The standard memcmp function on a Westmere system shows up hot in
profiles in the `git diff` workload (both parallel and single threaded),
and it is likely due to the costs associated with trapping into
microcode, and little opportunity to improve memory access (dentry
name is not likely to take up more than a cacheline).
So replace it with an open-coded byte comparison. This increases code
size by 8 bytes in the critical __d_lookup_rcu function, but the
speedup is huge, averaging 10 runs of each:
git diff st user sys elapsed CPU
before 1.15 2.57 3.82 97.1
after 1.14 2.35 3.61 96.8
git diff mt user sys elapsed CPU
before 1.27 3.85 1.46 349
after 1.26 3.54 1.43 333
Elapsed time for single threaded git diff at 95.0% confidence:
-0.21 +/- 0.01
-5.45% +/- 0.24%
It's -0.66% +/- 0.06% elapsed time on my Opteron, so rep cmp costs on the
fam10h seem to be relatively smaller, but there is still a win.
Signed-off-by: Nick Piggin <npiggin@kernel.dk>
This makes single threaded git diff -1.25% +/- 0.05% elapsed time on my
2s12c24t Westmere system, and -0.86% +/- 0.05% on my 2s8c Barcelona, by
prefetching the important first cacheline of the inode in while we do the
actual name compare and other operations on the dentry.
There was no measurable slowdown in the single file stat case, or the creat
case (where negative dentries would be common).
Signed-off-by: Nick Piggin <npiggin@kernel.dk>
Regardless of how much we possibly try to scale dcache, there is likely
always going to be some fundamental contention when adding or removing children
under the same parent. Pseudo filesystems do not seem need to have connected
dentries because by definition they are disconnected.
Signed-off-by: Nick Piggin <npiggin@kernel.dk>
dcache_inode_lock can be replaced with per-inode locking. Use existing
inode->i_lock for this. This is slightly non-trivial because we sometimes
need to find the inode from the dentry, which requires d_inode to be
stabilised (either with refcount or d_lock).
Signed-off-by: Nick Piggin <npiggin@kernel.dk>
Put dentry and inode fields into top of data structure. This allows RCU path
traversal to perform an RCU dentry lookup in a path walk by touching only the
first 56 bytes of the dentry.
We also fit in 8 bytes of inline name in the first 64 bytes, so for short
names, only 64 bytes needs to be touched to perform the lookup. We should
get rid of the hash->prev pointer from the first 64 bytes, and fit 16 bytes
of name in there, which will take care of 81% rather than 32% of the kernel
tree.
inode is also rearranged so that RCU lookup will only touch a single cacheline
in the inode, plus one in the i_ops structure.
This is important for directory component lookups in RCU path walking. In the
kernel source, directory names average is around 6 chars, so this works.
When we reach the last element of the lookup, we need to lock it and take its
refcount which requires another cacheline access.
Align dentry and inode operations structs, so members will be at predictable
offsets and we can group common operations into head of structure.
Signed-off-by: Nick Piggin <npiggin@kernel.dk>
Reduce some branches and memory accesses in dcache lookup by adding dentry
flags to indicate common d_ops are set, rather than having to check them.
This saves a pointer memory access (dentry->d_op) in common path lookup
situations, and saves another pointer load and branch in cases where we
have d_op but not the particular operation.
Patched with:
git grep -E '[.>]([[:space:]])*d_op([[:space:]])*=' | xargs sed -e 's/\([^\t ]*\)->d_op = \(.*\);/d_set_d_op(\1, \2);/' -e 's/\([^\t ]*\)\.d_op = \(.*\);/d_set_d_op(\&\1, \2);/' -i
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>
Perform common cases of path lookups without any stores or locking in the
ancestor dentry elements. This is called rcu-walk, as opposed to the current
algorithm which is a refcount based walk, or ref-walk.
This results in far fewer atomic operations on every path element,
significantly improving path lookup performance. It also avoids cacheline
bouncing on common dentries, significantly improving scalability.
The overall design is like this:
* LOOKUP_RCU is set in nd->flags, which distinguishes rcu-walk from ref-walk.
* Take the RCU lock for the entire path walk, starting with the acquiring
of the starting path (eg. root/cwd/fd-path). So now dentry refcounts are
not required for dentry persistence.
* synchronize_rcu is called when unregistering a filesystem, so we can
access d_ops and i_ops during rcu-walk.
