Rename to_shrink_list() into __move_to_shrink_list(), document and
export it. Switch d_dispose_if_unused() users to that and kill
d_dispose_if_unused() itself.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Refcount of a NORCU dentry must not be incremented after having dropped
to zero. Otherwise we might end up with the following race:
CPU1: in fast_dput(d), rcu_read_lock();
CPU1: decrements refcount of d to 0
CPU1: notice that it's unhashed
CPU2: grab a reference to d
CPU2: dput(d), freeing d
CPU1: ... looks like we need to evict d, let's grab ->d_lock, recheck
the refcount, etc.
and that spin_lock(&d->d_lock) ends up a UAF, despite still being in
an RCU read-side critical area started back when the refcount had been
positive. If not for DCACHE_NORCU in d->d_flags freeing would've been
RCU-delayed, so we'd have grabbed ->d_lock, noticed the negative value
stored into refcount by __dentry_kill(), dropped the locks and that would
be it. For NORCU dentries freeing is _not_ delayed, though.
Most of the non-counting references are excluded for NORCU dentries -
they are not allowed to be hashed, they never get placed on LRU, they
never get placed into anyone's list of children and while dput_to_list()
might put them into a shrink list, nobody bumps refcount of something
that had been reached that way.
However, inode's list of aliases can be a problem - it does not contribute
to dentry refcount (for obvious reasons) and we *do* have places that
grab references to something found on that list - that's precisely what
d_find_alias() is. In case of d_find_alias() we are safe - it skips
unhashed aliases, so all NORCU ones are ignored there. d_find_any_alias()
is *not* limited to hashed ones, though, and while it's usually called
for directories (which never get NORCU dentries), there are callers that
use it to get something for non-directories with no hashed aliases.
Having d_find_any_alias() hit a NORCU dentry is not impossible - it can
be easily arranged if you have CAP_DAC_READ_SEARCH (memfd_create() + mmap()
+ name_to_handle_at() for /proc/self/map_files/<...> + munmap() +
open_by_handle_at() will do that, and adding a second memfd_create() for
mount_fd makes it possible to do that without having memfd pinned).
The race window is narrow, and it's probably not feasible on bare hardware,
but...
It's not hard to fix, fortunately:
* separate __d_find_dir_alias() (== current __d_find_any_alias()) to
be used for directory inodes.
* provide dget_alias_ilocked() that would return false for NORCU
dentries with zero refcount and return true incrementing refcount otherwise
* make __d_find_any_alias() go over the list of aliases, using
dget_alias_ilocked() and returning the alias it succeeds on (normally the
first one). Any NORCU alias with zero refcount is going to be evicted by
the thread that had dropped the final reference; this makes __d_find_any_alias()
pretend it had lost the race with eviction.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Parallel lookup starts with a call of d_alloc_parallel(). That primitive
either returns a matching hashed dentry or allocates a new one in the
in-lookup state and returns it to the caller. Once the caller is done
with lookup, it indicates so either by call of d_{splice_alias,add}()
or by call of d_done_lookup(); at that point dentry leaves the in-lookup
state.
If d_alloc_parallel() finds a matching in-lookup dentry, it must wait for
that dentry to leave the in-lookup state, one way or another. Currently
by supplying wait_queue_head to d_alloc_parallel(). If d_alloc_parallel()
creates a new in-lookup dentry, the address of that wait_queue_head is stored
in ->d_wait of new dentry and stays there while it's in the in-lookup;
subsequent d_alloc_parallel() will wait on the queue found in the matching
in-lookup dentry. Transition out of in-lookup state wakes waiters on that
queue (if any).
That works, but the calling conventions are inconvenient - the caller must
supply wait_queue_head and make sure that it survives at least until the new
in-lookup dentry leaves the in-lookup state. That amounts to boilerplate
in the d_alloc_parallel() callers that are followed by a call of d_lookup_done()
in the same function; in cases like nfs asynchronous unlink it gets worse than
that.
This patch changes d_alloc_parallel() to use wake_up_var_locked() to
wake up waiters, and wait_var_event_spinlock() to wait. dentry->d_lock
is used for synchronisation as it is already held and the relevant
times.
That eliminates the need of caller-supplied wait_queue_head, simplifying
the calling conventions. Better yet, we only need one bit of information
stored in dentry itself: whether there are any waiters to be woken up,
and that can be easily stored in ->d_flags; ->d_wait goes away.
