The problem that this patch aims to fix is vfsmount refcounting scalability.
We need to take a reference on the vfsmount for every successful path lookup,
which often go to the same mount point.
The fundamental difficulty is that a "simple" reference count can never be made
scalable, because any time a reference is dropped, we must check whether that
was the last reference. To do that requires communication with all other CPUs
that may have taken a reference count.
We can make refcounts more scalable in a couple of ways, involving keeping
distributed counters, and checking for the global-zero condition less
frequently.
- check the global sum once every interval (this will delay zero detection
for some interval, so it's probably a showstopper for vfsmounts).
- keep a local count and only taking the global sum when local reaches 0 (this
is difficult for vfsmounts, because we can't hold preempt off for the life of
a reference, so a counter would need to be per-thread or tied strongly to a
particular CPU which requires more locking).
- keep a local difference of increments and decrements, which allows us to sum
the total difference and hence find the refcount when summing all CPUs. Then,
keep a single integer "long" refcount for slow and long lasting references,
and only take the global sum of local counters when the long refcount is 0.
This last scheme is what I implemented here. Attached mounts and process root
and working directory references are "long" references, and everything else is
a short reference.
This allows scalable vfsmount references during path walking over mounted
subtrees and unattached (lazy umounted) mounts with processes still running
in them.
This results in one fewer atomic op in the fastpath: mntget is now just a
per-CPU inc, rather than an atomic inc; and mntput just requires a spinlock
and non-atomic decrement in the common case. However code is otherwise bigger
and heavier, so single threaded performance is basically a wash.
Signed-off-by: Nick Piggin <npiggin@kernel.dk>
Suggested by Andreas, mnt_ prefix is clearer namespace, follows kernel
conventions better, and is easier for tab complete. I introduced these
names so I'll admit they were not good choices.
Signed-off-by: Nick Piggin <npiggin@kernel.dk>
Rather than keep a d_mounted count in the dentry, set a dentry flag instead.
The flag can be cleared by checking the hash table to see if there are any
mounts left, which is not time critical because it is performed at detach time.
The mounted state of a dentry is only used to speculatively take a look in the
mount hash table if it is set -- before following the mount, vfsmount lock is
taken and mount re-checked without races.
This saves 4 bytes on 32-bit, nothing on 64-bit but it does provide a hole I
might use later (and some configs have larger than 32-bit spinlocks which might
make use of the hole).
Autofs4 conversion and changelog by Ian Kent <raven@themaw.net>:
In autofs4, when expring direct (or offset) mounts we need to ensure that we
block user path walks into the autofs mount, which is covered by another mount.
To do this we clear the mounted status so that follows stop before walking into
the mount and are essentially blocked until the expire is completed. The
automount daemon still finds the correct dentry for the umount due to the
follow mount logic in fs/autofs4/root.c:autofs4_follow_link(), which is set as
an inode operation for direct and offset mounts only and is called following
the lookup that stopped at the covered mount.
At the end of the expire the covering mount probably has gone away so the
mounted status need not be restored. But we need to check this and only restore
the mounted status if the expire failed.
XXX: autofs may not work right if we have other mounts go over the top of it?
Signed-off-by: Nick Piggin <npiggin@kernel.dk>
The big kernel lock has been removed from all these files at some point,
leaving only the #include.
Remove this too as a cleanup.
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
If clone_mnt() happens while mnt_make_readonly() is running, the
cloned mount might have MNT_WRITE_HOLD flag set, which results in
mnt_want_write() spinning forever on this mount.
Needs CAP_SYS_ADMIN to trigger deliberately and unlikely to happen
accidentally. But if it does happen it can hang the machine.
Signed-off-by: Miklos Szeredi <mszeredi@suse.cz>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
After pushing down the BKL to the get_sb/fill_super operations of the
filesystems that still make usage of the BKL it is safe to remove it from
do_new_mount().
I've read through all the code formerly covered by the BKL inside
do_kern_mount() and have satisfied myself that it doesn't need the BKL
any more.
Signed-off-by: Jan Blunck <jblunck@infradead.org>
Cc: Matthew Wilcox <matthew@wil.cx>
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Sanity check the flags passed to change_mnt_propagation(). Exactly
one flag should be set. Return EINVAL otherwise.
Userspace can pass in arbitrary combinations of MS_* flags to mount().
do_change_type() is called if any of MS_SHARED, MS_PRIVATE, MS_SLAVE,
or MS_UNBINDABLE is set. do_change_type() clears MS_REC and then
calls change_mnt_propagation() with the rest of the user-supplied
flags. change_mnt_propagation() clearly assumes only one flag is set
but do_change_type() does not check that this is true. For example,
mount() with flags MS_SHARED | MS_RDONLY does not actually make the
mount shared or read-only but does clear MNT_UNBINDABLE.
Signed-off-by: Valerie Aurora <vaurora@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
fs: brlock vfsmount_lock
Use a brlock for the vfsmount lock. It must be taken for write whenever
modifying the mount hash or associated fields, and may be taken for read when
performing mount hash lookups.
A new lock is added for the mnt-id allocator, so it doesn't need to take
the heavy vfsmount write-lock.
The number of atomics should remain the same for fastpath rlock cases, though
code would be slightly slower due to per-cpu access. Scalability is not not be
much improved in common cases yet, due to other locks (ie. dcache_lock) getting
in the way. However path lookups crossing mountpoints should be one case where
scalability is improved (currently requiring the global lock).
