The inode structure layout is largely random, and some of the vfs paths
really do care.  The path lookup in particular is already quite D$
intensive, and profiles show that accessing the 'inode->i_op->xyz'
fields is quite costly.

We already optimized the dcache to not unnecessarily load the d_op
structure for members that are often NULL using the DCACHE_OP_xyz bits
in dentry->d_flags, and this does something very similar for the inode
ops that are used during pathname lookup.

It also re-orders the fields so that the fields accessed by 'stat' are
together at the beginning of the inode structure, and roughly in the
order accessed.

The effect of this seems to be in the 1-2% range for an empty kernel
"make -j" run (which is fairly kernel-intensive, mostly in filename
lookup), so it's visible.  The numbers are fairly noisy, though, and
likely depend a lot on exact microarchitecture.  So there's more tuning
to be done.
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Some inodes (pipes, sockets, ...) are not hashed, no need to take
contended inode_hash_lock at dismantle time.

nice speedup on SMP machines on socket intensive workloads.
Signed-off-by: NEric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>

Workloads using pipes and sockets hit inode_sb_list_lock contention.

superblock s_inodes list is needed for quota, dirty, pagecache and
fsnotify management. pipe/anon/socket fs are clearly not candidates for
these.
Signed-off-by: NEric Dumazet <eric.dumazet@gmail.com>
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>

Replace the ->check_acl method with a ->get_acl method that simply reads an
ACL from disk after having a cache miss.  This means we can replace the ACL
checking boilerplate code with a single implementation in namei.c.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>

For a number of file systems that don't have a mount point (e.g. sockfs
and pipefs), they are not marked as long term. Therefore in
mntput_no_expire, all locks in vfs_mount lock are taken instead of just
local cpu's lock to aggregate reference counts when we release
reference to file objects.  In fact, only local lock need to have been
taken to update ref counts as these file systems are in no danger of
going away until we are ready to unregister them.

The attached patch marks file systems using kern_mount without
mount point as long term.  The contentions of vfs_mount lock
is now eliminated.  Before un-registering such file system,
kern_unmount should be called to remove the long term flag and
make the mount point ready to be freed.
Signed-off-by: NTim Chen <tim.c.chen@linux.intel.com>
Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>

Fix build error when CONFIG_BLOCK is not enabled by providing a stub
inode_dio_wait() function.

mm/truncate.c:612: error: implicit declaration of function 'inode_dio_wait'
Signed-off-by: NRandy Dunlap <rdunlap@xenotime.net>
Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>

Big kernel lock had been removed and setlease now use the lock_flocks()
to hold a special spin lock file_lock_lock by Matthew.
So just remove the out-of-date NOTE.
Signed-off-by: NWanlong Gao <gaowanlong@cn.fujitsu.com>
Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>

Btrfs needs to be able to control how filemap_write_and_wait_range() is called
in fsync to make it less of a painful operation, so push down taking i_mutex and
the calling of filemap_write_and_wait() down into the ->fsync() handlers.  Some
file systems can drop taking the i_mutex altogether it seems, like ext3 and
ocfs2.  For correctness sake I just pushed everything down in all cases to make
sure that we keep the current behavior the same for everybody, and then each
individual fs maintainer can make up their mind about what to do from there.
Thanks,
Acked-by: NJan Kara <jack@suse.cz>
Signed-off-by: NJosef Bacik <josef@redhat.com>
Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>

This just gets us ready to support the SEEK_HOLE and SEEK_DATA flags.  Turns out
using fiemap in things like cp cause more problems than it solves, so lets try
and give userspace an interface that doesn't suck.  We need to match solaris
here, and the definitions are

*o* If /whence/ is SEEK_HOLE, the offset of the start of the
next hole greater than or equal to the supplied offset
is returned. The definition of a hole is provided near
the end of the DESCRIPTION.

*o* If /whence/ is SEEK_DATA, the file pointer is set to the
start of the next non-hole file region greater than or
equal to the supplied offset.

So in the generic case the entire file is data and there is a virtual hole at
the end.  That means we will just return i_size for SEEK_HOLE and will return
the same offset for SEEK_DATA.  This is how Solaris does it so we have to do it
the same way.