The reason we need that bit (DCACHE_LOOKUP_WAITERS) is that with wait_var
machinery the queues are shared with all kinds of stuff and there's
no way tell if any of the waiters have anything to do with our dentry;
most of the time none of them will be relevant, so we need to avoid the
pointless wakeups.
Another benefit of the new scheme comes from the fact that wakeups
have to be done outside of write-side critical areas of ->i_dir_seq;
with the old scheme we need to carry the value picked from ->d_wait from
__d_lookup_unhash() to the place where we actually wake the waiters up.
Now we can just leave DCACHE_LOOKUP_WAITERS in ->d_flags until we get
to doing wakeups - that's done within the same ->d_lock scope, so we
are fine; new bit is accessed only under ->d_lock and it's seen only
on dentries with DCACHE_PAR_LOOKUP in ->d_flags.
__d_lookup_unhash() no longer needs to re-init ->d_lru. That was
previously shared (in a union) with ->d_wait but ->d_wait is now gone
so it no longer corrupts ->d_lru.
Co-developed-by: Al Viro <viro@zeniv.linux.org.uk> # saner handling of flags
Signed-off-by: NeilBrown <neil@brown.name>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Pull dcache busy loop updates from Al Viro:
"Fix livelocks in shrink_dcache_tree()
If shrink_dcache_tree() finds a dentry in the middle of being killed
by another thread, it has to wait until the victim finishes dying,
gets detached from the tree and ceases to pin its parent.
The way we used to deal with that amounted to busy-wait;
unfortunately, it's not just inefficient but can lead to reliably
reproducible hard livelocks.
Solved by having shrink_dentry_tree() attach a completion to such
dentry, with dentry_unlist() calling complete() on all objects
attached to it. With a bit of care it can be done without growing
struct dentry or adding overhead in normal case"
* tag 'pull-dcache-busy-wait' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs:
get rid of busy-waiting in shrink_dcache_tree()
dcache.c: more idiomatic "positives are not allowed" sanity checks
struct dentry: make ->d_u anonymous
for_each_alias(): helper macro for iterating through dentries of given inode
If shrink_dcache_tree() runs into a potential victim that is already
dying, it must wait for that dentry to go away. To avoid busy-waiting
we need some object to wait on and a way for dentry_unlist() to see that
we need to be notified.
The obvious place for the object to wait on would be on our stack frame.
We will store a pointer to that object (struct completion_list) in victim
dentry; if there's more than one thread wanting to wait for the same
dentry to finish dying, we'll have their instances linked into a list,
with reference in dentry pointing to the head of that list.
* new object - struct completion_list. A pair of struct completion and
pointer to the next instance. That's what shrink_dcache_tree() will wait
on if needed.
* add a new member (->waiters, opaque pointer to struct completion_list)
to struct dentry. It is defined for negative live dentries that are
not in-lookup ones and it will remain NULL for almost all of them.
It does not conflict with ->d_rcu (defined for killed dentries), ->d_alias
(defined for positive dentries, all live) or ->d_in_lookup_hash (defined
for in-lookup dentries, all live negative). That allows to colocate
all four members.
* make sure that all places where dentry enters the state where ->waiters
is defined (live, negative, not-in-lookup) initialize ->waiters to NULL.
* if select_collect2() runs into a dentry that is already dying, have
its caller insert a local instance of struct completion_list into the
head of the list hanging off dentry->waiters and wait for completion.
* if dentry_unlist() sees non-NULL ->waiters, have it carefully walk
through the completion_list instances in that list, calling complete()
for each.
For now struct completion_list is local to fs/dcache.c; it's obviously
dentry-agnostic, and it can be trivially lifted into linux/completion.h
if somebody finds a reason to do so...
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Making ->d_rcu and (then) ->d_child overlapping dates back to
2006; anon unions support had been added to gcc only in 4.6
(2011) and the minimal gcc version hadn't been bumped to that
until 4.19 (2018).
These days there's no reason not to keep that union named.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Most of the places using d_alias are loops iterating through all aliases for
given inode; introduce a helper macro (for_each_alias(dentry, inode))
and convert open-coded instances of such loop to it.
They are easier to read that way and it reduces the noise on the next steps.