* Similarly take the vfsmount lock for the entire path walk. So now mnt
refcounts are not required for persistence. Also we are free to perform mount
lookups, and to assume dentry mount points and mount roots are stable up and
down the path.
* Have a per-dentry seqlock to protect the dentry name, parent, and inode,
so we can load this tuple atomically, and also check whether any of its
members have changed.
* Dentry lookups (based on parent, candidate string tuple) recheck the parent
sequence after the child is found in case anything changed in the parent
during the path walk.
* inode is also RCU protected so we can load d_inode and use the inode for
limited things.
* i_mode, i_uid, i_gid can be tested for exec permissions during path walk.
* i_op can be loaded.
When we reach the destination dentry, we lock it, recheck lookup sequence,
and increment its refcount and mountpoint refcount. RCU and vfsmount locks
are dropped. This is termed "dropping rcu-walk". If the dentry refcount does
not match, we can not drop rcu-walk gracefully at the current point in the
lokup, so instead return -ECHILD (for want of a better errno). This signals the
path walking code to re-do the entire lookup with a ref-walk.
Aside from the final dentry, there are other situations that may be encounted
where we cannot continue rcu-walk. In that case, we drop rcu-walk (ie. take
a reference on the last good dentry) and continue with a ref-walk. Again, if
we can drop rcu-walk gracefully, we return -ECHILD and do the whole lookup
using ref-walk. But it is very important that we can continue with ref-walk
for most cases, particularly to avoid the overhead of double lookups, and to
gain the scalability advantages on common path elements (like cwd and root).
The cases where rcu-walk cannot continue are:
* NULL dentry (ie. any uncached path element)
* parent with d_inode->i_op->permission or ACLs
* dentries with d_revalidate
* Following links
In future patches, permission checks and d_revalidate become rcu-walk aware. It
may be possible eventually to make following links rcu-walk aware.
Uncached path elements will always require dropping to ref-walk mode, at the
very least because i_mutex needs to be grabbed, and objects allocated.
Signed-off-by: Nick Piggin <npiggin@kernel.dk>
The tricky locking for disposing of a dentry is duplicated 3 times in the
dcache (dput, pruning a dentry from the LRU, and pruning its ancestors).
Consolidate them all into a single function dentry_kill.
Signed-off-by: Nick Piggin <npiggin@kernel.dk>
prune_one_dentry can avoid quite a bit of locking in the common case where
ancestors have an elevated refcount. Alternatively, we could have gone the
other way and made fewer trylocks in the case where d_count goes to zero, but
is probably less common.
Signed-off-by: Nick Piggin <npiggin@kernel.dk>
dget_locked was a shortcut to avoid the lazy lru manipulation when we already
held dcache_lock (lru manipulation was relatively cheap at that point).
However, how that the lru lock is an innermost one, we never hold it at any
caller, so the lock cost can now be avoided. We already have well working lazy
dcache LRU, so it should be fine to defer LRU manipulations to scan time.
Signed-off-by: Nick Piggin <npiggin@kernel.dk>
dcache_inode_lock can be avoided in d_delete() and d_materialise_unique()
in cases where it is not required.
Signed-off-by: Nick Piggin <npiggin@kernel.dk>
It is possible to run dput without taking data structure locks up-front. In
many cases where we don't kill the dentry anyway, these locks are not required.
Signed-off-by: Nick Piggin <npiggin@kernel.dk>
Long lived dcache "multi-step" operations which retry on rename seq can
be starved with a lot of rename activity. If they fail after the 1st pass,
take the rename_lock for writing to avoid further starvation.
Signed-off-by: Nick Piggin <npiggin@kernel.dk>
The remaining usages for dcache_lock is to allow atomic, multi-step read-side
operations over the directory tree by excluding modifications to the tree.
Also, to walk in the leaf->root direction in the tree where we don't have
a natural d_lock ordering.
This could be accomplished by taking every d_lock, but this would mean a
huge number of locks and actually gets very tricky.
Solve this instead by using the rename seqlock for multi-step read-side
operations, retry in case of a rename so we don't walk up the wrong parent.
Concurrent dentry insertions are not serialised against. Concurrent deletes
are tricky when walking up the directory: our parent might have been deleted
when dropping locks so also need to check and retry for that.
We can also use the rename lock in cases where livelock is a worry (and it
is introduced in subsequent patch).
Signed-off-by: Nick Piggin <npiggin@kernel.dk>
Add a new lock, dcache_inode_lock, to protect the inode's i_dentry list
from concurrent modification. d_alias is also protected by d_lock.