Thanks,
Signed-off-by: NJosef Bacik <josef@redhat.com>
Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>

Simple filesystems always pass inode->i_sb_bdev as the block device
argument, and never need a end_io handler.  Let's simply things for
them and for my grepping activity by dropping these arguments.  The
only thing not falling into that scheme is ext4, which passes and
end_io handler without needing special flags (yet), but given how
messy the direct I/O code there is use of __blockdev_direct_IO
in one instead of two out of three cases isn't going to make a large
difference anyway.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>

i_alloc_sem is a rather special rw_semaphore.  It's the last one that may
be released by a non-owner, and it's write side is always mirrored by
real exclusion.  It's intended use it to wait for all pending direct I/O
requests to finish before starting a truncate.

Replace it with a hand-grown construct:

 - exclusion for truncates is already guaranteed by i_mutex, so it can
   simply fall way
 - the reader side is replaced by an i_dio_count member in struct inode
   that counts the number of pending direct I/O requests.  Truncate can't
   proceed as long as it's non-zero
 - when i_dio_count reaches non-zero we wake up a pending truncate using
   wake_up_bit on a new bit in i_flags
 - new references to i_dio_count can't appear while we are waiting for
   it to read zero because the direct I/O count always needs i_mutex
   (or an equivalent like XFS's i_iolock) for starting a new operation.

This scheme is much simpler, and saves the space of a spinlock_t and a
struct list_head in struct inode (typically 160 bits on a non-debug 64-bit
system).
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>

Now we have a per-superblock shrinker implementation, we can add a
filesystem specific callout to it to allow filesystem internal
caches to be shrunk by the superblock shrinker.

Rather than perpetuate the multipurpose shrinker callback API (i.e.
nr_to_scan == 0 meaning "tell me how many objects freeable in the
cache), two operations will be added. The first will return the
number of objects that are freeable, the second is the actual
shrinker call.
Signed-off-by: NDave Chinner <dchinner@redhat.com>
Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>

With context based shrinkers, we can implement a per-superblock
shrinker that shrinks the caches attached to the superblock. We
currently have global shrinkers for the inode and dentry caches that
split up into per-superblock operations via a coarse proportioning
method that does not batch very well.  The global shrinkers also
have a dependency - dentries pin inodes - so we have to be very
careful about how we register the global shrinkers so that the
implicit call order is always correct.

With a per-sb shrinker callout, we can encode this dependency
directly into the per-sb shrinker, hence avoiding the need for
strictly ordering shrinker registrations. We also have no need for
any proportioning code for the shrinker subsystem already provides
this functionality across all shrinkers. Allowing the shrinker to
operate on a single superblock at a time means that we do less
superblock list traversals and locking and reclaim should batch more
effectively. This should result in less CPU overhead for reclaim and
potentially faster reclaim of items from each filesystem.
Signed-off-by: NDave Chinner <dchinner@redhat.com>
Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>

Both the filesystem and the lock manager can associate operations with a
lock.  Confusingly, one of them (fl_release_private) actually has the
same name in both operation structures.

It would save some confusion to give the lock-manager ops different
names.
Signed-off-by: NJ. Bruce Fields <bfields@redhat.com>

With the inode LRUs moving to per-sb structures, there is no longer
a need for a global inode_lru_lock. The locking can be made more
fine-grained by moving to a per-sb LRU lock, isolating the LRU
operations of different filesytsems completely from each other.
Signed-off-by: NDave Chinner <dchinner@redhat.com>
Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>

The inode unused list is currently a global LRU. This does not match
the other global filesystem cache - the dentry cache - which uses
per-superblock LRU lists. Hence we have related filesystem object
types using different LRU reclaimation schemes.

To enable a per-superblock filesystem cache shrinker, both of these
caches need to have per-sb unused object LRU lists. Hence this patch
converts the global inode LRU to per-sb LRUs.