You _must_ hold inode->i_lock over that thing.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Pull fuse updates from Miklos Szeredi:
- Add mechanism for cleaning out unused, stale dentries; controlled via
a module option (Luis Henriques)
- Fix various bugs
- Cleanups
* tag 'fuse-update-6.19' of git://git.kernel.org/pub/scm/linux/kernel/git/mszeredi/fuse:
fuse: Uninitialized variable in fuse_epoch_work()
fuse: fix io-uring list corruption for terminated non-committed requests
fuse: signal that a fuse inode should exhibit local fs behaviors
fuse: Always flush the page cache before FOPEN_DIRECT_IO write
fuse: Invalidate the page cache after FOPEN_DIRECT_IO write
fuse: rename 'namelen' to 'namesize'
fuse: use strscpy instead of strcpy
fuse: refactor fuse_conn_put() to remove negative logic.
fuse: new work queue to invalidate dentries from old epochs
fuse: new work queue to periodically invalidate expired dentries
dcache: export shrink_dentry_list() and add new helper d_dispose_if_unused()
fuse: add WARN_ON and comment for RCU revalidate
fuse: Fix whitespace for fuse_uring_args_to_ring() comment
fuse: missing copy_finish in fuse-over-io-uring argument copies
fuse: fix readahead reclaim deadlock
* d_make_persistent(dentry, inode) - bump refcount, mark persistent and
make hashed positive. Return value is a borrowed reference to dentry;
it can be used until something removes persistency (at the very least,
until the parent gets unlocked, but some filesystems may have stronger
exclusion).
* d_make_discardable() - remove persistency mark and drop reference.
d_make_persistent() is similar to combination of d_instantiate(), dget()
and setting flag. The only difference is that unlike d_instantiate()
it accepts hashed and unhashed negatives alike. It is always called in
strong locking environment (parent held exclusive, or, in some cases,
dentry coming from d_alloc_name()); if we ever start using it with parent
held only shared and dentry coming from d_alloc_parallel(), we'll need
to copy the in-lookup logics from __d_add().
d_make_discardable() is eqiuvalent to combination of removing flag and
dput(); since flag removal requires ->d_lock, there's no point trying
to avoid taking that for refcount decrement as fast_dput() does.
The slow path of dput() has been taken into a helper and reused in
d_make_discardable() instead.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Some filesystems use a kinda-sorta controlled dentry refcount leak to pin
dentries of created objects in dcache (and undo it when removing those).
Reference is grabbed and not released, but it's not actually _stored_
anywhere. That works, but it's hard to follow and verify; among other
things, we have no way to tell _which_ of the increments is intended
to be an unpaired one. Worse, on removal we need to decide whether
the reference had already been dropped, which can be non-trivial if
that removal is on umount and we need to figure out if this dentry is
pinned due to e.g. unlink() not done. Usually that is handled by using
kill_litter_super() as ->kill_sb(), but there are open-coded special
cases of the same (consider e.g. /proc/self).
Things get simpler if we introduce a new dentry flag (DCACHE_PERSISTENT)
marking those "leaked" dentries. Having it set claims responsibility
for +1 in refcount.
The end result this series is aiming for:
* get these unbalanced dget() and dput() replaced with new primitives that
would, in addition to adjusting refcount, set and clear persistency flag.
* instead of having kill_litter_super() mess with removing the remaining
"leaked" references (e.g. for all tmpfs files that hadn't been removed
prior to umount), have the regular shrink_dcache_for_umount() strip
DCACHE_PERSISTENT of all dentries, dropping the corresponding
reference if it had been set. After that kill_litter_super() becomes
an equivalent of kill_anon_super().
Doing that in a single step is not feasible - it would affect too many places
in too many filesystems. It has to be split into a series.
Here we
* introduce the new flag
* teach shrink_dcache_for_umount() to handle it (i.e. remove
and drop refcount on anything that survives to umount with that flag
still set)
* teach kill_litter_super() that anything with that flag does
*not* need to be unpinned.
Next commits will add primitives for maintaing that flag and convert the
common helpers to those. After that - a long series of per-filesystem
patches converting to those primitives.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Add and export a new helper d_dispose_if_unused() which is simply a wrapper
around to_shrink_list(), to add an entry to a dispose list if it's not used
anymore.
Also export shrink_dentry_list() to kill all dentries in a dispose list.
Suggested-by: Miklos Szeredi <miklos@szeredi.hu>
Signed-off-by: Luis Henriques <luis@igalia.com>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
Turn d_name into an anon union of const struct qstr d_name with
struct qstr __d_name. Very few places need to modify it (all
in fs/dcache.c); those are switched to use of ->__d_name.