Signed-off-by: Nick Piggin <npiggin@kernel.dk>
Protect d_subdirs and d_child with d_lock, except in filesystems that aren't
using dcache_lock for these anyway (eg. using i_mutex).
Note: if we change the locking rule in future so that ->d_child protection is
provided only with ->d_parent->d_lock, it may allow us to reduce some locking.
But it would be an exception to an otherwise regular locking scheme, so we'd
have to see some good results. Probably not worthwhile.
Signed-off-by: Nick Piggin <npiggin@kernel.dk>
Protect d_unhashed(dentry) condition with d_lock. This means keeping
DCACHE_UNHASHED bit in synch with hash manipulations.
Signed-off-by: Nick Piggin <npiggin@kernel.dk>
Make d_count non-atomic and protect it with d_lock. This allows us to ensure a
0 refcount dentry remains 0 without dcache_lock. It is also fairly natural when
we start protecting many other dentry members with d_lock.
Signed-off-by: Nick Piggin <npiggin@kernel.dk>
Add a new lock, dcache_lru_lock, to protect the dcache LRU list from concurrent
modification. d_lru is also protected by d_lock, which allows LRU lists to be
accessed without the lru lock, using RCU in future patches.
Signed-off-by: Nick Piggin <npiggin@kernel.dk>
Add a new lock, dcache_hash_lock, to protect the dcache hash table from
concurrent modification. d_hash is also protected by d_lock.
Signed-off-by: Nick Piggin <npiggin@kernel.dk>
Remove dcache_lock locking from hostfs filesystem, and move it into dcache
helpers. All that is required is a coherent path name. Protection from
concurrent modification of the namespace after path name generation is not
provided in current code, because dcache_lock is dropped before the path is
used.
Signed-off-by: Nick Piggin <npiggin@kernel.dk>
Change d_hash so it may be called from lock-free RCU lookups. See similar
patch for d_compare for details.
For in-tree filesystems, this is just a mechanical change.
Signed-off-by: Nick Piggin <npiggin@kernel.dk>
Change d_compare so it may be called from lock-free RCU lookups. This
does put significant restrictions on what may be done from the callback,
however there don't seem to have been any problems with in-tree fses.
If some strange use case pops up that _really_ cannot cope with the
rcu-walk rules, we can just add new rcu-unaware callbacks, which would
cause name lookup to drop out of rcu-walk mode.
For in-tree filesystems, this is just a mechanical change.
Signed-off-by: Nick Piggin <npiggin@kernel.dk>
smpfs and ncpfs want to update a live dentry name in-place. Rather than
have them open code the locking, provide a documented dcache API.
Signed-off-by: Nick Piggin <npiggin@kernel.dk>
Change d_delete from a dentry deletion notification to a dentry caching
advise, more like ->drop_inode. Require it to be constant and idempotent,
and not take d_lock. This is how all existing filesystems use the callback
anyway.
This makes fine grained dentry locking of dput and dentry lru scanning
much simpler.
Signed-off-by: Nick Piggin <npiggin@kernel.dk>
The nr_unused counters count the number of objects on an LRU, and as such they
are synchronized with LRU object insertion and removal and scanning, and
protected under the LRU lock.
Making it per-cpu does not actually get any concurrency improvements because of
this lock, and summing the counter is much slower, and
incrementing/decrementing it costs more code size and is slower too.
These counters should stay per-LRU, which currently means global.
Signed-off-by: Nick Piggin <npiggin@kernel.dk>
d_validate has been broken for a long time.
kmem_ptr_validate does not guarantee that a pointer can be dereferenced
if it can go away at any time. Even rcu_read_lock doesn't help, because
the pointer might be queued in RCU callbacks but not executed yet.
So the parent cannot be checked, nor the name hashed. The dentry pointer
can not be touched until it can be verified under lock. Hashing simply
cannot be used.
Instead, verify the parent/child relationship by traversing parent's
d_child list. It's slow, but only ncpfs and the destaged smbfs care
about it, at this point.
Signed-off-by: Nick Piggin <npiggin@kernel.dk>
This reverts commit 3825bdb7ed.
You cannot dget() a dentry without having a reference, or holding
a lock that guarantees it remains valid.