The patch only does rudimentary per-sb propotioning in the shrinker
infrastructure, as this gets removed when the per-sb shrinker
callouts are introduced later on.
Signed-off-by: NDave Chinner <dchinner@redhat.com>
Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>

Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>

not used anymore
Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>

not used by the instances anymore.
Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>

redundant; all callers get it duplicated in mask & MAY_NOT_BLOCK and none of
them removes that bit.
Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>

not used in the instances anymore.
Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>

Duplicate the flags argument into mask bitmap.
Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>

its value depends only on inode and does not change; we might as
well store it in ->i_op->check_acl and be done with that.
Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>

new helpers: atomic_inc_unless_negative()/atomic_dec_unless_positive()
Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>

convert the last remaining caller to inode_permission()
Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>

Call the given function for all superblocks of given type.  Function
gets a superblock (with s_umount locked shared) and (void *) argument
supplied by caller of iterator.
Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>

fs_excl is a poor man's priority inheritance for filesystems to hint to
the block layer that an operation is important. It was never clearly
specified, not widely adopted, and will not prevent starvation in many
cases (like across cgroups).

fs_excl was introduced with the time sliced CFQ IO scheduler, to
indicate when a process held FS exclusive resources and thus needed
a boost.

It doesn't cover all file systems, and it was never fully complete.
Lets kill it.
Signed-off-by: NJustin TerAvest <teravest@google.com>
Signed-off-by: NJens Axboe <jaxboe@fusionio.com>

Under heavy memory and filesystem load, users observe the assertion
mapping->nrpages == 0 in end_writeback() trigger.  This can be caused by
page reclaim reclaiming the last page from a mapping in the following
race:

	CPU0				CPU1
  ...
  shrink_page_list()
    __remove_mapping()
      __delete_from_page_cache()
        radix_tree_delete()
					evict_inode()
					  truncate_inode_pages()
					    truncate_inode_pages_range()
					      pagevec_lookup() - finds nothing
					  end_writeback()
					    mapping->nrpages != 0 -> BUG
        page->mapping = NULL
        mapping->nrpages--

Fix the problem by doing a reliable check of mapping->nrpages under
mapping->tree_lock in end_writeback().

Analyzed by Jay <jinshan.xiong@whamcloud.com>, lost in LKML, and dug out
by Miklos Szeredi <mszeredi@suse.de>.

Cc: Jay <jinshan.xiong@whamcloud.com>
Cc: Miklos Szeredi <mszeredi@suse.de>
Signed-off-by: NJan Kara <jack@suse.cz>
Cc: <stable@kernel.org>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Commit 13e12d14 ("vfs: reorganize 'struct inode' layout a bit")
moved things around a bit changed i_state to be unsigned int instead of
unsigned long.  That was to help structure layout for the 64-bit case,
and shrink 'struct inode' a bit (admittedly that only happened when
spinlock debugging was on and i_flags didn't pack with i_lock).

However, Meelis Roos reports that this results in unaligned exceptions
on sprc, and it turns out that the bit-locking primitives that we use
for the I_NEW bit want to use the bitops.  Which want 'unsigned long',
not 'unsigned int'.

We really should fix the bit locking code to not have that kind of
requirement, but that's a much bigger change.  So for now, revert that
field back to 'unsigned long' (but keep the other re-ordering changes
from the commit that caused this).

Andi points out that we have played games with this in 'struct page', so
it's solvable with other hacks too, but since right now the struct inode
size advantage only happens with some rare config options, it's not
worth fighting.

It _would_ be worth fixing the bitlocking code, though.  Especially
since there is no type safety in the bitlocking code (this never caused
any warnings, and worked fine on x86-64, because the bitlocks take a
'void *' and x86-64 doesn't care that deeply about alignment).  So it's
currently a very easy problem to trigger by mistake and never notice.
Reported-by: NMeelis Roos <mroos@linux.ee>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: David Miller <davem@davemloft.net>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Many stupid corrections of duplicated includes based on the output of
scripts/checkincludes.pl.
Signed-off-by: NVitaliy Ivanov <vitalivanov@gmail.com>
Signed-off-by: NJiri Kosina <jkosina@suse.cz>

This tries to make the 'struct inode' accesses denser in the data cache
by moving a commonly accessed field (i_security) closer to other fields
that are accessed often.

