happen to be buggy code.
Generalize the handle_inode_event() arguments and add the check for
FS_EXCL_UNLINK flag to the generic helper, so inotify backend could
be converted to use the simple interface.
Link: https://lore.kernel.org/r/20201202120713.702387-2-amir73il@gmail.com
CC: stable@vger.kernel.org
Fixes: b9a1b97725 ("fsnotify: create method handle_inode_event() in fsnotify_operations")
Signed-off-by: Amir Goldstein <amir73il@gmail.com>
Signed-off-by: Jan Kara <jack@suse.cz>
The method handle_event() grew a lot of complexity due to the design of
fanotify and merging of ignore masks.
Most backends do not care about this complex functionality, so we can hide
this complexity from them.
Introduce a method handle_inode_event() that serves those backends and
passes a single inode mark and less arguments.
This change converts all backends except fanotify and inotify to use the
simplified handle_inode_event() method. In pricipal, inotify could have
also used the new method, but that would require passing more arguments
on the simple helper (data, data_type, cookie), so we leave it with the
handle_event() method.
Link: https://lore.kernel.org/r/20200722125849.17418-9-amir73il@gmail.com
Suggested-by: Jan Kara <jack@suse.cz>
Signed-off-by: Amir Goldstein <amir73il@gmail.com>
Signed-off-by: Jan Kara <jack@suse.cz>
Similar to events "on child" to watching directory, send event
with parent/name info if sb/mount/non-dir marks are interested in
parent/name info.
The FS_EVENT_ON_CHILD flag can be set on sb/mount/non-dir marks to specify
interest in parent/name info for events on non-directory inodes.
Events on "orphan" children (disconnected dentries) are sent without
parent/name info.
Events on directories are sent with parent/name info only if the parent
directory is watching.
After this change, even groups that do not subscribe to events on
children could get an event with mark iterator type TYPE_CHILD and
without mark iterator type TYPE_INODE if fanotify has marks on the same
objects.
dnotify and inotify event handlers can already cope with that situation.
audit does not subscribe to events that are possible on child, so won't
get to this situation. nfsd does not access the marks iterator from its
event handler at the moment, so it is not affected.
This is a bit too fragile, so we should prepare all groups to cope with
mark type TYPE_CHILD preferably using a generic helper.
Link: https://lore.kernel.org/r/20200716084230.30611-16-amir73il@gmail.com
Signed-off-by: Amir Goldstein <amir73il@gmail.com>
Signed-off-by: Jan Kara <jack@suse.cz>
The arguments of fsnotify() are overloaded and mean different things
for different event types.
Replace the to_tell argument with separate arguments @dir and @inode,
because we may be sending to both dir and child. Using the @data
argument to pass the child is not enough, because dirent events pass
this argument (for audit), but we do not report to child.
Document the new fsnotify() function argumenets.
Link: https://lore.kernel.org/r/20200722125849.17418-7-amir73il@gmail.com
Signed-off-by: Amir Goldstein <amir73il@gmail.com>
Signed-off-by: Jan Kara <jack@suse.cz>
The object type iterator is used to collect all the marks of
a specific group that have interest in an event.
It is used by fanotify to get a single handle_event callback
when an event has a match to either of inode/sb/mount marks
of the group.
The nature of fsnotify events is that they are associated with
at most one sb at most one mount and at most one inode.
When a parent and child are both watching, two events are sent
to backend, one associated to parent inode and one associated
to the child inode.
This results in duplicate events in fanotify, which usually
get merged before user reads them, but this is sub-optimal.
It would be better if the same event is sent to backend with
an object type iterator that has both the child inode and its
parent, and let the backend decide if the event should be reported
once (fanotify) or twice (inotify).
Link: https://lore.kernel.org/r/20200716084230.30611-9-amir73il@gmail.com
Signed-off-by: Amir Goldstein <amir73il@gmail.com>
Signed-off-by: Jan Kara <jack@suse.cz>
It was never enabled in uapi and its functionality is about to be
superseded by events FAN_CREATE, FAN_DELETE, FAN_MOVE with group
flag FAN_REPORT_NAME.
Keep a place holder variable name_event instead of removing the
name recording code since it will be used by the new events.
Link: https://lore.kernel.org/r/20200708111156.24659-17-amir73il@gmail.com
Signed-off-by: Amir Goldstein <amir73il@gmail.com>
Signed-off-by: Jan Kara <jack@suse.cz>
The 'inode' argument to handle_event(), sometimes referred to as
'to_tell' is somewhat obsolete.
It is a remnant from the times when a group could only have an inode mark
associated with an event.
We now pass an iter_info array to the callback, with all marks associated
with an event.
Most backends ignore this argument, with two exceptions:
1. dnotify uses it for sanity check that event is on directory
2. fanotify uses it to report fid of directory on directory entry
modification events
Remove the 'inode' argument and add a 'dir' argument.