Signed-off-by: Nick Piggin <npiggin@kernel.dk>
d_validate does a purely read lookup in the dentry hash, so use RCU read side
locking instead of dcache_lock. Split out from a larget patch by
Nick Piggin <npiggin@suse.de>.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Always do a list_del_init on the LRU to make sure the list_empty invariant for
not beeing on the LRU always holds true, and fold dentry_lru_del_init into
dentry_lru_del. Replace the dentry_lru_add_tail primitive with a
dentry_lru_move_tail operations that simpler when the dentry already is one
the list, which is always is. Move the list_empty into dentry_lru_add to
fit the scheme of the other lru helpers, and simplify locking once we
move to a separate LRU lock.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Currently __shrink_dcache_sb has an extremly awkward calling convention
because it tries to please very different callers. Split out the
main loop into a shrink_dentry_list helper, which gets called directly
from shrink_dcache_sb for the cases where all dentries need to be pruned,
or from __shrink_dcache_sb for pruning only a certain number of dentries.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
dentry referenced bit is only set when installing the dentry back
onto the LRU. However with lazy LRU, the dentry can already be on
the LRU list at dput time, thus missing out on setting the referenced
bit. Fix this.
Signed-off-by: Nick Piggin <npiggin@suse.de>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
The nr_dentry stat is a globally touched cacheline and atomic operation
twice over the lifetime of a dentry. It is used for the benfit of userspace
only. Turn it into a per-cpu counter and always decrement it in d_free instead
of doing various batching operations to reduce lock hold times in the callers.
Based on an earlier patch from Nick Piggin <npiggin@suse.de>.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Remove d_callback and always call __d_free with a RCU head.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
All callers take dcache_lock just around the call to __d_path, so
take the lock into it in preparation of getting rid of dcache_lock.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
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>
fs: remove extra lookup in __lookup_hash
Optimize lookup for create operations, where no dentry should often be
common-case. In cases where it is not, such as unlink, the added overhead
is much smaller than the removed.
Also, move comments about __d_lookup racyness to the __d_lookup call site.
d_lookup is intuitive; __d_lookup is what needs commenting. So in that same
vein, add kerneldoc comments to __d_lookup and clean up some of the comments:
- We are interested in how the RCU lookup works here, particularly with
renames. Make that explicit, and point to the document where it is explained
in more detail.
- RCU is pretty standard now, and macros make implementations pretty mindless.
If we want to know about RCU barrier details, we look in RCU code.
- Delete some boring legacy comments because we don't care much about how the
code used to work, more about the interesting parts of how it works now. So
comments about lazy LRU may be interesting, but would better be done in the
LRU or refcount management code.
Signed-off-by: Nick Piggin <npiggin@kernel.dk>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Fix parameter name in kernel-doc notation (causes a warning).
Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Prepend "(unreachable)" to path strings if the path is not reachable
from the current root.
Two places updated are
- the return string from getcwd()
- and symlinks under /proc/$PID.
Other uses of d_path() are left unchanged (we know that some old
software crashes if /proc/mounts is changed).
Signed-off-by: Miklos Szeredi <mszeredi@suse.cz>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
__d_path() has 4 callers:
d_path()
sys_getcwd()
seq_path_root()
tomoyo_realpath_from_path2()
Of these the only one which needs the " (deleted)" ending is d_path().
sys_getcwd() checks for existence before calling __d_path().
seq_path_root() is used to show the mountpoint path in
/proc/PID/mountinfo, which is always a positive.
And tomoyo doesn't want the deleted ending.
Create a helper "path_with_deleted()" as subsequent patches will need
this in multiple places.
Signed-off-by: Miklos Szeredi <mszeredi@suse.cz>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Split off prepend_path() from __d_path(). This new helper takes an
end-of-buffer pointer and buffer-length pointer just like the other
prepend_* functions. Move the " (deleted)" postfix out to __d_path().
This patch doesn't change any functionality but paves the way for the
following patches.
Signed-off-by: Miklos Szeredi <mszeredi@suse.cz>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
In the old times pseudo-filesystems set the name of theroot dentry to
some prefix like "pipe:" and the name of the child dentry to "[123]"
and relied on a hack in __d_path() to replace the preceding slash with
the root's name to get "pipe:[123]".
Then the d_dname() dentry operation was introduced which solved the
same problem without having to pre-fill the name in each dentry.
Currently the following pseudo filesystems exist in the kernel:
perfmon
mtd
anon_inode
bdev
pipe
socket
Of these only perfmon, anon_inode, pipe and socket create
sub-dentries, all of which have now been switched to using d_dname().
bdev and mtd only create inodes.