It also makes 'i_state' just an 'unsigned int' rather than 'unsigned
long', since we only use a few bits of that field, and moves it next to
the existing 'i_flags' so that we potentially get better structure
layout (although depending on config options, i_flags may already have
packed in the same word as i_lock, so this improves packing only for the
case of spinlock debugging)

Out 'struct inode' is still way too big, and we should probably move
some other fields around too (the acl fields in particular) for better
data cache access density.  Other fields (like the inode hash) are
likely to be entirely irrelevant under most loads.
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Caching "we have already removed suid/caps" was overenthusiastic as merged.
On network filesystems we might have had suid/caps set on another client,
silently picked by this client on revalidate, all of that *without* clearing
the S_NOSEC flag.

AFAICS, the only reasonably sane way to deal with that is
	* new superblock flag; unless set, S_NOSEC is not going to be set.
	* local block filesystems set it in their ->mount() (more accurately,
mount_bdev() does, so does btrfs ->mount(), users of mount_bdev() other than
local block ones clear it)
	* if any network filesystem (or a cluster one) wants to use S_NOSEC,
it'll need to set MS_NOSEC in sb->s_flags *AND* take care to clear S_NOSEC when
inode attribute changes are picked from other clients.

It's not an earth-shattering hole (anybody that can set suid on another client
will almost certainly be able to write to the file before doing that anyway),
but it's a bug that needs fixing.
Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>

Some recent benchmarking on btrfs showed that a major scaling bottleneck
on large systems on btrfs is currently the xattr lookup on every write.

Why xattr lookup on every write I hear you ask?

write wants to drop suid and security related xattrs that could set o
capabilities for executables.  To do that it currently looks up
security.capability on EVERY write (even for non executables) to decide
whether to drop it or not.

In btrfs this causes an additional tree walk, hitting some per file system
locks and quite bad scalability. In a simple read workload on a 8S
system I saw over 90% CPU time in spinlocks related to that.

Chris Mason tells me this is also a problem in ext4, where it hits
the global mbcache lock.

This patch adds a simple per inode to avoid this problem.  We only
do the lookup once per file and then if there is no xattr cache
the decision. All xattr changes clear the flag.

I also used the same flag to avoid the suid check, although
that one is pretty cheap.

A file system can also set this flag when it creates the inode,
if it has a cheap way to do so.  This is done for some common file systems
in followon patches.

With this patch a major part of the lock contention disappears
for btrfs. Some testing on smaller systems didn't show significant
performance changes, but at least it helps the larger systems
and is generally more efficient.

v2: Rename is_sgid. add file system helper.
Cc: chris.mason@oracle.com
Cc: josef@redhat.com
Cc: viro@zeniv.linux.org.uk
Cc: agruen@linbit.com
Cc: Serge E. Hallyn <serue@us.ibm.com>
Signed-off-by: NAndi Kleen <ak@linux.intel.com>
Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>

Tell the filesystem if we just updated timestamp (I_DIRTY_SYNC) or
anything else, so that the filesystem can track internally if it
needs to push out a transaction for fdatasync or not.

This is just the prototype change with no user for it yet.  I plan
to push large XFS changes for the next merge window, and getting
this trivial infrastructure in this window would help a lot to avoid
tree interdependencies.

Also remove incorrect comments that ->dirty_inode can't block.  That
has been changed a long time ago, and many implementations rely on it.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>

This second patch of eight in this cleancache series adds a field to
the generic superblock to squirrel away a pool identifier that is
dynamically provided by cleancache-enabled filesystems at mount time
to uniquely identify files and pages belonging to this mounted filesystem.

Details and a FAQ can be found in Documentation/vm/cleancache.txt

[v8: trivial merge conflict update]
Signed-off-by: NDan Magenheimer <dan.magenheimer@oracle.com>
Reviewed-by: NJeremy Fitzhardinge <jeremy@goop.org>
Reviewed-by: NKonrad Rzeszutek Wilk <konrad.wilk@oracle.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Al Viro <viro@ZenIV.linux.org.uk>
Cc: Matthew Wilcox <matthew@wil.cx>
Cc: Nick Piggin <npiggin@kernel.dk>
Cc: Mel Gorman <mel@csn.ul.ie>
Cc: Rik Van Riel <riel@redhat.com>
Cc: Jan Beulich <JBeulich@novell.com>
Cc: Chris Mason <chris.mason@oracle.com>
Cc: Andreas Dilger <adilger@sun.com>
Cc: Ted Ts'o <tytso@mit.edu>
Cc: Mark Fasheh <mfasheh@suse.com>
Cc: Joel Becker <joel.becker@oracle.com>
Cc: Nitin Gupta <ngupta@vflare.org>