The callback function signature is deliberately changed, because
the meaning of the argument has changed and the arguments have
been documented.
The 'dir' argument is set to when 'file_name' is specified and it is
referring to the directory that the 'file_name' entry belongs to.
Signed-off-by: Amir Goldstein <amir73il@gmail.com>
Signed-off-by: Jan Kara <jack@suse.cz>
Return non const inode pointer from fsnotify_data_inode().
None of the fsnotify hooks pass const inode pointer as data and
callers often need to cast to a non const pointer.
Link: https://lore.kernel.org/r/20200708111156.24659-3-amir73il@gmail.com
Signed-off-by: Amir Goldstein <amir73il@gmail.com>
Signed-off-by: Jan Kara <jack@suse.cz>
The fsnotify paths are trivial to hit even when there are no watchers and
they are surprisingly expensive. For example, every successful vfs_write()
hits fsnotify_modify which calls both fsnotify_parent and fsnotify unless
FMODE_NONOTIFY is set which is an internal flag invisible to userspace.
As it stands, fsnotify_parent is a guaranteed functional call even if there
are no watchers and fsnotify() does a substantial amount of unnecessary
work before it checks if there are any watchers. A perf profile showed
that applying mnt->mnt_fsnotify_mask in fnotify() was almost half of the
total samples taken in that function during a test. This patch rearranges
the fast paths to reduce the amount of work done when there are no
watchers.
The test motivating this was "perf bench sched messaging --pipe". Despite
the fact the pipes are anonymous, fsnotify is still called a lot and
the overhead is noticeable even though it's completely pointless. It's
likely the overhead is negligible for real IO so this is an extreme
example. This is a comparison of hackbench using processes and pipes on
a 1-socket machine with 8 CPU threads without fanotify watchers.
5.7.0 5.7.0
vanilla fastfsnotify-v1r1
Amean 1 0.4837 ( 0.00%) 0.4630 * 4.27%*
Amean 3 1.5447 ( 0.00%) 1.4557 ( 5.76%)
Amean 5 2.6037 ( 0.00%) 2.4363 ( 6.43%)
Amean 7 3.5987 ( 0.00%) 3.4757 ( 3.42%)
Amean 12 5.8267 ( 0.00%) 5.6983 ( 2.20%)
Amean 18 8.4400 ( 0.00%) 8.1327 ( 3.64%)
Amean 24 11.0187 ( 0.00%) 10.0290 * 8.98%*
Amean 30 13.1013 ( 0.00%) 12.8510 ( 1.91%)
Amean 32 13.9190 ( 0.00%) 13.2410 ( 4.87%)
5.7.0 5.7.0
vanilla fastfsnotify-v1r1
Duration User 157.05 152.79
Duration System 1279.98 1219.32
Duration Elapsed 182.81 174.52
This is showing that the latencies are improved by roughly 2-9%. The
variability is not shown but some of these results are within the noise
as this workload heavily overloads the machine. That said, the system CPU
usage is reduced by quite a bit so it makes sense to avoid the overhead
even if it is a bit tricky to detect at times. A perf profile of just 1
group of tasks showed that 5.14% of samples taken were in either fsnotify()
or fsnotify_parent(). With the patch, 2.8% of samples were in fsnotify,
mostly function entry and the initial check for watchers. The check for
watchers is complicated enough that inlining it may be controversial.
[Amir] Slightly simplify with mnt_or_sb_mask => marks_mask
Link: https://lore.kernel.org/r/20200708111156.24659-1-amir73il@gmail.com
Signed-off-by: Mel Gorman <mgorman@techsingularity.net>
Signed-off-by: Amir Goldstein <amir73il@gmail.com>
Signed-off-by: Jan Kara <jack@suse.cz>
Dirent events are going to be supported in two flavors:
1. Directory fid info + mask that includes the specific event types
(e.g. FAN_CREATE) and an optional FAN_ONDIR flag.
2. Directory fid info + name + mask that includes only FAN_DIR_MODIFY.
To request the second event flavor, user needs to set the event type
FAN_DIR_MODIFY in the mark mask.
The first flavor is supported since kernel v5.1 for groups initialized
with flag FAN_REPORT_FID. It is intended to be used for watching
directories in "batch mode" - the watcher is notified when directory is
changed and re-scans the directory content in response. This event
flavor is stored more compactly in the event queue, so it is optimal
for workloads with frequent directory changes.
The second event flavor is intended to be used for watching large
directories, where the cost of re-scan of the directory on every change
is considered too high. The watcher getting the event with the directory
fid and entry name is expected to call fstatat(2) to query the content of
the entry after the change.
Legacy inotify events are reported with name and event mask (e.g. "foo",
FAN_CREATE | FAN_ONDIR). That can lead users to the conclusion that
there is *currently* an entry "foo" that is a sub-directory, when in fact
"foo" may be negative or non-dir by the time user gets the event.