This means that now the hack to overwrite the slash can be removed, so
for unreachable paths (e.g. within a detached mount) the path string
won't be polluted with garbage. For these cases a subsequent patch
will add a prefix, indicating that the path is unreachable.
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>
builds path relative to fs root, called under dcache_lock,
doesn't append any nonsense to unlinked ones.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
The current shrinker implementation requires the registered callback
to have global state to work from. This makes it difficult to shrink
caches that are not global (e.g. per-filesystem caches). Pass the shrinker
structure to the callback so that users can embed the shrinker structure
in the context the shrinker needs to operate on and get back to it in the
callback via container_of().
Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
list_for_each_entry_safe is not suitable to protect against concurrent
modification of the list. 6754af6 introduced a race in sb walking.
list_for_each_entry can use the trick of pinning the current entry in
the list before we drop and retake the lock because it subsequently
follows cur->next. However list_for_each_entry_safe saves n=cur->next
for following before entering the loop body, so when the lock is
dropped, n may be deleted.
Signed-off-by: Nick Piggin <npiggin@suse.de>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: John Stultz <johnstul@us.ibm.com>
Cc: Frank Mayhar <fmayhar@google.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
We used to remove from s_list and s_instances at the same
time. So let's *not* do the former and skip superblocks
that have empty s_instances in the loops over s_list.
The next step, of course, will be to get rid of rescan logics
in those loops.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
We set the "it's dead, don't mount on it" flag _and_ do not remove it if
we turn the damn thing negative and leave it around. And if it goes
positive afterwards, well...
Fortunately, there's only one place where that needs to be caught:
only d_delete() can turn the sucker negative without immediately freeing
it; all other places that can lead to ->d_iput() call are followed by
unconditionally freeing struct dentry in question. So the fix is obvious:
Addresses https://bugzilla.kernel.org/show_bug.cgi?id=16014
Reported-by: Adam Tkac <vonsch@gmail.com>
Tested-by: Adam Tkac <vonsch@gmail.com>
Cc: <stable@kernel.org> [2.6.34.x]
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
rehashing the negative placeholder opens a race with d_lookup();
we unhash it almost immediately (by d_move()), but the race
window is there. Since d_move() doesn't rely on target being
hashed, we don't need that d_rehash() at all.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Cleanup EXPORT* macros according to Documantation/CodingStyle.
Move EXPORT* macros to the line immediately after the closing
function brace.
Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
The EXPORT_SYMBOL for d_alloc_name is in fs/libfs.c but the function
is in fs/dcache.c. Move the EXPORT_SYMBOL to the line immediately
after the closing function brace line in fs/dcache.c as mentioned
in Documentation/CodingStyle.
Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
might_sleep() is called late-ish in cond_resched(), after the
need_resched()/preempt enabled/system running tests are
checked.
It's better to check the sleeps while atomic earlier and not
depend on some environment datas that reduce the chances to
detect a problem.
Also define cond_resched_*() helpers as macros, so that the
FILE/LINE reported in the sleeping while atomic warning
displays the real origin and not sched.h
Changes in v2:
- Call __might_sleep() directly instead of might_sleep() which
may call cond_resched()
- Turn cond_resched() into a macro so that the file:line
couple reported refers to the caller of cond_resched() and
not __cond_resched() itself.
Changes in v3:
- Also propagate this __might_sleep() pull up to
cond_resched_lock() and cond_resched_softirq()
Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <1247725694-6082-6-git-send-email-fweisbec@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
d_unlinked() will be used in middle-term to ban checkpointing when opened
but unlinked file is detected, and in long term, to detect such situation
and special case on it.
Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Fix ordering of LRU when moving referenced dentries to the head of the list
(they should go to the head of the list in the same order as they were found
from the tail, rather than reverse order).
Signed-off-by: Nick Piggin <npiggin@suse.de>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Don't pull it in sched.h; very few files actually need it and those
can include directly. sched.h itself only needs forward declaration
of struct fs_struct;
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Make sure that comments describe what's going on and not how, and always
use __d_instantiate instead of two separate branches, one with
d_instantiate and one with __d_instantiate.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Commit 4ea3ada295 declares d_obtain_alias()
as EXPORT_SYMBOL_GPL where it's supposed to replace d_alloc_anon which was
previously declared as EXPORT_SYMBOL and thus available to any loadable
module.
This patch reverts that.