Note that ->d_name can actually change under you unless you have
the right locking environment; this const just prohibits accidentally
doing stores without being easily spotted.
Reviewed-by: Christian Brauner <brauner@kernel.org>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Convert the last user (d_alloc_pseudo()) and be done with that.
Any out-of-tree filesystem using it should switch to d_splice_alias_ops()
or, better yet, check whether it really needs to have ->d_op vary among
its dentries.
Reviewed-by: Christian Brauner <brauner@kernel.org>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
... to be used instead of manually assigning to ->s_d_op.
All in-tree filesystem converted (and field itself is renamed,
so any out-of-tree ones in need of conversion will be caught
by compiler).
Reviewed-by: Christian Brauner <brauner@kernel.org>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Uses of d_set_d_op() on live dentry can be very dangerous; it is going
to be withdrawn and replaced with saner things.
The best way for a filesystem is to have the default dentry_operations
set at mount time and be done with that - __d_alloc() will use that.
Currently there are two cases when d_set_d_op() is used on a live dentry -
one is procfs, which has several genuinely different dentry_operations
instances (different ->d_revalidate(), etc.) and another is
simple_lookup(), where we would be better off without overriding ->d_op.
For procfs we have d_set_d_op() calls followed by d_splice_alias();
provide a new helper (d_splice_alias_ops(inode, dentry, d_ops)) that would
combine those two, and do the d_set_d_op() part while under ->d_lock.
That eliminates one of the places where ->d_flags had been modified
without holding ->d_lock; current behaviour is not racy, but the reasons
for that are far too brittle. Better move to uniform locking rules and
simpler proof of correctness...
The next commit will convert procfs to use of that helper; it is not
exported and won't be until somebody comes up with convincing modular
user for it.
Again, the best approach is to have default ->d_op and let __d_alloc()
do the right thing; filesystem _may_ need non-uniform ->d_op (procfs
does), but there'd better be good reasons for that.
Reviewed-by: Christian Brauner <brauner@kernel.org>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Pull vfs directory lookup updates from Christian Brauner:
"This contains cleanups for the lookup_one*() family of helpers.
We expose a set of functions with names containing "lookup_one_len"
and others without the "_len". This difference has nothing to do with
"len". It's rater a historical accident that can be confusing.
The functions without "_len" take a "mnt_idmap" pointer. This is found
in the "vfsmount" and that is an important question when choosing
which to use: do you have a vfsmount, or are you "inside" the
filesystem. A related question is "is permission checking relevant
here?".
nfsd and cachefiles *do* have a vfsmount but *don't* use the non-_len
functions. They pass nop_mnt_idmap and refuse to work on filesystems
which have any other idmap.
This work changes nfsd and cachefile to use the lookup_one family of
functions and to explictily pass &nop_mnt_idmap which is consistent
with all other vfs interfaces used where &nop_mnt_idmap is explicitly
passed.
The remaining uses of the "_one" functions do not require permission
checks so these are renamed to be "_noperm" and the permission
checking is removed.
This series also changes these lookup function to take a qstr instead
of separate name and len. In many cases this simplifies the call"
* tag 'vfs-6.16-rc1.async.dir' of git://git.kernel.org/pub/scm/linux/kernel/git/vfs/vfs:
VFS: change lookup_one_common and lookup_noperm_common to take a qstr
Use try_lookup_noperm() instead of d_hash_and_lookup() outside of VFS
VFS: rename lookup_one_len family to lookup_noperm and remove permission check
cachefiles: Use lookup_one() rather than lookup_one_len()
nfsd: Use lookup_one() rather than lookup_one_len()
VFS: improve interface for lookup_one functions
try_lookup_noperm() and d_hash_and_lookup() are nearly identical. The
former does some validation of the name where the latter doesn't.
Outside of the VFS that validation is likely valuable, and having only
one exported function for this task is certainly a good idea.
So make d_hash_and_lookup() local to VFS files and change all other
callers to try_lookup_noperm(). Note that the arguments are swapped.
Signed-off-by: NeilBrown <neilb@suse.de>
Link: https://lore.kernel.org/r/20250319031545.2999807-6-neil@brown.name
Signed-off-by: Christian Brauner <brauner@kernel.org>
The family of functions:
lookup_one()
lookup_one_unlocked()
lookup_one_positive_unlocked()
appear designed to be used by external clients of the filesystem rather
than by filesystems acting on themselves as the lookup_one_len family
are used.