To make it clear that the current state of the named entry is unknown,
when reporting an event with name info, fanotify obfuscates the specific
event types (e.g. create,delete,rename) and uses a common event type -
FAN_DIR_MODIFY to describe the change. This should make it harder for
users to make wrong assumptions and write buggy filesystem monitors.
At this point, name info reporting is not yet implemented, so trying to
set FAN_DIR_MODIFY in mark mask will return -EINVAL.
Link: https://lore.kernel.org/r/20200319151022.31456-12-amir73il@gmail.com
Signed-off-by: Amir Goldstein <amir73il@gmail.com>
Signed-off-by: Jan Kara <jack@suse.cz>
The event inode field is used only for comparison in queue merges and
cannot be dereferenced after handle_event(), because it does not hold a
refcount on the inode.
Replace it with an abstract id to do the same thing.
Link: https://lore.kernel.org/r/20200319151022.31456-8-amir73il@gmail.com
Signed-off-by: Amir Goldstein <amir73il@gmail.com>
Signed-off-by: Jan Kara <jack@suse.cz>
Instead of passing both dentry and path and having to figure out which
one to use, pass data/data_type to simplify the code.
Link: https://lore.kernel.org/r/20200319151022.31456-6-amir73il@gmail.com
Signed-off-by: Amir Goldstein <amir73il@gmail.com>
Signed-off-by: Jan Kara <jack@suse.cz>
The knfsd file cache will need to detect when files are unlinked, so that
it can close the associated cached files. Export a minimal set of notifier
functions to allow it to do so.
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>
Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
For all callers of fsnotify_{unlink,rmdir}(), we made sure that d_parent
and d_name are stable. Therefore, fsnotify_{unlink,rmdir}() do not need
the safety measures in fsnotify_nameremove() to stabilize parent and name.
We can now simplify those hooks and get rid of fsnotify_nameremove().
Signed-off-by: Amir Goldstein <amir73il@gmail.com>
Signed-off-by: Jan Kara <jack@suse.cz>
When implementing connector fsid cache, we only initialized the cache
when the first mark added to object was added by FAN_REPORT_FID group.
We forgot to update conn->fsid when the second mark is added by
FAN_REPORT_FID group to an already attached connector without fsid
cache.
Reported-and-tested-by: syzbot+c277e8e2f46414645508@syzkaller.appspotmail.com
Fixes: 77115225ac ("fanotify: cache fsid in fsnotify_mark_connector")
Signed-off-by: Amir Goldstein <amir73il@gmail.com>
Signed-off-by: Jan Kara <jack@suse.cz>
note that conditions surrounding accesses to dname in audit_watch_handle_event()
and audit_mark_handle_event() guarantee that dname won't have been NULL.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Note that in fnsotify_move() and fsnotify_link() we are guaranteed
that dentry->d_name won't change during the fsnotify() evaluation
(by having the parent directory locked exclusive), so we don't
need to fetch dentry->d_name.name in the callers. In fsnotify_dirent()
the same stability of dentry->d_name is also true, but it's a bit
more convoluted - there is one callchain (devpts_pty_new() ->
fsnotify_create() -> fsnotify_dirent()) where the parent is _not_
locked, but on devpts ->d_name of everything is unchanging; it
has neither explicit nor implicit renames.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Create function to remove event from the notification list. Later it will
be used from more places.
Reviewed-by: Amir Goldstein <amir73il@gmail.com>
Signed-off-by: Jan Kara <jack@suse.cz>
For FAN_REPORT_FID, we need to encode fid with fsid of the filesystem on
every event. To avoid having to call vfs_statfs() on every event to get
fsid, we store the fsid in fsnotify_mark_connector on the first time we
add a mark and on handle event we use the cached fsid.
Subsequent calls to add mark on the same object are expected to pass the
same fsid, so the call will fail on cached fsid mismatch.
If an event is reported on several mark types (inode, mount, filesystem),
all connectors should already have the same fsid, so we use the cached
fsid from the first connector.
[JK: Simplify code flow around fanotify_get_fid()
make fsid argument of fsnotify_add_mark_locked() unconditional]
Suggested-by: Jan Kara <jack@suse.cz>
Signed-off-by: Amir Goldstein <amir73il@gmail.com>
Signed-off-by: Jan Kara <jack@suse.cz>
Common fsnotify_event helpers have no need for the mask field.
It is only used by backend code, so move the field out of the
abstract fsnotify_event struct and into the concrete backend
event structs.
This change packs struct inotify_event_info better on 64bit
machine and will allow us to cram some more fields into
struct fanotify_event_info.
Signed-off-by: Amir Goldstein <amir73il@gmail.com>
Signed-off-by: Jan Kara <jack@suse.cz>