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
Acked-by: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: "J. Bruce Fields" <bfields@fieldses.org>
Cc: Trond Myklebust <trond.myklebust@fys.uio.no>
Acked-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Use the new generic implementation.
Signed-off-by: Wu Fengguang <fengguang.wu@intel.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Instead of creating the "filp" kmem_cache in vfs_caches_init(),
we can do it a litle be later in files_init(), so that filp_cachep
is static to fs/file_table.c
Acked-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Signed-off-by: Eric Dumazet <dada1@cosmosbay.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Explain that you really need to use the return value of d_path rather than
the buffer you passed into it.
Also fix the comment for seq_path(), the function arguments changed
recently but the comment hadn't been updated in sync.
Signed-off-by: Arjan van de Ven <arjan@linux.intel.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Christoph Hellwig <hch@lst.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
no function named d_put(), it should be dput().
Impact: fix document and comment, no functionality changed
Signed-off-by: Zhao Lei <zhaolei@cn.fuijtsu.com>
Signed-off-by: Randy Dunlap <rdunlap@xenotime.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
struct dentry is one of the most critical structures in the kernel. So it's
sad to see it going neglected.
With CONFIG_PROFILING turned on (which is probably the common case at least
for distros and kernel developers), sizeof(struct dcache) == 208 here
(64-bit). This gives 19 objects per slab.
I packed d_mounted into a hole, and took another 4 bytes off the inline
name length to take the padding out from the end of the structure. This
shinks it to 200 bytes. I could have gone the other way and increased the
length to 40, but I'm aiming for a magic number, read on...
I then got rid of the d_cookie pointer. This shrinks it to 192 bytes. Rant:
why was this ever a good idea? The cookie system should increase its hash
size or use a tree or something if lookups are a problem. Also the "fast
dcookie lookups" in oprofile should be moved into the dcookie code -- how
can oprofile possibly care about the dcookie_mutex? It gets dropped after
get_dcookie() returns so it can't be providing any sort of protection.
At 192 bytes, 21 objects fit into a 4K page, saving about 3MB on my system
with ~140 000 entries allocated. 192 is also a multiple of 64, so we get
nice cacheline alignment on 64 and 32 byte line systems -- any given dentry
will now require 3 cachelines to touch all fields wheras previously it
would require 4.
I know the inline name size was chosen quite carefully, however with the
reduction in cacheline footprint, it should actually be just about as fast
to do a name lookup for a 36 character name as it was before the patch (and
faster for other sizes). The memory footprint savings for names which are
<= 32 or > 36 bytes long should more than make up for the memory cost for
33-36 byte names.
Performance is a feature...
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Hi Al,
remember that debug session we did at KS? You suggested this patch back
then....
From 7751eaf30474b8cbfaea64795805a17eab05ac53 Mon Sep 17 00:00:00 2001
From: Arjan van de Ven <arjan@linux.intel.com>
Date: Tue, 16 Sep 2008 16:51:17 -0700
Subject: [PATCH] fs: add a sanity check in d_free
we're seeing some corruption in the dentry->d_alias list that
appears like a free of an entry still on the list; this patch
adds a WARN_ON() to catch this scenario, as suggested by Al Viro
Signed-off-by: Arjan van de Ven <arjan@linux.intel.com>
This adds __d_instantiate() for users which is already taking
dcache_lock, and replace with it.
The part of d_add_ci() isn't equivalent. But it should be needed
fsnotify_d_instantiate() actually, because the path is to add the
inode to negative dentry. fsnotify_d_instantiate() should be called
after change from negative to positive.
Signed-off-by: OGAWA Hirofumi <hirofumi@mail.parknet.co.jp>
This adds d_ancestor() instead of d_isparent(), then use it.
If new_dentry == old_dentry, is_subdir() returns 1, looks strange.
"new_dentry == old_dentry" is not subdir obviously. But I'm not
checking callers for now, so this keeps current behavior.
Signed-off-by: OGAWA Hirofumi <hirofumi@mail.parknet.co.jp>
The calling conventions of d_alloc_anon are rather unfortunate for all
users, and it's name is not very descriptive either.
Add d_obtain_alias as a new exported helper that drops the inode
reference in the failure case, too and allows to pass-through NULL
pointers and inodes to allow for tail-calls in the export operations.