The slowpath is slower due to use of brlock. On a 64 core, 64 socket, 32 node
Altix system (high latency to remote nodes), a simple umount microbenchmark
(mount --bind mnt mnt2 ; umount mnt2 loop 1000 times), before this patch it
took 6.8s, afterwards took 7.1s, about 5% slower.
Cc: Al Viro <viro@ZenIV.linux.org.uk>
Signed-off-by: Nick Piggin <npiggin@kernel.dk>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Commit d0adde574b added MNT_STRICTATIME
but it isn't actually used (MS_STRICTATIME clears MNT_RELATIME and
MNT_NOATIME rather than setting any mount flag).
Signed-off-by: Miklos Szeredi <mszeredi@suse.cz>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Add three helpers that retrieve a refcounted copy of the root and cwd
from the supplied fs_struct.
get_fs_root()
get_fs_pwd()
get_fs_root_and_pwd()
Signed-off-by: Miklos Szeredi <mszeredi@suse.cz>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
* 'for-linus' of git://git.infradead.org/users/eparis/notify: (132 commits)
fanotify: use both marks when possible
fsnotify: pass both the vfsmount mark and inode mark
fsnotify: walk the inode and vfsmount lists simultaneously
fsnotify: rework ignored mark flushing
fsnotify: remove global fsnotify groups lists
fsnotify: remove group->mask
fsnotify: remove the global masks
fsnotify: cleanup should_send_event
fanotify: use the mark in handler functions
audit: use the mark in handler functions
dnotify: use the mark in handler functions
inotify: use the mark in handler functions
fsnotify: send fsnotify_mark to groups in event handling functions
fsnotify: Exchange list heads instead of moving elements
fsnotify: srcu to protect read side of inode and vfsmount locks
fsnotify: use an explicit flag to indicate fsnotify_destroy_mark has been called
fsnotify: use _rcu functions for mark list traversal
fsnotify: place marks on object in order of group memory address
vfs/fsnotify: fsnotify_close can delay the final work in fput
fsnotify: store struct file not struct path
...
Fix up trivial delete/modify conflict in fs/notify/inotify/inotify.c.
If sget() finds a matching superblock being set up, it'll
grab an active reference to it and grab s_umount. That's
fine - we'll wait for completion of foofs_get_sb() that way.
However, if said foofs_get_sb() fails we'll end up holding
the halfway-created superblock. deactivate_locked_super()
called by foofs_get_sb() will just unlock the sucker since
we are holding another active reference to it.
What we need is a way to tell if superblock has been successfully
set up. Unfortunately, neither ->s_root nor the check for
MS_ACTIVE quite fit. Cheap and easy way, suitable for backport:
new flag set by the (only) caller of ->get_sb(). If that flag
isn't present by the time sget() grabbed s_umount on preexisting
superblock it has found, it's seeing a stillborn and should
just bury it with deactivate_locked_super() (and repeat the search).
Longer term we want to set that flag in ->get_sb() instances (and
check for it to distinguish between "sget() found us a live sb"
and "sget() has allocated an sb, we need to set it up" in there,
instead of checking ->s_root as we do now).
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Cc: stable@kernel.org
Per-mount watches allow groups to listen to fsnotify events on an entire
mount. This patch simply adds and initializes the fields needed in the
vfsmount struct to make this happen.
Signed-off-by: Andreas Gruenbacher <agruen@suse.de>
Signed-off-by: Eric Paris <eparis@redhat.com>
This patch adds the list and mask fields needed to support vfsmount marks.
These are the same fields fsnotify needs on an inode. They are not used,
just declared and we note where the cleanup hook should be (the function is
not yet defined)
Signed-off-by: Andreas Gruenbacher <agruen@suse.de>
Signed-off-by: Eric Paris <eparis@redhat.com>
1) i_flags simply doesn't work for mount/unlink race prevention;
we may have many links to file and rm on one of those obviously
shouldn't prevent bind on top of another later on. To fix it
right way we need to mark _dentry_ as unsuitable for mounting
upon; new flag (DCACHE_CANT_MOUNT) is protected by d_flags and
i_mutex on the inode in question. Set it (with dont_mount(dentry))
in unlink/rmdir/etc., check (with cant_mount(dentry)) in places
in namespace.c that used to check for S_DEAD. Setting S_DEAD
is still needed in places where we used to set it (for directories
getting killed), since we rely on it for readdir/rmdir race
prevention.
2) rename()/mount() protection has another bogosity - we unhash
the target before we'd checked that it's not a mountpoint. Fixed.
3) ancient bogosity in pivot_root() - we locked i_mutex on the
right directory, but checked S_DEAD on the different (and wrong)
one. Noticed and fixed.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Add a new UMOUNT_NOFOLLOW flag to umount(2). This is needed to prevent
symlink attacks in unprivileged unmounts (fuse, samba, ncpfs).
Additionally, return -EINVAL if an unknown flag is used (and specify
an explicitly unused flag: UMOUNT_UNUSED). This makes it possible for
the caller to determine if a flag is supported or not.
CC: Eugene Teo <eugene@redhat.com>
CC: Michael Kerrisk <mtk.manpages@gmail.com>
Signed-off-by: Miklos Szeredi <mszeredi@suse.cz>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
It hadn't been needed since we'd sanitized the logics in
mark_mounts_for_expiry() (which, in turn, used to be a
rudiment of bad old times when namespace_sem was per-ns).
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