They are used by:
btrfs/ioctl - which is a user-space interface rather than an internal
activity
exportfs - i.e. from nfsd or the open_by_handle_at interface
overlayfs - at access the underlying filesystems
smb/server - for file service
They should be used by nfsd (more than just the exportfs path) and
cachefs but aren't.
It would help if the documentation didn't claim they should "not be
called by generic code".
Also the path component name is passed as "name" and "len" which are
(confusingly?) separate by the "base". In some cases the len in simply
"strlen" and so passing a qstr using QSTR() would make the calling
clearer.
Other callers do pass separate name and len which are stored in a
struct. Sometimes these are already stored in a qstr, other times it
easily could be.
So this patch changes these three functions to receive a 'struct qstr *',
and improves the documentation.
QSTR_LEN() is added to make it easy to pass a QSTR containing a known
len.
[brauner@kernel.org: take a struct qstr pointer]
Signed-off-by: NeilBrown <neil@brown.name>
Link: https://lore.kernel.org/r/20250319031545.2999807-2-neil@brown.name
Signed-off-by: Christian Brauner <brauner@kernel.org>
Pull sysctl updates from Joel Granados:
- Move vm_table members out of kernel/sysctl.c
All vm_table array members have moved to their respective subsystems
leading to the removal of vm_table from kernel/sysctl.c. This
increases modularity by placing the ctl_tables closer to where they
are actually used and at the same time reducing the chances of merge
conflicts in kernel/sysctl.c.
- ctl_table range fixes
Replace the proc_handler function that checks variable ranges in
coredump_sysctls and vdso_table with the one that actually uses the
extra{1,2} pointers as min/max values. This tightens the range of the
values that users can pass into the kernel effectively preventing
{under,over}flows.
- Misc fixes
Correct grammar errors and typos in test messages. Update sysctl
files in MAINTAINERS. Constified and removed array size in
declaration for alignment_tbl
* tag 'sysctl-6.15-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/sysctl/sysctl: (22 commits)
selftests/sysctl: fix wording of help messages
selftests: fix spelling/grammar errors in sysctl/sysctl.sh
MAINTAINERS: Update sysctl file list in MAINTAINERS
sysctl: Fix underflow value setting risk in vm_table
coredump: Fixes core_pipe_limit sysctl proc_handler
sysctl: remove unneeded include
sysctl: remove the vm_table
sh: vdso: move the sysctl to arch/sh/kernel/vsyscall/vsyscall.c
x86: vdso: move the sysctl to arch/x86/entry/vdso/vdso32-setup.c
fs: dcache: move the sysctl to fs/dcache.c
sunrpc: simplify rpcauth_cache_shrink_count()
fs: drop_caches: move sysctl to fs/drop_caches.c
fs: fs-writeback: move sysctl to fs/fs-writeback.c
mm: nommu: move sysctl to mm/nommu.c
security: min_addr: move sysctl to security/min_addr.c
mm: mmap: move sysctl to mm/mmap.c
mm: util: move sysctls to mm/util.c
mm: vmscan: move vmscan sysctls to mm/vmscan.c
mm: swap: move sysctl to mm/swap.c
mm: filemap: move sysctl to mm/filemap.c
...
d_exact_alias() is a descendent of d_add_unique() which was introduced
20 years ago mostly likely to work around problems with NFS servers of
the time. It is now not used in several situations were it was
originally needed and there have been no reports of problems -
presumably the old NFS servers have been improved. This only place it
is now use is in NFSv4 code and the old problematic servers are thought
to have been v2/v3 only.
There is no clear benefit in reusing a unhashed() dentry which happens
to have the same name as the dentry we are adding.
So this patch removes d_exact_alias() and the one place that it is used.
Cc: Trond Myklebust <trond.myklebust@hammerspace.com>
Signed-off-by: NeilBrown <neilb@suse.de>
Link: https://lore.kernel.org/r/20250226062135.2043651-2-neilb@suse.de
Signed-off-by: Christian Brauner <brauner@kernel.org>
Al Viro
NeilBrownandAl Viro
Linus Torvalds
Paulo AlcantaraandSteve French
Luis HenriquesandMiklos Szeredi
Omar SandovalandChristian Brauner
NeilBrownandChristian Brauner