Incidentally this helper already existed as a private function in
libfs.c as exportfs_d_alloc so kill that one and switch the callers
to d_obtain_alias.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
The VFS interface for the 'd_compare()' is a bit special (read: 'odd'),
because it really just essentially replaces a memcmp(). The filesystem
is supposed to just compare the two names with whatever case-independent
or other function.
And when I say 'is supposed to', I obviously mean that 'procfs does odd
things, and actually looks at the dentry that we don't even pass down,
rather than just the name'. Which results in problems, because we
actually call d_compare before we have even verified that the dentry is
still hashed at all.
And that causes a problm since the inode that procfs looks at may have
been free'd and the d_inode pointer is NULL. procfs just assumes that
all dentries are positive, since procfs itself never generates a
negative one. But memory pressure will still result in the dentry
getting torn down, and as it is removed by RCU, it still remains visible
on some lists - and to d_compare.
If the filesystem just did a name comparison, we wouldn't care. And we
could just fix procfs to know about negative dentries too. But rather
than have the low-level filesystems know about internal VFS details,
just move the check for a unhashed dentry up a bit, so that we will only
call d_compare on dentries that are still active.
The actual oops this caused didn't look like a NULL pointer dereference
because procfs did a 'container_of(inode, struct proc_inode, vfs_inode)'
to get at its internal proc_inode information from the inode pointer,
and accessed a field below the inode. So the oops would look something
like
BUG: unable to handle kernel paging request at fffffffffffffff0
IP: [<ffffffff802bc6c6>] proc_sys_compare+0x36/0x50
and was seen on both x86-64 (Alexey Dobriyan and Hugh Dickins) and
ppc64 (Hugh Dickins).
Reported-by: Alexey Dobriyan <adobriyan@gmail.com>
Acked-by: Hugh Dickins <hugh@veritas.com>
Cc: Al Viro <viro@ZenIV.linux.org.uk>
Reviewed-by: "Eric W. Biederman" <ebiederm@xmission.com>
Signed-of-by: Linus Torvalds <torvalds@linux-foundation.org>
As pointed out during review d_add_ci argument order should match d_add,
so switch the dentry and inode arguments.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
This add a dcache entry to the dcache for lookup, but changing the name
that is associated with the entry rather than the one passed in to the
lookup routine.
First, it sees if the case-exact match already exists in the dcache and
uses it if one exists. Otherwise, it allocates a new node with the new
name and splices it into the dcache.
Original code from ntfs_lookup in fs/ntfs/namei.c by Anton Altaparmakov.
Signed-off-by: Barry Naujok <bnaujok@sgi.com>
Signed-off-by: Anton Altaparmakov <aia21@cantab.net>
Acked-by: Christoph Hellwig <hch@infradead.org>
[Summary]
Split LRU-list of unused dentries to one per superblock to avoid soft
lock up during NFS mounts and remounting of any filesystem.
Previously I posted here:
http://lkml.org/lkml/2008/3/5/590
[Descriptions]
- background
dentry_unused is a list of dentries which are not referenced.
dentry_unused grows up when references on directories or files are
released. This list can be very long if there is huge free memory.
- the problem
When shrink_dcache_sb() is called, it scans all dentry_unused linearly
under spin_lock(), and if dentry->d_sb is differnt from given
superblock, scan next dentry. This scan costs very much if there are
many entries, and very ineffective if there are many superblocks.
IOW, When we need to shrink unused dentries on one dentry, but scans
unused dentries on all superblocks in the system. For example, we scan
500 dentries to unmount a filesystem, but scans 1,000,000 or more unused
dentries on other superblocks.
In our case , At mounting NFS*, shrink_dcache_sb() is called to shrink
unused dentries on NFS, but scans 100,000,000 unused dentries on
superblocks in the system such as local ext3 filesystems. I hear NFS
mounting took 1 min on some system in use.
* : NFS uses virtual filesystem in rpc layer, so NFS is affected by
this problem.
100,000,000 is possible number on large systems.
Per-superblock LRU of unused dentried can reduce the cost in
reasonable manner.
- How to fix
I found this problem is solved by David Chinner's "Per-superblock
unused dentry LRU lists V3"(1), so I rebase it and add some fix to
reclaim with fairness, which is in Andrew Morton's comments(2).
1) http://lkml.org/lkml/2006/5/25/318
2) http://lkml.org/lkml/2006/5/25/320
Split LRU-list of unused dentries to each superblocks. Then, NFS
mounting will check dentries under a superblock instead of all. But
this spliting will break LRU of dentry-unused. So, I've attempted to
make reclaim unused dentrins with fairness by calculate number of
dentries to scan on this sb based on following way
number of dentries to scan on this sb =
count * (number of dentries on this sb / number of dentries in the machine)
- ToDo
- I have to measuring performance number and do stress tests.