Apps are increasingly using more than 1024 file descriptors.  See
discussion in several distro bug trackers, e.g.  BugLink:
http://bugs.launchpad.net/bugs/663090
https://issues.rpath.com/browse/RPL-2054

You don't want to raise the default soft limit, since that might break
apps that use select(), but it's safe to raise the default hard limit;
that way, apps that know they need lots of file descriptors can raise
their soft limit without needing root, and without user intervention.

Ubuntu is doing this with a kernel change because they have a policy of
not changing kernel defaults in userland.

While 4096 might not be enough for *all* apps, it seems to be plenty for
the apps I've seen lately that are unhappy with 1024.
Signed-off-by: NTim Gardner <tim.gardner@canonical.com>
Cc: Dan Kegel <dank@kegel.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Christoph Hellwig <hch@lst.de>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Straightforward conversion of i_mmap_lock to a mutex.
Signed-off-by: NPeter Zijlstra <a.p.zijlstra@chello.nl>
Acked-by: NHugh Dickins <hughd@google.com>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: David Miller <davem@davemloft.net>
Cc: Martin Schwidefsky <schwidefsky@de.ibm.com>
Cc: Russell King <rmk@arm.linux.org.uk>
Cc: Paul Mundt <lethal@linux-sh.org>
Cc: Jeff Dike <jdike@addtoit.com>
Cc: Richard Weinberger <richard@nod.at>
Cc: Tony Luck <tony.luck@intel.com>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: Mel Gorman <mel@csn.ul.ie>
Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: Nick Piggin <npiggin@kernel.dk>
Cc: Namhyung Kim <namhyung@gmail.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Hugh says:
 "The only significant loser, I think, would be page reclaim (when
  concurrent with truncation): could spin for a long time waiting for
  the i_mmap_mutex it expects would soon be dropped? "

Counter points:
 - cpu contention makes the spin stop (need_resched())
 - zap pages should be freeing pages at a higher rate than reclaim
   ever can

I think the simplification of the truncate code is definitely worth it.

Effectively reverts: 2aa15890 ("mm: prevent concurrent
unmap_mapping_range() on the same inode") and takes out the code that
caused its problem.
Signed-off-by: NPeter Zijlstra <a.p.zijlstra@chello.nl>
Reviewed-by: NKAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: Hugh Dickins <hughd@google.com>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: David Miller <davem@davemloft.net>
Cc: Martin Schwidefsky <schwidefsky@de.ibm.com>
Cc: Russell King <rmk@arm.linux.org.uk>
Cc: Paul Mundt <lethal@linux-sh.org>
Cc: Jeff Dike <jdike@addtoit.com>
Cc: Richard Weinberger <richard@nod.at>
Cc: Tony Luck <tony.luck@intel.com>
Cc: Mel Gorman <mel@csn.ul.ie>
Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: Nick Piggin <npiggin@kernel.dk>
Cc: Namhyung Kim <namhyung@gmail.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

FS_COW_FL and FS_NOCOW_FL were newly introduced to control per file
COW in btrfs, but FS_NOCOW_FL is sufficient.

The fact is we don't have corresponding BTRFS_INODE_COW flag.

COW is default, and FS_NOCOW_FL can be used to switch off COW for
a single file.

If we mount btrfs with nodatacow, a newly created file will be set with
the FS_NOCOW_FL flag. So to turn on COW for it, we can just clear the
FS_NOCOW_FL flag.
Signed-off-by: NLi Zefan <lizf@cn.fujitsu.com>
Signed-off-by: NChris Mason <chris.mason@oracle.com>

Gaah.  When commit be85bcca reverted the export of file system uuid
via /proc/<pid>/mountinfo, it also unintentionally removed the s_uuid
field in struct super_block.

I didn't mean to do that, since filesystems have been taught to fill it
in (and we want to keep it for future re-introduction in the mountinfo
file).

Stupid of me. This adds it back in.
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>