- When unmount occurs during prune_dcache(), scanning on same
superblock, It is unable to reach next superblock because it is gone
away. We restart scannig superblock from first one, it causes
unfairness of reclaim unused dentries on first superblock. But I think
this happens very rarely.
- Test Results
Result on 6GB boxes with excessive unused dentries.
Without patch:
$ cat /proc/sys/fs/dentry-state
10181835 10180203 45 0 0 0
# mount -t nfs 10.124.60.70:/work/kernel-src nfs
real 0m1.830s
user 0m0.001s
sys 0m1.653s
With this patch:
$ cat /proc/sys/fs/dentry-state
10236610 10234751 45 0 0 0
# mount -t nfs 10.124.60.70:/work/kernel-src nfs
real 0m0.106s
user 0m0.002s
sys 0m0.032s
[akpm@linux-foundation.org: fix comments]
Signed-off-by: Kentaro Makita <k-makita@np.css.fujitsu.com>
Cc: Neil Brown <neilb@suse.de>
Cc: Trond Myklebust <trond.myklebust@fys.uio.no>
Cc: David Chinner <dgc@sgi.com>
Cc: "J. Bruce Fields" <bfields@fieldses.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Fix the following sparse warnings:
fs/dcache.c:2183:19: warning: symbol 'filp_cachep' was not declared. Should it be static?
fs/dcache.c:115:3: warning: context imbalance in 'dentry_iput' - unexpected unlock
fs/dcache.c:188:2: warning: context imbalance in 'dput' - different lock contexts for basic block
fs/dcache.c:400:2: warning: context imbalance in 'prune_one_dentry' - different lock contexts for basic block
fs/dcache.c:431:22: warning: context imbalance in 'prune_dcache' - different lock contexts for basic block
fs/dcache.c:563:2: warning: context imbalance in 'shrink_dcache_sb' - different lock contexts for basic block
fs/dcache.c:1385:6: warning: context imbalance in 'd_delete' - wrong count at exit
fs/dcache.c:1636:2: warning: context imbalance in '__d_unalias' - unexpected unlock
fs/dcache.c:1735:2: warning: context imbalance in 'd_materialise_unique' - different lock contexts for basic block
Signed-off-by: Miklos Szeredi <mszeredi@suse.cz>
Reviewed-by: Matthew Wilcox <willy@linux.intel.com>
Acked-by: Christoph Hellwig <hch@infradead.org>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
The path that __d_path() computes can become slightly inconsistent when it
races with mount operations: it grabs the vfsmount_lock when traversing mount
points but immediately drops it again, only to re-grab it when it reaches the
next mount point. The result is that the filename computed is not always
consisent, and the file may never have had that name. (This is unlikely, but
still possible.)
Fix this by grabbing the vfsmount_lock for the whole duration of
__d_path().
Signed-off-by: Andreas Gruenbacher <agruen@suse.de>
Signed-off-by: John Johansen <jjohansen@suse.de>
Signed-off-by: Miklos Szeredi <mszeredi@suse.cz>
Acked-by: Christoph Hellwig <hch@infradead.org>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Jan Engelhardt <jengelh@medozas.de>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Add a new function:
seq_file_root()
This is similar to seq_path(), but calculates the path relative to the
given root, instead of current->fs->root. If the path was unreachable
from root, then modify the root parameter to reflect this.
Signed-off-by: Miklos Szeredi <mszeredi@suse.cz>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
[mszeredi@suse.cz] split big patch into managable chunks
Add the following functions:
dentry_path()
seq_dentry()
These are similar to d_path() and seq_path(). But instead of
calculating the path within a mount namespace, they calculate the path
from the root of the filesystem to a given dentry, ignoring mounts
completely.
Signed-off-by: Ram Pai <linuxram@us.ibm.com>
Signed-off-by: Miklos Szeredi <mszeredi@suse.cz>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Extract the common code to remove a dentry from the lru into a new function
dentry_lru_remove().
Two call sites used list_del() instead of list_del_init(). AFAIK the
performance of both is the same. dentry_lru_remove() does a list_del_init().
As a result dentry->d_lru is now always empty when a dentry is freed.
Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Al Viro